diff --git a/.vscode-remote/data/CachedProfilesData/__default__profile__/extensions.builtin.cache b/.vscode-remote/data/CachedProfilesData/__default__profile__/extensions.builtin.cache
new file mode 100644
index 0000000000000..7f2557fb08d5e
--- /dev/null
+++ b/.vscode-remote/data/CachedProfilesData/__default__profile__/extensions.builtin.cache
@@ -0,0 +1 @@
+{"input":{"location":{"$mid":1,"fsPath":"/ide/extensions","external":"file:///ide/extensions","path":"/ide/extensions","scheme":"file"},"mtime":1733414898000,"profile":false,"type":0,"validate":true,"productVersion":"1.97.2","productDate":"2025-02-27T07:12:40.470Z","productCommit":"e81ab749002377cf40b1551661a6b80725141ff0","devMode":false,"translations":{}},"result":[{"type":0,"identifier":{"id":"vscode.bat"},"manifest":{"name":"bat","displayName":"Windows Bat Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in Windows batch files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.52.0"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin mmims/language-batchfile grammars/batchfile.cson ./syntaxes/batchfile.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"bat","extensions":[".bat",".cmd"],"aliases":["Batch","bat"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"bat","scopeName":"source.batchfile","path":"./syntaxes/batchfile.tmLanguage.json"}],"snippets":[{"language":"bat","path":"./snippets/batchfile.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/bat","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.clojure"},"manifest":{"name":"clojure","displayName":"Clojure Language Basics","description":"Provides syntax highlighting and bracket matching in Clojure files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin atom/language-clojure grammars/clojure.cson ./syntaxes/clojure.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"clojure","aliases":["Clojure","clojure"],"extensions":[".clj",".cljs",".cljc",".cljx",".clojure",".edn"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"clojure","scopeName":"source.clojure","path":"./syntaxes/clojure.tmLanguage.json"}],"configurationDefaults":{"[clojure]":{"diffEditor.ignoreTrimWhitespace":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/clojure","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.coffeescript"},"manifest":{"name":"coffeescript","displayName":"CoffeeScript Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in CoffeeScript files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin atom/language-coffee-script grammars/coffeescript.cson ./syntaxes/coffeescript.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"coffeescript","extensions":[".coffee",".cson",".iced"],"aliases":["CoffeeScript","coffeescript","coffee"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"coffeescript","scopeName":"source.coffee","path":"./syntaxes/coffeescript.tmLanguage.json"}],"breakpoints":[{"language":"coffeescript"}],"snippets":[{"language":"coffeescript","path":"./snippets/coffeescript.code-snippets"}],"configurationDefaults":{"[coffeescript]":{"diffEditor.ignoreTrimWhitespace":false,"editor.defaultColorDecorators":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/coffeescript","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.configuration-editing"},"manifest":{"name":"configuration-editing","displayName":"Configuration Editing","description":"Provides capabilities (advanced IntelliSense, auto-fixing) in configuration files like settings, launch, and extension recommendation files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.0.0"},"icon":"images/icon.png","activationEvents":["onProfile","onProfile:github","onLanguage:json","onLanguage:jsonc"],"enabledApiProposals":["profileContentHandlers"],"main":"./dist/configurationEditingMain","browser":"./dist/browser/configurationEditingMain","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"languages":[{"id":"jsonc","extensions":[".code-workspace","language-configuration.json","icon-theme.json","color-theme.json"],"filenames":["settings.json","launch.json","tasks.json","keybindings.json","extensions.json","argv.json","profiles.json","devcontainer.json",".devcontainer.json"]},{"id":"json","extensions":[".code-profile"]}],"jsonValidation":[{"fileMatch":"vscode://defaultsettings/keybindings.json","url":"vscode://schemas/keybindings"},{"fileMatch":"%APP_SETTINGS_HOME%/keybindings.json","url":"vscode://schemas/keybindings"},{"fileMatch":"%APP_SETTINGS_HOME%/profiles/*/keybindings.json","url":"vscode://schemas/keybindings"},{"fileMatch":"vscode://defaultsettings/*.json","url":"vscode://schemas/settings/default"},{"fileMatch":"%APP_SETTINGS_HOME%/settings.json","url":"vscode://schemas/settings/user"},{"fileMatch":"%APP_SETTINGS_HOME%/profiles/*/settings.json","url":"vscode://schemas/settings/profile"},{"fileMatch":"%MACHINE_SETTINGS_HOME%/settings.json","url":"vscode://schemas/settings/machine"},{"fileMatch":"%APP_WORKSPACES_HOME%/*/workspace.json","url":"vscode://schemas/workspaceConfig"},{"fileMatch":"**/*.code-workspace","url":"vscode://schemas/workspaceConfig"},{"fileMatch":"**/argv.json","url":"vscode://schemas/argv"},{"fileMatch":"/.vscode/settings.json","url":"vscode://schemas/settings/folder"},{"fileMatch":"/.vscode/launch.json","url":"vscode://schemas/launch"},{"fileMatch":"/.vscode/tasks.json","url":"vscode://schemas/tasks"},{"fileMatch":"%APP_SETTINGS_HOME%/tasks.json","url":"vscode://schemas/tasks"},{"fileMatch":"%APP_SETTINGS_HOME%/snippets/*.json","url":"vscode://schemas/snippets"},{"fileMatch":"%APP_SETTINGS_HOME%/profiles/*/snippets/.json","url":"vscode://schemas/snippets"},{"fileMatch":"%APP_SETTINGS_HOME%/sync/snippets/preview/*.json","url":"vscode://schemas/snippets"},{"fileMatch":"**/*.code-snippets","url":"vscode://schemas/global-snippets"},{"fileMatch":"/.vscode/extensions.json","url":"vscode://schemas/extensions"},{"fileMatch":"devcontainer.json","url":"https://raw.githubusercontent.com/devcontainers/spec/main/schemas/devContainer.schema.json"},{"fileMatch":".devcontainer.json","url":"https://raw.githubusercontent.com/devcontainers/spec/main/schemas/devContainer.schema.json"},{"fileMatch":"%APP_SETTINGS_HOME%/globalStorage/ms-vscode-remote.remote-containers/nameConfigs/*.json","url":"./schemas/attachContainer.schema.json"},{"fileMatch":"%APP_SETTINGS_HOME%/globalStorage/ms-vscode-remote.remote-containers/imageConfigs/*.json","url":"./schemas/attachContainer.schema.json"},{"fileMatch":"**/quality/*/product.json","url":"vscode://schemas/vscode-product"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/configuration-editing","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.cpp"},"manifest":{"name":"cpp","displayName":"C/C++ Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in C/C++ files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammars.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"c","extensions":[".c",".i"],"aliases":["C","c"],"configuration":"./language-configuration.json"},{"id":"cpp","extensions":[".cpp",".cppm",".cc",".ccm",".cxx",".cxxm",".c++",".c++m",".hpp",".hh",".hxx",".h++",".h",".ii",".ino",".inl",".ipp",".ixx",".tpp",".txx",".hpp.in",".h.in"],"aliases":["C++","Cpp","cpp"],"configuration":"./language-configuration.json"},{"id":"cuda-cpp","extensions":[".cu",".cuh"],"aliases":["CUDA C++"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"c","scopeName":"source.c","path":"./syntaxes/c.tmLanguage.json"},{"language":"cpp","scopeName":"source.cpp.embedded.macro","path":"./syntaxes/cpp.embedded.macro.tmLanguage.json"},{"language":"cpp","scopeName":"source.cpp","path":"./syntaxes/cpp.tmLanguage.json"},{"scopeName":"source.c.platform","path":"./syntaxes/platform.tmLanguage.json"},{"language":"cuda-cpp","scopeName":"source.cuda-cpp","path":"./syntaxes/cuda-cpp.tmLanguage.json"}],"problemPatterns":[{"name":"nvcc-location","regexp":"^(.*)\\((\\d+)\\):\\s+(warning|error):\\s+(.*)","kind":"location","file":1,"location":2,"severity":3,"message":4}],"problemMatchers":[{"name":"nvcc","owner":"cuda-cpp","fileLocation":["relative","${workspaceFolder}"],"pattern":"$nvcc-location"}],"snippets":[{"language":"c","path":"./snippets/c.code-snippets"},{"language":"cpp","path":"./snippets/cpp.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/cpp","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.csharp"},"manifest":{"name":"csharp","displayName":"C# Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in C# files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin dotnet/csharp-tmLanguage grammars/csharp.tmLanguage ./syntaxes/csharp.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"configurationDefaults":{"[csharp]":{"editor.maxTokenizationLineLength":2500}},"languages":[{"id":"csharp","extensions":[".cs",".csx",".cake"],"aliases":["C#","csharp"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"csharp","scopeName":"source.cs","path":"./syntaxes/csharp.tmLanguage.json","tokenTypes":{"meta.interpolation":"other"}}],"snippets":[{"language":"csharp","path":"./snippets/csharp.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/csharp","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.css"},"manifest":{"name":"css","displayName":"CSS Language Basics","description":"Provides syntax highlighting and bracket matching for CSS, LESS and SCSS files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin microsoft/vscode-css grammars/css.cson ./syntaxes/css.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"css","aliases":["CSS","css"],"extensions":[".css"],"mimetypes":["text/css"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"css","scopeName":"source.css","path":"./syntaxes/css.tmLanguage.json","tokenTypes":{"meta.function.url string.quoted":"other"}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/css","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.css-language-features"},"manifest":{"name":"css-language-features","displayName":"CSS Language Features","description":"Provides rich language support for CSS, LESS and SCSS files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.77.0"},"icon":"icons/css.png","activationEvents":["onLanguage:css","onLanguage:less","onLanguage:scss","onCommand:_css.applyCodeAction"],"main":"./client/dist/node/cssClientMain","browser":"./client/dist/browser/cssClientMain","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"categories":["Programming Languages"],"contributes":{"configuration":[{"order":22,"id":"css","title":"CSS","properties":{"css.customData":{"type":"array","markdownDescription":"A list of relative file paths pointing to JSON files following the [custom data format](https://github.com/microsoft/vscode-css-languageservice/blob/master/docs/customData.md).\n\nVS Code loads custom data on startup to enhance its CSS support for CSS custom properties (variables), at-rules, pseudo-classes, and pseudo-elements you specify in the JSON files.\n\nThe file paths are relative to workspace and only workspace folder settings are considered.","default":[],"items":{"type":"string"},"scope":"resource"},"css.completion.triggerPropertyValueCompletion":{"type":"boolean","scope":"resource","default":true,"description":"By default, VS Code triggers property value completion after selecting a CSS property. Use this setting to disable this behavior."},"css.completion.completePropertyWithSemicolon":{"type":"boolean","scope":"resource","default":true,"description":"Insert semicolon at end of line when completing CSS properties."},"css.validate":{"type":"boolean","scope":"resource","default":true,"description":"Enables or disables all validations."},"css.hover.documentation":{"type":"boolean","scope":"resource","default":true,"description":"Show property and value documentation in CSS hovers."},"css.hover.references":{"type":"boolean","scope":"resource","default":true,"description":"Show references to MDN in CSS hovers."},"css.lint.compatibleVendorPrefixes":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"When using a vendor-specific prefix make sure to also include all other vendor-specific properties."},"css.lint.vendorPrefix":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"When using a vendor-specific prefix, also include the standard property."},"css.lint.duplicateProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Do not use duplicate style definitions."},"css.lint.emptyRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Do not use empty rulesets."},"css.lint.importStatement":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Import statements do not load in parallel."},"css.lint.boxModel":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Do not use `width` or `height` when using `padding` or `border`."},"css.lint.universalSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"The universal selector (`*`) is known to be slow."},"css.lint.zeroUnits":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"No unit for zero needed."},"css.lint.fontFaceProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"`@font-face` rule must define `src` and `font-family` properties."},"css.lint.hexColorLength":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Hex colors must consist of 3, 4, 6 or 8 hex numbers."},"css.lint.argumentsInColorFunction":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Invalid number of parameters."},"css.lint.unknownProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown property."},"css.lint.validProperties":{"type":"array","uniqueItems":true,"items":{"type":"string"},"scope":"resource","default":[],"description":"A list of properties that are not validated against the `unknownProperties` rule."},"css.lint.ieHack":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"IE hacks are only necessary when supporting IE7 and older."},"css.lint.unknownVendorSpecificProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Unknown vendor specific property."},"css.lint.propertyIgnoredDueToDisplay":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"Property is ignored due to the display. E.g. with `display: inline`, the `width`, `height`, `margin-top`, `margin-bottom`, and `float` properties have no effect."},"css.lint.important":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `!important`. It is an indication that the specificity of the entire CSS has gotten out of control and needs to be refactored."},"css.lint.float":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `float`. Floats lead to fragile CSS that is easy to break if one aspect of the layout changes."},"css.lint.idSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Selectors should not contain IDs because these rules are too tightly coupled with the HTML."},"css.lint.unknownAtRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown at-rule."},"css.trace.server":{"type":"string","scope":"window","enum":["off","messages","verbose"],"default":"off","description":"Traces the communication between VS Code and the CSS language server."},"css.format.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable default CSS formatter."},"css.format.newlineBetweenSelectors":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate selectors with a new line."},"css.format.newlineBetweenRules":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate rulesets by a blank line."},"css.format.spaceAroundSelectorSeparator":{"type":"boolean","scope":"resource","default":false,"markdownDescription":"Ensure a space character around selector separators '>', '+', '~' (e.g. `a > b`)."},"css.format.braceStyle":{"type":"string","scope":"resource","default":"collapse","enum":["collapse","expand"],"markdownDescription":"Put braces on the same line as rules (`collapse`) or put braces on own line (`expand`)."},"css.format.preserveNewLines":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Whether existing line breaks before rules and declarations should be preserved."},"css.format.maxPreserveNewLines":{"type":["number","null"],"scope":"resource","default":null,"markdownDescription":"Maximum number of line breaks to be preserved in one chunk, when `#css.format.preserveNewLines#` is enabled."}}},{"id":"scss","order":24,"title":"SCSS (Sass)","properties":{"scss.completion.triggerPropertyValueCompletion":{"type":"boolean","scope":"resource","default":true,"description":"By default, VS Code triggers property value completion after selecting a CSS property. Use this setting to disable this behavior."},"scss.completion.completePropertyWithSemicolon":{"type":"boolean","scope":"resource","default":true,"description":"Insert semicolon at end of line when completing CSS properties."},"scss.validate":{"type":"boolean","scope":"resource","default":true,"description":"Enables or disables all validations."},"scss.hover.documentation":{"type":"boolean","scope":"resource","default":true,"description":"Show property and value documentation in SCSS hovers."},"scss.hover.references":{"type":"boolean","scope":"resource","default":true,"description":"Show references to MDN in SCSS hovers."},"scss.lint.compatibleVendorPrefixes":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"When using a vendor-specific prefix make sure to also include all other vendor-specific properties."},"scss.lint.vendorPrefix":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"When using a vendor-specific prefix, also include the standard property."},"scss.lint.duplicateProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Do not use duplicate style definitions."},"scss.lint.emptyRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Do not use empty rulesets."},"scss.lint.importStatement":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Import statements do not load in parallel."},"scss.lint.boxModel":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Do not use `width` or `height` when using `padding` or `border`."},"scss.lint.universalSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"The universal selector (`*`) is known to be slow."},"scss.lint.zeroUnits":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"No unit for zero needed."},"scss.lint.fontFaceProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"`@font-face` rule must define `src` and `font-family` properties."},"scss.lint.hexColorLength":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Hex colors must consist of 3, 4, 6 or 8 hex numbers."},"scss.lint.argumentsInColorFunction":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Invalid number of parameters."},"scss.lint.unknownProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown property."},"scss.lint.validProperties":{"type":"array","uniqueItems":true,"items":{"type":"string"},"scope":"resource","default":[],"description":"A list of properties that are not validated against the `unknownProperties` rule."},"scss.lint.ieHack":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"IE hacks are only necessary when supporting IE7 and older."},"scss.lint.unknownVendorSpecificProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Unknown vendor specific property."},"scss.lint.propertyIgnoredDueToDisplay":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"Property is ignored due to the display. E.g. with `display: inline`, the `width`, `height`, `margin-top`, `margin-bottom`, and `float` properties have no effect."},"scss.lint.important":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `!important`. It is an indication that the specificity of the entire CSS has gotten out of control and needs to be refactored."},"scss.lint.float":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `float`. Floats lead to fragile CSS that is easy to break if one aspect of the layout changes."},"scss.lint.idSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Selectors should not contain IDs because these rules are too tightly coupled with the HTML."},"scss.lint.unknownAtRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown at-rule."},"scss.format.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable default SCSS formatter."},"scss.format.newlineBetweenSelectors":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate selectors with a new line."},"scss.format.newlineBetweenRules":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate rulesets by a blank line."},"scss.format.spaceAroundSelectorSeparator":{"type":"boolean","scope":"resource","default":false,"markdownDescription":"Ensure a space character around selector separators '>', '+', '~' (e.g. `a > b`)."},"scss.format.braceStyle":{"type":"string","scope":"resource","default":"collapse","enum":["collapse","expand"],"markdownDescription":"Put braces on the same line as rules (`collapse`) or put braces on own line (`expand`)."},"scss.format.preserveNewLines":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Whether existing line breaks before rules and declarations should be preserved."},"scss.format.maxPreserveNewLines":{"type":["number","null"],"scope":"resource","default":null,"markdownDescription":"Maximum number of line breaks to be preserved in one chunk, when `#scss.format.preserveNewLines#` is enabled."}}},{"id":"less","order":23,"type":"object","title":"LESS","properties":{"less.completion.triggerPropertyValueCompletion":{"type":"boolean","scope":"resource","default":true,"description":"By default, VS Code triggers property value completion after selecting a CSS property. Use this setting to disable this behavior."},"less.completion.completePropertyWithSemicolon":{"type":"boolean","scope":"resource","default":true,"description":"Insert semicolon at end of line when completing CSS properties."},"less.validate":{"type":"boolean","scope":"resource","default":true,"description":"Enables or disables all validations."},"less.hover.documentation":{"type":"boolean","scope":"resource","default":true,"description":"Show property and value documentation in LESS hovers."},"less.hover.references":{"type":"boolean","scope":"resource","default":true,"description":"Show references to MDN in LESS hovers."},"less.lint.compatibleVendorPrefixes":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"When using a vendor-specific prefix make sure to also include all other vendor-specific properties."},"less.lint.vendorPrefix":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"When using a vendor-specific prefix, also include the standard property."},"less.lint.duplicateProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Do not use duplicate style definitions."},"less.lint.emptyRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Do not use empty rulesets."},"less.lint.importStatement":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Import statements do not load in parallel."},"less.lint.boxModel":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Do not use `width` or `height` when using `padding` or `border`."},"less.lint.universalSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"The universal selector (`*`) is known to be slow."},"less.lint.zeroUnits":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"No unit for zero needed."},"less.lint.fontFaceProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"`@font-face` rule must define `src` and `font-family` properties."},"less.lint.hexColorLength":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Hex colors must consist of 3, 4, 6 or 8 hex numbers."},"less.lint.argumentsInColorFunction":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"error","description":"Invalid number of parameters."},"less.lint.unknownProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown property."},"less.lint.validProperties":{"type":"array","uniqueItems":true,"items":{"type":"string"},"scope":"resource","default":[],"description":"A list of properties that are not validated against the `unknownProperties` rule."},"less.lint.ieHack":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"IE hacks are only necessary when supporting IE7 and older."},"less.lint.unknownVendorSpecificProperties":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Unknown vendor specific property."},"less.lint.propertyIgnoredDueToDisplay":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","markdownDescription":"Property is ignored due to the display. E.g. with `display: inline`, the `width`, `height`, `margin-top`, `margin-bottom`, and `float` properties have no effect."},"less.lint.important":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `!important`. It is an indication that the specificity of the entire CSS has gotten out of control and needs to be refactored."},"less.lint.float":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","markdownDescription":"Avoid using `float`. Floats lead to fragile CSS that is easy to break if one aspect of the layout changes."},"less.lint.idSelector":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"ignore","description":"Selectors should not contain IDs because these rules are too tightly coupled with the HTML."},"less.lint.unknownAtRules":{"type":"string","scope":"resource","enum":["ignore","warning","error"],"default":"warning","description":"Unknown at-rule."},"less.format.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable default LESS formatter."},"less.format.newlineBetweenSelectors":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate selectors with a new line."},"less.format.newlineBetweenRules":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Separate rulesets by a blank line."},"less.format.spaceAroundSelectorSeparator":{"type":"boolean","scope":"resource","default":false,"markdownDescription":"Ensure a space character around selector separators '>', '+', '~' (e.g. `a > b`)."},"less.format.braceStyle":{"type":"string","scope":"resource","default":"collapse","enum":["collapse","expand"],"markdownDescription":"Put braces on the same line as rules (`collapse`) or put braces on own line (`expand`)."},"less.format.preserveNewLines":{"type":"boolean","scope":"resource","default":true,"markdownDescription":"Whether existing line breaks before rules and declarations should be preserved."},"less.format.maxPreserveNewLines":{"type":["number","null"],"scope":"resource","default":null,"markdownDescription":"Maximum number of line breaks to be preserved in one chunk, when `#less.format.preserveNewLines#` is enabled."}}}],"configurationDefaults":{"[css]":{"editor.suggest.insertMode":"replace"},"[scss]":{"editor.suggest.insertMode":"replace"},"[less]":{"editor.suggest.insertMode":"replace"}},"jsonValidation":[{"fileMatch":"*.css-data.json","url":"https://raw.githubusercontent.com/microsoft/vscode-css-languageservice/master/docs/customData.schema.json"},{"fileMatch":"package.json","url":"./schemas/package.schema.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/css-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.dart"},"manifest":{"name":"dart","displayName":"Dart Language Basics","description":"Provides syntax highlighting & bracket matching in Dart files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin dart-lang/dart-syntax-highlight grammars/dart.json ./syntaxes/dart.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"dart","extensions":[".dart"],"aliases":["Dart"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"dart","scopeName":"source.dart","path":"./syntaxes/dart.tmLanguage.json"}]}},"location":{"$mid":1,"path":"/ide/extensions/dart","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.debug-auto-launch"},"manifest":{"name":"debug-auto-launch","displayName":"Node Debug Auto-attach","description":"Helper for auto-attach feature when node-debug extensions are not active.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.5.0"},"icon":"media/icon.png","capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"activationEvents":["onStartupFinished"],"main":"./dist/extension","contributes":{"commands":[{"command":"extension.node-debug.toggleAutoAttach","title":"Toggle Auto Attach","category":"Debug"}]},"prettier":{"printWidth":100,"trailingComma":"all","singleQuote":true,"arrowParens":"avoid"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/debug-auto-launch","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.debug-server-ready"},"manifest":{"name":"debug-server-ready","displayName":"Server Ready Action","description":"Open URI in browser if server under debugging is ready.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.32.0"},"icon":"media/icon.png","activationEvents":["onDebugResolve"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"enabledApiProposals":["terminalDataWriteEvent"],"main":"./dist/extension","contributes":{"debuggers":[{"type":"*","configurationAttributes":{"launch":{"properties":{"serverReadyAction":{"oneOf":[{"type":"object","additionalProperties":false,"markdownDescription":"Act upon a URI when a server program under debugging is ready (indicated by sending output of the form 'listening on port 3000' or 'Now listening on: https://localhost:5001' to the debug console.)","default":{"action":"openExternally","killOnServerStop":false},"properties":{"action":{"type":"string","enum":["openExternally"],"enumDescriptions":["Open URI externally with the default application."],"markdownDescription":"What to do with the URI when the server is ready.","default":"openExternally"},"pattern":{"type":"string","markdownDescription":"Server is ready if this pattern appears on the debug console. The first capture group must include a URI or a port number.","default":"listening on port ([0-9]+)"},"uriFormat":{"type":"string","markdownDescription":"A format string used when constructing the URI from a port number. The first '%s' is substituted with the port number.","default":"http://localhost:%s"},"killOnServerStop":{"type":"boolean","markdownDescription":"Stop the child session when the parent session stopped.","default":false}}},{"type":"object","additionalProperties":false,"markdownDescription":"Act upon a URI when a server program under debugging is ready (indicated by sending output of the form 'listening on port 3000' or 'Now listening on: https://localhost:5001' to the debug console.)","default":{"action":"debugWithEdge","pattern":"listening on port ([0-9]+)","uriFormat":"http://localhost:%s","webRoot":"${workspaceFolder}","killOnServerStop":false},"properties":{"action":{"type":"string","enum":["debugWithChrome","debugWithEdge"],"enumDescriptions":["Start debugging with the 'Debugger for Chrome'."],"markdownDescription":"What to do with the URI when the server is ready.","default":"debugWithEdge"},"pattern":{"type":"string","markdownDescription":"Server is ready if this pattern appears on the debug console. The first capture group must include a URI or a port number.","default":"listening on port ([0-9]+)"},"uriFormat":{"type":"string","markdownDescription":"A format string used when constructing the URI from a port number. The first '%s' is substituted with the port number.","default":"http://localhost:%s"},"webRoot":{"type":"string","markdownDescription":"Value passed to the debug configuration for the 'Debugger for Chrome'.","default":"${workspaceFolder}"},"killOnServerStop":{"type":"boolean","markdownDescription":"Stop the child session when the parent session stopped.","default":false}}},{"type":"object","additionalProperties":false,"markdownDescription":"Act upon a URI when a server program under debugging is ready (indicated by sending output of the form 'listening on port 3000' or 'Now listening on: https://localhost:5001' to the debug console.)","default":{"action":"startDebugging","name":"<launch browser config name>","killOnServerStop":false},"required":["name"],"properties":{"action":{"type":"string","enum":["startDebugging"],"enumDescriptions":["Run another launch configuration."],"markdownDescription":"What to do with the URI when the server is ready.","default":"startDebugging"},"pattern":{"type":"string","markdownDescription":"Server is ready if this pattern appears on the debug console. The first capture group must include a URI or a port number.","default":"listening on port ([0-9]+)"},"name":{"type":"string","markdownDescription":"Name of the launch configuration to run.","default":"Launch Browser"},"killOnServerStop":{"type":"boolean","markdownDescription":"Stop the child session when the parent session stopped.","default":false}}},{"type":"object","additionalProperties":false,"markdownDescription":"Act upon a URI when a server program under debugging is ready (indicated by sending output of the form 'listening on port 3000' or 'Now listening on: https://localhost:5001' to the debug console.)","default":{"action":"startDebugging","config":{"type":"node","request":"launch"},"killOnServerStop":false},"required":["config"],"properties":{"action":{"type":"string","enum":["startDebugging"],"enumDescriptions":["Run another launch configuration."],"markdownDescription":"What to do with the URI when the server is ready.","default":"startDebugging"},"pattern":{"type":"string","markdownDescription":"Server is ready if this pattern appears on the debug console. The first capture group must include a URI or a port number.","default":"listening on port ([0-9]+)"},"config":{"type":"object","markdownDescription":"The debug configuration to run.","default":{}},"killOnServerStop":{"type":"boolean","markdownDescription":"Stop the child session when the parent session stopped.","default":false}}}]}}}}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/debug-server-ready","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.diff"},"manifest":{"name":"diff","displayName":"Diff Language Basics","description":"Provides syntax highlighting & bracket matching in Diff files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin textmate/diff.tmbundle Syntaxes/Diff.plist ./syntaxes/diff.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"diff","aliases":["Diff","diff"],"extensions":[".diff",".patch",".rej"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"diff","scopeName":"source.diff","path":"./syntaxes/diff.tmLanguage.json"}]}},"location":{"$mid":1,"path":"/ide/extensions/diff","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.docker"},"manifest":{"name":"docker","displayName":"Docker Language Basics","description":"Provides syntax highlighting and bracket matching in Docker files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin moby/moby contrib/syntax/textmate/Docker.tmbundle/Syntaxes/Dockerfile.tmLanguage ./syntaxes/docker.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"dockerfile","extensions":[".dockerfile",".containerfile"],"filenames":["Dockerfile","Containerfile"],"filenamePatterns":["Dockerfile.*","Containerfile.*"],"aliases":["Docker","Dockerfile","Containerfile"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"dockerfile","scopeName":"source.dockerfile","path":"./syntaxes/docker.tmLanguage.json"}],"configurationDefaults":{"[dockerfile]":{"editor.quickSuggestions":{"strings":true}}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/docker","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.emmet"},"manifest":{"name":"emmet","displayName":"Emmet","description":"Emmet support for VS Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.13.0"},"icon":"images/icon.png","categories":["Other"],"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"},"activationEvents":["onCommand:emmet.expandAbbreviation","onLanguage"],"main":"./dist/node/emmetNodeMain","browser":"./dist/browser/emmetBrowserMain","contributes":{"configuration":{"type":"object","title":"Emmet","properties":{"emmet.showExpandedAbbreviation":{"type":["string"],"enum":["never","always","inMarkupAndStylesheetFilesOnly"],"default":"always","markdownDescription":"Shows expanded Emmet abbreviations as suggestions.\nThe option `\"inMarkupAndStylesheetFilesOnly\"` applies to html, haml, jade, slim, xml, xsl, css, scss, sass, less and stylus.\nThe option `\"always\"` applies to all parts of the file regardless of markup/css."},"emmet.showAbbreviationSuggestions":{"type":"boolean","default":true,"scope":"language-overridable","markdownDescription":"Shows possible Emmet abbreviations as suggestions. Not applicable in stylesheets or when emmet.showExpandedAbbreviation is set to `\"never\"`."},"emmet.includeLanguages":{"type":"object","additionalProperties":{"type":"string"},"default":{},"markdownDescription":"Enable Emmet abbreviations in languages that are not supported by default. Add a mapping here between the language and Emmet supported language.\n For example: `{\"vue-html\": \"html\", \"javascript\": \"javascriptreact\"}`"},"emmet.variables":{"type":"object","properties":{"lang":{"type":"string","default":"en"},"charset":{"type":"string","default":"UTF-8"}},"additionalProperties":{"type":"string"},"default":{},"markdownDescription":"Variables to be used in Emmet snippets."},"emmet.syntaxProfiles":{"type":"object","default":{},"markdownDescription":"Define profile for specified syntax or use your own profile with specific rules."},"emmet.excludeLanguages":{"type":"array","items":{"type":"string"},"default":["markdown"],"markdownDescription":"An array of languages where Emmet abbreviations should not be expanded."},"emmet.extensionsPath":{"type":"array","items":{"type":"string","markdownDescription":"A path containing Emmet syntaxProfiles and/or snippets."},"default":[],"scope":"machine-overridable","markdownDescription":"An array of paths, where each path can contain Emmet syntaxProfiles and/or snippet files.\nIn case of conflicts, the profiles/snippets of later paths will override those of earlier paths.\nSee https://code.visualstudio.com/docs/editor/emmet for more information and an example snippet file."},"emmet.triggerExpansionOnTab":{"type":"boolean","default":false,"scope":"language-overridable","markdownDescription":"When enabled, Emmet abbreviations are expanded when pressing TAB, even when completions do not show up. When disabled, completions that show up can still be accepted by pressing TAB."},"emmet.useInlineCompletions":{"type":"boolean","default":false,"markdownDescription":"If `true`, Emmet will use inline completions to suggest expansions. To prevent the non-inline completion item provider from showing up as often while this setting is `true`, turn `#editor.quickSuggestions#` to `inline` or `off` for the `other` item."},"emmet.preferences":{"type":"object","default":{},"markdownDescription":"Preferences used to modify behavior of some actions and resolvers of Emmet.","properties":{"css.intUnit":{"type":"string","default":"px","markdownDescription":"Default unit for integer values."},"css.floatUnit":{"type":"string","default":"em","markdownDescription":"Default unit for float values."},"css.propertyEnd":{"type":"string","default":";","markdownDescription":"Symbol to be placed at the end of CSS property when expanding CSS abbreviations."},"sass.propertyEnd":{"type":"string","default":"","markdownDescription":"Symbol to be placed at the end of CSS property when expanding CSS abbreviations in Sass files."},"stylus.propertyEnd":{"type":"string","default":"","markdownDescription":"Symbol to be placed at the end of CSS property when expanding CSS abbreviations in Stylus files."},"css.valueSeparator":{"type":"string","default":": ","markdownDescription":"Symbol to be placed at the between CSS property and value when expanding CSS abbreviations."},"sass.valueSeparator":{"type":"string","default":": ","markdownDescription":"Symbol to be placed at the between CSS property and value when expanding CSS abbreviations in Sass files."},"stylus.valueSeparator":{"type":"string","default":" ","markdownDescription":"Symbol to be placed at the between CSS property and value when expanding CSS abbreviations in Stylus files."},"bem.elementSeparator":{"type":"string","default":"__","markdownDescription":"Element separator used for classes when using the BEM filter."},"bem.modifierSeparator":{"type":"string","default":"_","markdownDescription":"Modifier separator used for classes when using the BEM filter."},"filter.commentBefore":{"type":"string","default":"","markdownDescription":"A definition of comment that should be placed before matched element when comment filter is applied."},"filter.commentAfter":{"type":"string","default":"\n<!-- /[#ID][.CLASS] -->","markdownDescription":"A definition of comment that should be placed after matched element when comment filter is applied."},"filter.commentTrigger":{"type":"array","default":["id","class"],"markdownDescription":"A comma-separated list of attribute names that should exist in the abbreviation for the comment filter to be applied."},"format.noIndentTags":{"type":"array","default":["html"],"markdownDescription":"An array of tag names that should never get inner indentation."},"format.forceIndentationForTags":{"type":"array","default":["body"],"markdownDescription":"An array of tag names that should always get inner indentation."},"profile.allowCompactBoolean":{"type":"boolean","default":false,"markdownDescription":"If `true`, compact notation of boolean attributes are produced."},"css.webkitProperties":{"type":"string","default":null,"markdownDescription":"Comma separated CSS properties that get the 'webkit' vendor prefix when used in Emmet abbreviation that starts with `-`. Set to empty string to always avoid the 'webkit' prefix."},"css.mozProperties":{"type":"string","default":null,"markdownDescription":"Comma separated CSS properties that get the 'moz' vendor prefix when used in Emmet abbreviation that starts with `-`. Set to empty string to always avoid the 'moz' prefix."},"css.oProperties":{"type":"string","default":null,"markdownDescription":"Comma separated CSS properties that get the 'o' vendor prefix when used in Emmet abbreviation that starts with `-`. Set to empty string to always avoid the 'o' prefix."},"css.msProperties":{"type":"string","default":null,"markdownDescription":"Comma separated CSS properties that get the 'ms' vendor prefix when used in Emmet abbreviation that starts with `-`. Set to empty string to always avoid the 'ms' prefix."},"css.fuzzySearchMinScore":{"type":"number","default":0.3,"markdownDescription":"The minimum score (from 0 to 1) that fuzzy-matched abbreviation should achieve. Lower values may produce many false-positive matches, higher values may reduce possible matches."},"output.inlineBreak":{"type":"number","default":0,"markdownDescription":"The number of sibling inline elements needed for line breaks to be placed between those elements. If `0`, inline elements are always expanded onto a single line."},"output.reverseAttributes":{"type":"boolean","default":false,"markdownDescription":"If `true`, reverses attribute merging directions when resolving snippets."},"output.selfClosingStyle":{"type":"string","enum":["html","xhtml","xml"],"default":"html","markdownDescription":"Style of self-closing tags: html (`<br>`), xml (`<br/>`) or xhtml (`<br />`)."},"css.color.short":{"type":"boolean","default":true,"markdownDescription":"If `true`, color values like `#f` will be expanded to `#fff` instead of `#ffffff`."}}},"emmet.showSuggestionsAsSnippets":{"type":"boolean","default":false,"markdownDescription":"If `true`, then Emmet suggestions will show up as snippets allowing you to order them as per `#editor.snippetSuggestions#` setting."},"emmet.optimizeStylesheetParsing":{"type":"boolean","default":true,"markdownDescription":"When set to `false`, the whole file is parsed to determine if current position is valid for expanding Emmet abbreviations. When set to `true`, only the content around the current position in CSS/SCSS/Less files is parsed."}}},"commands":[{"command":"editor.emmet.action.wrapWithAbbreviation","title":"Wrap with Abbreviation","category":"Emmet"},{"command":"editor.emmet.action.removeTag","title":"Remove Tag","category":"Emmet"},{"command":"editor.emmet.action.updateTag","title":"Update Tag","category":"Emmet"},{"command":"editor.emmet.action.matchTag","title":"Go to Matching Pair","category":"Emmet"},{"command":"editor.emmet.action.balanceIn","title":"Balance (inward)","category":"Emmet"},{"command":"editor.emmet.action.balanceOut","title":"Balance (outward)","category":"Emmet"},{"command":"editor.emmet.action.prevEditPoint","title":"Go to Previous Edit Point","category":"Emmet"},{"command":"editor.emmet.action.nextEditPoint","title":"Go to Next Edit Point","category":"Emmet"},{"command":"editor.emmet.action.mergeLines","title":"Merge Lines","category":"Emmet"},{"command":"editor.emmet.action.selectPrevItem","title":"Select Previous Item","category":"Emmet"},{"command":"editor.emmet.action.selectNextItem","title":"Select Next Item","category":"Emmet"},{"command":"editor.emmet.action.splitJoinTag","title":"Split/Join Tag","category":"Emmet"},{"command":"editor.emmet.action.toggleComment","title":"Toggle Comment","category":"Emmet"},{"command":"editor.emmet.action.evaluateMathExpression","title":"Evaluate Math Expression","category":"Emmet"},{"command":"editor.emmet.action.updateImageSize","title":"Update Image Size","category":"Emmet"},{"command":"editor.emmet.action.incrementNumberByOneTenth","title":"Increment by 0.1","category":"Emmet"},{"command":"editor.emmet.action.incrementNumberByOne","title":"Increment by 1","category":"Emmet"},{"command":"editor.emmet.action.incrementNumberByTen","title":"Increment by 10","category":"Emmet"},{"command":"editor.emmet.action.decrementNumberByOneTenth","title":"Decrement by 0.1","category":"Emmet"},{"command":"editor.emmet.action.decrementNumberByOne","title":"Decrement by 1","category":"Emmet"},{"command":"editor.emmet.action.decrementNumberByTen","title":"Decrement by 10","category":"Emmet"},{"command":"editor.emmet.action.reflectCSSValue","title":"Reflect CSS Value","category":"Emmet"},{"command":"workbench.action.showEmmetCommands","title":"Show Emmet Commands","category":""}],"menus":{"commandPalette":[{"command":"editor.emmet.action.wrapWithAbbreviation","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.removeTag","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.updateTag","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.matchTag","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.balanceIn","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.balanceOut","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.prevEditPoint","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.nextEditPoint","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.mergeLines","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.selectPrevItem","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.selectNextItem","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.splitJoinTag","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.toggleComment","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.evaluateMathExpression","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.updateImageSize","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.incrementNumberByOneTenth","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.incrementNumberByOne","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.incrementNumberByTen","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.decrementNumberByOneTenth","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.decrementNumberByOne","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.decrementNumberByTen","when":"!activeEditorIsReadonly"},{"command":"editor.emmet.action.reflectCSSValue","when":"!activeEditorIsReadonly"}]}},"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}}},"location":{"$mid":1,"path":"/ide/extensions/emmet","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.extension-editing"},"manifest":{"name":"extension-editing","displayName":"Extension Authoring","description":"Provides linting capabilities for authoring extensions.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.4.0"},"icon":"images/icon.png","activationEvents":["onLanguage:json","onLanguage:markdown"],"main":"./dist/extensionEditingMain","browser":"./dist/browser/extensionEditingBrowserMain","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"jsonValidation":[{"fileMatch":"package.json","url":"vscode://schemas/vscode-extensions"},{"fileMatch":"*language-configuration.json","url":"vscode://schemas/language-configuration"},{"fileMatch":["*icon-theme.json","!*product-icon-theme.json"],"url":"vscode://schemas/icon-theme"},{"fileMatch":"*product-icon-theme.json","url":"vscode://schemas/product-icon-theme"},{"fileMatch":"*color-theme.json","url":"vscode://schemas/color-theme"}],"languages":[{"id":"ignore","filenames":[".vscodeignore"]}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/extension-editing","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.fsharp"},"manifest":{"name":"fsharp","displayName":"F# Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in F# files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin ionide/ionide-fsgrammar grammars/fsharp.json ./syntaxes/fsharp.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"fsharp","extensions":[".fs",".fsi",".fsx",".fsscript"],"aliases":["F#","FSharp","fsharp"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"fsharp","scopeName":"source.fsharp","path":"./syntaxes/fsharp.tmLanguage.json"}],"snippets":[{"language":"fsharp","path":"./snippets/fsharp.code-snippets"}],"configurationDefaults":{"[fsharp]":{"diffEditor.ignoreTrimWhitespace":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/fsharp","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.git"},"manifest":{"name":"git","displayName":"Git","description":"Git SCM Integration","publisher":"vscode","license":"MIT","version":"1.0.0","engines":{"vscode":"^1.5.0"},"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","enabledApiProposals":["canonicalUriProvider","contribEditSessions","contribEditorContentMenu","contribMergeEditorMenus","contribMultiDiffEditorMenus","contribDiffEditorGutterToolBarMenus","contribSourceControlHistoryItemMenu","contribSourceControlHistoryTitleMenu","contribSourceControlInputBoxMenu","contribSourceControlTitleMenu","contribViewsWelcome","diffCommand","editSessionIdentityProvider","quickDiffProvider","quickInputButtonLocation","quickPickSortByLabel","scmActionButton","scmHistoryProvider","scmMultiDiffEditor","scmSelectedProvider","scmTextDocument","scmValidation","statusBarItemTooltip","tabInputMultiDiff","tabInputTextMerge","textEditorDiffInformation","timeline"],"categories":["Other"],"activationEvents":["*","onEditSession:file","onFileSystem:git","onFileSystem:git-show"],"extensionDependencies":["vscode.git-base"],"main":"./dist/main","icon":"resources/icons/git.png","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":false}},"contributes":{"commands":[{"command":"git.continueInLocalClone","title":"Clone Repository Locally and Open on Desktop...","category":"Git","icon":"$(repo-clone)","enablement":"false && config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && remoteName"},{"command":"git.clone","title":"Clone","category":"Git","enablement":"!operationInProgress"},{"command":"git.cloneRecursive","title":"Clone (Recursive)","category":"Git","enablement":"!operationInProgress"},{"command":"git.init","title":"Initialize Repository","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.openRepository","title":"Open Repository","category":"Git","enablement":"!operationInProgress"},{"command":"git.reopenClosedRepositories","title":"Reopen Closed Repositories...","icon":"$(repo)","category":"Git","enablement":"!operationInProgress && git.closedRepositoryCount != 0"},{"command":"git.close","title":"Close Repository","category":"Git","enablement":"!operationInProgress"},{"command":"git.closeOtherRepositories","title":"Close Other Repositories","category":"Git","enablement":"!operationInProgress"},{"command":"git.refresh","title":"Refresh","category":"Git","icon":"$(refresh)","enablement":"!operationInProgress"},{"command":"git.openChange","title":"Open Changes","category":"Git","icon":"$(compare-changes)"},{"command":"git.openAllChanges","title":"Open All Changes","category":"Git"},{"command":"git.openFile","title":"Open File","category":"Git","icon":"$(go-to-file)"},{"command":"git.openFile2","title":"Open File","category":"Git","icon":"$(go-to-file)"},{"command":"git.openHEADFile","title":"Open File (HEAD)","category":"Git"},{"command":"git.stage","title":"Stage Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageAll","title":"Stage All Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageAllTracked","title":"Stage All Tracked Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageAllUntracked","title":"Stage All Untracked Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageAllMerge","title":"Stage All Merge Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageSelectedRanges","title":"Stage Selected Ranges","category":"Git","enablement":"!operationInProgress"},{"command":"git.diff.stageHunk","title":"Stage Block","category":"Git","icon":"$(plus)"},{"command":"git.diff.stageSelection","title":"Stage Selection","category":"Git","icon":"$(plus)"},{"command":"git.revertSelectedRanges","title":"Revert Selected Ranges","category":"Git","enablement":"!operationInProgress"},{"command":"git.stageChange","title":"Stage Change","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.stageFile","title":"Stage Changes","category":"Git","icon":"$(add)","enablement":"!operationInProgress"},{"command":"git.revertChange","title":"Revert Change","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.unstage","title":"Unstage Changes","category":"Git","icon":"$(remove)","enablement":"!operationInProgress"},{"command":"git.unstageAll","title":"Unstage All Changes","category":"Git","icon":"$(remove)","enablement":"!operationInProgress"},{"command":"git.unstageSelectedRanges","title":"Unstage Selected Ranges","category":"Git","enablement":"!operationInProgress"},{"command":"git.unstageFile","title":"Unstage Changes","category":"Git","icon":"$(remove)","enablement":"!operationInProgress"},{"command":"git.clean","title":"Discard Changes","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.cleanAll","title":"Discard All Changes","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.cleanAllTracked","title":"Discard All Tracked Changes","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.cleanAllUntracked","title":"Discard All Untracked Changes","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.rename","title":"Rename","category":"Git","icon":"$(discard)","enablement":"!operationInProgress"},{"command":"git.commit","title":"Commit","category":"Git","icon":"$(check)","enablement":"!operationInProgress"},{"command":"git.commitAmend","title":"Commit (Amend)","category":"Git","icon":"$(check)","enablement":"!operationInProgress"},{"command":"git.commitSigned","title":"Commit (Signed Off)","category":"Git","icon":"$(check)","enablement":"!operationInProgress"},{"command":"git.commitStaged","title":"Commit Staged","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitEmpty","title":"Commit Empty","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitStagedSigned","title":"Commit Staged (Signed Off)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitStagedAmend","title":"Commit Staged (Amend)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAll","title":"Commit All","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAllSigned","title":"Commit All (Signed Off)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAllAmend","title":"Commit All (Amend)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitNoVerify","title":"Commit (No Verify)","category":"Git","icon":"$(check)","enablement":"!operationInProgress"},{"command":"git.commitStagedNoVerify","title":"Commit Staged (No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitEmptyNoVerify","title":"Commit Empty (No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitStagedSignedNoVerify","title":"Commit Staged (Signed Off, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAmendNoVerify","title":"Commit (Amend, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitSignedNoVerify","title":"Commit (Signed Off, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitStagedAmendNoVerify","title":"Commit Staged (Amend, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAllNoVerify","title":"Commit All (No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAllSignedNoVerify","title":"Commit All (Signed Off, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitAllAmendNoVerify","title":"Commit All (Amend, No Verify)","category":"Git","enablement":"!operationInProgress"},{"command":"git.commitMessageAccept","title":"Accept Commit Message","icon":"$(check)","category":"Git"},{"command":"git.commitMessageDiscard","title":"Discard Commit Message","icon":"$(discard)","category":"Git"},{"command":"git.restoreCommitTemplate","title":"Restore Commit Template","category":"Git","enablement":"!operationInProgress"},{"command":"git.undoCommit","title":"Undo Last Commit","category":"Git","enablement":"!operationInProgress"},{"command":"git.checkout","title":"Checkout to...","category":"Git","enablement":"!operationInProgress"},{"command":"git.graph.checkout","title":"Checkout","category":"Git","enablement":"!operationInProgress"},{"command":"git.checkoutDetached","title":"Checkout to (Detached)...","category":"Git","enablement":"!operationInProgress"},{"command":"git.graph.checkoutDetached","title":"Checkout (Detached)","category":"Git","enablement":"!operationInProgress"},{"command":"git.branch","title":"Create Branch...","category":"Git","enablement":"!operationInProgress"},{"command":"git.branchFrom","title":"Create Branch From...","category":"Git","enablement":"!operationInProgress"},{"command":"git.deleteBranch","title":"Delete Branch...","category":"Git","enablement":"!operationInProgress"},{"command":"git.graph.deleteBranch","title":"Delete Branch","category":"Git","enablement":"!operationInProgress"},{"command":"git.renameBranch","title":"Rename Branch...","category":"Git","enablement":"!operationInProgress"},{"command":"git.merge","title":"Merge...","category":"Git","enablement":"!operationInProgress"},{"command":"git.mergeAbort","title":"Abort Merge","category":"Git","enablement":"gitMergeInProgress"},{"command":"git.rebase","title":"Rebase Branch...","category":"Git","enablement":"!operationInProgress"},{"command":"git.createTag","title":"Create Tag...","category":"Git","enablement":"!operationInProgress"},{"command":"git.deleteTag","title":"Delete Tag...","category":"Git","enablement":"!operationInProgress"},{"command":"git.graph.deleteTag","title":"Delete Tag","category":"Git","enablement":"!operationInProgress"},{"command":"git.deleteRemoteTag","title":"Delete Remote Tag...","category":"Git","enablement":"!operationInProgress"},{"command":"git.fetch","title":"Fetch","category":"Git","enablement":"!operationInProgress"},{"command":"git.fetchPrune","title":"Fetch (Prune)","category":"Git","enablement":"!operationInProgress"},{"command":"git.fetchAll","title":"Fetch From All Remotes","icon":"$(git-fetch)","category":"Git","enablement":"!operationInProgress"},{"command":"git.fetchRef","title":"Fetch","icon":"$(git-fetch)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pull","title":"Pull","category":"Git","enablement":"!operationInProgress"},{"command":"git.pullRebase","title":"Pull (Rebase)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pullFrom","title":"Pull from...","category":"Git","enablement":"!operationInProgress"},{"command":"git.pullRef","title":"Pull","icon":"$(repo-pull)","category":"Git","enablement":"!operationInProgress && scmCurrentHistoryItemRefInFilter && scmCurrentHistoryItemRefHasRemote"},{"command":"git.push","title":"Push","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushForce","title":"Push (Force)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushTo","title":"Push to...","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushToForce","title":"Push to... (Force)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushTags","title":"Push Tags","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushWithTags","title":"Push (Follow Tags)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushWithTagsForce","title":"Push (Follow Tags, Force)","category":"Git","enablement":"!operationInProgress"},{"command":"git.pushRef","title":"Push","icon":"$(repo-push)","category":"Git","enablement":"!operationInProgress && scmCurrentHistoryItemRefInFilter && scmCurrentHistoryItemRefHasRemote"},{"command":"git.cherryPick","title":"Cherry Pick...","category":"Git","enablement":"!operationInProgress"},{"command":"git.graph.cherryPick","title":"Cherry Pick","category":"Git","enablement":"!operationInProgress"},{"command":"git.cherryPickAbort","title":"Abort Cherry Pick","category":"Git","enablement":"!operationInProgress"},{"command":"git.addRemote","title":"Add Remote...","category":"Git","enablement":"!operationInProgress"},{"command":"git.removeRemote","title":"Remove Remote","category":"Git","enablement":"!operationInProgress"},{"command":"git.sync","title":"Sync","category":"Git","enablement":"!operationInProgress"},{"command":"git.syncRebase","title":"Sync (Rebase)","category":"Git","enablement":"!operationInProgress"},{"command":"git.publish","title":"Publish Branch...","category":"Git","icon":"$(cloud-upload)","enablement":"!operationInProgress"},{"command":"git.showOutput","title":"Show Git Output","category":"Git"},{"command":"git.ignore","title":"Add to .gitignore","category":"Git","enablement":"!operationInProgress"},{"command":"git.revealInExplorer","title":"Reveal in Explorer View","category":"Git"},{"command":"git.revealFileInOS.linux","title":"Open Containing Folder","category":"Git"},{"command":"git.revealFileInOS.mac","title":"Reveal in Finder","category":"Git"},{"command":"git.revealFileInOS.windows","title":"Reveal in File Explorer","category":"Git"},{"command":"git.stashIncludeUntracked","title":"Stash (Include Untracked)","category":"Git","enablement":"!operationInProgress"},{"command":"git.stash","title":"Stash","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashStaged","title":"Stash Staged","category":"Git","enablement":"!operationInProgress && gitVersion2.35"},{"command":"git.stashPop","title":"Pop Stash...","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashPopLatest","title":"Pop Latest Stash","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashPopEditor","title":"Pop Stash","icon":"$(git-stash-pop)","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashApply","title":"Apply Stash...","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashApplyLatest","title":"Apply Latest Stash","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashApplyEditor","title":"Apply Stash","icon":"$(git-stash-apply)","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashDrop","title":"Drop Stash...","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashDropAll","title":"Drop All Stashes...","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashDropEditor","title":"Drop Stash","icon":"$(trash)","category":"Git","enablement":"!operationInProgress"},{"command":"git.stashView","title":"View Stash...","category":"Git","enablement":"!operationInProgress"},{"command":"git.timeline.openDiff","title":"Open Changes","icon":"$(compare-changes)","category":"Git"},{"command":"git.timeline.copyCommitId","title":"Copy Commit ID","category":"Git"},{"command":"git.timeline.copyCommitMessage","title":"Copy Commit Message","category":"Git"},{"command":"git.timeline.selectForCompare","title":"Select for Compare","category":"Git"},{"command":"git.timeline.compareWithSelected","title":"Compare with Selected","category":"Git"},{"command":"git.timeline.viewCommit","title":"Open Commit","icon":"$(diff-multiple)","category":"Git"},{"command":"git.rebaseAbort","title":"Abort Rebase","category":"Git","enablement":"gitRebaseInProgress"},{"command":"git.closeAllDiffEditors","title":"Close All Diff Editors","category":"Git","enablement":"!operationInProgress"},{"command":"git.closeAllUnmodifiedEditors","title":"Close All Unmodified Editors","category":"Git","enablement":"!operationInProgress"},{"command":"git.api.getRepositories","title":"Get Repositories","category":"Git API"},{"command":"git.api.getRepositoryState","title":"Get Repository State","category":"Git API"},{"command":"git.api.getRemoteSources","title":"Get Remote Sources","category":"Git API"},{"command":"git.acceptMerge","title":"Complete Merge","category":"Git","enablement":"isMergeEditor && mergeEditorResultUri in git.mergeChanges"},{"command":"git.openMergeEditor","title":"Resolve in Merge Editor","category":"Git"},{"command":"git.runGitMerge","title":"Compute Conflicts With Git","category":"Git","enablement":"isMergeEditor"},{"command":"git.runGitMergeDiff3","title":"Compute Conflicts With Git (Diff3)","category":"Git","enablement":"isMergeEditor"},{"command":"git.manageUnsafeRepositories","title":"Manage Unsafe Repositories","category":"Git"},{"command":"git.openRepositoriesInParentFolders","title":"Open Repositories In Parent Folders","category":"Git"},{"command":"git.viewChanges","title":"Open Changes","icon":"$(diff-multiple)","category":"Git","enablement":"!operationInProgress"},{"command":"git.viewStagedChanges","title":"Open Staged Changes","icon":"$(diff-multiple)","category":"Git","enablement":"!operationInProgress"},{"command":"git.viewUntrackedChanges","title":"Open Untracked Changes","icon":"$(diff-multiple)","category":"Git","enablement":"!operationInProgress"},{"command":"git.viewCommit","title":"Open Commit","icon":"$(diff-multiple)","category":"Git","enablement":"!operationInProgress"},{"command":"git.copyCommitId","title":"Copy Commit ID","category":"Git"},{"command":"git.copyCommitMessage","title":"Copy Commit Message","category":"Git"},{"command":"git.blame.toggleEditorDecoration","title":"Toggle Git Blame Editor Decoration","category":"Git"},{"command":"git.blame.toggleStatusBarItem","title":"Toggle Git Blame Status Bar Item","category":"Git"}],"continueEditSession":[{"command":"git.continueInLocalClone","qualifiedName":"Continue Working in New Local Clone","when":"false && config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && remoteName","remoteGroup":"remote_42_git_0_local@0"}],"keybindings":[{"command":"git.stageSelectedRanges","key":"ctrl+k ctrl+alt+s","mac":"cmd+k cmd+alt+s","when":"editorTextFocus && resourceScheme == file"},{"command":"git.unstageSelectedRanges","key":"ctrl+k ctrl+n","mac":"cmd+k cmd+n","when":"editorTextFocus && isInDiffEditor && isInDiffRightEditor && resourceScheme == git"},{"command":"git.revertSelectedRanges","key":"ctrl+k ctrl+r","mac":"cmd+k cmd+r","when":"editorTextFocus && resourceScheme == file"}],"menus":{"commandPalette":[{"command":"git.continueInLocalClone","when":"false"},{"command":"git.clone","when":"config.git.enabled && !git.missing"},{"command":"git.cloneRecursive","when":"config.git.enabled && !git.missing"},{"command":"git.init","when":"config.git.enabled && !git.missing && remoteName != 'codespaces'"},{"command":"git.openRepository","when":"config.git.enabled && !git.missing"},{"command":"git.close","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.closeOtherRepositories","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount > 1"},{"command":"git.refresh","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.openFile","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file && scmActiveResourceHasChanges"},{"command":"git.openHEADFile","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file && scmActiveResourceHasChanges"},{"command":"git.openChange","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stage","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stageAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stageAllTracked","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stageAllUntracked","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stageAllMerge","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stageSelectedRanges","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file"},{"command":"git.stageChange","when":"false"},{"command":"git.revertSelectedRanges","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file"},{"command":"git.revertChange","when":"false"},{"command":"git.openFile2","when":"false"},{"command":"git.unstage","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.unstageAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.unstageSelectedRanges","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == git"},{"command":"git.clean","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.cleanAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.cleanAllTracked","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.cleanAllUntracked","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.rename","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file && scmActiveResourceRepository"},{"command":"git.commit","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitAmend","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitSigned","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitStaged","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitEmpty","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitStagedSigned","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitStagedAmend","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitAllSigned","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.commitAllAmend","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.rebaseAbort","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && gitRebaseInProgress"},{"command":"git.commitNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitStagedNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitEmptyNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitStagedSignedNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitAmendNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitSignedNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitStagedAmendNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitAllNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitAllSignedNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.commitAllAmendNoVerify","when":"config.git.enabled && !git.missing && config.git.allowNoVerifyCommit && gitOpenRepositoryCount != 0"},{"command":"git.restoreCommitTemplate","when":"false"},{"command":"git.commitMessageAccept","when":"false"},{"command":"git.commitMessageDiscard","when":"false"},{"command":"git.revealInExplorer","when":"false"},{"command":"git.revealFileInOS.linux","when":"false"},{"command":"git.revealFileInOS.mac","when":"false"},{"command":"git.revealFileInOS.windows","when":"false"},{"command":"git.undoCommit","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.checkout","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.branch","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.branchFrom","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.deleteBranch","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.renameBranch","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.cherryPick","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.cherryPickAbort","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && gitCherryPickInProgress"},{"command":"git.pull","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.pullFrom","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.pullRebase","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.merge","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.mergeAbort","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && gitMergeInProgress"},{"command":"git.rebase","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.createTag","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.deleteTag","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.deleteRemoteTag","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.fetch","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.fetchPrune","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.fetchAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.push","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.pushForce","when":"config.git.enabled && !git.missing && config.git.allowForcePush && gitOpenRepositoryCount != 0"},{"command":"git.pushTo","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.pushToForce","when":"config.git.enabled && !git.missing && config.git.allowForcePush && gitOpenRepositoryCount != 0"},{"command":"git.pushWithTags","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.pushWithTagsForce","when":"config.git.enabled && !git.missing && config.git.allowForcePush && gitOpenRepositoryCount != 0"},{"command":"git.pushTags","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.addRemote","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.removeRemote","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.sync","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.syncRebase","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.publish","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.showOutput","when":"config.git.enabled"},{"command":"git.ignore","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && resourceScheme == file && scmActiveResourceRepository"},{"command":"git.stashIncludeUntracked","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stash","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashStaged","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && gitVersion2.35"},{"command":"git.stashPop","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashPopLatest","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashPopEditor","when":"false"},{"command":"git.stashApply","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashApplyLatest","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashApplyEditor","when":"false"},{"command":"git.stashDrop","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashDropAll","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.stashDropEditor","when":"false"},{"command":"git.timeline.openDiff","when":"false"},{"command":"git.timeline.copyCommitId","when":"false"},{"command":"git.timeline.copyCommitMessage","when":"false"},{"command":"git.timeline.selectForCompare","when":"false"},{"command":"git.timeline.compareWithSelected","when":"false"},{"command":"git.timeline.viewCommit","when":"false"},{"command":"git.closeAllDiffEditors","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0"},{"command":"git.api.getRepositories","when":"false"},{"command":"git.api.getRepositoryState","when":"false"},{"command":"git.api.getRemoteSources","when":"false"},{"command":"git.openMergeEditor","when":"false"},{"command":"git.manageUnsafeRepositories","when":"config.git.enabled && !git.missing && git.unsafeRepositoryCount != 0"},{"command":"git.openRepositoriesInParentFolders","when":"config.git.enabled && !git.missing && git.parentRepositoryCount != 0"},{"command":"git.stashView","when":"config.git.enabled && !git.missing && config.multiDiffEditor.experimental.enabled"},{"command":"git.viewChanges","when":"config.git.enabled && !git.missing && config.multiDiffEditor.experimental.enabled"},{"command":"git.viewStagedChanges","when":"config.git.enabled && !git.missing && config.multiDiffEditor.experimental.enabled"},{"command":"git.viewUntrackedChanges","when":"config.git.enabled && !git.missing && config.multiDiffEditor.experimental.enabled && config.git.untrackedChanges == separate"},{"command":"git.viewCommit","when":"false"},{"command":"git.stageFile","when":"false"},{"command":"git.unstageFile","when":"false"},{"command":"git.fetchRef","when":"false"},{"command":"git.pullRef","when":"false"},{"command":"git.pushRef","when":"false"},{"command":"git.copyCommitId","when":"false"},{"command":"git.copyCommitMessage","when":"false"},{"command":"git.graph.checkout","when":"false"},{"command":"git.graph.checkoutDetached","when":"false"},{"command":"git.graph.deleteBranch","when":"false"},{"command":"git.graph.deleteTag","when":"false"},{"command":"git.graph.cherryPick","when":"false"},{"command":"git.diff.stageHunk","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && diffEditorOriginalUri =~ /^git\\:.*%22ref%22%3A%22~%22%7D$/"},{"command":"git.diff.stageSelection","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && diffEditorOriginalUri =~ /^git\\:.*%22ref%22%3A%22~%22%7D$/"}],"scm/title":[{"command":"git.commit","group":"navigation","when":"scmProvider == git"},{"command":"git.refresh","group":"navigation","when":"scmProvider == git"},{"command":"git.pull","group":"1_header@1","when":"scmProvider == git"},{"command":"git.push","group":"1_header@2","when":"scmProvider == git"},{"command":"git.clone","group":"1_header@3","when":"scmProvider == git"},{"command":"git.checkout","group":"1_header@4","when":"scmProvider == git"},{"command":"git.fetch","group":"1_header@5","when":"scmProvider == git"},{"submenu":"git.commit","group":"2_main@1","when":"scmProvider == git"},{"submenu":"git.changes","group":"2_main@2","when":"scmProvider == git"},{"submenu":"git.pullpush","group":"2_main@3","when":"scmProvider == git"},{"submenu":"git.branch","group":"2_main@4","when":"scmProvider == git"},{"submenu":"git.remotes","group":"2_main@5","when":"scmProvider == git"},{"submenu":"git.stash","group":"2_main@6","when":"scmProvider == git"},{"submenu":"git.tags","group":"2_main@7","when":"scmProvider == git"},{"command":"git.showOutput","group":"3_footer","when":"scmProvider == git"}],"scm/sourceControl/title":[{"command":"git.reopenClosedRepositories","group":"navigation@1","when":"git.closedRepositoryCount > 0"}],"scm/sourceControl":[{"command":"git.close","group":"navigation@1","when":"scmProvider == git"},{"command":"git.closeOtherRepositories","group":"navigation@2","when":"scmProvider == git && gitOpenRepositoryCount > 1"}],"scm/resourceGroup/context":[{"command":"git.stageAllMerge","when":"scmProvider == git && scmResourceGroup == merge","group":"1_modification"},{"command":"git.stageAllMerge","when":"scmProvider == git && scmResourceGroup == merge","group":"inline@2"},{"command":"git.unstageAll","when":"scmProvider == git && scmResourceGroup == index","group":"1_modification"},{"command":"git.unstageAll","when":"scmProvider == git && scmResourceGroup == index","group":"inline@2"},{"command":"git.viewStagedChanges","when":"scmProvider == git && scmResourceGroup == index && config.multiDiffEditor.experimental.enabled","group":"inline@1"},{"command":"git.viewChanges","when":"scmProvider == git && scmResourceGroup == workingTree && config.multiDiffEditor.experimental.enabled","group":"inline@1"},{"command":"git.cleanAll","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges == mixed","group":"1_modification"},{"command":"git.stageAll","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges == mixed","group":"1_modification"},{"command":"git.cleanAll","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges == mixed","group":"inline@2"},{"command":"git.stageAll","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges == mixed","group":"inline@2"},{"command":"git.cleanAllTracked","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges != mixed","group":"1_modification"},{"command":"git.stageAllTracked","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges != mixed","group":"1_modification"},{"command":"git.cleanAllTracked","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges != mixed","group":"inline@2"},{"command":"git.stageAllTracked","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.untrackedChanges != mixed","group":"inline@2"},{"command":"git.cleanAllUntracked","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification"},{"command":"git.stageAllUntracked","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification"},{"command":"git.viewUntrackedChanges","when":"scmProvider == git && scmResourceGroup == untracked && config.multiDiffEditor.experimental.enabled","group":"inline@1"},{"command":"git.cleanAllUntracked","when":"scmProvider == git && scmResourceGroup == untracked","group":"inline@2"},{"command":"git.stageAllUntracked","when":"scmProvider == git && scmResourceGroup == untracked","group":"inline@2"}],"scm/resourceFolder/context":[{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == merge","group":"1_modification"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == merge","group":"inline@2"},{"command":"git.unstage","when":"scmProvider == git && scmResourceGroup == index","group":"1_modification"},{"command":"git.unstage","when":"scmProvider == git && scmResourceGroup == index","group":"inline@2"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == workingTree","group":"inline@2"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == workingTree","group":"inline@2"},{"command":"git.ignore","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification@3"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == untracked","group":"inline@2"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == untracked","group":"inline@2"},{"command":"git.ignore","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification@3"}],"scm/resourceState/context":[{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == merge","group":"1_modification"},{"command":"git.openFile","when":"scmProvider == git && scmResourceGroup == merge","group":"navigation"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == merge","group":"inline@2"},{"command":"git.revealFileInOS.linux","when":"scmProvider == git && scmResourceGroup == merge && remoteName == '' && isLinux","group":"2_view@1"},{"command":"git.revealFileInOS.mac","when":"scmProvider == git && scmResourceGroup == merge && remoteName == '' && isMac","group":"2_view@1"},{"command":"git.revealFileInOS.windows","when":"scmProvider == git && scmResourceGroup == merge && remoteName == '' && isWindows","group":"2_view@1"},{"command":"git.revealInExplorer","when":"scmProvider == git && scmResourceGroup == merge","group":"2_view@2"},{"command":"git.openFile2","when":"scmProvider == git && scmResourceGroup == merge && config.git.showInlineOpenFileAction && config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == merge && config.git.showInlineOpenFileAction && !config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == index","group":"navigation"},{"command":"git.openFile","when":"scmProvider == git && scmResourceGroup == index","group":"navigation"},{"command":"git.openHEADFile","when":"scmProvider == git && scmResourceGroup == index","group":"navigation"},{"command":"git.unstage","when":"scmProvider == git && scmResourceGroup == index","group":"1_modification"},{"command":"git.unstage","when":"scmProvider == git && scmResourceGroup == index","group":"inline@2"},{"command":"git.revealFileInOS.linux","when":"scmProvider == git && scmResourceGroup == index && remoteName == '' && isLinux","group":"2_view@1"},{"command":"git.revealFileInOS.mac","when":"scmProvider == git && scmResourceGroup == index && remoteName == '' && isMac","group":"2_view@1"},{"command":"git.revealFileInOS.windows","when":"scmProvider == git && scmResourceGroup == index && remoteName == '' && isWindows","group":"2_view@1"},{"command":"git.revealInExplorer","when":"scmProvider == git && scmResourceGroup == index","group":"2_view@2"},{"command":"git.openFile2","when":"scmProvider == git && scmResourceGroup == index && config.git.showInlineOpenFileAction && config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == index && config.git.showInlineOpenFileAction && !config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == workingTree","group":"navigation"},{"command":"git.openHEADFile","when":"scmProvider == git && scmResourceGroup == workingTree","group":"navigation"},{"command":"git.openFile","when":"scmProvider == git && scmResourceGroup == workingTree","group":"navigation"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == workingTree","group":"inline@2"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == workingTree","group":"inline@2"},{"command":"git.openFile2","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.showInlineOpenFileAction && config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == workingTree && config.git.showInlineOpenFileAction && !config.git.openDiffOnClick","group":"inline@1"},{"command":"git.ignore","when":"scmProvider == git && scmResourceGroup == workingTree","group":"1_modification@3"},{"command":"git.revealFileInOS.linux","when":"scmProvider == git && scmResourceGroup == workingTree && remoteName == '' && isLinux","group":"2_view@1"},{"command":"git.revealFileInOS.mac","when":"scmProvider == git && scmResourceGroup == workingTree && remoteName == '' && isMac","group":"2_view@1"},{"command":"git.revealFileInOS.windows","when":"scmProvider == git && scmResourceGroup == workingTree && remoteName == '' && isWindows","group":"2_view@1"},{"command":"git.revealInExplorer","when":"scmProvider == git && scmResourceGroup == workingTree","group":"2_view@2"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == untracked","group":"navigation"},{"command":"git.openHEADFile","when":"scmProvider == git && scmResourceGroup == untracked","group":"navigation"},{"command":"git.openFile","when":"scmProvider == git && scmResourceGroup == untracked","group":"navigation"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == untracked && !gitFreshRepository","group":"1_modification"},{"command":"git.clean","when":"scmProvider == git && scmResourceGroup == untracked && !gitFreshRepository","group":"inline@2"},{"command":"git.stage","when":"scmProvider == git && scmResourceGroup == untracked","group":"inline@2"},{"command":"git.openFile2","when":"scmProvider == git && scmResourceGroup == untracked && config.git.showInlineOpenFileAction && config.git.openDiffOnClick","group":"inline@1"},{"command":"git.openChange","when":"scmProvider == git && scmResourceGroup == untracked && config.git.showInlineOpenFileAction && !config.git.openDiffOnClick","group":"inline@1"},{"command":"git.ignore","when":"scmProvider == git && scmResourceGroup == untracked","group":"1_modification@3"}],"scm/history/title":[{"command":"git.fetchAll","group":"navigation@900","when":"scmProvider == git"},{"command":"git.pullRef","group":"navigation@901","when":"scmProvider == git"},{"command":"git.pushRef","when":"scmProvider == git && scmCurrentHistoryItemRefHasRemote","group":"navigation@902"},{"command":"git.publish","when":"scmProvider == git && !scmCurrentHistoryItemRefHasRemote","group":"navigation@903"}],"scm/historyItem/context":[{"command":"git.graph.checkoutDetached","when":"scmProvider == git","group":"1_checkout@2"},{"command":"git.branch","when":"scmProvider == git","group":"2_branch@2"},{"command":"git.createTag","when":"scmProvider == git","group":"3_tag@1"},{"command":"git.graph.cherryPick","when":"scmProvider == git","group":"4_modify@1"},{"command":"git.copyCommitId","when":"scmProvider == git && !listMultiSelection","group":"9_copy@1"},{"command":"git.copyCommitMessage","when":"scmProvider == git && !listMultiSelection","group":"9_copy@2"}],"scm/historyItemRef/context":[{"command":"git.graph.checkout","when":"scmProvider == git","group":"1_checkout@1"},{"command":"git.graph.deleteBranch","when":"scmProvider == git && scmHistoryItemRef =~ /^refs\\/heads\\//","group":"2_branch@2"},{"command":"git.graph.deleteTag","when":"scmProvider == git && scmHistoryItemRef =~ /^refs\\/tags\\//","group":"3_tag@2"}],"editor/title":[{"command":"git.openFile","group":"navigation","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && resourceScheme =~ /^git$|^file$/"},{"command":"git.openFile","group":"navigation","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInNotebookTextDiffEditor && resourceScheme =~ /^git$|^file$/"},{"command":"git.openChange","group":"navigation@2","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && !isInDiffEditor && !isMergeEditor && resourceScheme == file && scmActiveResourceHasChanges"},{"command":"git.commitMessageAccept","group":"navigation","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && editorLangId == git-commit"},{"command":"git.commitMessageDiscard","group":"navigation","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && editorLangId == git-commit"},{"command":"git.stashApplyEditor","alt":"git.stashPopEditor","group":"navigation@1","when":"config.git.enabled && !git.missing && resourceScheme == git-stash"},{"command":"git.stashDropEditor","group":"navigation@2","when":"config.git.enabled && !git.missing && resourceScheme == git-stash"},{"command":"git.stage","group":"2_git@1","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && !isInDiffEditor && !isMergeEditor && resourceScheme == file && git.activeResourceHasUnstagedChanges"},{"command":"git.unstage","group":"2_git@2","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && !isInDiffEditor && !isMergeEditor && resourceScheme == file && git.activeResourceHasStagedChanges"},{"command":"git.stage","group":"2_git@1","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == file"},{"command":"git.stageSelectedRanges","group":"2_git@2","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == file"},{"command":"git.unstage","group":"2_git@3","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == git"},{"command":"git.unstageSelectedRanges","group":"2_git@4","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == git"},{"command":"git.revertSelectedRanges","group":"2_git@5","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == file"}],"editor/context":[{"command":"git.stageSelectedRanges","group":"2_git@1","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == file"},{"command":"git.unstageSelectedRanges","group":"2_git@2","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == git"},{"command":"git.revertSelectedRanges","group":"2_git@3","when":"config.git.enabled && !git.missing && gitOpenRepositoryCount != 0 && isInDiffEditor && isInDiffRightEditor && !isEmbeddedDiffEditor && resourceScheme == file"}],"editor/content":[{"command":"git.acceptMerge","when":"isMergeResultEditor && mergeEditorBaseUri =~ /^(git|file):/ && mergeEditorResultUri in git.mergeChanges"},{"command":"git.openMergeEditor","group":"navigation@-10","when":"config.git.enabled && !git.missing && !isInDiffEditor && !isMergeEditor && resource in git.mergeChanges"}],"multiDiffEditor/resource/title":[{"command":"git.stageFile","group":"navigation","when":"scmProvider == git && scmResourceGroup == workingTree"},{"command":"git.stageFile","group":"navigation","when":"scmProvider == git && scmResourceGroup == untracked"},{"command":"git.unstageFile","group":"navigation","when":"scmProvider == git && scmResourceGroup == index"}],"diffEditor/gutter/hunk":[{"command":"git.diff.stageHunk","group":"primary@10","when":"diffEditorOriginalUri =~ /^git\\:.*%22ref%22%3A%22~%22%7D$/"}],"diffEditor/gutter/selection":[{"command":"git.diff.stageSelection","group":"primary@10","when":"diffEditorOriginalUri =~ /^git\\:.*%22ref%22%3A%22~%22%7D$/"}],"scm/change/title":[{"command":"git.stageChange","when":"config.git.enabled && !git.missing && originalResourceScheme == git"},{"command":"git.revertChange","when":"config.git.enabled && !git.missing && originalResourceScheme == git"}],"timeline/item/context":[{"command":"git.timeline.viewCommit","group":"inline","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file:commit\\b/ && !listMultiSelection && config.multiDiffEditor.experimental.enabled"},{"command":"git.timeline.openDiff","group":"1_actions@1","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file\\b/ && !listMultiSelection"},{"command":"git.timeline.viewCommit","group":"1_actions@2","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file:commit\\b/ && !listMultiSelection && config.multiDiffEditor.experimental.enabled"},{"command":"git.timeline.compareWithSelected","group":"3_compare@1","when":"config.git.enabled && !git.missing && git.timeline.selectedForCompare && timelineItem =~ /git:file\\b/ && !listMultiSelection"},{"command":"git.timeline.selectForCompare","group":"3_compare@2","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file\\b/ && !listMultiSelection"},{"command":"git.timeline.copyCommitId","group":"5_copy@1","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file:commit\\b/ && !listMultiSelection"},{"command":"git.timeline.copyCommitMessage","group":"5_copy@2","when":"config.git.enabled && !git.missing && timelineItem =~ /git:file:commit\\b/ && !listMultiSelection"}],"git.commit":[{"command":"git.commit","group":"1_commit@1"},{"command":"git.commitStaged","group":"1_commit@2"},{"command":"git.commitAll","group":"1_commit@3"},{"command":"git.undoCommit","group":"1_commit@4"},{"command":"git.rebaseAbort","group":"1_commit@5"},{"command":"git.commitNoVerify","group":"2_commit_noverify@1","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitStagedNoVerify","group":"2_commit_noverify@2","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitAllNoVerify","group":"2_commit_noverify@3","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitAmend","group":"3_amend@1"},{"command":"git.commitStagedAmend","group":"3_amend@2"},{"command":"git.commitAllAmend","group":"3_amend@3"},{"command":"git.commitAmendNoVerify","group":"4_amend_noverify@1","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitStagedAmendNoVerify","group":"4_amend_noverify@2","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitAllAmendNoVerify","group":"4_amend_noverify@3","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitSigned","group":"5_signoff@1"},{"command":"git.commitStagedSigned","group":"5_signoff@2"},{"command":"git.commitAllSigned","group":"5_signoff@3"},{"command":"git.commitSignedNoVerify","group":"6_signoff_noverify@1","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitStagedSignedNoVerify","group":"6_signoff_noverify@2","when":"config.git.allowNoVerifyCommit"},{"command":"git.commitAllSignedNoVerify","group":"6_signoff_noverify@3","when":"config.git.allowNoVerifyCommit"}],"git.changes":[{"command":"git.stageAll","group":"changes@1"},{"command":"git.unstageAll","group":"changes@2"},{"command":"git.cleanAll","group":"changes@3"}],"git.pullpush":[{"command":"git.sync","group":"1_sync@1"},{"command":"git.syncRebase","when":"gitState == idle","group":"1_sync@2"},{"command":"git.pull","group":"2_pull@1"},{"command":"git.pullRebase","group":"2_pull@2"},{"command":"git.pullFrom","group":"2_pull@3"},{"command":"git.push","group":"3_push@1"},{"command":"git.pushForce","when":"config.git.allowForcePush","group":"3_push@2"},{"command":"git.pushTo","group":"3_push@3"},{"command":"git.pushToForce","when":"config.git.allowForcePush","group":"3_push@4"},{"command":"git.fetch","group":"4_fetch@1"},{"command":"git.fetchPrune","group":"4_fetch@2"},{"command":"git.fetchAll","group":"4_fetch@3"}],"git.branch":[{"command":"git.merge","group":"1_merge@1"},{"command":"git.rebase","group":"1_merge@2"},{"command":"git.branch","group":"2_branch@1"},{"command":"git.branchFrom","group":"2_branch@2"},{"command":"git.renameBranch","group":"3_modify@1"},{"command":"git.deleteBranch","group":"3_modify@2"},{"command":"git.publish","group":"4_publish@1"}],"git.remotes":[{"command":"git.addRemote","group":"remote@1"},{"command":"git.removeRemote","group":"remote@2"}],"git.stash":[{"command":"git.stash","group":"1_stash@1"},{"command":"git.stashIncludeUntracked","group":"1_stash@2"},{"command":"git.stashStaged","when":"gitVersion2.35","group":"1_stash@3"},{"command":"git.stashApplyLatest","group":"2_apply@1"},{"command":"git.stashApply","group":"2_apply@2"},{"command":"git.stashPopLatest","group":"3_pop@1"},{"command":"git.stashPop","group":"3_pop@2"},{"command":"git.stashDrop","group":"4_drop@1"},{"command":"git.stashDropAll","group":"4_drop@2"},{"command":"git.stashView","when":"config.multiDiffEditor.experimental.enabled","group":"5_preview@1"}],"git.tags":[{"command":"git.createTag","group":"tags@1"},{"command":"git.deleteTag","group":"tags@2"},{"command":"git.deleteRemoteTag","group":"tags@3"}]},"submenus":[{"id":"git.commit","label":"Commit"},{"id":"git.changes","label":"Changes"},{"id":"git.pullpush","label":"Pull, Push"},{"id":"git.branch","label":"Branch"},{"id":"git.remotes","label":"Remote"},{"id":"git.stash","label":"Stash"},{"id":"git.tags","label":"Tags"}],"configuration":{"title":"Git","properties":{"git.enabled":{"type":"boolean","scope":"resource","description":"Whether Git is enabled.","default":true},"git.path":{"type":["string","null","array"],"markdownDescription":"Path and filename of the git executable, e.g. `C:\\Program Files\\Git\\bin\\git.exe` (Windows). This can also be an array of string values containing multiple paths to look up.","default":null,"scope":"machine"},"git.autoRepositoryDetection":{"type":["boolean","string"],"enum":[true,false,"subFolders","openEditors"],"enumDescriptions":["Scan for both subfolders of the current opened folder and parent folders of open files.","Disable automatic repository scanning.","Scan for subfolders of the currently opened folder.","Scan for parent folders of open files."],"description":"Configures when repositories should be automatically detected.","default":true},"git.autorefresh":{"type":"boolean","description":"Whether auto refreshing is enabled.","default":true},"git.autofetch":{"type":["boolean","string"],"enum":[true,false,"all"],"scope":"resource","markdownDescription":"When set to true, commits will automatically be fetched from the default remote of the current Git repository. Setting to `all` will fetch from all remotes.","default":false,"tags":["usesOnlineServices"]},"git.autofetchPeriod":{"type":"number","scope":"resource","markdownDescription":"Duration in seconds between each automatic git fetch, when `#git.autofetch#` is enabled.","default":180},"git.defaultBranchName":{"type":"string","markdownDescription":"The name of the default branch (example: main, trunk, development) when initializing a new Git repository. When set to empty, the default branch name configured in Git will be used. **Note:** Requires Git version `2.28.0` or later.","default":"main","scope":"resource"},"git.branchPrefix":{"type":"string","description":"Prefix used when creating a new branch.","default":"","scope":"resource"},"git.branchProtection":{"type":"array","markdownDescription":"List of protected branches. By default, a prompt is shown before changes are committed to a protected branch. The prompt can be controlled using the `#git.branchProtectionPrompt#`  setting.","items":{"type":"string"},"default":[],"scope":"resource"},"git.branchProtectionPrompt":{"type":"string","description":"Controls whether a prompt is being shown before changes are committed to a protected branch.","enum":["alwaysCommit","alwaysCommitToNewBranch","alwaysPrompt"],"enumDescriptions":["Always commit changes to the protected branch.","Always commit changes to a new branch.","Always prompt before changes are committed to a protected branch."],"default":"alwaysPrompt","scope":"resource"},"git.branchValidationRegex":{"type":"string","description":"A regular expression to validate new branch names.","default":""},"git.branchWhitespaceChar":{"type":"string","description":"The character to replace whitespace in new branch names, and to separate segments of a randomly generated branch name.","default":"-"},"git.branchRandomName.enable":{"type":"boolean","description":"Controls whether a random name is generated when creating a new branch.","default":false,"scope":"resource"},"git.branchRandomName.dictionary":{"type":"array","markdownDescription":"List of dictionaries used for the randomly generated branch name. Each value represents the dictionary used to generate the segment of the branch name. Supported dictionaries: `adjectives`, `animals`, `colors` and `numbers`.","items":{"type":"string","enum":["adjectives","animals","colors","numbers"],"enumDescriptions":["A random adjective","A random animal name","A random color name","A random number between 100 and 999"]},"minItems":1,"maxItems":5,"default":["adjectives","animals"],"scope":"resource"},"git.confirmSync":{"type":"boolean","description":"Confirm before synchronizing Git repositories.","default":true},"git.countBadge":{"type":"string","enum":["all","tracked","off"],"enumDescriptions":["Count all changes.","Count only tracked changes.","Turn off counter."],"description":"Controls the Git count badge.","default":"all","scope":"resource"},"git.checkoutType":{"type":"array","items":{"type":"string","enum":["local","tags","remote"],"enumDescriptions":["Local branches","Tags","Remote branches"]},"uniqueItems":true,"markdownDescription":"Controls what type of Git refs are listed when running `Checkout to...`.","default":["local","remote","tags"]},"git.ignoreLegacyWarning":{"type":"boolean","description":"Ignores the legacy Git warning.","default":false},"git.ignoreMissingGitWarning":{"type":"boolean","description":"Ignores the warning when Git is missing.","default":false},"git.ignoreWindowsGit27Warning":{"type":"boolean","description":"Ignores the warning when Git 2.25 - 2.26 is installed on Windows.","default":false},"git.ignoreLimitWarning":{"type":"boolean","description":"Ignores the warning when there are too many changes in a repository.","default":false},"git.ignoreRebaseWarning":{"type":"boolean","description":"Ignores the warning when it looks like the branch might have been rebased when pulling.","default":false},"git.defaultCloneDirectory":{"type":["string","null"],"default":null,"scope":"machine","description":"The default location to clone a Git repository."},"git.useEditorAsCommitInput":{"type":"boolean","description":"Controls whether a full text editor will be used to author commit messages, whenever no message is provided in the commit input box.","default":true},"git.verboseCommit":{"type":"boolean","scope":"resource","markdownDescription":"Enable verbose output when `#git.useEditorAsCommitInput#` is enabled.","default":false},"git.enableSmartCommit":{"type":"boolean","scope":"resource","description":"Commit all changes when there are no staged changes.","default":false},"git.smartCommitChanges":{"type":"string","enum":["all","tracked"],"enumDescriptions":["Automatically stage all changes.","Automatically stage tracked changes only."],"scope":"resource","description":"Control which changes are automatically staged by Smart Commit.","default":"all"},"git.suggestSmartCommit":{"type":"boolean","scope":"resource","description":"Suggests to enable smart commit (commit all changes when there are no staged changes).","default":true},"git.enableCommitSigning":{"type":"boolean","scope":"resource","description":"Enables commit signing with GPG, X.509, or SSH.","default":false},"git.confirmEmptyCommits":{"type":"boolean","scope":"resource","description":"Always confirm the creation of empty commits for the 'Git: Commit Empty' command.","default":true},"git.decorations.enabled":{"type":"boolean","default":true,"description":"Controls whether Git contributes colors and badges to the Explorer and the Open Editors view."},"git.enableStatusBarSync":{"type":"boolean","default":true,"description":"Controls whether the Git Sync command appears in the status bar.","scope":"resource"},"git.followTagsWhenSync":{"type":"boolean","scope":"resource","default":false,"description":"Push all annotated tags when running the sync command."},"git.replaceTagsWhenPull":{"type":"boolean","scope":"resource","default":false,"description":"Automatically replace the local tags with the remote tags in case of a conflict when running the pull command."},"git.promptToSaveFilesBeforeStash":{"type":"string","enum":["always","staged","never"],"enumDescriptions":["Check for any unsaved files.","Check only for unsaved staged files.","Disable this check."],"scope":"resource","default":"always","description":"Controls whether Git should check for unsaved files before stashing changes."},"git.promptToSaveFilesBeforeCommit":{"type":"string","enum":["always","staged","never"],"enumDescriptions":["Check for any unsaved files.","Check only for unsaved staged files.","Disable this check."],"scope":"resource","default":"always","description":"Controls whether Git should check for unsaved files before committing."},"git.postCommitCommand":{"type":"string","enum":["none","push","sync"],"enumDescriptions":["Don't run any command after a commit.","Run 'git push' after a successful commit.","Run 'git pull' and 'git push' after a successful commit."],"markdownDescription":"Run a git command after a successful commit.","scope":"resource","default":"none"},"git.rememberPostCommitCommand":{"type":"boolean","description":"Remember the last git command that ran after a commit.","scope":"resource","default":false},"git.openAfterClone":{"type":"string","enum":["always","alwaysNewWindow","whenNoFolderOpen","prompt"],"enumDescriptions":["Always open in current window.","Always open in a new window.","Only open in current window when no folder is opened.","Always prompt for action."],"default":"prompt","description":"Controls whether to open a repository automatically after cloning."},"git.showInlineOpenFileAction":{"type":"boolean","default":true,"description":"Controls whether to show an inline Open File action in the Git changes view."},"git.showPushSuccessNotification":{"type":"boolean","description":"Controls whether to show a notification when a push is successful.","default":false},"git.inputValidation":{"type":"boolean","default":false,"description":"Controls whether to show commit message input validation diagnostics."},"git.inputValidationLength":{"type":"number","default":72,"description":"Controls the commit message length threshold for showing a warning."},"git.inputValidationSubjectLength":{"type":["number","null"],"default":50,"markdownDescription":"Controls the commit message subject length threshold for showing a warning. Unset it to inherit the value of `#git.inputValidationLength#`."},"git.detectSubmodules":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether to automatically detect Git submodules."},"git.detectSubmodulesLimit":{"type":"number","scope":"resource","default":10,"description":"Controls the limit of Git submodules detected."},"git.alwaysShowStagedChangesResourceGroup":{"type":"boolean","scope":"resource","default":false,"description":"Always show the Staged Changes resource group."},"git.alwaysSignOff":{"type":"boolean","scope":"resource","default":false,"description":"Controls the signoff flag for all commits."},"git.ignoreSubmodules":{"type":"boolean","scope":"resource","default":false,"description":"Ignore modifications to submodules in the file tree."},"git.ignoredRepositories":{"type":"array","items":{"type":"string"},"default":[],"scope":"window","description":"List of Git repositories to ignore."},"git.scanRepositories":{"type":"array","items":{"type":"string"},"default":[],"scope":"resource","description":"List of paths to search for Git repositories in."},"git.showProgress":{"type":"boolean","description":"Controls whether Git actions should show progress.","default":true,"scope":"resource"},"git.rebaseWhenSync":{"type":"boolean","scope":"resource","default":false,"description":"Force Git to use rebase when running the sync command."},"git.pullBeforeCheckout":{"type":"boolean","scope":"resource","default":false,"description":"Controls whether a branch that does not have outgoing commits is fast-forwarded before it is checked out."},"git.fetchOnPull":{"type":"boolean","scope":"resource","default":false,"description":"When enabled, fetch all branches when pulling. Otherwise, fetch just the current one."},"git.pruneOnFetch":{"type":"boolean","scope":"resource","default":false,"description":"Prune when fetching."},"git.pullTags":{"type":"boolean","scope":"resource","default":true,"description":"Fetch all tags when pulling."},"git.autoStash":{"type":"boolean","scope":"resource","default":false,"description":"Stash any changes before pulling and restore them after successful pull."},"git.allowForcePush":{"type":"boolean","default":false,"description":"Controls whether force push (with or without lease) is enabled."},"git.useForcePushWithLease":{"type":"boolean","default":true,"description":"Controls whether force pushing uses the safer force-with-lease variant."},"git.useForcePushIfIncludes":{"type":"boolean","default":true,"markdownDescription":"Controls whether force pushing uses the safer force-if-includes variant. Note: This setting requires the `#git.useForcePushWithLease#` setting to be enabled, and Git version `2.30.0` or later."},"git.confirmForcePush":{"type":"boolean","default":true,"description":"Controls whether to ask for confirmation before force-pushing."},"git.allowNoVerifyCommit":{"type":"boolean","default":false,"description":"Controls whether commits without running pre-commit and commit-msg hooks are allowed."},"git.confirmNoVerifyCommit":{"type":"boolean","default":true,"description":"Controls whether to ask for confirmation before committing without verification."},"git.closeDiffOnOperation":{"type":"boolean","scope":"resource","default":false,"description":"Controls whether the diff editor should be automatically closed when changes are stashed, committed, discarded, staged, or unstaged."},"git.openDiffOnClick":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether the diff editor should be opened when clicking a change. Otherwise the regular editor will be opened."},"git.supportCancellation":{"type":"boolean","scope":"resource","default":false,"description":"Controls whether a notification comes up when running the Sync action, which allows the user to cancel the operation."},"git.branchSortOrder":{"type":"string","enum":["committerdate","alphabetically"],"default":"committerdate","description":"Controls the sort order for branches."},"git.untrackedChanges":{"type":"string","enum":["mixed","separate","hidden"],"enumDescriptions":["All changes, tracked and untracked, appear together and behave equally.","Untracked changes appear separately in the Source Control view. They are also excluded from several actions.","Untracked changes are hidden and excluded from several actions."],"default":"mixed","description":"Controls how untracked changes behave.","scope":"resource"},"git.requireGitUserConfig":{"type":"boolean","description":"Controls whether to require explicit Git user configuration or allow Git to guess if missing.","default":true,"scope":"resource"},"git.showCommitInput":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether to show the commit input in the Git source control panel."},"git.terminalAuthentication":{"type":"boolean","default":true,"description":"Controls whether to enable VS Code to be the authentication handler for Git processes spawned in the Integrated Terminal. Note: Terminals need to be restarted to pick up a change in this setting."},"git.terminalGitEditor":{"type":"boolean","default":false,"description":"Controls whether to enable VS Code to be the Git editor for Git processes spawned in the integrated terminal. Note: Terminals need to be restarted to pick up a change in this setting."},"git.useCommitInputAsStashMessage":{"type":"boolean","scope":"resource","default":false,"description":"Controls whether to use the message from the commit input box as the default stash message."},"git.useIntegratedAskPass":{"type":"boolean","default":true,"description":"Controls whether GIT_ASKPASS should be overwritten to use the integrated version."},"git.githubAuthentication":{"markdownDeprecationMessage":"This setting is now deprecated, please use `#github.gitAuthentication#` instead."},"git.timeline.date":{"type":"string","enum":["committed","authored"],"enumDescriptions":["Use the committed date","Use the authored date"],"default":"committed","description":"Controls which date to use for items in the Timeline view.","scope":"window"},"git.timeline.showAuthor":{"type":"boolean","default":true,"description":"Controls whether to show the commit author in the Timeline view.","scope":"window"},"git.timeline.showUncommitted":{"type":"boolean","default":false,"description":"Controls whether to show uncommitted changes in the Timeline view.","scope":"window"},"git.showActionButton":{"type":"object","additionalProperties":false,"description":"Controls whether an action button is shown in the Source Control view.","properties":{"commit":{"type":"boolean","description":"Show an action button to commit changes when the local branch has modified files ready to be committed."},"publish":{"type":"boolean","description":"Show an action button to publish the local branch when it does not have a tracking remote branch."},"sync":{"type":"boolean","description":"Show an action button to synchronize changes when the local branch is either ahead or behind the remote branch."}},"default":{"commit":true,"publish":true,"sync":true},"scope":"resource"},"git.statusLimit":{"type":"number","scope":"resource","default":10000,"description":"Controls how to limit the number of changes that can be parsed from Git status command. Can be set to 0 for no limit."},"git.repositoryScanIgnoredFolders":{"type":"array","items":{"type":"string"},"default":["node_modules"],"scope":"resource","markdownDescription":"List of folders that are ignored while scanning for Git repositories when `#git.autoRepositoryDetection#` is set to `true` or `subFolders`."},"git.repositoryScanMaxDepth":{"type":"number","scope":"resource","default":1,"markdownDescription":"Controls the depth used when scanning workspace folders for Git repositories when `#git.autoRepositoryDetection#` is set to `true` or `subFolders`. Can be set to `-1` for no limit."},"git.commandsToLog":{"type":"array","items":{"type":"string"},"default":[],"markdownDescription":"List of git commands (ex: commit, push) that would have their `stdout` logged to the [git output](command:git.showOutput). If the git command has a client-side hook configured, the client-side hook's `stdout` will also be logged to the [git output](command:git.showOutput)."},"git.mergeEditor":{"type":"boolean","default":false,"markdownDescription":"Open the merge editor for files that are currently under conflict.","scope":"window"},"git.optimisticUpdate":{"type":"boolean","default":true,"markdownDescription":"Controls whether to optimistically update the state of the Source Control view after running git commands.","scope":"resource","tags":["experimental"]},"git.openRepositoryInParentFolders":{"type":"string","enum":["always","never","prompt"],"enumDescriptions":["Always open a repository in parent folders of workspaces or open files.","Never open a repository in parent folders of workspaces or open files.","Prompt before opening a repository the parent folders of workspaces or open files."],"default":"prompt","markdownDescription":"Control whether a repository in parent folders of workspaces or open files should be opened.","scope":"resource"},"git.similarityThreshold":{"type":"number","default":50,"minimum":0,"maximum":100,"markdownDescription":"Controls the threshold of the similarity index (the amount of additions/deletions compared to the file's size) for changes in a pair of added/deleted files to be considered a rename. **Note:** Requires Git version `2.18.0` or later.","scope":"resource"},"git.blame.editorDecoration.enabled":{"type":"boolean","default":false,"markdownDescription":"Controls whether to show blame information in the editor using editor decorations."},"git.blame.editorDecoration.template":{"type":"string","default":"${subject}, ${authorName} (${authorDateAgo})","markdownDescription":"Template for the blame information editor decoration. Supported variables:\n\n* `hash`: Commit hash\n\n* `hashShort`: First N characters of the commit hash according to `#git.commitShortHashLength#`\n\n* `subject`: First line of the commit message\n\n* `authorName`: Author name\n\n* `authorEmail`: Author email\n\n* `authorDate`: Author date\n\n* `authorDateAgo`: Time difference between now and the author date\n\n"},"git.blame.statusBarItem.enabled":{"type":"boolean","default":true,"markdownDescription":"Controls whether to show blame information in the status bar."},"git.blame.statusBarItem.template":{"type":"string","default":"${authorName} (${authorDateAgo})","markdownDescription":"Template for the blame information status bar item. Supported variables:\n\n* `hash`: Commit hash\n\n* `hashShort`: First N characters of the commit hash according to `#git.commitShortHashLength#`\n\n* `subject`: First line of the commit message\n\n* `authorName`: Author name\n\n* `authorEmail`: Author email\n\n* `authorDate`: Author date\n\n* `authorDateAgo`: Time difference between now and the author date\n\n"},"git.commitShortHashLength":{"type":"number","default":7,"minimum":7,"maximum":40,"markdownDescription":"Controls the length of the commit short hash.","scope":"resource"}}},"colors":[{"id":"gitDecoration.addedResourceForeground","description":"Color for added resources.","defaults":{"light":"#587c0c","dark":"#81b88b","highContrast":"#a1e3ad","highContrastLight":"#374e06"}},{"id":"gitDecoration.modifiedResourceForeground","description":"Color for modified resources.","defaults":{"light":"#895503","dark":"#E2C08D","highContrast":"#E2C08D","highContrastLight":"#895503"}},{"id":"gitDecoration.deletedResourceForeground","description":"Color for deleted resources.","defaults":{"light":"#ad0707","dark":"#c74e39","highContrast":"#c74e39","highContrastLight":"#ad0707"}},{"id":"gitDecoration.renamedResourceForeground","description":"Color for renamed or copied resources.","defaults":{"light":"#007100","dark":"#73C991","highContrast":"#73C991","highContrastLight":"#007100"}},{"id":"gitDecoration.untrackedResourceForeground","description":"Color for untracked resources.","defaults":{"light":"#007100","dark":"#73C991","highContrast":"#73C991","highContrastLight":"#007100"}},{"id":"gitDecoration.ignoredResourceForeground","description":"Color for ignored resources.","defaults":{"light":"#8E8E90","dark":"#8C8C8C","highContrast":"#A7A8A9","highContrastLight":"#8e8e90"}},{"id":"gitDecoration.stageModifiedResourceForeground","description":"Color for modified resources which have been staged.","defaults":{"light":"#895503","dark":"#E2C08D","highContrast":"#E2C08D","highContrastLight":"#895503"}},{"id":"gitDecoration.stageDeletedResourceForeground","description":"Color for deleted resources which have been staged.","defaults":{"light":"#ad0707","dark":"#c74e39","highContrast":"#c74e39","highContrastLight":"#ad0707"}},{"id":"gitDecoration.conflictingResourceForeground","description":"Color for resources with conflicts.","defaults":{"light":"#ad0707","dark":"#e4676b","highContrast":"#c74e39","highContrastLight":"#ad0707"}},{"id":"gitDecoration.submoduleResourceForeground","description":"Color for submodule resources.","defaults":{"light":"#1258a7","dark":"#8db9e2","highContrast":"#8db9e2","highContrastLight":"#1258a7"}},{"id":"git.blame.editorDecorationForeground","description":"Color for the blame editor decoration.","defaults":{"dark":"editorCodeLens.foreground","light":"editorCodeLens.foreground","highContrast":"editorCodeLens.foreground","highContrastLight":"editorCodeLens.foreground"}}],"configurationDefaults":{"[git-commit]":{"editor.rulers":[50,72],"editor.wordWrap":"off","workbench.editor.restoreViewState":false},"[git-rebase]":{"workbench.editor.restoreViewState":false}},"viewsWelcome":[{"view":"scm","contents":"If you would like to use Git features, please enable Git in your [settings](command:workbench.action.openSettings?%5B%22git.enabled%22%5D).\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"!config.git.enabled"},{"view":"scm","contents":"Install Git, a popular source control system, to track code changes and collaborate with others. Learn more in our [Git guides](https://aka.ms/vscode-scm).","when":"config.git.enabled && git.missing && remoteName != ''"},{"view":"scm","contents":"[Download Git for macOS](https://git-scm.com/download/mac)\nAfter installing, please [reload](command:workbench.action.reloadWindow) (or [troubleshoot](command:git.showOutput)). Additional source control providers can be installed [from the Marketplace](command:workbench.extensions.search?%22%40category%3A%5C%22scm%20providers%5C%22%22).","when":"config.git.enabled && git.missing && remoteName == '' && isMac"},{"view":"scm","contents":"[Download Git for Windows](https://git-scm.com/download/win)\nAfter installing, please [reload](command:workbench.action.reloadWindow) (or [troubleshoot](command:git.showOutput)). Additional source control providers can be installed [from the Marketplace](command:workbench.extensions.search?%22%40category%3A%5C%22scm%20providers%5C%22%22).","when":"config.git.enabled && git.missing && remoteName == '' && isWindows"},{"view":"scm","contents":"Source control depends on Git being installed.\n[Download Git for Linux](https://git-scm.com/download/linux)\nAfter installing, please [reload](command:workbench.action.reloadWindow) (or [troubleshoot](command:git.showOutput)). Additional source control providers can be installed [from the Marketplace](command:workbench.extensions.search?%22%40category%3A%5C%22scm%20providers%5C%22%22).","when":"config.git.enabled && git.missing && remoteName == '' && isLinux"},{"view":"scm","contents":"In order to use Git features, you can open a folder containing a Git repository or clone from a URL.\n[Open Folder](command:vscode.openFolder)\n[Clone Repository](command:git.cloneRecursive)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && workbenchState == empty && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0","enablement":"git.state == initialized","group":"2_open@1"},{"view":"scm","contents":"The workspace currently open doesn't have any folders containing Git repositories.\n[Add Folder to Workspace](command:workbench.action.addRootFolder)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && workbenchState == workspace && workspaceFolderCount == 0 && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0","enablement":"git.state == initialized","group":"2_open@1"},{"view":"scm","contents":"Scanning folder for Git repositories...","when":"config.git.enabled && !git.missing && workbenchState == folder && workspaceFolderCount != 0 && git.state != initialized"},{"view":"scm","contents":"Scanning workspace for Git repositories...","when":"config.git.enabled && !git.missing && workbenchState == workspace && workspaceFolderCount != 0 && git.state != initialized"},{"view":"scm","contents":"The folder currently open doesn't have a Git repository. You can initialize a repository which will enable source control features powered by Git.\n[Initialize Repository](command:git.init?%5Btrue%5D)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && git.state == initialized && workbenchState == folder && scm.providerCount == 0 && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0 && remoteName != 'codespaces'","group":"5_scm@1"},{"view":"scm","contents":"The workspace currently open doesn't have any folders containing Git repositories. You can initialize a repository on a folder which will enable source control features powered by Git.\n[Initialize Repository](command:git.init)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && git.state == initialized && workbenchState == workspace && workspaceFolderCount != 0 && scm.providerCount == 0 && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0 && remoteName != 'codespaces'","group":"5_scm@1"},{"view":"scm","contents":"A Git repository was found in the parent folders of the workspace or the open file(s).\n[Open Repository](command:git.openRepositoriesInParentFolders)\nUse the [git.openRepositoryInParentFolders](command:workbench.action.openSettings?%5B%22git.openRepositoryInParentFolders%22%5D) setting to control whether Git repositories in parent folders of workspaces or open files are opened. To learn more [read our docs](https://aka.ms/vscode-git-repository-in-parent-folders).","when":"config.git.enabled && !git.missing && git.state == initialized && git.parentRepositoryCount == 1"},{"view":"scm","contents":"Git repositories were found in the parent folders of the workspace or the open file(s).\n[Open Repository](command:git.openRepositoriesInParentFolders)\nUse the [git.openRepositoryInParentFolders](command:workbench.action.openSettings?%5B%22git.openRepositoryInParentFolders%22%5D) setting to control whether Git repositories in parent folders of workspace or open files are opened. To learn more [read our docs](https://aka.ms/vscode-git-repository-in-parent-folders).","when":"config.git.enabled && !git.missing && git.state == initialized && git.parentRepositoryCount > 1"},{"view":"scm","contents":"The detected Git repository is potentially unsafe as the folder is owned by someone other than the current user.\n[Manage Unsafe Repositories](command:git.manageUnsafeRepositories)\nTo learn more about unsafe repositories [read our docs](https://aka.ms/vscode-git-unsafe-repository).","when":"config.git.enabled && !git.missing && git.state == initialized && git.unsafeRepositoryCount == 1"},{"view":"scm","contents":"The detected Git repositories are potentially unsafe as the folders are owned by someone other than the current user.\n[Manage Unsafe Repositories](command:git.manageUnsafeRepositories)\nTo learn more about unsafe repositories [read our docs](https://aka.ms/vscode-git-unsafe-repository).","when":"config.git.enabled && !git.missing && git.state == initialized && git.unsafeRepositoryCount > 1"},{"view":"scm","contents":"A Git repository was found that was previously closed.\n[Reopen Closed Repository](command:git.reopenClosedRepositories)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && git.state == initialized && git.closedRepositoryCount == 1"},{"view":"scm","contents":"Git repositories were found that were previously closed.\n[Reopen Closed Repositories](command:git.reopenClosedRepositories)\nTo learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && !git.missing && git.state == initialized && git.closedRepositoryCount > 1"},{"view":"explorer","contents":"You can clone a repository locally.\n[Clone Repository](command:git.clone 'Clone a repository once the Git extension has activated')","when":"config.git.enabled && git.state == initialized && scm.providerCount == 0","group":"5_scm@1"},{"view":"explorer","contents":"To learn more about how to use Git and source control in VS Code [read our docs](https://aka.ms/vscode-scm).","when":"config.git.enabled && git.state == initialized && scm.providerCount == 0","group":"5_scm@10"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/git","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.git-base"},"manifest":{"name":"git-base","displayName":"Git Base","description":"Git static contributions and pickers.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"categories":["Other"],"activationEvents":["*"],"main":"./dist/extension.js","browser":"./dist/browser/extension.js","icon":"resources/icons/git.png","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"commands":[{"command":"git-base.api.getRemoteSources","title":"Get Remote Sources","category":"Git Base API"}],"menus":{"commandPalette":[{"command":"git-base.api.getRemoteSources","when":"false"}]},"languages":[{"id":"git-commit","aliases":["Git Commit Message","git-commit"],"filenames":["COMMIT_EDITMSG","MERGE_MSG"],"configuration":"./languages/git-commit.language-configuration.json"},{"id":"git-rebase","aliases":["Git Rebase Message","git-rebase"],"filenames":["git-rebase-todo"],"filenamePatterns":["**/rebase-merge/done"],"configuration":"./languages/git-rebase.language-configuration.json"},{"id":"ignore","aliases":["Ignore","ignore"],"extensions":[".gitignore_global",".gitignore",".git-blame-ignore-revs"],"configuration":"./languages/ignore.language-configuration.json"}],"grammars":[{"language":"git-commit","scopeName":"text.git-commit","path":"./syntaxes/git-commit.tmLanguage.json"},{"language":"git-rebase","scopeName":"text.git-rebase","path":"./syntaxes/git-rebase.tmLanguage.json"},{"language":"ignore","scopeName":"source.ignore","path":"./syntaxes/ignore.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/git-base","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.github"},"manifest":{"name":"github","displayName":"GitHub","description":"GitHub features for VS Code","publisher":"vscode","license":"MIT","version":"0.0.1","engines":{"vscode":"^1.41.0"},"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","icon":"images/icon.png","categories":["Other"],"activationEvents":["*"],"extensionDependencies":["vscode.git-base"],"main":"./dist/extension.js","capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"enabledApiProposals":["canonicalUriProvider","contribEditSessions","contribShareMenu","contribSourceControlHistoryItemMenu","scmHistoryProvider","shareProvider","timeline"],"contributes":{"commands":[{"command":"github.publish","title":"Publish to GitHub"},{"command":"github.copyVscodeDevLink","title":"Copy vscode.dev Link"},{"command":"github.copyVscodeDevLinkFile","title":"Copy vscode.dev Link"},{"command":"github.copyVscodeDevLinkWithoutRange","title":"Copy vscode.dev Link"},{"command":"github.openOnVscodeDev","title":"Open in vscode.dev","icon":"$(globe)"},{"command":"github.graph.openOnGitHub","title":"Open on GitHub","icon":"$(github)"},{"command":"github.timeline.openOnGitHub","title":"Open on GitHub","icon":"$(github)"}],"continueEditSession":[],"menus":{"commandPalette":[{"command":"github.publish","when":"git-base.gitEnabled && workspaceFolderCount != 0 && remoteName != 'codespaces'"},{"command":"github.graph.openOnGitHub","when":"false"},{"command":"github.copyVscodeDevLink","when":"false"},{"command":"github.copyVscodeDevLinkFile","when":"false"},{"command":"github.copyVscodeDevLinkWithoutRange","when":"false"},{"command":"github.openOnVscodeDev","when":"false"},{"command":"github.timeline.openOnGitHub","when":"false"}],"file/share":[{"command":"github.copyVscodeDevLinkFile","when":"github.hasGitHubRepo && remoteName != 'codespaces'","group":"0_vscode@0"}],"editor/context/share":[{"command":"github.copyVscodeDevLink","when":"github.hasGitHubRepo && resourceScheme != untitled && !isInEmbeddedEditor && remoteName != 'codespaces'","group":"0_vscode@0"}],"explorer/context/share":[{"command":"github.copyVscodeDevLinkWithoutRange","when":"github.hasGitHubRepo && resourceScheme != untitled && !isInEmbeddedEditor && remoteName != 'codespaces'","group":"0_vscode@0"}],"editor/lineNumber/context":[{"command":"github.copyVscodeDevLink","when":"github.hasGitHubRepo && resourceScheme != untitled && activeEditor == workbench.editors.files.textFileEditor && config.editor.lineNumbers == on && remoteName != 'codespaces'","group":"1_cutcopypaste@2"},{"command":"github.copyVscodeDevLink","when":"github.hasGitHubRepo && resourceScheme != untitled && activeEditor == workbench.editor.notebook && remoteName != 'codespaces'","group":"1_cutcopypaste@2"}],"editor/title/context/share":[{"command":"github.copyVscodeDevLinkWithoutRange","when":"github.hasGitHubRepo && resourceScheme != untitled && remoteName != 'codespaces'","group":"0_vscode@0"}],"scm/historyItem/context":[{"command":"github.graph.openOnGitHub","when":"github.hasGitHubRepo","group":"0_view@2"}],"scm/historyItem/hover":[{"command":"github.graph.openOnGitHub","when":"github.hasGitHubRepo","group":"1_open@1"}],"timeline/item/context":[{"command":"github.timeline.openOnGitHub","group":"1_actions@3","when":"github.hasGitHubRepo && timelineItem =~ /git:file:commit\\b/"}]},"configuration":[{"title":"GitHub","properties":{"github.branchProtection":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether to query repository rules for GitHub repositories"},"github.gitAuthentication":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether to enable automatic GitHub authentication for git commands within VS Code."},"github.gitProtocol":{"type":"string","enum":["https","ssh"],"default":"https","description":"Controls which protocol is used to clone a GitHub repository"},"github.showAvatar":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether to show the GitHub avatar of the commit author in various hovers (ex: Git blame, Timeline, Source Control Graph, etc.)"}}}],"viewsWelcome":[{"view":"scm","contents":"You can directly publish this folder to a GitHub repository. Once published, you'll have access to source control features powered by Git and GitHub.\n[$(github) Publish to GitHub](command:github.publish)","when":"config.git.enabled && git.state == initialized && workbenchState == folder && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0"},{"view":"scm","contents":"You can directly publish a workspace folder to a GitHub repository. Once published, you'll have access to source control features powered by Git and GitHub.\n[$(github) Publish to GitHub](command:github.publish)","when":"config.git.enabled && git.state == initialized && workbenchState == workspace && workspaceFolderCount != 0 && git.parentRepositoryCount == 0 && git.unsafeRepositoryCount == 0 && git.closedRepositoryCount == 0"}],"markdown.previewStyles":["./markdown.css"]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/github","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.github-authentication"},"manifest":{"name":"github-authentication","displayName":"GitHub Authentication","description":"Gitpod Override","publisher":"vscode","license":"MIT","version":"0.0.2","engines":{"vscode":"^1.41.0"},"categories":["Other"],"api":"none","extensionKind":["ui","workspace"],"activationEvents":["onAuthenticationRequest:github"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"main":"./extension.js"},"location":{"$mid":1,"path":"/ide/extensions/github-authentication","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"gitpod.gitpod-web"},"manifest":{"name":"gitpod-web","displayName":"Gitpod Web","description":"Gitpod Web Support","publisher":"gitpod","version":"0.0.7","license":"MIT","icon":"resources/gitpod.png","repository":{"type":"git","url":"https://github.com/gitpod-io/gitpod-code"},"bugs":{"url":"https://github.com/gitpod-io/gitpod/issues"},"engines":{"vscode":"^1.83.0"},"enabledApiProposals":["resolvers","tunnels","contribViewsRemote","contribRemoteHelp","contribMenuBarHome"],"categories":["Other"],"extensionKind":["workspace"],"activationEvents":["*","onAuthenticationRequest:gitpod","onAuthenticationRequest:github"],"main":"./out/extension.js","segmentKey":"untrusted-dummy-key","contributes":{"commands":[{"command":"gitpod.stop.ws","title":"Gitpod: Stop Workspace","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.open.settings","title":"Gitpod: Open Settings","enablement":"gitpod.inWorkspace == true"},{"command":"gitpod.open.accessControl","title":"Gitpod: Open Access Control","enablement":"gitpod.inWorkspace == true"},{"command":"gitpod.open.context","title":"Gitpod: Open Context","enablement":"gitpod.inWorkspace == true"},{"command":"gitpod.open.dashboard","title":"Gitpod: Open Dashboard","enablement":"gitpod.inWorkspace == true"},{"command":"gitpod.open.documentation","title":"Gitpod: Documentation","enablement":"gitpod.inWorkspace == true"},{"command":"gitpod.ExtendTimeout","title":"Gitpod: Extend Workspace Timeout (180m)","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.setWorkspaceTimeout","title":"Gitpod: Configure Workspace Timeout","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.takeSnapshot","title":"Gitpod: Share Workspace Snapshot","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.shareWorkspace","title":"Gitpod: Share Running Workspace","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true && gitpod.workspaceShared == false"},{"command":"gitpod.stopSharingWorkspace","title":"Gitpod: Stop Sharing Running Workspace","enablement":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true && gitpod.workspaceShared == true"},{"command":"gitpod.openInStable","title":"Gitpod: Open in VS Code","enablement":"gitpod.inWorkspace == true && gitpod.UIKind == 'web'"},{"command":"gitpod.openInInsiders","title":"Gitpod: Open in VS Code Insiders","enablement":"gitpod.inWorkspace == true && gitpod.UIKind == 'web'"},{"command":"gitpod.ports.openBrowser","title":"Open Browser","icon":"$(globe)"},{"command":"gitpod.ports.retryAutoExpose","title":"Retry to expose","icon":"$(refresh)"},{"command":"gitpod.ports.preview","title":"Open Preview","icon":"$(open-preview)"},{"command":"gitpod.ports.makePrivate","title":"Make Private","icon":"$(unlock)"},{"command":"gitpod.ports.makePublic","title":"Make Public","icon":"$(lock)"},{"command":"gitpod.ports.makeHTTPS","title":"Use HTTPS Protocol","icon":"$(workspace-untrusted)"},{"command":"gitpod.ports.makeHTTP","title":"Use HTTP Protocol","icon":"$(workspace-trusted)"},{"command":"gitpod.ports.tunnelNetwork","title":"Tunnel on all interfaces","icon":"$(eye)"},{"command":"gitpod.ports.tunnelHost","title":"Tunnel on localhost","icon":"$(eye-closed)"},{"command":"gitpod.extensions.addToConfig","title":"Add to .gitpod.yml"},{"command":"gitpod.dev.connectLocalApp","title":"Developer: Connect to Local Gitpod App"},{"command":"gitpod.dev.enableForwardedPortsView","title":"Developer: Enable Tunneled Ports View"},{"command":"gitpod.gitpodyml.run","title":"Gitpod: Validate Configuration","icon":"$(play)"},{"command":"gitpod.gitpodyml.editorContext.run","title":"Gitpod: Validate Configuration","icon":"$(play)"},{"command":"gitpod.gitpodyml.dockerfile.editorContext.run","title":"Gitpod: Validate Configuration","icon":"$(play)"},{"command":"gitpod.gitpodyml.editorTitle.run","title":"Gitpod: Validate Configuration","icon":"$(play)"},{"command":"gitpod.gitpodyml.dockerfile.editorTitle.run","title":"Gitpod: Validate Configuration","icon":"$(play)"}],"menus":{"menuBar/home":[{"command":"gitpod.open.context","group":"gitpod@10"},{"command":"gitpod.open.dashboard","group":"gitpod@20"},{"command":"gitpod.open.documentation","group":"gitpod@30","when":"gitpod.dataops !== true"},{"command":"gitpod.takeSnapshot","group":"gitpod@40","when":"gitpod.workspaceOwned == true"},{"command":"gitpod.shareWorkspace","group":"gitpod@50","when":"gitpod.workspaceOwned == true && gitpod.workspaceShared == false"},{"command":"gitpod.stopSharingWorkspace","group":"gitpod@50","when":"gitpod.workspaceOwned == true && gitpod.workspaceShared == true"},{"command":"gitpod.stop.ws","group":"gitpod@60","when":"gitpod.workspaceOwned == true"},{"command":"gitpod.openInStable","group":"gitpod@70"},{"command":"gitpod.openInInsiders","group":"gitpod@80"}],"accounts/context":[{"command":"gitpod.open.settings","group":"navigation@10"},{"command":"gitpod.open.accessControl","group":"navigation@20"},{"command":"gitpod.ExtendTimeout","group":"navigation@50","when":"gitpod.workspaceOwned == true"}],"extension/context":[{"command":"gitpod.extensions.addToConfig","group":"2_configure"}],"view/item/context":[{"command":"gitpod.ports.tunnelNetwork","when":"view == gitpod.workspace && viewItem =~ /host/ && viewItem =~ /tunneled/","group":"inline@1"},{"command":"gitpod.ports.tunnelHost","when":"view == gitpod.workspace && viewItem =~ /network/ && viewItem =~ /tunneled/","group":"inline@1"},{"command":"gitpod.ports.makePublic","when":"view == gitpod.workspace && viewItem =~ /private/","group":"inline@2"},{"command":"gitpod.ports.makePrivate","when":"view == gitpod.workspace && viewItem =~ /public/","group":"inline@2"},{"command":"gitpod.ports.preview","when":"view == gitpod.workspace && viewItem =~ /exposed/ || viewItem =~ /tunneled/","group":"inline@3"},{"command":"gitpod.ports.openBrowser","when":"view == gitpod.workspace && viewItem =~ /exposed/ || viewItem =~ /tunneled/","group":"inline@4"},{"command":"gitpod.ports.retryAutoExpose","when":"view == gitpod.workspace && viewItem =~ /failed/","group":"inline@5"}],"commandPalette":[{"command":"gitpod.extensions.addToConfig","when":"false"},{"command":"gitpod.ports.preview","when":"false"},{"command":"gitpod.ports.openBrowser","when":"false"},{"command":"gitpod.ports.retryAutoExpose","when":"false"},{"command":"gitpod.ports.makePublic","when":"false"},{"command":"gitpod.ports.makePrivate","when":"false"},{"command":"gitpod.ports.makeHTTPS","when":"false"},{"command":"gitpod.ports.makeHTTP","when":"false"},{"command":"gitpod.ports.tunnelNetwork","when":"false"},{"command":"gitpod.ports.tunnelHost","when":"false"},{"command":"gitpod.dev.connectLocalApp","when":"gitpod.localAppConnected == false"},{"command":"gitpod.gitpodyml.run","when":"gitpod.rebuild.enabled"},{"command":"gitpod.gitpodyml.editorContext.run","when":"false"},{"command":"gitpod.gitpodyml.dockerfile.editorContext.run","when":"false"},{"command":"gitpod.gitpodyml.editorTitle.run","when":"false"},{"command":"gitpod.gitpodyml.dockerfile.editorTitle.run","when":"false"},{"command":"gitpod.open.documentation","when":"gitpod.dataops !== true"}],"statusBar/remoteIndicator":[{"command":"gitpod.stop.ws","group":"remote_00_gitpod_navigation@10","when":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.open.settings","group":"remote_00_gitpod_navigation@20","when":"gitpod.inWorkspace == true"},{"command":"gitpod.open.accessControl","group":"remote_00_gitpod_navigation@30","when":"gitpod.inWorkspace == true"},{"command":"gitpod.open.context","group":"remote_00_gitpod_navigation@40","when":"gitpod.inWorkspace == true"},{"command":"gitpod.open.dashboard","group":"remote_00_gitpod_navigation@50","when":"gitpod.inWorkspace == true"},{"command":"gitpod.open.documentation","group":"remote_00_gitpod_navigation@60","when":"gitpod.inWorkspace == true && gitpod.dataops !== true"},{"command":"gitpod.ExtendTimeout","group":"remote_00_gitpod_navigation@110","when":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.takeSnapshot","group":"remote_00_gitpod_navigation@120","when":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true"},{"command":"gitpod.shareWorkspace","group":"remote_00_gitpod_navigation@130","when":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true && gitpod.workspaceShared == false"},{"command":"gitpod.stopSharingWorkspace","group":"remote_00_gitpod_navigation@130","when":"gitpod.inWorkspace == true && gitpod.workspaceOwned == true && gitpod.workspaceShared == true"},{"command":"gitpod.openInStable","group":"remote_00_gitpod_navigation@900","when":"gitpod.inWorkspace == true && gitpod.UIKind == 'web'"},{"command":"gitpod.openInInsiders","group":"remote_00_gitpod_navigation@1000","when":"gitpod.inWorkspace == true && gitpod.UIKind == 'web'"}],"editor/context":[{"command":"gitpod.gitpodyml.editorContext.run","when":"gitpod.rebuild.enabled && resourceFilename == '.gitpod.yml' && resourceScheme == file","group":"0_navigation@0"},{"command":"gitpod.gitpodyml.dockerfile.editorContext.run","when":"gitpod.rebuild.enabled && gitpod.rebuild.dockerfile","group":"0_navigation@0"}],"editor/title":[{"command":"gitpod.gitpodyml.editorTitle.run","when":"gitpod.rebuild.enabled && resourceFilename == '.gitpod.yml' && resourceScheme == file","group":"navigation"},{"command":"gitpod.gitpodyml.dockerfile.editorTitle.run","when":"gitpod.rebuild.enabled && gitpod.rebuild.dockerfile","group":"navigation"}]},"views":{"remote":[{"id":"gitpod.workspace","name":"Gitpod Workspace","when":"false"}],"portsView":[{"id":"gitpod.portsView","name":"Ports","type":"webview","icon":"$(plug)"}]},"authentication":[{"id":"gitpod","label":"Gitpod"},{"id":"github","label":"GitHub"}],"walkthroughs":[{"id":"gitpod-getstarted","title":"Get Started with Gitpod","description":"Explore what Gitpod is and how to get the most out of it.","steps":[{"id":"whatisgitpod","title":"What is Gitpod?","description":"","media":{"markdown":"resources/walkthroughs/getstarted/what-is-gitpod.md"}},{"id":"prebuilds","title":"Prebuilds do the heavy lifting","description":"","media":{"markdown":"resources/walkthroughs/getstarted/prebuilds.md"}},{"id":"workspaces","title":"A new workspace per task","description":"","media":{"markdown":"resources/walkthroughs/getstarted/workspaces.md"}},{"id":"terminal","title":"Full access to a terminal","description":"","media":{"markdown":"resources/walkthroughs/getstarted/terminal.md"}},{"id":"ports","title":"Open ports","description":"","media":{"markdown":"resources/walkthroughs/getstarted/ports.md"}},{"id":"vscodedesktopsupport","title":"VS Code desktop support","description":"","media":{"markdown":"resources/walkthroughs/getstarted/vscodedesktopsupport.md"}}]}],"configuration":{"title":"Gitpod","properties":{"gitpod.validate.neverPrompt":{"type":"boolean","description":"Control whether to prompt to validate the workspace configuration on change.","default":false},"gitpod.openInStable.neverPrompt":{"type":"boolean","description":"Control whether to prompt to open in VS Code Desktop on page load.","default":false}}},"viewsContainers":{"panel":[{"id":"portsView","title":"Ports","icon":"$(plug)"}]}},"scripts":{"vscode:prepublish":"yarn build:webview && webpack --mode production","inject-commit-hash":"jq '.commit = \"'$(git rev-parse --short HEAD)'\"' package.json > package.json.tmp && mv package.json.tmp package.json","webpack":"webpack --mode development","compile":"tsc -b","watch":"tsc -b -w","build:webview":"node esbuild.js","watch:webview":"node esbuild.js --watch","start:webview":"sirv public --no-clear","update-dts":"cd \"src/@types\" && npx vscode-dts dev","package":"npx vsce package --yarn --allow-star-activation"},"devDependencies":{"@parcel/watcher":"^2.1.0","@tsconfig/svelte":"^2.0.0","@types/js-yaml":"^4.0.5","@types/node":"16.x","@types/node-fetch":"^2.5.12","@types/uuid":"8.0.0","@types/vscode":"1.75.0","@types/vscode-webview":"^1.57.0","@types/webpack":"^5.28.0","@types/yauzl":"^2.9.1","@types/yazl":"^2.4.2","@typescript-eslint/eslint-plugin":"^5.19.0","@typescript-eslint/parser":"^5.19.0","esbuild":"^0.17.11","esbuild-plugin-alias-path":"^2.0.2","esbuild-svelte":"^0.7.3","eslint":"^8.13.0","sirv-cli":"^2.0.0","svelte":"^3.0.0","svelte-check":"^2.0.0","svelte-preprocess":"^5.0.1","ts-loader":"^9.2.7","ttf-loader":"^1.0.2","typescript":"^4.6.3","webpack":"^5.42.0","webpack-cli":"^4.7.2"},"dependencies":{"gitpod-shared":"link:./../gitpod-shared","js-yaml":"^4.1.0","node-fetch":"2.6.7","uuid":"8.1.0","vscode-nls":"^5.0.0","yauzl":"^2.9.2","yazl":"^2.5.1"},"commit":"3953e81"},"location":{"$mid":1,"path":"/ide/extensions/gitpod-web","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"gitpod.gitpod-theme","uuid":"985c019b-4620-45a7-92c5-b721d8c11bc0"},"manifest":{"name":"gitpod-theme","displayName":"Gitpod Theme","description":"Gitpod Theme for VS Code","publisher":"gitpod","version":"0.0.2","license":"MIT","icon":"resources/icon.png","repository":{"type":"git","url":"https://github.com/gitpod-io/gitpod-vscode-theme"},"bugs":{"url":"https://github.com/gitpod-io/gitpod/issues"},"engines":{"vscode":"^1.62.0"},"categories":["Themes"],"keywords":["theme","dark","light","gitpod"],"contributes":{"themes":[{"id":"Gitpod Dark","label":"Gitpod Dark","uiTheme":"vs-dark","path":"./themes/gitpod-dark-color-theme.json"},{"id":"Gitpod Light","label":"Gitpod Light","uiTheme":"vs","path":"./themes/gitpod-light-color-theme.json"}]}},"location":{"$mid":1,"path":"/ide/extensions/gitpod.gitpod-theme","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","publisherDisplayName":"Gitpod","metadata":{"id":"985c019b-4620-45a7-92c5-b721d8c11bc0","publisherId":{"publisherId":"452246e5-9ce6-485c-b35f-5a50d93f81df","publisherName":"gitpod","displayName":"Gitpod","flags":"verified"},"publisherDisplayName":"Gitpod"},"isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.go"},"manifest":{"name":"go","displayName":"Go Language Basics","description":"Provides syntax highlighting and bracket matching in Go files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin worlpaker/go-syntax syntaxes/go.tmLanguage.json ./syntaxes/go.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"go","extensions":[".go"],"aliases":["Go"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"go","scopeName":"source.go","path":"./syntaxes/go.tmLanguage.json"}],"configurationDefaults":{"[go]":{"editor.insertSpaces":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/go","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.groovy"},"manifest":{"name":"groovy","displayName":"Groovy Language Basics","description":"Provides snippets, syntax highlighting and bracket matching in Groovy files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin textmate/groovy.tmbundle Syntaxes/Groovy.tmLanguage ./syntaxes/groovy.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"groovy","aliases":["Groovy","groovy"],"extensions":[".groovy",".gvy",".gradle",".jenkinsfile",".nf"],"filenames":["Jenkinsfile"],"filenamePatterns":["Jenkinsfile*"],"firstLine":"^#!.*\\bgroovy\\b","configuration":"./language-configuration.json"}],"grammars":[{"language":"groovy","scopeName":"source.groovy","path":"./syntaxes/groovy.tmLanguage.json"}],"snippets":[{"language":"groovy","path":"./snippets/groovy.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/groovy","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.grunt"},"manifest":{"name":"grunt","publisher":"vscode","description":"Extension to add Grunt capabilities to VS Code.","displayName":"Grunt support for VS Code","version":"1.0.0","private":true,"icon":"images/grunt.png","license":"MIT","engines":{"vscode":"*"},"categories":["Other"],"main":"./dist/main","activationEvents":["onTaskType:grunt"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"contributes":{"configuration":{"id":"grunt","type":"object","title":"Grunt","properties":{"grunt.autoDetect":{"scope":"application","type":"string","enum":["off","on"],"default":"off","description":"Controls enablement of Grunt task detection. Grunt task detection can cause files in any open workspace to be executed."}}},"taskDefinitions":[{"type":"grunt","required":["task"],"properties":{"task":{"type":"string","description":"The Grunt task to customize."},"args":{"type":"array","description":"Command line arguments to pass to the grunt task"},"file":{"type":"string","description":"The Grunt file that provides the task. Can be omitted."}},"when":"shellExecutionSupported"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/grunt","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.gulp"},"manifest":{"name":"gulp","publisher":"vscode","description":"Extension to add Gulp capabilities to VSCode.","displayName":"Gulp support for VSCode","version":"1.0.0","icon":"images/gulp.png","license":"MIT","engines":{"vscode":"*"},"categories":["Other"],"main":"./dist/main","activationEvents":["onTaskType:gulp"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"contributes":{"configuration":{"id":"gulp","type":"object","title":"Gulp","properties":{"gulp.autoDetect":{"scope":"application","type":"string","enum":["off","on"],"default":"off","description":"Controls enablement of Gulp task detection. Gulp task detection can cause files in any open workspace to be executed."}}},"taskDefinitions":[{"type":"gulp","required":["task"],"properties":{"task":{"type":"string","description":"The Gulp task to customize."},"file":{"type":"string","description":"The Gulp file that provides the task. Can be omitted."}},"when":"shellExecutionSupported"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/gulp","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.handlebars"},"manifest":{"name":"handlebars","displayName":"Handlebars Language Basics","description":"Provides syntax highlighting and bracket matching in Handlebars files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin daaain/Handlebars grammars/Handlebars.json ./syntaxes/Handlebars.tmLanguage.json"},"categories":["Programming Languages"],"extensionKind":["ui","workspace"],"contributes":{"languages":[{"id":"handlebars","extensions":[".handlebars",".hbs",".hjs"],"aliases":["Handlebars","handlebars"],"mimetypes":["text/x-handlebars-template"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"handlebars","scopeName":"text.html.handlebars","path":"./syntaxes/Handlebars.tmLanguage.json"}],"htmlLanguageParticipants":[{"languageId":"handlebars","autoInsert":true}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/handlebars","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[[2,"property `extensionKind` can be defined only if property `main` is also defined."]],"preRelease":false},{"type":0,"identifier":{"id":"vscode.hlsl"},"manifest":{"name":"hlsl","displayName":"HLSL Language Basics","description":"Provides syntax highlighting and bracket matching in HLSL files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin tgjones/shaders-tmLanguage grammars/hlsl.json ./syntaxes/hlsl.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"hlsl","extensions":[".hlsl",".hlsli",".fx",".fxh",".vsh",".psh",".cginc",".compute"],"aliases":["HLSL","hlsl"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"hlsl","path":"./syntaxes/hlsl.tmLanguage.json","scopeName":"source.hlsl"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/hlsl","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.html"},"manifest":{"name":"html","displayName":"HTML Language Basics","description":"Provides syntax highlighting, bracket matching & snippets in HTML files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ./build/update-grammar.mjs"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"html","extensions":[".html",".htm",".shtml",".xhtml",".xht",".mdoc",".jsp",".asp",".aspx",".jshtm",".volt",".ejs",".rhtml"],"aliases":["HTML","htm","html","xhtml"],"mimetypes":["text/html","text/x-jshtm","text/template","text/ng-template","application/xhtml+xml"],"configuration":"./language-configuration.json"}],"grammars":[{"scopeName":"text.html.basic","path":"./syntaxes/html.tmLanguage.json","embeddedLanguages":{"text.html":"html","source.css":"css","source.js":"javascript","source.python":"python","source.smarty":"smarty"},"tokenTypes":{"meta.tag string.quoted":"other"}},{"language":"html","scopeName":"text.html.derivative","path":"./syntaxes/html-derivative.tmLanguage.json","embeddedLanguages":{"text.html":"html","source.css":"css","source.js":"javascript","source.python":"python","source.smarty":"smarty"},"tokenTypes":{"meta.tag string.quoted":"other"}}],"snippets":[{"language":"html","path":"./snippets/html.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/html","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.html-language-features"},"manifest":{"name":"html-language-features","displayName":"HTML Language Features","description":"Provides rich language support for HTML and Handlebar files","version":"1.0.0","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.77.0"},"icon":"icons/html.png","activationEvents":["onLanguage:html","onLanguage:handlebars"],"enabledApiProposals":["extensionsAny"],"main":"./client/dist/node/htmlClientMain","browser":"./client/dist/browser/htmlClientMain","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"categories":["Programming Languages"],"contributes":{"configuration":{"id":"html","order":20,"type":"object","title":"HTML","properties":{"html.completion.attributeDefaultValue":{"type":"string","scope":"resource","enum":["doublequotes","singlequotes","empty"],"enumDescriptions":["Attribute value is set to \"\".","Attribute value is set to ''.","Attribute value is not set."],"default":"doublequotes","markdownDescription":"Controls the default value for attributes when completion is accepted."},"html.customData":{"type":"array","markdownDescription":"A list of relative file paths pointing to JSON files following the [custom data format](https://github.com/microsoft/vscode-html-languageservice/blob/master/docs/customData.md).\n\nVS Code loads custom data on startup to enhance its HTML support for the custom HTML tags, attributes and attribute values you specify in the JSON files.\n\nThe file paths are relative to workspace and only workspace folder settings are considered.","default":[],"items":{"type":"string"},"scope":"resource"},"html.format.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable default HTML formatter."},"html.format.wrapLineLength":{"type":"integer","scope":"resource","default":120,"description":"Maximum amount of characters per line (0 = disable)."},"html.format.unformatted":{"type":["string","null"],"scope":"resource","default":"wbr","markdownDescription":"List of tags, comma separated, that shouldn't be reformatted. `null` defaults to all tags listed at https://www.w3.org/TR/html5/dom.html#phrasing-content."},"html.format.contentUnformatted":{"type":["string","null"],"scope":"resource","default":"pre,code,textarea","markdownDescription":"List of tags, comma separated, where the content shouldn't be reformatted. `null` defaults to the `pre` tag."},"html.format.indentInnerHtml":{"type":"boolean","scope":"resource","default":false,"markdownDescription":"Indent `<head>` and `<body>` sections."},"html.format.preserveNewLines":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether existing line breaks before elements should be preserved. Only works before elements, not inside tags or for text."},"html.format.maxPreserveNewLines":{"type":["number","null"],"scope":"resource","default":null,"markdownDescription":"Maximum number of line breaks to be preserved in one chunk. Use `null` for unlimited."},"html.format.indentHandlebars":{"type":"boolean","scope":"resource","default":false,"markdownDescription":"Format and indent `{{#foo}}` and `{{/foo}}`."},"html.format.extraLiners":{"type":["string","null"],"scope":"resource","default":"head, body, /html","markdownDescription":"List of tags, comma separated, that should have an extra newline before them. `null` defaults to `\"head, body, /html\"`."},"html.format.wrapAttributes":{"type":"string","scope":"resource","default":"auto","enum":["auto","force","force-aligned","force-expand-multiline","aligned-multiple","preserve","preserve-aligned"],"enumDescriptions":["Wrap attributes only when line length is exceeded.","Wrap each attribute except first.","Wrap each attribute except first and keep aligned.","Wrap each attribute.","Wrap when line length is exceeded, align attributes vertically.","Preserve wrapping of attributes.","Preserve wrapping of attributes but align."],"description":"Wrap attributes."},"html.format.wrapAttributesIndentSize":{"type":["number","null"],"scope":"resource","default":null,"markdownDescription":"Indent wrapped attributes to after N characters. Use `null` to use the default indent size. Ignored if `#html.format.wrapAttributes#` is set to `aligned`."},"html.format.templating":{"type":"boolean","scope":"resource","default":false,"description":"Honor django, erb, handlebars and php templating language tags."},"html.format.unformattedContentDelimiter":{"type":"string","scope":"resource","default":"","markdownDescription":"Keep text content together between this string."},"html.suggest.html5":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether the built-in HTML language support suggests HTML5 tags, properties and values."},"html.validate.scripts":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether the built-in HTML language support validates embedded scripts."},"html.validate.styles":{"type":"boolean","scope":"resource","default":true,"description":"Controls whether the built-in HTML language support validates embedded styles."},"html.autoCreateQuotes":{"type":"boolean","scope":"resource","default":true,"description":"Enable/disable auto creation of quotes for HTML attribute assignment. The type of quotes can be configured by `#html.completion.attributeDefaultValue#`."},"html.autoClosingTags":{"type":"boolean","scope":"resource","default":true,"description":"Enable/disable autoclosing of HTML tags."},"html.hover.documentation":{"type":"boolean","scope":"resource","default":true,"description":"Show tag and attribute documentation in hover."},"html.hover.references":{"type":"boolean","scope":"resource","default":true,"description":"Show references to MDN in hover."},"html.mirrorCursorOnMatchingTag":{"type":"boolean","scope":"resource","default":false,"description":"Enable/disable mirroring cursor on matching HTML tag.","deprecationMessage":"Deprecated in favor of `editor.linkedEditing`"},"html.trace.server":{"type":"string","scope":"window","enum":["off","messages","verbose"],"default":"off","description":"Traces the communication between VS Code and the HTML language server."}}},"configurationDefaults":{"[html]":{"editor.suggest.insertMode":"replace"},"[handlebars]":{"editor.suggest.insertMode":"replace"}},"jsonValidation":[{"fileMatch":"*.html-data.json","url":"https://raw.githubusercontent.com/microsoft/vscode-html-languageservice/master/docs/customData.schema.json"},{"fileMatch":"package.json","url":"./schemas/package.schema.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/html-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.ini"},"manifest":{"name":"ini","displayName":"Ini Language Basics","description":"Provides syntax highlighting and bracket matching in Ini files.","version":"1.0.0","private":true,"publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin textmate/ini.tmbundle Syntaxes/Ini.plist ./syntaxes/ini.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"ini","extensions":[".ini"],"aliases":["Ini","ini"],"configuration":"./ini.language-configuration.json"},{"id":"properties","extensions":[".conf",".properties",".cfg",".directory",".gitattributes",".gitconfig",".gitmodules",".editorconfig",".repo"],"filenames":["gitconfig",".env"],"filenamePatterns":["**/.config/git/config","**/.git/config"],"aliases":["Properties","properties"],"configuration":"./properties.language-configuration.json"}],"grammars":[{"language":"ini","scopeName":"source.ini","path":"./syntaxes/ini.tmLanguage.json"},{"language":"properties","scopeName":"source.ini","path":"./syntaxes/ini.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/ini","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.ipynb"},"manifest":{"name":"ipynb","displayName":".ipynb Support","description":"Provides basic support for opening and reading Jupyter's .ipynb notebook files","publisher":"vscode","version":"1.0.0","license":"MIT","icon":"media/icon.png","engines":{"vscode":"^1.57.0"},"enabledApiProposals":["diffContentOptions"],"activationEvents":["onNotebook:jupyter-notebook","onNotebookSerializer:interactive","onNotebookSerializer:repl"],"extensionKind":["workspace","ui"],"main":"./dist/ipynbMain.node.js","browser":"./dist/browser/ipynbMain.browser.js","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"configuration":[{"properties":{"ipynb.pasteImagesAsAttachments.enabled":{"type":"boolean","scope":"resource","markdownDescription":"Enable/disable pasting of images into Markdown cells in ipynb notebook files. Pasted images are inserted as attachments to the cell.","default":true},"ipynb.experimental.serialization":{"type":"boolean","scope":"resource","markdownDescription":"Experimental feature to serialize the Jupyter notebook in a worker thread.","default":false,"tags":["experimental"]}}}],"commands":[{"command":"ipynb.newUntitledIpynb","title":"New Jupyter Notebook","shortTitle":"Jupyter Notebook","category":"Create"},{"command":"ipynb.openIpynbInNotebookEditor","title":"Open IPYNB File In Notebook Editor"},{"command":"ipynb.cleanInvalidImageAttachment","title":"Clean Invalid Image Attachment Reference"},{"command":"notebook.cellOutput.copy","title":"Copy Cell Output","category":"Notebook"},{"command":"notebook.cellOutput.openInTextEditor","title":"Open Cell Output in Text Editor","category":"Notebook"}],"notebooks":[{"type":"jupyter-notebook","displayName":"Jupyter Notebook","selector":[{"filenamePattern":"*.ipynb"}],"priority":"default"}],"notebookRenderer":[{"id":"vscode.markdown-it-cell-attachment-renderer","displayName":"Markdown-It ipynb Cell Attachment renderer","entrypoint":{"extends":"vscode.markdown-it-renderer","path":"./notebook-out/cellAttachmentRenderer.js"}}],"menus":{"file/newFile":[{"command":"ipynb.newUntitledIpynb","group":"notebook"}],"commandPalette":[{"command":"ipynb.newUntitledIpynb"},{"command":"ipynb.openIpynbInNotebookEditor","when":"false"},{"command":"ipynb.cleanInvalidImageAttachment","when":"false"},{"command":"notebook.cellOutput.copy","when":"notebookCellHasOutputs"},{"command":"notebook.cellOutput.openInTextEditor","when":"false"}],"webview/context":[{"command":"notebook.cellOutput.copy","when":"webviewId == 'notebook.output' && webviewSection == 'image'"},{"command":"notebook.cellOutput.copy","when":"webviewId == 'notebook.output' && webviewSection == 'text'"},{"command":"notebook.cellOutput.openInTextEditor","when":"webviewId == 'notebook.output' && webviewSection == 'text'"}]}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/ipynb","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.jake"},"manifest":{"name":"jake","publisher":"vscode","description":"Extension to add Jake capabilities to VS Code.","displayName":"Jake support for VS Code","icon":"images/cowboy_hat.png","version":"1.0.0","license":"MIT","engines":{"vscode":"*"},"categories":["Other"],"main":"./dist/main","activationEvents":["onTaskType:jake"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"contributes":{"configuration":{"id":"jake","type":"object","title":"Jake","properties":{"jake.autoDetect":{"scope":"application","type":"string","enum":["off","on"],"default":"off","description":"Controls enablement of Jake task detection. Jake task detection can cause files in any open workspace to be executed."}}},"taskDefinitions":[{"type":"jake","required":["task"],"properties":{"task":{"type":"string","description":"The Jake task to customize."},"file":{"type":"string","description":"The Jake file that provides the task. Can be omitted."}},"when":"shellExecutionSupported"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/jake","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.java"},"manifest":{"name":"java","displayName":"Java Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in Java files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin redhat-developer/vscode-java language-support/java/java.tmLanguage.json ./syntaxes/java.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"java","extensions":[".java",".jav"],"aliases":["Java","java"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"java","scopeName":"source.java","path":"./syntaxes/java.tmLanguage.json"}],"snippets":[{"language":"java","path":"./snippets/java.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/java","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.javascript"},"manifest":{"name":"javascript","displayName":"JavaScript Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in JavaScript files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"categories":["Programming Languages"],"contributes":{"configurationDefaults":{"[javascript]":{"editor.maxTokenizationLineLength":2500}},"languages":[{"id":"javascriptreact","aliases":["JavaScript JSX","JavaScript React","jsx"],"extensions":[".jsx"],"configuration":"./javascript-language-configuration.json"},{"id":"javascript","aliases":["JavaScript","javascript","js"],"extensions":[".js",".es6",".mjs",".cjs",".pac"],"filenames":["jakefile"],"firstLine":"^#!.*\\bnode","mimetypes":["text/javascript"],"configuration":"./javascript-language-configuration.json"},{"id":"jsx-tags","aliases":[],"configuration":"./tags-language-configuration.json"}],"grammars":[{"language":"javascriptreact","scopeName":"source.js.jsx","path":"./syntaxes/JavaScriptReact.tmLanguage.json","embeddedLanguages":{"meta.tag.js":"jsx-tags","meta.tag.without-attributes.js":"jsx-tags","meta.tag.attributes.js.jsx":"javascriptreact","meta.embedded.expression.js":"javascriptreact"},"tokenTypes":{"meta.template.expression":"other","meta.template.expression string":"string","meta.template.expression comment":"comment","entity.name.type.instance.jsdoc":"other","entity.name.function.tagged-template":"other","meta.import string.quoted":"other","variable.other.jsdoc":"other"}},{"language":"javascript","scopeName":"source.js","path":"./syntaxes/JavaScript.tmLanguage.json","embeddedLanguages":{"meta.tag.js":"jsx-tags","meta.tag.without-attributes.js":"jsx-tags","meta.tag.attributes.js":"javascript","meta.embedded.expression.js":"javascript"},"tokenTypes":{"meta.template.expression":"other","meta.template.expression string":"string","meta.template.expression comment":"comment","entity.name.type.instance.jsdoc":"other","entity.name.function.tagged-template":"other","meta.import string.quoted":"other","variable.other.jsdoc":"other"}},{"scopeName":"source.js.regexp","path":"./syntaxes/Regular Expressions (JavaScript).tmLanguage"}],"semanticTokenScopes":[{"language":"javascript","scopes":{"property":["variable.other.property.js"],"property.readonly":["variable.other.constant.property.js"],"variable":["variable.other.readwrite.js"],"variable.readonly":["variable.other.constant.object.js"],"function":["entity.name.function.js"],"namespace":["entity.name.type.module.js"],"variable.defaultLibrary":["support.variable.js"],"function.defaultLibrary":["support.function.js"]}},{"language":"javascriptreact","scopes":{"property":["variable.other.property.jsx"],"property.readonly":["variable.other.constant.property.jsx"],"variable":["variable.other.readwrite.jsx"],"variable.readonly":["variable.other.constant.object.jsx"],"function":["entity.name.function.jsx"],"namespace":["entity.name.type.module.jsx"],"variable.defaultLibrary":["support.variable.js"],"function.defaultLibrary":["support.function.js"]}}],"snippets":[{"language":"javascript","path":"./snippets/javascript.code-snippets"},{"language":"javascriptreact","path":"./snippets/javascript.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/javascript","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.json"},"manifest":{"name":"json","displayName":"JSON Language Basics","description":"Provides syntax highlighting & bracket matching in JSON files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ./build/update-grammars.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"json","aliases":["JSON","json"],"extensions":[".json",".bowerrc",".jscsrc",".webmanifest",".js.map",".css.map",".ts.map",".har",".jslintrc",".jsonld",".geojson",".ipynb",".vuerc"],"filenames":["composer.lock",".watchmanconfig"],"mimetypes":["application/json","application/manifest+json"],"configuration":"./language-configuration.json"},{"id":"jsonc","aliases":["JSON with Comments"],"extensions":[".jsonc",".eslintrc",".eslintrc.json",".jsfmtrc",".jshintrc",".swcrc",".hintrc",".babelrc"],"filenames":["babel.config.json","bun.lock",".babelrc.json",".ember-cli","typedoc.json"],"configuration":"./language-configuration.json"},{"id":"jsonl","aliases":["JSON Lines"],"extensions":[".jsonl",".ndjson"],"filenames":[],"configuration":"./language-configuration.json"},{"id":"snippets","aliases":["Code Snippets"],"extensions":[".code-snippets"],"filenamePatterns":["**/User/snippets/*.json","**/User/profiles/*/snippets/*.json"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"json","scopeName":"source.json","path":"./syntaxes/JSON.tmLanguage.json"},{"language":"jsonc","scopeName":"source.json.comments","path":"./syntaxes/JSONC.tmLanguage.json"},{"language":"jsonl","scopeName":"source.json.lines","path":"./syntaxes/JSONL.tmLanguage.json"},{"language":"snippets","scopeName":"source.json.comments.snippets","path":"./syntaxes/snippets.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/json","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.json-language-features"},"manifest":{"name":"json-language-features","displayName":"JSON Language Features","description":"Provides rich language support for JSON files.","version":"1.0.0","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.77.0"},"enabledApiProposals":["extensionsAny"],"icon":"icons/json.png","activationEvents":["onLanguage:json","onLanguage:jsonc","onLanguage:snippets"],"main":"./client/dist/node/jsonClientMain","browser":"./client/dist/browser/jsonClientMain","capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"categories":["Programming Languages"],"contributes":{"configuration":{"id":"json","order":20,"type":"object","title":"JSON","properties":{"json.schemas":{"type":"array","scope":"resource","description":"Associate schemas to JSON files in the current project.","items":{"type":"object","default":{"fileMatch":["/myfile"],"url":"schemaURL"},"properties":{"url":{"type":"string","default":"/user.schema.json","description":"A URL or absolute file path to a schema. Can be a relative path (starting with './') in workspace and workspace folder settings."},"fileMatch":{"type":"array","items":{"type":"string","default":"MyFile.json","description":"A file pattern that can contain '*' and '**' to match against when resolving JSON files to schemas. When beginning with '!', it defines an exclusion pattern."},"minItems":1,"description":"An array of file patterns to match against when resolving JSON files to schemas. `*` and '**' can be used as a wildcard. Exclusion patterns can also be defined and start with '!'. A file matches when there is at least one matching pattern and the last matching pattern is not an exclusion pattern."},"schema":{"$ref":"http://json-schema.org/draft-07/schema#","description":"The schema definition for the given URL. The schema only needs to be provided to avoid accesses to the schema URL."}}}},"json.validate.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable JSON validation."},"json.format.enable":{"type":"boolean","scope":"window","default":true,"description":"Enable/disable default JSON formatter"},"json.format.keepLines":{"type":"boolean","scope":"window","default":false,"description":"Keep all existing new lines when formatting."},"json.trace.server":{"type":"string","scope":"window","enum":["off","messages","verbose"],"default":"off","description":"Traces the communication between VS Code and the JSON language server."},"json.colorDecorators.enable":{"type":"boolean","scope":"window","default":true,"description":"Enables or disables color decorators","deprecationMessage":"The setting `json.colorDecorators.enable` has been deprecated in favor of `editor.colorDecorators`."},"json.maxItemsComputed":{"type":"number","default":5000,"description":"The maximum number of outline symbols and folding regions computed (limited for performance reasons)."},"json.schemaDownload.enable":{"type":"boolean","default":true,"description":"When enabled, JSON schemas can be fetched from http and https locations.","tags":["usesOnlineServices"]}}},"configurationDefaults":{"[json]":{"editor.quickSuggestions":{"strings":true},"editor.suggest.insertMode":"replace"},"[jsonc]":{"editor.quickSuggestions":{"strings":true},"editor.suggest.insertMode":"replace"},"[snippets]":{"editor.quickSuggestions":{"strings":true},"editor.suggest.insertMode":"replace"}},"jsonValidation":[{"fileMatch":"*.schema.json","url":"http://json-schema.org/draft-07/schema#"}],"commands":[{"command":"json.clearCache","title":"Clear Schema Cache","category":"JSON"},{"command":"json.sort","title":"Sort Document","category":"JSON"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/json-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.julia"},"manifest":{"name":"julia","displayName":"Julia Language Basics","description":"Provides syntax highlighting & bracket matching in Julia files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin JuliaEditorSupport/atom-language-julia grammars/julia_vscode.json ./syntaxes/julia.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"julia","aliases":["Julia","julia"],"extensions":[".jl"],"firstLine":"^#!\\s*/.*\\bjulia[0-9.-]*\\b","configuration":"./language-configuration.json"},{"id":"juliamarkdown","aliases":["Julia Markdown","juliamarkdown"],"extensions":[".jmd"]}],"grammars":[{"language":"julia","scopeName":"source.julia","path":"./syntaxes/julia.tmLanguage.json","embeddedLanguages":{"meta.embedded.inline.cpp":"cpp","meta.embedded.inline.javascript":"javascript","meta.embedded.inline.python":"python","meta.embedded.inline.r":"r","meta.embedded.inline.sql":"sql"}}],"configurationDefaults":{"[julia]":{"editor.defaultColorDecorators":false}}}},"location":{"$mid":1,"path":"/ide/extensions/julia","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.latex"},"manifest":{"name":"latex","displayName":"LaTeX Language Basics","description":"Provides syntax highlighting and bracket matching for TeX, LaTeX and BibTeX.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammars.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"tex","aliases":["TeX","tex"],"extensions":[".sty",".cls",".bbx",".cbx"],"configuration":"latex-language-configuration.json"},{"id":"latex","aliases":["LaTeX","latex"],"extensions":[".tex",".ltx",".ctx"],"configuration":"latex-language-configuration.json"},{"id":"bibtex","aliases":["BibTeX","bibtex"],"extensions":[".bib"]},{"id":"cpp_embedded_latex","configuration":"latex-cpp-embedded-language-configuration.json","aliases":[]},{"id":"markdown_latex_combined","configuration":"markdown-latex-combined-language-configuration.json","aliases":[]}],"grammars":[{"language":"tex","scopeName":"text.tex","path":"./syntaxes/TeX.tmLanguage.json"},{"language":"latex","scopeName":"text.tex.latex","path":"./syntaxes/LaTeX.tmLanguage.json","embeddedLanguages":{"source.cpp":"cpp_embedded_latex","source.css":"css","text.html":"html","source.java":"java","source.js":"javascript","source.julia":"julia","source.lua":"lua","source.python":"python","source.ruby":"ruby","source.ts":"typescript","text.xml":"xml","source.yaml":"yaml","meta.embedded.markdown_latex_combined":"markdown_latex_combined"}},{"language":"bibtex","scopeName":"text.bibtex","path":"./syntaxes/Bibtex.tmLanguage.json"},{"language":"markdown_latex_combined","scopeName":"text.tex.markdown_latex_combined","path":"./syntaxes/markdown-latex-combined.tmLanguage.json"},{"language":"cpp_embedded_latex","scopeName":"source.cpp.embedded.latex","path":"./syntaxes/cpp-grammar-bailout.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/latex","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.less"},"manifest":{"name":"less","displayName":"Less Language Basics","description":"Provides syntax highlighting, bracket matching and folding in Less files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammar.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"less","aliases":["Less","less"],"extensions":[".less"],"mimetypes":["text/x-less","text/less"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"less","scopeName":"source.css.less","path":"./syntaxes/less.tmLanguage.json"}],"problemMatchers":[{"name":"lessc","label":"Lessc compiler","owner":"lessc","source":"less","fileLocation":"absolute","pattern":{"regexp":"(.*)\\sin\\s(.*)\\son line\\s(\\d+),\\scolumn\\s(\\d+)","message":1,"file":2,"line":3,"column":4}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/less","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.log"},"manifest":{"name":"log","displayName":"Log","description":"Provides syntax highlighting for files with .log extension.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin emilast/vscode-logfile-highlighter syntaxes/log.tmLanguage ./syntaxes/log.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"log","extensions":[".log","*.log.?"],"aliases":["Log"]}],"grammars":[{"language":"log","scopeName":"text.log","path":"./syntaxes/log.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/log","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.lua"},"manifest":{"name":"lua","displayName":"Lua Language Basics","description":"Provides syntax highlighting and bracket matching in Lua files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin sumneko/lua.tmbundle Syntaxes/Lua.plist ./syntaxes/lua.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"lua","extensions":[".lua"],"aliases":["Lua","lua"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"lua","scopeName":"source.lua","path":"./syntaxes/lua.tmLanguage.json","tokenTypes":{"comment.line.double-dash.doc.lua":"other"}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/lua","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.make"},"manifest":{"name":"make","displayName":"Make Language Basics","description":"Provides syntax highlighting and bracket matching in Make files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin fadeevab/make.tmbundle Syntaxes/Makefile.plist ./syntaxes/make.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"makefile","aliases":["Makefile","makefile"],"extensions":[".mak",".mk"],"filenames":["Makefile","makefile","GNUmakefile","OCamlMakefile"],"firstLine":"^#!\\s*/usr/bin/make","configuration":"./language-configuration.json"}],"grammars":[{"language":"makefile","scopeName":"source.makefile","path":"./syntaxes/make.tmLanguage.json","tokenTypes":{"string.interpolated":"other"}}],"configurationDefaults":{"[makefile]":{"editor.insertSpaces":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/make","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.markdown"},"manifest":{"name":"markdown","displayName":"Markdown Language Basics","description":"Provides snippets and syntax highlighting for Markdown.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.20.0"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"markdown","aliases":["Markdown","markdown"],"extensions":[".md",".mkd",".mdwn",".mdown",".markdown",".markdn",".mdtxt",".mdtext",".workbook"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"markdown","scopeName":"text.html.markdown","path":"./syntaxes/markdown.tmLanguage.json","embeddedLanguages":{"meta.embedded.block.html":"html","source.js":"javascript","source.css":"css","meta.embedded.block.frontmatter":"yaml","meta.embedded.block.css":"css","meta.embedded.block.ini":"ini","meta.embedded.block.java":"java","meta.embedded.block.lua":"lua","meta.embedded.block.makefile":"makefile","meta.embedded.block.perl":"perl","meta.embedded.block.r":"r","meta.embedded.block.ruby":"ruby","meta.embedded.block.php":"php","meta.embedded.block.sql":"sql","meta.embedded.block.vs_net":"vs_net","meta.embedded.block.xml":"xml","meta.embedded.block.xsl":"xsl","meta.embedded.block.yaml":"yaml","meta.embedded.block.dosbatch":"dosbatch","meta.embedded.block.clojure":"clojure","meta.embedded.block.coffee":"coffee","meta.embedded.block.c":"c","meta.embedded.block.cpp":"cpp","meta.embedded.block.diff":"diff","meta.embedded.block.dockerfile":"dockerfile","meta.embedded.block.go":"go","meta.embedded.block.groovy":"groovy","meta.embedded.block.pug":"jade","meta.embedded.block.javascript":"javascript","meta.embedded.block.json":"json","meta.embedded.block.jsonc":"jsonc","meta.embedded.block.latex":"latex","meta.embedded.block.less":"less","meta.embedded.block.objc":"objc","meta.embedded.block.scss":"scss","meta.embedded.block.perl6":"perl6","meta.embedded.block.powershell":"powershell","meta.embedded.block.python":"python","meta.embedded.block.rust":"rust","meta.embedded.block.scala":"scala","meta.embedded.block.shellscript":"shellscript","meta.embedded.block.typescript":"typescript","meta.embedded.block.typescriptreact":"typescriptreact","meta.embedded.block.csharp":"csharp","meta.embedded.block.fsharp":"fsharp"},"unbalancedBracketScopes":["markup.underline.link.markdown","punctuation.definition.list.begin.markdown"]}],"snippets":[{"language":"markdown","path":"./snippets/markdown.code-snippets"}],"configurationDefaults":{"[markdown]":{"editor.unicodeHighlight.ambiguousCharacters":false,"editor.unicodeHighlight.invisibleCharacters":false,"diffEditor.ignoreTrimWhitespace":false}}},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin microsoft/vscode-markdown-tm-grammar syntaxes/markdown.tmLanguage ./syntaxes/markdown.tmLanguage.json"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/markdown-basics","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.markdown-language-features"},"manifest":{"name":"markdown-language-features","displayName":"Markdown Language Features","description":"Provides rich language support for Markdown.","version":"1.0.0","icon":"icon.png","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.70.0"},"main":"./dist/extension","browser":"./dist/browser/extension","categories":["Programming Languages"],"activationEvents":["onLanguage:markdown","onCommand:markdown.api.render","onCommand:markdown.api.reloadPlugins","onWebviewPanel:markdown.preview"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":"limited","description":"Required for loading styles configured in the workspace.","restrictedConfigurations":["markdown.styles"]}},"contributes":{"notebookRenderer":[{"id":"vscode.markdown-it-renderer","displayName":"Markdown it renderer","entrypoint":"./notebook-out/index.js","mimeTypes":["text/markdown","text/latex","text/x-css","text/x-html","text/x-json","text/x-typescript","text/x-abap","text/x-apex","text/x-azcli","text/x-bat","text/x-cameligo","text/x-clojure","text/x-coffee","text/x-cpp","text/x-csharp","text/x-csp","text/x-css","text/x-dart","text/x-dockerfile","text/x-ecl","text/x-fsharp","text/x-go","text/x-graphql","text/x-handlebars","text/x-hcl","text/x-html","text/x-ini","text/x-java","text/x-javascript","text/x-julia","text/x-kotlin","text/x-less","text/x-lexon","text/x-lua","text/x-m3","text/x-markdown","text/x-mips","text/x-msdax","text/x-mysql","text/x-objective-c/objective","text/x-pascal","text/x-pascaligo","text/x-perl","text/x-pgsql","text/x-php","text/x-postiats","text/x-powerquery","text/x-powershell","text/x-pug","text/x-python","text/x-r","text/x-razor","text/x-redis","text/x-redshift","text/x-restructuredtext","text/x-ruby","text/x-rust","text/x-sb","text/x-scala","text/x-scheme","text/x-scss","text/x-shell","text/x-solidity","text/x-sophia","text/x-sql","text/x-st","text/x-swift","text/x-systemverilog","text/x-tcl","text/x-twig","text/x-typescript","text/x-vb","text/x-xml","text/x-yaml","application/json"]}],"commands":[{"command":"_markdown.copyImage","title":"Copy Image","category":"Markdown"},{"command":"_markdown.openImage","title":"Open Image","category":"Markdown"},{"command":"markdown.showPreview","title":"Open Preview","category":"Markdown","icon":{"light":"./media/preview-light.svg","dark":"./media/preview-dark.svg"}},{"command":"markdown.showPreviewToSide","title":"Open Preview to the Side","category":"Markdown","icon":"$(open-preview)"},{"command":"markdown.showLockedPreviewToSide","title":"Open Locked Preview to the Side","category":"Markdown","icon":"$(open-preview)"},{"command":"markdown.showSource","title":"Show Source","category":"Markdown","icon":"$(go-to-file)"},{"command":"markdown.showPreviewSecuritySelector","title":"Change Preview Security Settings","category":"Markdown"},{"command":"markdown.preview.refresh","title":"Refresh Preview","category":"Markdown"},{"command":"markdown.preview.toggleLock","title":"Toggle Preview Locking","category":"Markdown"},{"command":"markdown.findAllFileReferences","title":"Find File References","category":"Markdown"},{"command":"markdown.editor.insertLinkFromWorkspace","title":"Insert Link to File in Workspace","category":"Markdown","enablement":"editorLangId == markdown && !activeEditorIsReadonly"},{"command":"markdown.editor.insertImageFromWorkspace","title":"Insert Image from Workspace","category":"Markdown","enablement":"editorLangId == markdown && !activeEditorIsReadonly"}],"menus":{"webview/context":[{"command":"_markdown.copyImage","when":"webviewId == 'markdown.preview' && (webviewSection == 'image' || webviewSection == 'localImage')"},{"command":"_markdown.openImage","when":"webviewId == 'markdown.preview' && webviewSection == 'localImage'"}],"editor/title":[{"command":"markdown.showPreviewToSide","when":"editorLangId == markdown && !notebookEditorFocused && !hasCustomMarkdownPreview","alt":"markdown.showPreview","group":"navigation"},{"command":"markdown.showSource","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'","group":"navigation"},{"command":"markdown.preview.refresh","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'","group":"1_markdown"},{"command":"markdown.preview.toggleLock","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'","group":"1_markdown"},{"command":"markdown.showPreviewSecuritySelector","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'","group":"1_markdown"}],"explorer/context":[{"command":"markdown.showPreview","when":"resourceLangId == markdown && !hasCustomMarkdownPreview","group":"navigation"},{"command":"markdown.findAllFileReferences","when":"resourceLangId == markdown","group":"4_search"}],"editor/title/context":[{"command":"markdown.showPreview","when":"resourceLangId == markdown && !hasCustomMarkdownPreview","group":"1_open"},{"command":"markdown.findAllFileReferences","when":"resourceLangId == markdown"}],"commandPalette":[{"command":"_markdown.openImage","when":"false"},{"command":"_markdown.copyImage","when":"false"},{"command":"markdown.showPreview","when":"editorLangId == markdown && !notebookEditorFocused","group":"navigation"},{"command":"markdown.showPreviewToSide","when":"editorLangId == markdown && !notebookEditorFocused","group":"navigation"},{"command":"markdown.showLockedPreviewToSide","when":"editorLangId == markdown && !notebookEditorFocused","group":"navigation"},{"command":"markdown.showSource","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'","group":"navigation"},{"command":"markdown.showPreviewSecuritySelector","when":"editorLangId == markdown && !notebookEditorFocused"},{"command":"markdown.showPreviewSecuritySelector","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'"},{"command":"markdown.preview.toggleLock","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'"},{"command":"markdown.preview.refresh","when":"editorLangId == markdown && !notebookEditorFocused"},{"command":"markdown.preview.refresh","when":"activeWebviewPanelId == 'markdown.preview' || activeCustomEditorId == 'vscode.markdown.preview.editor'"},{"command":"markdown.findAllFileReferences","when":"editorLangId == markdown"}]},"keybindings":[{"command":"markdown.showPreview","key":"shift+ctrl+v","mac":"shift+cmd+v","when":"editorLangId == markdown && !notebookEditorFocused"},{"command":"markdown.showPreviewToSide","key":"ctrl+k v","mac":"cmd+k v","when":"editorLangId == markdown && !notebookEditorFocused"}],"configuration":{"type":"object","title":"Markdown","order":20,"properties":{"markdown.styles":{"type":"array","items":{"type":"string"},"default":[],"description":"A list of URLs or local paths to CSS style sheets to use from the Markdown preview. Relative paths are interpreted relative to the folder open in the Explorer. If there is no open folder, they are interpreted relative to the location of the Markdown file. All '\\' need to be written as '\\\\'.","scope":"resource"},"markdown.preview.breaks":{"type":"boolean","default":false,"markdownDescription":"Sets how line-breaks are rendered in the Markdown preview. Setting it to `true` creates a `<br>` for newlines inside paragraphs.","scope":"resource"},"markdown.preview.linkify":{"type":"boolean","default":true,"description":"Convert URL-like text to links in the Markdown preview.","scope":"resource"},"markdown.preview.typographer":{"type":"boolean","default":false,"description":"Enable some language-neutral replacement and quotes beautification in the Markdown preview.","scope":"resource"},"markdown.preview.fontFamily":{"type":"string","default":"-apple-system, BlinkMacSystemFont, 'Segoe WPC', 'Segoe UI', system-ui, 'Ubuntu', 'Droid Sans', sans-serif","description":"Controls the font family used in the Markdown preview.","scope":"resource"},"markdown.preview.fontSize":{"type":"number","default":14,"description":"Controls the font size in pixels used in the Markdown preview.","scope":"resource"},"markdown.preview.lineHeight":{"type":"number","default":1.6,"description":"Controls the line height used in the Markdown preview. This number is relative to the font size.","scope":"resource"},"markdown.preview.scrollPreviewWithEditor":{"type":"boolean","default":true,"description":"When a Markdown editor is scrolled, update the view of the preview.","scope":"resource"},"markdown.preview.markEditorSelection":{"type":"boolean","default":true,"description":"Mark the current editor selection in the Markdown preview.","scope":"resource"},"markdown.preview.scrollEditorWithPreview":{"type":"boolean","default":true,"description":"When a Markdown preview is scrolled, update the view of the editor.","scope":"resource"},"markdown.preview.doubleClickToSwitchToEditor":{"type":"boolean","default":true,"description":"Double-click in the Markdown preview to switch to the editor.","scope":"resource"},"markdown.preview.openMarkdownLinks":{"type":"string","default":"inPreview","description":"Controls how links to other Markdown files in the Markdown preview should be opened.","scope":"resource","enum":["inPreview","inEditor"],"enumDescriptions":["Try to open links in the Markdown preview.","Try to open links in the editor."]},"markdown.links.openLocation":{"type":"string","default":"currentGroup","description":"Controls where links in Markdown files should be opened.","scope":"resource","enum":["currentGroup","beside"],"enumDescriptions":["Open links in the active editor group.","Open links beside the active editor."]},"markdown.suggest.paths.enabled":{"type":"boolean","default":true,"description":"Enable path suggestions while writing links in Markdown files.","scope":"resource"},"markdown.suggest.paths.includeWorkspaceHeaderCompletions":{"type":"string","default":"onDoubleHash","scope":"resource","markdownDescription":"Enable suggestions for headers in other Markdown files in the current workspace. Accepting one of these suggestions inserts the full path to header in that file, for example: `[link text](/path/to/file.md#header)`.","enum":["never","onDoubleHash","onSingleOrDoubleHash"],"markdownEnumDescriptions":["Disable workspace header suggestions.","Enable workspace header suggestions after typing `##` in a path, for example: `[link text](##`.","Enable workspace header suggestions after typing either `##` or `#` in a path, for example: `[link text](#` or `[link text](##`."]},"markdown.trace.extension":{"type":"string","enum":["off","verbose"],"default":"off","description":"Enable debug logging for the Markdown extension.","scope":"window"},"markdown.trace.server":{"type":"string","scope":"window","enum":["off","messages","verbose"],"default":"off","description":"Traces the communication between VS Code and the Markdown language server."},"markdown.server.log":{"type":"string","scope":"window","enum":["off","debug","trace"],"default":"off","description":"Controls the logging level of the Markdown language server."},"markdown.editor.drop.enabled":{"type":"string","scope":"resource","markdownDescription":"Enable dropping files into a Markdown editor while holding Shift. Requires enabling `#editor.dropIntoEditor.enabled#`.","default":"smart","enum":["always","smart","never"],"markdownEnumDescriptions":["Always insert Markdown links.","Smartly create Markdown links by default when not dropping into a code block or other special element. Use the drop widget to switch between pasting as plain text or as Markdown links.","Never create Markdown links."]},"markdown.editor.drop.copyIntoWorkspace":{"type":"string","markdownDescription":"Controls if files outside of the workspace that are dropped into a Markdown editor should be copied into the workspace.\n\nUse `#markdown.copyFiles.destination#` to configure where copied dropped files should be created","default":"mediaFiles","enum":["mediaFiles","never"],"markdownEnumDescriptions":["Try to copy external image and video files into the workspace.","Do not copy external files into the workspace."]},"markdown.editor.filePaste.enabled":{"type":"string","scope":"resource","markdownDescription":"Enable pasting files into a Markdown editor to create Markdown links. Requires enabling `#editor.pasteAs.enabled#`.","default":"smart","enum":["always","smart","never"],"markdownEnumDescriptions":["Always insert Markdown links.","Smartly create Markdown links by default when not pasting into a code block or other special element. Use the paste widget to switch between pasting as plain text or as Markdown links.","Never create Markdown links."]},"markdown.editor.filePaste.copyIntoWorkspace":{"type":"string","markdownDescription":"Controls if files outside of the workspace that are pasted into a Markdown editor should be copied into the workspace.\n\nUse `#markdown.copyFiles.destination#` to configure where copied files should be created.","default":"mediaFiles","enum":["mediaFiles","never"],"markdownEnumDescriptions":["Try to copy external image and video files into the workspace.","Do not copy external files into the workspace."]},"markdown.editor.filePaste.videoSnippet":{"type":"string","markdownDescription":"Snippet used when adding videos to Markdown. This snippet can use the following variables:\n- `${src}` — The resolved path of the video file.\n- `${title}` — The title used for the video. A snippet placeholder will automatically be created for this variable.","default":"<video controls src=\"${src}\" title=\"${title}\"></video>"},"markdown.editor.filePaste.audioSnippet":{"type":"string","markdownDescription":"Snippet used when adding audio to Markdown. This snippet can use the following variables:\n- `${src}` — The resolved path of the audio  file.\n- `${title}` — The title used for the audio. A snippet placeholder will automatically be created for this variable.","default":"<audio controls src=\"${src}\" title=\"${title}\"></audio>"},"markdown.editor.pasteUrlAsFormattedLink.enabled":{"type":"string","scope":"resource","markdownDescription":"Controls if Markdown links are created when URLs are pasted into a Markdown editor. Requires enabling `#editor.pasteAs.enabled#`.","default":"smartWithSelection","enum":["always","smart","smartWithSelection","never"],"markdownEnumDescriptions":["Always insert Markdown links.","Smartly create Markdown links by default when not pasting into a code block or other special element. Use the paste widget to switch between pasting as plain text or as Markdown links.","Smartly create Markdown links by default when you have selected text and are not pasting into a code block or other special element. Use the paste widget to switch between pasting as plain text or as Markdown links.","Never create Markdown links."]},"markdown.validate.enabled":{"type":"boolean","scope":"resource","description":"Enable all error reporting in Markdown files.","default":false},"markdown.validate.referenceLinks.enabled":{"type":"string","scope":"resource","markdownDescription":"Validate reference links in Markdown files, for example: `[link][ref]`. Requires enabling `#markdown.validate.enabled#`.","default":"warning","enum":["ignore","warning","error"]},"markdown.validate.fragmentLinks.enabled":{"type":"string","scope":"resource","markdownDescription":"Validate fragment links to headers in the current Markdown file, for example: `[link](#header)`. Requires enabling `#markdown.validate.enabled#`.","default":"warning","enum":["ignore","warning","error"]},"markdown.validate.fileLinks.enabled":{"type":"string","scope":"resource","markdownDescription":"Validate links to other files in Markdown files, for example `[link](/path/to/file.md)`. This checks that the target files exists. Requires enabling `#markdown.validate.enabled#`.","default":"warning","enum":["ignore","warning","error"]},"markdown.validate.fileLinks.markdownFragmentLinks":{"type":"string","scope":"resource","markdownDescription":"Validate the fragment part of links to headers in other files in Markdown files, for example: `[link](/path/to/file.md#header)`. Inherits the setting value from `#markdown.validate.fragmentLinks.enabled#` by default.","default":"inherit","enum":["inherit","ignore","warning","error"]},"markdown.validate.ignoredLinks":{"type":"array","scope":"resource","markdownDescription":"Configure links that should not be validated. For example adding `/about` would not validate the link `[about](/about)`, while the glob `/assets/**/*.svg` would let you skip validation for any link to `.svg` files under the `assets` directory.","items":{"type":"string"}},"markdown.validate.unusedLinkDefinitions.enabled":{"type":"string","scope":"resource","markdownDescription":"Validate link definitions that are unused in the current file.","default":"hint","enum":["ignore","hint","warning","error"]},"markdown.validate.duplicateLinkDefinitions.enabled":{"type":"string","scope":"resource","markdownDescription":"Validate duplicated definitions in the current file.","default":"warning","enum":["ignore","warning","error"]},"markdown.updateLinksOnFileMove.enabled":{"type":"string","enum":["prompt","always","never"],"markdownEnumDescriptions":["Prompt on each file move.","Always update links automatically.","Never try to update link and don't prompt."],"default":"never","markdownDescription":"Try to update links in Markdown files when a file is renamed/moved in the workspace. Use `#markdown.updateLinksOnFileMove.include#` to configure which files trigger link updates.","scope":"window"},"markdown.updateLinksOnFileMove.include":{"type":"array","markdownDescription":"Glob patterns that specifies files that trigger automatic link updates. See `#markdown.updateLinksOnFileMove.enabled#` for details about this feature.","scope":"window","items":{"type":"string","description":"The glob pattern to match file paths against. Set to true to enable the pattern."},"default":["**/*.{md,mkd,mdwn,mdown,markdown,markdn,mdtxt,mdtext,workbook}","**/*.{jpg,jpe,jpeg,png,bmp,gif,ico,webp,avif,tiff,svg,mp4}"]},"markdown.updateLinksOnFileMove.enableForDirectories":{"type":"boolean","default":true,"description":"Enable updating links when a directory is moved or renamed in the workspace.","scope":"window"},"markdown.occurrencesHighlight.enabled":{"type":"boolean","default":false,"description":"Enable highlighting link occurrences in the current document.","scope":"resource"},"markdown.copyFiles.destination":{"type":"object","markdownDescription":"Configures the path and file name of files created by copy/paste or drag and drop. This is a map of globs that match against a Markdown document path to the destination path where the new file should be created.\n\nThe destination path may use the following variables:\n\n- `${documentDirName}` — Absolute parent directory path of the Markdown document, e.g. `/Users/me/myProject/docs`.\n- `${documentRelativeDirName}` — Relative parent directory path of the Markdown document, e.g. `docs`. This is the same as `${documentDirName}` if the file is not part of a workspace.\n- `${documentFileName}` — The full filename of the Markdown document, e.g. `README.md`.\n- `${documentBaseName}` — The basename of the Markdown document, e.g. `README`.\n- `${documentExtName}` — The extension of the Markdown document, e.g. `md`.\n- `${documentFilePath}` — Absolute path of the Markdown document, e.g. `/Users/me/myProject/docs/README.md`.\n- `${documentRelativeFilePath}` — Relative path of the Markdown document, e.g. `docs/README.md`. This is the same as `${documentFilePath}` if the file is not part of a workspace.\n- `${documentWorkspaceFolder}` — The workspace folder for the Markdown document, e.g. `/Users/me/myProject`. This is the same as `${documentDirName}` if the file is not part of a workspace.\n- `${fileName}` — The file name of the dropped file, e.g. `image.png`.\n- `${fileExtName}` — The extension of the dropped file, e.g. `png`.\n- `${unixTime}` — The current Unix timestamp in milliseconds.","additionalProperties":{"type":"string"}},"markdown.copyFiles.overwriteBehavior":{"type":"string","markdownDescription":"Controls if files created by drop or paste should overwrite existing files.","default":"nameIncrementally","enum":["nameIncrementally","overwrite"],"markdownEnumDescriptions":["If a file with the same name already exists, append a number to the file name, for example: `image.png` becomes `image-1.png`.","If a file with the same name already exists, overwrite it."]},"markdown.preferredMdPathExtensionStyle":{"type":"string","default":"auto","markdownDescription":"Controls if file extensions (for example `.md`) are added or not for links to Markdown files. This setting is used when file paths are added by tooling such as path completions or file renames.","enum":["auto","includeExtension","removeExtension"],"markdownEnumDescriptions":["For existing paths, try to maintain the file extension style. For new paths, add file extensions.","Prefer including the file extension. For example, path completions to a file named `file.md` will insert `file.md`.","Prefer removing the file extension. For example, path completions to a file named `file.md` will insert `file` without the `.md`."]},"markdown.editor.updateLinksOnPaste.enabled":{"type":"boolean","markdownDescription":"Enable/disable a paste option that updates links and reference in text that is copied and pasted between Markdown editors.\n\nTo use this feature, after pasting text that contains updatable links, just click on the Paste Widget and select `Paste and update pasted links`.","scope":"resource","default":true}}},"configurationDefaults":{"[markdown]":{"editor.wordWrap":"on","editor.quickSuggestions":{"comments":"off","strings":"off","other":"off"}}},"jsonValidation":[{"fileMatch":"package.json","url":"./schemas/package.schema.json"}],"markdown.previewStyles":["./media/markdown.css","./media/highlight.css"],"markdown.previewScripts":["./media/index.js"],"customEditors":[{"viewType":"vscode.markdown.preview.editor","displayName":"Markdown Preview","priority":"option","selector":[{"filenamePattern":"*.md"}]}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/markdown-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.markdown-math"},"manifest":{"name":"markdown-math","displayName":"Markdown Math","description":"Adds math support to Markdown in notebooks.","version":"1.0.0","icon":"icon.png","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.54.0"},"categories":["Other","Programming Languages"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"main":"./dist/extension","browser":"./dist/browser/extension","activationEvents":[],"contributes":{"languages":[{"id":"markdown-math","aliases":[]}],"grammars":[{"language":"markdown-math","scopeName":"text.html.markdown.math","path":"./syntaxes/md-math.tmLanguage.json"},{"scopeName":"markdown.math.block","path":"./syntaxes/md-math-block.tmLanguage.json","injectTo":["text.html.markdown"],"embeddedLanguages":{"meta.embedded.math.markdown":"latex"}},{"scopeName":"markdown.math.inline","path":"./syntaxes/md-math-inline.tmLanguage.json","injectTo":["text.html.markdown"],"embeddedLanguages":{"meta.embedded.math.markdown":"latex","punctuation.definition.math.end.markdown":"latex"}},{"scopeName":"markdown.math.codeblock","path":"./syntaxes/md-math-fence.tmLanguage.json","injectTo":["text.html.markdown"],"embeddedLanguages":{"meta.embedded.math.markdown":"latex"}}],"notebookRenderer":[{"id":"vscode.markdown-it-katex-extension","displayName":"Markdown it KaTeX renderer","entrypoint":{"extends":"vscode.markdown-it-renderer","path":"./notebook-out/katex.js"}}],"markdown.markdownItPlugins":true,"markdown.previewStyles":["./notebook-out/katex.min.css","./preview-styles/index.css"],"configuration":[{"title":"Markdown Math","properties":{"markdown.math.enabled":{"type":"boolean","default":true,"description":"Enable/disable rendering math in the built-in Markdown preview."},"markdown.math.macros":{"type":"object","additionalProperties":{"type":"string"},"default":{},"description":"A collection of custom macros. Each macro is a key-value pair where the key is a new command name and the value is the expansion of the macro.","scope":"resource"}}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/markdown-math","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.media-preview"},"manifest":{"name":"media-preview","displayName":"Media Preview","description":"Provides VS Code's built-in previews for images, audio, and video","extensionKind":["ui","workspace"],"version":"1.0.0","publisher":"vscode","icon":"icon.png","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.70.0"},"main":"./dist/extension","browser":"./dist/browser/extension.js","categories":["Other"],"activationEvents":[],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"configuration":{"type":"object","title":"Media Previewer","properties":{"mediaPreview.video.autoPlay":{"type":"boolean","default":false,"markdownDescription":"Start playing videos on mute automatically."},"mediaPreview.video.loop":{"type":"boolean","default":false,"markdownDescription":"Loop videos over again automatically."}}},"customEditors":[{"viewType":"imagePreview.previewEditor","displayName":"Image Preview","priority":"builtin","selector":[{"filenamePattern":"*.{jpg,jpe,jpeg,png,bmp,gif,ico,webp,avif,svg}"}]},{"viewType":"vscode.audioPreview","displayName":"Audio Preview","priority":"builtin","selector":[{"filenamePattern":"*.{mp3,wav,ogg,oga}"}]},{"viewType":"vscode.videoPreview","displayName":"Video Preview","priority":"builtin","selector":[{"filenamePattern":"*.{mp4,webm}"}]}],"commands":[{"command":"imagePreview.zoomIn","title":"Zoom in","category":"Image Preview"},{"command":"imagePreview.zoomOut","title":"Zoom out","category":"Image Preview"},{"command":"imagePreview.copyImage","title":"Copy","category":"Image Preview"}],"menus":{"commandPalette":[{"command":"imagePreview.zoomIn","when":"activeCustomEditorId == 'imagePreview.previewEditor'","group":"1_imagePreview"},{"command":"imagePreview.zoomOut","when":"activeCustomEditorId == 'imagePreview.previewEditor'","group":"1_imagePreview"},{"command":"imagePreview.copyImage","when":"false"}],"webview/context":[{"command":"imagePreview.copyImage","when":"webviewId == 'imagePreview.previewEditor'"}]}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/media-preview","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.merge-conflict"},"manifest":{"name":"merge-conflict","publisher":"vscode","displayName":"Merge Conflict","description":"Highlighting and commands for inline merge conflicts.","icon":"media/icon.png","version":"1.0.0","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.5.0"},"categories":["Other"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"activationEvents":["onStartupFinished"],"main":"./dist/mergeConflictMain","browser":"./dist/browser/mergeConflictMain","contributes":{"commands":[{"category":"Merge Conflict","title":"Accept All Current","original":"Accept All Current","command":"merge-conflict.accept.all-current","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept All Incoming","original":"Accept All Incoming","command":"merge-conflict.accept.all-incoming","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept All Both","original":"Accept All Both","command":"merge-conflict.accept.all-both","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept Current","original":"Accept Current","command":"merge-conflict.accept.current","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept Incoming","original":"Accept Incoming","command":"merge-conflict.accept.incoming","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept Selection","original":"Accept Selection","command":"merge-conflict.accept.selection","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Accept Both","original":"Accept Both","command":"merge-conflict.accept.both","enablement":"!isMergeEditor"},{"category":"Merge Conflict","title":"Next Conflict","original":"Next Conflict","command":"merge-conflict.next","enablement":"!isMergeEditor","icon":"$(arrow-down)"},{"category":"Merge Conflict","title":"Previous Conflict","original":"Previous Conflict","command":"merge-conflict.previous","enablement":"!isMergeEditor","icon":"$(arrow-up)"},{"category":"Merge Conflict","title":"Compare Current Conflict","original":"Compare Current Conflict","command":"merge-conflict.compare","enablement":"!isMergeEditor"}],"menus":{"scm/resourceState/context":[{"command":"merge-conflict.accept.all-current","when":"scmProvider == git && scmResourceGroup == merge","group":"1_modification"},{"command":"merge-conflict.accept.all-incoming","when":"scmProvider == git && scmResourceGroup == merge","group":"1_modification"}],"editor/title":[{"command":"merge-conflict.previous","group":"navigation@1","when":"!isMergeEditor && mergeConflictsCount && mergeConflictsCount != 0"},{"command":"merge-conflict.next","group":"navigation@2","when":"!isMergeEditor && mergeConflictsCount && mergeConflictsCount != 0"}]},"configuration":{"title":"Merge Conflict","properties":{"merge-conflict.codeLens.enabled":{"type":"boolean","description":"Create a CodeLens for merge conflict blocks within editor.","default":true},"merge-conflict.decorators.enabled":{"type":"boolean","description":"Create decorators for merge conflict blocks within editor.","default":true},"merge-conflict.autoNavigateNextConflict.enabled":{"type":"boolean","description":"Whether to automatically navigate to the next merge conflict after resolving a merge conflict.","default":false},"merge-conflict.diffViewPosition":{"type":"string","enum":["Current","Beside","Below"],"description":"Controls where the diff view should be opened when comparing changes in merge conflicts.","enumDescriptions":["Open the diff view in the current editor group.","Open the diff view next to the current editor group.","Open the diff view below the current editor group."],"default":"Current"}}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/merge-conflict","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.microsoft-authentication"},"manifest":{"name":"microsoft-authentication","publisher":"vscode","license":"MIT","displayName":"Microsoft Account","description":"Microsoft authentication provider","version":"0.0.1","engines":{"vscode":"^1.42.0"},"icon":"media/icon.png","categories":["Other"],"activationEvents":[],"enabledApiProposals":["idToken","nativeWindowHandle"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"extensionKind":["ui","workspace"],"contributes":{"authentication":[{"label":"Microsoft","id":"microsoft"},{"label":"Microsoft Sovereign Cloud","id":"microsoft-sovereign-cloud"}],"configuration":[{"title":"Microsoft Sovereign Cloud","properties":{"microsoft-sovereign-cloud.environment":{"type":"string","markdownDescription":"The Sovereign Cloud to use for authentication. If you select `custom`, you must also set the `#microsoft-sovereign-cloud.customEnvironment#` setting.","enum":["ChinaCloud","USGovernment","custom"],"enumDescriptions":["Azure China","Azure US Government","A custom Microsoft Sovereign Cloud"]},"microsoft-sovereign-cloud.customEnvironment":{"type":"object","additionalProperties":true,"markdownDescription":"The custom configuration for the Sovereign Cloud to use with the Microsoft Sovereign Cloud authentication provider. This along with setting `#microsoft-sovereign-cloud.environment#` to `custom` is required to use this feature.","properties":{"name":{"type":"string","description":"The name of the custom Sovereign Cloud."},"portalUrl":{"type":"string","description":"The portal URL for the custom Sovereign Cloud."},"managementEndpointUrl":{"type":"string","description":"The management endpoint for the custom Sovereign Cloud."},"resourceManagerEndpointUrl":{"type":"string","description":"The resource manager endpoint for the custom Sovereign Cloud."},"activeDirectoryEndpointUrl":{"type":"string","description":"The Active Directory endpoint for the custom Sovereign Cloud."},"activeDirectoryResourceId":{"type":"string","description":"The Active Directory resource ID for the custom Sovereign Cloud."}},"required":["name","portalUrl","managementEndpointUrl","resourceManagerEndpointUrl","activeDirectoryEndpointUrl","activeDirectoryResourceId"]}}},{"title":"Microsoft","properties":{"microsoft-authentication.implementation":{"type":"string","default":"msal","enum":["msal","classic"],"enumDescriptions":["Use the Microsoft Authentication Library (MSAL) to sign in with a Microsoft account.","(deprecated) Use the classic authentication flow to sign in with a Microsoft account."],"markdownDescription":"The authentication implementation to use for signing in with a Microsoft account.\n\n*NOTE: The `classic` implementation is deprecated and will be removed, along with this setting, in a future release. If only the `classic` implementation works for you, please [open an issue](command:workbench.action.openIssueReporter) and explain what you are trying to log in to.*","tags":["onExP"]},"microsoft-authentication.clientIdVersion":{"type":"string","default":"v1","enum":["v2","v1"],"enumDescriptions":["Use the v2 Microsoft Account client ID to sign in with a Microsoft account.","Use the v1 Microsoft Account client ID to sign in with a Microsoft account."],"markdownDescription":"The version of the Microsoft Account client ID to use for signing in with a Microsoft account. Only change this if you have been asked to. The default is `v1`.","tags":["onExP","experimental"]}}}]},"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","main":"./dist/extension.js","browser":"./dist/browser/extension.js","repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/microsoft-authentication","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"ms-vscode.js-debug","uuid":"25629058-ddac-4e17-abba-74678e126c5d"},"manifest":{"name":"js-debug","displayName":"JavaScript Debugger","version":"1.97.1","publisher":"ms-vscode","author":{"name":"Microsoft Corporation"},"keywords":["pwa","javascript","node","chrome","debugger"],"description":"An extension for debugging Node.js programs and Chrome.","license":"MIT","engines":{"vscode":"^1.80.0","node":">=10"},"icon":"resources/logo.png","categories":["Debuggers"],"private":true,"repository":{"type":"git","url":"https://github.com/Microsoft/vscode-pwa.git"},"bugs":{"url":"https://github.com/Microsoft/vscode-pwa/issues"},"main":"./src/extension.js","enabledApiProposals":["portsAttributes","workspaceTrust","tunnels"],"extensionKind":["workspace"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":"limited","description":"Trust is required to debug code in this workspace."}},"activationEvents":["onDebugDynamicConfigurations","onDebugInitialConfigurations","onFileSystem:jsDebugNetworkFs","onDebugResolve:pwa-node","onDebugResolve:node-terminal","onDebugResolve:pwa-extensionHost","onDebugResolve:pwa-chrome","onDebugResolve:pwa-msedge","onDebugResolve:node","onDebugResolve:chrome","onDebugResolve:extensionHost","onDebugResolve:msedge","onCommand:extension.js-debug.clearAutoAttachVariables","onCommand:extension.js-debug.setAutoAttachVariables","onCommand:extension.js-debug.autoAttachToProcess","onCommand:extension.js-debug.pickNodeProcess","onCommand:extension.js-debug.requestCDPProxy","onCommand:extension.js-debug.completion.nodeTool"],"contributes":{"menus":{"commandPalette":[{"command":"extension.js-debug.prettyPrint","title":"Pretty print for debugging","when":"debugType == pwa-extensionHost && debugState == stopped || debugType == node-terminal && debugState == stopped || debugType == pwa-node && debugState == stopped || debugType == pwa-chrome && debugState == stopped || debugType == pwa-msedge && debugState == stopped"},{"command":"extension.js-debug.startProfile","title":"Take Performance Profile","when":"debugType == pwa-extensionHost && inDebugMode && !jsDebugIsProfiling || debugType == node-terminal && inDebugMode && !jsDebugIsProfiling || debugType == pwa-node && inDebugMode && !jsDebugIsProfiling || debugType == pwa-chrome && inDebugMode && !jsDebugIsProfiling || debugType == pwa-msedge && inDebugMode && !jsDebugIsProfiling"},{"command":"extension.js-debug.stopProfile","title":"Stop Performance Profile","when":"debugType == pwa-extensionHost && inDebugMode && jsDebugIsProfiling || debugType == node-terminal && inDebugMode && jsDebugIsProfiling || debugType == pwa-node && inDebugMode && jsDebugIsProfiling || debugType == pwa-chrome && inDebugMode && jsDebugIsProfiling || debugType == pwa-msedge && inDebugMode && jsDebugIsProfiling"},{"command":"extension.js-debug.revealPage","when":"false"},{"command":"extension.js-debug.debugLink","title":"Open Link","when":"!isWeb"},{"command":"extension.js-debug.createDiagnostics","title":"Diagnose Breakpoint Problems","when":"debugType == pwa-extensionHost && inDebugMode || debugType == node-terminal && inDebugMode || debugType == pwa-node && inDebugMode || debugType == pwa-chrome && inDebugMode || debugType == pwa-msedge && inDebugMode"},{"command":"extension.js-debug.getDiagnosticLogs","title":"Save Diagnostic JS Debug Logs","when":"debugType == pwa-extensionHost && inDebugMode || debugType == node-terminal && inDebugMode || debugType == pwa-node && inDebugMode || debugType == pwa-chrome && inDebugMode || debugType == pwa-msedge && inDebugMode"},{"command":"extension.js-debug.openEdgeDevTools","title":"Open Browser Devtools","when":"debugType == pwa-msedge"},{"command":"extension.js-debug.callers.add","title":"Exclude caller from pausing in the current location","when":"debugType == pwa-extensionHost && debugState == \"stopped\" || debugType == node-terminal && debugState == \"stopped\" || debugType == pwa-node && debugState == \"stopped\" || debugType == pwa-chrome && debugState == \"stopped\" || debugType == pwa-msedge && debugState == \"stopped\""},{"command":"extension.js-debug.callers.goToCaller","when":"false"},{"command":"extension.js-debug.callers.gotToTarget","when":"false"},{"command":"extension.js-debug.network.copyUri","when":"false"},{"command":"extension.js-debug.network.openBody","when":"false"},{"command":"extension.js-debug.network.openBodyInHex","when":"false"},{"command":"extension.js-debug.network.replayXHR","when":"false"},{"command":"extension.js-debug.network.viewRequest","when":"false"},{"command":"extension.js-debug.network.clear","when":"false"},{"command":"extension.js-debug.enableSourceMapStepping","when":"jsDebugIsMapSteppingDisabled"},{"command":"extension.js-debug.disableSourceMapStepping","when":"!jsDebugIsMapSteppingDisabled"}],"debug/callstack/context":[{"command":"extension.js-debug.revealPage","group":"navigation","when":"debugType == pwa-chrome && callStackItemType == 'session' || debugType == pwa-msedge && callStackItemType == 'session'"},{"command":"extension.js-debug.toggleSkippingFile","group":"navigation","when":"debugType == pwa-extensionHost && callStackItemType == 'session' || debugType == node-terminal && callStackItemType == 'session' || debugType == pwa-node && callStackItemType == 'session' || debugType == pwa-chrome && callStackItemType == 'session' || debugType == pwa-msedge && callStackItemType == 'session'"},{"command":"extension.js-debug.startProfile","group":"navigation","when":"debugType == pwa-extensionHost && !jsDebugIsProfiling && callStackItemType == 'session' || debugType == node-terminal && !jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-node && !jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-chrome && !jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-msedge && !jsDebugIsProfiling && callStackItemType == 'session'"},{"command":"extension.js-debug.stopProfile","group":"navigation","when":"debugType == pwa-extensionHost && jsDebugIsProfiling && callStackItemType == 'session' || debugType == node-terminal && jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-node && jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-chrome && jsDebugIsProfiling && callStackItemType == 'session' || debugType == pwa-msedge && jsDebugIsProfiling && callStackItemType == 'session'"},{"command":"extension.js-debug.startProfile","group":"inline","when":"debugType == pwa-extensionHost && !jsDebugIsProfiling || debugType == node-terminal && !jsDebugIsProfiling || debugType == pwa-node && !jsDebugIsProfiling || debugType == pwa-chrome && !jsDebugIsProfiling || debugType == pwa-msedge && !jsDebugIsProfiling"},{"command":"extension.js-debug.stopProfile","group":"inline","when":"debugType == pwa-extensionHost && jsDebugIsProfiling || debugType == node-terminal && jsDebugIsProfiling || debugType == pwa-node && jsDebugIsProfiling || debugType == pwa-chrome && jsDebugIsProfiling || debugType == pwa-msedge && jsDebugIsProfiling"},{"command":"extension.js-debug.callers.add","when":"debugType == pwa-extensionHost && callStackItemType == 'stackFrame' || debugType == node-terminal && callStackItemType == 'stackFrame' || debugType == pwa-node && callStackItemType == 'stackFrame' || debugType == pwa-chrome && callStackItemType == 'stackFrame' || debugType == pwa-msedge && callStackItemType == 'stackFrame'"}],"debug/toolBar":[{"command":"extension.js-debug.stopProfile","when":"debugType == pwa-extensionHost && jsDebugIsProfiling || debugType == node-terminal && jsDebugIsProfiling || debugType == pwa-node && jsDebugIsProfiling || debugType == pwa-chrome && jsDebugIsProfiling || debugType == pwa-msedge && jsDebugIsProfiling"},{"command":"extension.js-debug.openEdgeDevTools","when":"debugType == pwa-msedge"},{"command":"extension.js-debug.enableSourceMapStepping","when":"jsDebugIsMapSteppingDisabled"}],"view/title":[{"command":"extension.js-debug.addCustomBreakpoints","when":"view == jsBrowserBreakpoints","group":"navigation"},{"command":"extension.js-debug.removeAllCustomBreakpoints","when":"view == jsBrowserBreakpoints","group":"navigation"},{"command":"extension.js-debug.callers.removeAll","group":"navigation","when":"view == jsExcludedCallers"},{"command":"extension.js-debug.disableSourceMapStepping","group":"navigation","when":"debugType == pwa-extensionHost && view == workbench.debug.callStackView && !jsDebugIsMapSteppingDisabled || debugType == node-terminal && view == workbench.debug.callStackView && !jsDebugIsMapSteppingDisabled || debugType == pwa-node && view == workbench.debug.callStackView && !jsDebugIsMapSteppingDisabled || debugType == pwa-chrome && view == workbench.debug.callStackView && !jsDebugIsMapSteppingDisabled || debugType == pwa-msedge && view == workbench.debug.callStackView && !jsDebugIsMapSteppingDisabled"},{"command":"extension.js-debug.enableSourceMapStepping","group":"navigation","when":"debugType == pwa-extensionHost && view == workbench.debug.callStackView && jsDebugIsMapSteppingDisabled || debugType == node-terminal && view == workbench.debug.callStackView && jsDebugIsMapSteppingDisabled || debugType == pwa-node && view == workbench.debug.callStackView && jsDebugIsMapSteppingDisabled || debugType == pwa-chrome && view == workbench.debug.callStackView && jsDebugIsMapSteppingDisabled || debugType == pwa-msedge && view == workbench.debug.callStackView && jsDebugIsMapSteppingDisabled"},{"command":"extension.js-debug.network.clear","group":"navigation","when":"view == jsDebugNetworkTree"}],"view/item/context":[{"command":"extension.js-debug.addXHRBreakpoints","when":"view == jsBrowserBreakpoints && viewItem == xhrBreakpoint"},{"command":"extension.js-debug.editXHRBreakpoints","when":"view == jsBrowserBreakpoints && viewItem == xhrBreakpoint","group":"inline"},{"command":"extension.js-debug.editXHRBreakpoints","when":"view == jsBrowserBreakpoints && viewItem == xhrBreakpoint"},{"command":"extension.js-debug.removeXHRBreakpoint","when":"view == jsBrowserBreakpoints && viewItem == xhrBreakpoint","group":"inline"},{"command":"extension.js-debug.removeXHRBreakpoint","when":"view == jsBrowserBreakpoints && viewItem == xhrBreakpoint"},{"command":"extension.js-debug.addXHRBreakpoints","when":"view == jsBrowserBreakpoints && viewItem == xhrCategory","group":"inline"},{"command":"extension.js-debug.callers.goToCaller","group":"inline","when":"view == jsExcludedCallers"},{"command":"extension.js-debug.callers.gotToTarget","group":"inline","when":"view == jsExcludedCallers"},{"command":"extension.js-debug.callers.remove","group":"inline","when":"view == jsExcludedCallers"},{"command":"extension.js-debug.network.viewRequest","group":"inline@1","when":"view == jsDebugNetworkTree"},{"command":"extension.js-debug.network.openBody","group":"body@1","when":"view == jsDebugNetworkTree"},{"command":"extension.js-debug.network.openBodyInHex","group":"body@2","when":"view == jsDebugNetworkTree"},{"command":"extension.js-debug.network.copyUri","group":"other@1","when":"view == jsDebugNetworkTree"},{"command":"extension.js-debug.network.replayXHR","group":"other@2","when":"view == jsDebugNetworkTree"}],"editor/title":[{"command":"extension.js-debug.prettyPrint","group":"navigation","when":"jsDebugCanPrettyPrint"}]},"breakpoints":[{"language":"javascript"},{"language":"typescript"},{"language":"typescriptreact"},{"language":"javascriptreact"},{"language":"fsharp"},{"language":"html"},{"language":"wat"},{"language":"c"},{"language":"cpp"},{"language":"rust"},{"language":"zig"}],"debuggers":[{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"TCP/IP address of process to be debugged. Default is 'localhost'.","type":"string"},"attachExistingChildren":{"default":false,"description":"Whether to attempt to attach to already-spawned child processes.","type":"boolean"},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"continueOnAttach":{"default":true,"markdownDescription":"If true, we'll automatically resume programs launched and waiting on `--inspect-brk`","type":"boolean"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"port":{"default":9229,"description":"Debug port to attach to. Default is 9229.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"processId":{"default":"${command:PickProcess}","description":"ID of process to attach to.","type":"string"},"remoteHostHeader":{"description":"Explicit Host header to use when connecting to the websocket of inspector. If unspecified, the host header will be set to 'localhost'. This is useful when the inspector is running behind a proxy that only accept particular Host header.","type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"websocketAddress":{"description":"Exact websocket address to attach to. If unspecified, it will be discovered from the address and port.","type":"string"}}},"launch":{"properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}}}},"configurationSnippets":[],"deprecated":"Please use type node instead","label":"Node.js","languages":["javascript","typescript","javascriptreact","typescriptreact"],"strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"pwa-node","variables":{"PickProcess":"extension.js-debug.pickNodeProcess"}},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"TCP/IP address of process to be debugged. Default is 'localhost'.","type":"string"},"attachExistingChildren":{"default":false,"description":"Whether to attempt to attach to already-spawned child processes.","type":"boolean"},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"continueOnAttach":{"default":true,"markdownDescription":"If true, we'll automatically resume programs launched and waiting on `--inspect-brk`","type":"boolean"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"port":{"default":9229,"description":"Debug port to attach to. Default is 9229.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"processId":{"default":"${command:PickProcess}","description":"ID of process to attach to.","type":"string"},"remoteHostHeader":{"description":"Explicit Host header to use when connecting to the websocket of inspector. If unspecified, the host header will be set to 'localhost'. This is useful when the inspector is running behind a proxy that only accept particular Host header.","type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"websocketAddress":{"description":"Exact websocket address to attach to. If unspecified, it will be discovered from the address and port.","type":"string"}}},"launch":{"properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}}}},"configurationSnippets":[{"body":{"name":"${1:Attach}","port":9229,"request":"attach","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Attach to a running node program","label":"Node.js: Attach"},{"body":{"address":"${2:TCP/IP address of process to be debugged}","localRoot":"^\"\\${workspaceFolder}\"","name":"${1:Attach to Remote}","port":9229,"remoteRoot":"${3:Absolute path to the remote directory containing the program}","request":"attach","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Attach to the debug port of a remote node program","label":"Node.js: Attach to Remote Program"},{"body":{"name":"${1:Attach by Process ID}","processId":"^\"\\${command:PickProcess}\"","request":"attach","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Open process picker to select node process to attach to","label":"Node.js: Attach to Process"},{"body":{"name":"${2:Launch Program}","program":"^\"\\${workspaceFolder}/${1:app.js}\"","request":"launch","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Launch a node program in debug mode","label":"Node.js: Launch Program"},{"body":{"name":"${1:Launch via NPM}","request":"launch","runtimeArgs":["run-script","debug"],"runtimeExecutable":"npm","skipFiles":["<node_internals>/**"],"type":"node"},"label":"Node.js: Launch via npm","markdownDescription":"Launch a node program through an npm `debug` script"},{"body":{"console":"integratedTerminal","internalConsoleOptions":"neverOpen","name":"nodemon","program":"^\"\\${workspaceFolder}/${1:app.js}\"","request":"launch","restart":true,"runtimeExecutable":"nodemon","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Use nodemon to relaunch a debug session on source changes","label":"Node.js: Nodemon Setup"},{"body":{"args":["-u","tdd","--timeout","999999","--colors","^\"\\${workspaceFolder}/${1:test}\""],"internalConsoleOptions":"openOnSessionStart","name":"Mocha Tests","program":"^\"mocha\"","request":"launch","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Debug mocha tests","label":"Node.js: Mocha Tests"},{"body":{"args":["${1:generator}"],"console":"integratedTerminal","internalConsoleOptions":"neverOpen","name":"Yeoman ${1:generator}","program":"^\"\\${workspaceFolder}/node_modules/yo/lib/cli.js\"","request":"launch","skipFiles":["<node_internals>/**"],"type":"node"},"label":"Node.js: Yeoman generator","markdownDescription":"Debug yeoman generator (install by running `npm link` in project folder)"},{"body":{"args":["${1:task}"],"name":"Gulp ${1:task}","program":"^\"\\${workspaceFolder}/node_modules/gulp/bin/gulp.js\"","request":"launch","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Debug gulp task (make sure to have a local gulp installed in your project)","label":"Node.js: Gulp task"},{"body":{"name":"Electron Main","program":"^\"\\${workspaceFolder}/main.js\"","request":"launch","runtimeExecutable":"^\"electron\"","skipFiles":["<node_internals>/**"],"type":"node"},"description":"Debug the Electron main process","label":"Node.js: Electron Main"}],"label":"Node.js","strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"node","variables":{"PickProcess":"extension.js-debug.pickNodeProcess"}},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"launch":{"properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}}}},"configurationSnippets":[{"body":{"command":"npm start","name":"Run npm start","request":"launch","type":"node-terminal"},"description":"Run \"npm start\" in a debug terminal","label":"Run \"npm start\" in a debug terminal"}],"label":"JavaScript Debug Terminal","languages":[],"strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"node-terminal"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"launch":{"properties":{"args":{"default":["--extensionDevelopmentPath=${workspaceFolder}"],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":"array"},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"debugWebWorkerHost":{"default":true,"markdownDescription":"Configures whether we should try to attach to the web worker extension host.","type":["boolean"]},"debugWebviews":{"default":true,"markdownDescription":"Configures whether we should try to attach to webviews in the launched VS Code instance. This will only work in desktop VS Code.","type":["boolean"]},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"rendererDebugOptions":{"default":{"webRoot":"${workspaceFolder}"},"markdownDescription":"Chrome launch options used when attaching to the renderer process, with `debugWebviews` or `debugWebWorkerHost`.","properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}},"type":"object"},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeExecutable":{"default":"node","markdownDescription":"Absolute path to VS Code.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"testConfiguration":{"default":"${workspaceFolder}/.vscode-test.js","markdownDescription":"Path to a test configuration file for the [test CLI](https://code.visualstudio.com/api/working-with-extensions/testing-extension#quick-setup-the-test-cli).","type":"string"},"testConfigurationLabel":{"default":"","markdownDescription":"A single configuration to run from the file. If not specified, you may be asked to pick.","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"required":[]}},"configurationSnippets":[],"deprecated":"Please use type extensionHost instead","label":"VS Code Extension Development","languages":["javascript","typescript","javascriptreact","typescriptreact"],"strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"pwa-extensionHost"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"launch":{"properties":{"args":{"default":["--extensionDevelopmentPath=${workspaceFolder}"],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":"array"},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"debugWebWorkerHost":{"default":true,"markdownDescription":"Configures whether we should try to attach to the web worker extension host.","type":["boolean"]},"debugWebviews":{"default":true,"markdownDescription":"Configures whether we should try to attach to webviews in the launched VS Code instance. This will only work in desktop VS Code.","type":["boolean"]},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"rendererDebugOptions":{"default":{"webRoot":"${workspaceFolder}"},"markdownDescription":"Chrome launch options used when attaching to the renderer process, with `debugWebviews` or `debugWebWorkerHost`.","properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}},"type":"object"},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeExecutable":{"default":"node","markdownDescription":"Absolute path to VS Code.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"testConfiguration":{"default":"${workspaceFolder}/.vscode-test.js","markdownDescription":"Path to a test configuration file for the [test CLI](https://code.visualstudio.com/api/working-with-extensions/testing-extension#quick-setup-the-test-cli).","type":"string"},"testConfigurationLabel":{"default":"","markdownDescription":"A single configuration to run from the file. If not specified, you may be asked to pick.","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"required":[]}},"configurationSnippets":[{"body":{"args":["^\"--extensionDevelopmentPath=\\${workspaceFolder}\""],"name":"Launch Extension","outFiles":["^\"\\${workspaceFolder}/out/**/*.js\""],"preLaunchTask":"npm","request":"launch","type":"extensionHost"},"description":"Launch a VS Code extension in debug mode","label":"VS Code Extension Development"}],"label":"VS Code Extension Development","strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"extensionHost"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}},"launch":{"properties":{"browserLaunchLocation":{"default":null,"description":"Forces the browser to be launched in one location. In a remote workspace (through ssh or WSL, for example) this can be used to open the browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"cleanUp":{"default":"wholeBrowser","description":"What clean-up to do after the debugging session finishes. Close only the tab being debug, vs. close the whole browser.","enum":["wholeBrowser","onlyTab"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":null,"description":"Optional working directory for the runtime executable.","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"default":{},"description":"Optional dictionary of environment key/value pairs for the browser.","type":"object"},"file":{"default":"${workspaceFolder}/index.html","description":"A local html file to open in the browser","tags":["setup"],"type":"string"},"includeDefaultArgs":{"default":true,"description":"Whether default browser launch arguments (to disable features that may make debugging harder) will be included in the launch.","type":"boolean"},"includeLaunchArgs":{"default":true,"description":"Advanced: whether any default launch/debugging arguments are set on the browser. The debugger will assume the browser will use pipe debugging such as that which is provided with `--remote-debugging-pipe`.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":0,"description":"Port for the browser to listen on. Defaults to \"0\", which will cause the browser to be debugged via pipes, which is generally more secure and should be chosen unless you need to attach to the browser from another tool.","type":"number"},"profileStartup":{"default":true,"description":"If true, will start profiling soon as the process launches","type":"boolean"},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"type":"array"},"runtimeExecutable":{"default":"stable","description":"Either 'canary', 'stable', 'custom' or path to the browser executable. Custom means a custom wrapper, custom build or CHROME_PATH environment variable.","type":["string","null"]},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"userDataDir":{"default":true,"description":"By default, the browser is launched with a separate user profile in a temp folder. Use this option to override it. Set to false to launch with your default user profile. A new browser can't be launched if an instance is already running from `userDataDir`.","type":["string","boolean"]},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}}},"configurationSnippets":[],"deprecated":"Please use type chrome instead","label":"Web App (Chrome)","languages":["javascript","typescript","javascriptreact","typescriptreact","html","css","coffeescript","handlebars","vue"],"strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"pwa-chrome"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}},"launch":{"properties":{"browserLaunchLocation":{"default":null,"description":"Forces the browser to be launched in one location. In a remote workspace (through ssh or WSL, for example) this can be used to open the browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"cleanUp":{"default":"wholeBrowser","description":"What clean-up to do after the debugging session finishes. Close only the tab being debug, vs. close the whole browser.","enum":["wholeBrowser","onlyTab"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":null,"description":"Optional working directory for the runtime executable.","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"default":{},"description":"Optional dictionary of environment key/value pairs for the browser.","type":"object"},"file":{"default":"${workspaceFolder}/index.html","description":"A local html file to open in the browser","tags":["setup"],"type":"string"},"includeDefaultArgs":{"default":true,"description":"Whether default browser launch arguments (to disable features that may make debugging harder) will be included in the launch.","type":"boolean"},"includeLaunchArgs":{"default":true,"description":"Advanced: whether any default launch/debugging arguments are set on the browser. The debugger will assume the browser will use pipe debugging such as that which is provided with `--remote-debugging-pipe`.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":0,"description":"Port for the browser to listen on. Defaults to \"0\", which will cause the browser to be debugged via pipes, which is generally more secure and should be chosen unless you need to attach to the browser from another tool.","type":"number"},"profileStartup":{"default":true,"description":"If true, will start profiling soon as the process launches","type":"boolean"},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"type":"array"},"runtimeExecutable":{"default":"stable","description":"Either 'canary', 'stable', 'custom' or path to the browser executable. Custom means a custom wrapper, custom build or CHROME_PATH environment variable.","type":["string","null"]},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"userDataDir":{"default":true,"description":"By default, the browser is launched with a separate user profile in a temp folder. Use this option to override it. Set to false to launch with your default user profile. A new browser can't be launched if an instance is already running from `userDataDir`.","type":["string","boolean"]},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}}},"configurationSnippets":[{"body":{"name":"Launch Chrome","request":"launch","type":"chrome","url":"http://localhost:8080","webRoot":"^\"${2:\\${workspaceFolder\\}}\""},"description":"Launch Chrome to debug a URL","label":"Chrome: Launch"},{"body":{"name":"Attach to Chrome","port":9222,"request":"attach","type":"chrome","webRoot":"^\"${2:\\${workspaceFolder\\}}\""},"description":"Attach to an instance of Chrome already in debug mode","label":"Chrome: Attach"}],"label":"Web App (Chrome)","strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"chrome"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"useWebView":{"default":{"pipeName":"MyPipeName"},"description":"An object containing the `pipeName` of a debug pipe for a UWP hosted Webview2. This is the \"MyTestSharedMemory\" when creating the pipe \"\\\\.\\pipe\\LOCAL\\MyTestSharedMemory\"","properties":{"pipeName":{"type":"string"}},"type":"object"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}},"launch":{"properties":{"address":{"default":"localhost","description":"When debugging webviews, the IP address or hostname the webview is listening on. Will be automatically discovered if not set.","type":"string"},"browserLaunchLocation":{"default":null,"description":"Forces the browser to be launched in one location. In a remote workspace (through ssh or WSL, for example) this can be used to open the browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"cleanUp":{"default":"wholeBrowser","description":"What clean-up to do after the debugging session finishes. Close only the tab being debug, vs. close the whole browser.","enum":["wholeBrowser","onlyTab"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":null,"description":"Optional working directory for the runtime executable.","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"default":{},"description":"Optional dictionary of environment key/value pairs for the browser.","type":"object"},"file":{"default":"${workspaceFolder}/index.html","description":"A local html file to open in the browser","tags":["setup"],"type":"string"},"includeDefaultArgs":{"default":true,"description":"Whether default browser launch arguments (to disable features that may make debugging harder) will be included in the launch.","type":"boolean"},"includeLaunchArgs":{"default":true,"description":"Advanced: whether any default launch/debugging arguments are set on the browser. The debugger will assume the browser will use pipe debugging such as that which is provided with `--remote-debugging-pipe`.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"When debugging webviews, the port the webview debugger is listening on. Will be automatically discovered if not set.","type":"number"},"profileStartup":{"default":true,"description":"If true, will start profiling soon as the process launches","type":"boolean"},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"type":"array"},"runtimeExecutable":{"default":"stable","description":"Either 'canary', 'stable', 'dev', 'custom' or path to the browser executable. Custom means a custom wrapper, custom build or EDGE_PATH environment variable.","type":["string","null"]},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"useWebView":{"default":false,"description":"When 'true', the debugger will treat the runtime executable as a host application that contains a WebView allowing you to debug the WebView script content.","type":"boolean"},"userDataDir":{"default":true,"description":"By default, the browser is launched with a separate user profile in a temp folder. Use this option to override it. Set to false to launch with your default user profile. A new browser can't be launched if an instance is already running from `userDataDir`.","type":["string","boolean"]},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}}},"configurationSnippets":[],"deprecated":"Please use type msedge instead","label":"Web App (Edge)","languages":["javascript","typescript","javascriptreact","typescriptreact","html","css","coffeescript","handlebars","vue"],"strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"pwa-msedge"},{"aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","configurationAttributes":{"attach":{"properties":{"address":{"default":"localhost","description":"IP address or hostname the debugged browser is listening on.","type":"string"},"browserAttachLocation":{"default":null,"description":"Forces the browser to attach in one location. In a remote workspace (through ssh or WSL, for example) this can be used to attach to a browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"Port to use to remote debugging the browser, given as `--remote-debugging-port` when launching the browser.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}],"tags":["setup"]},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":false,"markdownDescription":"Whether to reconnect if the browser connection is closed","type":"boolean"},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"targetSelection":{"default":"automatic","enum":["pick","automatic"],"markdownDescription":"Whether to attach to all targets that match the URL filter (\"automatic\") or ask to pick one (\"pick\").","type":"string"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"useWebView":{"default":{"pipeName":"MyPipeName"},"description":"An object containing the `pipeName` of a debug pipe for a UWP hosted Webview2. This is the \"MyTestSharedMemory\" when creating the pipe \"\\\\.\\pipe\\LOCAL\\MyTestSharedMemory\"","properties":{"pipeName":{"type":"string"}},"type":"object"},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}},"launch":{"properties":{"address":{"default":"localhost","description":"When debugging webviews, the IP address or hostname the webview is listening on. Will be automatically discovered if not set.","type":"string"},"browserLaunchLocation":{"default":null,"description":"Forces the browser to be launched in one location. In a remote workspace (through ssh or WSL, for example) this can be used to open the browser on the remote machine rather than locally.","oneOf":[{"type":"null"},{"enum":["ui","workspace"],"type":"string"}]},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"cleanUp":{"default":"wholeBrowser","description":"What clean-up to do after the debugging session finishes. Close only the tab being debug, vs. close the whole browser.","enum":["wholeBrowser","onlyTab"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":null,"description":"Optional working directory for the runtime executable.","type":"string"},"disableNetworkCache":{"default":true,"description":"Controls whether to skip the network cache for each request","type":"boolean"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"default":{},"description":"Optional dictionary of environment key/value pairs for the browser.","type":"object"},"file":{"default":"${workspaceFolder}/index.html","description":"A local html file to open in the browser","tags":["setup"],"type":"string"},"includeDefaultArgs":{"default":true,"description":"Whether default browser launch arguments (to disable features that may make debugging harder) will be included in the launch.","type":"boolean"},"includeLaunchArgs":{"default":true,"description":"Advanced: whether any default launch/debugging arguments are set on the browser. The debugger will assume the browser will use pipe debugging such as that which is provided with `--remote-debugging-pipe`.","type":"boolean"},"inspectUri":{"default":null,"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","type":["string","null"]},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pathMapping":{"default":{},"description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","type":"object"},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"perScriptSourcemaps":{"default":"auto","description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate.","enum":["yes","no","auto"],"type":"string"},"port":{"default":9229,"description":"When debugging webviews, the port the webview debugger is listening on. Will be automatically discovered if not set.","type":"number"},"profileStartup":{"default":true,"description":"If true, will start profiling soon as the process launches","type":"boolean"},"resolveSourceMapLocations":{"default":null,"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"type":"array"},"runtimeExecutable":{"default":"stable","description":"Either 'canary', 'stable', 'dev', 'custom' or path to the browser executable. Custom means a custom wrapper, custom build or EDGE_PATH environment variable.","type":["string","null"]},"server":{"oneOf":[{"additionalProperties":false,"default":{"program":"node my-server.js"},"description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","properties":{"args":{"default":[],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"tags":["setup"],"type":["array","string"]},"attachSimplePort":{"default":9229,"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","oneOf":[{"type":"integer"},{"pattern":"^\\${.*}$","type":"string"}]},"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"console":{"default":"internalConsole","description":"Where to launch the debug target.","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"type":"string"},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"experimentalNetworking":{"default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"],"type":"string"},"killBehavior":{"default":"forceful","enum":["forceful","polite","none"],"markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"profileStartup":{"default":true,"description":"If true, will start profiling as soon as the process launches","type":"boolean"},"program":{"default":"","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","tags":["setup"],"type":"string"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"restart":{"default":true,"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","oneOf":[{"type":"boolean"},{"properties":{"delay":{"default":1000,"minimum":0,"type":"number"},"maxAttempts":{"default":10,"minimum":0,"type":"number"}},"type":"object"}]},"runtimeArgs":{"default":[],"description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"tags":["setup"],"type":"array"},"runtimeExecutable":{"default":"node","markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","type":["string","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"runtimeVersion":{"default":"default","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","type":"string"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"stopOnEntry":{"default":true,"description":"Automatically stop program after launch.","type":["boolean","string"]},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"},{"additionalProperties":false,"default":{"program":"npm start"},"description":"JavaScript Debug Terminal","properties":{"autoAttachChildProcesses":{"default":true,"description":"Attach debugger to new child processes automatically.","type":"boolean"},"cascadeTerminateToConfigurations":{"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped.","items":{"type":"string","uniqueItems":true},"type":"array"},"command":{"default":"npm start","description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","tags":["setup"],"type":["string","null"]},"customDescriptionGenerator":{"description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      ","type":"string"},"customPropertiesGenerator":{"deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181","type":"string"},"cwd":{"default":"${workspaceFolder}","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"],"type":"string"},"enableContentValidation":{"default":true,"description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example.","type":"boolean"},"enableDWARF":{"default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function.","type":"boolean"},"env":{"additionalProperties":{"type":["string","null"]},"default":{},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","tags":["setup"],"type":"object"},"envFile":{"default":"${workspaceFolder}/.env","description":"Absolute path to a file containing environment variable definitions.","type":"string"},"localRoot":{"default":null,"description":"Path to the local directory containing the program.","type":["string","null"]},"nodeVersionHint":{"default":12,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","minimum":8,"type":"number"},"outFiles":{"default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","items":{"type":"string"},"tags":["setup"],"type":["array"]},"outputCapture":{"default":"console","enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`."},"pauseForSourceMap":{"default":false,"markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","type":"boolean"},"remoteRoot":{"default":null,"description":"Absolute path to the remote directory containing the program.","type":["string","null"]},"resolveSourceMapLocations":{"default":["${workspaceFolder}/**","!**/node_modules/**"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","items":{"type":"string"},"type":["array","null"]},"runtimeSourcemapPausePatterns":{"default":[],"items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","type":"array"},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]}},"type":"object"}]},"showAsyncStacks":{"default":true,"description":"Show the async calls that led to the current call stack.","oneOf":[{"type":"boolean"},{"properties":{"onAttach":{"default":32,"type":"number"}},"required":["onAttach"],"type":"object"},{"properties":{"onceBreakpointResolved":{"default":32,"type":"number"}},"required":["onceBreakpointResolved"],"type":"object"}]},"skipFiles":{"default":["${/**"],"description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","type":"array"},"smartStep":{"default":true,"description":"Automatically step through generated code that cannot be mapped back to the original source.","type":"boolean"},"sourceMapPathOverrides":{"default":{"meteor://💻app/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","webpack://?:*/*":"${workspaceFolder}/*"},"description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","type":"object"},"sourceMapRenames":{"default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers.","type":"boolean"},"sourceMaps":{"default":true,"description":"Use JavaScript source maps (if they exist).","type":"boolean"},"timeout":{"default":10000,"description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","type":"number"},"timeouts":{"additionalProperties":false,"default":{},"description":"Timeouts for several debugger operations.","markdownDescription":"Timeouts for several debugger operations.","properties":{"hoverEvaluation":{"default":500,"description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","type":"number"},"sourceMapCumulativePause":{"default":1000,"description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","type":"number"},"sourceMapMinPause":{"default":1000,"description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","type":"number"}},"type":"object"},"trace":{"default":true,"description":"Configures what diagnostic output is produced.","oneOf":[{"description":"Trace may be set to 'true' to write diagnostic logs to the disk.","type":"boolean"},{"additionalProperties":false,"properties":{"logFile":{"description":"Configures where on disk logs are written.","type":["string","null"]},"stdio":{"description":"Whether to return trace data from the launched application or browser.","type":"boolean"}},"type":"object"}]},"url":{"default":"http://localhost:8080","description":"Will search for a tab with this exact url and attach to it, if found","tags":["setup"],"type":"string"},"urlFilter":{"default":"","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","type":"string"},"useWebView":{"default":false,"description":"When 'true', the debugger will treat the runtime executable as a host application that contains a WebView allowing you to debug the WebView script content.","type":"boolean"},"userDataDir":{"default":true,"description":"By default, the browser is launched with a separate user profile in a temp folder. Use this option to override it. Set to false to launch with your default user profile. A new browser can't be launched if an instance is already running from `userDataDir`.","type":["string","boolean"]},"vueComponentPaths":{"default":["${workspaceFolder}/**/*.vue"],"description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","type":"array"},"webRoot":{"default":"${workspaceFolder}","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","tags":["setup"],"type":"string"}}}},"configurationSnippets":[{"body":{"name":"Launch Edge","request":"launch","type":"msedge","url":"http://localhost:8080","webRoot":"^\"${2:\\${workspaceFolder\\}}\""},"description":"Launch Edge to debug a URL","label":"Edge: Launch"},{"body":{"name":"Attach to Edge","port":9222,"request":"attach","type":"msedge","webRoot":"^\"${2:\\${workspaceFolder\\}}\""},"description":"Attach to an instance of Edge already in debug mode","label":"Edge: Attach"}],"label":"Web App (Edge)","strings":{"unverifiedBreakpoints":"Some of your breakpoints could not be set. If you're having an issue, you can [troubleshoot your launch configuration](command:extension.js-debug.createDiagnostics)."},"type":"msedge"}],"commands":[{"command":"extension.js-debug.prettyPrint","title":"Pretty print for debugging","category":"Debug","icon":"$(json)"},{"command":"extension.js-debug.toggleSkippingFile","title":"Toggle Skipping this File","category":"Debug"},{"command":"extension.js-debug.addCustomBreakpoints","title":"Toggle Event Listener Breakpoints","icon":"$(add)"},{"command":"extension.js-debug.removeAllCustomBreakpoints","title":"Remove All Event Listener Breakpoints","icon":"$(close-all)"},{"command":"extension.js-debug.addXHRBreakpoints","title":"Add XHR/fetch Breakpoint","icon":"$(add)"},{"command":"extension.js-debug.removeXHRBreakpoint","title":"Remove XHR/fetch Breakpoint","icon":"$(remove)"},{"command":"extension.js-debug.editXHRBreakpoints","title":"Edit XHR/fetch Breakpoint","icon":"$(edit)"},{"command":"extension.pwa-node-debug.attachNodeProcess","title":"Attach to Node Process","category":"Debug"},{"command":"extension.js-debug.npmScript","title":"Debug npm Script","category":"Debug"},{"command":"extension.js-debug.createDebuggerTerminal","title":"JavaScript Debug Terminal","category":"Debug"},{"command":"extension.js-debug.startProfile","title":"Take Performance Profile","category":"Debug","icon":"$(record)"},{"command":"extension.js-debug.stopProfile","title":"Stop Performance Profile","category":"Debug","icon":"resources/dark/stop-profiling.svg"},{"command":"extension.js-debug.revealPage","title":"Focus Tab","category":"Debug"},{"command":"extension.js-debug.debugLink","title":"Open Link","category":"Debug"},{"command":"extension.js-debug.createDiagnostics","title":"Diagnose Breakpoint Problems","category":"Debug"},{"command":"extension.js-debug.getDiagnosticLogs","title":"Save Diagnostic JS Debug Logs","category":"Debug"},{"command":"extension.node-debug.startWithStopOnEntry","title":"Start Debugging and Stop on Entry","category":"Debug"},{"command":"extension.js-debug.openEdgeDevTools","title":"Open Browser Devtools","icon":"$(inspect)","category":"Debug"},{"command":"extension.js-debug.callers.add","title":"Exclude Caller","category":"Debug"},{"command":"extension.js-debug.callers.remove","title":"Remove excluded caller","icon":"$(close)"},{"command":"extension.js-debug.callers.removeAll","title":"Remove all excluded callers","icon":"$(clear-all)"},{"command":"extension.js-debug.callers.goToCaller","title":"Go to caller location","icon":"$(call-outgoing)"},{"command":"extension.js-debug.callers.gotToTarget","title":"Go to target location","icon":"$(call-incoming)"},{"command":"extension.js-debug.enableSourceMapStepping","title":"Enable Source Mapped Stepping","icon":"$(compass-dot)"},{"command":"extension.js-debug.disableSourceMapStepping","title":"Disable Source Mapped Stepping","icon":"$(compass)"},{"command":"extension.js-debug.network.viewRequest","title":"View Request as cURL","icon":"$(arrow-right)"},{"command":"extension.js-debug.network.clear","title":"Clear Network Log","icon":"$(clear-all)"},{"command":"extension.js-debug.network.openBody","title":"Open Response Body"},{"command":"extension.js-debug.network.openBodyInHex","title":"Open Response Body in Hex Editor"},{"command":"extension.js-debug.network.replayXHR","title":"Replay Request"},{"command":"extension.js-debug.network.copyUri","title":"Copy Request URL"}],"keybindings":[{"command":"extension.node-debug.startWithStopOnEntry","key":"F10","mac":"F10","when":"debugConfigurationType == pwa-node && !inDebugMode || debugConfigurationType == pwa-extensionHost && !inDebugMode || debugConfigurationType == node && !inDebugMode"},{"command":"extension.node-debug.startWithStopOnEntry","key":"F11","mac":"F11","when":"debugConfigurationType == pwa-node && !inDebugMode && activeViewlet == workbench.view.debug || debugConfigurationType == pwa-extensionHost && !inDebugMode && activeViewlet == workbench.view.debug || debugConfigurationType == node && !inDebugMode && activeViewlet == workbench.view.debug"}],"configuration":{"title":"JavaScript Debugger","properties":{"debug.javascript.codelens.npmScripts":{"enum":["top","all","never"],"default":"top","description":"Where a \"Run\" and \"Debug\" code lens should be shown in your npm scripts. It may be on \"all\", scripts, on \"top\" of the script section, or \"never\"."},"debug.javascript.terminalOptions":{"type":"object","description":"Default launch options for the JavaScript debug terminal and npm scripts.","default":{},"properties":{"resolveSourceMapLocations":{"type":["array","null"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","default":["${workspaceFolder}/**","!**/node_modules/**"],"items":{"type":"string"}},"outFiles":{"type":["array"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"items":{"type":"string"},"tags":["setup"]},"pauseForSourceMap":{"type":"boolean","markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","default":false},"showAsyncStacks":{"description":"Show the async calls that led to the current call stack.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","required":["onAttach"],"properties":{"onAttach":{"type":"number","default":32}}},{"type":"object","required":["onceBreakpointResolved"],"properties":{"onceBreakpointResolved":{"type":"number","default":32}}}]},"skipFiles":{"type":"array","description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","default":["${/**"]},"smartStep":{"type":"boolean","description":"Automatically step through generated code that cannot be mapped back to the original source.","default":true},"sourceMaps":{"type":"boolean","description":"Use JavaScript source maps (if they exist).","default":true},"sourceMapRenames":{"type":"boolean","default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers."},"sourceMapPathOverrides":{"type":"object","description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","default":{"webpack://?:*/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","meteor://💻app/*":"${workspaceFolder}/*"}},"timeout":{"type":"number","description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","default":10000},"timeouts":{"type":"object","description":"Timeouts for several debugger operations.","default":{},"properties":{"sourceMapMinPause":{"type":"number","description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","default":1000},"sourceMapCumulativePause":{"type":"number","description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","default":1000},"hoverEvaluation":{"type":"number","description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","default":500}},"additionalProperties":false,"markdownDescription":"Timeouts for several debugger operations."},"trace":{"description":"Configures what diagnostic output is produced.","default":true,"oneOf":[{"type":"boolean","description":"Trace may be set to 'true' to write diagnostic logs to the disk."},{"type":"object","additionalProperties":false,"properties":{"stdio":{"type":"boolean","description":"Whether to return trace data from the launched application or browser."},"logFile":{"type":["string","null"],"description":"Configures where on disk logs are written."}}}]},"outputCapture":{"enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`.","default":"console"},"enableContentValidation":{"default":true,"type":"boolean","description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example."},"customDescriptionGenerator":{"type":"string","description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      "},"customPropertiesGenerator":{"type":"string","deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181"},"cascadeTerminateToConfigurations":{"type":"array","items":{"type":"string","uniqueItems":true},"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped."},"enableDWARF":{"type":"boolean","default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function."},"cwd":{"type":"string","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","default":"${workspaceFolder}","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"]},"localRoot":{"type":["string","null"],"description":"Path to the local directory containing the program.","default":null},"remoteRoot":{"type":["string","null"],"description":"Absolute path to the remote directory containing the program.","default":null},"autoAttachChildProcesses":{"type":"boolean","description":"Attach debugger to new child processes automatically.","default":true},"env":{"type":"object","additionalProperties":{"type":["string","null"]},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","default":{},"tags":["setup"]},"envFile":{"type":"string","description":"Absolute path to a file containing environment variable definitions.","default":"${workspaceFolder}/.env"},"runtimeSourcemapPausePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","default":[]},"nodeVersionHint":{"type":"number","minimum":8,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","default":12},"command":{"type":["string","null"],"description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","default":"npm start","tags":["setup"]}}},"debug.javascript.automaticallyTunnelRemoteServer":{"type":"boolean","description":"When debugging a remote web app, configures whether to automatically tunnel the remote server to your local machine.","default":true},"debug.javascript.debugByLinkOptions":{"default":"on","description":"Options used when debugging open links clicked from inside the JavaScript Debug Terminal. Can be set to \"off\" to disable this behavior, or \"always\" to enable debugging in all terminals.","oneOf":[{"type":"string","enum":["on","off","always"]},{"type":"object","properties":{"resolveSourceMapLocations":{"type":["array","null"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","default":null,"items":{"type":"string"}},"outFiles":{"type":["array"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"items":{"type":"string"},"tags":["setup"]},"pauseForSourceMap":{"type":"boolean","markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","default":false},"showAsyncStacks":{"description":"Show the async calls that led to the current call stack.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","required":["onAttach"],"properties":{"onAttach":{"type":"number","default":32}}},{"type":"object","required":["onceBreakpointResolved"],"properties":{"onceBreakpointResolved":{"type":"number","default":32}}}]},"skipFiles":{"type":"array","description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","default":["${/**"]},"smartStep":{"type":"boolean","description":"Automatically step through generated code that cannot be mapped back to the original source.","default":true},"sourceMaps":{"type":"boolean","description":"Use JavaScript source maps (if they exist).","default":true},"sourceMapRenames":{"type":"boolean","default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers."},"sourceMapPathOverrides":{"type":"object","description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","default":{"webpack://?:*/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","meteor://💻app/*":"${workspaceFolder}/*"}},"timeout":{"type":"number","description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","default":10000},"timeouts":{"type":"object","description":"Timeouts for several debugger operations.","default":{},"properties":{"sourceMapMinPause":{"type":"number","description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","default":1000},"sourceMapCumulativePause":{"type":"number","description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","default":1000},"hoverEvaluation":{"type":"number","description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","default":500}},"additionalProperties":false,"markdownDescription":"Timeouts for several debugger operations."},"trace":{"description":"Configures what diagnostic output is produced.","default":true,"oneOf":[{"type":"boolean","description":"Trace may be set to 'true' to write diagnostic logs to the disk."},{"type":"object","additionalProperties":false,"properties":{"stdio":{"type":"boolean","description":"Whether to return trace data from the launched application or browser."},"logFile":{"type":["string","null"],"description":"Configures where on disk logs are written."}}}]},"outputCapture":{"enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`.","default":"console"},"enableContentValidation":{"default":true,"type":"boolean","description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example."},"customDescriptionGenerator":{"type":"string","description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      "},"customPropertiesGenerator":{"type":"string","deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181"},"cascadeTerminateToConfigurations":{"type":"array","items":{"type":"string","uniqueItems":true},"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped."},"enableDWARF":{"type":"boolean","default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function."},"disableNetworkCache":{"type":"boolean","description":"Controls whether to skip the network cache for each request","default":true},"pathMapping":{"type":"object","description":"A mapping of URLs/paths to local folders, to resolve scripts in the Browser to scripts on disk","default":{}},"webRoot":{"type":"string","description":"This specifies the workspace absolute path to the webserver root. Used to resolve paths like `/app.js` to files on disk. Shorthand for a pathMapping for \"/\"","default":"${workspaceFolder}","tags":["setup"]},"urlFilter":{"type":"string","description":"Will search for a page with this url and attach to it, if found. Can have * wildcards.","default":""},"url":{"type":"string","description":"Will search for a tab with this exact url and attach to it, if found","default":"http://localhost:8080","tags":["setup"]},"inspectUri":{"type":["string","null"],"description":"Format to use to rewrite the inspectUri: It's a template string that interpolates keys in `{curlyBraces}`. Available keys are:\n - `url.*` is the parsed address of the running application. For instance, `{url.port}`, `{url.hostname}`\n - `port` is the debug port that Chrome is listening on.\n - `browserInspectUri` is the inspector URI on the launched browser\n - `browserInspectUriPath` is the path part of the inspector URI on the launched browser (e.g.: \"/devtools/browser/e9ec0098-306e-472a-8133-5e42488929c2\").\n - `wsProtocol` is the hinted websocket protocol. This is set to `wss` if the original URL is `https`, or `ws` otherwise.\n","default":null},"vueComponentPaths":{"type":"array","description":"A list of file glob patterns to find `*.vue` components. By default, searches the entire workspace. This needs to be specified due to extra lookups that Vue's sourcemaps require in Vue CLI 4. You can disable this special handling by setting this to an empty array.","default":["${workspaceFolder}/**/*.vue"]},"server":{"oneOf":[{"type":"object","description":"Configures a web server to start up. Takes the same configuration as the 'node' launch task.","additionalProperties":false,"default":{"program":"node my-server.js"},"properties":{"resolveSourceMapLocations":{"type":["array","null"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","default":["${workspaceFolder}/**","!**/node_modules/**"],"items":{"type":"string"}},"outFiles":{"type":["array"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"items":{"type":"string"},"tags":["setup"]},"pauseForSourceMap":{"type":"boolean","markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","default":false},"showAsyncStacks":{"description":"Show the async calls that led to the current call stack.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","required":["onAttach"],"properties":{"onAttach":{"type":"number","default":32}}},{"type":"object","required":["onceBreakpointResolved"],"properties":{"onceBreakpointResolved":{"type":"number","default":32}}}]},"skipFiles":{"type":"array","description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","default":["${/**"]},"smartStep":{"type":"boolean","description":"Automatically step through generated code that cannot be mapped back to the original source.","default":true},"sourceMaps":{"type":"boolean","description":"Use JavaScript source maps (if they exist).","default":true},"sourceMapRenames":{"type":"boolean","default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers."},"sourceMapPathOverrides":{"type":"object","description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","default":{"webpack://?:*/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","meteor://💻app/*":"${workspaceFolder}/*"}},"timeout":{"type":"number","description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","default":10000},"timeouts":{"type":"object","description":"Timeouts for several debugger operations.","default":{},"properties":{"sourceMapMinPause":{"type":"number","description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","default":1000},"sourceMapCumulativePause":{"type":"number","description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","default":1000},"hoverEvaluation":{"type":"number","description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","default":500}},"additionalProperties":false,"markdownDescription":"Timeouts for several debugger operations."},"trace":{"description":"Configures what diagnostic output is produced.","default":true,"oneOf":[{"type":"boolean","description":"Trace may be set to 'true' to write diagnostic logs to the disk."},{"type":"object","additionalProperties":false,"properties":{"stdio":{"type":"boolean","description":"Whether to return trace data from the launched application or browser."},"logFile":{"type":["string","null"],"description":"Configures where on disk logs are written."}}}]},"outputCapture":{"enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`.","default":"console"},"enableContentValidation":{"default":true,"type":"boolean","description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example."},"customDescriptionGenerator":{"type":"string","description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      "},"customPropertiesGenerator":{"type":"string","deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181"},"cascadeTerminateToConfigurations":{"type":"array","items":{"type":"string","uniqueItems":true},"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped."},"enableDWARF":{"type":"boolean","default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function."},"cwd":{"type":"string","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","default":"${workspaceFolder}","tags":["setup"]},"localRoot":{"type":["string","null"],"description":"Path to the local directory containing the program.","default":null},"remoteRoot":{"type":["string","null"],"description":"Absolute path to the remote directory containing the program.","default":null},"autoAttachChildProcesses":{"type":"boolean","description":"Attach debugger to new child processes automatically.","default":true},"env":{"type":"object","additionalProperties":{"type":["string","null"]},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","default":{},"tags":["setup"]},"envFile":{"type":"string","description":"Absolute path to a file containing environment variable definitions.","default":"${workspaceFolder}/.env"},"runtimeSourcemapPausePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","default":[]},"nodeVersionHint":{"type":"number","minimum":8,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","default":12},"program":{"type":"string","description":"Absolute path to the program. Generated value is guessed by looking at package.json and opened files. Edit this attribute.","default":"","tags":["setup"]},"stopOnEntry":{"type":["boolean","string"],"description":"Automatically stop program after launch.","default":true},"console":{"type":"string","enum":["internalConsole","integratedTerminal","externalTerminal"],"enumDescriptions":["VS Code Debug Console (which doesn't support to read input from a program)","VS Code's integrated terminal","External terminal that can be configured via user settings"],"description":"Where to launch the debug target.","default":"internalConsole"},"args":{"type":["array","string"],"description":"Command line arguments passed to the program.\n\nCan be an array of strings or a single string. When the program is launched in a terminal, setting this property to a single string will result in the arguments not being escaped for the shell.","items":{"type":"string"},"default":[],"tags":["setup"]},"restart":{"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","properties":{"delay":{"type":"number","minimum":0,"default":1000},"maxAttempts":{"type":"number","minimum":0,"default":10}}}]},"runtimeExecutable":{"type":["string","null"],"markdownDescription":"Runtime to use. Either an absolute path or the name of a runtime available on the PATH. If omitted `node` is assumed.","default":"node"},"runtimeVersion":{"type":"string","markdownDescription":"Version of `node` runtime to use. Requires `nvm`.","default":"default"},"runtimeArgs":{"type":"array","description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"default":[],"tags":["setup"]},"profileStartup":{"type":"boolean","description":"If true, will start profiling as soon as the process launches","default":true},"attachSimplePort":{"oneOf":[{"type":"integer"},{"type":"string","pattern":"^\\${.*}$"}],"description":"If set, attaches to the process via the given port. This is generally no longer necessary for Node.js programs and loses the ability to debug child processes, but can be useful in more esoteric scenarios such as with Deno and Docker launches. If set to 0, a random port will be chosen and --inspect-brk added to the launch arguments automatically.","default":9229},"killBehavior":{"type":"string","enum":["forceful","polite","none"],"default":"forceful","markdownDescription":"Configures how debug processes are killed when stopping the session. Can be:\n\n- forceful (default): forcefully tears down the process tree. Sends SIGKILL on posix, or `taskkill.exe /F` on Windows.\n- polite: gracefully tears down the process tree. It's possible that misbehaving processes continue to run after shutdown in this way. Sends SIGTERM on posix, or `taskkill.exe` with no `/F` (force) flag on Windows.\n- none: no termination will happen."},"experimentalNetworking":{"type":"string","default":"auto","description":"Enable experimental inspection in Node.js. When set to `auto` this is enabled for versions of Node.js that support it. It can be set to `on` or `off` to enable or disable it explicitly.","enum":["auto","on","off"]}}},{"type":"object","description":"JavaScript Debug Terminal","additionalProperties":false,"default":{"program":"npm start"},"properties":{"resolveSourceMapLocations":{"type":["array","null"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","default":["${workspaceFolder}/**","!**/node_modules/**"],"items":{"type":"string"}},"outFiles":{"type":["array"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"items":{"type":"string"},"tags":["setup"]},"pauseForSourceMap":{"type":"boolean","markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","default":false},"showAsyncStacks":{"description":"Show the async calls that led to the current call stack.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","required":["onAttach"],"properties":{"onAttach":{"type":"number","default":32}}},{"type":"object","required":["onceBreakpointResolved"],"properties":{"onceBreakpointResolved":{"type":"number","default":32}}}]},"skipFiles":{"type":"array","description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","default":["${/**"]},"smartStep":{"type":"boolean","description":"Automatically step through generated code that cannot be mapped back to the original source.","default":true},"sourceMaps":{"type":"boolean","description":"Use JavaScript source maps (if they exist).","default":true},"sourceMapRenames":{"type":"boolean","default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers."},"sourceMapPathOverrides":{"type":"object","description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","default":{"webpack://?:*/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","meteor://💻app/*":"${workspaceFolder}/*"}},"timeout":{"type":"number","description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","default":10000},"timeouts":{"type":"object","description":"Timeouts for several debugger operations.","default":{},"properties":{"sourceMapMinPause":{"type":"number","description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","default":1000},"sourceMapCumulativePause":{"type":"number","description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","default":1000},"hoverEvaluation":{"type":"number","description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","default":500}},"additionalProperties":false,"markdownDescription":"Timeouts for several debugger operations."},"trace":{"description":"Configures what diagnostic output is produced.","default":true,"oneOf":[{"type":"boolean","description":"Trace may be set to 'true' to write diagnostic logs to the disk."},{"type":"object","additionalProperties":false,"properties":{"stdio":{"type":"boolean","description":"Whether to return trace data from the launched application or browser."},"logFile":{"type":["string","null"],"description":"Configures where on disk logs are written."}}}]},"outputCapture":{"enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`.","default":"console"},"enableContentValidation":{"default":true,"type":"boolean","description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example."},"customDescriptionGenerator":{"type":"string","description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      "},"customPropertiesGenerator":{"type":"string","deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181"},"cascadeTerminateToConfigurations":{"type":"array","items":{"type":"string","uniqueItems":true},"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped."},"enableDWARF":{"type":"boolean","default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function."},"cwd":{"type":"string","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","default":"${workspaceFolder}","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"]},"localRoot":{"type":["string","null"],"description":"Path to the local directory containing the program.","default":null},"remoteRoot":{"type":["string","null"],"description":"Absolute path to the remote directory containing the program.","default":null},"autoAttachChildProcesses":{"type":"boolean","description":"Attach debugger to new child processes automatically.","default":true},"env":{"type":"object","additionalProperties":{"type":["string","null"]},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","default":{},"tags":["setup"]},"envFile":{"type":"string","description":"Absolute path to a file containing environment variable definitions.","default":"${workspaceFolder}/.env"},"runtimeSourcemapPausePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","default":[]},"nodeVersionHint":{"type":"number","minimum":8,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","default":12},"command":{"type":["string","null"],"description":"Command to run in the launched terminal. If not provided, the terminal will open without launching a program.","default":"npm start","tags":["setup"]}}}]},"perScriptSourcemaps":{"type":"string","default":"auto","enum":["yes","no","auto"],"description":"Whether scripts are loaded individually with unique sourcemaps containing the basename of the source file. This can be set to optimize sourcemap handling when dealing with lots of small scripts. If set to \"auto\", we'll detect known cases where this is appropriate."},"port":{"type":"number","description":"Port for the browser to listen on. Defaults to \"0\", which will cause the browser to be debugged via pipes, which is generally more secure and should be chosen unless you need to attach to the browser from another tool.","default":0},"file":{"type":"string","description":"A local html file to open in the browser","default":"${workspaceFolder}/index.html","tags":["setup"]},"userDataDir":{"type":["string","boolean"],"description":"By default, the browser is launched with a separate user profile in a temp folder. Use this option to override it. Set to false to launch with your default user profile. A new browser can't be launched if an instance is already running from `userDataDir`.","default":true},"includeDefaultArgs":{"type":"boolean","description":"Whether default browser launch arguments (to disable features that may make debugging harder) will be included in the launch.","default":true},"includeLaunchArgs":{"type":"boolean","description":"Advanced: whether any default launch/debugging arguments are set on the browser. The debugger will assume the browser will use pipe debugging such as that which is provided with `--remote-debugging-pipe`.","default":true},"runtimeExecutable":{"type":["string","null"],"description":"Either 'canary', 'stable', 'custom' or path to the browser executable. Custom means a custom wrapper, custom build or CHROME_PATH environment variable.","default":"stable"},"runtimeArgs":{"type":"array","description":"Optional arguments passed to the runtime executable.","items":{"type":"string"},"default":[]},"env":{"type":"object","description":"Optional dictionary of environment key/value pairs for the browser.","default":{}},"cwd":{"type":"string","description":"Optional working directory for the runtime executable.","default":null},"profileStartup":{"type":"boolean","description":"If true, will start profiling soon as the process launches","default":true},"cleanUp":{"type":"string","enum":["wholeBrowser","onlyTab"],"description":"What clean-up to do after the debugging session finishes. Close only the tab being debug, vs. close the whole browser.","default":"wholeBrowser"},"browserLaunchLocation":{"description":"Forces the browser to be launched in one location. In a remote workspace (through ssh or WSL, for example) this can be used to open the browser on the remote machine rather than locally.","default":null,"oneOf":[{"type":"null"},{"type":"string","enum":["ui","workspace"]}]},"enabled":{"type":"string","enum":["on","off","always"]}}}]},"debug.javascript.pickAndAttachOptions":{"type":"object","default":{},"markdownDescription":"Default options used when debugging a process through the `Debug: Attach to Node.js Process` command","properties":{"resolveSourceMapLocations":{"type":["array","null"],"description":"A list of minimatch patterns for locations (folders and URLs) in which source maps can be used to resolve local files. This can be used to avoid incorrectly breaking in external source mapped code. Patterns can be prefixed with \"!\" to exclude them. May be set to an empty array or null to avoid restriction.","default":["${workspaceFolder}/**","!**/node_modules/**"],"items":{"type":"string"}},"outFiles":{"type":["array"],"description":"If source maps are enabled, these glob patterns specify the generated JavaScript files. If a pattern starts with `!` the files are excluded. If not specified, the generated code is expected in the same directory as its source.","default":["${workspaceFolder}/**/*.(m|c|)js","!**/node_modules/**"],"items":{"type":"string"},"tags":["setup"]},"pauseForSourceMap":{"type":"boolean","markdownDescription":"Whether to wait for source maps to load for each incoming script. This has a performance overhead, and might be safely disabled when running off of disk, so long as `rootPath` is not disabled.","default":false},"showAsyncStacks":{"description":"Show the async calls that led to the current call stack.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","required":["onAttach"],"properties":{"onAttach":{"type":"number","default":32}}},{"type":"object","required":["onceBreakpointResolved"],"properties":{"onceBreakpointResolved":{"type":"number","default":32}}}]},"skipFiles":{"type":"array","description":"An array of file or folder names, or path globs, to skip when debugging. Star patterns and negations are allowed, for example, `[\"**/node_modules/**\", \"!**/node_modules/my-module/**\"]`","default":["${/**"]},"smartStep":{"type":"boolean","description":"Automatically step through generated code that cannot be mapped back to the original source.","default":true},"sourceMaps":{"type":"boolean","description":"Use JavaScript source maps (if they exist).","default":true},"sourceMapRenames":{"type":"boolean","default":true,"description":"Whether to use the \"names\" mapping in sourcemaps. This requires requesting source content, which can be slow with certain debuggers."},"sourceMapPathOverrides":{"type":"object","description":"A set of mappings for rewriting the locations of source files from what the sourcemap says, to their locations on disk.","default":{"webpack://?:*/*":"${workspaceFolder}/*","webpack:///./~/*":"${workspaceFolder}/node_modules/*","meteor://💻app/*":"${workspaceFolder}/*"}},"timeout":{"type":"number","description":"Retry for this number of milliseconds to connect to Node.js. Default is 10000 ms.","default":10000},"timeouts":{"type":"object","description":"Timeouts for several debugger operations.","default":{},"properties":{"sourceMapMinPause":{"type":"number","description":"Minimum time in milliseconds spent waiting for each source-map to be processed when a script is being parsed","default":1000},"sourceMapCumulativePause":{"type":"number","description":"Extra time in milliseconds allowed per session to be spent waiting for source-maps to be processed, after the minimum time (sourceMapMinPause) has been exhausted","default":1000},"hoverEvaluation":{"type":"number","description":"Time until value evaluation for hovered symbols is aborted. If set to 0, hover evaluation does never time out.","default":500}},"additionalProperties":false,"markdownDescription":"Timeouts for several debugger operations."},"trace":{"description":"Configures what diagnostic output is produced.","default":true,"oneOf":[{"type":"boolean","description":"Trace may be set to 'true' to write diagnostic logs to the disk."},{"type":"object","additionalProperties":false,"properties":{"stdio":{"type":"boolean","description":"Whether to return trace data from the launched application or browser."},"logFile":{"type":["string","null"],"description":"Configures where on disk logs are written."}}}]},"outputCapture":{"enum":["console","std"],"markdownDescription":"From where to capture output messages: the default debug API if set to `console`, or stdout/stderr streams if set to `std`.","default":"console"},"enableContentValidation":{"default":true,"type":"boolean","description":"Toggles whether we verify the contents of files on disk match the ones loaded in the runtime. This is useful in a variety of scenarios and required in some, but can cause issues if you have server-side transformation of scripts, for example."},"customDescriptionGenerator":{"type":"string","description":"Customize the textual description the debugger shows for objects (local variables, etc...). Samples:\n      1. this.toString() // will call toString to print all objects\n      2. this.customDescription ? this.customDescription() : defaultValue // Use customDescription method if available, if not return defaultValue\n      3. function (def) { return this.customDescription ? this.customDescription() : def } // Use customDescription method if available, if not return defaultValue\n      "},"customPropertiesGenerator":{"type":"string","deprecated":true,"description":"Customize the properties shown for an object in the debugger (local variables, etc...). Samples:\n    1. { ...this, extraProperty: '12345' } // Add an extraProperty 12345 to all objects\n    2. this.customProperties ? this.customProperties() : this // Use customProperties method if available, if not use the properties in this (the default properties)\n    3. function () { return this.customProperties ? this.customProperties() : this } // Use customDescription method if available, if not return the default properties\n\n    Deprecated: This is a temporary implementation of this feature until we have time to implement it in the way described here: https://github.com/microsoft/vscode/issues/102181"},"cascadeTerminateToConfigurations":{"type":"array","items":{"type":"string","uniqueItems":true},"default":[],"description":"A list of debug sessions which, when this debug session is terminated, will also be stopped."},"enableDWARF":{"type":"boolean","default":true,"markdownDescription":"Toggles whether the debugger will try to read DWARF debug symbols from WebAssembly, which can be resource intensive. Requires the `ms-vscode.wasm-dwarf-debugging` extension to function."},"cwd":{"type":"string","description":"Absolute path to the working directory of the program being debugged. If you've set localRoot then cwd will match that value otherwise it falls back to your workspaceFolder","default":"${workspaceFolder}","docDefault":"localRoot || ${workspaceFolder}","tags":["setup"]},"localRoot":{"type":["string","null"],"description":"Path to the local directory containing the program.","default":null},"remoteRoot":{"type":["string","null"],"description":"Absolute path to the remote directory containing the program.","default":null},"autoAttachChildProcesses":{"type":"boolean","description":"Attach debugger to new child processes automatically.","default":true},"env":{"type":"object","additionalProperties":{"type":["string","null"]},"markdownDescription":"Environment variables passed to the program. The value `null` removes the variable from the environment.","default":{},"tags":["setup"]},"envFile":{"type":"string","description":"Absolute path to a file containing environment variable definitions.","default":"${workspaceFolder}/.env"},"runtimeSourcemapPausePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"A list of patterns at which to manually insert entrypoint breakpoints. This can be useful to give the debugger an opportunity to set breakpoints when using sourcemaps that don't exist or can't be detected before launch, such as [with the Serverless framework](https://github.com/microsoft/vscode-js-debug/issues/492).","default":[]},"nodeVersionHint":{"type":"number","minimum":8,"description":"Allows you to explicitly specify the Node version that's running, which can be used to disable or enable certain behaviors in cases where the automatic version detection does not work.","default":12},"address":{"type":"string","description":"TCP/IP address of process to be debugged. Default is 'localhost'.","default":"localhost"},"port":{"description":"Debug port to attach to. Default is 9229.","default":9229,"oneOf":[{"type":"integer"},{"type":"string","pattern":"^\\${.*}$"}],"tags":["setup"]},"websocketAddress":{"type":"string","description":"Exact websocket address to attach to. If unspecified, it will be discovered from the address and port."},"remoteHostHeader":{"type":"string","description":"Explicit Host header to use when connecting to the websocket of inspector. If unspecified, the host header will be set to 'localhost'. This is useful when the inspector is running behind a proxy that only accept particular Host header."},"restart":{"description":"Try to reconnect to the program if we lose connection. If set to `true`, we'll try once a second, forever. You can customize the interval and maximum number of attempts by specifying the `delay` and `maxAttempts` in an object instead.","default":true,"oneOf":[{"type":"boolean"},{"type":"object","properties":{"delay":{"type":"number","minimum":0,"default":1000},"maxAttempts":{"type":"number","minimum":0,"default":10}}}]},"processId":{"type":"string","description":"ID of process to attach to.","default":"${command:PickProcess}"},"attachExistingChildren":{"type":"boolean","description":"Whether to attempt to attach to already-spawned child processes.","default":false},"continueOnAttach":{"type":"boolean","markdownDescription":"If true, we'll automatically resume programs launched and waiting on `--inspect-brk`","default":true}}},"debug.javascript.autoAttachFilter":{"type":"string","default":"disabled","enum":["always","smart","onlyWithFlag","disabled"],"enumDescriptions":["Auto attach to every Node.js process launched in the terminal.","Auto attach when running scripts that aren't in a node_modules folder.","Only auto attach when the `--inspect` is given.","Auto attach is disabled and not shown in status bar."],"markdownDescription":"Configures which processes to automatically attach and debug when `#debug.node.autoAttach#` is on. A Node process launched with the `--inspect` flag will always be attached to, regardless of this setting."},"debug.javascript.autoAttachSmartPattern":{"type":"array","items":{"type":"string"},"default":["${workspaceFolder}/**","!**/node_modules/**","**/$KNOWN_TOOLS$/**"],"markdownDescription":"Configures glob patterns for determining when to attach in \"smart\" `#debug.javascript.autoAttachFilter#` mode. `$KNOWN_TOOLS$` is replaced with a list of names of common test and code runners. [Read more on the VS Code docs](https://code.visualstudio.com/docs/nodejs/nodejs-debugging#_auto-attach-smart-patterns)."},"debug.javascript.breakOnConditionalError":{"type":"boolean","default":false,"markdownDescription":"Whether to stop when conditional breakpoints throw an error."},"debug.javascript.unmapMissingSources":{"type":"boolean","default":false,"description":"Configures whether sourcemapped file where the original file can't be read will automatically be unmapped. If this is false (default), a prompt is shown."},"debug.javascript.defaultRuntimeExecutable":{"type":"object","default":{"pwa-node":"node"},"markdownDescription":"The default `runtimeExecutable` used for launch configurations, if unspecified. This can be used to config custom paths to Node.js or browser installations.","properties":{"pwa-node":{"type":"string"},"pwa-chrome":{"type":"string"},"pwa-msedge":{"type":"string"}}},"debug.javascript.resourceRequestOptions":{"type":"object","default":{},"markdownDescription":"Request options to use when loading resources, such as source maps, in the debugger. You may need to configure this if your sourcemaps require authentication or use a self-signed certificate, for instance. Options are used to create a request using the [`got`](https://github.com/sindresorhus/got) library.\n\nA common case to disable certificate verification can be done by passing `{ \"https\": { \"rejectUnauthorized\": false } }`."},"debug.javascript.enableNetworkView":{"type":"boolean","default":false,"description":"Enables the experimental network view for targets that support it."}}},"grammars":[{"language":"wat","scopeName":"text.wat","path":"./src/ui/basic-wat.tmLanguage.json"}],"languages":[{"id":"wat","extensions":[".wat",".wasm"],"aliases":["WebAssembly Text Format"],"firstLine":"^\\(module","mimetypes":["text/wat"],"configuration":"./src/ui/basic-wat.configuration.json"}],"terminal":{"profiles":[{"id":"extension.js-debug.debugTerminal","title":"JavaScript Debug Terminal","icon":"$(debug)"}]},"views":{"debug":[{"id":"jsBrowserBreakpoints","name":"Event Listener Breakpoints","when":"debugType == pwa-chrome || debugType == pwa-msedge"},{"id":"jsExcludedCallers","name":"Excluded Callers","when":"debugType == pwa-extensionHost && jsDebugHasExcludedCallers || debugType == node-terminal && jsDebugHasExcludedCallers || debugType == pwa-node && jsDebugHasExcludedCallers || debugType == pwa-chrome && jsDebugHasExcludedCallers || debugType == pwa-msedge && jsDebugHasExcludedCallers"},{"id":"jsDebugNetworkTree","name":"Network","when":"jsDebugNetworkAvailable"}]},"viewsWelcome":[{"view":"debug","contents":"[JavaScript Debug Terminal](command:extension.js-debug.createDebuggerTerminal)\n\nYou can use the JavaScript Debug Terminal to debug Node.js processes run on the command line.\n\n[Debug URL](command:extension.js-debug.debugLink)","when":"debugStartLanguage == javascript && !isWeb || debugStartLanguage == typescript && !isWeb || debugStartLanguage == javascriptreact && !isWeb || debugStartLanguage == typescriptreact && !isWeb"},{"view":"debug","contents":"[JavaScript Debug Terminal](command:extension.js-debug.createDebuggerTerminal)\n\nYou can use the JavaScript Debug Terminal to debug Node.js processes run on the command line.","when":"debugStartLanguage == javascript && isWeb || debugStartLanguage == typescript && isWeb || debugStartLanguage == javascriptreact && isWeb || debugStartLanguage == typescriptreact && isWeb"}]}},"location":{"$mid":1,"path":"/ide/extensions/ms-vscode.js-debug","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","publisherDisplayName":"Microsoft","metadata":{"id":"25629058-ddac-4e17-abba-74678e126c5d","publisherId":{"publisherId":"5f5636e7-69ed-4afe-b5d6-8d231fb3d3ee","publisherName":"ms-vscode","displayName":"Microsoft","flags":"verified"},"publisherDisplayName":"Microsoft"},"isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"ms-vscode.js-debug-companion","uuid":"99cb0b7f-7354-4278-b8da-6cc79972169d"},"manifest":{"name":"js-debug-companion","displayName":"JavaScript Debugger Companion Extension","description":"Companion extension to js-debug that provides capability for remote debugging","version":"1.1.3","publisher":"ms-vscode","engines":{"vscode":"^1.90.0"},"icon":"resources/logo.png","categories":["Other"],"repository":{"type":"git","url":"https://github.com/microsoft/vscode-js-debug-companion.git"},"author":"Connor Peet <connor@peet.io>","license":"MIT","bugs":{"url":"https://github.com/microsoft/vscode-js-debug-companion/issues"},"homepage":"https://github.com/microsoft/vscode-js-debug-companion#readme","capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":true}},"activationEvents":["onCommand:js-debug-companion.launchAndAttach","onCommand:js-debug-companion.kill","onCommand:js-debug-companion.launch","onCommand:js-debug-companion.defaultBrowser"],"main":"./out/extension.js","contributes":{},"extensionKind":["ui"],"api":"none","prettier":{"trailingComma":"all","singleQuote":true,"printWidth":100,"tabWidth":2,"arrowParens":"avoid"}},"location":{"$mid":1,"path":"/ide/extensions/ms-vscode.js-debug-companion","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","publisherDisplayName":"Microsoft","metadata":{"id":"99cb0b7f-7354-4278-b8da-6cc79972169d","publisherId":{"publisherId":"5f5636e7-69ed-4afe-b5d6-8d231fb3d3ee","publisherName":"ms-vscode","displayName":"Microsoft","flags":"verified"},"publisherDisplayName":"Microsoft"},"isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"ms-vscode.vscode-js-profile-table","uuid":"7e52b41b-71ad-457b-ab7e-0620f1fc4feb"},"manifest":{"name":"vscode-js-profile-table","version":"1.0.10","displayName":"Table Visualizer for JavaScript Profiles","description":"Text visualizer for profiles taken from the JavaScript debugger","author":"Connor Peet <connor@peet.io>","homepage":"https://github.com/microsoft/vscode-js-profile-visualizer#readme","license":"MIT","main":"out/extension.js","browser":"out/extension.web.js","repository":{"type":"git","url":"https://github.com/microsoft/vscode-js-profile-visualizer.git"},"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"icon":"resources/icon.png","publisher":"ms-vscode","sideEffects":false,"engines":{"vscode":"^1.74.0"},"contributes":{"customEditors":[{"viewType":"jsProfileVisualizer.cpuprofile.table","displayName":"CPU Profile Table Visualizer","priority":"default","selector":[{"filenamePattern":"*.cpuprofile"}]},{"viewType":"jsProfileVisualizer.heapprofile.table","displayName":"Heap Profile Table Visualizer","priority":"default","selector":[{"filenamePattern":"*.heapprofile"}]},{"viewType":"jsProfileVisualizer.heapsnapshot.table","displayName":"Heap Snapshot Table Visualizer","priority":"default","selector":[{"filenamePattern":"*.heapsnapshot"}]}],"commands":[{"command":"extension.jsProfileVisualizer.table.clearCodeLenses","title":"Clear Profile Code Lenses"}],"menus":{"commandPalette":[{"command":"extension.jsProfileVisualizer.table.clearCodeLenses","when":"jsProfileVisualizer.hasCodeLenses == true"}]}},"bugs":{"url":"https://github.com/microsoft/vscode-js-profile-visualizer/issues"}},"location":{"$mid":1,"path":"/ide/extensions/ms-vscode.vscode-js-profile-table","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","publisherDisplayName":"Microsoft","metadata":{"id":"7e52b41b-71ad-457b-ab7e-0620f1fc4feb","publisherId":{"publisherId":"5f5636e7-69ed-4afe-b5d6-8d231fb3d3ee","publisherName":"ms-vscode","displayName":"Microsoft","flags":"verified"},"publisherDisplayName":"Microsoft"},"isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.builtin-notebook-renderers"},"manifest":{"name":"builtin-notebook-renderers","displayName":"Builtin Notebook Output Renderers","description":"Provides basic output renderers for notebooks","publisher":"vscode","version":"1.0.0","license":"MIT","icon":"media/icon.png","engines":{"vscode":"^1.57.0"},"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"notebookRenderer":[{"id":"vscode.builtin-renderer","entrypoint":"./renderer-out/index.js","displayName":"VS Code Builtin Notebook Output Renderer","requiresMessaging":"never","mimeTypes":["image/gif","image/png","image/jpeg","image/git","image/svg+xml","text/html","application/javascript","application/vnd.code.notebook.error","application/vnd.code.notebook.stdout","application/x.notebook.stdout","application/x.notebook.stream","application/vnd.code.notebook.stderr","application/x.notebook.stderr","text/plain"]}]},"scripts":{"compile":"npx gulp compile-extension:notebook-renderers && npm run build-notebook","watch":"npx gulp compile-watch:notebook-renderers","build-notebook":"node ./esbuild"},"dependencies":{},"devDependencies":{"@types/jsdom":"^21.1.0","@types/vscode-notebook-renderer":"^1.60.0","jsdom":"^21.1.1"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/notebook-renderers","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.npm"},"manifest":{"name":"npm","publisher":"vscode","displayName":"NPM support for VS Code","description":"Extension to add task support for npm scripts.","version":"1.0.1","private":true,"license":"MIT","engines":{"vscode":"0.10.x"},"icon":"images/npm_icon.png","categories":["Other"],"enabledApiProposals":["terminalQuickFixProvider"],"main":"./dist/npmMain","browser":"./dist/browser/npmBrowserMain","activationEvents":["onTaskType:npm","onLanguage:json","workspaceContains:package.json"],"capabilities":{"virtualWorkspaces":{"supported":"limited","description":"Functionality that requires running the 'npm' command is not available in virtual workspaces."},"untrustedWorkspaces":{"supported":"limited","description":"This extension executes tasks, which require trust to run."}},"contributes":{"languages":[{"id":"ignore","extensions":[".npmignore"]},{"id":"properties","extensions":[".npmrc"]}],"views":{"explorer":[{"id":"npm","name":"NPM Scripts","when":"npm:showScriptExplorer","icon":"$(json)","visibility":"hidden","contextualTitle":"NPM Scripts"}]},"commands":[{"command":"npm.runScript","title":"Run","icon":"$(run)"},{"command":"npm.debugScript","title":"Debug","icon":"$(debug)"},{"command":"npm.openScript","title":"Open"},{"command":"npm.runInstall","title":"Run Install"},{"command":"npm.refresh","title":"Refresh","icon":"$(refresh)"},{"command":"npm.runSelectedScript","title":"Run Script"},{"command":"npm.runScriptFromFolder","title":"Run NPM Script in Folder..."},{"command":"npm.packageManager","title":"Get Configured Package Manager"}],"menus":{"commandPalette":[{"command":"npm.refresh","when":"false"},{"command":"npm.runScript","when":"false"},{"command":"npm.debugScript","when":"false"},{"command":"npm.openScript","when":"false"},{"command":"npm.runInstall","when":"false"},{"command":"npm.runSelectedScript","when":"false"},{"command":"npm.runScriptFromFolder","when":"false"},{"command":"npm.packageManager","when":"false"}],"editor/context":[{"command":"npm.runSelectedScript","when":"resourceFilename == 'package.json' && resourceScheme == file","group":"navigation@+1"}],"view/title":[{"command":"npm.refresh","when":"view == npm","group":"navigation"}],"view/item/context":[{"command":"npm.openScript","when":"view == npm && viewItem == packageJSON","group":"navigation@1"},{"command":"npm.runInstall","when":"view == npm && viewItem == packageJSON","group":"navigation@2"},{"command":"npm.openScript","when":"view == npm && viewItem == script","group":"navigation@1"},{"command":"npm.runScript","when":"view == npm && viewItem == script","group":"navigation@2"},{"command":"npm.runScript","when":"view == npm && viewItem == script","group":"inline"},{"command":"npm.debugScript","when":"view == npm && viewItem == script","group":"inline"},{"command":"npm.debugScript","when":"view == npm && viewItem == script","group":"navigation@3"}],"explorer/context":[{"when":"config.npm.enableRunFromFolder && explorerViewletVisible && explorerResourceIsFolder && resourceScheme == file","command":"npm.runScriptFromFolder","group":"2_workspace"}]},"configuration":{"id":"npm","type":"object","title":"Npm","properties":{"npm.autoDetect":{"type":"string","enum":["off","on"],"default":"on","scope":"resource","description":"Controls whether npm scripts should be automatically detected."},"npm.runSilent":{"type":"boolean","default":false,"scope":"resource","markdownDescription":"Run npm commands with the `--silent` option."},"npm.packageManager":{"scope":"resource","type":"string","enum":["auto","npm","yarn","pnpm","bun"],"enumDescriptions":["Auto-detect which package manager to use for running scripts based on lock files and installed package managers.","Use npm as the package manager for running scripts.","Use yarn as the package manager for running scripts.","Use pnpm as the package manager for running scripts.","Use bun as the package manager for running scripts."],"default":"auto","description":"The package manager used to run scripts."},"npm.exclude":{"type":["string","array"],"items":{"type":"string"},"description":"Configure glob patterns for folders that should be excluded from automatic script detection.","scope":"resource"},"npm.enableScriptExplorer":{"type":"boolean","default":false,"scope":"resource","deprecationMessage":"The NPM Script Explorer is now available in 'Views' menu in the Explorer in all folders.","description":"Enable an explorer view for npm scripts when there is no top-level 'package.json' file."},"npm.enableRunFromFolder":{"type":"boolean","default":false,"scope":"resource","description":"Enable running npm scripts contained in a folder from the Explorer context menu."},"npm.scriptExplorerAction":{"type":"string","enum":["open","run"],"markdownDescription":"The default click action used in the NPM Scripts Explorer: `open` or `run`, the default is `open`.","scope":"window","default":"open"},"npm.scriptExplorerExclude":{"type":"array","items":{"type":"string"},"markdownDescription":"An array of regular expressions that indicate which scripts should be excluded from the NPM Scripts view.","scope":"resource","default":[]},"npm.fetchOnlinePackageInfo":{"type":"boolean","description":"Fetch data from https://registry.npmjs.org and https://registry.bower.io to provide auto-completion and information on hover features on npm dependencies.","default":true,"scope":"window","tags":["usesOnlineServices"]},"npm.scriptHover":{"type":"boolean","description":"Display hover with 'Run' and 'Debug' commands for scripts.","default":true,"scope":"window"}}},"jsonValidation":[{"fileMatch":"package.json","url":"https://json.schemastore.org/package"},{"fileMatch":"bower.json","url":"https://json.schemastore.org/bower"}],"taskDefinitions":[{"type":"npm","required":["script"],"properties":{"script":{"type":"string","description":"The npm script to customize."},"path":{"type":"string","description":"The path to the folder of the package.json file that provides the script. Can be omitted."}},"when":"shellExecutionSupported"}],"terminalQuickFixes":[{"id":"ms-vscode.npm-command","commandLineMatcher":"npm","commandExitResult":"error","outputMatcher":{"anchor":"bottom","length":8,"lineMatcher":"Did you mean (?:this|one of these)\\?((?:\\n.+?npm .+ #.+)+)","offset":2}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/npm","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.objective-c"},"manifest":{"name":"objective-c","displayName":"Objective-C Language Basics","description":"Provides syntax highlighting and bracket matching in Objective-C files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammars.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"objective-c","extensions":[".m"],"aliases":["Objective-C"],"configuration":"./language-configuration.json"},{"id":"objective-cpp","extensions":[".mm"],"aliases":["Objective-C++"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"objective-c","scopeName":"source.objc","path":"./syntaxes/objective-c.tmLanguage.json"},{"language":"objective-cpp","scopeName":"source.objcpp","path":"./syntaxes/objective-c++.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/objective-c","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.perl"},"manifest":{"name":"perl","displayName":"Perl Language Basics","description":"Provides syntax highlighting and bracket matching in Perl files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin textmate/perl.tmbundle Syntaxes/Perl.plist ./syntaxes/perl.tmLanguage.json Syntaxes/Perl%206.tmLanguage ./syntaxes/perl6.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"perl","aliases":["Perl","perl"],"extensions":[".pl",".pm",".pod",".t",".PL",".psgi"],"firstLine":"^#!.*\\bperl\\b","configuration":"./perl.language-configuration.json"},{"id":"raku","aliases":["Raku","Perl6","perl6"],"extensions":[".raku",".rakumod",".rakutest",".rakudoc",".nqp",".p6",".pl6",".pm6"],"firstLine":"(^#!.*\\bperl6\\b)|use\\s+v6|raku|=begin\\spod|my\\sclass","configuration":"./perl6.language-configuration.json"}],"grammars":[{"language":"perl","scopeName":"source.perl","path":"./syntaxes/perl.tmLanguage.json","unbalancedBracketScopes":["variable.other.predefined.perl"]},{"language":"raku","scopeName":"source.perl.6","path":"./syntaxes/perl6.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/perl","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.php"},"manifest":{"name":"php","displayName":"PHP Language Basics","description":"Provides syntax highlighting and bracket matching for PHP files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"php","extensions":[".php",".php4",".php5",".phtml",".ctp"],"aliases":["PHP","php"],"firstLine":"^#!\\s*/.*\\bphp\\b","mimetypes":["application/x-php"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"php","scopeName":"source.php","path":"./syntaxes/php.tmLanguage.json"},{"language":"php","scopeName":"text.html.php","path":"./syntaxes/html.tmLanguage.json","embeddedLanguages":{"text.html":"html","source.php":"php","source.sql":"sql","text.xml":"xml","source.js":"javascript","source.json":"json","source.css":"css"}}],"snippets":[{"language":"php","path":"./snippets/php.code-snippets"}]},"scripts":{"update-grammar":"node ./build/update-grammar.mjs"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/php","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.php-language-features"},"manifest":{"name":"php-language-features","displayName":"PHP Language Features","description":"Provides rich language support for PHP files.","version":"1.0.0","publisher":"vscode","license":"MIT","icon":"icons/logo.png","engines":{"vscode":"0.10.x"},"activationEvents":["onLanguage:php"],"main":"./dist/phpMain","categories":["Programming Languages"],"capabilities":{"virtualWorkspaces":false,"untrustedWorkspaces":{"supported":"limited","description":"The extension requires workspace trust when the `php.validate.executablePath` setting will load a version of PHP in the workspace.","restrictedConfigurations":["php.validate.executablePath"]}},"contributes":{"configuration":{"title":"PHP","type":"object","order":20,"properties":{"php.suggest.basic":{"type":"boolean","default":true,"description":"Controls whether the built-in PHP language suggestions are enabled. The support suggests PHP globals and variables."},"php.validate.enable":{"type":"boolean","default":true,"description":"Enable/disable built-in PHP validation."},"php.validate.executablePath":{"type":["string","null"],"default":null,"description":"Points to the PHP executable.","scope":"machine-overridable"},"php.validate.run":{"type":"string","enum":["onSave","onType"],"default":"onSave","description":"Whether the linter is run on save or on type."}}},"jsonValidation":[{"fileMatch":"composer.json","url":"https://getcomposer.org/schema.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/php-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.powershell"},"manifest":{"name":"powershell","displayName":"Powershell Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in Powershell files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"powershell","extensions":[".ps1",".psm1",".psd1",".pssc",".psrc"],"aliases":["PowerShell","powershell","ps","ps1","pwsh"],"firstLine":"^#!\\s*/.*\\bpwsh\\b","configuration":"./language-configuration.json"}],"grammars":[{"language":"powershell","scopeName":"source.powershell","path":"./syntaxes/powershell.tmLanguage.json"}]},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin PowerShell/EditorSyntax PowerShellSyntax.tmLanguage ./syntaxes/powershell.tmLanguage.json"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/powershell","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.pug"},"manifest":{"name":"pug","displayName":"Pug Language Basics","description":"Provides syntax highlighting and bracket matching in Pug files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin davidrios/pug-tmbundle Syntaxes/Pug.JSON-tmLanguage ./syntaxes/pug.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"jade","extensions":[".pug",".jade"],"aliases":["Pug","Jade","jade"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"jade","scopeName":"text.pug","path":"./syntaxes/pug.tmLanguage.json"}],"configurationDefaults":{"[jade]":{"diffEditor.ignoreTrimWhitespace":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/pug","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.python"},"manifest":{"name":"python","displayName":"Python Language Basics","description":"Provides syntax highlighting, bracket matching and folding in Python files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"python","extensions":[".py",".rpy",".pyw",".cpy",".gyp",".gypi",".pyi",".ipy",".pyt"],"aliases":["Python","py"],"filenames":["SConstruct","SConscript"],"firstLine":"^#!\\s*/?.*\\bpython[0-9.-]*\\b","configuration":"./language-configuration.json"}],"grammars":[{"language":"python","scopeName":"source.python","path":"./syntaxes/MagicPython.tmLanguage.json"},{"scopeName":"source.regexp.python","path":"./syntaxes/MagicRegExp.tmLanguage.json"}],"configurationDefaults":{"[python]":{"diffEditor.ignoreTrimWhitespace":false,"editor.defaultColorDecorators":false}}},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin MagicStack/MagicPython grammars/MagicPython.tmLanguage ./syntaxes/MagicPython.tmLanguage.json grammars/MagicRegExp.tmLanguage ./syntaxes/MagicRegExp.tmLanguage.json"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/python","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.r"},"manifest":{"name":"r","displayName":"R Language Basics","description":"Provides syntax highlighting and bracket matching in R files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin REditorSupport/vscode-R syntax/r.json ./syntaxes/r.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"r","extensions":[".r",".rhistory",".rprofile",".rt"],"aliases":["R","r"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"r","scopeName":"source.r","path":"./syntaxes/r.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/r","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.razor"},"manifest":{"name":"razor","displayName":"Razor Language Basics","description":"Provides syntax highlighting, bracket matching and folding in Razor files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"0.10.x"},"scripts":{"update-grammar":"node ./build/update-grammar.mjs"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"razor","extensions":[".cshtml",".razor"],"aliases":["Razor","razor"],"mimetypes":["text/x-cshtml"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"razor","scopeName":"text.html.cshtml","path":"./syntaxes/cshtml.tmLanguage.json","embeddedLanguages":{"section.embedded.source.cshtml":"csharp","source.css":"css","source.js":"javascript"}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/razor","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"redhat.vscode-yaml","uuid":"2061917f-f76a-458a-8da9-f162de22b97e"},"manifest":{"name":"vscode-yaml","displayName":"YAML","description":"YAML Language Support by Red Hat, with built-in Kubernetes syntax support","author":"Red Hat","contributors":[{"name":"Joshua Pinkney","email":"jpinkney@redhat.com"},{"name":"Yevhen Vydolob","email":"yvydolob@redhat.com"}],"license":"MIT","version":"1.14.0","publisher":"redhat","bugs":"https://github.com/redhat-developer/vscode-yaml/issues","repository":{"type":"git","url":"https://github.com/redhat-developer/vscode-yaml"},"icon":"icon/icon128.png","engines":{"npm":">=7.0.0","vscode":"^1.63.0"},"categories":["Programming Languages","Linters","Snippets","Formatters"],"capabilities":{"untrustedWorkspaces":{"supported":true},"virtualWorkspaces":true},"activationEvents":["onLanguage:yaml","onLanguage:dockercompose"],"keywords":["kubernetes","yaml","autocompletion","validation"],"main":"./dist/extension","browser":"./dist/extension-web","contributes":{"languages":[{"id":"yaml","aliases":["YAML","yaml"],"extensions":[".yml",".eyaml",".eyml",".yaml"],"filenames":[".clang-format","_clang-format"],"firstLine":"^#cloud-config","configuration":"./language-configuration.json"}],"grammars":[{"language":"yaml","scopeName":"source.yaml","path":"./syntaxes/yaml.tmLanguage.json"}],"configuration":{"title:":"YAML","properties":{"redhat.telemetry.enabled":{"type":"boolean","default":null,"markdownDescription":"Enable usage data and errors to be sent to Red Hat servers. Read our [privacy statement](https://developers.redhat.com/article/tool-data-collection).","scope":"window","tags":["telemetry","usesOnlineServices"]},"yaml.yamlVersion":{"type":"string","default":"1.2","enum":["1.1","1.2"],"markdownDescription":"Default YAML spec version"},"yaml.trace.server":{"type":"string","enum":["off","messages","verbose"],"default":"off","description":"Traces the communication between VSCode and the YAML language service."},"yaml.schemas":{"type":"object","default":{},"description":"Associate schemas to YAML files in the current workspace"},"yaml.format.enable":{"type":"boolean","default":true,"description":"Enable/disable default YAML formatter"},"yaml.format.singleQuote":{"type":"boolean","default":false,"description":"Use single quotes instead of double quotes"},"yaml.format.bracketSpacing":{"type":"boolean","default":true,"description":"Print spaces between brackets in objects"},"yaml.format.proseWrap":{"type":"string","default":"preserve","enum":["preserve","never","always"],"description":"Always: wrap prose if it exceeds the print width, Never: never wrap the prose, Preserve: wrap prose as-is"},"yaml.format.printWidth":{"type":"integer","default":80,"description":"Specify the line length that the printer will wrap on"},"yaml.validate":{"type":"boolean","default":true,"description":"Enable/disable validation feature"},"yaml.hover":{"type":"boolean","default":true,"description":"Enable/disable hover feature"},"yaml.completion":{"type":"boolean","default":true,"description":"Enable/disable completion feature"},"yaml.customTags":{"type":"array","default":[],"description":"Custom tags for the parser to use"},"yaml.schemaStore.enable":{"type":"boolean","default":true,"description":"Automatically pull available YAML schemas from JSON Schema Store"},"yaml.schemaStore.url":{"type":"string","default":"https://www.schemastore.org/api/json/catalog.json","description":"URL of schema store catalog to use"},"yaml.disableAdditionalProperties":{"type":"boolean","default":false,"description":"Globally set additionalProperties to false for all objects. So if its true, no extra properties are allowed inside yaml."},"yaml.disableDefaultProperties":{"type":"boolean","default":false,"description":"Disable adding not required properties with default values into completion text."},"yaml.maxItemsComputed":{"type":"integer","default":5000,"description":"The maximum number of outline symbols and folding regions computed (limited for performance reasons)."},"yaml.suggest.parentSkeletonSelectedFirst":{"type":"boolean","default":false,"description":"If true, the user must select some parent skeleton first before autocompletion starts to suggest the rest of the properties. When yaml object is not empty, autocompletion ignores this setting and returns all properties and skeletons"},"yaml.style.flowMapping":{"type":"string","enum":["allow","forbid"],"default":"allow","description":"Forbid flow style mappings"},"yaml.style.flowSequence":{"type":"string","enum":["allow","forbid"],"default":"allow","description":"Forbid flow style sequences"},"yaml.keyOrdering":{"type":"boolean","default":false,"description":"Enforces alphabetical ordering of keys in mappings when set to true"},"yaml.extension.recommendations":{"type":"boolean","default":"true","description":"Suggest additional extensions based on YAML usage."}}},"configurationDefaults":{"[yaml]":{"editor.insertSpaces":true,"editor.tabSize":2,"editor.quickSuggestions":{"other":true,"comments":false,"strings":true},"editor.autoIndent":"keep"}}},"extensionDependencies":[]},"location":{"$mid":1,"path":"/ide/extensions/redhat.vscode-yaml","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","publisherDisplayName":"Red Hat","metadata":{"id":"2061917f-f76a-458a-8da9-f162de22b97e","publisherId":{"publisherId":"eed56242-9699-4317-8bc7-e9f4b9bdd3ff","publisherName":"redhat","displayName":"Red Hat","flags":"verified"},"publisherDisplayName":"Red Hat"},"isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.references-view"},"manifest":{"name":"references-view","displayName":"Reference Search View","description":"Reference Search results as separate, stable view in the sidebar","icon":"media/icon.png","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"^1.67.0"},"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"repository":{"type":"git","url":"https://github.com/Microsoft/vscode-references-view"},"bugs":{"url":"https://github.com/Microsoft/vscode-references-view/issues"},"activationEvents":["onCommand:references-view.find","onCommand:editor.action.showReferences"],"main":"./dist/extension","browser":"./dist/extension.js","contributes":{"configuration":{"properties":{"references.preferredLocation":{"description":"Controls whether 'Peek References' or 'Find References' is invoked when selecting CodeLens references.","type":"string","default":"peek","enum":["peek","view"],"enumDescriptions":["Show references in peek editor.","Show references in separate view."]}}},"viewsContainers":{"activitybar":[{"id":"references-view","icon":"$(references)","title":"References"}]},"views":{"references-view":[{"id":"references-view.tree","name":"Reference Search Results","when":"reference-list.isActive"}]},"commands":[{"command":"references-view.findReferences","title":"Find All References","category":"References"},{"command":"references-view.findImplementations","title":"Find All Implementations","category":"References"},{"command":"references-view.clearHistory","title":"Clear History","category":"References","icon":"$(clear-all)"},{"command":"references-view.clear","title":"Clear","category":"References","icon":"$(clear-all)"},{"command":"references-view.refresh","title":"Refresh","category":"References","icon":"$(refresh)"},{"command":"references-view.pickFromHistory","title":"Show History","category":"References"},{"command":"references-view.removeReferenceItem","title":"Dismiss","icon":"$(close)"},{"command":"references-view.copy","title":"Copy"},{"command":"references-view.copyAll","title":"Copy All"},{"command":"references-view.copyPath","title":"Copy Path"},{"command":"references-view.refind","title":"Rerun","icon":"$(refresh)"},{"command":"references-view.showCallHierarchy","title":"Show Call Hierarchy","category":"Calls"},{"command":"references-view.showOutgoingCalls","title":"Show Outgoing Calls","category":"Calls","icon":"$(call-outgoing)"},{"command":"references-view.showIncomingCalls","title":"Show Incoming Calls","category":"Calls","icon":"$(call-incoming)"},{"command":"references-view.removeCallItem","title":"Dismiss","icon":"$(close)"},{"command":"references-view.next","title":"Go to Next Reference","enablement":"references-view.canNavigate"},{"command":"references-view.prev","title":"Go to Previous Reference","enablement":"references-view.canNavigate"},{"command":"references-view.showTypeHierarchy","title":"Show Type Hierarchy","category":"Types"},{"command":"references-view.showSupertypes","title":"Show Supertypes","category":"Types","icon":"$(type-hierarchy-super)"},{"command":"references-view.showSubtypes","title":"Show Subtypes","category":"Types","icon":"$(type-hierarchy-sub)"},{"command":"references-view.removeTypeItem","title":"Dismiss","icon":"$(close)"}],"menus":{"editor/context":[{"command":"references-view.findReferences","when":"editorHasReferenceProvider","group":"0_navigation@1"},{"command":"references-view.findImplementations","when":"editorHasImplementationProvider","group":"0_navigation@2"},{"command":"references-view.showCallHierarchy","when":"editorHasCallHierarchyProvider","group":"0_navigation@3"},{"command":"references-view.showTypeHierarchy","when":"editorHasTypeHierarchyProvider","group":"0_navigation@4"}],"view/title":[{"command":"references-view.clear","group":"navigation@3","when":"view == references-view.tree && reference-list.hasResult"},{"command":"references-view.clearHistory","group":"navigation@3","when":"view == references-view.tree && reference-list.hasHistory && !reference-list.hasResult"},{"command":"references-view.refresh","group":"navigation@2","when":"view == references-view.tree && reference-list.hasResult"},{"command":"references-view.showOutgoingCalls","group":"navigation@1","when":"view == references-view.tree && reference-list.hasResult && reference-list.source == callHierarchy &&  references-view.callHierarchyMode == showIncoming"},{"command":"references-view.showIncomingCalls","group":"navigation@1","when":"view == references-view.tree && reference-list.hasResult && reference-list.source == callHierarchy &&  references-view.callHierarchyMode == showOutgoing"},{"command":"references-view.showSupertypes","group":"navigation@1","when":"view == references-view.tree && reference-list.hasResult && reference-list.source == typeHierarchy &&  references-view.typeHierarchyMode != supertypes"},{"command":"references-view.showSubtypes","group":"navigation@1","when":"view == references-view.tree && reference-list.hasResult && reference-list.source == typeHierarchy &&  references-view.typeHierarchyMode != subtypes"}],"view/item/context":[{"command":"references-view.removeReferenceItem","group":"inline","when":"view == references-view.tree && viewItem == file-item || view == references-view.tree && viewItem == reference-item"},{"command":"references-view.removeCallItem","group":"inline","when":"view == references-view.tree && viewItem == call-item"},{"command":"references-view.removeTypeItem","group":"inline","when":"view == references-view.tree && viewItem == type-item"},{"command":"references-view.refind","group":"inline","when":"view == references-view.tree && viewItem == history-item"},{"command":"references-view.removeReferenceItem","group":"1","when":"view == references-view.tree && viewItem == file-item || view == references-view.tree && viewItem == reference-item"},{"command":"references-view.removeCallItem","group":"1","when":"view == references-view.tree && viewItem == call-item"},{"command":"references-view.removeTypeItem","group":"1","when":"view == references-view.tree && viewItem == type-item"},{"command":"references-view.refind","group":"1","when":"view == references-view.tree && viewItem == history-item"},{"command":"references-view.copy","group":"2@1","when":"view == references-view.tree && viewItem == file-item || view == references-view.tree && viewItem == reference-item"},{"command":"references-view.copyPath","group":"2@2","when":"view == references-view.tree && viewItem == file-item"},{"command":"references-view.copyAll","group":"2@3","when":"view == references-view.tree && viewItem == file-item || view == references-view.tree && viewItem == reference-item"},{"command":"references-view.showOutgoingCalls","group":"1","when":"view == references-view.tree && viewItem == call-item"},{"command":"references-view.showIncomingCalls","group":"1","when":"view == references-view.tree && viewItem == call-item"},{"command":"references-view.showSupertypes","group":"1","when":"view == references-view.tree && viewItem == type-item"},{"command":"references-view.showSubtypes","group":"1","when":"view == references-view.tree && viewItem == type-item"}],"commandPalette":[{"command":"references-view.removeReferenceItem","when":"never"},{"command":"references-view.removeCallItem","when":"never"},{"command":"references-view.removeTypeItem","when":"never"},{"command":"references-view.copy","when":"never"},{"command":"references-view.copyAll","when":"never"},{"command":"references-view.copyPath","when":"never"},{"command":"references-view.refind","when":"never"},{"command":"references-view.findReferences","when":"editorHasReferenceProvider"},{"command":"references-view.clear","when":"reference-list.hasResult"},{"command":"references-view.clearHistory","when":"reference-list.isActive && !reference-list.hasResult"},{"command":"references-view.refresh","when":"reference-list.hasResult"},{"command":"references-view.pickFromHistory","when":"reference-list.isActive"},{"command":"references-view.next","when":"never"},{"command":"references-view.prev","when":"never"}]},"keybindings":[{"command":"references-view.findReferences","when":"editorHasReferenceProvider","key":"shift+alt+f12"},{"command":"references-view.next","when":"reference-list.hasResult","key":"f4"},{"command":"references-view.prev","when":"reference-list.hasResult","key":"shift+f4"},{"command":"references-view.showCallHierarchy","when":"editorHasCallHierarchyProvider","key":"shift+alt+h"}]}},"location":{"$mid":1,"path":"/ide/extensions/references-view","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.restructuredtext"},"manifest":{"name":"restructuredtext","displayName":"reStructuredText Language Basics","description":"Provides syntax highlighting in reStructuredText files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin trond-snekvik/vscode-rst syntaxes/rst.tmLanguage.json ./syntaxes/rst.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"restructuredtext","aliases":["reStructuredText"],"configuration":"./language-configuration.json","extensions":[".rst"]}],"grammars":[{"language":"restructuredtext","scopeName":"source.rst","path":"./syntaxes/rst.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/restructuredtext","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.ruby"},"manifest":{"name":"ruby","displayName":"Ruby Language Basics","description":"Provides syntax highlighting and bracket matching in Ruby files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin Shopify/ruby-lsp vscode/grammars/ruby.cson.json ./syntaxes/ruby.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"ruby","extensions":[".rb",".rbx",".rjs",".gemspec",".rake",".ru",".erb",".podspec",".rbi"],"filenames":["rakefile","gemfile","guardfile","podfile","capfile","cheffile","hobofile","vagrantfile","appraisals","rantfile","berksfile","berksfile.lock","thorfile","puppetfile","dangerfile","brewfile","fastfile","appfile","deliverfile","matchfile","scanfile","snapfile","gymfile"],"aliases":["Ruby","rb"],"firstLine":"^#!\\s*/.*\\bruby\\b","configuration":"./language-configuration.json"}],"grammars":[{"language":"ruby","scopeName":"source.ruby","path":"./syntaxes/ruby.tmLanguage.json"}],"configurationDefaults":{"[ruby]":{"editor.defaultColorDecorators":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/ruby","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.rust"},"manifest":{"name":"rust","displayName":"Rust Language Basics","description":"Provides syntax highlighting and bracket matching in Rust files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammar.mjs"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"rust","extensions":[".rs"],"aliases":["Rust","rust"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"rust","path":"./syntaxes/rust.tmLanguage.json","scopeName":"source.rust"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/rust","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.scss"},"manifest":{"name":"scss","displayName":"SCSS Language Basics","description":"Provides syntax highlighting, bracket matching and folding in SCSS files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin atom/language-sass grammars/scss.cson ./syntaxes/scss.tmLanguage.json grammars/sassdoc.cson ./syntaxes/sassdoc.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"scss","aliases":["SCSS","scss"],"extensions":[".scss"],"mimetypes":["text/x-scss","text/scss"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"scss","scopeName":"source.css.scss","path":"./syntaxes/scss.tmLanguage.json"},{"scopeName":"source.sassdoc","path":"./syntaxes/sassdoc.tmLanguage.json"}],"problemMatchers":[{"name":"node-sass","label":"Node Sass Compiler","owner":"node-sass","fileLocation":"absolute","pattern":[{"regexp":"^{$"},{"regexp":"\\s*\"status\":\\s\\d+,"},{"regexp":"\\s*\"file\":\\s\"(.*)\",","file":1},{"regexp":"\\s*\"line\":\\s(\\d+),","line":1},{"regexp":"\\s*\"column\":\\s(\\d+),","column":1},{"regexp":"\\s*\"message\":\\s\"(.*)\",","message":1},{"regexp":"\\s*\"formatted\":\\s(.*)"},{"regexp":"^}$"}]}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/scss","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.search-result"},"manifest":{"name":"search-result","displayName":"Search Result","description":"Provides syntax highlighting and language features for tabbed search results.","version":"1.0.0","publisher":"vscode","license":"MIT","icon":"images/icon.png","engines":{"vscode":"^1.39.0"},"main":"./dist/extension.js","browser":"./dist/extension.js","activationEvents":["onLanguage:search-result"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"enabledApiProposals":["documentFiltersExclusive"],"contributes":{"configurationDefaults":{"[search-result]":{"editor.lineNumbers":"off"}},"languages":[{"id":"search-result","extensions":[".code-search"],"aliases":["Search Result"]}],"grammars":[{"language":"search-result","scopeName":"text.searchResult","path":"./syntaxes/searchResult.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/search-result","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.shaderlab"},"manifest":{"name":"shaderlab","displayName":"Shaderlab Language Basics","description":"Provides syntax highlighting and bracket matching in Shaderlab files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin tgjones/shaders-tmLanguage grammars/shaderlab.json ./syntaxes/shaderlab.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"shaderlab","extensions":[".shader"],"aliases":["ShaderLab","shaderlab"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"shaderlab","path":"./syntaxes/shaderlab.tmLanguage.json","scopeName":"source.shaderlab"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/shaderlab","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.shellscript"},"manifest":{"name":"shellscript","displayName":"Shell Script Language Basics","description":"Provides syntax highlighting and bracket matching in Shell Script files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin jeff-hykin/better-shell-syntax autogenerated/shell.tmLanguage.json ./syntaxes/shell-unix-bash.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"shellscript","aliases":["Shell Script","shellscript","bash","fish","sh","zsh","ksh","csh"],"extensions":[".sh",".bash",".bashrc",".bash_aliases",".bash_profile",".bash_login",".ebuild",".eclass",".profile",".bash_logout",".xprofile",".xsession",".xsessionrc",".Xsession",".zsh",".zshrc",".zprofile",".zlogin",".zlogout",".zshenv",".zsh-theme",".fish",".ksh",".csh",".cshrc",".tcshrc",".yashrc",".yash_profile"],"filenames":["APKBUILD","PKGBUILD",".envrc",".hushlogin","zshrc","zshenv","zlogin","zprofile","zlogout","bashrc_Apple_Terminal","zshrc_Apple_Terminal"],"filenamePatterns":[".env.*"],"firstLine":"^#!.*\\b(bash|fish|zsh|sh|ksh|dtksh|pdksh|mksh|ash|dash|yash|sh|csh|jcsh|tcsh|itcsh).*|^#\\s*-\\*-[^*]*mode:\\s*shell-script[^*]*-\\*-","configuration":"./language-configuration.json","mimetypes":["text/x-shellscript"]}],"grammars":[{"language":"shellscript","scopeName":"source.shell","path":"./syntaxes/shell-unix-bash.tmLanguage.json","balancedBracketScopes":["*"],"unbalancedBracketScopes":["meta.scope.case-pattern.shell"]}],"configurationDefaults":{"[shellscript]":{"files.eol":"\n","editor.defaultColorDecorators":false}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/shellscript","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.simple-browser"},"manifest":{"name":"simple-browser","displayName":"Simple Browser","description":"A very basic built-in webview for displaying web content.","enabledApiProposals":["externalUriOpener"],"version":"1.0.0","icon":"media/icon.png","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","engines":{"vscode":"^1.70.0"},"main":"./dist/extension","browser":"./dist/browser/extension","categories":["Other"],"extensionKind":["ui","workspace"],"activationEvents":["onCommand:simpleBrowser.api.open","onOpenExternalUri:http","onOpenExternalUri:https","onWebviewPanel:simpleBrowser.view"],"capabilities":{"virtualWorkspaces":true,"untrustedWorkspaces":{"supported":true}},"contributes":{"commands":[{"command":"simpleBrowser.show","title":"Show","category":"Simple Browser"}],"configuration":[{"title":"Simple Browser","properties":{"simpleBrowser.focusLockIndicator.enabled":{"type":"boolean","default":true,"title":"Focus Lock Indicator Enabled","description":"Enable/disable the floating indicator that shows when focused in the simple browser."}}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/simple-browser","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.sql"},"manifest":{"name":"sql","displayName":"SQL Language Basics","description":"Provides syntax highlighting and bracket matching in SQL files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammar.mjs"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"sql","extensions":[".sql",".dsql"],"aliases":["MS SQL","T-SQL"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"sql","scopeName":"source.sql","path":"./syntaxes/sql.tmLanguage.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/sql","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.swift"},"manifest":{"name":"swift","displayName":"Swift Language Basics","description":"Provides snippets, syntax highlighting and bracket matching in Swift files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin jtbandes/swift-tmlanguage Swift.tmLanguage.json ./syntaxes/swift.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"swift","aliases":["Swift","swift"],"extensions":[".swift"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"swift","scopeName":"source.swift","path":"./syntaxes/swift.tmLanguage.json"}],"snippets":[{"language":"swift","path":"./snippets/swift.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/swift","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.terminal-suggest"},"manifest":{"name":"terminal-suggest","publisher":"vscode","displayName":"Terminal Suggest for VS Code","description":"Extension to add terminal completions for zsh, bash, and fish terminals.","version":"1.0.1","private":true,"license":"MIT","icon":"./src/media/icon.png","engines":{"vscode":"^1.95.0"},"categories":["Other"],"enabledApiProposals":["terminalCompletionProvider","terminalShellEnv","terminalShellType"],"main":"./dist/terminalSuggestMain","activationEvents":["onTerminalCompletionsRequested"],"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/terminal-suggest","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-abyss"},"manifest":{"name":"theme-abyss","displayName":"Abyss Theme","description":"Abyss theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Abyss","label":"Abyss","uiTheme":"vs-dark","path":"./themes/abyss-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-abyss","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-defaults"},"manifest":{"name":"theme-defaults","displayName":"Default Themes","description":"The default Visual Studio light and dark themes","categories":["Themes"],"version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"contributes":{"themes":[{"id":"Default Dark+","label":"Dark+","uiTheme":"vs-dark","path":"./themes/dark_plus.json"},{"id":"Default Dark Modern","label":"Dark Modern","uiTheme":"vs-dark","path":"./themes/dark_modern.json"},{"id":"Default Light+","label":"Light+","uiTheme":"vs","path":"./themes/light_plus.json"},{"id":"Default Light Modern","label":"Light Modern","uiTheme":"vs","path":"./themes/light_modern.json"},{"id":"Visual Studio Dark","label":"Dark (Visual Studio)","uiTheme":"vs-dark","path":"./themes/dark_vs.json"},{"id":"Visual Studio Light","label":"Light (Visual Studio)","uiTheme":"vs","path":"./themes/light_vs.json"},{"id":"Default High Contrast","label":"Dark High Contrast","uiTheme":"hc-black","path":"./themes/hc_black.json"},{"id":"Default High Contrast Light","label":"Light High Contrast","uiTheme":"hc-light","path":"./themes/hc_light.json"}],"iconThemes":[{"id":"vs-minimal","label":"Minimal (Visual Studio Code)","path":"./fileicons/vs_minimal-icon-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-defaults","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-kimbie-dark"},"manifest":{"name":"theme-kimbie-dark","displayName":"Kimbie Dark Theme","description":"Kimbie dark theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Kimbie Dark","label":"Kimbie Dark","uiTheme":"vs-dark","path":"./themes/kimbie-dark-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-kimbie-dark","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-monokai"},"manifest":{"name":"theme-monokai","displayName":"Monokai Theme","description":"Monokai theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Monokai","label":"Monokai","uiTheme":"vs-dark","path":"./themes/monokai-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-monokai","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-monokai-dimmed"},"manifest":{"name":"theme-monokai-dimmed","displayName":"Monokai Dimmed Theme","description":"Monokai dimmed theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Monokai Dimmed","label":"Monokai Dimmed","uiTheme":"vs-dark","path":"./themes/dimmed-monokai-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-monokai-dimmed","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-quietlight"},"manifest":{"name":"theme-quietlight","displayName":"Quiet Light Theme","description":"Quiet light theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Quiet Light","label":"Quiet Light","uiTheme":"vs","path":"./themes/quietlight-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-quietlight","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-red"},"manifest":{"name":"theme-red","displayName":"Red Theme","description":"Red theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Red","label":"Red","uiTheme":"vs-dark","path":"./themes/Red-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-red","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.vscode-theme-seti"},"manifest":{"name":"vscode-theme-seti","private":true,"version":"1.0.0","displayName":"Seti File Icon Theme","description":"A file icon theme made out of the Seti UI file icons","publisher":"vscode","license":"MIT","icon":"icons/seti-circular-128x128.png","scripts":{"update":"node ./build/update-icon-theme.js"},"engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"iconThemes":[{"id":"vs-seti","label":"Seti (Visual Studio Code)","path":"./icons/vs-seti-icon-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-seti","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-solarized-dark"},"manifest":{"name":"theme-solarized-dark","displayName":"Solarized Dark Theme","description":"Solarized dark theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Solarized Dark","label":"Solarized Dark","uiTheme":"vs-dark","path":"./themes/solarized-dark-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-solarized-dark","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-solarized-light"},"manifest":{"name":"theme-solarized-light","displayName":"Solarized Light Theme","description":"Solarized light theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Solarized Light","label":"Solarized Light","uiTheme":"vs","path":"./themes/solarized-light-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-solarized-light","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.theme-tomorrow-night-blue"},"manifest":{"name":"theme-tomorrow-night-blue","displayName":"Tomorrow Night Blue Theme","description":"Tomorrow night blue theme for Visual Studio Code","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Themes"],"contributes":{"themes":[{"id":"Tomorrow Night Blue","label":"Tomorrow Night Blue","uiTheme":"vs-dark","path":"./themes/tomorrow-night-blue-color-theme.json"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/theme-tomorrow-night-blue","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.typescript"},"manifest":{"name":"typescript","description":"Provides snippets, syntax highlighting, bracket matching and folding in TypeScript files.","displayName":"TypeScript Language Basics","version":"1.0.0","author":"vscode","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammars.mjs"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"typescript","aliases":["TypeScript","ts","typescript"],"extensions":[".ts",".cts",".mts"],"configuration":"./language-configuration.json"},{"id":"typescriptreact","aliases":["TypeScript JSX","TypeScript React","tsx"],"extensions":[".tsx"],"configuration":"./language-configuration.json"},{"id":"jsonc","filenames":["tsconfig.json","jsconfig.json"],"filenamePatterns":["tsconfig.*.json","jsconfig.*.json","tsconfig-*.json","jsconfig-*.json"]},{"id":"json","extensions":[".tsbuildinfo"]}],"grammars":[{"language":"typescript","scopeName":"source.ts","path":"./syntaxes/TypeScript.tmLanguage.json","unbalancedBracketScopes":["keyword.operator.relational","storage.type.function.arrow","keyword.operator.bitwise.shift","meta.brace.angle","punctuation.definition.tag","keyword.operator.assignment.compound.bitwise.ts"],"tokenTypes":{"meta.template.expression":"other","meta.template.expression string":"string","meta.template.expression comment":"comment","entity.name.type.instance.jsdoc":"other","entity.name.function.tagged-template":"other","meta.import string.quoted":"other","variable.other.jsdoc":"other"}},{"language":"typescriptreact","scopeName":"source.tsx","path":"./syntaxes/TypeScriptReact.tmLanguage.json","unbalancedBracketScopes":["keyword.operator.relational","storage.type.function.arrow","keyword.operator.bitwise.shift","punctuation.definition.tag","keyword.operator.assignment.compound.bitwise.ts"],"embeddedLanguages":{"meta.tag.tsx":"jsx-tags","meta.tag.without-attributes.tsx":"jsx-tags","meta.tag.attributes.tsx":"typescriptreact","meta.embedded.expression.tsx":"typescriptreact"},"tokenTypes":{"meta.template.expression":"other","meta.template.expression string":"string","meta.template.expression comment":"comment","entity.name.type.instance.jsdoc":"other","entity.name.function.tagged-template":"other","meta.import string.quoted":"other","variable.other.jsdoc":"other"}},{"scopeName":"documentation.injection.ts","path":"./syntaxes/jsdoc.ts.injection.tmLanguage.json","injectTo":["source.ts","source.tsx"]},{"scopeName":"documentation.injection.js.jsx","path":"./syntaxes/jsdoc.js.injection.tmLanguage.json","injectTo":["source.js","source.js.jsx"]}],"semanticTokenScopes":[{"language":"typescript","scopes":{"property":["variable.other.property.ts"],"property.readonly":["variable.other.constant.property.ts"],"variable":["variable.other.readwrite.ts"],"variable.readonly":["variable.other.constant.object.ts"],"function":["entity.name.function.ts"],"namespace":["entity.name.type.module.ts"],"variable.defaultLibrary":["support.variable.ts"],"function.defaultLibrary":["support.function.ts"]}},{"language":"typescriptreact","scopes":{"property":["variable.other.property.tsx"],"property.readonly":["variable.other.constant.property.tsx"],"variable":["variable.other.readwrite.tsx"],"variable.readonly":["variable.other.constant.object.tsx"],"function":["entity.name.function.tsx"],"namespace":["entity.name.type.module.tsx"],"variable.defaultLibrary":["support.variable.tsx"],"function.defaultLibrary":["support.function.tsx"]}}],"snippets":[{"language":"typescript","path":"./snippets/typescript.code-snippets"},{"language":"typescriptreact","path":"./snippets/typescript.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/typescript-basics","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.typescript-language-features"},"manifest":{"name":"typescript-language-features","description":"Provides rich language support for JavaScript and TypeScript.","displayName":"TypeScript and JavaScript Language Features","version":"1.0.0","author":"vscode","publisher":"vscode","license":"MIT","aiKey":"0c6ae279ed8443289764825290e4f9e2-1a736e7c-1324-4338-be46-fc2a58ae4d14-7255","enabledApiProposals":["workspaceTrust","multiDocumentHighlightProvider","codeActionAI","codeActionRanges","editorHoverVerbosityLevel"],"capabilities":{"virtualWorkspaces":{"supported":"limited","description":"In virtual workspaces, resolving and finding references across files is not supported."},"untrustedWorkspaces":{"supported":"limited","description":"The extension requires workspace trust when the workspace version is used because it executes code specified by the workspace.","restrictedConfigurations":["typescript.tsdk","typescript.tsserver.pluginPaths","typescript.npm","typescript.tsserver.nodePath"]}},"engines":{"vscode":"^1.30.0"},"icon":"media/icon.png","categories":["Programming Languages"],"activationEvents":["onLanguage:javascript","onLanguage:javascriptreact","onLanguage:typescript","onLanguage:typescriptreact","onLanguage:jsx-tags","onCommand:typescript.tsserverRequest","onCommand:_typescript.configurePlugin","onCommand:_typescript.learnMoreAboutRefactorings","onCommand:typescript.fileReferences","onTaskType:typescript","onLanguage:jsonc","onWalkthrough:nodejsWelcome"],"main":"./dist/extension","browser":"./dist/browser/extension","contributes":{"jsonValidation":[{"fileMatch":"package.json","url":"./schemas/package.schema.json"},{"fileMatch":"tsconfig.json","url":"https://json.schemastore.org/tsconfig"},{"fileMatch":"tsconfig.json","url":"./schemas/tsconfig.schema.json"},{"fileMatch":"tsconfig.*.json","url":"https://json.schemastore.org/tsconfig"},{"fileMatch":"tsconfig-*.json","url":"./schemas/tsconfig.schema.json"},{"fileMatch":"tsconfig-*.json","url":"https://json.schemastore.org/tsconfig"},{"fileMatch":"tsconfig.*.json","url":"./schemas/tsconfig.schema.json"},{"fileMatch":"typings.json","url":"https://json.schemastore.org/typings"},{"fileMatch":".bowerrc","url":"https://json.schemastore.org/bowerrc"},{"fileMatch":".babelrc","url":"https://json.schemastore.org/babelrc"},{"fileMatch":".babelrc.json","url":"https://json.schemastore.org/babelrc"},{"fileMatch":"babel.config.json","url":"https://json.schemastore.org/babelrc"},{"fileMatch":"jsconfig.json","url":"https://json.schemastore.org/jsconfig"},{"fileMatch":"jsconfig.json","url":"./schemas/jsconfig.schema.json"},{"fileMatch":"jsconfig.*.json","url":"https://json.schemastore.org/jsconfig"},{"fileMatch":"jsconfig.*.json","url":"./schemas/jsconfig.schema.json"},{"fileMatch":".swcrc","url":"https://swc.rs/schema.json"},{"fileMatch":"typedoc.json","url":"https://typedoc.org/schema.json"}],"configuration":{"type":"object","title":"TypeScript","order":20,"properties":{"typescript.tsdk":{"type":"string","markdownDescription":"Specifies the folder path to the tsserver and `lib*.d.ts` files under a TypeScript install to use for IntelliSense, for example: `./node_modules/typescript/lib`.\n\n- When specified as a user setting, the TypeScript version from `typescript.tsdk` automatically replaces the built-in TypeScript version.\n- When specified as a workspace setting, `typescript.tsdk` allows you to switch to use that workspace version of TypeScript for IntelliSense with the `TypeScript: Select TypeScript version` command.\n\nSee the [TypeScript documentation](https://code.visualstudio.com/docs/typescript/typescript-compiling#_using-newer-typescript-versions) for more detail about managing TypeScript versions.","scope":"window"},"typescript.disableAutomaticTypeAcquisition":{"type":"boolean","default":false,"markdownDescription":"Disables [automatic type acquisition](https://code.visualstudio.com/docs/nodejs/working-with-javascript#_typings-and-automatic-type-acquisition). Automatic type acquisition fetches `@types` packages from npm to improve IntelliSense for external libraries.","scope":"window","tags":["usesOnlineServices"]},"typescript.enablePromptUseWorkspaceTsdk":{"type":"boolean","default":false,"description":"Enables prompting of users to use the TypeScript version configured in the workspace for Intellisense.","scope":"window"},"typescript.npm":{"type":"string","markdownDescription":"Specifies the path to the npm executable used for [Automatic Type Acquisition](https://code.visualstudio.com/docs/nodejs/working-with-javascript#_typings-and-automatic-type-acquisition).","scope":"machine"},"typescript.check.npmIsInstalled":{"type":"boolean","default":true,"markdownDescription":"Check if npm is installed for [Automatic Type Acquisition](https://code.visualstudio.com/docs/nodejs/working-with-javascript#_typings-and-automatic-type-acquisition).","scope":"window"},"javascript.referencesCodeLens.enabled":{"type":"boolean","default":false,"description":"Enable/disable references CodeLens in JavaScript files.","scope":"window"},"javascript.referencesCodeLens.showOnAllFunctions":{"type":"boolean","default":false,"description":"Enable/disable references CodeLens on all functions in JavaScript files.","scope":"window"},"typescript.referencesCodeLens.enabled":{"type":"boolean","default":false,"description":"Enable/disable references CodeLens in TypeScript files.","scope":"window"},"typescript.referencesCodeLens.showOnAllFunctions":{"type":"boolean","default":false,"description":"Enable/disable references CodeLens on all functions in TypeScript files.","scope":"window"},"typescript.implementationsCodeLens.enabled":{"type":"boolean","default":false,"description":"Enable/disable implementations CodeLens. This CodeLens shows the implementers of an interface.","scope":"window"},"typescript.implementationsCodeLens.showOnInterfaceMethods":{"type":"boolean","default":false,"description":"Enable/disable implementations CodeLens on interface methods.","scope":"window"},"typescript.tsserver.enableTracing":{"type":"boolean","default":false,"description":"Enables tracing TS server performance to a directory. These trace files can be used to diagnose TS Server performance issues. The log may contain file paths, source code, and other potentially sensitive information from your project.","scope":"window"},"typescript.tsserver.log":{"type":"string","enum":["off","terse","normal","verbose"],"default":"off","description":"Enables logging of the TS server to a file. This log can be used to diagnose TS Server issues. The log may contain file paths, source code, and other potentially sensitive information from your project.","scope":"window"},"typescript.tsserver.pluginPaths":{"type":"array","items":{"type":"string","description":"Either an absolute or relative path. Relative path will be resolved against workspace folder(s)."},"default":[],"description":"Additional paths to discover TypeScript Language Service plugins.","scope":"machine"},"javascript.suggest.completeFunctionCalls":{"type":"boolean","default":false,"description":"Complete functions with their parameter signature.","scope":"resource"},"typescript.suggest.completeFunctionCalls":{"type":"boolean","default":false,"description":"Complete functions with their parameter signature.","scope":"resource"},"javascript.suggest.includeAutomaticOptionalChainCompletions":{"type":"boolean","default":true,"description":"Enable/disable showing completions on potentially undefined values that insert an optional chain call. Requires strict null checks to be enabled.","scope":"resource"},"typescript.suggest.includeAutomaticOptionalChainCompletions":{"type":"boolean","default":true,"description":"Enable/disable showing completions on potentially undefined values that insert an optional chain call. Requires strict null checks to be enabled.","scope":"resource"},"typescript.inlayHints.parameterNames.enabled":{"type":"string","enum":["none","literals","all"],"enumDescriptions":["Disable parameter name hints.","Enable parameter name hints only for literal arguments.","Enable parameter name hints for literal and non-literal arguments."],"default":"none","markdownDescription":"Enable/disable inlay hints for parameter names:\n```typescript\n\nparseInt(/* str: */ '123', /* radix: */ 8)\n \n```","scope":"resource"},"typescript.inlayHints.parameterNames.suppressWhenArgumentMatchesName":{"type":"boolean","default":true,"markdownDescription":"Suppress parameter name hints on arguments whose text is identical to the parameter name.","scope":"resource"},"typescript.inlayHints.parameterTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit parameter types:\n```typescript\n\nel.addEventListener('click', e /* :MouseEvent */ => ...)\n \n```","scope":"resource"},"typescript.inlayHints.variableTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit variable types:\n```typescript\n\nconst foo /* :number */ = Date.now();\n \n```","scope":"resource"},"typescript.inlayHints.variableTypes.suppressWhenTypeMatchesName":{"type":"boolean","default":true,"markdownDescription":"Suppress type hints on variables whose name is identical to the type name.","scope":"resource"},"typescript.inlayHints.propertyDeclarationTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit types on property declarations:\n```typescript\n\nclass Foo {\n\tprop /* :number */ = Date.now();\n}\n \n```","scope":"resource"},"typescript.inlayHints.functionLikeReturnTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit return types on function signatures:\n```typescript\n\nfunction foo() /* :number */ {\n\treturn Date.now();\n} \n \n```","scope":"resource"},"typescript.inlayHints.enumMemberValues.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for member values in enum declarations:\n```typescript\n\nenum MyValue {\n\tA /* = 0 */;\n\tB /* = 1 */;\n}\n \n```","scope":"resource"},"javascript.inlayHints.parameterNames.enabled":{"type":"string","enum":["none","literals","all"],"enumDescriptions":["Disable parameter name hints.","Enable parameter name hints only for literal arguments.","Enable parameter name hints for literal and non-literal arguments."],"default":"none","markdownDescription":"Enable/disable inlay hints for parameter names:\n```typescript\n\nparseInt(/* str: */ '123', /* radix: */ 8)\n \n```","scope":"resource"},"javascript.inlayHints.parameterNames.suppressWhenArgumentMatchesName":{"type":"boolean","default":true,"markdownDescription":"Suppress parameter name hints on arguments whose text is identical to the parameter name.","scope":"resource"},"javascript.inlayHints.parameterTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit parameter types:\n```typescript\n\nel.addEventListener('click', e /* :MouseEvent */ => ...)\n \n```","scope":"resource"},"javascript.inlayHints.variableTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit variable types:\n```typescript\n\nconst foo /* :number */ = Date.now();\n \n```","scope":"resource"},"javascript.inlayHints.variableTypes.suppressWhenTypeMatchesName":{"type":"boolean","default":true,"markdownDescription":"Suppress type hints on variables whose name is identical to the type name.","scope":"resource"},"javascript.inlayHints.propertyDeclarationTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit types on property declarations:\n```typescript\n\nclass Foo {\n\tprop /* :number */ = Date.now();\n}\n \n```","scope":"resource"},"javascript.inlayHints.functionLikeReturnTypes.enabled":{"type":"boolean","default":false,"markdownDescription":"Enable/disable inlay hints for implicit return types on function signatures:\n```typescript\n\nfunction foo() /* :number */ {\n\treturn Date.now();\n} \n \n```","scope":"resource"},"javascript.suggest.includeCompletionsForImportStatements":{"type":"boolean","default":true,"description":"Enable/disable auto-import-style completions on partially-typed import statements.","scope":"resource"},"typescript.suggest.includeCompletionsForImportStatements":{"type":"boolean","default":true,"description":"Enable/disable auto-import-style completions on partially-typed import statements.","scope":"resource"},"typescript.reportStyleChecksAsWarnings":{"type":"boolean","default":true,"description":"Report style checks as warnings.","scope":"window"},"typescript.validate.enable":{"type":"boolean","default":true,"description":"Enable/disable TypeScript validation.","scope":"window"},"typescript.format.enable":{"type":"boolean","default":true,"description":"Enable/disable default TypeScript formatter.","scope":"window"},"typescript.format.insertSpaceAfterCommaDelimiter":{"type":"boolean","default":true,"description":"Defines space handling after a comma delimiter.","scope":"resource"},"typescript.format.insertSpaceAfterConstructor":{"type":"boolean","default":false,"description":"Defines space handling after the constructor keyword.","scope":"resource"},"typescript.format.insertSpaceAfterSemicolonInForStatements":{"type":"boolean","default":true,"description":"Defines space handling after a semicolon in a for statement.","scope":"resource"},"typescript.format.insertSpaceBeforeAndAfterBinaryOperators":{"type":"boolean","default":true,"description":"Defines space handling after a binary operator.","scope":"resource"},"typescript.format.insertSpaceAfterKeywordsInControlFlowStatements":{"type":"boolean","default":true,"description":"Defines space handling after keywords in a control flow statement.","scope":"resource"},"typescript.format.insertSpaceAfterFunctionKeywordForAnonymousFunctions":{"type":"boolean","default":true,"description":"Defines space handling after function keyword for anonymous functions.","scope":"resource"},"typescript.format.insertSpaceBeforeFunctionParenthesis":{"type":"boolean","default":false,"description":"Defines space handling before function argument parentheses.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyParenthesis":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing non-empty parenthesis.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyBrackets":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing non-empty brackets.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyBraces":{"type":"boolean","default":true,"description":"Defines space handling after opening and before closing non-empty braces.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingEmptyBraces":{"type":"boolean","default":true,"description":"Defines space handling after opening and before closing empty braces.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingTemplateStringBraces":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing template string braces.","scope":"resource"},"typescript.format.insertSpaceAfterOpeningAndBeforeClosingJsxExpressionBraces":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing JSX expression braces.","scope":"resource"},"typescript.format.insertSpaceAfterTypeAssertion":{"type":"boolean","default":false,"description":"Defines space handling after type assertions in TypeScript.","scope":"resource"},"typescript.format.placeOpenBraceOnNewLineForFunctions":{"type":"boolean","default":false,"description":"Defines whether an open brace is put onto a new line for functions or not.","scope":"resource"},"typescript.format.placeOpenBraceOnNewLineForControlBlocks":{"type":"boolean","default":false,"description":"Defines whether an open brace is put onto a new line for control blocks or not.","scope":"resource"},"typescript.format.semicolons":{"type":"string","default":"ignore","description":"Defines handling of optional semicolons.","scope":"resource","enum":["ignore","insert","remove"],"enumDescriptions":["Don't insert or remove any semicolons.","Insert semicolons at statement ends.","Remove unnecessary semicolons."]},"typescript.format.indentSwitchCase":{"type":"boolean","default":true,"description":"Indent case clauses in switch statements. Requires using TypeScript 5.1+ in the workspace.","scope":"resource"},"javascript.format.indentSwitchCase":{"type":"boolean","default":true,"description":"Indent case clauses in switch statements. Requires using TypeScript 5.1+ in the workspace.","scope":"resource"},"javascript.validate.enable":{"type":"boolean","default":true,"description":"Enable/disable JavaScript validation.","scope":"window"},"javascript.format.enable":{"type":"boolean","default":true,"description":"Enable/disable default JavaScript formatter.","scope":"window"},"javascript.format.insertSpaceAfterCommaDelimiter":{"type":"boolean","default":true,"description":"Defines space handling after a comma delimiter.","scope":"resource"},"javascript.format.insertSpaceAfterConstructor":{"type":"boolean","default":false,"description":"Defines space handling after the constructor keyword.","scope":"resource"},"javascript.format.insertSpaceAfterSemicolonInForStatements":{"type":"boolean","default":true,"description":"Defines space handling after a semicolon in a for statement.","scope":"resource"},"javascript.format.insertSpaceBeforeAndAfterBinaryOperators":{"type":"boolean","default":true,"description":"Defines space handling after a binary operator.","scope":"resource"},"javascript.format.insertSpaceAfterKeywordsInControlFlowStatements":{"type":"boolean","default":true,"description":"Defines space handling after keywords in a control flow statement.","scope":"resource"},"javascript.format.insertSpaceAfterFunctionKeywordForAnonymousFunctions":{"type":"boolean","default":true,"description":"Defines space handling after function keyword for anonymous functions.","scope":"resource"},"javascript.format.insertSpaceBeforeFunctionParenthesis":{"type":"boolean","default":false,"description":"Defines space handling before function argument parentheses.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyParenthesis":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing non-empty parenthesis.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyBrackets":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing non-empty brackets.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingNonemptyBraces":{"type":"boolean","default":true,"description":"Defines space handling after opening and before closing non-empty braces.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingEmptyBraces":{"type":"boolean","default":true,"description":"Defines space handling after opening and before closing empty braces.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingTemplateStringBraces":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing template string braces.","scope":"resource"},"javascript.format.insertSpaceAfterOpeningAndBeforeClosingJsxExpressionBraces":{"type":"boolean","default":false,"description":"Defines space handling after opening and before closing JSX expression braces.","scope":"resource"},"javascript.format.placeOpenBraceOnNewLineForFunctions":{"type":"boolean","default":false,"description":"Defines whether an open brace is put onto a new line for functions or not.","scope":"resource"},"javascript.format.placeOpenBraceOnNewLineForControlBlocks":{"type":"boolean","default":false,"description":"Defines whether an open brace is put onto a new line for control blocks or not.","scope":"resource"},"javascript.format.semicolons":{"type":"string","default":"ignore","description":"Defines handling of optional semicolons.","scope":"resource","enum":["ignore","insert","remove"],"enumDescriptions":["Don't insert or remove any semicolons.","Insert semicolons at statement ends.","Remove unnecessary semicolons."]},"js/ts.implicitProjectConfig.module":{"type":"string","markdownDescription":"Sets the module system for the program. See more: https://www.typescriptlang.org/tsconfig#module.","default":"ESNext","enum":["CommonJS","AMD","System","UMD","ES6","ES2015","ES2020","ESNext","None","ES2022","Node12","NodeNext"],"scope":"window"},"js/ts.implicitProjectConfig.target":{"type":"string","default":"ES2022","markdownDescription":"Set target JavaScript language version for emitted JavaScript and include library declarations. See more: https://www.typescriptlang.org/tsconfig#target.","enum":["ES3","ES5","ES6","ES2015","ES2016","ES2017","ES2018","ES2019","ES2020","ES2021","ES2022","ES2023","ES2024","ESNext"],"scope":"window"},"javascript.implicitProjectConfig.checkJs":{"type":"boolean","default":false,"markdownDescription":"Enable/disable semantic checking of JavaScript files. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","markdownDeprecationMessage":"This setting has been deprecated in favor of `js/ts.implicitProjectConfig.checkJs`.","scope":"window"},"js/ts.implicitProjectConfig.checkJs":{"type":"boolean","default":false,"markdownDescription":"Enable/disable semantic checking of JavaScript files. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","scope":"window"},"javascript.implicitProjectConfig.experimentalDecorators":{"type":"boolean","default":false,"markdownDescription":"Enable/disable `experimentalDecorators` in JavaScript files that are not part of a project. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","markdownDeprecationMessage":"This setting has been deprecated in favor of `js/ts.implicitProjectConfig.experimentalDecorators`.","scope":"window"},"js/ts.implicitProjectConfig.experimentalDecorators":{"type":"boolean","default":false,"markdownDescription":"Enable/disable `experimentalDecorators` in JavaScript files that are not part of a project. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","scope":"window"},"js/ts.implicitProjectConfig.strictNullChecks":{"type":"boolean","default":true,"markdownDescription":"Enable/disable [strict null checks](https://www.typescriptlang.org/tsconfig#strictNullChecks) in JavaScript and TypeScript files that are not part of a project. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","scope":"window"},"js/ts.implicitProjectConfig.strictFunctionTypes":{"type":"boolean","default":true,"markdownDescription":"Enable/disable [strict function types](https://www.typescriptlang.org/tsconfig#strictFunctionTypes) in JavaScript and TypeScript files that are not part of a project. Existing `jsconfig.json` or `tsconfig.json` files override this setting.","scope":"window"},"javascript.suggest.names":{"type":"boolean","default":true,"markdownDescription":"Enable/disable including unique names from the file in JavaScript suggestions. Note that name suggestions are always disabled in JavaScript code that is semantically checked using `@ts-check` or `checkJs`.","scope":"resource"},"typescript.tsc.autoDetect":{"type":"string","default":"on","enum":["on","off","build","watch"],"markdownEnumDescriptions":["Create both build and watch tasks.","Disable this feature.","Only create single run compile tasks.","Only create compile and watch tasks."],"description":"Controls auto detection of tsc tasks.","scope":"window"},"javascript.suggest.paths":{"type":"boolean","default":true,"description":"Enable/disable suggestions for paths in import statements and require calls.","scope":"resource"},"typescript.suggest.paths":{"type":"boolean","default":true,"description":"Enable/disable suggestions for paths in import statements and require calls.","scope":"resource"},"javascript.suggest.autoImports":{"type":"boolean","default":true,"description":"Enable/disable auto import suggestions.","scope":"resource"},"typescript.suggest.autoImports":{"type":"boolean","default":true,"description":"Enable/disable auto import suggestions.","scope":"resource"},"javascript.suggest.completeJSDocs":{"type":"boolean","default":true,"description":"Enable/disable suggestion to complete JSDoc comments.","scope":"language-overridable"},"typescript.suggest.completeJSDocs":{"type":"boolean","default":true,"description":"Enable/disable suggestion to complete JSDoc comments.","scope":"language-overridable"},"javascript.suggest.jsdoc.generateReturns":{"type":"boolean","default":true,"markdownDescription":"Enable/disable generating `@returns` annotations for JSDoc templates.","scope":"language-overridable"},"typescript.suggest.jsdoc.generateReturns":{"type":"boolean","default":true,"markdownDescription":"Enable/disable generating `@returns` annotations for JSDoc templates.","scope":"language-overridable"},"typescript.locale":{"type":"string","default":"auto","enum":["auto","de","es","en","fr","it","ja","ko","ru","zh-CN","zh-TW"],"enumDescriptions":["Use VS Code's configured display language.","Deutsch","español","English","français","italiano","日本語","한국어","русский","中文(简体)","中文(繁體)"],"markdownDescription":"Sets the locale used to report JavaScript and TypeScript errors. Defaults to use VS Code's locale.","scope":"window"},"javascript.suggestionActions.enabled":{"type":"boolean","default":true,"description":"Enable/disable suggestion diagnostics for JavaScript files in the editor.","scope":"resource"},"typescript.suggestionActions.enabled":{"type":"boolean","default":true,"description":"Enable/disable suggestion diagnostics for TypeScript files in the editor.","scope":"resource"},"javascript.preferences.quoteStyle":{"type":"string","enum":["auto","single","double"],"default":"auto","markdownDescription":"Preferred quote style to use for Quick Fixes.","markdownEnumDescriptions":["Infer quote type from existing code","Always use single quotes: `'`","Always use double quotes: `\"`"],"scope":"language-overridable"},"typescript.preferences.quoteStyle":{"type":"string","enum":["auto","single","double"],"default":"auto","markdownDescription":"Preferred quote style to use for Quick Fixes.","markdownEnumDescriptions":["Infer quote type from existing code","Always use single quotes: `'`","Always use double quotes: `\"`"],"scope":"language-overridable"},"javascript.preferences.importModuleSpecifier":{"type":"string","enum":["shortest","relative","non-relative","project-relative"],"markdownEnumDescriptions":["Prefers a non-relative import only if one is available that has fewer path segments than a relative import.","Prefers a relative path to the imported file location.","Prefers a non-relative import based on the `baseUrl` or `paths` configured in your `jsconfig.json` / `tsconfig.json`.","Prefers a non-relative import only if the relative import path would leave the package or project directory."],"default":"shortest","description":"Preferred path style for auto imports.","scope":"language-overridable"},"typescript.preferences.importModuleSpecifier":{"type":"string","enum":["shortest","relative","non-relative","project-relative"],"markdownEnumDescriptions":["Prefers a non-relative import only if one is available that has fewer path segments than a relative import.","Prefers a relative path to the imported file location.","Prefers a non-relative import based on the `baseUrl` or `paths` configured in your `jsconfig.json` / `tsconfig.json`.","Prefers a non-relative import only if the relative import path would leave the package or project directory."],"default":"shortest","description":"Preferred path style for auto imports.","scope":"language-overridable"},"javascript.preferences.importModuleSpecifierEnding":{"type":"string","enum":["auto","minimal","index","js"],"enumItemLabels":[null,null,null,".js / .ts"],"markdownEnumDescriptions":["Use project settings to select a default.","Shorten `./component/index.js` to `./component`.","Shorten `./component/index.js` to `./component/index`.","Do not shorten path endings; include the `.js` or `.ts` extension."],"default":"auto","description":"Preferred path ending for auto imports.","scope":"language-overridable"},"typescript.preferences.importModuleSpecifierEnding":{"type":"string","enum":["auto","minimal","index","js"],"enumItemLabels":[null,null,null,".js / .ts"],"markdownEnumDescriptions":["Use project settings to select a default.","Shorten `./component/index.js` to `./component`.","Shorten `./component/index.js` to `./component/index`.","Do not shorten path endings; include the `.js` or `.ts` extension."],"default":"auto","description":"Preferred path ending for auto imports.","scope":"language-overridable"},"javascript.preferences.jsxAttributeCompletionStyle":{"type":"string","enum":["auto","braces","none"],"markdownEnumDescriptions":["Insert `={}` or `=\"\"` after attribute names based on the prop type. See `#javascript.preferences.quoteStyle#` to control the type of quotes used for string attributes.","Insert `={}` after attribute names.","Only insert attribute names."],"default":"auto","description":"Preferred style for JSX attribute completions.","scope":"language-overridable"},"typescript.preferences.jsxAttributeCompletionStyle":{"type":"string","enum":["auto","braces","none"],"markdownEnumDescriptions":["Insert `={}` or `=\"\"` after attribute names based on the prop type. See `#typescript.preferences.quoteStyle#` to control the type of quotes used for string attributes.","Insert `={}` after attribute names.","Only insert attribute names."],"default":"auto","description":"Preferred style for JSX attribute completions.","scope":"language-overridable"},"typescript.preferences.includePackageJsonAutoImports":{"type":"string","enum":["auto","on","off"],"enumDescriptions":["Search dependencies based on estimated performance impact.","Always search dependencies.","Never search dependencies."],"default":"auto","markdownDescription":"Enable/disable searching `package.json` dependencies for available auto imports.","scope":"window"},"typescript.preferences.autoImportFileExcludePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"Specify glob patterns of files to exclude from auto imports. Relative paths are resolved relative to the workspace root. Patterns are evaluated using tsconfig.json [`exclude`](https://www.typescriptlang.org/tsconfig#exclude) semantics.","scope":"resource"},"javascript.preferences.autoImportFileExcludePatterns":{"type":"array","items":{"type":"string"},"markdownDescription":"Specify glob patterns of files to exclude from auto imports. Relative paths are resolved relative to the workspace root. Patterns are evaluated using tsconfig.json [`exclude`](https://www.typescriptlang.org/tsconfig#exclude) semantics.","scope":"resource"},"typescript.preferences.autoImportSpecifierExcludeRegexes":{"type":"array","items":{"type":"string"},"markdownDescription":"Specify regular expressions to exclude auto imports with matching import specifiers. Examples:\n\n- `^node:`\n- `lib/internal` (slashes don't need to be escaped...)\n- `/lib\\/internal/i` (...unless including surrounding slashes for `i` or `u` flags)\n- `^lodash$` (only allow subpath imports from lodash)","scope":"resource"},"javascript.preferences.autoImportSpecifierExcludeRegexes":{"type":"array","items":{"type":"string"},"markdownDescription":"Specify regular expressions to exclude auto imports with matching import specifiers. Examples:\n\n- `^node:`\n- `lib/internal` (slashes don't need to be escaped...)\n- `/lib\\/internal/i` (...unless including surrounding slashes for `i` or `u` flags)\n- `^lodash$` (only allow subpath imports from lodash)","scope":"resource"},"typescript.preferences.preferTypeOnlyAutoImports":{"type":"boolean","default":false,"markdownDescription":"Include the `type` keyword in auto-imports whenever possible. Requires using TypeScript 5.3+ in the workspace.","scope":"resource"},"javascript.preferences.renameShorthandProperties":{"type":"boolean","default":true,"description":"Enable/disable introducing aliases for object shorthand properties during renames.","deprecationMessage":"The setting 'typescript.preferences.renameShorthandProperties' has been deprecated in favor of 'typescript.preferences.useAliasesForRenames'","scope":"language-overridable"},"typescript.preferences.renameShorthandProperties":{"type":"boolean","default":true,"description":"Enable/disable introducing aliases for object shorthand properties during renames.","deprecationMessage":"The setting 'typescript.preferences.renameShorthandProperties' has been deprecated in favor of 'typescript.preferences.useAliasesForRenames'","scope":"language-overridable"},"javascript.preferences.useAliasesForRenames":{"type":"boolean","default":true,"description":"Enable/disable introducing aliases for object shorthand properties during renames.","scope":"language-overridable"},"typescript.preferences.useAliasesForRenames":{"type":"boolean","default":true,"description":"Enable/disable introducing aliases for object shorthand properties during renames.","scope":"language-overridable"},"javascript.preferences.renameMatchingJsxTags":{"type":"boolean","default":true,"description":"When on a JSX tag, try to rename the matching tag instead of renaming the symbol. Requires using TypeScript 5.1+ in the workspace.","scope":"language-overridable"},"typescript.preferences.renameMatchingJsxTags":{"type":"boolean","default":true,"description":"When on a JSX tag, try to rename the matching tag instead of renaming the symbol. Requires using TypeScript 5.1+ in the workspace.","scope":"language-overridable"},"typescript.preferences.organizeImports":{"type":"object","markdownDescription":"Advanced preferences that control how imports are ordered.","properties":{"caseSensitivity":{"type":"string","markdownDescription":"Specifies how imports should be sorted with regards to case-sensitivity. If `auto` or unspecified, we will detect the case-sensitivity per file","enum":["auto","caseInsensitive","caseSensitive"],"markdownEnumDescriptions":["Detect case-sensitivity for import sorting.","%typescript.preferences.organizeImports.caseSensitivity.insensitive","Sort imports case-sensitively."],"default":"auto"},"typeOrder":{"type":"string","markdownDescription":"Specify how type-only named imports should be sorted.","enum":["auto","last","inline","first"],"default":"auto","markdownEnumDescriptions":["Detect where type-only named imports should be sorted.","Type only named imports are sorted to the end of the import list. E.g. `import { B, Z, type A, type Y } from 'module';`","Named imports are sorted by name only. E.g. `import { type A, B, type Y, Z } from 'module';`","Type only named imports are sorted to the beginning of the import list. E.g. `import { type A, type Y, B, Z } from 'module';`"]},"unicodeCollation":{"type":"string","markdownDescription":"Specify whether to sort imports using Unicode or Ordinal collation.","enum":["ordinal","unicode"],"markdownEnumDescriptions":["Sort imports using the numeric value of each code point.","Sort imports using the Unicode code collation."],"default":"ordinal"},"locale":{"type":"string","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Overrides the locale used for collation. Specify `auto` to use the UI locale."},"numericCollation":{"type":"boolean","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Sort numeric strings by integer value."},"accentCollation":{"type":"boolean","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Compare characters with diacritical marks as unequal to base character."},"caseFirst":{"type":"string","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`, and `organizeImports.caseSensitivity` is not `caseInsensitive`. Indicates whether upper-case will sort before lower-case.","enum":["default","upper","lower"],"markdownEnumDescriptions":["Default order given by `locale`.","Upper-case comes before lower-case. E.g. ` A, a, B, b`.","Lower-case comes before upper-case. E.g.` a, A, z, Z`."],"default":"default"}}},"javascript.preferences.organizeImports":{"type":"object","markdownDescription":"Advanced preferences that control how imports are ordered.","properties":{"caseSensitivity":{"type":"string","markdownDescription":"Specifies how imports should be sorted with regards to case-sensitivity. If `auto` or unspecified, we will detect the case-sensitivity per file","enum":["auto","caseInsensitive","caseSensitive"],"markdownEnumDescriptions":["Detect case-sensitivity for import sorting.","%typescript.preferences.organizeImports.caseSensitivity.insensitive","Sort imports case-sensitively."],"default":"auto"},"typeOrder":{"type":"string","markdownDescription":"Specify how type-only named imports should be sorted.","enum":["auto","last","inline","first"],"default":"auto","markdownEnumDescriptions":["Detect where type-only named imports should be sorted.","Type only named imports are sorted to the end of the import list. E.g. `import { B, Z, type A, type Y } from 'module';`","Named imports are sorted by name only. E.g. `import { type A, B, type Y, Z } from 'module';`","Type only named imports are sorted to the beginning of the import list. E.g. `import { type A, type Y, B, Z } from 'module';`"]},"unicodeCollation":{"type":"string","markdownDescription":"Specify whether to sort imports using Unicode or Ordinal collation.","enum":["ordinal","unicode"],"markdownEnumDescriptions":["Sort imports using the numeric value of each code point.","Sort imports using the Unicode code collation."],"default":"ordinal"},"locale":{"type":"string","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Overrides the locale used for collation. Specify `auto` to use the UI locale."},"numericCollation":{"type":"boolean","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Sort numeric strings by integer value."},"accentCollation":{"type":"boolean","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`. Compare characters with diacritical marks as unequal to base character."},"caseFirst":{"type":"string","markdownDescription":"Requires `organizeImports.unicodeCollation: 'unicode'`, and `organizeImports.caseSensitivity` is not `caseInsensitive`. Indicates whether upper-case will sort before lower-case.","enum":["default","upper","lower"],"markdownEnumDescriptions":["Default order given by `locale`.","Upper-case comes before lower-case. E.g. ` A, a, B, b`.","Lower-case comes before upper-case. E.g.` a, A, z, Z`."],"default":"default"}}},"typescript.updateImportsOnFileMove.enabled":{"type":"string","enum":["prompt","always","never"],"markdownEnumDescriptions":["Prompt on each rename.","Always update paths automatically.","Never rename paths and don't prompt."],"default":"prompt","description":"Enable/disable automatic updating of import paths when you rename or move a file in VS Code.","scope":"resource"},"javascript.updateImportsOnFileMove.enabled":{"type":"string","enum":["prompt","always","never"],"markdownEnumDescriptions":["Prompt on each rename.","Always update paths automatically.","Never rename paths and don't prompt."],"default":"prompt","description":"Enable/disable automatic updating of import paths when you rename or move a file in VS Code.","scope":"resource"},"typescript.autoClosingTags":{"type":"boolean","default":true,"description":"Enable/disable automatic closing of JSX tags.","scope":"language-overridable"},"javascript.autoClosingTags":{"type":"boolean","default":true,"description":"Enable/disable automatic closing of JSX tags.","scope":"language-overridable"},"javascript.suggest.enabled":{"type":"boolean","default":true,"description":"Enabled/disable autocomplete suggestions.","scope":"language-overridable"},"typescript.suggest.enabled":{"type":"boolean","default":true,"description":"Enabled/disable autocomplete suggestions.","scope":"language-overridable"},"typescript.surveys.enabled":{"type":"boolean","default":true,"description":"Enabled/disable occasional surveys that help us improve VS Code's JavaScript and TypeScript support.","scope":"window"},"typescript.tsserver.useSeparateSyntaxServer":{"type":"boolean","default":true,"description":"Enable/disable spawning a separate TypeScript server that can more quickly respond to syntax related operations, such as calculating folding or computing document symbols.","markdownDeprecationMessage":"This setting has been deprecated in favor of `typescript.tsserver.useSyntaxServer`.","scope":"window"},"typescript.tsserver.useSyntaxServer":{"type":"string","scope":"window","description":"Controls if TypeScript launches a dedicated server to more quickly handle syntax related operations, such as computing code folding.","default":"auto","enum":["always","never","auto"],"enumDescriptions":["Use a lighter weight syntax server to handle all IntelliSense operations. This syntax server can only provide IntelliSense for opened files.","Don't use a dedicated syntax server. Use a single server to handle all IntelliSense operations.","Spawn both a full server and a lighter weight server dedicated to syntax operations. The syntax server is used to speed up syntax operations and provide IntelliSense while projects are loading."]},"typescript.tsserver.maxTsServerMemory":{"type":"number","default":3072,"markdownDescription":"The maximum amount of memory (in MB) to allocate to the TypeScript server process. To use a memory limit greater than 4 GB, use `#typescript.tsserver.nodePath#` to run TS Server with a custom Node installation.","scope":"window"},"typescript.tsserver.experimental.enableProjectDiagnostics":{"type":"boolean","default":false,"description":"Enables project wide error reporting.","scope":"window","tags":["experimental"]},"typescript.tsserver.experimental.useVsCodeWatcher":{"type":"boolean","description":"Use VS Code's file watchers instead of TypeScript's. Requires using TypeScript 5.4+ in the workspace.","deprecationMessage":"Please use the `#typescript.tsserver.watchOptions#` setting instead.","default":true},"typescript.tsserver.watchOptions":{"description":"Configure which watching strategies should be used to keep track of files and directories.","scope":"window","default":"vscode","oneOf":[{"type":"string","const":"vscode","description":"Use VS Code's file watchers instead of TypeScript's. Requires using TypeScript 5.4+ in the workspace."},{"type":"object","properties":{"watchFile":{"type":"string","description":"Strategy for how individual files are watched.","enum":["fixedChunkSizePolling","fixedPollingInterval","priorityPollingInterval","dynamicPriorityPolling","useFsEvents","useFsEventsOnParentDirectory"],"enumDescriptions":["Polls files in chunks at regular interval.","Check every file for changes several times a second at a fixed interval.","Check every file for changes several times a second, but use heuristics to check certain types of files less frequently than others.","Use a dynamic queue where less-frequently modified files will be checked less often.","Attempt to use the operating system/file system's native events for file changes.","Attempt to use the operating system/file system's native events to listen for changes on a file's containing directories. This can use fewer file watchers, but might be less accurate."],"default":"useFsEvents"},"watchDirectory":{"type":"string","description":"Strategy for how entire directory trees are watched under systems that lack recursive file-watching functionality.","enum":["fixedChunkSizePolling","fixedPollingInterval","dynamicPriorityPolling","useFsEvents"],"enumDescriptions":["Polls directories in chunks at regular interval.","Check every directory for changes several times a second at a fixed interval.","Use a dynamic queue where less-frequently modified directories will be checked less often.","Attempt to use the operating system/file system's native events for directory changes."],"default":"useFsEvents"},"fallbackPolling":{"type":"string","description":"When using file system events, this option specifies the polling strategy that gets used when the system runs out of native file watchers and/or doesn't support native file watchers.","enum":["fixedPollingInterval","priorityPollingInterval","dynamicPriorityPolling"],"enumDescriptions":["configuration.tsserver.watchOptions.fallbackPolling.fixedPollingInterval","configuration.tsserver.watchOptions.fallbackPolling.priorityPollingInterval","configuration.tsserver.watchOptions.fallbackPolling.dynamicPriorityPolling"]},"synchronousWatchDirectory":{"type":"boolean","description":"Disable deferred watching on directories. Deferred watching is useful when lots of file changes might occur at once (e.g. a change in node_modules from running npm install), but you might want to disable it with this flag for some less-common setups."}}}]},"typescript.workspaceSymbols.scope":{"type":"string","enum":["allOpenProjects","currentProject"],"enumDescriptions":["Search all open JavaScript or TypeScript projects for symbols.","Only search for symbols in the current JavaScript or TypeScript project."],"default":"allOpenProjects","markdownDescription":"Controls which files are searched by [Go to Symbol in Workspace](https://code.visualstudio.com/docs/editor/editingevolved#_open-symbol-by-name).","scope":"window"},"javascript.suggest.classMemberSnippets.enabled":{"type":"boolean","default":true,"description":"Enable/disable snippet completions for class members.","scope":"resource"},"typescript.suggest.classMemberSnippets.enabled":{"type":"boolean","default":true,"description":"Enable/disable snippet completions for class members.","scope":"resource"},"typescript.suggest.objectLiteralMethodSnippets.enabled":{"type":"boolean","default":true,"description":"Enable/disable snippet completions for methods in object literals.","scope":"resource"},"typescript.tsserver.web.projectWideIntellisense.enabled":{"type":"boolean","default":true,"description":"Enable/disable project-wide IntelliSense on web. Requires that VS Code is running in a trusted context.","scope":"window"},"typescript.tsserver.web.projectWideIntellisense.suppressSemanticErrors":{"type":"boolean","default":false,"description":"Suppresses semantic errors on web even when project wide IntelliSense is enabled. This is always on when project wide IntelliSense is not enabled or available. See `#typescript.tsserver.web.projectWideIntellisense.enabled#`","scope":"window"},"typescript.tsserver.web.typeAcquisition.enabled":{"type":"boolean","default":true,"description":"Enable/disable package acquisition on the web. This enables IntelliSense for imported packages. Requires `#typescript.tsserver.web.projectWideIntellisense.enabled#`. Currently not supported for Safari.","scope":"window"},"typescript.tsserver.nodePath":{"type":"string","description":"Run TS Server on a custom Node installation. This can be a path to a Node executable, or 'node' if you want VS Code to detect a Node installation.","scope":"window"},"typescript.preferGoToSourceDefinition":{"type":"boolean","default":false,"description":"Makes `Go to Definition` avoid type declaration files when possible by triggering `Go to Source Definition` instead. This allows `Go to Source Definition` to be triggered with the mouse gesture.","scope":"window"},"javascript.preferGoToSourceDefinition":{"type":"boolean","default":false,"description":"Makes `Go to Definition` avoid type declaration files when possible by triggering `Go to Source Definition` instead. This allows `Go to Source Definition` to be triggered with the mouse gesture.","scope":"window"},"typescript.workspaceSymbols.excludeLibrarySymbols":{"type":"boolean","default":true,"markdownDescription":"Exclude symbols that come from library files in `Go to Symbol in Workspace` results. Requires using TypeScript 5.3+ in the workspace.","scope":"window"},"typescript.tsserver.enableRegionDiagnostics":{"type":"boolean","default":true,"description":"Enables region-based diagnostics in TypeScript. Requires using TypeScript 5.6+ in the workspace.","scope":"window"},"javascript.updateImportsOnPaste.enabled":{"scope":"window","type":"boolean","default":true,"markdownDescription":"Enable updating imports when pasting code. Requires TypeScript 5.7+.\n\nBy default this shows a option to update imports after pasting. You can use the `#editor.pasteAs.preferences#` setting to update imports automatically when pasting: `\"editor.pasteAs.preferences\": [ \"text.updateImports.jsts\" ]`."},"typescript.updateImportsOnPaste.enabled":{"scope":"window","type":"boolean","default":true,"markdownDescription":"Enable updating imports when pasting code. Requires TypeScript 5.7+.\n\nBy default this shows a option to update imports after pasting. You can use the `#editor.pasteAs.preferences#` setting to update imports automatically when pasting: `\"editor.pasteAs.preferences\": [ \"text.updateImports.jsts\" ]`."},"typescript.experimental.expandableHover":{"type":"boolean","default":false,"description":"Enable expanding/contracting the hover to reveal more/less information from the TS server.","scope":"window","tags":["experimental"]}}},"commands":[{"command":"typescript.reloadProjects","title":"Reload Project","category":"TypeScript"},{"command":"javascript.reloadProjects","title":"Reload Project","category":"JavaScript"},{"command":"typescript.selectTypeScriptVersion","title":"Select TypeScript Version...","category":"TypeScript"},{"command":"typescript.goToProjectConfig","title":"Go to Project Configuration (tsconfig)","category":"TypeScript"},{"command":"javascript.goToProjectConfig","title":"Go to Project Configuration (jsconfig / tsconfig)","category":"JavaScript"},{"command":"typescript.openTsServerLog","title":"Open TS Server log","category":"TypeScript"},{"command":"typescript.restartTsServer","title":"Restart TS Server","category":"TypeScript"},{"command":"typescript.findAllFileReferences","title":"Find File References","category":"TypeScript"},{"command":"typescript.goToSourceDefinition","title":"Go to Source Definition","category":"TypeScript"},{"command":"typescript.sortImports","title":"Sort Imports","category":"TypeScript"},{"command":"javascript.sortImports","title":"Sort Imports","category":"JavaScript"},{"command":"typescript.removeUnusedImports","title":"Remove Unused Imports","category":"TypeScript"},{"command":"javascript.removeUnusedImports","title":"Remove Unused Imports","category":"JavaScript"}],"menus":{"commandPalette":[{"command":"typescript.reloadProjects","when":"editorLangId == typescript && typescript.isManagedFile"},{"command":"typescript.reloadProjects","when":"editorLangId == typescriptreact && typescript.isManagedFile"},{"command":"javascript.reloadProjects","when":"editorLangId == javascript && typescript.isManagedFile"},{"command":"javascript.reloadProjects","when":"editorLangId == javascriptreact && typescript.isManagedFile"},{"command":"typescript.goToProjectConfig","when":"editorLangId == typescript && typescript.isManagedFile"},{"command":"typescript.goToProjectConfig","when":"editorLangId == typescriptreact"},{"command":"javascript.goToProjectConfig","when":"editorLangId == javascript && typescript.isManagedFile"},{"command":"javascript.goToProjectConfig","when":"editorLangId == javascriptreact && typescript.isManagedFile"},{"command":"typescript.selectTypeScriptVersion","when":"typescript.isManagedFile"},{"command":"typescript.openTsServerLog","when":"typescript.isManagedFile"},{"command":"typescript.restartTsServer","when":"typescript.isManagedFile"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && typescript.isManagedFile"},{"command":"typescript.goToSourceDefinition","when":"tsSupportsSourceDefinition && typescript.isManagedFile"},{"command":"typescript.sortImports","when":"supportedCodeAction =~ /(\\s|^)source\\.sortImports\\b/ && editorLangId =~ /^typescript(react)?$/"},{"command":"javascript.sortImports","when":"supportedCodeAction =~ /(\\s|^)source\\.sortImports\\b/ && editorLangId =~ /^javascript(react)?$/"},{"command":"typescript.removeUnusedImports","when":"supportedCodeAction =~ /(\\s|^)source\\.removeUnusedImports\\b/ && editorLangId =~ /^typescript(react)?$/"},{"command":"javascript.removeUnusedImports","when":"supportedCodeAction =~ /(\\s|^)source\\.removeUnusedImports\\b/ && editorLangId =~ /^javascript(react)?$/"}],"editor/context":[{"command":"typescript.goToSourceDefinition","when":"tsSupportsSourceDefinition && (resourceLangId == typescript || resourceLangId == typescriptreact || resourceLangId == javascript || resourceLangId == javascriptreact)","group":"navigation@1.41"}],"explorer/context":[{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == typescript","group":"4_search"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == typescriptreact","group":"4_search"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == javascript","group":"4_search"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == javascriptreact","group":"4_search"}],"editor/title/context":[{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == javascript"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == javascriptreact"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == typescript"},{"command":"typescript.findAllFileReferences","when":"tsSupportsFileReferences && resourceLangId == typescriptreact"}]},"breakpoints":[{"language":"typescript"},{"language":"typescriptreact"}],"taskDefinitions":[{"type":"typescript","required":["tsconfig"],"properties":{"tsconfig":{"type":"string","description":"The tsconfig file that defines the TS build."},"option":{"type":"string"}},"when":"shellExecutionSupported"}],"problemPatterns":[{"name":"tsc","regexp":"^([^\\s].*)[\\(:](\\d+)[,:](\\d+)(?:\\):\\s+|\\s+-\\s+)(error|warning|info)\\s+TS(\\d+)\\s*:\\s*(.*)$","file":1,"line":2,"column":3,"severity":4,"code":5,"message":6}],"problemMatchers":[{"name":"tsc","label":"TypeScript problems","owner":"typescript","source":"ts","applyTo":"closedDocuments","fileLocation":["relative","${cwd}"],"pattern":"$tsc"},{"name":"tsc-watch","label":"TypeScript problems (watch mode)","owner":"typescript","source":"ts","applyTo":"closedDocuments","fileLocation":["relative","${cwd}"],"pattern":"$tsc","background":{"activeOnStart":true,"beginsPattern":{"regexp":"^\\s*(?:message TS6032:|\\[?\\D*.{1,2}[:.].{1,2}[:.].{1,2}\\D*(├\\D*\\d{1,2}\\D+┤)?(?:\\]| -)) (Starting compilation in watch mode|File change detected\\. Starting incremental compilation)\\.\\.\\."},"endsPattern":{"regexp":"^\\s*(?:message TS6042:|\\[?\\D*.{1,2}[:.].{1,2}[:.].{1,2}\\D*(├\\D*\\d{1,2}\\D+┤)?(?:\\]| -)) (?:Compilation complete\\.|Found \\d+ errors?\\.) Watching for file changes\\."}}}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/typescript-language-features","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.vb"},"manifest":{"name":"vb","displayName":"Visual Basic Language Basics","description":"Provides snippets, syntax highlighting, bracket matching and folding in Visual Basic files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin textmate/asp.vb.net.tmbundle Syntaxes/ASP%20VB.net.plist ./syntaxes/asp-vb-net.tmLanguage.json"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"vb","extensions":[".vb",".brs",".vbs",".bas",".vba"],"aliases":["Visual Basic","vb"],"configuration":"./language-configuration.json"}],"grammars":[{"language":"vb","scopeName":"source.asp.vb.net","path":"./syntaxes/asp-vb-net.tmLanguage.json"}],"snippets":[{"language":"vb","path":"./snippets/vb.code-snippets"}]},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/vb","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.xml"},"manifest":{"name":"xml","displayName":"XML Language Basics","description":"Provides syntax highlighting and bracket matching in XML files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"xml","extensions":[".xml",".xsd",".ascx",".atom",".axml",".axaml",".bpmn",".cpt",".csl",".csproj",".csproj.user",".dita",".ditamap",".dtd",".ent",".mod",".dtml",".fsproj",".fxml",".iml",".isml",".jmx",".launch",".menu",".mxml",".nuspec",".opml",".owl",".proj",".props",".pt",".publishsettings",".pubxml",".pubxml.user",".rbxlx",".rbxmx",".rdf",".rng",".rss",".shproj",".storyboard",".svg",".targets",".tld",".tmx",".vbproj",".vbproj.user",".vcxproj",".vcxproj.filters",".wsdl",".wxi",".wxl",".wxs",".xaml",".xbl",".xib",".xlf",".xliff",".xpdl",".xul",".xoml"],"firstLine":"(\\<\\?xml.*)|(\\<svg)|(\\<\\!doctype\\s+svg)","aliases":["XML","xml"],"configuration":"./xml.language-configuration.json"},{"id":"xsl","extensions":[".xsl",".xslt"],"aliases":["XSL","xsl"],"configuration":"./xsl.language-configuration.json"}],"grammars":[{"language":"xml","scopeName":"text.xml","path":"./syntaxes/xml.tmLanguage.json"},{"language":"xsl","scopeName":"text.xml.xsl","path":"./syntaxes/xsl.tmLanguage.json"}]},"scripts":{"update-grammar":"node ../node_modules/vscode-grammar-updater/bin atom/language-xml grammars/xml.cson ./syntaxes/xml.tmLanguage.json grammars/xsl.cson ./syntaxes/xsl.tmLanguage.json"},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/xml","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false},{"type":0,"identifier":{"id":"vscode.yaml"},"manifest":{"name":"yaml","displayName":"YAML Language Basics","description":"Provides syntax highlighting and bracket matching in YAML files.","version":"1.0.0","publisher":"vscode","license":"MIT","engines":{"vscode":"*"},"scripts":{"update-grammar":"node ./build/update-grammar.js"},"categories":["Programming Languages"],"contributes":{"languages":[{"id":"dockercompose","aliases":["Compose","compose"],"filenamePatterns":["compose.yml","compose.yaml","compose.*.yml","compose.*.yaml","*docker*compose*.yml","*docker*compose*.yaml"],"configuration":"./language-configuration.json"},{"id":"yaml","aliases":["YAML","yaml"],"extensions":[".yaml",".yml",".eyaml",".eyml",".cff",".yaml-tmlanguage",".yaml-tmpreferences",".yaml-tmtheme",".winget"],"firstLine":"^#cloud-config","configuration":"./language-configuration.json"}],"grammars":[{"language":"dockercompose","scopeName":"source.yaml","path":"./syntaxes/yaml.tmLanguage.json"},{"scopeName":"source.yaml.1.3","path":"./syntaxes/yaml-1.3.tmLanguage.json"},{"scopeName":"source.yaml.1.2","path":"./syntaxes/yaml-1.2.tmLanguage.json"},{"scopeName":"source.yaml.1.1","path":"./syntaxes/yaml-1.1.tmLanguage.json"},{"scopeName":"source.yaml.1.0","path":"./syntaxes/yaml-1.0.tmLanguage.json"},{"scopeName":"source.yaml.embedded","path":"./syntaxes/yaml-embedded.tmLanguage.json"},{"language":"yaml","scopeName":"source.yaml","path":"./syntaxes/yaml.tmLanguage.json","unbalancedBracketScopes":["invalid.illegal","meta.scalar.yaml","storage.type.tag.shorthand.yaml","keyword.control.flow"]}],"configurationDefaults":{"[yaml]":{"editor.insertSpaces":true,"editor.tabSize":2,"editor.autoIndent":"advanced","diffEditor.ignoreTrimWhitespace":false,"editor.defaultColorDecorators":false},"[dockercompose]":{"editor.insertSpaces":true,"editor.tabSize":2,"editor.autoIndent":"advanced"}}},"repository":{"type":"git","url":"https://github.com/microsoft/vscode.git"}},"location":{"$mid":1,"path":"/ide/extensions/yaml","scheme":"file"},"isBuiltin":true,"targetPlatform":"undefined","isValid":true,"validations":[],"preRelease":false}]}
\ No newline at end of file
diff --git a/.vscode-remote/extensions/extensions.json b/.vscode-remote/extensions/extensions.json
new file mode 100644
index 0000000000000..0637a088a01e8
--- /dev/null
+++ b/.vscode-remote/extensions/extensions.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.vscode-remote/extensions/ms-python.debugpy-2025.0.1-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_frame_eval/pydevd_frame_evaluator.pyx b/.vscode-remote/extensions/ms-python.debugpy-2025.0.1-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_frame_eval/pydevd_frame_evaluator.pyx
new file mode 100644
index 0000000000000..8e7b9965af1c4
--- /dev/null
+++ b/.vscode-remote/extensions/ms-python.debugpy-2025.0.1-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_frame_eval/pydevd_frame_evaluator.pyx
@@ -0,0 +1,747 @@
+from __future__ import print_function
+from _pydev_bundle._pydev_saved_modules import threading, thread
+from _pydevd_bundle.pydevd_constants import GlobalDebuggerHolder
+import dis
+import sys
+from _pydevd_frame_eval.pydevd_frame_tracing import update_globals_dict, dummy_tracing_holder
+from _pydevd_frame_eval.pydevd_modify_bytecode import DebugHelper, insert_pydevd_breaks
+from pydevd_file_utils import get_abs_path_real_path_and_base_from_frame, NORM_PATHS_AND_BASE_CONTAINER
+from _pydevd_bundle.pydevd_trace_dispatch import fix_top_level_trace_and_get_trace_func
+
+from _pydevd_bundle.pydevd_additional_thread_info import _set_additional_thread_info_lock
+from _pydevd_bundle.pydevd_cython cimport PyDBAdditionalThreadInfo
+from pydevd_tracing import SetTrace
+
+_get_ident = threading.get_ident  # Note this is py3 only, if py2 needed to be supported, _get_ident would be needed.
+_thread_local_info = threading.local()
+_thread_active = threading._active
+
+def clear_thread_local_info():
+    global _thread_local_info
+    _thread_local_info = threading.local()
+
+
+cdef class ThreadInfo:
+
+    cdef public PyDBAdditionalThreadInfo additional_info
+    cdef public bint is_pydevd_thread
+    cdef public int inside_frame_eval
+    cdef public bint fully_initialized
+    cdef public object thread_trace_func
+    cdef bint _can_create_dummy_thread
+
+    # Note: whenever get_func_code_info is called, this value is reset (we're using
+    # it as a thread-local value info).
+    # If True the debugger should not go into trace mode even if the new
+    # code for a function is None and there are breakpoints.
+    cdef public bint force_stay_in_untraced_mode
+
+    cdef initialize(self, PyFrameObject * frame_obj):
+        # Places that create a ThreadInfo should verify that
+        # a current Python frame is being executed!
+        assert frame_obj != NULL
+
+        self.additional_info = None
+        self.is_pydevd_thread = False
+        self.inside_frame_eval = 0
+        self.fully_initialized = False
+        self.thread_trace_func = None
+
+        # Get the root (if it's not a Thread initialized from the threading
+        # module, create the dummy thread entry so that we can debug it --
+        # otherwise, we have to wait for the threading module itself to
+        # create the Thread entry).
+        while frame_obj.f_back != NULL:
+            frame_obj = frame_obj.f_back
+
+        basename = <str> frame_obj.f_code.co_filename
+        i = basename.rfind('/')
+        j = basename.rfind('\\')
+        if j > i:
+            i = j
+        if i >= 0:
+            basename = basename[i + 1:]
+        # remove ext
+        i = basename.rfind('.')
+        if i >= 0:
+            basename = basename[:i]
+
+        co_name = <str> frame_obj.f_code.co_name
+
+        # In these cases we cannot create a dummy thread (an actual
+        # thread will be created later or tracing will already be set).
+        if basename == 'threading' and co_name in ('__bootstrap', '_bootstrap', '__bootstrap_inner', '_bootstrap_inner'):
+            self._can_create_dummy_thread = False
+        elif basename == 'pydev_monkey' and co_name == '__call__':
+            self._can_create_dummy_thread = False
+        elif basename == 'pydevd' and co_name in ('run', 'main', '_exec'):
+            self._can_create_dummy_thread = False
+        elif basename == 'pydevd_tracing':
+            self._can_create_dummy_thread = False
+        else:
+            self._can_create_dummy_thread = True
+
+        # print('Can create dummy thread for thread started in: %s %s' % (basename, co_name))
+
+    cdef initialize_if_possible(self):
+        # Don't call threading.currentThread because if we're too early in the process
+        # we may create a dummy thread.
+        self.inside_frame_eval += 1
+
+        try:
+            thread_ident = _get_ident()
+            t = _thread_active.get(thread_ident)
+            if t is None:
+                if self._can_create_dummy_thread:
+                    # Initialize the dummy thread and set the tracing (both are needed to
+                    # actually stop on breakpoints).
+                    t = threading.current_thread()
+                    SetTrace(dummy_trace_dispatch)
+                else:
+                    return  # Cannot initialize until thread becomes active.
+
+            if getattr(t, 'is_pydev_daemon_thread', False):
+                self.is_pydevd_thread = True
+                self.fully_initialized = True
+            else:
+                try:
+                    additional_info = t.additional_info
+                    if additional_info is None:
+                        raise AttributeError()
+                except:
+                    with _set_additional_thread_info_lock:
+                        # If it's not there, set it within a lock to avoid any racing
+                        # conditions.
+                        additional_info = getattr(thread, 'additional_info', None)
+                        if additional_info is None:
+                            additional_info = PyDBAdditionalThreadInfo()
+                        t.additional_info = additional_info
+                self.additional_info = additional_info
+                self.fully_initialized = True
+        finally:
+            self.inside_frame_eval -= 1
+
+
+cdef class FuncCodeInfo:
+
+    cdef public str co_filename
+    cdef public str co_name
+    cdef public str canonical_normalized_filename
+    cdef bint always_skip_code
+    cdef public bint breakpoint_found
+    cdef public object new_code
+
+    # When breakpoints_mtime != PyDb.mtime the validity of breakpoints have
+    # to be re-evaluated (if invalid a new FuncCodeInfo must be created and
+    # tracing can't be disabled for the related frames).
+    cdef public int breakpoints_mtime
+
+    def __init__(self):
+        self.co_filename = ''
+        self.canonical_normalized_filename = ''
+        self.always_skip_code = False
+
+        # If breakpoints are found but new_code is None,
+        # this means we weren't able to actually add the code
+        # where needed, so, fallback to tracing.
+        self.breakpoint_found = False
+        self.new_code = None
+        self.breakpoints_mtime = -1
+
+
+def dummy_trace_dispatch(frame, str event, arg):
+    if event == 'call':
+        if frame.f_trace is not None:
+            return frame.f_trace(frame, event, arg)
+    return None
+
+
+def get_thread_info_py() -> ThreadInfo:
+    return get_thread_info(PyEval_GetFrame())
+
+
+cdef ThreadInfo get_thread_info(PyFrameObject * frame_obj):
+    '''
+    Provides thread-related info.
+
+    May return None if the thread is still not active.
+    '''
+    cdef ThreadInfo thread_info
+    try:
+        # Note: changing to a `dict[thread.ident] = thread_info` had almost no
+        # effect in the performance.
+        thread_info = _thread_local_info.thread_info
+    except:
+        if frame_obj == NULL:
+            return None
+        thread_info = ThreadInfo()
+        thread_info.initialize(frame_obj)
+        thread_info.inside_frame_eval += 1
+        try:
+            _thread_local_info.thread_info = thread_info
+
+            # Note: _code_extra_index is not actually thread-related,
+            # but this is a good point to initialize it.
+            global _code_extra_index
+            if _code_extra_index == -1:
+                _code_extra_index = <int> _PyEval_RequestCodeExtraIndex(release_co_extra)
+
+            thread_info.initialize_if_possible()
+        finally:
+            thread_info.inside_frame_eval -= 1
+
+    return thread_info
+
+
+def decref_py(obj):
+    '''
+    Helper to be called from Python.
+    '''
+    Py_DECREF(obj)
+
+
+def get_func_code_info_py(thread_info, frame, code_obj) -> FuncCodeInfo:
+    '''
+    Helper to be called from Python.
+    '''
+    return get_func_code_info(<ThreadInfo> thread_info, <PyFrameObject *> frame, <PyCodeObject *> code_obj)
+
+
+cdef int _code_extra_index = -1
+
+cdef FuncCodeInfo get_func_code_info(ThreadInfo thread_info, PyFrameObject * frame_obj, PyCodeObject * code_obj):
+    '''
+    Provides code-object related info.
+
+    Stores the gathered info in a cache in the code object itself. Note that
+    multiple threads can get the same info.
+
+    get_thread_info() *must* be called at least once before get_func_code_info()
+    to initialize _code_extra_index.
+
+    '''
+    # f_code = <object> code_obj
+    # DEBUG = f_code.co_filename.endswith('_debugger_case_multiprocessing.py')
+    # if DEBUG:
+    #     print('get_func_code_info', f_code.co_name, f_code.co_filename)
+
+    cdef object main_debugger = GlobalDebuggerHolder.global_dbg
+    thread_info.force_stay_in_untraced_mode = False  # This is an output value of the function.
+
+    cdef PyObject * extra
+    _PyCode_GetExtra(<PyObject *> code_obj, _code_extra_index, & extra)
+    if extra is not NULL:
+        extra_obj = <PyObject *> extra
+        if extra_obj is not NULL:
+            func_code_info_obj = <FuncCodeInfo> extra_obj
+            if func_code_info_obj.breakpoints_mtime == main_debugger.mtime:
+                # if DEBUG:
+                #     print('get_func_code_info: matched mtime', f_code.co_name, f_code.co_filename)
+
+                return func_code_info_obj
+
+    cdef str co_filename = <str> code_obj.co_filename
+    cdef str co_name = <str> code_obj.co_name
+    cdef dict cache_file_type
+    cdef tuple cache_file_type_key
+
+    func_code_info = FuncCodeInfo()
+    func_code_info.breakpoints_mtime = main_debugger.mtime
+
+    func_code_info.co_filename = co_filename
+    func_code_info.co_name = co_name
+
+    if not func_code_info.always_skip_code:
+        try:
+            abs_path_real_path_and_base = NORM_PATHS_AND_BASE_CONTAINER[co_filename]
+        except:
+            abs_path_real_path_and_base = get_abs_path_real_path_and_base_from_frame(<object>frame_obj)
+
+        func_code_info.canonical_normalized_filename = abs_path_real_path_and_base[1]
+
+        cache_file_type = main_debugger.get_cache_file_type()
+        # Note: this cache key must be the same from PyDB.get_file_type() -- see it for comments
+        # on the cache.
+        cache_file_type_key = (frame_obj.f_code.co_firstlineno, abs_path_real_path_and_base[0], <object>frame_obj.f_code)
+        try:
+            file_type = cache_file_type[cache_file_type_key]  # Make it faster
+        except:
+            file_type = main_debugger.get_file_type(<object>frame_obj, abs_path_real_path_and_base)  # we don't want to debug anything related to pydevd
+
+        if file_type is not None:
+            func_code_info.always_skip_code = True
+
+    if not func_code_info.always_skip_code:
+        if main_debugger is not None:
+
+            breakpoints: dict = main_debugger.breakpoints.get(func_code_info.canonical_normalized_filename)
+            function_breakpoint: object = main_debugger.function_breakpoint_name_to_breakpoint.get(func_code_info.co_name)
+            # print('\n---')
+            # print(main_debugger.breakpoints)
+            # print(func_code_info.canonical_normalized_filename)
+            # print(main_debugger.breakpoints.get(func_code_info.canonical_normalized_filename))
+            code_obj_py: object = <object> code_obj
+            cached_code_obj_info: object = _cache.get(code_obj_py)
+            if cached_code_obj_info:
+                # The cache is for new code objects, so, in this case it's already
+                # using the new code and we can't change it as this is a generator!
+                # There's still a catch though: even though we don't replace the code,
+                # we may not want to go into tracing mode (as would usually happen
+                # when the new_code is None).
+                func_code_info.new_code = None
+                breakpoint_found, thread_info.force_stay_in_untraced_mode = \
+                    cached_code_obj_info.compute_force_stay_in_untraced_mode(breakpoints)
+                func_code_info.breakpoint_found = breakpoint_found
+
+            elif function_breakpoint:
+                # Go directly into tracing mode
+                func_code_info.breakpoint_found = True
+                func_code_info.new_code = None
+                
+            elif breakpoints:
+                # if DEBUG:
+                #    print('found breakpoints', code_obj_py.co_name, breakpoints)
+
+                # Note: new_code can be None if unable to generate.
+                # It should automatically put the new code object in the cache.
+                breakpoint_found, func_code_info.new_code = generate_code_with_breakpoints(code_obj_py, breakpoints)
+                func_code_info.breakpoint_found = breakpoint_found
+
+    Py_INCREF(func_code_info)
+    _PyCode_SetExtra(<PyObject *> code_obj, _code_extra_index, <PyObject *> func_code_info)
+
+    return func_code_info
+
+
+cdef class _CodeLineInfo:
+
+    cdef public dict line_to_offset
+    cdef public int first_line
+    cdef public int last_line
+
+    def __init__(self, dict line_to_offset,  int first_line,  int last_line):
+        self.line_to_offset = line_to_offset
+        self.first_line = first_line
+        self.last_line = last_line
+
+
+# Note: this method has a version in pure-python too.
+def _get_code_line_info(code_obj):
+    line_to_offset: dict = {}
+    first_line: int = None
+    last_line: int = None
+
+    cdef int offset
+    cdef int line
+
+    for offset, line in dis.findlinestarts(code_obj):
+        line_to_offset[line] = offset
+
+    if line_to_offset:
+        first_line = min(line_to_offset)
+        last_line = max(line_to_offset)
+    return _CodeLineInfo(line_to_offset, first_line, last_line)
+
+
+# Note: this is a cache where the key is the code objects we create ourselves so that
+# we always return the same code object for generators.
+# (so, we don't have a cache from the old code to the new info -- that's actually
+# handled by the cython side in `FuncCodeInfo get_func_code_info` by providing the
+# same code info if the debugger mtime is still the same).
+_cache: dict = {}
+
+def get_cached_code_obj_info_py(code_obj_py):
+    '''
+    :return _CacheValue:
+    :note: on cython use _cache.get(code_obj_py) directly.
+    '''
+    return _cache.get(code_obj_py)
+
+
+cdef class _CacheValue(object):
+
+    cdef public object code_obj_py
+    cdef public _CodeLineInfo code_line_info
+    cdef public set breakpoints_hit_at_lines
+    cdef public set code_lines_as_set
+
+    def __init__(self, object code_obj_py, _CodeLineInfo code_line_info, set breakpoints_hit_at_lines):
+        '''
+        :param code_obj_py:
+        :param _CodeLineInfo code_line_info:
+        :param set[int] breakpoints_hit_at_lines:
+        '''
+        self.code_obj_py = code_obj_py
+        self.code_line_info = code_line_info
+        self.breakpoints_hit_at_lines = breakpoints_hit_at_lines
+        self.code_lines_as_set = set(code_line_info.line_to_offset)
+
+    cpdef compute_force_stay_in_untraced_mode(self, breakpoints):
+        '''
+        :param breakpoints:
+            set(breakpoint_lines) or dict(breakpoint_line->breakpoint info)
+        :return tuple(breakpoint_found, force_stay_in_untraced_mode)
+        '''
+        cdef bint force_stay_in_untraced_mode
+        cdef bint breakpoint_found
+        cdef set target_breakpoints
+
+        force_stay_in_untraced_mode = False
+
+        target_breakpoints = self.code_lines_as_set.intersection(breakpoints)
+        breakpoint_found = bool(target_breakpoints)
+
+        if not breakpoint_found:
+            force_stay_in_untraced_mode = True
+        else:
+            force_stay_in_untraced_mode = self.breakpoints_hit_at_lines.issuperset(set(breakpoints))
+
+        return breakpoint_found, force_stay_in_untraced_mode
+
+def generate_code_with_breakpoints_py(object code_obj_py, dict breakpoints):
+    return generate_code_with_breakpoints(code_obj_py, breakpoints)
+
+# DEBUG = True
+# debug_helper = DebugHelper()
+
+cdef generate_code_with_breakpoints(object code_obj_py, dict breakpoints):
+    '''
+    :param breakpoints:
+        dict where the keys are the breakpoint lines.
+    :return tuple(breakpoint_found, new_code)
+    '''
+    # The cache is needed for generator functions, because after each yield a new frame
+    # is created but the former code object is used (so, check if code_to_modify is
+    # already there and if not cache based on the new code generated).
+
+    cdef bint success
+    cdef int breakpoint_line
+    cdef bint breakpoint_found
+    cdef _CacheValue cache_value
+    cdef set breakpoints_hit_at_lines
+    cdef dict line_to_offset
+
+    assert code_obj_py not in _cache, 'If a code object is cached, that same code object must be reused.'
+
+#     if DEBUG:
+#         initial_code_obj_py = code_obj_py
+
+    code_line_info = _get_code_line_info(code_obj_py)
+
+    success = True
+
+    breakpoints_hit_at_lines = set()
+    line_to_offset = code_line_info.line_to_offset
+
+    for breakpoint_line in breakpoints:
+        if breakpoint_line in line_to_offset:
+            breakpoints_hit_at_lines.add(breakpoint_line)
+
+    if breakpoints_hit_at_lines:
+        success, new_code = insert_pydevd_breaks(
+            code_obj_py,
+            breakpoints_hit_at_lines,
+            code_line_info
+        )
+
+        if not success:
+            code_obj_py = None
+        else:
+            code_obj_py = new_code
+
+    breakpoint_found = bool(breakpoints_hit_at_lines)
+    if breakpoint_found and success:
+#         if DEBUG:
+#             op_number = debug_helper.write_dis(
+#                 'inserting code, breaks at: %s' % (list(breakpoints),),
+#                 initial_code_obj_py
+#             )
+#
+#             debug_helper.write_dis(
+#                 'after inserting code, breaks at: %s' % (list(breakpoints,)),
+#                 code_obj_py,
+#                 op_number=op_number,
+#             )
+
+        cache_value = _CacheValue(code_obj_py, code_line_info, breakpoints_hit_at_lines)
+        _cache[code_obj_py] = cache_value
+
+    return breakpoint_found, code_obj_py
+
+import sys
+
+cdef bint IS_PY_39_OWNARDS = sys.version_info[:2] >= (3, 9)
+
+def frame_eval_func():
+    cdef PyThreadState *state = PyThreadState_Get()
+    if IS_PY_39_OWNARDS:
+        state.interp.eval_frame = <_PyFrameEvalFunction *> get_bytecode_while_frame_eval_39
+    else:
+        state.interp.eval_frame = <_PyFrameEvalFunction *> get_bytecode_while_frame_eval_38
+    dummy_tracing_holder.set_trace_func(dummy_trace_dispatch)
+
+
+def stop_frame_eval():
+    cdef PyThreadState *state = PyThreadState_Get()
+    state.interp.eval_frame = _PyEval_EvalFrameDefault
+
+# During the build we'll generate 2 versions of the code below so that we're compatible with
+# Python 3.9, which receives a "PyThreadState* tstate" as the first parameter and Python 3.6-3.8
+# which doesn't.
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+cdef PyObject * get_bytecode_while_frame_eval_38(PyFrameObject * frame_obj, int exc):
+    '''
+    This function makes the actual evaluation and changes the bytecode to a version
+    where programmatic breakpoints are added.
+    '''
+    if GlobalDebuggerHolder is None or _thread_local_info is None or exc:
+        # Sometimes during process shutdown these global variables become None
+        return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    # co_filename: str = <str>frame_obj.f_code.co_filename
+    # if co_filename.endswith('threading.py'):
+    #     return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    cdef ThreadInfo thread_info
+    cdef int STATE_SUSPEND = 2
+    cdef int CMD_STEP_INTO = 107
+    cdef int CMD_STEP_OVER = 108
+    cdef int CMD_STEP_OVER_MY_CODE = 159
+    cdef int CMD_STEP_INTO_MY_CODE = 144
+    cdef int CMD_STEP_INTO_COROUTINE = 206
+    cdef int CMD_SMART_STEP_INTO = 128
+    cdef bint can_skip = True
+    try:
+        thread_info = _thread_local_info.thread_info
+    except:
+        thread_info = get_thread_info(frame_obj)
+        if thread_info is None:
+            return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    if thread_info.inside_frame_eval:
+        return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    if not thread_info.fully_initialized:
+        thread_info.initialize_if_possible()
+        if not thread_info.fully_initialized:
+            return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    # Can only get additional_info when fully initialized.
+    cdef PyDBAdditionalThreadInfo additional_info = thread_info.additional_info
+    if thread_info.is_pydevd_thread or additional_info.is_tracing:
+        # Make sure that we don't trace pydevd threads or inside our own calls.
+        return CALL_EvalFrameDefault_38(frame_obj, exc)
+
+    # frame = <object> frame_obj
+    # DEBUG = frame.f_code.co_filename.endswith('_debugger_case_tracing.py')
+    # if DEBUG:
+    #     print('get_bytecode_while_frame_eval', frame.f_lineno, frame.f_code.co_name, frame.f_code.co_filename)
+
+    thread_info.inside_frame_eval += 1
+    additional_info.is_tracing = True
+    try:
+        main_debugger: object = GlobalDebuggerHolder.global_dbg
+        if main_debugger is None:
+            return CALL_EvalFrameDefault_38(frame_obj, exc)
+        frame = <object> frame_obj
+
+        if thread_info.thread_trace_func is None:
+            trace_func, apply_to_global = fix_top_level_trace_and_get_trace_func(main_debugger, frame)
+            if apply_to_global:
+                thread_info.thread_trace_func = trace_func
+
+        if additional_info.pydev_step_cmd in (CMD_STEP_INTO, CMD_STEP_INTO_MY_CODE, CMD_STEP_INTO_COROUTINE, CMD_SMART_STEP_INTO) or \
+                main_debugger.break_on_caught_exceptions or \
+                main_debugger.break_on_user_uncaught_exceptions or \
+                main_debugger.has_plugin_exception_breaks or \
+                main_debugger.signature_factory or \
+                additional_info.pydev_step_cmd in (CMD_STEP_OVER, CMD_STEP_OVER_MY_CODE) and main_debugger.show_return_values and frame.f_back is additional_info.pydev_step_stop:
+
+            # if DEBUG:
+            #     print('get_bytecode_while_frame_eval enabled trace')
+            if thread_info.thread_trace_func is not None:
+                frame.f_trace = thread_info.thread_trace_func
+            else:
+                frame.f_trace = <object> main_debugger.trace_dispatch
+        else:
+            func_code_info: FuncCodeInfo = get_func_code_info(thread_info, frame_obj, frame_obj.f_code)
+            # if DEBUG:
+            #     print('get_bytecode_while_frame_eval always skip', func_code_info.always_skip_code)
+            if not func_code_info.always_skip_code:
+
+                if main_debugger.has_plugin_line_breaks or main_debugger.has_plugin_exception_breaks:
+                    can_skip = main_debugger.plugin.can_skip(main_debugger, <object> frame_obj)
+
+                    if not can_skip:
+                        # if DEBUG:
+                        #     print('get_bytecode_while_frame_eval not can_skip')
+                        if thread_info.thread_trace_func is not None:
+                            frame.f_trace = thread_info.thread_trace_func
+                        else:
+                            frame.f_trace = <object> main_debugger.trace_dispatch
+
+                if can_skip and func_code_info.breakpoint_found:
+                    # if DEBUG:
+                    #     print('get_bytecode_while_frame_eval new_code', func_code_info.new_code)
+                    if not thread_info.force_stay_in_untraced_mode:
+                        # If breakpoints are found but new_code is None,
+                        # this means we weren't able to actually add the code
+                        # where needed, so, fallback to tracing.
+                        if func_code_info.new_code is None:
+                            if thread_info.thread_trace_func is not None:
+                                frame.f_trace = thread_info.thread_trace_func
+                            else:
+                                frame.f_trace = <object> main_debugger.trace_dispatch
+                        else:
+                            # print('Using frame eval break for', <object> frame_obj.f_code.co_name)
+                            update_globals_dict(<object> frame_obj.f_globals)
+                            Py_INCREF(func_code_info.new_code)
+                            old = <object> frame_obj.f_code
+                            frame_obj.f_code = <PyCodeObject *> func_code_info.new_code
+                            Py_DECREF(old)
+                    else:
+                        # When we're forcing to stay in traced mode we need to
+                        # update the globals dict (because this means that we're reusing
+                        # a previous code which had breakpoints added in a new frame).
+                        update_globals_dict(<object> frame_obj.f_globals)
+
+    finally:
+        thread_info.inside_frame_eval -= 1
+        additional_info.is_tracing = False
+
+    return CALL_EvalFrameDefault_38(frame_obj, exc)
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+
+
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+cdef PyObject * get_bytecode_while_frame_eval_39(PyThreadState* tstate, PyFrameObject * frame_obj, int exc):
+    '''
+    This function makes the actual evaluation and changes the bytecode to a version
+    where programmatic breakpoints are added.
+    '''
+    if GlobalDebuggerHolder is None or _thread_local_info is None or exc:
+        # Sometimes during process shutdown these global variables become None
+        return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    # co_filename: str = <str>frame_obj.f_code.co_filename
+    # if co_filename.endswith('threading.py'):
+    #     return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    cdef ThreadInfo thread_info
+    cdef int STATE_SUSPEND = 2
+    cdef int CMD_STEP_INTO = 107
+    cdef int CMD_STEP_OVER = 108
+    cdef int CMD_STEP_OVER_MY_CODE = 159
+    cdef int CMD_STEP_INTO_MY_CODE = 144
+    cdef int CMD_STEP_INTO_COROUTINE = 206
+    cdef int CMD_SMART_STEP_INTO = 128
+    cdef bint can_skip = True
+    try:
+        thread_info = _thread_local_info.thread_info
+    except:
+        thread_info = get_thread_info(frame_obj)
+        if thread_info is None:
+            return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    if thread_info.inside_frame_eval:
+        return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    if not thread_info.fully_initialized:
+        thread_info.initialize_if_possible()
+        if not thread_info.fully_initialized:
+            return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    # Can only get additional_info when fully initialized.
+    cdef PyDBAdditionalThreadInfo additional_info = thread_info.additional_info
+    if thread_info.is_pydevd_thread or additional_info.is_tracing:
+        # Make sure that we don't trace pydevd threads or inside our own calls.
+        return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+
+    # frame = <object> frame_obj
+    # DEBUG = frame.f_code.co_filename.endswith('_debugger_case_tracing.py')
+    # if DEBUG:
+    #     print('get_bytecode_while_frame_eval', frame.f_lineno, frame.f_code.co_name, frame.f_code.co_filename)
+
+    thread_info.inside_frame_eval += 1
+    additional_info.is_tracing = True
+    try:
+        main_debugger: object = GlobalDebuggerHolder.global_dbg
+        if main_debugger is None:
+            return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+        frame = <object> frame_obj
+
+        if thread_info.thread_trace_func is None:
+            trace_func, apply_to_global = fix_top_level_trace_and_get_trace_func(main_debugger, frame)
+            if apply_to_global:
+                thread_info.thread_trace_func = trace_func
+
+        if additional_info.pydev_step_cmd in (CMD_STEP_INTO, CMD_STEP_INTO_MY_CODE, CMD_STEP_INTO_COROUTINE, CMD_SMART_STEP_INTO) or \
+                main_debugger.break_on_caught_exceptions or \
+                main_debugger.break_on_user_uncaught_exceptions or \
+                main_debugger.has_plugin_exception_breaks or \
+                main_debugger.signature_factory or \
+                additional_info.pydev_step_cmd in (CMD_STEP_OVER, CMD_STEP_OVER_MY_CODE) and main_debugger.show_return_values and frame.f_back is additional_info.pydev_step_stop:
+
+            # if DEBUG:
+            #     print('get_bytecode_while_frame_eval enabled trace')
+            if thread_info.thread_trace_func is not None:
+                frame.f_trace = thread_info.thread_trace_func
+            else:
+                frame.f_trace = <object> main_debugger.trace_dispatch
+        else:
+            func_code_info: FuncCodeInfo = get_func_code_info(thread_info, frame_obj, frame_obj.f_code)
+            # if DEBUG:
+            #     print('get_bytecode_while_frame_eval always skip', func_code_info.always_skip_code)
+            if not func_code_info.always_skip_code:
+
+                if main_debugger.has_plugin_line_breaks or main_debugger.has_plugin_exception_breaks:
+                    can_skip = main_debugger.plugin.can_skip(main_debugger, <object> frame_obj)
+
+                    if not can_skip:
+                        # if DEBUG:
+                        #     print('get_bytecode_while_frame_eval not can_skip')
+                        if thread_info.thread_trace_func is not None:
+                            frame.f_trace = thread_info.thread_trace_func
+                        else:
+                            frame.f_trace = <object> main_debugger.trace_dispatch
+
+                if can_skip and func_code_info.breakpoint_found:
+                    # if DEBUG:
+                    #     print('get_bytecode_while_frame_eval new_code', func_code_info.new_code)
+                    if not thread_info.force_stay_in_untraced_mode:
+                        # If breakpoints are found but new_code is None,
+                        # this means we weren't able to actually add the code
+                        # where needed, so, fallback to tracing.
+                        if func_code_info.new_code is None:
+                            if thread_info.thread_trace_func is not None:
+                                frame.f_trace = thread_info.thread_trace_func
+                            else:
+                                frame.f_trace = <object> main_debugger.trace_dispatch
+                        else:
+                            # print('Using frame eval break for', <object> frame_obj.f_code.co_name)
+                            update_globals_dict(<object> frame_obj.f_globals)
+                            Py_INCREF(func_code_info.new_code)
+                            old = <object> frame_obj.f_code
+                            frame_obj.f_code = <PyCodeObject *> func_code_info.new_code
+                            Py_DECREF(old)
+                    else:
+                        # When we're forcing to stay in traced mode we need to
+                        # update the globals dict (because this means that we're reusing
+                        # a previous code which had breakpoints added in a new frame).
+                        update_globals_dict(<object> frame_obj.f_globals)
+
+    finally:
+        thread_info.inside_frame_eval -= 1
+        additional_info.is_tracing = False
+
+    return CALL_EvalFrameDefault_39(tstate, frame_obj, exc)
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+### WARNING: GENERATED CODE, DO NOT EDIT!
+
diff --git a/pandas/.devcontainer.json b/pandas/.devcontainer.json
new file mode 100644
index 0000000000000..54ddfa1a130f8
--- /dev/null
+++ b/pandas/.devcontainer.json
@@ -0,0 +1,28 @@
+// For format details, see https://aka.ms/vscode-remote/devcontainer.json or the definition README at
+// https://github.com/microsoft/vscode-dev-containers/tree/master/containers/python-3-miniconda
+{
+	"name": "pandas",
+	"context": ".",
+	"dockerFile": "Dockerfile",
+
+	// Use 'settings' to set *default* container specific settings.json values on container create.
+	// You can edit these settings after create using File > Preferences > Settings > Remote.
+	"settings": {
+		"python.pythonPath": "/usr/local/bin/python",
+		"python.formatting.provider": "black",
+		"python.linting.enabled": true,
+		"python.linting.flake8Enabled": true,
+		"python.linting.pylintEnabled": false,
+		"python.linting.mypyEnabled": true,
+		"python.testing.pytestEnabled": true,
+		"python.testing.pytestArgs": [
+			"pandas"
+		]
+	},
+
+	// Add the IDs of extensions you want installed when the container is created in the array below.
+	"extensions": [
+		"ms-python.python",
+		"ms-vscode.cpptools"
+	]
+}
diff --git a/pandas/.gitattributes b/pandas/.gitattributes
new file mode 100644
index 0000000000000..d94c19e7edb1f
--- /dev/null
+++ b/pandas/.gitattributes
@@ -0,0 +1,88 @@
+* text=auto
+# enforce text on certain files
+*.py text
+*.pyx text
+*.pyd text
+*.c text
+*.h text
+*.html text
+*.csv text
+*.json text
+*.pickle binary
+*.h5 binary
+*.dta binary
+*.xls binary
+*.xlsx binary
+pandas/_version.py export-subst
+
+
+*.bz2 export-ignore
+*.csv export-ignore
+*.data export-ignore
+*.dta export-ignore
+*.feather export-ignore
+*.tar export-ignore
+*.gz export-ignore
+*.h5 export-ignore
+*.html export-ignore
+*.json export-ignore
+*.jsonl export-ignore
+*.kml export-ignore
+*.msgpack export-ignore
+*.pdf export-ignore
+*.parquet export-ignore
+*.pickle export-ignore
+*.pkl export-ignore
+*.png export-ignore
+*.pptx export-ignore
+*.ods export-ignore
+*.odt export-ignore
+*.orc export-ignore
+*.sas7bdat export-ignore
+*.sav export-ignore
+*.so export-ignore
+*.txt export-ignore
+*.xls export-ignore
+*.xlsb export-ignore
+*.xlsm export-ignore
+*.xlsx export-ignore
+*.xpt export-ignore
+*.cpt export-ignore
+*.xml export-ignore
+*.xsl export-ignore
+*.xz export-ignore
+*.zip export-ignore
+*.zst export-ignore
+*~ export-ignore
+.DS_Store export-ignore
+.git* export-ignore
+
+*.py[ocd] export-ignore
+*.pxi export-ignore
+
+# Ignoring stuff from the top level
+.github export-ignore
+asv_bench export-ignore
+ci export-ignore
+doc export-ignore
+gitpod export-ignore
+MANIFEST.in export-ignore
+scripts/** export-ignore
+typings export-ignore
+web export-ignore
+CITATION.cff export-ignore
+codecov.yml export-ignore
+Dockerfile export-ignore
+environment.yml export-ignore
+setup.py export-ignore
+
+
+# GH 39321
+# csv_dir_path fixture checks the existence of the directory
+# exclude the whole directory to avoid running related tests in sdist
+pandas/tests/io/parser/data export-ignore
+
+# Include cibw script in sdist since it's needed for building wheels
+scripts/cibw_before_build.sh -export-ignore
+scripts/cibw_before_build_windows.sh -export-ignore
+scripts/cibw_before_test_windows.sh -export-ignore
diff --git a/pandas/.github/CODEOWNERS b/pandas/.github/CODEOWNERS
new file mode 100644
index 0000000000000..39f52bb3edd8e
--- /dev/null
+++ b/pandas/.github/CODEOWNERS
@@ -0,0 +1,20 @@
+# github
+.github/                          @mroeschke
+
+# ci
+ci/                               @mroeschke
+
+# docs
+doc/cheatsheet                    @Dr-Irv
+doc/source/development            @noatamir
+
+# pandas
+pandas/_libs/                     @WillAyd
+pandas/_libs/tslibs/*             @MarcoGorelli
+pandas/_typing.py                 @Dr-Irv
+pandas/core/groupby/*             @rhshadrach
+pandas/core/tools/datetimes.py    @MarcoGorelli
+pandas/io/excel/*                 @rhshadrach
+pandas/io/formats/style.py        @attack68
+pandas/io/formats/style_render.py @attack68
+pandas/io/formats/templates       @attack68
diff --git a/pandas/.github/ISSUE_TEMPLATE/bug_report.yaml b/pandas/.github/ISSUE_TEMPLATE/bug_report.yaml
new file mode 100644
index 0000000000000..4e1bc8f61d04e
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/bug_report.yaml
@@ -0,0 +1,70 @@
+name: Bug Report
+description: Report incorrect behavior in the pandas library
+title: "BUG: "
+labels: [Bug, Needs Triage]
+
+body:
+  - type: checkboxes
+    id: checks
+    attributes:
+      label: Pandas version checks
+      options:
+        - label: >
+            I have checked that this issue has not already been reported.
+          required: true
+        - label: >
+            I have confirmed this bug exists on the
+            [latest version](https://pandas.pydata.org/docs/whatsnew/index.html) of pandas.
+          required: true
+        - label: >
+            I have confirmed this bug exists on the
+            [main branch](https://pandas.pydata.org/docs/dev/getting_started/install.html#installing-the-development-version-of-pandas)
+            of pandas.
+  - type: textarea
+    id: example
+    attributes:
+      label: Reproducible Example
+      description: >
+        Please follow [this guide](https://matthewrocklin.com/blog/work/2018/02/28/minimal-bug-reports) on how to
+        provide a minimal, copy-pastable example.
+      placeholder: >
+        import pandas as pd
+
+        df = pd.DataFrame(range(5))
+
+        ...
+      render: python
+    validations:
+      required: true
+  - type: textarea
+    id: problem
+    attributes:
+      label: Issue Description
+      description: >
+        Please provide a description of the issue shown in the reproducible example.
+    validations:
+      required: true
+  - type: textarea
+    id: expected-behavior
+    attributes:
+      label: Expected Behavior
+      description: >
+        Please describe or show a code example of the expected behavior.
+    validations:
+      required: true
+  - type: textarea
+    id: version
+    attributes:
+      label: Installed Versions
+      description: >
+        Please paste the output of ``pd.show_versions()``
+      value: >
+        <details>
+
+
+        Replace this line with the output of pd.show_versions()
+
+
+        </details>
+    validations:
+      required: true
diff --git a/pandas/.github/ISSUE_TEMPLATE/documentation_improvement.yaml b/pandas/.github/ISSUE_TEMPLATE/documentation_improvement.yaml
new file mode 100644
index 0000000000000..b89600f8598e7
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/documentation_improvement.yaml
@@ -0,0 +1,41 @@
+name: Documentation Improvement
+description: Report wrong or missing documentation
+title: "DOC: "
+labels: [Docs, Needs Triage]
+
+body:
+  - type: checkboxes
+    attributes:
+      label: Pandas version checks
+      options:
+        - label: >
+            I have checked that the issue still exists on the latest versions of the docs
+            on `main` [here](https://pandas.pydata.org/docs/dev/)
+          required: true
+  - type: textarea
+    id: location
+    attributes:
+      label: Location of the documentation
+      description: >
+        Please provide the location of the documentation, e.g. "pandas.read_csv" or the
+        URL of the documentation, e.g.
+        "https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html"
+      placeholder: https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html
+    validations:
+      required: true
+  - type: textarea
+    id: problem
+    attributes:
+      label: Documentation problem
+      description: >
+        Please provide a description of what documentation you believe needs to be fixed/improved
+    validations:
+      required: true
+  - type: textarea
+    id: suggested-fix
+    attributes:
+      label: Suggested fix for documentation
+      description: >
+        Please explain the suggested fix and **why** it's better than the existing documentation
+    validations:
+      required: true
diff --git a/pandas/.github/ISSUE_TEMPLATE/feature_request.yaml b/pandas/.github/ISSUE_TEMPLATE/feature_request.yaml
new file mode 100644
index 0000000000000..6e6cd78ace11d
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/feature_request.yaml
@@ -0,0 +1,72 @@
+name: Feature Request
+description: Suggest an idea for pandas
+title: "ENH: "
+labels: [Enhancement, Needs Triage]
+
+body:
+  - type: checkboxes
+    id: checks
+    attributes:
+      label: Feature Type
+      description: Please check what type of feature request you would like to propose.
+      options:
+        - label: >
+            Adding new functionality to pandas
+        - label: >
+            Changing existing functionality in pandas
+        - label: >
+            Removing existing functionality in pandas
+  - type: textarea
+    id: description
+    attributes:
+      label: Problem Description
+      description: >
+        Please describe what problem the feature would solve, e.g. "I wish I could use pandas to ..."
+      placeholder: >
+        I wish I could use pandas to return a Series from a DataFrame when possible.
+    validations:
+      required: true
+  - type: textarea
+    id: feature
+    attributes:
+      label: Feature Description
+      description: >
+        Please describe how the new feature would be implemented, using psudocode if relevant.
+      placeholder: >
+        Add a new parameter to DataFrame, to_series, to return a Series if possible.
+
+        def __init__(self, ..., to_series: bool=False):
+            """
+            Parameters
+            ----------
+            ...
+
+            to_series : bool, default False
+                Return a Series if possible
+            """
+            if to_series:
+                return Series(data)
+    validations:
+      required: true
+  - type: textarea
+    id: alternative
+    attributes:
+      label: Alternative Solutions
+      description: >
+        Please describe any alternative solution (existing functionality, 3rd party package, etc.)
+        that would satisfy the feature request.
+      placeholder: >
+        Write a custom function to return a Series when possible.
+
+        def to_series(...)
+           result = pd.DataFrame(...)
+           ...
+    validations:
+      required: true
+  - type: textarea
+    id: context
+    attributes:
+      label: Additional Context
+      description: >
+        Please provide any relevant GitHub issues, code examples or references that help describe and support
+        the feature request.
diff --git a/pandas/.github/ISSUE_TEMPLATE/installation_issue.yaml b/pandas/.github/ISSUE_TEMPLATE/installation_issue.yaml
new file mode 100644
index 0000000000000..a80269ff0f12d
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/installation_issue.yaml
@@ -0,0 +1,66 @@
+name: Installation Issue
+description: Report issues installing the pandas library on your system
+title: "BUILD: "
+labels: [Build, Needs Triage]
+
+body:
+  - type: checkboxes
+    id: checks
+    attributes:
+      label: Installation check
+      options:
+        - label: >
+            I have read the [installation guide](https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html#installing-pandas).
+          required: true
+  - type: input
+    id: platform
+    attributes:
+      label: Platform
+      description: >
+        Please provide the output of ``import platform; print(platform.platform())``
+    validations:
+      required: true
+  - type: dropdown
+    id: method
+    attributes:
+      label: Installation Method
+      description: >
+        Please provide how you tried to install pandas from a clean environment.
+      options:
+        - pip install
+        - conda install
+        - apt-get install
+        - Built from source
+        - Other
+    validations:
+      required: true
+  - type: input
+    id: pandas
+    attributes:
+      label: pandas Version
+      description: >
+        Please provide the version of pandas you are trying to install.
+    validations:
+      required: true
+  - type: input
+    id: python
+    attributes:
+      label: Python Version
+      description: >
+        Please provide the installed version of Python.
+    validations:
+      required: true
+  - type: textarea
+    id: logs
+    attributes:
+      label: Installation Logs
+      description: >
+        If possible, please copy and paste the installation logs when attempting to install pandas.
+      value: >
+        <details>
+
+
+        Replace this line with the installation logs.
+
+
+        </details>
diff --git a/pandas/.github/ISSUE_TEMPLATE/pdep_vote.yaml b/pandas/.github/ISSUE_TEMPLATE/pdep_vote.yaml
new file mode 100644
index 0000000000000..6dcbd76eb0f74
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/pdep_vote.yaml
@@ -0,0 +1,74 @@
+name: PDEP Vote
+description: Call for a vote on a PDEP
+title: "VOTE: "
+labels: [Vote]
+
+body:
+  - type: markdown
+    attributes:
+      value: >
+        As per [PDEP-1](https://pandas.pydata.org/pdeps/0001-purpose-and-guidelines.html), the following issue template should be used when a
+        maintainer has opened a PDEP discussion and is ready to call for a vote.
+  - type: checkboxes
+    attributes:
+      label: Locked issue
+      options:
+        - label: >
+            I locked this voting issue so that only voting members are able to cast their votes or
+            comment on this issue.
+          required: true
+  - type: input
+    id: PDEP-name
+    attributes:
+      label: PDEP number and title
+      placeholder: >
+        PDEP-1: Purpose and guidelines
+    validations:
+      required: true
+  - type: input
+    id: PDEP-link
+    attributes:
+      label: Pull request with discussion
+      description: e.g. https://github.com/pandas-dev/pandas/pull/47444
+    validations:
+      required: true
+  - type: input
+    id: PDEP-rendered-link
+    attributes:
+      label: Rendered PDEP for easy reading
+      description: e.g. https://github.com/pandas-dev/pandas/pull/47444/files?short_path=7c449e6#diff-7c449e698132205b235c501f7e47ebba38da4d2b7f9492c98f16745dba787041
+    validations:
+      required: true
+  - type: input
+    id: PDEP-number-of-discussion-participants
+    attributes:
+      label: Discussion participants
+      description: >
+        You may find it useful to list or total the number of participating members in the
+        PDEP discussion PR. This would be the maximum possible disapprove votes.
+      placeholder: >
+        14 voting members participated in the PR discussion thus far.
+  - type: input
+    id: PDEP-vote-end
+    attributes:
+      label: Voting will close in 15 days.
+      description: The voting period end date. ('Voting will close in 15 days.' will be automatically written)
+  - type: markdown
+    attributes:
+      value: ---
+  - type: textarea
+    id: Vote
+    attributes:
+      label: Vote
+      value: |
+        Cast your vote in a comment below.
+        * +1: approve.
+        * 0: abstain.
+            * Reason: A one sentence reason is required.
+        * -1: disapprove
+            * Reason: A one sentence reason is required.
+        A disapprove vote requires prior participation in the linked discussion PR.
+
+        @pandas-dev/pandas-core
+    validations:
+      required: true
diff --git a/pandas/.github/ISSUE_TEMPLATE/performance_issue.yaml b/pandas/.github/ISSUE_TEMPLATE/performance_issue.yaml
new file mode 100644
index 0000000000000..096e012f4ee0f
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/performance_issue.yaml
@@ -0,0 +1,53 @@
+name: Performance Issue
+description: Report slow performance or memory issues when running pandas code
+title: "PERF: "
+labels: [Performance, Needs Triage]
+
+body:
+  - type: checkboxes
+    id: checks
+    attributes:
+      label: Pandas version checks
+      options:
+        - label: >
+            I have checked that this issue has not already been reported.
+          required: true
+        - label: >
+            I have confirmed this issue exists on the
+            [latest version](https://pandas.pydata.org/docs/whatsnew/index.html) of pandas.
+          required: true
+        - label: >
+            I have confirmed this issue exists on the main branch of pandas.
+  - type: textarea
+    id: example
+    attributes:
+      label: Reproducible Example
+      description: >
+        Please provide a minimal, copy-pastable example that quantifies
+        [slow runtime](https://docs.python.org/3/library/timeit.html) or
+        [memory](https://pypi.org/project/memory-profiler/) issues.
+    validations:
+      required: true
+  - type: textarea
+    id: version
+    attributes:
+      label: Installed Versions
+      description: >
+        Please paste the output of ``pd.show_versions()``
+      value: >
+        <details>
+
+
+        Replace this line with the output of pd.show_versions()
+
+
+        </details>
+    validations:
+      required: true
+  - type: textarea
+    id: prior-performance
+    attributes:
+      label: Prior Performance
+      description: >
+        If applicable, please provide the prior version of pandas and output
+        of the same reproducible example where the performance issue did not exist.
diff --git a/pandas/.github/ISSUE_TEMPLATE/submit_question.yml b/pandas/.github/ISSUE_TEMPLATE/submit_question.yml
new file mode 100644
index 0000000000000..6f73041b0f527
--- /dev/null
+++ b/pandas/.github/ISSUE_TEMPLATE/submit_question.yml
@@ -0,0 +1,44 @@
+name: Submit Question
+description: Ask a general question about pandas
+title: "QST: "
+labels: [Usage Question, Needs Triage]
+
+body:
+  - type: markdown
+    attributes:
+      value: >
+        Since [StackOverflow](https://stackoverflow.com) is better suited towards answering
+        usage questions, we ask that all usage questions are first asked on StackOverflow.
+  - type: checkboxes
+    attributes:
+      label: Research
+      options:
+        - label: >
+            I have searched the [[pandas] tag](https://stackoverflow.com/questions/tagged/pandas)
+            on StackOverflow for similar questions.
+          required: true
+        - label: >
+            I have asked my usage related question on [StackOverflow](https://stackoverflow.com).
+          required: true
+  - type: input
+    id: question-link
+    attributes:
+      label: Link to question on StackOverflow
+    validations:
+      required: true
+  - type: markdown
+    attributes:
+      value: ---
+  - type: textarea
+    id: question
+    attributes:
+      label: Question about pandas
+      description: >
+        **Note**: If you'd still like to submit a question, please read [this guide](
+        https://matthewrocklin.com/blog/work/2018/02/28/minimal-bug-reports) detailing
+        how to provide the necessary information for us to reproduce your question.
+      placeholder: |
+        ```python
+        # Your code here, if applicable
+
+        ```
diff --git a/pandas/.github/PULL_REQUEST_TEMPLATE.md b/pandas/.github/PULL_REQUEST_TEMPLATE.md
new file mode 100644
index 0000000000000..8eca91c692710
--- /dev/null
+++ b/pandas/.github/PULL_REQUEST_TEMPLATE.md
@@ -0,0 +1,5 @@
+- [ ] closes #xxxx (Replace xxxx with the GitHub issue number)
+- [ ] [Tests added and passed](https://pandas.pydata.org/pandas-docs/dev/development/contributing_codebase.html#writing-tests) if fixing a bug or adding a new feature
+- [ ] All [code checks passed](https://pandas.pydata.org/pandas-docs/dev/development/contributing_codebase.html#pre-commit).
+- [ ] Added [type annotations](https://pandas.pydata.org/pandas-docs/dev/development/contributing_codebase.html#type-hints) to new arguments/methods/functions.
+- [ ] Added an entry in the latest `doc/source/whatsnew/vX.X.X.rst` file if fixing a bug or adding a new feature.
diff --git a/pandas/.github/actions/build_pandas/action.yml b/pandas/.github/actions/build_pandas/action.yml
new file mode 100644
index 0000000000000..b92bacd1a537c
--- /dev/null
+++ b/pandas/.github/actions/build_pandas/action.yml
@@ -0,0 +1,34 @@
+name: Build pandas
+description: Rebuilds the C extensions and installs pandas
+inputs:
+  editable:
+    description: Whether to build pandas in editable mode (default true)
+    default: true
+runs:
+  using: composite
+  steps:
+
+    - name: Environment Detail
+      run: |
+        micromamba info
+        micromamba list
+        pip list --pre
+      shell: bash -el {0}
+
+    - name: Uninstall existing Pandas installation
+      run: |
+        if pip show pandas 1>/dev/null; then
+          pip uninstall -y pandas
+        fi
+      shell: bash -el {0}
+
+    - name: Build Pandas
+      run: |
+        if [[ ${{ inputs.editable }} == "true" ]]; then
+          pip install -e . --no-build-isolation -v --no-deps \
+            -Csetup-args="--werror"
+        else
+          pip install . --no-build-isolation -v --no-deps \
+            -Csetup-args="--werror"
+        fi
+      shell: bash -el {0}
diff --git a/pandas/.github/actions/run-tests/action.yml b/pandas/.github/actions/run-tests/action.yml
new file mode 100644
index 0000000000000..f5d6abdf0f186
--- /dev/null
+++ b/pandas/.github/actions/run-tests/action.yml
@@ -0,0 +1,22 @@
+name: Run tests and report results
+runs:
+  using: composite
+  steps:
+    - name: Test
+      run: ci/run_tests.sh
+      shell: bash -el {0}
+
+    - name: Publish test results
+      uses: actions/upload-artifact@v4
+      with:
+        name: Test results
+        path: test-data.xml
+      if: failure()
+
+    - name: Upload coverage to Codecov
+      uses: codecov/codecov-action@v4
+      with:
+        flags: unittests
+        name: codecov-pandas
+        fail_ci_if_error: false
+      if: failure()
diff --git a/pandas/.github/actions/setup-conda/action.yml b/pandas/.github/actions/setup-conda/action.yml
new file mode 100644
index 0000000000000..3eb68bdd2a15c
--- /dev/null
+++ b/pandas/.github/actions/setup-conda/action.yml
@@ -0,0 +1,22 @@
+name: Set up Conda environment
+inputs:
+  environment-file:
+    description: Conda environment file to use.
+    default: environment.yml
+runs:
+  using: composite
+  steps:
+    - name: Install ${{ inputs.environment-file }}
+      uses: mamba-org/setup-micromamba@v1
+      with:
+        environment-file: ${{ inputs.environment-file }}
+        environment-name: test
+        condarc-file: ci/.condarc
+        cache-environment: true
+        cache-downloads: true
+
+    - name: Uninstall pyarrow
+      if: ${{ env.REMOVE_PYARROW == '1' }}
+      run: |
+        micromamba remove -y pyarrow
+      shell: bash -el {0}
diff --git a/pandas/.github/dependabot.yml b/pandas/.github/dependabot.yml
new file mode 100644
index 0000000000000..784206dfe67ff
--- /dev/null
+++ b/pandas/.github/dependabot.yml
@@ -0,0 +1,9 @@
+version: 2
+updates:
+  - package-ecosystem: github-actions
+    directory: /
+    schedule:
+      interval: weekly
+    labels:
+      - "CI"
+      - "Dependencies"
diff --git a/pandas/.github/workflows/broken-linkcheck.yml b/pandas/.github/workflows/broken-linkcheck.yml
new file mode 100644
index 0000000000000..191252cccf1c3
--- /dev/null
+++ b/pandas/.github/workflows/broken-linkcheck.yml
@@ -0,0 +1,39 @@
+name: Linkcheck
+on:
+  schedule:
+  # Run monthly on the 1st day of the month
+    - cron: '0 0 1 * *'
+  pull_request:
+    paths:
+      - ".github/workflows/broken-linkcheck.yml"
+      - "doc/make.py"
+jobs:
+  linkcheck:
+    if: false
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Conda
+        uses: ./.github/actions/setup-conda
+
+      - name: Build Pandas
+        uses: ./.github/actions/build_pandas
+
+      - name: Run linkcheck script
+        working-directory: ./doc
+        run: |
+          set -o pipefail
+          python make.py linkcheck | tee linkcheck.txt
+
+      - name: Display broken links
+        if: failure()
+        working-directory: ./doc
+        run: grep broken linkcheck.txt
diff --git a/pandas/.github/workflows/cache-cleanup-daily.yml b/pandas/.github/workflows/cache-cleanup-daily.yml
new file mode 100644
index 0000000000000..8eadfb2ccd2a9
--- /dev/null
+++ b/pandas/.github/workflows/cache-cleanup-daily.yml
@@ -0,0 +1,32 @@
+name: Purge caches daily
+on:
+  schedule:
+    # 4:10 UTC daily
+    - cron: "10 4 * * *"
+
+jobs:
+  cleanup:
+    runs-on: ubuntu-latest
+    if: github.repository_owner == 'pandas-dev'
+    permissions:
+      actions: write
+    steps:
+      - name: Clean Cache
+        run: |
+          gh extension install actions/gh-actions-cache
+
+          REPO=${{ github.repository }}
+
+          echo "Fetching list of cache key"
+          allCaches=$(gh actions-cache list -L 100 -R $REPO | cut -f 1 )
+
+          ## Setting this to not fail the workflow while deleting cache keys.
+          set +e
+          echo "Deleting caches..."
+          for cacheKey in $allCaches
+          do
+              gh actions-cache delete $cacheKey -R $REPO --confirm
+          done
+          echo "Done"
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
diff --git a/pandas/.github/workflows/cache-cleanup.yml b/pandas/.github/workflows/cache-cleanup.yml
new file mode 100644
index 0000000000000..099974141c1d1
--- /dev/null
+++ b/pandas/.github/workflows/cache-cleanup.yml
@@ -0,0 +1,30 @@
+name: Clean closed branch caches
+on:
+  pull_request:
+    types:
+      - closed
+
+jobs:
+  cleanup:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clean Cache
+        run: |
+          gh extension install actions/gh-actions-cache
+
+          REPO=${{ github.repository }}
+          BRANCH="refs/pull/${{ github.event.pull_request.number }}/merge"
+
+          echo "Fetching list of cache key"
+          cacheKeysForPR=$(gh actions-cache list -R $REPO -B $BRANCH | cut -f 1 )
+
+          ## Setting this to not fail the workflow while deleting cache keys.
+          set +e
+          echo "Deleting caches..."
+          for cacheKey in $cacheKeysForPR
+          do
+              gh actions-cache delete $cacheKey -R $REPO -B $BRANCH --confirm
+          done
+          echo "Done"
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
diff --git a/pandas/.github/workflows/code-checks.yml b/pandas/.github/workflows/code-checks.yml
new file mode 100644
index 0000000000000..e1d2d1ea846b8
--- /dev/null
+++ b/pandas/.github/workflows/code-checks.yml
@@ -0,0 +1,188 @@
+name: Code Checks
+
+on:
+  push:
+    branches:
+      - main
+      - 2.3.x
+  pull_request:
+    branches:
+      - main
+      - 2.3.x
+
+env:
+  ENV_FILE: environment.yml
+  PANDAS_CI: 1
+
+permissions:
+  contents: read
+
+# pre-commit run by https://pre-commit.ci/
+jobs:
+  docstring_typing_manual_hooks:
+    name: Docstring validation, typing, and other manual pre-commit hooks
+    runs-on: ubuntu-22.04
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-code-checks
+      cancel-in-progress: true
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+
+    - name: Set up Conda
+      uses: ./.github/actions/setup-conda
+
+    - name: Build Pandas
+      id: build
+      uses: ./.github/actions/build_pandas
+      with:
+        editable: false
+
+    # The following checks are independent of each other and should still be run if one fails
+
+    # TODO: The doctests have to be run first right now, since the Cython doctests only work
+    # with pandas installed in non-editable mode
+    # This can be removed once pytest-cython doesn't require C extensions to be installed inplace
+
+    - name: Extra installs
+      # https://pytest-qt.readthedocs.io/en/latest/troubleshooting.html#github-actions-azure-pipelines-travis-ci-and-gitlab-ci-cd
+      run: sudo apt-get update && sudo apt-get install -y libegl1 libopengl0
+
+    - name: Run doctests
+      run: cd ci && ./code_checks.sh doctests
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Install pandas in editable mode
+      id: build-editable
+      if: ${{ steps.build.outcome == 'success' && always() }}
+      uses: ./.github/actions/build_pandas
+      with:
+        editable: true
+
+    - name: Check for no warnings when building single-page docs
+      run: ci/code_checks.sh single-docs
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Run checks on imported code
+      run: ci/code_checks.sh code
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Run docstring validation
+      run: ci/code_checks.sh docstrings
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Run check of documentation notebooks
+      run: ci/code_checks.sh notebooks
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Use existing environment for type checking
+      run: |
+        echo $PATH >> $GITHUB_PATH
+        echo "PYTHONHOME=$PYTHONHOME" >> $GITHUB_ENV
+        echo "PYTHONPATH=$PYTHONPATH" >> $GITHUB_ENV
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Typing
+      uses: pre-commit/action@v3.0.1
+      with:
+        extra_args: --verbose --hook-stage manual --all-files
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+    - name: Run docstring validation script tests
+      run: pytest scripts
+      if: ${{ steps.build.outcome == 'success' && always() }}
+
+  asv-benchmarks:
+    name: ASV Benchmarks
+    runs-on: ubuntu-22.04
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-asv-benchmarks
+      cancel-in-progress: true
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+
+    - name: Set up Conda
+      uses: ./.github/actions/setup-conda
+
+    - name: Build Pandas
+      id: build
+      uses: ./.github/actions/build_pandas
+
+    - name: Run ASV benchmarks
+      run: |
+        cd asv_bench
+        asv machine --yes
+        asv run --quick --dry-run --durations=30 --python=same --show-stderr
+
+  build_docker_dev_environment:
+    name: Build Docker Dev Environment
+    runs-on: ubuntu-22.04
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-build_docker_dev_environment
+      cancel-in-progress: true
+
+    steps:
+      - name: Clean up dangling images
+        run: docker image prune -f
+
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Build image
+        run: docker build --pull --no-cache --tag pandas-dev-env .
+
+      - name: Show environment
+        run: docker run --rm pandas-dev-env python -c "import pandas as pd; print(pd.show_versions())"
+
+  requirements-dev-text-installable:
+    name: Test install requirements-dev.txt
+    runs-on: ubuntu-22.04
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-requirements-dev-text-installable
+      cancel-in-progress: true
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Setup Python
+        id: setup_python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+          cache: 'pip'
+          cache-dependency-path: 'requirements-dev.txt'
+
+      - name: Install requirements-dev.txt
+        run: pip install -r requirements-dev.txt
+
+      - name: Check Pip Cache Hit
+        run: echo ${{ steps.setup_python.outputs.cache-hit }}
diff --git a/pandas/.github/workflows/codeql.yml b/pandas/.github/workflows/codeql.yml
new file mode 100644
index 0000000000000..4d0066bc0b48d
--- /dev/null
+++ b/pandas/.github/workflows/codeql.yml
@@ -0,0 +1,35 @@
+name: CodeQL
+on:
+  schedule:
+    # every day at midnight
+    - cron: "0 0 * * *"
+
+concurrency:
+  group: ${{ github.repository }}-${{ github.head_ref || github.sha }}-${{ github.workflow }}
+  cancel-in-progress: true
+
+permissions:
+  contents: read
+
+jobs:
+  analyze:
+    runs-on: ubuntu-22.04
+    permissions:
+      actions: read
+      contents: read
+      security-events: write
+    if: github.repository_owner == 'pandas-dev'
+
+    strategy:
+      fail-fast: false
+      matrix:
+        language:
+          - python
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: github/codeql-action/init@v3
+        with:
+          languages: ${{ matrix.language }}
+      - uses: github/codeql-action/autobuild@v3
+      - uses: github/codeql-action/analyze@v3
diff --git a/pandas/.github/workflows/comment-commands.yml b/pandas/.github/workflows/comment-commands.yml
new file mode 100644
index 0000000000000..62956f5825782
--- /dev/null
+++ b/pandas/.github/workflows/comment-commands.yml
@@ -0,0 +1,91 @@
+name: Comment Commands
+on:
+  issue_comment:
+    types: created
+
+permissions:
+  contents: read
+  issues: write
+  pull-requests: write
+
+jobs:
+  issue_assign:
+    runs-on: ubuntu-22.04
+    if: (!github.event.issue.pull_request) && github.event.comment.body == 'take'
+    concurrency:
+      group: ${{ github.actor }}-issue-assign
+    steps:
+      - run: |
+          echo "Assigning issue ${{ github.event.issue.number }} to ${{ github.event.comment.user.login }}"
+          curl -H "Authorization: token ${{ secrets.GITHUB_TOKEN }}" -d '{"assignees": ["${{ github.event.comment.user.login }}"]}' https://api.github.com/repos/${{ github.repository }}/issues/${{ github.event.issue.number }}/assignees
+  preview_docs:
+    runs-on: ubuntu-22.04
+    if: github.event.issue.pull_request && github.event.comment.body == '/preview'
+    concurrency:
+      group: ${{ github.actor }}-preview-docs
+    steps:
+      - uses: pandas-dev/github-doc-previewer@v0.3.2
+        with:
+          previewer-server: "https://pandas.pydata.org/preview"
+          artifact-job: "Doc Build and Upload"
+  asv_run:
+    runs-on: ubuntu-22.04
+    # TODO: Support more benchmarking options later, against different branches, against self, etc
+    if: github.event.issue.pull_request && startsWith(github.event.comment.body, '@github-actions benchmark')
+    defaults:
+      run:
+        shell: bash -el {0}
+    env:
+      ENV_FILE: environment.yml
+      COMMENT: ${{github.event.comment.body}}
+
+    concurrency:
+      # Set concurrency to prevent abuse(full runs are ~5.5 hours !!!)
+      # each user can only run one concurrent benchmark bot at a time
+      # We don't cancel in progress jobs, but if you want to benchmark multiple PRs, you're gonna have
+      # to wait
+      group: ${{ github.actor }}-asv
+      cancel-in-progress: false
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+        # Although asv sets up its own env, deps are still needed
+        # during discovery process
+      - name: Set up Conda
+        uses: ./.github/actions/setup-conda
+
+      - name: Run benchmarks
+        id: bench
+        continue-on-error: true # asv will exit code 1 for regressions
+        run: |
+          # extracting the regex, see https://stackoverflow.com/a/36798723
+          REGEX=$(echo "$COMMENT" | sed -n "s/^.*-b\s*\(\S*\).*$/\1/p")
+          cd asv_bench
+          asv check -E existing
+          git remote add upstream https://github.com/pandas-dev/pandas.git
+          git fetch upstream
+          asv machine --yes
+          asv continuous -f 1.1 -b $REGEX upstream/main HEAD
+          echo 'BENCH_OUTPUT<<EOF' >> $GITHUB_ENV
+          asv compare -f 1.1 upstream/main HEAD >> $GITHUB_ENV
+          echo 'EOF' >> $GITHUB_ENV
+          echo "REGEX=$REGEX" >> $GITHUB_ENV
+
+      - uses: actions/github-script@v7
+        env:
+          BENCH_OUTPUT: ${{env.BENCH_OUTPUT}}
+          REGEX: ${{env.REGEX}}
+        with:
+          script: |
+            const ENV_VARS = process.env
+            const run_url = `https://github.com/${context.repo.owner}/${context.repo.repo}/actions/runs/${context.runId}`
+            github.rest.issues.createComment({
+              issue_number: context.issue.number,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body: '\nBenchmarks completed. View runner logs here.' + run_url + '\nRegex used: '+ 'regex ' + ENV_VARS["REGEX"] + '\n' + ENV_VARS["BENCH_OUTPUT"]
+            })
diff --git a/pandas/.github/workflows/deprecation-tracking-bot.yml b/pandas/.github/workflows/deprecation-tracking-bot.yml
new file mode 100644
index 0000000000000..3d4cab7be09c5
--- /dev/null
+++ b/pandas/.github/workflows/deprecation-tracking-bot.yml
@@ -0,0 +1,65 @@
+# This bot updates the issue with number DEPRECATION_TRACKER_ISSUE
+# with the PR number that issued the deprecation.
+
+# It runs on commits to main, and will trigger if the PR linked to a merged commit has the "Deprecate" label
+name: Deprecations Bot
+
+on:
+  push:
+    branches:
+      - main
+
+
+permissions:
+  contents: read
+
+jobs:
+  deprecation_update:
+    permissions:
+      issues: write
+    runs-on: ubuntu-22.04
+    env:
+      DEPRECATION_TRACKER_ISSUE: 56596
+    steps:
+    - uses: actions/github-script@v7
+      id: update-deprecation-issue
+      with:
+        script: |
+          body = await github.rest.issues.get({
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            issue_number: ${{ env.DEPRECATION_TRACKER_ISSUE }},
+          })
+          body = body["data"]["body"];
+          linkedPRs = await github.rest.repos.listPullRequestsAssociatedWithCommit({
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            commit_sha: '${{ github.sha }}'
+          })
+          linkedPRs = linkedPRs["data"];
+          console.log(linkedPRs);
+          if (linkedPRs.length > 0) {
+            console.log("Found linked PR");
+            linkedPR = linkedPRs[0]
+            isDeprecation = false
+            for (label of linkedPR["labels"]) {
+              if (label["name"] == "Deprecate") {
+                isDeprecation = true;
+                break;
+              }
+            }
+
+            PR_NUMBER = linkedPR["number"];
+
+            body += ("\n- [ ] #" + PR_NUMBER);
+            if (isDeprecation) {
+              console.log("PR is a deprecation PR. Printing new body of issue");
+              console.log(body);
+              github.rest.issues.update({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: ${{ env.DEPRECATION_TRACKER_ISSUE }},
+                body: body
+              })
+            }
+          }
diff --git a/pandas/.github/workflows/docbuild-and-upload.yml b/pandas/.github/workflows/docbuild-and-upload.yml
new file mode 100644
index 0000000000000..908baa87815ab
--- /dev/null
+++ b/pandas/.github/workflows/docbuild-and-upload.yml
@@ -0,0 +1,96 @@
+name: Doc Build and Upload
+
+on:
+  push:
+    branches:
+      - main
+      - 2.3.x
+    tags:
+      - '*'
+  pull_request:
+    branches:
+      - main
+      - 2.3.x
+
+env:
+  ENV_FILE: environment.yml
+  PANDAS_CI: 1
+  GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+permissions:
+  contents: read
+
+jobs:
+  web_and_docs:
+    name: Doc Build and Upload
+    runs-on: ubuntu-22.04
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-web-docs
+      cancel-in-progress: true
+
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+
+    - name: Set up Conda
+      uses: ./.github/actions/setup-conda
+
+    - name: Build Pandas
+      uses: ./.github/actions/build_pandas
+
+    - name: Extra installs
+      # https://pytest-qt.readthedocs.io/en/latest/troubleshooting.html#github-actions-azure-pipelines-travis-ci-and-gitlab-ci-cd
+      run: sudo apt-get update && sudo apt-get install -y libegl1 libopengl0
+
+    - name: Test website
+      run: python -m pytest web/
+
+    - name: Build website
+      run: python web/pandas_web.py web/pandas --target-path=web/build
+
+    - name: Build documentation
+      run: doc/make.py --warnings-are-errors
+
+    - name: Build documentation zip
+      run: doc/make.py zip_html
+
+    - name: Install ssh key
+      run: |
+        mkdir -m 700 -p ~/.ssh
+        echo "${{ secrets.server_ssh_key }}" > ~/.ssh/id_rsa
+        chmod 600 ~/.ssh/id_rsa
+        echo "${{ secrets.server_ip }} ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBFjYkJBk7sos+r7yATODogQc3jUdW1aascGpyOD4bohj8dWjzwLJv/OJ/fyOQ5lmj81WKDk67tGtqNJYGL9acII=" > ~/.ssh/known_hosts
+      if: github.event_name == 'push' && (github.ref == 'refs/heads/main' || startsWith(github.ref, 'refs/tags/'))
+
+    - name: Copy cheatsheets into site directory
+      run: cp doc/cheatsheet/Pandas_Cheat_Sheet* web/build/
+
+    - name: Upload web
+      run: rsync -az --delete --exclude='pandas-docs' --exclude='docs' --exclude='benchmarks' web/build/ web@${{ secrets.server_ip }}:/var/www/html
+      if: github.event_name == 'push' && github.ref == 'refs/heads/main'
+
+    - name: Upload dev docs
+      run: rsync -az --delete doc/build/html/ web@${{ secrets.server_ip }}:/var/www/html/pandas-docs/dev
+      if: github.event_name == 'push' && github.ref == 'refs/heads/main'
+
+    - name: Upload prod docs
+      run: rsync -az --delete doc/build/html/ web@${{ secrets.server_ip }}:/var/www/html/pandas-docs/version/${GITHUB_REF_NAME:1}
+      if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/')
+
+    - name: Move docs into site directory
+      run: mv doc/build/html web/build/docs
+
+    - name: Save website as an artifact
+      uses: actions/upload-artifact@v4
+      with:
+        name: website
+        path: web/build
+        retention-days: 14
diff --git a/pandas/.github/workflows/package-checks.yml b/pandas/.github/workflows/package-checks.yml
new file mode 100644
index 0000000000000..331af6e05b650
--- /dev/null
+++ b/pandas/.github/workflows/package-checks.yml
@@ -0,0 +1,81 @@
+name: Package Checks
+
+on:
+  push:
+    branches:
+      - main
+      - 2.3.x
+  pull_request:
+    branches:
+      - main
+      - 2.3.x
+    types: [ labeled, opened, synchronize, reopened ]
+
+permissions:
+  contents: read
+
+defaults:
+  run:
+    shell: bash -el {0}
+
+jobs:
+  pip:
+    if: ${{ github.event.label.name == 'Build' || contains(github.event.pull_request.labels.*.name, 'Build') || github.event_name == 'push'}}
+    runs-on: ubuntu-22.04
+    strategy:
+      matrix:
+        extra: ["test", "pyarrow", "performance", "computation", "fss", "aws", "gcp", "excel", "parquet", "feather", "hdf5", "spss", "postgresql", "mysql", "sql-other", "html", "xml", "plot", "output-formatting", "clipboard", "compression", "all"]
+      fail-fast: false
+    name: Install Extras - ${{ matrix.extra }}
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-pip-extras-${{ matrix.extra }}
+      cancel-in-progress: true
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Setup Python
+        id: setup_python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+
+      - name: Pip install with extra
+        run: |
+          python -m pip install .[${{ matrix.extra }}] -v
+        shell: bash -el {0}
+  conda_forge_recipe:
+    if: ${{ github.event.label.name == 'Build' || contains(github.event.pull_request.labels.*.name, 'Build') || github.event_name == 'push'}}
+    runs-on: ubuntu-22.04
+    strategy:
+      matrix:
+        python-version: ['3.10', '3.11']
+      fail-fast: false
+    name: Test Conda Forge Recipe - Python ${{ matrix.python-version }}
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-conda-forge-recipe-${{ matrix.python-version }}
+      cancel-in-progress: true
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python
+        uses: mamba-org/setup-micromamba@v2
+        with:
+          environment-name: recipe-test
+          create-args: >-
+            python=${{ matrix.python-version }}
+            boa
+            conda-verify
+          cache-downloads: true
+          cache-environment: true
+
+      - name: Build conda package
+        run: conda mambabuild ci --no-anaconda-upload --verify --strict-verify --output --output-folder .
diff --git a/pandas/.github/workflows/stale-pr.yml b/pandas/.github/workflows/stale-pr.yml
new file mode 100644
index 0000000000000..792afe8f4faf5
--- /dev/null
+++ b/pandas/.github/workflows/stale-pr.yml
@@ -0,0 +1,27 @@
+name: "Stale PRs"
+on:
+  schedule:
+  # * is a special character in YAML so you have to quote this string
+  - cron: "0 0 * * *"
+
+permissions:
+  contents: read
+
+jobs:
+  stale:
+    permissions:
+      pull-requests: write
+    if: github.repository_owner == 'pandas-dev'
+    runs-on: ubuntu-22.04
+    steps:
+    - uses: actions/stale@v9
+      with:
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+        stale-pr-message: "This pull request is stale because it has been open for thirty days with no activity. Please [update](https://pandas.pydata.org/pandas-docs/stable/development/contributing.html#updating-your-pull-request) and respond to this comment if you're still interested in working on this."
+        stale-pr-label: "Stale"
+        exempt-pr-labels: "Needs Review,Blocked,Needs Discussion"
+        days-before-issue-stale: -1
+        days-before-pr-stale: 30
+        days-before-close: -1
+        remove-stale-when-updated: false
+        debug-only: false
diff --git a/pandas/.github/workflows/unit-tests.yml b/pandas/.github/workflows/unit-tests.yml
new file mode 100644
index 0000000000000..08c41a1eeb21f
--- /dev/null
+++ b/pandas/.github/workflows/unit-tests.yml
@@ -0,0 +1,468 @@
+name: Unit Tests
+
+on:
+  push:
+    branches:
+      - main
+      - 2.3.x
+  pull_request:
+    branches:
+      - main
+      - 2.3.x
+    paths-ignore:
+      - "doc/**"
+      - "web/**"
+
+permissions:
+  contents: read
+
+defaults:
+  run:
+    shell: bash -el {0}
+
+jobs:
+  ubuntu:
+    runs-on: ${{ matrix.platform }}
+    timeout-minutes: 90
+    strategy:
+      matrix:
+        platform: [ubuntu-22.04, ubuntu-24.04-arm]
+        env_file: [actions-310.yaml, actions-311.yaml, actions-312.yaml]
+        # Prevent the include jobs from overriding other jobs
+        pattern: [""]
+        pandas_future_infer_string: ["0"]
+        include:
+          - name: "Downstream Compat"
+            env_file: actions-311-downstream_compat.yaml
+            pattern: "not slow and not network and not single_cpu"
+            pytest_target: "pandas/tests/test_downstream.py"
+            platform: ubuntu-22.04
+          - name: "Minimum Versions"
+            env_file: actions-310-minimum_versions.yaml
+            pattern: "not slow and not network and not single_cpu"
+            platform: ubuntu-22.04
+          - name: "Locale: it_IT"
+            env_file: actions-311.yaml
+            pattern: "not slow and not network and not single_cpu"
+            extra_apt: "language-pack-it"
+            # Use the utf8 version as the default, it has no bad side-effect.
+            lang: "it_IT.utf8"
+            lc_all: "it_IT.utf8"
+            # Also install it_IT (its encoding is ISO8859-1) but do not activate it.
+            # It will be temporarily activated during tests with locale.setlocale
+            extra_loc: "it_IT"
+            platform: ubuntu-22.04
+          - name: "Locale: zh_CN"
+            env_file: actions-311.yaml
+            pattern: "not slow and not network and not single_cpu"
+            extra_apt: "language-pack-zh-hans"
+            # Use the utf8 version as the default, it has no bad side-effect.
+            lang: "zh_CN.utf8"
+            lc_all: "zh_CN.utf8"
+            # Also install zh_CN (its encoding is gb2312) but do not activate it.
+            # It will be temporarily activated during tests with locale.setlocale
+            extra_loc: "zh_CN"
+            platform: ubuntu-22.04
+          - name: "Future infer strings"
+            env_file: actions-312.yaml
+            pandas_future_infer_string: "1"
+            platform: ubuntu-22.04
+          - name: "Future infer strings (without pyarrow)"
+            env_file: actions-311.yaml
+            pandas_future_infer_string: "1"
+            platform: ubuntu-22.04
+          - name: "Pypy"
+            env_file: actions-pypy-39.yaml
+            pattern: "not slow and not network and not single_cpu"
+            test_args: "--max-worker-restart 0"
+            platform: ubuntu-22.04
+          - name: "Numpy Dev"
+            env_file: actions-311-numpydev.yaml
+            pattern: "not slow and not network and not single_cpu"
+            test_args: "-W error::DeprecationWarning -W error::FutureWarning"
+            platform: ubuntu-22.04
+          - name: "Pyarrow Nightly"
+            env_file: actions-311-pyarrownightly.yaml
+            pattern: "not slow and not network and not single_cpu"
+            pandas_future_infer_string: "1"
+            platform: ubuntu-22.04
+      fail-fast: false
+    name: ${{ matrix.name || format('{0} {1}', matrix.platform, matrix.env_file) }}
+    env:
+      PATTERN: ${{ matrix.pattern }}
+      LANG: ${{ matrix.lang || 'C.UTF-8' }}
+      LC_ALL: ${{ matrix.lc_all || '' }}
+      PANDAS_CI: '1'
+      PANDAS_FUTURE_INFER_STRING: ${{ matrix.pandas_future_infer_string || '0' }}
+      TEST_ARGS: ${{ matrix.test_args || '' }}
+      PYTEST_WORKERS: 'auto'
+      PYTEST_TARGET: ${{ matrix.pytest_target || 'pandas' }}
+      # Clipboard tests
+      QT_QPA_PLATFORM: offscreen
+      REMOVE_PYARROW: ${{ matrix.name == 'Future infer strings (without pyarrow)' && '1' || '0' }}
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-${{ matrix.env_file }}-${{ matrix.pattern }}-${{ matrix.extra_apt || '' }}-${{ matrix.pandas_future_infer_string }}-${{ matrix.platform }}
+      cancel-in-progress: true
+
+    services:
+      mysql:
+        image: mysql:9
+        env:
+          MYSQL_ALLOW_EMPTY_PASSWORD: yes
+          MYSQL_DATABASE: pandas
+        options: >-
+          --health-cmd "mysqladmin ping"
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 3306:3306
+
+      postgres:
+        image: postgres:17
+        env:
+          PGUSER: postgres
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: pandas
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 5432:5432
+
+      moto:
+        image: motoserver/moto:5.0.27
+        env:
+          AWS_ACCESS_KEY_ID: foobar_key
+          AWS_SECRET_ACCESS_KEY: foobar_secret
+        ports:
+          - 5000:5000
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+
+    - name: Extra installs
+      # https://pytest-qt.readthedocs.io/en/latest/troubleshooting.html#github-actions-azure-pipelines-travis-ci-and-gitlab-ci-cd
+      run: sudo apt-get update && sudo apt-get install -y libegl1 libopengl0 ${{ matrix.extra_apt || ''}}
+
+    - name: Generate extra locales
+      # These extra locales will be available for locale.setlocale() calls in tests
+      run: sudo locale-gen ${{ matrix.extra_loc }}
+      if: ${{ matrix.extra_loc }}
+
+    - name: Set up Conda
+      uses: ./.github/actions/setup-conda
+      with:
+        environment-file: ci/deps/${{ matrix.env_file }}
+
+    - name: Build Pandas
+      id: build
+      uses: ./.github/actions/build_pandas
+      # TODO: Re-enable once Pypy has Pypy 3.10 on conda-forge
+      if: ${{ matrix.name != 'Pypy' }}
+
+    - name: Test (not single_cpu)
+      uses: ./.github/actions/run-tests
+      if: ${{ matrix.name != 'Pypy' }}
+      env:
+        # Set pattern to not single_cpu if not already set
+        PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
+
+    - name: Test (single_cpu)
+      uses: ./.github/actions/run-tests
+      env:
+        PATTERN: 'single_cpu'
+        PYTEST_WORKERS: 0
+      if: ${{ matrix.pattern == '' && (always() && steps.build.outcome == 'success')}}
+
+  macos-windows:
+    timeout-minutes: 90
+    strategy:
+      matrix:
+        # Note: Don't use macOS latest since macos 14 appears to be arm64 only
+        os: [macos-13, macos-14, windows-latest]
+        env_file: [actions-310.yaml, actions-311.yaml, actions-312.yaml]
+      fail-fast: false
+    runs-on: ${{ matrix.os }}
+    name: ${{ format('{0} {1}', matrix.os, matrix.env_file) }}
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-${{ matrix.env_file }}-${{ matrix.os }}
+      cancel-in-progress: true
+    env:
+      PANDAS_CI: 1
+      PYTEST_TARGET: pandas
+      PATTERN: "not slow and not db and not network and not single_cpu"
+      PYTEST_WORKERS: 'auto'
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Conda
+        uses: ./.github/actions/setup-conda
+        with:
+          environment-file: ci/deps/${{ matrix.env_file }}
+
+      - name: Build Pandas
+        uses: ./.github/actions/build_pandas
+
+      - name: Test
+        uses: ./.github/actions/run-tests
+
+  Linux-32-bit:
+    runs-on: ubuntu-22.04
+    container:
+      image: quay.io/pypa/manylinux2014_i686
+      options: --platform linux/386
+    steps:
+      - name: Checkout pandas Repo
+        # actions/checkout does not work since it requires node
+        run: |
+          git config --global --add safe.directory $PWD
+
+          if [ $GITHUB_EVENT_NAME != pull_request ]; then
+              git clone --recursive --branch=$GITHUB_REF_NAME https://github.com/${GITHUB_REPOSITORY}.git $GITHUB_WORKSPACE
+              git reset --hard $GITHUB_SHA
+          else
+              git clone --recursive https://github.com/${GITHUB_REPOSITORY}.git $GITHUB_WORKSPACE
+              git fetch origin $GITHUB_REF:my_ref_name
+              git checkout $GITHUB_BASE_REF
+              git -c user.email="you@example.com" merge --no-commit my_ref_name
+          fi
+      - name: Build environment and Run Tests
+        # https://github.com/numpy/numpy/issues/24703#issuecomment-1722379388
+        run: |
+          /opt/python/cp313-cp313/bin/python -m venv ~/virtualenvs/pandas-dev
+          . ~/virtualenvs/pandas-dev/bin/activate
+          python -m pip install --no-cache-dir -U pip wheel setuptools meson[ninja]==1.2.1 meson-python==0.13.1
+          python -m pip install numpy -Csetup-args="-Dallow-noblas=true"
+          python -m pip install --no-cache-dir versioneer[toml] cython python-dateutil pytest>=7.3.2 pytest-xdist>=3.4.0 hypothesis>=6.84.0
+          python -m pip install --no-cache-dir --no-build-isolation -e . -Csetup-args="--werror"
+          python -m pip list --no-cache-dir
+          PANDAS_CI=1 python -m pytest -m 'not slow and not network and not clipboard and not single_cpu' pandas --junitxml=test-data.xml
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-32bit
+      cancel-in-progress: true
+
+  Linux-Musl:
+    runs-on: ubuntu-22.04
+    container:
+      image: quay.io/pypa/musllinux_1_2_x86_64
+    steps:
+      - name: Checkout pandas Repo
+        # actions/checkout does not work since it requires node
+        run: |
+          git config --global --add safe.directory $PWD
+
+          if [ $GITHUB_EVENT_NAME != pull_request ]; then
+              git clone --recursive --branch=$GITHUB_REF_NAME https://github.com/${GITHUB_REPOSITORY}.git $GITHUB_WORKSPACE
+              git reset --hard $GITHUB_SHA
+          else
+              git clone --recursive https://github.com/${GITHUB_REPOSITORY}.git $GITHUB_WORKSPACE
+              git fetch origin $GITHUB_REF:my_ref_name
+              git checkout $GITHUB_BASE_REF
+              git -c user.email="you@example.com" merge --no-commit my_ref_name
+          fi
+      - name: Configure System Packages
+        run: |
+          apk update
+          apk add musl-locales
+      - name: Build environment
+        run: |
+          /opt/python/cp313-cp313/bin/python -m venv ~/virtualenvs/pandas-dev
+          . ~/virtualenvs/pandas-dev/bin/activate
+          python -m pip install --no-cache-dir -U pip wheel setuptools meson-python==0.13.1 meson[ninja]==1.2.1
+          python -m pip install --no-cache-dir versioneer[toml] cython numpy python-dateutil pytest>=7.3.2 pytest-xdist>=3.4.0 hypothesis>=6.84.0
+          python -m pip install --no-cache-dir --no-build-isolation -e . -Csetup-args="--werror"
+          python -m pip list --no-cache-dir
+
+      - name: Run Tests
+        run: |
+          . ~/virtualenvs/pandas-dev/bin/activate
+          PANDAS_CI=1 python -m pytest -m 'not slow and not network and not clipboard and not single_cpu' pandas --junitxml=test-data.xml
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-musl
+      cancel-in-progress: true
+
+  python-dev:
+    # This job may or may not run depending on the state of the next
+    # unreleased Python version. DO NOT DELETE IT.
+    #
+    # In general, this will remain frozen(present, but not running) until:
+    #    - The next unreleased Python version has released beta 1
+    #      - This version should be available on GitHub Actions.
+    #    - Our required build/runtime dependencies(numpy, Cython, python-dateutil)
+    #      support that unreleased Python version.
+    #    To unfreeze, comment out the ``if: false`` condition, and make sure you update
+    #    the name of the workflow and Python version in actions/setup-python ``python-version:``
+    #
+    # After it has been unfrozen, this file should remain unfrozen(present, and running) until:
+    #    - The next Python version has been officially released.
+    #    OR
+    #    - Most/All of our optional dependencies support the next Python version AND
+    #    - The next Python version has released a rc(we are guaranteed a stable ABI).
+    #    To freeze this file, uncomment out the ``if: false`` condition, and migrate the jobs
+    #    to the corresponding posix/windows-macos/sdist etc. workflows.
+    # Feel free to modify this comment as necessary.
+    # if: false # Uncomment this to freeze the workflow, comment it to unfreeze
+    defaults:
+      run:
+        shell: bash -eou pipefail {0}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        # Separate out macOS 13 and 14, since macOS 14 is arm64 only
+        os: [ubuntu-22.04, macOS-13, macOS-14, windows-latest]
+
+    timeout-minutes: 90
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-${{ matrix.os }}-python-dev
+      cancel-in-progress: true
+
+    env:
+      PYTEST_WORKERS: "auto"
+      PANDAS_CI: 1
+      PATTERN: "not slow and not network and not clipboard and not single_cpu"
+      PYTEST_TARGET: pandas
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python Dev Version
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.13-dev'
+
+      - name: Build Environment
+        run: |
+          python --version
+          python -m pip install --upgrade pip setuptools wheel meson[ninja]==1.2.1 meson-python==0.13.1
+          python -m pip install --pre --extra-index-url https://pypi.anaconda.org/scientific-python-nightly-wheels/simple numpy
+          python -m pip install versioneer[toml] python-dateutil tzdata cython hypothesis>=6.84.0 pytest>=7.3.2 pytest-xdist>=3.4.0 pytest-cov
+          python -m pip install -ve . --no-build-isolation --no-index --no-deps -Csetup-args="--werror"
+          python -m pip list
+
+      - name: Run Tests
+        uses: ./.github/actions/run-tests
+
+  python-freethreading:
+    defaults:
+      run:
+        shell: bash -eou pipefail {0}
+    runs-on: ubuntu-22.04
+
+    timeout-minutes: 90
+
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-python-freethreading-dev
+      cancel-in-progress: true
+
+    env:
+      PYTEST_WORKERS: "auto"
+      PANDAS_CI: 1
+      PATTERN: "not slow and not network and not clipboard and not single_cpu"
+      PYTEST_TARGET: pandas
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python Free-threading Version
+        uses: deadsnakes/action@v3.2.0
+        with:
+          python-version: 3.13-dev
+          nogil: true
+
+      - name: Build Environment
+        run: |
+          python --version
+          python -m pip install --upgrade pip setuptools wheel numpy meson[ninja]==1.2.1 meson-python==0.13.1
+          python -m pip install --pre --extra-index-url https://pypi.anaconda.org/scientific-python-nightly-wheels/simple cython
+          python -m pip install versioneer[toml] python-dateutil pytz tzdata hypothesis>=6.84.0 pytest>=7.3.2 pytest-xdist>=3.4.0 pytest-cov
+          python -m pip install -ve . --no-build-isolation --no-index --no-deps -Csetup-args="--werror"
+          python -m pip list
+
+      - name: Run Tests
+        uses: ./.github/actions/run-tests
+
+  # NOTE: this job must be kept in sync with the Pyodide build job in wheels.yml
+  emscripten:
+    # Note: the Python version, Emscripten toolchain version are determined
+    # by the Pyodide version. The appropriate versions can be found in the
+    # Pyodide repodata.json "info" field, or in the Makefile.envs file:
+    # https://github.com/pyodide/pyodide/blob/stable/Makefile.envs#L2
+    # The Node.js version can be determined via Pyodide:
+    # https://pyodide.org/en/stable/usage/index.html#node-js
+    name: Pyodide build
+    runs-on: ubuntu-22.04
+    concurrency:
+      # https://github.community/t/concurrecy-not-work-for-push/183068/7
+      group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-wasm
+      cancel-in-progress: true
+    steps:
+      - name: Checkout pandas Repo
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python for pyodide-build
+        id: setup-python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+
+      - name: Set up Emscripten toolchain
+        uses: mymindstorm/setup-emsdk@v14
+        with:
+          version: '3.1.58'
+          actions-cache-folder: emsdk-cache
+
+      - name: Install pyodide-build
+        run: pip install "pyodide-build>=0.29.2"
+
+      - name: Build pandas for Pyodide
+        run: |
+          pyodide build
+
+      - name: Set up Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Set up Pyodide virtual environment
+        env:
+          pyodide-version: '0.27.1'
+        run: |
+          pyodide xbuildenv install ${{ env.pyodide-version }}
+          pyodide venv .venv-pyodide
+          source .venv-pyodide/bin/activate
+          pip install dist/*.whl
+
+      - name: Test pandas for Pyodide
+        env:
+          PANDAS_CI: 1
+        run: |
+          source .venv-pyodide/bin/activate
+          pip install pytest hypothesis
+          # do not import pandas from the checked out repo
+          cd ..
+          python -c 'import pandas as pd; pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db"])'
diff --git a/pandas/.github/workflows/wheels.yml b/pandas/.github/workflows/wheels.yml
new file mode 100644
index 0000000000000..1cb4fbcc726cd
--- /dev/null
+++ b/pandas/.github/workflows/wheels.yml
@@ -0,0 +1,201 @@
+# Workflow to build wheels for upload to PyPI.
+# Inspired by numpy's cibuildwheel config https://github.com/numpy/numpy/blob/main/.github/workflows/wheels.yml
+#
+# In an attempt to save CI resources, wheel builds do
+# not run on each push but only weekly and for releases.
+# Wheel builds can be triggered from the Actions page
+# (if you have the permissions) on a commit to main.
+#
+# Alternatively, you can add labels to the pull request in order to trigger wheel
+# builds.
+# The label(s) that trigger builds are:
+# - Build
+name: Wheel builder
+
+on:
+  schedule:
+  # 3:27 UTC every day
+  - cron: "27 3 * * *"
+  push:
+  pull_request:
+    types: [labeled, opened, synchronize, reopened]
+    paths-ignore:
+      - "doc/**"
+      - "web/**"
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
+  cancel-in-progress: true
+
+permissions:
+  contents: read
+
+jobs:
+  build_sdist:
+    name: Build sdist
+    if: >-
+      (github.event_name == 'schedule') ||
+      github.event_name == 'workflow_dispatch' ||
+      (github.event_name == 'pull_request' &&
+      contains(github.event.pull_request.labels.*.name, 'Build')) ||
+      (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v') && ( ! endsWith(github.ref, 'dev0')))
+    runs-on: ubuntu-22.04
+    env:
+      IS_PUSH: ${{ github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v') }}
+      IS_SCHEDULE_DISPATCH: ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' }}
+    outputs:
+      sdist_file: ${{ steps.save-path.outputs.sdist_name }}
+    steps:
+      - name: Checkout pandas
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.11'
+
+      - name: Build sdist
+        run: |
+          python -m pip install build
+          python -m build --sdist
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: sdist
+          path: ./dist/*
+
+      - name: Sanity check sdist files
+        run: |
+          ls ./dist
+
+      - name: Output sdist name
+        id: save-path
+        shell: bash -el {0}
+        run: echo "sdist_name=$(ls ./dist)" >> "$GITHUB_OUTPUT"
+
+  build_wheels:
+    needs: build_sdist
+    name: Build wheel for ${{ matrix.python[0] }}-${{ matrix.buildplat[1] }}
+    if: >-
+      (github.event_name == 'schedule') ||
+      github.event_name == 'workflow_dispatch' ||
+      (github.event_name == 'pull_request' &&
+      contains(github.event.pull_request.labels.*.name, 'Build')) ||
+      (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v') && ( ! endsWith(github.ref, 'dev0')))
+    runs-on: ${{ matrix.buildplat[0] }}
+    strategy:
+      fail-fast: false
+      matrix:
+        # GitHub Actions doesn't support pairing matrix values together, let's improvise
+        # https://github.com/github/feedback/discussions/7835#discussioncomment-1769026
+        buildplat:
+        - [ubuntu-22.04, manylinux_x86_64]
+        - [ubuntu-22.04, musllinux_x86_64]
+        - [ubuntu-24.04-arm, manylinux_aarch64]
+        - [macos-13, macosx_x86_64]
+        # Note: M1 images on Github Actions start from macOS 14
+        - [macos-14, macosx_arm64]
+        - [windows-2022, win_amd64]
+        # TODO: support PyPy?
+        python: [["cp310", "3.10"], ["cp311", "3.11"], ["cp312", "3.12"], ["cp313", "3.13"], ["cp313t", "3.13"]]
+        include:
+        # TODO: Remove this plus installing build deps in cibw_before_build.sh
+        # after pandas can be built with a released NumPy/Cython
+        - python: ["cp313t", "3.13"]
+          cibw_build_frontend: 'pip; args: --no-build-isolation'
+        # Build Pyodide wheels and upload them to Anaconda.org
+        # NOTE: this job is similar to the one in unit-tests.yml except for the fact
+        # that it uses cibuildwheel instead of a standard Pyodide xbuildenv setup.
+        - buildplat: [ubuntu-22.04, pyodide_wasm32]
+          python: ["cp312", "3.12"]
+          cibw_build_frontend: 'build'
+
+    env:
+      IS_PUSH: ${{ github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v') }}
+      IS_SCHEDULE_DISPATCH: ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' }}
+    steps:
+      - name: Checkout pandas
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      # TODO: Build wheels from sdist again
+      # There's some sort of weird race condition?
+      # within Github that makes the sdist be missing files
+
+      # We need to build wheels from the sdist since the sdist
+      # removes unnecessary files from the release
+      - name: Download sdist (not macOS)
+        #if: ${{ matrix.buildplat[1] != 'macosx_*' }}
+        uses: actions/download-artifact@v4
+        with:
+          name: sdist
+          path: ./dist
+
+      - name: Output sdist name (macOS)
+        id: save-path
+        shell: bash -el {0}
+        run: echo "sdist_name=$(ls ./dist)" >> "$GITHUB_ENV"
+
+      # Python version used to build sdist doesn't matter
+      # wheel will be built from sdist with the correct version
+      - name: Unzip sdist (macOS)
+        if: ${{ startsWith(matrix.buildplat[1], 'macosx') }}
+        run: |
+          tar -xzf ./dist/${{ env.sdist_name }} -C ./dist
+
+      - name: Output sdist name (macOS)
+        id: save-path2
+        shell: bash -el {0}
+        run: echo "sdist_name=$(cd ./dist && ls -d */)" >> "$GITHUB_ENV"
+
+      - name: Build wheels
+        uses: pypa/cibuildwheel@v2.22.0
+        with:
+         package-dir: ./dist/${{ startsWith(matrix.buildplat[1], 'macosx') && env.sdist_name || needs.build_sdist.outputs.sdist_file }}
+        env:
+          CIBW_BUILD: ${{ matrix.python[0] }}-${{ matrix.buildplat[1] }}
+          CIBW_BUILD_FRONTEND: ${{ matrix.cibw_build_frontend || 'pip' }}
+          CIBW_PLATFORM: ${{ matrix.buildplat[1] == 'pyodide_wasm32' && 'pyodide' || 'auto' }}
+
+      - name: Set up Python
+        uses: mamba-org/setup-micromamba@v2
+        with:
+          environment-name: wheel-env
+          # Use a fixed Python, since we might have an unreleased Python not
+          # yet present on conda-forge
+          create-args: >-
+            python=3.11
+            anaconda-client
+            wheel
+          cache-downloads: true
+          cache-environment: true
+
+      - name: Validate wheel RECORD
+        shell: bash -el {0}
+        run: for whl in $(ls wheelhouse); do wheel unpack wheelhouse/$whl -d /tmp; done
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: ${{ matrix.python[0] }}-${{ matrix.buildplat[1] }}
+          path: ./wheelhouse/*.whl
+
+      - name: Upload wheels & sdist
+        if: ${{ success() && (env.IS_SCHEDULE_DISPATCH == 'true' || env.IS_PUSH == 'true') }}
+        shell: bash -el {0}
+        env:
+          PANDAS_STAGING_UPLOAD_TOKEN: ${{ secrets.PANDAS_STAGING_UPLOAD_TOKEN }}
+          PANDAS_NIGHTLY_UPLOAD_TOKEN: ${{ secrets.PANDAS_NIGHTLY_UPLOAD_TOKEN }}
+        # trigger an upload to
+        # https://anaconda.org/scientific-python-nightly-wheels/pandas
+        # for cron jobs or "Run workflow" (restricted to main branch).
+        # Tags will upload to
+        # https://anaconda.org/multibuild-wheels-staging/pandas
+        # The tokens were originally generated at anaconda.org
+        run: |
+          source ci/upload_wheels.sh
+          set_upload_vars
+          upload_wheels
diff --git a/pandas/.gitignore b/pandas/.gitignore
new file mode 100644
index 0000000000000..d951f3fb9cbad
--- /dev/null
+++ b/pandas/.gitignore
@@ -0,0 +1,143 @@
+#########################################
+# Editor temporary/working/backup files #
+.#*
+*\#*\#
+[#]*#
+*~
+*$
+*.bak
+*flymake*
+*.iml
+*.kdev4
+*.log
+*.swp
+*.pdb
+*.zip
+.project
+.pydevproject
+.settings
+.idea
+.vagrant
+.noseids
+.ipynb_checkpoints
+.tags
+.cache/
+.vscode/
+
+# Compiled source #
+###################
+*.a
+*.com
+*.class
+*.dll
+*.exe
+*.pxi
+*.o
+*.py[ocd]
+*.so
+.build_cache_dir
+.mesonpy-native-file.ini
+MANIFEST
+compile_commands.json
+debug
+.debug
+
+# Python files #
+################
+# setup.py working directory
+build
+# sphinx build directory
+doc/_build
+# setup.py dist directory
+dist
+# Egg metadata
+*.egg-info
+.eggs
+.pypirc
+# type checkers
+pandas/py.typed
+
+# pyenv
+.python-version
+
+# tox testing tool
+.tox
+# rope
+.ropeproject
+# wheel files
+*.whl
+**/wheelhouse/*
+pip-wheel-metadata
+# coverage
+.coverage
+coverage.xml
+coverage_html_report
+.mypy_cache
+*.pytest_cache
+.ruff_cache
+# hypothesis test database
+.hypothesis/
+__pycache__
+# pytest-monkeytype
+monkeytype.sqlite3
+# meson editable install folder
+.mesonpy
+
+
+# OS generated files #
+######################
+.directory
+.gdb_history
+.DS_Store
+ehthumbs.db
+Icon?
+Thumbs.db
+
+# Data files #
+##############
+*.dta
+*.xpt
+*.h5
+pandas/io/*.dat
+pandas/io/*.json
+scikits
+
+# Generated Sources #
+#####################
+!skts.c
+*.c
+*.cpp
+!pandas/_libs/src/**/*.c
+!pandas/_libs/src/**/*.h
+!pandas/_libs/include/**/*.h
+
+# Unit / Performance Testing #
+##############################
+asv_bench/env/
+asv_bench/html/
+asv_bench/results/
+asv_bench/pandas/
+test-data.xml
+
+# Documentation generated files #
+#################################
+doc/source/generated
+doc/source/user_guide/styled.xlsx
+doc/source/reference/api
+doc/source/_static/*.html
+doc/source/vbench
+doc/source/vbench.rst
+doc/source/index.rst
+doc/build/html/index.html
+# Windows specific leftover:
+doc/tmp.sv
+env/
+doc/source/savefig/
+
+# Interactive terminal generated files #
+########################################
+.jupyterlite.doit.db
+
+# Pyodide/WASM related files #
+##############################
+/.pyodide-xbuildenv-*
diff --git a/pandas/.gitpod.yml b/pandas/.gitpod.yml
new file mode 100644
index 0000000000000..5bf028750f30f
--- /dev/null
+++ b/pandas/.gitpod.yml
@@ -0,0 +1,60 @@
+# Building pandas on init
+# Might delegate this later to prebuild with Q2 improvements on gitpod
+# https://www.gitpod.io/docs/config-start-tasks/#configuring-the-terminal
+# -------------------------------------------------------------------------
+
+# images for gitpod pandas are in https://hub.docker.com/r/pandas/pandas-gitpod/tags
+# we're using the Dockerfile in the base of the repo
+image:
+  file: Dockerfile
+tasks:
+  - name: Prepare development environment
+    init: |
+      mkdir -p .vscode
+      cp gitpod/settings.json .vscode/settings.json
+      git fetch --tags
+      python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true
+      pre-commit install --install-hooks
+    command: |
+      python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true
+      echo "✨ Pre-build complete! You can close this terminal ✨ "
+
+# --------------------------------------------------------
+# exposing ports for liveserve
+ports:
+  - port: 5500
+    onOpen: notify
+
+# --------------------------------------------------------
+# some useful extensions to have
+vscode:
+  extensions:
+    - ms-python.python
+    - yzhang.markdown-all-in-one
+    - eamodio.gitlens
+    - lextudio.restructuredtext
+    - ritwickdey.liveserver
+    # add or remove what you think is generally useful to most contributors
+    # avoid adding too many. they each open a pop-up window
+
+# --------------------------------------------------------
+# Using prebuilds for the container
+# With this configuration the prebuild will happen on push to main
+github:
+  prebuilds:
+    # enable for main/default branch
+    main: true
+    # enable for other branches (defaults to false)
+    branches: false
+    # enable for pull requests coming from this repo (defaults to true)
+    pullRequests: false
+    # enable for pull requests coming from forks (defaults to false)
+    pullRequestsFromForks: false
+    # add a check to pull requests (defaults to true)
+    addCheck: false
+    # add a "Review in Gitpod" button as a comment to pull requests (defaults to false)
+    addComment: false
+    # add a "Review in Gitpod" button to the pull request's description (defaults to false)
+    addBadge: false
+    # add a label once the prebuild is ready to pull requests (defaults to false)
+    addLabel: false
diff --git a/pandas/.pre-commit-config.yaml b/pandas/.pre-commit-config.yaml
new file mode 100644
index 0000000000000..c13c38b20a7f8
--- /dev/null
+++ b/pandas/.pre-commit-config.yaml
@@ -0,0 +1,315 @@
+minimum_pre_commit_version: 2.15.0
+exclude: ^LICENSES/|\.(html|csv|svg)$
+# reserve "manual" for relatively slow hooks which we still want to run in CI
+default_stages: [
+    pre-commit,
+    pre-merge-commit,
+    pre-push,
+    prepare-commit-msg,
+    commit-msg,
+    post-checkout,
+    post-commit,
+    post-merge,
+    post-rewrite
+]
+ci:
+    autofix_prs: false
+    autoupdate_schedule: monthly
+    # manual stage hooks
+    skip: [pyright, mypy]
+repos:
+-   repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.4
+    hooks:
+    -   id: ruff
+        args: [--exit-non-zero-on-fix]
+        exclude: ^pandas/tests/frame/test_query_eval.py
+    -   id: ruff
+        # TODO: remove autofixe-only rules when they are checked by ruff
+        name: ruff-selected-autofixes
+        alias: ruff-selected-autofixes
+        files: ^pandas
+        exclude: ^pandas/tests
+        args: [--select, "ANN001,ANN2", --fix-only, --exit-non-zero-on-fix]
+    -   id: ruff-format
+        exclude: ^scripts|^pandas/tests/frame/test_query_eval.py
+-   repo: https://github.com/jendrikseipp/vulture
+    rev: 'v2.14'
+    hooks:
+      - id: vulture
+        entry: python scripts/run_vulture.py
+        pass_filenames: true
+        require_serial: false
+-   repo: https://github.com/codespell-project/codespell
+    rev: v2.4.1
+    hooks:
+    -   id: codespell
+        types_or: [python, rst, markdown, cython, c]
+        additional_dependencies: [tomli]
+-   repo: https://github.com/MarcoGorelli/cython-lint
+    rev: v0.16.6
+    hooks:
+    -   id: cython-lint
+    -   id: double-quote-cython-strings
+-   repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v5.0.0
+    hooks:
+    -   id: check-case-conflict
+    -   id: check-toml
+    -   id: check-xml
+    -   id: check-yaml
+        exclude: ^ci/meta.yaml$
+    -   id: end-of-file-fixer
+        exclude: \.txt$
+    -   id: mixed-line-ending
+        args: [--fix=auto]
+        exclude: ^pandas/tests/io/parser/data/utf16_ex.txt$
+    -   id: fix-byte-order-marker
+    -   id: fix-encoding-pragma
+        args: [--remove]
+    -   id: trailing-whitespace
+        args: [--markdown-linebreak-ext=md]
+-   repo: https://github.com/PyCQA/isort
+    rev: 6.0.0
+    hooks:
+    -   id: isort
+-   repo: https://github.com/asottile/pyupgrade
+    rev: v3.19.1
+    hooks:
+    -   id: pyupgrade
+        args: [--py310-plus]
+-   repo: https://github.com/pre-commit/pygrep-hooks
+    rev: v1.10.0
+    hooks:
+      - id: rst-backticks
+      - id: rst-directive-colons
+        types: [text]  # overwrite types: [rst]
+        types_or: [python, rst]
+      - id: rst-inline-touching-normal
+        exclude: ^pandas/tests/frame/test_query_eval.py
+        types: [text]  # overwrite types: [rst]
+        types_or: [python, rst]
+-   repo: https://github.com/sphinx-contrib/sphinx-lint
+    rev: v1.0.0
+    hooks:
+    - id: sphinx-lint
+      args: ["--enable", "all", "--disable", "line-too-long"]
+-   repo: https://github.com/pre-commit/mirrors-clang-format
+    rev: v19.1.7
+    hooks:
+    - id: clang-format
+      files: ^pandas/_libs/src|^pandas/_libs/include
+      args: [-i]
+      types_or: [c, c++]
+-   repo: https://github.com/trim21/pre-commit-mirror-meson
+    rev: v1.7.0
+    hooks:
+    - id: meson-fmt
+      args: ['--inplace']
+-   repo: https://github.com/shellcheck-py/shellcheck-py
+    rev: v0.10.0.1
+    hooks:
+    -   id: shellcheck
+        args: ["--severity=warning"]
+-   repo: local
+    hooks:
+    -   id: pyright
+        # note: assumes python env is setup and activated
+        name: pyright
+        entry: pyright
+        language: node
+        pass_filenames: false
+        types: [python]
+        stages: [manual]
+        additional_dependencies: &pyright_dependencies
+        - pyright@1.1.383
+    -   id: pyright
+        # note: assumes python env is setup and activated
+        name: pyright reportGeneralTypeIssues
+        entry: pyright -p pyright_reportGeneralTypeIssues.json --level warning
+        language: node
+        pass_filenames: false
+        types: [python]
+        stages: [manual]
+        additional_dependencies: *pyright_dependencies
+    -   id: mypy
+        # note: assumes python env is setup and activated
+        name: mypy
+        entry: mypy
+        language: system
+        pass_filenames: false
+        types: [python]
+        stages: [manual]
+    -   id: mypy
+        # note: assumes python env is setup and activated
+        # note: requires pandas dev to be installed
+        name: mypy (stubtest)
+        entry: python
+        language: system
+        pass_filenames: false
+        types: [pyi]
+        args: [scripts/run_stubtest.py]
+        stages: [manual]
+    -   id: inconsistent-namespace-usage
+        name: 'Check for inconsistent use of pandas namespace'
+        entry: python scripts/check_for_inconsistent_pandas_namespace.py
+        exclude: ^pandas/core/interchange/
+        language: python
+        types: [python]
+    -   id: unwanted-patterns
+        name: Unwanted patterns
+        language: pygrep
+        entry: |
+            (?x)
+            # outdated annotation syntax
+            \#\ type:\ (?!ignore)
+
+            # foo._class__ instead of type(foo)
+            |\.__class__
+
+            # Numpy
+            |from\ numpy\ import\ random
+            |from\ numpy\.random\ import
+
+            # Incorrect code-block / IPython directives
+            |\.\.\ code-block\ ::
+            |\.\.\ ipython\ ::
+            # directive should not have a space before ::
+            |\.\.\ \w+\ ::
+
+            # Check for deprecated messages without sphinx directive
+            |(DEPRECATED|DEPRECATE|Deprecated)(:|,|\.)
+
+            # builtin filter function
+            |(?<!def)[\(\s]filter\(
+        types_or: [python, cython, rst]
+        exclude: ^doc/source/development/code_style\.rst  # contains examples of patterns to avoid
+    -   id: incorrect-backticks
+        name: Check for backticks incorrectly rendering because of missing spaces
+        language: pygrep
+        entry: '[a-zA-Z0-9]\`\`?[a-zA-Z0-9]'
+        types: [rst]
+        files: ^doc/source/
+    -   id: seed-check-asv
+        name: Check for unnecessary random seeds in asv benchmarks
+        language: pygrep
+        entry: 'np\.random\.seed'
+        files: ^asv_bench/benchmarks
+        exclude: ^asv_bench/benchmarks/pandas_vb_common\.py
+    -   id: unwanted-patterns-in-tests
+        name: Unwanted patterns in tests
+        language: pygrep
+        entry: |
+            (?x)
+            # imports from pandas._testing instead of `import pandas._testing as tm`
+            from\ pandas\._testing\ import
+            |from\ pandas\ import\ _testing\ as\ tm
+
+            # pandas.testing instead of tm
+            |pd\.testing\.
+
+            # pd.api.types instead of from pandas.api.types import ...
+            |(pd|pandas)\.api\.types\.
+
+            # np.array_equal
+            |(numpy|np)\.array_equal
+
+            # pytest raises without context
+            |\s\ pytest.raises
+
+            # Unseeded numpy default_rng
+            |default_rng\(\)
+        files: ^pandas/tests/
+        types_or: [python, cython, rst]
+    -   id: unwanted-patterns-in-cython
+        name: Unwanted patterns in Cython code
+        language: pygrep
+        entry: |
+            (?x)
+            # `<type>obj` as opposed to `<type> obj`
+            [a-zA-Z0-9*]>[ ]
+        types: [cython]
+    -   id: pip-to-conda
+        name: Generate pip dependency from conda
+        language: python
+        entry: python scripts/generate_pip_deps_from_conda.py
+        files: ^(environment.yml|requirements-dev.txt)$
+        pass_filenames: false
+        additional_dependencies: [tomli, pyyaml]
+    -   id: title-capitalization
+        name: Validate correct capitalization among titles in documentation
+        entry: python scripts/validate_rst_title_capitalization.py
+        language: python
+        types: [rst]
+        files: ^doc/source/(development|reference)/
+    -   id: unwanted-patterns-private-function-across-module
+        name: Check for use of private functions across modules
+        language: python
+        entry: python scripts/validate_unwanted_patterns.py --validation-type="private_function_across_module"
+        types: [python]
+        exclude: ^(asv_bench|pandas/tests|doc)/
+    -   id: unwanted-patterns-private-import-across-module
+        name: Check for import of private attributes across modules
+        language: python
+        entry: python scripts/validate_unwanted_patterns.py --validation-type="private_import_across_module"
+        types: [python]
+        exclude: |
+            (?x)
+            ^(asv_bench|pandas/tests|doc)/
+            |scripts/validate_min_versions_in_sync\.py$
+    -   id: unwanted-patterns-strings-with-misplaced-whitespace
+        name: Check for strings with misplaced spaces
+        language: python
+        entry: python scripts/validate_unwanted_patterns.py --validation-type="strings_with_wrong_placed_whitespace"
+        types_or: [python, cython]
+    -   id: unwanted-patterns-nodefault-used-not-only-for-typing
+        name: Check that `pandas._libs.lib.NoDefault` is used only for typing
+        language: python
+        entry: python scripts/validate_unwanted_patterns.py --validation-type="nodefault_used_not_only_for_typing"
+        types: [python]
+    -   id: no-return-exception
+        name: Use raise instead of return for exceptions
+        language: pygrep
+        entry: 'return [A-Za-z]+(Error|Exit|Interrupt|Exception|Iteration)'
+        files: ^pandas/
+        types: [python]
+        exclude: ^pandas/tests/
+    -   id: pandas-errors-documented
+        name: Ensure pandas errors are documented in doc/source/reference/testing.rst
+        entry: python scripts/pandas_errors_documented.py
+        language: python
+        files: ^pandas/errors/__init__.py$
+    -   id: pg8000-not-installed-CI
+        name: Check for pg8000 not installed on CI for test_pg8000_sqlalchemy_passthrough_error
+        language: pygrep
+        entry: 'pg8000'
+        files: ^ci/deps
+        types: [yaml]
+    -   id: validate-min-versions-in-sync
+        name: Check minimum version of dependencies are aligned
+        entry: python -m scripts.validate_min_versions_in_sync
+        language: python
+        files: ^(ci/deps/actions-.*-minimum_versions\.yaml|pandas/compat/_optional\.py)$
+        additional_dependencies: [tomli, pyyaml]
+        pass_filenames: false
+    -   id: validate-errors-locations
+        name: Validate errors locations
+        description: Validate errors are in appropriate locations.
+        entry: python scripts/validate_exception_location.py
+        language: python
+        files: ^pandas/
+        exclude: ^(pandas/_libs/|pandas/tests/|pandas/errors/__init__.py$|pandas/_version.py)
+        types: [python]
+    -   id: check-test-naming
+        name: check that test names start with 'test'
+        entry: python -m scripts.check_test_naming
+        types: [python]
+        files: ^pandas/tests
+        language: python
+    -   id: sort-whatsnew-items
+        name: sort whatsnew entries alphabetically
+        entry: python -m scripts.sort_whatsnew_note
+        types: [rst]
+        language: python
+        files: ^doc/source/whatsnew/v
+        exclude: ^doc/source/whatsnew/v(0|1|2\.0\.0)
diff --git a/pandas/AUTHORS.md b/pandas/AUTHORS.md
new file mode 100644
index 0000000000000..84fcfe05e3043
--- /dev/null
+++ b/pandas/AUTHORS.md
@@ -0,0 +1,56 @@
+About the Copyright Holders
+===========================
+
+*   Copyright (c) 2008-2011 AQR Capital Management, LLC
+
+    AQR Capital Management began pandas development in 2008. Development was
+    led by Wes McKinney. AQR released the source under this license in 2009.
+*   Copyright (c) 2011-2012, Lambda Foundry, Inc.
+
+    Wes is now an employee of Lambda Foundry, and remains the pandas project
+    lead.
+*   Copyright (c) 2011-2012, PyData Development Team
+
+    The PyData Development Team is the collection of developers of the PyData
+    project. This includes all of the PyData sub-projects, including pandas. The
+    core team that coordinates development on GitHub can be found here:
+    https://github.com/pydata.
+
+Full credits for pandas contributors can be found in the documentation.
+
+Our Copyright Policy
+====================
+
+PyData uses a shared copyright model. Each contributor maintains copyright
+over their contributions to PyData. However, it is important to note that
+these contributions are typically only changes to the repositories. Thus,
+the PyData source code, in its entirety, is not the copyright of any single
+person or institution. Instead, it is the collective copyright of the
+entire PyData Development Team. If individual contributors want to maintain
+a record of what changes/contributions they have specific copyright on,
+they should indicate their copyright in the commit message of the change
+when they commit the change to one of the PyData repositories.
+
+With this in mind, the following banner should be used in any source code
+file to indicate the copyright and license terms:
+
+```
+#-----------------------------------------------------------------------------
+# Copyright (c) 2012, PyData Development Team
+# All rights reserved.
+#
+# Distributed under the terms of the BSD Simplified License.
+#
+# The full license is in the LICENSE file, distributed with this software.
+#-----------------------------------------------------------------------------
+```
+
+Other licenses can be found in the LICENSES directory.
+
+License
+=======
+
+pandas is distributed under a 3-clause ("Simplified" or "New") BSD
+license. Parts of NumPy, SciPy, numpydoc, bottleneck, which all have
+BSD-compatible licenses, are included. Their licenses follow the pandas
+license.
diff --git a/pandas/CITATION.cff b/pandas/CITATION.cff
new file mode 100644
index 0000000000000..11f45b0d87ec7
--- /dev/null
+++ b/pandas/CITATION.cff
@@ -0,0 +1,52 @@
+cff-version: 1.2.0
+title: 'pandas-dev/pandas: Pandas'
+message: 'If you use this software, please cite it as below.'
+authors:
+  - name: "The pandas development team"
+    website: "https://pandas.pydata.org/about/team.html"
+abstract: "Pandas is a powerful data structures for data analysis, time series, and statistics."
+doi: 10.5281/zenodo.3509134
+license: BSD-3-Clause
+license-url: "https://github.com/pandas-dev/pandas/blob/main/LICENSE"
+repository-code: "https://github.com/pandas-dev/pandas"
+keywords:
+  - python
+  - data science
+  - flexible
+  - pandas
+  - alignment
+  - data analysis
+type: software
+url: "https://pandas.pydata.org/"
+references:
+  - type: article
+    authors:
+      - given-names: Wes
+        family-names: McKinney
+        affiliation: AQR Capital Management, LLC
+        email: wesmckinn@gmail.com
+    title: Data Structures for Statistical Computing in Python
+    doi: 10.25080/Majora-92bf1922-00a
+    license: CC-BY-3.0
+    start: 56
+    end: 61
+    year: 2010
+    collection-title: Proceedings of the 9th Python in Science Conference
+    collection-doi: 10.25080/Majora-92bf1922-012
+    collection-type: proceedings
+    editors:
+      - given-names: Stéfan
+        name-particle: van der
+        family-names: Walt
+      - given-names: Jarrod
+        family-names: Millman
+    conference:
+      name: 9th Python in Science Conference (SciPy 2010)
+      city: Austin, TX
+      country: US
+      date-start: "2010-06-28"
+      date-end: "2010-07-03"
+    keywords:
+      - data structure
+      - statistics
+      - R
diff --git a/pandas/Dockerfile b/pandas/Dockerfile
new file mode 100644
index 0000000000000..dead3a494e52d
--- /dev/null
+++ b/pandas/Dockerfile
@@ -0,0 +1,17 @@
+FROM python:3.10.8
+WORKDIR /home/pandas
+
+RUN apt-get update && apt-get -y upgrade
+RUN apt-get install -y build-essential bash-completion
+
+# hdf5 needed for pytables installation
+# libgles2-mesa needed for pytest-qt
+RUN apt-get install -y libhdf5-dev libgles2-mesa-dev
+
+RUN python -m pip install --upgrade pip
+COPY requirements-dev.txt /tmp
+RUN python -m pip install -r /tmp/requirements-dev.txt
+RUN git config --global --add safe.directory /home/pandas
+
+ENV SHELL "/bin/bash"
+CMD ["/bin/bash"]
diff --git a/pandas/LICENSE b/pandas/LICENSE
new file mode 100644
index 0000000000000..c343da2ebe870
--- /dev/null
+++ b/pandas/LICENSE
@@ -0,0 +1,31 @@
+BSD 3-Clause License
+
+Copyright (c) 2008-2011, AQR Capital Management, LLC, Lambda Foundry, Inc. and PyData Development Team
+All rights reserved.
+
+Copyright (c) 2011-2025, Open source contributors.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+* Neither the name of the copyright holder nor the names of its
+  contributors may be used to endorse or promote products derived from
+  this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/pandas/LICENSES/BOTTLENECK_LICENCE b/pandas/LICENSES/BOTTLENECK_LICENCE
new file mode 100644
index 0000000000000..f4bdbb1647ee6
--- /dev/null
+++ b/pandas/LICENSES/BOTTLENECK_LICENCE
@@ -0,0 +1,25 @@
+Copyright (c) 2010-2019 Keith Goodman
+Copyright (c) 2019 Bottleneck Developers
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
\ No newline at end of file
diff --git a/pandas/LICENSES/DATEUTIL_LICENSE b/pandas/LICENSES/DATEUTIL_LICENSE
new file mode 100644
index 0000000000000..1e65815cf0b31
--- /dev/null
+++ b/pandas/LICENSES/DATEUTIL_LICENSE
@@ -0,0 +1,54 @@
+Copyright 2017- Paul Ganssle <paul@ganssle.io>
+Copyright 2017- dateutil contributors (see AUTHORS file)
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+
+The above license applies to all contributions after 2017-12-01, as well as
+all contributions that have been re-licensed (see AUTHORS file for the list of
+contributors who have re-licensed their code).
+--------------------------------------------------------------------------------
+dateutil - Extensions to the standard Python datetime module.
+
+Copyright (c) 2003-2011 - Gustavo Niemeyer <gustavo@niemeyer.net>
+Copyright (c) 2012-2014 - Tomi Pieviläinen <tomi.pievilainen@iki.fi>
+Copyright (c) 2014-2016 - Yaron de Leeuw <me@jarondl.net>
+Copyright (c) 2015-     - Paul Ganssle <paul@ganssle.io>
+Copyright (c) 2015-     - dateutil contributors (see AUTHORS file)
+
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright notice,
+      this list of conditions and the following disclaimer in the documentation
+      and/or other materials provided with the distribution.
+    * Neither the name of the copyright holder nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+The above BSD License Applies to all code, even that also covered by Apache 2.0.
\ No newline at end of file
diff --git a/pandas/LICENSES/HAVEN_LICENSE b/pandas/LICENSES/HAVEN_LICENSE
new file mode 100644
index 0000000000000..ce1b07b783e74
--- /dev/null
+++ b/pandas/LICENSES/HAVEN_LICENSE
@@ -0,0 +1,21 @@
+# MIT License
+
+Copyright (c) 2019 Hadley Wickham; RStudio; and Evan Miller
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/pandas/LICENSES/HAVEN_MIT b/pandas/LICENSES/HAVEN_MIT
new file mode 100644
index 0000000000000..b03d0e640627a
--- /dev/null
+++ b/pandas/LICENSES/HAVEN_MIT
@@ -0,0 +1,32 @@
+Based on http://opensource.org/licenses/MIT
+
+This is a template. Complete and ship as file LICENSE the following 2
+lines (only)
+
+YEAR:
+COPYRIGHT HOLDER:
+
+and specify as
+
+License: MIT + file LICENSE
+
+Copyright (c) <YEAR>, <COPYRIGHT HOLDER>
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/pandas/LICENSES/KLIB_LICENSE b/pandas/LICENSES/KLIB_LICENSE
new file mode 100644
index 0000000000000..2de13e402643b
--- /dev/null
+++ b/pandas/LICENSES/KLIB_LICENSE
@@ -0,0 +1,23 @@
+The MIT License
+
+Copyright (c) 2008-     Attractive Chaos <attractor@live.co.uk>
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
\ No newline at end of file
diff --git a/pandas/LICENSES/MUSL_LICENSE b/pandas/LICENSES/MUSL_LICENSE
new file mode 100644
index 0000000000000..5ff71b5f5a7fe
--- /dev/null
+++ b/pandas/LICENSES/MUSL_LICENSE
@@ -0,0 +1,193 @@
+musl as a whole is licensed under the following standard MIT license:
+
+----------------------------------------------------------------------
+Copyright © 2005-2020 Rich Felker, et al.
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+----------------------------------------------------------------------
+
+Authors/contributors include:
+
+A. Wilcox
+Ada Worcester
+Alex Dowad
+Alex Suykov
+Alexander Monakov
+Andre McCurdy
+Andrew Kelley
+Anthony G. Basile
+Aric Belsito
+Arvid Picciani
+Bartosz Brachaczek
+Benjamin Peterson
+Bobby Bingham
+Boris Brezillon
+Brent Cook
+Chris Spiegel
+Clément Vasseur
+Daniel Micay
+Daniel Sabogal
+Daurnimator
+David Carlier
+David Edelsohn
+Denys Vlasenko
+Dmitry Ivanov
+Dmitry V. Levin
+Drew DeVault
+Emil Renner Berthing
+Fangrui Song
+Felix Fietkau
+Felix Janda
+Gianluca Anzolin
+Hauke Mehrtens
+He X
+Hiltjo Posthuma
+Isaac Dunham
+Jaydeep Patil
+Jens Gustedt
+Jeremy Huntwork
+Jo-Philipp Wich
+Joakim Sindholt
+John Spencer
+Julien Ramseier
+Justin Cormack
+Kaarle Ritvanen
+Khem Raj
+Kylie McClain
+Leah Neukirchen
+Luca Barbato
+Luka Perkov
+M Farkas-Dyck (Strake)
+Mahesh Bodapati
+Markus Wichmann
+Masanori Ogino
+Michael Clark
+Michael Forney
+Mikhail Kremnyov
+Natanael Copa
+Nicholas J. Kain
+orc
+Pascal Cuoq
+Patrick Oppenlander
+Petr Hosek
+Petr Skocik
+Pierre Carrier
+Reini Urban
+Rich Felker
+Richard Pennington
+Ryan Fairfax
+Samuel Holland
+Segev Finer
+Shiz
+sin
+Solar Designer
+Stefan Kristiansson
+Stefan O'Rear
+Szabolcs Nagy
+Timo Teräs
+Trutz Behn
+Valentin Ochs
+Will Dietz
+William Haddon
+William Pitcock
+
+Portions of this software are derived from third-party works licensed
+under terms compatible with the above MIT license:
+
+The TRE regular expression implementation (src/regex/reg* and
+src/regex/tre*) is Copyright © 2001-2008 Ville Laurikari and licensed
+under a 2-clause BSD license (license text in the source files). The
+included version has been heavily modified by Rich Felker in 2012, in
+the interests of size, simplicity, and namespace cleanliness.
+
+Much of the math library code (src/math/* and src/complex/*) is
+Copyright © 1993,2004 Sun Microsystems or
+Copyright © 2003-2011 David Schultz or
+Copyright © 2003-2009 Steven G. Kargl or
+Copyright © 2003-2009 Bruce D. Evans or
+Copyright © 2008 Stephen L. Moshier or
+Copyright © 2017-2018 Arm Limited
+and labelled as such in comments in the individual source files. All
+have been licensed under extremely permissive terms.
+
+The ARM memcpy code (src/string/arm/memcpy.S) is Copyright © 2008
+The Android Open Source Project and is licensed under a two-clause BSD
+license. It was taken from Bionic libc, used on Android.
+
+The AArch64 memcpy and memset code (src/string/aarch64/*) are
+Copyright © 1999-2019, Arm Limited.
+
+The implementation of DES for crypt (src/crypt/crypt_des.c) is
+Copyright © 1994 David Burren. It is licensed under a BSD license.
+
+The implementation of blowfish crypt (src/crypt/crypt_blowfish.c) was
+originally written by Solar Designer and placed into the public
+domain. The code also comes with a fallback permissive license for use
+in jurisdictions that may not recognize the public domain.
+
+The smoothsort implementation (src/stdlib/qsort.c) is Copyright © 2011
+Valentin Ochs and is licensed under an MIT-style license.
+
+The x86_64 port was written by Nicholas J. Kain and is licensed under
+the standard MIT terms.
+
+The mips and microblaze ports were originally written by Richard
+Pennington for use in the ellcc project. The original code was adapted
+by Rich Felker for build system and code conventions during upstream
+integration. It is licensed under the standard MIT terms.
+
+The mips64 port was contributed by Imagination Technologies and is
+licensed under the standard MIT terms.
+
+The powerpc port was also originally written by Richard Pennington,
+and later supplemented and integrated by John Spencer. It is licensed
+under the standard MIT terms.
+
+All other files which have no copyright comments are original works
+produced specifically for use as part of this library, written either
+by Rich Felker, the main author of the library, or by one or more
+contibutors listed above. Details on authorship of individual files
+can be found in the git version control history of the project. The
+omission of copyright and license comments in each file is in the
+interest of source tree size.
+
+In addition, permission is hereby granted for all public header files
+(include/* and arch/*/bits/*) and crt files intended to be linked into
+applications (crt/*, ldso/dlstart.c, and arch/*/crt_arch.h) to omit
+the copyright notice and permission notice otherwise required by the
+license, and to use these files without any requirement of
+attribution. These files include substantial contributions from:
+
+Bobby Bingham
+John Spencer
+Nicholas J. Kain
+Rich Felker
+Richard Pennington
+Stefan Kristiansson
+Szabolcs Nagy
+
+all of whom have explicitly granted such permission.
+
+This file previously contained text expressing a belief that most of
+the files covered by the above exception were sufficiently trivial not
+to be subject to copyright, resulting in confusion over whether it
+negated the permissions granted in the license. In the spirit of
+permissive licensing, and of not having licensing issues being an
+obstacle to adoption, that text has been removed.
\ No newline at end of file
diff --git a/pandas/LICENSES/NUMPY_LICENSE b/pandas/LICENSES/NUMPY_LICENSE
new file mode 100644
index 0000000000000..f2d647bf0bc48
--- /dev/null
+++ b/pandas/LICENSES/NUMPY_LICENSE
@@ -0,0 +1,30 @@
+Copyright (c) 2005-2023, NumPy Developers.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+    * Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above
+       copyright notice, this list of conditions and the following
+       disclaimer in the documentation and/or other materials provided
+       with the distribution.
+
+    * Neither the name of the NumPy Developers nor the names of any
+       contributors may be used to endorse or promote products derived
+       from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
\ No newline at end of file
diff --git a/pandas/LICENSES/PACKAGING_LICENSE b/pandas/LICENSES/PACKAGING_LICENSE
new file mode 100644
index 0000000000000..2bfcd5297c470
--- /dev/null
+++ b/pandas/LICENSES/PACKAGING_LICENSE
@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+
+Copyright (c) Donald Stufft and individual contributors.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    1. Redistributions of source code must retain the above copyright notice,
+       this list of conditions and the following disclaimer.
+
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
\ No newline at end of file
diff --git a/pandas/LICENSES/PSF_LICENSE b/pandas/LICENSES/PSF_LICENSE
new file mode 100644
index 0000000000000..f26bcf4d2de6e
--- /dev/null
+++ b/pandas/LICENSES/PSF_LICENSE
@@ -0,0 +1,279 @@
+A. HISTORY OF THE SOFTWARE
+==========================
+
+Python was created in the early 1990s by Guido van Rossum at Stichting
+Mathematisch Centrum (CWI, see https://www.cwi.nl) in the Netherlands
+as a successor of a language called ABC.  Guido remains Python's
+principal author, although it includes many contributions from others.
+
+In 1995, Guido continued his work on Python at the Corporation for
+National Research Initiatives (CNRI, see https://www.cnri.reston.va.us)
+in Reston, Virginia where he released several versions of the
+software.
+
+In May 2000, Guido and the Python core development team moved to
+BeOpen.com to form the BeOpen PythonLabs team.  In October of the same
+year, the PythonLabs team moved to Digital Creations, which became
+Zope Corporation.  In 2001, the Python Software Foundation (PSF, see
+https://www.python.org/psf/) was formed, a non-profit organization
+created specifically to own Python-related Intellectual Property.
+Zope Corporation was a sponsoring member of the PSF.
+
+All Python releases are Open Source (see https://opensource.org for
+the Open Source Definition).  Historically, most, but not all, Python
+releases have also been GPL-compatible; the table below summarizes
+the various releases.
+
+    Release         Derived     Year        Owner       GPL-
+                    from                                compatible? (1)
+
+    0.9.0 thru 1.2              1991-1995   CWI         yes
+    1.3 thru 1.5.2  1.2         1995-1999   CNRI        yes
+    1.6             1.5.2       2000        CNRI        no
+    2.0             1.6         2000        BeOpen.com  no
+    1.6.1           1.6         2001        CNRI        yes (2)
+    2.1             2.0+1.6.1   2001        PSF         no
+    2.0.1           2.0+1.6.1   2001        PSF         yes
+    2.1.1           2.1+2.0.1   2001        PSF         yes
+    2.1.2           2.1.1       2002        PSF         yes
+    2.1.3           2.1.2       2002        PSF         yes
+    2.2 and above   2.1.1       2001-now    PSF         yes
+
+Footnotes:
+
+(1) GPL-compatible doesn't mean that we're distributing Python under
+    the GPL.  All Python licenses, unlike the GPL, let you distribute
+    a modified version without making your changes open source.  The
+    GPL-compatible licenses make it possible to combine Python with
+    other software that is released under the GPL; the others don't.
+
+(2) According to Richard Stallman, 1.6.1 is not GPL-compatible,
+    because its license has a choice of law clause.  According to
+    CNRI, however, Stallman's lawyer has told CNRI's lawyer that 1.6.1
+    is "not incompatible" with the GPL.
+
+Thanks to the many outside volunteers who have worked under Guido's
+direction to make these releases possible.
+
+
+B. TERMS AND CONDITIONS FOR ACCESSING OR OTHERWISE USING PYTHON
+===============================================================
+
+Python software and documentation are licensed under the
+Python Software Foundation License Version 2.
+
+Starting with Python 3.8.6, examples, recipes, and other code in
+the documentation are dual licensed under the PSF License Version 2
+and the Zero-Clause BSD license.
+
+Some software incorporated into Python is under different licenses.
+The licenses are listed with code falling under that license.
+
+
+PYTHON SOFTWARE FOUNDATION LICENSE VERSION 2
+--------------------------------------------
+
+1. This LICENSE AGREEMENT is between the Python Software Foundation
+("PSF"), and the Individual or Organization ("Licensee") accessing and
+otherwise using this software ("Python") in source or binary form and
+its associated documentation.
+
+2. Subject to the terms and conditions of this License Agreement, PSF hereby
+grants Licensee a nonexclusive, royalty-free, world-wide license to reproduce,
+analyze, test, perform and/or display publicly, prepare derivative works,
+distribute, and otherwise use Python alone or in any derivative version,
+provided, however, that PSF's License Agreement and PSF's notice of copyright,
+i.e., "Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010,
+2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023 Python Software Foundation;
+All Rights Reserved" are retained in Python alone or in any derivative version
+prepared by Licensee.
+
+3. In the event Licensee prepares a derivative work that is based on
+or incorporates Python or any part thereof, and wants to make
+the derivative work available to others as provided herein, then
+Licensee hereby agrees to include in any such work a brief summary of
+the changes made to Python.
+
+4. PSF is making Python available to Licensee on an "AS IS"
+basis.  PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
+IMPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND
+DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
+FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON WILL NOT
+INFRINGE ANY THIRD PARTY RIGHTS.
+
+5. PSF SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF PYTHON
+FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS AS
+A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING PYTHON,
+OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.
+
+6. This License Agreement will automatically terminate upon a material
+breach of its terms and conditions.
+
+7. Nothing in this License Agreement shall be deemed to create any
+relationship of agency, partnership, or joint venture between PSF and
+Licensee.  This License Agreement does not grant permission to use PSF
+trademarks or trade name in a trademark sense to endorse or promote
+products or services of Licensee, or any third party.
+
+8. By copying, installing or otherwise using Python, Licensee
+agrees to be bound by the terms and conditions of this License
+Agreement.
+
+
+BEOPEN.COM LICENSE AGREEMENT FOR PYTHON 2.0
+-------------------------------------------
+
+BEOPEN PYTHON OPEN SOURCE LICENSE AGREEMENT VERSION 1
+
+1. This LICENSE AGREEMENT is between BeOpen.com ("BeOpen"), having an
+office at 160 Saratoga Avenue, Santa Clara, CA 95051, and the
+Individual or Organization ("Licensee") accessing and otherwise using
+this software in source or binary form and its associated
+documentation ("the Software").
+
+2. Subject to the terms and conditions of this BeOpen Python License
+Agreement, BeOpen hereby grants Licensee a non-exclusive,
+royalty-free, world-wide license to reproduce, analyze, test, perform
+and/or display publicly, prepare derivative works, distribute, and
+otherwise use the Software alone or in any derivative version,
+provided, however, that the BeOpen Python License is retained in the
+Software, alone or in any derivative version prepared by Licensee.
+
+3. BeOpen is making the Software available to Licensee on an "AS IS"
+basis.  BEOPEN MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
+IMPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, BEOPEN MAKES NO AND
+DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
+FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF THE SOFTWARE WILL NOT
+INFRINGE ANY THIRD PARTY RIGHTS.
+
+4. BEOPEN SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF THE
+SOFTWARE FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS
+AS A RESULT OF USING, MODIFYING OR DISTRIBUTING THE SOFTWARE, OR ANY
+DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.
+
+5. This License Agreement will automatically terminate upon a material
+breach of its terms and conditions.
+
+6. This License Agreement shall be governed by and interpreted in all
+respects by the law of the State of California, excluding conflict of
+law provisions.  Nothing in this License Agreement shall be deemed to
+create any relationship of agency, partnership, or joint venture
+between BeOpen and Licensee.  This License Agreement does not grant
+permission to use BeOpen trademarks or trade names in a trademark
+sense to endorse or promote products or services of Licensee, or any
+third party.  As an exception, the "BeOpen Python" logos available at
+http://www.pythonlabs.com/logos.html may be used according to the
+permissions granted on that web page.
+
+7. By copying, installing or otherwise using the software, Licensee
+agrees to be bound by the terms and conditions of this License
+Agreement.
+
+
+CNRI LICENSE AGREEMENT FOR PYTHON 1.6.1
+---------------------------------------
+
+1. This LICENSE AGREEMENT is between the Corporation for National
+Research Initiatives, having an office at 1895 Preston White Drive,
+Reston, VA 20191 ("CNRI"), and the Individual or Organization
+("Licensee") accessing and otherwise using Python 1.6.1 software in
+source or binary form and its associated documentation.
+
+2. Subject to the terms and conditions of this License Agreement, CNRI
+hereby grants Licensee a nonexclusive, royalty-free, world-wide
+license to reproduce, analyze, test, perform and/or display publicly,
+prepare derivative works, distribute, and otherwise use Python 1.6.1
+alone or in any derivative version, provided, however, that CNRI's
+License Agreement and CNRI's notice of copyright, i.e., "Copyright (c)
+1995-2001 Corporation for National Research Initiatives; All Rights
+Reserved" are retained in Python 1.6.1 alone or in any derivative
+version prepared by Licensee.  Alternately, in lieu of CNRI's License
+Agreement, Licensee may substitute the following text (omitting the
+quotes): "Python 1.6.1 is made available subject to the terms and
+conditions in CNRI's License Agreement.  This Agreement together with
+Python 1.6.1 may be located on the internet using the following
+unique, persistent identifier (known as a handle): 1895.22/1013.  This
+Agreement may also be obtained from a proxy server on the internet
+using the following URL: http://hdl.handle.net/1895.22/1013".
+
+3. In the event Licensee prepares a derivative work that is based on
+or incorporates Python 1.6.1 or any part thereof, and wants to make
+the derivative work available to others as provided herein, then
+Licensee hereby agrees to include in any such work a brief summary of
+the changes made to Python 1.6.1.
+
+4. CNRI is making Python 1.6.1 available to Licensee on an "AS IS"
+basis.  CNRI MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
+IMPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, CNRI MAKES NO AND
+DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
+FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON 1.6.1 WILL NOT
+INFRINGE ANY THIRD PARTY RIGHTS.
+
+5. CNRI SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF PYTHON
+1.6.1 FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS AS
+A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING PYTHON 1.6.1,
+OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.
+
+6. This License Agreement will automatically terminate upon a material
+breach of its terms and conditions.
+
+7. This License Agreement shall be governed by the federal
+intellectual property law of the United States, including without
+limitation the federal copyright law, and, to the extent such
+U.S. federal law does not apply, by the law of the Commonwealth of
+Virginia, excluding Virginia's conflict of law provisions.
+Notwithstanding the foregoing, with regard to derivative works based
+on Python 1.6.1 that incorporate non-separable material that was
+previously distributed under the GNU General Public License (GPL), the
+law of the Commonwealth of Virginia shall govern this License
+Agreement only as to issues arising under or with respect to
+Paragraphs 4, 5, and 7 of this License Agreement.  Nothing in this
+License Agreement shall be deemed to create any relationship of
+agency, partnership, or joint venture between CNRI and Licensee.  This
+License Agreement does not grant permission to use CNRI trademarks or
+trade name in a trademark sense to endorse or promote products or
+services of Licensee, or any third party.
+
+8. By clicking on the "ACCEPT" button where indicated, or by copying,
+installing or otherwise using Python 1.6.1, Licensee agrees to be
+bound by the terms and conditions of this License Agreement.
+
+        ACCEPT
+
+
+CWI LICENSE AGREEMENT FOR PYTHON 0.9.0 THROUGH 1.2
+--------------------------------------------------
+
+Copyright (c) 1991 - 1995, Stichting Mathematisch Centrum Amsterdam,
+The Netherlands.  All rights reserved.
+
+Permission to use, copy, modify, and distribute this software and its
+documentation for any purpose and without fee is hereby granted,
+provided that the above copyright notice appear in all copies and that
+both that copyright notice and this permission notice appear in
+supporting documentation, and that the name of Stichting Mathematisch
+Centrum or CWI not be used in advertising or publicity pertaining to
+distribution of the software without specific, written prior
+permission.
+
+STICHTING MATHEMATISCH CENTRUM DISCLAIMS ALL WARRANTIES WITH REGARD TO
+THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
+FITNESS, IN NO EVENT SHALL STICHTING MATHEMATISCH CENTRUM BE LIABLE
+FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
+OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+
+ZERO-CLAUSE BSD LICENSE FOR CODE IN THE PYTHON DOCUMENTATION
+----------------------------------------------------------------------
+
+Permission to use, copy, modify, and/or distribute this software for any
+purpose with or without fee is hereby granted.
+
+THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH
+REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
+AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT,
+INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
+OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+PERFORMANCE OF THIS SOFTWARE.
diff --git a/pandas/LICENSES/PYPERCLIP_LICENSE b/pandas/LICENSES/PYPERCLIP_LICENSE
new file mode 100644
index 0000000000000..07cc746cd5ad6
--- /dev/null
+++ b/pandas/LICENSES/PYPERCLIP_LICENSE
@@ -0,0 +1,27 @@
+Copyright (c) 2014, Al Sweigart
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+* Neither the name of the {organization} nor the names of its
+  contributors may be used to endorse or promote products derived from
+  this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
\ No newline at end of file
diff --git a/pandas/LICENSES/PYUPGRADE_LICENSE b/pandas/LICENSES/PYUPGRADE_LICENSE
new file mode 100644
index 0000000000000..edeac73dade04
--- /dev/null
+++ b/pandas/LICENSES/PYUPGRADE_LICENSE
@@ -0,0 +1,19 @@
+Copyright (c) 2017 Anthony Sottile
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
\ No newline at end of file
diff --git a/pandas/LICENSES/SAS7BDAT_LICENSE b/pandas/LICENSES/SAS7BDAT_LICENSE
new file mode 100644
index 0000000000000..94b7fe934e85c
--- /dev/null
+++ b/pandas/LICENSES/SAS7BDAT_LICENSE
@@ -0,0 +1,19 @@
+Copyright (c) 2015-2019 Jared Hobbs
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
\ No newline at end of file
diff --git a/pandas/LICENSES/ULTRAJSON_LICENSE b/pandas/LICENSES/ULTRAJSON_LICENSE
new file mode 100644
index 0000000000000..58fc9dfc0a35b
--- /dev/null
+++ b/pandas/LICENSES/ULTRAJSON_LICENSE
@@ -0,0 +1,114 @@
+Developed by ESN, an Electronic Arts Inc. studio.
+Copyright (c) 2014, Electronic Arts Inc.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+* Neither the name of ESN, Electronic Arts Inc. nor the
+names of its contributors may be used to endorse or promote products
+derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ELECTRONIC ARTS INC. BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+----
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+
+  Copyright 2005, 2006, 2007
+  Nick Galbreath -- nickg [at] modp [dot] com
+  All rights reserved.
+
+  Redistribution and use in source and binary forms, with or without
+  modification, are permitted provided that the following conditions are
+  met:
+
+    Redistributions of source code must retain the above copyright
+    notice, this list of conditions and the following disclaimer.
+
+    Redistributions in binary form must reproduce the above copyright
+    notice, this list of conditions and the following disclaimer in the
+    documentation and/or other materials provided with the distribution.
+
+    Neither the name of the modp.com nor the names of its
+    contributors may be used to endorse or promote products derived from
+    this software without specific prior written permission.
+
+  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+  A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+  OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+  SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+  LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+  DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+  THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+  OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+  This is the standard "new" BSD license:
+  http://www.opensource.org/licenses/bsd-license.php
+
+https://github.com/client9/stringencoders/blob/cfd5c1507325ae497ea9bacdacba12c0ffd79d30/COPYING
+
+----
+
+Numeric decoder derived from from TCL library
+https://opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+ * Copyright (c) 1988-1993 The Regents of the University of California.
+ * Copyright (c) 1994 Sun Microsystems, Inc.
+
+  This software is copyrighted by the Regents of the University of
+  California, Sun Microsystems, Inc., Scriptics Corporation, ActiveState
+  Corporation and other parties.  The following terms apply to all files
+  associated with the software unless explicitly disclaimed in
+  individual files.
+
+  The authors hereby grant permission to use, copy, modify, distribute,
+  and license this software and its documentation for any purpose, provided
+  that existing copyright notices are retained in all copies and that this
+  notice is included verbatim in any distributions. No written agreement,
+  license, or royalty fee is required for any of the authorized uses.
+  Modifications to this software may be copyrighted by their authors
+  and need not follow the licensing terms described here, provided that
+  the new terms are clearly indicated on the first page of each file where
+  they apply.
+
+  IN NO EVENT SHALL THE AUTHORS OR DISTRIBUTORS BE LIABLE TO ANY PARTY
+  FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES
+  ARISING OUT OF THE USE OF THIS SOFTWARE, ITS DOCUMENTATION, OR ANY
+  DERIVATIVES THEREOF, EVEN IF THE AUTHORS HAVE BEEN ADVISED OF THE
+  POSSIBILITY OF SUCH DAMAGE.
+
+  THE AUTHORS AND DISTRIBUTORS SPECIFICALLY DISCLAIM ANY WARRANTIES,
+  INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY,
+  FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.  THIS SOFTWARE
+  IS PROVIDED ON AN "AS IS" BASIS, AND THE AUTHORS AND DISTRIBUTORS HAVE
+  NO OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR
+  MODIFICATIONS.
+
+  GOVERNMENT USE: If you are acquiring this software on behalf of the
+  U.S. government, the Government shall have only "Restricted Rights"
+  in the software and related documentation as defined in the Federal
+  Acquisition Regulations (FARs) in Clause 52.227.19 (c) (2).  If you
+  are acquiring the software on behalf of the Department of Defense, the
+  software shall be classified as "Commercial Computer Software" and the
+  Government shall have only "Restricted Rights" as defined in Clause
+  252.227-7013 (c) (1) of DFARs.  Notwithstanding the foregoing, the
+  authors grant the U.S. Government and others acting in its behalf
+  permission to use and distribute the software in accordance with the
+  terms specified in this license.
\ No newline at end of file
diff --git a/pandas/MANIFEST.in b/pandas/MANIFEST.in
new file mode 100644
index 0000000000000..c59151f340545
--- /dev/null
+++ b/pandas/MANIFEST.in
@@ -0,0 +1,69 @@
+graft doc
+prune doc/build
+
+graft LICENSES
+
+graft pandas
+
+global-exclude *.bz2
+global-exclude *.csv
+global-exclude *.data
+global-exclude *.dta
+global-exclude *.feather
+global-exclude *.tar
+global-exclude *.gz
+global-exclude *.h5
+global-exclude *.html
+global-exclude *.json
+global-exclude *.jsonl
+global-exclude *.kml
+global-exclude *.msgpack
+global-exclude *.pdf
+global-exclude *.parquet
+global-exclude *.pickle
+global-exclude *.pkl
+global-exclude *.png
+global-exclude *.pptx
+global-exclude *.ods
+global-exclude *.odt
+global-exclude *.orc
+global-exclude *.sas7bdat
+global-exclude *.sav
+global-exclude *.so
+global-exclude *.txt
+global-exclude *.xls
+global-exclude *.xlsb
+global-exclude *.xlsm
+global-exclude *.xlsx
+global-exclude *.xpt
+global-exclude *.cpt
+global-exclude *.xml
+global-exclude *.xsl
+global-exclude *.xz
+global-exclude *.zip
+global-exclude *.zst
+global-exclude *~
+global-exclude .DS_Store
+global-exclude .git*
+global-exclude \#*
+
+global-exclude *.c
+global-exclude *.cpp
+global-exclude *.h
+
+global-exclude *.py[ocd]
+global-exclude *.pxi
+
+# GH 39321
+# csv_dir_path fixture checks the existence of the directory
+# exclude the whole directory to avoid running related tests in sdist
+prune pandas/tests/io/parser/data
+
+# Selectively re-add *.cxx files that were excluded above
+graft pandas/_libs/src
+graft pandas/_libs/include
+
+# Include cibw script in sdist since it's needed for building wheels
+include scripts/cibw_before_build.sh
+include scripts/cibw_before_build_windows.sh
+include scripts/cibw_before_test_windows.sh
diff --git a/pandas/README.md b/pandas/README.md
new file mode 100644
index 0000000000000..1a273fdb896c5
--- /dev/null
+++ b/pandas/README.md
@@ -0,0 +1,190 @@
+<picture align="center">
+  <source media="(prefers-color-scheme: dark)" srcset="https://pandas.pydata.org/static/img/pandas_white.svg">
+  <img alt="Pandas Logo" src="https://pandas.pydata.org/static/img/pandas.svg">
+</picture>
+
+-----------------
+
+# pandas: powerful Python data analysis toolkit
+
+| | |
+| --- | --- |
+| Testing | [![CI - Test](https://github.com/pandas-dev/pandas/actions/workflows/unit-tests.yml/badge.svg)](https://github.com/pandas-dev/pandas/actions/workflows/unit-tests.yml) [![Coverage](https://codecov.io/github/pandas-dev/pandas/coverage.svg?branch=main)](https://codecov.io/gh/pandas-dev/pandas) |
+| Package | [![PyPI Latest Release](https://img.shields.io/pypi/v/pandas.svg)](https://pypi.org/project/pandas/) [![PyPI Downloads](https://img.shields.io/pypi/dm/pandas.svg?label=PyPI%20downloads)](https://pypi.org/project/pandas/) [![Conda Latest Release](https://anaconda.org/conda-forge/pandas/badges/version.svg)](https://anaconda.org/conda-forge/pandas) [![Conda Downloads](https://img.shields.io/conda/dn/conda-forge/pandas.svg?label=Conda%20downloads)](https://anaconda.org/conda-forge/pandas) |
+| Meta | [![Powered by NumFOCUS](https://img.shields.io/badge/powered%20by-NumFOCUS-orange.svg?style=flat&colorA=E1523D&colorB=007D8A)](https://numfocus.org) [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.3509134.svg)](https://doi.org/10.5281/zenodo.3509134) [![License - BSD 3-Clause](https://img.shields.io/pypi/l/pandas.svg)](https://github.com/pandas-dev/pandas/blob/main/LICENSE) [![Slack](https://img.shields.io/badge/join_Slack-information-brightgreen.svg?logo=slack)](https://pandas.pydata.org/docs/dev/development/community.html?highlight=slack#community-slack) |
+
+
+## What is it?
+
+**pandas** is a Python package that provides fast, flexible, and expressive data
+structures designed to make working with "relational" or "labeled" data both
+easy and intuitive. It aims to be the fundamental high-level building block for
+doing practical, **real world** data analysis in Python. Additionally, it has
+the broader goal of becoming **the most powerful and flexible open source data
+analysis / manipulation tool available in any language**. It is already well on
+its way towards this goal.
+
+## Table of Contents
+
+- [Main Features](#main-features)
+- [Where to get it](#where-to-get-it)
+- [Dependencies](#dependencies)
+- [Installation from sources](#installation-from-sources)
+- [License](#license)
+- [Documentation](#documentation)
+- [Background](#background)
+- [Getting Help](#getting-help)
+- [Discussion and Development](#discussion-and-development)
+- [Contributing to pandas](#contributing-to-pandas)
+
+## Main Features
+Here are just a few of the things that pandas does well:
+
+  - Easy handling of [**missing data**][missing-data] (represented as
+    `NaN`, `NA`, or `NaT`) in floating point as well as non-floating point data
+  - Size mutability: columns can be [**inserted and
+    deleted**][insertion-deletion] from DataFrame and higher dimensional
+    objects
+  - Automatic and explicit [**data alignment**][alignment]: objects can
+    be explicitly aligned to a set of labels, or the user can simply
+    ignore the labels and let `Series`, `DataFrame`, etc. automatically
+    align the data for you in computations
+  - Powerful, flexible [**group by**][groupby] functionality to perform
+    split-apply-combine operations on data sets, for both aggregating
+    and transforming data
+  - Make it [**easy to convert**][conversion] ragged,
+    differently-indexed data in other Python and NumPy data structures
+    into DataFrame objects
+  - Intelligent label-based [**slicing**][slicing], [**fancy
+    indexing**][fancy-indexing], and [**subsetting**][subsetting] of
+    large data sets
+  - Intuitive [**merging**][merging] and [**joining**][joining] data
+    sets
+  - Flexible [**reshaping**][reshape] and [**pivoting**][pivot-table] of
+    data sets
+  - [**Hierarchical**][mi] labeling of axes (possible to have multiple
+    labels per tick)
+  - Robust IO tools for loading data from [**flat files**][flat-files]
+    (CSV and delimited), [**Excel files**][excel], [**databases**][db],
+    and saving/loading data from the ultrafast [**HDF5 format**][hdfstore]
+  - [**Time series**][timeseries]-specific functionality: date range
+    generation and frequency conversion, moving window statistics,
+    date shifting and lagging
+
+
+   [missing-data]: https://pandas.pydata.org/pandas-docs/stable/user_guide/missing_data.html
+   [insertion-deletion]: https://pandas.pydata.org/pandas-docs/stable/user_guide/dsintro.html#column-selection-addition-deletion
+   [alignment]: https://pandas.pydata.org/pandas-docs/stable/user_guide/dsintro.html?highlight=alignment#intro-to-data-structures
+   [groupby]: https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html#group-by-split-apply-combine
+   [conversion]: https://pandas.pydata.org/pandas-docs/stable/user_guide/dsintro.html#dataframe
+   [slicing]: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#slicing-ranges
+   [fancy-indexing]: https://pandas.pydata.org/pandas-docs/stable/user_guide/advanced.html#advanced
+   [subsetting]: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#boolean-indexing
+   [merging]: https://pandas.pydata.org/pandas-docs/stable/user_guide/merging.html#database-style-dataframe-or-named-series-joining-merging
+   [joining]: https://pandas.pydata.org/pandas-docs/stable/user_guide/merging.html#joining-on-index
+   [reshape]: https://pandas.pydata.org/pandas-docs/stable/user_guide/reshaping.html
+   [pivot-table]: https://pandas.pydata.org/pandas-docs/stable/user_guide/reshaping.html
+   [mi]: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#hierarchical-indexing-multiindex
+   [flat-files]: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#csv-text-files
+   [excel]: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#excel-files
+   [db]: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#sql-queries
+   [hdfstore]: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#hdf5-pytables
+   [timeseries]: https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#time-series-date-functionality
+
+## Where to get it
+The source code is currently hosted on GitHub at:
+https://github.com/pandas-dev/pandas
+
+Binary installers for the latest released version are available at the [Python
+Package Index (PyPI)](https://pypi.org/project/pandas) and on [Conda](https://anaconda.org/conda-forge/pandas).
+
+```sh
+# conda
+conda install -c conda-forge pandas
+```
+
+```sh
+# or PyPI
+pip install pandas
+```
+
+The list of changes to pandas between each release can be found
+[here](https://pandas.pydata.org/pandas-docs/stable/whatsnew/index.html). For full
+details, see the commit logs at https://github.com/pandas-dev/pandas.
+
+## Dependencies
+- [NumPy - Adds support for large, multi-dimensional arrays, matrices and high-level mathematical functions to operate on these arrays](https://www.numpy.org)
+- [python-dateutil - Provides powerful extensions to the standard datetime module](https://dateutil.readthedocs.io/en/stable/index.html)
+- [pytz - Brings the Olson tz database into Python which allows accurate and cross platform timezone calculations](https://github.com/stub42/pytz)
+
+See the [full installation instructions](https://pandas.pydata.org/pandas-docs/stable/install.html#dependencies) for minimum supported versions of required, recommended and optional dependencies.
+
+## Installation from sources
+To install pandas from source you need [Cython](https://cython.org/) in addition to the normal
+dependencies above. Cython can be installed from PyPI:
+
+```sh
+pip install cython
+```
+
+In the `pandas` directory (same one where you found this file after
+cloning the git repo), execute:
+
+```sh
+pip install .
+```
+
+or for installing in [development mode](https://pip.pypa.io/en/latest/cli/pip_install/#install-editable):
+
+
+```sh
+python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true
+```
+
+See the full instructions for [installing from source](https://pandas.pydata.org/docs/dev/development/contributing_environment.html).
+
+## License
+[BSD 3](LICENSE)
+
+## Documentation
+The official documentation is hosted on [PyData.org](https://pandas.pydata.org/pandas-docs/stable/).
+
+## Background
+Work on ``pandas`` started at [AQR](https://www.aqr.com/) (a quantitative hedge fund) in 2008 and
+has been under active development since then.
+
+## Getting Help
+
+For usage questions, the best place to go to is [StackOverflow](https://stackoverflow.com/questions/tagged/pandas).
+Further, general questions and discussions can also take place on the [pydata mailing list](https://groups.google.com/forum/?fromgroups#!forum/pydata).
+
+## Discussion and Development
+Most development discussions take place on GitHub in this repo, via the [GitHub issue tracker](https://github.com/pandas-dev/pandas/issues).
+
+Further, the [pandas-dev mailing list](https://mail.python.org/mailman/listinfo/pandas-dev) can also be used for specialized discussions or design issues, and a [Slack channel](https://pandas.pydata.org/docs/dev/development/community.html?highlight=slack#community-slack) is available for quick development related questions.
+
+There are also frequent [community meetings](https://pandas.pydata.org/docs/dev/development/community.html#community-meeting) for project maintainers open to the community as well as monthly [new contributor meetings](https://pandas.pydata.org/docs/dev/development/community.html#new-contributor-meeting) to help support new contributors.
+
+Additional information on the communication channels can be found on the [contributor community](https://pandas.pydata.org/docs/development/community.html) page.
+
+## Contributing to pandas
+
+[![Open Source Helpers](https://www.codetriage.com/pandas-dev/pandas/badges/users.svg)](https://www.codetriage.com/pandas-dev/pandas)
+
+All contributions, bug reports, bug fixes, documentation improvements, enhancements, and ideas are welcome.
+
+A detailed overview on how to contribute can be found in the **[contributing guide](https://pandas.pydata.org/docs/dev/development/contributing.html)**.
+
+If you are simply looking to start working with the pandas codebase, navigate to the [GitHub "issues" tab](https://github.com/pandas-dev/pandas/issues) and start looking through interesting issues. There are a number of issues listed under [Docs](https://github.com/pandas-dev/pandas/issues?labels=Docs&sort=updated&state=open) and [good first issue](https://github.com/pandas-dev/pandas/issues?labels=good+first+issue&sort=updated&state=open) where you could start out.
+
+You can also triage issues which may include reproducing bug reports, or asking for vital information such as version numbers or reproduction instructions. If you would like to start triaging issues, one easy way to get started is to [subscribe to pandas on CodeTriage](https://www.codetriage.com/pandas-dev/pandas).
+
+Or maybe through using pandas you have an idea of your own or are looking for something in the documentation and thinking ‘this can be improved’...you can do something about it!
+
+Feel free to ask questions on the [mailing list](https://groups.google.com/forum/?fromgroups#!forum/pydata) or on [Slack](https://pandas.pydata.org/docs/dev/development/community.html?highlight=slack#community-slack).
+
+As contributors and maintainers to this project, you are expected to abide by pandas' code of conduct. More information can be found at: [Contributor Code of Conduct](https://github.com/pandas-dev/.github/blob/master/CODE_OF_CONDUCT.md)
+
+<hr>
+
+[Go to Top](#table-of-contents)
diff --git a/pandas/__init__.py b/pandas/__init__.py
index c570fb8d70204..7a98b2207b6c6 100644
--- a/pandas/__init__.py
+++ b/pandas/__init__.py
@@ -20,8 +20,8 @@
 
 try:
     # numpy compat
-    from pandas.compat import (
-        is_numpy_dev as _is_numpy_dev,  # pyright: ignore[reportUnusedImport] # noqa: F401
+    from pandas.compat import (  # pyright: ignore[reportUnusedImport] # noqa: F401
+        is_numpy_dev as _is_numpy_dev,
     )
 except ImportError as _err:  # pragma: no cover
     _module = _err.name
@@ -33,151 +33,141 @@
     ) from _err
 
 from pandas._config import (
-    get_option,
-    set_option,
-    reset_option,
     describe_option,
+    get_option,
     option_context,
     options,
+    reset_option,
+    set_option,
 )
 
-# let init-time option registration happen
-import pandas.core.config_init  # pyright: ignore[reportUnusedImport] # noqa: F401
+from pandas.util._print_versions import show_versions
+from pandas.util._tester import test
+
+from pandas.core.dtypes.dtypes import SparseDtype
 
-from pandas.core.api import (
-    # dtype
+from pandas import (
+    api,
+    arrays,
+    errors,
+    io,
+    plotting,
+    testing,
+    tseries,
+)
+from pandas.core.api import (  # dtype; missing; indexes; tseries; conversion; misc
+    NA,
     ArrowDtype,
+    BooleanDtype,
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    DateOffset,
+    DatetimeIndex,
+    DatetimeTZDtype,
+    Flags,
+    Float32Dtype,
+    Float64Dtype,
+    Grouper,
+    Index,
+    IndexSlice,
     Int8Dtype,
     Int16Dtype,
     Int32Dtype,
     Int64Dtype,
+    Interval,
+    IntervalDtype,
+    IntervalIndex,
+    MultiIndex,
+    NamedAgg,
+    NaT,
+    Period,
+    PeriodDtype,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    StringDtype,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
     UInt8Dtype,
     UInt16Dtype,
     UInt32Dtype,
     UInt64Dtype,
-    Float32Dtype,
-    Float64Dtype,
-    CategoricalDtype,
-    PeriodDtype,
-    IntervalDtype,
-    DatetimeTZDtype,
-    StringDtype,
-    BooleanDtype,
-    # missing
-    NA,
+    array,
+    bdate_range,
+    date_range,
+    factorize,
+    interval_range,
     isna,
     isnull,
     notna,
     notnull,
-    # indexes
-    Index,
-    CategoricalIndex,
-    RangeIndex,
-    MultiIndex,
-    IntervalIndex,
-    TimedeltaIndex,
-    DatetimeIndex,
-    PeriodIndex,
-    IndexSlice,
-    # tseries
-    NaT,
-    Period,
     period_range,
-    Timedelta,
+    set_eng_float_format,
     timedelta_range,
-    Timestamp,
-    date_range,
-    bdate_range,
-    Interval,
-    interval_range,
-    DateOffset,
-    # conversion
-    to_numeric,
     to_datetime,
+    to_numeric,
     to_timedelta,
-    # misc
-    Flags,
-    Grouper,
-    factorize,
     unique,
-    NamedAgg,
-    array,
-    Categorical,
-    set_eng_float_format,
-    Series,
-    DataFrame,
 )
-
-from pandas.core.dtypes.dtypes import SparseDtype
-
-from pandas.tseries.api import infer_freq
-from pandas.tseries import offsets
-
 from pandas.core.computation.api import eval
 
+# let init-time option registration happen
+import pandas.core.config_init  # pyright: ignore[reportUnusedImport] # noqa: F401
 from pandas.core.reshape.api import (
     concat,
+    crosstab,
+    cut,
+    from_dummies,
+    get_dummies,
     lreshape,
     melt,
-    wide_to_long,
     merge,
     merge_asof,
     merge_ordered,
-    crosstab,
     pivot,
     pivot_table,
-    get_dummies,
-    from_dummies,
-    cut,
     qcut,
+    wide_to_long,
 )
 
-from pandas import api, arrays, errors, io, plotting, tseries
-from pandas import testing
-from pandas.util._print_versions import show_versions
-
-from pandas.io.api import (
-    # excel
+from pandas.io.api import (  # excel; parsers; pickle; pytables; sql; misc
     ExcelFile,
     ExcelWriter,
-    read_excel,
-    # parsers
-    read_csv,
-    read_fwf,
-    read_table,
-    # pickle
-    read_pickle,
-    to_pickle,
-    # pytables
     HDFStore,
-    read_hdf,
-    # sql
-    read_sql,
-    read_sql_query,
-    read_sql_table,
-    # misc
     read_clipboard,
-    read_parquet,
-    read_orc,
+    read_csv,
+    read_excel,
     read_feather,
+    read_fwf,
+    read_hdf,
     read_html,
-    read_xml,
     read_json,
-    read_stata,
+    read_orc,
+    read_parquet,
+    read_pickle,
     read_sas,
     read_spss,
+    read_sql,
+    read_sql_query,
+    read_sql_table,
+    read_stata,
+    read_table,
+    read_xml,
+    to_pickle,
 )
-
 from pandas.io.json._normalize import json_normalize
-
-from pandas.util._tester import test
+from pandas.tseries import offsets
+from pandas.tseries.api import infer_freq
 
 # use the closest tagged version if possible
 _built_with_meson = False
 try:
     from pandas._version_meson import (  # pyright: ignore [reportMissingImports]
-        __version__,
         __git_version__,
+        __version__,
     )
 
     _built_with_meson = True
diff --git a/pandas/asv_bench/asv.conf.json b/pandas/asv_bench/asv.conf.json
new file mode 100644
index 0000000000000..30c692115eab1
--- /dev/null
+++ b/pandas/asv_bench/asv.conf.json
@@ -0,0 +1,132 @@
+{
+    // The version of the config file format.  Do not change, unless
+    // you know what you are doing.
+    "version": 1,
+
+    // The name of the project being benchmarked
+    "project": "pandas",
+
+    // The project's homepage
+    "project_url": "https://pandas.pydata.org/",
+
+    // The URL of the source code repository for the project being
+    // benchmarked
+    "repo": "..",
+
+    // List of branches to benchmark. If not provided, defaults to "master"
+    // (for git) or "default" (for mercurial).
+    "branches": ["main"],
+
+    // The tool to use to create environments.  May be "conda",
+    // "virtualenv" or other value depending on the plugins in use.
+    // If missing or the empty string, the tool will be automatically
+    // determined by looking for tools on the PATH environment
+    // variable.
+    "environment_type": "conda",
+
+    // the base URL to show a commit for the project.
+    "show_commit_url": "https://github.com/pandas-dev/pandas/commit/",
+
+    // The Pythons you'd like to test against.  If not provided, defaults
+    // to the current version of Python used to run `asv`.
+    "pythons": ["3.10"],
+
+    // The matrix of dependencies to test.  Each key is the name of a
+    // package (in PyPI) and the values are version numbers.  An empty
+    // list or empty string indicates to just test against the default
+    // (latest) version. null indicates that the package is to not be
+    // installed. If the package to be tested is only available from
+    // PyPi, and the 'environment_type' is conda, then you can preface
+    // the package name by 'pip+', and the package will be installed via
+    // pip (with all the conda available packages installed first,
+    // followed by the pip installed packages).
+    "matrix": {
+        "pip+build": [],
+        "Cython": ["3.0"],
+        "matplotlib": [],
+        "sqlalchemy": [],
+        "scipy": [],
+        "numba": [],
+        "numexpr": [],
+        "pytables": [null, ""],  // platform dependent, see excludes below
+        "pyarrow": [],
+        "tables": [null, ""],
+        "openpyxl": [],
+        "xlsxwriter": [],
+        "xlrd": [],
+        "odfpy": [],
+        "jinja2": [],
+        "meson": [],
+        "meson-python": [],
+        "python-build": [],
+    },
+    "conda_channels": ["conda-forge"],
+    // Combinations of libraries/python versions can be excluded/included
+    // from the set to test. Each entry is a dictionary containing additional
+    // key-value pairs to include/exclude.
+    //
+    // An exclude entry excludes entries where all values match. The
+    // values are regexps that should match the whole string.
+    //
+    // An include entry adds an environment. Only the packages listed
+    // are installed. The 'python' key is required. The exclude rules
+    // do not apply to includes.
+    //
+    // In addition to package names, the following keys are available:
+    //
+    // - python
+    //     Python version, as in the *pythons* variable above.
+    // - environment_type
+    //     Environment type, as above.
+    // - sys_platform
+    //     Platform, as in sys.platform. Possible values for the common
+    //     cases: 'linux2', 'win32', 'cygwin', 'darwin'.
+    "exclude": [
+        // On conda install pytables, otherwise tables
+        {"environment_type": "conda", "tables": ""},
+        {"environment_type": "conda", "pytables": null},
+        {"environment_type": "(?!conda).*", "tables": null},
+        {"environment_type": "(?!conda).*", "pytables": ""},
+    ],
+    "include": [],
+
+    // The directory (relative to the current directory) that benchmarks are
+    // stored in.  If not provided, defaults to "benchmarks"
+    // "benchmark_dir": "benchmarks",
+
+    // The directory (relative to the current directory) to cache the Python
+    // environments in.  If not provided, defaults to "env"
+    // "env_dir": "env",
+
+    // The directory (relative to the current directory) that raw benchmark
+    // results are stored in.  If not provided, defaults to "results".
+    // "results_dir": "results",
+
+    // The directory (relative to the current directory) that the html tree
+    // should be written to.  If not provided, defaults to "html".
+    // "html_dir": "html",
+
+    // The number of characters to retain in the commit hashes.
+    // "hash_length": 8,
+
+    // `asv` will cache wheels of the recent builds in each
+    // environment, making them faster to install next time.  This is
+    // number of builds to keep, per environment.
+    "build_cache_size": 8,
+
+    // The commits after which the regression search in `asv publish`
+    // should start looking for regressions. Dictionary whose keys are
+    // regexps matching to benchmark names, and values corresponding to
+    // the commit (exclusive) after which to start looking for
+    // regressions.  The default is to start from the first commit
+    // with results. If the commit is `null`, regression detection is
+    // skipped for the matching benchmark.
+    //
+    "regressions_first_commits": {
+        ".*": "0409521665"
+    },
+    "regression_thresholds": {
+    },
+    "build_command":
+    ["python -m build -Cbuilddir=builddir --wheel --outdir {build_cache_dir} {build_dir}"]
+}
diff --git a/pandas/asv_bench/benchmarks/__init__.py b/pandas/asv_bench/benchmarks/__init__.py
new file mode 100644
index 0000000000000..eada147852fe1
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/__init__.py
@@ -0,0 +1 @@
+"""Pandas benchmarks."""
diff --git a/pandas/asv_bench/benchmarks/algorithms.py b/pandas/asv_bench/benchmarks/algorithms.py
new file mode 100644
index 0000000000000..933e8fbc175d8
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/algorithms.py
@@ -0,0 +1,210 @@
+from importlib import import_module
+
+import numpy as np
+
+import pandas as pd
+
+for imp in ["pandas.util", "pandas.tools.hashing"]:
+    try:
+        hashing = import_module(imp)
+        break
+    except (ImportError, TypeError, ValueError):
+        pass
+
+
+class Factorize:
+    params = [
+        [True, False],
+        [True, False],
+        [
+            "int64",
+            "uint64",
+            "float64",
+            "object",
+            "object_str",
+            "datetime64[ns]",
+            "datetime64[ns, tz]",
+            "Int64",
+            "boolean",
+            "string[pyarrow]",
+        ],
+    ]
+    param_names = ["unique", "sort", "dtype"]
+
+    def setup(self, unique, sort, dtype):
+        N = 10**5
+
+        if dtype in ["int64", "uint64", "Int64", "object"]:
+            data = pd.Index(np.arange(N), dtype=dtype)
+        elif dtype == "float64":
+            data = pd.Index(np.random.randn(N), dtype=dtype)
+        elif dtype == "boolean":
+            data = pd.array(np.random.randint(0, 2, N), dtype=dtype)
+        elif dtype == "datetime64[ns]":
+            data = pd.date_range("2011-01-01", freq="h", periods=N)
+        elif dtype == "datetime64[ns, tz]":
+            data = pd.date_range("2011-01-01", freq="h", periods=N, tz="Asia/Tokyo")
+        elif dtype == "object_str":
+            data = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
+        elif dtype == "string[pyarrow]":
+            data = pd.array(
+                pd.Index([f"i-{i}" for i in range(N)], dtype=object),
+                dtype="string[pyarrow]",
+            )
+        else:
+            raise NotImplementedError
+
+        if not unique:
+            data = data.repeat(5)
+        self.data = data
+
+    def time_factorize(self, unique, sort, dtype):
+        pd.factorize(self.data, sort=sort)
+
+    def peakmem_factorize(self, unique, sort, dtype):
+        pd.factorize(self.data, sort=sort)
+
+
+class Duplicated:
+    params = [
+        [True, False],
+        ["first", "last", False],
+        [
+            "int64",
+            "uint64",
+            "float64",
+            "string",
+            "datetime64[ns]",
+            "datetime64[ns, tz]",
+            "timestamp[ms][pyarrow]",
+            "duration[s][pyarrow]",
+        ],
+    ]
+    param_names = ["unique", "keep", "dtype"]
+
+    def setup(self, unique, keep, dtype):
+        N = 10**5
+        if dtype in ["int64", "uint64"]:
+            data = pd.Index(np.arange(N), dtype=dtype)
+        elif dtype == "float64":
+            data = pd.Index(np.random.randn(N), dtype="float64")
+        elif dtype == "string":
+            data = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
+        elif dtype == "datetime64[ns]":
+            data = pd.date_range("2011-01-01", freq="h", periods=N)
+        elif dtype == "datetime64[ns, tz]":
+            data = pd.date_range("2011-01-01", freq="h", periods=N, tz="Asia/Tokyo")
+        elif dtype in ["timestamp[ms][pyarrow]", "duration[s][pyarrow]"]:
+            data = pd.Index(np.arange(N), dtype=dtype)
+        else:
+            raise NotImplementedError
+        if not unique:
+            data = data.repeat(5)
+        self.idx = data
+        # cache is_unique
+        self.idx.is_unique
+
+    def time_duplicated(self, unique, keep, dtype):
+        self.idx.duplicated(keep=keep)
+
+
+class DuplicatedMaskedArray:
+    params = [
+        [True, False],
+        ["first", "last", False],
+        ["Int64", "Float64"],
+    ]
+    param_names = ["unique", "keep", "dtype"]
+
+    def setup(self, unique, keep, dtype):
+        N = 10**5
+        data = pd.Series(np.arange(N), dtype=dtype)
+        data[list(range(1, N, 100))] = pd.NA
+        if not unique:
+            data = data.repeat(5)
+        self.ser = data
+        # cache is_unique
+        self.ser.is_unique
+
+    def time_duplicated(self, unique, keep, dtype):
+        self.ser.duplicated(keep=keep)
+
+
+class Hashing:
+    def setup_cache(self):
+        N = 10**5
+
+        df = pd.DataFrame(
+            {
+                "strings": pd.Series(
+                    pd.Index([f"i-{i}" for i in range(10000)], dtype=object).take(
+                        np.random.randint(0, 10000, size=N)
+                    )
+                ),
+                "floats": np.random.randn(N),
+                "ints": np.arange(N),
+                "dates": pd.date_range("20110101", freq="s", periods=N),
+                "timedeltas": pd.timedelta_range("1 day", freq="s", periods=N),
+            }
+        )
+        df["categories"] = df["strings"].astype("category")
+        df.iloc[10:20] = np.nan
+        return df
+
+    def time_frame(self, df):
+        hashing.hash_pandas_object(df)
+
+    def time_series_int(self, df):
+        hashing.hash_pandas_object(df["ints"])
+
+    def time_series_string(self, df):
+        hashing.hash_pandas_object(df["strings"])
+
+    def time_series_float(self, df):
+        hashing.hash_pandas_object(df["floats"])
+
+    def time_series_categorical(self, df):
+        hashing.hash_pandas_object(df["categories"])
+
+    def time_series_timedeltas(self, df):
+        hashing.hash_pandas_object(df["timedeltas"])
+
+    def time_series_dates(self, df):
+        hashing.hash_pandas_object(df["dates"])
+
+
+class Quantile:
+    params = [
+        [0, 0.5, 1],
+        ["linear", "nearest", "lower", "higher", "midpoint"],
+        ["float64", "int64", "uint64"],
+    ]
+    param_names = ["quantile", "interpolation", "dtype"]
+
+    def setup(self, quantile, interpolation, dtype):
+        N = 10**5
+        if dtype in ["int64", "uint64"]:
+            data = np.arange(N, dtype=dtype)
+        elif dtype == "float64":
+            data = np.random.randn(N)
+        else:
+            raise NotImplementedError
+        self.ser = pd.Series(data.repeat(5))
+
+    def time_quantile(self, quantile, interpolation, dtype):
+        self.ser.quantile(quantile, interpolation=interpolation)
+
+
+class SortIntegerArray:
+    params = [10**3, 10**5]
+
+    def setup(self, N):
+        data = np.arange(N, dtype=float)
+        data[40] = np.nan
+        self.array = pd.array(data, dtype="Int64")
+
+    def time_argsort(self, N):
+        self.array.argsort()
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/algos/__init__.py b/pandas/asv_bench/benchmarks/algos/__init__.py
new file mode 100644
index 0000000000000..97c9ab09b9c6b
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/algos/__init__.py
@@ -0,0 +1,12 @@
+"""
+algos/ directory is intended for individual functions from core.algorithms
+
+In many cases these algorithms are reachable in multiple ways:
+   algos.foo(x, y)
+   Series(x).foo(y)
+   Index(x).foo(y)
+   pd.array(x).foo(y)
+
+In most cases we profile the Series variant directly, trusting the performance
+of the others to be highly correlated.
+"""
diff --git a/pandas/asv_bench/benchmarks/algos/isin.py b/pandas/asv_bench/benchmarks/algos/isin.py
new file mode 100644
index 0000000000000..a17732c70c2c7
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/algos/isin.py
@@ -0,0 +1,341 @@
+import numpy as np
+
+from pandas import (
+    Categorical,
+    Index,
+    NaT,
+    Series,
+    date_range,
+)
+
+
+class IsIn:
+    params = [
+        "int64",
+        "uint64",
+        "object",
+        "Int64",
+        "boolean",
+        "bool",
+        "datetime64[ns]",
+        "category[object]",
+        "category[int]",
+        "str",
+        "string[python]",
+        "string[pyarrow]",
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10000
+
+        self.mismatched = [NaT.to_datetime64()] * 2
+
+        if dtype in ["boolean", "bool"]:
+            self.series = Series(np.random.randint(0, 2, N)).astype(dtype)
+            self.values = [True, False]
+
+        elif dtype == "datetime64[ns]":
+            # Note: values here is much larger than non-dt64ns cases
+
+            # dti has length=115777
+            dti = date_range(start="2015-10-26", end="2016-01-01", freq="50s")
+            self.series = Series(dti)
+            self.values = self.series._values[::3]
+            self.mismatched = [1, 2]
+
+        elif dtype in ["category[object]", "category[int]"]:
+            # Note: sizes are different in this case than others
+            n = 5 * 10**5
+            sample_size = 100
+
+            arr = list(np.random.randint(0, n // 10, size=n))
+            if dtype == "category[object]":
+                arr = [f"s{i:04d}" for i in arr]
+
+            self.values = np.random.choice(arr, sample_size)
+            self.series = Series(arr).astype("category")
+
+        elif dtype in ["str", "string[python]", "string[pyarrow]"]:
+            try:
+                self.series = Series(
+                    Index([f"i-{i}" for i in range(N)], dtype=object)._values,
+                    dtype=dtype,
+                )
+            except ImportError as err:
+                raise NotImplementedError from err
+            self.values = list(self.series[:2])
+
+        else:
+            self.series = Series(np.random.randint(1, 10, N)).astype(dtype)
+            self.values = [1, 2]
+
+        self.cat_values = Categorical(self.values)
+
+    def time_isin(self, dtype):
+        self.series.isin(self.values)
+
+    def time_isin_categorical(self, dtype):
+        self.series.isin(self.cat_values)
+
+    def time_isin_empty(self, dtype):
+        self.series.isin([])
+
+    def time_isin_mismatched_dtype(self, dtype):
+        self.series.isin(self.mismatched)
+
+
+class IsinAlmostFullWithRandomInt:
+    params = [
+        [np.float64, np.int64, np.uint64, np.object_],
+        range(10, 21),
+        ["inside", "outside"],
+    ]
+    param_names = ["dtype", "exponent", "title"]
+
+    def setup(self, dtype, exponent, title):
+        M = 3 * 2 ** (exponent - 2)
+        # 0.77-the maximal share of occupied buckets
+        self.series = Series(np.random.randint(0, M, M)).astype(dtype)
+
+        values = np.random.randint(0, M, M).astype(dtype)
+        if title == "inside":
+            self.values = values
+        elif title == "outside":
+            self.values = values + M
+        else:
+            raise ValueError(title)
+
+    def time_isin(self, dtype, exponent, title):
+        self.series.isin(self.values)
+
+
+class IsinWithRandomFloat:
+    params = [
+        [np.float64, np.object_],
+        [
+            1_300,
+            2_000,
+            7_000,
+            8_000,
+            70_000,
+            80_000,
+            750_000,
+            900_000,
+        ],
+        ["inside", "outside"],
+    ]
+    param_names = ["dtype", "size", "title"]
+
+    def setup(self, dtype, size, title):
+        self.values = np.random.rand(size)
+        self.series = Series(self.values).astype(dtype)
+        np.random.shuffle(self.values)
+
+        if title == "outside":
+            self.values = self.values + 0.1
+
+    def time_isin(self, dtype, size, title):
+        self.series.isin(self.values)
+
+
+class IsinWithArangeSorted:
+    params = [
+        [np.float64, np.int64, np.uint64, np.object_],
+        [
+            1_000,
+            2_000,
+            8_000,
+            100_000,
+            1_000_000,
+        ],
+    ]
+    param_names = ["dtype", "size"]
+
+    def setup(self, dtype, size):
+        self.series = Series(np.arange(size)).astype(dtype)
+        self.values = np.arange(size).astype(dtype)
+
+    def time_isin(self, dtype, size):
+        self.series.isin(self.values)
+
+
+class IsinWithArange:
+    params = [
+        [np.float64, np.int64, np.uint64, np.object_],
+        [
+            1_000,
+            2_000,
+            8_000,
+        ],
+        [-2, 0, 2],
+    ]
+    param_names = ["dtype", "M", "offset_factor"]
+
+    def setup(self, dtype, M, offset_factor):
+        offset = int(M * offset_factor)
+        tmp = Series(np.random.randint(offset, M + offset, 10**6))
+        self.series = tmp.astype(dtype)
+        self.values = np.arange(M).astype(dtype)
+
+    def time_isin(self, dtype, M, offset_factor):
+        self.series.isin(self.values)
+
+
+class IsInFloat64:
+    params = [
+        [np.float64, "Float64"],
+        ["many_different_values", "few_different_values", "only_nans_values"],
+    ]
+    param_names = ["dtype", "title"]
+
+    def setup(self, dtype, title):
+        N_many = 10**5
+        N_few = 10**6
+        self.series = Series([1, 2], dtype=dtype)
+
+        if title == "many_different_values":
+            # runtime is dominated by creation of the lookup-table
+            self.values = np.arange(N_many, dtype=np.float64)
+        elif title == "few_different_values":
+            # runtime is dominated by creation of the lookup-table
+            self.values = np.zeros(N_few, dtype=np.float64)
+        elif title == "only_nans_values":
+            # runtime is dominated by creation of the lookup-table
+            self.values = np.full(N_few, np.nan, dtype=np.float64)
+        else:
+            raise ValueError(title)
+
+    def time_isin(self, dtype, title):
+        self.series.isin(self.values)
+
+
+class IsInForObjects:
+    """
+    A subset of the cartesian product of cases have special motivations:
+
+    "nans" x "nans"
+        if nan-objects are different objects,
+        this has the potential to trigger O(n^2) running time
+
+    "short" x "long"
+        running time dominated by the preprocessing
+
+    "long" x "short"
+        running time dominated by look-up
+
+    "long" x "long"
+        no dominating part
+
+    "long_floats" x "long_floats"
+        because of nans floats are special
+        no dominating part
+
+    """
+
+    variants = ["nans", "short", "long", "long_floats"]
+
+    params = [variants, variants]
+    param_names = ["series_type", "vals_type"]
+
+    def setup(self, series_type, vals_type):
+        N_many = 10**5
+
+        if series_type == "nans":
+            ser_vals = np.full(10**4, np.nan)
+        elif series_type == "short":
+            ser_vals = np.arange(2)
+        elif series_type == "long":
+            ser_vals = np.arange(N_many)
+        elif series_type == "long_floats":
+            ser_vals = np.arange(N_many, dtype=np.float64)
+
+        self.series = Series(ser_vals).astype(object)
+
+        if vals_type == "nans":
+            values = np.full(10**4, np.nan)
+        elif vals_type == "short":
+            values = np.arange(2)
+        elif vals_type == "long":
+            values = np.arange(N_many)
+        elif vals_type == "long_floats":
+            values = np.arange(N_many, dtype=np.float64)
+
+        self.values = values.astype(object)
+
+    def time_isin(self, series_type, vals_type):
+        self.series.isin(self.values)
+
+
+class IsInLongSeriesLookUpDominates:
+    params = [
+        ["int64", "int32", "float64", "float32", "object", "Int64", "Float64"],
+        [5, 1000],
+        ["random_hits", "random_misses", "monotone_hits", "monotone_misses"],
+    ]
+    param_names = ["dtype", "MaxNumber", "series_type"]
+
+    def setup(self, dtype, MaxNumber, series_type):
+        N = 10**7
+
+        if series_type == "random_hits":
+            array = np.random.randint(0, MaxNumber, N)
+        if series_type == "random_misses":
+            array = np.random.randint(0, MaxNumber, N) + MaxNumber
+        if series_type == "monotone_hits":
+            array = np.repeat(np.arange(MaxNumber), N // MaxNumber)
+        if series_type == "monotone_misses":
+            array = np.arange(N) + MaxNumber
+
+        self.series = Series(array).astype(dtype)
+
+        self.values = np.arange(MaxNumber).astype(dtype.lower())
+
+    def time_isin(self, dtypes, MaxNumber, series_type):
+        self.series.isin(self.values)
+
+
+class IsInLongSeriesValuesDominate:
+    params = [
+        ["int64", "int32", "float64", "float32", "object", "Int64", "Float64"],
+        ["random", "monotone"],
+    ]
+    param_names = ["dtype", "series_type"]
+
+    def setup(self, dtype, series_type):
+        N = 10**7
+
+        if series_type == "random":
+            vals = np.random.randint(0, 10 * N, N)
+        if series_type == "monotone":
+            vals = np.arange(N)
+
+        self.values = vals.astype(dtype.lower())
+        M = 10**6 + 1
+        self.series = Series(np.arange(M)).astype(dtype)
+
+    def time_isin(self, dtypes, series_type):
+        self.series.isin(self.values)
+
+
+class IsInWithLongTupples:
+    def setup(self):
+        t = tuple(range(1000))
+        self.series = Series([t] * 1000)
+        self.values = [t]
+
+    def time_isin(self):
+        self.series.isin(self.values)
+
+
+class IsInIndexes:
+    def setup(self):
+        self.range_idx = Index(range(1000))
+        self.index = Index(list(range(1000)))
+        self.series = Series(np.random.randint(100_000, size=1000))
+
+    def time_isin_range_index(self):
+        self.series.isin(self.range_idx)
+
+    def time_isin_index(self):
+        self.series.isin(self.index)
diff --git a/pandas/asv_bench/benchmarks/arithmetic.py b/pandas/asv_bench/benchmarks/arithmetic.py
new file mode 100644
index 0000000000000..6b1f75187f887
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/arithmetic.py
@@ -0,0 +1,479 @@
+import operator
+import warnings
+
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+    to_timedelta,
+)
+
+from .pandas_vb_common import numeric_dtypes
+
+try:
+    import pandas.core.computation.expressions as expr
+except ImportError:
+    import pandas.computation.expressions as expr
+try:
+    import pandas.tseries.holiday
+except ImportError:
+    pass
+
+
+class IntFrameWithScalar:
+    params = [
+        [np.float64, np.int64],
+        [2, 3.0, np.int32(4), np.float64(5)],
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.truediv,
+            operator.floordiv,
+            operator.pow,
+            operator.mod,
+            operator.eq,
+            operator.ne,
+            operator.gt,
+            operator.ge,
+            operator.lt,
+            operator.le,
+        ],
+    ]
+    param_names = ["dtype", "scalar", "op"]
+
+    def setup(self, dtype, scalar, op):
+        arr = np.random.randn(20000, 100)
+        self.df = DataFrame(arr.astype(dtype))
+
+    def time_frame_op_with_scalar(self, dtype, scalar, op):
+        op(self.df, scalar)
+
+
+class OpWithFillValue:
+    def setup(self):
+        # GH#31300
+        arr = np.arange(10**6)
+        df = DataFrame({"A": arr})
+        ser = df["A"]
+
+        self.df = df
+        self.ser = ser
+
+    def time_frame_op_with_fill_value_no_nas(self):
+        self.df.add(self.df, fill_value=4)
+
+    def time_series_op_with_fill_value_no_nas(self):
+        self.ser.add(self.ser, fill_value=4)
+
+
+class MixedFrameWithSeriesAxis:
+    params = [
+        [
+            "eq",
+            "ne",
+            "lt",
+            "le",
+            "ge",
+            "gt",
+            "add",
+            "sub",
+            "truediv",
+            "floordiv",
+            "mul",
+            "pow",
+        ]
+    ]
+    param_names = ["opname"]
+
+    def setup(self, opname):
+        arr = np.arange(10**6).reshape(1000, -1)
+        df = DataFrame(arr)
+        df["C"] = 1.0
+        self.df = df
+        self.ser = df[0]
+        self.row = df.iloc[0]
+
+    def time_frame_op_with_series_axis0(self, opname):
+        getattr(self.df, opname)(self.ser, axis=0)
+
+    def time_frame_op_with_series_axis1(self, opname):
+        getattr(operator, opname)(self.df, self.ser)
+
+    # exclude comparisons from the params for time_frame_op_with_series_axis1
+    #  since they do not do alignment so raise
+    time_frame_op_with_series_axis1.params = [params[0][6:]]
+
+
+class FrameWithFrameWide:
+    # Many-columns, mixed dtypes
+
+    params = [
+        [
+            # GH#32779 has discussion of which operators are included here
+            operator.add,
+            operator.floordiv,
+            operator.gt,
+        ],
+        [
+            # (n_rows, n_columns)
+            (1_000_000, 10),
+            (100_000, 100),
+            (10_000, 1000),
+            (1000, 10_000),
+        ],
+    ]
+    param_names = ["op", "shape"]
+
+    def setup(self, op, shape):
+        # we choose dtypes so as to make the blocks
+        #  a) not perfectly match between right and left
+        #  b) appreciably bigger than single columns
+        n_rows, n_cols = shape
+
+        if op is operator.floordiv:
+            # floordiv is much slower than the other operations -> use less data
+            n_rows = n_rows // 10
+
+        # construct dataframe with 2 blocks
+        arr1 = np.random.randn(n_rows, n_cols // 2).astype("f8")
+        arr2 = np.random.randn(n_rows, n_cols // 2).astype("f4")
+        df = pd.concat([DataFrame(arr1), DataFrame(arr2)], axis=1, ignore_index=True)
+        # should already be the case, but just to be sure
+        df._consolidate_inplace()
+
+        # TODO: GH#33198 the setting here shouldn't need two steps
+        arr1 = np.random.randn(n_rows, max(n_cols // 4, 3)).astype("f8")
+        arr2 = np.random.randn(n_rows, n_cols // 2).astype("i8")
+        arr3 = np.random.randn(n_rows, n_cols // 4).astype("f8")
+        df2 = pd.concat(
+            [DataFrame(arr1), DataFrame(arr2), DataFrame(arr3)],
+            axis=1,
+            ignore_index=True,
+        )
+        # should already be the case, but just to be sure
+        df2._consolidate_inplace()
+
+        self.left = df
+        self.right = df2
+
+    def time_op_different_blocks(self, op, shape):
+        # blocks (and dtypes) are not aligned
+        op(self.left, self.right)
+
+    def time_op_same_blocks(self, op, shape):
+        # blocks (and dtypes) are aligned
+        op(self.left, self.left)
+
+
+class Ops:
+    params = [[True, False], ["default", 1]]
+    param_names = ["use_numexpr", "threads"]
+
+    def setup(self, use_numexpr, threads):
+        self.df = DataFrame(np.random.randn(20000, 100))
+        self.df2 = DataFrame(np.random.randn(20000, 100))
+
+        if threads != "default":
+            expr.set_numexpr_threads(threads)
+        if not use_numexpr:
+            expr.set_use_numexpr(False)
+
+    def time_frame_add(self, use_numexpr, threads):
+        self.df + self.df2
+
+    def time_frame_mult(self, use_numexpr, threads):
+        self.df * self.df2
+
+    def time_frame_multi_and(self, use_numexpr, threads):
+        self.df[(self.df > 0) & (self.df2 > 0)]
+
+    def time_frame_comparison(self, use_numexpr, threads):
+        self.df > self.df2
+
+    def teardown(self, use_numexpr, threads):
+        expr.set_use_numexpr(True)
+        expr.set_numexpr_threads()
+
+
+class Ops2:
+    def setup(self):
+        N = 10**3
+        self.df = DataFrame(np.random.randn(N, N))
+        self.df2 = DataFrame(np.random.randn(N, N))
+
+        self.df_int = DataFrame(
+            np.random.randint(
+                np.iinfo(np.int16).min, np.iinfo(np.int16).max, size=(N, N)
+            )
+        )
+        self.df2_int = DataFrame(
+            np.random.randint(
+                np.iinfo(np.int16).min, np.iinfo(np.int16).max, size=(N, N)
+            )
+        )
+
+        self.s = Series(np.random.randn(N))
+
+    # Division
+
+    def time_frame_float_div(self):
+        self.df // self.df2
+
+    def time_frame_float_div_by_zero(self):
+        self.df / 0
+
+    def time_frame_float_floor_by_zero(self):
+        self.df // 0
+
+    def time_frame_int_div_by_zero(self):
+        self.df_int / 0
+
+    # Modulo
+
+    def time_frame_int_mod(self):
+        self.df_int % self.df2_int
+
+    def time_frame_float_mod(self):
+        self.df % self.df2
+
+    # Dot product
+
+    def time_frame_dot(self):
+        self.df.dot(self.df2)
+
+    def time_series_dot(self):
+        self.s.dot(self.s)
+
+    def time_frame_series_dot(self):
+        self.df.dot(self.s)
+
+
+class Timeseries:
+    params = [None, "US/Eastern"]
+    param_names = ["tz"]
+
+    def setup(self, tz):
+        N = 10**6
+        halfway = (N // 2) - 1
+        self.s = Series(date_range("20010101", periods=N, freq="min", tz=tz))
+        self.ts = self.s[halfway]
+
+        self.s2 = Series(date_range("20010101", periods=N, freq="s", tz=tz))
+        self.ts_different_reso = Timestamp("2001-01-02", tz=tz)
+
+    def time_series_timestamp_compare(self, tz):
+        self.s <= self.ts
+
+    def time_series_timestamp_different_reso_compare(self, tz):
+        self.s <= self.ts_different_reso
+
+    def time_timestamp_series_compare(self, tz):
+        self.ts >= self.s
+
+    def time_timestamp_ops_diff(self, tz):
+        self.s2.diff()
+
+    def time_timestamp_ops_diff_with_shift(self, tz):
+        self.s - self.s.shift()
+
+
+class IrregularOps:
+    def setup(self):
+        N = 10**5
+        idx = date_range(start="1/1/2000", periods=N, freq="s")
+        s = Series(np.random.randn(N), index=idx)
+        self.left = s.sample(frac=1)
+        self.right = s.sample(frac=1)
+
+    def time_add(self):
+        self.left + self.right
+
+
+class TimedeltaOps:
+    def setup(self):
+        self.td = to_timedelta(np.arange(1000000))
+        self.ts = Timestamp("2000")
+
+    def time_add_td_ts(self):
+        self.td + self.ts
+
+
+class CategoricalComparisons:
+    params = ["__lt__", "__le__", "__eq__", "__ne__", "__ge__", "__gt__"]
+    param_names = ["op"]
+
+    def setup(self, op):
+        N = 10**5
+        self.cat = pd.Categorical(list("aabbcd") * N, ordered=True)
+
+    def time_categorical_op(self, op):
+        getattr(self.cat, op)("b")
+
+
+class IndexArithmetic:
+    params = ["float", "int"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**6
+        if dtype == "float":
+            self.index = Index(np.arange(N), dtype=np.float64)
+        elif dtype == "int":
+            self.index = Index(np.arange(N), dtype=np.int64)
+
+    def time_add(self, dtype):
+        self.index + 2
+
+    def time_subtract(self, dtype):
+        self.index - 2
+
+    def time_multiply(self, dtype):
+        self.index * 2
+
+    def time_divide(self, dtype):
+        self.index / 2
+
+    def time_modulo(self, dtype):
+        self.index % 2
+
+
+class NumericInferOps:
+    # from GH 7332
+    params = numeric_dtypes
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 5 * 10**5
+        self.df = DataFrame(
+            {"A": np.arange(N).astype(dtype), "B": np.arange(N).astype(dtype)}
+        )
+
+    def time_add(self, dtype):
+        self.df["A"] + self.df["B"]
+
+    def time_subtract(self, dtype):
+        self.df["A"] - self.df["B"]
+
+    def time_multiply(self, dtype):
+        self.df["A"] * self.df["B"]
+
+    def time_divide(self, dtype):
+        self.df["A"] / self.df["B"]
+
+    def time_modulo(self, dtype):
+        self.df["A"] % self.df["B"]
+
+
+class DateInferOps:
+    # from GH 7332
+    def setup_cache(self):
+        N = 5 * 10**5
+        df = DataFrame({"datetime64": np.arange(N).astype("datetime64[ms]")})
+        df["timedelta"] = df["datetime64"] - df["datetime64"]
+        return df
+
+    def time_subtract_datetimes(self, df):
+        df["datetime64"] - df["datetime64"]
+
+    def time_timedelta_plus_datetime(self, df):
+        df["timedelta"] + df["datetime64"]
+
+    def time_add_timedeltas(self, df):
+        df["timedelta"] + df["timedelta"]
+
+
+hcal = pd.tseries.holiday.USFederalHolidayCalendar()
+# These offsets currently raise a NotImplementedError with .apply_index()
+non_apply = [
+    pd.offsets.Day(),
+    pd.offsets.BYearEnd(),
+    pd.offsets.BYearBegin(),
+    pd.offsets.BQuarterEnd(),
+    pd.offsets.BQuarterBegin(),
+    pd.offsets.BMonthEnd(),
+    pd.offsets.BMonthBegin(),
+    pd.offsets.CustomBusinessDay(),
+    pd.offsets.CustomBusinessDay(calendar=hcal),
+    pd.offsets.CustomBusinessMonthBegin(calendar=hcal),
+    pd.offsets.CustomBusinessMonthEnd(calendar=hcal),
+    pd.offsets.CustomBusinessMonthEnd(calendar=hcal),
+]
+other_offsets = [
+    pd.offsets.YearEnd(),
+    pd.offsets.YearBegin(),
+    pd.offsets.QuarterEnd(),
+    pd.offsets.QuarterBegin(),
+    pd.offsets.MonthEnd(),
+    pd.offsets.MonthBegin(),
+    pd.offsets.DateOffset(months=2, days=2),
+    pd.offsets.BusinessDay(),
+    pd.offsets.SemiMonthEnd(),
+    pd.offsets.SemiMonthBegin(),
+]
+offsets = non_apply + other_offsets
+
+
+class OffsetArrayArithmetic:
+    params = offsets
+    param_names = ["offset"]
+
+    def setup(self, offset):
+        N = 10000
+        rng = date_range(start="1/1/2000", periods=N, freq="min")
+        self.rng = rng
+        self.ser = Series(rng)
+
+    def time_add_series_offset(self, offset):
+        with warnings.catch_warnings(record=True):
+            self.ser + offset
+
+    def time_add_dti_offset(self, offset):
+        with warnings.catch_warnings(record=True):
+            self.rng + offset
+
+
+class ApplyIndex:
+    params = other_offsets
+    param_names = ["offset"]
+
+    def setup(self, offset):
+        N = 10000
+        rng = date_range(start="1/1/2000", periods=N, freq="min")
+        self.rng = rng
+
+    def time_apply_index(self, offset):
+        self.rng + offset
+
+
+class BinaryOpsMultiIndex:
+    params = ["sub", "add", "mul", "div"]
+    param_names = ["func"]
+
+    def setup(self, func):
+        array = date_range("20200101 00:00", "20200102 0:00", freq="s")
+        level_0_names = [str(i) for i in range(30)]
+
+        index = pd.MultiIndex.from_product([level_0_names, array])
+        column_names = ["col_1", "col_2"]
+
+        self.df = DataFrame(
+            np.random.rand(len(index), 2), index=index, columns=column_names
+        )
+
+        self.arg_df = DataFrame(
+            np.random.randint(1, 10, (len(level_0_names), 2)),
+            index=level_0_names,
+            columns=column_names,
+        )
+
+    def time_binary_op_multiindex(self, func):
+        getattr(self.df, func)(self.arg_df, level=0)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/array.py b/pandas/asv_bench/benchmarks/array.py
new file mode 100644
index 0000000000000..953af5c868356
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/array.py
@@ -0,0 +1,140 @@
+import numpy as np
+
+import pandas as pd
+
+
+class BooleanArray:
+    def setup(self):
+        self.values_bool = np.array([True, False, True, False])
+        self.values_float = np.array([1.0, 0.0, 1.0, 0.0])
+        self.values_integer = np.array([1, 0, 1, 0])
+        self.values_integer_like = [1, 0, 1, 0]
+        self.data = np.array([True, False, True, False])
+        self.mask = np.array([False, False, True, False])
+
+    def time_constructor(self):
+        pd.arrays.BooleanArray(self.data, self.mask)
+
+    def time_from_bool_array(self):
+        pd.array(self.values_bool, dtype="boolean")
+
+    def time_from_integer_array(self):
+        pd.array(self.values_integer, dtype="boolean")
+
+    def time_from_integer_like(self):
+        pd.array(self.values_integer_like, dtype="boolean")
+
+    def time_from_float_array(self):
+        pd.array(self.values_float, dtype="boolean")
+
+
+class IntegerArray:
+    def setup(self):
+        N = 250_000
+        self.values_integer = np.tile(np.array([1, 0, 1, 0]), N)
+        self.data = np.tile(np.array([1, 2, 3, 4], dtype="int64"), N)
+        self.mask = np.tile(np.array([False, False, True, False]), N)
+
+    def time_constructor(self):
+        pd.arrays.IntegerArray(self.data, self.mask)
+
+    def time_from_integer_array(self):
+        pd.array(self.values_integer, dtype="Int64")
+
+
+class IntervalArray:
+    def setup(self):
+        N = 10_000
+        self.tuples = [(i, i + 1) for i in range(N)]
+
+    def time_from_tuples(self):
+        pd.arrays.IntervalArray.from_tuples(self.tuples)
+
+
+class StringArray:
+    def setup(self):
+        N = 100_000
+        values = np.array([str(i) for i in range(N)], dtype=object)
+        self.values_obj = np.array(values, dtype="object")
+        self.values_str = np.array(values, dtype="U")
+        self.values_list = values.tolist()
+
+    def time_from_np_object_array(self):
+        pd.array(self.values_obj, dtype="string")
+
+    def time_from_np_str_array(self):
+        pd.array(self.values_str, dtype="string")
+
+    def time_from_list(self):
+        pd.array(self.values_list, dtype="string")
+
+
+class ArrowStringArray:
+    params = [False, True]
+    param_names = ["multiple_chunks"]
+
+    def setup(self, multiple_chunks):
+        try:
+            import pyarrow as pa
+        except ImportError as err:
+            raise NotImplementedError from err
+        strings = np.array([str(i) for i in range(10_000)], dtype=object)
+        if multiple_chunks:
+            chunks = [strings[i : i + 100] for i in range(0, len(strings), 100)]
+            self.array = pd.arrays.ArrowStringArray(pa.chunked_array(chunks))
+        else:
+            self.array = pd.arrays.ArrowStringArray(pa.array(strings))
+
+    def time_setitem(self, multiple_chunks):
+        for i in range(200):
+            self.array[i] = "foo"
+
+    def time_setitem_list(self, multiple_chunks):
+        indexer = list(range(50)) + list(range(-1000, 0, 50))
+        self.array[indexer] = ["foo"] * len(indexer)
+
+    def time_setitem_slice(self, multiple_chunks):
+        self.array[::10] = "foo"
+
+    def time_setitem_null_slice(self, multiple_chunks):
+        self.array[:] = "foo"
+
+    def time_tolist(self, multiple_chunks):
+        self.array.tolist()
+
+
+class ArrowExtensionArray:
+    params = [
+        [
+            "boolean[pyarrow]",
+            "float64[pyarrow]",
+            "int64[pyarrow]",
+            "string[pyarrow]",
+            "timestamp[ns][pyarrow]",
+        ],
+        [False, True],
+    ]
+    param_names = ["dtype", "hasna"]
+
+    def setup(self, dtype, hasna):
+        N = 100_000
+        if dtype == "boolean[pyarrow]":
+            data = np.random.choice([True, False], N, replace=True)
+        elif dtype == "float64[pyarrow]":
+            data = np.random.randn(N)
+        elif dtype == "int64[pyarrow]":
+            data = np.arange(N)
+        elif dtype == "string[pyarrow]":
+            data = np.array([str(i) for i in range(N)], dtype=object)
+        elif dtype == "timestamp[ns][pyarrow]":
+            data = pd.date_range("2000-01-01", freq="s", periods=N)
+        else:
+            raise NotImplementedError
+
+        arr = pd.array(data, dtype=dtype)
+        if hasna:
+            arr[::2] = pd.NA
+        self.arr = arr
+
+    def time_to_numpy(self, dtype, hasna):
+        self.arr.to_numpy()
diff --git a/pandas/asv_bench/benchmarks/attrs_caching.py b/pandas/asv_bench/benchmarks/attrs_caching.py
new file mode 100644
index 0000000000000..2a004113d1b91
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/attrs_caching.py
@@ -0,0 +1,50 @@
+import numpy as np
+
+import pandas as pd
+from pandas import DataFrame
+
+try:
+    from pandas.core.construction import extract_array
+except ImportError:
+    extract_array = None
+
+
+class DataFrameAttributes:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(10, 6))
+        self.cur_index = self.df.index
+
+    def time_get_index(self):
+        self.df.index
+
+    def time_set_index(self):
+        self.df.index = self.cur_index
+
+
+class SeriesArrayAttribute:
+    params = [["numeric", "object", "category", "datetime64", "datetime64tz"]]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        if dtype == "numeric":
+            self.series = pd.Series([1, 2, 3])
+        elif dtype == "object":
+            self.series = pd.Series(["a", "b", "c"], dtype=object)
+        elif dtype == "category":
+            self.series = pd.Series(["a", "b", "c"], dtype="category")
+        elif dtype == "datetime64":
+            self.series = pd.Series(pd.date_range("2013", periods=3))
+        elif dtype == "datetime64tz":
+            self.series = pd.Series(pd.date_range("2013", periods=3, tz="UTC"))
+
+    def time_array(self, dtype):
+        self.series.array
+
+    def time_extract_array(self, dtype):
+        extract_array(self.series)
+
+    def time_extract_array_numpy(self, dtype):
+        extract_array(self.series, extract_numpy=True)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/boolean.py b/pandas/asv_bench/benchmarks/boolean.py
new file mode 100644
index 0000000000000..71c422c641775
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/boolean.py
@@ -0,0 +1,32 @@
+import numpy as np
+
+import pandas as pd
+
+
+class TimeLogicalOps:
+    def setup(self):
+        N = 10_000
+        left, right, lmask, rmask = np.random.randint(0, 2, size=(4, N)).astype("bool")
+        self.left = pd.arrays.BooleanArray(left, lmask)
+        self.right = pd.arrays.BooleanArray(right, rmask)
+
+    def time_or_scalar(self):
+        self.left | True
+        self.left | False
+
+    def time_or_array(self):
+        self.left | self.right
+
+    def time_and_scalar(self):
+        self.left & True
+        self.left & False
+
+    def time_and_array(self):
+        self.left & self.right
+
+    def time_xor_scalar(self):
+        self.left ^ True
+        self.left ^ False
+
+    def time_xor_array(self):
+        self.left ^ self.right
diff --git a/pandas/asv_bench/benchmarks/categoricals.py b/pandas/asv_bench/benchmarks/categoricals.py
new file mode 100644
index 0000000000000..7d5b250c7b157
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/categoricals.py
@@ -0,0 +1,337 @@
+import string
+import sys
+import warnings
+
+import numpy as np
+
+import pandas as pd
+
+try:
+    from pandas.api.types import union_categoricals
+except ImportError:
+    try:
+        from pandas.types.concat import union_categoricals
+    except ImportError:
+        pass
+
+
+class Constructor:
+    def setup(self):
+        N = 10**5
+        self.categories = list("abcde")
+        self.cat_idx = pd.Index(self.categories)
+        self.values = np.tile(self.categories, N)
+        self.codes = np.tile(range(len(self.categories)), N)
+
+        self.datetimes = pd.Series(
+            pd.date_range("1995-01-01 00:00:00", periods=N // 10, freq="s")
+        )
+        self.datetimes_with_nat = self.datetimes.copy()
+        self.datetimes_with_nat.iloc[-1] = pd.NaT
+
+        self.values_some_nan = list(np.tile(self.categories + [np.nan], N))
+        self.values_all_nan = [np.nan] * len(self.values)
+        self.values_all_int8 = np.ones(N, "int8")
+        self.categorical = pd.Categorical(self.values, self.categories)
+        self.series = pd.Series(self.categorical)
+        self.intervals = pd.interval_range(0, 1, periods=N // 10)
+
+    def time_regular(self):
+        pd.Categorical(self.values, self.categories)
+
+    def time_fastpath(self):
+        dtype = pd.CategoricalDtype(categories=self.cat_idx)
+        pd.Categorical._simple_new(self.codes, dtype)
+
+    def time_datetimes(self):
+        pd.Categorical(self.datetimes)
+
+    def time_interval(self):
+        pd.Categorical(self.datetimes, categories=self.datetimes)
+
+    def time_datetimes_with_nat(self):
+        pd.Categorical(self.datetimes_with_nat)
+
+    def time_with_nan(self):
+        pd.Categorical(self.values_some_nan)
+
+    def time_all_nan(self):
+        pd.Categorical(self.values_all_nan)
+
+    def time_from_codes_all_int8(self):
+        pd.Categorical.from_codes(self.values_all_int8, self.categories)
+
+    def time_existing_categorical(self):
+        pd.Categorical(self.categorical)
+
+    def time_existing_series(self):
+        pd.Categorical(self.series)
+
+
+class AsType:
+    def setup(self):
+        N = 10**5
+
+        random_pick = np.random.default_rng().choice
+
+        categories = {
+            "str": list(string.ascii_letters),
+            "int": np.random.randint(2**16, size=154),
+            "float": sys.maxsize * np.random.random((38,)),
+            "timestamp": [
+                pd.Timestamp(x, unit="s") for x in np.random.randint(2**18, size=578)
+            ],
+        }
+
+        self.df = pd.DataFrame(
+            {col: random_pick(cats, N) for col, cats in categories.items()}
+        )
+
+        for col in ("int", "float", "timestamp"):
+            self.df[f"{col}_as_str"] = self.df[col].astype(str)
+
+        for col in self.df.columns:
+            self.df[col] = self.df[col].astype("category")
+
+    def astype_str(self):
+        [self.df[col].astype("str") for col in "int float timestamp".split()]
+
+    def astype_int(self):
+        [self.df[col].astype("int") for col in "int_as_str timestamp".split()]
+
+    def astype_float(self):
+        [
+            self.df[col].astype("float")
+            for col in "float_as_str int int_as_str timestamp".split()
+        ]
+
+    def astype_datetime(self):
+        self.df["float"].astype(pd.DatetimeTZDtype(tz="US/Pacific"))
+
+
+class Concat:
+    def setup(self):
+        N = 10**5
+        self.s = pd.Series(list("aabbcd") * N).astype("category")
+
+        self.a = pd.Categorical(list("aabbcd") * N)
+        self.b = pd.Categorical(list("bbcdjk") * N)
+
+        self.idx_a = pd.CategoricalIndex(range(N), range(N))
+        self.idx_b = pd.CategoricalIndex(range(N + 1), range(N + 1))
+        self.df_a = pd.DataFrame(range(N), columns=["a"], index=self.idx_a)
+        self.df_b = pd.DataFrame(range(N + 1), columns=["a"], index=self.idx_b)
+
+    def time_concat(self):
+        pd.concat([self.s, self.s])
+
+    def time_union(self):
+        union_categoricals([self.a, self.b])
+
+    def time_append_overlapping_index(self):
+        self.idx_a.append(self.idx_a)
+
+    def time_append_non_overlapping_index(self):
+        self.idx_a.append(self.idx_b)
+
+    def time_concat_overlapping_index(self):
+        pd.concat([self.df_a, self.df_a])
+
+    def time_concat_non_overlapping_index(self):
+        pd.concat([self.df_a, self.df_b])
+
+
+class ValueCounts:
+    params = [True, False]
+    param_names = ["dropna"]
+
+    def setup(self, dropna):
+        n = 5 * 10**5
+        arr = [f"s{i:04d}" for i in np.random.randint(0, n // 10, size=n)]
+        self.ts = pd.Series(arr).astype("category")
+
+    def time_value_counts(self, dropna):
+        self.ts.value_counts(dropna=dropna)
+
+
+class Repr:
+    def setup(self):
+        self.sel = pd.Series(["s1234"]).astype("category")
+
+    def time_rendering(self):
+        str(self.sel)
+
+
+class SetCategories:
+    def setup(self):
+        n = 5 * 10**5
+        arr = [f"s{i:04d}" for i in np.random.randint(0, n // 10, size=n)]
+        self.ts = pd.Series(arr).astype("category")
+
+    def time_set_categories(self):
+        self.ts.cat.set_categories(self.ts.cat.categories[::2])
+
+
+class RemoveCategories:
+    def setup(self):
+        n = 5 * 10**5
+        arr = [f"s{i:04d}" for i in np.random.randint(0, n // 10, size=n)]
+        self.ts = pd.Series(arr).astype("category")
+
+    def time_remove_categories(self):
+        self.ts.cat.remove_categories(self.ts.cat.categories[::2])
+
+
+class Rank:
+    def setup(self):
+        N = 10**5
+        ncats = 15
+
+        self.s_str = pd.Series(np.random.randint(0, ncats, size=N).astype(str))
+        self.s_str_cat = pd.Series(self.s_str, dtype="category")
+        with warnings.catch_warnings(record=True):
+            str_cat_type = pd.CategoricalDtype(set(self.s_str), ordered=True)
+            self.s_str_cat_ordered = self.s_str.astype(str_cat_type)
+
+        self.s_int = pd.Series(np.random.randint(0, ncats, size=N))
+        self.s_int_cat = pd.Series(self.s_int, dtype="category")
+        with warnings.catch_warnings(record=True):
+            int_cat_type = pd.CategoricalDtype(set(self.s_int), ordered=True)
+            self.s_int_cat_ordered = self.s_int.astype(int_cat_type)
+
+    def time_rank_string(self):
+        self.s_str.rank()
+
+    def time_rank_string_cat(self):
+        self.s_str_cat.rank()
+
+    def time_rank_string_cat_ordered(self):
+        self.s_str_cat_ordered.rank()
+
+    def time_rank_int(self):
+        self.s_int.rank()
+
+    def time_rank_int_cat(self):
+        self.s_int_cat.rank()
+
+    def time_rank_int_cat_ordered(self):
+        self.s_int_cat_ordered.rank()
+
+
+class IsMonotonic:
+    def setup(self):
+        N = 1000
+        self.c = pd.CategoricalIndex(list("a" * N + "b" * N + "c" * N))
+        self.s = pd.Series(self.c)
+
+    def time_categorical_index_is_monotonic_increasing(self):
+        self.c.is_monotonic_increasing
+
+    def time_categorical_index_is_monotonic_decreasing(self):
+        self.c.is_monotonic_decreasing
+
+    def time_categorical_series_is_monotonic_increasing(self):
+        self.s.is_monotonic_increasing
+
+    def time_categorical_series_is_monotonic_decreasing(self):
+        self.s.is_monotonic_decreasing
+
+
+class Contains:
+    def setup(self):
+        N = 10**5
+        self.ci = pd.CategoricalIndex(np.arange(N))
+        self.c = self.ci.values
+        self.key = self.ci.categories[0]
+
+    def time_categorical_index_contains(self):
+        self.key in self.ci
+
+    def time_categorical_contains(self):
+        self.key in self.c
+
+
+class CategoricalSlicing:
+    params = ["monotonic_incr", "monotonic_decr", "non_monotonic"]
+    param_names = ["index"]
+
+    def setup(self, index):
+        N = 10**6
+        categories = ["a", "b", "c"]
+        if index == "monotonic_incr":
+            codes = np.repeat([0, 1, 2], N)
+        elif index == "monotonic_decr":
+            codes = np.repeat([2, 1, 0], N)
+        elif index == "non_monotonic":
+            codes = np.tile([0, 1, 2], N)
+        else:
+            raise ValueError(f"Invalid index param: {index}")
+
+        self.data = pd.Categorical.from_codes(codes, categories=categories)
+        self.scalar = 10000
+        self.list = list(range(10000))
+        self.cat_scalar = "b"
+
+    def time_getitem_scalar(self, index):
+        self.data[self.scalar]
+
+    def time_getitem_slice(self, index):
+        self.data[: self.scalar]
+
+    def time_getitem_list_like(self, index):
+        self.data[[self.scalar]]
+
+    def time_getitem_list(self, index):
+        self.data[self.list]
+
+    def time_getitem_bool_array(self, index):
+        self.data[self.data == self.cat_scalar]
+
+
+class Indexing:
+    def setup(self):
+        N = 10**5
+        self.index = pd.CategoricalIndex(range(N), range(N))
+        self.series = pd.Series(range(N), index=self.index).sort_index()
+        self.category = self.index[500]
+
+    def time_get_loc(self):
+        self.index.get_loc(self.category)
+
+    def time_shallow_copy(self):
+        self.index._view()
+
+    def time_align(self):
+        pd.DataFrame({"a": self.series, "b": self.series[:500]})
+
+    def time_intersection(self):
+        self.index[:750].intersection(self.index[250:])
+
+    def time_unique(self):
+        self.index.unique()
+
+    def time_reindex(self):
+        self.index.reindex(self.index[:500])
+
+    def time_reindex_missing(self):
+        self.index.reindex(["a", "b", "c", "d"])
+
+    def time_sort_values(self):
+        self.index.sort_values(ascending=False)
+
+
+class SearchSorted:
+    def setup(self):
+        N = 10**5
+        self.ci = pd.CategoricalIndex(np.arange(N)).sort_values()
+        self.c = self.ci.values
+        self.key = self.ci.categories[1]
+
+    def time_categorical_index_contains(self):
+        self.ci.searchsorted(self.key)
+
+    def time_categorical_contains(self):
+        self.c.searchsorted(self.key)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/ctors.py b/pandas/asv_bench/benchmarks/ctors.py
new file mode 100644
index 0000000000000..77c9faf3d3a87
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/ctors.py
@@ -0,0 +1,145 @@
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+
+
+def no_change(arr):
+    return arr
+
+
+def list_of_str(arr):
+    return list(arr.astype(str))
+
+
+def gen_of_str(arr):
+    return (x for x in arr.astype(str))
+
+
+def arr_dict(arr):
+    return dict(zip(range(len(arr)), arr))
+
+
+def list_of_tuples(arr):
+    return [(i, -i) for i in arr]
+
+
+def gen_of_tuples(arr):
+    return ((i, -i) for i in arr)
+
+
+def list_of_lists(arr):
+    return [[i, -i] for i in arr]
+
+
+def list_of_tuples_with_none(arr):
+    return [(i, -i) for i in arr][:-1] + [None]
+
+
+def list_of_lists_with_none(arr):
+    return [[i, -i] for i in arr][:-1] + [None]
+
+
+class SeriesConstructors:
+    param_names = ["data_fmt", "with_index", "dtype"]
+    params = [
+        [
+            no_change,
+            list,
+            list_of_str,
+            gen_of_str,
+            arr_dict,
+            list_of_tuples,
+            gen_of_tuples,
+            list_of_lists,
+            list_of_tuples_with_none,
+            list_of_lists_with_none,
+        ],
+        [False, True],
+        ["float", "int"],
+    ]
+
+    # Generators get exhausted on use, so run setup before every call
+    number = 1
+    repeat = (3, 250, 10)
+
+    def setup(self, data_fmt, with_index, dtype):
+        if data_fmt in (gen_of_str, gen_of_tuples) and with_index:
+            raise NotImplementedError(
+                "Series constructors do not support using generators with indexes"
+            )
+        N = 10**4
+        if dtype == "float":
+            arr = np.random.randn(N)
+        else:
+            arr = np.arange(N)
+        self.data = data_fmt(arr)
+        self.index = np.arange(N) if with_index else None
+
+    def time_series_constructor(self, data_fmt, with_index, dtype):
+        Series(self.data, index=self.index)
+
+
+class SeriesDtypesConstructors:
+    def setup(self):
+        N = 10**4
+        self.arr = np.random.randn(N)
+        self.arr_str = np.array(["foo", "bar", "baz"], dtype=object)
+        self.s = Series(
+            [Timestamp("20110101"), Timestamp("20120101"), Timestamp("20130101")]
+            * N
+            * 10
+        )
+
+    def time_index_from_array_string(self):
+        Index(self.arr_str)
+
+    def time_index_from_array_floats(self):
+        Index(self.arr)
+
+    def time_dtindex_from_series(self):
+        DatetimeIndex(self.s)
+
+    def time_dtindex_from_index_with_series(self):
+        Index(self.s)
+
+
+class MultiIndexConstructor:
+    def setup(self):
+        N = 10**4
+        self.iterables = [Index([f"i-{i}" for i in range(N)], dtype=object), range(20)]
+
+    def time_multiindex_from_iterables(self):
+        MultiIndex.from_product(self.iterables)
+
+
+class DatetimeIndexConstructor:
+    def setup(self):
+        N = 20_000
+        dti = date_range("1900-01-01", periods=N)
+
+        self.list_of_timestamps = dti.tolist()
+        self.list_of_dates = dti.date.tolist()
+        self.list_of_datetimes = dti.to_pydatetime().tolist()
+        self.list_of_str = dti.strftime("%Y-%m-%d").tolist()
+
+    def time_from_list_of_timestamps(self):
+        DatetimeIndex(self.list_of_timestamps)
+
+    def time_from_list_of_dates(self):
+        DatetimeIndex(self.list_of_dates)
+
+    def time_from_list_of_datetimes(self):
+        DatetimeIndex(self.list_of_datetimes)
+
+    def time_from_list_of_str(self):
+        DatetimeIndex(self.list_of_str)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/dtypes.py b/pandas/asv_bench/benchmarks/dtypes.py
new file mode 100644
index 0000000000000..7f3429b5e3882
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/dtypes.py
@@ -0,0 +1,129 @@
+import string
+
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+from pandas.api.types import (
+    is_extension_array_dtype,
+    pandas_dtype,
+)
+
+from .pandas_vb_common import (
+    datetime_dtypes,
+    extension_dtypes,
+    numeric_dtypes,
+    string_dtypes,
+)
+
+_numpy_dtypes = [
+    np.dtype(dtype) for dtype in (numeric_dtypes + datetime_dtypes + string_dtypes)
+]
+_dtypes = _numpy_dtypes + extension_dtypes
+
+
+class Dtypes:
+    params = _dtypes + [dt.name for dt in _dtypes]
+    param_names = ["dtype"]
+
+    def time_pandas_dtype(self, dtype):
+        pandas_dtype(dtype)
+
+
+class DtypesInvalid:
+    param_names = ["dtype"]
+    params = ["scalar-string", "scalar-int", "list-string", "array-string"]
+    data_dict = {
+        "scalar-string": "foo",
+        "scalar-int": 1,
+        "list-string": ["foo"] * 1000,
+        "array-string": np.array(["foo"] * 1000),
+    }
+
+    def time_pandas_dtype_invalid(self, dtype):
+        try:
+            pandas_dtype(self.data_dict[dtype])
+        except TypeError:
+            pass
+
+
+class SelectDtypes:
+    try:
+        params = [
+            tm.ALL_INT_NUMPY_DTYPES
+            + tm.ALL_INT_EA_DTYPES
+            + tm.FLOAT_NUMPY_DTYPES
+            + tm.COMPLEX_DTYPES
+            + tm.DATETIME64_DTYPES
+            + tm.TIMEDELTA64_DTYPES
+            + tm.BOOL_DTYPES
+        ]
+    except AttributeError:
+        params = [
+            tm.ALL_INT_DTYPES
+            + tm.ALL_EA_INT_DTYPES
+            + tm.FLOAT_DTYPES
+            + tm.COMPLEX_DTYPES
+            + tm.DATETIME64_DTYPES
+            + tm.TIMEDELTA64_DTYPES
+            + tm.BOOL_DTYPES
+        ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N, K = 5000, 50
+        self.index = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.columns = Index([f"i-{i}" for i in range(K)], dtype=object)
+
+        def create_df(data):
+            return DataFrame(data, index=self.index, columns=self.columns)
+
+        self.df_int = create_df(np.random.randint(low=100, size=(N, K)))
+        self.df_float = create_df(np.random.randn(N, K))
+        self.df_bool = create_df(np.random.choice([True, False], size=(N, K)))
+        self.df_string = create_df(
+            np.random.choice(list(string.ascii_letters), size=(N, K))
+        )
+
+    def time_select_dtype_int_include(self, dtype):
+        self.df_int.select_dtypes(include=dtype)
+
+    def time_select_dtype_int_exclude(self, dtype):
+        self.df_int.select_dtypes(exclude=dtype)
+
+    def time_select_dtype_float_include(self, dtype):
+        self.df_float.select_dtypes(include=dtype)
+
+    def time_select_dtype_float_exclude(self, dtype):
+        self.df_float.select_dtypes(exclude=dtype)
+
+    def time_select_dtype_bool_include(self, dtype):
+        self.df_bool.select_dtypes(include=dtype)
+
+    def time_select_dtype_bool_exclude(self, dtype):
+        self.df_bool.select_dtypes(exclude=dtype)
+
+    def time_select_dtype_string_include(self, dtype):
+        self.df_string.select_dtypes(include=dtype)
+
+    def time_select_dtype_string_exclude(self, dtype):
+        self.df_string.select_dtypes(exclude=dtype)
+
+
+class CheckDtypes:
+    def setup(self):
+        self.ext_dtype = pd.Int64Dtype()
+        self.np_dtype = np.dtype("int64")
+
+    def time_is_extension_array_dtype_true(self):
+        is_extension_array_dtype(self.ext_dtype)
+
+    def time_is_extension_array_dtype_false(self):
+        is_extension_array_dtype(self.np_dtype)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/eval.py b/pandas/asv_bench/benchmarks/eval.py
new file mode 100644
index 0000000000000..656d16a910a9f
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/eval.py
@@ -0,0 +1,65 @@
+import numpy as np
+
+import pandas as pd
+
+try:
+    import pandas.core.computation.expressions as expr
+except ImportError:
+    import pandas.computation.expressions as expr
+
+
+class Eval:
+    params = [["numexpr", "python"], [1, "all"]]
+    param_names = ["engine", "threads"]
+
+    def setup(self, engine, threads):
+        self.df = pd.DataFrame(np.random.randn(20000, 100))
+        self.df2 = pd.DataFrame(np.random.randn(20000, 100))
+        self.df3 = pd.DataFrame(np.random.randn(20000, 100))
+        self.df4 = pd.DataFrame(np.random.randn(20000, 100))
+
+        if threads == 1:
+            expr.set_numexpr_threads(1)
+
+    def time_add(self, engine, threads):
+        pd.eval("self.df + self.df2 + self.df3 + self.df4", engine=engine)
+
+    def time_and(self, engine, threads):
+        pd.eval(
+            "(self.df > 0) & (self.df2 > 0) & (self.df3 > 0) & (self.df4 > 0)",
+            engine=engine,
+        )
+
+    def time_chained_cmp(self, engine, threads):
+        pd.eval("self.df < self.df2 < self.df3 < self.df4", engine=engine)
+
+    def time_mult(self, engine, threads):
+        pd.eval("self.df * self.df2 * self.df3 * self.df4", engine=engine)
+
+    def teardown(self, engine, threads):
+        expr.set_numexpr_threads()
+
+
+class Query:
+    def setup(self):
+        N = 10**6
+        halfway = (N // 2) - 1
+        index = pd.date_range("20010101", periods=N, freq="min")
+        s = pd.Series(index)
+        self.ts = s.iloc[halfway]
+        self.df = pd.DataFrame({"a": np.random.randn(N), "dates": index}, index=index)
+        data = np.random.randn(N)
+        self.min_val = data.min()
+        self.max_val = data.max()
+
+    def time_query_datetime_index(self):
+        self.df.query("index < @self.ts")
+
+    def time_query_datetime_column(self):
+        self.df.query("dates < @self.ts")
+
+    def time_query_with_boolean_selection(self):
+        self.df.query("(a >= @self.min_val) & (a <= @self.max_val)")
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/finalize.py b/pandas/asv_bench/benchmarks/finalize.py
new file mode 100644
index 0000000000000..dc06f55cc6ca0
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/finalize.py
@@ -0,0 +1,16 @@
+import pandas as pd
+
+
+class Finalize:
+    param_names = ["series", "frame"]
+    params = [pd.Series, pd.DataFrame]
+
+    def setup(self, param):
+        N = 1000
+        obj = param(dtype=float)
+        for i in range(N):
+            obj.attrs[i] = i
+        self.obj = obj
+
+    def time_finalize_micro(self, param):
+        self.obj.__finalize__(self.obj, method="__finalize__")
diff --git a/pandas/asv_bench/benchmarks/frame_ctor.py b/pandas/asv_bench/benchmarks/frame_ctor.py
new file mode 100644
index 0000000000000..f938f7eb0d951
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/frame_ctor.py
@@ -0,0 +1,201 @@
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    NA,
+    Categorical,
+    DataFrame,
+    Float64Dtype,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+
+try:
+    from pandas.tseries.offsets import (
+        Hour,
+        Nano,
+    )
+except ImportError:
+    # For compatibility with older versions
+    from pandas.core.datetools import (
+        Hour,
+        Nano,
+    )
+
+
+class FromDicts:
+    def setup(self):
+        N, K = 5000, 50
+        self.index = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.columns = pd.Index([f"i-{i}" for i in range(K)], dtype=object)
+        frame = DataFrame(np.random.randn(N, K), index=self.index, columns=self.columns)
+        self.data = frame.to_dict()
+        self.dict_list = frame.to_dict(orient="records")
+        self.data2 = {i: {j: float(j) for j in range(100)} for i in range(2000)}
+
+        # arrays which we won't consolidate
+        self.dict_of_categoricals = {i: Categorical(np.arange(N)) for i in range(K)}
+
+    def time_list_of_dict(self):
+        DataFrame(self.dict_list)
+
+    def time_nested_dict(self):
+        DataFrame(self.data)
+
+    def time_nested_dict_index(self):
+        DataFrame(self.data, index=self.index)
+
+    def time_nested_dict_columns(self):
+        DataFrame(self.data, columns=self.columns)
+
+    def time_nested_dict_index_columns(self):
+        DataFrame(self.data, index=self.index, columns=self.columns)
+
+    def time_nested_dict_int64(self):
+        # nested dict, integer indexes, regression described in #621
+        DataFrame(self.data2)
+
+    def time_dict_of_categoricals(self):
+        # dict of arrays that we won't consolidate
+        DataFrame(self.dict_of_categoricals)
+
+
+class FromSeries:
+    def setup(self):
+        mi = MultiIndex.from_product([range(100), range(100)])
+        self.s = Series(np.random.randn(10000), index=mi)
+
+    def time_mi_series(self):
+        DataFrame(self.s)
+
+
+class FromDictwithTimestamp:
+    params = [Nano(1), Hour(1)]
+    param_names = ["offset"]
+
+    def setup(self, offset):
+        N = 10**3
+        idx = date_range(Timestamp("1/1/1900"), freq=offset, periods=N)
+        df = DataFrame(np.random.randn(N, 10), index=idx)
+        self.d = df.to_dict()
+
+    def time_dict_with_timestamp_offsets(self, offset):
+        DataFrame(self.d)
+
+
+class FromRecords:
+    params = [None, 1000]
+    param_names = ["nrows"]
+
+    # Generators get exhausted on use, so run setup before every call
+    number = 1
+    repeat = (3, 250, 10)
+
+    def setup(self, nrows):
+        N = 100000
+        self.gen = ((x, (x * 20), (x * 100)) for x in range(N))
+
+    def time_frame_from_records_generator(self, nrows):
+        # issue-6700
+        self.df = DataFrame.from_records(self.gen, nrows=nrows)
+
+
+class FromNDArray:
+    def setup(self):
+        N = 100000
+        self.data = np.random.randn(N)
+
+    def time_frame_from_ndarray(self):
+        self.df = DataFrame(self.data)
+
+
+class FromLists:
+    goal_time = 0.2
+
+    def setup(self):
+        N = 1000
+        M = 100
+        self.data = [list(range(M)) for i in range(N)]
+
+    def time_frame_from_lists(self):
+        self.df = DataFrame(self.data)
+
+
+class FromRange:
+    goal_time = 0.2
+
+    def setup(self):
+        N = 1_000_000
+        self.data = range(N)
+
+    def time_frame_from_range(self):
+        self.df = DataFrame(self.data)
+
+
+class FromScalar:
+    def setup(self):
+        self.nrows = 100_000
+
+    def time_frame_from_scalar_ea_float64(self):
+        DataFrame(
+            1.0,
+            index=range(self.nrows),
+            columns=list("abc"),
+            dtype=Float64Dtype(),
+        )
+
+    def time_frame_from_scalar_ea_float64_na(self):
+        DataFrame(
+            NA,
+            index=range(self.nrows),
+            columns=list("abc"),
+            dtype=Float64Dtype(),
+        )
+
+
+class FromArrays:
+    goal_time = 0.2
+
+    def setup(self):
+        N_rows = 1000
+        N_cols = 1000
+        self.float_arrays = [np.random.randn(N_rows) for _ in range(N_cols)]
+        self.sparse_arrays = [
+            pd.arrays.SparseArray(np.random.randint(0, 2, N_rows), dtype="float64")
+            for _ in range(N_cols)
+        ]
+        self.int_arrays = [
+            pd.array(np.random.randint(1000, size=N_rows), dtype="Int64")
+            for _ in range(N_cols)
+        ]
+        self.index = pd.Index(range(N_rows))
+        self.columns = pd.Index(range(N_cols))
+
+    def time_frame_from_arrays_float(self):
+        self.df = DataFrame._from_arrays(
+            self.float_arrays,
+            index=self.index,
+            columns=self.columns,
+            verify_integrity=False,
+        )
+
+    def time_frame_from_arrays_int(self):
+        self.df = DataFrame._from_arrays(
+            self.int_arrays,
+            index=self.index,
+            columns=self.columns,
+            verify_integrity=False,
+        )
+
+    def time_frame_from_arrays_sparse(self):
+        self.df = DataFrame._from_arrays(
+            self.sparse_arrays,
+            index=self.index,
+            columns=self.columns,
+            verify_integrity=False,
+        )
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/frame_methods.py b/pandas/asv_bench/benchmarks/frame_methods.py
new file mode 100644
index 0000000000000..6a2ab24df26fe
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/frame_methods.py
@@ -0,0 +1,889 @@
+import string
+import warnings
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    NaT,
+    Series,
+    date_range,
+    isnull,
+    period_range,
+    timedelta_range,
+)
+
+
+class AsType:
+    params = [
+        [
+            # from_dtype == to_dtype
+            ("Float64", "Float64"),
+            ("float64[pyarrow]", "float64[pyarrow]"),
+            # from non-EA to EA
+            ("float64", "Float64"),
+            ("float64", "float64[pyarrow]"),
+            # from EA to non-EA
+            ("Float64", "float64"),
+            ("float64[pyarrow]", "float64"),
+            # from EA to EA
+            ("Int64", "Float64"),
+            ("int64[pyarrow]", "float64[pyarrow]"),
+        ],
+        [False, True],
+    ]
+    param_names = ["from_to_dtypes", "copy"]
+
+    def setup(self, from_to_dtypes, copy):
+        from_dtype = from_to_dtypes[0]
+        if from_dtype in ("float64", "Float64", "float64[pyarrow]"):
+            data = np.random.randn(100, 100)
+        elif from_dtype in ("int64", "Int64", "int64[pyarrow]"):
+            data = np.random.randint(0, 1000, (100, 100))
+        else:
+            raise NotImplementedError
+        self.df = DataFrame(data, dtype=from_dtype)
+
+    def time_astype(self, from_to_dtypes, copy):
+        self.df.astype(from_to_dtypes[1], copy=copy)
+
+
+class Clip:
+    params = [
+        ["float64", "Float64", "float64[pyarrow]"],
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        data = np.random.randn(100_000, 10)
+        df = DataFrame(data, dtype=dtype)
+        self.df = df
+
+    def time_clip(self, dtype):
+        self.df.clip(-1.0, 1.0)
+
+
+class GetNumericData:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(10000, 25))
+        self.df["foo"] = "bar"
+        self.df["bar"] = "baz"
+        self.df = self.df._consolidate()
+
+    def time_frame_get_numeric_data(self):
+        self.df._get_numeric_data()
+
+
+class Reindex:
+    def setup(self):
+        N = 10**3
+        self.df = DataFrame(np.random.randn(N * 10, N))
+        self.idx = np.arange(4 * N, 7 * N)
+        self.idx_cols = np.random.randint(0, N, N)
+        self.df2 = DataFrame(
+            {
+                c: {
+                    0: np.random.randint(0, 2, N).astype(np.bool_),
+                    1: np.random.randint(0, N, N).astype(np.int16),
+                    2: np.random.randint(0, N, N).astype(np.int32),
+                    3: np.random.randint(0, N, N).astype(np.int64),
+                }[np.random.randint(0, 4)]
+                for c in range(N)
+            }
+        )
+
+    def time_reindex_axis0(self):
+        self.df.reindex(self.idx)
+
+    def time_reindex_axis1(self):
+        self.df.reindex(columns=self.idx_cols)
+
+    def time_reindex_axis1_missing(self):
+        self.df.reindex(columns=self.idx)
+
+    def time_reindex_both_axes(self):
+        self.df.reindex(index=self.idx, columns=self.idx_cols)
+
+    def time_reindex_upcast(self):
+        self.df2.reindex(np.random.permutation(range(1200)))
+
+
+class Rename:
+    def setup(self):
+        N = 10**3
+        self.df = DataFrame(np.random.randn(N * 10, N))
+        self.idx = np.arange(4 * N, 7 * N)
+        self.dict_idx = {k: k for k in self.idx}
+        self.df2 = DataFrame(
+            {
+                c: {
+                    0: np.random.randint(0, 2, N).astype(np.bool_),
+                    1: np.random.randint(0, N, N).astype(np.int16),
+                    2: np.random.randint(0, N, N).astype(np.int32),
+                    3: np.random.randint(0, N, N).astype(np.int64),
+                }[np.random.randint(0, 4)]
+                for c in range(N)
+            }
+        )
+
+    def time_rename_single(self):
+        self.df.rename({0: 0})
+
+    def time_rename_axis0(self):
+        self.df.rename(self.dict_idx)
+
+    def time_rename_axis1(self):
+        self.df.rename(columns=self.dict_idx)
+
+    def time_rename_both_axes(self):
+        self.df.rename(index=self.dict_idx, columns=self.dict_idx)
+
+    def time_dict_rename_both_axes(self):
+        self.df.rename(index=self.dict_idx, columns=self.dict_idx)
+
+
+class Iteration:
+    # mem_itertuples_* benchmarks are slow
+    timeout = 120
+
+    def setup(self):
+        N = 1000
+        self.df = DataFrame(np.random.randn(N * 10, N))
+        self.df2 = DataFrame(np.random.randn(N * 50, 10))
+        self.df3 = DataFrame(
+            np.random.randn(N, 5 * N), columns=["C" + str(c) for c in range(N * 5)]
+        )
+        self.df4 = DataFrame(np.random.randn(N * 1000, 10))
+
+    def time_items(self):
+        # (monitor no-copying behaviour)
+        for name, col in self.df.items():
+            pass
+
+    def time_iteritems_indexing(self):
+        for col in self.df3:
+            self.df3[col]
+
+    def time_itertuples_start(self):
+        self.df4.itertuples()
+
+    def time_itertuples_read_first(self):
+        next(self.df4.itertuples())
+
+    def time_itertuples(self):
+        for row in self.df4.itertuples():
+            pass
+
+    def time_itertuples_to_list(self):
+        list(self.df4.itertuples())
+
+    def mem_itertuples_start(self):
+        return self.df4.itertuples()
+
+    def peakmem_itertuples_start(self):
+        self.df4.itertuples()
+
+    def mem_itertuples_read_first(self):
+        return next(self.df4.itertuples())
+
+    def peakmem_itertuples(self):
+        for row in self.df4.itertuples():
+            pass
+
+    def mem_itertuples_to_list(self):
+        return list(self.df4.itertuples())
+
+    def peakmem_itertuples_to_list(self):
+        list(self.df4.itertuples())
+
+    def time_itertuples_raw_start(self):
+        self.df4.itertuples(index=False, name=None)
+
+    def time_itertuples_raw_read_first(self):
+        next(self.df4.itertuples(index=False, name=None))
+
+    def time_itertuples_raw_tuples(self):
+        for row in self.df4.itertuples(index=False, name=None):
+            pass
+
+    def time_itertuples_raw_tuples_to_list(self):
+        list(self.df4.itertuples(index=False, name=None))
+
+    def mem_itertuples_raw_start(self):
+        return self.df4.itertuples(index=False, name=None)
+
+    def peakmem_itertuples_raw_start(self):
+        self.df4.itertuples(index=False, name=None)
+
+    def peakmem_itertuples_raw_read_first(self):
+        next(self.df4.itertuples(index=False, name=None))
+
+    def peakmem_itertuples_raw(self):
+        for row in self.df4.itertuples(index=False, name=None):
+            pass
+
+    def mem_itertuples_raw_to_list(self):
+        return list(self.df4.itertuples(index=False, name=None))
+
+    def peakmem_itertuples_raw_to_list(self):
+        list(self.df4.itertuples(index=False, name=None))
+
+    def time_iterrows(self):
+        for row in self.df.iterrows():
+            pass
+
+
+class ToString:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(100, 10))
+
+    def time_to_string_floats(self):
+        self.df.to_string()
+
+
+class ToHTML:
+    def setup(self):
+        nrows = 500
+        self.df2 = DataFrame(np.random.randn(nrows, 10))
+        self.df2[0] = period_range("2000", periods=nrows)
+        self.df2[1] = range(nrows)
+
+    def time_to_html_mixed(self):
+        self.df2.to_html()
+
+
+class ToDict:
+    params = [["dict", "list", "series", "split", "records", "index"]]
+    param_names = ["orient"]
+
+    def setup(self, orient):
+        data = np.random.randint(0, 1000, size=(10000, 4))
+        self.int_df = DataFrame(data)
+        self.datetimelike_df = self.int_df.astype("timedelta64[ns]")
+
+    def time_to_dict_ints(self, orient):
+        self.int_df.to_dict(orient=orient)
+
+    def time_to_dict_datetimelike(self, orient):
+        self.datetimelike_df.to_dict(orient=orient)
+
+
+class ToNumpy:
+    def setup(self):
+        N = 10000
+        M = 10
+        self.df_tall = DataFrame(np.random.randn(N, M))
+        self.df_wide = DataFrame(np.random.randn(M, N))
+        self.df_mixed_tall = self.df_tall.copy()
+        self.df_mixed_tall["foo"] = "bar"
+        self.df_mixed_tall[0] = period_range("2000", periods=N)
+        self.df_mixed_tall[1] = range(N)
+        self.df_mixed_wide = self.df_wide.copy()
+        self.df_mixed_wide["foo"] = "bar"
+        self.df_mixed_wide[0] = period_range("2000", periods=M)
+        self.df_mixed_wide[1] = range(M)
+
+    def time_to_numpy_tall(self):
+        self.df_tall.to_numpy()
+
+    def time_to_numpy_wide(self):
+        self.df_wide.to_numpy()
+
+    def time_to_numpy_mixed_tall(self):
+        self.df_mixed_tall.to_numpy()
+
+    def time_to_numpy_mixed_wide(self):
+        self.df_mixed_wide.to_numpy()
+
+    def time_values_tall(self):
+        self.df_tall.values
+
+    def time_values_wide(self):
+        self.df_wide.values
+
+    def time_values_mixed_tall(self):
+        self.df_mixed_tall.values
+
+    def time_values_mixed_wide(self):
+        self.df_mixed_wide.values
+
+
+class ToRecords:
+    def setup(self):
+        N = 100_000
+        data = np.random.randn(N, 2)
+        mi = MultiIndex.from_arrays(
+            [
+                np.arange(N),
+                date_range("1970-01-01", periods=N, freq="ms"),
+            ]
+        )
+        self.df = DataFrame(data)
+        self.df_mi = DataFrame(data, index=mi)
+
+    def time_to_records(self):
+        self.df.to_records(index=True)
+
+    def time_to_records_multiindex(self):
+        self.df_mi.to_records(index=True)
+
+
+class Repr:
+    def setup(self):
+        nrows = 10000
+        data = np.random.randn(nrows, 10)
+        arrays = np.tile(np.random.randn(3, nrows // 100), 100)
+        idx = MultiIndex.from_arrays(arrays)
+        self.df3 = DataFrame(data, index=idx)
+        self.df4 = DataFrame(data, index=np.random.randn(nrows))
+        self.df_tall = DataFrame(np.random.randn(nrows, 10))
+        self.df_wide = DataFrame(np.random.randn(10, nrows))
+
+    def time_html_repr_trunc_mi(self):
+        self.df3._repr_html_()
+
+    def time_html_repr_trunc_si(self):
+        self.df4._repr_html_()
+
+    def time_repr_tall(self):
+        repr(self.df_tall)
+
+    def time_frame_repr_wide(self):
+        repr(self.df_wide)
+
+
+class MaskBool:
+    def setup(self):
+        data = np.random.randn(1000, 500)
+        df = DataFrame(data)
+        df = df.where(df > 0)
+        self.bools = df > 0
+        self.mask = isnull(df)
+
+    def time_frame_mask_bools(self):
+        self.bools.mask(self.mask)
+
+    def time_frame_mask_floats(self):
+        self.bools.astype(float).mask(self.mask)
+
+
+class Isnull:
+    def setup(self):
+        N = 10**3
+        self.df_no_null = DataFrame(np.random.randn(N, N))
+
+        sample = np.array([np.nan, 1.0])
+        data = np.random.choice(sample, (N, N))
+        self.df = DataFrame(data)
+
+        sample = np.array(list(string.ascii_letters + string.whitespace))
+        data = np.random.choice(sample, (N, N))
+        self.df_strings = DataFrame(data)
+
+        sample = np.array(
+            [
+                NaT,
+                np.nan,
+                None,
+                np.datetime64("NaT"),
+                np.timedelta64("NaT"),
+                0,
+                1,
+                2.0,
+                "",
+                "abcd",
+            ]
+        )
+        data = np.random.choice(sample, (N, N))
+        self.df_obj = DataFrame(data)
+
+    def time_isnull_floats_no_null(self):
+        isnull(self.df_no_null)
+
+    def time_isnull(self):
+        isnull(self.df)
+
+    def time_isnull_strngs(self):
+        isnull(self.df_strings)
+
+    def time_isnull_obj(self):
+        isnull(self.df_obj)
+
+
+class Fillna:
+    params = (
+        [True, False],
+        [
+            "float64",
+            "float32",
+            "object",
+            "Int64",
+            "Float64",
+            "datetime64[ns]",
+            "datetime64[ns, tz]",
+            "timedelta64[ns]",
+        ],
+    )
+    param_names = ["inplace", "dtype"]
+
+    def setup(self, inplace, dtype):
+        N, M = 10000, 100
+        if dtype in ("datetime64[ns]", "datetime64[ns, tz]", "timedelta64[ns]"):
+            data = {
+                "datetime64[ns]": date_range("2011-01-01", freq="h", periods=N),
+                "datetime64[ns, tz]": date_range(
+                    "2011-01-01", freq="h", periods=N, tz="Asia/Tokyo"
+                ),
+                "timedelta64[ns]": timedelta_range(start="1 day", periods=N, freq="1D"),
+            }
+            self.df = DataFrame({f"col_{i}": data[dtype] for i in range(M)})
+            self.df[::2] = None
+        else:
+            values = np.random.randn(N, M)
+            values[::2] = np.nan
+            if dtype == "Int64":
+                values = values.round()
+            self.df = DataFrame(values, dtype=dtype)
+        self.fill_values = self.df.iloc[self.df.first_valid_index()].to_dict()
+
+    def time_fillna(self, inplace, dtype):
+        self.df.fillna(value=self.fill_values, inplace=inplace)
+
+    def time_ffill(self, inplace, dtype):
+        self.df.ffill(inplace=inplace)
+
+    def time_bfill(self, inplace, dtype):
+        self.df.bfill(inplace=inplace)
+
+
+class Dropna:
+    params = (["all", "any"], [0, 1])
+    param_names = ["how", "axis"]
+
+    def setup(self, how, axis):
+        self.df = DataFrame(np.random.randn(10000, 1000))
+        self.df.iloc[50:1000, 20:50] = np.nan
+        self.df.iloc[2000:3000] = np.nan
+        self.df.iloc[:, 60:70] = np.nan
+        self.df_mixed = self.df.copy()
+        self.df_mixed["foo"] = "bar"
+
+    def time_dropna(self, how, axis):
+        self.df.dropna(how=how, axis=axis)
+
+    def time_dropna_axis_mixed_dtypes(self, how, axis):
+        self.df_mixed.dropna(how=how, axis=axis)
+
+
+class Isna:
+    params = ["float64", "Float64", "float64[pyarrow]"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        data = np.random.randn(10000, 1000)
+        # all-na columns
+        data[:, 600:800] = np.nan
+        # partial-na columns
+        data[800:1000, 4000:5000] = np.nan
+        self.df = DataFrame(data, dtype=dtype)
+
+    def time_isna(self, dtype):
+        self.df.isna()
+
+
+class Count:
+    params = [0, 1]
+    param_names = ["axis"]
+
+    def setup(self, axis):
+        self.df = DataFrame(np.random.randn(10000, 1000))
+        self.df.iloc[50:1000, 20:50] = np.nan
+        self.df.iloc[2000:3000] = np.nan
+        self.df.iloc[:, 60:70] = np.nan
+        self.df_mixed = self.df.copy()
+        self.df_mixed["foo"] = "bar"
+
+    def time_count(self, axis):
+        self.df.count(axis=axis)
+
+    def time_count_mixed_dtypes(self, axis):
+        self.df_mixed.count(axis=axis)
+
+
+class Apply:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(1000, 100))
+
+        self.s = Series(np.arange(1028.0))
+        self.df2 = DataFrame({i: self.s for i in range(1028)})
+        self.df3 = DataFrame(np.random.randn(1000, 3), columns=list("ABC"))
+
+    def time_apply_user_func(self):
+        self.df2.apply(lambda x: np.corrcoef(x, self.s)[(0, 1)])
+
+    def time_apply_axis_1(self):
+        self.df.apply(lambda x: x + 1, axis=1)
+
+    def time_apply_lambda_mean(self):
+        self.df.apply(lambda x: x.mean())
+
+    def time_apply_str_mean(self):
+        self.df.apply("mean")
+
+    def time_apply_pass_thru(self):
+        self.df.apply(lambda x: x)
+
+    def time_apply_ref_by_name(self):
+        self.df3.apply(lambda x: x["A"] + x["B"], axis=1)
+
+
+class Dtypes:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(1000, 1000))
+
+    def time_frame_dtypes(self):
+        self.df.dtypes
+
+
+class Equals:
+    def setup(self):
+        N = 10**3
+        self.float_df = DataFrame(np.random.randn(N, N))
+        self.float_df_nan = self.float_df.copy()
+        self.float_df_nan.iloc[-1, -1] = np.nan
+
+        self.object_df = DataFrame("foo", index=range(N), columns=range(N))
+        self.object_df_nan = self.object_df.copy()
+        self.object_df_nan.iloc[-1, -1] = np.nan
+
+        self.nonunique_cols = self.object_df.copy()
+        self.nonunique_cols.columns = ["A"] * len(self.nonunique_cols.columns)
+        self.nonunique_cols_nan = self.nonunique_cols.copy()
+        self.nonunique_cols_nan.iloc[-1, -1] = np.nan
+
+    def time_frame_float_equal(self):
+        self.float_df.equals(self.float_df)
+
+    def time_frame_float_unequal(self):
+        self.float_df.equals(self.float_df_nan)
+
+    def time_frame_nonunique_equal(self):
+        self.nonunique_cols.equals(self.nonunique_cols)
+
+    def time_frame_nonunique_unequal(self):
+        self.nonunique_cols.equals(self.nonunique_cols_nan)
+
+    def time_frame_object_equal(self):
+        self.object_df.equals(self.object_df)
+
+    def time_frame_object_unequal(self):
+        self.object_df.equals(self.object_df_nan)
+
+
+class Interpolate:
+    def setup(self):
+        N = 10000
+        # this is the worst case, where every column has NaNs.
+        arr = np.random.randn(N, 100)
+        arr[::2] = np.nan
+
+        self.df = DataFrame(arr)
+
+        self.df2 = DataFrame(
+            {
+                "A": np.arange(0, N),
+                "B": np.random.randint(0, 100, N),
+                "C": np.random.randn(N),
+                "D": np.random.randn(N),
+            }
+        )
+        self.df2.loc[1::5, "A"] = np.nan
+        self.df2.loc[1::5, "C"] = np.nan
+
+    def time_interpolate(self):
+        self.df.interpolate()
+
+    def time_interpolate_some_good(self):
+        self.df2.interpolate()
+
+
+class Shift:
+    # frame shift speedup issue-5609
+    params = [0, 1]
+    param_names = ["axis"]
+
+    def setup(self, axis):
+        self.df = DataFrame(np.random.rand(10000, 500))
+
+    def time_shift(self, axis):
+        self.df.shift(1, axis=axis)
+
+
+class Nunique:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(10000, 1000))
+
+    def time_frame_nunique(self):
+        self.df.nunique()
+
+
+class SeriesNuniqueWithNan:
+    def setup(self):
+        values = 100 * [np.nan] + list(range(100))
+        self.ser = Series(np.tile(values, 10000), dtype=float)
+
+    def time_series_nunique_nan(self):
+        self.ser.nunique()
+
+
+class Duplicated:
+    def setup(self):
+        n = 1 << 20
+        t = date_range("2015-01-01", freq="s", periods=(n // 64))
+        xs = np.random.randn(n // 64).round(2)
+        self.df = DataFrame(
+            {
+                "a": np.random.randint(-1 << 8, 1 << 8, n),
+                "b": np.random.choice(t, n),
+                "c": np.random.choice(xs, n),
+            }
+        )
+        self.df2 = DataFrame(np.random.randn(1000, 100).astype(str)).T
+
+    def time_frame_duplicated(self):
+        self.df.duplicated()
+
+    def time_frame_duplicated_wide(self):
+        self.df2.duplicated()
+
+    def time_frame_duplicated_subset(self):
+        self.df.duplicated(subset=["a"])
+
+
+class XS:
+    params = [0, 1]
+    param_names = ["axis"]
+
+    def setup(self, axis):
+        self.N = 10**4
+        self.df = DataFrame(np.random.randn(self.N, self.N))
+
+    def time_frame_xs(self, axis):
+        self.df.xs(self.N / 2, axis=axis)
+
+
+class SortValues:
+    params = [True, False]
+    param_names = ["ascending"]
+
+    def setup(self, ascending):
+        self.df = DataFrame(np.random.randn(1000000, 2), columns=list("AB"))
+
+    def time_frame_sort_values(self, ascending):
+        self.df.sort_values(by="A", ascending=ascending)
+
+
+class SortMultiKey:
+    params = [True, False]
+    param_names = ["monotonic"]
+
+    def setup(self, monotonic):
+        N = 10000
+        K = 10
+        df = DataFrame(
+            {
+                "key1": Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(
+                    K
+                ),
+                "key2": Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(
+                    K
+                ),
+                "value": np.random.randn(N * K),
+            }
+        )
+        if monotonic:
+            df = df.sort_values(["key1", "key2"])
+        self.df_by_columns = df
+        self.df_by_index = df.set_index(["key1", "key2"])
+
+    def time_sort_values(self, monotonic):
+        self.df_by_columns.sort_values(by=["key1", "key2"])
+
+    def time_sort_index(self, monotonic):
+        self.df_by_index.sort_index()
+
+
+class Quantile:
+    params = [0, 1]
+    param_names = ["axis"]
+
+    def setup(self, axis):
+        self.df = DataFrame(np.random.randn(1000, 3), columns=list("ABC"))
+
+    def time_frame_quantile(self, axis):
+        self.df.quantile([0.1, 0.5], axis=axis)
+
+
+class Rank:
+    param_names = ["dtype"]
+    params = [
+        ["int", "uint", "float", "object"],
+    ]
+
+    def setup(self, dtype):
+        self.df = DataFrame(
+            np.random.randn(10000, 10).astype(dtype), columns=range(10), dtype=dtype
+        )
+
+    def time_rank(self, dtype):
+        self.df.rank()
+
+
+class GetDtypeCounts:
+    # 2807
+    def setup(self):
+        self.df = DataFrame(np.random.randn(10, 10000))
+
+    def time_frame_get_dtype_counts(self):
+        with warnings.catch_warnings(record=True):
+            self.df.dtypes.value_counts()
+
+    def time_info(self):
+        self.df.info()
+
+
+class NSort:
+    params = ["first", "last", "all"]
+    param_names = ["keep"]
+
+    def setup(self, keep):
+        self.df = DataFrame(np.random.randn(100000, 3), columns=list("ABC"))
+
+    def time_nlargest_one_column(self, keep):
+        self.df.nlargest(100, "A", keep=keep)
+
+    def time_nlargest_two_columns(self, keep):
+        self.df.nlargest(100, ["A", "B"], keep=keep)
+
+    def time_nsmallest_one_column(self, keep):
+        self.df.nsmallest(100, "A", keep=keep)
+
+    def time_nsmallest_two_columns(self, keep):
+        self.df.nsmallest(100, ["A", "B"], keep=keep)
+
+
+class Describe:
+    def setup(self):
+        self.df = DataFrame(
+            {
+                "a": np.random.randint(0, 100, 10**6),
+                "b": np.random.randint(0, 100, 10**6),
+                "c": np.random.randint(0, 100, 10**6),
+            }
+        )
+
+    def time_series_describe(self):
+        self.df["a"].describe()
+
+    def time_dataframe_describe(self):
+        self.df.describe()
+
+
+class MemoryUsage:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(100000, 2), columns=list("AB"))
+        self.df2 = self.df.copy()
+        self.df2["A"] = self.df2["A"].astype("object")
+
+    def time_memory_usage(self):
+        self.df.memory_usage(deep=True)
+
+    def time_memory_usage_object_dtype(self):
+        self.df2.memory_usage(deep=True)
+
+
+class Round:
+    def setup(self):
+        self.df = DataFrame(np.random.randn(10000, 10))
+        self.df_t = self.df.transpose(copy=True)
+
+    def time_round(self):
+        self.df.round()
+
+    def time_round_transposed(self):
+        self.df_t.round()
+
+    def peakmem_round(self):
+        self.df.round()
+
+    def peakmem_round_transposed(self):
+        self.df_t.round()
+
+
+class Where:
+    params = (
+        [True, False],
+        ["float64", "Float64", "float64[pyarrow]"],
+    )
+    param_names = ["dtype"]
+
+    def setup(self, inplace, dtype):
+        self.df = DataFrame(np.random.randn(100_000, 10), dtype=dtype)
+        self.mask = self.df < 0
+
+    def time_where(self, inplace, dtype):
+        self.df.where(self.mask, other=0.0, inplace=inplace)
+
+
+class FindValidIndex:
+    param_names = ["dtype"]
+    params = [
+        ["float", "Float64", "float64[pyarrow]"],
+    ]
+
+    def setup(self, dtype):
+        df = DataFrame(
+            np.random.randn(100000, 2),
+            columns=list("AB"),
+            dtype=dtype,
+        )
+        df.iloc[:100, 0] = None
+        df.iloc[:200, 1] = None
+        df.iloc[-100:, 0] = None
+        df.iloc[-200:, 1] = None
+        self.df = df
+
+    def time_first_valid_index(self, dtype):
+        self.df.first_valid_index()
+
+    def time_last_valid_index(self, dtype):
+        self.df.last_valid_index()
+
+
+class Update:
+    def setup(self):
+        rng = np.random.default_rng()
+        self.df = DataFrame(rng.uniform(size=(1_000_000, 10)))
+
+        idx = rng.choice(range(1_000_000), size=1_000_000, replace=False)
+        self.df_random = DataFrame(self.df, index=idx)
+
+        idx = rng.choice(range(1_000_000), size=100_000, replace=False)
+        cols = rng.choice(range(10), size=2, replace=False)
+        self.df_sample = DataFrame(
+            rng.uniform(size=(100_000, 2)), index=idx, columns=cols
+        )
+
+    def time_to_update_big_frame_small_arg(self):
+        self.df.update(self.df_sample)
+
+    def time_to_update_random_indices(self):
+        self.df_random.update(self.df_sample)
+
+    def time_to_update_small_frame_big_arg(self):
+        self.df_sample.update(self.df)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/gil.py b/pandas/asv_bench/benchmarks/gil.py
new file mode 100644
index 0000000000000..a0c4189c72d0e
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/gil.py
@@ -0,0 +1,327 @@
+from functools import wraps
+import threading
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    factorize,
+    read_csv,
+)
+from pandas.core.algorithms import take_nd
+
+try:
+    from pandas import (
+        rolling_kurt,
+        rolling_max,
+        rolling_mean,
+        rolling_median,
+        rolling_min,
+        rolling_skew,
+        rolling_std,
+        rolling_var,
+    )
+
+    have_rolling_methods = True
+except ImportError:
+    have_rolling_methods = False
+try:
+    from pandas._libs import algos
+except ImportError:
+    from pandas import algos
+
+from .pandas_vb_common import BaseIO  # isort:skip
+
+
+def test_parallel(num_threads=2, kwargs_list=None):
+    """
+    Decorator to run the same function multiple times in parallel.
+
+    Parameters
+    ----------
+    num_threads : int, optional
+        The number of times the function is run in parallel.
+    kwargs_list : list of dicts, optional
+        The list of kwargs to update original
+        function kwargs on different threads.
+
+    Notes
+    -----
+    This decorator does not pass the return value of the decorated function.
+
+    Original from scikit-image:
+
+    https://github.com/scikit-image/scikit-image/pull/1519
+
+    """
+    assert num_threads > 0
+    has_kwargs_list = kwargs_list is not None
+    if has_kwargs_list:
+        assert len(kwargs_list) == num_threads
+
+    def wrapper(func):
+        @wraps(func)
+        def inner(*args, **kwargs):
+            if has_kwargs_list:
+                update_kwargs = lambda i: dict(kwargs, **kwargs_list[i])
+            else:
+                update_kwargs = lambda i: kwargs
+            threads = []
+            for i in range(num_threads):
+                updated_kwargs = update_kwargs(i)
+                thread = threading.Thread(target=func, args=args, kwargs=updated_kwargs)
+                threads.append(thread)
+            for thread in threads:
+                thread.start()
+            for thread in threads:
+                thread.join()
+
+        return inner
+
+    return wrapper
+
+
+class ParallelGroupbyMethods:
+    params = ([2, 4, 8], ["count", "last", "max", "mean", "min", "prod", "sum", "var"])
+    param_names = ["threads", "method"]
+
+    def setup(self, threads, method):
+        N = 10**6
+        ngroups = 10**3
+        df = DataFrame(
+            {"key": np.random.randint(0, ngroups, size=N), "data": np.random.randn(N)}
+        )
+
+        @test_parallel(num_threads=threads)
+        def parallel():
+            getattr(df.groupby("key")["data"], method)()
+
+        self.parallel = parallel
+
+        def loop():
+            getattr(df.groupby("key")["data"], method)()
+
+        self.loop = loop
+
+    def time_parallel(self, threads, method):
+        self.parallel()
+
+    def time_loop(self, threads, method):
+        for i in range(threads):
+            self.loop()
+
+
+class ParallelGroups:
+    params = [2, 4, 8]
+    param_names = ["threads"]
+
+    def setup(self, threads):
+        size = 2**22
+        ngroups = 10**3
+        data = Series(np.random.randint(0, ngroups, size=size))
+
+        @test_parallel(num_threads=threads)
+        def get_groups():
+            data.groupby(data).groups
+
+        self.get_groups = get_groups
+
+    def time_get_groups(self, threads):
+        self.get_groups()
+
+
+class ParallelTake1D:
+    params = ["int64", "float64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**6
+        df = DataFrame({"col": np.arange(N, dtype=dtype)})
+        indexer = np.arange(100, len(df) - 100)
+
+        @test_parallel(num_threads=2)
+        def parallel_take1d():
+            take_nd(df["col"].values, indexer)
+
+        self.parallel_take1d = parallel_take1d
+
+    def time_take1d(self, dtype):
+        self.parallel_take1d()
+
+
+class ParallelKth:
+    # This depends exclusively on code in _libs/, could go in libs.py
+
+    number = 1
+    repeat = 5
+
+    def setup(self):
+        N = 10**7
+        k = 5 * 10**5
+        kwargs_list = [{"arr": np.random.randn(N)}, {"arr": np.random.randn(N)}]
+
+        @test_parallel(num_threads=2, kwargs_list=kwargs_list)
+        def parallel_kth_smallest(arr):
+            algos.kth_smallest(arr, k)
+
+        self.parallel_kth_smallest = parallel_kth_smallest
+
+    def time_kth_smallest(self):
+        self.parallel_kth_smallest()
+
+
+class ParallelDatetimeFields:
+    def setup(self):
+        N = 10**6
+        self.dti = date_range("1900-01-01", periods=N, freq="min")
+        self.period = self.dti.to_period("D")
+
+    def time_datetime_field_year(self):
+        @test_parallel(num_threads=2)
+        def run(dti):
+            dti.year
+
+        run(self.dti)
+
+    def time_datetime_field_day(self):
+        @test_parallel(num_threads=2)
+        def run(dti):
+            dti.day
+
+        run(self.dti)
+
+    def time_datetime_field_daysinmonth(self):
+        @test_parallel(num_threads=2)
+        def run(dti):
+            dti.days_in_month
+
+        run(self.dti)
+
+    def time_datetime_field_normalize(self):
+        @test_parallel(num_threads=2)
+        def run(dti):
+            dti.normalize()
+
+        run(self.dti)
+
+    def time_datetime_to_period(self):
+        @test_parallel(num_threads=2)
+        def run(dti):
+            dti.to_period("s")
+
+        run(self.dti)
+
+    def time_period_to_datetime(self):
+        @test_parallel(num_threads=2)
+        def run(period):
+            period.to_timestamp()
+
+        run(self.period)
+
+
+class ParallelRolling:
+    params = ["median", "mean", "min", "max", "var", "skew", "kurt", "std"]
+    param_names = ["method"]
+
+    def setup(self, method):
+        win = 100
+        arr = np.random.rand(100000)
+        if hasattr(DataFrame, "rolling"):
+            df = DataFrame(arr).rolling(win)
+
+            @test_parallel(num_threads=2)
+            def parallel_rolling():
+                getattr(df, method)()
+
+            self.parallel_rolling = parallel_rolling
+        elif have_rolling_methods:
+            rolling = {
+                "median": rolling_median,
+                "mean": rolling_mean,
+                "min": rolling_min,
+                "max": rolling_max,
+                "var": rolling_var,
+                "skew": rolling_skew,
+                "kurt": rolling_kurt,
+                "std": rolling_std,
+            }
+
+            @test_parallel(num_threads=2)
+            def parallel_rolling():
+                rolling[method](arr, win)
+
+            self.parallel_rolling = parallel_rolling
+        else:
+            raise NotImplementedError
+
+    def time_rolling(self, method):
+        self.parallel_rolling()
+
+
+class ParallelReadCSV(BaseIO):
+    number = 1
+    repeat = 5
+    params = ["float", "object", "datetime"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        rows = 10000
+        cols = 50
+        if dtype == "float":
+            df = DataFrame(np.random.randn(rows, cols))
+        elif dtype == "datetime":
+            df = DataFrame(
+                np.random.randn(rows, cols), index=date_range("1/1/2000", periods=rows)
+            )
+        elif dtype == "object":
+            df = DataFrame(
+                "foo", index=range(rows), columns=["object%03d" for _ in range(5)]
+            )
+        else:
+            raise NotImplementedError
+
+        self.fname = f"__test_{dtype}__.csv"
+        df.to_csv(self.fname)
+
+        @test_parallel(num_threads=2)
+        def parallel_read_csv():
+            read_csv(self.fname)
+
+        self.parallel_read_csv = parallel_read_csv
+
+    def time_read_csv(self, dtype):
+        self.parallel_read_csv()
+
+
+class ParallelFactorize:
+    number = 1
+    repeat = 5
+    params = [2, 4, 8]
+    param_names = ["threads"]
+
+    def setup(self, threads):
+        strings = Index([f"i-{i}" for i in range(100000)], dtype=object)
+
+        @test_parallel(num_threads=threads)
+        def parallel():
+            factorize(strings)
+
+        self.parallel = parallel
+
+        def loop():
+            factorize(strings)
+
+        self.loop = loop
+
+    def time_parallel(self, threads):
+        self.parallel()
+
+    def time_loop(self, threads):
+        for i in range(threads):
+            self.loop()
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/groupby.py b/pandas/asv_bench/benchmarks/groupby.py
new file mode 100644
index 0000000000000..19c556dfe9d1f
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/groupby.py
@@ -0,0 +1,1151 @@
+from functools import partial
+from itertools import product
+from string import ascii_letters
+
+import numpy as np
+
+from pandas import (
+    NA,
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    to_timedelta,
+)
+
+method_blocklist = {
+    "object": {
+        "diff",
+        "median",
+        "prod",
+        "sem",
+        "cumsum",
+        "sum",
+        "cummin",
+        "mean",
+        "max",
+        "skew",
+        "cumprod",
+        "cummax",
+        "pct_change",
+        "min",
+        "var",
+        "describe",
+        "std",
+        "quantile",
+    },
+    "datetime": {
+        "median",
+        "prod",
+        "sem",
+        "cumsum",
+        "sum",
+        "mean",
+        "skew",
+        "cumprod",
+        "cummax",
+        "pct_change",
+        "var",
+        "describe",
+        "std",
+    },
+}
+
+# These aggregations don't have a kernel implemented for them yet
+_numba_unsupported_methods = [
+    "all",
+    "any",
+    "bfill",
+    "count",
+    "cumcount",
+    "cummax",
+    "cummin",
+    "cumprod",
+    "cumsum",
+    "describe",
+    "diff",
+    "ffill",
+    "first",
+    "head",
+    "idxmax",
+    "idxmin",
+    "last",
+    "median",
+    "nunique",
+    "pct_change",
+    "prod",
+    "quantile",
+    "rank",
+    "sem",
+    "shift",
+    "size",
+    "skew",
+    "tail",
+    "unique",
+    "value_counts",
+]
+
+
+class ApplyDictReturn:
+    def setup(self):
+        self.labels = np.arange(1000).repeat(10)
+        self.data = Series(np.random.randn(len(self.labels)))
+
+    def time_groupby_apply_dict_return(self):
+        self.data.groupby(self.labels).apply(
+            lambda x: {"first": x.values[0], "last": x.values[-1]}
+        )
+
+
+class Apply:
+    param_names = ["factor"]
+    params = [4, 5]
+
+    def setup(self, factor):
+        N = 10**factor
+        # two cases:
+        # - small groups: small data (N**4) + many labels (2000) -> average group
+        #   size of 5 (-> larger overhead of slicing method)
+        # - larger groups: larger data (N**5) + fewer labels (20) -> average group
+        #   size of 5000
+        labels = np.random.randint(0, 2000 if factor == 4 else 20, size=N)
+        labels2 = np.random.randint(0, 3, size=N)
+        df = DataFrame(
+            {
+                "key": labels,
+                "key2": labels2,
+                "value1": np.random.randn(N),
+                "value2": ["foo", "bar", "baz", "qux"] * (N // 4),
+            }
+        )
+        self.df = df
+
+    def time_scalar_function_multi_col(self, factor):
+        self.df.groupby(["key", "key2"]).apply(lambda x: 1)
+
+    def time_scalar_function_single_col(self, factor):
+        self.df.groupby("key").apply(lambda x: 1)
+
+    @staticmethod
+    def df_copy_function(g):
+        # ensure that the group name is available (see GH #15062)
+        g.name
+        return g.copy()
+
+    def time_copy_function_multi_col(self, factor):
+        self.df.groupby(["key", "key2"]).apply(self.df_copy_function)
+
+    def time_copy_overhead_single_col(self, factor):
+        self.df.groupby("key").apply(self.df_copy_function)
+
+
+class ApplyNonUniqueUnsortedIndex:
+    def setup(self):
+        # GH 46527
+        # unsorted and non-unique index
+        idx = np.arange(100)[::-1]
+        idx = Index(np.repeat(idx, 200), name="key")
+        self.df = DataFrame(np.random.randn(len(idx), 10), index=idx)
+
+    def time_groupby_apply_non_unique_unsorted_index(self):
+        self.df.groupby("key", group_keys=False).apply(lambda x: x)
+
+
+class Groups:
+    param_names = ["key"]
+    params = ["int64_small", "int64_large", "object_small", "object_large"]
+
+    def setup_cache(self):
+        size = 10**6
+        data = {
+            "int64_small": Series(np.random.randint(0, 100, size=size)),
+            "int64_large": Series(np.random.randint(0, 10000, size=size)),
+            "object_small": Series(
+                Index([f"i-{i}" for i in range(100)], dtype=object).take(
+                    np.random.randint(0, 100, size=size)
+                )
+            ),
+            "object_large": Series(
+                Index([f"i-{i}" for i in range(10000)], dtype=object).take(
+                    np.random.randint(0, 10000, size=size)
+                )
+            ),
+        }
+        return data
+
+    def setup(self, data, key):
+        self.ser = data[key]
+
+    def time_series_groups(self, data, key):
+        self.ser.groupby(self.ser).groups
+
+    def time_series_indices(self, data, key):
+        self.ser.groupby(self.ser).indices
+
+
+class GroupManyLabels:
+    params = [1, 1000]
+    param_names = ["ncols"]
+
+    def setup(self, ncols):
+        N = 1000
+        data = np.random.randn(N, ncols)
+        self.labels = np.random.randint(0, 100, size=N)
+        self.df = DataFrame(data)
+
+    def time_sum(self, ncols):
+        self.df.groupby(self.labels).sum()
+
+
+class Nth:
+    param_names = ["dtype"]
+    params = ["float32", "float64", "datetime", "object"]
+
+    def setup(self, dtype):
+        N = 10**5
+        # with datetimes (GH7555)
+        if dtype == "datetime":
+            values = date_range("1/1/2011", periods=N, freq="s")
+        elif dtype == "object":
+            values = ["foo"] * N
+        else:
+            values = np.arange(N).astype(dtype)
+
+        key = np.arange(N)
+        self.df = DataFrame({"key": key, "values": values})
+        self.df.iloc[1, 1] = np.nan  # insert missing data
+
+    def time_frame_nth_any(self, dtype):
+        self.df.groupby("key").nth(0, dropna="any")
+
+    def time_groupby_nth_all(self, dtype):
+        self.df.groupby("key").nth(0, dropna="all")
+
+    def time_frame_nth(self, dtype):
+        self.df.groupby("key").nth(0)
+
+    def time_series_nth_any(self, dtype):
+        self.df["values"].groupby(self.df["key"]).nth(0, dropna="any")
+
+    def time_series_nth_all(self, dtype):
+        self.df["values"].groupby(self.df["key"]).nth(0, dropna="all")
+
+    def time_series_nth(self, dtype):
+        self.df["values"].groupby(self.df["key"]).nth(0)
+
+
+class DateAttributes:
+    def setup(self):
+        rng = date_range("1/1/2000", "12/31/2005", freq="h")
+        self.year, self.month, self.day = rng.year, rng.month, rng.day
+        self.ts = Series(np.random.randn(len(rng)), index=rng)
+
+    def time_len_groupby_object(self):
+        len(self.ts.groupby([self.year, self.month, self.day]))
+
+
+class Int64:
+    def setup(self):
+        arr = np.random.randint(-1 << 12, 1 << 12, (1 << 17, 5))
+        i = np.random.choice(len(arr), len(arr) * 5)
+        arr = np.vstack((arr, arr[i]))
+        i = np.random.permutation(len(arr))
+        arr = arr[i]
+        self.cols = list("abcde")
+        self.df = DataFrame(arr, columns=self.cols)
+        self.df["jim"], self.df["joe"] = np.random.randn(2, len(self.df)) * 10
+
+    def time_overflow(self):
+        self.df.groupby(self.cols).max()
+
+
+class CountMultiDtype:
+    def setup_cache(self):
+        n = 10000
+        offsets = np.random.randint(n, size=n).astype("timedelta64[ns]")
+        dates = np.datetime64("now") + offsets
+        dates[np.random.rand(n) > 0.5] = np.datetime64("nat")
+        offsets[np.random.rand(n) > 0.5] = np.timedelta64("nat")
+        value2 = np.random.randn(n)
+        value2[np.random.rand(n) > 0.5] = np.nan
+        obj = np.random.choice(list("ab"), size=n).astype(object)
+        obj[np.random.randn(n) > 0.5] = np.nan
+        df = DataFrame(
+            {
+                "key1": np.random.randint(0, 500, size=n),
+                "key2": np.random.randint(0, 100, size=n),
+                "dates": dates,
+                "value2": value2,
+                "value3": np.random.randn(n),
+                "ints": np.random.randint(0, 1000, size=n),
+                "obj": obj,
+                "offsets": offsets,
+            }
+        )
+        return df
+
+    def time_multi_count(self, df):
+        df.groupby(["key1", "key2"]).count()
+
+
+class CountMultiInt:
+    def setup_cache(self):
+        n = 10000
+        df = DataFrame(
+            {
+                "key1": np.random.randint(0, 500, size=n),
+                "key2": np.random.randint(0, 100, size=n),
+                "ints": np.random.randint(0, 1000, size=n),
+                "ints2": np.random.randint(0, 1000, size=n),
+            }
+        )
+        return df
+
+    def time_multi_int_count(self, df):
+        df.groupby(["key1", "key2"]).count()
+
+    def time_multi_int_nunique(self, df):
+        df.groupby(["key1", "key2"]).nunique()
+
+
+class AggFunctions:
+    def setup_cache(self):
+        N = 10**5
+        fac1 = np.array(["A", "B", "C"], dtype="O")
+        fac2 = np.array(["one", "two"], dtype="O")
+        df = DataFrame(
+            {
+                "key1": fac1.take(np.random.randint(0, 3, size=N)),
+                "key2": fac2.take(np.random.randint(0, 2, size=N)),
+                "value1": np.random.randn(N),
+                "value2": np.random.randn(N),
+                "value3": np.random.randn(N),
+            }
+        )
+        return df
+
+    def time_different_str_functions(self, df):
+        df.groupby(["key1", "key2"]).agg(
+            {"value1": "mean", "value2": "var", "value3": "sum"}
+        )
+
+    def time_different_str_functions_multicol(self, df):
+        df.groupby(["key1", "key2"]).agg(["sum", "min", "max"])
+
+    def time_different_str_functions_singlecol(self, df):
+        df.groupby("key1").agg({"value1": "mean", "value2": "var", "value3": "sum"})
+
+
+class GroupStrings:
+    def setup(self):
+        n = 2 * 10**5
+        alpha = list(map("".join, product(ascii_letters, repeat=4)))
+        data = np.random.choice(alpha, (n // 5, 4), replace=False)
+        data = np.repeat(data, 5, axis=0)
+        self.df = DataFrame(data, columns=list("abcd"))
+        self.df["joe"] = (np.random.randn(len(self.df)) * 10).round(3)
+        self.df = self.df.sample(frac=1).reset_index(drop=True)
+
+    def time_multi_columns(self):
+        self.df.groupby(list("abcd")).max()
+
+
+class MultiColumn:
+    def setup_cache(self):
+        N = 10**5
+        key1 = np.tile(np.arange(100, dtype=object), 1000)
+        key2 = key1.copy()
+        np.random.shuffle(key1)
+        np.random.shuffle(key2)
+        df = DataFrame(
+            {
+                "key1": key1,
+                "key2": key2,
+                "data1": np.random.randn(N),
+                "data2": np.random.randn(N),
+            }
+        )
+        return df
+
+    def time_lambda_sum(self, df):
+        df.groupby(["key1", "key2"]).agg(lambda x: x.values.sum())
+
+    def time_cython_sum(self, df):
+        df.groupby(["key1", "key2"]).sum()
+
+    def time_col_select_lambda_sum(self, df):
+        df.groupby(["key1", "key2"])["data1"].agg(lambda x: x.values.sum())
+
+    def time_col_select_str_sum(self, df):
+        df.groupby(["key1", "key2"])["data1"].agg("sum")
+
+
+class Size:
+    def setup(self):
+        n = 10**5
+        offsets = np.random.randint(n, size=n).astype("timedelta64[ns]")
+        dates = np.datetime64("now") + offsets
+        self.df = DataFrame(
+            {
+                "key1": np.random.randint(0, 500, size=n),
+                "key2": np.random.randint(0, 100, size=n),
+                "value1": np.random.randn(n),
+                "value2": np.random.randn(n),
+                "value3": np.random.randn(n),
+                "dates": dates,
+            }
+        )
+        self.draws = Series(np.random.randn(n))
+        labels = Series(["foo", "bar", "baz", "qux"] * (n // 4))
+        self.cats = labels.astype("category")
+
+    def time_multi_size(self):
+        self.df.groupby(["key1", "key2"]).size()
+
+    def time_category_size(self):
+        self.draws.groupby(self.cats, observed=True).size()
+
+
+class Shift:
+    def setup(self):
+        N = 18
+        self.df = DataFrame({"g": ["a", "b"] * 9, "v": list(range(N))})
+
+    def time_defaults(self):
+        self.df.groupby("g").shift()
+
+    def time_fill_value(self):
+        self.df.groupby("g").shift(fill_value=99)
+
+
+class Fillna:
+    def setup(self):
+        N = 100
+        self.df = DataFrame(
+            {"group": [1] * N + [2] * N, "value": [np.nan, 1.0] * N}
+        ).set_index("group")
+
+    def time_df_ffill(self):
+        self.df.groupby("group").ffill()
+
+    def time_df_bfill(self):
+        self.df.groupby("group").bfill()
+
+    def time_srs_ffill(self):
+        self.df.groupby("group")["value"].ffill()
+
+    def time_srs_bfill(self):
+        self.df.groupby("group")["value"].bfill()
+
+
+class GroupByMethods:
+    param_names = ["dtype", "method", "application", "ncols"]
+    params = [
+        ["int", "int16", "float", "object", "datetime", "uint"],
+        [
+            "all",
+            "any",
+            "bfill",
+            "count",
+            "cumcount",
+            "cummax",
+            "cummin",
+            "cumprod",
+            "cumsum",
+            "describe",
+            "diff",
+            "ffill",
+            "first",
+            "head",
+            "last",
+            "max",
+            "min",
+            "median",
+            "mean",
+            "nunique",
+            "pct_change",
+            "prod",
+            "quantile",
+            "rank",
+            "sem",
+            "shift",
+            "size",
+            "skew",
+            "std",
+            "sum",
+            "tail",
+            "unique",
+            "value_counts",
+            "var",
+        ],
+        ["direct", "transformation"],
+        [1, 5],
+        ["cython", "numba"],
+    ]
+
+    def setup(self, dtype, method, application, ncols, engine):
+        if method in method_blocklist.get(dtype, {}):
+            raise NotImplementedError  # skip benchmark
+
+        if ncols != 1 and method in ["value_counts", "unique"]:
+            # DataFrameGroupBy doesn't have these methods
+            raise NotImplementedError
+
+        if application == "transformation" and method in [
+            "describe",
+            "head",
+            "tail",
+            "unique",
+            "value_counts",
+            "size",
+        ]:
+            # DataFrameGroupBy doesn't have these methods
+            raise NotImplementedError
+
+        # Numba currently doesn't support
+        # multiple transform functions or strs for transform,
+        # grouping on multiple columns
+        # and we lack kernels for a bunch of methods
+        if (
+            (engine == "numba" and method in _numba_unsupported_methods)
+            or ncols > 1
+            or application == "transformation"
+            or dtype == "datetime"
+        ):
+            raise NotImplementedError
+
+        if method == "describe":
+            ngroups = 20
+        elif method == "skew":
+            ngroups = 100
+        else:
+            ngroups = 1000
+        size = ngroups * 2
+        rng = np.arange(ngroups).reshape(-1, 1)
+        rng = np.broadcast_to(rng, (len(rng), ncols))
+        taker = np.random.randint(0, ngroups, size=size)
+        values = rng.take(taker, axis=0)
+        if dtype == "int":
+            key = np.random.randint(0, size, size=size)
+        elif dtype in ("int16", "uint"):
+            key = np.random.randint(0, size, size=size, dtype=dtype)
+        elif dtype == "float":
+            key = np.concatenate(
+                [np.random.random(ngroups) * 0.1, np.random.random(ngroups) * 10.0]
+            )
+        elif dtype == "object":
+            key = ["foo"] * size
+        elif dtype == "datetime":
+            key = date_range("1/1/2011", periods=size, freq="s")
+
+        cols = [f"values{n}" for n in range(ncols)]
+        df = DataFrame(values, columns=cols)
+        df["key"] = key
+
+        if len(cols) == 1:
+            cols = cols[0]
+
+        # Not everything supports the engine keyword yet
+        kwargs = {}
+        if engine == "numba":
+            kwargs["engine"] = engine
+
+        if application == "transformation":
+            self.as_group_method = lambda: df.groupby("key")[cols].transform(
+                method, **kwargs
+            )
+            self.as_field_method = lambda: df.groupby(cols)["key"].transform(
+                method, **kwargs
+            )
+        else:
+            self.as_group_method = partial(
+                getattr(df.groupby("key")[cols], method), **kwargs
+            )
+            self.as_field_method = partial(
+                getattr(df.groupby(cols)["key"], method), **kwargs
+            )
+
+    def time_dtype_as_group(self, dtype, method, application, ncols, engine):
+        self.as_group_method()
+
+    def time_dtype_as_field(self, dtype, method, application, ncols, engine):
+        self.as_field_method()
+
+
+class GroupByCythonAgg:
+    """
+    Benchmarks specifically targeting our cython aggregation algorithms
+    (using a big enough dataframe with simple key, so a large part of the
+    time is actually spent in the grouped aggregation).
+    """
+
+    param_names = ["dtype", "method"]
+    params = [
+        ["float64"],
+        [
+            "sum",
+            "prod",
+            "min",
+            "max",
+            "idxmin",
+            "idxmax",
+            "mean",
+            "median",
+            "var",
+            "first",
+            "last",
+            "any",
+            "all",
+        ],
+    ]
+
+    def setup(self, dtype, method):
+        N = 1_000_000
+        df = DataFrame(np.random.randn(N, 10), columns=list("abcdefghij"))
+        df["key"] = np.random.randint(0, 100, size=N)
+        self.df = df
+
+    def time_frame_agg(self, dtype, method):
+        self.df.groupby("key").agg(method)
+
+
+class GroupByNumbaAgg(GroupByCythonAgg):
+    """
+    Benchmarks specifically targeting our numba aggregation algorithms
+    (using a big enough dataframe with simple key, so a large part of the
+    time is actually spent in the grouped aggregation).
+    """
+
+    def setup(self, dtype, method):
+        if method in _numba_unsupported_methods:
+            raise NotImplementedError
+        super().setup(dtype, method)
+
+    def time_frame_agg(self, dtype, method):
+        self.df.groupby("key").agg(method, engine="numba")
+
+
+class GroupByCythonAggEaDtypes:
+    """
+    Benchmarks specifically targeting our cython aggregation algorithms
+    (using a big enough dataframe with simple key, so a large part of the
+    time is actually spent in the grouped aggregation).
+    """
+
+    param_names = ["dtype", "method"]
+    params = [
+        ["Float64", "Int64", "Int32"],
+        [
+            "sum",
+            "prod",
+            "min",
+            "max",
+            "mean",
+            "median",
+            "var",
+            "first",
+            "last",
+            "any",
+            "all",
+        ],
+    ]
+
+    def setup(self, dtype, method):
+        N = 1_000_000
+        df = DataFrame(
+            np.random.randint(0, high=100, size=(N, 10)),
+            columns=list("abcdefghij"),
+            dtype=dtype,
+        )
+        df.loc[list(range(1, N, 5)), list("abcdefghij")] = NA
+        df["key"] = np.random.randint(0, 100, size=N)
+        self.df = df
+
+    def time_frame_agg(self, dtype, method):
+        self.df.groupby("key").agg(method)
+
+
+class Cumulative:
+    param_names = ["dtype", "method", "with_nans"]
+    params = [
+        ["float64", "int64", "Float64", "Int64"],
+        ["cummin", "cummax", "cumsum"],
+        [True, False],
+    ]
+
+    def setup(self, dtype, method, with_nans):
+        if with_nans and dtype == "int64":
+            raise NotImplementedError("Construction of df would raise")
+
+        N = 500_000
+        keys = np.random.randint(0, 100, size=N)
+        vals = np.random.randint(-10, 10, (N, 5))
+
+        if with_nans:
+            null_vals = vals.astype(float, copy=True)
+            null_vals[::2, :] = np.nan
+            null_vals[::3, :] = np.nan
+            df = DataFrame(null_vals, columns=list("abcde"), dtype=dtype)
+            df["key"] = keys
+            self.df = df
+        else:
+            df = DataFrame(vals, columns=list("abcde")).astype(dtype, copy=False)
+            df["key"] = keys
+            self.df = df
+
+    def time_frame_transform(self, dtype, method, with_nans):
+        self.df.groupby("key").transform(method)
+
+
+class RankWithTies:
+    # GH 21237
+    param_names = ["dtype", "tie_method"]
+    params = [
+        ["float64", "float32", "int64", "datetime64"],
+        ["first", "average", "dense", "min", "max"],
+    ]
+
+    def setup(self, dtype, tie_method):
+        N = 10**4
+        if dtype == "datetime64":
+            data = np.array([Timestamp("2011/01/01")] * N, dtype=dtype)
+        else:
+            data = np.ones(N, dtype=dtype)
+        self.df = DataFrame({"values": data, "key": ["foo"] * N})
+
+    def time_rank_ties(self, dtype, tie_method):
+        self.df.groupby("key").rank(method=tie_method)
+
+
+class Float32:
+    # GH 13335
+    def setup(self):
+        tmp1 = (np.random.random(10000) * 0.1).astype(np.float32)
+        tmp2 = (np.random.random(10000) * 10.0).astype(np.float32)
+        tmp = np.concatenate((tmp1, tmp2))
+        arr = np.repeat(tmp, 10)
+        self.df = DataFrame({"a": arr, "b": arr})
+
+    def time_sum(self):
+        self.df.groupby(["a"])["b"].sum()
+
+
+class String:
+    # GH#41596
+    param_names = ["dtype", "method"]
+    params = [
+        ["str", "string[python]"],
+        [
+            "sum",
+            "min",
+            "max",
+            "first",
+            "last",
+            "any",
+            "all",
+        ],
+    ]
+
+    def setup(self, dtype, method):
+        cols = list("abcdefghjkl")
+        self.df = DataFrame(
+            np.random.randint(0, 100, size=(10_000, len(cols))),
+            columns=cols,
+            dtype=dtype,
+        )
+
+    def time_str_func(self, dtype, method):
+        self.df.groupby("a")[self.df.columns[1:]].agg(method)
+
+
+class Categories:
+    params = [True, False]
+    param_names = ["observed"]
+
+    def setup(self, observed):
+        N = 10**5
+        arr = np.random.random(N)
+        data = {"a": Categorical(np.random.randint(10000, size=N)), "b": arr}
+        self.df = DataFrame(data)
+        data = {
+            "a": Categorical(np.random.randint(10000, size=N), ordered=True),
+            "b": arr,
+        }
+        self.df_ordered = DataFrame(data)
+        data = {
+            "a": Categorical(
+                np.random.randint(100, size=N), categories=np.arange(10000)
+            ),
+            "b": arr,
+        }
+        self.df_extra_cat = DataFrame(data)
+
+    def time_groupby_sort(self, observed):
+        self.df.groupby("a", observed=observed)["b"].count()
+
+    def time_groupby_nosort(self, observed):
+        self.df.groupby("a", observed=observed, sort=False)["b"].count()
+
+    def time_groupby_ordered_sort(self, observed):
+        self.df_ordered.groupby("a", observed=observed)["b"].count()
+
+    def time_groupby_ordered_nosort(self, observed):
+        self.df_ordered.groupby("a", observed=observed, sort=False)["b"].count()
+
+    def time_groupby_extra_cat_sort(self, observed):
+        self.df_extra_cat.groupby("a", observed=observed)["b"].count()
+
+    def time_groupby_extra_cat_nosort(self, observed):
+        self.df_extra_cat.groupby("a", observed=observed, sort=False)["b"].count()
+
+
+class MultipleCategories:
+    def setup(self):
+        N = 10**3
+        arr = np.random.random(N)
+        data = {
+            "a1": Categorical(np.random.randint(10000, size=N)),
+            "a2": Categorical(np.random.randint(10000, size=N)),
+            "b": arr,
+        }
+        self.df = DataFrame(data)
+        data = {
+            "a1": Categorical(np.random.randint(10000, size=N), ordered=True),
+            "a2": Categorical(np.random.randint(10000, size=N), ordered=True),
+            "b": arr,
+        }
+        self.df_ordered = DataFrame(data)
+        data = {
+            "a1": Categorical(np.random.randint(100, size=N), categories=np.arange(N)),
+            "a2": Categorical(np.random.randint(100, size=N), categories=np.arange(N)),
+            "b": arr,
+        }
+        self.df_extra_cat = DataFrame(data)
+
+    def time_groupby_sort(self):
+        self.df.groupby(["a1", "a2"], observed=False)["b"].count()
+
+    def time_groupby_nosort(self):
+        self.df.groupby(["a1", "a2"], observed=False, sort=False)["b"].count()
+
+    def time_groupby_ordered_sort(self):
+        self.df_ordered.groupby(["a1", "a2"], observed=False)["b"].count()
+
+    def time_groupby_ordered_nosort(self):
+        self.df_ordered.groupby(["a1", "a2"], observed=False, sort=False)["b"].count()
+
+    def time_groupby_extra_cat_sort(self):
+        self.df_extra_cat.groupby(["a1", "a2"], observed=False)["b"].count()
+
+    def time_groupby_extra_cat_nosort(self):
+        self.df_extra_cat.groupby(["a1", "a2"], observed=False, sort=False)["b"].count()
+
+    def time_groupby_transform(self):
+        self.df_extra_cat.groupby(["a1", "a2"], observed=False)["b"].cumsum()
+
+
+class Datelike:
+    # GH 14338
+    params = ["period_range", "date_range", "date_range_tz"]
+    param_names = ["grouper"]
+
+    def setup(self, grouper):
+        N = 10**4
+        rng_map = {
+            "period_range": period_range,
+            "date_range": date_range,
+            "date_range_tz": partial(date_range, tz="US/Central"),
+        }
+        self.grouper = rng_map[grouper]("1900-01-01", freq="D", periods=N)
+        self.df = DataFrame(np.random.randn(10**4, 2))
+
+    def time_sum(self, grouper):
+        self.df.groupby(self.grouper).sum()
+
+
+class SumBools:
+    # GH 2692
+    def setup(self):
+        N = 500
+        self.df = DataFrame({"ii": range(N), "bb": [True] * N})
+
+    def time_groupby_sum_booleans(self):
+        self.df.groupby("ii").sum()
+
+
+class SumMultiLevel:
+    # GH 9049
+    timeout = 120.0
+
+    def setup(self):
+        N = 50
+        self.df = DataFrame(
+            {"A": list(range(N)) * 2, "B": range(N * 2), "C": 1}
+        ).set_index(["A", "B"])
+
+    def time_groupby_sum_multiindex(self):
+        self.df.groupby(level=[0, 1]).sum()
+
+
+class SumTimeDelta:
+    # GH 20660
+    def setup(self):
+        N = 10**4
+        self.df = DataFrame(
+            np.random.randint(1000, 100000, (N, 100)),
+            index=np.random.randint(200, size=(N,)),
+        ).astype("timedelta64[ns]")
+        self.df_int = self.df.copy().astype("int64")
+
+    def time_groupby_sum_timedelta(self):
+        self.df.groupby(lambda x: x).sum()
+
+    def time_groupby_sum_int(self):
+        self.df_int.groupby(lambda x: x).sum()
+
+
+class Transform:
+    def setup(self):
+        n1 = 400
+        n2 = 250
+        index = MultiIndex(
+            levels=[np.arange(n1), Index([f"i-{i}" for i in range(n2)], dtype=object)],
+            codes=[np.repeat(range(n1), n2).tolist(), list(range(n2)) * n1],
+            names=["lev1", "lev2"],
+        )
+        arr = np.random.randn(n1 * n2, 3)
+        arr[::10000, 0] = np.nan
+        arr[1::10000, 1] = np.nan
+        arr[2::10000, 2] = np.nan
+        data = DataFrame(arr, index=index, columns=["col1", "col20", "col3"])
+        self.df = data
+
+        n = 1000
+        self.df_wide = DataFrame(
+            np.random.randn(n, n),
+            index=np.random.choice(range(10), n),
+        )
+
+        n = 1_000_000
+        self.df_tall = DataFrame(
+            np.random.randn(n, 3),
+            index=np.random.randint(0, 5, n),
+        )
+
+        n = 20000
+        self.df1 = DataFrame(
+            np.random.randint(1, n, (n, 3)), columns=["jim", "joe", "jolie"]
+        )
+        self.df2 = self.df1.copy()
+        self.df2["jim"] = self.df2["joe"]
+
+        self.df3 = DataFrame(
+            np.random.randint(1, (n / 10), (n, 3)), columns=["jim", "joe", "jolie"]
+        )
+        self.df4 = self.df3.copy()
+        self.df4["jim"] = self.df4["joe"]
+
+    def time_transform_lambda_max(self):
+        self.df.groupby(level="lev1").transform(lambda x: max(x))
+
+    def time_transform_str_max(self):
+        self.df.groupby(level="lev1").transform("max")
+
+    def time_transform_lambda_max_tall(self):
+        self.df_tall.groupby(level=0).transform(lambda x: np.max(x, axis=0))
+
+    def time_transform_lambda_max_wide(self):
+        self.df_wide.groupby(level=0).transform(lambda x: np.max(x, axis=0))
+
+    def time_transform_multi_key1(self):
+        self.df1.groupby(["jim", "joe"])["jolie"].transform("max")
+
+    def time_transform_multi_key2(self):
+        self.df2.groupby(["jim", "joe"])["jolie"].transform("max")
+
+    def time_transform_multi_key3(self):
+        self.df3.groupby(["jim", "joe"])["jolie"].transform("max")
+
+    def time_transform_multi_key4(self):
+        self.df4.groupby(["jim", "joe"])["jolie"].transform("max")
+
+
+class TransformBools:
+    def setup(self):
+        N = 120000
+        transition_points = np.sort(np.random.choice(np.arange(N), 1400))
+        transitions = np.zeros(N, dtype=np.bool_)
+        transitions[transition_points] = True
+        self.g = transitions.cumsum()
+        self.df = DataFrame({"signal": np.random.rand(N)})
+
+    def time_transform_mean(self):
+        self.df["signal"].groupby(self.g).transform("mean")
+
+
+class TransformNaN:
+    # GH 12737
+    def setup(self):
+        self.df_nans = DataFrame(
+            {"key": np.repeat(np.arange(1000), 10), "B": np.nan, "C": np.nan}
+        )
+        self.df_nans.loc[4::10, "B":"C"] = 5
+
+    def time_first(self):
+        self.df_nans.groupby("key").transform("first")
+
+
+class TransformEngine:
+    param_names = ["parallel"]
+    params = [[True, False]]
+
+    def setup(self, parallel):
+        N = 10**3
+        data = DataFrame(
+            {0: [str(i) for i in range(100)] * N, 1: list(range(100)) * N},
+            columns=[0, 1],
+        )
+        self.parallel = parallel
+        self.grouper = data.groupby(0)
+
+    def time_series_numba(self, parallel):
+        def function(values, index):
+            return values * 5
+
+        self.grouper[1].transform(
+            function, engine="numba", engine_kwargs={"parallel": self.parallel}
+        )
+
+    def time_series_cython(self, parallel):
+        def function(values):
+            return values * 5
+
+        self.grouper[1].transform(function, engine="cython")
+
+    def time_dataframe_numba(self, parallel):
+        def function(values, index):
+            return values * 5
+
+        self.grouper.transform(
+            function, engine="numba", engine_kwargs={"parallel": self.parallel}
+        )
+
+    def time_dataframe_cython(self, parallel):
+        def function(values):
+            return values * 5
+
+        self.grouper.transform(function, engine="cython")
+
+
+class AggEngine:
+    param_names = ["parallel"]
+    params = [[True, False]]
+
+    def setup(self, parallel):
+        N = 10**3
+        data = DataFrame(
+            {0: [str(i) for i in range(100)] * N, 1: list(range(100)) * N},
+            columns=[0, 1],
+        )
+        self.parallel = parallel
+        self.grouper = data.groupby(0)
+
+    def time_series_numba(self, parallel):
+        def function(values, index):
+            total = 0
+            for i, value in enumerate(values):
+                if i % 2:
+                    total += value + 5
+                else:
+                    total += value * 2
+            return total
+
+        self.grouper[1].agg(
+            function, engine="numba", engine_kwargs={"parallel": self.parallel}
+        )
+
+    def time_series_cython(self, parallel):
+        def function(values):
+            total = 0
+            for i, value in enumerate(values):
+                if i % 2:
+                    total += value + 5
+                else:
+                    total += value * 2
+            return total
+
+        self.grouper[1].agg(function, engine="cython")
+
+    def time_dataframe_numba(self, parallel):
+        def function(values, index):
+            total = 0
+            for i, value in enumerate(values):
+                if i % 2:
+                    total += value + 5
+                else:
+                    total += value * 2
+            return total
+
+        self.grouper.agg(
+            function, engine="numba", engine_kwargs={"parallel": self.parallel}
+        )
+
+    def time_dataframe_cython(self, parallel):
+        def function(values):
+            total = 0
+            for i, value in enumerate(values):
+                if i % 2:
+                    total += value + 5
+                else:
+                    total += value * 2
+            return total
+
+        self.grouper.agg(function, engine="cython")
+
+
+class Sample:
+    def setup(self):
+        N = 10**3
+        self.df = DataFrame({"a": np.zeros(N)})
+        self.groups = np.arange(0, N)
+        self.weights = np.ones(N)
+
+    def time_sample(self):
+        self.df.groupby(self.groups).sample(n=1)
+
+    def time_sample_weights(self):
+        self.df.groupby(self.groups).sample(n=1, weights=self.weights)
+
+
+class Resample:
+    # GH 28635
+    def setup(self):
+        num_timedeltas = 20_000
+        num_groups = 3
+
+        index = MultiIndex.from_product(
+            [
+                np.arange(num_groups),
+                to_timedelta(np.arange(num_timedeltas), unit="s"),
+            ],
+            names=["groups", "timedeltas"],
+        )
+        data = np.random.randint(0, 1000, size=(len(index)))
+
+        self.df = DataFrame(data, index=index).reset_index("timedeltas")
+        self.df_multiindex = DataFrame(data, index=index)
+
+    def time_resample(self):
+        self.df.groupby(level="groups").resample("10s", on="timedeltas").mean()
+
+    def time_resample_multiindex(self):
+        self.df_multiindex.groupby(level="groups").resample(
+            "10s", level="timedeltas"
+        ).mean()
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/hash_functions.py b/pandas/asv_bench/benchmarks/hash_functions.py
new file mode 100644
index 0000000000000..d2c5b4dfbef70
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/hash_functions.py
@@ -0,0 +1,89 @@
+import numpy as np
+
+import pandas as pd
+
+
+class UniqueForLargePyObjectInts:
+    def setup(self):
+        lst = [x << 32 for x in range(5000)]
+        self.arr = np.array(lst, dtype=np.object_)
+
+    def time_unique(self):
+        pd.unique(self.arr)
+
+
+class Float64GroupIndex:
+    # GH28303
+    def setup(self):
+        self.df = pd.date_range(
+            start="1/1/2018", end="1/2/2018", periods=10**6
+        ).to_frame()
+        self.group_index = np.round(self.df.index.astype(int) / 10**9)
+
+    def time_groupby(self):
+        self.df.groupby(self.group_index).last()
+
+
+class UniqueAndFactorizeArange:
+    params = range(4, 16)
+    param_names = ["exponent"]
+
+    def setup(self, exponent):
+        a = np.arange(10**4, dtype="float64")
+        self.a2 = (a + 10**exponent).repeat(100)
+
+    def time_factorize(self, exponent):
+        pd.factorize(self.a2)
+
+    def time_unique(self, exponent):
+        pd.unique(self.a2)
+
+
+class Unique:
+    params = ["Int64", "Float64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        self.ser = pd.Series(([1, pd.NA, 2] + list(range(100_000))) * 3, dtype=dtype)
+        self.ser_unique = pd.Series(list(range(300_000)) + [pd.NA], dtype=dtype)
+
+    def time_unique_with_duplicates(self, exponent):
+        pd.unique(self.ser)
+
+    def time_unique(self, exponent):
+        pd.unique(self.ser_unique)
+
+
+class NumericSeriesIndexing:
+    params = [
+        (np.int64, np.uint64, np.float64),
+        (10**4, 10**5, 5 * 10**5, 10**6, 5 * 10**6),
+    ]
+    param_names = ["dtype", "N"]
+
+    def setup(self, dtype, N):
+        vals = np.array(list(range(55)) + [54] + list(range(55, N - 1)), dtype=dtype)
+        indices = pd.Index(vals)
+        self.data = pd.Series(np.arange(N), index=indices)
+
+    def time_loc_slice(self, index, N):
+        # trigger building of mapping
+        self.data.loc[:800]
+
+
+class NumericSeriesIndexingShuffled:
+    params = [
+        (np.int64, np.uint64, np.float64),
+        (10**4, 10**5, 5 * 10**5, 10**6, 5 * 10**6),
+    ]
+    param_names = ["dtype", "N"]
+
+    def setup(self, dtype, N):
+        vals = np.array(list(range(55)) + [54] + list(range(55, N - 1)), dtype=dtype)
+        np.random.shuffle(vals)
+        indices = pd.Index(vals)
+        self.data = pd.Series(np.arange(N), index=indices)
+
+    def time_loc_slice(self, index, N):
+        # trigger building of mapping
+        self.data.loc[:800]
diff --git a/pandas/asv_bench/benchmarks/index_cached_properties.py b/pandas/asv_bench/benchmarks/index_cached_properties.py
new file mode 100644
index 0000000000000..d21bbe15c4cc8
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/index_cached_properties.py
@@ -0,0 +1,72 @@
+import pandas as pd
+
+
+class IndexCache:
+    number = 1
+    repeat = (3, 100, 20)
+
+    params = [
+        [
+            "CategoricalIndex",
+            "DatetimeIndex",
+            "Float64Index",
+            "IntervalIndex",
+            "Int64Index",
+            "MultiIndex",
+            "PeriodIndex",
+            "RangeIndex",
+            "TimedeltaIndex",
+            "UInt64Index",
+        ]
+    ]
+    param_names = ["index_type"]
+
+    def setup(self, index_type):
+        N = 10**5
+        if index_type == "MultiIndex":
+            self.idx = pd.MultiIndex.from_product(
+                [pd.date_range("1/1/2000", freq="min", periods=N // 2), ["a", "b"]]
+            )
+        elif index_type == "DatetimeIndex":
+            self.idx = pd.date_range("1/1/2000", freq="min", periods=N)
+        elif index_type == "Int64Index":
+            self.idx = pd.Index(range(N), dtype="int64")
+        elif index_type == "PeriodIndex":
+            self.idx = pd.period_range("1/1/2000", freq="min", periods=N)
+        elif index_type == "RangeIndex":
+            self.idx = pd.RangeIndex(start=0, stop=N)
+        elif index_type == "IntervalIndex":
+            self.idx = pd.IntervalIndex.from_arrays(range(N), range(1, N + 1))
+        elif index_type == "TimedeltaIndex":
+            self.idx = pd.TimedeltaIndex(range(N))
+        elif index_type == "Float64Index":
+            self.idx = pd.Index(range(N), dtype="float64")
+        elif index_type == "UInt64Index":
+            self.idx = pd.Index(range(N), dtype="uint64")
+        elif index_type == "CategoricalIndex":
+            self.idx = pd.CategoricalIndex(range(N), range(N))
+        else:
+            raise ValueError
+        assert len(self.idx) == N
+        self.idx._cache = {}
+
+    def time_values(self, index_type):
+        self.idx._values
+
+    def time_shape(self, index_type):
+        self.idx.shape
+
+    def time_is_monotonic_decreasing(self, index_type):
+        self.idx.is_monotonic_decreasing
+
+    def time_is_monotonic_increasing(self, index_type):
+        self.idx.is_monotonic_increasing
+
+    def time_is_unique(self, index_type):
+        self.idx.is_unique
+
+    def time_engine(self, index_type):
+        self.idx._engine
+
+    def time_inferred_type(self, index_type):
+        self.idx.inferred_type
diff --git a/pandas/asv_bench/benchmarks/index_object.py b/pandas/asv_bench/benchmarks/index_object.py
new file mode 100644
index 0000000000000..9c1e9656503f7
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/index_object.py
@@ -0,0 +1,262 @@
+import gc
+
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    date_range,
+)
+
+
+class SetOperations:
+    params = (
+        ["monotonic", "non_monotonic"],
+        ["datetime", "date_string", "int", "strings", "ea_int"],
+        ["intersection", "union", "symmetric_difference"],
+    )
+    param_names = ["index_structure", "dtype", "method"]
+
+    def setup(self, index_structure, dtype, method):
+        N = 10**5
+        dates_left = date_range("1/1/2000", periods=N, freq="min")
+        fmt = "%Y-%m-%d %H:%M:%S"
+        date_str_left = Index(dates_left.strftime(fmt))
+        int_left = Index(np.arange(N))
+        ea_int_left = Index(np.arange(N), dtype="Int64")
+        str_left = Index([f"i-{i}" for i in range(N)], dtype=object)
+
+        data = {
+            "datetime": dates_left,
+            "date_string": date_str_left,
+            "int": int_left,
+            "strings": str_left,
+            "ea_int": ea_int_left,
+        }
+
+        if index_structure == "non_monotonic":
+            data = {k: mi[::-1] for k, mi in data.items()}
+
+        data = {k: {"left": idx, "right": idx[:-1]} for k, idx in data.items()}
+
+        self.left = data[dtype]["left"]
+        self.right = data[dtype]["right"]
+
+    def time_operation(self, index_structure, dtype, method):
+        getattr(self.left, method)(self.right)
+
+
+class SetDisjoint:
+    def setup(self):
+        N = 10**5
+        B = N + 20000
+        self.datetime_left = DatetimeIndex(range(N))
+        self.datetime_right = DatetimeIndex(range(N, B))
+
+    def time_datetime_difference_disjoint(self):
+        self.datetime_left.difference(self.datetime_right)
+
+
+class UnionWithDuplicates:
+    def setup(self):
+        self.left = Index(np.repeat(np.arange(1000), 100))
+        self.right = Index(np.tile(np.arange(500, 1500), 50))
+
+    def time_union_with_duplicates(self):
+        self.left.union(self.right)
+
+
+class Range:
+    def setup(self):
+        self.idx_inc = RangeIndex(start=0, stop=10**6, step=3)
+        self.idx_dec = RangeIndex(start=10**6, stop=-1, step=-3)
+
+    def time_max(self):
+        self.idx_inc.max()
+
+    def time_max_trivial(self):
+        self.idx_dec.max()
+
+    def time_min(self):
+        self.idx_dec.min()
+
+    def time_min_trivial(self):
+        self.idx_inc.min()
+
+    def time_get_loc_inc(self):
+        self.idx_inc.get_loc(900_000)
+
+    def time_get_loc_dec(self):
+        self.idx_dec.get_loc(100_000)
+
+    def time_iter_inc(self):
+        for _ in self.idx_inc:
+            pass
+
+    def time_iter_dec(self):
+        for _ in self.idx_dec:
+            pass
+
+    def time_sort_values_asc(self):
+        self.idx_inc.sort_values()
+
+    def time_sort_values_des(self):
+        self.idx_inc.sort_values(ascending=False)
+
+
+class IndexEquals:
+    def setup(self):
+        idx_large_fast = RangeIndex(100_000)
+        idx_small_slow = date_range(start="1/1/2012", periods=1)
+        self.mi_large_slow = MultiIndex.from_product([idx_large_fast, idx_small_slow])
+
+        self.idx_non_object = RangeIndex(1)
+
+    def time_non_object_equals_multiindex(self):
+        self.idx_non_object.equals(self.mi_large_slow)
+
+
+class IndexAppend:
+    def setup(self):
+        N = 10_000
+        self.range_idx = RangeIndex(0, 100)
+        self.int_idx = self.range_idx.astype(int)
+        self.obj_idx = self.int_idx.astype(str)
+        self.range_idxs = []
+        self.int_idxs = []
+        self.object_idxs = []
+        for i in range(1, N):
+            r_idx = RangeIndex(i * 100, (i + 1) * 100)
+            self.range_idxs.append(r_idx)
+            i_idx = r_idx.astype(int)
+            self.int_idxs.append(i_idx)
+            o_idx = i_idx.astype(str)
+            self.object_idxs.append(o_idx)
+        self.same_range_idx = [self.range_idx] * N
+
+    def time_append_range_list(self):
+        self.range_idx.append(self.range_idxs)
+
+    def time_append_range_list_same(self):
+        self.range_idx.append(self.same_range_idx)
+
+    def time_append_int_list(self):
+        self.int_idx.append(self.int_idxs)
+
+    def time_append_obj_list(self):
+        self.obj_idx.append(self.object_idxs)
+
+
+class Indexing:
+    params = ["String", "Float", "Int"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**6
+        if dtype == "String":
+            self.idx = Index([f"i-{i}" for i in range(N)], dtype=object)
+        elif dtype == "Float":
+            self.idx = Index(np.arange(N), dtype=np.float64)
+        elif dtype == "Int":
+            self.idx = Index(np.arange(N), dtype=np.int64)
+        self.array_mask = (np.arange(N) % 3) == 0
+        self.series_mask = Series(self.array_mask)
+        self.sorted = self.idx.sort_values()
+        half = N // 2
+        self.non_unique = self.idx[:half].append(self.idx[:half])
+        self.non_unique_sorted = self.sorted[:half].repeat(2)
+        self.key = self.sorted[N // 4]
+
+    def time_boolean_array(self, dtype):
+        self.idx[self.array_mask]
+
+    def time_boolean_series(self, dtype):
+        self.idx[self.series_mask]
+
+    def time_get(self, dtype):
+        self.idx[1]
+
+    def time_slice(self, dtype):
+        self.idx[:-1]
+
+    def time_slice_step(self, dtype):
+        self.idx[::2]
+
+    def time_get_loc(self, dtype):
+        self.idx.get_loc(self.key)
+
+    def time_get_loc_sorted(self, dtype):
+        self.sorted.get_loc(self.key)
+
+    def time_get_loc_non_unique(self, dtype):
+        self.non_unique.get_loc(self.key)
+
+    def time_get_loc_non_unique_sorted(self, dtype):
+        self.non_unique_sorted.get_loc(self.key)
+
+
+class Float64IndexMethod:
+    # GH 13166
+    def setup(self):
+        N = 100_000
+        a = np.arange(N, dtype=np.float64)
+        self.ind = Index(a * 4.8000000418824129e-08)
+
+    def time_get_loc(self):
+        self.ind.get_loc(0)
+
+
+class IntervalIndexMethod:
+    # GH 24813
+    params = [10**3, 10**5]
+
+    def setup(self, N):
+        left = np.append(np.arange(N), np.array(0))
+        right = np.append(np.arange(1, N + 1), np.array(1))
+        self.intv = IntervalIndex.from_arrays(left, right)
+        self.intv._engine
+
+        self.intv2 = IntervalIndex.from_arrays(left + 1, right + 1)
+        self.intv2._engine
+
+        self.left = IntervalIndex.from_breaks(np.arange(N))
+        self.right = IntervalIndex.from_breaks(np.arange(N - 3, 2 * N - 3))
+
+    def time_monotonic_inc(self, N):
+        self.intv.is_monotonic_increasing
+
+    def time_is_unique(self, N):
+        self.intv.is_unique
+
+    def time_intersection(self, N):
+        self.left.intersection(self.right)
+
+    def time_intersection_one_duplicate(self, N):
+        self.intv.intersection(self.right)
+
+    def time_intersection_both_duplicate(self, N):
+        self.intv.intersection(self.intv2)
+
+
+class GC:
+    params = [1, 2, 5]
+
+    def create_use_drop(self):
+        idx = Index(list(range(1_000_000)))
+        idx._engine
+
+    def peakmem_gc_instances(self, N):
+        try:
+            gc.disable()
+
+            for _ in range(N):
+                self.create_use_drop()
+        finally:
+            gc.enable()
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/indexing.py b/pandas/asv_bench/benchmarks/indexing.py
new file mode 100644
index 0000000000000..b2495356f134c
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/indexing.py
@@ -0,0 +1,562 @@
+"""
+These benchmarks are for Series and DataFrame indexing methods.  For the
+lower-level methods directly on Index and subclasses, see index_object.py,
+indexing_engine.py, and index_cached.py
+"""
+
+from datetime import datetime
+import warnings
+
+import numpy as np
+
+from pandas import (
+    NA,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    Series,
+    concat,
+    date_range,
+    option_context,
+    period_range,
+)
+
+
+class NumericSeriesIndexing:
+    params = [
+        (np.int64, np.uint64, np.float64),
+        ("unique_monotonic_inc", "nonunique_monotonic_inc"),
+    ]
+    param_names = ["dtype", "index_structure"]
+
+    def setup(self, dtype, index_structure):
+        N = 10**6
+        indices = {
+            "unique_monotonic_inc": Index(range(N), dtype=dtype),
+            "nonunique_monotonic_inc": Index(
+                list(range(55)) + [54] + list(range(55, N - 1)), dtype=dtype
+            ),
+        }
+        self.data = Series(np.random.rand(N), index=indices[index_structure])
+        self.array = np.arange(10000)
+        self.array_list = self.array.tolist()
+
+    def time_getitem_scalar(self, index, index_structure):
+        self.data[800000]
+
+    def time_getitem_slice(self, index, index_structure):
+        self.data[:800000]
+
+    def time_getitem_list_like(self, index, index_structure):
+        self.data[[800000]]
+
+    def time_getitem_array(self, index, index_structure):
+        self.data[self.array]
+
+    def time_getitem_lists(self, index, index_structure):
+        self.data[self.array_list]
+
+    def time_iloc_array(self, index, index_structure):
+        self.data.iloc[self.array]
+
+    def time_iloc_list_like(self, index, index_structure):
+        self.data.iloc[[800000]]
+
+    def time_iloc_scalar(self, index, index_structure):
+        self.data.iloc[800000]
+
+    def time_iloc_slice(self, index, index_structure):
+        self.data.iloc[:800000]
+
+    def time_loc_array(self, index, index_structure):
+        self.data.loc[self.array]
+
+    def time_loc_list_like(self, index, index_structure):
+        self.data.loc[[800000]]
+
+    def time_loc_scalar(self, index, index_structure):
+        self.data.loc[800000]
+
+    def time_loc_slice(self, index, index_structure):
+        self.data.loc[:800000]
+
+
+class NumericMaskedIndexing:
+    monotonic_list = list(range(10**6))
+    non_monotonic_list = list(range(50)) + [54, 53, 52, 51] + list(range(55, 10**6 - 1))
+
+    params = [
+        ("Int64", "UInt64", "Float64"),
+        (True, False),
+    ]
+    param_names = ["dtype", "monotonic"]
+
+    def setup(self, dtype, monotonic):
+        indices = {
+            True: Index(self.monotonic_list, dtype=dtype),
+            False: Index(self.non_monotonic_list, dtype=dtype).append(
+                Index([NA], dtype=dtype)
+            ),
+        }
+        self.data = indices[monotonic]
+        self.indexer = np.arange(300, 1_000)
+        self.data_dups = self.data.append(self.data)
+
+    def time_get_indexer(self, dtype, monotonic):
+        self.data.get_indexer(self.indexer)
+
+    def time_get_indexer_dups(self, dtype, monotonic):
+        self.data.get_indexer_for(self.indexer)
+
+
+class NonNumericSeriesIndexing:
+    params = [
+        ("string", "datetime", "period"),
+        ("unique_monotonic_inc", "nonunique_monotonic_inc", "non_monotonic"),
+    ]
+    param_names = ["index_dtype", "index_structure"]
+
+    def setup(self, index, index_structure):
+        N = 10**6
+        if index == "string":
+            index = Index([f"i-{i}" for i in range(N)], dtype=object)
+        elif index == "datetime":
+            index = date_range("1900", periods=N, freq="s")
+        elif index == "period":
+            index = period_range("1900", periods=N, freq="s")
+        index = index.sort_values()
+        assert index.is_unique and index.is_monotonic_increasing
+        if index_structure == "nonunique_monotonic_inc":
+            index = index.insert(item=index[2], loc=2)[:-1]
+        elif index_structure == "non_monotonic":
+            index = index[::2].append(index[1::2])
+            assert len(index) == N
+        self.s = Series(np.random.rand(N), index=index)
+        self.lbl = index[80000]
+        # warm up index mapping
+        self.s[self.lbl]
+
+    def time_getitem_label_slice(self, index, index_structure):
+        self.s[: self.lbl]
+
+    def time_getitem_pos_slice(self, index, index_structure):
+        self.s[:80000]
+
+    def time_getitem_scalar(self, index, index_structure):
+        self.s[self.lbl]
+
+    def time_getitem_list_like(self, index, index_structure):
+        self.s[[self.lbl]]
+
+
+class DataFrameStringIndexing:
+    def setup(self):
+        index = Index([f"i-{i}" for i in range(1000)], dtype=object)
+        columns = Index([f"i-{i}" for i in range(30)], dtype=object)
+        with warnings.catch_warnings(record=True):
+            self.df = DataFrame(np.random.randn(1000, 30), index=index, columns=columns)
+        self.idx_scalar = index[100]
+        self.col_scalar = columns[10]
+        self.bool_indexer = self.df[self.col_scalar] > 0
+        self.bool_obj_indexer = self.bool_indexer.astype(object)
+        self.boolean_indexer = (self.df[self.col_scalar] > 0).astype("boolean")
+
+    def time_loc(self):
+        self.df.loc[self.idx_scalar, self.col_scalar]
+
+    def time_at(self):
+        self.df.at[self.idx_scalar, self.col_scalar]
+
+    def time_at_setitem(self):
+        self.df.at[self.idx_scalar, self.col_scalar] = 0.0
+
+    def time_getitem_scalar(self):
+        self.df[self.col_scalar][self.idx_scalar]
+
+    def time_boolean_rows(self):
+        self.df[self.bool_indexer]
+
+    def time_boolean_rows_object(self):
+        self.df[self.bool_obj_indexer]
+
+    def time_boolean_rows_boolean(self):
+        self.df[self.boolean_indexer]
+
+
+class DataFrameNumericIndexing:
+    params = [
+        (np.int64, np.uint64, np.float64),
+        ("unique_monotonic_inc", "nonunique_monotonic_inc"),
+    ]
+    param_names = ["dtype", "index_structure"]
+
+    def setup(self, dtype, index_structure):
+        N = 10**5
+        indices = {
+            "unique_monotonic_inc": Index(range(N), dtype=dtype),
+            "nonunique_monotonic_inc": Index(
+                list(range(55)) + [54] + list(range(55, N - 1)), dtype=dtype
+            ),
+        }
+        self.idx_dupe = np.array(range(30)) * 99
+        self.df = DataFrame(np.random.randn(N, 5), index=indices[index_structure])
+        self.df_dup = concat([self.df, 2 * self.df, 3 * self.df])
+        self.bool_indexer = [True] * (N // 2) + [False] * (N - N // 2)
+
+    def time_iloc_dups(self, index, index_structure):
+        self.df_dup.iloc[self.idx_dupe]
+
+    def time_loc_dups(self, index, index_structure):
+        self.df_dup.loc[self.idx_dupe]
+
+    def time_iloc(self, index, index_structure):
+        self.df.iloc[:100, 0]
+
+    def time_loc(self, index, index_structure):
+        self.df.loc[:100, 0]
+
+    def time_bool_indexer(self, index, index_structure):
+        self.df[self.bool_indexer]
+
+
+class Take:
+    params = ["int", "datetime"]
+    param_names = ["index"]
+
+    def setup(self, index):
+        N = 100000
+        indexes = {
+            "int": Index(np.arange(N), dtype=np.int64),
+            "datetime": date_range("2011-01-01", freq="s", periods=N),
+        }
+        index = indexes[index]
+        self.s = Series(np.random.rand(N), index=index)
+        self.indexer = np.random.randint(0, N, size=N)
+
+    def time_take(self, index):
+        self.s.take(self.indexer)
+
+
+class MultiIndexing:
+    params = [True, False]
+    param_names = ["unique_levels"]
+
+    def setup(self, unique_levels):
+        self.nlevels = 2
+        if unique_levels:
+            mi = MultiIndex.from_arrays([range(1000000)] * self.nlevels)
+        else:
+            mi = MultiIndex.from_product([range(1000)] * self.nlevels)
+        self.df = DataFrame(np.random.randn(len(mi)), index=mi)
+
+        self.tgt_slice = slice(200, 800)
+        self.tgt_null_slice = slice(None)
+        self.tgt_list = list(range(0, 1000, 10))
+        self.tgt_scalar = 500
+
+        bool_indexer = np.zeros(len(mi), dtype=np.bool_)
+        bool_indexer[slice(0, len(mi), 100)] = True
+        self.tgt_bool_indexer = bool_indexer
+
+    def time_loc_partial_key_slice(self, unique_levels):
+        self.df.loc[self.tgt_slice, :]
+
+    def time_loc_partial_key_null_slice(self, unique_levels):
+        self.df.loc[self.tgt_null_slice, :]
+
+    def time_loc_partial_key_list(self, unique_levels):
+        self.df.loc[self.tgt_list, :]
+
+    def time_loc_partial_key_scalar(self, unique_levels):
+        self.df.loc[self.tgt_scalar, :]
+
+    def time_loc_partial_key_bool_indexer(self, unique_levels):
+        self.df.loc[self.tgt_bool_indexer, :]
+
+    def time_loc_all_slices(self, unique_levels):
+        target = tuple([self.tgt_slice] * self.nlevels)
+        self.df.loc[target, :]
+
+    def time_loc_all_null_slices(self, unique_levels):
+        target = tuple([self.tgt_null_slice] * self.nlevels)
+        self.df.loc[target, :]
+
+    def time_loc_all_lists(self, unique_levels):
+        target = tuple([self.tgt_list] * self.nlevels)
+        self.df.loc[target, :]
+
+    def time_loc_all_scalars(self, unique_levels):
+        target = tuple([self.tgt_scalar] * self.nlevels)
+        self.df.loc[target, :]
+
+    def time_loc_all_bool_indexers(self, unique_levels):
+        target = tuple([self.tgt_bool_indexer] * self.nlevels)
+        self.df.loc[target, :]
+
+    def time_loc_slice_plus_null_slice(self, unique_levels):
+        target = (self.tgt_slice, self.tgt_null_slice)
+        self.df.loc[target, :]
+
+    def time_loc_null_slice_plus_slice(self, unique_levels):
+        target = (self.tgt_null_slice, self.tgt_slice)
+        self.df.loc[target, :]
+
+    def time_loc_multiindex(self, unique_levels):
+        target = self.df.index[::10]
+        self.df.loc[target]
+
+    def time_xs_level_0(self, unique_levels):
+        target = self.tgt_scalar
+        self.df.xs(target, level=0)
+
+    def time_xs_level_1(self, unique_levels):
+        target = self.tgt_scalar
+        self.df.xs(target, level=1)
+
+    def time_xs_full_key(self, unique_levels):
+        target = tuple([self.tgt_scalar] * self.nlevels)
+        self.df.xs(target)
+
+
+class IntervalIndexing:
+    def setup_cache(self):
+        idx = IntervalIndex.from_breaks(np.arange(1000001))
+        monotonic = Series(np.arange(1000000), index=idx)
+        return monotonic
+
+    def time_getitem_scalar(self, monotonic):
+        monotonic[80000]
+
+    def time_loc_scalar(self, monotonic):
+        monotonic.loc[80000]
+
+    def time_getitem_list(self, monotonic):
+        monotonic[80000:]
+
+    def time_loc_list(self, monotonic):
+        monotonic.loc[80000:]
+
+
+class DatetimeIndexIndexing:
+    def setup(self):
+        dti = date_range("2016-01-01", periods=10000, tz="US/Pacific")
+        dti2 = dti.tz_convert("UTC")
+        self.dti = dti
+        self.dti2 = dti2
+
+    def time_get_indexer_mismatched_tz(self):
+        # reached via e.g.
+        #  ser = Series(range(len(dti)), index=dti)
+        #  ser[dti2]
+        self.dti.get_indexer(self.dti2)
+
+
+class SortedAndUnsortedDatetimeIndexLoc:
+    def setup(self):
+        dti = date_range("2016-01-01", periods=10000, tz="US/Pacific")
+        index = np.array(dti)
+
+        unsorted_index = index.copy()
+        unsorted_index[10] = unsorted_index[20]
+
+        self.df_unsorted = DataFrame(index=unsorted_index, data={"a": 1})
+        self.df_sort = DataFrame(index=index, data={"a": 1})
+
+    def time_loc_unsorted(self):
+        self.df_unsorted.loc["2016-6-11"]
+
+    def time_loc_sorted(self):
+        self.df_sort.loc["2016-6-11"]
+
+
+class CategoricalIndexIndexing:
+    params = ["monotonic_incr", "monotonic_decr", "non_monotonic"]
+    param_names = ["index"]
+
+    def setup(self, index):
+        N = 10**5
+        values = list("a" * N + "b" * N + "c" * N)
+        indices = {
+            "monotonic_incr": CategoricalIndex(values),
+            "monotonic_decr": CategoricalIndex(reversed(values)),
+            "non_monotonic": CategoricalIndex(list("abc" * N)),
+        }
+        self.data = indices[index]
+        self.data_unique = CategoricalIndex([str(i) for i in range(N * 3)])
+
+        self.int_scalar = 10000
+        self.int_list = list(range(10000))
+
+        self.cat_scalar = "b"
+        self.cat_list = ["1", "3"]
+
+    def time_getitem_scalar(self, index):
+        self.data[self.int_scalar]
+
+    def time_getitem_slice(self, index):
+        self.data[: self.int_scalar]
+
+    def time_getitem_list_like(self, index):
+        self.data[[self.int_scalar]]
+
+    def time_getitem_list(self, index):
+        self.data[self.int_list]
+
+    def time_getitem_bool_array(self, index):
+        self.data[self.data == self.cat_scalar]
+
+    def time_get_loc_scalar(self, index):
+        self.data.get_loc(self.cat_scalar)
+
+    def time_get_indexer_list(self, index):
+        self.data_unique.get_indexer(self.cat_list)
+
+
+class MethodLookup:
+    def setup_cache(self):
+        s = Series()
+        return s
+
+    def time_lookup_iloc(self, s):
+        s.iloc
+
+    def time_lookup_loc(self, s):
+        s.loc
+
+
+class GetItemSingleColumn:
+    def setup(self):
+        self.df_string_col = DataFrame(np.random.randn(3000, 1), columns=["A"])
+        self.df_int_col = DataFrame(np.random.randn(3000, 1))
+
+    def time_frame_getitem_single_column_label(self):
+        self.df_string_col["A"]
+
+    def time_frame_getitem_single_column_int(self):
+        self.df_int_col[0]
+
+
+class IndexSingleRow:
+    params = [True, False]
+    param_names = ["unique_cols"]
+
+    def setup(self, unique_cols):
+        arr = np.arange(10**7).reshape(-1, 10)
+        df = DataFrame(arr)
+        dtypes = ["u1", "u2", "u4", "u8", "i1", "i2", "i4", "i8", "f8", "f4"]
+        for i, d in enumerate(dtypes):
+            df[i] = df[i].astype(d)
+
+        if not unique_cols:
+            # GH#33032 single-row lookups with non-unique columns were
+            #  15x slower than with unique columns
+            df.columns = ["A", "A"] + list(df.columns[2:])
+
+        self.df = df
+
+    def time_iloc_row(self, unique_cols):
+        self.df.iloc[10000]
+
+    def time_loc_row(self, unique_cols):
+        self.df.loc[10000]
+
+
+class AssignTimeseriesIndex:
+    def setup(self):
+        N = 100000
+        idx = date_range("1/1/2000", periods=N, freq="h")
+        self.df = DataFrame(np.random.randn(N, 1), columns=["A"], index=idx)
+
+    def time_frame_assign_timeseries_index(self):
+        self.df["date"] = self.df.index
+
+
+class InsertColumns:
+    def setup(self):
+        self.N = 10**3
+        self.df = DataFrame(index=range(self.N))
+        self.df2 = DataFrame(np.random.randn(self.N, 2))
+
+    def time_insert(self):
+        for i in range(100):
+            self.df.insert(0, i, np.random.randn(self.N), allow_duplicates=True)
+
+    def time_insert_middle(self):
+        # same as time_insert but inserting to a middle column rather than
+        #  front or back (which have fast-paths)
+        for i in range(100):
+            self.df2.insert(
+                1, "colname", np.random.randn(self.N), allow_duplicates=True
+            )
+
+    def time_assign_with_setitem(self):
+        for i in range(100):
+            self.df[i] = np.random.randn(self.N)
+
+    def time_assign_list_like_with_setitem(self):
+        self.df[list(range(100))] = np.random.randn(self.N, 100)
+
+    def time_assign_list_of_columns_concat(self):
+        df = DataFrame(np.random.randn(self.N, 100))
+        concat([self.df, df], axis=1)
+
+
+class Setitem:
+    def setup(self):
+        N = 500_000
+        cols = 500
+        self.df = DataFrame(np.random.rand(N, cols))
+
+    def time_setitem(self):
+        self.df[100] = 100
+
+    def time_setitem_list(self):
+        self.df[[100, 200, 300]] = 100
+
+
+class SetitemObjectDtype:
+    # GH#19299
+
+    def setup(self):
+        N = 1000
+        cols = 500
+        self.df = DataFrame(index=range(N), columns=range(cols), dtype=object)
+
+    def time_setitem_object_dtype(self):
+        self.df.loc[0, 1] = 1.0
+
+
+class ChainIndexing:
+    params = [None, "warn"]
+    param_names = ["mode"]
+
+    def setup(self, mode):
+        self.N = 1000000
+        self.df = DataFrame({"A": np.arange(self.N), "B": "foo"})
+
+    def time_chained_indexing(self, mode):
+        df = self.df
+        N = self.N
+        with warnings.catch_warnings(record=True):
+            with option_context("mode.chained_assignment", mode):
+                df2 = df[df.A > N // 2]
+                df2["C"] = 1.0
+
+
+class Block:
+    def setup(self):
+        self.df = DataFrame(
+            False,
+            columns=np.arange(500).astype(str),
+            index=date_range("2010-01-01", "2011-01-01"),
+        )
+
+    def time_test(self):
+        start = datetime(2010, 5, 1)
+        end = datetime(2010, 9, 1)
+        self.df.loc[start:end, :] = True
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/indexing_engines.py b/pandas/asv_bench/benchmarks/indexing_engines.py
new file mode 100644
index 0000000000000..5e3c593e269cb
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/indexing_engines.py
@@ -0,0 +1,176 @@
+"""
+Benchmarks in this file depend mostly on code in _libs/
+
+We have to created masked arrays to test the masked engine though. The
+array is unpacked on the Cython level.
+
+If a PR does not edit anything in _libs, it is very unlikely that benchmarks
+in this file will be affected.
+"""
+
+import numpy as np
+
+from pandas._libs import index as libindex
+
+from pandas.core.arrays import BaseMaskedArray
+
+
+def _get_numeric_engines():
+    engine_names = [
+        ("Int64Engine", np.int64),
+        ("Int32Engine", np.int32),
+        ("Int16Engine", np.int16),
+        ("Int8Engine", np.int8),
+        ("UInt64Engine", np.uint64),
+        ("UInt32Engine", np.uint32),
+        ("UInt16engine", np.uint16),
+        ("UInt8Engine", np.uint8),
+        ("Float64Engine", np.float64),
+        ("Float32Engine", np.float32),
+    ]
+    return [
+        (getattr(libindex, engine_name), dtype)
+        for engine_name, dtype in engine_names
+        if hasattr(libindex, engine_name)
+    ]
+
+
+def _get_masked_engines():
+    engine_names = [
+        ("MaskedInt64Engine", "Int64"),
+        ("MaskedInt32Engine", "Int32"),
+        ("MaskedInt16Engine", "Int16"),
+        ("MaskedInt8Engine", "Int8"),
+        ("MaskedUInt64Engine", "UInt64"),
+        ("MaskedUInt32Engine", "UInt32"),
+        ("MaskedUInt16engine", "UInt16"),
+        ("MaskedUInt8Engine", "UInt8"),
+        ("MaskedFloat64Engine", "Float64"),
+        ("MaskedFloat32Engine", "Float32"),
+    ]
+    return [
+        (getattr(libindex, engine_name), dtype)
+        for engine_name, dtype in engine_names
+        if hasattr(libindex, engine_name)
+    ]
+
+
+class NumericEngineIndexing:
+    params = [
+        _get_numeric_engines(),
+        ["monotonic_incr", "monotonic_decr", "non_monotonic"],
+        [True, False],
+        [10**5, 2 * 10**6],  # 2e6 is above SIZE_CUTOFF
+    ]
+    param_names = ["engine_and_dtype", "index_type", "unique", "N"]
+
+    def setup(self, engine_and_dtype, index_type, unique, N):
+        engine, dtype = engine_and_dtype
+
+        if index_type == "monotonic_incr":
+            if unique:
+                arr = np.arange(N * 3, dtype=dtype)
+            else:
+                arr = np.array([1, 2, 3], dtype=dtype).repeat(N)
+        elif index_type == "monotonic_decr":
+            if unique:
+                arr = np.arange(N * 3, dtype=dtype)[::-1]
+            else:
+                arr = np.array([3, 2, 1], dtype=dtype).repeat(N)
+        else:
+            assert index_type == "non_monotonic"
+            if unique:
+                arr = np.empty(N * 3, dtype=dtype)
+                arr[:N] = np.arange(N * 2, N * 3, dtype=dtype)
+                arr[N:] = np.arange(N * 2, dtype=dtype)
+            else:
+                arr = np.array([1, 2, 3], dtype=dtype).repeat(N)
+
+        self.data = engine(arr)
+        # code below avoids populating the mapping etc. while timing.
+        self.data.get_loc(2)
+
+        self.key_middle = arr[len(arr) // 2]
+        self.key_early = arr[2]
+
+    def time_get_loc(self, engine_and_dtype, index_type, unique, N):
+        self.data.get_loc(self.key_early)
+
+    def time_get_loc_near_middle(self, engine_and_dtype, index_type, unique, N):
+        # searchsorted performance may be different near the middle of a range
+        #  vs near an endpoint
+        self.data.get_loc(self.key_middle)
+
+
+class MaskedNumericEngineIndexing:
+    params = [
+        _get_masked_engines(),
+        ["monotonic_incr", "monotonic_decr", "non_monotonic"],
+        [True, False],
+        [10**5, 2 * 10**6],  # 2e6 is above SIZE_CUTOFF
+    ]
+    param_names = ["engine_and_dtype", "index_type", "unique", "N"]
+
+    def setup(self, engine_and_dtype, index_type, unique, N):
+        engine, dtype = engine_and_dtype
+        dtype = dtype.lower()
+
+        if index_type == "monotonic_incr":
+            if unique:
+                arr = np.arange(N * 3, dtype=dtype)
+            else:
+                arr = np.array([1, 2, 3], dtype=dtype).repeat(N)
+            mask = np.zeros(N * 3, dtype=np.bool_)
+        elif index_type == "monotonic_decr":
+            if unique:
+                arr = np.arange(N * 3, dtype=dtype)[::-1]
+            else:
+                arr = np.array([3, 2, 1], dtype=dtype).repeat(N)
+            mask = np.zeros(N * 3, dtype=np.bool_)
+        else:
+            assert index_type == "non_monotonic"
+            if unique:
+                arr = np.zeros(N * 3, dtype=dtype)
+                arr[:N] = np.arange(N * 2, N * 3, dtype=dtype)
+                arr[N:] = np.arange(N * 2, dtype=dtype)
+
+            else:
+                arr = np.array([1, 2, 3], dtype=dtype).repeat(N)
+            mask = np.zeros(N * 3, dtype=np.bool_)
+            mask[-1] = True
+
+        self.data = engine(BaseMaskedArray(arr, mask))
+        # code below avoids populating the mapping etc. while timing.
+        self.data.get_loc(2)
+
+        self.key_middle = arr[len(arr) // 2]
+        self.key_early = arr[2]
+
+    def time_get_loc(self, engine_and_dtype, index_type, unique, N):
+        self.data.get_loc(self.key_early)
+
+    def time_get_loc_near_middle(self, engine_and_dtype, index_type, unique, N):
+        # searchsorted performance may be different near the middle of a range
+        #  vs near an endpoint
+        self.data.get_loc(self.key_middle)
+
+
+class ObjectEngineIndexing:
+    params = [("monotonic_incr", "monotonic_decr", "non_monotonic")]
+    param_names = ["index_type"]
+
+    def setup(self, index_type):
+        N = 10**5
+        values = list("a" * N + "b" * N + "c" * N)
+        arr = {
+            "monotonic_incr": np.array(values, dtype=object),
+            "monotonic_decr": np.array(list(reversed(values)), dtype=object),
+            "non_monotonic": np.array(list("abc") * N, dtype=object),
+        }[index_type]
+
+        self.data = libindex.ObjectEngine(arr)
+        # code below avoids populating the mapping etc. while timing.
+        self.data.get_loc("b")
+
+    def time_get_loc(self, index_type):
+        self.data.get_loc("b")
diff --git a/pandas/asv_bench/benchmarks/inference.py b/pandas/asv_bench/benchmarks/inference.py
new file mode 100644
index 0000000000000..ce3935d2cd0ac
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/inference.py
@@ -0,0 +1,297 @@
+"""
+The functions benchmarked in this file depend _almost_ exclusively on
+_libs, but not in a way that is easy to formalize.
+
+If a PR does not change anything in pandas/_libs/ or pandas/core/tools/, then
+it is likely that these benchmarks will be unaffected.
+"""
+
+import numpy as np
+
+from pandas import (
+    Index,
+    NaT,
+    Series,
+    date_range,
+    to_datetime,
+    to_numeric,
+    to_timedelta,
+)
+
+from .pandas_vb_common import lib
+
+
+class ToNumeric:
+    def setup(self):
+        N = 10000
+        self.float = Series(np.random.randn(N))
+        self.numstr = self.float.astype("str")
+        self.str = Series(Index([f"i-{i}" for i in range(N)], dtype=object))
+
+    def time_from_float(self):
+        to_numeric(self.float, errors="coerce")
+
+    def time_from_numeric_str(self):
+        to_numeric(self.numstr, errors="coerce")
+
+    def time_from_str(self):
+        to_numeric(self.str, errors="coerce")
+
+
+class ToNumericDowncast:
+    param_names = ["dtype", "downcast"]
+    params = [
+        [
+            "string-float",
+            "string-int",
+            "string-nint",
+            "datetime64",
+            "int-list",
+            "int32",
+        ],
+        [None, "integer", "signed", "unsigned", "float"],
+    ]
+
+    N = 500000
+    N2 = N // 2
+
+    data_dict = {
+        "string-int": ["1"] * N2 + [2] * N2,
+        "string-nint": ["-1"] * N2 + [2] * N2,
+        "datetime64": np.repeat(
+            np.array(["1970-01-01", "1970-01-02"], dtype="datetime64[D]"), N
+        ),
+        "string-float": ["1.1"] * N2 + [2] * N2,
+        "int-list": [1] * N2 + [2] * N2,
+        "int32": np.repeat(np.int32(1), N),
+    }
+
+    def setup(self, dtype, downcast):
+        self.data = self.data_dict[dtype]
+
+    def time_downcast(self, dtype, downcast):
+        to_numeric(self.data, downcast=downcast)
+
+
+class MaybeConvertNumeric:
+    # maybe_convert_numeric depends _exclusively_ on _libs, could
+    #  go in benchmarks/libs.py
+
+    def setup_cache(self):
+        N = 10**6
+        arr = np.repeat([2**63], N) + np.arange(N).astype("uint64")
+        data = arr.astype(object)
+        data[1::2] = arr[1::2].astype(str)
+        data[-1] = -1
+        return data
+
+    def time_convert(self, data):
+        lib.maybe_convert_numeric(data, set(), coerce_numeric=False)
+
+
+class MaybeConvertObjects:
+    # maybe_convert_objects depends _almost_ exclusively on _libs, but
+    #  does have some run-time imports from outside of _libs
+
+    def setup(self):
+        N = 10**5
+
+        data = list(range(N))
+        data[0] = NaT
+        data = np.array(data)
+        self.data = data
+
+    def time_maybe_convert_objects(self):
+        lib.maybe_convert_objects(self.data)
+
+
+class ToDatetimeFromIntsFloats:
+    def setup(self):
+        self.ts_sec = Series(range(1521080307, 1521685107), dtype="int64")
+        self.ts_sec_uint = Series(range(1521080307, 1521685107), dtype="uint64")
+        self.ts_sec_float = self.ts_sec.astype("float64")
+
+        self.ts_nanosec = 1_000_000 * self.ts_sec
+        self.ts_nanosec_uint = 1_000_000 * self.ts_sec_uint
+        self.ts_nanosec_float = self.ts_nanosec.astype("float64")
+
+    # speed of int64, uint64 and float64 paths should be comparable
+
+    def time_nanosec_int64(self):
+        to_datetime(self.ts_nanosec, unit="ns")
+
+    def time_nanosec_uint64(self):
+        to_datetime(self.ts_nanosec_uint, unit="ns")
+
+    def time_nanosec_float64(self):
+        to_datetime(self.ts_nanosec_float, unit="ns")
+
+    def time_sec_uint64(self):
+        to_datetime(self.ts_sec_uint, unit="s")
+
+    def time_sec_int64(self):
+        to_datetime(self.ts_sec, unit="s")
+
+    def time_sec_float64(self):
+        to_datetime(self.ts_sec_float, unit="s")
+
+
+class ToDatetimeYYYYMMDD:
+    def setup(self):
+        rng = date_range(start="1/1/2000", periods=10000, freq="D")
+        self.stringsD = Series(rng.strftime("%Y%m%d"))
+
+    def time_format_YYYYMMDD(self):
+        to_datetime(self.stringsD, format="%Y%m%d")
+
+
+class ToDatetimeCacheSmallCount:
+    params = ([True, False], [50, 500, 5000, 100000])
+    param_names = ["cache", "count"]
+
+    def setup(self, cache, count):
+        rng = date_range(start="1/1/1971", periods=count)
+        self.unique_date_strings = rng.strftime("%Y-%m-%d").tolist()
+
+    def time_unique_date_strings(self, cache, count):
+        to_datetime(self.unique_date_strings, cache=cache)
+
+
+class ToDatetimeISO8601:
+    def setup(self):
+        rng = date_range(start="1/1/2000", periods=20000, freq="h")
+        self.strings = rng.strftime("%Y-%m-%d %H:%M:%S").tolist()
+        self.strings_nosep = rng.strftime("%Y%m%d %H:%M:%S").tolist()
+        self.strings_tz_space = [
+            x.strftime("%Y-%m-%d %H:%M:%S") + " -0800" for x in rng
+        ]
+        self.strings_zero_tz = [x.strftime("%Y-%m-%d %H:%M:%S") + "Z" for x in rng]
+
+    def time_iso8601(self):
+        to_datetime(self.strings)
+
+    def time_iso8601_nosep(self):
+        to_datetime(self.strings_nosep)
+
+    def time_iso8601_format(self):
+        to_datetime(self.strings, format="%Y-%m-%d %H:%M:%S")
+
+    def time_iso8601_format_no_sep(self):
+        to_datetime(self.strings_nosep, format="%Y%m%d %H:%M:%S")
+
+    def time_iso8601_tz_spaceformat(self):
+        to_datetime(self.strings_tz_space)
+
+    def time_iso8601_infer_zero_tz_fromat(self):
+        # GH 41047
+        to_datetime(self.strings_zero_tz)
+
+
+class ToDatetimeNONISO8601:
+    def setup(self):
+        N = 10000
+        half = N // 2
+        ts_string_1 = "March 1, 2018 12:00:00+0400"
+        ts_string_2 = "March 1, 2018 12:00:00+0500"
+        self.same_offset = [ts_string_1] * N
+        self.diff_offset = [ts_string_1] * half + [ts_string_2] * half
+
+    def time_same_offset(self):
+        to_datetime(self.same_offset)
+
+    def time_different_offset(self):
+        to_datetime(self.diff_offset, utc=True)
+
+
+class ToDatetimeFormatQuarters:
+    def setup(self):
+        self.s = Series(["2Q2005", "2Q05", "2005Q1", "05Q1"] * 10000)
+
+    def time_infer_quarter(self):
+        to_datetime(self.s)
+
+
+class ToDatetimeFormat:
+    def setup(self):
+        N = 100000
+        self.s = Series(["19MAY11", "19MAY11:00:00:00"] * N)
+        self.s2 = self.s.str.replace(":\\S+$", "", regex=True)
+
+        self.same_offset = ["10/11/2018 00:00:00.045-07:00"] * N
+        self.diff_offset = [
+            f"10/11/2018 00:00:00.045-0{offset}:00" for offset in range(10)
+        ] * (N // 10)
+
+    def time_exact(self):
+        to_datetime(self.s2, format="%d%b%y")
+
+    def time_no_exact(self):
+        to_datetime(self.s, format="%d%b%y", exact=False)
+
+    def time_same_offset(self):
+        to_datetime(self.same_offset, format="%m/%d/%Y %H:%M:%S.%f%z")
+
+    def time_same_offset_to_utc(self):
+        to_datetime(self.same_offset, format="%m/%d/%Y %H:%M:%S.%f%z", utc=True)
+
+    def time_different_offset_to_utc(self):
+        to_datetime(self.diff_offset, format="%m/%d/%Y %H:%M:%S.%f%z", utc=True)
+
+
+class ToDatetimeCache:
+    params = [True, False]
+    param_names = ["cache"]
+
+    def setup(self, cache):
+        N = 10000
+        self.unique_numeric_seconds = list(range(N))
+        self.dup_numeric_seconds = [1000] * N
+        self.dup_string_dates = ["2000-02-11"] * N
+        self.dup_string_with_tz = ["2000-02-11 15:00:00-0800"] * N
+
+    def time_unique_seconds_and_unit(self, cache):
+        to_datetime(self.unique_numeric_seconds, unit="s", cache=cache)
+
+    def time_dup_seconds_and_unit(self, cache):
+        to_datetime(self.dup_numeric_seconds, unit="s", cache=cache)
+
+    def time_dup_string_dates(self, cache):
+        to_datetime(self.dup_string_dates, cache=cache)
+
+    def time_dup_string_dates_and_format(self, cache):
+        to_datetime(self.dup_string_dates, format="%Y-%m-%d", cache=cache)
+
+    def time_dup_string_tzoffset_dates(self, cache):
+        to_datetime(self.dup_string_with_tz, cache=cache)
+
+
+class ToTimedelta:
+    def setup(self):
+        self.ints = np.random.randint(0, 60, size=10000)
+        self.str_days = []
+        self.str_seconds = []
+        for i in self.ints:
+            self.str_days.append(f"{i} days")
+            self.str_seconds.append(f"00:00:{i:02d}")
+
+    def time_convert_int(self):
+        to_timedelta(self.ints, unit="s")
+
+    def time_convert_string_days(self):
+        to_timedelta(self.str_days)
+
+    def time_convert_string_seconds(self):
+        to_timedelta(self.str_seconds)
+
+
+class ToTimedeltaErrors:
+    def setup(self):
+        ints = np.random.randint(0, 60, size=10000)
+        self.arr = [f"{i} days" for i in ints]
+        self.arr[-1] = "apple"
+
+    def time_convert(self):
+        to_timedelta(self.arr, errors="coerce")
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/__init__.py b/pandas/asv_bench/benchmarks/io/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/asv_bench/benchmarks/io/csv.py b/pandas/asv_bench/benchmarks/io/csv.py
new file mode 100644
index 0000000000000..3a15f754ae523
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/csv.py
@@ -0,0 +1,626 @@
+from io import (
+    BytesIO,
+    StringIO,
+)
+import random
+import string
+
+import numpy as np
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    concat,
+    date_range,
+    period_range,
+    read_csv,
+    to_datetime,
+)
+
+from ..pandas_vb_common import BaseIO
+
+
+class ToCSV(BaseIO):
+    fname = "__test__.csv"
+    params = ["wide", "long", "mixed"]
+    param_names = ["kind"]
+
+    def setup(self, kind):
+        wide_frame = DataFrame(np.random.randn(3000, 30))
+        long_frame = DataFrame(
+            {
+                "A": np.arange(50000),
+                "B": np.arange(50000) + 1.0,
+                "C": np.arange(50000) + 2.0,
+                "D": np.arange(50000) + 3.0,
+            }
+        )
+        mixed_frame = DataFrame(
+            {
+                "float": np.random.randn(5000),
+                "int": np.random.randn(5000).astype(int),
+                "bool": (np.arange(5000) % 2) == 0,
+                "datetime": date_range("2001", freq="s", periods=5000),
+                "object": ["foo"] * 5000,
+            }
+        )
+        mixed_frame.loc[30:500, "float"] = np.nan
+        data = {"wide": wide_frame, "long": long_frame, "mixed": mixed_frame}
+        self.df = data[kind]
+
+    def time_frame(self, kind):
+        self.df.to_csv(self.fname)
+
+
+class ToCSVMultiIndexUnusedLevels(BaseIO):
+    fname = "__test__.csv"
+
+    def setup(self):
+        df = DataFrame({"a": np.random.randn(100_000), "b": 1, "c": 1})
+        self.df = df.set_index(["a", "b"])
+        self.df_unused_levels = self.df.iloc[:10_000]
+        self.df_single_index = df.set_index(["a"]).iloc[:10_000]
+
+    def time_full_frame(self):
+        self.df.to_csv(self.fname)
+
+    def time_sliced_frame(self):
+        self.df_unused_levels.to_csv(self.fname)
+
+    def time_single_index_frame(self):
+        self.df_single_index.to_csv(self.fname)
+
+
+class ToCSVDatetime(BaseIO):
+    fname = "__test__.csv"
+
+    def setup(self):
+        rng = date_range("1/1/2000", periods=1000)
+        self.data = DataFrame(rng, index=rng)
+
+    def time_frame_date_formatting(self):
+        self.data.to_csv(self.fname, date_format="%Y%m%d")
+
+
+class ToCSVDatetimeIndex(BaseIO):
+    fname = "__test__.csv"
+
+    def setup(self):
+        rng = date_range("2000", periods=100_000, freq="s")
+        self.data = DataFrame({"a": 1}, index=rng)
+
+    def time_frame_date_formatting_index(self):
+        self.data.to_csv(self.fname, date_format="%Y-%m-%d %H:%M:%S")
+
+    def time_frame_date_no_format_index(self):
+        self.data.to_csv(self.fname)
+
+
+class ToCSVPeriod(BaseIO):
+    fname = "__test__.csv"
+
+    params = ([1000, 10000], ["D", "h"])
+    param_names = ["nobs", "freq"]
+
+    def setup(self, nobs, freq):
+        rng = period_range(start="2000-01-01", periods=nobs, freq=freq)
+        self.data = DataFrame(rng)
+        if freq == "D":
+            self.default_fmt = "%Y-%m-%d"
+        elif freq == "h":
+            self.default_fmt = "%Y-%m-%d %H:00"
+
+    def time_frame_period_formatting_default(self, nobs, freq):
+        self.data.to_csv(self.fname)
+
+    def time_frame_period_formatting_default_explicit(self, nobs, freq):
+        self.data.to_csv(self.fname, date_format=self.default_fmt)
+
+    def time_frame_period_formatting(self, nobs, freq):
+        # Nb: `date_format` is not actually taken into account here today, so the
+        # performance is currently identical to `time_frame_period_formatting_default`
+        # above. This timer is therefore expected to degrade when GH#51621 is fixed.
+        # (Remove this comment when GH#51621 is fixed.)
+        self.data.to_csv(self.fname, date_format="%Y-%m-%d___%H:%M:%S")
+
+
+class ToCSVPeriodIndex(BaseIO):
+    fname = "__test__.csv"
+
+    params = ([1000, 10000], ["D", "h"])
+    param_names = ["nobs", "freq"]
+
+    def setup(self, nobs, freq):
+        rng = period_range(start="2000-01-01", periods=nobs, freq=freq)
+        self.data = DataFrame({"a": 1}, index=rng)
+        if freq == "D":
+            self.default_fmt = "%Y-%m-%d"
+        elif freq == "h":
+            self.default_fmt = "%Y-%m-%d %H:00"
+
+    def time_frame_period_formatting_index(self, nobs, freq):
+        self.data.to_csv(self.fname, date_format="%Y-%m-%d___%H:%M:%S")
+
+    def time_frame_period_formatting_index_default(self, nobs, freq):
+        self.data.to_csv(self.fname)
+
+    def time_frame_period_formatting_index_default_explicit(self, nobs, freq):
+        self.data.to_csv(self.fname, date_format=self.default_fmt)
+
+
+class ToCSVDatetimeBig(BaseIO):
+    fname = "__test__.csv"
+    timeout = 1500
+    params = [1000, 10000, 100000]
+    param_names = ["nobs"]
+
+    def setup(self, nobs):
+        d = "2018-11-29"
+        dt = "2018-11-26 11:18:27.0"
+        self.data = DataFrame(
+            {
+                "dt": [np.datetime64(dt)] * nobs,
+                "d": [np.datetime64(d)] * nobs,
+                "r": [np.random.uniform()] * nobs,
+            }
+        )
+
+    def time_frame(self, nobs):
+        self.data.to_csv(self.fname)
+
+
+class ToCSVIndexes(BaseIO):
+    fname = "__test__.csv"
+
+    @staticmethod
+    def _create_df(rows, cols):
+        index_cols = {
+            "index1": np.random.randint(0, rows, rows),
+            "index2": np.full(rows, 1, dtype=int),
+            "index3": np.full(rows, 1, dtype=int),
+        }
+        data_cols = {
+            f"col{i}": np.random.uniform(0, 100000.0, rows) for i in range(cols)
+        }
+        df = DataFrame({**index_cols, **data_cols})
+        return df
+
+    def setup(self):
+        ROWS = 100000
+        COLS = 5
+        # For tests using .head(), create an initial dataframe with this many times
+        # more rows
+        HEAD_ROW_MULTIPLIER = 10
+
+        self.df_standard_index = self._create_df(ROWS, COLS)
+
+        self.df_custom_index_then_head = (
+            self._create_df(ROWS * HEAD_ROW_MULTIPLIER, COLS)
+            .set_index(["index1", "index2", "index3"])
+            .head(ROWS)
+        )
+
+        self.df_head_then_custom_index = (
+            self._create_df(ROWS * HEAD_ROW_MULTIPLIER, COLS)
+            .head(ROWS)
+            .set_index(["index1", "index2", "index3"])
+        )
+
+    def time_standard_index(self):
+        self.df_standard_index.to_csv(self.fname)
+
+    def time_multiindex(self):
+        self.df_head_then_custom_index.to_csv(self.fname)
+
+    def time_head_of_multiindex(self):
+        self.df_custom_index_then_head.to_csv(self.fname)
+
+
+class StringIORewind:
+    def data(self, stringio_object):
+        stringio_object.seek(0)
+        return stringio_object
+
+
+class ReadCSVDInferDatetimeFormat(StringIORewind):
+    params = [None, "custom", "iso8601", "ymd"]
+    param_names = ["format"]
+
+    def setup(self, format):
+        rng = date_range("1/1/2000", periods=1000)
+        formats = {
+            None: None,
+            "custom": "%m/%d/%Y %H:%M:%S.%f",
+            "iso8601": "%Y-%m-%d %H:%M:%S",
+            "ymd": "%Y%m%d",
+        }
+        dt_format = formats[format]
+        self.StringIO_input = StringIO("\n".join(rng.strftime(dt_format).tolist()))
+
+    def time_read_csv(self, format):
+        read_csv(
+            self.data(self.StringIO_input),
+            header=None,
+            names=["foo"],
+            parse_dates=["foo"],
+        )
+
+
+class ReadCSVConcatDatetime(StringIORewind):
+    iso8601 = "%Y-%m-%d %H:%M:%S"
+
+    def setup(self):
+        rng = date_range("1/1/2000", periods=50000, freq="s")
+        self.StringIO_input = StringIO("\n".join(rng.strftime(self.iso8601).tolist()))
+
+    def time_read_csv(self):
+        read_csv(
+            self.data(self.StringIO_input),
+            header=None,
+            names=["foo"],
+            parse_dates=["foo"],
+        )
+
+
+class ReadCSVConcatDatetimeBadDateValue(StringIORewind):
+    params = (["nan", "0", ""],)
+    param_names = ["bad_date_value"]
+
+    def setup(self, bad_date_value):
+        self.StringIO_input = StringIO((f"{bad_date_value},\n") * 50000)
+
+    def time_read_csv(self, bad_date_value):
+        read_csv(
+            self.data(self.StringIO_input),
+            header=None,
+            names=["foo", "bar"],
+            parse_dates=["foo"],
+        )
+
+
+class ReadCSVSkipRows(BaseIO):
+    fname = "__test__.csv"
+    params = ([None, 10000], ["c", "python", "pyarrow"])
+    param_names = ["skiprows", "engine"]
+
+    def setup(self, skiprows, engine):
+        N = 20000
+        index = Index([f"i-{i}" for i in range(N)], dtype=object)
+        df = DataFrame(
+            {
+                "float1": np.random.randn(N),
+                "float2": np.random.randn(N),
+                "string1": ["foo"] * N,
+                "bool1": [True] * N,
+                "int1": np.random.randint(0, N, size=N),
+            },
+            index=index,
+        )
+        df.to_csv(self.fname)
+
+    def time_skipprows(self, skiprows, engine):
+        read_csv(self.fname, skiprows=skiprows, engine=engine)
+
+
+class ReadUint64Integers(StringIORewind):
+    def setup(self):
+        self.na_values = [2**63 + 500]
+        arr = np.arange(10000).astype("uint64") + 2**63
+        self.data1 = StringIO("\n".join(arr.astype(str).tolist()))
+        arr = arr.astype(object)
+        arr[500] = -1
+        self.data2 = StringIO("\n".join(arr.astype(str).tolist()))
+
+    def time_read_uint64(self):
+        read_csv(self.data(self.data1), header=None, names=["foo"])
+
+    def time_read_uint64_neg_values(self):
+        read_csv(self.data(self.data2), header=None, names=["foo"])
+
+    def time_read_uint64_na_values(self):
+        read_csv(
+            self.data(self.data1), header=None, names=["foo"], na_values=self.na_values
+        )
+
+
+class ReadCSVThousands(BaseIO):
+    fname = "__test__.csv"
+    params = ([",", "|"], [None, ","], ["c", "python"])
+    param_names = ["sep", "thousands", "engine"]
+
+    def setup(self, sep, thousands, engine):
+        N = 10000
+        K = 8
+        data = np.random.randn(N, K) * np.random.randint(100, 10000, (N, K))
+        df = DataFrame(data)
+        if thousands is not None:
+            fmt = f":{thousands}"
+            fmt = "{" + fmt + "}"
+            df = df.map(lambda x: fmt.format(x))
+        df.to_csv(self.fname, sep=sep)
+
+    def time_thousands(self, sep, thousands, engine):
+        read_csv(self.fname, sep=sep, thousands=thousands, engine=engine)
+
+
+class ReadCSVComment(StringIORewind):
+    params = ["c", "python"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        data = ["A,B,C"] + (["1,2,3 # comment"] * 100000)
+        self.StringIO_input = StringIO("\n".join(data))
+
+    def time_comment(self, engine):
+        read_csv(
+            self.data(self.StringIO_input), comment="#", header=None, names=list("abc")
+        )
+
+
+class ReadCSVFloatPrecision(StringIORewind):
+    params = ([",", ";"], [".", "_"], [None, "high", "round_trip"])
+    param_names = ["sep", "decimal", "float_precision"]
+
+    def setup(self, sep, decimal, float_precision):
+        floats = [
+            "".join([random.choice(string.digits) for _ in range(28)])
+            for _ in range(15)
+        ]
+        rows = sep.join([f"0{decimal}{{}}"] * 3) + "\n"
+        data = rows * 5
+        data = data.format(*floats) * 200  # 1000 x 3 strings csv
+        self.StringIO_input = StringIO(data)
+
+    def time_read_csv(self, sep, decimal, float_precision):
+        read_csv(
+            self.data(self.StringIO_input),
+            sep=sep,
+            header=None,
+            names=list("abc"),
+            float_precision=float_precision,
+        )
+
+    def time_read_csv_python_engine(self, sep, decimal, float_precision):
+        read_csv(
+            self.data(self.StringIO_input),
+            sep=sep,
+            header=None,
+            engine="python",
+            float_precision=None,
+            names=list("abc"),
+        )
+
+
+class ReadCSVEngine(StringIORewind):
+    params = ["c", "python", "pyarrow"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        data = ["A,B,C,D,E"] + (["1,2,3,4,5"] * 100000)
+        self.StringIO_input = StringIO("\n".join(data))
+        # simulate reading from file
+        self.BytesIO_input = BytesIO(self.StringIO_input.read().encode("utf-8"))
+
+    def time_read_stringcsv(self, engine):
+        read_csv(self.data(self.StringIO_input), engine=engine)
+
+    def time_read_bytescsv(self, engine):
+        read_csv(self.data(self.BytesIO_input), engine=engine)
+
+    def peakmem_read_csv(self, engine):
+        read_csv(self.data(self.BytesIO_input), engine=engine)
+
+
+class ReadCSVCategorical(BaseIO):
+    fname = "__test__.csv"
+    params = ["c", "python"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        N = 100000
+        group1 = ["aaaaaaaa", "bbbbbbb", "cccccccc", "dddddddd", "eeeeeeee"]
+        df = DataFrame(np.random.choice(group1, (N, 3)), columns=list("abc"))
+        df.to_csv(self.fname, index=False)
+
+    def time_convert_post(self, engine):
+        read_csv(self.fname, engine=engine).apply(Categorical)
+
+    def time_convert_direct(self, engine):
+        read_csv(self.fname, engine=engine, dtype="category")
+
+
+class ReadCSVParseDates(StringIORewind):
+    params = ["c", "python"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        data = """{},19:00:00,18:56:00,0.8100,2.8100,7.2000,0.0000,280.0000\n
+                  {},20:00:00,19:56:00,0.0100,2.2100,7.2000,0.0000,260.0000\n
+                  {},21:00:00,20:56:00,-0.5900,2.2100,5.7000,0.0000,280.0000\n
+                  {},21:00:00,21:18:00,-0.9900,2.0100,3.6000,0.0000,270.0000\n
+                  {},22:00:00,21:56:00,-0.5900,1.7100,5.1000,0.0000,290.0000\n
+               """
+        two_cols = ["KORD,19990127"] * 5
+        data = data.format(*two_cols)
+        self.StringIO_input = StringIO(data)
+
+    def time_baseline(self, engine):
+        read_csv(
+            self.data(self.StringIO_input),
+            engine=engine,
+            sep=",",
+            header=None,
+            parse_dates=[1],
+            names=list(string.digits[:9]),
+        )
+
+
+class ReadCSVCachedParseDates(StringIORewind):
+    params = ([True, False], ["c", "python"])
+    param_names = ["do_cache", "engine"]
+
+    def setup(self, do_cache, engine):
+        data = ("\n".join([f"10/{year}" for year in range(2000, 2100)]) + "\n") * 10
+        self.StringIO_input = StringIO(data)
+
+    def time_read_csv_cached(self, do_cache, engine):
+        try:
+            read_csv(
+                self.data(self.StringIO_input),
+                engine=engine,
+                header=None,
+                parse_dates=[0],
+                cache_dates=do_cache,
+            )
+        except TypeError:
+            # cache_dates is a new keyword in 0.25
+            pass
+
+
+class ReadCSVMemoryGrowth(BaseIO):
+    chunksize = 20
+    num_rows = 1000
+    fname = "__test__.csv"
+    params = ["c", "python"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        with open(self.fname, "w", encoding="utf-8") as f:
+            for i in range(self.num_rows):
+                f.write(f"{i}\n")
+
+    def mem_parser_chunks(self, engine):
+        # see gh-24805.
+        result = read_csv(self.fname, chunksize=self.chunksize, engine=engine)
+
+        for _ in result:
+            pass
+
+
+class ReadCSVParseSpecialDate(StringIORewind):
+    params = (["mY", "mdY", "hm"], ["c", "python"])
+    param_names = ["value", "engine"]
+    objects = {
+        "mY": "01-2019\n10-2019\n02/2000\n",
+        "mdY": "12/02/2010\n",
+        "hm": "21:34\n",
+    }
+
+    def setup(self, value, engine):
+        count_elem = 10000
+        data = self.objects[value] * count_elem
+        self.StringIO_input = StringIO(data)
+
+    def time_read_special_date(self, value, engine):
+        read_csv(
+            self.data(self.StringIO_input),
+            engine=engine,
+            sep=",",
+            header=None,
+            names=["Date"],
+            parse_dates=["Date"],
+        )
+
+
+class ReadCSVMemMapUTF8:
+    fname = "__test__.csv"
+    number = 5
+
+    def setup(self):
+        lines = []
+        line_length = 128
+        start_char = " "
+        end_char = "\U00010080"
+        # This for loop creates a list of 128-char strings
+        # consisting of consecutive Unicode chars
+        for lnum in range(ord(start_char), ord(end_char), line_length):
+            line = "".join([chr(c) for c in range(lnum, lnum + 0x80)]) + "\n"
+            try:
+                line.encode("utf-8")
+            except UnicodeEncodeError:
+                # Some 16-bit words are not valid Unicode chars and must be skipped
+                continue
+            lines.append(line)
+        df = DataFrame(lines)
+        df = concat([df for n in range(100)], ignore_index=True)
+        df.to_csv(self.fname, index=False, header=False, encoding="utf-8")
+
+    def time_read_memmapped_utf8(self):
+        read_csv(self.fname, header=None, memory_map=True, encoding="utf-8", engine="c")
+
+
+class ParseDateComparison(StringIORewind):
+    params = ([False, True],)
+    param_names = ["cache_dates"]
+
+    def setup(self, cache_dates):
+        count_elem = 10000
+        data = "12-02-2010\n" * count_elem
+        self.StringIO_input = StringIO(data)
+
+    def time_read_csv_dayfirst(self, cache_dates):
+        try:
+            read_csv(
+                self.data(self.StringIO_input),
+                sep=",",
+                header=None,
+                names=["Date"],
+                parse_dates=["Date"],
+                cache_dates=cache_dates,
+                dayfirst=True,
+            )
+        except TypeError:
+            # cache_dates is a new keyword in 0.25
+            pass
+
+    def time_to_datetime_dayfirst(self, cache_dates):
+        df = read_csv(
+            self.data(self.StringIO_input), dtype={"date": str}, names=["date"]
+        )
+        to_datetime(df["date"], cache=cache_dates, dayfirst=True)
+
+    def time_to_datetime_format_DD_MM_YYYY(self, cache_dates):
+        df = read_csv(
+            self.data(self.StringIO_input), dtype={"date": str}, names=["date"]
+        )
+        to_datetime(df["date"], cache=cache_dates, format="%d-%m-%Y")
+
+
+class ReadCSVIndexCol(StringIORewind):
+    def setup(self):
+        count_elem = 100_000
+        data = "a,b\n" + "1,2\n" * count_elem
+        self.StringIO_input = StringIO(data)
+
+    def time_read_csv_index_col(self):
+        read_csv(self.data(self.StringIO_input), index_col="a")
+
+
+class ReadCSVDatePyarrowEngine(StringIORewind):
+    def setup(self):
+        count_elem = 100_000
+        data = "a\n" + "2019-12-31\n" * count_elem
+        self.StringIO_input = StringIO(data)
+
+    def time_read_csv_index_col(self):
+        read_csv(
+            self.data(self.StringIO_input),
+            parse_dates=["a"],
+            engine="pyarrow",
+            dtype_backend="pyarrow",
+        )
+
+
+class ReadCSVCParserLowMemory:
+    # GH 16798
+    def setup(self):
+        self.csv = StringIO(
+            "strings\n" + "\n".join(["x" * (1 << 20) for _ in range(2100)])
+        )
+
+    def peakmem_over_2gb_input(self):
+        read_csv(self.csv, engine="c", low_memory=False)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/excel.py b/pandas/asv_bench/benchmarks/io/excel.py
new file mode 100644
index 0000000000000..902a61be901bd
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/excel.py
@@ -0,0 +1,108 @@
+from io import BytesIO
+
+import numpy as np
+from odf.opendocument import OpenDocumentSpreadsheet
+from odf.table import (
+    Table,
+    TableCell,
+    TableRow,
+)
+from odf.text import P
+
+from pandas import (
+    DataFrame,
+    ExcelWriter,
+    Index,
+    date_range,
+    read_excel,
+)
+
+
+def _generate_dataframe():
+    N = 2000
+    C = 5
+    df = DataFrame(
+        np.random.randn(N, C),
+        columns=[f"float{i}" for i in range(C)],
+        index=date_range("20000101", periods=N, freq="h"),
+    )
+    df["object"] = Index([f"i-{i}" for i in range(N)], dtype=object)
+    return df
+
+
+class WriteExcel:
+    params = ["openpyxl", "xlsxwriter"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        self.df = _generate_dataframe()
+
+    def time_write_excel(self, engine):
+        bio = BytesIO()
+        bio.seek(0)
+        with ExcelWriter(bio, engine=engine) as writer:
+            self.df.to_excel(writer, sheet_name="Sheet1")
+
+
+class WriteExcelStyled:
+    params = ["openpyxl", "xlsxwriter"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        self.df = _generate_dataframe()
+
+    def time_write_excel_style(self, engine):
+        bio = BytesIO()
+        bio.seek(0)
+        with ExcelWriter(bio, engine=engine) as writer:
+            df_style = self.df.style
+            df_style.map(lambda x: "border: red 1px solid;")
+            df_style.map(lambda x: "color: blue")
+            df_style.map(lambda x: "border-color: green black", subset=["float1"])
+            df_style.to_excel(writer, sheet_name="Sheet1")
+
+
+class ReadExcel:
+    params = ["openpyxl", "odf"]
+    param_names = ["engine"]
+    fname_excel = "spreadsheet.xlsx"
+    fname_odf = "spreadsheet.ods"
+
+    def _create_odf(self):
+        doc = OpenDocumentSpreadsheet()
+        table = Table(name="Table1")
+        for row in self.df.values:
+            tr = TableRow()
+            for val in row:
+                tc = TableCell(valuetype="string")
+                tc.addElement(P(text=val))
+                tr.addElement(tc)
+            table.addElement(tr)
+
+        doc.spreadsheet.addElement(table)
+        doc.save(self.fname_odf)
+
+    def setup_cache(self):
+        self.df = _generate_dataframe()
+
+        self.df.to_excel(self.fname_excel, sheet_name="Sheet1")
+        self._create_odf()
+
+    def time_read_excel(self, engine):
+        if engine == "odf":
+            fname = self.fname_odf
+        else:
+            fname = self.fname_excel
+        read_excel(fname, engine=engine)
+
+
+class ReadExcelNRows(ReadExcel):
+    def time_read_excel(self, engine):
+        if engine == "odf":
+            fname = self.fname_odf
+        else:
+            fname = self.fname_excel
+        read_excel(fname, engine=engine, nrows=10)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/hdf.py b/pandas/asv_bench/benchmarks/io/hdf.py
new file mode 100644
index 0000000000000..2eb4c8c7f674b
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/hdf.py
@@ -0,0 +1,143 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    date_range,
+    read_hdf,
+)
+
+from ..pandas_vb_common import BaseIO
+
+
+class HDFStoreDataFrame(BaseIO):
+    def setup(self):
+        N = 25000
+        index = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.df = DataFrame(
+            {"float1": np.random.randn(N), "float2": np.random.randn(N)}, index=index
+        )
+        self.df_mixed = DataFrame(
+            {
+                "float1": np.random.randn(N),
+                "float2": np.random.randn(N),
+                "string1": ["foo"] * N,
+                "bool1": [True] * N,
+                "int1": np.random.randint(0, N, size=N),
+            },
+            index=index,
+        )
+        self.df_wide = DataFrame(np.random.randn(N, 100))
+        self.start_wide = self.df_wide.index[10000]
+        self.stop_wide = self.df_wide.index[15000]
+        self.df2 = DataFrame(
+            {"float1": np.random.randn(N), "float2": np.random.randn(N)},
+            index=date_range("1/1/2000", periods=N),
+        )
+        self.start = self.df2.index[10000]
+        self.stop = self.df2.index[15000]
+        self.df_wide2 = DataFrame(
+            np.random.randn(N, 100), index=date_range("1/1/2000", periods=N)
+        )
+        self.df_dc = DataFrame(
+            np.random.randn(N, 10), columns=[f"C{i:03d}" for i in range(10)]
+        )
+
+        self.fname = "__test__.h5"
+
+        self.store = HDFStore(self.fname)
+        self.store.put("fixed", self.df)
+        self.store.put("fixed_mixed", self.df_mixed)
+        self.store.append("table", self.df2)
+        self.store.append("table_mixed", self.df_mixed)
+        self.store.append("table_wide", self.df_wide)
+        self.store.append("table_wide2", self.df_wide2)
+
+    def teardown(self):
+        self.store.close()
+        self.remove(self.fname)
+
+    def time_read_store(self):
+        self.store.get("fixed")
+
+    def time_read_store_mixed(self):
+        self.store.get("fixed_mixed")
+
+    def time_write_store(self):
+        self.store.put("fixed_write", self.df)
+
+    def time_write_store_mixed(self):
+        self.store.put("fixed_mixed_write", self.df_mixed)
+
+    def time_read_store_table_mixed(self):
+        self.store.select("table_mixed")
+
+    def time_write_store_table_mixed(self):
+        self.store.append("table_mixed_write", self.df_mixed)
+
+    def time_read_store_table(self):
+        self.store.select("table")
+
+    def time_write_store_table(self):
+        self.store.append("table_write", self.df)
+
+    def time_read_store_table_wide(self):
+        self.store.select("table_wide")
+
+    def time_write_store_table_wide(self):
+        self.store.append("table_wide_write", self.df_wide)
+
+    def time_write_store_table_dc(self):
+        self.store.append("table_dc_write", self.df_dc, data_columns=True)
+
+    def time_query_store_table_wide(self):
+        self.store.select(
+            "table_wide", where="index > self.start_wide and index < self.stop_wide"
+        )
+
+    def time_query_store_table(self):
+        self.store.select("table", where="index > self.start and index < self.stop")
+
+    def time_store_repr(self):
+        repr(self.store)
+
+    def time_store_str(self):
+        str(self.store)
+
+    def time_store_info(self):
+        self.store.info()
+
+
+class HDF(BaseIO):
+    params = ["table", "fixed"]
+    param_names = ["format"]
+
+    def setup(self, format):
+        self.fname = "__test__.h5"
+        N = 100000
+        C = 5
+        self.df = DataFrame(
+            np.random.randn(N, C),
+            columns=[f"float{i}" for i in range(C)],
+            index=date_range("20000101", periods=N, freq="h"),
+        )
+        self.df["object"] = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.df.to_hdf(self.fname, key="df", format=format)
+
+        # Numeric df
+        self.df1 = self.df.copy()
+        self.df1 = self.df1.reset_index()
+        self.df1.to_hdf(self.fname, key="df1", format=format)
+
+    def time_read_hdf(self, format):
+        read_hdf(self.fname, "df")
+
+    def peakmem_read_hdf(self, format):
+        read_hdf(self.fname, "df")
+
+    def time_write_hdf(self, format):
+        self.df.to_hdf(self.fname, key="df", format=format)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/json.py b/pandas/asv_bench/benchmarks/io/json.py
new file mode 100644
index 0000000000000..bcbfcdea42dd9
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/json.py
@@ -0,0 +1,312 @@
+import sys
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    concat,
+    date_range,
+    json_normalize,
+    read_json,
+    timedelta_range,
+)
+
+from ..pandas_vb_common import BaseIO
+
+
+class ReadJSON(BaseIO):
+    fname = "__test__.json"
+    params = (["split", "index", "records"], ["int", "datetime"])
+    param_names = ["orient", "index"]
+
+    def setup(self, orient, index):
+        N = 100000
+        indexes = {
+            "int": np.arange(N),
+            "datetime": date_range("20000101", periods=N, freq="h"),
+        }
+        df = DataFrame(
+            np.random.randn(N, 5),
+            columns=[f"float_{i}" for i in range(5)],
+            index=indexes[index],
+        )
+        df.to_json(self.fname, orient=orient)
+
+    def time_read_json(self, orient, index):
+        read_json(self.fname, orient=orient)
+
+
+class ReadJSONLines(BaseIO):
+    fname = "__test_lines__.json"
+    params = ["int", "datetime"]
+    param_names = ["index"]
+
+    def setup(self, index):
+        N = 100000
+        indexes = {
+            "int": np.arange(N),
+            "datetime": date_range("20000101", periods=N, freq="h"),
+        }
+        df = DataFrame(
+            np.random.randn(N, 5),
+            columns=[f"float_{i}" for i in range(5)],
+            index=indexes[index],
+        )
+        df.to_json(self.fname, orient="records", lines=True)
+
+    def time_read_json_lines(self, index):
+        read_json(self.fname, orient="records", lines=True)
+
+    def time_read_json_lines_concat(self, index):
+        concat(read_json(self.fname, orient="records", lines=True, chunksize=25000))
+
+    def time_read_json_lines_nrows(self, index):
+        read_json(self.fname, orient="records", lines=True, nrows=25000)
+
+    def peakmem_read_json_lines(self, index):
+        read_json(self.fname, orient="records", lines=True)
+
+    def peakmem_read_json_lines_concat(self, index):
+        concat(read_json(self.fname, orient="records", lines=True, chunksize=25000))
+
+    def peakmem_read_json_lines_nrows(self, index):
+        read_json(self.fname, orient="records", lines=True, nrows=15000)
+
+
+class NormalizeJSON(BaseIO):
+    fname = "__test__.json"
+    params = [
+        ["split", "columns", "index", "values", "records"],
+        ["df", "df_date_idx", "df_td_int_ts", "df_int_floats", "df_int_float_str"],
+    ]
+    param_names = ["orient", "frame"]
+
+    def setup(self, orient, frame):
+        data = {
+            "hello": ["thisisatest", 999898, "mixed types"],
+            "nest1": {"nest2": {"nest3": "nest3_value", "nest3_int": 3445}},
+            "nest1_list": {"nest2": ["blah", 32423, 546456.876, 92030234]},
+            "hello2": "string",
+        }
+        self.data = [data for i in range(10000)]
+
+    def time_normalize_json(self, orient, frame):
+        json_normalize(self.data)
+
+
+class ToJSON(BaseIO):
+    fname = "__test__.json"
+    params = [
+        ["split", "columns", "index", "values", "records"],
+        ["df", "df_date_idx", "df_td_int_ts", "df_int_floats", "df_int_float_str"],
+    ]
+    param_names = ["orient", "frame"]
+
+    def setup(self, orient, frame):
+        N = 10**5
+        ncols = 5
+        index = date_range("20000101", periods=N, freq="h")
+        timedeltas = timedelta_range(start=1, periods=N, freq="s")
+        datetimes = date_range(start=1, periods=N, freq="s")
+        ints = np.random.randint(100000000, size=N)
+        longints = sys.maxsize * np.random.randint(100000000, size=N)
+        floats = np.random.randn(N)
+        strings = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.df = DataFrame(np.random.randn(N, ncols), index=np.arange(N))
+        self.df_date_idx = DataFrame(np.random.randn(N, ncols), index=index)
+        self.df_td_int_ts = DataFrame(
+            {
+                "td_1": timedeltas,
+                "td_2": timedeltas,
+                "int_1": ints,
+                "int_2": ints,
+                "ts_1": datetimes,
+                "ts_2": datetimes,
+            },
+            index=index,
+        )
+        self.df_int_floats = DataFrame(
+            {
+                "int_1": ints,
+                "int_2": ints,
+                "int_3": ints,
+                "float_1": floats,
+                "float_2": floats,
+                "float_3": floats,
+            },
+            index=index,
+        )
+        self.df_int_float_str = DataFrame(
+            {
+                "int_1": ints,
+                "int_2": ints,
+                "float_1": floats,
+                "float_2": floats,
+                "str_1": strings,
+                "str_2": strings,
+            },
+            index=index,
+        )
+
+        self.df_longint_float_str = DataFrame(
+            {
+                "longint_1": longints,
+                "longint_2": longints,
+                "float_1": floats,
+                "float_2": floats,
+                "str_1": strings,
+                "str_2": strings,
+            },
+            index=index,
+        )
+
+    def time_to_json(self, orient, frame):
+        getattr(self, frame).to_json(self.fname, orient=orient)
+
+    def peakmem_to_json(self, orient, frame):
+        getattr(self, frame).to_json(self.fname, orient=orient)
+
+
+class ToJSONWide(ToJSON):
+    def setup(self, orient, frame):
+        super().setup(orient, frame)
+        base_df = getattr(self, frame).copy()
+        df_wide = concat([base_df.iloc[:100]] * 1000, ignore_index=True, axis=1)
+        self.df_wide = df_wide
+
+    def time_to_json_wide(self, orient, frame):
+        self.df_wide.to_json(self.fname, orient=orient)
+
+    def peakmem_to_json_wide(self, orient, frame):
+        self.df_wide.to_json(self.fname, orient=orient)
+
+
+class ToJSONISO(BaseIO):
+    fname = "__test__.json"
+    params = [["split", "columns", "index", "values", "records"]]
+    param_names = ["orient"]
+
+    def setup(self, orient):
+        N = 10**5
+        index = date_range("20000101", periods=N, freq="h")
+        timedeltas = timedelta_range(start=1, periods=N, freq="s")
+        datetimes = date_range(start=1, periods=N, freq="s")
+        self.df = DataFrame(
+            {
+                "td_1": timedeltas,
+                "td_2": timedeltas,
+                "ts_1": datetimes,
+                "ts_2": datetimes,
+            },
+            index=index,
+        )
+
+    def time_iso_format(self, orient):
+        self.df.to_json(orient=orient, date_format="iso")
+
+
+class ToJSONLines(BaseIO):
+    fname = "__test__.json"
+
+    def setup(self):
+        N = 10**5
+        ncols = 5
+        index = date_range("20000101", periods=N, freq="h")
+        timedeltas = timedelta_range(start=1, periods=N, freq="s")
+        datetimes = date_range(start=1, periods=N, freq="s")
+        ints = np.random.randint(100000000, size=N)
+        longints = sys.maxsize * np.random.randint(100000000, size=N)
+        floats = np.random.randn(N)
+        strings = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.df = DataFrame(np.random.randn(N, ncols), index=np.arange(N))
+        self.df_date_idx = DataFrame(np.random.randn(N, ncols), index=index)
+        self.df_td_int_ts = DataFrame(
+            {
+                "td_1": timedeltas,
+                "td_2": timedeltas,
+                "int_1": ints,
+                "int_2": ints,
+                "ts_1": datetimes,
+                "ts_2": datetimes,
+            },
+            index=index,
+        )
+        self.df_int_floats = DataFrame(
+            {
+                "int_1": ints,
+                "int_2": ints,
+                "int_3": ints,
+                "float_1": floats,
+                "float_2": floats,
+                "float_3": floats,
+            },
+            index=index,
+        )
+        self.df_int_float_str = DataFrame(
+            {
+                "int_1": ints,
+                "int_2": ints,
+                "float_1": floats,
+                "float_2": floats,
+                "str_1": strings,
+                "str_2": strings,
+            },
+            index=index,
+        )
+        self.df_longint_float_str = DataFrame(
+            {
+                "longint_1": longints,
+                "longint_2": longints,
+                "float_1": floats,
+                "float_2": floats,
+                "str_1": strings,
+                "str_2": strings,
+            },
+            index=index,
+        )
+
+    def time_floats_with_int_idex_lines(self):
+        self.df.to_json(self.fname, orient="records", lines=True)
+
+    def time_floats_with_dt_index_lines(self):
+        self.df_date_idx.to_json(self.fname, orient="records", lines=True)
+
+    def time_delta_int_tstamp_lines(self):
+        self.df_td_int_ts.to_json(self.fname, orient="records", lines=True)
+
+    def time_float_int_lines(self):
+        self.df_int_floats.to_json(self.fname, orient="records", lines=True)
+
+    def time_float_int_str_lines(self):
+        self.df_int_float_str.to_json(self.fname, orient="records", lines=True)
+
+    def time_float_longint_str_lines(self):
+        self.df_longint_float_str.to_json(self.fname, orient="records", lines=True)
+
+
+class ToJSONMem:
+    def setup_cache(self):
+        df = DataFrame([[1]])
+        df2 = DataFrame(range(8), date_range("1/1/2000", periods=8, freq="min"))
+        frames = {"int": df, "float": df.astype(float), "datetime": df2}
+
+        return frames
+
+    def peakmem_int(self, frames):
+        df = frames["int"]
+        for _ in range(100_000):
+            df.to_json()
+
+    def peakmem_float(self, frames):
+        df = frames["float"]
+        for _ in range(100_000):
+            df.to_json()
+
+    def peakmem_time(self, frames):
+        df = frames["datetime"]
+        for _ in range(10_000):
+            df.to_json(orient="table")
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/parsers.py b/pandas/asv_bench/benchmarks/io/parsers.py
new file mode 100644
index 0000000000000..d3fd5075a4707
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/parsers.py
@@ -0,0 +1,17 @@
+try:
+    from pandas._libs.tslibs.parsing import _does_string_look_like_datetime
+except ImportError:
+    # Avoid whole benchmark suite import failure on asv (currently 0.4)
+    pass
+
+
+class DoesStringLookLikeDatetime:
+    params = (["2Q2005", "0.0", "10000"],)
+    param_names = ["value"]
+
+    def setup(self, value):
+        self.objects = [value] * 1000000
+
+    def time_check_datetimes(self, value):
+        for obj in self.objects:
+            _does_string_look_like_datetime(obj)
diff --git a/pandas/asv_bench/benchmarks/io/pickle.py b/pandas/asv_bench/benchmarks/io/pickle.py
new file mode 100644
index 0000000000000..4787b57b54756
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/pickle.py
@@ -0,0 +1,39 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+    read_pickle,
+)
+
+from ..pandas_vb_common import BaseIO
+
+
+class Pickle(BaseIO):
+    def setup(self):
+        self.fname = "__test__.pkl"
+        N = 100000
+        C = 5
+        self.df = DataFrame(
+            np.random.randn(N, C),
+            columns=[f"float{i}" for i in range(C)],
+            index=date_range("20000101", periods=N, freq="h"),
+        )
+        self.df["object"] = Index([f"i-{i}" for i in range(N)], dtype=object)
+        self.df.to_pickle(self.fname)
+
+    def time_read_pickle(self):
+        read_pickle(self.fname)
+
+    def time_write_pickle(self):
+        self.df.to_pickle(self.fname)
+
+    def peakmem_read_pickle(self):
+        read_pickle(self.fname)
+
+    def peakmem_write_pickle(self):
+        self.df.to_pickle(self.fname)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/sas.py b/pandas/asv_bench/benchmarks/io/sas.py
new file mode 100644
index 0000000000000..411e5b6099f76
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/sas.py
@@ -0,0 +1,23 @@
+from pathlib import Path
+
+from pandas import read_sas
+
+ROOT = Path(__file__).parents[3] / "pandas" / "tests" / "io" / "sas" / "data"
+
+
+class SAS:
+    def time_read_sas7bdat(self):
+        read_sas(ROOT / "test1.sas7bdat")
+
+    def time_read_xpt(self):
+        read_sas(ROOT / "paxraw_d_short.xpt")
+
+    def time_read_sas7bdat_2(self):
+        next(read_sas(ROOT / "0x00controlbyte.sas7bdat.bz2", chunksize=11000))
+
+    def time_read_sas7bdat_2_chunked(self):
+        for i, _ in enumerate(
+            read_sas(ROOT / "0x00controlbyte.sas7bdat.bz2", chunksize=1000)
+        ):
+            if i == 10:
+                break
diff --git a/pandas/asv_bench/benchmarks/io/sql.py b/pandas/asv_bench/benchmarks/io/sql.py
new file mode 100644
index 0000000000000..e87cc4aaa80c7
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/sql.py
@@ -0,0 +1,173 @@
+import sqlite3
+
+import numpy as np
+from sqlalchemy import create_engine
+
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+    read_sql_query,
+    read_sql_table,
+)
+
+
+class SQL:
+    params = ["sqlalchemy", "sqlite"]
+    param_names = ["connection"]
+
+    def setup(self, connection):
+        N = 10000
+        con = {
+            "sqlalchemy": create_engine("sqlite:///:memory:"),
+            "sqlite": sqlite3.connect(":memory:"),
+        }
+        self.table_name = "test_type"
+        self.query_all = f"SELECT * FROM {self.table_name}"
+        self.con = con[connection]
+        self.df = DataFrame(
+            {
+                "float": np.random.randn(N),
+                "float_with_nan": np.random.randn(N),
+                "string": ["foo"] * N,
+                "bool": [True] * N,
+                "int": np.random.randint(0, N, size=N),
+                "datetime": date_range("2000-01-01", periods=N, freq="s"),
+            },
+            index=Index([f"i-{i}" for i in range(N)], dtype=object),
+        )
+        self.df.iloc[1000:3000, 1] = np.nan
+        self.df["date"] = self.df["datetime"].dt.date
+        self.df["time"] = self.df["datetime"].dt.time
+        self.df["datetime_string"] = self.df["datetime"].astype(str)
+        self.df.to_sql(self.table_name, self.con, if_exists="replace")
+
+    def time_to_sql_dataframe(self, connection):
+        self.df.to_sql("test1", self.con, if_exists="replace")
+
+    def time_read_sql_query(self, connection):
+        read_sql_query(self.query_all, self.con)
+
+
+class WriteSQLDtypes:
+    params = (
+        ["sqlalchemy", "sqlite"],
+        [
+            "float",
+            "float_with_nan",
+            "string",
+            "bool",
+            "int",
+            "date",
+            "time",
+            "datetime",
+        ],
+    )
+    param_names = ["connection", "dtype"]
+
+    def setup(self, connection, dtype):
+        N = 10000
+        con = {
+            "sqlalchemy": create_engine("sqlite:///:memory:"),
+            "sqlite": sqlite3.connect(":memory:"),
+        }
+        self.table_name = "test_type"
+        self.query_col = f"SELECT {dtype} FROM {self.table_name}"
+        self.con = con[connection]
+        self.df = DataFrame(
+            {
+                "float": np.random.randn(N),
+                "float_with_nan": np.random.randn(N),
+                "string": ["foo"] * N,
+                "bool": [True] * N,
+                "int": np.random.randint(0, N, size=N),
+                "datetime": date_range("2000-01-01", periods=N, freq="s"),
+            },
+            index=Index([f"i-{i}" for i in range(N)], dtype=object),
+        )
+        self.df.iloc[1000:3000, 1] = np.nan
+        self.df["date"] = self.df["datetime"].dt.date
+        self.df["time"] = self.df["datetime"].dt.time
+        self.df["datetime_string"] = self.df["datetime"].astype(str)
+        self.df.to_sql(self.table_name, self.con, if_exists="replace")
+
+    def time_to_sql_dataframe_column(self, connection, dtype):
+        self.df[[dtype]].to_sql("test1", self.con, if_exists="replace")
+
+    def time_read_sql_query_select_column(self, connection, dtype):
+        read_sql_query(self.query_col, self.con)
+
+
+class ReadSQLTable:
+    def setup(self):
+        N = 10000
+        self.table_name = "test"
+        self.con = create_engine("sqlite:///:memory:")
+        self.df = DataFrame(
+            {
+                "float": np.random.randn(N),
+                "float_with_nan": np.random.randn(N),
+                "string": ["foo"] * N,
+                "bool": [True] * N,
+                "int": np.random.randint(0, N, size=N),
+                "datetime": date_range("2000-01-01", periods=N, freq="s"),
+            },
+            index=Index([f"i-{i}" for i in range(N)], dtype=object),
+        )
+        self.df.iloc[1000:3000, 1] = np.nan
+        self.df["date"] = self.df["datetime"].dt.date
+        self.df["time"] = self.df["datetime"].dt.time
+        self.df["datetime_string"] = self.df["datetime"].astype(str)
+        self.df.to_sql(self.table_name, self.con, if_exists="replace")
+
+    def time_read_sql_table_all(self):
+        read_sql_table(self.table_name, self.con)
+
+    def time_read_sql_table_parse_dates(self):
+        read_sql_table(
+            self.table_name,
+            self.con,
+            columns=["datetime_string"],
+            parse_dates=["datetime_string"],
+        )
+
+
+class ReadSQLTableDtypes:
+    params = [
+        "float",
+        "float_with_nan",
+        "string",
+        "bool",
+        "int",
+        "date",
+        "time",
+        "datetime",
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10000
+        self.table_name = "test"
+        self.con = create_engine("sqlite:///:memory:")
+        self.df = DataFrame(
+            {
+                "float": np.random.randn(N),
+                "float_with_nan": np.random.randn(N),
+                "string": ["foo"] * N,
+                "bool": [True] * N,
+                "int": np.random.randint(0, N, size=N),
+                "datetime": date_range("2000-01-01", periods=N, freq="s"),
+            },
+            index=Index([f"i-{i}" for i in range(N)], dtype=object),
+        )
+        self.df.iloc[1000:3000, 1] = np.nan
+        self.df["date"] = self.df["datetime"].dt.date
+        self.df["time"] = self.df["datetime"].dt.time
+        self.df["datetime_string"] = self.df["datetime"].astype(str)
+        self.df.to_sql(self.table_name, self.con, if_exists="replace")
+
+    def time_read_sql_table_column(self, dtype):
+        read_sql_table(self.table_name, self.con, columns=[dtype])
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/stata.py b/pandas/asv_bench/benchmarks/io/stata.py
new file mode 100644
index 0000000000000..ff33ededdfed9
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/stata.py
@@ -0,0 +1,57 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+    read_stata,
+)
+
+from ..pandas_vb_common import BaseIO
+
+
+class Stata(BaseIO):
+    params = ["tc", "td", "tm", "tw", "th", "tq", "ty"]
+    param_names = ["convert_dates"]
+
+    def setup(self, convert_dates):
+        self.fname = "__test__.dta"
+        N = self.N = 100000
+        C = self.C = 5
+        self.df = DataFrame(
+            np.random.randn(N, C),
+            columns=[f"float{i}" for i in range(C)],
+            index=date_range("20000101", periods=N, freq="h"),
+        )
+        self.df["object"] = Index([f"i-{i}" for i in range(self.N)], dtype=object)
+        self.df["int8_"] = np.random.randint(
+            np.iinfo(np.int8).min, np.iinfo(np.int8).max - 27, N
+        )
+        self.df["int16_"] = np.random.randint(
+            np.iinfo(np.int16).min, np.iinfo(np.int16).max - 27, N
+        )
+        self.df["int32_"] = np.random.randint(
+            np.iinfo(np.int32).min, np.iinfo(np.int32).max - 27, N
+        )
+        self.df["float32_"] = np.array(np.random.randn(N), dtype=np.float32)
+        self.convert_dates = {"index": convert_dates}
+        self.df.to_stata(self.fname, convert_dates=self.convert_dates)
+
+    def time_read_stata(self, convert_dates):
+        read_stata(self.fname)
+
+    def time_write_stata(self, convert_dates):
+        self.df.to_stata(self.fname, convert_dates=self.convert_dates)
+
+
+class StataMissing(Stata):
+    def setup(self, convert_dates):
+        super().setup(convert_dates)
+        for i in range(10):
+            missing_data = np.random.randn(self.N)
+            missing_data[missing_data < 0] = np.nan
+            self.df[f"missing_{i}"] = missing_data
+        self.df.to_stata(self.fname, convert_dates=self.convert_dates)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/io/style.py b/pandas/asv_bench/benchmarks/io/style.py
new file mode 100644
index 0000000000000..0486cabb29845
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/io/style.py
@@ -0,0 +1,93 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    IndexSlice,
+)
+
+
+class Render:
+    params = [[12, 24, 36], [12, 120]]
+    param_names = ["cols", "rows"]
+
+    def setup(self, cols, rows):
+        self.df = DataFrame(
+            np.random.randn(rows, cols),
+            columns=[f"float_{i + 1}" for i in range(cols)],
+            index=[f"row_{i + 1}" for i in range(rows)],
+        )
+
+    def time_apply_render(self, cols, rows):
+        self._style_apply()
+        self.st._render_html(True, True)
+
+    def peakmem_apply_render(self, cols, rows):
+        self._style_apply()
+        self.st._render_html(True, True)
+
+    def time_classes_render(self, cols, rows):
+        self._style_classes()
+        self.st._render_html(True, True)
+
+    def peakmem_classes_render(self, cols, rows):
+        self._style_classes()
+        self.st._render_html(True, True)
+
+    def time_tooltips_render(self, cols, rows):
+        self._style_tooltips()
+        self.st._render_html(True, True)
+
+    def peakmem_tooltips_render(self, cols, rows):
+        self._style_tooltips()
+        self.st._render_html(True, True)
+
+    def time_format_render(self, cols, rows):
+        self._style_format()
+        self.st._render_html(True, True)
+
+    def peakmem_format_render(self, cols, rows):
+        self._style_format()
+        self.st._render_html(True, True)
+
+    def time_apply_format_hide_render(self, cols, rows):
+        self._style_apply_format_hide()
+        self.st._render_html(True, True)
+
+    def peakmem_apply_format_hide_render(self, cols, rows):
+        self._style_apply_format_hide()
+        self.st._render_html(True, True)
+
+    def _style_apply(self):
+        def _apply_func(s):
+            return [
+                "background-color: lightcyan" if s.name == "row_1" else "" for v in s
+            ]
+
+        self.st = self.df.style.apply(_apply_func, axis=1)
+
+    def _style_classes(self):
+        classes = self.df.map(lambda v: ("cls-1" if v > 0 else ""))
+        classes.index, classes.columns = self.df.index, self.df.columns
+        self.st = self.df.style.set_td_classes(classes)
+
+    def _style_format(self):
+        ic = int(len(self.df.columns) / 4 * 3)
+        ir = int(len(self.df.index) / 4 * 3)
+        # apply a formatting function
+        # subset is flexible but hinders vectorised solutions
+        self.st = self.df.style.format(
+            "{:,.3f}",
+            subset=IndexSlice["row_1" : f"row_{ir}", "float_1" : f"float_{ic}"],
+        )
+
+    def _style_apply_format_hide(self):
+        self.st = self.df.style.map(lambda v: "color: red;")
+        self.st.format("{:.3f}")
+        self.st.hide(self.st.index[1:], axis=0)
+        self.st.hide(self.st.columns[1:], axis=1)
+
+    def _style_tooltips(self):
+        ttips = DataFrame("abc", index=self.df.index[::2], columns=self.df.columns[::2])
+        self.st = self.df.style.set_tooltips(ttips)
+        self.st.hide(self.st.index[12:], axis=0)
+        self.st.hide(self.st.columns[12:], axis=1)
diff --git a/pandas/asv_bench/benchmarks/join_merge.py b/pandas/asv_bench/benchmarks/join_merge.py
new file mode 100644
index 0000000000000..a6c6990892d38
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/join_merge.py
@@ -0,0 +1,590 @@
+import string
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    array,
+    concat,
+    date_range,
+    merge,
+    merge_asof,
+)
+
+try:
+    from pandas import merge_ordered
+except ImportError:
+    from pandas import ordered_merge as merge_ordered
+
+
+class Concat:
+    params = [0, 1]
+    param_names = ["axis"]
+
+    def setup(self, axis):
+        N = 1000
+        s = Series(N, index=Index([f"i-{i}" for i in range(N)], dtype=object))
+        self.series = [s[i:-i] for i in range(1, 10)] * 50
+        self.small_frames = [DataFrame(np.random.randn(5, 4))] * 1000
+        df = DataFrame(
+            {"A": range(N)}, index=date_range("20130101", periods=N, freq="s")
+        )
+        self.empty_left = [DataFrame(), df]
+        self.empty_right = [df, DataFrame()]
+        self.mixed_ndims = [df, df.head(N // 2)]
+
+    def time_concat_series(self, axis):
+        concat(self.series, axis=axis, sort=False)
+
+    def time_concat_small_frames(self, axis):
+        concat(self.small_frames, axis=axis)
+
+    def time_concat_empty_right(self, axis):
+        concat(self.empty_right, axis=axis)
+
+    def time_concat_empty_left(self, axis):
+        concat(self.empty_left, axis=axis)
+
+    def time_concat_mixed_ndims(self, axis):
+        concat(self.mixed_ndims, axis=axis)
+
+
+class ConcatDataFrames:
+    params = ([0, 1], [True, False])
+    param_names = ["axis", "ignore_index"]
+
+    def setup(self, axis, ignore_index):
+        frame_c = DataFrame(np.zeros((10000, 200), dtype=np.float32, order="C"))
+        self.frame_c = [frame_c] * 20
+        frame_f = DataFrame(np.zeros((10000, 200), dtype=np.float32, order="F"))
+        self.frame_f = [frame_f] * 20
+
+    def time_c_ordered(self, axis, ignore_index):
+        concat(self.frame_c, axis=axis, ignore_index=ignore_index)
+
+    def time_f_ordered(self, axis, ignore_index):
+        concat(self.frame_f, axis=axis, ignore_index=ignore_index)
+
+
+class ConcatIndexDtype:
+    params = (
+        [
+            "datetime64[ns]",
+            "int64",
+            "Int64",
+            "int64[pyarrow]",
+            "string[python]",
+            "string[pyarrow]",
+        ],
+        ["monotonic", "non_monotonic", "has_na"],
+        [0, 1],
+        [True, False],
+    )
+    param_names = ["dtype", "structure", "axis", "sort"]
+
+    def setup(self, dtype, structure, axis, sort):
+        N = 10_000
+        if dtype == "datetime64[ns]":
+            vals = date_range("1970-01-01", periods=N)
+        elif dtype in ("int64", "Int64", "int64[pyarrow]"):
+            vals = np.arange(N, dtype=np.int64)
+        elif dtype in ("string[python]", "string[pyarrow]"):
+            vals = Index([f"i-{i}" for i in range(N)], dtype=object)
+        else:
+            raise NotImplementedError
+
+        idx = Index(vals, dtype=dtype)
+
+        if structure == "monotonic":
+            idx = idx.sort_values()
+        elif structure == "non_monotonic":
+            idx = idx[::-1]
+        elif structure == "has_na":
+            if not idx._can_hold_na:
+                raise NotImplementedError
+            idx = Index([None], dtype=dtype).append(idx)
+        else:
+            raise NotImplementedError
+
+        self.series = [Series(i, idx[:-i]) for i in range(1, 6)]
+
+    def time_concat_series(self, dtype, structure, axis, sort):
+        concat(self.series, axis=axis, sort=sort)
+
+
+class Join:
+    params = [True, False]
+    param_names = ["sort"]
+
+    def setup(self, sort):
+        level1 = Index([f"i-{i}" for i in range(10)], dtype=object).values
+        level2 = Index([f"i-{i}" for i in range(1000)], dtype=object).values
+        codes1 = np.arange(10).repeat(1000)
+        codes2 = np.tile(np.arange(1000), 10)
+        index2 = MultiIndex(levels=[level1, level2], codes=[codes1, codes2])
+        self.df_multi = DataFrame(
+            np.random.randn(len(index2), 4), index=index2, columns=["A", "B", "C", "D"]
+        )
+
+        self.key1 = np.tile(level1.take(codes1), 10)
+        self.key2 = np.tile(level2.take(codes2), 10)
+        self.df = DataFrame(
+            {
+                "data1": np.random.randn(100000),
+                "data2": np.random.randn(100000),
+                "key1": self.key1,
+                "key2": self.key2,
+            }
+        )
+
+        self.df_key1 = DataFrame(
+            np.random.randn(len(level1), 4), index=level1, columns=["A", "B", "C", "D"]
+        )
+        self.df_key2 = DataFrame(
+            np.random.randn(len(level2), 4), index=level2, columns=["A", "B", "C", "D"]
+        )
+
+        shuf = np.arange(100000)
+        np.random.shuffle(shuf)
+        self.df_shuf = self.df.reindex(self.df.index[shuf])
+
+    def time_join_dataframe_index_multi(self, sort):
+        self.df.join(self.df_multi, on=["key1", "key2"], sort=sort)
+
+    def time_join_dataframe_index_single_key_bigger(self, sort):
+        self.df.join(self.df_key2, on="key2", sort=sort)
+
+    def time_join_dataframe_index_single_key_small(self, sort):
+        self.df.join(self.df_key1, on="key1", sort=sort)
+
+    def time_join_dataframe_index_shuffle_key_bigger_sort(self, sort):
+        self.df_shuf.join(self.df_key2, on="key2", sort=sort)
+
+    def time_join_dataframes_cross(self, sort):
+        self.df.loc[:2000].join(self.df_key1, how="cross", sort=sort)
+
+
+class JoinIndex:
+    def setup(self):
+        N = 5000
+        self.left = DataFrame(
+            np.random.randint(1, N / 50, (N, 2)), columns=["jim", "joe"]
+        )
+        self.right = DataFrame(
+            np.random.randint(1, N / 50, (N, 2)), columns=["jolie", "jolia"]
+        ).set_index("jolie")
+
+    def time_left_outer_join_index(self):
+        self.left.join(self.right, on="jim")
+
+
+class JoinMultiindexSubset:
+    def setup(self):
+        N = 100_000
+        mi1 = MultiIndex.from_arrays([np.arange(N)] * 4, names=["a", "b", "c", "d"])
+        mi2 = MultiIndex.from_arrays([np.arange(N)] * 2, names=["a", "b"])
+        self.left = DataFrame({"col1": 1}, index=mi1)
+        self.right = DataFrame({"col2": 2}, index=mi2)
+
+    def time_join_multiindex_subset(self):
+        self.left.join(self.right)
+
+
+class JoinEmpty:
+    def setup(self):
+        N = 100_000
+        self.df = DataFrame({"A": np.arange(N)})
+        self.df_empty = DataFrame(columns=["B", "C"], dtype="int64")
+
+    def time_inner_join_left_empty(self):
+        self.df_empty.join(self.df, how="inner")
+
+    def time_inner_join_right_empty(self):
+        self.df.join(self.df_empty, how="inner")
+
+
+class JoinNonUnique:
+    # outer join of non-unique
+    # GH 6329
+    def setup(self):
+        date_index = date_range("01-Jan-2013", "23-Jan-2013", freq="min")
+        daily_dates = date_index.to_period("D").to_timestamp("s", "s")
+        self.fracofday = date_index.values - daily_dates.values
+        self.fracofday = self.fracofday.astype("timedelta64[ns]")
+        self.fracofday = self.fracofday.astype(np.float64) / 86_400_000_000_000
+        self.fracofday = Series(self.fracofday, daily_dates)
+        index = date_range(date_index.min(), date_index.max(), freq="D")
+        self.temp = Series(1.0, index)[self.fracofday.index]
+
+    def time_join_non_unique_equal(self):
+        self.fracofday * self.temp
+
+
+class Merge:
+    params = [True, False]
+    param_names = ["sort"]
+
+    def setup(self, sort):
+        N = 10000
+        indices = Index([f"i-{i}" for i in range(N)], dtype=object).values
+        indices2 = Index([f"i-{i}" for i in range(N)], dtype=object).values
+        key = np.tile(indices[:8000], 10)
+        key2 = np.tile(indices2[:8000], 10)
+        self.left = DataFrame(
+            {"key": key, "key2": key2, "value": np.random.randn(80000)}
+        )
+        self.right = DataFrame(
+            {
+                "key": indices[2000:],
+                "key2": indices2[2000:],
+                "value2": np.random.randn(8000),
+            }
+        )
+
+        self.df = DataFrame(
+            {
+                "key1": np.tile(np.arange(500).repeat(10), 2),
+                "key2": np.tile(np.arange(250).repeat(10), 4),
+                "value": np.random.randn(10000),
+            }
+        )
+        self.df2 = DataFrame({"key1": np.arange(500), "value2": np.random.randn(500)})
+        self.df3 = self.df[:5000]
+
+    def time_merge_2intkey(self, sort):
+        merge(self.left, self.right, sort=sort)
+
+    def time_merge_dataframe_integer_2key(self, sort):
+        merge(self.df, self.df3, sort=sort)
+
+    def time_merge_dataframe_integer_key(self, sort):
+        merge(self.df, self.df2, on="key1", sort=sort)
+
+    def time_merge_dataframe_empty_right(self, sort):
+        merge(self.left, self.right.iloc[:0], sort=sort)
+
+    def time_merge_dataframe_empty_left(self, sort):
+        merge(self.left.iloc[:0], self.right, sort=sort)
+
+    def time_merge_dataframes_cross(self, sort):
+        merge(self.left.loc[:2000], self.right.loc[:2000], how="cross", sort=sort)
+
+
+class MergeEA:
+    params = [
+        [
+            "Int64",
+            "Int32",
+            "Int16",
+            "UInt64",
+            "UInt32",
+            "UInt16",
+            "Float64",
+            "Float32",
+        ],
+        [True, False],
+    ]
+    param_names = ["dtype", "monotonic"]
+
+    def setup(self, dtype, monotonic):
+        N = 10_000
+        indices = np.arange(1, N)
+        key = np.tile(indices[:8000], 10)
+        self.left = DataFrame(
+            {"key": Series(key, dtype=dtype), "value": np.random.randn(80000)}
+        )
+        self.right = DataFrame(
+            {
+                "key": Series(indices[2000:], dtype=dtype),
+                "value2": np.random.randn(7999),
+            }
+        )
+        if monotonic:
+            self.left = self.left.sort_values("key")
+            self.right = self.right.sort_values("key")
+
+    def time_merge(self, dtype, monotonic):
+        merge(self.left, self.right)
+
+
+class I8Merge:
+    params = ["inner", "outer", "left", "right"]
+    param_names = ["how"]
+
+    def setup(self, how):
+        low, high, n = -1000, 1000, 10**6
+        self.left = DataFrame(
+            np.random.randint(low, high, (n, 7)), columns=list("ABCDEFG")
+        )
+        self.left["left"] = self.left.sum(axis=1)
+        self.right = self.left.sample(frac=1).rename({"left": "right"}, axis=1)
+        self.right = self.right.reset_index(drop=True)
+        self.right["right"] *= -1
+
+    def time_i8merge(self, how):
+        merge(self.left, self.right, how=how)
+
+
+class UniqueMerge:
+    params = [4_000_000, 1_000_000]
+    param_names = ["unique_elements"]
+
+    def setup(self, unique_elements):
+        N = 1_000_000
+        self.left = DataFrame({"a": np.random.randint(1, unique_elements, (N,))})
+        self.right = DataFrame({"a": np.random.randint(1, unique_elements, (N,))})
+        uniques = self.right.a.drop_duplicates()
+        self.right["a"] = concat(
+            [uniques, Series(np.arange(0, -(N - len(uniques)), -1))], ignore_index=True
+        )
+
+    def time_unique_merge(self, unique_elements):
+        merge(self.left, self.right, how="inner")
+
+
+class MergeDatetime:
+    params = [
+        [
+            ("ns", "ns"),
+            ("ms", "ms"),
+            ("ns", "ms"),
+        ],
+        [None, "Europe/Brussels"],
+        [True, False],
+    ]
+    param_names = ["units", "tz", "monotonic"]
+
+    def setup(self, units, tz, monotonic):
+        unit_left, unit_right = units
+        N = 10_000
+        keys = Series(date_range("2012-01-01", freq="min", periods=N, tz=tz))
+        self.left = DataFrame(
+            {
+                "key": keys.sample(N * 10, replace=True).dt.as_unit(unit_left),
+                "value1": np.random.randn(N * 10),
+            }
+        )
+        self.right = DataFrame(
+            {
+                "key": keys[:8000].dt.as_unit(unit_right),
+                "value2": np.random.randn(8000),
+            }
+        )
+        if monotonic:
+            self.left = self.left.sort_values("key")
+            self.right = self.right.sort_values("key")
+
+    def time_merge(self, units, tz, monotonic):
+        merge(self.left, self.right)
+
+
+class MergeCategoricals:
+    def setup(self):
+        self.left_object = DataFrame(
+            {
+                "X": np.random.choice(range(10), size=(10000,)),
+                "Y": np.random.choice(["one", "two", "three"], size=(10000,)),
+            }
+        )
+
+        self.right_object = DataFrame(
+            {
+                "X": np.random.choice(range(10), size=(10000,)),
+                "Z": np.random.choice(["jjj", "kkk", "sss"], size=(10000,)),
+            }
+        )
+
+        self.left_cat = self.left_object.assign(
+            Y=self.left_object["Y"].astype("category")
+        )
+        self.right_cat = self.right_object.assign(
+            Z=self.right_object["Z"].astype("category")
+        )
+
+        self.left_cat_col = self.left_object.astype({"X": "category"})
+        self.right_cat_col = self.right_object.astype({"X": "category"})
+
+        self.left_cat_idx = self.left_cat_col.set_index("X")
+        self.right_cat_idx = self.right_cat_col.set_index("X")
+
+    def time_merge_object(self):
+        merge(self.left_object, self.right_object, on="X")
+
+    def time_merge_cat(self):
+        merge(self.left_cat, self.right_cat, on="X")
+
+    def time_merge_on_cat_col(self):
+        merge(self.left_cat_col, self.right_cat_col, on="X")
+
+    def time_merge_on_cat_idx(self):
+        merge(self.left_cat_idx, self.right_cat_idx, on="X")
+
+
+class MergeOrdered:
+    def setup(self):
+        groups = Index([f"i-{i}" for i in range(10)], dtype=object).values
+        self.left = DataFrame(
+            {
+                "group": groups.repeat(5000),
+                "key": np.tile(np.arange(0, 10000, 2), 10),
+                "lvalue": np.random.randn(50000),
+            }
+        )
+        self.right = DataFrame(
+            {"key": np.arange(10000), "rvalue": np.random.randn(10000)}
+        )
+
+    def time_merge_ordered(self):
+        merge_ordered(self.left, self.right, on="key", left_by="group")
+
+
+class MergeAsof:
+    params = [["backward", "forward", "nearest"], [None, 5]]
+    param_names = ["direction", "tolerance"]
+
+    def setup(self, direction, tolerance):
+        one_count = 200000
+        two_count = 1000000
+
+        df1 = DataFrame(
+            {
+                "time": np.random.randint(0, one_count / 20, one_count),
+                "key": np.random.choice(list(string.ascii_uppercase), one_count),
+                "key2": np.random.randint(0, 25, one_count),
+                "value1": np.random.randn(one_count),
+            }
+        )
+        df2 = DataFrame(
+            {
+                "time": np.random.randint(0, two_count / 20, two_count),
+                "key": np.random.choice(list(string.ascii_uppercase), two_count),
+                "key2": np.random.randint(0, 25, two_count),
+                "value2": np.random.randn(two_count),
+            }
+        )
+
+        df1 = df1.sort_values("time")
+        df2 = df2.sort_values("time")
+
+        df1["time32"] = np.int32(df1.time)
+        df2["time32"] = np.int32(df2.time)
+
+        df1["timeu64"] = np.uint64(df1.time)
+        df2["timeu64"] = np.uint64(df2.time)
+
+        self.df1a = df1[["time", "value1"]]
+        self.df2a = df2[["time", "value2"]]
+        self.df1b = df1[["time", "key", "value1"]]
+        self.df2b = df2[["time", "key", "value2"]]
+        self.df1c = df1[["time", "key2", "value1"]]
+        self.df2c = df2[["time", "key2", "value2"]]
+        self.df1d = df1[["time32", "value1"]]
+        self.df2d = df2[["time32", "value2"]]
+        self.df1e = df1[["time", "key", "key2", "value1"]]
+        self.df2e = df2[["time", "key", "key2", "value2"]]
+        self.df1f = df1[["timeu64", "value1"]]
+        self.df2f = df2[["timeu64", "value2"]]
+
+    def time_on_int(self, direction, tolerance):
+        merge_asof(
+            self.df1a, self.df2a, on="time", direction=direction, tolerance=tolerance
+        )
+
+    def time_on_int32(self, direction, tolerance):
+        merge_asof(
+            self.df1d, self.df2d, on="time32", direction=direction, tolerance=tolerance
+        )
+
+    def time_on_uint64(self, direction, tolerance):
+        merge_asof(
+            self.df1f, self.df2f, on="timeu64", direction=direction, tolerance=tolerance
+        )
+
+    def time_by_object(self, direction, tolerance):
+        merge_asof(
+            self.df1b,
+            self.df2b,
+            on="time",
+            by="key",
+            direction=direction,
+            tolerance=tolerance,
+        )
+
+    def time_by_int(self, direction, tolerance):
+        merge_asof(
+            self.df1c,
+            self.df2c,
+            on="time",
+            by="key2",
+            direction=direction,
+            tolerance=tolerance,
+        )
+
+    def time_multiby(self, direction, tolerance):
+        merge_asof(
+            self.df1e,
+            self.df2e,
+            on="time",
+            by=["key", "key2"],
+            direction=direction,
+            tolerance=tolerance,
+        )
+
+
+class MergeMultiIndex:
+    params = [
+        [
+            ("int64", "int64"),
+            ("datetime64[ns]", "int64"),
+            ("Int64", "Int64"),
+        ],
+        ["left", "right", "inner", "outer"],
+    ]
+    param_names = ["dtypes", "how"]
+
+    def setup(self, dtypes, how):
+        n = 100_000
+        offset = 50_000
+        mi1 = MultiIndex.from_arrays(
+            [
+                array(np.arange(n), dtype=dtypes[0]),
+                array(np.arange(n), dtype=dtypes[1]),
+            ]
+        )
+        mi2 = MultiIndex.from_arrays(
+            [
+                array(np.arange(offset, n + offset), dtype=dtypes[0]),
+                array(np.arange(offset, n + offset), dtype=dtypes[1]),
+            ]
+        )
+        self.df1 = DataFrame({"col1": 1}, index=mi1)
+        self.df2 = DataFrame({"col2": 2}, index=mi2)
+
+    def time_merge_sorted_multiindex(self, dtypes, how):
+        # copy to avoid MultiIndex._values caching
+        df1 = self.df1.copy()
+        df2 = self.df2.copy()
+        merge(df1, df2, how=how, left_index=True, right_index=True)
+
+
+class Align:
+    def setup(self):
+        size = 5 * 10**5
+        rng = np.arange(0, 10**13, 10**7)
+        stamps = np.datetime64("now").view("i8") + rng
+        idx1 = np.sort(np.random.choice(stamps, size, replace=False))
+        idx2 = np.sort(np.random.choice(stamps, size, replace=False))
+        self.ts1 = Series(np.random.randn(size), idx1)
+        self.ts2 = Series(np.random.randn(size), idx2)
+
+    def time_series_align_int64_index(self):
+        self.ts1 + self.ts2
+
+    def time_series_align_left_monotonic(self):
+        self.ts1.align(self.ts2, join="left")
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/libs.py b/pandas/asv_bench/benchmarks/libs.py
new file mode 100644
index 0000000000000..7da2d27d98dbb
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/libs.py
@@ -0,0 +1,105 @@
+"""
+Benchmarks for code in pandas/_libs, excluding pandas/_libs/tslibs,
+which has its own directory.
+
+If a PR does not edit anything in _libs/, then it is unlikely that the
+benchmarks will be affected.
+"""
+
+import numpy as np
+
+from pandas._libs.lib import (
+    infer_dtype,
+    is_list_like,
+    is_scalar,
+)
+
+from pandas import (
+    NA,
+    Index,
+    NaT,
+)
+
+from .pandas_vb_common import lib
+
+try:
+    from pandas.util import cache_readonly
+except ImportError:
+    from pandas.util.decorators import cache_readonly
+
+
+# TODO: share with something in pd._testing?
+scalars = [
+    0,
+    1.0,
+    1 + 2j,
+    True,
+    "foo",
+    b"bar",
+    None,
+    np.datetime64(123, "ns"),
+    np.timedelta64(123, "ns"),
+    NaT,
+    NA,
+]
+zero_dims = [np.array("123")]
+listlikes = [np.array([1, 2, 3]), {0: 1}, {1, 2, 3}, [1, 2, 3], (1, 2, 3)]
+
+
+class ScalarListLike:
+    params = scalars + zero_dims + listlikes
+
+    def time_is_list_like(self, param):
+        is_list_like(param)
+
+    def time_is_scalar(self, param):
+        is_scalar(param)
+
+
+class FastZip:
+    def setup(self):
+        N = 10000
+        K = 10
+        key1 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
+        key2 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
+        col_array = np.vstack([key1, key2, np.random.randn(N * K)])
+        col_array2 = col_array.copy()
+        col_array2[:, :10000] = np.nan
+        self.col_array_list = list(col_array)
+
+    def time_lib_fast_zip(self):
+        lib.fast_zip(self.col_array_list)
+
+
+class InferDtype:
+    param_names = ["dtype"]
+    data_dict = {
+        "np-object": np.array([1] * 100000, dtype="O"),
+        "py-object": [1] * 100000,
+        "np-null": np.array([1] * 50000 + [np.nan] * 50000),
+        "py-null": [1] * 50000 + [None] * 50000,
+        "np-int": np.array([1] * 100000, dtype=int),
+        "np-floating": np.array([1.0] * 100000, dtype=float),
+        "empty": [],
+        "bytes": [b"a"] * 100000,
+    }
+    params = list(data_dict.keys())
+
+    def time_infer_dtype_skipna(self, dtype):
+        infer_dtype(self.data_dict[dtype], skipna=True)
+
+    def time_infer_dtype(self, dtype):
+        infer_dtype(self.data_dict[dtype], skipna=False)
+
+
+class CacheReadonly:
+    def setup(self):
+        class Foo:
+            @cache_readonly
+            def prop(self):
+                return 5
+
+        self.obj = Foo()
+
+    def time_cache_readonly(self):
+        self.obj.prop
diff --git a/pandas/asv_bench/benchmarks/multiindex_object.py b/pandas/asv_bench/benchmarks/multiindex_object.py
new file mode 100644
index 0000000000000..0a588c9a2e22e
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/multiindex_object.py
@@ -0,0 +1,434 @@
+import string
+
+import numpy as np
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    array,
+    date_range,
+)
+
+
+class GetLoc:
+    def setup(self):
+        self.mi_large = MultiIndex.from_product(
+            [np.arange(1000), np.arange(20), list(string.ascii_letters)],
+            names=["one", "two", "three"],
+        )
+        self.mi_med = MultiIndex.from_product(
+            [np.arange(1000), np.arange(10), list("A")], names=["one", "two", "three"]
+        )
+        self.mi_small = MultiIndex.from_product(
+            [np.arange(100), list("A"), list("A")], names=["one", "two", "three"]
+        )
+
+    def time_large_get_loc(self):
+        self.mi_large.get_loc((999, 19, "Z"))
+
+    def time_large_get_loc_warm(self):
+        for _ in range(1000):
+            self.mi_large.get_loc((999, 19, "Z"))
+
+    def time_med_get_loc(self):
+        self.mi_med.get_loc((999, 9, "A"))
+
+    def time_med_get_loc_warm(self):
+        for _ in range(1000):
+            self.mi_med.get_loc((999, 9, "A"))
+
+    def time_string_get_loc(self):
+        self.mi_small.get_loc((99, "A", "A"))
+
+    def time_small_get_loc_warm(self):
+        for _ in range(1000):
+            self.mi_small.get_loc((99, "A", "A"))
+
+
+class GetLocs:
+    def setup(self):
+        self.mi_large = MultiIndex.from_product(
+            [np.arange(1000), np.arange(20), list(string.ascii_letters)],
+            names=["one", "two", "three"],
+        )
+        self.mi_med = MultiIndex.from_product(
+            [np.arange(1000), np.arange(10), list("A")], names=["one", "two", "three"]
+        )
+        self.mi_small = MultiIndex.from_product(
+            [np.arange(100), list("A"), list("A")], names=["one", "two", "three"]
+        )
+
+    def time_large_get_locs(self):
+        self.mi_large.get_locs([999, 19, "Z"])
+
+    def time_med_get_locs(self):
+        self.mi_med.get_locs([999, 9, "A"])
+
+    def time_small_get_locs(self):
+        self.mi_small.get_locs([99, "A", "A"])
+
+
+class Duplicates:
+    def setup(self):
+        size = 65536
+        arrays = [np.random.randint(0, 8192, size), np.random.randint(0, 1024, size)]
+        mask = np.random.rand(size) < 0.1
+        self.mi_unused_levels = MultiIndex.from_arrays(arrays)
+        self.mi_unused_levels = self.mi_unused_levels[mask]
+
+    def time_remove_unused_levels(self):
+        self.mi_unused_levels.remove_unused_levels()
+
+
+class Integer:
+    def setup(self):
+        self.mi_int = MultiIndex.from_product(
+            [np.arange(1000), np.arange(1000)], names=["one", "two"]
+        )
+        self.obj_index = np.array(
+            [
+                (0, 10),
+                (0, 11),
+                (0, 12),
+                (0, 13),
+                (0, 14),
+                (0, 15),
+                (0, 16),
+                (0, 17),
+                (0, 18),
+                (0, 19),
+            ],
+            dtype=object,
+        )
+        self.other_mi_many_mismatches = MultiIndex.from_tuples(
+            [
+                (-7, 41),
+                (-2, 3),
+                (-0.7, 5),
+                (0, 0),
+                (0, 1.5),
+                (0, 340),
+                (0, 1001),
+                (1, -4),
+                (1, 20),
+                (1, 1040),
+                (432, -5),
+                (432, 17),
+                (439, 165.5),
+                (998, -4),
+                (998, 24065),
+                (999, 865.2),
+                (999, 1000),
+                (1045, -843),
+            ]
+        )
+
+    def time_get_indexer(self):
+        self.mi_int.get_indexer(self.obj_index)
+
+    def time_get_indexer_and_backfill(self):
+        self.mi_int.get_indexer(self.other_mi_many_mismatches, method="backfill")
+
+    def time_get_indexer_and_pad(self):
+        self.mi_int.get_indexer(self.other_mi_many_mismatches, method="pad")
+
+    def time_is_monotonic(self):
+        self.mi_int.is_monotonic_increasing
+
+
+class Duplicated:
+    def setup(self):
+        n, k = 200, 5000
+        levels = [
+            np.arange(n),
+            Index([f"i-{i}" for i in range(n)], dtype=object).values,
+            1000 + np.arange(n),
+        ]
+        codes = [np.random.choice(n, (k * n)) for lev in levels]
+        self.mi = MultiIndex(levels=levels, codes=codes)
+
+    def time_duplicated(self):
+        self.mi.duplicated()
+
+
+class Sortlevel:
+    def setup(self):
+        n = 1182720
+        low, high = -4096, 4096
+        arrs = [
+            np.repeat(np.random.randint(low, high, (n // k)), k)
+            for k in [11, 7, 5, 3, 1]
+        ]
+        self.mi_int = MultiIndex.from_arrays(arrs)[np.random.permutation(n)]
+
+        a = np.repeat(np.arange(100), 1000)
+        b = np.tile(np.arange(1000), 100)
+        self.mi = MultiIndex.from_arrays([a, b])
+        self.mi = self.mi.take(np.random.permutation(np.arange(100000)))
+
+    def time_sortlevel_int64(self):
+        self.mi_int.sortlevel()
+
+    def time_sortlevel_zero(self):
+        self.mi.sortlevel(0)
+
+    def time_sortlevel_one(self):
+        self.mi.sortlevel(1)
+
+
+class SortValues:
+    params = ["int64", "Int64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        a = array(np.tile(np.arange(100), 1000), dtype=dtype)
+        b = array(np.tile(np.arange(1000), 100), dtype=dtype)
+        self.mi = MultiIndex.from_arrays([a, b])
+
+    def time_sort_values(self, dtype):
+        self.mi.sort_values()
+
+
+class Values:
+    def setup_cache(self):
+        level1 = range(1000)
+        level2 = date_range(start="1/1/2012", periods=100)
+        mi = MultiIndex.from_product([level1, level2])
+        return mi
+
+    def time_datetime_level_values_copy(self, mi):
+        mi.copy().values
+
+    def time_datetime_level_values_sliced(self, mi):
+        mi[:10].values
+
+
+class CategoricalLevel:
+    def setup(self):
+        self.df = DataFrame(
+            {
+                "a": np.arange(1_000_000, dtype=np.int32),
+                "b": np.arange(1_000_000, dtype=np.int64),
+                "c": np.arange(1_000_000, dtype=float),
+            }
+        ).astype({"a": "category", "b": "category"})
+
+    def time_categorical_level(self):
+        self.df.set_index(["a", "b"])
+
+
+class Equals:
+    def setup(self):
+        self.mi = MultiIndex.from_product(
+            [
+                date_range("2000-01-01", periods=1000),
+                RangeIndex(1000),
+            ]
+        )
+        self.mi_deepcopy = self.mi.copy(deep=True)
+        self.idx_non_object = RangeIndex(1)
+
+    def time_equals_deepcopy(self):
+        self.mi.equals(self.mi_deepcopy)
+
+    def time_equals_non_object_index(self):
+        self.mi.equals(self.idx_non_object)
+
+
+class SetOperations:
+    params = [
+        ("monotonic", "non_monotonic"),
+        ("datetime", "int", "string", "ea_int"),
+        ("intersection", "union", "symmetric_difference"),
+        (False, None),
+    ]
+    param_names = ["index_structure", "dtype", "method", "sort"]
+
+    def setup(self, index_structure, dtype, method, sort):
+        N = 10**5
+        level1 = range(1000)
+
+        level2 = date_range(start="1/1/2000", periods=N // 1000)
+        dates_left = MultiIndex.from_product([level1, level2])
+
+        level2 = range(N // 1000)
+        int_left = MultiIndex.from_product([level1, level2])
+
+        level2 = Index([f"i-{i}" for i in range(N // 1000)], dtype=object).values
+        str_left = MultiIndex.from_product([level1, level2])
+
+        level2 = range(N // 1000)
+        ea_int_left = MultiIndex.from_product([level1, Series(level2, dtype="Int64")])
+
+        data = {
+            "datetime": dates_left,
+            "int": int_left,
+            "string": str_left,
+            "ea_int": ea_int_left,
+        }
+
+        if index_structure == "non_monotonic":
+            data = {k: mi[::-1] for k, mi in data.items()}
+
+        data = {k: {"left": mi, "right": mi[:-1]} for k, mi in data.items()}
+        self.left = data[dtype]["left"]
+        self.right = data[dtype]["right"]
+
+    def time_operation(self, index_structure, dtype, method, sort):
+        getattr(self.left, method)(self.right, sort=sort)
+
+
+class Difference:
+    params = [
+        ("datetime", "int", "string", "ea_int"),
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**4 * 2
+        level1 = range(1000)
+
+        level2 = date_range(start="1/1/2000", periods=N // 1000)
+        dates_left = MultiIndex.from_product([level1, level2])
+
+        level2 = range(N // 1000)
+        int_left = MultiIndex.from_product([level1, level2])
+
+        level2 = Series(range(N // 1000), dtype="Int64")
+        level2[0] = NA
+        ea_int_left = MultiIndex.from_product([level1, level2])
+
+        level2 = Index([f"i-{i}" for i in range(N // 1000)], dtype=object).values
+        str_left = MultiIndex.from_product([level1, level2])
+
+        data = {
+            "datetime": dates_left,
+            "int": int_left,
+            "ea_int": ea_int_left,
+            "string": str_left,
+        }
+
+        data = {k: {"left": mi, "right": mi[:5]} for k, mi in data.items()}
+        self.left = data[dtype]["left"]
+        self.right = data[dtype]["right"]
+
+    def time_difference(self, dtype):
+        self.left.difference(self.right)
+
+
+class Unique:
+    params = [
+        (("Int64", NA), ("int64", 0)),
+    ]
+    param_names = ["dtype_val"]
+
+    def setup(self, dtype_val):
+        level = Series(
+            [1, 2, dtype_val[1], dtype_val[1]] + list(range(1_000_000)),
+            dtype=dtype_val[0],
+        )
+        self.midx = MultiIndex.from_arrays([level, level])
+
+        level_dups = Series(
+            [1, 2, dtype_val[1], dtype_val[1]] + list(range(500_000)) * 2,
+            dtype=dtype_val[0],
+        )
+
+        self.midx_dups = MultiIndex.from_arrays([level_dups, level_dups])
+
+    def time_unique(self, dtype_val):
+        self.midx.unique()
+
+    def time_unique_dups(self, dtype_val):
+        self.midx_dups.unique()
+
+
+class Isin:
+    params = [
+        ("string", "int", "datetime"),
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**5
+        level1 = range(1000)
+
+        level2 = date_range(start="1/1/2000", periods=N // 1000)
+        dates_midx = MultiIndex.from_product([level1, level2])
+
+        level2 = range(N // 1000)
+        int_midx = MultiIndex.from_product([level1, level2])
+
+        level2 = Index([f"i-{i}" for i in range(N // 1000)], dtype=object).values
+        str_midx = MultiIndex.from_product([level1, level2])
+
+        data = {
+            "datetime": dates_midx,
+            "int": int_midx,
+            "string": str_midx,
+        }
+
+        self.midx = data[dtype]
+        self.values_small = self.midx[:100]
+        self.values_large = self.midx[100:]
+
+    def time_isin_small(self, dtype):
+        self.midx.isin(self.values_small)
+
+    def time_isin_large(self, dtype):
+        self.midx.isin(self.values_large)
+
+
+class Putmask:
+    def setup(self):
+        N = 10**5
+        level1 = range(1_000)
+
+        level2 = date_range(start="1/1/2000", periods=N // 1000)
+        self.midx = MultiIndex.from_product([level1, level2])
+
+        level1 = range(1_000, 2_000)
+        self.midx_values = MultiIndex.from_product([level1, level2])
+
+        level2 = date_range(start="1/1/2010", periods=N // 1000)
+        self.midx_values_different = MultiIndex.from_product([level1, level2])
+        self.mask = np.array([True, False] * (N // 2))
+
+    def time_putmask(self):
+        self.midx.putmask(self.mask, self.midx_values)
+
+    def time_putmask_all_different(self):
+        self.midx.putmask(self.mask, self.midx_values_different)
+
+
+class Append:
+    params = ["datetime64[ns]", "int64", "string"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N1 = 1000
+        N2 = 500
+        left_level1 = range(N1)
+        right_level1 = range(N1, N1 + N1)
+
+        if dtype == "datetime64[ns]":
+            level2 = date_range(start="2000-01-01", periods=N2)
+        elif dtype == "int64":
+            level2 = range(N2)
+        elif dtype == "string":
+            level2 = Index([f"i-{i}" for i in range(N2)], dtype=object)
+        else:
+            raise NotImplementedError
+
+        self.left = MultiIndex.from_product([left_level1, level2])
+        self.right = MultiIndex.from_product([right_level1, level2])
+
+    def time_append(self, dtype):
+        self.left.append(self.right)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/package.py b/pandas/asv_bench/benchmarks/package.py
new file mode 100644
index 0000000000000..f8b51a523dab8
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/package.py
@@ -0,0 +1,20 @@
+"""
+Benchmarks for pandas at the package-level.
+"""
+
+import subprocess
+import sys
+
+
+class TimeImport:
+    def time_import(self):
+        # on py37+ we the "-X importtime" usage gives us a more precise
+        #  measurement of the import time we actually care about,
+        #  without the subprocess or interpreter overhead
+        cmd = [sys.executable, "-X", "importtime", "-c", "import pandas as pd"]
+        p = subprocess.run(cmd, stderr=subprocess.PIPE, check=True)
+
+        line = p.stderr.splitlines()[-1]
+        field = line.split(b"|")[-2].strip()
+        total = int(field)  # microseconds
+        return total
diff --git a/pandas/asv_bench/benchmarks/pandas_vb_common.py b/pandas/asv_bench/benchmarks/pandas_vb_common.py
new file mode 100644
index 0000000000000..4bd56ccb1b5ce
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/pandas_vb_common.py
@@ -0,0 +1,80 @@
+from importlib import import_module
+import os
+
+import numpy as np
+
+import pandas as pd
+
+# Compatibility import for lib
+for imp in ["pandas._libs.lib", "pandas.lib"]:
+    try:
+        lib = import_module(imp)
+        break
+    except (ImportError, TypeError, ValueError):
+        pass
+
+# Compatibility import for the testing module
+try:
+    import pandas._testing as tm
+except ImportError:
+    import pandas.util.testing as tm  # noqa: F401
+
+
+numeric_dtypes = [
+    np.int64,
+    np.int32,
+    np.uint32,
+    np.uint64,
+    np.float32,
+    np.float64,
+    np.int16,
+    np.int8,
+    np.uint16,
+    np.uint8,
+]
+datetime_dtypes = [np.datetime64, np.timedelta64]
+string_dtypes = [object]
+try:
+    extension_dtypes = [
+        pd.Int8Dtype,
+        pd.Int16Dtype,
+        pd.Int32Dtype,
+        pd.Int64Dtype,
+        pd.UInt8Dtype,
+        pd.UInt16Dtype,
+        pd.UInt32Dtype,
+        pd.UInt64Dtype,
+        pd.CategoricalDtype,
+        pd.IntervalDtype,
+        pd.DatetimeTZDtype("ns", "UTC"),
+        pd.PeriodDtype("D"),
+    ]
+except AttributeError:
+    extension_dtypes = []
+
+
+def setup(*args, **kwargs):
+    # This function just needs to be imported into each benchmark file to
+    # set up the random seed before each function.
+    # https://asv.readthedocs.io/en/latest/writing_benchmarks.html
+    np.random.seed(1234)
+
+
+class BaseIO:
+    """
+    Base class for IO benchmarks
+    """
+
+    fname = None
+
+    def remove(self, f):
+        """Remove created files"""
+        try:
+            os.remove(f)
+        except OSError:
+            # On Windows, attempting to remove a file that is in use
+            # causes an exception to be raised
+            pass
+
+    def teardown(self, *args, **kwargs):
+        self.remove(self.fname)
diff --git a/pandas/asv_bench/benchmarks/period.py b/pandas/asv_bench/benchmarks/period.py
new file mode 100644
index 0000000000000..3b8b60e790380
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/period.py
@@ -0,0 +1,107 @@
+"""
+Period benchmarks with non-tslibs dependencies.  See
+benchmarks.tslibs.period for benchmarks that rely only on tslibs.
+"""
+
+from pandas import (
+    DataFrame,
+    Period,
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+)
+
+from pandas.tseries.frequencies import to_offset
+
+
+class PeriodIndexConstructor:
+    params = [["D"], [True, False]]
+    param_names = ["freq", "is_offset"]
+
+    def setup(self, freq, is_offset):
+        self.rng = date_range("1985", periods=1000)
+        self.rng2 = date_range("1985", periods=1000).to_pydatetime()
+        self.ints = list(range(2000, 3000))
+        self.daily_ints = (
+            date_range("1/1/2000", periods=1000, freq=freq).strftime("%Y%m%d").map(int)
+        )
+        if is_offset:
+            self.freq = to_offset(freq)
+        else:
+            self.freq = freq
+
+    def time_from_date_range(self, freq, is_offset):
+        PeriodIndex(self.rng, freq=freq)
+
+    def time_from_pydatetime(self, freq, is_offset):
+        PeriodIndex(self.rng2, freq=freq)
+
+    def time_from_ints(self, freq, is_offset):
+        PeriodIndex(self.ints, freq=freq)
+
+    def time_from_ints_daily(self, freq, is_offset):
+        PeriodIndex(self.daily_ints, freq=freq)
+
+
+class DataFramePeriodColumn:
+    def setup(self):
+        self.rng = period_range(start="1/1/1990", freq="s", periods=20000)
+        self.df = DataFrame(index=range(len(self.rng)))
+
+    def time_setitem_period_column(self):
+        self.df["col"] = self.rng
+
+    def time_set_index(self):
+        # GH#21582 limited by comparisons of Period objects
+        self.df["col2"] = self.rng
+        self.df.set_index("col2", append=True)
+
+
+class Algorithms:
+    params = ["index", "series"]
+    param_names = ["typ"]
+
+    def setup(self, typ):
+        data = [
+            Period("2011-01", freq="M"),
+            Period("2011-02", freq="M"),
+            Period("2011-03", freq="M"),
+            Period("2011-04", freq="M"),
+        ]
+
+        if typ == "index":
+            self.vector = PeriodIndex(data * 1000, freq="M")
+        elif typ == "series":
+            self.vector = Series(data * 1000)
+
+    def time_drop_duplicates(self, typ):
+        self.vector.drop_duplicates()
+
+    def time_value_counts(self, typ):
+        self.vector.value_counts()
+
+
+class Indexing:
+    def setup(self):
+        self.index = period_range(start="1985", periods=1000, freq="D")
+        self.series = Series(range(1000), index=self.index)
+        self.period = self.index[500]
+
+    def time_get_loc(self):
+        self.index.get_loc(self.period)
+
+    def time_shallow_copy(self):
+        self.index._view()
+
+    def time_series_loc(self):
+        self.series.loc[self.period]
+
+    def time_align(self):
+        DataFrame({"a": self.series, "b": self.series[:500]})
+
+    def time_intersection(self):
+        self.index[:750].intersection(self.index[250:])
+
+    def time_unique(self):
+        self.index.unique()
diff --git a/pandas/asv_bench/benchmarks/plotting.py b/pandas/asv_bench/benchmarks/plotting.py
new file mode 100644
index 0000000000000..789bb8d8533b1
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/plotting.py
@@ -0,0 +1,164 @@
+import contextlib
+import importlib.machinery
+import importlib.util
+import os
+import pathlib
+import sys
+import tempfile
+from unittest import mock
+
+import matplotlib
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    date_range,
+)
+
+try:
+    from pandas.plotting import andrews_curves
+except ImportError:
+    from pandas.tools.plotting import andrews_curves
+
+from pandas.plotting._core import _get_plot_backend
+
+matplotlib.use("Agg")
+
+
+class SeriesPlotting:
+    params = [["line", "bar", "area", "barh", "hist", "kde", "pie"]]
+    param_names = ["kind"]
+
+    def setup(self, kind):
+        if kind in ["bar", "barh", "pie"]:
+            n = 100
+        elif kind in ["kde"]:
+            n = 10000
+        else:
+            n = 1000000
+
+        self.s = Series(np.random.randn(n))
+        if kind in ["area", "pie"]:
+            self.s = self.s.abs()
+
+    def time_series_plot(self, kind):
+        self.s.plot(kind=kind)
+
+
+class FramePlotting:
+    params = [
+        ["line", "bar", "area", "barh", "hist", "kde", "pie", "scatter", "hexbin"]
+    ]
+    param_names = ["kind"]
+
+    def setup(self, kind):
+        if kind in ["bar", "barh", "pie"]:
+            n = 100
+        elif kind in ["kde", "scatter", "hexbin"]:
+            n = 10000
+        else:
+            n = 1000000
+
+        self.x = Series(np.random.randn(n))
+        self.y = Series(np.random.randn(n))
+        if kind in ["area", "pie"]:
+            self.x = self.x.abs()
+            self.y = self.y.abs()
+        self.df = DataFrame({"x": self.x, "y": self.y})
+
+    def time_frame_plot(self, kind):
+        self.df.plot(x="x", y="y", kind=kind)
+
+
+class TimeseriesPlotting:
+    def setup(self):
+        N = 2000
+        M = 5
+        idx = date_range("1/1/1975", periods=N)
+        self.df = DataFrame(np.random.randn(N, M), index=idx)
+
+        idx_irregular = DatetimeIndex(
+            np.concatenate((idx.values[0:10], idx.values[12:]))
+        )
+        self.df2 = DataFrame(
+            np.random.randn(len(idx_irregular), M), index=idx_irregular
+        )
+
+    def time_plot_regular(self):
+        self.df.plot()
+
+    def time_plot_regular_compat(self):
+        self.df.plot(x_compat=True)
+
+    def time_plot_irregular(self):
+        self.df2.plot()
+
+    def time_plot_table(self):
+        self.df.plot(table=True)
+
+
+class Misc:
+    def setup(self):
+        N = 500
+        M = 10
+        self.df = DataFrame(np.random.randn(N, M))
+        self.df["Name"] = ["A"] * N
+
+    def time_plot_andrews_curves(self):
+        andrews_curves(self.df, "Name")
+
+
+class BackendLoading:
+    repeat = 1
+    number = 1
+    warmup_time = 0
+
+    def setup(self):
+        mod = importlib.util.module_from_spec(
+            importlib.machinery.ModuleSpec("pandas_dummy_backend", None)
+        )
+        mod.plot = lambda *args, **kwargs: 1
+
+        with contextlib.ExitStack() as stack:
+            stack.enter_context(
+                mock.patch.dict(sys.modules, {"pandas_dummy_backend": mod})
+            )
+            tmp_path = pathlib.Path(stack.enter_context(tempfile.TemporaryDirectory()))
+
+            sys.path.insert(0, os.fsdecode(tmp_path))
+            stack.callback(sys.path.remove, os.fsdecode(tmp_path))
+
+            dist_info = tmp_path / "my_backend-0.0.0.dist-info"
+            dist_info.mkdir()
+            (dist_info / "entry_points.txt").write_bytes(
+                b"[pandas_plotting_backends]\n"
+                b"my_ep_backend = pandas_dummy_backend\n"
+                b"my_ep_backend0 = pandas_dummy_backend\n"
+                b"my_ep_backend1 = pandas_dummy_backend\n"
+                b"my_ep_backend2 = pandas_dummy_backend\n"
+                b"my_ep_backend3 = pandas_dummy_backend\n"
+                b"my_ep_backend4 = pandas_dummy_backend\n"
+                b"my_ep_backend5 = pandas_dummy_backend\n"
+                b"my_ep_backend6 = pandas_dummy_backend\n"
+                b"my_ep_backend7 = pandas_dummy_backend\n"
+                b"my_ep_backend8 = pandas_dummy_backend\n"
+                b"my_ep_backend9 = pandas_dummy_backend\n"
+            )
+            self.stack = stack.pop_all()
+
+    def teardown(self):
+        self.stack.close()
+
+    def time_get_plot_backend(self):
+        # finds the first my_ep_backend
+        _get_plot_backend("my_ep_backend")
+
+    def time_get_plot_backend_fallback(self):
+        # iterates through all the my_ep_backend[0-9] before falling back
+        # to importlib.import_module
+        _get_plot_backend("pandas_dummy_backend")
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/reindex.py b/pandas/asv_bench/benchmarks/reindex.py
new file mode 100644
index 0000000000000..3d22bfce7e2b2
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/reindex.py
@@ -0,0 +1,148 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    period_range,
+)
+
+
+class Reindex:
+    def setup(self):
+        rng = date_range(start="1/1/1970", periods=10000, freq="1min")
+        self.df = DataFrame(np.random.rand(10000, 10), index=rng, columns=range(10))
+        self.df["foo"] = "bar"
+        self.rng_subset = Index(rng[::2])
+        self.df2 = DataFrame(
+            index=range(10000), data=np.random.rand(10000, 30), columns=range(30)
+        )
+        N = 5000
+        K = 200
+        level1 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
+        level2 = np.tile(Index([f"i-{i}" for i in range(K)], dtype=object).values, N)
+        index = MultiIndex.from_arrays([level1, level2])
+        self.s = Series(np.random.randn(N * K), index=index)
+        self.s_subset = self.s[::2]
+        self.s_subset_no_cache = self.s[::2].copy()
+
+        mi = MultiIndex.from_product([rng, range(100)])
+        self.s2 = Series(np.random.randn(len(mi)), index=mi)
+        self.s2_subset = self.s2[::2].copy()
+
+    def time_reindex_dates(self):
+        self.df.reindex(self.rng_subset)
+
+    def time_reindex_columns(self):
+        self.df2.reindex(columns=self.df.columns[1:5])
+
+    def time_reindex_multiindex_with_cache(self):
+        # MultiIndex._values gets cached
+        self.s.reindex(self.s_subset.index)
+
+    def time_reindex_multiindex_no_cache(self):
+        # Copy to avoid MultiIndex._values getting cached
+        self.s.reindex(self.s_subset_no_cache.index.copy())
+
+    def time_reindex_multiindex_no_cache_dates(self):
+        # Copy to avoid MultiIndex._values getting cached
+        self.s2_subset.reindex(self.s2.index.copy())
+
+
+class ReindexMethod:
+    params = [["pad", "backfill"], [date_range, period_range]]
+    param_names = ["method", "constructor"]
+
+    def setup(self, method, constructor):
+        N = 100000
+        self.idx = constructor("1/1/2000", periods=N, freq="1min")
+        self.ts = Series(np.random.randn(N), index=self.idx)[::2]
+
+    def time_reindex_method(self, method, constructor):
+        self.ts.reindex(self.idx, method=method)
+
+
+class LevelAlign:
+    def setup(self):
+        self.index = MultiIndex(
+            levels=[np.arange(10), np.arange(100), np.arange(100)],
+            codes=[
+                np.arange(10).repeat(10000),
+                np.tile(np.arange(100).repeat(100), 10),
+                np.tile(np.tile(np.arange(100), 100), 10),
+            ],
+        )
+        self.df = DataFrame(np.random.randn(len(self.index), 4), index=self.index)
+        self.df_level = DataFrame(np.random.randn(100, 4), index=self.index.levels[1])
+
+    def time_align_level(self):
+        self.df.align(self.df_level, level=1, copy=False)
+
+    def time_reindex_level(self):
+        self.df_level.reindex(self.index, level=1)
+
+
+class DropDuplicates:
+    params = [True, False]
+    param_names = ["inplace"]
+
+    def setup(self, inplace):
+        N = 10000
+        K = 10
+        key1 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
+        key2 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
+        self.df = DataFrame(
+            {"key1": key1, "key2": key2, "value": np.random.randn(N * K)}
+        )
+        self.df_nan = self.df.copy()
+        self.df_nan.iloc[:10000, :] = np.nan
+
+        self.s = Series(np.random.randint(0, 1000, size=10000))
+        self.s_str = Series(
+            np.tile(Index([f"i-{i}" for i in range(1000)], dtype=object).values, 10)
+        )
+
+        N = 1000000
+        K = 10000
+        key1 = np.random.randint(0, K, size=N)
+        self.df_int = DataFrame({"key1": key1})
+        self.df_bool = DataFrame(np.random.randint(0, 2, size=(K, 10), dtype=bool))
+
+    def time_frame_drop_dups(self, inplace):
+        self.df.drop_duplicates(["key1", "key2"], inplace=inplace)
+
+    def time_frame_drop_dups_na(self, inplace):
+        self.df_nan.drop_duplicates(["key1", "key2"], inplace=inplace)
+
+    def time_series_drop_dups_int(self, inplace):
+        self.s.drop_duplicates(inplace=inplace)
+
+    def time_series_drop_dups_string(self, inplace):
+        self.s_str.drop_duplicates(inplace=inplace)
+
+    def time_frame_drop_dups_int(self, inplace):
+        self.df_int.drop_duplicates(inplace=inplace)
+
+    def time_frame_drop_dups_bool(self, inplace):
+        self.df_bool.drop_duplicates(inplace=inplace)
+
+
+class Align:
+    # blog "pandas escaped the zoo"
+    def setup(self):
+        n = 50000
+        indices = Index([f"i-{i}" for i in range(n)], dtype=object)
+        subsample_size = 40000
+        self.x = Series(np.random.randn(n), indices)
+        self.y = Series(
+            np.random.randn(subsample_size),
+            index=np.random.choice(indices, subsample_size, replace=False),
+        )
+
+    def time_align_series_irregular_string(self):
+        self.x + self.y
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/replace.py b/pandas/asv_bench/benchmarks/replace.py
new file mode 100644
index 0000000000000..a9276b7dc32ce
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/replace.py
@@ -0,0 +1,74 @@
+import numpy as np
+
+import pandas as pd
+
+
+class FillNa:
+    params = [True, False]
+    param_names = ["inplace"]
+
+    def setup(self, inplace):
+        N = 10**6
+        rng = pd.date_range("1/1/2000", periods=N, freq="min")
+        data = np.random.randn(N)
+        data[::2] = np.nan
+        self.ts = pd.Series(data, index=rng)
+
+    def time_fillna(self, inplace):
+        self.ts.fillna(0.0, inplace=inplace)
+
+    def time_replace(self, inplace):
+        self.ts.replace(np.nan, 0.0, inplace=inplace)
+
+
+class ReplaceDict:
+    params = [True, False]
+    param_names = ["inplace"]
+
+    def setup(self, inplace):
+        N = 10**5
+        start_value = 10**5
+        self.to_rep = dict(enumerate(np.arange(N) + start_value))
+        self.s = pd.Series(np.random.randint(N, size=10**3))
+
+    def time_replace_series(self, inplace):
+        self.s.replace(self.to_rep, inplace=inplace)
+
+
+class ReplaceList:
+    # GH#28099
+
+    params = [(True, False)]
+    param_names = ["inplace"]
+
+    def setup(self, inplace):
+        self.df = pd.DataFrame({"A": 0, "B": 0}, index=range(10**7))
+
+    def time_replace_list(self, inplace):
+        self.df.replace([np.inf, -np.inf], np.nan, inplace=inplace)
+
+    def time_replace_list_one_match(self, inplace):
+        # the 1 can be held in self._df.blocks[0], while the inf and -inf can't
+        self.df.replace([np.inf, -np.inf, 1], np.nan, inplace=inplace)
+
+
+class Convert:
+    params = (["DataFrame", "Series"], ["Timestamp", "Timedelta"])
+    param_names = ["constructor", "replace_data"]
+
+    def setup(self, constructor, replace_data):
+        N = 10**3
+        data = {
+            "Series": pd.Series(np.random.randint(N, size=N)),
+            "DataFrame": pd.DataFrame(
+                {"A": np.random.randint(N, size=N), "B": np.random.randint(N, size=N)}
+            ),
+        }
+        self.to_replace = {i: getattr(pd, replace_data) for i in range(N)}
+        self.data = data[constructor]
+
+    def time_replace(self, constructor, replace_data):
+        self.data.replace(self.to_replace)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/reshape.py b/pandas/asv_bench/benchmarks/reshape.py
new file mode 100644
index 0000000000000..54326a4433756
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/reshape.py
@@ -0,0 +1,338 @@
+from itertools import product
+import string
+
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    date_range,
+    melt,
+    wide_to_long,
+)
+from pandas.api.types import CategoricalDtype
+
+
+class Melt:
+    params = ["float64", "Float64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        self.df = DataFrame(
+            np.random.randn(100_000, 3), columns=["A", "B", "C"], dtype=dtype
+        )
+        self.df["id1"] = pd.Series(np.random.randint(0, 10, 10000))
+        self.df["id2"] = pd.Series(np.random.randint(100, 1000, 10000))
+
+    def time_melt_dataframe(self, dtype):
+        melt(self.df, id_vars=["id1", "id2"])
+
+
+class Pivot:
+    def setup(self):
+        N = 10000
+        index = date_range("1/1/2000", periods=N, freq="h")
+        data = {
+            "value": np.random.randn(N * 50),
+            "variable": np.arange(50).repeat(N),
+            "date": np.tile(index.values, 50),
+        }
+        self.df = DataFrame(data)
+
+    def time_reshape_pivot_time_series(self):
+        self.df.pivot(index="date", columns="variable", values="value")
+
+
+class SimpleReshape:
+    def setup(self):
+        arrays = [np.arange(100).repeat(100), np.roll(np.tile(np.arange(100), 100), 25)]
+        index = MultiIndex.from_arrays(arrays)
+        self.df = DataFrame(np.random.randn(10000, 4), index=index)
+        self.udf = self.df.unstack(1)
+
+    def time_stack(self):
+        self.udf.stack()
+
+    def time_unstack(self):
+        self.df.unstack(1)
+
+
+class ReshapeExtensionDtype:
+    params = ["datetime64[ns, US/Pacific]", "Period[s]"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        lev = pd.Index(list("ABCDEFGHIJ"))
+        ri = pd.Index(range(1000))
+        mi = MultiIndex.from_product([lev, ri], names=["foo", "bar"])
+
+        index = date_range("2016-01-01", periods=10000, freq="s", tz="US/Pacific")
+        if dtype == "Period[s]":
+            index = index.tz_localize(None).to_period("s")
+
+        ser = pd.Series(index, index=mi)
+        df = ser.unstack("bar")
+        # roundtrips -> df.stack().equals(ser)
+
+        self.ser = ser
+        self.df = df
+
+    def time_stack(self, dtype):
+        self.df.stack()
+
+    def time_unstack_fast(self, dtype):
+        # last level -> doesn't have to make copies
+        self.ser.unstack("bar")
+
+    def time_unstack_slow(self, dtype):
+        # first level -> must make copies
+        self.ser.unstack("foo")
+
+    def time_transpose(self, dtype):
+        self.df.T
+
+
+class ReshapeMaskedArrayDtype(ReshapeExtensionDtype):
+    params = ["Int64", "Float64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        lev = pd.Index(list("ABCDEFGHIJ"))
+        ri = pd.Index(range(1000))
+        mi = MultiIndex.from_product([lev, ri], names=["foo", "bar"])
+
+        values = np.random.randn(10_000).astype(int)
+
+        ser = pd.Series(values, dtype=dtype, index=mi)
+        df = ser.unstack("bar")
+        # roundtrips -> df.stack().equals(ser)
+
+        self.ser = ser
+        self.df = df
+
+
+class Unstack:
+    params = ["int", "category"]
+
+    def setup(self, dtype):
+        m = 100
+        n = 1000
+
+        levels = np.arange(m)
+        index = MultiIndex.from_product([levels] * 2)
+        columns = np.arange(n)
+        if dtype == "int":
+            values = np.arange(m * m * n).reshape(m * m, n)
+            self.df = DataFrame(values, index, columns)
+        else:
+            # the category branch is ~20x slower than int. So we
+            # cut down the size a bit. Now it's only ~3x slower.
+            n = 50
+            columns = columns[:n]
+            indices = np.random.randint(0, 52, size=(m * m, n))
+            values = np.take(list(string.ascii_letters), indices)
+            values = [pd.Categorical(v) for v in values.T]
+
+            self.df = DataFrame(dict(enumerate(values)), index, columns)
+
+        self.df2 = self.df.iloc[:-1]
+
+    def time_full_product(self, dtype):
+        self.df.unstack()
+
+    def time_without_last_row(self, dtype):
+        self.df2.unstack()
+
+
+class SparseIndex:
+    def setup(self):
+        NUM_ROWS = 1000
+        self.df = DataFrame(
+            {
+                "A": np.random.randint(50, size=NUM_ROWS),
+                "B": np.random.randint(50, size=NUM_ROWS),
+                "C": np.random.randint(-10, 10, size=NUM_ROWS),
+                "D": np.random.randint(-10, 10, size=NUM_ROWS),
+                "E": np.random.randint(10, size=NUM_ROWS),
+                "F": np.random.randn(NUM_ROWS),
+            }
+        )
+        self.df = self.df.set_index(["A", "B", "C", "D", "E"])
+
+    def time_unstack(self):
+        self.df.unstack()
+
+
+class WideToLong:
+    def setup(self):
+        nyrs = 20
+        nidvars = 20
+        N = 5000
+        self.letters = list("ABCD")
+        yrvars = [
+            letter + str(num)
+            for letter, num in product(self.letters, range(1, nyrs + 1))
+        ]
+        columns = [str(i) for i in range(nidvars)] + yrvars
+        self.df = DataFrame(np.random.randn(N, nidvars + len(yrvars)), columns=columns)
+        self.df["id"] = self.df.index
+
+    def time_wide_to_long_big(self):
+        wide_to_long(self.df, self.letters, i="id", j="year")
+
+
+class PivotTable:
+    def setup(self):
+        N = 100000
+        fac1 = np.array(["A", "B", "C"], dtype="O")
+        fac2 = np.array(["one", "two"], dtype="O")
+        ind1 = np.random.randint(0, 3, size=N)
+        ind2 = np.random.randint(0, 2, size=N)
+        self.df = DataFrame(
+            {
+                "key1": fac1.take(ind1),
+                "key2": fac2.take(ind2),
+                "key3": fac2.take(ind2),
+                "value1": np.random.randn(N),
+                "value2": np.random.randn(N),
+                "value3": np.random.randn(N),
+            }
+        )
+        self.df2 = DataFrame(
+            {"col1": list("abcde"), "col2": list("fghij"), "col3": [1, 2, 3, 4, 5]}
+        )
+        self.df2.col1 = self.df2.col1.astype("category")
+        self.df2.col2 = self.df2.col2.astype("category")
+
+    def time_pivot_table(self):
+        self.df.pivot_table(index="key1", columns=["key2", "key3"])
+
+    def time_pivot_table_agg(self):
+        self.df.pivot_table(
+            index="key1", columns=["key2", "key3"], aggfunc=["sum", "mean"]
+        )
+
+    def time_pivot_table_margins(self):
+        self.df.pivot_table(index="key1", columns=["key2", "key3"], margins=True)
+
+    def time_pivot_table_categorical(self):
+        self.df2.pivot_table(
+            index="col1", values="col3", columns="col2", aggfunc="sum", fill_value=0
+        )
+
+    def time_pivot_table_categorical_observed(self):
+        self.df2.pivot_table(
+            index="col1",
+            values="col3",
+            columns="col2",
+            aggfunc="sum",
+            fill_value=0,
+            observed=True,
+        )
+
+    def time_pivot_table_margins_only_column(self):
+        self.df.pivot_table(columns=["key1", "key2", "key3"], margins=True)
+
+
+class Crosstab:
+    def setup(self):
+        N = 100000
+        fac1 = np.array(["A", "B", "C"], dtype="O")
+        fac2 = np.array(["one", "two"], dtype="O")
+        self.ind1 = np.random.randint(0, 3, size=N)
+        self.ind2 = np.random.randint(0, 2, size=N)
+        self.vec1 = fac1.take(self.ind1)
+        self.vec2 = fac2.take(self.ind2)
+
+    def time_crosstab(self):
+        pd.crosstab(self.vec1, self.vec2)
+
+    def time_crosstab_values(self):
+        pd.crosstab(self.vec1, self.vec2, values=self.ind1, aggfunc="sum")
+
+    def time_crosstab_normalize(self):
+        pd.crosstab(self.vec1, self.vec2, normalize=True)
+
+    def time_crosstab_normalize_margins(self):
+        pd.crosstab(self.vec1, self.vec2, normalize=True, margins=True)
+
+
+class GetDummies:
+    def setup(self):
+        categories = list(string.ascii_letters[:12])
+        s = pd.Series(
+            np.random.choice(categories, size=1000000),
+            dtype=CategoricalDtype(categories),
+        )
+        self.s = s
+
+    def time_get_dummies_1d(self):
+        pd.get_dummies(self.s, sparse=False)
+
+    def time_get_dummies_1d_sparse(self):
+        pd.get_dummies(self.s, sparse=True)
+
+
+class Cut:
+    params = [[4, 10, 1000]]
+    param_names = ["bins"]
+
+    def setup(self, bins):
+        N = 10**5
+        self.int_series = pd.Series(np.arange(N).repeat(5))
+        self.float_series = pd.Series(np.random.randn(N).repeat(5))
+        self.timedelta_series = pd.Series(
+            np.random.randint(N, size=N), dtype="timedelta64[ns]"
+        )
+        self.datetime_series = pd.Series(
+            np.random.randint(N, size=N), dtype="datetime64[ns]"
+        )
+        self.interval_bins = pd.IntervalIndex.from_breaks(np.linspace(0, N, bins))
+
+    def time_cut_int(self, bins):
+        pd.cut(self.int_series, bins)
+
+    def time_cut_float(self, bins):
+        pd.cut(self.float_series, bins)
+
+    def time_cut_timedelta(self, bins):
+        pd.cut(self.timedelta_series, bins)
+
+    def time_cut_datetime(self, bins):
+        pd.cut(self.datetime_series, bins)
+
+    def time_qcut_int(self, bins):
+        pd.qcut(self.int_series, bins)
+
+    def time_qcut_float(self, bins):
+        pd.qcut(self.float_series, bins)
+
+    def time_qcut_timedelta(self, bins):
+        pd.qcut(self.timedelta_series, bins)
+
+    def time_qcut_datetime(self, bins):
+        pd.qcut(self.datetime_series, bins)
+
+    def time_cut_interval(self, bins):
+        # GH 27668
+        pd.cut(self.int_series, self.interval_bins)
+
+    def peakmem_cut_interval(self, bins):
+        # GH 27668
+        pd.cut(self.int_series, self.interval_bins)
+
+
+class Explode:
+    param_names = ["n_rows", "max_list_length"]
+    params = [[100, 1000, 10000], [3, 5, 10]]
+
+    def setup(self, n_rows, max_list_length):
+        data = [np.arange(np.random.randint(max_list_length)) for _ in range(n_rows)]
+        self.series = pd.Series(data)
+
+    def time_explode(self, n_rows, max_list_length):
+        self.series.explode()
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/rolling.py b/pandas/asv_bench/benchmarks/rolling.py
new file mode 100644
index 0000000000000..bd4da00bfd2ad
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/rolling.py
@@ -0,0 +1,370 @@
+import warnings
+
+import numpy as np
+
+import pandas as pd
+
+
+class Methods:
+    params = (
+        ["DataFrame", "Series"],
+        [("rolling", {"window": 10}), ("rolling", {"window": 1000}), ("expanding", {})],
+        ["int", "float"],
+        ["median", "mean", "max", "min", "std", "count", "skew", "kurt", "sum", "sem"],
+    )
+    param_names = ["constructor", "window_kwargs", "dtype", "method"]
+
+    def setup(self, constructor, window_kwargs, dtype, method):
+        N = 10**5
+        window, kwargs = window_kwargs
+        arr = (100 * np.random.random(N)).astype(dtype)
+        obj = getattr(pd, constructor)(arr)
+        self.window = getattr(obj, window)(**kwargs)
+
+    def time_method(self, constructor, window_kwargs, dtype, method):
+        getattr(self.window, method)()
+
+    def peakmem_method(self, constructor, window_kwargs, dtype, method):
+        getattr(self.window, method)()
+
+
+class Apply:
+    params = (
+        ["DataFrame", "Series"],
+        [3, 300],
+        ["int", "float"],
+        [sum, np.sum, lambda x: np.sum(x) + 5],
+        [True, False],
+    )
+    param_names = ["constructor", "window", "dtype", "function", "raw"]
+
+    def setup(self, constructor, window, dtype, function, raw):
+        N = 10**3
+        arr = (100 * np.random.random(N)).astype(dtype)
+        self.roll = getattr(pd, constructor)(arr).rolling(window)
+
+    def time_rolling(self, constructor, window, dtype, function, raw):
+        self.roll.apply(function, raw=raw)
+
+
+class NumbaEngineMethods:
+    params = (
+        ["DataFrame", "Series"],
+        ["int", "float"],
+        [("rolling", {"window": 10}), ("expanding", {})],
+        ["sum", "max", "min", "median", "mean", "var", "std"],
+        [True, False],
+        [None, 100],
+    )
+    param_names = [
+        "constructor",
+        "dtype",
+        "window_kwargs",
+        "method",
+        "parallel",
+        "cols",
+    ]
+
+    def setup(self, constructor, dtype, window_kwargs, method, parallel, cols):
+        N = 10**3
+        window, kwargs = window_kwargs
+        shape = (N, cols) if cols is not None and constructor != "Series" else N
+        arr = (100 * np.random.random(shape)).astype(dtype)
+        data = getattr(pd, constructor)(arr)
+
+        # Warm the cache
+        with warnings.catch_warnings(record=True):
+            # Catch parallel=True not being applicable e.g. 1D data
+            self.window = getattr(data, window)(**kwargs)
+            getattr(self.window, method)(
+                engine="numba", engine_kwargs={"parallel": parallel}
+            )
+
+    def test_method(self, constructor, dtype, window_kwargs, method, parallel, cols):
+        with warnings.catch_warnings(record=True):
+            getattr(self.window, method)(
+                engine="numba", engine_kwargs={"parallel": parallel}
+            )
+
+
+class NumbaEngineApply:
+    params = (
+        ["DataFrame", "Series"],
+        ["int", "float"],
+        [("rolling", {"window": 10}), ("expanding", {})],
+        [np.sum, lambda x: np.sum(x) + 5],
+        [True, False],
+        [None, 100],
+    )
+    param_names = [
+        "constructor",
+        "dtype",
+        "window_kwargs",
+        "function",
+        "parallel",
+        "cols",
+    ]
+
+    def setup(self, constructor, dtype, window_kwargs, function, parallel, cols):
+        N = 10**3
+        window, kwargs = window_kwargs
+        shape = (N, cols) if cols is not None and constructor != "Series" else N
+        arr = (100 * np.random.random(shape)).astype(dtype)
+        data = getattr(pd, constructor)(arr)
+
+        # Warm the cache
+        with warnings.catch_warnings(record=True):
+            # Catch parallel=True not being applicable e.g. 1D data
+            self.window = getattr(data, window)(**kwargs)
+            self.window.apply(
+                function, raw=True, engine="numba", engine_kwargs={"parallel": parallel}
+            )
+
+    def test_method(self, constructor, dtype, window_kwargs, function, parallel, cols):
+        with warnings.catch_warnings(record=True):
+            self.window.apply(
+                function, raw=True, engine="numba", engine_kwargs={"parallel": parallel}
+            )
+
+
+class EWMMethods:
+    params = (
+        ["DataFrame", "Series"],
+        [
+            ({"halflife": 10}, "mean"),
+            ({"halflife": 10}, "std"),
+            ({"halflife": 1000}, "mean"),
+            ({"halflife": 1000}, "std"),
+            (
+                {
+                    "halflife": "1 Day",
+                    "times": pd.date_range("1900", periods=10**5, freq="23s"),
+                },
+                "mean",
+            ),
+        ],
+        ["int", "float"],
+    )
+    param_names = ["constructor", "kwargs_method", "dtype"]
+
+    def setup(self, constructor, kwargs_method, dtype):
+        N = 10**5
+        kwargs, method = kwargs_method
+        arr = (100 * np.random.random(N)).astype(dtype)
+        self.method = method
+        self.ewm = getattr(pd, constructor)(arr).ewm(**kwargs)
+
+    def time_ewm(self, constructor, kwargs_method, dtype):
+        getattr(self.ewm, self.method)()
+
+
+class VariableWindowMethods(Methods):
+    params = (
+        ["DataFrame", "Series"],
+        ["50s", "1h", "1d"],
+        ["int", "float"],
+        ["median", "mean", "max", "min", "std", "count", "skew", "kurt", "sum", "sem"],
+    )
+    param_names = ["constructor", "window", "dtype", "method"]
+
+    def setup(self, constructor, window, dtype, method):
+        N = 10**5
+        arr = (100 * np.random.random(N)).astype(dtype)
+        index = pd.date_range("2017-01-01", periods=N, freq="5s")
+        self.window = getattr(pd, constructor)(arr, index=index).rolling(window)
+
+
+class Pairwise:
+    params = (
+        [({"window": 10}, "rolling"), ({"window": 1000}, "rolling"), ({}, "expanding")],
+        ["corr", "cov"],
+        [True, False],
+    )
+    param_names = ["window_kwargs", "method", "pairwise"]
+
+    def setup(self, kwargs_window, method, pairwise):
+        N = 10**4
+        n_groups = 20
+        kwargs, window = kwargs_window
+        groups = [i for _ in range(N // n_groups) for i in range(n_groups)]
+        arr = np.random.random(N)
+        self.df = pd.DataFrame(arr)
+        self.window = getattr(self.df, window)(**kwargs)
+        self.window_group = getattr(
+            pd.DataFrame({"A": groups, "B": arr}).groupby("A"), window
+        )(**kwargs)
+
+    def time_pairwise(self, kwargs_window, method, pairwise):
+        getattr(self.window, method)(self.df, pairwise=pairwise)
+
+    def time_groupby(self, kwargs_window, method, pairwise):
+        getattr(self.window_group, method)(self.df, pairwise=pairwise)
+
+
+class Quantile:
+    params = (
+        ["DataFrame", "Series"],
+        [10, 1000],
+        ["int", "float"],
+        [0, 0.5, 1],
+        ["linear", "nearest", "lower", "higher", "midpoint"],
+    )
+    param_names = ["constructor", "window", "dtype", "percentile"]
+
+    def setup(self, constructor, window, dtype, percentile, interpolation):
+        N = 10**5
+        arr = np.random.random(N).astype(dtype)
+        self.roll = getattr(pd, constructor)(arr).rolling(window)
+
+    def time_quantile(self, constructor, window, dtype, percentile, interpolation):
+        self.roll.quantile(percentile, interpolation=interpolation)
+
+
+class Rank:
+    params = (
+        ["DataFrame", "Series"],
+        [10, 1000],
+        ["int", "float"],
+        [True, False],
+        [True, False],
+        ["min", "max", "average"],
+    )
+    param_names = [
+        "constructor",
+        "window",
+        "dtype",
+        "percentile",
+        "ascending",
+        "method",
+    ]
+
+    def setup(self, constructor, window, dtype, percentile, ascending, method):
+        N = 10**5
+        arr = np.random.random(N).astype(dtype)
+        self.roll = getattr(pd, constructor)(arr).rolling(window)
+
+    def time_rank(self, constructor, window, dtype, percentile, ascending, method):
+        self.roll.rank(pct=percentile, ascending=ascending, method=method)
+
+
+class PeakMemFixedWindowMinMax:
+    params = ["min", "max"]
+
+    def setup(self, operation):
+        N = 10**6
+        arr = np.random.random(N)
+        self.roll = pd.Series(arr).rolling(2)
+
+    def peakmem_fixed(self, operation):
+        for x in range(5):
+            getattr(self.roll, operation)()
+
+
+class ForwardWindowMethods:
+    params = (
+        ["DataFrame", "Series"],
+        [10, 1000],
+        ["int", "float"],
+        ["median", "mean", "max", "min", "kurt", "sum"],
+    )
+    param_names = ["constructor", "window_size", "dtype", "method"]
+
+    def setup(self, constructor, window_size, dtype, method):
+        N = 10**5
+        arr = np.random.random(N).astype(dtype)
+        indexer = pd.api.indexers.FixedForwardWindowIndexer(window_size=window_size)
+        self.roll = getattr(pd, constructor)(arr).rolling(window=indexer)
+
+    def time_rolling(self, constructor, window_size, dtype, method):
+        getattr(self.roll, method)()
+
+    def peakmem_rolling(self, constructor, window_size, dtype, method):
+        getattr(self.roll, method)()
+
+
+class Groupby:
+    params = (
+        ["sum", "median", "mean", "max", "min", "kurt", "sum"],
+        [
+            ("rolling", {"window": 2}),
+            ("rolling", {"window": "30s"}),
+            ("expanding", {}),
+        ],
+    )
+
+    def setup(self, method, window_kwargs):
+        N = 1000
+        window, kwargs = window_kwargs
+        df = pd.DataFrame(
+            {
+                "A": [str(i) for i in range(N)] * 10,
+                "B": list(range(N)) * 10,
+            }
+        )
+        if isinstance(kwargs.get("window", None), str):
+            df.index = pd.date_range(start="1900-01-01", freq="1min", periods=N * 10)
+        self.groupby_window = getattr(df.groupby("A"), window)(**kwargs)
+
+    def time_method(self, method, window_kwargs):
+        getattr(self.groupby_window, method)()
+
+
+class GroupbyLargeGroups:
+    # https://github.com/pandas-dev/pandas/issues/38038
+    # specific example where the rolling operation on a larger dataframe
+    # is relatively cheap (few but large groups), but creation of
+    # MultiIndex of result can be expensive
+
+    def setup(self):
+        N = 100000
+        self.df = pd.DataFrame({"A": [1, 2] * (N // 2), "B": np.random.randn(N)})
+
+    def time_rolling_multiindex_creation(self):
+        self.df.groupby("A").rolling(3).mean()
+
+
+class GroupbyEWM:
+    params = ["var", "std", "cov", "corr"]
+    param_names = ["method"]
+
+    def setup(self, method):
+        df = pd.DataFrame({"A": range(50), "B": range(50)})
+        self.gb_ewm = df.groupby("A").ewm(com=1.0)
+
+    def time_groupby_method(self, method):
+        getattr(self.gb_ewm, method)()
+
+
+class GroupbyEWMEngine:
+    params = ["cython", "numba"]
+    param_names = ["engine"]
+
+    def setup(self, engine):
+        df = pd.DataFrame({"A": range(50), "B": range(50)})
+        self.gb_ewm = df.groupby("A").ewm(com=1.0)
+
+    def time_groupby_mean(self, engine):
+        self.gb_ewm.mean(engine=engine)
+
+
+def table_method_func(x):
+    return np.sum(x, axis=0) + 1
+
+
+class TableMethod:
+    params = ["single", "table"]
+    param_names = ["method"]
+
+    def setup(self, method):
+        self.df = pd.DataFrame(np.random.randn(10, 1000))
+
+    def time_apply(self, method):
+        self.df.rolling(2, method=method).apply(
+            table_method_func, raw=True, engine="numba"
+        )
+
+    def time_ewm_mean(self, method):
+        self.df.ewm(1, method=method).mean(engine="numba")
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/series_methods.py b/pandas/asv_bench/benchmarks/series_methods.py
new file mode 100644
index 0000000000000..85d34cac5a7bf
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/series_methods.py
@@ -0,0 +1,425 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas import (
+    NA,
+    Index,
+    NaT,
+    Series,
+    date_range,
+)
+
+
+class SeriesConstructor:
+    def setup(self):
+        self.idx = date_range(
+            start=datetime(2015, 10, 26), end=datetime(2016, 1, 1), freq="50s"
+        )
+        self.data = dict(zip(self.idx, range(len(self.idx))))
+        self.array = np.array([1, 2, 3])
+        self.idx2 = Index(["a", "b", "c"])
+
+    def time_constructor_dict(self):
+        Series(data=self.data, index=self.idx)
+
+    def time_constructor_no_data(self):
+        Series(data=None, index=self.idx)
+
+
+class ToFrame:
+    params = [["int64", "datetime64[ns]", "category", "Int64"], [None, "foo"]]
+    param_names = ["dtype", "name"]
+
+    def setup(self, dtype, name):
+        arr = np.arange(10**5)
+        ser = Series(arr, dtype=dtype)
+        self.ser = ser
+
+    def time_to_frame(self, dtype, name):
+        self.ser.to_frame(name)
+
+
+class NSort:
+    params = ["first", "last", "all"]
+    param_names = ["keep"]
+
+    def setup(self, keep):
+        self.s = Series(np.random.randint(1, 10, 100000))
+
+    def time_nlargest(self, keep):
+        self.s.nlargest(3, keep=keep)
+
+    def time_nsmallest(self, keep):
+        self.s.nsmallest(3, keep=keep)
+
+
+class Dropna:
+    params = ["int", "datetime"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**6
+        data = {
+            "int": np.random.randint(1, 10, N),
+            "datetime": date_range("2000-01-01", freq="s", periods=N),
+        }
+        self.s = Series(data[dtype])
+        if dtype == "datetime":
+            self.s[np.random.randint(1, N, 100)] = NaT
+
+    def time_dropna(self, dtype):
+        self.s.dropna()
+
+
+class Fillna:
+    params = [
+        [
+            "datetime64[ns]",
+            "float32",
+            "float64",
+            "Float64",
+            "Int64",
+            "int64[pyarrow]",
+            "string",
+            "string[pyarrow]",
+        ],
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**6
+        if dtype == "datetime64[ns]":
+            data = date_range("2000-01-01", freq="s", periods=N)
+            na_value = NaT
+        elif dtype in ("float64", "Float64"):
+            data = np.random.randn(N)
+            na_value = np.nan
+        elif dtype in ("Int64", "int64[pyarrow]"):
+            data = np.arange(N)
+            na_value = NA
+        elif dtype in ("string", "string[pyarrow]"):
+            data = np.array([str(i) * 5 for i in range(N)], dtype=object)
+            na_value = NA
+        else:
+            raise NotImplementedError
+        fill_value = data[0]
+        ser = Series(data, dtype=dtype)
+        ser[::2] = na_value
+        self.ser = ser
+        self.fill_value = fill_value
+
+    def time_fillna(self, dtype):
+        self.ser.fillna(value=self.fill_value)
+
+    def time_ffill(self, dtype):
+        self.ser.ffill()
+
+    def time_bfill(self, dtype):
+        self.ser.bfill()
+
+
+class SearchSorted:
+    goal_time = 0.2
+    params = [
+        "int8",
+        "int16",
+        "int32",
+        "int64",
+        "uint8",
+        "uint16",
+        "uint32",
+        "uint64",
+        "float16",
+        "float32",
+        "float64",
+        "str",
+    ]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        N = 10**5
+        data = np.array([1] * N + [2] * N + [3] * N).astype(dtype)
+        self.s = Series(data)
+
+    def time_searchsorted(self, dtype):
+        key = "2" if dtype == "str" else 2
+        self.s.searchsorted(key)
+
+
+class Map:
+    params = (
+        ["dict", "Series", "lambda"],
+        ["object", "category", "int"],
+        [None, "ignore"],
+    )
+    param_names = ["mapper", "dtype", "na_action"]
+
+    def setup(self, mapper, dtype, na_action):
+        map_size = 1000
+        map_data = Series(map_size - np.arange(map_size), dtype=dtype)
+
+        # construct mapper
+        if mapper == "Series":
+            self.map_data = map_data
+        elif mapper == "dict":
+            self.map_data = map_data.to_dict()
+        elif mapper == "lambda":
+            map_dict = map_data.to_dict()
+            self.map_data = lambda x: map_dict[x]
+        else:
+            raise NotImplementedError
+
+        self.s = Series(np.random.randint(0, map_size, 10000), dtype=dtype)
+
+    def time_map(self, mapper, dtype, na_action):
+        self.s.map(self.map_data, na_action=na_action)
+
+
+class Clip:
+    params = [50, 1000, 10**5]
+    param_names = ["n"]
+
+    def setup(self, n):
+        self.s = Series(np.random.randn(n))
+
+    def time_clip(self, n):
+        self.s.clip(0, 1)
+
+
+class ClipDt:
+    def setup(self):
+        dr = date_range("20220101", periods=100_000, freq="s", tz="UTC")
+        self.clipper_dt = dr[0:1_000].repeat(100)
+        self.s = Series(dr)
+
+    def time_clip(self):
+        self.s.clip(upper=self.clipper_dt)
+
+
+class ValueCounts:
+    params = [[10**3, 10**4, 10**5], ["int", "uint", "float", "object"]]
+    param_names = ["N", "dtype"]
+
+    def setup(self, N, dtype):
+        self.s = Series(np.random.randint(0, N, size=10 * N)).astype(dtype)
+
+    def time_value_counts(self, N, dtype):
+        self.s.value_counts()
+
+
+class ValueCountsEA:
+    params = [[10**3, 10**4, 10**5], [True, False]]
+    param_names = ["N", "dropna"]
+
+    def setup(self, N, dropna):
+        self.s = Series(np.random.randint(0, N, size=10 * N), dtype="Int64")
+        self.s.loc[1] = NA
+
+    def time_value_counts(self, N, dropna):
+        self.s.value_counts(dropna=dropna)
+
+
+class ValueCountsObjectDropNAFalse:
+    params = [10**3, 10**4, 10**5]
+    param_names = ["N"]
+
+    def setup(self, N):
+        self.s = Series(np.random.randint(0, N, size=10 * N)).astype("object")
+
+    def time_value_counts(self, N):
+        self.s.value_counts(dropna=False)
+
+
+class Mode:
+    params = [[10**3, 10**4, 10**5], ["int", "uint", "float", "object"]]
+    param_names = ["N", "dtype"]
+
+    def setup(self, N, dtype):
+        self.s = Series(np.random.randint(0, N, size=10 * N)).astype(dtype)
+
+    def time_mode(self, N, dtype):
+        self.s.mode()
+
+
+class ModeObjectDropNAFalse:
+    params = [10**3, 10**4, 10**5]
+    param_names = ["N"]
+
+    def setup(self, N):
+        self.s = Series(np.random.randint(0, N, size=10 * N)).astype("object")
+
+    def time_mode(self, N):
+        self.s.mode(dropna=False)
+
+
+class Dir:
+    def setup(self):
+        self.s = Series(index=Index([f"i-{i}" for i in range(10000)], dtype=object))
+
+    def time_dir_strings(self):
+        dir(self.s)
+
+
+class SeriesGetattr:
+    # https://github.com/pandas-dev/pandas/issues/19764
+    def setup(self):
+        self.s = Series(1, index=date_range("2012-01-01", freq="s", periods=10**6))
+
+    def time_series_datetimeindex_repr(self):
+        getattr(self.s, "a", None)
+
+
+class All:
+    params = [[10**3, 10**6], ["fast", "slow"], ["bool", "boolean"]]
+    param_names = ["N", "case", "dtype"]
+
+    def setup(self, N, case, dtype):
+        val = case != "fast"
+        self.s = Series([val] * N, dtype=dtype)
+
+    def time_all(self, N, case, dtype):
+        self.s.all()
+
+
+class Any:
+    params = [[10**3, 10**6], ["fast", "slow"], ["bool", "boolean"]]
+    param_names = ["N", "case", "dtype"]
+
+    def setup(self, N, case, dtype):
+        val = case == "fast"
+        self.s = Series([val] * N, dtype=dtype)
+
+    def time_any(self, N, case, dtype):
+        self.s.any()
+
+
+class NanOps:
+    params = [
+        [
+            "var",
+            "mean",
+            "median",
+            "max",
+            "min",
+            "sum",
+            "std",
+            "sem",
+            "argmax",
+            "skew",
+            "kurt",
+            "prod",
+        ],
+        [10**3, 10**6],
+        ["int8", "int32", "int64", "float64", "Int64", "boolean"],
+    ]
+    param_names = ["func", "N", "dtype"]
+
+    def setup(self, func, N, dtype):
+        if func == "argmax" and dtype in {"Int64", "boolean"}:
+            # Skip argmax for nullable int since this doesn't work yet (GH-24382)
+            raise NotImplementedError
+        self.s = Series(np.ones(N), dtype=dtype)
+        self.func = getattr(self.s, func)
+
+    def time_func(self, func, N, dtype):
+        self.func()
+
+
+class Rank:
+    param_names = ["dtype"]
+    params = [
+        ["int", "uint", "float", "object"],
+    ]
+
+    def setup(self, dtype):
+        self.s = Series(np.random.randint(0, 1000, size=100000), dtype=dtype)
+
+    def time_rank(self, dtype):
+        self.s.rank()
+
+
+class Iter:
+    param_names = ["dtype"]
+    params = [
+        "bool",
+        "boolean",
+        "int64",
+        "Int64",
+        "float64",
+        "Float64",
+        "datetime64[ns]",
+    ]
+
+    def setup(self, dtype):
+        N = 10**5
+        if dtype in ["bool", "boolean"]:
+            data = np.repeat([True, False], N // 2)
+        elif dtype in ["int64", "Int64"]:
+            data = np.arange(N)
+        elif dtype in ["float64", "Float64"]:
+            data = np.random.randn(N)
+        elif dtype == "datetime64[ns]":
+            data = date_range("2000-01-01", freq="s", periods=N)
+        else:
+            raise NotImplementedError
+
+        self.s = Series(data, dtype=dtype)
+
+    def time_iter(self, dtype):
+        for v in self.s:
+            pass
+
+
+class ToNumpy:
+    def setup(self):
+        N = 1_000_000
+        self.ser = Series(
+            np.random.randn(
+                N,
+            )
+        )
+
+    def time_to_numpy(self):
+        self.ser.to_numpy()
+
+    def time_to_numpy_double_copy(self):
+        self.ser.to_numpy(dtype="float64", copy=True)
+
+    def time_to_numpy_copy(self):
+        self.ser.to_numpy(copy=True)
+
+    def time_to_numpy_float_with_nan(self):
+        self.ser.to_numpy(dtype="float64", na_value=np.nan)
+
+
+class Replace:
+    param_names = ["num_to_replace"]
+    params = [100, 1000]
+
+    def setup(self, num_to_replace):
+        N = 1_000_000
+        self.arr = np.random.randn(N)
+        self.arr1 = self.arr.copy()
+        np.random.shuffle(self.arr1)
+        self.ser = Series(self.arr)
+
+        self.to_replace_list = np.random.choice(self.arr, num_to_replace)
+        self.values_list = np.random.choice(self.arr1, num_to_replace)
+
+        self.replace_dict = dict(zip(self.to_replace_list, self.values_list))
+
+    def time_replace_dict(self, num_to_replace):
+        self.ser.replace(self.replace_dict)
+
+    def peakmem_replace_dict(self, num_to_replace):
+        self.ser.replace(self.replace_dict)
+
+    def time_replace_list(self, num_to_replace):
+        self.ser.replace(self.to_replace_list, self.values_list)
+
+    def peakmem_replace_list(self, num_to_replace):
+        self.ser.replace(self.to_replace_list, self.values_list)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/sparse.py b/pandas/asv_bench/benchmarks/sparse.py
new file mode 100644
index 0000000000000..22a5511e4c678
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/sparse.py
@@ -0,0 +1,227 @@
+import numpy as np
+import scipy.sparse
+
+import pandas as pd
+from pandas import (
+    MultiIndex,
+    Series,
+    date_range,
+)
+from pandas.arrays import SparseArray
+
+
+def make_array(size, dense_proportion, fill_value, dtype):
+    dense_size = int(size * dense_proportion)
+    arr = np.full(size, fill_value, dtype)
+    indexer = np.random.choice(np.arange(size), dense_size, replace=False)
+    arr[indexer] = np.random.choice(np.arange(100, dtype=dtype), dense_size)
+    return arr
+
+
+class SparseSeriesToFrame:
+    def setup(self):
+        K = 50
+        N = 50001
+        rng = date_range("1/1/2000", periods=N, freq="min")
+        self.series = {}
+        for i in range(1, K):
+            data = np.random.randn(N)[:-i]
+            idx = rng[:-i]
+            data[100:] = np.nan
+            self.series[i] = Series(SparseArray(data), index=idx)
+
+    def time_series_to_frame(self):
+        pd.DataFrame(self.series)
+
+
+class SparseArrayConstructor:
+    params = ([0.1, 0.01], [0, np.nan], [np.int64, np.float64, object])
+    param_names = ["dense_proportion", "fill_value", "dtype"]
+
+    def setup(self, dense_proportion, fill_value, dtype):
+        N = 10**6
+        self.array = make_array(N, dense_proportion, fill_value, dtype)
+
+    def time_sparse_array(self, dense_proportion, fill_value, dtype):
+        SparseArray(self.array, fill_value=fill_value, dtype=dtype)
+
+
+class SparseDataFrameConstructor:
+    def setup(self):
+        N = 1000
+        self.sparse = scipy.sparse.rand(N, N, 0.005)
+
+    def time_from_scipy(self):
+        pd.DataFrame.sparse.from_spmatrix(self.sparse)
+
+
+class FromCoo:
+    def setup(self):
+        self.matrix = scipy.sparse.coo_matrix(
+            ([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])), shape=(100, 100)
+        )
+
+    def time_sparse_series_from_coo(self):
+        Series.sparse.from_coo(self.matrix)
+
+
+class ToCoo:
+    params = [True, False]
+    param_names = ["sort_labels"]
+
+    def setup(self, sort_labels):
+        s = Series([np.nan] * 10000)
+        s[0] = 3.0
+        s[100] = -1.0
+        s[999] = 12.1
+
+        s_mult_lvl = s.set_axis(MultiIndex.from_product([range(10)] * 4))
+        self.ss_mult_lvl = s_mult_lvl.astype("Sparse")
+
+        s_two_lvl = s.set_axis(MultiIndex.from_product([range(100)] * 2))
+        self.ss_two_lvl = s_two_lvl.astype("Sparse")
+
+    def time_sparse_series_to_coo(self, sort_labels):
+        self.ss_mult_lvl.sparse.to_coo(
+            row_levels=[0, 1], column_levels=[2, 3], sort_labels=sort_labels
+        )
+
+    def time_sparse_series_to_coo_single_level(self, sort_labels):
+        self.ss_two_lvl.sparse.to_coo(sort_labels=sort_labels)
+
+
+class ToCooFrame:
+    def setup(self):
+        N = 10000
+        k = 10
+        arr = np.zeros((N, k), dtype=float)
+        arr[0, 0] = 3.0
+        arr[12, 7] = -1.0
+        arr[0, 9] = 11.2
+        self.df = pd.DataFrame(arr, dtype=pd.SparseDtype("float", fill_value=0.0))
+
+    def time_to_coo(self):
+        self.df.sparse.to_coo()
+
+
+class Arithmetic:
+    params = ([0.1, 0.01], [0, np.nan])
+    param_names = ["dense_proportion", "fill_value"]
+
+    def setup(self, dense_proportion, fill_value):
+        N = 10**6
+        arr1 = make_array(N, dense_proportion, fill_value, np.int64)
+        self.array1 = SparseArray(arr1, fill_value=fill_value)
+        arr2 = make_array(N, dense_proportion, fill_value, np.int64)
+        self.array2 = SparseArray(arr2, fill_value=fill_value)
+
+    def time_make_union(self, dense_proportion, fill_value):
+        self.array1.sp_index.make_union(self.array2.sp_index)
+
+    def time_intersect(self, dense_proportion, fill_value):
+        self.array1.sp_index.intersect(self.array2.sp_index)
+
+    def time_add(self, dense_proportion, fill_value):
+        self.array1 + self.array2
+
+    def time_divide(self, dense_proportion, fill_value):
+        self.array1 / self.array2
+
+
+class ArithmeticBlock:
+    params = [np.nan, 0]
+    param_names = ["fill_value"]
+
+    def setup(self, fill_value):
+        N = 10**6
+        self.arr1 = self.make_block_array(
+            length=N, num_blocks=1000, block_size=10, fill_value=fill_value
+        )
+        self.arr2 = self.make_block_array(
+            length=N, num_blocks=1000, block_size=10, fill_value=fill_value
+        )
+
+    def make_block_array(self, length, num_blocks, block_size, fill_value):
+        arr = np.full(length, fill_value)
+        indices = np.random.choice(
+            np.arange(0, length, block_size), num_blocks, replace=False
+        )
+        for ind in indices:
+            arr[ind : ind + block_size] = np.random.randint(0, 100, block_size)
+        return SparseArray(arr, fill_value=fill_value)
+
+    def time_make_union(self, fill_value):
+        self.arr1.sp_index.make_union(self.arr2.sp_index)
+
+    def time_intersect(self, fill_value):
+        self.arr2.sp_index.intersect(self.arr2.sp_index)
+
+    def time_addition(self, fill_value):
+        self.arr1 + self.arr2
+
+    def time_division(self, fill_value):
+        self.arr1 / self.arr2
+
+
+class MinMax:
+    params = (["min", "max"], [0.0, np.nan])
+    param_names = ["func", "fill_value"]
+
+    def setup(self, func, fill_value):
+        N = 1_000_000
+        arr = make_array(N, 1e-5, fill_value, np.float64)
+        self.sp_arr = SparseArray(arr, fill_value=fill_value)
+
+    def time_min_max(self, func, fill_value):
+        getattr(self.sp_arr, func)()
+
+
+class Take:
+    params = ([np.array([0]), np.arange(100_000), np.full(100_000, -1)], [True, False])
+    param_names = ["indices", "allow_fill"]
+
+    def setup(self, indices, allow_fill):
+        N = 1_000_000
+        fill_value = 0.0
+        arr = make_array(N, 1e-5, fill_value, np.float64)
+        self.sp_arr = SparseArray(arr, fill_value=fill_value)
+
+    def time_take(self, indices, allow_fill):
+        self.sp_arr.take(indices, allow_fill=allow_fill)
+
+
+class GetItem:
+    def setup(self):
+        N = 1_000_000
+        d = 1e-5
+        arr = make_array(N, d, np.nan, np.float64)
+        self.sp_arr = SparseArray(arr)
+
+    def time_integer_indexing(self):
+        self.sp_arr[78]
+
+    def time_slice(self):
+        self.sp_arr[1:]
+
+
+class GetItemMask:
+    params = [True, False, np.nan]
+    param_names = ["fill_value"]
+
+    def setup(self, fill_value):
+        N = 1_000_000
+        d = 1e-5
+        arr = make_array(N, d, np.nan, np.float64)
+        self.sp_arr = SparseArray(arr)
+        b_arr = np.full(shape=N, fill_value=fill_value, dtype=np.bool_)
+        fv_inds = np.unique(
+            np.random.randint(low=0, high=N - 1, size=int(N * d), dtype=np.int32)
+        )
+        b_arr[fv_inds] = True if pd.isna(fill_value) else not fill_value
+        self.sp_b_arr = SparseArray(b_arr, dtype=np.bool_, fill_value=fill_value)
+
+    def time_mask(self, fill_value):
+        self.sp_arr[self.sp_b_arr]
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/stat_ops.py b/pandas/asv_bench/benchmarks/stat_ops.py
new file mode 100644
index 0000000000000..8913293dfa20e
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/stat_ops.py
@@ -0,0 +1,168 @@
+import numpy as np
+
+import pandas as pd
+
+ops = ["mean", "sum", "median", "std", "skew", "kurt", "prod", "sem", "var"]
+
+
+class FrameOps:
+    params = [ops, ["float", "int", "Int64"], [0, 1, None]]
+    param_names = ["op", "dtype", "axis"]
+
+    def setup(self, op, dtype, axis):
+        values = np.random.randn(100000, 4)
+        if dtype == "Int64":
+            values = values.astype(int)
+        df = pd.DataFrame(values).astype(dtype)
+        self.df_func = getattr(df, op)
+
+    def time_op(self, op, dtype, axis):
+        self.df_func(axis=axis)
+
+
+class FrameMixedDtypesOps:
+    params = [ops, [0, 1, None]]
+    param_names = ["op", "axis"]
+
+    def setup(self, op, axis):
+        if op in ("sum", "skew", "kurt", "prod", "sem", "var") or (
+            (op, axis)
+            in (
+                ("mean", 1),
+                ("mean", None),
+                ("median", 1),
+                ("median", None),
+                ("std", 1),
+                ("std", None),
+            )
+        ):
+            # Skipping cases where datetime aggregations are not implemented
+            raise NotImplementedError
+
+        N = 1_000_000
+        df = pd.DataFrame(
+            {
+                "f": np.random.normal(0.0, 1.0, N),
+                "i": np.random.randint(0, N, N),
+                "ts": pd.date_range(start="1/1/2000", periods=N, freq="h"),
+            }
+        )
+
+        self.df_func = getattr(df, op)
+
+    def time_op(self, op, axis):
+        self.df_func(axis=axis)
+
+
+class FrameMultiIndexOps:
+    params = [ops]
+    param_names = ["op"]
+
+    def setup(self, op):
+        levels = [np.arange(10), np.arange(100), np.arange(100)]
+        codes = [
+            np.arange(10).repeat(10000),
+            np.tile(np.arange(100).repeat(100), 10),
+            np.tile(np.tile(np.arange(100), 100), 10),
+        ]
+        index = pd.MultiIndex(levels=levels, codes=codes)
+        df = pd.DataFrame(np.random.randn(len(index), 4), index=index)
+        self.df_func = getattr(df, op)
+
+    def time_op(self, op):
+        self.df_func()
+
+
+class SeriesOps:
+    params = [ops, ["float", "int"]]
+    param_names = ["op", "dtype"]
+
+    def setup(self, op, dtype):
+        s = pd.Series(np.random.randn(100000)).astype(dtype)
+        self.s_func = getattr(s, op)
+
+    def time_op(self, op, dtype):
+        self.s_func()
+
+
+class SeriesMultiIndexOps:
+    params = [ops]
+    param_names = ["op"]
+
+    def setup(self, op):
+        levels = [np.arange(10), np.arange(100), np.arange(100)]
+        codes = [
+            np.arange(10).repeat(10000),
+            np.tile(np.arange(100).repeat(100), 10),
+            np.tile(np.tile(np.arange(100), 100), 10),
+        ]
+        index = pd.MultiIndex(levels=levels, codes=codes)
+        s = pd.Series(np.random.randn(len(index)), index=index)
+        self.s_func = getattr(s, op)
+
+    def time_op(self, op):
+        self.s_func()
+
+
+class Rank:
+    params = [["DataFrame", "Series"], [True, False]]
+    param_names = ["constructor", "pct"]
+
+    def setup(self, constructor, pct):
+        values = np.random.randn(10**5)
+        self.data = getattr(pd, constructor)(values)
+
+    def time_rank(self, constructor, pct):
+        self.data.rank(pct=pct)
+
+    def time_average_old(self, constructor, pct):
+        self.data.rank(pct=pct) / len(self.data)
+
+
+class Correlation:
+    params = [["spearman", "kendall", "pearson"]]
+    param_names = ["method"]
+
+    def setup(self, method):
+        self.df = pd.DataFrame(np.random.randn(500, 15))
+        self.df2 = pd.DataFrame(np.random.randn(500, 15))
+        self.df_wide = pd.DataFrame(np.random.randn(500, 100))
+        self.df_wide_nans = self.df_wide.where(np.random.random((500, 100)) < 0.9)
+        self.s = pd.Series(np.random.randn(500))
+        self.s2 = pd.Series(np.random.randn(500))
+
+    def time_corr(self, method):
+        self.df.corr(method=method)
+
+    def time_corr_wide(self, method):
+        self.df_wide.corr(method=method)
+
+    def time_corr_wide_nans(self, method):
+        self.df_wide_nans.corr(method=method)
+
+    def peakmem_corr_wide(self, method):
+        self.df_wide.corr(method=method)
+
+    def time_corr_series(self, method):
+        self.s.corr(self.s2, method=method)
+
+    def time_corrwith_cols(self, method):
+        self.df.corrwith(self.df2, method=method)
+
+    def time_corrwith_rows(self, method):
+        self.df.corrwith(self.df2, axis=1, method=method)
+
+
+class Covariance:
+    params = []
+    param_names = []
+
+    def setup(self):
+        self.s = pd.Series(np.random.randn(100000))
+        self.s2 = pd.Series(np.random.randn(100000))
+
+    def time_cov_series(self):
+        self.s.cov(self.s2)
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/strftime.py b/pandas/asv_bench/benchmarks/strftime.py
new file mode 100644
index 0000000000000..5f2832e361eb5
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/strftime.py
@@ -0,0 +1,109 @@
+import numpy as np
+
+import pandas as pd
+from pandas import offsets
+
+
+class DatetimeStrftime:
+    timeout = 1500
+    params = [1000, 10000]
+    param_names = ["nobs"]
+
+    def setup(self, nobs):
+        d = "2018-11-29"
+        dt = "2018-11-26 11:18:27.0"
+        self.data = pd.DataFrame(
+            {
+                "dt": [np.datetime64(dt)] * nobs,
+                "d": [np.datetime64(d)] * nobs,
+                "r": [np.random.uniform()] * nobs,
+            }
+        )
+
+    def time_frame_date_to_str(self, nobs):
+        self.data["d"].astype(str)
+
+    def time_frame_date_formatting_default(self, nobs):
+        self.data["d"].dt.strftime(date_format=None)
+
+    def time_frame_date_formatting_default_explicit(self, nobs):
+        self.data["d"].dt.strftime(date_format="%Y-%m-%d")
+
+    def time_frame_date_formatting_custom(self, nobs):
+        self.data["d"].dt.strftime(date_format="%Y---%m---%d")
+
+    def time_frame_datetime_to_str(self, nobs):
+        self.data["dt"].astype(str)
+
+    def time_frame_datetime_formatting_default(self, nobs):
+        self.data["dt"].dt.strftime(date_format=None)
+
+    def time_frame_datetime_formatting_default_explicit_date_only(self, nobs):
+        self.data["dt"].dt.strftime(date_format="%Y-%m-%d")
+
+    def time_frame_datetime_formatting_default_explicit(self, nobs):
+        self.data["dt"].dt.strftime(date_format="%Y-%m-%d %H:%M:%S")
+
+    def time_frame_datetime_formatting_default_with_float(self, nobs):
+        self.data["dt"].dt.strftime(date_format="%Y-%m-%d %H:%M:%S.%f")
+
+    def time_frame_datetime_formatting_custom(self, nobs):
+        self.data["dt"].dt.strftime(date_format="%Y-%m-%d --- %H:%M:%S")
+
+
+class PeriodStrftime:
+    timeout = 1500
+    params = ([1000, 10000], ["D", "h"])
+    param_names = ["nobs", "freq"]
+
+    def setup(self, nobs, freq):
+        self.data = pd.DataFrame(
+            {
+                "p": pd.period_range(start="2000-01-01", periods=nobs, freq=freq),
+                "r": [np.random.uniform()] * nobs,
+            }
+        )
+        self.data["i"] = self.data["p"]
+        self.data.set_index("i", inplace=True)
+        if freq == "D":
+            self.default_fmt = "%Y-%m-%d"
+        elif freq == "h":
+            self.default_fmt = "%Y-%m-%d %H:00"
+
+    def time_frame_period_to_str(self, nobs, freq):
+        self.data["p"].astype(str)
+
+    def time_frame_period_formatting_default(self, nobs, freq):
+        self.data["p"].dt.strftime(date_format=None)
+
+    def time_frame_period_formatting_default_explicit(self, nobs, freq):
+        self.data["p"].dt.strftime(date_format=self.default_fmt)
+
+    def time_frame_period_formatting_custom(self, nobs, freq):
+        self.data["p"].dt.strftime(date_format="%Y-%m-%d --- %H:%M:%S")
+
+    def time_frame_period_formatting_iso8601_strftime_Z(self, nobs, freq):
+        self.data["p"].dt.strftime(date_format="%Y-%m-%dT%H:%M:%SZ")
+
+    def time_frame_period_formatting_iso8601_strftime_offset(self, nobs, freq):
+        """Not optimized yet as %z is not supported by `convert_strftime_format`"""
+        self.data["p"].dt.strftime(date_format="%Y-%m-%dT%H:%M:%S%z")
+
+
+class BusinessHourStrftime:
+    timeout = 1500
+    params = [1000, 10000]
+    param_names = ["nobs"]
+
+    def setup(self, nobs):
+        self.data = pd.DataFrame(
+            {
+                "off": [offsets.BusinessHour()] * nobs,
+            }
+        )
+
+    def time_frame_offset_str(self, nobs):
+        self.data["off"].apply(str)
+
+    def time_frame_offset_repr(self, nobs):
+        self.data["off"].apply(repr)
diff --git a/pandas/asv_bench/benchmarks/strings.py b/pandas/asv_bench/benchmarks/strings.py
new file mode 100644
index 0000000000000..467fab857d306
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/strings.py
@@ -0,0 +1,299 @@
+import warnings
+
+import numpy as np
+
+from pandas import (
+    NA,
+    Categorical,
+    DataFrame,
+    Index,
+    Series,
+)
+from pandas.arrays import StringArray
+
+
+class Dtypes:
+    params = ["str", "string[python]", "string[pyarrow]"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        try:
+            self.s = Series(
+                Index([f"i-{i}" for i in range(10000)], dtype=object)._values,
+                dtype=dtype,
+            )
+        except ImportError as err:
+            raise NotImplementedError from err
+
+
+class Construction:
+    params = (
+        ["series", "frame", "categorical_series"],
+        ["str", "string[python]", "string[pyarrow]"],
+    )
+    param_names = ["pd_type", "dtype"]
+    pd_mapping = {"series": Series, "frame": DataFrame, "categorical_series": Series}
+    dtype_mapping = {"str": "str", "string[python]": object, "string[pyarrow]": object}
+
+    def setup(self, pd_type, dtype):
+        series_arr = np.array(
+            [str(i) * 10 for i in range(100_000)], dtype=self.dtype_mapping[dtype]
+        )
+        if pd_type == "series":
+            self.arr = series_arr
+        elif pd_type == "frame":
+            self.arr = series_arr.reshape((50_000, 2)).copy()
+        elif pd_type == "categorical_series":
+            # GH37371. Testing construction of string series/frames from ExtensionArrays
+            self.arr = Categorical(series_arr)
+
+    def time_construction(self, pd_type, dtype):
+        self.pd_mapping[pd_type](self.arr, dtype=dtype)
+
+    def peakmem_construction(self, pd_type, dtype):
+        self.pd_mapping[pd_type](self.arr, dtype=dtype)
+
+
+class Methods(Dtypes):
+    def time_center(self, dtype):
+        self.s.str.center(100)
+
+    def time_count(self, dtype):
+        self.s.str.count("A")
+
+    def time_endswith(self, dtype):
+        self.s.str.endswith("A")
+
+    def time_extract(self, dtype):
+        with warnings.catch_warnings(record=True):
+            self.s.str.extract("(\\w*)A(\\w*)")
+
+    def time_findall(self, dtype):
+        self.s.str.findall("[A-Z]+")
+
+    def time_find(self, dtype):
+        self.s.str.find("[A-Z]+")
+
+    def time_rfind(self, dtype):
+        self.s.str.rfind("[A-Z]+")
+
+    def time_fullmatch(self, dtype):
+        self.s.str.fullmatch("A")
+
+    def time_get(self, dtype):
+        self.s.str.get(0)
+
+    def time_len(self, dtype):
+        self.s.str.len()
+
+    def time_join(self, dtype):
+        self.s.str.join(" ")
+
+    def time_match(self, dtype):
+        self.s.str.match("A")
+
+    def time_normalize(self, dtype):
+        self.s.str.normalize("NFC")
+
+    def time_pad(self, dtype):
+        self.s.str.pad(100, side="both")
+
+    def time_partition(self, dtype):
+        self.s.str.partition("A")
+
+    def time_rpartition(self, dtype):
+        self.s.str.rpartition("A")
+
+    def time_replace(self, dtype):
+        self.s.str.replace("A", "\x01\x01")
+
+    def time_translate(self, dtype):
+        self.s.str.translate({"A": "\x01\x01"})
+
+    def time_slice(self, dtype):
+        self.s.str.slice(5, 15, 2)
+
+    def time_startswith(self, dtype):
+        self.s.str.startswith("A")
+
+    def time_strip(self, dtype):
+        self.s.str.strip("A")
+
+    def time_rstrip(self, dtype):
+        self.s.str.rstrip("A")
+
+    def time_lstrip(self, dtype):
+        self.s.str.lstrip("A")
+
+    def time_title(self, dtype):
+        self.s.str.title()
+
+    def time_upper(self, dtype):
+        self.s.str.upper()
+
+    def time_lower(self, dtype):
+        self.s.str.lower()
+
+    def time_wrap(self, dtype):
+        self.s.str.wrap(10)
+
+    def time_zfill(self, dtype):
+        self.s.str.zfill(10)
+
+    def time_isalnum(self, dtype):
+        self.s.str.isalnum()
+
+    def time_isalpha(self, dtype):
+        self.s.str.isalpha()
+
+    def time_isdecimal(self, dtype):
+        self.s.str.isdecimal()
+
+    def time_isdigit(self, dtype):
+        self.s.str.isdigit()
+
+    def time_islower(self, dtype):
+        self.s.str.islower()
+
+    def time_isnumeric(self, dtype):
+        self.s.str.isnumeric()
+
+    def time_isspace(self, dtype):
+        self.s.str.isspace()
+
+    def time_istitle(self, dtype):
+        self.s.str.istitle()
+
+    def time_isupper(self, dtype):
+        self.s.str.isupper()
+
+
+class Repeat:
+    params = ["int", "array"]
+    param_names = ["repeats"]
+
+    def setup(self, repeats):
+        N = 10**5
+        self.s = Series(Index([f"i-{i}" for i in range(N)], dtype=object))
+        repeat = {"int": 1, "array": np.random.randint(1, 3, N)}
+        self.values = repeat[repeats]
+
+    def time_repeat(self, repeats):
+        self.s.str.repeat(self.values)
+
+
+class Cat:
+    params = ([0, 3], [None, ","], [None, "-"], [0.0, 0.001, 0.15])
+    param_names = ["other_cols", "sep", "na_rep", "na_frac"]
+
+    def setup(self, other_cols, sep, na_rep, na_frac):
+        N = 10**5
+        mask_gen = lambda: np.random.choice([True, False], N, p=[1 - na_frac, na_frac])
+        self.s = Series(Index([f"i-{i}" for i in range(N)], dtype=object)).where(
+            mask_gen()
+        )
+        if other_cols == 0:
+            # str.cat self-concatenates only for others=None
+            self.others = None
+        else:
+            self.others = DataFrame(
+                {
+                    i: Index([f"i-{i}" for i in range(N)], dtype=object).where(
+                        mask_gen()
+                    )
+                    for i in range(other_cols)
+                }
+            )
+
+    def time_cat(self, other_cols, sep, na_rep, na_frac):
+        # before the concatenation (one caller + other_cols columns), the total
+        # expected fraction of rows containing any NaN is:
+        # reduce(lambda t, _: t + (1 - t) * na_frac, range(other_cols + 1), 0)
+        # for other_cols=3 and na_frac=0.15, this works out to ~48%
+        self.s.str.cat(others=self.others, sep=sep, na_rep=na_rep)
+
+
+class Contains(Dtypes):
+    params = (Dtypes.params, [True, False])
+    param_names = ["dtype", "regex"]
+
+    def setup(self, dtype, regex):
+        super().setup(dtype)
+
+    def time_contains(self, dtype, regex):
+        self.s.str.contains("A", regex=regex)
+
+
+class Split(Dtypes):
+    params = (Dtypes.params, [True, False])
+    param_names = ["dtype", "expand"]
+
+    def setup(self, dtype, expand):
+        super().setup(dtype)
+        self.s = self.s.str.join("--")
+
+    def time_split(self, dtype, expand):
+        self.s.str.split("--", expand=expand)
+
+    def time_rsplit(self, dtype, expand):
+        self.s.str.rsplit("--", expand=expand)
+
+
+class Extract(Dtypes):
+    params = (Dtypes.params, [True, False])
+    param_names = ["dtype", "expand"]
+
+    def setup(self, dtype, expand):
+        super().setup(dtype)
+
+    def time_extract_single_group(self, dtype, expand):
+        with warnings.catch_warnings(record=True):
+            self.s.str.extract("(\\w*)A", expand=expand)
+
+
+class Dummies(Dtypes):
+    def setup(self, dtype):
+        super().setup(dtype)
+        N = len(self.s) // 5
+        self.s = self.s[:N].str.join("|")
+
+    def time_get_dummies(self, dtype):
+        self.s.str.get_dummies("|")
+
+
+class Encode:
+    def setup(self):
+        self.ser = Series(Index([f"i-{i}" for i in range(10_000)], dtype=object))
+
+    def time_encode_decode(self):
+        self.ser.str.encode("utf-8").str.decode("utf-8")
+
+
+class Slice:
+    def setup(self):
+        self.s = Series(["abcdefg", np.nan] * 500000)
+
+    def time_vector_slice(self):
+        # GH 2602
+        self.s.str[:5]
+
+
+class Iter(Dtypes):
+    def time_iter(self, dtype):
+        for i in self.s:
+            pass
+
+
+class StringArrayConstruction:
+    def setup(self):
+        self.series_arr = np.array([str(i) * 10 for i in range(10**5)], dtype=object)
+        self.series_arr_nan = np.concatenate([self.series_arr, np.array([NA] * 1000)])
+
+    def time_string_array_construction(self):
+        StringArray(self.series_arr)
+
+    def time_string_array_with_nan_construction(self):
+        StringArray(self.series_arr_nan)
+
+    def peakmem_stringarray_construction(self):
+        StringArray(self.series_arr)
diff --git a/pandas/asv_bench/benchmarks/timedelta.py b/pandas/asv_bench/benchmarks/timedelta.py
new file mode 100644
index 0000000000000..cb0e4455e1a56
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/timedelta.py
@@ -0,0 +1,61 @@
+"""
+Timedelta benchmarks with non-tslibs dependencies.  See
+benchmarks.tslibs.timedelta for benchmarks that rely only on tslibs.
+"""
+
+from pandas import (
+    DataFrame,
+    Series,
+    timedelta_range,
+)
+
+
+class DatetimeAccessor:
+    def setup_cache(self):
+        N = 100000
+        series = Series(timedelta_range("1 days", periods=N, freq="h"))
+        return series
+
+    def time_dt_accessor(self, series):
+        series.dt
+
+    def time_timedelta_days(self, series):
+        series.dt.days
+
+    def time_timedelta_seconds(self, series):
+        series.dt.seconds
+
+    def time_timedelta_microseconds(self, series):
+        series.dt.microseconds
+
+    def time_timedelta_nanoseconds(self, series):
+        series.dt.nanoseconds
+
+
+class TimedeltaIndexing:
+    def setup(self):
+        self.index = timedelta_range(start="1985", periods=1000, freq="D")
+        self.index2 = timedelta_range(start="1986", periods=1000, freq="D")
+        self.series = Series(range(1000), index=self.index)
+        self.timedelta = self.index[500]
+
+    def time_get_loc(self):
+        self.index.get_loc(self.timedelta)
+
+    def time_shallow_copy(self):
+        self.index._view()
+
+    def time_series_loc(self):
+        self.series.loc[self.timedelta]
+
+    def time_align(self):
+        DataFrame({"a": self.series, "b": self.series[:500]})
+
+    def time_intersection(self):
+        self.index.intersection(self.index2)
+
+    def time_union(self):
+        self.index.union(self.index2)
+
+    def time_unique(self):
+        self.index.unique()
diff --git a/pandas/asv_bench/benchmarks/timeseries.py b/pandas/asv_bench/benchmarks/timeseries.py
new file mode 100644
index 0000000000000..8deec502898d9
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/timeseries.py
@@ -0,0 +1,297 @@
+from datetime import timedelta
+
+import dateutil
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+
+from pandas.tseries.frequencies import infer_freq
+
+try:
+    from pandas.plotting._matplotlib.converter import DatetimeConverter
+except ImportError:
+    from pandas.tseries.converter import DatetimeConverter
+
+
+class DatetimeIndex:
+    params = ["dst", "repeated", "tz_aware", "tz_local", "tz_naive"]
+    param_names = ["index_type"]
+
+    def setup(self, index_type):
+        N = 100000
+        dtidxes = {
+            "dst": date_range(
+                start="10/29/2000 1:00:00", end="10/29/2000 1:59:59", freq="s"
+            ),
+            "repeated": date_range(start="2000", periods=N // 10, freq="s").repeat(10),
+            "tz_aware": date_range(start="2000", periods=N, freq="s", tz="US/Eastern"),
+            "tz_local": date_range(
+                start="2000", periods=N, freq="s", tz=dateutil.tz.tzlocal()
+            ),
+            "tz_naive": date_range(start="2000", periods=N, freq="s"),
+        }
+        self.index = dtidxes[index_type]
+
+    def time_add_timedelta(self, index_type):
+        self.index + timedelta(minutes=2)
+
+    def time_normalize(self, index_type):
+        self.index.normalize()
+
+    def time_unique(self, index_type):
+        self.index.unique()
+
+    def time_to_time(self, index_type):
+        self.index.time
+
+    def time_get(self, index_type):
+        self.index[0]
+
+    def time_timeseries_is_month_start(self, index_type):
+        self.index.is_month_start
+
+    def time_to_date(self, index_type):
+        self.index.date
+
+    def time_to_pydatetime(self, index_type):
+        self.index.to_pydatetime()
+
+    def time_is_dates_only(self, index_type):
+        self.index._is_dates_only
+
+
+class TzLocalize:
+    params = [None, "US/Eastern", "UTC", dateutil.tz.tzutc()]
+    param_names = "tz"
+
+    def setup(self, tz):
+        dst_rng = date_range(
+            start="10/29/2000 1:00:00", end="10/29/2000 1:59:59", freq="s"
+        )
+        self.index = date_range(start="10/29/2000", end="10/29/2000 00:59:59", freq="s")
+        self.index = self.index.append(dst_rng)
+        self.index = self.index.append(dst_rng)
+        self.index = self.index.append(
+            date_range(start="10/29/2000 2:00:00", end="10/29/2000 3:00:00", freq="s")
+        )
+
+    def time_infer_dst(self, tz):
+        self.index.tz_localize(tz, ambiguous="infer")
+
+
+class ResetIndex:
+    params = [None, "US/Eastern"]
+    param_names = "tz"
+
+    def setup(self, tz):
+        idx = date_range(start="1/1/2000", periods=1000, freq="h", tz=tz)
+        self.df = DataFrame(np.random.randn(1000, 2), index=idx)
+
+    def time_reset_datetimeindex(self, tz):
+        self.df.reset_index()
+
+
+class InferFreq:
+    # This depends mostly on code in _libs/, tseries/, and core.algos.unique
+    params = [None, "D", "B"]
+    param_names = ["freq"]
+
+    def setup(self, freq):
+        if freq is None:
+            self.idx = date_range(start="1/1/1700", freq="D", periods=10000)
+            self.idx._data._freq = None
+        else:
+            self.idx = date_range(start="1/1/1700", freq=freq, periods=10000)
+
+    def time_infer_freq(self, freq):
+        infer_freq(self.idx)
+
+
+class TimeDatetimeConverter:
+    def setup(self):
+        N = 100000
+        self.rng = date_range(start="1/1/2000", periods=N, freq="min")
+
+    def time_convert(self):
+        DatetimeConverter.convert(self.rng, None, None)
+
+
+class Iteration:
+    params = [date_range, period_range, timedelta_range]
+    param_names = ["time_index"]
+
+    def setup(self, time_index):
+        N = 10**6
+        if time_index is timedelta_range:
+            self.idx = time_index(start=0, freq="min", periods=N)
+        else:
+            self.idx = time_index(start="20140101", freq="min", periods=N)
+        self.exit = 10000
+
+    def time_iter(self, time_index):
+        for _ in self.idx:
+            pass
+
+    def time_iter_preexit(self, time_index):
+        for i, _ in enumerate(self.idx):
+            if i > self.exit:
+                break
+
+
+class ResampleDataFrame:
+    params = ["max", "mean", "min"]
+    param_names = ["method"]
+
+    def setup(self, method):
+        rng = date_range(start="20130101", periods=100000, freq="50ms")
+        df = DataFrame(np.random.randn(100000, 2), index=rng)
+        self.resample = getattr(df.resample("1s"), method)
+
+    def time_method(self, method):
+        self.resample()
+
+
+class ResampleSeries:
+    params = (["period", "datetime"], ["5min", "1D"], ["mean", "ohlc"])
+    param_names = ["index", "freq", "method"]
+
+    def setup(self, index, freq, method):
+        indexes = {
+            "period": period_range(start="1/1/2000", end="1/1/2001", freq="min"),
+            "datetime": date_range(start="1/1/2000", end="1/1/2001", freq="min"),
+        }
+        idx = indexes[index]
+        ts = Series(np.random.randn(len(idx)), index=idx)
+        self.resample = getattr(ts.resample(freq), method)
+
+    def time_resample(self, index, freq, method):
+        self.resample()
+
+
+class ResampleDatetetime64:
+    # GH 7754
+    def setup(self):
+        rng3 = date_range(
+            start="2000-01-01 00:00:00", end="2000-01-01 10:00:00", freq="555000us"
+        )
+        self.dt_ts = Series(5, rng3, dtype="datetime64[ns]")
+
+    def time_resample(self):
+        self.dt_ts.resample("1s").last()
+
+
+class AsOf:
+    params = ["DataFrame", "Series"]
+    param_names = ["constructor"]
+
+    def setup(self, constructor):
+        N = 10000
+        M = 10
+        rng = date_range(start="1/1/1990", periods=N, freq="53s")
+        data = {
+            "DataFrame": DataFrame(np.random.randn(N, M)),
+            "Series": Series(np.random.randn(N)),
+        }
+        self.ts = data[constructor]
+        self.ts.index = rng
+        self.ts2 = self.ts.copy()
+        self.ts2.iloc[250:5000] = np.nan
+        self.ts3 = self.ts.copy()
+        self.ts3.iloc[-5000:] = np.nan
+        self.dates = date_range(start="1/1/1990", periods=N * 10, freq="5s")
+        self.date = self.dates[0]
+        self.date_last = self.dates[-1]
+        self.date_early = self.date - timedelta(10)
+
+    # test speed of pre-computing NAs.
+    def time_asof(self, constructor):
+        self.ts.asof(self.dates)
+
+    # should be roughly the same as above.
+    def time_asof_nan(self, constructor):
+        self.ts2.asof(self.dates)
+
+    # test speed of the code path for a scalar index
+    # without *while* loop
+    def time_asof_single(self, constructor):
+        self.ts.asof(self.date)
+
+    # test speed of the code path for a scalar index
+    # before the start. should be the same as above.
+    def time_asof_single_early(self, constructor):
+        self.ts.asof(self.date_early)
+
+    # test the speed of the code path for a scalar index
+    # with a long *while* loop. should still be much
+    # faster than pre-computing all the NAs.
+    def time_asof_nan_single(self, constructor):
+        self.ts3.asof(self.date_last)
+
+
+class SortIndex:
+    params = [True, False]
+    param_names = ["monotonic"]
+
+    def setup(self, monotonic):
+        N = 10**5
+        idx = date_range(start="1/1/2000", periods=N, freq="s")
+        self.s = Series(np.random.randn(N), index=idx)
+        if not monotonic:
+            self.s = self.s.sample(frac=1)
+
+    def time_sort_index(self, monotonic):
+        self.s.sort_index()
+
+    def time_get_slice(self, monotonic):
+        self.s[:10000]
+
+
+class Lookup:
+    def setup(self):
+        N = 1500000
+        rng = date_range(start="1/1/2000", periods=N, freq="s")
+        self.ts = Series(1, index=rng)
+        self.lookup_val = rng[N // 2]
+
+    def time_lookup_and_cleanup(self):
+        self.ts[self.lookup_val]
+        self.ts.index._cleanup()
+
+
+class DatetimeAccessor:
+    params = [None, "US/Eastern", "UTC", dateutil.tz.tzutc()]
+    param_names = "tz"
+
+    def setup(self, tz):
+        N = 100000
+        self.series = Series(date_range(start="1/1/2000", periods=N, freq="min", tz=tz))
+
+    def time_dt_accessor(self, tz):
+        self.series.dt
+
+    def time_dt_accessor_normalize(self, tz):
+        self.series.dt.normalize()
+
+    def time_dt_accessor_month_name(self, tz):
+        self.series.dt.month_name()
+
+    def time_dt_accessor_day_name(self, tz):
+        self.series.dt.day_name()
+
+    def time_dt_accessor_time(self, tz):
+        self.series.dt.time
+
+    def time_dt_accessor_date(self, tz):
+        self.series.dt.date
+
+    def time_dt_accessor_year(self, tz):
+        self.series.dt.year
+
+
+from .pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/tslibs/__init__.py b/pandas/asv_bench/benchmarks/tslibs/__init__.py
new file mode 100644
index 0000000000000..815cf55b291cd
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/__init__.py
@@ -0,0 +1,7 @@
+"""
+Benchmarks in this directory should depend only on tslibs, tseries.offsets,
+and to_offset.
+
+i.e. any code changes that do not touch those files should not need to
+run these benchmarks.
+"""
diff --git a/pandas/asv_bench/benchmarks/tslibs/fields.py b/pandas/asv_bench/benchmarks/tslibs/fields.py
new file mode 100644
index 0000000000000..fe31879e67a67
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/fields.py
@@ -0,0 +1,82 @@
+import numpy as np
+
+from pandas._libs.tslibs.fields import (
+    get_date_field,
+    get_start_end_field,
+    get_timedelta_field,
+)
+
+from .tslib import _sizes
+
+
+class TimeGetTimedeltaField:
+    params = [
+        _sizes,
+        ["seconds", "microseconds", "nanoseconds"],
+    ]
+    param_names = ["size", "field"]
+
+    def setup(self, size, field):
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        self.i8data = arr
+        arr = np.random.randint(-86400 * 1_000_000_000, 0, size=size, dtype="i8")
+        self.i8data_negative = arr
+
+    def time_get_timedelta_field(self, size, field):
+        get_timedelta_field(self.i8data, field)
+
+    def time_get_timedelta_field_negative_td(self, size, field):
+        get_timedelta_field(self.i8data_negative, field)
+
+
+class TimeGetDateField:
+    params = [
+        _sizes,
+        [
+            "Y",
+            "M",
+            "D",
+            "h",
+            "m",
+            "s",
+            "us",
+            "ns",
+            "doy",
+            "dow",
+            "woy",
+            "q",
+            "dim",
+            "is_leap_year",
+        ],
+    ]
+    param_names = ["size", "field"]
+
+    def setup(self, size, field):
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        self.i8data = arr
+
+    def time_get_date_field(self, size, field):
+        get_date_field(self.i8data, field)
+
+
+class TimeGetStartEndField:
+    params = [
+        _sizes,
+        ["start", "end"],
+        ["month", "quarter", "year"],
+        ["B", None, "QS"],
+        [12, 3, 5],
+    ]
+    param_names = ["size", "side", "period", "freqstr", "month_kw"]
+
+    def setup(self, size, side, period, freqstr, month_kw):
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        self.i8data = arr
+
+        self.attrname = f"is_{period}_{side}"
+
+    def time_get_start_end_field(self, size, side, period, freqstr, month_kw):
+        get_start_end_field(self.i8data, self.attrname, freqstr, month_kw=month_kw)
+
+
+from ..pandas_vb_common import setup  # noqa: F401 isort:skip
diff --git a/pandas/asv_bench/benchmarks/tslibs/normalize.py b/pandas/asv_bench/benchmarks/tslibs/normalize.py
new file mode 100644
index 0000000000000..b263ae21422b6
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/normalize.py
@@ -0,0 +1,45 @@
+try:
+    from pandas._libs.tslibs import (
+        is_date_array_normalized,
+        normalize_i8_timestamps,
+    )
+except ImportError:
+    from pandas._libs.tslibs.conversion import (
+        normalize_i8_timestamps,
+        is_date_array_normalized,
+    )
+
+import pandas as pd
+
+from .tslib import (
+    _sizes,
+    _tzs,
+    tzlocal_obj,
+)
+
+
+class Normalize:
+    params = [
+        _sizes,
+        _tzs,
+    ]
+    param_names = ["size", "tz"]
+
+    def setup(self, size, tz):
+        # use an array that will have is_date_array_normalized give True,
+        #  so we do not short-circuit early.
+        dti = pd.date_range("2016-01-01", periods=10, tz=tz).repeat(size // 10)
+        self.i8data = dti.asi8
+
+        if size == 10**6 and tz is tzlocal_obj:
+            # tzlocal is cumbersomely slow, so skip to keep runtime in check
+            raise NotImplementedError
+
+    def time_normalize_i8_timestamps(self, size, tz):
+        # 10 i.e. NPY_FR_ns
+        normalize_i8_timestamps(self.i8data, tz, 10)
+
+    def time_is_date_array_normalized(self, size, tz):
+        # TODO: cases with different levels of short-circuiting
+        # 10 i.e. NPY_FR_ns
+        is_date_array_normalized(self.i8data, tz, 10)
diff --git a/pandas/asv_bench/benchmarks/tslibs/offsets.py b/pandas/asv_bench/benchmarks/tslibs/offsets.py
new file mode 100644
index 0000000000000..55bd3c31c055c
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/offsets.py
@@ -0,0 +1,86 @@
+"""
+offsets benchmarks that rely only on tslibs.  See benchmarks.offset for
+offsets benchmarks that rely on other parts of pandas.
+"""
+
+from datetime import datetime
+
+import numpy as np
+
+from pandas import offsets
+
+try:
+    import pandas.tseries.holiday
+except ImportError:
+    pass
+
+hcal = pandas.tseries.holiday.USFederalHolidayCalendar()
+# These offsets currently raise a NotImplementedError with .apply_index()
+non_apply = [
+    offsets.Day(),
+    offsets.BYearEnd(),
+    offsets.BYearBegin(),
+    offsets.BQuarterEnd(),
+    offsets.BQuarterBegin(),
+    offsets.BMonthEnd(),
+    offsets.BMonthBegin(),
+    offsets.CustomBusinessDay(),
+    offsets.CustomBusinessDay(calendar=hcal),
+    offsets.CustomBusinessMonthBegin(calendar=hcal),
+    offsets.CustomBusinessMonthEnd(calendar=hcal),
+    offsets.CustomBusinessMonthEnd(calendar=hcal),
+]
+other_offsets = [
+    offsets.YearEnd(),
+    offsets.YearBegin(),
+    offsets.QuarterEnd(),
+    offsets.QuarterBegin(),
+    offsets.MonthEnd(),
+    offsets.MonthBegin(),
+    offsets.DateOffset(months=2, days=2),
+    offsets.BusinessDay(),
+    offsets.SemiMonthEnd(),
+    offsets.SemiMonthBegin(),
+]
+offset_objs = non_apply + other_offsets
+
+
+class OnOffset:
+    params = offset_objs
+    param_names = ["offset"]
+
+    def setup(self, offset):
+        self.dates = [
+            datetime(2016, m, d)
+            for m in [10, 11, 12]
+            for d in [1, 2, 3, 28, 29, 30, 31]
+            if not (m == 11 and d == 31)
+        ]
+
+    def time_on_offset(self, offset):
+        for date in self.dates:
+            offset.is_on_offset(date)
+
+
+class OffestDatetimeArithmetic:
+    params = offset_objs
+    param_names = ["offset"]
+
+    def setup(self, offset):
+        self.date = datetime(2011, 1, 1)
+        self.dt64 = np.datetime64("2011-01-01 09:00Z")
+
+    def time_add_np_dt64(self, offset):
+        offset + self.dt64
+
+    def time_add(self, offset):
+        self.date + offset
+
+    def time_add_10(self, offset):
+        self.date + (10 * offset)
+
+    def time_subtract(self, offset):
+        self.date - offset
+
+    def time_subtract_10(self, offset):
+        self.date - (10 * offset)
diff --git a/pandas/asv_bench/benchmarks/tslibs/period.py b/pandas/asv_bench/benchmarks/tslibs/period.py
new file mode 100644
index 0000000000000..af3bfac6d3d01
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/period.py
@@ -0,0 +1,162 @@
+"""
+Period benchmarks that rely only on tslibs. See benchmarks.period for
+Period benchmarks that rely on other parts of pandas.
+"""
+
+import numpy as np
+
+from pandas._libs.tslibs.period import (
+    Period,
+    periodarr_to_dt64arr,
+)
+
+from pandas.tseries.frequencies import to_offset
+
+from .tslib import (
+    _sizes,
+    _tzs,
+    tzlocal_obj,
+)
+
+try:
+    from pandas._libs.tslibs.vectorized import dt64arr_to_periodarr
+except ImportError:
+    from pandas._libs.tslibs.period import dt64arr_to_periodarr
+
+
+class PeriodProperties:
+    params = (
+        ["M", "min"],
+        [
+            "year",
+            "month",
+            "day",
+            "hour",
+            "minute",
+            "second",
+            "is_leap_year",
+            "quarter",
+            "qyear",
+            "week",
+            "daysinmonth",
+            "dayofweek",
+            "dayofyear",
+            "start_time",
+            "end_time",
+        ],
+    )
+    param_names = ["freq", "attr"]
+
+    def setup(self, freq, attr):
+        self.per = Period("2012-06-01", freq=freq)
+
+    def time_property(self, freq, attr):
+        getattr(self.per, attr)
+
+
+class PeriodUnaryMethods:
+    params = ["M", "min"]
+    param_names = ["freq"]
+
+    def setup(self, freq):
+        self.per = Period("2012-06-01", freq=freq)
+        if freq == "M":
+            self.default_fmt = "%Y-%m"
+        elif freq == "min":
+            self.default_fmt = "%Y-%m-%d %H:%M"
+
+    def time_to_timestamp(self, freq):
+        self.per.to_timestamp()
+
+    def time_now(self, freq):
+        self.per.now(freq)
+
+    def time_asfreq(self, freq):
+        self.per.asfreq("Y")
+
+    def time_str(self, freq):
+        str(self.per)
+
+    def time_repr(self, freq):
+        repr(self.per)
+
+    def time_strftime_default(self, freq):
+        self.per.strftime(None)
+
+    def time_strftime_default_explicit(self, freq):
+        self.per.strftime(self.default_fmt)
+
+    def time_strftime_custom(self, freq):
+        self.per.strftime("%b. %d, %Y was a %A")
+
+
+class PeriodConstructor:
+    params = [["D"], [True, False]]
+    param_names = ["freq", "is_offset"]
+
+    def setup(self, freq, is_offset):
+        if is_offset:
+            self.freq = to_offset(freq)
+        else:
+            self.freq = freq
+
+    def time_period_constructor(self, freq, is_offset):
+        Period("2012-06-01", freq=freq)
+
+
+_freq_ints = [
+    1000,
+    1011,  # Annual - November End
+    2000,
+    2011,  # Quarterly - November End
+    3000,
+    4000,
+    4006,  # Weekly - Saturday End
+    5000,
+    6000,
+    7000,
+    8000,
+    9000,
+    10000,
+    11000,
+    12000,
+]
+
+
+class TimePeriodArrToDT64Arr:
+    params = [
+        _sizes,
+        _freq_ints,
+    ]
+    param_names = ["size", "freq"]
+
+    def setup(self, size, freq):
+        arr = np.arange(10, dtype="i8").repeat(size // 10)
+        self.i8values = arr
+
+    def time_periodarray_to_dt64arr(self, size, freq):
+        periodarr_to_dt64arr(self.i8values, freq)
+
+
+class TimeDT64ArrToPeriodArr:
+    params = [
+        _sizes,
+        _freq_ints,
+        _tzs,
+    ]
+    param_names = ["size", "freq", "tz"]
+
+    def setup(self, size, freq, tz):
+        if size == 10**6 and tz is tzlocal_obj:
+            # tzlocal is cumbersomely slow, so skip to keep runtime in check
+            raise NotImplementedError
+
+        # we pick 2**55 because smaller values end up returning
+        # -1 from npy_datetimestruct_to_datetime with NPY_FR_Y frequency
+        # this artificially slows down functions since -1 is also the
+        # error sentinel
+        arr = np.arange(2**55, 2**55 + 10, dtype="i8").repeat(size // 10)
+        self.i8values = arr
+
+    def time_dt64arr_to_periodarr(self, size, freq, tz):
+        dt64arr_to_periodarr(self.i8values, freq, tz)
diff --git a/pandas/asv_bench/benchmarks/tslibs/resolution.py b/pandas/asv_bench/benchmarks/tslibs/resolution.py
new file mode 100644
index 0000000000000..6317d299379d3
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/resolution.py
@@ -0,0 +1,53 @@
+"""
+ipython analogue:
+
+tr = TimeResolution()
+mi = pd.MultiIndex.from_product(tr.params[:-1] + ([str(x) for x in tr.params[-1]],))
+df = pd.DataFrame(np.nan, index=mi, columns=["mean", "stdev"])
+
+for unit in tr.params[0]:
+    for size in tr.params[1]:
+        for tz in tr.params[2]:
+            tr.setup(unit, size, tz)
+            key = (unit, size, str(tz))
+            print(key)
+
+            val = %timeit -o tr.time_get_resolution(unit, size, tz)
+
+            df.loc[key] = (val.average, val.stdev)
+
+"""
+
+import numpy as np
+
+try:
+    from pandas._libs.tslibs import get_resolution
+except ImportError:
+    from pandas._libs.tslibs.resolution import get_resolution
+
+from .tslib import (
+    _sizes,
+    _tzs,
+    tzlocal_obj,
+)
+
+
+class TimeResolution:
+    params = (
+        ["D", "h", "m", "s", "us", "ns"],
+        _sizes,
+        _tzs,
+    )
+    param_names = ["unit", "size", "tz"]
+
+    def setup(self, unit, size, tz):
+        if size == 10**6 and tz is tzlocal_obj:
+            # tzlocal is cumbersomely slow, so skip to keep runtime in check
+            raise NotImplementedError
+
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        arr = arr.view(f"M8[{unit}]").astype("M8[ns]").view("i8")
+        self.i8data = arr
+
+    def time_get_resolution(self, unit, size, tz):
+        get_resolution(self.i8data, tz)
diff --git a/pandas/asv_bench/benchmarks/tslibs/timedelta.py b/pandas/asv_bench/benchmarks/tslibs/timedelta.py
new file mode 100644
index 0000000000000..9d9689fcfa94b
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/timedelta.py
@@ -0,0 +1,70 @@
+"""
+Timedelta benchmarks that rely only on tslibs. See benchmarks.timedeltas for
+Timedelta benchmarks that rely on other parts of pandas.
+"""
+
+import datetime
+
+import numpy as np
+
+from pandas import Timedelta
+
+
+class TimedeltaConstructor:
+    def setup(self):
+        self.nptimedelta64 = np.timedelta64(3600)
+        self.dttimedelta = datetime.timedelta(seconds=3600)
+        self.td = Timedelta(3600, unit="s")
+
+    def time_from_int(self):
+        Timedelta(123456789)
+
+    def time_from_unit(self):
+        Timedelta(1, unit="D")
+
+    def time_from_components(self):
+        Timedelta(
+            days=1,
+            hours=2,
+            minutes=3,
+            seconds=4,
+            milliseconds=5,
+            microseconds=6,
+            nanoseconds=7,
+        )
+
+    def time_from_datetime_timedelta(self):
+        Timedelta(self.dttimedelta)
+
+    def time_from_np_timedelta(self):
+        Timedelta(self.nptimedelta64)
+
+    def time_from_string(self):
+        Timedelta("1 days")
+
+    def time_from_iso_format(self):
+        Timedelta("P4DT12H30M5S")
+
+    def time_from_missing(self):
+        Timedelta("nat")
+
+    def time_from_pd_timedelta(self):
+        Timedelta(self.td)
+
+
+class TimedeltaProperties:
+    def setup_cache(self):
+        td = Timedelta(days=365, minutes=35, seconds=25, milliseconds=35)
+        return td
+
+    def time_timedelta_days(self, td):
+        td.days
+
+    def time_timedelta_seconds(self, td):
+        td.seconds
+
+    def time_timedelta_microseconds(self, td):
+        td.microseconds
+
+    def time_timedelta_nanoseconds(self, td):
+        td.nanoseconds
diff --git a/pandas/asv_bench/benchmarks/tslibs/timestamp.py b/pandas/asv_bench/benchmarks/tslibs/timestamp.py
new file mode 100644
index 0000000000000..6145966fb6a0e
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/timestamp.py
@@ -0,0 +1,155 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+import zoneinfo
+
+import numpy as np
+
+from pandas import Timestamp
+
+from .tslib import _tzs
+
+
+class TimestampConstruction:
+    def setup(self):
+        self.npdatetime64 = np.datetime64("2020-01-01 00:00:00")
+        self.dttime_unaware = datetime(2020, 1, 1, 0, 0, 0)
+        self.dttime_aware = datetime(2020, 1, 1, 0, 0, 0, 0, timezone.utc)
+        self.ts = Timestamp("2020-01-01 00:00:00")
+
+    def time_parse_iso8601_no_tz(self):
+        Timestamp("2017-08-25 08:16:14")
+
+    def time_parse_iso8601_tz(self):
+        Timestamp("2017-08-25 08:16:14-0500")
+
+    def time_parse_dateutil(self):
+        Timestamp("2017/08/25 08:16:14 AM")
+
+    def time_parse_today(self):
+        Timestamp("today")
+
+    def time_parse_now(self):
+        Timestamp("now")
+
+    def time_fromordinal(self):
+        Timestamp.fromordinal(730120)
+
+    def time_fromtimestamp(self):
+        Timestamp.fromtimestamp(1515448538)
+
+    def time_from_npdatetime64(self):
+        Timestamp(self.npdatetime64)
+
+    def time_from_datetime_unaware(self):
+        Timestamp(self.dttime_unaware)
+
+    def time_from_datetime_aware(self):
+        Timestamp(self.dttime_aware)
+
+    def time_from_pd_timestamp(self):
+        Timestamp(self.ts)
+
+
+class TimestampProperties:
+    params = [_tzs]
+    param_names = ["tz"]
+
+    def setup(self, tz):
+        self.ts = Timestamp("2017-08-25 08:16:14", tzinfo=tz)
+
+    def time_tz(self, tz):
+        self.ts.tz
+
+    def time_dayofweek(self, tz):
+        self.ts.dayofweek
+
+    def time_dayofyear(self, tz):
+        self.ts.dayofyear
+
+    def time_week(self, tz):
+        self.ts.week
+
+    def time_quarter(self, tz):
+        self.ts.quarter
+
+    def time_days_in_month(self, tz):
+        self.ts.days_in_month
+
+    def time_is_month_start(self, tz):
+        self.ts.is_month_start
+
+    def time_is_month_end(self, tz):
+        self.ts.is_month_end
+
+    def time_is_quarter_start(self, tz):
+        self.ts.is_quarter_start
+
+    def time_is_quarter_end(self, tz):
+        self.ts.is_quarter_end
+
+    def time_is_year_start(self, tz):
+        self.ts.is_year_start
+
+    def time_is_year_end(self, tz):
+        self.ts.is_year_end
+
+    def time_is_leap_year(self, tz):
+        self.ts.is_leap_year
+
+    def time_microsecond(self, tz):
+        self.ts.microsecond
+
+    def time_month_name(self, tz):
+        self.ts.month_name()
+
+    def time_weekday_name(self, tz):
+        self.ts.day_name()
+
+
+class TimestampOps:
+    params = _tzs
+    param_names = ["tz"]
+
+    def setup(self, tz):
+        self.ts = Timestamp("2017-08-25 08:16:14", tz=tz)
+
+    def time_replace_tz(self, tz):
+        self.ts.replace(tzinfo=zoneinfo.ZoneInfo("US/Eastern"))
+
+    def time_replace_None(self, tz):
+        self.ts.replace(tzinfo=None)
+
+    def time_to_pydatetime(self, tz):
+        self.ts.to_pydatetime()
+
+    def time_normalize(self, tz):
+        self.ts.normalize()
+
+    def time_tz_convert(self, tz):
+        if self.ts.tz is not None:
+            self.ts.tz_convert(tz)
+
+    def time_tz_localize(self, tz):
+        if self.ts.tz is None:
+            self.ts.tz_localize(tz)
+
+    def time_to_julian_date(self, tz):
+        self.ts.to_julian_date()
+
+    def time_floor(self, tz):
+        self.ts.floor("5min")
+
+    def time_ceil(self, tz):
+        self.ts.ceil("5min")
+
+
+class TimestampAcrossDst:
+    def setup(self):
+        dt = datetime(2016, 3, 27, 1, fold=0)
+        self.tzinfo = dt.astimezone(zoneinfo.ZoneInfo("Europe/Berlin")).tzinfo
+        self.ts2 = Timestamp(dt)
+
+    def time_replace_across_dst(self):
+        self.ts2.replace(tzinfo=self.tzinfo)
diff --git a/pandas/asv_bench/benchmarks/tslibs/tslib.py b/pandas/asv_bench/benchmarks/tslibs/tslib.py
new file mode 100644
index 0000000000000..885cf48d01743
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/tslib.py
@@ -0,0 +1,69 @@
+"""
+ipython analogue:
+
+tr = TimeIntsToPydatetime()
+mi = pd.MultiIndex.from_product(
+    tr.params[:-1] + ([str(x) for x in tr.params[-1]],)
+)
+df = pd.DataFrame(np.nan, index=mi, columns=["mean", "stdev"])
+for box in tr.params[0]:
+    for size in tr.params[1]:
+        for tz in tr.params[2]:
+            tr.setup(box, size, tz)
+            key = (box, size, str(tz))
+            print(key)
+            val = %timeit -o tr.time_ints_to_pydatetime(box, size, tz)
+            df.loc[key] = (val.average, val.stdev)
+"""
+
+from datetime import (
+    timedelta,
+    timezone,
+)
+import zoneinfo
+
+from dateutil.tz import (
+    gettz,
+    tzlocal,
+)
+import numpy as np
+
+try:
+    from pandas._libs.tslibs import ints_to_pydatetime
+except ImportError:
+    from pandas._libs.tslib import ints_to_pydatetime
+
+tzlocal_obj = tzlocal()
+_tzs = [
+    None,
+    timezone.utc,
+    timezone(timedelta(minutes=60)),
+    zoneinfo.ZoneInfo("US/Pacific"),
+    gettz("Asia/Tokyo"),
+    tzlocal_obj,
+]
+_sizes = [0, 1, 100, 10**4, 10**6]
+
+
+class TimeIntsToPydatetime:
+    params = (
+        ["time", "date", "datetime", "timestamp"],
+        _sizes,
+        _tzs,
+    )
+    param_names = ["box", "size", "tz"]
+    # TODO: fold?
+
+    def setup(self, box, size, tz):
+        if box == "date" and tz is not None:
+            # tz is ignored, so avoid running redundant benchmarks
+            raise NotImplementedError  # skip benchmark
+        if size == 10**6 and tz is _tzs[-1]:
+            # This is cumbersomely-slow, so skip to trim runtime
+            raise NotImplementedError  # skip benchmark
+
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        self.i8data = arr
+
+    def time_ints_to_pydatetime(self, box, size, tz):
+        ints_to_pydatetime(self.i8data, tz, box=box)
diff --git a/pandas/asv_bench/benchmarks/tslibs/tz_convert.py b/pandas/asv_bench/benchmarks/tslibs/tz_convert.py
new file mode 100644
index 0000000000000..c87adb5e5d0e9
--- /dev/null
+++ b/pandas/asv_bench/benchmarks/tslibs/tz_convert.py
@@ -0,0 +1,53 @@
+from datetime import timezone
+
+import numpy as np
+
+from pandas._libs.tslibs.tzconversion import tz_localize_to_utc
+
+from .tslib import (
+    _sizes,
+    _tzs,
+    tzlocal_obj,
+)
+
+try:
+    old_sig = False
+    from pandas._libs.tslibs import tz_convert_from_utc
+except ImportError:
+    try:
+        old_sig = False
+        from pandas._libs.tslibs.tzconversion import tz_convert_from_utc
+    except ImportError:
+        old_sig = True
+        from pandas._libs.tslibs.tzconversion import tz_convert as tz_convert_from_utc
+
+
+class TimeTZConvert:
+    params = [
+        _sizes,
+        [x for x in _tzs if x is not None],
+    ]
+    param_names = ["size", "tz"]
+
+    def setup(self, size, tz):
+        if size == 10**6 and tz is tzlocal_obj:
+            # tzlocal is cumbersomely slow, so skip to keep runtime in check
+            raise NotImplementedError
+
+        arr = np.random.randint(0, 10, size=size, dtype="i8")
+        self.i8data = arr
+
+    def time_tz_convert_from_utc(self, size, tz):
+        # effectively:
+        #  dti = DatetimeIndex(self.i8data, tz=tz)
+        #  dti.tz_localize(None)
+        if old_sig:
+            tz_convert_from_utc(self.i8data, timezone.utc, tz)
+        else:
+            tz_convert_from_utc(self.i8data, tz)
+
+    def time_tz_localize_to_utc(self, size, tz):
+        # effectively:
+        #  dti = DatetimeIndex(self.i8data)
+        #  dti.tz_localize(tz, ambiguous="NaT", nonexistent="NaT")
+        tz_localize_to_utc(self.i8data, tz, ambiguous="NaT", nonexistent="NaT")
diff --git a/pandas/ci/.condarc b/pandas/ci/.condarc
new file mode 100644
index 0000000000000..f5fb60b208a9c
--- /dev/null
+++ b/pandas/ci/.condarc
@@ -0,0 +1,32 @@
+# https://docs.conda.io/projects/conda/en/latest/configuration.html
+
+# always_yes (NoneType, bool)
+#   aliases: yes
+#   Automatically choose the 'yes' option whenever asked to proceed with a
+#   conda operation, such as when running `conda install`.
+#
+always_yes: true
+
+# remote_connect_timeout_secs (float)
+#   The number seconds conda will wait for your client to establish a
+#   connection to a remote url resource.
+#
+remote_connect_timeout_secs: 30
+
+# remote_max_retries (int)
+#   The maximum number of retries each HTTP connection should attempt.
+#
+remote_max_retries: 10
+
+# remote_backoff_factor (int)
+#   The factor determines the time HTTP connection should wait for
+#   attempt.
+#
+remote_backoff_factor: 3
+
+# remote_read_timeout_secs (float)
+#   Once conda has connected to a remote resource and sent an HTTP
+#   request, the read timeout is the number of seconds conda will wait for
+#   the server to send a response.
+#
+remote_read_timeout_secs: 60.0
diff --git a/pandas/ci/code_checks.sh b/pandas/ci/code_checks.sh
new file mode 100755
index 0000000000000..5782b2b171e07
--- /dev/null
+++ b/pandas/ci/code_checks.sh
@@ -0,0 +1,279 @@
+#!/bin/bash
+#
+# Run checks related to code quality.
+#
+# This script is intended for both the CI and to check locally that code standards are
+# respected. We run doctests here (currently some files only), and we
+# validate formatting error in docstrings.
+#
+# Usage:
+#   $ ./ci/code_checks.sh               # run all checks
+#   $ ./ci/code_checks.sh code          # checks on imported code
+#   $ ./ci/code_checks.sh doctests      # run doctests
+#   $ ./ci/code_checks.sh docstrings    # validate docstring errors
+#   $ ./ci/code_checks.sh single-docs   # check single-page docs build warning-free
+#   $ ./ci/code_checks.sh notebooks     # check execution of documentation notebooks
+
+set -uo pipefail
+
+if [[ -v 1 ]]; then
+    CHECK=$1
+else
+    # script will fail if it uses an unset variable (i.e. $1 is not provided)
+    CHECK=""
+fi
+
+[[ -z "$CHECK" || "$CHECK" == "code" || "$CHECK" == "doctests" || "$CHECK" == "docstrings" || "$CHECK" == "single-docs" || "$CHECK" == "notebooks" ]] || \
+    { echo "Unknown command $1. Usage: $0 [code|doctests|docstrings|single-docs|notebooks]"; exit 1; }
+
+BASE_DIR="$(dirname "$0")/.."
+RET=0
+
+### CODE ###
+if [[ -z "$CHECK" || "$CHECK" == "code" ]]; then
+
+    MSG='Check import. No warnings, and blocklist some optional dependencies' ; echo "$MSG"
+    python -W error -c "
+import sys
+import pandas
+
+blocklist = {'bs4', 'gcsfs', 'html5lib', 'http', 'ipython', 'jinja2', 'hypothesis',
+             'lxml', 'matplotlib', 'openpyxl', 'py', 'pytest', 's3fs', 'scipy',
+             'tables', 'urllib.request', 'xlrd', 'xlsxwriter'}
+
+# GH#28227 for some of these check for top-level modules, while others are
+#  more specific (e.g. urllib.request)
+import_mods = set(m.split('.')[0] for m in sys.modules) | set(sys.modules)
+mods = blocklist & import_mods
+if mods:
+    sys.stderr.write('err: pandas should not import: {}\n'.format(', '.join(mods)))
+    sys.exit(len(mods))
+    "
+    RET=$(($RET + $?)) ; echo "$MSG" "DONE"
+
+fi
+
+### DOCTESTS ###
+if [[ -z "$CHECK" || "$CHECK" == "doctests" ]]; then
+
+    MSG='Python and Cython Doctests' ; echo "$MSG"
+    python -c 'import pandas as pd; pd.test(run_doctests=True)'
+    RET=$(($RET + $?)) ; echo "$MSG" "DONE"
+
+fi
+
+### DOCSTRINGS ###
+if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
+
+    MSG='Validate Docstrings' ; echo "$MSG"
+    "$BASE_DIR"/scripts/validate_docstrings.py \
+        --format=actions \
+        -i ES01 `# For now it is ok if docstrings are missing the extended summary` \
+        -i "pandas.Series.dt PR01" `# Accessors are implemented as classes, but we do not document the Parameters section` \
+        -i "pandas.Period.freq GL08" \
+        -i "pandas.Period.ordinal GL08" \
+        -i "pandas.Timedelta.max PR02" \
+        -i "pandas.Timedelta.min PR02" \
+        -i "pandas.Timedelta.resolution PR02" \
+        -i "pandas.Timestamp.max PR02" \
+        -i "pandas.Timestamp.min PR02" \
+        -i "pandas.Timestamp.resolution PR02" \
+        -i "pandas.Timestamp.tzinfo GL08" \
+        -i "pandas.core.groupby.DataFrameGroupBy.plot PR02" \
+        -i "pandas.core.groupby.SeriesGroupBy.plot PR02" \
+        -i "pandas.core.resample.Resampler.quantile PR01,PR07" \
+        -i "pandas.tseries.offsets.BDay PR02,SA01" \
+        -i "pandas.tseries.offsets.BQuarterBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BQuarterBegin.n GL08" \
+        -i "pandas.tseries.offsets.BQuarterBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.BQuarterBegin.rule_code GL08" \
+        -i "pandas.tseries.offsets.BQuarterBegin.startingMonth GL08" \
+        -i "pandas.tseries.offsets.BQuarterEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BQuarterEnd.n GL08" \
+        -i "pandas.tseries.offsets.BQuarterEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.BQuarterEnd.rule_code GL08" \
+        -i "pandas.tseries.offsets.BQuarterEnd.startingMonth GL08" \
+        -i "pandas.tseries.offsets.BYearBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BYearBegin.month GL08" \
+        -i "pandas.tseries.offsets.BYearBegin.n GL08" \
+        -i "pandas.tseries.offsets.BYearBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.BYearEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BYearEnd.month GL08" \
+        -i "pandas.tseries.offsets.BYearEnd.n GL08" \
+        -i "pandas.tseries.offsets.BYearEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.BusinessDay PR02,SA01" \
+        -i "pandas.tseries.offsets.BusinessDay.calendar GL08" \
+        -i "pandas.tseries.offsets.BusinessDay.holidays GL08" \
+        -i "pandas.tseries.offsets.BusinessDay.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BusinessDay.n GL08" \
+        -i "pandas.tseries.offsets.BusinessDay.normalize GL08" \
+        -i "pandas.tseries.offsets.BusinessDay.weekmask GL08" \
+        -i "pandas.tseries.offsets.BusinessHour PR02,SA01" \
+        -i "pandas.tseries.offsets.BusinessHour.calendar GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.end GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.holidays GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.n GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.normalize GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.start GL08" \
+        -i "pandas.tseries.offsets.BusinessHour.weekmask GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthBegin.n GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthEnd.n GL08" \
+        -i "pandas.tseries.offsets.BusinessMonthEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.CBMonthBegin PR02" \
+        -i "pandas.tseries.offsets.CBMonthEnd PR02" \
+        -i "pandas.tseries.offsets.CDay PR02,SA01" \
+        -i "pandas.tseries.offsets.CustomBusinessDay PR02,SA01" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.calendar GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.holidays GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.n GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.normalize GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessDay.weekmask GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour PR02,SA01" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.calendar GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.end GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.holidays GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.n GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.normalize GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.start GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessHour.weekmask GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin PR02" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.calendar GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.holidays GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.m_offset GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.n GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthBegin.weekmask GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd PR02" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.calendar GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.holidays GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.m_offset GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.n GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.CustomBusinessMonthEnd.weekmask GL08" \
+        -i "pandas.tseries.offsets.DateOffset.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.DateOffset.n GL08" \
+        -i "pandas.tseries.offsets.DateOffset.normalize GL08" \
+        -i "pandas.tseries.offsets.Day.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Day.n GL08" \
+        -i "pandas.tseries.offsets.Day.normalize GL08" \
+        -i "pandas.tseries.offsets.Easter.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Easter.n GL08" \
+        -i "pandas.tseries.offsets.Easter.normalize GL08" \
+        -i "pandas.tseries.offsets.FY5253.get_rule_code_suffix GL08" \
+        -i "pandas.tseries.offsets.FY5253.get_year_end GL08" \
+        -i "pandas.tseries.offsets.FY5253.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.FY5253.n GL08" \
+        -i "pandas.tseries.offsets.FY5253.normalize GL08" \
+        -i "pandas.tseries.offsets.FY5253.rule_code GL08" \
+        -i "pandas.tseries.offsets.FY5253.startingMonth GL08" \
+        -i "pandas.tseries.offsets.FY5253.variation GL08" \
+        -i "pandas.tseries.offsets.FY5253.weekday GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.get_rule_code_suffix GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.get_weeks GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.n GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.normalize GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.qtr_with_extra_week GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.rule_code GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.startingMonth GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.variation GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.weekday GL08" \
+        -i "pandas.tseries.offsets.FY5253Quarter.year_has_extra_week GL08" \
+        -i "pandas.tseries.offsets.Hour.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Hour.n GL08" \
+        -i "pandas.tseries.offsets.Hour.normalize GL08" \
+        -i "pandas.tseries.offsets.LastWeekOfMonth.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.LastWeekOfMonth.n GL08" \
+        -i "pandas.tseries.offsets.LastWeekOfMonth.normalize GL08" \
+        -i "pandas.tseries.offsets.LastWeekOfMonth.week GL08" \
+        -i "pandas.tseries.offsets.LastWeekOfMonth.weekday GL08" \
+        -i "pandas.tseries.offsets.Micro.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Micro.n GL08" \
+        -i "pandas.tseries.offsets.Micro.normalize GL08" \
+        -i "pandas.tseries.offsets.Milli.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Milli.n GL08" \
+        -i "pandas.tseries.offsets.Milli.normalize GL08" \
+        -i "pandas.tseries.offsets.Minute.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Minute.n GL08" \
+        -i "pandas.tseries.offsets.Minute.normalize GL08" \
+        -i "pandas.tseries.offsets.MonthBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.MonthBegin.n GL08" \
+        -i "pandas.tseries.offsets.MonthBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.MonthEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.MonthEnd.n GL08" \
+        -i "pandas.tseries.offsets.MonthEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.Nano.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Nano.normalize GL08" \
+        -i "pandas.tseries.offsets.Nano.n GL08" \
+        -i "pandas.tseries.offsets.QuarterBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.QuarterBegin.n GL08" \
+        -i "pandas.tseries.offsets.QuarterBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.QuarterBegin.rule_code GL08" \
+        -i "pandas.tseries.offsets.QuarterBegin.startingMonth GL08" \
+        -i "pandas.tseries.offsets.QuarterEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.QuarterEnd.n GL08" \
+        -i "pandas.tseries.offsets.QuarterEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.QuarterEnd.rule_code GL08" \
+        -i "pandas.tseries.offsets.QuarterEnd.startingMonth GL08" \
+        -i "pandas.tseries.offsets.Second.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Second.n GL08" \
+        -i "pandas.tseries.offsets.Second.normalize GL08" \
+        -i "pandas.tseries.offsets.SemiMonthBegin.day_of_month GL08" \
+        -i "pandas.tseries.offsets.SemiMonthBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.SemiMonthBegin.n GL08" \
+        -i "pandas.tseries.offsets.SemiMonthBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.SemiMonthBegin.rule_code GL08" \
+        -i "pandas.tseries.offsets.SemiMonthEnd.day_of_month GL08" \
+        -i "pandas.tseries.offsets.SemiMonthEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.SemiMonthEnd.n GL08" \
+        -i "pandas.tseries.offsets.SemiMonthEnd.normalize GL08" \
+        -i "pandas.tseries.offsets.SemiMonthEnd.rule_code GL08" \
+        -i "pandas.tseries.offsets.Tick GL08" \
+        -i "pandas.tseries.offsets.Tick.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Tick.n GL08" \
+        -i "pandas.tseries.offsets.Tick.normalize GL08" \
+        -i "pandas.tseries.offsets.Week.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.Week.n GL08" \
+        -i "pandas.tseries.offsets.Week.normalize GL08" \
+        -i "pandas.tseries.offsets.Week.weekday GL08" \
+        -i "pandas.tseries.offsets.WeekOfMonth.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.WeekOfMonth.n GL08" \
+        -i "pandas.tseries.offsets.WeekOfMonth.normalize GL08" \
+        -i "pandas.tseries.offsets.WeekOfMonth.week GL08" \
+        -i "pandas.tseries.offsets.WeekOfMonth.weekday GL08" \
+        -i "pandas.tseries.offsets.YearBegin.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.YearBegin.month GL08" \
+        -i "pandas.tseries.offsets.YearBegin.n GL08" \
+        -i "pandas.tseries.offsets.YearBegin.normalize GL08" \
+        -i "pandas.tseries.offsets.YearEnd.is_on_offset GL08" \
+        -i "pandas.tseries.offsets.YearEnd.month GL08" \
+        -i "pandas.tseries.offsets.YearEnd.n GL08" \
+        -i "pandas.tseries.offsets.YearEnd.normalize GL08" \
+        -i "pandas.util.hash_pandas_object PR07,SA01" # There should be no backslash in the final line, please keep this comment in the last ignored function
+
+    RET=$(($RET + $?)) ; echo $MSG "DONE"
+
+fi
+
+### DOCUMENTATION NOTEBOOKS ###
+if [[ -z "$CHECK" || "$CHECK" == "notebooks" ]]; then
+
+    MSG='Notebooks' ; echo $MSG
+    jupyter nbconvert --execute "$(find doc/source -name '*.ipynb')" --to notebook
+    RET=$(($RET + $?)) ; echo $MSG "DONE"
+
+fi
+
+### SINGLE-PAGE DOCS ###
+if [[ -z "$CHECK" || "$CHECK" == "single-docs" ]]; then
+    python doc/make.py --warnings-are-errors --no-browser --single pandas.Series.value_counts
+    python doc/make.py --warnings-are-errors --no-browser --single pandas.Series.str.split
+fi
+
+exit $RET
diff --git a/pandas/ci/deps/actions-310-minimum_versions.yaml b/pandas/ci/deps/actions-310-minimum_versions.yaml
new file mode 100644
index 0000000000000..c7c72828db481
--- /dev/null
+++ b/pandas/ci/deps/actions-310-minimum_versions.yaml
@@ -0,0 +1,65 @@
+# Minimum version of required + optional dependencies
+# Aligned with getting_started/install.rst and compat/_optional.py
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.10
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-localserver>=0.8.1
+  - pytest-qt>=4.4.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil=2.8.2
+  - numpy=1.23.5
+
+  # optional dependencies
+  - beautifulsoup4=4.11.2
+  - blosc=1.21.3
+  - bottleneck=1.3.6
+  - fastparquet=2023.10.0
+  - fsspec=2022.11.0
+  - html5lib=1.1
+  - hypothesis=6.84.0
+  - gcsfs=2022.11.0
+  - jinja2=3.1.2
+  - lxml=4.9.2
+  - matplotlib=3.6.3
+  - numba=0.56.4
+  - numexpr=2.8.4
+  - odfpy=1.4.1
+  - qtpy=2.3.0
+  - openpyxl=3.1.0
+  - psycopg2=2.9.6
+  - pyarrow=10.0.1
+  - pymysql=1.0.2
+  - pyqt=5.15.9
+  - pyreadstat=1.2.0
+  - pytables=3.8.0
+  - python-calamine=0.1.7
+  - pytz=2023.4
+  - pyxlsb=1.0.10
+  - s3fs=2022.11.0
+  - scipy=1.10.0
+  - sqlalchemy=2.0.0
+  - tabulate=0.9.0
+  - xarray=2022.12.0
+  - xlrd=2.0.1
+  - xlsxwriter=3.0.5
+  - zstandard=0.19.0
+
+  - pip:
+    - adbc-driver-postgresql==0.10.0
+    - adbc-driver-sqlite==0.8.0
+    - tzdata==2022.7
diff --git a/pandas/ci/deps/actions-310.yaml b/pandas/ci/deps/actions-310.yaml
new file mode 100644
index 0000000000000..74cab4e0970dc
--- /dev/null
+++ b/pandas/ci/deps/actions-310.yaml
@@ -0,0 +1,63 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.10
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-localserver>=0.8.1
+  - pytest-qt>=4.4.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+
+  # optional dependencies
+  - beautifulsoup4>=4.11.2
+  - blosc>=1.21.3
+  - bottleneck>=1.3.6
+  - fastparquet>=2023.10.0
+  - fsspec>=2022.11.0
+  - html5lib>=1.1
+  - hypothesis>=6.84.0
+  - gcsfs>=2022.11.0
+  - jinja2>=3.1.2
+  - lxml>=4.9.2
+  - matplotlib>=3.6.3
+  - numba>=0.56.4
+  - numexpr>=2.8.4
+  - odfpy>=1.4.1
+  - qtpy>=2.3.0
+  - openpyxl>=3.1.0
+  - psycopg2>=2.9.6
+  - pyarrow>=10.0.1
+  - pymysql>=1.0.2
+  - pyqt>=5.15.9
+  - pyreadstat>=1.2.0
+  - pytables>=3.8.0
+  - python-calamine>=0.1.7
+  - pytz>=2023.4
+  - pyxlsb>=1.0.10
+  - s3fs>=2022.11.0
+  - scipy>=1.10.0
+  - sqlalchemy>=2.0.0
+  - tabulate>=0.9.0
+  - xarray>=2022.12.0, <=2024.9.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=3.0.5
+  - zstandard>=0.19.0
+
+  - pip:
+    - adbc-driver-postgresql>=0.10.0
+    - adbc-driver-sqlite>=0.8.0
+    - tzdata>=2022.7
diff --git a/pandas/ci/deps/actions-311-downstream_compat.yaml b/pandas/ci/deps/actions-311-downstream_compat.yaml
new file mode 100644
index 0000000000000..092ca18d61259
--- /dev/null
+++ b/pandas/ci/deps/actions-311-downstream_compat.yaml
@@ -0,0 +1,77 @@
+# Non-dependencies that pandas utilizes or has compatibility with pandas objects
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-localserver>=0.8.1
+  - pytest-qt>=4.4.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+
+  # optional dependencies
+  - beautifulsoup4>=4.11.2
+  - blosc>=1.21.3
+  - bottleneck>=1.3.6
+  - fastparquet>=2023.10.0
+  - fsspec>=2022.11.0
+  - html5lib>=1.1
+  - hypothesis>=6.84.0
+  - gcsfs>=2022.11.0
+  - jinja2>=3.1.2
+  - lxml>=4.9.2
+  - matplotlib>=3.6.3
+  - numba>=0.56.4
+  - numexpr>=2.8.4
+  - odfpy>=1.4.1
+  - qtpy>=2.3.0
+  - openpyxl>=3.1.0
+  - psycopg2>=2.9.6
+  - pyarrow>=10.0.1
+  - pymysql>=1.0.2
+  - pyqt>=5.15.9
+  - pyreadstat>=1.2.0
+  - pytables>=3.8.0
+  - python-calamine>=0.1.7
+  - pytz>=2023.4
+  - pyxlsb>=1.0.10
+  - s3fs>=2022.11.0
+  - scipy>=1.10.0
+  - sqlalchemy>=2.0.0
+  - tabulate>=0.9.0
+  - xarray>=2022.12.0, <=2024.9.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=3.0.5
+  - zstandard>=0.19.0
+
+  # downstream packages
+  - botocore
+  - cftime
+  - dask
+  - ipython
+  - geopandas-base
+  - seaborn
+  - scikit-learn
+  - statsmodels
+  - coverage
+  - pandas-datareader
+  - pyyaml
+  - py
+  - pip:
+    - adbc-driver-postgresql>=0.10.0
+    - adbc-driver-sqlite>=0.8.0
+    - tzdata>=2022.7
diff --git a/pandas/ci/deps/actions-311-numpydev.yaml b/pandas/ci/deps/actions-311-numpydev.yaml
new file mode 100644
index 0000000000000..325a6d45d74fd
--- /dev/null
+++ b/pandas/ci/deps/actions-311-numpydev.yaml
@@ -0,0 +1,27 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+
+  # build dependencies
+  - versioneer
+  - meson=1.2.1
+  - meson-python=0.13.1
+  - cython>=0.29.33
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - hypothesis>=6.84.0
+
+  # pandas dependencies
+  - python-dateutil
+  - pip
+
+  - pip:
+    - "--extra-index-url https://pypi.anaconda.org/scientific-python-nightly-wheels/simple"
+    - "--pre"
+    - "numpy"
+    - "tzdata>=2022.7"
diff --git a/pandas/ci/deps/actions-311-pyarrownightly.yaml b/pandas/ci/deps/actions-311-pyarrownightly.yaml
new file mode 100644
index 0000000000000..2d3d11c294e12
--- /dev/null
+++ b/pandas/ci/deps/actions-311-pyarrownightly.yaml
@@ -0,0 +1,29 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+
+  # build dependencies
+  - versioneer
+  - meson=1.2.1
+  - cython>=0.29.33
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - hypothesis>=6.84.0
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+  - pip
+
+  - pip:
+    - "tzdata>=2022.7"
+    - "--extra-index-url https://pypi.anaconda.org/scientific-python-nightly-wheels/simple"
+    - "--prefer-binary"
+    - "--pre"
+    - "pyarrow"
diff --git a/pandas/ci/deps/actions-311.yaml b/pandas/ci/deps/actions-311.yaml
new file mode 100644
index 0000000000000..b6f515dceaea9
--- /dev/null
+++ b/pandas/ci/deps/actions-311.yaml
@@ -0,0 +1,62 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-localserver>=0.8.1
+  - pytest-qt>=4.4.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+
+  # optional dependencies
+  - beautifulsoup4>=4.11.2
+  - blosc>=1.21.3
+  - bottleneck>=1.3.6
+  - fastparquet>=2023.10.0
+  - fsspec>=2022.11.0
+  - html5lib>=1.1
+  - hypothesis>=6.84.0
+  - gcsfs>=2022.11.0
+  - jinja2>=3.1.2
+  - lxml>=4.9.2
+  - matplotlib>=3.6.3
+  - numba>=0.56.4
+  - numexpr>=2.8.4
+  - odfpy>=1.4.1
+  - qtpy>=2.3.0
+  - pyqt>=5.15.9
+  - openpyxl>=3.1.0
+  - psycopg2>=2.9.6
+  - pyarrow>=10.0.1
+  - pymysql>=1.0.2
+  - pyreadstat>=1.2.0
+  - pytables>=3.8.0
+  - python-calamine>=0.1.7
+  - pytz>=2023.4
+  - pyxlsb>=1.0.10
+  - s3fs>=2022.11.0
+  - scipy>=1.10.0
+  - sqlalchemy>=2.0.0
+  - tabulate>=0.9.0
+  - xarray>=2022.12.0, <=2024.9.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=3.0.5
+  - zstandard>=0.19.0
+
+  - pip:
+    - adbc-driver-postgresql>=0.10.0
+    - adbc-driver-sqlite>=0.8.0
diff --git a/pandas/ci/deps/actions-312.yaml b/pandas/ci/deps/actions-312.yaml
new file mode 100644
index 0000000000000..bc66f8a5382c9
--- /dev/null
+++ b/pandas/ci/deps/actions-312.yaml
@@ -0,0 +1,63 @@
+name: pandas-dev-312
+channels:
+  - conda-forge
+dependencies:
+  - python=3.12
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-localserver>=0.8.1
+  - pytest-qt>=4.4.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+
+  # optional dependencies
+  - beautifulsoup4>=4.11.2
+  - blosc>=1.21.3
+  - bottleneck>=1.3.6
+  - fastparquet>=2023.10.0
+  - fsspec>=2022.11.0
+  - html5lib>=1.1
+  - hypothesis>=6.84.0
+  - gcsfs>=2022.11.0
+  - jinja2>=3.1.2
+  - lxml>=4.9.2
+  - matplotlib>=3.6.3
+  - numba>=0.56.4
+  - numexpr>=2.8.4
+  - odfpy>=1.4.1
+  - qtpy>=2.3.0
+  - pyqt>=5.15.9
+  - openpyxl>=3.1.0
+  - psycopg2>=2.9.6
+  - pyarrow>=10.0.1
+  - pymysql>=1.0.2
+  - pyreadstat>=1.2.0
+  - pytables>=3.8.0
+  - python-calamine>=0.1.7
+  - pytz>=2023.4
+  - pyxlsb>=1.0.10
+  - s3fs>=2022.11.0
+  - scipy>=1.10.0
+  - sqlalchemy>=2.0.0
+  - tabulate>=0.9.0
+  - xarray>=2022.12.0, <=2024.9.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=3.0.5
+  - zstandard>=0.19.0
+
+  - pip:
+    - adbc-driver-postgresql>=0.10.0
+    - adbc-driver-sqlite>=0.8.0
+    - tzdata>=2022.7
diff --git a/pandas/ci/deps/actions-pypy-39.yaml b/pandas/ci/deps/actions-pypy-39.yaml
new file mode 100644
index 0000000000000..90933b24b88db
--- /dev/null
+++ b/pandas/ci/deps/actions-pypy-39.yaml
@@ -0,0 +1,26 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  # TODO: Add the rest of the dependencies in here
+  # once the other plentiful failures/segfaults
+  # with base pandas has been dealt with
+  - python=3.9[build=*_pypy]
+
+  # build dependencies
+  - versioneer
+  - cython>=0.29.33
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - hypothesis>=6.84.0
+
+  # required
+  - numpy
+  - python-dateutil
+  - pip:
+    - tzdata>=2022.7
diff --git a/pandas/ci/meta.yaml b/pandas/ci/meta.yaml
new file mode 100644
index 0000000000000..9d434991b12c1
--- /dev/null
+++ b/pandas/ci/meta.yaml
@@ -0,0 +1,92 @@
+{% set version = "2.0.1" %}
+
+package:
+  name: pandas
+  version: {{ version }}
+
+source:
+  git_url: ../..
+
+build:
+  number: 1
+  script:
+    - export PYTHONUNBUFFERED=1  # [ppc64le]
+    - {{ PYTHON }} -m pip install -vv --no-deps --ignore-installed .  # [not unix]
+    - {{ PYTHON }} -m pip install -vv --no-deps --ignore-installed . --global-option="build_ext" --global-option="-j4" --no-use-pep517  # [unix]
+  skip: true  # [py<39]
+
+requirements:
+  build:
+    - python                                 # [build_platform != target_platform]
+    - cross-python_{{ target_platform }}     # [build_platform != target_platform]
+    - cython                                 # [build_platform != target_platform]
+    - numpy                                  # [build_platform != target_platform]
+    - {{ compiler('c') }}
+    - {{ compiler('cxx') }}
+  host:
+    - python
+    - pip
+    - setuptools >=61.0.0
+    - cython >=0.29.33,<3
+    - numpy >=1.21.6  # [py<311]
+    - numpy >=1.23.2  # [py>=311]
+    - versioneer
+    - tomli  # [py<311]
+  run:
+    - python
+    - numpy >=1.21.6  # [py<311]
+    - numpy >=1.23.2  # [py>=311]
+    - python-dateutil >=2.8.2
+    - python-tzdata >=2022.7
+
+test:
+  imports:
+    - pandas
+  commands:
+    - pip check
+    # Skip test suite on PyPy as it segfaults there
+    # xref: https://github.com/conda-forge/pandas-feedstock/issues/148
+    #
+    # Also skip `test_rolling_var_numerical_issues` on `ppc64le` as it is a known test failure.
+    # xref: https://github.com/conda-forge/pandas-feedstock/issues/149
+    {% set markers = ["not clipboard", "not single_cpu", "not db", "not network", "not slow"] %}
+    {% set markers = markers + ["not arm_slow"] %}   # [aarch64 or ppc64le]
+    {% set extra_args = ["-n=2 -m " + " and ".join(markers)] %}
+    {% set tests_to_skip = "_not_a_real_test" %}
+    {% set tests_to_skip = tests_to_skip + " or test_rolling_var_numerical_issues" %}           # [ppc64le]
+    {% set tests_to_skip = tests_to_skip + " or test_std_timedelta64_skipna_false" %}           # [ppc64le]
+    {% set tests_to_skip = tests_to_skip + " or test_value_counts_normalized[M8[ns]]" %}        # [ppc64le]
+    {% set tests_to_skip = tests_to_skip + " or test_to_datetime_format_YYYYMMDD_with_nat" %}   # [ppc64le]
+    {% set tests_to_skip = tests_to_skip + " or (TestReductions and test_median_2d)" %}         # [ppc64le]
+    {% set extra_args = extra_args + ["-k", "not (" + tests_to_skip + ")"] %}
+    - python -c "import pandas; pandas.test(extra_args={{ extra_args }})"  # [python_impl == "cpython"]
+  requires:
+    - pip
+    - pytest >=7.3.2
+    - pytest-xdist >=3.4.0
+    - pytest-cov
+    - hypothesis >=6.84.0
+    - tomli  # [py<311]
+
+about:
+  home: http://pandas.pydata.org
+  license: BSD-3-Clause
+  license_file: LICENSE
+  summary: Powerful data structures for data analysis, time series, and statistics
+  doc_url: https://pandas.pydata.org/docs/
+  dev_url: https://github.com/pandas-dev/pandas
+
+extra:
+  recipe-maintainers:
+    - jreback
+    - jorisvandenbossche
+    - msarahan
+    - ocefpaf
+    - TomAugspurger
+    - WillAyd
+    - simonjayhawkins
+    - mroeschke
+    - datapythonista
+    - phofl
+    - lithomas1
+    - marcogorelli
diff --git a/pandas/ci/run_tests.sh b/pandas/ci/run_tests.sh
new file mode 100755
index 0000000000000..16292beec612b
--- /dev/null
+++ b/pandas/ci/run_tests.sh
@@ -0,0 +1,18 @@
+#!/bin/bash -e
+
+# Workaround for pytest-xdist (it collects different tests in the workers if PYTHONHASHSEED is not set)
+# https://github.com/pytest-dev/pytest/issues/920
+# https://github.com/pytest-dev/pytest/issues/1075
+PYTHONHASHSEED=$(python -c 'import random; print(random.randint(1, 4294967295))')
+export PYTHONHASHSEED
+
+COVERAGE="-s --cov=pandas --cov-report=xml --cov-append --cov-config=pyproject.toml"
+
+PYTEST_CMD="MESONPY_EDITABLE_VERBOSE=1 PYTHONDEVMODE=1 PYTHONWARNDEFAULTENCODING=1 pytest -r fE -n $PYTEST_WORKERS --dist=worksteal $TEST_ARGS $COVERAGE $PYTEST_TARGET"
+
+if [[ "$PATTERN" ]]; then
+  PYTEST_CMD="$PYTEST_CMD -m \"$PATTERN\""
+fi
+
+echo "$PYTEST_CMD"
+sh -c "$PYTEST_CMD"
diff --git a/pandas/ci/upload_wheels.sh b/pandas/ci/upload_wheels.sh
new file mode 100644
index 0000000000000..c7c7ca00ee466
--- /dev/null
+++ b/pandas/ci/upload_wheels.sh
@@ -0,0 +1,42 @@
+#!/bin/bash
+# Modified from numpy's https://github.com/numpy/numpy/blob/main/tools/wheels/upload_wheels.sh
+
+set_upload_vars() {
+    echo "IS_PUSH is $IS_PUSH"
+    echo "IS_SCHEDULE_DISPATCH is $IS_SCHEDULE_DISPATCH"
+    if [[ "$IS_PUSH" == "true" ]]; then
+        echo push and tag event
+        export ANACONDA_ORG="multibuild-wheels-staging"
+        export TOKEN="$PANDAS_STAGING_UPLOAD_TOKEN"
+        export ANACONDA_UPLOAD="true"
+    elif [[ "$IS_SCHEDULE_DISPATCH" == "true" ]]; then
+        echo scheduled or dispatched event
+        export ANACONDA_ORG="scientific-python-nightly-wheels"
+        export TOKEN="$PANDAS_NIGHTLY_UPLOAD_TOKEN"
+        export ANACONDA_UPLOAD="true"
+    else
+        echo non-dispatch event
+        export ANACONDA_UPLOAD="false"
+    fi
+}
+upload_wheels() {
+    echo "${PWD}"
+    if [[ ${ANACONDA_UPLOAD} == true ]]; then
+        if [ -z "${TOKEN}" ]; then
+            echo no token set, not uploading
+        else
+            # sdists are located under dist folder when built through setup.py
+            if compgen -G "./dist/*.gz"; then
+                echo "Found sdist"
+                anaconda -q -t "${TOKEN}" upload --skip -u "${ANACONDA_ORG}" ./dist/*.gz
+                echo "Uploaded sdist"
+            fi
+            if compgen -G "./wheelhouse/*.whl"; then
+                echo "Found wheel"
+                anaconda -q -t "${TOKEN}" upload --skip -u "${ANACONDA_ORG}" ./wheelhouse/*.whl
+                echo "Uploaded wheel"
+            fi
+            echo "PyPI-style index: https://pypi.anaconda.org/$ANACONDA_ORG/simple"
+        fi
+    fi
+}
diff --git a/pandas/codecov.yml b/pandas/codecov.yml
new file mode 100644
index 0000000000000..d893bdbdc9298
--- /dev/null
+++ b/pandas/codecov.yml
@@ -0,0 +1,18 @@
+codecov:
+  branch: main
+  notify:
+    after_n_builds: 10
+comment: false
+
+coverage:
+  status:
+    project:
+      default:
+        target: '82'
+    patch:
+      default:
+        target: '50'
+        informational: true
+
+github_checks:
+    annotations: false
diff --git a/pandas/core/_numba/kernels/min_max_.py b/pandas/core/_numba/kernels/min_max_.py
index 68aa1446bbe3c..62763b73cda10 100644
--- a/pandas/core/_numba/kernels/min_max_.py
+++ b/pandas/core/_numba/kernels/min_max_.py
@@ -71,11 +71,10 @@ def sliding_min_max(
         # Save output based on index in input value array
         if Q and curr_win_size > 0 and nobs >= min_periods:
             output[i] = values[Q[0]]
+        elif values.dtype.kind != "i":
+            output[i] = np.nan
         else:
-            if values.dtype.kind != "i":
-                output[i] = np.nan
-            else:
-                na_pos.append(i)
+            na_pos.append(i)
 
     return output, na_pos
 
@@ -118,9 +117,8 @@ def grouped_min_max(
         if is_max:
             if val > output[lab]:
                 output[lab] = val
-        else:
-            if val < output[lab]:
-                output[lab] = val
+        elif val < output[lab]:
+            output[lab] = val
 
     # Set labels that don't satisfy min_periods as np.nan
     for lab, count in enumerate(nobs):
diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index aafd802b827a5..ac76805fd18cd 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -882,38 +882,35 @@ def value_counts_internal(
         # normalizing is by len of all (regardless of dropna)
         counts = np.array([len(ii)])
 
-    else:
-        if is_extension_array_dtype(values):
-            # handle Categorical and sparse,
-            result = Series(values, copy=False)._values.value_counts(dropna=dropna)
-            result.name = name
-            result.index.name = index_name
-            counts = result._values
-            if not isinstance(counts, np.ndarray):
-                # e.g. ArrowExtensionArray
-                counts = np.asarray(counts)
-
-        elif isinstance(values, ABCMultiIndex):
-            # GH49558
-            levels = list(range(values.nlevels))
-            result = (
-                Series(index=values, name=name)
-                .groupby(level=levels, dropna=dropna)
-                .size()
-            )
-            result.index.names = values.names
-            counts = result._values
+    elif is_extension_array_dtype(values):
+        # handle Categorical and sparse,
+        result = Series(values, copy=False)._values.value_counts(dropna=dropna)
+        result.name = name
+        result.index.name = index_name
+        counts = result._values
+        if not isinstance(counts, np.ndarray):
+            # e.g. ArrowExtensionArray
+            counts = np.asarray(counts)
 
-        else:
-            values = _ensure_arraylike(values, func_name="value_counts")
-            keys, counts, _ = value_counts_arraylike(values, dropna)
-            if keys.dtype == np.float16:
-                keys = keys.astype(np.float32)
-
-            # Starting in 3.0, we no longer perform dtype inference on the
-            #  Index object we construct here, xref GH#56161
-            idx = Index(keys, dtype=keys.dtype, name=index_name)
-            result = Series(counts, index=idx, name=name, copy=False)
+    elif isinstance(values, ABCMultiIndex):
+        # GH49558
+        levels = list(range(values.nlevels))
+        result = (
+            Series(index=values, name=name).groupby(level=levels, dropna=dropna).size()
+        )
+        result.index.names = values.names
+        counts = result._values
+
+    else:
+        values = _ensure_arraylike(values, func_name="value_counts")
+        keys, counts, _ = value_counts_arraylike(values, dropna)
+        if keys.dtype == np.float16:
+            keys = keys.astype(np.float32)
+
+        # Starting in 3.0, we no longer perform dtype inference on the
+        #  Index object we construct here, xref GH#56161
+        idx = Index(keys, dtype=keys.dtype, name=index_name)
+        result = Series(counts, index=idx, name=name, copy=False)
 
     if sort:
         result = result.sort_values(ascending=ascending)
diff --git a/pandas/core/arraylike.py b/pandas/core/arraylike.py
index 51ddd9e91b227..1371ae7afe855 100644
--- a/pandas/core/arraylike.py
+++ b/pandas/core/arraylike.py
@@ -397,19 +397,18 @@ def _reconstruct(result):
         # ufunc(series, ...)
         inputs = tuple(extract_array(x, extract_numpy=True) for x in inputs)
         result = getattr(ufunc, method)(*inputs, **kwargs)
+    # ufunc(dataframe)
+    elif method == "__call__" and not kwargs:
+        # for np.<ufunc>(..) calls
+        # kwargs cannot necessarily be handled block-by-block, so only
+        # take this path if there are no kwargs
+        mgr = inputs[0]._mgr  # pyright: ignore[reportGeneralTypeIssues]
+        result = mgr.apply(getattr(ufunc, method))
     else:
-        # ufunc(dataframe)
-        if method == "__call__" and not kwargs:
-            # for np.<ufunc>(..) calls
-            # kwargs cannot necessarily be handled block-by-block, so only
-            # take this path if there are no kwargs
-            mgr = inputs[0]._mgr  # pyright: ignore[reportGeneralTypeIssues]
-            result = mgr.apply(getattr(ufunc, method))
-        else:
-            # otherwise specific ufunc methods (eg np.<ufunc>.accumulate(..))
-            # Those can have an axis keyword and thus can't be called block-by-block
-            result = default_array_ufunc(inputs[0], ufunc, method, *inputs, **kwargs)  # pyright: ignore[reportGeneralTypeIssues]
-            # e.g. np.negative (only one reached), with "where" and "out" in kwargs
+        # otherwise specific ufunc methods (eg np.<ufunc>.accumulate(..))
+        # Those can have an axis keyword and thus can't be called block-by-block
+        result = default_array_ufunc(inputs[0], ufunc, method, *inputs, **kwargs)  # pyright: ignore[reportGeneralTypeIssues]
+        # e.g. np.negative (only one reached), with "where" and "out" in kwargs
 
     result = reconstruct(result)
     return result
diff --git a/pandas/core/arrays/_arrow_string_mixins.py b/pandas/core/arrays/_arrow_string_mixins.py
index 1ca52ce64bd77..5cd1d3067d948 100644
--- a/pandas/core/arrays/_arrow_string_mixins.py
+++ b/pandas/core/arrays/_arrow_string_mixins.py
@@ -216,16 +216,15 @@ def _str_startswith(
     ):
         if isinstance(pat, str):
             result = pc.starts_with(self._pa_array, pattern=pat)
+        elif len(pat) == 0:
+            # For empty tuple we return null for missing values and False
+            #  for valid values.
+            result = pc.if_else(pc.is_null(self._pa_array), None, False)
         else:
-            if len(pat) == 0:
-                # For empty tuple we return null for missing values and False
-                #  for valid values.
-                result = pc.if_else(pc.is_null(self._pa_array), None, False)
-            else:
-                result = pc.starts_with(self._pa_array, pattern=pat[0])
+            result = pc.starts_with(self._pa_array, pattern=pat[0])
 
-                for p in pat[1:]:
-                    result = pc.or_(result, pc.starts_with(self._pa_array, pattern=p))
+            for p in pat[1:]:
+                result = pc.or_(result, pc.starts_with(self._pa_array, pattern=p))
         return self._convert_bool_result(result, na=na, method_name="startswith")
 
     def _str_endswith(
@@ -233,16 +232,15 @@ def _str_endswith(
     ):
         if isinstance(pat, str):
             result = pc.ends_with(self._pa_array, pattern=pat)
+        elif len(pat) == 0:
+            # For empty tuple we return null for missing values and False
+            #  for valid values.
+            result = pc.if_else(pc.is_null(self._pa_array), None, False)
         else:
-            if len(pat) == 0:
-                # For empty tuple we return null for missing values and False
-                #  for valid values.
-                result = pc.if_else(pc.is_null(self._pa_array), None, False)
-            else:
-                result = pc.ends_with(self._pa_array, pattern=pat[0])
+            result = pc.ends_with(self._pa_array, pattern=pat[0])
 
-                for p in pat[1:]:
-                    result = pc.or_(result, pc.ends_with(self._pa_array, pattern=p))
+            for p in pat[1:]:
+                result = pc.or_(result, pc.ends_with(self._pa_array, pattern=p))
         return self._convert_bool_result(result, na=na, method_name="endswith")
 
     def _str_isalnum(self):
diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index 4e6f20e6ad3dd..6022e2108f250 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -320,11 +320,10 @@ def _pad_or_backfill(
             else:
                 new_values = self
 
+        elif copy:
+            new_values = self.copy()
         else:
-            if copy:
-                new_values = self.copy()
-            else:
-                new_values = self
+            new_values = self
         return new_values
 
     @doc(ExtensionArray.fillna)
diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index e2feda495c103..d5f05b0a8fbb9 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -1802,11 +1802,10 @@ def pyarrow_meth(data, skip_nulls, min_count=0):  # type: ignore[misc]
                         # binary_join returns null if there is any null ->
                         # have to filter out any nulls
                         data = data.filter(pc.invert(mask))
-                else:
-                    if mask is not None or check_below_min_count(
-                        (len(data),), None, min_count
-                    ):
-                        return pa.scalar(None, type=data.type)
+                elif mask is not None or check_below_min_count(
+                    (len(data),), None, min_count
+                ):
+                    return pa.scalar(None, type=data.type)
 
                 if pa.types.is_large_string(data.type):
                     # binary_join only supports string, not large_string
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 33745438e2aea..d5b45482d0b20 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -1250,11 +1250,10 @@ def fillna(
             else:
                 new_values = self.copy()
             new_values[mask] = value
+        elif not copy:
+            new_values = self[:]
         else:
-            if not copy:
-                new_values = self[:]
-            else:
-                new_values = self.copy()
+            new_values = self.copy()
         return new_values
 
     def dropna(self) -> Self:
diff --git a/pandas/core/arrays/boolean.py b/pandas/core/arrays/boolean.py
index 87c18fe346c62..14a592f9dd632 100644
--- a/pandas/core/arrays/boolean.py
+++ b/pandas/core/arrays/boolean.py
@@ -226,17 +226,15 @@ def coerce_to_array(
         mask = np.zeros(values.shape, dtype=bool)
     elif mask is None:
         mask = mask_values
+    elif isinstance(mask, np.ndarray) and mask.dtype == np.bool_:
+        if mask_values is not None:
+            mask = mask | mask_values
+        elif copy:
+            mask = mask.copy()
     else:
-        if isinstance(mask, np.ndarray) and mask.dtype == np.bool_:
-            if mask_values is not None:
-                mask = mask | mask_values
-            else:
-                if copy:
-                    mask = mask.copy()
-        else:
-            mask = np.array(mask, dtype=bool)
-            if mask_values is not None:
-                mask = mask | mask_values
+        mask = np.array(mask, dtype=bool)
+        if mask_values is not None:
+            mask = mask | mask_values
 
     if values.shape != mask.shape:
         raise ValueError("values.shape and mask.shape must match")
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 8a79ab53442c3..76d0e07c0569a 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -1675,10 +1675,9 @@ def _groupby_op(
                     f"'{how}' with PeriodDtype is no longer supported. "
                     f"Use (obj != pd.Period(0, freq)).{how}() instead."
                 )
-        else:
-            # timedeltas we can add but not multiply
-            if how in ["prod", "cumprod", "skew", "kurt", "var"]:
-                raise TypeError(f"timedelta64 type does not support {how} operations")
+        # timedeltas we can add but not multiply
+        elif how in ["prod", "cumprod", "skew", "kurt", "var"]:
+            raise TypeError(f"timedelta64 type does not support {how} operations")
 
         # All of the functions implemented here are ordinal, so we can
         #  operate on the tz-naive equivalents
diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index f3a0cc0dccdb3..e0386579fccc6 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -222,11 +222,10 @@ def _pad_or_backfill(
                 return self._simple_new(npvalues.T, new_mask.T)
             else:
                 return self
+        elif copy:
+            new_values = self.copy()
         else:
-            if copy:
-                new_values = self.copy()
-            else:
-                new_values = self
+            new_values = self
         return new_values
 
     @doc(ExtensionArray.fillna)
@@ -248,11 +247,10 @@ def fillna(self, value, limit: int | None = None, copy: bool = True) -> Self:
             else:
                 new_values = self[:]
             new_values[mask] = value
+        elif copy:
+            new_values = self.copy()
         else:
-            if copy:
-                new_values = self.copy()
-            else:
-                new_values = self[:]
+            new_values = self[:]
         return new_values
 
     @classmethod
@@ -279,9 +277,8 @@ def _validate_setitem_value(self, value):
             if lib.is_integer(value) or lib.is_float(value):
                 return value
 
-        else:
-            if lib.is_integer(value) or (lib.is_float(value) and value.is_integer()):
-                return value
+        elif lib.is_integer(value) or (lib.is_float(value) and value.is_integer()):
+            return value
             # TODO: unsigned checks
 
         # Note: without the "str" here, the f-string rendering raises in
@@ -1413,11 +1410,10 @@ def any(
         result = values.any()
         if skipna:
             return result
+        elif result or len(self) == 0 or not self._mask.any():
+            return result
         else:
-            if result or len(self) == 0 or not self._mask.any():
-                return result
-            else:
-                return self.dtype.na_value
+            return self.dtype.na_value
 
     @overload
     def all(
@@ -1502,11 +1498,10 @@ def all(
 
         if skipna:
             return result
+        elif not result or len(self) == 0 or not self._mask.any():
+            return result
         else:
-            if not result or len(self) == 0 or not self._mask.any():
-                return result
-            else:
-                return self.dtype.na_value
+            return self.dtype.na_value
 
     def interpolate(
         self,
diff --git a/pandas/core/arrays/string_.py b/pandas/core/arrays/string_.py
index 7227ea77ca433..da600ff08e079 100644
--- a/pandas/core/arrays/string_.py
+++ b/pandas/core/arrays/string_.py
@@ -674,11 +674,10 @@ def _from_sequence(
         if dtype and not (isinstance(dtype, str) and dtype == "string"):
             dtype = pandas_dtype(dtype)
             assert isinstance(dtype, StringDtype) and dtype.storage == "python"
+        elif using_string_dtype():
+            dtype = StringDtype(storage="python", na_value=np.nan)
         else:
-            if using_string_dtype():
-                dtype = StringDtype(storage="python", na_value=np.nan)
-            else:
-                dtype = StringDtype(storage="python")
+            dtype = StringDtype(storage="python")
 
         from pandas.core.arrays.masked import BaseMaskedArray
 
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index d35083fd892a8..8ab4862b1cea7 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -247,9 +247,8 @@ def _convert_bool_result(self, values, na=lib.no_default, method_name=None):
             else:
                 values = values.fill_null(na)
             return values.to_numpy()
-        else:
-            if na is not lib.no_default and not isna(na):  # pyright: ignore [reportGeneralTypeIssues]
-                values = values.fill_null(na)
+        elif na is not lib.no_default and not isna(na):  # pyright: ignore [reportGeneralTypeIssues]
+            values = values.fill_null(na)
         return BooleanDtype().__from_arrow__(values)
 
     def _maybe_convert_setitem_value(self, value):
diff --git a/pandas/core/dtypes/cast.py b/pandas/core/dtypes/cast.py
index f11aefeeaaa00..de97f0525f305 100644
--- a/pandas/core/dtypes/cast.py
+++ b/pandas/core/dtypes/cast.py
@@ -379,9 +379,8 @@ def trans(x):
                 # np.allclose may raise TypeError on object-dtype
                 if (new_result == result).all():
                     return new_result
-            else:
-                if np.allclose(new_result, result, rtol=0):
-                    return new_result
+            elif np.allclose(new_result, result, rtol=0):
+                return new_result
 
     elif (
         issubclass(dtype.type, np.floating)
diff --git a/pandas/core/dtypes/missing.py b/pandas/core/dtypes/missing.py
index f20ca44728664..7db972cfd3f85 100644
--- a/pandas/core/dtypes/missing.py
+++ b/pandas/core/dtypes/missing.py
@@ -263,13 +263,12 @@ def _isna_string_dtype(values: np.ndarray) -> npt.NDArray[np.bool_]:
 
     if dtype.kind in ("S", "U"):
         result = np.zeros(values.shape, dtype=bool)
+    elif values.ndim in {1, 2}:
+        result = libmissing.isnaobj(values)
     else:
-        if values.ndim in {1, 2}:
-            result = libmissing.isnaobj(values)
-        else:
-            # 0-D, reached via e.g. mask_missing
-            result = libmissing.isnaobj(values.ravel())
-            result = result.reshape(values.shape)
+        # 0-D, reached via e.g. mask_missing
+        result = libmissing.isnaobj(values.ravel())
+        result = result.reshape(values.shape)
 
     return result
 
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 4a86048bc20e2..257a907a28ab2 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4205,29 +4205,28 @@ def _setitem_array(self, key, value) -> None:
                 value = value.reindex(self.index.take(indexer))
             self.iloc[indexer] = value
 
-        else:
-            # Note: unlike self.iloc[:, indexer] = value, this will
-            #  never try to overwrite values inplace
+        # Note: unlike self.iloc[:, indexer] = value, this will
+        #  never try to overwrite values inplace
 
-            if isinstance(value, DataFrame):
-                check_key_length(self.columns, key, value)
-                for k1, k2 in zip(key, value.columns):
-                    self[k1] = value[k2]
+        elif isinstance(value, DataFrame):
+            check_key_length(self.columns, key, value)
+            for k1, k2 in zip(key, value.columns):
+                self[k1] = value[k2]
 
-            elif not is_list_like(value):
-                for col in key:
-                    self[col] = value
+        elif not is_list_like(value):
+            for col in key:
+                self[col] = value
 
-            elif isinstance(value, np.ndarray) and value.ndim == 2:
-                self._iset_not_inplace(key, value)
+        elif isinstance(value, np.ndarray) and value.ndim == 2:
+            self._iset_not_inplace(key, value)
 
-            elif np.ndim(value) > 1:
-                # list of lists
-                value = DataFrame(value).values
-                self._setitem_array(key, value)
+        elif np.ndim(value) > 1:
+            # list of lists
+            value = DataFrame(value).values
+            self._setitem_array(key, value)
 
-            else:
-                self._iset_not_inplace(key, value)
+        else:
+            self._iset_not_inplace(key, value)
 
     def _iset_not_inplace(self, key, value) -> None:
         # GH#39510 when setting with df[key] = obj with a list-like key and
@@ -7094,11 +7093,10 @@ def sort_values(
             indexer = nargsort(
                 k, kind=kind, ascending=ascending, na_position=na_position, key=key
             )
+        elif inplace:
+            return self._update_inplace(self)
         else:
-            if inplace:
-                return self._update_inplace(self)
-            else:
-                return self.copy(deep=False)
+            return self.copy(deep=False)
 
         if is_range_indexer(indexer, len(indexer)):
             result = self.copy(deep=False)
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 0c3f535df9ce2..4b4e8e5b2ce82 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -1026,12 +1026,11 @@ def _rename(
                 raise TypeError(
                     "Cannot specify both 'mapper' and any of 'index' or 'columns'"
                 )
+        # use the mapper argument
+        elif axis and self._get_axis_number(axis) == 1:
+            columns = mapper
         else:
-            # use the mapper argument
-            if axis and self._get_axis_number(axis) == 1:
-                columns = mapper
-            else:
-                index = mapper
+            index = mapper
 
         self._check_inplace_and_allows_duplicate_labels(inplace)
         result = self if inplace else self.copy(deep=False)
@@ -5371,11 +5370,10 @@ def reindex(
                     columns = labels
                 else:
                     index = labels
+        elif axis and self._get_axis_number(axis) == 1:
+            columns = labels
         else:
-            if axis and self._get_axis_number(axis) == 1:
-                columns = labels
-            else:
-                index = labels
+            index = labels
         axes: dict[Literal["index", "columns"], Any] = {
             "index": index,
             "columns": columns,
@@ -7089,40 +7087,37 @@ def fillna(
 
                 if not inplace:
                     result[k] = res_k
+                # We can write into our existing column(s) iff dtype
+                #  was preserved.
+                elif isinstance(res_k, ABCSeries):
+                    # i.e. 'k' only shows up once in self.columns
+                    if res_k.dtype == result[k].dtype:
+                        result.loc[:, k] = res_k
+                    else:
+                        # Different dtype -> no way to do inplace.
+                        result[k] = res_k
                 else:
-                    # We can write into our existing column(s) iff dtype
-                    #  was preserved.
-                    if isinstance(res_k, ABCSeries):
-                        # i.e. 'k' only shows up once in self.columns
-                        if res_k.dtype == result[k].dtype:
-                            result.loc[:, k] = res_k
+                    # see test_fillna_dict_inplace_nonunique_columns
+                    locs = result.columns.get_loc(k)
+                    if isinstance(locs, slice):
+                        locs = range(self.shape[1])[locs]
+                    elif isinstance(locs, np.ndarray) and locs.dtype.kind == "b":
+                        locs = locs.nonzero()[0]
+                    elif not (isinstance(locs, np.ndarray) and locs.dtype.kind == "i"):
+                        # Should never be reached, but let's cover our bases
+                        raise NotImplementedError(
+                            "Unexpected get_loc result, please report a bug at "
+                            "https://github.com/pandas-dev/pandas"
+                        )
+
+                    for i, loc in enumerate(locs):
+                        res_loc = res_k.iloc[:, i]
+                        target = self.iloc[:, loc]
+
+                        if res_loc.dtype == target.dtype:
+                            result.iloc[:, loc] = res_loc
                         else:
-                            # Different dtype -> no way to do inplace.
-                            result[k] = res_k
-                    else:
-                        # see test_fillna_dict_inplace_nonunique_columns
-                        locs = result.columns.get_loc(k)
-                        if isinstance(locs, slice):
-                            locs = range(self.shape[1])[locs]
-                        elif isinstance(locs, np.ndarray) and locs.dtype.kind == "b":
-                            locs = locs.nonzero()[0]
-                        elif not (
-                            isinstance(locs, np.ndarray) and locs.dtype.kind == "i"
-                        ):
-                            # Should never be reached, but let's cover our bases
-                            raise NotImplementedError(
-                                "Unexpected get_loc result, please report a bug at "
-                                "https://github.com/pandas-dev/pandas"
-                            )
-
-                        for i, loc in enumerate(locs):
-                            res_loc = res_k.iloc[:, i]
-                            target = self.iloc[:, loc]
-
-                            if res_loc.dtype == target.dtype:
-                                result.iloc[:, loc] = res_loc
-                            else:
-                                result.isetitem(loc, res_loc)
+                            result.isetitem(loc, res_loc)
             if inplace:
                 return self._update_inplace(result)
             else:
@@ -7605,34 +7600,32 @@ def replace(
                         f"you passed a {type(regex).__name__!r}"
                     )
                 return self.replace(regex, value, inplace=inplace, regex=True)
-            else:
-                # dest iterable dict-like
-                if is_dict_like(value):  # NA -> {'A' : 0, 'B' : -1}
-                    # Operate column-wise
-                    if self.ndim == 1:
-                        raise ValueError(
-                            "Series.replace cannot use dict-value and "
-                            "non-None to_replace"
-                        )
-                    mapping = {col: (to_replace, val) for col, val in value.items()}
-                    return self._replace_columnwise(mapping, inplace, regex)
-
-                elif not is_list_like(value):  # NA -> 0
-                    regex = should_use_regex(regex, to_replace)
-                    if regex:
-                        new_data = self._mgr.replace_regex(
-                            to_replace=to_replace,
-                            value=value,
-                            inplace=inplace,
-                        )
-                    else:
-                        new_data = self._mgr.replace(
-                            to_replace=to_replace, value=value, inplace=inplace
-                        )
+            # dest iterable dict-like
+            elif is_dict_like(value):  # NA -> {'A' : 0, 'B' : -1}
+                # Operate column-wise
+                if self.ndim == 1:
+                    raise ValueError(
+                        "Series.replace cannot use dict-value and non-None to_replace"
+                    )
+                mapping = {col: (to_replace, val) for col, val in value.items()}
+                return self._replace_columnwise(mapping, inplace, regex)
+
+            elif not is_list_like(value):  # NA -> 0
+                regex = should_use_regex(regex, to_replace)
+                if regex:
+                    new_data = self._mgr.replace_regex(
+                        to_replace=to_replace,
+                        value=value,
+                        inplace=inplace,
+                    )
                 else:
-                    raise TypeError(
-                        f'Invalid "to_replace" type: {type(to_replace).__name__!r}'
+                    new_data = self._mgr.replace(
+                        to_replace=to_replace, value=value, inplace=inplace
                     )
+            else:
+                raise TypeError(
+                    f'Invalid "to_replace" type: {type(to_replace).__name__!r}'
+                )
 
         result = self._constructor_from_mgr(new_data, axes=new_data.axes)
         if inplace:
diff --git a/pandas/core/groupby/grouper.py b/pandas/core/groupby/grouper.py
index c9d874fc08dbe..69cfb15944e0b 100644
--- a/pandas/core/groupby/grouper.py
+++ b/pandas/core/groupby/grouper.py
@@ -366,9 +366,8 @@ def _set_grouper(
                     level = ax._get_level_number(level)
                     ax = Index(ax._get_level_values(level), name=ax.names[level])
 
-                else:
-                    if level not in (0, ax.name):
-                        raise ValueError(f"The level {level} is not valid")
+                elif level not in (0, ax.name):
+                    raise ValueError(f"The level {level} is not valid")
 
         # possibly sort
         indexer: npt.NDArray[np.intp] | None = None
diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index c4c7f73ee166c..8ec10a39929dc 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -276,11 +276,10 @@ def _get_out_dtype(self, dtype: np.dtype) -> np.dtype:
             # The Cython implementation only produces the row number; we'll take
             # from the index using this in post processing
             out_dtype = "intp"
+        elif dtype.kind in "iufcb":
+            out_dtype = f"{dtype.kind}{dtype.itemsize}"
         else:
-            if dtype.kind in "iufcb":
-                out_dtype = f"{dtype.kind}{dtype.itemsize}"
-            else:
-                out_dtype = "object"
+            out_dtype = "object"
         return np.dtype(out_dtype)
 
     def _get_result_dtype(self, dtype: np.dtype) -> np.dtype:
diff --git a/pandas/core/indexers/utils.py b/pandas/core/indexers/utils.py
index b089be3469d87..e95f572dbe0c6 100644
--- a/pandas/core/indexers/utils.py
+++ b/pandas/core/indexers/utils.py
@@ -389,10 +389,9 @@ def check_key_length(columns: Index, key, value: DataFrame) -> None:
     if columns.is_unique:
         if len(value.columns) != len(key):
             raise ValueError("Columns must be same length as key")
-    else:
-        # Missing keys in columns are represented as -1
-        if len(columns.get_indexer_non_unique(key)[0]) != len(value.columns):
-            raise ValueError("Columns must be same length as key")
+    # Missing keys in columns are represented as -1
+    elif len(columns.get_indexer_non_unique(key)[0]) != len(value.columns):
+        raise ValueError("Columns must be same length as key")
 
 
 def unpack_tuple_and_ellipses(item: tuple):
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index c17c8c1d9a172..6518b324f0b87 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -3163,12 +3163,11 @@ def _union(self, other: Index, sort: bool | None):
             # Preserve MultiIndex to avoid losing dtypes
             result = self.append(other.take(missing))
 
+        elif len(missing) > 0:
+            other_diff = rvals.take(missing)
+            result = concat_compat((lvals, other_diff))
         else:
-            if len(missing) > 0:
-                other_diff = rvals.take(missing)
-                result = concat_compat((lvals, other_diff))
-            else:
-                result = lvals
+            result = lvals
 
         if not self.is_monotonic_increasing or not other.is_monotonic_increasing:
             # if both are monotonic then result should already be sorted
@@ -4187,21 +4186,19 @@ def reindex(
                 target, level, how="right", keep_order=not self._is_multi
             )
 
+        elif self.equals(target):
+            indexer = None
+        elif self._index_as_unique:
+            indexer = self.get_indexer(
+                target, method=method, limit=limit, tolerance=tolerance
+            )
+        elif self._is_multi:
+            raise ValueError("cannot handle a non-unique multi-index!")
+        elif not self.is_unique:
+            # GH#42568
+            raise ValueError("cannot reindex on an axis with duplicate labels")
         else:
-            if self.equals(target):
-                indexer = None
-            else:
-                if self._index_as_unique:
-                    indexer = self.get_indexer(
-                        target, method=method, limit=limit, tolerance=tolerance
-                    )
-                elif self._is_multi:
-                    raise ValueError("cannot handle a non-unique multi-index!")
-                elif not self.is_unique:
-                    # GH#42568
-                    raise ValueError("cannot reindex on an axis with duplicate labels")
-                else:
-                    indexer, _ = self.get_indexer_non_unique(target)
+            indexer, _ = self.get_indexer_non_unique(target)
 
         target = self._wrap_reindex_result(target, indexer, preserve_names)
         return target, indexer
@@ -4739,31 +4736,30 @@ def _get_leaf_sorter(labels: list[np.ndarray]) -> npt.NDArray[np.intp]:
                     new_codes = [lab[mask] for lab in new_codes]
                     left_indexer = left_indexer[mask]
 
-            else:  # tie out the order with other
-                if level == 0:  # outer most level, take the fast route
-                    max_new_lev = 0 if len(new_lev_codes) == 0 else new_lev_codes.max()
-                    ngroups = 1 + max_new_lev
-                    left_indexer, counts = libalgos.groupsort_indexer(
-                        new_lev_codes, ngroups
-                    )
+            elif level == 0:  # outer most level, take the fast route
+                max_new_lev = 0 if len(new_lev_codes) == 0 else new_lev_codes.max()
+                ngroups = 1 + max_new_lev
+                left_indexer, counts = libalgos.groupsort_indexer(
+                    new_lev_codes, ngroups
+                )
 
-                    # missing values are placed first; drop them!
-                    left_indexer = left_indexer[counts[0] :]
-                    new_codes = [lab[left_indexer] for lab in new_codes]
+                # missing values are placed first; drop them!
+                left_indexer = left_indexer[counts[0] :]
+                new_codes = [lab[left_indexer] for lab in new_codes]
 
-                else:  # sort the leaves
-                    mask = new_lev_codes != -1
-                    mask_all = mask.all()
-                    if not mask_all:
-                        new_codes = [lab[mask] for lab in new_codes]
+            else:  # sort the leaves
+                mask = new_lev_codes != -1
+                mask_all = mask.all()
+                if not mask_all:
+                    new_codes = [lab[mask] for lab in new_codes]
 
-                    left_indexer = _get_leaf_sorter(new_codes[: level + 1])
-                    new_codes = [lab[left_indexer] for lab in new_codes]
+                left_indexer = _get_leaf_sorter(new_codes[: level + 1])
+                new_codes = [lab[left_indexer] for lab in new_codes]
 
-                    # left_indexers are w.r.t masked frame.
-                    # reverse to original frame!
-                    if not mask_all:
-                        left_indexer = mask.nonzero()[0][left_indexer]
+                # left_indexers are w.r.t masked frame.
+                # reverse to original frame!
+                if not mask_all:
+                    left_indexer = mask.nonzero()[0][left_indexer]
 
             join_index = MultiIndex(
                 levels=new_levels,
@@ -6727,11 +6723,10 @@ def get_slice_bound(self, label, side: Literal["left", "right"]) -> int:
                 return slc.start
             else:
                 return slc.stop
+        elif side == "right":
+            return slc + 1
         else:
-            if side == "right":
-                return slc + 1
-            else:
-                return slc
+            return slc
 
     def slice_locs(
         self,
diff --git a/pandas/core/indexes/datetimelike.py b/pandas/core/indexes/datetimelike.py
index 8b316de30662c..9da2f0766c83c 100644
--- a/pandas/core/indexes/datetimelike.py
+++ b/pandas/core/indexes/datetimelike.py
@@ -804,14 +804,13 @@ def _get_insert_freq(self, loc: int, item):
                     freq = self.freq
                 elif (loc == len(self)) and item - self.freq == self[-1]:
                     freq = self.freq
-            else:
-                # Adding a single item to an empty index may preserve freq
-                if isinstance(self.freq, Tick):
-                    # all TimedeltaIndex cases go through here; is_on_offset
-                    #  would raise TypeError
-                    freq = self.freq
-                elif self.freq.is_on_offset(item):
-                    freq = self.freq
+            # Adding a single item to an empty index may preserve freq
+            elif isinstance(self.freq, Tick):
+                # all TimedeltaIndex cases go through here; is_on_offset
+                #  would raise TypeError
+                freq = self.freq
+            elif self.freq.is_on_offset(item):
+                freq = self.freq
         return freq
 
     @doc(NDArrayBackedExtensionIndex.delete)
diff --git a/pandas/core/indexes/interval.py b/pandas/core/indexes/interval.py
index 13811c28e6c1e..edbb9d9035b9a 100644
--- a/pandas/core/indexes/interval.py
+++ b/pandas/core/indexes/interval.py
@@ -1287,12 +1287,11 @@ def interval_range(
                 breaks,  # type: ignore[arg-type]
                 dtype,
             )
+    # delegate to the appropriate range function
+    elif isinstance(endpoint, Timestamp):
+        breaks = date_range(start=start, end=end, periods=periods, freq=freq)
     else:
-        # delegate to the appropriate range function
-        if isinstance(endpoint, Timestamp):
-            breaks = date_range(start=start, end=end, periods=periods, freq=freq)
-        else:
-            breaks = timedelta_range(start=start, end=end, periods=periods, freq=freq)
+        breaks = timedelta_range(start=start, end=end, periods=periods, freq=freq)
 
     return IntervalIndex.from_breaks(
         breaks,
diff --git a/pandas/core/indexes/range.py b/pandas/core/indexes/range.py
index 2db50bbbdfa37..fff00abf81949 100644
--- a/pandas/core/indexes/range.py
+++ b/pandas/core/indexes/range.py
@@ -722,10 +722,9 @@ def sort_values(
                 if self.step < 0:
                     sorted_index = self[::-1]
                     inverse_indexer = True
-            else:
-                if self.step > 0:
-                    sorted_index = self[::-1]
-                    inverse_indexer = True
+            elif self.step > 0:
+                sorted_index = self[::-1]
+                inverse_indexer = True
 
         if return_indexer:
             if inverse_indexer:
diff --git a/pandas/core/methods/describe.py b/pandas/core/methods/describe.py
index 17d4d38c97f33..bec064d89c1ff 100644
--- a/pandas/core/methods/describe.py
+++ b/pandas/core/methods/describe.py
@@ -354,9 +354,8 @@ def _refine_percentiles(
     # get them all to be in [0, 1]
     validate_percentile(percentiles)
 
-    # median should always be included
-    if 0.5 not in percentiles:
-        percentiles.append(0.5)
+    if percentiles == []:
+        percentiles.append(0.5)  # By default if percentiles is empty then append 50th percentile.
 
     percentiles = np.asarray(percentiles)
 
diff --git a/pandas/core/methods/selectn.py b/pandas/core/methods/selectn.py
index 02e7445f1d275..de968b763cfc5 100644
--- a/pandas/core/methods/selectn.py
+++ b/pandas/core/methods/selectn.py
@@ -163,11 +163,10 @@ def compute(self, method: str) -> Series:
         if self.keep != "all":
             inds = inds[:n]
             findex = nbase
+        elif len(inds) < nbase <= len(nan_index) + len(inds):
+            findex = len(nan_index) + len(inds)
         else:
-            if len(inds) < nbase <= len(nan_index) + len(inds):
-                findex = len(nan_index) + len(inds)
-            else:
-                findex = len(inds)
+            findex = len(inds)
 
         if self.keep == "last":
             # reverse indices
diff --git a/pandas/core/nanops.py b/pandas/core/nanops.py
index d1dc0ff809497..69800bc91a19e 100644
--- a/pandas/core/nanops.py
+++ b/pandas/core/nanops.py
@@ -195,17 +195,15 @@ def _get_fill_value(
     if _na_ok_dtype(dtype):
         if fill_value_typ is None:
             return np.nan
+        elif fill_value_typ == "+inf":
+            return np.inf
         else:
-            if fill_value_typ == "+inf":
-                return np.inf
-            else:
-                return -np.inf
+            return -np.inf
+    elif fill_value_typ == "+inf":
+        # need the max int here
+        return lib.i8max
     else:
-        if fill_value_typ == "+inf":
-            # need the max int here
-            return lib.i8max
-        else:
-            return iNaT
+        return iNaT
 
 
 def _maybe_get_mask(
@@ -654,9 +652,8 @@ def _mask_datetimelike_result(
         # error: Unsupported target for indexed assignment ("Union[ndarray[Any, Any],
         # datetime64, timedelta64]")
         result[axis_mask] = iNaT  # type: ignore[index]
-    else:
-        if mask.any():
-            return np.int64(iNaT).view(orig_values.dtype)
+    elif mask.any():
+        return np.int64(iNaT).view(orig_values.dtype)
     return result
 
 
@@ -1447,11 +1444,10 @@ def _maybe_arg_null_out(
             raise ValueError("Encountered all NA values")
         elif not skipna and mask.any():
             raise ValueError("Encountered an NA value with skipna=False")
-    else:
-        if skipna and mask.all(axis).any():
-            raise ValueError("Encountered all NA values")
-        elif not skipna and mask.any(axis).any():
-            raise ValueError("Encountered an NA value with skipna=False")
+    elif skipna and mask.all(axis).any():
+        raise ValueError("Encountered all NA values")
+    elif not skipna and mask.any(axis).any():
+        raise ValueError("Encountered an NA value with skipna=False")
     return result
 
 
diff --git a/pandas/core/ops/mask_ops.py b/pandas/core/ops/mask_ops.py
index 86a93e5a3ca2b..5d2bc341d41fc 100644
--- a/pandas/core/ops/mask_ops.py
+++ b/pandas/core/ops/mask_ops.py
@@ -67,14 +67,13 @@ def kleene_or(
             | (right_false & left_mask)
             | (left_mask & right_mask)
         )
+    elif right is True:
+        mask = np.zeros_like(left_mask)
+    elif right is libmissing.NA:
+        mask = (~left & ~left_mask) | left_mask
     else:
-        if right is True:
-            mask = np.zeros_like(left_mask)
-        elif right is libmissing.NA:
-            mask = (~left & ~left_mask) | left_mask
-        else:
-            # False
-            mask = left_mask.copy()
+        # False
+        mask = left_mask.copy()
 
     return result, mask
 
diff --git a/pandas/core/resample.py b/pandas/core/resample.py
index 1cfc75ea11725..a3417f76508f1 100644
--- a/pandas/core/resample.py
+++ b/pandas/core/resample.py
@@ -2191,23 +2191,22 @@ def __init__(
                 closed = "right"
             if label is None:
                 label = "right"
+        # The backward resample sets ``closed`` to ``'right'`` by default
+        # since the last value should be considered as the edge point for
+        # the last bin. When origin in "end" or "end_day", the value for a
+        # specific ``Timestamp`` index stands for the resample result from
+        # the current ``Timestamp`` minus ``freq`` to the current
+        # ``Timestamp`` with a right close.
+        elif origin in ["end", "end_day"]:
+            if closed is None:
+                closed = "right"
+            if label is None:
+                label = "right"
         else:
-            # The backward resample sets ``closed`` to ``'right'`` by default
-            # since the last value should be considered as the edge point for
-            # the last bin. When origin in "end" or "end_day", the value for a
-            # specific ``Timestamp`` index stands for the resample result from
-            # the current ``Timestamp`` minus ``freq`` to the current
-            # ``Timestamp`` with a right close.
-            if origin in ["end", "end_day"]:
-                if closed is None:
-                    closed = "right"
-                if label is None:
-                    label = "right"
-            else:
-                if closed is None:
-                    closed = "left"
-                if label is None:
-                    label = "left"
+            if closed is None:
+                closed = "left"
+            if label is None:
+                label = "left"
 
         self.closed = closed
         self.label = label
diff --git a/pandas/core/reshape/tile.py b/pandas/core/reshape/tile.py
index 034b861a83f43..8345daaec11be 100644
--- a/pandas/core/reshape/tile.py
+++ b/pandas/core/reshape/tile.py
@@ -522,11 +522,10 @@ def _bins_to_cuts(
                 "labels must be unique if ordered=True; pass ordered=False "
                 "for duplicate labels"
             )
-        else:
-            if len(labels) != len(bins) - 1:
-                raise ValueError(
-                    "Bin labels must be one fewer than the number of bin edges"
-                )
+        elif len(labels) != len(bins) - 1:
+            raise ValueError(
+                "Bin labels must be one fewer than the number of bin edges"
+            )
 
         if not isinstance(getattr(labels, "dtype", None), CategoricalDtype):
             labels = Categorical(
diff --git a/pandas/core/series.py b/pandas/core/series.py
index da46f8ede3409..834f1a2acf8de 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -1554,12 +1554,11 @@ def to_string(
 
         if buf is None:
             return result
+        elif hasattr(buf, "write"):
+            buf.write(result)
         else:
-            if hasattr(buf, "write"):
-                buf.write(result)
-            else:
-                with open(buf, "w", encoding="utf-8") as f:
-                    f.write(result)
+            with open(buf, "w", encoding="utf-8") as f:
+                f.write(result)
         return None
 
     @overload
@@ -5615,11 +5614,10 @@ def dropna(
 
         if self._can_hold_na:
             result = remove_na_arraylike(self)
+        elif not inplace:
+            result = self.copy(deep=False)
         else:
-            if not inplace:
-                result = self.copy(deep=False)
-            else:
-                result = self
+            result = self
 
         if ignore_index:
             result.index = default_index(len(result))
diff --git a/pandas/core/strings/object_array.py b/pandas/core/strings/object_array.py
index 0adb7b51cf2b7..a0a73c528f7f1 100644
--- a/pandas/core/strings/object_array.py
+++ b/pandas/core/strings/object_array.py
@@ -148,12 +148,11 @@ def _str_contains(
             pat = re.compile(pat, flags=flags)
 
             f = lambda x: pat.search(x) is not None
+        elif case:
+            f = lambda x: pat in x
         else:
-            if case:
-                f = lambda x: pat in x
-            else:
-                upper_pat = pat.upper()
-                f = lambda x: upper_pat in x.upper()
+            upper_pat = pat.upper()
+            f = lambda x: upper_pat in x.upper()
         if na is not lib.no_default and not isna(na) and not isinstance(na, bool):
             # GH#59561
             warnings.warn(
@@ -382,11 +381,10 @@ def _str_split(
             elif regex is False:
                 new_pat = pat
             # regex is None so link to old behavior #43563
+            elif len(pat) == 1:
+                new_pat = pat
             else:
-                if len(pat) == 1:
-                    new_pat = pat
-                else:
-                    new_pat = re.compile(pat)
+                new_pat = re.compile(pat)
 
             if isinstance(new_pat, re.Pattern):
                 if n is None or n == -1:
diff --git a/pandas/core/tools/datetimes.py b/pandas/core/tools/datetimes.py
index 0a10001a3113f..9b90dfae9f5e8 100644
--- a/pandas/core/tools/datetimes.py
+++ b/pandas/core/tools/datetimes.py
@@ -379,12 +379,11 @@ def _convert_listlike_datetimes(
                 else:
                     arg_array = arg_array._dt_tz_localize("UTC")
                 arg = Index(arg_array)
+            # ArrowExtensionArray
+            elif arg_dtype.pyarrow_dtype.tz is not None:
+                arg = arg._dt_tz_convert("UTC")
             else:
-                # ArrowExtensionArray
-                if arg_dtype.pyarrow_dtype.tz is not None:
-                    arg = arg._dt_tz_convert("UTC")
-                else:
-                    arg = arg._dt_tz_localize("UTC")
+                arg = arg._dt_tz_localize("UTC")
         return arg
 
     elif lib.is_np_dtype(arg_dtype, "M"):
diff --git a/pandas/core/window/common.py b/pandas/core/window/common.py
index 004a3555f0212..766d8d3d2f4e6 100644
--- a/pandas/core/window/common.py
+++ b/pandas/core/window/common.py
@@ -155,9 +155,8 @@ def zsqrt(x):
     if isinstance(x, ABCDataFrame):
         if mask._values.any():
             result[mask] = 0
-    else:
-        if mask.any():
-            result[mask] = 0
+    elif mask.any():
+        result[mask] = 0
 
     return result
 
diff --git a/pandas/doc/.gitignore b/pandas/doc/.gitignore
new file mode 100644
index 0000000000000..e23892d6100e8
--- /dev/null
+++ b/pandas/doc/.gitignore
@@ -0,0 +1,4 @@
+data/
+timeseries.csv
+timeseries.parquet
+timeseries_wide.parquet
diff --git a/pandas/doc/_templates/api_redirect.html b/pandas/doc/_templates/api_redirect.html
new file mode 100644
index 0000000000000..c04a8b58ce544
--- /dev/null
+++ b/pandas/doc/_templates/api_redirect.html
@@ -0,0 +1,10 @@
+{% set redirect = redirects[pagename.split("/")[-1]] %}
+<html>
+    <head>
+        <meta http-equiv="Refresh" content="0; url={{ redirect }}.html" />
+        <title>This API page has moved</title>
+    </head>
+    <body>
+        <p>This API page has moved <a href="{{ redirect }}.html">here</a>.</p>
+    </body>
+</html>
diff --git a/pandas/doc/_templates/autosummary/accessor.rst b/pandas/doc/_templates/autosummary/accessor.rst
new file mode 100644
index 0000000000000..4ba745cd6fdba
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/accessor.rst
@@ -0,0 +1,6 @@
+{{ fullname }}
+{{ underline }}
+
+.. currentmodule:: {{ module.split('.')[0] }}
+
+.. autoaccessor:: {{ (module.split('.')[1:] + [objname]) | join('.') }}
diff --git a/pandas/doc/_templates/autosummary/accessor_attribute.rst b/pandas/doc/_templates/autosummary/accessor_attribute.rst
new file mode 100644
index 0000000000000..b5ad65d6a736f
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/accessor_attribute.rst
@@ -0,0 +1,6 @@
+{{ fullname }}
+{{ underline }}
+
+.. currentmodule:: {{ module.split('.')[0] }}
+
+.. autoaccessorattribute:: {{ (module.split('.')[1:] + [objname]) | join('.') }}
diff --git a/pandas/doc/_templates/autosummary/accessor_callable.rst b/pandas/doc/_templates/autosummary/accessor_callable.rst
new file mode 100644
index 0000000000000..7a3301814f5f4
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/accessor_callable.rst
@@ -0,0 +1,6 @@
+{{ fullname }}
+{{ underline }}
+
+.. currentmodule:: {{ module.split('.')[0] }}
+
+.. autoaccessorcallable:: {{ (module.split('.')[1:] + [objname]) | join('.') }}.__call__
diff --git a/pandas/doc/_templates/autosummary/accessor_method.rst b/pandas/doc/_templates/autosummary/accessor_method.rst
new file mode 100644
index 0000000000000..aefbba6ef1bbc
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/accessor_method.rst
@@ -0,0 +1,6 @@
+{{ fullname }}
+{{ underline }}
+
+.. currentmodule:: {{ module.split('.')[0] }}
+
+.. autoaccessormethod:: {{ (module.split('.')[1:] + [objname]) | join('.') }}
diff --git a/pandas/doc/_templates/autosummary/class.rst b/pandas/doc/_templates/autosummary/class.rst
new file mode 100644
index 0000000000000..79c2e37b0192f
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/class.rst
@@ -0,0 +1,32 @@
+{{ fullname | escape | underline}}
+
+.. currentmodule:: {{ module }}
+
+.. autoclass:: {{ objname }}
+
+   {% block methods %}
+
+   {% block attributes %}
+   {% if attributes %}
+   .. rubric:: {{ _('Attributes') }}
+
+   .. autosummary::
+   {% for item in attributes %}
+      {% if item in members and not item.startswith('_') %}
+        ~{{ name }}.{{ item }}
+      {% endif %}
+   {%- endfor %}
+   {% endif %}
+   {% endblock %}
+
+   {% if methods %}
+   .. rubric:: {{ _('Methods') }}
+
+   .. autosummary::
+   {% for item in methods %}
+      {% if item in members and (not item.startswith('_') or item in ['__call__']) %}
+        ~{{ name }}.{{ item }}
+      {% endif %}
+   {%- endfor %}
+   {% endif %}
+   {% endblock %}
diff --git a/pandas/doc/_templates/autosummary/class_without_autosummary.rst b/pandas/doc/_templates/autosummary/class_without_autosummary.rst
new file mode 100644
index 0000000000000..6676c672b206d
--- /dev/null
+++ b/pandas/doc/_templates/autosummary/class_without_autosummary.rst
@@ -0,0 +1,6 @@
+{{ fullname }}
+{{ underline }}
+
+.. currentmodule:: {{ module }}
+
+.. autoclass:: {{ objname }}
diff --git a/pandas/doc/_templates/pandas_footer.html b/pandas/doc/_templates/pandas_footer.html
new file mode 100644
index 0000000000000..6d8caa4d6c741
--- /dev/null
+++ b/pandas/doc/_templates/pandas_footer.html
@@ -0,0 +1,3 @@
+<p class="copyright">
+    &copy {{copyright}} pandas via <a href="https://numfocus.org">NumFOCUS, Inc.</a> Hosted by <a href="https://www.ovhcloud.com">OVHcloud</a>.
+</p>
diff --git a/pandas/doc/_templates/sidebar-nav-bs.html b/pandas/doc/_templates/sidebar-nav-bs.html
new file mode 100644
index 0000000000000..8298b66568e20
--- /dev/null
+++ b/pandas/doc/_templates/sidebar-nav-bs.html
@@ -0,0 +1,9 @@
+<nav class="bd-links" id="bd-docs-nav" aria-label="Main navigation">
+  <div class="bd-toc-item navbar-nav">
+    {% if pagename.startswith("reference") %}
+    {{ generate_toctree_html("sidebar", maxdepth=4, collapse=True, includehidden=True, titles_only=True) }}
+    {% else %}
+    {{ generate_toctree_html("sidebar", maxdepth=4, collapse=False, includehidden=True, titles_only=True) }}
+    {% endif %}
+  </div>
+</nav>
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pdf b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pdf
new file mode 100644
index 0000000000000..3582e0c0dabf9
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pdf differ
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pptx b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pptx
new file mode 100644
index 0000000000000..746f508516964
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet.pptx differ
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pdf b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pdf
new file mode 100644
index 0000000000000..ea356385e9fb1
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pdf differ
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pptx b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pptx
new file mode 100644
index 0000000000000..1b812c1a2595a
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pptx differ
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pdf b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pdf
new file mode 100644
index 0000000000000..cf1e40e627f33
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pdf differ
diff --git a/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pptx b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pptx
new file mode 100644
index 0000000000000..564d92ddbb56a
Binary files /dev/null and b/pandas/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pptx differ
diff --git a/pandas/doc/cheatsheet/README.md b/pandas/doc/cheatsheet/README.md
new file mode 100644
index 0000000000000..b8599acff2f6e
--- /dev/null
+++ b/pandas/doc/cheatsheet/README.md
@@ -0,0 +1,24 @@
+# Pandas Cheat Sheet
+
+The Pandas Cheat Sheet was created using Microsoft Powerpoint 2013.
+To create the PDF version, within Powerpoint, simply do a "Save As"
+and pick "PDF" as the format.
+
+This cheat sheet, originally written by Irv Lustig, [Princeton Consultants](https://www.princetonoptimization.com/), was inspired by the [RStudio Data Wrangling Cheatsheet](https://www.rstudio.com/wp-content/uploads/2015/02/data-wrangling-cheatsheet.pdf).
+
+| Topic                  | Language    | PDF                                                                                                                                                                                                                                     | PPT                                                                                                                                                                                                                                               |
+|------------------------|-------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| Pandas_Cheat_Sheet     | English     | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet.pdf" target="_parent"><img src="https://img.shields.io/badge/Open in PDF-%23FF0000.svg?style=flat-square&logo=adobe&logoColor=white"/></a>    | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet.pptx" target="_parent"><img src="https://img.shields.io/badge/Open in PPT-B7472A?style=flat-square&logo=microsoft-powerpoint&logoColor=white"/></a>     |
+| Pandas_Cheat_Sheet_JA  | Japanese    | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pdf" target="_parent"><img src="https://img.shields.io/badge/Open in PDF-%23FF0000.svg?style=flat-square&logo=adobe&logoColor=white"/></a> | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet_JA.pptx" target="_parent"><img  src="https://img.shields.io/badge/Open in PPT-B7472A?style=flat-square&logo=microsoft-powerpoint&logoColor=white"/></a> |
+| Pandas_Cheat_Sheet_FA  | Persian     | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pdf" target="_parent"><img src="https://img.shields.io/badge/Open in PDF-%23FF0000.svg?style=flat-square&logo=adobe&logoColor=white"/></a> | <a href="https://github.com/pandas-dev/pandas/blob/main/doc/cheatsheet/Pandas_Cheat_Sheet_FA.pptx" target="_parent"><img  src="https://img.shields.io/badge/Open in PPT-B7472A?style=flat-square&logo=microsoft-powerpoint&logoColor=white"/></a> |
+
+
+
+**Alternative**
+
+Alternatively, if you want to complement your learning, you can use the Pandas Cheat sheets
+developed by [DataCamp](https://www.datacamp.com/) in "PDF", "Google Colab" and "Streamlit" formats.
+
+| Topic       | PDF                                                                                                                                                                                                                                  | Streamlit                                                                                                                                                        | Google Colab                                                                                                                                                                                                                                   |
+|-------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| Pandas      | <a href="https://github.com/fralfaro/DS-Cheat-Sheets/blob/main/docs/files/pandas_cs.pdf" target="_parent"><img src="https://img.shields.io/badge/Open in PDF-%23FF0000.svg?style=flat-square&logo=adobe&logoColor=white"/></a>       | <a href="https://ds-cheat-sheets-pandas.streamlit.app/" target="_parent"><img src="https://static.streamlit.io/badges/streamlit_badge_black_white.svg"/></a>     | <a href="https://colab.research.google.com/github/fralfaro/DS-Cheat-Sheets/blob/main/docs/examples/pandas/pandas.ipynb" target="_parent"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a>         |
diff --git a/pandas/doc/make.py b/pandas/doc/make.py
new file mode 100755
index 0000000000000..9542563dc037b
--- /dev/null
+++ b/pandas/doc/make.py
@@ -0,0 +1,391 @@
+#!/usr/bin/env python3
+"""
+Python script for building documentation.
+
+To build the docs you must have all optional dependencies for pandas
+installed. See the installation instructions for a list of these.
+
+Usage
+-----
+    $ python make.py clean
+    $ python make.py html
+    $ python make.py latex
+"""
+
+import argparse
+import csv
+import importlib
+import os
+import shutil
+import subprocess
+import sys
+import webbrowser
+
+import docutils
+import docutils.parsers.rst
+
+DOC_PATH = os.path.dirname(os.path.abspath(__file__))
+SOURCE_PATH = os.path.join(DOC_PATH, "source")
+BUILD_PATH = os.path.join(DOC_PATH, "build")
+REDIRECTS_FILE = os.path.join(DOC_PATH, "redirects.csv")
+
+
+class DocBuilder:
+    """
+    Class to wrap the different commands of this script.
+
+    All public methods of this class can be called as parameters of the
+    script.
+    """
+
+    def __init__(
+        self,
+        num_jobs="auto",
+        include_api=True,
+        whatsnew=False,
+        single_doc=None,
+        verbosity=0,
+        warnings_are_errors=False,
+        no_browser=False,
+    ) -> None:
+        self.num_jobs = num_jobs
+        self.include_api = include_api
+        self.whatsnew = whatsnew
+        self.verbosity = verbosity
+        self.warnings_are_errors = warnings_are_errors
+        self.no_browser = no_browser
+
+        if single_doc:
+            single_doc = self._process_single_doc(single_doc)
+            os.environ["SPHINX_PATTERN"] = single_doc
+        elif not include_api:
+            os.environ["SPHINX_PATTERN"] = "-api"
+        elif whatsnew:
+            os.environ["SPHINX_PATTERN"] = "whatsnew"
+
+        self.single_doc_html = None
+        if single_doc and single_doc.endswith(".rst"):
+            self.single_doc_html = os.path.splitext(single_doc)[0] + ".html"
+        elif single_doc:
+            self.single_doc_html = f"reference/api/pandas.{single_doc}.html"
+
+    def _process_single_doc(self, single_doc):
+        """
+        Make sure the provided value for --single is a path to an existing
+        .rst/.ipynb file, or a pandas object that can be imported.
+
+        For example, categorial.rst or pandas.DataFrame.head. For the latter,
+        return the corresponding file path
+        (e.g. reference/api/pandas.DataFrame.head.rst).
+        """
+        base_name, extension = os.path.splitext(single_doc)
+        if extension in (".rst", ".ipynb"):
+            if os.path.exists(os.path.join(SOURCE_PATH, single_doc)):
+                return single_doc
+            else:
+                raise FileNotFoundError(f"File {single_doc} not found")
+
+        elif single_doc.startswith("pandas."):
+            try:
+                obj = pandas  # noqa: F821
+                for name in single_doc.split("."):
+                    obj = getattr(obj, name)
+            except AttributeError as err:
+                raise ImportError(f"Could not import {single_doc}") from err
+            else:
+                return single_doc[len("pandas.") :]
+        else:
+            raise ValueError(
+                f"--single={single_doc} not understood. "
+                "Value should be a valid path to a .rst or .ipynb file, "
+                "or a valid pandas object "
+                "(e.g. categorical.rst or pandas.DataFrame.head)"
+            )
+
+    @staticmethod
+    def _run_os(*args) -> None:
+        """
+        Execute a command as a OS terminal.
+
+        Parameters
+        ----------
+        *args : list of str
+            Command and parameters to be executed
+
+        Examples
+        --------
+        >>> DocBuilder()._run_os("python", "--version")
+        """
+        subprocess.check_call(args, stdout=sys.stdout, stderr=sys.stderr)
+
+    def _sphinx_build(self, kind: str):
+        """
+        Call sphinx to build documentation.
+
+        Attribute `num_jobs` from the class is used.
+
+        Parameters
+        ----------
+        kind : {'html', 'latex', 'linkcheck'}
+
+        Examples
+        --------
+        >>> DocBuilder(num_jobs=4)._sphinx_build("html")
+        """
+        if kind not in ("html", "latex", "linkcheck"):
+            raise ValueError(f"kind must be html, latex or linkcheck, not {kind}")
+
+        cmd = ["sphinx-build", "-b", kind]
+        if self.num_jobs:
+            cmd += ["-j", self.num_jobs]
+        if self.warnings_are_errors:
+            cmd += ["-W", "--keep-going"]
+        if self.verbosity:
+            cmd.append(f"-{'v' * self.verbosity}")
+        cmd += [
+            "-d",
+            os.path.join(BUILD_PATH, "doctrees"),
+            SOURCE_PATH,
+            os.path.join(BUILD_PATH, kind),
+        ]
+        return subprocess.call(cmd)
+
+    def _open_browser(self, single_doc_html) -> None:
+        """
+        Open a browser tab showing single
+        """
+        url = os.path.join("file://", DOC_PATH, "build", "html", single_doc_html)
+        webbrowser.open(url, new=2)
+
+    def _get_page_title(self, page):
+        """
+        Open the rst file `page` and extract its title.
+        """
+        fname = os.path.join(SOURCE_PATH, f"{page}.rst")
+        doc = docutils.utils.new_document(
+            "<doc>",
+            docutils.frontend.get_default_settings(docutils.parsers.rst.Parser),
+        )
+        with open(fname, encoding="utf-8") as f:
+            data = f.read()
+
+        parser = docutils.parsers.rst.Parser()
+        # do not generate any warning when parsing the rst
+        with open(os.devnull, "a", encoding="utf-8") as f:
+            doc.reporter.stream = f
+            parser.parse(data, doc)
+
+        section = next(
+            node for node in doc.children if isinstance(node, docutils.nodes.section)
+        )
+        title = next(
+            node for node in section.children if isinstance(node, docutils.nodes.title)
+        )
+
+        return title.astext()
+
+    def _add_redirects(self) -> None:
+        """
+        Create in the build directory an html file with a redirect,
+        for every row in REDIRECTS_FILE.
+        """
+        with open(REDIRECTS_FILE, encoding="utf-8") as mapping_fd:
+            reader = csv.reader(mapping_fd)
+            for row in reader:
+                if not row or row[0].strip().startswith("#"):
+                    continue
+
+                html_path = os.path.join(BUILD_PATH, "html")
+                path = os.path.join(html_path, *row[0].split("/")) + ".html"
+
+                if not self.include_api and (
+                    os.path.join(html_path, "reference") in path
+                    or os.path.join(html_path, "generated") in path
+                ):
+                    continue
+
+                try:
+                    title = self._get_page_title(row[1])
+                except Exception:
+                    # the file can be an ipynb and not an rst, or docutils
+                    # may not be able to read the rst because it has some
+                    # sphinx specific stuff
+                    title = "this page"
+
+                with open(path, "w", encoding="utf-8") as moved_page_fd:
+                    html = f"""\
+<html>
+    <head>
+        <meta http-equiv="refresh" content="0;URL={row[1]}.html"/>
+    </head>
+    <body>
+        <p>
+            The page has been moved to <a href="{row[1]}.html">{title}</a>
+        </p>
+    </body>
+<html>"""
+
+                    moved_page_fd.write(html)
+
+    def html(self):
+        """
+        Build HTML documentation.
+        """
+        ret_code = self._sphinx_build("html")
+        zip_fname = os.path.join(BUILD_PATH, "html", "pandas.zip")
+        if os.path.exists(zip_fname):
+            os.remove(zip_fname)  # noqa: TID251
+
+        if ret_code == 0:
+            if self.single_doc_html is not None:
+                if not self.no_browser:
+                    self._open_browser(self.single_doc_html)
+            else:
+                self._add_redirects()
+                if self.whatsnew and not self.no_browser:
+                    self._open_browser(os.path.join("whatsnew", "index.html"))
+
+        return ret_code
+
+    def latex(self, force=False):
+        """
+        Build PDF documentation.
+        """
+        if sys.platform == "win32":
+            sys.stderr.write("latex build has not been tested on windows\n")
+        else:
+            ret_code = self._sphinx_build("latex")
+            os.chdir(os.path.join(BUILD_PATH, "latex"))
+            if force:
+                for i in range(3):
+                    self._run_os("pdflatex", "-interaction=nonstopmode", "pandas.tex")
+                raise SystemExit(
+                    'You should check the file "build/latex/pandas.pdf" for problems.'
+                )
+            self._run_os("make")
+            return ret_code
+
+    def latex_forced(self):
+        """
+        Build PDF documentation with retries to find missing references.
+        """
+        return self.latex(force=True)
+
+    @staticmethod
+    def clean() -> None:
+        """
+        Clean documentation generated files.
+        """
+        shutil.rmtree(BUILD_PATH, ignore_errors=True)
+        shutil.rmtree(os.path.join(SOURCE_PATH, "reference", "api"), ignore_errors=True)
+
+    def zip_html(self) -> None:
+        """
+        Compress HTML documentation into a zip file.
+        """
+        zip_fname = os.path.join(BUILD_PATH, "html", "pandas.zip")
+        if os.path.exists(zip_fname):
+            os.remove(zip_fname)  # noqa: TID251
+        dirname = os.path.join(BUILD_PATH, "html")
+        fnames = os.listdir(dirname)
+        os.chdir(dirname)
+        self._run_os("zip", zip_fname, "-r", "-q", *fnames)
+
+    def linkcheck(self):
+        """
+        Check for broken links in the documentation.
+        """
+        return self._sphinx_build("linkcheck")
+
+
+def main():
+    cmds = [method for method in dir(DocBuilder) if not method.startswith("_")]
+
+    joined = ",".join(cmds)
+    argparser = argparse.ArgumentParser(
+        description="pandas documentation builder", epilog=f"Commands: {joined}"
+    )
+
+    joined = ", ".join(cmds)
+    argparser.add_argument(
+        "command", nargs="?", default="html", help=f"command to run: {joined}"
+    )
+    argparser.add_argument(
+        "--num-jobs", default="auto", help="number of jobs used by sphinx-build"
+    )
+    argparser.add_argument(
+        "--no-api", default=False, help="omit api and autosummary", action="store_true"
+    )
+    argparser.add_argument(
+        "--whatsnew",
+        default=False,
+        help="only build whatsnew (and api for links)",
+        action="store_true",
+    )
+    argparser.add_argument(
+        "--single",
+        metavar="FILENAME",
+        type=str,
+        default=None,
+        help=(
+            "filename (relative to the 'source' folder) of section or method name to "
+            "compile, e.g. 'development/contributing.rst', "
+            "'pandas.DataFrame.join'"
+        ),
+    )
+    argparser.add_argument(
+        "--python-path", type=str, default=os.path.dirname(DOC_PATH), help="path"
+    )
+    argparser.add_argument(
+        "-v",
+        action="count",
+        dest="verbosity",
+        default=0,
+        help=(
+            "increase verbosity (can be repeated), passed to the sphinx build command"
+        ),
+    )
+    argparser.add_argument(
+        "--warnings-are-errors",
+        "-W",
+        action="store_true",
+        help="fail if warnings are raised",
+    )
+    argparser.add_argument(
+        "--no-browser",
+        help="Don't open browser",
+        default=False,
+        action="store_true",
+    )
+    args = argparser.parse_args()
+
+    if args.command not in cmds:
+        joined = ", ".join(cmds)
+        raise ValueError(f"Unknown command {args.command}. Available options: {joined}")
+
+    # Below we update both os.environ and sys.path. The former is used by
+    # external libraries (namely Sphinx) to compile this module and resolve
+    # the import of `python_path` correctly. The latter is used to resolve
+    # the import within the module, injecting it into the global namespace
+    os.environ["PYTHONPATH"] = args.python_path
+    sys.path.insert(0, args.python_path)
+    globals()["pandas"] = importlib.import_module("pandas")
+
+    # Set the matplotlib backend to the non-interactive Agg backend for all
+    # child processes.
+    os.environ["MPLBACKEND"] = "module://matplotlib.backends.backend_agg"
+
+    builder = DocBuilder(
+        args.num_jobs,
+        not args.no_api,
+        args.whatsnew,
+        args.single,
+        args.verbosity,
+        args.warnings_are_errors,
+        args.no_browser,
+    )
+    return getattr(builder, args.command)()
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/pandas/doc/redirects.csv b/pandas/doc/redirects.csv
new file mode 100644
index 0000000000000..c11e4e242f128
--- /dev/null
+++ b/pandas/doc/redirects.csv
@@ -0,0 +1,1434 @@
+# This file should contain all the redirects in the documentation
+# in the format `<old_path>,<new_path>`
+
+# whatsnew
+whatsnew,whatsnew/index
+release,whatsnew/index
+
+# getting started
+install,getting_started/install
+comparison_with_r,getting_started/comparison/comparison_with_r
+comparison_with_sql,getting_started/comparison/comparison_with_sql
+comparison_with_sas,getting_started/comparison/comparison_with_sas
+comparison_with_stata,getting_started/comparison/comparison_with_stata
+overview,getting_started/overview
+tutorials,getting_started/tutorials
+
+# user guide
+advanced,user_guide/advanced
+categorical,user_guide/categorical
+computation,user_guide/computation
+cookbook,user_guide/cookbook
+enhancingperf,user_guide/enhancingperf
+gotchas,user_guide/gotchas
+groupby,user_guide/groupby
+indexing,user_guide/indexing
+integer_na,user_guide/integer_na
+io,user_guide/io
+merging,user_guide/merging
+missing_data,user_guide/missing_data
+options,user_guide/options
+reshaping,user_guide/reshaping
+sparse,user_guide/sparse
+style,user_guide/style
+text,user_guide/text
+timedeltas,user_guide/timedeltas
+timeseries,user_guide/timeseries
+visualization,user_guide/visualization
+10min,user_guide/10min
+basics,user_guide/basics
+dsintro,user_guide/dsintro
+
+# development
+contributing,development/contributing
+contributing_docstring,development/contributing_docstring
+developer,development/developer
+extending,development/extending
+internals,development/internals
+development/meeting,community
+
+# api moved function
+reference/api/pandas.io.json.json_normalize,pandas.json_normalize
+
+# rename due to refactors
+reference/api/pandas.core.window.Rolling,pandas.core.window.rolling.Rolling
+reference/api/pandas.core.window.Rolling.aggregate,pandas.core.window.rolling.Rolling.aggregate
+reference/api/pandas.core.window.Rolling.apply,pandas.core.window.rolling.Rolling.apply
+reference/api/pandas.core.window.Rolling.corr,pandas.core.window.rolling.Rolling.corr
+reference/api/pandas.core.window.Rolling.count,pandas.core.window.rolling.Rolling.count
+reference/api/pandas.core.window.Rolling.cov,pandas.core.window.rolling.Rolling.cov
+reference/api/pandas.core.window.Rolling.kurt,pandas.core.window.rolling.Rolling.kurt
+reference/api/pandas.core.window.Rolling.max,pandas.core.window.rolling.Rolling.max
+reference/api/pandas.core.window.Rolling.mean,pandas.core.window.rolling.Rolling.mean
+reference/api/pandas.core.window.Rolling.median,pandas.core.window.rolling.Rolling.median
+reference/api/pandas.core.window.Rolling.min,pandas.core.window.rolling.Rolling.min
+reference/api/pandas.core.window.Rolling.quantile,pandas.core.window.rolling.Rolling.quantile
+reference/api/pandas.core.window.Rolling.skew,pandas.core.window.rolling.Rolling.skew
+reference/api/pandas.core.window.Rolling.std,pandas.core.window.rolling.Rolling.std
+reference/api/pandas.core.window.Rolling.sum,pandas.core.window.rolling.Rolling.sum
+reference/api/pandas.core.window.Rolling.var,pandas.core.window.rolling.Rolling.var
+
+# api url change (generated -> reference/api rename)
+api,reference/index
+generated/pandas.api.extensions.ExtensionArray.argsort,../reference/api/pandas.api.extensions.ExtensionArray.argsort
+generated/pandas.api.extensions.ExtensionArray.astype,../reference/api/pandas.api.extensions.ExtensionArray.astype
+generated/pandas.api.extensions.ExtensionArray.copy,../reference/api/pandas.api.extensions.ExtensionArray.copy
+generated/pandas.api.extensions.ExtensionArray.dropna,../reference/api/pandas.api.extensions.ExtensionArray.dropna
+generated/pandas.api.extensions.ExtensionArray.dtype,../reference/api/pandas.api.extensions.ExtensionArray.dtype
+generated/pandas.api.extensions.ExtensionArray.factorize,../reference/api/pandas.api.extensions.ExtensionArray.factorize
+generated/pandas.api.extensions.ExtensionArray.fillna,../reference/api/pandas.api.extensions.ExtensionArray.fillna
+generated/pandas.api.extensions.ExtensionArray,../reference/api/pandas.api.extensions.ExtensionArray
+generated/pandas.api.extensions.ExtensionArray.isna,../reference/api/pandas.api.extensions.ExtensionArray.isna
+generated/pandas.api.extensions.ExtensionArray.nbytes,../reference/api/pandas.api.extensions.ExtensionArray.nbytes
+generated/pandas.api.extensions.ExtensionArray.ndim,../reference/api/pandas.api.extensions.ExtensionArray.ndim
+generated/pandas.api.extensions.ExtensionArray.shape,../reference/api/pandas.api.extensions.ExtensionArray.shape
+generated/pandas.api.extensions.ExtensionArray.take,../reference/api/pandas.api.extensions.ExtensionArray.take
+generated/pandas.api.extensions.ExtensionArray.unique,../reference/api/pandas.api.extensions.ExtensionArray.unique
+generated/pandas.api.extensions.ExtensionDtype.construct_array_type,../reference/api/pandas.api.extensions.ExtensionDtype.construct_array_type
+generated/pandas.api.extensions.ExtensionDtype.construct_from_string,../reference/api/pandas.api.extensions.ExtensionDtype.construct_from_string
+generated/pandas.api.extensions.ExtensionDtype,../reference/api/pandas.api.extensions.ExtensionDtype
+generated/pandas.api.extensions.ExtensionDtype.is_dtype,../reference/api/pandas.api.extensions.ExtensionDtype.is_dtype
+generated/pandas.api.extensions.ExtensionDtype.kind,../reference/api/pandas.api.extensions.ExtensionDtype.kind
+generated/pandas.api.extensions.ExtensionDtype.name,../reference/api/pandas.api.extensions.ExtensionDtype.name
+generated/pandas.api.extensions.ExtensionDtype.names,../reference/api/pandas.api.extensions.ExtensionDtype.names
+generated/pandas.api.extensions.ExtensionDtype.na_value,../reference/api/pandas.api.extensions.ExtensionDtype.na_value
+generated/pandas.api.extensions.ExtensionDtype.type,../reference/api/pandas.api.extensions.ExtensionDtype.type
+generated/pandas.api.extensions.register_dataframe_accessor,../reference/api/pandas.api.extensions.register_dataframe_accessor
+generated/pandas.api.extensions.register_extension_dtype,../reference/api/pandas.api.extensions.register_extension_dtype
+generated/pandas.api.extensions.register_index_accessor,../reference/api/pandas.api.extensions.register_index_accessor
+generated/pandas.api.extensions.register_series_accessor,../reference/api/pandas.api.extensions.register_series_accessor
+generated/pandas.api.types.infer_dtype,../reference/api/pandas.api.types.infer_dtype
+generated/pandas.api.types.is_bool_dtype,../reference/api/pandas.api.types.is_bool_dtype
+generated/pandas.api.types.is_bool,../reference/api/pandas.api.types.is_bool
+generated/pandas.api.types.is_complex_dtype,../reference/api/pandas.api.types.is_complex_dtype
+generated/pandas.api.types.is_complex,../reference/api/pandas.api.types.is_complex
+generated/pandas.api.types.is_datetime64_any_dtype,../reference/api/pandas.api.types.is_datetime64_any_dtype
+generated/pandas.api.types.is_datetime64_dtype,../reference/api/pandas.api.types.is_datetime64_dtype
+generated/pandas.api.types.is_datetime64_ns_dtype,../reference/api/pandas.api.types.is_datetime64_ns_dtype
+generated/pandas.api.types.is_datetime64tz_dtype,../reference/api/pandas.api.types.is_datetime64tz_dtype
+generated/pandas.api.types.is_datetimetz,../reference/api/pandas.api.types.is_datetimetz
+generated/pandas.api.types.is_dict_like,../reference/api/pandas.api.types.is_dict_like
+generated/pandas.api.types.is_extension_array_dtype,../reference/api/pandas.api.types.is_extension_array_dtype
+generated/pandas.api.types.is_extension_type,../reference/api/pandas.api.types.is_extension_type
+generated/pandas.api.types.is_file_like,../reference/api/pandas.api.types.is_file_like
+generated/pandas.api.types.is_float_dtype,../reference/api/pandas.api.types.is_float_dtype
+generated/pandas.api.types.is_float,../reference/api/pandas.api.types.is_float
+generated/pandas.api.types.is_hashable,../reference/api/pandas.api.types.is_hashable
+generated/pandas.api.types.is_int64_dtype,../reference/api/pandas.api.types.is_int64_dtype
+generated/pandas.api.types.is_integer_dtype,../reference/api/pandas.api.types.is_integer_dtype
+generated/pandas.api.types.is_integer,../reference/api/pandas.api.types.is_integer
+generated/pandas.api.types.is_interval_dtype,../reference/api/pandas.api.types.is_interval_dtype
+generated/pandas.api.types.is_iterator,../reference/api/pandas.api.types.is_iterator
+generated/pandas.api.types.is_list_like,../reference/api/pandas.api.types.is_list_like
+generated/pandas.api.types.is_named_tuple,../reference/api/pandas.api.types.is_named_tuple
+generated/pandas.api.types.is_number,../reference/api/pandas.api.types.is_number
+generated/pandas.api.types.is_numeric_dtype,../reference/api/pandas.api.types.is_numeric_dtype
+generated/pandas.api.types.is_object_dtype,../reference/api/pandas.api.types.is_object_dtype
+generated/pandas.api.types.is_period_dtype,../reference/api/pandas.api.types.is_period_dtype
+generated/pandas.api.types.is_re_compilable,../reference/api/pandas.api.types.is_re_compilable
+generated/pandas.api.types.is_re,../reference/api/pandas.api.types.is_re
+generated/pandas.api.types.is_scalar,../reference/api/pandas.api.types.is_scalar
+generated/pandas.api.types.is_signed_integer_dtype,../reference/api/pandas.api.types.is_signed_integer_dtype
+generated/pandas.api.types.is_sparse,../reference/api/pandas.api.types.is_sparse
+generated/pandas.api.types.is_string_dtype,../reference/api/pandas.api.types.is_string_dtype
+generated/pandas.api.types.is_timedelta64_dtype,../reference/api/pandas.api.types.is_timedelta64_dtype
+generated/pandas.api.types.is_timedelta64_ns_dtype,../reference/api/pandas.api.types.is_timedelta64_ns_dtype
+generated/pandas.api.types.is_unsigned_integer_dtype,../reference/api/pandas.api.types.is_unsigned_integer_dtype
+generated/pandas.api.types.pandas_dtype,../reference/api/pandas.api.types.pandas_dtype
+generated/pandas.api.types.union_categoricals,../reference/api/pandas.api.types.union_categoricals
+generated/pandas.bdate_range,../reference/api/pandas.bdate_range
+generated/pandas.Categorical.__array__,../reference/api/pandas.Categorical.__array__
+generated/pandas.Categorical.categories,../reference/api/pandas.Categorical.categories
+generated/pandas.Categorical.codes,../reference/api/pandas.Categorical.codes
+generated/pandas.CategoricalDtype.categories,../reference/api/pandas.CategoricalDtype.categories
+generated/pandas.Categorical.dtype,../reference/api/pandas.Categorical.dtype
+generated/pandas.CategoricalDtype,../reference/api/pandas.CategoricalDtype
+generated/pandas.CategoricalDtype.ordered,../reference/api/pandas.CategoricalDtype.ordered
+generated/pandas.Categorical.from_codes,../reference/api/pandas.Categorical.from_codes
+generated/pandas.Categorical,../reference/api/pandas.Categorical
+generated/pandas.CategoricalIndex.add_categories,../reference/api/pandas.CategoricalIndex.add_categories
+generated/pandas.CategoricalIndex.as_ordered,../reference/api/pandas.CategoricalIndex.as_ordered
+generated/pandas.CategoricalIndex.as_unordered,../reference/api/pandas.CategoricalIndex.as_unordered
+generated/pandas.CategoricalIndex.categories,../reference/api/pandas.CategoricalIndex.categories
+generated/pandas.CategoricalIndex.codes,../reference/api/pandas.CategoricalIndex.codes
+generated/pandas.CategoricalIndex.equals,../reference/api/pandas.CategoricalIndex.equals
+generated/pandas.CategoricalIndex,../reference/api/pandas.CategoricalIndex
+generated/pandas.CategoricalIndex.map,../reference/api/pandas.CategoricalIndex.map
+generated/pandas.CategoricalIndex.ordered,../reference/api/pandas.CategoricalIndex.ordered
+generated/pandas.CategoricalIndex.remove_categories,../reference/api/pandas.CategoricalIndex.remove_categories
+generated/pandas.CategoricalIndex.remove_unused_categories,../reference/api/pandas.CategoricalIndex.remove_unused_categories
+generated/pandas.CategoricalIndex.rename_categories,../reference/api/pandas.CategoricalIndex.rename_categories
+generated/pandas.CategoricalIndex.reorder_categories,../reference/api/pandas.CategoricalIndex.reorder_categories
+generated/pandas.CategoricalIndex.set_categories,../reference/api/pandas.CategoricalIndex.set_categories
+generated/pandas.Categorical.ordered,../reference/api/pandas.Categorical.ordered
+generated/pandas.concat,../reference/api/pandas.concat
+generated/pandas.core.groupby.DataFrameGroupBy.all,../reference/api/pandas.core.groupby.DataFrameGroupBy.all
+generated/pandas.core.groupby.DataFrameGroupBy.any,../reference/api/pandas.core.groupby.DataFrameGroupBy.any
+generated/pandas.core.groupby.DataFrameGroupBy.bfill,../reference/api/pandas.core.groupby.DataFrameGroupBy.bfill
+generated/pandas.core.groupby.DataFrameGroupBy.boxplot,../reference/api/pandas.core.groupby.DataFrameGroupBy.boxplot
+generated/pandas.core.groupby.DataFrameGroupBy.corr,../reference/api/pandas.core.groupby.DataFrameGroupBy.corr
+generated/pandas.core.groupby.DataFrameGroupBy.corrwith,../reference/api/pandas.core.groupby.DataFrameGroupBy.corrwith
+generated/pandas.core.groupby.DataFrameGroupBy.count,../reference/api/pandas.core.groupby.DataFrameGroupBy.count
+generated/pandas.core.groupby.DataFrameGroupBy.cov,../reference/api/pandas.core.groupby.DataFrameGroupBy.cov
+generated/pandas.core.groupby.DataFrameGroupBy.cummax,../reference/api/pandas.core.groupby.DataFrameGroupBy.cummax
+generated/pandas.core.groupby.DataFrameGroupBy.cummin,../reference/api/pandas.core.groupby.DataFrameGroupBy.cummin
+generated/pandas.core.groupby.DataFrameGroupBy.cumprod,../reference/api/pandas.core.groupby.DataFrameGroupBy.cumprod
+generated/pandas.core.groupby.DataFrameGroupBy.cumsum,../reference/api/pandas.core.groupby.DataFrameGroupBy.cumsum
+generated/pandas.core.groupby.DataFrameGroupBy.describe,../reference/api/pandas.core.groupby.DataFrameGroupBy.describe
+generated/pandas.core.groupby.DataFrameGroupBy.diff,../reference/api/pandas.core.groupby.DataFrameGroupBy.diff
+generated/pandas.core.groupby.DataFrameGroupBy.ffill,../reference/api/pandas.core.groupby.DataFrameGroupBy.ffill
+generated/pandas.core.groupby.DataFrameGroupBy.fillna,../reference/api/pandas.core.groupby.DataFrameGroupBy.fillna
+generated/pandas.core.groupby.DataFrameGroupBy.filter,../reference/api/pandas.core.groupby.DataFrameGroupBy.filter
+generated/pandas.core.groupby.DataFrameGroupBy.hist,../reference/api/pandas.core.groupby.DataFrameGroupBy.hist
+generated/pandas.core.groupby.DataFrameGroupBy.idxmax,../reference/api/pandas.core.groupby.DataFrameGroupBy.idxmax
+generated/pandas.core.groupby.DataFrameGroupBy.idxmin,../reference/api/pandas.core.groupby.DataFrameGroupBy.idxmin
+generated/pandas.core.groupby.DataFrameGroupBy.pct_change,../reference/api/pandas.core.groupby.DataFrameGroupBy.pct_change
+generated/pandas.core.groupby.DataFrameGroupBy.plot,../reference/api/pandas.core.groupby.DataFrameGroupBy.plot
+generated/pandas.core.groupby.DataFrameGroupBy.quantile,../reference/api/pandas.core.groupby.DataFrameGroupBy.quantile
+generated/pandas.core.groupby.DataFrameGroupBy.rank,../reference/api/pandas.core.groupby.DataFrameGroupBy.rank
+generated/pandas.core.groupby.DataFrameGroupBy.resample,../reference/api/pandas.core.groupby.DataFrameGroupBy.resample
+generated/pandas.core.groupby.DataFrameGroupBy.shift,../reference/api/pandas.core.groupby.DataFrameGroupBy.shift
+generated/pandas.core.groupby.DataFrameGroupBy.size,../reference/api/pandas.core.groupby.DataFrameGroupBy.size
+generated/pandas.core.groupby.DataFrameGroupBy.skew,../reference/api/pandas.core.groupby.DataFrameGroupBy.skew
+generated/pandas.core.groupby.DataFrameGroupBy.take,../reference/api/pandas.core.groupby.DataFrameGroupBy.take
+generated/pandas.core.groupby.GroupBy.agg,../reference/api/pandas.core.groupby.DataFrameGroupBy.agg
+generated/pandas.core.groupby.GroupBy.aggregate,../reference/api/pandas.core.groupby.DataFrameGroupBy.aggregate
+generated/pandas.core.groupby.GroupBy.all,../reference/api/pandas.core.groupby.DataFrameGroupBy.all
+generated/pandas.core.groupby.GroupBy.any,../reference/api/pandas.core.groupby.DataFrameGroupBy.any
+generated/pandas.core.groupby.GroupBy.apply,../reference/api/pandas.core.groupby.DataFrameGroupBy.apply
+generated/pandas.core.groupby.GroupBy.bfill,../reference/api/pandas.core.groupby.DataFrameGroupBy.bfill
+generated/pandas.core.groupby.GroupBy.count,../reference/api/pandas.core.groupby.DataFrameGroupBy.count
+generated/pandas.core.groupby.GroupBy.cumcount,../reference/api/pandas.core.groupby.DataFrameGroupBy.cumcount
+generated/pandas.core.groupby.GroupBy.ffill,../reference/api/pandas.core.groupby.DataFrameGroupBy.ffill
+generated/pandas.core.groupby.GroupBy.first,../reference/api/pandas.core.groupby.DataFrameGroupBy.first
+generated/pandas.core.groupby.GroupBy.get_group,../reference/api/pandas.core.groupby.DataFrameGroupBy.get_group
+generated/pandas.core.groupby.GroupBy.groups,../reference/api/pandas.core.groupby.DataFrameGroupBy.groups
+generated/pandas.core.groupby.GroupBy.head,../reference/api/pandas.core.groupby.DataFrameGroupBy.head
+generated/pandas.core.groupby.GroupBy.indices,../reference/api/pandas.core.groupby.DataFrameGroupBy.indices
+generated/pandas.core.groupby.GroupBy.__iter__,../reference/api/pandas.core.groupby.DataFrameGroupBy.__iter__
+generated/pandas.core.groupby.GroupBy.last,../reference/api/pandas.core.groupby.DataFrameGroupBy.last
+generated/pandas.core.groupby.GroupBy.max,../reference/api/pandas.core.groupby.DataFrameGroupBy.max
+generated/pandas.core.groupby.GroupBy.mean,../reference/api/pandas.core.groupby.DataFrameGroupBy.mean
+generated/pandas.core.groupby.GroupBy.median,../reference/api/pandas.core.groupby.DataFrameGroupBy.median
+generated/pandas.core.groupby.GroupBy.min,../reference/api/pandas.core.groupby.DataFrameGroupBy.min
+generated/pandas.core.groupby.GroupBy.ngroup,../reference/api/pandas.core.groupby.DataFrameGroupBy.ngroup
+generated/pandas.core.groupby.GroupBy.nth,../reference/api/pandas.core.groupby.DataFrameGroupBy.nth
+generated/pandas.core.groupby.GroupBy.ohlc,../reference/api/pandas.core.groupby.DataFrameGroupBy.ohlc
+generated/pandas.core.groupby.GroupBy.pct_change,../reference/api/pandas.core.groupby.DataFrameGroupBy.pct_change
+generated/pandas.core.groupby.GroupBy.pipe,../reference/api/pandas.core.groupby.DataFrameGroupBy.pipe
+generated/pandas.core.groupby.GroupBy.prod,../reference/api/pandas.core.groupby.DataFrameGroupBy.prod
+generated/pandas.core.groupby.GroupBy.rank,../reference/api/pandas.core.groupby.DataFrameGroupBy.rank
+generated/pandas.core.groupby.GroupBy.size,../reference/api/pandas.core.groupby.DataFrameGroupBy.size
+generated/pandas.core.groupby.GroupBy.std,../reference/api/pandas.core.groupby.DataFrameGroupBy.std
+generated/pandas.core.groupby.GroupBy.sum,../reference/api/pandas.core.groupby.DataFrameGroupBy.sum
+generated/pandas.core.groupby.GroupBy.tail,../reference/api/pandas.core.groupby.DataFrameGroupBy.tail
+generated/pandas.core.groupby.GroupBy.transform,../reference/api/pandas.core.groupby.DataFrameGroupBy.transform
+generated/pandas.core.groupby.GroupBy.var,../reference/api/pandas.core.groupby.DataFrameGroupBy.var
+generated/pandas.core.groupby.SeriesGroupBy.is_monotonic_decreasing,../reference/api/pandas.core.groupby.SeriesGroupBy.is_monotonic_decreasing
+generated/pandas.core.groupby.SeriesGroupBy.is_monotonic_increasing,../reference/api/pandas.core.groupby.SeriesGroupBy.is_monotonic_increasing
+generated/pandas.core.groupby.SeriesGroupBy.nlargest,../reference/api/pandas.core.groupby.SeriesGroupBy.nlargest
+generated/pandas.core.groupby.SeriesGroupBy.nsmallest,../reference/api/pandas.core.groupby.SeriesGroupBy.nsmallest
+generated/pandas.core.groupby.SeriesGroupBy.nunique,../reference/api/pandas.core.groupby.SeriesGroupBy.nunique
+generated/pandas.core.groupby.SeriesGroupBy.unique,../reference/api/pandas.core.groupby.SeriesGroupBy.unique
+generated/pandas.core.groupby.SeriesGroupBy.value_counts,../reference/api/pandas.core.groupby.SeriesGroupBy.value_counts
+generated/pandas.core.resample.Resampler.aggregate,../reference/api/pandas.core.resample.Resampler.aggregate
+generated/pandas.core.resample.Resampler.apply,../reference/api/pandas.core.resample.Resampler.apply
+generated/pandas.core.resample.Resampler.asfreq,../reference/api/pandas.core.resample.Resampler.asfreq
+generated/pandas.core.resample.Resampler.backfill,../reference/api/pandas.core.resample.Resampler.bfill
+generated/pandas.core.resample.Resampler.bfill,../reference/api/pandas.core.resample.Resampler.bfill
+generated/pandas.core.resample.Resampler.count,../reference/api/pandas.core.resample.Resampler.count
+generated/pandas.core.resample.Resampler.ffill,../reference/api/pandas.core.resample.Resampler.ffill
+generated/pandas.core.resample.Resampler.first,../reference/api/pandas.core.resample.Resampler.first
+generated/pandas.core.resample.Resampler.get_group,../reference/api/pandas.core.resample.Resampler.get_group
+generated/pandas.core.resample.Resampler.groups,../reference/api/pandas.core.resample.Resampler.groups
+generated/pandas.core.resample.Resampler.indices,../reference/api/pandas.core.resample.Resampler.indices
+generated/pandas.core.resample.Resampler.interpolate,../reference/api/pandas.core.resample.Resampler.interpolate
+generated/pandas.core.resample.Resampler.__iter__,../reference/api/pandas.core.resample.Resampler.__iter__
+generated/pandas.core.resample.Resampler.last,../reference/api/pandas.core.resample.Resampler.last
+generated/pandas.core.resample.Resampler.max,../reference/api/pandas.core.resample.Resampler.max
+generated/pandas.core.resample.Resampler.mean,../reference/api/pandas.core.resample.Resampler.mean
+generated/pandas.core.resample.Resampler.median,../reference/api/pandas.core.resample.Resampler.median
+generated/pandas.core.resample.Resampler.min,../reference/api/pandas.core.resample.Resampler.min
+generated/pandas.core.resample.Resampler.nearest,../reference/api/pandas.core.resample.Resampler.nearest
+generated/pandas.core.resample.Resampler.nunique,../reference/api/pandas.core.resample.Resampler.nunique
+generated/pandas.core.resample.Resampler.ohlc,../reference/api/pandas.core.resample.Resampler.ohlc
+generated/pandas.core.resample.Resampler.pipe,../reference/api/pandas.core.resample.Resampler.pipe
+generated/pandas.core.resample.Resampler.prod,../reference/api/pandas.core.resample.Resampler.prod
+generated/pandas.core.resample.Resampler.quantile,../reference/api/pandas.core.resample.Resampler.quantile
+generated/pandas.core.resample.Resampler.sem,../reference/api/pandas.core.resample.Resampler.sem
+generated/pandas.core.resample.Resampler.size,../reference/api/pandas.core.resample.Resampler.size
+generated/pandas.core.resample.Resampler.std,../reference/api/pandas.core.resample.Resampler.std
+generated/pandas.core.resample.Resampler.sum,../reference/api/pandas.core.resample.Resampler.sum
+generated/pandas.core.resample.Resampler.transform,../reference/api/pandas.core.resample.Resampler.transform
+generated/pandas.core.resample.Resampler.var,../reference/api/pandas.core.resample.Resampler.var
+generated/pandas.core.window.ExponentialMovingWindow.corr,../reference/api/pandas.core.window.ExponentialMovingWindow.corr
+generated/pandas.core.window.ExponentialMovingWindow.cov,../reference/api/pandas.core.window.ExponentialMovingWindow.cov
+generated/pandas.core.window.ExponentialMovingWindow.mean,../reference/api/pandas.core.window.ExponentialMovingWindow.mean
+generated/pandas.core.window.ExponentialMovingWindow.std,../reference/api/pandas.core.window.ExponentialMovingWindow.std
+generated/pandas.core.window.ExponentialMovingWindow.var,../reference/api/pandas.core.window.ExponentialMovingWindow.var
+generated/pandas.core.window.Expanding.aggregate,../reference/api/pandas.core.window.Expanding.aggregate
+generated/pandas.core.window.Expanding.apply,../reference/api/pandas.core.window.Expanding.apply
+generated/pandas.core.window.Expanding.corr,../reference/api/pandas.core.window.Expanding.corr
+generated/pandas.core.window.Expanding.count,../reference/api/pandas.core.window.Expanding.count
+generated/pandas.core.window.Expanding.cov,../reference/api/pandas.core.window.Expanding.cov
+generated/pandas.core.window.Expanding.kurt,../reference/api/pandas.core.window.Expanding.kurt
+generated/pandas.core.window.Expanding.max,../reference/api/pandas.core.window.Expanding.max
+generated/pandas.core.window.Expanding.mean,../reference/api/pandas.core.window.Expanding.mean
+generated/pandas.core.window.Expanding.median,../reference/api/pandas.core.window.Expanding.median
+generated/pandas.core.window.Expanding.min,../reference/api/pandas.core.window.Expanding.min
+generated/pandas.core.window.Expanding.quantile,../reference/api/pandas.core.window.Expanding.quantile
+generated/pandas.core.window.Expanding.skew,../reference/api/pandas.core.window.Expanding.skew
+generated/pandas.core.window.Expanding.std,../reference/api/pandas.core.window.Expanding.std
+generated/pandas.core.window.Expanding.sum,../reference/api/pandas.core.window.Expanding.sum
+generated/pandas.core.window.Expanding.var,../reference/api/pandas.core.window.Expanding.var
+generated/pandas.core.window.Rolling.aggregate,../reference/api/pandas.core.window.rolling.Rolling.aggregate
+generated/pandas.core.window.Rolling.apply,../reference/api/pandas.core.window.rolling.Rolling.apply
+generated/pandas.core.window.Rolling.corr,../reference/api/pandas.core.window.rolling.Rolling.corr
+generated/pandas.core.window.Rolling.count,../reference/api/pandas.core.window.rolling.Rolling.count
+generated/pandas.core.window.Rolling.cov,../reference/api/pandas.core.window.rolling.Rolling.cov
+generated/pandas.core.window.Rolling.kurt,../reference/api/pandas.core.window.rolling.Rolling.kurt
+generated/pandas.core.window.Rolling.max,../reference/api/pandas.core.window.rolling.Rolling.max
+generated/pandas.core.window.Rolling.mean,../reference/api/pandas.core.window.rolling.Rolling.mean
+generated/pandas.core.window.Rolling.median,../reference/api/pandas.core.window.rolling.Rolling.median
+generated/pandas.core.window.Rolling.min,../reference/api/pandas.core.window.rolling.Rolling.min
+generated/pandas.core.window.Rolling.quantile,../reference/api/pandas.core.window.rolling.Rolling.quantile
+generated/pandas.core.window.Rolling.skew,../reference/api/pandas.core.window.rolling.Rolling.skew
+generated/pandas.core.window.Rolling.std,../reference/api/pandas.core.window.rolling.Rolling.std
+generated/pandas.core.window.Rolling.sum,../reference/api/pandas.core.window.rolling.Rolling.sum
+generated/pandas.core.window.Rolling.var,../reference/api/pandas.core.window.rolling.Rolling.var
+generated/pandas.core.window.Window.mean,../reference/api/pandas.core.window.Window.mean
+generated/pandas.core.window.Window.sum,../reference/api/pandas.core.window.Window.sum
+generated/pandas.crosstab,../reference/api/pandas.crosstab
+generated/pandas.cut,../reference/api/pandas.cut
+generated/pandas.DataFrame.abs,../reference/api/pandas.DataFrame.abs
+generated/pandas.DataFrame.add,../reference/api/pandas.DataFrame.add
+generated/pandas.DataFrame.add_prefix,../reference/api/pandas.DataFrame.add_prefix
+generated/pandas.DataFrame.add_suffix,../reference/api/pandas.DataFrame.add_suffix
+generated/pandas.DataFrame.agg,../reference/api/pandas.DataFrame.agg
+generated/pandas.DataFrame.aggregate,../reference/api/pandas.DataFrame.aggregate
+generated/pandas.DataFrame.align,../reference/api/pandas.DataFrame.align
+generated/pandas.DataFrame.all,../reference/api/pandas.DataFrame.all
+generated/pandas.DataFrame.any,../reference/api/pandas.DataFrame.any
+generated/pandas.DataFrame.apply,../reference/api/pandas.DataFrame.apply
+generated/pandas.DataFrame.as_blocks,../reference/api/pandas.DataFrame.as_blocks
+generated/pandas.DataFrame.asfreq,../reference/api/pandas.DataFrame.asfreq
+generated/pandas.DataFrame.as_matrix,../reference/api/pandas.DataFrame.as_matrix
+generated/pandas.DataFrame.asof,../reference/api/pandas.DataFrame.asof
+generated/pandas.DataFrame.assign,../reference/api/pandas.DataFrame.assign
+generated/pandas.DataFrame.astype,../reference/api/pandas.DataFrame.astype
+generated/pandas.DataFrame.at,../reference/api/pandas.DataFrame.at
+generated/pandas.DataFrame.at_time,../reference/api/pandas.DataFrame.at_time
+generated/pandas.DataFrame.axes,../reference/api/pandas.DataFrame.axes
+generated/pandas.DataFrame.between_time,../reference/api/pandas.DataFrame.between_time
+generated/pandas.DataFrame.bfill,../reference/api/pandas.DataFrame.bfill
+generated/pandas.DataFrame.blocks,../reference/api/pandas.DataFrame.blocks
+generated/pandas.DataFrame.boxplot,../reference/api/pandas.DataFrame.boxplot
+generated/pandas.DataFrame.clip,../reference/api/pandas.DataFrame.clip
+generated/pandas.DataFrame.clip_lower,../reference/api/pandas.DataFrame.clip_lower
+generated/pandas.DataFrame.clip_upper,../reference/api/pandas.DataFrame.clip_upper
+generated/pandas.DataFrame.columns,../reference/api/pandas.DataFrame.columns
+generated/pandas.DataFrame.combine_first,../reference/api/pandas.DataFrame.combine_first
+generated/pandas.DataFrame.combine,../reference/api/pandas.DataFrame.combine
+generated/pandas.DataFrame.convert_objects,../reference/api/pandas.DataFrame.convert_objects
+generated/pandas.DataFrame.copy,../reference/api/pandas.DataFrame.copy
+generated/pandas.DataFrame.corr,../reference/api/pandas.DataFrame.corr
+generated/pandas.DataFrame.corrwith,../reference/api/pandas.DataFrame.corrwith
+generated/pandas.DataFrame.count,../reference/api/pandas.DataFrame.count
+generated/pandas.DataFrame.cov,../reference/api/pandas.DataFrame.cov
+generated/pandas.DataFrame.cummax,../reference/api/pandas.DataFrame.cummax
+generated/pandas.DataFrame.cummin,../reference/api/pandas.DataFrame.cummin
+generated/pandas.DataFrame.cumprod,../reference/api/pandas.DataFrame.cumprod
+generated/pandas.DataFrame.cumsum,../reference/api/pandas.DataFrame.cumsum
+generated/pandas.DataFrame.describe,../reference/api/pandas.DataFrame.describe
+generated/pandas.DataFrame.diff,../reference/api/pandas.DataFrame.diff
+generated/pandas.DataFrame.div,../reference/api/pandas.DataFrame.div
+generated/pandas.DataFrame.divide,../reference/api/pandas.DataFrame.divide
+generated/pandas.DataFrame.dot,../reference/api/pandas.DataFrame.dot
+generated/pandas.DataFrame.drop_duplicates,../reference/api/pandas.DataFrame.drop_duplicates
+generated/pandas.DataFrame.drop,../reference/api/pandas.DataFrame.drop
+generated/pandas.DataFrame.droplevel,../reference/api/pandas.DataFrame.droplevel
+generated/pandas.DataFrame.dropna,../reference/api/pandas.DataFrame.dropna
+generated/pandas.DataFrame.dtypes,../reference/api/pandas.DataFrame.dtypes
+generated/pandas.DataFrame.duplicated,../reference/api/pandas.DataFrame.duplicated
+generated/pandas.DataFrame.empty,../reference/api/pandas.DataFrame.empty
+generated/pandas.DataFrame.eq,../reference/api/pandas.DataFrame.eq
+generated/pandas.DataFrame.equals,../reference/api/pandas.DataFrame.equals
+generated/pandas.DataFrame.eval,../reference/api/pandas.DataFrame.eval
+generated/pandas.DataFrame.ewm,../reference/api/pandas.DataFrame.ewm
+generated/pandas.DataFrame.expanding,../reference/api/pandas.DataFrame.expanding
+generated/pandas.DataFrame.ffill,../reference/api/pandas.DataFrame.ffill
+generated/pandas.DataFrame.fillna,../reference/api/pandas.DataFrame.fillna
+generated/pandas.DataFrame.filter,../reference/api/pandas.DataFrame.filter
+generated/pandas.DataFrame.first,../reference/api/pandas.DataFrame.first
+generated/pandas.DataFrame.first_valid_index,../reference/api/pandas.DataFrame.first_valid_index
+generated/pandas.DataFrame.floordiv,../reference/api/pandas.DataFrame.floordiv
+generated/pandas.DataFrame.from_csv,../reference/api/pandas.DataFrame.from_csv
+generated/pandas.DataFrame.from_dict,../reference/api/pandas.DataFrame.from_dict
+generated/pandas.DataFrame.from_items,../reference/api/pandas.DataFrame.from_items
+generated/pandas.DataFrame.from_records,../reference/api/pandas.DataFrame.from_records
+generated/pandas.DataFrame.ge,../reference/api/pandas.DataFrame.ge
+generated/pandas.DataFrame.get,../reference/api/pandas.DataFrame.get
+generated/pandas.DataFrame.get_value,../reference/api/pandas.DataFrame.get_value
+generated/pandas.DataFrame.groupby,../reference/api/pandas.DataFrame.groupby
+generated/pandas.DataFrame.gt,../reference/api/pandas.DataFrame.gt
+generated/pandas.DataFrame.head,../reference/api/pandas.DataFrame.head
+generated/pandas.DataFrame.hist,../reference/api/pandas.DataFrame.hist
+generated/pandas.DataFrame,../reference/api/pandas.DataFrame
+generated/pandas.DataFrame.iat,../reference/api/pandas.DataFrame.iat
+generated/pandas.DataFrame.idxmax,../reference/api/pandas.DataFrame.idxmax
+generated/pandas.DataFrame.idxmin,../reference/api/pandas.DataFrame.idxmin
+generated/pandas.DataFrame.iloc,../reference/api/pandas.DataFrame.iloc
+generated/pandas.DataFrame.index,../reference/api/pandas.DataFrame.index
+generated/pandas.DataFrame.infer_objects,../reference/api/pandas.DataFrame.infer_objects
+generated/pandas.DataFrame.info,../reference/api/pandas.DataFrame.info
+generated/pandas.DataFrame.insert,../reference/api/pandas.DataFrame.insert
+generated/pandas.DataFrame.interpolate,../reference/api/pandas.DataFrame.interpolate
+generated/pandas.DataFrame.is_copy,../reference/api/pandas.DataFrame.is_copy
+generated/pandas.DataFrame.isin,../reference/api/pandas.DataFrame.isin
+generated/pandas.DataFrame.isna,../reference/api/pandas.DataFrame.isna
+generated/pandas.DataFrame.isnull,../reference/api/pandas.DataFrame.isnull
+generated/pandas.DataFrame.items,../reference/api/pandas.DataFrame.items
+generated/pandas.DataFrame.__iter__,../reference/api/pandas.DataFrame.__iter__
+generated/pandas.DataFrame.iterrows,../reference/api/pandas.DataFrame.iterrows
+generated/pandas.DataFrame.itertuples,../reference/api/pandas.DataFrame.itertuples
+generated/pandas.DataFrame.join,../reference/api/pandas.DataFrame.join
+generated/pandas.DataFrame.keys,../reference/api/pandas.DataFrame.keys
+generated/pandas.DataFrame.kurt,../reference/api/pandas.DataFrame.kurt
+generated/pandas.DataFrame.kurtosis,../reference/api/pandas.DataFrame.kurtosis
+generated/pandas.DataFrame.last,../reference/api/pandas.DataFrame.last
+generated/pandas.DataFrame.last_valid_index,../reference/api/pandas.DataFrame.last_valid_index
+generated/pandas.DataFrame.le,../reference/api/pandas.DataFrame.le
+generated/pandas.DataFrame.loc,../reference/api/pandas.DataFrame.loc
+generated/pandas.DataFrame.lt,../reference/api/pandas.DataFrame.lt
+generated/pandas.DataFrame.mask,../reference/api/pandas.DataFrame.mask
+generated/pandas.DataFrame.max,../reference/api/pandas.DataFrame.max
+generated/pandas.DataFrame.mean,../reference/api/pandas.DataFrame.mean
+generated/pandas.DataFrame.median,../reference/api/pandas.DataFrame.median
+generated/pandas.DataFrame.melt,../reference/api/pandas.DataFrame.melt
+generated/pandas.DataFrame.memory_usage,../reference/api/pandas.DataFrame.memory_usage
+generated/pandas.DataFrame.merge,../reference/api/pandas.DataFrame.merge
+generated/pandas.DataFrame.min,../reference/api/pandas.DataFrame.min
+generated/pandas.DataFrame.mode,../reference/api/pandas.DataFrame.mode
+generated/pandas.DataFrame.mod,../reference/api/pandas.DataFrame.mod
+generated/pandas.DataFrame.mul,../reference/api/pandas.DataFrame.mul
+generated/pandas.DataFrame.multiply,../reference/api/pandas.DataFrame.multiply
+generated/pandas.DataFrame.ndim,../reference/api/pandas.DataFrame.ndim
+generated/pandas.DataFrame.ne,../reference/api/pandas.DataFrame.ne
+generated/pandas.DataFrame.nlargest,../reference/api/pandas.DataFrame.nlargest
+generated/pandas.DataFrame.notna,../reference/api/pandas.DataFrame.notna
+generated/pandas.DataFrame.notnull,../reference/api/pandas.DataFrame.notnull
+generated/pandas.DataFrame.nsmallest,../reference/api/pandas.DataFrame.nsmallest
+generated/pandas.DataFrame.nunique,../reference/api/pandas.DataFrame.nunique
+generated/pandas.DataFrame.pct_change,../reference/api/pandas.DataFrame.pct_change
+generated/pandas.DataFrame.pipe,../reference/api/pandas.DataFrame.pipe
+generated/pandas.DataFrame.pivot,../reference/api/pandas.DataFrame.pivot
+generated/pandas.DataFrame.pivot_table,../reference/api/pandas.DataFrame.pivot_table
+generated/pandas.DataFrame.plot.barh,../reference/api/pandas.DataFrame.plot.barh
+generated/pandas.DataFrame.plot.bar,../reference/api/pandas.DataFrame.plot.bar
+generated/pandas.DataFrame.plot.box,../reference/api/pandas.DataFrame.plot.box
+generated/pandas.DataFrame.plot.density,../reference/api/pandas.DataFrame.plot.density
+generated/pandas.DataFrame.plot.hexbin,../reference/api/pandas.DataFrame.plot.hexbin
+generated/pandas.DataFrame.plot.hist,../reference/api/pandas.DataFrame.plot.hist
+generated/pandas.DataFrame.plot,../reference/api/pandas.DataFrame.plot
+generated/pandas.DataFrame.plot.kde,../reference/api/pandas.DataFrame.plot.kde
+generated/pandas.DataFrame.plot.line,../reference/api/pandas.DataFrame.plot.line
+generated/pandas.DataFrame.plot.pie,../reference/api/pandas.DataFrame.plot.pie
+generated/pandas.DataFrame.plot.scatter,../reference/api/pandas.DataFrame.plot.scatter
+generated/pandas.DataFrame.pop,../reference/api/pandas.DataFrame.pop
+generated/pandas.DataFrame.pow,../reference/api/pandas.DataFrame.pow
+generated/pandas.DataFrame.prod,../reference/api/pandas.DataFrame.prod
+generated/pandas.DataFrame.product,../reference/api/pandas.DataFrame.product
+generated/pandas.DataFrame.quantile,../reference/api/pandas.DataFrame.quantile
+generated/pandas.DataFrame.query,../reference/api/pandas.DataFrame.query
+generated/pandas.DataFrame.radd,../reference/api/pandas.DataFrame.radd
+generated/pandas.DataFrame.rank,../reference/api/pandas.DataFrame.rank
+generated/pandas.DataFrame.rdiv,../reference/api/pandas.DataFrame.rdiv
+generated/pandas.DataFrame.reindex_axis,../reference/api/pandas.DataFrame.reindex_axis
+generated/pandas.DataFrame.reindex,../reference/api/pandas.DataFrame.reindex
+generated/pandas.DataFrame.reindex_like,../reference/api/pandas.DataFrame.reindex_like
+generated/pandas.DataFrame.rename_axis,../reference/api/pandas.DataFrame.rename_axis
+generated/pandas.DataFrame.rename,../reference/api/pandas.DataFrame.rename
+generated/pandas.DataFrame.reorder_levels,../reference/api/pandas.DataFrame.reorder_levels
+generated/pandas.DataFrame.replace,../reference/api/pandas.DataFrame.replace
+generated/pandas.DataFrame.resample,../reference/api/pandas.DataFrame.resample
+generated/pandas.DataFrame.reset_index,../reference/api/pandas.DataFrame.reset_index
+generated/pandas.DataFrame.rfloordiv,../reference/api/pandas.DataFrame.rfloordiv
+generated/pandas.DataFrame.rmod,../reference/api/pandas.DataFrame.rmod
+generated/pandas.DataFrame.rmul,../reference/api/pandas.DataFrame.rmul
+generated/pandas.DataFrame.rolling,../reference/api/pandas.DataFrame.rolling
+generated/pandas.DataFrame.round,../reference/api/pandas.DataFrame.round
+generated/pandas.DataFrame.rpow,../reference/api/pandas.DataFrame.rpow
+generated/pandas.DataFrame.rsub,../reference/api/pandas.DataFrame.rsub
+generated/pandas.DataFrame.rtruediv,../reference/api/pandas.DataFrame.rtruediv
+generated/pandas.DataFrame.sample,../reference/api/pandas.DataFrame.sample
+generated/pandas.DataFrame.select_dtypes,../reference/api/pandas.DataFrame.select_dtypes
+generated/pandas.DataFrame.select,../reference/api/pandas.DataFrame.select
+generated/pandas.DataFrame.sem,../reference/api/pandas.DataFrame.sem
+generated/pandas.DataFrame.set_axis,../reference/api/pandas.DataFrame.set_axis
+generated/pandas.DataFrame.set_index,../reference/api/pandas.DataFrame.set_index
+generated/pandas.DataFrame.set_value,../reference/api/pandas.DataFrame.set_value
+generated/pandas.DataFrame.shape,../reference/api/pandas.DataFrame.shape
+generated/pandas.DataFrame.shift,../reference/api/pandas.DataFrame.shift
+generated/pandas.DataFrame.size,../reference/api/pandas.DataFrame.size
+generated/pandas.DataFrame.skew,../reference/api/pandas.DataFrame.skew
+generated/pandas.DataFrame.sort_index,../reference/api/pandas.DataFrame.sort_index
+generated/pandas.DataFrame.sort_values,../reference/api/pandas.DataFrame.sort_values
+generated/pandas.DataFrame.squeeze,../reference/api/pandas.DataFrame.squeeze
+generated/pandas.DataFrame.stack,../reference/api/pandas.DataFrame.stack
+generated/pandas.DataFrame.std,../reference/api/pandas.DataFrame.std
+generated/pandas.DataFrame.style,../reference/api/pandas.DataFrame.style
+generated/pandas.DataFrame.sub,../reference/api/pandas.DataFrame.sub
+generated/pandas.DataFrame.subtract,../reference/api/pandas.DataFrame.subtract
+generated/pandas.DataFrame.sum,../reference/api/pandas.DataFrame.sum
+generated/pandas.DataFrame.swaplevel,../reference/api/pandas.DataFrame.swaplevel
+generated/pandas.DataFrame.tail,../reference/api/pandas.DataFrame.tail
+generated/pandas.DataFrame.take,../reference/api/pandas.DataFrame.take
+generated/pandas.DataFrame.T,../reference/api/pandas.DataFrame.T
+generated/pandas.DataFrame.timetuple,../reference/api/pandas.DataFrame.timetuple
+generated/pandas.DataFrame.to_clipboard,../reference/api/pandas.DataFrame.to_clipboard
+generated/pandas.DataFrame.to_csv,../reference/api/pandas.DataFrame.to_csv
+generated/pandas.DataFrame.to_dict,../reference/api/pandas.DataFrame.to_dict
+generated/pandas.DataFrame.to_excel,../reference/api/pandas.DataFrame.to_excel
+generated/pandas.DataFrame.to_feather,../reference/api/pandas.DataFrame.to_feather
+generated/pandas.DataFrame.to_hdf,../reference/api/pandas.DataFrame.to_hdf
+generated/pandas.DataFrame.to,../reference/api/pandas.DataFrame.to
+generated/pandas.DataFrame.to_json,../reference/api/pandas.DataFrame.to_json
+generated/pandas.DataFrame.to_latex,../reference/api/pandas.DataFrame.to_latex
+generated/pandas.DataFrame.to_numpy,../reference/api/pandas.DataFrame.to_numpy
+generated/pandas.DataFrame.to_panel,../reference/api/pandas.DataFrame.to_panel
+generated/pandas.DataFrame.to_parquet,../reference/api/pandas.DataFrame.to_parquet
+generated/pandas.DataFrame.to_period,../reference/api/pandas.DataFrame.to_period
+generated/pandas.DataFrame.to_pickle,../reference/api/pandas.DataFrame.to_pickle
+generated/pandas.DataFrame.to_records,../reference/api/pandas.DataFrame.to_records
+generated/pandas.DataFrame.to_sql,../reference/api/pandas.DataFrame.to_sql
+generated/pandas.DataFrame.to_stata,../reference/api/pandas.DataFrame.to_stata
+generated/pandas.DataFrame.to_string,../reference/api/pandas.DataFrame.to_string
+generated/pandas.DataFrame.to_timestamp,../reference/api/pandas.DataFrame.to_timestamp
+generated/pandas.DataFrame.to_xarray,../reference/api/pandas.DataFrame.to_xarray
+generated/pandas.DataFrame.transform,../reference/api/pandas.DataFrame.transform
+generated/pandas.DataFrame.transpose,../reference/api/pandas.DataFrame.transpose
+generated/pandas.DataFrame.truediv,../reference/api/pandas.DataFrame.truediv
+generated/pandas.DataFrame.truncate,../reference/api/pandas.DataFrame.truncate
+generated/pandas.DataFrame.tz_convert,../reference/api/pandas.DataFrame.tz_convert
+generated/pandas.DataFrame.tz_localize,../reference/api/pandas.DataFrame.tz_localize
+generated/pandas.DataFrame.unstack,../reference/api/pandas.DataFrame.unstack
+generated/pandas.DataFrame.update,../reference/api/pandas.DataFrame.update
+generated/pandas.DataFrame.values,../reference/api/pandas.DataFrame.values
+generated/pandas.DataFrame.var,../reference/api/pandas.DataFrame.var
+generated/pandas.DataFrame.where,../reference/api/pandas.DataFrame.where
+generated/pandas.DataFrame.xs,../reference/api/pandas.DataFrame.xs
+generated/pandas.date_range,../reference/api/pandas.date_range
+generated/pandas.DatetimeIndex.ceil,../reference/api/pandas.DatetimeIndex.ceil
+generated/pandas.DatetimeIndex.date,../reference/api/pandas.DatetimeIndex.date
+generated/pandas.DatetimeIndex.day,../reference/api/pandas.DatetimeIndex.day
+generated/pandas.DatetimeIndex.day_name,../reference/api/pandas.DatetimeIndex.day_name
+generated/pandas.DatetimeIndex.dayofweek,../reference/api/pandas.DatetimeIndex.dayofweek
+generated/pandas.DatetimeIndex.day_of_week,../reference/api/pandas.DatetimeIndex.day_of_week
+generated/pandas.DatetimeIndex.dayofyear,../reference/api/pandas.DatetimeIndex.dayofyear
+generated/pandas.DatetimeIndex.day_of_year,../reference/api/pandas.DatetimeIndex.day_of_year
+generated/pandas.DatetimeIndex.floor,../reference/api/pandas.DatetimeIndex.floor
+generated/pandas.DatetimeIndex.freq,../reference/api/pandas.DatetimeIndex.freq
+generated/pandas.DatetimeIndex.freqstr,../reference/api/pandas.DatetimeIndex.freqstr
+generated/pandas.DatetimeIndex.hour,../reference/api/pandas.DatetimeIndex.hour
+generated/pandas.DatetimeIndex,../reference/api/pandas.DatetimeIndex
+generated/pandas.DatetimeIndex.indexer_at_time,../reference/api/pandas.DatetimeIndex.indexer_at_time
+generated/pandas.DatetimeIndex.indexer_between_time,../reference/api/pandas.DatetimeIndex.indexer_between_time
+generated/pandas.DatetimeIndex.inferred_freq,../reference/api/pandas.DatetimeIndex.inferred_freq
+generated/pandas.DatetimeIndex.is_leap_year,../reference/api/pandas.DatetimeIndex.is_leap_year
+generated/pandas.DatetimeIndex.is_month_end,../reference/api/pandas.DatetimeIndex.is_month_end
+generated/pandas.DatetimeIndex.is_month_start,../reference/api/pandas.DatetimeIndex.is_month_start
+generated/pandas.DatetimeIndex.is_quarter_end,../reference/api/pandas.DatetimeIndex.is_quarter_end
+generated/pandas.DatetimeIndex.is_quarter_start,../reference/api/pandas.DatetimeIndex.is_quarter_start
+generated/pandas.DatetimeIndex.is_year_end,../reference/api/pandas.DatetimeIndex.is_year_end
+generated/pandas.DatetimeIndex.is_year_start,../reference/api/pandas.DatetimeIndex.is_year_start
+generated/pandas.DatetimeIndex.microsecond,../reference/api/pandas.DatetimeIndex.microsecond
+generated/pandas.DatetimeIndex.minute,../reference/api/pandas.DatetimeIndex.minute
+generated/pandas.DatetimeIndex.month,../reference/api/pandas.DatetimeIndex.month
+generated/pandas.DatetimeIndex.month_name,../reference/api/pandas.DatetimeIndex.month_name
+generated/pandas.DatetimeIndex.nanosecond,../reference/api/pandas.DatetimeIndex.nanosecond
+generated/pandas.DatetimeIndex.normalize,../reference/api/pandas.DatetimeIndex.normalize
+generated/pandas.DatetimeIndex.quarter,../reference/api/pandas.DatetimeIndex.quarter
+generated/pandas.DatetimeIndex.round,../reference/api/pandas.DatetimeIndex.round
+generated/pandas.DatetimeIndex.second,../reference/api/pandas.DatetimeIndex.second
+generated/pandas.DatetimeIndex.snap,../reference/api/pandas.DatetimeIndex.snap
+generated/pandas.DatetimeIndex.strftime,../reference/api/pandas.DatetimeIndex.strftime
+generated/pandas.DatetimeIndex.time,../reference/api/pandas.DatetimeIndex.time
+generated/pandas.DatetimeIndex.timetz,../reference/api/pandas.DatetimeIndex.timetz
+generated/pandas.DatetimeIndex.to_frame,../reference/api/pandas.DatetimeIndex.to_frame
+generated/pandas.DatetimeIndex.to_period,../reference/api/pandas.DatetimeIndex.to_period
+generated/pandas.DatetimeIndex.to_pydatetime,../reference/api/pandas.DatetimeIndex.to_pydatetime
+generated/pandas.DatetimeIndex.to_series,../reference/api/pandas.DatetimeIndex.to_series
+generated/pandas.DatetimeIndex.tz_convert,../reference/api/pandas.DatetimeIndex.tz_convert
+generated/pandas.DatetimeIndex.tz,../reference/api/pandas.DatetimeIndex.tz
+generated/pandas.DatetimeIndex.tz_localize,../reference/api/pandas.DatetimeIndex.tz_localize
+generated/pandas.DatetimeIndex.weekday,../reference/api/pandas.DatetimeIndex.weekday
+generated/pandas.DatetimeIndex.week,../reference/api/pandas.DatetimeIndex.week
+generated/pandas.DatetimeIndex.weekofyear,../reference/api/pandas.DatetimeIndex.weekofyear
+generated/pandas.DatetimeIndex.year,../reference/api/pandas.DatetimeIndex.year
+generated/pandas.DatetimeTZDtype.base,../reference/api/pandas.DatetimeTZDtype.base
+generated/pandas.DatetimeTZDtype.construct_array_type,../reference/api/pandas.DatetimeTZDtype.construct_array_type
+generated/pandas.DatetimeTZDtype.construct_from_string,../reference/api/pandas.DatetimeTZDtype.construct_from_string
+generated/pandas.DatetimeTZDtype,../reference/api/pandas.DatetimeTZDtype
+generated/pandas.DatetimeTZDtype.isbuiltin,../reference/api/pandas.DatetimeTZDtype.isbuiltin
+generated/pandas.DatetimeTZDtype.is_dtype,../reference/api/pandas.DatetimeTZDtype.is_dtype
+generated/pandas.DatetimeTZDtype.isnative,../reference/api/pandas.DatetimeTZDtype.isnative
+generated/pandas.DatetimeTZDtype.itemsize,../reference/api/pandas.DatetimeTZDtype.itemsize
+generated/pandas.DatetimeTZDtype.kind,../reference/api/pandas.DatetimeTZDtype.kind
+generated/pandas.DatetimeTZDtype.name,../reference/api/pandas.DatetimeTZDtype.name
+generated/pandas.DatetimeTZDtype.names,../reference/api/pandas.DatetimeTZDtype.names
+generated/pandas.DatetimeTZDtype.na_value,../reference/api/pandas.DatetimeTZDtype.na_value
+generated/pandas.DatetimeTZDtype.num,../reference/api/pandas.DatetimeTZDtype.num
+generated/pandas.DatetimeTZDtype.reset_cache,../reference/api/pandas.DatetimeTZDtype.reset_cache
+generated/pandas.DatetimeTZDtype.shape,../reference/api/pandas.DatetimeTZDtype.shape
+generated/pandas.DatetimeTZDtype.str,../reference/api/pandas.DatetimeTZDtype.str
+generated/pandas.DatetimeTZDtype.subdtype,../reference/api/pandas.DatetimeTZDtype.subdtype
+generated/pandas.DatetimeTZDtype.tz,../reference/api/pandas.DatetimeTZDtype.tz
+generated/pandas.DatetimeTZDtype.unit,../reference/api/pandas.DatetimeTZDtype.unit
+generated/pandas.describe_option,../reference/api/pandas.describe_option
+generated/pandas.errors.DtypeWarning,../reference/api/pandas.errors.DtypeWarning
+generated/pandas.errors.EmptyDataError,../reference/api/pandas.errors.EmptyDataError
+generated/pandas.errors.OutOfBoundsDatetime,../reference/api/pandas.errors.OutOfBoundsDatetime
+generated/pandas.errors.ParserError,../reference/api/pandas.errors.ParserError
+generated/pandas.errors.ParserWarning,../reference/api/pandas.errors.ParserWarning
+generated/pandas.errors.PerformanceWarning,../reference/api/pandas.errors.PerformanceWarning
+generated/pandas.errors.UnsortedIndexError,../reference/api/pandas.errors.UnsortedIndexError
+generated/pandas.errors.UnsupportedFunctionCall,../reference/api/pandas.errors.UnsupportedFunctionCall
+generated/pandas.eval,../reference/api/pandas.eval
+generated/pandas.ExcelFile.parse,../reference/api/pandas.ExcelFile.parse
+generated/pandas.ExcelWriter,../reference/api/pandas.ExcelWriter
+generated/pandas.factorize,../reference/api/pandas.factorize
+generated/pandas.Float64Index,../reference/api/pandas.Float64Index
+generated/pandas.get_dummies,../reference/api/pandas.get_dummies
+generated/pandas.get_option,../reference/api/pandas.get_option
+generated/pandas.Grouper,../reference/api/pandas.Grouper
+generated/pandas.HDFStore.append,../reference/api/pandas.HDFStore.append
+generated/pandas.HDFStore.get,../reference/api/pandas.HDFStore.get
+generated/pandas.HDFStore.groups,../reference/api/pandas.HDFStore.groups
+generated/pandas.HDFStore.info,../reference/api/pandas.HDFStore.info
+generated/pandas.HDFStore.keys,../reference/api/pandas.HDFStore.keys
+generated/pandas.HDFStore.put,../reference/api/pandas.HDFStore.put
+generated/pandas.HDFStore.select,../reference/api/pandas.HDFStore.select
+generated/pandas.HDFStore.walk,../reference/api/pandas.HDFStore.walk
+generated/pandas.Index.all,../reference/api/pandas.Index.all
+generated/pandas.Index.any,../reference/api/pandas.Index.any
+generated/pandas.Index.append,../reference/api/pandas.Index.append
+generated/pandas.Index.argmax,../reference/api/pandas.Index.argmax
+generated/pandas.Index.argmin,../reference/api/pandas.Index.argmin
+generated/pandas.Index.argsort,../reference/api/pandas.Index.argsort
+generated/pandas.Index.array,../reference/api/pandas.Index.array
+generated/pandas.Index.asof,../reference/api/pandas.Index.asof
+generated/pandas.Index.asof_locs,../reference/api/pandas.Index.asof_locs
+generated/pandas.Index.astype,../reference/api/pandas.Index.astype
+generated/pandas.Index.copy,../reference/api/pandas.Index.copy
+generated/pandas.Index.data,../reference/api/pandas.Index.data
+generated/pandas.Index.delete,../reference/api/pandas.Index.delete
+generated/pandas.Index.difference,../reference/api/pandas.Index.difference
+generated/pandas.Index.drop_duplicates,../reference/api/pandas.Index.drop_duplicates
+generated/pandas.Index.drop,../reference/api/pandas.Index.drop
+generated/pandas.Index.droplevel,../reference/api/pandas.Index.droplevel
+generated/pandas.Index.dropna,../reference/api/pandas.Index.dropna
+generated/pandas.Index.dtype,../reference/api/pandas.Index.dtype
+generated/pandas.Index.duplicated,../reference/api/pandas.Index.duplicated
+generated/pandas.Index.empty,../reference/api/pandas.Index.empty
+generated/pandas.Index.equals,../reference/api/pandas.Index.equals
+generated/pandas.Index.factorize,../reference/api/pandas.Index.factorize
+generated/pandas.Index.fillna,../reference/api/pandas.Index.fillna
+generated/pandas.Index.format,../reference/api/pandas.Index.format
+generated/pandas.Index.get_indexer_for,../reference/api/pandas.Index.get_indexer_for
+generated/pandas.Index.get_indexer,../reference/api/pandas.Index.get_indexer
+generated/pandas.Index.get_indexer_non_unique,../reference/api/pandas.Index.get_indexer_non_unique
+generated/pandas.Index.get_level_values,../reference/api/pandas.Index.get_level_values
+generated/pandas.Index.get_loc,../reference/api/pandas.Index.get_loc
+generated/pandas.Index.get_slice_bound,../reference/api/pandas.Index.get_slice_bound
+generated/pandas.Index.groupby,../reference/api/pandas.Index.groupby
+generated/pandas.Index.has_duplicates,../reference/api/pandas.Index.has_duplicates
+generated/pandas.Index.hasnans,../reference/api/pandas.Index.hasnans
+generated/pandas.Index.holds_integer,../reference/api/pandas.Index.holds_integer
+generated/pandas.Index,../reference/api/pandas.Index
+generated/pandas.Index.identical,../reference/api/pandas.Index.identical
+generated/pandas.Index.inferred_type,../reference/api/pandas.Index.inferred_type
+generated/pandas.Index.insert,../reference/api/pandas.Index.insert
+generated/pandas.Index.intersection,../reference/api/pandas.Index.intersection
+generated/pandas.Index.is_boolean,../reference/api/pandas.Index.is_boolean
+generated/pandas.Index.is_categorical,../reference/api/pandas.Index.is_categorical
+generated/pandas.Index.is_floating,../reference/api/pandas.Index.is_floating
+generated/pandas.Index.is_,../reference/api/pandas.Index.is_
+generated/pandas.Index.isin,../reference/api/pandas.Index.isin
+generated/pandas.Index.is_integer,../reference/api/pandas.Index.is_integer
+generated/pandas.Index.is_interval,../reference/api/pandas.Index.is_interval
+generated/pandas.Index.is_lexsorted_for_tuple,../reference/api/pandas.Index.is_lexsorted_for_tuple
+generated/pandas.Index.is_monotonic_decreasing,../reference/api/pandas.Index.is_monotonic_decreasing
+generated/pandas.Index.is_monotonic_increasing,../reference/api/pandas.Index.is_monotonic_increasing
+generated/pandas.Index.isna,../reference/api/pandas.Index.isna
+generated/pandas.Index.isnull,../reference/api/pandas.Index.isnull
+generated/pandas.Index.is_numeric,../reference/api/pandas.Index.is_numeric
+generated/pandas.Index.is_object,../reference/api/pandas.Index.is_object
+generated/pandas.Index.is_unique,../reference/api/pandas.Index.is_unique
+generated/pandas.Index.item,../reference/api/pandas.Index.item
+generated/pandas.Index.join,../reference/api/pandas.Index.join
+generated/pandas.Index.map,../reference/api/pandas.Index.map
+generated/pandas.Index.max,../reference/api/pandas.Index.max
+generated/pandas.Index.memory_usage,../reference/api/pandas.Index.memory_usage
+generated/pandas.Index.min,../reference/api/pandas.Index.min
+generated/pandas.Index.name,../reference/api/pandas.Index.name
+generated/pandas.Index.names,../reference/api/pandas.Index.names
+generated/pandas.Index.nbytes,../reference/api/pandas.Index.nbytes
+generated/pandas.Index.ndim,../reference/api/pandas.Index.ndim
+generated/pandas.Index.nlevels,../reference/api/pandas.Index.nlevels
+generated/pandas.Index.notna,../reference/api/pandas.Index.notna
+generated/pandas.Index.notnull,../reference/api/pandas.Index.notnull
+generated/pandas.Index.nunique,../reference/api/pandas.Index.nunique
+generated/pandas.Index.putmask,../reference/api/pandas.Index.putmask
+generated/pandas.Index.ravel,../reference/api/pandas.Index.ravel
+generated/pandas.Index.reindex,../reference/api/pandas.Index.reindex
+generated/pandas.Index.rename,../reference/api/pandas.Index.rename
+generated/pandas.Index.repeat,../reference/api/pandas.Index.repeat
+generated/pandas.Index.searchsorted,../reference/api/pandas.Index.searchsorted
+generated/pandas.Index.set_names,../reference/api/pandas.Index.set_names
+generated/pandas.Index.shape,../reference/api/pandas.Index.shape
+generated/pandas.Index.shift,../reference/api/pandas.Index.shift
+generated/pandas.Index.size,../reference/api/pandas.Index.size
+generated/pandas.IndexSlice,../reference/api/pandas.IndexSlice
+generated/pandas.Index.slice_indexer,../reference/api/pandas.Index.slice_indexer
+generated/pandas.Index.slice_locs,../reference/api/pandas.Index.slice_locs
+generated/pandas.Index.sort,../reference/api/pandas.Index.sort
+generated/pandas.Index.sortlevel,../reference/api/pandas.Index.sortlevel
+generated/pandas.Index.sort_values,../reference/api/pandas.Index.sort_values
+generated/pandas.Index.str,../reference/api/pandas.Index.str
+generated/pandas.Index.summary,../reference/api/pandas.Index.summary
+generated/pandas.Index.symmetric_difference,../reference/api/pandas.Index.symmetric_difference
+generated/pandas.Index.take,../reference/api/pandas.Index.take
+generated/pandas.Index.T,../reference/api/pandas.Index.T
+generated/pandas.Index.to_flat_index,../reference/api/pandas.MultiIndex.to_flat_index
+generated/pandas.Index.to_frame,../reference/api/pandas.Index.to_frame
+generated/pandas.Index.to_list,../reference/api/pandas.Index.to_list
+generated/pandas.Index.tolist,../reference/api/pandas.Index.tolist
+generated/pandas.Index.to_numpy,../reference/api/pandas.Index.to_numpy
+generated/pandas.Index.to_series,../reference/api/pandas.Index.to_series
+generated/pandas.Index.transpose,../reference/api/pandas.Index.transpose
+generated/pandas.Index.union,../reference/api/pandas.Index.union
+generated/pandas.Index.unique,../reference/api/pandas.Index.unique
+generated/pandas.Index.value_counts,../reference/api/pandas.Index.value_counts
+generated/pandas.Index.values,../reference/api/pandas.Index.values
+generated/pandas.Index.view,../reference/api/pandas.Index.view
+generated/pandas.Index.where,../reference/api/pandas.Index.where
+generated/pandas.infer_freq,../reference/api/pandas.infer_freq
+generated/pandas.Interval.closed,../reference/api/pandas.Interval.closed
+generated/pandas.Interval.closed_left,../reference/api/pandas.Interval.closed_left
+generated/pandas.Interval.closed_right,../reference/api/pandas.Interval.closed_right
+generated/pandas.Interval,../reference/api/pandas.Interval
+generated/pandas.IntervalIndex.closed,../reference/api/pandas.IntervalIndex.closed
+generated/pandas.IntervalIndex.contains,../reference/api/pandas.IntervalIndex.contains
+generated/pandas.IntervalIndex.from_arrays,../reference/api/pandas.IntervalIndex.from_arrays
+generated/pandas.IntervalIndex.from_breaks,../reference/api/pandas.IntervalIndex.from_breaks
+generated/pandas.IntervalIndex.from_tuples,../reference/api/pandas.IntervalIndex.from_tuples
+generated/pandas.IntervalIndex.get_indexer,../reference/api/pandas.IntervalIndex.get_indexer
+generated/pandas.IntervalIndex.get_loc,../reference/api/pandas.IntervalIndex.get_loc
+generated/pandas.IntervalIndex,../reference/api/pandas.IntervalIndex
+generated/pandas.IntervalIndex.is_non_overlapping_monotonic,../reference/api/pandas.IntervalIndex.is_non_overlapping_monotonic
+generated/pandas.IntervalIndex.is_overlapping,../reference/api/pandas.IntervalIndex.is_overlapping
+generated/pandas.IntervalIndex.left,../reference/api/pandas.IntervalIndex.left
+generated/pandas.IntervalIndex.length,../reference/api/pandas.IntervalIndex.length
+generated/pandas.IntervalIndex.mid,../reference/api/pandas.IntervalIndex.mid
+generated/pandas.IntervalIndex.overlaps,../reference/api/pandas.IntervalIndex.overlaps
+generated/pandas.IntervalIndex.right,../reference/api/pandas.IntervalIndex.right
+generated/pandas.IntervalIndex.set_closed,../reference/api/pandas.IntervalIndex.set_closed
+generated/pandas.IntervalIndex.to_tuples,../reference/api/pandas.IntervalIndex.to_tuples
+generated/pandas.IntervalIndex.values,../reference/api/pandas.IntervalIndex.values
+generated/pandas.Interval.left,../reference/api/pandas.Interval.left
+generated/pandas.Interval.length,../reference/api/pandas.Interval.length
+generated/pandas.Interval.mid,../reference/api/pandas.Interval.mid
+generated/pandas.Interval.open_left,../reference/api/pandas.Interval.open_left
+generated/pandas.Interval.open_right,../reference/api/pandas.Interval.open_right
+generated/pandas.Interval.overlaps,../reference/api/pandas.Interval.overlaps
+generated/pandas.interval_range,../reference/api/pandas.interval_range
+generated/pandas.Interval.right,../reference/api/pandas.Interval.right
+generated/pandas.io.formats.style.Styler.apply,../reference/api/pandas.io.formats.style.Styler.apply
+generated/pandas.io.formats.style.Styler.background_gradient,../reference/api/pandas.io.formats.style.Styler.background_gradient
+generated/pandas.io.formats.style.Styler.bar,../reference/api/pandas.io.formats.style.Styler.bar
+generated/pandas.io.formats.style.Styler.clear,../reference/api/pandas.io.formats.style.Styler.clear
+generated/pandas.io.formats.style.Styler.env,../reference/api/pandas.io.formats.style.Styler.env
+generated/pandas.io.formats.style.Styler.export,../reference/api/pandas.io.formats.style.Styler.export
+generated/pandas.io.formats.style.Styler.format,../reference/api/pandas.io.formats.style.Styler.format
+generated/pandas.io.formats.style.Styler.from_custom_template,../reference/api/pandas.io.formats.style.Styler.from_custom_template
+generated/pandas.io.formats.style.Styler.hide_columns,../reference/api/pandas.io.formats.style.Styler.hide_columns
+generated/pandas.io.formats.style.Styler.hide_index,../reference/api/pandas.io.formats.style.Styler.hide_index
+generated/pandas.io.formats.style.Styler.highlight_max,../reference/api/pandas.io.formats.style.Styler.highlight_max
+generated/pandas.io.formats.style.Styler.highlight_min,../reference/api/pandas.io.formats.style.Styler.highlight_min
+generated/pandas.io.formats.style.Styler.highlight_null,../reference/api/pandas.io.formats.style.Styler.highlight_null
+generated/pandas.io.formats.style.Styler,../reference/api/pandas.io.formats.style.Styler
+generated/pandas.io.formats.style.Styler.loader,../reference/api/pandas.io.formats.style.Styler.loader
+generated/pandas.io.formats.style.Styler.pipe,../reference/api/pandas.io.formats.style.Styler.pipe
+generated/pandas.io.formats.style.Styler.render,../reference/api/pandas.io.formats.style.Styler.render
+generated/pandas.io.formats.style.Styler.set_caption,../reference/api/pandas.io.formats.style.Styler.set_caption
+generated/pandas.io.formats.style.Styler.set_precision,../reference/api/pandas.io.formats.style.Styler.set_precision
+generated/pandas.io.formats.style.Styler.set_properties,../reference/api/pandas.io.formats.style.Styler.set_properties
+generated/pandas.io.formats.style.Styler.set_table_attributes,../reference/api/pandas.io.formats.style.Styler.set_table_attributes
+generated/pandas.io.formats.style.Styler.set_table_styles,../reference/api/pandas.io.formats.style.Styler.set_table_styles
+generated/pandas.io.formats.style.Styler.set_uuid,../reference/api/pandas.io.formats.style.Styler.set_uuid
+generated/pandas.io.formats.style.Styler.template,../reference/api/pandas.io.formats.style.Styler.template
+generated/pandas.io.formats.style.Styler.to_excel,../reference/api/pandas.io.formats.style.Styler.to_excel
+generated/pandas.io.formats.style.Styler.use,../reference/api/pandas.io.formats.style.Styler.use
+generated/pandas.io.formats.style.Styler.where,../reference/api/pandas.io.formats.style.Styler.where
+generated/pandas.io.json.build_table_schema,../reference/api/pandas.io.json.build_table_schema
+generated/pandas.io.json.json_normalize,../reference/api/pandas.json_normalize
+generated/pandas.io.stata.StataReader.data_label,../reference/api/pandas.io.stata.StataReader.data_label
+generated/pandas.io.stata.StataReader.value_labels,../reference/api/pandas.io.stata.StataReader.value_labels
+generated/pandas.io.stata.StataReader.variable_labels,../reference/api/pandas.io.stata.StataReader.variable_labels
+generated/pandas.io.stata.StataWriter.write_file,../reference/api/pandas.io.stata.StataWriter.write_file
+generated/pandas.isna,../reference/api/pandas.isna
+generated/pandas.isnull,../reference/api/pandas.isnull
+generated/pandas.melt,../reference/api/pandas.melt
+generated/pandas.merge_asof,../reference/api/pandas.merge_asof
+generated/pandas.merge,../reference/api/pandas.merge
+generated/pandas.merge_ordered,../reference/api/pandas.merge_ordered
+generated/pandas.MultiIndex.codes,../reference/api/pandas.MultiIndex.codes
+generated/pandas.MultiIndex.droplevel,../reference/api/pandas.MultiIndex.droplevel
+generated/pandas.MultiIndex.from_arrays,../reference/api/pandas.MultiIndex.from_arrays
+generated/pandas.MultiIndex.from_frame,../reference/api/pandas.MultiIndex.from_frame
+generated/pandas.MultiIndex.from_product,../reference/api/pandas.MultiIndex.from_product
+generated/pandas.MultiIndex.from_tuples,../reference/api/pandas.MultiIndex.from_tuples
+generated/pandas.MultiIndex.get_indexer,../reference/api/pandas.MultiIndex.get_indexer
+generated/pandas.MultiIndex.get_level_values,../reference/api/pandas.MultiIndex.get_level_values
+generated/pandas.MultiIndex.get_loc,../reference/api/pandas.MultiIndex.get_loc
+generated/pandas.MultiIndex.get_loc_level,../reference/api/pandas.MultiIndex.get_loc_level
+generated/pandas.MultiIndex,../reference/api/pandas.MultiIndex
+generated/pandas.MultiIndex.is_lexsorted,../reference/api/pandas.MultiIndex.is_lexsorted
+generated/pandas.MultiIndex.levels,../reference/api/pandas.MultiIndex.levels
+generated/pandas.MultiIndex.levshape,../reference/api/pandas.MultiIndex.levshape
+generated/pandas.MultiIndex.names,../reference/api/pandas.MultiIndex.names
+generated/pandas.MultiIndex.nlevels,../reference/api/pandas.MultiIndex.nlevels
+generated/pandas.MultiIndex.remove_unused_levels,../reference/api/pandas.MultiIndex.remove_unused_levels
+generated/pandas.MultiIndex.reorder_levels,../reference/api/pandas.MultiIndex.reorder_levels
+generated/pandas.MultiIndex.set_codes,../reference/api/pandas.MultiIndex.set_codes
+generated/pandas.MultiIndex.set_levels,../reference/api/pandas.MultiIndex.set_levels
+generated/pandas.MultiIndex.sortlevel,../reference/api/pandas.MultiIndex.sortlevel
+generated/pandas.MultiIndex.swaplevel,../reference/api/pandas.MultiIndex.swaplevel
+generated/pandas.MultiIndex.to_flat_index,../reference/api/pandas.MultiIndex.to_flat_index
+generated/pandas.MultiIndex.to_frame,../reference/api/pandas.MultiIndex.to_frame
+generated/pandas.notna,../reference/api/pandas.notna
+generated/pandas.notnull,../reference/api/pandas.notnull
+generated/pandas.option_context,../reference/api/pandas.option_context
+generated/pandas.Period.asfreq,../reference/api/pandas.Period.asfreq
+generated/pandas.Period.day,../reference/api/pandas.Period.day
+generated/pandas.Period.dayofweek,../reference/api/pandas.Period.dayofweek
+generated/pandas.Period.day_of_week,../reference/api/pandas.Period.day_of_week
+generated/pandas.Period.dayofyear,../reference/api/pandas.Period.dayofyear
+generated/pandas.Period.day_of_year,../reference/api/pandas.Period.day_of_year
+generated/pandas.Period.days_in_month,../reference/api/pandas.Period.days_in_month
+generated/pandas.Period.daysinmonth,../reference/api/pandas.Period.daysinmonth
+generated/pandas.Period.end_time,../reference/api/pandas.Period.end_time
+generated/pandas.Period.freq,../reference/api/pandas.Period.freq
+generated/pandas.Period.freqstr,../reference/api/pandas.Period.freqstr
+generated/pandas.Period.hour,../reference/api/pandas.Period.hour
+generated/pandas.Period,../reference/api/pandas.Period
+generated/pandas.PeriodIndex.asfreq,../reference/api/pandas.PeriodIndex.asfreq
+generated/pandas.PeriodIndex.day,../reference/api/pandas.PeriodIndex.day
+generated/pandas.PeriodIndex.dayofweek,../reference/api/pandas.PeriodIndex.dayofweek
+generated/pandas.PeriodIndex.day_of_week,../reference/api/pandas.PeriodIndex.day_of_week
+generated/pandas.PeriodIndex.dayofyear,../reference/api/pandas.PeriodIndex.dayofyear
+generated/pandas.PeriodIndex.day_of_year,../reference/api/pandas.PeriodIndex.day_of_year
+generated/pandas.PeriodIndex.days_in_month,../reference/api/pandas.PeriodIndex.days_in_month
+generated/pandas.PeriodIndex.daysinmonth,../reference/api/pandas.PeriodIndex.daysinmonth
+generated/pandas.PeriodIndex.end_time,../reference/api/pandas.PeriodIndex.end_time
+generated/pandas.PeriodIndex.freq,../reference/api/pandas.PeriodIndex.freq
+generated/pandas.PeriodIndex.freqstr,../reference/api/pandas.PeriodIndex.freqstr
+generated/pandas.PeriodIndex.hour,../reference/api/pandas.PeriodIndex.hour
+generated/pandas.PeriodIndex,../reference/api/pandas.PeriodIndex
+generated/pandas.PeriodIndex.is_leap_year,../reference/api/pandas.PeriodIndex.is_leap_year
+generated/pandas.PeriodIndex.minute,../reference/api/pandas.PeriodIndex.minute
+generated/pandas.PeriodIndex.month,../reference/api/pandas.PeriodIndex.month
+generated/pandas.PeriodIndex.quarter,../reference/api/pandas.PeriodIndex.quarter
+generated/pandas.PeriodIndex.qyear,../reference/api/pandas.PeriodIndex.qyear
+generated/pandas.PeriodIndex.second,../reference/api/pandas.PeriodIndex.second
+generated/pandas.PeriodIndex.start_time,../reference/api/pandas.PeriodIndex.start_time
+generated/pandas.PeriodIndex.strftime,../reference/api/pandas.PeriodIndex.strftime
+generated/pandas.PeriodIndex.to_timestamp,../reference/api/pandas.PeriodIndex.to_timestamp
+generated/pandas.PeriodIndex.weekday,../reference/api/pandas.PeriodIndex.weekday
+generated/pandas.PeriodIndex.week,../reference/api/pandas.PeriodIndex.week
+generated/pandas.PeriodIndex.weekofyear,../reference/api/pandas.PeriodIndex.weekofyear
+generated/pandas.PeriodIndex.year,../reference/api/pandas.PeriodIndex.year
+generated/pandas.Period.is_leap_year,../reference/api/pandas.Period.is_leap_year
+generated/pandas.Period.minute,../reference/api/pandas.Period.minute
+generated/pandas.Period.month,../reference/api/pandas.Period.month
+generated/pandas.Period.now,../reference/api/pandas.Period.now
+generated/pandas.Period.ordinal,../reference/api/pandas.Period.ordinal
+generated/pandas.Period.quarter,../reference/api/pandas.Period.quarter
+generated/pandas.Period.qyear,../reference/api/pandas.Period.qyear
+generated/pandas.period_range,../reference/api/pandas.period_range
+generated/pandas.Period.second,../reference/api/pandas.Period.second
+generated/pandas.Period.start_time,../reference/api/pandas.Period.start_time
+generated/pandas.Period.strftime,../reference/api/pandas.Period.strftime
+generated/pandas.Period.to_timestamp,../reference/api/pandas.Period.to_timestamp
+generated/pandas.Period.weekday,../reference/api/pandas.Period.weekday
+generated/pandas.Period.week,../reference/api/pandas.Period.week
+generated/pandas.Period.weekofyear,../reference/api/pandas.Period.weekofyear
+generated/pandas.Period.year,../reference/api/pandas.Period.year
+generated/pandas.pivot,../reference/api/pandas.pivot
+generated/pandas.pivot_table,../reference/api/pandas.pivot_table
+generated/pandas.plotting.andrews_curves,../reference/api/pandas.plotting.andrews_curves
+generated/pandas.plotting.bootstrap_plot,../reference/api/pandas.plotting.bootstrap_plot
+generated/pandas.plotting.deregister_matplotlib_converters,../reference/api/pandas.plotting.deregister_matplotlib_converters
+generated/pandas.plotting.lag_plot,../reference/api/pandas.plotting.lag_plot
+generated/pandas.plotting.parallel_coordinates,../reference/api/pandas.plotting.parallel_coordinates
+generated/pandas.plotting.radviz,../reference/api/pandas.plotting.radviz
+generated/pandas.plotting.register_matplotlib_converters,../reference/api/pandas.plotting.register_matplotlib_converters
+generated/pandas.plotting.scatter_matrix,../reference/api/pandas.plotting.scatter_matrix
+generated/pandas.qcut,../reference/api/pandas.qcut
+generated/pandas.RangeIndex.from_range,../reference/api/pandas.RangeIndex.from_range
+generated/pandas.RangeIndex,../reference/api/pandas.RangeIndex
+generated/pandas.read_clipboard,../reference/api/pandas.read_clipboard
+generated/pandas.read_csv,../reference/api/pandas.read_csv
+generated/pandas.read_excel,../reference/api/pandas.read_excel
+generated/pandas.read_feather,../reference/api/pandas.read_feather
+generated/pandas.read_fwf,../reference/api/pandas.read_fwf
+generated/pandas.read_hdf,../reference/api/pandas.read_hdf
+generated/pandas.read,../reference/api/pandas.read
+generated/pandas.read_json,../reference/api/pandas.read_json
+generated/pandas.read_parquet,../reference/api/pandas.read_parquet
+generated/pandas.read_pickle,../reference/api/pandas.read_pickle
+generated/pandas.read_sas,../reference/api/pandas.read_sas
+generated/pandas.read_sql,../reference/api/pandas.read_sql
+generated/pandas.read_sql_query,../reference/api/pandas.read_sql_query
+generated/pandas.read_sql_table,../reference/api/pandas.read_sql_table
+generated/pandas.read_stata,../reference/api/pandas.read_stata
+generated/pandas.read_table,../reference/api/pandas.read_table
+generated/pandas.reset_option,../reference/api/pandas.reset_option
+generated/pandas.Series.abs,../reference/api/pandas.Series.abs
+generated/pandas.Series.add,../reference/api/pandas.Series.add
+generated/pandas.Series.add_prefix,../reference/api/pandas.Series.add_prefix
+generated/pandas.Series.add_suffix,../reference/api/pandas.Series.add_suffix
+generated/pandas.Series.agg,../reference/api/pandas.Series.agg
+generated/pandas.Series.aggregate,../reference/api/pandas.Series.aggregate
+generated/pandas.Series.align,../reference/api/pandas.Series.align
+generated/pandas.Series.all,../reference/api/pandas.Series.all
+generated/pandas.Series.any,../reference/api/pandas.Series.any
+generated/pandas.Series.apply,../reference/api/pandas.Series.apply
+generated/pandas.Series.argmax,../reference/api/pandas.Series.argmax
+generated/pandas.Series.argmin,../reference/api/pandas.Series.argmin
+generated/pandas.Series.argsort,../reference/api/pandas.Series.argsort
+generated/pandas.Series.__array__,../reference/api/pandas.Series.__array__
+generated/pandas.Series.array,../reference/api/pandas.Series.array
+generated/pandas.Series.as_blocks,../reference/api/pandas.Series.as_blocks
+generated/pandas.Series.asfreq,../reference/api/pandas.Series.asfreq
+generated/pandas.Series.as_matrix,../reference/api/pandas.Series.as_matrix
+generated/pandas.Series.asobject,../reference/api/pandas.Series.asobject
+generated/pandas.Series.asof,../reference/api/pandas.Series.asof
+generated/pandas.Series.astype,../reference/api/pandas.Series.astype
+generated/pandas.Series.at,../reference/api/pandas.Series.at
+generated/pandas.Series.at_time,../reference/api/pandas.Series.at_time
+generated/pandas.Series.autocorr,../reference/api/pandas.Series.autocorr
+generated/pandas.Series.axes,../reference/api/pandas.Series.axes
+generated/pandas.Series.between,../reference/api/pandas.Series.between
+generated/pandas.Series.between_time,../reference/api/pandas.Series.between_time
+generated/pandas.Series.bfill,../reference/api/pandas.Series.bfill
+generated/pandas.Series.blocks,../reference/api/pandas.Series.blocks
+generated/pandas.Series.cat.add_categories,../reference/api/pandas.Series.cat.add_categories
+generated/pandas.Series.cat.as_ordered,../reference/api/pandas.Series.cat.as_ordered
+generated/pandas.Series.cat.as_unordered,../reference/api/pandas.Series.cat.as_unordered
+generated/pandas.Series.cat.categories,../reference/api/pandas.Series.cat.categories
+generated/pandas.Series.cat.codes,../reference/api/pandas.Series.cat.codes
+generated/pandas.Series.cat,../reference/api/pandas.Series.cat
+generated/pandas.Series.cat.ordered,../reference/api/pandas.Series.cat.ordered
+generated/pandas.Series.cat.remove_categories,../reference/api/pandas.Series.cat.remove_categories
+generated/pandas.Series.cat.remove_unused_categories,../reference/api/pandas.Series.cat.remove_unused_categories
+generated/pandas.Series.cat.rename_categories,../reference/api/pandas.Series.cat.rename_categories
+generated/pandas.Series.cat.reorder_categories,../reference/api/pandas.Series.cat.reorder_categories
+generated/pandas.Series.cat.set_categories,../reference/api/pandas.Series.cat.set_categories
+generated/pandas.Series.clip,../reference/api/pandas.Series.clip
+generated/pandas.Series.clip_lower,../reference/api/pandas.Series.clip_lower
+generated/pandas.Series.clip_upper,../reference/api/pandas.Series.clip_upper
+generated/pandas.Series.combine_first,../reference/api/pandas.Series.combine_first
+generated/pandas.Series.combine,../reference/api/pandas.Series.combine
+generated/pandas.Series.compress,../reference/api/pandas.Series.compress
+generated/pandas.Series.convert_objects,../reference/api/pandas.Series.convert_objects
+generated/pandas.Series.copy,../reference/api/pandas.Series.copy
+generated/pandas.Series.corr,../reference/api/pandas.Series.corr
+generated/pandas.Series.count,../reference/api/pandas.Series.count
+generated/pandas.Series.cov,../reference/api/pandas.Series.cov
+generated/pandas.Series.cummax,../reference/api/pandas.Series.cummax
+generated/pandas.Series.cummin,../reference/api/pandas.Series.cummin
+generated/pandas.Series.cumprod,../reference/api/pandas.Series.cumprod
+generated/pandas.Series.cumsum,../reference/api/pandas.Series.cumsum
+generated/pandas.Series.data,../reference/api/pandas.Series.data
+generated/pandas.Series.describe,../reference/api/pandas.Series.describe
+generated/pandas.Series.diff,../reference/api/pandas.Series.diff
+generated/pandas.Series.div,../reference/api/pandas.Series.div
+generated/pandas.Series.divide,../reference/api/pandas.Series.divide
+generated/pandas.Series.divmod,../reference/api/pandas.Series.divmod
+generated/pandas.Series.dot,../reference/api/pandas.Series.dot
+generated/pandas.Series.drop_duplicates,../reference/api/pandas.Series.drop_duplicates
+generated/pandas.Series.drop,../reference/api/pandas.Series.drop
+generated/pandas.Series.droplevel,../reference/api/pandas.Series.droplevel
+generated/pandas.Series.dropna,../reference/api/pandas.Series.dropna
+generated/pandas.Series.dt.ceil,../reference/api/pandas.Series.dt.ceil
+generated/pandas.Series.dt.components,../reference/api/pandas.Series.dt.components
+generated/pandas.Series.dt.date,../reference/api/pandas.Series.dt.date
+generated/pandas.Series.dt.day,../reference/api/pandas.Series.dt.day
+generated/pandas.Series.dt.day_name,../reference/api/pandas.Series.dt.day_name
+generated/pandas.Series.dt.dayofweek,../reference/api/pandas.Series.dt.dayofweek
+generated/pandas.Series.dt.day_of_week,../reference/api/pandas.Series.dt.day_of_week
+generated/pandas.Series.dt.dayofyear,../reference/api/pandas.Series.dt.dayofyear
+generated/pandas.Series.dt.day_of_year,../reference/api/pandas.Series.dt.day_of_year
+generated/pandas.Series.dt.days,../reference/api/pandas.Series.dt.days
+generated/pandas.Series.dt.days_in_month,../reference/api/pandas.Series.dt.days_in_month
+generated/pandas.Series.dt.daysinmonth,../reference/api/pandas.Series.dt.daysinmonth
+generated/pandas.Series.dt.end_time,../reference/api/pandas.Series.dt.end_time
+generated/pandas.Series.dt.floor,../reference/api/pandas.Series.dt.floor
+generated/pandas.Series.dt.freq,../reference/api/pandas.Series.dt.freq
+generated/pandas.Series.dt.hour,../reference/api/pandas.Series.dt.hour
+generated/pandas.Series.dt,../reference/api/pandas.Series.dt
+generated/pandas.Series.dt.is_leap_year,../reference/api/pandas.Series.dt.is_leap_year
+generated/pandas.Series.dt.is_month_end,../reference/api/pandas.Series.dt.is_month_end
+generated/pandas.Series.dt.is_month_start,../reference/api/pandas.Series.dt.is_month_start
+generated/pandas.Series.dt.is_quarter_end,../reference/api/pandas.Series.dt.is_quarter_end
+generated/pandas.Series.dt.is_quarter_start,../reference/api/pandas.Series.dt.is_quarter_start
+generated/pandas.Series.dt.is_year_end,../reference/api/pandas.Series.dt.is_year_end
+generated/pandas.Series.dt.is_year_start,../reference/api/pandas.Series.dt.is_year_start
+generated/pandas.Series.dt.microsecond,../reference/api/pandas.Series.dt.microsecond
+generated/pandas.Series.dt.microseconds,../reference/api/pandas.Series.dt.microseconds
+generated/pandas.Series.dt.minute,../reference/api/pandas.Series.dt.minute
+generated/pandas.Series.dt.month,../reference/api/pandas.Series.dt.month
+generated/pandas.Series.dt.month_name,../reference/api/pandas.Series.dt.month_name
+generated/pandas.Series.dt.nanosecond,../reference/api/pandas.Series.dt.nanosecond
+generated/pandas.Series.dt.nanoseconds,../reference/api/pandas.Series.dt.nanoseconds
+generated/pandas.Series.dt.normalize,../reference/api/pandas.Series.dt.normalize
+generated/pandas.Series.dt.quarter,../reference/api/pandas.Series.dt.quarter
+generated/pandas.Series.dt.qyear,../reference/api/pandas.Series.dt.qyear
+generated/pandas.Series.dt.round,../reference/api/pandas.Series.dt.round
+generated/pandas.Series.dt.second,../reference/api/pandas.Series.dt.second
+generated/pandas.Series.dt.seconds,../reference/api/pandas.Series.dt.seconds
+generated/pandas.Series.dt.start_time,../reference/api/pandas.Series.dt.start_time
+generated/pandas.Series.dt.strftime,../reference/api/pandas.Series.dt.strftime
+generated/pandas.Series.dt.time,../reference/api/pandas.Series.dt.time
+generated/pandas.Series.dt.timetz,../reference/api/pandas.Series.dt.timetz
+generated/pandas.Series.dt.to_period,../reference/api/pandas.Series.dt.to_period
+generated/pandas.Series.dt.to_pydatetime,../reference/api/pandas.Series.dt.to_pydatetime
+generated/pandas.Series.dt.to_pytimedelta,../reference/api/pandas.Series.dt.to_pytimedelta
+generated/pandas.Series.dt.total_seconds,../reference/api/pandas.Series.dt.total_seconds
+generated/pandas.Series.dt.tz_convert,../reference/api/pandas.Series.dt.tz_convert
+generated/pandas.Series.dt.tz,../reference/api/pandas.Series.dt.tz
+generated/pandas.Series.dt.tz_localize,../reference/api/pandas.Series.dt.tz_localize
+generated/pandas.Series.dt.weekday,../reference/api/pandas.Series.dt.weekday
+generated/pandas.Series.dt.week,../reference/api/pandas.Series.dt.week
+generated/pandas.Series.dt.weekofyear,../reference/api/pandas.Series.dt.weekofyear
+generated/pandas.Series.dt.year,../reference/api/pandas.Series.dt.year
+generated/pandas.Series.dtype,../reference/api/pandas.Series.dtype
+generated/pandas.Series.dtypes,../reference/api/pandas.Series.dtypes
+generated/pandas.Series.duplicated,../reference/api/pandas.Series.duplicated
+generated/pandas.Series.empty,../reference/api/pandas.Series.empty
+generated/pandas.Series.eq,../reference/api/pandas.Series.eq
+generated/pandas.Series.equals,../reference/api/pandas.Series.equals
+generated/pandas.Series.ewm,../reference/api/pandas.Series.ewm
+generated/pandas.Series.expanding,../reference/api/pandas.Series.expanding
+generated/pandas.Series.factorize,../reference/api/pandas.Series.factorize
+generated/pandas.Series.ffill,../reference/api/pandas.Series.ffill
+generated/pandas.Series.fillna,../reference/api/pandas.Series.fillna
+generated/pandas.Series.filter,../reference/api/pandas.Series.filter
+generated/pandas.Series.first,../reference/api/pandas.Series.first
+generated/pandas.Series.first_valid_index,../reference/api/pandas.Series.first_valid_index
+generated/pandas.Series.floordiv,../reference/api/pandas.Series.floordiv
+generated/pandas.Series.from_array,../reference/api/pandas.Series.from_array
+generated/pandas.Series.from_csv,../reference/api/pandas.Series.from_csv
+generated/pandas.Series.ge,../reference/api/pandas.Series.ge
+generated/pandas.Series.get,../reference/api/pandas.Series.get
+generated/pandas.Series.get_value,../reference/api/pandas.Series.get_value
+generated/pandas.Series.groupby,../reference/api/pandas.Series.groupby
+generated/pandas.Series.gt,../reference/api/pandas.Series.gt
+generated/pandas.Series.hasnans,../reference/api/pandas.Series.hasnans
+generated/pandas.Series.head,../reference/api/pandas.Series.head
+generated/pandas.Series.hist,../reference/api/pandas.Series.hist
+generated/pandas.Series,../reference/api/pandas.Series
+generated/pandas.Series.iat,../reference/api/pandas.Series.iat
+generated/pandas.Series.idxmax,../reference/api/pandas.Series.idxmax
+generated/pandas.Series.idxmin,../reference/api/pandas.Series.idxmin
+generated/pandas.Series.iloc,../reference/api/pandas.Series.iloc
+generated/pandas.Series.imag,../reference/api/pandas.Series.imag
+generated/pandas.Series.index,../reference/api/pandas.Series.index
+generated/pandas.Series.infer_objects,../reference/api/pandas.Series.infer_objects
+generated/pandas.Series.interpolate,../reference/api/pandas.Series.interpolate
+generated/pandas.Series.is_copy,../reference/api/pandas.Series.is_copy
+generated/pandas.Series.isin,../reference/api/pandas.Series.isin
+generated/pandas.Series.is_monotonic_decreasing,../reference/api/pandas.Series.is_monotonic_decreasing
+generated/pandas.Series.is_monotonic_increasing,../reference/api/pandas.Series.is_monotonic_increasing
+generated/pandas.Series.isna,../reference/api/pandas.Series.isna
+generated/pandas.Series.isnull,../reference/api/pandas.Series.isnull
+generated/pandas.Series.is_unique,../reference/api/pandas.Series.is_unique
+generated/pandas.Series.item,../reference/api/pandas.Series.item
+generated/pandas.Series.items,../reference/api/pandas.Series.items
+generated/pandas.Series.__iter__,../reference/api/pandas.Series.__iter__
+generated/pandas.Series.keys,../reference/api/pandas.Series.keys
+generated/pandas.Series.kurt,../reference/api/pandas.Series.kurt
+generated/pandas.Series.kurtosis,../reference/api/pandas.Series.kurtosis
+generated/pandas.Series.last,../reference/api/pandas.Series.last
+generated/pandas.Series.last_valid_index,../reference/api/pandas.Series.last_valid_index
+generated/pandas.Series.le,../reference/api/pandas.Series.le
+generated/pandas.Series.loc,../reference/api/pandas.Series.loc
+generated/pandas.Series.lt,../reference/api/pandas.Series.lt
+generated/pandas.Series.map,../reference/api/pandas.Series.map
+generated/pandas.Series.mask,../reference/api/pandas.Series.mask
+generated/pandas.Series.max,../reference/api/pandas.Series.max
+generated/pandas.Series.mean,../reference/api/pandas.Series.mean
+generated/pandas.Series.median,../reference/api/pandas.Series.median
+generated/pandas.Series.memory_usage,../reference/api/pandas.Series.memory_usage
+generated/pandas.Series.min,../reference/api/pandas.Series.min
+generated/pandas.Series.mode,../reference/api/pandas.Series.mode
+generated/pandas.Series.mod,../reference/api/pandas.Series.mod
+generated/pandas.Series.mul,../reference/api/pandas.Series.mul
+generated/pandas.Series.multiply,../reference/api/pandas.Series.multiply
+generated/pandas.Series.name,../reference/api/pandas.Series.name
+generated/pandas.Series.nbytes,../reference/api/pandas.Series.nbytes
+generated/pandas.Series.ndim,../reference/api/pandas.Series.ndim
+generated/pandas.Series.ne,../reference/api/pandas.Series.ne
+generated/pandas.Series.nlargest,../reference/api/pandas.Series.nlargest
+generated/pandas.Series.nonzero,../reference/api/pandas.Series.nonzero
+generated/pandas.Series.notna,../reference/api/pandas.Series.notna
+generated/pandas.Series.notnull,../reference/api/pandas.Series.notnull
+generated/pandas.Series.nsmallest,../reference/api/pandas.Series.nsmallest
+generated/pandas.Series.nunique,../reference/api/pandas.Series.nunique
+generated/pandas.Series.pct_change,../reference/api/pandas.Series.pct_change
+generated/pandas.Series.pipe,../reference/api/pandas.Series.pipe
+generated/pandas.Series.plot.area,../reference/api/pandas.Series.plot.area
+generated/pandas.Series.plot.barh,../reference/api/pandas.Series.plot.barh
+generated/pandas.Series.plot.bar,../reference/api/pandas.Series.plot.bar
+generated/pandas.Series.plot.box,../reference/api/pandas.Series.plot.box
+generated/pandas.Series.plot.density,../reference/api/pandas.Series.plot.density
+generated/pandas.Series.plot.hist,../reference/api/pandas.Series.plot.hist
+generated/pandas.Series.plot,../reference/api/pandas.Series.plot
+generated/pandas.Series.plot.kde,../reference/api/pandas.Series.plot.kde
+generated/pandas.Series.plot.line,../reference/api/pandas.Series.plot.line
+generated/pandas.Series.plot.pie,../reference/api/pandas.Series.plot.pie
+generated/pandas.Series.pop,../reference/api/pandas.Series.pop
+generated/pandas.Series.pow,../reference/api/pandas.Series.pow
+generated/pandas.Series.prod,../reference/api/pandas.Series.prod
+generated/pandas.Series.product,../reference/api/pandas.Series.product
+generated/pandas.Series.ptp,../reference/api/pandas.Series.ptp
+generated/pandas.Series.quantile,../reference/api/pandas.Series.quantile
+generated/pandas.Series.radd,../reference/api/pandas.Series.radd
+generated/pandas.Series.rank,../reference/api/pandas.Series.rank
+generated/pandas.Series.rdiv,../reference/api/pandas.Series.rdiv
+generated/pandas.Series.rdivmod,../reference/api/pandas.Series.rdivmod
+generated/pandas.Series.real,../reference/api/pandas.Series.real
+generated/pandas.Series.reindex_axis,../reference/api/pandas.Series.reindex_axis
+generated/pandas.Series.reindex,../reference/api/pandas.Series.reindex
+generated/pandas.Series.reindex_like,../reference/api/pandas.Series.reindex_like
+generated/pandas.Series.rename_axis,../reference/api/pandas.Series.rename_axis
+generated/pandas.Series.rename,../reference/api/pandas.Series.rename
+generated/pandas.Series.reorder_levels,../reference/api/pandas.Series.reorder_levels
+generated/pandas.Series.repeat,../reference/api/pandas.Series.repeat
+generated/pandas.Series.replace,../reference/api/pandas.Series.replace
+generated/pandas.Series.resample,../reference/api/pandas.Series.resample
+generated/pandas.Series.reset_index,../reference/api/pandas.Series.reset_index
+generated/pandas.Series.rfloordiv,../reference/api/pandas.Series.rfloordiv
+generated/pandas.Series.rmod,../reference/api/pandas.Series.rmod
+generated/pandas.Series.rmul,../reference/api/pandas.Series.rmul
+generated/pandas.Series.rolling,../reference/api/pandas.Series.rolling
+generated/pandas.Series.round,../reference/api/pandas.Series.round
+generated/pandas.Series.rpow,../reference/api/pandas.Series.rpow
+generated/pandas.Series.rsub,../reference/api/pandas.Series.rsub
+generated/pandas.Series.rtruediv,../reference/api/pandas.Series.rtruediv
+generated/pandas.Series.sample,../reference/api/pandas.Series.sample
+generated/pandas.Series.searchsorted,../reference/api/pandas.Series.searchsorted
+generated/pandas.Series.select,../reference/api/pandas.Series.select
+generated/pandas.Series.sem,../reference/api/pandas.Series.sem
+generated/pandas.Series.set_axis,../reference/api/pandas.Series.set_axis
+generated/pandas.Series.set_value,../reference/api/pandas.Series.set_value
+generated/pandas.Series.shape,../reference/api/pandas.Series.shape
+generated/pandas.Series.shift,../reference/api/pandas.Series.shift
+generated/pandas.Series.size,../reference/api/pandas.Series.size
+generated/pandas.Series.skew,../reference/api/pandas.Series.skew
+generated/pandas.Series.sort_index,../reference/api/pandas.Series.sort_index
+generated/pandas.Series.sort_values,../reference/api/pandas.Series.sort_values
+generated/pandas.Series.sparse.density,../reference/api/pandas.Series.sparse.density
+generated/pandas.Series.sparse.fill_value,../reference/api/pandas.Series.sparse.fill_value
+generated/pandas.Series.sparse.from_coo,../reference/api/pandas.Series.sparse.from_coo
+generated/pandas.Series.sparse.npoints,../reference/api/pandas.Series.sparse.npoints
+generated/pandas.Series.sparse.sp_values,../reference/api/pandas.Series.sparse.sp_values
+generated/pandas.Series.sparse.to_coo,../reference/api/pandas.Series.sparse.to_coo
+generated/pandas.Series.squeeze,../reference/api/pandas.Series.squeeze
+generated/pandas.Series.std,../reference/api/pandas.Series.std
+generated/pandas.Series.str.capitalize,../reference/api/pandas.Series.str.capitalize
+generated/pandas.Series.str.cat,../reference/api/pandas.Series.str.cat
+generated/pandas.Series.str.center,../reference/api/pandas.Series.str.center
+generated/pandas.Series.str.contains,../reference/api/pandas.Series.str.contains
+generated/pandas.Series.str.count,../reference/api/pandas.Series.str.count
+generated/pandas.Series.str.decode,../reference/api/pandas.Series.str.decode
+generated/pandas.Series.str.encode,../reference/api/pandas.Series.str.encode
+generated/pandas.Series.str.endswith,../reference/api/pandas.Series.str.endswith
+generated/pandas.Series.str.extractall,../reference/api/pandas.Series.str.extractall
+generated/pandas.Series.str.extract,../reference/api/pandas.Series.str.extract
+generated/pandas.Series.str.findall,../reference/api/pandas.Series.str.findall
+generated/pandas.Series.str.find,../reference/api/pandas.Series.str.find
+generated/pandas.Series.str.fullmatch,../reference/api/pandas.Series.str.fullmatch
+generated/pandas.Series.str.get_dummies,../reference/api/pandas.Series.str.get_dummies
+generated/pandas.Series.str.get,../reference/api/pandas.Series.str.get
+generated/pandas.Series.str,../reference/api/pandas.Series.str
+generated/pandas.Series.str.index,../reference/api/pandas.Series.str.index
+generated/pandas.Series.str.isalnum,../reference/api/pandas.Series.str.isalnum
+generated/pandas.Series.str.isalpha,../reference/api/pandas.Series.str.isalpha
+generated/pandas.Series.str.isdecimal,../reference/api/pandas.Series.str.isdecimal
+generated/pandas.Series.str.isdigit,../reference/api/pandas.Series.str.isdigit
+generated/pandas.Series.str.islower,../reference/api/pandas.Series.str.islower
+generated/pandas.Series.str.isnumeric,../reference/api/pandas.Series.str.isnumeric
+generated/pandas.Series.str.isspace,../reference/api/pandas.Series.str.isspace
+generated/pandas.Series.str.istitle,../reference/api/pandas.Series.str.istitle
+generated/pandas.Series.str.isupper,../reference/api/pandas.Series.str.isupper
+generated/pandas.Series.str.join,../reference/api/pandas.Series.str.join
+generated/pandas.Series.str.len,../reference/api/pandas.Series.str.len
+generated/pandas.Series.str.ljust,../reference/api/pandas.Series.str.ljust
+generated/pandas.Series.str.lower,../reference/api/pandas.Series.str.lower
+generated/pandas.Series.str.lstrip,../reference/api/pandas.Series.str.lstrip
+generated/pandas.Series.str.match,../reference/api/pandas.Series.str.match
+generated/pandas.Series.str.normalize,../reference/api/pandas.Series.str.normalize
+generated/pandas.Series.str.pad,../reference/api/pandas.Series.str.pad
+generated/pandas.Series.str.partition,../reference/api/pandas.Series.str.partition
+generated/pandas.Series.str.repeat,../reference/api/pandas.Series.str.repeat
+generated/pandas.Series.str.replace,../reference/api/pandas.Series.str.replace
+generated/pandas.Series.str.rfind,../reference/api/pandas.Series.str.rfind
+generated/pandas.Series.str.rindex,../reference/api/pandas.Series.str.rindex
+generated/pandas.Series.str.rjust,../reference/api/pandas.Series.str.rjust
+generated/pandas.Series.str.rpartition,../reference/api/pandas.Series.str.rpartition
+generated/pandas.Series.str.rsplit,../reference/api/pandas.Series.str.rsplit
+generated/pandas.Series.str.rstrip,../reference/api/pandas.Series.str.rstrip
+generated/pandas.Series.str.slice,../reference/api/pandas.Series.str.slice
+generated/pandas.Series.str.slice_replace,../reference/api/pandas.Series.str.slice_replace
+generated/pandas.Series.str.split,../reference/api/pandas.Series.str.split
+generated/pandas.Series.str.startswith,../reference/api/pandas.Series.str.startswith
+generated/pandas.Series.str.strip,../reference/api/pandas.Series.str.strip
+generated/pandas.Series.str.swapcase,../reference/api/pandas.Series.str.swapcase
+generated/pandas.Series.str.title,../reference/api/pandas.Series.str.title
+generated/pandas.Series.str.translate,../reference/api/pandas.Series.str.translate
+generated/pandas.Series.str.upper,../reference/api/pandas.Series.str.upper
+generated/pandas.Series.str.wrap,../reference/api/pandas.Series.str.wrap
+generated/pandas.Series.str.zfill,../reference/api/pandas.Series.str.zfill
+generated/pandas.Series.sub,../reference/api/pandas.Series.sub
+generated/pandas.Series.subtract,../reference/api/pandas.Series.subtract
+generated/pandas.Series.sum,../reference/api/pandas.Series.sum
+generated/pandas.Series.swaplevel,../reference/api/pandas.Series.swaplevel
+generated/pandas.Series.tail,../reference/api/pandas.Series.tail
+generated/pandas.Series.take,../reference/api/pandas.Series.take
+generated/pandas.Series.T,../reference/api/pandas.Series.T
+generated/pandas.Series.timetuple,../reference/api/pandas.Series.timetuple
+generated/pandas.Series.to_clipboard,../reference/api/pandas.Series.to_clipboard
+generated/pandas.Series.to_csv,../reference/api/pandas.Series.to_csv
+generated/pandas.Series.to_dict,../reference/api/pandas.Series.to_dict
+generated/pandas.Series.to_excel,../reference/api/pandas.Series.to_excel
+generated/pandas.Series.to_frame,../reference/api/pandas.Series.to_frame
+generated/pandas.Series.to_hdf,../reference/api/pandas.Series.to_hdf
+generated/pandas.Series.to_json,../reference/api/pandas.Series.to_json
+generated/pandas.Series.to_latex,../reference/api/pandas.Series.to_latex
+generated/pandas.Series.to_list,../reference/api/pandas.Series.to_list
+generated/pandas.Series.tolist,../reference/api/pandas.Series.tolist
+generated/pandas.Series.to_numpy,../reference/api/pandas.Series.to_numpy
+generated/pandas.Series.to_period,../reference/api/pandas.Series.to_period
+generated/pandas.Series.to_pickle,../reference/api/pandas.Series.to_pickle
+generated/pandas.Series.to_sql,../reference/api/pandas.Series.to_sql
+generated/pandas.Series.to_string,../reference/api/pandas.Series.to_string
+generated/pandas.Series.to_timestamp,../reference/api/pandas.Series.to_timestamp
+generated/pandas.Series.to_xarray,../reference/api/pandas.Series.to_xarray
+generated/pandas.Series.transform,../reference/api/pandas.Series.transform
+generated/pandas.Series.transpose,../reference/api/pandas.Series.transpose
+generated/pandas.Series.truediv,../reference/api/pandas.Series.truediv
+generated/pandas.Series.truncate,../reference/api/pandas.Series.truncate
+generated/pandas.Series.tz_convert,../reference/api/pandas.Series.tz_convert
+generated/pandas.Series.tz_localize,../reference/api/pandas.Series.tz_localize
+generated/pandas.Series.unique,../reference/api/pandas.Series.unique
+generated/pandas.Series.unstack,../reference/api/pandas.Series.unstack
+generated/pandas.Series.update,../reference/api/pandas.Series.update
+generated/pandas.Series.valid,../reference/api/pandas.Series.valid
+generated/pandas.Series.value_counts,../reference/api/pandas.Series.value_counts
+generated/pandas.Series.values,../reference/api/pandas.Series.values
+generated/pandas.Series.var,../reference/api/pandas.Series.var
+generated/pandas.Series.where,../reference/api/pandas.Series.where
+generated/pandas.Series.xs,../reference/api/pandas.Series.xs
+generated/pandas.set_option,../reference/api/pandas.set_option
+generated/pandas.SparseDataFrame.to_coo,../reference/api/pandas.SparseDataFrame.to_coo
+generated/pandas.SparseSeries.from_coo,../reference/api/pandas.SparseSeries.from_coo
+generated/pandas.SparseSeries.to_coo,../reference/api/pandas.SparseSeries.to_coo
+generated/pandas.test,../reference/api/pandas.test
+generated/pandas.testing.assert_frame_equal,../reference/api/pandas.testing.assert_frame_equal
+generated/pandas.testing.assert_index_equal,../reference/api/pandas.testing.assert_index_equal
+generated/pandas.testing.assert_series_equal,../reference/api/pandas.testing.assert_series_equal
+generated/pandas.Timedelta.asm8,../reference/api/pandas.Timedelta.asm8
+generated/pandas.Timedelta.ceil,../reference/api/pandas.Timedelta.ceil
+generated/pandas.Timedelta.components,../reference/api/pandas.Timedelta.components
+generated/pandas.Timedelta.days,../reference/api/pandas.Timedelta.days
+generated/pandas.Timedelta.delta,../reference/api/pandas.Timedelta.delta
+generated/pandas.Timedelta.floor,../reference/api/pandas.Timedelta.floor
+generated/pandas.Timedelta.freq,../reference/api/pandas.Timedelta.freq
+generated/pandas.Timedelta,../reference/api/pandas.Timedelta
+generated/pandas.TimedeltaIndex.ceil,../reference/api/pandas.TimedeltaIndex.ceil
+generated/pandas.TimedeltaIndex.components,../reference/api/pandas.TimedeltaIndex.components
+generated/pandas.TimedeltaIndex.days,../reference/api/pandas.TimedeltaIndex.days
+generated/pandas.TimedeltaIndex.floor,../reference/api/pandas.TimedeltaIndex.floor
+generated/pandas.TimedeltaIndex,../reference/api/pandas.TimedeltaIndex
+generated/pandas.TimedeltaIndex.inferred_freq,../reference/api/pandas.TimedeltaIndex.inferred_freq
+generated/pandas.TimedeltaIndex.microseconds,../reference/api/pandas.TimedeltaIndex.microseconds
+generated/pandas.TimedeltaIndex.nanoseconds,../reference/api/pandas.TimedeltaIndex.nanoseconds
+generated/pandas.TimedeltaIndex.round,../reference/api/pandas.TimedeltaIndex.round
+generated/pandas.TimedeltaIndex.seconds,../reference/api/pandas.TimedeltaIndex.seconds
+generated/pandas.TimedeltaIndex.to_frame,../reference/api/pandas.TimedeltaIndex.to_frame
+generated/pandas.TimedeltaIndex.to_pytimedelta,../reference/api/pandas.TimedeltaIndex.to_pytimedelta
+generated/pandas.TimedeltaIndex.to_series,../reference/api/pandas.TimedeltaIndex.to_series
+generated/pandas.Timedelta.isoformat,../reference/api/pandas.Timedelta.isoformat
+generated/pandas.Timedelta.is_populated,../reference/api/pandas.Timedelta.is_populated
+generated/pandas.Timedelta.max,../reference/api/pandas.Timedelta.max
+generated/pandas.Timedelta.microseconds,../reference/api/pandas.Timedelta.microseconds
+generated/pandas.Timedelta.min,../reference/api/pandas.Timedelta.min
+generated/pandas.Timedelta.nanoseconds,../reference/api/pandas.Timedelta.nanoseconds
+generated/pandas.timedelta_range,../reference/api/pandas.timedelta_range
+generated/pandas.Timedelta.resolution,../reference/api/pandas.Timedelta.resolution
+generated/pandas.Timedelta.round,../reference/api/pandas.Timedelta.round
+generated/pandas.Timedelta.seconds,../reference/api/pandas.Timedelta.seconds
+generated/pandas.Timedelta.to_pytimedelta,../reference/api/pandas.Timedelta.to_pytimedelta
+generated/pandas.Timedelta.total_seconds,../reference/api/pandas.Timedelta.total_seconds
+generated/pandas.Timedelta.to_timedelta64,../reference/api/pandas.Timedelta.to_timedelta64
+generated/pandas.Timedelta.value,../reference/api/pandas.Timedelta.value
+generated/pandas.Timedelta.view,../reference/api/pandas.Timedelta.view
+generated/pandas.Timestamp.asm8,../reference/api/pandas.Timestamp.asm8
+generated/pandas.Timestamp.astimezone,../reference/api/pandas.Timestamp.astimezone
+generated/pandas.Timestamp.ceil,../reference/api/pandas.Timestamp.ceil
+generated/pandas.Timestamp.combine,../reference/api/pandas.Timestamp.combine
+generated/pandas.Timestamp.ctime,../reference/api/pandas.Timestamp.ctime
+generated/pandas.Timestamp.date,../reference/api/pandas.Timestamp.date
+generated/pandas.Timestamp.day,../reference/api/pandas.Timestamp.day
+generated/pandas.Timestamp.day_name,../reference/api/pandas.Timestamp.day_name
+generated/pandas.Timestamp.dayofweek,../reference/api/pandas.Timestamp.dayofweek
+generated/pandas.Timestamp.day_of_week,../reference/api/pandas.Timestamp.day_of_week
+generated/pandas.Timestamp.dayofyear,../reference/api/pandas.Timestamp.dayofyear
+generated/pandas.Timestamp.day_of_year,../reference/api/pandas.Timestamp.day_of_year
+generated/pandas.Timestamp.days_in_month,../reference/api/pandas.Timestamp.days_in_month
+generated/pandas.Timestamp.daysinmonth,../reference/api/pandas.Timestamp.daysinmonth
+generated/pandas.Timestamp.dst,../reference/api/pandas.Timestamp.dst
+generated/pandas.Timestamp.floor,../reference/api/pandas.Timestamp.floor
+generated/pandas.Timestamp.fold,../reference/api/pandas.Timestamp.fold
+generated/pandas.Timestamp.fromisoformat,../reference/api/pandas.Timestamp.fromisoformat
+generated/pandas.Timestamp.fromordinal,../reference/api/pandas.Timestamp.fromordinal
+generated/pandas.Timestamp.fromtimestamp,../reference/api/pandas.Timestamp.fromtimestamp
+generated/pandas.Timestamp.hour,../reference/api/pandas.Timestamp.hour
+generated/pandas.Timestamp,../reference/api/pandas.Timestamp
+generated/pandas.Timestamp.is_leap_year,../reference/api/pandas.Timestamp.is_leap_year
+generated/pandas.Timestamp.is_month_end,../reference/api/pandas.Timestamp.is_month_end
+generated/pandas.Timestamp.is_month_start,../reference/api/pandas.Timestamp.is_month_start
+generated/pandas.Timestamp.isocalendar,../reference/api/pandas.Timestamp.isocalendar
+generated/pandas.Timestamp.isoformat,../reference/api/pandas.Timestamp.isoformat
+generated/pandas.Timestamp.isoweekday,../reference/api/pandas.Timestamp.isoweekday
+generated/pandas.Timestamp.is_quarter_end,../reference/api/pandas.Timestamp.is_quarter_end
+generated/pandas.Timestamp.is_quarter_start,../reference/api/pandas.Timestamp.is_quarter_start
+generated/pandas.Timestamp.is_year_end,../reference/api/pandas.Timestamp.is_year_end
+generated/pandas.Timestamp.is_year_start,../reference/api/pandas.Timestamp.is_year_start
+generated/pandas.Timestamp.max,../reference/api/pandas.Timestamp.max
+generated/pandas.Timestamp.microsecond,../reference/api/pandas.Timestamp.microsecond
+generated/pandas.Timestamp.min,../reference/api/pandas.Timestamp.min
+generated/pandas.Timestamp.minute,../reference/api/pandas.Timestamp.minute
+generated/pandas.Timestamp.month,../reference/api/pandas.Timestamp.month
+generated/pandas.Timestamp.month_name,../reference/api/pandas.Timestamp.month_name
+generated/pandas.Timestamp.nanosecond,../reference/api/pandas.Timestamp.nanosecond
+generated/pandas.Timestamp.normalize,../reference/api/pandas.Timestamp.normalize
+generated/pandas.Timestamp.now,../reference/api/pandas.Timestamp.now
+generated/pandas.Timestamp.quarter,../reference/api/pandas.Timestamp.quarter
+generated/pandas.Timestamp.replace,../reference/api/pandas.Timestamp.replace
+generated/pandas.Timestamp.resolution,../reference/api/pandas.Timestamp.resolution
+generated/pandas.Timestamp.round,../reference/api/pandas.Timestamp.round
+generated/pandas.Timestamp.second,../reference/api/pandas.Timestamp.second
+generated/pandas.Timestamp.strftime,../reference/api/pandas.Timestamp.strftime
+generated/pandas.Timestamp.strptime,../reference/api/pandas.Timestamp.strptime
+generated/pandas.Timestamp.time,../reference/api/pandas.Timestamp.time
+generated/pandas.Timestamp.timestamp,../reference/api/pandas.Timestamp.timestamp
+generated/pandas.Timestamp.timetuple,../reference/api/pandas.Timestamp.timetuple
+generated/pandas.Timestamp.timetz,../reference/api/pandas.Timestamp.timetz
+generated/pandas.Timestamp.to_datetime64,../reference/api/pandas.Timestamp.to_datetime64
+generated/pandas.Timestamp.today,../reference/api/pandas.Timestamp.today
+generated/pandas.Timestamp.to_julian_date,../reference/api/pandas.Timestamp.to_julian_date
+generated/pandas.Timestamp.toordinal,../reference/api/pandas.Timestamp.toordinal
+generated/pandas.Timestamp.to_period,../reference/api/pandas.Timestamp.to_period
+generated/pandas.Timestamp.to_pydatetime,../reference/api/pandas.Timestamp.to_pydatetime
+generated/pandas.Timestamp.tz_convert,../reference/api/pandas.Timestamp.tz_convert
+generated/pandas.Timestamp.tz,../reference/api/pandas.Timestamp.tz
+generated/pandas.Timestamp.tzinfo,../reference/api/pandas.Timestamp.tzinfo
+generated/pandas.Timestamp.tz_localize,../reference/api/pandas.Timestamp.tz_localize
+generated/pandas.Timestamp.tzname,../reference/api/pandas.Timestamp.tzname
+generated/pandas.Timestamp.utcfromtimestamp,../reference/api/pandas.Timestamp.utcfromtimestamp
+generated/pandas.Timestamp.utcnow,../reference/api/pandas.Timestamp.utcnow
+generated/pandas.Timestamp.utcoffset,../reference/api/pandas.Timestamp.utcoffset
+generated/pandas.Timestamp.utctimetuple,../reference/api/pandas.Timestamp.utctimetuple
+generated/pandas.Timestamp.value,../reference/api/pandas.Timestamp.value
+generated/pandas.Timestamp.weekday,../reference/api/pandas.Timestamp.weekday
+generated/pandas.Timestamp.weekday_name,../reference/api/pandas.Timestamp.weekday_name
+generated/pandas.Timestamp.week,../reference/api/pandas.Timestamp.week
+generated/pandas.Timestamp.weekofyear,../reference/api/pandas.Timestamp.weekofyear
+generated/pandas.Timestamp.year,../reference/api/pandas.Timestamp.year
+generated/pandas.to_datetime,../reference/api/pandas.to_datetime
+generated/pandas.to_numeric,../reference/api/pandas.to_numeric
+generated/pandas.to_timedelta,../reference/api/pandas.to_timedelta
+generated/pandas.tseries.frequencies.to_offset,../reference/api/pandas.tseries.frequencies.to_offset
+generated/pandas.unique,../reference/api/pandas.unique
+generated/pandas.util.hash_array,../reference/api/pandas.util.hash_array
+generated/pandas.util.hash_pandas_object,../reference/api/pandas.util.hash_pandas_object
+generated/pandas.wide_to_long,../reference/api/pandas.wide_to_long
+
+# Cached searches
+reference/api/pandas.DataFrame.from_csv,pandas.read_csv
+
+# GroupBy -> DataFrameGroupBy
+reference/api/pandas.core.groupby.GroupBy.__iter__,pandas.core.groupby.DataFrameGroupBy.__iter__
+reference/api/pandas.core.groupby.GroupBy.agg,pandas.core.groupby.DataFrameGroupBy.agg
+reference/api/pandas.core.groupby.GroupBy.aggregate,pandas.core.groupby.DataFrameGroupBy.aggregate
+reference/api/pandas.core.groupby.GroupBy.all,pandas.core.groupby.DataFrameGroupBy.all
+reference/api/pandas.core.groupby.GroupBy.any,pandas.core.groupby.DataFrameGroupBy.any
+reference/api/pandas.core.groupby.GroupBy.apply,pandas.core.groupby.DataFrameGroupBy.apply
+reference/api/pandas.core.groupby.GroupBy.bfill,pandas.core.groupby.DataFrameGroupBy.bfill
+reference/api/pandas.core.groupby.GroupBy.count,pandas.core.groupby.DataFrameGroupBy.count
+reference/api/pandas.core.groupby.GroupBy.cumcount,pandas.core.groupby.DataFrameGroupBy.cumcount
+reference/api/pandas.core.groupby.GroupBy.cummax,pandas.core.groupby.DataFrameGroupBy.cummax
+reference/api/pandas.core.groupby.GroupBy.cummin,pandas.core.groupby.DataFrameGroupBy.cummin
+reference/api/pandas.core.groupby.GroupBy.cumprod,pandas.core.groupby.DataFrameGroupBy.cumprod
+reference/api/pandas.core.groupby.GroupBy.cumsum,pandas.core.groupby.DataFrameGroupBy.cumsum
+reference/api/pandas.core.groupby.GroupBy.ffill,pandas.core.groupby.DataFrameGroupBy.ffill
+reference/api/pandas.core.groupby.GroupBy.first,pandas.core.groupby.DataFrameGroupBy.first
+reference/api/pandas.core.groupby.GroupBy.get_group,pandas.core.groupby.DataFrameGroupBy.get_group
+reference/api/pandas.core.groupby.GroupBy.groups,pandas.core.groupby.DataFrameGroupBy.groups
+reference/api/pandas.core.groupby.GroupBy.head,pandas.core.groupby.DataFrameGroupBy.head
+reference/api/pandas.core.groupby.GroupBy.indices,pandas.core.groupby.DataFrameGroupBy.indices
+reference/api/pandas.core.groupby.GroupBy.last,pandas.core.groupby.DataFrameGroupBy.last
+reference/api/pandas.core.groupby.GroupBy.max,pandas.core.groupby.DataFrameGroupBy.max
+reference/api/pandas.core.groupby.GroupBy.mean,pandas.core.groupby.DataFrameGroupBy.mean
+reference/api/pandas.core.groupby.GroupBy.median,pandas.core.groupby.DataFrameGroupBy.median
+reference/api/pandas.core.groupby.GroupBy.min,pandas.core.groupby.DataFrameGroupBy.min
+reference/api/pandas.core.groupby.GroupBy.ngroup,pandas.core.groupby.DataFrameGroupBy.ngroup
+reference/api/pandas.core.groupby.GroupBy.nth,pandas.core.groupby.DataFrameGroupBy.nth
+reference/api/pandas.core.groupby.GroupBy.ohlc,pandas.core.groupby.DataFrameGroupBy.ohlc
+reference/api/pandas.core.groupby.GroupBy.pct_change,pandas.core.groupby.DataFrameGroupBy.pct_change
+reference/api/pandas.core.groupby.GroupBy.pipe,pandas.core.groupby.DataFrameGroupBy.pipe
+reference/api/pandas.core.groupby.GroupBy.prod,pandas.core.groupby.DataFrameGroupBy.prod
+reference/api/pandas.core.groupby.GroupBy.rank,pandas.core.groupby.DataFrameGroupBy.rank
+reference/api/pandas.core.groupby.GroupBy.sem,pandas.core.groupby.DataFrameGroupBy.sem
+reference/api/pandas.core.groupby.GroupBy.size,pandas.core.groupby.DataFrameGroupBy.size
+reference/api/pandas.core.groupby.GroupBy.std,pandas.core.groupby.DataFrameGroupBy.std
+reference/api/pandas.core.groupby.GroupBy.sum,pandas.core.groupby.DataFrameGroupBy.sum
+reference/api/pandas.core.groupby.GroupBy.tail,pandas.core.groupby.DataFrameGroupBy.tail
+reference/api/pandas.core.groupby.GroupBy.transform,pandas.core.groupby.DataFrameGroupBy.transform
+reference/api/pandas.core.groupby.GroupBy.var,pandas.core.groupby.DataFrameGroupBy.var
+
+# Renamed or alias doc page was removed
+reference/api/pandas.DataFrame.subtract,pandas.DataFrame.sub
+reference/api/pandas.DataFrame.multiply,pandas.DataFrame.mul
+reference/api/pandas.DataFrame.divide,pandas.DataFrame.div
+reference/api/pandas.Series.subtract,pandas.Series.sub
+reference/api/pandas.Series.multiply,pandas.Series.mul
+reference/api/pandas.Series.divide,pandas.Series.div
+reference/api/pandas.Series.tolist,pandas.Series.to_list
+reference/api/pandas.Series.transpose,pandas.Series.T
+reference/api/pandas.Index.transpose,pandas.Index.T
+reference/api/pandas.Index.notnull,pandas.Index.notna
+reference/api/pandas.Index.tolist,pandas.Index.to_list
+reference/api/pandas.core.groupby.DataFrameGroupBy.backfill,pandas.core.groupby.DataFrameGroupBy.bfill
+reference/api/pandas.core.groupby.GroupBy.backfill,pandas.core.groupby.DataFrameGroupBy.bfill
+reference/api/pandas.core.resample.Resampler.backfill,pandas.core.resample.Resampler.bfill
+
+# EWM -> ExponentialMovingWindow
+reference/api/pandas.core.window.ewm.EWM.corr,pandas.core.window.ewm.ExponentialMovingWindow.corr
+reference/api/pandas.core.window.ewm.EWM.cov,pandas.core.window.ewm.ExponentialMovingWindow.cov
+reference/api/pandas.core.window.ewm.EWM.mean,pandas.core.window.ewm.ExponentialMovingWindow.mean
+reference/api/pandas.core.window.ewm.EWM.std,pandas.core.window.ewm.ExponentialMovingWindow.std
+reference/api/pandas.core.window.ewm.EWM.var,pandas.core.window.ewm.ExponentialMovingWindow.var
diff --git a/pandas/doc/scripts/eval_performance.py b/pandas/doc/scripts/eval_performance.py
new file mode 100644
index 0000000000000..0383d4d598d55
--- /dev/null
+++ b/pandas/doc/scripts/eval_performance.py
@@ -0,0 +1,108 @@
+from timeit import repeat as timeit
+
+import numpy as np
+import seaborn as sns
+
+from pandas import DataFrame
+
+setup_common = """from pandas import DataFrame
+import numpy as np
+df = DataFrame(np.random.randn(%d, 3), columns=list('abc'))
+%s"""
+
+setup_with = "s = 'a + b * (c ** 2 + b ** 2 - a) / (a * c) ** 3'"
+
+
+def bench_with(n, times=10, repeat=3, engine="numexpr"):
+    return (
+        np.array(
+            timeit(
+                f"df.eval(s, engine={engine!r})",
+                setup=setup_common % (n, setup_with),
+                repeat=repeat,
+                number=times,
+            )
+        )
+        / times
+    )
+
+
+setup_subset = "s = 'a <= b <= c ** 2 + b ** 2 - a and b > c'"
+
+
+def bench_subset(n, times=20, repeat=3, engine="numexpr"):
+    return (
+        np.array(
+            timeit(
+                f"df.query(s, engine={engine!r})",
+                setup=setup_common % (n, setup_subset),
+                repeat=repeat,
+                number=times,
+            )
+        )
+        / times
+    )
+
+
+def bench(mn=3, mx=7, num=100, engines=("python", "numexpr"), verbose=False):
+    r = np.logspace(mn, mx, num=num).round().astype(int)
+
+    ev = DataFrame(np.empty((num, len(engines))), columns=engines)
+    qu = ev.copy(deep=True)
+
+    ev["size"] = qu["size"] = r
+
+    for engine in engines:
+        for i, n in enumerate(r):
+            if verbose & (i % 10 == 0):
+                print(f"engine: {engine!r}, i == {i:d}")
+            ev_times = bench_with(n, times=1, repeat=1, engine=engine)
+            ev.loc[i, engine] = np.mean(ev_times)
+            qu_times = bench_subset(n, times=1, repeat=1, engine=engine)
+            qu.loc[i, engine] = np.mean(qu_times)
+
+    return ev, qu
+
+
+def plot_perf(df, engines, title, filename=None) -> None:
+    from matplotlib.pyplot import figure
+
+    sns.set()
+    sns.set_palette("Set2")
+
+    fig = figure(figsize=(4, 3), dpi=120)
+    ax = fig.add_subplot(111)
+
+    for engine in engines:
+        ax.loglog(df["size"], df[engine], label=engine, lw=2)
+
+    ax.set_xlabel("Number of Rows")
+    ax.set_ylabel("Time (s)")
+    ax.set_title(title)
+    ax.legend(loc="best")
+    ax.tick_params(top=False, right=False)
+
+    fig.tight_layout()
+
+    if filename is not None:
+        fig.savefig(filename)
+
+
+if __name__ == "__main__":
+    import os
+
+    pandas_dir = os.path.dirname(
+        os.path.dirname(os.path.abspath(os.path.dirname(__file__)))
+    )
+    static_path = os.path.join(pandas_dir, "doc", "source", "_static")
+
+    join = lambda p: os.path.join(static_path, p)
+
+    fn = join("eval-query-perf-data.h5")
+
+    engines = "python", "numexpr"
+
+    ev, qu = bench(verbose=True)  # only this one
+
+    plot_perf(ev, engines, "DataFrame.eval()", filename=join("eval-perf.png"))
+    plot_perf(qu, engines, "DataFrame.query()", filename=join("query-perf.png"))
diff --git a/pandas/doc/source/_static/ci.png b/pandas/doc/source/_static/ci.png
new file mode 100644
index 0000000000000..4754dc2945db5
Binary files /dev/null and b/pandas/doc/source/_static/ci.png differ
diff --git a/pandas/doc/source/_static/css/getting_started.css b/pandas/doc/source/_static/css/getting_started.css
new file mode 100644
index 0000000000000..b02311eb66080
--- /dev/null
+++ b/pandas/doc/source/_static/css/getting_started.css
@@ -0,0 +1,279 @@
+/* Getting started pages */
+
+/* data intro */
+.gs-data {
+  font-size: 0.9rem;
+}
+
+.gs-data-title {
+  align-items: center;
+  font-size: 0.9rem;
+}
+
+.gs-data-header {
+  background-color: var(--pst-color-on-surface);
+}
+
+.gs-data-list {
+  background-color: var(--pst-color-on-background);
+}
+
+.gs-data-title .badge {
+  margin: 10px;
+  padding: 5px;
+}
+
+.gs-data .badge {
+  cursor: pointer;
+  padding: 10px;
+  border: none;
+  text-align: left;
+  outline: none;
+  font-size: 12px;
+}
+
+.gs-data .btn {
+ background-color: grey;
+ border: none;
+}
+
+/* note/alert properties */
+
+.alert-heading {
+  font-size: 1.2rem;
+}
+
+/* callout properties */
+.gs-callout {
+  padding: 20px;
+  margin: 20px 0;
+  border: 1px solid #eee;
+  border-left-width: 5px;
+  border-radius: 3px;
+}
+.gs-callout h4 {
+  margin-top: 0;
+  margin-bottom: 5px;
+}
+.gs-callout p:last-child {
+  margin-bottom: 0;
+}
+.gs-callout code {
+  border-radius: 3px;
+}
+.gs-callout+.gs-callout {
+  margin-top: -5px;
+}
+.gs-callout-remember {
+  border-left-color: var(--pst-color-secondary);
+  align-items: center;
+  font-size: 1.2rem;
+}
+.gs-callout-remember h4 {
+  color:  var(--pst-color-secondary);
+}
+
+/* reference to user guide */
+.gs-torefguide {
+  align-items: center;
+  font-size: 0.9rem;
+  background-color: var(--pst-color-on-background);
+  border-radius: .25rem;
+  padding: 2px;
+}
+
+.gs-torefguide .badge {
+  background-color: var(--pst-color-primary);
+  margin: 10px 10px 10px 10px;
+  padding: 5px;
+}
+
+.gs-torefguide p {
+  margin-top: 1rem;
+}
+
+/* question-task environment */
+
+ul.task-bullet, ol.custom-bullet{
+  list-style:none;
+  padding-left: 0;
+  margin-top: 2em;
+}
+
+ul.task-bullet > li:before {
+    content:"";
+    height:2em;
+    width:2em;
+    display:block;
+    float:left;
+    margin-left:-2em;
+    background-position:center;
+    background-repeat:no-repeat;
+    background-color: var(--pst-color-primary);
+    border-radius: 50%;
+    background-size:100%;
+    background-image:url('../question_mark_noback.svg');
+  }
+
+ul.task-bullet > li {
+  border-left: 1px solid var(--pst-color-primary);
+  padding-left:1em;
+}
+
+ul.task-bullet > li > p:first-child {
+  font-size: 1.1rem;
+  padding-left: 0.75rem;
+}
+
+/* Getting started index page */
+
+.comparison-card {
+  background-color: var(--pst-color-background);
+  border-radius:0;
+  padding: 30px 10px 10px 10px;
+  margin: 10px 0px;
+}
+
+.comparison-card p.card-text {
+  margin: 0px;
+}
+
+.comparison-card .sd-card-img-top {
+  margin: 10px;
+  margin-bottom: 20px;
+  height: 52px;
+  background: none !important;
+}
+
+.comparison-card .sd-btn-secondary {
+  background-color: #6c757d !important;
+  border-color: #6c757d !important;
+}
+
+.comparison-card .sd-btn-secondary:hover {
+  background-color: #5a6268 !important;
+  border-color: #545b62 !important;
+}
+
+.comparison-card .card-footer {
+  border: none;
+  background-color: var(--pst-color-background);
+}
+
+.install-block {
+  padding-bottom: 30px;
+}
+
+.install-card .card-header {
+  border: none;
+  background-color: transparent;
+  padding: 1rem 1rem 0rem 1rem;
+}
+
+.install-card .card-header p.card-text {
+  font-size: 1.1rem;
+  font-weight: bold;
+}
+
+.install-card .card-footer {
+  border: none;
+  background-color: transparent;
+}
+
+.install-card pre {
+  margin: 0 1em 1em 1em;
+}
+
+.install-card .sd-btn-secondary {
+  background-color: #6c757d !important;
+  border-color: #6c757d !important;
+}
+
+.install-card .sd-btn-secondary:hover {
+  background-color: #5a6268 !important;
+  border-color: #545b62 !important;
+}
+
+.custom-button {
+  background-color:#DCDCDC;
+  border: none;
+  color: #484848;
+  text-align: center;
+  text-decoration: none;
+  display: inline-block;
+  font-size: 0.9rem;
+  border-radius: 0.5rem;
+  max-width: 120px;
+  padding: 0.5rem 0rem;
+}
+
+.custom-button a {
+  color: #484848;
+}
+
+.custom-button p {
+  margin-top: 0;
+  margin-bottom: 0rem;
+  color: #484848;
+}
+
+/* intro to tutorial collapsed cards */
+
+.tutorial-accordion {
+  margin-top: 20px;
+  margin-bottom: 20px;
+}
+
+.tutorial-card .card-header.card-link .btn {
+  margin-right: 12px;
+}
+
+.tutorial-card .card-header.card-link .btn:after {
+  content: "-";
+}
+
+.tutorial-card .card-header.card-link.collapsed .btn:after {
+  content: "+";
+}
+
+.tutorial-card-header-1 {
+  justify-content: space-between;
+  align-items: center;
+}
+
+.tutorial-card-header-2 {
+  justify-content: flex-start;
+  align-items: center;
+  font-size: 1.3rem;
+}
+
+.tutorial-card .card-header {
+  --bs-card-cap-color: var(--pst-color-text-base);
+  cursor: pointer;
+  background-color: var(--pst-color-surface);
+  border: 1px solid var(--pst-color-border)
+}
+
+.tutorial-card .card-body {
+  background-color: var(--pst-color-on-background);
+}
+
+.tutorial-card .badge {
+  background-color: var(--pst-color-primary);
+  margin: 10px 10px 10px 10px;
+  padding: 5px;
+}
+
+.tutorial-card .gs-badge-link p {
+  margin: 0px;
+}
+
+
+.tutorial-card .gs-badge-link a {
+  color: var(--pst-color-primary-text);
+  text-decoration: none;
+}
+
+.tutorial-card .badge:hover {
+  background-color: grey;
+}
diff --git a/pandas/doc/source/_static/css/pandas.css b/pandas/doc/source/_static/css/pandas.css
new file mode 100644
index 0000000000000..1145177898737
--- /dev/null
+++ b/pandas/doc/source/_static/css/pandas.css
@@ -0,0 +1,52 @@
+/* Override some aspects of the pydata-sphinx-theme */
+
+:root {
+  /* Use softer blue from bootstrap's default info color */
+  --pst-color-info: 23, 162, 184;
+}
+
+table {
+  width: auto; /* Override fit-content which breaks Styler user guide ipynb */
+}
+
+/* Main index page overview cards */
+
+.intro-card {
+  padding: 30px 10px 20px 10px;
+}
+
+.intro-card .sd-card-img-top {
+  margin: 10px;
+  height: 52px;
+  background: none !important;
+}
+
+.intro-card .sd-card-title {
+  color: var(--pst-color-primary);
+  font-size: var(--pst-font-size-h5);
+  padding: 1rem 0rem 0.5rem 0rem;
+}
+
+.intro-card .sd-card-footer {
+  border: none !important;
+}
+
+.intro-card .sd-card-footer p.sd-card-text {
+  max-width: 220px;
+  margin-left: auto;
+  margin-right: auto;
+}
+
+.intro-card .sd-btn-secondary {
+  background-color: #6c757d !important;
+  border-color: #6c757d !important;
+}
+
+.intro-card .sd-btn-secondary:hover {
+  background-color: #5a6268 !important;
+  border-color: #545b62 !important;
+}
+
+.card, .card img {
+  background-color: var(--pst-color-background);
+}
diff --git a/pandas/doc/source/_static/df_repr_truncated.png b/pandas/doc/source/_static/df_repr_truncated.png
new file mode 100644
index 0000000000000..8f60270358761
Binary files /dev/null and b/pandas/doc/source/_static/df_repr_truncated.png differ
diff --git a/pandas/doc/source/_static/eval-perf.png b/pandas/doc/source/_static/eval-perf.png
new file mode 100644
index 0000000000000..ed92337c1d995
Binary files /dev/null and b/pandas/doc/source/_static/eval-perf.png differ
diff --git a/pandas/doc/source/_static/index_api.svg b/pandas/doc/source/_static/index_api.svg
new file mode 100644
index 0000000000000..69f7ba1d2d114
--- /dev/null
+++ b/pandas/doc/source/_static/index_api.svg
@@ -0,0 +1,97 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="99.058548mm"
+   height="89.967583mm"
+   viewBox="0 0 99.058554 89.967582"
+   version="1.1"
+   id="svg1040"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="index_api.svg">
+  <defs
+     id="defs1034" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.35"
+     inkscape:cx="533.74914"
+     inkscape:cy="10.90433"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="930"
+     inkscape:window-height="472"
+     inkscape:window-x="2349"
+     inkscape:window-y="267"
+     inkscape:window-maximized="0" />
+  <metadata
+     id="metadata1037">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(195.19933,-1.0492759)">
+    <g
+       id="g1008"
+       transform="matrix(1.094977,0,0,1.094977,-521.5523,-198.34055)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path899"
+         d="M 324.96812,187.09499 H 303.0455 v 72.1639 h 22.67969"
+         style="fill:none;stroke:#459DB9;stroke-width:10;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path899-3"
+         d="m 361.58921,187.09499 h 21.92262 v 72.1639 h -22.67969"
+         style="fill:none;stroke:#459DB9;stroke-width:10;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+      <g
+         transform="translate(415.87139,46.162126)"
+         id="g944">
+        <circle
+           style="fill:#459DB9;fill-opacity:1;stroke:#459DB9;stroke-width:4.53704548;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path918"
+           cx="-84.40152"
+           cy="189.84375"
+           r="2.2293637" />
+        <circle
+           style="fill:#459DB9;fill-opacity:1;stroke:#459DB9;stroke-width:4.53704548;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path918-5"
+           cx="-72.949402"
+           cy="189.84375"
+           r="2.2293637" />
+        <circle
+           style="fill:#459DB9;fill-opacity:1;stroke:#459DB9;stroke-width:4.53704548;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path918-6"
+           cx="-61.497284"
+           cy="189.84375"
+           r="2.2293637" />
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/index_contribute.svg b/pandas/doc/source/_static/index_contribute.svg
new file mode 100644
index 0000000000000..de3d902379383
--- /dev/null
+++ b/pandas/doc/source/_static/index_contribute.svg
@@ -0,0 +1,76 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="89.624855mm"
+   height="89.96759mm"
+   viewBox="0 0 89.62486 89.96759"
+   version="1.1"
+   id="svg1040"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="index_contribute.svg">
+  <defs
+     id="defs1034" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.35"
+     inkscape:cx="683.11893"
+     inkscape:cy="-59.078181"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="930"
+     inkscape:window-height="472"
+     inkscape:window-x="2349"
+     inkscape:window-y="267"
+     inkscape:window-maximized="0" />
+  <metadata
+     id="metadata1037">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(234.72009,17.466935)">
+    <g
+       id="g875"
+       transform="matrix(0.99300176,0,0,0.99300176,-133.24106,-172.58804)">
+      <path
+         sodipodi:nodetypes="ccc"
+         inkscape:connector-curvature="0"
+         id="path869"
+         d="m -97.139881,161.26069 47.247024,40.25446 -47.247024,40.25446"
+         style="fill:none;stroke:#459DB9;stroke-width:10;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path871"
+         d="m -49.514879,241.81547 h 32.505951"
+         style="fill:none;stroke:#459DB9;stroke-width:10;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/index_getting_started.svg b/pandas/doc/source/_static/index_getting_started.svg
new file mode 100644
index 0000000000000..2d36622cb7e55
--- /dev/null
+++ b/pandas/doc/source/_static/index_getting_started.svg
@@ -0,0 +1,66 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="101.09389mm"
+   height="89.96759mm"
+   viewBox="0 0 101.09389 89.96759"
+   version="1.1"
+   id="svg1040"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="index_getting_started.svg">
+  <defs
+     id="defs1034" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.35"
+     inkscape:cx="-93.242129"
+     inkscape:cy="-189.9825"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1875"
+     inkscape:window-height="1056"
+     inkscape:window-x="1965"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata1037">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(2.9219487,-8.5995374)">
+    <path
+       style="fill:#459DB9;fill-opacity:1;stroke-width:0.20233451"
+       d="M 37.270955,98.335591 C 33.358064,97.07991 31.237736,92.52319 32.964256,89.08022 c 0.18139,-0.361738 4.757999,-5.096629 10.17021,-10.521968 l 9.84041,-9.864254 -4.03738,-4.041175 -4.037391,-4.041172 -4.96415,4.916665 c -3.61569,3.581096 -5.238959,5.04997 -5.975818,5.407377 l -1.011682,0.490718 H 17.267525 1.5866055 L 0.65034544,70.96512 C -2.2506745,69.535833 -3.5952145,66.18561 -2.5925745,62.884631 c 0.53525,-1.762217 1.61699004,-3.050074 3.22528014,-3.839847 l 1.15623996,-0.56778 13.2591094,-0.05613 13.259111,-0.05613 11.5262,-11.527539 11.526199,-11.527528 H 40.622647 c -12.145542,0 -12.189222,-0.0046 -13.752801,-1.445851 -2.229871,-2.055423 -2.162799,-5.970551 0.135998,-7.938238 1.475193,-1.262712 1.111351,-1.238469 18.588522,-1.238469 12.899229,0 16.035311,0.05193 16.692589,0.276494 0.641832,0.219264 2.590731,2.051402 9.416301,8.852134 l 8.606941,8.575638 h 6.848168 c 4.837422,0 7.092281,0.07311 7.679571,0.249094 0.48064,0.144008 1.22985,0.634863 1.77578,1.163429 2.383085,2.307333 1.968685,6.539886 -0.804989,8.221882 -0.571871,0.346781 -1.38284,0.687226 -1.80217,0.756523 -0.41933,0.06928 -4.2741,0.127016 -8.56615,0.128238 -6.56998,0.0016 -7.977492,-0.04901 -8.902732,-0.321921 -0.975569,-0.287742 -1.400468,-0.622236 -3.783999,-2.978832 l -2.685021,-2.654679 -5.05411,5.051071 -5.0541,5.051081 3.926292,3.947202 c 2.365399,2.378001 4.114289,4.309171 4.399158,4.857713 0.39266,0.75606 0.47311,1.219412 0.474321,2.731516 0.003,3.083647 0.620779,2.331942 -13.598011,16.531349 -10.273768,10.259761 -12.679778,12.563171 -13.500979,12.92519 -1.267042,0.55857 -3.156169,0.681342 -4.390271,0.285321 z m 40.130741,-65.45839 c -2.212909,-0.579748 -3.782711,-1.498393 -5.51275,-3.226063 -2.522111,-2.518633 -3.633121,-5.181304 -3.633121,-8.707194 0,-3.530699 1.11238,-6.197124 3.631161,-8.704043 4.866751,-4.8438383 12.324781,-4.8550953 17.211791,-0.026 3.908758,3.862461 4.818578,9.377999 2.372188,14.380771 -0.846209,1.730481 -3.39493,4.326384 -5.143839,5.239072 -2.69708,1.407492 -6.042829,1.798628 -8.92543,1.043434 z"
+       id="path1000"
+       inkscape:connector-curvature="0" />
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/index_user_guide.svg b/pandas/doc/source/_static/index_user_guide.svg
new file mode 100644
index 0000000000000..bd170535170a3
--- /dev/null
+++ b/pandas/doc/source/_static/index_user_guide.svg
@@ -0,0 +1,67 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="123.72241mm"
+   height="89.96759mm"
+   viewBox="0 0 123.72242 89.96759"
+   version="1.1"
+   id="svg1040"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="index_userguide.svg">
+  <defs
+     id="defs1034" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.35"
+     inkscape:cx="332.26618"
+     inkscape:cy="83.744004"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="930"
+     inkscape:window-height="472"
+     inkscape:window-x="2349"
+     inkscape:window-y="267"
+     inkscape:window-maximized="0" />
+  <metadata
+     id="metadata1037">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(141.8903,-20.32143)">
+    <path
+       style="fill:#459DB9;fill-opacity:1;stroke-width:0.20483544"
+       d="m -139.53374,110.1657 c -0.80428,-0.24884 -1.71513,-1.11296 -2.07107,-1.96486 -0.23905,-0.57214 -0.28453,-6.28104 -0.28453,-35.720988 0,-38.274546 -0.079,-35.840728 1.19849,-36.91568 0.58869,-0.495345 4.63766,-2.187548 8.47998,-3.544073 l 1.58749,-0.560453 v -3.309822 c 0,-3.025538 0.0396,-3.388179 0.46086,-4.222122 0.68808,-1.362003 1.38671,-1.714455 4.60319,-2.322195 4.12797,-0.779966 5.13304,-0.912766 8.81544,-1.16476 11.80964,-0.808168 22.80911,2.509277 30.965439,9.3392 1.750401,1.465747 3.840861,3.5635 5.0903,5.108065 l 0.659122,0.814805 0.659109,-0.814805 c 1.249431,-1.544565 3.33988,-3.642318 5.09029,-5.108065 8.156331,-6.829923 19.155791,-10.147368 30.965441,-9.3392 3.682389,0.251994 4.68748,0.384794 8.81544,1.16476 3.21647,0.60774 3.91511,0.960192 4.60318,2.322195 0.4213,0.833943 0.46087,1.196584 0.46087,4.222122 v 3.309822 l 1.58748,0.560453 c 4.10165,1.448077 7.98852,3.072753 8.5259,3.563743 1.22643,1.120567 1.15258,-1.245868 1.15258,36.927177 0,34.567591 -0.005,35.083151 -0.40663,35.903991 -0.22365,0.45804 -0.73729,1.05665 -1.14143,1.33024 -1.22281,0.82783 -2.17721,0.70485 -5.86813,-0.7561 -9.19595,-3.63998 -18.956011,-6.38443 -26.791332,-7.53353 -3.02827,-0.44412 -9.26189,-0.61543 -11.77821,-0.3237 -5.19357,0.60212 -8.736108,2.05527 -11.700039,4.79936 -0.684501,0.63371 -1.466141,1.23646 -1.736979,1.33942 -0.63859,0.2428 -4.236521,0.2428 -4.875112,0 -0.27083,-0.10296 -1.05247,-0.70571 -1.73696,-1.33942 -2.96395,-2.74409 -6.50648,-4.19724 -11.700058,-4.79936 -2.516312,-0.29173 -8.749941,-0.12042 -11.778201,0.3237 -7.78194,1.14127 -17.39965,3.83907 -26.73341,7.49883 -3.38325,1.32658 -4.15525,1.50926 -5.11851,1.21125 z m 4.2107,-5.34052 c 5.86759,-2.29858 14.40398,-4.922695 20.2018,-6.210065 6.31584,-1.402418 8.5236,-1.646248 14.91592,-1.647338 4.68699,-7.94e-4 6.013661,0.0632 7.257809,0.3497 0.837332,0.19286 1.561052,0.312028 1.60828,0.264819 0.147111,-0.147119 -1.803289,-1.307431 -4.154879,-2.471801 -8.12511,-4.023029 -18.27311,-4.986568 -29.0861,-2.761718 -1.09536,0.22538 -2.32708,0.40827 -2.73715,0.406418 -1.12787,-0.005 -2.3054,-0.76382 -2.84516,-1.8332 l -0.46086,-0.913098 V 62.99179 35.97471 l -0.56331,0.138329 c -0.30981,0.07608 -1.89985,0.665075 -3.5334,1.308881 -2.27551,0.896801 -2.96414,1.252878 -2.94452,1.522563 0.014,0.193604 0.0372,15.284513 0.0512,33.535345 0.014,18.250839 0.0538,33.183322 0.0884,33.183322 0.0346,0 1.02543,-0.3771 2.20198,-0.83801 z m 113.006991,-32.697216 -0.0518,-33.535203 -3.17495,-1.272156 c -1.74623,-0.699685 -3.33627,-1.278755 -3.53341,-1.286819 -0.33966,-0.01389 -0.35847,1.401778 -0.35847,26.980216 v 26.994863 l -0.46087,0.913112 c -0.53976,1.06939 -1.71729,1.828088 -2.84515,1.833189 -0.41008,0.0021 -1.6418,-0.181031 -2.73716,-0.406421 -11.888201,-2.446089 -22.84337,-1.046438 -31.491022,4.02332 -1.68175,0.985941 -2.216748,1.467501 -1.36534,1.228942 1.575181,-0.441362 4.990592,-0.73864 8.524862,-0.742011 5.954408,-0.005 11.43046,0.791951 19.10874,2.78333 3.9516,1.024874 12.1555,3.687454 15.6699,5.085704 1.23926,0.49306 2.36869,0.90517 2.50985,0.9158 0.20489,0.0155 0.2462,-6.745894 0.20483,-33.515866 z m -59.76135,-2.233777 V 40.065438 l -0.95972,-1.357442 c -1.380522,-1.952627 -5.376262,-5.847994 -7.64336,-7.45136 -3.778692,-2.672401 -9.063392,-4.943324 -13.672511,-5.875304 -3.19731,-0.646503 -5.23069,-0.833103 -9.05886,-0.831312 -4.37716,0.0021 -7.70223,0.349169 -11.83461,1.235469 l -1.07538,0.230645 v 31.242342 c 0,26.565778 0.0426,31.226011 0.28429,31.133261 0.15637,-0.06 1.42379,-0.297169 2.81648,-0.527026 12.37657,-2.042634 23.21658,-0.346861 32.521639,5.087596 2.10018,1.226558 5.20202,3.618878 6.880942,5.30692 0.788609,0.792909 1.502978,1.446609 1.587468,1.452679 0.0845,0.006 0.153622,-13.411893 0.153622,-29.817719 z m 5.80221,28.3766 c 6.21476,-6.141601 15.08488,-10.061509 25.025529,-11.05933 4.262419,-0.427849 11.579921,-0.0054 16.017661,0.924912 0.75932,0.15916 1.45259,0.244888 1.54058,0.190498 0.088,-0.05434 0.16003,-14.060382 0.16003,-31.124436 V 26.176883 l -0.52136,-0.198219 c -0.66893,-0.254325 -4.77649,-0.95482 -7.159981,-1.221048 -2.41372,-0.269605 -8.559851,-0.266589 -10.759229,0.0052 -6.458111,0.798299 -12.584091,3.083792 -17.405651,6.49374 -2.267091,1.603366 -6.262831,5.498733 -7.64336,7.45136 l -0.959721,1.357438 v 29.828747 c 0,16.405812 0.0532,29.828746 0.11802,29.828746 0.065,0 0.77928,-0.65347 1.587482,-1.452149 z"
+       id="path845"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="csscccscsssscsssssscscsccsccsccscsscccccccscccccccccsccscscscccscccsccssccsscccscccccsccccsccscsccsscc" />
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/logo_r.svg b/pandas/doc/source/_static/logo_r.svg
new file mode 100644
index 0000000000000..389b03c113e0f
--- /dev/null
+++ b/pandas/doc/source/_static/logo_r.svg
@@ -0,0 +1,14 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" preserveAspectRatio="xMidYMid" width="724" height="561" viewBox="0 0 724 561">
+  <defs>
+    <linearGradient id="gradientFill-1" x1="0" x2="1" y1="0" y2="1" gradientUnits="objectBoundingBox" spreadMethod="pad">
+      <stop offset="0" stop-color="rgb(203,206,208)" stop-opacity="1"/>
+      <stop offset="1" stop-color="rgb(132,131,139)" stop-opacity="1"/>
+    </linearGradient>
+    <linearGradient id="gradientFill-2" x1="0" x2="1" y1="0" y2="1" gradientUnits="objectBoundingBox" spreadMethod="pad">
+      <stop offset="0" stop-color="rgb(39,109,195)" stop-opacity="1"/>
+      <stop offset="1" stop-color="rgb(22,92,170)" stop-opacity="1"/>
+    </linearGradient>
+  </defs>
+  <path d="M361.453,485.937 C162.329,485.937 0.906,377.828 0.906,244.469 C0.906,111.109 162.329,3.000 361.453,3.000 C560.578,3.000 722.000,111.109 722.000,244.469 C722.000,377.828 560.578,485.937 361.453,485.937 ZM416.641,97.406 C265.289,97.406 142.594,171.314 142.594,262.484 C142.594,353.654 265.289,427.562 416.641,427.562 C567.992,427.562 679.687,377.033 679.687,262.484 C679.687,147.971 567.992,97.406 416.641,97.406 Z" fill="url(#gradientFill-1)" fill-rule="evenodd"/>
+  <path d="M550.000,377.000 C550.000,377.000 571.822,383.585 584.500,390.000 C588.899,392.226 596.510,396.668 602.000,402.500 C607.378,408.212 610.000,414.000 610.000,414.000 L696.000,559.000 L557.000,559.062 L492.000,437.000 C492.000,437.000 478.690,414.131 470.500,407.500 C463.668,401.969 460.755,400.000 454.000,400.000 C449.298,400.000 420.974,400.000 420.974,400.000 L421.000,558.974 L298.000,559.026 L298.000,152.938 L545.000,152.938 C545.000,152.938 657.500,154.967 657.500,262.000 C657.500,369.033 550.000,377.000 550.000,377.000 ZM496.500,241.024 L422.037,240.976 L422.000,310.026 L496.500,310.002 C496.500,310.002 531.000,309.895 531.000,274.877 C531.000,239.155 496.500,241.024 496.500,241.024 Z" fill="url(#gradientFill-2)" fill-rule="evenodd"/>
+</svg>
diff --git a/pandas/doc/source/_static/logo_sas.svg b/pandas/doc/source/_static/logo_sas.svg
new file mode 100644
index 0000000000000..d14fa105d49d6
--- /dev/null
+++ b/pandas/doc/source/_static/logo_sas.svg
@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) by Marsupilami -->
+<svg xmlns:svg="http://www.w3.org/2000/svg" xmlns="http://www.w3.org/2000/svg" version="1.0" width="1024" height="420" viewBox="0 0 129.24898 53.067852" id="svg2320">
+  <defs id="defs2322"/>
+  <g transform="translate(-310.37551,-505.82825)" id="layer1">
+    <path d="M 18.46875,0 C 16.638866,-0.041377236 14.748438,0.1725 12.8125,0.65625 C 3.86,2.8925 -6.16125,14.40875 4.78125,27.6875 L 11.3125,35.59375 L 13.15625,37.84375 C 14.39375,39.315 16.41875,39.28875 17.90625,38.0625 C 19.40125,36.829998 19.82625,34.80875 18.59375,33.3125 C 18.592173,33.310397 18.071121,32.679752 17.84375,32.40625 L 17.875,32.40625 C 17.402936,31.838361 17.300473,31.743127 16.8125,31.15625 C 14.448752,28.313409 11.75,25.03125 11.75,25.03125 C 6.9987503,19.265 9.11875,12.14125 15.5625,8.09375 C 21.24,4.5275001 31.65875,5.80125 35.25,11.65625 C 32.988202,4.9805465 26.398246,0.17930136 18.46875,0 z M 19.78125,13.9375 C 18.937031,13.90875 18.055625,14.230625 17.3125,14.84375 C 15.815001,16.0775 15.39375,18.10125 16.625,19.59375 C 16.627499,19.597501 16.77625,19.7825 17.03125,20.09375 C 19.863614,23.496657 23.625,28.0625 23.625,28.0625 C 28.3775,33.82875 26.25625,40.92125 19.8125,44.96875 C 14.136251,48.53375 3.71625,47.2575 0.125,41.40625 C 2.90875,49.618751 12.23875,54.9875 22.5625,52.40625 C 31.5175,50.166248 41.53625,38.6825 30.59375,25.40625 L 22.9375,16.15625 L 22.0625,15.0625 C 21.44375,14.326875 20.625469,13.96625 19.78125,13.9375 z " transform="translate(310.37551,505.82825)" style="fill:#007cc2;fill-opacity:1;fill-rule:nonzero;stroke:none" id="path2440"/>
+    <path d="M 53.53125,6.3125 C 47.8625,6.3125 41.374998,9.2362506 41.375,16.28125 C 41.375,22.9875 46.708752,24.869999 52,26.15625 C 57.355,27.4425 62.656249,28.187498 62.65625,32.65625 C 62.65625,37.05875 58.121251,37.875002 54.78125,37.875 C 50.3725,37.875 46.218751,36.27125 46.03125,31.125 L 40.6875,31.125 C 41,39.79375 47.161249,42.968749 54.46875,42.96875 C 61.085,42.96875 68.343749,40.27125 68.34375,31.9375 C 68.34375,25.16375 63.041251,23.25625 57.6875,21.96875 C 52.7125,20.6825 47.031249,20.00625 47.03125,15.875 C 47.03125,12.3525 50.75625,11.40625 53.96875,11.40625 C 57.49875,11.40625 61.151251,12.810001 61.53125,17.28125 L 66.875,17.28125 C 66.435,8.74625 60.712498,6.3125001 53.53125,6.3125 z M 84.40625,6.3125 C 77.159998,6.3125001 70.90625,9.3625 70.59375,18.03125 L 75.9375,18.03125 C 76.190003,12.883749 79.5275,11.40625 84.0625,11.40625 C 87.466253,11.40625 91.3125,12.20625 91.3125,17.21875 C 91.312501,21.553749 86.2975,21.155 80.375,22.375 C 74.833751,23.526251 69.34375,25.23 69.34375,33.15625 C 69.343748,40.133748 74.20125,42.96875 80.125,42.96875 C 84.656249,42.968749 88.6025,41.255 91.5625,37.53125 C 91.562499,41.322498 93.3525,42.96875 96.125,42.96875 C 97.823751,42.968749 98.9925,42.60875 99.9375,42 L 99.9375,37.53125 C 99.244997,37.802498 98.7525,37.875 98.3125,37.875 C 96.612501,37.875002 96.625,36.68 96.625,33.96875 L 96.625,15.9375 C 96.624998,7.7424996 90.265,6.3125 84.40625,6.3125 z M 112.40625,6.3125 C 106.7375,6.3125 100.25,9.2362506 100.25,16.28125 C 100.25,22.9875 105.61625,24.869999 110.90625,26.15625 C 116.2625,27.4425 121.5625,28.187498 121.5625,32.65625 C 121.5625,37.05875 117.02625,37.875002 113.6875,37.875 C 109.2775,37.875 105.125,36.27125 104.9375,31.125 L 99.5625,31.125 C 99.87625,39.79375 106.06875,42.968749 113.375,42.96875 C 119.9875,42.96875 127.21875,40.27125 127.21875,31.9375 C 127.21875,25.16375 121.91625,23.25625 116.5625,21.96875 C 111.58875,20.6825 105.9375,20.00625 105.9375,15.875 C 105.9375,12.3525 109.63125,11.40625 112.84375,11.40625 C 116.37,11.40625 120.025,12.810001 120.40625,17.28125 L 125.78125,17.28125 C 125.3425,8.74625 119.59,6.3125001 112.40625,6.3125 z M 91.25,24.0625 L 91.25,29.96875 C 91.25,33.1525 88.36875,37.875 81.3125,37.875 C 78.040002,37.875002 75,36.51375 75,32.71875 C 75.000003,28.452501 78.0375,27.115 81.5625,26.4375 C 85.15375,25.761251 89.1725,25.6875 91.25,24.0625 z M 38.21875,39.40625 C 37.088748,39.406249 36.125,40.28375 36.125,41.46875 C 36.125001,42.658749 37.08875,43.53125 38.21875,43.53125 C 39.338748,43.53125 40.28125,42.65875 40.28125,41.46875 C 40.281252,40.283749 39.33875,39.40625 38.21875,39.40625 z M 127.15625,39.40625 C 126.0225,39.406249 125.0625,40.285 125.0625,41.46875 C 125.0625,42.66 126.0225,43.53125 127.15625,43.53125 C 128.275,43.53125 129.25,42.66 129.25,41.46875 C 129.25,40.285 128.275,39.40625 127.15625,39.40625 z M 38.21875,39.75 C 39.146248,39.750002 39.875,40.49 39.875,41.46875 C 39.875,42.456249 39.14625,43.1875 38.21875,43.1875 C 37.273748,43.187501 36.53125,42.45625 36.53125,41.46875 C 36.53125,40.489999 37.27375,39.75 38.21875,39.75 z M 127.15625,39.75 C 128.08375,39.750002 128.84375,40.49125 128.84375,41.46875 C 128.84375,42.4575 128.08375,43.1875 127.15625,43.1875 C 126.21,43.187499 125.5,42.4575 125.5,41.46875 C 125.5,40.49125 126.21,39.75 127.15625,39.75 z M 37.40625,40.28125 L 37.40625,42.65625 L 37.78125,42.65625 L 37.78125,41.625 L 38.1875,41.625 L 38.8125,42.65625 L 39.21875,42.65625 L 38.53125,41.59375 C 38.88375,41.553751 39.15625,41.395 39.15625,40.96875 C 39.156251,40.49875 38.8775,40.28125 38.3125,40.28125 L 37.40625,40.28125 z M 126.375,40.28125 L 126.375,42.65625 L 126.71875,42.65625 L 126.71875,41.625 L 127.15625,41.625 L 127.78125,42.65625 L 128.1875,42.65625 L 127.5,41.59375 C 127.84625,41.554998 128.125,41.395 128.125,40.96875 C 128.125,40.49875 127.8425,40.28125 127.28125,40.28125 L 126.375,40.28125 z M 37.78125,40.59375 L 38.28125,40.59375 C 38.528749,40.593749 38.78125,40.6425 38.78125,40.9375 C 38.78125,41.300001 38.5275,41.3125 38.21875,41.3125 L 37.78125,41.3125 L 37.78125,40.59375 z M 126.71875,40.59375 L 127.21875,40.59375 C 127.47125,40.593749 127.75,40.64125 127.75,40.9375 C 127.75,41.300001 127.4625,41.3125 127.15625,41.3125 L 126.71875,41.3125 L 126.71875,40.59375 z " transform="translate(310.37551,505.82825)" style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none" id="path2448"/>
+  </g>
+<head xmlns=""/></svg>
\ No newline at end of file
diff --git a/pandas/doc/source/_static/logo_sas_white.svg b/pandas/doc/source/_static/logo_sas_white.svg
new file mode 100644
index 0000000000000..8f023fa831b6b
--- /dev/null
+++ b/pandas/doc/source/_static/logo_sas_white.svg
@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) by Marsupilami -->
+<svg xmlns:svg="http://www.w3.org/2000/svg" xmlns="http://www.w3.org/2000/svg" version="1.0" width="1024" height="420" viewBox="0 0 129.24898 53.067852" id="svg2320">
+  <defs id="defs2322"/>
+  <g transform="translate(-310.37551,-505.82825)" id="layer1">
+    <path d="M 18.46875,0 C 16.638866,-0.041377236 14.748438,0.1725 12.8125,0.65625 C 3.86,2.8925 -6.16125,14.40875 4.78125,27.6875 L 11.3125,35.59375 L 13.15625,37.84375 C 14.39375,39.315 16.41875,39.28875 17.90625,38.0625 C 19.40125,36.829998 19.82625,34.80875 18.59375,33.3125 C 18.592173,33.310397 18.071121,32.679752 17.84375,32.40625 L 17.875,32.40625 C 17.402936,31.838361 17.300473,31.743127 16.8125,31.15625 C 14.448752,28.313409 11.75,25.03125 11.75,25.03125 C 6.9987503,19.265 9.11875,12.14125 15.5625,8.09375 C 21.24,4.5275001 31.65875,5.80125 35.25,11.65625 C 32.988202,4.9805465 26.398246,0.17930136 18.46875,0 z M 19.78125,13.9375 C 18.937031,13.90875 18.055625,14.230625 17.3125,14.84375 C 15.815001,16.0775 15.39375,18.10125 16.625,19.59375 C 16.627499,19.597501 16.77625,19.7825 17.03125,20.09375 C 19.863614,23.496657 23.625,28.0625 23.625,28.0625 C 28.3775,33.82875 26.25625,40.92125 19.8125,44.96875 C 14.136251,48.53375 3.71625,47.2575 0.125,41.40625 C 2.90875,49.618751 12.23875,54.9875 22.5625,52.40625 C 31.5175,50.166248 41.53625,38.6825 30.59375,25.40625 L 22.9375,16.15625 L 22.0625,15.0625 C 21.44375,14.326875 20.625469,13.96625 19.78125,13.9375 z " transform="translate(310.37551,505.82825)" style="fill:#007cc2;fill-opacity:1;fill-rule:nonzero;stroke:none" id="path2440"/>
+    <path d="M 53.53125,6.3125 C 47.8625,6.3125 41.374998,9.2362506 41.375,16.28125 C 41.375,22.9875 46.708752,24.869999 52,26.15625 C 57.355,27.4425 62.656249,28.187498 62.65625,32.65625 C 62.65625,37.05875 58.121251,37.875002 54.78125,37.875 C 50.3725,37.875 46.218751,36.27125 46.03125,31.125 L 40.6875,31.125 C 41,39.79375 47.161249,42.968749 54.46875,42.96875 C 61.085,42.96875 68.343749,40.27125 68.34375,31.9375 C 68.34375,25.16375 63.041251,23.25625 57.6875,21.96875 C 52.7125,20.6825 47.031249,20.00625 47.03125,15.875 C 47.03125,12.3525 50.75625,11.40625 53.96875,11.40625 C 57.49875,11.40625 61.151251,12.810001 61.53125,17.28125 L 66.875,17.28125 C 66.435,8.74625 60.712498,6.3125001 53.53125,6.3125 z M 84.40625,6.3125 C 77.159998,6.3125001 70.90625,9.3625 70.59375,18.03125 L 75.9375,18.03125 C 76.190003,12.883749 79.5275,11.40625 84.0625,11.40625 C 87.466253,11.40625 91.3125,12.20625 91.3125,17.21875 C 91.312501,21.553749 86.2975,21.155 80.375,22.375 C 74.833751,23.526251 69.34375,25.23 69.34375,33.15625 C 69.343748,40.133748 74.20125,42.96875 80.125,42.96875 C 84.656249,42.968749 88.6025,41.255 91.5625,37.53125 C 91.562499,41.322498 93.3525,42.96875 96.125,42.96875 C 97.823751,42.968749 98.9925,42.60875 99.9375,42 L 99.9375,37.53125 C 99.244997,37.802498 98.7525,37.875 98.3125,37.875 C 96.612501,37.875002 96.625,36.68 96.625,33.96875 L 96.625,15.9375 C 96.624998,7.7424996 90.265,6.3125 84.40625,6.3125 z M 112.40625,6.3125 C 106.7375,6.3125 100.25,9.2362506 100.25,16.28125 C 100.25,22.9875 105.61625,24.869999 110.90625,26.15625 C 116.2625,27.4425 121.5625,28.187498 121.5625,32.65625 C 121.5625,37.05875 117.02625,37.875002 113.6875,37.875 C 109.2775,37.875 105.125,36.27125 104.9375,31.125 L 99.5625,31.125 C 99.87625,39.79375 106.06875,42.968749 113.375,42.96875 C 119.9875,42.96875 127.21875,40.27125 127.21875,31.9375 C 127.21875,25.16375 121.91625,23.25625 116.5625,21.96875 C 111.58875,20.6825 105.9375,20.00625 105.9375,15.875 C 105.9375,12.3525 109.63125,11.40625 112.84375,11.40625 C 116.37,11.40625 120.025,12.810001 120.40625,17.28125 L 125.78125,17.28125 C 125.3425,8.74625 119.59,6.3125001 112.40625,6.3125 z M 91.25,24.0625 L 91.25,29.96875 C 91.25,33.1525 88.36875,37.875 81.3125,37.875 C 78.040002,37.875002 75,36.51375 75,32.71875 C 75.000003,28.452501 78.0375,27.115 81.5625,26.4375 C 85.15375,25.761251 89.1725,25.6875 91.25,24.0625 z M 38.21875,39.40625 C 37.088748,39.406249 36.125,40.28375 36.125,41.46875 C 36.125001,42.658749 37.08875,43.53125 38.21875,43.53125 C 39.338748,43.53125 40.28125,42.65875 40.28125,41.46875 C 40.281252,40.283749 39.33875,39.40625 38.21875,39.40625 z M 127.15625,39.40625 C 126.0225,39.406249 125.0625,40.285 125.0625,41.46875 C 125.0625,42.66 126.0225,43.53125 127.15625,43.53125 C 128.275,43.53125 129.25,42.66 129.25,41.46875 C 129.25,40.285 128.275,39.40625 127.15625,39.40625 z M 38.21875,39.75 C 39.146248,39.750002 39.875,40.49 39.875,41.46875 C 39.875,42.456249 39.14625,43.1875 38.21875,43.1875 C 37.273748,43.187501 36.53125,42.45625 36.53125,41.46875 C 36.53125,40.489999 37.27375,39.75 38.21875,39.75 z M 127.15625,39.75 C 128.08375,39.750002 128.84375,40.49125 128.84375,41.46875 C 128.84375,42.4575 128.08375,43.1875 127.15625,43.1875 C 126.21,43.187499 125.5,42.4575 125.5,41.46875 C 125.5,40.49125 126.21,39.75 127.15625,39.75 z M 37.40625,40.28125 L 37.40625,42.65625 L 37.78125,42.65625 L 37.78125,41.625 L 38.1875,41.625 L 38.8125,42.65625 L 39.21875,42.65625 L 38.53125,41.59375 C 38.88375,41.553751 39.15625,41.395 39.15625,40.96875 C 39.156251,40.49875 38.8775,40.28125 38.3125,40.28125 L 37.40625,40.28125 z M 126.375,40.28125 L 126.375,42.65625 L 126.71875,42.65625 L 126.71875,41.625 L 127.15625,41.625 L 127.78125,42.65625 L 128.1875,42.65625 L 127.5,41.59375 C 127.84625,41.554998 128.125,41.395 128.125,40.96875 C 128.125,40.49875 127.8425,40.28125 127.28125,40.28125 L 126.375,40.28125 z M 37.78125,40.59375 L 38.28125,40.59375 C 38.528749,40.593749 38.78125,40.6425 38.78125,40.9375 C 38.78125,41.300001 38.5275,41.3125 38.21875,41.3125 L 37.78125,41.3125 L 37.78125,40.59375 z M 126.71875,40.59375 L 127.21875,40.59375 C 127.47125,40.593749 127.75,40.64125 127.75,40.9375 C 127.75,41.300001 127.4625,41.3125 127.15625,41.3125 L 126.71875,41.3125 L 126.71875,40.59375 z " transform="translate(310.37551,505.82825)" style="fill:#ffffff;fill-opacity:1;fill-rule:nonzero;stroke:none" id="path2448"/>
+  </g>
+<head xmlns=""/></svg>
\ No newline at end of file
diff --git a/pandas/doc/source/_static/logo_sql.svg b/pandas/doc/source/_static/logo_sql.svg
new file mode 100644
index 0000000000000..38b3b2c726214
--- /dev/null
+++ b/pandas/doc/source/_static/logo_sql.svg
@@ -0,0 +1,73 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Generated by IcoMoon.io -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   version="1.1"
+   width="20.558033"
+   height="29.264381"
+   viewBox="0 0 20.558033 29.264382"
+   id="svg4"
+   sodipodi:docname="logo_sql.svg"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)">
+  <metadata
+     id="metadata10">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <defs
+     id="defs8" />
+  <sodipodi:namedview
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1"
+     objecttolerance="10"
+     gridtolerance="10"
+     guidetolerance="10"
+     inkscape:pageopacity="0"
+     inkscape:pageshadow="2"
+     inkscape:window-width="1871"
+     inkscape:window-height="1029"
+     id="namedview6"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:zoom="16.68772"
+     inkscape:cx="-2.295492"
+     inkscape:cy="16.594944"
+     inkscape:window-x="1649"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="svg4" />
+  <path
+     d="m 18.846017,1.608 c -0.497,-0.326 -1.193,-0.615 -2.069,-0.858 -1.742,-0.484 -4.05,-0.75 -6.498,-0.75 -2.4480004,0 -4.7560004,0.267 -6.4980004,0.75 -0.877,0.243 -1.573,0.532 -2.069,0.858 -0.619,0.407 -0.93299996,0.874 -0.93299996,1.391 v 12 c 0,0.517 0.31399996,0.985 0.93299996,1.391 0.497,0.326 1.193,0.615 2.069,0.858 1.742,0.484 4.05,0.75 6.4980004,0.75 2.448,0 4.756,-0.267 6.498,-0.751 0.877,-0.243 1.573,-0.532 2.069,-0.858 0.619,-0.406 0.933,-0.874 0.933,-1.391 v -12 c 0,-0.517 -0.314,-0.985 -0.933,-1.391 z M 4.0490166,1.713 c 1.658,-0.46 3.87,-0.714 6.2300004,-0.714 2.36,0 4.573,0.254 6.23,0.714 1.795,0.499 2.27,1.059 2.27,1.286 0,0.227 -0.474,0.787 -2.27,1.286 -1.658,0.46 -3.87,0.714 -6.23,0.714 -2.3600004,0 -4.5730004,-0.254 -6.2300004,-0.714 -1.795,-0.499 -2.27,-1.059 -2.27,-1.286 0,-0.227 0.474,-0.787 2.27,-1.286 z M 16.509017,16.285 c -1.658,0.46 -3.87,0.714 -6.23,0.714 -2.3600004,0 -4.5730004,-0.254 -6.2300004,-0.714 -1.795,-0.499 -2.27,-1.059 -2.27,-1.286 v -2.566 c 0.492,0.309 1.164,0.583 2.002,0.816 1.742,0.484 4.05,0.75 6.4980004,0.75 2.448,0 4.756,-0.267 6.498,-0.751 0.838,-0.233 1.511,-0.507 2.002,-0.816 v 2.566 c 0,0.227 -0.474,0.787 -2.27,1.286 z m 0,-4 c -1.658,0.46 -3.87,0.714 -6.23,0.714 -2.3600004,0 -4.5730004,-0.254 -6.2300004,-0.714 -1.795,-0.499 -2.27,-1.059 -2.27,-1.286 V 8.433 c 0.492,0.309 1.164,0.583 2.002,0.816 1.742,0.484 4.05,0.75 6.4980004,0.75 2.448,0 4.756,-0.267 6.498,-0.75 0.838,-0.233 1.511,-0.507 2.002,-0.816 v 2.566 c 0,0.227 -0.474,0.787 -2.27,1.286 z m 0,-4 c -1.658,0.46 -3.87,0.714 -6.23,0.714 -2.3600004,0 -4.5730004,-0.254 -6.2300004,-0.714 -1.795,-0.499 -2.27,-1.059 -2.27,-1.286 V 4.433 c 0.492,0.309 1.164,0.583 2.002,0.816 1.742,0.484 4.05,0.75 6.4980004,0.75 2.448,0 4.756,-0.267 6.498,-0.75 0.838,-0.233 1.511,-0.507 2.002,-0.816 v 2.566 c 0,0.227 -0.474,0.787 -2.27,1.286 z"
+     id="path2"
+     inkscape:connector-curvature="0"
+     style="fill:#888888" />
+  <text
+     xml:space="preserve"
+     style="font-style:normal;font-weight:normal;font-size:32.32878494px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#888888;fill-opacity:1;stroke:none;stroke-width:0.99999994"
+     x="-0.71034926"
+     y="27.875254"
+     id="text819"><tspan
+       sodipodi:role="line"
+       id="tspan817"
+       x="-0.71034926"
+       y="27.875254"
+       style="font-size:10.77626133px;stroke-width:0.99999994">SQL</tspan></text>
+</svg>
diff --git a/pandas/doc/source/_static/logo_stata.svg b/pandas/doc/source/_static/logo_stata.svg
new file mode 100644
index 0000000000000..a6e3f1d221959
--- /dev/null
+++ b/pandas/doc/source/_static/logo_stata.svg
@@ -0,0 +1,17 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 415.27 95.89">
+  <defs>
+    <style>
+      .cls-1 {
+        fill: #135f90;
+      }
+    </style>
+  </defs>
+  <title>stata-logo-blue</title>
+  <g id="Foreground">
+    <path class="cls-1" d="M94.23,0l-8,21.31H29.3a8,8,0,0,0,0,16H64.93a29.3,29.3,0,0,1,0,58.6H0L8,74.58H64.93a8,8,0,1,0,0-16H29.3A29.3,29.3,0,0,1,29.3,0Z"/>
+    <polygon class="cls-1" points="259.9 0 251.91 21.31 278.55 21.31 278.55 95.89 299.86 95.89 299.86 21.31 321.16 21.31 329.15 0 259.9 0"/>
+    <path class="cls-1" d="M386,58.6H345a8,8,0,0,0,0,16h49v-16ZM366.31,37.29H394V29a8,8,0,0,0-8-7.65H323.69l8-21.31H386a29.3,29.3,0,0,1,29.3,29.3V95.89H345a29.3,29.3,0,0,1,0-58.6Z"/>
+    <path class="cls-1" d="M222.8,58.6h-41a8,8,0,1,0,0,16h48.95v-16ZM212.14,37.29H230.8V29a8,8,0,0,0-8-7.65H160.53l8-21.31H222.8a29.3,29.3,0,0,1,29.3,29.3V95.89H181.84a29.3,29.3,0,0,1,0-58.6Z"/>
+    <polygon class="cls-1" points="96.73 0 88.74 21.31 115.38 21.31 115.38 95.89 136.69 95.89 136.69 21.31 158 21.31 165.99 0 96.73 0"/>
+  </g>
+<head xmlns=""/></svg>
\ No newline at end of file
diff --git a/pandas/doc/source/_static/new-excel-index.png b/pandas/doc/source/_static/new-excel-index.png
new file mode 100644
index 0000000000000..479237c3712d2
Binary files /dev/null and b/pandas/doc/source/_static/new-excel-index.png differ
diff --git a/pandas/doc/source/_static/old-excel-index.png b/pandas/doc/source/_static/old-excel-index.png
new file mode 100644
index 0000000000000..5281367a5ad9d
Binary files /dev/null and b/pandas/doc/source/_static/old-excel-index.png differ
diff --git a/pandas/doc/source/_static/print_df_new.png b/pandas/doc/source/_static/print_df_new.png
new file mode 100644
index 0000000000000..767d7d3f0ef06
Binary files /dev/null and b/pandas/doc/source/_static/print_df_new.png differ
diff --git a/pandas/doc/source/_static/print_df_old.png b/pandas/doc/source/_static/print_df_old.png
new file mode 100644
index 0000000000000..5f458722f1269
Binary files /dev/null and b/pandas/doc/source/_static/print_df_old.png differ
diff --git a/pandas/doc/source/_static/query-perf.png b/pandas/doc/source/_static/query-perf.png
new file mode 100644
index 0000000000000..c52849a0edd53
Binary files /dev/null and b/pandas/doc/source/_static/query-perf.png differ
diff --git a/pandas/doc/source/_static/question_mark_noback.svg b/pandas/doc/source/_static/question_mark_noback.svg
new file mode 100644
index 0000000000000..3abb4b806d20a
--- /dev/null
+++ b/pandas/doc/source/_static/question_mark_noback.svg
@@ -0,0 +1,72 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="6.1681423mm"
+   height="6.1681423mm"
+   viewBox="0 0 6.1681423 6.1681423"
+   version="1.1"
+   id="svg856"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="question_mark_noback.svg">
+  <defs
+     id="defs850" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="11.2"
+     inkscape:cx="4.3447038"
+     inkscape:cy="5.995975"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1600"
+     inkscape:window-height="876"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata853">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-71.755217,-98.124272)">
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:1.1868099px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335"
+       x="73.403717"
+       y="103.38733"
+       id="text1405"><tspan
+         sodipodi:role="line"
+         id="tspan1403"
+         x="73.403717"
+         y="103.38733"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.6966877px;font-family:'Noto Sans Mono CJK HK';-inkscape-font-specification:'Noto Sans Mono CJK HK';fill:#ffffff;fill-opacity:1;stroke-width:0.26458335">?</tspan></text>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/reshaping_melt.png b/pandas/doc/source/_static/reshaping_melt.png
new file mode 100644
index 0000000000000..d0c4e77655e60
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_melt.png differ
diff --git a/pandas/doc/source/_static/reshaping_pivot.png b/pandas/doc/source/_static/reshaping_pivot.png
new file mode 100644
index 0000000000000..6d779562adcac
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_pivot.png differ
diff --git a/pandas/doc/source/_static/reshaping_stack.png b/pandas/doc/source/_static/reshaping_stack.png
new file mode 100644
index 0000000000000..924f916ae0d37
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_stack.png differ
diff --git a/pandas/doc/source/_static/reshaping_unstack.png b/pandas/doc/source/_static/reshaping_unstack.png
new file mode 100644
index 0000000000000..3e14cdd1ee1f7
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_unstack.png differ
diff --git a/pandas/doc/source/_static/reshaping_unstack_0.png b/pandas/doc/source/_static/reshaping_unstack_0.png
new file mode 100644
index 0000000000000..eceddf73eea9e
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_unstack_0.png differ
diff --git a/pandas/doc/source/_static/reshaping_unstack_1.png b/pandas/doc/source/_static/reshaping_unstack_1.png
new file mode 100644
index 0000000000000..ab0ae3796dcc1
Binary files /dev/null and b/pandas/doc/source/_static/reshaping_unstack_1.png differ
diff --git a/pandas/doc/source/_static/schemas/01_table_dataframe.svg b/pandas/doc/source/_static/schemas/01_table_dataframe.svg
new file mode 100644
index 0000000000000..9bd1c217b3ca2
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/01_table_dataframe.svg
@@ -0,0 +1,262 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="143.19496mm"
+   height="104.30615mm"
+   viewBox="0 0 143.19496 104.30615"
+   version="1.1"
+   id="svg10280"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="01_table_dataframe.svg">
+  <defs
+     id="defs10274" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="355.65317"
+     inkscape:cy="142.80245"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0" />
+  <metadata
+     id="metadata10277">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-22.613419,-96.097789)">
+    <g
+       id="g888"
+       transform="matrix(0.89990753,0,0,0.9,9.4364163,14.825088)"
+       style="stroke-width:1.11116815">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 23.647349,141.16281 H 48.53529 V 128.97485 H 23.647349 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 50.063339,127.4468 H 74.951291 V 115.25884 H 50.063339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 50.063339,141.16281 H 74.951291 V 128.97485 H 50.063339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 23.647339,153.86281 H 48.53529 V 141.67486 H 23.647339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 50.063339,153.86281 H 74.951291 V 141.67486 H 50.063339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 75.463341,127.4468 H 100.3513 V 115.25884 H 75.463341 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 75.463341,141.16281 H 100.3513 V 128.97485 H 75.463341 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 75.463341,153.86281 H 100.3513 V 141.67486 H 75.463341 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 23.647349,179.26284 H 48.53529 V 167.07487 H 23.647349 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 50.063339,179.26284 H 74.951291 V 167.07487 H 50.063339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 75.463341,179.26284 H 100.3513 V 167.07487 H 75.463341 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 23.647349,191.96284 H 48.53529 V 179.77488 H 23.647349 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 50.063339,191.96284 H 74.951291 V 179.77488 H 50.063339 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 75.463341,191.96284 H 100.3513 V 179.77488 H 75.463341 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 100.86334,127.4468 h 24.88794 v -12.18796 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 100.86335,141.16281 h 24.88793 v -12.18796 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 100.86334,153.86281 h 24.88794 v -12.18795 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 100.86335,179.26284 h 24.88793 v -12.18797 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 100.86335,191.96284 h 24.88793 v -12.18796 h -24.88793 z" />
+      <g
+         id="g935"
+         style="stroke-width:1.11116815">
+        <path
+           d="M 23.647349,166.56283 H 48.53529 V 154.37487 H 23.647349 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 50.063339,166.56283 H 74.951291 V 154.37487 H 50.063339 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 75.463341,166.56283 H 100.3513 V 154.37487 H 75.463341 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 100.86335,166.56283 h 24.88793 v -12.18796 h -24.88793 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 126.26334,166.56283 h 24.88792 v -12.18796 h -24.88792 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-4"
+           inkscape:connector-curvature="0" />
+      </g>
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 126.26333,127.4468 h 24.88793 v -12.18796 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 126.26334,141.16281 h 24.88792 v -12.18796 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 126.26333,153.86281 h 24.88793 v -12.18795 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 126.26334,179.26284 h 24.88792 v -12.18797 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56897205;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 126.26334,191.96284 h 24.88792 v -12.18796 h -24.88792 z" />
+      <text
+         id="text47247-9"
+         y="200.30403"
+         x="100.55013"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29399657"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.05555534px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29399657"
+           y="200.30403"
+           x="100.55013"
+           id="tspan47245-7"
+           sodipodi:role="line">column</tspan></text>
+      <text
+         id="text47247-1"
+         y="103.81308"
+         x="76.306671"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29399657"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29399657"
+           y="103.81308"
+           x="76.306671"
+           id="tspan47245-3"
+           sodipodi:role="line">DataFrame</tspan></text>
+      <rect
+         y="113.61689"
+         x="100.55073"
+         height="79.987907"
+         width="25.513165"
+         id="rect850"
+         style="fill:none;fill-opacity:1;stroke:#000000;stroke-width:0.29399657;stroke-opacity:1" />
+      <rect
+         y="154.10715"
+         x="22.74571"
+         height="12.771029"
+         width="129.30719"
+         id="rect850-3"
+         style="fill:none;fill-opacity:1;stroke:#000000;stroke-width:0.29399657;stroke-opacity:1" />
+      <text
+         id="text47247-9-6"
+         y="162.41847"
+         x="153.07187"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29399657"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.05555534px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29399657"
+           y="162.41847"
+           x="153.07187"
+           id="tspan47245-7-7"
+           sodipodi:role="line">row</tspan></text>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/01_table_series.svg b/pandas/doc/source/_static/schemas/01_table_series.svg
new file mode 100644
index 0000000000000..d52c882f26868
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/01_table_series.svg
@@ -0,0 +1,127 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="51.815998mm"
+   height="81.350258mm"
+   viewBox="0 0 51.815998 81.350261"
+   version="1.1"
+   id="svg10280"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="01_table_series.svg">
+  <defs
+     id="defs10274" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="2.1693488"
+     inkscape:cx="97.919996"
+     inkscape:cy="153.73277"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0" />
+  <metadata
+     id="metadata10277">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-79.092005,-111.90219)">
+    <g
+       id="g836"
+       transform="matrix(0.89900193,0,0,0.89968429,10.604797,15.293015)"
+       style="stroke-width:1.11192274">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 79.348032,142.1964 H 104.23598 V 130.00844 H 79.348032 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 105.76403,142.1964 h 24.88795 v -12.18796 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-90"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 79.348029,154.8964 H 104.23598 V 142.70845 H 79.348029 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 105.76403,154.8964 h 24.88795 v -12.18795 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 79.348032,167.59641 H 104.23598 V 155.40846 H 79.348032 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 105.76403,167.59641 h 24.88795 v -12.18795 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 79.348032,180.29643 H 104.23598 V 168.10846 H 79.348032 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 105.76403,180.29643 h 24.88795 v -12.18797 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 79.348032,192.99642 H 104.23598 V 180.80847 H 79.348032 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56935841;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 105.76403,192.99642 h 24.88795 v -12.18795 h -24.88795 z" />
+      <text
+         id="text47247-1-3"
+         y="119.94304"
+         x="86.063171"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29419622"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29419622"
+           y="119.94304"
+           x="86.063171"
+           id="tspan47245-3-1"
+           sodipodi:role="line">Series</tspan></text>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/01_table_spreadsheet.png b/pandas/doc/source/_static/schemas/01_table_spreadsheet.png
new file mode 100644
index 0000000000000..4e3497879de31
Binary files /dev/null and b/pandas/doc/source/_static/schemas/01_table_spreadsheet.png differ
diff --git a/pandas/doc/source/_static/schemas/02_io_readwrite.svg b/pandas/doc/source/_static/schemas/02_io_readwrite.svg
new file mode 100644
index 0000000000000..a99a6d731a6ad
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/02_io_readwrite.svg
@@ -0,0 +1,1401 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:xlink="http://www.w3.org/1999/xlink"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="295.1355mm"
+   height="79.820137mm"
+   viewBox="0 0 295.1355 79.82014"
+   version="1.1"
+   id="svg11307"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="02_io_readwrite.svg">
+  <defs
+     id="defs11301">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-2"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <linearGradient
+       y2="120.7894"
+       x2="63.999699"
+       y1="7.0335999"
+       x1="63.999699"
+       gradientUnits="userSpaceOnUse"
+       id="SVGID_1_">
+      <stop
+         id="stop56586"
+         style="stop-color:#4387FD"
+         offset="0" />
+      <stop
+         id="stop56588"
+         style="stop-color:#4683EA"
+         offset="1" />
+    </linearGradient>
+    <clipPath
+       id="SVGID_2_">
+      <use
+         id="use56597"
+         style="overflow:visible"
+         xlink:href="#SVGID_6_"
+         x="0"
+         y="0"
+         width="100%"
+         height="100%" />
+    </clipPath>
+    <path
+       d="M 27.7908,115.2166 1.5403,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7908,12.7831 C 29.8449,9.2253 33.641,7.0336 37.7493,7.0336 h 52.501 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2505,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2505,45.4672 c -2.0541,3.5578 -5.8502,5.7495 -9.9585,5.7495 h -52.501 c -4.1083,1e-4 -7.9044,-2.1916 -9.9585,-5.7494 z"
+       id="path56636"
+       inkscape:connector-curvature="0" />
+    <linearGradient
+       y2="120.7894"
+       x2="63.999699"
+       y1="7.0335999"
+       x1="63.999699"
+       gradientUnits="userSpaceOnUse"
+       id="SVGID_1_-2">
+      <stop
+         id="stop56586-7"
+         style="stop-color:#4387FD"
+         offset="0" />
+      <stop
+         id="stop56588-0"
+         style="stop-color:#4683EA"
+         offset="1" />
+    </linearGradient>
+    <clipPath
+       id="SVGID_2_-9">
+      <use
+         id="use56597-3"
+         style="overflow:visible"
+         xlink:href="#SVGID_6_"
+         x="0"
+         y="0"
+         width="100%"
+         height="100%" />
+    </clipPath>
+    <path
+       inkscape:connector-curvature="0"
+       id="path12064-9"
+       d="M 27.7908,115.2166 1.5403,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7908,12.7831 C 29.8449,9.2253 33.641,7.0336 37.7493,7.0336 h 52.501 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2505,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2505,45.4672 c -2.0541,3.5578 -5.8502,5.7495 -9.9585,5.7495 h -52.501 c -4.1083,1e-4 -7.9044,-2.1916 -9.9585,-5.7494 z" />
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-0"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-9"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="609.1577"
+     inkscape:cy="-5.1149931"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata11304">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(45.015806,-5.4401261)">
+    <g
+       id="g1332"
+       transform="matrix(0.89992508,0,0,0.89964684,10.262878,4.5510367)"
+       style="stroke-width:1.11137545">
+      <text
+         inkscape:export-ydpi="96"
+         inkscape:export-xdpi="96"
+         id="text47247-6"
+         y="37.291649"
+         x="25.401798"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.05555534px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29405141"
+           y="37.291649"
+           x="25.401798"
+           id="tspan47245-1"
+           sodipodi:role="line">read_*</tspan></text>
+      <text
+         inkscape:export-ydpi="96"
+         inkscape:export-xdpi="96"
+         id="text47247-6-5"
+         y="37.229637"
+         x="158.19954"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.05555534px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.29405141"
+           y="37.229637"
+           x="158.19954"
+           id="tspan47245-1-5"
+           sodipodi:role="line">to_*</tspan></text>
+      <g
+         transform="matrix(0.87620685,0,0,0.87620685,-48.432527,-55.405379)"
+         id="g12198"
+         style="stroke-width:1.11137545">
+        <path
+           d="m 121.26405,115.97458 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-80"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 147.68004,102.25857 h 24.88795 V 90.070615 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 147.68004,115.97458 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 121.26404,128.67458 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 147.68004,128.67458 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-69"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 173.08004,102.25857 H 197.968 V 90.070615 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-21"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 173.08004,115.97458 H 197.968 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 173.08004,128.67458 H 197.968 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 121.26405,141.37459 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 147.68004,141.37459 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 173.08004,141.37459 H 197.968 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 198.48005,141.37459 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 198.48004,102.25857 h 24.88795 V 90.070615 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-65"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 198.48005,115.97458 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-68"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 198.48004,128.67458 h 24.88795 v -12.18795 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56907815;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-4"
+           inkscape:connector-curvature="0" />
+      </g>
+      <g
+         transform="matrix(0.78210044,0,0,0.78210044,252.41915,32.616607)"
+         id="g13221-6"
+         style="stroke-width:1.11137545">
+        <g
+           id="g13046-0"
+           transform="translate(0,3.4017917)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.76397361,0,0,0.76397361,301.89818,-142.00712)"
+             id="g57096-9-6"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -506.51101,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-1-2"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -473.73325,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-2-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-499.97018"
+               y="155.93517"
+               id="text55709-5-8-7-1"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-0-8"
+                 x="-499.97018"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">CSV</tspan></text>
+            <path
+               style="fill:#755075;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -506.27749,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-9-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <g
+               transform="translate(-681.73064,244.47732)"
+               id="g56212-3-9"
+               style="stroke-width:1.11137545">
+              <rect
+                 y="-83.565544"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-60-2"
+                 height="2.2165694" />
+              <rect
+                 y="-79.919731"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-6-6-0"
+                 height="2.2165694" />
+              <rect
+                 y="-76.273918"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-4-2-2"
+                 height="2.2165694" />
+              <rect
+                 y="-72.628105"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-5-6-3"
+                 height="2.2165694" />
+            </g>
+          </g>
+          <g
+             transform="matrix(0.76397361,0,0,0.76397361,389.69361,-142.00712)"
+             id="g57123-7"
+             style="stroke-width:1.11137545">
+            <g
+               id="g56942-5"
+               style="stroke-width:1.11137545">
+              <path
+                 sodipodi:nodetypes="cccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-9"
+                 d="m -584.13959,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+                 style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <path
+                 sodipodi:nodetypes="cccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55701-2"
+                 d="m -551.36183,144.95855 h -6.949 v -6.88219 z"
+                 style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686" />
+              <text
+                 id="text55709-2"
+                 y="155.93515"
+                 x="-577.10522"
+                 style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+                 xml:space="preserve"><tspan
+                   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141"
+                   y="155.93515"
+                   x="-577.10522"
+                   id="tspan55707-8"
+                   sodipodi:role="line">XLS</tspan></text>
+              <g
+                 id="g4140-9"
+                 transform="matrix(0.15295911,0,0,0.15295911,-574.53339,160.91154)"
+                 style="stroke-width:1.11137545">
+                <path
+                   d="m 46.04,0 h 5.94 c 0,2.67 0,5.33 0,8 10.01,0 20.02,0.02 30.03,-0.03 1.69,0.07 3.55,-0.05 5.02,0.96 1.03,1.48 0.91,3.36 0.98,5.06 -0.05,17.36 -0.03,34.71 -0.02,52.06 -0.05,2.91 0.27,5.88 -0.34,8.75 -0.4,2.08 -2.9,2.13 -4.57,2.2 -10.36,0.03 -20.73,-0.02 -31.1,0 0,3 0,6 0,9 H 45.77 C 30.53,83.23 15.26,80.67 0,78 0,54.67 0,31.34 0,8.01 15.35,5.34 30.7,2.71 46.04,0 Z"
+                   id="path10-9-7"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 51.98,11 c 11,0 22,0 33,0 0,21 0,42 0,63 -11,0 -22,0 -33,0 0,-2 0,-4 0,-6 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-2 0,-4 0,-6 z"
+                   id="path48-1-3"
+                   inkscape:connector-curvature="0"
+                   style="fill:#ffffff;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,17 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path58-2-6"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 29.62,26.37 c 2.26,-0.16 4.53,-0.3 6.8,-0.41 -2.67,5.47 -5.35,10.94 -8.07,16.39 2.75,5.6 5.56,11.16 8.32,16.76 -2.41,-0.14 -4.81,-0.29 -7.22,-0.46 -1.7,-4.17 -3.77,-8.2 -4.99,-12.56 -1.36,4.06 -3.3,7.89 -4.86,11.87 -2.19,-0.03 -4.38,-0.12 -6.57,-0.21 2.57,-5.03 5.05,-10.1 7.7,-15.1 -2.25,-5.15 -4.72,-10.2 -7.04,-15.32 2.2,-0.13 4.4,-0.26 6.6,-0.38 1.49,3.91 3.12,7.77 4.35,11.78 1.32,-4.25 3.29,-8.25 4.98,-12.36 z"
+                   id="path72-7-1"
+                   inkscape:connector-curvature="0"
+                   style="fill:#ffffff;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,28 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path90-2"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,39 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path108-9"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,50 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path114-3"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,61 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path120-1"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+              </g>
+              <path
+                 sodipodi:nodetypes="ccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-1-9"
+                 d="m -583.90607,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+                 style="fill:#207245;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+            </g>
+          </g>
+          <g
+             transform="translate(32.905868,-67.171919)"
+             id="g12301-4"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.39916864;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -60.99023,30.65149 h 19.73249 l 5.23218,5.13176 v 25.6129 h -24.96467 z"
+               id="rect55679-5-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#6995f4;fill-opacity:1;stroke-width:0.01482968"
+               d="m -48.45626,58.61909 c -0.0716,-0.0731 -0.1361,-0.133 -0.14344,-0.133 -0.008,0 -0.0134,-0.009 -0.0134,-0.02 0,-0.0109 -0.009,-0.02 -0.02,-0.02 -0.011,0 -0.02,-0.009 -0.02,-0.02 0,-0.0109 -0.009,-0.02 -0.02,-0.02 -0.0109,0 -0.02,-0.009 -0.02,-0.02 0,-0.0109 -0.006,-0.02 -0.0134,-0.02 -0.008,0 -0.0134,-0.006 -0.0134,-0.0134 0,-0.008 -0.0331,-0.0461 -0.0733,-0.0862 -0.0403,-0.04 -0.0734,-0.0791 -0.0734,-0.0867 0,-0.008 -0.007,-0.014 -0.0145,-0.014 -0.0215,0 -0.0793,-0.0484 -0.0694,-0.0583 0.005,-0.005 7.7e-4,-0.009 -0.008,-0.009 -0.0162,0 -0.086,-0.0679 -0.1372,-0.13343 -0.0143,-0.0183 -0.0616,-0.0705 -0.10521,-0.11607 -0.0435,-0.0455 -0.0792,-0.0894 -0.0792,-0.0975 0,-0.009 -0.006,-0.0112 -0.0134,-0.007 -0.008,0.005 -0.0134,-7.6e-4 -0.0134,-0.0126 0,-0.0115 -0.009,-0.0208 -0.02,-0.0208 -0.0111,0 -0.02,-0.006 -0.02,-0.0134 0,-0.008 -0.006,-0.0134 -0.0141,-0.0134 -0.0157,0 -0.0926,-0.08 -0.0926,-0.0964 0,-0.006 -0.006,-0.0104 -0.0133,-0.0104 -0.0211,0 -0.0935,-0.0818 -0.0935,-0.10572 0,-0.023 0.11047,-0.13446 0.1333,-0.13446 0.008,0 0.0135,-0.009 0.0135,-0.02 0,-0.0111 0.006,-0.02 0.0128,-0.02 0.007,0 0.0594,-0.045 0.11617,-0.10007 0.0568,-0.0551 0.10848,-0.10008 0.11472,-0.10008 0.0137,0 0.0899,-0.0817 0.0899,-0.0964 0,-0.006 0.006,-0.0104 0.014,-0.0104 0.008,0 0.0292,-0.0164 0.0478,-0.0365 0.0186,-0.02 0.0484,-0.0461 0.0662,-0.0577 0.0178,-0.0117 0.0712,-0.0583 0.11856,-0.10364 0.0474,-0.0453 0.0924,-0.0824 0.0999,-0.0824 0.008,0 0.0138,-0.007 0.0138,-0.0143 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.006,-0.02 0.0134,-0.02 0.008,0 0.0539,-0.0405 0.10352,-0.0901 0.0496,-0.0496 0.12006,-0.11409 0.15669,-0.14344 0.0367,-0.0293 0.0686,-0.0579 0.071,-0.0633 0.002,-0.005 0.0129,-0.01 0.0234,-0.01 0.0105,0 0.0189,-0.006 0.0189,-0.0142 0,-0.014 0.0338,-0.0477 0.1495,-0.14964 0.15388,-0.13549 0.17813,-0.15569 0.18742,-0.15601 0.005,-2e-4 0.01,-0.007 0.01,-0.0147 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.011 0.006,-0.02 0.0142,-0.02 0.008,0 0.0356,-0.0225 0.0617,-0.05 0.0261,-0.0276 0.0527,-0.0485 0.0591,-0.0468 0.006,0.002 0.0117,-0.003 0.0117,-0.0118 0,-0.015 0.0267,-0.0414 0.0974,-0.0961 0.0205,-0.0159 0.0686,-0.0603 0.10676,-0.0986 0.0382,-0.0383 0.0769,-0.0699 0.0861,-0.0701 0.009,-1.5e-4 0.0167,-0.006 0.0167,-0.0136 0,-0.008 0.012,-0.0254 0.0267,-0.0401 0.0147,-0.0147 0.0267,-0.0237 0.0267,-0.02 0,0.003 0.012,-0.005 0.0267,-0.02 0.0147,-0.0147 0.0267,-0.0327 0.0267,-0.04 0,-0.008 0.008,-0.0134 0.0168,-0.0134 0.009,0 0.0331,-0.015 0.0528,-0.0334 0.0198,-0.0183 0.0426,-0.0334 0.0507,-0.0334 0.008,0 0.0114,-0.005 0.008,-0.0119 -0.005,-0.007 0.008,-0.0219 0.0268,-0.0342 0.0455,-0.0298 0.0857,-0.0708 0.0857,-0.0873 0,-0.008 0.007,-0.0134 0.0144,-0.0134 0.008,0 0.05,-0.0361 0.0933,-0.08 0.0433,-0.044 0.0814,-0.08 0.0846,-0.08 0.003,0 0.0232,-0.0165 0.0445,-0.0367 0.0844,-0.0799 0.12162,-0.11009 0.13552,-0.11009 0.008,0 0.0145,-0.007 0.0145,-0.0143 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.006,-0.02 0.013,-0.02 0.007,0 0.0455,-0.0315 0.0852,-0.07 0.0806,-0.0782 0.15505,-0.14345 0.16376,-0.14345 0.003,0 0.0502,-0.045 0.10452,-0.10008 0.0543,-0.0551 0.10416,-0.10007 0.11073,-0.10007 0.007,0 0.0199,-0.015 0.0298,-0.0334 0.01,-0.0183 0.0226,-0.0334 0.0284,-0.0334 0.006,0 0.0374,-0.0255 0.0701,-0.0567 0.0808,-0.0772 0.17035,-0.15678 0.17609,-0.15678 0.002,0 0.0439,-0.039 0.0919,-0.0867 0.048,-0.0477 0.0935,-0.0867 0.10133,-0.0867 0.008,0 0.0105,-0.006 0.006,-0.0134 -0.005,-0.008 -0.002,-0.0134 0.005,-0.0134 0.0157,0 0.0761,-0.0506 0.17152,-0.14344 0.0396,-0.0385 0.0777,-0.0701 0.0848,-0.0701 0.007,0 0.0128,-0.007 0.0128,-0.0143 0,-0.0163 0.0332,-0.0524 0.048,-0.0524 0.0113,0 0.0189,-0.007 0.14815,-0.13125 0.0529,-0.051 0.0994,-0.0892 0.10342,-0.0852 0.005,0.005 0.008,-0.002 0.008,-0.0149 0,-0.0122 0.009,-0.0222 0.02,-0.0222 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0109 0.006,-0.02 0.0134,-0.02 0.008,0 0.0134,-0.009 0.0134,-0.02 0,-0.011 0.008,-0.02 0.0169,-0.02 0.009,0 0.028,-0.015 0.0419,-0.0334 0.0137,-0.0183 0.0306,-0.0334 0.0375,-0.0334 0.017,0 0.0506,-0.0348 0.0506,-0.0524 0,-0.008 0.006,-0.0143 0.0134,-0.0143 0.008,0 0.0243,-0.0105 0.0377,-0.0234 0.0423,-0.0404 0.0956,-0.0834 0.10351,-0.0834 0.0113,0 0.11227,-0.11049 0.11227,-0.12284 0,-0.006 0.009,-0.0106 0.02,-0.0106 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.009,-0.02 0.02,-0.02 0.0109,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0106,-0.0158 0.0236,-0.0199 0.013,-0.005 0.0784,-0.0612 0.14534,-0.12689 0.0669,-0.0657 0.12534,-0.11573 0.1298,-0.11126 0.005,0.005 0.008,-0.002 0.008,-0.0141 0,-0.0122 0.009,-0.0222 0.02,-0.0222 0.0109,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.009,-0.02 0.0208,-0.02 0.0115,0 0.0171,-0.006 0.0126,-0.0134 -0.005,-0.008 7.6e-4,-0.0134 0.0126,-0.0134 0.0115,0 0.0208,-0.009 0.0208,-0.02 0,-0.011 0.005,-0.02 0.0121,-0.02 0.0116,0 0.054,-0.0324 0.10929,-0.0834 0.0139,-0.0128 0.0312,-0.0234 0.0386,-0.0234 0.008,0 0.0134,-0.009 0.0134,-0.02 0,-0.0111 0.006,-0.02 0.0134,-0.02 0.017,0 0.16013,-0.14365 0.16013,-0.16071 0,-0.007 0.007,-0.0127 0.0152,-0.0127 0.009,0 0.0403,-0.0255 0.0708,-0.0567 0.0306,-0.0312 0.0727,-0.0688 0.0937,-0.0834 0.021,-0.0147 0.0703,-0.0582 0.10963,-0.0967 0.0393,-0.0385 0.0739,-0.0701 0.0768,-0.0701 0.0109,0 0.0875,-0.0826 0.0875,-0.0944 0,-0.007 0.007,-0.0123 0.0149,-0.0123 0.009,0 0.039,-0.0241 0.0683,-0.0533 0.0293,-0.0293 0.0583,-0.0533 0.0644,-0.0533 0.006,0 0.0744,-0.0626 0.15188,-0.13903 0.0775,-0.0765 0.14089,-0.13535 0.14089,-0.13084 0,0.005 0.015,-0.0121 0.0334,-0.037 0.0183,-0.0248 0.0334,-0.0411 0.0334,-0.0361 0,0.009 0.0347,-0.0167 0.0739,-0.054 0.01,-0.009 0.024,-0.0167 0.0319,-0.0167 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.007,-0.02 0.0143,-0.02 0.0185,0 0.0524,-0.0339 0.0524,-0.0524 0,-0.008 0.009,-0.0143 0.02,-0.0143 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.005,-0.0185 0.0105,-0.0167 0.006,0.002 0.0355,-0.0207 0.0662,-0.05 0.0823,-0.0786 0.10166,-0.0938 0.11416,-0.0898 0.006,0.002 0.008,-0.002 0.003,-0.009 -0.009,-0.0144 0.049,-0.0552 0.0785,-0.0552 0.0112,0 0.0205,0.006 0.0205,0.0134 0,0.008 0.009,0.0134 0.0205,0.0134 0.0112,0 0.0239,0.009 0.028,0.02 0.005,0.0111 0.0221,0.02 0.0396,0.02 0.0176,0 0.0319,0.006 0.0319,0.0125 0,0.007 0.0135,0.0167 0.03,0.0219 0.0165,0.005 0.0331,0.013 0.0368,0.0176 0.0107,0.0132 0.0581,0.0417 0.0834,0.0502 0.0128,0.005 0.0234,0.0131 0.0234,0.0196 0,0.007 0.0113,0.0118 0.025,0.0118 0.0138,0 0.0325,0.009 0.0417,0.02 0.009,0.0111 0.0272,0.02 0.0403,0.02 0.013,0 0.0271,0.009 0.0313,0.02 0.005,0.0111 0.0189,0.02 0.0327,0.02 0.0286,0 0.10484,0.0355 0.1164,0.0542 0.005,0.007 0.0193,0.0125 0.0337,0.0125 0.0142,0 0.0258,0.006 0.0258,0.0131 0,0.007 0.0151,0.0169 0.0334,0.0215 0.0183,0.005 0.0334,0.0137 0.0334,0.0203 0,0.007 0.0113,0.0119 0.025,0.0119 0.0137,0 0.0325,0.009 0.0417,0.02 0.009,0.0111 0.0279,0.02 0.0417,0.02 0.0138,0 0.025,0.006 0.025,0.0134 0,0.008 0.009,0.0134 0.0191,0.0134 0.0105,0 0.027,0.008 0.0367,0.0167 0.01,0.009 0.0416,0.0287 0.071,0.0434 0.0293,0.0147 0.0554,0.0312 0.0577,0.0367 0.002,0.005 0.0189,0.01 0.0367,0.01 0.0198,0 0.0323,0.008 0.0323,0.02 0,0.0109 0.007,0.02 0.0155,0.02 0.009,0 -0.01,0.024 -0.0406,0.0533 -0.0308,0.0293 -0.0618,0.0533 -0.0689,0.0533 -0.007,0 -0.0128,0.009 -0.0128,0.0205 0,0.0112 -0.009,0.0239 -0.02,0.0281 -0.0111,0.005 -0.02,0.016 -0.02,0.0263 0,0.0102 -0.007,0.0186 -0.0149,0.0186 -0.009,0 -0.0368,0.0251 -0.0633,0.056 -0.0597,0.0689 -0.17284,0.1842 -0.18074,0.1842 -0.006,0 -0.0879,0.0797 -0.0879,0.086 0,0.0121 -0.25327,0.26091 -0.26551,0.26091 -0.008,0 -0.0147,0.006 -0.0147,0.0142 0,0.008 -0.0315,0.0459 -0.0701,0.0846 -0.0385,0.0387 -0.10127,0.10436 -0.13943,0.1459 -0.0382,0.0416 -0.0757,0.0755 -0.0834,0.0755 -0.008,0 -0.0141,0.007 -0.0141,0.0143 0,0.0185 -0.034,0.0524 -0.0524,0.0524 -0.008,0 -0.0143,0.008 -0.0143,0.0162 0,0.0183 -0.1426,0.15729 -0.16133,0.15729 -0.007,0 -0.0122,0.009 -0.0122,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.018,0 -0.026,0.0158 -0.0215,0.0433 7.7e-4,0.005 -0.005,0.01 -0.0105,0.01 -0.007,0 -0.0623,0.0512 -0.12343,0.11381 -0.0612,0.0626 -0.19038,0.19416 -0.28704,0.29238 -0.0967,0.0982 -0.17223,0.1821 -0.16791,0.18641 0.005,0.005 7.6e-4,0.008 -0.007,0.008 -0.0184,0 -0.20301,0.1823 -0.20301,0.20044 0,0.007 -0.007,0.0131 -0.0143,0.0131 -0.0148,0 -0.0524,0.0326 -0.0524,0.0454 0,0.0128 -0.12034,0.12807 -0.13379,0.12807 -0.007,0 -0.013,0.006 -0.013,0.0134 0,0.008 -0.0241,0.0374 -0.0533,0.0667 -0.0293,0.0293 -0.0594,0.0533 -0.0667,0.0533 -0.008,0 -0.0134,0.006 -0.0134,0.0125 0,0.007 -0.0466,0.0601 -0.10342,0.11848 -0.0568,0.0584 -0.14023,0.14437 -0.18525,0.19113 -0.13555,0.1408 -0.24885,0.24535 -0.26166,0.24146 -0.007,-0.002 -0.009,0.002 -0.003,0.01 0.005,0.008 -0.034,0.0554 -0.0858,0.10662 -0.0517,0.0512 -0.0941,0.0994 -0.0941,0.10694 0,0.008 -0.008,0.0141 -0.0167,0.0143 -0.009,2.5e-4 -0.0527,0.04 -0.0967,0.0882 -0.1391,0.15243 -0.21768,0.22954 -0.22573,0.22151 -0.005,-0.005 -0.008,-7.7e-4 -0.008,0.008 0,0.009 -0.0355,0.0511 -0.0789,0.0945 -0.0434,0.0434 -0.0824,0.0754 -0.0867,0.0711 -0.005,-0.005 -0.008,7.6e-4 -0.008,0.0115 0,0.0224 -0.0474,0.0668 -0.0589,0.0552 -0.005,-0.005 -0.008,0.002 -0.008,0.0126 0,0.0112 -0.015,0.0317 -0.0334,0.0454 -0.0183,0.0137 -0.0334,0.0326 -0.0334,0.0419 0,0.009 -0.009,0.0169 -0.02,0.0169 -0.011,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.006,0.02 -0.0134,0.02 -0.0156,0 -0.16011,0.14299 -0.16011,0.15845 0,0.006 -0.015,0.0185 -0.0334,0.0283 -0.0183,0.01 -0.0334,0.0259 -0.0334,0.0356 0,0.01 -0.008,0.0177 -0.0177,0.0177 -0.01,0 -0.0257,0.015 -0.0356,0.0334 -0.01,0.0183 -0.0258,0.0334 -0.0356,0.0334 -0.01,0 -0.0177,0.007 -0.0177,0.0144 0,0.0186 -0.0817,0.10008 -0.0959,0.0957 -0.006,-0.002 -0.0109,0.003 -0.0109,0.011 0,0.0189 -0.0341,0.0524 -0.0533,0.0524 -0.009,0 -0.0113,0.006 -0.007,0.0134 0.005,0.008 -7.6e-4,0.017 -0.0121,0.0212 -0.0112,0.005 -0.024,0.0192 -0.0284,0.0332 -0.005,0.014 -0.0227,0.0291 -0.0405,0.0335 -0.0178,0.005 -0.0325,0.014 -0.0325,0.021 0,0.0173 -0.0347,0.051 -0.0524,0.051 -0.008,0 -0.0143,0.006 -0.0143,0.0142 0,0.0157 -0.0301,0.0497 -0.16559,0.18634 -0.0506,0.0512 -0.0973,0.093 -0.10342,0.093 -0.006,0 -0.0112,0.006 -0.0112,0.0142 0,0.008 -0.0211,0.0341 -0.0468,0.0583 -0.0257,0.0242 -0.0468,0.0421 -0.0468,0.0397 0,-0.002 -0.0271,0.0246 -0.06,0.0598 -0.0331,0.0353 -0.0601,0.0685 -0.0601,0.0738 0,0.013 -0.0355,0.0477 -0.049,0.0478 -0.006,5e-5 -0.024,0.0166 -0.0399,0.0367 -0.0481,0.0609 -0.13316,0.14999 -0.1431,0.14999 -0.0147,0 -0.0748,0.0639 -0.0748,0.0795 0,0.008 -0.009,0.014 -0.02,0.014 -0.0109,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.009,0.02 -0.0208,0.02 -0.0115,0 -0.0177,0.005 -0.0141,0.011 0.008,0.0122 -0.0382,0.0557 -0.0588,0.0557 -0.007,0 -0.0132,0.005 -0.0132,0.0121 0,0.017 -0.20964,0.22804 -0.22646,0.22804 -0.008,0 -0.0137,0.006 -0.0137,0.0142 0,0.008 -0.0451,0.0587 -0.10008,0.11299 -0.0551,0.0544 -0.10008,0.10498 -0.10008,0.11256 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.005 -0.02,0.0105 0,0.0115 -0.0503,0.0642 -0.16561,0.17358 -0.041,0.0389 -0.0746,0.0743 -0.0746,0.0786 0,0.0131 -0.0788,0.0842 -0.0933,0.0842 -0.008,0 -0.0135,0.008 -0.0135,0.0167 -2e-5,0.009 -0.0361,0.0547 -0.0801,0.10128 -0.044,0.0465 -0.08,0.0818 -0.08,0.0783 0,-0.003 -0.051,0.0448 -0.11343,0.10739 -0.0624,0.0625 -0.11342,0.11572 -0.11342,0.11822 0,0.0121 -0.18633,0.19192 -0.19895,0.19192 -0.008,0 -0.0145,0.007 -0.0145,0.0151 0,0.009 -0.0211,0.0349 -0.0467,0.0591 -0.0257,0.0242 -0.0468,0.0504 -0.0468,0.0583 0,0.008 -0.009,0.0142 -0.02,0.0142 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.0396 -0.0608,-0.002 -0.1768,-0.12053 z m -1.54346,-1.58412 c -5.2e-4,-0.009 -0.0101,-0.0167 -0.0211,-0.0167 -0.011,0 -0.02,-0.009 -0.02,-0.02 0,-0.0111 -0.009,-0.02 -0.02,-0.02 -0.0111,0 -0.02,-0.005 -0.02,-0.011 0,-0.0142 -0.0729,-0.0958 -0.0855,-0.0958 -0.0148,0 -0.0478,-0.0362 -0.0478,-0.0524 0,-0.008 -0.006,-0.0143 -0.0139,-0.0143 -0.019,0 -0.06,-0.0412 -0.0708,-0.0713 -0.005,-0.0135 -0.0208,-0.0277 -0.0355,-0.0316 -0.0147,-0.003 -0.0267,-0.0111 -0.0267,-0.0161 0,-0.0131 -0.11633,-0.14283 -0.18547,-0.20691 -0.10595,-0.0982 -0.23492,-0.23707 -0.23152,-0.24927 0.002,-0.007 -0.002,-0.012 -0.0105,-0.012 -0.014,0 -0.093,-0.0749 -0.093,-0.0882 0,-0.0127 -0.0377,-0.0453 -0.0524,-0.0453 -0.008,0 -0.0143,-0.009 -0.0143,-0.0186 0,-0.0102 -0.0151,-0.0223 -0.0334,-0.027 -0.0183,-0.005 -0.0334,-0.0144 -0.0334,-0.0219 0,-0.008 -0.019,-0.0321 -0.0421,-0.0547 -0.0369,-0.0361 -0.0394,-0.0427 -0.02,-0.0536 0.0121,-0.007 0.0446,-0.0331 0.0722,-0.0584 0.0276,-0.0253 0.061,-0.0461 0.0743,-0.0461 0.0133,-5e-5 0.021,-0.005 0.0171,-0.0114 -0.003,-0.006 0.0149,-0.0287 0.0416,-0.05 0.0267,-0.0213 0.0561,-0.0462 0.0653,-0.0554 0.009,-0.009 0.0204,-0.0167 0.0249,-0.0167 0.0102,0 0.0419,-0.0257 0.0714,-0.0578 0.0123,-0.0134 0.0569,-0.0488 0.0991,-0.0787 0.0422,-0.0298 0.0767,-0.0594 0.0767,-0.0657 0,-0.006 0.006,-0.0114 0.0141,-0.0114 0.008,0 0.0356,-0.0211 0.0621,-0.0468 0.0264,-0.0257 0.0526,-0.0468 0.0581,-0.0468 0.005,0 0.0377,-0.0268 0.0713,-0.0595 0.0338,-0.0328 0.0613,-0.0558 0.0613,-0.051 0,0.005 0.015,-0.006 0.0334,-0.0228 0.0183,-0.0172 0.0334,-0.0277 0.0334,-0.0234 0,0.005 0.0253,-0.0156 0.0562,-0.0444 0.0308,-0.0288 0.0609,-0.0523 0.0667,-0.0523 0.006,0 0.0346,-0.0241 0.064,-0.0533 0.0293,-0.0293 0.0564,-0.0533 0.0602,-0.0533 0.006,0 0.0676,-0.0513 0.10754,-0.0901 0.009,-0.009 0.0236,-0.0167 0.0315,-0.0167 0.008,0 0.0143,-0.006 0.0143,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.008,-0.02 0.0168,-0.02 0.009,0 0.0329,-0.015 0.0528,-0.0334 0.0198,-0.0183 0.0422,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0111 0.006,-0.02 0.0139,-0.02 0.008,0 0.0389,-0.0241 0.0694,-0.0533 0.0306,-0.0293 0.0603,-0.0533 0.0661,-0.0533 0.006,0 0.0241,-0.0116 0.0406,-0.0259 0.0918,-0.0789 0.11089,-0.0942 0.11796,-0.0942 0.005,-10e-6 0.0339,-0.0241 0.0656,-0.0534 0.0318,-0.0293 0.0623,-0.0533 0.0679,-0.0533 0.005,0 0.0281,-0.0179 0.0502,-0.04 0.022,-0.022 0.0449,-0.0402 0.051,-0.0403 0.006,-1.4e-4 0.0253,-0.0152 0.0426,-0.0335 0.0173,-0.0183 0.0489,-0.0435 0.0701,-0.056 0.0212,-0.0125 0.0384,-0.0305 0.0384,-0.0399 0,-0.009 0.009,-0.0171 0.02,-0.0171 0.0111,0 0.02,-0.005 0.02,-0.0109 0,-0.006 0.0243,-0.0246 0.054,-0.0412 0.0297,-0.0167 0.0612,-0.0419 0.07,-0.0559 0.009,-0.014 0.0194,-0.0254 0.0238,-0.0254 0.005,0 0.0339,-0.0241 0.0658,-0.0533 0.0318,-0.0293 0.0652,-0.0533 0.0743,-0.0533 0.009,0 0.0203,-0.01 0.0248,-0.0219 0.005,-0.0121 0.0118,-0.0185 0.016,-0.0144 0.007,0.007 0.0423,-0.0224 0.14025,-0.11385 0.0137,-0.0128 0.031,-0.0234 0.0384,-0.0234 0.008,0 0.0134,-0.006 0.0134,-0.0134 0,-0.008 0.005,-0.0134 0.0123,-0.0134 0.007,0 0.0304,-0.0165 0.0526,-0.0367 0.0941,-0.0858 0.12369,-0.11009 0.13441,-0.11009 0.006,0 0.0148,-0.009 0.019,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.006,-0.0134 0.0124,-0.0134 0.007,0 0.0347,-0.0211 0.0619,-0.0468 0.0272,-0.0257 0.0542,-0.0468 0.0599,-0.0468 0.006,0 0.0374,-0.027 0.0704,-0.06 0.0331,-0.0331 0.0643,-0.0601 0.0696,-0.0601 0.005,0 0.0235,-0.015 0.0404,-0.0334 0.017,-0.0183 0.0358,-0.0334 0.0417,-0.0334 0.006,0 0.027,-0.015 0.0468,-0.0334 0.0198,-0.0183 0.0422,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0116 0.0121,-0.02 0.029,-0.02 0.0159,0 0.0251,-0.003 0.0205,-0.009 -0.008,-0.008 0.0462,-0.0583 0.0629,-0.0583 0.005,0 0.0319,-0.0241 0.0613,-0.0533 0.0293,-0.0293 0.0627,-0.0533 0.0742,-0.0533 0.0115,0 0.0171,-0.006 0.0126,-0.0134 -0.005,-0.008 7.6e-4,-0.0134 0.0126,-0.0134 0.0115,0 0.0208,-0.005 0.0208,-0.0111 0,-0.006 0.0328,-0.0346 0.0731,-0.0633 0.0403,-0.0289 0.0732,-0.0568 0.0734,-0.0624 1.5e-4,-0.005 0.007,-0.01 0.0143,-0.01 0.008,0 0.0279,-0.0151 0.0448,-0.0334 0.017,-0.0183 0.0438,-0.0334 0.0594,-0.0334 0.0157,0 0.027,-0.005 0.0251,-0.0105 -0.002,-0.006 0.0163,-0.0298 0.0405,-0.0533 0.046,-0.045 0.0963,-0.057 0.0963,-0.0229 0,0.0111 0.009,0.02 0.02,0.02 0.0111,0 0.02,0.005 0.02,0.0119 0,0.007 0.0105,0.0154 0.0234,0.0196 0.0267,0.009 0.0672,0.0343 0.0938,0.0587 0.01,0.009 0.0222,0.0167 0.0272,0.0167 0.005,1e-5 0.0265,0.015 0.0479,0.0334 0.0213,0.0183 0.0441,0.0333 0.0506,0.0333 0.007,0 0.0258,0.015 0.0428,0.0334 0.017,0.0183 0.0352,0.0334 0.0403,0.0334 0.005,0 0.0178,0.008 0.028,0.0167 0.0463,0.0415 0.0598,0.05 0.079,0.05 0.0113,0 0.0206,0.009 0.0206,0.02 0,0.0111 0.0114,0.02 0.0253,0.02 0.0139,0 0.0287,0.009 0.033,0.02 0.005,0.0111 0.0143,0.02 0.0225,0.02 0.0192,0 0.0528,0.0339 0.0528,0.0532 0,0.009 0.005,0.0118 0.0119,0.008 0.007,-0.005 0.0267,0.005 0.0449,0.0188 0.0183,0.0144 0.0369,0.0222 0.0416,0.0177 0.005,-0.005 0.009,0.002 0.009,0.0138 0,0.0122 0.007,0.0222 0.0143,0.0222 0.008,0 0.0227,0.008 0.0328,0.0167 0.0393,0.0352 0.0587,0.05 0.0656,0.05 0.007,0 0.0825,0.0551 0.16703,0.12109 0.0182,0.0141 0.0377,0.0257 0.0434,0.0257 0.006,0 -0.0201,0.03 -0.0576,0.0667 -0.0375,0.0367 -0.0736,0.0667 -0.0801,0.0667 -0.007,0 -0.012,0.006 -0.012,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.0147,0 -0.0859,0.0613 -0.18222,0.1568 -0.0315,0.0312 -0.0604,0.0567 -0.0644,0.0567 -0.005,0 -0.0362,0.03 -0.0718,0.0667 -0.0355,0.0367 -0.0714,0.0667 -0.0796,0.0667 -0.009,0 -0.015,0.009 -0.015,0.0208 0,0.0115 -0.005,0.0177 -0.0111,0.014 -0.006,-0.003 -0.0305,0.0109 -0.0542,0.0326 -0.0237,0.0216 -0.0488,0.0393 -0.0556,0.0393 -0.007,0 -0.0125,0.005 -0.0125,0.0123 0,0.0129 -0.077,0.0944 -0.0891,0.0944 -0.006,0 -0.0482,0.0377 -0.15182,0.13677 -0.0211,0.0203 -0.0419,0.0367 -0.0461,0.0367 -0.005,0 -0.021,0.0137 -0.0371,0.0305 -0.0162,0.0168 -0.0564,0.0511 -0.0894,0.0763 -0.0331,0.0251 -0.0791,0.0655 -0.10251,0.0896 -0.0234,0.0241 -0.0489,0.0438 -0.0567,0.0438 -0.008,0 -0.0142,0.009 -0.0142,0.02 0,0.0111 -0.008,0.02 -0.0177,0.02 -0.01,0 -0.0258,0.015 -0.0356,0.0334 -0.01,0.0183 -0.0236,0.0334 -0.0307,0.0334 -0.007,0 -0.0268,0.015 -0.0439,0.0334 -0.017,0.0183 -0.0376,0.0334 -0.0457,0.0334 -0.008,0 -0.0118,0.005 -0.008,0.0109 0.003,0.006 -0.0118,0.0219 -0.0343,0.0351 -0.0226,0.0134 -0.0636,0.0474 -0.0911,0.0759 -0.0276,0.0283 -0.0568,0.0516 -0.065,0.0516 -0.009,0 -0.0149,0.009 -0.0149,0.02 0,0.011 -0.005,0.02 -0.01,0.02 -0.0126,4e-5 -0.11046,0.0821 -0.18695,0.15674 -0.032,0.0312 -0.0636,0.0567 -0.0704,0.0567 -0.007,0 -0.0221,0.015 -0.034,0.0332 -0.012,0.0183 -0.0271,0.03 -0.0337,0.0259 -0.007,-0.005 -0.012,5.2e-4 -0.0122,0.0101 -1.5e-4,0.01 -0.03,0.0385 -0.0665,0.0642 -0.0364,0.0257 -0.0662,0.0512 -0.0665,0.0567 -1.5e-4,0.005 -0.009,0.01 -0.0203,0.01 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.007,0.02 -0.0143,0.02 -0.008,0 -0.0229,0.009 -0.0334,0.019 -0.0105,0.0105 -0.019,0.0255 -0.019,0.0334 0,0.008 -0.006,0.0143 -0.0138,0.0143 -0.008,0 -0.0305,0.0165 -0.0508,0.0367 -0.0463,0.0459 -0.14661,0.12862 -0.16316,0.13454 -0.007,0.002 -0.0125,0.0129 -0.0125,0.0234 0,0.0105 -0.009,0.0189 -0.02,0.0189 -0.0109,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0109,0 -0.02,0.008 -0.02,0.0177 0,0.01 -0.015,0.0257 -0.0334,0.0356 -0.0183,0.01 -0.0334,0.0262 -0.0334,0.0364 0,0.0102 -0.005,0.0155 -0.0109,0.0119 -0.006,-0.003 -0.0388,0.0199 -0.0729,0.0526 -0.0341,0.0326 -0.0671,0.0594 -0.0736,0.0594 -0.006,0 -0.0196,0.015 -0.0294,0.0334 -0.01,0.0183 -0.0231,0.0334 -0.0296,0.0334 -0.007,0 -0.0249,0.0137 -0.0411,0.0305 -0.0162,0.0168 -0.0562,0.0513 -0.0891,0.0767 -0.0328,0.0254 -0.0799,0.0658 -0.10455,0.0896 -0.0247,0.0239 -0.0503,0.0433 -0.057,0.0433 -0.007,0 -0.0122,0.009 -0.0122,0.02 0,0.011 -0.006,0.02 -0.0134,0.02 -0.008,0 -0.0389,0.024 -0.0701,0.0532 -0.0312,0.0292 -0.073,0.0682 -0.0929,0.0867 -0.0199,0.0185 -0.0424,0.0335 -0.05,0.0335 -0.008,0 -0.0139,0.009 -0.0139,0.02 0,0.0111 -0.008,0.02 -0.0169,0.02 -0.009,0 -0.0281,0.0151 -0.0419,0.0334 -0.0138,0.0183 -0.0301,0.0334 -0.0363,0.0334 -0.0116,0 -0.055,0.0364 -0.14677,0.12343 -0.029,0.0276 -0.0589,0.05 -0.0665,0.05 -0.008,0 -0.01,0.006 -0.005,0.0134 0.005,0.008 -7.7e-4,0.0134 -0.0126,0.0134 -0.0115,0 -0.0208,0.009 -0.0208,0.02 0,0.0111 -0.006,0.02 -0.0129,0.02 -0.007,0 -0.0439,0.03 -0.0818,0.0667 -0.0378,0.0368 -0.074,0.0667 -0.0805,0.0667 -0.006,0 -0.0449,0.0331 -0.0856,0.0734 -0.0406,0.0403 -0.0792,0.0733 -0.0856,0.0733 -0.006,0 -0.0147,0.008 -0.0184,0.0167 -0.005,0.0134 -0.007,0.0134 -0.008,0 z m -1.50221,-1.53083 c -0.022,-0.0251 -0.0475,-0.0461 -0.0567,-0.0464 -0.009,-2.6e-4 -0.0167,-0.01 -0.0167,-0.0206 0,-0.0111 -0.009,-0.02 -0.02,-0.02 -0.0111,0 -0.02,-0.007 -0.02,-0.0149 0,-0.009 -0.0751,-0.0894 -0.1668,-0.18044 -0.0917,-0.0911 -0.16679,-0.17146 -0.16679,-0.17862 0,-0.007 -0.008,-0.0131 -0.0167,-0.0132 -0.009,-5e-5 -0.0377,-0.0247 -0.0633,-0.0548 -0.0428,-0.0502 -0.15564,-0.1732 -0.18819,-0.20522 -0.12475,-0.12271 -0.19877,-0.19935 -0.19877,-0.20581 0,-0.009 0.043,-0.0335 0.0834,-0.0487 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.006,-0.0118 0.014,-0.0118 0.008,0 0.0302,-0.0152 0.05,-0.0338 0.0199,-0.0186 0.0661,-0.053 0.10283,-0.0765 0.0368,-0.0235 0.0688,-0.0473 0.0711,-0.0529 0.002,-0.006 0.0123,-0.0103 0.0219,-0.0103 0.01,0 0.021,-0.009 0.0252,-0.02 0.005,-0.0109 0.0154,-0.02 0.0249,-0.02 0.009,0 0.0334,-0.015 0.0532,-0.0334 0.0198,-0.0183 0.0423,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.0208 0,-0.0115 0.005,-0.0176 0.0117,-0.0135 0.007,0.005 0.032,-0.0105 0.0567,-0.0322 0.0248,-0.0216 0.0645,-0.05 0.0883,-0.0629 0.0239,-0.0129 0.0434,-0.0281 0.0434,-0.0338 0,-0.006 0.009,-0.0102 0.0189,-0.0102 0.0105,0 0.0209,-0.005 0.0234,-0.0101 0.006,-0.0135 0.11264,-0.0967 0.12407,-0.0967 0.005,0 0.0174,-0.008 0.0277,-0.0167 0.0442,-0.0396 0.0593,-0.05 0.0724,-0.05 0.008,0 0.014,-0.006 0.014,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0137 0,-0.008 0.015,-0.0193 0.0334,-0.0264 0.0183,-0.007 0.0334,-0.0188 0.0334,-0.0264 0,-0.008 0.009,-0.0137 0.02,-0.0137 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.0204,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.013,-0.02 0.0194,-0.02 0.0112,0 0.0473,-0.0267 0.086,-0.0633 0.01,-0.009 0.024,-0.0167 0.0319,-0.0167 0.008,0 0.0143,-0.006 0.0143,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.006,-0.0119 0.0139,-0.0119 0.0131,0 0.0282,-0.0105 0.0724,-0.05 0.0102,-0.009 0.025,-0.0167 0.0328,-0.0167 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.011 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0131,-0.0166 0.029,-0.0216 0.0161,-0.005 0.0445,-0.0235 0.0633,-0.041 0.0189,-0.0175 0.0418,-0.0317 0.051,-0.0317 0.009,0 0.0169,-0.005 0.0169,-0.01 0,-0.005 0.0315,-0.031 0.07,-0.0567 0.0384,-0.0257 0.0849,-0.0602 0.10325,-0.0767 0.0387,-0.0348 0.0646,-0.0383 0.0755,-0.01 0.005,0.0111 0.0134,0.02 0.0205,0.02 0.0173,0 0.051,0.0346 0.051,0.0524 0,0.008 0.007,0.0143 0.0144,0.0143 0.008,0 0.0324,0.018 0.0544,0.0401 0.022,0.022 0.0453,0.04 0.0517,0.04 0.007,0 0.0574,0.045 0.11343,0.10008 0.0559,0.0551 0.10637,0.10008 0.11206,0.10008 0.006,0 0.0265,0.015 0.0463,0.0334 0.0198,0.0183 0.0417,0.0334 0.0487,0.0334 0.007,0 0.0127,0.009 0.0127,0.02 0,0.0111 0.006,0.02 0.0139,0.02 0.0149,0 0.0316,0.0125 0.0941,0.0701 0.0219,0.0202 0.0433,0.0367 0.0477,0.0367 0.005,0 0.0278,0.021 0.052,0.0468 0.0242,0.0257 0.0504,0.0468 0.0583,0.0468 0.008,0 0.0142,0.009 0.0142,0.02 0,0.011 0.006,0.02 0.014,0.02 0.008,0 0.0279,0.0135 0.0449,0.03 0.017,0.0165 0.0422,0.0367 0.056,0.0448 0.0215,0.0127 0.0225,0.0178 0.007,0.0367 -0.01,0.0121 -0.0264,0.0219 -0.0367,0.0219 -0.0101,0 -0.0184,0.009 -0.0184,0.02 0,0.0111 -0.005,0.02 -0.0119,0.02 -0.007,0 -0.0261,0.0149 -0.0434,0.0332 -0.0173,0.0183 -0.0377,0.0332 -0.0453,0.0334 -0.008,1.5e-4 -0.0362,0.0212 -0.0635,0.0469 -0.0272,0.0257 -0.054,0.0468 -0.0595,0.0468 -0.005,5e-5 -0.0292,0.0181 -0.0526,0.04 -0.0235,0.022 -0.049,0.04 -0.0567,0.04 -0.008,0 -0.0141,0.009 -0.0141,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.007,0.0134 -0.0143,0.0134 -0.008,0 -0.0227,0.008 -0.0329,0.0167 -0.04,0.0359 -0.0587,0.05 -0.0662,0.05 -0.009,0 -0.0213,0.01 -0.0764,0.0575 -0.0199,0.0174 -0.052,0.0431 -0.0711,0.0571 -0.0348,0.0255 -0.0786,0.0598 -0.1088,0.0853 -0.009,0.008 -0.0223,0.0136 -0.03,0.0136 -0.008,0 -0.014,0.009 -0.014,0.02 0,0.0111 -0.009,0.02 -0.0189,0.02 -0.0105,0 -0.0208,0.005 -0.0234,0.01 -0.008,0.0165 -0.11042,0.0902 -0.12121,0.0867 -0.005,-0.002 -0.01,0.006 -0.01,0.0167 0,0.0111 -0.008,0.02 -0.0169,0.02 -0.009,0 -0.0281,0.0151 -0.0419,0.0334 -0.0138,0.0183 -0.0334,0.0334 -0.0436,0.0334 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0194,0.02 -0.0338,0.02 -0.0143,0 -0.0223,0.006 -0.0178,0.0134 0.005,0.008 9e-5,0.0134 -0.01,0.0134 -0.01,0 -0.0432,0.0241 -0.0736,0.0533 -0.0306,0.0293 -0.0618,0.0533 -0.0694,0.0533 -0.008,0 -0.0138,0.005 -0.0138,0.0111 0,0.006 -0.0285,0.0315 -0.0633,0.0565 -0.0348,0.0249 -0.0754,0.0581 -0.0901,0.0738 -0.0147,0.0157 -0.0402,0.0326 -0.0567,0.0378 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.006,0.0125 -0.014,0.0125 -0.008,0 -0.0301,0.015 -0.0499,0.0334 -0.0198,0.0183 -0.0469,0.0334 -0.0603,0.0334 -0.0134,0 -0.0212,0.005 -0.0175,0.0109 0.008,0.0122 -0.0382,0.0558 -0.0587,0.0558 -0.007,0 -0.0132,0.008 -0.0132,0.0171 0,0.009 -0.0191,0.0284 -0.0425,0.0422 -0.0397,0.0234 -0.0852,0.0588 -0.12519,0.0974 -0.0226,0.0219 -0.0198,0.023 -0.0659,-0.0297 z m -1.31892,-1.34259 c -0.0415,-0.0434 -0.0754,-0.0814 -0.0754,-0.0845 0,-0.0111 -0.0824,-0.0877 -0.0944,-0.0877 -0.007,0 -0.0123,-0.009 -0.0123,-0.02 0,-0.0111 -0.007,-0.02 -0.0143,-0.02 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.007,-0.0143 -0.0143,-0.0143 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.009,-0.0143 -0.02,-0.0143 -0.0111,0 -0.02,-0.009 -0.02,-0.02 0,-0.0111 -0.006,-0.02 -0.0138,-0.02 -0.014,0 -0.093,-0.075 -0.093,-0.0881 0,-0.0128 -0.0377,-0.0453 -0.0524,-0.0453 -0.008,0 -0.0143,-0.007 -0.0143,-0.0154 0,-0.017 -0.10466,-0.11804 -0.12232,-0.11804 -0.006,0 -0.0111,-0.007 -0.0111,-0.0144 0,-0.008 -0.0219,-0.0357 -0.0487,-0.0618 l -0.0487,-0.0474 0.0321,-0.0247 c 0.0177,-0.0135 0.039,-0.0248 0.0474,-0.0249 0.009,-1.5e-4 0.0187,-0.009 0.0229,-0.0203 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.012,-0.0134 0.0267,-0.0134 0.0148,0 0.0267,-0.009 0.0267,-0.02 0,-0.0111 0.009,-0.02 0.0184,-0.02 0.0101,0 0.0254,-0.009 0.034,-0.0187 0.009,-0.0104 0.0363,-0.0286 0.0616,-0.0406 0.0254,-0.0121 0.0461,-0.0276 0.0461,-0.0346 0,-0.007 0.0144,-0.0128 0.0319,-0.0128 0.0176,0 0.0355,-0.009 0.0399,-0.0209 0.005,-0.0115 0.0141,-0.0171 0.0214,-0.0126 0.008,0.005 0.0135,-7.6e-4 0.0135,-0.0125 0,-0.0115 0.009,-0.0208 0.02,-0.0208 0.0111,0 0.02,-0.005 0.02,-0.0108 0,-0.006 0.0241,-0.0214 0.0534,-0.0343 0.0293,-0.0129 0.0609,-0.0306 0.0701,-0.0393 0.009,-0.009 0.0391,-0.0276 0.0667,-0.0421 0.0275,-0.0144 0.05,-0.031 0.05,-0.0367 0,-0.006 0.009,-0.0102 0.02,-0.0102 0.0111,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0135,-0.0167 0.03,-0.0219 0.0165,-0.005 0.0432,-0.0236 0.0594,-0.0409 0.0162,-0.0173 0.0389,-0.0315 0.0504,-0.0315 0.0116,0 0.0243,-0.009 0.0283,-0.0189 0.005,-0.0104 0.0346,-0.0269 0.068,-0.0368 0.0334,-0.01 0.0636,-0.0254 0.0672,-0.0345 0.003,-0.009 0.0126,-0.0167 0.0201,-0.0167 0.008,0 0.0273,-0.0147 0.0439,-0.0326 0.0166,-0.0179 0.0328,-0.0299 0.0361,-0.0266 0.003,0.003 0.0169,-0.006 0.0302,-0.0208 0.0134,-0.0147 0.0322,-0.0268 0.0422,-0.0268 0.01,0 0.0179,-0.005 0.0179,-0.0102 0,-0.0102 0.0396,-0.0343 0.0834,-0.0509 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.009,-0.0118 0.0205,-0.0118 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0221,-0.02 0.0397,-0.02 0.0176,0 0.0319,-0.007 0.0319,-0.0152 0,-0.0205 0.0412,-0.0505 0.0701,-0.0511 0.0128,-2.6e-4 0.0234,-0.009 0.0234,-0.0205 0,-0.0111 0.0119,-0.02 0.0267,-0.02 0.0147,0 0.0267,-0.006 0.0267,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0109 0.0184,-0.02 0.0315,-0.02 0.0131,0 0.0273,-0.009 0.0315,-0.02 0.005,-0.0111 0.0141,-0.02 0.0218,-0.02 0.008,0 0.0432,-0.0196 0.0788,-0.0434 0.0356,-0.0239 0.0847,-0.0554 0.10912,-0.0701 0.0244,-0.0147 0.0552,-0.0332 0.0683,-0.0412 0.0132,-0.008 0.036,-0.0254 0.0507,-0.0388 0.0147,-0.0134 0.0268,-0.0196 0.0268,-0.0137 0,0.006 0.0159,-0.005 0.0354,-0.0249 0.0196,-0.0196 0.0495,-0.039 0.0667,-0.0433 0.0172,-0.005 0.0312,-0.0132 0.0312,-0.0198 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0109,0 0.02,-0.005 0.02,-0.0121 0,-0.007 0.0165,-0.0179 0.0368,-0.0253 0.0202,-0.007 0.0499,-0.0258 0.0662,-0.0413 0.0162,-0.0155 0.0381,-0.0282 0.0486,-0.0282 0.0105,0 0.0226,-0.009 0.0268,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0129,-0.0165 0.0288,-0.0215 0.0158,-0.005 0.042,-0.0235 0.0582,-0.0409 0.0162,-0.0175 0.0433,-0.0318 0.0602,-0.0318 0.0169,0 0.0267,-0.005 0.0219,-0.009 -0.008,-0.008 0.0192,-0.0283 0.0711,-0.0531 0.0111,-0.005 0.0248,-0.0175 0.0306,-0.0272 0.006,-0.01 0.0218,-0.0177 0.0353,-0.0177 0.0135,0 0.0281,-0.009 0.0322,-0.02 0.005,-0.011 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.005 0.0204,-0.0118 0,-0.007 0.0105,-0.0156 0.0234,-0.0203 0.0327,-0.0118 0.0797,-0.0403 0.11527,-0.0697 0.0432,-0.0357 0.078,-0.058 0.10491,-0.067 0.0128,-0.005 0.0234,-0.0124 0.0234,-0.018 0,-0.006 0.0105,-0.0138 0.0234,-0.0181 0.0313,-0.0105 0.0651,-0.0326 0.12631,-0.0827 0.009,-0.008 0.0261,-0.0183 0.0371,-0.0234 0.0111,-0.005 0.038,-0.021 0.0601,-0.035 0.022,-0.0141 0.0539,-0.026 0.0708,-0.0265 0.0171,-5.1e-4 0.0271,-0.007 0.0225,-0.0144 -0.005,-0.008 0.002,-0.0176 0.0159,-0.0226 0.0133,-0.005 0.0315,-0.0155 0.0404,-0.0233 0.0468,-0.0406 0.0578,-0.0474 0.0766,-0.0474 0.0113,0 0.0206,-0.005 0.0206,-0.0106 0,-0.006 0.0239,-0.0254 0.053,-0.0434 0.0292,-0.018 0.0532,-0.0371 0.0533,-0.0427 2.1e-4,-0.005 0.0118,-0.01 0.0256,-0.01 0.0138,0 0.0283,-0.008 0.0321,-0.0179 0.003,-0.01 0.0284,-0.0282 0.0548,-0.0406 0.0264,-0.0125 0.048,-0.0285 0.048,-0.0354 0,-0.007 0.009,-0.0127 0.02,-0.0127 0.0111,0 0.02,-0.005 0.02,-0.0102 0,-0.0102 0.0396,-0.0343 0.0834,-0.0509 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.008,-0.0118 0.018,-0.0118 0.01,0 0.0354,-0.0151 0.0567,-0.0335 0.0213,-0.0184 0.0503,-0.0371 0.0645,-0.0417 0.0141,-0.005 0.0411,-0.0225 0.0599,-0.0399 0.0189,-0.0175 0.0418,-0.0317 0.051,-0.0317 0.009,0 0.0168,-0.005 0.0168,-0.0109 0,-0.006 0.027,-0.0245 0.06,-0.041 0.0331,-0.0165 0.06,-0.0342 0.06,-0.0392 0,-0.009 0.0109,-0.0163 0.10117,-0.0623 0.0287,-0.0147 0.0544,-0.0313 0.0568,-0.037 0.002,-0.006 0.028,-0.0223 0.0567,-0.037 0.0286,-0.0147 0.0521,-0.0325 0.0521,-0.0396 0,-0.007 0.0144,-0.0131 0.0319,-0.0131 0.0176,0 0.0354,-0.009 0.0397,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.01,-0.0134 0.0219,-0.0134 0.012,0 0.0357,-0.015 0.0527,-0.0334 0.017,-0.0183 0.0404,-0.0334 0.052,-0.0334 0.0115,0 0.0244,-0.009 0.0287,-0.02 0.005,-0.0111 0.016,-0.02 0.0263,-0.02 0.0102,0 0.0186,-0.006 0.0186,-0.0124 0,-0.007 0.0135,-0.0167 0.03,-0.0219 0.0165,-0.005 0.0331,-0.0134 0.0367,-0.0182 0.0108,-0.0144 0.17561,-0.12104 0.1869,-0.12104 0.006,0 0.0139,-0.009 0.018,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.009,-0.0134 0.021,-0.0134 0.0115,0 0.0299,-0.009 0.041,-0.02 0.011,-0.0111 0.0306,-0.02 0.0435,-0.02 0.0129,0 0.0307,-0.0135 0.0396,-0.03 0.009,-0.0165 0.0399,-0.0415 0.0689,-0.0555 0.029,-0.0141 0.0528,-0.0306 0.0528,-0.0368 0,-0.0247 0.0756,-0.009 0.10964,0.0222 0.0198,0.0183 0.0455,0.0334 0.057,0.0334 0.0115,0 0.0245,0.009 0.0286,0.02 0.005,0.011 0.016,0.02 0.0263,0.02 0.0102,0 0.0186,0.006 0.0186,0.0134 0,0.008 0.0144,0.0134 0.0319,0.0134 0.0176,0 0.0354,0.009 0.0397,0.02 0.005,0.0109 0.0169,0.02 0.028,0.02 0.0112,0 0.0204,0.006 0.0204,0.0131 0,0.007 0.015,0.0169 0.0333,0.0214 0.0183,0.005 0.0299,0.0137 0.026,0.0203 -0.005,0.007 -1.6e-4,0.0119 0.009,0.0119 0.009,0 0.054,0.0241 0.10054,0.0533 0.0465,0.0293 0.0902,0.0533 0.0971,0.0533 0.007,0 0.0145,0.005 0.017,0.01 0.007,0.0145 0.12964,0.0967 0.14494,0.0967 0.007,0 0.0129,0.006 0.0129,0.0134 0,0.008 0.0122,0.0134 0.0271,0.0134 0.0149,0 0.0306,0.009 0.0348,0.02 0.005,0.011 0.016,0.02 0.0263,0.02 0.0102,0 0.0186,0.006 0.0186,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0124 0,0.007 -0.0131,0.0166 -0.029,0.0217 -0.016,0.005 -0.0445,0.0234 -0.0633,0.041 -0.0189,0.0175 -0.039,0.0317 -0.0449,0.0317 -0.006,0 -0.0248,0.0151 -0.0417,0.0334 -0.017,0.0183 -0.0377,0.0334 -0.0461,0.0334 -0.009,0 -0.0151,0.006 -0.0151,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.009 -0.02,0.0184 0,0.0102 -0.0105,0.022 -0.0234,0.0263 -0.025,0.009 -0.0724,0.0368 -0.0834,0.05 -0.003,0.005 -0.0324,0.0241 -0.0639,0.0439 -0.0314,0.0197 -0.0692,0.0478 -0.0839,0.0625 -0.0147,0.0147 -0.0325,0.0231 -0.0396,0.0187 -0.007,-0.005 -0.0128,0.002 -0.0128,0.0128 0,0.0115 -0.009,0.0208 -0.02,0.0208 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.0134,0.0134 -0.008,0 -0.0425,0.0209 -0.078,0.0464 -0.0356,0.0256 -0.0687,0.0426 -0.0734,0.0378 -0.005,-0.005 -0.009,-0.002 -0.009,0.007 0,0.009 -0.006,0.0156 -0.014,0.0156 -0.0154,0 -0.0366,0.0155 -0.0953,0.0701 -0.0218,0.0202 -0.0448,0.0367 -0.0512,0.0367 -0.006,0 -0.0164,0.008 -0.0223,0.0167 -0.006,0.009 -0.0286,0.0257 -0.0506,0.0367 -0.0219,0.011 -0.0526,0.0335 -0.0682,0.05 -0.0156,0.0165 -0.0384,0.03 -0.0506,0.03 -0.0122,0 -0.0257,0.009 -0.0299,0.02 -0.005,0.011 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.006 -0.0186,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.0208,0.0119 -0.0115,0 -0.0173,0.006 -0.0132,0.0124 0.005,0.007 -0.006,0.0167 -0.0225,0.0219 -0.0166,0.005 -0.0434,0.0236 -0.0596,0.041 -0.0162,0.0174 -0.038,0.0315 -0.0487,0.0315 -0.0105,0 -0.0227,0.009 -0.0269,0.02 -0.005,0.0109 -0.0169,0.02 -0.0281,0.02 -0.0112,0 -0.0205,0.005 -0.0205,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0144 -0.065,0.0399 -0.0833,0.0564 -0.0183,0.0165 -0.0381,0.03 -0.044,0.03 -0.006,0 -0.0247,0.015 -0.0417,0.0334 -0.017,0.0183 -0.037,0.0334 -0.0446,0.0334 -0.008,0 -0.0172,0.009 -0.0214,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0134 0,0.008 -0.009,0.0134 -0.0186,0.0134 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0158,0.02 -0.0257,0.02 -0.01,0 -0.0382,0.018 -0.0629,0.0401 -0.0247,0.022 -0.0483,0.04 -0.0526,0.04 -0.005,0 -0.0317,0.0209 -0.061,0.0464 -0.0293,0.0255 -0.061,0.0466 -0.0705,0.0467 -0.009,1.5e-4 -0.0306,0.0142 -0.0467,0.0312 -0.0162,0.017 -0.0384,0.0352 -0.0494,0.0402 -0.0111,0.005 -0.0381,0.0212 -0.0601,0.0359 -0.022,0.0147 -0.0505,0.0303 -0.0633,0.0348 -0.0128,0.005 -0.0234,0.0134 -0.0234,0.0199 0,0.007 -0.007,0.0118 -0.0143,0.0118 -0.008,0 -0.0227,0.008 -0.0328,0.0167 -0.0442,0.0397 -0.0593,0.05 -0.0724,0.05 -0.008,0 -0.0138,0.009 -0.0138,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0132 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0128 -0.0334,0.0182 0,0.005 -0.0211,0.0212 -0.0468,0.0349 -0.0257,0.0139 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0103 -0.0186,0.0103 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0166,0.02 -0.0275,0.02 -0.0109,0 -0.0342,0.0143 -0.0516,0.0318 -0.0175,0.0175 -0.0448,0.0359 -0.0606,0.041 -0.0159,0.005 -0.0289,0.0147 -0.0289,0.0215 0,0.007 -0.009,0.0125 -0.02,0.0125 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.008,0 -0.0302,0.0152 -0.05,0.0338 -0.0199,0.0186 -0.0812,0.0618 -0.13619,0.0961 -0.0551,0.0342 -0.10209,0.067 -0.10452,0.073 -0.002,0.006 -0.0129,0.0107 -0.0234,0.0107 -0.0105,0 -0.0189,0.005 -0.019,0.01 -6e-5,0.005 -0.0226,0.0249 -0.05,0.0432 -0.0275,0.0183 -0.0594,0.0433 -0.0707,0.0558 -0.0115,0.0125 -0.037,0.0267 -0.0567,0.0317 -0.0198,0.005 -0.0359,0.0144 -0.0359,0.0209 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.005 -0.02,0.0103 0,0.006 -0.0211,0.0215 -0.0468,0.0354 -0.0257,0.0139 -0.0468,0.0332 -0.0468,0.0431 0,0.01 -0.0121,0.0179 -0.0271,0.0179 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.005 -0.0186,0.0112 0,0.006 -0.0211,0.0225 -0.0468,0.0364 -0.0257,0.0138 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0102 -0.02,0.0102 -0.0111,0 -0.02,0.006 -0.02,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0264 -0.0183,0.007 -0.0334,0.0188 -0.0334,0.0264 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.0151,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.007,0.0119 -0.0151,0.0119 -0.009,0 -0.0278,0.0135 -0.0434,0.03 -0.0155,0.0165 -0.0612,0.0509 -0.10161,0.0763 -0.0403,0.0254 -0.0764,0.0494 -0.08,0.0533 -0.0125,0.0133 -0.0599,0.0412 -0.0834,0.049 -0.0128,0.005 -0.0234,0.0134 -0.0234,0.02 0,0.007 -0.0106,0.0162 -0.0234,0.0211 -0.0128,0.005 -0.0306,0.0151 -0.0396,0.0229 -0.0428,0.0372 -0.0567,0.0474 -0.0643,0.0474 -0.005,0 -0.0165,0.008 -0.0268,0.0167 -0.0447,0.0401 -0.0594,0.05 -0.0737,0.05 -0.017,0 -0.17693,0.11791 -0.18389,0.13552 -0.002,0.006 -0.0129,0.0113 -0.0234,0.0113 -0.0105,0 -0.019,0.005 -0.019,0.0118 0,0.007 -0.0105,0.0153 -0.0234,0.0196 -0.0313,0.0105 -0.0686,0.0357 -0.10657,0.072 -0.0173,0.0165 -0.0374,0.03 -0.0448,0.03 -0.008,0 -0.0267,0.0135 -0.0428,0.03 -0.0339,0.0345 -0.0725,0.0607 -0.10602,0.072 -0.0128,0.005 -0.0234,0.0132 -0.0234,0.0196 0,0.007 -0.006,0.0118 -0.014,0.0118 -0.0151,0 -0.0419,0.0199 -0.0767,0.0567 -0.0121,0.0128 -0.0297,0.0234 -0.0391,0.0234 -0.009,0 -0.017,0.006 -0.017,0.0125 0,0.007 -0.0121,0.0163 -0.0269,0.0209 -0.0148,0.005 -0.0235,0.0141 -0.0193,0.0209 0.005,0.007 -0.008,0.0124 -0.0274,0.0124 -0.0199,0 -0.0314,0.006 -0.0267,0.0134 0.005,0.008 -7.6e-4,0.0134 -0.0126,0.0134 -0.0115,0 -0.0208,0.006 -0.0208,0.0125 0,0.007 -0.0129,0.0165 -0.0287,0.0215 -0.0158,0.005 -0.0413,0.0225 -0.0567,0.0386 -0.0507,0.0534 -0.0658,0.0499 -0.14595,-0.0338 z m 3.73064,-0.57254 c -0.0306,-0.0293 -0.0618,-0.0533 -0.0694,-0.0533 -0.008,0 -0.0139,-0.006 -0.0139,-0.0134 0,-0.008 -0.005,-0.0134 -0.012,-0.0134 -0.0122,0 -0.0685,-0.0452 -0.13702,-0.11008 -0.0213,-0.0202 -0.0435,-0.0367 -0.0493,-0.0367 -0.006,0 -0.0186,-0.0151 -0.0285,-0.0334 -0.01,-0.0183 -0.0263,-0.0334 -0.0367,-0.0334 -0.0104,0 -0.0363,-0.0176 -0.0577,-0.039 -0.0215,-0.0215 -0.039,-0.0354 -0.039,-0.0312 0,0.005 -0.0225,-0.0164 -0.05,-0.0462 -0.0276,-0.0297 -0.0621,-0.0566 -0.0767,-0.0597 -0.0147,-0.003 -0.0251,-0.0128 -0.0234,-0.0215 0.002,-0.009 -0.002,-0.0159 -0.0106,-0.0159 -0.008,0 -0.0279,-0.015 -0.0448,-0.0334 -0.017,-0.0183 -0.0343,-0.0334 -0.0384,-0.0334 -0.005,0 -0.0298,-0.021 -0.057,-0.0468 -0.0272,-0.0257 -0.0538,-0.0467 -0.0591,-0.0467 -0.0147,0 -0.0801,-0.0639 -0.0801,-0.0783 0,-0.007 -0.008,-0.0159 -0.0167,-0.0196 -0.0102,-0.005 -0.008,-0.007 0.006,-0.008 0.0125,-5.1e-4 0.0422,-0.0191 0.0664,-0.0411 0.0242,-0.022 0.0516,-0.0401 0.0609,-0.0401 0.009,0 0.017,-0.006 0.017,-0.0134 0,-0.008 0.009,-0.0134 0.021,-0.0134 0.0115,0 0.0293,-0.008 0.0396,-0.0167 0.0442,-0.0396 0.0592,-0.05 0.0724,-0.05 0.008,0 0.0139,-0.009 0.0139,-0.02 0,-0.0126 0.0127,-0.02 0.0341,-0.02 0.0189,0 0.0306,-0.006 0.0263,-0.0127 -0.005,-0.007 0.005,-0.0189 0.0182,-0.0267 0.0143,-0.008 0.034,-0.0215 0.0438,-0.0307 0.01,-0.009 0.0402,-0.0318 0.0676,-0.0501 0.0275,-0.0183 0.05,-0.0378 0.05,-0.0433 5e-5,-0.005 0.006,-0.01 0.014,-0.01 0.008,0 0.027,-0.0122 0.0431,-0.0272 0.0161,-0.015 0.0428,-0.0306 0.0595,-0.0348 0.0167,-0.005 0.0303,-0.0164 0.0303,-0.027 0,-0.0106 0.005,-0.016 0.0119,-0.012 0.007,0.005 0.0154,-0.002 0.0196,-0.0126 0.005,-0.0109 0.02,-0.0199 0.0352,-0.0199 0.0151,0 0.031,-0.009 0.0352,-0.02 0.005,-0.0111 0.0164,-0.02 0.027,-0.02 0.0107,0 0.0157,-0.006 0.0112,-0.0134 -0.005,-0.008 -0.002,-0.0134 0.007,-0.0134 0.0158,0 0.0423,-0.0193 0.0775,-0.0567 0.0121,-0.0128 0.0312,-0.0235 0.0425,-0.0238 0.0208,-5.1e-4 0.0813,-0.0396 0.0611,-0.0396 -0.006,0 0.0111,-0.015 0.0381,-0.0334 0.027,-0.0183 0.0549,-0.0334 0.0619,-0.0334 0.007,0 0.0128,-0.006 0.0128,-0.0134 0,-0.008 0.006,-0.0134 0.0128,-0.0134 0.007,0 0.0313,-0.018 0.0539,-0.0401 0.0225,-0.022 0.0475,-0.04 0.0555,-0.04 0.008,0 0.0244,-0.0105 0.0362,-0.0234 0.0118,-0.0128 0.0448,-0.0352 0.0732,-0.0497 0.0283,-0.0144 0.0545,-0.034 0.0581,-0.0434 0.003,-0.009 0.0134,-0.017 0.0219,-0.017 0.009,0 0.0316,-0.015 0.0513,-0.0334 0.0198,-0.0183 0.0409,-0.0334 0.0468,-0.0334 0.006,0 0.0269,-0.015 0.0468,-0.0334 0.0198,-0.0183 0.0425,-0.0334 0.0504,-0.0334 0.008,0 0.0178,-0.009 0.022,-0.02 0.005,-0.0111 0.0154,-0.02 0.0248,-0.02 0.009,0 0.0206,-0.009 0.0248,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.005 0.02,-0.0102 0,-0.006 0.0241,-0.025 0.0533,-0.0432 0.0293,-0.0182 0.0533,-0.0376 0.0533,-0.0432 0,-0.006 0.007,-0.0102 0.0143,-0.0102 0.008,0 0.0227,-0.008 0.0328,-0.0167 0.0408,-0.0367 0.0588,-0.05 0.0671,-0.05 0.0112,0 0.11784,-0.0834 0.12368,-0.0966 0.002,-0.005 0.0129,-0.0101 0.0234,-0.0101 0.0105,0 0.0189,-0.005 0.0189,-0.0112 0,-0.006 0.024,-0.0227 0.0532,-0.0368 0.0293,-0.014 0.0532,-0.0316 0.0533,-0.0391 1e-4,-0.008 0.0209,-0.0248 0.0462,-0.0384 0.0254,-0.0136 0.0597,-0.0375 0.0764,-0.0531 0.0167,-0.0155 0.0335,-0.0286 0.0374,-0.0291 0.0167,-0.002 0.0936,-0.0534 0.0936,-0.0625 4e-5,-0.005 0.0121,-0.01 0.0268,-0.01 0.0148,0 0.0267,-0.009 0.0267,-0.02 0,-0.0109 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.011 0.0164,-0.02 0.027,-0.02 0.0106,0 0.0213,-0.005 0.0238,-0.01 0.008,-0.0167 0.11532,-0.0967 0.13043,-0.0967 0.008,0 0.0141,-0.005 0.0141,-0.0105 0,-0.006 0.0195,-0.0219 0.0433,-0.0357 0.0239,-0.0138 0.0464,-0.0289 0.05,-0.0332 0.0186,-0.0225 0.11921,-0.0941 0.13202,-0.0941 0.008,0 0.0147,-0.006 0.0147,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.024,-0.0238 0.0533,-0.0364 0.0293,-0.0127 0.0533,-0.0307 0.0533,-0.04 0,-0.009 0.007,-0.0169 0.0143,-0.0169 0.008,0 0.0227,-0.008 0.0328,-0.0167 0.0423,-0.0379 0.059,-0.05 0.069,-0.0501 0.006,-3e-5 0.0298,-0.018 0.0533,-0.04 0.0234,-0.022 0.049,-0.04 0.0567,-0.04 0.008,0 0.0141,-0.006 0.0141,-0.0134 0,-0.008 0.009,-0.0134 0.0189,-0.0134 0.0105,0 0.0208,-0.005 0.0234,-0.01 0.007,-0.0148 0.12679,-0.0967 0.14197,-0.0967 0.007,0 0.0165,-0.009 0.0207,-0.02 0.005,-0.0111 0.0141,-0.02 0.0221,-0.02 0.008,0 0.0293,-0.015 0.0477,-0.0334 0.0183,-0.0183 0.0393,-0.0334 0.0466,-0.0334 0.007,0 0.0308,-0.018 0.0523,-0.04 0.0215,-0.022 0.047,-0.0401 0.0567,-0.0401 0.01,0 0.0262,-0.012 0.0368,-0.0267 0.0106,-0.0148 0.0269,-0.0267 0.0363,-0.0267 0.009,0 0.017,-0.006 0.017,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.009 0.02,-0.0185 0,-0.0102 0.0105,-0.022 0.0234,-0.0263 0.0299,-0.01 0.0665,-0.0342 0.10925,-0.072 0.0186,-0.0165 0.0416,-0.03 0.0508,-0.03 0.009,0 0.0168,-0.006 0.0168,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0134,-0.02 0.0205,-0.02 0.007,0 0.0212,-0.008 0.0315,-0.0167 0.037,-0.0332 0.0537,-0.0452 0.0984,-0.0708 0.025,-0.0143 0.0695,-0.0485 0.0989,-0.076 0.0293,-0.0276 0.0613,-0.05 0.0712,-0.05 0.01,0 0.0179,-0.005 0.0179,-0.0105 0,-0.0107 0.1639,-0.12293 0.17953,-0.12293 0.005,0 0.0225,-0.015 0.0395,-0.0334 0.0305,-0.0328 0.0879,-0.046 0.0879,-0.02 0,0.008 0.009,0.0134 0.0186,0.0134 0.0102,0 0.0221,0.009 0.0263,0.02 0.005,0.0111 0.0199,0.02 0.0348,0.02 0.0149,0 0.0271,0.005 0.0271,0.0119 0,0.007 0.0211,0.0158 0.0468,0.0206 0.0257,0.005 0.0468,0.0144 0.0468,0.0215 0,0.007 0.012,0.0128 0.0267,0.0128 0.0147,0 0.0267,0.005 0.0267,0.0108 0,0.006 0.0255,0.0217 0.0567,0.035 0.0312,0.0134 0.0642,0.0336 0.0733,0.0452 0.0108,0.0135 0.0167,0.0151 0.0167,0.005 0,-0.0101 0.0102,-0.007 0.0269,0.009 0.0275,0.0256 0.0606,0.0448 0.11986,0.0693 0.0183,0.008 0.0513,0.0257 0.0733,0.0402 0.022,0.0144 0.049,0.0299 0.06,0.0343 0.0506,0.0203 0.0882,0.0403 0.10436,0.0556 0.01,0.009 0.0262,0.0167 0.0367,0.0167 0.0105,0 0.0191,0.006 0.0191,0.0134 0,0.008 0.015,0.0134 0.0334,0.0134 0.0183,0 0.0332,0.005 0.0329,0.01 -5.2e-4,0.0144 -0.0342,0.0567 -0.045,0.0567 -0.005,0 -0.0231,0.015 -0.0401,0.0334 -0.017,0.0183 -0.0352,0.0334 -0.0403,0.0334 -0.005,0 -0.0178,0.008 -0.028,0.0167 -0.0401,0.036 -0.0587,0.05 -0.0662,0.05 -0.005,0 -0.0211,0.0139 -0.0372,0.0309 -0.0162,0.017 -0.0504,0.0436 -0.0761,0.0593 -0.0257,0.0157 -0.0555,0.0409 -0.0664,0.0559 -0.0108,0.0151 -0.0272,0.0275 -0.0365,0.0275 -0.009,0 -0.0331,0.018 -0.0526,0.04 -0.0196,0.022 -0.0437,0.04 -0.0535,0.04 -0.01,0 -0.0178,0.006 -0.0178,0.0134 0,0.008 -0.008,0.0134 -0.0173,0.0134 -0.009,0 -0.0423,0.0241 -0.0729,0.0533 -0.0306,0.0293 -0.0613,0.0533 -0.0683,0.0533 -0.007,0 -0.0147,0.005 -0.0172,0.0106 -0.002,0.006 -0.0311,0.0313 -0.0636,0.0567 -0.0326,0.0254 -0.069,0.0552 -0.0811,0.0662 -0.0406,0.0371 -0.099,0.0801 -0.10895,0.0801 -0.005,0 -0.0351,0.0241 -0.066,0.0533 -0.0308,0.0293 -0.062,0.0533 -0.0691,0.0533 -0.007,0 -0.0269,0.015 -0.0439,0.0334 -0.017,0.0183 -0.0347,0.0334 -0.0391,0.0334 -0.005,0 -0.0166,0.008 -0.0269,0.0167 -0.0442,0.0396 -0.0593,0.05 -0.0724,0.05 -0.008,0 -0.0139,0.009 -0.0139,0.02 0,0.0111 -0.006,0.02 -0.014,0.02 -0.008,0 -0.0302,0.015 -0.0499,0.0334 -0.0198,0.0183 -0.0436,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.008,0.02 -0.018,0.02 -0.01,0 -0.0397,0.0211 -0.066,0.0468 -0.0264,0.0257 -0.0537,0.0468 -0.0604,0.0468 -0.007,0 -0.0355,0.0212 -0.064,0.0473 -0.0283,0.026 -0.0652,0.0506 -0.0818,0.0548 -0.0166,0.005 -0.0301,0.0159 -0.0301,0.0261 0,0.0102 -0.007,0.0186 -0.0144,0.0186 -0.008,0 -0.0382,0.024 -0.0672,0.0533 -0.0291,0.0293 -0.0594,0.0533 -0.0675,0.0533 -0.008,0 -0.0285,0.015 -0.0455,0.0334 -0.017,0.0183 -0.0361,0.0334 -0.0426,0.0334 -0.006,0 -0.0257,0.015 -0.043,0.0334 -0.0172,0.0183 -0.0368,0.0334 -0.0434,0.0334 -0.007,0 -0.0307,0.0186 -0.0535,0.0415 -0.0228,0.0228 -0.0644,0.0566 -0.0924,0.0752 -0.028,0.0185 -0.051,0.0389 -0.051,0.0452 0,0.006 -0.009,0.0115 -0.02,0.0115 -0.011,0 -0.02,0.005 -0.02,0.0105 0,0.006 -0.0165,0.0205 -0.0368,0.033 -0.0202,0.0123 -0.0577,0.0406 -0.0834,0.0627 -0.0759,0.0655 -0.0783,0.0673 -0.0928,0.0673 -0.008,0 -0.014,0.006 -0.014,0.0125 0,0.007 -0.0121,0.0163 -0.0269,0.0209 -0.0148,0.005 -0.0235,0.0141 -0.0193,0.0209 0.005,0.007 -0.002,0.0124 -0.0132,0.0124 -0.0115,0 -0.0208,0.006 -0.0208,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0264 -0.0183,0.007 -0.0334,0.0188 -0.0334,0.0264 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.008,0 -0.0301,0.0151 -0.0499,0.0334 -0.0198,0.0183 -0.0436,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.009,0.02 -0.0186,0.02 -0.0102,0 -0.0221,0.009 -0.0263,0.02 -0.005,0.011 -0.0203,0.02 -0.0356,0.02 -0.0154,0 -0.0244,0.006 -0.0202,0.0124 0.005,0.007 -0.0105,0.0232 -0.0326,0.0362 -0.0221,0.0131 -0.0403,0.0285 -0.0403,0.0342 0,0.006 -0.009,0.0105 -0.02,0.0105 -0.0111,0 -0.02,0.005 -0.02,0.0111 0,0.006 -0.0285,0.0314 -0.0633,0.0562 -0.0348,0.0248 -0.0694,0.0517 -0.0767,0.0597 -0.0243,0.0267 -0.11139,0.0865 -0.12587,0.0865 -0.008,0 -0.0142,0.009 -0.0142,0.02 0,0.0109 -0.007,0.02 -0.0151,0.02 -0.009,0 -0.029,0.015 -0.0461,0.0334 -0.017,0.0183 -0.0375,0.0334 -0.0455,0.0334 -0.008,0 -0.0285,0.0151 -0.0455,0.0334 -0.017,0.0183 -0.0347,0.0334 -0.0393,0.0334 -0.009,0 -0.0607,0.0417 -0.10427,0.0834 -0.0135,0.0128 -0.0304,0.0234 -0.0377,0.0234 -0.008,0 -0.0134,0.006 -0.0134,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.007 -0.02,0.0143 0,0.0185 -0.0339,0.0524 -0.0524,0.0524 -0.008,0 -0.0143,0.006 -0.0143,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0263 -0.0183,0.007 -0.0334,0.0189 -0.0334,0.0264 0,0.008 -0.007,0.0137 -0.0151,0.0137 -0.009,0 -0.0293,0.0154 -0.0468,0.0341 -0.0173,0.0187 -0.0315,0.0311 -0.0315,0.0274 0,-0.003 -0.0193,0.0111 -0.043,0.0327 -0.0236,0.0216 -0.0474,0.0393 -0.0528,0.0393 -0.005,0 -0.0343,0.024 -0.0643,0.0533 -0.0299,0.0293 -0.061,0.0533 -0.0691,0.0533 -0.008,0 -0.0285,0.015 -0.0455,0.0334 -0.017,0.0183 -0.0345,0.0334 -0.0388,0.0334 -0.005,0 -0.0241,0.0151 -0.0439,0.0334 -0.0198,0.0183 -0.0435,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.005 -0.02,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0145 -0.0234,0.0215 0,0.0172 -0.0347,0.0509 -0.0524,0.0509 -0.008,0 -0.0143,0.009 -0.0143,0.0185 0,0.0101 -0.0105,0.0225 -0.0234,0.0272 -0.0438,0.0165 -0.0834,0.0406 -0.0834,0.0509 0,0.006 -0.007,0.0102 -0.0151,0.0102 -0.009,0 -0.028,0.0138 -0.0437,0.0307 -0.0157,0.017 -0.0389,0.0348 -0.0516,0.0397 -0.0127,0.005 -0.023,0.0151 -0.023,0.0226 0,0.0281 -0.0363,0.0121 -0.0902,-0.0396 z m -4.80688,-0.51843 c 0,-0.006 -0.0595,-0.0718 -0.13233,-0.14566 -0.0728,-0.0739 -0.12882,-0.13779 -0.12454,-0.14208 0.005,-0.005 0.002,-0.008 -0.007,-0.008 -0.0182,0 -0.10368,-0.083 -0.0996,-0.0967 0.002,-0.005 -0.002,-0.009 -0.007,-0.007 -0.0111,0.003 -0.0967,-0.0745 -0.0967,-0.0882 0,-0.0151 -0.0509,-0.06 -0.0589,-0.052 -0.005,0.005 -0.008,-0.002 -0.008,-0.0144 0,-0.0122 -0.007,-0.0222 -0.0143,-0.0222 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.009,-0.0143 -0.02,-0.0143 -0.0109,0 -0.02,-0.005 -0.02,-0.0115 0,-0.006 -0.0156,-0.027 -0.0345,-0.0461 -0.0368,-0.0368 -0.0336,-0.0566 0.0125,-0.0766 0.0352,-0.0154 0.0749,-0.0404 0.0992,-0.0626 0.01,-0.009 0.0268,-0.0167 0.0372,-0.0167 0.0105,0 0.0191,-0.005 0.0191,-0.0101 0,-0.005 0.0285,-0.0237 0.0633,-0.0404 0.0754,-0.0361 0.10607,-0.0548 0.14245,-0.0868 0.0149,-0.0131 0.0407,-0.0273 0.0575,-0.0315 0.0166,-0.005 0.0303,-0.012 0.0303,-0.0174 0,-0.005 0.024,-0.0212 0.0532,-0.0353 0.0293,-0.0141 0.0532,-0.0293 0.0533,-0.0341 10e-5,-0.005 0.0151,-0.0123 0.0335,-0.017 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.0144,-0.0131 0.0319,-0.0131 0.0176,0 0.0354,-0.009 0.0396,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.005 0.0271,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0267,-0.009 0.0672,-0.0343 0.0938,-0.0587 0.01,-0.009 0.0268,-0.0167 0.0373,-0.0167 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.0153,-0.0134 0.0341,-0.0134 0.0208,0 0.0307,-0.006 0.0254,-0.0142 -0.005,-0.008 0.003,-0.0176 0.0192,-0.0215 0.0154,-0.005 0.028,-0.012 0.028,-0.0177 0,-0.006 0.015,-0.0141 0.0334,-0.0186 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.021,-0.0131 0.0115,0 0.0293,-0.008 0.0396,-0.0167 0.0483,-0.0433 0.062,-0.0519 0.0863,-0.0537 0.0143,-7.6e-4 0.029,-0.01 0.0326,-0.0192 0.003,-0.009 0.0158,-0.0172 0.0271,-0.0172 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.0122,-0.0134 0.0271,-0.0134 0.0149,0 0.0306,-0.009 0.0348,-0.02 0.005,-0.011 0.019,-0.02 0.0329,-0.02 0.0139,0 0.0253,-0.006 0.0253,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.0184,-0.0119 0.0101,0 0.0259,-0.009 0.035,-0.02 0.009,-0.011 0.0254,-0.02 0.036,-0.02 0.0106,0 0.0277,-0.008 0.0379,-0.0167 0.0397,-0.0356 0.0588,-0.05 0.0659,-0.05 0.005,0 0.0312,-0.0149 0.0601,-0.0332 0.0289,-0.0183 0.0596,-0.0348 0.0682,-0.0368 0.009,-0.002 0.0309,-0.0172 0.0496,-0.0339 0.0186,-0.0167 0.0479,-0.0338 0.0649,-0.0382 0.017,-0.005 0.0311,-0.0132 0.0311,-0.0197 0,-0.007 0.0122,-0.0119 0.0274,-0.0119 0.0151,0 0.0423,-0.015 0.0606,-0.0334 0.0183,-0.0183 0.0452,-0.0334 0.0594,-0.0334 0.0144,0 0.0261,-0.006 0.0261,-0.0125 0,-0.007 0.0133,-0.0166 0.0294,-0.0218 0.0162,-0.005 0.0376,-0.0176 0.0477,-0.0276 0.01,-0.01 0.0268,-0.0182 0.0374,-0.0182 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.015,-0.0134 0.0334,-0.0134 0.0207,0 0.0334,-0.008 0.0334,-0.02 0,-0.0112 0.0119,-0.02 0.0267,-0.02 0.0147,0 0.0267,-0.005 0.0267,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0321,-0.0108 0.0698,-0.0357 0.12907,-0.0854 0.0109,-0.009 0.0264,-0.0167 0.0343,-0.0167 0.0133,0 0.0783,-0.0369 0.0935,-0.0532 0.003,-0.005 0.0306,-0.019 0.0598,-0.0336 0.0293,-0.0147 0.0612,-0.0341 0.071,-0.0432 0.01,-0.009 0.0264,-0.0167 0.0369,-0.0167 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.0109 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.006 0.0271,-0.0134 0,-0.008 0.0116,-0.0134 0.0258,-0.0134 0.0142,0 0.0299,-0.006 0.0347,-0.0141 0.005,-0.008 0.0217,-0.0198 0.0375,-0.0267 0.0352,-0.0153 0.075,-0.0404 0.0992,-0.0626 0.01,-0.009 0.0256,-0.0167 0.0348,-0.0167 0.009,0 0.0286,-0.0112 0.0432,-0.0248 0.0148,-0.0137 0.0504,-0.036 0.0793,-0.0495 0.029,-0.0136 0.0604,-0.0324 0.0702,-0.0419 0.01,-0.009 0.0262,-0.0172 0.0367,-0.0172 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.009,-0.0134 0.0184,-0.0134 0.0101,0 0.0258,-0.009 0.035,-0.02 0.009,-0.0111 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.025,-0.006 0.025,-0.0131 0,-0.007 0.0151,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0122 0.0335,-0.017 10e-5,-0.005 0.0241,-0.02 0.0533,-0.0341 0.0293,-0.0141 0.0532,-0.0306 0.0532,-0.0367 0,-0.006 0.0121,-0.0112 0.0271,-0.0112 0.0149,0 0.0306,-0.009 0.0348,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0276,-0.01 0.0363,-0.0217 0.0319,-0.0436 0.12628,-0.0125 0.19549,0.0644 0.0118,0.0132 0.0281,0.024 0.0361,0.024 0.008,0 0.0192,0.009 0.0251,0.0183 0.006,0.01 0.0302,0.0282 0.054,0.0404 0.0239,0.0122 0.0434,0.027 0.0434,0.0331 0,0.01 0.0449,0.0355 0.0701,0.04 0.005,7.7e-4 0.0215,0.0123 0.0357,0.0251 0.0484,0.0439 0.0665,0.0567 0.0805,0.0567 0.008,0 0.014,0.006 0.014,0.0134 0,0.008 0.006,0.0134 0.0134,0.0134 0.008,0 0.0249,0.0106 0.039,0.0234 0.0484,0.0439 0.0665,0.0567 0.0805,0.0567 0.008,0 0.0139,0.006 0.0139,0.0134 0,0.008 0.009,0.0134 0.0205,0.0134 0.0112,0 0.0239,0.009 0.0281,0.02 0.005,0.0111 0.0133,0.02 0.0203,0.02 0.007,0 0.025,0.0116 0.0403,0.026 0.0239,0.0222 0.0251,0.0284 0.009,0.0434 -0.0271,0.025 -0.0675,0.0503 -0.0945,0.0594 -0.0128,0.005 -0.0234,0.0119 -0.0234,0.0166 0,0.005 -0.0225,0.0205 -0.05,0.035 -0.0275,0.0144 -0.0548,0.0339 -0.0607,0.0431 -0.006,0.009 -0.0223,0.0168 -0.0367,0.0168 -0.0143,0 -0.026,0.005 -0.026,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0372,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.0118,0.02 -0.026,0.02 -0.0143,0 -0.0308,0.008 -0.0368,0.0174 -0.006,0.01 -0.0332,0.0278 -0.0607,0.0406 -0.0276,0.0128 -0.0501,0.0275 -0.0502,0.0326 -1.1e-4,0.01 -0.064,0.0427 -0.0831,0.0427 -0.0113,0 -0.0307,0.0141 -0.0765,0.0554 -0.0134,0.0121 -0.046,0.0298 -0.0724,0.0393 -0.0264,0.009 -0.048,0.0221 -0.048,0.0277 0,0.006 -0.0241,0.0227 -0.0533,0.0376 -0.0293,0.015 -0.0533,0.0332 -0.0533,0.0403 0,0.007 -0.0114,0.0131 -0.0253,0.0131 -0.0139,0 -0.0286,0.009 -0.0329,0.02 -0.005,0.0111 -0.0169,0.02 -0.0281,0.02 -0.0112,0 -0.0204,0.006 -0.0204,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.005 -0.02,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.007,0.0131 -0.0143,0.0131 -0.008,0 -0.0227,0.008 -0.0328,0.0167 -0.0464,0.0416 -0.0597,0.05 -0.0795,0.05 -0.0115,0 -0.0244,0.009 -0.0286,0.02 -0.005,0.0111 -0.0191,0.02 -0.033,0.02 -0.0138,0 -0.0253,0.006 -0.0253,0.0134 0,0.008 -0.008,0.0134 -0.0168,0.0134 -0.009,0 -0.033,0.0151 -0.0529,0.0334 -0.0198,0.0183 -0.0418,0.0334 -0.049,0.0334 -0.007,0 -0.0209,0.008 -0.0306,0.0167 -0.01,0.009 -0.0415,0.0287 -0.0708,0.0433 -0.0293,0.0147 -0.0605,0.0327 -0.0696,0.04 -0.009,0.008 -0.0254,0.0208 -0.0363,0.03 -0.0109,0.009 -0.0255,0.0167 -0.0324,0.0167 -0.007,0 -0.0159,0.009 -0.0202,0.02 -0.005,0.0111 -0.0173,0.02 -0.0291,0.02 -0.0118,0 -0.0298,0.008 -0.04,0.0167 -0.0462,0.0415 -0.0597,0.05 -0.079,0.05 -0.0113,0 -0.0206,0.006 -0.0206,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.013 -0.0334,0.0186 0,0.006 -0.0105,0.0139 -0.0234,0.0183 -0.0244,0.009 -0.072,0.0365 -0.0834,0.0497 -0.003,0.005 -0.0275,0.0192 -0.053,0.0334 -0.0499,0.0277 -0.058,0.0332 -0.10414,0.0711 -0.0167,0.0137 -0.0497,0.0328 -0.0733,0.0426 -0.0236,0.01 -0.0495,0.0239 -0.0574,0.0315 -0.008,0.008 -0.0424,0.0316 -0.0767,0.0533 -0.0343,0.0217 -0.0623,0.0442 -0.0623,0.05 0,0.006 -0.0114,0.0107 -0.0253,0.0107 -0.0139,0 -0.0287,0.009 -0.033,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0134 0,0.008 -0.009,0.0134 -0.0193,0.0134 -0.0106,0 -0.0246,0.009 -0.0309,0.0206 -0.006,0.0114 -0.0247,0.0216 -0.0406,0.0228 -0.016,0.002 -0.0291,0.0103 -0.0291,0.0202 0,0.01 -0.006,0.0142 -0.0135,0.01 -0.008,-0.005 -0.017,7.7e-4 -0.0214,0.0126 -0.005,0.0115 -0.0193,0.0209 -0.0333,0.0209 -0.0139,0 -0.0253,0.009 -0.0253,0.02 0,0.0109 -0.009,0.02 -0.0202,0.02 -0.0247,0 -0.0764,0.0279 -0.0897,0.0485 -0.005,0.009 -0.0249,0.019 -0.0434,0.0236 -0.0184,0.005 -0.0335,0.0143 -0.0335,0.0215 0,0.007 -0.009,0.0131 -0.0208,0.0131 -0.0115,0 -0.0171,0.006 -0.0126,0.0134 0.005,0.008 -0.005,0.0134 -0.0196,0.0134 -0.0154,0 -0.0309,0.008 -0.0345,0.0172 -0.003,0.009 -0.0334,0.029 -0.0662,0.0434 -0.0328,0.0144 -0.0597,0.0306 -0.0597,0.0362 -6e-5,0.005 -0.009,0.01 -0.0211,0.01 -0.0115,0 -0.0293,0.008 -0.0396,0.0167 -0.039,0.0349 -0.0587,0.05 -0.0652,0.05 -0.007,0 -0.12791,0.0802 -0.14105,0.0934 -0.003,0.003 -0.0292,0.0182 -0.0567,0.0321 -0.0275,0.014 -0.05,0.0299 -0.05,0.0353 0,0.005 -0.015,0.0136 -0.0334,0.0182 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.02,0.0131 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.012,0.0119 -0.0267,0.0119 -0.0147,0 -0.0267,-0.005 -0.0267,-0.0114 z m 8.3197,-0.0215 c -0.0147,-0.0179 -0.0365,-0.0328 -0.0487,-0.0332 -0.0121,-2.7e-4 -0.0254,-0.01 -0.0296,-0.0206 -0.005,-0.0111 -0.0143,-0.02 -0.0225,-0.02 -0.008,0 -0.0231,-0.008 -0.0333,-0.0167 -0.0442,-0.0396 -0.0593,-0.05 -0.0724,-0.05 -0.008,0 -0.0138,-0.006 -0.0138,-0.0134 0,-0.008 -0.009,-0.0134 -0.0186,-0.0134 -0.0102,0 -0.022,-0.009 -0.0263,-0.02 -0.005,-0.011 -0.0154,-0.0201 -0.0248,-0.0202 -0.009,-1.1e-4 -0.0381,-0.0181 -0.0638,-0.0399 -0.0257,-0.0219 -0.0539,-0.0457 -0.0628,-0.0532 -0.009,-0.008 -0.0223,-0.0135 -0.03,-0.0135 -0.008,0 -0.0139,-0.005 -0.0139,-0.0119 0,-0.007 -0.0135,-0.0154 -0.03,-0.0197 -0.0165,-0.005 -0.0474,-0.0229 -0.0688,-0.0414 -0.0213,-0.0186 -0.0441,-0.0338 -0.0506,-0.0338 -0.007,0 -0.0258,-0.0151 -0.0428,-0.0334 -0.017,-0.0183 -0.0352,-0.0334 -0.0403,-0.0334 -0.005,0 -0.0178,-0.008 -0.028,-0.0167 -0.0453,-0.0406 -0.0595,-0.0501 -0.0757,-0.0503 -0.009,-9e-5 -0.0361,-0.0166 -0.0592,-0.0367 -0.0231,-0.02 -0.0554,-0.0455 -0.0717,-0.0565 -0.0163,-0.0111 -0.0375,-0.0276 -0.0471,-0.0367 -0.01,-0.009 -0.0239,-0.0167 -0.0318,-0.0167 -0.008,0 -0.0143,-0.012 -0.0143,-0.0267 0,-0.0147 0.007,-0.0267 0.0144,-0.0267 0.008,0 0.0318,-0.0165 0.0529,-0.0368 0.0596,-0.0567 0.0771,-0.0701 0.0922,-0.0701 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0111 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0141 0,-0.008 0.006,-0.0106 0.0125,-0.006 0.007,0.005 0.0187,-0.002 0.0263,-0.0158 0.008,-0.013 0.014,-0.0187 0.0141,-0.0127 2.1e-4,0.006 0.0206,-0.0105 0.0454,-0.0367 0.0247,-0.0262 0.0517,-0.0477 0.0601,-0.0477 0.009,0 0.0151,-0.009 0.0151,-0.02 0,-0.0111 0.005,-0.0185 0.0103,-0.0167 0.006,0.002 0.028,-0.0132 0.0499,-0.0334 0.0219,-0.0203 0.0467,-0.0367 0.0553,-0.0367 0.009,0 0.0177,-0.005 0.0203,-0.0101 0.002,-0.005 0.0314,-0.0306 0.0645,-0.0558 0.0331,-0.0251 0.0728,-0.0596 0.0883,-0.0766 0.0155,-0.017 0.0345,-0.0309 0.0423,-0.0309 0.008,0 0.016,-0.005 0.0184,-0.0105 0.002,-0.006 0.0308,-0.0313 0.0631,-0.0567 0.0323,-0.0254 0.0831,-0.0688 0.11283,-0.0962 0.0298,-0.0275 0.0598,-0.05 0.0666,-0.05 0.007,0 0.0263,-0.015 0.0433,-0.0334 0.017,-0.0183 0.0373,-0.0334 0.0448,-0.0334 0.008,0 0.014,-0.009 0.014,-0.0208 0,-0.0115 0.006,-0.017 0.0135,-0.0125 0.008,0.005 0.017,-7.6e-4 0.0214,-0.0125 0.005,-0.0114 0.0121,-0.0183 0.017,-0.0152 0.005,0.003 0.0477,-0.0299 0.0949,-0.0734 0.0473,-0.0435 0.0981,-0.0791 0.11302,-0.0793 0.0223,-1.6e-4 0.0236,-0.002 0.007,-0.0131 -0.0178,-0.0114 -0.0178,-0.013 0,-0.0134 0.0109,-2e-4 0.053,-0.0303 0.0934,-0.0669 0.0403,-0.0366 0.0839,-0.0699 0.0968,-0.0742 0.0128,-0.005 0.0234,-0.0133 0.0234,-0.0201 0,-0.007 0.006,-0.0125 0.0134,-0.0125 0.008,0 0.0243,-0.0105 0.0377,-0.0234 0.0413,-0.0396 0.0955,-0.0834 0.10309,-0.0834 0.005,0 0.0212,-0.015 0.0381,-0.0334 0.017,-0.0183 0.0377,-0.0334 0.0461,-0.0334 0.009,0 0.0151,-0.009 0.0151,-0.02 0,-0.011 0.009,-0.02 0.0206,-0.02 0.0114,0 0.0346,-0.015 0.0516,-0.0334 0.017,-0.0183 0.0345,-0.0334 0.039,-0.0334 0.005,0 0.0277,-0.0183 0.0518,-0.0405 0.0241,-0.0222 0.0599,-0.0553 0.0797,-0.0733 0.0197,-0.018 0.0392,-0.0329 0.0432,-0.0329 0.005,0 0.021,-0.0116 0.0375,-0.0259 0.0928,-0.0798 0.11101,-0.0942 0.11918,-0.0942 0.005,-1e-5 0.026,-0.0166 0.0467,-0.0369 0.0207,-0.0203 0.0647,-0.0578 0.0978,-0.0833 0.0331,-0.0255 0.0621,-0.051 0.0645,-0.0565 0.002,-0.005 0.0102,-0.0101 0.0173,-0.0101 0.007,0 0.0268,-0.0135 0.0439,-0.03 0.0434,-0.0419 0.0867,-0.0767 0.0956,-0.0767 0.005,0 0.0158,-0.008 0.0261,-0.0167 0.0465,-0.0417 0.0598,-0.05 0.0798,-0.05 0.0118,0 0.018,-0.005 0.0141,-0.0119 -0.005,-0.007 0.009,-0.0225 0.0283,-0.0353 0.0196,-0.0128 0.0587,-0.0461 0.0866,-0.0738 0.028,-0.0277 0.051,-0.0448 0.051,-0.0381 0,0.007 0.0115,0.0123 0.0257,0.0123 0.0141,0 0.0338,0.008 0.0436,0.0167 0.01,0.009 0.0418,0.0286 0.071,0.0432 0.0293,0.0147 0.0562,0.0297 0.0598,0.0336 0.0182,0.0193 0.0815,0.0532 0.0996,0.0532 0.0113,0 0.0205,0.005 0.0205,0.0119 0,0.007 0.015,0.0157 0.0334,0.0203 0.0183,0.005 0.0334,0.0143 0.0334,0.0215 0,0.007 0.01,0.0131 0.0216,0.0131 0.024,0 0.18258,0.0809 0.18962,0.0967 0.002,0.005 0.0159,0.01 0.03,0.01 0.014,0 0.0256,0.006 0.0256,0.0134 0,0.008 0.012,0.0134 0.0267,0.0134 0.0147,0 0.0267,0.007 0.0267,0.0156 0,0.009 0.003,0.0119 0.008,0.008 0.005,-0.005 0.021,0.005 0.0368,0.0188 0.0157,0.0147 0.0451,0.0301 0.0653,0.0342 0.0202,0.005 0.0368,0.0118 0.0368,0.0173 0,0.005 0.03,0.0241 0.0667,0.0415 0.0367,0.0174 0.0667,0.0363 0.0667,0.0419 0,0.006 0.0165,0.0104 0.0367,0.0106 l 0.0367,5.2e-4 -0.0367,0.0322 c -0.0202,0.0178 -0.0367,0.0386 -0.0367,0.0462 0,0.008 -0.006,0.014 -0.0138,0.014 -0.008,0 -0.0301,0.0166 -0.0501,0.0369 -0.0199,0.0203 -0.0603,0.0549 -0.0896,0.0767 -0.0709,0.0529 -0.0867,0.0685 -0.0867,0.0855 0,0.008 -0.007,0.0143 -0.0147,0.0143 -0.008,0 -0.0265,0.015 -0.041,0.0334 -0.0144,0.0183 -0.0348,0.0334 -0.0453,0.0334 -0.0105,0 -0.0191,0.007 -0.0191,0.0143 0,0.015 -0.0326,0.0524 -0.0456,0.0524 -0.008,0 -0.0614,0.0432 -0.10342,0.0834 -0.0134,0.0128 -0.0304,0.0234 -0.0377,0.0234 -0.008,0 -0.0134,0.009 -0.0134,0.02 0,0.011 -0.006,0.02 -0.0142,0.02 -0.008,0 -0.0333,0.0196 -0.0567,0.0435 -0.0234,0.024 -0.0634,0.0591 -0.0892,0.0783 -0.0257,0.019 -0.0715,0.0594 -0.10183,0.0898 -0.0303,0.0304 -0.0618,0.0552 -0.0701,0.0552 -0.009,0 -0.0149,0.009 -0.0149,0.02 0,0.0116 -0.0121,0.02 -0.029,0.02 -0.0159,0 -0.0254,0.003 -0.0212,0.008 0.0105,0.0104 -0.14918,0.16569 -0.17027,0.16569 -0.003,0 -0.0458,0.039 -0.0938,0.0867 -0.048,0.0477 -0.0902,0.0867 -0.094,0.0867 -0.003,0 -0.0208,0.0135 -0.0379,0.03 -0.0509,0.0491 -0.0873,0.0767 -0.1013,0.0767 -0.007,0 -0.0132,0.007 -0.0132,0.0143 0,0.0167 -0.0332,0.0523 -0.049,0.0527 -0.006,1.5e-4 -0.0455,0.0332 -0.0877,0.0734 -0.0422,0.0403 -0.0795,0.0731 -0.0828,0.0731 -0.003,0 -0.0264,0.021 -0.0513,0.0467 -0.0249,0.0257 -0.0523,0.0468 -0.0607,0.0468 -0.009,0 -0.0154,0.006 -0.0154,0.0128 0,0.014 -0.074,0.0939 -0.087,0.0939 -0.005,0 -0.027,0.0179 -0.0505,0.04 -0.0235,0.022 -0.0476,0.0399 -0.0536,0.04 -0.0178,1e-4 -0.0486,0.0361 -0.039,0.0458 0.005,0.005 -0.003,0.0121 -0.019,0.0161 -0.0154,0.005 -0.0355,0.0218 -0.045,0.0395 -0.009,0.0177 -0.0253,0.0322 -0.035,0.0322 -0.01,0 -0.0177,0.006 -0.0177,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.019 0,0.0105 -0.005,0.0208 -0.0118,0.0234 -0.014,0.005 -0.11677,0.0878 -0.15108,0.12121 -0.0132,0.0128 -0.0271,0.0234 -0.0311,0.0234 -0.005,0 -0.0242,0.0165 -0.0452,0.0367 -0.0824,0.0796 -0.15021,0.13678 -0.1621,0.13678 -0.007,0 -0.0125,0.009 -0.0125,0.02 0,0.0109 -0.006,0.02 -0.0134,0.02 -0.008,0 -0.0489,0.0361 -0.0923,0.08 -0.0433,0.044 -0.0898,0.08 -0.10332,0.08 -0.0134,-1e-4 -0.0364,-0.0148 -0.0512,-0.0327 z m -9.30545,-1.00488 c -0.0234,-0.009 -0.0603,-0.0451 -0.0893,-0.0893 -0.0125,-0.0191 -0.0249,-0.0325 -0.0277,-0.0296 -0.007,0.007 -0.17819,-0.16457 -0.17819,-0.17825 0,-0.006 -0.0121,-0.0148 -0.0269,-0.0194 -0.0148,-0.005 -0.0235,-0.0141 -0.0192,-0.0209 0.005,-0.007 -0.002,-0.0124 -0.0124,-0.0124 -0.0111,0 -0.0306,-0.018 -0.0437,-0.04 -0.013,-0.022 -0.0324,-0.0401 -0.043,-0.0401 -0.0106,0 -0.016,-0.003 -0.012,-0.008 0.005,-0.005 -0.0301,-0.0459 -0.0761,-0.0929 -0.0459,-0.0471 -0.0808,-0.0882 -0.0776,-0.0915 0.003,-0.003 -0.007,-0.01 -0.0219,-0.0147 -0.026,-0.009 -0.0263,-0.0102 -0.005,-0.0312 0.0123,-0.0123 0.0267,-0.0225 0.0319,-0.0225 0.005,0 0.0347,-0.0165 0.0655,-0.0367 0.0308,-0.0203 0.0646,-0.0368 0.0748,-0.0368 0.0104,0 0.0189,-0.005 0.0189,-0.0121 0,-0.007 0.0331,-0.0257 0.0733,-0.0425 0.0403,-0.0167 0.0733,-0.0353 0.0733,-0.0413 0,-0.006 0.012,-0.0109 0.0267,-0.0109 0.0147,0 0.0267,-0.005 0.0268,-0.01 1.6e-4,-0.0154 0.0788,-0.0567 0.10774,-0.0567 0.0141,0 0.0256,-0.005 0.0256,-0.0112 0,-0.006 0.0331,-0.0251 0.0733,-0.0422 0.0403,-0.017 0.0733,-0.0361 0.0733,-0.0423 0,-0.006 0.012,-0.0112 0.0267,-0.0112 0.0147,0 0.0267,-0.005 0.0267,-0.0109 0,-0.006 0.0285,-0.0244 0.0633,-0.041 0.12095,-0.0574 0.19013,-0.0971 0.19013,-0.1094 0,-0.007 0.015,-0.0123 0.0334,-0.0123 0.0183,0 0.0334,-0.005 0.0334,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0264,-0.009 0.0733,-0.0374 0.0836,-0.051 0.003,-0.005 0.0233,-0.0131 0.0434,-0.0181 0.0201,-0.005 0.0366,-0.0145 0.0366,-0.0211 0,-0.007 0.009,-0.0119 0.0184,-0.0119 0.0101,0 0.0258,-0.009 0.035,-0.02 0.009,-0.0111 0.0271,-0.02 0.04,-0.02 0.0128,0 0.0309,-0.009 0.04,-0.02 0.009,-0.0109 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.025,-0.006 0.025,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0135,-0.0164 0.03,-0.0211 0.0568,-0.0163 0.11675,-0.0478 0.11675,-0.0613 0,-0.008 0.005,-0.0101 0.0122,-0.006 0.007,0.005 0.0203,4e-5 0.03,-0.009 0.01,-0.009 0.0477,-0.0313 0.0845,-0.049 0.0368,-0.0177 0.0697,-0.036 0.0733,-0.0405 0.003,-0.005 0.0487,-0.0296 0.10008,-0.0556 0.15823,-0.0803 0.20666,-0.10734 0.21435,-0.11976 0.005,-0.007 0.0221,-0.012 0.04,-0.012 0.0179,0 0.0326,-0.006 0.0326,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.012,-0.0119 0.0267,-0.0119 0.0147,0 0.0267,-0.005 0.0267,-0.0109 0,-0.006 0.0296,-0.0237 0.0656,-0.0395 0.0361,-0.0157 0.0694,-0.0349 0.0741,-0.0426 0.005,-0.008 0.0205,-0.0138 0.0349,-0.0138 0.0144,0 0.0298,-0.009 0.034,-0.02 0.005,-0.0111 0.0155,-0.02 0.0251,-0.02 0.01,0 0.0196,-0.005 0.0219,-0.01 0.006,-0.0136 0.14712,-0.0839 0.1679,-0.0836 0.009,9e-5 0.0167,-0.006 0.0167,-0.0128 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0133 0.0334,-0.0193 0,-0.006 0.0133,-0.0152 0.0293,-0.0203 0.0162,-0.005 0.0379,-0.0179 0.0484,-0.0283 0.0105,-0.0105 0.0272,-0.0158 0.0374,-0.012 0.0104,0.005 0.0183,-0.002 0.0183,-0.0126 0,-0.012 0.0129,-0.0196 0.0334,-0.0196 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.009,-0.0134 0.0191,-0.0134 0.0105,0 0.027,-0.008 0.0367,-0.0168 0.01,-0.009 0.0431,-0.0288 0.0743,-0.0434 0.0311,-0.0147 0.0566,-0.0311 0.0567,-0.0365 1e-4,-0.005 0.0119,-0.01 0.0262,-0.01 0.0143,0 0.0321,-0.0105 0.0396,-0.0234 0.008,-0.0128 0.0138,-0.0189 0.0141,-0.0135 5.2e-4,0.0149 0.1122,-0.032 0.1245,-0.0523 0.006,-0.01 0.0186,-0.0177 0.0283,-0.0177 0.01,0 0.0252,-0.009 0.0343,-0.02 0.009,-0.0111 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.0251,-0.005 0.0251,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.02,-0.0131 0.0111,0 0.02,-0.006 0.02,-0.0128 0,-0.007 0.0211,-0.0167 0.0468,-0.0215 0.0257,-0.005 0.0502,-0.0141 0.0542,-0.0206 0.005,-0.007 0.0234,-0.0119 0.0432,-0.0121 0.027,-1.6e-4 0.0311,-0.003 0.0166,-0.0125 -0.0151,-0.01 -0.0115,-0.0141 0.0167,-0.0212 0.0198,-0.005 0.0361,-0.0144 0.0361,-0.021 0,-0.007 0.0112,-0.0119 0.025,-0.0119 0.0138,0 0.0326,-0.009 0.0417,-0.02 0.009,-0.0111 0.0251,-0.02 0.0354,-0.02 0.0104,0 0.0222,-0.009 0.0264,-0.02 0.005,-0.0111 0.0235,-0.0201 0.043,-0.0203 0.0276,-1.5e-4 0.031,-0.002 0.0152,-0.0131 -0.0163,-0.0105 -0.0145,-0.013 0.009,-0.0131 0.0161,-10e-5 0.0326,-0.006 0.0368,-0.0125 0.005,-0.007 0.0256,-0.0163 0.0475,-0.0211 0.0219,-0.005 0.0399,-0.0143 0.0399,-0.0211 0,-0.007 0.0114,-0.0122 0.0253,-0.0122 0.0139,0 0.0287,-0.009 0.033,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.005 0.0271,-0.0105 0,-0.006 0.0196,-0.0198 0.0434,-0.0311 0.0239,-0.0113 0.0614,-0.03 0.0834,-0.0417 0.022,-0.0116 0.0626,-0.032 0.0901,-0.0453 0.0276,-0.0134 0.05,-0.0289 0.05,-0.0345 0,-0.006 0.0112,-0.0103 0.025,-0.0103 0.0137,0 0.0325,-0.009 0.0417,-0.02 0.009,-0.0111 0.0249,-0.02 0.035,-0.02 0.0101,0 0.0184,-0.006 0.0184,-0.0134 0,-0.008 0.0152,-0.0134 0.0338,-0.0134 0.0186,0 0.0372,-0.009 0.0414,-0.02 0.005,-0.0111 0.019,-0.02 0.033,-0.02 0.0138,0 0.0253,-0.005 0.0253,-0.0101 0,-0.005 0.0298,-0.0237 0.0662,-0.0404 0.0364,-0.0167 0.0697,-0.0362 0.0741,-0.0433 0.005,-0.007 0.0196,-0.0129 0.0339,-0.0129 0.0143,0 0.0259,-0.005 0.0261,-0.01 9e-5,-0.005 0.0256,-0.0219 0.0567,-0.0364 0.0311,-0.0145 0.0596,-0.0302 0.0632,-0.0348 0.003,-0.005 0.0306,-0.0187 0.06,-0.0317 0.0925,-0.0406 0.14655,-0.0669 0.15345,-0.0749 0.003,-0.005 0.0157,-0.0118 0.0267,-0.0168 0.0111,-0.005 0.038,-0.0212 0.06,-0.0359 0.022,-0.0147 0.0505,-0.0303 0.0633,-0.0348 0.0128,-0.005 0.0234,-0.0134 0.0234,-0.0199 0,-0.007 0.015,-0.0118 0.0334,-0.0118 0.0183,0 0.0334,-0.006 0.0334,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0122 0.0335,-0.017 5e-5,-0.005 0.027,-0.0204 0.06,-0.0349 0.033,-0.0144 0.0599,-0.0306 0.0599,-0.0358 0,-0.005 0.0133,-0.0136 0.0293,-0.0189 0.0162,-0.005 0.0377,-0.0177 0.0477,-0.0277 0.0101,-0.0101 0.0232,-0.0154 0.0291,-0.0117 0.006,0.003 0.0182,-0.002 0.0272,-0.0132 0.009,-0.011 0.0277,-0.0199 0.0416,-0.0199 0.0138,0 0.0251,-0.006 0.0251,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.008,-0.0134 0.0176,-0.0134 0.0277,0 0.12927,-0.0526 0.12927,-0.067 0,-0.007 0.0116,-0.013 0.026,-0.013 0.0142,0 0.0297,-0.006 0.0341,-0.0134 0.005,-0.008 0.0201,-0.0134 0.0346,-0.0134 0.0144,0 0.0298,-0.009 0.034,-0.02 0.005,-0.0109 0.019,-0.02 0.0329,-0.02 0.0139,0 0.0253,-0.006 0.0253,-0.0131 0,-0.007 0.0164,-0.0172 0.0366,-0.0222 0.0201,-0.005 0.0439,-0.0157 0.0529,-0.0236 0.047,-0.0416 0.059,-0.0477 0.0909,-0.0477 0.0191,0 0.0312,-0.006 0.027,-0.0126 -0.005,-0.007 0.0149,-0.0227 0.0426,-0.035 0.12252,-0.0547 0.14915,-0.0693 0.14115,-0.0773 -0.005,-0.005 0.008,-0.009 0.0275,-0.009 0.0198,0 0.0395,-0.009 0.0436,-0.02 0.005,-0.0111 0.0221,-0.02 0.0396,-0.02 0.0176,0 0.0319,-0.005 0.0319,-0.0123 0,-0.0138 0.0751,-0.0544 0.10071,-0.0544 0.009,0 0.0199,-0.009 0.0242,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.006 0.0271,-0.0134 0,-0.0225 0.0842,-0.013 0.12677,0.0141 0.022,0.0141 0.049,0.0293 0.0601,0.0339 0.0766,0.0315 0.10674,0.0483 0.10674,0.0594 0,0.007 0.015,0.0128 0.0334,0.0128 0.0183,0 0.0334,0.005 0.0334,0.0108 0,0.006 0.0329,0.0254 0.0733,0.0434 0.0403,0.0179 0.0733,0.0371 0.0734,0.0426 1e-4,0.005 0.009,0.01 0.0202,0.01 0.0111,0 0.02,0.007 0.02,0.0156 0,0.009 0.003,0.0118 0.009,0.008 0.006,-0.006 0.16717,0.0673 0.20405,0.0932 0.0102,0.007 -0.043,0.0574 -0.061,0.0574 -0.01,0 -0.018,0.005 -0.0181,0.01 -5e-5,0.005 -0.027,0.0219 -0.0601,0.0364 -0.033,0.0144 -0.06,0.031 -0.06,0.0367 0,0.006 -0.01,0.0104 -0.0214,0.0104 -0.0118,0 -0.0433,0.0167 -0.0701,0.037 -0.0267,0.0204 -0.0711,0.0474 -0.0986,0.0598 -0.0276,0.0125 -0.05,0.0289 -0.05,0.0363 0,0.008 -0.015,0.0136 -0.0334,0.0136 -0.0183,0 -0.0334,0.006 -0.0334,0.0134 0,0.008 -0.009,0.0134 -0.0184,0.0134 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.0111 -0.0281,0.02 -0.0421,0.02 -0.0141,0 -0.0289,0.009 -0.0332,0.0199 -0.005,0.0109 -0.0127,0.0169 -0.0187,0.0131 -0.006,-0.003 -0.0173,0.003 -0.0249,0.0168 -0.008,0.013 -0.014,0.0189 -0.0141,0.013 -2e-4,-0.006 -0.026,0.006 -0.0571,0.0267 -0.0312,0.0206 -0.0807,0.0493 -0.11007,0.0639 -0.0293,0.0147 -0.0613,0.0342 -0.071,0.0434 -0.01,0.009 -0.0262,0.0167 -0.0367,0.0167 -0.0105,0 -0.0191,0.005 -0.0191,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.012,0.0131 -0.0267,0.0131 -0.0147,0 -0.0267,0.005 -0.0267,0.0113 0,0.006 -0.0255,0.0225 -0.0567,0.0361 -0.0312,0.0137 -0.10867,0.0566 -0.17216,0.0954 -0.0635,0.0388 -0.12204,0.0705 -0.13011,0.0705 -0.008,0 -0.0147,0.005 -0.0147,0.0105 0,0.006 -0.039,0.0311 -0.0867,0.0562 -0.0477,0.0253 -0.0867,0.0506 -0.0867,0.0562 0,0.006 -0.009,0.0105 -0.0184,0.0105 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.011 -0.0279,0.02 -0.0417,0.02 -0.0138,0 -0.0251,0.006 -0.0251,0.0134 0,0.008 -0.009,0.0134 -0.0184,0.0134 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.0111 -0.0279,0.02 -0.0417,0.02 -0.0138,0 -0.025,0.005 -0.025,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.02,0.0131 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.0122,0.0119 -0.0271,0.0119 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.0221,0.02 -0.0396,0.02 -0.0176,0 -0.0319,0.005 -0.0319,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0254,0.009 -0.0727,0.037 -0.0834,0.0502 -0.003,0.005 -0.0202,0.0124 -0.0367,0.0176 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.006,0.0125 -0.0131,0.0125 -0.007,0 -0.0523,0.0241 -0.10031,0.0533 -0.0479,0.0293 -0.0991,0.0533 -0.11365,0.0533 -0.0145,0 -0.0264,0.006 -0.0264,0.0125 0,0.007 -0.0133,0.0166 -0.0293,0.0218 -0.0162,0.005 -0.0376,0.0176 -0.0477,0.0276 -0.01,0.01 -0.0268,0.0183 -0.0374,0.0183 -0.0105,0 -0.019,0.006 -0.019,0.0134 0,0.008 -0.009,0.0134 -0.0186,0.0134 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.0121,0.0134 -0.0271,0.0134 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.0187,0.02 -0.0322,0.02 -0.0135,0 -0.0298,0.009 -0.0361,0.0206 -0.006,0.0114 -0.0247,0.0216 -0.0406,0.0229 -0.016,0.002 -0.0291,0.007 -0.0291,0.0127 0,0.006 -0.009,0.0105 -0.0186,0.0105 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.011 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.005 -0.0271,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.012,0.0131 -0.0267,0.0131 -0.0147,0 -0.0267,0.006 -0.0267,0.013 0,0.007 -0.0255,0.0254 -0.0567,0.0406 -0.10769,0.0524 -0.23019,0.12272 -0.23463,0.13459 -0.002,0.007 -0.0125,0.0119 -0.0225,0.0119 -0.01,0 -0.0259,0.008 -0.0358,0.0167 -0.01,0.009 -0.0418,0.0286 -0.071,0.0432 -0.0293,0.0147 -0.0562,0.0299 -0.0598,0.0341 -0.0157,0.0177 -0.16149,0.0928 -0.17994,0.0928 -0.0111,0 -0.0202,0.005 -0.0202,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.0206,0.0134 -0.0113,0 -0.0368,0.0111 -0.0567,0.0246 -0.0199,0.0135 -0.0632,0.042 -0.0962,0.0633 -0.0331,0.0213 -0.063,0.0419 -0.0667,0.0459 -0.003,0.005 -0.0534,0.0306 -0.1105,0.0593 -0.0572,0.0286 -0.11039,0.06 -0.11839,0.0697 -0.008,0.01 -0.0262,0.0176 -0.0404,0.0176 -0.0142,0 -0.0225,0.005 -0.0185,0.0119 0.005,0.007 -0.009,0.0158 -0.0275,0.0206 -0.0192,0.005 -0.0437,0.0176 -0.0544,0.0283 -0.0107,0.0107 -0.025,0.0161 -0.0318,0.0119 -0.007,-0.005 -0.0122,0.002 -0.0122,0.0133 0,0.0115 -0.0105,0.0212 -0.0234,0.0218 -0.0128,5.1e-4 -0.0413,0.0122 -0.0633,0.026 -0.022,0.0138 -0.067,0.0388 -0.10008,0.0558 -0.0331,0.0169 -0.063,0.0345 -0.0667,0.039 -0.003,0.005 -0.0202,0.0124 -0.0367,0.0177 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.009,0.0125 -0.02,0.0125 -0.0111,0 -0.02,0.005 -0.02,0.0113 0,0.006 -0.027,0.0225 -0.0601,0.0361 -0.0331,0.0136 -0.06,0.0293 -0.06,0.0348 0,0.005 -0.0225,0.0198 -0.05,0.0318 -0.0276,0.012 -0.0591,0.0273 -0.0701,0.0341 -0.0111,0.007 -0.029,0.0163 -0.0401,0.0213 -0.011,0.005 -0.023,0.0124 -0.0267,0.0166 -0.003,0.005 -0.0338,0.0209 -0.0668,0.037 -0.0331,0.0161 -0.0734,0.0416 -0.0895,0.0567 -0.0162,0.0151 -0.0383,0.0274 -0.0491,0.0274 -0.0109,0 -0.0237,0.006 -0.0285,0.0141 -0.005,0.008 -0.0217,0.0198 -0.0375,0.0267 -0.0158,0.007 -0.0405,0.02 -0.0549,0.0292 -0.0809,0.0516 -0.12902,0.0768 -0.14702,0.0768 -0.0111,0 -0.0203,0.005 -0.0203,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0145 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0262,0.0166 -0.0367,0.0166 -0.0105,0 -0.019,0.005 -0.019,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0141 -0.0234,0.0206 0,0.007 -0.0144,0.0118 -0.0319,0.0118 -0.0176,0 -0.0354,0.009 -0.0396,0.02 -0.005,0.011 -0.0163,0.02 -0.027,0.02 -0.0106,0 -0.0213,0.005 -0.0238,0.01 -0.002,0.005 -0.0328,0.0239 -0.0675,0.0409 -0.0347,0.017 -0.0665,0.0364 -0.0708,0.0434 -0.005,0.007 -0.0194,0.0126 -0.0336,0.0126 -0.0142,0 -0.026,0.006 -0.026,0.0127 0,0.007 -0.0225,0.0233 -0.05,0.0362 -0.0275,0.0129 -0.058,0.0312 -0.0676,0.0406 -0.01,0.009 -0.0265,0.0172 -0.0375,0.0172 -0.0109,0 -0.0162,0.006 -0.0115,0.0134 0.005,0.008 -0.003,0.0134 -0.0177,0.0134 -0.0142,0 -0.0333,0.009 -0.0425,0.02 -0.009,0.0111 -0.0309,0.02 -0.0483,0.02 -0.0175,0 -0.0318,0.005 -0.0318,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0276,0.0144 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0262,0.0167 -0.0368,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.0118,0.0134 -0.026,0.0134 -0.0143,0 -0.0308,0.008 -0.0367,0.0177 -0.0134,0.0221 -0.0449,0.0338 -0.069,0.0254 z m 5.35576,-1.75103 c -0.008,-0.007 -0.0269,-0.0243 -0.0433,-0.0394 -0.0165,-0.0151 -0.0361,-0.0274 -0.0434,-0.0274 -0.008,0 -0.0134,-0.005 -0.0134,-0.0119 0,-0.007 -0.015,-0.0157 -0.0334,-0.0203 -0.0183,-0.005 -0.0334,-0.0143 -0.0334,-0.0215 0,-0.007 -0.009,-0.0131 -0.02,-0.0131 -0.0111,0 -0.02,-0.005 -0.02,-0.0119 0,-0.007 -0.015,-0.0157 -0.0334,-0.0203 -0.0183,-0.005 -0.0334,-0.0143 -0.0334,-0.0215 0,-0.007 -0.006,-0.0131 -0.014,-0.0131 -0.0144,0 -0.029,-0.0108 -0.0901,-0.0667 -0.02,-0.0183 -0.0589,-0.0448 -0.0863,-0.0588 -0.0274,-0.014 -0.0499,-0.0298 -0.0499,-0.0351 0,-0.005 -0.0105,-0.0132 -0.0234,-0.0176 -0.0335,-0.0113 -0.0721,-0.0374 -0.10601,-0.072 -0.0163,-0.0165 -0.0367,-0.03 -0.0454,-0.03 -0.016,0 -0.051,-0.0383 -0.0517,-0.0567 -2.4e-4,-0.005 0.0115,-0.01 0.0263,-0.01 0.0147,0 0.0267,-0.005 0.0267,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.012,-0.0131 0.0267,-0.0131 0.0147,0 0.0267,-0.006 0.0267,-0.013 0,-0.007 0.0267,-0.0251 0.0592,-0.04 0.0326,-0.0148 0.063,-0.0329 0.0675,-0.0403 0.005,-0.008 0.0199,-0.0134 0.0341,-0.0134 0.0142,0 0.026,-0.006 0.026,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0133,-0.0166 0.0293,-0.0218 0.0162,-0.005 0.0376,-0.0176 0.0477,-0.0276 0.01,-0.01 0.0268,-0.0183 0.0374,-0.0183 0.0105,0 0.019,-0.005 0.019,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0227,-0.008 0.07,-0.0352 0.0834,-0.0486 0.003,-0.003 0.0352,-0.0201 0.0701,-0.0366 0.0348,-0.0164 0.0633,-0.036 0.0633,-0.0434 0,-0.008 0.015,-0.0135 0.0334,-0.0135 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0184,-0.02 0.0315,-0.02 0.0131,0 0.0269,-0.008 0.0306,-0.0178 0.003,-0.01 0.0419,-0.0348 0.0849,-0.0556 0.0428,-0.0208 0.081,-0.0413 0.0847,-0.0455 0.0113,-0.0132 0.0588,-0.0414 0.0834,-0.0497 0.0128,-0.005 0.0234,-0.0127 0.0234,-0.0186 0,-0.006 0.0185,-0.0148 0.041,-0.0197 0.0225,-0.005 0.0376,-0.0125 0.0335,-0.0165 -0.009,-0.009 0.0239,-0.0284 0.11572,-0.0707 0.0348,-0.0161 0.0633,-0.0345 0.0633,-0.0409 0,-0.007 0.0105,-0.0118 0.0234,-0.0118 0.0129,0 0.0397,-0.015 0.0595,-0.0334 0.0198,-0.0183 0.0445,-0.0334 0.055,-0.0334 0.0105,0 0.0264,-0.009 0.0356,-0.02 0.009,-0.011 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.0251,-0.006 0.0251,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.0109 0.0209,-0.02 0.037,-0.02 0.0161,0 0.0255,-0.003 0.0208,-0.009 -0.005,-0.005 0.0169,-0.021 0.0479,-0.0361 0.031,-0.0151 0.0643,-0.0354 0.0739,-0.0448 0.01,-0.009 0.0262,-0.0172 0.0368,-0.0172 0.0105,0 0.0191,-0.006 0.0191,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.015,-0.0119 0.0334,-0.0119 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.0241,-0.0238 0.0533,-0.0364 0.0293,-0.0128 0.0533,-0.0307 0.0533,-0.04 0,-0.009 0.015,-0.0169 0.0334,-0.0169 0.0183,0 0.0334,-0.006 0.0334,-0.0126 0,-0.0122 0.0222,-0.0264 0.12009,-0.0768 0.0257,-0.0132 0.0497,-0.027 0.0533,-0.0306 0.01,-0.01 0.0567,-0.0387 0.08,-0.0494 0.0111,-0.005 0.0279,-0.017 0.0377,-0.0267 0.01,-0.01 0.0262,-0.0175 0.0367,-0.0175 0.0105,0 0.0191,-0.005 0.0191,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.012,-0.0131 0.0267,-0.0131 0.0147,0 0.0267,-0.005 0.0267,-0.0118 0,-0.007 0.0106,-0.0154 0.0234,-0.0199 0.0128,-0.005 0.0413,-0.0202 0.0633,-0.0351 0.022,-0.0148 0.0505,-0.0306 0.0633,-0.0351 0.0128,-0.005 0.0234,-0.0128 0.0234,-0.0186 0,-0.0167 0.0956,0.0171 0.10295,0.0364 0.003,0.009 0.0151,0.0173 0.0255,0.0173 0.0167,0 0.059,0.0222 0.14558,0.0765 0.0144,0.009 0.0547,0.0293 0.0896,0.045 0.0348,0.0157 0.0633,0.0339 0.0633,0.0403 0,0.007 0.012,0.0118 0.0267,0.0118 0.0147,0 0.0267,0.006 0.0267,0.0131 0,0.007 0.015,0.0169 0.0334,0.0215 0.0183,0.005 0.0334,0.0137 0.0334,0.0203 0,0.007 0.0122,0.0119 0.0271,0.0119 0.0149,0 0.0306,0.009 0.0348,0.02 0.005,0.0111 0.0221,0.02 0.0397,0.02 0.0176,0 0.0319,0.005 0.0319,0.0119 0,0.007 0.0105,0.0156 0.0234,0.0202 0.0313,0.0114 0.0845,0.0429 0.0876,0.052 0.003,0.0101 -0.0842,0.0685 -0.14652,0.0978 -0.0282,0.0132 -0.0584,0.0303 -0.0674,0.0378 -0.0345,0.0295 -0.0523,0.0406 -0.0711,0.0449 -0.0107,0.002 -0.0243,0.0129 -0.0302,0.0234 -0.006,0.0105 -0.0168,0.019 -0.0243,0.019 -0.008,0 -0.022,0.008 -0.0322,0.0167 -0.0335,0.0301 -0.053,0.0431 -0.10329,0.0688 -0.0273,0.014 -0.0497,0.0298 -0.0497,0.0352 0,0.005 -0.0105,0.0131 -0.0234,0.0176 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.019,0.006 -0.019,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.013 -0.0334,0.0187 0,0.006 -0.0105,0.0139 -0.0234,0.0182 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0372,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0109 -0.009,0.02 -0.0203,0.02 -0.0111,0 -0.0441,0.015 -0.0731,0.0334 -0.0291,0.0183 -0.0584,0.0334 -0.0653,0.0334 -0.007,0 -0.0144,0.005 -0.017,0.01 -0.008,0.0179 -0.13146,0.0967 -0.15162,0.0967 -0.0108,0 -0.0196,0.007 -0.0196,0.0152 0,0.0207 -0.0413,0.0506 -0.0706,0.0511 -0.0132,2.6e-4 -0.0565,0.0244 -0.0962,0.0538 -0.0397,0.0293 -0.0815,0.0533 -0.0928,0.0533 -0.0113,0 -0.0206,0.009 -0.0206,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0141 0,0.008 -0.006,0.0106 -0.0128,0.006 -0.007,-0.005 -0.0246,0.003 -0.039,0.0182 -0.0143,0.0143 -0.0284,0.0237 -0.0312,0.0208 -0.002,-0.002 -0.0248,0.0127 -0.0487,0.0345 -0.0239,0.0219 -0.0513,0.0397 -0.061,0.0397 -0.01,0 -0.025,0.009 -0.0342,0.02 -0.009,0.0111 -0.0249,0.02 -0.035,0.02 -0.0101,0 -0.0184,0.006 -0.0184,0.0134 0,0.008 -0.009,0.0134 -0.0205,0.0134 -0.0112,0 -0.0239,0.009 -0.0281,0.02 -0.005,0.0111 -0.019,0.02 -0.033,0.02 -0.0138,0 -0.0253,0.006 -0.0253,0.0134 0,0.008 -0.0114,0.0134 -0.0253,0.0134 -0.0139,0 -0.0286,0.009 -0.0329,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0153,0 -0.0271,0.009 -0.0271,0.02 0,0.011 -0.009,0.02 -0.02,0.02 -0.011,0 -0.02,0.005 -0.02,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0141 -0.0234,0.0206 0,0.007 -0.009,0.0118 -0.0191,0.0118 -0.0105,0 -0.0272,0.008 -0.0373,0.0167 -0.0265,0.0243 -0.067,0.0497 -0.0938,0.0587 -0.0128,0.005 -0.0234,0.0124 -0.0234,0.018 0,0.006 -0.0105,0.0138 -0.0234,0.018 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.005 -0.02,0.0119 0,0.007 -0.0155,0.0157 -0.0345,0.0206 -0.0325,0.009 -0.0463,0.0169 -0.0918,0.0576 -0.0102,0.009 -0.0243,0.0167 -0.0312,0.0167 -0.007,0 -0.0201,0.009 -0.0293,0.02 -0.0183,0.0219 -0.0657,0.0256 -0.0867,0.007 z m 2.92224,-0.36528 c -0.0147,-0.009 -0.0297,-0.0181 -0.0334,-0.0217 -0.0201,-0.0201 -0.0943,-0.0667 -0.10621,-0.0667 -0.008,0 -0.0138,-0.006 -0.0138,-0.0134 0,-0.008 -0.0114,-0.0134 -0.0253,-0.0134 -0.0139,0 -0.0286,-0.009 -0.0329,-0.02 -0.005,-0.0111 -0.0199,-0.02 -0.0348,-0.02 -0.0149,0 -0.0271,-0.005 -0.0271,-0.0118 0,-0.007 -0.0105,-0.0154 -0.0234,-0.0196 -0.0254,-0.009 -0.0727,-0.037 -0.0834,-0.0502 -0.003,-0.005 -0.0203,-0.0124 -0.0367,-0.0176 -0.0165,-0.005 -0.03,-0.015 -0.03,-0.0219 0,-0.007 -0.009,-0.0124 -0.02,-0.0124 -0.0111,0 -0.02,-0.006 -0.02,-0.0125 0,-0.007 -0.0135,-0.0167 -0.03,-0.0219 -0.0165,-0.005 -0.0331,-0.013 -0.0367,-0.0176 -0.0107,-0.0132 -0.0581,-0.0417 -0.0834,-0.0502 -0.0128,-0.005 -0.0234,-0.0131 -0.0234,-0.0196 0,-0.007 -0.009,-0.0118 -0.019,-0.0118 -0.0105,0 -0.0272,-0.008 -0.0372,-0.0167 -0.0265,-0.0243 -0.0671,-0.0497 -0.0938,-0.0587 -0.0128,-0.005 -0.0234,-0.0107 -0.0234,-0.0141 0,-0.0144 0.0617,-0.0564 0.0834,-0.0568 0.0128,-2.6e-4 0.0234,-0.007 0.0234,-0.0138 0,-0.008 0.009,-0.0134 0.019,-0.0134 0.0105,0 0.0208,-0.005 0.0234,-0.0106 0.002,-0.006 0.027,-0.0243 0.0545,-0.0411 0.0275,-0.0168 0.05,-0.0361 0.05,-0.0428 0,-0.007 0.0128,-0.0122 0.0285,-0.0122 0.0157,0 0.0413,-0.0139 0.0571,-0.0309 0.0157,-0.017 0.0494,-0.0425 0.075,-0.0567 0.0255,-0.0142 0.0493,-0.0293 0.053,-0.0335 0.003,-0.005 0.0157,-0.0119 0.0267,-0.0172 0.0521,-0.0249 0.0791,-0.045 0.071,-0.0531 -0.005,-0.005 7.6e-4,-0.009 0.0125,-0.009 0.0118,0 0.0294,-0.008 0.0395,-0.0167 0.0265,-0.0243 0.067,-0.0497 0.0938,-0.0587 0.0128,-0.005 0.0234,-0.0125 0.0234,-0.0182 0,-0.006 0.015,-0.0141 0.0334,-0.0187 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.02,-0.0131 0.0109,0 0.02,-0.005 0.02,-0.0103 0,-0.006 0.0225,-0.0222 0.05,-0.0366 0.0276,-0.0144 0.0549,-0.0338 0.0607,-0.0431 0.006,-0.009 0.0193,-0.0168 0.03,-0.0168 0.0106,0 0.0193,-0.006 0.0193,-0.0134 0,-0.008 0.009,-0.0134 0.0204,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0221,-0.02 0.0396,-0.02 0.0176,0 0.0319,-0.005 0.0319,-0.0105 0,-0.0128 0.0828,-0.0636 0.0967,-0.0594 0.005,0.002 0.01,-0.002 0.01,-0.009 0,-0.007 0.018,-0.005 0.04,0.002 0.022,0.009 0.0401,0.0193 0.0401,0.0244 0,0.005 0.018,0.0134 0.04,0.0182 0.022,0.005 0.04,0.0143 0.04,0.0211 0,0.007 0.009,0.0122 0.0184,0.0122 0.0101,0 0.0258,0.009 0.035,0.02 0.009,0.0109 0.0279,0.02 0.0417,0.02 0.0137,0 0.025,0.006 0.025,0.0134 0,0.008 0.0152,0.0134 0.0339,0.0134 0.0186,0 0.0373,0.009 0.0414,0.02 0.005,0.0111 0.019,0.02 0.033,0.02 0.0139,0 0.0253,0.005 0.0253,0.0118 0,0.007 0.0105,0.0156 0.0234,0.0202 0.0316,0.0114 0.0823,0.0416 0.10054,0.0598 0.009,0.009 0.0257,0.0149 0.039,0.0149 0.0132,0 0.0241,0.006 0.0241,0.0134 0,0.008 0.0136,0.0134 0.0301,0.0134 0.0166,0 0.0463,0.015 0.0661,0.0334 0.0198,0.0183 0.0484,0.0334 0.0637,0.0334 0.0152,0 0.0312,0.009 0.0354,0.02 0.005,0.0111 0.019,0.02 0.0329,0.02 0.0485,0 0.0281,0.0473 -0.0338,0.0782 -0.0326,0.0162 -0.073,0.0442 -0.0901,0.0622 -0.017,0.018 -0.0371,0.0328 -0.0445,0.0328 -0.008,1.6e-4 -0.0296,0.0153 -0.0495,0.0335 -0.0198,0.0183 -0.0422,0.0334 -0.0499,0.0334 -0.008,0 -0.014,0.005 -0.014,0.0102 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0144 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0257,0.0167 -0.0357,0.0167 -0.01,0 -0.029,0.0138 -0.0424,0.0307 -0.0134,0.0169 -0.0333,0.0352 -0.0444,0.0406 -0.0484,0.0237 -0.0856,0.0484 -0.0733,0.0487 0.008,9e-5 -0.0121,0.0128 -0.0434,0.0283 -0.0312,0.0154 -0.0567,0.0335 -0.0567,0.04 0,0.007 -0.009,0.012 -0.02,0.012 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0112 -0.0119,0.02 -0.0271,0.02 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.005 -0.0186,0.0112 0,0.006 -0.0211,0.0225 -0.0468,0.0364 -0.0257,0.0137 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0103 -0.0186,0.0103 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0192,0.02 -0.0334,0.02 -0.0141,0 -0.0293,0.009 -0.0336,0.0209 -0.005,0.0115 -0.0141,0.0171 -0.0214,0.0126 -0.008,-0.005 -0.0134,5.2e-4 -0.0134,0.0111 0,0.0106 -0.015,0.0232 -0.0334,0.0277 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.0189,0.0131 -0.0105,0 -0.0209,0.005 -0.0234,0.0118 -0.002,0.007 -0.0164,0.005 -0.0312,-0.003 z"
+               id="path56252-8"
+               inkscape:connector-curvature="0" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.4313232;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -35.94889,35.9093 h -5.30885 v -5.25781 z"
+               id="rect55701-5-4"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:4.9705863px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.13810469"
+               x="-59.061882"
+               y="42.855503"
+               id="text55709-6-5"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-0-0"
+                 x="-59.061882"
+                 y="42.855503"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.13810469">PARQUET</tspan></text>
+            <path
+               style="fill:#6995f4;fill-opacity:1;stroke:none;stroke-width:0.31363189;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -60.81183,30.65149 h 19.55409 l -0.003,2.44581 h -19.55143 z"
+               id="rect55679-1-5-3"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+          </g>
+        </g>
+        <g
+           id="g13072-6"
+           transform="translate(0,-0.47244895)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,141.24395,-103.38038)"
+             id="g5002-1"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -296.22335,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-7-0"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -263.44559,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-7-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-292.76242"
+               y="155.93517"
+               id="text55709-5-8-1-3"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-21-2"
+                 x="-292.76242"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">HTML</tspan></text>
+            <path
+               style="fill:#d45500;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -295.98983,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-0-0"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               id="text55879-4-6"
+               y="171.957"
+               x="-291.82764"
+               style="font-style:normal;font-weight:normal;font-size:14.25364685px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#d45500;fill-opacity:1;stroke:none;stroke-width:0.39602885"
+               xml:space="preserve"><tspan
+                 style="fill:#d45500;fill-opacity:1;stroke-width:0.39602885"
+                 y="171.957"
+                 x="-291.82764"
+                 id="tspan55877-8-1"
+                 sodipodi:role="line">&lt;&gt;</tspan></text>
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,132.71412,-103.38038)"
+             id="g4993-5"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -247.76738,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-7-6-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -214.98962,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-7-7-4"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-244.30646"
+               y="155.93517"
+               id="text55709-5-8-1-5-7"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-21-3-6"
+                 x="-244.30646"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">HDF5</tspan></text>
+            <path
+               style="fill:#2cab28;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -247.53386,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-0-5-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <path
+               style="fill:#2cab28;fill-opacity:1;stroke-width:0.01142396"
+               d="m -228.2809,172.80263 c -0.202,-0.0438 -0.37838,-0.14187 -0.50896,-0.28291 -0.0769,-0.083 -0.11316,-0.13774 -0.16538,-0.24931 -0.0851,-0.18181 -0.0809,-0.13496 -0.0871,-0.97682 l -0.006,-0.74885 -0.0888,0.10126 c -0.12247,0.13959 -0.41022,0.41948 -0.55874,0.54347 -0.84282,0.70368 -1.86652,1.17525 -3.06769,1.41315 -0.52549,0.10407 -1.00462,0.15148 -1.63022,0.16129 -0.53258,0.008 -0.64972,-0.005 -0.84254,-0.098 -0.26642,-0.12809 -0.46447,-0.38731 -0.51956,-0.68003 -0.0177,-0.094 -0.0151,-0.2674 0.005,-0.36491 0.076,-0.36089 0.3608,-0.6475 0.72486,-0.72956 0.065,-0.0147 0.1512,-0.0179 0.54992,-0.0205 0.47948,-0.004 0.60576,-0.009 0.95596,-0.0474 1.30447,-0.14149 2.34564,-0.6184 2.99398,-1.37141 0.39145,-0.45465 0.62298,-0.97115 0.70781,-1.579 0.0247,-0.17684 0.0247,-0.57128 -10e-5,-0.74923 -0.0492,-0.3547 -0.13728,-0.63945 -0.29551,-0.95596 -0.48894,-0.97817 -1.47794,-1.6518 -2.8511,-1.94196 -0.23021,-0.0487 -0.4358,-0.0806 -0.76323,-0.11859 l -0.0539,-0.006 v 2.13841 2.13841 h -2.71369 -2.71369 l -8e-5,1.75516 c -5e-5,1.16801 -0.004,1.77751 -0.0108,1.82197 -0.0153,0.0952 -0.0615,0.22303 -0.11373,0.31501 -0.0524,0.0922 -0.18048,0.23228 -0.27212,0.29768 -0.18509,0.13208 -0.43523,0.19653 -0.65358,0.1684 -0.27855,-0.0359 -0.52289,-0.18249 -0.67666,-0.40601 -0.0501,-0.0728 -0.11087,-0.20816 -0.13607,-0.30313 l -0.0231,-0.0874 v -4.48683 -4.48684 l 0.0237,-0.0867 c 0.0906,-0.33162 0.32687,-0.5747 0.65683,-0.67579 0.0816,-0.025 0.10651,-0.0278 0.25488,-0.0283 0.19819,-4.8e-4 0.25702,0.0121 0.42144,0.0916 0.27365,0.13243 0.46516,0.38659 0.51858,0.68819 0.007,0.041 0.0108,0.6273 0.0108,1.79114 v 1.73014 h 1.76777 1.76779 l 0.003,-1.76544 0.003,-1.76544 0.0236,-0.0822 c 0.0312,-0.10896 0.10123,-0.2515 0.16408,-0.33431 0.17162,-0.22611 0.44991,-0.37003 0.71444,-0.36946 0.0465,1e-4 2.81826,0.004 6.15949,0.008 l 6.07496,0.007 0.0722,0.0225 c 0.18274,0.057 0.31346,0.13732 0.43953,0.27005 0.14446,0.15211 0.23163,0.34134 0.25287,0.54896 0.0476,0.46593 -0.24024,0.88175 -0.70295,1.01524 l -0.0822,0.0237 -2.29996,0.003 -2.29995,0.003 v 1.2331 1.23311 l 1.66778,0.003 1.66779,0.003 0.0925,0.0288 c 0.17469,0.0546 0.32388,0.148 0.44081,0.27613 0.1602,0.17551 0.24276,0.39165 0.24276,0.63553 0,0.24752 -0.0819,0.46012 -0.24645,0.63968 -0.11647,0.12711 -0.26043,0.21641 -0.43936,0.27253 l -0.0903,0.0283 -1.66756,0.003 -1.66757,0.003 -0.003,1.80634 -0.003,1.80635 -0.0288,0.0925 c -0.11257,0.36055 -0.37513,0.60464 -0.72877,0.67751 -0.0965,0.0199 -0.27614,0.0201 -0.36676,4.9e-4 z"
+               id="path4972-6"
+               inkscape:connector-curvature="0" />
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,389.89266,-103.38038)"
+             id="g57106-9"
+             style="stroke-width:1.11137545">
+            <g
+               id="g56951-3"
+               style="stroke-width:1.11137545">
+              <path
+                 sodipodi:nodetypes="cccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-4-7"
+                 d="m -547.10925,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+                 style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <path
+                 sodipodi:nodetypes="cccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55701-3-4"
+                 d="m -514.33149,144.95855 h -6.949 v -6.88219 z"
+                 style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686" />
+              <text
+                 id="text55709-5-5"
+                 y="155.93515"
+                 x="-543.47522"
+                 style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+                 xml:space="preserve"><tspan
+                   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141"
+                   y="155.93515"
+                   x="-543.47522"
+                   id="tspan55707-4-2"
+                   sodipodi:role="line">JSON</tspan></text>
+              <path
+                 sodipodi:nodetypes="ccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-1-7-5"
+                 d="m -546.87573,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+                 style="fill:#7c4515;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <text
+                 xml:space="preserve"
+                 style="font-style:normal;font-weight:normal;font-size:14.25364685px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#7c4515;fill-opacity:1;stroke:none;stroke-width:0.39602885"
+                 x="-539.08789"
+                 y="171.74121"
+                 id="text55879-47"><tspan
+                   sodipodi:role="line"
+                   id="tspan55877-4"
+                   x="-539.08789"
+                   y="171.74121"
+                   style="fill:#7c4515;fill-opacity:1;stroke-width:0.39602885">{}</tspan></text>
+            </g>
+          </g>
+        </g>
+        <g
+           id="g13116-4"
+           transform="translate(0,-4.3467227)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,209.56487,-64.753609)"
+             id="g56797-3"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -385.65173,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-5-3-4-0"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <g
+               id="g56653-7"
+               transform="matrix(0.12939252,0,0,0.12939252,-377.59411,159.20773)"
+               style="stroke-width:1.11137545">
+              <g
+                 style="display:none;stroke-width:1.11137545"
+                 id="Placement_ONLY-8" />
+              <g
+                 id="BASE-6"
+                 style="stroke-width:1.11137545">
+                <linearGradient
+                   y2="120.7894"
+                   x2="63.999699"
+                   y1="7.0335999"
+                   x1="63.999699"
+                   gradientUnits="userSpaceOnUse"
+                   id="linearGradient13523">
+                  <stop
+                     id="stop11214-8"
+                     style="stop-color:#4387FD"
+                     offset="0" />
+                  <stop
+                     id="stop11216-8"
+                     style="stop-color:#4683EA"
+                     offset="1" />
+                </linearGradient>
+                <path
+                   id="path56591-4"
+                   d="M 27.7906,115.2166 1.54,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7906,12.7831 c 2.0541,-3.5578 5.8503,-5.7495 9.9585,-5.7495 h 52.5012 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2506,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2506,45.4672 c -2.0541,3.5578 -5.8503,5.7495 -9.9585,5.7495 H 37.7491 c -4.1082,1e-4 -7.9043,-2.1916 -9.9585,-5.7494 z"
+                   style="fill:url(#SVGID_1_-2);stroke-width:1.11137545"
+                   inkscape:connector-curvature="0" />
+              </g>
+              <g
+                 id="shadow-3"
+                 style="stroke-width:1.11137545">
+                <g
+                   id="g56602-1"
+                   style="stroke-width:1.11137545">
+                  <defs
+                     id="defs56595-4">
+                    <path
+                       d="M 27.7908,115.2166 1.5403,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7908,12.7831 C 29.8449,9.2253 33.641,7.0336 37.7493,7.0336 h 52.501 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2505,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2505,45.4672 c -2.0541,3.5578 -5.8502,5.7495 -9.9585,5.7495 h -52.501 c -4.1083,1e-4 -7.9044,-2.1916 -9.9585,-5.7494 z"
+                       id="path13527"
+                       inkscape:connector-curvature="0" />
+                  </defs>
+                  <clipPath
+                     id="clipPath11226-2">
+                    <use
+                       id="use11224-0"
+                       style="overflow:visible"
+                       xlink:href="#SVGID_6_"
+                       x="0"
+                       y="0"
+                       width="100%"
+                       height="100%" />
+                  </clipPath>
+                  <polygon
+                     id="polygon56600-6"
+                     clip-path="url(#SVGID_2_-9)"
+                     points="88.8747,121.6665 97.5622,121.2811 119.2289,86.4791 80.6247,47.8749 63.9997,43.4256 49.0668,48.9745 43.3004,63.9999 47.9372,80.729 "
+                     style="opacity:0.07000002;stroke-width:1.11137545" />
+                </g>
+              </g>
+              <g
+                 id="art-8"
+                 style="stroke-width:1.11137545">
+                <g
+                   id="g56617-9"
+                   style="stroke-width:1.11137545">
+                  <g
+                     id="g56615-2"
+                     style="stroke-width:1.11137545">
+                    <path
+                       id="path56605-6"
+                       d="m 63.9997,40.804 c -12.8097,0 -23.1947,10.3849 -23.1947,23.1959 0,12.8097 10.385,23.1947 23.1947,23.1947 12.8097,0 23.1947,-10.385 23.1947,-23.1947 C 87.1945,51.1889 76.8095,40.804 63.9997,40.804 m 0,40.7948 c -9.72,0 -17.599,-7.879 -17.599,-17.599 0,-9.72 7.879,-17.6007 17.599,-17.6007 9.72,0 17.6001,7.8807 17.6001,17.6007 0,9.72 -7.8801,17.599 -17.6001,17.599"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56607-6"
+                       d="m 52.9898,63.1041 v 7.2091 c 1.0659,1.8326 2.5751,3.3702 4.381,4.4754 V 63.1041 Z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56609-4"
+                       d="m 61.6754,57.0263 v 19.4109 c 0.7448,0.1363 1.5067,0.2199 2.2892,0.2199 0.7145,0 1.4098,-0.0752 2.093,-0.189 V 57.0263 Z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56611-9"
+                       d="m 70.7656,66.1008 v 8.5614 c 1.8325,-1.1695 3.3478,-2.7822 4.3822,-4.7002 v -3.8612 z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56613-5"
+                       d="m 80.6914,78.2867 -2.403,2.4049 c -0.4239,0.4227 -0.4239,1.1132 0,1.5371 l 9.1143,9.112 c 0.4227,0.4238 1.1144,0.4238 1.5371,0 l 2.403,-2.4013 c 0.4214,-0.4227 0.4214,-1.1143 0,-1.5365 l -9.1156,-9.1162 c -0.4215,-0.4221 -1.1143,-0.4221 -1.5358,0"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                  </g>
+                </g>
+              </g>
+              <g
+                 id="Guides-0"
+                 style="stroke-width:1.11137545" />
+            </g>
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -352.87397,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-5-6-9-4"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#4486f6;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -385.41821,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-5-6-6-8"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-379.1109"
+               y="155.33165"
+               id="text55709-5-8-0-4-7"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2-6-1"
+                 x="-379.1109"
+                 y="155.33165"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">GBQ</tspan></text>
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,270.63242,-64.753609)"
+             id="g56805-7"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -428.29495,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-5-3-2"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -395.51719,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-5-6-72"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#d22b28;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -428.06143,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-5-6-2"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <path
+               style="fill:#d22b28;fill-opacity:1;stroke-width:0.03987985"
+               d="m -412.94012,174.83899 c -2.39286,-0.17337 -4.25462,-0.92363 -4.80603,-1.93675 -0.20796,-0.38207 -0.1997,-0.15791 -0.1997,-5.4163 0,-5.2584 -0.008,-5.03422 0.1997,-5.41631 0.52229,-0.95963 2.12948,-1.65202 4.42926,-1.90817 0.59425,-0.0661 2.22107,-0.055 2.85272,0.0195 2.42842,0.28682 3.99755,1.04433 4.40698,2.12751 l 0.0784,0.20754 0.01,4.80836 c 0.007,3.25426 -0.003,4.88804 -0.0279,5.05486 -0.18694,1.21275 -1.86095,2.11808 -4.44966,2.40645 -0.62283,0.0695 -1.89574,0.0965 -2.49389,0.0532 z m 2.42212,-1.10913 c 1.01274,-0.12241 1.89648,-0.35364 2.52978,-0.66193 0.45831,-0.2231 0.73601,-0.43117 0.86172,-0.64568 0.0872,-0.14886 0.0893,-0.16846 0.0893,-0.89661 0,-0.40943 -0.0124,-0.73888 -0.0269,-0.73213 -0.0149,0.007 -0.16397,0.0828 -0.33145,0.16913 -1.54197,0.7942 -4.13331,1.10309 -6.47743,0.7721 -1.10764,-0.1564 -2.14234,-0.47058 -2.79149,-0.84762 -0.0946,-0.055 -0.1794,-0.0999 -0.18839,-0.0999 -0.01,0 -0.0164,0.34474 -0.0164,0.7661 v 0.76611 l 0.10825,0.14942 c 0.2826,0.39014 1.05764,0.77963 2.03963,1.02501 0.50715,0.12668 0.98591,0.20062 1.85311,0.28599 0.27949,0.0275 2.0144,-0.01 2.35036,-0.0499 z m -0.0897,-2.99878 c 1.48756,-0.16113 2.68968,-0.55239 3.26059,-1.06123 0.30547,-0.27227 0.30981,-0.28973 0.30981,-1.2474 0,-0.45588 -0.007,-0.82886 -0.0139,-0.82886 -0.008,0 -0.14893,0.0711 -0.31398,0.15803 -1.28011,0.67416 -3.50461,1.02839 -5.53905,0.88204 -1.40344,-0.10096 -2.61466,-0.38428 -3.49862,-0.81837 -0.22696,-0.11141 -0.42477,-0.20828 -0.43957,-0.21517 -0.0147,-0.007 -0.027,0.36911 -0.027,0.83552 0,0.96882 -0.004,0.94996 0.30074,1.22903 0.49729,0.45451 1.6456,0.85736 2.94812,1.03429 0.84879,0.11526 2.11973,0.12888 3.01279,0.0321 z m -0.25117,-3.17465 c 1.79501,-0.15181 3.38835,-0.72508 3.73976,-1.34554 0.0773,-0.13644 0.0806,-0.17795 0.0813,-0.99242 l 4.3e-4,-0.85007 -0.29605,0.15736 c -0.83154,0.442 -2.06898,0.76261 -3.39994,0.88088 -0.64577,0.0574 -2.00104,0.048 -2.62283,-0.0182 -1.30016,-0.13837 -2.35469,-0.41553 -3.14541,-0.82671 l -0.36781,-0.19125 v 0.87316 0.87316 l 0.11828,0.155 c 0.46867,0.61447 1.93789,1.13255 3.63158,1.28057 0.51111,0.0447 1.75415,0.0469 2.26066,0.004 z m -0.19736,-3.19261 c 1.57372,-0.11691 2.89719,-0.48754 3.5985,-1.00787 0.23246,-0.17245 0.42043,-0.42724 0.42043,-0.56986 0,-0.14316 -0.18674,-0.40773 -0.40076,-0.5678 -0.59209,-0.44282 -1.65343,-0.78705 -2.97228,-0.96401 -0.61144,-0.0821 -2.35152,-0.0922 -2.96037,-0.0173 -1.07536,0.13231 -1.92804,0.35436 -2.5836,0.67278 -0.55306,0.26864 -0.91503,0.60618 -0.91503,0.85329 0,0.28182 0.3789,0.63844 0.97521,0.91787 1.15871,0.543 3.07651,0.81371 4.8379,0.6829 z"
+               id="path56342-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccscccccscccccccsccscssccccccccssccccsccccccccccccccccsccccsscccsscc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-421.75409"
+               y="155.33165"
+               id="text55709-5-8-0-1"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2-0"
+                 x="-421.75409"
+                 y="155.33165"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">SQL</tspan></text>
+          </g>
+          <g
+             transform="translate(-4.7030536,-0.28414145)"
+             id="g12598-6"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.39916864;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="M -23.381309,41.017226 H -3.6488181 L 1.5833621,46.14899 V 71.761885 H -23.381309 Z"
+               id="rect55679-5-3-6-1"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.4313232;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="M 1.6600345,46.275038 H -3.6488181 V 41.017226 Z"
+               id="rect55701-5-6-7-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.31363189;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -23.202906,41.017226 h 19.5540879 l -0.00306,2.445808 H -23.203307 Z"
+               id="rect55679-1-5-6-5-9"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.08538723px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.22464751"
+               x="-18.192799"
+               y="60.704777"
+               id="text55709-5-8-0-5-9-4"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2-62-1-9"
+                 x="-18.192799"
+                 y="60.704777"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#333333;fill-opacity:1;stroke-width:0.22464751">...</tspan></text>
+          </g>
+        </g>
+      </g>
+      <path
+         sodipodi:nodetypes="cc"
+         inkscape:connector-curvature="0"
+         id="path6109"
+         d="M 24.80494,44.790275 H 51.474804"
+         style="fill:none;stroke:#000000;stroke-width:0.44455019;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend)" />
+      <g
+         transform="matrix(0.78210044,0,0,0.78210044,21.652355,31.482679)"
+         id="g13221"
+         style="stroke-width:1.11137545">
+        <g
+           id="g13046"
+           transform="translate(0,3.4017917)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.76397361,0,0,0.76397361,301.89818,-142.00712)"
+             id="g57096-9"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -506.51101,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-1"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -473.73325,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-2"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-499.97018"
+               y="155.93517"
+               id="text55709-5-8-7"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-0"
+                 x="-499.97018"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">CSV</tspan></text>
+            <path
+               style="fill:#755075;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -506.27749,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-9"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <g
+               transform="translate(-681.73064,244.47732)"
+               id="g56212-3"
+               style="stroke-width:1.11137545">
+              <rect
+                 y="-83.565544"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-60"
+                 height="2.2165694" />
+              <rect
+                 y="-79.919731"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-6-6"
+                 height="2.2165694" />
+              <rect
+                 y="-76.273918"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-4-2"
+                 height="2.2165694" />
+              <rect
+                 y="-72.628105"
+                 x="179.92134"
+                 width="23.273981"
+                 style="display:inline;overflow:visible;visibility:visible;opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:none;stroke-width:0.37059268;stroke-opacity:1;marker:none"
+                 id="rect2277-8-3-4-5-6"
+                 height="2.2165694" />
+            </g>
+          </g>
+          <g
+             transform="matrix(0.76397361,0,0,0.76397361,389.69361,-142.00712)"
+             id="g57123"
+             style="stroke-width:1.11137545">
+            <g
+               id="g56942"
+               style="stroke-width:1.11137545">
+              <path
+                 sodipodi:nodetypes="cccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679"
+                 d="m -584.13959,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+                 style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <path
+                 sodipodi:nodetypes="cccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55701"
+                 d="m -551.36183,144.95855 h -6.949 v -6.88219 z"
+                 style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686" />
+              <text
+                 id="text55709"
+                 y="155.93515"
+                 x="-577.10522"
+                 style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+                 xml:space="preserve"><tspan
+                   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141"
+                   y="155.93515"
+                   x="-577.10522"
+                   id="tspan55707"
+                   sodipodi:role="line">XLS</tspan></text>
+              <g
+                 id="g4140"
+                 transform="matrix(0.15295911,0,0,0.15295911,-574.53339,160.91154)"
+                 style="stroke-width:1.11137545">
+                <path
+                   d="m 46.04,0 h 5.94 c 0,2.67 0,5.33 0,8 10.01,0 20.02,0.02 30.03,-0.03 1.69,0.07 3.55,-0.05 5.02,0.96 1.03,1.48 0.91,3.36 0.98,5.06 -0.05,17.36 -0.03,34.71 -0.02,52.06 -0.05,2.91 0.27,5.88 -0.34,8.75 -0.4,2.08 -2.9,2.13 -4.57,2.2 -10.36,0.03 -20.73,-0.02 -31.1,0 0,3 0,6 0,9 H 45.77 C 30.53,83.23 15.26,80.67 0,78 0,54.67 0,31.34 0,8.01 15.35,5.34 30.7,2.71 46.04,0 Z"
+                   id="path10-9"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 51.98,11 c 11,0 22,0 33,0 0,21 0,42 0,63 -11,0 -22,0 -33,0 0,-2 0,-4 0,-6 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-1.33 0,-2.67 0,-4 2.67,0 5.33,0 8,0 0,-2.33 0,-4.67 0,-7 -2.67,0 -5.33,0 -8,0 0,-2 0,-4 0,-6 z"
+                   id="path48-1"
+                   inkscape:connector-curvature="0"
+                   style="fill:#ffffff;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,17 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path58-2"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 29.62,26.37 c 2.26,-0.16 4.53,-0.3 6.8,-0.41 -2.67,5.47 -5.35,10.94 -8.07,16.39 2.75,5.6 5.56,11.16 8.32,16.76 -2.41,-0.14 -4.81,-0.29 -7.22,-0.46 -1.7,-4.17 -3.77,-8.2 -4.99,-12.56 -1.36,4.06 -3.3,7.89 -4.86,11.87 -2.19,-0.03 -4.38,-0.12 -6.57,-0.21 2.57,-5.03 5.05,-10.1 7.7,-15.1 -2.25,-5.15 -4.72,-10.2 -7.04,-15.32 2.2,-0.13 4.4,-0.26 6.6,-0.38 1.49,3.91 3.12,7.77 4.35,11.78 1.32,-4.25 3.29,-8.25 4.98,-12.36 z"
+                   id="path72-7"
+                   inkscape:connector-curvature="0"
+                   style="fill:#ffffff;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,28 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path90"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,39 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path108"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,50 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path114"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+                <path
+                   d="m 63.98,61 c 4.67,0 9.33,0 14,0 0,2.33 0,4.67 0,7 -4.67,0 -9.33,0 -14,0 0,-2.33 0,-4.67 0,-7 z"
+                   id="path120"
+                   inkscape:connector-curvature="0"
+                   style="fill:#207245;stroke-width:1.11137545" />
+              </g>
+              <path
+                 sodipodi:nodetypes="ccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-1"
+                 d="m -583.90607,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+                 style="fill:#207245;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+            </g>
+          </g>
+          <g
+             transform="translate(32.905868,-67.171919)"
+             id="g12301"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.39916864;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -60.99023,30.65149 h 19.73249 l 5.23218,5.13176 v 25.6129 h -24.96467 z"
+               id="rect55679-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#6995f4;fill-opacity:1;stroke-width:0.01482968"
+               d="m -48.45626,58.61909 c -0.0716,-0.0731 -0.1361,-0.133 -0.14344,-0.133 -0.008,0 -0.0134,-0.009 -0.0134,-0.02 0,-0.0109 -0.009,-0.02 -0.02,-0.02 -0.011,0 -0.02,-0.009 -0.02,-0.02 0,-0.0109 -0.009,-0.02 -0.02,-0.02 -0.0109,0 -0.02,-0.009 -0.02,-0.02 0,-0.0109 -0.006,-0.02 -0.0134,-0.02 -0.008,0 -0.0134,-0.006 -0.0134,-0.0134 0,-0.008 -0.0331,-0.0461 -0.0733,-0.0862 -0.0403,-0.04 -0.0734,-0.0791 -0.0734,-0.0867 0,-0.008 -0.007,-0.014 -0.0145,-0.014 -0.0215,0 -0.0793,-0.0484 -0.0694,-0.0583 0.005,-0.005 7.7e-4,-0.009 -0.008,-0.009 -0.0162,0 -0.086,-0.0679 -0.1372,-0.13343 -0.0143,-0.0183 -0.0616,-0.0705 -0.10521,-0.11607 -0.0435,-0.0455 -0.0792,-0.0894 -0.0792,-0.0975 0,-0.009 -0.006,-0.0112 -0.0134,-0.007 -0.008,0.005 -0.0134,-7.6e-4 -0.0134,-0.0126 0,-0.0115 -0.009,-0.0208 -0.02,-0.0208 -0.0111,0 -0.02,-0.006 -0.02,-0.0134 0,-0.008 -0.006,-0.0134 -0.0141,-0.0134 -0.0157,0 -0.0926,-0.08 -0.0926,-0.0964 0,-0.006 -0.006,-0.0104 -0.0133,-0.0104 -0.0211,0 -0.0935,-0.0818 -0.0935,-0.10572 0,-0.023 0.11047,-0.13446 0.1333,-0.13446 0.008,0 0.0135,-0.009 0.0135,-0.02 0,-0.0111 0.006,-0.02 0.0128,-0.02 0.007,0 0.0594,-0.045 0.11617,-0.10007 0.0568,-0.0551 0.10848,-0.10008 0.11472,-0.10008 0.0137,0 0.0899,-0.0817 0.0899,-0.0964 0,-0.006 0.006,-0.0104 0.014,-0.0104 0.008,0 0.0292,-0.0164 0.0478,-0.0365 0.0186,-0.02 0.0484,-0.0461 0.0662,-0.0577 0.0178,-0.0117 0.0712,-0.0583 0.11856,-0.10364 0.0474,-0.0453 0.0924,-0.0824 0.0999,-0.0824 0.008,0 0.0138,-0.007 0.0138,-0.0143 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.006,-0.02 0.0134,-0.02 0.008,0 0.0539,-0.0405 0.10352,-0.0901 0.0496,-0.0496 0.12006,-0.11409 0.15669,-0.14344 0.0367,-0.0293 0.0686,-0.0579 0.071,-0.0633 0.002,-0.005 0.0129,-0.01 0.0234,-0.01 0.0105,0 0.0189,-0.006 0.0189,-0.0142 0,-0.014 0.0338,-0.0477 0.1495,-0.14964 0.15388,-0.13549 0.17813,-0.15569 0.18742,-0.15601 0.005,-2e-4 0.01,-0.007 0.01,-0.0147 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.011 0.006,-0.02 0.0142,-0.02 0.008,0 0.0356,-0.0225 0.0617,-0.05 0.0261,-0.0276 0.0527,-0.0485 0.0591,-0.0468 0.006,0.002 0.0117,-0.003 0.0117,-0.0118 0,-0.015 0.0267,-0.0414 0.0974,-0.0961 0.0205,-0.0159 0.0686,-0.0603 0.10676,-0.0986 0.0382,-0.0383 0.0769,-0.0699 0.0861,-0.0701 0.009,-1.5e-4 0.0167,-0.006 0.0167,-0.0136 0,-0.008 0.012,-0.0254 0.0267,-0.0401 0.0147,-0.0147 0.0267,-0.0237 0.0267,-0.02 0,0.003 0.012,-0.005 0.0267,-0.02 0.0147,-0.0147 0.0267,-0.0327 0.0267,-0.04 0,-0.008 0.008,-0.0134 0.0168,-0.0134 0.009,0 0.0331,-0.015 0.0528,-0.0334 0.0198,-0.0183 0.0426,-0.0334 0.0507,-0.0334 0.008,0 0.0114,-0.005 0.008,-0.0119 -0.005,-0.007 0.008,-0.0219 0.0268,-0.0342 0.0455,-0.0298 0.0857,-0.0708 0.0857,-0.0873 0,-0.008 0.007,-0.0134 0.0144,-0.0134 0.008,0 0.05,-0.0361 0.0933,-0.08 0.0433,-0.044 0.0814,-0.08 0.0846,-0.08 0.003,0 0.0232,-0.0165 0.0445,-0.0367 0.0844,-0.0799 0.12162,-0.11009 0.13552,-0.11009 0.008,0 0.0145,-0.007 0.0145,-0.0143 0,-0.0185 0.0339,-0.0524 0.0524,-0.0524 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.006,-0.02 0.013,-0.02 0.007,0 0.0455,-0.0315 0.0852,-0.07 0.0806,-0.0782 0.15505,-0.14345 0.16376,-0.14345 0.003,0 0.0502,-0.045 0.10452,-0.10008 0.0543,-0.0551 0.10416,-0.10007 0.11073,-0.10007 0.007,0 0.0199,-0.015 0.0298,-0.0334 0.01,-0.0183 0.0226,-0.0334 0.0284,-0.0334 0.006,0 0.0374,-0.0255 0.0701,-0.0567 0.0808,-0.0772 0.17035,-0.15678 0.17609,-0.15678 0.002,0 0.0439,-0.039 0.0919,-0.0867 0.048,-0.0477 0.0935,-0.0867 0.10133,-0.0867 0.008,0 0.0105,-0.006 0.006,-0.0134 -0.005,-0.008 -0.002,-0.0134 0.005,-0.0134 0.0157,0 0.0761,-0.0506 0.17152,-0.14344 0.0396,-0.0385 0.0777,-0.0701 0.0848,-0.0701 0.007,0 0.0128,-0.007 0.0128,-0.0143 0,-0.0163 0.0332,-0.0524 0.048,-0.0524 0.0113,0 0.0189,-0.007 0.14815,-0.13125 0.0529,-0.051 0.0994,-0.0892 0.10342,-0.0852 0.005,0.005 0.008,-0.002 0.008,-0.0149 0,-0.0122 0.009,-0.0222 0.02,-0.0222 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0109 0.006,-0.02 0.0134,-0.02 0.008,0 0.0134,-0.009 0.0134,-0.02 0,-0.011 0.008,-0.02 0.0169,-0.02 0.009,0 0.028,-0.015 0.0419,-0.0334 0.0137,-0.0183 0.0306,-0.0334 0.0375,-0.0334 0.017,0 0.0506,-0.0348 0.0506,-0.0524 0,-0.008 0.006,-0.0143 0.0134,-0.0143 0.008,0 0.0243,-0.0105 0.0377,-0.0234 0.0423,-0.0404 0.0956,-0.0834 0.10351,-0.0834 0.0113,0 0.11227,-0.11049 0.11227,-0.12284 0,-0.006 0.009,-0.0106 0.02,-0.0106 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.009,-0.02 0.02,-0.02 0.0109,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0106,-0.0158 0.0236,-0.0199 0.013,-0.005 0.0784,-0.0612 0.14534,-0.12689 0.0669,-0.0657 0.12534,-0.11573 0.1298,-0.11126 0.005,0.005 0.008,-0.002 0.008,-0.0141 0,-0.0122 0.009,-0.0222 0.02,-0.0222 0.0109,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.009,-0.02 0.0208,-0.02 0.0115,0 0.0171,-0.006 0.0126,-0.0134 -0.005,-0.008 7.6e-4,-0.0134 0.0126,-0.0134 0.0115,0 0.0208,-0.009 0.0208,-0.02 0,-0.011 0.005,-0.02 0.0121,-0.02 0.0116,0 0.054,-0.0324 0.10929,-0.0834 0.0139,-0.0128 0.0312,-0.0234 0.0386,-0.0234 0.008,0 0.0134,-0.009 0.0134,-0.02 0,-0.0111 0.006,-0.02 0.0134,-0.02 0.017,0 0.16013,-0.14365 0.16013,-0.16071 0,-0.007 0.007,-0.0127 0.0152,-0.0127 0.009,0 0.0403,-0.0255 0.0708,-0.0567 0.0306,-0.0312 0.0727,-0.0688 0.0937,-0.0834 0.021,-0.0147 0.0703,-0.0582 0.10963,-0.0967 0.0393,-0.0385 0.0739,-0.0701 0.0768,-0.0701 0.0109,0 0.0875,-0.0826 0.0875,-0.0944 0,-0.007 0.007,-0.0123 0.0149,-0.0123 0.009,0 0.039,-0.0241 0.0683,-0.0533 0.0293,-0.0293 0.0583,-0.0533 0.0644,-0.0533 0.006,0 0.0744,-0.0626 0.15188,-0.13903 0.0775,-0.0765 0.14089,-0.13535 0.14089,-0.13084 0,0.005 0.015,-0.0121 0.0334,-0.037 0.0183,-0.0248 0.0334,-0.0411 0.0334,-0.0361 0,0.009 0.0347,-0.0167 0.0739,-0.054 0.01,-0.009 0.024,-0.0167 0.0319,-0.0167 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.0111 0.007,-0.02 0.0143,-0.02 0.0185,0 0.0524,-0.0339 0.0524,-0.0524 0,-0.008 0.009,-0.0143 0.02,-0.0143 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.005,-0.0185 0.0105,-0.0167 0.006,0.002 0.0355,-0.0207 0.0662,-0.05 0.0823,-0.0786 0.10166,-0.0938 0.11416,-0.0898 0.006,0.002 0.008,-0.002 0.003,-0.009 -0.009,-0.0144 0.049,-0.0552 0.0785,-0.0552 0.0112,0 0.0205,0.006 0.0205,0.0134 0,0.008 0.009,0.0134 0.0205,0.0134 0.0112,0 0.0239,0.009 0.028,0.02 0.005,0.0111 0.0221,0.02 0.0396,0.02 0.0176,0 0.0319,0.006 0.0319,0.0125 0,0.007 0.0135,0.0167 0.03,0.0219 0.0165,0.005 0.0331,0.013 0.0368,0.0176 0.0107,0.0132 0.0581,0.0417 0.0834,0.0502 0.0128,0.005 0.0234,0.0131 0.0234,0.0196 0,0.007 0.0113,0.0118 0.025,0.0118 0.0138,0 0.0325,0.009 0.0417,0.02 0.009,0.0111 0.0272,0.02 0.0403,0.02 0.013,0 0.0271,0.009 0.0313,0.02 0.005,0.0111 0.0189,0.02 0.0327,0.02 0.0286,0 0.10484,0.0355 0.1164,0.0542 0.005,0.007 0.0193,0.0125 0.0337,0.0125 0.0142,0 0.0258,0.006 0.0258,0.0131 0,0.007 0.0151,0.0169 0.0334,0.0215 0.0183,0.005 0.0334,0.0137 0.0334,0.0203 0,0.007 0.0113,0.0119 0.025,0.0119 0.0137,0 0.0325,0.009 0.0417,0.02 0.009,0.0111 0.0279,0.02 0.0417,0.02 0.0138,0 0.025,0.006 0.025,0.0134 0,0.008 0.009,0.0134 0.0191,0.0134 0.0105,0 0.027,0.008 0.0367,0.0167 0.01,0.009 0.0416,0.0287 0.071,0.0434 0.0293,0.0147 0.0554,0.0312 0.0577,0.0367 0.002,0.005 0.0189,0.01 0.0367,0.01 0.0198,0 0.0323,0.008 0.0323,0.02 0,0.0109 0.007,0.02 0.0155,0.02 0.009,0 -0.01,0.024 -0.0406,0.0533 -0.0308,0.0293 -0.0618,0.0533 -0.0689,0.0533 -0.007,0 -0.0128,0.009 -0.0128,0.0205 0,0.0112 -0.009,0.0239 -0.02,0.0281 -0.0111,0.005 -0.02,0.016 -0.02,0.0263 0,0.0102 -0.007,0.0186 -0.0149,0.0186 -0.009,0 -0.0368,0.0251 -0.0633,0.056 -0.0597,0.0689 -0.17284,0.1842 -0.18074,0.1842 -0.006,0 -0.0879,0.0797 -0.0879,0.086 0,0.0121 -0.25327,0.26091 -0.26551,0.26091 -0.008,0 -0.0147,0.006 -0.0147,0.0142 0,0.008 -0.0315,0.0459 -0.0701,0.0846 -0.0385,0.0387 -0.10127,0.10436 -0.13943,0.1459 -0.0382,0.0416 -0.0757,0.0755 -0.0834,0.0755 -0.008,0 -0.0141,0.007 -0.0141,0.0143 0,0.0185 -0.034,0.0524 -0.0524,0.0524 -0.008,0 -0.0143,0.008 -0.0143,0.0162 0,0.0183 -0.1426,0.15729 -0.16133,0.15729 -0.007,0 -0.0122,0.009 -0.0122,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.018,0 -0.026,0.0158 -0.0215,0.0433 7.7e-4,0.005 -0.005,0.01 -0.0105,0.01 -0.007,0 -0.0623,0.0512 -0.12343,0.11381 -0.0612,0.0626 -0.19038,0.19416 -0.28704,0.29238 -0.0967,0.0982 -0.17223,0.1821 -0.16791,0.18641 0.005,0.005 7.6e-4,0.008 -0.007,0.008 -0.0184,0 -0.20301,0.1823 -0.20301,0.20044 0,0.007 -0.007,0.0131 -0.0143,0.0131 -0.0148,0 -0.0524,0.0326 -0.0524,0.0454 0,0.0128 -0.12034,0.12807 -0.13379,0.12807 -0.007,0 -0.013,0.006 -0.013,0.0134 0,0.008 -0.0241,0.0374 -0.0533,0.0667 -0.0293,0.0293 -0.0594,0.0533 -0.0667,0.0533 -0.008,0 -0.0134,0.006 -0.0134,0.0125 0,0.007 -0.0466,0.0601 -0.10342,0.11848 -0.0568,0.0584 -0.14023,0.14437 -0.18525,0.19113 -0.13555,0.1408 -0.24885,0.24535 -0.26166,0.24146 -0.007,-0.002 -0.009,0.002 -0.003,0.01 0.005,0.008 -0.034,0.0554 -0.0858,0.10662 -0.0517,0.0512 -0.0941,0.0994 -0.0941,0.10694 0,0.008 -0.008,0.0141 -0.0167,0.0143 -0.009,2.5e-4 -0.0527,0.04 -0.0967,0.0882 -0.1391,0.15243 -0.21768,0.22954 -0.22573,0.22151 -0.005,-0.005 -0.008,-7.7e-4 -0.008,0.008 0,0.009 -0.0355,0.0511 -0.0789,0.0945 -0.0434,0.0434 -0.0824,0.0754 -0.0867,0.0711 -0.005,-0.005 -0.008,7.6e-4 -0.008,0.0115 0,0.0224 -0.0474,0.0668 -0.0589,0.0552 -0.005,-0.005 -0.008,0.002 -0.008,0.0126 0,0.0112 -0.015,0.0317 -0.0334,0.0454 -0.0183,0.0137 -0.0334,0.0326 -0.0334,0.0419 0,0.009 -0.009,0.0169 -0.02,0.0169 -0.011,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.006,0.02 -0.0134,0.02 -0.0156,0 -0.16011,0.14299 -0.16011,0.15845 0,0.006 -0.015,0.0185 -0.0334,0.0283 -0.0183,0.01 -0.0334,0.0259 -0.0334,0.0356 0,0.01 -0.008,0.0177 -0.0177,0.0177 -0.01,0 -0.0257,0.015 -0.0356,0.0334 -0.01,0.0183 -0.0258,0.0334 -0.0356,0.0334 -0.01,0 -0.0177,0.007 -0.0177,0.0144 0,0.0186 -0.0817,0.10008 -0.0959,0.0957 -0.006,-0.002 -0.0109,0.003 -0.0109,0.011 0,0.0189 -0.0341,0.0524 -0.0533,0.0524 -0.009,0 -0.0113,0.006 -0.007,0.0134 0.005,0.008 -7.6e-4,0.017 -0.0121,0.0212 -0.0112,0.005 -0.024,0.0192 -0.0284,0.0332 -0.005,0.014 -0.0227,0.0291 -0.0405,0.0335 -0.0178,0.005 -0.0325,0.014 -0.0325,0.021 0,0.0173 -0.0347,0.051 -0.0524,0.051 -0.008,0 -0.0143,0.006 -0.0143,0.0142 0,0.0157 -0.0301,0.0497 -0.16559,0.18634 -0.0506,0.0512 -0.0973,0.093 -0.10342,0.093 -0.006,0 -0.0112,0.006 -0.0112,0.0142 0,0.008 -0.0211,0.0341 -0.0468,0.0583 -0.0257,0.0242 -0.0468,0.0421 -0.0468,0.0397 0,-0.002 -0.0271,0.0246 -0.06,0.0598 -0.0331,0.0353 -0.0601,0.0685 -0.0601,0.0738 0,0.013 -0.0355,0.0477 -0.049,0.0478 -0.006,5e-5 -0.024,0.0166 -0.0399,0.0367 -0.0481,0.0609 -0.13316,0.14999 -0.1431,0.14999 -0.0147,0 -0.0748,0.0639 -0.0748,0.0795 0,0.008 -0.009,0.014 -0.02,0.014 -0.0109,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.009,0.02 -0.0208,0.02 -0.0115,0 -0.0177,0.005 -0.0141,0.011 0.008,0.0122 -0.0382,0.0557 -0.0588,0.0557 -0.007,0 -0.0132,0.005 -0.0132,0.0121 0,0.017 -0.20964,0.22804 -0.22646,0.22804 -0.008,0 -0.0137,0.006 -0.0137,0.0142 0,0.008 -0.0451,0.0587 -0.10008,0.11299 -0.0551,0.0544 -0.10008,0.10498 -0.10008,0.11256 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.005 -0.02,0.0105 0,0.0115 -0.0503,0.0642 -0.16561,0.17358 -0.041,0.0389 -0.0746,0.0743 -0.0746,0.0786 0,0.0131 -0.0788,0.0842 -0.0933,0.0842 -0.008,0 -0.0135,0.008 -0.0135,0.0167 -2e-5,0.009 -0.0361,0.0547 -0.0801,0.10128 -0.044,0.0465 -0.08,0.0818 -0.08,0.0783 0,-0.003 -0.051,0.0448 -0.11343,0.10739 -0.0624,0.0625 -0.11342,0.11572 -0.11342,0.11822 0,0.0121 -0.18633,0.19192 -0.19895,0.19192 -0.008,0 -0.0145,0.007 -0.0145,0.0151 0,0.009 -0.0211,0.0349 -0.0467,0.0591 -0.0257,0.0242 -0.0468,0.0504 -0.0468,0.0583 0,0.008 -0.009,0.0142 -0.02,0.0142 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.0396 -0.0608,-0.002 -0.1768,-0.12053 z m -1.54346,-1.58412 c -5.2e-4,-0.009 -0.0101,-0.0167 -0.0211,-0.0167 -0.011,0 -0.02,-0.009 -0.02,-0.02 0,-0.0111 -0.009,-0.02 -0.02,-0.02 -0.0111,0 -0.02,-0.005 -0.02,-0.011 0,-0.0142 -0.0729,-0.0958 -0.0855,-0.0958 -0.0148,0 -0.0478,-0.0362 -0.0478,-0.0524 0,-0.008 -0.006,-0.0143 -0.0139,-0.0143 -0.019,0 -0.06,-0.0412 -0.0708,-0.0713 -0.005,-0.0135 -0.0208,-0.0277 -0.0355,-0.0316 -0.0147,-0.003 -0.0267,-0.0111 -0.0267,-0.0161 0,-0.0131 -0.11633,-0.14283 -0.18547,-0.20691 -0.10595,-0.0982 -0.23492,-0.23707 -0.23152,-0.24927 0.002,-0.007 -0.002,-0.012 -0.0105,-0.012 -0.014,0 -0.093,-0.0749 -0.093,-0.0882 0,-0.0127 -0.0377,-0.0453 -0.0524,-0.0453 -0.008,0 -0.0143,-0.009 -0.0143,-0.0186 0,-0.0102 -0.0151,-0.0223 -0.0334,-0.027 -0.0183,-0.005 -0.0334,-0.0144 -0.0334,-0.0219 0,-0.008 -0.019,-0.0321 -0.0421,-0.0547 -0.0369,-0.0361 -0.0394,-0.0427 -0.02,-0.0536 0.0121,-0.007 0.0446,-0.0331 0.0722,-0.0584 0.0276,-0.0253 0.061,-0.0461 0.0743,-0.0461 0.0133,-5e-5 0.021,-0.005 0.0171,-0.0114 -0.003,-0.006 0.0149,-0.0287 0.0416,-0.05 0.0267,-0.0213 0.0561,-0.0462 0.0653,-0.0554 0.009,-0.009 0.0204,-0.0167 0.0249,-0.0167 0.0102,0 0.0419,-0.0257 0.0714,-0.0578 0.0123,-0.0134 0.0569,-0.0488 0.0991,-0.0787 0.0422,-0.0298 0.0767,-0.0594 0.0767,-0.0657 0,-0.006 0.006,-0.0114 0.0141,-0.0114 0.008,0 0.0356,-0.0211 0.0621,-0.0468 0.0264,-0.0257 0.0526,-0.0468 0.0581,-0.0468 0.005,0 0.0377,-0.0268 0.0713,-0.0595 0.0338,-0.0328 0.0613,-0.0558 0.0613,-0.051 0,0.005 0.015,-0.006 0.0334,-0.0228 0.0183,-0.0172 0.0334,-0.0277 0.0334,-0.0234 0,0.005 0.0253,-0.0156 0.0562,-0.0444 0.0308,-0.0288 0.0609,-0.0523 0.0667,-0.0523 0.006,0 0.0346,-0.0241 0.064,-0.0533 0.0293,-0.0293 0.0564,-0.0533 0.0602,-0.0533 0.006,0 0.0676,-0.0513 0.10754,-0.0901 0.009,-0.009 0.0236,-0.0167 0.0315,-0.0167 0.008,0 0.0143,-0.006 0.0143,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.009 0.02,-0.02 0,-0.0111 0.008,-0.02 0.0168,-0.02 0.009,0 0.0329,-0.015 0.0528,-0.0334 0.0198,-0.0183 0.0422,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0111 0.006,-0.02 0.0139,-0.02 0.008,0 0.0389,-0.0241 0.0694,-0.0533 0.0306,-0.0293 0.0603,-0.0533 0.0661,-0.0533 0.006,0 0.0241,-0.0116 0.0406,-0.0259 0.0918,-0.0789 0.11089,-0.0942 0.11796,-0.0942 0.005,-10e-6 0.0339,-0.0241 0.0656,-0.0534 0.0318,-0.0293 0.0623,-0.0533 0.0679,-0.0533 0.005,0 0.0281,-0.0179 0.0502,-0.04 0.022,-0.022 0.0449,-0.0402 0.051,-0.0403 0.006,-1.4e-4 0.0253,-0.0152 0.0426,-0.0335 0.0173,-0.0183 0.0489,-0.0435 0.0701,-0.056 0.0212,-0.0125 0.0384,-0.0305 0.0384,-0.0399 0,-0.009 0.009,-0.0171 0.02,-0.0171 0.0111,0 0.02,-0.005 0.02,-0.0109 0,-0.006 0.0243,-0.0246 0.054,-0.0412 0.0297,-0.0167 0.0612,-0.0419 0.07,-0.0559 0.009,-0.014 0.0194,-0.0254 0.0238,-0.0254 0.005,0 0.0339,-0.0241 0.0658,-0.0533 0.0318,-0.0293 0.0652,-0.0533 0.0743,-0.0533 0.009,0 0.0203,-0.01 0.0248,-0.0219 0.005,-0.0121 0.0118,-0.0185 0.016,-0.0144 0.007,0.007 0.0423,-0.0224 0.14025,-0.11385 0.0137,-0.0128 0.031,-0.0234 0.0384,-0.0234 0.008,0 0.0134,-0.006 0.0134,-0.0134 0,-0.008 0.005,-0.0134 0.0123,-0.0134 0.007,0 0.0304,-0.0165 0.0526,-0.0367 0.0941,-0.0858 0.12369,-0.11009 0.13441,-0.11009 0.006,0 0.0148,-0.009 0.019,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.006,-0.0134 0.0124,-0.0134 0.007,0 0.0347,-0.0211 0.0619,-0.0468 0.0272,-0.0257 0.0542,-0.0468 0.0599,-0.0468 0.006,0 0.0374,-0.027 0.0704,-0.06 0.0331,-0.0331 0.0643,-0.0601 0.0696,-0.0601 0.005,0 0.0235,-0.015 0.0404,-0.0334 0.017,-0.0183 0.0358,-0.0334 0.0417,-0.0334 0.006,0 0.027,-0.015 0.0468,-0.0334 0.0198,-0.0183 0.0422,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0116 0.0121,-0.02 0.029,-0.02 0.0159,0 0.0251,-0.003 0.0205,-0.009 -0.008,-0.008 0.0462,-0.0583 0.0629,-0.0583 0.005,0 0.0319,-0.0241 0.0613,-0.0533 0.0293,-0.0293 0.0627,-0.0533 0.0742,-0.0533 0.0115,0 0.0171,-0.006 0.0126,-0.0134 -0.005,-0.008 7.6e-4,-0.0134 0.0126,-0.0134 0.0115,0 0.0208,-0.005 0.0208,-0.0111 0,-0.006 0.0328,-0.0346 0.0731,-0.0633 0.0403,-0.0289 0.0732,-0.0568 0.0734,-0.0624 1.5e-4,-0.005 0.007,-0.01 0.0143,-0.01 0.008,0 0.0279,-0.0151 0.0448,-0.0334 0.017,-0.0183 0.0438,-0.0334 0.0594,-0.0334 0.0157,0 0.027,-0.005 0.0251,-0.0105 -0.002,-0.006 0.0163,-0.0298 0.0405,-0.0533 0.046,-0.045 0.0963,-0.057 0.0963,-0.0229 0,0.0111 0.009,0.02 0.02,0.02 0.0111,0 0.02,0.005 0.02,0.0119 0,0.007 0.0105,0.0154 0.0234,0.0196 0.0267,0.009 0.0672,0.0343 0.0938,0.0587 0.01,0.009 0.0222,0.0167 0.0272,0.0167 0.005,1e-5 0.0265,0.015 0.0479,0.0334 0.0213,0.0183 0.0441,0.0333 0.0506,0.0333 0.007,0 0.0258,0.015 0.0428,0.0334 0.017,0.0183 0.0352,0.0334 0.0403,0.0334 0.005,0 0.0178,0.008 0.028,0.0167 0.0463,0.0415 0.0598,0.05 0.079,0.05 0.0113,0 0.0206,0.009 0.0206,0.02 0,0.0111 0.0114,0.02 0.0253,0.02 0.0139,0 0.0287,0.009 0.033,0.02 0.005,0.0111 0.0143,0.02 0.0225,0.02 0.0192,0 0.0528,0.0339 0.0528,0.0532 0,0.009 0.005,0.0118 0.0119,0.008 0.007,-0.005 0.0267,0.005 0.0449,0.0188 0.0183,0.0144 0.0369,0.0222 0.0416,0.0177 0.005,-0.005 0.009,0.002 0.009,0.0138 0,0.0122 0.007,0.0222 0.0143,0.0222 0.008,0 0.0227,0.008 0.0328,0.0167 0.0393,0.0352 0.0587,0.05 0.0656,0.05 0.007,0 0.0825,0.0551 0.16703,0.12109 0.0182,0.0141 0.0377,0.0257 0.0434,0.0257 0.006,0 -0.0201,0.03 -0.0576,0.0667 -0.0375,0.0367 -0.0736,0.0667 -0.0801,0.0667 -0.007,0 -0.012,0.006 -0.012,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.0147,0 -0.0859,0.0613 -0.18222,0.1568 -0.0315,0.0312 -0.0604,0.0567 -0.0644,0.0567 -0.005,0 -0.0362,0.03 -0.0718,0.0667 -0.0355,0.0367 -0.0714,0.0667 -0.0796,0.0667 -0.009,0 -0.015,0.009 -0.015,0.0208 0,0.0115 -0.005,0.0177 -0.0111,0.014 -0.006,-0.003 -0.0305,0.0109 -0.0542,0.0326 -0.0237,0.0216 -0.0488,0.0393 -0.0556,0.0393 -0.007,0 -0.0125,0.005 -0.0125,0.0123 0,0.0129 -0.077,0.0944 -0.0891,0.0944 -0.006,0 -0.0482,0.0377 -0.15182,0.13677 -0.0211,0.0203 -0.0419,0.0367 -0.0461,0.0367 -0.005,0 -0.021,0.0137 -0.0371,0.0305 -0.0162,0.0168 -0.0564,0.0511 -0.0894,0.0763 -0.0331,0.0251 -0.0791,0.0655 -0.10251,0.0896 -0.0234,0.0241 -0.0489,0.0438 -0.0567,0.0438 -0.008,0 -0.0142,0.009 -0.0142,0.02 0,0.0111 -0.008,0.02 -0.0177,0.02 -0.01,0 -0.0258,0.015 -0.0356,0.0334 -0.01,0.0183 -0.0236,0.0334 -0.0307,0.0334 -0.007,0 -0.0268,0.015 -0.0439,0.0334 -0.017,0.0183 -0.0376,0.0334 -0.0457,0.0334 -0.008,0 -0.0118,0.005 -0.008,0.0109 0.003,0.006 -0.0118,0.0219 -0.0343,0.0351 -0.0226,0.0134 -0.0636,0.0474 -0.0911,0.0759 -0.0276,0.0283 -0.0568,0.0516 -0.065,0.0516 -0.009,0 -0.0149,0.009 -0.0149,0.02 0,0.011 -0.005,0.02 -0.01,0.02 -0.0126,4e-5 -0.11046,0.0821 -0.18695,0.15674 -0.032,0.0312 -0.0636,0.0567 -0.0704,0.0567 -0.007,0 -0.0221,0.015 -0.034,0.0332 -0.012,0.0183 -0.0271,0.03 -0.0337,0.0259 -0.007,-0.005 -0.012,5.2e-4 -0.0122,0.0101 -1.5e-4,0.01 -0.03,0.0385 -0.0665,0.0642 -0.0364,0.0257 -0.0662,0.0512 -0.0665,0.0567 -1.5e-4,0.005 -0.009,0.01 -0.0203,0.01 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.007,0.02 -0.0143,0.02 -0.008,0 -0.0229,0.009 -0.0334,0.019 -0.0105,0.0105 -0.019,0.0255 -0.019,0.0334 0,0.008 -0.006,0.0143 -0.0138,0.0143 -0.008,0 -0.0305,0.0165 -0.0508,0.0367 -0.0463,0.0459 -0.14661,0.12862 -0.16316,0.13454 -0.007,0.002 -0.0125,0.0129 -0.0125,0.0234 0,0.0105 -0.009,0.0189 -0.02,0.0189 -0.0109,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0109,0 -0.02,0.008 -0.02,0.0177 0,0.01 -0.015,0.0257 -0.0334,0.0356 -0.0183,0.01 -0.0334,0.0262 -0.0334,0.0364 0,0.0102 -0.005,0.0155 -0.0109,0.0119 -0.006,-0.003 -0.0388,0.0199 -0.0729,0.0526 -0.0341,0.0326 -0.0671,0.0594 -0.0736,0.0594 -0.006,0 -0.0196,0.015 -0.0294,0.0334 -0.01,0.0183 -0.0231,0.0334 -0.0296,0.0334 -0.007,0 -0.0249,0.0137 -0.0411,0.0305 -0.0162,0.0168 -0.0562,0.0513 -0.0891,0.0767 -0.0328,0.0254 -0.0799,0.0658 -0.10455,0.0896 -0.0247,0.0239 -0.0503,0.0433 -0.057,0.0433 -0.007,0 -0.0122,0.009 -0.0122,0.02 0,0.011 -0.006,0.02 -0.0134,0.02 -0.008,0 -0.0389,0.024 -0.0701,0.0532 -0.0312,0.0292 -0.073,0.0682 -0.0929,0.0867 -0.0199,0.0185 -0.0424,0.0335 -0.05,0.0335 -0.008,0 -0.0139,0.009 -0.0139,0.02 0,0.0111 -0.008,0.02 -0.0169,0.02 -0.009,0 -0.0281,0.0151 -0.0419,0.0334 -0.0138,0.0183 -0.0301,0.0334 -0.0363,0.0334 -0.0116,0 -0.055,0.0364 -0.14677,0.12343 -0.029,0.0276 -0.0589,0.05 -0.0665,0.05 -0.008,0 -0.01,0.006 -0.005,0.0134 0.005,0.008 -7.7e-4,0.0134 -0.0126,0.0134 -0.0115,0 -0.0208,0.009 -0.0208,0.02 0,0.0111 -0.006,0.02 -0.0129,0.02 -0.007,0 -0.0439,0.03 -0.0818,0.0667 -0.0378,0.0368 -0.074,0.0667 -0.0805,0.0667 -0.006,0 -0.0449,0.0331 -0.0856,0.0734 -0.0406,0.0403 -0.0792,0.0733 -0.0856,0.0733 -0.006,0 -0.0147,0.008 -0.0184,0.0167 -0.005,0.0134 -0.007,0.0134 -0.008,0 z m -1.50221,-1.53083 c -0.022,-0.0251 -0.0475,-0.0461 -0.0567,-0.0464 -0.009,-2.6e-4 -0.0167,-0.01 -0.0167,-0.0206 0,-0.0111 -0.009,-0.02 -0.02,-0.02 -0.0111,0 -0.02,-0.007 -0.02,-0.0149 0,-0.009 -0.0751,-0.0894 -0.1668,-0.18044 -0.0917,-0.0911 -0.16679,-0.17146 -0.16679,-0.17862 0,-0.007 -0.008,-0.0131 -0.0167,-0.0132 -0.009,-5e-5 -0.0377,-0.0247 -0.0633,-0.0548 -0.0428,-0.0502 -0.15564,-0.1732 -0.18819,-0.20522 -0.12475,-0.12271 -0.19877,-0.19935 -0.19877,-0.20581 0,-0.009 0.043,-0.0335 0.0834,-0.0487 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.006,-0.0118 0.014,-0.0118 0.008,0 0.0302,-0.0152 0.05,-0.0338 0.0199,-0.0186 0.0661,-0.053 0.10283,-0.0765 0.0368,-0.0235 0.0688,-0.0473 0.0711,-0.0529 0.002,-0.006 0.0123,-0.0103 0.0219,-0.0103 0.01,0 0.021,-0.009 0.0252,-0.02 0.005,-0.0109 0.0154,-0.02 0.0249,-0.02 0.009,0 0.0334,-0.015 0.0532,-0.0334 0.0198,-0.0183 0.0423,-0.0334 0.0499,-0.0334 0.008,0 0.014,-0.009 0.014,-0.0208 0,-0.0115 0.005,-0.0176 0.0117,-0.0135 0.007,0.005 0.032,-0.0105 0.0567,-0.0322 0.0248,-0.0216 0.0645,-0.05 0.0883,-0.0629 0.0239,-0.0129 0.0434,-0.0281 0.0434,-0.0338 0,-0.006 0.009,-0.0102 0.0189,-0.0102 0.0105,0 0.0209,-0.005 0.0234,-0.0101 0.006,-0.0135 0.11264,-0.0967 0.12407,-0.0967 0.005,0 0.0174,-0.008 0.0277,-0.0167 0.0442,-0.0396 0.0593,-0.05 0.0724,-0.05 0.008,0 0.014,-0.006 0.014,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0137 0,-0.008 0.015,-0.0193 0.0334,-0.0264 0.0183,-0.007 0.0334,-0.0188 0.0334,-0.0264 0,-0.008 0.009,-0.0137 0.02,-0.0137 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.0204,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.013,-0.02 0.0194,-0.02 0.0112,0 0.0473,-0.0267 0.086,-0.0633 0.01,-0.009 0.024,-0.0167 0.0319,-0.0167 0.008,0 0.0143,-0.006 0.0143,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.006,-0.0119 0.0139,-0.0119 0.0131,0 0.0282,-0.0105 0.0724,-0.05 0.0102,-0.009 0.025,-0.0167 0.0328,-0.0167 0.008,0 0.0143,-0.009 0.0143,-0.02 0,-0.011 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0131,-0.0166 0.029,-0.0216 0.0161,-0.005 0.0445,-0.0235 0.0633,-0.041 0.0189,-0.0175 0.0418,-0.0317 0.051,-0.0317 0.009,0 0.0169,-0.005 0.0169,-0.01 0,-0.005 0.0315,-0.031 0.07,-0.0567 0.0384,-0.0257 0.0849,-0.0602 0.10325,-0.0767 0.0387,-0.0348 0.0646,-0.0383 0.0755,-0.01 0.005,0.0111 0.0134,0.02 0.0205,0.02 0.0173,0 0.051,0.0346 0.051,0.0524 0,0.008 0.007,0.0143 0.0144,0.0143 0.008,0 0.0324,0.018 0.0544,0.0401 0.022,0.022 0.0453,0.04 0.0517,0.04 0.007,0 0.0574,0.045 0.11343,0.10008 0.0559,0.0551 0.10637,0.10008 0.11206,0.10008 0.006,0 0.0265,0.015 0.0463,0.0334 0.0198,0.0183 0.0417,0.0334 0.0487,0.0334 0.007,0 0.0127,0.009 0.0127,0.02 0,0.0111 0.006,0.02 0.0139,0.02 0.0149,0 0.0316,0.0125 0.0941,0.0701 0.0219,0.0202 0.0433,0.0367 0.0477,0.0367 0.005,0 0.0278,0.021 0.052,0.0468 0.0242,0.0257 0.0504,0.0468 0.0583,0.0468 0.008,0 0.0142,0.009 0.0142,0.02 0,0.011 0.006,0.02 0.014,0.02 0.008,0 0.0279,0.0135 0.0449,0.03 0.017,0.0165 0.0422,0.0367 0.056,0.0448 0.0215,0.0127 0.0225,0.0178 0.007,0.0367 -0.01,0.0121 -0.0264,0.0219 -0.0367,0.0219 -0.0101,0 -0.0184,0.009 -0.0184,0.02 0,0.0111 -0.005,0.02 -0.0119,0.02 -0.007,0 -0.0261,0.0149 -0.0434,0.0332 -0.0173,0.0183 -0.0377,0.0332 -0.0453,0.0334 -0.008,1.5e-4 -0.0362,0.0212 -0.0635,0.0469 -0.0272,0.0257 -0.054,0.0468 -0.0595,0.0468 -0.005,5e-5 -0.0292,0.0181 -0.0526,0.04 -0.0235,0.022 -0.049,0.04 -0.0567,0.04 -0.008,0 -0.0141,0.009 -0.0141,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.007,0.0134 -0.0143,0.0134 -0.008,0 -0.0227,0.008 -0.0329,0.0167 -0.04,0.0359 -0.0587,0.05 -0.0662,0.05 -0.009,0 -0.0213,0.01 -0.0764,0.0575 -0.0199,0.0174 -0.052,0.0431 -0.0711,0.0571 -0.0348,0.0255 -0.0786,0.0598 -0.1088,0.0853 -0.009,0.008 -0.0223,0.0136 -0.03,0.0136 -0.008,0 -0.014,0.009 -0.014,0.02 0,0.0111 -0.009,0.02 -0.0189,0.02 -0.0105,0 -0.0208,0.005 -0.0234,0.01 -0.008,0.0165 -0.11042,0.0902 -0.12121,0.0867 -0.005,-0.002 -0.01,0.006 -0.01,0.0167 0,0.0111 -0.008,0.02 -0.0169,0.02 -0.009,0 -0.0281,0.0151 -0.0419,0.0334 -0.0138,0.0183 -0.0334,0.0334 -0.0436,0.0334 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0194,0.02 -0.0338,0.02 -0.0143,0 -0.0223,0.006 -0.0178,0.0134 0.005,0.008 9e-5,0.0134 -0.01,0.0134 -0.01,0 -0.0432,0.0241 -0.0736,0.0533 -0.0306,0.0293 -0.0618,0.0533 -0.0694,0.0533 -0.008,0 -0.0138,0.005 -0.0138,0.0111 0,0.006 -0.0285,0.0315 -0.0633,0.0565 -0.0348,0.0249 -0.0754,0.0581 -0.0901,0.0738 -0.0147,0.0157 -0.0402,0.0326 -0.0567,0.0378 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.006,0.0125 -0.014,0.0125 -0.008,0 -0.0301,0.015 -0.0499,0.0334 -0.0198,0.0183 -0.0469,0.0334 -0.0603,0.0334 -0.0134,0 -0.0212,0.005 -0.0175,0.0109 0.008,0.0122 -0.0382,0.0558 -0.0587,0.0558 -0.007,0 -0.0132,0.008 -0.0132,0.0171 0,0.009 -0.0191,0.0284 -0.0425,0.0422 -0.0397,0.0234 -0.0852,0.0588 -0.12519,0.0974 -0.0226,0.0219 -0.0198,0.023 -0.0659,-0.0297 z m -1.31892,-1.34259 c -0.0415,-0.0434 -0.0754,-0.0814 -0.0754,-0.0845 0,-0.0111 -0.0824,-0.0877 -0.0944,-0.0877 -0.007,0 -0.0123,-0.009 -0.0123,-0.02 0,-0.0111 -0.007,-0.02 -0.0143,-0.02 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.007,-0.0143 -0.0143,-0.0143 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.009,-0.0143 -0.02,-0.0143 -0.0111,0 -0.02,-0.009 -0.02,-0.02 0,-0.0111 -0.006,-0.02 -0.0138,-0.02 -0.014,0 -0.093,-0.075 -0.093,-0.0881 0,-0.0128 -0.0377,-0.0453 -0.0524,-0.0453 -0.008,0 -0.0143,-0.007 -0.0143,-0.0154 0,-0.017 -0.10466,-0.11804 -0.12232,-0.11804 -0.006,0 -0.0111,-0.007 -0.0111,-0.0144 0,-0.008 -0.0219,-0.0357 -0.0487,-0.0618 l -0.0487,-0.0474 0.0321,-0.0247 c 0.0177,-0.0135 0.039,-0.0248 0.0474,-0.0249 0.009,-1.5e-4 0.0187,-0.009 0.0229,-0.0203 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.012,-0.0134 0.0267,-0.0134 0.0148,0 0.0267,-0.009 0.0267,-0.02 0,-0.0111 0.009,-0.02 0.0184,-0.02 0.0101,0 0.0254,-0.009 0.034,-0.0187 0.009,-0.0104 0.0363,-0.0286 0.0616,-0.0406 0.0254,-0.0121 0.0461,-0.0276 0.0461,-0.0346 0,-0.007 0.0144,-0.0128 0.0319,-0.0128 0.0176,0 0.0355,-0.009 0.0399,-0.0209 0.005,-0.0115 0.0141,-0.0171 0.0214,-0.0126 0.008,0.005 0.0135,-7.6e-4 0.0135,-0.0125 0,-0.0115 0.009,-0.0208 0.02,-0.0208 0.0111,0 0.02,-0.005 0.02,-0.0108 0,-0.006 0.0241,-0.0214 0.0534,-0.0343 0.0293,-0.0129 0.0609,-0.0306 0.0701,-0.0393 0.009,-0.009 0.0391,-0.0276 0.0667,-0.0421 0.0275,-0.0144 0.05,-0.031 0.05,-0.0367 0,-0.006 0.009,-0.0102 0.02,-0.0102 0.0111,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0135,-0.0167 0.03,-0.0219 0.0165,-0.005 0.0432,-0.0236 0.0594,-0.0409 0.0162,-0.0173 0.0389,-0.0315 0.0504,-0.0315 0.0116,0 0.0243,-0.009 0.0283,-0.0189 0.005,-0.0104 0.0346,-0.0269 0.068,-0.0368 0.0334,-0.01 0.0636,-0.0254 0.0672,-0.0345 0.003,-0.009 0.0126,-0.0167 0.0201,-0.0167 0.008,0 0.0273,-0.0147 0.0439,-0.0326 0.0166,-0.0179 0.0328,-0.0299 0.0361,-0.0266 0.003,0.003 0.0169,-0.006 0.0302,-0.0208 0.0134,-0.0147 0.0322,-0.0268 0.0422,-0.0268 0.01,0 0.0179,-0.005 0.0179,-0.0102 0,-0.0102 0.0396,-0.0343 0.0834,-0.0509 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.009,-0.0118 0.0205,-0.0118 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0221,-0.02 0.0397,-0.02 0.0176,0 0.0319,-0.007 0.0319,-0.0152 0,-0.0205 0.0412,-0.0505 0.0701,-0.0511 0.0128,-2.6e-4 0.0234,-0.009 0.0234,-0.0205 0,-0.0111 0.0119,-0.02 0.0267,-0.02 0.0147,0 0.0267,-0.006 0.0267,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0109 0.0184,-0.02 0.0315,-0.02 0.0131,0 0.0273,-0.009 0.0315,-0.02 0.005,-0.0111 0.0141,-0.02 0.0218,-0.02 0.008,0 0.0432,-0.0196 0.0788,-0.0434 0.0356,-0.0239 0.0847,-0.0554 0.10912,-0.0701 0.0244,-0.0147 0.0552,-0.0332 0.0683,-0.0412 0.0132,-0.008 0.036,-0.0254 0.0507,-0.0388 0.0147,-0.0134 0.0268,-0.0196 0.0268,-0.0137 0,0.006 0.0159,-0.005 0.0354,-0.0249 0.0196,-0.0196 0.0495,-0.039 0.0667,-0.0433 0.0172,-0.005 0.0312,-0.0132 0.0312,-0.0198 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0109,0 0.02,-0.005 0.02,-0.0121 0,-0.007 0.0165,-0.0179 0.0368,-0.0253 0.0202,-0.007 0.0499,-0.0258 0.0662,-0.0413 0.0162,-0.0155 0.0381,-0.0282 0.0486,-0.0282 0.0105,0 0.0226,-0.009 0.0268,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0129,-0.0165 0.0288,-0.0215 0.0158,-0.005 0.042,-0.0235 0.0582,-0.0409 0.0162,-0.0175 0.0433,-0.0318 0.0602,-0.0318 0.0169,0 0.0267,-0.005 0.0219,-0.009 -0.008,-0.008 0.0192,-0.0283 0.0711,-0.0531 0.0111,-0.005 0.0248,-0.0175 0.0306,-0.0272 0.006,-0.01 0.0218,-0.0177 0.0353,-0.0177 0.0135,0 0.0281,-0.009 0.0322,-0.02 0.005,-0.011 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.005 0.0204,-0.0118 0,-0.007 0.0105,-0.0156 0.0234,-0.0203 0.0327,-0.0118 0.0797,-0.0403 0.11527,-0.0697 0.0432,-0.0357 0.078,-0.058 0.10491,-0.067 0.0128,-0.005 0.0234,-0.0124 0.0234,-0.018 0,-0.006 0.0105,-0.0138 0.0234,-0.0181 0.0313,-0.0105 0.0651,-0.0326 0.12631,-0.0827 0.009,-0.008 0.0261,-0.0183 0.0371,-0.0234 0.0111,-0.005 0.038,-0.021 0.0601,-0.035 0.022,-0.0141 0.0539,-0.026 0.0708,-0.0265 0.0171,-5.1e-4 0.0271,-0.007 0.0225,-0.0144 -0.005,-0.008 0.002,-0.0176 0.0159,-0.0226 0.0133,-0.005 0.0315,-0.0155 0.0404,-0.0233 0.0468,-0.0406 0.0578,-0.0474 0.0766,-0.0474 0.0113,0 0.0206,-0.005 0.0206,-0.0106 0,-0.006 0.0239,-0.0254 0.053,-0.0434 0.0292,-0.018 0.0532,-0.0371 0.0533,-0.0427 2.1e-4,-0.005 0.0118,-0.01 0.0256,-0.01 0.0138,0 0.0283,-0.008 0.0321,-0.0179 0.003,-0.01 0.0284,-0.0282 0.0548,-0.0406 0.0264,-0.0125 0.048,-0.0285 0.048,-0.0354 0,-0.007 0.009,-0.0127 0.02,-0.0127 0.0111,0 0.02,-0.005 0.02,-0.0102 0,-0.0102 0.0396,-0.0343 0.0834,-0.0509 0.0128,-0.005 0.0234,-0.0141 0.0234,-0.0206 0,-0.007 0.008,-0.0118 0.018,-0.0118 0.01,0 0.0354,-0.0151 0.0567,-0.0335 0.0213,-0.0184 0.0503,-0.0371 0.0645,-0.0417 0.0141,-0.005 0.0411,-0.0225 0.0599,-0.0399 0.0189,-0.0175 0.0418,-0.0317 0.051,-0.0317 0.009,0 0.0168,-0.005 0.0168,-0.0109 0,-0.006 0.027,-0.0245 0.06,-0.041 0.0331,-0.0165 0.06,-0.0342 0.06,-0.0392 0,-0.009 0.0109,-0.0163 0.10117,-0.0623 0.0287,-0.0147 0.0544,-0.0313 0.0568,-0.037 0.002,-0.006 0.028,-0.0223 0.0567,-0.037 0.0286,-0.0147 0.0521,-0.0325 0.0521,-0.0396 0,-0.007 0.0144,-0.0131 0.0319,-0.0131 0.0176,0 0.0354,-0.009 0.0397,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.01,-0.0134 0.0219,-0.0134 0.012,0 0.0357,-0.015 0.0527,-0.0334 0.017,-0.0183 0.0404,-0.0334 0.052,-0.0334 0.0115,0 0.0244,-0.009 0.0287,-0.02 0.005,-0.0111 0.016,-0.02 0.0263,-0.02 0.0102,0 0.0186,-0.006 0.0186,-0.0124 0,-0.007 0.0135,-0.0167 0.03,-0.0219 0.0165,-0.005 0.0331,-0.0134 0.0367,-0.0182 0.0108,-0.0144 0.17561,-0.12104 0.1869,-0.12104 0.006,0 0.0139,-0.009 0.018,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.009,-0.0134 0.021,-0.0134 0.0115,0 0.0299,-0.009 0.041,-0.02 0.011,-0.0111 0.0306,-0.02 0.0435,-0.02 0.0129,0 0.0307,-0.0135 0.0396,-0.03 0.009,-0.0165 0.0399,-0.0415 0.0689,-0.0555 0.029,-0.0141 0.0528,-0.0306 0.0528,-0.0368 0,-0.0247 0.0756,-0.009 0.10964,0.0222 0.0198,0.0183 0.0455,0.0334 0.057,0.0334 0.0115,0 0.0245,0.009 0.0286,0.02 0.005,0.011 0.016,0.02 0.0263,0.02 0.0102,0 0.0186,0.006 0.0186,0.0134 0,0.008 0.0144,0.0134 0.0319,0.0134 0.0176,0 0.0354,0.009 0.0397,0.02 0.005,0.0109 0.0169,0.02 0.028,0.02 0.0112,0 0.0204,0.006 0.0204,0.0131 0,0.007 0.015,0.0169 0.0333,0.0214 0.0183,0.005 0.0299,0.0137 0.026,0.0203 -0.005,0.007 -1.6e-4,0.0119 0.009,0.0119 0.009,0 0.054,0.0241 0.10054,0.0533 0.0465,0.0293 0.0902,0.0533 0.0971,0.0533 0.007,0 0.0145,0.005 0.017,0.01 0.007,0.0145 0.12964,0.0967 0.14494,0.0967 0.007,0 0.0129,0.006 0.0129,0.0134 0,0.008 0.0122,0.0134 0.0271,0.0134 0.0149,0 0.0306,0.009 0.0348,0.02 0.005,0.011 0.016,0.02 0.0263,0.02 0.0102,0 0.0186,0.006 0.0186,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0124 0,0.007 -0.0131,0.0166 -0.029,0.0217 -0.016,0.005 -0.0445,0.0234 -0.0633,0.041 -0.0189,0.0175 -0.039,0.0317 -0.0449,0.0317 -0.006,0 -0.0248,0.0151 -0.0417,0.0334 -0.017,0.0183 -0.0377,0.0334 -0.0461,0.0334 -0.009,0 -0.0151,0.006 -0.0151,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.009 -0.02,0.0184 0,0.0102 -0.0105,0.022 -0.0234,0.0263 -0.025,0.009 -0.0724,0.0368 -0.0834,0.05 -0.003,0.005 -0.0324,0.0241 -0.0639,0.0439 -0.0314,0.0197 -0.0692,0.0478 -0.0839,0.0625 -0.0147,0.0147 -0.0325,0.0231 -0.0396,0.0187 -0.007,-0.005 -0.0128,0.002 -0.0128,0.0128 0,0.0115 -0.009,0.0208 -0.02,0.0208 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.0134,0.0134 -0.008,0 -0.0425,0.0209 -0.078,0.0464 -0.0356,0.0256 -0.0687,0.0426 -0.0734,0.0378 -0.005,-0.005 -0.009,-0.002 -0.009,0.007 0,0.009 -0.006,0.0156 -0.014,0.0156 -0.0154,0 -0.0366,0.0155 -0.0953,0.0701 -0.0218,0.0202 -0.0448,0.0367 -0.0512,0.0367 -0.006,0 -0.0164,0.008 -0.0223,0.0167 -0.006,0.009 -0.0286,0.0257 -0.0506,0.0367 -0.0219,0.011 -0.0526,0.0335 -0.0682,0.05 -0.0156,0.0165 -0.0384,0.03 -0.0506,0.03 -0.0122,0 -0.0257,0.009 -0.0299,0.02 -0.005,0.011 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.006 -0.0186,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.0208,0.0119 -0.0115,0 -0.0173,0.006 -0.0132,0.0124 0.005,0.007 -0.006,0.0167 -0.0225,0.0219 -0.0166,0.005 -0.0434,0.0236 -0.0596,0.041 -0.0162,0.0174 -0.038,0.0315 -0.0487,0.0315 -0.0105,0 -0.0227,0.009 -0.0269,0.02 -0.005,0.0109 -0.0169,0.02 -0.0281,0.02 -0.0112,0 -0.0205,0.005 -0.0205,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0144 -0.065,0.0399 -0.0833,0.0564 -0.0183,0.0165 -0.0381,0.03 -0.044,0.03 -0.006,0 -0.0247,0.015 -0.0417,0.0334 -0.017,0.0183 -0.037,0.0334 -0.0446,0.0334 -0.008,0 -0.0172,0.009 -0.0214,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0134 0,0.008 -0.009,0.0134 -0.0186,0.0134 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0158,0.02 -0.0257,0.02 -0.01,0 -0.0382,0.018 -0.0629,0.0401 -0.0247,0.022 -0.0483,0.04 -0.0526,0.04 -0.005,0 -0.0317,0.0209 -0.061,0.0464 -0.0293,0.0255 -0.061,0.0466 -0.0705,0.0467 -0.009,1.5e-4 -0.0306,0.0142 -0.0467,0.0312 -0.0162,0.017 -0.0384,0.0352 -0.0494,0.0402 -0.0111,0.005 -0.0381,0.0212 -0.0601,0.0359 -0.022,0.0147 -0.0505,0.0303 -0.0633,0.0348 -0.0128,0.005 -0.0234,0.0134 -0.0234,0.0199 0,0.007 -0.007,0.0118 -0.0143,0.0118 -0.008,0 -0.0227,0.008 -0.0328,0.0167 -0.0442,0.0397 -0.0593,0.05 -0.0724,0.05 -0.008,0 -0.0138,0.009 -0.0138,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0132 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0128 -0.0334,0.0182 0,0.005 -0.0211,0.0212 -0.0468,0.0349 -0.0257,0.0139 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0103 -0.0186,0.0103 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0166,0.02 -0.0275,0.02 -0.0109,0 -0.0342,0.0143 -0.0516,0.0318 -0.0175,0.0175 -0.0448,0.0359 -0.0606,0.041 -0.0159,0.005 -0.0289,0.0147 -0.0289,0.0215 0,0.007 -0.009,0.0125 -0.02,0.0125 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.008,0 -0.0302,0.0152 -0.05,0.0338 -0.0199,0.0186 -0.0812,0.0618 -0.13619,0.0961 -0.0551,0.0342 -0.10209,0.067 -0.10452,0.073 -0.002,0.006 -0.0129,0.0107 -0.0234,0.0107 -0.0105,0 -0.0189,0.005 -0.019,0.01 -6e-5,0.005 -0.0226,0.0249 -0.05,0.0432 -0.0275,0.0183 -0.0594,0.0433 -0.0707,0.0558 -0.0115,0.0125 -0.037,0.0267 -0.0567,0.0317 -0.0198,0.005 -0.0359,0.0144 -0.0359,0.0209 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.005 -0.02,0.0103 0,0.006 -0.0211,0.0215 -0.0468,0.0354 -0.0257,0.0139 -0.0468,0.0332 -0.0468,0.0431 0,0.01 -0.0121,0.0179 -0.0271,0.0179 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.005 -0.0186,0.0112 0,0.006 -0.0211,0.0225 -0.0468,0.0364 -0.0257,0.0138 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0102 -0.02,0.0102 -0.0111,0 -0.02,0.006 -0.02,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0264 -0.0183,0.007 -0.0334,0.0188 -0.0334,0.0264 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.0151,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.007,0.0119 -0.0151,0.0119 -0.009,0 -0.0278,0.0135 -0.0434,0.03 -0.0155,0.0165 -0.0612,0.0509 -0.10161,0.0763 -0.0403,0.0254 -0.0764,0.0494 -0.08,0.0533 -0.0125,0.0133 -0.0599,0.0412 -0.0834,0.049 -0.0128,0.005 -0.0234,0.0134 -0.0234,0.02 0,0.007 -0.0106,0.0162 -0.0234,0.0211 -0.0128,0.005 -0.0306,0.0151 -0.0396,0.0229 -0.0428,0.0372 -0.0567,0.0474 -0.0643,0.0474 -0.005,0 -0.0165,0.008 -0.0268,0.0167 -0.0447,0.0401 -0.0594,0.05 -0.0737,0.05 -0.017,0 -0.17693,0.11791 -0.18389,0.13552 -0.002,0.006 -0.0129,0.0113 -0.0234,0.0113 -0.0105,0 -0.019,0.005 -0.019,0.0118 0,0.007 -0.0105,0.0153 -0.0234,0.0196 -0.0313,0.0105 -0.0686,0.0357 -0.10657,0.072 -0.0173,0.0165 -0.0374,0.03 -0.0448,0.03 -0.008,0 -0.0267,0.0135 -0.0428,0.03 -0.0339,0.0345 -0.0725,0.0607 -0.10602,0.072 -0.0128,0.005 -0.0234,0.0132 -0.0234,0.0196 0,0.007 -0.006,0.0118 -0.014,0.0118 -0.0151,0 -0.0419,0.0199 -0.0767,0.0567 -0.0121,0.0128 -0.0297,0.0234 -0.0391,0.0234 -0.009,0 -0.017,0.006 -0.017,0.0125 0,0.007 -0.0121,0.0163 -0.0269,0.0209 -0.0148,0.005 -0.0235,0.0141 -0.0193,0.0209 0.005,0.007 -0.008,0.0124 -0.0274,0.0124 -0.0199,0 -0.0314,0.006 -0.0267,0.0134 0.005,0.008 -7.6e-4,0.0134 -0.0126,0.0134 -0.0115,0 -0.0208,0.006 -0.0208,0.0125 0,0.007 -0.0129,0.0165 -0.0287,0.0215 -0.0158,0.005 -0.0413,0.0225 -0.0567,0.0386 -0.0507,0.0534 -0.0658,0.0499 -0.14595,-0.0338 z m 3.73064,-0.57254 c -0.0306,-0.0293 -0.0618,-0.0533 -0.0694,-0.0533 -0.008,0 -0.0139,-0.006 -0.0139,-0.0134 0,-0.008 -0.005,-0.0134 -0.012,-0.0134 -0.0122,0 -0.0685,-0.0452 -0.13702,-0.11008 -0.0213,-0.0202 -0.0435,-0.0367 -0.0493,-0.0367 -0.006,0 -0.0186,-0.0151 -0.0285,-0.0334 -0.01,-0.0183 -0.0263,-0.0334 -0.0367,-0.0334 -0.0104,0 -0.0363,-0.0176 -0.0577,-0.039 -0.0215,-0.0215 -0.039,-0.0354 -0.039,-0.0312 0,0.005 -0.0225,-0.0164 -0.05,-0.0462 -0.0276,-0.0297 -0.0621,-0.0566 -0.0767,-0.0597 -0.0147,-0.003 -0.0251,-0.0128 -0.0234,-0.0215 0.002,-0.009 -0.002,-0.0159 -0.0106,-0.0159 -0.008,0 -0.0279,-0.015 -0.0448,-0.0334 -0.017,-0.0183 -0.0343,-0.0334 -0.0384,-0.0334 -0.005,0 -0.0298,-0.021 -0.057,-0.0468 -0.0272,-0.0257 -0.0538,-0.0467 -0.0591,-0.0467 -0.0147,0 -0.0801,-0.0639 -0.0801,-0.0783 0,-0.007 -0.008,-0.0159 -0.0167,-0.0196 -0.0102,-0.005 -0.008,-0.007 0.006,-0.008 0.0125,-5.1e-4 0.0422,-0.0191 0.0664,-0.0411 0.0242,-0.022 0.0516,-0.0401 0.0609,-0.0401 0.009,0 0.017,-0.006 0.017,-0.0134 0,-0.008 0.009,-0.0134 0.021,-0.0134 0.0115,0 0.0293,-0.008 0.0396,-0.0167 0.0442,-0.0396 0.0592,-0.05 0.0724,-0.05 0.008,0 0.0139,-0.009 0.0139,-0.02 0,-0.0126 0.0127,-0.02 0.0341,-0.02 0.0189,0 0.0306,-0.006 0.0263,-0.0127 -0.005,-0.007 0.005,-0.0189 0.0182,-0.0267 0.0143,-0.008 0.034,-0.0215 0.0438,-0.0307 0.01,-0.009 0.0402,-0.0318 0.0676,-0.0501 0.0275,-0.0183 0.05,-0.0378 0.05,-0.0433 5e-5,-0.005 0.006,-0.01 0.014,-0.01 0.008,0 0.027,-0.0122 0.0431,-0.0272 0.0161,-0.015 0.0428,-0.0306 0.0595,-0.0348 0.0167,-0.005 0.0303,-0.0164 0.0303,-0.027 0,-0.0106 0.005,-0.016 0.0119,-0.012 0.007,0.005 0.0154,-0.002 0.0196,-0.0126 0.005,-0.0109 0.02,-0.0199 0.0352,-0.0199 0.0151,0 0.031,-0.009 0.0352,-0.02 0.005,-0.0111 0.0164,-0.02 0.027,-0.02 0.0107,0 0.0157,-0.006 0.0112,-0.0134 -0.005,-0.008 -0.002,-0.0134 0.007,-0.0134 0.0158,0 0.0423,-0.0193 0.0775,-0.0567 0.0121,-0.0128 0.0312,-0.0235 0.0425,-0.0238 0.0208,-5.1e-4 0.0813,-0.0396 0.0611,-0.0396 -0.006,0 0.0111,-0.015 0.0381,-0.0334 0.027,-0.0183 0.0549,-0.0334 0.0619,-0.0334 0.007,0 0.0128,-0.006 0.0128,-0.0134 0,-0.008 0.006,-0.0134 0.0128,-0.0134 0.007,0 0.0313,-0.018 0.0539,-0.0401 0.0225,-0.022 0.0475,-0.04 0.0555,-0.04 0.008,0 0.0244,-0.0105 0.0362,-0.0234 0.0118,-0.0128 0.0448,-0.0352 0.0732,-0.0497 0.0283,-0.0144 0.0545,-0.034 0.0581,-0.0434 0.003,-0.009 0.0134,-0.017 0.0219,-0.017 0.009,0 0.0316,-0.015 0.0513,-0.0334 0.0198,-0.0183 0.0409,-0.0334 0.0468,-0.0334 0.006,0 0.0269,-0.015 0.0468,-0.0334 0.0198,-0.0183 0.0425,-0.0334 0.0504,-0.0334 0.008,0 0.0178,-0.009 0.022,-0.02 0.005,-0.0111 0.0154,-0.02 0.0248,-0.02 0.009,0 0.0206,-0.009 0.0248,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0204,-0.006 0.0204,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.005 0.02,-0.0102 0,-0.006 0.0241,-0.025 0.0533,-0.0432 0.0293,-0.0182 0.0533,-0.0376 0.0533,-0.0432 0,-0.006 0.007,-0.0102 0.0143,-0.0102 0.008,0 0.0227,-0.008 0.0328,-0.0167 0.0408,-0.0367 0.0588,-0.05 0.0671,-0.05 0.0112,0 0.11784,-0.0834 0.12368,-0.0966 0.002,-0.005 0.0129,-0.0101 0.0234,-0.0101 0.0105,0 0.0189,-0.005 0.0189,-0.0112 0,-0.006 0.024,-0.0227 0.0532,-0.0368 0.0293,-0.014 0.0532,-0.0316 0.0533,-0.0391 1e-4,-0.008 0.0209,-0.0248 0.0462,-0.0384 0.0254,-0.0136 0.0597,-0.0375 0.0764,-0.0531 0.0167,-0.0155 0.0335,-0.0286 0.0374,-0.0291 0.0167,-0.002 0.0936,-0.0534 0.0936,-0.0625 4e-5,-0.005 0.0121,-0.01 0.0268,-0.01 0.0148,0 0.0267,-0.009 0.0267,-0.02 0,-0.0109 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.011 0.0164,-0.02 0.027,-0.02 0.0106,0 0.0213,-0.005 0.0238,-0.01 0.008,-0.0167 0.11532,-0.0967 0.13043,-0.0967 0.008,0 0.0141,-0.005 0.0141,-0.0105 0,-0.006 0.0195,-0.0219 0.0433,-0.0357 0.0239,-0.0138 0.0464,-0.0289 0.05,-0.0332 0.0186,-0.0225 0.11921,-0.0941 0.13202,-0.0941 0.008,0 0.0147,-0.006 0.0147,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.024,-0.0238 0.0533,-0.0364 0.0293,-0.0127 0.0533,-0.0307 0.0533,-0.04 0,-0.009 0.007,-0.0169 0.0143,-0.0169 0.008,0 0.0227,-0.008 0.0328,-0.0167 0.0423,-0.0379 0.059,-0.05 0.069,-0.0501 0.006,-3e-5 0.0298,-0.018 0.0533,-0.04 0.0234,-0.022 0.049,-0.04 0.0567,-0.04 0.008,0 0.0141,-0.006 0.0141,-0.0134 0,-0.008 0.009,-0.0134 0.0189,-0.0134 0.0105,0 0.0208,-0.005 0.0234,-0.01 0.007,-0.0148 0.12679,-0.0967 0.14197,-0.0967 0.007,0 0.0165,-0.009 0.0207,-0.02 0.005,-0.0111 0.0141,-0.02 0.0221,-0.02 0.008,0 0.0293,-0.015 0.0477,-0.0334 0.0183,-0.0183 0.0393,-0.0334 0.0466,-0.0334 0.007,0 0.0308,-0.018 0.0523,-0.04 0.0215,-0.022 0.047,-0.0401 0.0567,-0.0401 0.01,0 0.0262,-0.012 0.0368,-0.0267 0.0106,-0.0148 0.0269,-0.0267 0.0363,-0.0267 0.009,0 0.017,-0.006 0.017,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.009 0.02,-0.0185 0,-0.0102 0.0105,-0.022 0.0234,-0.0263 0.0299,-0.01 0.0665,-0.0342 0.10925,-0.072 0.0186,-0.0165 0.0416,-0.03 0.0508,-0.03 0.009,0 0.0168,-0.006 0.0168,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0134,-0.02 0.0205,-0.02 0.007,0 0.0212,-0.008 0.0315,-0.0167 0.037,-0.0332 0.0537,-0.0452 0.0984,-0.0708 0.025,-0.0143 0.0695,-0.0485 0.0989,-0.076 0.0293,-0.0276 0.0613,-0.05 0.0712,-0.05 0.01,0 0.0179,-0.005 0.0179,-0.0105 0,-0.0107 0.1639,-0.12293 0.17953,-0.12293 0.005,0 0.0225,-0.015 0.0395,-0.0334 0.0305,-0.0328 0.0879,-0.046 0.0879,-0.02 0,0.008 0.009,0.0134 0.0186,0.0134 0.0102,0 0.0221,0.009 0.0263,0.02 0.005,0.0111 0.0199,0.02 0.0348,0.02 0.0149,0 0.0271,0.005 0.0271,0.0119 0,0.007 0.0211,0.0158 0.0468,0.0206 0.0257,0.005 0.0468,0.0144 0.0468,0.0215 0,0.007 0.012,0.0128 0.0267,0.0128 0.0147,0 0.0267,0.005 0.0267,0.0108 0,0.006 0.0255,0.0217 0.0567,0.035 0.0312,0.0134 0.0642,0.0336 0.0733,0.0452 0.0108,0.0135 0.0167,0.0151 0.0167,0.005 0,-0.0101 0.0102,-0.007 0.0269,0.009 0.0275,0.0256 0.0606,0.0448 0.11986,0.0693 0.0183,0.008 0.0513,0.0257 0.0733,0.0402 0.022,0.0144 0.049,0.0299 0.06,0.0343 0.0506,0.0203 0.0882,0.0403 0.10436,0.0556 0.01,0.009 0.0262,0.0167 0.0367,0.0167 0.0105,0 0.0191,0.006 0.0191,0.0134 0,0.008 0.015,0.0134 0.0334,0.0134 0.0183,0 0.0332,0.005 0.0329,0.01 -5.2e-4,0.0144 -0.0342,0.0567 -0.045,0.0567 -0.005,0 -0.0231,0.015 -0.0401,0.0334 -0.017,0.0183 -0.0352,0.0334 -0.0403,0.0334 -0.005,0 -0.0178,0.008 -0.028,0.0167 -0.0401,0.036 -0.0587,0.05 -0.0662,0.05 -0.005,0 -0.0211,0.0139 -0.0372,0.0309 -0.0162,0.017 -0.0504,0.0436 -0.0761,0.0593 -0.0257,0.0157 -0.0555,0.0409 -0.0664,0.0559 -0.0108,0.0151 -0.0272,0.0275 -0.0365,0.0275 -0.009,0 -0.0331,0.018 -0.0526,0.04 -0.0196,0.022 -0.0437,0.04 -0.0535,0.04 -0.01,0 -0.0178,0.006 -0.0178,0.0134 0,0.008 -0.008,0.0134 -0.0173,0.0134 -0.009,0 -0.0423,0.0241 -0.0729,0.0533 -0.0306,0.0293 -0.0613,0.0533 -0.0683,0.0533 -0.007,0 -0.0147,0.005 -0.0172,0.0106 -0.002,0.006 -0.0311,0.0313 -0.0636,0.0567 -0.0326,0.0254 -0.069,0.0552 -0.0811,0.0662 -0.0406,0.0371 -0.099,0.0801 -0.10895,0.0801 -0.005,0 -0.0351,0.0241 -0.066,0.0533 -0.0308,0.0293 -0.062,0.0533 -0.0691,0.0533 -0.007,0 -0.0269,0.015 -0.0439,0.0334 -0.017,0.0183 -0.0347,0.0334 -0.0391,0.0334 -0.005,0 -0.0166,0.008 -0.0269,0.0167 -0.0442,0.0396 -0.0593,0.05 -0.0724,0.05 -0.008,0 -0.0139,0.009 -0.0139,0.02 0,0.0111 -0.006,0.02 -0.014,0.02 -0.008,0 -0.0302,0.015 -0.0499,0.0334 -0.0198,0.0183 -0.0436,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.008,0.02 -0.018,0.02 -0.01,0 -0.0397,0.0211 -0.066,0.0468 -0.0264,0.0257 -0.0537,0.0468 -0.0604,0.0468 -0.007,0 -0.0355,0.0212 -0.064,0.0473 -0.0283,0.026 -0.0652,0.0506 -0.0818,0.0548 -0.0166,0.005 -0.0301,0.0159 -0.0301,0.0261 0,0.0102 -0.007,0.0186 -0.0144,0.0186 -0.008,0 -0.0382,0.024 -0.0672,0.0533 -0.0291,0.0293 -0.0594,0.0533 -0.0675,0.0533 -0.008,0 -0.0285,0.015 -0.0455,0.0334 -0.017,0.0183 -0.0361,0.0334 -0.0426,0.0334 -0.006,0 -0.0257,0.015 -0.043,0.0334 -0.0172,0.0183 -0.0368,0.0334 -0.0434,0.0334 -0.007,0 -0.0307,0.0186 -0.0535,0.0415 -0.0228,0.0228 -0.0644,0.0566 -0.0924,0.0752 -0.028,0.0185 -0.051,0.0389 -0.051,0.0452 0,0.006 -0.009,0.0115 -0.02,0.0115 -0.011,0 -0.02,0.005 -0.02,0.0105 0,0.006 -0.0165,0.0205 -0.0368,0.033 -0.0202,0.0123 -0.0577,0.0406 -0.0834,0.0627 -0.0759,0.0655 -0.0783,0.0673 -0.0928,0.0673 -0.008,0 -0.014,0.006 -0.014,0.0125 0,0.007 -0.0121,0.0163 -0.0269,0.0209 -0.0148,0.005 -0.0235,0.0141 -0.0193,0.0209 0.005,0.007 -0.002,0.0124 -0.0132,0.0124 -0.0115,0 -0.0208,0.006 -0.0208,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0264 -0.0183,0.007 -0.0334,0.0188 -0.0334,0.0264 0,0.008 -0.009,0.0137 -0.02,0.0137 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.006,0.0134 -0.014,0.0134 -0.008,0 -0.0301,0.0151 -0.0499,0.0334 -0.0198,0.0183 -0.0436,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.009,0.02 -0.0186,0.02 -0.0102,0 -0.0221,0.009 -0.0263,0.02 -0.005,0.011 -0.0203,0.02 -0.0356,0.02 -0.0154,0 -0.0244,0.006 -0.0202,0.0124 0.005,0.007 -0.0105,0.0232 -0.0326,0.0362 -0.0221,0.0131 -0.0403,0.0285 -0.0403,0.0342 0,0.006 -0.009,0.0105 -0.02,0.0105 -0.0111,0 -0.02,0.005 -0.02,0.0111 0,0.006 -0.0285,0.0314 -0.0633,0.0562 -0.0348,0.0248 -0.0694,0.0517 -0.0767,0.0597 -0.0243,0.0267 -0.11139,0.0865 -0.12587,0.0865 -0.008,0 -0.0142,0.009 -0.0142,0.02 0,0.0109 -0.007,0.02 -0.0151,0.02 -0.009,0 -0.029,0.015 -0.0461,0.0334 -0.017,0.0183 -0.0375,0.0334 -0.0455,0.0334 -0.008,0 -0.0285,0.0151 -0.0455,0.0334 -0.017,0.0183 -0.0347,0.0334 -0.0393,0.0334 -0.009,0 -0.0607,0.0417 -0.10427,0.0834 -0.0135,0.0128 -0.0304,0.0234 -0.0377,0.0234 -0.008,0 -0.0134,0.006 -0.0134,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.007 -0.02,0.0143 0,0.0185 -0.0339,0.0524 -0.0524,0.0524 -0.008,0 -0.0143,0.006 -0.0143,0.0137 0,0.008 -0.015,0.0193 -0.0334,0.0263 -0.0183,0.007 -0.0334,0.0189 -0.0334,0.0264 0,0.008 -0.007,0.0137 -0.0151,0.0137 -0.009,0 -0.0293,0.0154 -0.0468,0.0341 -0.0173,0.0187 -0.0315,0.0311 -0.0315,0.0274 0,-0.003 -0.0193,0.0111 -0.043,0.0327 -0.0236,0.0216 -0.0474,0.0393 -0.0528,0.0393 -0.005,0 -0.0343,0.024 -0.0643,0.0533 -0.0299,0.0293 -0.061,0.0533 -0.0691,0.0533 -0.008,0 -0.0285,0.015 -0.0455,0.0334 -0.017,0.0183 -0.0345,0.0334 -0.0388,0.0334 -0.005,0 -0.0241,0.0151 -0.0439,0.0334 -0.0198,0.0183 -0.0435,0.0334 -0.0528,0.0334 -0.009,0 -0.0168,0.009 -0.0168,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.005 -0.02,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0145 -0.0234,0.0215 0,0.0172 -0.0347,0.0509 -0.0524,0.0509 -0.008,0 -0.0143,0.009 -0.0143,0.0185 0,0.0101 -0.0105,0.0225 -0.0234,0.0272 -0.0438,0.0165 -0.0834,0.0406 -0.0834,0.0509 0,0.006 -0.007,0.0102 -0.0151,0.0102 -0.009,0 -0.028,0.0138 -0.0437,0.0307 -0.0157,0.017 -0.0389,0.0348 -0.0516,0.0397 -0.0127,0.005 -0.023,0.0151 -0.023,0.0226 0,0.0281 -0.0363,0.0121 -0.0902,-0.0396 z m -4.80688,-0.51843 c 0,-0.006 -0.0595,-0.0718 -0.13233,-0.14566 -0.0728,-0.0739 -0.12882,-0.13779 -0.12454,-0.14208 0.005,-0.005 0.002,-0.008 -0.007,-0.008 -0.0182,0 -0.10368,-0.083 -0.0996,-0.0967 0.002,-0.005 -0.002,-0.009 -0.007,-0.007 -0.0111,0.003 -0.0967,-0.0745 -0.0967,-0.0882 0,-0.0151 -0.0509,-0.06 -0.0589,-0.052 -0.005,0.005 -0.008,-0.002 -0.008,-0.0144 0,-0.0122 -0.007,-0.0222 -0.0143,-0.0222 -0.0185,0 -0.0524,-0.0339 -0.0524,-0.0524 0,-0.008 -0.009,-0.0143 -0.02,-0.0143 -0.0109,0 -0.02,-0.005 -0.02,-0.0115 0,-0.006 -0.0156,-0.027 -0.0345,-0.0461 -0.0368,-0.0368 -0.0336,-0.0566 0.0125,-0.0766 0.0352,-0.0154 0.0749,-0.0404 0.0992,-0.0626 0.01,-0.009 0.0268,-0.0167 0.0372,-0.0167 0.0105,0 0.0191,-0.005 0.0191,-0.0101 0,-0.005 0.0285,-0.0237 0.0633,-0.0404 0.0754,-0.0361 0.10607,-0.0548 0.14245,-0.0868 0.0149,-0.0131 0.0407,-0.0273 0.0575,-0.0315 0.0166,-0.005 0.0303,-0.012 0.0303,-0.0174 0,-0.005 0.024,-0.0212 0.0532,-0.0353 0.0293,-0.0141 0.0532,-0.0293 0.0533,-0.0341 10e-5,-0.005 0.0151,-0.0123 0.0335,-0.017 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.0144,-0.0131 0.0319,-0.0131 0.0176,0 0.0354,-0.009 0.0396,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.005 0.0271,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0267,-0.009 0.0672,-0.0343 0.0938,-0.0587 0.01,-0.009 0.0268,-0.0167 0.0373,-0.0167 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.0153,-0.0134 0.0341,-0.0134 0.0208,0 0.0307,-0.006 0.0254,-0.0142 -0.005,-0.008 0.003,-0.0176 0.0192,-0.0215 0.0154,-0.005 0.028,-0.012 0.028,-0.0177 0,-0.006 0.015,-0.0141 0.0334,-0.0186 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.021,-0.0131 0.0115,0 0.0293,-0.008 0.0396,-0.0167 0.0483,-0.0433 0.062,-0.0519 0.0863,-0.0537 0.0143,-7.6e-4 0.029,-0.01 0.0326,-0.0192 0.003,-0.009 0.0158,-0.0172 0.0271,-0.0172 0.0112,0 0.0205,-0.006 0.0205,-0.0134 0,-0.008 0.0122,-0.0134 0.0271,-0.0134 0.0149,0 0.0306,-0.009 0.0348,-0.02 0.005,-0.011 0.019,-0.02 0.0329,-0.02 0.0139,0 0.0253,-0.006 0.0253,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.0184,-0.0119 0.0101,0 0.0259,-0.009 0.035,-0.02 0.009,-0.011 0.0254,-0.02 0.036,-0.02 0.0106,0 0.0277,-0.008 0.0379,-0.0167 0.0397,-0.0356 0.0588,-0.05 0.0659,-0.05 0.005,0 0.0312,-0.0149 0.0601,-0.0332 0.0289,-0.0183 0.0596,-0.0348 0.0682,-0.0368 0.009,-0.002 0.0309,-0.0172 0.0496,-0.0339 0.0186,-0.0167 0.0479,-0.0338 0.0649,-0.0382 0.017,-0.005 0.0311,-0.0132 0.0311,-0.0197 0,-0.007 0.0122,-0.0119 0.0274,-0.0119 0.0151,0 0.0423,-0.015 0.0606,-0.0334 0.0183,-0.0183 0.0452,-0.0334 0.0594,-0.0334 0.0144,0 0.0261,-0.006 0.0261,-0.0125 0,-0.007 0.0133,-0.0166 0.0294,-0.0218 0.0162,-0.005 0.0376,-0.0176 0.0477,-0.0276 0.01,-0.01 0.0268,-0.0182 0.0374,-0.0182 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.015,-0.0134 0.0334,-0.0134 0.0207,0 0.0334,-0.008 0.0334,-0.02 0,-0.0112 0.0119,-0.02 0.0267,-0.02 0.0147,0 0.0267,-0.005 0.0267,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0321,-0.0108 0.0698,-0.0357 0.12907,-0.0854 0.0109,-0.009 0.0264,-0.0167 0.0343,-0.0167 0.0133,0 0.0783,-0.0369 0.0935,-0.0532 0.003,-0.005 0.0306,-0.019 0.0598,-0.0336 0.0293,-0.0147 0.0612,-0.0341 0.071,-0.0432 0.01,-0.009 0.0264,-0.0167 0.0369,-0.0167 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.0109 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.006 0.0271,-0.0134 0,-0.008 0.0116,-0.0134 0.0258,-0.0134 0.0142,0 0.0299,-0.006 0.0347,-0.0141 0.005,-0.008 0.0217,-0.0198 0.0375,-0.0267 0.0352,-0.0153 0.075,-0.0404 0.0992,-0.0626 0.01,-0.009 0.0256,-0.0167 0.0348,-0.0167 0.009,0 0.0286,-0.0112 0.0432,-0.0248 0.0148,-0.0137 0.0504,-0.036 0.0793,-0.0495 0.029,-0.0136 0.0604,-0.0324 0.0702,-0.0419 0.01,-0.009 0.0262,-0.0172 0.0367,-0.0172 0.0105,0 0.0191,-0.006 0.0191,-0.0134 0,-0.008 0.009,-0.0134 0.0184,-0.0134 0.0101,0 0.0258,-0.009 0.035,-0.02 0.009,-0.0111 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.025,-0.006 0.025,-0.0131 0,-0.007 0.0151,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0122 0.0335,-0.017 10e-5,-0.005 0.0241,-0.02 0.0533,-0.0341 0.0293,-0.0141 0.0532,-0.0306 0.0532,-0.0367 0,-0.006 0.0121,-0.0112 0.0271,-0.0112 0.0149,0 0.0306,-0.009 0.0348,-0.02 0.005,-0.0111 0.0169,-0.02 0.0281,-0.02 0.0112,0 0.0276,-0.01 0.0363,-0.0217 0.0319,-0.0436 0.12628,-0.0125 0.19549,0.0644 0.0118,0.0132 0.0281,0.024 0.0361,0.024 0.008,0 0.0192,0.009 0.0251,0.0183 0.006,0.01 0.0302,0.0282 0.054,0.0404 0.0239,0.0122 0.0434,0.027 0.0434,0.0331 0,0.01 0.0449,0.0355 0.0701,0.04 0.005,7.7e-4 0.0215,0.0123 0.0357,0.0251 0.0484,0.0439 0.0665,0.0567 0.0805,0.0567 0.008,0 0.014,0.006 0.014,0.0134 0,0.008 0.006,0.0134 0.0134,0.0134 0.008,0 0.0249,0.0106 0.039,0.0234 0.0484,0.0439 0.0665,0.0567 0.0805,0.0567 0.008,0 0.0139,0.006 0.0139,0.0134 0,0.008 0.009,0.0134 0.0205,0.0134 0.0112,0 0.0239,0.009 0.0281,0.02 0.005,0.0111 0.0133,0.02 0.0203,0.02 0.007,0 0.025,0.0116 0.0403,0.026 0.0239,0.0222 0.0251,0.0284 0.009,0.0434 -0.0271,0.025 -0.0675,0.0503 -0.0945,0.0594 -0.0128,0.005 -0.0234,0.0119 -0.0234,0.0166 0,0.005 -0.0225,0.0205 -0.05,0.035 -0.0275,0.0144 -0.0548,0.0339 -0.0607,0.0431 -0.006,0.009 -0.0223,0.0168 -0.0367,0.0168 -0.0143,0 -0.026,0.005 -0.026,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0372,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0111 -0.0118,0.02 -0.026,0.02 -0.0143,0 -0.0308,0.008 -0.0368,0.0174 -0.006,0.01 -0.0332,0.0278 -0.0607,0.0406 -0.0276,0.0128 -0.0501,0.0275 -0.0502,0.0326 -1.1e-4,0.01 -0.064,0.0427 -0.0831,0.0427 -0.0113,0 -0.0307,0.0141 -0.0765,0.0554 -0.0134,0.0121 -0.046,0.0298 -0.0724,0.0393 -0.0264,0.009 -0.048,0.0221 -0.048,0.0277 0,0.006 -0.0241,0.0227 -0.0533,0.0376 -0.0293,0.015 -0.0533,0.0332 -0.0533,0.0403 0,0.007 -0.0114,0.0131 -0.0253,0.0131 -0.0139,0 -0.0286,0.009 -0.0329,0.02 -0.005,0.0111 -0.0169,0.02 -0.0281,0.02 -0.0112,0 -0.0204,0.006 -0.0204,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.005 -0.02,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.007,0.0131 -0.0143,0.0131 -0.008,0 -0.0227,0.008 -0.0328,0.0167 -0.0464,0.0416 -0.0597,0.05 -0.0795,0.05 -0.0115,0 -0.0244,0.009 -0.0286,0.02 -0.005,0.0111 -0.0191,0.02 -0.033,0.02 -0.0138,0 -0.0253,0.006 -0.0253,0.0134 0,0.008 -0.008,0.0134 -0.0168,0.0134 -0.009,0 -0.033,0.0151 -0.0529,0.0334 -0.0198,0.0183 -0.0418,0.0334 -0.049,0.0334 -0.007,0 -0.0209,0.008 -0.0306,0.0167 -0.01,0.009 -0.0415,0.0287 -0.0708,0.0433 -0.0293,0.0147 -0.0605,0.0327 -0.0696,0.04 -0.009,0.008 -0.0254,0.0208 -0.0363,0.03 -0.0109,0.009 -0.0255,0.0167 -0.0324,0.0167 -0.007,0 -0.0159,0.009 -0.0202,0.02 -0.005,0.0111 -0.0173,0.02 -0.0291,0.02 -0.0118,0 -0.0298,0.008 -0.04,0.0167 -0.0462,0.0415 -0.0597,0.05 -0.079,0.05 -0.0113,0 -0.0206,0.006 -0.0206,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.013 -0.0334,0.0186 0,0.006 -0.0105,0.0139 -0.0234,0.0183 -0.0244,0.009 -0.072,0.0365 -0.0834,0.0497 -0.003,0.005 -0.0275,0.0192 -0.053,0.0334 -0.0499,0.0277 -0.058,0.0332 -0.10414,0.0711 -0.0167,0.0137 -0.0497,0.0328 -0.0733,0.0426 -0.0236,0.01 -0.0495,0.0239 -0.0574,0.0315 -0.008,0.008 -0.0424,0.0316 -0.0767,0.0533 -0.0343,0.0217 -0.0623,0.0442 -0.0623,0.05 0,0.006 -0.0114,0.0107 -0.0253,0.0107 -0.0139,0 -0.0287,0.009 -0.033,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0134 0,0.008 -0.009,0.0134 -0.0193,0.0134 -0.0106,0 -0.0246,0.009 -0.0309,0.0206 -0.006,0.0114 -0.0247,0.0216 -0.0406,0.0228 -0.016,0.002 -0.0291,0.0103 -0.0291,0.0202 0,0.01 -0.006,0.0142 -0.0135,0.01 -0.008,-0.005 -0.017,7.7e-4 -0.0214,0.0126 -0.005,0.0115 -0.0193,0.0209 -0.0333,0.0209 -0.0139,0 -0.0253,0.009 -0.0253,0.02 0,0.0109 -0.009,0.02 -0.0202,0.02 -0.0247,0 -0.0764,0.0279 -0.0897,0.0485 -0.005,0.009 -0.0249,0.019 -0.0434,0.0236 -0.0184,0.005 -0.0335,0.0143 -0.0335,0.0215 0,0.007 -0.009,0.0131 -0.0208,0.0131 -0.0115,0 -0.0171,0.006 -0.0126,0.0134 0.005,0.008 -0.005,0.0134 -0.0196,0.0134 -0.0154,0 -0.0309,0.008 -0.0345,0.0172 -0.003,0.009 -0.0334,0.029 -0.0662,0.0434 -0.0328,0.0144 -0.0597,0.0306 -0.0597,0.0362 -6e-5,0.005 -0.009,0.01 -0.0211,0.01 -0.0115,0 -0.0293,0.008 -0.0396,0.0167 -0.039,0.0349 -0.0587,0.05 -0.0652,0.05 -0.007,0 -0.12791,0.0802 -0.14105,0.0934 -0.003,0.003 -0.0292,0.0182 -0.0567,0.0321 -0.0275,0.014 -0.05,0.0299 -0.05,0.0353 0,0.005 -0.015,0.0136 -0.0334,0.0182 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.02,0.0131 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.012,0.0119 -0.0267,0.0119 -0.0147,0 -0.0267,-0.005 -0.0267,-0.0114 z m 8.3197,-0.0215 c -0.0147,-0.0179 -0.0365,-0.0328 -0.0487,-0.0332 -0.0121,-2.7e-4 -0.0254,-0.01 -0.0296,-0.0206 -0.005,-0.0111 -0.0143,-0.02 -0.0225,-0.02 -0.008,0 -0.0231,-0.008 -0.0333,-0.0167 -0.0442,-0.0396 -0.0593,-0.05 -0.0724,-0.05 -0.008,0 -0.0138,-0.006 -0.0138,-0.0134 0,-0.008 -0.009,-0.0134 -0.0186,-0.0134 -0.0102,0 -0.022,-0.009 -0.0263,-0.02 -0.005,-0.011 -0.0154,-0.0201 -0.0248,-0.0202 -0.009,-1.1e-4 -0.0381,-0.0181 -0.0638,-0.0399 -0.0257,-0.0219 -0.0539,-0.0457 -0.0628,-0.0532 -0.009,-0.008 -0.0223,-0.0135 -0.03,-0.0135 -0.008,0 -0.0139,-0.005 -0.0139,-0.0119 0,-0.007 -0.0135,-0.0154 -0.03,-0.0197 -0.0165,-0.005 -0.0474,-0.0229 -0.0688,-0.0414 -0.0213,-0.0186 -0.0441,-0.0338 -0.0506,-0.0338 -0.007,0 -0.0258,-0.0151 -0.0428,-0.0334 -0.017,-0.0183 -0.0352,-0.0334 -0.0403,-0.0334 -0.005,0 -0.0178,-0.008 -0.028,-0.0167 -0.0453,-0.0406 -0.0595,-0.0501 -0.0757,-0.0503 -0.009,-9e-5 -0.0361,-0.0166 -0.0592,-0.0367 -0.0231,-0.02 -0.0554,-0.0455 -0.0717,-0.0565 -0.0163,-0.0111 -0.0375,-0.0276 -0.0471,-0.0367 -0.01,-0.009 -0.0239,-0.0167 -0.0318,-0.0167 -0.008,0 -0.0143,-0.012 -0.0143,-0.0267 0,-0.0147 0.007,-0.0267 0.0144,-0.0267 0.008,0 0.0318,-0.0165 0.0529,-0.0368 0.0596,-0.0567 0.0771,-0.0701 0.0922,-0.0701 0.008,0 0.014,-0.009 0.014,-0.02 0,-0.0111 0.009,-0.02 0.02,-0.02 0.0111,0 0.02,-0.006 0.02,-0.0141 0,-0.008 0.006,-0.0106 0.0125,-0.006 0.007,0.005 0.0187,-0.002 0.0263,-0.0158 0.008,-0.013 0.014,-0.0187 0.0141,-0.0127 2.1e-4,0.006 0.0206,-0.0105 0.0454,-0.0367 0.0247,-0.0262 0.0517,-0.0477 0.0601,-0.0477 0.009,0 0.0151,-0.009 0.0151,-0.02 0,-0.0111 0.005,-0.0185 0.0103,-0.0167 0.006,0.002 0.028,-0.0132 0.0499,-0.0334 0.0219,-0.0203 0.0467,-0.0367 0.0553,-0.0367 0.009,0 0.0177,-0.005 0.0203,-0.0101 0.002,-0.005 0.0314,-0.0306 0.0645,-0.0558 0.0331,-0.0251 0.0728,-0.0596 0.0883,-0.0766 0.0155,-0.017 0.0345,-0.0309 0.0423,-0.0309 0.008,0 0.016,-0.005 0.0184,-0.0105 0.002,-0.006 0.0308,-0.0313 0.0631,-0.0567 0.0323,-0.0254 0.0831,-0.0688 0.11283,-0.0962 0.0298,-0.0275 0.0598,-0.05 0.0666,-0.05 0.007,0 0.0263,-0.015 0.0433,-0.0334 0.017,-0.0183 0.0373,-0.0334 0.0448,-0.0334 0.008,0 0.014,-0.009 0.014,-0.0208 0,-0.0115 0.006,-0.017 0.0135,-0.0125 0.008,0.005 0.017,-7.6e-4 0.0214,-0.0125 0.005,-0.0114 0.0121,-0.0183 0.017,-0.0152 0.005,0.003 0.0477,-0.0299 0.0949,-0.0734 0.0473,-0.0435 0.0981,-0.0791 0.11302,-0.0793 0.0223,-1.6e-4 0.0236,-0.002 0.007,-0.0131 -0.0178,-0.0114 -0.0178,-0.013 0,-0.0134 0.0109,-2e-4 0.053,-0.0303 0.0934,-0.0669 0.0403,-0.0366 0.0839,-0.0699 0.0968,-0.0742 0.0128,-0.005 0.0234,-0.0133 0.0234,-0.0201 0,-0.007 0.006,-0.0125 0.0134,-0.0125 0.008,0 0.0243,-0.0105 0.0377,-0.0234 0.0413,-0.0396 0.0955,-0.0834 0.10309,-0.0834 0.005,0 0.0212,-0.015 0.0381,-0.0334 0.017,-0.0183 0.0377,-0.0334 0.0461,-0.0334 0.009,0 0.0151,-0.009 0.0151,-0.02 0,-0.011 0.009,-0.02 0.0206,-0.02 0.0114,0 0.0346,-0.015 0.0516,-0.0334 0.017,-0.0183 0.0345,-0.0334 0.039,-0.0334 0.005,0 0.0277,-0.0183 0.0518,-0.0405 0.0241,-0.0222 0.0599,-0.0553 0.0797,-0.0733 0.0197,-0.018 0.0392,-0.0329 0.0432,-0.0329 0.005,0 0.021,-0.0116 0.0375,-0.0259 0.0928,-0.0798 0.11101,-0.0942 0.11918,-0.0942 0.005,-1e-5 0.026,-0.0166 0.0467,-0.0369 0.0207,-0.0203 0.0647,-0.0578 0.0978,-0.0833 0.0331,-0.0255 0.0621,-0.051 0.0645,-0.0565 0.002,-0.005 0.0102,-0.0101 0.0173,-0.0101 0.007,0 0.0268,-0.0135 0.0439,-0.03 0.0434,-0.0419 0.0867,-0.0767 0.0956,-0.0767 0.005,0 0.0158,-0.008 0.0261,-0.0167 0.0465,-0.0417 0.0598,-0.05 0.0798,-0.05 0.0118,0 0.018,-0.005 0.0141,-0.0119 -0.005,-0.007 0.009,-0.0225 0.0283,-0.0353 0.0196,-0.0128 0.0587,-0.0461 0.0866,-0.0738 0.028,-0.0277 0.051,-0.0448 0.051,-0.0381 0,0.007 0.0115,0.0123 0.0257,0.0123 0.0141,0 0.0338,0.008 0.0436,0.0167 0.01,0.009 0.0418,0.0286 0.071,0.0432 0.0293,0.0147 0.0562,0.0297 0.0598,0.0336 0.0182,0.0193 0.0815,0.0532 0.0996,0.0532 0.0113,0 0.0205,0.005 0.0205,0.0119 0,0.007 0.015,0.0157 0.0334,0.0203 0.0183,0.005 0.0334,0.0143 0.0334,0.0215 0,0.007 0.01,0.0131 0.0216,0.0131 0.024,0 0.18258,0.0809 0.18962,0.0967 0.002,0.005 0.0159,0.01 0.03,0.01 0.014,0 0.0256,0.006 0.0256,0.0134 0,0.008 0.012,0.0134 0.0267,0.0134 0.0147,0 0.0267,0.007 0.0267,0.0156 0,0.009 0.003,0.0119 0.008,0.008 0.005,-0.005 0.021,0.005 0.0368,0.0188 0.0157,0.0147 0.0451,0.0301 0.0653,0.0342 0.0202,0.005 0.0368,0.0118 0.0368,0.0173 0,0.005 0.03,0.0241 0.0667,0.0415 0.0367,0.0174 0.0667,0.0363 0.0667,0.0419 0,0.006 0.0165,0.0104 0.0367,0.0106 l 0.0367,5.2e-4 -0.0367,0.0322 c -0.0202,0.0178 -0.0367,0.0386 -0.0367,0.0462 0,0.008 -0.006,0.014 -0.0138,0.014 -0.008,0 -0.0301,0.0166 -0.0501,0.0369 -0.0199,0.0203 -0.0603,0.0549 -0.0896,0.0767 -0.0709,0.0529 -0.0867,0.0685 -0.0867,0.0855 0,0.008 -0.007,0.0143 -0.0147,0.0143 -0.008,0 -0.0265,0.015 -0.041,0.0334 -0.0144,0.0183 -0.0348,0.0334 -0.0453,0.0334 -0.0105,0 -0.0191,0.007 -0.0191,0.0143 0,0.015 -0.0326,0.0524 -0.0456,0.0524 -0.008,0 -0.0614,0.0432 -0.10342,0.0834 -0.0134,0.0128 -0.0304,0.0234 -0.0377,0.0234 -0.008,0 -0.0134,0.009 -0.0134,0.02 0,0.011 -0.006,0.02 -0.0142,0.02 -0.008,0 -0.0333,0.0196 -0.0567,0.0435 -0.0234,0.024 -0.0634,0.0591 -0.0892,0.0783 -0.0257,0.019 -0.0715,0.0594 -0.10183,0.0898 -0.0303,0.0304 -0.0618,0.0552 -0.0701,0.0552 -0.009,0 -0.0149,0.009 -0.0149,0.02 0,0.0116 -0.0121,0.02 -0.029,0.02 -0.0159,0 -0.0254,0.003 -0.0212,0.008 0.0105,0.0104 -0.14918,0.16569 -0.17027,0.16569 -0.003,0 -0.0458,0.039 -0.0938,0.0867 -0.048,0.0477 -0.0902,0.0867 -0.094,0.0867 -0.003,0 -0.0208,0.0135 -0.0379,0.03 -0.0509,0.0491 -0.0873,0.0767 -0.1013,0.0767 -0.007,0 -0.0132,0.007 -0.0132,0.0143 0,0.0167 -0.0332,0.0523 -0.049,0.0527 -0.006,1.5e-4 -0.0455,0.0332 -0.0877,0.0734 -0.0422,0.0403 -0.0795,0.0731 -0.0828,0.0731 -0.003,0 -0.0264,0.021 -0.0513,0.0467 -0.0249,0.0257 -0.0523,0.0468 -0.0607,0.0468 -0.009,0 -0.0154,0.006 -0.0154,0.0128 0,0.014 -0.074,0.0939 -0.087,0.0939 -0.005,0 -0.027,0.0179 -0.0505,0.04 -0.0235,0.022 -0.0476,0.0399 -0.0536,0.04 -0.0178,1e-4 -0.0486,0.0361 -0.039,0.0458 0.005,0.005 -0.003,0.0121 -0.019,0.0161 -0.0154,0.005 -0.0355,0.0218 -0.045,0.0395 -0.009,0.0177 -0.0253,0.0322 -0.035,0.0322 -0.01,0 -0.0177,0.006 -0.0177,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.019 0,0.0105 -0.005,0.0208 -0.0118,0.0234 -0.014,0.005 -0.11677,0.0878 -0.15108,0.12121 -0.0132,0.0128 -0.0271,0.0234 -0.0311,0.0234 -0.005,0 -0.0242,0.0165 -0.0452,0.0367 -0.0824,0.0796 -0.15021,0.13678 -0.1621,0.13678 -0.007,0 -0.0125,0.009 -0.0125,0.02 0,0.0109 -0.006,0.02 -0.0134,0.02 -0.008,0 -0.0489,0.0361 -0.0923,0.08 -0.0433,0.044 -0.0898,0.08 -0.10332,0.08 -0.0134,-1e-4 -0.0364,-0.0148 -0.0512,-0.0327 z m -9.30545,-1.00488 c -0.0234,-0.009 -0.0603,-0.0451 -0.0893,-0.0893 -0.0125,-0.0191 -0.0249,-0.0325 -0.0277,-0.0296 -0.007,0.007 -0.17819,-0.16457 -0.17819,-0.17825 0,-0.006 -0.0121,-0.0148 -0.0269,-0.0194 -0.0148,-0.005 -0.0235,-0.0141 -0.0192,-0.0209 0.005,-0.007 -0.002,-0.0124 -0.0124,-0.0124 -0.0111,0 -0.0306,-0.018 -0.0437,-0.04 -0.013,-0.022 -0.0324,-0.0401 -0.043,-0.0401 -0.0106,0 -0.016,-0.003 -0.012,-0.008 0.005,-0.005 -0.0301,-0.0459 -0.0761,-0.0929 -0.0459,-0.0471 -0.0808,-0.0882 -0.0776,-0.0915 0.003,-0.003 -0.007,-0.01 -0.0219,-0.0147 -0.026,-0.009 -0.0263,-0.0102 -0.005,-0.0312 0.0123,-0.0123 0.0267,-0.0225 0.0319,-0.0225 0.005,0 0.0347,-0.0165 0.0655,-0.0367 0.0308,-0.0203 0.0646,-0.0368 0.0748,-0.0368 0.0104,0 0.0189,-0.005 0.0189,-0.0121 0,-0.007 0.0331,-0.0257 0.0733,-0.0425 0.0403,-0.0167 0.0733,-0.0353 0.0733,-0.0413 0,-0.006 0.012,-0.0109 0.0267,-0.0109 0.0147,0 0.0267,-0.005 0.0268,-0.01 1.6e-4,-0.0154 0.0788,-0.0567 0.10774,-0.0567 0.0141,0 0.0256,-0.005 0.0256,-0.0112 0,-0.006 0.0331,-0.0251 0.0733,-0.0422 0.0403,-0.017 0.0733,-0.0361 0.0733,-0.0423 0,-0.006 0.012,-0.0112 0.0267,-0.0112 0.0147,0 0.0267,-0.005 0.0267,-0.0109 0,-0.006 0.0285,-0.0244 0.0633,-0.041 0.12095,-0.0574 0.19013,-0.0971 0.19013,-0.1094 0,-0.007 0.015,-0.0123 0.0334,-0.0123 0.0183,0 0.0334,-0.005 0.0334,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0264,-0.009 0.0733,-0.0374 0.0836,-0.051 0.003,-0.005 0.0233,-0.0131 0.0434,-0.0181 0.0201,-0.005 0.0366,-0.0145 0.0366,-0.0211 0,-0.007 0.009,-0.0119 0.0184,-0.0119 0.0101,0 0.0258,-0.009 0.035,-0.02 0.009,-0.0111 0.0271,-0.02 0.04,-0.02 0.0128,0 0.0309,-0.009 0.04,-0.02 0.009,-0.0109 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.025,-0.006 0.025,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0111,0 0.02,-0.006 0.02,-0.0125 0,-0.007 0.0135,-0.0164 0.03,-0.0211 0.0568,-0.0163 0.11675,-0.0478 0.11675,-0.0613 0,-0.008 0.005,-0.0101 0.0122,-0.006 0.007,0.005 0.0203,4e-5 0.03,-0.009 0.01,-0.009 0.0477,-0.0313 0.0845,-0.049 0.0368,-0.0177 0.0697,-0.036 0.0733,-0.0405 0.003,-0.005 0.0487,-0.0296 0.10008,-0.0556 0.15823,-0.0803 0.20666,-0.10734 0.21435,-0.11976 0.005,-0.007 0.0221,-0.012 0.04,-0.012 0.0179,0 0.0326,-0.006 0.0326,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.012,-0.0119 0.0267,-0.0119 0.0147,0 0.0267,-0.005 0.0267,-0.0109 0,-0.006 0.0296,-0.0237 0.0656,-0.0395 0.0361,-0.0157 0.0694,-0.0349 0.0741,-0.0426 0.005,-0.008 0.0205,-0.0138 0.0349,-0.0138 0.0144,0 0.0298,-0.009 0.034,-0.02 0.005,-0.0111 0.0155,-0.02 0.0251,-0.02 0.01,0 0.0196,-0.005 0.0219,-0.01 0.006,-0.0136 0.14712,-0.0839 0.1679,-0.0836 0.009,9e-5 0.0167,-0.006 0.0167,-0.0128 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0133 0.0334,-0.0193 0,-0.006 0.0133,-0.0152 0.0293,-0.0203 0.0162,-0.005 0.0379,-0.0179 0.0484,-0.0283 0.0105,-0.0105 0.0272,-0.0158 0.0374,-0.012 0.0104,0.005 0.0183,-0.002 0.0183,-0.0126 0,-0.012 0.0129,-0.0196 0.0334,-0.0196 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.009,-0.0134 0.0191,-0.0134 0.0105,0 0.027,-0.008 0.0367,-0.0168 0.01,-0.009 0.0431,-0.0288 0.0743,-0.0434 0.0311,-0.0147 0.0566,-0.0311 0.0567,-0.0365 1e-4,-0.005 0.0119,-0.01 0.0262,-0.01 0.0143,0 0.0321,-0.0105 0.0396,-0.0234 0.008,-0.0128 0.0138,-0.0189 0.0141,-0.0135 5.2e-4,0.0149 0.1122,-0.032 0.1245,-0.0523 0.006,-0.01 0.0186,-0.0177 0.0283,-0.0177 0.01,0 0.0252,-0.009 0.0343,-0.02 0.009,-0.0111 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.0251,-0.005 0.0251,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.02,-0.0131 0.0111,0 0.02,-0.006 0.02,-0.0128 0,-0.007 0.0211,-0.0167 0.0468,-0.0215 0.0257,-0.005 0.0502,-0.0141 0.0542,-0.0206 0.005,-0.007 0.0234,-0.0119 0.0432,-0.0121 0.027,-1.6e-4 0.0311,-0.003 0.0166,-0.0125 -0.0151,-0.01 -0.0115,-0.0141 0.0167,-0.0212 0.0198,-0.005 0.0361,-0.0144 0.0361,-0.021 0,-0.007 0.0112,-0.0119 0.025,-0.0119 0.0138,0 0.0326,-0.009 0.0417,-0.02 0.009,-0.0111 0.0251,-0.02 0.0354,-0.02 0.0104,0 0.0222,-0.009 0.0264,-0.02 0.005,-0.0111 0.0235,-0.0201 0.043,-0.0203 0.0276,-1.5e-4 0.031,-0.002 0.0152,-0.0131 -0.0163,-0.0105 -0.0145,-0.013 0.009,-0.0131 0.0161,-10e-5 0.0326,-0.006 0.0368,-0.0125 0.005,-0.007 0.0256,-0.0163 0.0475,-0.0211 0.0219,-0.005 0.0399,-0.0143 0.0399,-0.0211 0,-0.007 0.0114,-0.0122 0.0253,-0.0122 0.0139,0 0.0287,-0.009 0.033,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.005 0.0271,-0.0105 0,-0.006 0.0196,-0.0198 0.0434,-0.0311 0.0239,-0.0113 0.0614,-0.03 0.0834,-0.0417 0.022,-0.0116 0.0626,-0.032 0.0901,-0.0453 0.0276,-0.0134 0.05,-0.0289 0.05,-0.0345 0,-0.006 0.0112,-0.0103 0.025,-0.0103 0.0137,0 0.0325,-0.009 0.0417,-0.02 0.009,-0.0111 0.0249,-0.02 0.035,-0.02 0.0101,0 0.0184,-0.006 0.0184,-0.0134 0,-0.008 0.0152,-0.0134 0.0338,-0.0134 0.0186,0 0.0372,-0.009 0.0414,-0.02 0.005,-0.0111 0.019,-0.02 0.033,-0.02 0.0138,0 0.0253,-0.005 0.0253,-0.0101 0,-0.005 0.0298,-0.0237 0.0662,-0.0404 0.0364,-0.0167 0.0697,-0.0362 0.0741,-0.0433 0.005,-0.007 0.0196,-0.0129 0.0339,-0.0129 0.0143,0 0.0259,-0.005 0.0261,-0.01 9e-5,-0.005 0.0256,-0.0219 0.0567,-0.0364 0.0311,-0.0145 0.0596,-0.0302 0.0632,-0.0348 0.003,-0.005 0.0306,-0.0187 0.06,-0.0317 0.0925,-0.0406 0.14655,-0.0669 0.15345,-0.0749 0.003,-0.005 0.0157,-0.0118 0.0267,-0.0168 0.0111,-0.005 0.038,-0.0212 0.06,-0.0359 0.022,-0.0147 0.0505,-0.0303 0.0633,-0.0348 0.0128,-0.005 0.0234,-0.0134 0.0234,-0.0199 0,-0.007 0.015,-0.0118 0.0334,-0.0118 0.0183,0 0.0334,-0.006 0.0334,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0122 0.0335,-0.017 5e-5,-0.005 0.027,-0.0204 0.06,-0.0349 0.033,-0.0144 0.0599,-0.0306 0.0599,-0.0358 0,-0.005 0.0133,-0.0136 0.0293,-0.0189 0.0162,-0.005 0.0377,-0.0177 0.0477,-0.0277 0.0101,-0.0101 0.0232,-0.0154 0.0291,-0.0117 0.006,0.003 0.0182,-0.002 0.0272,-0.0132 0.009,-0.011 0.0277,-0.0199 0.0416,-0.0199 0.0138,0 0.0251,-0.006 0.0251,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.009,-0.0119 0.02,-0.0119 0.0111,0 0.02,-0.006 0.02,-0.0134 0,-0.008 0.008,-0.0134 0.0176,-0.0134 0.0277,0 0.12927,-0.0526 0.12927,-0.067 0,-0.007 0.0116,-0.013 0.026,-0.013 0.0142,0 0.0297,-0.006 0.0341,-0.0134 0.005,-0.008 0.0201,-0.0134 0.0346,-0.0134 0.0144,0 0.0298,-0.009 0.034,-0.02 0.005,-0.0109 0.019,-0.02 0.0329,-0.02 0.0139,0 0.0253,-0.006 0.0253,-0.0131 0,-0.007 0.0164,-0.0172 0.0366,-0.0222 0.0201,-0.005 0.0439,-0.0157 0.0529,-0.0236 0.047,-0.0416 0.059,-0.0477 0.0909,-0.0477 0.0191,0 0.0312,-0.006 0.027,-0.0126 -0.005,-0.007 0.0149,-0.0227 0.0426,-0.035 0.12252,-0.0547 0.14915,-0.0693 0.14115,-0.0773 -0.005,-0.005 0.008,-0.009 0.0275,-0.009 0.0198,0 0.0395,-0.009 0.0436,-0.02 0.005,-0.0111 0.0221,-0.02 0.0396,-0.02 0.0176,0 0.0319,-0.005 0.0319,-0.0123 0,-0.0138 0.0751,-0.0544 0.10071,-0.0544 0.009,0 0.0199,-0.009 0.0242,-0.02 0.005,-0.0111 0.0199,-0.02 0.0348,-0.02 0.0149,0 0.0271,-0.006 0.0271,-0.0134 0,-0.0225 0.0842,-0.013 0.12677,0.0141 0.022,0.0141 0.049,0.0293 0.0601,0.0339 0.0766,0.0315 0.10674,0.0483 0.10674,0.0594 0,0.007 0.015,0.0128 0.0334,0.0128 0.0183,0 0.0334,0.005 0.0334,0.0108 0,0.006 0.0329,0.0254 0.0733,0.0434 0.0403,0.0179 0.0733,0.0371 0.0734,0.0426 1e-4,0.005 0.009,0.01 0.0202,0.01 0.0111,0 0.02,0.007 0.02,0.0156 0,0.009 0.003,0.0118 0.009,0.008 0.006,-0.006 0.16717,0.0673 0.20405,0.0932 0.0102,0.007 -0.043,0.0574 -0.061,0.0574 -0.01,0 -0.018,0.005 -0.0181,0.01 -5e-5,0.005 -0.027,0.0219 -0.0601,0.0364 -0.033,0.0144 -0.06,0.031 -0.06,0.0367 0,0.006 -0.01,0.0104 -0.0214,0.0104 -0.0118,0 -0.0433,0.0167 -0.0701,0.037 -0.0267,0.0204 -0.0711,0.0474 -0.0986,0.0598 -0.0276,0.0125 -0.05,0.0289 -0.05,0.0363 0,0.008 -0.015,0.0136 -0.0334,0.0136 -0.0183,0 -0.0334,0.006 -0.0334,0.0134 0,0.008 -0.009,0.0134 -0.0184,0.0134 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.0111 -0.0281,0.02 -0.0421,0.02 -0.0141,0 -0.0289,0.009 -0.0332,0.0199 -0.005,0.0109 -0.0127,0.0169 -0.0187,0.0131 -0.006,-0.003 -0.0173,0.003 -0.0249,0.0168 -0.008,0.013 -0.014,0.0189 -0.0141,0.013 -2e-4,-0.006 -0.026,0.006 -0.0571,0.0267 -0.0312,0.0206 -0.0807,0.0493 -0.11007,0.0639 -0.0293,0.0147 -0.0613,0.0342 -0.071,0.0434 -0.01,0.009 -0.0262,0.0167 -0.0367,0.0167 -0.0105,0 -0.0191,0.005 -0.0191,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.012,0.0131 -0.0267,0.0131 -0.0147,0 -0.0267,0.005 -0.0267,0.0113 0,0.006 -0.0255,0.0225 -0.0567,0.0361 -0.0312,0.0137 -0.10867,0.0566 -0.17216,0.0954 -0.0635,0.0388 -0.12204,0.0705 -0.13011,0.0705 -0.008,0 -0.0147,0.005 -0.0147,0.0105 0,0.006 -0.039,0.0311 -0.0867,0.0562 -0.0477,0.0253 -0.0867,0.0506 -0.0867,0.0562 0,0.006 -0.009,0.0105 -0.0184,0.0105 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.011 -0.0279,0.02 -0.0417,0.02 -0.0138,0 -0.0251,0.006 -0.0251,0.0134 0,0.008 -0.009,0.0134 -0.0184,0.0134 -0.0101,0 -0.0258,0.009 -0.035,0.02 -0.009,0.0111 -0.0279,0.02 -0.0417,0.02 -0.0138,0 -0.025,0.005 -0.025,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.02,0.0131 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.0122,0.0119 -0.0271,0.0119 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.0221,0.02 -0.0396,0.02 -0.0176,0 -0.0319,0.005 -0.0319,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0254,0.009 -0.0727,0.037 -0.0834,0.0502 -0.003,0.005 -0.0202,0.0124 -0.0367,0.0176 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.006,0.0125 -0.0131,0.0125 -0.007,0 -0.0523,0.0241 -0.10031,0.0533 -0.0479,0.0293 -0.0991,0.0533 -0.11365,0.0533 -0.0145,0 -0.0264,0.006 -0.0264,0.0125 0,0.007 -0.0133,0.0166 -0.0293,0.0218 -0.0162,0.005 -0.0376,0.0176 -0.0477,0.0276 -0.01,0.01 -0.0268,0.0183 -0.0374,0.0183 -0.0105,0 -0.019,0.006 -0.019,0.0134 0,0.008 -0.009,0.0134 -0.0186,0.0134 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.006 -0.0271,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.0137 -0.0334,0.0203 0,0.007 -0.009,0.0119 -0.02,0.0119 -0.0111,0 -0.02,0.006 -0.02,0.0134 0,0.008 -0.0121,0.0134 -0.0271,0.0134 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.0187,0.02 -0.0322,0.02 -0.0135,0 -0.0298,0.009 -0.0361,0.0206 -0.006,0.0114 -0.0247,0.0216 -0.0406,0.0229 -0.016,0.002 -0.0291,0.007 -0.0291,0.0127 0,0.006 -0.009,0.0105 -0.0186,0.0105 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.011 -0.0199,0.02 -0.0348,0.02 -0.0149,0 -0.0271,0.005 -0.0271,0.0119 0,0.007 -0.015,0.0157 -0.0334,0.0203 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.012,0.0131 -0.0267,0.0131 -0.0147,0 -0.0267,0.006 -0.0267,0.013 0,0.007 -0.0255,0.0254 -0.0567,0.0406 -0.10769,0.0524 -0.23019,0.12272 -0.23463,0.13459 -0.002,0.007 -0.0125,0.0119 -0.0225,0.0119 -0.01,0 -0.0259,0.008 -0.0358,0.0167 -0.01,0.009 -0.0418,0.0286 -0.071,0.0432 -0.0293,0.0147 -0.0562,0.0299 -0.0598,0.0341 -0.0157,0.0177 -0.16149,0.0928 -0.17994,0.0928 -0.0111,0 -0.0202,0.005 -0.0202,0.0118 0,0.007 -0.0105,0.0154 -0.0234,0.0196 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.0206,0.0134 -0.0113,0 -0.0368,0.0111 -0.0567,0.0246 -0.0199,0.0135 -0.0632,0.042 -0.0962,0.0633 -0.0331,0.0213 -0.063,0.0419 -0.0667,0.0459 -0.003,0.005 -0.0534,0.0306 -0.1105,0.0593 -0.0572,0.0286 -0.11039,0.06 -0.11839,0.0697 -0.008,0.01 -0.0262,0.0176 -0.0404,0.0176 -0.0142,0 -0.0225,0.005 -0.0185,0.0119 0.005,0.007 -0.009,0.0158 -0.0275,0.0206 -0.0192,0.005 -0.0437,0.0176 -0.0544,0.0283 -0.0107,0.0107 -0.025,0.0161 -0.0318,0.0119 -0.007,-0.005 -0.0122,0.002 -0.0122,0.0133 0,0.0115 -0.0105,0.0212 -0.0234,0.0218 -0.0128,5.1e-4 -0.0413,0.0122 -0.0633,0.026 -0.022,0.0138 -0.067,0.0388 -0.10008,0.0558 -0.0331,0.0169 -0.063,0.0345 -0.0667,0.039 -0.003,0.005 -0.0202,0.0124 -0.0367,0.0177 -0.0165,0.005 -0.03,0.015 -0.03,0.0219 0,0.007 -0.009,0.0125 -0.02,0.0125 -0.0111,0 -0.02,0.005 -0.02,0.0113 0,0.006 -0.027,0.0225 -0.0601,0.0361 -0.0331,0.0136 -0.06,0.0293 -0.06,0.0348 0,0.005 -0.0225,0.0198 -0.05,0.0318 -0.0276,0.012 -0.0591,0.0273 -0.0701,0.0341 -0.0111,0.007 -0.029,0.0163 -0.0401,0.0213 -0.011,0.005 -0.023,0.0124 -0.0267,0.0166 -0.003,0.005 -0.0338,0.0209 -0.0668,0.037 -0.0331,0.0161 -0.0734,0.0416 -0.0895,0.0567 -0.0162,0.0151 -0.0383,0.0274 -0.0491,0.0274 -0.0109,0 -0.0237,0.006 -0.0285,0.0141 -0.005,0.008 -0.0217,0.0198 -0.0375,0.0267 -0.0158,0.007 -0.0405,0.02 -0.0549,0.0292 -0.0809,0.0516 -0.12902,0.0768 -0.14702,0.0768 -0.0111,0 -0.0203,0.005 -0.0203,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0145 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0262,0.0166 -0.0367,0.0166 -0.0105,0 -0.019,0.005 -0.019,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0141 -0.0234,0.0206 0,0.007 -0.0144,0.0118 -0.0319,0.0118 -0.0176,0 -0.0354,0.009 -0.0396,0.02 -0.005,0.011 -0.0163,0.02 -0.027,0.02 -0.0106,0 -0.0213,0.005 -0.0238,0.01 -0.002,0.005 -0.0328,0.0239 -0.0675,0.0409 -0.0347,0.017 -0.0665,0.0364 -0.0708,0.0434 -0.005,0.007 -0.0194,0.0126 -0.0336,0.0126 -0.0142,0 -0.026,0.006 -0.026,0.0127 0,0.007 -0.0225,0.0233 -0.05,0.0362 -0.0275,0.0129 -0.058,0.0312 -0.0676,0.0406 -0.01,0.009 -0.0265,0.0172 -0.0375,0.0172 -0.0109,0 -0.0162,0.006 -0.0115,0.0134 0.005,0.008 -0.003,0.0134 -0.0177,0.0134 -0.0142,0 -0.0333,0.009 -0.0425,0.02 -0.009,0.0111 -0.0309,0.02 -0.0483,0.02 -0.0175,0 -0.0318,0.005 -0.0318,0.0103 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0276,0.0144 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0262,0.0167 -0.0368,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.0118,0.0134 -0.026,0.0134 -0.0143,0 -0.0308,0.008 -0.0367,0.0177 -0.0134,0.0221 -0.0449,0.0338 -0.069,0.0254 z m 5.35576,-1.75103 c -0.008,-0.007 -0.0269,-0.0243 -0.0433,-0.0394 -0.0165,-0.0151 -0.0361,-0.0274 -0.0434,-0.0274 -0.008,0 -0.0134,-0.005 -0.0134,-0.0119 0,-0.007 -0.015,-0.0157 -0.0334,-0.0203 -0.0183,-0.005 -0.0334,-0.0143 -0.0334,-0.0215 0,-0.007 -0.009,-0.0131 -0.02,-0.0131 -0.0111,0 -0.02,-0.005 -0.02,-0.0119 0,-0.007 -0.015,-0.0157 -0.0334,-0.0203 -0.0183,-0.005 -0.0334,-0.0143 -0.0334,-0.0215 0,-0.007 -0.006,-0.0131 -0.014,-0.0131 -0.0144,0 -0.029,-0.0108 -0.0901,-0.0667 -0.02,-0.0183 -0.0589,-0.0448 -0.0863,-0.0588 -0.0274,-0.014 -0.0499,-0.0298 -0.0499,-0.0351 0,-0.005 -0.0105,-0.0132 -0.0234,-0.0176 -0.0335,-0.0113 -0.0721,-0.0374 -0.10601,-0.072 -0.0163,-0.0165 -0.0367,-0.03 -0.0454,-0.03 -0.016,0 -0.051,-0.0383 -0.0517,-0.0567 -2.4e-4,-0.005 0.0115,-0.01 0.0263,-0.01 0.0147,0 0.0267,-0.005 0.0267,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.012,-0.0131 0.0267,-0.0131 0.0147,0 0.0267,-0.006 0.0267,-0.013 0,-0.007 0.0267,-0.0251 0.0592,-0.04 0.0326,-0.0148 0.063,-0.0329 0.0675,-0.0403 0.005,-0.008 0.0199,-0.0134 0.0341,-0.0134 0.0142,0 0.026,-0.006 0.026,-0.0134 0,-0.008 0.009,-0.0134 0.02,-0.0134 0.0109,0 0.02,-0.006 0.02,-0.0124 0,-0.007 0.0133,-0.0166 0.0293,-0.0218 0.0162,-0.005 0.0376,-0.0176 0.0477,-0.0276 0.01,-0.01 0.0268,-0.0183 0.0374,-0.0183 0.0105,0 0.019,-0.005 0.019,-0.0118 0,-0.007 0.0105,-0.0154 0.0234,-0.0196 0.0227,-0.008 0.07,-0.0352 0.0834,-0.0486 0.003,-0.003 0.0352,-0.0201 0.0701,-0.0366 0.0348,-0.0164 0.0633,-0.036 0.0633,-0.0434 0,-0.008 0.015,-0.0135 0.0334,-0.0135 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.009,-0.0134 0.0205,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0184,-0.02 0.0315,-0.02 0.0131,0 0.0269,-0.008 0.0306,-0.0178 0.003,-0.01 0.0419,-0.0348 0.0849,-0.0556 0.0428,-0.0208 0.081,-0.0413 0.0847,-0.0455 0.0113,-0.0132 0.0588,-0.0414 0.0834,-0.0497 0.0128,-0.005 0.0234,-0.0127 0.0234,-0.0186 0,-0.006 0.0185,-0.0148 0.041,-0.0197 0.0225,-0.005 0.0376,-0.0125 0.0335,-0.0165 -0.009,-0.009 0.0239,-0.0284 0.11572,-0.0707 0.0348,-0.0161 0.0633,-0.0345 0.0633,-0.0409 0,-0.007 0.0105,-0.0118 0.0234,-0.0118 0.0129,0 0.0397,-0.015 0.0595,-0.0334 0.0198,-0.0183 0.0445,-0.0334 0.055,-0.0334 0.0105,0 0.0264,-0.009 0.0356,-0.02 0.009,-0.011 0.0279,-0.02 0.0417,-0.02 0.0137,0 0.0251,-0.006 0.0251,-0.0134 0,-0.008 0.009,-0.0134 0.0186,-0.0134 0.0102,0 0.022,-0.009 0.0263,-0.02 0.005,-0.0109 0.0209,-0.02 0.037,-0.02 0.0161,0 0.0255,-0.003 0.0208,-0.009 -0.005,-0.005 0.0169,-0.021 0.0479,-0.0361 0.031,-0.0151 0.0643,-0.0354 0.0739,-0.0448 0.01,-0.009 0.0262,-0.0172 0.0368,-0.0172 0.0105,0 0.0191,-0.006 0.0191,-0.0131 0,-0.007 0.015,-0.0169 0.0334,-0.0215 0.0183,-0.005 0.0334,-0.0137 0.0334,-0.0203 0,-0.007 0.015,-0.0119 0.0334,-0.0119 0.0183,0 0.0334,-0.006 0.0334,-0.0134 0,-0.008 0.0241,-0.0238 0.0533,-0.0364 0.0293,-0.0128 0.0533,-0.0307 0.0533,-0.04 0,-0.009 0.015,-0.0169 0.0334,-0.0169 0.0183,0 0.0334,-0.006 0.0334,-0.0126 0,-0.0122 0.0222,-0.0264 0.12009,-0.0768 0.0257,-0.0132 0.0497,-0.027 0.0533,-0.0306 0.01,-0.01 0.0567,-0.0387 0.08,-0.0494 0.0111,-0.005 0.0279,-0.017 0.0377,-0.0267 0.01,-0.01 0.0262,-0.0175 0.0367,-0.0175 0.0105,0 0.0191,-0.005 0.0191,-0.0119 0,-0.007 0.015,-0.0157 0.0334,-0.0203 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.012,-0.0131 0.0267,-0.0131 0.0147,0 0.0267,-0.005 0.0267,-0.0118 0,-0.007 0.0106,-0.0154 0.0234,-0.0199 0.0128,-0.005 0.0413,-0.0202 0.0633,-0.0351 0.022,-0.0148 0.0505,-0.0306 0.0633,-0.0351 0.0128,-0.005 0.0234,-0.0128 0.0234,-0.0186 0,-0.0167 0.0956,0.0171 0.10295,0.0364 0.003,0.009 0.0151,0.0173 0.0255,0.0173 0.0167,0 0.059,0.0222 0.14558,0.0765 0.0144,0.009 0.0547,0.0293 0.0896,0.045 0.0348,0.0157 0.0633,0.0339 0.0633,0.0403 0,0.007 0.012,0.0118 0.0267,0.0118 0.0147,0 0.0267,0.006 0.0267,0.0131 0,0.007 0.015,0.0169 0.0334,0.0215 0.0183,0.005 0.0334,0.0137 0.0334,0.0203 0,0.007 0.0122,0.0119 0.0271,0.0119 0.0149,0 0.0306,0.009 0.0348,0.02 0.005,0.0111 0.0221,0.02 0.0397,0.02 0.0176,0 0.0319,0.005 0.0319,0.0119 0,0.007 0.0105,0.0156 0.0234,0.0202 0.0313,0.0114 0.0845,0.0429 0.0876,0.052 0.003,0.0101 -0.0842,0.0685 -0.14652,0.0978 -0.0282,0.0132 -0.0584,0.0303 -0.0674,0.0378 -0.0345,0.0295 -0.0523,0.0406 -0.0711,0.0449 -0.0107,0.002 -0.0243,0.0129 -0.0302,0.0234 -0.006,0.0105 -0.0168,0.019 -0.0243,0.019 -0.008,0 -0.022,0.008 -0.0322,0.0167 -0.0335,0.0301 -0.053,0.0431 -0.10329,0.0688 -0.0273,0.014 -0.0497,0.0298 -0.0497,0.0352 0,0.005 -0.0105,0.0131 -0.0234,0.0176 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.019,0.006 -0.019,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.006 -0.02,0.0131 0,0.007 -0.015,0.0169 -0.0334,0.0215 -0.0183,0.005 -0.0334,0.013 -0.0334,0.0187 0,0.006 -0.0105,0.0139 -0.0234,0.0182 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0372,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0109 -0.009,0.02 -0.0203,0.02 -0.0111,0 -0.0441,0.015 -0.0731,0.0334 -0.0291,0.0183 -0.0584,0.0334 -0.0653,0.0334 -0.007,0 -0.0144,0.005 -0.017,0.01 -0.008,0.0179 -0.13146,0.0967 -0.15162,0.0967 -0.0108,0 -0.0196,0.007 -0.0196,0.0152 0,0.0207 -0.0413,0.0506 -0.0706,0.0511 -0.0132,2.6e-4 -0.0565,0.0244 -0.0962,0.0538 -0.0397,0.0293 -0.0815,0.0533 -0.0928,0.0533 -0.0113,0 -0.0206,0.009 -0.0206,0.02 0,0.0111 -0.009,0.02 -0.02,0.02 -0.0111,0 -0.02,0.006 -0.02,0.0141 0,0.008 -0.006,0.0106 -0.0128,0.006 -0.007,-0.005 -0.0246,0.003 -0.039,0.0182 -0.0143,0.0143 -0.0284,0.0237 -0.0312,0.0208 -0.002,-0.002 -0.0248,0.0127 -0.0487,0.0345 -0.0239,0.0219 -0.0513,0.0397 -0.061,0.0397 -0.01,0 -0.025,0.009 -0.0342,0.02 -0.009,0.0111 -0.0249,0.02 -0.035,0.02 -0.0101,0 -0.0184,0.006 -0.0184,0.0134 0,0.008 -0.009,0.0134 -0.0205,0.0134 -0.0112,0 -0.0239,0.009 -0.0281,0.02 -0.005,0.0111 -0.019,0.02 -0.033,0.02 -0.0138,0 -0.0253,0.006 -0.0253,0.0134 0,0.008 -0.0114,0.0134 -0.0253,0.0134 -0.0139,0 -0.0286,0.009 -0.0329,0.02 -0.005,0.0111 -0.0199,0.02 -0.0348,0.02 -0.0153,0 -0.0271,0.009 -0.0271,0.02 0,0.011 -0.009,0.02 -0.02,0.02 -0.011,0 -0.02,0.005 -0.02,0.0102 0,0.0102 -0.0396,0.0343 -0.0834,0.0509 -0.0128,0.005 -0.0234,0.0141 -0.0234,0.0206 0,0.007 -0.009,0.0118 -0.0191,0.0118 -0.0105,0 -0.0272,0.008 -0.0373,0.0167 -0.0265,0.0243 -0.067,0.0497 -0.0938,0.0587 -0.0128,0.005 -0.0234,0.0124 -0.0234,0.018 0,0.006 -0.0105,0.0138 -0.0234,0.018 -0.0267,0.009 -0.0672,0.0343 -0.0938,0.0587 -0.01,0.009 -0.0268,0.0167 -0.0373,0.0167 -0.0105,0 -0.0191,0.006 -0.0191,0.0134 0,0.008 -0.009,0.0134 -0.02,0.0134 -0.0111,0 -0.02,0.005 -0.02,0.0119 0,0.007 -0.0155,0.0157 -0.0345,0.0206 -0.0325,0.009 -0.0463,0.0169 -0.0918,0.0576 -0.0102,0.009 -0.0243,0.0167 -0.0312,0.0167 -0.007,0 -0.0201,0.009 -0.0293,0.02 -0.0183,0.0219 -0.0657,0.0256 -0.0867,0.007 z m 2.92224,-0.36528 c -0.0147,-0.009 -0.0297,-0.0181 -0.0334,-0.0217 -0.0201,-0.0201 -0.0943,-0.0667 -0.10621,-0.0667 -0.008,0 -0.0138,-0.006 -0.0138,-0.0134 0,-0.008 -0.0114,-0.0134 -0.0253,-0.0134 -0.0139,0 -0.0286,-0.009 -0.0329,-0.02 -0.005,-0.0111 -0.0199,-0.02 -0.0348,-0.02 -0.0149,0 -0.0271,-0.005 -0.0271,-0.0118 0,-0.007 -0.0105,-0.0154 -0.0234,-0.0196 -0.0254,-0.009 -0.0727,-0.037 -0.0834,-0.0502 -0.003,-0.005 -0.0203,-0.0124 -0.0367,-0.0176 -0.0165,-0.005 -0.03,-0.015 -0.03,-0.0219 0,-0.007 -0.009,-0.0124 -0.02,-0.0124 -0.0111,0 -0.02,-0.006 -0.02,-0.0125 0,-0.007 -0.0135,-0.0167 -0.03,-0.0219 -0.0165,-0.005 -0.0331,-0.013 -0.0367,-0.0176 -0.0107,-0.0132 -0.0581,-0.0417 -0.0834,-0.0502 -0.0128,-0.005 -0.0234,-0.0131 -0.0234,-0.0196 0,-0.007 -0.009,-0.0118 -0.019,-0.0118 -0.0105,0 -0.0272,-0.008 -0.0372,-0.0167 -0.0265,-0.0243 -0.0671,-0.0497 -0.0938,-0.0587 -0.0128,-0.005 -0.0234,-0.0107 -0.0234,-0.0141 0,-0.0144 0.0617,-0.0564 0.0834,-0.0568 0.0128,-2.6e-4 0.0234,-0.007 0.0234,-0.0138 0,-0.008 0.009,-0.0134 0.019,-0.0134 0.0105,0 0.0208,-0.005 0.0234,-0.0106 0.002,-0.006 0.027,-0.0243 0.0545,-0.0411 0.0275,-0.0168 0.05,-0.0361 0.05,-0.0428 0,-0.007 0.0128,-0.0122 0.0285,-0.0122 0.0157,0 0.0413,-0.0139 0.0571,-0.0309 0.0157,-0.017 0.0494,-0.0425 0.075,-0.0567 0.0255,-0.0142 0.0493,-0.0293 0.053,-0.0335 0.003,-0.005 0.0157,-0.0119 0.0267,-0.0172 0.0521,-0.0249 0.0791,-0.045 0.071,-0.0531 -0.005,-0.005 7.6e-4,-0.009 0.0125,-0.009 0.0118,0 0.0294,-0.008 0.0395,-0.0167 0.0265,-0.0243 0.067,-0.0497 0.0938,-0.0587 0.0128,-0.005 0.0234,-0.0125 0.0234,-0.0182 0,-0.006 0.015,-0.0141 0.0334,-0.0187 0.0183,-0.005 0.0334,-0.0143 0.0334,-0.0215 0,-0.007 0.009,-0.0131 0.02,-0.0131 0.0109,0 0.02,-0.005 0.02,-0.0103 0,-0.006 0.0225,-0.0222 0.05,-0.0366 0.0276,-0.0144 0.0549,-0.0338 0.0607,-0.0431 0.006,-0.009 0.0193,-0.0168 0.03,-0.0168 0.0106,0 0.0193,-0.006 0.0193,-0.0134 0,-0.008 0.009,-0.0134 0.0204,-0.0134 0.0112,0 0.0239,-0.009 0.0281,-0.02 0.005,-0.0111 0.0221,-0.02 0.0396,-0.02 0.0176,0 0.0319,-0.005 0.0319,-0.0105 0,-0.0128 0.0828,-0.0636 0.0967,-0.0594 0.005,0.002 0.01,-0.002 0.01,-0.009 0,-0.007 0.018,-0.005 0.04,0.002 0.022,0.009 0.0401,0.0193 0.0401,0.0244 0,0.005 0.018,0.0134 0.04,0.0182 0.022,0.005 0.04,0.0143 0.04,0.0211 0,0.007 0.009,0.0122 0.0184,0.0122 0.0101,0 0.0258,0.009 0.035,0.02 0.009,0.0109 0.0279,0.02 0.0417,0.02 0.0137,0 0.025,0.006 0.025,0.0134 0,0.008 0.0152,0.0134 0.0339,0.0134 0.0186,0 0.0373,0.009 0.0414,0.02 0.005,0.0111 0.019,0.02 0.033,0.02 0.0139,0 0.0253,0.005 0.0253,0.0118 0,0.007 0.0105,0.0156 0.0234,0.0202 0.0316,0.0114 0.0823,0.0416 0.10054,0.0598 0.009,0.009 0.0257,0.0149 0.039,0.0149 0.0132,0 0.0241,0.006 0.0241,0.0134 0,0.008 0.0136,0.0134 0.0301,0.0134 0.0166,0 0.0463,0.015 0.0661,0.0334 0.0198,0.0183 0.0484,0.0334 0.0637,0.0334 0.0152,0 0.0312,0.009 0.0354,0.02 0.005,0.0111 0.019,0.02 0.0329,0.02 0.0485,0 0.0281,0.0473 -0.0338,0.0782 -0.0326,0.0162 -0.073,0.0442 -0.0901,0.0622 -0.017,0.018 -0.0371,0.0328 -0.0445,0.0328 -0.008,1.6e-4 -0.0296,0.0153 -0.0495,0.0335 -0.0198,0.0183 -0.0422,0.0334 -0.0499,0.0334 -0.008,0 -0.014,0.005 -0.014,0.0102 0,0.006 -0.0225,0.0222 -0.05,0.0367 -0.0275,0.0144 -0.058,0.0339 -0.0676,0.0431 -0.01,0.009 -0.0257,0.0167 -0.0357,0.0167 -0.01,0 -0.029,0.0138 -0.0424,0.0307 -0.0134,0.0169 -0.0333,0.0352 -0.0444,0.0406 -0.0484,0.0237 -0.0856,0.0484 -0.0733,0.0487 0.008,9e-5 -0.0121,0.0128 -0.0434,0.0283 -0.0312,0.0154 -0.0567,0.0335 -0.0567,0.04 0,0.007 -0.009,0.012 -0.02,0.012 -0.0111,0 -0.02,0.009 -0.02,0.02 0,0.0112 -0.0119,0.02 -0.0271,0.02 -0.0149,0 -0.0306,0.009 -0.0348,0.02 -0.005,0.0111 -0.016,0.02 -0.0263,0.02 -0.0102,0 -0.0186,0.005 -0.0186,0.0112 0,0.006 -0.0211,0.0225 -0.0468,0.0364 -0.0257,0.0137 -0.0468,0.0298 -0.0468,0.0354 0,0.006 -0.009,0.0103 -0.0186,0.0103 -0.0102,0 -0.022,0.009 -0.0263,0.02 -0.005,0.0111 -0.0192,0.02 -0.0334,0.02 -0.0141,0 -0.0293,0.009 -0.0336,0.0209 -0.005,0.0115 -0.0141,0.0171 -0.0214,0.0126 -0.008,-0.005 -0.0134,5.2e-4 -0.0134,0.0111 0,0.0106 -0.015,0.0232 -0.0334,0.0277 -0.0183,0.005 -0.0334,0.0143 -0.0334,0.0215 0,0.007 -0.009,0.0131 -0.0189,0.0131 -0.0105,0 -0.0209,0.005 -0.0234,0.0118 -0.002,0.007 -0.0164,0.005 -0.0312,-0.003 z"
+               id="path56252"
+               inkscape:connector-curvature="0" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.4313232;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -35.94889,35.9093 h -5.30885 v -5.25781 z"
+               id="rect55701-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:4.9705863px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.13810469"
+               x="-59.061882"
+               y="42.855503"
+               id="text55709-6"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-0"
+                 x="-59.061882"
+                 y="42.855503"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.13810469">PARQUET</tspan></text>
+            <path
+               style="fill:#6995f4;fill-opacity:1;stroke:none;stroke-width:0.31363189;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -60.81183,30.65149 h 19.55409 l -0.003,2.44581 h -19.55143 z"
+               id="rect55679-1-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+          </g>
+        </g>
+        <g
+           id="g13072"
+           transform="translate(0,-0.47244895)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,141.24395,-103.38038)"
+             id="g5002"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -296.22335,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -263.44559,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-292.76242"
+               y="155.93517"
+               id="text55709-5-8-1"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-21"
+                 x="-292.76242"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">HTML</tspan></text>
+            <path
+               style="fill:#d45500;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -295.98983,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-0"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               id="text55879-4"
+               y="171.957"
+               x="-291.82764"
+               style="font-style:normal;font-weight:normal;font-size:14.25364685px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#d45500;fill-opacity:1;stroke:none;stroke-width:0.39602885"
+               xml:space="preserve"><tspan
+                 style="fill:#d45500;fill-opacity:1;stroke-width:0.39602885"
+                 y="171.957"
+                 x="-291.82764"
+                 id="tspan55877-8"
+                 sodipodi:role="line">&lt;&gt;</tspan></text>
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,132.71412,-103.38038)"
+             id="g4993"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -247.76738,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-4-0-7-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -214.98962,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-3-7-7-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-244.30646"
+               y="155.93517"
+               id="text55709-5-8-1-5"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-21-3"
+                 x="-244.30646"
+                 y="155.93517"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">HDF5</tspan></text>
+            <path
+               style="fill:#2cab28;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -247.53386,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-7-9-0-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <path
+               style="fill:#2cab28;fill-opacity:1;stroke-width:0.01142396"
+               d="m -228.2809,172.80263 c -0.202,-0.0438 -0.37838,-0.14187 -0.50896,-0.28291 -0.0769,-0.083 -0.11316,-0.13774 -0.16538,-0.24931 -0.0851,-0.18181 -0.0809,-0.13496 -0.0871,-0.97682 l -0.006,-0.74885 -0.0888,0.10126 c -0.12247,0.13959 -0.41022,0.41948 -0.55874,0.54347 -0.84282,0.70368 -1.86652,1.17525 -3.06769,1.41315 -0.52549,0.10407 -1.00462,0.15148 -1.63022,0.16129 -0.53258,0.008 -0.64972,-0.005 -0.84254,-0.098 -0.26642,-0.12809 -0.46447,-0.38731 -0.51956,-0.68003 -0.0177,-0.094 -0.0151,-0.2674 0.005,-0.36491 0.076,-0.36089 0.3608,-0.6475 0.72486,-0.72956 0.065,-0.0147 0.1512,-0.0179 0.54992,-0.0205 0.47948,-0.004 0.60576,-0.009 0.95596,-0.0474 1.30447,-0.14149 2.34564,-0.6184 2.99398,-1.37141 0.39145,-0.45465 0.62298,-0.97115 0.70781,-1.579 0.0247,-0.17684 0.0247,-0.57128 -10e-5,-0.74923 -0.0492,-0.3547 -0.13728,-0.63945 -0.29551,-0.95596 -0.48894,-0.97817 -1.47794,-1.6518 -2.8511,-1.94196 -0.23021,-0.0487 -0.4358,-0.0806 -0.76323,-0.11859 l -0.0539,-0.006 v 2.13841 2.13841 h -2.71369 -2.71369 l -8e-5,1.75516 c -5e-5,1.16801 -0.004,1.77751 -0.0108,1.82197 -0.0153,0.0952 -0.0615,0.22303 -0.11373,0.31501 -0.0524,0.0922 -0.18048,0.23228 -0.27212,0.29768 -0.18509,0.13208 -0.43523,0.19653 -0.65358,0.1684 -0.27855,-0.0359 -0.52289,-0.18249 -0.67666,-0.40601 -0.0501,-0.0728 -0.11087,-0.20816 -0.13607,-0.30313 l -0.0231,-0.0874 v -4.48683 -4.48684 l 0.0237,-0.0867 c 0.0906,-0.33162 0.32687,-0.5747 0.65683,-0.67579 0.0816,-0.025 0.10651,-0.0278 0.25488,-0.0283 0.19819,-4.8e-4 0.25702,0.0121 0.42144,0.0916 0.27365,0.13243 0.46516,0.38659 0.51858,0.68819 0.007,0.041 0.0108,0.6273 0.0108,1.79114 v 1.73014 h 1.76777 1.76779 l 0.003,-1.76544 0.003,-1.76544 0.0236,-0.0822 c 0.0312,-0.10896 0.10123,-0.2515 0.16408,-0.33431 0.17162,-0.22611 0.44991,-0.37003 0.71444,-0.36946 0.0465,1e-4 2.81826,0.004 6.15949,0.008 l 6.07496,0.007 0.0722,0.0225 c 0.18274,0.057 0.31346,0.13732 0.43953,0.27005 0.14446,0.15211 0.23163,0.34134 0.25287,0.54896 0.0476,0.46593 -0.24024,0.88175 -0.70295,1.01524 l -0.0822,0.0237 -2.29996,0.003 -2.29995,0.003 v 1.2331 1.23311 l 1.66778,0.003 1.66779,0.003 0.0925,0.0288 c 0.17469,0.0546 0.32388,0.148 0.44081,0.27613 0.1602,0.17551 0.24276,0.39165 0.24276,0.63553 0,0.24752 -0.0819,0.46012 -0.24645,0.63968 -0.11647,0.12711 -0.26043,0.21641 -0.43936,0.27253 l -0.0903,0.0283 -1.66756,0.003 -1.66757,0.003 -0.003,1.80634 -0.003,1.80635 -0.0288,0.0925 c -0.11257,0.36055 -0.37513,0.60464 -0.72877,0.67751 -0.0965,0.0199 -0.27614,0.0201 -0.36676,4.9e-4 z"
+               id="path4972"
+               inkscape:connector-curvature="0" />
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,389.89266,-103.38038)"
+             id="g57106"
+             style="stroke-width:1.11137545">
+            <g
+               id="g56951"
+               style="stroke-width:1.11137545">
+              <path
+                 sodipodi:nodetypes="cccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-4"
+                 d="m -547.10925,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+                 style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <path
+                 sodipodi:nodetypes="cccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55701-3"
+                 d="m -514.33149,144.95855 h -6.949 v -6.88219 z"
+                 style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686" />
+              <text
+                 id="text55709-5"
+                 y="155.93515"
+                 x="-543.47522"
+                 style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+                 xml:space="preserve"><tspan
+                   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141"
+                   y="155.93515"
+                   x="-543.47522"
+                   id="tspan55707-4"
+                   sodipodi:role="line">JSON</tspan></text>
+              <path
+                 sodipodi:nodetypes="ccccc"
+                 inkscape:connector-curvature="0"
+                 id="rect55679-1-7"
+                 d="m -546.87573,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+                 style="fill:#7c4515;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />
+              <text
+                 xml:space="preserve"
+                 style="font-style:normal;font-weight:normal;font-size:14.25364685px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#7c4515;fill-opacity:1;stroke:none;stroke-width:0.39602885"
+                 x="-539.08789"
+                 y="171.74121"
+                 id="text55879"><tspan
+                   sodipodi:role="line"
+                   id="tspan55877"
+                   x="-539.08789"
+                   y="171.74121"
+                   style="fill:#7c4515;fill-opacity:1;stroke-width:0.39602885">{}</tspan></text>
+            </g>
+          </g>
+        </g>
+        <g
+           id="g13116"
+           transform="translate(0,-4.3467227)"
+           style="stroke-width:1.11137545">
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,209.56487,-64.753609)"
+             id="g56797"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -385.65173,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-5-3-4"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <g
+               id="g56653"
+               transform="matrix(0.12939252,0,0,0.12939252,-377.59411,159.20773)"
+               style="stroke-width:1.11137545">
+              <g
+                 style="display:none;stroke-width:1.11137545"
+                 id="Placement_ONLY" />
+              <g
+                 id="BASE"
+                 style="stroke-width:1.11137545">
+                <linearGradient
+                   y2="120.7894"
+                   x2="63.999699"
+                   y1="7.0335999"
+                   x1="63.999699"
+                   gradientUnits="userSpaceOnUse"
+                   id="linearGradient11218">
+                  <stop
+                     id="stop11214"
+                     style="stop-color:#4387FD"
+                     offset="0" />
+                  <stop
+                     id="stop11216"
+                     style="stop-color:#4683EA"
+                     offset="1" />
+                </linearGradient>
+                <path
+                   id="path56591"
+                   d="M 27.7906,115.2166 1.54,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7906,12.7831 c 2.0541,-3.5578 5.8503,-5.7495 9.9585,-5.7495 h 52.5012 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2506,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2506,45.4672 c -2.0541,3.5578 -5.8503,5.7495 -9.9585,5.7495 H 37.7491 c -4.1082,1e-4 -7.9043,-2.1916 -9.9585,-5.7494 z"
+                   style="fill:url(#SVGID_1_);stroke-width:1.11137545"
+                   inkscape:connector-curvature="0" />
+              </g>
+              <g
+                 id="shadow"
+                 style="stroke-width:1.11137545">
+                <g
+                   id="g56602"
+                   style="stroke-width:1.11137545">
+                  <defs
+                     id="defs56595">
+                    <path
+                       d="M 27.7908,115.2166 1.5403,69.7493 c -2.0541,-3.5578 -2.0541,-7.9412 0,-11.499 L 27.7908,12.7831 C 29.8449,9.2253 33.641,7.0336 37.7493,7.0336 h 52.501 c 4.1082,0 7.9044,2.1917 9.9585,5.7495 l 26.2505,45.4672 c 2.0541,3.5578 2.0541,7.9412 0,11.499 l -26.2505,45.4672 c -2.0541,3.5578 -5.8502,5.7495 -9.9585,5.7495 h -52.501 c -4.1083,1e-4 -7.9044,-2.1916 -9.9585,-5.7494 z"
+                       id="path12064"
+                       inkscape:connector-curvature="0" />
+                  </defs>
+                  <clipPath
+                     id="clipPath11226">
+                    <use
+                       id="use11224"
+                       style="overflow:visible"
+                       xlink:href="#SVGID_6_"
+                       x="0"
+                       y="0"
+                       width="100%"
+                       height="100%" />
+                  </clipPath>
+                  <polygon
+                     id="polygon56600"
+                     clip-path="url(#SVGID_2_)"
+                     points="119.2289,86.4791 80.6247,47.8749 63.9997,43.4256 49.0668,48.9745 43.3004,63.9999 47.9372,80.729 88.8747,121.6665 97.5622,121.2811 "
+                     style="opacity:0.07000002;stroke-width:1.11137545" />
+                </g>
+              </g>
+              <g
+                 id="art"
+                 style="stroke-width:1.11137545">
+                <g
+                   id="g56617"
+                   style="stroke-width:1.11137545">
+                  <g
+                     id="g56615"
+                     style="stroke-width:1.11137545">
+                    <path
+                       id="path56605"
+                       d="m 63.9997,40.804 c -12.8097,0 -23.1947,10.3849 -23.1947,23.1959 0,12.8097 10.385,23.1947 23.1947,23.1947 12.8097,0 23.1947,-10.385 23.1947,-23.1947 C 87.1945,51.1889 76.8095,40.804 63.9997,40.804 m 0,40.7948 c -9.72,0 -17.599,-7.879 -17.599,-17.599 0,-9.72 7.879,-17.6007 17.599,-17.6007 9.72,0 17.6001,7.8807 17.6001,17.6007 0,9.72 -7.8801,17.599 -17.6001,17.599"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56607"
+                       d="m 52.9898,63.1041 v 7.2091 c 1.0659,1.8326 2.5751,3.3702 4.381,4.4754 V 63.1041 Z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56609"
+                       d="m 61.6754,57.0263 v 19.4109 c 0.7448,0.1363 1.5067,0.2199 2.2892,0.2199 0.7145,0 1.4098,-0.0752 2.093,-0.189 V 57.0263 Z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56611"
+                       d="m 70.7656,66.1008 v 8.5614 c 1.8325,-1.1695 3.3478,-2.7822 4.3822,-4.7002 v -3.8612 z"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                    <path
+                       id="path56613"
+                       d="m 80.6914,78.2867 -2.403,2.4049 c -0.4239,0.4227 -0.4239,1.1132 0,1.5371 l 9.1143,9.112 c 0.4227,0.4238 1.1144,0.4238 1.5371,0 l 2.403,-2.4013 c 0.4214,-0.4227 0.4214,-1.1143 0,-1.5365 l -9.1156,-9.1162 c -0.4215,-0.4221 -1.1143,-0.4221 -1.5358,0"
+                       style="fill:#ffffff;stroke-width:1.11137545"
+                       inkscape:connector-curvature="0" />
+                  </g>
+                </g>
+              </g>
+              <g
+                 id="Guides"
+                 style="stroke-width:1.11137545" />
+            </g>
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -352.87397,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-5-6-9"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#4486f6;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -385.41821,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-5-6-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-379.1109"
+               y="155.33165"
+               id="text55709-5-8-0-4"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2-6"
+                 x="-379.1109"
+                 y="155.33165"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">GBQ</tspan></text>
+          </g>
+          <g
+             transform="matrix(0.7639736,0,0,0.7639736,270.63242,-64.753609)"
+             id="g56805"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.52249008;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -428.29495,138.07636 h 25.82876 l 6.84864,6.7172 v 33.52589 h -32.6774 z"
+               id="rect55679-5-3"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.56457865;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="m -395.51719,144.95855 h -6.949 v -6.88219 z"
+               id="rect55701-5-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#d22b28;fill-opacity:1;stroke:none;stroke-width:0.41052711;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -428.06143,138.07636 h 25.59524 l -0.004,3.20143 h -25.59176 z"
+               id="rect55679-1-5-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <path
+               style="fill:#d22b28;fill-opacity:1;stroke-width:0.03987985"
+               d="m -412.94012,174.83899 c -2.39286,-0.17337 -4.25462,-0.92363 -4.80603,-1.93675 -0.20796,-0.38207 -0.1997,-0.15791 -0.1997,-5.4163 0,-5.2584 -0.008,-5.03422 0.1997,-5.41631 0.52229,-0.95963 2.12948,-1.65202 4.42926,-1.90817 0.59425,-0.0661 2.22107,-0.055 2.85272,0.0195 2.42842,0.28682 3.99755,1.04433 4.40698,2.12751 l 0.0784,0.20754 0.01,4.80836 c 0.007,3.25426 -0.003,4.88804 -0.0279,5.05486 -0.18694,1.21275 -1.86095,2.11808 -4.44966,2.40645 -0.62283,0.0695 -1.89574,0.0965 -2.49389,0.0532 z m 2.42212,-1.10913 c 1.01274,-0.12241 1.89648,-0.35364 2.52978,-0.66193 0.45831,-0.2231 0.73601,-0.43117 0.86172,-0.64568 0.0872,-0.14886 0.0893,-0.16846 0.0893,-0.89661 0,-0.40943 -0.0124,-0.73888 -0.0269,-0.73213 -0.0149,0.007 -0.16397,0.0828 -0.33145,0.16913 -1.54197,0.7942 -4.13331,1.10309 -6.47743,0.7721 -1.10764,-0.1564 -2.14234,-0.47058 -2.79149,-0.84762 -0.0946,-0.055 -0.1794,-0.0999 -0.18839,-0.0999 -0.01,0 -0.0164,0.34474 -0.0164,0.7661 v 0.76611 l 0.10825,0.14942 c 0.2826,0.39014 1.05764,0.77963 2.03963,1.02501 0.50715,0.12668 0.98591,0.20062 1.85311,0.28599 0.27949,0.0275 2.0144,-0.01 2.35036,-0.0499 z m -0.0897,-2.99878 c 1.48756,-0.16113 2.68968,-0.55239 3.26059,-1.06123 0.30547,-0.27227 0.30981,-0.28973 0.30981,-1.2474 0,-0.45588 -0.007,-0.82886 -0.0139,-0.82886 -0.008,0 -0.14893,0.0711 -0.31398,0.15803 -1.28011,0.67416 -3.50461,1.02839 -5.53905,0.88204 -1.40344,-0.10096 -2.61466,-0.38428 -3.49862,-0.81837 -0.22696,-0.11141 -0.42477,-0.20828 -0.43957,-0.21517 -0.0147,-0.007 -0.027,0.36911 -0.027,0.83552 0,0.96882 -0.004,0.94996 0.30074,1.22903 0.49729,0.45451 1.6456,0.85736 2.94812,1.03429 0.84879,0.11526 2.11973,0.12888 3.01279,0.0321 z m -0.25117,-3.17465 c 1.79501,-0.15181 3.38835,-0.72508 3.73976,-1.34554 0.0773,-0.13644 0.0806,-0.17795 0.0813,-0.99242 l 4.3e-4,-0.85007 -0.29605,0.15736 c -0.83154,0.442 -2.06898,0.76261 -3.39994,0.88088 -0.64577,0.0574 -2.00104,0.048 -2.62283,-0.0182 -1.30016,-0.13837 -2.35469,-0.41553 -3.14541,-0.82671 l -0.36781,-0.19125 v 0.87316 0.87316 l 0.11828,0.155 c 0.46867,0.61447 1.93789,1.13255 3.63158,1.28057 0.51111,0.0447 1.75415,0.0469 2.26066,0.004 z m -0.19736,-3.19261 c 1.57372,-0.11691 2.89719,-0.48754 3.5985,-1.00787 0.23246,-0.17245 0.42043,-0.42724 0.42043,-0.56986 0,-0.14316 -0.18674,-0.40773 -0.40076,-0.5678 -0.59209,-0.44282 -1.65343,-0.78705 -2.97228,-0.96401 -0.61144,-0.0821 -2.35152,-0.0922 -2.96037,-0.0173 -1.07536,0.13231 -1.92804,0.35436 -2.5836,0.67278 -0.55306,0.26864 -0.91503,0.60618 -0.91503,0.85329 0,0.28182 0.3789,0.63844 0.97521,0.91787 1.15871,0.543 3.07651,0.81371 4.8379,0.6829 z"
+               id="path56342"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccscccccscccccccsccscssccccccccssccccsccccccccccccccccsccccsscccsscc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.58333302px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.29405141"
+               x="-421.75409"
+               y="155.33165"
+               id="text55709-5-8-0"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2"
+                 x="-421.75409"
+                 y="155.33165"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#afabab;fill-opacity:1;stroke-width:0.29405141">SQL</tspan></text>
+          </g>
+          <g
+             transform="translate(-4.7030536,-0.28414145)"
+             id="g12598"
+             style="stroke-width:1.11137545">
+            <path
+               style="fill:#adabab;fill-opacity:0.39215686;stroke:#ffffff;stroke-width:0.39916864;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="M -23.381309,41.017226 H -3.6488181 L 1.5833621,46.14899 V 71.761885 H -23.381309 Z"
+               id="rect55679-5-3-6"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccccc" />
+            <path
+               style="fill:#afabab;fill-opacity:1;stroke:none;stroke-width:0.4313232;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:0.39215686"
+               d="M 1.6600345,46.275038 H -3.6488181 V 41.017226 Z"
+               id="rect55701-5-6-7"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="cccc" />
+            <path
+               style="fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.31363189;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               d="m -23.202906,41.017226 h 19.5540879 l -0.00306,2.445808 H -23.203307 Z"
+               id="rect55679-1-5-6-5"
+               inkscape:connector-curvature="0"
+               sodipodi:nodetypes="ccccc" />
+            <text
+               xml:space="preserve"
+               style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.08538723px;line-height:1.25;font-family:monospace;-inkscape-font-specification:monospace;letter-spacing:0px;word-spacing:0px;fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.22464751"
+               x="-18.192799"
+               y="60.704777"
+               id="text55709-5-8-0-5-9"><tspan
+                 sodipodi:role="line"
+                 id="tspan55707-4-9-2-62-1"
+                 x="-18.192799"
+                 y="60.704777"
+                 style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:monospace;-inkscape-font-specification:monospace;fill:#333333;fill-opacity:1;stroke-width:0.22464751">...</tspan></text>
+          </g>
+        </g>
+      </g>
+      <path
+         sodipodi:nodetypes="cc"
+         inkscape:connector-curvature="0"
+         id="path6109-1"
+         d="m 152.96041,44.790275 h 26.66986"
+         style="fill:none;stroke:#000000;stroke-width:0.44455019;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-0)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/03_subset_columns.svg b/pandas/doc/source/_static/schemas/03_subset_columns.svg
new file mode 100644
index 0000000000000..5495d3f67bcfc
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/03_subset_columns.svg
@@ -0,0 +1,327 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="305.39435mm"
+   height="77.216072mm"
+   viewBox="0 0 305.39435 77.216072"
+   version="1.1"
+   id="svg8981"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="03_subset_columns.svg">
+  <defs
+     id="defs8975">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="515.3968"
+     inkscape:cy="188.43624"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata8978">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(221.9424,-112.49315)">
+    <g
+       id="g10981"
+       transform="matrix(0.89983205,0,0,0.89933244,-6.9361517,15.210989)"
+       style="stroke-width:1.1116271">
+      <g
+         transform="translate(3.4400191e-6)"
+         id="g10300"
+         style="stroke-width:1.1116271">
+        <path
+           d="M 6.4919652,138.65319 H 31.379905 V 126.46518 H 6.4919652 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-5-6-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,124.93718 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-47-3-7-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,138.65319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-1-5-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 6.4919552,151.35319 H 31.379905 V 139.16518 H 6.4919552 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-4-3-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,151.35319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-20-5-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,124.93718 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-0-1-6-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,138.65319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-3-2-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,151.35319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-0-9-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 6.4919652,164.05319 H 31.379905 V 151.86528 H 6.4919652 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-5-0-1-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,164.05319 h 24.88796 v -12.18791 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-8-4-2-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,164.05319 h 24.88796 v -12.18791 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-10-8-7-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 6.4919652,176.75319 H 31.379905 V 164.56518 H 6.4919652 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-8-8-1-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,176.75319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-5-2-8-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 6.4919552,189.45319 H 31.379905 V 177.26518 H 6.4919552 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-0-5-7-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 32.907955,189.45319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-2-4-9-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,176.75319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-2-7-2-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 58.307955,189.45319 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-8-5-0-3"
+           inkscape:connector-curvature="0" />
+      </g>
+      <g
+         transform="translate(-1.765534e-6,2.601185)"
+         id="g10331"
+         style="stroke-width:1.1116271">
+        <path
+           d="m -221.68636,136.052 h 24.88794 v -12.188 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-5-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,122.33599 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-47-3-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,136.052 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-1-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -221.68637,148.75201 h 24.88795 v -12.18802 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-4-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,148.75201 h 24.88796 V 136.564 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-20-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,122.33599 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-0-1-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,136.052 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-3-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,148.75201 h 24.88796 V 136.564 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-0-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -221.68636,161.45201 h 24.88794 V 149.2641 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-5-0-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,161.45201 h 24.88796 V 149.2641 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-8-4-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,161.45201 h 24.88796 V 149.2641 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-10-8-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,161.45201 h 24.88796 V 149.2641 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-67-2-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,122.33599 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-0-2-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,136.052 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-0-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,148.75201 h 24.88796 v -12.18802 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-19-24-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,161.45201 h 24.88793 V 149.2641 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,122.33599 h 24.88793 v -12.188 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-2-5-01-8-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,136.052 h 24.88793 v -12.18801 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-4-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,148.75201 h 24.88793 V 136.564 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -221.68636,174.152 h 24.88794 v -12.18801 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-8-8-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,174.152 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-5-2-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -221.68637,186.85201 h 24.88795 v -12.18802 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-0-5-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -195.27037,186.85201 h 24.88796 V 174.664 h -24.88796 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-2-4-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,174.152 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-2-7-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -169.87037,186.85201 h 24.88796 V 174.664 h -24.88796 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-8-5-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,174.152 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-5-5-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -144.47037,186.85201 h 24.88796 v -12.18802 h -24.88796 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-19-2-2-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,174.152 h 24.88793 v -12.18801 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -119.07037,186.85201 h 24.88793 V 174.664 h -24.88793 z"
+           style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-3-6-5"
+           inkscape:connector-curvature="0" />
+      </g>
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-7"
+         d="m -68.124379,151.10823 h 47.88959"
+         style="fill:none;stroke:#000000;stroke-width:0.44465086;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/03_subset_columns_rows.svg b/pandas/doc/source/_static/schemas/03_subset_columns_rows.svg
new file mode 100644
index 0000000000000..5ea9d609ec1c3
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/03_subset_columns_rows.svg
@@ -0,0 +1,272 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="247.37685mm"
+   height="62.54488mm"
+   viewBox="0 0 247.37685 62.54488"
+   version="1.1"
+   id="svg8981"
+   sodipodi:docname="03_subset_columns_rows.svg"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)">
+  <defs
+     id="defs8975">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="382.49796"
+     inkscape:cy="12.478764"
+     inkscape:document-units="mm"
+     inkscape:current-layer="g10981"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata8978">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(190.64444,-119.82874)">
+    <g
+       id="g10981"
+       transform="matrix(0.89983205,0,0,0.89933244,-6.9361517,15.210989)"
+       style="stroke-width:1.1116271">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-5-6-6"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -5.3791094,156.43886 H 14.771891 v -9.8551 H -5.3791094 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-47-3-7-4"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.009106,145.34823 h 20.151016 v -9.85509 H 16.009106 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-8-1-5-9"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.009106,156.43886 h 20.151016 v -9.8551 H 16.009106 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-6-4-3-5"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -5.3791175,166.70796 H 14.771891 v -9.85511 H -5.3791175 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-0-20-5-0"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.009106,166.70796 h 20.151016 v -9.8551 H 16.009106 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-0-1-6-4"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 36.574705,145.34823 h 20.151017 v -9.85509 H 36.574705 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-64-3-2-8"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 36.574705,156.43886 h 20.151017 v -9.8551 H 36.574705 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-37-0-9-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 36.574705,166.70796 h 20.151017 v -9.8551 H 36.574705 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-5-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -190.12809,141.03585 h 20.15101 v -9.85509 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-47-3-7"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,129.94522 h 20.15101 v -9.85509 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-8-1-5"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,141.03585 h 20.15101 v -9.8551 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-6-4-3"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -190.12809,151.30495 h 20.15101 v -9.85511 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-0-20-5"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,151.30495 h 20.15101 v -9.85511 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-0-1-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,129.94522 h 20.15102 v -9.85509 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-64-3-2"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,141.03585 h 20.15102 v -9.8551 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-37-0-9"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,151.30495 h 20.15102 v -9.85511 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-5-0-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -190.12809,161.57404 h 20.15101 v -9.85502 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-8-4-2"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,161.57404 h 20.15101 v -9.85502 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-10-8-7"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,161.57404 h 20.15102 v -9.85502 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-67-2-0"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,161.57404 h 20.15102 v -9.85502 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-0-2-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,129.94522 h 20.15102 v -9.85509 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-8-0-3"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,141.03585 h 20.15102 v -9.85509 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-19-24-6"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,151.30495 h 20.15102 v -9.85511 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-0"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,161.57404 h 20.150992 v -9.85502 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-01-8-6"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,129.94522 h 20.150992 v -9.85509 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-59-4-2"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,141.03585 h 20.150992 v -9.8551 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-7-0-6"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,151.30495 h 20.150992 v -9.85511 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-8-8-1"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -190.12809,171.84313 h 20.15101 v -9.8551 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-8-5-2-8"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,171.84313 h 20.15101 v -9.8551 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-6-0-5-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -190.12809,182.11223 h 20.15101 v -9.85511 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-0-2-4-9"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -168.73986,182.11223 h 20.15101 v -9.8551 h -20.15101 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-64-2-7-2"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,171.84313 h 20.15102 v -9.8551 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-37-8-5-0"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -148.17427,182.11223 h 20.15102 v -9.8551 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-8-5-5-2"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,171.84313 h 20.15102 v -9.8551 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-19-2-2-3"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -127.60867,182.11223 h 20.15102 v -9.85511 h -20.15102 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,171.84313 h 20.150992 v -9.8551 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-7-3-6-5"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920707;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -107.04307,182.11223 h 20.150992 v -9.8551 h -20.150992 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-7"
+         d="m -65.823888,151.10688 h 38.774729"
+         style="fill:none;stroke:#000000;stroke-width:0.44465086;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/03_subset_rows.svg b/pandas/doc/source/_static/schemas/03_subset_rows.svg
new file mode 100644
index 0000000000000..41fe07d7fc34e
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/03_subset_rows.svg
@@ -0,0 +1,316 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="320.56647mm"
+   height="69.494316mm"
+   viewBox="0 0 320.56647 69.494316"
+   version="1.1"
+   id="svg8981"
+   sodipodi:docname="03_subset_rows.svg"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)">
+  <defs
+     id="defs8975">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="838.29471"
+     inkscape:cy="34.832455"
+     inkscape:document-units="mm"
+     inkscape:current-layer="g10981"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata8978">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(206.67275,-116.35402)">
+    <g
+       id="g10981"
+       transform="matrix(0.89983205,0,0,0.89933244,-6.9361517,15.210989)"
+       style="stroke-width:1.1116271">
+      <g
+         id="g11450"
+         transform="matrix(0.89983997,0,0,0.89925792,-4.3915493,15.222243)"
+         style="stroke-width:1.23576057">
+        <g
+           transform="translate(-1.1404361e-5)"
+           id="g11347"
+           style="stroke-width:1.23576057">
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-74-5-5-6-6"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 6.49197,157.7024 h 24.88794 v -12.188 H 6.49197 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-17-47-3-7-4"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 32.90796,143.98639 h 24.88796 v -12.188 H 32.90796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-1-8-1-5-9"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M 32.90796,157.7024 H 57.79592 V 145.51439 H 32.90796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-1-6-4-3-5"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M 6.49196,170.40241 H 31.37991 V 158.21439 H 6.49196 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-1-0-20-5-0"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M 32.90796,170.40241 H 57.79592 V 158.2144 H 32.90796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-6-7-0-1-6-4"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 58.30796,143.98639 h 24.88796 v -12.188 H 58.30796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-5-0-64-3-2-8"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M 58.30796,157.7024 H 83.19592 V 145.51439 H 58.30796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-3-4-37-0-9-7"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M 58.30796,170.40241 H 83.19592 V 158.2144 H 58.30796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-8-8-0-2-9-2"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.70796,143.98639 h 24.88796 v -12.188 H 83.70796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-1-8-0-3-2"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.70796,157.7024 h 24.88796 v -12.188 H 83.70796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-0-19-24-6-6"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.70796,170.40241 h 24.88796 V 158.21439 H 83.70796 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-8-2-5-01-8-6-0"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.10796,143.98639 h 24.88793 v -12.188 h -24.88793 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-2-6-59-4-2-6"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.10796,157.7024 h 24.88793 v -12.18801 h -24.88793 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-1-4-7-0-6-1"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.10796,170.40241 h 24.88793 V 158.2144 h -24.88793 z" />
+        </g>
+        <g
+           id="g11378"
+           style="stroke-width:1.23576057">
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-74-5-5-6"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -221.68636,138.65318 h 24.88794 v -12.188 h -24.88794 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-17-47-3-7"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,124.93717 h 24.88796 v -12.188 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-1-8-1-5"
+             style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,138.65318 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-1-6-4-3"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -221.68637,151.35319 h 24.88795 v -12.18802 h -24.88795 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-1-0-20-5"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,151.35319 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-6-7-0-1-6"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,124.93717 h 24.88796 v -12.188 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-5-0-64-3-2"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,138.65318 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-3-4-37-0-9"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,151.35319 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-7-0-5-0-1"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -221.68636,164.05319 h 24.88794 v -12.18791 h -24.88794 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-8-8-4-2"
+             style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,164.05319 h 24.88796 v -12.18791 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-7-8-51-10-8-7"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,164.05319 h 24.88796 v -12.18791 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-2-0-67-2-0"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,164.05319 h 24.88796 v -12.18791 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-8-8-0-2-9"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,124.93717 h 24.88796 v -12.188 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-1-8-0-3"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,138.65318 h 24.88796 v -12.188 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-0-19-24-6"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,151.35319 h 24.88796 v -12.18802 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-0"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,164.05319 h 24.887928 v -12.18791 h -24.887928 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-8-2-5-01-8-6"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,124.93717 h 24.887928 v -12.188 h -24.887928 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-2-6-59-4-2"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,138.65318 h 24.887928 v -12.18801 h -24.887928 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-1-4-7-0-6"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,151.35319 h 24.887928 v -12.18801 h -24.887928 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-74-5-8-8-1"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -221.68636,176.75318 h 24.88794 v -12.18801 h -24.88794 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-1-8-5-2-8"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,176.75318 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-1-6-0-5-7"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -221.68637,189.45319 h 24.88795 v -12.18802 h -24.88795 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-1-0-2-4-9"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -195.27037,189.45319 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-5-0-64-2-7-2"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,176.75318 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-3-4-37-8-5-0"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -169.87037,189.45319 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-1-8-5-5-2"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,176.75318 h 24.88796 v -12.18801 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-0-19-2-2-3"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -144.47037,189.45319 h 24.88796 v -12.18802 h -24.88796 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-7"
+             style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,176.75318 h 24.887928 v -12.18801 h -24.887928 z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-5-1-4-7-3-6-5"
+             style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.63276947;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m -119.07037,189.45319 h 24.887928 v -12.18801 h -24.887928 z" />
+        </g>
+        <path
+           style="fill:none;stroke:#000000;stroke-width:0.49430427;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1)"
+           d="m -68.161706,151.10823 h 47.88959"
+           id="path6109-2-9-6-9-7"
+           inkscape:connector-curvature="0" />
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/04_plot_overview.svg b/pandas/doc/source/_static/schemas/04_plot_overview.svg
new file mode 100644
index 0000000000000..44ae5b6ae5e33
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/04_plot_overview.svg
@@ -0,0 +1,6443 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="279.05978mm"
+   height="79.448936mm"
+   viewBox="0 0 279.05978 79.448937"
+   version="1.1"
+   id="svg15565"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="04_plot_overview.svg">
+  <defs
+     id="defs15559">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <clipPath
+       id="clipPath23666-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23664-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23678-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23676-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23690-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23688-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23702-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23700-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23714-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23712-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23726-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23724-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23738-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23736-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23750-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23748-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23762-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23760-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23774-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23772-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23786-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23784-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23798-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23796-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23810-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23808-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23822-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23820-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23834-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23832-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23846-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23844-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23858-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23856-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23870-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23868-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23882-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23880-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23894-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23892-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23906-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23904-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23918-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23916-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23930-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23928-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23942-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23940-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23954-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23952-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23966-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23964-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23978-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23976-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23990-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23988-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24002-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24000-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24014-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24012-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24026-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24024-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24038-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24036-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24050-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24048-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24062-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24060-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24074-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24072-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24086-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24084-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24098-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24096-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24110-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24108-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24122-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24120-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24134-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24132-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24146-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24144-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24158-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24156-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24170-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24168-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24182-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24180-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24194-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24192-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24206-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24204-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24218-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24216-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24230-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24228-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24242-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24240-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24254-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24252-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24266-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24264-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24278-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24276-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24290-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24288-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24302-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24300-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24314-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24312-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24326-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24324-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24338-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24336-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24350-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24348-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24362-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24360-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24374-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24372-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24386-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24384-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24398-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24396-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24410-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24408-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24422-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24420-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24434-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24432-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24446-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24444-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24458-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24456-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24470-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24468-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24482-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24480-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24494-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24492-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24506-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24504-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24518-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24516-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24530-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24528-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24542-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24540-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24554-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24552-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24566-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24564-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24578-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24576-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24590-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24588-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24602-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24600-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24614-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24612-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24626-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24624-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24638-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24636-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24650-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24648-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24662-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24660-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24674-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24672-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24686-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24684-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24698-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24696-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24710-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24708-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24722-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24720-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24734-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24732-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24746-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24744-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24758-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24756-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24770-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24768-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24782-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24780-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24794-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24792-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24806-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24804-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24818-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24816-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24830-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24828-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24842-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24840-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24854-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24852-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24866-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24864-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24878-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24876-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24890-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24888-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24902-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24900-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24914-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24912-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24926-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24924-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24938-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24936-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24950-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24948-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24962-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24960-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24974-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24972-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24986-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24984-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath24998-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path24996-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25010-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25008-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25022-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25020-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25034-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25032-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25046-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25044-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25058-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25056-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25070-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25068-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25082-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25080-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25094-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25092-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25106-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25104-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25118-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25116-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25130-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25128-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25142-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25140-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25154-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25152-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25166-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25164-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25178-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25176-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25190-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25188-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25202-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25200-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25214-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25212-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25226-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25224-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25238-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25236-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25250-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25248-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25262-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25260-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25274-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25272-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25286-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25284-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25298-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25296-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25310-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25308-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25322-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25320-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25334-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25332-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25346-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25344-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25358-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25356-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25370-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25368-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25382-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25380-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25394-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25392-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25406-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25404-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25418-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25416-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25430-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25428-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25442-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25440-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25454-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25452-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25466-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25464-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25478-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25476-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25490-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25488-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25502-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25500-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25514-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25512-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25526-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25524-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25538-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25536-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25550-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25548-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25562-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25560-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25574-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25572-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25586-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25584-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25598-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25596-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25610-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25608-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25622-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25620-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25634-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25632-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25646-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25644-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25658-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25656-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25670-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25668-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25682-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25680-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25694-2"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25692-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25706-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25704-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25718-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25716-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25730-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25728-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25742-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25740-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25754-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25752-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25766-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25764-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25778-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25776-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25790-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25788-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25802-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25800-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25814-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25812-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25826-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25824-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25838-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25836-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25850-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25848-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25862-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25860-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25874-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25872-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25886-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25884-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25898-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25896-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25910-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25908-6"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25922-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25920-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25934-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25932-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25946-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25944-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25958-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25956-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25970-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25968-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25982-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25980-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath25994-7"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path25992-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26006-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26004-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26018-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26016-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26030-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26028-1"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26042-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26040-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26054-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26052-8"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26066-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26064-5"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26078-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26076-4"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26090-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26088-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26102-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26100-3"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26114-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26112-0"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26126-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26124-7"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26138-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26136-9"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath26150-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path26148-2"
+         d="M -2.736068,-2.736068 H 2.736068 V 2.736068 H -2.736068 Z" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath272">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path270"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath132">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path130"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath142">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path140"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath154">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path152"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath166">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path164"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath178">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path176"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath190">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path188"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath202">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path200"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath214">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path212"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath226">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path224"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath238">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path236"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath250">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path248"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath262">
+      <path
+         d="m -3.5,-3.5 h 7 v 7 h -7 z"
+         id="path260"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath122">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path120"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath112">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path110"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath102">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path100"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath92">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path90"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath82">
+      <path
+         d="M 54,36 H 388.8 V 253.44 H 54 Z"
+         id="path80"
+         inkscape:connector-curvature="0" />
+    </clipPath>
+    <clipPath
+       id="clipPath20926"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20924"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20936"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20934"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20946"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20944"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20956"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20954"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20966"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20964"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20976"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20974"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20986"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20984"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20996"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20994"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21006"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21004"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21016"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21014"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21026"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21024"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21036"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21034"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21048"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21046"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21060"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21058"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21072"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21070"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21084"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21082"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21096"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21094"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21108"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21106"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21118"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21116"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21128"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21126"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23791"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23789"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23801"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23799"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23811"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23809"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23821"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23819"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23831"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23829"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23841"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23839"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23851"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23849"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23861"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23859"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23871"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23869"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath23881"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path23879"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath42476"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path42474"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath42486"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path42484"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath42496"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path42494"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21118-6"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21116-7"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21128-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21126-6"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20926-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20924-6"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20936-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20934-9"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20946-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20944-8"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20956-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20954-2"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20966-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20964-3"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20976-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20974-0"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20986-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20984-8"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath20996-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path20994-0"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21006-9"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21004-6"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21016-3"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21014-8"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21026-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21024-6"
+         d="M 54,36 H 388.8 V 253.44 H 54 Z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21036-1"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21034-1"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21048-5"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21046-9"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21060-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21058-4"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21072-8"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21070-1"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21084-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21082-3"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21096-0"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21094-4"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+    <clipPath
+       id="clipPath21108-4"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path21106-4"
+         d="m -3.5,-3.5 h 7 v 7 h -7 z" />
+    </clipPath>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.52850025"
+     inkscape:cx="304.48721"
+     inkscape:cy="323.02301"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     showguides="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0" />
+  <metadata
+     id="metadata15562">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(2.9900145,-83.072472)">
+    <g
+       id="g3126"
+       transform="matrix(0.89985527,0,0,0.89947476,13.672209,12.344233)"
+       style="stroke-width:1.11152482">
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2"
+         d="M 99.794627,121.59047 H 147.68421"
+         style="fill:none;stroke:#000000;stroke-width:0.44460994;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7)" />
+      <text
+         id="text47247-6-7"
+         y="115.60715"
+         x="106.41215"
+         style="font-style:normal;font-weight:normal;font-size:12.12705898px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.33698818"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.08470631px;font-family:monospace;-inkscape-font-specification:monospace;stroke-width:0.33698818"
+           y="115.60715"
+           x="106.41215"
+           id="tspan47245-1-51"
+           sodipodi:role="line">.plot.*</tspan></text>
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -2.8037002,112.52508 H 15.307077 v -8.8691 H -2.8037002 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-4"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 16.419028,102.54403 H 34.529812 V 93.674949 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-6"
+         style="fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.419028,112.52508 h 18.110784 v -8.8691 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -2.8037074,121.76678 H 15.307077 v -8.86909 H -2.8037074 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-6"
+         style="fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.419028,121.76678 h 18.110784 v -8.86909 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 34.902427,102.54403 H 53.013218 V 93.674949 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 34.902427,112.52508 h 18.110791 v -8.8691 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 34.902427,121.76678 h 18.110791 v -8.86909 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -2.8037002,131.00849 H 15.307077 V 122.1394 H -2.8037002 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-3"
+         style="fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 16.419028,131.00849 H 34.529812 V 122.1394 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 34.902427,131.00849 H 53.013218 V 122.1394 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -2.8037002,140.2502 H 15.307077 v -8.86909 H -2.8037002 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-2"
+         style="fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.419028,140.2502 h 18.110784 v -8.86909 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 34.902427,140.2502 h 18.110791 v -8.86909 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -2.8037002,149.4919 H 15.307077 v -8.86909 H -2.8037002 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-9"
+         style="fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 16.419028,149.4919 h 18.110784 v -8.86909 H 16.419028 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 34.902427,149.4919 h 18.110791 v -8.86909 H 34.902427 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-2"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.385833,131.00849 h 18.11077 v -8.86909 h -18.11077 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-4"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 53.385826,102.54403 H 71.496603 V 93.674949 H 53.385826 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.385833,112.52508 h 18.11077 v -8.8691 h -18.11077 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-6"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.385826,121.76678 h 18.110777 v -8.86909 H 53.385826 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-5"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.385833,140.2502 h 18.11077 v -8.86909 h -18.11077 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.385833,149.4919 h 18.11077 v -8.86909 h -18.11077 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-4-7-1"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 71.869225,131.00849 H 89.979987 V 122.1394 H 71.869225 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-3"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 71.869218,102.54403 H 89.979987 V 93.674949 H 71.869218 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-3"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 71.869225,112.52508 h 18.110762 v -8.8691 H 71.869225 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-3"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 71.869218,121.76678 h 18.110769 v -8.86909 H 71.869218 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-6-6-8"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 71.869225,140.2502 h 18.110762 v -8.86909 H 71.869225 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-2-3-5"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.41416982;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 71.869225,149.4919 h 18.110762 v -8.86909 H 71.869225 Z" />
+      <g
+         style="fill:#ffca00;fill-opacity:1;stroke-width:1.11152482"
+         id="g3833">
+        <path
+           inkscape:connector-curvature="0"
+           id="path23812"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:0.35804299;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           d="m 189.7326,108.3544 c 0.18739,0 0.36713,-0.0744 0.49963,-0.20696 0.13251,-0.1325 0.20696,-0.31224 0.20696,-0.49963 0,-0.18739 -0.0744,-0.36713 -0.20696,-0.49964 -0.1325,-0.1325 -0.31224,-0.20695 -0.49963,-0.20695 -0.18739,0 -0.36713,0.0744 -0.49963,0.20695 -0.13251,0.13251 -0.20697,0.31225 -0.20697,0.49964 0,0.18739 0.0744,0.36713 0.20697,0.49963 0.1325,0.13251 0.31224,0.20696 0.49963,0.20696 z" />
+        <path
+           d="m 187.00776,102.79094 c 0.18739,0 0.36713,-0.0745 0.49963,-0.20695 0.13251,-0.13251 0.20696,-0.31225 0.20696,-0.49964 0,-0.18739 -0.0744,-0.36713 -0.20696,-0.49963 -0.1325,-0.13251 -0.31224,-0.20696 -0.49963,-0.20696 -0.18739,0 -0.36713,0.0744 -0.49963,0.20696 -0.13251,0.1325 -0.20696,0.31224 -0.20696,0.49963 0,0.18739 0.0744,0.36713 0.20696,0.49964 0.1325,0.1325 0.31224,0.20695 0.49963,0.20695 z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:0.35804299;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           id="path23824"
+           inkscape:connector-curvature="0" />
+        <g
+           id="g23830"
+           clip-path="url(#clipPath23834-3)"
+           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           transform="matrix(0.31599668,0,0,-0.31599668,163.8466,114.13986)">
+          <path
+             d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+             id="path23836"
+             inkscape:connector-curvature="0" />
+        </g>
+        <path
+           inkscape:connector-curvature="0"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:0.35804299;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           d="m 180.19566,113.17061 c 0.18739,0 0.36713,-0.0744 0.49963,-0.20695 0.13251,-0.13251 0.20696,-0.31225 0.20696,-0.49964 0,-0.18739 -0.0744,-0.36713 -0.20696,-0.49963 -0.1325,-0.13251 -0.31224,-0.20696 -0.49963,-0.20696 -0.18739,0 -0.36713,0.0744 -0.49963,0.20696 -0.13251,0.1325 -0.20697,0.31224 -0.20697,0.49963 0,0.18739 0.0745,0.36713 0.20697,0.49964 0.1325,0.1325 0.31224,0.20695 0.49963,0.20695 z"
+           id="path23860" />
+        <g
+           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           clip-path="url(#clipPath23870-0)"
+           id="g23866"
+           transform="matrix(0.31599668,0,0,-0.31599668,201.9944,116.85251)">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23872"
+             style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+             d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+        </g>
+        <g
+           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           id="g23876"
+           transform="matrix(0.31599668,0,0,-0.31599668,208.80651,114.90184)">
+          <g
+             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+             clip-path="url(#clipPath23882-9)"
+             id="g23878">
+            <path
+               inkscape:connector-curvature="0"
+               id="path23884"
+               style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+               d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+          </g>
+        </g>
+        <g
+           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+           transform="matrix(0.31599668,0,0,-0.31599668,193.39883,118.19034)"
+           id="g23886">
+          <g
+             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+             id="g23888">
+            <g
+               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+               clip-path="url(#clipPath23894-6)"
+               id="g23890">
+              <path
+                 inkscape:connector-curvature="0"
+                 id="path23896"
+                 style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                 d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+            </g>
+          </g>
+          <g
+             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+             transform="translate(-8.6230106,6.5685171)"
+             id="g23898">
+            <g
+               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+               id="g23900">
+              <g
+                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                 clip-path="url(#clipPath23906-2)"
+                 id="g23902">
+                <path
+                   inkscape:connector-curvature="0"
+                   id="path23908"
+                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+              </g>
+            </g>
+            <g
+               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+               transform="translate(-150.90269,22.264784)"
+               id="g23910">
+              <g
+                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                 id="g23912">
+                <g
+                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                   clip-path="url(#clipPath23918-4)"
+                   id="g23914" />
+              </g>
+              <g
+                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                 transform="translate(163.8372,-21.440103)"
+                 id="g23922">
+                <g
+                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                   id="g23924">
+                  <g
+                     style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                     clip-path="url(#clipPath23930-7)"
+                     id="g23926">
+                    <path
+                       inkscape:connector-curvature="0"
+                       id="path23932"
+                       style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                       d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                  </g>
+                </g>
+                <g
+                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                   transform="translate(56.049569,-3.9690819)"
+                   id="g23934">
+                  <g
+                     style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                     id="g23936">
+                    <g
+                       style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                       clip-path="url(#clipPath23942-4)"
+                       id="g23938" />
+                  </g>
+                  <g
+                     style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                     transform="translate(-224.19828,9.8688641)"
+                     id="g23946">
+                    <g
+                       style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                       id="g23948">
+                      <g
+                         style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                         clip-path="url(#clipPath23954-1)"
+                         id="g23950" />
+                    </g>
+                    <g
+                       style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                       transform="translate(60.361074,-12.984328)"
+                       id="g23958">
+                      <g
+                         style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                         id="g23960">
+                        <g
+                           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                           clip-path="url(#clipPath23966-8)"
+                           id="g23962">
+                          <path
+                             inkscape:connector-curvature="0"
+                             id="path23968"
+                             style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                             d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                        </g>
+                      </g>
+                      <g
+                         style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                         transform="translate(4.3115053,2.5190697)"
+                         id="g23970">
+                        <g
+                           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                           id="g23972">
+                          <g
+                             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                             clip-path="url(#clipPath23978-3)"
+                             id="g23974">
+                            <path
+                               inkscape:connector-curvature="0"
+                               id="path23980"
+                               style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                               d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                          </g>
+                        </g>
+                        <g
+                           style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                           transform="translate(129.34516,14.48577)"
+                           id="g23982">
+                          <g
+                             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                             id="g23984">
+                            <g
+                               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                               clip-path="url(#clipPath23990-8)"
+                               id="g23986">
+                              <path
+                                 inkscape:connector-curvature="0"
+                                 id="path23992"
+                                 style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                 d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                            </g>
+                          </g>
+                          <g
+                             style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                             transform="translate(-25.869032)"
+                             id="g23994">
+                            <g
+                               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                               id="g23996">
+                              <g
+                                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                 clip-path="url(#clipPath24002-2)"
+                                 id="g23998">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24004"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                              </g>
+                            </g>
+                            <g
+                               style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                               transform="translate(-90.541612,81.166037)"
+                               id="g24006">
+                              <g
+                                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                 id="g24008">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24014-0)"
+                                   id="g24010" />
+                              </g>
+                              <g
+                                 style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                 transform="translate(-30.180537,-85.916396)"
+                                 id="g24018">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24020">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24026-1)"
+                                   id="g24022" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(43.115053,-7.8819118)"
+                                   id="g24030">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24032">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24038-0)"
+                                   id="g24034">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24040"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,-4.3725682)"
+                                   id="g24042">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24044">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24050-5)"
+                                   id="g24046">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24052"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-38.803548,-0.0578724)"
+                                   id="g24054">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24056">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24062-6)"
+                                   id="g24058">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24064"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,19.369583)"
+                                   id="g24066">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24068">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24074-6)"
+                                   id="g24070">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24076"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(73.29559,-19.322976)"
+                                   id="g24078">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24080">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24086-5)"
+                                   id="g24082">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24088"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(6.467258,-0.56263544)"
+                                   id="g24090">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24092">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24098-6)"
+                                   id="g24094">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24100"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-6.467258,5.4351826)"
+                                   id="g24102">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24104">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24110-8)"
+                                   id="g24106">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24112"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-68.984085,34.096487)"
+                                   id="g24114">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24116">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24122-7)"
+                                   id="g24118">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24124"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-25.869032,-34.766842)"
+                                   id="g24126">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24128">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24134-4)"
+                                   id="g24130" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(73.29559,-3.1974499)"
+                                   id="g24138">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24140">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24146-6)"
+                                   id="g24142">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24148"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(60.361074,16.225182)"
+                                   id="g24150">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24152">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24158-9)"
+                                   id="g24154">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24160"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-86.230106,-17.479084)"
+                                   id="g24162">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24164">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24170-0)"
+                                   id="g24166">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24172"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(17.246021,4.4513519)"
+                                   id="g24174">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24176">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24182-1)"
+                                   id="g24178">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24184"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(17.246021,-4.2021919)"
+                                   id="g24186">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24188">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24194-1)"
+                                   id="g24190">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24196"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(8.6230106,6.9848897)"
+                                   id="g24198">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24200">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24206-0)"
+                                   id="g24202">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24208"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-43.115053,-7.243695)"
+                                   id="g24210">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24212">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24218-4)"
+                                   id="g24214">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24220"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(47.426558,1.9483707)"
+                                   id="g24222">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24224">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24230-3)"
+                                   id="g24226">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24232"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-51.738064,-1.6300725)"
+                                   id="g24234">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24236">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24242-1)"
+                                   id="g24238">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24244"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(25.869032,-0.41475218)"
+                                   id="g24246">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24248">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24254-6)"
+                                   id="g24250">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24256"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(116.41064,17.318315)"
+                                   id="g24258">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24260">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24266-3)"
+                                   id="g24262" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-81.918601,0.0530497)"
+                                   id="g24270">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24272">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24278-8)"
+                                   id="g24274">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24280"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-90.541612,-8.4606358)"
+                                   id="g24282">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24284">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24290-5)"
+                                   id="g24286" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(120.72215,-6.5861874)"
+                                   id="g24294">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24296">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24302-6)"
+                                   id="g24298">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24304"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-81.918601,-1.9692819)"
+                                   id="g24306">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24308">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24314-0)"
+                                   id="g24310">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24316"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-4.3115053,-0.29738696)"
+                                   id="g24318">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24320">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24326-4)"
+                                   id="g24322">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24328"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-25.869032,30.624143)"
+                                   id="g24330">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24332">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24338-2)"
+                                   id="g24334">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24340"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-34.492042,-22.136192)"
+                                   id="g24342">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24344">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24350-7)"
+                                   id="g24346" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(0,-1.1092209)"
+                                   id="g24354">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24356">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24362-6)"
+                                   id="g24358" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(103.47613,1.6397179)"
+                                   id="g24366">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24368">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24374-8)"
+                                   id="g24370">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24376"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-120.72215,-1.0031215)"
+                                   id="g24378">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24380">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24386-2)"
+                                   id="g24382" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(120.72215,-7.774192)"
+                                   id="g24390">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24392">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24398-2)"
+                                   id="g24394">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24400"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-30.180537,7.0411416)"
+                                   id="g24402">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24404">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24410-9)"
+                                   id="g24406">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24412"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(21.557527,26.042579)"
+                                   id="g24414">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24416">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24422-0)"
+                                   id="g24418">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24424"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,-17.901861)"
+                                   id="g24426">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24428">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24434-7)"
+                                   id="g24430">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24436"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-43.115053,-15.164188)"
+                                   id="g24438">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24440">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24446-1)"
+                                   id="g24442">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24448"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-73.640511,55.462667)"
+                                   id="g24450">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24452">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24458-2)"
+                                   id="g24454" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(99.509543,42.999191)"
+                                   id="g24462">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24464">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24470-5)"
+                                   id="g24466">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24472"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(159.5257,-99.063075)"
+                                   id="g24474">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24476">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24482-9)"
+                                   id="g24478" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-159.5257,93.090953)"
+                                   id="g24486">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24488">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24494-4)"
+                                   id="g24490">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24496"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(86.230106,-89.285843)"
+                                   id="g24498">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24500">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24506-1)"
+                                   id="g24502">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24508"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-17.246021,5.7486581)"
+                                   id="g24510">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24512">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24518-7)"
+                                   id="g24514">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24520"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(43.115053,9.6068179)"
+                                   id="g24522">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24524">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24530-8)"
+                                   id="g24526">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24532"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-94.853117,-17.901861)"
+                                   id="g24534">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24536">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24542-0)"
+                                   id="g24538">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24544"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(56.049569,11.64682)"
+                                   id="g24546">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24548">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24554-8)"
+                                   id="g24550">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24556"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-155.21419,-10.499017)"
+                                   id="g24558">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24560">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24566-4)"
+                                   id="g24562" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(90.541612,-1.7699308)"
+                                   id="g24570">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24572">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24578-9)"
+                                   id="g24574">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24580"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,6.9880919)"
+                                   id="g24582">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24584">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24590-1)"
+                                   id="g24586">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24592"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-47.426558,-1.1574479)"
+                                   id="g24594">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24596">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24602-4)"
+                                   id="g24598">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24604"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(4.3115053,92.595835)"
+                                   id="g24606">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24608">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24614-2)"
+                                   id="g24610">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24616"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(30.180537,-95.257965)"
+                                   id="g24618">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24620">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24626-0)"
+                                   id="g24622">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24628"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(94.853117,-1.1960295)"
+                                   id="g24630">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24632">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24638-5)"
+                                   id="g24634">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24640"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-103.47613,4.3018482)"
+                                   id="g24642">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24644">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24650-9)"
+                                   id="g24646">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24652"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-90.541612,0.71375956)"
+                                   id="g24654">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24656">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24662-2)"
+                                   id="g24658" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(133.65666,-7.525032)"
+                                   id="g24666">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24668">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24674-3)"
+                                   id="g24670">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24676"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-30.180537,27.097168)"
+                                   id="g24678">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24680">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24686-0)"
+                                   id="g24682">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24688"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,-26.54576)"
+                                   id="g24690">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24692">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24698-0)"
+                                   id="g24694">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24700"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,1.9580161)"
+                                   id="g24702">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24704">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24710-1)"
+                                   id="g24706">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24712"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(34.492042,5.0156077)"
+                                   id="g24714">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24716">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24722-6)"
+                                   id="g24718">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24724"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-47.426558,1.5432639)"
+                                   id="g24726">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24728">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24734-5)"
+                                   id="g24730">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24736"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-4.3115053,20.509322)"
+                                   id="g24738">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24740">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24746-4)"
+                                   id="g24742">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24748"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(25.869032,-10.119606)"
+                                   id="g24750">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24752">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24758-9)"
+                                   id="g24754">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24760"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(68.984085,-11.93298)"
+                                   id="g24762">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24764">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24770-6)"
+                                   id="g24766">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24772"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(56.049569,-4.8226997)"
+                                   id="g24774">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24776">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24782-5)"
+                                   id="g24778" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-241.4443,2.922556)"
+                                   id="g24786">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24788">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24794-2)"
+                                   id="g24790" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(159.5257,2.6331941)"
+                                   id="g24798">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24800">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24806-4)"
+                                   id="g24802">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24808"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-142.27968,-0.63659637)"
+                                   id="g24810">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24812">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24818-5)"
+                                   id="g24814" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(64.67258,15.567675)"
+                                   id="g24822">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24824">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24830-7)"
+                                   id="g24826">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24832"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(0,-22.906203)"
+                                   id="g24834">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24836">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24842-3)"
+                                   id="g24838">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24844"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(8.6230106,4.4449087)"
+                                   id="g24846">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24848">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24854-4)"
+                                   id="g24850">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24856"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,-4.3324048)"
+                                   id="g24858">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24860">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24866-9)"
+                                   id="g24862">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24868"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,0.15594682)"
+                                   id="g24870">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24872">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24878-2)"
+                                   id="g24874">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24880"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(68.984085,-0.01126583)"
+                                   id="g24882">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24884">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24890-6)"
+                                   id="g24886">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24892"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-56.049569,0.01126583)"
+                                   id="g24894">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24896">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24902-1)"
+                                   id="g24898">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24904"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(172.46021,27.019966)"
+                                   id="g24906">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24908">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24914-8)"
+                                   id="g24910" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-120.72215,167.58723)"
+                                   id="g24918">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24920">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24926-9)"
+                                   id="g24922" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-34.492042,-185.8974)"
+                                   id="g24930">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24932">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24938-8)"
+                                   id="g24934">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24940"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(125.03365,1.9290799)"
+                                   id="g24942">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24944">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24950-8)"
+                                   id="g24946" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-25.869032,-9.6453995)"
+                                   id="g24954">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24956">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24962-8)"
+                                   id="g24958">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24964"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-103.47613,-1.0207919)"
+                                   id="g24966">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24968">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24974-8)"
+                                   id="g24970">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24976"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,25.327199)"
+                                   id="g24978">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24980">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24986-3)"
+                                   id="g24982">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path24988"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(68.984085,-25.569954)"
+                                   id="g24990">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g24992">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath24998-8)"
+                                   id="g24994">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25000"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(38.803548,26.815831)"
+                                   id="g25002">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25004">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25010-4)"
+                                   id="g25006">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25012"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(4.3115053,-19.572136)"
+                                   id="g25014">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25016">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25022-6)"
+                                   id="g25018">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25024"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-68.984085,-5.0156077)"
+                                   id="g25026">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25028">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25034-9)"
+                                   id="g25030">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25036"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(60.361074,5.2278065)"
+                                   id="g25038">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25040">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25046-6)"
+                                   id="g25042">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25048"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-112.09914,-7.4462484)"
+                                   id="g25050">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25052">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25058-7)"
+                                   id="g25054">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25060"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,0.15432639)"
+                                   id="g25062">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25064">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25070-0)"
+                                   id="g25066">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25072"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-30.180537,25.405982)"
+                                   id="g25074">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25076">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25082-3)"
+                                   id="g25078">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25084"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(142.27968,-18.11406)"
+                                   id="g25086">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25088">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25094-7)"
+                                   id="g25090">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25096"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-77.607096,-7.1970884)"
+                                   id="g25098">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25100">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25106-2)"
+                                   id="g25102">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25108"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(43.115053,78.491059)"
+                                   id="g25110">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25112">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25118-5)"
+                                   id="g25114">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25120"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-43.115053,-78.198456)"
+                                   id="g25122">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25124">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25130-6)"
+                                   id="g25126">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25132"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-38.803548,-0.29260284)"
+                                   id="g25134">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25136">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25142-8)"
+                                   id="g25138">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25144"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(120.72215,27.150218)"
+                                   id="g25146">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25148">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25154-9)"
+                                   id="g25150">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25156"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-185.39473,-24.31447)"
+                                   id="g25158">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25160">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25166-0)"
+                                   id="g25162" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(103.47613,4.1491422)"
+                                   id="g25170">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25172">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25178-1)"
+                                   id="g25174">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25180"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(64.67258,-7.0009396)"
+                                   id="g25182">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25184">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25190-4)"
+                                   id="g25186">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25192"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-137.96817,-0.04664515)"
+                                   id="g25194">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25196">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25202-7)"
+                                   id="g25198">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25204"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(60.361074,0.12218792)"
+                                   id="g25206">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25208">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25214-8)"
+                                   id="g25210">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25216"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-25.869032,-0.21381922)"
+                                   id="g25218">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25220">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25226-2)"
+                                   id="g25222">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25228"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,0.09807442)"
+                                   id="g25230">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25232">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25238-7)"
+                                   id="g25234">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25240"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,0.0675178)"
+                                   id="g25242">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25244">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25250-3)"
+                                   id="g25246">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25252"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(53.893816,8.5442035)"
+                                   id="g25254">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25256">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25262-2)"
+                                   id="g25258">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25264"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(66.828332,18.001556)"
+                                   id="g25266">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25268">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25274-3)"
+                                   id="g25270">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25276"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-77.607096,-21.501216)"
+                                   id="g25278">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25280">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25286-1)"
+                                   id="g25282">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25288"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,-5.1024163)"
+                                   id="g25290">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25292">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25298-8)"
+                                   id="g25294">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25300"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-140.46884,0.28615971)"
+                                   id="g25302">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25304">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25310-1)"
+                                   id="g25306" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(243.94497,1.4789484)"
+                                   id="g25314">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25316">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25322-4)"
+                                   id="g25318" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-129.34516,5.2663881)"
+                                   id="g25326">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25328">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25334-2)"
+                                   id="g25330">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25336"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,-5.0156077)"
+                                   id="g25338">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25340">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25346-7)"
+                                   id="g25342">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25348"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25350">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25352">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25358-9)"
+                                   id="g25354">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25360"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(66.828332,5.9801477)"
+                                   id="g25362">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25364">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25370-4)"
+                                   id="g25366">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25372"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-40.9593,-10.995755)"
+                                   id="g25374">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25376">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25382-9)"
+                                   id="g25378">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25384"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-77.607096,7.7983055)"
+                                   id="g25386">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25388">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25394-5)"
+                                   id="g25390">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25396"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(51.738064,33.26216)"
+                                   id="g25398">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25400">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25406-0)"
+                                   id="g25402">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25408"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-30.180537,-37.501313)"
+                                   id="g25410">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25412">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25418-1)"
+                                   id="g25414">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25420"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(4.3115053,0.10609939)"
+                                   id="g25422">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25424">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25430-9)"
+                                   id="g25426">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25432"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(34.492042,-0.60766017)"
+                                   id="g25434">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25436">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25442-8)"
+                                   id="g25438">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25444"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-4.3115053,18.933919)"
+                                   id="g25446">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25448">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25454-5)"
+                                   id="g25450">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25456"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-43.115053,-18.961235)"
+                                   id="g25458">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25460">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25466-4)"
+                                   id="g25462">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25468"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(189.70623,-0.04020203)"
+                                   id="g25470">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25472">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25478-0)"
+                                   id="g25474" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-129.34516,17.496755)"
+                                   id="g25482">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25484">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25490-0)"
+                                   id="g25486">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25492"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(23.713279,-17.496755)"
+                                   id="g25494">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25496">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25502-9)"
+                                   id="g25498">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25504"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-84.074354,-0.00644313)"
+                                   id="g25506">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25508">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25514-2)"
+                                   id="g25510">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25516"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-8.6230106,7.0475848)"
+                                   id="g25518">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25520">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25526-2)"
+                                   id="g25522">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25528"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(81.918601,17.471021)"
+                                   id="g25530">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25532">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25538-7)"
+                                   id="g25534">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25540"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-34.492042,-18.18478)"
+                                   id="g25542">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25544">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25550-1)"
+                                   id="g25546">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25552"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(17.246021,-3.7617058)"
+                                   id="g25554">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25556">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25562-9)"
+                                   id="g25558">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25564"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(28.024785,0.0385816)"
+                                   id="g25566">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25568">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25574-5)"
+                                   id="g25570">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25576"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-165.99295,2.5367401)"
+                                   id="g25578">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25580">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25586-7)"
+                                   id="g25582" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(146.59118,2.5753217)"
+                                   id="g25590">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25592">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25598-4)"
+                                   id="g25594">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25600"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-64.67258,-7.6986492)"
+                                   id="g25602">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25604">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25610-6)"
+                                   id="g25606">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25612"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(12.934516,-0.28774156)"
+                                   id="g25614">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25616">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25622-7)"
+                                   id="g25618">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25624"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(0,24.017045)"
+                                   id="g25626">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25628">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25634-8)"
+                                   id="g25630">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25636"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-12.934516,-23.939882)"
+                                   id="g25638">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25640">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25646-8)"
+                                   id="g25642">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25648"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(107.78763,27.75946)"
+                                   id="g25650">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25652">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25658-6)"
+                                   id="g25654">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25660"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-34.492042,28.651659)"
+                                   id="g25662">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25664">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25670-6)"
+                                   id="g25666">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25672"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-56.049569,-56.221413)"
+                                   id="g25674">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25676">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25682-4)"
+                                   id="g25678">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25684"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-51.738064,-0.19773069)"
+                                   id="g25686">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25688">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25694-2)"
+                                   id="g25690" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(56.049569,7.2420746)"
+                                   id="g25698">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25700">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25706-9)"
+                                   id="g25702">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25708"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(43.115053,20.390375)"
+                                   id="g25710">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25712">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25718-0)"
+                                   id="g25714">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25720"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-38.803548,-25.608536)"
+                                   id="g25722">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25724">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25730-0)"
+                                   id="g25726">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25732"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(4.3115053,5.4303599)"
+                                   id="g25734">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25736">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25742-0)"
+                                   id="g25738">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25744"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(30.180537,-0.21219879)"
+                                   id="g25746">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25748">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25754-3)"
+                                   id="g25750">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25756"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-43.115053,6.0107043)"
+                                   id="g25758">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25760">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25766-7)"
+                                   id="g25762">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25768"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(30.180537,5.7550626)"
+                                   id="g25770">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25772">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25778-6)"
+                                   id="g25774">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25780"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-112.09914,-9.0650551)"
+                                   id="g25782">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25784">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25790-5)"
+                                   id="g25786" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(73.29559,11.712717)"
+                                   id="g25794">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25796">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25802-0)"
+                                   id="g25798">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25804"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-17.246021,-21.338826)"
+                                   id="g25806">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25808">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25814-9)"
+                                   id="g25810">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25816"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(64.67258,9.4332007)"
+                                   id="g25818">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25820">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25826-9)"
+                                   id="g25822">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25828"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(25.869032,-1.8438917)"
+                                   id="g25830">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25832">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25838-4)"
+                                   id="g25834">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25840"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(133.65666,2.6750165)"
+                                   id="g25842">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25844">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25850-1)"
+                                   id="g25846" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-206.95225,-8.3545364)"
+                                   id="g25854">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25856">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25862-3)"
+                                   id="g25858">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25864"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(25.869032,0.86646553)"
+                                   id="g25866">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25868">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25874-8)"
+                                   id="g25870">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25876"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(64.67258,28.841365)"
+                                   id="g25878">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25880">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25886-6)"
+                                   id="g25882">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25888"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(25.869032,-30.67237)"
+                                   id="g25890">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25892">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25898-4)"
+                                   id="g25894">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25900"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-198.32924,0.24433726)"
+                                   id="g25902">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25904">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25910-7)"
+                                   id="g25906" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(107.78763,-0.24433726)"
+                                   id="g25914">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25916">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25922-0)"
+                                   id="g25918">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25924"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(86.230106,17.913127)"
+                                   id="g25926">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25928">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25934-7)"
+                                   id="g25930">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25936"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-107.78763,8.5924305)"
+                                   id="g25938">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25940">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25946-9)"
+                                   id="g25942">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25948"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-21.557527,-27.409023)"
+                                   id="g25950">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25952">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25958-8)"
+                                   id="g25954">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25960"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(51.738064,-0.15752866)"
+                                   id="g25962">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25964">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25970-3)"
+                                   id="g25966">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25972"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(8.6230106,7.7163196)"
+                                   id="g25974">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25976">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25982-8)"
+                                   id="g25978">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path25984"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(129.34516,-7.0073827)"
+                                   id="g25986">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g25988">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath25994-7)"
+                                   id="g25990" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-159.5257,1.8776506)"
+                                   id="g25998">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26000">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26006-3)"
+                                   id="g26002">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26008"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(155.21419,-1.5255935)"
+                                   id="g26010">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26012">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26018-8)"
+                                   id="g26014" />
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-155.21419,17.745915)"
+                                   id="g26022">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26024">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26030-4)"
+                                   id="g26026">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26032"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(56.049569,-21.796982)"
+                                   id="g26034">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26036">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26042-9)"
+                                   id="g26038">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26044"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-10.778763,10.031215)"
+                                   id="g26046">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26048">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26054-9)"
+                                   id="g26050">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26056"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-66.828332,-7.0314962)"
+                                   id="g26058">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26060">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26066-8)"
+                                   id="g26062">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26068"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(81.918601,56.208566)"
+                                   id="g26070">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26072">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26078-8)"
+                                   id="g26074">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26080"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-21.557527,-51.106149)"
+                                   id="g26082">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26084">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26090-3)"
+                                   id="g26086">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26092"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-64.67258,-4.3034686)"
+                                   id="g26094">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26096">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26102-1)"
+                                   id="g26098">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26104"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(86.230106,-0.14306057)"
+                                   id="g26106">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26108">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26114-8)"
+                                   id="g26110">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26116"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(-17.246021,11.82526)"
+                                   id="g26118">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26120">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26126-9)"
+                                   id="g26122">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26128"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(4.3115053,5.0059623)"
+                                   id="g26130">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26132">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26138-9)"
+                                   id="g26134">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26140"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   transform="translate(51.738064,22.29372)"
+                                   id="g26142">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   id="g26144">
+                                <g
+                                   style="fill:#ffca00;fill-opacity:1;stroke:none;stroke-width:1.13305926;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   clip-path="url(#clipPath26150-3)"
+                                   id="g26146">
+                                <path
+                                   inkscape:connector-curvature="0"
+                                   id="path26152"
+                                   style="fill:#ffca00;fill-opacity:1;fill-rule:nonzero;stroke:none;stroke-width:1.13305926;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:0.78431373"
+                                   d="m 0,-2.236068 c 0.593012,0 1.161816,0.235606 1.581139,0.654929 C 2.000462,-1.161816 2.236068,-0.593012 2.236068,0 c 0,0.593012 -0.235606,1.161816 -0.654929,1.581139 C 1.161816,2.000462 0.593012,2.236068 0,2.236068 c -0.593012,0 -1.161816,-0.235606 -1.581139,-0.654929 C -2.000462,1.161816 -2.236068,0.593012 -2.236068,0 c 0,-0.593012 0.235606,-1.161816 0.654929,-1.581139 C -1.161816,-2.000462 -0.593012,-2.236068 0,-2.236068 Z" />
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                                </g>
+                              </g>
+                            </g>
+                          </g>
+                        </g>
+                      </g>
+                    </g>
+                  </g>
+                </g>
+              </g>
+            </g>
+          </g>
+        </g>
+        <path
+           inkscape:connector-curvature="0"
+           id="path26248"
+           style="fill:#ffca00;fill-opacity:1;stroke:#505050;stroke-width:0.48219368;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           d="M 158.19812,120.30361 V 83.289427" />
+      </g>
+      <path
+         d="m 158.19812,120.30361 h 56.99205"
+         style="fill:none;stroke:#505050;stroke-width:0.48219368;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         id="path26252"
+         inkscape:connector-curvature="0" />
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g21112">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath21118)"
+           id="g21114">
+          <path
+             inkscape:connector-curvature="0"
+             id="path21120"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 125.145,111.90702 h 25.11" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g21122">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath21128)"
+           id="g21124">
+          <path
+             inkscape:connector-curvature="0"
+             id="path21130"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 292.545,116.87492 h 25.11" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20920">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20926)"
+           id="g20922">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20928"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 125.145,89.551474 h 25.11 v 47.195046 h -25.11 V 89.551474" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20930">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20936)"
+           id="g20932">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20938"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="M 137.7,89.551474 V 46.70334" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20940">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20946)"
+           id="g20942">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20948"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 137.7,136.74652 v 64.5827" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20950">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20956)"
+           id="g20952">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20958"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 131.4225,46.70334 h 12.555" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20960">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20966)"
+           id="g20962">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20968"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 131.4225,201.32922 h 12.555" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20970">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20976)"
+           id="g20972">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20978"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 292.545,97.003324 h 25.11 v 44.711096 h -25.11 V 97.003324" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20980">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20986)"
+           id="g20982">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20988"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="M 305.1,97.003324 V 45.883636" />
+        </g>
+      </g>
+      <g
+         style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g20990">
+        <g
+           style="stroke:#150458;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath20996)"
+           id="g20992">
+          <path
+             inkscape:connector-curvature="0"
+             id="path20998"
+             style="fill:none;stroke:#150458;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 305.1,141.71442 v 67.06665" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g21000">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath21006)"
+           id="g21002">
+          <path
+             inkscape:connector-curvature="0"
+             id="path21008"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 298.8225,45.883636 h 12.555" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g21010">
+        <g
+           style="stroke:#505050;stroke-width:3.39917612;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath21016)"
+           id="g21012">
+          <path
+             inkscape:connector-curvature="0"
+             id="path21018"
+             style="fill:none;stroke:#505050;stroke-width:3.39917612;stroke-linecap:square;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 298.8225,208.78107 h 12.555" />
+        </g>
+      </g>
+      <g
+         style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.17022713,0,0,-0.17022713,209.66847,126.43174)"
+         id="g21020">
+        <g
+           style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath21026)"
+           id="g21022">
+          <g
+             style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             transform="translate(305.1,221.20082)"
+             id="g21028">
+            <g
+               style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+               id="g21030">
+              <g
+                 style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                 clip-path="url(#clipPath21036)"
+                 id="g21032">
+                <path
+                   inkscape:connector-curvature="0"
+                   id="path21038"
+                   style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                   d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+              </g>
+            </g>
+            <g
+               style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+               transform="translate(0,-1.2419749)"
+               id="g21040">
+              <g
+                 style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                 id="g21042">
+                <g
+                   style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                   clip-path="url(#clipPath21048)"
+                   id="g21044">
+                  <path
+                     inkscape:connector-curvature="0"
+                     id="path21050"
+                     style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                     d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                </g>
+              </g>
+              <g
+                 style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                 transform="translate(0,1.2419749)"
+                 id="g21052">
+                <g
+                   style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                   id="g21054">
+                  <g
+                     style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                     clip-path="url(#clipPath21060)"
+                     id="g21056">
+                    <path
+                       inkscape:connector-curvature="0"
+                       id="path21062"
+                       style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                       d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                  </g>
+                </g>
+                <g
+                   style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                   transform="translate(0,22.355548)"
+                   id="g21064">
+                  <g
+                     style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                     id="g21066">
+                    <g
+                       style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                       clip-path="url(#clipPath21072)"
+                       id="g21068">
+                      <path
+                         inkscape:connector-curvature="0"
+                         id="path21074"
+                         style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                         d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                    </g>
+                  </g>
+                  <g
+                     style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                     transform="translate(0,-24.839498)"
+                     id="g21076">
+                    <g
+                       style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                       id="g21078">
+                      <g
+                         style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                         clip-path="url(#clipPath21084)"
+                         id="g21080">
+                        <path
+                           inkscape:connector-curvature="0"
+                           id="path21086"
+                           style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                           d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                      </g>
+                    </g>
+                    <g
+                       style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                       id="g21088">
+                      <g
+                         style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                         id="g21090">
+                        <g
+                           style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                           clip-path="url(#clipPath21096)"
+                           id="g21092">
+                          <path
+                             inkscape:connector-curvature="0"
+                             id="path21098"
+                             style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                             d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                        </g>
+                      </g>
+                      <g
+                         style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                         transform="translate(0,9.9357993)"
+                         id="g21100">
+                        <g
+                           style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                           id="g21102">
+                          <g
+                             style="stroke:#505050;stroke-width:1.13305879;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                             clip-path="url(#clipPath21108)"
+                             id="g21104">
+                            <path
+                               inkscape:connector-curvature="0"
+                               id="path21110"
+                               style="fill:none;stroke:#505050;stroke-width:1.13305879;stroke-linecap:butt;stroke-linejoin:round;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+                               d="m 0,-3 c 0.795609,0 1.55874,0.316099 2.12132,0.87868 C 2.683901,-1.55874 3,-0.795609 3,0 3,0.795609 2.683901,1.55874 2.12132,2.12132 1.55874,2.683901 0.795609,3 0,3 -0.795609,3 -1.55874,2.683901 -2.12132,2.12132 -2.683901,1.55874 -3,0.795609 -3,0 -3,-0.795609 -2.683901,-1.55874 -2.12132,-2.12132 -1.55874,-2.683901 -0.795609,-3 0,-3 Z" />
+                          </g>
+                        </g>
+                      </g>
+                    </g>
+                  </g>
+                </g>
+              </g>
+            </g>
+          </g>
+        </g>
+      </g>
+      <path
+         d="M 218.86073,120.30357 V 83.289378"
+         style="fill:none;stroke:#505050;stroke-width:0.48219332;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         id="path21134"
+         inkscape:connector-curvature="0" />
+      <path
+         d="m 218.86073,120.30357 h 56.99204"
+         style="fill:none;stroke:#505050;stroke-width:0.48240176;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         id="path21138"
+         inkscape:connector-curvature="0" />
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23785">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23791)"
+           id="g23787">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23793"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="M 69.218182,36 H 99.654545 V 99.178692 H 69.218182 Z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23795">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23801)"
+           id="g23797">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23803"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="M 99.654545,36 H 130.09091 V 89.818886 H 99.654545 Z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23805">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23811)"
+           id="g23807">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23813"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 130.09091,36 h 30.43636 v 207.08571 h -30.43636 z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23815">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23821)"
+           id="g23817">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23823"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 160.52727,36 h 30.43637 v 197.72591 h -30.43637 z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23825">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23831)"
+           id="g23827">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23833"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="M 190.96364,36 H 221.4 v 138.05714 h -30.43636 z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23835">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23841)"
+           id="g23837">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23843"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 221.4,36 h 30.43636 v 81.8983 H 221.4 Z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23845">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23851)"
+           id="g23847">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23853"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 251.83636,36 h 30.43637 v 52.64891 h -30.43637 z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23855">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23861)"
+           id="g23857">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23863"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 282.27273,36 h 30.43636 v 28.079419 h -30.43636 z" />
+        </g>
+      </g>
+      <g
+         style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         transform="matrix(0.1702344,0,0,-0.1702344,149.00467,168.43373)"
+         id="g23865">
+        <g
+           style="fill:#150458;fill-opacity:1;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           clip-path="url(#clipPath23871)"
+           id="g23867">
+          <path
+             inkscape:connector-curvature="0"
+             id="path23873"
+             style="fill:#150458;fill-opacity:1;fill-rule:nonzero;stroke:#ffffd8;stroke-width:1.57538962;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+             d="m 312.70909,36 h 30.43637 v 10.529782 h -30.43637 z" />
+        </g>
+      </g>
+      <path
+         d="M 158.19733,162.30529 V 125.28952"
+         style="fill:none;stroke:#505050;stroke-width:0.48043513;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         id="path23993"
+         inkscape:connector-curvature="0" />
+      <path
+         d="m 158.19733,162.30529 h 56.99448"
+         style="fill:none;stroke:#505050;stroke-width:0.48043513;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+         id="path23997"
+         inkscape:connector-curvature="0" />
+      <g
+         transform="matrix(0.91342101,0,0,0.91342101,-19.440724,14.747995)"
+         id="g44700-2"
+         style="stroke-width:1.11152482">
+        <path
+           inkscape:connector-curvature="0"
+           id="path21134-8-9-1"
+           style="fill:none;stroke:#505050;stroke-width:0.48219332;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           d="M 260.86839,161.56318 V 124.54899" />
+        <path
+           inkscape:connector-curvature="0"
+           id="path21138-1-3-0"
+           style="fill:none;stroke:#505050;stroke-width:0.48240176;stroke-linecap:square;stroke-linejoin:miter;stroke-miterlimit:10;stroke-dasharray:none;stroke-opacity:1"
+           d="m 260.86839,161.56318 h 56.99204" />
+      </g>
+      <text
+         id="text47247-6-7-5"
+         y="146.26187"
+         x="234.65596"
+         style="font-style:normal;font-weight:normal;font-size:16.98972702px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#505050;fill-opacity:1;stroke:none;stroke-width:0.4721126"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:11.32648563px;font-family:monospace;-inkscape-font-specification:monospace;fill:#505050;fill-opacity:1;stroke-width:0.4721126"
+           y="146.26187"
+           x="234.65596"
+           id="tspan47245-1-51-1"
+           sodipodi:role="line">...</tspan></text>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/05_newcolumn_1.svg b/pandas/doc/source/_static/schemas/05_newcolumn_1.svg
new file mode 100644
index 0000000000000..c158aa932d38e
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/05_newcolumn_1.svg
@@ -0,0 +1,347 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="324.44537mm"
+   height="77.726311mm"
+   viewBox="0 0 324.44537 77.726311"
+   version="1.1"
+   id="svg9265"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="05_newcolumn_1.svg">
+  <defs
+     id="defs9259">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="662.36942"
+     inkscape:cy="1.4822257"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata9262">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(113.44795,-29.083275)">
+    <g
+       id="g938"
+       transform="matrix(0.89983835,0,0,0.89933325,4.8853597,6.8399466)"
+       style="stroke-width:1.11162269">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,55.75353 h 24.88795 V 43.56556 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-73-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-72-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23746,68.45353 h 24.88796 V 56.26557 H 83.23746 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-13-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-4-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-94-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,81.15354 h 24.88795 V 68.96558 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-74-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-22-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,93.85355 h 24.88795 V 81.66559 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-62-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-0-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,106.55356 h 24.88795 V 94.3656 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-2-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-9-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,81.15354 h 24.88795 V 68.96558 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-7-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-4-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,55.75353 h 24.88795 V 43.56556 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-56-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,93.85355 h 24.88795 V 81.66559 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,106.55356 h 24.88795 V 94.3656 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-4-7-4-9-1"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,81.15354 H 210.7414 V 68.96558 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-03-2-2"
+         style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,42.03751 H 210.7414 V 29.84956 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-1-4-8"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,55.75353 H 210.7414 V 43.56556 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-2-4-9"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,68.45353 H 210.7414 V 56.26557 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-6-6-7-5-6"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,93.85355 H 210.7414 V 81.66559 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-2-3-7-9-0"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,106.55356 H 210.7414 V 94.3656 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-9-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,55.24327 h 24.88795 V 43.0553 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-73-8-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,41.52725 h 24.88795 V 29.3393 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-2-1-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,55.24327 h 24.88795 V 43.0553 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-72-0-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19193,67.94327 h 24.88796 V 55.75531 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-13-7-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,67.94327 h 24.88795 V 55.75531 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-4-6-7"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,41.52725 h 24.88796 V 29.3393 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-94-0-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,55.24327 h 24.88796 V 43.0553 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-1-2"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,67.94327 h 24.88796 V 55.75531 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-0-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,80.64328 h 24.88795 V 68.45532 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-74-2-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,80.64328 h 24.88795 V 68.45532 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-5-2"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,80.64328 h 24.88796 V 68.45532 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-22-1-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,93.34329 h 24.88795 V 81.15533 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-1-9-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,93.34329 h 24.88795 V 81.15533 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-62-7-7"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,93.34329 h 24.88796 V 81.15533 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-0-1-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,106.0433 h 24.88795 V 93.85534 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-2-3-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,106.0433 h 24.88795 V 93.85534 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-2-1-1"
+         style="fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,106.0433 h 24.88796 V 93.85534 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-9-2-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,80.64328 h 24.88794 V 68.45532 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-7-0-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97592,41.52725 h 24.88795 V 29.3393 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-4-6-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,55.24327 h 24.88794 V 43.0553 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-8-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97592,67.94327 h 24.88795 V 55.75531 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-56-7-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,93.34329 h 24.88794 V 81.15533 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920481;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,106.0433 h 24.88794 V 93.85534 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9"
+         d="M 14.43931,73.73504 H 62.32889"
+         style="fill:none;stroke:#000000;stroke-width:0.4446491;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/05_newcolumn_2.svg b/pandas/doc/source/_static/schemas/05_newcolumn_2.svg
new file mode 100644
index 0000000000000..8bd5ad9a26994
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/05_newcolumn_2.svg
@@ -0,0 +1,347 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="324.44537mm"
+   height="77.726311mm"
+   viewBox="0 0 324.44537 77.726311"
+   version="1.1"
+   id="svg9265"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="05_newcolumn_2.svg">
+  <defs
+     id="defs9259">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="419.63138"
+     inkscape:cy="230.09877"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata9262">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(113.44795,-29.083275)">
+    <g
+       id="g938"
+       transform="matrix(0.89984193,0,0,0.89933685,4.8851851,6.839702)"
+       style="stroke-width:1.11161828">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,55.75353 h 24.88795 V 43.56556 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-73-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-72-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23746,68.45353 h 24.88796 V 56.26557 H 83.23746 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-13-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-4-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-94-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,81.15354 h 24.88795 V 68.96558 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-74-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-22-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,93.85355 h 24.88795 V 81.66559 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-62-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-0-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,106.55356 h 24.88795 V 94.3656 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-2-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-9-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,81.15354 h 24.88795 V 68.96558 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-7-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-4-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,55.75353 h 24.88795 V 43.56556 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-56-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,93.85355 h 24.88795 V 81.66559 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,106.55356 h 24.88795 V 94.3656 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-4-7-4-9-1"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,81.15354 H 210.7414 V 68.96558 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-03-2-2"
+         style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,42.03751 H 210.7414 V 29.84956 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-1-4-8"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,55.75353 H 210.7414 V 43.56556 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-2-4-9"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,68.45353 H 210.7414 V 56.26557 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-6-6-7-5-6"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,93.85355 H 210.7414 V 81.66559 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-2-3-7-9-0"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,106.55356 H 210.7414 V 94.3656 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-9-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,55.24327 h 24.88795 V 43.0553 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-73-8-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,41.52725 h 24.88795 V 29.3393 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-2-1-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,55.24327 h 24.88795 V 43.0553 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-72-0-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19193,67.94327 h 24.88796 V 55.75531 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-13-7-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,67.94327 h 24.88795 V 55.75531 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-4-6-7"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,41.52725 h 24.88796 V 29.3393 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-94-0-7"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,55.24327 h 24.88796 V 43.0553 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-1-2"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,67.94327 h 24.88796 V 55.75531 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-0-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,80.64328 h 24.88795 V 68.45532 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-74-2-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,80.64328 h 24.88795 V 68.45532 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-5-2"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,80.64328 h 24.88796 V 68.45532 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-22-1-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,93.34329 h 24.88795 V 81.15533 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-1-9-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,93.34329 h 24.88795 V 81.15533 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-62-7-7"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,93.34329 h 24.88796 V 81.15533 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-0-1-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.19192,106.0433 h 24.88795 V 93.85534 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-2-3-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.77592,106.0433 h 24.88795 V 93.85534 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-2-1-1"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.37592,106.0433 h 24.88796 V 93.85534 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-9-2-3"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,80.64328 h 24.88794 V 68.45532 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-7-0-6"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97592,41.52725 h 24.88795 V 29.3393 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-4-6-4"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,55.24327 h 24.88794 V 43.0553 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-8-0"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97592,67.94327 h 24.88795 V 55.75531 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-56-7-3"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,93.34329 h 24.88794 V 81.15533 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4-5"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -35.97591,106.0433 h 24.88794 V 93.85534 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9"
+         d="M 14.43931,73.73504 H 62.32889"
+         style="fill:none;stroke:#000000;stroke-width:0.44464734;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/05_newcolumn_3.svg b/pandas/doc/source/_static/schemas/05_newcolumn_3.svg
new file mode 100644
index 0000000000000..45272d8c9a368
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/05_newcolumn_3.svg
@@ -0,0 +1,352 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="324.44537mm"
+   height="77.726311mm"
+   viewBox="0 0 324.44537 77.726311"
+   version="1.1"
+   id="svg9265"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="05_newcolumn_3.svg">
+  <defs
+     id="defs9259">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="398.22496"
+     inkscape:cy="-188.87908"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata9262">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(113.44795,-29.083275)">
+    <g
+       id="g940"
+       transform="matrix(0.89984193,0,0,0.89933685,4.8851851,6.839702)"
+       style="stroke-width:1.11161828">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-5-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,55.75353 h 24.88795 V 43.56556 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-73-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-72-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23746,68.45353 h 24.88796 V 56.26557 H 83.23746 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-13-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-4-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-94-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,55.75353 h 24.88796 V 43.56556 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,81.15354 h 24.88795 V 68.96558 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-74-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,81.15354 h 24.88796 V 68.96558 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-22-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,93.85355 h 24.88795 V 81.66559 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-62-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,93.85355 h 24.88796 V 81.66559 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-0-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 83.23747,106.55356 h 24.88795 V 94.3656 H 83.23747 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-2-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 109.65347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-2-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 135.05347,106.55356 h 24.88796 V 94.3656 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-9-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,81.15354 h 24.88795 V 68.96558 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-7-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,42.03751 h 24.88796 V 29.84956 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-4-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,55.75353 h 24.88795 V 43.56556 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45347,68.45353 h 24.88796 V 56.26557 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-56-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,93.85355 h 24.88795 V 81.66559 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 160.45348,106.55356 h 24.88795 V 94.3656 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-4-7-4-9-1"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,81.15354 H 210.7414 V 68.96558 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-03-2-2"
+         style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,42.03751 H 210.7414 V 29.84956 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-1-4-8"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,55.75353 H 210.7414 V 43.56556 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-2-4-9"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,68.45353 H 210.7414 V 56.26557 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-6-6-7-5-6"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,93.85355 H 210.7414 V 81.66559 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-2-3-7-9-0"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 185.85348,106.55356 H 210.7414 V 94.3656 h -24.88792 z" />
+      <g
+         transform="translate(200.11428,-607.97617)"
+         id="g50992"
+         style="stroke-width:1.11161828">
+        <path
+           d="m -313.3062,663.21944 h 24.88795 v -12.18797 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-9-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,649.50342 h 24.88795 v -12.18795 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-73-8-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,663.21944 h 24.88795 v -12.18797 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-2-1-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -313.30621,675.91944 h 24.88796 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-72-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,675.91944 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-13-7-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,649.50342 h 24.88796 v -12.18795 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-4-6-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,663.21944 h 24.88796 v -12.18797 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-94-0-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,675.91944 h 24.88796 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-1-1-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -313.3062,688.61945 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-95-0-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,688.61945 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-74-2-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,688.61945 h 24.88796 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-1-5-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -313.3062,701.31946 h 24.88795 V 689.1315 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-6-22-1-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,701.31946 h 24.88795 V 689.1315 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-6-1-9-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,701.31946 h 24.88796 V 689.1315 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-21-62-7-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -313.3062,714.01947 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-2-96-0-1-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -286.8902,714.01947 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-7-4-2-3-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -261.4902,714.01947 h 24.88796 v -12.18796 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-2-8-2-1-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.09019,688.61945 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-9-2-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.0902,649.50342 h 24.88795 v -12.18795 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-7-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.09019,663.21944 h 24.88794 v -12.18797 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-4-6-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.0902,675.91944 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3-8-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.09019,701.31946 h 24.88794 V 689.1315 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-2-56-7-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -236.09019,714.01947 h 24.88794 v -12.18796 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56920254;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-1-29-1-4-5"
+           inkscape:connector-curvature="0" />
+      </g>
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9"
+         d="M 14.43931,73.73504 H 62.32889"
+         style="fill:none;stroke:#000000;stroke-width:0.44464734;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4)" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_aggregate.svg b/pandas/doc/source/_static/schemas/06_aggregate.svg
new file mode 100644
index 0000000000000..14428feda44ec
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_aggregate.svg
@@ -0,0 +1,211 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="252.24091mm"
+   height="77.216049mm"
+   viewBox="0 0 252.2409 77.216049"
+   version="1.1"
+   id="svg11151"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_aggregate.svg">
+  <defs
+     id="defs11145">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="248.04006"
+     inkscape:cy="-37.739686"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata11148">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-3.5465457,-106.44555)">
+    <g
+       id="g910"
+       transform="matrix(0.89979659,0,0,0.89933244,12.993074,14.602189)"
+       style="stroke-width:1.11164904">
+      <g
+         id="g882"
+         style="stroke-width:1.11164904">
+        <path
+           d="m 230.64348,151.14755 h 24.88795 v -12.18796 h -24.88795 z"
+           style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-6-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 3.80258,132.60554 H 28.69052 V 120.41757 H 3.80258 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,118.88952 H 55.10652 V 106.70157 H 30.21857 Z"
+           style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-40"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,132.60554 H 55.10652 V 120.41757 H 30.21857 Z"
+           style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-94"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 3.80257,145.30554 H 28.69052 V 133.11758 H 3.80257 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,145.30554 H 55.10652 V 133.11758 H 30.21857 Z"
+           style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,118.88952 H 80.50653 V 106.70157 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,132.60554 H 80.50653 V 120.41757 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,145.30554 H 80.50653 V 133.11758 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-39"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 3.80258,158.00555 H 28.69052 V 145.81759 H 3.80258 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-95"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,158.00555 H 55.10652 V 145.81759 H 30.21857 Z"
+           style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,158.00555 H 80.50653 V 145.81759 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 3.80258,170.70556 H 28.69052 V 158.5176 H 3.80258 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-6-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,170.70556 H 55.10652 V 158.5176 H 30.21857 Z"
+           style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-6-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,170.70556 H 80.50653 V 158.5176 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-21-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 3.80258,183.40557 H 28.69052 V 171.21761 H 3.80258 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-2-96-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 30.21857,183.40557 H 55.10652 V 171.21761 H 30.21857 Z"
+           style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-7-4-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 55.61857,183.40557 H 80.50653 V 171.21761 H 55.61857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-2-8-34"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01858,158.00555 h 24.88794 V 145.81759 H 81.01858 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-58"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01857,118.88952 h 24.88795 V 106.70157 H 81.01857 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01858,132.60554 h 24.88794 V 120.41757 H 81.01858 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01857,145.30554 h 24.88795 V 133.11758 H 81.01857 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01858,170.70556 h 24.88794 V 158.5176 H 81.01858 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-2-89"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 81.01858,183.40557 h 24.88794 V 171.21761 H 81.01858 Z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921828;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-1-29-55"
+           inkscape:connector-curvature="0" />
+        <path
+           style="fill:none;stroke:#000000;stroke-width:0.44465962;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2)"
+           d="m 149.73091,145.06062 h 47.88958"
+           id="path6109-2-9-6-9-0"
+           inkscape:connector-curvature="0" />
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_groupby.svg b/pandas/doc/source/_static/schemas/06_groupby.svg
new file mode 100644
index 0000000000000..ca4d32be7084b
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_groupby.svg
@@ -0,0 +1,307 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="464.94458mm"
+   height="165.19176mm"
+   viewBox="0 0 464.94458 165.19176"
+   version="1.1"
+   id="svg14732"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_groupby.svg">
+  <defs
+     id="defs14726">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-86"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1-7"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8-3"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="808.73401"
+     inkscape:cy="127.43429"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata14729">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(120.59136,-57.166026)">
+    <path
+       d="m -107.12734,127.31387 h 24.887965 V 115.1259 h -24.887965 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-94-33-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -107.12734,140.01387 h 24.887965 v -12.18796 h -24.887965 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-4-9-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,113.59785 h 24.88796 V 101.4099 h -24.88796 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-6-4-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,127.31387 h 24.88796 V 115.1259 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-6-9-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,140.01387 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-39-9-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -107.12734,152.71388 h 24.887965 v -12.18796 h -24.887965 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-8-2-9-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,152.71388 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-51-1-4-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -107.12734,165.41389 h 24.887965 v -12.18796 h -24.887965 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-0-6-0-0-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,165.41389 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-5-21-9-6-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -107.12734,178.1139 h 24.887965 v -12.18796 h -24.887965 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-0-7-4-1-9-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -81.727335,178.1139 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-5-2-8-34-5-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327325,152.71388 h 24.88795 v -12.18796 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-0-58-6-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327335,113.59785 h 24.88796 V 101.4099 h -24.88796 z"
+       style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-5-1-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327325,127.31387 h 24.88795 V 115.1259 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327335,140.01387 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-3-4-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327325,165.41389 h 24.88795 v -12.18796 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-7-2-89-3-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -56.327325,178.1139 h 24.88795 v -12.18796 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-22"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 279.58525,146.36389 h 24.88794 v -12.18797 h -24.88794 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-6-3-4-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 306.00124,132.64787 H 330.8892 V 120.45992 H 306.00124 Z"
+       style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-17-40-0-4-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 306.00124,146.36389 H 330.8892 V 134.17592 H 306.00124 Z"
+       style="fill:#e50387;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-94-33-4-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 279.58524,159.06389 h 24.88795 v -12.18796 h -24.88795 z"
+       style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-5-7-9-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 306.00124,159.06389 H 330.8892 V 146.87593 H 306.00124 Z"
+       style="fill:#e50387;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-4-9-9-9"
+       inkscape:connector-curvature="0" />
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-3)"
+       d="m 203.45015,139.76895 h 47.88958"
+       id="path6109-2-9-6-9-0-9-4"
+       inkscape:connector-curvature="0" />
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-1-7)"
+       d="M -3.8625116,139.76895 H 44.027068"
+       id="path6109-2-9-6-9-0-9-7-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 86.228953,184.50217 h 24.887957 v -12.1879 H 86.228953 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-94-33-49-2-90"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 86.228953,197.20217 h 24.887957 v -12.1879 H 86.228953 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-4-9-5-7-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62895,170.78617 h 24.88796 v -12.1879 h -24.88796 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-6-4-7-9-85"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62895,184.50217 h 24.88796 v -12.1879 h -24.88796 z"
+       style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-6-9-3-5-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62895,197.20217 h 24.88796 v -12.1879 h -24.88796 z"
+       style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-39-9-9-6-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02895,170.78617 h 24.88796 v -12.1879 h -24.88796 z"
+       style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-5-1-1-21-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02896,184.50217 h 24.88795 v -12.1879 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-0-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02895,197.20217 h 24.88796 v -12.1879 h -24.88796 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-3-4-9-5-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 86.228963,108.22561 H 111.11692 V 96.037639 H 86.228963 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-94-33-8-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 86.228963,120.92561 H 111.11692 V 108.73765 H 86.228963 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-4-9-6-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62896,94.509589 h 24.88796 v -12.18795 h -24.88796 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-6-4-5-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62896,108.22561 h 24.88796 V 96.037639 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-6-9-2-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62896,120.92561 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-39-9-4-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 86.228963,133.62562 H 111.11692 V 121.43766 H 86.228963 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-8-2-9-0-1"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 111.62896,133.62562 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-51-1-4-6-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02897,133.62562 h 24.88795 v -12.18796 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-0-58-6-8-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02896,94.509589 h 24.88796 v -12.18795 h -24.88796 z"
+       style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-5-1-3-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02897,108.22561 h 24.88795 V 96.037639 h -24.88795 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-2-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 137.02896,120.92561 h 24.88796 v -12.18796 h -24.88796 z"
+       style="fill:#ffc900;fill-opacity:0.39215687;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-3-4-3-3"
+       inkscape:connector-curvature="0" />
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_groupby_agg_detail.svg b/pandas/doc/source/_static/schemas/06_groupby_agg_detail.svg
new file mode 100644
index 0000000000000..23a78d3ed2a9e
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_groupby_agg_detail.svg
@@ -0,0 +1,619 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="636.6601mm"
+   height="189.74704mm"
+   viewBox="0 0 636.6601 189.74704"
+   version="1.1"
+   id="svg17926"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_groupby_agg_detail.svg">
+  <defs
+     id="defs17920">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-12"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1-5"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8-0"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1-5-6"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8-0-3"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.24748737"
+     inkscape:cx="1089.6106"
+     inkscape:cy="-100.5864"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     inkscape:window-width="1875"
+     inkscape:window-height="1029"
+     inkscape:window-x="45"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0" />
+  <metadata
+     id="metadata17923">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(335.37904,-157.22519)">
+    <g
+       style="stroke-width:1.1112442"
+       id="g1993"
+       transform="matrix(0.89991951,0,0,0.89986489,18.49749,25.231115)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -357.57363,270.5441 h 24.88794 v -12.18799 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,256.82806 h 24.88797 v -12.18795 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,270.5441 h 24.88797 v -12.18799 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -357.57364,283.2441 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,283.2441 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,256.82806 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,270.5441 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,283.2441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -357.57363,295.9441 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,295.9441 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,295.9441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-3-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -357.57363,308.6441 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-0-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,308.6441 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-9-6"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,308.6441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-9-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -357.57363,321.3441 h 24.88794 v -12.1879 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -331.15764,321.3441 h 24.88797 v -12.1879 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-34-5"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.75763,321.3441 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,295.9441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,256.82806 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,270.5441 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,283.2441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-89-3"
+         style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,308.6441 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -280.35763,321.3441 h 24.88796 v -12.1879 h -24.88796 z" />
+    </g>
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3)"
+       d="m 135.58895,279.89212 h 43.09677"
+       id="path6109-2-9-6-9-0-9"
+       inkscape:connector-curvature="0" />
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-1)"
+       d="m -190.52748,279.89212 h 43.09677"
+       id="path6109-2-9-6-9-0-9-7"
+       inkscape:connector-curvature="0" />
+    <g
+       aria-label=" titanic      .groupby(&quot;Sex&quot;)   .mean()"
+       transform="scale(1.0000303,0.99996965)"
+       style="font-style:normal;font-weight:normal;font-size:18.18744659px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.50526738"
+       id="text56748-3">
+      <path
+         d="m -279.70486,171.2281 h -3.49199 v -1.80056 c 0,-0.873 -0.0728,-1.23675 -0.70931,-1.23675 -0.47287,0 -0.69112,0.25463 -0.69112,0.74569 0,0.23644 0,0.40012 0,0.47287 v 1.81875 h -0.98212 c -0.85481,0 -1.21856,0.0546 -1.21856,0.63656 0,0.4365 0.21825,0.61837 0.70931,0.61837 h 1.49137 v 2.96456 c 0,0.0909 0,0.18187 0,0.27281 0,0.81843 0.0182,1.52774 0.32738,2.05518 0.50924,0.89118 1.45499,1.32768 2.81905,1.32768 0.83662,0 1.76418,-0.23643 2.87362,-0.67293 0.65474,-0.25463 0.98212,-0.50925 0.98212,-0.92756 0,-0.32738 -0.291,-0.63656 -0.61838,-0.63656 -0.70931,0 -1.89149,0.94574 -3.3283,0.94574 -1.56412,0 -1.67324,-0.94574 -1.67324,-2.36436 v -2.96456 h 3.60111 c 0.49106,0 0.74569,-0.20006 0.74569,-0.61837 0,-0.45469 -0.27281,-0.63656 -0.83663,-0.63656 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1031" />
+      <path
+         d="m -270.36447,177.68464 v -6.58385 c 0,-0.21825 -0.16368,-0.36375 -0.50924,-0.36375 h -2.60081 c -0.47287,0 -0.70931,0.20006 -0.70931,0.61837 0,0.41831 0.25463,0.63656 0.74569,0.63656 h 1.7278 v 5.69267 h -2.70993 c -0.49106,0 -0.74568,0.20006 -0.74568,0.61837 0,0.41832 0.23644,0.63657 0.70931,0.63657 h 6.89304 c 0.52744,0 0.7275,-0.12732 0.7275,-0.63657 0,-0.43649 -0.21825,-0.61837 -0.67294,-0.61837 z m -1.65505,-9.51203 c 0,0.72749 0.18187,0.87299 0.92756,0.87299 0.85481,0 0.90937,-0.20006 0.90937,-1.12762 0,-0.92756 -0.0909,-1.20037 -0.90937,-1.20037 -0.81844,0 -0.92756,0.27281 -0.92756,1.455 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1033" />
+      <path
+         d="m -257.82309,171.2281 h -3.49199 v -1.80056 c 0,-0.873 -0.0727,-1.23675 -0.70931,-1.23675 -0.47287,0 -0.69112,0.25463 -0.69112,0.74569 0,0.23644 0,0.40012 0,0.47287 v 1.81875 h -0.98212 c -0.85481,0 -1.21856,0.0546 -1.21856,0.63656 0,0.4365 0.21825,0.61837 0.70931,0.61837 h 1.49137 v 2.96456 c 0,0.0909 0,0.18187 0,0.27281 0,0.81843 0.0182,1.52774 0.32737,2.05518 0.50925,0.89118 1.455,1.32768 2.81906,1.32768 0.83662,0 1.76418,-0.23643 2.87362,-0.67293 0.65474,-0.25463 0.98212,-0.50925 0.98212,-0.92756 0,-0.32738 -0.291,-0.63656 -0.61838,-0.63656 -0.70931,0 -1.89149,0.94574 -3.3283,0.94574 -1.56412,0 -1.67324,-0.94574 -1.67324,-2.36436 v -2.96456 h 3.60111 c 0.49106,0 0.74569,-0.20006 0.74569,-0.61837 0,-0.45469 -0.27282,-0.63656 -0.83663,-0.63656 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1035" />
+      <path
+         d="m -247.28232,177.81195 0.0909,0.54563 c 0.0728,0.38193 0.291,0.582 0.61838,0.582 h 1.40043 c 0.47287,0 0.70931,-0.21825 0.70931,-0.63657 0,-0.52743 -0.30919,-0.61837 -1.05487,-0.61837 h -0.45469 v -4.01942 c 0,-2.14612 -1.05487,-3.14643 -3.3283,-3.14643 -2.00062,0 -3.31011,0.63656 -3.31011,1.25493 0,0.45469 0.27281,0.74569 0.60018,0.74569 0.69112,0 1.41862,-0.74569 2.67356,-0.74569 1.36405,0 2.00061,0.63656 2.00061,2.00062 0,0.0182 0,0.0364 0,0.0546 -0.72749,-0.18187 -1.41862,-0.27281 -2.10974,-0.27281 -2.63718,0 -4.11036,1.10944 -4.11036,2.94637 0,1.56412 1.16399,2.65536 2.94636,2.65536 1.21856,0 2.36437,-0.47287 3.32831,-1.34587 z m -0.0546,-2.80086 v 0.96393 c 0,0.80025 -1.45499,1.81875 -3.05549,1.81875 -1.03668,0 -1.74599,-0.60019 -1.74599,-1.38225 0,-1.00031 1.03668,-1.70962 2.92818,-1.70962 0.65475,0 1.27312,0.10912 1.8733,0.30919 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1037" />
+      <path
+         d="m -241.67036,177.68464 c -0.0728,0 -0.12731,0 -0.18187,0 -0.7275,0 -1.03669,0.10913 -1.03669,0.61837 0,0.41832 0.23644,0.63657 0.74569,0.63657 0.23644,0 0.40012,0 0.47287,0 h 1.61869 c 0.67293,0 0.94574,-0.0364 0.94574,-0.63657 0,-0.49106 -0.23643,-0.63656 -0.85481,-0.63656 -0.10912,0 -0.23643,0.0182 -0.38193,0.0182 v -3.69205 c 0,-1.32768 1.23674,-2.12793 2.3098,-2.12793 1.164,0 1.81875,0.70931 1.81875,2.12793 v 3.69205 c -0.10913,0 -0.20007,0 -0.291,0 -0.63656,0 -0.89119,0.12731 -0.89119,0.61837 0,0.582 0.25463,0.63657 0.89119,0.63657 h 1.96424 c 0.63656,0 0.94575,-0.12732 0.94575,-0.63657 0,-0.45468 -0.27281,-0.61837 -0.83662,-0.61837 -0.12732,0 -0.27282,0 -0.40013,0 v -4.01942 c 0,-2.05519 -1.20037,-3.16462 -2.8918,-3.16462 -0.92756,0 -1.746,0.40012 -2.60081,1.20037 v -0.60018 c 0,-0.21825 -0.16368,-0.36375 -0.50924,-0.36375 h -0.83663 c -0.81843,0 -1.21856,0 -1.21856,0.61837 0,0.45469 0.25463,0.63656 0.81844,0.63656 0.12731,0 0.27281,0 0.40012,0 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1039" />
+      <path
+         d="m -226.60093,177.68464 v -6.58385 c 0,-0.21825 -0.16368,-0.36375 -0.50925,-0.36375 h -2.6008 c -0.47287,0 -0.70931,0.20006 -0.70931,0.61837 0,0.41831 0.25462,0.63656 0.74568,0.63656 h 1.72781 v 5.69267 h -2.70993 c -0.49106,0 -0.74568,0.20006 -0.74568,0.61837 0,0.41832 0.23643,0.63657 0.70931,0.63657 h 6.89304 c 0.52744,0 0.7275,-0.12732 0.7275,-0.63657 0,-0.43649 -0.21825,-0.61837 -0.67294,-0.61837 z m -1.65505,-9.51203 c 0,0.72749 0.18187,0.87299 0.92756,0.87299 0.85481,0 0.90937,-0.20006 0.90937,-1.12762 0,-0.92756 -0.0909,-1.20037 -0.90937,-1.20037 -0.81844,0 -0.92756,0.27281 -0.92756,1.455 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1041" />
+      <path
+         d="m -212.47724,173.41059 v -1.65506 c 0,-0.76387 -0.0909,-1.09124 -0.63656,-1.09124 -0.45468,0 -0.69112,0.16368 -0.69112,0.56381 0,0 0,0.0182 0,0.0364 -0.98212,-0.49106 -1.87331,-0.74568 -2.80087,-0.74568 -2.54624,0 -4.47411,1.8733 -4.47411,4.32861 0,2.45531 1.90968,4.31042 4.4923,4.31042 2.14612,0 4.14674,-1.32768 4.14674,-2.1643 0,-0.36375 -0.25463,-0.61837 -0.61837,-0.61837 -0.47288,0 -0.81844,0.43649 -1.36406,0.80024 -0.63656,0.41831 -1.36406,0.65475 -2.07337,0.65475 -1.85512,0 -3.03731,-1.23675 -3.03731,-3.00093 0,-1.70962 1.27313,-2.92818 3.11006,-2.92818 1.05487,0 1.89149,0.49106 2.43712,1.455 0.27281,0.49106 0.45468,0.70931 0.83662,0.70931 0.4365,0 0.67293,-0.21825 0.67293,-0.65475 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1043" />
+      <path
+         d="m -141.63827,177.48458 c 0,0.90937 0.69112,1.58231 1.74599,1.58231 1.05487,0 1.746,-0.65475 1.746,-1.58231 0,-0.92756 -0.69113,-1.58231 -1.746,-1.58231 -1.05487,0 -1.74599,0.67294 -1.74599,1.58231 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1045" />
+      <path
+         d="m -125.35028,178.4667 v -6.47473 c 0.0546,0 0.10913,0 0.16369,0 0.74569,0 1.05487,-0.12731 1.05487,-0.63656 0,-0.41831 -0.23643,-0.61837 -0.70931,-0.61837 h -1.34587 c -0.34556,0 -0.50925,0.10912 -0.50925,0.36375 v 0.63656 c -0.85481,-0.67294 -1.70962,-1.0185 -2.61899,-1.0185 -2.14612,0 -3.9103,1.61868 -3.9103,3.9103 0,2.32799 1.63687,3.94668 3.83755,3.94668 1.14581,0 1.98243,-0.38194 2.70993,-1.21856 v 1.07306 c 0,1.83693 -0.56381,2.67355 -2.41893,2.67355 -0.47287,0 -0.96393,-0.12731 -1.49137,-0.12731 -0.41831,0 -0.76387,0.32737 -0.76387,0.69112 0,0.582 0.63656,0.81844 1.94605,0.81844 1.65506,0 2.85543,-0.49106 3.51018,-1.34587 0.52744,-0.69113 0.54562,-1.47319 0.54562,-2.4735 0,-0.0728 0,-0.12731 0,-0.20006 z m -3.85573,-6.42017 c 1.56412,0 2.63718,1.09125 2.63718,2.58262 0,1.50956 -1.09125,2.58262 -2.63718,2.58262 -1.56413,0 -2.63718,-1.09125 -2.63718,-2.58262 0,-1.52775 1.09124,-2.58262 2.63718,-2.58262 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1047" />
+      <path
+         d="m -118.8835,177.68464 v -3.67386 c 1.18218,-1.34587 2.18249,-2.05518 3.16462,-2.05518 0.69112,0 1.07305,0.54562 1.54593,0.54562 0.38193,0 0.78206,-0.40012 0.78206,-0.83662 0,-0.65475 -0.65475,-1.14581 -1.76418,-1.14581 -1.41862,0 -2.619,0.69112 -3.72843,2.09155 v -1.50955 c 0,-0.25463 -0.16369,-0.36375 -0.49106,-0.36375 h -2.01881 c -0.47287,0 -0.70931,0.20006 -0.70931,0.61837 0,0.45469 0.25463,0.63656 0.83662,0.63656 0.0909,0 0.25463,0 0.45469,0 h 0.60019 v 5.69267 h -1.56412 c -0.49107,0 -0.74569,0.20006 -0.74569,0.61837 0,0.41832 0.25462,0.63657 0.7275,0.63657 h 5.80179 c 0.52744,0 0.7275,-0.12732 0.7275,-0.63657 0,-0.43649 -0.21825,-0.61837 -0.67293,-0.61837 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1049" />
+      <path
+         d="m -102.43182,174.8474 c 0,-2.45531 -1.81874,-4.31043 -4.6378,-4.31043 -2.81905,0 -4.65598,1.83694 -4.65598,4.31043 0,2.47349 1.83693,4.31042 4.65598,4.31042 2.81906,0 4.6378,-1.83693 4.6378,-4.31042 z m -4.6378,3.07368 c -1.83693,0 -3.11005,-1.3095 -3.11005,-3.07368 0,-1.76418 1.27312,-3.09187 3.11005,-3.09187 1.83694,0 3.11006,1.32769 3.11006,3.09187 0,1.78237 -1.27312,3.07368 -3.11006,3.07368 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1051" />
+      <path
+         d="m -98.383971,175.66583 v -4.25586 c 0,-0.0364 0,-0.0546 0,-0.0909 0,-0.40013 -0.01819,-0.582 -0.509249,-0.582 h -1.34587 c -0.47287,0 -0.70931,0.20006 -0.70931,0.61837 0,0.45469 0.25462,0.63656 0.81843,0.63656 0.12732,0 0.272815,0 0.400128,0 v 4.00124 c 0,2.07337 0.982122,3.12824 2.928178,3.12824 1.07306,0 1.836933,-0.41831 2.764492,-1.14581 v 0.60019 c 0,0.21825 0.163687,0.36375 0.509249,0.36375 h 1.345871 c 0.472874,0 0.70931,-0.21825 0.70931,-0.63657 0,-0.45468 -0.254624,-0.61837 -0.818435,-0.61837 -0.127312,0 -0.272811,0 -0.400124,0 v -6.09279 c 0,-0.52744 0,-0.85481 -0.618373,-0.85481 h -1.891494 c -0.491061,0 -0.709311,0.18187 -0.709311,0.61837 0,0.4365 0.181875,0.63656 0.654748,0.63656 h 1.218559 v 3.67386 c 0,1.3095 -1.145809,2.14612 -2.34618,2.14612 -1.382246,0 -2.000619,-0.69112 -2.000619,-2.14612 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1053" />
+      <path
+         d="m -84.733154,177.64827 c -1.655058,0 -2.800867,-1.18219 -2.800867,-2.78268 0,-1.6005 1.145809,-2.78268 2.800867,-2.78268 1.673245,0 2.819054,1.18218 2.819054,2.78268 0,1.65505 -1.182184,2.78268 -2.819054,2.78268 z m -2.764492,3.56474 v -3.54656 c 0.727498,0.83663 1.818744,1.29131 3.073678,1.29131 2.327993,0 4.037613,-1.63687 4.037613,-4.00124 0,-2.49168 -1.855119,-4.23767 -4.183112,-4.23767 -1.127622,0 -2.109744,0.41831 -2.909992,1.23675 v -0.85481 c 0,-0.25463 -0.163687,-0.36375 -0.509248,-0.36375 h -1.364059 c -0.472873,0 -0.70931,0.20006 -0.70931,0.61837 0,0.45469 0.272811,0.63656 0.836622,0.63656 0.127312,0 0.272812,0 0.400124,0 v 9.22104 c -0.05456,0 -0.109125,0 -0.163687,0 -0.745685,0 -1.073059,0.12731 -1.073059,0.63656 0,0.41831 0.236437,0.63656 0.70931,0.63656 h 3.928489 c 0.527436,0 0.745685,-0.1455 0.745685,-0.63656 0,-0.41831 -0.218249,-0.63656 -0.672936,-0.63656 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1055" />
+      <path
+         d="m -76.538571,172.01016 v -4.32861 c 0,-0.25463 -0.163687,-0.36375 -0.509249,-0.36375 h -1.345871 c -0.472874,0 -0.70931,0.21825 -0.70931,0.63656 0,0.45468 0.254624,0.61837 0.818435,0.61837 0.127312,0 0.272811,0.0182 0.400124,0.0182 v 9.09372 c -0.07275,0 -0.127313,0 -0.181875,0 -0.727498,0 -1.036684,0.10913 -1.036684,0.61837 0,0.41832 0.236436,0.63657 0.70931,0.63657 h 1.618683 c 0.272812,0 0.363749,-0.1455 0.363749,-0.45469 v -0.70931 c 0.600186,0.89118 1.527745,1.32768 2.764492,1.32768 2.382555,0 4.2013,-1.8733 4.2013,-4.18311 0,-2.34618 -1.78237,-4.18311 -4.164925,-4.18311 -1.254934,0 -2.255244,0.4365 -2.928179,1.27312 z m 2.764492,5.69267 c -1.655058,0 -2.800867,-1.20037 -2.800867,-2.83724 0,-1.65506 1.163996,-2.83724 2.800867,-2.83724 1.655057,0 2.800866,1.21855 2.800866,2.83724 0,1.69143 -1.145809,2.83724 -2.800866,2.83724 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1057" />
+      <path
+         d="m -63.979016,181.21301 4.674173,-9.22104 c 0.618374,-0.0182 0.891185,-0.16369 0.891185,-0.63656 0,-0.41831 -0.236437,-0.61837 -0.70931,-0.61837 h -2.200681 c -0.581999,0 -0.836623,0.0909 -0.836623,0.61837 0,0.50925 0.309187,0.63656 1.07306,0.63656 0.109124,0 0.218249,0 0.345561,0 l -2.473493,4.89242 -2.382555,-4.89242 c 0.109125,0 0.218249,0 0.327374,0 0.727498,0 1.036684,-0.12731 1.036684,-0.63656 0,-0.52744 -0.254624,-0.61837 -0.836622,-0.61837 h -2.473493 c -0.472874,0 -0.70931,0.20006 -0.70931,0.61837 0,0.50925 0.290999,0.63656 1.018497,0.63656 l 3.255553,6.38379 -1.454996,2.83725 h -1.927869 c -0.454687,0 -0.672936,0.21825 -0.672936,0.63656 0,0.49106 0.218249,0.63656 0.745685,0.63656 h 4.073988 c 0.472874,0 0.727498,-0.21825 0.727498,-0.63656 0,-0.52744 -0.327374,-0.63656 -1.109434,-0.63656 -0.109125,0 -0.236437,0 -0.381936,0 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1059" />
+      <path
+         d="m -50.564636,168.49998 c 0,-0.25462 -0.1455,-0.40012 -0.345562,-0.40012 -0.418311,0 -1.000309,0.56381 -1.691432,1.69143 -1.054872,1.70962 -1.564121,3.51018 -1.564121,5.5108 0,2.00062 0.509249,3.80117 1.564121,5.51079 0.691123,1.12762 1.273121,1.69143 1.691432,1.69143 0.200062,0 0.345562,-0.16368 0.345562,-0.41831 0,-0.34556 -0.491061,-1.07306 -1.00031,-2.36436 -0.563811,-1.455 -0.85481,-2.85543 -0.85481,-4.41955 0,-2.00062 0.472874,-3.51018 1.109434,-5.01974 0.381937,-0.90937 0.745686,-1.47318 0.745686,-1.78237 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1061" />
+      <path
+         d="m -43.352175,172.79222 h 0.545623 c 0.345562,0 0.418311,-0.16369 0.454686,-0.54563 l 0.272812,-3.29192 c 0,-0.0364 0,-0.0727 0,-0.10913 0,-0.291 -0.1455,-0.4365 -0.418311,-0.4365 h -1.182184 c -0.272812,0 -0.400124,0.1455 -0.400124,0.4365 0,0.0364 0,0.0728 0,0.10913 l 0.272812,3.29192 c 0.03637,0.38194 0.109124,0.54563 0.454686,0.54563 z m 3.310115,0 h 0.545623 c 0.345562,0 0.418312,-0.16369 0.454687,-0.54563 l 0.272811,-3.29192 c 0,-0.0364 0,-0.0727 0,-0.10913 0,-0.291 -0.127312,-0.4365 -0.400124,-0.4365 h -1.182184 c -0.272811,0 -0.418311,0.1455 -0.418311,0.4365 0,0.0364 0,0.0728 0,0.10913 l 0.272812,3.29192 c 0.03637,0.38194 0.109124,0.54563 0.454686,0.54563 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1063" />
+      <path
+         d="m -33.084239,178.63039 c 0.891185,0.40012 1.70962,0.61837 2.618993,0.61837 2.49168,0 4.092175,-1.32768 4.092175,-3.29193 0,-1.67324 -1.054872,-2.67355 -3.128241,-2.94636 l -1.418621,-0.18188 c -1.364058,-0.18187 -2.036994,-0.70931 -2.036994,-1.60049 0,-1.12762 0.963935,-1.92787 2.418931,-1.92787 0.945747,0 1.70962,0.36375 2.091556,0.92756 0.381937,0.56381 0.272812,1.21856 0.982122,1.21856 0.418312,0 0.672936,-0.23644 0.672936,-0.65475 0,-0.0364 0,-0.0546 0,-0.0909 l -0.127312,-1.85512 c -0.03638,-0.47287 -0.09094,-0.67293 -0.727498,-0.67293 -0.400124,0 -0.563811,0.10912 -0.563811,0.45468 -0.818435,-0.32737 -1.527745,-0.52743 -2.255243,-0.52743 -2.291619,0 -3.928489,1.43681 -3.928489,3.20099 0,1.78237 1.163997,2.63718 3.528365,2.96455 l 1.309496,0.18188 c 1.127622,0.16368 1.727807,0.74568 1.727807,1.67324 0,1.09125 -1.000309,1.92787 -2.582617,1.92787 -1.273121,0 -2.200681,-0.50925 -2.582617,-1.29131 -0.327374,-0.65475 -0.236437,-1.34587 -0.982123,-1.34587 -0.545623,0 -0.691123,0.25462 -0.691123,0.76387 0,0.0546 0,0.10913 0,0.18188 l 0.109125,2.05518 c 0.03637,0.63656 0.254624,0.92756 0.763873,0.92756 0.491061,0 0.654748,-0.18188 0.70931,-0.70931 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1065" />
+      <path
+         d="m -22.5071,174.06534 c 0.345561,-1.47318 1.49137,-2.34618 3.055491,-2.34618 1.436808,0 2.56443,0.92756 2.728117,2.34618 z m -0.03637,1.164 h 5.965482 c 1.091247,0 1.473184,0 1.473184,-0.83663 0,-1.98243 -1.655058,-3.89211 -4.3468,-3.89211 -2.746305,0 -4.637799,1.76418 -4.637799,4.31042 0,2.65537 1.70962,4.3468 4.455924,4.3468 1.200372,0 2.491681,-0.29099 3.71024,-0.89118 0.527435,-0.25462 0.78206,-0.52744 0.78206,-0.89119 0,-0.32737 -0.272812,-0.58199 -0.618373,-0.58199 -0.800248,0 -1.946057,1.10943 -3.746614,1.10943 -1.836933,0 -2.946367,-0.96393 -3.037304,-2.67355 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1067" />
+      <path
+         d="m -10.547716,177.68464 2.0369943,-2.18249 2.237056,2.18249 h -0.4910611 c -0.5274359,0 -0.8002476,0.25463 -0.8002476,0.63656 0,0.4365 0.3091866,0.61838 0.9093723,0.61838 h 1.8187446 c 0.7820603,0 1.0730594,-0.0909 1.0730594,-0.61838 0,-0.40012 -0.2364368,-0.63656 -0.691123,-0.63656 -0.018187,0 -0.054562,0 -0.090937,0 l -3.1100534,-3.0373 2.5462425,-2.65537 c 0.090937,0 0.1818745,0.0182 0.2546243,0.0182 0.6183732,0 0.9275598,-0.20006 0.9275598,-0.65475 0,-0.38194 -0.2546243,-0.61837 -0.6183732,-0.61837 H -6.855664 c -0.6183732,0 -0.9093724,0.18187 -0.9093724,0.63656 0,0.41831 0.3273741,0.61837 0.9821222,0.61837 0.07275,0 0.1273121,0 0.1818744,0 l -1.8187446,1.87331 -1.9096816,-1.87331 c 0.6729352,0 1.0184967,-0.20006 1.0184967,-0.61837 0,-0.45469 -0.2909992,-0.63656 -0.9093727,-0.63656 h -1.818744 c -0.782061,0 -1.091247,0.10912 -1.091247,0.63656 0,0.47287 0.254624,0.61837 0.909372,0.61837 0.03638,0 0.09094,0 0.127312,0 l 2.8190546,2.74631 -2.8736166,2.94636 c -0.07275,0 -0.127312,0 -0.181874,0 -0.654749,0 -1.00031,0.18188 -1.00031,0.582 0,0.4365 0.254624,0.67294 0.618373,0.67294 h 2.437118 c 0.6001858,0 0.9275598,-0.18188 0.9275598,-0.61838 0,-0.40012 -0.3455615,-0.65475 -0.9457468,-0.65475 -0.09094,0 -0.163687,0.0182 -0.254625,0.0182 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1069" />
+      <path
+         d="m 0.41135921,172.79222 h 0.5456234 c 0.34556149,0 0.41831129,-0.16369 0.45468619,-0.54563 l 0.2728117,-3.29192 c 0,-0.0364 0,-0.0727 0,-0.10913 0,-0.291 -0.1454996,-0.4365 -0.4183113,-0.4365 H 0.08398518 c -0.2728117,0 -0.40012383,0.1455 -0.40012383,0.4365 0,0.0364 0,0.0728 0,0.10913 l 0.2728117,3.29192 c 0.03637489,0.38194 0.10912468,0.54563 0.45468616,0.54563 z m 3.31011529,0 h 0.5456234 c 0.3455615,0 0.4183113,-0.16369 0.4546862,-0.54563 l 0.2728117,-3.29192 c 0,-0.0364 0,-0.0727 0,-0.10913 0,-0.291 -0.1273122,-0.4365 -0.4001239,-0.4365 h -1.182184 c -0.2728117,0 -0.4183113,0.1455 -0.4183113,0.4365 0,0.0364 0,0.0728 0,0.10913 l 0.2728117,3.29192 c 0.036375,0.38194 0.1091247,0.54563 0.4546862,0.54563 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1071" />
+      <path
+         d="m 11.461371,182.086 c 0,0.25463 0.1455,0.41831 0.345562,0.41831 0.418311,0 1.000309,-0.56381 1.691432,-1.69143 1.054872,-1.70962 1.582308,-3.51017 1.582308,-5.51079 0,-2.00062 -0.527436,-3.80118 -1.582308,-5.5108 -0.691123,-1.12762 -1.273121,-1.69143 -1.691432,-1.69143 -0.200062,0 -0.345562,0.1455 -0.345562,0.40012 0,0.34556 0.491061,1.07306 1.00031,2.36437 0.563811,1.45499 0.85481,2.87362 0.85481,4.43774 0,2.00062 -0.454686,3.51017 -1.091247,5.01973 -0.381936,0.90937 -0.763873,1.455 -0.763873,1.76418 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1073" />
+      <path
+         d="m 55.297671,177.48458 c 0,0.90937 0.691123,1.58231 1.745995,1.58231 1.054872,0 1.745995,-0.65475 1.745995,-1.58231 0,-0.92756 -0.691123,-1.58231 -1.745995,-1.58231 -1.054872,0 -1.745995,0.67294 -1.745995,1.58231 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1075" />
+      <path
+         d="m 71.221919,173.99259 v 4.0558 c 0,0.582 0.254624,0.89119 0.745685,0.89119 h 1.073059 c 0.418312,0 0.636561,-0.21825 0.636561,-0.63657 0,-0.50924 -0.254624,-0.61837 -0.92756,-0.61837 -0.05456,0 -0.109124,0 -0.163687,0 v -4.01942 c 0,-2.07337 -0.545623,-3.16462 -2.182493,-3.16462 -0.85481,0 -1.491371,0.36375 -1.982432,1.09125 -0.254624,-0.7275 -0.800248,-1.09125 -1.618683,-1.09125 -0.691123,0 -1.236746,0.23644 -1.70962,0.76387 -0.03637,-0.4365 -0.127312,-0.52743 -0.509248,-0.52743 h -1.364059 c -0.472873,0 -0.691123,0.20006 -0.691123,0.61837 0,0.45469 0.254625,0.63656 0.818435,0.63656 0.127313,0 0.272812,0 0.400124,0 v 5.69267 c -0.07275,0 -0.127312,0 -0.181874,0 -0.727498,0 -1.036685,0.10913 -1.036685,0.61837 0,0.41832 0.21825,0.63657 0.691123,0.63657 h 2.364368 c 0.527436,0 0.727498,-0.12732 0.727498,-0.63657 0,-0.49106 -0.236437,-0.63656 -0.85481,-0.63656 -0.109124,0 -0.236437,0.0182 -0.381936,0.0182 v -4.89242 c 0.327374,-0.61837 0.763873,-0.92756 1.309496,-0.92756 0.672935,0 1.163997,0.52743 1.163997,1.43681 0,0.10912 0,0.34556 0,0.69112 v 3.74661 c 0,0.0909 0,0.16369 0,0.23644 0,0.69112 0.03637,0.96394 0.618373,0.96394 h 1.127621 c 0.418312,0 0.618374,-0.21825 0.618374,-0.63657 0,-0.50924 -0.254625,-0.61837 -0.92756,-0.61837 -0.05456,0 -0.109125,0 -0.163687,0 v -3.69205 c 0,-1.41862 0.618373,-2.12793 1.436808,-2.12793 0.891185,0 0.963935,0.78206 0.963935,2.12793 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1077" />
+      <path
+         d="m 75.960887,174.06534 c 0.345561,-1.47318 1.491371,-2.34618 3.055491,-2.34618 1.436808,0 2.56443,0.92756 2.728117,2.34618 z m -0.03638,1.164 h 5.965483 c 1.091246,0 1.473183,0 1.473183,-0.83663 0,-1.98243 -1.655058,-3.89211 -4.3468,-3.89211 -2.746304,0 -4.637799,1.76418 -4.637799,4.31042 0,2.65537 1.70962,4.3468 4.455924,4.3468 1.200372,0 2.491681,-0.29099 3.71024,-0.89118 0.527436,-0.25462 0.78206,-0.52744 0.78206,-0.89119 0,-0.32737 -0.272812,-0.58199 -0.618373,-0.58199 -0.800248,0 -1.946057,1.10943 -3.746614,1.10943 -1.836933,0 -2.946367,-0.96393 -3.037304,-2.67355 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1079" />
+      <path
+         d="m 91.885135,177.81195 0.09094,0.54563 c 0.07275,0.38193 0.290999,0.582 0.618373,0.582 h 1.400434 c 0.472873,0 0.70931,-0.21825 0.70931,-0.63657 0,-0.52743 -0.309187,-0.61837 -1.054872,-0.61837 h -0.454686 v -4.01942 c 0,-2.14612 -1.054872,-3.14643 -3.328303,-3.14643 -2.000619,0 -3.310115,0.63656 -3.310115,1.25493 0,0.45469 0.272812,0.74569 0.600186,0.74569 0.691123,0 1.41862,-0.74569 2.673554,-0.74569 1.364059,0 2.000619,0.63656 2.000619,2.00062 0,0.0182 0,0.0364 0,0.0546 -0.727497,-0.18187 -1.41862,-0.27281 -2.109743,-0.27281 -2.63718,0 -4.110363,1.10944 -4.110363,2.94637 0,1.56412 1.163996,2.65536 2.946366,2.65536 1.218559,0 2.364368,-0.47287 3.328303,-1.34587 z m -0.05456,-2.80086 v 0.96393 c 0,0.80025 -1.454995,1.81875 -3.055491,1.81875 -1.036684,0 -1.745995,-0.60019 -1.745995,-1.38225 0,-1.00031 1.036685,-1.70962 2.928179,-1.70962 0.654748,0 1.273122,0.10912 1.873307,0.30919 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1081" />
+      <path
+         d="m 97.4971,177.68464 c -0.07275,0 -0.127312,0 -0.181874,0 -0.727498,0 -1.036685,0.10913 -1.036685,0.61837 0,0.41832 0.236437,0.63657 0.745686,0.63657 0.236437,0 0.400124,0 0.472873,0 h 1.618683 c 0.672936,0 0.945747,-0.0364 0.945747,-0.63657 0,-0.49106 -0.236437,-0.63656 -0.85481,-0.63656 -0.109124,0 -0.236436,0.0182 -0.381936,0.0182 v -3.69205 c 0,-1.32768 1.236746,-2.12793 2.309806,-2.12793 1.164,0 1.81874,0.70931 1.81874,2.12793 v 3.69205 c -0.10912,0 -0.20006,0 -0.29099,0 -0.63657,0 -0.89119,0.12731 -0.89119,0.61837 0,0.582 0.25462,0.63657 0.89119,0.63657 h 1.96424 c 0.63656,0 0.94575,-0.12732 0.94575,-0.63657 0,-0.45468 -0.27282,-0.61837 -0.83663,-0.61837 -0.12731,0 -0.27281,0 -0.40012,0 v -4.01942 c 0,-2.05519 -1.20037,-3.16462 -2.8918,-3.16462 -0.92756,0 -1.745999,0.40012 -2.600809,1.20037 v -0.60018 c 0,-0.21825 -0.163687,-0.36375 -0.509248,-0.36375 H 97.4971 c -0.818435,0 -1.218559,0 -1.218559,0.61837 0,0.45469 0.254625,0.63656 0.818436,0.63656 0.127312,0 0.272811,0 0.400123,0 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1083" />
+      <path
+         d="m 113.54866,168.49998 c 0,-0.25462 -0.1455,-0.40012 -0.34556,-0.40012 -0.41831,0 -1.00031,0.56381 -1.69143,1.69143 -1.05488,1.70962 -1.56412,3.51018 -1.56412,5.5108 0,2.00062 0.50924,3.80117 1.56412,5.51079 0.69112,1.12762 1.27312,1.69143 1.69143,1.69143 0.20006,0 0.34556,-0.16368 0.34556,-0.41831 0,-0.34556 -0.49106,-1.07306 -1.00031,-2.36436 -0.56381,-1.455 -0.85481,-2.85543 -0.85481,-4.41955 0,-2.00062 0.47287,-3.51018 1.10943,-5.01974 0.38194,-0.90937 0.74569,-1.47318 0.74569,-1.78237 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1085" />
+      <path
+         d="m 120.87025,182.086 c 0,0.25463 0.1455,0.41831 0.34556,0.41831 0.41831,0 1.00031,-0.56381 1.69143,-1.69143 1.05487,-1.70962 1.58231,-3.51017 1.58231,-5.51079 0,-2.00062 -0.52744,-3.80118 -1.58231,-5.5108 -0.69112,-1.12762 -1.27312,-1.69143 -1.69143,-1.69143 -0.20006,0 -0.34556,0.1455 -0.34556,0.40012 0,0.34556 0.49106,1.07306 1.00031,2.36437 0.56381,1.45499 0.85481,2.87362 0.85481,4.43774 0,2.00062 -0.45469,3.51017 -1.09125,5.01973 -0.38194,0.90937 -0.76387,1.455 -0.76387,1.76418 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.50526738"
+         id="path1087" />
+    </g>
+    <g
+       style="stroke-width:1.1112442"
+       id="g1962"
+       transform="matrix(0.89991951,0,0,0.89986489,12.004634,25.231115)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0-48"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 37.425333,257.87198 H 62.31326 v -12.188 H 37.425333 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 63.84131,244.15598 h 24.88795 v -12.188 H 63.84131 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-62"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 63.84131,257.87198 h 24.88795 v -12.188 H 63.84131 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-5-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 89.24131,244.15598 h 24.88795 v -12.188 H 89.24131 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-62-9"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 89.24131,257.87198 h 24.88795 v -12.188 H 89.24131 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0-4-8"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 37.425323,340.36628 H 62.31327 V 328.17827 H 37.425323 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-1-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 63.84132,326.65027 h 24.88795 v -12.188 H 63.84132 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-6-9"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 63.84132,340.36628 H 88.72927 V 328.17827 H 63.84132 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-1-7-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 89.24132,326.65027 h 24.88795 v -12.188 H 89.24132 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-6-9-9"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 89.24132,340.36628 h 24.88795 V 328.17827 H 89.24132 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-43-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -153.29861,334.01621 h 24.88793 v -12.1879 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-9-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,320.30021 h 24.88796 v -12.18786 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-49-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,334.01621 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-5-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -153.29862,346.71621 h 24.88794 v -12.1879 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-5-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,346.71621 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-7-9"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48263,320.30021 h 24.887953 v -12.18786 h -24.887953 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-3-5"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48263,334.01621 h 24.887953 v -12.1879 h -24.887953 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-9-6"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48263,346.71621 h 24.887953 v -12.1879 h -24.887953 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1-1-21"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082627,320.30021 h 24.88795 v -12.18786 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082627,334.01621 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-9-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082627,346.71621 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -153.29862,245.17197 h 24.88794 v -12.18799 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,231.45593 h 24.88797 v -12.18795 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-6-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,245.17197 h 24.88797 v -12.18799 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -153.29863,257.87197 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,257.87197 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-1"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48262,231.45593 h 24.887956 v -12.18795 h -24.887956 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-8"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48262,245.17197 h 24.887956 v -12.18799 h -24.887956 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-7"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48262,257.87197 h 24.887956 v -12.188 h -24.887956 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -153.29862,270.57197 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2-9-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -126.88263,270.57197 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4-0"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -101.48262,270.57197 h 24.887956 v -12.188 h -24.887956 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082624,270.57197 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082624,231.45593 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082624,245.17197 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.082624,257.87197 h 24.88796 v -12.188 h -24.88796 z" />
+    </g>
+    <g
+       style="stroke-width:1.1112442"
+       id="g1924"
+       transform="matrix(0.89991951,0,0,0.89986489,-23.834263,25.231115)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0-48-4-7"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 248.90994,289.59406 h 24.88793 v -12.188 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-5-5-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 275.32592,275.87806 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-62-5-3"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 275.32592,289.59406 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-5-8-9-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 300.72592,275.87806 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-62-9-2-6"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 300.72592,289.59406 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0-4-8-5-2"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 248.90993,302.29417 h 24.88795 v -12.18801 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-6-9-0-9"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 275.32593,302.29417 h 24.88795 v -12.18801 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-6-9-9-6-1"
+         style="fill:#ed7d30;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56901097;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 300.72593,302.29417 h 24.88795 v -12.18801 h -24.88795 z" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_groupby_select_detail.svg b/pandas/doc/source/_static/schemas/06_groupby_select_detail.svg
new file mode 100644
index 0000000000000..589c3add26e6f
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_groupby_select_detail.svg
@@ -0,0 +1,697 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="745.23389mm"
+   height="189.74706mm"
+   viewBox="0 0 745.23389 189.74705"
+   version="1.1"
+   id="svg15800"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_groupby_select_detail.svg">
+  <defs
+     id="defs15794">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.35"
+     inkscape:cx="903.54999"
+     inkscape:cy="155.22496"
+     inkscape:document-units="mm"
+     inkscape:current-layer="g2193"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1875"
+     inkscape:window-height="1029"
+     inkscape:window-x="45"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata15797">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(306.84909,-0.94988499)">
+    <g
+       id="g2193"
+       transform="matrix(0.89996563,0,0,0.89986489,6.5662338,9.5824747)"
+       style="stroke-width:1.11121571">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.70378,114.2688 h 24.88794 v -12.18799 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,100.55276 h 24.88797 V 88.364806 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,114.2688 h 24.88797 v -12.18799 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.70379,126.9688 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,126.9688 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,100.55276 h 24.88796 V 88.364806 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,114.2688 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,126.9688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.70378,139.6688 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,139.6688 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,139.6688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-3-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.70378,152.3688 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-0-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,152.3688 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-9-6"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,152.3688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-9-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -305.70378,165.0688 h 24.88794 v -12.1879 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -279.28779,165.0688 h 24.88797 v -12.1879 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-34-5"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -253.88778,165.0688 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,139.6688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,100.55276 h 24.88796 V 88.364806 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,114.2688 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,126.9688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-89-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,152.3688 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -228.48778,165.0688 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 386.82482,133.31877 h 24.88795 v -12.18795 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-4"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 413.24082,119.60277 h 24.88795 v -12.18794 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-4"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 413.24082,133.31877 h 24.88795 v -12.18795 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-9"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 386.82482,146.01877 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-9"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 413.24082,146.01877 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-0-9"
+         d="m 317.4389,126.72385 h 47.88958"
+         style="fill:none;stroke:#000000;stroke-width:0.44448629;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3)" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-0-9-7"
+         d="m -182.77222,126.72385 h 47.88958"
+         style="fill:none;stroke:#000000;stroke-width:0.44448629;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-1)" />
+      <g
+         aria-label=" titanic     .groupby(&quot;Sex&quot;)    [&quot;Age&quot;]     .mean()"
+         style="font-style:normal;font-weight:normal;font-size:20.21069527px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.56146109"
+         id="text56748-3">
+        <path
+           d="m -285.91965,5.9621374 h -3.88046 V 3.9612786 c 0,-0.9701134 -0.0808,-1.3743273 -0.78821,-1.3743273 -0.52548,0 -0.76801,0.2829497 -0.76801,0.8286385 0,0.262739 0,0.4446353 0,0.5254781 v 2.0210695 h -1.09138 c -0.9499,0 -1.35411,0.060632 -1.35411,0.7073743 0,0.4850567 0.24252,0.6871637 0.78821,0.6871637 h 1.65728 v 3.2943436 c 0,0.101053 0,0.202107 0,0.30316 0,0.909481 0.0202,1.697699 0.36379,2.283809 0.5659,0.990324 1.61686,1.47538 3.13266,1.47538 0.92969,0 1.96044,-0.262739 3.19329,-0.747795 0.72759,-0.28295 1.09138,-0.5659 1.09138,-1.030746 0,-0.363792 -0.32337,-0.707374 -0.68717,-0.707374 -0.78821,0 -2.10191,1.050956 -3.69855,1.050956 -1.73812,0 -1.85939,-1.050956 -1.85939,-2.62739 V 7.3566754 h 4.00172 c 0.54569,0 0.82864,-0.2223177 0.82864,-0.6871637 0,-0.5052673 -0.30316,-0.7073743 -0.92969,-0.7073743 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path935" />
+        <path
+           d="M -275.5402,13.136934 V 5.8206625 c 0,-0.2425283 -0.18189,-0.4042139 -0.5659,-0.4042139 h -2.89013 c -0.52547,0 -0.78821,0.2223177 -0.78821,0.6871637 0,0.464846 0.28295,0.7073743 0.82863,0.7073743 h 1.92002 v 6.3259474 h -3.01139 c -0.54569,0 -0.82864,0.222318 -0.82864,0.687164 0,0.464846 0.26274,0.707374 0.78822,0.707374 h 7.65985 c 0.58611,0 0.80843,-0.141475 0.80843,-0.707374 0,-0.485057 -0.24253,-0.687164 -0.7478,-0.687164 z m -1.83917,-10.5701934 c 0,0.8084278 0.20211,0.9701134 1.03074,0.9701134 0.94991,0 1.01054,-0.2223177 1.01054,-1.2530631 0,-1.0307455 -0.10105,-1.33390591 -1.01054,-1.33390591 -0.90948,0 -1.03074,0.30316041 -1.03074,1.61685561 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path937" />
+        <path
+           d="m -261.60366,5.9621374 h -3.88045 V 3.9612786 c 0,-0.9701134 -0.0808,-1.3743273 -0.78822,-1.3743273 -0.52548,0 -0.76801,0.2829497 -0.76801,0.8286385 0,0.262739 0,0.4446353 0,0.5254781 v 2.0210695 h -1.09138 c -0.9499,0 -1.35411,0.060632 -1.35411,0.7073743 0,0.4850567 0.24253,0.6871637 0.78821,0.6871637 h 1.65728 v 3.2943436 c 0,0.101053 0,0.202107 0,0.30316 0,0.909481 0.0202,1.697699 0.36379,2.283809 0.5659,0.990324 1.61686,1.47538 3.13266,1.47538 0.92969,0 1.96044,-0.262739 3.19329,-0.747795 0.72759,-0.28295 1.09138,-0.5659 1.09138,-1.030746 0,-0.363792 -0.32337,-0.707374 -0.68716,-0.707374 -0.78822,0 -2.10192,1.050956 -3.69856,1.050956 -1.73812,0 -1.85939,-1.050956 -1.85939,-2.62739 V 7.3566754 h 4.00172 c 0.54569,0 0.82864,-0.2223177 0.82864,-0.6871637 0,-0.5052673 -0.30316,-0.7073743 -0.92969,-0.7073743 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path939" />
+        <path
+           d="m -249.8903,13.278409 0.10105,0.606321 c 0.0808,0.424425 0.32338,0.646742 0.68717,0.646742 h 1.55622 c 0.52548,0 0.78822,-0.242528 0.78822,-0.707374 0,-0.58611 -0.34358,-0.687164 -1.17222,-0.687164 h -0.50527 V 8.6703706 c 0,-2.3848621 -1.17222,-3.4964503 -3.69856,-3.4964503 -2.22317,0 -3.67834,0.7073743 -3.67834,1.394538 0,0.5052674 0.30316,0.8286385 0.66695,0.8286385 0.76801,0 1.57643,-0.8286385 2.97097,-0.8286385 1.5158,0 2.22318,0.7073743 2.22318,2.2231765 0,0.020211 0,0.040421 0,0.060632 -0.80843,-0.2021069 -1.57644,-0.3031604 -2.34444,-0.3031604 -2.93055,0 -4.56762,1.2328524 -4.56762,3.2741326 0,1.73812 1.29349,2.950762 3.27413,2.950762 1.35412,0 2.62739,-0.525479 3.69856,-1.495592 z m -0.0606,-3.112447 v 1.071167 c 0,0.88927 -1.61686,2.021069 -3.3954,2.021069 -1.15201,0 -1.94023,-0.666953 -1.94023,-1.536012 0,-1.111589 1.15201,-1.8998058 3.25393,-1.8998058 0.72758,0 1.41475,0.1212642 2.0817,0.3435818 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path941" />
+        <path
+           d="m -243.65404,13.136934 c -0.0808,0 -0.14147,0 -0.2021,0 -0.80843,0 -1.15201,0.121264 -1.15201,0.687164 0,0.464846 0.26274,0.707374 0.82864,0.707374 0.26273,0 0.44463,0 0.52547,0 h 1.79876 c 0.74779,0 1.05095,-0.04042 1.05095,-0.707374 0,-0.545689 -0.26274,-0.707374 -0.9499,-0.707374 -0.12126,0 -0.26274,0.02021 -0.42443,0.02021 V 9.0341631 c 0,-1.4753808 1.37433,-2.3646514 2.56676,-2.3646514 1.29349,0 2.02107,0.7882172 2.02107,2.3646514 v 4.1027709 c -0.12126,0 -0.22232,0 -0.32337,0 -0.70737,0 -0.99032,0.141475 -0.99032,0.687164 0,0.646742 0.28295,0.707374 0.99032,0.707374 h 2.18276 c 0.70737,0 1.05095,-0.141475 1.05095,-0.707374 0,-0.505268 -0.30316,-0.687164 -0.92969,-0.687164 -0.14147,0 -0.30316,0 -0.44463,0 V 8.6703706 c 0,-2.2838086 -1.33391,-3.516661 -3.21351,-3.516661 -1.03074,0 -1.94022,0.4446353 -2.89012,1.3339059 v -0.666953 c 0,-0.2425283 -0.1819,-0.4042139 -0.5659,-0.4042139 h -0.9297 c -0.90948,0 -1.35411,0 -1.35411,0.6871637 0,0.5052674 0.28295,0.7073743 0.90948,0.7073743 0.14147,0 0.30316,0 0.44463,0 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path943" />
+        <path
+           d="M -226.90821,13.136934 V 5.8206625 c 0,-0.2425283 -0.1819,-0.4042139 -0.5659,-0.4042139 h -2.89013 c -0.52548,0 -0.78822,0.2223177 -0.78822,0.6871637 0,0.464846 0.28295,0.7073743 0.82864,0.7073743 h 1.92001 v 6.3259474 h -3.01139 c -0.54569,0 -0.82864,0.222318 -0.82864,0.687164 0,0.464846 0.26274,0.707374 0.78822,0.707374 h 7.65985 c 0.58611,0 0.80843,-0.141475 0.80843,-0.707374 0,-0.485057 -0.24253,-0.687164 -0.7478,-0.687164 z m -1.83918,-10.5701934 c 0,0.8084278 0.20211,0.9701134 1.03075,0.9701134 0.9499,0 1.01053,-0.2223177 1.01053,-1.2530631 0,-1.0307455 -0.10105,-1.33390591 -1.01053,-1.33390591 -0.90948,0 -1.03075,0.30316041 -1.03075,1.61685561 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path945" />
+        <path
+           d="M -211.21335,8.3874208 V 6.5482476 c 0,-0.8488492 -0.10105,-1.2126417 -0.70737,-1.2126417 -0.50527,0 -0.76801,0.1818962 -0.76801,0.6265315 0,0 0,0.020211 0,0.040421 -1.09137,-0.5456888 -2.0817,-0.8286385 -3.11244,-0.8286385 -2.8295,0 -4.97183,2.0817016 -4.97183,4.8101455 0,2.7284442 2.12212,4.7899352 4.99204,4.7899352 2.38486,0 4.60804,-1.475381 4.60804,-2.405073 0,-0.404214 -0.28295,-0.687164 -0.68717,-0.687164 -0.52548,0 -0.90948,0.485057 -1.5158,0.889271 -0.70737,0.464846 -1.5158,0.727585 -2.30402,0.727585 -2.06149,0 -3.37519,-1.374327 -3.37519,-3.3347649 0,-1.8998054 1.41475,-3.253922 3.45603,-3.253922 1.17222,0 2.10192,0.5456888 2.70824,1.6168557 0.30316,0.5456887 0.50526,0.7882171 0.92969,0.7882171 0.48505,0 0.74779,-0.2425284 0.74779,-0.7275851 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path947" />
+        <path
+           d="m -144.65194,12.914617 c 0,1.010534 0.768,1.75833 1.94022,1.75833 1.17222,0 1.94023,-0.727585 1.94023,-1.75833 0,-1.030746 -0.76801,-1.758331 -1.94023,-1.758331 -1.17222,0 -1.94022,0.747796 -1.94022,1.758331 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path949" />
+        <path
+           d="M -126.55201,14.005994 V 6.8109866 c 0.0606,0 0.12127,0 0.1819,0 0.82864,0 1.17222,-0.1414749 1.17222,-0.7073743 0,-0.464846 -0.26274,-0.6871637 -0.78822,-0.6871637 h -1.49559 c -0.384,0 -0.5659,0.1212642 -0.5659,0.4042139 v 0.7073744 c -0.9499,-0.7477957 -1.89981,-1.131799 -2.91034,-1.131799 -2.38486,0 -4.3453,1.7987519 -4.3453,4.3452995 0,2.5869686 1.81896,4.3857206 4.26446,4.3857206 1.27327,0 2.20296,-0.424424 3.01139,-1.354116 v 1.192431 c 0,2.04128 -0.62653,2.970972 -2.68802,2.970972 -0.52548,0 -1.07117,-0.141475 -1.65728,-0.141475 -0.46485,0 -0.84885,0.363793 -0.84885,0.768006 0,0.646743 0.70738,0.909482 2.16255,0.909482 1.83917,0 3.17308,-0.545689 3.90066,-1.495592 0.58611,-0.768006 0.60632,-1.637066 0.60632,-2.748654 0,-0.08084 0,-0.141475 0,-0.222318 z m -4.28467,-7.1343753 c 1.73812,0 2.93055,1.2126417 2.93055,2.8699187 0,1.6774876 -1.21264,2.8699186 -2.93055,2.8699186 -1.73812,0 -2.93055,-1.212642 -2.93055,-2.8699186 0,-1.6976984 1.21264,-2.8699187 2.93055,-2.8699187 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path951" />
+        <path
+           d="M -119.36584,13.136934 V 9.0543738 c 1.3137,-1.4955915 2.42529,-2.2838086 3.51667,-2.2838086 0.768,0 1.19243,0.6063209 1.7179,0.6063209 0.42443,0 0.86906,-0.4446353 0.86906,-0.929692 0,-0.727585 -0.72758,-1.2732738 -1.96043,-1.2732738 -1.57644,0 -2.91034,0.7680064 -4.1432,2.32423 V 5.8206625 c 0,-0.2829497 -0.18189,-0.4042139 -0.54568,-0.4042139 h -2.24339 c -0.52548,0 -0.78822,0.2223177 -0.78822,0.6871637 0,0.5052674 0.28295,0.7073743 0.92969,0.7073743 0.10106,0 0.28295,0 0.50527,0 h 0.66695 v 6.3259474 h -1.73812 c -0.54568,0 -0.82863,0.222318 -0.82863,0.687164 0,0.464846 0.28295,0.707374 0.80842,0.707374 h 6.44722 c 0.58611,0 0.80842,-0.141475 0.80842,-0.707374 0,-0.485057 -0.24253,-0.687164 -0.74779,-0.687164 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path953" />
+        <path
+           d="m -101.08401,9.9840658 c 0,-2.7284439 -2.02106,-4.7899348 -5.15372,-4.7899348 -3.13266,0 -5.17394,2.0412802 -5.17394,4.7899348 0,2.7486542 2.04128,4.7899352 5.17394,4.7899352 3.13266,0 5.15372,-2.041281 5.15372,-4.7899352 z m -5.15372,3.4156072 c -2.04128,0 -3.45603,-1.45517 -3.45603,-3.4156072 0,-1.9604375 1.41475,-3.4358182 3.45603,-3.4358182 2.04128,0 3.45603,1.4753807 3.45603,3.4358182 0,1.9806482 -1.41475,3.4156072 -3.45603,3.4156072 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path955" />
+        <path
+           d="M -96.585869,10.893547 V 6.1642444 c 0,-0.040421 0,-0.060632 0,-0.1010535 0,-0.4446353 -0.02021,-0.6467423 -0.5659,-0.6467423 h -1.495591 c -0.525478,0 -0.788217,0.2223177 -0.788217,0.6871637 0,0.5052674 0.28295,0.7073743 0.909481,0.7073743 0.141475,0 0.303161,0 0.444635,0 V 11.25734 c 0,2.304019 1.091378,3.476239 3.253922,3.476239 1.192431,0 2.041281,-0.464846 3.072026,-1.273274 v 0.666953 c 0,0.242529 0.181896,0.404214 0.565899,0.404214 h 1.495592 c 0.525478,0 0.788217,-0.242528 0.788217,-0.707374 0,-0.505268 -0.28295,-0.687164 -0.909481,-0.687164 -0.141475,0 -0.303161,0 -0.444636,0 V 6.3663513 c 0,-0.5861101 0,-0.9499027 -0.687163,-0.9499027 h -2.101912 c -0.545689,0 -0.788218,0.202107 -0.788218,0.6871637 0,0.4850567 0.202107,0.7073743 0.727585,0.7073743 h 1.354117 v 4.0825604 c 0,1.45517 -1.273274,2.384862 -2.60718,2.384862 -1.536013,0 -2.223176,-0.768006 -2.223176,-2.384862 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path957" />
+        <path
+           d="m -81.416471,13.096513 c -1.839173,0 -3.112447,-1.313695 -3.112447,-3.092237 0,-1.7785407 1.273274,-3.0922359 3.112447,-3.0922359 1.859384,0 3.132658,1.3136952 3.132658,3.0922359 0,1.839174 -1.313695,3.092237 -3.132658,3.092237 z m -3.072025,3.961296 v -3.941085 c 0.808427,0.929692 2.021069,1.434959 3.415607,1.434959 2.586969,0 4.486774,-1.818963 4.486774,-4.446353 0,-2.7688653 -2.061491,-4.7090921 -4.64846,-4.7090921 -1.253063,0 -2.34444,0.464846 -3.233711,1.3743273 V 5.8206625 c 0,-0.2829497 -0.181896,-0.4042139 -0.565899,-0.4042139 h -1.515802 c -0.525478,0 -0.788218,0.2223177 -0.788218,0.6871637 0,0.5052674 0.303161,0.7073743 0.929692,0.7073743 0.141475,0 0.303161,0 0.444636,0 V 17.057809 c -0.06063,0 -0.121264,0 -0.181897,0 -0.828638,0 -1.192431,0.141475 -1.192431,0.707374 0,0.464846 0.26274,0.707375 0.788218,0.707375 h 4.36551 c 0.58611,0 0.828638,-0.161686 0.828638,-0.707375 0,-0.464846 -0.242528,-0.707374 -0.747795,-0.707374 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path959" />
+        <path
+           d="M -72.310296,6.8311973 V 2.0210518 c 0,-0.2829497 -0.181897,-0.4042139 -0.5659,-0.4042139 h -1.495591 c -0.525478,0 -0.788217,0.2425284 -0.788217,0.7073744 0,0.5052673 0.282949,0.6871636 0.909481,0.6871636 0.141475,0 0.30316,0.020211 0.444635,0.020211 V 13.136934 c -0.08084,0 -0.141475,0 -0.202107,0 -0.808428,0 -1.152009,0.121264 -1.152009,0.687164 0,0.464846 0.262739,0.707374 0.788217,0.707374 h 1.798752 c 0.30316,0 0.404214,-0.161685 0.404214,-0.505267 v -0.788217 c 0.666953,0.990324 1.697698,1.47538 3.072025,1.47538 2.647601,0 4.668671,-2.081701 4.668671,-4.648459 0,-2.6071801 -1.980648,-4.6484604 -4.628249,-4.6484604 -1.394538,0 -2.506127,0.4850567 -3.253922,1.4147487 z m 3.072025,6.3259477 c -1.839173,0 -3.112447,-1.333906 -3.112447,-3.152869 0,-1.8391728 1.293485,-3.152868 3.112447,-3.152868 1.839174,0 3.112447,1.3541166 3.112447,3.152868 0,1.879595 -1.273273,3.152869 -3.112447,3.152869 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path961" />
+        <path
+           d="m -58.35354,17.057809 5.194149,-10.2468224 c 0.687164,-0.020211 0.990324,-0.1818962 0.990324,-0.7073743 0,-0.464846 -0.262739,-0.6871637 -0.788217,-0.6871637 h -2.445494 c -0.646742,0 -0.929692,0.1010535 -0.929692,0.6871637 0,0.5658994 0.343582,0.7073743 1.192431,0.7073743 0.121264,0 0.242528,0 0.384003,0 l -2.748654,5.4366774 -2.647602,-5.4366774 c 0.121265,0 0.242529,0 0.363793,0 0.808428,0 1.15201,-0.1414749 1.15201,-0.7073743 0,-0.5861102 -0.28295,-0.6871637 -0.929692,-0.6871637 h -2.748655 c -0.525478,0 -0.788217,0.2223177 -0.788217,0.6871637 0,0.5658994 0.323371,0.7073743 1.131799,0.7073743 l 3.617714,7.0939544 -1.616855,3.152868 h -2.142334 c -0.505267,0 -0.747796,0.242528 -0.747796,0.707374 0,0.545689 0.242529,0.707375 0.828639,0.707375 h 4.527196 c 0.525478,0 0.808427,-0.242529 0.808427,-0.707375 0,-0.58611 -0.363792,-0.707374 -1.232852,-0.707374 -0.121264,0 -0.262739,0 -0.424425,0 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path963" />
+        <path
+           d="m -43.446896,2.9305331 c 0,-0.2829497 -0.161685,-0.4446353 -0.384003,-0.4446353 -0.464846,0 -1.111588,0.6265316 -1.879594,1.8795947 -1.172221,1.8998053 -1.73812,3.9006642 -1.73812,6.1238405 0,2.223177 0.565899,4.224035 1.73812,6.123841 0.768006,1.253063 1.414748,1.879594 1.879594,1.879594 0.222318,0 0.384003,-0.181896 0.384003,-0.464846 0,-0.384003 -0.545688,-1.192431 -1.111588,-2.62739 -0.626531,-1.616856 -0.949903,-3.173079 -0.949903,-4.911199 0,-2.2231763 0.525479,-3.900664 1.232853,-5.5781517 0.424424,-1.0105348 0.828638,-1.6370664 0.828638,-1.9806482 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path965" />
+        <path
+           d="m -35.432083,7.7002572 h 0.60632 c 0.384004,0 0.464846,-0.1818963 0.505268,-0.6063209 l 0.30316,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.161685,-0.4850567 -0.464846,-0.4850567 h -1.313695 c -0.30316,0 -0.444635,0.1616856 -0.444635,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.30316,3.6581358 c 0.04042,0.4244246 0.121264,0.6063209 0.505268,0.6063209 z m 3.678346,0 h 0.606321 c 0.384003,0 0.464846,-0.1818963 0.505267,-0.6063209 l 0.303161,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.141475,-0.4850567 -0.444635,-0.4850567 h -1.313696 c -0.30316,0 -0.464846,0.1616856 -0.464846,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.303161,3.6581358 c 0.04042,0.4244246 0.121264,0.6063209 0.505267,0.6063209 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path967" />
+        <path
+           d="m -24.02189,14.18789 c 0.990324,0.444636 1.899806,0.687164 2.910341,0.687164 2.768865,0 4.547406,-1.475381 4.547406,-3.658136 0,-1.8593838 -1.17222,-2.970972 -3.47624,-3.2741325 l -1.576434,-0.2021069 c -1.515802,-0.202107 -2.263598,-0.7882171 -2.263598,-1.7785412 0,-1.2530631 1.071167,-2.1423337 2.688023,-2.1423337 1.050956,0 1.899805,0.4042139 2.32423,1.0307455 0.424424,0.6265315 0.30316,1.3541166 1.091377,1.3541166 0.464846,0 0.747796,-0.2627391 0.747796,-0.7275851 0,-0.040421 0,-0.060632 0,-0.1010534 l -0.141475,-2.061491 c -0.04042,-0.525478 -0.101053,-0.7477957 -0.808428,-0.7477957 -0.444635,0 -0.626531,0.1212642 -0.626531,0.5052674 -0.909482,-0.3637925 -1.697699,-0.5861102 -2.506126,-0.5861102 -2.546548,0 -4.365511,1.596645 -4.365511,3.5570824 0,1.9806481 1.293485,2.9305508 3.920875,3.2943433 l 1.45517,0.202107 c 1.253063,0.1818962 1.920016,0.8286385 1.920016,1.8593835 0,1.212642 -1.111588,2.142334 -2.869918,2.142334 -1.414749,0 -2.445495,-0.565899 -2.869919,-1.434959 -0.363793,-0.727585 -0.262739,-1.495592 -1.091378,-1.495592 -0.606321,0 -0.768006,0.28295 -0.768006,0.84885 0,0.06063 0,0.121264 0,0.202106 l 0.121264,2.283809 c 0.04042,0.707374 0.28295,1.030746 0.848849,1.030746 0.545689,0 0.727585,-0.202107 0.788217,-0.788218 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path969" />
+        <path
+           d="m -12.268114,9.1150059 c 0.384003,-1.6370663 1.657277,-2.6071797 3.3953968,-2.6071797 1.5966449,0 2.849708,1.0307454 3.0316043,2.6071797 z m -0.04042,1.2934841 h 6.6291077 c 1.2126417,0 1.6370663,0 1.6370663,-0.9296916 0,-2.2029658 -1.8391733,-4.3250888 -4.8303562,-4.3250888 -3.0518148,0 -5.1537268,1.9604374 -5.1537268,4.7899348 0,2.9507616 1.899805,4.8303566 4.9516199,4.8303566 1.3339058,0 2.7688652,-0.323372 4.1229818,-0.990325 0.5861102,-0.282949 0.8690599,-0.58611 0.8690599,-0.990324 0,-0.363792 -0.3031604,-0.646742 -0.6871637,-0.646742 -0.8892705,0 -2.1625443,1.232853 -4.1634032,1.232853 -2.0412797,0 -3.2741327,-1.071167 -3.3751857,-2.970973 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path971" />
+        <path
+           d="m 1.0216744,13.136934 2.2635979,-2.425283 2.4859155,2.425283 H 5.2254991 c -0.5861102,0 -0.8892706,0.28295 -0.8892706,0.707375 0,0.485056 0.3435818,0.687163 1.0105347,0.687163 h 2.0210696 c 0.8690599,0 1.192431,-0.101053 1.192431,-0.687163 0,-0.444636 -0.2627391,-0.707375 -0.7680064,-0.707375 -0.020211,0 -0.060632,0 -0.1010535,0 L 4.235175,9.7617481 7.0646723,6.8109866 c 0.1010535,0 0.202107,0.020211 0.2829498,0.020211 0.6871636,0 1.0307454,-0.2223176 1.0307454,-0.727585 0,-0.4244246 -0.2829497,-0.6871637 -0.6871636,-0.6871637 H 5.1244456 c -0.6871637,0 -1.0105348,0.202107 -1.0105348,0.7073744 0,0.464846 0.3637925,0.6871636 1.0913776,0.6871636 0.080843,0 0.1414748,0 0.2021069,0 L 3.3863258,8.8926882 1.2642028,6.8109866 c 0.7477957,0 1.1317989,-0.2223176 1.1317989,-0.6871636 0,-0.5052674 -0.3233711,-0.7073744 -1.0105347,-0.7073744 h -2.02106957 c -0.86905993,0 -1.21264173,0.1212642 -1.21264173,0.7073744 0,0.5254781 0.2829497,0.6871636 1.01053478,0.6871636 0.0404214,0 0.10105347,0 0.14147486,0 l 3.13265776,3.051815 -3.19328984,3.2741324 c -0.0808428,0 -0.14147487,0 -0.20210695,0 -0.72758501,0 -1.11158821,0.202107 -1.11158821,0.646742 0,0.485057 0.2829497,0.747796 0.6871636,0.747796 h 2.7082332 c 0.6669529,0 1.0307454,-0.202107 1.0307454,-0.687163 0,-0.444636 -0.3840032,-0.727585 -1.0509561,-0.727585 -0.1010535,0 -0.1818963,0.02021 -0.2829498,0.02021 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path973" />
+        <path
+           d="m 13.199905,7.7002572 h 0.606321 c 0.384003,0 0.464846,-0.1818963 0.505267,-0.6063209 l 0.303161,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.161686,-0.4850567 -0.464846,-0.4850567 h -1.313695 c -0.303161,0 -0.444636,0.1616856 -0.444636,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.303161,3.6581358 c 0.04042,0.4244246 0.121264,0.6063209 0.505267,0.6063209 z m 3.678347,0 h 0.606321 c 0.384003,0 0.464846,-0.1818963 0.505267,-0.6063209 L 18.293,3.4358005 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.141475,-0.4850567 -0.444635,-0.4850567 H 16.53467 c -0.303161,0 -0.464846,0.1616856 -0.464846,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.30316,3.6581358 c 0.04042,0.4244246 0.121264,0.6063209 0.505268,0.6063209 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path975" />
+        <path
+           d="m 25.479159,18.027922 c 0,0.28295 0.161685,0.464846 0.384003,0.464846 0.464846,0 1.111588,-0.626531 1.879595,-1.879594 1.17222,-1.899806 1.75833,-3.900664 1.75833,-6.123841 0,-2.2231763 -0.58611,-4.2240352 -1.75833,-6.1238405 -0.768007,-1.2530631 -1.414749,-1.8795947 -1.879595,-1.8795947 -0.222318,0 -0.384003,0.1616856 -0.384003,0.4446353 0,0.3840032 0.545689,1.192431 1.111588,2.6273904 0.626532,1.6168556 0.949903,3.1932899 0.949903,4.9314095 0,2.223177 -0.505268,3.900664 -1.212642,5.578152 -0.424425,1.010535 -0.848849,1.616856 -0.848849,1.960437 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path977" />
+        <path
+           d="M 87.987046,16.976966 V 4.0421214 h 1.657277 c 0.384003,0 0.646742,-0.2425284 0.646742,-0.6063209 0,-0.3840032 -0.242528,-0.6063209 -0.626531,-0.6063209 h -2.667812 c -0.424425,0 -0.707374,0.2829498 -0.707374,0.6467423 V 17.542866 c 0,0.363792 0.282949,0.646742 0.707374,0.646742 h 2.667812 c 0.384003,0 0.626531,-0.242528 0.626531,-0.626532 0,-0.363792 -0.262739,-0.58611 -0.646742,-0.58611 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path979" />
+        <path
+           d="m 98.305862,7.7002572 h 0.606321 c 0.384003,0 0.464846,-0.1818963 0.505267,-0.6063209 l 0.303161,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.161686,-0.4850567 -0.464846,-0.4850567 H 97.94207 c -0.303161,0 -0.444636,0.1616856 -0.444636,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.303161,3.6581358 c 0.04042,0.4244246 0.121264,0.6063209 0.505267,0.6063209 z m 3.678348,0 h 0.60632 c 0.384,0 0.46485,-0.1818963 0.50527,-0.6063209 l 0.30316,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.14148,-0.4850567 -0.44464,-0.4850567 h -1.31369 c -0.30316,0 -0.46485,0.1616856 -0.46485,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.30316,3.6581358 c 0.0404,0.4244246 0.12127,0.6063209 0.50527,0.6063209 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path981" />
+        <path
+           d="m 111.07017,4.2442283 -3.35497,8.8927057 c -0.0808,0 -0.16169,0 -0.24253,0 -0.66695,0 -0.92969,0.161686 -0.92969,0.687164 0,0.606321 0.28295,0.707374 1.03074,0.707374 h 2.58697 c 0.76801,0 1.15201,-0.06063 1.15201,-0.707374 0,-0.5659 -0.36379,-0.687164 -1.1318,-0.687164 h -0.9499 l 0.84885,-2.34444 h 5.01225 l 0.9499,2.34444 h -0.90948 c -0.768,0 -1.11159,0.121264 -1.11159,0.687164 0,0.646742 0.38401,0.707374 1.17222,0.707374 h 2.68803 c 0.64674,0 0.88927,-0.141475 0.88927,-0.707374 0,-0.545689 -0.24253,-0.687164 -0.92969,-0.687164 -0.0606,0 -0.10106,0 -0.16169,0 l -3.86024,-9.6405014 c -0.26274,-0.6467423 -0.52548,-0.666953 -1.09138,-0.666953 h -2.80929 c -0.92969,0 -1.37432,0 -1.37432,0.7073744 0,0.5456888 0.34358,0.7073743 1.09137,0.7073743 z m -0.50526,5.2345701 1.8998,-5.2345701 2.14233,5.2345701 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path983" />
+        <path
+           d="M 128.76592,14.005994 V 6.8109866 c 0.0606,0 0.12127,0 0.1819,0 0.82864,0 1.17222,-0.1414749 1.17222,-0.7073743 0,-0.464846 -0.26274,-0.6871637 -0.78822,-0.6871637 h -1.49559 c -0.384,0 -0.5659,0.1212642 -0.5659,0.4042139 v 0.7073744 c -0.9499,-0.7477957 -1.8998,-1.131799 -2.91034,-1.131799 -2.38486,0 -4.3453,1.7987519 -4.3453,4.3452995 0,2.5869686 1.81896,4.3857206 4.26446,4.3857206 1.27327,0 2.20296,-0.424424 3.01139,-1.354116 v 1.192431 c 0,2.04128 -0.62653,2.970972 -2.68802,2.970972 -0.52548,0 -1.07117,-0.141475 -1.65728,-0.141475 -0.46484,0 -0.84885,0.363793 -0.84885,0.768006 0,0.646743 0.70738,0.909482 2.16255,0.909482 1.83917,0 3.17308,-0.545689 3.90066,-1.495592 0.58611,-0.768006 0.60632,-1.637066 0.60632,-2.748654 0,-0.08084 0,-0.141475 0,-0.222318 z m -4.28466,-7.1343753 c 1.73812,0 2.93055,1.2126417 2.93055,2.8699187 0,1.6774876 -1.21265,2.8699186 -2.93055,2.8699186 -1.73812,0 -2.93056,-1.212642 -2.93056,-2.8699186 0,-1.6976984 1.21265,-2.8699187 2.93056,-2.8699187 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path985" />
+        <path
+           d="m 133.62785,9.1150059 c 0.384,-1.6370663 1.65728,-2.6071797 3.3954,-2.6071797 1.59664,0 2.84971,1.0307454 3.0316,2.6071797 z m -0.0404,1.2934841 h 6.62911 c 1.21264,0 1.63706,0 1.63706,-0.9296916 0,-2.2029658 -1.83917,-4.3250888 -4.83035,-4.3250888 -3.05182,0 -5.15373,1.9604374 -5.15373,4.7899348 0,2.9507616 1.89981,4.8303566 4.95162,4.8303566 1.33391,0 2.76887,-0.323372 4.12298,-0.990325 0.58611,-0.282949 0.86906,-0.58611 0.86906,-0.990324 0,-0.363792 -0.30316,-0.646742 -0.68716,-0.646742 -0.88927,0 -2.16254,1.232853 -4.1634,1.232853 -2.04128,0 -3.27414,-1.071167 -3.37519,-2.970973 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path987" />
+        <path
+           d="m 146.93785,7.7002572 h 0.60632 c 0.384,0 0.46485,-0.1818963 0.50527,-0.6063209 l 0.30316,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.16169,-0.4850567 -0.46485,-0.4850567 h -1.31369 c -0.30316,0 -0.44464,0.1616856 -0.44464,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.30316,3.6581358 c 0.0404,0.4244246 0.12127,0.6063209 0.50527,0.6063209 z m 3.67835,0 h 0.60632 c 0.384,0 0.46484,-0.1818963 0.50527,-0.6063209 l 0.30316,-3.6581358 c 0,-0.040421 0,-0.080843 0,-0.1212642 0,-0.3233711 -0.14148,-0.4850567 -0.44464,-0.4850567 h -1.31369 c -0.30316,0 -0.46485,0.1616856 -0.46485,0.4850567 0,0.040421 0,0.080843 0,0.1212642 l 0.30316,3.6581358 c 0.0404,0.4244246 0.12126,0.6063209 0.50527,0.6063209 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path989" />
+        <path
+           d="M 161.54133,4.0421214 V 16.976966 h -1.65727 c -0.38401,0 -0.64674,0.222318 -0.64674,0.58611 0,0.384004 0.24252,0.626532 0.62653,0.626532 h 2.6476 c 0.42442,0 0.72758,-0.28295 0.72758,-0.646742 V 3.4762219 c 0,-0.3637925 -0.30316,-0.6467423 -0.72758,-0.6467423 h -2.6476 c -0.38401,0 -0.62653,0.2223177 -0.62653,0.6063209 0,0.3637925 0.26273,0.6063209 0.64674,0.6063209 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path991" />
+        <path
+           d="m 232.24595,12.914617 c 0,1.010534 0.768,1.75833 1.94022,1.75833 1.17222,0 1.94023,-0.727585 1.94023,-1.75833 0,-1.030746 -0.76801,-1.758331 -1.94023,-1.758331 -1.17222,0 -1.94022,0.747796 -1.94022,1.758331 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path993" />
+        <path
+           d="m 249.94163,9.0341631 v 4.5069849 c 0,0.646742 0.28295,0.990324 0.82864,0.990324 h 1.19243 c 0.46485,0 0.70738,-0.242528 0.70738,-0.707374 0,-0.5659 -0.28295,-0.687164 -1.03075,-0.687164 -0.0606,0 -0.12126,0 -0.18189,0 V 8.6703706 c 0,-2.3040193 -0.60632,-3.516661 -2.42529,-3.516661 -0.9499,0 -1.65727,0.4042139 -2.20296,1.2126417 -0.28295,-0.8084278 -0.88927,-1.2126417 -1.79875,-1.2126417 -0.76801,0 -1.37433,0.262739 -1.89981,0.8488492 -0.0404,-0.4850567 -0.14147,-0.5861102 -0.5659,-0.5861102 h -1.5158 c -0.52548,0 -0.76801,0.2223177 -0.76801,0.6871637 0,0.5052674 0.28295,0.7073743 0.90948,0.7073743 0.14148,0 0.30316,0 0.44464,0 v 6.3259474 c -0.0808,0 -0.14148,0 -0.20211,0 -0.80843,0 -1.15201,0.121264 -1.15201,0.687164 0,0.464846 0.24253,0.707374 0.76801,0.707374 h 2.62739 c 0.58611,0 0.80843,-0.141475 0.80843,-0.707374 0,-0.545689 -0.26274,-0.707374 -0.94991,-0.707374 -0.12126,0 -0.26273,0.02021 -0.42442,0.02021 V 7.7002572 c 0.36379,-0.6871636 0.84885,-1.0307455 1.45517,-1.0307455 0.7478,0 1.29348,0.5861102 1.29348,1.596645 0,0.1212641 0,0.3840032 0,0.7680064 v 4.1634029 c 0,0.101054 0,0.181897 0,0.262739 0,0.768007 0.0404,1.071167 0.68717,1.071167 h 1.25306 c 0.46485,0 0.68716,-0.242528 0.68716,-0.707374 0,-0.5659 -0.28295,-0.687164 -1.03074,-0.687164 -0.0606,0 -0.12126,0 -0.1819,0 V 9.0341631 c 0,-1.5764342 0.68717,-2.3646514 1.59665,-2.3646514 0.99032,0 1.07116,0.8690599 1.07116,2.3646514 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path995" />
+        <path
+           d="m 255.20781,9.1150059 c 0.384,-1.6370663 1.65727,-2.6071797 3.39539,-2.6071797 1.59665,0 2.84971,1.0307454 3.03161,2.6071797 z m -0.0404,1.2934841 h 6.6291 c 1.21265,0 1.63707,0 1.63707,-0.9296916 0,-2.2029658 -1.83917,-4.3250888 -4.83036,-4.3250888 -3.05181,0 -5.15372,1.9604374 -5.15372,4.7899348 0,2.9507616 1.8998,4.8303566 4.95162,4.8303566 1.3339,0 2.76886,-0.323372 4.12298,-0.990325 0.58611,-0.282949 0.86906,-0.58611 0.86906,-0.990324 0,-0.363792 -0.30316,-0.646742 -0.68716,-0.646742 -0.88928,0 -2.16255,1.232853 -4.16341,1.232853 -2.04128,0 -3.27413,-1.071167 -3.37518,-2.970973 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path997" />
+        <path
+           d="m 272.90356,13.278409 0.10105,0.606321 c 0.0808,0.424425 0.32337,0.646742 0.68717,0.646742 h 1.55622 c 0.52548,0 0.78822,-0.242528 0.78822,-0.707374 0,-0.58611 -0.34359,-0.687164 -1.17222,-0.687164 h -0.50527 V 8.6703706 c 0,-2.3848621 -1.17222,-3.4964503 -3.69856,-3.4964503 -2.22318,0 -3.67835,0.7073743 -3.67835,1.394538 0,0.5052674 0.30316,0.8286385 0.66696,0.8286385 0.768,0 1.57643,-0.8286385 2.97097,-0.8286385 1.5158,0 2.22318,0.7073743 2.22318,2.2231765 0,0.020211 0,0.040421 0,0.060632 -0.80843,-0.2021069 -1.57644,-0.3031604 -2.34444,-0.3031604 -2.93056,0 -4.56762,1.2328524 -4.56762,3.2741326 0,1.73812 1.29348,2.950762 3.27413,2.950762 1.35412,0 2.62739,-0.525479 3.69856,-1.495592 z m -0.0606,-3.112447 v 1.071167 c 0,0.88927 -1.61686,2.021069 -3.3954,2.021069 -1.15201,0 -1.94023,-0.666953 -1.94023,-1.536012 0,-1.111589 1.15201,-1.8998058 3.25392,-1.8998058 0.72759,0 1.41475,0.1212642 2.08171,0.3435818 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path999" />
+        <path
+           d="m 279.13978,13.136934 c -0.0808,0 -0.14147,0 -0.2021,0 -0.80843,0 -1.15201,0.121264 -1.15201,0.687164 0,0.464846 0.26274,0.707374 0.82864,0.707374 0.26273,0 0.44463,0 0.52547,0 h 1.79876 c 0.74779,0 1.05095,-0.04042 1.05095,-0.707374 0,-0.545689 -0.26274,-0.707374 -0.9499,-0.707374 -0.12127,0 -0.26274,0.02021 -0.42443,0.02021 V 9.0341631 c 0,-1.4753808 1.37433,-2.3646514 2.56676,-2.3646514 1.29349,0 2.02107,0.7882172 2.02107,2.3646514 v 4.1027709 c -0.12126,0 -0.22232,0 -0.32337,0 -0.70737,0 -0.99032,0.141475 -0.99032,0.687164 0,0.646742 0.28295,0.707374 0.99032,0.707374 h 2.18276 c 0.70737,0 1.05095,-0.141475 1.05095,-0.707374 0,-0.505268 -0.30316,-0.687164 -0.92969,-0.687164 -0.14147,0 -0.30316,0 -0.44464,0 V 8.6703706 c 0,-2.2838086 -1.3339,-3.516661 -3.2135,-3.516661 -1.03074,0 -1.94022,0.4446353 -2.89013,1.3339059 v -0.666953 c 0,-0.2425283 -0.18189,-0.4042139 -0.56589,-0.4042139 h -0.9297 c -0.90948,0 -1.35411,0 -1.35411,0.6871637 0,0.5052674 0.28295,0.7073743 0.90948,0.7073743 0.14147,0 0.30316,0 0.44463,0 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path1001" />
+        <path
+           d="m 296.97701,2.9305331 c 0,-0.2829497 -0.16169,-0.4446353 -0.384,-0.4446353 -0.46485,0 -1.11159,0.6265316 -1.8796,1.8795947 -1.17222,1.8998053 -1.73812,3.9006642 -1.73812,6.1238405 0,2.223177 0.5659,4.224035 1.73812,6.123841 0.76801,1.253063 1.41475,1.879594 1.8796,1.879594 0.22231,0 0.384,-0.181896 0.384,-0.464846 0,-0.384003 -0.54569,-1.192431 -1.11159,-2.62739 -0.62653,-1.616856 -0.9499,-3.173079 -0.9499,-4.911199 0,-2.2231763 0.52548,-3.900664 1.23285,-5.5781517 0.42442,-1.0105348 0.82864,-1.6370664 0.82864,-1.9806482 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path1003" />
+        <path
+           d="m 305.11304,18.027922 c 0,0.28295 0.16169,0.464846 0.384,0.464846 0.46485,0 1.11159,-0.626531 1.8796,-1.879594 1.17222,-1.899806 1.75833,-3.900664 1.75833,-6.123841 0,-2.2231763 -0.58611,-4.2240352 -1.75833,-6.1238405 -0.76801,-1.2530631 -1.41475,-1.8795947 -1.8796,-1.8795947 -0.22231,0 -0.384,0.1616856 -0.384,0.4446353 0,0.3840032 0.54569,1.192431 1.11159,2.6273904 0.62653,1.6168556 0.9499,3.1932899 0.9499,4.9314095 0,2.223177 -0.50527,3.900664 -1.21264,5.578152 -0.42443,1.010535 -0.84885,1.616856 -0.84885,1.960437 z"
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';stroke-width:0.56146109"
+           id="path1005" />
+      </g>
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 242.66155,98.421652 H 267.5495 V 86.233682 H 242.66155 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 269.07755,84.705634 H 293.9655 V 72.517682 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 269.07755,98.421652 H 293.9655 V 86.233682 H 269.07755 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-0-4"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 245.30738,180.91592 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-6-1"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 271.72338,167.19992 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-3-6"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 271.72338,180.91592 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-43-4"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.38633,177.74091 h 24.887936 v -12.1879 h -24.887936 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-9-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970344,164.02491 h 24.88796 v -12.18786 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-49-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970344,177.74091 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-5-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.38634,190.44091 h 24.887946 v -12.1879 h -24.887946 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-5-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970344,190.44091 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-7-9"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570344,164.02491 h 24.88795 v -12.18786 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-3-5"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570344,177.74091 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-9-6"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570344,190.44091 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1-1-21"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170344,164.02491 h 24.88795 v -12.18786 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-0"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170344,177.74091 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-9-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170344,190.44091 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.38634,88.896676 h 24.887939 V 76.708687 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970351,75.180636 h 24.88797 V 62.992687 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970351,88.896676 h 24.88797 V 76.708687 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.38635,101.59668 h 24.887949 V 89.408676 h -24.887949 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-6"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970351,101.59668 h 24.88797 V 89.408676 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-1"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570341,75.180636 h 24.88796 V 62.992687 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-8"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570341,88.896676 h 24.88796 V 76.708687 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-7"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570341,101.59668 h 24.88796 V 89.408676 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -113.38634,114.29668 h 24.887939 v -12.188 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2-9-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -86.970351,114.29668 h 24.88797 v -12.188 h -24.88797 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4-0"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -61.570341,114.29668 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6-2"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170341,114.29668 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170341,75.180636 h 24.88796 V 62.992687 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170341,88.896676 h 24.88796 V 76.708687 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -36.170341,101.59668 h 24.88796 V 89.408676 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-43-4-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 63.889263,177.74092 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-9-7-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 90.305262,164.02492 h 24.887958 v -12.1879 H 90.305262 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-49-2-5"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 90.305262,177.74092 h 24.887958 v -12.1879 H 90.305262 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-5-1-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 63.889263,190.44092 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-5-7-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 90.305262,190.44092 h 24.887958 v -12.1879 H 90.305262 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-7-9-8"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 115.70527,164.02492 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-3-5-9"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 115.70527,177.74092 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-9-6-4"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 115.70527,190.44092 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-2-5-6-7-1-9-3"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 142.60199,164.02492 h 24.88792 v -12.1879 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-2-6-0-3-2-7-5"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 142.60199,177.74092 h 24.88792 v -12.1879 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-1-4-33-3-5-6-4"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 142.60199,190.44092 h 24.88792 v -12.1879 h -24.88792 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-6-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 64.637609,88.896667 h 24.88794 v -12.18799 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-0-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 91.053599,75.180627 H 115.94157 V 62.992677 H 91.053599 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-6-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 91.053599,88.896667 H 115.94157 V 76.708677 H 91.053599 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-2-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 64.637599,101.59667 h 24.88795 V 89.408667 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-6-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 91.053599,101.59667 H 115.94157 V 89.408667 H 91.053599 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-1-3"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 116.45361,75.180627 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-8-6"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 116.45361,88.896667 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-7-1"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 116.45361,101.59667 h 24.88796 V 89.408667 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1-9-2"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 64.637609,114.29667 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2-9-2-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 91.053599,114.29667 h 24.887971 v -12.188 H 91.053599 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4-0-3"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 116.45361,114.29667 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6-2-1"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 141.85361,114.29667 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5-1-3-9"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 141.85361,75.180627 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-4"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 141.85361,88.896667 h 24.88796 v -12.18799 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-5-7"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56899637;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 141.85361,101.59667 h 24.88796 V 89.408667 h -24.88796 z" />
+    </g>
+    <flowRoot
+       xml:space="preserve"
+       id="flowRoot1848"
+       style="font-style:normal;font-weight:normal;font-size:40px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none"
+       transform="matrix(0.26458333,0,0,0.26458333,-306.84909,0.94988499)"><flowRegion
+         id="flowRegion1850"><rect
+           id="rect1852"
+           width="78.933067"
+           height="54.952229"
+           x="901.05609"
+           y="1308.1863" /></flowRegion><flowPara
+         id="flowPara1854" /></flowRoot>  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_reduction.svg b/pandas/doc/source/_static/schemas/06_reduction.svg
new file mode 100644
index 0000000000000..6ee808b953f7e
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_reduction.svg
@@ -0,0 +1,222 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="278.65692mm"
+   height="77.216049mm"
+   viewBox="0 0 278.65692 77.216049"
+   version="1.1"
+   id="svg11151"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_reduction.svg">
+  <defs
+     id="defs11145">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="626.80804"
+     inkscape:cy="-104.50802"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata11148">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-3.5465457,-106.44555)">
+    <g
+       id="g921"
+       transform="matrix(0.89981591,0,0,0.89933244,14.313808,14.602194)"
+       style="stroke-width:1.11163712">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 3.80258,132.60554 H 28.69052 V 120.41757 H 3.80258 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,118.88952 H 55.10652 V 106.70157 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,132.60554 H 55.10652 V 120.41757 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 3.80257,145.30554 H 28.69052 V 133.11758 H 3.80257 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,145.30554 H 55.10652 V 133.11758 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,118.88952 H 80.50653 V 106.70157 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,132.60554 H 80.50653 V 120.41757 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,145.30554 H 80.50653 V 133.11758 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 3.80258,158.00555 H 28.69052 V 145.81759 H 3.80258 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-2"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,158.00555 H 55.10652 V 145.81759 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,158.00555 H 80.50653 V 145.81759 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 3.80258,170.70556 H 28.69052 V 158.5176 H 3.80258 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-0"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,170.70556 H 55.10652 V 158.5176 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-9"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,170.70556 H 80.50653 V 158.5176 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 3.80258,183.40557 H 28.69052 V 171.21761 H 3.80258 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-1"
+         style="fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 30.21857,183.40557 H 55.10652 V 171.21761 H 30.21857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-34"
+         style="fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 55.61857,183.40557 H 80.50653 V 171.21761 H 55.61857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01858,158.00555 h 24.88794 V 145.81759 H 81.01858 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01857,118.88952 h 24.88795 V 106.70157 H 81.01857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01858,132.60554 h 24.88794 V 120.41757 H 81.01858 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01857,145.30554 h 24.88795 V 133.11758 H 81.01857 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-89"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01858,170.70556 h 24.88794 V 158.5176 H 81.01858 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-55"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 81.01858,183.40557 h 24.88794 V 171.21761 H 81.01858 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-0"
+         d="m 149.73091,145.06062 h 47.88958"
+         style="fill:none;stroke:#000000;stroke-width:0.44465485;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2)" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-4-6"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 230.64348,144.79755 h 24.88795 v -12.18796 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-6-8"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 257.05948,144.79755 h 24.88795 v -12.18796 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-90-2"
+         style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 230.64348,157.49755 h 24.88795 V 145.3096 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-4-3"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56921214;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 257.05948,157.49755 h 24.88795 V 145.3096 h -24.88795 z" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/06_valuecounts.svg b/pandas/doc/source/_static/schemas/06_valuecounts.svg
new file mode 100644
index 0000000000000..6d7439b45ae6f
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/06_valuecounts.svg
@@ -0,0 +1,269 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="370.33118mm"
+   height="128.24377mm"
+   viewBox="0 0 370.33118 128.24378"
+   version="1.1"
+   id="svg14732"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="06_valuecounts.svg">
+  <defs
+     id="defs14726">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-86"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-2-3-1-7"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-9-2-8-3"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="654.27439"
+     inkscape:cy="146.60032"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata14729">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(72.482839,-75.64002)">
+    <g
+       id="g984"
+       transform="matrix(0.89986154,0,0,0.89959912,11.283883,14.032219)"
+       style="stroke-width:1.11144412">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-5"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,113.59785 h 24.88796 V 101.4099 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-2"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,127.31387 h 24.88796 V 115.1259 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-4"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,140.01387 h 24.88796 v -12.18796 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4-6"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,152.71388 h 24.88796 v -12.18796 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-9-6-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,165.41389 h 24.88796 v -12.18796 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-34-5-6"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -45.810816,178.1139 h 24.88796 v -12.18796 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -72.226805,127.31387 h 24.887939 V 115.1259 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -72.226815,140.01387 h 24.887949 v -12.18796 h -24.887949 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1-0"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -72.226805,152.71388 h 24.887939 v -12.18796 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-3-2-8"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -72.226805,165.41389 h 24.887939 v -12.18796 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-9-5-9"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -72.226805,178.1139 h 24.887939 v -12.18796 h -24.887939 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-0-9-4"
+         d="m 170.15323,139.76895 h 47.88958"
+         style="fill:none;stroke:#000000;stroke-width:0.44457766;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-3)" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-0-9-7-0"
+         d="m 9.345491,139.76895 h 47.88958"
+         style="fill:none;stroke:#000000;stroke-width:0.44457766;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-2-3-1-7)" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-4-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 246.28835,146.36389 h 24.88794 v -12.18797 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-4-7"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 272.70434,132.64787 H 297.5923 V 120.45992 H 272.70434 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-94-33-4-0"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 272.70434,146.36389 H 297.5923 V 134.17592 H 272.70434 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-9-4"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 246.28834,159.06389 h 24.88795 v -12.18796 h -24.88795 z" />
+      <text
+         id="text17627"
+         y="143.61424"
+         x="282.14795"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.29406959"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#333333;fill-opacity:1;stroke-width:0.29406959"
+           y="143.61424"
+           x="282.14795"
+           id="tspan17625"
+           sodipodi:role="line">3</tspan></text>
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-4-9-9-9"
+         style="fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M 272.70434,159.06389 H 297.5923 V 146.87593 H 272.70434 Z" />
+      <text
+         id="text17627-7"
+         y="156.31425"
+         x="282.14795"
+         style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#333333;fill-opacity:1;stroke:none;stroke-width:0.29406959"
+         xml:space="preserve"><tspan
+           style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#333333;fill-opacity:1;stroke-width:0.29406959"
+           y="156.31425"
+           x="282.14795"
+           id="tspan17625-9"
+           sodipodi:role="line">2</tspan></text>
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-7-9-85"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,177.21177 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-3-5-8"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,190.92777 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-9-6-2"
+         style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,203.62777 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-43-4-21"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 85.480629,190.92777 h 24.887941 v -12.1879 H 85.480629 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-5-1-13"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 85.480619,203.62777 h 24.887951 v -12.1879 H 85.480619 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-5-8"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,88.083994 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-6-9-2-4"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,101.80001 h 24.88796 V 89.612044 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-39-9-4-5"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,114.50001 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-1-4-6-0"
+         style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 111.89662,127.20002 h 24.88796 v -12.18795 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-3-3"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 85.48063,101.80001 h 24.88794 V 89.612044 H 85.48063 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-7-6"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 85.48062,114.50001 h 24.88795 V 102.31206 H 85.48062 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-95-1-0-1"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56911337;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 85.48063,127.20002 h 24.88794 V 115.01207 H 85.48063 Z" />
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/07_melt.svg b/pandas/doc/source/_static/schemas/07_melt.svg
new file mode 100644
index 0000000000000..c4551b48c5001
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/07_melt.svg
@@ -0,0 +1,315 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="275.26425mm"
+   height="64.67041mm"
+   viewBox="0 0 275.26425 64.67041"
+   version="1.1"
+   id="svg19055"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="07_melt.svg">
+  <defs
+     id="defs19049">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <clipPath
+       id="clipPath1036"
+       clipPathUnits="userSpaceOnUse">
+      <path
+         inkscape:connector-curvature="0"
+         id="path1034"
+         d="M 0,1080 H 1920 V 0 H 0 Z" />
+    </clipPath>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.7"
+     inkscape:cx="568.82093"
+     inkscape:cy="-47.502894"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     showguides="false" />
+  <metadata
+     id="metadata19052">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(68.457043,-84.703481)">
+    <g
+       id="g1283"
+       transform="translate(13.763216,-3.2373883)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-5"
+         d="m 22.137533,120.23996 h 43.0917"
+         style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-1)" />
+      <g
+         style="stroke-width:1.11179423"
+         transform="matrix(0.89981363,0,0,0.89908051,-6.8328066,15.041111)"
+         id="g1236">
+        <path
+           d="m 104.44721,148.96344 h 24.88794 v -12.1879 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-2-96-9-5-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,148.96344 h 24.88797 v -12.1879 H 130.8632 Z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-7-4-1-9-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,148.96344 h 24.88796 v -12.1879 h -24.88796 z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-2-8-34-5-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,148.96344 24.88796,-12.1879 h -24.88796 z"
+           style="opacity:1;fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-6"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,149.0407 h 24.88796 v -12.18799 z"
+           style="opacity:1;fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 104.44721,110.8635 h 24.88794 V 98.675505 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-6-3-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,97.147455 h 24.88797 V 84.959505 H 130.8632 Z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-40-0-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,110.8635 h 24.88797 V 98.675505 H 130.8632 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-94-33-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 104.4472,123.5635 h 24.88795 v -12.188 H 104.4472 Z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-5-7-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,123.5635 h 24.88797 v -12.188 H 130.8632 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-4-9-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,97.147455 h 24.88796 v -12.18795 h -24.88796 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,110.8635 h 24.88796 V 98.675505 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-6-9-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,123.5635 h 24.88796 v -12.188 h -24.88796 z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-39-9-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 104.44721,136.26349 h 24.88794 V 124.0755 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-95-1-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,136.26349 h 24.88797 V 124.0755 H 130.8632 Z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-2-9-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,136.26349 h 24.88796 V 124.0755 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-1-4-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,97.147455 h 24.88796 v -12.18795 h -24.88796 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-5-1-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,123.5635 24.88796,-12.188 h -24.88796 z"
+           style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3-4-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M 181.66321,110.8635 206.55117,98.675505 H 181.66321 Z"
+           style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-6"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,110.94071 h 24.88796 V 98.752722 Z"
+           style="opacity:1;fill:#47e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,123.64071 h 24.88796 v -12.188 z"
+           style="opacity:1;fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-9"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M 181.66321,136.26349 206.55117,124.0755 H 181.66321 Z"
+           style="opacity:1;fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-58-6-5"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,136.34071 h 24.88796 v -12.18799 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-3"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+      </g>
+      <g
+         style="stroke-width:1.11179423"
+         transform="matrix(0.89981363,0,0,0.89908051,-6.832813,15.041111)"
+         id="g1211">
+        <path
+           d="m -41.785016,123.56348 24.887959,-12.188 h -24.887959 z"
+           style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-3"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.856727,123.64069 h 24.887959 V 111.4527 Z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-7"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -16.385018,123.56348 24.8879597,-12.188 H -16.385018 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3-4-4"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M -16.456728,123.64069 H 8.4312317 V 111.4527 Z"
+           style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-9-0"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.785016,136.26348 24.887959,-12.188 h -24.887959 z"
+           style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-58-6-6"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.856727,136.34069 h 24.887958 V 124.1527 Z"
+           style="opacity:1;fill:#47e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-3-3"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -68.201008,123.56343 h 24.88795 v -12.1879 h -24.88795 z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-6-3-5-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -41.785018,109.84743 h 24.88796 V 97.659523 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-40-0-2-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -68.201018,136.26343 h 24.88796 v -12.1879 h -24.88796 z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-5-7-91-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M -16.385018,109.84743 H 8.5029417 V 97.659523 H -16.385018 Z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-3-5"
+           inkscape:connector-curvature="0" />
+        <g
+           style="stroke-width:1.11179423"
+           id="g1156">
+          <path
+             sodipodi:nodetypes="cccc"
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-5"
+             style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M -16.385018,136.26343 8.5029417,124.07553 H -16.385018 Z" />
+          <path
+             sodipodi:nodetypes="cccc"
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-1-73-1-7-2-8"
+             style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.5692926;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M -16.456728,136.34069 H 8.4312317 V 124.1527 Z" />
+        </g>
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/07_pivot.svg b/pandas/doc/source/_static/schemas/07_pivot.svg
new file mode 100644
index 0000000000000..14b61c5f9a73b
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/07_pivot.svg
@@ -0,0 +1,338 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="272.51166mm"
+   height="64.025223mm"
+   viewBox="0 0 272.51166 64.025223"
+   version="1.1"
+   id="svg19055"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="07_pivot.svg">
+  <defs
+     id="defs19049">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1-2"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-9"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="494.55926"
+     inkscape:cy="32.273014"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     showguides="false" />
+  <metadata
+     id="metadata19052">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(73.058798,-82.073475)">
+    <g
+       id="g1246"
+       transform="matrix(0.98997936,0,0,0.98990821,12.896057,-1.7340547)"
+       style="stroke-width:1.01015842">
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-5"
+         d="M 37.958629,117.00715 H 85.848212"
+         style="fill:none;stroke:#000000;stroke-width:0.40406337;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-1)" />
+      <g
+         style="stroke-width:1.12308824"
+         transform="matrix(0.89981363,0,0,0.89908051,-166.78578,11.765142)"
+         id="g1236">
+        <path
+           d="m 104.44721,148.96344 h 24.88794 v -12.1879 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-5-2-96-9-5-2-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,148.96344 h 24.88797 v -12.1879 H 130.8632 Z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-0-7-4-1-9-9-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,148.96344 h 24.88796 v -12.1879 h -24.88796 z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-5-2-8-34-5-3-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,148.96344 24.88796,-12.1879 h -24.88796 z"
+           style="opacity:1;fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-6-6"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,149.0407 h 24.88796 v -12.18799 z"
+           style="opacity:1;fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-2-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 104.44721,110.8635 h 24.88794 V 98.675505 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-6-3-0-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,97.147455 h 24.88797 V 84.959505 H 130.8632 Z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-40-0-7-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,110.8635 h 24.88797 V 98.675505 H 130.8632 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-94-33-6-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 104.4472,123.5635 h 24.88795 v -12.188 H 104.4472 Z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-5-7-1-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,123.5635 h 24.88797 v -12.188 H 130.8632 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-4-9-0-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,97.147455 h 24.88796 v -12.18795 h -24.88796 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-9-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,110.8635 h 24.88796 V 98.675505 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-6-9-1-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,123.5635 h 24.88796 v -12.188 h -24.88796 z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-39-9-1-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 104.44721,136.26349 h 24.88794 V 124.0755 h -24.88794 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-95-1-2-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 130.8632,136.26349 h 24.88797 V 124.0755 H 130.8632 Z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-2-9-9-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 156.26321,136.26349 h 24.88796 V 124.0755 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-1-4-8-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,97.147455 h 24.88796 v -12.18795 h -24.88796 z"
+           style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-5-1-9-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 181.66321,123.5635 24.88796,-12.188 h -24.88796 z"
+           style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3-4-2-1"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M 181.66321,110.8635 206.55117,98.675505 H 181.66321 Z"
+           style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-6-8"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,110.94071 h 24.88796 V 98.752722 Z"
+           style="opacity:1;fill:#47e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-79"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,123.64071 h 24.88796 v -12.188 z"
+           style="opacity:1;fill:#e50387;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-9-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M 181.66321,136.26349 206.55117,124.0755 H 181.66321 Z"
+           style="opacity:1;fill:#140357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-58-6-5-0"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m 181.5915,136.34071 h 24.88796 v -12.18799 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-3-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+      </g>
+      <g
+         style="stroke-width:1.12308824"
+         transform="matrix(0.89981363,0,0,0.89908051,166.77195,11.765115)"
+         id="g1211">
+        <path
+           d="m -41.785016,123.56348 24.887959,-12.188 h -24.887959 z"
+           style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-3-3"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.856727,123.64069 h 24.887959 V 111.4527 Z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-7-7"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -16.385018,123.56348 24.8879597,-12.188 H -16.385018 Z"
+           style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-3-4-4-5"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M -16.456728,123.64069 H 8.4312317 V 111.4527 Z"
+           style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-9-0-9"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.785016,136.26348 24.887959,-12.188 h -24.887959 z"
+           style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-58-6-6-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -41.856727,136.34069 h 24.887958 V 124.1527 Z"
+           style="opacity:1;fill:#47e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-3-3-2"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="m -68.201008,123.56343 h 24.88795 v -12.1879 h -24.88795 z"
+           style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-6-3-5-2-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -41.785018,109.84743 h 24.88796 V 97.659523 h -24.88796 z"
+           style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-40-0-2-0-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -68.201018,136.26343 h 24.88796 v -12.1879 h -24.88796 z"
+           style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-5-7-91-2-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M -16.385018,109.84743 H 8.5029417 V 97.659523 H -16.385018 Z"
+           style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-3-5-3"
+           inkscape:connector-curvature="0" />
+        <g
+           style="stroke-width:1.12308824"
+           id="g1156">
+          <path
+             sodipodi:nodetypes="cccc"
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-5-6"
+             style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M -16.385018,136.26343 8.5029417,124.07553 H -16.385018 Z" />
+          <path
+             sodipodi:nodetypes="cccc"
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-4-1-73-1-7-2-8-1"
+             style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.57507569;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="M -16.456728,136.34069 H 8.4312317 V 124.1527 Z" />
+        </g>
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/07_pivot_table.svg b/pandas/doc/source/_static/schemas/07_pivot_table.svg
new file mode 100644
index 0000000000000..81ddb8b7f9288
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/07_pivot_table.svg
@@ -0,0 +1,455 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="310.98999mm"
+   height="77.370468mm"
+   viewBox="0 0 310.98998 77.370469"
+   version="1.1"
+   id="svg19055"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="07_pivot_table.svg">
+  <defs
+     id="defs19049">
+    <marker
+       inkscape:stockid="Arrow1Mend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="marker14101"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         id="path14103"
+         d="M 0,0 5,-5 -12.5,0 5,5 Z"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:1.00000003pt;stroke-opacity:1"
+         transform="matrix(-0.4,0,0,-0.4,-4,0)"
+         inkscape:connector-curvature="0" />
+    </marker>
+    <marker
+       inkscape:isstock="true"
+       style="overflow:visible"
+       id="marker14061"
+       refX="0"
+       refY="0"
+       orient="auto"
+       inkscape:stockid="Arrow2Mend">
+      <path
+         transform="scale(-0.6)"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         id="path14063"
+         inkscape:connector-curvature="0" />
+    </marker>
+    <marker
+       inkscape:isstock="true"
+       style="overflow:visible"
+       id="marker13985"
+       refX="0"
+       refY="0"
+       orient="auto"
+       inkscape:stockid="Arrow1Mend"
+       inkscape:collect="always">
+      <path
+         transform="matrix(-0.4,0,0,-0.4,-4,0)"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:1.00000003pt;stroke-opacity:1"
+         d="M 0,0 5,-5 -12.5,0 5,5 Z"
+         id="path13987"
+         inkscape:connector-curvature="0" />
+    </marker>
+    <marker
+       inkscape:isstock="true"
+       style="overflow:visible"
+       id="marker13901"
+       refX="0"
+       refY="0"
+       orient="auto"
+       inkscape:stockid="Arrow2Mend">
+      <path
+         transform="scale(-0.6)"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         id="path13903"
+         inkscape:connector-curvature="0" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-1-3"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-6"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.44945336"
+     inkscape:cx="392.70209"
+     inkscape:cy="-168.6873"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1875"
+     inkscape:window-height="1029"
+     inkscape:window-x="45"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     showguides="false" />
+  <metadata
+     id="metadata19052">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(90.105873,-78.392086)">
+    <g
+       id="g1211"
+       transform="matrix(0.88189511,0,0,0.88119418,197.57988,12.204016)"
+       style="stroke-width:1.13437259">
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-3-3"
+         style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -41.785016,123.56348 24.887959,-12.188 h -24.887959 z" />
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-7-7"
+         style="opacity:1;fill:#48e5ac;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -41.856727,123.64069 h 24.887959 V 111.4527 Z" />
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-3-4-4-5"
+         style="opacity:1;fill:#ffca00;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -16.385018,123.56348 24.8879597,-12.188 H -16.385018 Z" />
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-9-0-9"
+         style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -16.456728,123.64069 H 8.4312317 V 111.4527 Z" />
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-58-6-6-2"
+         style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -41.785016,136.26348 24.887959,-12.188 h -24.887959 z" />
+      <path
+         sodipodi:nodetypes="cccc"
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-73-1-7-3-3-2"
+         style="opacity:1;fill:#47e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -41.856727,136.34069 h 24.887958 V 124.1527 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-6-3-5-2-8"
+         style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -68.201008,123.56343 h 24.88795 v -12.1879 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-40-0-2-0-9"
+         style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -41.785018,109.84743 h 24.88796 V 97.659523 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-5-7-91-2-7"
+         style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -68.201018,136.26343 h 24.88796 v -12.1879 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-6-4-3-5-3"
+         style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -16.385018,109.84743 H 8.5029417 V 97.659523 H -16.385018 Z" />
+      <g
+         id="g1156"
+         style="stroke-width:1.13437259">
+        <path
+           d="M -16.385018,136.26343 8.5029417,124.07553 H -16.385018 Z"
+           style="opacity:1;fill:#150458;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7-5-6"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+        <path
+           d="M -16.456728,136.34069 H 8.4312317 V 124.1527 Z"
+           style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.58085382;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-73-1-7-2-8-1"
+           inkscape:connector-curvature="0"
+           sodipodi:nodetypes="cccc" />
+      </g>
+    </g>
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-1)"
+       d="M 65.052483,117.00647 H 108.14521"
+       id="path6109-2-9-6-9-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -74.300346,105.80648 h 22.395042 V 94.846642 h -22.395042 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-6-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,93.472569 h 22.39507 V 82.512767 h -22.39507 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-17-40-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,105.80648 h 22.39507 V 94.846642 h -22.39507 z"
+       style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-94-33"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -74.300355,117.22673 h 22.395051 v -10.95984 h -22.395051 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-5-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,117.22673 h 22.39507 v -10.95984 h -22.39507 z"
+       style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-4-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,93.472569 H -5.2794326 V 82.512767 H -27.674491 Z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-6-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,105.80648 H -5.2794326 V 94.846642 H -27.674491 Z"
+       style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-6-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,117.22673 H -5.2794326 V 106.26689 H -27.674491 Z"
+       style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-39-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -74.300346,128.64699 h 22.395042 v -10.95985 h -22.395042 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-0-95-1"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,128.64699 h 22.39507 v -10.95985 h -22.39507 z"
+       style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-8-2-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,128.64699 H -5.2794326 V 117.68714 H -27.674491 Z"
+       style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-51-1-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -74.300346,140.06724 h 22.395042 v -10.95985 h -22.395042 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-5-6-3-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,140.06724 h 22.39507 v -10.95985 h -22.39507 z"
+       style="opacity:1;fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-0-6-0-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,140.06724 H -5.2794326 V 129.10739 H -27.674491 Z"
+       style="opacity:1;fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-5-21-9-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -74.300346,151.4875 h 22.395042 v -10.95976 h -22.395042 z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-5-2-96-9-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m -50.530312,151.4875 h 22.39507 v -10.95976 h -22.39507 z"
+       style="opacity:1;fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-0-7-4-1-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -27.674491,151.4875 H -5.2794326 V 140.52774 H -27.674491 Z"
+       style="opacity:1;fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-5-2-8-34-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -4.8186801,93.472569 H 17.576378 V 82.512767 H -4.8186801 Z"
+       style="opacity:1;fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-5-1"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M -4.8186801,117.22673 17.576378,106.26689 H -4.8186801 Z"
+       style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-3-4"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8186801,140.06724 17.576378,129.10739 H -4.8186801 Z"
+       style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-7-2-89-3"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8186801,151.4875 17.576378,140.52774 H -4.8186801 Z"
+       style="opacity:1;fill:#120357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-7-1-29-55-7"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8186801,105.80648 17.576378,94.846642 H -4.8186801 Z"
+       style="opacity:1;fill:#ffc900;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8832082,105.87591 H 17.51185 V 94.916078 Z"
+       style="opacity:1;fill:#45e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8832082,117.29616 H 17.511849 v -10.95983 z"
+       style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-9"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8186801,128.64699 17.576378,117.68714 H -4.8186801 Z"
+       style="opacity:1;fill:#120357;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-0-58-6"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8832082,128.71642 H 17.511849 v -10.95984 z"
+       style="opacity:1;fill:#45e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-3"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8832082,140.13667 H 17.511849 v -10.95984 z"
+       style="opacity:1;fill:#45e5ab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-1"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="M -4.8832082,151.55697 H 17.511849 v -10.95984 z"
+       style="opacity:1;fill:#e70488;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-2"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       style="fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:0.28199998;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#marker13985)"
+       d="M 15.392997,135.10777 C 95.323619,82.428706 110.04977,99.372596 162.49825,112.63178"
+       id="path13592"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cc" />
+    <path
+       style="fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:0.28199998;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#marker14101)"
+       d="M 12.483482,101.05648 C 63.450395,83.124053 112.26633,94.573723 162.49825,112.63178"
+       id="path13592-1"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cc" />
+    <g
+       aria-label="mean(  )"
+       transform="scale(1.0003349,0.99966518)"
+       style="font-style:normal;font-weight:normal;font-size:10.79440498px;line-height:0%;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.99999994px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1"
+       id="text14840">
+      <path
+         d="m 99.273083,91.613281 v -3.40144 q 0,-0.763588 -0.123959,-1.066049 -0.119001,-0.307419 -0.441295,-0.307419 -0.317335,0 -0.51567,0.476004 -0.193376,0.476003 -0.193376,1.289175 v 3.009729 h -0.837964 v -4.21957 q 0,-0.937132 -0.02975,-1.145383 h 0.738796 l 0.02975,0.629712 v 0.238002 h 0.0099 q 0.168585,-0.505753 0.42642,-0.733838 0.257835,-0.233044 0.644588,-0.233044 0.436336,0 0.649546,0.238002 0.218168,0.238002 0.312377,0.733838 h 0.0099 q 0.19833,-0.525587 0.47104,-0.748713 0.27767,-0.223127 0.69913,-0.223127 0.58509,0 0.83797,0.42642 0.25288,0.42642 0.25288,1.462718 v 3.574983 h -0.83301 v -3.40144 q 0,-0.763588 -0.12396,-1.066049 -0.119,-0.307419 -0.44129,-0.307419 -0.32726,0 -0.52063,0.416503 -0.18842,0.416503 -0.18842,1.249509 v 3.108896 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1009" />
+      <path
+         d="m 104.32566,89.119222 q 0,0.902423 0.39667,1.413135 0.40163,0.510712 1.0958,0.510712 0.51071,0 0.8925,-0.218168 0.38676,-0.223127 0.51567,-0.604921 l 0.78343,0.223127 q -0.21817,0.614837 -0.80326,0.942089 -0.58013,0.327253 -1.38834,0.327253 -1.17018,0 -1.79989,-0.72888 -0.62971,-0.72888 -0.62971,-2.087473 0,-1.323884 0.61484,-2.032931 0.61979,-0.714005 1.78501,-0.714005 1.16521,0 1.76518,0.709047 0.59996,0.709046 0.59996,2.142015 v 0.119 z m 1.47263,-2.310599 q -0.66442,0 -1.05117,0.436337 -0.38675,0.431378 -0.41154,1.190008 h 2.89568 q -0.13883,-1.626345 -1.43297,-1.626345 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1011" />
+      <path
+         d="m 114.2771,91.062902 q 0.12892,0 0.29255,-0.03471 v 0.555337 q -0.33717,0.07933 -0.68922,0.07933 -0.49583,0 -0.72392,-0.257835 -0.22312,-0.262794 -0.25287,-0.818131 h -0.0297 q -0.3223,0.599963 -0.76359,0.862756 -0.43634,0.262794 -1.08093,0.262794 -0.78342,0 -1.18009,-0.42642 -0.39667,-0.42642 -0.39667,-1.170175 0,-1.73047 2.2511,-1.755262 l 1.17018,-0.01983 v -0.292544 q 0,-0.649546 -0.2628,-0.932173 -0.26279,-0.287585 -0.83796,-0.287585 -0.58509,0 -0.84292,0.208251 -0.25784,0.208252 -0.30742,0.644588 l -0.93218,-0.08429 q 0.22809,-1.447844 2.09739,-1.447844 0.99168,0 1.48751,0.466087 0.5008,0.461128 0.5008,1.338759 v 2.310599 q 0,0.39667 0.10412,0.599963 0.10413,0.198334 0.39667,0.198334 z m -3.01964,-0.02975 q 0.476,0 0.84292,-0.228085 0.36692,-0.228085 0.57021,-0.609879 0.2033,-0.381794 0.2033,-0.78838 v -0.441295 l -0.94209,0.01983 q -0.58509,0.0099 -0.89251,0.128917 -0.30742,0.119001 -0.48096,0.366919 -0.16859,0.24296 -0.16859,0.649546 0,0.406587 0.21817,0.654505 0.22313,0.247918 0.64955,0.247918 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1013" />
+      <path
+         d="m 119.21068,91.613281 v -3.446065 q 0,-0.674338 -0.26279,-1.00159 -0.25784,-0.327253 -0.82805,-0.327253 -0.61484,0 -1.01151,0.451212 -0.39171,0.446253 -0.39171,1.2148 v 3.108896 h -0.89251 v -4.21957 q 0,-0.937132 -0.0298,-1.145383 h 0.84293 q 0.005,0.02479 0.01,0.133876 0.005,0.109084 0.01,0.252876 0.01,0.138835 0.0198,0.530546 h 0.0149 q 0.52063,-1.016466 1.71559,-1.016466 0.8578,0 1.27926,0.466087 0.42146,0.461128 0.42146,1.423051 v 3.574983 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1015" />
+      <path
+         d="m 123.61867,88.985347 q 0,1.363551 0.42146,2.464308 0.42146,1.100758 1.37347,2.270932 h -0.94209 q -0.95201,-1.170174 -1.36851,-2.265974 -0.41155,-1.100757 -0.41155,-2.479183 0,-1.353635 0.40659,-2.439517 0.40659,-1.090841 1.37347,-2.280849 h 0.94209 q -0.95201,1.170175 -1.37347,2.275891 -0.42146,1.100757 -0.42146,2.454392 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1017" />
+      <path
+         d="m 143.65047,88.97543 q 0,1.388343 -0.41154,2.484142 -0.40659,1.090841 -1.35859,2.261015 h -0.95201 q 0.97184,-1.194966 1.38834,-2.295724 0.4165,-1.105716 0.4165,-2.439516 0,-1.328843 -0.4165,-2.429601 -0.4165,-1.105715 -1.38834,-2.300682 h 0.95201 q 0.96688,1.190008 1.36851,2.270932 0.40162,1.080924 0.40162,2.449434 z"
+         style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:10.15473652px;line-height:1.25;font-family:'Courier New';-inkscape-font-specification:'Courier New';stroke-width:0.99999994px"
+         id="path1019" />
+    </g>
+    <path
+       d="m 126.62955,92.397099 13.40642,-6.560918 h -13.40642 z"
+       style="fill:#ffca00;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.3065289;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-2"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+    <path
+       d="m 126.59093,92.438661 h 13.40641 v -6.560913 z"
+       style="fill:#48e5ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.3065289;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-73-1-7-0"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccc" />
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/08_concat_column.svg b/pandas/doc/source/_static/schemas/08_concat_column.svg
new file mode 100644
index 0000000000000..8c3e92a36d8ef
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/08_concat_column.svg
@@ -0,0 +1,465 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="474.70731mm"
+   height="77.216072mm"
+   viewBox="0 0 474.7073 77.216072"
+   version="1.1"
+   id="svg11623"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="08_concat_column.svg">
+  <defs
+     id="defs11617">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1-0"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="717.85316"
+     inkscape:cy="232.01409"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1875"
+     inkscape:window-height="1056"
+     inkscape:window-x="1965"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata11620">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(129.02778,-98.13007)">
+    <g
+       id="g4921"
+       transform="matrix(0.9,0,0,0.9,10.832587,13.673811)">
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-52"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -128.77175,124.2901 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-07"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,110.574 h 24.88795 V 98.3861 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,124.2901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-64"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -128.77176,136.9901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,136.9901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-5"
+         style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="M -76.95576,110.574 H -52.0678 V 98.3861 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-35"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.95576,124.2901 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.95576,136.9901 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-1"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -128.77175,149.6901 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-00"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,149.6901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-7"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.95576,149.6901 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-0"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -128.77175,162.3901 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-8"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,162.3901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-16"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.95576,162.3901 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-4"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -128.77175,175.0901 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-14"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -102.35576,175.0901 h 24.88795 v -12.188 h -24.88795 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-9"
+         style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -76.95576,175.0901 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-74-52-9"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -23.678541,124.29012 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-17-07-7"
+         style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,110.57401 h 24.88796 V 98.386112 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-1-1-9"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,124.29012 h 24.88796 v -12.188 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-1-64-2"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -23.678561,136.99012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-1-9-8"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,136.99012 h 24.88796 v -12.188 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-6-7-5-9"
+         style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,110.57401 h 24.88796 V 98.386112 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-5-0-35-9"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,124.29012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-2-3-4-1-8"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,136.99012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-0-1-9"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -23.678541,149.69012 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-8-00-3"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,149.69012 h 24.88796 v -12.188 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-51-7-6"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,149.69012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-6-0-3"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -23.678541,162.39012 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-6-8-6"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,162.39012 h 24.88796 v -12.188 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-21-16-0"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,162.39012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-5-2-96-4-4"
+         style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m -23.678541,175.09012 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-0-7-4-14-4"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 2.737439,175.09012 h 24.88796 v -12.188 H 2.737439 Z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-7-8-5-2-8-9-9"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 28.137449,175.09012 h 24.88796 v -12.188 h -24.88796 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-0-8-6"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537459,149.69012 h 24.88793 v -12.188 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-8-8-62-6"
+         style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537449,110.57401 h 24.88794 V 98.386112 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-4-1-2-4"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537459,124.29012 h 24.88793 v -12.188 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-7-5-5-0-5-1"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537449,136.99012 h 24.88794 v -12.188 h -24.88794 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-2-23-9"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537459,162.39012 h 24.88793 v -12.188 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path4891-1-50-8-2-1-9-8-2-7-1-29-2-8"
+         style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+         d="m 53.537459,175.09012 h 24.88793 v -12.188 h -24.88793 z" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-7-6"
+         d="m 110.92608,136.74514 h 47.88959"
+         style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1-0)" />
+      <g
+         transform="translate(-22.985179)"
+         id="g4842">
+        <path
+           d="m 215.50479,124.2901 h 24.88794 v -12.18801 h -24.88794 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-5-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,110.57409 h 24.88795 V 98.386094 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-47-3-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,124.2901 h 24.88795 v -12.18801 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-1-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 215.50477,136.99009 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-4-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,136.99009 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-20-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,110.57409 h 24.88797 V 98.386094 h -24.88797 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-0-1-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,124.2901 h 24.88797 v -12.18801 h -24.88797 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-3-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,136.99009 h 24.88797 v -12.188 h -24.88797 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-0-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 215.50479,149.69009 h 24.88794 V 137.5022 h -24.88794 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-5-0-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,149.69009 h 24.88795 V 137.5022 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-8-4-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,149.69009 h 24.88797 V 137.5022 h -24.88797 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-10-8-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72079,149.69009 h 24.88796 V 137.5022 h -24.88796 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-67-2-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72077,110.57409 h 24.88798 V 98.386094 h -24.88798 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-0-2-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72079,124.2901 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72077,136.99009 h 24.88798 v -12.188 h -24.88798 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-19-24-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,149.69009 h 24.88793 V 137.5022 H 318.1208 Z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,110.57409 h 24.88793 V 98.386094 H 318.1208 Z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-2-5-01-8-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,124.2901 h 24.88793 V 112.10209 H 318.1208 Z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-4-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,136.99009 h 24.88793 v -12.188 H 318.1208 Z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-0-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 215.50479,162.3901 h 24.88794 v -12.18801 h -24.88794 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-8-8-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,162.3901 h 24.88795 v -12.18801 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-5-2-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 215.50477,175.0901 h 24.88795 v -12.18801 h -24.88795 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-0-5-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 241.92077,175.0901 h 24.88795 v -12.18801 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-2-4-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,162.3901 h 24.88797 v -12.18801 h -24.88797 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-2-7-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 267.32078,175.0901 h 24.88797 v -12.18801 h -24.88797 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-8-5-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72079,162.3901 h 24.88796 v -12.18801 h -24.88796 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-5-5-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 292.72077,175.0901 h 24.88798 v -12.18801 h -24.88798 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-19-2-2-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,162.3901 h 24.88793 V 150.20209 H 318.1208 Z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 318.1208,175.0901 h 24.88793 V 162.90209 H 318.1208 Z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-3-6-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,149.69009 h 24.88791 V 137.5022 h -24.88791 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-8-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,110.57409 h 24.88791 V 98.386094 h -24.88791 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-2-5-01-8-0-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,124.2901 h 24.88791 v -12.18801 h -24.88791 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-4-4-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,136.99009 h 24.88791 v -12.188 h -24.88791 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-0-9-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,162.3901 h 24.88791 v -12.18801 h -24.88791 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-0-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 343.52077,175.0901 h 24.88791 v -12.18801 h -24.88791 z"
+           style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-3-6-9-7"
+           inkscape:connector-curvature="0" />
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/08_concat_row.svg b/pandas/doc/source/_static/schemas/08_concat_row.svg
new file mode 100644
index 0000000000000..116afc8f89890
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/08_concat_row.svg
@@ -0,0 +1,392 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="432.81134mm"
+   height="119.66626mm"
+   viewBox="0 0 432.81134 119.66626"
+   version="1.1"
+   id="svg10627"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="08_concat_row.svg">
+  <defs
+     id="defs10621">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.49497475"
+     inkscape:cx="569.23237"
+     inkscape:cy="-49.001989"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1875"
+     inkscape:window-height="1056"
+     inkscape:window-x="1965"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1" />
+  <metadata
+     id="metadata10624">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-2.8750229e-8,-63.297826)">
+    <path
+       d="M 21.896609,92.839632 H 44.292807 V 81.875669 H 21.896609 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,80.501126 H 68.064079 V 69.537164 H 45.66787 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-17-47"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,92.839632 H 68.064079 V 81.875669 H 45.66787 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 21.8966,104.26417 H 44.292807 V 93.300211 H 21.8966 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-6"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,104.26417 H 68.064079 V 93.300211 H 45.66787 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,80.501126 H 90.921078 V 69.537164 H 68.524862 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,92.839632 H 90.921078 V 81.875669 H 68.524862 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-64"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,104.26417 H 90.921078 V 93.300211 H 68.524862 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-37"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 21.896609,115.68872 H 44.292807 V 104.72484 H 21.896609 Z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-0-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,115.68872 H 68.064079 V 104.72484 H 45.66787 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-8-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,115.68872 H 90.921078 V 104.72484 H 68.524862 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-51-10"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381863,115.68872 H 113.77807 V 104.72484 H 91.381863 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-0-67"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381853,80.501126 H 113.77807 V 69.537164 H 91.381853 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381863,92.839632 H 113.77807 V 81.875669 H 91.381863 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381853,104.26417 H 113.77807 V 93.300211 H 91.381853 Z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-19"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,115.68872 h 22.39618 v -10.96388 h -22.39618 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-4-7-85"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,80.501126 h 22.39618 V 69.537164 h -22.39618 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-2-5-01"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,92.839632 h 22.39618 V 81.875669 h -22.39618 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-2-6-59"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,104.26417 h 22.39618 V 93.300211 h -22.39618 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-1-4-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 21.896609,165.30021 H 44.292807 V 154.33623 H 21.896609 Z"
+       style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-5-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,152.96169 H 68.064079 V 141.99773 H 45.66787 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-17-47-9"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,165.30021 H 68.064079 V 154.33623 H 45.66787 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-8-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 21.8966,176.72475 H 44.292807 V 165.76079 H 21.8966 Z"
+       style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-6-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 45.66787,176.72475 H 68.064079 V 165.76079 H 45.66787 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-0-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,152.96169 H 90.921078 V 141.99773 H 68.524862 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-0-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,165.30021 H 90.921078 V 154.33623 H 68.524862 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-64-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 68.524862,176.72475 H 90.921078 V 165.76079 H 68.524862 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-37-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381853,152.96169 H 113.77807 V 141.99773 H 91.381853 Z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-0-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381863,165.30021 H 113.77807 V 154.33623 H 91.381863 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-8-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="M 91.381853,176.72475 H 113.77807 V 165.76079 H 91.381853 Z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-19-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,152.96169 h 22.39618 v -10.96396 h -22.39618 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-2-5-01-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,165.30021 h 22.39618 v -10.96398 h -22.39618 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-2-6-59-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 114.23885,176.72475 h 22.39618 v -10.96396 h -22.39618 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-1-4-7-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 296.17632,111.9331 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-5-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,99.594594 h 22.39621 V 88.630632 h -22.39621 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-17-47-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,111.9331 h 22.39621 v -10.96396 h -22.39621 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-8-1"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 296.17631,123.35764 h 22.3962 v -10.96396 h -22.3962 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-6-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,123.35764 h 22.39621 v -10.96396 h -22.39621 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-0-20"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,99.594594 h 22.39622 V 88.630632 h -22.39622 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-6-7-0-1"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,111.9331 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-64-3"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,123.35764 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-37-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 296.17632,134.78219 h 22.39619 v -10.96388 h -22.39619 z"
+       style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-0-5-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,134.78219 h 22.39621 v -10.96388 h -22.39621 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-8-8-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,134.78219 h 22.39622 v -10.96388 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-7-8-51-10-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,134.78219 h 22.39622 v -10.96388 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-0-67-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,99.594594 h 22.39622 V 88.630632 h -22.39622 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-8-0-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,111.9331 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-8-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,123.35764 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-19-24"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,134.78219 h 22.39619 v -10.96388 h -22.39619 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-8-2-4-7-85-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,99.594594 h 22.39619 V 88.630632 h -22.39619 z"
+       style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-8-2-5-01-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,111.9331 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-2-6-59-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,123.35764 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-1-4-7-0"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 296.17632,146.20673 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-74-5-8-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,146.20673 h 22.39621 v -10.96396 h -22.39621 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-1-8-5-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 296.17631,157.63127 h 22.3962 v -10.96396 h -22.3962 z"
+       style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-1-6-0-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 319.94758,157.63127 h 22.39621 v -10.96396 h -22.39621 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-1-0-2-4"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,146.20673 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-5-0-64-2-7"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 342.80457,157.63127 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-2-3-4-37-8-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,146.20673 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-1-8-5-5"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 365.66156,157.63127 h 22.39622 v -10.96396 h -22.39622 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-0-19-2-2"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,146.20673 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-1-9-4-2-6-59-7-8"
+       inkscape:connector-curvature="0" />
+    <path
+       d="m 388.51855,157.63127 h 22.39619 v -10.96396 h -22.39619 z"
+       style="fill:#969696;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.51204854;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+       id="path4891-1-50-8-2-7-5-5-1-4-7-3-6"
+       inkscape:connector-curvature="0" />
+    <path
+       style="fill:none;stroke:#000000;stroke-width:0.40000001;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1)"
+       d="m 194.55732,123.13725 h 43.09496"
+       id="path6109-2-9-6-9-7"
+       inkscape:connector-curvature="0" />
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/schemas/08_merge_left.svg b/pandas/doc/source/_static/schemas/08_merge_left.svg
new file mode 100644
index 0000000000000..d06fcf2319a09
--- /dev/null
+++ b/pandas/doc/source/_static/schemas/08_merge_left.svg
@@ -0,0 +1,608 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="647.59534mm"
+   height="86.101425mm"
+   viewBox="0 0 647.59534 86.101425"
+   version="1.1"
+   id="svg12741"
+   inkscape:version="0.92.4 (f8dce91, 2019-08-02)"
+   sodipodi:docname="08_merge_left.svg">
+  <defs
+     id="defs12735">
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1-0-4-1"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1-1-3-0"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+    <marker
+       inkscape:stockid="Arrow2Lend"
+       orient="auto"
+       refY="0"
+       refX="0"
+       id="Arrow2Lend-7-6-9-4-6-6-1-0-4"
+       style="overflow:visible"
+       inkscape:isstock="true">
+      <path
+         inkscape:connector-curvature="0"
+         id="path7253-1-4-3-6-0-5-1-1-3"
+         style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1"
+         d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z"
+         transform="matrix(-1.1,0,0,-1.1,-1.1,0)" />
+    </marker>
+  </defs>
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="0.3946822"
+     inkscape:cx="1172.6503"
+     inkscape:cy="52.059868"
+     inkscape:document-units="mm"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0"
+     inkscape:window-width="1551"
+     inkscape:window-height="849"
+     inkscape:window-x="49"
+     inkscape:window-y="27"
+     inkscape:window-maximized="1"
+     showguides="true"
+     inkscape:guide-bbox="true" />
+  <metadata
+     id="metadata12738">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title></dc:title>
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(213.15146,-94.034376)">
+    <g
+       id="g3627"
+       transform="matrix(0.89992087,0,0,0.89940174,11.073374,13.790523)"
+       style="stroke-width:1.11152947">
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-7-6-0"
+         d="M -5.9124488,126.29946 H 41.977141"
+         style="fill:none;stroke:#000000;stroke-width:0.44461179;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1-0-4)" />
+      <path
+         inkscape:connector-curvature="0"
+         id="path6109-2-9-6-9-7-6-0-8"
+         d="m 241.87403,126.29946 h 47.88959"
+         style="fill:none;stroke:#000000;stroke-width:0.44461179;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Lend-7-6-9-4-6-6-1-0-4-1)" />
+      <g
+         transform="translate(12.798637,-6.3500197)"
+         id="g2802"
+         style="stroke-width:1.11152947">
+        <path
+           d="m -173.87807,112.82842 h 24.88797 v -12.188 h -24.88797 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-51"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -173.87807,126.54442 h 24.88797 v -12.188 h -24.88797 z"
+           style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-6-9-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -173.87807,139.24442 h 24.88797 v -12.188 h -24.88797 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-39-9-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -173.87807,151.94442 h 24.88797 v -12.188 h -24.88797 z"
+           style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-1-4-91"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -225.69406,126.54432 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-57-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -199.27807,112.82832 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-6-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -199.27807,126.54432 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-69-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -225.69407,139.24432 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-14-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -199.27807,139.24432 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-3-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -225.69406,151.94432 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-0-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -199.27807,151.94432 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-9-9"
+           inkscape:connector-curvature="0" />
+        <text
+           transform="scale(1.0854314,0.92129267)"
+           id="text56748-3"
+           y="117.81081"
+           x="-156.77634"
+           style="font-style:normal;font-weight:normal;font-size:14.32942963px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#ffffff;fill-opacity:1;stroke:none;stroke-width:0.56161964"
+           xml:space="preserve"><tspan
+             style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.81944466px;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#ffffff;stroke-width:0.56161964"
+             y="117.81081"
+             x="-156.77634"
+             id="tspan56746-6"
+             sodipodi:role="line">key</tspan></text>
+      </g>
+      <g
+         id="g2825"
+         style="stroke-width:1.11152947">
+        <path
+           d="m -124.26805,120.19444 h 24.887959 v -12.1879 h -24.887959 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-57-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -97.852061,106.47844 h 24.88797 V 94.290537 h -24.88797 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-6-61"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -97.852061,120.19444 h 24.88797 v -12.1879 h -24.88797 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-69-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -124.26806,132.89444 h 24.887969 v -12.1879 h -24.887969 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-14-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -97.852061,132.89444 h 24.88797 v -12.1879 h -24.88797 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-3-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -72.452051,106.47844 h 24.88796 V 94.290537 h -24.88796 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-67-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -72.452051,120.19444 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-69-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -72.452051,132.89444 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-76-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -124.26805,145.59444 h 24.887959 v -12.1879 h -24.887959 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-0-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -97.852061,145.59444 h 24.88797 v -12.1879 h -24.88797 z"
+           style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-9-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -72.452051,145.59444 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-07-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -47.052031,145.59444 h 24.88793 v -12.1879 h -24.88793 z"
+           style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-65-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -47.052051,106.47844 h 24.88795 V 94.290537 h -24.88795 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-37-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -47.052031,120.19444 h 24.88793 v -12.1879 h -24.88793 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-69-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -47.052051,132.89444 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-7-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -124.26803,158.2943 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-0-6-24"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -97.85204,158.2943 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-9-7-02"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -72.45204,158.2943 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-07-5-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m -47.05202,158.2943 h 24.88794 v -12.1879 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-65-5-2"
+           inkscape:connector-curvature="0" />
+        <text
+           transform="scale(1.0854314,0.92129267)"
+           id="text56748-3-5"
+           y="110.9184"
+           x="-86.734131"
+           style="font-style:normal;font-weight:normal;font-size:14.32942963px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#ffffff;fill-opacity:1;stroke:none;stroke-width:0.56161964"
+           xml:space="preserve"><tspan
+             style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.81944466px;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#ffffff;stroke-width:0.56161964"
+             y="110.9184"
+             x="-86.734131"
+             id="tspan56746-6-3"
+             sodipodi:role="line">key</tspan></text>
+      </g>
+      <g
+         transform="translate(-45.506234)"
+         id="g3141"
+         style="stroke-width:1.11152947">
+        <path
+           d="m 352.19017,120.1944 h 24.88794 v -12.188 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-5-9-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 378.60616,106.4784 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-47-3-7-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 378.60616,120.1944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-1-7-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 352.19016,132.8944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-1-6-4-6-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 378.60616,132.8944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-1-0-20-8-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 404.00616,106.4784 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-0-1-1-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 404.00616,120.1944 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-3-2-1"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 404.00616,132.8944 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-2-3-4-37-0-0-3"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 352.19017,145.5944 h 24.88794 v -12.1879 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-0-5-0-3-2"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 378.60616,145.5944 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-8-8-4-2-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 404.00616,145.5944 h 24.88796 v -12.1879 h -24.88796 z"
+           style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-7-8-51-10-8-8-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 429.40617,145.5944 h 24.88795 v -12.1879 h -24.88795 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-0-67-2-7-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 429.40616,106.4784 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-8-0-2-4-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 429.40617,120.1944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-0-6-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 429.40616,132.8944 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-0-19-24-3-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 454.80617,145.5944 h 24.88792 v -12.1879 h -24.88792 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-8-2-4-7-85-5-8-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 454.80617,106.4784 h 24.88792 v -12.188 h -24.88792 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-8-2-5-01-8-0-8"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 454.80617,120.1944 h 24.88792 v -12.188 h -24.88792 z"
+           style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-4-4-6"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 454.80617,132.8944 h 24.88792 v -12.188 h -24.88792 z"
+           style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-5-5-1-4-7-0-9-7"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 352.19017,158.2944 h 24.88794 v -12.188 h -24.88794 z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-74-5-8-8-7-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 378.60616,158.2944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#adabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-1-8-5-2-9-4"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 404.00616,158.2944 h 24.88796 v -12.188 h -24.88796 z"
+           style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-5-0-64-2-7-1-0"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 429.40617,158.2944 h 24.88795 v -12.188 h -24.88795 z"
+           style="fill:#000000;fill-opacity:0;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-1-8-5-5-7-5"
+           inkscape:connector-curvature="0" />
+        <path
+           d="m 454.80617,158.2944 h 24.88792 v -12.188 h -24.88792 z"
+           style="fill:#000000;fill-opacity:0;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-1-9-4-2-6-59-7-8-0-9"
+           inkscape:connector-curvature="0" />
+        <text
+           transform="scale(1.0854314,0.92129267)"
+           id="text56748-3-5-1"
+           y="110.9183"
+           x="375.62418"
+           style="font-style:normal;font-weight:normal;font-size:14.32942963px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#ffffff;fill-opacity:1;stroke:none;stroke-width:0.56161964"
+           xml:space="preserve"><tspan
+             style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.81944466px;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#ffffff;stroke-width:0.56161964"
+             y="110.9183"
+             x="375.62418"
+             id="tspan56746-6-3-2"
+             sodipodi:role="line">key</tspan></text>
+      </g>
+      <g
+         transform="translate(2.7116079)"
+         id="g3229"
+         style="stroke-width:1.11152947">
+        <path
+           d="m 108.00186,106.47854 h 24.88797 V 94.290537 h -24.88797 z"
+           style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-6-7-6-4-51-9"
+           inkscape:connector-curvature="0" />
+        <path
+           d="M 82.601856,106.47844 H 107.48982 V 94.290537 H 82.601856 Z"
+           style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+           id="path4891-1-50-8-2-7-17-6-6-2"
+           inkscape:connector-curvature="0" />
+        <text
+           transform="scale(1.0854314,0.92129267)"
+           id="text56748-3-5-5"
+           y="110.91845"
+           x="102.91758"
+           style="font-style:normal;font-weight:normal;font-size:14.32942963px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#ffffff;fill-opacity:1;stroke:none;stroke-width:0.56161964"
+           xml:space="preserve"><tspan
+             style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.81944466px;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#ffffff;stroke-width:0.56161964"
+             y="110.91845"
+             x="102.91758"
+             id="tspan56746-6-3-6"
+             sodipodi:role="line">key</tspan></text>
+        <g
+           id="g3193"
+           transform="translate(-8.5324243)"
+           style="stroke-width:1.11152947">
+          <g
+             transform="translate(-27.019344,-1.530803)"
+             id="g2722"
+             style="stroke-width:1.11152947">
+            <path
+               d="m 182.77459,132.3093 h 24.88797 v -12.1879 h -24.88797 z"
+               style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-9-8-8-9-7-3"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 208.1746,132.3093 h 24.88796 V 120.1214 H 208.1746 Z"
+               style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-7-8-51-07-5-0"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 233.57462,132.3093 h 24.88793 v -12.1879 h -24.88793 z"
+               style="fill:#fc6cc0;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-9-8-2-0-65-5-8"
+               inkscape:connector-curvature="0" />
+          </g>
+          <g
+             transform="translate(-27.019344,-0.0159648)"
+             id="g2730"
+             style="stroke-width:1.11152947">
+            <path
+               d="m 182.77459,148.9951 h 24.88797 v -12.1879 h -24.88797 z"
+               style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-9-1-69-5-6"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 182.77459,161.6951 h 24.88797 v -12.1879 h -24.88797 z"
+               style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-7-5-1-3-4-3"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 208.1746,148.9951 h 24.88796 V 136.8072 H 208.1746 Z"
+               style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-5-0-69-6-1"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 208.1746,161.6951 h 24.88796 V 149.5072 H 208.1746 Z"
+               style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-2-3-4-76-9-7"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 233.57462,148.9951 h 24.88793 v -12.1879 h -24.88793 z"
+               style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-1-9-4-1-69-8-9"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 233.5746,161.6951 h 24.88795 V 149.5072 H 233.5746 Z"
+               style="fill:#e70488;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-7-5-5-0-7-5-0"
+               inkscape:connector-curvature="0" />
+          </g>
+          <g
+             transform="translate(-27.019344,3.400837)"
+             id="g2752"
+             style="stroke-width:1.11152947">
+            <path
+               d="m 182.77459,103.0776 h 24.88797 V 90.8897 h -24.88797 z"
+               style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-7-17-6-61-5"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 208.1746,103.0776 h 24.88796 V 90.8897 H 208.1746 Z"
+               style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-6-7-67-6-4"
+               inkscape:connector-curvature="0" />
+            <path
+               d="m 233.5746,103.0776 h 24.88795 V 90.8897 H 233.5746 Z"
+               style="fill:#646464;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+               id="path4891-1-50-8-2-7-8-8-37-3-2"
+               inkscape:connector-curvature="0" />
+            <text
+               transform="scale(1.0854314,0.92129267)"
+               id="text56748-3-5-2"
+               y="107.22701"
+               x="171.80515"
+               style="font-style:normal;font-weight:normal;font-size:14.32942963px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#ffffff;fill-opacity:1;stroke:none;stroke-width:0.56161964"
+               xml:space="preserve"><tspan
+                 style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:8.81944466px;font-family:'Courier 10 Pitch';-inkscape-font-specification:'Courier 10 Pitch';fill:#ffffff;stroke-width:0.56161964"
+                 y="107.22701"
+                 x="171.80515"
+                 id="tspan56746-6-3-9"
+                 sodipodi:role="line">key</tspan></text>
+          </g>
+        </g>
+        <g
+           id="g2745"
+           transform="translate(-27.019344,3.400837)"
+           style="stroke-width:1.11152947">
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-5-0-6-9-6-7"
+             style="fill:#8b6ff9;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 135.0212,127.37776 h 24.88797 v -12.188 H 135.0212 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-74-57-1-3"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.205209,127.37766 h 24.887951 v -12.1879 H 83.205209 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-1-69-0-6"
+             style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.6212,127.37766 h 24.88796 v -12.1879 H 109.6212 Z" />
+        </g>
+        <g
+           id="g2740"
+           transform="translate(-27.019339,2.9268152)"
+           style="stroke-width:1.11152947">
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-3-4-39-9-0-0"
+             style="fill:#150458;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 135.0212,152.40237 h 24.88797 v -12.188 H 135.0212 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-2-1-14-9-0"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.205199,152.40227 h 24.887961 v -12.1879 H 83.205199 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-7-5-1-3-1-6"
+             style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.6212,152.40227 h 24.88796 v -12.1879 H 109.6212 Z" />
+        </g>
+        <g
+           id="g2735"
+           transform="translate(-27.019344,-0.865376)"
+           style="stroke-width:1.11152947">
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-7-8-51-1-4-91-9"
+             style="fill:#2808ac;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 135.0212,180.74515 h 24.88797 v -12.188 H 135.0212 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-7-0-0-0-2"
+             style="fill:#afabab;fill-opacity:1;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 83.205209,180.74505 h 24.887951 v -12.1879 H 83.205209 Z" />
+          <path
+             inkscape:connector-curvature="0"
+             id="path4891-1-50-8-2-1-9-8-8-9-9-6"
+             style="fill:#afabab;fill-opacity:0.39215686;fill-rule:evenodd;stroke:#ffffff;stroke-width:0.56915706;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"
+             d="m 109.6212,180.74505 h 24.88796 v -12.1879 H 109.6212 Z" />
+        </g>
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/doc/source/_static/spreadsheets/conditional.png b/pandas/doc/source/_static/spreadsheets/conditional.png
new file mode 100644
index 0000000000000..d518ff19dc760
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/conditional.png differ
diff --git a/pandas/doc/source/_static/spreadsheets/filter.png b/pandas/doc/source/_static/spreadsheets/filter.png
new file mode 100644
index 0000000000000..b4c929793ca44
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/filter.png differ
diff --git a/pandas/doc/source/_static/spreadsheets/find.png b/pandas/doc/source/_static/spreadsheets/find.png
new file mode 100644
index 0000000000000..223b2e6fc762f
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/find.png differ
diff --git a/pandas/doc/source/_static/spreadsheets/logo_excel.svg b/pandas/doc/source/_static/spreadsheets/logo_excel.svg
new file mode 100644
index 0000000000000..ffb25108df67c
--- /dev/null
+++ b/pandas/doc/source/_static/spreadsheets/logo_excel.svg
@@ -0,0 +1,27 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Generator: Adobe Illustrator 23.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Livello_1" xmlns:x="http://ns.adobe.com/Extensibility/1.0/" xmlns:i="http://ns.adobe.com/AdobeIllustrator/10.0/" xmlns:graph="http://ns.adobe.com/Graphs/1.0/" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px" viewBox="0 0 2289.75 2130" enable-background="new 0 0 2289.75 2130" xml:space="preserve">
+<metadata>
+	<sfw xmlns="http://ns.adobe.com/SaveForWeb/1.0/">
+		<slices/>
+		<sliceSourceBounds bottomLeftOrigin="true" height="2130" width="2289.75" x="-1147.5" y="-1041"/>
+	</sfw>
+</metadata>
+<path fill="#185C37" d="M1437.75,1011.75L532.5,852v1180.393c0,53.907,43.7,97.607,97.607,97.607l0,0h1562.036  c53.907,0,97.607-43.7,97.607-97.607l0,0V1597.5L1437.75,1011.75z"/>
+<path fill="#21A366" d="M1437.75,0H630.107C576.2,0,532.5,43.7,532.5,97.607c0,0,0,0,0,0V532.5l905.25,532.5L1917,1224.75  L2289.75,1065V532.5L1437.75,0z"/>
+<path fill="#107C41" d="M532.5,532.5h905.25V1065H532.5V532.5z"/>
+<path opacity="0.1" enable-background="new    " d="M1180.393,426H532.5v1331.25h647.893c53.834-0.175,97.432-43.773,97.607-97.607  V523.607C1277.825,469.773,1234.227,426.175,1180.393,426z"/>
+<path opacity="0.2" enable-background="new    " d="M1127.143,479.25H532.5V1810.5h594.643  c53.834-0.175,97.432-43.773,97.607-97.607V576.857C1224.575,523.023,1180.977,479.425,1127.143,479.25z"/>
+<path opacity="0.2" enable-background="new    " d="M1127.143,479.25H532.5V1704h594.643c53.834-0.175,97.432-43.773,97.607-97.607  V576.857C1224.575,523.023,1180.977,479.425,1127.143,479.25z"/>
+<path opacity="0.2" enable-background="new    " d="M1073.893,479.25H532.5V1704h541.393c53.834-0.175,97.432-43.773,97.607-97.607  V576.857C1171.325,523.023,1127.727,479.425,1073.893,479.25z"/>
+<linearGradient id="SVGID_1_" gradientUnits="userSpaceOnUse" x1="203.5132" y1="1729.0183" x2="967.9868" y2="404.9817" gradientTransform="matrix(1 0 0 -1 0 2132)">
+	<stop offset="0" style="stop-color:#18884F"/>
+	<stop offset="0.5" style="stop-color:#117E43"/>
+	<stop offset="1" style="stop-color:#0B6631"/>
+</linearGradient>
+<path fill="url(#SVGID_1_)" d="M97.607,479.25h976.285c53.907,0,97.607,43.7,97.607,97.607v976.285  c0,53.907-43.7,97.607-97.607,97.607H97.607C43.7,1650.75,0,1607.05,0,1553.143V576.857C0,522.95,43.7,479.25,97.607,479.25z"/>
+<path fill="#FFFFFF" d="M302.3,1382.264l205.332-318.169L319.5,747.683h151.336l102.666,202.35  c9.479,19.223,15.975,33.494,19.49,42.919h1.331c6.745-15.336,13.845-30.228,21.3-44.677L725.371,747.79h138.929l-192.925,314.548  L869.2,1382.263H721.378L602.79,1160.158c-5.586-9.45-10.326-19.376-14.164-29.66h-1.757c-3.474,10.075-8.083,19.722-13.739,28.755  l-122.102,223.011H302.3z"/>
+<path fill="#33C481" d="M2192.143,0H1437.75v532.5h852V97.607C2289.75,43.7,2246.05,0,2192.143,0L2192.143,0z"/>
+<path fill="#107C41" d="M1437.75,1065h852v532.5h-852V1065z"/>
+</svg>
\ No newline at end of file
diff --git a/pandas/doc/source/_static/spreadsheets/pivot.png b/pandas/doc/source/_static/spreadsheets/pivot.png
new file mode 100644
index 0000000000000..beacc90bc313e
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/pivot.png differ
diff --git a/pandas/doc/source/_static/spreadsheets/sort.png b/pandas/doc/source/_static/spreadsheets/sort.png
new file mode 100644
index 0000000000000..253f2f3bfb9ba
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/sort.png differ
diff --git a/pandas/doc/source/_static/spreadsheets/vlookup.png b/pandas/doc/source/_static/spreadsheets/vlookup.png
new file mode 100644
index 0000000000000..e96da01da1eeb
Binary files /dev/null and b/pandas/doc/source/_static/spreadsheets/vlookup.png differ
diff --git a/pandas/doc/source/_static/style-excel.png b/pandas/doc/source/_static/style-excel.png
new file mode 100644
index 0000000000000..f946949e8bcf9
Binary files /dev/null and b/pandas/doc/source/_static/style-excel.png differ
diff --git a/pandas/doc/source/_static/style/appmaphead1.png b/pandas/doc/source/_static/style/appmaphead1.png
new file mode 100644
index 0000000000000..905bcaa63e900
Binary files /dev/null and b/pandas/doc/source/_static/style/appmaphead1.png differ
diff --git a/pandas/doc/source/_static/style/appmaphead2.png b/pandas/doc/source/_static/style/appmaphead2.png
new file mode 100644
index 0000000000000..9adde61908378
Binary files /dev/null and b/pandas/doc/source/_static/style/appmaphead2.png differ
diff --git a/pandas/doc/source/_static/style/bg_ax0.png b/pandas/doc/source/_static/style/bg_ax0.png
new file mode 100644
index 0000000000000..1767d34136a02
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_ax0.png differ
diff --git a/pandas/doc/source/_static/style/bg_axNone.png b/pandas/doc/source/_static/style/bg_axNone.png
new file mode 100644
index 0000000000000..8882c6f689773
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_axNone.png differ
diff --git a/pandas/doc/source/_static/style/bg_axNone_gmap.png b/pandas/doc/source/_static/style/bg_axNone_gmap.png
new file mode 100644
index 0000000000000..bdd2b55e8c6b4
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_axNone_gmap.png differ
diff --git a/pandas/doc/source/_static/style/bg_axNone_lowhigh.png b/pandas/doc/source/_static/style/bg_axNone_lowhigh.png
new file mode 100644
index 0000000000000..c37a707e73692
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_axNone_lowhigh.png differ
diff --git a/pandas/doc/source/_static/style/bg_axNone_vminvmax.png b/pandas/doc/source/_static/style/bg_axNone_vminvmax.png
new file mode 100644
index 0000000000000..4ca958de15ec3
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_axNone_vminvmax.png differ
diff --git a/pandas/doc/source/_static/style/bg_gmap.png b/pandas/doc/source/_static/style/bg_gmap.png
new file mode 100644
index 0000000000000..039ff6b78958e
Binary files /dev/null and b/pandas/doc/source/_static/style/bg_gmap.png differ
diff --git a/pandas/doc/source/_static/style/df_pipe.png b/pandas/doc/source/_static/style/df_pipe.png
new file mode 100644
index 0000000000000..071a481ad5acc
Binary files /dev/null and b/pandas/doc/source/_static/style/df_pipe.png differ
diff --git a/pandas/doc/source/_static/style/df_pipe_applydata.png b/pandas/doc/source/_static/style/df_pipe_applydata.png
new file mode 100644
index 0000000000000..a2d5aa514e311
Binary files /dev/null and b/pandas/doc/source/_static/style/df_pipe_applydata.png differ
diff --git a/pandas/doc/source/_static/style/df_pipe_applymap.png b/pandas/doc/source/_static/style/df_pipe_applymap.png
new file mode 100644
index 0000000000000..cd493c78452ef
Binary files /dev/null and b/pandas/doc/source/_static/style/df_pipe_applymap.png differ
diff --git a/pandas/doc/source/_static/style/df_pipe_hl.png b/pandas/doc/source/_static/style/df_pipe_hl.png
new file mode 100644
index 0000000000000..2238a55ab1ce3
Binary files /dev/null and b/pandas/doc/source/_static/style/df_pipe_hl.png differ
diff --git a/pandas/doc/source/_static/style/df_pipe_hl2.png b/pandas/doc/source/_static/style/df_pipe_hl2.png
new file mode 100644
index 0000000000000..7025a7c373d92
Binary files /dev/null and b/pandas/doc/source/_static/style/df_pipe_hl2.png differ
diff --git a/pandas/doc/source/_static/style/footer_extended.png b/pandas/doc/source/_static/style/footer_extended.png
new file mode 100644
index 0000000000000..3699d61ad4346
Binary files /dev/null and b/pandas/doc/source/_static/style/footer_extended.png differ
diff --git a/pandas/doc/source/_static/style/footer_simple.png b/pandas/doc/source/_static/style/footer_simple.png
new file mode 100644
index 0000000000000..56dc3c09cc700
Binary files /dev/null and b/pandas/doc/source/_static/style/footer_simple.png differ
diff --git a/pandas/doc/source/_static/style/format_excel_css.png b/pandas/doc/source/_static/style/format_excel_css.png
new file mode 100644
index 0000000000000..0bd4662c3f2d0
Binary files /dev/null and b/pandas/doc/source/_static/style/format_excel_css.png differ
diff --git a/pandas/doc/source/_static/style/hbetw_axNone.png b/pandas/doc/source/_static/style/hbetw_axNone.png
new file mode 100644
index 0000000000000..2918131b40bde
Binary files /dev/null and b/pandas/doc/source/_static/style/hbetw_axNone.png differ
diff --git a/pandas/doc/source/_static/style/hbetw_basic.png b/pandas/doc/source/_static/style/hbetw_basic.png
new file mode 100644
index 0000000000000..1d8e015aec37f
Binary files /dev/null and b/pandas/doc/source/_static/style/hbetw_basic.png differ
diff --git a/pandas/doc/source/_static/style/hbetw_props.png b/pandas/doc/source/_static/style/hbetw_props.png
new file mode 100644
index 0000000000000..56bbe8479d564
Binary files /dev/null and b/pandas/doc/source/_static/style/hbetw_props.png differ
diff --git a/pandas/doc/source/_static/style/hbetw_seq.png b/pandas/doc/source/_static/style/hbetw_seq.png
new file mode 100644
index 0000000000000..0fc3108a7968c
Binary files /dev/null and b/pandas/doc/source/_static/style/hbetw_seq.png differ
diff --git a/pandas/doc/source/_static/style/hq_ax1.png b/pandas/doc/source/_static/style/hq_ax1.png
new file mode 100644
index 0000000000000..95d840b7c8f99
Binary files /dev/null and b/pandas/doc/source/_static/style/hq_ax1.png differ
diff --git a/pandas/doc/source/_static/style/hq_axNone.png b/pandas/doc/source/_static/style/hq_axNone.png
new file mode 100644
index 0000000000000..40a33b194e640
Binary files /dev/null and b/pandas/doc/source/_static/style/hq_axNone.png differ
diff --git a/pandas/doc/source/_static/style/hq_props.png b/pandas/doc/source/_static/style/hq_props.png
new file mode 100644
index 0000000000000..1f11749096690
Binary files /dev/null and b/pandas/doc/source/_static/style/hq_props.png differ
diff --git a/pandas/doc/source/_static/style/latex_1.png b/pandas/doc/source/_static/style/latex_1.png
new file mode 100644
index 0000000000000..8b901878a0ec9
Binary files /dev/null and b/pandas/doc/source/_static/style/latex_1.png differ
diff --git a/pandas/doc/source/_static/style/latex_2.png b/pandas/doc/source/_static/style/latex_2.png
new file mode 100644
index 0000000000000..7d6baa681575e
Binary files /dev/null and b/pandas/doc/source/_static/style/latex_2.png differ
diff --git a/pandas/doc/source/_static/style/latex_stocks.png b/pandas/doc/source/_static/style/latex_stocks.png
new file mode 100644
index 0000000000000..c8906c33b810b
Binary files /dev/null and b/pandas/doc/source/_static/style/latex_stocks.png differ
diff --git a/pandas/doc/source/_static/style/latex_stocks_html.png b/pandas/doc/source/_static/style/latex_stocks_html.png
new file mode 100644
index 0000000000000..11b30faddf47c
Binary files /dev/null and b/pandas/doc/source/_static/style/latex_stocks_html.png differ
diff --git a/pandas/doc/source/_static/style/tg_ax0.png b/pandas/doc/source/_static/style/tg_ax0.png
new file mode 100644
index 0000000000000..3460329352282
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_ax0.png differ
diff --git a/pandas/doc/source/_static/style/tg_axNone.png b/pandas/doc/source/_static/style/tg_axNone.png
new file mode 100644
index 0000000000000..00357f7eb016b
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_axNone.png differ
diff --git a/pandas/doc/source/_static/style/tg_axNone_gmap.png b/pandas/doc/source/_static/style/tg_axNone_gmap.png
new file mode 100644
index 0000000000000..d06a4b244a23d
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_axNone_gmap.png differ
diff --git a/pandas/doc/source/_static/style/tg_axNone_lowhigh.png b/pandas/doc/source/_static/style/tg_axNone_lowhigh.png
new file mode 100644
index 0000000000000..bc3fb16ee8e40
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_axNone_lowhigh.png differ
diff --git a/pandas/doc/source/_static/style/tg_axNone_vminvmax.png b/pandas/doc/source/_static/style/tg_axNone_vminvmax.png
new file mode 100644
index 0000000000000..42579c2840fb9
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_axNone_vminvmax.png differ
diff --git a/pandas/doc/source/_static/style/tg_gmap.png b/pandas/doc/source/_static/style/tg_gmap.png
new file mode 100644
index 0000000000000..fb73529544180
Binary files /dev/null and b/pandas/doc/source/_static/style/tg_gmap.png differ
diff --git a/pandas/doc/source/_static/trunc_after.png b/pandas/doc/source/_static/trunc_after.png
new file mode 100644
index 0000000000000..950690de8d1ee
Binary files /dev/null and b/pandas/doc/source/_static/trunc_after.png differ
diff --git a/pandas/doc/source/_static/trunc_before.png b/pandas/doc/source/_static/trunc_before.png
new file mode 100644
index 0000000000000..36ac203422e76
Binary files /dev/null and b/pandas/doc/source/_static/trunc_before.png differ
diff --git a/pandas/doc/source/_static/whatsnew_assign.png b/pandas/doc/source/_static/whatsnew_assign.png
new file mode 100644
index 0000000000000..84546cadf0276
Binary files /dev/null and b/pandas/doc/source/_static/whatsnew_assign.png differ
diff --git a/pandas/doc/source/_static/whatsnew_plot_submethods.png b/pandas/doc/source/_static/whatsnew_plot_submethods.png
new file mode 100644
index 0000000000000..c321a0f02e72b
Binary files /dev/null and b/pandas/doc/source/_static/whatsnew_plot_submethods.png differ
diff --git a/pandas/doc/source/conf.py b/pandas/doc/source/conf.py
new file mode 100644
index 0000000000000..677ee6274b093
--- /dev/null
+++ b/pandas/doc/source/conf.py
@@ -0,0 +1,850 @@
+#
+# pandas documentation build configuration file, created by
+#
+# This file is execfile()d with the current directory set to its containing
+# dir.
+#
+# Note that not all possible configuration values are present in this
+# autogenerated file.
+#
+# All configuration values have a default; values that are commented out
+# serve to show the default.
+from datetime import datetime
+import importlib
+import inspect
+import logging
+import os
+import re
+import sys
+import warnings
+
+import jinja2
+from numpydoc.docscrape import NumpyDocString
+from sphinx.ext.autosummary import _import_by_name
+
+logger = logging.getLogger(__name__)
+
+# https://github.com/sphinx-doc/sphinx/pull/2325/files
+# Workaround for sphinx-build recursion limit overflow:
+# pickle.dump(doctree, f, pickle.HIGHEST_PROTOCOL)
+#  RuntimeError: maximum recursion depth exceeded while pickling an object
+#
+# Python's default allowed recursion depth is 1000.
+sys.setrecursionlimit(5000)
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+# sys.path.append(os.path.abspath('.'))
+sys.path.insert(0, os.path.abspath("../sphinxext"))
+sys.path.extend(
+    [
+        # numpy standard doc extensions
+        os.path.join(os.path.dirname(__file__), "..", "../..", "sphinxext")
+    ]
+)
+
+# -- General configuration -----------------------------------------------
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
+# sphinxext.
+
+extensions = [
+    "contributors",  # custom pandas extension
+    "IPython.sphinxext.ipython_directive",
+    "IPython.sphinxext.ipython_console_highlighting",
+    "matplotlib.sphinxext.plot_directive",
+    "numpydoc",
+    "sphinx_copybutton",
+    "sphinx_design",
+    "sphinx.ext.autodoc",
+    "sphinx.ext.autosummary",
+    "sphinx.ext.coverage",
+    "sphinx.ext.doctest",
+    "sphinx.ext.extlinks",
+    "sphinx.ext.ifconfig",
+    "sphinx.ext.intersphinx",
+    "sphinx.ext.linkcode",
+    "sphinx.ext.mathjax",
+    "sphinx.ext.todo",
+    "nbsphinx",
+]
+
+exclude_patterns = [
+    "**.ipynb_checkpoints",
+    # to ensure that include files (partial pages) aren't built, exclude them
+    # https://github.com/sphinx-doc/sphinx/issues/1965#issuecomment-124732907
+    "**/includes/**",
+]
+try:
+    import nbconvert
+except ImportError:
+    logger.warning("nbconvert not installed. Skipping notebooks.")
+    exclude_patterns.append("**/*.ipynb")
+else:
+    try:
+        nbconvert.utils.pandoc.get_pandoc_version()
+    except nbconvert.utils.pandoc.PandocMissing:
+        logger.warning("Pandoc not installed. Skipping notebooks.")
+        exclude_patterns.append("**/*.ipynb")
+
+# sphinx_pattern can be '-api' to exclude the API pages,
+# the path to a file, or a Python object
+# (e.g. '10min.rst' or 'pandas.DataFrame.head')
+source_path = os.path.dirname(os.path.abspath(__file__))
+pattern = os.environ.get("SPHINX_PATTERN")
+single_doc = pattern is not None and pattern not in ("-api", "whatsnew")
+include_api = pattern is None or pattern == "whatsnew"
+if pattern:
+    for dirname, dirs, fnames in os.walk(source_path):
+        reldir = os.path.relpath(dirname, source_path)
+        for fname in fnames:
+            if os.path.splitext(fname)[-1] in (".rst", ".ipynb"):
+                rel_fname = os.path.relpath(os.path.join(dirname, fname), source_path)
+
+                if rel_fname == "index.rst" and os.path.abspath(dirname) == source_path:
+                    continue
+                if pattern == "-api" and reldir.startswith("reference"):
+                    exclude_patterns.append(rel_fname)
+                elif (
+                    pattern == "whatsnew"
+                    and not reldir.startswith("reference")
+                    and reldir != "whatsnew"
+                ):
+                    exclude_patterns.append(rel_fname)
+                elif single_doc and rel_fname != pattern:
+                    exclude_patterns.append(rel_fname)
+
+with open(os.path.join(source_path, "index.rst.template"), encoding="utf-8") as f:
+    t = jinja2.Template(f.read())
+with open(os.path.join(source_path, "index.rst"), "w", encoding="utf-8") as f:
+    f.write(
+        t.render(
+            include_api=include_api,
+            single_doc=(pattern if single_doc else None),
+        )
+    )
+autosummary_generate = True if include_api else ["index"]
+autodoc_typehints = "none"
+
+# numpydoc
+numpydoc_show_class_members = False
+numpydoc_show_inherited_class_members = False
+numpydoc_attributes_as_param_list = False
+
+# matplotlib plot directive
+plot_include_source = True
+plot_formats = [("png", 90)]
+plot_html_show_formats = False
+plot_html_show_source_link = False
+plot_pre_code = """import numpy as np
+import pandas as pd"""
+
+# nbsphinx do not use requirejs (breaks bootstrap)
+nbsphinx_requirejs_path = ""
+
+# https://sphinx-toggleprompt.readthedocs.io/en/stable/#offset
+toggleprompt_offset_right = 35
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ["../_templates"]
+
+# The suffix of source filenames.
+source_suffix = [".rst"]
+
+# The encoding of source files.
+source_encoding = "utf-8"
+
+# The master toctree document.
+master_doc = "index"
+
+# General information about the project.
+project = "pandas"
+# We have our custom "pandas_footer.html" template, using copyright for the current year
+copyright = f"{datetime.now().year},"
+
+# The version info for the project you're documenting, acts as replacement for
+# |version| and |release|, also used in various other places throughout the
+# built documents.
+#
+# The short X.Y version.
+import pandas  # isort:skip
+
+# version = '%s r%s' % (pandas.__version__, svn_version())
+version = str(pandas.__version__)
+
+# The full version, including alpha/beta/rc tags.
+release = version
+
+# The language for content autogenerated by Sphinx. Refer to documentation
+# for a list of supported languages.
+language = "en"
+
+# There are two options for replacing |today|: either, you set today to some
+# non-false value, then it is used:
+# today = ''
+# Else, today_fmt is used as the format for a strftime call.
+# today_fmt = '%B %d, %Y'
+
+# List of documents that shouldn't be included in the build.
+# unused_docs = []
+
+# List of directories, relative to source directory, that shouldn't be searched
+# for source files.
+exclude_trees = []
+
+# The reST default role (used for this markup: `text`) to use for all
+# documents. default_role = None
+
+# If true, '()' will be appended to :func: etc. cross-reference text.
+# add_function_parentheses = True
+
+# If true, the current module name will be prepended to all description
+# unit titles (such as .. function::).
+# add_module_names = True
+
+# If true, sectionauthor and moduleauthor directives will be shown in the
+# output. They are ignored by default.
+# show_authors = False
+
+# The name of the Pygments (syntax highlighting) style to use.
+pygments_style = "sphinx"
+
+# A list of ignored prefixes for module index sorting.
+# modindex_common_prefix = []
+
+
+# -- Options for HTML output ---------------------------------------------
+
+# The theme to use for HTML and HTML Help pages.  Major themes that come with
+# Sphinx are currently 'default' and 'sphinxdoc'.
+html_theme = "pydata_sphinx_theme"
+
+# The style sheet to use for HTML and HTML Help pages. A file of that name
+# must exist either in Sphinx' static/ path, or in one of the custom paths
+# given in html_static_path.
+# html_style = 'statsmodels.css'
+
+# Theme options are theme-specific and customize the look and feel of a theme
+# further.  For a list of options available for each theme, see the
+# documentation.
+
+if ".dev" in version:
+    switcher_version = "dev"
+elif "rc" in version:
+    switcher_version = version.split("rc", maxsplit=1)[0] + " (rc)"
+else:
+    # only keep major.minor version number to match versions.json
+    switcher_version = ".".join(version.split(".")[:2])
+
+html_theme_options = {
+    "external_links": [],
+    "footer_start": ["pandas_footer", "sphinx-version"],
+    "github_url": "https://github.com/pandas-dev/pandas",
+    "analytics": {
+        "plausible_analytics_domain": "pandas.pydata.org",
+        "plausible_analytics_url": "https://views.scientific-python.org/js/script.js",
+    },
+    "logo": {"image_dark": "https://pandas.pydata.org/static/img/pandas_white.svg"},
+    "navbar_align": "left",
+    "navbar_end": ["version-switcher", "theme-switcher", "navbar-icon-links"],
+    "switcher": {
+        "json_url": "https://pandas.pydata.org/versions.json",
+        "version_match": switcher_version,
+    },
+    # This shows a warning for patch releases since the
+    # patch version doesn't compare as equal (e.g. 2.2.1 != 2.2.0 but it should be)
+    "show_version_warning_banner": False,
+    "icon_links": [
+        {
+            "name": "X",
+            "url": "https://x.com/pandas_dev",
+            "icon": "fa-brands fa-square-x-twitter",
+        },
+        {
+            "name": "Mastodon",
+            "url": "https://fosstodon.org/@pandas_dev",
+            "icon": "fa-brands fa-mastodon",
+        },
+    ],
+}
+
+# Add any paths that contain custom themes here, relative to this directory.
+# html_theme_path = ["themes"]
+
+# The name for this set of Sphinx documents.  If None, it defaults to
+# "<project> v<release> documentation".
+# html_title = None
+
+# A shorter title for the navigation bar.  Default is the same as html_title.
+# html_short_title = None
+
+# The name of an image file (relative to this directory) to place at the top
+# of the sidebar.
+html_logo = "../../web/pandas/static/img/pandas.svg"
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ["_static"]
+
+html_css_files = [
+    "css/getting_started.css",
+    "css/pandas.css",
+]
+
+# The name of an image file (within the static path) to use as favicon of the
+# docs.  This file should be a Windows icon file (.ico) being 16x16 or 32x32
+# pixels large.
+html_favicon = "../../web/pandas/static/img/favicon.ico"
+
+# If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
+# using the given strftime format.
+# html_last_updated_fmt = '%b %d, %Y'
+
+# If true, SmartyPants will be used to convert quotes and dashes to
+# typographically correct entities.
+# html_use_smartypants = True
+
+# Custom sidebar templates, maps document names to template names.
+# html_sidebars = {}
+
+# Additional templates that should be rendered to pages, maps page names to
+# template names.
+
+# Add redirect for previously existing API pages
+# each item is like `(from_old, to_new)`
+# To redirect a class and all its methods, see below
+# https://github.com/pandas-dev/pandas/issues/16186
+
+moved_api_pages = [
+    ("pandas.core.common.isnull", "pandas.isna"),
+    ("pandas.core.common.notnull", "pandas.notna"),
+    ("pandas.core.reshape.get_dummies", "pandas.get_dummies"),
+    ("pandas.tools.merge.concat", "pandas.concat"),
+    ("pandas.tools.merge.merge", "pandas.merge"),
+    ("pandas.tools.pivot.pivot_table", "pandas.pivot_table"),
+    ("pandas.tseries.tools.to_datetime", "pandas.to_datetime"),
+    ("pandas.io.clipboard.read_clipboard", "pandas.read_clipboard"),
+    ("pandas.io.excel.ExcelFile.parse", "pandas.ExcelFile.parse"),
+    ("pandas.io.excel.read_excel", "pandas.read_excel"),
+    ("pandas.io.html.read_html", "pandas.read_html"),
+    ("pandas.io.json.read_json", "pandas.read_json"),
+    ("pandas.io.parsers.read_csv", "pandas.read_csv"),
+    ("pandas.io.parsers.read_fwf", "pandas.read_fwf"),
+    ("pandas.io.parsers.read_table", "pandas.read_table"),
+    ("pandas.io.pickle.read_pickle", "pandas.read_pickle"),
+    ("pandas.io.pytables.HDFStore.append", "pandas.HDFStore.append"),
+    ("pandas.io.pytables.HDFStore.get", "pandas.HDFStore.get"),
+    ("pandas.io.pytables.HDFStore.put", "pandas.HDFStore.put"),
+    ("pandas.io.pytables.HDFStore.select", "pandas.HDFStore.select"),
+    ("pandas.io.pytables.read_hdf", "pandas.read_hdf"),
+    ("pandas.io.sql.read_sql", "pandas.read_sql"),
+    ("pandas.io.sql.read_frame", "pandas.read_frame"),
+    ("pandas.io.sql.write_frame", "pandas.write_frame"),
+    ("pandas.io.stata.read_stata", "pandas.read_stata"),
+]
+
+# Again, tuples of (from_old, to_new)
+moved_classes = [
+    ("pandas.tseries.resample.Resampler", "pandas.core.resample.Resampler"),
+    ("pandas.formats.style.Styler", "pandas.io.formats.style.Styler"),
+]
+
+for old, new in moved_classes:
+    # the class itself...
+    moved_api_pages.append((old, new))
+
+    mod, classname = new.rsplit(".", 1)
+    klass = getattr(importlib.import_module(mod), classname)
+    methods = [
+        x for x in dir(klass) if not x.startswith("_") or x in ("__iter__", "__array__")
+    ]
+    # ... and each of its public methods
+    moved_api_pages.extend((f"{old}.{method}", f"{new}.{method}") for method in methods)
+
+if include_api:
+    html_additional_pages = {
+        "generated/" + page[0]: "api_redirect.html" for page in moved_api_pages
+    }
+
+
+header = f"""\
+.. currentmodule:: pandas
+
+.. ipython:: python
+   :suppress:
+
+   import numpy as np
+   import pandas as pd
+
+   np.random.seed(123456)
+   np.set_printoptions(precision=4, suppress=True)
+   pd.options.display.max_rows = 15
+
+   import os
+   os.chdir(r'{os.path.dirname(os.path.dirname(__file__))}')
+"""
+
+
+html_context = {
+    "redirects": dict(moved_api_pages),
+    "header": header,
+}
+
+# If false, no module index is generated.
+html_use_modindex = True
+
+# If false, no index is generated.
+# html_use_index = True
+
+# If true, the index is split into individual pages for each letter.
+# html_split_index = False
+
+# If true, links to the reST sources are added to the pages.
+# html_show_sourcelink = True
+
+# If true, an OpenSearch description file will be output, and all pages will
+# contain a <link> tag referring to it.  The value of this option must be the
+# base URL from which the finished HTML is served.
+# html_use_opensearch = ''
+
+# If nonempty, this is the file name suffix for HTML files (e.g. ".xhtml").
+# html_file_suffix = ''
+
+# Output file base name for HTML help builder.
+htmlhelp_basename = "pandas"
+
+# -- Options for nbsphinx ------------------------------------------------
+
+nbsphinx_allow_errors = True
+
+# -- Options for LaTeX output --------------------------------------------
+
+latex_elements = {}
+
+# The paper size ('letter' or 'a4').
+# latex_paper_size = 'letter'
+
+# The font size ('10pt', '11pt' or '12pt').
+# latex_font_size = '10pt'
+
+# Grouping the document tree into LaTeX files. List of tuples (source start
+# file, target name, title, author, documentclass [howto/manual]).
+latex_documents = [
+    (
+        "index",
+        "pandas.tex",
+        "pandas: powerful Python data analysis toolkit",
+        "Wes McKinney and the pandas Development Team",
+        "manual",
+    )
+]
+
+# The name of an image file (relative to this directory) to place at the top of
+# the title page.
+# latex_logo = None
+
+# For "manual" documents, if this is true, then toplevel headings are parts,
+# not chapters.
+# latex_use_parts = False
+
+# Additional stuff for the LaTeX preamble.
+# latex_preamble = ''
+
+# Documents to append as an appendix to all manuals.
+# latex_appendices = []
+
+# If false, no module index is generated.
+# latex_use_modindex = True
+
+
+if include_api:
+    intersphinx_mapping = {
+        "dateutil": ("https://dateutil.readthedocs.io/en/latest/", None),
+        "matplotlib": ("https://matplotlib.org/stable/", None),
+        "numpy": ("https://numpy.org/doc/stable/", None),
+        "python": ("https://docs.python.org/3/", None),
+        "scipy": ("https://docs.scipy.org/doc/scipy/", None),
+        "pyarrow": ("https://arrow.apache.org/docs/", None),
+    }
+
+# extlinks alias
+extlinks = {
+    "issue": ("https://github.com/pandas-dev/pandas/issues/%s", "GH %s"),
+}
+
+
+ipython_execlines = [
+    "import numpy as np",
+    "import pandas as pd",
+    # This ensures correct rendering on system with console encoding != utf8
+    # (windows). It forces pandas to encode its output reprs using utf8
+    # wherever the docs are built. The docs' target is the browser, not
+    # the console, so this is fine.
+    'pd.options.display.encoding="utf8"',
+]
+
+
+# Add custom Documenter to handle attributes/methods of an AccessorProperty
+# eg pandas.Series.str and pandas.Series.dt (see GH9322)
+
+import sphinx  # isort:skip
+from sphinx.ext.autodoc import (  # isort:skip
+    AttributeDocumenter,
+    Documenter,
+    MethodDocumenter,
+)
+from sphinx.ext.autosummary import Autosummary  # isort:skip
+
+
+class AccessorDocumenter(MethodDocumenter):
+    """
+    Specialized Documenter subclass for accessors.
+    """
+
+    objtype = "accessor"
+    directivetype = "method"
+
+    # lower than MethodDocumenter so this is not chosen for normal methods
+    priority = 0.6
+
+    def format_signature(self) -> str:
+        # this method gives an error/warning for the accessors, therefore
+        # overriding it (accessor has no arguments)
+        return ""
+
+
+class AccessorLevelDocumenter(Documenter):
+    """
+    Specialized Documenter subclass for objects on accessor level (methods,
+    attributes).
+    """
+
+    # This is the simple straightforward version
+    # modname is None, base the last elements (eg 'hour')
+    # and path the part before (eg 'Series.dt')
+    # def resolve_name(self, modname, parents, path, base):
+    #     modname = 'pandas'
+    #     mod_cls = path.rstrip('.')
+    #     mod_cls = mod_cls.split('.')
+    #
+    #     return modname, mod_cls + [base]
+    def resolve_name(self, modname, parents, path, base):
+        if modname is None:
+            if path:
+                mod_cls = path.rstrip(".")
+            else:
+                mod_cls = None
+                # if documenting a class-level object without path,
+                # there must be a current class, either from a parent
+                # auto directive ...
+                mod_cls = self.env.temp_data.get("autodoc:class")
+                # ... or from a class directive
+                if mod_cls is None:
+                    mod_cls = self.env.temp_data.get("py:class")
+                # ... if still None, there's no way to know
+                if mod_cls is None:
+                    return None, []
+            # HACK: this is added in comparison to ClassLevelDocumenter
+            # mod_cls still exists of class.accessor, so an extra
+            # rpartition is needed
+            modname, _, accessor = mod_cls.rpartition(".")
+            modname, _, cls = modname.rpartition(".")
+            parents = [cls, accessor]
+            # if the module name is still missing, get it like above
+            if not modname:
+                modname = self.env.temp_data.get("autodoc:module")
+            if not modname:
+                if sphinx.__version__ > "1.3":
+                    modname = self.env.ref_context.get("py:module")
+                else:
+                    modname = self.env.temp_data.get("py:module")
+            # ... else, it stays None, which means invalid
+        return modname, parents + [base]
+
+
+class AccessorAttributeDocumenter(AccessorLevelDocumenter, AttributeDocumenter):
+    objtype = "accessorattribute"
+    directivetype = "attribute"
+
+    # lower than AttributeDocumenter so this is not chosen for normal
+    # attributes
+    priority = 0.6
+
+
+class AccessorMethodDocumenter(AccessorLevelDocumenter, MethodDocumenter):
+    objtype = "accessormethod"
+    directivetype = "method"
+
+    # lower than MethodDocumenter so this is not chosen for normal methods
+    priority = 0.6
+
+
+class AccessorCallableDocumenter(AccessorLevelDocumenter, MethodDocumenter):
+    """
+    This documenter lets us removes .__call__ from the method signature for
+    callable accessors like Series.plot
+    """
+
+    objtype = "accessorcallable"
+    directivetype = "method"
+
+    # lower than MethodDocumenter; otherwise the doc build prints warnings
+    priority = 0.5
+
+    def format_name(self):
+        return MethodDocumenter.format_name(self).removesuffix(".__call__")
+
+
+class PandasAutosummary(Autosummary):
+    """
+    This alternative autosummary class lets us override the table summary for
+    Series.plot and DataFrame.plot in the API docs.
+    """
+
+    def _replace_pandas_items(self, display_name, sig, summary, real_name):
+        # this a hack: ideally we should extract the signature from the
+        # .__call__ method instead of hard coding this
+        if display_name == "DataFrame.plot":
+            sig = "([x, y, kind, ax, ....])"
+            summary = "DataFrame plotting accessor and method"
+        elif display_name == "Series.plot":
+            sig = "([kind, ax, figsize, ....])"
+            summary = "Series plotting accessor and method"
+        return (display_name, sig, summary, real_name)
+
+    @staticmethod
+    def _is_deprecated(real_name):
+        try:
+            obj, parent, modname = _import_by_name(real_name)
+        except ImportError:
+            return False
+        doc = NumpyDocString(obj.__doc__ or "")
+        summary = "".join(doc["Summary"] + doc["Extended Summary"])
+        return ".. deprecated::" in summary
+
+    def _add_deprecation_prefixes(self, items):
+        for item in items:
+            display_name, sig, summary, real_name = item
+            if self._is_deprecated(real_name):
+                summary = f"(DEPRECATED) {summary}"
+            yield display_name, sig, summary, real_name
+
+    def get_items(self, names):
+        items = Autosummary.get_items(self, names)
+        items = [self._replace_pandas_items(*item) for item in items]
+        items = list(self._add_deprecation_prefixes(items))
+        return items
+
+
+# based on numpy doc/source/conf.py
+def linkcode_resolve(domain, info) -> str | None:
+    """
+    Determine the URL corresponding to Python object
+    """
+    if domain != "py":
+        return None
+
+    modname = info["module"]
+    fullname = info["fullname"]
+
+    submod = sys.modules.get(modname)
+    if submod is None:
+        return None
+
+    obj = submod
+    for part in fullname.split("."):
+        try:
+            with warnings.catch_warnings():
+                # Accessing deprecated objects will generate noisy warnings
+                warnings.simplefilter("ignore", FutureWarning)
+                obj = getattr(obj, part)
+        except AttributeError:
+            return None
+
+    try:
+        fn = inspect.getsourcefile(inspect.unwrap(obj))
+    except TypeError:
+        try:  # property
+            fn = inspect.getsourcefile(inspect.unwrap(obj.fget))
+        except (AttributeError, TypeError):
+            fn = None
+    if not fn:
+        return None
+
+    try:
+        source, lineno = inspect.getsourcelines(obj)
+    except TypeError:
+        try:  # property
+            source, lineno = inspect.getsourcelines(obj.fget)
+        except (AttributeError, TypeError):
+            lineno = None
+    except OSError:
+        lineno = None
+
+    if lineno:
+        linespec = f"#L{lineno}-L{lineno + len(source) - 1}"
+    else:
+        linespec = ""
+
+    fn = os.path.relpath(fn, start=os.path.dirname(pandas.__file__))
+
+    if "+" in pandas.__version__:
+        return f"https://github.com/pandas-dev/pandas/blob/main/pandas/{fn}{linespec}"
+    else:
+        return (
+            f"https://github.com/pandas-dev/pandas/blob/"
+            f"v{pandas.__version__}/pandas/{fn}{linespec}"
+        )
+
+
+# remove the docstring of the flags attribute (inherited from numpy ndarray)
+# because these give doc build errors (see GH issue 5331)
+def remove_flags_docstring(app, what, name, obj, options, lines) -> None:
+    if what == "attribute" and name.endswith(".flags"):
+        del lines[:]
+
+
+def process_class_docstrings(app, what, name, obj, options, lines) -> None:
+    """
+    For those classes for which we use ::
+
+    :template: autosummary/class_without_autosummary.rst
+
+    the documented attributes/methods have to be listed in the class
+    docstring. However, if one of those lists is empty, we use 'None',
+    which then generates warnings in sphinx / ugly html output.
+    This "autodoc-process-docstring" event connector removes that part
+    from the processed docstring.
+
+    """
+    if what == "class":
+        joined = "\n".join(lines)
+
+        templates = [
+            """.. rubric:: Attributes
+
+.. autosummary::
+   :toctree:
+
+   None
+""",
+            """.. rubric:: Methods
+
+.. autosummary::
+   :toctree:
+
+   None
+""",
+        ]
+
+        for template in templates:
+            if template in joined:
+                joined = joined.replace(template, "")
+        lines[:] = joined.split("\n")
+
+
+_BUSINED_ALIASES = [
+    "pandas.tseries.offsets." + name
+    for name in [
+        "BDay",
+        "CDay",
+        "BMonthEnd",
+        "BMonthBegin",
+        "CBMonthEnd",
+        "CBMonthBegin",
+    ]
+]
+
+
+def process_business_alias_docstrings(app, what, name, obj, options, lines) -> None:
+    """
+    Starting with sphinx 3.4, the "autodoc-process-docstring" event also
+    gets called for alias classes. This results in numpydoc adding the
+    methods/attributes to the docstring, which we don't want (+ this
+    causes warnings with sphinx).
+    """
+    if name in _BUSINED_ALIASES:
+        lines[:] = []
+
+
+suppress_warnings = [
+    # We "overwrite" autosummary with our PandasAutosummary, but
+    # still want the regular autosummary setup to run. So we just
+    # suppress this warning.
+    "app.add_directive"
+]
+if pattern:
+    # When building a single document we don't want to warn because references
+    # to other documents are unknown, as it's expected
+    suppress_warnings.append("ref.ref")
+
+
+def rstjinja(app, docname, source) -> None:
+    """
+    Render our pages as a jinja template for fancy templating goodness.
+    """
+    # https://www.ericholscher.com/blog/2016/jul/25/integrating-jinja-rst-sphinx/
+    # Make sure we're outputting HTML
+    if app.builder.format != "html":
+        return
+    src = source[0]
+    rendered = app.builder.templates.render_string(src, app.config.html_context)
+    source[0] = rendered
+
+
+def setup(app) -> None:
+    app.connect("source-read", rstjinja)
+    app.connect("autodoc-process-docstring", remove_flags_docstring)
+    app.connect("autodoc-process-docstring", process_class_docstrings)
+    app.connect("autodoc-process-docstring", process_business_alias_docstrings)
+    app.add_autodocumenter(AccessorDocumenter)
+    app.add_autodocumenter(AccessorAttributeDocumenter)
+    app.add_autodocumenter(AccessorMethodDocumenter)
+    app.add_autodocumenter(AccessorCallableDocumenter)
+    app.add_directive("autosummary", PandasAutosummary)
+
+
+# Ignore list for broken links,found in CI run checks for broken-linkcheck.yml
+
+linkcheck_ignore = [
+    "^http://$",
+    "^https://$",
+    *[
+        re.escape(link)
+        for link in [
+            "http://scatterci.github.io/pydata/pandas",
+            "http://specs.frictionlessdata.io/json-table-schema/",
+            "https://crates.io/crates/calamine",
+            "https://devguide.python.org/setup/#macos",
+            "https://en.wikipedia.org/wiki/Imputation_statistics",
+            "https://en.wikipedia.org/wiki/Imputation_(statistics",
+            "https://github.com/noatamir/pandas-dev",
+            "https://github.com/pandas-dev/pandas/blob/main/pandas/plotting/__init__.py#L1",
+            "https://github.com/pandas-dev/pandas/blob/v0.20.2/pandas/core/generic.py#L568",
+            "https://github.com/pandas-dev/pandas/blob/v0.20.2/pandas/core/frame.py#L1495",
+            "https://github.com/pandas-dev/pandas/issues/174151",
+            "https://gitpod.io/#https://github.com/USERNAME/pandas",
+            "https://manishamde.github.io/blog/2013/03/07/pandas-and-python-top-10/",
+            "https://matplotlib.org/api/axes_api.html#matplotlib.axes.Axes.table",
+            "https://nipunbatra.github.io/blog/visualisation/2013/05/01/aggregation-timeseries.html",
+            "https://nbviewer.ipython.org/gist/metakermit/5720498",
+            "https://numpy.org/doc/stable/user/basics.byteswapping.html",
+            "https://pandas.pydata.org/pandas-docs/stable/io.html#io-chunking",
+            "https://pandas.pydata.org/pandas-docs/stable/ecosystem.html",
+            "https://sqlalchemy.readthedocs.io/en/latest/dialects/index.html",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000245912.htm",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000214639.htm",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002283942.htm",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000245965.htm",
+            "https://support.sas.com/documentation/cdl/en/imlug/66845/HTML/default/viewer.htm#imlug_langref_sect455.htm",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002284668.htm",
+            "https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002978282.htm",
+            "https://wesmckinney.com/blog/update-on-upcoming-pandas-v0-10-new-file-parser-other-performance-wins/",
+            "https://visualstudio.microsoft.com/downloads/#build-tools-for-visual-studio-2022",
+            "pandas.zip",
+        ]
+    ],
+]
diff --git a/pandas/doc/source/development/community.rst b/pandas/doc/source/development/community.rst
new file mode 100644
index 0000000000000..ab8294b8f135a
--- /dev/null
+++ b/pandas/doc/source/development/community.rst
@@ -0,0 +1,123 @@
+.. _community:
+
+=====================
+Contributor community
+=====================
+
+pandas is a community-driven open source project developed by a large group
+of `contributors <https://github.com/pandas-dev/pandas/graphs/contributors>`_
+and a smaller group of `maintainers <https://pandas.pydata.org/about/team.html>`_.
+The pandas leadership has made a strong commitment to creating an open,
+inclusive, and positive community. Please read the pandas `Code of Conduct
+<https://pandas.pydata.org/community/coc.html>`_ for guidance on how to
+interact with others in a way that makes the community thrive.
+
+We offer several meetings and communication channels to share knowledge and
+connect with others within the pandas community.
+
+Community meeting
+-----------------
+
+The pandas Community Meeting is a regular sync meeting for the project's
+maintainers which is open to the community. Everyone is welcome to attend and
+contribute to conversations.
+
+The meetings take place on the second and fourth Wednesdays of each month at 18:00 UTC.
+
+The minutes of past meetings are available in `this Google Document <https://docs.google.com/document/d/1tGbTiYORHiSPgVMXawiweGJlBw5dOkVJLY-licoBmBU/edit?usp=sharing>`__.
+
+
+New contributor meeting
+-----------------------
+
+On the third Wednesday of the month, we hold meetings to welcome and support
+new contributors in our community.
+
+| 👋 you all are invited
+| 💬 everyone can present (add yourself to the hackMD agenda)
+| 👀 anyone can sit in and listen
+
+Attendees are new and experienced contributors, as well as a few maintainers.
+We aim to answer questions about getting started, or help with work in
+progress when possible, as well as get to know each other and share our
+learnings and experiences.
+
+The agenda for the next meeting and minutes of past meetings are available in
+`this HackMD <https://hackmd.io/@pandas-dev/HJgQt1Tei>`__.
+
+Calendar
+--------
+
+This calendar shows all the community meetings. Our community meetings are
+ideal for anyone wanting to contribute to pandas, or just curious to know how
+current development is going.
+
+.. raw:: html
+
+   <iframe src="https://calendar.google.com/calendar/embed?src=pgbn14p6poja8a1cf2dv2jhrmg%40group.calendar.google.com" style="border: 0" width="800" height="600" frameborder="0" scrolling="no"></iframe>
+
+You can subscribe to this calendar with the following links:
+
+* `iCal <https://calendar.google.com/calendar/ical/pgbn14p6poja8a1cf2dv2jhrmg%40group.calendar.google.com/public/basic.ics>`__
+* `Google calendar <https://calendar.google.com/calendar/r?cid=pgbn14p6poja8a1cf2dv2jhrmg@group.calendar.google.com>`__
+
+Additionally, we'll sometimes have one-off meetings on specific topics.
+These will be published on the same calendar.
+
+`GitHub issue tracker <https://github.com/pandas-dev/pandas/issues>`_
+----------------------------------------------------------------------
+
+The pandas contributor community conducts conversations mainly via this channel.
+Any community member can open issues to:
+
+- Report bugs, e.g. "I noticed the behavior of a certain function is
+  incorrect"
+- Request features, e.g. "I would like this error message to be more readable"
+- Request documentation improvements, e.g. "I found this section unclear"
+- Ask questions, e.g. "I noticed the behavior of a certain function
+  changed between versions. Is this expected?".
+
+    Ideally, your questions should be related to how pandas works rather
+    than how you use pandas. `StackOverflow <https://stackoverflow.com/>`_ is
+    better suited for answering usage questions, and we ask that all usage
+    questions are first asked on StackOverflow. Thank you for respecting our
+    time and wishes. 🙇
+
+Maintainers and frequent contributors might also open issues to discuss the
+ongoing development of the project. For example:
+
+- Report issues with the CI, GitHub Actions, or the performance of pandas
+- Open issues relating to the internals
+- Start roadmap discussion aligning on proposals for what to do in future
+  releases or changes to the API.
+- Open issues relating to the project's website, logo, or governance
+
+The developer mailing list
+--------------------------
+
+The pandas mailing list `pandas-dev@python.org <mailto://pandas-dev@python
+.org>`_ is used for long form
+conversations and to engage people in the wider community who might not
+be active on the issue tracker but we would like to include in discussions.
+
+Join the mailing list and view the archives `here <https://mail.python.org/mailman/listinfo/pandas-dev>`_.
+
+.. _community.slack:
+
+Community slack
+---------------
+
+We have a chat platform for contributors, maintainers and potential
+contributors. This is not a space for user questions, rather for questions about
+contributing to pandas. The slack is a private space, specifically meant for
+people who are hesitant to bring up their questions or ideas on a large public
+mailing list or GitHub.
+
+If this sounds like the right place for you, you are welcome to join using
+`this link <https://join.slack.com/t/pandas-dev-community/shared_invite/zt-2blg6u9k3-K6_XvMRDZWeH7Id274UeIg>`_!
+Please remember to follow our `Code of Conduct <https://pandas.pydata.org/community/coc.html>`_,
+and be aware that our admins are monitoring for irrelevant messages and will remove folks who use
+our
+slack for spam, advertisements and messages not related to the pandas contributing community. And
+please remember that slack is not meant to replace the mailing list or issue tracker - all important
+announcements and conversations should still happen there.
diff --git a/pandas/doc/source/development/contributing.rst b/pandas/doc/source/development/contributing.rst
new file mode 100644
index 0000000000000..4d99f282aa695
--- /dev/null
+++ b/pandas/doc/source/development/contributing.rst
@@ -0,0 +1,341 @@
+.. _contributing:
+
+{{ header }}
+
+**********************
+Contributing to pandas
+**********************
+
+.. contents:: Table of contents:
+   :local:
+
+
+All contributions, bug reports, bug fixes, documentation improvements,
+enhancements, and ideas are welcome.
+
+.. _contributing.bug_reports:
+
+Bug reports and enhancement requests
+====================================
+
+Bug reports and enhancement requests are an important part of making pandas more stable and
+are curated though Github issues. When reporting an issue or request, please select the `appropriate
+category and fill out the issue form fully <https://github.com/pandas-dev/pandas/issues/new/choose>`_
+to ensure others and the core development team can fully understand the scope of the issue.
+
+The issue will then show up to the pandas community and be open to comments/ideas from others.
+
+Finding an issue to contribute to
+=================================
+
+If you are brand new to pandas or open-source development, we recommend searching
+the `GitHub "issues" tab <https://github.com/pandas-dev/pandas/issues>`_
+to find issues that interest you. Unassigned issues labeled `Docs
+<https://github.com/pandas-dev/pandas/issues?q=is%3Aopen+sort%3Aupdated-desc+label%3ADocs+no%3Aassignee>`_
+and `good first issue
+<https://github.com/pandas-dev/pandas/issues?q=is%3Aopen+sort%3Aupdated-desc+label%3A%22good+first+issue%22+no%3Aassignee>`_
+are typically good for newer contributors.
+
+Once you've found an interesting issue, it's a good idea to assign the issue to yourself,
+so nobody else duplicates the work on it. On the Github issue, a comment with the exact
+text ``take`` to automatically assign you the issue
+(this will take seconds and may require refreshing the page to see it).
+
+If for whatever reason you are not able to continue working with the issue, please
+unassign it, so other people know it's available again. You can check the list of
+assigned issues, since people may not be working in them anymore. If you want to work on one
+that is assigned, feel free to kindly ask the current assignee if you can take it
+(please allow at least a week of inactivity before considering work in the issue discontinued).
+
+We have several :ref:`contributor community <community>` communication channels, which you are
+welcome to join, and ask questions as you figure things out. Among them are regular meetings for
+new contributors, dev meetings, a dev mailing list, and a Slack for the contributor community.
+All pandas contributors are welcome to these spaces, where they can connect with each other. Even
+maintainers who have been with us for a long time felt just like you when they started out, and
+are happy to welcome you and support you as you get to know how we work, and where things are.
+Take a look at the next sections to learn more.
+
+.. _contributing.github:
+
+Submitting a pull request
+=========================
+
+.. _contributing.version_control:
+
+Version control, Git, and GitHub
+--------------------------------
+
+pandas is hosted on `GitHub <https://www.github.com/pandas-dev/pandas>`_, and to
+contribute, you will need to sign up for a `free GitHub account
+<https://github.com/signup/free>`_. We use `Git <https://git-scm.com/>`_ for
+version control to allow many people to work together on the project.
+
+If you are new to Git, you can reference some of these resources for learning Git. Feel free to reach out
+to the :ref:`contributor community <community>` for help if needed:
+
+* `Git documentation <https://git-scm.com/doc>`_.
+
+Also, the project follows a forking workflow further described on this page whereby
+contributors fork the repository, make changes and then create a pull request.
+So please be sure to read and follow all the instructions in this guide.
+
+If you are new to contributing to projects through forking on GitHub,
+take a look at the `GitHub documentation for contributing to projects <https://docs.github.com/en/get-started/quickstart/contributing-to-projects>`_.
+GitHub provides a quick tutorial using a test repository that may help you become more familiar
+with forking a repository, cloning a fork, creating a feature branch, pushing changes and
+making pull requests.
+
+Below are some useful resources for learning more about forking and pull requests on GitHub:
+
+* the `GitHub documentation for forking a repo <https://docs.github.com/en/get-started/quickstart/fork-a-repo>`_.
+* the `GitHub documentation for collaborating with pull requests <https://docs.github.com/en/pull-requests/collaborating-with-pull-requests>`_.
+* the `GitHub documentation for working with forks <https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks>`_.
+
+Getting started with Git
+------------------------
+
+`GitHub has instructions <https://docs.github.com/en/get-started/quickstart/set-up-git>`__ for installing git,
+setting up your SSH key, and configuring git.  All these steps need to be completed before
+you can work seamlessly between your local repository and GitHub.
+
+.. _contributing.forking:
+
+Create a fork of pandas
+-----------------------
+
+You will need your own copy of pandas (aka fork) to work on the code. Go to the `pandas project
+page <https://github.com/pandas-dev/pandas>`_ and hit the ``Fork`` button. Please uncheck the box to copy only the main branch before selecting ``Create Fork``.
+You will want to clone your fork to your machine
+
+.. code-block:: shell
+
+    git clone https://github.com/your-user-name/pandas.git pandas-yourname
+    cd pandas-yourname
+    git remote add upstream https://github.com/pandas-dev/pandas.git
+    git fetch upstream
+
+This creates the directory ``pandas-yourname`` and connects your repository to
+the upstream (main project) *pandas* repository.
+
+.. note::
+
+    Performing a shallow clone (with ``--depth==N``, for some ``N`` greater
+    or equal to 1) might break some tests and features as ``pd.show_versions()``
+    as the version number cannot be computed anymore.
+
+Creating a feature branch
+-------------------------
+
+Your local ``main`` branch should always reflect the current state of pandas repository.
+First ensure it's up-to-date with the main pandas repository.
+
+.. code-block:: shell
+
+    git checkout main
+    git pull upstream main --ff-only
+
+Then, create a feature branch for making your changes. For example
+
+.. code-block:: shell
+
+    git checkout -b shiny-new-feature
+
+This changes your working branch from ``main`` to the ``shiny-new-feature`` branch.  Keep any
+changes in this branch specific to one bug or feature so it is clear
+what the branch brings to pandas. You can have many feature branches
+and switch in between them using the ``git checkout`` command.
+
+When you want to update the feature branch with changes in main after
+you created the branch, check the section on
+:ref:`updating a PR <contributing.update-pr>`.
+
+.. _contributing.commit-code:
+
+Making code changes
+-------------------
+
+Before modifying any code, ensure you follow the :ref:`contributing environment <contributing_environment>`
+guidelines to set up an appropriate development environment.
+
+Then once you have made code changes, you can see all the changes you've currently made by running.
+
+.. code-block:: shell
+
+    git status
+
+For files you intended to modify or add, run.
+
+.. code-block:: shell
+
+    git add path/to/file-to-be-added-or-changed.py
+
+Running ``git status`` again should display
+
+.. code-block:: shell
+
+    On branch shiny-new-feature
+
+         modified:   /relative/path/to/file-to-be-added-or-changed.py
+
+
+Finally, commit your changes to your local repository with an explanatory commit
+message
+
+.. code-block:: shell
+
+    git commit -m "your commit message goes here"
+
+.. _contributing.push-code:
+
+Pushing your changes
+--------------------
+
+When you want your changes to appear publicly on your GitHub page, push your
+forked feature branch's commits
+
+.. code-block:: shell
+
+    git push origin shiny-new-feature
+
+Here ``origin`` is the default name given to your remote repository on GitHub.
+You can see the remote repositories
+
+.. code-block:: shell
+
+    git remote -v
+
+If you added the upstream repository as described above you will see something
+like
+
+.. code-block:: shell
+
+    origin  git@github.com:yourname/pandas.git (fetch)
+    origin  git@github.com:yourname/pandas.git (push)
+    upstream        git://github.com/pandas-dev/pandas.git (fetch)
+    upstream        git://github.com/pandas-dev/pandas.git (push)
+
+Now your code is on GitHub, but it is not yet a part of the pandas project. For that to
+happen, a pull request needs to be submitted on GitHub.
+
+Making a pull request
+---------------------
+
+One you have finished your code changes, your code change will need to follow the
+:ref:`pandas contribution guidelines <contributing_codebase>` to be successfully accepted.
+
+If everything looks good, you are ready to make a pull request. A pull request is how
+code from your local repository becomes available to the GitHub community to review
+and merged into project to appear the in the next release. To submit a pull request:
+
+#. Navigate to your repository on GitHub
+#. Click on the ``Compare & pull request`` button
+#. You can then click on ``Commits`` and ``Files Changed`` to make sure everything looks
+   okay one last time
+#. Write a descriptive title that includes prefixes. pandas uses a convention for title
+   prefixes. Here are some common ones along with general guidelines for when to use them:
+
+    * ENH: Enhancement, new functionality
+    * BUG: Bug fix
+    * DOC: Additions/updates to documentation
+    * TST: Additions/updates to tests
+    * BLD: Updates to the build process/scripts
+    * PERF: Performance improvement
+    * TYP: Type annotations
+    * CLN: Code cleanup
+
+#. Write a description of your changes in the ``Preview Discussion`` tab
+#. Click ``Send Pull Request``.
+
+This request then goes to the repository maintainers, and they will review
+the code.
+
+.. _contributing.update-pr:
+
+Updating your pull request
+--------------------------
+
+Based on the review you get on your pull request, you will probably need to make
+some changes to the code. You can follow the :ref:`code committing steps <contributing.commit-code>`
+again to address any feedback and update your pull request.
+
+It is also important that updates in the pandas ``main`` branch are reflected in your pull request.
+To update your feature branch with changes in the pandas ``main`` branch, run:
+
+.. code-block:: shell
+
+    git checkout shiny-new-feature
+    git fetch upstream
+    git merge upstream/main
+
+If there are no conflicts (or they could be fixed automatically), a file with a
+default commit message will open, and you can simply save and quit this file.
+
+If there are merge conflicts, you need to solve those conflicts. See for
+example at https://help.github.com/articles/resolving-a-merge-conflict-using-the-command-line/
+for an explanation on how to do this.
+
+Once the conflicts are resolved, run:
+
+#. ``git add -u`` to stage any files you've updated;
+#. ``git commit`` to finish the merge.
+
+.. note::
+
+    If you have uncommitted changes at the moment you want to update the branch with
+    ``main``, you will need to ``stash`` them prior to updating (see the
+    `stash docs <https://git-scm.com/book/en/v2/Git-Tools-Stashing-and-Cleaning>`__).
+    This will effectively store your changes and they can be reapplied after updating.
+
+After the feature branch has been update locally, you can now update your pull
+request by pushing to the branch on GitHub:
+
+.. code-block:: shell
+
+    git push origin shiny-new-feature
+
+Any ``git push`` will automatically update your pull request with your branch's changes
+and restart the :ref:`Continuous Integration <contributing.ci>` checks.
+
+.. _contributing.update-dev:
+
+Updating the development environment
+------------------------------------
+
+It is important to periodically update your local ``main`` branch with updates from the pandas ``main``
+branch and update your development environment to reflect any changes to the various packages that
+are used during development.
+
+If using :ref:`conda <contributing.conda>`, run:
+
+.. code-block:: shell
+
+    git checkout main
+    git fetch upstream
+    git merge upstream/main
+    conda activate pandas-dev
+    conda env update -f environment.yml --prune
+
+If using :ref:`pip <contributing.pip>` , do:
+
+.. code-block:: shell
+
+    git checkout main
+    git fetch upstream
+    git merge upstream/main
+    # activate the virtual environment based on your platform
+    python -m pip install --upgrade -r requirements-dev.txt
+
+Tips for a successful pull request
+==================================
+
+If you have made it to the `Making a pull request`_ phase, one of the core contributors may
+take a look. Please note however that a handful of people are responsible for reviewing
+all of the contributions, which can often lead to bottlenecks.
+
+To improve the chances of your pull request being reviewed, you should:
+
+- **Reference an open issue** for non-trivial changes to clarify the PR's purpose
+- **Ensure you have appropriate tests**. These should be the first part of any PR
+- **Keep your pull requests as simple as possible**. Larger PRs take longer to review
+- **Ensure that CI is in a green state**. Reviewers may not even look otherwise
+- **Keep** `Updating your pull request`_, either by request or every few days
diff --git a/pandas/doc/source/development/contributing_codebase.rst b/pandas/doc/source/development/contributing_codebase.rst
new file mode 100644
index 0000000000000..45d4e24b0df51
--- /dev/null
+++ b/pandas/doc/source/development/contributing_codebase.rst
@@ -0,0 +1,964 @@
+.. _contributing_codebase:
+
+{{ header }}
+
+=============================
+Contributing to the code base
+=============================
+
+.. contents:: Table of Contents:
+   :local:
+
+Code standards
+--------------
+
+Writing good code is not just about what you write. It is also about *how* you
+write it. During :ref:`Continuous Integration <contributing.ci>` testing, several
+tools will be run to check your code for stylistic errors.
+Generating any warnings will cause the test to fail.
+Thus, good style is a requirement for submitting code to pandas.
+
+There are a couple of tools in pandas to help contributors verify their changes
+before contributing to the project
+
+- ``./ci/code_checks.sh``: a script validates the doctests, formatting in docstrings,
+  and imported modules. It is possible to run the checks independently by using the
+  parameters ``docstrings``, ``code``, and ``doctests``
+  (e.g. ``./ci/code_checks.sh doctests``);
+- ``pre-commit``, which we go into detail on in the next section.
+
+In addition, because a lot of people use our library, it is important that we
+do not make sudden changes to the code that could have the potential to break
+a lot of user code as a result, that is, we need it to be as *backwards compatible*
+as possible to avoid mass breakages.
+
+.. _contributing.pre-commit:
+
+Pre-commit
+----------
+
+Additionally, :ref:`Continuous Integration <contributing.ci>` will run code formatting checks
+like ``ruff``,
+``isort``, and ``clang-format`` and more using `pre-commit hooks <https://pre-commit.com/>`_.
+Any warnings from these checks will cause the :ref:`Continuous Integration <contributing.ci>` to fail; therefore,
+it is helpful to run the check yourself before submitting code. This
+can be done by installing ``pre-commit`` (which should already have happened if you followed the instructions
+in :ref:`Setting up your development environment <contributing_environment>`) and then running::
+
+    pre-commit install
+
+from the root of the pandas repository. Now all of the styling checks will be
+run each time you commit changes without your needing to run each one manually.
+In addition, using ``pre-commit`` will also allow you to more easily
+remain up-to-date with our code checks as they change.
+
+Note that if needed, you can skip these checks with ``git commit --no-verify``.
+
+If you don't want to use ``pre-commit`` as part of your workflow, you can still use it
+to run its checks with one of the following::
+
+    pre-commit run --files <files you have modified>
+    pre-commit run --from-ref=upstream/main --to-ref=HEAD --all-files
+
+without needing to have done ``pre-commit install`` beforehand.
+
+Finally, we also have some slow pre-commit checks, which don't run on each commit
+but which do run during continuous integration. You can trigger them manually with::
+
+    pre-commit run --hook-stage manual --all-files
+
+.. note::
+
+    You may want to periodically run ``pre-commit gc``, to clean up repos
+    which are no longer used.
+
+.. note::
+
+    If you have conflicting installations of ``virtualenv``, then you may get an
+    error - see `here <https://github.com/pypa/virtualenv/issues/1875>`_.
+
+    Also, due to a `bug in virtualenv <https://github.com/pypa/virtualenv/issues/1986>`_,
+    you may run into issues if you're using conda. To solve this, you can downgrade
+    ``virtualenv`` to version ``20.0.33``.
+
+.. note::
+
+    If you have recently merged in main from the upstream branch, some of the
+    dependencies used by ``pre-commit`` may have changed.  Make sure to
+    :ref:`update your development environment <contributing.update-dev>`.
+
+Optional dependencies
+---------------------
+
+Optional dependencies (e.g. matplotlib) should be imported with the private helper
+``pandas.compat._optional.import_optional_dependency``. This ensures a
+consistent error message when the dependency is not met.
+
+All methods using an optional dependency should include a test asserting that an
+``ImportError`` is raised when the optional dependency is not found. This test
+should be skipped if the library is present.
+
+All optional dependencies should be documented in
+:ref:`install.optional_dependencies` and the minimum required version should be
+set in the ``pandas.compat._optional.VERSIONS`` dict.
+
+Backwards compatibility
+-----------------------
+
+Please try to maintain backward compatibility. pandas has lots of users with lots of
+existing code, so don't break it if at all possible.  If you think breakage is required,
+clearly state why as part of the pull request.  Also, be careful when changing method
+signatures and add deprecation warnings where needed. Also, add the deprecated sphinx
+directive to the deprecated functions or methods.
+
+If a function with the same arguments as the one being deprecated exist, you can use
+the ``pandas.util._decorators.deprecate``:
+
+.. code-block:: python
+
+    from pandas.util._decorators import deprecate
+
+    deprecate('old_func', 'new_func', '1.1.0')
+
+Otherwise, you need to do it manually:
+
+.. code-block:: python
+
+    import warnings
+    from pandas.util._exceptions import find_stack_level
+
+
+    def old_func():
+        """Summary of the function.
+
+        .. deprecated:: 1.1.0
+           Use new_func instead.
+        """
+        warnings.warn(
+            'Use new_func instead.',
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        new_func()
+
+
+    def new_func():
+        pass
+
+You'll also need to
+
+1. Write a new test that asserts a warning is issued when calling with the deprecated argument
+2. Update all of pandas existing tests and code to use the new argument
+
+See :ref:`contributing.warnings` for more.
+
+.. _contributing.type_hints:
+
+Type hints
+----------
+
+pandas strongly encourages the use of :pep:`484` style type hints. New development should contain type hints and pull requests to annotate existing code are accepted as well!
+
+Style guidelines
+~~~~~~~~~~~~~~~~
+
+Type imports should follow the ``from typing import ...`` convention.
+Your code may be automatically re-written to use some modern constructs (e.g. using the built-in ``list`` instead of ``typing.List``)
+by the :ref:`pre-commit checks <contributing.pre-commit>`.
+
+In some cases in the code base classes may define class variables that shadow builtins. This causes an issue as described in `Mypy 1775 <https://github.com/python/mypy/issues/1775#issuecomment-310969854>`_. The defensive solution here is to create an unambiguous alias of the builtin and use that without your annotation. For example, if you come across a definition like
+
+.. code-block:: python
+
+   class SomeClass1:
+       str = None
+
+The appropriate way to annotate this would be as follows
+
+.. code-block:: python
+
+   str_type = str
+
+   class SomeClass2:
+       str: str_type = None
+
+In some cases you may be tempted to use ``cast`` from the typing module when you know better than the analyzer. This occurs particularly when using custom inference functions. For example
+
+.. code-block:: python
+
+   from typing import cast
+
+   from pandas.core.dtypes.common import is_number
+
+   def cannot_infer_bad(obj: Union[str, int, float]):
+
+       if is_number(obj):
+           ...
+       else:  # Reasonably only str objects would reach this but...
+           obj = cast(str, obj)  # Mypy complains without this!
+           return obj.upper()
+
+The limitation here is that while a human can reasonably understand that ``is_number`` would catch the ``int`` and ``float`` types mypy cannot make that same inference just yet (see `mypy #5206 <https://github.com/python/mypy/issues/5206>`_). While the above works, the use of ``cast`` is **strongly discouraged**. Where applicable a refactor of the code to appease static analysis is preferable
+
+.. code-block:: python
+
+   def cannot_infer_good(obj: Union[str, int, float]):
+
+       if isinstance(obj, str):
+           return obj.upper()
+       else:
+           ...
+
+With custom types and inference this is not always possible so exceptions are made, but every effort should be exhausted to avoid ``cast`` before going down such paths.
+
+pandas-specific types
+~~~~~~~~~~~~~~~~~~~~~
+
+Commonly used types specific to pandas will appear in `pandas._typing <https://github.com/pandas-dev/pandas/blob/main/pandas/_typing.py>`_ and you should use these where applicable. This module is private for now but ultimately this should be exposed to third party libraries who want to implement type checking against pandas.
+
+For example, quite a few functions in pandas accept a ``dtype`` argument. This can be expressed as a string like ``"object"``, a ``numpy.dtype`` like ``np.int64`` or even a pandas ``ExtensionDtype`` like ``pd.CategoricalDtype``. Rather than burden the user with having to constantly annotate all of those options, this can simply be imported and reused from the pandas._typing module
+
+.. code-block:: python
+
+   from pandas._typing import Dtype
+
+   def as_type(dtype: Dtype) -> ...:
+       ...
+
+This module will ultimately house types for repeatedly used concepts like "path-like", "array-like", "numeric", etc... and can also hold aliases for commonly appearing parameters like ``axis``. Development of this module is active so be sure to refer to the source for the most up to date list of available types.
+
+Validating type hints
+~~~~~~~~~~~~~~~~~~~~~
+
+pandas uses `mypy <http://mypy-lang.org>`_ and `pyright <https://github.com/microsoft/pyright>`_ to statically analyze the code base and type hints. After making any change you can ensure your type hints are consistent by running
+
+.. code-block:: shell
+
+    pre-commit run --hook-stage manual --all-files mypy
+    pre-commit run --hook-stage manual --all-files pyright
+    pre-commit run --hook-stage manual --all-files pyright_reportGeneralTypeIssues
+    # the following might fail if the installed pandas version does not correspond to your local git version
+    pre-commit run --hook-stage manual --all-files stubtest
+
+in your python environment.
+
+.. warning::
+
+    * Please be aware that the above commands will use the current python environment. If your python packages are older/newer than those installed by the pandas CI, the above commands might fail. This is often the case when the ``mypy`` or ``numpy`` versions do not match. Please see :ref:`how to setup the python environment <contributing.conda>` or select a `recently succeeded workflow <https://github.com/pandas-dev/pandas/actions/workflows/code-checks.yml?query=branch%3Amain+is%3Asuccess>`_, select the "Docstring validation, typing, and other manual pre-commit hooks" job, then click on "Set up Conda" and "Environment info" to see which versions the pandas CI installs.
+
+.. _contributing.ci:
+
+Testing type hints in code using pandas
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. warning::
+
+    * pandas is not yet a py.typed library (:pep:`561`)!
+      The primary purpose of locally declaring pandas as a py.typed library is to test and
+      improve the pandas-builtin type annotations.
+
+Until pandas becomes a py.typed library, it is possible to easily experiment with the type
+annotations shipped with pandas by creating an empty file named "py.typed" in the pandas
+installation folder:
+
+.. code-block:: none
+
+   python -c "import pandas; import pathlib; (pathlib.Path(pandas.__path__[0]) / 'py.typed').touch()"
+
+The existence of the py.typed file signals to type checkers that pandas is already a py.typed
+library. This makes type checkers aware of the type annotations shipped with pandas.
+
+Testing with continuous integration
+-----------------------------------
+
+The pandas test suite will run automatically on `GitHub Actions <https://github.com/features/actions/>`__
+continuous integration services, once your pull request is submitted.
+However, if you wish to run the test suite on a branch prior to submitting the pull request,
+then the continuous integration services need to be hooked to your GitHub repository. Instructions are here
+for `GitHub Actions <https://docs.github.com/en/actions/>`__.
+
+A pull-request will be considered for merging when you have an all 'green' build. If any tests are failing,
+then you will get a red 'X', where you can click through to see the individual failed tests.
+This is an example of a green build.
+
+.. image:: ../_static/ci.png
+
+.. _contributing.tdd:
+
+
+Test-driven development
+-----------------------
+
+pandas is serious about testing and strongly encourages contributors to embrace
+`test-driven development (TDD) <https://en.wikipedia.org/wiki/Test-driven_development>`_.
+This development process "relies on the repetition of a very short development cycle:
+first the developer writes an (initially failing) automated test case that defines a desired
+improvement or new function, then produces the minimum amount of code to pass that test."
+So, before actually writing any code, you should write your tests.  Often the test can be
+taken from the original GitHub issue.  However, it is always worth considering additional
+use cases and writing corresponding tests.
+
+We use `code coverage <https://en.wikipedia.org/wiki/Code_coverage>`_ to help understand
+the amount of code which is covered by a test. We recommend striving to ensure code
+you add or change within Pandas is covered by a test. Please see our
+`code coverage dashboard through Codecov <https://app.codecov.io/github/pandas-dev/pandas>`_
+for more information.
+
+Adding tests is one of the most common requests after code is pushed to pandas.  Therefore,
+it is worth getting in the habit of writing tests ahead of time so this is never an issue.
+
+Writing tests
+~~~~~~~~~~~~~
+
+All tests should go into the ``tests`` subdirectory of the specific package.
+This folder contains many current examples of tests, and we suggest looking to these for
+inspiration.
+
+As a general tip, you can use the search functionality in your integrated development
+environment (IDE) or the git grep command in a terminal to find test files in which the method
+is called. If you are unsure of the best location to put your test, take your best guess,
+but note that reviewers may request that you move the test to a different location.
+
+To use git grep, you can run the following command in a terminal:
+
+``git grep "function_name("``
+
+This will search through all files in your repository for the text ``function_name(``.
+This can be a useful way to quickly locate the function in the
+codebase and determine the best location to add a test for it.
+
+Ideally, there should be one, and only one, obvious place for a test to reside.
+Until we reach that ideal, these are some rules of thumb for where a test should
+be located.
+
+1. Does your test depend only on code in ``pd._libs.tslibs``?
+   This test likely belongs in one of:
+
+   - tests.tslibs
+
+     .. note::
+
+          No file in ``tests.tslibs`` should import from any pandas modules
+          outside of ``pd._libs.tslibs``
+
+   - tests.scalar
+   - tests.tseries.offsets
+
+2. Does your test depend only on code in ``pd._libs``?
+   This test likely belongs in one of:
+
+   - tests.libs
+   - tests.groupby.test_libgroupby
+
+3. Is your test for an arithmetic or comparison method?
+   This test likely belongs in one of:
+
+   - tests.arithmetic
+
+     .. note::
+
+         These are intended for tests that can be shared to test the behavior
+         of DataFrame/Series/Index/ExtensionArray using the ``box_with_array``
+         fixture.
+
+   - tests.frame.test_arithmetic
+   - tests.series.test_arithmetic
+
+4. Is your test for a reduction method (min, max, sum, prod, ...)?
+   This test likely belongs in one of:
+
+   - tests.reductions
+
+     .. note::
+
+         These are intended for tests that can be shared to test the behavior
+         of DataFrame/Series/Index/ExtensionArray.
+
+   - tests.frame.test_reductions
+   - tests.series.test_reductions
+   - tests.test_nanops
+
+5. Is your test for an indexing method?
+   This is the most difficult case for deciding where a test belongs, because
+   there are many of these tests, and many of them test more than one method
+   (e.g. both ``Series.__getitem__`` and ``Series.loc.__getitem__``)
+
+   A) Is the test specifically testing an Index method (e.g. ``Index.get_loc``,
+      ``Index.get_indexer``)?
+      This test likely belongs in one of:
+
+      - tests.indexes.test_indexing
+      - tests.indexes.fooindex.test_indexing
+
+      Within that files there should be a method-specific test class e.g.
+      ``TestGetLoc``.
+
+      In most cases, neither ``Series`` nor ``DataFrame`` objects should be
+      needed in these tests.
+
+   B) Is the test for a Series or DataFrame indexing method *other* than
+      ``__getitem__`` or ``__setitem__``, e.g. ``xs``, ``where``, ``take``,
+      ``mask``, ``lookup``, or ``insert``?
+      This test likely belongs in one of:
+
+      - tests.frame.indexing.test_methodname
+      - tests.series.indexing.test_methodname
+
+   C) Is the test for any of ``loc``, ``iloc``, ``at``, or ``iat``?
+      This test likely belongs in one of:
+
+      - tests.indexing.test_loc
+      - tests.indexing.test_iloc
+      - tests.indexing.test_at
+      - tests.indexing.test_iat
+
+      Within the appropriate file, test classes correspond to either types of
+      indexers (e.g. ``TestLocBooleanMask``) or major use cases
+      (e.g. ``TestLocSetitemWithExpansion``).
+
+      See the note in section D) about tests that test multiple indexing methods.
+
+   D) Is the test for ``Series.__getitem__``, ``Series.__setitem__``,
+      ``DataFrame.__getitem__``, or ``DataFrame.__setitem__``?
+      This test likely belongs in one of:
+
+      - tests.series.test_getitem
+      - tests.series.test_setitem
+      - tests.frame.test_getitem
+      - tests.frame.test_setitem
+
+      If many cases such a test may test multiple similar methods, e.g.
+
+      .. code-block:: python
+
+           import pandas as pd
+           import pandas._testing as tm
+
+           def test_getitem_listlike_of_ints():
+               ser = pd.Series(range(5))
+
+               result = ser[[3, 4]]
+               expected = pd.Series([2, 3])
+               tm.assert_series_equal(result, expected)
+
+               result = ser.loc[[3, 4]]
+               tm.assert_series_equal(result, expected)
+
+    In cases like this, the test location should be based on the *underlying*
+    method being tested.  Or in the case of a test for a bugfix, the location
+    of the actual bug.  So in this example, we know that ``Series.__getitem__``
+    calls ``Series.loc.__getitem__``, so this is *really* a test for
+    ``loc.__getitem__``.  So this test belongs in ``tests.indexing.test_loc``.
+
+6. Is your test for a DataFrame or Series method?
+
+   A) Is the method a plotting method?
+      This test likely belongs in one of:
+
+      - tests.plotting
+
+   B) Is the method an IO method?
+      This test likely belongs in one of:
+
+      - tests.io
+
+        .. note::
+
+            This includes ``to_string`` but excludes ``__repr__``, which is
+            tested in ``tests.frame.test_repr`` and ``tests.series.test_repr``.
+            Other classes often have a ``test_formats`` file.
+
+   C) Otherwise
+      This test likely belongs in one of:
+
+      - tests.series.methods.test_mymethod
+      - tests.frame.methods.test_mymethod
+
+        .. note::
+
+            If a test can be shared between DataFrame/Series using the
+            ``frame_or_series`` fixture, by convention it goes in the
+            ``tests.frame`` file.
+
+7. Is your test for an Index method, not depending on Series/DataFrame?
+   This test likely belongs in one of:
+
+   - tests.indexes
+
+8) Is your test for one of the pandas-provided ExtensionArrays (``Categorical``,
+   ``DatetimeArray``, ``TimedeltaArray``, ``PeriodArray``, ``IntervalArray``,
+   ``NumpyExtensionArray``, ``FloatArray``, ``BoolArray``, ``StringArray``)?
+   This test likely belongs in one of:
+
+   - tests.arrays
+
+9) Is your test for *all* ExtensionArray subclasses (the "EA Interface")?
+   This test likely belongs in one of:
+
+   - tests.extension
+
+Using ``pytest``
+~~~~~~~~~~~~~~~~
+
+Test structure
+^^^^^^^^^^^^^^
+
+pandas existing test structure is *mostly* class-based, meaning that you will typically find tests wrapped in a class.
+
+.. code-block:: python
+
+   class TestReallyCoolFeature:
+       def test_cool_feature_aspect(self):
+           pass
+
+We prefer a more *functional* style using the `pytest <https://docs.pytest.org/en/latest/>`__ framework, which offers a richer testing
+framework that will facilitate testing and developing. Thus, instead of writing test classes, we will write test functions like this:
+
+.. code-block:: python
+
+    def test_really_cool_feature():
+        pass
+
+Preferred ``pytest`` idioms
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+* Functional tests named ``def test_*`` and *only* take arguments that are either fixtures or parameters.
+* Use a bare ``assert`` for testing scalars and truth-testing
+* Use ``tm.assert_series_equal(result, expected)`` and ``tm.assert_frame_equal(result, expected)`` for comparing :class:`Series` and :class:`DataFrame` results respectively.
+* Use `@pytest.mark.parameterize <https://docs.pytest.org/en/latest/how-to/parametrize.html>`__ when testing multiple cases.
+* Use `pytest.mark.xfail <https://docs.pytest.org/en/latest/reference/reference.html?#pytest.mark.xfail>`__ when a test case is expected to fail.
+* Use `pytest.mark.skip <https://docs.pytest.org/en/latest/reference/reference.html?#pytest.mark.skip>`__ when a test case is never expected to pass.
+* Use `pytest.param <https://docs.pytest.org/en/latest/reference/reference.html?#pytest-param>`__ when a test case needs a particular mark.
+* Use `@pytest.fixture <https://docs.pytest.org/en/latest/reference/reference.html?#pytest-fixture>`__ if multiple tests can share a setup object.
+
+.. warning::
+
+    Do not use ``pytest.xfail`` (which is different than ``pytest.mark.xfail``) since it immediately stops the
+    test and does not check if the test will fail. If this is the behavior you desire, use ``pytest.skip`` instead.
+
+If a test is known to fail but the manner in which it fails
+is not meant to be captured, use ``pytest.mark.xfail`` It is common to use this method for a test that
+exhibits buggy behavior or a non-implemented feature. If
+the failing test has flaky behavior, use the argument ``strict=False``. This
+will make it so pytest does not fail if the test happens to pass. Using ``strict=False`` is highly undesirable, please use it only as a last resort.
+
+Prefer the decorator ``@pytest.mark.xfail`` and the argument ``pytest.param``
+over usage within a test so that the test is appropriately marked during the
+collection phase of pytest. For xfailing a test that involves multiple
+parameters, a fixture, or a combination of these, it is only possible to
+xfail during the testing phase. To do so, use the ``request`` fixture:
+
+.. code-block:: python
+
+    def test_xfail(request):
+        mark = pytest.mark.xfail(raises=TypeError, reason="Indicate why here")
+        request.applymarker(mark)
+
+xfail is not to be used for tests involving failure due to invalid user arguments.
+For these tests, we need to verify the correct exception type and error message
+is being raised, using ``pytest.raises`` instead.
+
+.. _contributing.warnings:
+
+Testing a warning
+^^^^^^^^^^^^^^^^^
+
+Use ``tm.assert_produces_warning`` as a context manager to check that a block of code raises a warning
+and specify the warning message using the ``match`` argument.
+
+.. code-block:: python
+
+    with tm.assert_produces_warning(DeprecationWarning, match="the warning message"):
+        pd.deprecated_function()
+
+If a warning should specifically not happen in a block of code, pass ``False`` into the context manager.
+
+.. code-block:: python
+
+    with tm.assert_produces_warning(False):
+        pd.no_warning_function()
+
+If you have a test that would emit a warning, but you aren't actually testing the
+warning itself (say because it's going to be removed in the future, or because we're
+matching a 3rd-party library's behavior), then use ``pytest.mark.filterwarnings`` to
+ignore the error.
+
+.. code-block:: python
+
+    @pytest.mark.filterwarnings("ignore:msg:category")
+    def test_thing(self):
+        pass
+
+Testing an exception
+^^^^^^^^^^^^^^^^^^^^
+
+Use `pytest.raises <https://docs.pytest.org/en/latest/reference/reference.html#pytest-raises>`_ as a context manager
+with the specific exception subclass (i.e. never use :py:class:`Exception`) and the exception message in ``match``.
+
+.. code-block:: python
+
+    with pytest.raises(ValueError, match="an error"):
+        raise ValueError("an error")
+
+Testing involving files
+^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``temp_file`` pytest fixture creates a temporary file :py:class:`Pathlib` object for testing:
+
+.. code-block:: python
+
+    def test_something(temp_file):
+        pd.DataFrame([1]).to_csv(str(temp_file))
+
+Please reference `pytest's documentation <https://docs.pytest.org/en/latest/how-to/tmp_path.html#the-default-base-temporary-directory>`_
+for the file retention policy.
+
+Testing involving network connectivity
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A unit test should not access a public data set over the internet due to flakiness of network connections and
+lack of ownership of the server that is being connected to. To mock this interaction, use the ``httpserver`` fixture from the
+`pytest-localserver plugin. <https://github.com/pytest-dev/pytest-localserver>`_ with synthetic data.
+
+.. code-block:: python
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_network(httpserver):
+        httpserver.serve_content(content="content")
+        result = pd.read_html(httpserver.url)
+
+Example
+^^^^^^^
+
+Here is an example of a self-contained set of tests in a file ``pandas/tests/test_cool_feature.py``
+that illustrate multiple features that we like to use. Please remember to add the GitHub Issue Number
+as a comment to a new test.
+
+.. code-block:: python
+
+   import pytest
+   import numpy as np
+   import pandas as pd
+
+
+   @pytest.mark.parametrize('dtype', ['int8', 'int16', 'int32', 'int64'])
+   def test_dtypes(dtype):
+       assert str(np.dtype(dtype)) == dtype
+
+
+   @pytest.mark.parametrize(
+       'dtype', ['float32', pytest.param('int16', marks=pytest.mark.skip),
+                 pytest.param('int32', marks=pytest.mark.xfail(
+                     reason='to show how it works'))])
+   def test_mark(dtype):
+       assert str(np.dtype(dtype)) == 'float32'
+
+
+   @pytest.fixture
+   def series():
+       return pd.Series([1, 2, 3])
+
+
+   @pytest.fixture(params=['int8', 'int16', 'int32', 'int64'])
+   def dtype(request):
+       return request.param
+
+
+   def test_series(series, dtype):
+       # GH <issue_number>
+       result = series.astype(dtype)
+       assert result.dtype == dtype
+
+       expected = pd.Series([1, 2, 3], dtype=dtype)
+       tm.assert_series_equal(result, expected)
+
+
+A test run of this yields
+
+.. code-block:: shell
+
+   ((pandas) bash-3.2$ pytest  test_cool_feature.py  -v
+   =========================== test session starts ===========================
+   platform darwin -- Python 3.6.2, pytest-3.6.0, py-1.4.31, pluggy-0.4.0
+   collected 11 items
+
+   tester.py::test_dtypes[int8] PASSED
+   tester.py::test_dtypes[int16] PASSED
+   tester.py::test_dtypes[int32] PASSED
+   tester.py::test_dtypes[int64] PASSED
+   tester.py::test_mark[float32] PASSED
+   tester.py::test_mark[int16] SKIPPED
+   tester.py::test_mark[int32] xfail
+   tester.py::test_series[int8] PASSED
+   tester.py::test_series[int16] PASSED
+   tester.py::test_series[int32] PASSED
+   tester.py::test_series[int64] PASSED
+
+Tests that we have ``parametrized`` are now accessible via the test name, for example we could run these with ``-k int8`` to sub-select *only* those tests which match ``int8``.
+
+
+.. code-block:: shell
+
+   ((pandas) bash-3.2$ pytest  test_cool_feature.py  -v -k int8
+   =========================== test session starts ===========================
+   platform darwin -- Python 3.6.2, pytest-3.6.0, py-1.4.31, pluggy-0.4.0
+   collected 11 items
+
+   test_cool_feature.py::test_dtypes[int8] PASSED
+   test_cool_feature.py::test_series[int8] PASSED
+
+
+.. _using-hypothesis:
+
+Using ``hypothesis``
+~~~~~~~~~~~~~~~~~~~~
+
+Hypothesis is a library for property-based testing. Instead of explicitly
+parametrizing a test, you can describe *all* valid inputs and let Hypothesis
+try to find a failing input.  Even better, no matter how many random examples
+it tries, Hypothesis always reports a single minimal counterexample to your
+assertions - often an example that you would never have thought to test.
+
+See `Getting Started with Hypothesis <https://hypothesis.works/articles/getting-started-with-hypothesis/>`_
+for more of an introduction, then `refer to the Hypothesis documentation
+for details <https://hypothesis.readthedocs.io/en/latest/index.html>`_.
+
+.. code-block:: python
+
+    import json
+    from hypothesis import given, strategies as st
+
+    any_json_value = st.deferred(lambda: st.one_of(
+        st.none(), st.booleans(), st.floats(allow_nan=False), st.text(),
+        st.lists(any_json_value), st.dictionaries(st.text(), any_json_value)
+    ))
+
+
+    @given(value=any_json_value)
+    def test_json_roundtrip(value):
+        result = json.loads(json.dumps(value))
+        assert value == result
+
+This test shows off several useful features of Hypothesis, as well as
+demonstrating a good use-case: checking properties that should hold over
+a large or complicated domain of inputs.
+
+To keep the pandas test suite running quickly, parametrized tests are
+preferred if the inputs or logic are simple, with Hypothesis tests reserved
+for cases with complex logic or where there are too many combinations of
+options or subtle interactions to test (or think of!) all of them.
+
+.. _contributing.running_tests:
+
+Running the test suite
+----------------------
+
+The tests can then be run directly inside your Git clone (without having to
+install pandas) by typing::
+
+    pytest pandas
+
+.. note::
+
+    If a handful of tests don't pass, it may not be an issue with your pandas installation.
+    Some tests (e.g. some SQLAlchemy ones) require additional setup, others might start
+    failing because a non-pinned library released a new version, and others might be flaky
+    if run in parallel. As long as you can import pandas from your locally built version,
+    your installation is probably fine and you can start contributing!
+
+Often it is worth running only a subset of tests first around your changes before running the
+entire suite.
+
+The easiest way to do this is with::
+
+    pytest pandas/path/to/test.py -k regex_matching_test_name
+
+Or with one of the following constructs::
+
+    pytest pandas/tests/[test-module].py
+    pytest pandas/tests/[test-module].py::[TestClass]
+    pytest pandas/tests/[test-module].py::[TestClass]::[test_method]
+
+Using `pytest-xdist <https://pypi.org/project/pytest-xdist>`_, which is
+included in our 'pandas-dev' environment, one can speed up local testing on
+multicore machines. The ``-n`` number flag then can be specified when running
+pytest to parallelize a test run across the number of specified cores or auto to
+utilize all the available cores on your machine.
+
+.. code-block:: bash
+
+   # Utilize 4 cores
+   pytest -n 4 pandas
+
+   # Utilizes all available cores
+   pytest -n auto pandas
+
+If you'd like to speed things along further a more advanced use of this
+command would look like this
+
+.. code-block:: bash
+
+    pytest pandas -n 4 -m "not slow and not network and not db and not single_cpu" -r sxX
+
+In addition to the multithreaded performance increase this improves test
+speed by skipping some tests using the ``-m`` mark flag:
+
+- slow: any test taking long (think seconds rather than milliseconds)
+- network: tests requiring network connectivity
+- db: tests requiring a database (mysql or postgres)
+- single_cpu: tests that should run on a single cpu only
+
+You might want to enable the following option if it's relevant for you:
+
+- arm_slow: any test taking long on arm64 architecture
+
+These markers are defined `in this toml file <https://github.com/pandas-dev/pandas/blob/main/pyproject.toml>`_
+, under ``[tool.pytest.ini_options]`` in a list called ``markers``, in case
+you want to check if new ones have been created which are of interest to you.
+
+The ``-r`` report flag will display a short summary info (see `pytest
+documentation <https://docs.pytest.org/en/4.6.x/usage.html#detailed-summary-report>`_)
+. Here we are displaying the number of:
+
+- s: skipped tests
+- x: xfailed tests
+- X: xpassed tests
+
+The summary is optional and can be removed if you don't need the added
+information. Using the parallelization option can significantly reduce the
+time it takes to locally run tests before submitting a pull request.
+
+If you require assistance with the results,
+which has happened in the past, please set a seed before running the command
+and opening a bug report, that way we can reproduce it. Here's an example
+for setting a seed on windows
+
+.. code-block:: bash
+
+    set PYTHONHASHSEED=314159265
+    pytest pandas -n 4 -m "not slow and not network and not db and not single_cpu" -r sxX
+
+On Unix use
+
+.. code-block:: bash
+
+    export PYTHONHASHSEED=314159265
+    pytest pandas -n 4 -m "not slow and not network and not db and not single_cpu" -r sxX
+
+For more, see the `pytest <https://docs.pytest.org/en/latest/>`_ documentation.
+
+Furthermore one can run
+
+.. code-block:: python
+
+   pd.test()
+
+with an imported pandas to run tests similarly.
+
+Running the performance test suite
+----------------------------------
+
+Performance matters and it is worth considering whether your code has introduced
+performance regressions. pandas is in the process of migrating to
+`asv benchmarks <https://github.com/airspeed-velocity/asv>`__
+to enable easy monitoring of the performance of critical pandas operations.
+These benchmarks are all found in the ``pandas/asv_bench`` directory, and the
+test results can be found `here <https://asv-runner.github.io/asv-collection/pandas>`__.
+
+To use all features of asv, you will need either ``conda`` or
+``virtualenv``. For more details please check the `asv installation
+webpage <https://asv.readthedocs.io/en/latest/installing.html>`_.
+
+To install asv::
+
+    pip install git+https://github.com/airspeed-velocity/asv
+
+If you need to run a benchmark, change your directory to ``asv_bench/`` and run::
+
+    asv continuous -f 1.1 upstream/main HEAD
+
+You can replace ``HEAD`` with the name of the branch you are working on,
+and report benchmarks that changed by more than 10%.
+The command uses ``conda`` by default for creating the benchmark
+environments. If you want to use virtualenv instead, write::
+
+    asv continuous -f 1.1 -E virtualenv upstream/main HEAD
+
+The ``-E virtualenv`` option should be added to all ``asv`` commands
+that run benchmarks. The default value is defined in ``asv.conf.json``.
+
+Running the full benchmark suite can be an all-day process, depending on your
+hardware and its resource utilization. However, usually it is sufficient to paste
+only a subset of the results into the pull request to show that the committed changes
+do not cause unexpected performance regressions.  You can run specific benchmarks
+using the ``-b`` flag, which takes a regular expression. For example, this will
+only run benchmarks from a ``pandas/asv_bench/benchmarks/groupby.py`` file::
+
+    asv continuous -f 1.1 upstream/main HEAD -b ^groupby
+
+If you want to only run a specific group of benchmarks from a file, you can do it
+using ``.`` as a separator. For example::
+
+    asv continuous -f 1.1 upstream/main HEAD -b groupby.GroupByMethods
+
+will only run the ``GroupByMethods`` benchmark defined in ``groupby.py``.
+
+You can also run the benchmark suite using the version of ``pandas``
+already installed in your current Python environment. This can be
+useful if you do not have virtualenv or conda, or are using the
+``setup.py develop`` approach discussed above; for the in-place build
+you need to set ``PYTHONPATH``, e.g.
+``PYTHONPATH="$PWD/.." asv [remaining arguments]``.
+You can run benchmarks using an existing Python
+environment by::
+
+    asv run -e -E existing
+
+or, to use a specific Python interpreter,::
+
+    asv run -e -E existing:python3.6
+
+This will display stderr from the benchmarks, and use your local
+``python`` that comes from your ``$PATH``.
+
+Information on how to write a benchmark and how to use asv can be found in the
+`asv documentation <https://asv.readthedocs.io/en/latest/writing_benchmarks.html>`_.
+
+Documenting your code
+---------------------
+
+Changes should be reflected in the release notes located in ``doc/source/whatsnew/vx.y.z.rst``.
+This file contains an ongoing change log for each release.  Add an entry to this file to
+document your fix, enhancement or (unavoidable) breaking change.  Make sure to include the
+GitHub issue number when adding your entry (using ``:issue:`1234``` where ``1234`` is the
+issue/pull request number). Your entry should be written using full sentences and proper
+grammar.
+
+When mentioning parts of the API, use a Sphinx ``:func:``, ``:meth:``, or ``:class:``
+directive as appropriate. Not all public API functions and methods have a
+documentation page; ideally links would only be added if they resolve. You can
+usually find similar examples by checking the release notes for one of the previous
+versions.
+
+If your code is a bugfix, add your entry to the relevant bugfix section. Avoid
+adding to the ``Other`` section; only in rare cases should entries go there.
+Being as concise as possible, the description of the bug should include how the
+user may encounter it and an indication of the bug itself, e.g.
+"produces incorrect results" or "incorrectly raises". It may be necessary to also
+indicate the new behavior.
+
+If your code is an enhancement, it is most likely necessary to add usage
+examples to the existing documentation.  This can be done following the section
+regarding :ref:`documentation <contributing_documentation>`.
+Further, to let users know when this feature was added, the ``versionadded``
+directive is used. The sphinx syntax for that is:
+
+.. code-block:: rst
+
+    .. versionadded:: 2.1.0
+
+This will put the text *New in version 2.1.0* wherever you put the sphinx
+directive. This should also be put in the docstring when adding a new function
+or method (`example <https://github.com/pandas-dev/pandas/blob/v0.20.2/pandas/core/frame.py#L1495>`__)
+or a new keyword argument (`example <https://github.com/pandas-dev/pandas/blob/v0.20.2/pandas/core/generic.py#L568>`__).
diff --git a/pandas/doc/source/development/contributing_docstring.rst b/pandas/doc/source/development/contributing_docstring.rst
new file mode 100644
index 0000000000000..59d7957275e15
--- /dev/null
+++ b/pandas/doc/source/development/contributing_docstring.rst
@@ -0,0 +1,999 @@
+.. _docstring:
+
+{{ header }}
+
+======================
+pandas docstring guide
+======================
+
+About docstrings and standards
+------------------------------
+
+A Python docstring is a string used to document a Python module, class,
+function or method, so programmers can understand what it does without having
+to read the details of the implementation.
+
+Also, it is a common practice to generate online (html) documentation
+automatically from docstrings. `Sphinx <https://www.sphinx-doc.org>`_ serves
+this purpose.
+
+The next example gives an idea of what a docstring looks like:
+
+.. code-block:: python
+
+    def add(num1, num2):
+        """
+        Add up two integer numbers.
+
+        This function simply wraps the ``+`` operator, and does not
+        do anything interesting, except for illustrating what
+        the docstring of a very simple function looks like.
+
+        Parameters
+        ----------
+        num1 : int
+            First number to add.
+        num2 : int
+            Second number to add.
+
+        Returns
+        -------
+        int
+            The sum of ``num1`` and ``num2``.
+
+        See Also
+        --------
+        subtract : Subtract one integer from another.
+
+        Examples
+        --------
+        >>> add(2, 2)
+        4
+        >>> add(25, 0)
+        25
+        >>> add(10, -10)
+        0
+        """
+        return num1 + num2
+
+Some standards regarding docstrings exist, which make them easier to read, and allow them
+be easily exported to other formats such as html or pdf.
+
+The first conventions every Python docstring should follow are defined in
+`PEP-257 <https://www.python.org/dev/peps/pep-0257/>`_.
+
+As PEP-257 is quite broad, other more specific standards also exist. In the
+case of pandas, the NumPy docstring convention is followed. These conventions are
+explained in this document:
+
+* `numpydoc docstring guide <https://numpydoc.readthedocs.io/en/latest/format.html>`_
+
+numpydoc is a Sphinx extension to support the NumPy docstring convention.
+
+The standard uses reStructuredText (reST). reStructuredText is a markup
+language that allows encoding styles in plain text files. Documentation
+about reStructuredText can be found in:
+
+* `Sphinx reStructuredText primer <https://www.sphinx-doc.org/en/stable/rest.html>`_
+* `Quick reStructuredText reference <https://docutils.sourceforge.io/docs/user/rst/quickref.html>`_
+* `Full reStructuredText specification <https://docutils.sourceforge.io/docs/ref/rst/restructuredtext.html>`_
+
+pandas has some helpers for sharing docstrings between related classes, see
+:ref:`docstring.sharing`.
+
+The rest of this document will summarize all the above guidelines, and will
+provide additional conventions specific to the pandas project.
+
+.. _docstring.tutorial:
+
+Writing a docstring
+-------------------
+
+.. _docstring.general:
+
+General rules
+~~~~~~~~~~~~~
+
+Docstrings must be defined with three double-quotes. No blank lines should be
+left before or after the docstring. The text starts in the next line after the
+opening quotes. The closing quotes have their own line
+(meaning that they are not at the end of the last sentence).
+
+On rare occasions reST styles like bold text or italics will be used in
+docstrings, but is it common to have inline code, which is presented between
+backticks. The following are considered inline code:
+
+* The name of a parameter
+* Python code, a module, function, built-in, type, literal... (e.g. ``os``,
+  ``list``, ``numpy.abs``, ``datetime.date``, ``True``)
+* A pandas class (in the form ``:class:`pandas.Series```)
+* A pandas method (in the form ``:meth:`pandas.Series.sum```)
+* A pandas function (in the form ``:func:`pandas.to_datetime```)
+
+.. note::
+    To display only the last component of the linked class, method or
+    function, prefix it with ``~``. For example, ``:class:`~pandas.Series```
+    will link to ``pandas.Series`` but only display the last part, ``Series``
+    as the link text. See `Sphinx cross-referencing syntax
+    <https://www.sphinx-doc.org/en/stable/domains.html#cross-referencing-syntax>`_
+    for details.
+
+**Good:**
+
+.. code-block:: python
+
+    def add_values(arr):
+        """
+        Add the values in ``arr``.
+
+        This is equivalent to Python ``sum`` of :meth:`pandas.Series.sum`.
+
+        Some sections are omitted here for simplicity.
+        """
+        return sum(arr)
+
+**Bad:**
+
+.. code-block:: python
+
+    def func():
+
+        """Some function.
+
+        With several mistakes in the docstring.
+
+        It has a blank line after the signature ``def func():``.
+
+        The text 'Some function' should go in the line after the
+        opening quotes of the docstring, not in the same line.
+
+        There is a blank line between the docstring and the first line
+        of code ``foo = 1``.
+
+        The closing quotes should be in the next line, not in this one."""
+
+        foo = 1
+        bar = 2
+        return foo + bar
+
+.. _docstring.short_summary:
+
+Section 1: short summary
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+The short summary is a single sentence that expresses what the function does in
+a concise way.
+
+The short summary must start with a capital letter, end with a dot, and fit in
+a single line. It needs to express what the object does without providing
+details. For functions and methods, the short summary must start with an
+infinitive verb.
+
+**Good:**
+
+.. code-block:: python
+
+    def astype(dtype):
+        """
+        Cast Series type.
+
+        This section will provide further details.
+        """
+        pass
+
+**Bad:**
+
+.. code-block:: python
+
+    def astype(dtype):
+        """
+        Casts Series type.
+
+        Verb in third-person of the present simple, should be infinitive.
+        """
+        pass
+
+.. code-block:: python
+
+    def astype(dtype):
+        """
+        Method to cast Series type.
+
+        Does not start with verb.
+        """
+        pass
+
+.. code-block:: python
+
+    def astype(dtype):
+        """
+        Cast Series type
+
+        Missing dot at the end.
+        """
+        pass
+
+.. code-block:: python
+
+    def astype(dtype):
+        """
+        Cast Series type from its current type to the new type defined in
+        the parameter dtype.
+
+        Summary is too verbose and doesn't fit in a single line.
+        """
+        pass
+
+.. _docstring.extended_summary:
+
+Section 2: extended summary
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The extended summary provides details on what the function does. It should not
+go into the details of the parameters, or discuss implementation notes, which
+go in other sections.
+
+A blank line is left between the short summary and the extended summary.
+Every paragraph in the extended summary ends with a dot.
+
+The extended summary should provide details on why the function is useful and
+their use cases, if it is not too generic.
+
+.. code-block:: python
+
+    def unstack():
+        """
+        Pivot a row index to columns.
+
+        When using a MultiIndex, a level can be pivoted so each value in
+        the index becomes a column. This is especially useful when a subindex
+        is repeated for the main index, and data is easier to visualize as a
+        pivot table.
+
+        The index level will be automatically removed from the index when added
+        as columns.
+        """
+        pass
+
+.. _docstring.parameters:
+
+Section 3: parameters
+~~~~~~~~~~~~~~~~~~~~~
+
+The details of the parameters will be added in this section. This section has
+the title "Parameters", followed by a line with a hyphen under each letter of
+the word "Parameters". A blank line is left before the section title, but not
+after, and not between the line with the word "Parameters" and the one with
+the hyphens.
+
+After the title, each parameter in the signature must be documented, including
+``*args`` and ``**kwargs``, but not ``self``.
+
+The parameters are defined by their name, followed by a space, a colon, another
+space, and the type (or types). Note that the space between the name and the
+colon is important. Types are not defined for ``*args`` and ``**kwargs``, but must
+be defined for all other parameters. After the parameter definition, it is
+required to have a line with the parameter description, which is indented, and
+can have multiple lines. The description must start with a capital letter, and
+finish with a dot.
+
+For keyword arguments with a default value, the default will be listed after a
+comma at the end of the type. The exact form of the type in this case will be
+"int, default 0". In some cases it may be useful to explain what the default
+argument means, which can be added after a comma "int, default -1, meaning all
+cpus".
+
+In cases where the default value is ``None``, meaning that the value will not be
+used. Instead of ``"str, default None"``, it is preferred to write ``"str, optional"``.
+When ``None`` is a value being used, we will keep the form "str, default None".
+For example, in ``df.to_csv(compression=None)``, ``None`` is not a value being used,
+but means that compression is optional, and no compression is being used if not
+provided. In this case we will use ``"str, optional"``. Only in cases like
+``func(value=None)`` and ``None`` is being used in the same way as ``0`` or ``foo``
+would be used, then we will specify "str, int or None, default None".
+
+**Good:**
+
+.. code-block:: python
+
+    class Series:
+        def plot(self, kind, color='blue', **kwargs):
+            """
+            Generate a plot.
+
+            Render the data in the Series as a matplotlib plot of the
+            specified kind.
+
+            Parameters
+            ----------
+            kind : str
+                Kind of matplotlib plot.
+            color : str, default 'blue'
+                Color name or rgb code.
+            **kwargs
+                These parameters will be passed to the matplotlib plotting
+                function.
+            """
+            pass
+
+**Bad:**
+
+.. code-block:: python
+
+    class Series:
+        def plot(self, kind, **kwargs):
+            """
+            Generate a plot.
+
+            Render the data in the Series as a matplotlib plot of the
+            specified kind.
+
+            Note the blank line between the parameters title and the first
+            parameter. Also, note that after the name of the parameter ``kind``
+            and before the colon, a space is missing.
+
+            Also, note that the parameter descriptions do not start with a
+            capital letter, and do not finish with a dot.
+
+            Finally, the ``**kwargs`` parameter is missing.
+
+            Parameters
+            ----------
+
+            kind: str
+                kind of matplotlib plot
+            """
+            pass
+
+.. _docstring.parameter_types:
+
+Parameter types
+^^^^^^^^^^^^^^^
+
+When specifying the parameter types, Python built-in data types can be used
+directly (the Python type is preferred to the more verbose string, integer,
+boolean, etc):
+
+* int
+* float
+* str
+* bool
+
+For complex types, define the subtypes. For ``dict`` and ``tuple``, as more than
+one type is present, we use the brackets to help read the type (curly brackets
+for ``dict`` and normal brackets for ``tuple``):
+
+* list of int
+* dict of {str : int}
+* tuple of (str, int, int)
+* tuple of (str,)
+* set of str
+
+In case where there are just a set of values allowed, list them in curly
+brackets and separated by commas (followed by a space). If the values are
+ordinal and they have an order, list them in this order. Otherwise, list
+the default value first, if there is one:
+
+* {0, 10, 25}
+* {'simple', 'advanced'}
+* {'low', 'medium', 'high'}
+* {'cat', 'dog', 'bird'}
+
+If the type is defined in a Python module, the module must be specified:
+
+* datetime.date
+* datetime.datetime
+* decimal.Decimal
+
+If the type is in a package, the module must be also specified:
+
+* numpy.ndarray
+* scipy.sparse.coo_matrix
+
+If the type is a pandas type, also specify pandas except for Series and
+DataFrame:
+
+* Series
+* DataFrame
+* pandas.Index
+* pandas.Categorical
+* pandas.arrays.SparseArray
+
+If the exact type is not relevant, but must be compatible with a NumPy
+array, array-like can be specified. If Any type that can be iterated is
+accepted, iterable can be used:
+
+* array-like
+* iterable
+
+If more than one type is accepted, separate them by commas, except the
+last two types, that need to be separated by the word 'or':
+
+* int or float
+* float, decimal.Decimal or None
+* str or list of str
+
+If ``None`` is one of the accepted values, it always needs to be the last in
+the list.
+
+For axis, the convention is to use something like:
+
+* axis : {0 or 'index', 1 or 'columns', None}, default None
+
+.. _docstring.returns:
+
+Section 4: returns or yields
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If the method returns a value, it will be documented in this section. Also
+if the method yields its output.
+
+The title of the section will be defined in the same way as the "Parameters".
+With the names "Returns" or "Yields" followed by a line with as many hyphens
+as the letters in the preceding word.
+
+The documentation of the return is also similar to the parameters. But in this
+case, no name will be provided, unless the method returns or yields more than
+one value (a tuple of values).
+
+The types for "Returns" and "Yields" are the same as the ones for the
+"Parameters". Also, the description must finish with a dot.
+
+For example, with a single value:
+
+.. code-block:: python
+
+    def sample():
+        """
+        Generate and return a random number.
+
+        The value is sampled from a continuous uniform distribution between
+        0 and 1.
+
+        Returns
+        -------
+        float
+            Random number generated.
+        """
+        return np.random.random()
+
+With more than one value:
+
+.. code-block:: python
+
+    import string
+
+    def random_letters():
+        """
+        Generate and return a sequence of random letters.
+
+        The length of the returned string is also random, and is also
+        returned.
+
+        Returns
+        -------
+        length : int
+            Length of the returned string.
+        letters : str
+            String of random letters.
+        """
+        length = np.random.randint(1, 10)
+        letters = ''.join(np.random.choice(string.ascii_lowercase)
+                          for i in range(length))
+        return length, letters
+
+If the method yields its value:
+
+.. code-block:: python
+
+    def sample_values():
+        """
+        Generate an infinite sequence of random numbers.
+
+        The values are sampled from a continuous uniform distribution between
+        0 and 1.
+
+        Yields
+        ------
+        float
+            Random number generated.
+        """
+        while True:
+            yield np.random.random()
+
+.. _docstring.see_also:
+
+Section 5: see also
+~~~~~~~~~~~~~~~~~~~
+
+This section is used to let users know about pandas functionality
+related to the one being documented. In rare cases, if no related methods
+or functions can be found at all, this section can be skipped.
+
+An obvious example would be the ``head()`` and ``tail()`` methods. As ``tail()`` does
+the equivalent as ``head()`` but at the end of the ``Series`` or ``DataFrame``
+instead of at the beginning, it is good to let the users know about it.
+
+To give an intuition on what can be considered related, here there are some
+examples:
+
+* ``loc`` and ``iloc``, as they do the same, but in one case providing indices
+  and in the other positions
+* ``max`` and ``min``, as they do the opposite
+* ``iterrows``, ``itertuples`` and ``items``, as it is easy that a user
+  looking for the method to iterate over columns ends up in the method to
+  iterate over rows, and vice-versa
+* ``fillna`` and ``dropna``, as both methods are used to handle missing values
+* ``read_csv`` and ``to_csv``, as they are complementary
+* ``merge`` and ``join``, as one is a generalization of the other
+* ``astype`` and ``pandas.to_datetime``, as users may be reading the
+  documentation of ``astype`` to know how to cast as a date, and the way to do
+  it is with ``pandas.to_datetime``
+* ``where`` is related to ``numpy.where``, as its functionality is based on it
+
+When deciding what is related, you should mainly use your common sense and
+think about what can be useful for the users reading the documentation,
+especially the less experienced ones.
+
+When relating to other libraries (mainly ``numpy``), use the name of the module
+first (not an alias like ``np``). If the function is in a module which is not
+the main one, like ``scipy.sparse``, list the full module (e.g.
+``scipy.sparse.coo_matrix``).
+
+This section has a header, "See Also" (note the capital
+S and A), followed by the line with hyphens and preceded by a blank line.
+
+After the header, we will add a line for each related method or function,
+followed by a space, a colon, another space, and a short description that
+illustrates what this method or function does, why is it relevant in this
+context, and what the key differences are between the documented function and
+the one being referenced. The description must also end with a dot.
+
+Note that in "Returns" and "Yields", the description is located on the line
+after the type. In this section, however, it is located on the same
+line, with a colon in between. If the description does not fit on the same
+line, it can continue onto other lines which must be further indented.
+
+For example:
+
+.. code-block:: python
+
+    class Series:
+        def head(self):
+            """
+            Return the first 5 elements of the Series.
+
+            This function is mainly useful to preview the values of the
+            Series without displaying the whole of it.
+
+            Returns
+            -------
+            Series
+                Subset of the original series with the 5 first values.
+
+            See Also
+            --------
+            Series.tail : Return the last 5 elements of the Series.
+            Series.iloc : Return a slice of the elements in the Series,
+                which can also be used to return the first or last n.
+            """
+            return self.iloc[:5]
+
+.. _docstring.notes:
+
+Section 6: notes
+~~~~~~~~~~~~~~~~
+
+This is an optional section used for notes about the implementation of the
+algorithm, or to document technical aspects of the function behavior.
+
+Feel free to skip it, unless you are familiar with the implementation of the
+algorithm, or you discover some counter-intuitive behavior while writing the
+examples for the function.
+
+This section follows the same format as the extended summary section.
+
+.. _docstring.examples:
+
+Section 7: examples
+~~~~~~~~~~~~~~~~~~~
+
+This is one of the most important sections of a docstring, despite being
+placed in the last position, as often people understand concepts better
+by example than through accurate explanations.
+
+Examples in docstrings, besides illustrating the usage of the function or
+method, must be valid Python code, that returns the given output in a
+deterministic way, and that can be copied and run by users.
+
+Examples are presented as a session in the Python terminal. ``>>>`` is used to
+present code. ``...`` is used for code continuing from the previous line.
+Output is presented immediately after the last line of code generating the
+output (no blank lines in between). Comments describing the examples can
+be added with blank lines before and after them.
+
+The way to present examples is as follows:
+
+1. Import required libraries (except ``numpy`` and ``pandas``)
+
+2. Create the data required for the example
+
+3. Show a very basic example that gives an idea of the most common use case
+
+4. Add examples with explanations that illustrate how the parameters can be
+   used for extended functionality
+
+A simple example could be:
+
+.. code-block:: python
+
+    class Series:
+
+        def head(self, n=5):
+            """
+            Return the first elements of the Series.
+
+            This function is mainly useful to preview the values of the
+            Series without displaying all of it.
+
+            Parameters
+            ----------
+            n : int
+                Number of values to return.
+
+            Return
+            ------
+            pandas.Series
+                Subset of the original series with the n first values.
+
+            See Also
+            --------
+            tail : Return the last n elements of the Series.
+
+            Examples
+            --------
+            >>> ser = pd.Series(['Ant', 'Bear', 'Cow', 'Dog', 'Falcon',
+            ...                'Lion', 'Monkey', 'Rabbit', 'Zebra'])
+            >>> ser.head()
+            0   Ant
+            1   Bear
+            2   Cow
+            3   Dog
+            4   Falcon
+            dtype: object
+
+            With the ``n`` parameter, we can change the number of returned rows:
+
+            >>> ser.head(n=3)
+            0   Ant
+            1   Bear
+            2   Cow
+            dtype: object
+            """
+            return self.iloc[:n]
+
+The examples should be as concise as possible. In cases where the complexity of
+the function requires long examples, is recommended to use blocks with headers
+in bold. Use double star ``**`` to make a text bold, like in ``**this example**``.
+
+.. _docstring.example_conventions:
+
+Conventions for the examples
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Code in examples is assumed to always start with these two lines which are not
+shown:
+
+.. code-block:: python
+
+    import numpy as np
+    import pandas as pd
+
+Any other module used in the examples must be explicitly imported, one per line (as
+recommended in :pep:`8#imports`)
+and avoiding aliases. Avoid excessive imports, but if needed, imports from
+the standard library go first, followed by third-party libraries (like
+matplotlib).
+
+When illustrating examples with a single ``Series`` use the name ``ser``, and if
+illustrating with a single ``DataFrame`` use the name ``df``. For indices,
+``idx`` is the preferred name. If a set of homogeneous ``Series`` or
+``DataFrame`` is used, name them ``ser1``, ``ser2``, ``ser3``...  or ``df1``,
+``df2``, ``df3``... If the data is not homogeneous, and more than one structure
+is needed, name them with something meaningful, for example ``df_main`` and
+``df_to_join``.
+
+Data used in the example should be as compact as possible. The number of rows
+is recommended to be around 4, but make it a number that makes sense for the
+specific example. For example in the ``head`` method, it requires to be higher
+than 5, to show the example with the default values. If doing the ``mean``, we
+could use something like ``[1, 2, 3]``, so it is easy to see that the value
+returned is the mean.
+
+For more complex examples (grouping for example), avoid using data without
+interpretation, like a matrix of random numbers with columns A, B, C, D...
+And instead use a meaningful example, which makes it easier to understand the
+concept. Unless required by the example, use names of animals, to keep examples
+consistent. And numerical properties of them.
+
+When calling the method, keywords arguments ``head(n=3)`` are preferred to
+positional arguments ``head(3)``.
+
+**Good:**
+
+.. code-block:: python
+
+    class Series:
+
+        def mean(self):
+            """
+            Compute the mean of the input.
+
+            Examples
+            --------
+            >>> ser = pd.Series([1, 2, 3])
+            >>> ser.mean()
+            2
+            """
+            pass
+
+
+        def fillna(self, value):
+            """
+            Replace missing values by ``value``.
+
+            Examples
+            --------
+            >>> ser = pd.Series([1, np.nan, 3])
+            >>> ser.fillna(0)
+            [1, 0, 3]
+            """
+            pass
+
+        def groupby_mean(self):
+            """
+            Group by index and return mean.
+
+            Examples
+            --------
+            >>> ser = pd.Series([380., 370., 24., 26],
+            ...               name='max_speed',
+            ...               index=['falcon', 'falcon', 'parrot', 'parrot'])
+            >>> ser.groupby_mean()
+            index
+            falcon    375.0
+            parrot     25.0
+            Name: max_speed, dtype: float64
+            """
+            pass
+
+        def contains(self, pattern, case_sensitive=True, na=numpy.nan):
+            """
+            Return whether each value contains ``pattern``.
+
+            In this case, we are illustrating how to use sections, even
+            if the example is simple enough and does not require them.
+
+            Examples
+            --------
+            >>> ser = pd.Series('Antelope', 'Lion', 'Zebra', np.nan)
+            >>> ser.contains(pattern='a')
+            0    False
+            1    False
+            2     True
+            3      NaN
+            dtype: bool
+
+            **Case sensitivity**
+
+            With ``case_sensitive`` set to ``False`` we can match ``a`` with both
+            ``a`` and ``A``:
+
+            >>> s.contains(pattern='a', case_sensitive=False)
+            0     True
+            1    False
+            2     True
+            3      NaN
+            dtype: bool
+
+            **Missing values**
+
+            We can fill missing values in the output using the ``na`` parameter:
+
+            >>> ser.contains(pattern='a', na=False)
+            0    False
+            1    False
+            2     True
+            3    False
+            dtype: bool
+            """
+            pass
+
+**Bad:**
+
+.. code-block:: python
+
+    def method(foo=None, bar=None):
+        """
+        A sample DataFrame method.
+
+        Do not import NumPy and pandas.
+
+        Try to use meaningful data, when it makes the example easier
+        to understand.
+
+        Try to avoid positional arguments like in ``df.method(1)``. They
+        can be all right if previously defined with a meaningful name,
+        like in ``present_value(interest_rate)``, but avoid them otherwise.
+
+        When presenting the behavior with different parameters, do not place
+        all the calls one next to the other. Instead, add a short sentence
+        explaining what the example shows.
+
+        Examples
+        --------
+        >>> import numpy as np
+        >>> import pandas as pd
+        >>> df = pd.DataFrame(np.random.randn(3, 3),
+        ...                   columns=('a', 'b', 'c'))
+        >>> df.method(1)
+        21
+        >>> df.method(bar=14)
+        123
+        """
+        pass
+
+
+.. _docstring.doctest_tips:
+
+Tips for getting your examples pass the doctests
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Getting the examples pass the doctests in the validation script can sometimes
+be tricky. Here are some attention points:
+
+* Import all needed libraries (except for pandas and NumPy, those are already
+  imported as ``import pandas as pd`` and ``import numpy as np``) and define
+  all variables you use in the example.
+
+* Try to avoid using random data. However random data might be OK in some
+  cases, like if the function you are documenting deals with probability
+  distributions, or if the amount of data needed to make the function result
+  meaningful is too much, such that creating it manually is very cumbersome.
+  In those cases, always use a fixed random seed to make the generated examples
+  predictable. Example::
+
+    >>> np.random.seed(42)
+    >>> df = pd.DataFrame({'normal': np.random.normal(100, 5, 20)})
+
+* If you have a code snippet that wraps multiple lines, you need to use '...'
+  on the continued lines: ::
+
+    >>> df = pd.DataFrame([[1, 2, 3], [4, 5, 6]], index=['a', 'b', 'c'],
+    ...                   columns=['A', 'B'])
+
+* If you want to show a case where an exception is raised, you can do::
+
+    >>> pd.to_datetime(["712-01-01"])
+    Traceback (most recent call last):
+    OutOfBoundsDatetime: Out of bounds nanosecond timestamp: 712-01-01 00:00:00
+
+  It is essential to include the "Traceback (most recent call last):", but for
+  the actual error only the error name is sufficient.
+
+* If there is a small part of the result that can vary (e.g. a hash in an object
+  representation), you can use ``...`` to represent this part.
+
+  If you want to show that ``s.plot()`` returns a matplotlib AxesSubplot object,
+  this will fail the doctest ::
+
+    >>> s.plot()
+    <matplotlib.axes._subplots.AxesSubplot at 0x7efd0c0b0690>
+
+  However, you can do (notice the comment that needs to be added) ::
+
+    >>> s.plot()  # doctest: +ELLIPSIS
+    <matplotlib.axes._subplots.AxesSubplot at ...>
+
+
+.. _docstring.example_plots:
+
+Plots in examples
+^^^^^^^^^^^^^^^^^
+
+There are some methods in pandas returning plots. To render the plots generated
+by the examples in the documentation, the ``.. plot::`` directive exists.
+
+To use it, place the next code after the "Examples" header as shown below. The
+plot will be generated automatically when building the documentation.
+
+.. code-block:: python
+
+    class Series:
+        def plot(self):
+            """
+            Generate a plot with the ``Series`` data.
+
+            Examples
+            --------
+
+            .. plot::
+                :context: close-figs
+
+                >>> ser = pd.Series([1, 2, 3])
+                >>> ser.plot()
+            """
+            pass
+
+.. _docstring.sharing:
+
+Sharing docstrings
+------------------
+
+pandas has a system for sharing docstrings, with slight variations, between
+classes. This helps us keep docstrings consistent, while keeping things clear
+for the user reading. It comes at the cost of some complexity when writing.
+
+Each shared docstring will have a base template with variables, like
+``{klass}``. The variables filled in later on using the ``doc`` decorator.
+Finally, docstrings can also be appended to with the ``doc`` decorator.
+
+In this example, we'll create a parent docstring normally (this is like
+``pandas.core.generic.NDFrame``). Then we'll have two children (like
+``pandas.Series`` and ``pandas.DataFrame``). We'll
+substitute the class names in this docstring.
+
+.. code-block:: python
+
+   class Parent:
+       @doc(klass="Parent")
+       def my_function(self):
+           """Apply my function to {klass}."""
+           ...
+
+
+   class ChildA(Parent):
+       @doc(Parent.my_function, klass="ChildA")
+       def my_function(self):
+           ...
+
+
+   class ChildB(Parent):
+       @doc(Parent.my_function, klass="ChildB")
+       def my_function(self):
+           ...
+
+The resulting docstrings are
+
+.. code-block:: python
+
+   >>> print(Parent.my_function.__doc__)
+   Apply my function to Parent.
+   >>> print(ChildA.my_function.__doc__)
+   Apply my function to ChildA.
+   >>> print(ChildB.my_function.__doc__)
+   Apply my function to ChildB.
+
+Notice:
+
+1. We "append" the parent docstring to the children docstrings, which are
+   initially empty.
+
+Our files will often contain a module-level ``_shared_doc_kwargs`` with some
+common substitution values (things like ``klass``, ``axes``, etc).
+
+You can substitute and append in one shot with something like
+
+.. code-block:: python
+
+   @doc(template, **_shared_doc_kwargs)
+   def my_function(self):
+       ...
+
+where ``template`` may come from a module-level ``_shared_docs`` dictionary
+mapping function names to docstrings. Wherever possible, we prefer using
+``doc``, since the docstring-writing processes is slightly closer to normal.
+
+See ``pandas.core.generic.NDFrame.fillna`` for an example template, and
+``pandas.Series.fillna`` and ``pandas.core.generic.frame.fillna``
+for the filled versions.
diff --git a/pandas/doc/source/development/contributing_documentation.rst b/pandas/doc/source/development/contributing_documentation.rst
new file mode 100644
index 0000000000000..443470e6c50f9
--- /dev/null
+++ b/pandas/doc/source/development/contributing_documentation.rst
@@ -0,0 +1,215 @@
+.. _contributing_documentation:
+
+{{ header }}
+
+=================================
+Contributing to the documentation
+=================================
+
+Contributing to the documentation benefits everyone who uses pandas.
+We encourage you to help us improve the documentation, and
+you don't have to be an expert on pandas to do so! In fact,
+there are sections of the docs that are worse off after being written by
+experts. If something in the docs doesn't make sense to you, updating the
+relevant section after you figure it out is a great way to ensure it will help
+the next person. Please visit the `issues page <https://github.com/pandas-dev/pandas/issues?page=1&q=is%3Aopen+sort%3Aupdated-desc+label%3ADocs>`__
+for a full list of issues that are currently open regarding the
+pandas documentation.
+
+
+
+.. contents:: Documentation:
+   :local:
+
+
+About the pandas documentation
+--------------------------------
+
+The documentation is written in **reStructuredText**, which is almost like writing
+in plain English, and built using `Sphinx <https://www.sphinx-doc.org/en/master/>`__. The
+Sphinx Documentation has an excellent `introduction to reST
+<https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html>`__. Review the Sphinx docs to perform more
+complex changes to the documentation as well.
+
+Some other important things to know about the docs:
+
+* The pandas documentation consists of two parts: the docstrings in the code
+  itself and the docs in this folder ``doc/``.
+
+  The docstrings provide a clear explanation of the usage of the individual
+  functions, while the documentation in this folder consists of tutorial-like
+  overviews per topic together with some other information (what's new,
+  installation, etc).
+
+* The docstrings follow a pandas convention, based on the **Numpy Docstring
+  Standard**. Follow the :ref:`pandas docstring guide <docstring>` for detailed
+  instructions on how to write a correct docstring.
+
+  .. toctree::
+     :maxdepth: 2
+
+     contributing_docstring.rst
+
+* The tutorials make heavy use of the `IPython directive
+  <https://matplotlib.org/sampledoc/ipython_directive.html>`_ sphinx extension.
+  This directive lets you put code in the documentation which will be run
+  during the doc build. For example::
+
+      .. ipython:: python
+
+          x = 2
+          x**3
+
+  will be rendered as::
+
+      In [1]: x = 2
+
+      In [2]: x**3
+      Out[2]: 8
+
+  Almost all code examples in the docs are run (and the output saved) during the
+  doc build. This approach means that code examples will always be up to date,
+  but it does make the doc building a bit more complex.
+
+* Our API documentation files in ``doc/source/reference`` house the auto-generated
+  documentation from the docstrings. For classes, there are a few subtleties
+  around controlling which methods and attributes have pages auto-generated.
+
+  We have two autosummary templates for classes.
+
+  1. ``_templates/autosummary/class.rst``. Use this when you want to
+     automatically generate a page for every public method and attribute on the
+     class. The ``Attributes`` and ``Methods`` sections will be automatically
+     added to the class' rendered documentation by numpydoc. See ``DataFrame``
+     for an example.
+
+  2. ``_templates/autosummary/class_without_autosummary``. Use this when you
+     want to pick a subset of methods / attributes to auto-generate pages for.
+     When using this template, you should include an ``Attributes`` and
+     ``Methods`` section in the class docstring. See ``CategoricalIndex`` for an
+     example.
+
+  Every method should be included in a ``toctree`` in one of the documentation files in
+  ``doc/source/reference``, else Sphinx
+  will emit a warning.
+
+The utility script ``scripts/validate_docstrings.py`` can be used to get a csv
+summary of the API documentation. And also validate common errors in the docstring
+of a specific class, function or method. The summary also compares the list of
+methods documented in the files in ``doc/source/reference`` (which is used to generate
+the `API Reference <https://pandas.pydata.org/pandas-docs/stable/api.html>`_ page)
+and the actual public methods.
+This will identify methods documented in ``doc/source/reference`` that are not actually
+class methods, and existing methods that are not documented in ``doc/source/reference``.
+
+
+Updating a pandas docstring
+-----------------------------
+
+When improving a single function or method's docstring, it is not necessarily
+needed to build the full documentation (see next section).
+However, there is a script that checks a docstring (for example for the ``DataFrame.mean`` method)::
+
+    python scripts/validate_docstrings.py pandas.DataFrame.mean
+
+This script will indicate some formatting errors if present, and will also
+run and test the examples included in the docstring.
+Check the :ref:`pandas docstring guide <docstring>` for a detailed guide
+on how to format the docstring.
+
+The examples in the docstring ('doctests') must be valid Python code,
+that in a deterministic way returns the presented output, and that can be
+copied and run by users. This can be checked with the script above, and is
+also tested on Travis. A failing doctest will be a blocker for merging a PR.
+Check the :ref:`examples <docstring.examples>` section in the docstring guide
+for some tips and tricks to get the doctests passing.
+
+When doing a PR with a docstring update, it is good to post the
+output of the validation script in a comment on github.
+
+.. _contributing.howto-build-docs:
+
+How to build the pandas documentation
+---------------------------------------
+
+Requirements
+~~~~~~~~~~~~
+
+First, you need to have a development environment to be able to build pandas
+(see the docs on :ref:`creating a development environment <contributing_environment>`).
+
+Building the documentation
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+So how do you build the docs? Navigate to your local
+``doc/`` directory in the console and run::
+
+    python make.py html
+
+Then you can find the HTML output in the folder ``doc/build/html/``.
+
+The first time you build the docs, it will take quite a while because it has to run
+all the code examples and build all the generated docstring pages. In subsequent
+evocations, sphinx will try to only build the pages that have been modified.
+
+If you want to do a full clean build, do::
+
+    python make.py clean
+    python make.py html
+
+You can tell ``make.py`` to compile only a single section of the docs, greatly
+reducing the turn-around time for checking your changes.
+
+::
+
+    # omit autosummary and API section
+    python make.py clean
+    python make.py --no-api
+
+    # compile the docs with only a single section, relative to the "source" folder.
+    # For example, compiling only this guide (doc/source/development/contributing.rst)
+    python make.py clean
+    python make.py --single development/contributing.rst
+
+    # compile the reference docs for a single function
+    python make.py clean
+    python make.py --single pandas.DataFrame.join
+
+    # compile whatsnew and API section (to resolve links in the whatsnew)
+    python make.py clean
+    python make.py --whatsnew
+
+For comparison, a full documentation build may take 15 minutes, but a single
+section may take 15 seconds. Subsequent builds, which only process portions
+you have changed, will be faster.
+
+The build will automatically use the number of cores available on your machine
+to speed up the documentation build. You can override this::
+
+    python make.py html --num-jobs 4
+
+Open the following file in a web browser to see the full documentation you
+just built ``doc/build/html/index.html``.
+
+And you'll have the satisfaction of seeing your new and improved documentation!
+
+.. _contributing.dev_docs:
+
+Building main branch documentation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+When pull requests are merged into the pandas ``main`` branch, the main parts of
+the documentation are also built by Travis-CI. These docs are then hosted `here
+<https://pandas.pydata.org/docs/dev/>`__, see also
+the :any:`Continuous Integration <contributing.ci>` section.
+
+Previewing changes
+------------------
+
+Once, the pull request is submitted, GitHub Actions will automatically build the
+documentation. To view the built site:
+
+#. Wait for the ``CI / Web and docs`` check to complete.
+#. Click ``Details`` next to it.
+#. From the ``Artifacts`` drop-down, click ``docs`` or ``website`` to download
+   the site as a ZIP file.
diff --git a/pandas/doc/source/development/contributing_environment.rst b/pandas/doc/source/development/contributing_environment.rst
new file mode 100644
index 0000000000000..98bd4b00d016b
--- /dev/null
+++ b/pandas/doc/source/development/contributing_environment.rst
@@ -0,0 +1,313 @@
+.. _contributing_environment:
+
+{{ header }}
+
+==================================
+Creating a development environment
+==================================
+
+To test out code changes, you'll need to build pandas from source, which
+requires a C/C++ compiler and Python environment. If you're making documentation
+changes, you can skip to :ref:`contributing to the documentation <contributing_documentation>` but if you skip
+creating the development environment you won't be able to build the documentation
+locally before pushing your changes. It's recommended to also install the :ref:`pre-commit hooks <contributing.pre-commit>`.
+
+.. toctree::
+    :maxdepth: 2
+    :hidden:
+
+    contributing_gitpod.rst
+
+Step 1: install a C compiler
+----------------------------
+
+How to do this will depend on your platform. If you choose to use ``Docker`` or ``GitPod``
+in the next step, then you can skip this step.
+
+**Windows**
+
+You will need `Build Tools for Visual Studio 2022
+<https://visualstudio.microsoft.com/downloads/#build-tools-for-visual-studio-2022>`_.
+
+.. note::
+        You DO NOT need to install Visual Studio 2022.
+        You only need "Build Tools for Visual Studio 2022" found by
+        scrolling down to "All downloads" -> "Tools for Visual Studio".
+        In the installer, select the "Desktop development with C++" Workloads.
+
+        If you encounter an error indicating ``cl.exe`` is not found when building with Meson,
+        reopen the installer and also select the optional component
+        **MSVC v142 - VS 2019 C++ x64/x86 build tools** in the right pane for installation.
+
+Alternatively, you can install the necessary components on the commandline using
+`vs_BuildTools.exe <https://learn.microsoft.com/en-us/visualstudio/install/use-command-line-parameters-to-install-visual-studio?source=recommendations&view=vs-2022>`_
+
+Alternatively, you could use the `WSL <https://learn.microsoft.com/en-us/windows/wsl/install>`_
+and consult the ``Linux`` instructions below.
+
+**macOS**
+
+To use the :ref:`conda <contributing.conda>`-based compilers, you will need to install the
+Developer Tools using ``xcode-select --install``.
+
+If you prefer to use a different compiler, general information can be found here:
+https://devguide.python.org/setup/#macos
+
+**Linux**
+
+For Linux-based :ref:`conda <contributing.conda>` installations, you won't have to install any
+additional components outside of the conda environment. The instructions
+below are only needed if your setup isn't based on conda environments.
+
+Some Linux distributions will come with a pre-installed C compiler. To find out
+which compilers (and versions) are installed on your system::
+
+    # for Debian/Ubuntu:
+    dpkg --list | grep compiler
+    # for Red Hat/RHEL/CentOS/Fedora:
+    yum list installed | grep -i --color compiler
+
+`GCC (GNU Compiler Collection) <https://gcc.gnu.org/>`_, is a widely used
+compiler, which supports C and a number of other languages. If GCC is listed
+as an installed compiler nothing more is required.
+
+If no C compiler is installed, or you wish to upgrade, or you're using a different
+Linux distribution, consult your favorite search engine for compiler installation/update
+instructions.
+
+Let us know if you have any difficulties by opening an issue or reaching out on our contributor
+community :ref:`Slack <community.slack>`.
+
+Step 2: create an isolated environment
+----------------------------------------
+
+Before we begin, please:
+
+* Make sure that you have :any:`cloned the repository <contributing.forking>`
+* ``cd`` to the pandas source directory you just created with the clone command
+
+.. _contributing.conda:
+
+Option 1: using conda (recommended)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+* Install miniforge to get `conda <https://github.com/conda-forge/miniforge?tab=readme-ov-file#download>`_
+* Create and activate the ``pandas-dev`` conda environment using the following commands:
+
+.. code-block:: bash
+
+   conda env create --file environment.yml
+   conda activate pandas-dev
+
+.. _contributing.pip:
+
+Option 2: using pip
+~~~~~~~~~~~~~~~~~~~
+
+You'll need to have at least the :ref:`minimum Python version <install.version>` that pandas supports.
+You also need to have ``setuptools`` 51.0.0 or later to build pandas.
+
+**Unix**/**macOS with virtualenv**
+
+.. code-block:: bash
+
+   # Create a virtual environment
+   # Use an ENV_DIR of your choice. We'll use ~/virtualenvs/pandas-dev
+   # Any parent directories should already exist
+   python3 -m venv ~/virtualenvs/pandas-dev
+
+   # Activate the virtualenv
+   . ~/virtualenvs/pandas-dev/bin/activate
+
+   # Install the build dependencies
+   python -m pip install -r requirements-dev.txt
+
+**Unix**/**macOS with pyenv**
+
+Consult the docs for setting up pyenv `here <https://github.com/pyenv/pyenv>`__.
+
+.. code-block:: bash
+
+   # Create a virtual environment
+   # Use an ENV_DIR of your choice. We'll use ~/Users/<yourname>/.pyenv/versions/pandas-dev
+   pyenv virtualenv <version> <name-to-give-it>
+
+   # For instance:
+   pyenv virtualenv 3.10 pandas-dev
+
+   # Activate the virtualenv
+   pyenv activate pandas-dev
+
+   # Now install the build dependencies in the cloned pandas repo
+   python -m pip install -r requirements-dev.txt
+
+**Windows**
+
+Below is a brief overview on how to set-up a virtual environment with Powershell
+under Windows. For details please refer to the
+`official virtualenv user guide <https://virtualenv.pypa.io/en/latest/user_guide.html#activators>`__.
+
+Use an ENV_DIR of your choice. We'll use ``~\\virtualenvs\\pandas-dev`` where
+``~`` is the folder pointed to by either ``$env:USERPROFILE`` (Powershell) or
+``%USERPROFILE%`` (cmd.exe) environment variable. Any parent directories
+should already exist.
+
+.. code-block:: powershell
+
+   # Create a virtual environment
+   python -m venv $env:USERPROFILE\virtualenvs\pandas-dev
+
+   # Activate the virtualenv. Use activate.bat for cmd.exe
+   ~\virtualenvs\pandas-dev\Scripts\Activate.ps1
+
+   # Install the build dependencies
+   python -m pip install -r requirements-dev.txt
+
+Option 3: using Docker
+~~~~~~~~~~~~~~~~~~~~~~
+
+pandas provides a ``DockerFile`` in the root directory to build a Docker image
+with a full pandas development environment.
+
+**Docker Commands**
+
+Build the Docker image::
+
+    # Build the image
+    docker build -t pandas-dev .
+
+Run Container::
+
+    # Run a container and bind your local repo to the container
+    # This command assumes you are running from your local repo
+    # but if not alter ${PWD} to match your local repo path
+    docker run -it --rm -v ${PWD}:/home/pandas pandas-dev
+
+*Even easier, you can integrate Docker with the following IDEs:*
+
+**Visual Studio Code**
+
+You can use the DockerFile to launch a remote session with Visual Studio Code,
+a popular free IDE, using the ``.devcontainer.json`` file.
+See https://code.visualstudio.com/docs/remote/containers for details.
+
+**PyCharm (Professional)**
+
+Enable Docker support and use the Services tool window to build and manage images as well as
+run and interact with containers.
+See https://www.jetbrains.com/help/pycharm/docker.html for details.
+
+Option 4: using Gitpod
+~~~~~~~~~~~~~~~~~~~~~~
+
+Gitpod is an open-source platform that automatically creates the correct development
+environment right in your browser, reducing the need to install local development
+environments and deal with incompatible dependencies.
+
+If you are a Windows user, unfamiliar with using the command line or building pandas
+for the first time, it is often faster to build with Gitpod. Here are the in-depth instructions
+for :ref:`building pandas with GitPod <contributing-gitpod>`.
+
+Step 3: build and install pandas
+--------------------------------
+
+There are currently two supported ways of building pandas, pip/meson and setuptools(setup.py).
+Historically, pandas has only supported using setuptools to build pandas. However, this method
+requires a lot of convoluted code in setup.py and also has many issues in compiling pandas in parallel
+due to limitations in setuptools.
+
+The newer build system, invokes the meson backend through pip (via a `PEP 517 <https://peps.python.org/pep-0517/>`_ build).
+It automatically uses all available cores on your CPU, and also avoids the need for manual rebuilds by
+rebuilding automatically whenever pandas is imported (with an editable install).
+
+For these reasons, you should compile pandas with meson.
+Because the meson build system is newer, you may find bugs/minor issues as it matures. You can report these bugs
+`here <https://github.com/pandas-dev/pandas/issues/49683>`_.
+
+To compile pandas with meson, run::
+
+   # Build and install pandas
+   # By default, this will print verbose output
+   # showing the "rebuild" taking place on import (see section below for explanation)
+   # If you do not want to see this, omit everything after --no-build-isolation
+   python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true
+
+.. note::
+   The version number is pulled from the latest repository tag. Be sure to fetch the latest tags from upstream
+   before building::
+
+      # set the upstream repository, if not done already, and fetch the latest tags
+      git remote add upstream https://github.com/pandas-dev/pandas.git
+      git fetch upstream --tags
+
+**Build options**
+
+It is possible to pass options from the pip frontend to the meson backend if you would like to configure your
+install. Occasionally, you'll want to use this to adjust the build directory, and/or toggle debug/optimization levels.
+
+You can pass a build directory to pandas by appending ``-Cbuilddir="your builddir here"`` to your pip command.
+This option allows you to configure where meson stores your built C extensions, and allows for fast rebuilds.
+
+Sometimes, it might be useful to compile pandas with debugging symbols, when debugging C extensions.
+Appending ``-Csetup-args="-Ddebug=true"`` will do the trick.
+
+With pip, it is possible to chain together multiple config settings (for example specifying both a build directory
+and building with debug symbols would look like
+``-Cbuilddir="your builddir here" -Csetup-args="-Dbuildtype=debug"``.
+
+**Compiling pandas with setup.py**
+
+.. note::
+   This method of compiling pandas will be deprecated and removed very soon, as the meson backend matures.
+
+To compile pandas with setuptools, run::
+
+   python setup.py develop
+
+.. note::
+   If pandas is already installed (via meson), you have to uninstall it first::
+
+        python -m pip uninstall pandas
+
+This is because python setup.py develop will not uninstall the loader script that ``meson-python``
+uses to import the extension from the build folder, which may cause errors such as an
+``FileNotFoundError`` to be raised.
+
+.. note::
+   You will need to repeat this step each time the C extensions change, for example
+   if you modified any file in ``pandas/_libs`` or if you did a fetch and merge from ``upstream/main``.
+
+**Checking the build**
+
+At this point you should be able to import pandas from your locally built version::
+
+   $ python
+   >>> import pandas
+   >>> print(pandas.__version__)  # note: the exact output may differ
+   2.0.0.dev0+880.g2b9e661fbb.dirty
+
+
+At this point you may want to try
+`running the test suite <https://pandas.pydata.org/docs/dev/development/contributing_codebase.html#running-the-test-suite>`_.
+
+**Keeping up to date with the latest build**
+
+When building pandas with meson, importing pandas will automatically trigger a rebuild, even when C/Cython files are modified.
+By default, no output will be produced by this rebuild (the import will just take longer). If you would like to see meson's
+output when importing pandas, you can set the environment variable ``MESONPY_EDITABLE_VERBOSE``. For example, this would be::
+
+   # On Linux/macOS
+   MESONPY_EDITABLE_VERBOSE=1 python
+
+   # Windows
+   set MESONPY_EDITABLE_VERBOSE=1 # Only need to set this once per session
+   python
+
+If you would like to see this verbose output every time, you can set the ``editable-verbose`` config setting to ``true`` like so::
+
+   python -m pip install -ve . -Ceditable-verbose=true
+
+.. tip::
+   If you ever find yourself wondering whether setuptools or meson was used to build your pandas,
+   you can check the value of ``pandas._built_with_meson``, which will be true if meson was used
+   to compile pandas.
diff --git a/pandas/doc/source/development/contributing_gitpod.rst b/pandas/doc/source/development/contributing_gitpod.rst
new file mode 100644
index 0000000000000..b70981b4d307d
--- /dev/null
+++ b/pandas/doc/source/development/contributing_gitpod.rst
@@ -0,0 +1,274 @@
+.. _contributing-gitpod:
+
+Using Gitpod for pandas development
+===================================
+
+This section of the documentation will guide you through:
+
+*  using Gitpod for your pandas development environment
+*  creating a personal fork of the pandas repository on GitHub
+*  a quick tour of pandas and VSCode
+*  working on the pandas documentation in Gitpod
+
+Gitpod
+------
+
+`Gitpod`_  is an open-source platform for automated and ready-to-code
+development environments. It enables developers to describe their dev
+environment as code and start instant and fresh development environments for
+each new task directly from your browser. This reduces the need to install local
+development environments and deal with incompatible dependencies.
+
+
+Gitpod GitHub integration
+-------------------------
+
+To be able to use Gitpod, you will need to have the Gitpod app installed on your
+GitHub account, so if
+you do not have an account yet, you will need to create one first.
+
+To get started just login at `Gitpod`_, and grant the appropriate permissions to GitHub.
+
+We have built a python 3.10 environment and all development dependencies will
+install when the environment starts.
+
+
+Forking the pandas repository
+-----------------------------
+
+The best way to work on pandas as a contributor is by making a fork of the
+repository first.
+
+#. Browse to the `pandas repository on GitHub`_ and `create your own fork`_.
+
+#. Browse to your fork. Your fork will have a URL like
+   https://github.com/noatamir/pandas-dev, except with your GitHub username in place of
+   ``noatamir``.
+
+Starting Gitpod
+---------------
+Once you have authenticated to Gitpod through GitHub, you can install the
+`Gitpod Chromium or Firefox browser extension <https://www.gitpod.io/docs/browser-extension>`_
+which will add a **Gitpod** button next to the **Code** button in the
+repository:
+
+.. image:: ./gitpod-imgs/pandas-github.png
+   :alt: pandas repository with Gitpod button screenshot
+
+#. If you install the extension - you can click the **Gitpod** button to start
+   a new workspace.
+
+#. Alternatively, if you do not want to install the browser extension, you can
+   visit https://gitpod.io/#https://github.com/USERNAME/pandas replacing
+   ``USERNAME`` with your GitHub username.
+
+#. In both cases, this will open a new tab on your web browser and start
+   building your development environment. Please note this can take a few
+   minutes.
+
+#. Once the build is complete, you will be directed to your workspace,
+   including the VSCode editor and all the dependencies you need to work on
+   pandas. The first time you start your workspace, you will notice that there
+   might be some actions running. This will ensure that you have a development
+   version of pandas installed.
+
+#. When your workspace is ready, you can :ref:`test the build<contributing.running_tests>` by
+   entering::
+
+      $ python -m pytest pandas
+
+   Note that this command takes a while to run, so once you've confirmed it's running you may want to cancel it using ctrl-c.
+
+Quick workspace tour
+--------------------
+Gitpod uses VSCode as the editor. If you have not used this editor before, you
+can check the Getting started `VSCode docs`_ to familiarize yourself with it.
+
+Your workspace will look similar to the image below:
+
+.. image:: ./gitpod-imgs/gitpod-workspace.png
+   :alt: Gitpod workspace screenshot
+
+We have marked some important sections in the editor:
+
+#. Your current Python interpreter - by default, this is ``pandas-dev`` and
+   should be displayed in the status bar and on your terminal. You do not need
+   to activate the conda environment as this will always be activated for you.
+#. Your current branch is always displayed in the status bar. You can also use
+   this button to change or create branches.
+#. GitHub Pull Requests extension - you can use this to work with Pull Requests
+   from your workspace.
+#. Marketplace extensions - we have added some essential extensions to the pandas
+   Gitpod. Still, you can also install other extensions or syntax highlighting
+   themes for your user, and these will be preserved for you.
+#. Your workspace directory - by default, it is ``/workspace/pandas-dev``. **Do not
+   change this** as this is the only directory preserved in Gitpod.
+
+We have also pre-installed a few tools and VSCode extensions to help with the
+development experience:
+
+*  `VSCode rst extension <https://marketplace.visualstudio.com/items?itemName=lextudio.restructuredtext>`_
+*  `Markdown All in One <https://marketplace.visualstudio.com/items?itemName=yzhang.markdown-all-in-one>`_
+*  `VSCode GitLens extension <https://marketplace.visualstudio.com/items?itemName=eamodio.gitlens>`_
+*  `VSCode Git Graph extension <https://marketplace.visualstudio.com/items?itemName=mhutchie.git-graph>`_
+
+Development workflow with Gitpod
+--------------------------------
+The  :ref:`contributing` section of this documentation contains
+information regarding the pandas development workflow. Make sure to check this
+before working on your contributions.
+
+When using Gitpod, git is pre configured for you:
+
+#. You do not need to configure your git username, and email as this should be
+   done for you as you authenticated through GitHub. Unless you are using GitHub
+   feature to keep email address private. You can check the git
+   configuration with the command ``git config --list`` in your terminal. Use
+   ``git config --global user.email “your-secret-email@users.noreply.github.com”``
+   to set your email address to the one you use to make commits with your github
+   profile.
+#. As you started your workspace from your own pandas fork, you will by default
+   have both ``upstream`` and ``origin`` added as remotes. You can verify this by
+   typing ``git remote`` on your terminal or by clicking on the **branch name**
+   on the status bar (see image below).
+
+   .. image:: ./gitpod-imgs/pandas-gitpod-branches.png
+      :alt: Gitpod workspace branches plugin screenshot
+
+Rendering the pandas documentation
+----------------------------------
+You can find the detailed documentation on how rendering the documentation with
+Sphinx works in the :ref:`contributing.howto-build-docs` section. To build the full
+docs you need to run the following command in the ``/doc`` directory::
+
+    $ cd doc
+    $ python make.py html
+
+Alternatively you can build a single page with::
+
+    python make.py --single development/contributing_gitpod.rst
+
+You have two main options to render the documentation in Gitpod.
+
+Option 1: using Liveserve
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+#. View the documentation in ``pandas/doc/build/html``.
+#. To see the rendered version of a page, you can right-click on the ``.html``
+   file and click on **Open with Live Serve**. Alternatively, you can open the
+   file in the editor and click on the **Go live** button on the status bar.
+
+    .. image:: ./gitpod-imgs/vscode-statusbar.png
+        :alt: Gitpod workspace VSCode start live serve screenshot
+
+#. A simple browser will open to the right-hand side of the editor. We recommend
+   closing it and click on the **Open in browser** button in the pop-up.
+#. To stop the server click on the **Port: 5500** button on the status bar.
+
+Option 2: using the rst extension
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A quick and easy way to see live changes in a ``.rst`` file as you work on it
+uses the rst extension with docutils.
+
+.. note:: This will generate a simple live preview of the document without the
+    ``html`` theme, and some backlinks might not be added correctly. But it is an
+    easy and lightweight way to get instant feedback on your work, without
+    building the html files.
+
+#. Open any of the source documentation files located in ``doc/source`` in the
+   editor.
+#. Open VSCode Command Palette with :kbd:`Cmd-Shift-P` in Mac or
+   :kbd:`Ctrl-Shift-P` in Linux and Windows. Start typing "restructured"
+   and choose either "Open preview" or "Open preview to the Side".
+
+    .. image:: ./gitpod-imgs/vscode-rst.png
+        :alt: Gitpod workspace VSCode open rst screenshot
+
+#. As you work on the document, you will see a live rendering of it on the editor.
+
+    .. image:: ./gitpod-imgs/rst-rendering.png
+        :alt: Gitpod workspace VSCode rst rendering screenshot
+
+If you want to see the final output with the ``html`` theme you will need to
+rebuild the docs with ``make html`` and use Live Serve as described in option 1.
+
+FAQ's and troubleshooting
+-------------------------
+
+How long is my Gitpod workspace kept for?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Your stopped workspace will be kept for 14 days and deleted afterwards if you do
+not use them.
+
+Can I come back to a previous workspace?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Yes, let's say you stepped away for a while and you want to carry on working on
+your pandas contributions. You need to visit https://gitpod.io/workspaces and
+click on the workspace you want to spin up again. All your changes will be there
+as you last left them.
+
+Can I install additional VSCode extensions?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Absolutely! Any extensions you installed will be installed in your own workspace
+and preserved.
+
+I registered on Gitpod but I still cannot see a ``Gitpod`` button in my repositories.
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Head to https://gitpod.io/integrations and make sure you are logged in.
+Hover over GitHub and click on the three buttons that appear on the right.
+Click on edit permissions and make sure you have ``user:email``,
+``read:user``, and ``public_repo`` checked. Click on **Update Permissions**
+and confirm the changes in the GitHub application page.
+
+.. image:: ./gitpod-imgs/gitpod-edit-permissions-gh.png
+   :alt: Gitpod integrations - edit GH permissions screenshot
+
+How long does my workspace stay active if I'm not using it?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If you keep your workspace open in a browser tab but don't interact with it,
+it will shut down after 30 minutes. If you close the browser tab, it will
+shut down after 3 minutes.
+
+My terminal is blank - there is no cursor and it's completely unresponsive
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Unfortunately this is a known-issue on Gitpod's side. You can sort this
+issue in two ways:
+
+#. Create a new Gitpod workspace altogether.
+#. Head to your `Gitpod dashboard <https://gitpod.io/workspaces>`_ and locate
+   the running workspace. Hover on it and click on the **three dots menu**
+   and then click on **Stop**. When the workspace is completely stopped you
+   can click on its name to restart it again.
+
+.. image:: ./gitpod-imgs/gitpod-dashboard-stop.png
+   :alt: Gitpod dashboard and workspace menu screenshot
+
+I authenticated through GitHub but I still cannot commit to the repository through Gitpod.
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Head to https://gitpod.io/integrations and make sure you are logged in.
+Hover over GitHub and click on the three buttons that appear on the right.
+Click on edit permissions and make sure you have ``public_repo`` checked.
+Click on **Update Permissions** and confirm the changes in the
+GitHub application page.
+
+.. image:: ./gitpod-imgs/gitpod-edit-permissions-repo.png
+   :alt: Gitpod integrations - edit GH repository permissions screenshot
+
+Acknowledgments
+---------------
+
+This page is lightly adapted from the `NumPy`_ project .
+
+.. _Gitpod: https://www.gitpod.io/
+.. _pandas repository on GitHub: https://github.com/pandas-dev/pandas
+.. _create your own fork: https://help.github.com/en/articles/fork-a-repo
+.. _VSCode docs: https://code.visualstudio.com/docs/getstarted/tips-and-tricks
+.. _NumPy: https://www.numpy.org/
diff --git a/pandas/doc/source/development/copy_on_write.rst b/pandas/doc/source/development/copy_on_write.rst
new file mode 100644
index 0000000000000..9a2309b8a77a3
--- /dev/null
+++ b/pandas/doc/source/development/copy_on_write.rst
@@ -0,0 +1,42 @@
+.. _copy_on_write_dev:
+
+{{ header }}
+
+*************
+Copy on write
+*************
+
+Copy on Write is a mechanism to simplify the indexing API and improve
+performance through avoiding copies if possible.
+CoW means that any DataFrame or Series derived from another in any way always
+behaves as a copy. An explanation on how to use Copy on Write efficiently can be
+found :ref:`here <copy_on_write>`.
+
+Reference tracking
+------------------
+
+To be able to determine if we have to make a copy when writing into a DataFrame,
+we have to be aware if the values are shared with another DataFrame. pandas
+keeps track of all ``Blocks`` that share values with another block internally to
+be able to tell when a copy needs to be triggered. The reference tracking
+mechanism is implemented on the Block level.
+
+We use a custom reference tracker object, ``BlockValuesRefs``, that keeps
+track of every block, whose values share memory with each other. The reference
+is held through a weak-reference. Every pair of blocks that share some memory should
+point to the same ``BlockValuesRefs`` object. If one block goes out of
+scope, the reference to this block dies. As a consequence, the reference tracker
+object always knows how many blocks are alive and share memory.
+
+Whenever a :class:`DataFrame` or :class:`Series` object is sharing data with another
+object, it is required that each of those objects have its own BlockManager and Block
+objects. Thus, in other words, one Block instance (that is held by a DataFrame, not
+necessarily for intermediate objects) should always be uniquely used for only
+a single DataFrame/Series object. For example, when you want to use the same
+Block for another object, you can create a shallow copy of the Block instance
+with ``block.copy(deep=False)`` (which will create a new Block instance with
+the same underlying values and which will correctly set up the references).
+
+We can ask the reference tracking object if there is another block alive that shares
+data with us before writing into the values. We can trigger a copy before
+writing if there is in fact another block alive.
diff --git a/pandas/doc/source/development/debugging_extensions.rst b/pandas/doc/source/development/debugging_extensions.rst
new file mode 100644
index 0000000000000..0ea1c112cb55b
--- /dev/null
+++ b/pandas/doc/source/development/debugging_extensions.rst
@@ -0,0 +1,63 @@
+.. _debugging_c_extensions:
+
+{{ header }}
+
+======================
+Debugging C extensions
+======================
+
+pandas uses Cython and C/C++ `extension modules <https://docs.python.org/3/extending/extending.html>`_ to optimize performance. Unfortunately, the standard Python debugger does not allow you to step into these extensions. Cython extensions can be debugged with the `Cython debugger <https://docs.cython.org/en/latest/src/userguide/debugging.html>`_ and C/C++ extensions can be debugged using the tools shipped with your platform's compiler.
+
+For Python developers with limited or no C/C++ experience this can seem a daunting task. Core developer Will Ayd has written a 3 part blog series to help guide you from the standard Python debugger into these other tools:
+
+  1. `Fundamental Python Debugging Part 1 - Python <https://willayd.com/fundamental-python-debugging-part-1-python.html>`_
+  2. `Fundamental Python Debugging Part 2 - Python Extensions <https://willayd.com/fundamental-python-debugging-part-2-python-extensions.html>`_
+  3. `Fundamental Python Debugging Part 3 - Cython Extensions <https://willayd.com/fundamental-python-debugging-part-3-cython-extensions.html>`_
+
+Debugging locally
+-----------------
+
+By default building pandas from source will generate a release build. To generate a development build you can type::
+
+    pip install -ve . --no-build-isolation -Cbuilddir="debug" -Csetup-args="-Dbuildtype=debug"
+
+.. note::
+
+   conda environments update CFLAGS/CPPFLAGS with flags that are geared towards generating releases. If using conda, you may need to set ``CFLAGS="$CFLAGS -O0"`` and ``CPPFLAGS="$CPPFLAGS -O0"`` to ensure optimizations are turned off for debugging
+
+By specifying ``builddir="debug"`` all of the targets will be built and placed in the debug directory relative to the project root. This helps to keep your debug and release artifacts separate; you are of course able to choose a different directory name or omit altogether if you do not care to separate build types.
+
+Using Docker
+------------
+
+To simplify the debugging process, pandas has created a Docker image with a debug build of Python and the gdb/Cython debuggers pre-installed. You may either ``docker pull pandas/pandas-debug`` to get access to this image or build it from the ``tooling/debug`` folder locally.
+
+You can then mount your pandas repository into this image via:
+
+.. code-block:: sh
+
+   docker run --rm -it -w /data -v ${PWD}:/data pandas/pandas-debug
+
+Inside the image, you can use meson to build/install pandas and place the build artifacts into a ``debug`` folder using a command as follows:
+
+.. code-block:: sh
+
+    python -m pip install -ve . --no-build-isolation -Cbuilddir="debug" -Csetup-args="-Dbuildtype=debug"
+
+If planning to use cygdb, the files required by that application are placed within the build folder. So you have to first ``cd`` to the build folder, then start that application.
+
+.. code-block:: sh
+
+   cd debug
+   cygdb
+
+Within the debugger you can use `cygdb commands <https://docs.cython.org/en/latest/src/userguide/debugging.html#using-the-debugger>`_ to navigate cython extensions.
+
+Editor support
+--------------
+
+The meson build system generates a `compilation database <https://clang.llvm.org/docs/JSONCompilationDatabase.html>`_ automatically and places it in the build directory. Many language servers and IDEs can use this information to provide code-completion, go-to-definition and error checking support as you type.
+
+How each language server / IDE chooses to look for the compilation database may vary. When in doubt you may want to create a symlink at the root of the project that points to the compilation database in your build directory. Assuming you used *debug* as your directory name, you can run::
+
+    ln -s debug/compile_commands.json .
diff --git a/pandas/doc/source/development/developer.rst b/pandas/doc/source/development/developer.rst
new file mode 100644
index 0000000000000..c5c4b7c449ce7
--- /dev/null
+++ b/pandas/doc/source/development/developer.rst
@@ -0,0 +1,187 @@
+.. _developer:
+
+{{ header }}
+
+.. currentmodule:: pandas
+
+*********
+Developer
+*********
+
+This section will focus on downstream applications of pandas.
+
+.. _apache.parquet:
+
+Storing pandas DataFrame objects in Apache Parquet format
+---------------------------------------------------------
+
+The `Apache Parquet <https://github.com/apache/parquet-format>`__ format
+provides key-value metadata at the file and column level, stored in the footer
+of the Parquet file:
+
+.. code-block:: shell
+
+  5: optional list<KeyValue> key_value_metadata
+
+where ``KeyValue`` is
+
+.. code-block:: shell
+
+   struct KeyValue {
+     1: required string key
+     2: optional string value
+   }
+
+So that a ``pandas.DataFrame`` can be faithfully reconstructed, we store a
+``pandas`` metadata key in the ``FileMetaData`` with the value stored as :
+
+.. code-block:: text
+
+   {'index_columns': [<descr0>, <descr1>, ...],
+    'column_indexes': [<ci0>, <ci1>, ..., <ciN>],
+    'columns': [<c0>, <c1>, ...],
+    'pandas_version': $VERSION,
+    'creator': {
+      'library': $LIBRARY,
+      'version': $LIBRARY_VERSION
+    }}
+
+The "descriptor" values ``<descr0>`` in the ``'index_columns'`` field are
+strings (referring to a column) or dictionaries with values as described below.
+
+The ``<c0>``/``<ci0>`` and so forth are dictionaries containing the metadata
+for each column, *including the index columns*. This has JSON form:
+
+.. code-block:: text
+
+   {'name': column_name,
+    'field_name': parquet_column_name,
+    'pandas_type': pandas_type,
+    'numpy_type': numpy_type,
+    'metadata': metadata}
+
+See below for the detailed specification for these.
+
+Index metadata descriptors
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``RangeIndex`` can be stored as metadata only, not requiring serialization. The
+descriptor format for these as is follows:
+
+.. code-block:: python
+
+   index = pd.RangeIndex(0, 10, 2)
+   {
+       "kind": "range",
+       "name": index.name,
+       "start": index.start,
+       "stop": index.stop,
+       "step": index.step,
+   }
+
+Other index types must be serialized as data columns along with the other
+DataFrame columns. The metadata for these is a string indicating the name of
+the field in the data columns, for example ``'__index_level_0__'``.
+
+If an index has a non-None ``name`` attribute, and there is no other column
+with a name matching that value, then the ``index.name`` value can be used as
+the descriptor. Otherwise (for unnamed indexes and ones with names colliding
+with other column names) a disambiguating name with pattern matching
+``__index_level_\d+__`` should be used. In cases of named indexes as data
+columns, ``name`` attribute is always stored in the column descriptors as
+above.
+
+Column metadata
+~~~~~~~~~~~~~~~
+
+``pandas_type`` is the logical type of the column, and is one of:
+
+* Boolean: ``'bool'``
+* Integers: ``'int8', 'int16', 'int32', 'int64', 'uint8', 'uint16', 'uint32', 'uint64'``
+* Floats: ``'float16', 'float32', 'float64'``
+* Date and Time Types: ``'datetime', 'datetimetz', 'timedelta'``
+* String: ``'unicode', 'bytes'``
+* Categorical: ``'categorical'``
+* Other Python objects: ``'object'``
+
+The ``numpy_type`` is the physical storage type of the column, which is the
+result of ``str(dtype)`` for the underlying NumPy array that holds the data. So
+for ``datetimetz`` this is ``datetime64[ns]`` and for categorical, it may be
+any of the supported integer categorical types.
+
+The ``metadata`` field is ``None`` except for:
+
+* ``datetimetz``: ``{'timezone': zone, 'unit': 'ns'}``, e.g. ``{'timezone',
+  'America/New_York', 'unit': 'ns'}``. The ``'unit'`` is optional, and if
+  omitted it is assumed to be nanoseconds.
+* ``categorical``: ``{'num_categories': K, 'ordered': is_ordered, 'type': $TYPE}``
+
+    * Here ``'type'`` is optional, and can be a nested pandas type specification
+      here (but not categorical)
+
+* ``unicode``: ``{'encoding': encoding}``
+
+    * The encoding is optional, and if not present is UTF-8
+
+* ``object``: ``{'encoding': encoding}``. Objects can be serialized and stored
+  in ``BYTE_ARRAY`` Parquet columns. The encoding can be one of:
+
+    * ``'pickle'``
+    * ``'bson'``
+    * ``'json'``
+
+* ``timedelta``: ``{'unit': 'ns'}``. The ``'unit'`` is optional, and if omitted
+  it is assumed to be nanoseconds. This metadata is optional altogether
+
+For types other than these, the ``'metadata'`` key can be
+omitted. Implementations can assume ``None`` if the key is not present.
+
+As an example of fully-formed metadata:
+
+.. code-block:: text
+
+   {'index_columns': ['__index_level_0__'],
+    'column_indexes': [
+        {'name': None,
+         'field_name': 'None',
+         'pandas_type': 'unicode',
+         'numpy_type': 'object',
+         'metadata': {'encoding': 'UTF-8'}}
+    ],
+    'columns': [
+        {'name': 'c0',
+         'field_name': 'c0',
+         'pandas_type': 'int8',
+         'numpy_type': 'int8',
+         'metadata': None},
+        {'name': 'c1',
+         'field_name': 'c1',
+         'pandas_type': 'bytes',
+         'numpy_type': 'object',
+         'metadata': None},
+        {'name': 'c2',
+         'field_name': 'c2',
+         'pandas_type': 'categorical',
+         'numpy_type': 'int16',
+         'metadata': {'num_categories': 1000, 'ordered': False}},
+        {'name': 'c3',
+         'field_name': 'c3',
+         'pandas_type': 'datetimetz',
+         'numpy_type': 'datetime64[ns]',
+         'metadata': {'timezone': 'America/Los_Angeles'}},
+        {'name': 'c4',
+         'field_name': 'c4',
+         'pandas_type': 'object',
+         'numpy_type': 'object',
+         'metadata': {'encoding': 'pickle'}},
+        {'name': None,
+         'field_name': '__index_level_0__',
+         'pandas_type': 'int64',
+         'numpy_type': 'int64',
+         'metadata': None}
+    ],
+    'pandas_version': '1.4.0',
+    'creator': {
+      'library': 'pyarrow',
+      'version': '0.13.0'
+    }}
diff --git a/pandas/doc/source/development/extending.rst b/pandas/doc/source/development/extending.rst
new file mode 100644
index 0000000000000..e67829b8805eb
--- /dev/null
+++ b/pandas/doc/source/development/extending.rst
@@ -0,0 +1,536 @@
+.. _extending:
+
+{{ header }}
+
+****************
+Extending pandas
+****************
+
+While pandas provides a rich set of methods, containers, and data types, your
+needs may not be fully satisfied. pandas offers a few options for extending
+pandas.
+
+.. _extending.register-accessors:
+
+Registering custom accessors
+----------------------------
+
+Libraries can use the decorators
+:func:`pandas.api.extensions.register_dataframe_accessor`,
+:func:`pandas.api.extensions.register_series_accessor`, and
+:func:`pandas.api.extensions.register_index_accessor`, to add additional
+"namespaces" to pandas objects. All of these follow a similar convention: you
+decorate a class, providing the name of attribute to add. The class's
+``__init__`` method gets the object being decorated. For example:
+
+.. code-block:: python
+
+   @pd.api.extensions.register_dataframe_accessor("geo")
+   class GeoAccessor:
+       def __init__(self, pandas_obj):
+           self._validate(pandas_obj)
+           self._obj = pandas_obj
+
+       @staticmethod
+       def _validate(obj):
+           # verify there is a column latitude and a column longitude
+           if "latitude" not in obj.columns or "longitude" not in obj.columns:
+               raise AttributeError("Must have 'latitude' and 'longitude'.")
+
+       @property
+       def center(self):
+           # return the geographic center point of this DataFrame
+           lat = self._obj.latitude
+           lon = self._obj.longitude
+           return (float(lon.mean()), float(lat.mean()))
+
+       def plot(self):
+           # plot this array's data on a map, e.g., using Cartopy
+           pass
+
+Now users can access your methods using the ``geo`` namespace:
+
+      >>> ds = pd.DataFrame(
+      ...     {"longitude": np.linspace(0, 10), "latitude": np.linspace(0, 20)}
+      ... )
+      >>> ds.geo.center
+      (5.0, 10.0)
+      >>> ds.geo.plot()
+      # plots data on a map
+
+This can be a convenient way to extend pandas objects without subclassing them.
+If you write a custom accessor, make a pull request adding it to our
+`ecosystem <https://pandas.pydata.org/community/ecosystem.html>`_ page.
+
+We highly recommend validating the data in your accessor's ``__init__``.
+In our ``GeoAccessor``, we validate that the data contains the expected columns,
+raising an ``AttributeError`` when the validation fails.
+For a ``Series`` accessor, you should validate the ``dtype`` if the accessor
+applies only to certain dtypes.
+
+
+.. _extending.extension-types:
+
+Extension types
+---------------
+
+.. note::
+
+   The :class:`pandas.api.extensions.ExtensionDtype` and :class:`pandas.api.extensions.ExtensionArray` APIs were
+   experimental prior to pandas 1.5. Starting with version 1.5, future changes will follow
+   the :ref:`pandas deprecation policy <policies.version>`.
+
+pandas defines an interface for implementing data types and arrays that *extend*
+NumPy's type system. pandas itself uses the extension system for some types
+that aren't built into NumPy (categorical, period, interval, datetime with
+timezone).
+
+Libraries can define a custom array and data type. When pandas encounters these
+objects, they will be handled properly (i.e. not converted to an ndarray of
+objects). Many methods like :func:`pandas.isna` will dispatch to the extension
+type's implementation.
+
+If you're building a library that implements the interface, please publicize it
+on `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_.
+
+The interface consists of two classes.
+
+:class:`~pandas.api.extensions.ExtensionDtype`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A :class:`pandas.api.extensions.ExtensionDtype` is similar to a ``numpy.dtype`` object. It describes the
+data type. Implementers are responsible for a few unique items like the name.
+
+One particularly important item is the ``type`` property. This should be the
+class that is the scalar type for your data. For example, if you were writing an
+extension array for IP Address data, this might be ``ipaddress.IPv4Address``.
+
+See the `extension dtype source`_ for interface definition.
+
+:class:`pandas.api.extensions.ExtensionDtype` can be registered to pandas to allow creation via a string dtype name.
+This allows one to instantiate ``Series`` and ``.astype()`` with a registered string name, for
+example ``'category'`` is a registered string accessor for the ``CategoricalDtype``.
+
+See the `extension dtype dtypes`_ for more on how to register dtypes.
+
+:class:`~pandas.api.extensions.ExtensionArray`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+This class provides all the array-like functionality. ExtensionArrays are
+limited to 1 dimension. An ExtensionArray is linked to an ExtensionDtype via the
+``dtype`` attribute.
+
+pandas makes no restrictions on how an extension array is created via its
+``__new__`` or ``__init__``, and puts no restrictions on how you store your
+data. We do require that your array be convertible to a NumPy array, even if
+this is relatively expensive (as it is for ``Categorical``).
+
+They may be backed by none, one, or many NumPy arrays. For example,
+:class:`pandas.Categorical` is an extension array backed by two arrays,
+one for codes and one for categories. An array of IPv6 addresses may
+be backed by a NumPy structured array with two fields, one for the
+lower 64 bits and one for the upper 64 bits. Or they may be backed
+by some other storage type, like Python lists.
+
+See the `extension array source`_ for the interface definition. The docstrings
+and comments contain guidance for properly implementing the interface.
+
+.. _extending.extension.operator:
+
+:class:`~pandas.api.extensions.ExtensionArray` operator support
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+By default, there are no operators defined for the class :class:`~pandas.api.extensions.ExtensionArray`.
+There are two approaches for providing operator support for your ExtensionArray:
+
+1. Define each of the operators on your ``ExtensionArray`` subclass.
+2. Use an operator implementation from pandas that depends on operators that are already defined
+   on the underlying elements (scalars) of the ExtensionArray.
+
+.. note::
+
+   Regardless of the approach, you may want to set ``__array_priority__``
+   if you want your implementation to be called when involved in binary operations
+   with NumPy arrays.
+
+For the first approach, you define selected operators, e.g., ``__add__``, ``__le__``, etc. that
+you want your ``ExtensionArray`` subclass to support.
+
+The second approach assumes that the underlying elements (i.e., scalar type) of the ``ExtensionArray``
+have the individual operators already defined.  In other words, if your ``ExtensionArray``
+named ``MyExtensionArray`` is implemented so that each element is an instance
+of the class ``MyExtensionElement``, then if the operators are defined
+for ``MyExtensionElement``, the second approach will automatically
+define the operators for ``MyExtensionArray``.
+
+A mixin class, :class:`~pandas.api.extensions.ExtensionScalarOpsMixin` supports this second
+approach.  If developing an ``ExtensionArray`` subclass, for example ``MyExtensionArray``,
+can simply include ``ExtensionScalarOpsMixin`` as a parent class of ``MyExtensionArray``,
+and then call the methods :meth:`~MyExtensionArray._add_arithmetic_ops` and/or
+:meth:`~MyExtensionArray._add_comparison_ops` to hook the operators into
+your ``MyExtensionArray`` class, as follows:
+
+.. code-block:: python
+
+    from pandas.api.extensions import ExtensionArray, ExtensionScalarOpsMixin
+
+
+    class MyExtensionArray(ExtensionArray, ExtensionScalarOpsMixin):
+        pass
+
+
+    MyExtensionArray._add_arithmetic_ops()
+    MyExtensionArray._add_comparison_ops()
+
+
+.. note::
+
+   Since ``pandas`` automatically calls the underlying operator on each
+   element one-by-one, this might not be as performant as implementing your own
+   version of the associated operators directly on the ``ExtensionArray``.
+
+For arithmetic operations, this implementation will try to reconstruct a new
+``ExtensionArray`` with the result of the element-wise operation. Whether
+or not that succeeds depends on whether the operation returns a result
+that's valid for the ``ExtensionArray``. If an ``ExtensionArray`` cannot
+be reconstructed, an ndarray containing the scalars returned instead.
+
+For ease of implementation and consistency with operations between pandas
+and NumPy ndarrays, we recommend *not* handling Series and Indexes in your binary ops.
+Instead, you should detect these cases and return ``NotImplemented``.
+When pandas encounters an operation like ``op(Series, ExtensionArray)``, pandas
+will
+
+1. unbox the array from the ``Series`` (``Series.array``)
+2. call ``result = op(values, ExtensionArray)``
+3. re-box the result in a ``Series``
+
+.. _extending.extension.ufunc:
+
+NumPy universal functions
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Series` implements ``__array_ufunc__``. As part of the implementation,
+pandas unboxes the ``ExtensionArray`` from the :class:`Series`, applies the ufunc,
+and re-boxes it if necessary.
+
+If applicable, we highly recommend that you implement ``__array_ufunc__`` in your
+extension array to avoid coercion to an ndarray. See
+`the NumPy documentation <https://numpy.org/doc/stable/reference/generated/numpy.lib.mixins.NDArrayOperatorsMixin.html>`__
+for an example.
+
+As part of your implementation, we require that you defer to pandas when a pandas
+container (:class:`Series`, :class:`DataFrame`, :class:`Index`) is detected in ``inputs``.
+If any of those is present, you should return ``NotImplemented``. pandas will take care of
+unboxing the array from the container and re-calling the ufunc with the unwrapped input.
+
+.. _extending.extension.testing:
+
+Testing extension arrays
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+We provide a test suite for ensuring that your extension arrays satisfy the expected
+behavior. To use the test suite, you must provide several pytest fixtures and inherit
+from the base test class. The required fixtures are found in
+https://github.com/pandas-dev/pandas/blob/main/pandas/tests/extension/conftest.py.
+
+To use a test, subclass it:
+
+.. code-block:: python
+
+   from pandas.tests.extension import base
+
+
+   class TestConstructors(base.BaseConstructorsTests):
+       pass
+
+
+See https://github.com/pandas-dev/pandas/blob/main/pandas/tests/extension/base/__init__.py
+for a list of all the tests available.
+
+.. _extending.extension.arrow:
+
+Compatibility with Apache Arrow
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+An ``ExtensionArray`` can support conversion to / from ``pyarrow`` arrays
+(and thus support for example serialization to the Parquet file format)
+by implementing two methods: ``ExtensionArray.__arrow_array__`` and
+``ExtensionDtype.__from_arrow__``.
+
+The ``ExtensionArray.__arrow_array__`` ensures that ``pyarrow`` knowns how
+to convert the specific extension array into a ``pyarrow.Array`` (also when
+included as a column in a pandas DataFrame):
+
+.. code-block:: python
+
+    class MyExtensionArray(ExtensionArray):
+        ...
+
+        def __arrow_array__(self, type=None):
+            # convert the underlying array values to a pyarrow Array
+            import pyarrow
+
+            return pyarrow.array(..., type=type)
+
+The ``ExtensionDtype.__from_arrow__`` method then controls the conversion
+back from pyarrow to a pandas ExtensionArray. This method receives a pyarrow
+``Array`` or ``ChunkedArray`` as only argument and is expected to return the
+appropriate pandas ``ExtensionArray`` for this dtype and the passed values:
+
+.. code-block:: none
+
+    class ExtensionDtype:
+        ...
+
+        def __from_arrow__(self, array: pyarrow.Array/ChunkedArray) -> ExtensionArray:
+            ...
+
+See more in the `Arrow documentation <https://arrow.apache.org/docs/python/extending_types.html>`__.
+
+Those methods have been implemented for the nullable integer and string extension
+dtypes included in pandas, and ensure roundtrip to pyarrow and the Parquet file format.
+
+.. _extension dtype dtypes: https://github.com/pandas-dev/pandas/blob/main/pandas/core/dtypes/dtypes.py
+.. _extension dtype source: https://github.com/pandas-dev/pandas/blob/main/pandas/core/dtypes/base.py
+.. _extension array source: https://github.com/pandas-dev/pandas/blob/main/pandas/core/arrays/base.py
+
+.. _extending.subclassing-pandas:
+
+Subclassing pandas data structures
+----------------------------------
+
+.. warning:: There are some easier alternatives before considering subclassing ``pandas`` data structures.
+
+  1. Extensible method chains with :ref:`pipe <basics.pipe>`
+
+  2. Use *composition*. See `here <https://en.wikipedia.org/wiki/Composition_over_inheritance>`_.
+
+  3. Extending by :ref:`registering an accessor <extending.register-accessors>`
+
+  4. Extending by :ref:`extension type <extending.extension-types>`
+
+This section describes how to subclass ``pandas`` data structures to meet more specific needs. There are two points that need attention:
+
+1. Override constructor properties.
+2. Define original properties
+
+.. note::
+
+   You can find a nice example in `geopandas <https://github.com/geopandas/geopandas>`_ project.
+
+Override constructor properties
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Each data structure has several *constructor properties* for returning a new
+data structure as the result of an operation. By overriding these properties,
+you can retain subclasses through ``pandas`` data manipulations.
+
+There are 3 possible constructor properties to be defined on a subclass:
+
+* ``DataFrame/Series._constructor``: Used when a manipulation result has the same dimension as the original.
+* ``DataFrame._constructor_sliced``: Used when a ``DataFrame`` (sub-)class manipulation result should be a ``Series`` (sub-)class.
+* ``Series._constructor_expanddim``: Used when a ``Series`` (sub-)class manipulation result should be a ``DataFrame`` (sub-)class, e.g. ``Series.to_frame()``.
+
+Below example shows how to define ``SubclassedSeries`` and ``SubclassedDataFrame`` overriding constructor properties.
+
+.. code-block:: python
+
+   class SubclassedSeries(pd.Series):
+       @property
+       def _constructor(self):
+           return SubclassedSeries
+
+       @property
+       def _constructor_expanddim(self):
+           return SubclassedDataFrame
+
+
+   class SubclassedDataFrame(pd.DataFrame):
+       @property
+       def _constructor(self):
+           return SubclassedDataFrame
+
+       @property
+       def _constructor_sliced(self):
+           return SubclassedSeries
+
+.. code-block:: python
+
+   >>> s = SubclassedSeries([1, 2, 3])
+   >>> type(s)
+   <class '__main__.SubclassedSeries'>
+
+   >>> to_framed = s.to_frame()
+   >>> type(to_framed)
+   <class '__main__.SubclassedDataFrame'>
+
+   >>> df = SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+   >>> df
+      A  B  C
+   0  1  4  7
+   1  2  5  8
+   2  3  6  9
+
+   >>> type(df)
+   <class '__main__.SubclassedDataFrame'>
+
+   >>> sliced1 = df[["A", "B"]]
+   >>> sliced1
+      A  B
+   0  1  4
+   1  2  5
+   2  3  6
+
+   >>> type(sliced1)
+   <class '__main__.SubclassedDataFrame'>
+
+   >>> sliced2 = df["A"]
+   >>> sliced2
+   0    1
+   1    2
+   2    3
+   Name: A, dtype: int64
+
+   >>> type(sliced2)
+   <class '__main__.SubclassedSeries'>
+
+Define original properties
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+To let original data structures have additional properties, you should let ``pandas`` know what properties are added. ``pandas`` maps unknown properties to data names overriding ``__getattribute__``. Defining original properties can be done in one of 2 ways:
+
+1. Define ``_internal_names`` and ``_internal_names_set`` for temporary properties which WILL NOT be passed to manipulation results.
+2. Define ``_metadata`` for normal properties which will be passed to manipulation results.
+
+Below is an example to define two original properties, "internal_cache" as a temporary property and "added_property" as a normal property
+
+.. code-block:: python
+
+   class SubclassedDataFrame2(pd.DataFrame):
+
+       # temporary properties
+       _internal_names = pd.DataFrame._internal_names + ["internal_cache"]
+       _internal_names_set = set(_internal_names)
+
+       # normal properties
+       _metadata = ["added_property"]
+
+       @property
+       def _constructor(self):
+           return SubclassedDataFrame2
+
+.. code-block:: python
+
+   >>> df = SubclassedDataFrame2({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+   >>> df
+      A  B  C
+   0  1  4  7
+   1  2  5  8
+   2  3  6  9
+
+   >>> df.internal_cache = "cached"
+   >>> df.added_property = "property"
+
+   >>> df.internal_cache
+   cached
+   >>> df.added_property
+   property
+
+   # properties defined in _internal_names is reset after manipulation
+   >>> df[["A", "B"]].internal_cache
+   AttributeError: 'SubclassedDataFrame2' object has no attribute 'internal_cache'
+
+   # properties defined in _metadata are retained
+   >>> df[["A", "B"]].added_property
+   property
+
+.. _extending.plotting-backends:
+
+Plotting backends
+-----------------
+
+pandas can be extended with third-party plotting backends. The
+main idea is letting users select a plotting backend different than the provided
+one based on Matplotlib. For example:
+
+.. code-block:: python
+
+    >>> pd.set_option("plotting.backend", "backend.module")
+    >>> pd.Series([1, 2, 3]).plot()
+
+This would be more or less equivalent to:
+
+.. code-block:: python
+
+    >>> import backend.module
+    >>> backend.module.plot(pd.Series([1, 2, 3]))
+
+The backend module can then use other visualization tools (Bokeh, Altair,...)
+to generate the plots.
+
+Libraries implementing the plotting backend should use `entry points <https://setuptools.pypa.io/en/latest/userguide/entry_point.html>`__
+to make their backend discoverable to pandas. The key is ``"pandas_plotting_backends"``. For example, pandas
+registers the default "matplotlib" backend as follows.
+
+.. code-block:: python
+
+   # in setup.py
+   setup(  # noqa: F821
+       ...,
+       entry_points={
+           "pandas_plotting_backends": [
+               "matplotlib = pandas:plotting._matplotlib",
+           ],
+       },
+   )
+
+
+More information on how to implement a third-party plotting backend can be found at
+https://github.com/pandas-dev/pandas/blob/main/pandas/plotting/__init__.py#L1.
+
+.. _extending.pandas_priority:
+
+Arithmetic with 3rd party types
+-------------------------------
+
+In order to control how arithmetic works between a custom type and a pandas type,
+implement ``__pandas_priority__``.  Similar to numpy's ``__array_priority__``
+semantics, arithmetic methods on :class:`DataFrame`, :class:`Series`, and :class:`Index`
+objects will delegate to ``other``, if it has an attribute ``__pandas_priority__`` with a higher value.
+
+By default, pandas objects try to operate with other objects, even if they are not types known to pandas:
+
+.. code-block:: python
+
+    >>> pd.Series([1, 2]) + [10, 20]
+    0    11
+    1    22
+    dtype: int64
+
+In the example above, if ``[10, 20]`` was a custom type that can be understood as a list, pandas objects will still operate with it in the same way.
+
+In some cases, it is useful to delegate to the other type the operation. For example, consider I implement a
+custom list object, and I want the result of adding my custom list with a pandas :class:`Series` to be an instance of my list
+and not a :class:`Series` as seen in the previous example. This is now possible by defining the ``__pandas_priority__`` attribute
+of my custom list, and setting it to a higher value, than the priority of the pandas objects I want to operate with.
+
+The ``__pandas_priority__`` of :class:`DataFrame`, :class:`Series`, and :class:`Index` are ``4000``, ``3000``, and ``2000`` respectively.  The base ``ExtensionArray.__pandas_priority__`` is ``1000``.
+
+.. code-block:: python
+
+    class CustomList(list):
+        __pandas_priority__ = 5000
+
+        def __radd__(self, other):
+            # return `self` and not the addition for simplicity
+            return self
+
+    custom = CustomList()
+    series = pd.Series([1, 2, 3])
+
+    # Series refuses to add custom, since it's an unknown type with higher priority
+    assert series.__add__(custom) is NotImplemented
+
+    # This will cause the custom class `__radd__` being used instead
+    assert series + custom is custom
diff --git a/pandas/doc/source/development/gitpod-imgs/gitpod-dashboard-stop.png b/pandas/doc/source/development/gitpod-imgs/gitpod-dashboard-stop.png
new file mode 100644
index 0000000000000..b64790a986646
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/gitpod-dashboard-stop.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-gh.png b/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-gh.png
new file mode 100644
index 0000000000000..ec21a9064c83d
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-gh.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-repo.png b/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-repo.png
new file mode 100644
index 0000000000000..8bfaff81cfb69
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/gitpod-edit-permissions-repo.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/gitpod-workspace.png b/pandas/doc/source/development/gitpod-imgs/gitpod-workspace.png
new file mode 100644
index 0000000000000..daf763e9adb05
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/gitpod-workspace.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/pandas-github.png b/pandas/doc/source/development/gitpod-imgs/pandas-github.png
new file mode 100644
index 0000000000000..010b0fc5ea33d
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/pandas-github.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/pandas-gitpod-branches.png b/pandas/doc/source/development/gitpod-imgs/pandas-gitpod-branches.png
new file mode 100644
index 0000000000000..f95c66056ca37
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/pandas-gitpod-branches.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/rst-rendering.png b/pandas/doc/source/development/gitpod-imgs/rst-rendering.png
new file mode 100644
index 0000000000000..b613c621c398b
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/rst-rendering.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/vscode-rst.png b/pandas/doc/source/development/gitpod-imgs/vscode-rst.png
new file mode 100644
index 0000000000000..5b574c115a2b7
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/vscode-rst.png differ
diff --git a/pandas/doc/source/development/gitpod-imgs/vscode-statusbar.png b/pandas/doc/source/development/gitpod-imgs/vscode-statusbar.png
new file mode 100644
index 0000000000000..dad25369fedfd
Binary files /dev/null and b/pandas/doc/source/development/gitpod-imgs/vscode-statusbar.png differ
diff --git a/pandas/doc/source/development/index.rst b/pandas/doc/source/development/index.rst
new file mode 100644
index 0000000000000..aa7e7845bfa7a
--- /dev/null
+++ b/pandas/doc/source/development/index.rst
@@ -0,0 +1,26 @@
+{{ header }}
+
+.. _development:
+
+===========
+Development
+===========
+
+.. If you update this toctree, also update the manual toctree in the
+   main index.rst.template
+
+.. toctree::
+    :maxdepth: 2
+
+    contributing
+    contributing_environment
+    contributing_documentation
+    contributing_codebase
+    maintaining
+    internals
+    copy_on_write
+    debugging_extensions
+    extending
+    developer
+    policies
+    community
diff --git a/pandas/doc/source/development/internals.rst b/pandas/doc/source/development/internals.rst
new file mode 100644
index 0000000000000..e3468746ce177
--- /dev/null
+++ b/pandas/doc/source/development/internals.rst
@@ -0,0 +1,78 @@
+.. _internals:
+
+{{ header }}
+
+*********
+Internals
+*********
+
+This section will provide a look into some of pandas internals. It's primarily
+intended for developers of pandas itself.
+
+Indexing
+--------
+
+In pandas there are a few objects implemented which can serve as valid
+containers for the axis labels:
+
+* :class:`Index`: the generic "ordered set" object, an ndarray of object dtype
+  assuming nothing about its contents. The labels must be hashable (and
+  likely immutable) and unique. Populates a dict of label to location in
+  Cython to do ``O(1)`` lookups.
+* :class:`MultiIndex`: the standard hierarchical index object
+* :class:`DatetimeIndex`: An Index object with :class:`Timestamp` boxed elements (impl are the int64 values)
+* :class:`TimedeltaIndex`: An Index object with :class:`Timedelta` boxed elements (impl are the in64 values)
+* :class:`PeriodIndex`: An Index object with Period elements
+
+There are functions that make the creation of a regular index easy:
+
+* :func:`date_range`: fixed frequency date range generated from a time rule or
+  DateOffset. An ndarray of Python datetime objects
+* :func:`period_range`: fixed frequency date range generated from a time rule or
+  DateOffset. An ndarray of :class:`Period` objects, representing timespans
+
+.. warning::
+
+   Custom :class:`Index` subclasses are not supported, custom behavior should be implemented using the :class:`ExtensionArray` interface instead.
+
+MultiIndex
+~~~~~~~~~~
+
+Internally, the :class:`MultiIndex` consists of a few things: the **levels**, the
+integer **codes**, and the level **names**:
+
+.. ipython:: python
+
+   index = pd.MultiIndex.from_product(
+       [range(3), ["one", "two"]], names=["first", "second"]
+   )
+   index
+   index.levels
+   index.codes
+   index.names
+
+You can probably guess that the codes determine which unique element is
+identified with that location at each layer of the index. It's important to
+note that sortedness is determined **solely** from the integer codes and does
+not check (or care) whether the levels themselves are sorted. Fortunately, the
+constructors :meth:`~MultiIndex.from_tuples` and :meth:`~MultiIndex.from_arrays` ensure
+that this is true, but if you compute the levels and codes yourself, please be careful.
+
+Values
+~~~~~~
+
+pandas extends NumPy's type system with custom types, like :class:`Categorical` or
+datetimes with a timezone, so we have multiple notions of "values". For 1-D
+containers (``Index`` classes and ``Series``) we have the following convention:
+
+* ``cls._values`` refers is the "best possible" array. This could be an
+  ``ndarray`` or ``ExtensionArray``.
+
+So, for example, ``Series[category]._values`` is a ``Categorical``.
+
+.. _ref-subclassing-pandas:
+
+Subclassing pandas data structures
+----------------------------------
+
+This section has been moved to :ref:`extending.subclassing-pandas`.
diff --git a/pandas/doc/source/development/maintaining.rst b/pandas/doc/source/development/maintaining.rst
new file mode 100644
index 0000000000000..c572559dcc3e0
--- /dev/null
+++ b/pandas/doc/source/development/maintaining.rst
@@ -0,0 +1,497 @@
+.. _maintaining:
+
+******************
+pandas maintenance
+******************
+
+This guide is for pandas' maintainers. It may also be interesting to contributors
+looking to understand the pandas development process and what steps are necessary
+to become a maintainer.
+
+The main contributing guide is available at :ref:`contributing`.
+
+Roles
+-----
+
+pandas uses two levels of permissions: **triage** and **core** team members.
+
+Triage members can label and close issues and pull requests.
+
+Core team members can label and close issues and pull request, and can merge
+pull requests.
+
+GitHub publishes the full `list of permissions`_.
+
+Tasks
+-----
+
+pandas is largely a volunteer project, so these tasks shouldn't be read as
+"expectations" of triage and maintainers. Rather, they're general descriptions
+of what it means to be a maintainer.
+
+* Triage newly filed issues (see :ref:`maintaining.triage`)
+* Review newly opened pull requests
+* Respond to updates on existing issues and pull requests
+* Drive discussion and decisions on stalled issues and pull requests
+* Provide experience / wisdom on API design questions to ensure consistency and maintainability
+* Project organization (run / attend developer meetings, represent pandas)
+
+https://matthewrocklin.com/blog/2019/05/18/maintainer may be interesting background
+reading.
+
+.. _maintaining.triage:
+
+Issue triage
+------------
+
+Triage is an important first step in addressing issues reported by the community, and even
+partial contributions are a great way to help maintain pandas. Only remove the "Needs Triage"
+tag once all of the steps below have been completed.
+
+Here's a typical workflow for triaging a newly opened issue.
+
+1. **Thank the reporter for opening an issue**
+
+   The issue tracker is many people's first interaction with the pandas project itself,
+   beyond just using the library. As such, we want it to be a welcoming, pleasant
+   experience.
+
+2. **Is the necessary information provided?**
+
+   Ideally reporters would fill out the issue template, but many don't.
+   If crucial information (like the version of pandas they used), is missing
+   feel free to ask for that and label the issue with "Needs info". The
+   report should follow the guidelines in :ref:`contributing.bug_reports`.
+   You may want to link to that if they didn't follow the template.
+
+   Make sure that the title accurately reflects the issue. Edit it yourself
+   if it's not clear.
+
+3. **Is this a duplicate issue?**
+
+   We have many open issues. If a new issue is clearly a duplicate, label the
+   new issue as "Duplicate" and close the issue with a link to the original issue.
+   Make sure to still thank the reporter, and encourage them to chime in on the
+   original issue, and perhaps try to fix it.
+
+   If the new issue provides relevant information, such as a better or slightly
+   different example, add it to the original issue as a comment or an edit to
+   the original post.
+
+4. **Is the issue minimal and reproducible**?
+
+   For bug reports, we ask that the reporter provide a minimal reproducible
+   example. See https://matthewrocklin.com/blog/work/2018/02/28/minimal-bug-reports
+   for a good explanation. If the example is not reproducible, or if it's
+   *clearly* not minimal, feel free to ask the reporter if they can provide
+   an example or simplify the provided one. Do acknowledge that writing
+   minimal reproducible examples is hard work. If the reporter is struggling,
+   you can try to write one yourself and we'll edit the original post to include it.
+
+   If a reproducible example can't be provided, add the "Needs info" label.
+
+   If a reproducible example is provided, but you see a simplification,
+   edit the original post with your simpler reproducible example.
+
+   If this is a regression report, post the result of a ``git bisect`` run.
+   More info on this can be found in the :ref:`maintaining.regressions` section.
+
+   Ensure the issue exists on the main branch and that it has the "Needs Triage" tag
+   until all steps have been completed. Add a comment to the issue once you have
+   verified it exists on the main branch, so others know it has been confirmed.
+
+5. **Is this a clearly defined feature request?**
+
+   Generally, pandas prefers to discuss and design new features in issues, before
+   a pull request is made. Encourage the submitter to include a proposed API
+   for the new feature. Having them write a full docstring is a good way to
+   pin down specifics.
+
+   Tag new feature requests with "Needs Discussion", as we'll need a discussion
+   from several pandas maintainers before deciding whether the proposal is in
+   scope for pandas.
+
+6. **Is this a usage question?**
+
+   We prefer that usage questions are asked on StackOverflow with the pandas
+   tag. https://stackoverflow.com/questions/tagged/pandas
+
+   If it's easy to answer, feel free to link to the relevant documentation section,
+   let them know that in the future this kind of question should be on
+   StackOverflow, and close the issue.
+
+7. **What labels and milestones should I add?**
+
+   Apply the relevant labels. This is a bit of an art, and comes with experience.
+   Look at similar issues to get a feel for how things are labeled.
+
+   If the issue is clearly defined and the fix seems relatively straightforward,
+   label the issue as "Good first issue".
+
+   If the issue is a regression report, add the "Regression" label and the next patch
+   release milestone.
+
+   Once you have completed the above, make sure to remove the "Needs Triage" label.
+
+.. _maintaining.regressions:
+
+Investigating regressions
+-------------------------
+
+Regressions are bugs that unintentionally break previously working code. The common way
+to  investigate regressions is by using
+`git bisect <https://git-scm.com/docs/git-bisect>`_,
+which finds the first commit that introduced the bug.
+
+For example: a user reports that ``pd.Series([1, 1]).sum()`` returns ``3``
+in pandas version ``1.5.0`` while in version ``1.4.0`` it returned ``2``. To begin,
+create a file ``t.py`` in your pandas directory, which contains
+
+.. code-block:: python
+
+    import pandas as pd
+    assert pd.Series([1, 1]).sum() == 2
+
+and then run::
+
+    git bisect start
+    git bisect good v1.4.0
+    git bisect bad v1.5.0
+    git bisect run bash -c "python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true; python t.py"
+
+This finds the first commit that changed the behavior. The C extensions have to be
+rebuilt at every step, so the search can take a while.
+
+Exit bisect and rebuild the current version::
+
+    git bisect reset
+    python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true
+
+Report your findings under the corresponding issue and ping the commit author to get
+their input.
+
+.. note::
+    In the ``bisect run`` command above, commits are considered good if ``t.py`` exits
+    with ``0`` and bad otherwise. When raising an exception is the desired behavior,
+    wrap the code in an appropriate ``try/except`` statement. See :issue:`35685` for
+    more examples.
+
+.. _maintaining.closing:
+
+Closing issues
+--------------
+
+Be delicate here: many people interpret closing an issue as us saying that the
+conversation is over. It's typically best to give the reporter some time to
+respond or self-close their issue if it's determined that the behavior is not a bug,
+or the feature is out of scope. Sometimes reporters just go away though, and
+we'll close the issue after the conversation has died.
+If you think an issue should be closed but are not completely sure, please apply
+the "closing candidate" label and wait for other maintainers to take a look.
+
+.. _maintaining.reviewing:
+
+Reviewing pull requests
+-----------------------
+
+Anybody can review a pull request: regular contributors, triagers, or core-team
+members. But only core-team members can merge pull requests when they're ready.
+
+Here are some things to check when reviewing a pull request.
+
+* Tests should be in a sensible location: in the same file as closely related tests.
+* New public APIs should be included somewhere in ``doc/source/reference/``.
+* New / changed API should use the ``versionadded`` or ``versionchanged`` directives in the docstring.
+* User-facing changes should have a whatsnew in the appropriate file.
+* Regression tests should reference the original GitHub issue number like ``# GH-1234``.
+* The pull request should be labeled and assigned the appropriate milestone (the next patch release
+  for regression fixes and small bug fixes, the next minor milestone otherwise)
+* Changes should comply with our :ref:`policies.version`.
+
+
+.. _maintaining.backporting:
+
+Backporting
+-----------
+
+pandas supports point releases (e.g. ``1.4.3``) that aim to:
+
+1. Fix bugs in new features introduced in the first minor version release.
+
+  * e.g. If a new feature was added in ``1.4`` and contains a bug, a fix can be applied in ``1.4.3``
+
+2. Fix bugs that used to work in a few minor releases prior. There should be agreement between core team members that a backport is appropriate.
+
+  * e.g. If a feature worked in ``1.2`` and stopped working since ``1.3``, a fix can be applied in ``1.4.3``.
+
+Since pandas minor releases are based on GitHub branches (e.g. point release of ``1.4`` are based off the ``1.4.x`` branch),
+"backporting" means merging a pull request fix to the ``main`` branch and correct minor branch associated with the next point release.
+
+By default, if a pull request is assigned to the next point release milestone within the GitHub interface,
+the backporting process should happen automatically by the ``@meeseeksdev`` bot once the pull request is merged.
+A new pull request will be made backporting the pull request to the correct version branch.
+Sometimes due to merge conflicts, a manual pull request will need to be made addressing the code conflict.
+
+If the bot does not automatically start the backporting process, you can also write a GitHub comment in the merged pull request
+to trigger the backport::
+
+    @meeseeksdev backport version-branch
+
+This will trigger a workflow which will backport a given change to a branch
+(e.g. @meeseeksdev backport 1.4.x)
+
+Cleaning up old issues
+----------------------
+
+Every open issue in pandas has a cost. Open issues make finding duplicates harder,
+and can make it harder to know what needs to be done in pandas. That said, closing
+issues isn't a goal on its own. Our goal is to make pandas the best it can be,
+and that's best done by ensuring that the quality of our open issues is high.
+
+Occasionally, bugs are fixed but the issue isn't linked to in the Pull Request.
+In these cases, comment that "This has been fixed, but could use a test." and
+label the issue as "Good First Issue" and "Needs Test".
+
+If an older issue doesn't follow our issue template, edit the original post to
+include a minimal example, the actual output, and the expected output. Uniformity
+in issue reports is valuable.
+
+If an older issue lacks a reproducible example, label it as "Needs Info" and
+ask them to provide one (or write one yourself if possible). If one isn't
+provide reasonably soon, close it according to the policies in :ref:`maintaining.closing`.
+
+Cleaning up old pull requests
+-----------------------------
+
+Occasionally, contributors are unable to finish off a pull request.
+If some time has passed (two weeks, say) since the last review requesting changes,
+gently ask if they're still interested in working on this. If another two weeks or
+so passes with no response, thank them for their work and then either:
+
+- close the pull request;
+- push to the contributor's branch to push their work over the finish line (if
+  you're part of ``pandas-core``). This can be helpful for pushing an important PR
+  across the line, or for fixing a small merge conflict.
+
+If closing the pull request, then please comment on the original issue that
+"There's a stalled PR at #1234 that may be helpful.", and perhaps label the issue
+as "Good first issue" if the PR was relatively close to being accepted.
+
+Becoming a pandas maintainer
+----------------------------
+
+The full process is outlined in our `governance documents`_. In summary,
+we're happy to give triage permissions to anyone who shows interest by
+being helpful on the issue tracker.
+
+The required steps for adding a maintainer are:
+
+1. Contact the contributor and ask their interest to join.
+2. Add the contributor to the appropriate `GitHub Team <https://github.com/orgs/pandas-dev/teams>`_ if accepted the invitation.
+
+  * ``pandas-core`` is for core team members
+  * ``pandas-triage`` is for pandas triage members
+
+If adding to ``pandas-core``, there are two additional steps:
+
+3. Add the contributor to the pandas Google group.
+4. Create a pull request to add the contributor's GitHub handle to ``pandas-dev/pandas/web/pandas/config.yml``.
+
+The current list of core-team members is at
+https://github.com/pandas-dev/pandas/blob/main/web/pandas/config.yml
+
+
+.. _maintaining.merging:
+
+Merging pull requests
+---------------------
+
+Only core team members can merge pull requests. We have a few guidelines.
+
+1. You should typically not self-merge your own pull requests without approval.
+   Exceptions include things like small changes to fix CI
+   (e.g. pinning a package version). Self-merging with approval from other
+   core team members is fine if the change is something you're very confident
+   about.
+2. You should not merge pull requests that have an active discussion, or pull
+   requests that has any ``-1`` votes from a core maintainer. pandas operates
+   by consensus.
+3. For larger changes, it's good to have a +1 from at least two core team members.
+
+In addition to the items listed in :ref:`maintaining.closing`, you should verify
+that the pull request is assigned the correct milestone.
+
+Pull requests merged with a patch-release milestone will typically be backported
+by our bot. Verify that the bot noticed the merge (it will leave a comment within
+a minute typically). If a manual backport is needed please do that, and remove
+the "Needs backport" label once you've done it manually. If you forget to assign
+a milestone before tagging, you can request the bot to backport it with:
+
+.. code-block:: console
+
+   @Meeseeksdev backport <branch>
+
+
+.. _maintaining.release:
+
+Release process
+---------------
+
+The release process makes a snapshot of pandas (a git commit) available to users with
+a particular version number. After the release the new pandas version will be available
+in the next places:
+
+- Git repo with a `new tag <https://github.com/pandas-dev/pandas/tags>`_
+- Source distribution in a `GitHub release <https://github.com/pandas-dev/pandas/releases>`_
+- Pip packages in the `PyPI <https://pypi.org/project/pandas/>`_
+- Conda packages in `conda-forge <https://anaconda.org/conda-forge/pandas>`_
+
+The process for releasing a new version of pandas is detailed next section.
+
+The instructions contain ``<version>`` which needs to be replaced with the version
+to be released (e.g. ``1.5.2``). Also the branch to be released ``<branch>``, which
+depends on whether the version being released is the release candidate of a new version,
+or any other version. Release candidates are released from ``main``, while other
+versions are released from their branch (e.g. ``1.5.x``).
+
+
+Prerequisites
+`````````````
+
+In order to be able to release a new pandas version, the next permissions are needed:
+
+- Merge rights to the `pandas <https://github.com/pandas-dev/pandas/>`_ and
+  `pandas-feedstock <https://github.com/conda-forge/pandas-feedstock/>`_ repositories.
+  For the latter, open a PR adding your GitHub username to the conda-forge recipe.
+- Permissions to push to ``main`` in the pandas repository, to push the new tags.
+- `Write permissions to PyPI <https://github.com/conda-forge/pandas-feedstock/pulls>`_.
+- Access to our website / documentation server. Share your public key with the
+  infrastructure committee to be added to the ``authorized_keys`` file of the main
+  server user.
+- Access to the social media accounts, to publish the announcements.
+
+Pre-release
+```````````
+
+1. Agree with the core team on the next topics:
+
+   - Release date (major/minor releases happen usually every 6 months, and patch releases
+     monthly until x.x.5, just before the next major/minor)
+   - Blockers (issues and PRs that must be part of the release)
+   - Next version after the one being released
+
+2. Update and clean release notes for the version to be released, including:
+
+   - Set the final date of the release
+   - Remove any unused bullet point
+   - Make sure there are no formatting issues, typos, etc.
+
+3. Make sure the CI is green for the last commit of the branch being released.
+
+4. If not a release candidate, make sure all backporting pull requests to the branch
+   being released are merged.
+
+5. Create a new issue and milestone for the version after the one being released.
+   If the release was a release candidate, we would usually want to create issues and
+   milestones for both the next major/minor, and the next patch release. In the
+   milestone of a patch release, we add the description ``on-merge: backport to <branch>``,
+   so tagged PRs are automatically backported to the release branch by our bot.
+
+6. Change the milestone of all issues and PRs in the milestone being released to the
+   next milestone.
+
+Release
+```````
+
+1. Create an empty commit and a tag in the last commit of the branch to be released::
+
+    git checkout <branch>
+    git pull --ff-only upstream <branch>
+    git clean -xdf
+    git commit --allow-empty --author="pandas Development Team <pandas-dev@python.org>" -m "RLS: <version>"
+    git tag -a v<version> -m "Version <version>"  # NOTE that the tag is v1.5.2 with "v" not 1.5.2
+    git push upstream <branch> --follow-tags
+
+The docs for the new version will be built and published automatically with the docs job in the CI,
+which will be triggered when the tag is pushed.
+
+2. Only if the release is a release candidate, we want to create a new branch for it, immediately
+   after creating the tag. For example, if we are releasing pandas 1.4.0rc0, we would like to
+   create the branch 1.4.x to backport commits to the 1.4 versions. As well as create a tag to
+   mark the start of the development of 1.5.0 (assuming it is the next version)::
+
+    git checkout -b 1.4.x
+    git push upstream 1.4.x
+    git checkout main
+    git commit --allow-empty -m "Start 1.5.0"
+    git tag -a v1.5.0.dev0 -m "DEV: Start 1.5.0"
+    git push upstream main --follow-tags
+
+3. Download the source distribution and wheels from the `wheel staging area <https://anaconda.org/scientific-python-nightly-wheels/pandas>`_.
+   Be careful to make sure that no wheels are missing (e.g. due to failed builds).
+
+   Running scripts/download_wheels.sh with the version that you want to download wheels/the sdist for should do the trick.
+   This script will make a ``dist`` folder inside your clone of pandas and put the downloaded wheels and sdist there::
+
+    scripts/download_wheels.sh <VERSION>
+
+4. Create a `new GitHub release <https://github.com/pandas-dev/pandas/releases/new>`_:
+
+   - Tag: ``<version>``
+   - Title: ``pandas <version>``
+   - Description: Copy the description of the last release of the same kind (release candidate, major/minor or patch release)
+   - Files: ``pandas-<version>.tar.gz`` source distribution just generated
+   - Set as a pre-release: Only check for a release candidate
+   - Set as the latest release: Leave checked, unless releasing a patch release for an older version
+     (e.g. releasing 1.4.5 after 1.5 has been released)
+
+5. Upload wheels to PyPI::
+
+    twine upload pandas/dist/pandas-<version>*.{whl,tar.gz} --skip-existing
+
+6. The GitHub release will after some hours trigger an
+   `automated conda-forge PR <https://github.com/conda-forge/pandas-feedstock/pulls>`_.
+   (If you don't want to wait, you can open an issue titled ``@conda-forge-admin, please update version`` to trigger the bot.)
+   Merge it once the CI is green, and it will generate the conda-forge packages.
+
+   In case a manual PR needs to be done, the version, sha256 and build fields are the
+   ones that usually need to be changed. If anything else in the recipe has changed since
+   the last release, those changes should be available in ``ci/meta.yaml``.
+
+Post-Release
+````````````
+
+1. Update symlinks to stable documentation by logging in to our web server, and
+   editing ``/var/www/html/pandas-docs/stable`` to point to ``version/<latest-version>``
+   for major and minor releases, or ``version/<minor>`` to ``version/<patch>`` for
+   patch releases. The exact instructions are (replace the example version numbers by
+   the appropriate ones for the version you are releasing):
+
+    - Log in to the server and use the correct user.
+    - ``cd /var/www/html/pandas-docs/``
+    - ``ln -sfn version/2.1 stable`` (for a major or minor release)
+    - ``ln -sfn version/2.0.3 version/2.0`` (for a patch release)
+
+2. If releasing a major or minor release, open a PR in our source code to update
+   ``web/pandas/versions.json``, to have the desired versions in the documentation
+   dropdown menu.
+
+3. Close the milestone and the issue for the released version.
+
+4. Create a new issue for the next release, with the estimated date of release.
+
+5. Open a PR with the placeholder for the release notes of the next version. See
+   for example `the PR for 1.5.3 <https://github.com/pandas-dev/pandas/pull/49843/files>`_.
+   Note that the template to use depends on whether it is a major, minor or patch release.
+
+6. Announce the new release in the official channels (use previous announcements
+   for reference):
+
+    - The pandas-dev and pydata mailing lists
+    - X, Mastodon, Telegram and LinkedIn
+
+7. Update this release instructions to fix anything incorrect and to update about any
+   change since the last release.
+
+.. _governance documents: https://github.com/pandas-dev/pandas/blob/main/web/pandas/about/governance.md
+.. _list of permissions: https://docs.github.com/en/organizations/managing-access-to-your-organizations-repositories/repository-roles-for-an-organization
diff --git a/pandas/doc/source/development/policies.rst b/pandas/doc/source/development/policies.rst
new file mode 100644
index 0000000000000..a3665c5bb4d1f
--- /dev/null
+++ b/pandas/doc/source/development/policies.rst
@@ -0,0 +1,61 @@
+.. _develop.policies:
+
+********
+Policies
+********
+
+.. _policies.version:
+
+Version policy
+~~~~~~~~~~~~~~
+
+pandas uses a loose variant of semantic versioning (`SemVer`_) to govern
+deprecations, API compatibility, and version numbering.
+
+A pandas release number is made up of ``MAJOR.MINOR.PATCH``.
+
+API breaking changes should only occur in **major** releases. These changes
+will be documented, with clear guidance on what is changing, why it's changing,
+and how to migrate existing code to the new behavior.
+
+Whenever possible, a deprecation path will be provided rather than an outright
+breaking change.
+
+pandas will introduce deprecations in **minor** releases. These deprecations
+will preserve the existing behavior while emitting a warning that provide
+guidance on:
+
+* How to achieve similar behavior if an alternative is available
+* The pandas version in which the deprecation will be enforced.
+
+We will not introduce new deprecations in patch releases.
+
+Deprecations will only be enforced in **major** releases. For example, if a
+behavior is deprecated in pandas 1.2.0, it will continue to work, with a
+warning, for all releases in the 1.x series. The behavior will change and the
+deprecation removed in the next major release (2.0.0).
+
+.. note::
+
+   pandas will sometimes make *behavior changing* bug fixes, as part of
+   minor or patch releases. Whether or not a change is a bug fix or an
+   API-breaking change is a judgement call. We'll do our best, and we
+   invite you to participate in development discussion on the issue
+   tracker or mailing list.
+
+These policies do not apply to features marked as **experimental** in the documentation.
+pandas may change the behavior of experimental features at any time.
+
+.. _policies.python_support:
+
+Python support
+~~~~~~~~~~~~~~
+
+pandas mirrors the `SPEC 0 guideline for Python support <https://scientific-python.org/specs/spec-0000>`__.
+
+Security policy
+~~~~~~~~~~~~~~~
+
+To report a security vulnerability to pandas, please go to https://github.com/pandas-dev/pandas/security/policy and see the instructions there.
+
+.. _SemVer: https://semver.org
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_r.rst b/pandas/doc/source/getting_started/comparison/comparison_with_r.rst
new file mode 100644
index 0000000000000..d9d7d916b0238
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_r.rst
@@ -0,0 +1,577 @@
+.. _compare_with_r:
+
+{{ header }}
+
+Comparison with R / R libraries
+*******************************
+
+Since pandas aims to provide a lot of the data manipulation and analysis
+functionality that people use `R <https://www.r-project.org/>`__ for, this page
+was started to provide a more detailed look at the `R language
+<https://en.wikipedia.org/wiki/R_(programming_language)>`__ and its many third
+party libraries as they relate to pandas. In comparisons with R and CRAN
+libraries, we care about the following things:
+
+* **Functionality / flexibility**: what can/cannot be done with each tool
+* **Performance**: how fast are operations. Hard numbers/benchmarks are
+  preferable
+* **Ease-of-use**: Is one tool easier/harder to use (you may have to be
+  the judge of this, given side-by-side code comparisons)
+
+This page is also here to offer a bit of a translation guide for users of these
+R packages.
+
+
+Quick reference
+---------------
+
+We'll start off with a quick reference guide pairing some common R
+operations using `dplyr
+<https://cran.r-project.org/web/packages/dplyr/index.html>`__ with
+pandas equivalents.
+
+
+Querying, filtering, sampling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+===========================================  ===========================================
+R                                            pandas
+===========================================  ===========================================
+``dim(df)``                                  ``df.shape``
+``head(df)``                                 ``df.head()``
+``slice(df, 1:10)``                          ``df.iloc[:9]``
+``filter(df, col1 == 1, col2 == 1)``         ``df.query('col1 == 1 & col2 == 1')``
+``df[df$col1 == 1 & df$col2 == 1,]``         ``df[(df.col1 == 1) & (df.col2 == 1)]``
+``select(df, col1, col2)``                   ``df[['col1', 'col2']]``
+``select(df, col1:col3)``                    ``df.loc[:, 'col1':'col3']``
+``select(df, -(col1:col3))``                 ``df.drop(cols_to_drop, axis=1)`` but see [#select_range]_
+``distinct(select(df, col1))``               ``df[['col1']].drop_duplicates()``
+``distinct(select(df, col1, col2))``         ``df[['col1', 'col2']].drop_duplicates()``
+``sample_n(df, 10)``                         ``df.sample(n=10)``
+``sample_frac(df, 0.01)``                    ``df.sample(frac=0.01)``
+===========================================  ===========================================
+
+.. [#select_range] R's shorthand for a subrange of columns
+                   (``select(df, col1:col3)``) can be approached
+                   cleanly in pandas, if you have the list of columns,
+                   for example ``df[cols[1:3]]`` or
+                   ``df.drop(cols[1:3])``, but doing this by column
+                   name is a bit messy.
+
+
+Sorting
+~~~~~~~
+
+===========================================  ===========================================
+R                                            pandas
+===========================================  ===========================================
+``arrange(df, col1, col2)``                  ``df.sort_values(['col1', 'col2'])``
+``arrange(df, desc(col1))``                  ``df.sort_values('col1', ascending=False)``
+===========================================  ===========================================
+
+Transforming
+~~~~~~~~~~~~
+
+===========================================  ===========================================
+R                                            pandas
+===========================================  ===========================================
+``select(df, col_one = col1)``               ``df.rename(columns={'col1': 'col_one'})['col_one']``
+``rename(df, col_one = col1)``               ``df.rename(columns={'col1': 'col_one'})``
+``mutate(df, c=a-b)``                        ``df.assign(c=df['a']-df['b'])``
+===========================================  ===========================================
+
+
+Grouping and summarizing
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+==============================================  ===========================================
+R                                               pandas
+==============================================  ===========================================
+``summary(df)``                                 ``df.describe()``
+``gdf <- group_by(df, col1)``                   ``gdf = df.groupby('col1')``
+``summarise(gdf, avg=mean(col1, na.rm=TRUE))``  ``df.groupby('col1').agg({'col1': 'mean'})``
+``summarise(gdf, total=sum(col1))``             ``df.groupby('col1').sum()``
+==============================================  ===========================================
+
+
+Base R
+------
+
+Slicing with R's |c|_
+~~~~~~~~~~~~~~~~~~~~~
+
+R makes it easy to access ``data.frame`` columns by name
+
+.. code-block:: r
+
+   df <- data.frame(a=rnorm(5), b=rnorm(5), c=rnorm(5), d=rnorm(5), e=rnorm(5))
+   df[, c("a", "c", "e")]
+
+or by integer location
+
+.. code-block:: r
+
+   df <- data.frame(matrix(rnorm(1000), ncol=100))
+   df[, c(1:10, 25:30, 40, 50:100)]
+
+Selecting multiple columns by name in pandas is straightforward
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 3), columns=list("abc"))
+   df[["a", "c"]]
+   df.loc[:, ["a", "c"]]
+
+Selecting multiple noncontiguous columns by integer location can be achieved
+with a combination of the ``iloc`` indexer attribute and ``numpy.r_``.
+
+.. ipython:: python
+
+   named = list("abcdefg")
+   n = 30
+   columns = named + np.arange(len(named), n).tolist()
+   df = pd.DataFrame(np.random.randn(n, n), columns=columns)
+
+   df.iloc[:, np.r_[:10, 24:30]]
+
+|aggregate|_
+~~~~~~~~~~~~
+
+In R you may want to split data into subsets and compute the mean for each.
+Using a data.frame called ``df`` and splitting it into groups ``by1`` and
+``by2``:
+
+.. code-block:: r
+
+   df <- data.frame(
+     v1 = c(1,3,5,7,8,3,5,NA,4,5,7,9),
+     v2 = c(11,33,55,77,88,33,55,NA,44,55,77,99),
+     by1 = c("red", "blue", 1, 2, NA, "big", 1, 2, "red", 1, NA, 12),
+     by2 = c("wet", "dry", 99, 95, NA, "damp", 95, 99, "red", 99, NA, NA))
+   aggregate(x=df[, c("v1", "v2")], by=list(mydf2$by1, mydf2$by2), FUN = mean)
+
+The :meth:`~pandas.DataFrame.groupby` method is similar to base R ``aggregate``
+function.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "v1": [1, 3, 5, 7, 8, 3, 5, np.nan, 4, 5, 7, 9],
+           "v2": [11, 33, 55, 77, 88, 33, 55, np.nan, 44, 55, 77, 99],
+           "by1": ["red", "blue", 1, 2, np.nan, "big", 1, 2, "red", 1, np.nan, 12],
+           "by2": [
+               "wet",
+               "dry",
+               99,
+               95,
+               np.nan,
+               "damp",
+               95,
+               99,
+               "red",
+               99,
+               np.nan,
+               np.nan,
+           ],
+       }
+   )
+
+   g = df.groupby(["by1", "by2"])
+   g[["v1", "v2"]].mean()
+
+For more details and examples see :ref:`the groupby documentation
+<groupby.split>`.
+
+|match|_
+~~~~~~~~~~~~
+
+A common way to select data in R is using ``%in%`` which is defined using the
+function ``match``. The operator ``%in%`` is used to return a logical vector
+indicating if there is a match or not:
+
+.. code-block:: r
+
+   s <- 0:4
+   s %in% c(2,4)
+
+The :meth:`~pandas.DataFrame.isin` method is similar to R ``%in%`` operator:
+
+.. ipython:: python
+
+   s = pd.Series(np.arange(5), dtype=np.float32)
+   s.isin([2, 4])
+
+The ``match`` function returns a vector of the positions of matches
+of its first argument in its second:
+
+.. code-block:: r
+
+   s <- 0:4
+   match(s, c(2,4))
+
+For more details and examples see :ref:`the reshaping documentation
+<indexing.basics.indexing_isin>`.
+
+|tapply|_
+~~~~~~~~~
+
+``tapply`` is similar to ``aggregate``, but data can be in a ragged array,
+since the subclass sizes are possibly irregular. Using a data.frame called
+``baseball``, and retrieving information based on the array ``team``:
+
+.. code-block:: r
+
+   baseball <-
+     data.frame(team = gl(5, 5,
+                labels = paste("Team", LETTERS[1:5])),
+                player = sample(letters, 25),
+                batting.average = runif(25, .200, .400))
+
+   tapply(baseball$batting.average, baseball.example$team,
+          max)
+
+In pandas we may use :meth:`~pandas.pivot_table` method to handle this:
+
+.. ipython:: python
+
+   import random
+   import string
+
+   baseball = pd.DataFrame(
+       {
+           "team": ["team %d" % (x + 1) for x in range(5)] * 5,
+           "player": random.sample(list(string.ascii_lowercase), 25),
+           "batting avg": np.random.uniform(0.200, 0.400, 25),
+       }
+   )
+
+   baseball.pivot_table(values="batting avg", columns="team", aggfunc="max")
+
+For more details and examples see :ref:`the reshaping documentation
+<reshaping.pivot>`.
+
+|subset|_
+~~~~~~~~~~
+
+The :meth:`~pandas.DataFrame.query` method is similar to the base R ``subset``
+function. In R you might want to get the rows of a ``data.frame`` where one
+column's values are less than another column's values:
+
+.. code-block:: r
+
+   df <- data.frame(a=rnorm(10), b=rnorm(10))
+   subset(df, a <= b)
+   df[df$a <= df$b,]  # note the comma
+
+In pandas, there are a few ways to perform subsetting. You can use
+:meth:`~pandas.DataFrame.query` or pass an expression as if it were an
+index/slice as well as standard boolean indexing:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": np.random.randn(10), "b": np.random.randn(10)})
+   df.query("a <= b")
+   df[df["a"] <= df["b"]]
+   df.loc[df["a"] <= df["b"]]
+
+For more details and examples see :ref:`the query documentation
+<indexing.query>`.
+
+
+|with|_
+~~~~~~~~
+
+An expression using a data.frame called ``df`` in R with the columns ``a`` and
+``b`` would be evaluated using ``with`` like so:
+
+.. code-block:: r
+
+   df <- data.frame(a=rnorm(10), b=rnorm(10))
+   with(df, a + b)
+   df$a + df$b  # same as the previous expression
+
+In pandas the equivalent expression, using the
+:meth:`~pandas.DataFrame.eval` method, would be:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": np.random.randn(10), "b": np.random.randn(10)})
+   df.eval("a + b")
+   df["a"] + df["b"]  # same as the previous expression
+
+In certain cases :meth:`~pandas.DataFrame.eval` will be much faster than
+evaluation in pure Python. For more details and examples see :ref:`the eval
+documentation <enhancingperf.eval>`.
+
+plyr
+----
+
+``plyr`` is an R library for the split-apply-combine strategy for data
+analysis. The functions revolve around three data structures in R, ``a``
+for ``arrays``, ``l`` for ``lists``, and ``d`` for ``data.frame``. The
+table below shows how these data structures could be mapped in Python.
+
++------------+-------------------------------+
+| R          | Python                        |
++============+===============================+
+| array      | list                          |
++------------+-------------------------------+
+| lists      | dictionary or list of objects |
++------------+-------------------------------+
+| data.frame | dataframe                     |
++------------+-------------------------------+
+
+ddply
+~~~~~
+
+An expression using a data.frame called ``df`` in R where you want to
+summarize ``x`` by ``month``:
+
+.. code-block:: r
+
+   require(plyr)
+   df <- data.frame(
+     x = runif(120, 1, 168),
+     y = runif(120, 7, 334),
+     z = runif(120, 1.7, 20.7),
+     month = rep(c(5,6,7,8),30),
+     week = sample(1:4, 120, TRUE)
+   )
+
+   ddply(df, .(month, week), summarize,
+         mean = round(mean(x), 2),
+         sd = round(sd(x), 2))
+
+In pandas the equivalent expression, using the
+:meth:`~pandas.DataFrame.groupby` method, would be:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "x": np.random.uniform(1.0, 168.0, 120),
+           "y": np.random.uniform(7.0, 334.0, 120),
+           "z": np.random.uniform(1.7, 20.7, 120),
+           "month": [5, 6, 7, 8] * 30,
+           "week": np.random.randint(1, 4, 120),
+       }
+   )
+
+   grouped = df.groupby(["month", "week"])
+   grouped["x"].agg(["mean", "std"])
+
+
+For more details and examples see :ref:`the groupby documentation
+<groupby.aggregate>`.
+
+reshape / reshape2
+------------------
+
+meltarray
+~~~~~~~~~
+
+An expression using a 3 dimensional array called ``a`` in R where you want to
+melt it into a data.frame:
+
+.. code-block:: r
+
+   a <- array(c(1:23, NA), c(2,3,4))
+   data.frame(melt(a))
+
+In Python, since ``a`` is a list, you can simply use list comprehension.
+
+.. ipython:: python
+
+   a = np.array(list(range(1, 24)) + [np.NAN]).reshape(2, 3, 4)
+   pd.DataFrame([tuple(list(x) + [val]) for x, val in np.ndenumerate(a)])
+
+meltlist
+~~~~~~~~
+
+An expression using a list called ``a`` in R where you want to melt it
+into a data.frame:
+
+.. code-block:: r
+
+   a <- as.list(c(1:4, NA))
+   data.frame(melt(a))
+
+In Python, this list would be a list of tuples, so
+:meth:`~pandas.DataFrame` method would convert it to a dataframe as required.
+
+.. ipython:: python
+
+   a = list(enumerate(list(range(1, 5)) + [np.NAN]))
+   pd.DataFrame(a)
+
+For more details and examples see :ref:`the Intro to Data Structures
+documentation <dsintro>`.
+
+meltdf
+~~~~~~
+
+An expression using a data.frame called ``cheese`` in R where you want to
+reshape the data.frame:
+
+.. code-block:: r
+
+   cheese <- data.frame(
+     first = c('John', 'Mary'),
+     last = c('Doe', 'Bo'),
+     height = c(5.5, 6.0),
+     weight = c(130, 150)
+   )
+   melt(cheese, id=c("first", "last"))
+
+In Python, the :meth:`~pandas.melt` method is the R equivalent:
+
+.. ipython:: python
+
+   cheese = pd.DataFrame(
+       {
+           "first": ["John", "Mary"],
+           "last": ["Doe", "Bo"],
+           "height": [5.5, 6.0],
+           "weight": [130, 150],
+       }
+   )
+
+   pd.melt(cheese, id_vars=["first", "last"])
+   cheese.set_index(["first", "last"]).stack()  # alternative way
+
+For more details and examples see :ref:`the reshaping documentation
+<reshaping.melt>`.
+
+cast
+~~~~
+
+In R ``acast`` is an expression using a data.frame called ``df`` in R to cast
+into a higher dimensional array:
+
+.. code-block:: r
+
+   df <- data.frame(
+     x = runif(12, 1, 168),
+     y = runif(12, 7, 334),
+     z = runif(12, 1.7, 20.7),
+     month = rep(c(5,6,7),4),
+     week = rep(c(1,2), 6)
+   )
+
+   mdf <- melt(df, id=c("month", "week"))
+   acast(mdf, week ~ month ~ variable, mean)
+
+In Python the best way is to make use of :meth:`~pandas.pivot_table`:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "x": np.random.uniform(1.0, 168.0, 12),
+           "y": np.random.uniform(7.0, 334.0, 12),
+           "z": np.random.uniform(1.7, 20.7, 12),
+           "month": [5, 6, 7] * 4,
+           "week": [1, 2] * 6,
+       }
+   )
+
+   mdf = pd.melt(df, id_vars=["month", "week"])
+   pd.pivot_table(
+       mdf,
+       values="value",
+       index=["variable", "week"],
+       columns=["month"],
+       aggfunc="mean",
+   )
+
+Similarly for ``dcast`` which uses a data.frame called ``df`` in R to
+aggregate information based on ``Animal`` and ``FeedType``:
+
+.. code-block:: r
+
+   df <- data.frame(
+     Animal = c('Animal1', 'Animal2', 'Animal3', 'Animal2', 'Animal1',
+                'Animal2', 'Animal3'),
+     FeedType = c('A', 'B', 'A', 'A', 'B', 'B', 'A'),
+     Amount = c(10, 7, 4, 2, 5, 6, 2)
+   )
+
+   dcast(df, Animal ~ FeedType, sum, fill=NaN)
+   # Alternative method using base R
+   with(df, tapply(Amount, list(Animal, FeedType), sum))
+
+Python can approach this in two different ways. Firstly, similar to above
+using :meth:`~pandas.pivot_table`:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "Animal": [
+               "Animal1",
+               "Animal2",
+               "Animal3",
+               "Animal2",
+               "Animal1",
+               "Animal2",
+               "Animal3",
+           ],
+           "FeedType": ["A", "B", "A", "A", "B", "B", "A"],
+           "Amount": [10, 7, 4, 2, 5, 6, 2],
+       }
+   )
+
+   df.pivot_table(values="Amount", index="Animal", columns="FeedType", aggfunc="sum")
+
+The second approach is to use the :meth:`~pandas.DataFrame.groupby` method:
+
+.. ipython:: python
+
+   df.groupby(["Animal", "FeedType"])["Amount"].sum()
+
+For more details and examples see :ref:`the reshaping documentation
+<reshaping.pivot>` or :ref:`the groupby documentation<groupby.split>`.
+
+|factor|_
+~~~~~~~~~
+
+pandas has a data type for categorical data.
+
+.. code-block:: r
+
+   cut(c(1,2,3,4,5,6), 3)
+   factor(c(1,2,3,2,2,3))
+
+In pandas this is accomplished with ``pd.cut`` and ``astype("category")``:
+
+.. ipython:: python
+
+   pd.cut(pd.Series([1, 2, 3, 4, 5, 6]), 3)
+   pd.Series([1, 2, 3, 2, 2, 3]).astype("category")
+
+For more details and examples see :ref:`categorical introduction <categorical>` and the
+:ref:`API documentation <api.arrays.categorical>`. There is also a documentation regarding the
+:ref:`differences to R's factor <categorical.rfactor>`.
+
+
+.. |c| replace:: ``c``
+.. _c: https://stat.ethz.ch/R-manual/R-patched/library/base/html/c.html
+
+.. |aggregate| replace:: ``aggregate``
+.. _aggregate: https://stat.ethz.ch/R-manual/R-patched/library/stats/html/aggregate.html
+
+.. |match| replace:: ``match`` / ``%in%``
+.. _match: https://stat.ethz.ch/R-manual/R-patched/library/base/html/match.html
+
+.. |tapply| replace:: ``tapply``
+.. _tapply: https://stat.ethz.ch/R-manual/R-patched/library/base/html/tapply.html
+
+.. |with| replace:: ``with``
+.. _with: https://stat.ethz.ch/R-manual/R-patched/library/base/html/with.html
+
+.. |subset| replace:: ``subset``
+.. _subset: https://stat.ethz.ch/R-manual/R-patched/library/base/html/subset.html
+
+.. |factor| replace:: ``factor``
+.. _factor: https://stat.ethz.ch/R-manual/R-devel/library/base/html/factor.html
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_sas.rst b/pandas/doc/source/getting_started/comparison/comparison_with_sas.rst
new file mode 100644
index 0000000000000..595f3c85a9dc2
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_sas.rst
@@ -0,0 +1,584 @@
+.. _compare_with_sas:
+
+{{ header }}
+
+Comparison with SAS
+********************
+
+For potential users coming from `SAS <https://en.wikipedia.org/wiki/SAS_(software)>`__
+this page is meant to demonstrate how different SAS operations would be
+performed in pandas.
+
+.. include:: includes/introduction.rst
+
+
+Data structures
+---------------
+
+General terminology translation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. csv-table::
+    :header: "pandas", "SAS"
+    :widths: 20, 20
+
+    ``DataFrame``, data set
+    column, variable
+    row, observation
+    groupby, BY-group
+    ``NaN``, ``.``
+
+
+``DataFrame``
+~~~~~~~~~~~~~
+
+A ``DataFrame`` in pandas is analogous to a SAS data set - a two-dimensional
+data source with labeled columns that can be of different types. As will be
+shown in this document, almost any operation that can be applied to a data set
+using SAS's ``DATA`` step, can also be accomplished in pandas.
+
+``Series``
+~~~~~~~~~~
+
+A ``Series`` is the data structure that represents one column of a
+``DataFrame``. SAS doesn't have a separate data structure for a single column,
+but in general, working with a ``Series`` is analogous to referencing a column
+in the ``DATA`` step.
+
+``Index``
+~~~~~~~~~
+
+Every ``DataFrame`` and ``Series`` has an ``Index`` - which are labels on the
+*rows* of the data. SAS does not have an exactly analogous concept. A data set's
+rows are essentially unlabeled, other than an implicit integer index that can be
+accessed during the ``DATA`` step (``_N_``).
+
+In pandas, if no index is specified, an integer index is also used by default
+(first row = 0, second row = 1, and so on). While using a labeled ``Index`` or
+``MultiIndex`` can enable sophisticated analyses and is ultimately an important
+part of pandas to understand, for this comparison we will essentially ignore the
+``Index`` and just treat the ``DataFrame`` as a collection of columns. Please
+see the :ref:`indexing documentation<indexing>` for much more on how to use an
+``Index`` effectively.
+
+
+Copies vs. in place operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. include:: includes/copies.rst
+
+
+Data input / output
+-------------------
+
+Constructing a DataFrame from values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A SAS data set can be built from specified values by
+placing the data after a ``datalines`` statement and
+specifying the column names.
+
+.. code-block:: sas
+
+   data df;
+       input x y;
+       datalines;
+       1 2
+       3 4
+       5 6
+       ;
+   run;
+
+.. include:: includes/construct_dataframe.rst
+
+Reading external data
+~~~~~~~~~~~~~~~~~~~~~
+
+Like SAS, pandas provides utilities for reading in data from
+many formats.  The ``tips`` dataset, found within the pandas
+tests (`csv <https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/csv/tips.csv>`_)
+will be used in many of the following examples.
+
+SAS provides ``PROC IMPORT`` to read csv data into a data set.
+
+.. code-block:: sas
+
+   proc import datafile='tips.csv' dbms=csv out=tips replace;
+       getnames=yes;
+   run;
+
+The pandas method is :func:`read_csv`, which works similarly.
+
+.. ipython:: python
+
+   url = (
+       "https://raw.githubusercontent.com/pandas-dev/"
+       "pandas/main/pandas/tests/io/data/csv/tips.csv"
+   )
+   tips = pd.read_csv(url)
+   tips
+
+
+Like ``PROC IMPORT``, ``read_csv`` can take a number of parameters to specify
+how the data should be parsed.  For example, if the data was instead tab delimited,
+and did not have column names, the pandas command would be:
+
+.. code-block:: python
+
+   tips = pd.read_csv("tips.csv", sep="\t", header=None)
+
+   # alternatively, read_table is an alias to read_csv with tab delimiter
+   tips = pd.read_table("tips.csv", header=None)
+
+In addition to text/csv, pandas supports a variety of other data formats
+such as Excel, HDF5, and SQL databases.  These are all read via a ``pd.read_*``
+function.  See the :ref:`IO documentation<io>` for more details.
+
+Limiting output
+~~~~~~~~~~~~~~~
+
+.. include:: includes/limit.rst
+
+The equivalent in SAS would be:
+
+.. code-block:: sas
+
+   proc print data=df(obs=5);
+   run;
+
+
+Exporting data
+~~~~~~~~~~~~~~
+
+The inverse of ``PROC IMPORT`` in SAS is ``PROC EXPORT``
+
+.. code-block:: sas
+
+   proc export data=tips outfile='tips2.csv' dbms=csv;
+   run;
+
+Similarly in pandas, the opposite of ``read_csv`` is :meth:`~DataFrame.to_csv`,
+and other data formats follow a similar api.
+
+.. code-block:: python
+
+   tips.to_csv("tips2.csv")
+
+
+Data operations
+---------------
+
+Operations on columns
+~~~~~~~~~~~~~~~~~~~~~
+
+In the ``DATA`` step, arbitrary math expressions can
+be used on new or existing columns.
+
+.. code-block:: sas
+
+   data tips;
+       set tips;
+       total_bill = total_bill - 2;
+       new_bill = total_bill / 2;
+   run;
+
+.. include:: includes/column_operations.rst
+
+
+Filtering
+~~~~~~~~~
+
+Filtering in SAS is done with an ``if`` or ``where`` statement, on one
+or more columns.
+
+.. code-block:: sas
+
+   data tips;
+       set tips;
+       if total_bill > 10;
+   run;
+
+   data tips;
+       set tips;
+       where total_bill > 10;
+       /* equivalent in this case - where happens before the
+          DATA step begins and can also be used in PROC statements */
+   run;
+
+.. include:: includes/filtering.rst
+
+If/then logic
+~~~~~~~~~~~~~
+
+In SAS, if/then logic can be used to create new columns.
+
+.. code-block:: sas
+
+   data tips;
+       set tips;
+       format bucket $4.;
+
+       if total_bill < 10 then bucket = 'low';
+       else bucket = 'high';
+   run;
+
+.. include:: includes/if_then.rst
+
+Date functionality
+~~~~~~~~~~~~~~~~~~
+
+SAS provides a variety of functions to do operations on
+date/datetime columns.
+
+.. code-block:: sas
+
+   data tips;
+       set tips;
+       format date1 date2 date1_plusmonth mmddyy10.;
+       date1 = mdy(1, 15, 2013);
+       date2 = mdy(2, 15, 2015);
+       date1_year = year(date1);
+       date2_month = month(date2);
+       * shift date to beginning of next interval;
+       date1_next = intnx('MONTH', date1, 1);
+       * count intervals between dates;
+       months_between = intck('MONTH', date1, date2);
+   run;
+
+The equivalent pandas operations are shown below.  In addition to these
+functions pandas supports other Time Series features
+not available in Base SAS (such as resampling and custom offsets) -
+see the :ref:`timeseries documentation<timeseries>` for more details.
+
+.. include:: includes/time_date.rst
+
+Selection of columns
+~~~~~~~~~~~~~~~~~~~~
+
+SAS provides keywords in the ``DATA`` step to select,
+drop, and rename columns.
+
+.. code-block:: sas
+
+   data tips;
+       set tips;
+       keep sex total_bill tip;
+   run;
+
+   data tips;
+       set tips;
+       drop sex;
+   run;
+
+   data tips;
+       set tips;
+       rename total_bill=total_bill_2;
+   run;
+
+.. include:: includes/column_selection.rst
+
+
+Sorting by values
+~~~~~~~~~~~~~~~~~
+
+Sorting in SAS is accomplished via ``PROC SORT``
+
+.. code-block:: sas
+
+   proc sort data=tips;
+       by sex total_bill;
+   run;
+
+.. include:: includes/sorting.rst
+
+String processing
+-----------------
+
+Finding length of string
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+SAS determines the length of a character string with the
+`LENGTHN <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002284668.htm>`__
+and `LENGTHC <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002283942.htm>`__
+functions. ``LENGTHN`` excludes trailing blanks and ``LENGTHC`` includes trailing blanks.
+
+.. code-block:: sas
+
+   data _null_;
+   set tips;
+   put(LENGTHN(time));
+   put(LENGTHC(time));
+   run;
+
+.. include:: includes/length.rst
+
+
+Finding position of substring
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+SAS determines the position of a character in a string with the
+`FINDW <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a002978282.htm>`__ function.
+``FINDW`` takes the string defined by the first argument and searches for the first position of the substring
+you supply as the second argument.
+
+.. code-block:: sas
+
+   data _null_;
+   set tips;
+   put(FINDW(sex,'ale'));
+   run;
+
+.. include:: includes/find_substring.rst
+
+
+Extracting substring by position
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+SAS extracts a substring from a string based on its position with the
+`SUBSTR <https://support.sas.com/documentation/cdl/en/imlug/66845/HTML/default/viewer.htm#imlug_langref_sect455.htm>`__ function.
+
+.. code-block:: sas
+
+   data _null_;
+   set tips;
+   put(substr(sex,1,1));
+   run;
+
+.. include:: includes/extract_substring.rst
+
+
+Extracting nth word
+~~~~~~~~~~~~~~~~~~~
+
+The SAS `SCAN <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000214639.htm>`__
+function returns the nth word from a string. The first argument is the string you want to parse and the
+second argument specifies which word you want to extract.
+
+.. code-block:: sas
+
+   data firstlast;
+   input String $60.;
+   First_Name = scan(string, 1);
+   Last_Name = scan(string, -1);
+   datalines2;
+   John Smith;
+   Jane Cook;
+   ;;;
+   run;
+
+.. include:: includes/nth_word.rst
+
+
+Changing case
+~~~~~~~~~~~~~
+
+The SAS `UPCASE <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000245965.htm>`__
+`LOWCASE <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000245912.htm>`__ and
+`PROPCASE <https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/a002598106.htm>`__
+functions change the case of the argument.
+
+.. code-block:: sas
+
+   data firstlast;
+   input String $60.;
+   string_up = UPCASE(string);
+   string_low = LOWCASE(string);
+   string_prop = PROPCASE(string);
+   datalines2;
+   John Smith;
+   Jane Cook;
+   ;;;
+   run;
+
+.. include:: includes/case.rst
+
+
+Merging
+-------
+
+.. include:: includes/merge_setup.rst
+
+In SAS, data must be explicitly sorted before merging.  Different
+types of joins are accomplished using the ``in=`` dummy
+variables to track whether a match was found in one or both
+input frames.
+
+.. code-block:: sas
+
+   proc sort data=df1;
+       by key;
+   run;
+
+   proc sort data=df2;
+       by key;
+   run;
+
+   data left_join inner_join right_join outer_join;
+       merge df1(in=a) df2(in=b);
+
+       if a and b then output inner_join;
+       if a then output left_join;
+       if b then output right_join;
+       if a or b then output outer_join;
+   run;
+
+.. include:: includes/merge.rst
+
+
+Missing data
+------------
+
+Both pandas and SAS have a representation for missing data.
+
+.. include:: includes/missing_intro.rst
+
+One difference is that missing data cannot be compared to its sentinel value.
+For example, in SAS you could do this to filter missing values.
+
+.. code-block:: sas
+
+   data outer_join_nulls;
+       set outer_join;
+       if value_x = .;
+   run;
+
+   data outer_join_no_nulls;
+       set outer_join;
+       if value_x ^= .;
+   run;
+
+.. include:: includes/missing.rst
+
+
+GroupBy
+-------
+
+Aggregation
+~~~~~~~~~~~
+
+SAS's ``PROC SUMMARY`` can be used to group by one or
+more key variables and compute aggregations on
+numeric columns.
+
+.. code-block:: sas
+
+   proc summary data=tips nway;
+       class sex smoker;
+       var total_bill tip;
+       output out=tips_summed sum=;
+   run;
+
+.. include:: includes/groupby.rst
+
+
+Transformation
+~~~~~~~~~~~~~~
+
+In SAS, if the group aggregations need to be used with
+the original frame, it must be merged back together.  For
+example, to subtract the mean for each observation by smoker group.
+
+.. code-block:: sas
+
+   proc summary data=tips missing nway;
+       class smoker;
+       var total_bill;
+       output out=smoker_means mean(total_bill)=group_bill;
+   run;
+
+   proc sort data=tips;
+       by smoker;
+   run;
+
+   data tips;
+       merge tips(in=a) smoker_means(in=b);
+       by smoker;
+       adj_total_bill = total_bill - group_bill;
+       if a and b;
+   run;
+
+.. include:: includes/transform.rst
+
+
+By group processing
+~~~~~~~~~~~~~~~~~~~
+
+In addition to aggregation, pandas ``groupby`` can be used to
+replicate most other by group processing from SAS. For example,
+this ``DATA`` step reads the data by sex/smoker group and filters to
+the first entry for each.
+
+.. code-block:: sas
+
+   proc sort data=tips;
+      by sex smoker;
+   run;
+
+   data tips_first;
+       set tips;
+       by sex smoker;
+       if FIRST.sex or FIRST.smoker then output;
+   run;
+
+In pandas this would be written as:
+
+.. ipython:: python
+
+   tips.groupby(["sex", "smoker"]).first()
+
+
+Other considerations
+--------------------
+
+Disk vs memory
+~~~~~~~~~~~~~~
+
+pandas operates exclusively in memory, where a SAS data set exists on disk.
+This means that the size of data able to be loaded in pandas is limited by your
+machine's memory, but also that the operations on that data may be faster.
+
+If out of core processing is needed, one possibility is the
+`dask.dataframe <https://docs.dask.org/en/latest/dataframe.html>`_
+library (currently in development) which
+provides a subset of pandas functionality for an on-disk ``DataFrame``
+
+Data interop
+~~~~~~~~~~~~
+
+pandas provides a :func:`read_sas` method that can read SAS data saved in
+the XPORT or SAS7BDAT binary format.
+
+.. code-block:: sas
+
+   libname xportout xport 'transport-file.xpt';
+   data xportout.tips;
+       set tips(rename=(total_bill=tbill));
+       * xport variable names limited to 6 characters;
+   run;
+
+.. code-block:: python
+
+   df = pd.read_sas("transport-file.xpt")
+   df = pd.read_sas("binary-file.sas7bdat")
+
+You can also specify the file format directly. By default, pandas will try
+to infer the file format based on its extension.
+
+.. code-block:: python
+
+   df = pd.read_sas("transport-file.xpt", format="xport")
+   df = pd.read_sas("binary-file.sas7bdat", format="sas7bdat")
+
+XPORT is a relatively limited format and the parsing of it is not as
+optimized as some of the other pandas readers. An alternative way
+to interop data between SAS and pandas is to serialize to csv.
+
+.. code-block:: ipython
+
+   # version 0.17, 10M rows
+
+   In [8]: %time df = pd.read_sas('big.xpt')
+   Wall time: 14.6 s
+
+   In [9]: %time df = pd.read_csv('big.csv')
+   Wall time: 4.86 s
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_spreadsheets.rst b/pandas/doc/source/getting_started/comparison/comparison_with_spreadsheets.rst
new file mode 100644
index 0000000000000..d55b669d94a87
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_spreadsheets.rst
@@ -0,0 +1,465 @@
+.. _compare_with_spreadsheets:
+
+{{ header }}
+
+Comparison with spreadsheets
+****************************
+
+Since many potential pandas users have some familiarity with spreadsheet programs like
+`Excel <https://support.microsoft.com/en-us/excel>`_, this page is meant to provide some examples
+of how various spreadsheet operations would be performed using pandas. This page will use
+terminology and link to documentation for Excel, but much will be the same/similar in
+`Google Sheets <https://support.google.com/a/users/answer/9282959>`_,
+`LibreOffice Calc <https://help.libreoffice.org/latest/en-US/text/scalc/main0000.html?DbPAR=CALC>`_,
+`Apple Numbers <https://www.apple.com/numbers/compatibility/>`_, and other
+Excel-compatible spreadsheet software.
+
+.. include:: includes/introduction.rst
+
+Data structures
+---------------
+
+General terminology translation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. csv-table::
+    :header: "pandas", "Excel"
+    :widths: 20, 20
+
+    ``DataFrame``, worksheet
+    ``Series``, column
+    ``Index``, row headings
+    row, row
+    ``NaN``, empty cell
+
+``DataFrame``
+~~~~~~~~~~~~~
+
+A ``DataFrame`` in pandas is analogous to an Excel worksheet. While an Excel workbook can contain
+multiple worksheets, pandas ``DataFrame``\s exist independently.
+
+``Series``
+~~~~~~~~~~
+
+A ``Series`` is the data structure that represents one column of a ``DataFrame``. Working with a
+``Series`` is analogous to referencing a column of a spreadsheet.
+
+``Index``
+~~~~~~~~~
+
+Every ``DataFrame`` and ``Series`` has an ``Index``, which are labels on the *rows* of the data. In
+pandas, if no index is specified, a :class:`~pandas.RangeIndex` is used by default (first row = 0,
+second row = 1, and so on), analogous to row headings/numbers in spreadsheets.
+
+In pandas, indexes can be set to one (or multiple) unique values, which is like having a column that
+is used as the row identifier in a worksheet. Unlike most spreadsheets, these ``Index`` values can
+actually be used to reference the rows. (Note that `this can be done in Excel with structured
+references
+<https://support.microsoft.com/en-us/office/using-structured-references-with-excel-tables-f5ed2452-2337-4f71-bed3-c8ae6d2b276e>`_.)
+For example, in spreadsheets, you would reference the first row as ``A1:Z1``, while in pandas you
+could use ``populations.loc['Chicago']``.
+
+Index values are also persistent, so if you re-order the rows in a ``DataFrame``, the label for a
+particular row don't change.
+
+See the :ref:`indexing documentation<indexing>` for much more on how to use an ``Index``
+effectively.
+
+
+Copies vs. in place operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. include:: includes/copies.rst
+
+
+Data input / output
+-------------------
+
+Constructing a DataFrame from values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In a spreadsheet, `values can be typed directly into cells <https://support.microsoft.com/en-us/office/enter-data-manually-in-worksheet-cells-c798181d-d75a-41b1-92ad-6c0800f80038>`_.
+
+.. include:: includes/construct_dataframe.rst
+
+Reading external data
+~~~~~~~~~~~~~~~~~~~~~
+
+Both `Excel <https://support.microsoft.com/en-us/office/import-data-from-data-sources-power-query-be4330b3-5356-486c-a168-b68e9e616f5a>`__
+and :ref:`pandas <10min_tut_02_read_write>` can import data from various sources in various
+formats.
+
+CSV
+'''
+
+Let's load and display the `tips <https://github.com/pandas-dev/pandas/blob/main/pandas/tests/io/data/csv/tips.csv>`_
+dataset from the pandas tests, which is a CSV file. In Excel, you would download and then
+`open the CSV <https://support.microsoft.com/en-us/office/import-or-export-text-txt-or-csv-files-5250ac4c-663c-47ce-937b-339e391393ba>`_.
+In pandas, you pass the URL or local path of the CSV file to :func:`~pandas.read_csv`:
+
+.. ipython:: python
+
+   url = (
+       "https://raw.githubusercontent.com/pandas-dev"
+       "/pandas/main/pandas/tests/io/data/csv/tips.csv"
+   )
+   tips = pd.read_csv(url)
+   tips
+
+Like `Excel's Text Import Wizard <https://support.microsoft.com/en-us/office/text-import-wizard-c5b02af6-fda1-4440-899f-f78bafe41857>`_,
+``read_csv`` can take a number of parameters to specify how the data should be parsed. For
+example, if the data was instead tab delimited, and did not have column names, the pandas command
+would be:
+
+.. code-block:: python
+
+   tips = pd.read_csv("tips.csv", sep="\t", header=None)
+
+   # alternatively, read_table is an alias to read_csv with tab delimiter
+   tips = pd.read_table("tips.csv", header=None)
+
+Excel files
+'''''''''''
+
+Excel opens `various Excel file formats <https://support.microsoft.com/en-us/office/file-formats-that-are-supported-in-excel-0943ff2c-6014-4e8d-aaea-b83d51d46247>`_
+by double-clicking them, or using `the Open menu <https://support.microsoft.com/en-us/office/open-files-from-the-file-menu-97f087d8-3136-4485-8e86-c5b12a8c4176>`_.
+In pandas, you use :ref:`special methods for reading and writing from/to Excel files <io.excel>`.
+
+Let's first :ref:`create a new Excel file <io.excel_writer>` based on the ``tips`` dataframe in the above example:
+
+.. code-block:: python
+
+    tips.to_excel("./tips.xlsx")
+
+Should you wish to subsequently access the data in the ``tips.xlsx`` file, you can read it into your module using
+
+.. code-block:: python
+
+    tips_df = pd.read_excel("./tips.xlsx", index_col=0)
+
+You have just read in an Excel file using pandas!
+
+
+Limiting output
+~~~~~~~~~~~~~~~
+
+Spreadsheet programs will only show one screenful of data at a time and then allow you to scroll, so
+there isn't really a need to limit output. In pandas, you'll need to put a little more thought into
+controlling how your ``DataFrame``\s are displayed.
+
+.. include:: includes/limit.rst
+
+
+Exporting data
+~~~~~~~~~~~~~~
+
+By default, desktop spreadsheet software will save to its respective file format (``.xlsx``, ``.ods``, etc). You can, however, `save to other file formats <https://support.microsoft.com/en-us/office/save-a-workbook-in-another-file-format-6a16c862-4a36-48f9-a300-c2ca0065286e>`_.
+
+:ref:`pandas can create Excel files <io.excel_writer>`, :ref:`CSV <io.store_in_csv>`, or :ref:`a number of other formats <io>`.
+
+Data operations
+---------------
+
+Operations on columns
+~~~~~~~~~~~~~~~~~~~~~
+
+In spreadsheets, `formulas
+<https://support.microsoft.com/en-us/office/overview-of-formulas-in-excel-ecfdc708-9162-49e8-b993-c311f47ca173>`_
+are often created in individual cells and then `dragged
+<https://support.microsoft.com/en-us/office/copy-a-formula-by-dragging-the-fill-handle-in-excel-for-mac-dd928259-622b-473f-9a33-83aa1a63e218>`_
+into other cells to compute them for other columns. In pandas, you're able to do operations on whole
+columns directly.
+
+.. include:: includes/column_operations.rst
+
+Note that we aren't having to tell it to do that subtraction cell-by-cell — pandas handles that for
+us. See :ref:`how to create new columns derived from existing columns <10min_tut_05_columns>`.
+
+
+Filtering
+~~~~~~~~~
+
+`In Excel, filtering is done through a graphical menu. <https://support.microsoft.com/en-us/office/filter-data-in-a-range-or-table-01832226-31b5-4568-8806-38c37dcc180e>`_
+
+.. image:: ../../_static/spreadsheets/filter.png
+   :alt: Screenshot showing filtering of the total_bill column to values greater than 10
+   :align: center
+
+.. include:: includes/filtering.rst
+
+If/then logic
+~~~~~~~~~~~~~
+
+Let's say we want to make a ``bucket`` column with values of ``low`` and ``high``, based on whether
+the ``total_bill`` is less or more than $10.
+
+In spreadsheets, logical comparison can be done with `conditional formulas
+<https://support.microsoft.com/en-us/office/create-conditional-formulas-ca916c57-abd8-4b44-997c-c309b7307831>`_.
+We'd use a formula of ``=IF(A2 < 10, "low", "high")``, dragged to all cells in a new ``bucket``
+column.
+
+.. image:: ../../_static/spreadsheets/conditional.png
+   :alt: Screenshot showing the formula from above in a bucket column of the tips spreadsheet
+   :align: center
+
+.. include:: includes/if_then.rst
+
+Date functionality
+~~~~~~~~~~~~~~~~~~
+
+*This section will refer to "dates", but timestamps are handled similarly.*
+
+We can think of date functionality in two parts: parsing, and output. In spreadsheets, date values
+are generally parsed automatically, though there is a `DATEVALUE
+<https://support.microsoft.com/en-us/office/datevalue-function-df8b07d4-7761-4a93-bc33-b7471bbff252>`_
+function if you need it. In pandas, you need to explicitly convert plain text to datetime objects,
+either :ref:`while reading from a CSV <io.read_csv_table.datetime>` or :ref:`once in a DataFrame
+<10min_tut_09_timeseries.properties>`.
+
+Once parsed, spreadsheets display the dates in a default format, though `the format can be changed
+<https://support.microsoft.com/en-us/office/format-a-date-the-way-you-want-8e10019e-d5d8-47a1-ba95-db95123d273e>`_.
+In pandas, you'll generally want to keep dates as ``datetime`` objects while you're doing
+calculations with them. Outputting *parts* of dates (such as the year) is done through `date
+functions
+<https://support.microsoft.com/en-us/office/date-and-time-functions-reference-fd1b5961-c1ae-4677-be58-074152f97b81>`_
+in spreadsheets, and :ref:`datetime properties <10min_tut_09_timeseries.properties>` in pandas.
+
+Given ``date1`` and ``date2`` in columns ``A`` and ``B`` of a spreadsheet, you might have these
+formulas:
+
+.. list-table::
+    :header-rows: 1
+    :widths: auto
+
+    * - column
+      - formula
+    * - ``date1_year``
+      - ``=YEAR(A2)``
+    * - ``date2_month``
+      - ``=MONTH(B2)``
+    * - ``date1_next``
+      - ``=DATE(YEAR(A2),MONTH(A2)+1,1)``
+    * - ``months_between``
+      - ``=DATEDIF(A2,B2,"M")``
+
+The equivalent pandas operations are shown below.
+
+.. include:: includes/time_date.rst
+
+See :ref:`timeseries` for more details.
+
+
+Selection of columns
+~~~~~~~~~~~~~~~~~~~~
+
+In spreadsheets, you can select columns you want by:
+
+- `Hiding columns <https://support.microsoft.com/en-us/office/hide-or-show-rows-or-columns-659c2cad-802e-44ee-a614-dde8443579f8>`_
+- `Deleting columns <https://support.microsoft.com/en-us/office/insert-or-delete-rows-and-columns-6f40e6e4-85af-45e0-b39d-65dd504a3246>`_
+- `Referencing a range <https://support.microsoft.com/en-us/office/create-or-change-a-cell-reference-c7b8b95d-c594-4488-947e-c835903cebaa>`_ from one worksheet into another
+
+Since spreadsheet columns are typically `named in a header row
+<https://support.microsoft.com/en-us/office/turn-excel-table-headers-on-or-off-c91d1742-312c-4480-820f-cf4b534c8b3b>`_,
+renaming a column is simply a matter of changing the text in that first cell.
+
+.. include:: includes/column_selection.rst
+
+
+Sorting by values
+~~~~~~~~~~~~~~~~~
+
+Sorting in spreadsheets is accomplished via `the sort dialog <https://support.microsoft.com/en-us/office/sort-data-in-a-range-or-table-62d0b95d-2a90-4610-a6ae-2e545c4a4654>`_.
+
+.. image:: ../../_static/spreadsheets/sort.png
+   :alt: Screenshot of dialog from Excel showing sorting by the sex then total_bill columns
+   :align: center
+
+.. include:: includes/sorting.rst
+
+String processing
+-----------------
+
+Finding length of string
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+In spreadsheets, the number of characters in text can be found with the `LEN
+<https://support.microsoft.com/en-us/office/len-lenb-functions-29236f94-cedc-429d-affd-b5e33d2c67cb>`_
+function. This can be used with the `TRIM
+<https://support.microsoft.com/en-us/office/trim-function-410388fa-c5df-49c6-b16c-9e5630b479f9>`_
+function to remove extra whitespace.
+
+::
+
+   =LEN(TRIM(A2))
+
+.. include:: includes/length.rst
+
+Note this will still include multiple spaces within the string, so isn't 100% equivalent.
+
+
+Finding position of substring
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The `FIND
+<https://support.microsoft.com/en-us/office/find-findb-functions-c7912941-af2a-4bdf-a553-d0d89b0a0628>`_
+spreadsheet function returns the position of a substring, with the first character being ``1``.
+
+.. image:: ../../_static/spreadsheets/sort.png
+   :alt: Screenshot of FIND formula being used in Excel
+   :align: center
+
+.. include:: includes/find_substring.rst
+
+
+Extracting substring by position
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Spreadsheets have a `MID
+<https://support.microsoft.com/en-us/office/mid-midb-functions-d5f9e25c-d7d6-472e-b568-4ecb12433028>`_
+formula for extracting a substring from a given position. To get the first character::
+
+   =MID(A2,1,1)
+
+.. include:: includes/extract_substring.rst
+
+
+Extracting nth word
+~~~~~~~~~~~~~~~~~~~
+
+In Excel, you might use the `Text to Columns Wizard
+<https://support.microsoft.com/en-us/office/split-text-into-different-columns-with-the-convert-text-to-columns-wizard-30b14928-5550-41f5-97ca-7a3e9c363ed7>`_
+for splitting text and retrieving a specific column. (Note `it's possible to do so through a formula
+as well <https://exceljet.net/formula/extract-nth-word-from-text-string>`_.)
+
+.. include:: includes/nth_word.rst
+
+
+Changing case
+~~~~~~~~~~~~~
+
+Spreadsheets provide `UPPER, LOWER, and PROPER functions
+<https://support.microsoft.com/en-us/office/change-the-case-of-text-01481046-0fa7-4f3b-a693-496795a7a44d>`_
+for converting text to upper, lower, and title case, respectively.
+
+.. include:: includes/case.rst
+
+
+Merging
+-------
+
+.. include:: includes/merge_setup.rst
+
+In Excel, there are `merging of tables can be done through a VLOOKUP
+<https://support.microsoft.com/en-us/office/how-can-i-merge-two-or-more-tables-c80a9fce-c1ab-4425-bb96-497dd906d656>`_.
+
+.. image:: ../../_static/spreadsheets/vlookup.png
+   :alt: Screenshot showing a VLOOKUP formula between two tables in Excel, with some values being filled in and others with "#N/A"
+   :align: center
+
+.. include:: includes/merge.rst
+
+``merge`` has a number of advantages over ``VLOOKUP``:
+
+* The lookup value doesn't need to be the first column of the lookup table
+* If multiple rows are matched, there will be one row for each match, instead of just the first
+* It will include all columns from the lookup table, instead of just a single specified column
+* It supports :ref:`more complex join operations <merging.join>`
+
+
+Other considerations
+--------------------
+
+Fill Handle
+~~~~~~~~~~~
+
+Create a series of numbers following a set pattern in a certain set of cells. In
+a spreadsheet, this would be done by shift+drag after entering the first number or by
+entering the first two or three values and then dragging.
+
+This can be achieved by creating a series and assigning it to the desired cells.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"AAA": [1] * 8, "BBB": list(range(0, 8))})
+    df
+
+    series = list(range(1, 5))
+    series
+
+    df.loc[2:5, "AAA"] = series
+
+    df
+
+Drop Duplicates
+~~~~~~~~~~~~~~~
+
+Excel has built-in functionality for `removing duplicate values <https://support.microsoft.com/en-us/office/find-and-remove-duplicates-00e35bea-b46a-4d5d-b28e-66a552dc138d>`_.
+This is supported in pandas via :meth:`~DataFrame.drop_duplicates`.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {
+            "class": ["A", "A", "A", "B", "C", "D"],
+            "student_count": [42, 35, 42, 50, 47, 45],
+            "all_pass": ["Yes", "Yes", "Yes", "No", "No", "Yes"],
+        }
+    )
+
+    df.drop_duplicates()
+
+    df.drop_duplicates(["class", "student_count"])
+
+Pivot Tables
+~~~~~~~~~~~~
+
+`PivotTables <https://support.microsoft.com/en-us/office/create-a-pivottable-to-analyze-worksheet-data-a9a84538-bfe9-40a9-a8e9-f99134456576>`_
+from spreadsheets can be replicated in pandas through :ref:`reshaping`. Using the ``tips`` dataset again,
+let's find the average gratuity by size of the party and sex of the server.
+
+In Excel, we use the following configuration for the PivotTable:
+
+.. image:: ../../_static/spreadsheets/pivot.png
+   :alt: Screenshot showing a PivotTable in Excel, using sex as the column, size as the rows, then average tip as the values
+   :align: center
+
+The equivalent in pandas:
+
+.. ipython:: python
+
+    pd.pivot_table(
+        tips, values="tip", index=["size"], columns=["sex"], aggfunc=np.average
+    )
+
+
+Adding a row
+~~~~~~~~~~~~
+
+Assuming we are using a :class:`~pandas.RangeIndex` (numbered ``0``, ``1``, etc.), we can use :func:`concat` to add a row to the bottom of a ``DataFrame``.
+
+.. ipython:: python
+
+    df
+    new_row = pd.DataFrame([["E", 51, True]],
+                           columns=["class", "student_count", "all_pass"])
+    pd.concat([df, new_row])
+
+
+Find and Replace
+~~~~~~~~~~~~~~~~
+
+`Excel's Find dialog <https://support.microsoft.com/en-us/office/find-or-replace-text-and-numbers-on-a-worksheet-0e304ca5-ecef-4808-b90f-fdb42f892e90>`_
+takes you to cells that match, one by one. In pandas, this operation is generally done for an
+entire column or ``DataFrame`` at once through :ref:`conditional expressions <10min_tut_03_subset.rows_and_columns>`.
+
+.. ipython:: python
+
+    tips
+    tips == "Sun"
+    tips["day"].str.contains("S")
+
+pandas' :meth:`~DataFrame.replace` is comparable to Excel's ``Replace All``.
+
+.. ipython:: python
+
+    tips.replace("Thu", "Thursday")
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_spss.rst b/pandas/doc/source/getting_started/comparison/comparison_with_spss.rst
new file mode 100644
index 0000000000000..12c64bfd180a3
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_spss.rst
@@ -0,0 +1,229 @@
+.. _compare_with_spss:
+
+{{ header }}
+
+Comparison with SPSS
+********************
+For potential users coming from `SPSS <https://www.ibm.com/spss>`__, this page is meant to demonstrate
+how various SPSS operations would be performed using pandas.
+
+.. include:: includes/introduction.rst
+
+Data structures
+---------------
+
+General terminology translation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. csv-table::
+    :header: "pandas", "SPSS"
+    :widths: 20, 20
+
+    :class:`DataFrame`, data file
+    column, variable
+    row, case
+    groupby, split file
+    :class:`NaN`, system-missing
+
+:class:`DataFrame`
+~~~~~~~~~~~~~~~~~~
+
+A :class:`DataFrame` in pandas is analogous to an SPSS data file - a two-dimensional
+data source with labeled columns that can be of different types. As will be shown in this
+document, almost any operation that can be performed in SPSS can also be accomplished in pandas.
+
+:class:`Series`
+~~~~~~~~~~~~~~~
+
+A :class:`Series` is the data structure that represents one column of a :class:`DataFrame`. SPSS doesn't have a
+separate data structure for a single variable, but in general, working with a :class:`Series` is analogous
+to working with a variable in SPSS.
+
+:class:`Index`
+~~~~~~~~~~~~~~
+
+Every :class:`DataFrame` and :class:`Series` has an :class:`Index` -- labels on the *rows* of the data. SPSS does not
+have an exact analogue, as cases are simply numbered sequentially from 1. In pandas, if no index is
+specified, a :class:`RangeIndex` is used by default (first row = 0, second row = 1, and so on).
+
+While using a labeled :class:`Index` or :class:`MultiIndex` can enable sophisticated analyses and is ultimately an
+important part of pandas to understand, for this comparison we will essentially ignore the :class:`Index` and
+just treat the :class:`DataFrame` as a collection of columns. Please see the :ref:`indexing documentation<indexing>`
+for much more on how to use an :class:`Index` effectively.
+
+
+Copies vs. in place operations
+------------------------------
+
+.. include:: includes/copies.rst
+
+
+Data input / output
+-------------------
+
+Reading external data
+~~~~~~~~~~~~~~~~~~~~~
+
+Like SPSS, pandas provides utilities for reading in data from many formats. The ``tips`` dataset, found within
+the pandas tests (`csv <https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/csv/tips.csv>`_)
+will be used in many of the following examples.
+
+In SPSS, you would use File > Open > Data to import a CSV file:
+
+.. code-block:: text
+
+    FILE > OPEN > DATA
+    /TYPE=CSV
+    /FILE='tips.csv'
+    /DELIMITERS=","
+    /FIRSTCASE=2
+    /VARIABLES=col1 col2 col3.
+
+The pandas equivalent would use :func:`read_csv`:
+
+.. code-block:: python
+
+   url = (
+       "https://raw.githubusercontent.com/pandas-dev"
+       "/pandas/main/pandas/tests/io/data/csv/tips.csv"
+   )
+   tips = pd.read_csv(url)
+   tips
+
+Like SPSS's data import wizard, ``read_csv`` can take a number of parameters to specify how the data should be parsed.
+For example, if the data was instead tab delimited, and did not have column names, the pandas command would be:
+
+.. code-block:: python
+
+   tips = pd.read_csv("tips.csv", sep="\t", header=None)
+
+   # alternatively, read_table is an alias to read_csv with tab delimiter
+   tips = pd.read_table("tips.csv", header=None)
+
+
+Data operations
+---------------
+
+Filtering
+~~~~~~~~~
+
+In SPSS, filtering is done through Data > Select Cases:
+
+.. code-block:: text
+
+    SELECT IF (total_bill > 10).
+    EXECUTE.
+
+In pandas, boolean indexing can be used:
+
+.. code-block:: python
+
+    tips[tips["total_bill"] > 10]
+
+
+Sorting
+~~~~~~~
+
+In SPSS, sorting is done through Data > Sort Cases:
+
+.. code-block:: text
+
+    SORT CASES BY sex total_bill.
+    EXECUTE.
+
+In pandas, this would be written as:
+
+.. code-block:: python
+
+    tips.sort_values(["sex", "total_bill"])
+
+
+String processing
+-----------------
+
+Finding length of string
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+In SPSS:
+
+.. code-block:: text
+
+    COMPUTE length = LENGTH(time).
+    EXECUTE.
+
+.. include:: includes/length.rst
+
+
+Changing case
+~~~~~~~~~~~~~
+
+In SPSS:
+
+.. code-block:: text
+
+    COMPUTE upper = UPCASE(time).
+    COMPUTE lower = LOWER(time).
+    EXECUTE.
+
+.. include:: includes/case.rst
+
+
+Merging
+-------
+
+In SPSS, merging data files is done through Data > Merge Files.
+
+.. include:: includes/merge_setup.rst
+.. include:: includes/merge.rst
+
+
+GroupBy operations
+------------------
+
+Split-file processing
+~~~~~~~~~~~~~~~~~~~~~
+
+In SPSS, split-file analysis is done through Data > Split File:
+
+.. code-block:: text
+
+    SORT CASES BY sex.
+    SPLIT FILE BY sex.
+    DESCRIPTIVES VARIABLES=total_bill tip
+      /STATISTICS=MEAN STDDEV MIN MAX.
+
+The pandas equivalent would be:
+
+.. code-block:: python
+
+    tips.groupby("sex")[["total_bill", "tip"]].agg(["mean", "std", "min", "max"])
+
+
+Missing data
+------------
+
+SPSS uses the period (``.``) for numeric missing values and blank spaces for string missing values.
+pandas uses ``NaN`` (Not a Number) for numeric missing values and ``None`` or ``NaN`` for string
+missing values.
+
+.. include:: includes/missing.rst
+
+
+Other considerations
+--------------------
+
+Output management
+-----------------
+
+While pandas does not have a direct equivalent to SPSS's Output Management System (OMS), you can
+capture and export results in various ways:
+
+.. code-block:: python
+
+   # Save summary statistics to CSV
+   tips.groupby('sex')[['total_bill', 'tip']].mean().to_csv('summary.csv')
+
+   # Save multiple results to Excel sheets
+   with pd.ExcelWriter('results.xlsx') as writer:
+       tips.describe().to_excel(writer, sheet_name='Descriptives')
+       tips.groupby('sex').mean().to_excel(writer, sheet_name='Means by Gender')
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_sql.rst b/pandas/doc/source/getting_started/comparison/comparison_with_sql.rst
new file mode 100644
index 0000000000000..dc0590f18751a
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_sql.rst
@@ -0,0 +1,512 @@
+.. _compare_with_sql:
+
+{{ header }}
+
+Comparison with SQL
+********************
+Since many potential pandas users have some familiarity with
+`SQL <https://en.wikipedia.org/wiki/SQL>`_, this page is meant to provide some examples of how
+various SQL operations would be performed using pandas.
+
+.. include:: includes/introduction.rst
+
+Most of the examples will utilize the ``tips`` dataset found within pandas tests.  We'll read
+the data into a DataFrame called ``tips`` and assume we have a database table of the same name and
+structure.
+
+.. ipython:: python
+
+    url = (
+        "https://raw.githubusercontent.com/pandas-dev"
+        "/pandas/main/pandas/tests/io/data/csv/tips.csv"
+    )
+    tips = pd.read_csv(url)
+    tips
+
+
+Copies vs. in place operations
+------------------------------
+
+.. include:: includes/copies.rst
+
+
+SELECT
+------
+In SQL, selection is done using a comma-separated list of columns you'd like to select (or a ``*``
+to select all columns):
+
+.. code-block:: sql
+
+    SELECT total_bill, tip, smoker, time
+    FROM tips;
+
+With pandas, column selection is done by passing a list of column names to your DataFrame:
+
+.. ipython:: python
+
+    tips[["total_bill", "tip", "smoker", "time"]]
+
+Calling the DataFrame without the list of column names would display all columns (akin to SQL's
+``*``).
+
+In SQL, you can add a calculated column:
+
+.. code-block:: sql
+
+    SELECT *, tip/total_bill as tip_rate
+    FROM tips;
+
+With pandas, you can use the :meth:`DataFrame.assign` method of a DataFrame to append a new column:
+
+.. ipython:: python
+
+    tips.assign(tip_rate=tips["tip"] / tips["total_bill"])
+
+WHERE
+-----
+Filtering in SQL is done via a WHERE clause.
+
+.. code-block:: sql
+
+    SELECT *
+    FROM tips
+    WHERE time = 'Dinner';
+
+.. include:: includes/filtering.rst
+
+Just like SQL's ``OR`` and ``AND``, multiple conditions can be passed to a DataFrame using ``|``
+(``OR``) and ``&`` (``AND``).
+
+Tips of more than $5 at Dinner meals:
+
+.. code-block:: sql
+
+    SELECT *
+    FROM tips
+    WHERE time = 'Dinner' AND tip > 5.00;
+
+.. ipython:: python
+
+    tips[(tips["time"] == "Dinner") & (tips["tip"] > 5.00)]
+
+Tips by parties of at least 5 diners OR bill total was more than $45:
+
+.. code-block:: sql
+
+    SELECT *
+    FROM tips
+    WHERE size >= 5 OR total_bill > 45;
+
+.. ipython:: python
+
+    tips[(tips["size"] >= 5) | (tips["total_bill"] > 45)]
+
+NULL checking is done using the :meth:`~pandas.Series.notna` and :meth:`~pandas.Series.isna`
+methods.
+
+.. ipython:: python
+
+    frame = pd.DataFrame(
+        {"col1": ["A", "B", np.nan, "C", "D"], "col2": ["F", np.nan, "G", "H", "I"]}
+    )
+    frame
+
+Assume we have a table of the same structure as our DataFrame above. We can see only the records
+where ``col2`` IS NULL with the following query:
+
+.. code-block:: sql
+
+    SELECT *
+    FROM frame
+    WHERE col2 IS NULL;
+
+.. ipython:: python
+
+    frame[frame["col2"].isna()]
+
+Getting items where ``col1`` IS NOT NULL can be done with :meth:`~pandas.Series.notna`.
+
+.. code-block:: sql
+
+    SELECT *
+    FROM frame
+    WHERE col1 IS NOT NULL;
+
+.. ipython:: python
+
+    frame[frame["col1"].notna()]
+
+
+GROUP BY
+--------
+In pandas, SQL's ``GROUP BY`` operations are performed using the similarly named
+:meth:`~pandas.DataFrame.groupby` method. :meth:`~pandas.DataFrame.groupby` typically refers to a
+process where we'd like to split a dataset into groups, apply some function (typically aggregation)
+, and then combine the groups together.
+
+A common SQL operation would be getting the count of records in each group throughout a dataset.
+For instance, a query getting us the number of tips left by sex:
+
+.. code-block:: sql
+
+    SELECT sex, count(*)
+    FROM tips
+    GROUP BY sex;
+    /*
+    Female     87
+    Male      157
+    */
+
+
+The pandas equivalent would be:
+
+.. ipython:: python
+
+    tips.groupby("sex").size()
+
+Notice that in the pandas code we used :meth:`.DataFrameGroupBy.size` and not
+:meth:`.DataFrameGroupBy.count`. This is because
+:meth:`.DataFrameGroupBy.count` applies the function to each column, returning
+the number of ``NOT NULL`` records within each.
+
+.. ipython:: python
+
+    tips.groupby("sex").count()
+
+Alternatively, we could have applied the :meth:`.DataFrameGroupBy.count` method
+to an individual column:
+
+.. ipython:: python
+
+    tips.groupby("sex")["total_bill"].count()
+
+Multiple functions can also be applied at once. For instance, say we'd like to see how tip amount
+differs by day of the week - :meth:`.DataFrameGroupBy.agg` allows you to pass a dictionary
+to your grouped DataFrame, indicating which functions to apply to specific columns.
+
+.. code-block:: sql
+
+    SELECT day, AVG(tip), COUNT(*)
+    FROM tips
+    GROUP BY day;
+    /*
+    Fri   2.734737   19
+    Sat   2.993103   87
+    Sun   3.255132   76
+    Thu  2.771452   62
+    */
+
+.. ipython:: python
+
+    tips.groupby("day").agg({"tip": "mean", "day": "size"})
+
+Grouping by more than one column is done by passing a list of columns to the
+:meth:`~pandas.DataFrame.groupby` method.
+
+.. code-block:: sql
+
+    SELECT smoker, day, COUNT(*), AVG(tip)
+    FROM tips
+    GROUP BY smoker, day;
+    /*
+    smoker day
+    No     Fri      4  2.812500
+           Sat     45  3.102889
+           Sun     57  3.167895
+           Thu    45  2.673778
+    Yes    Fri     15  2.714000
+           Sat     42  2.875476
+           Sun     19  3.516842
+           Thu    17  3.030000
+    */
+
+.. ipython:: python
+
+    tips.groupby(["smoker", "day"]).agg({"tip": ["size", "mean"]})
+
+.. _compare_with_sql.join:
+
+JOIN
+----
+``JOIN``\s can be performed with :meth:`~pandas.DataFrame.join` or :meth:`~pandas.merge`. By
+default, :meth:`~pandas.DataFrame.join` will join the DataFrames on their indices. Each method has
+parameters allowing you to specify the type of join to perform (``LEFT``, ``RIGHT``, ``INNER``,
+``FULL``) or the columns to join on (column names or indices).
+
+.. warning::
+
+    If both key columns contain rows where the key is a null value, those
+    rows will be matched against each other. This is different from usual SQL
+    join behaviour and can lead to unexpected results.
+
+.. ipython:: python
+
+    df1 = pd.DataFrame({"key": ["A", "B", "C", "D"], "value": np.random.randn(4)})
+    df2 = pd.DataFrame({"key": ["B", "D", "D", "E"], "value": np.random.randn(4)})
+
+Assume we have two database tables of the same name and structure as our DataFrames.
+
+Now let's go over the various types of ``JOIN``\s.
+
+INNER JOIN
+~~~~~~~~~~
+.. code-block:: sql
+
+    SELECT *
+    FROM df1
+    INNER JOIN df2
+      ON df1.key = df2.key;
+
+.. ipython:: python
+
+    # merge performs an INNER JOIN by default
+    pd.merge(df1, df2, on="key")
+
+:meth:`~pandas.merge` also offers parameters for cases when you'd like to join one DataFrame's
+column with another DataFrame's index.
+
+.. ipython:: python
+
+    indexed_df2 = df2.set_index("key")
+    pd.merge(df1, indexed_df2, left_on="key", right_index=True)
+
+LEFT OUTER JOIN
+~~~~~~~~~~~~~~~
+
+Show all records from ``df1``.
+
+.. code-block:: sql
+
+    SELECT *
+    FROM df1
+    LEFT OUTER JOIN df2
+      ON df1.key = df2.key;
+
+.. ipython:: python
+
+    pd.merge(df1, df2, on="key", how="left")
+
+RIGHT JOIN
+~~~~~~~~~~
+
+Show all records from ``df2``.
+
+.. code-block:: sql
+
+    SELECT *
+    FROM df1
+    RIGHT OUTER JOIN df2
+      ON df1.key = df2.key;
+
+.. ipython:: python
+
+    pd.merge(df1, df2, on="key", how="right")
+
+FULL JOIN
+~~~~~~~~~
+pandas also allows for ``FULL JOIN``\s, which display both sides of the dataset, whether or not the
+joined columns find a match. As of writing, ``FULL JOIN``\s are not supported in all RDBMS (MySQL).
+
+Show all records from both tables.
+
+.. code-block:: sql
+
+    SELECT *
+    FROM df1
+    FULL OUTER JOIN df2
+      ON df1.key = df2.key;
+
+.. ipython:: python
+
+    pd.merge(df1, df2, on="key", how="outer")
+
+
+UNION
+-----
+
+``UNION ALL`` can be performed using :meth:`~pandas.concat`.
+
+.. ipython:: python
+
+    df1 = pd.DataFrame(
+        {"city": ["Chicago", "San Francisco", "New York City"], "rank": range(1, 4)}
+    )
+    df2 = pd.DataFrame(
+        {"city": ["Chicago", "Boston", "Los Angeles"], "rank": [1, 4, 5]}
+    )
+
+.. code-block:: sql
+
+    SELECT city, rank
+    FROM df1
+    UNION ALL
+    SELECT city, rank
+    FROM df2;
+    /*
+             city  rank
+          Chicago     1
+    San Francisco     2
+    New York City     3
+          Chicago     1
+           Boston     4
+      Los Angeles     5
+    */
+
+.. ipython:: python
+
+    pd.concat([df1, df2])
+
+SQL's ``UNION`` is similar to ``UNION ALL``, however ``UNION`` will remove duplicate rows.
+
+.. code-block:: sql
+
+    SELECT city, rank
+    FROM df1
+    UNION
+    SELECT city, rank
+    FROM df2;
+    -- notice that there is only one Chicago record this time
+    /*
+             city  rank
+          Chicago     1
+    San Francisco     2
+    New York City     3
+           Boston     4
+      Los Angeles     5
+    */
+
+In pandas, you can use :meth:`~pandas.concat` in conjunction with
+:meth:`~pandas.DataFrame.drop_duplicates`.
+
+.. ipython:: python
+
+    pd.concat([df1, df2]).drop_duplicates()
+
+
+LIMIT
+-----
+
+.. code-block:: sql
+
+    SELECT * FROM tips
+    LIMIT 10;
+
+.. ipython:: python
+
+    tips.head(10)
+
+
+pandas equivalents for some SQL analytic and aggregate functions
+----------------------------------------------------------------
+
+Top n rows with offset
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: sql
+
+    -- MySQL
+    SELECT * FROM tips
+    ORDER BY tip DESC
+    LIMIT 10 OFFSET 5;
+
+.. ipython:: python
+
+    tips.nlargest(10 + 5, columns="tip").tail(10)
+
+Top n rows per group
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: sql
+
+    -- Oracle's ROW_NUMBER() analytic function
+    SELECT * FROM (
+      SELECT
+        t.*,
+        ROW_NUMBER() OVER(PARTITION BY day ORDER BY total_bill DESC) AS rn
+      FROM tips t
+    )
+    WHERE rn < 3
+    ORDER BY day, rn;
+
+
+.. ipython:: python
+
+    (
+        tips.assign(
+            rn=tips.sort_values(["total_bill"], ascending=False)
+            .groupby(["day"])
+            .cumcount()
+            + 1
+        )
+        .query("rn < 3")
+        .sort_values(["day", "rn"])
+    )
+
+the same using ``rank(method='first')`` function
+
+.. ipython:: python
+
+    (
+        tips.assign(
+            rnk=tips.groupby(["day"])["total_bill"].rank(
+                method="first", ascending=False
+            )
+        )
+        .query("rnk < 3")
+        .sort_values(["day", "rnk"])
+    )
+
+.. code-block:: sql
+
+    -- Oracle's RANK() analytic function
+    SELECT * FROM (
+      SELECT
+        t.*,
+        RANK() OVER(PARTITION BY sex ORDER BY tip) AS rnk
+      FROM tips t
+      WHERE tip < 2
+    )
+    WHERE rnk < 3
+    ORDER BY sex, rnk;
+
+Let's find tips with (rank < 3) per gender group for (tips < 2).
+Notice that when using ``rank(method='min')`` function
+``rnk_min`` remains the same for the same ``tip``
+(as Oracle's ``RANK()`` function)
+
+.. ipython:: python
+
+    (
+        tips[tips["tip"] < 2]
+        .assign(rnk_min=tips.groupby(["sex"])["tip"].rank(method="min"))
+        .query("rnk_min < 3")
+        .sort_values(["sex", "rnk_min"])
+    )
+
+
+UPDATE
+------
+
+.. code-block:: sql
+
+    UPDATE tips
+    SET tip = tip*2
+    WHERE tip < 2;
+
+.. ipython:: python
+
+    tips.loc[tips["tip"] < 2, "tip"] *= 2
+
+DELETE
+------
+
+.. code-block:: sql
+
+    DELETE FROM tips
+    WHERE tip > 9;
+
+In pandas we select the rows that should remain instead of deleting the rows that should be removed:
+
+.. ipython:: python
+
+    tips = tips.loc[tips["tip"] <= 9]
diff --git a/pandas/doc/source/getting_started/comparison/comparison_with_stata.rst b/pandas/doc/source/getting_started/comparison/comparison_with_stata.rst
new file mode 100644
index 0000000000000..b4b0c42d1db1d
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/comparison_with_stata.rst
@@ -0,0 +1,501 @@
+.. _compare_with_stata:
+
+{{ header }}
+
+Comparison with Stata
+*********************
+For potential users coming from `Stata <https://en.wikipedia.org/wiki/Stata>`__
+this page is meant to demonstrate how different Stata operations would be
+performed in pandas.
+
+.. include:: includes/introduction.rst
+
+
+Data structures
+---------------
+
+General terminology translation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. csv-table::
+    :header: "pandas", "Stata"
+    :widths: 20, 20
+
+    ``DataFrame``, data set
+    column, variable
+    row, observation
+    groupby, bysort
+    ``NaN``, ``.``
+
+
+``DataFrame``
+~~~~~~~~~~~~~
+
+A ``DataFrame`` in pandas is analogous to a Stata data set -- a two-dimensional
+data source with labeled columns that can be of different types. As will be
+shown in this document, almost any operation that can be applied to a data set
+in Stata can also be accomplished in pandas.
+
+``Series``
+~~~~~~~~~~
+
+A ``Series`` is the data structure that represents one column of a
+``DataFrame``. Stata doesn't have a separate data structure for a single column,
+but in general, working with a ``Series`` is analogous to referencing a column
+of a data set in Stata.
+
+``Index``
+~~~~~~~~~
+
+Every ``DataFrame`` and ``Series`` has an ``Index`` -- labels on the
+*rows* of the data. Stata does not have an exactly analogous concept. In Stata, a data set's
+rows are essentially unlabeled, other than an implicit integer index that can be
+accessed with ``_n``.
+
+In pandas, if no index is specified, an integer index is also used by default
+(first row = 0, second row = 1, and so on). While using a labeled ``Index`` or
+``MultiIndex`` can enable sophisticated analyses and is ultimately an important
+part of pandas to understand, for this comparison we will essentially ignore the
+``Index`` and just treat the ``DataFrame`` as a collection of columns. Please
+see the :ref:`indexing documentation<indexing>` for much more on how to use an
+``Index`` effectively.
+
+
+Copies vs. in place operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. include:: includes/copies.rst
+
+
+Data input / output
+-------------------
+
+Constructing a DataFrame from values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A Stata data set can be built from specified values by
+placing the data after an ``input`` statement and
+specifying the column names.
+
+.. code-block:: stata
+
+   input x y
+   1 2
+   3 4
+   5 6
+   end
+
+.. include:: includes/construct_dataframe.rst
+
+Reading external data
+~~~~~~~~~~~~~~~~~~~~~
+
+Like Stata, pandas provides utilities for reading in data from
+many formats.  The ``tips`` data set, found within the pandas
+tests (`csv <https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/csv/tips.csv>`_)
+will be used in many of the following examples.
+
+Stata provides ``import delimited`` to read csv data into a data set in memory.
+If the ``tips.csv`` file is in the current working directory, we can import it as follows.
+
+.. code-block:: stata
+
+   import delimited tips.csv
+
+The pandas method is :func:`read_csv`, which works similarly. Additionally, it will automatically download
+the data set if presented with a url.
+
+.. ipython:: python
+
+   url = (
+       "https://raw.githubusercontent.com/pandas-dev"
+       "/pandas/main/pandas/tests/io/data/csv/tips.csv"
+   )
+   tips = pd.read_csv(url)
+   tips
+
+Like ``import delimited``, :func:`read_csv` can take a number of parameters to specify
+how the data should be parsed.  For example, if the data were instead tab delimited,
+did not have column names, and existed in the current working directory,
+the pandas command would be:
+
+.. code-block:: python
+
+   tips = pd.read_csv("tips.csv", sep="\t", header=None)
+
+   # alternatively, read_table is an alias to read_csv with tab delimiter
+   tips = pd.read_table("tips.csv", header=None)
+
+pandas can also read Stata data sets in ``.dta`` format with the :func:`read_stata` function.
+
+.. code-block:: python
+
+   df = pd.read_stata("data.dta")
+
+In addition to text/csv and Stata files, pandas supports a variety of other data formats
+such as Excel, SAS, HDF5, Parquet, and SQL databases.  These are all read via a ``pd.read_*``
+function.  See the :ref:`IO documentation<io>` for more details.
+
+
+Limiting output
+~~~~~~~~~~~~~~~
+
+.. include:: includes/limit.rst
+
+The equivalent in Stata would be:
+
+.. code-block:: stata
+
+   list in 1/5
+
+
+Exporting data
+~~~~~~~~~~~~~~
+
+The inverse of ``import delimited`` in Stata is ``export delimited``
+
+.. code-block:: stata
+
+   export delimited tips2.csv
+
+Similarly in pandas, the opposite of ``read_csv`` is :meth:`DataFrame.to_csv`.
+
+.. code-block:: python
+
+   tips.to_csv("tips2.csv")
+
+pandas can also export to Stata file format with the :meth:`DataFrame.to_stata` method.
+
+.. code-block:: python
+
+   tips.to_stata("tips2.dta")
+
+
+Data operations
+---------------
+
+Operations on columns
+~~~~~~~~~~~~~~~~~~~~~
+
+In Stata, arbitrary math expressions can be used with the ``generate`` and
+``replace`` commands on new or existing columns. The ``drop`` command drops
+the column from the data set.
+
+.. code-block:: stata
+
+   replace total_bill = total_bill - 2
+   generate new_bill = total_bill / 2
+   drop new_bill
+
+.. include:: includes/column_operations.rst
+
+
+Filtering
+~~~~~~~~~
+
+Filtering in Stata is done with an ``if`` clause on one or more columns.
+
+.. code-block:: stata
+
+   list if total_bill > 10
+
+.. include:: includes/filtering.rst
+
+If/then logic
+~~~~~~~~~~~~~
+
+In Stata, an ``if`` clause can also be used to create new columns.
+
+.. code-block:: stata
+
+   generate bucket = "low" if total_bill < 10
+   replace bucket = "high" if total_bill >= 10
+
+.. include:: includes/if_then.rst
+
+Date functionality
+~~~~~~~~~~~~~~~~~~
+
+Stata provides a variety of functions to do operations on
+date/datetime columns.
+
+.. code-block:: stata
+
+   generate date1 = mdy(1, 15, 2013)
+   generate date2 = date("Feb152015", "MDY")
+
+   generate date1_year = year(date1)
+   generate date2_month = month(date2)
+
+   * shift date to beginning of next month
+   generate date1_next = mdy(month(date1) + 1, 1, year(date1)) if month(date1) != 12
+   replace date1_next = mdy(1, 1, year(date1) + 1) if month(date1) == 12
+   generate months_between = mofd(date2) - mofd(date1)
+
+   list date1 date2 date1_year date2_month date1_next months_between
+
+The equivalent pandas operations are shown below.  In addition to these
+functions, pandas supports other Time Series features
+not available in Stata (such as time zone handling and custom offsets) --
+see the :ref:`timeseries documentation<timeseries>` for more details.
+
+.. include:: includes/time_date.rst
+
+Selection of columns
+~~~~~~~~~~~~~~~~~~~~
+
+Stata provides keywords to select, drop, and rename columns.
+
+.. code-block:: stata
+
+   keep sex total_bill tip
+
+   drop sex
+
+   rename total_bill total_bill_2
+
+.. include:: includes/column_selection.rst
+
+
+Sorting by values
+~~~~~~~~~~~~~~~~~
+
+Sorting in Stata is accomplished via ``sort``
+
+.. code-block:: stata
+
+   sort sex total_bill
+
+.. include:: includes/sorting.rst
+
+String processing
+-----------------
+
+Finding length of string
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Stata determines the length of a character string with the :func:`strlen` and
+:func:`ustrlen` functions for ASCII and Unicode strings, respectively.
+
+.. code-block:: stata
+
+   generate strlen_time = strlen(time)
+   generate ustrlen_time = ustrlen(time)
+
+.. include:: includes/length.rst
+
+
+Finding position of substring
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Stata determines the position of a character in a string with the :func:`strpos` function.
+This takes the string defined by the first argument and searches for the
+first position of the substring you supply as the second argument.
+
+.. code-block:: stata
+
+   generate str_position = strpos(sex, "ale")
+
+.. include:: includes/find_substring.rst
+
+
+Extracting substring by position
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Stata extracts a substring from a string based on its position with the :func:`substr` function.
+
+.. code-block:: stata
+
+   generate short_sex = substr(sex, 1, 1)
+
+.. include:: includes/extract_substring.rst
+
+
+Extracting nth word
+~~~~~~~~~~~~~~~~~~~
+
+The Stata :func:`word` function returns the nth word from a string.
+The first argument is the string you want to parse and the
+second argument specifies which word you want to extract.
+
+.. code-block:: stata
+
+   clear
+   input str20 string
+   "John Smith"
+   "Jane Cook"
+   end
+
+   generate first_name = word(name, 1)
+   generate last_name = word(name, -1)
+
+.. include:: includes/nth_word.rst
+
+
+Changing case
+~~~~~~~~~~~~~
+
+The Stata :func:`strupper`, :func:`strlower`, :func:`strproper`,
+:func:`ustrupper`, :func:`ustrlower`, and :func:`ustrtitle` functions
+change the case of ASCII and Unicode strings, respectively.
+
+.. code-block:: stata
+
+   clear
+   input str20 string
+   "John Smith"
+   "Jane Cook"
+   end
+
+   generate upper = strupper(string)
+   generate lower = strlower(string)
+   generate title = strproper(string)
+   list
+
+.. include:: includes/case.rst
+
+
+Merging
+-------
+
+.. include:: includes/merge_setup.rst
+
+In Stata, to perform a merge, one data set must be in memory
+and the other must be referenced as a file name on disk. In
+contrast, Python must have both ``DataFrames`` already in memory.
+
+By default, Stata performs an outer join, where all observations
+from both data sets are left in memory after the merge. One can
+keep only observations from the initial data set, the merged data set,
+or the intersection of the two by using the values created in the
+``_merge`` variable.
+
+.. code-block:: stata
+
+   * First create df2 and save to disk
+   clear
+   input str1 key
+   B
+   D
+   D
+   E
+   end
+   generate value = rnormal()
+   save df2.dta
+
+   * Now create df1 in memory
+   clear
+   input str1 key
+   A
+   B
+   C
+   D
+   end
+   generate value = rnormal()
+
+   preserve
+
+   * Left join
+   merge 1:n key using df2.dta
+   keep if _merge == 1
+
+   * Right join
+   restore, preserve
+   merge 1:n key using df2.dta
+   keep if _merge == 2
+
+   * Inner join
+   restore, preserve
+   merge 1:n key using df2.dta
+   keep if _merge == 3
+
+   * Outer join
+   restore
+   merge 1:n key using df2.dta
+
+.. include:: includes/merge.rst
+
+
+Missing data
+------------
+
+Both pandas and Stata have a representation for missing data.
+
+.. include:: includes/missing_intro.rst
+
+One difference is that missing data cannot be compared to its sentinel value.
+For example, in Stata you could do this to filter missing values.
+
+.. code-block:: stata
+
+   * Keep missing values
+   list if value_x == .
+   * Keep non-missing values
+   list if value_x != .
+
+.. include:: includes/missing.rst
+
+
+GroupBy
+-------
+
+Aggregation
+~~~~~~~~~~~
+
+Stata's ``collapse`` can be used to group by one or
+more key variables and compute aggregations on
+numeric columns.
+
+.. code-block:: stata
+
+   collapse (sum) total_bill tip, by(sex smoker)
+
+.. include:: includes/groupby.rst
+
+
+Transformation
+~~~~~~~~~~~~~~
+
+In Stata, if the group aggregations need to be used with the
+original data set, one would usually use ``bysort`` with :func:`egen`.
+For example, to subtract the mean for each observation by smoker group.
+
+.. code-block:: stata
+
+   bysort sex smoker: egen group_bill = mean(total_bill)
+   generate adj_total_bill = total_bill - group_bill
+
+.. include:: includes/transform.rst
+
+
+By group processing
+~~~~~~~~~~~~~~~~~~~
+
+In addition to aggregation, pandas ``groupby`` can be used to
+replicate most other ``bysort`` processing from Stata. For example,
+the following example lists the first observation in the current
+sort order by sex/smoker group.
+
+.. code-block:: stata
+
+   bysort sex smoker: list if _n == 1
+
+In pandas this would be written as:
+
+.. ipython:: python
+
+   tips.groupby(["sex", "smoker"]).first()
+
+
+Other considerations
+--------------------
+
+Disk vs memory
+~~~~~~~~~~~~~~
+
+pandas and Stata both operate exclusively in memory. This means that the size of
+data able to be loaded in pandas is limited by your machine's memory.
+If out of core processing is needed, one possibility is the
+`dask.dataframe <https://docs.dask.org/en/latest/dataframe.html>`_
+library, which provides a subset of pandas functionality for an
+on-disk ``DataFrame``.
diff --git a/pandas/doc/source/getting_started/comparison/includes/case.rst b/pandas/doc/source/getting_started/comparison/includes/case.rst
new file mode 100644
index 0000000000000..c00a830bc8511
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/case.rst
@@ -0,0 +1,10 @@
+The equivalent pandas methods are :meth:`Series.str.upper`, :meth:`Series.str.lower`, and
+:meth:`Series.str.title`.
+
+.. ipython:: python
+
+   firstlast = pd.DataFrame({"string": ["John Smith", "Jane Cook"]})
+   firstlast["upper"] = firstlast["string"].str.upper()
+   firstlast["lower"] = firstlast["string"].str.lower()
+   firstlast["title"] = firstlast["string"].str.title()
+   firstlast
diff --git a/pandas/doc/source/getting_started/comparison/includes/column_operations.rst b/pandas/doc/source/getting_started/comparison/includes/column_operations.rst
new file mode 100644
index 0000000000000..b23b931ed2db1
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/column_operations.rst
@@ -0,0 +1,11 @@
+pandas provides vectorized operations by specifying the individual ``Series`` in the
+``DataFrame``. New columns can be assigned in the same way. The :meth:`DataFrame.drop` method drops
+a column from the ``DataFrame``.
+
+.. ipython:: python
+
+   tips["total_bill"] = tips["total_bill"] - 2
+   tips["new_bill"] = tips["total_bill"] / 2
+   tips
+
+   tips = tips.drop("new_bill", axis=1)
diff --git a/pandas/doc/source/getting_started/comparison/includes/column_selection.rst b/pandas/doc/source/getting_started/comparison/includes/column_selection.rst
new file mode 100644
index 0000000000000..071645c9718cb
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/column_selection.rst
@@ -0,0 +1,22 @@
+The same operations are expressed in pandas below.
+
+Keep certain columns
+''''''''''''''''''''
+
+.. ipython:: python
+
+   tips[["sex", "total_bill", "tip"]]
+
+Drop a column
+'''''''''''''
+
+.. ipython:: python
+
+   tips.drop("sex", axis=1)
+
+Rename a column
+'''''''''''''''
+
+.. ipython:: python
+
+   tips.rename(columns={"total_bill": "total_bill_2"})
diff --git a/pandas/doc/source/getting_started/comparison/includes/construct_dataframe.rst b/pandas/doc/source/getting_started/comparison/includes/construct_dataframe.rst
new file mode 100644
index 0000000000000..4d066c7962d98
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/construct_dataframe.rst
@@ -0,0 +1,9 @@
+A pandas ``DataFrame`` can be constructed in many different ways,
+but for a small number of values, it is often convenient to specify it as
+a Python dictionary, where the keys are the column names
+and the values are the data.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"x": [1, 3, 5], "y": [2, 4, 6]})
+   df
diff --git a/pandas/doc/source/getting_started/comparison/includes/copies.rst b/pandas/doc/source/getting_started/comparison/includes/copies.rst
new file mode 100644
index 0000000000000..4f49c3a1a762e
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/copies.rst
@@ -0,0 +1,28 @@
+Most pandas operations return copies of the ``Series``/``DataFrame``. To make the changes "stick",
+you'll need to either assign to a new variable:
+
+   .. code-block:: python
+
+      sorted_df = df.sort_values("col1")
+
+
+or overwrite the original one:
+
+   .. code-block:: python
+
+      df = df.sort_values("col1")
+
+.. note::
+
+   You will see an ``inplace=True`` or ``copy=False`` keyword argument available for
+   some methods:
+
+   .. code-block:: python
+
+      df.replace(5, inplace=True)
+
+   There is an active discussion about deprecating and removing ``inplace`` and ``copy`` for
+   most methods (e.g. ``dropna``) except for a very small subset of methods
+   (including ``replace``). Both keywords won't be
+   necessary anymore in the context of Copy-on-Write. The proposal can be found
+   `here <https://github.com/pandas-dev/pandas/pull/51466>`_.
diff --git a/pandas/doc/source/getting_started/comparison/includes/extract_substring.rst b/pandas/doc/source/getting_started/comparison/includes/extract_substring.rst
new file mode 100644
index 0000000000000..1ba0dfac2317a
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/extract_substring.rst
@@ -0,0 +1,7 @@
+With pandas you can use ``[]`` notation to extract a substring
+from a string by position locations. Keep in mind that Python
+indexes are zero-based.
+
+.. ipython:: python
+
+   tips["sex"].str[0:1]
diff --git a/pandas/doc/source/getting_started/comparison/includes/filtering.rst b/pandas/doc/source/getting_started/comparison/includes/filtering.rst
new file mode 100644
index 0000000000000..8ddf7c0d2fa39
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/filtering.rst
@@ -0,0 +1,16 @@
+DataFrames can be filtered in multiple ways; the most intuitive of which is using
+:ref:`boolean indexing <indexing.boolean>`.
+
+.. ipython:: python
+
+   tips[tips["total_bill"] > 10]
+
+The above statement is simply passing a ``Series`` of ``True``/``False`` objects to the DataFrame,
+returning all rows with ``True``.
+
+.. ipython:: python
+
+    is_dinner = tips["time"] == "Dinner"
+    is_dinner
+    is_dinner.value_counts()
+    tips[is_dinner]
diff --git a/pandas/doc/source/getting_started/comparison/includes/find_substring.rst b/pandas/doc/source/getting_started/comparison/includes/find_substring.rst
new file mode 100644
index 0000000000000..42543d05a0014
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/find_substring.rst
@@ -0,0 +1,8 @@
+You can find the position of a character in a column of strings with the :meth:`Series.str.find`
+method. ``find`` searches for the first position of the substring. If the substring is found, the
+method returns its position. If not found, it returns ``-1``. Keep in mind that Python indexes are
+zero-based.
+
+.. ipython:: python
+
+   tips["sex"].str.find("ale")
diff --git a/pandas/doc/source/getting_started/comparison/includes/groupby.rst b/pandas/doc/source/getting_started/comparison/includes/groupby.rst
new file mode 100644
index 0000000000000..93d5d51e3fb00
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/groupby.rst
@@ -0,0 +1,7 @@
+pandas provides a flexible ``groupby`` mechanism that allows similar aggregations. See the
+:ref:`groupby documentation<groupby>` for more details and examples.
+
+.. ipython:: python
+
+   tips_summed = tips.groupby(["sex", "smoker"])[["total_bill", "tip"]].sum()
+   tips_summed
diff --git a/pandas/doc/source/getting_started/comparison/includes/if_then.rst b/pandas/doc/source/getting_started/comparison/includes/if_then.rst
new file mode 100644
index 0000000000000..f94e7588827f5
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/if_then.rst
@@ -0,0 +1,12 @@
+The same operation in pandas can be accomplished using
+the ``where`` method from ``numpy``.
+
+.. ipython:: python
+
+   tips["bucket"] = np.where(tips["total_bill"] < 10, "low", "high")
+   tips
+
+.. ipython:: python
+   :suppress:
+
+   tips = tips.drop("bucket", axis=1)
diff --git a/pandas/doc/source/getting_started/comparison/includes/introduction.rst b/pandas/doc/source/getting_started/comparison/includes/introduction.rst
new file mode 100644
index 0000000000000..aedf2875dc452
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/introduction.rst
@@ -0,0 +1,9 @@
+If you're new to pandas, you might want to first read through :ref:`10 Minutes to pandas<10min>`
+to familiarize yourself with the library.
+
+As is customary, we import pandas and NumPy as follows:
+
+.. ipython:: python
+
+    import pandas as pd
+    import numpy as np
diff --git a/pandas/doc/source/getting_started/comparison/includes/length.rst b/pandas/doc/source/getting_started/comparison/includes/length.rst
new file mode 100644
index 0000000000000..9141fd4ea582a
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/length.rst
@@ -0,0 +1,8 @@
+You can find the length of a character string with :meth:`Series.str.len`.
+In Python 3, all strings are Unicode strings. ``len`` includes trailing blanks.
+Use ``len`` and ``rstrip`` to exclude trailing blanks.
+
+.. ipython:: python
+
+   tips["time"].str.len()
+   tips["time"].str.rstrip().str.len()
diff --git a/pandas/doc/source/getting_started/comparison/includes/limit.rst b/pandas/doc/source/getting_started/comparison/includes/limit.rst
new file mode 100644
index 0000000000000..4efeb4e43d07c
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/limit.rst
@@ -0,0 +1,7 @@
+By default, pandas will truncate output of large ``DataFrame``\s to show the first and last rows.
+This can be overridden by :ref:`changing the pandas options <options>`, or using
+:meth:`DataFrame.head` or :meth:`DataFrame.tail`.
+
+.. ipython:: python
+
+   tips.head(5)
diff --git a/pandas/doc/source/getting_started/comparison/includes/merge.rst b/pandas/doc/source/getting_started/comparison/includes/merge.rst
new file mode 100644
index 0000000000000..b8e3f54fd132b
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/merge.rst
@@ -0,0 +1,17 @@
+pandas DataFrames have a :meth:`~DataFrame.merge` method, which provides similar functionality. The
+data does not have to be sorted ahead of time, and different join types are accomplished via the
+``how`` keyword.
+
+.. ipython:: python
+
+   inner_join = df1.merge(df2, on=["key"], how="inner")
+   inner_join
+
+   left_join = df1.merge(df2, on=["key"], how="left")
+   left_join
+
+   right_join = df1.merge(df2, on=["key"], how="right")
+   right_join
+
+   outer_join = df1.merge(df2, on=["key"], how="outer")
+   outer_join
diff --git a/pandas/doc/source/getting_started/comparison/includes/merge_setup.rst b/pandas/doc/source/getting_started/comparison/includes/merge_setup.rst
new file mode 100644
index 0000000000000..f115cd58f7a94
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/merge_setup.rst
@@ -0,0 +1,8 @@
+The following tables will be used in the merge examples:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"key": ["A", "B", "C", "D"], "value": np.random.randn(4)})
+   df1
+   df2 = pd.DataFrame({"key": ["B", "D", "D", "E"], "value": np.random.randn(4)})
+   df2
diff --git a/pandas/doc/source/getting_started/comparison/includes/missing.rst b/pandas/doc/source/getting_started/comparison/includes/missing.rst
new file mode 100644
index 0000000000000..ab5d90166e7b0
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/missing.rst
@@ -0,0 +1,31 @@
+In pandas, :meth:`Series.isna` and :meth:`Series.notna` can be used to filter the rows.
+
+.. ipython:: python
+
+   outer_join[outer_join["value_x"].isna()]
+   outer_join[outer_join["value_x"].notna()]
+
+pandas provides :ref:`a variety of methods to work with missing data <missing_data>`. Here are some examples:
+
+Drop rows with missing values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. ipython:: python
+
+   outer_join.dropna()
+
+Forward fill from previous rows
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. ipython:: python
+
+   outer_join.ffill()
+
+Replace missing values with a specified value
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Using the mean:
+
+.. ipython:: python
+
+   outer_join["value_x"].fillna(outer_join["value_x"].mean())
diff --git a/pandas/doc/source/getting_started/comparison/includes/missing_intro.rst b/pandas/doc/source/getting_started/comparison/includes/missing_intro.rst
new file mode 100644
index 0000000000000..366aa43d1264c
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/missing_intro.rst
@@ -0,0 +1,9 @@
+pandas represents missing data with the special float value ``NaN`` (not a number).  Many of the
+semantics are the same; for example missing data propagates through numeric operations, and is
+ignored by default for aggregations.
+
+.. ipython:: python
+
+   outer_join
+   outer_join["value_x"] + outer_join["value_y"]
+   outer_join["value_x"].sum()
diff --git a/pandas/doc/source/getting_started/comparison/includes/nth_word.rst b/pandas/doc/source/getting_started/comparison/includes/nth_word.rst
new file mode 100644
index 0000000000000..20e2ec47a8c9d
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/nth_word.rst
@@ -0,0 +1,9 @@
+The simplest way to extract words in pandas is to split the strings by spaces, then reference the
+word by index. Note there are more powerful approaches should you need them.
+
+.. ipython:: python
+
+   firstlast = pd.DataFrame({"String": ["John Smith", "Jane Cook"]})
+   firstlast["First_Name"] = firstlast["String"].str.split(" ", expand=True)[0]
+   firstlast["Last_Name"] = firstlast["String"].str.rsplit(" ", expand=True)[1]
+   firstlast
diff --git a/pandas/doc/source/getting_started/comparison/includes/sorting.rst b/pandas/doc/source/getting_started/comparison/includes/sorting.rst
new file mode 100644
index 0000000000000..4e2e40a18adbd
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/sorting.rst
@@ -0,0 +1,6 @@
+pandas has a :meth:`DataFrame.sort_values` method, which takes a list of columns to sort by.
+
+.. ipython:: python
+
+   tips = tips.sort_values(["sex", "total_bill"])
+   tips
diff --git a/pandas/doc/source/getting_started/comparison/includes/time_date.rst b/pandas/doc/source/getting_started/comparison/includes/time_date.rst
new file mode 100644
index 0000000000000..fb9ee2e216cd7
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/time_date.rst
@@ -0,0 +1,22 @@
+.. ipython:: python
+
+   tips["date1"] = pd.Timestamp("2013-01-15")
+   tips["date2"] = pd.Timestamp("2015-02-15")
+   tips["date1_year"] = tips["date1"].dt.year
+   tips["date2_month"] = tips["date2"].dt.month
+   tips["date1_next"] = tips["date1"] + pd.offsets.MonthBegin()
+   tips["months_between"] = tips["date2"].dt.to_period("M") - tips[
+       "date1"
+   ].dt.to_period("M")
+
+   tips[
+       ["date1", "date2", "date1_year", "date2_month", "date1_next", "months_between"]
+   ]
+
+.. ipython:: python
+   :suppress:
+
+   tips = tips.drop(
+       ["date1", "date2", "date1_year", "date2_month", "date1_next", "months_between"],
+       axis=1,
+   )
diff --git a/pandas/doc/source/getting_started/comparison/includes/transform.rst b/pandas/doc/source/getting_started/comparison/includes/transform.rst
new file mode 100644
index 0000000000000..b7599471432ad
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/includes/transform.rst
@@ -0,0 +1,8 @@
+pandas provides a :ref:`groupby.transform` mechanism that allows these type of operations to be
+succinctly expressed in one operation.
+
+.. ipython:: python
+
+   gb = tips.groupby("smoker")["total_bill"]
+   tips["adj_total_bill"] = tips["total_bill"] - gb.transform("mean")
+   tips
diff --git a/pandas/doc/source/getting_started/comparison/index.rst b/pandas/doc/source/getting_started/comparison/index.rst
new file mode 100644
index 0000000000000..3133d74afa3db
--- /dev/null
+++ b/pandas/doc/source/getting_started/comparison/index.rst
@@ -0,0 +1,17 @@
+{{ header }}
+
+.. _comparison:
+
+===========================
+Comparison with other tools
+===========================
+
+.. toctree::
+    :maxdepth: 2
+
+    comparison_with_r
+    comparison_with_sql
+    comparison_with_spreadsheets
+    comparison_with_sas
+    comparison_with_stata
+    comparison_with_spss
diff --git a/pandas/doc/source/getting_started/index.rst b/pandas/doc/source/getting_started/index.rst
new file mode 100644
index 0000000000000..a17699a71fbd3
--- /dev/null
+++ b/pandas/doc/source/getting_started/index.rst
@@ -0,0 +1,671 @@
+{{ header }}
+
+.. _getting_started:
+
+===============
+Getting started
+===============
+
+Installation
+------------
+
+.. grid:: 1 2 2 2
+    :gutter: 4
+
+    .. grid-item-card:: Working with conda?
+        :class-card: install-card
+        :columns: 12 12 6 6
+        :padding: 3
+
+        pandas can be installed via conda from `conda-forge <https://anaconda.org/conda-forge/pandas>`__.
+
+        ++++++++++++++++++++++
+
+        .. code-block:: bash
+
+            conda install -c conda-forge pandas
+
+    .. grid-item-card:: Prefer pip?
+        :class-card: install-card
+        :columns: 12 12 6 6
+        :padding: 3
+
+        pandas can be installed via pip from `PyPI <https://pypi.org/project/pandas>`__.
+
+        ++++
+
+        .. code-block:: bash
+
+            pip install pandas
+
+    .. grid-item-card:: In-depth instructions?
+        :class-card: install-card
+        :columns: 12
+        :padding: 3
+
+        Installing a specific version? Installing from source? Check the advanced
+        installation page.
+
+        +++
+
+        .. button-ref:: install
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+
+.. _gentle_intro:
+
+Intro to pandas
+---------------
+
+.. raw:: html
+
+    <div class="container">
+    <div id="accordion" class="shadow tutorial-accordion">
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseOne">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        What kind of data does pandas handle?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_01_tableoriented>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseOne" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+When working with tabular data, such as data stored in spreadsheets or databases, pandas is the right tool for you. pandas will help you
+to explore, clean, and process your data. In pandas, a data table is called a :class:`DataFrame`.
+
+.. image:: ../_static/schemas/01_table_dataframe.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_01_tableoriented>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <dsintro>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseTwo">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How do I read and write tabular data?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_02_read_write>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseTwo" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+pandas supports the integration with many file formats or data sources out of the box (csv, excel, sql, json, parquet,…). The ability to import data from each of these
+data sources is provided by functions with the prefix, ``read_*``. Similarly, the ``to_*`` methods are used to store data.
+
+.. image:: ../_static/schemas/02_io_readwrite.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_02_read_write>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <io>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseThree">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How do I select a subset of a table?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_03_subset>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseThree" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+Selecting or filtering specific rows and/or columns? Filtering the data on a particular condition? Methods for slicing, selecting, and extracting the
+data you need are available in pandas.
+
+.. image:: ../_static/schemas/03_subset_columns_rows.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_03_subset>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <indexing>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseFour">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to create plots in pandas?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_04_plotting>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseFour" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+pandas provides plotting for your data right out of the box with the power of Matplotlib. Simply pick the plot type (scatter, bar, boxplot,...)
+corresponding to your data.
+
+.. image:: ../_static/schemas/04_plot_overview.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_04_plotting>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <visualization>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseFive">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to create new columns derived from existing columns?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_05_columns>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseFive" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+There's no need to loop over all rows of your data table to do calculations. Column data manipulations work elementwise in pandas.
+Adding a column to a :class:`DataFrame` based on existing data in other columns is straightforward.
+
+.. image:: ../_static/schemas/05_newcolumn_2.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_05_columns>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <basics.dataframe.sel_add_del>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseSix">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to calculate summary statistics?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_06_stats>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseSix" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+Basic statistics (mean, median, min, max, counts...) are easily calculable across data frames. These, or even custom aggregations, can be applied on the entire
+data set, a sliding window of the data, or grouped by categories. The latter is also known as the split-apply-combine approach.
+
+.. image:: ../_static/schemas/06_groupby.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_06_stats>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <groupby>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseSeven">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to reshape the layout of tables?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_07_reshape>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseSeven" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+Change the structure of your data table in a variety of ways. You can use :func:`~pandas.melt` to reshape your data from a wide format to a long and tidy one. Use :func:`~pandas.pivot`
+ to go from long to wide format. With aggregations built-in, a pivot table can be created with a single command.
+
+.. image:: ../_static/schemas/07_melt.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_07_reshape>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <reshaping>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseEight">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to combine data from multiple tables?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_08_combine>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseEight" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+Multiple tables can be concatenated column wise or row wise with pandas' database-like join and merge operations.
+
+.. image:: ../_static/schemas/08_concat_row.svg
+   :align: center
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_08_combine>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <merging>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseNine">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to handle time series data?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_09_timeseries>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseNine" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+pandas has great support for time series and has an extensive set of tools for working with dates, times, and time-indexed data.
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_09_timeseries>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <timeseries>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+        <div class="card tutorial-card">
+            <div class="card-header collapsed card-link" data-bs-toggle="collapse" data-bs-target="#collapseTen">
+                <div class="d-flex flex-row tutorial-card-header-1">
+                    <div class="d-flex flex-row tutorial-card-header-2">
+                        <button class="btn btn-dark btn-sm"></button>
+                        How to manipulate textual data?
+                    </div>
+                    <span class="badge gs-badge-link">
+
+:ref:`Straight to tutorial...<10min_tut_10_text>`
+
+.. raw:: html
+
+                    </span>
+                </div>
+            </div>
+            <div id="collapseTen" class="collapse" data-parent="#accordion">
+                <div class="card-body">
+
+Data sets often contain more than just numerical data. pandas provides a wide range of functions to clean textual data and extract useful information from it.
+
+.. raw:: html
+
+                    <div class="d-flex flex-row">
+                        <span class="badge gs-badge-link">
+
+:ref:`To introduction tutorial <10min_tut_10_text>`
+
+.. raw:: html
+
+                        </span>
+                        <span class="badge gs-badge-link">
+
+:ref:`To user guide <text>`
+
+.. raw:: html
+
+                        </span>
+                    </div>
+                </div>
+            </div>
+        </div>
+
+    </div>
+    </div>
+
+
+.. _comingfrom:
+
+Coming from...
+--------------
+
+Are you familiar with other software for manipulating tabular data? Learn
+the pandas-equivalent operations compared to software you already know:
+
+.. grid:: 1 2 2 2
+    :gutter: 4
+    :class-container: sd-text-center sd-d-inline-flex
+
+    .. grid-item-card::
+        :img-top: ../_static/logo_r.svg
+        :columns: 12 6 6 6
+        :class-card: comparison-card
+        :shadow: md
+
+        The `R programming language <https://www.r-project.org/>`__ provides a
+        ``data.frame`` data structure as well as packages like
+        `tidyverse <https://www.tidyverse.org>`__ which use and extend ``data.frame``
+        for convenient data handling functionalities similar to pandas.
+
+        +++
+
+        .. button-ref:: compare_with_r
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+    .. grid-item-card::
+        :img-top: ../_static/logo_sql.svg
+        :columns: 12 6 6 6
+        :class-card: comparison-card
+        :shadow: md
+
+        Already familiar with ``SELECT``, ``GROUP BY``, ``JOIN``, etc.?
+        Many SQL manipulations have equivalents in pandas.
+
+        +++
+
+        .. button-ref:: compare_with_sql
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+    .. grid-item-card::
+        :img-top: ../_static/logo_stata.svg
+        :columns: 12 6 6 6
+        :class-card: comparison-card
+        :shadow: md
+
+        The ``data set`` included in the `STATA <https://en.wikipedia.org/wiki/Stata>`__
+        statistical software suite corresponds to the pandas ``DataFrame``.
+        Many of the operations known from STATA have an equivalent in pandas.
+
+        +++
+
+        .. button-ref:: compare_with_stata
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+    .. grid-item-card::
+        :img-top: ../_static/spreadsheets/logo_excel.svg
+        :columns: 12 6 6 6
+        :class-card: comparison-card
+        :shadow: md
+
+        Users of `Excel <https://en.wikipedia.org/wiki/Microsoft_Excel>`__
+        or other spreadsheet programs will find that many of the concepts are
+        transferable to pandas.
+
+        +++
+
+        .. button-ref:: compare_with_spreadsheets
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+    .. grid-item-card::
+        :img-top: ../_static/logo_sas.svg
+        :columns: 12 6 6 6
+        :class-card: comparison-card
+        :shadow: md
+
+        `SAS <https://en.wikipedia.org/wiki/SAS_(software)>`__, the statistical software suite,
+        uses the ``data set`` structure, which closely corresponds pandas' ``DataFrame``.
+        Also SAS vectorized operations such as filtering or string processing operations
+        have similar functions in pandas.
+
+        +++
+
+        .. button-ref:: compare_with_sas
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            Learn more
+
+Tutorials
+---------
+
+For a quick overview of pandas functionality, see :ref:`10 Minutes to pandas<10min>`.
+
+You can also reference the pandas `cheat sheet <https://pandas.pydata.org/Pandas_Cheat_Sheet.pdf>`_
+for a succinct guide for manipulating data with pandas.
+
+The community produces a wide variety of tutorials available online. Some of the
+material is enlisted in the community contributed :ref:`communitytutorials`.
+
+
+.. If you update this toctree, also update the manual toctree in the
+   main index.rst.template
+
+.. toctree::
+    :maxdepth: 2
+    :hidden:
+
+    install
+    overview
+    intro_tutorials/index
+    comparison/index
+    tutorials
diff --git a/pandas/doc/source/getting_started/install.rst b/pandas/doc/source/getting_started/install.rst
new file mode 100644
index 0000000000000..bda959f380e8a
--- /dev/null
+++ b/pandas/doc/source/getting_started/install.rst
@@ -0,0 +1,376 @@
+.. _install:
+
+{{ header }}
+
+============
+Installation
+============
+
+The pandas development team officially distributes pandas for installation
+through the following methods:
+
+* Available on `conda-forge <https://anaconda.org/conda-forge/pandas>`__ for installation with the conda package manager.
+* Available on `PyPI <https://pypi.org/project/pandas/>`__ for installation with pip.
+* Available on `Github <https://github.com/pandas-dev/pandas>`__ for installation from source.
+
+.. note::
+    pandas may be installable from other sources besides the ones listed above,
+    but they are **not** managed by the pandas development team.
+
+.. _install.version:
+
+Python version support
+----------------------
+
+See :ref:`Python support policy <policies.python_support>`.
+
+Installing pandas
+-----------------
+
+.. _install.conda:
+
+Installing with Conda
+~~~~~~~~~~~~~~~~~~~~~
+
+For users working with the `Conda <https://conda.io/en/latest/>`__ package manager,
+pandas can be installed from the ``conda-forge`` channel.
+
+.. code-block:: shell
+
+    conda install -c conda-forge pandas
+
+To install the Conda package manager on your system, the
+`Miniforge distribution <https://github.com/conda-forge/miniforge?tab=readme-ov-file#install>`__
+is recommended.
+
+Additionally, it is recommended to install and run pandas from a virtual environment.
+
+.. code-block:: shell
+
+    conda create -c conda-forge -n name_of_my_env python pandas
+    # On Linux or MacOS
+    source activate name_of_my_env
+    # On Windows
+    activate name_of_my_env
+
+.. tip::
+    For users that are new to Python, the easiest way to install Python, pandas, and the
+    packages that make up the `PyData <https://pydata.org/>`__ stack such as
+    `SciPy <https://scipy.org/>`__, `NumPy <https://numpy.org/>`__ and
+    `Matplotlib <https://matplotlib.org/>`__
+    is with `Anaconda <https://docs.anaconda.com/anaconda/install/>`__, a cross-platform
+    (Linux, macOS, Windows) Python distribution for data analytics and
+    scientific computing.
+
+    However, pandas from Anaconda is **not** officially managed by the pandas development team.
+
+.. _install.pip:
+
+Installing with pip
+~~~~~~~~~~~~~~~~~~~
+
+For users working with the `pip <https://pip.pypa.io/en/stable/>`__ package manager,
+pandas can be installed from `PyPI <https://pypi.org/project/pandas/>`__.
+
+.. code-block:: shell
+
+    pip install pandas
+
+pandas can also be installed with sets of optional dependencies to enable certain functionality. For example,
+to install pandas with the optional dependencies to read Excel files.
+
+.. code-block:: shell
+
+    pip install "pandas[excel]"
+
+The full list of extras that can be installed can be found in the :ref:`dependency section.<install.optional_dependencies>`
+
+Additionally, it is recommended to install and run pandas from a virtual environment, for example,
+using the Python standard library's `venv <https://docs.python.org/3/library/venv.html>`__
+
+.. _install.source:
+
+Installing from source
+~~~~~~~~~~~~~~~~~~~~~~
+
+See the :ref:`contributing guide <contributing>` for complete instructions on building from the git source tree.
+Further, see :ref:`creating a development environment <contributing_environment>` if you wish to create
+a pandas development environment.
+
+.. _install.dev:
+
+Installing the development version of pandas
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Installing the development version is the quickest way to:
+
+* Try a new feature that will be shipped in the next release (that is, a feature from a pull-request that was recently merged to the main branch).
+* Check whether a bug you encountered has been fixed since the last release.
+
+The development version is usually uploaded daily to the scientific-python-nightly-wheels
+index from the PyPI registry of anaconda.org. You can install it by running.
+
+.. code-block:: shell
+
+    pip install --pre --extra-index https://pypi.anaconda.org/scientific-python-nightly-wheels/simple pandas
+
+.. note::
+    You might be required to uninstall an existing version of pandas to install the development version.
+
+    .. code-block:: shell
+
+        pip uninstall pandas -y
+
+Running the test suite
+----------------------
+
+If pandas has been installed :ref:`from source <install.source>`, running ``pytest pandas`` will run all of pandas unit tests.
+
+The unit tests can also be run from the pandas module itself with the :func:`test` function. The packages required to run the tests
+can be installed with ``pip install "pandas[test]"``.
+
+.. note::
+
+    Test failures are not necessarily indicative of a broken pandas installation.
+
+.. _install.dependencies:
+
+Dependencies
+------------
+
+.. _install.required_dependencies:
+
+Required dependencies
+~~~~~~~~~~~~~~~~~~~~~
+
+pandas requires the following dependencies.
+
+================================================================ ==========================
+Package                                                          Minimum supported version
+================================================================ ==========================
+`NumPy <https://numpy.org>`__                                    1.23.5
+`python-dateutil <https://dateutil.readthedocs.io/en/stable/>`__ 2.8.2
+`tzdata <https://pypi.org/project/tzdata/>`__                    2022.7
+================================================================ ==========================
+
+.. _install.optional_dependencies:
+
+Optional dependencies
+~~~~~~~~~~~~~~~~~~~~~
+
+pandas has many optional dependencies that are only used for specific methods.
+For example, :func:`pandas.read_hdf` requires the ``pytables`` package, while
+:meth:`DataFrame.to_markdown` requires the ``tabulate`` package. If the
+optional dependency is not installed, pandas will raise an ``ImportError`` when
+the method requiring that dependency is called.
+
+With pip, optional pandas dependencies can be installed or managed in a file (e.g. requirements.txt or pyproject.toml)
+as optional extras (e.g. ``pandas[performance, aws]``). All optional dependencies can be installed with ``pandas[all]``,
+and specific sets of dependencies are listed in the sections below.
+
+.. _install.recommended_dependencies:
+
+Performance dependencies (recommended)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. note::
+
+   You are highly encouraged to install these libraries, as they provide speed improvements, especially
+   when working with large data sets.
+
+Installable with ``pip install "pandas[performance]"``
+
+===================================================== ================== ================== ===================================================================================================================================================================================
+Dependency                                            Minimum Version    pip extra          Notes
+===================================================== ================== ================== ===================================================================================================================================================================================
+`numexpr <https://github.com/pydata/numexpr>`__       2.8.4              performance        Accelerates certain numerical operations by using multiple cores as well as smart chunking and caching to achieve large speedups
+`bottleneck <https://github.com/pydata/bottleneck>`__ 1.3.6              performance        Accelerates certain types of ``nan`` by using specialized cython routines to achieve large speedup.
+`numba <https://github.com/numba/numba>`__            0.56.4             performance        Alternative execution engine for operations that accept ``engine="numba"`` using a JIT compiler that translates Python functions to optimized machine code using the LLVM compiler.
+===================================================== ================== ================== ===================================================================================================================================================================================
+
+Visualization
+^^^^^^^^^^^^^
+
+Installable with ``pip install "pandas[plot, output-formatting]"``.
+
+========================================================== ================== ================== =======================================================
+Dependency                                                 Minimum Version    pip extra          Notes
+========================================================== ================== ================== =======================================================
+`matplotlib <https://github.com/matplotlib/matplotlib>`__  3.6.3              plot               Plotting library
+`Jinja2 <https://github.com/pallets/jinja>`__              3.1.2              output-formatting  Conditional formatting with DataFrame.style
+`tabulate <https://github.com/astanin/python-tabulate>`__  0.9.0              output-formatting  Printing in Markdown-friendly format (see `tabulate`_)
+========================================================== ================== ================== =======================================================
+
+Computation
+^^^^^^^^^^^
+
+Installable with ``pip install "pandas[computation]"``.
+
+============================================== ================== =============== =======================================
+Dependency                                     Minimum Version    pip extra       Notes
+============================================== ================== =============== =======================================
+`SciPy <https://github.com/scipy/scipy>`__     1.10.0             computation     Miscellaneous statistical functions
+`xarray <https://github.com/pydata/xarray>`__  2022.12.0          computation     pandas-like API for N-dimensional data
+============================================== ================== =============== =======================================
+
+.. _install.excel_dependencies:
+
+Excel files
+^^^^^^^^^^^
+
+Installable with ``pip install "pandas[excel]"``.
+
+================================================================== ================== =============== =============================================================
+Dependency                                                         Minimum Version    pip extra       Notes
+================================================================== ================== =============== =============================================================
+`xlrd <https://github.com/python-excel/xlrd>`__                    2.0.1              excel           Reading for xls files
+`xlsxwriter <https://github.com/jmcnamara/XlsxWriter>`__           3.0.5              excel           Writing for xlsx files
+`openpyxl <https://github.com/theorchard/openpyxl>`__              3.1.0              excel           Reading / writing for Excel 2010 xlsx/xlsm/xltx/xltm files
+`pyxlsb <https://github.com/willtrnr/pyxlsb>`__                    1.0.10             excel           Reading for xlsb files
+`python-calamine <https://github.com/dimastbk/python-calamine>`__  0.1.7              excel           Reading for xls/xlsx/xlsm/xlsb/xla/xlam/ods files
+`odfpy <https://github.com/eea/odfpy>`__                           1.4.1              excel           Reading / writing for OpenDocument 1.2 files
+================================================================== ================== =============== =============================================================
+
+HTML
+^^^^
+
+Installable with ``pip install "pandas[html]"``.
+
+=============================================================== ================== =============== ==========================
+Dependency                                                      Minimum Version    pip extra       Notes
+=============================================================== ================== =============== ==========================
+`BeautifulSoup4 <https://github.com/wention/BeautifulSoup4>`__  4.11.2             html            HTML parser for read_html
+`html5lib <https://github.com/html5lib/html5lib-python>`__      1.1                html            HTML parser for read_html
+`lxml <https://github.com/lxml/lxml>`__                         4.9.2              html            HTML parser for read_html
+=============================================================== ================== =============== ==========================
+
+One of the following combinations of libraries is needed to use the
+top-level :func:`~pandas.read_html` function:
+
+* `BeautifulSoup4`_ and `html5lib`_
+* `BeautifulSoup4`_ and `lxml`_
+* `BeautifulSoup4`_ and `html5lib`_ and `lxml`_
+* Only `lxml`_, although see :ref:`HTML Table Parsing <io.html.gotchas>`
+  for reasons as to why you should probably **not** take this approach.
+
+.. warning::
+
+    * if you install `BeautifulSoup4`_ you must install either
+      `lxml`_ or `html5lib`_ or both.
+      :func:`~pandas.read_html` will **not** work with *only*
+      `BeautifulSoup4`_ installed.
+    * You are highly encouraged to read :ref:`HTML Table Parsing gotchas <io.html.gotchas>`.
+      It explains issues surrounding the installation and
+      usage of the above three libraries.
+
+.. _html5lib: https://github.com/html5lib/html5lib-python
+.. _BeautifulSoup4: https://www.crummy.com/software/BeautifulSoup
+.. _lxml: https://lxml.de
+.. _tabulate: https://github.com/astanin/python-tabulate
+
+XML
+^^^
+
+Installable with ``pip install "pandas[xml]"``.
+
+======================================== ================== =============== ====================================================
+Dependency                               Minimum Version    pip extra       Notes
+======================================== ================== =============== ====================================================
+`lxml <https://github.com/lxml/lxml>`__  4.9.2              xml             XML parser for read_xml and tree builder for to_xml
+======================================== ================== =============== ====================================================
+
+SQL databases
+^^^^^^^^^^^^^
+
+Traditional drivers are installable with ``pip install "pandas[postgresql, mysql, sql-other]"``
+
+================================================================== ================== =============== ============================================
+Dependency                                                         Minimum Version    pip extra       Notes
+================================================================== ================== =============== ============================================
+`SQLAlchemy <https://github.com/sqlalchemy/sqlalchemy>`__          2.0.0              postgresql,     SQL support for databases other than sqlite
+                                                                                      mysql,
+                                                                                      sql-other
+`psycopg2 <https://github.com/psycopg/psycopg2>`__                 2.9.6              postgresql      PostgreSQL engine for sqlalchemy
+`pymysql <https://github.com/PyMySQL/PyMySQL>`__                   1.0.2              mysql           MySQL engine for sqlalchemy
+`adbc-driver-postgresql <https://github.com/apache/arrow-adbc>`__  0.10.0             postgresql      ADBC Driver for PostgreSQL
+`adbc-driver-sqlite <https://github.com/apache/arrow-adbc>`__      0.8.0              sql-other       ADBC Driver for SQLite
+================================================================== ================== =============== ============================================
+
+Other data sources
+^^^^^^^^^^^^^^^^^^
+
+Installable with ``pip install "pandas[hdf5, parquet, feather, spss, excel]"``
+
+====================================================== ================== ================ ==========================================================
+Dependency                                             Minimum Version    pip extra        Notes
+====================================================== ================== ================ ==========================================================
+`PyTables <https://github.com/PyTables/PyTables>`__    3.8.0              hdf5             HDF5-based reading / writing
+`blosc <https://github.com/Blosc/c-blosc>`__           1.21.3             hdf5             Compression for HDF5; only available on ``conda``
+`zlib <https://github.com/madler/zlib>`__                                 hdf5             Compression for HDF5
+`fastparquet <https://github.com/dask/fastparquet>`__  2023.10.0          -                Parquet reading / writing (pyarrow is default)
+`pyarrow <https://github.com/apache/arrow>`__          10.0.1             parquet, feather Parquet, ORC, and feather reading / writing
+`pyreadstat <https://github.com/Roche/pyreadstat>`__   1.2.0              spss             SPSS files (.sav) reading
+`odfpy <https://github.com/eea/odfpy>`__               1.4.1              excel            Open document format (.odf, .ods, .odt) reading / writing
+====================================================== ================== ================ ==========================================================
+
+.. _install.warn_orc:
+
+.. warning::
+
+    * If you want to use :func:`~pandas.read_orc`, it is highly recommended to install pyarrow using conda.
+      :func:`~pandas.read_orc` may fail if pyarrow was installed from pypi, and :func:`~pandas.read_orc` is
+      not compatible with Windows OS.
+
+Access data in the cloud
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+Installable with ``pip install "pandas[fss, aws, gcp]"``
+
+============================================ ================== =============== ==========================================================
+Dependency                                   Minimum Version    pip extra       Notes
+============================================ ================== =============== ==========================================================
+`fsspec <https://github.com/fsspec>`__       2022.11.0          fss, gcp, aws   Handling files aside from simple local and HTTP (required
+                                                                                dependency of s3fs, gcsfs).
+`gcsfs <https://github.com/fsspec/gcsfs>`__  2022.11.0          gcp             Google Cloud Storage access
+`s3fs <https://github.com/fsspec/s3fs>`__    2022.11.0          aws             Amazon S3 access
+============================================ ================== =============== ==========================================================
+
+Clipboard
+^^^^^^^^^
+
+Installable with ``pip install "pandas[clipboard]"``.
+
+======================================================================================== ================== =============== ==============
+Dependency                                                                               Minimum Version    pip extra       Notes
+======================================================================================== ================== =============== ==============
+`PyQt4 <https://pypi.org/project/PyQt4/>`__/`PyQt5 <https://pypi.org/project/PyQt5/>`__  5.15.9             clipboard       Clipboard I/O
+`qtpy <https://github.com/spyder-ide/qtpy>`__                                            2.3.0              clipboard       Clipboard I/O
+======================================================================================== ================== =============== ==============
+
+.. note::
+
+   Depending on operating system, system-level packages may need to installed.
+   For clipboard to operate on Linux one of the CLI tools ``xclip`` or ``xsel`` must be installed on your system.
+
+
+Compression
+^^^^^^^^^^^
+
+Installable with ``pip install "pandas[compression]"``
+
+================================================= ================== =============== ======================
+Dependency                                        Minimum Version    pip extra       Notes
+================================================= ================== =============== ======================
+`Zstandard <https://github.com/facebook/zstd>`__  0.19.0             compression     Zstandard compression
+================================================= ================== =============== ======================
+
+Timezone
+^^^^^^^^
+
+Installable with ``pip install "pandas[timezone]"``
+
+========================================== ================== =================== ==============================================
+Dependency                                 Minimum Version    pip extra           Notes
+========================================== ================== =================== ==============================================
+`pytz <https://github.com/stub42/pytz>`__  2023.4             timezone            Alternative timezone library to ``zoneinfo``.
+========================================== ================== =================== ==============================================
diff --git a/pandas/doc/source/getting_started/intro_tutorials/01_table_oriented.rst b/pandas/doc/source/getting_started/intro_tutorials/01_table_oriented.rst
new file mode 100644
index 0000000000000..efcdb22778ef4
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/01_table_oriented.rst
@@ -0,0 +1,222 @@
+.. _10min_tut_01_tableoriented:
+
+{{ header }}
+
+What kind of data does pandas handle?
+=====================================
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to start using pandas
+
+.. ipython:: python
+
+    import pandas as pd
+
+To load the pandas package and start working with it, import the
+package. The community agreed alias for pandas is ``pd``, so loading
+pandas as ``pd`` is assumed standard practice for all of the pandas
+documentation.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+pandas data table representation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/01_table_dataframe.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to store passenger data of the Titanic. For a number of passengers, I know the name (characters), age (integers) and sex (male/female) data.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {
+            "Name": [
+                "Braund, Mr. Owen Harris",
+                "Allen, Mr. William Henry",
+                "Bonnell, Miss Elizabeth",
+            ],
+            "Age": [22, 35, 58],
+            "Sex": ["male", "male", "female"],
+        }
+    )
+    df
+
+To manually store data in a table, create a ``DataFrame``. When using a Python dictionary of lists, the dictionary keys will be used as column headers and
+the values in each list as columns of the ``DataFrame``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+A :class:`DataFrame` is a 2-dimensional data structure that can store data of
+different types (including characters, integers, floating point values,
+categorical data and more) in columns. It is similar to a spreadsheet, a
+SQL table or the ``data.frame`` in R.
+
+-  The table has 3 columns, each of them with a column label. The column
+   labels are respectively ``Name``, ``Age`` and ``Sex``.
+-  The column ``Name`` consists of textual data with each value a
+   string, the column ``Age`` are numbers and the column ``Sex`` is
+   textual data.
+
+In spreadsheet software, the table representation of our data would look
+very similar:
+
+.. image:: ../../_static/schemas/01_table_spreadsheet.png
+   :align: center
+
+Each column in a ``DataFrame`` is a ``Series``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/01_table_series.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m just interested in working with the data in the column ``Age``
+
+.. ipython:: python
+
+    df["Age"]
+
+When selecting a single column of a pandas :class:`DataFrame`, the result is
+a pandas :class:`Series`. To select the column, use the column label in
+between square brackets ``[]``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+    If you are familiar with Python
+    :ref:`dictionaries <python:tut-dictionaries>`, the selection of a
+    single column is very similar to the selection of dictionary values based on
+    the key.
+
+You can create a ``Series`` from scratch as well:
+
+.. ipython:: python
+
+    ages = pd.Series([22, 35, 58], name="Age")
+    ages
+
+A pandas ``Series`` has no column labels, as it is just a single column
+of a ``DataFrame``. A Series does have row labels.
+
+Do something with a DataFrame or Series
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to know the maximum Age of the passengers
+
+We can do this on the ``DataFrame`` by selecting the ``Age`` column and
+applying ``max()``:
+
+.. ipython:: python
+
+    df["Age"].max()
+
+Or to the ``Series``:
+
+.. ipython:: python
+
+    ages.max()
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+As illustrated by the ``max()`` method, you can *do* things with a
+``DataFrame`` or ``Series``. pandas provides a lot of functionalities,
+each of them a *method* you can apply to a ``DataFrame`` or ``Series``.
+As methods are functions, do not forget to use parentheses ``()``.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in some basic statistics of the numerical data of my data table
+
+.. ipython:: python
+
+    df.describe()
+
+The :func:`~DataFrame.describe` method provides a quick overview of the numerical data in
+a ``DataFrame``. As the ``Name`` and ``Sex`` columns are textual data,
+these are by default not taken into account by the :func:`~DataFrame.describe` method.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Many pandas operations return a ``DataFrame`` or a ``Series``. The
+:func:`~DataFrame.describe` method is an example of a pandas operation returning a
+pandas ``Series`` or a pandas ``DataFrame``.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+Check more options on ``describe`` in the user guide section about :ref:`aggregations with describe <basics.describe>`
+
+.. raw:: html
+
+    </div>
+
+.. note::
+    This is just a starting point. Similar to spreadsheet
+    software, pandas represents data as a table with columns and rows. Apart
+    from the representation, the data manipulations and calculations
+    you would do in spreadsheet software are also supported by pandas. Continue
+    reading the next tutorials to get started!
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Import the package, aka ``import pandas as pd``
+-  A table of data is stored as a pandas ``DataFrame``
+-  Each column in a ``DataFrame`` is a ``Series``
+-  You can do things by applying a method on a ``DataFrame`` or ``Series``
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A more extended explanation of ``DataFrame`` and ``Series`` is provided in the :ref:`introduction to data structures <dsintro>` page.
+
+.. raw:: html
+
+    </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/02_read_write.rst b/pandas/doc/source/getting_started/intro_tutorials/02_read_write.rst
new file mode 100644
index 0000000000000..0549c17a1013c
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/02_read_write.rst
@@ -0,0 +1,214 @@
+.. _10min_tut_02_read_write:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/titanic.rst
+
+.. raw:: html
+
+            </li>
+        </ul>
+    </div>
+
+How do I read and write tabular data?
+=====================================
+
+.. image:: ../../_static/schemas/02_io_readwrite.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to analyze the Titanic passenger data, available as a CSV file.
+
+.. ipython:: python
+
+    titanic = pd.read_csv("data/titanic.csv")
+
+pandas provides the :func:`read_csv` function to read data stored as a csv
+file into a pandas ``DataFrame``. pandas supports many different file
+formats or data sources out of the box (csv, excel, sql, json, parquet,
+…), each of them with the prefix ``read_*``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Make sure to always have a check on the data after reading in the
+data. When displaying a ``DataFrame``, the first and last 5 rows will be
+shown by default:
+
+.. ipython:: python
+
+    titanic
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to see the first 8 rows of a pandas DataFrame.
+
+.. ipython:: python
+
+    titanic.head(8)
+
+To see the first N rows of a ``DataFrame``, use the :meth:`~DataFrame.head` method with
+the required number of rows (in this case 8) as argument.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+
+    Interested in the last N rows instead? pandas also provides a
+    :meth:`~DataFrame.tail` method. For example, ``titanic.tail(10)`` will return the last
+    10 rows of the DataFrame.
+
+A check on how pandas interpreted each of the column data types can be
+done by requesting the pandas ``dtypes`` attribute:
+
+.. ipython:: python
+
+    titanic.dtypes
+
+For each of the columns, the used data type is enlisted. The data types
+in this ``DataFrame`` are integers (``int64``), floats (``float64``) and
+strings (``object``).
+
+.. note::
+    When asking for the ``dtypes``, no parentheses ``()`` are used!
+    ``dtypes`` is an attribute of a ``DataFrame`` and ``Series``. Attributes
+    of a ``DataFrame`` or ``Series`` do not need ``()``. Attributes
+    represent a characteristic of a ``DataFrame``/``Series``, whereas
+    methods (which require parentheses ``()``) *do* something with the
+    ``DataFrame``/``Series`` as introduced in the :ref:`first tutorial <10min_tut_01_tableoriented>`.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+My colleague requested the Titanic data as a spreadsheet.
+
+.. note::
+    If you want to use :func:`~pandas.to_excel` and :func:`~pandas.read_excel`,
+    you need to install an Excel reader as outlined in the
+    :ref:`Excel files <install.excel_dependencies>` section of the
+    installation documentation.
+
+.. ipython:: python
+
+    titanic.to_excel("titanic.xlsx", sheet_name="passengers", index=False)
+
+Whereas ``read_*`` functions are used to read data to pandas, the
+``to_*`` methods are used to store data. The :meth:`~DataFrame.to_excel` method stores
+the data as an excel file. In the example here, the ``sheet_name`` is
+named *passengers* instead of the default *Sheet1*. By setting
+``index=False`` the row index labels are not saved in the spreadsheet.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The equivalent read function :meth:`~DataFrame.read_excel` will reload the data to a
+``DataFrame``:
+
+.. ipython:: python
+
+    titanic = pd.read_excel("titanic.xlsx", sheet_name="passengers")
+
+.. ipython:: python
+
+    titanic.head()
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("titanic.xlsx")
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in a technical summary of a ``DataFrame``
+
+.. ipython:: python
+
+    titanic.info()
+
+
+The method :meth:`~DataFrame.info` provides technical information about a
+``DataFrame``, so let’s explain the output in more detail:
+
+-  It is indeed a :class:`DataFrame`.
+-  There are 891 entries, i.e. 891 rows.
+-  Each row has a row label (aka the ``index``) with values ranging from
+   0 to 890.
+-  The table has 12 columns. Most columns have a value for each of the
+   rows (all 891 values are ``non-null``). Some columns do have missing
+   values and less than 891 ``non-null`` values.
+-  The columns ``Name``, ``Sex``, ``Cabin`` and ``Embarked`` consist of
+   textual data (strings, aka ``object``). The other columns are
+   numerical data, some of them are whole numbers (``integer``) and
+   others are real numbers (``float``).
+-  The kind of data (characters, integers, …) in the different columns
+   are summarized by listing the ``dtypes``.
+-  The approximate amount of RAM used to hold the DataFrame is provided
+   as well.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Getting data in to pandas from many different file formats or data
+   sources is supported by ``read_*`` functions.
+-  Exporting data out of pandas is provided by different
+   ``to_*`` methods.
+-  The ``head``/``tail``/``info`` methods and the ``dtypes`` attribute
+   are convenient for a first check.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For a complete overview of the input and output possibilities from and to pandas, see the user guide section about :ref:`reader and writer functions <io>`.
+
+.. raw:: html
+
+    </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/03_subset_data.rst b/pandas/doc/source/getting_started/intro_tutorials/03_subset_data.rst
new file mode 100644
index 0000000000000..ced976f680885
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/03_subset_data.rst
@@ -0,0 +1,377 @@
+.. _10min_tut_03_subset:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/titanic.rst
+
+.. ipython:: python
+
+    titanic = pd.read_csv("data/titanic.csv")
+    titanic.head()
+
+.. raw:: html
+
+            </li>
+        </ul>
+    </div>
+
+How do I select a subset of a ``DataFrame``?
+============================================
+
+How do I select specific columns from a ``DataFrame``?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/03_subset_columns.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in the age of the Titanic passengers.
+
+.. ipython:: python
+
+    ages = titanic["Age"]
+    ages.head()
+
+To select a single column, use square brackets ``[]`` with the column
+name of the column of interest.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Each column in a :class:`DataFrame` is a :class:`Series`. As a single column is
+selected, the returned object is a pandas :class:`Series`. We can verify this
+by checking the type of the output:
+
+.. ipython:: python
+
+    type(titanic["Age"])
+
+And have a look at the ``shape`` of the output:
+
+.. ipython:: python
+
+    titanic["Age"].shape
+
+:attr:`DataFrame.shape` is an attribute (remember :ref:`tutorial on reading and writing <10min_tut_02_read_write>`, do not use parentheses for attributes) of a
+pandas ``Series`` and ``DataFrame`` containing the number of rows and
+columns: *(nrows, ncolumns)*. A pandas Series is 1-dimensional and only
+the number of rows is returned.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in the age and sex of the Titanic passengers.
+
+.. ipython:: python
+
+    age_sex = titanic[["Age", "Sex"]]
+    age_sex.head()
+
+To select multiple columns, use a list of column names within the
+selection brackets ``[]``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+    The inner square brackets define a
+    :ref:`Python list <python:tut-morelists>` with column names, whereas
+    the outer square brackets are used to select the data from a pandas
+    ``DataFrame`` as seen in the previous example.
+
+The returned data type is a pandas DataFrame:
+
+.. ipython:: python
+
+    type(titanic[["Age", "Sex"]])
+
+.. ipython:: python
+
+    titanic[["Age", "Sex"]].shape
+
+The selection returned a ``DataFrame`` with 891 rows and 2 columns. Remember, a
+``DataFrame`` is 2-dimensional with both a row and column dimension.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For basic information on indexing, see the user guide section on :ref:`indexing and selecting data <indexing.basics>`.
+
+.. raw:: html
+
+    </div>
+
+How do I filter specific rows from a ``DataFrame``?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/03_subset_rows.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in the passengers older than 35 years.
+
+.. ipython:: python
+
+    above_35 = titanic[titanic["Age"] > 35]
+    above_35.head()
+
+To select rows based on a conditional expression, use a condition inside
+the selection brackets ``[]``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The condition inside the selection
+brackets ``titanic["Age"] > 35`` checks for which rows the ``Age``
+column has a value larger than 35:
+
+.. ipython:: python
+
+    titanic["Age"] > 35
+
+The output of the conditional expression (``>``, but also ``==``,
+``!=``, ``<``, ``<=``,… would work) is actually a pandas ``Series`` of
+boolean values (either ``True`` or ``False``) with the same number of
+rows as the original ``DataFrame``. Such a ``Series`` of boolean values
+can be used to filter the ``DataFrame`` by putting it in between the
+selection brackets ``[]``. Only rows for which the value is ``True``
+will be selected.
+
+We know from before that the original Titanic ``DataFrame`` consists of
+891 rows. Let’s have a look at the number of rows which satisfy the
+condition by checking the ``shape`` attribute of the resulting
+``DataFrame`` ``above_35``:
+
+.. ipython:: python
+
+    above_35.shape
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in the Titanic passengers from cabin class 2 and 3.
+
+.. ipython:: python
+
+    class_23 = titanic[titanic["Pclass"].isin([2, 3])]
+    class_23.head()
+
+Similar to the conditional expression, the :func:`~Series.isin` conditional function
+returns a ``True`` for each row the values are in the provided list. To
+filter the rows based on such a function, use the conditional function
+inside the selection brackets ``[]``. In this case, the condition inside
+the selection brackets ``titanic["Pclass"].isin([2, 3])`` checks for
+which rows the ``Pclass`` column is either 2 or 3.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The above is equivalent to filtering by rows for which the class is
+either 2 or 3 and combining the two statements with an ``|`` (or)
+operator:
+
+.. ipython:: python
+
+    class_23 = titanic[(titanic["Pclass"] == 2) | (titanic["Pclass"] == 3)]
+    class_23.head()
+
+.. note::
+    When combining multiple conditional statements, each condition
+    must be surrounded by parentheses ``()``. Moreover, you can not use
+    ``or``/``and`` but need to use the ``or`` operator ``|`` and the ``and``
+    operator ``&``.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+See the dedicated section in the user guide about :ref:`boolean indexing <indexing.boolean>` or about the :ref:`isin function <indexing.basics.indexing_isin>`.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to work with passenger data for which the age is known.
+
+.. ipython:: python
+
+    age_no_na = titanic[titanic["Age"].notna()]
+    age_no_na.head()
+
+The :meth:`~Series.notna` conditional function returns a ``True`` for each row the
+values are not a ``Null`` value. As such, this can be combined with the
+selection brackets ``[]`` to filter the data table.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+You might wonder what actually changed, as the first 5 lines are still
+the same values. One way to verify is to check if the shape has changed:
+
+.. ipython:: python
+
+    age_no_na.shape
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For more dedicated functions on missing values, see the user guide section about :ref:`handling missing data <missing_data>`.
+
+.. raw:: html
+
+    </div>
+
+.. _10min_tut_03_subset.rows_and_columns:
+
+How do I select specific rows and columns from a ``DataFrame``?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/03_subset_columns_rows.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in the names of the passengers older than 35 years.
+
+.. ipython:: python
+
+    adult_names = titanic.loc[titanic["Age"] > 35, "Name"]
+    adult_names.head()
+
+In this case, a subset of both rows and columns is made in one go and
+just using selection brackets ``[]`` is not sufficient anymore. The
+``loc``/``iloc`` operators are required in front of the selection
+brackets ``[]``. When using ``loc``/``iloc``, the part before the comma
+is the rows you want, and the part after the comma is the columns you
+want to select.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+When using column names, row labels or a condition expression, use
+the ``loc`` operator in front of the selection brackets ``[]``. For both
+the part before and after the comma, you can use a single label, a list
+of labels, a slice of labels, a conditional expression or a colon. Using
+a colon specifies you want to select all rows or columns.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I’m interested in rows 10 till 25 and columns 3 to 5.
+
+.. ipython:: python
+
+    titanic.iloc[9:25, 2:5]
+
+Again, a subset of both rows and columns is made in one go and just
+using selection brackets ``[]`` is not sufficient anymore. When
+specifically interested in certain rows and/or columns based on their
+position in the table, use the ``iloc`` operator in front of the
+selection brackets ``[]``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+When selecting specific rows and/or columns with ``loc`` or ``iloc``,
+new values can be assigned to the selected data. For example, to assign
+the name ``anonymous`` to the first 3 elements of the fourth column:
+
+.. ipython:: python
+
+    titanic.iloc[0:3, 3] = "anonymous"
+    titanic.iloc[:5, 3]
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+See the user guide section on :ref:`different choices for indexing <indexing.choice>` to get more insight into the usage of ``loc`` and ``iloc``.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  When selecting subsets of data, square brackets ``[]`` are used.
+-  Inside these square brackets, you can use a single column/row label, a list
+   of column/row labels, a slice of labels, a conditional expression or
+   a colon.
+-  Use ``loc`` for label-based selection (using row/column names).
+-  Use ``iloc`` for position-based selection (using table positions).
+-  You can assign new values to a selection based on ``loc``/``iloc``.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full overview of indexing is provided in the user guide pages on :ref:`indexing and selecting data <indexing>`.
+
+.. raw:: html
+
+    </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/04_plotting.rst b/pandas/doc/source/getting_started/intro_tutorials/04_plotting.rst
new file mode 100644
index 0000000000000..e9f83c602d086
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/04_plotting.rst
@@ -0,0 +1,259 @@
+.. _10min_tut_04_plotting:
+
+{{ header }}
+
+How do I create plots in pandas?
+----------------------------------
+
+.. image:: ../../_static/schemas/04_plot_overview.svg
+   :align: center
+
+.. ipython:: python
+
+    import pandas as pd
+    import matplotlib.pyplot as plt
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/air_quality_no2.rst
+
+.. ipython:: python
+
+    air_quality = pd.read_csv("data/air_quality_no2.csv", index_col=0, parse_dates=True)
+    air_quality.head()
+
+.. note::
+    The ``index_col=0`` and ``parse_dates=True`` parameters passed to the ``read_csv`` function define
+    the first (0th) column as index of the resulting ``DataFrame`` and convert the dates in the column
+    to :class:`Timestamp`  objects, respectively.
+
+
+.. raw:: html
+
+        </li>
+    </ul>
+    </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want a quick visual check of the data.
+
+.. ipython:: python
+   :okwarning:
+
+    @savefig 04_airqual_quick.png
+    air_quality.plot()
+    plt.show()
+
+With a ``DataFrame``, pandas creates by default one line plot for each of
+the columns with numeric data.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to plot only the columns of the data table with the data from Paris.
+
+.. ipython:: python
+   :suppress:
+
+    # We need to clear the figure here as, within doc generation, the plot
+    # accumulates data on each plot(). This is not needed when running
+    # in a notebook, so is suppressed from output.
+    plt.clf()
+
+.. ipython:: python
+   :okwarning:
+
+    @savefig 04_airqual_paris.png
+    air_quality["station_paris"].plot()
+    plt.show()
+
+To plot a specific column, use a selection method from the
+:ref:`subset data tutorial <10min_tut_03_subset>` in combination with the :meth:`~DataFrame.plot`
+method. Hence, the :meth:`~DataFrame.plot` method works on both ``Series`` and
+``DataFrame``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to visually compare the :math:`NO_2` values measured in London versus Paris.
+
+.. ipython:: python
+   :okwarning:
+
+    @savefig 04_airqual_scatter.png
+    air_quality.plot.scatter(x="station_london", y="station_paris", alpha=0.5)
+    plt.show()
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Apart from the default ``line`` plot when using the ``plot`` function, a
+number of alternatives are available to plot data. Let’s use some
+standard Python to get an overview of the available plot methods:
+
+.. ipython:: python
+
+    [
+        method_name
+        for method_name in dir(air_quality.plot)
+        if not method_name.startswith("_")
+    ]
+
+.. note::
+    In many development environments such as IPython and
+    Jupyter Notebook, use the TAB button to get an overview of the available
+    methods, for example ``air_quality.plot.`` + TAB.
+
+One of the options is :meth:`DataFrame.plot.box`, which refers to a
+`boxplot <https://en.wikipedia.org/wiki/Box_plot>`__. The ``box``
+method is applicable on the air quality example data:
+
+.. ipython:: python
+   :okwarning:
+
+    @savefig 04_airqual_boxplot.png
+    air_quality.plot.box()
+    plt.show()
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For an introduction to plots other than the default line plot, see the user guide section about :ref:`supported plot styles <visualization.other>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want each of the columns in a separate subplot.
+
+.. ipython:: python
+   :okwarning:
+
+    @savefig 04_airqual_area_subplot.png
+    axs = air_quality.plot.area(figsize=(12, 4), subplots=True)
+    plt.show()
+
+Separate subplots for each of the data columns are supported by the ``subplots`` argument
+of the ``plot`` functions. The builtin options available in each of the pandas plot
+functions are worth reviewing.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+Some more formatting options are explained in the user guide section on :ref:`plot formatting <visualization.formatting>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to further customize, extend or save the resulting plot.
+
+.. ipython:: python
+   :okwarning:
+
+    fig, axs = plt.subplots(figsize=(12, 4))
+    air_quality.plot.area(ax=axs)
+    axs.set_ylabel("NO$_2$ concentration")
+    @savefig 04_airqual_customized.png
+    fig.savefig("no2_concentrations.png")
+    plt.show()
+
+.. ipython:: python
+   :suppress:
+   :okwarning:
+
+   import os
+
+   os.remove("no2_concentrations.png")
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Each of the plot objects created by pandas is a
+`Matplotlib <https://matplotlib.org/>`__ object. As Matplotlib provides
+plenty of options to customize plots, making the link between pandas and
+Matplotlib explicit enables all the power of Matplotlib to the plot.
+This strategy is applied in the previous example:
+
+::
+
+   fig, axs = plt.subplots(figsize=(12, 4))        # Create an empty Matplotlib Figure and Axes
+   air_quality.plot.area(ax=axs)                   # Use pandas to put the area plot on the prepared Figure/Axes
+   axs.set_ylabel("NO$_2$ concentration")          # Do any Matplotlib customization you like
+   fig.savefig("no2_concentrations.png")           # Save the Figure/Axes using the existing Matplotlib method.
+   plt.show()                                      # Display the plot
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  The ``.plot.*`` methods are applicable on both Series and DataFrames.
+-  By default, each of the columns is plotted as a different element
+   (line, boxplot, …).
+-  Any plot created by pandas is a Matplotlib object.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full overview of plotting in pandas is provided in the :ref:`visualization pages <visualization>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/05_add_columns.rst b/pandas/doc/source/getting_started/intro_tutorials/05_add_columns.rst
new file mode 100644
index 0000000000000..481c094870e12
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/05_add_columns.rst
@@ -0,0 +1,173 @@
+.. _10min_tut_05_columns:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/air_quality_no2.rst
+
+.. ipython:: python
+
+    air_quality = pd.read_csv("data/air_quality_no2.csv", index_col=0, parse_dates=True)
+    air_quality.head()
+
+.. raw:: html
+
+        </li>
+    </ul>
+    </div>
+
+How to create new columns derived from existing columns
+-------------------------------------------------------
+
+.. image:: ../../_static/schemas/05_newcolumn_1.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to express the :math:`NO_2` concentration of the station in London in mg/m\ :math:`^3`.
+
+(*If we assume temperature of 25 degrees Celsius and pressure of 1013
+hPa, the conversion factor is 1.882*)
+
+.. ipython:: python
+
+    air_quality["london_mg_per_cubic"] = air_quality["station_london"] * 1.882
+    air_quality.head()
+
+To create a new column, use the square brackets ``[]`` with the new column name
+at the left side of the assignment.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+    The calculation of the values is done **element-wise**. This
+    means all values in the given column are multiplied by the value 1.882
+    at once. You do not need to use a loop to iterate each of the rows!
+
+.. image:: ../../_static/schemas/05_newcolumn_2.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to check the ratio of the values in Paris versus Antwerp and save the result in a new column.
+
+.. ipython:: python
+
+    air_quality["ratio_paris_antwerp"] = (
+        air_quality["station_paris"] / air_quality["station_antwerp"]
+    )
+    air_quality.head()
+
+The calculation is again element-wise, so the ``/`` is applied *for the
+values in each row*.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Other mathematical operators (``+``, ``-``, ``*``, ``/``, …) and logical
+operators (``<``, ``>``, ``==``, …) also work element-wise. The latter was already
+used in the :ref:`subset data tutorial <10min_tut_03_subset>` to filter
+rows of a table using a conditional expression.
+
+If you need more advanced logic, you can use arbitrary Python code via :meth:`~DataFrame.apply`.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to rename the data columns to the corresponding station identifiers used by `OpenAQ <https://openaq.org/>`__.
+
+.. ipython:: python
+
+    air_quality_renamed = air_quality.rename(
+        columns={
+            "station_antwerp": "BETR801",
+            "station_paris": "FR04014",
+            "station_london": "London Westminster",
+        }
+    )
+
+.. ipython:: python
+
+    air_quality_renamed.head()
+
+The :meth:`~DataFrame.rename` function can be used for both row labels and column
+labels. Provide a dictionary with the keys the current names and the
+values the new names to update the corresponding names.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The mapping should not be restricted to fixed names only, but can be a
+mapping function as well. For example, converting the column names to
+lowercase letters can be done using a function as well:
+
+.. ipython:: python
+
+    air_quality_renamed = air_quality_renamed.rename(columns=str.lower)
+    air_quality_renamed.head()
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+Details about column or row label renaming is provided in the user guide section on :ref:`renaming labels <basics.rename>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Create a new column by assigning the output to the DataFrame with a
+   new column name in between the ``[]``.
+-  Operations are element-wise, no need to loop over rows.
+-  Use ``rename`` with a dictionary or function to rename row labels or
+   column names.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+The user guide contains a separate section on :ref:`column addition and deletion <basics.dataframe.sel_add_del>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/06_calculate_statistics.rst b/pandas/doc/source/getting_started/intro_tutorials/06_calculate_statistics.rst
new file mode 100644
index 0000000000000..1399ab66426f4
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/06_calculate_statistics.rst
@@ -0,0 +1,286 @@
+.. _10min_tut_06_stats:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/titanic.rst
+
+.. ipython:: python
+
+    titanic = pd.read_csv("data/titanic.csv")
+    titanic.head()
+
+.. raw:: html
+
+            </li>
+        </ul>
+    </div>
+
+How to calculate summary statistics
+-----------------------------------
+
+Aggregating statistics
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/06_aggregate.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the average age of the Titanic passengers?
+
+.. ipython:: python
+
+    titanic["Age"].mean()
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Different statistics are available and can be applied to columns with
+numerical data. Operations in general exclude missing data and operate
+across rows by default.
+
+.. image:: ../../_static/schemas/06_reduction.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the median age and ticket fare price of the Titanic passengers?
+
+.. ipython:: python
+
+    titanic[["Age", "Fare"]].median()
+
+The statistic applied to multiple columns of a ``DataFrame`` (the selection of two columns
+returns a ``DataFrame``, see the :ref:`subset data tutorial <10min_tut_03_subset>`) is calculated for each numeric column.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The aggregating statistic can be calculated for multiple columns at the
+same time. Remember the ``describe`` function from the :ref:`first tutorial <10min_tut_01_tableoriented>`?
+
+.. ipython:: python
+
+    titanic[["Age", "Fare"]].describe()
+
+Instead of the predefined statistics, specific combinations of
+aggregating statistics for given columns can be defined using the
+:func:`DataFrame.agg` method:
+
+.. ipython:: python
+
+    titanic.agg(
+        {
+            "Age": ["min", "max", "median", "skew"],
+            "Fare": ["min", "max", "median", "mean"],
+        }
+    )
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+Details about descriptive statistics are provided in the user guide section on :ref:`descriptive statistics <basics.stats>`.
+
+.. raw:: html
+
+   </div>
+
+
+Aggregating statistics grouped by category
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/06_groupby.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the average age for male versus female Titanic passengers?
+
+.. ipython:: python
+
+    titanic[["Sex", "Age"]].groupby("Sex").mean()
+
+As our interest is the average age for each gender, a subselection on
+these two columns is made first: ``titanic[["Sex", "Age"]]``. Next, the
+:meth:`~DataFrame.groupby` method is applied on the ``Sex`` column to make a group per
+category. The average age *for each gender* is calculated and
+returned.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Calculating a given statistic (e.g. ``mean`` age) *for each category in
+a column* (e.g. male/female in the ``Sex`` column) is a common pattern.
+The ``groupby`` method is used to support this type of operations. This
+fits in the more general ``split-apply-combine`` pattern:
+
+-  **Split** the data into groups
+-  **Apply** a function to each group independently
+-  **Combine** the results into a data structure
+
+The apply and combine steps are typically done together in pandas.
+
+In the previous example, we explicitly selected the 2 columns first. If
+not, the ``mean`` method is applied to each column containing numerical
+columns by passing ``numeric_only=True``:
+
+.. ipython:: python
+
+    titanic.groupby("Sex").mean(numeric_only=True)
+
+It does not make much sense to get the average value of the ``Pclass``.
+If we are only interested in the average age for each gender, the
+selection of columns (square brackets ``[]`` as usual) is supported
+on the grouped data as well:
+
+.. ipython:: python
+
+    titanic.groupby("Sex")["Age"].mean()
+
+.. image:: ../../_static/schemas/06_groupby_select_detail.svg
+   :align: center
+
+.. note::
+    The ``Pclass`` column contains numerical data but actually
+    represents 3 categories (or factors) with respectively the labels ‘1’,
+    ‘2’ and ‘3’. Calculating statistics on these does not make much sense.
+    Therefore, pandas provides a ``Categorical`` data type to handle this
+    type of data. More information is provided in the user guide
+    :ref:`categorical` section.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the mean ticket fare price for each of the sex and cabin class combinations?
+
+.. ipython:: python
+
+    titanic.groupby(["Sex", "Pclass"])["Fare"].mean()
+
+Grouping can be done by multiple columns at the same time. Provide the
+column names as a list to the :meth:`~DataFrame.groupby` method.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full description on the split-apply-combine approach is provided in the user guide section on :ref:`groupby operations <groupby>`.
+
+.. raw:: html
+
+   </div>
+
+Count number of records by category
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/06_valuecounts.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the number of passengers in each of the cabin classes?
+
+.. ipython:: python
+
+    titanic["Pclass"].value_counts()
+
+The :meth:`~Series.value_counts` method counts the number of records for each
+category in a column.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The function is a shortcut, it is actually a groupby operation in combination with counting the number of records
+within each group:
+
+.. ipython:: python
+
+    titanic.groupby("Pclass")["Pclass"].count()
+
+.. note::
+    Both ``size`` and ``count`` can be used in combination with
+    ``groupby``. Whereas ``size`` includes ``NaN`` values and just provides
+    the number of rows (size of the table), ``count`` excludes the missing
+    values. In the ``value_counts`` method, use the ``dropna`` argument to
+    include or exclude the ``NaN`` values.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+The user guide has a dedicated section on ``value_counts`` , see the page on :ref:`discretization <basics.discretization>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Aggregation statistics can be calculated on entire columns or rows.
+-  ``groupby`` provides the power of the *split-apply-combine* pattern.
+-  ``value_counts`` is a convenient shortcut to count the number of
+   entries in each category of a variable.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full description on the split-apply-combine approach is provided in the user guide pages about :ref:`groupby operations <groupby>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/07_reshape_table_layout.rst b/pandas/doc/source/getting_started/intro_tutorials/07_reshape_table_layout.rst
new file mode 100644
index 0000000000000..e4b34e3af57bf
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/07_reshape_table_layout.rst
@@ -0,0 +1,380 @@
+.. _10min_tut_07_reshape:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+
+.. include:: includes/titanic.rst
+
+.. ipython:: python
+
+    titanic = pd.read_csv("data/titanic.csv")
+    titanic.head()
+
+.. raw:: html
+
+        </li>
+        <li class="list-group-item gs-data-list">
+            <div data-bs-toggle="collapse" href="#collapsedata2" role="button" aria-expanded="false" aria-controls="collapsedata2">
+                <span class="badge bg-secondary">Air quality data</span>
+            </div>
+            <div class="collapse" id="collapsedata2">
+                <div class="card-body">
+                    <p class="card-text">
+
+This tutorial uses air quality data about :math:`NO_2` and Particulate matter less than 2.5
+micrometers, made available by
+`OpenAQ <https://openaq.org>`__ and using the
+`py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+The ``air_quality_long.csv`` data set provides :math:`NO_2` and
+:math:`PM_{25}` values for the measurement stations *FR04014*, *BETR801*
+and *London Westminster* in respectively Paris, Antwerp and London.
+
+The air-quality data set has the following columns:
+
+-  city: city where the sensor is used, either Paris, Antwerp or London
+-  country: country where the sensor is used, either FR, BE or GB
+-  location: the id of the sensor, either *FR04014*, *BETR801* or
+   *London Westminster*
+-  parameter: the parameter measured by the sensor, either :math:`NO_2`
+   or Particulate matter
+-  value: the measured value
+-  unit: the unit of the measured parameter, in this case ‘µg/m³’
+
+and the index of the ``DataFrame`` is ``datetime``, the datetime of the
+measurement.
+
+.. note::
+    The air-quality data is provided in a so-called *long format*
+    data representation with each observation on a separate row and each
+    variable a separate column of the data table. The long/narrow format is
+    also known as the `tidy data
+    format <https://www.jstatsoft.org/article/view/v059i10>`__.
+
+.. raw:: html
+
+                    </p>
+                <a href="https://github.com/pandas-dev/pandas/tree/main/doc/data/air_quality_long.csv" class="btn btn-dark btn-sm">To raw data</a>
+            </div>
+        </div>
+
+.. ipython:: python
+
+    air_quality = pd.read_csv(
+        "data/air_quality_long.csv", index_col="date.utc", parse_dates=True
+    )
+    air_quality.head()
+
+.. raw:: html
+
+            </li>
+        </ul>
+    </div>
+
+How to reshape the layout of tables
+-----------------------------------
+
+Sort table rows
+~~~~~~~~~~~~~~~
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to sort the Titanic data according to the age of the passengers.
+
+.. ipython:: python
+
+    titanic.sort_values(by="Age").head()
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to sort the Titanic data according to the cabin class and age in descending order.
+
+.. ipython:: python
+
+    titanic.sort_values(by=['Pclass', 'Age'], ascending=False).head()
+
+With :meth:`DataFrame.sort_values`, the rows in the table are sorted according to the
+defined column(s). The index will follow the row order.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More details about sorting of tables is provided in the user guide section on :ref:`sorting data <basics.sorting>`.
+
+.. raw:: html
+
+   </div>
+
+Long to wide table format
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Let’s use a small subset of the air quality data set. We focus on
+:math:`NO_2` data and only use the first two measurements of each
+location (i.e. the head of each group). The subset of data will be
+called ``no2_subset``.
+
+.. ipython:: python
+
+    # filter for no2 data only
+    no2 = air_quality[air_quality["parameter"] == "no2"]
+
+.. ipython:: python
+
+    # use 2 measurements (head) for each location (groupby)
+    no2_subset = no2.sort_index().groupby(["location"]).head(2)
+    no2_subset
+
+.. image:: ../../_static/schemas/07_pivot.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want the values for the three stations as separate columns next to each other.
+
+.. ipython:: python
+
+    no2_subset.pivot(columns="location", values="value")
+
+The :meth:`~pandas.pivot` function is purely reshaping of the data: a single value
+for each index/column combination is required.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+As pandas supports plotting of multiple columns (see :ref:`plotting tutorial <10min_tut_04_plotting>`) out of the box, the conversion from
+*long* to *wide* table format enables the plotting of the different time
+series at the same time:
+
+.. ipython:: python
+
+    no2.head()
+
+.. ipython:: python
+
+    @savefig 7_reshape_columns.png
+    no2.pivot(columns="location", values="value").plot()
+
+.. note::
+    When the ``index`` parameter is not defined, the existing
+    index (row labels) is used.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For more information about :meth:`~DataFrame.pivot`, see the user guide section on :ref:`pivoting DataFrame objects <reshaping.reshaping>`.
+
+.. raw:: html
+
+   </div>
+
+Pivot table
+~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/07_pivot_table.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want the mean concentrations for :math:`NO_2` and :math:`PM_{2.5}` in each of the stations in table form.
+
+.. ipython:: python
+
+    air_quality.pivot_table(
+        values="value", index="location", columns="parameter", aggfunc="mean"
+    )
+
+In the case of :meth:`~DataFrame.pivot`, the data is only rearranged. When multiple
+values need to be aggregated (in this specific case, the values on
+different time steps), :meth:`~DataFrame.pivot_table` can be used, providing an
+aggregation function (e.g. mean) on how to combine these values.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Pivot table is a well known concept in spreadsheet software. When
+interested in the row/column margins (subtotals) for each variable, set
+the ``margins`` parameter to ``True``:
+
+.. ipython:: python
+
+    air_quality.pivot_table(
+        values="value",
+        index="location",
+        columns="parameter",
+        aggfunc="mean",
+        margins=True,
+    )
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+For more information about :meth:`~DataFrame.pivot_table`, see the user guide section on :ref:`pivot tables <reshaping.pivot>`.
+
+.. raw:: html
+
+   </div>
+
+.. note::
+    In case you are wondering, :meth:`~DataFrame.pivot_table` is indeed directly linked
+    to :meth:`~DataFrame.groupby`. The same result can be derived by grouping on both
+    ``parameter`` and ``location``:
+
+    ::
+
+        air_quality.groupby(["parameter", "location"])[["value"]].mean()
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+.. raw:: html
+
+   </div>
+
+Wide to long format
+~~~~~~~~~~~~~~~~~~~
+
+Starting again from the wide format table created in the previous
+section, we add a new index to the ``DataFrame`` with :meth:`~DataFrame.reset_index`.
+
+.. ipython:: python
+
+    no2_pivoted = no2.pivot(columns="location", values="value").reset_index()
+    no2_pivoted.head()
+
+.. image:: ../../_static/schemas/07_melt.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to collect all air quality :math:`NO_2` measurements in a single column (long format).
+
+.. ipython:: python
+
+    no_2 = no2_pivoted.melt(id_vars="date.utc")
+    no_2.head()
+
+The :func:`pandas.melt` method on a ``DataFrame`` converts the data table from wide
+format to long format. The column headers become the variable names in a
+newly created column.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The solution is the short version on how to apply :func:`pandas.melt`. The method
+will *melt* all columns NOT mentioned in ``id_vars`` together into two
+columns: A column with the column header names and a column with the
+values itself. The latter column gets by default the name ``value``.
+
+The parameters passed to :func:`pandas.melt` can be defined in more detail:
+
+.. ipython:: python
+
+    no_2 = no2_pivoted.melt(
+        id_vars="date.utc",
+        value_vars=["BETR801", "FR04014", "London Westminster"],
+        value_name="NO_2",
+        var_name="id_location",
+    )
+    no_2.head()
+
+The additional parameters have the following effects:
+
+-  ``value_vars`` defines which columns to *melt* together
+-  ``value_name`` provides a custom column name for the values column
+   instead of the default column name ``value``
+-  ``var_name`` provides a custom column name for the column collecting
+   the column header names. Otherwise it takes the index name or a
+   default ``variable``
+
+Hence, the arguments ``value_name`` and ``var_name`` are just
+user-defined names for the two generated columns. The columns to melt
+are defined by ``id_vars`` and ``value_vars``.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+Conversion from wide to long format with :func:`pandas.melt` is explained in the user guide section on :ref:`reshaping by melt <reshaping.melt>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Sorting by one or more columns is supported by ``sort_values``.
+-  The ``pivot`` function is purely restructuring of the data,
+   ``pivot_table`` supports aggregations.
+-  The reverse of ``pivot`` (long to wide format) is ``melt`` (wide to
+   long format).
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full overview is available in the user guide on the pages about :ref:`reshaping and pivoting <reshaping>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/08_combine_dataframes.rst b/pandas/doc/source/getting_started/intro_tutorials/08_combine_dataframes.rst
new file mode 100644
index 0000000000000..024300bb8a9b0
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/08_combine_dataframes.rst
@@ -0,0 +1,321 @@
+.. _10min_tut_08_combine:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+                <div data-bs-toggle="collapse" href="#collapsedata" role="button" aria-expanded="false" aria-controls="collapsedata">
+                    <span class="badge bg-secondary">Air quality Nitrate data</span>
+                </div>
+                <div class="collapse" id="collapsedata">
+                    <div class="card-body">
+                        <p class="card-text">
+
+For this tutorial, air quality data about :math:`NO_2` is used, made available by
+`OpenAQ <https://openaq.org>`__ and downloaded using the
+`py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+
+The ``air_quality_no2_long.csv`` data set provides :math:`NO_2`
+values for the measurement stations *FR04014*, *BETR801* and *London
+Westminster* in respectively Paris, Antwerp and London.
+
+.. raw:: html
+
+                        </p>
+                    <a href="https://github.com/pandas-dev/pandas/tree/main/doc/data/air_quality_no2_long.csv" class="btn btn-dark btn-sm">To raw data</a>
+                </div>
+            </div>
+
+.. ipython:: python
+
+    air_quality_no2 = pd.read_csv("data/air_quality_no2_long.csv",
+                                  parse_dates=True)
+    air_quality_no2 = air_quality_no2[["date.utc", "location",
+                                       "parameter", "value"]]
+    air_quality_no2.head()
+
+.. raw:: html
+
+        </li>
+        <li class="list-group-item gs-data-list">
+            <div data-bs-toggle="collapse" href="#collapsedata2" role="button" aria-expanded="false" aria-controls="collapsedata2">
+                <span class="badge bg-secondary">Air quality Particulate matter data</span>
+            </div>
+            <div class="collapse" id="collapsedata2">
+                <div class="card-body">
+                    <p class="card-text">
+
+For this tutorial, air quality data about Particulate
+matter less than 2.5 micrometers is used, made available by
+`OpenAQ <https://openaq.org>`__ and downloaded using the
+`py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+
+The ``air_quality_pm25_long.csv`` data set provides :math:`PM_{25}`
+values for the measurement stations *FR04014*, *BETR801* and *London
+Westminster* in respectively Paris, Antwerp and London.
+
+.. raw:: html
+
+                    </p>
+                <a href="https://github.com/pandas-dev/pandas/tree/main/doc/data/air_quality_pm25_long.csv" class="btn btn-dark btn-sm">To raw data</a>
+            </div>
+        </div>
+
+.. ipython:: python
+
+    air_quality_pm25 = pd.read_csv("data/air_quality_pm25_long.csv",
+                                   parse_dates=True)
+    air_quality_pm25 = air_quality_pm25[["date.utc", "location",
+                                         "parameter", "value"]]
+    air_quality_pm25.head()
+
+.. raw:: html
+
+        </li>
+    </ul>
+    </div>
+
+
+How to combine data from multiple tables
+----------------------------------------
+
+Concatenating objects
+~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/08_concat_row.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to combine the measurements of :math:`NO_2` and :math:`PM_{25}`, two tables with a similar structure, in a single table.
+
+.. ipython:: python
+
+    air_quality = pd.concat([air_quality_pm25, air_quality_no2], axis=0)
+    air_quality.head()
+
+The :func:`~pandas.concat` function performs concatenation operations of multiple
+tables along one of the axes (row-wise or column-wise).
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+By default concatenation is along axis 0, so the resulting table combines the rows
+of the input tables. Let’s check the shape of the original and the
+concatenated tables to verify the operation:
+
+.. ipython:: python
+
+    print('Shape of the ``air_quality_pm25`` table: ', air_quality_pm25.shape)
+    print('Shape of the ``air_quality_no2`` table: ', air_quality_no2.shape)
+    print('Shape of the resulting ``air_quality`` table: ', air_quality.shape)
+
+Hence, the resulting table has 3178 = 1110 + 2068 rows.
+
+.. note::
+    The **axis** argument will return in a number of pandas
+    methods that can be applied **along an axis**. A ``DataFrame`` has two
+    corresponding axes: the first running vertically downwards across rows
+    (axis 0), and the second running horizontally across columns (axis 1).
+    Most operations like concatenation or summary statistics are by default
+    across rows (axis 0), but can be applied across columns as well.
+
+Sorting the table on the datetime information also illustrates the
+combination of both tables, with the ``parameter`` column defining the
+origin of the table (either ``no2`` from table ``air_quality_no2`` or
+``pm25`` from table ``air_quality_pm25``):
+
+.. ipython:: python
+
+    air_quality = air_quality.sort_values("date.utc")
+    air_quality.head()
+
+In this specific example, the ``parameter`` column provided by the data
+ensures that each of the original tables can be identified. This is not
+always the case. The ``concat`` function provides a convenient solution
+with the ``keys`` argument, adding an additional (hierarchical) row
+index. For example:
+
+.. ipython:: python
+
+    air_quality_ = pd.concat([air_quality_pm25, air_quality_no2], keys=["PM25", "NO2"])
+    air_quality_.head()
+
+.. note::
+    The existence of multiple row/column indices at the same time
+    has not been mentioned within these tutorials. *Hierarchical indexing*
+    or *MultiIndex* is an advanced and powerful pandas feature to analyze
+    higher dimensional data.
+
+    Multi-indexing is out of scope for this pandas introduction. For the
+    moment, remember that the function ``reset_index`` can be used to
+    convert any level of an index to a column, e.g.
+    ``air_quality.reset_index(level=0)``
+
+    .. raw:: html
+
+        <div class="d-flex flex-row  gs-torefguide">
+            <span class="badge badge-info">To user guide</span>
+
+    Feel free to dive into the world of multi-indexing at the user guide section on :ref:`advanced indexing <advanced>`.
+
+    .. raw:: html
+
+        </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More options on table concatenation (row and column
+wise) and how ``concat`` can be used to define the logic (union or
+intersection) of the indexes on the other axes is provided at the section on
+:ref:`object concatenation <merging.concat>`.
+
+.. raw:: html
+
+    </div>
+
+Join tables using a common identifier
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. image:: ../../_static/schemas/08_merge_left.svg
+   :align: center
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Add the station coordinates, provided by the stations metadata table, to the corresponding rows in the measurements table.
+
+.. warning::
+    The air quality measurement station coordinates are stored in a data
+    file ``air_quality_stations.csv``, downloaded using the
+    `py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+
+.. ipython:: python
+
+    stations_coord = pd.read_csv("data/air_quality_stations.csv")
+    stations_coord.head()
+
+.. note::
+    The stations used in this example (FR04014, BETR801 and London
+    Westminster) are just three entries enlisted in the metadata table. We
+    only want to add the coordinates of these three to the measurements
+    table, each on the corresponding rows of the ``air_quality`` table.
+
+.. ipython:: python
+
+    air_quality.head()
+
+.. ipython:: python
+
+    air_quality = pd.merge(air_quality, stations_coord, how="left", on="location")
+    air_quality.head()
+
+Using the :meth:`~pandas.merge` function, for each of the rows in the
+``air_quality`` table, the corresponding coordinates are added from the
+``air_quality_stations_coord`` table. Both tables have the column
+``location`` in common which is used as a key to combine the
+information. By choosing the ``left`` join, only the locations available
+in the ``air_quality`` (left) table, i.e. FR04014, BETR801 and London
+Westminster, end up in the resulting table. The ``merge`` function
+supports multiple join options similar to database-style operations.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Add the parameters' full description and name, provided by the parameters metadata table, to the measurements table.
+
+.. warning::
+    The air quality parameters metadata are stored in a data file
+    ``air_quality_parameters.csv``, downloaded using the
+    `py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+
+.. ipython:: python
+
+    air_quality_parameters = pd.read_csv("data/air_quality_parameters.csv")
+    air_quality_parameters.head()
+
+.. ipython:: python
+
+    air_quality = pd.merge(air_quality, air_quality_parameters,
+                           how='left', left_on='parameter', right_on='id')
+    air_quality.head()
+
+Compared to the previous example, there is no common column name.
+However, the ``parameter`` column in the ``air_quality`` table and the
+``id`` column in the ``air_quality_parameters`` table both provide the
+measured variable in a common format. The ``left_on`` and ``right_on``
+arguments are used here (instead of just ``on``) to make the link
+between the two tables.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+pandas also supports inner, outer, and right joins.
+More information on join/merge of tables is provided in the user guide section on
+:ref:`database style merging of tables <merging.join>`. Or have a look at the
+:ref:`comparison with SQL<compare_with_sql.join>` page.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Multiple tables can be concatenated column-wise or row-wise using
+   the ``concat`` function.
+-  For database-like merging/joining of tables, use the ``merge``
+   function.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+See the user guide for a full description of the various :ref:`facilities to combine data tables <merging>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/09_timeseries.rst b/pandas/doc/source/getting_started/intro_tutorials/09_timeseries.rst
new file mode 100644
index 0000000000000..6ba3c17fac3c3
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/09_timeseries.rst
@@ -0,0 +1,390 @@
+.. _10min_tut_09_timeseries:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+    import matplotlib.pyplot as plt
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+                <div data-bs-toggle="collapse" href="#collapsedata" role="button" aria-expanded="false" aria-controls="collapsedata">
+                    <span class="badge bg-secondary">Air quality data</span>
+                </div>
+                <div class="collapse" id="collapsedata">
+                    <div class="card-body">
+                        <p class="card-text">
+
+For this tutorial, air quality data about :math:`NO_2` and Particulate
+matter less than 2.5 micrometers is used, made available by
+`OpenAQ <https://openaq.org>`__ and downloaded using the
+`py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+The ``air_quality_no2_long.csv"`` data set provides :math:`NO_2` values
+for the measurement stations *FR04014*, *BETR801* and *London
+Westminster* in respectively Paris, Antwerp and London.
+
+.. raw:: html
+
+                        </p>
+                    <a href="https://github.com/pandas-dev/pandas/tree/main/doc/data/air_quality_no2_long.csv" class="btn btn-dark btn-sm">To raw data</a>
+                </div>
+            </div>
+
+.. ipython:: python
+
+    air_quality = pd.read_csv("data/air_quality_no2_long.csv")
+    air_quality = air_quality.rename(columns={"date.utc": "datetime"})
+    air_quality.head()
+
+.. ipython:: python
+
+    air_quality.city.unique()
+
+.. raw:: html
+
+        </li>
+    </ul>
+    </div>
+
+How to handle time series data with ease
+----------------------------------------
+
+.. _10min_tut_09_timeseries.properties:
+
+Using pandas datetime properties
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to work with the dates in the column ``datetime`` as datetime objects instead of plain text
+
+.. ipython:: python
+
+    air_quality["datetime"] = pd.to_datetime(air_quality["datetime"])
+    air_quality["datetime"]
+
+Initially, the values in ``datetime`` are character strings and do not
+provide any datetime operations (e.g. extract the year, day of the
+week, …). By applying the ``to_datetime`` function, pandas interprets the
+strings and convert these to datetime (i.e. ``datetime64[ns, UTC]``)
+objects. In pandas we call these datetime objects that are similar to
+``datetime.datetime`` from the standard library as :class:`pandas.Timestamp`.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+    As many data sets do contain datetime information in one of
+    the columns, pandas input function like :func:`pandas.read_csv` and :func:`pandas.read_json`
+    can do the transformation to dates when reading the data using the
+    ``parse_dates`` parameter with a list of the columns to read as
+    Timestamp:
+
+    ::
+
+        pd.read_csv("../data/air_quality_no2_long.csv", parse_dates=["datetime"])
+
+Why are these :class:`pandas.Timestamp` objects useful? Let’s illustrate the added
+value with some example cases.
+
+   What is the start and end date of the time series data set we are working
+   with?
+
+.. ipython:: python
+
+    air_quality["datetime"].min(), air_quality["datetime"].max()
+
+Using :class:`pandas.Timestamp` for datetimes enables us to calculate with date
+information and make them comparable. Hence, we can use this to get the
+length of our time series:
+
+.. ipython:: python
+
+    air_quality["datetime"].max() - air_quality["datetime"].min()
+
+The result is a :class:`pandas.Timedelta` object, similar to ``datetime.timedelta``
+from the standard Python library which defines a time duration.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+The various time concepts supported by pandas are explained in the user guide section on :ref:`time related concepts <timeseries.overview>`.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+I want to add a new column to the ``DataFrame`` containing only the month of the measurement
+
+.. ipython:: python
+
+    air_quality["month"] = air_quality["datetime"].dt.month
+    air_quality.head()
+
+By using ``Timestamp`` objects for dates, a lot of time-related
+properties are provided by pandas. For example the ``month``, but also
+``year``, ``quarter``,… All of these properties are
+accessible by the ``dt`` accessor.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+An overview of the existing date properties is given in the
+:ref:`time and date components overview table <timeseries.components>`. More details about the ``dt`` accessor
+to return datetime like properties are explained in a dedicated section on the  :ref:`dt accessor <basics.dt_accessors>`.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+What is the average :math:`NO_2` concentration for each day of the week for each of the measurement locations?
+
+.. ipython:: python
+
+    air_quality.groupby(
+        [air_quality["datetime"].dt.weekday, "location"])["value"].mean()
+
+Remember the split-apply-combine pattern provided by ``groupby`` from the
+:ref:`tutorial on statistics calculation <10min_tut_06_stats>`?
+Here, we want to calculate a given statistic (e.g. mean :math:`NO_2`)
+**for each weekday** and **for each measurement location**. To group on
+weekdays, we use the datetime property ``weekday`` (with Monday=0 and
+Sunday=6) of pandas ``Timestamp``, which is also accessible by the
+``dt`` accessor. The grouping on both locations and weekdays can be done
+to split the calculation of the mean on each of these combinations.
+
+.. danger::
+    As we are working with a very short time series in these
+    examples, the analysis does not provide a long-term representative
+    result!
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Plot the typical :math:`NO_2` pattern during the day of our time series of all stations together. In other words, what is the average value for each hour of the day?
+
+.. ipython:: python
+
+    fig, axs = plt.subplots(figsize=(12, 4))
+    air_quality.groupby(air_quality["datetime"].dt.hour)["value"].mean().plot(
+        kind='bar', rot=0, ax=axs
+    )
+    plt.xlabel("Hour of the day");  # custom x label using Matplotlib
+    @savefig 09_bar_chart.png
+    plt.ylabel("$NO_2 (µg/m^3)$");
+
+Similar to the previous case, we want to calculate a given statistic
+(e.g. mean :math:`NO_2`) **for each hour of the day** and we can use the
+split-apply-combine approach again. For this case, we use the datetime property ``hour``
+of pandas ``Timestamp``, which is also accessible by the ``dt`` accessor.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Datetime as index
+~~~~~~~~~~~~~~~~~
+
+In the :ref:`tutorial on reshaping <10min_tut_07_reshape>`,
+:meth:`~pandas.pivot` was introduced to reshape the data table with each of the
+measurements locations as a separate column:
+
+.. ipython:: python
+
+    no_2 = air_quality.pivot(index="datetime", columns="location", values="value")
+    no_2.head()
+
+.. note::
+    By pivoting the data, the datetime information became the
+    index of the table. In general, setting a column as an index can be
+    achieved by the ``set_index`` function.
+
+Working with a datetime index (i.e. ``DatetimeIndex``) provides powerful
+functionalities. For example, we do not need the ``dt`` accessor to get
+the time series properties, but have these properties available on the
+index directly:
+
+.. ipython:: python
+
+    no_2.index.year, no_2.index.weekday
+
+Some other advantages are the convenient subsetting of time period or
+the adapted time scale on plots. Let’s apply this on our data.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Create a plot of the :math:`NO_2` values in the different stations from May 20th till the end of May 21st.
+
+.. ipython:: python
+    :okwarning:
+
+    @savefig 09_time_section.png
+    no_2["2019-05-20":"2019-05-21"].plot();
+
+By providing a **string that parses to a datetime**, a specific subset of the data can be selected on a ``DatetimeIndex``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More information on the ``DatetimeIndex`` and the slicing by using strings is provided in the section on :ref:`time series indexing <timeseries.datetimeindex>`.
+
+.. raw:: html
+
+    </div>
+
+Resample a time series to another frequency
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Aggregate the current hourly time series values to the monthly maximum value in each of the stations.
+
+.. ipython:: python
+
+    monthly_max = no_2.resample("MS").max()
+    monthly_max
+
+A very powerful method on time series data with a datetime index, is the
+ability to :meth:`~Series.resample` time series to another frequency (e.g.,
+converting secondly data into 5-minutely data).
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+The :meth:`~Series.resample` method is similar to a groupby operation:
+
+-  it provides a time-based grouping, by using a string (e.g. ``M``,
+   ``5H``, …) that defines the target frequency
+-  it requires an aggregation function such as ``mean``, ``max``,…
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+An overview of the aliases used to define time series frequencies is given in the :ref:`offset aliases overview table <timeseries.offset_aliases>`.
+
+.. raw:: html
+
+    </div>
+
+When defined, the frequency of the time series is provided by the
+``freq`` attribute:
+
+.. ipython:: python
+
+    monthly_max.index.freq
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Make a plot of the daily mean :math:`NO_2` value in each of the stations.
+
+.. ipython:: python
+    :okwarning:
+
+    @savefig 09_resample_mean.png
+    no_2.resample("D").mean().plot(style="-o", figsize=(10, 5));
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More details on the power of time series ``resampling`` is provided in the user guide section on :ref:`resampling <timeseries.resampling>`.
+
+.. raw:: html
+
+    </div>
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  Valid date strings can be converted to datetime objects using
+   ``to_datetime`` function or as part of read functions.
+-  Datetime objects in pandas support calculations, logical operations
+   and convenient date-related properties using the ``dt`` accessor.
+-  A ``DatetimeIndex`` contains these date-related properties and
+   supports convenient slicing.
+-  ``Resample`` is a powerful method to change the frequency of a time
+   series.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full overview on time series is given on the pages on :ref:`time series and date functionality <timeseries>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/10_text_data.rst b/pandas/doc/source/getting_started/intro_tutorials/10_text_data.rst
new file mode 100644
index 0000000000000..8493a071863c4
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/10_text_data.rst
@@ -0,0 +1,248 @@
+.. _10min_tut_10_text:
+
+{{ header }}
+
+.. ipython:: python
+
+    import pandas as pd
+
+.. raw:: html
+
+    <div class="card gs-data">
+        <div class="card-header gs-data-header">
+            <div class="gs-data-title">
+                Data used for this tutorial:
+            </div>
+        </div>
+        <ul class="list-group list-group-flush">
+            <li class="list-group-item gs-data-list">
+.. include:: includes/titanic.rst
+
+.. ipython:: python
+
+    titanic = pd.read_csv("data/titanic.csv")
+    titanic.head()
+
+.. raw:: html
+
+            </li>
+        </ul>
+    </div>
+
+How to manipulate textual data
+------------------------------
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Make all name characters lowercase.
+
+.. ipython:: python
+
+    titanic["Name"].str.lower()
+
+To make each of the strings in the ``Name`` column lowercase, select the ``Name`` column
+(see the :ref:`tutorial on selection of data <10min_tut_03_subset>`), add the ``str`` accessor and
+apply the ``lower`` method. As such, each of the strings is converted element-wise.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+Similar to datetime objects in the :ref:`time series tutorial <10min_tut_09_timeseries>`
+having a ``dt`` accessor, a number of
+specialized string methods are available when using the ``str``
+accessor. These methods have in general matching names with the
+equivalent built-in string methods for single elements, but are applied
+element-wise (remember :ref:`element-wise calculations <10min_tut_05_columns>`?)
+on each of the values of the columns.
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Create a new column ``Surname`` that contains the surname of the passengers by extracting the part before the comma.
+
+.. ipython:: python
+
+    titanic["Name"].str.split(",")
+
+Using the :meth:`Series.str.split` method, each of the values is returned as a list of
+2 elements. The first element is the part before the comma and the
+second element is the part after the comma.
+
+.. ipython:: python
+
+    titanic["Surname"] = titanic["Name"].str.split(",").str.get(0)
+    titanic["Surname"]
+
+As we are only interested in the first part representing the surname
+(element 0), we can again use the ``str`` accessor and apply :meth:`Series.str.get` to
+extract the relevant part. Indeed, these string functions can be
+concatenated to combine multiple functions at once!
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More information on extracting parts of strings is available in the user guide section on :ref:`splitting and replacing strings <text.split>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Extract the passenger data about the countesses on board of the Titanic.
+
+.. ipython:: python
+
+    titanic["Name"].str.contains("Countess")
+
+.. ipython:: python
+
+    titanic[titanic["Name"].str.contains("Countess")]
+
+(*Interested in her story? See* `Wikipedia <https://en.wikipedia.org/wiki/No%C3%ABl_Leslie,_Countess_of_Rothes>`__\ *!*)
+
+The string method :meth:`Series.str.contains` checks for each of the values in the
+column ``Name`` if the string contains the word ``Countess`` and returns
+for each of the values ``True`` (``Countess`` is part of the name) or
+``False`` (``Countess`` is not part of the name). This output can be used
+to subselect the data using conditional (boolean) indexing introduced in
+the :ref:`subsetting of data tutorial <10min_tut_03_subset>`. As there was
+only one countess on the Titanic, we get one row as a result.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. note::
+    More powerful extractions on strings are supported, as the
+    :meth:`Series.str.contains` and :meth:`Series.str.extract` methods accept `regular
+    expressions <https://docs.python.org/3/library/re.html>`__, but are out of
+    the scope of this tutorial.
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+More information on extracting parts of strings is available in the user guide section on :ref:`string matching and extracting <text.extract>`.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+Which passenger of the Titanic has the longest name?
+
+.. ipython:: python
+
+    titanic["Name"].str.len()
+
+To get the longest name we first have to get the lengths of each of the
+names in the ``Name`` column. By using pandas string methods, the
+:meth:`Series.str.len` function is applied to each of the names individually
+(element-wise).
+
+.. ipython:: python
+
+    titanic["Name"].str.len().idxmax()
+
+Next, we need to get the corresponding location, preferably the index
+label, in the table for which the name length is the largest. The
+:meth:`~Series.idxmax` method does exactly that. It is not a string method and is
+applied to integers, so no ``str`` is used.
+
+.. ipython:: python
+
+    titanic.loc[titanic["Name"].str.len().idxmax(), "Name"]
+
+Based on the index name of the row (``307``) and the column (``Name``),
+we can do a selection using the ``loc`` operator, introduced in the
+:ref:`tutorial on subsetting <10min_tut_03_subset>`.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. raw:: html
+
+    <ul class="task-bullet">
+        <li>
+
+In the "Sex" column, replace values of "male" by "M" and values of "female" by "F".
+
+.. ipython:: python
+
+    titanic["Sex_short"] = titanic["Sex"].replace({"male": "M", "female": "F"})
+    titanic["Sex_short"]
+
+Whereas :meth:`~Series.replace` is not a string method, it provides a convenient way
+to use mappings or vocabularies to translate certain values. It requires
+a ``dictionary`` to define the mapping ``{from: to}``.
+
+.. raw:: html
+
+        </li>
+    </ul>
+
+.. warning::
+    There is also a :meth:`~Series.str.replace` method available to replace a
+    specific set of characters. However, when having a mapping of multiple
+    values, this would become:
+
+    ::
+
+        titanic["Sex_short"] = titanic["Sex"].str.replace("female", "F")
+        titanic["Sex_short"] = titanic["Sex_short"].str.replace("male", "M")
+
+    This would become cumbersome and easily lead to mistakes. Just think (or
+    try out yourself) what would happen if those two statements are applied
+    in the opposite order…
+
+.. raw:: html
+
+    <div class="shadow gs-callout gs-callout-remember">
+        <h4>REMEMBER</h4>
+
+-  String methods are available using the ``str`` accessor.
+-  String methods work element-wise and can be used for conditional
+   indexing.
+-  The ``replace`` method is a convenient method to convert values
+   according to a given dictionary.
+
+.. raw:: html
+
+   </div>
+
+.. raw:: html
+
+    <div class="d-flex flex-row gs-torefguide">
+        <span class="badge badge-info">To user guide</span>
+
+A full overview is provided in the user guide pages on :ref:`working with text data <text>`.
+
+.. raw:: html
+
+   </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/includes/air_quality_no2.rst b/pandas/doc/source/getting_started/intro_tutorials/includes/air_quality_no2.rst
new file mode 100644
index 0000000000000..b699dd9b23f68
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/includes/air_quality_no2.rst
@@ -0,0 +1,22 @@
+.. raw:: html
+
+    <div data-bs-toggle="collapse" href="#collapsedata" role="button" aria-expanded="false" aria-controls="collapsedata">
+        <span class="badge bg-secondary">Air quality data</span>
+    </div>
+    <div class="collapse" id="collapsedata">
+        <div class="card-body">
+            <p class="card-text">
+
+For this tutorial, air quality data about :math:`NO_2` is used, made
+available by `OpenAQ <https://openaq.org>`__ and using the
+`py-openaq <http://dhhagan.github.io/py-openaq/index.html>`__ package.
+The ``air_quality_no2.csv`` data set provides :math:`NO_2` values for
+the measurement stations *FR04014*, *BETR801* and *London Westminster*
+in respectively Paris, Antwerp and London.
+
+.. raw:: html
+
+            </p>
+            <a href="https://github.com/pandas-dev/pandas/tree/main/doc/data/air_quality_no2.csv" class="btn btn-dark btn-sm">To raw data</a>
+        </div>
+    </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/includes/titanic.rst b/pandas/doc/source/getting_started/intro_tutorials/includes/titanic.rst
new file mode 100644
index 0000000000000..6e03b848aab06
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/includes/titanic.rst
@@ -0,0 +1,31 @@
+.. raw:: html
+
+    <div data-bs-toggle="collapse" href="#collapsedata" role="button" aria-expanded="false" aria-controls="collapsedata">
+        <span class="badge bg-secondary">Titanic data</span>
+    </div>
+    <div class="collapse" id="collapsedata">
+        <div class="card-body">
+            <p class="card-text">
+
+This tutorial uses the Titanic data set, stored as CSV. The data
+consists of the following data columns:
+
+-  PassengerId: Id of every passenger.
+-  Survived: Indication whether passenger survived. ``0`` for yes and ``1`` for no.
+-  Pclass: One out of the 3 ticket classes: Class ``1``, Class ``2`` and Class ``3``.
+-  Name: Name of passenger.
+-  Sex: Gender of passenger.
+-  Age: Age of passenger in years.
+-  SibSp: Number of siblings or spouses aboard.
+-  Parch: Number of parents or children aboard.
+-  Ticket: Ticket number of passenger.
+-  Fare: Indicating the fare.
+-  Cabin: Cabin number of passenger.
+-  Embarked: Port of embarkation.
+
+.. raw:: html
+
+            </p>
+            <a href="https://github.com/pandas-dev/pandas/raw/main/doc/data/titanic.csv" class="btn btn-dark btn-sm">To raw data</a>
+        </div>
+    </div>
diff --git a/pandas/doc/source/getting_started/intro_tutorials/index.rst b/pandas/doc/source/getting_started/intro_tutorials/index.rst
new file mode 100644
index 0000000000000..c67e18043c175
--- /dev/null
+++ b/pandas/doc/source/getting_started/intro_tutorials/index.rst
@@ -0,0 +1,21 @@
+{{ header }}
+
+.. _10times1minute:
+
+=========================
+Getting started tutorials
+=========================
+
+.. toctree::
+    :maxdepth: 1
+
+    01_table_oriented
+    02_read_write
+    03_subset_data
+    04_plotting
+    05_add_columns
+    06_calculate_statistics
+    07_reshape_table_layout
+    08_combine_dataframes
+    09_timeseries
+    10_text_data
diff --git a/pandas/doc/source/getting_started/overview.rst b/pandas/doc/source/getting_started/overview.rst
new file mode 100644
index 0000000000000..a8b7a387d80ec
--- /dev/null
+++ b/pandas/doc/source/getting_started/overview.rst
@@ -0,0 +1,176 @@
+.. _overview:
+
+{{ header }}
+
+****************
+Package overview
+****************
+
+pandas is a `Python <https://www.python.org>`__ package that provides fast,
+flexible, and expressive data structures designed to make working with
+"relational" or "labeled" data both easy and intuitive. It aims to be the
+fundamental high-level building block for Python's practical, **real-world** data
+analysis. Additionally, it seeks to become **the
+most powerful and flexible open source data analysis/manipulation tool
+available in any language**. It is already well on its way toward this goal.
+
+pandas is well suited for many different kinds of data:
+
+  - Tabular data with heterogeneously-typed columns, as in an SQL table or
+    Excel spreadsheet
+  - Ordered and unordered (not necessarily fixed-frequency) time series data.
+  - Arbitrary matrix data (homogeneously typed or heterogeneous) with row and
+    column labels
+  - Any other form of observational / statistical data sets. The data
+    need not be labeled at all to be placed into a pandas data structure
+
+The two primary data structures of pandas, :class:`Series` (1-dimensional)
+and :class:`DataFrame` (2-dimensional), handle the vast majority of typical use
+cases in finance, statistics, social science, and many areas of
+engineering. For R users, :class:`DataFrame` provides everything that R's
+``data.frame`` provides and much more. pandas is built on top of `NumPy
+<https://numpy.org>`__ and is intended to integrate well within a scientific
+computing environment with many other 3rd party libraries.
+
+Here are just a few of the things that pandas does well:
+
+  - Easy handling of **missing data** (represented as NaN) in floating point as
+    well as non-floating point data
+  - Size mutability: columns can be **inserted and deleted** from DataFrame and
+    higher dimensional objects
+  - Automatic and explicit **data alignment**: objects can be explicitly
+    aligned to a set of labels, or the user can simply ignore the labels and
+    let ``Series``, ``DataFrame``, etc. automatically align the data for you in
+    computations
+  - Powerful, flexible **group by** functionality to perform
+    split-apply-combine operations on data sets, for both aggregating and
+    transforming data
+  - Make it **easy to convert** ragged, differently-indexed data in other
+    Python and NumPy data structures into DataFrame objects
+  - Intelligent label-based **slicing**, **fancy indexing**, and **subsetting**
+    of large data sets
+  - Intuitive **merging** and **joining** data sets
+  - Flexible **reshaping** and pivoting of data sets
+  - **Hierarchical** labeling of axes (possible to have multiple labels per
+    tick)
+  - Robust IO tools for loading data from **flat files** (CSV and delimited),
+    Excel files, databases, and saving / loading data from the ultrafast **HDF5
+    format**
+  - **Time series**-specific functionality: date range generation and frequency
+    conversion, moving window statistics, date shifting, and lagging.
+
+Many of these principles are here to address the shortcomings frequently
+experienced using other languages / scientific research environments. For data
+scientists, working with data is typically divided into multiple stages:
+munging and cleaning data, analyzing / modeling it, then organizing the results
+of the analysis into a form suitable for plotting or tabular display. pandas
+is the ideal tool for all of these tasks.
+
+Some other notes
+
+ - pandas is **fast**. Many of the low-level algorithmic bits have been
+   extensively tweaked in `Cython <https://cython.org>`__ code. However, as with
+   anything else generalization usually sacrifices performance. So if you focus
+   on one feature for your application you may be able to create a faster
+   specialized tool.
+
+ - pandas is a dependency of `statsmodels
+   <https://www.statsmodels.org/>`__, making it an important part of the
+   statistical computing ecosystem in Python.
+
+ - pandas has been used extensively in production in financial applications.
+
+Data structures
+---------------
+
+.. csv-table::
+    :header: "Dimensions", "Name", "Description"
+    :widths: 15, 20, 50
+
+    1, "Series", "1D labeled homogeneously-typed array"
+    2, "DataFrame", "General 2D labeled, size-mutable tabular structure with potentially heterogeneously-typed column"
+
+Why more than one data structure?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The best way to think about the pandas data structures is as flexible
+containers for lower dimensional data. For example, DataFrame is a container
+for Series, and Series is a container for scalars. We would like to be
+able to insert and remove objects from these containers in a dictionary-like
+fashion.
+
+Also, we would like sensible default behaviors for the common API functions
+which take into account the typical orientation of time series and
+cross-sectional data sets. When using the N-dimensional array (ndarrays) to store 2- and 3-dimensional
+data, a burden is placed on the user to consider the orientation of the data
+set when writing functions; axes are considered more or less equivalent (except
+when C- or Fortran-contiguousness matters for performance). In pandas, the axes
+are intended to lend more semantic meaning to the data; i.e., for a particular
+data set, there is likely to be a "right" way to orient the data. The goal,
+then, is to reduce the amount of mental effort required to code up data
+transformations in downstream functions.
+
+For example, with tabular data (DataFrame) it is more semantically helpful to
+think of the **index** (the rows) and the **columns** rather than axis 0 and
+axis 1. Iterating through the columns of the DataFrame thus results in more
+readable code:
+
+::
+
+    for col in df.columns:
+        series = df[col]
+        # do something with series
+
+Mutability and copying of data
+------------------------------
+
+All pandas data structures are value-mutable (the values they contain can be
+altered) but not always size-mutable. The length of a Series cannot be
+changed, but, for example, columns can be inserted into a DataFrame. However,
+the vast majority of methods produce new objects and leave the input data
+untouched. In general we like to **favor immutability** where sensible.
+
+Getting support
+---------------
+
+The first stop for pandas issues and ideas is the `GitHub Issue Tracker
+<https://github.com/pandas-dev/pandas/issues>`__. If you have a general question,
+pandas community experts can answer through `Stack Overflow
+<https://stackoverflow.com/questions/tagged/pandas>`__.
+
+Community
+---------
+
+pandas is actively supported today by a community of like-minded individuals around
+the world who contribute their valuable time and energy to help make open source
+pandas possible. Thanks to `all of our contributors <https://github.com/pandas-dev/pandas/graphs/contributors>`__.
+
+If you're interested in contributing, please visit the :ref:`contributing guide <contributing>`.
+
+pandas is a `NumFOCUS <https://numfocus.org/sponsored-projects>`__ sponsored project.
+This will help ensure the success of the development of pandas as a world-class open-source
+project and makes it possible to `donate <https://pandas.pydata.org/donate.html>`__ to the project.
+
+Project governance
+------------------
+
+The governance process that pandas project has used informally since its inception in 2008 is formalized in `Project Governance documents <https://github.com/pandas-dev/pandas/blob/main/web/pandas/about/governance.md>`__.
+The documents clarify how decisions are made and how the various elements of our community interact, including the relationship between open source collaborative development and work that may be funded by for-profit or non-profit entities.
+
+Wes McKinney is the Benevolent Dictator for Life (BDFL).
+
+Development team
+-----------------
+
+The list of the Core Team members and more detailed information can be found on the `pandas website <https://pandas.pydata.org/about/team.html>`__.
+
+
+Institutional partners
+----------------------
+
+The information about current institutional partners can be found on `pandas website page <https://pandas.pydata.org/about/sponsors.html>`__.
+
+License
+-------
+
+.. literalinclude:: ../../../LICENSE
diff --git a/pandas/doc/source/getting_started/tutorials.rst b/pandas/doc/source/getting_started/tutorials.rst
new file mode 100644
index 0000000000000..eae7771418485
--- /dev/null
+++ b/pandas/doc/source/getting_started/tutorials.rst
@@ -0,0 +1,120 @@
+.. _communitytutorials:
+
+{{ header }}
+
+*******************
+Community tutorials
+*******************
+
+This is a guide to many pandas tutorials by the community, geared mainly for new users.
+
+pandas cookbook by Julia Evans
+------------------------------
+
+The goal of this 2015 cookbook (by `Julia Evans <https://jvns.ca>`_) is to
+give you some concrete examples for getting started with pandas. These
+are examples with real-world data, and all the bugs and weirdness that
+entails.
+For the table of contents, see the `pandas-cookbook GitHub
+repository <https://github.com/jvns/pandas-cookbook>`_.
+
+pandas workshop by Stefanie Molin
+---------------------------------
+
+An introductory workshop by `Stefanie Molin <https://github.com/stefmolin>`_
+designed to quickly get you up to speed with pandas using real-world datasets.
+It covers getting started with pandas, data wrangling, and data visualization
+(with some exposure to matplotlib and seaborn). The
+`pandas-workshop GitHub repository <https://github.com/stefmolin/pandas-workshop>`_
+features detailed environment setup instructions (including a Binder environment),
+slides and notebooks for following along, and exercises to practice the concepts.
+There is also a lab with new exercises on a dataset not covered in the workshop for
+additional practice.
+
+Learn pandas by Hernan Rojas
+----------------------------
+
+A set of lesson for new pandas users: https://bitbucket.org/hrojas/learn-pandas
+
+Practical data analysis with Python
+-----------------------------------
+
+This `guide <https://wavedatalab.github.io/datawithpython>`_ is an introduction to the data analysis process using the Python data ecosystem and an interesting open dataset.
+There are four sections covering selected topics as `munging data <https://wavedatalab.github.io/datawithpython/munge.html>`__,
+`aggregating data <https://wavedatalab.github.io/datawithpython/aggregate.html>`_, `visualizing data <https://wavedatalab.github.io/datawithpython/visualize.html>`_
+and `time series <https://wavedatalab.github.io/datawithpython/timeseries.html>`_.
+
+.. _tutorial-exercises-new-users:
+
+Exercises for new users
+-----------------------
+Practice your skills with real data sets and exercises.
+For more resources, please visit the main `repository <https://github.com/guipsamora/pandas_exercises>`__.
+
+
+.. _tutorial-modern:
+
+Modern pandas
+-------------
+
+Tutorial series written in 2016 by
+`Tom Augspurger <https://github.com/TomAugspurger>`_.
+The source may be found in the GitHub repository
+`TomAugspurger/effective-pandas <https://github.com/TomAugspurger/effective-pandas>`_.
+
+* `Modern Pandas <https://tomaugspurger.github.io/modern-1-intro.html>`_
+* `Method Chaining <https://tomaugspurger.github.io/method-chaining.html>`_
+* `Indexes <https://tomaugspurger.github.io/modern-3-indexes.html>`_
+* `Performance <https://tomaugspurger.github.io/modern-4-performance.html>`_
+* `Tidy Data <https://tomaugspurger.github.io/modern-5-tidy.html>`_
+* `Visualization <https://tomaugspurger.github.io/modern-6-visualization.html>`_
+* `Timeseries <https://tomaugspurger.github.io/modern-7-timeseries.html>`_
+
+Excel charts with pandas, vincent and xlsxwriter
+------------------------------------------------
+
+*  `Using Pandas and XlsxWriter to create Excel charts <https://pandas-xlsxwriter-charts.readthedocs.io/>`_
+
+Joyful pandas
+-------------
+
+A tutorial written in Chinese by Yuanhao Geng. It covers the basic operations
+for NumPy and pandas, 4 main data manipulation methods (including indexing, groupby, reshaping
+and concatenation) and 4 main data types (including missing data, string data, categorical
+data and time series data). At the end of each chapter, corresponding exercises are posted.
+All the datasets and related materials can be found in the GitHub repository
+`datawhalechina/joyful-pandas <https://github.com/datawhalechina/joyful-pandas>`_.
+
+Video tutorials
+---------------
+
+* `Pandas From The Ground Up <https://www.youtube.com/watch?v=5JnMutdy6Fw>`_
+  (2015) (2:24)
+  `GitHub repo <https://github.com/brandon-rhodes/pycon-pandas-tutorial>`__
+* `Introduction Into Pandas <https://www.youtube.com/watch?v=-NR-ynQg0YM>`_
+  (2016) (1:28)
+  `GitHub repo <https://github.com/chendaniely/2016-pydata-carolinas-pandas>`__
+* `Pandas: .head() to .tail() <https://www.youtube.com/watch?v=7vuO9QXDN50>`_
+  (2016) (1:26)
+  `GitHub repo <https://github.com/TomAugspurger/pydata-chi-h2t>`__
+* `Data analysis in Python with pandas <https://www.youtube.com/playlist?list=PL5-da3qGB5ICCsgW1MxlZ0Hq8LL5U3u9y>`_
+  (2016-2018)
+  `GitHub repo <https://github.com/justmarkham/pandas-videos>`__ and
+  `Jupyter Notebook <https://nbviewer.org/github/justmarkham/pandas-videos/blob/master/pandas.ipynb>`__
+* `Best practices with pandas <https://www.youtube.com/playlist?list=PL5-da3qGB5IBITZj_dYSFqnd_15JgqwA6>`_
+  (2018)
+  `GitHub repo <https://github.com/justmarkham/pycon-2018-tutorial>`__ and
+  `Jupyter Notebook <https://nbviewer.org/github/justmarkham/pycon-2018-tutorial/blob/master/tutorial.ipynb>`__
+
+
+Various tutorials
+-----------------
+
+* `Wes McKinney's (pandas BDFL) blog <https://wesmckinney.com/archives.html>`_
+* `Statistical analysis made easy in Python with SciPy and pandas DataFrames, by Randal Olson <http://www.randalolson.com/2012/08/06/statistical-analysis-made-easy-in-python/>`_
+* `Statistical Data Analysis in Python, tutorial by Christopher Fonnesbeck from SciPy 2013 <https://github.com/fonnesbeck/statistical-analysis-python-tutorial>`_
+* `Financial analysis in Python, by Thomas Wiecki <https://nbviewer.org/github/twiecki/financial-analysis-python-tutorial/blob/master/1.%20Pandas%20Basics.ipynb>`_
+* `Intro to pandas data structures, by Greg Reda <http://www.gregreda.com/2013/10/26/intro-to-pandas-data-structures/>`_
+* `Pandas DataFrames Tutorial, by Karlijn Willems <https://www.datacamp.com/community/tutorials/pandas-tutorial-dataframe-python>`_
+* `A concise tutorial with real life examples <https://tutswiki.com/pandas-cookbook/chapter1/>`_
+* `430+ Searchable Pandas recipes by Isshin Inada <https://skytowner.com/explore/pandas_recipes_reference>`_
diff --git a/pandas/doc/source/index.rst.template b/pandas/doc/source/index.rst.template
new file mode 100644
index 0000000000000..8a72df0661fe4
--- /dev/null
+++ b/pandas/doc/source/index.rst.template
@@ -0,0 +1,143 @@
+:notoc:
+
+.. pandas documentation master file, created by
+
+.. module:: pandas
+
+********************
+pandas documentation
+********************
+
+**Date**: |today| **Version**: |version|
+
+**Download documentation**: `Zipped HTML <pandas.zip>`__
+
+**Previous versions**: Documentation of previous pandas versions is available at
+`pandas.pydata.org <https://pandas.pydata.org/>`__.
+
+**Useful links**:
+`Binary Installers <https://pypi.org/project/pandas>`__ |
+`Source Repository <https://github.com/pandas-dev/pandas>`__ |
+`Issues & Ideas <https://github.com/pandas-dev/pandas/issues>`__ |
+`Q&A Support <https://stackoverflow.com/questions/tagged/pandas>`__ |
+`Mailing List <https://groups.google.com/g/pydata>`__
+
+:mod:`pandas` is an open source, BSD-licensed library providing high-performance,
+easy-to-use data structures and data analysis tools for the `Python <https://www.python.org/>`__
+programming language.
+
+{% if not single_doc or single_doc == "index.rst" -%}
+.. grid:: 1 2 2 2
+    :gutter: 4
+    :padding: 2 2 0 0
+    :class-container: sd-text-center
+
+    .. grid-item-card:: Getting started
+        :img-top: _static/index_getting_started.svg
+        :class-card: intro-card
+        :shadow: md
+
+        New to *pandas*? Check out the getting started guides. They contain an
+        introduction to *pandas'* main concepts and links to additional tutorials.
+
+        +++
+
+        .. button-ref:: getting_started
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            To the getting started guides
+
+    .. grid-item-card::  User guide
+        :img-top: _static/index_user_guide.svg
+        :class-card: intro-card
+        :shadow: md
+
+        The user guide provides in-depth information on the
+        key concepts of pandas with useful background information and explanation.
+
+        +++
+
+        .. button-ref:: user_guide
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            To the user guide
+
+    .. grid-item-card::  API reference
+        :img-top: _static/index_api.svg
+        :class-card: intro-card
+        :shadow: md
+
+        The reference guide contains a detailed description of
+        the pandas API. The reference describes how the methods work and which parameters can
+        be used. It assumes that you have an understanding of the key concepts.
+
+        +++
+
+        .. button-ref:: api
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            To the reference guide
+
+    .. grid-item-card::  Developer guide
+        :img-top: _static/index_contribute.svg
+        :class-card: intro-card
+        :shadow: md
+
+        Saw a typo in the documentation? Want to improve
+        existing functionalities? The contributing guidelines will guide
+        you through the process of improving pandas.
+
+        +++
+
+        .. button-ref:: development
+            :ref-type: ref
+            :click-parent:
+            :color: secondary
+            :expand:
+
+            To the development guide
+
+{% endif %}
+{% if single_doc and single_doc.endswith('.rst') -%}
+.. toctree::
+    :maxdepth: 3
+    :titlesonly:
+
+    {{ single_doc[:-4] }}
+{% elif single_doc and single_doc.count('.') <= 1 %}
+.. autosummary::
+    :toctree: reference/api/
+
+    {{ single_doc }}
+{% elif single_doc %}
+.. autosummary::
+    :toctree: reference/api/
+    :template: autosummary/accessor_method.rst
+
+    {{ single_doc }}
+{% else -%}
+.. toctree::
+    :maxdepth: 3
+    :hidden:
+    :titlesonly:
+{% endif %}
+{% if not single_doc %}
+    getting_started/index
+    user_guide/index
+    {% endif -%}
+    {% if include_api -%}
+    reference/index
+    {% endif -%}
+    {% if not single_doc -%}
+    development/index
+    whatsnew/index
+{% endif %}
diff --git a/pandas/doc/source/reference/arrays.rst b/pandas/doc/source/reference/arrays.rst
new file mode 100644
index 0000000000000..5be08f163e6ce
--- /dev/null
+++ b/pandas/doc/source/reference/arrays.rst
@@ -0,0 +1,706 @@
+{{ header }}
+
+.. _api.arrays:
+
+======================================
+pandas arrays, scalars, and data types
+======================================
+
+*******
+Objects
+*******
+
+.. currentmodule:: pandas
+
+For most data types, pandas uses NumPy arrays as the concrete
+objects contained with a :class:`Index`, :class:`Series`, or
+:class:`DataFrame`.
+
+For some data types, pandas extends NumPy's type system. String aliases for these types
+can be found at :ref:`basics.dtypes`.
+
+=================== ========================== ============================= =============================
+Kind of Data        pandas Data Type           Scalar                        Array
+=================== ========================== ============================= =============================
+TZ-aware datetime   :class:`DatetimeTZDtype`   :class:`Timestamp`            :ref:`api.arrays.datetime`
+Timedeltas          (none)                     :class:`Timedelta`            :ref:`api.arrays.timedelta`
+Period (time spans) :class:`PeriodDtype`       :class:`Period`               :ref:`api.arrays.period`
+Intervals           :class:`IntervalDtype`     :class:`Interval`             :ref:`api.arrays.interval`
+Nullable Integer    :class:`Int64Dtype`, ...   (none)                        :ref:`api.arrays.integer_na`
+Nullable Float      :class:`Float64Dtype`, ... (none)                        :ref:`api.arrays.float_na`
+Categorical         :class:`CategoricalDtype`  (none)                        :ref:`api.arrays.categorical`
+Sparse              :class:`SparseDtype`       (none)                        :ref:`api.arrays.sparse`
+Strings             :class:`StringDtype`       :class:`str`                  :ref:`api.arrays.string`
+Nullable Boolean    :class:`BooleanDtype`      :class:`bool`                 :ref:`api.arrays.bool`
+PyArrow             :class:`ArrowDtype`        Python Scalars or :class:`NA` :ref:`api.arrays.arrow`
+=================== ========================== ============================= =============================
+
+pandas and third-party libraries can extend NumPy's type system (see :ref:`extending.extension-types`).
+The top-level :meth:`array` method can be used to create a new array, which may be
+stored in a :class:`Series`, :class:`Index`, or as a column in a :class:`DataFrame`.
+
+.. autosummary::
+   :toctree: api/
+
+   array
+
+.. _api.arrays.arrow:
+
+PyArrow
+-------
+
+.. warning::
+
+    This feature is experimental, and the API can change in a future release without warning.
+
+The :class:`arrays.ArrowExtensionArray` is backed by a :external+pyarrow:py:class:`pyarrow.ChunkedArray` with a
+:external+pyarrow:py:class:`pyarrow.DataType` instead of a NumPy array and data type. The ``.dtype`` of a :class:`arrays.ArrowExtensionArray`
+is an :class:`ArrowDtype`.
+
+`Pyarrow <https://arrow.apache.org/docs/python/index.html>`__ provides similar array and `data type <https://arrow.apache.org/docs/python/api/datatypes.html>`__
+support as NumPy including first-class nullability support for all data types, immutability and more.
+
+The table below shows the equivalent pyarrow-backed (``pa``), pandas extension, and numpy (``np``) types that are recognized by pandas.
+Pyarrow-backed types below need to be passed into :class:`ArrowDtype` to be recognized by pandas e.g. ``pd.ArrowDtype(pa.bool_())``.
+
+=============================================== ========================== ===================
+PyArrow type                                    pandas extension type      NumPy type
+=============================================== ========================== ===================
+:external+pyarrow:py:func:`pyarrow.bool_`       :class:`BooleanDtype`      ``np.bool_``
+:external+pyarrow:py:func:`pyarrow.int8`        :class:`Int8Dtype`         ``np.int8``
+:external+pyarrow:py:func:`pyarrow.int16`       :class:`Int16Dtype`        ``np.int16``
+:external+pyarrow:py:func:`pyarrow.int32`       :class:`Int32Dtype`        ``np.int32``
+:external+pyarrow:py:func:`pyarrow.int64`       :class:`Int64Dtype`        ``np.int64``
+:external+pyarrow:py:func:`pyarrow.uint8`       :class:`UInt8Dtype`        ``np.uint8``
+:external+pyarrow:py:func:`pyarrow.uint16`      :class:`UInt16Dtype`       ``np.uint16``
+:external+pyarrow:py:func:`pyarrow.uint32`      :class:`UInt32Dtype`       ``np.uint32``
+:external+pyarrow:py:func:`pyarrow.uint64`      :class:`UInt64Dtype`       ``np.uint64``
+:external+pyarrow:py:func:`pyarrow.float32`     :class:`Float32Dtype`      ``np.float32``
+:external+pyarrow:py:func:`pyarrow.float64`     :class:`Float64Dtype`      ``np.float64``
+:external+pyarrow:py:func:`pyarrow.time32`      (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.time64`      (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.timestamp`   :class:`DatetimeTZDtype`   ``np.datetime64``
+:external+pyarrow:py:func:`pyarrow.date32`      (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.date64`      (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.duration`    (none)                     ``np.timedelta64``
+:external+pyarrow:py:func:`pyarrow.binary`      (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.string`      :class:`StringDtype`       ``np.str_``
+:external+pyarrow:py:func:`pyarrow.decimal128`  (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.list_`       (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.map_`        (none)                     (none)
+:external+pyarrow:py:func:`pyarrow.dictionary`  :class:`CategoricalDtype`  (none)
+=============================================== ========================== ===================
+
+.. note::
+
+    Pyarrow-backed string support is provided by both ``pd.StringDtype("pyarrow")`` and ``pd.ArrowDtype(pa.string())``.
+    ``pd.StringDtype("pyarrow")`` is described below in the :ref:`string section <api.arrays.string>`
+    and will be returned if the string alias ``"string[pyarrow]"`` is specified. ``pd.ArrowDtype(pa.string())``
+    generally has better interoperability with :class:`ArrowDtype` of different types.
+
+While individual values in an :class:`arrays.ArrowExtensionArray` are stored as a PyArrow objects, scalars are **returned**
+as Python scalars corresponding to the data type, e.g. a PyArrow int64 will be returned as Python int, or :class:`NA` for missing
+values.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.ArrowExtensionArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   ArrowDtype
+
+For more information, please see the :ref:`PyArrow user guide <pyarrow>`.
+
+.. _api.arrays.datetime:
+
+Datetimes
+---------
+
+NumPy cannot natively represent timezone-aware datetimes. pandas supports this
+with the :class:`arrays.DatetimeArray` extension array, which can hold timezone-naive
+or timezone-aware values.
+
+:class:`Timestamp`, a subclass of :class:`datetime.datetime`, is pandas'
+scalar type for timezone-naive or timezone-aware datetime data. :class:`NaT`
+is the missing value for datetime data.
+
+.. autosummary::
+   :toctree: api/
+
+   Timestamp
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Timestamp.asm8
+   Timestamp.day
+   Timestamp.dayofweek
+   Timestamp.day_of_week
+   Timestamp.dayofyear
+   Timestamp.day_of_year
+   Timestamp.days_in_month
+   Timestamp.daysinmonth
+   Timestamp.fold
+   Timestamp.hour
+   Timestamp.is_leap_year
+   Timestamp.is_month_end
+   Timestamp.is_month_start
+   Timestamp.is_quarter_end
+   Timestamp.is_quarter_start
+   Timestamp.is_year_end
+   Timestamp.is_year_start
+   Timestamp.max
+   Timestamp.microsecond
+   Timestamp.min
+   Timestamp.minute
+   Timestamp.month
+   Timestamp.nanosecond
+   Timestamp.quarter
+   Timestamp.resolution
+   Timestamp.second
+   Timestamp.tz
+   Timestamp.tzinfo
+   Timestamp.unit
+   Timestamp.value
+   Timestamp.week
+   Timestamp.weekofyear
+   Timestamp.year
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Timestamp.as_unit
+   Timestamp.astimezone
+   Timestamp.ceil
+   Timestamp.combine
+   Timestamp.ctime
+   Timestamp.date
+   Timestamp.day_name
+   Timestamp.dst
+   Timestamp.floor
+   Timestamp.fromordinal
+   Timestamp.fromtimestamp
+   Timestamp.isocalendar
+   Timestamp.isoformat
+   Timestamp.isoweekday
+   Timestamp.month_name
+   Timestamp.normalize
+   Timestamp.now
+   Timestamp.replace
+   Timestamp.round
+   Timestamp.strftime
+   Timestamp.strptime
+   Timestamp.time
+   Timestamp.timestamp
+   Timestamp.timetuple
+   Timestamp.timetz
+   Timestamp.to_datetime64
+   Timestamp.to_numpy
+   Timestamp.to_julian_date
+   Timestamp.to_period
+   Timestamp.to_pydatetime
+   Timestamp.today
+   Timestamp.toordinal
+   Timestamp.tz_convert
+   Timestamp.tz_localize
+   Timestamp.tzname
+   Timestamp.utcfromtimestamp
+   Timestamp.utcnow
+   Timestamp.utcoffset
+   Timestamp.utctimetuple
+   Timestamp.weekday
+
+A collection of timestamps may be stored in a :class:`arrays.DatetimeArray`.
+For timezone-aware data, the ``.dtype`` of a :class:`arrays.DatetimeArray` is a
+:class:`DatetimeTZDtype`. For timezone-naive data, ``np.dtype("datetime64[ns]")``
+is used.
+
+If the data are timezone-aware, then every value in the array must have the same timezone.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.DatetimeArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   DatetimeTZDtype
+
+.. _api.arrays.timedelta:
+
+Timedeltas
+----------
+
+NumPy can natively represent timedeltas. pandas provides :class:`Timedelta`
+for symmetry with :class:`Timestamp`. :class:`NaT`
+is the missing value for timedelta data.
+
+.. autosummary::
+   :toctree: api/
+
+   Timedelta
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Timedelta.asm8
+   Timedelta.components
+   Timedelta.days
+   Timedelta.max
+   Timedelta.microseconds
+   Timedelta.min
+   Timedelta.nanoseconds
+   Timedelta.resolution
+   Timedelta.seconds
+   Timedelta.unit
+   Timedelta.value
+   Timedelta.view
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Timedelta.as_unit
+   Timedelta.ceil
+   Timedelta.floor
+   Timedelta.isoformat
+   Timedelta.round
+   Timedelta.to_pytimedelta
+   Timedelta.to_timedelta64
+   Timedelta.to_numpy
+   Timedelta.total_seconds
+
+A collection of :class:`Timedelta` may be stored in a :class:`TimedeltaArray`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.TimedeltaArray
+
+.. _api.arrays.period:
+
+Periods
+-------
+
+pandas represents spans of times as :class:`Period` objects.
+
+Period
+------
+.. autosummary::
+   :toctree: api/
+
+   Period
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Period.day
+   Period.dayofweek
+   Period.day_of_week
+   Period.dayofyear
+   Period.day_of_year
+   Period.days_in_month
+   Period.daysinmonth
+   Period.end_time
+   Period.freq
+   Period.freqstr
+   Period.hour
+   Period.is_leap_year
+   Period.minute
+   Period.month
+   Period.ordinal
+   Period.quarter
+   Period.qyear
+   Period.second
+   Period.start_time
+   Period.week
+   Period.weekday
+   Period.weekofyear
+   Period.year
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Period.asfreq
+   Period.now
+   Period.strftime
+   Period.to_timestamp
+
+A collection of :class:`Period` may be stored in a :class:`arrays.PeriodArray`.
+Every period in a :class:`arrays.PeriodArray` must have the same ``freq``.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.PeriodArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   PeriodDtype
+
+.. _api.arrays.interval:
+
+Intervals
+---------
+
+Arbitrary intervals can be represented as :class:`Interval` objects.
+
+.. autosummary::
+   :toctree: api/
+
+    Interval
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Interval.closed
+   Interval.closed_left
+   Interval.closed_right
+   Interval.is_empty
+   Interval.left
+   Interval.length
+   Interval.mid
+   Interval.open_left
+   Interval.open_right
+   Interval.overlaps
+   Interval.right
+
+A collection of intervals may be stored in an :class:`arrays.IntervalArray`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.IntervalArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   IntervalDtype
+
+
+.. Those attributes and methods are included in the API because the docstrings
+.. of IntervalIndex and IntervalArray are shared. Including it here to make
+.. sure a docstring page is built for them to avoid warnings
+
+..
+    .. autosummary::
+      :toctree: api/
+
+      arrays.IntervalArray.left
+      arrays.IntervalArray.right
+      arrays.IntervalArray.closed
+      arrays.IntervalArray.mid
+      arrays.IntervalArray.length
+      arrays.IntervalArray.is_empty
+      arrays.IntervalArray.is_non_overlapping_monotonic
+      arrays.IntervalArray.from_arrays
+      arrays.IntervalArray.from_tuples
+      arrays.IntervalArray.from_breaks
+      arrays.IntervalArray.contains
+      arrays.IntervalArray.overlaps
+      arrays.IntervalArray.set_closed
+      arrays.IntervalArray.to_tuples
+
+
+.. _api.arrays.integer_na:
+
+Nullable integer
+----------------
+
+:class:`numpy.ndarray` cannot natively represent integer-data with missing values.
+pandas provides this through :class:`arrays.IntegerArray`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.IntegerArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   Int8Dtype
+   Int16Dtype
+   Int32Dtype
+   Int64Dtype
+   UInt8Dtype
+   UInt16Dtype
+   UInt32Dtype
+   UInt64Dtype
+
+.. _api.arrays.float_na:
+
+Nullable float
+--------------
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.FloatingArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   Float32Dtype
+   Float64Dtype
+
+.. _api.arrays.categorical:
+
+Categoricals
+------------
+
+pandas defines a custom data type for representing data that can take only a
+limited, fixed set of values. The dtype of a :class:`Categorical` can be described by
+a :class:`CategoricalDtype`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   CategoricalDtype
+
+.. autosummary::
+   :toctree: api/
+
+   CategoricalDtype.categories
+   CategoricalDtype.ordered
+
+Categorical data can be stored in a :class:`pandas.Categorical`:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   Categorical
+
+The alternative :meth:`Categorical.from_codes` constructor can be used when you
+have the categories and integer codes already:
+
+.. autosummary::
+   :toctree: api/
+
+   Categorical.from_codes
+
+The dtype information is available on the :class:`Categorical`
+
+.. autosummary::
+   :toctree: api/
+
+   Categorical.dtype
+   Categorical.categories
+   Categorical.ordered
+   Categorical.codes
+
+``np.asarray(categorical)`` works by implementing the array interface. Be aware, that this converts
+the :class:`Categorical` back to a NumPy array, so categories and order information is not preserved!
+
+.. autosummary::
+   :toctree: api/
+
+   Categorical.__array__
+
+A :class:`Categorical` can be stored in a :class:`Series` or :class:`DataFrame`.
+To create a Series of dtype ``category``, use ``cat = s.astype(dtype)`` or
+``Series(..., dtype=dtype)`` where ``dtype`` is either
+
+* the string ``'category'``
+* an instance of :class:`CategoricalDtype`.
+
+If the :class:`Series` is of dtype :class:`CategoricalDtype`, ``Series.cat`` can be used to change the categorical
+data. See :ref:`api.series.cat` for more.
+
+More methods are available on :class:`Categorical`:
+
+.. autosummary::
+   :toctree: api/
+
+   Categorical.as_ordered
+   Categorical.as_unordered
+   Categorical.set_categories
+   Categorical.rename_categories
+   Categorical.reorder_categories
+   Categorical.add_categories
+   Categorical.remove_categories
+   Categorical.remove_unused_categories
+   Categorical.map
+
+.. _api.arrays.sparse:
+
+Sparse
+------
+
+Data where a single value is repeated many times (e.g. ``0`` or ``NaN``) may
+be stored efficiently as a :class:`arrays.SparseArray`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.SparseArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   SparseDtype
+
+The ``Series.sparse`` accessor may be used to access sparse-specific attributes
+and methods if the :class:`Series` contains sparse values. See
+:ref:`api.series.sparse` and :ref:`the user guide <sparse>` for more.
+
+
+.. _api.arrays.string:
+
+Strings
+-------
+
+When working with text data, where each valid element is a string or missing,
+we recommend using :class:`StringDtype` (with the alias ``"string"``).
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.StringArray
+   arrays.ArrowStringArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   StringDtype
+
+The ``Series.str`` accessor is available for :class:`Series` backed by a :class:`arrays.StringArray`.
+See :ref:`api.series.str` for more.
+
+
+.. _api.arrays.bool:
+
+Nullable Boolean
+----------------
+
+The boolean dtype (with the alias ``"boolean"``) provides support for storing
+boolean data (``True``, ``False``) with missing values, which is not possible
+with a bool :class:`numpy.ndarray`.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   arrays.BooleanArray
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   BooleanDtype
+
+
+.. Dtype attributes which are manually listed in their docstrings: including
+.. it here to make sure a docstring page is built for them
+
+..
+    .. autosummary::
+      :toctree: api/
+
+      DatetimeTZDtype.unit
+      DatetimeTZDtype.tz
+      PeriodDtype.freq
+      IntervalDtype.subtype
+
+*********
+Utilities
+*********
+
+Constructors
+------------
+.. autosummary::
+   :toctree: api/
+
+   api.types.union_categoricals
+   api.types.infer_dtype
+   api.types.pandas_dtype
+
+Data type introspection
+~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    api.types.is_any_real_numeric_dtype
+    api.types.is_bool_dtype
+    api.types.is_categorical_dtype
+    api.types.is_complex_dtype
+    api.types.is_datetime64_any_dtype
+    api.types.is_datetime64_dtype
+    api.types.is_datetime64_ns_dtype
+    api.types.is_datetime64tz_dtype
+    api.types.is_extension_array_dtype
+    api.types.is_float_dtype
+    api.types.is_int64_dtype
+    api.types.is_integer_dtype
+    api.types.is_interval_dtype
+    api.types.is_numeric_dtype
+    api.types.is_object_dtype
+    api.types.is_period_dtype
+    api.types.is_signed_integer_dtype
+    api.types.is_string_dtype
+    api.types.is_timedelta64_dtype
+    api.types.is_timedelta64_ns_dtype
+    api.types.is_unsigned_integer_dtype
+    api.types.is_sparse
+
+Iterable introspection
+~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    api.types.is_dict_like
+    api.types.is_file_like
+    api.types.is_list_like
+    api.types.is_named_tuple
+    api.types.is_iterator
+
+Scalar introspection
+~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    api.types.is_bool
+    api.types.is_complex
+    api.types.is_float
+    api.types.is_hashable
+    api.types.is_integer
+    api.types.is_number
+    api.types.is_re
+    api.types.is_re_compilable
+    api.types.is_scalar
diff --git a/pandas/doc/source/reference/extensions.rst b/pandas/doc/source/reference/extensions.rst
new file mode 100644
index 0000000000000..e412793a328a3
--- /dev/null
+++ b/pandas/doc/source/reference/extensions.rst
@@ -0,0 +1,83 @@
+{{ header }}
+
+.. _api.extensions:
+
+==========
+Extensions
+==========
+.. currentmodule:: pandas
+
+These are primarily intended for library authors looking to extend pandas
+objects.
+
+.. autosummary::
+   :toctree: api/
+
+   api.extensions.register_extension_dtype
+   api.extensions.register_dataframe_accessor
+   api.extensions.register_series_accessor
+   api.extensions.register_index_accessor
+   api.extensions.ExtensionDtype
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   api.extensions.ExtensionArray
+   arrays.NumpyExtensionArray
+
+.. We need this autosummary so that methods and attributes are generated.
+.. Separate block, since they aren't classes.
+
+   .. autosummary::
+      :toctree: api/
+
+      api.extensions.ExtensionArray._accumulate
+      api.extensions.ExtensionArray._concat_same_type
+      api.extensions.ExtensionArray._explode
+      api.extensions.ExtensionArray._formatter
+      api.extensions.ExtensionArray._from_factorized
+      api.extensions.ExtensionArray._from_sequence
+      api.extensions.ExtensionArray._from_sequence_of_strings
+      api.extensions.ExtensionArray._hash_pandas_object
+      api.extensions.ExtensionArray._pad_or_backfill
+      api.extensions.ExtensionArray._reduce
+      api.extensions.ExtensionArray._values_for_argsort
+      api.extensions.ExtensionArray._values_for_factorize
+      api.extensions.ExtensionArray.argsort
+      api.extensions.ExtensionArray.astype
+      api.extensions.ExtensionArray.copy
+      api.extensions.ExtensionArray.view
+      api.extensions.ExtensionArray.dropna
+      api.extensions.ExtensionArray.duplicated
+      api.extensions.ExtensionArray.equals
+      api.extensions.ExtensionArray.factorize
+      api.extensions.ExtensionArray.fillna
+      api.extensions.ExtensionArray.insert
+      api.extensions.ExtensionArray.interpolate
+      api.extensions.ExtensionArray.isin
+      api.extensions.ExtensionArray.isna
+      api.extensions.ExtensionArray.ravel
+      api.extensions.ExtensionArray.repeat
+      api.extensions.ExtensionArray.searchsorted
+      api.extensions.ExtensionArray.shift
+      api.extensions.ExtensionArray.take
+      api.extensions.ExtensionArray.unique
+      api.extensions.ExtensionArray.dtype
+      api.extensions.ExtensionArray.nbytes
+      api.extensions.ExtensionArray.ndim
+      api.extensions.ExtensionArray.shape
+      api.extensions.ExtensionArray.tolist
+
+Additionally, we have some utility methods for ensuring your object
+behaves correctly.
+
+.. autosummary::
+  :toctree: api/
+
+  api.indexers.check_array_indexer
+
+
+The sentinel ``pandas.api.extensions.no_default`` is used as the default
+value in some methods. Use an ``is`` comparison to check if the user
+provides a non-default value.
diff --git a/pandas/doc/source/reference/frame.rst b/pandas/doc/source/reference/frame.rst
new file mode 100644
index 0000000000000..e701d48a89db7
--- /dev/null
+++ b/pandas/doc/source/reference/frame.rst
@@ -0,0 +1,383 @@
+{{ header }}
+
+.. _api.dataframe:
+
+=========
+DataFrame
+=========
+.. currentmodule:: pandas
+
+Constructor
+~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame
+
+Attributes and underlying data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+**Axes**
+
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.index
+   DataFrame.columns
+
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.dtypes
+   DataFrame.info
+   DataFrame.select_dtypes
+   DataFrame.values
+   DataFrame.axes
+   DataFrame.ndim
+   DataFrame.size
+   DataFrame.shape
+   DataFrame.memory_usage
+   DataFrame.empty
+   DataFrame.set_flags
+
+Conversion
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.astype
+   DataFrame.convert_dtypes
+   DataFrame.infer_objects
+   DataFrame.copy
+   DataFrame.to_numpy
+
+Indexing, iteration
+~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.head
+   DataFrame.at
+   DataFrame.iat
+   DataFrame.loc
+   DataFrame.iloc
+   DataFrame.insert
+   DataFrame.__iter__
+   DataFrame.items
+   DataFrame.keys
+   DataFrame.iterrows
+   DataFrame.itertuples
+   DataFrame.pop
+   DataFrame.tail
+   DataFrame.xs
+   DataFrame.get
+   DataFrame.isin
+   DataFrame.where
+   DataFrame.mask
+   DataFrame.query
+   DataFrame.isetitem
+
+For more information on ``.at``, ``.iat``, ``.loc``, and
+``.iloc``,  see the :ref:`indexing documentation <indexing>`.
+
+Binary operator functions
+~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.__add__
+   DataFrame.add
+   DataFrame.sub
+   DataFrame.mul
+   DataFrame.div
+   DataFrame.truediv
+   DataFrame.floordiv
+   DataFrame.mod
+   DataFrame.pow
+   DataFrame.dot
+   DataFrame.radd
+   DataFrame.rsub
+   DataFrame.rmul
+   DataFrame.rdiv
+   DataFrame.rtruediv
+   DataFrame.rfloordiv
+   DataFrame.rmod
+   DataFrame.rpow
+   DataFrame.lt
+   DataFrame.gt
+   DataFrame.le
+   DataFrame.ge
+   DataFrame.ne
+   DataFrame.eq
+   DataFrame.combine
+   DataFrame.combine_first
+
+Function application, GroupBy & window
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.apply
+   DataFrame.map
+   DataFrame.pipe
+   DataFrame.agg
+   DataFrame.aggregate
+   DataFrame.transform
+   DataFrame.groupby
+   DataFrame.rolling
+   DataFrame.expanding
+   DataFrame.ewm
+
+.. _api.dataframe.stats:
+
+Computations / descriptive stats
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.abs
+   DataFrame.all
+   DataFrame.any
+   DataFrame.clip
+   DataFrame.corr
+   DataFrame.corrwith
+   DataFrame.count
+   DataFrame.cov
+   DataFrame.cummax
+   DataFrame.cummin
+   DataFrame.cumprod
+   DataFrame.cumsum
+   DataFrame.describe
+   DataFrame.diff
+   DataFrame.eval
+   DataFrame.kurt
+   DataFrame.kurtosis
+   DataFrame.max
+   DataFrame.mean
+   DataFrame.median
+   DataFrame.min
+   DataFrame.mode
+   DataFrame.pct_change
+   DataFrame.prod
+   DataFrame.product
+   DataFrame.quantile
+   DataFrame.rank
+   DataFrame.round
+   DataFrame.sem
+   DataFrame.skew
+   DataFrame.sum
+   DataFrame.std
+   DataFrame.var
+   DataFrame.nunique
+   DataFrame.value_counts
+
+Reindexing / selection / label manipulation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.add_prefix
+   DataFrame.add_suffix
+   DataFrame.align
+   DataFrame.at_time
+   DataFrame.between_time
+   DataFrame.drop
+   DataFrame.drop_duplicates
+   DataFrame.duplicated
+   DataFrame.equals
+   DataFrame.filter
+   DataFrame.idxmax
+   DataFrame.idxmin
+   DataFrame.reindex
+   DataFrame.reindex_like
+   DataFrame.rename
+   DataFrame.rename_axis
+   DataFrame.reset_index
+   DataFrame.sample
+   DataFrame.set_axis
+   DataFrame.set_index
+   DataFrame.take
+   DataFrame.truncate
+
+.. _api.dataframe.missing:
+
+Missing data handling
+~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.bfill
+   DataFrame.dropna
+   DataFrame.ffill
+   DataFrame.fillna
+   DataFrame.interpolate
+   DataFrame.isna
+   DataFrame.isnull
+   DataFrame.notna
+   DataFrame.notnull
+   DataFrame.replace
+
+Reshaping, sorting, transposing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.droplevel
+   DataFrame.pivot
+   DataFrame.pivot_table
+   DataFrame.reorder_levels
+   DataFrame.sort_values
+   DataFrame.sort_index
+   DataFrame.nlargest
+   DataFrame.nsmallest
+   DataFrame.swaplevel
+   DataFrame.stack
+   DataFrame.unstack
+   DataFrame.melt
+   DataFrame.explode
+   DataFrame.squeeze
+   DataFrame.to_xarray
+   DataFrame.T
+   DataFrame.transpose
+
+Combining / comparing / joining / merging
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.assign
+   DataFrame.compare
+   DataFrame.join
+   DataFrame.merge
+   DataFrame.update
+
+Time Series-related
+~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.asfreq
+   DataFrame.asof
+   DataFrame.shift
+   DataFrame.first_valid_index
+   DataFrame.last_valid_index
+   DataFrame.resample
+   DataFrame.to_period
+   DataFrame.to_timestamp
+   DataFrame.tz_convert
+   DataFrame.tz_localize
+
+.. _api.frame.flags:
+
+Flags
+~~~~~
+
+Flags refer to attributes of the pandas object. Properties of the dataset (like
+the date is was recorded, the URL it was accessed from, etc.) should be stored
+in :attr:`DataFrame.attrs`.
+
+.. autosummary::
+   :toctree: api/
+
+   Flags
+
+
+.. _api.frame.metadata:
+
+Metadata
+~~~~~~~~
+
+:attr:`DataFrame.attrs` is a dictionary for storing global metadata for this DataFrame.
+
+.. warning:: ``DataFrame.attrs`` is considered experimental and may change without warning.
+
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.attrs
+
+
+.. _api.dataframe.plotting:
+
+Plotting
+~~~~~~~~
+``DataFrame.plot`` is both a callable method and a namespace attribute for
+specific plotting methods of the form ``DataFrame.plot.<kind>``.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_callable.rst
+
+   DataFrame.plot
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   DataFrame.plot.area
+   DataFrame.plot.bar
+   DataFrame.plot.barh
+   DataFrame.plot.box
+   DataFrame.plot.density
+   DataFrame.plot.hexbin
+   DataFrame.plot.hist
+   DataFrame.plot.kde
+   DataFrame.plot.line
+   DataFrame.plot.pie
+   DataFrame.plot.scatter
+
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.boxplot
+   DataFrame.hist
+
+
+.. _api.frame.sparse:
+
+Sparse accessor
+~~~~~~~~~~~~~~~
+
+Sparse-dtype specific methods and attributes are provided under the
+``DataFrame.sparse`` accessor.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   DataFrame.sparse.density
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   DataFrame.sparse.from_spmatrix
+   DataFrame.sparse.to_coo
+   DataFrame.sparse.to_dense
+
+
+Serialization / IO / conversion
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.from_dict
+   DataFrame.from_records
+   DataFrame.to_orc
+   DataFrame.to_parquet
+   DataFrame.to_pickle
+   DataFrame.to_csv
+   DataFrame.to_hdf
+   DataFrame.to_sql
+   DataFrame.to_dict
+   DataFrame.to_excel
+   DataFrame.to_json
+   DataFrame.to_html
+   DataFrame.to_feather
+   DataFrame.to_latex
+   DataFrame.to_stata
+   DataFrame.to_records
+   DataFrame.to_string
+   DataFrame.to_clipboard
+   DataFrame.to_markdown
+   DataFrame.style
+   DataFrame.__dataframe__
diff --git a/pandas/doc/source/reference/general_functions.rst b/pandas/doc/source/reference/general_functions.rst
new file mode 100644
index 0000000000000..e93514de5f762
--- /dev/null
+++ b/pandas/doc/source/reference/general_functions.rst
@@ -0,0 +1,96 @@
+{{ header }}
+
+.. _api.general_functions:
+
+=================
+General functions
+=================
+.. currentmodule:: pandas
+
+Data manipulations
+~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   melt
+   pivot
+   pivot_table
+   crosstab
+   cut
+   qcut
+   merge
+   merge_ordered
+   merge_asof
+   concat
+   get_dummies
+   from_dummies
+   factorize
+   unique
+   lreshape
+   wide_to_long
+
+Top-level missing data
+~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   isna
+   isnull
+   notna
+   notnull
+
+Top-level dealing with numeric data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   to_numeric
+
+Top-level dealing with datetimelike data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   to_datetime
+   to_timedelta
+   date_range
+   bdate_range
+   period_range
+   timedelta_range
+   infer_freq
+
+Top-level dealing with Interval data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   interval_range
+
+Top-level evaluation
+~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   eval
+
+Datetime formats
+~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   tseries.api.guess_datetime_format
+
+Hashing
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   util.hash_array
+   util.hash_pandas_object
+
+Importing from other DataFrame libraries
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   api.interchange.from_dataframe
diff --git a/pandas/doc/source/reference/groupby.rst b/pandas/doc/source/reference/groupby.rst
new file mode 100644
index 0000000000000..fc180c8161a7e
--- /dev/null
+++ b/pandas/doc/source/reference/groupby.rst
@@ -0,0 +1,180 @@
+{{ header }}
+
+.. _api.groupby:
+
+=======
+GroupBy
+=======
+.. currentmodule:: pandas.core.groupby
+
+:class:`pandas.api.typing.DataFrameGroupBy` and :class:`pandas.api.typing.SeriesGroupBy`
+instances are returned by groupby calls :func:`pandas.DataFrame.groupby` and
+:func:`pandas.Series.groupby` respectively.
+
+Indexing, iteration
+-------------------
+.. autosummary::
+   :toctree: api/
+
+   DataFrameGroupBy.__iter__
+   SeriesGroupBy.__iter__
+   DataFrameGroupBy.groups
+   SeriesGroupBy.groups
+   DataFrameGroupBy.indices
+   SeriesGroupBy.indices
+   DataFrameGroupBy.get_group
+   SeriesGroupBy.get_group
+
+.. currentmodule:: pandas
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   Grouper
+
+Function application helper
+---------------------------
+.. autosummary::
+   :toctree: api/
+
+   NamedAgg
+
+.. currentmodule:: pandas.core.groupby
+
+Function application
+--------------------
+.. autosummary::
+   :toctree: api/
+
+   SeriesGroupBy.apply
+   DataFrameGroupBy.apply
+   SeriesGroupBy.agg
+   DataFrameGroupBy.agg
+   SeriesGroupBy.aggregate
+   DataFrameGroupBy.aggregate
+   SeriesGroupBy.transform
+   DataFrameGroupBy.transform
+   SeriesGroupBy.pipe
+   DataFrameGroupBy.pipe
+   DataFrameGroupBy.filter
+   SeriesGroupBy.filter
+
+``DataFrameGroupBy`` computations / descriptive stats
+-----------------------------------------------------
+.. autosummary::
+   :toctree: api/
+
+   DataFrameGroupBy.all
+   DataFrameGroupBy.any
+   DataFrameGroupBy.bfill
+   DataFrameGroupBy.corr
+   DataFrameGroupBy.corrwith
+   DataFrameGroupBy.count
+   DataFrameGroupBy.cov
+   DataFrameGroupBy.cumcount
+   DataFrameGroupBy.cummax
+   DataFrameGroupBy.cummin
+   DataFrameGroupBy.cumprod
+   DataFrameGroupBy.cumsum
+   DataFrameGroupBy.describe
+   DataFrameGroupBy.diff
+   DataFrameGroupBy.ffill
+   DataFrameGroupBy.first
+   DataFrameGroupBy.head
+   DataFrameGroupBy.idxmax
+   DataFrameGroupBy.idxmin
+   DataFrameGroupBy.last
+   DataFrameGroupBy.max
+   DataFrameGroupBy.mean
+   DataFrameGroupBy.median
+   DataFrameGroupBy.min
+   DataFrameGroupBy.ngroup
+   DataFrameGroupBy.nth
+   DataFrameGroupBy.nunique
+   DataFrameGroupBy.ohlc
+   DataFrameGroupBy.pct_change
+   DataFrameGroupBy.prod
+   DataFrameGroupBy.quantile
+   DataFrameGroupBy.rank
+   DataFrameGroupBy.resample
+   DataFrameGroupBy.rolling
+   DataFrameGroupBy.sample
+   DataFrameGroupBy.sem
+   DataFrameGroupBy.shift
+   DataFrameGroupBy.size
+   DataFrameGroupBy.skew
+   DataFrameGroupBy.kurt
+   DataFrameGroupBy.std
+   DataFrameGroupBy.sum
+   DataFrameGroupBy.var
+   DataFrameGroupBy.tail
+   DataFrameGroupBy.take
+   DataFrameGroupBy.value_counts
+
+``SeriesGroupBy`` computations / descriptive stats
+--------------------------------------------------
+.. autosummary::
+   :toctree: api/
+
+   SeriesGroupBy.all
+   SeriesGroupBy.any
+   SeriesGroupBy.bfill
+   SeriesGroupBy.corr
+   SeriesGroupBy.count
+   SeriesGroupBy.cov
+   SeriesGroupBy.cumcount
+   SeriesGroupBy.cummax
+   SeriesGroupBy.cummin
+   SeriesGroupBy.cumprod
+   SeriesGroupBy.cumsum
+   SeriesGroupBy.describe
+   SeriesGroupBy.diff
+   SeriesGroupBy.ffill
+   SeriesGroupBy.first
+   SeriesGroupBy.head
+   SeriesGroupBy.last
+   SeriesGroupBy.idxmax
+   SeriesGroupBy.idxmin
+   SeriesGroupBy.is_monotonic_increasing
+   SeriesGroupBy.is_monotonic_decreasing
+   SeriesGroupBy.max
+   SeriesGroupBy.mean
+   SeriesGroupBy.median
+   SeriesGroupBy.min
+   SeriesGroupBy.ngroup
+   SeriesGroupBy.nlargest
+   SeriesGroupBy.nsmallest
+   SeriesGroupBy.nth
+   SeriesGroupBy.nunique
+   SeriesGroupBy.unique
+   SeriesGroupBy.ohlc
+   SeriesGroupBy.pct_change
+   SeriesGroupBy.prod
+   SeriesGroupBy.quantile
+   SeriesGroupBy.rank
+   SeriesGroupBy.resample
+   SeriesGroupBy.rolling
+   SeriesGroupBy.sample
+   SeriesGroupBy.sem
+   SeriesGroupBy.shift
+   SeriesGroupBy.size
+   SeriesGroupBy.skew
+   SeriesGroupBy.kurt
+   SeriesGroupBy.std
+   SeriesGroupBy.sum
+   SeriesGroupBy.var
+   SeriesGroupBy.tail
+   SeriesGroupBy.take
+   SeriesGroupBy.value_counts
+
+Plotting and visualization
+--------------------------
+.. autosummary::
+   :toctree: api/
+
+   DataFrameGroupBy.boxplot
+   DataFrameGroupBy.hist
+   SeriesGroupBy.hist
+   DataFrameGroupBy.plot
+   SeriesGroupBy.plot
diff --git a/pandas/doc/source/reference/index.rst b/pandas/doc/source/reference/index.rst
new file mode 100644
index 0000000000000..639bac4d40b70
--- /dev/null
+++ b/pandas/doc/source/reference/index.rst
@@ -0,0 +1,75 @@
+{{ header }}
+
+.. _api:
+
+=============
+API reference
+=============
+
+This page gives an overview of all public pandas objects, functions and
+methods. All classes and functions exposed in ``pandas.*`` namespace are public.
+
+The following subpackages are public.
+
+- ``pandas.errors``: Custom exception and warnings classes that are raised by pandas.
+- ``pandas.plotting``: Plotting public API.
+- ``pandas.testing``: Functions that are useful for writing tests involving pandas objects.
+- ``pandas.api.extensions``: Functions and classes for extending pandas objects.
+- ``pandas.api.indexers``: Functions and classes for rolling window indexers.
+- ``pandas.api.interchange``: DataFrame interchange protocol.
+- ``pandas.api.types``: Datatype classes and functions.
+- ``pandas.api.typing``: Classes that may be necessary for type-hinting.
+  These are classes that are encountered as intermediate results but should not be instantiated
+  directly by users. These classes are not to be confused with classes from the
+  `pandas-stubs <https://github.com/pandas-dev/pandas-stubs>`_ package
+  which has classes in addition to those that occur in pandas for type-hinting.
+
+In addition, public functions in ``pandas.io``, ``pandas.tseries``, ``pandas.util`` submodules
+are explicitly mentioned in the documentation. Further APIs in these modules are not guaranteed
+to be stable.
+
+
+.. warning::
+
+    The ``pandas.core``, ``pandas.compat`` top-level modules are PRIVATE. Stable functionality in such modules is not guaranteed.
+
+.. If you update this toctree, also update the manual toctree in the
+.. main index.rst.template
+
+.. toctree::
+   :maxdepth: 2
+
+   io
+   general_functions
+   series
+   frame
+   arrays
+   indexing
+   offset_frequency
+   window
+   groupby
+   resampling
+   style
+   plotting
+   options
+   extensions
+   testing
+   missing_value
+
+.. This is to prevent warnings in the doc build. We don't want to encourage
+.. these methods.
+
+..
+    .. toctree::
+
+        api/pandas.Index.nlevels
+        api/pandas.Index.sort
+
+
+.. Can't convince sphinx to generate toctree for this class attribute.
+.. So we do it manually to avoid a warning
+
+..
+    .. toctree::
+
+        api/pandas.api.extensions.ExtensionDtype.na_value
diff --git a/pandas/doc/source/reference/indexing.rst b/pandas/doc/source/reference/indexing.rst
new file mode 100644
index 0000000000000..7a4bc0f467f9a
--- /dev/null
+++ b/pandas/doc/source/reference/indexing.rst
@@ -0,0 +1,488 @@
+{{ header }}
+
+.. _api.indexing:
+
+=============
+Index objects
+=============
+
+Index
+-----
+.. currentmodule:: pandas
+
+**Many of these methods or variants thereof are available on the objects
+that contain an index (Series/DataFrame) and those should most likely be
+used before calling these methods directly.**
+
+.. autosummary::
+   :toctree: api/
+
+   Index
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.values
+   Index.is_monotonic_increasing
+   Index.is_monotonic_decreasing
+   Index.is_unique
+   Index.has_duplicates
+   Index.hasnans
+   Index.dtype
+   Index.inferred_type
+   Index.shape
+   Index.name
+   Index.names
+   Index.nbytes
+   Index.ndim
+   Index.size
+   Index.empty
+   Index.T
+   Index.memory_usage
+   Index.array
+
+Modifying and computations
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.all
+   Index.any
+   Index.argmin
+   Index.argmax
+   Index.copy
+   Index.delete
+   Index.drop
+   Index.drop_duplicates
+   Index.duplicated
+   Index.equals
+   Index.factorize
+   Index.identical
+   Index.insert
+   Index.is_
+   Index.min
+   Index.max
+   Index.reindex
+   Index.rename
+   Index.repeat
+   Index.where
+   Index.take
+   Index.putmask
+   Index.unique
+   Index.nunique
+   Index.value_counts
+
+Compatibility with MultiIndex
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.set_names
+   Index.droplevel
+
+Missing values
+~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.fillna
+   Index.dropna
+   Index.isna
+   Index.notna
+
+Conversion
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.astype
+   Index.item
+   Index.map
+   Index.ravel
+   Index.to_list
+   Index.to_series
+   Index.to_frame
+   Index.to_numpy
+   Index.view
+
+Sorting
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.argsort
+   Index.searchsorted
+   Index.sort_values
+
+Time-specific operations
+~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.shift
+
+Combining / joining / set operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.append
+   Index.join
+   Index.intersection
+   Index.union
+   Index.difference
+   Index.symmetric_difference
+
+Selecting
+~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Index.asof
+   Index.asof_locs
+   Index.get_indexer
+   Index.get_indexer_for
+   Index.get_indexer_non_unique
+   Index.get_level_values
+   Index.get_loc
+   Index.get_slice_bound
+   Index.isin
+   Index.slice_indexer
+   Index.slice_locs
+
+.. _api.numericindex:
+
+Numeric Index
+-------------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   RangeIndex
+
+.. We need this autosummary so that the methods are generated.
+.. Separate block, since they aren't classes.
+
+.. autosummary::
+   :toctree: api/
+
+   RangeIndex.start
+   RangeIndex.stop
+   RangeIndex.step
+   RangeIndex.from_range
+
+.. _api.categoricalindex:
+
+CategoricalIndex
+----------------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   CategoricalIndex
+
+Categorical components
+~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   CategoricalIndex.codes
+   CategoricalIndex.categories
+   CategoricalIndex.ordered
+   CategoricalIndex.rename_categories
+   CategoricalIndex.reorder_categories
+   CategoricalIndex.add_categories
+   CategoricalIndex.remove_categories
+   CategoricalIndex.remove_unused_categories
+   CategoricalIndex.set_categories
+   CategoricalIndex.as_ordered
+   CategoricalIndex.as_unordered
+
+Modifying and computations
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   CategoricalIndex.map
+   CategoricalIndex.equals
+
+.. _api.intervalindex:
+
+IntervalIndex
+-------------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   IntervalIndex
+
+IntervalIndex components
+~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   IntervalIndex.from_arrays
+   IntervalIndex.from_tuples
+   IntervalIndex.from_breaks
+   IntervalIndex.left
+   IntervalIndex.right
+   IntervalIndex.mid
+   IntervalIndex.closed
+   IntervalIndex.length
+   IntervalIndex.values
+   IntervalIndex.is_empty
+   IntervalIndex.is_non_overlapping_monotonic
+   IntervalIndex.is_overlapping
+   IntervalIndex.get_loc
+   IntervalIndex.get_indexer
+   IntervalIndex.set_closed
+   IntervalIndex.contains
+   IntervalIndex.overlaps
+   IntervalIndex.to_tuples
+
+.. _api.multiindex:
+
+MultiIndex
+----------
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   MultiIndex
+
+MultiIndex constructors
+~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   MultiIndex.from_arrays
+   MultiIndex.from_tuples
+   MultiIndex.from_product
+   MultiIndex.from_frame
+
+MultiIndex properties
+~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   MultiIndex.names
+   MultiIndex.levels
+   MultiIndex.codes
+   MultiIndex.nlevels
+   MultiIndex.levshape
+   MultiIndex.dtypes
+
+MultiIndex components
+~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   MultiIndex.set_levels
+   MultiIndex.set_codes
+   MultiIndex.to_flat_index
+   MultiIndex.to_frame
+   MultiIndex.sortlevel
+   MultiIndex.droplevel
+   MultiIndex.swaplevel
+   MultiIndex.reorder_levels
+   MultiIndex.remove_unused_levels
+   MultiIndex.drop
+   MultiIndex.copy
+   MultiIndex.append
+   MultiIndex.truncate
+
+MultiIndex selecting
+~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   MultiIndex.get_loc
+   MultiIndex.get_locs
+   MultiIndex.get_loc_level
+   MultiIndex.get_indexer
+   MultiIndex.get_level_values
+
+.. autosummary::
+   :toctree: api/
+
+   IndexSlice
+
+.. _api.datetimeindex:
+
+DatetimeIndex
+-------------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   DatetimeIndex
+
+Time/date components
+~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DatetimeIndex.year
+   DatetimeIndex.month
+   DatetimeIndex.day
+   DatetimeIndex.hour
+   DatetimeIndex.minute
+   DatetimeIndex.second
+   DatetimeIndex.microsecond
+   DatetimeIndex.nanosecond
+   DatetimeIndex.date
+   DatetimeIndex.time
+   DatetimeIndex.timetz
+   DatetimeIndex.dayofyear
+   DatetimeIndex.day_of_year
+   DatetimeIndex.dayofweek
+   DatetimeIndex.day_of_week
+   DatetimeIndex.weekday
+   DatetimeIndex.quarter
+   DatetimeIndex.tz
+   DatetimeIndex.freq
+   DatetimeIndex.freqstr
+   DatetimeIndex.is_month_start
+   DatetimeIndex.is_month_end
+   DatetimeIndex.is_quarter_start
+   DatetimeIndex.is_quarter_end
+   DatetimeIndex.is_year_start
+   DatetimeIndex.is_year_end
+   DatetimeIndex.is_leap_year
+   DatetimeIndex.inferred_freq
+
+Selecting
+~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DatetimeIndex.indexer_at_time
+   DatetimeIndex.indexer_between_time
+
+
+Time-specific operations
+~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DatetimeIndex.normalize
+   DatetimeIndex.strftime
+   DatetimeIndex.snap
+   DatetimeIndex.tz_convert
+   DatetimeIndex.tz_localize
+   DatetimeIndex.round
+   DatetimeIndex.floor
+   DatetimeIndex.ceil
+   DatetimeIndex.month_name
+   DatetimeIndex.day_name
+
+Conversion
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DatetimeIndex.as_unit
+   DatetimeIndex.to_period
+   DatetimeIndex.to_pydatetime
+   DatetimeIndex.to_series
+   DatetimeIndex.to_frame
+
+Methods
+~~~~~~~
+.. autosummary::
+    :toctree: api/
+
+    DatetimeIndex.mean
+    DatetimeIndex.std
+
+TimedeltaIndex
+--------------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   TimedeltaIndex
+
+Components
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   TimedeltaIndex.days
+   TimedeltaIndex.seconds
+   TimedeltaIndex.microseconds
+   TimedeltaIndex.nanoseconds
+   TimedeltaIndex.components
+   TimedeltaIndex.inferred_freq
+
+Conversion
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   TimedeltaIndex.as_unit
+   TimedeltaIndex.to_pytimedelta
+   TimedeltaIndex.to_series
+   TimedeltaIndex.round
+   TimedeltaIndex.floor
+   TimedeltaIndex.ceil
+   TimedeltaIndex.to_frame
+
+Methods
+~~~~~~~
+.. autosummary::
+    :toctree: api/
+
+    TimedeltaIndex.mean
+
+.. currentmodule:: pandas
+
+PeriodIndex
+-----------
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   PeriodIndex
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+    :toctree: api/
+
+    PeriodIndex.day
+    PeriodIndex.dayofweek
+    PeriodIndex.day_of_week
+    PeriodIndex.dayofyear
+    PeriodIndex.day_of_year
+    PeriodIndex.days_in_month
+    PeriodIndex.daysinmonth
+    PeriodIndex.end_time
+    PeriodIndex.freq
+    PeriodIndex.freqstr
+    PeriodIndex.hour
+    PeriodIndex.is_leap_year
+    PeriodIndex.minute
+    PeriodIndex.month
+    PeriodIndex.quarter
+    PeriodIndex.qyear
+    PeriodIndex.second
+    PeriodIndex.start_time
+    PeriodIndex.week
+    PeriodIndex.weekday
+    PeriodIndex.weekofyear
+    PeriodIndex.year
+
+Methods
+~~~~~~~
+.. autosummary::
+    :toctree: api/
+
+    PeriodIndex.asfreq
+    PeriodIndex.strftime
+    PeriodIndex.to_timestamp
+    PeriodIndex.from_fields
+    PeriodIndex.from_ordinals
diff --git a/pandas/doc/source/reference/io.rst b/pandas/doc/source/reference/io.rst
new file mode 100644
index 0000000000000..805fb8b783459
--- /dev/null
+++ b/pandas/doc/source/reference/io.rst
@@ -0,0 +1,207 @@
+{{ header }}
+
+.. _api.io:
+
+============
+Input/output
+============
+.. currentmodule:: pandas
+
+Pickling
+~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_pickle
+   DataFrame.to_pickle
+
+Flat file
+~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_table
+   read_csv
+   DataFrame.to_csv
+   read_fwf
+
+Clipboard
+~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_clipboard
+   DataFrame.to_clipboard
+
+Excel
+~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_excel
+   DataFrame.to_excel
+   ExcelFile
+   ExcelFile.book
+   ExcelFile.sheet_names
+   ExcelFile.parse
+
+.. currentmodule:: pandas.io.formats.style
+
+.. autosummary::
+   :toctree: api/
+
+   Styler.to_excel
+
+.. currentmodule:: pandas
+
+.. autosummary::
+   :toctree: api/
+
+   ExcelWriter
+
+.. currentmodule:: pandas
+
+JSON
+~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_json
+   json_normalize
+   DataFrame.to_json
+
+.. currentmodule:: pandas.io.json
+
+.. autosummary::
+   :toctree: api/
+
+   build_table_schema
+
+.. currentmodule:: pandas
+
+HTML
+~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_html
+   DataFrame.to_html
+
+.. currentmodule:: pandas.io.formats.style
+
+.. autosummary::
+   :toctree: api/
+
+   Styler.to_html
+
+.. currentmodule:: pandas
+
+XML
+~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_xml
+   DataFrame.to_xml
+
+Latex
+~~~~~
+.. autosummary::
+   :toctree: api/
+
+   DataFrame.to_latex
+
+.. currentmodule:: pandas.io.formats.style
+
+.. autosummary::
+   :toctree: api/
+
+   Styler.to_latex
+
+.. currentmodule:: pandas
+
+HDFStore: PyTables (HDF5)
+~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_hdf
+   HDFStore.put
+   HDFStore.append
+   HDFStore.get
+   HDFStore.select
+   HDFStore.info
+   HDFStore.keys
+   HDFStore.groups
+   HDFStore.walk
+
+.. warning::
+
+   One can store a subclass of :class:`DataFrame` or :class:`Series` to HDF5,
+   but the type of the subclass is lost upon storing.
+
+Feather
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_feather
+   DataFrame.to_feather
+
+Parquet
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_parquet
+   DataFrame.to_parquet
+
+ORC
+~~~
+.. autosummary::
+   :toctree: api/
+
+   read_orc
+   DataFrame.to_orc
+
+SAS
+~~~
+.. autosummary::
+   :toctree: api/
+
+   read_sas
+
+SPSS
+~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_spss
+
+SQL
+~~~
+.. autosummary::
+   :toctree: api/
+
+   read_sql_table
+   read_sql_query
+   read_sql
+   DataFrame.to_sql
+
+STATA
+~~~~~
+.. autosummary::
+   :toctree: api/
+
+   read_stata
+   DataFrame.to_stata
+
+.. currentmodule:: pandas.io.stata
+
+.. autosummary::
+   :toctree: api/
+
+   StataReader.data_label
+   StataReader.value_labels
+   StataReader.variable_labels
+   StataWriter.write_file
diff --git a/pandas/doc/source/reference/missing_value.rst b/pandas/doc/source/reference/missing_value.rst
new file mode 100644
index 0000000000000..3bf22aef765d1
--- /dev/null
+++ b/pandas/doc/source/reference/missing_value.rst
@@ -0,0 +1,24 @@
+{{ header }}
+
+.. _api.missing_value:
+
+==============
+Missing values
+==============
+.. currentmodule:: pandas
+
+NA is the way to represent missing values for nullable dtypes (see below):
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   NA
+
+NaT is the missing value for timedelta and datetime data (see below):
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   NaT
diff --git a/pandas/doc/source/reference/offset_frequency.rst b/pandas/doc/source/reference/offset_frequency.rst
new file mode 100644
index 0000000000000..8bb2c6ffe73be
--- /dev/null
+++ b/pandas/doc/source/reference/offset_frequency.rst
@@ -0,0 +1,1317 @@
+{{ header }}
+
+.. _api.dateoffsets:
+
+============
+Date offsets
+============
+.. currentmodule:: pandas.tseries.offsets
+
+DateOffset
+----------
+.. autosummary::
+   :toctree: api/
+
+    DateOffset
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    DateOffset.freqstr
+    DateOffset.kwds
+    DateOffset.name
+    DateOffset.nanos
+    DateOffset.normalize
+    DateOffset.rule_code
+    DateOffset.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    DateOffset.copy
+    DateOffset.is_on_offset
+    DateOffset.is_month_start
+    DateOffset.is_month_end
+    DateOffset.is_quarter_start
+    DateOffset.is_quarter_end
+    DateOffset.is_year_start
+    DateOffset.is_year_end
+
+BusinessDay
+-----------
+
+.. autosummary::
+   :toctree: api/
+
+    BusinessDay
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   BDay
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessDay.freqstr
+    BusinessDay.kwds
+    BusinessDay.name
+    BusinessDay.nanos
+    BusinessDay.normalize
+    BusinessDay.rule_code
+    BusinessDay.n
+    BusinessDay.weekmask
+    BusinessDay.holidays
+    BusinessDay.calendar
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessDay.copy
+    BusinessDay.is_on_offset
+    BusinessDay.is_month_start
+    BusinessDay.is_month_end
+    BusinessDay.is_quarter_start
+    BusinessDay.is_quarter_end
+    BusinessDay.is_year_start
+    BusinessDay.is_year_end
+
+BusinessHour
+------------
+.. autosummary::
+   :toctree: api/
+
+    BusinessHour
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessHour.freqstr
+    BusinessHour.kwds
+    BusinessHour.name
+    BusinessHour.nanos
+    BusinessHour.normalize
+    BusinessHour.rule_code
+    BusinessHour.n
+    BusinessHour.start
+    BusinessHour.end
+    BusinessHour.weekmask
+    BusinessHour.holidays
+    BusinessHour.calendar
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessHour.copy
+    BusinessHour.is_on_offset
+    BusinessHour.is_month_start
+    BusinessHour.is_month_end
+    BusinessHour.is_quarter_start
+    BusinessHour.is_quarter_end
+    BusinessHour.is_year_start
+    BusinessHour.is_year_end
+
+CustomBusinessDay
+-----------------
+
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessDay
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   CDay
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessDay.freqstr
+    CustomBusinessDay.kwds
+    CustomBusinessDay.name
+    CustomBusinessDay.nanos
+    CustomBusinessDay.normalize
+    CustomBusinessDay.rule_code
+    CustomBusinessDay.n
+    CustomBusinessDay.weekmask
+    CustomBusinessDay.calendar
+    CustomBusinessDay.holidays
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessDay.copy
+    CustomBusinessDay.is_on_offset
+    CustomBusinessDay.is_month_start
+    CustomBusinessDay.is_month_end
+    CustomBusinessDay.is_quarter_start
+    CustomBusinessDay.is_quarter_end
+    CustomBusinessDay.is_year_start
+    CustomBusinessDay.is_year_end
+
+CustomBusinessHour
+------------------
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessHour
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessHour.freqstr
+    CustomBusinessHour.kwds
+    CustomBusinessHour.name
+    CustomBusinessHour.nanos
+    CustomBusinessHour.normalize
+    CustomBusinessHour.rule_code
+    CustomBusinessHour.n
+    CustomBusinessHour.weekmask
+    CustomBusinessHour.calendar
+    CustomBusinessHour.holidays
+    CustomBusinessHour.start
+    CustomBusinessHour.end
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessHour.copy
+    CustomBusinessHour.is_on_offset
+    CustomBusinessHour.is_month_start
+    CustomBusinessHour.is_month_end
+    CustomBusinessHour.is_quarter_start
+    CustomBusinessHour.is_quarter_end
+    CustomBusinessHour.is_year_start
+    CustomBusinessHour.is_year_end
+
+MonthEnd
+--------
+.. autosummary::
+   :toctree: api/
+
+    MonthEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    MonthEnd.freqstr
+    MonthEnd.kwds
+    MonthEnd.name
+    MonthEnd.nanos
+    MonthEnd.normalize
+    MonthEnd.rule_code
+    MonthEnd.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    MonthEnd.copy
+    MonthEnd.is_on_offset
+    MonthEnd.is_month_start
+    MonthEnd.is_month_end
+    MonthEnd.is_quarter_start
+    MonthEnd.is_quarter_end
+    MonthEnd.is_year_start
+    MonthEnd.is_year_end
+
+MonthBegin
+----------
+.. autosummary::
+   :toctree: api/
+
+    MonthBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    MonthBegin.freqstr
+    MonthBegin.kwds
+    MonthBegin.name
+    MonthBegin.nanos
+    MonthBegin.normalize
+    MonthBegin.rule_code
+    MonthBegin.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    MonthBegin.copy
+    MonthBegin.is_on_offset
+    MonthBegin.is_month_start
+    MonthBegin.is_month_end
+    MonthBegin.is_quarter_start
+    MonthBegin.is_quarter_end
+    MonthBegin.is_year_start
+    MonthBegin.is_year_end
+
+BusinessMonthEnd
+----------------
+
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthEnd
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   BMonthEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthEnd.freqstr
+    BusinessMonthEnd.kwds
+    BusinessMonthEnd.name
+    BusinessMonthEnd.nanos
+    BusinessMonthEnd.normalize
+    BusinessMonthEnd.rule_code
+    BusinessMonthEnd.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthEnd.copy
+    BusinessMonthEnd.is_on_offset
+    BusinessMonthEnd.is_month_start
+    BusinessMonthEnd.is_month_end
+    BusinessMonthEnd.is_quarter_start
+    BusinessMonthEnd.is_quarter_end
+    BusinessMonthEnd.is_year_start
+    BusinessMonthEnd.is_year_end
+
+BusinessMonthBegin
+------------------
+
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthBegin
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   BMonthBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthBegin.freqstr
+    BusinessMonthBegin.kwds
+    BusinessMonthBegin.name
+    BusinessMonthBegin.nanos
+    BusinessMonthBegin.normalize
+    BusinessMonthBegin.rule_code
+    BusinessMonthBegin.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BusinessMonthBegin.copy
+    BusinessMonthBegin.is_on_offset
+    BusinessMonthBegin.is_month_start
+    BusinessMonthBegin.is_month_end
+    BusinessMonthBegin.is_quarter_start
+    BusinessMonthBegin.is_quarter_end
+    BusinessMonthBegin.is_year_start
+    BusinessMonthBegin.is_year_end
+
+CustomBusinessMonthEnd
+----------------------
+
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthEnd
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   CBMonthEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthEnd.freqstr
+    CustomBusinessMonthEnd.kwds
+    CustomBusinessMonthEnd.m_offset
+    CustomBusinessMonthEnd.name
+    CustomBusinessMonthEnd.nanos
+    CustomBusinessMonthEnd.normalize
+    CustomBusinessMonthEnd.rule_code
+    CustomBusinessMonthEnd.n
+    CustomBusinessMonthEnd.weekmask
+    CustomBusinessMonthEnd.calendar
+    CustomBusinessMonthEnd.holidays
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthEnd.copy
+    CustomBusinessMonthEnd.is_on_offset
+    CustomBusinessMonthEnd.is_month_start
+    CustomBusinessMonthEnd.is_month_end
+    CustomBusinessMonthEnd.is_quarter_start
+    CustomBusinessMonthEnd.is_quarter_end
+    CustomBusinessMonthEnd.is_year_start
+    CustomBusinessMonthEnd.is_year_end
+
+CustomBusinessMonthBegin
+------------------------
+
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthBegin
+
+Alias:
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/class_without_autosummary.rst
+
+   CBMonthBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthBegin.freqstr
+    CustomBusinessMonthBegin.kwds
+    CustomBusinessMonthBegin.m_offset
+    CustomBusinessMonthBegin.name
+    CustomBusinessMonthBegin.nanos
+    CustomBusinessMonthBegin.normalize
+    CustomBusinessMonthBegin.rule_code
+    CustomBusinessMonthBegin.n
+    CustomBusinessMonthBegin.weekmask
+    CustomBusinessMonthBegin.calendar
+    CustomBusinessMonthBegin.holidays
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    CustomBusinessMonthBegin.copy
+    CustomBusinessMonthBegin.is_on_offset
+    CustomBusinessMonthBegin.is_month_start
+    CustomBusinessMonthBegin.is_month_end
+    CustomBusinessMonthBegin.is_quarter_start
+    CustomBusinessMonthBegin.is_quarter_end
+    CustomBusinessMonthBegin.is_year_start
+    CustomBusinessMonthBegin.is_year_end
+
+SemiMonthEnd
+------------
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthEnd.freqstr
+    SemiMonthEnd.kwds
+    SemiMonthEnd.name
+    SemiMonthEnd.nanos
+    SemiMonthEnd.normalize
+    SemiMonthEnd.rule_code
+    SemiMonthEnd.n
+    SemiMonthEnd.day_of_month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthEnd.copy
+    SemiMonthEnd.is_on_offset
+    SemiMonthEnd.is_month_start
+    SemiMonthEnd.is_month_end
+    SemiMonthEnd.is_quarter_start
+    SemiMonthEnd.is_quarter_end
+    SemiMonthEnd.is_year_start
+    SemiMonthEnd.is_year_end
+
+SemiMonthBegin
+--------------
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthBegin.freqstr
+    SemiMonthBegin.kwds
+    SemiMonthBegin.name
+    SemiMonthBegin.nanos
+    SemiMonthBegin.normalize
+    SemiMonthBegin.rule_code
+    SemiMonthBegin.n
+    SemiMonthBegin.day_of_month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    SemiMonthBegin.copy
+    SemiMonthBegin.is_on_offset
+    SemiMonthBegin.is_month_start
+    SemiMonthBegin.is_month_end
+    SemiMonthBegin.is_quarter_start
+    SemiMonthBegin.is_quarter_end
+    SemiMonthBegin.is_year_start
+    SemiMonthBegin.is_year_end
+
+Week
+----
+.. autosummary::
+   :toctree: api/
+
+    Week
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Week.freqstr
+    Week.kwds
+    Week.name
+    Week.nanos
+    Week.normalize
+    Week.rule_code
+    Week.n
+    Week.weekday
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Week.copy
+    Week.is_on_offset
+    Week.is_month_start
+    Week.is_month_end
+    Week.is_quarter_start
+    Week.is_quarter_end
+    Week.is_year_start
+    Week.is_year_end
+
+WeekOfMonth
+-----------
+.. autosummary::
+   :toctree: api/
+
+    WeekOfMonth
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    WeekOfMonth.freqstr
+    WeekOfMonth.kwds
+    WeekOfMonth.name
+    WeekOfMonth.nanos
+    WeekOfMonth.normalize
+    WeekOfMonth.rule_code
+    WeekOfMonth.n
+    WeekOfMonth.week
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    WeekOfMonth.copy
+    WeekOfMonth.is_on_offset
+    WeekOfMonth.weekday
+    WeekOfMonth.is_month_start
+    WeekOfMonth.is_month_end
+    WeekOfMonth.is_quarter_start
+    WeekOfMonth.is_quarter_end
+    WeekOfMonth.is_year_start
+    WeekOfMonth.is_year_end
+
+LastWeekOfMonth
+---------------
+.. autosummary::
+   :toctree: api/
+
+    LastWeekOfMonth
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    LastWeekOfMonth.freqstr
+    LastWeekOfMonth.kwds
+    LastWeekOfMonth.name
+    LastWeekOfMonth.nanos
+    LastWeekOfMonth.normalize
+    LastWeekOfMonth.rule_code
+    LastWeekOfMonth.n
+    LastWeekOfMonth.weekday
+    LastWeekOfMonth.week
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    LastWeekOfMonth.copy
+    LastWeekOfMonth.is_on_offset
+    LastWeekOfMonth.is_month_start
+    LastWeekOfMonth.is_month_end
+    LastWeekOfMonth.is_quarter_start
+    LastWeekOfMonth.is_quarter_end
+    LastWeekOfMonth.is_year_start
+    LastWeekOfMonth.is_year_end
+
+BQuarterEnd
+-----------
+.. autosummary::
+   :toctree: api/
+
+    BQuarterEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BQuarterEnd.freqstr
+    BQuarterEnd.kwds
+    BQuarterEnd.name
+    BQuarterEnd.nanos
+    BQuarterEnd.normalize
+    BQuarterEnd.rule_code
+    BQuarterEnd.n
+    BQuarterEnd.startingMonth
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BQuarterEnd.copy
+    BQuarterEnd.is_on_offset
+    BQuarterEnd.is_month_start
+    BQuarterEnd.is_month_end
+    BQuarterEnd.is_quarter_start
+    BQuarterEnd.is_quarter_end
+    BQuarterEnd.is_year_start
+    BQuarterEnd.is_year_end
+
+BQuarterBegin
+-------------
+.. autosummary::
+   :toctree: api/
+
+    BQuarterBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BQuarterBegin.freqstr
+    BQuarterBegin.kwds
+    BQuarterBegin.name
+    BQuarterBegin.nanos
+    BQuarterBegin.normalize
+    BQuarterBegin.rule_code
+    BQuarterBegin.n
+    BQuarterBegin.startingMonth
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BQuarterBegin.copy
+    BQuarterBegin.is_on_offset
+    BQuarterBegin.is_month_start
+    BQuarterBegin.is_month_end
+    BQuarterBegin.is_quarter_start
+    BQuarterBegin.is_quarter_end
+    BQuarterBegin.is_year_start
+    BQuarterBegin.is_year_end
+
+QuarterEnd
+----------
+.. autosummary::
+   :toctree: api/
+
+    QuarterEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    QuarterEnd.freqstr
+    QuarterEnd.kwds
+    QuarterEnd.name
+    QuarterEnd.nanos
+    QuarterEnd.normalize
+    QuarterEnd.rule_code
+    QuarterEnd.n
+    QuarterEnd.startingMonth
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    QuarterEnd.copy
+    QuarterEnd.is_on_offset
+    QuarterEnd.is_month_start
+    QuarterEnd.is_month_end
+    QuarterEnd.is_quarter_start
+    QuarterEnd.is_quarter_end
+    QuarterEnd.is_year_start
+    QuarterEnd.is_year_end
+
+QuarterBegin
+------------
+.. autosummary::
+   :toctree: api/
+
+    QuarterBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    QuarterBegin.freqstr
+    QuarterBegin.kwds
+    QuarterBegin.name
+    QuarterBegin.nanos
+    QuarterBegin.normalize
+    QuarterBegin.rule_code
+    QuarterBegin.n
+    QuarterBegin.startingMonth
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    QuarterBegin.copy
+    QuarterBegin.is_on_offset
+    QuarterBegin.is_month_start
+    QuarterBegin.is_month_end
+    QuarterBegin.is_quarter_start
+    QuarterBegin.is_quarter_end
+    QuarterBegin.is_year_start
+    QuarterBegin.is_year_end
+
+BYearEnd
+--------
+.. autosummary::
+   :toctree: api/
+
+    BYearEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BYearEnd.freqstr
+    BYearEnd.kwds
+    BYearEnd.name
+    BYearEnd.nanos
+    BYearEnd.normalize
+    BYearEnd.rule_code
+    BYearEnd.n
+    BYearEnd.month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BYearEnd.copy
+    BYearEnd.is_on_offset
+    BYearEnd.is_month_start
+    BYearEnd.is_month_end
+    BYearEnd.is_quarter_start
+    BYearEnd.is_quarter_end
+    BYearEnd.is_year_start
+    BYearEnd.is_year_end
+
+BYearBegin
+----------
+.. autosummary::
+   :toctree: api/
+
+    BYearBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BYearBegin.freqstr
+    BYearBegin.kwds
+    BYearBegin.name
+    BYearBegin.nanos
+    BYearBegin.normalize
+    BYearBegin.rule_code
+    BYearBegin.n
+    BYearBegin.month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    BYearBegin.copy
+    BYearBegin.is_on_offset
+    BYearBegin.is_month_start
+    BYearBegin.is_month_end
+    BYearBegin.is_quarter_start
+    BYearBegin.is_quarter_end
+    BYearBegin.is_year_start
+    BYearBegin.is_year_end
+
+YearEnd
+-------
+.. autosummary::
+   :toctree: api/
+
+    YearEnd
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    YearEnd.freqstr
+    YearEnd.kwds
+    YearEnd.name
+    YearEnd.nanos
+    YearEnd.normalize
+    YearEnd.rule_code
+    YearEnd.n
+    YearEnd.month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    YearEnd.copy
+    YearEnd.is_on_offset
+    YearEnd.is_month_start
+    YearEnd.is_month_end
+    YearEnd.is_quarter_start
+    YearEnd.is_quarter_end
+    YearEnd.is_year_start
+    YearEnd.is_year_end
+
+YearBegin
+---------
+.. autosummary::
+   :toctree: api/
+
+    YearBegin
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    YearBegin.freqstr
+    YearBegin.kwds
+    YearBegin.name
+    YearBegin.nanos
+    YearBegin.normalize
+    YearBegin.rule_code
+    YearBegin.n
+    YearBegin.month
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    YearBegin.copy
+    YearBegin.is_on_offset
+    YearBegin.is_month_start
+    YearBegin.is_month_end
+    YearBegin.is_quarter_start
+    YearBegin.is_quarter_end
+    YearBegin.is_year_start
+    YearBegin.is_year_end
+
+FY5253
+------
+.. autosummary::
+   :toctree: api/
+
+    FY5253
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    FY5253.freqstr
+    FY5253.kwds
+    FY5253.name
+    FY5253.nanos
+    FY5253.normalize
+    FY5253.rule_code
+    FY5253.n
+    FY5253.startingMonth
+    FY5253.variation
+    FY5253.weekday
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    FY5253.copy
+    FY5253.get_rule_code_suffix
+    FY5253.get_year_end
+    FY5253.is_on_offset
+    FY5253.is_month_start
+    FY5253.is_month_end
+    FY5253.is_quarter_start
+    FY5253.is_quarter_end
+    FY5253.is_year_start
+    FY5253.is_year_end
+
+FY5253Quarter
+-------------
+.. autosummary::
+   :toctree: api/
+
+    FY5253Quarter
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    FY5253Quarter.freqstr
+    FY5253Quarter.kwds
+    FY5253Quarter.name
+    FY5253Quarter.nanos
+    FY5253Quarter.normalize
+    FY5253Quarter.rule_code
+    FY5253Quarter.n
+    FY5253Quarter.qtr_with_extra_week
+    FY5253Quarter.startingMonth
+    FY5253Quarter.variation
+    FY5253Quarter.weekday
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    FY5253Quarter.copy
+    FY5253Quarter.get_rule_code_suffix
+    FY5253Quarter.get_weeks
+    FY5253Quarter.is_on_offset
+    FY5253Quarter.year_has_extra_week
+    FY5253Quarter.is_month_start
+    FY5253Quarter.is_month_end
+    FY5253Quarter.is_quarter_start
+    FY5253Quarter.is_quarter_end
+    FY5253Quarter.is_year_start
+    FY5253Quarter.is_year_end
+
+Easter
+------
+.. autosummary::
+   :toctree: api/
+
+    Easter
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Easter.freqstr
+    Easter.kwds
+    Easter.name
+    Easter.nanos
+    Easter.normalize
+    Easter.rule_code
+    Easter.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Easter.copy
+    Easter.is_on_offset
+    Easter.is_month_start
+    Easter.is_month_end
+    Easter.is_quarter_start
+    Easter.is_quarter_end
+    Easter.is_year_start
+    Easter.is_year_end
+
+Tick
+----
+.. autosummary::
+   :toctree: api/
+
+    Tick
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Tick.freqstr
+    Tick.kwds
+    Tick.name
+    Tick.nanos
+    Tick.normalize
+    Tick.rule_code
+    Tick.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Tick.copy
+    Tick.is_on_offset
+    Tick.is_month_start
+    Tick.is_month_end
+    Tick.is_quarter_start
+    Tick.is_quarter_end
+    Tick.is_year_start
+    Tick.is_year_end
+
+Day
+---
+.. autosummary::
+   :toctree: api/
+
+    Day
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Day.freqstr
+    Day.kwds
+    Day.name
+    Day.nanos
+    Day.normalize
+    Day.rule_code
+    Day.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Day.copy
+    Day.is_on_offset
+    Day.is_month_start
+    Day.is_month_end
+    Day.is_quarter_start
+    Day.is_quarter_end
+    Day.is_year_start
+    Day.is_year_end
+
+Hour
+----
+.. autosummary::
+   :toctree: api/
+
+    Hour
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Hour.freqstr
+    Hour.kwds
+    Hour.name
+    Hour.nanos
+    Hour.normalize
+    Hour.rule_code
+    Hour.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Hour.copy
+    Hour.is_on_offset
+    Hour.is_month_start
+    Hour.is_month_end
+    Hour.is_quarter_start
+    Hour.is_quarter_end
+    Hour.is_year_start
+    Hour.is_year_end
+
+Minute
+------
+.. autosummary::
+   :toctree: api/
+
+    Minute
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Minute.freqstr
+    Minute.kwds
+    Minute.name
+    Minute.nanos
+    Minute.normalize
+    Minute.rule_code
+    Minute.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Minute.copy
+    Minute.is_on_offset
+    Minute.is_month_start
+    Minute.is_month_end
+    Minute.is_quarter_start
+    Minute.is_quarter_end
+    Minute.is_year_start
+    Minute.is_year_end
+
+Second
+------
+.. autosummary::
+   :toctree: api/
+
+    Second
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Second.freqstr
+    Second.kwds
+    Second.name
+    Second.nanos
+    Second.normalize
+    Second.rule_code
+    Second.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Second.copy
+    Second.is_on_offset
+    Second.is_month_start
+    Second.is_month_end
+    Second.is_quarter_start
+    Second.is_quarter_end
+    Second.is_year_start
+    Second.is_year_end
+
+Milli
+-----
+.. autosummary::
+   :toctree: api/
+
+    Milli
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Milli.freqstr
+    Milli.kwds
+    Milli.name
+    Milli.nanos
+    Milli.normalize
+    Milli.rule_code
+    Milli.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Milli.copy
+    Milli.is_on_offset
+    Milli.is_month_start
+    Milli.is_month_end
+    Milli.is_quarter_start
+    Milli.is_quarter_end
+    Milli.is_year_start
+    Milli.is_year_end
+
+Micro
+-----
+.. autosummary::
+   :toctree: api/
+
+    Micro
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Micro.freqstr
+    Micro.kwds
+    Micro.name
+    Micro.nanos
+    Micro.normalize
+    Micro.rule_code
+    Micro.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Micro.copy
+    Micro.is_on_offset
+    Micro.is_month_start
+    Micro.is_month_end
+    Micro.is_quarter_start
+    Micro.is_quarter_end
+    Micro.is_year_start
+    Micro.is_year_end
+
+Nano
+----
+.. autosummary::
+   :toctree: api/
+
+    Nano
+
+Properties
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Nano.freqstr
+    Nano.kwds
+    Nano.name
+    Nano.nanos
+    Nano.normalize
+    Nano.rule_code
+    Nano.n
+
+Methods
+~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+    Nano.copy
+    Nano.is_on_offset
+    Nano.is_month_start
+    Nano.is_month_end
+    Nano.is_quarter_start
+    Nano.is_quarter_end
+    Nano.is_year_start
+    Nano.is_year_end
+
+.. _api.frequencies:
+
+===========
+Frequencies
+===========
+.. currentmodule:: pandas.tseries.frequencies
+
+.. _api.offsets:
+
+.. autosummary::
+   :toctree: api/
+
+   to_offset
diff --git a/pandas/doc/source/reference/options.rst b/pandas/doc/source/reference/options.rst
new file mode 100644
index 0000000000000..b91ea3ffb62f5
--- /dev/null
+++ b/pandas/doc/source/reference/options.rst
@@ -0,0 +1,28 @@
+{{ header }}
+
+.. _api.options:
+
+====================
+Options and settings
+====================
+.. currentmodule:: pandas
+
+API for configuring global behavior. See :ref:`the User Guide <options>` for more.
+
+Working with options
+--------------------
+.. autosummary::
+   :toctree: api/
+
+   describe_option
+   reset_option
+   get_option
+   set_option
+   option_context
+
+Numeric formatting
+------------------
+.. autosummary::
+   :toctree: api/
+
+   set_eng_float_format
diff --git a/pandas/doc/source/reference/plotting.rst b/pandas/doc/source/reference/plotting.rst
new file mode 100644
index 0000000000000..632b39a1fa858
--- /dev/null
+++ b/pandas/doc/source/reference/plotting.rst
@@ -0,0 +1,26 @@
+{{ header }}
+
+.. _api.plotting:
+
+========
+Plotting
+========
+.. currentmodule:: pandas.plotting
+
+The following functions are contained in the ``pandas.plotting`` module.
+
+.. autosummary::
+   :toctree: api/
+
+   andrews_curves
+   autocorrelation_plot
+   bootstrap_plot
+   boxplot
+   deregister_matplotlib_converters
+   lag_plot
+   parallel_coordinates
+   plot_params
+   radviz
+   register_matplotlib_converters
+   scatter_matrix
+   table
diff --git a/pandas/doc/source/reference/resampling.rst b/pandas/doc/source/reference/resampling.rst
new file mode 100644
index 0000000000000..2e0717081b129
--- /dev/null
+++ b/pandas/doc/source/reference/resampling.rst
@@ -0,0 +1,64 @@
+{{ header }}
+
+.. _api.resampling:
+
+==========
+Resampling
+==========
+.. currentmodule:: pandas.core.resample
+
+:class:`pandas.api.typing.Resampler` instances are returned by
+resample calls: :func:`pandas.DataFrame.resample`, :func:`pandas.Series.resample`.
+
+Indexing, iteration
+~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Resampler.__iter__
+   Resampler.groups
+   Resampler.indices
+   Resampler.get_group
+
+Function application
+~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Resampler.apply
+   Resampler.aggregate
+   Resampler.transform
+   Resampler.pipe
+
+Upsampling
+~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Resampler.ffill
+   Resampler.bfill
+   Resampler.nearest
+   Resampler.asfreq
+   Resampler.interpolate
+
+Computations / descriptive stats
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autosummary::
+   :toctree: api/
+
+   Resampler.count
+   Resampler.nunique
+   Resampler.first
+   Resampler.last
+   Resampler.max
+   Resampler.mean
+   Resampler.median
+   Resampler.min
+   Resampler.ohlc
+   Resampler.prod
+   Resampler.size
+   Resampler.sem
+   Resampler.std
+   Resampler.sum
+   Resampler.var
+   Resampler.quantile
diff --git a/pandas/doc/source/reference/series.rst b/pandas/doc/source/reference/series.rst
new file mode 100644
index 0000000000000..6006acc8f5e16
--- /dev/null
+++ b/pandas/doc/source/reference/series.rst
@@ -0,0 +1,638 @@
+{{ header }}
+
+.. _api.series:
+
+======
+Series
+======
+.. currentmodule:: pandas
+
+Constructor
+-----------
+.. autosummary::
+   :toctree: api/
+
+   Series
+
+Attributes
+----------
+**Axes**
+
+.. autosummary::
+   :toctree: api/
+
+   Series.index
+   Series.array
+   Series.values
+   Series.dtype
+   Series.info
+   Series.shape
+   Series.nbytes
+   Series.ndim
+   Series.size
+   Series.T
+   Series.memory_usage
+   Series.hasnans
+   Series.empty
+   Series.dtypes
+   Series.name
+   Series.flags
+   Series.set_flags
+
+Conversion
+----------
+.. autosummary::
+   :toctree: api/
+
+   Series.astype
+   Series.convert_dtypes
+   Series.infer_objects
+   Series.copy
+   Series.to_numpy
+   Series.to_period
+   Series.to_timestamp
+   Series.to_list
+   Series.__array__
+
+Indexing, iteration
+-------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.get
+   Series.at
+   Series.iat
+   Series.loc
+   Series.iloc
+   Series.__iter__
+   Series.items
+   Series.keys
+   Series.pop
+   Series.item
+   Series.xs
+
+For more information on ``.at``, ``.iat``, ``.loc``, and
+``.iloc``,  see the :ref:`indexing documentation <indexing>`.
+
+Binary operator functions
+-------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.add
+   Series.sub
+   Series.mul
+   Series.div
+   Series.truediv
+   Series.floordiv
+   Series.mod
+   Series.pow
+   Series.radd
+   Series.rsub
+   Series.rmul
+   Series.rdiv
+   Series.rtruediv
+   Series.rfloordiv
+   Series.rmod
+   Series.rpow
+   Series.combine
+   Series.combine_first
+   Series.round
+   Series.lt
+   Series.gt
+   Series.le
+   Series.ge
+   Series.ne
+   Series.eq
+   Series.product
+   Series.dot
+
+Function application, GroupBy & window
+--------------------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.apply
+   Series.agg
+   Series.aggregate
+   Series.transform
+   Series.map
+   Series.groupby
+   Series.rolling
+   Series.expanding
+   Series.ewm
+   Series.pipe
+
+.. _api.series.stats:
+
+Computations / descriptive stats
+--------------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.abs
+   Series.all
+   Series.any
+   Series.autocorr
+   Series.between
+   Series.clip
+   Series.corr
+   Series.count
+   Series.cov
+   Series.cummax
+   Series.cummin
+   Series.cumprod
+   Series.cumsum
+   Series.describe
+   Series.diff
+   Series.factorize
+   Series.kurt
+   Series.max
+   Series.mean
+   Series.median
+   Series.min
+   Series.mode
+   Series.nlargest
+   Series.nsmallest
+   Series.pct_change
+   Series.prod
+   Series.quantile
+   Series.rank
+   Series.sem
+   Series.skew
+   Series.std
+   Series.sum
+   Series.var
+   Series.kurtosis
+   Series.unique
+   Series.nunique
+   Series.is_unique
+   Series.is_monotonic_increasing
+   Series.is_monotonic_decreasing
+   Series.value_counts
+
+Reindexing / selection / label manipulation
+-------------------------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.align
+   Series.case_when
+   Series.drop
+   Series.droplevel
+   Series.drop_duplicates
+   Series.duplicated
+   Series.equals
+   Series.head
+   Series.idxmax
+   Series.idxmin
+   Series.isin
+   Series.reindex
+   Series.reindex_like
+   Series.rename
+   Series.rename_axis
+   Series.reset_index
+   Series.sample
+   Series.set_axis
+   Series.take
+   Series.tail
+   Series.truncate
+   Series.where
+   Series.mask
+   Series.add_prefix
+   Series.add_suffix
+   Series.filter
+
+Missing data handling
+---------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.bfill
+   Series.dropna
+   Series.ffill
+   Series.fillna
+   Series.interpolate
+   Series.isna
+   Series.isnull
+   Series.notna
+   Series.notnull
+   Series.replace
+
+Reshaping, sorting
+------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.argsort
+   Series.argmin
+   Series.argmax
+   Series.reorder_levels
+   Series.sort_values
+   Series.sort_index
+   Series.swaplevel
+   Series.unstack
+   Series.explode
+   Series.searchsorted
+   Series.repeat
+   Series.squeeze
+
+Combining / comparing / joining / merging
+-----------------------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.compare
+   Series.update
+
+Time Series-related
+-------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.asfreq
+   Series.asof
+   Series.shift
+   Series.first_valid_index
+   Series.last_valid_index
+   Series.resample
+   Series.tz_convert
+   Series.tz_localize
+   Series.at_time
+   Series.between_time
+
+Accessors
+---------
+
+pandas provides dtype-specific methods under various accessors.
+These are separate namespaces within :class:`Series` that only apply
+to specific data types.
+
+.. autosummary::
+    :toctree: api/
+    :nosignatures:
+    :template: autosummary/accessor.rst
+
+    Series.str
+    Series.cat
+    Series.dt
+    Series.sparse
+    DataFrame.sparse
+    Index.str
+
+
+=========================== =================================
+Data Type                   Accessor
+=========================== =================================
+Datetime, Timedelta, Period :ref:`dt <api.series.dt>`
+String                      :ref:`str <api.series.str>`
+Categorical                 :ref:`cat <api.series.cat>`
+Sparse                      :ref:`sparse <api.series.sparse>`
+=========================== =================================
+
+.. _api.series.dt:
+
+Datetimelike properties
+~~~~~~~~~~~~~~~~~~~~~~~
+
+``Series.dt`` can be used to access the values of the series as
+datetimelike and return several properties.
+These can be accessed like ``Series.dt.<property>``.
+
+Datetime properties
+^^^^^^^^^^^^^^^^^^^
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.dt.date
+   Series.dt.time
+   Series.dt.timetz
+   Series.dt.year
+   Series.dt.month
+   Series.dt.day
+   Series.dt.hour
+   Series.dt.minute
+   Series.dt.second
+   Series.dt.microsecond
+   Series.dt.nanosecond
+   Series.dt.dayofweek
+   Series.dt.day_of_week
+   Series.dt.weekday
+   Series.dt.dayofyear
+   Series.dt.day_of_year
+   Series.dt.days_in_month
+   Series.dt.quarter
+   Series.dt.is_month_start
+   Series.dt.is_month_end
+   Series.dt.is_quarter_start
+   Series.dt.is_quarter_end
+   Series.dt.is_year_start
+   Series.dt.is_year_end
+   Series.dt.is_leap_year
+   Series.dt.daysinmonth
+   Series.dt.days_in_month
+   Series.dt.tz
+   Series.dt.freq
+   Series.dt.unit
+
+Datetime methods
+^^^^^^^^^^^^^^^^
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.dt.isocalendar
+   Series.dt.to_period
+   Series.dt.to_pydatetime
+   Series.dt.tz_localize
+   Series.dt.tz_convert
+   Series.dt.normalize
+   Series.dt.strftime
+   Series.dt.round
+   Series.dt.floor
+   Series.dt.ceil
+   Series.dt.month_name
+   Series.dt.day_name
+   Series.dt.as_unit
+
+Period properties
+^^^^^^^^^^^^^^^^^
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.dt.qyear
+   Series.dt.start_time
+   Series.dt.end_time
+
+Timedelta properties
+^^^^^^^^^^^^^^^^^^^^
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.dt.days
+   Series.dt.seconds
+   Series.dt.microseconds
+   Series.dt.nanoseconds
+   Series.dt.components
+   Series.dt.unit
+
+Timedelta methods
+^^^^^^^^^^^^^^^^^
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.dt.to_pytimedelta
+   Series.dt.total_seconds
+   Series.dt.as_unit
+
+
+.. _api.series.str:
+
+String handling
+~~~~~~~~~~~~~~~
+
+``Series.str`` can be used to access the values of the series as
+strings and apply several methods to it. These can be accessed like
+``Series.str.<function/property>``.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.str.capitalize
+   Series.str.casefold
+   Series.str.cat
+   Series.str.center
+   Series.str.contains
+   Series.str.count
+   Series.str.decode
+   Series.str.encode
+   Series.str.endswith
+   Series.str.extract
+   Series.str.extractall
+   Series.str.find
+   Series.str.findall
+   Series.str.fullmatch
+   Series.str.get
+   Series.str.index
+   Series.str.join
+   Series.str.len
+   Series.str.ljust
+   Series.str.lower
+   Series.str.lstrip
+   Series.str.match
+   Series.str.normalize
+   Series.str.pad
+   Series.str.partition
+   Series.str.removeprefix
+   Series.str.removesuffix
+   Series.str.repeat
+   Series.str.replace
+   Series.str.rfind
+   Series.str.rindex
+   Series.str.rjust
+   Series.str.rpartition
+   Series.str.rstrip
+   Series.str.slice
+   Series.str.slice_replace
+   Series.str.split
+   Series.str.rsplit
+   Series.str.startswith
+   Series.str.strip
+   Series.str.swapcase
+   Series.str.title
+   Series.str.translate
+   Series.str.upper
+   Series.str.wrap
+   Series.str.zfill
+   Series.str.isalnum
+   Series.str.isalpha
+   Series.str.isdigit
+   Series.str.isspace
+   Series.str.islower
+   Series.str.isupper
+   Series.str.istitle
+   Series.str.isnumeric
+   Series.str.isdecimal
+   Series.str.get_dummies
+
+.. _api.series.cat:
+
+Categorical accessor
+~~~~~~~~~~~~~~~~~~~~
+
+Categorical-dtype specific methods and attributes are available under
+the ``Series.cat`` accessor.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.cat.categories
+   Series.cat.ordered
+   Series.cat.codes
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.cat.rename_categories
+   Series.cat.reorder_categories
+   Series.cat.add_categories
+   Series.cat.remove_categories
+   Series.cat.remove_unused_categories
+   Series.cat.set_categories
+   Series.cat.as_ordered
+   Series.cat.as_unordered
+
+
+.. _api.series.sparse:
+
+Sparse accessor
+~~~~~~~~~~~~~~~
+
+Sparse-dtype specific methods and attributes are provided under the
+``Series.sparse`` accessor.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.sparse.npoints
+   Series.sparse.density
+   Series.sparse.fill_value
+   Series.sparse.sp_values
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.sparse.from_coo
+   Series.sparse.to_coo
+
+
+.. _api.series.list:
+
+List accessor
+~~~~~~~~~~~~~
+
+Arrow list-dtype specific methods and attributes are provided under the
+``Series.list`` accessor.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.list.flatten
+   Series.list.len
+   Series.list.__getitem__
+
+
+.. _api.series.struct:
+
+Struct accessor
+~~~~~~~~~~~~~~~
+
+Arrow struct-dtype specific methods and attributes are provided under the
+``Series.struct`` accessor.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_attribute.rst
+
+   Series.struct.dtypes
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.struct.field
+   Series.struct.explode
+
+
+.. _api.series.flags:
+
+Flags
+~~~~~
+
+Flags refer to attributes of the pandas object. Properties of the dataset (like
+the date is was recorded, the URL it was accessed from, etc.) should be stored
+in :attr:`Series.attrs`.
+
+.. autosummary::
+   :toctree: api/
+
+   Flags
+
+.. _api.series.metadata:
+
+Metadata
+~~~~~~~~
+
+:attr:`Series.attrs` is a dictionary for storing global metadata for this Series.
+
+.. warning:: ``Series.attrs`` is considered experimental and may change without warning.
+
+.. autosummary::
+   :toctree: api/
+
+   Series.attrs
+
+
+Plotting
+--------
+``Series.plot`` is both a callable method and a namespace attribute for
+specific plotting methods of the form ``Series.plot.<kind>``.
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_callable.rst
+
+   Series.plot
+
+.. autosummary::
+   :toctree: api/
+   :template: autosummary/accessor_method.rst
+
+   Series.plot.area
+   Series.plot.bar
+   Series.plot.barh
+   Series.plot.box
+   Series.plot.density
+   Series.plot.hist
+   Series.plot.kde
+   Series.plot.line
+   Series.plot.pie
+
+.. autosummary::
+   :toctree: api/
+
+   Series.hist
+
+Serialization / IO / conversion
+-------------------------------
+.. autosummary::
+   :toctree: api/
+
+   Series.to_pickle
+   Series.to_csv
+   Series.to_dict
+   Series.to_excel
+   Series.to_frame
+   Series.to_xarray
+   Series.to_hdf
+   Series.to_sql
+   Series.to_json
+   Series.to_string
+   Series.to_clipboard
+   Series.to_latex
+   Series.to_markdown
diff --git a/pandas/doc/source/reference/style.rst b/pandas/doc/source/reference/style.rst
new file mode 100644
index 0000000000000..742263c788c2f
--- /dev/null
+++ b/pandas/doc/source/reference/style.rst
@@ -0,0 +1,85 @@
+{{ header }}
+
+.. _api.style:
+
+=====
+Style
+=====
+.. currentmodule:: pandas.io.formats.style
+
+``Styler`` objects are returned by :attr:`pandas.DataFrame.style`.
+
+Styler constructor
+------------------
+.. autosummary::
+   :toctree: api/
+
+   Styler
+   Styler.from_custom_template
+
+Styler properties
+-----------------
+.. autosummary::
+   :toctree: api/
+
+   Styler.env
+   Styler.template_html
+   Styler.template_html_style
+   Styler.template_html_table
+   Styler.template_latex
+   Styler.template_typst
+   Styler.template_string
+   Styler.loader
+
+Style application
+-----------------
+.. autosummary::
+   :toctree: api/
+
+   Styler.apply
+   Styler.map
+   Styler.apply_index
+   Styler.map_index
+   Styler.format
+   Styler.format_index
+   Styler.format_index_names
+   Styler.relabel_index
+   Styler.hide
+   Styler.concat
+   Styler.set_td_classes
+   Styler.set_table_styles
+   Styler.set_table_attributes
+   Styler.set_tooltips
+   Styler.set_caption
+   Styler.set_sticky
+   Styler.set_properties
+   Styler.set_uuid
+   Styler.clear
+   Styler.pipe
+
+Builtin styles
+--------------
+.. autosummary::
+   :toctree: api/
+
+   Styler.highlight_null
+   Styler.highlight_max
+   Styler.highlight_min
+   Styler.highlight_between
+   Styler.highlight_quantile
+   Styler.background_gradient
+   Styler.text_gradient
+   Styler.bar
+
+Style export and import
+-----------------------
+.. autosummary::
+   :toctree: api/
+
+   Styler.to_html
+   Styler.to_latex
+   Styler.to_typst
+   Styler.to_excel
+   Styler.to_string
+   Styler.export
+   Styler.use
diff --git a/pandas/doc/source/reference/testing.rst b/pandas/doc/source/reference/testing.rst
new file mode 100644
index 0000000000000..1f164d1aa98b4
--- /dev/null
+++ b/pandas/doc/source/reference/testing.rst
@@ -0,0 +1,79 @@
+{{ header }}
+
+.. _api.testing:
+
+=======
+Testing
+=======
+.. currentmodule:: pandas
+
+.. _api.general.testing:
+
+Assertion functions
+-------------------
+.. autosummary::
+   :toctree: api/
+
+   testing.assert_frame_equal
+   testing.assert_series_equal
+   testing.assert_index_equal
+   testing.assert_extension_array_equal
+
+Exceptions and warnings
+-----------------------
+.. autosummary::
+   :toctree: api/
+
+   errors.AbstractMethodError
+   errors.AttributeConflictWarning
+   errors.CategoricalConversionWarning
+   errors.ChainedAssignmentError
+   errors.ClosedFileError
+   errors.CSSWarning
+   errors.DatabaseError
+   errors.DataError
+   errors.DtypeWarning
+   errors.DuplicateLabelError
+   errors.EmptyDataError
+   errors.IncompatibilityWarning
+   errors.IndexingError
+   errors.InvalidColumnName
+   errors.InvalidComparison
+   errors.InvalidIndexError
+   errors.InvalidVersion
+   errors.IntCastingNaNError
+   errors.LossySetitemError
+   errors.MergeError
+   errors.NoBufferPresent
+   errors.NullFrequencyError
+   errors.NumbaUtilError
+   errors.NumExprClobberingError
+   errors.OptionError
+   errors.OutOfBoundsDatetime
+   errors.OutOfBoundsTimedelta
+   errors.ParserError
+   errors.ParserWarning
+   errors.PerformanceWarning
+   errors.PossibleDataLossError
+   errors.PossiblePrecisionLoss
+   errors.PyperclipException
+   errors.PyperclipWindowsException
+   errors.SpecificationError
+   errors.UndefinedVariableError
+   errors.UnsortedIndexError
+   errors.UnsupportedFunctionCall
+   errors.ValueLabelTypeMismatch
+
+Bug report function
+-------------------
+.. autosummary::
+   :toctree: api/
+
+   show_versions
+
+Test suite runner
+-----------------
+.. autosummary::
+   :toctree: api/
+
+   test
diff --git a/pandas/doc/source/reference/window.rst b/pandas/doc/source/reference/window.rst
new file mode 100644
index 0000000000000..2aeb57faac112
--- /dev/null
+++ b/pandas/doc/source/reference/window.rst
@@ -0,0 +1,119 @@
+{{ header }}
+
+.. _api.window:
+
+======
+Window
+======
+
+:class:`pandas.api.typing.Rolling` instances are returned by ``.rolling`` calls:
+:func:`pandas.DataFrame.rolling` and :func:`pandas.Series.rolling`.
+:class:`pandas.api.typing.Expanding` instances are returned by ``.expanding`` calls:
+:func:`pandas.DataFrame.expanding` and :func:`pandas.Series.expanding`.
+:class:`pandas.api.typing.ExponentialMovingWindow` instances are returned by ``.ewm``
+calls: :func:`pandas.DataFrame.ewm` and :func:`pandas.Series.ewm`.
+
+.. _api.functions_rolling:
+
+Rolling window functions
+------------------------
+.. currentmodule:: pandas.core.window.rolling
+
+.. autosummary::
+   :toctree: api/
+
+   Rolling.count
+   Rolling.sum
+   Rolling.mean
+   Rolling.median
+   Rolling.var
+   Rolling.std
+   Rolling.min
+   Rolling.max
+   Rolling.first
+   Rolling.last
+   Rolling.corr
+   Rolling.cov
+   Rolling.skew
+   Rolling.kurt
+   Rolling.apply
+   Rolling.pipe
+   Rolling.aggregate
+   Rolling.quantile
+   Rolling.sem
+   Rolling.rank
+
+.. _api.functions_window:
+
+Weighted window functions
+-------------------------
+.. currentmodule:: pandas.core.window.rolling
+
+.. autosummary::
+   :toctree: api/
+
+   Window.mean
+   Window.sum
+   Window.var
+   Window.std
+
+.. _api.functions_expanding:
+
+Expanding window functions
+--------------------------
+.. currentmodule:: pandas.core.window.expanding
+
+.. autosummary::
+   :toctree: api/
+
+   Expanding.count
+   Expanding.sum
+   Expanding.mean
+   Expanding.median
+   Expanding.var
+   Expanding.std
+   Expanding.min
+   Expanding.max
+   Expanding.first
+   Expanding.last
+   Expanding.corr
+   Expanding.cov
+   Expanding.skew
+   Expanding.kurt
+   Expanding.apply
+   Expanding.pipe
+   Expanding.aggregate
+   Expanding.quantile
+   Expanding.sem
+   Expanding.rank
+
+.. _api.functions_ewm:
+
+Exponentially-weighted window functions
+---------------------------------------
+.. currentmodule:: pandas.core.window.ewm
+
+.. autosummary::
+   :toctree: api/
+
+   ExponentialMovingWindow.mean
+   ExponentialMovingWindow.sum
+   ExponentialMovingWindow.std
+   ExponentialMovingWindow.var
+   ExponentialMovingWindow.corr
+   ExponentialMovingWindow.cov
+
+.. _api.indexers_window:
+
+Window indexer
+--------------
+.. currentmodule:: pandas
+
+Base class for defining custom window boundaries.
+
+.. autosummary::
+   :toctree: api/
+
+   api.indexers.BaseIndexer
+   api.indexers.FixedForwardWindowIndexer
+   api.indexers.VariableOffsetWindowIndexer
diff --git a/pandas/doc/source/user_guide/10min.rst b/pandas/doc/source/user_guide/10min.rst
new file mode 100644
index 0000000000000..72bb93d21a99f
--- /dev/null
+++ b/pandas/doc/source/user_guide/10min.rst
@@ -0,0 +1,814 @@
+.. _10min:
+
+{{ header }}
+
+********************
+10 minutes to pandas
+********************
+
+This is a short introduction to pandas, geared mainly for new users.
+You can see more complex recipes in the :ref:`Cookbook<cookbook>`.
+
+Customarily, we import as follows:
+
+.. ipython:: python
+
+   import numpy as np
+   import pandas as pd
+
+Basic data structures in pandas
+-------------------------------
+
+pandas provides two types of classes for handling data:
+
+1. :class:`Series`: a one-dimensional labeled array holding data of any type
+    such as integers, strings, Python objects etc.
+2. :class:`DataFrame`: a two-dimensional data structure that holds data like
+   a two-dimension array or a table with rows and columns.
+
+Object creation
+---------------
+
+See the :ref:`Intro to data structures section <dsintro>`.
+
+Creating a :class:`Series` by passing a list of values, letting pandas create
+a default :class:`RangeIndex`.
+
+.. ipython:: python
+
+   s = pd.Series([1, 3, 5, np.nan, 6, 8])
+   s
+
+Creating a :class:`DataFrame` by passing a NumPy array with a datetime index using :func:`date_range`
+and labeled columns:
+
+.. ipython:: python
+
+   dates = pd.date_range("20130101", periods=6)
+   dates
+   df = pd.DataFrame(np.random.randn(6, 4), index=dates, columns=list("ABCD"))
+   df
+
+Creating a :class:`DataFrame` by passing a dictionary of objects where the keys are the column
+labels and the values are the column values.
+
+.. ipython:: python
+
+   df2 = pd.DataFrame(
+       {
+           "A": 1.0,
+           "B": pd.Timestamp("20130102"),
+           "C": pd.Series(1, index=list(range(4)), dtype="float32"),
+           "D": np.array([3] * 4, dtype="int32"),
+           "E": pd.Categorical(["test", "train", "test", "train"]),
+           "F": "foo",
+       }
+   )
+   df2
+
+The columns of the resulting :class:`DataFrame` have different
+:ref:`dtypes <basics.dtypes>`:
+
+.. ipython:: python
+
+   df2.dtypes
+
+If you're using IPython, tab completion for column names (as well as public
+attributes) is automatically enabled. Here's a subset of the attributes that
+will be completed:
+
+.. ipython::
+
+   @verbatim
+   In [1]: df2.<TAB>  # noqa: E225, E999
+   df2.A                  df2.bool
+   df2.abs                df2.boxplot
+   df2.add                df2.C
+   df2.add_prefix         df2.clip
+   df2.add_suffix         df2.columns
+   df2.align              df2.copy
+   df2.all                df2.count
+   df2.any                df2.combine
+   df2.append             df2.D
+   df2.apply              df2.describe
+   df2.B                  df2.duplicated
+   df2.diff
+
+As you can see, the columns ``A``, ``B``, ``C``, and ``D`` are automatically
+tab completed. ``E`` and ``F`` are there as well; the rest of the attributes have been
+truncated for brevity.
+
+Viewing data
+------------
+
+See the :ref:`Essential basic functionality section <basics>`.
+
+Use :meth:`DataFrame.head` and :meth:`DataFrame.tail` to view the top and bottom rows of the frame
+respectively:
+
+.. ipython:: python
+
+   df.head()
+   df.tail(3)
+
+Display the :attr:`DataFrame.index` or :attr:`DataFrame.columns`:
+
+.. ipython:: python
+
+   df.index
+   df.columns
+
+Return a NumPy representation of the underlying data with :meth:`DataFrame.to_numpy`
+without the index or column labels:
+
+.. ipython:: python
+
+   df.to_numpy()
+
+.. note::
+
+   **NumPy arrays have one dtype for the entire array while pandas DataFrames
+   have one dtype per column**. When you call :meth:`DataFrame.to_numpy`, pandas will
+   find the NumPy dtype that can hold *all* of the dtypes in the DataFrame.
+   If the common data type is ``object``, :meth:`DataFrame.to_numpy` will require
+   copying data.
+
+   .. ipython:: python
+
+      df2.dtypes
+      df2.to_numpy()
+
+:func:`~DataFrame.describe` shows a quick statistic summary of your data:
+
+.. ipython:: python
+
+   df.describe()
+
+Transposing your data:
+
+.. ipython:: python
+
+   df.T
+
+:meth:`DataFrame.sort_index` sorts by an axis:
+
+.. ipython:: python
+
+   df.sort_index(axis=1, ascending=False)
+
+:meth:`DataFrame.sort_values` sorts by values:
+
+.. ipython:: python
+
+   df.sort_values(by="B")
+
+Selection
+---------
+
+.. note::
+
+   While standard Python / NumPy expressions for selecting and setting are
+   intuitive and come in handy for interactive work, for production code, we
+   recommend the optimized pandas data access methods, :meth:`DataFrame.at`, :meth:`DataFrame.iat`,
+   :meth:`DataFrame.loc` and :meth:`DataFrame.iloc`.
+
+See the indexing documentation :ref:`Indexing and Selecting Data <indexing>` and :ref:`MultiIndex / Advanced Indexing <advanced>`.
+
+Getitem (``[]``)
+~~~~~~~~~~~~~~~~
+
+For a :class:`DataFrame`, passing a single label selects a column and
+yields a :class:`Series` equivalent to ``df.A``:
+
+.. ipython:: python
+
+   df["A"]
+
+For a :class:`DataFrame`, passing a slice ``:`` selects matching rows:
+
+.. ipython:: python
+
+   df[0:3]
+   df["20130102":"20130104"]
+
+Selection by label
+~~~~~~~~~~~~~~~~~~
+
+See more in :ref:`Selection by Label <indexing.label>` using :meth:`DataFrame.loc` or :meth:`DataFrame.at`.
+
+Selecting a row matching a label:
+
+.. ipython:: python
+
+   df.loc[dates[0]]
+
+Selecting all rows (``:``) with a select column labels:
+
+.. ipython:: python
+
+   df.loc[:, ["A", "B"]]
+
+For label slicing, both endpoints are *included*:
+
+.. ipython:: python
+
+   df.loc["20130102":"20130104", ["A", "B"]]
+
+Selecting a single row and column label returns a scalar:
+
+.. ipython:: python
+
+   df.loc[dates[0], "A"]
+
+For getting fast access to a scalar (equivalent to the prior method):
+
+.. ipython:: python
+
+   df.at[dates[0], "A"]
+
+Selection by position
+~~~~~~~~~~~~~~~~~~~~~
+
+See more in :ref:`Selection by Position <indexing.integer>` using :meth:`DataFrame.iloc` or :meth:`DataFrame.iat`.
+
+Select via the position of the passed integers:
+
+.. ipython:: python
+
+   df.iloc[3]
+
+Integer slices acts similar to NumPy/Python:
+
+.. ipython:: python
+
+   df.iloc[3:5, 0:2]
+
+Lists of integer position locations:
+
+.. ipython:: python
+
+   df.iloc[[1, 2, 4], [0, 2]]
+
+For slicing rows explicitly:
+
+.. ipython:: python
+
+   df.iloc[1:3, :]
+
+For slicing columns explicitly:
+
+.. ipython:: python
+
+   df.iloc[:, 1:3]
+
+For getting a value explicitly:
+
+.. ipython:: python
+
+   df.iloc[1, 1]
+
+For getting fast access to a scalar (equivalent to the prior method):
+
+.. ipython:: python
+
+   df.iat[1, 1]
+
+Boolean indexing
+~~~~~~~~~~~~~~~~
+
+Select rows where ``df.A`` is greater than ``0``.
+
+.. ipython:: python
+
+   df[df["A"] > 0]
+
+Selecting values from a :class:`DataFrame` where a boolean condition is met:
+
+.. ipython:: python
+
+   df[df > 0]
+
+Using :func:`~Series.isin` method for filtering:
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2["E"] = ["one", "one", "two", "three", "four", "three"]
+   df2
+   df2[df2["E"].isin(["two", "four"])]
+
+Setting
+~~~~~~~
+
+Setting a new column automatically aligns the data by the indexes:
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, 3, 4, 5, 6], index=pd.date_range("20130102", periods=6))
+   s1
+   df["F"] = s1
+
+Setting values by label:
+
+.. ipython:: python
+
+   df.at[dates[0], "A"] = 0
+
+Setting values by position:
+
+.. ipython:: python
+
+   df.iat[0, 1] = 0
+
+Setting by assigning with a NumPy array:
+
+.. ipython:: python
+   :okwarning:
+
+   df.loc[:, "D"] = np.array([5] * len(df))
+
+The result of the prior setting operations:
+
+.. ipython:: python
+
+   df
+
+A ``where`` operation with setting:
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2[df2 > 0] = -df2
+   df2
+
+
+Missing data
+------------
+
+For NumPy data types, ``np.nan`` represents missing data. It is by
+default not included in computations. See the :ref:`Missing Data section
+<missing_data>`.
+
+Reindexing allows you to change/add/delete the index on a specified axis. This
+returns a copy of the data:
+
+.. ipython:: python
+
+   df1 = df.reindex(index=dates[0:4], columns=list(df.columns) + ["E"])
+   df1.loc[dates[0] : dates[1], "E"] = 1
+   df1
+
+:meth:`DataFrame.dropna` drops any rows that have missing data:
+
+.. ipython:: python
+
+   df1.dropna(how="any")
+
+:meth:`DataFrame.fillna` fills missing data:
+
+.. ipython:: python
+
+   df1.fillna(value=5)
+
+:func:`isna` gets the boolean mask where values are ``nan``:
+
+.. ipython:: python
+
+   pd.isna(df1)
+
+
+Operations
+----------
+
+See the :ref:`Basic section on Binary Ops <basics.binop>`.
+
+Stats
+~~~~~
+
+Operations in general *exclude* missing data.
+
+Calculate the mean value for each column:
+
+.. ipython:: python
+
+   df.mean()
+
+Calculate the mean value for each row:
+
+.. ipython:: python
+
+   df.mean(axis=1)
+
+Operating with another :class:`Series` or :class:`DataFrame` with a different index or column
+will align the result with the union of the index or column labels. In addition, pandas
+automatically broadcasts along the specified dimension and will fill unaligned labels with ``np.nan``.
+
+.. ipython:: python
+
+   s = pd.Series([1, 3, 5, np.nan, 6, 8], index=dates).shift(2)
+   s
+   df.sub(s, axis="index")
+
+
+User defined functions
+~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`DataFrame.agg` and :meth:`DataFrame.transform` applies a user defined function
+that reduces or broadcasts its result respectively.
+
+.. ipython:: python
+
+   df.agg(lambda x: np.mean(x) * 5.6)
+   df.transform(lambda x: x * 101.2)
+
+Value Counts
+~~~~~~~~~~~~~
+
+See more at :ref:`Histogramming and Discretization <basics.discretization>`.
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randint(0, 7, size=10))
+   s
+   s.value_counts()
+
+String Methods
+~~~~~~~~~~~~~~
+
+:class:`Series` is equipped with a set of string processing methods in the ``str``
+attribute that make it easy to operate on each element of the array, as in the
+code snippet below. See more at :ref:`Vectorized String Methods
+<text.string_methods>`.
+
+.. ipython:: python
+
+   s = pd.Series(["A", "B", "C", "Aaba", "Baca", np.nan, "CABA", "dog", "cat"])
+   s.str.lower()
+
+Merge
+-----
+
+Concat
+~~~~~~
+
+pandas provides various facilities for easily combining together :class:`Series` and
+:class:`DataFrame` objects with various kinds of set logic for the indexes
+and relational algebra functionality in the case of join / merge-type
+operations.
+
+See the :ref:`Merging section <merging>`.
+
+Concatenating pandas objects together row-wise with :func:`concat`:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 4))
+   df
+
+   # break it into pieces
+   pieces = [df[:3], df[3:7], df[7:]]
+
+   pd.concat(pieces)
+
+.. note::
+
+   Adding a column to a :class:`DataFrame` is relatively fast. However, adding
+   a row requires a copy, and may be expensive. We recommend passing a
+   pre-built list of records to the :class:`DataFrame` constructor instead
+   of building a :class:`DataFrame` by iteratively appending records to it.
+
+Join
+~~~~
+
+:func:`merge` enables SQL style join types along specific columns. See the :ref:`Database style joining <merging.join>` section.
+
+.. ipython:: python
+
+   left = pd.DataFrame({"key": ["foo", "foo"], "lval": [1, 2]})
+   right = pd.DataFrame({"key": ["foo", "foo"], "rval": [4, 5]})
+   left
+   right
+   pd.merge(left, right, on="key")
+
+:func:`merge` on unique keys:
+
+.. ipython:: python
+
+   left = pd.DataFrame({"key": ["foo", "bar"], "lval": [1, 2]})
+   right = pd.DataFrame({"key": ["foo", "bar"], "rval": [4, 5]})
+   left
+   right
+   pd.merge(left, right, on="key")
+
+Grouping
+--------
+
+By "group by" we are referring to a process involving one or more of the
+following steps:
+
+* **Splitting** the data into groups based on some criteria
+* **Applying** a function to each group independently
+* **Combining** the results into a data structure
+
+See the :ref:`Grouping section <groupby>`.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+           "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+           "C": np.random.randn(8),
+           "D": np.random.randn(8),
+       }
+   )
+   df
+
+Grouping by a column label, selecting column labels, and then applying the
+:meth:`.DataFrameGroupBy.sum` function to the resulting
+groups:
+
+.. ipython:: python
+
+   df.groupby("A")[["C", "D"]].sum()
+
+Grouping by multiple columns label forms :class:`MultiIndex`.
+
+.. ipython:: python
+
+   df.groupby(["A", "B"]).sum()
+
+Reshaping
+---------
+
+See the sections on :ref:`Hierarchical Indexing <advanced.hierarchical>` and
+:ref:`Reshaping <reshaping.stacking>`.
+
+Stack
+~~~~~
+
+.. ipython:: python
+
+   arrays = [
+      ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+      ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+   index = pd.MultiIndex.from_arrays(arrays, names=["first", "second"])
+   df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=["A", "B"])
+   df2 = df[:4]
+   df2
+
+The :meth:`~DataFrame.stack` method "compresses" a level in the DataFrame's
+columns:
+
+.. ipython:: python
+
+   stacked = df2.stack()
+   stacked
+
+With a "stacked" DataFrame or Series (having a :class:`MultiIndex` as the
+``index``), the inverse operation of :meth:`~DataFrame.stack` is
+:meth:`~DataFrame.unstack`, which by default unstacks the **last level**:
+
+.. ipython:: python
+
+   stacked.unstack()
+   stacked.unstack(1)
+   stacked.unstack(0)
+
+Pivot tables
+~~~~~~~~~~~~
+See the section on :ref:`Pivot Tables <reshaping.pivot>`.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": ["one", "one", "two", "three"] * 3,
+           "B": ["A", "B", "C"] * 4,
+           "C": ["foo", "foo", "foo", "bar", "bar", "bar"] * 2,
+           "D": np.random.randn(12),
+           "E": np.random.randn(12),
+       }
+   )
+   df
+
+:func:`pivot_table` pivots a :class:`DataFrame` specifying the ``values``, ``index`` and ``columns``
+
+.. ipython:: python
+
+   pd.pivot_table(df, values="D", index=["A", "B"], columns=["C"])
+
+
+Time series
+-----------
+
+pandas has simple, powerful, and efficient functionality for performing
+resampling operations during frequency conversion (e.g., converting secondly
+data into 5-minutely data). This is extremely common in, but not limited to,
+financial applications. See the :ref:`Time Series section <timeseries>`.
+
+.. ipython:: python
+
+   rng = pd.date_range("1/1/2012", periods=100, freq="s")
+   ts = pd.Series(np.random.randint(0, 500, len(rng)), index=rng)
+   ts.resample("5Min").sum()
+
+:meth:`Series.tz_localize` localizes a time series to a time zone:
+
+.. ipython:: python
+
+   rng = pd.date_range("3/6/2012 00:00", periods=5, freq="D")
+   ts = pd.Series(np.random.randn(len(rng)), rng)
+   ts
+   ts_utc = ts.tz_localize("UTC")
+   ts_utc
+
+:meth:`Series.tz_convert` converts a timezones aware time series to another time zone:
+
+.. ipython:: python
+
+   ts_utc.tz_convert("US/Eastern")
+
+Adding a non-fixed duration (:class:`~pandas.tseries.offsets.BusinessDay`) to a time series:
+
+.. ipython:: python
+
+   rng
+   rng + pd.offsets.BusinessDay(5)
+
+Categoricals
+------------
+
+pandas can include categorical data in a :class:`DataFrame`. For full docs, see the
+:ref:`categorical introduction <categorical>` and the :ref:`API documentation <api.arrays.categorical>`.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {"id": [1, 2, 3, 4, 5, 6], "raw_grade": ["a", "b", "b", "a", "a", "e"]}
+    )
+
+Converting the raw grades to a categorical data type:
+
+.. ipython:: python
+
+    df["grade"] = df["raw_grade"].astype("category")
+    df["grade"]
+
+Rename the categories to more meaningful names:
+
+.. ipython:: python
+
+    new_categories = ["very good", "good", "very bad"]
+    df["grade"] = df["grade"].cat.rename_categories(new_categories)
+
+Reorder the categories and simultaneously add the missing categories (methods under :meth:`Series.cat` return a new :class:`Series` by default):
+
+.. ipython:: python
+
+    df["grade"] = df["grade"].cat.set_categories(
+        ["very bad", "bad", "medium", "good", "very good"]
+    )
+    df["grade"]
+
+Sorting is per order in the categories, not lexical order:
+
+.. ipython:: python
+
+    df.sort_values(by="grade")
+
+Grouping by a categorical column with ``observed=False`` also shows empty categories:
+
+.. ipython:: python
+
+    df.groupby("grade", observed=False).size()
+
+
+Plotting
+--------
+
+See the :ref:`Plotting <visualization>` docs.
+
+We use the standard convention for referencing the matplotlib API:
+
+.. ipython:: python
+
+   import matplotlib.pyplot as plt
+
+   plt.close("all")
+
+The ``plt.close`` method is used to `close <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.close.html>`__ a figure window:
+
+.. ipython:: python
+
+   ts = pd.Series(np.random.randn(1000), index=pd.date_range("1/1/2000", periods=1000))
+   ts = ts.cumsum()
+
+   @savefig series_plot_basic.png
+   ts.plot();
+
+.. note::
+
+   When using Jupyter, the plot will appear using :meth:`~Series.plot`.  Otherwise use
+   `matplotlib.pyplot.show <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.show.html>`__ to show it or
+   `matplotlib.pyplot.savefig <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.savefig.html>`__ to write it to a file.
+
+:meth:`~DataFrame.plot` plots all columns:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.random.randn(1000, 4), index=ts.index, columns=["A", "B", "C", "D"]
+   )
+
+   df = df.cumsum()
+
+   plt.figure();
+   df.plot();
+   @savefig frame_plot_basic.png
+   plt.legend(loc='best');
+
+Importing and exporting data
+----------------------------
+
+See the :ref:`IO Tools <io>` section.
+
+CSV
+~~~
+
+:ref:`Writing to a csv file: <io.store_in_csv>` using :meth:`DataFrame.to_csv`
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randint(0, 5, (10, 5)))
+   df.to_csv("foo.csv")
+
+:ref:`Reading from a csv file: <io.read_csv_table>` using :func:`read_csv`
+
+.. ipython:: python
+
+   pd.read_csv("foo.csv")
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("foo.csv")
+
+Parquet
+~~~~~~~
+
+Writing to a Parquet file:
+
+.. ipython:: python
+
+   df.to_parquet("foo.parquet")
+
+Reading from a Parquet file Store using :func:`read_parquet`:
+
+.. ipython:: python
+
+   pd.read_parquet("foo.parquet")
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("foo.parquet")
+
+Excel
+~~~~~
+
+Reading and writing to :ref:`Excel <io.excel>`.
+
+Writing to an excel file using :meth:`DataFrame.to_excel`:
+
+.. ipython:: python
+
+   df.to_excel("foo.xlsx", sheet_name="Sheet1")
+
+Reading from an excel file using :func:`read_excel`:
+
+.. ipython:: python
+
+   pd.read_excel("foo.xlsx", "Sheet1", index_col=None, na_values=["NA"])
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("foo.xlsx")
+
+Gotchas
+-------
+
+If you are attempting to perform a boolean operation on a :class:`Series` or :class:`DataFrame`
+you might see an exception like:
+
+.. ipython:: python
+   :okexcept:
+
+    if pd.Series([False, True, False]):
+        print("I was true")
+
+See :ref:`Comparisons<basics.compare>` and :ref:`Gotchas<gotchas>` for an explanation and what to do.
diff --git a/pandas/doc/source/user_guide/advanced.rst b/pandas/doc/source/user_guide/advanced.rst
new file mode 100644
index 0000000000000..f7ab466e92d93
--- /dev/null
+++ b/pandas/doc/source/user_guide/advanced.rst
@@ -0,0 +1,1140 @@
+.. _advanced:
+
+{{ header }}
+
+******************************
+MultiIndex / advanced indexing
+******************************
+
+This section covers :ref:`indexing with a MultiIndex <advanced.hierarchical>`
+and :ref:`other advanced indexing features <advanced.index_types>`.
+
+See the :ref:`Indexing and Selecting Data <indexing>` for general indexing documentation.
+
+See the :ref:`cookbook<cookbook.selection>` for some advanced strategies.
+
+.. _advanced.hierarchical:
+
+Hierarchical indexing (MultiIndex)
+----------------------------------
+
+Hierarchical / Multi-level indexing is very exciting as it opens the door to some
+quite sophisticated data analysis and manipulation, especially for working with
+higher dimensional data. In essence, it enables you to store and manipulate
+data with an arbitrary number of dimensions in lower dimensional data
+structures like ``Series`` (1d) and ``DataFrame`` (2d).
+
+In this section, we will show what exactly we mean by "hierarchical" indexing
+and how it integrates with all of the pandas indexing functionality
+described above and in prior sections. Later, when discussing :ref:`group by
+<groupby>` and :ref:`pivoting and reshaping data <reshaping>`, we'll show
+non-trivial applications to illustrate how it aids in structuring data for
+analysis.
+
+See the :ref:`cookbook<cookbook.multi_index>` for some advanced strategies.
+
+Creating a MultiIndex (hierarchical index) object
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :class:`MultiIndex` object is the hierarchical analogue of the standard
+:class:`Index` object which typically stores the axis labels in pandas objects. You
+can think of ``MultiIndex`` as an array of tuples where each tuple is unique. A
+``MultiIndex`` can be created from a list of arrays (using
+:meth:`MultiIndex.from_arrays`), an array of tuples (using
+:meth:`MultiIndex.from_tuples`), a crossed set of iterables (using
+:meth:`MultiIndex.from_product`), or a :class:`DataFrame` (using
+:meth:`MultiIndex.from_frame`).  The ``Index`` constructor will attempt to return
+a ``MultiIndex`` when it is passed a list of tuples.  The following examples
+demonstrate different ways to initialize MultiIndexes.
+
+
+.. ipython:: python
+
+   arrays = [
+       ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+       ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+   tuples = list(zip(*arrays))
+   tuples
+
+   index = pd.MultiIndex.from_tuples(tuples, names=["first", "second"])
+   index
+
+   s = pd.Series(np.random.randn(8), index=index)
+   s
+
+When you want every pairing of the elements in two iterables, it can be easier
+to use the :meth:`MultiIndex.from_product` method:
+
+.. ipython:: python
+
+   iterables = [["bar", "baz", "foo", "qux"], ["one", "two"]]
+   pd.MultiIndex.from_product(iterables, names=["first", "second"])
+
+You can also construct a ``MultiIndex`` from a ``DataFrame`` directly, using
+the method :meth:`MultiIndex.from_frame`. This is a complementary method to
+:meth:`MultiIndex.to_frame`.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       [["bar", "one"], ["bar", "two"], ["foo", "one"], ["foo", "two"]],
+       columns=["first", "second"],
+   )
+   pd.MultiIndex.from_frame(df)
+
+As a convenience, you can pass a list of arrays directly into ``Series`` or
+``DataFrame`` to construct a ``MultiIndex`` automatically:
+
+.. ipython:: python
+
+   arrays = [
+       np.array(["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"]),
+       np.array(["one", "two", "one", "two", "one", "two", "one", "two"]),
+   ]
+   s = pd.Series(np.random.randn(8), index=arrays)
+   s
+   df = pd.DataFrame(np.random.randn(8, 4), index=arrays)
+   df
+
+All of the ``MultiIndex`` constructors accept a ``names`` argument which stores
+string names for the levels themselves. If no names are provided, ``None`` will
+be assigned:
+
+.. ipython:: python
+
+   df.index.names
+
+This index can back any axis of a pandas object, and the number of **levels**
+of the index is up to you:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(3, 8), index=["A", "B", "C"], columns=index)
+   df
+   pd.DataFrame(np.random.randn(6, 6), index=index[:6], columns=index[:6])
+
+We've "sparsified" the higher levels of the indexes to make the console output a
+bit easier on the eyes. Note that how the index is displayed can be controlled using the
+``multi_sparse`` option in ``pandas.set_options()``:
+
+.. ipython:: python
+
+   with pd.option_context("display.multi_sparse", False):
+       df
+
+It's worth keeping in mind that there's nothing preventing you from using
+tuples as atomic labels on an axis:
+
+.. ipython:: python
+
+   pd.Series(np.random.randn(8), index=tuples)
+
+The reason that the ``MultiIndex`` matters is that it can allow you to do
+grouping, selection, and reshaping operations as we will describe below and in
+subsequent areas of the documentation. As you will see in later sections, you
+can find yourself working with hierarchically-indexed data without creating a
+``MultiIndex`` explicitly yourself. However, when loading data from a file, you
+may wish to generate your own ``MultiIndex`` when preparing the data set.
+
+.. _advanced.get_level_values:
+
+Reconstructing the level labels
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The method :meth:`~MultiIndex.get_level_values` will return a vector of the labels for each
+location at a particular level:
+
+.. ipython:: python
+
+   index.get_level_values(0)
+   index.get_level_values("second")
+
+Basic indexing on axis with MultiIndex
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+One of the important features of hierarchical indexing is that you can select
+data by a "partial" label identifying a subgroup in the data. **Partial**
+selection "drops" levels of the hierarchical index in the result in a
+completely analogous way to selecting a column in a regular DataFrame:
+
+.. ipython:: python
+
+   df["bar"]
+   df["bar", "one"]
+   df["bar"]["one"]
+   s["qux"]
+
+See :ref:`Cross-section with hierarchical index <advanced.xs>` for how to select
+on a deeper level.
+
+.. _advanced.shown_levels:
+
+Defined levels
+~~~~~~~~~~~~~~
+
+The :class:`MultiIndex` keeps all the defined levels of an index, even
+if they are not actually used. When slicing an index, you may notice this.
+For example:
+
+.. ipython:: python
+
+   df.columns.levels  # original MultiIndex
+
+   df[["foo","qux"]].columns.levels  # sliced
+
+This is done to avoid a recomputation of the levels in order to make slicing
+highly performant. If you want to see only the used levels, you can use the
+:meth:`~MultiIndex.get_level_values` method.
+
+.. ipython:: python
+
+   df[["foo", "qux"]].columns.to_numpy()
+
+   # for a specific level
+   df[["foo", "qux"]].columns.get_level_values(0)
+
+To reconstruct the ``MultiIndex`` with only the used levels, the
+:meth:`~MultiIndex.remove_unused_levels` method may be used.
+
+.. ipython:: python
+
+   new_mi = df[["foo", "qux"]].columns.remove_unused_levels()
+   new_mi.levels
+
+Data alignment and using ``reindex``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Operations between differently-indexed objects having ``MultiIndex`` on the
+axes will work as you expect; data alignment will work the same as an Index of
+tuples:
+
+.. ipython:: python
+
+   s + s[:-2]
+   s + s[::2]
+
+The :meth:`~DataFrame.reindex` method of ``Series``/``DataFrames`` can be
+called with another ``MultiIndex``, or even a list or array of tuples:
+
+.. ipython:: python
+
+   s.reindex(index[:3])
+   s.reindex([("foo", "two"), ("bar", "one"), ("qux", "one"), ("baz", "one")])
+
+.. _advanced.advanced_hierarchical:
+
+Advanced indexing with hierarchical index
+-----------------------------------------
+
+Syntactically integrating ``MultiIndex`` in advanced indexing with ``.loc`` is a
+bit challenging, but we've made every effort to do so. In general, MultiIndex
+keys take the form of tuples. For example, the following works as you would expect:
+
+.. ipython:: python
+
+   df = df.T
+   df
+   df.loc[("bar", "two")]
+
+Note that ``df.loc['bar', 'two']`` would also work in this example, but this shorthand
+notation can lead to ambiguity in general.
+
+If you also want to index a specific column with ``.loc``, you must use a tuple
+like this:
+
+.. ipython:: python
+
+   df.loc[("bar", "two"), "A"]
+
+You don't have to specify all levels of the ``MultiIndex`` by passing only the
+first elements of the tuple. For example, you can use "partial" indexing to
+get all elements with ``bar`` in the first level as follows:
+
+.. ipython:: python
+
+   df.loc["bar"]
+
+This is a shortcut for the slightly more verbose notation ``df.loc[('bar',),]`` (equivalent
+to ``df.loc['bar',]`` in this example).
+
+"Partial" slicing also works quite nicely.
+
+.. ipython:: python
+
+   df.loc["baz":"foo"]
+
+You can slice with a 'range' of values, by providing a slice of tuples.
+
+.. ipython:: python
+
+   df.loc[("baz", "two"):("qux", "one")]
+   df.loc[("baz", "two"):"foo"]
+
+Passing a list of labels or tuples works similar to reindexing:
+
+.. ipython:: python
+
+   df.loc[[("bar", "two"), ("qux", "one")]]
+
+.. note::
+
+   It is important to note that tuples and lists are not treated identically
+   in pandas when it comes to indexing. Whereas a tuple is interpreted as one
+   multi-level key, a list is used to specify several keys. Or in other words,
+   tuples go horizontally (traversing levels), lists go vertically (scanning levels).
+
+Importantly, a list of tuples indexes several complete ``MultiIndex`` keys,
+whereas a tuple of lists refer to several values within a level:
+
+.. ipython:: python
+
+   s = pd.Series(
+       [1, 2, 3, 4, 5, 6],
+       index=pd.MultiIndex.from_product([["A", "B"], ["c", "d", "e"]]),
+   )
+   s.loc[[("A", "c"), ("B", "d")]]  # list of tuples
+   s.loc[(["A", "B"], ["c", "d"])]  # tuple of lists
+
+
+.. _advanced.mi_slicers:
+
+Using slicers
+~~~~~~~~~~~~~
+
+You can slice a ``MultiIndex`` by providing multiple indexers.
+
+You can provide any of the selectors as if you are indexing by label, see :ref:`Selection by Label <indexing.label>`,
+including slices, lists of labels, labels, and boolean indexers.
+
+You can use ``slice(None)`` to select all the contents of *that* level. You do not need to specify all the
+*deeper* levels, they will be implied as ``slice(None)``.
+
+As usual, **both sides** of the slicers are included as this is label indexing.
+
+.. warning::
+
+   You should specify all axes in the ``.loc`` specifier, meaning the indexer for the **index** and
+   for the **columns**. There are some ambiguous cases where the passed indexer could be misinterpreted
+   as indexing *both* axes, rather than into say the ``MultiIndex`` for the rows.
+
+   You should do this:
+
+   .. code-block:: python
+
+      df.loc[(slice("A1", "A3"), ...), :]  # noqa: E999
+
+   You should **not** do this:
+ 
+   .. code-block:: python
+
+      df.loc[(slice("A1", "A3"), ...)]  # noqa: E999
+
+.. ipython:: python
+
+   def mklbl(prefix, n):
+       return ["%s%s" % (prefix, i) for i in range(n)]
+
+
+   miindex = pd.MultiIndex.from_product(
+       [mklbl("A", 4), mklbl("B", 2), mklbl("C", 4), mklbl("D", 2)]
+   )
+   micolumns = pd.MultiIndex.from_tuples(
+       [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")], names=["lvl0", "lvl1"]
+   )
+   dfmi = (
+       pd.DataFrame(
+           np.arange(len(miindex) * len(micolumns)).reshape(
+               (len(miindex), len(micolumns))
+           ),
+           index=miindex,
+           columns=micolumns,
+       )
+       .sort_index()
+       .sort_index(axis=1)
+   )
+   dfmi
+
+Basic MultiIndex slicing using slices, lists, and labels.
+
+.. ipython:: python
+
+   dfmi.loc[(slice("A1", "A3"), slice(None), ["C1", "C3"]), :]
+
+
+You can use :class:`pandas.IndexSlice` to facilitate a more natural syntax
+using ``:``, rather than using ``slice(None)``.
+
+.. ipython:: python
+
+   idx = pd.IndexSlice
+   dfmi.loc[idx[:, :, ["C1", "C3"]], idx[:, "foo"]]
+
+It is possible to perform quite complicated selections using this method on multiple
+axes at the same time.
+
+.. ipython:: python
+
+   dfmi.loc["A1", (slice(None), "foo")]
+   dfmi.loc[idx[:, :, ["C1", "C3"]], idx[:, "foo"]]
+
+Using a boolean indexer you can provide selection related to the *values*.
+
+.. ipython:: python
+
+   mask = dfmi[("a", "foo")] > 200
+   dfmi.loc[idx[mask, :, ["C1", "C3"]], idx[:, "foo"]]
+
+You can also specify the ``axis`` argument to ``.loc`` to interpret the passed
+slicers on a single axis.
+
+.. ipython:: python
+
+   dfmi.loc(axis=0)[:, :, ["C1", "C3"]]
+
+Furthermore, you can *set* the values using the following methods.
+
+.. ipython:: python
+   :okwarning:
+
+   df2 = dfmi.copy()
+   df2.loc(axis=0)[:, :, ["C1", "C3"]] = -10
+   df2
+
+You can use a right-hand-side of an alignable object as well.
+
+.. ipython:: python
+
+   df2 = dfmi.copy()
+   df2.loc[idx[:, :, ["C1", "C3"]], :] = df2 * 1000
+   df2
+
+.. _advanced.xs:
+
+Cross-section
+~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.xs` method of ``DataFrame`` additionally takes a level argument to make
+selecting data at a particular level of a ``MultiIndex`` easier.
+
+.. ipython:: python
+
+   df
+   df.xs("one", level="second")
+
+.. ipython:: python
+
+   # using the slicers
+   df.loc[(slice(None), "one"), :]
+
+You can also select on the columns with ``xs``, by
+providing the axis argument.
+
+.. ipython:: python
+
+   df = df.T
+   df.xs("one", level="second", axis=1)
+
+.. ipython:: python
+
+   # using the slicers
+   df.loc[:, (slice(None), "one")]
+
+``xs`` also allows selection with multiple keys.
+
+.. ipython:: python
+
+   df.xs(("one", "bar"), level=("second", "first"), axis=1)
+
+.. ipython:: python
+
+   # using the slicers
+   df.loc[:, ("bar", "one")]
+
+You can pass ``drop_level=False`` to ``xs`` to retain
+the level that was selected.
+
+.. ipython:: python
+
+   df.xs("one", level="second", axis=1, drop_level=False)
+
+Compare the above with the result using ``drop_level=True`` (the default value).
+
+.. ipython:: python
+
+   df.xs("one", level="second", axis=1, drop_level=True)
+
+.. _advanced.advanced_reindex:
+
+Advanced reindexing and alignment
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Using the parameter ``level`` in the :meth:`~DataFrame.reindex` and
+:meth:`~DataFrame.align` methods of pandas objects is useful to broadcast
+values across a level. For instance:
+
+.. ipython:: python
+
+   midx = pd.MultiIndex(
+       levels=[["zero", "one"], ["x", "y"]], codes=[[1, 1, 0, 0], [1, 0, 1, 0]]
+   )
+   df = pd.DataFrame(np.random.randn(4, 2), index=midx)
+   df
+   df2 = df.groupby(level=0).mean()
+   df2
+   df2.reindex(df.index, level=0)
+
+   # aligning
+   df_aligned, df2_aligned = df.align(df2, level=0)
+   df_aligned
+   df2_aligned
+
+
+Swapping levels with ``swaplevel``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~MultiIndex.swaplevel` method can switch the order of two levels:
+
+.. ipython:: python
+
+   df[:5]
+   df[:5].swaplevel(0, 1, axis=0)
+
+.. _advanced.reorderlevels:
+
+Reordering levels with ``reorder_levels``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~MultiIndex.reorder_levels` method generalizes the ``swaplevel``
+method, allowing you to permute the hierarchical index levels in one step:
+
+.. ipython:: python
+
+   df[:5].reorder_levels([1, 0], axis=0)
+
+.. _advanced.index_names:
+
+Renaming names of an ``Index`` or ``MultiIndex``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.rename` method is used to rename the labels of a
+``MultiIndex``, and is typically used to rename the columns of a ``DataFrame``.
+The ``columns`` argument of ``rename`` allows a dictionary to be specified
+that includes only the columns you wish to rename.
+
+.. ipython:: python
+
+   df.rename(columns={0: "col0", 1: "col1"})
+
+This method can also be used to rename specific labels of the main index
+of the ``DataFrame``.
+
+.. ipython:: python
+
+   df.rename(index={"one": "two", "y": "z"})
+
+The :meth:`~DataFrame.rename_axis` method is used to rename the name of a
+``Index`` or ``MultiIndex``. In particular, the names of the levels of a
+``MultiIndex`` can be specified, which is useful if ``reset_index()`` is later
+used to move the values from the ``MultiIndex`` to a column.
+
+.. ipython:: python
+
+   df.rename_axis(index=["abc", "def"])
+
+Note that the columns of a ``DataFrame`` are an index, so that using
+``rename_axis`` with the ``columns`` argument will change the name of that
+index.
+
+.. ipython:: python
+
+   df.rename_axis(columns="Cols").columns
+
+Both ``rename`` and ``rename_axis`` support specifying a dictionary,
+``Series`` or a mapping function to map labels/names to new values.
+
+When working with an ``Index`` object directly, rather than via a ``DataFrame``,
+:meth:`Index.set_names` can be used to change the names.
+
+.. ipython:: python
+
+   mi = pd.MultiIndex.from_product([[1, 2], ["a", "b"]], names=["x", "y"])
+   mi.names
+
+   mi2 = mi.rename("new name", level=0)
+   mi2
+
+
+You cannot set the names of the MultiIndex via a level.
+
+.. ipython:: python
+   :okexcept:
+
+   mi.levels[0].name = "name via level"
+
+Use :meth:`Index.set_names` instead.
+
+Sorting a ``MultiIndex``
+------------------------
+
+For :class:`MultiIndex`-ed objects to be indexed and sliced effectively,
+they need to be sorted. As with any index, you can use :meth:`~DataFrame.sort_index`.
+
+.. ipython:: python
+
+   import random
+
+   random.shuffle(tuples)
+   s = pd.Series(np.random.randn(8), index=pd.MultiIndex.from_tuples(tuples))
+   s
+   s.sort_index()
+   s.sort_index(level=0)
+   s.sort_index(level=1)
+
+.. _advanced.sortlevel_byname:
+
+You may also pass a level name to ``sort_index`` if the ``MultiIndex`` levels
+are named.
+
+.. ipython:: python
+
+   s.index = s.index.set_names(["L1", "L2"])
+   s.sort_index(level="L1")
+   s.sort_index(level="L2")
+
+On higher dimensional objects, you can sort any of the other axes by level if
+they have a ``MultiIndex``:
+
+.. ipython:: python
+
+   df.T.sort_index(level=1, axis=1)
+
+Indexing will work even if the data are not sorted, but will be rather
+inefficient (and show a ``PerformanceWarning``). It will also
+return a copy of the data rather than a view:
+
+.. ipython:: python
+   :okwarning:
+
+   dfm = pd.DataFrame(
+       {"jim": [0, 0, 1, 1], "joe": ["x", "x", "z", "y"], "jolie": np.random.rand(4)}
+   )
+   dfm = dfm.set_index(["jim", "joe"])
+   dfm
+   dfm.loc[(1, 'z')]
+
+.. _advanced.unsorted:
+
+Furthermore, if you try to index something that is not fully lexsorted, this can raise:
+
+.. ipython:: python
+   :okexcept:
+
+   dfm.loc[(0, 'y'):(1, 'z')]
+
+The :meth:`~MultiIndex.is_monotonic_increasing` method on a ``MultiIndex`` shows if the
+index is sorted:
+
+.. ipython:: python
+
+   dfm.index.is_monotonic_increasing
+
+.. ipython:: python
+
+   dfm = dfm.sort_index()
+   dfm
+   dfm.index.is_monotonic_increasing
+
+And now selection works as expected.
+
+.. ipython:: python
+
+   dfm.loc[(0, "y"):(1, "z")]
+
+Take methods
+------------
+
+.. _advanced.take:
+
+Similar to NumPy ndarrays, pandas ``Index``, ``Series``, and ``DataFrame`` also provides
+the :meth:`~DataFrame.take` method that retrieves elements along a given axis at the given
+indices. The given indices must be either a list or an ndarray of integer
+index positions. ``take`` will also accept negative integers as relative positions to the end of the object.
+
+.. ipython:: python
+
+   index = pd.Index(np.random.randint(0, 1000, 10))
+   index
+
+   positions = [0, 9, 3]
+
+   index[positions]
+   index.take(positions)
+
+   ser = pd.Series(np.random.randn(10))
+
+   ser.iloc[positions]
+   ser.take(positions)
+
+For DataFrames, the given indices should be a 1d list or ndarray that specifies
+row or column positions.
+
+.. ipython:: python
+
+   frm = pd.DataFrame(np.random.randn(5, 3))
+
+   frm.take([1, 4, 3])
+
+   frm.take([0, 2], axis=1)
+
+It is important to note that the ``take`` method on pandas objects are not
+intended to work on boolean indices and may return unexpected results.
+
+.. ipython:: python
+
+   arr = np.random.randn(10)
+   arr.take([False, False, True, True])
+   arr[[0, 1]]
+
+   ser = pd.Series(np.random.randn(10))
+   ser.take([False, False, True, True])
+   ser.iloc[[0, 1]]
+
+Finally, as a small note on performance, because the ``take`` method handles
+a narrower range of inputs, it can offer performance that is a good deal
+faster than fancy indexing.
+
+.. ipython:: python
+
+   arr = np.random.randn(10000, 5)
+   indexer = np.arange(10000)
+   random.shuffle(indexer)
+
+   %timeit arr[indexer]
+   %timeit arr.take(indexer, axis=0)
+
+.. ipython:: python
+
+   ser = pd.Series(arr[:, 0])
+   %timeit ser.iloc[indexer]
+   %timeit ser.take(indexer)
+
+.. _advanced.index_types:
+
+Index types
+-----------
+
+We have discussed ``MultiIndex`` in the previous sections pretty extensively.
+Documentation about ``DatetimeIndex`` and ``PeriodIndex`` are shown :ref:`here <timeseries.overview>`,
+and documentation about ``TimedeltaIndex`` is found :ref:`here <timedeltas.index>`.
+
+In the following sub-sections we will highlight some other index types.
+
+.. _advanced.categoricalindex:
+
+CategoricalIndex
+~~~~~~~~~~~~~~~~
+
+:class:`CategoricalIndex` is a type of index that is useful for supporting
+indexing with duplicates. This is a container around a :class:`Categorical`
+and allows efficient indexing and storage of an index with a large number of duplicated elements.
+
+.. ipython:: python
+
+   from pandas.api.types import CategoricalDtype
+
+   df = pd.DataFrame({"A": np.arange(6), "B": list("aabbca")})
+   df["B"] = df["B"].astype(CategoricalDtype(list("cab")))
+   df
+   df.dtypes
+   df["B"].cat.categories
+
+Setting the index will create a ``CategoricalIndex``.
+
+.. ipython:: python
+
+   df2 = df.set_index("B")
+   df2.index
+
+Indexing with ``__getitem__/.iloc/.loc`` works similarly to an ``Index`` with duplicates.
+The indexers **must** be in the category or the operation will raise a ``KeyError``.
+
+.. ipython:: python
+
+   df2.loc["a"]
+
+The ``CategoricalIndex`` is **preserved** after indexing:
+
+.. ipython:: python
+
+   df2.loc["a"].index
+
+Sorting the index will sort by the order of the categories (recall that we
+created the index with ``CategoricalDtype(list('cab'))``, so the sorted
+order is ``cab``).
+
+.. ipython:: python
+
+   df2.sort_index()
+
+Groupby operations on the index will preserve the index nature as well.
+
+.. ipython:: python
+
+   df2.groupby(level=0, observed=True).sum()
+   df2.groupby(level=0, observed=True).sum().index
+
+Reindexing operations will return a resulting index based on the type of the passed
+indexer. Passing a list will return a plain-old ``Index``; indexing with
+a ``Categorical`` will return a ``CategoricalIndex``, indexed according to the categories
+of the **passed** ``Categorical`` dtype. This allows one to arbitrarily index these even with
+values **not** in the categories, similarly to how you can reindex **any** pandas index.
+
+.. ipython:: python
+
+   df3 = pd.DataFrame(
+       {"A": np.arange(3), "B": pd.Series(list("abc")).astype("category")}
+   )
+   df3 = df3.set_index("B")
+   df3
+
+.. ipython:: python
+
+   df3.reindex(["a", "e"])
+   df3.reindex(["a", "e"]).index
+   df3.reindex(pd.Categorical(["a", "e"], categories=list("abe")))
+   df3.reindex(pd.Categorical(["a", "e"], categories=list("abe"))).index
+
+.. warning::
+
+   Reshaping and Comparison operations on a ``CategoricalIndex`` must have the same categories
+   or a ``TypeError`` will be raised.
+
+   .. ipython:: python
+
+      df4 = pd.DataFrame({"A": np.arange(2), "B": list("ba")})
+      df4["B"] = df4["B"].astype(CategoricalDtype(list("ab")))
+      df4 = df4.set_index("B")
+      df4.index
+
+      df5 = pd.DataFrame({"A": np.arange(2), "B": list("bc")})
+      df5["B"] = df5["B"].astype(CategoricalDtype(list("bc")))
+      df5 = df5.set_index("B")
+      df5.index
+
+   .. ipython:: python
+      :okexcept:
+
+      pd.concat([df4, df5])
+
+.. _advanced.rangeindex:
+
+RangeIndex
+~~~~~~~~~~
+
+:class:`RangeIndex` is a sub-class of :class:`Index`  that provides the default index for all :class:`DataFrame` and :class:`Series` objects.
+``RangeIndex`` is an optimized version of ``Index`` that can represent a monotonic ordered set. These are analogous to Python `range types <https://docs.python.org/3/library/stdtypes.html#typesseq-range>`__.
+A ``RangeIndex`` will always have an ``int64`` dtype.
+
+.. ipython:: python
+
+   idx = pd.RangeIndex(5)
+   idx
+
+``RangeIndex`` is the default index for all :class:`DataFrame` and :class:`Series` objects:
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3])
+   ser.index
+   df = pd.DataFrame([[1, 2], [3, 4]])
+   df.index
+   df.columns
+
+A ``RangeIndex`` will behave similarly to a :class:`Index` with an ``int64`` dtype and operations on a ``RangeIndex``,
+whose result cannot be represented by a ``RangeIndex``, but should have an integer dtype, will be converted to an ``Index`` with ``int64``.
+For example:
+
+.. ipython:: python
+
+   idx[[0, 2]]
+
+
+.. _advanced.intervalindex:
+
+IntervalIndex
+~~~~~~~~~~~~~
+
+:class:`IntervalIndex` together with its own dtype, :class:`~pandas.api.types.IntervalDtype`
+as well as the :class:`Interval` scalar type,  allow first-class support in pandas
+for interval notation.
+
+The ``IntervalIndex`` allows some unique indexing and is also used as a
+return type for the categories in :func:`cut` and :func:`qcut`.
+
+Indexing with an ``IntervalIndex``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+An ``IntervalIndex`` can be used in ``Series`` and in ``DataFrame`` as the index.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"A": [1, 2, 3, 4]}, index=pd.IntervalIndex.from_breaks([0, 1, 2, 3, 4])
+   )
+   df
+
+Label based indexing via ``.loc`` along the edges of an interval works as you would expect,
+selecting that particular interval.
+
+.. ipython:: python
+
+   df.loc[2]
+   df.loc[[2, 3]]
+
+If you select a label *contained* within an interval, this will also select the interval.
+
+.. ipython:: python
+
+   df.loc[2.5]
+   df.loc[[2.5, 3.5]]
+
+Selecting using an ``Interval`` will only return exact matches.
+
+.. ipython:: python
+
+   df.loc[pd.Interval(1, 2)]
+
+Trying to select an ``Interval`` that is not exactly contained in the ``IntervalIndex`` will raise a ``KeyError``.
+
+.. ipython:: python
+   :okexcept:
+
+   df.loc[pd.Interval(0.5, 2.5)]
+
+Selecting all ``Intervals`` that overlap a given ``Interval`` can be performed using the
+:meth:`~IntervalIndex.overlaps` method to create a boolean indexer.
+
+.. ipython:: python
+
+   idxr = df.index.overlaps(pd.Interval(0.5, 2.5))
+   idxr
+   df[idxr]
+
+Binning data with ``cut`` and ``qcut``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`cut` and :func:`qcut` both return a ``Categorical`` object, and the bins they
+create are stored as an ``IntervalIndex`` in its ``.categories`` attribute.
+
+.. ipython:: python
+
+   c = pd.cut(range(4), bins=2)
+   c
+   c.categories
+
+:func:`cut` also accepts an ``IntervalIndex`` for its ``bins`` argument, which enables
+a useful pandas idiom. First, We call :func:`cut` with some data and ``bins`` set to a
+fixed number, to generate the bins. Then, we pass the values of ``.categories`` as the
+``bins`` argument in subsequent calls to :func:`cut`, supplying new data which will be
+binned into the same bins.
+
+.. ipython:: python
+
+   pd.cut([0, 3, 5, 1], bins=c.categories)
+
+Any value which falls outside all bins will be assigned a ``NaN`` value.
+
+Generating ranges of intervals
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+If we need intervals on a regular frequency, we can use the :func:`interval_range` function
+to create an ``IntervalIndex`` using various combinations of ``start``, ``end``, and ``periods``.
+The default frequency for ``interval_range`` is a 1 for numeric intervals, and calendar day for
+datetime-like intervals:
+
+.. ipython:: python
+
+   pd.interval_range(start=0, end=5)
+
+   pd.interval_range(start=pd.Timestamp("2017-01-01"), periods=4)
+
+   pd.interval_range(end=pd.Timedelta("3 days"), periods=3)
+
+The ``freq`` parameter can used to specify non-default frequencies, and can utilize a variety
+of :ref:`frequency aliases <timeseries.offset_aliases>` with datetime-like intervals:
+
+.. ipython:: python
+
+   pd.interval_range(start=0, periods=5, freq=1.5)
+
+   pd.interval_range(start=pd.Timestamp("2017-01-01"), periods=4, freq="W")
+
+   pd.interval_range(start=pd.Timedelta("0 days"), periods=3, freq="9h")
+
+Additionally, the ``closed`` parameter can be used to specify which side(s) the intervals
+are closed on.  Intervals are closed on the right side by default.
+
+.. ipython:: python
+
+   pd.interval_range(start=0, end=4, closed="both")
+
+   pd.interval_range(start=0, end=4, closed="neither")
+
+Specifying ``start``, ``end``, and ``periods`` will generate a range of evenly spaced
+intervals from ``start`` to ``end`` inclusively, with ``periods`` number of elements
+in the resulting ``IntervalIndex``:
+
+.. ipython:: python
+
+   pd.interval_range(start=0, end=6, periods=4)
+
+   pd.interval_range(pd.Timestamp("2018-01-01"), pd.Timestamp("2018-02-28"), periods=3)
+
+Miscellaneous indexing FAQ
+--------------------------
+
+Integer indexing
+~~~~~~~~~~~~~~~~
+
+Label-based indexing with integer axis labels is a thorny topic. It has been
+discussed heavily on mailing lists and among various members of the scientific
+Python community. In pandas, our general viewpoint is that labels matter more
+than integer locations. Therefore, with an integer axis index *only*
+label-based indexing is possible with the standard tools like ``.loc``. The
+following code will generate exceptions:
+
+.. ipython:: python
+   :okexcept:
+
+   s = pd.Series(range(5))
+   s[-1]
+   df = pd.DataFrame(np.random.randn(5, 4))
+   df
+   df.loc[-2:]
+
+This deliberate decision was made to prevent ambiguities and subtle bugs (many
+users reported finding bugs when the API change was made to stop "falling back"
+on position-based indexing).
+
+Non-monotonic indexes require exact matches
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If the index of a ``Series`` or ``DataFrame`` is monotonically increasing or decreasing, then the bounds
+of a label-based slice can be outside the range of the index, much like slice indexing a
+normal Python ``list``. Monotonicity of an index can be tested with the :meth:`~Index.is_monotonic_increasing` and
+:meth:`~Index.is_monotonic_decreasing` attributes.
+
+.. ipython:: python
+
+    df = pd.DataFrame(index=[2, 3, 3, 4, 5], columns=["data"], data=list(range(5)))
+    df.index.is_monotonic_increasing
+
+    # no rows 0 or 1, but still returns rows 2, 3 (both of them), and 4:
+    df.loc[0:4, :]
+
+    # slice is are outside the index, so empty DataFrame is returned
+    df.loc[13:15, :]
+
+On the other hand, if the index is not monotonic, then both slice bounds must be
+*unique* members of the index.
+
+.. ipython:: python
+
+    df = pd.DataFrame(index=[2, 3, 1, 4, 3, 5], columns=["data"], data=list(range(6)))
+    df.index.is_monotonic_increasing
+
+    # OK because 2 and 4 are in the index
+    df.loc[2:4, :]
+
+.. ipython:: python
+   :okexcept:
+
+    # 0 is not in the index
+    df.loc[0:4, :]
+
+    # 3 is not a unique label
+    df.loc[2:3, :]
+
+``Index.is_monotonic_increasing`` and ``Index.is_monotonic_decreasing`` only check that
+an index is weakly monotonic. To check for strict monotonicity, you can combine one of those with
+the :meth:`~Index.is_unique` attribute.
+
+.. ipython:: python
+
+   weakly_monotonic = pd.Index(["a", "b", "c", "c"])
+   weakly_monotonic
+   weakly_monotonic.is_monotonic_increasing
+   weakly_monotonic.is_monotonic_increasing & weakly_monotonic.is_unique
+
+.. _advanced.endpoints_are_inclusive:
+
+Endpoints are inclusive
+~~~~~~~~~~~~~~~~~~~~~~~
+
+Compared with standard Python sequence slicing in which the slice endpoint is
+not inclusive, label-based slicing in pandas **is inclusive**. The primary
+reason for this is that it is often not possible to easily determine the
+"successor" or next element after a particular label in an index. For example,
+consider the following ``Series``:
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(6), index=list("abcdef"))
+   s
+
+Suppose we wished to slice from ``c`` to ``e``, using integers this would be
+accomplished as such:
+
+.. ipython:: python
+
+   s[2:5]
+
+However, if you only had ``c`` and ``e``, determining the next element in the
+index can be somewhat complicated. For example, the following does not work:
+
+.. ipython:: python
+   :okexcept:
+
+    s.loc['c':'e' + 1]
+
+A very common use case is to limit a time series to start and end at two
+specific dates. To enable this, we made the design choice to make label-based
+slicing include both endpoints:
+
+.. ipython:: python
+
+    s.loc["c":"e"]
+
+This is most definitely a "practicality beats purity" sort of thing, but it is
+something to watch out for if you expect label-based slicing to behave exactly
+in the way that standard Python integer slicing works.
+
+
+Indexing potentially changes underlying Series dtype
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The different indexing operation can potentially change the dtype of a ``Series``.
+
+.. ipython:: python
+
+   series1 = pd.Series([1, 2, 3])
+   series1.dtype
+   res = series1.reindex([0, 4])
+   res.dtype
+   res
+
+.. ipython:: python
+
+   series2 = pd.Series([True])
+   series2.dtype
+   res = series2.reindex_like(series1)
+   res.dtype
+   res
+
+This is because the (re)indexing operations above silently inserts ``NaNs`` and the ``dtype``
+changes accordingly.  This can cause some issues when using ``numpy`` ``ufuncs``
+such as ``numpy.logical_and``.
+
+See the :issue:`2388` for a more
+detailed discussion.
diff --git a/pandas/doc/source/user_guide/basics.rst b/pandas/doc/source/user_guide/basics.rst
new file mode 100644
index 0000000000000..ffd7a2ad7bb01
--- /dev/null
+++ b/pandas/doc/source/user_guide/basics.rst
@@ -0,0 +1,2401 @@
+.. _basics:
+
+{{ header }}
+
+==============================
+ Essential basic functionality
+==============================
+
+Here we discuss a lot of the essential functionality common to the pandas data
+structures. To begin, let's create some example objects like we did in
+the :ref:`10 minutes to pandas <10min>` section:
+
+.. ipython:: python
+
+   index = pd.date_range("1/1/2000", periods=8)
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   df = pd.DataFrame(np.random.randn(8, 3), index=index, columns=["A", "B", "C"])
+
+.. _basics.head_tail:
+
+Head and tail
+-------------
+
+To view a small sample of a Series or DataFrame object, use the
+:meth:`~DataFrame.head` and :meth:`~DataFrame.tail` methods. The default number
+of elements to display is five, but you may pass a custom number.
+
+.. ipython:: python
+
+   long_series = pd.Series(np.random.randn(1000))
+   long_series.head()
+   long_series.tail(3)
+
+.. _basics.attrs:
+
+Attributes and underlying data
+------------------------------
+
+pandas objects have a number of attributes enabling you to access the metadata
+
+* **shape**: gives the axis dimensions of the object, consistent with ndarray
+* Axis labels
+    * **Series**: *index* (only axis)
+    * **DataFrame**: *index* (rows) and *columns*
+
+Note, **these attributes can be safely assigned to**!
+
+.. ipython:: python
+
+   df[:2]
+   df.columns = [x.lower() for x in df.columns]
+   df
+
+pandas objects (:class:`Index`, :class:`Series`, :class:`DataFrame`) can be
+thought of as containers for arrays, which hold the actual data and do the
+actual computation. For many types, the underlying array is a
+:class:`numpy.ndarray`. However, pandas and 3rd party libraries may *extend*
+NumPy's type system to add support for custom arrays
+(see :ref:`basics.dtypes`).
+
+To get the actual data inside a :class:`Index` or :class:`Series`, use
+the ``.array`` property
+
+.. ipython:: python
+
+   s.array
+   s.index.array
+
+:attr:`~Series.array` will always be an :class:`~pandas.api.extensions.ExtensionArray`.
+The exact details of what an :class:`~pandas.api.extensions.ExtensionArray` is and why pandas uses them are a bit
+beyond the scope of this introduction. See :ref:`basics.dtypes` for more.
+
+If you know you need a NumPy array, use :meth:`~Series.to_numpy`
+or :meth:`numpy.asarray`.
+
+.. ipython:: python
+
+   s.to_numpy()
+   np.asarray(s)
+
+When the Series or Index is backed by
+an :class:`~pandas.api.extensions.ExtensionArray`, :meth:`~Series.to_numpy`
+may involve copying data and coercing values. See :ref:`basics.dtypes` for more.
+
+:meth:`~Series.to_numpy` gives some control over the ``dtype`` of the
+resulting :class:`numpy.ndarray`. For example, consider datetimes with timezones.
+NumPy doesn't have a dtype to represent timezone-aware datetimes, so there
+are two possibly useful representations:
+
+1. An object-dtype :class:`numpy.ndarray` with :class:`Timestamp` objects, each
+   with the correct ``tz``
+2. A ``datetime64[ns]`` -dtype :class:`numpy.ndarray`, where the values have
+   been converted to UTC and the timezone discarded
+
+Timezones may be preserved with ``dtype=object``
+
+.. ipython:: python
+
+   ser = pd.Series(pd.date_range("2000", periods=2, tz="CET"))
+   ser.to_numpy(dtype=object)
+
+Or thrown away with ``dtype='datetime64[ns]'``
+
+.. ipython:: python
+
+   ser.to_numpy(dtype="datetime64[ns]")
+
+Getting the "raw data" inside a :class:`DataFrame` is possibly a bit more
+complex. When your ``DataFrame`` only has a single data type for all the
+columns, :meth:`DataFrame.to_numpy` will return the underlying data:
+
+.. ipython:: python
+
+   df.to_numpy()
+
+If a DataFrame contains homogeneously-typed data, the ndarray can
+actually be modified in-place, and the changes will be reflected in the data
+structure. For heterogeneous data (e.g. some of the DataFrame's columns are not
+all the same dtype), this will not be the case. The values attribute itself,
+unlike the axis labels, cannot be assigned to.
+
+.. note::
+
+    When working with heterogeneous data, the dtype of the resulting ndarray
+    will be chosen to accommodate all of the data involved. For example, if
+    strings are involved, the result will be of object dtype. If there are only
+    floats and integers, the resulting array will be of float dtype.
+
+In the past, pandas recommended :attr:`Series.values` or :attr:`DataFrame.values`
+for extracting the data from a Series or DataFrame. You'll still find references
+to these in old code bases and online. Going forward, we recommend avoiding
+``.values`` and using ``.array`` or ``.to_numpy()``. ``.values`` has the following
+drawbacks:
+
+1. When your Series contains an :ref:`extension type <extending.extension-types>`, it's
+   unclear whether :attr:`Series.values` returns a NumPy array or the extension array.
+   :attr:`Series.array` will always return an :class:`~pandas.api.extensions.ExtensionArray`, and will never
+   copy data. :meth:`Series.to_numpy` will always return a NumPy array,
+   potentially at the cost of copying / coercing values.
+2. When your DataFrame contains a mixture of data types, :attr:`DataFrame.values` may
+   involve copying data and coercing values to a common dtype, a relatively expensive
+   operation. :meth:`DataFrame.to_numpy`, being a method, makes it clearer that the
+   returned NumPy array may not be a view on the same data in the DataFrame.
+
+.. _basics.accelerate:
+
+Accelerated operations
+----------------------
+
+pandas has support for accelerating certain types of binary numerical and boolean operations using
+the ``numexpr`` library and the ``bottleneck`` libraries.
+
+These libraries are especially useful when dealing with large data sets, and provide large
+speedups. ``numexpr`` uses smart chunking, caching, and multiple cores. ``bottleneck`` is
+a set of specialized cython routines that are especially fast when dealing with arrays that have
+``nans``.
+
+You are highly encouraged to install both libraries. See the section
+:ref:`Recommended Dependencies <install.recommended_dependencies>` for more installation info.
+
+These are both enabled to be used by default, you can control this by setting the options:
+
+.. code-block:: python
+
+   pd.set_option("compute.use_bottleneck", False)
+   pd.set_option("compute.use_numexpr", False)
+
+.. _basics.binop:
+
+Flexible binary operations
+--------------------------
+
+With binary operations between pandas data structures, there are two key points
+of interest:
+
+* Broadcasting behavior between higher- (e.g. DataFrame) and
+  lower-dimensional (e.g. Series) objects.
+* Missing data in computations.
+
+We will demonstrate how to manage these issues independently, though they can
+be handled simultaneously.
+
+Matching / broadcasting behavior
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+DataFrame has the methods :meth:`~DataFrame.add`, :meth:`~DataFrame.sub`,
+:meth:`~DataFrame.mul`, :meth:`~DataFrame.div` and related functions
+:meth:`~DataFrame.radd`, :meth:`~DataFrame.rsub`, ...
+for carrying out binary operations. For broadcasting behavior,
+Series input is of primary interest. Using these functions, you can use to
+either match on the *index* or *columns* via the **axis** keyword:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "one": pd.Series(np.random.randn(3), index=["a", "b", "c"]),
+           "two": pd.Series(np.random.randn(4), index=["a", "b", "c", "d"]),
+           "three": pd.Series(np.random.randn(3), index=["b", "c", "d"]),
+       }
+   )
+   df
+   row = df.iloc[1]
+   column = df["two"]
+
+   df.sub(row, axis="columns")
+   df.sub(row, axis=1)
+
+   df.sub(column, axis="index")
+   df.sub(column, axis=0)
+
+Furthermore you can align a level of a MultiIndexed DataFrame with a Series.
+
+.. ipython:: python
+
+   dfmi = df.copy()
+   dfmi.index = pd.MultiIndex.from_tuples(
+       [(1, "a"), (1, "b"), (1, "c"), (2, "a")], names=["first", "second"]
+   )
+   dfmi.sub(column, axis=0, level="second")
+
+Series and Index also support the :func:`divmod` builtin. This function takes
+the floor division and modulo operation at the same time returning a two-tuple
+of the same type as the left hand side. For example:
+
+.. ipython:: python
+
+   s = pd.Series(np.arange(10))
+   s
+   div, rem = divmod(s, 3)
+   div
+   rem
+
+   idx = pd.Index(np.arange(10))
+   idx
+   div, rem = divmod(idx, 3)
+   div
+   rem
+
+We can also do elementwise :func:`divmod`:
+
+.. ipython:: python
+
+   div, rem = divmod(s, [2, 2, 3, 3, 4, 4, 5, 5, 6, 6])
+   div
+   rem
+
+Missing data / operations with fill values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In Series and DataFrame, the arithmetic functions have the option of inputting
+a *fill_value*, namely a value to substitute when at most one of the values at
+a location are missing. For example, when adding two DataFrame objects, you may
+wish to treat NaN as 0 unless both DataFrames are missing that value, in which
+case the result will be NaN (you can later replace NaN with some other value
+using ``fillna`` if you wish).
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2.loc["a", "three"] = 1.0
+   df
+   df2
+   df + df2
+   df.add(df2, fill_value=0)
+
+.. _basics.compare:
+
+Flexible comparisons
+~~~~~~~~~~~~~~~~~~~~
+
+Series and DataFrame have the binary comparison methods ``eq``, ``ne``, ``lt``, ``gt``,
+``le``, and ``ge`` whose behavior is analogous to the binary
+arithmetic operations described above:
+
+.. ipython:: python
+
+   df.gt(df2)
+   df2.ne(df)
+
+These operations produce a pandas object of the same type as the left-hand-side
+input that is of dtype ``bool``. These ``boolean`` objects can be used in
+indexing operations, see the section on :ref:`Boolean indexing<indexing.boolean>`.
+
+.. _basics.reductions:
+
+Boolean reductions
+~~~~~~~~~~~~~~~~~~
+
+You can apply the reductions: :attr:`~DataFrame.empty`, :meth:`~DataFrame.any`,
+:meth:`~DataFrame.all`.
+
+.. ipython:: python
+
+   (df > 0).all()
+   (df > 0).any()
+
+You can reduce to a final boolean value.
+
+.. ipython:: python
+
+   (df > 0).any().any()
+
+You can test if a pandas object is empty, via the :attr:`~DataFrame.empty` property.
+
+.. ipython:: python
+
+   df.empty
+   pd.DataFrame(columns=list("ABC")).empty
+
+.. warning::
+
+   Asserting the truthiness of a pandas object will raise an error, as the testing of the emptiness
+   or values is ambiguous.
+
+   .. ipython:: python
+      :okexcept:
+
+      if df:
+          print(True)
+
+   .. ipython:: python
+      :okexcept:
+
+      df and df2
+
+   See :ref:`gotchas<gotchas.truth>` for a more detailed discussion.
+
+.. _basics.equals:
+
+Comparing if objects are equivalent
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Often you may find that there is more than one way to compute the same
+result.  As a simple example, consider ``df + df`` and ``df * 2``. To test
+that these two computations produce the same result, given the tools
+shown above, you might imagine using ``(df + df == df * 2).all()``. But in
+fact, this expression is False:
+
+.. ipython:: python
+
+   df + df == df * 2
+   (df + df == df * 2).all()
+
+Notice that the boolean DataFrame ``df + df == df * 2`` contains some False values!
+This is because NaNs do not compare as equals:
+
+.. ipython:: python
+
+   np.nan == np.nan
+
+So, NDFrames (such as Series and DataFrames)
+have an :meth:`~DataFrame.equals` method for testing equality, with NaNs in
+corresponding locations treated as equal.
+
+.. ipython:: python
+
+   (df + df).equals(df * 2)
+
+Note that the Series or DataFrame index needs to be in the same order for
+equality to be True:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"col": ["foo", 0, np.nan]})
+   df2 = pd.DataFrame({"col": [np.nan, 0, "foo"]}, index=[2, 1, 0])
+   df1.equals(df2)
+   df1.equals(df2.sort_index())
+
+Comparing array-like objects
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can conveniently perform element-wise comparisons when comparing a pandas
+data structure with a scalar value:
+
+.. ipython:: python
+
+   pd.Series(["foo", "bar", "baz"]) == "foo"
+   pd.Index(["foo", "bar", "baz"]) == "foo"
+
+pandas also handles element-wise comparisons between different array-like
+objects of the same length:
+
+.. ipython:: python
+
+    pd.Series(["foo", "bar", "baz"]) == pd.Index(["foo", "bar", "qux"])
+    pd.Series(["foo", "bar", "baz"]) == np.array(["foo", "bar", "qux"])
+
+Trying to compare ``Index`` or ``Series`` objects of different lengths will
+raise a ValueError:
+
+.. ipython:: python
+   :okexcept:
+
+    pd.Series(['foo', 'bar', 'baz']) == pd.Series(['foo', 'bar'])
+
+    pd.Series(['foo', 'bar', 'baz']) == pd.Series(['foo'])
+
+Combining overlapping data sets
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A problem occasionally arising is the combination of two similar data sets
+where values in one are preferred over the other. An example would be two data
+series representing a particular economic indicator where one is considered to
+be of "higher quality". However, the lower quality series might extend further
+back in history or have more complete data coverage. As such, we would like to
+combine two DataFrame objects where missing values in one DataFrame are
+conditionally filled with like-labeled values from the other DataFrame. The
+function implementing this operation is :meth:`~DataFrame.combine_first`,
+which we illustrate:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(
+       {"A": [1.0, np.nan, 3.0, 5.0, np.nan], "B": [np.nan, 2.0, 3.0, np.nan, 6.0]}
+   )
+   df2 = pd.DataFrame(
+       {
+           "A": [5.0, 2.0, 4.0, np.nan, 3.0, 7.0],
+           "B": [np.nan, np.nan, 3.0, 4.0, 6.0, 8.0],
+       }
+   )
+   df1
+   df2
+   df1.combine_first(df2)
+
+General DataFrame combine
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.combine_first` method above calls the more general
+:meth:`DataFrame.combine`. This method takes another DataFrame
+and a combiner function, aligns the input DataFrame and then passes the combiner
+function pairs of Series (i.e., columns whose names are the same).
+
+So, for instance, to reproduce :meth:`~DataFrame.combine_first` as above:
+
+.. ipython:: python
+
+   def combiner(x, y):
+       return np.where(pd.isna(x), y, x)
+
+
+   df1.combine(df2, combiner)
+
+.. _basics.stats:
+
+Descriptive statistics
+----------------------
+
+There exists a large number of methods for computing descriptive statistics and
+other related operations on :ref:`Series <api.series.stats>`, :ref:`DataFrame
+<api.dataframe.stats>`. Most of these
+are aggregations (hence producing a lower-dimensional result) like
+:meth:`~DataFrame.sum`, :meth:`~DataFrame.mean`, and :meth:`~DataFrame.quantile`,
+but some of them, like :meth:`~DataFrame.cumsum` and :meth:`~DataFrame.cumprod`,
+produce an object of the same size. Generally speaking, these methods take an
+**axis** argument, just like *ndarray.{sum, std, ...}*, but the axis can be
+specified by name or integer:
+
+* **Series**: no axis argument needed
+* **DataFrame**: "index" (axis=0, default), "columns" (axis=1)
+
+For example:
+
+.. ipython:: python
+
+   df
+   df.mean(axis=0)
+   df.mean(axis=1)
+
+All such methods have a ``skipna`` option signaling whether to exclude missing
+data (``True`` by default):
+
+.. ipython:: python
+
+   df.sum(axis=0, skipna=False)
+   df.sum(axis=1, skipna=True)
+
+Combined with the broadcasting / arithmetic behavior, one can describe various
+statistical procedures, like standardization (rendering data zero mean and
+standard deviation of 1), very concisely:
+
+.. ipython:: python
+
+   ts_stand = (df - df.mean()) / df.std()
+   ts_stand.std()
+   xs_stand = df.sub(df.mean(axis=1), axis=0).div(df.std(axis=1), axis=0)
+   xs_stand.std(axis=1)
+
+Note that methods like :meth:`~DataFrame.cumsum` and :meth:`~DataFrame.cumprod`
+preserve the location of ``NaN`` values. This is somewhat different from
+:meth:`~DataFrame.expanding` and :meth:`~DataFrame.rolling` since ``NaN`` behavior
+is furthermore dictated by a ``min_periods`` parameter.
+
+.. ipython:: python
+
+   df.cumsum()
+
+Here is a quick reference summary table of common functions. Each also takes an
+optional ``level`` parameter which applies only if the object has a
+:ref:`hierarchical index<advanced.hierarchical>`.
+
+.. csv-table::
+    :header: "Function", "Description"
+    :widths: 20, 80
+
+    ``count``, Number of non-NA observations
+    ``sum``, Sum of values
+    ``mean``, Mean of values
+    ``median``, Arithmetic median of values
+    ``min``, Minimum
+    ``max``, Maximum
+    ``mode``, Mode
+    ``abs``, Absolute Value
+    ``prod``, Product of values
+    ``std``, Bessel-corrected sample standard deviation
+    ``var``, Unbiased variance
+    ``sem``, Standard error of the mean
+    ``skew``, Sample skewness (3rd moment)
+    ``kurt``, Sample kurtosis (4th moment)
+    ``quantile``, Sample quantile (value at %)
+    ``cumsum``, Cumulative sum
+    ``cumprod``, Cumulative product
+    ``cummax``, Cumulative maximum
+    ``cummin``, Cumulative minimum
+
+Note that by chance some NumPy methods, like ``mean``, ``std``, and ``sum``,
+will exclude NAs on Series input by default:
+
+.. ipython:: python
+
+   np.mean(df["one"])
+   np.mean(df["one"].to_numpy())
+
+:meth:`Series.nunique` will return the number of unique non-NA values in a
+Series:
+
+.. ipython:: python
+
+   series = pd.Series(np.random.randn(500))
+   series[20:500] = np.nan
+   series[10:20] = 5
+   series.nunique()
+
+.. _basics.describe:
+
+Summarizing data: describe
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+There is a convenient :meth:`~DataFrame.describe` function which computes a variety of summary
+statistics about a Series or the columns of a DataFrame (excluding NAs of
+course):
+
+.. ipython:: python
+
+    series = pd.Series(np.random.randn(1000))
+    series[::2] = np.nan
+    series.describe()
+    frame = pd.DataFrame(np.random.randn(1000, 5), columns=["a", "b", "c", "d", "e"])
+    frame.iloc[::2] = np.nan
+    frame.describe()
+
+You can select specific percentiles to include in the output:
+
+.. ipython:: python
+
+    series.describe(percentiles=[0.05, 0.25, 0.75, 0.95])
+
+By default, the median is always included.
+
+For a non-numerical Series object, :meth:`~Series.describe` will give a simple
+summary of the number of unique values and most frequently occurring values:
+
+.. ipython:: python
+
+   s = pd.Series(["a", "a", "b", "b", "a", "a", np.nan, "c", "d", "a"])
+   s.describe()
+
+Note that on a mixed-type DataFrame object, :meth:`~DataFrame.describe` will
+restrict the summary to include only numerical columns or, if none are, only
+categorical columns:
+
+.. ipython:: python
+
+    frame = pd.DataFrame({"a": ["Yes", "Yes", "No", "No"], "b": range(4)})
+    frame.describe()
+
+This behavior can be controlled by providing a list of types as ``include``/``exclude``
+arguments. The special value ``all`` can also be used:
+
+.. ipython:: python
+
+    frame.describe(include=["object"])
+    frame.describe(include=["number"])
+    frame.describe(include="all")
+
+That feature relies on :ref:`select_dtypes <basics.selectdtypes>`. Refer to
+there for details about accepted inputs.
+
+.. _basics.idxmin:
+
+Index of min/max values
+~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.idxmin` and :meth:`~DataFrame.idxmax` functions on Series
+and DataFrame compute the index labels with the minimum and maximum
+corresponding values:
+
+.. ipython:: python
+
+   s1 = pd.Series(np.random.randn(5))
+   s1
+   s1.idxmin(), s1.idxmax()
+
+   df1 = pd.DataFrame(np.random.randn(5, 3), columns=["A", "B", "C"])
+   df1
+   df1.idxmin(axis=0)
+   df1.idxmax(axis=1)
+
+When there are multiple rows (or columns) matching the minimum or maximum
+value, :meth:`~DataFrame.idxmin` and :meth:`~DataFrame.idxmax` return the first
+matching index:
+
+.. ipython:: python
+
+   df3 = pd.DataFrame([2, 1, 1, 3, np.nan], columns=["A"], index=list("edcba"))
+   df3
+   df3["A"].idxmin()
+
+.. note::
+
+   ``idxmin`` and ``idxmax`` are called ``argmin`` and ``argmax`` in NumPy.
+
+.. _basics.discretization:
+
+Value counts (histogramming) / mode
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~Series.value_counts` Series method computes a histogram
+of a 1D array of values. It can also be used as a function on regular arrays:
+
+.. ipython:: python
+
+   data = np.random.randint(0, 7, size=50)
+   data
+   s = pd.Series(data)
+   s.value_counts()
+
+The :meth:`~DataFrame.value_counts` method can be used to count combinations across multiple columns.
+By default all columns are used but a subset can be selected using the ``subset`` argument.
+
+.. ipython:: python
+
+    data = {"a": [1, 2, 3, 4], "b": ["x", "x", "y", "y"]}
+    frame = pd.DataFrame(data)
+    frame.value_counts()
+
+Similarly, you can get the most frequently occurring value(s), i.e. the mode, of the values in a Series or DataFrame:
+
+.. ipython:: python
+
+    s5 = pd.Series([1, 1, 3, 3, 3, 5, 5, 7, 7, 7])
+    s5.mode()
+    df5 = pd.DataFrame(
+        {
+            "A": np.random.randint(0, 7, size=50),
+            "B": np.random.randint(-10, 15, size=50),
+        }
+    )
+    df5.mode()
+
+
+Discretization and quantiling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Continuous values can be discretized using the :func:`cut` (bins based on values)
+and :func:`qcut` (bins based on sample quantiles) functions:
+
+.. ipython:: python
+
+   arr = np.random.randn(20)
+   factor = pd.cut(arr, 4)
+   factor
+
+   factor = pd.cut(arr, [-5, -1, 0, 1, 5])
+   factor
+
+:func:`qcut` computes sample quantiles. For example, we could slice up some
+normally distributed data into equal-size quartiles like so:
+
+.. ipython:: python
+
+   arr = np.random.randn(30)
+   factor = pd.qcut(arr, [0, 0.25, 0.5, 0.75, 1])
+   factor
+
+We can also pass infinite values to define the bins:
+
+.. ipython:: python
+
+   arr = np.random.randn(20)
+   factor = pd.cut(arr, [-np.inf, 0, np.inf])
+   factor
+
+.. _basics.apply:
+
+Function application
+--------------------
+
+To apply your own or another library's functions to pandas objects,
+you should be aware of the three methods below. The appropriate
+method to use depends on whether your function expects to operate
+on an entire ``DataFrame`` or ``Series``, row- or column-wise, or elementwise.
+
+1. `Tablewise Function Application`_: :meth:`~DataFrame.pipe`
+2. `Row or Column-wise Function Application`_: :meth:`~DataFrame.apply`
+3. `Aggregation API`_: :meth:`~DataFrame.agg` and :meth:`~DataFrame.transform`
+4. `Applying Elementwise Functions`_: :meth:`~DataFrame.map`
+
+.. _basics.pipe:
+
+Tablewise function application
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``DataFrames`` and ``Series`` can be passed into functions.
+However, if the function needs to be called in a chain, consider using the :meth:`~DataFrame.pipe` method.
+
+First some setup:
+
+.. ipython:: python
+
+    def extract_city_name(df):
+        """
+        Chicago, IL -> Chicago for city_name column
+        """
+        df["city_name"] = df["city_and_code"].str.split(",").str.get(0)
+        return df
+
+
+    def add_country_name(df, country_name=None):
+        """
+        Chicago -> Chicago-US for city_name column
+        """
+        col = "city_name"
+        df["city_and_country"] = df[col] + country_name
+        return df
+
+
+    df_p = pd.DataFrame({"city_and_code": ["Chicago, IL"]})
+
+
+``extract_city_name`` and ``add_country_name`` are functions taking and returning ``DataFrames``.
+
+Now compare the following:
+
+.. ipython:: python
+
+    add_country_name(extract_city_name(df_p), country_name="US")
+
+Is equivalent to:
+
+.. ipython:: python
+
+    df_p.pipe(extract_city_name).pipe(add_country_name, country_name="US")
+
+pandas encourages the second style, which is known as method chaining.
+``pipe`` makes it easy to use your own or another library's functions
+in method chains, alongside pandas' methods.
+
+In the example above, the functions ``extract_city_name`` and ``add_country_name`` each expected a ``DataFrame`` as the first positional argument.
+What if the function you wish to apply takes its data as, say, the second argument?
+In this case, provide ``pipe`` with a tuple of ``(callable, data_keyword)``.
+``.pipe`` will route the ``DataFrame`` to the argument specified in the tuple.
+
+For example, we can fit a regression using statsmodels. Their API expects a formula first and a ``DataFrame`` as the second argument, ``data``. We pass in the function, keyword pair ``(sm.ols, 'data')`` to ``pipe``:
+
+.. code-block:: ipython
+
+   In [147]: import statsmodels.formula.api as sm
+
+   In [148]: bb = pd.read_csv("data/baseball.csv", index_col="id")
+
+   In [149]: (
+      .....:     bb.query("h > 0")
+      .....:     .assign(ln_h=lambda df: np.log(df.h))
+      .....:     .pipe((sm.ols, "data"), "hr ~ ln_h + year + g + C(lg)")
+      .....:     .fit()
+      .....:     .summary()
+      .....: )
+      .....:
+   Out[149]:
+   <class 'statsmodels.iolib.summary.Summary'>
+   """
+                              OLS Regression Results
+   ==============================================================================
+   Dep. Variable:                     hr   R-squared:                       0.685
+   Model:                            OLS   Adj. R-squared:                  0.665
+   Method:                 Least Squares   F-statistic:                     34.28
+   Date:                Tue, 22 Nov 2022   Prob (F-statistic):           3.48e-15
+   Time:                        05:34:17   Log-Likelihood:                -205.92
+   No. Observations:                  68   AIC:                             421.8
+   Df Residuals:                      63   BIC:                             432.9
+   Df Model:                           4
+   Covariance Type:            nonrobust
+   ===============================================================================
+                     coef    std err          t      P>|t|      [0.025      0.975]
+   -------------------------------------------------------------------------------
+   Intercept   -8484.7720   4664.146     -1.819      0.074   -1.78e+04     835.780
+   C(lg)[T.NL]    -2.2736      1.325     -1.716      0.091      -4.922       0.375
+   ln_h           -1.3542      0.875     -1.547      0.127      -3.103       0.395
+   year            4.2277      2.324      1.819      0.074      -0.417       8.872
+   g               0.1841      0.029      6.258      0.000       0.125       0.243
+   ==============================================================================
+   Omnibus:                       10.875   Durbin-Watson:                   1.999
+   Prob(Omnibus):                  0.004   Jarque-Bera (JB):               17.298
+   Skew:                           0.537   Prob(JB):                     0.000175
+   Kurtosis:                       5.225   Cond. No.                     1.49e+07
+   ==============================================================================
+
+   Notes:
+   [1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+   [2] The condition number is large, 1.49e+07. This might indicate that there are
+   strong multicollinearity or other numerical problems.
+   """
+
+The pipe method is inspired by unix pipes and more recently dplyr_ and magrittr_, which
+have introduced the popular ``(%>%)`` (read pipe) operator for R_.
+The implementation of ``pipe`` here is quite clean and feels right at home in Python.
+We encourage you to view the source code of :meth:`~DataFrame.pipe`.
+
+.. _dplyr: https://github.com/tidyverse/dplyr
+.. _magrittr: https://github.com/tidyverse/magrittr
+.. _R: https://www.r-project.org
+
+
+Row or column-wise function application
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Arbitrary functions can be applied along the axes of a DataFrame
+using the :meth:`~DataFrame.apply` method, which, like the descriptive
+statistics methods, takes an optional ``axis`` argument:
+
+.. ipython:: python
+
+   df.apply(lambda x: np.mean(x))
+   df.apply(lambda x: np.mean(x), axis=1)
+   df.apply(lambda x: x.max() - x.min())
+   df.apply(np.cumsum)
+   df.apply(np.exp)
+
+The :meth:`~DataFrame.apply` method will also dispatch on a string method name.
+
+.. ipython:: python
+
+   df.apply("mean")
+   df.apply("mean", axis=1)
+
+The return type of the function passed to :meth:`~DataFrame.apply` affects the
+type of the final output from ``DataFrame.apply`` for the default behaviour:
+
+* If the applied function returns a ``Series``, the final output is a ``DataFrame``.
+  The columns match the index of the ``Series`` returned by the applied function.
+* If the applied function returns any other type, the final output is a ``Series``.
+
+This default behaviour can be overridden using the ``result_type``, which
+accepts three options: ``reduce``, ``broadcast``, and ``expand``.
+These will determine how list-likes return values expand (or not) to a ``DataFrame``.
+
+:meth:`~DataFrame.apply` combined with some cleverness can be used to answer many questions
+about a data set. For example, suppose we wanted to extract the date where the
+maximum value for each column occurred:
+
+.. ipython:: python
+
+   tsdf = pd.DataFrame(
+       np.random.randn(1000, 3),
+       columns=["A", "B", "C"],
+       index=pd.date_range("1/1/2000", periods=1000),
+   )
+   tsdf.apply(lambda x: x.idxmax())
+
+You may also pass additional arguments and keyword arguments to the :meth:`~DataFrame.apply`
+method.
+
+.. ipython:: python
+
+   def subtract_and_divide(x, sub, divide=1):
+       return (x - sub) / divide
+
+   df_udf = pd.DataFrame(np.ones((2, 2)))
+   df_udf.apply(subtract_and_divide, args=(5,), divide=3)
+
+Another useful feature is the ability to pass Series methods to carry out some
+Series operation on each column or row:
+
+.. ipython:: python
+
+   tsdf = pd.DataFrame(
+       np.random.randn(10, 3),
+       columns=["A", "B", "C"],
+       index=pd.date_range("1/1/2000", periods=10),
+   )
+   tsdf.iloc[3:7] = np.nan
+   tsdf
+   tsdf.apply(pd.Series.interpolate)
+
+
+Finally, :meth:`~DataFrame.apply` takes an argument ``raw`` which is False by default, which
+converts each row or column into a Series before applying the function. When
+set to True, the passed function will instead receive an ndarray object, which
+has positive performance implications if you do not need the indexing
+functionality.
+
+.. _basics.aggregate:
+
+Aggregation API
+~~~~~~~~~~~~~~~
+
+The aggregation API allows one to express possibly multiple aggregation operations in a single concise way.
+This API is similar across pandas objects, see :ref:`groupby API <groupby.aggregate>`, the
+:ref:`window API <window.overview>`, and the :ref:`resample API <timeseries.aggregate>`.
+The entry point for aggregation is :meth:`DataFrame.aggregate`, or the alias
+:meth:`DataFrame.agg`.
+
+We will use a similar starting frame from above:
+
+.. ipython:: python
+
+   tsdf = pd.DataFrame(
+       np.random.randn(10, 3),
+       columns=["A", "B", "C"],
+       index=pd.date_range("1/1/2000", periods=10),
+   )
+   tsdf.iloc[3:7] = np.nan
+   tsdf
+
+Using a single function is equivalent to :meth:`~DataFrame.apply`. You can also
+pass named methods as strings. These will return a ``Series`` of the aggregated
+output:
+
+.. ipython:: python
+
+   tsdf.agg(lambda x: np.sum(x))
+
+   tsdf.agg("sum")
+
+   # these are equivalent to a ``.sum()`` because we are aggregating
+   # on a single function
+   tsdf.sum()
+
+Single aggregations on a ``Series`` this will return a scalar value:
+
+.. ipython:: python
+
+   tsdf["A"].agg("sum")
+
+
+Aggregating with multiple functions
++++++++++++++++++++++++++++++++++++
+
+You can pass multiple aggregation arguments as a list.
+The results of each of the passed functions will be a row in the resulting ``DataFrame``.
+These are naturally named from the aggregation function.
+
+.. ipython:: python
+
+   tsdf.agg(["sum"])
+
+Multiple functions yield multiple rows:
+
+.. ipython:: python
+
+   tsdf.agg(["sum", "mean"])
+
+On a ``Series``, multiple functions return a ``Series``, indexed by the function names:
+
+.. ipython:: python
+
+   tsdf["A"].agg(["sum", "mean"])
+
+Passing a ``lambda`` function will yield a ``<lambda>`` named row:
+
+.. ipython:: python
+
+   tsdf["A"].agg(["sum", lambda x: x.mean()])
+
+Passing a named function will yield that name for the row:
+
+.. ipython:: python
+
+   def mymean(x):
+       return x.mean()
+
+
+   tsdf["A"].agg(["sum", mymean])
+
+Aggregating with a dict
++++++++++++++++++++++++
+
+Passing a dictionary of column names to a scalar or a list of scalars, to ``DataFrame.agg``
+allows you to customize which functions are applied to which columns. Note that the results
+are not in any particular order, you can use an ``OrderedDict`` instead to guarantee ordering.
+
+.. ipython:: python
+
+   tsdf.agg({"A": "mean", "B": "sum"})
+
+Passing a list-like will generate a ``DataFrame`` output. You will get a matrix-like output
+of all of the aggregators. The output will consist of all unique functions. Those that are
+not noted for a particular column will be ``NaN``:
+
+.. ipython:: python
+
+   tsdf.agg({"A": ["mean", "min"], "B": "sum"})
+
+.. _basics.aggregation.custom_describe:
+
+Custom describe
++++++++++++++++
+
+With ``.agg()`` it is possible to easily create a custom describe function, similar
+to the built in :ref:`describe function <basics.describe>`.
+
+.. ipython:: python
+
+   from functools import partial
+
+   q_25 = partial(pd.Series.quantile, q=0.25)
+   q_25.__name__ = "25%"
+   q_75 = partial(pd.Series.quantile, q=0.75)
+   q_75.__name__ = "75%"
+
+   tsdf.agg(["count", "mean", "std", "min", q_25, "median", q_75, "max"])
+
+.. _basics.transform:
+
+Transform API
+~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.transform` method returns an object that is indexed the same (same size)
+as the original. This API allows you to provide *multiple* operations at the same
+time rather than one-by-one. Its API is quite similar to the ``.agg`` API.
+
+We create a frame similar to the one used in the above sections.
+
+.. ipython:: python
+
+   tsdf = pd.DataFrame(
+       np.random.randn(10, 3),
+       columns=["A", "B", "C"],
+       index=pd.date_range("1/1/2000", periods=10),
+   )
+   tsdf.iloc[3:7] = np.nan
+   tsdf
+
+Transform the entire frame. ``.transform()`` allows input functions as: a NumPy function, a string
+function name or a user defined function.
+
+.. ipython:: python
+   :okwarning:
+
+   tsdf.transform(np.abs)
+   tsdf.transform("abs")
+   tsdf.transform(lambda x: x.abs())
+
+Here :meth:`~DataFrame.transform` received a single function; this is equivalent to a `ufunc
+<https://numpy.org/doc/stable/reference/ufuncs.html>`__ application.
+
+.. ipython:: python
+
+   np.abs(tsdf)
+
+Passing a single function to ``.transform()`` with a ``Series`` will yield a single ``Series`` in return.
+
+.. ipython:: python
+
+   tsdf["A"].transform(np.abs)
+
+
+Transform with multiple functions
++++++++++++++++++++++++++++++++++
+
+Passing multiple functions will yield a column MultiIndexed DataFrame.
+The first level will be the original frame column names; the second level
+will be the names of the transforming functions.
+
+.. ipython:: python
+
+   tsdf.transform([np.abs, lambda x: x + 1])
+
+Passing multiple functions to a Series will yield a DataFrame. The
+resulting column names will be the transforming functions.
+
+.. ipython:: python
+
+   tsdf["A"].transform([np.abs, lambda x: x + 1])
+
+
+Transforming with a dict
+++++++++++++++++++++++++
+
+
+Passing a dict of functions will allow selective transforming per column.
+
+.. ipython:: python
+
+   tsdf.transform({"A": np.abs, "B": lambda x: x + 1})
+
+Passing a dict of lists will generate a MultiIndexed DataFrame with these
+selective transforms.
+
+.. ipython:: python
+   :okwarning:
+
+   tsdf.transform({"A": np.abs, "B": [lambda x: x + 1, "sqrt"]})
+
+.. _basics.elementwise:
+
+Applying elementwise functions
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Since not all functions can be vectorized (accept NumPy arrays and return
+another array or value), the methods :meth:`~DataFrame.map` on DataFrame
+and analogously :meth:`~Series.map` on Series accept any Python function taking
+a single value and returning a single value. For example:
+
+.. ipython:: python
+
+   df4 = df.copy()
+   df4
+
+   def f(x):
+       return len(str(x))
+
+   df4["one"].map(f)
+   df4.map(f)
+
+:meth:`Series.map` has an additional feature; it can be used to easily
+"link" or "map" values defined by a secondary series. This is closely related
+to :ref:`merging/joining functionality <merging>`:
+
+.. ipython:: python
+
+   s = pd.Series(
+       ["six", "seven", "six", "seven", "six"], index=["a", "b", "c", "d", "e"]
+   )
+   t = pd.Series({"six": 6.0, "seven": 7.0})
+   s
+   s.map(t)
+
+
+.. _basics.reindexing:
+
+Reindexing and altering labels
+------------------------------
+
+:meth:`~Series.reindex` is the fundamental data alignment method in pandas.
+It is used to implement nearly all other features relying on label-alignment
+functionality. To *reindex* means to conform the data to match a given set of
+labels along a particular axis. This accomplishes several things:
+
+* Reorders the existing data to match a new set of labels
+* Inserts missing value (NA) markers in label locations where no data for
+  that label existed
+* If specified, **fill** data for missing labels using logic (highly relevant
+  to working with time series data)
+
+Here is a simple example:
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   s
+   s.reindex(["e", "b", "f", "d"])
+
+Here, the ``f`` label was not contained in the Series and hence appears as
+``NaN`` in the result.
+
+With a DataFrame, you can simultaneously reindex the index and columns:
+
+.. ipython:: python
+
+   df
+   df.reindex(index=["c", "f", "b"], columns=["three", "two", "one"])
+
+Note that the ``Index`` objects containing the actual axis labels can be
+**shared** between objects. So if we have a Series and a DataFrame, the
+following can be done:
+
+.. ipython:: python
+
+   rs = s.reindex(df.index)
+   rs
+   rs.index is df.index
+
+This means that the reindexed Series's index is the same Python object as the
+DataFrame's index.
+
+:meth:`DataFrame.reindex` also supports an "axis-style" calling convention,
+where you specify a single ``labels`` argument and the ``axis`` it applies to.
+
+.. ipython:: python
+
+   df.reindex(["c", "f", "b"], axis="index")
+   df.reindex(["three", "two", "one"], axis="columns")
+
+.. seealso::
+
+   :ref:`MultiIndex / Advanced Indexing <advanced>` is an even more concise way of
+   doing reindexing.
+
+.. note::
+
+    When writing performance-sensitive code, there is a good reason to spend
+    some time becoming a reindexing ninja: **many operations are faster on
+    pre-aligned data**. Adding two unaligned DataFrames internally triggers a
+    reindexing step. For exploratory analysis you will hardly notice the
+    difference (because ``reindex`` has been heavily optimized), but when CPU
+    cycles matter sprinkling a few explicit ``reindex`` calls here and there can
+    have an impact.
+
+.. _basics.reindex_like:
+
+Reindexing to align with another object
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You may wish to take an object and reindex its axes to be labeled the same as
+another object. While the syntax for this is straightforward albeit verbose, it
+is a common enough operation that the :meth:`~DataFrame.reindex_like` method is
+available to make this simpler:
+
+.. ipython:: python
+
+   df2 = df.reindex(["a", "b", "c"], columns=["one", "two"])
+   df3 = df2 - df2.mean()
+   df2
+   df3
+   df.reindex_like(df2)
+
+.. _basics.align:
+
+Aligning objects with each other with ``align``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~Series.align` method is the fastest way to simultaneously align two objects. It
+supports a ``join`` argument (related to :ref:`joining and merging <merging>`):
+
+  - ``join='outer'``: take the union of the indexes (default)
+  - ``join='left'``: use the calling object's index
+  - ``join='right'``: use the passed object's index
+  - ``join='inner'``: intersect the indexes
+
+It returns a tuple with both of the reindexed Series:
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   s1 = s[:4]
+   s2 = s[1:]
+   s1.align(s2)
+   s1.align(s2, join="inner")
+   s1.align(s2, join="left")
+
+.. _basics.df_join:
+
+For DataFrames, the join method will be applied to both the index and the
+columns by default:
+
+.. ipython:: python
+
+   df.align(df2, join="inner")
+
+You can also pass an ``axis`` option to only align on the specified axis:
+
+.. ipython:: python
+
+   df.align(df2, join="inner", axis=0)
+
+.. _basics.align.frame.series:
+
+If you pass a Series to :meth:`DataFrame.align`, you can choose to align both
+objects either on the DataFrame's index or columns using the ``axis`` argument:
+
+.. ipython:: python
+
+   df.align(df2.iloc[0], axis=1)
+
+.. _basics.reindex_fill:
+
+Filling while reindexing
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`~Series.reindex` takes an optional parameter ``method`` which is a
+filling method chosen from the following table:
+
+.. csv-table::
+    :header: "Method", "Action"
+    :widths: 30, 50
+
+    ffill, Fill values forward
+    bfill, Fill values backward
+    nearest, Fill from the nearest index value
+
+We illustrate these fill methods on a simple Series:
+
+.. ipython:: python
+
+   rng = pd.date_range("1/3/2000", periods=8)
+   ts = pd.Series(np.random.randn(8), index=rng)
+   ts2 = ts.iloc[[0, 3, 6]]
+   ts
+   ts2
+
+   ts2.reindex(ts.index)
+   ts2.reindex(ts.index, method="ffill")
+   ts2.reindex(ts.index, method="bfill")
+   ts2.reindex(ts.index, method="nearest")
+
+These methods require that the indexes are **ordered** increasing or
+decreasing.
+
+Note that the same result could have been achieved using
+:ref:`ffill <missing_data.fillna>` (except for ``method='nearest'``) or
+:ref:`interpolate <missing_data.interpolate>`:
+
+.. ipython:: python
+
+   ts2.reindex(ts.index).ffill()
+
+:meth:`~Series.reindex` will raise a ValueError if the index is not monotonically
+increasing or decreasing. :meth:`~Series.fillna` and :meth:`~Series.interpolate`
+will not perform any checks on the order of the index.
+
+.. _basics.limits_on_reindex_fill:
+
+Limits on filling while reindexing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The ``limit`` and ``tolerance`` arguments provide additional control over
+filling while reindexing. Limit specifies the maximum count of consecutive
+matches:
+
+.. ipython:: python
+
+   ts2.reindex(ts.index, method="ffill", limit=1)
+
+In contrast, tolerance specifies the maximum distance between the index and
+indexer values:
+
+.. ipython:: python
+
+   ts2.reindex(ts.index, method="ffill", tolerance="1 day")
+
+Notice that when used on a ``DatetimeIndex``, ``TimedeltaIndex`` or
+``PeriodIndex``, ``tolerance`` will coerced into a ``Timedelta`` if possible.
+This allows you to specify tolerance with appropriate strings.
+
+.. _basics.drop:
+
+Dropping labels from an axis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A method closely related to ``reindex`` is the :meth:`~DataFrame.drop` function.
+It removes a set of labels from an axis:
+
+.. ipython:: python
+
+   df
+   df.drop(["a", "d"], axis=0)
+   df.drop(["one"], axis=1)
+
+Note that the following also works, but is a bit less obvious / clean:
+
+.. ipython:: python
+
+   df.reindex(df.index.difference(["a", "d"]))
+
+.. _basics.rename:
+
+Renaming / mapping labels
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :meth:`~DataFrame.rename` method allows you to relabel an axis based on some
+mapping (a dict or Series) or an arbitrary function.
+
+.. ipython:: python
+
+   s
+   s.rename(str.upper)
+
+If you pass a function, it must return a value when called with any of the
+labels (and must produce a set of unique values). A dict or
+Series can also be used:
+
+.. ipython:: python
+
+   df.rename(
+       columns={"one": "foo", "two": "bar"},
+       index={"a": "apple", "b": "banana", "d": "durian"},
+   )
+
+If the mapping doesn't include a column/index label, it isn't renamed. Note that
+extra labels in the mapping don't throw an error.
+
+:meth:`DataFrame.rename` also supports an "axis-style" calling convention, where
+you specify a single ``mapper`` and the ``axis`` to apply that mapping to.
+
+.. ipython:: python
+
+   df.rename({"one": "foo", "two": "bar"}, axis="columns")
+   df.rename({"a": "apple", "b": "banana", "d": "durian"}, axis="index")
+
+Finally, :meth:`~Series.rename` also accepts a scalar or list-like
+for altering the ``Series.name`` attribute.
+
+.. ipython:: python
+
+   s.rename("scalar-name")
+
+.. _basics.rename_axis:
+
+The methods :meth:`DataFrame.rename_axis` and :meth:`Series.rename_axis`
+allow specific names of a ``MultiIndex`` to be changed (as opposed to the
+labels).
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"x": [1, 2, 3, 4, 5, 6], "y": [10, 20, 30, 40, 50, 60]},
+       index=pd.MultiIndex.from_product(
+           [["a", "b", "c"], [1, 2]], names=["let", "num"]
+       ),
+   )
+   df
+   df.rename_axis(index={"let": "abc"})
+   df.rename_axis(index=str.upper)
+
+.. _basics.iteration:
+
+Iteration
+---------
+
+The behavior of basic iteration over pandas objects depends on the type.
+When iterating over a Series, it is regarded as array-like, and basic iteration
+produces the values. DataFrames follow the dict-like convention of iterating
+over the "keys" of the objects.
+
+In short, basic iteration (``for i in object``) produces:
+
+* **Series**: values
+* **DataFrame**: column labels
+
+Thus, for example, iterating over a DataFrame gives you the column names:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"col1": np.random.randn(3), "col2": np.random.randn(3)}, index=["a", "b", "c"]
+   )
+
+   for col in df:
+       print(col)
+
+
+pandas objects also have the dict-like :meth:`~DataFrame.items` method to
+iterate over the (key, value) pairs.
+
+To iterate over the rows of a DataFrame, you can use the following methods:
+
+* :meth:`~DataFrame.iterrows`: Iterate over the rows of a DataFrame as (index, Series) pairs.
+  This converts the rows to Series objects, which can change the dtypes and has some
+  performance implications.
+* :meth:`~DataFrame.itertuples`: Iterate over the rows of a DataFrame
+  as namedtuples of the values.  This is a lot faster than
+  :meth:`~DataFrame.iterrows`, and is in most cases preferable to use
+  to iterate over the values of a DataFrame.
+
+.. warning::
+
+  Iterating through pandas objects is generally **slow**. In many cases,
+  iterating manually over the rows is not needed and can be avoided with
+  one of the following approaches:
+
+  * Look for a *vectorized* solution: many operations can be performed using
+    built-in methods or NumPy functions, (boolean) indexing, ...
+
+  * When you have a function that cannot work on the full DataFrame/Series
+    at once, it is better to use :meth:`~DataFrame.apply` instead of iterating
+    over the values. See the docs on :ref:`function application <basics.apply>`.
+
+  * If you need to do iterative manipulations on the values but performance is
+    important, consider writing the inner loop with cython or numba.
+    See the :ref:`enhancing performance <enhancingperf>` section for some
+    examples of this approach.
+
+.. warning::
+
+  You should **never modify** something you are iterating over.
+  This is not guaranteed to work in all cases. Depending on the
+  data types, the iterator returns a copy and not a view, and writing
+  to it will have no effect!
+
+  For example, in the following case setting the value has no effect:
+
+  .. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2, 3], "b": ["a", "b", "c"]})
+
+    for index, row in df.iterrows():
+        row["a"] = 10
+
+    df
+
+items
+~~~~~
+
+Consistent with the dict-like interface, :meth:`~DataFrame.items` iterates
+through key-value pairs:
+
+* **Series**: (index, scalar value) pairs
+* **DataFrame**: (column, Series) pairs
+
+For example:
+
+.. ipython:: python
+
+   for label, ser in df.items():
+       print(label)
+       print(ser)
+
+.. _basics.iterrows:
+
+iterrows
+~~~~~~~~
+
+:meth:`~DataFrame.iterrows` allows you to iterate through the rows of a
+DataFrame as Series objects. It returns an iterator yielding each
+index value along with a Series containing the data in each row:
+
+.. ipython:: python
+
+   for row_index, row in df.iterrows():
+       print(row_index, row, sep="\n")
+
+.. note::
+
+   Because :meth:`~DataFrame.iterrows` returns a Series for each row,
+   it does **not** preserve dtypes across the rows (dtypes are
+   preserved across columns for DataFrames). For example,
+
+   .. ipython:: python
+
+      df_orig = pd.DataFrame([[1, 1.5]], columns=["int", "float"])
+      df_orig.dtypes
+      row = next(df_orig.iterrows())[1]
+      row
+
+   All values in ``row``, returned as a Series, are now upcasted
+   to floats, also the original integer value in column ``x``:
+
+   .. ipython:: python
+
+      row["int"].dtype
+      df_orig["int"].dtype
+
+   To preserve dtypes while iterating over the rows, it is better
+   to use :meth:`~DataFrame.itertuples` which returns namedtuples of the values
+   and which is generally much faster than :meth:`~DataFrame.iterrows`.
+
+For instance, a contrived way to transpose the DataFrame would be:
+
+.. ipython:: python
+
+   df2 = pd.DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+   print(df2)
+   print(df2.T)
+
+   df2_t = pd.DataFrame({idx: values for idx, values in df2.iterrows()})
+   print(df2_t)
+
+itertuples
+~~~~~~~~~~
+
+The :meth:`~DataFrame.itertuples` method will return an iterator
+yielding a namedtuple for each row in the DataFrame. The first element
+of the tuple will be the row's corresponding index value, while the
+remaining values are the row values.
+
+For instance:
+
+.. ipython:: python
+
+   for row in df.itertuples():
+       print(row)
+
+This method does not convert the row to a Series object; it merely
+returns the values inside a namedtuple. Therefore,
+:meth:`~DataFrame.itertuples` preserves the data type of the values
+and is generally faster than :meth:`~DataFrame.iterrows`.
+
+.. note::
+
+   The column names will be renamed to positional names if they are
+   invalid Python identifiers, repeated, or start with an underscore.
+   With a large number of columns (>255), regular tuples are returned.
+
+.. _basics.dt_accessors:
+
+.dt accessor
+------------
+
+``Series`` has an accessor to succinctly return datetime like properties for the
+*values* of the Series, if it is a datetime/period like Series.
+This will return a Series, indexed like the existing Series.
+
+.. ipython:: python
+
+   # datetime
+   s = pd.Series(pd.date_range("20130101 09:10:12", periods=4))
+   s
+   s.dt.hour
+   s.dt.second
+   s.dt.day
+
+This enables nice expressions like this:
+
+.. ipython:: python
+
+   s[s.dt.day == 2]
+
+You can easily produces tz aware transformations:
+
+.. ipython:: python
+
+   stz = s.dt.tz_localize("US/Eastern")
+   stz
+   stz.dt.tz
+
+You can also chain these types of operations:
+
+.. ipython:: python
+
+   s.dt.tz_localize("UTC").dt.tz_convert("US/Eastern")
+
+You can also format datetime values as strings with :meth:`Series.dt.strftime` which
+supports the same format as the standard :meth:`~datetime.datetime.strftime`.
+
+.. ipython:: python
+
+   # DatetimeIndex
+   s = pd.Series(pd.date_range("20130101", periods=4))
+   s
+   s.dt.strftime("%Y/%m/%d")
+
+.. ipython:: python
+
+   # PeriodIndex
+   s = pd.Series(pd.period_range("20130101", periods=4))
+   s
+   s.dt.strftime("%Y/%m/%d")
+
+The ``.dt`` accessor works for period and timedelta dtypes.
+
+.. ipython:: python
+
+   # period
+   s = pd.Series(pd.period_range("20130101", periods=4, freq="D"))
+   s
+   s.dt.year
+   s.dt.day
+
+.. ipython:: python
+
+   # timedelta
+   s = pd.Series(pd.timedelta_range("1 day 00:00:05", periods=4, freq="s"))
+   s
+   s.dt.days
+   s.dt.seconds
+   s.dt.components
+
+.. note::
+
+   ``Series.dt`` will raise a ``TypeError`` if you access with a non-datetime-like values.
+
+Vectorized string methods
+-------------------------
+
+Series is equipped with a set of string processing methods that make it easy to
+operate on each element of the array. Perhaps most importantly, these methods
+exclude missing/NA values automatically. These are accessed via the Series's
+``str`` attribute and generally have names matching the equivalent (scalar)
+built-in string methods. For example:
+
+ .. ipython:: python
+
+  s = pd.Series(
+      ["A", "B", "C", "Aaba", "Baca", np.nan, "CABA", "dog", "cat"], dtype="string"
+  )
+  s.str.lower()
+
+Powerful pattern-matching methods are provided as well, but note that
+pattern-matching generally uses `regular expressions
+<https://docs.python.org/3/library/re.html>`__ by default (and in some cases
+always uses them).
+
+.. note::
+
+   Prior to pandas 1.0, string methods were only available on ``object`` -dtype
+   ``Series``. pandas 1.0 added the :class:`StringDtype` which is dedicated
+   to strings. See :ref:`text.types` for more.
+
+Please see :ref:`Vectorized String Methods <text.string_methods>` for a complete
+description.
+
+.. _basics.sorting:
+
+Sorting
+-------
+
+pandas supports three kinds of sorting: sorting by index labels,
+sorting by column values, and sorting by a combination of both.
+
+.. _basics.sort_index:
+
+By index
+~~~~~~~~
+
+The :meth:`Series.sort_index` and :meth:`DataFrame.sort_index` methods are
+used to sort a pandas object by its index levels.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "one": pd.Series(np.random.randn(3), index=["a", "b", "c"]),
+           "two": pd.Series(np.random.randn(4), index=["a", "b", "c", "d"]),
+           "three": pd.Series(np.random.randn(3), index=["b", "c", "d"]),
+       }
+   )
+
+   unsorted_df = df.reindex(
+       index=["a", "d", "c", "b"], columns=["three", "two", "one"]
+   )
+   unsorted_df
+
+   # DataFrame
+   unsorted_df.sort_index()
+   unsorted_df.sort_index(ascending=False)
+   unsorted_df.sort_index(axis=1)
+
+   # Series
+   unsorted_df["three"].sort_index()
+
+.. _basics.sort_index_key:
+
+Sorting by index also supports a ``key`` parameter that takes a callable
+function to apply to the index being sorted. For ``MultiIndex`` objects,
+the key is applied per-level to the levels specified by ``level``.
+
+.. ipython:: python
+
+   s1 = pd.DataFrame({"a": ["B", "a", "C"], "b": [1, 2, 3], "c": [2, 3, 4]}).set_index(
+       list("ab")
+   )
+   s1
+
+.. ipython:: python
+
+   s1.sort_index(level="a")
+   s1.sort_index(level="a", key=lambda idx: idx.str.lower())
+
+For information on key sorting by value, see :ref:`value sorting
+<basics.sort_value_key>`.
+
+.. _basics.sort_values:
+
+By values
+~~~~~~~~~
+
+The :meth:`Series.sort_values` method is used to sort a ``Series`` by its values. The
+:meth:`DataFrame.sort_values` method is used to sort a ``DataFrame`` by its column or row values.
+The optional ``by`` parameter to :meth:`DataFrame.sort_values` may used to specify one or more columns
+to use to determine the sorted order.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(
+       {"one": [2, 1, 1, 1], "two": [1, 3, 2, 4], "three": [5, 4, 3, 2]}
+   )
+   df1.sort_values(by="two")
+
+The ``by`` parameter can take a list of column names, e.g.:
+
+.. ipython:: python
+
+   df1[["one", "two", "three"]].sort_values(by=["one", "two"])
+
+These methods have special treatment of NA values via the ``na_position``
+argument:
+
+.. ipython:: python
+
+   s[2] = np.nan
+   s.sort_values()
+   s.sort_values(na_position="first")
+
+.. _basics.sort_value_key:
+
+Sorting also supports a ``key`` parameter that takes a callable function
+to apply to the values being sorted.
+
+.. ipython:: python
+
+   s1 = pd.Series(["B", "a", "C"])
+
+.. ipython:: python
+
+   s1.sort_values()
+   s1.sort_values(key=lambda x: x.str.lower())
+
+``key`` will be given the :class:`Series` of values and should return a ``Series``
+or array of the same shape with the transformed values. For ``DataFrame`` objects,
+the key is applied per column, so the key should still expect a Series and return
+a Series, e.g.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": ["B", "a", "C"], "b": [1, 2, 3]})
+
+.. ipython:: python
+
+   df.sort_values(by="a")
+   df.sort_values(by="a", key=lambda col: col.str.lower())
+
+The name or type of each column can be used to apply different functions to
+different columns.
+
+.. _basics.sort_indexes_and_values:
+
+By indexes and values
+~~~~~~~~~~~~~~~~~~~~~
+
+Strings passed as the ``by`` parameter to :meth:`DataFrame.sort_values` may
+refer to either columns or index level names.
+
+.. ipython:: python
+
+   # Build MultiIndex
+   idx = pd.MultiIndex.from_tuples(
+       [("a", 1), ("a", 2), ("a", 2), ("b", 2), ("b", 1), ("b", 1)]
+   )
+   idx.names = ["first", "second"]
+
+   # Build DataFrame
+   df_multi = pd.DataFrame({"A": np.arange(6, 0, -1)}, index=idx)
+   df_multi
+
+Sort by 'second' (index) and 'A' (column)
+
+.. ipython:: python
+
+   df_multi.sort_values(by=["second", "A"])
+
+.. note::
+
+   If a string matches both a column name and an index level name then a
+   warning is issued and the column takes precedence. This will result in an
+   ambiguity error in a future version.
+
+.. _basics.searchsorted:
+
+searchsorted
+~~~~~~~~~~~~
+
+Series has the :meth:`~Series.searchsorted` method, which works similarly to
+:meth:`numpy.ndarray.searchsorted`.
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3])
+   ser.searchsorted([0, 3])
+   ser.searchsorted([0, 4])
+   ser.searchsorted([1, 3], side="right")
+   ser.searchsorted([1, 3], side="left")
+   ser = pd.Series([3, 1, 2])
+   ser.searchsorted([0, 3], sorter=np.argsort(ser))
+
+.. _basics.nsorted:
+
+smallest / largest values
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``Series`` has the :meth:`~Series.nsmallest` and :meth:`~Series.nlargest` methods which return the
+smallest or largest :math:`n` values. For a large ``Series`` this can be much
+faster than sorting the entire Series and calling ``head(n)`` on the result.
+
+.. ipython:: python
+
+   s = pd.Series(np.random.permutation(10))
+   s
+   s.sort_values()
+   s.nsmallest(3)
+   s.nlargest(3)
+
+``DataFrame`` also has the ``nlargest`` and ``nsmallest`` methods.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "a": [-2, -1, 1, 10, 8, 11, -1],
+           "b": list("abdceff"),
+           "c": [1.0, 2.0, 4.0, 3.2, np.nan, 3.0, 4.0],
+       }
+   )
+   df.nlargest(3, "a")
+   df.nlargest(5, ["a", "c"])
+   df.nsmallest(3, "a")
+   df.nsmallest(5, ["a", "c"])
+
+
+.. _basics.multiindex_sorting:
+
+Sorting by a MultiIndex column
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You must be explicit about sorting when the column is a MultiIndex, and fully specify
+all levels to ``by``.
+
+.. ipython:: python
+
+   df1.columns = pd.MultiIndex.from_tuples(
+       [("a", "one"), ("a", "two"), ("b", "three")]
+   )
+   df1.sort_values(by=("a", "two"))
+
+
+Copying
+-------
+
+The :meth:`~DataFrame.copy` method on pandas objects copies the underlying data (though not
+the axis indexes, since they are immutable) and returns a new object. Note that
+**it is seldom necessary to copy objects**. For example, there are only a
+handful of ways to alter a DataFrame *in-place*:
+
+* Inserting, deleting, or modifying a column.
+* Assigning to the ``index`` or ``columns`` attributes.
+* For homogeneous data, directly modifying the values via the ``values``
+  attribute or advanced indexing.
+
+To be clear, no pandas method has the side effect of modifying your data;
+almost every method returns a new object, leaving the original object
+untouched. If the data is modified, it is because you did so explicitly.
+
+.. _basics.dtypes:
+
+dtypes
+------
+
+For the most part, pandas uses NumPy arrays and dtypes for Series or individual
+columns of a DataFrame. NumPy provides support for ``float``,
+``int``, ``bool``, ``timedelta64[ns]`` and ``datetime64[ns]`` (note that NumPy
+does not support timezone-aware datetimes).
+
+pandas and third-party libraries *extend* NumPy's type system in a few places.
+This section describes the extensions pandas has made internally.
+See :ref:`extending.extension-types` for how to write your own extension that
+works with pandas. See `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_ for a list of third-party
+libraries that have implemented an extension.
+
+The following table lists all of pandas extension types. For methods requiring ``dtype``
+arguments, strings can be specified as indicated. See the respective
+documentation sections for more on each type.
+
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| Kind of Data                                    | Data Type                 | Scalar             | Array                         | String Aliases                         |
++=================================================+===============+===========+========+===========+===============================+========================================+
+| :ref:`tz-aware datetime <timeseries.timezone>`  | :class:`DatetimeTZDtype`  | :class:`Timestamp` | :class:`arrays.DatetimeArray` | ``'datetime64[ns, <tz>]'``             |
+|                                                 |                           |                    |                               |                                        |
++-------------------------------------------------+---------------+-----------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`Categorical <categorical>`                | :class:`CategoricalDtype` | (none)             | :class:`Categorical`          | ``'category'``                         |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`period (time spans) <timeseries.periods>` | :class:`PeriodDtype`      | :class:`Period`    | :class:`arrays.PeriodArray`   | ``'period[<freq>]'``,                  |
+|                                                 |                           |                    | ``'Period[<freq>]'``          |                                        |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`sparse <sparse>`                          | :class:`SparseDtype`      | (none)             | :class:`arrays.SparseArray`   | ``'Sparse'``, ``'Sparse[int]'``,       |
+|                                                 |                           |                    |                               | ``'Sparse[float]'``                    |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`intervals <advanced.intervalindex>`       | :class:`IntervalDtype`    | :class:`Interval`  | :class:`arrays.IntervalArray` | ``'interval'``, ``'Interval'``,        |
+|                                                 |                           |                    |                               | ``'Interval[<numpy_dtype>]'``,         |
+|                                                 |                           |                    |                               | ``'Interval[datetime64[ns, <tz>]]'``,  |
+|                                                 |                           |                    |                               | ``'Interval[timedelta64[<freq>]]'``    |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`nullable integer <integer_na>`            | :class:`Int64Dtype`, ...  | (none)             | :class:`arrays.IntegerArray`  | ``'Int8'``, ``'Int16'``, ``'Int32'``,  |
+|                                                 |                           |                    |                               | ``'Int64'``, ``'UInt8'``, ``'UInt16'``,|
+|                                                 |                           |                    |                               | ``'UInt32'``, ``'UInt64'``             |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`nullable float <api.arrays.float_na>`     | :class:`Float64Dtype`, ...| (none)             | :class:`arrays.FloatingArray` | ``'Float32'``, ``'Float64'``           |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`Strings <text>`                           | :class:`StringDtype`      | :class:`str`       | :class:`arrays.StringArray`   | ``'string'``                           |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+| :ref:`Boolean (with NA) <api.arrays.bool>`      | :class:`BooleanDtype`     | :class:`bool`      | :class:`arrays.BooleanArray`  | ``'boolean'``                          |
++-------------------------------------------------+---------------------------+--------------------+-------------------------------+----------------------------------------+
+
+pandas has two ways to store strings.
+
+1. ``object`` dtype, which can hold any Python object, including strings.
+2. :class:`StringDtype`, which is dedicated to strings.
+
+Generally, we recommend using :class:`StringDtype`. See :ref:`text.types` for more.
+
+Finally, arbitrary objects may be stored using the ``object`` dtype, but should
+be avoided to the extent possible (for performance and interoperability with
+other libraries and methods. See :ref:`basics.object_conversion`).
+
+A convenient :attr:`~DataFrame.dtypes` attribute for DataFrame returns a Series
+with the data type of each column.
+
+.. ipython:: python
+
+   dft = pd.DataFrame(
+       {
+           "A": np.random.rand(3),
+           "B": 1,
+           "C": "foo",
+           "D": pd.Timestamp("20010102"),
+           "E": pd.Series([1.0] * 3).astype("float32"),
+           "F": False,
+           "G": pd.Series([1] * 3, dtype="int8"),
+       }
+   )
+   dft
+   dft.dtypes
+
+On a ``Series`` object, use the :attr:`~Series.dtype` attribute.
+
+.. ipython:: python
+
+   dft["A"].dtype
+
+If a pandas object contains data with multiple dtypes *in a single column*, the
+dtype of the column will be chosen to accommodate all of the data types
+(``object`` is the most general).
+
+.. ipython:: python
+
+   # these ints are coerced to floats
+   pd.Series([1, 2, 3, 4, 5, 6.0])
+
+   # string data forces an ``object`` dtype
+   pd.Series([1, 2, 3, 6.0, "foo"])
+
+The number of columns of each type in a ``DataFrame`` can be found by calling
+``DataFrame.dtypes.value_counts()``.
+
+.. ipython:: python
+
+   dft.dtypes.value_counts()
+
+Numeric dtypes will propagate and can coexist in DataFrames.
+If a dtype is passed (either directly via the ``dtype`` keyword, a passed ``ndarray``,
+or a passed ``Series``), then it will be preserved in DataFrame operations. Furthermore,
+different numeric dtypes will **NOT** be combined. The following example will give you a taste.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.random.randn(8, 1), columns=["A"], dtype="float32")
+   df1
+   df1.dtypes
+   df2 = pd.DataFrame(
+       {
+           "A": pd.Series(np.random.randn(8), dtype="float16"),
+           "B": pd.Series(np.random.randn(8)),
+           "C": pd.Series(np.random.randint(0, 255, size=8), dtype="uint8"),  # [0,255] (range of uint8)
+       }
+   )
+   df2
+   df2.dtypes
+
+defaults
+~~~~~~~~
+
+By default integer types are ``int64`` and float types are ``float64``,
+*regardless* of platform (32-bit or 64-bit).
+The following will all result in ``int64`` dtypes.
+
+.. ipython:: python
+
+   pd.DataFrame([1, 2], columns=["a"]).dtypes
+   pd.DataFrame({"a": [1, 2]}).dtypes
+   pd.DataFrame({"a": 1}, index=list(range(2))).dtypes
+
+Note that Numpy will choose *platform-dependent* types when creating arrays.
+The following **WILL** result in ``int32`` on 32-bit platform.
+
+.. ipython:: python
+
+   frame = pd.DataFrame(np.array([1, 2]))
+
+
+upcasting
+~~~~~~~~~
+
+Types can potentially be *upcasted* when combined with other types, meaning they are promoted
+from the current type (e.g. ``int`` to ``float``).
+
+.. ipython:: python
+
+   df3 = df1.reindex_like(df2).fillna(value=0.0) + df2
+   df3
+   df3.dtypes
+
+:meth:`DataFrame.to_numpy` will return the *lower-common-denominator* of the dtypes, meaning
+the dtype that can accommodate **ALL** of the types in the resulting homogeneous dtyped NumPy array. This can
+force some *upcasting*.
+
+.. ipython:: python
+
+   df3.to_numpy().dtype
+
+astype
+~~~~~~
+
+.. _basics.cast:
+
+You can use the :meth:`~DataFrame.astype` method to explicitly convert dtypes from one to another. These will by default return a copy,
+even if the dtype was unchanged (pass ``copy=False`` to change this behavior). In addition, they will raise an
+exception if the astype operation is invalid.
+
+Upcasting is always according to the **NumPy** rules. If two different dtypes are involved in an operation,
+then the more *general* one will be used as the result of the operation.
+
+.. ipython:: python
+
+   df3
+   df3.dtypes
+
+   # conversion of dtypes
+   df3.astype("float32").dtypes
+
+
+Convert a subset of columns to a specified type using :meth:`~DataFrame.astype`.
+
+.. ipython:: python
+
+   dft = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+   dft[["a", "b"]] = dft[["a", "b"]].astype(np.uint8)
+   dft
+   dft.dtypes
+
+Convert certain columns to a specific dtype by passing a dict to :meth:`~DataFrame.astype`.
+
+.. ipython:: python
+
+   dft1 = pd.DataFrame({"a": [1, 0, 1], "b": [4, 5, 6], "c": [7, 8, 9]})
+   dft1 = dft1.astype({"a": np.bool_, "c": np.float64})
+   dft1
+   dft1.dtypes
+
+.. note::
+
+    When trying to convert a subset of columns to a specified type using :meth:`~DataFrame.astype`  and :meth:`~DataFrame.loc`, upcasting occurs.
+
+    :meth:`~DataFrame.loc` tries to fit in what we are assigning to the current dtypes, while ``[]`` will overwrite them taking the dtype from the right hand side. Therefore the following piece of code produces the unintended result.
+
+    .. ipython:: python
+
+       dft = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+       dft.loc[:, ["a", "b"]].astype(np.uint8).dtypes
+       dft.loc[:, ["a", "b"]] = dft.loc[:, ["a", "b"]].astype(np.uint8)
+       dft.dtypes
+
+.. _basics.object_conversion:
+
+object conversion
+~~~~~~~~~~~~~~~~~
+
+pandas offers various functions to try to force conversion of types from the ``object`` dtype to other types.
+In cases where the data is already of the correct type, but stored in an ``object`` array, the
+:meth:`DataFrame.infer_objects` and :meth:`Series.infer_objects` methods can be used to soft convert
+to the correct type.
+
+  .. ipython:: python
+
+     import datetime
+
+     df = pd.DataFrame(
+         [
+             [1, 2],
+             ["a", "b"],
+             [datetime.datetime(2016, 3, 2), datetime.datetime(2016, 3, 2)],
+         ]
+     )
+     df = df.T
+     df
+     df.dtypes
+
+Because the data was transposed the original inference stored all columns as object, which
+``infer_objects`` will correct.
+
+  .. ipython:: python
+
+     df.infer_objects().dtypes
+
+The following functions are available for one dimensional object arrays or scalars to perform
+hard conversion of objects to a specified type:
+
+* :meth:`~pandas.to_numeric` (conversion to numeric dtypes)
+
+  .. ipython:: python
+
+     m = ["1.1", 2, 3]
+     pd.to_numeric(m)
+
+* :meth:`~pandas.to_datetime` (conversion to datetime objects)
+
+  .. ipython:: python
+
+     import datetime
+
+     m = ["2016-07-09", datetime.datetime(2016, 3, 2)]
+     pd.to_datetime(m)
+
+* :meth:`~pandas.to_timedelta` (conversion to timedelta objects)
+
+  .. ipython:: python
+
+     m = ["5us", pd.Timedelta("1day")]
+     pd.to_timedelta(m)
+
+To force a conversion, we can pass in an ``errors`` argument, which specifies how pandas should deal with elements
+that cannot be converted to desired dtype or object. By default, ``errors='raise'``, meaning that any errors encountered
+will be raised during the conversion process. However, if ``errors='coerce'``, these errors will be ignored and pandas
+will convert problematic elements to ``pd.NaT`` (for datetime and timedelta) or ``np.nan`` (for numeric). This might be
+useful if you are reading in data which is mostly of the desired dtype (e.g. numeric, datetime), but occasionally has
+non-conforming elements intermixed that you want to represent as missing:
+
+.. ipython:: python
+   :okwarning:
+
+    import datetime
+
+    m = ["apple", datetime.datetime(2016, 3, 2)]
+    pd.to_datetime(m, errors="coerce")
+
+    m = ["apple", 2, 3]
+    pd.to_numeric(m, errors="coerce")
+
+    m = ["apple", pd.Timedelta("1day")]
+    pd.to_timedelta(m, errors="coerce")
+
+In addition to object conversion, :meth:`~pandas.to_numeric` provides another argument ``downcast``, which gives the
+option of downcasting the newly (or already) numeric data to a smaller dtype, which can conserve memory:
+
+.. ipython:: python
+
+    m = ["1", 2, 3]
+    pd.to_numeric(m, downcast="integer")  # smallest signed int dtype
+    pd.to_numeric(m, downcast="signed")  # same as 'integer'
+    pd.to_numeric(m, downcast="unsigned")  # smallest unsigned int dtype
+    pd.to_numeric(m, downcast="float")  # smallest float dtype
+
+As these methods apply only to one-dimensional arrays, lists or scalars; they cannot be used directly on multi-dimensional objects such
+as DataFrames. However, with :meth:`~pandas.DataFrame.apply`, we can "apply" the function over each column efficiently:
+
+.. ipython:: python
+
+    import datetime
+
+    df = pd.DataFrame([["2016-07-09", datetime.datetime(2016, 3, 2)]] * 2, dtype="O")
+    df
+    df.apply(pd.to_datetime)
+
+    df = pd.DataFrame([["1.1", 2, 3]] * 2, dtype="O")
+    df
+    df.apply(pd.to_numeric)
+
+    df = pd.DataFrame([["5us", pd.Timedelta("1day")]] * 2, dtype="O")
+    df
+    df.apply(pd.to_timedelta)
+
+gotchas
+~~~~~~~
+
+Performing selection operations on ``integer`` type data can easily upcast the data to ``floating``.
+The dtype of the input data will be preserved in cases where ``nans`` are not introduced.
+See also :ref:`Support for integer NA <gotchas.intna>`.
+
+.. ipython:: python
+
+   dfi = df3.astype("int32")
+   dfi["E"] = 1
+   dfi
+   dfi.dtypes
+
+   casted = dfi[dfi > 0]
+   casted
+   casted.dtypes
+
+While float dtypes are unchanged.
+
+.. ipython:: python
+
+   dfa = df3.copy()
+   dfa["A"] = dfa["A"].astype("float32")
+   dfa.dtypes
+
+   casted = dfa[df2 > 0]
+   casted
+   casted.dtypes
+
+Selecting columns based on ``dtype``
+------------------------------------
+
+.. _basics.selectdtypes:
+
+The :meth:`~DataFrame.select_dtypes` method implements subsetting of columns
+based on their ``dtype``.
+
+First, let's create a :class:`DataFrame` with a slew of different
+dtypes:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "string": list("abc"),
+           "int64": list(range(1, 4)),
+           "uint8": np.arange(3, 6).astype("u1"),
+           "float64": np.arange(4.0, 7.0),
+           "bool1": [True, False, True],
+           "bool2": [False, True, False],
+           "dates": pd.date_range("now", periods=3),
+           "category": pd.Series(list("ABC")).astype("category"),
+       }
+   )
+   df["tdeltas"] = df.dates.diff()
+   df["uint64"] = np.arange(3, 6).astype("u8")
+   df["other_dates"] = pd.date_range("20130101", periods=3)
+   df["tz_aware_dates"] = pd.date_range("20130101", periods=3, tz="US/Eastern")
+   df
+
+And the dtypes:
+
+.. ipython:: python
+
+   df.dtypes
+
+:meth:`~DataFrame.select_dtypes` has two parameters ``include`` and ``exclude`` that allow you to
+say "give me the columns *with* these dtypes" (``include``) and/or "give the
+columns *without* these dtypes" (``exclude``).
+
+For example, to select ``bool`` columns:
+
+.. ipython:: python
+
+   df.select_dtypes(include=[bool])
+
+You can also pass the name of a dtype in the `NumPy dtype hierarchy
+<https://numpy.org/doc/stable/reference/arrays.scalars.html>`__:
+
+.. ipython:: python
+
+   df.select_dtypes(include=["bool"])
+
+:meth:`~pandas.DataFrame.select_dtypes` also works with generic dtypes as well.
+
+For example, to select all numeric and boolean columns while excluding unsigned
+integers:
+
+.. ipython:: python
+
+   df.select_dtypes(include=["number", "bool"], exclude=["unsignedinteger"])
+
+To select string columns you must use the ``object`` dtype:
+
+.. ipython:: python
+
+   df.select_dtypes(include=["object"])
+
+To see all the child dtypes of a generic ``dtype`` like ``numpy.number`` you
+can define a function that returns a tree of child dtypes:
+
+.. ipython:: python
+
+   def subdtypes(dtype):
+       subs = dtype.__subclasses__()
+       if not subs:
+           return dtype
+       return [dtype, [subdtypes(dt) for dt in subs]]
+
+All NumPy dtypes are subclasses of ``numpy.generic``:
+
+.. ipython:: python
+
+    subdtypes(np.generic)
+
+.. note::
+
+    pandas also defines the types ``category``, and ``datetime64[ns, tz]``, which are not integrated into the normal
+    NumPy hierarchy and won't show up with the above function.
diff --git a/pandas/doc/source/user_guide/boolean.rst b/pandas/doc/source/user_guide/boolean.rst
new file mode 100644
index 0000000000000..7de0430123fd2
--- /dev/null
+++ b/pandas/doc/source/user_guide/boolean.rst
@@ -0,0 +1,115 @@
+.. currentmodule:: pandas
+
+.. ipython:: python
+   :suppress:
+
+   import pandas as pd
+   import numpy as np
+
+.. _boolean:
+
+**************************
+Nullable Boolean data type
+**************************
+
+.. note::
+
+   BooleanArray is currently experimental. Its API or implementation may
+   change without warning.
+
+.. _boolean.indexing:
+
+Indexing with NA values
+-----------------------
+
+pandas allows indexing with ``NA`` values in a boolean array, which are treated as ``False``.
+
+.. ipython:: python
+   :okexcept:
+
+   s = pd.Series([1, 2, 3])
+   mask = pd.array([True, False, pd.NA], dtype="boolean")
+   s[mask]
+
+If you would prefer to keep the ``NA`` values you can manually fill them with ``fillna(True)``.
+
+.. ipython:: python
+
+   s[mask.fillna(True)]
+
+If you create a column of ``NA`` values (for example to fill them later)
+with ``df['new_col'] = pd.NA``, the ``dtype`` would be set to ``object`` in the
+new column. The performance on this column will be worse than with
+the appropriate type. It's better to use
+``df['new_col'] = pd.Series(pd.NA, dtype="boolean")``
+(or another ``dtype`` that supports ``NA``).
+
+.. ipython:: python
+
+   df = pd.DataFrame()
+   df['objects'] = pd.NA
+   df.dtypes
+
+.. _boolean.kleene:
+
+Kleene logical operations
+-------------------------
+
+:class:`arrays.BooleanArray` implements `Kleene Logic`_ (sometimes called three-value logic) for
+logical operations like ``&`` (and), ``|`` (or) and ``^`` (exclusive-or).
+
+This table demonstrates the results for every combination. These operations are symmetrical,
+so flipping the left- and right-hand side makes no difference in the result.
+
+================= =========
+Expression        Result
+================= =========
+``True & True``   ``True``
+``True & False``  ``False``
+``True & NA``     ``NA``
+``False & False`` ``False``
+``False & NA``    ``False``
+``NA & NA``       ``NA``
+``True | True``   ``True``
+``True | False``  ``True``
+``True | NA``     ``True``
+``False | False`` ``False``
+``False | NA``    ``NA``
+``NA | NA``       ``NA``
+``True ^ True``   ``False``
+``True ^ False``  ``True``
+``True ^ NA``     ``NA``
+``False ^ False`` ``False``
+``False ^ NA``    ``NA``
+``NA ^ NA``       ``NA``
+================= =========
+
+When an ``NA`` is present in an operation, the output value is ``NA`` only if
+the result cannot be determined solely based on the other input. For example,
+``True | NA`` is ``True``, because both ``True | True`` and ``True | False``
+are ``True``. In that case, we don't actually need to consider the value
+of the ``NA``.
+
+On the other hand, ``True & NA`` is ``NA``. The result depends on whether
+the ``NA`` really is ``True`` or ``False``, since ``True & True`` is ``True``,
+but ``True & False`` is ``False``, so we can't determine the output.
+
+
+This differs from how ``np.nan`` behaves in logical operations. pandas treated
+``np.nan`` is *always false in the output*.
+
+In ``or``
+
+.. ipython:: python
+
+   pd.Series([True, False, np.nan], dtype="object") | True
+   pd.Series([True, False, np.nan], dtype="boolean") | True
+
+In ``and``
+
+.. ipython:: python
+
+   pd.Series([True, False, np.nan], dtype="object") & True
+   pd.Series([True, False, np.nan], dtype="boolean") & True
+
+.. _Kleene Logic: https://en.wikipedia.org/wiki/Three-valued_logic#Kleene_and_Priest_logics
diff --git a/pandas/doc/source/user_guide/categorical.rst b/pandas/doc/source/user_guide/categorical.rst
new file mode 100644
index 0000000000000..1e7d66dfeb142
--- /dev/null
+++ b/pandas/doc/source/user_guide/categorical.rst
@@ -0,0 +1,1180 @@
+.. _categorical:
+
+{{ header }}
+
+****************
+Categorical data
+****************
+
+This is an introduction to pandas categorical data type, including a short comparison
+with R's ``factor``.
+
+``Categoricals`` are a pandas data type corresponding to categorical variables in
+statistics. A categorical variable takes on a limited, and usually fixed,
+number of possible values (``categories``; ``levels`` in R). Examples are gender,
+social class, blood type, country affiliation, observation time or rating via
+Likert scales.
+
+In contrast to statistical categorical variables, categorical data might have an order (e.g.
+'strongly agree' vs 'agree' or 'first observation' vs. 'second observation'), but numerical
+operations (additions, divisions, ...) are not possible.
+
+All values of categorical data are either in ``categories`` or ``np.nan``. Order is defined by
+the order of ``categories``, not lexical order of the values. Internally, the data structure
+consists of a ``categories`` array and an integer array of ``codes`` which point to the real value in
+the ``categories`` array.
+
+The categorical data type is useful in the following cases:
+
+* A string variable consisting of only a few different values. Converting such a string
+  variable to a categorical variable will save some memory, see :ref:`here <categorical.memory>`.
+* The lexical order of a variable is not the same as the logical order ("one", "two", "three").
+  By converting to a categorical and specifying an order on the categories, sorting and
+  min/max will use the logical order instead of the lexical order, see :ref:`here <categorical.sort>`.
+* As a signal to other Python libraries that this column should be treated as a categorical
+  variable (e.g. to use suitable statistical methods or plot types).
+
+See also the :ref:`API docs on categoricals<api.arrays.categorical>`.
+
+.. _categorical.objectcreation:
+
+Object creation
+---------------
+
+Series creation
+~~~~~~~~~~~~~~~
+
+Categorical ``Series`` or columns in a ``DataFrame`` can be created in several ways:
+
+By specifying ``dtype="category"`` when constructing a ``Series``:
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "a"], dtype="category")
+    s
+
+By converting an existing ``Series`` or column to a ``category`` dtype:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": ["a", "b", "c", "a"]})
+    df["B"] = df["A"].astype("category")
+    df
+
+By using special functions, such as :func:`~pandas.cut`, which groups data into
+discrete bins. See the :ref:`example on tiling <reshaping.tile.cut>` in the docs.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"value": np.random.randint(0, 100, 20)})
+    labels = ["{0} - {1}".format(i, i + 9) for i in range(0, 100, 10)]
+
+    df["group"] = pd.cut(df.value, range(0, 105, 10), right=False, labels=labels)
+    df.head(10)
+
+By passing a :class:`pandas.Categorical` object to a ``Series`` or assigning it to a ``DataFrame``.
+
+.. ipython:: python
+
+    raw_cat = pd.Categorical(
+        ["a", "b", "c", "a"], categories=["b", "c", "d"], ordered=False
+    )
+    s = pd.Series(raw_cat)
+    s
+    df = pd.DataFrame({"A": ["a", "b", "c", "a"]})
+    df["B"] = raw_cat
+    df
+
+Categorical data has a specific ``category`` :ref:`dtype <basics.dtypes>`:
+
+.. ipython:: python
+
+    df.dtypes
+
+DataFrame creation
+~~~~~~~~~~~~~~~~~~
+
+Similar to the previous section where a single column was converted to categorical, all columns in a
+``DataFrame`` can be batch converted to categorical either during or after construction.
+
+This can be done during construction by specifying ``dtype="category"`` in the ``DataFrame`` constructor:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": list("abca"), "B": list("bccd")}, dtype="category")
+    df.dtypes
+
+Note that the categories present in each column differ; the conversion is done column by column, so
+only labels present in a given column are categories:
+
+.. ipython:: python
+
+    df["A"]
+    df["B"]
+
+
+Analogously, all columns in an existing ``DataFrame`` can be batch converted using :meth:`DataFrame.astype`:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": list("abca"), "B": list("bccd")})
+    df_cat = df.astype("category")
+    df_cat.dtypes
+
+This conversion is likewise done column by column:
+
+.. ipython:: python
+
+    df_cat["A"]
+    df_cat["B"]
+
+
+Controlling behavior
+~~~~~~~~~~~~~~~~~~~~
+
+In the examples above where we passed ``dtype='category'``, we used the default
+behavior:
+
+1. Categories are inferred from the data.
+2. Categories are unordered.
+
+To control those behaviors, instead of passing ``'category'``, use an instance
+of :class:`~pandas.api.types.CategoricalDtype`.
+
+.. ipython:: python
+
+    from pandas.api.types import CategoricalDtype
+
+    s = pd.Series(["a", "b", "c", "a"])
+    cat_type = CategoricalDtype(categories=["b", "c", "d"], ordered=True)
+    s_cat = s.astype(cat_type)
+    s_cat
+
+Similarly, a ``CategoricalDtype`` can be used with a ``DataFrame`` to ensure that categories
+are consistent among all columns.
+
+.. ipython:: python
+
+    from pandas.api.types import CategoricalDtype
+
+    df = pd.DataFrame({"A": list("abca"), "B": list("bccd")})
+    cat_type = CategoricalDtype(categories=list("abcd"), ordered=True)
+    df_cat = df.astype(cat_type)
+    df_cat["A"]
+    df_cat["B"]
+
+.. note::
+
+    To perform table-wise conversion, where all labels in the entire ``DataFrame`` are used as
+    categories for each column, the ``categories`` parameter can be determined programmatically by
+    ``categories = pd.unique(df.to_numpy().ravel())``.
+
+If you already have ``codes`` and ``categories``, you can use the
+:func:`~pandas.Categorical.from_codes` constructor to save the factorize step
+during normal constructor mode:
+
+.. ipython:: python
+
+    splitter = np.random.choice([0, 1], 5, p=[0.5, 0.5])
+    s = pd.Series(pd.Categorical.from_codes(splitter, categories=["train", "test"]))
+
+
+Regaining original data
+~~~~~~~~~~~~~~~~~~~~~~~
+
+To get back to the original ``Series`` or NumPy array, use
+``Series.astype(original_dtype)`` or ``np.asarray(categorical)``:
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "a"])
+    s
+    s2 = s.astype("category")
+    s2
+    s2.astype(str)
+    np.asarray(s2)
+
+.. note::
+
+    In contrast to R's ``factor`` function, categorical data is not converting input values to
+    strings; categories will end up the same data type as the original values.
+
+.. note::
+
+    In contrast to R's ``factor`` function, there is currently no way to assign/change labels at
+    creation time. Use ``categories`` to change the categories after creation time.
+
+.. _categorical.categoricaldtype:
+
+CategoricalDtype
+----------------
+
+A categorical's type is fully described by
+
+1. ``categories``: a sequence of unique values and no missing values
+2. ``ordered``: a boolean
+
+This information can be stored in a :class:`~pandas.api.types.CategoricalDtype`.
+The ``categories`` argument is optional, which implies that the actual categories
+should be inferred from whatever is present in the data when the
+:class:`pandas.Categorical` is created. The categories are assumed to be unordered
+by default.
+
+.. ipython:: python
+
+   from pandas.api.types import CategoricalDtype
+
+   CategoricalDtype(["a", "b", "c"])
+   CategoricalDtype(["a", "b", "c"], ordered=True)
+   CategoricalDtype()
+
+A :class:`~pandas.api.types.CategoricalDtype` can be used in any place pandas
+expects a ``dtype``. For example :func:`pandas.read_csv`,
+:func:`pandas.DataFrame.astype`, or in the ``Series`` constructor.
+
+.. note::
+
+    As a convenience, you can use the string ``'category'`` in place of a
+    :class:`~pandas.api.types.CategoricalDtype` when you want the default behavior of
+    the categories being unordered, and equal to the set values present in the
+    array. In other words, ``dtype='category'`` is equivalent to
+    ``dtype=CategoricalDtype()``.
+
+Equality semantics
+~~~~~~~~~~~~~~~~~~
+
+Two instances of :class:`~pandas.api.types.CategoricalDtype` compare equal
+whenever they have the same categories and order. When comparing two
+unordered categoricals, the order of the ``categories`` is not considered. Note
+that categories with different dtypes are not the same.
+
+.. ipython:: python
+
+   c1 = CategoricalDtype(["a", "b", "c"], ordered=False)
+
+   # Equal, since order is not considered when ordered=False
+   c1 == CategoricalDtype(["b", "c", "a"], ordered=False)
+
+   # Unequal, since the second CategoricalDtype is ordered
+   c1 == CategoricalDtype(["a", "b", "c"], ordered=True)
+
+All instances of ``CategoricalDtype`` compare equal to the string ``'category'``.
+
+.. ipython:: python
+
+   c1 == "category"
+
+Notice that the ``categories_dtype`` should be considered, especially when comparing with
+two empty ``CategoricalDtype`` instances.
+
+.. ipython:: python
+
+    c2 = pd.Categorical(np.array([], dtype=object))
+    c3 = pd.Categorical(np.array([], dtype=float))
+
+    c2.dtype == c3.dtype
+
+Description
+-----------
+
+Using :meth:`~DataFrame.describe` on categorical data will produce similar
+output to a ``Series`` or ``DataFrame`` of type ``string``.
+
+.. ipython:: python
+
+    cat = pd.Categorical(["a", "c", "c", np.nan], categories=["b", "a", "c"])
+    df = pd.DataFrame({"cat": cat, "s": ["a", "c", "c", np.nan]})
+    df.describe()
+    df["cat"].describe()
+
+.. _categorical.cat:
+
+Working with categories
+-----------------------
+
+Categorical data has a ``categories`` and a ``ordered`` property, which list their
+possible values and whether the ordering matters or not. These properties are
+exposed as ``s.cat.categories`` and ``s.cat.ordered``. If you don't manually
+specify categories and ordering, they are inferred from the passed arguments.
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "a"], dtype="category")
+    s.cat.categories
+    s.cat.ordered
+
+It's also possible to pass in the categories in a specific order:
+
+.. ipython:: python
+
+    s = pd.Series(pd.Categorical(["a", "b", "c", "a"], categories=["c", "b", "a"]))
+    s.cat.categories
+    s.cat.ordered
+
+.. note::
+
+    New categorical data are **not** automatically ordered. You must explicitly
+    pass ``ordered=True`` to indicate an ordered ``Categorical``.
+
+
+.. note::
+
+    The result of :meth:`~Series.unique` is not always the same as ``Series.cat.categories``,
+    because ``Series.unique()`` has a couple of guarantees, namely that it returns categories
+    in the order of appearance, and it only includes values that are actually present.
+
+    .. ipython:: python
+
+         s = pd.Series(list("babc")).astype(CategoricalDtype(list("abcd")))
+         s
+
+         # categories
+         s.cat.categories
+
+         # uniques
+         s.unique()
+
+Renaming categories
+~~~~~~~~~~~~~~~~~~~
+
+Renaming categories is done by using the
+:meth:`~pandas.Categorical.rename_categories` method:
+
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "a"], dtype="category")
+    s
+    new_categories = ["Group %s" % g for g in s.cat.categories]
+    s = s.cat.rename_categories(new_categories)
+    s
+    # You can also pass a dict-like object to map the renaming
+    s = s.cat.rename_categories({1: "x", 2: "y", 3: "z"})
+    s
+
+.. note::
+
+    In contrast to R's ``factor``, categorical data can have categories of other types than string.
+
+Categories must be unique or a ``ValueError`` is raised:
+
+.. ipython:: python
+
+    try:
+        s = s.cat.rename_categories([1, 1, 1])
+    except ValueError as e:
+        print("ValueError:", str(e))
+
+Categories must also not be ``NaN`` or a ``ValueError`` is raised:
+
+.. ipython:: python
+
+    try:
+        s = s.cat.rename_categories([1, 2, np.nan])
+    except ValueError as e:
+        print("ValueError:", str(e))
+
+Appending new categories
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Appending categories can be done by using the
+:meth:`~pandas.Categorical.add_categories` method:
+
+.. ipython:: python
+
+    s = s.cat.add_categories([4])
+    s.cat.categories
+    s
+
+Removing categories
+~~~~~~~~~~~~~~~~~~~
+
+Removing categories can be done by using the
+:meth:`~pandas.Categorical.remove_categories` method. Values which are removed
+are replaced by ``np.nan``.:
+
+.. ipython:: python
+
+    s = s.cat.remove_categories([4])
+    s
+
+Removing unused categories
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Removing unused categories can also be done:
+
+.. ipython:: python
+
+    s = pd.Series(pd.Categorical(["a", "b", "a"], categories=["a", "b", "c", "d"]))
+    s
+    s.cat.remove_unused_categories()
+
+Setting categories
+~~~~~~~~~~~~~~~~~~
+
+If you want to do remove and add new categories in one step (which has some
+speed advantage), or simply set the categories to a predefined scale,
+use :meth:`~pandas.Categorical.set_categories`.
+
+
+.. ipython:: python
+
+    s = pd.Series(["one", "two", "four", "-"], dtype="category")
+    s
+    s = s.cat.set_categories(["one", "two", "three", "four"])
+    s
+
+.. note::
+    Be aware that :func:`Categorical.set_categories` cannot know whether some category is omitted
+    intentionally or because it is misspelled or (under Python3) due to a type difference (e.g.,
+    NumPy S1 dtype and Python strings). This can result in surprising behaviour!
+
+Sorting and order
+-----------------
+
+.. _categorical.sort:
+
+If categorical data is ordered (``s.cat.ordered == True``), then the order of the categories has a
+meaning and certain operations are possible. If the categorical is unordered, ``.min()/.max()`` will raise a ``TypeError``.
+
+.. ipython:: python
+
+    s = pd.Series(pd.Categorical(["a", "b", "c", "a"], ordered=False))
+    s = s.sort_values()
+    s = pd.Series(["a", "b", "c", "a"]).astype(CategoricalDtype(ordered=True))
+    s = s.sort_values()
+    s
+    s.min(), s.max()
+
+You can set categorical data to be ordered by using ``as_ordered()`` or unordered by using ``as_unordered()``. These will by
+default return a *new* object.
+
+.. ipython:: python
+
+    s.cat.as_ordered()
+    s.cat.as_unordered()
+
+Sorting will use the order defined by categories, not any lexical order present on the data type.
+This is even true for strings and numeric data:
+
+.. ipython:: python
+
+    s = pd.Series([1, 2, 3, 1], dtype="category")
+    s = s.cat.set_categories([2, 3, 1], ordered=True)
+    s
+    s = s.sort_values()
+    s
+    s.min(), s.max()
+
+
+Reordering
+~~~~~~~~~~
+
+Reordering the categories is possible via the :meth:`Categorical.reorder_categories` and
+the :meth:`Categorical.set_categories` methods. For :meth:`Categorical.reorder_categories`, all
+old categories must be included in the new categories and no new categories are allowed. This will
+necessarily make the sort order the same as the categories order.
+
+.. ipython:: python
+
+    s = pd.Series([1, 2, 3, 1], dtype="category")
+    s = s.cat.reorder_categories([2, 3, 1], ordered=True)
+    s
+    s = s.sort_values()
+    s
+    s.min(), s.max()
+
+.. note::
+
+    Note the difference between assigning new categories and reordering the categories: the first
+    renames categories and therefore the individual values in the ``Series``, but if the first
+    position was sorted last, the renamed value will still be sorted last. Reordering means that the
+    way values are sorted is different afterwards, but not that individual values in the
+    ``Series`` are changed.
+
+.. note::
+
+    If the ``Categorical`` is not ordered, :meth:`Series.min` and :meth:`Series.max` will raise
+    ``TypeError``. Numeric operations like ``+``, ``-``, ``*``, ``/`` and operations based on them
+    (e.g. :meth:`Series.median`, which would need to compute the mean between two values if the length
+    of an array is even) do not work and raise a ``TypeError``.
+
+Multi column sorting
+~~~~~~~~~~~~~~~~~~~~
+
+A categorical dtyped column will participate in a multi-column sort in a similar manner to other columns.
+The ordering of the categorical is determined by the ``categories`` of that column.
+
+.. ipython:: python
+
+   dfs = pd.DataFrame(
+       {
+           "A": pd.Categorical(
+               list("bbeebbaa"),
+               categories=["e", "a", "b"],
+               ordered=True,
+           ),
+           "B": [1, 2, 1, 2, 2, 1, 2, 1],
+       }
+   )
+   dfs.sort_values(by=["A", "B"])
+
+Reordering the ``categories`` changes a future sort.
+
+.. ipython:: python
+
+   dfs["A"] = dfs["A"].cat.reorder_categories(["a", "b", "e"])
+   dfs.sort_values(by=["A", "B"])
+
+Comparisons
+-----------
+
+Comparing categorical data with other objects is possible in three cases:
+
+* Comparing equality (``==`` and ``!=``) to a list-like object (list, Series, array,
+  ...) of the same length as the categorical data.
+* All comparisons (``==``, ``!=``, ``>``, ``>=``, ``<``, and ``<=``) of categorical data to
+  another categorical Series, when ``ordered==True`` and the ``categories`` are the same.
+* All comparisons of a categorical data to a scalar.
+
+All other comparisons, especially "non-equality" comparisons of two categoricals with different
+categories or a categorical with any list-like object, will raise a ``TypeError``.
+
+.. note::
+
+    Any "non-equality" comparisons of categorical data with a ``Series``, ``np.array``, ``list`` or
+    categorical data with different categories or ordering will raise a ``TypeError`` because custom
+    categories ordering could be interpreted in two ways: one with taking into account the
+    ordering and one without.
+
+.. ipython:: python
+
+    cat = pd.Series([1, 2, 3]).astype(CategoricalDtype([3, 2, 1], ordered=True))
+    cat_base = pd.Series([2, 2, 2]).astype(CategoricalDtype([3, 2, 1], ordered=True))
+    cat_base2 = pd.Series([2, 2, 2]).astype(CategoricalDtype(ordered=True))
+
+    cat
+    cat_base
+    cat_base2
+
+Comparing to a categorical with the same categories and ordering or to a scalar works:
+
+.. ipython:: python
+
+    cat > cat_base
+    cat > 2
+
+Equality comparisons work with any list-like object of same length and scalars:
+
+.. ipython:: python
+
+    cat == cat_base
+    cat == np.array([1, 2, 3])
+    cat == 2
+
+This doesn't work because the categories are not the same:
+
+.. ipython:: python
+
+    try:
+        cat > cat_base2
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+If you want to do a "non-equality" comparison of a categorical series with a list-like object
+which is not categorical data, you need to be explicit and convert the categorical data back to
+the original values:
+
+.. ipython:: python
+
+    base = np.array([1, 2, 3])
+
+    try:
+        cat > base
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+    np.asarray(cat) > base
+
+When you compare two unordered categoricals with the same categories, the order is not considered:
+
+.. ipython:: python
+
+   c1 = pd.Categorical(["a", "b"], categories=["a", "b"], ordered=False)
+   c2 = pd.Categorical(["a", "b"], categories=["b", "a"], ordered=False)
+   c1 == c2
+
+Operations
+----------
+
+Apart from :meth:`Series.min`, :meth:`Series.max` and :meth:`Series.mode`, the
+following operations are possible with categorical data:
+
+``Series`` methods like :meth:`Series.value_counts` will use all categories,
+even if some categories are not present in the data:
+
+.. ipython:: python
+
+    s = pd.Series(pd.Categorical(["a", "b", "c", "c"], categories=["c", "a", "b", "d"]))
+    s.value_counts()
+
+``DataFrame`` methods like :meth:`DataFrame.sum` also show "unused" categories when ``observed=False``.
+
+.. ipython:: python
+
+    columns = pd.Categorical(
+        ["One", "One", "Two"], categories=["One", "Two", "Three"], ordered=True
+    )
+    df = pd.DataFrame(
+        data=[[1, 2, 3], [4, 5, 6]],
+        columns=pd.MultiIndex.from_arrays([["A", "B", "B"], columns]),
+    ).T
+    df.groupby(level=1, observed=False).sum()
+
+Groupby will also show "unused" categories when ``observed=False``:
+
+.. ipython:: python
+
+    cats = pd.Categorical(
+        ["a", "b", "b", "b", "c", "c", "c"], categories=["a", "b", "c", "d"]
+    )
+    df = pd.DataFrame({"cats": cats, "values": [1, 2, 2, 2, 3, 4, 5]})
+    df.groupby("cats", observed=False).mean()
+
+    cats2 = pd.Categorical(["a", "a", "b", "b"], categories=["a", "b", "c"])
+    df2 = pd.DataFrame(
+        {
+            "cats": cats2,
+            "B": ["c", "d", "c", "d"],
+            "values": [1, 2, 3, 4],
+        }
+    )
+    df2.groupby(["cats", "B"], observed=False).mean()
+
+
+Pivot tables:
+
+.. ipython:: python
+
+    raw_cat = pd.Categorical(["a", "a", "b", "b"], categories=["a", "b", "c"])
+    df = pd.DataFrame({"A": raw_cat, "B": ["c", "d", "c", "d"], "values": [1, 2, 3, 4]})
+    pd.pivot_table(df, values="values", index=["A", "B"], observed=False)
+
+Data munging
+------------
+
+The optimized pandas data access methods  ``.loc``, ``.iloc``, ``.at``, and ``.iat``,
+work as normal. The only difference is the return type (for getting) and
+that only values already in ``categories`` can be assigned.
+
+Getting
+~~~~~~~
+
+If the slicing operation returns either a ``DataFrame`` or a column of type
+``Series``, the ``category`` dtype is preserved.
+
+.. ipython:: python
+
+    idx = pd.Index(["h", "i", "j", "k", "l", "m", "n"])
+    cats = pd.Series(["a", "b", "b", "b", "c", "c", "c"], dtype="category", index=idx)
+    values = [1, 2, 2, 2, 3, 4, 5]
+    df = pd.DataFrame({"cats": cats, "values": values}, index=idx)
+    df.iloc[2:4, :]
+    df.iloc[2:4, :].dtypes
+    df.loc["h":"j", "cats"]
+    df[df["cats"] == "b"]
+
+An example where the category type is not preserved is if you take one single
+row: the resulting ``Series`` is of dtype ``object``:
+
+.. ipython:: python
+
+    # get the complete "h" row as a Series
+    df.loc["h", :]
+
+Returning a single item from categorical data will also return the value, not a categorical
+of length "1".
+
+.. ipython:: python
+
+    df.iat[0, 0]
+    df["cats"] = df["cats"].cat.rename_categories(["x", "y", "z"])
+    df.at["h", "cats"]  # returns a string
+
+.. note::
+    The is in contrast to R's ``factor`` function, where ``factor(c(1,2,3))[1]``
+    returns a single value ``factor``.
+
+To get a single value ``Series`` of type ``category``, you pass in a list with
+a single value:
+
+.. ipython:: python
+
+    df.loc[["h"], "cats"]
+
+String and datetime accessors
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The accessors  ``.dt`` and ``.str`` will work if the ``s.cat.categories`` are of
+an appropriate type:
+
+
+.. ipython:: python
+
+    str_s = pd.Series(list("aabb"))
+    str_cat = str_s.astype("category")
+    str_cat
+    str_cat.str.contains("a")
+
+    date_s = pd.Series(pd.date_range("1/1/2015", periods=5))
+    date_cat = date_s.astype("category")
+    date_cat
+    date_cat.dt.day
+
+.. note::
+
+    The returned ``Series`` (or ``DataFrame``) is of the same type as if you used the
+    ``.str.<method>`` / ``.dt.<method>`` on a ``Series`` of that type (and not of
+    type ``category``!).
+
+That means, that the returned values from methods and properties on the accessors of a
+``Series`` and the returned values from methods and properties on the accessors of this
+``Series`` transformed to one of type ``category`` will be equal:
+
+.. ipython:: python
+
+    ret_s = str_s.str.contains("a")
+    ret_cat = str_cat.str.contains("a")
+    ret_s.dtype == ret_cat.dtype
+    ret_s == ret_cat
+
+.. note::
+
+    The work is done on the ``categories`` and then a new ``Series`` is constructed. This has
+    some performance implication if you have a ``Series`` of type string, where lots of elements
+    are repeated (i.e. the number of unique elements in the ``Series`` is a lot smaller than the
+    length of the ``Series``). In this case it can be faster to convert the original ``Series``
+    to one of type ``category`` and use ``.str.<method>`` or ``.dt.<property>`` on that.
+
+Setting
+~~~~~~~
+
+Setting values in a categorical column (or ``Series``) works as long as the
+value is included in the ``categories``:
+
+.. ipython:: python
+
+    idx = pd.Index(["h", "i", "j", "k", "l", "m", "n"])
+    cats = pd.Categorical(["a", "a", "a", "a", "a", "a", "a"], categories=["a", "b"])
+    values = [1, 1, 1, 1, 1, 1, 1]
+    df = pd.DataFrame({"cats": cats, "values": values}, index=idx)
+
+    df.iloc[2:4, :] = [["b", 2], ["b", 2]]
+    df
+    try:
+        df.iloc[2:4, :] = [["c", 3], ["c", 3]]
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+Setting values by assigning categorical data will also check that the ``categories`` match:
+
+.. ipython:: python
+
+    df.loc["j":"k", "cats"] = pd.Categorical(["a", "a"], categories=["a", "b"])
+    df
+    try:
+        df.loc["j":"k", "cats"] = pd.Categorical(["b", "b"], categories=["a", "b", "c"])
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+Assigning a ``Categorical`` to parts of a column of other types will use the values:
+
+.. ipython:: python
+    :okwarning:
+
+    df = pd.DataFrame({"a": [1, 1, 1, 1, 1], "b": ["a", "a", "a", "a", "a"]})
+    df.loc[1:2, "a"] = pd.Categorical([2, 2], categories=[2, 3])
+    df.loc[2:3, "b"] = pd.Categorical(["b", "b"], categories=["a", "b"])
+    df
+    df.dtypes
+
+.. _categorical.merge:
+.. _categorical.concat:
+
+Merging / concatenation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+By default, combining ``Series`` or ``DataFrames`` which contain the same
+categories results in ``category`` dtype, otherwise results will depend on the
+dtype of the underlying categories. Merges that result in non-categorical
+dtypes will likely have higher memory usage. Use ``.astype`` or
+``union_categoricals`` to ensure ``category`` results.
+
+.. ipython:: python
+
+   from pandas.api.types import union_categoricals
+
+   # same categories
+   s1 = pd.Series(["a", "b"], dtype="category")
+   s2 = pd.Series(["a", "b", "a"], dtype="category")
+   pd.concat([s1, s2])
+
+   # different categories
+   s3 = pd.Series(["b", "c"], dtype="category")
+   pd.concat([s1, s3])
+
+   # Output dtype is inferred based on categories values
+   int_cats = pd.Series([1, 2], dtype="category")
+   float_cats = pd.Series([3.0, 4.0], dtype="category")
+   pd.concat([int_cats, float_cats])
+
+   pd.concat([s1, s3]).astype("category")
+   union_categoricals([s1.array, s3.array])
+
+The following table summarizes the results of merging ``Categoricals``:
+
++-------------------+------------------------+----------------------+-----------------------------+
+| arg1              | arg2                   |      identical       | result                      |
++===================+========================+======================+=============================+
+| category          | category               | True                 | category                    |
++-------------------+------------------------+----------------------+-----------------------------+
+| category (object) | category (object)      | False                | object (dtype is inferred)  |
++-------------------+------------------------+----------------------+-----------------------------+
+| category (int)    | category (float)       | False                | float (dtype is inferred)   |
++-------------------+------------------------+----------------------+-----------------------------+
+
+.. _categorical.union:
+
+Unioning
+~~~~~~~~
+
+If you want to combine categoricals that do not necessarily have the same
+categories, the :func:`~pandas.api.types.union_categoricals` function will
+combine a list-like of categoricals. The new categories will be the union of
+the categories being combined.
+
+.. ipython:: python
+
+    from pandas.api.types import union_categoricals
+
+    a = pd.Categorical(["b", "c"])
+    b = pd.Categorical(["a", "b"])
+    union_categoricals([a, b])
+
+By default, the resulting categories will be ordered as
+they appear in the data. If you want the categories to
+be lexsorted, use ``sort_categories=True`` argument.
+
+.. ipython:: python
+
+    union_categoricals([a, b], sort_categories=True)
+
+``union_categoricals`` also works with the "easy" case of combining two
+categoricals of the same categories and order information
+(e.g. what you could also ``append`` for).
+
+.. ipython:: python
+
+    a = pd.Categorical(["a", "b"], ordered=True)
+    b = pd.Categorical(["a", "b", "a"], ordered=True)
+    union_categoricals([a, b])
+
+The below raises ``TypeError`` because the categories are ordered and not identical.
+
+.. ipython:: python
+   :okexcept:
+
+   a = pd.Categorical(["a", "b"], ordered=True)
+   b = pd.Categorical(["a", "b", "c"], ordered=True)
+   union_categoricals([a, b])
+
+Ordered categoricals with different categories or orderings can be combined by
+using the ``ignore_ordered=True`` argument.
+
+.. ipython:: python
+
+    a = pd.Categorical(["a", "b", "c"], ordered=True)
+    b = pd.Categorical(["c", "b", "a"], ordered=True)
+    union_categoricals([a, b], ignore_order=True)
+
+:func:`~pandas.api.types.union_categoricals` also works with a
+``CategoricalIndex``, or ``Series`` containing categorical data, but note that
+the resulting array will always be a plain ``Categorical``:
+
+.. ipython:: python
+
+    a = pd.Series(["b", "c"], dtype="category")
+    b = pd.Series(["a", "b"], dtype="category")
+    union_categoricals([a, b])
+
+.. note::
+
+   ``union_categoricals`` may recode the integer codes for categories
+   when combining categoricals.  This is likely what you want,
+   but if you are relying on the exact numbering of the categories, be
+   aware.
+
+   .. ipython:: python
+
+      c1 = pd.Categorical(["b", "c"])
+      c2 = pd.Categorical(["a", "b"])
+
+      c1
+      # "b" is coded to 0
+      c1.codes
+
+      c2
+      # "b" is coded to 1
+      c2.codes
+
+      c = union_categoricals([c1, c2])
+      c
+      # "b" is coded to 0 throughout, same as c1, different from c2
+      c.codes
+
+
+Getting data in/out
+-------------------
+
+You can write data that contains ``category`` dtypes to a ``HDFStore``.
+See :ref:`here <io.hdf5-categorical>` for an example and caveats.
+
+It is also possible to write data to and reading data from *Stata* format files.
+See :ref:`here <io.stata-categorical>` for an example and caveats.
+
+Writing to a CSV file will convert the data, effectively removing any information about the
+categorical (categories and ordering). So if you read back the CSV file you have to convert the
+relevant columns back to ``category`` and assign the right categories and categories ordering.
+
+.. ipython:: python
+
+    import io
+
+    s = pd.Series(pd.Categorical(["a", "b", "b", "a", "a", "d"]))
+    # rename the categories
+    s = s.cat.rename_categories(["very good", "good", "bad"])
+    # reorder the categories and add missing categories
+    s = s.cat.set_categories(["very bad", "bad", "medium", "good", "very good"])
+    df = pd.DataFrame({"cats": s, "vals": [1, 2, 3, 4, 5, 6]})
+    csv = io.StringIO()
+    df.to_csv(csv)
+    df2 = pd.read_csv(io.StringIO(csv.getvalue()))
+    df2.dtypes
+    df2["cats"]
+    # Redo the category
+    df2["cats"] = df2["cats"].astype("category")
+    df2["cats"] = df2["cats"].cat.set_categories(
+        ["very bad", "bad", "medium", "good", "very good"]
+    )
+    df2.dtypes
+    df2["cats"]
+
+The same holds for writing to a SQL database with ``to_sql``.
+
+Missing data
+------------
+
+pandas primarily uses the value ``np.nan`` to represent missing data. It is by
+default not included in computations. See the :ref:`Missing Data section
+<missing_data>`.
+
+Missing values should **not** be included in the Categorical's ``categories``,
+only in the ``values``.
+Instead, it is understood that NaN is different, and is always a possibility.
+When working with the Categorical's ``codes``, missing values will always have
+a code of ``-1``.
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", np.nan, "a"], dtype="category")
+    # only two categories
+    s
+    s.cat.codes
+
+
+Methods for working with missing data, e.g. :meth:`~Series.isna`, :meth:`~Series.fillna`,
+:meth:`~Series.dropna`, all work normally:
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", np.nan], dtype="category")
+    s
+    pd.isna(s)
+    s.fillna("a")
+
+Differences to R's ``factor``
+-----------------------------
+
+The following differences to R's factor functions can be observed:
+
+* R's ``levels`` are named ``categories``.
+* R's ``levels`` are always of type string, while ``categories`` in pandas can be of any dtype.
+* It's not possible to specify labels at creation time. Use ``s.cat.rename_categories(new_labels)``
+  afterwards.
+* In contrast to R's ``factor`` function, using categorical data as the sole input to create a
+  new categorical series will *not* remove unused categories but create a new categorical series
+  which is equal to the passed in one!
+* R allows for missing values to be included in its ``levels`` (pandas' ``categories``). pandas
+  does not allow ``NaN`` categories, but missing values can still be in the ``values``.
+
+
+Gotchas
+-------
+
+.. _categorical.rfactor:
+
+Memory usage
+~~~~~~~~~~~~
+
+.. _categorical.memory:
+
+The memory usage of a ``Categorical`` is proportional to the number of categories plus the length of the data. In contrast,
+an ``object`` dtype is a constant times the length of the data.
+
+.. ipython:: python
+
+   s = pd.Series(["foo", "bar"] * 1000)
+
+   # object dtype
+   s.nbytes
+
+   # category dtype
+   s.astype("category").nbytes
+
+.. note::
+
+   If the number of categories approaches the length of the data, the ``Categorical`` will use nearly the same or
+   more memory than an equivalent ``object`` dtype representation.
+
+   .. ipython:: python
+
+      s = pd.Series(["foo%04d" % i for i in range(2000)])
+
+      # object dtype
+      s.nbytes
+
+      # category dtype
+      s.astype("category").nbytes
+
+
+``Categorical`` is not a ``numpy`` array
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Currently, categorical data and the underlying ``Categorical`` is implemented as a Python
+object and not as a low-level NumPy array dtype. This leads to some problems.
+
+NumPy itself doesn't know about the new ``dtype``:
+
+.. ipython:: python
+
+    try:
+        np.dtype("category")
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+    dtype = pd.Categorical(["a"]).dtype
+    try:
+        np.dtype(dtype)
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+Dtype comparisons work:
+
+.. ipython:: python
+
+    dtype == np.str_
+    np.str_ == dtype
+
+To check if a Series contains Categorical data, use ``hasattr(s, 'cat')``:
+
+.. ipython:: python
+
+    hasattr(pd.Series(["a"], dtype="category"), "cat")
+    hasattr(pd.Series(["a"]), "cat")
+
+Using NumPy functions on a ``Series`` of type ``category`` should not work as ``Categoricals``
+are not numeric data (even in the case that ``.categories`` is numeric).
+
+.. ipython:: python
+
+    s = pd.Series(pd.Categorical([1, 2, 3, 4]))
+    try:
+        np.sum(s)
+        # same with np.log(s),...
+    except TypeError as e:
+        print("TypeError:", str(e))
+
+.. note::
+    If such a function works, please file a bug at https://github.com/pandas-dev/pandas!
+
+dtype in apply
+~~~~~~~~~~~~~~
+
+pandas currently does not preserve the dtype in apply functions: If you apply along rows you get
+a ``Series`` of ``object`` ``dtype`` (same as getting a row -> getting one element will return a
+basic type) and applying along columns will also convert to object. ``NaN`` values are unaffected.
+You can use ``fillna`` to handle missing values before applying a function.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {
+            "a": [1, 2, 3, 4],
+            "b": ["a", "b", "c", "d"],
+            "cats": pd.Categorical([1, 2, 3, 2]),
+        }
+    )
+    df.apply(lambda row: type(row["cats"]), axis=1)
+    df.apply(lambda col: col.dtype, axis=0)
+
+Categorical index
+~~~~~~~~~~~~~~~~~
+
+``CategoricalIndex`` is a type of index that is useful for supporting
+indexing with duplicates. This is a container around a ``Categorical``
+and allows efficient indexing and storage of an index with a large number of duplicated elements.
+See the :ref:`advanced indexing docs <advanced.categoricalindex>` for a more detailed
+explanation.
+
+Setting the index will create a ``CategoricalIndex``:
+
+.. ipython:: python
+
+    cats = pd.Categorical([1, 2, 3, 4], categories=[4, 2, 3, 1])
+    strings = ["a", "b", "c", "d"]
+    values = [4, 2, 3, 1]
+    df = pd.DataFrame({"strings": strings, "values": values}, index=cats)
+    df.index
+    # This now sorts by the categories order
+    df.sort_index()
+
+Side effects
+~~~~~~~~~~~~
+
+Constructing a ``Series`` from a ``Categorical`` will not copy the input
+``Categorical``. This means that changes to the ``Series`` will in most cases
+change the original ``Categorical``:
+
+.. ipython:: python
+
+    cat = pd.Categorical([1, 2, 3, 10], categories=[1, 2, 3, 4, 10])
+    s = pd.Series(cat, name="cat")
+    cat
+    s.iloc[0:2] = 10
+    cat
+
+Use ``copy=True`` to prevent such a behaviour or simply don't reuse ``Categoricals``:
+
+.. ipython:: python
+
+    cat = pd.Categorical([1, 2, 3, 10], categories=[1, 2, 3, 4, 10])
+    s = pd.Series(cat, name="cat", copy=True)
+    cat
+    s.iloc[0:2] = 10
+    cat
+
+.. note::
+
+    This also happens in some cases when you supply a NumPy array instead of a ``Categorical``:
+    using an int array (e.g. ``np.array([1,2,3,4])``) will exhibit the same behavior, while using
+    a string array (e.g. ``np.array(["a","b","c","a"])``) will not.
diff --git a/pandas/doc/source/user_guide/cookbook.rst b/pandas/doc/source/user_guide/cookbook.rst
new file mode 100644
index 0000000000000..91a0b4a4fe967
--- /dev/null
+++ b/pandas/doc/source/user_guide/cookbook.rst
@@ -0,0 +1,1518 @@
+.. _cookbook:
+
+{{ header }}
+
+********
+Cookbook
+********
+
+This is a repository for *short and sweet* examples and links for useful pandas recipes.
+We encourage users to add to this documentation.
+
+Adding interesting links and/or inline examples to this section is a great *First Pull Request*.
+
+Simplified, condensed, new-user friendly, in-line examples have been inserted where possible to
+augment the Stack-Overflow and GitHub links.  Many of the links contain expanded information,
+above what the in-line examples offer.
+
+pandas (pd) and NumPy (np) are the only two abbreviated imported modules. The rest are kept
+explicitly imported for newer users.
+
+Idioms
+------
+
+.. _cookbook.idioms:
+
+These are some neat pandas ``idioms``
+
+`if-then/if-then-else on one column, and assignment to another one or more columns:
+<https://stackoverflow.com/questions/17128302/python-pandas-idiom-for-if-then-else>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+If-then...
+**********
+
+An if-then on one column
+
+.. ipython:: python
+
+   df.loc[df.AAA >= 5, "BBB"] = -1
+   df
+
+An if-then with assignment to 2 columns:
+
+.. ipython:: python
+
+   df.loc[df.AAA >= 5, ["BBB", "CCC"]] = 555
+   df
+
+Add another line with different logic, to do the -else
+
+.. ipython:: python
+
+   df.loc[df.AAA < 5, ["BBB", "CCC"]] = 2000
+   df
+
+Or use pandas where after you've set up a mask
+
+.. ipython:: python
+
+   df_mask = pd.DataFrame(
+       {"AAA": [True] * 4, "BBB": [False] * 4, "CCC": [True, False] * 2}
+   )
+   df.where(df_mask, -1000)
+
+`if-then-else using NumPy's where()
+<https://stackoverflow.com/questions/19913659/pandas-conditional-creation-of-a-series-dataframe-column>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+   df["logic"] = np.where(df["AAA"] > 5, "high", "low")
+   df
+
+Splitting
+*********
+
+`Split a frame with a boolean criterion
+<https://stackoverflow.com/questions/14957116/how-to-split-a-dataframe-according-to-a-boolean-criterion>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+   df[df.AAA <= 5]
+   df[df.AAA > 5]
+
+Building criteria
+*****************
+
+`Select with multi-column criteria
+<https://stackoverflow.com/questions/15315452/selecting-with-complex-criteria-from-pandas-dataframe>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+...and (without assignment returns a Series)
+
+.. ipython:: python
+
+   df.loc[(df["BBB"] < 25) & (df["CCC"] >= -40), "AAA"]
+
+...or (without assignment returns a Series)
+
+.. ipython:: python
+
+   df.loc[(df["BBB"] > 25) | (df["CCC"] >= -40), "AAA"]
+
+...or (with assignment modifies the DataFrame.)
+
+.. ipython:: python
+
+   df.loc[(df["BBB"] > 25) | (df["CCC"] >= 75), "AAA"] = 999
+   df
+
+`Select rows with data closest to certain value using argsort
+<https://stackoverflow.com/questions/17758023/return-rows-in-a-dataframe-closest-to-a-user-defined-number>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+   aValue = 43.0
+   df.loc[(df.CCC - aValue).abs().argsort()]
+
+`Dynamically reduce a list of criteria using a binary operators
+<https://stackoverflow.com/questions/21058254/pandas-boolean-operation-in-a-python-list/21058331>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+   Crit1 = df.AAA <= 5.5
+   Crit2 = df.BBB == 10.0
+   Crit3 = df.CCC > -40.0
+
+One could hard code:
+
+.. ipython:: python
+
+   AllCrit = Crit1 & Crit2 & Crit3
+
+...Or it can be done with a list of dynamically built criteria
+
+.. ipython:: python
+
+   import functools
+
+   CritList = [Crit1, Crit2, Crit3]
+   AllCrit = functools.reduce(lambda x, y: x & y, CritList)
+
+   df[AllCrit]
+
+.. _cookbook.selection:
+
+Selection
+---------
+
+DataFrames
+**********
+
+The :ref:`indexing <indexing>` docs.
+
+`Using both row labels and value conditionals
+<https://stackoverflow.com/questions/14725068/pandas-using-row-labels-in-boolean-indexing>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+   df[(df.AAA <= 6) & (df.index.isin([0, 2, 4]))]
+
+Use loc for label-oriented slicing and iloc positional slicing :issue:`2904`
+
+.. ipython:: python
+
+  df = pd.DataFrame(
+      {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]},
+      index=["foo", "bar", "boo", "kar"],
+  )
+
+
+There are 2 explicit slicing methods, with a third general case
+
+1. Positional-oriented (Python slicing style : exclusive of end)
+2. Label-oriented (Non-Python slicing style : inclusive of end)
+3. General (Either slicing style : depends on if the slice contains labels or positions)
+
+.. ipython:: python
+   df.iloc[0:3]  # Positional
+
+   df.loc["bar":"kar"]  # Label
+
+   # Generic
+   df[0:3]
+   df["bar":"kar"]
+
+Ambiguity arises when an index consists of integers with a non-zero start or non-unit increment.
+
+.. ipython:: python
+
+   data = {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   df2 = pd.DataFrame(data=data, index=[1, 2, 3, 4])  # Note index starts at 1.
+   df2.iloc[1:3]  # Position-oriented
+   df2.loc[1:3]  # Label-oriented
+
+`Using inverse operator (~) to take the complement of a mask
+<https://stackoverflow.com/q/14986510>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [4, 5, 6, 7], "BBB": [10, 20, 30, 40], "CCC": [100, 50, -30, -50]}
+   )
+   df
+
+   df[~((df.AAA <= 6) & (df.index.isin([0, 2, 4])))]
+
+New columns
+***********
+
+`Efficiently and dynamically creating new columns using DataFrame.map (previously named applymap)
+<https://stackoverflow.com/questions/16575868/efficiently-creating-additional-columns-in-a-pandas-dataframe-using-map>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame({"AAA": [1, 2, 1, 3], "BBB": [1, 1, 2, 2], "CCC": [2, 1, 3, 1]})
+   df
+
+   source_cols = df.columns  # Or some subset would work too
+   new_cols = [str(x) + "_cat" for x in source_cols]
+   categories = {1: "Alpha", 2: "Beta", 3: "Charlie"}
+
+   df[new_cols] = df[source_cols].map(categories.get)
+   df
+
+`Keep other columns when using min() with groupby
+<https://stackoverflow.com/q/23394476>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"AAA": [1, 1, 1, 2, 2, 2, 3, 3], "BBB": [2, 1, 3, 4, 5, 1, 2, 3]}
+   )
+   df
+
+Method 1 : idxmin() to get the index of the minimums
+
+.. ipython:: python
+
+   df.loc[df.groupby("AAA")["BBB"].idxmin()]
+
+Method 2 : sort then take first of each
+
+.. ipython:: python
+
+   df.sort_values(by="BBB").groupby("AAA", as_index=False).first()
+
+Notice the same results, with the exception of the index.
+
+.. _cookbook.multi_index:
+
+Multiindexing
+-------------
+
+The :ref:`multindexing <advanced.hierarchical>` docs.
+
+`Creating a MultiIndex from a labeled frame
+<https://stackoverflow.com/questions/14916358/reshaping-dataframes-in-pandas-based-on-column-labels>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "row": [0, 1, 2],
+           "One_X": [1.1, 1.1, 1.1],
+           "One_Y": [1.2, 1.2, 1.2],
+           "Two_X": [1.11, 1.11, 1.11],
+           "Two_Y": [1.22, 1.22, 1.22],
+       }
+   )
+   df
+
+   # As Labelled Index
+   df = df.set_index("row")
+   df
+   # With Hierarchical Columns
+   df.columns = pd.MultiIndex.from_tuples([tuple(c.split("_")) for c in df.columns])
+   df
+   # Now stack & Reset
+   df = df.stack(0).reset_index(1)
+   df
+   # And fix the labels (Notice the label 'level_1' got added automatically)
+   df.columns = ["Sample", "All_X", "All_Y"]
+   df
+
+Arithmetic
+**********
+
+`Performing arithmetic with a MultiIndex that needs broadcasting
+<https://stackoverflow.com/questions/19501510/divide-entire-pandas-multiindex-dataframe-by-dataframe-variable/19502176#19502176>`__
+
+.. ipython:: python
+
+   cols = pd.MultiIndex.from_tuples(
+       [(x, y) for x in ["A", "B", "C"] for y in ["O", "I"]]
+   )
+   df = pd.DataFrame(np.random.randn(2, 6), index=["n", "m"], columns=cols)
+   df
+   df = df.div(df["C"], level=1)
+   df
+
+Slicing
+*******
+
+`Slicing a MultiIndex with xs
+<https://stackoverflow.com/questions/12590131/how-to-slice-multindex-columns-in-pandas-dataframes>`__
+
+.. ipython:: python
+
+   coords = [("AA", "one"), ("AA", "six"), ("BB", "one"), ("BB", "two"), ("BB", "six")]
+   index = pd.MultiIndex.from_tuples(coords)
+   df = pd.DataFrame([11, 22, 33, 44, 55], index, ["MyData"])
+   df
+
+To take the cross section of the 1st level and 1st axis the index:
+
+.. ipython:: python
+
+   # Note : level and axis are optional, and default to zero
+   df.xs("BB", level=0, axis=0)
+
+...and now the 2nd level of the 1st axis.
+
+.. ipython:: python
+
+   df.xs("six", level=1, axis=0)
+
+`Slicing a MultiIndex with xs, method #2
+<https://stackoverflow.com/questions/14964493/multiindex-based-indexing-in-pandas>`__
+
+.. ipython:: python
+
+   import itertools
+
+   index = list(itertools.product(["Ada", "Quinn", "Violet"], ["Comp", "Math", "Sci"]))
+   headr = list(itertools.product(["Exams", "Labs"], ["I", "II"]))
+   indx = pd.MultiIndex.from_tuples(index, names=["Student", "Course"])
+   cols = pd.MultiIndex.from_tuples(headr)  # Notice these are un-named
+   data = [[70 + x + y + (x * y) % 3 for x in range(4)] for y in range(9)]
+   df = pd.DataFrame(data, indx, cols)
+   df
+
+   All = slice(None)
+   df.loc["Violet"]
+   df.loc[(All, "Math"), All]
+   df.loc[(slice("Ada", "Quinn"), "Math"), All]
+   df.loc[(All, "Math"), ("Exams")]
+   df.loc[(All, "Math"), (All, "II")]
+
+`Setting portions of a MultiIndex with xs
+<https://stackoverflow.com/questions/19319432/pandas-selecting-a-lower-level-in-a-dataframe-to-do-a-ffill>`__
+
+Sorting
+*******
+
+`Sort by specific column or an ordered list of columns, with a MultiIndex
+<https://stackoverflow.com/q/14733871>`__
+
+.. ipython:: python
+
+   df.sort_values(by=("Labs", "II"), ascending=False)
+
+Partial selection, the need for sortedness :issue:`2995`
+
+Levels
+******
+
+`Prepending a level to a multiindex
+<https://stackoverflow.com/questions/14744068/prepend-a-level-to-a-pandas-multiindex>`__
+
+`Flatten Hierarchical columns
+<https://stackoverflow.com/q/14507794>`__
+
+.. _cookbook.missing_data:
+
+Missing data
+------------
+
+The :ref:`missing data<missing_data>` docs.
+
+Fill forward a reversed timeseries
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.random.randn(6, 1),
+       index=pd.date_range("2013-08-01", periods=6, freq="B"),
+       columns=list("A"),
+   )
+   df.loc[df.index[3], "A"] = np.nan
+   df
+   df.bfill()
+
+`cumsum reset at NaN values
+<https://stackoverflow.com/questions/18196811/cumsum-reset-at-nan>`__
+
+Replace
+*******
+
+`Using replace with backrefs
+<https://stackoverflow.com/questions/16818871/extracting-value-and-creating-new-column-out-of-it>`__
+
+.. _cookbook.grouping:
+
+Grouping
+--------
+
+The :ref:`grouping <groupby>` docs.
+
+`Basic grouping with apply
+<https://stackoverflow.com/questions/15322632/python-pandas-df-groupy-agg-column-reference-in-agg>`__
+
+Unlike agg, apply's callable is passed a sub-DataFrame which gives you access to all the columns
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "animal": "cat dog cat fish dog cat cat".split(),
+           "size": list("SSMMMLL"),
+           "weight": [8, 10, 11, 1, 20, 12, 12],
+           "adult": [False] * 5 + [True] * 2,
+       }
+   )
+   df
+
+   # List the size of the animals with the highest weight.
+   df.groupby("animal").apply(lambda subf: subf["size"][subf["weight"].idxmax()])
+
+`Using get_group
+<https://stackoverflow.com/questions/14734533/how-to-access-pandas-groupby-dataframe-by-key>`__
+
+.. ipython:: python
+
+   gb = df.groupby("animal")
+   gb.get_group("cat")
+
+`Apply to different items in a group
+<https://stackoverflow.com/questions/15262134/apply-different-functions-to-different-items-in-group-object-python-pandas>`__
+
+.. ipython:: python
+
+   def GrowUp(x):
+       avg_weight = sum(x[x["size"] == "S"].weight * 1.5)
+       avg_weight += sum(x[x["size"] == "M"].weight * 1.25)
+       avg_weight += sum(x[x["size"] == "L"].weight)
+       avg_weight /= len(x)
+       return pd.Series(["L", avg_weight, True], index=["size", "weight", "adult"])
+
+
+   expected_df = gb.apply(GrowUp)
+   expected_df
+
+`Expanding apply
+<https://stackoverflow.com/questions/14542145/reductions-down-a-column-in-pandas>`__
+
+.. ipython:: python
+
+   S = pd.Series([i / 100.0 for i in range(1, 11)])
+
+   def cum_ret(x, y):
+       return x * (1 + y)
+
+   def red(x):
+       return functools.reduce(cum_ret, x, 1.0)
+
+   S.expanding().apply(red, raw=True)
+
+
+`Replacing some values with mean of the rest of a group
+<https://stackoverflow.com/questions/14760757/replacing-values-with-groupby-means>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 1, 2, 2], "B": [1, -1, 1, 2]})
+   gb = df.groupby("A")
+
+   def replace(g):
+       mask = g < 0
+       return g.where(~mask, g[~mask].mean())
+
+   gb.transform(replace)
+
+`Sort groups by aggregated data
+<https://stackoverflow.com/questions/14941366/pandas-sort-by-group-aggregate-and-column>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "code": ["foo", "bar", "baz"] * 2,
+           "data": [0.16, -0.21, 0.33, 0.45, -0.59, 0.62],
+           "flag": [False, True] * 3,
+       }
+   )
+
+   code_groups = df.groupby("code")
+
+   agg_n_sort_order = code_groups[["data"]].transform("sum").sort_values(by="data")
+
+   sorted_df = df.loc[agg_n_sort_order.index]
+
+   sorted_df
+
+`Create multiple aggregated columns
+<https://stackoverflow.com/questions/14897100/create-multiple-columns-in-pandas-aggregation-function>`__
+
+.. ipython:: python
+
+   rng = pd.date_range(start="2014-10-07", periods=10, freq="2min")
+   ts = pd.Series(data=list(range(10)), index=rng)
+
+   def MyCust(x):
+       if len(x) > 2:
+           return x.iloc[1] * 1.234
+       return pd.NaT
+
+   mhc = {"Mean": "mean", "Max": "max", "Custom": MyCust}
+   ts.resample("5min").apply(mhc)
+   ts
+
+`Create a value counts column and reassign back to the DataFrame
+<https://stackoverflow.com/q/17709270>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"Color": "Red Red Red Blue".split(), "Value": [100, 150, 50, 50]}
+   )
+   df
+   df["Counts"] = df.groupby(["Color"]).transform(len)
+   df
+
+`Shift groups of the values in a column based on the index
+<https://stackoverflow.com/q/23198053/190597>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"line_race": [10, 10, 8, 10, 10, 8], "beyer": [99, 102, 103, 103, 88, 100]},
+       index=[
+           "Last Gunfighter",
+           "Last Gunfighter",
+           "Last Gunfighter",
+           "Paynter",
+           "Paynter",
+           "Paynter",
+       ],
+   )
+   df
+   df["beyer_shifted"] = df.groupby(level=0)["beyer"].shift(1)
+   df
+
+`Select row with maximum value from each group
+<https://stackoverflow.com/q/26701849/190597>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "host": ["other", "other", "that", "this", "this"],
+           "service": ["mail", "web", "mail", "mail", "web"],
+           "no": [1, 2, 1, 2, 1],
+       }
+   ).set_index(["host", "service"])
+   mask = df.groupby(level=0).agg("idxmax")
+   df_count = df.loc[mask["no"]].reset_index()
+   df_count
+
+`Grouping like Python's itertools.groupby
+<https://stackoverflow.com/q/29142487/846892>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame([0, 1, 0, 1, 1, 1, 0, 1, 1], columns=["A"])
+   df["A"].groupby((df["A"] != df["A"].shift()).cumsum()).groups
+   df["A"].groupby((df["A"] != df["A"].shift()).cumsum()).cumsum()
+
+Expanding data
+**************
+
+`Alignment and to-date
+<https://stackoverflow.com/questions/15489011/python-time-series-alignment-and-to-date-functions>`__
+
+`Rolling Computation window based on values instead of counts
+<https://stackoverflow.com/questions/14300768/pandas-rolling-computation-with-window-based-on-values-instead-of-counts>`__
+
+`Rolling Mean by Time Interval
+<https://stackoverflow.com/questions/15771472/pandas-rolling-mean-by-time-interval>`__
+
+Splitting
+*********
+
+`Splitting a frame
+<https://stackoverflow.com/questions/13353233/best-way-to-split-a-dataframe-given-an-edge/15449992#15449992>`__
+
+Create a list of dataframes, split using a delineation based on logic included in rows.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data={
+           "Case": ["A", "A", "A", "B", "A", "A", "B", "A", "A"],
+           "Data": np.random.randn(9),
+       }
+   )
+
+   dfs = list(
+       zip(
+           *df.groupby(
+               (1 * (df["Case"] == "B"))
+               .cumsum()
+               .rolling(window=3, min_periods=1)
+               .median()
+           )
+       )
+   )[-1]
+
+   dfs[0]
+   dfs[1]
+   dfs[2]
+
+.. _cookbook.pivot:
+
+Pivot
+*****
+The :ref:`Pivot <reshaping.pivot>` docs.
+
+`Partial sums and subtotals
+<https://stackoverflow.com/a/15574875>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data={
+           "Province": ["ON", "QC", "BC", "AL", "AL", "MN", "ON"],
+           "City": [
+               "Toronto",
+               "Montreal",
+               "Vancouver",
+               "Calgary",
+               "Edmonton",
+               "Winnipeg",
+               "Windsor",
+           ],
+           "Sales": [13, 6, 16, 8, 4, 3, 1],
+       }
+   )
+   table = pd.pivot_table(
+       df,
+       values=["Sales"],
+       index=["Province"],
+       columns=["City"],
+       aggfunc="sum",
+       margins=True,
+   )
+   table.stack("City")
+
+`Frequency table like plyr in R
+<https://stackoverflow.com/questions/15589354/frequency-tables-in-pandas-like-plyr-in-r>`__
+
+.. ipython:: python
+
+   grades = [48, 99, 75, 80, 42, 80, 72, 68, 36, 78]
+   df = pd.DataFrame(
+       {
+           "ID": ["x%d" % r for r in range(10)],
+           "Gender": ["F", "M", "F", "M", "F", "M", "F", "M", "M", "M"],
+           "ExamYear": [
+               "2007",
+               "2007",
+               "2007",
+               "2008",
+               "2008",
+               "2008",
+               "2008",
+               "2009",
+               "2009",
+               "2009",
+           ],
+           "Class": [
+               "algebra",
+               "stats",
+               "bio",
+               "algebra",
+               "algebra",
+               "stats",
+               "stats",
+               "algebra",
+               "bio",
+               "bio",
+           ],
+           "Participated": [
+               "yes",
+               "yes",
+               "yes",
+               "yes",
+               "no",
+               "yes",
+               "yes",
+               "yes",
+               "yes",
+               "yes",
+           ],
+           "Passed": ["yes" if x > 50 else "no" for x in grades],
+           "Employed": [
+               True,
+               True,
+               True,
+               False,
+               False,
+               False,
+               False,
+               True,
+               True,
+               False,
+           ],
+           "Grade": grades,
+       }
+   )
+
+   df.groupby("ExamYear").agg(
+       {
+           "Participated": lambda x: x.value_counts()["yes"],
+           "Passed": lambda x: sum(x == "yes"),
+           "Employed": lambda x: sum(x),
+           "Grade": lambda x: sum(x) / len(x),
+       }
+   )
+
+`Plot pandas DataFrame with year over year data
+<https://stackoverflow.com/questions/30379789/plot-pandas-data-frame-with-year-over-year-data>`__
+
+To create year and month cross tabulation:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"value": np.random.randn(36)},
+       index=pd.date_range("2011-01-01", freq="ME", periods=36),
+   )
+
+   pd.pivot_table(
+       df, index=df.index.month, columns=df.index.year, values="value", aggfunc="sum"
+   )
+
+Apply
+*****
+
+`Rolling apply to organize - Turning embedded lists into a MultiIndex frame
+<https://stackoverflow.com/questions/17349981/converting-pandas-dataframe-with-categorical-values-into-binary-values>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data={
+           "A": [[2, 4, 8, 16], [100, 200], [10, 20, 30]],
+           "B": [["a", "b", "c"], ["jj", "kk"], ["ccc"]],
+       },
+       index=["I", "II", "III"],
+   )
+
+   def SeriesFromSubList(aList):
+       return pd.Series(aList)
+
+   df_orgz = pd.concat(
+       {ind: row.apply(SeriesFromSubList) for ind, row in df.iterrows()}
+   )
+   df_orgz
+
+`Rolling apply with a DataFrame returning a Series
+<https://stackoverflow.com/questions/19121854/using-rolling-apply-on-a-dataframe-object>`__
+
+Rolling Apply to multiple columns where function calculates a Series before a Scalar from the Series is returned
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data=np.random.randn(2000, 2) / 10000,
+       index=pd.date_range("2001-01-01", periods=2000),
+       columns=["A", "B"],
+   )
+   df
+
+   def gm(df, const):
+       v = ((((df["A"] + df["B"]) + 1).cumprod()) - 1) * const
+       return v.iloc[-1]
+
+   s = pd.Series(
+       {
+           df.index[i]: gm(df.iloc[i: min(i + 51, len(df) - 1)], 5)
+           for i in range(len(df) - 50)
+       }
+   )
+   s
+
+`Rolling apply with a DataFrame returning a Scalar
+<https://stackoverflow.com/questions/21040766/python-pandas-rolling-apply-two-column-input-into-function/21045831#21045831>`__
+
+Rolling Apply to multiple columns where function returns a Scalar (Volume Weighted Average Price)
+
+.. ipython:: python
+
+   rng = pd.date_range(start="2014-01-01", periods=100)
+   df = pd.DataFrame(
+       {
+           "Open": np.random.randn(len(rng)),
+           "Close": np.random.randn(len(rng)),
+           "Volume": np.random.randint(100, 2000, len(rng)),
+       },
+       index=rng,
+   )
+   df
+
+   def vwap(bars):
+       return (bars.Close * bars.Volume).sum() / bars.Volume.sum()
+
+   window = 5
+   s = pd.concat(
+       [
+           (pd.Series(vwap(df.iloc[i: i + window]), index=[df.index[i + window]]))
+           for i in range(len(df) - window)
+       ]
+   )
+   s.round(2)
+
+Timeseries
+----------
+
+`Between times
+<https://stackoverflow.com/questions/14539992/pandas-drop-rows-outside-of-time-range>`__
+
+`Using indexer between time
+<https://stackoverflow.com/questions/17559885/pandas-dataframe-mask-based-on-index>`__
+
+`Constructing a datetime range that excludes weekends and includes only certain times
+<https://stackoverflow.com/a/24014440>`__
+
+`Vectorized Lookup
+<https://stackoverflow.com/questions/13893227/vectorized-look-up-of-values-in-pandas-dataframe>`__
+
+`Aggregation and plotting time series
+<https://nipunbatra.github.io/blog/posts/2013-05-01-aggregation-timeseries.html>`__
+
+Turn a matrix with hours in columns and days in rows into a continuous row sequence in the form of a time series.
+`How to rearrange a Python pandas DataFrame?
+<https://stackoverflow.com/questions/15432659/how-to-rearrange-a-python-pandas-dataframe>`__
+
+`Dealing with duplicates when reindexing a timeseries to a specified frequency
+<https://stackoverflow.com/questions/22244383/pandas-df-refill-adding-two-columns-of-different-shape>`__
+
+Calculate the first day of the month for each entry in a DatetimeIndex
+
+.. ipython:: python
+
+   dates = pd.date_range("2000-01-01", periods=5)
+   dates.to_period(freq="M").to_timestamp()
+
+.. _cookbook.resample:
+
+Resampling
+**********
+
+The :ref:`Resample <timeseries.resampling>` docs.
+
+`Using Grouper instead of TimeGrouper for time grouping of values
+<https://stackoverflow.com/questions/15297053/how-can-i-divide-single-values-of-a-dataframe-by-monthly-averages>`__
+
+`Time grouping with some missing values
+<https://stackoverflow.com/questions/33637312/pandas-grouper-by-frequency-with-completeness-requirement>`__
+
+Valid frequency arguments to Grouper :ref:`Timeseries <timeseries.offset_aliases>`
+
+`Grouping using a MultiIndex
+<https://stackoverflow.com/questions/41483763/pandas-timegrouper-on-multiindex>`__
+
+Using TimeGrouper and another grouping to create subgroups, then apply a custom function :issue:`3791`
+
+`Resampling with custom periods
+<https://stackoverflow.com/questions/15408156/resampling-with-custom-periods>`__
+
+`Resample intraday frame without adding new days
+<https://stackoverflow.com/questions/14898574/resample-intraday-pandas-dataframe-without-add-new-days>`__
+
+`Resample minute data
+<https://stackoverflow.com/questions/14861023/resampling-minute-data>`__
+
+`Resample with groupby <https://stackoverflow.com/q/18677271/564538>`__
+
+.. _cookbook.merge:
+
+Merge
+-----
+
+The :ref:`Join <merging.join>` docs.
+
+`Concatenate two dataframes with overlapping index (emulate R rbind)
+<https://stackoverflow.com/questions/14988480/pandas-version-of-rbind>`__
+
+.. ipython:: python
+
+   rng = pd.date_range("2000-01-01", periods=6)
+   df1 = pd.DataFrame(np.random.randn(6, 3), index=rng, columns=["A", "B", "C"])
+   df2 = df1.copy()
+
+Depending on df construction, ``ignore_index`` may be needed
+
+.. ipython:: python
+
+   df = pd.concat([df1, df2], ignore_index=True)
+   df
+
+Self Join of a DataFrame :issue:`2996`
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data={
+           "Area": ["A"] * 5 + ["C"] * 2,
+           "Bins": [110] * 2 + [160] * 3 + [40] * 2,
+           "Test_0": [0, 1, 0, 1, 2, 0, 1],
+           "Data": np.random.randn(7),
+       }
+   )
+   df
+
+   df["Test_1"] = df["Test_0"] - 1
+
+   pd.merge(
+       df,
+       df,
+       left_on=["Bins", "Area", "Test_0"],
+       right_on=["Bins", "Area", "Test_1"],
+       suffixes=("_L", "_R"),
+   )
+
+`How to set the index and join
+<https://stackoverflow.com/questions/14341805/pandas-merge-pd-merge-how-to-set-the-index-and-join>`__
+
+`KDB like asof join
+<https://stackoverflow.com/questions/12322289/kdb-like-asof-join-for-timeseries-data-in-pandas/12336039#12336039>`__
+
+`Join with a criteria based on the values
+<https://stackoverflow.com/questions/15581829/how-to-perform-an-inner-or-outer-join-of-dataframes-with-pandas-on-non-simplisti>`__
+
+`Using searchsorted to merge based on values inside a range
+<https://stackoverflow.com/questions/25125626/pandas-merge-with-logic/2512764>`__
+
+.. _cookbook.plotting:
+
+Plotting
+--------
+
+The :ref:`Plotting <visualization>` docs.
+
+`Make Matplotlib look like R
+<https://stackoverflow.com/questions/14349055/making-matplotlib-graphs-look-like-r-by-default>`__
+
+`Setting x-axis major and minor labels
+<https://stackoverflow.com/questions/12945971/pandas-timeseries-plot-setting-x-axis-major-and-minor-ticks-and-labels>`__
+
+`Plotting multiple charts in an IPython Jupyter notebook
+<https://stackoverflow.com/questions/16392921/make-more-than-one-chart-in-same-ipython-notebook-cell>`__
+
+`Creating a multi-line plot
+<https://stackoverflow.com/questions/16568964/make-a-multiline-plot-from-csv-file-in-matplotlib>`__
+
+`Plotting a heatmap
+<https://stackoverflow.com/questions/17050202/plot-timeseries-of-histograms-in-python>`__
+
+`Annotate a time-series plot
+<https://stackoverflow.com/questions/11067368/annotate-time-series-plot-in-matplotlib>`__
+
+`Annotate a time-series plot #2
+<https://stackoverflow.com/questions/17891493/annotating-points-from-a-pandas-dataframe-in-matplotlib-plot>`__
+
+`Generate Embedded plots in excel files using Pandas, Vincent and xlsxwriter
+<https://pandas-xlsxwriter-charts.readthedocs.io/>`__
+
+`Boxplot for each quartile of a stratifying variable
+<https://stackoverflow.com/questions/23232989/boxplot-stratified-by-column-in-python-pandas>`__
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "stratifying_var": np.random.uniform(0, 100, 20),
+           "price": np.random.normal(100, 5, 20),
+       }
+   )
+
+   df["quartiles"] = pd.qcut(
+       df["stratifying_var"], 4, labels=["0-25%", "25-50%", "50-75%", "75-100%"]
+   )
+
+   @savefig quartile_boxplot.png
+   df.boxplot(column="price", by="quartiles")
+
+Data in/out
+-----------
+
+`Performance comparison of SQL vs HDF5
+<https://stackoverflow.com/q/16628329>`__
+
+.. _cookbook.csv:
+
+CSV
+***
+
+The :ref:`CSV <io.read_csv_table>` docs
+
+`read_csv in action <https://www.datacamp.com/tutorial/pandas-read-csv>`__
+
+`appending to a csv
+<https://stackoverflow.com/questions/17134942/pandas-dataframe-output-end-of-csv>`__
+
+`Reading a csv chunk-by-chunk
+<https://stackoverflow.com/questions/11622652/large-persistent-dataframe-in-pandas/12193309#12193309>`__
+
+`Reading only certain rows of a csv chunk-by-chunk
+<https://stackoverflow.com/questions/19674212/pandas-data-frame-select-rows-and-clear-memory>`__
+
+`Reading the first few lines of a frame
+<https://stackoverflow.com/questions/15008970/way-to-read-first-few-lines-for-pandas-dataframe>`__
+
+Reading a file that is compressed but not by ``gzip/bz2`` (the native compressed formats which ``read_csv`` understands).
+This example shows a ``WinZipped`` file, but is a general application of opening the file within a context manager and
+using that handle to read.
+`See here
+<https://stackoverflow.com/questions/17789907/pandas-convert-winzipped-csv-file-to-data-frame>`__
+
+`Inferring dtypes from a file
+<https://stackoverflow.com/questions/15555005/get-inferred-dataframe-types-iteratively-using-chunksize>`__
+
+Dealing with bad lines :issue:`2886`
+
+`Write a multi-row index CSV without writing duplicates
+<https://stackoverflow.com/questions/17349574/pandas-write-multiindex-rows-with-to-csv>`__
+
+.. _cookbook.csv.multiple_files:
+
+Reading multiple files to create a single DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The best way to combine multiple files into a single DataFrame is to read the individual frames one by one, put all
+of the individual frames into a list, and then combine the frames in the list using :func:`pd.concat`:
+
+.. ipython:: python
+
+    for i in range(3):
+        data = pd.DataFrame(np.random.randn(10, 4))
+        data.to_csv("file_{}.csv".format(i))
+
+    files = ["file_0.csv", "file_1.csv", "file_2.csv"]
+    result = pd.concat([pd.read_csv(f) for f in files], ignore_index=True)
+
+You can use the same approach to read all files matching a pattern.  Here is an example using ``glob``:
+
+.. ipython:: python
+
+    import glob
+    import os
+
+    files = glob.glob("file_*.csv")
+    result = pd.concat([pd.read_csv(f) for f in files], ignore_index=True)
+
+Finally, this strategy will work with the other ``pd.read_*(...)`` functions described in the :ref:`io docs<io>`.
+
+.. ipython:: python
+    :suppress:
+
+    for i in range(3):
+        os.remove("file_{}.csv".format(i))
+
+Parsing date components in multi-columns
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Parsing date components in multi-columns is faster with a format
+
+.. ipython:: python
+
+    i = pd.date_range("20000101", periods=10000)
+    df = pd.DataFrame({"year": i.year, "month": i.month, "day": i.day})
+    df.head()
+
+    %timeit pd.to_datetime(df.year * 10000 + df.month * 100 + df.day, format='%Y%m%d')
+    ds = df.apply(lambda x: "%04d%02d%02d" % (x["year"], x["month"], x["day"]), axis=1)
+    ds.head()
+    %timeit pd.to_datetime(ds)
+
+
+Skip row between header and data
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. ipython:: python
+
+    data = """;;;;
+     ;;;;
+     ;;;;
+     ;;;;
+     ;;;;
+     ;;;;
+    ;;;;
+     ;;;;
+     ;;;;
+    ;;;;
+    date;Param1;Param2;Param4;Param5
+        ;m²;°C;m²;m
+    ;;;;
+    01.01.1990 00:00;1;1;2;3
+    01.01.1990 01:00;5;3;4;5
+    01.01.1990 02:00;9;5;6;7
+    01.01.1990 03:00;13;7;8;9
+    01.01.1990 04:00;17;9;10;11
+    01.01.1990 05:00;21;11;12;13
+    """
+
+Option 1: pass rows explicitly to skip rows
+"""""""""""""""""""""""""""""""""""""""""""
+
+.. ipython:: python
+
+    from io import StringIO
+
+    pd.read_csv(
+        StringIO(data),
+        sep=";",
+        skiprows=[11, 12],
+        index_col=0,
+        parse_dates=True,
+        header=10,
+    )
+
+Option 2: read column names and then data
+"""""""""""""""""""""""""""""""""""""""""
+
+.. ipython:: python
+
+    pd.read_csv(StringIO(data), sep=";", header=10, nrows=10).columns
+    columns = pd.read_csv(StringIO(data), sep=";", header=10, nrows=10).columns
+    pd.read_csv(
+        StringIO(data), sep=";", index_col=0, header=12, parse_dates=True, names=columns
+    )
+
+
+.. _cookbook.sql:
+
+SQL
+***
+
+The :ref:`SQL <io.sql>` docs
+
+`Reading from databases with SQL
+<https://stackoverflow.com/questions/10065051/python-pandas-and-databases-like-mysql>`__
+
+.. _cookbook.excel:
+
+Excel
+*****
+
+The :ref:`Excel <io.excel>` docs
+
+`Reading from a filelike handle
+<https://stackoverflow.com/questions/15588713/sheets-of-excel-workbook-from-a-url-into-a-pandas-dataframe>`__
+
+`Modifying formatting in XlsxWriter output
+<https://pbpython.com/improve-pandas-excel-output.html>`__
+
+Loading only visible sheets :issue:`19842#issuecomment-892150745`
+
+.. _cookbook.html:
+
+HTML
+****
+
+`Reading HTML tables from a server that cannot handle the default request
+header <https://stackoverflow.com/a/18939272/564538>`__
+
+.. _cookbook.hdf:
+
+HDFStore
+********
+
+The :ref:`HDFStores <io.hdf5>` docs
+
+`Simple queries with a Timestamp Index
+<https://stackoverflow.com/questions/13926089/selecting-columns-from-pandas-hdfstore-table>`__
+
+Managing heterogeneous data using a linked multiple table hierarchy :issue:`3032`
+
+`Merging on-disk tables with millions of rows
+<https://stackoverflow.com/questions/14614512/merging-two-tables-with-millions-of-rows-in-python/14617925#14617925>`__
+
+`Avoiding inconsistencies when writing to a store from multiple processes/threads
+<https://stackoverflow.com/a/29014295/2858145>`__
+
+De-duplicating a large store by chunks, essentially a recursive reduction operation. Shows a function for taking in data from
+csv file and creating a store by chunks, with date parsing as well.
+`See here
+<https://stackoverflow.com/questions/16110252/need-to-compare-very-large-files-around-1-5gb-in-python/16110391#16110391>`__
+
+`Creating a store chunk-by-chunk from a csv file
+<https://stackoverflow.com/questions/20428355/appending-column-to-frame-of-hdf-file-in-pandas/20428786#20428786>`__
+
+`Appending to a store, while creating a unique index
+<https://stackoverflow.com/questions/16997048/how-does-one-append-large-amounts-of-data-to-a-pandas-hdfstore-and-get-a-natural/16999397#16999397>`__
+
+`Large Data work flows
+<https://stackoverflow.com/q/14262433>`__
+
+`Reading in a sequence of files, then providing a global unique index to a store while appending
+<https://stackoverflow.com/questions/16997048/how-does-one-append-large-amounts-of-data-to-a-pandas-hdfstore-and-get-a-natural>`__
+
+`Groupby on a HDFStore with low group density
+<https://stackoverflow.com/questions/15798209/pandas-group-by-query-on-large-data-in-hdfstore>`__
+
+`Groupby on a HDFStore with high group density
+<https://stackoverflow.com/questions/25459982/trouble-with-grouby-on-millions-of-keys-on-a-chunked-file-in-python-pandas/25471765#25471765>`__
+
+`Hierarchical queries on a HDFStore
+<https://stackoverflow.com/questions/22777284/improve-query-performance-from-a-large-hdfstore-table-with-pandas/22820780#22820780>`__
+
+`Counting with a HDFStore
+<https://stackoverflow.com/questions/20497897/converting-dict-of-dicts-into-pandas-dataframe-memory-issues>`__
+
+`Troubleshoot HDFStore exceptions
+<https://stackoverflow.com/questions/15488809/how-to-trouble-shoot-hdfstore-exception-cannot-find-the-correct-atom-type>`__
+
+`Setting min_itemsize with strings
+<https://stackoverflow.com/questions/15988871/hdfstore-appendstring-dataframe-fails-when-string-column-contents-are-longer>`__
+
+`Using ptrepack to create a completely-sorted-index on a store
+<https://stackoverflow.com/questions/17893370/ptrepack-sortby-needs-full-index>`__
+
+Storing Attributes to a group node
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(8, 3))
+   store = pd.HDFStore("test.h5")
+   store.put("df", df)
+
+   # you can store an arbitrary Python object via pickle
+   store.get_storer("df").attrs.my_attribute = {"A": 10}
+   store.get_storer("df").attrs.my_attribute
+
+.. ipython:: python
+   :suppress:
+
+   store.close()
+   os.remove("test.h5")
+
+You can create or load a HDFStore in-memory  by passing the ``driver``
+parameter to PyTables. Changes are only written to disk when the HDFStore
+is closed.
+
+.. ipython:: python
+
+   store = pd.HDFStore("test.h5", "w", driver="H5FD_CORE")
+
+   df = pd.DataFrame(np.random.randn(8, 3))
+   store["test"] = df
+
+   # only after closing the store, data is written to disk:
+   store.close()
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("test.h5")
+
+.. _cookbook.binary:
+
+Binary files
+************
+
+pandas readily accepts NumPy record arrays, if you need to read in a binary
+file consisting of an array of C structs. For example, given this C program
+in a file called ``main.c`` compiled with ``gcc main.c -std=gnu99`` on a
+64-bit machine,
+
+.. code-block:: c
+
+   #include <stdio.h>
+   #include <stdint.h>
+
+   typedef struct _Data
+   {
+       int32_t count;
+       double avg;
+       float scale;
+   } Data;
+
+   int main(int argc, const char *argv[])
+   {
+       size_t n = 10;
+       Data d[n];
+
+       for (int i = 0; i < n; ++i)
+       {
+           d[i].count = i;
+           d[i].avg = i + 1.0;
+           d[i].scale = (float) i + 2.0f;
+       }
+
+       FILE *file = fopen("binary.dat", "wb");
+       fwrite(&d, sizeof(Data), n, file);
+       fclose(file);
+
+       return 0;
+   }
+
+the following Python code will read the binary file ``'binary.dat'`` into a
+pandas ``DataFrame``, where each element of the struct corresponds to a column
+in the frame:
+
+.. code-block:: python
+
+   names = "count", "avg", "scale"
+
+   # note that the offsets are larger than the size of the type because of
+   # struct padding
+   offsets = 0, 8, 16
+   formats = "i4", "f8", "f4"
+   dt = np.dtype({"names": names, "offsets": offsets, "formats": formats}, align=True)
+   df = pd.DataFrame(np.fromfile("binary.dat", dt))
+
+.. note::
+
+   The offsets of the structure elements may be different depending on the
+   architecture of the machine on which the file was created. Using a raw
+   binary file format like this for general data storage is not recommended, as
+   it is not cross platform. We recommended either HDF5 or parquet, both of
+   which are supported by pandas' IO facilities.
+
+Computation
+-----------
+
+`Numerical integration (sample-based) of a time series
+<https://nbviewer.ipython.org/gist/metakermit/5720498>`__
+
+Correlation
+***********
+
+Often it's useful to obtain the lower (or upper) triangular form of a correlation matrix calculated from :func:`DataFrame.corr`.  This can be achieved by passing a boolean mask to ``where`` as follows:
+
+.. ipython:: python
+
+    df = pd.DataFrame(np.random.random(size=(100, 5)))
+
+    corr_mat = df.corr()
+    mask = np.tril(np.ones_like(corr_mat, dtype=np.bool_), k=-1)
+
+    corr_mat.where(mask)
+
+The ``method`` argument within ``DataFrame.corr`` can accept a callable in addition to the named correlation types.  Here we compute the `distance correlation <https://en.wikipedia.org/wiki/Distance_correlation>`__ matrix for a ``DataFrame`` object.
+
+.. ipython:: python
+
+   def distcorr(x, y):
+       n = len(x)
+       a = np.zeros(shape=(n, n))
+       b = np.zeros(shape=(n, n))
+       for i in range(n):
+           for j in range(i + 1, n):
+               a[i, j] = abs(x[i] - x[j])
+               b[i, j] = abs(y[i] - y[j])
+       a += a.T
+       b += b.T
+       a_bar = np.vstack([np.nanmean(a, axis=0)] * n)
+       b_bar = np.vstack([np.nanmean(b, axis=0)] * n)
+       A = a - a_bar - a_bar.T + np.full(shape=(n, n), fill_value=a_bar.mean())
+       B = b - b_bar - b_bar.T + np.full(shape=(n, n), fill_value=b_bar.mean())
+       cov_ab = np.sqrt(np.nansum(A * B)) / n
+       std_a = np.sqrt(np.sqrt(np.nansum(A ** 2)) / n)
+       std_b = np.sqrt(np.sqrt(np.nansum(B ** 2)) / n)
+       return cov_ab / std_a / std_b
+
+
+   df = pd.DataFrame(np.random.normal(size=(100, 3)))
+   df.corr(method=distcorr)
+
+Timedeltas
+----------
+
+The :ref:`Timedeltas <timedeltas.timedeltas>` docs.
+
+`Using timedeltas
+<https://github.com/pandas-dev/pandas/pull/2899>`__
+
+.. ipython:: python
+
+   import datetime
+
+   s = pd.Series(pd.date_range("2012-1-1", periods=3, freq="D"))
+
+   s - s.max()
+
+   s.max() - s
+
+   s - datetime.datetime(2011, 1, 1, 3, 5)
+
+   s + datetime.timedelta(minutes=5)
+
+   datetime.datetime(2011, 1, 1, 3, 5) - s
+
+   datetime.timedelta(minutes=5) + s
+
+`Adding and subtracting deltas and dates
+<https://stackoverflow.com/questions/16385785/add-days-to-dates-in-dataframe>`__
+
+.. ipython:: python
+
+   deltas = pd.Series([datetime.timedelta(days=i) for i in range(3)])
+
+   df = pd.DataFrame({"A": s, "B": deltas})
+   df
+
+   df["New Dates"] = df["A"] + df["B"]
+
+   df["Delta"] = df["A"] - df["New Dates"]
+   df
+
+   df.dtypes
+
+`Another example
+<https://stackoverflow.com/questions/15683588/iterating-through-a-pandas-dataframe>`__
+
+Values can be set to NaT using np.nan, similar to datetime
+
+.. ipython:: python
+
+   y = s - s.shift()
+   y
+
+   y[1] = np.nan
+   y
+
+Creating example data
+---------------------
+
+To create a dataframe from every combination of some given values, like R's ``expand.grid()``
+function, we can create a dict where the keys are column names and the values are lists
+of the data values:
+
+.. ipython:: python
+
+   def expand_grid(data_dict):
+       rows = itertools.product(*data_dict.values())
+       return pd.DataFrame.from_records(rows, columns=data_dict.keys())
+
+
+   df = expand_grid(
+       {"height": [60, 70], "weight": [100, 140, 180], "sex": ["Male", "Female"]}
+   )
+   df
+
+Constant Series
+---------------
+
+To assess if a series has a constant value, we can check if ``series.nunique() <= 1``.
+However, a more performant approach, that does not count all unique values first, is:
+
+.. ipython:: python
+
+   v = s.to_numpy()
+   is_constant = v.shape[0] == 0 or (s[0] == s).all()
+
+This approach assumes that the series does not contain missing values.
+For the case that we would drop NA values, we can simply remove those values first:
+
+.. ipython:: python
+
+   v = s.dropna().to_numpy()
+   is_constant = v.shape[0] == 0 or (s[0] == s).all()
+
+If missing values are considered distinct from any other value, then one could use:
+
+.. ipython:: python
+
+   v = s.to_numpy()
+   is_constant = v.shape[0] == 0 or (s[0] == s).all() or not pd.notna(v).any()
+
+(Note that this example does not disambiguate between ``np.nan``, ``pd.NA`` and ``None``)
diff --git a/pandas/doc/source/user_guide/copy_on_write.rst b/pandas/doc/source/user_guide/copy_on_write.rst
new file mode 100644
index 0000000000000..90353d9f49f00
--- /dev/null
+++ b/pandas/doc/source/user_guide/copy_on_write.rst
@@ -0,0 +1,357 @@
+.. _copy_on_write:
+
+{{ header }}
+
+*******************
+Copy-on-Write (CoW)
+*******************
+
+.. note::
+
+    Copy-on-Write is now the default with pandas 3.0.
+
+Copy-on-Write was first introduced in version 1.5.0. Starting from version 2.0 most of the
+optimizations that become possible through CoW are implemented and supported. All possible
+optimizations are supported starting from pandas 2.1.
+
+CoW will lead to more predictable behavior since it is not possible to update more than
+one object with one statement, e.g. indexing operations or methods won't have side-effects. Additionally, through
+delaying copies as long as possible, the average performance and memory usage will improve.
+
+Previous behavior
+-----------------
+
+pandas indexing behavior is tricky to understand. Some operations return views while
+other return copies. Depending on the result of the operation, mutating one object
+might accidentally mutate another:
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    In [2]: subset = df["foo"]
+    In [3]: subset.iloc[0] = 100
+    In [4]: df
+    Out[4]:
+       foo  bar
+    0  100    4
+    1    2    5
+    2    3    6
+
+
+Mutating ``subset``, e.g. updating its values, also updated ``df``. The exact behavior was
+hard to predict. Copy-on-Write solves accidentally modifying more than one object,
+it explicitly disallows this. ``df`` is unchanged:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    subset = df["foo"]
+    subset.iloc[0] = 100
+    df
+
+The following sections will explain what this means and how it impacts existing
+applications.
+
+.. _copy_on_write.migration_guide:
+
+Migrating to Copy-on-Write
+--------------------------
+
+Copy-on-Write is the default and only mode in pandas 3.0. This means that users
+need to migrate their code to be compliant with CoW rules.
+
+The default mode in pandas < 3.0 raises warnings for certain cases that will actively
+change behavior and thus change user intended behavior.
+
+pandas 2.2 has a warning mode
+
+.. code-block:: python
+
+    pd.options.mode.copy_on_write = "warn"
+
+that will warn for every operation that will change behavior with CoW. We expect this mode
+to be very noisy, since many cases that we don't expect that they will influence users will
+also emit a warning. We recommend checking this mode and analyzing the warnings, but it is
+not necessary to address all of these warning. The first two items of the following lists
+are the only cases that need to be addressed to make existing code work with CoW.
+
+The following few items describe the user visible changes:
+
+**Chained assignment will never work**
+
+``loc`` should be used as an alternative. Check the
+:ref:`chained assignment section <copy_on_write_chained_assignment>` for more details.
+
+**Accessing the underlying array of a pandas object will return a read-only view**
+
+.. ipython:: python
+
+    ser = pd.Series([1, 2, 3])
+    ser.to_numpy()
+
+This example returns a NumPy array that is a view of the Series object. This view can
+be modified and thus also modify the pandas object. This is not compliant with CoW
+rules. The returned array is set to non-writeable to protect against this behavior.
+Creating a copy of this array allows modification. You can also make the array
+writeable again if you don't care about the pandas object anymore.
+
+See the section about :ref:`read-only NumPy arrays <copy_on_write_read_only_na>`
+for more details.
+
+**Only one pandas object is updated at once**
+
+The following code snippet updated both ``df`` and ``subset`` without CoW:
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    In [2]: subset = df["foo"]
+    In [3]: subset.iloc[0] = 100
+    In [4]: df
+    Out[4]:
+       foo  bar
+    0  100    4
+    1    2    5
+    2    3    6
+
+This is not possible anymore with CoW, since the CoW rules explicitly forbid this.
+This includes updating a single column as a :class:`Series` and relying on the change
+propagating back to the parent :class:`DataFrame`.
+This statement can be rewritten into a single statement with ``loc`` or ``iloc`` if
+this behavior is necessary. :meth:`DataFrame.where` is another suitable alternative
+for this case.
+
+Updating a column selected from a :class:`DataFrame` with an inplace method will
+also not work anymore.
+
+.. ipython:: python
+    :okwarning:
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df["foo"].replace(1, 5, inplace=True)
+    df
+
+This is another form of chained assignment. This can generally be rewritten in 2
+different forms:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df.replace({"foo": {1: 5}}, inplace=True)
+    df
+
+A different alternative would be to not use ``inplace``:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df["foo"] = df["foo"].replace(1, 5)
+    df
+
+**Constructors now copy NumPy arrays by default**
+
+The Series and DataFrame constructors now copies a NumPy array by default when not
+otherwise specified. This was changed to avoid mutating a pandas object when the
+NumPy array is changed inplace outside of pandas. You can set ``copy=False`` to
+avoid this copy.
+
+Description
+-----------
+
+CoW means that any DataFrame or Series derived from another in any way always
+behaves as a copy. As a consequence, we can only change the values of an object
+through modifying the object itself. CoW disallows updating a DataFrame or a Series
+that shares data with another DataFrame or Series object inplace.
+
+This avoids side-effects when modifying values and hence, most methods can avoid
+actually copying the data and only trigger a copy when necessary.
+
+The following example will operate inplace:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df.iloc[0, 0] = 100
+    df
+
+The object ``df`` does not share any data with any other object and hence no
+copy is triggered when updating the values. In contrast, the following operation
+triggers a copy of the data under CoW:
+
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df2 = df.reset_index(drop=True)
+    df2.iloc[0, 0] = 100
+
+    df
+    df2
+
+``reset_index`` returns a lazy copy with CoW while it copies the data without CoW.
+Since both objects, ``df`` and ``df2`` share the same data, a copy is triggered
+when modifying ``df2``. The object ``df`` still has the same values as initially
+while ``df2`` was modified.
+
+If the object ``df`` isn't needed anymore after performing the ``reset_index`` operation,
+you can emulate an inplace-like operation through assigning the output of ``reset_index``
+to the same variable:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df = df.reset_index(drop=True)
+    df.iloc[0, 0] = 100
+    df
+
+The initial object gets out of scope as soon as the result of ``reset_index`` is
+reassigned and hence ``df`` does not share data with any other object. No copy
+is necessary when modifying the object. This is generally true for all methods
+listed in :ref:`Copy-on-Write optimizations <copy_on_write.optimizations>`.
+
+Previously, when operating on views, the view and the parent object was modified:
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    In [2]: subset = df["foo"]
+    In [3]: subset.iloc[0] = 100
+    In [4]: df
+    Out[4]:
+       foo  bar
+    0  100    4
+    1    2    5
+    2    3    6
+
+CoW triggers a copy when ``df`` is changed to avoid mutating ``view`` as well:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    view = df[:]
+    df.iloc[0, 0] = 100
+
+    df
+    view
+
+.. _copy_on_write_chained_assignment:
+
+Chained Assignment
+------------------
+
+Chained assignment references a technique where an object is updated through
+two subsequent indexing operations, e.g.
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    In [2]: df["foo"][df["bar"] > 5] = 100
+    In [3]: df
+    Out[3]:
+       foo  bar
+    0  100    4
+    1    2    5
+    2    3    6
+
+The column ``foo`` was updated where the column ``bar`` is greater than 5.
+This violated the CoW principles though, because it would have to modify the
+view ``df["foo"]`` and ``df`` in one step. Hence, chained assignment will
+consistently never work and raise a ``ChainedAssignmentError`` warning
+with CoW enabled:
+
+.. ipython:: python
+    :okwarning:
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+    df["foo"][df["bar"] > 5] = 100
+
+With copy on write this can be done by using ``loc``.
+
+.. ipython:: python
+
+    df.loc[df["bar"] > 5, "foo"] = 100
+
+.. _copy_on_write_read_only_na:
+
+Read-only NumPy arrays
+----------------------
+
+Accessing the underlying NumPy array of a DataFrame will return a read-only array if the array
+shares data with the initial DataFrame:
+
+The array is a copy if the initial DataFrame consists of more than one array:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2], "b": [1.5, 2.5]})
+    df.to_numpy()
+
+The array shares data with the DataFrame if the DataFrame consists of only one NumPy array:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2], "b": [3, 4]})
+    df.to_numpy()
+
+This array is read-only, which means that it can't be modified inplace:
+
+.. ipython:: python
+    :okexcept:
+
+    arr = df.to_numpy()
+    arr[0, 0] = 100
+
+The same holds true for a Series, since a Series always consists of a single array.
+
+There are two potential solutions to this:
+
+- Trigger a copy manually if you want to avoid updating DataFrames that share memory with your array.
+- Make the array writeable. This is a more performant solution but circumvents Copy-on-Write rules, so
+  it should be used with caution.
+
+.. ipython:: python
+
+    arr = df.to_numpy()
+    arr.flags.writeable = True
+    arr[0, 0] = 100
+    arr
+
+Patterns to avoid
+-----------------
+
+No defensive copy will be performed if two objects share the same data while
+you are modifying one object inplace.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df2 = df.reset_index(drop=True)
+    df2.iloc[0, 0] = 100
+
+This creates two objects that share data and thus the setitem operation will trigger a
+copy. This is not necessary if the initial object ``df`` isn't needed anymore.
+Simply reassigning to the same variable will invalidate the reference that is
+held by the object.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df = df.reset_index(drop=True)
+    df.iloc[0, 0] = 100
+
+No copy is necessary in this example.
+Creating multiple references keeps unnecessary references alive
+and thus will hurt performance with Copy-on-Write.
+
+.. _copy_on_write.optimizations:
+
+Copy-on-Write optimizations
+---------------------------
+
+A new lazy copy mechanism that defers the copy until the object in question is modified
+and only if this object shares data with another object. This mechanism was added to
+methods that don't require a copy of the underlying data. Popular examples are :meth:`DataFrame.drop` for ``axis=1``
+and :meth:`DataFrame.rename`.
+
+These methods return views when Copy-on-Write is enabled, which provides a significant
+performance improvement compared to the regular execution.
diff --git a/pandas/doc/source/user_guide/dsintro.rst b/pandas/doc/source/user_guide/dsintro.rst
new file mode 100644
index 0000000000000..89981786d60b5
--- /dev/null
+++ b/pandas/doc/source/user_guide/dsintro.rst
@@ -0,0 +1,846 @@
+.. _dsintro:
+
+{{ header }}
+
+************************
+Intro to data structures
+************************
+
+We'll start with a quick, non-comprehensive overview of the fundamental data
+structures in pandas to get you started. The fundamental behavior about data
+types, indexing, axis labeling, and alignment apply across all of the
+objects. To get started, import NumPy and load pandas into your namespace:
+
+.. ipython:: python
+
+   import numpy as np
+   import pandas as pd
+
+Fundamentally, **data alignment is intrinsic**. The link
+between labels and data will not be broken unless done so explicitly by you.
+
+We'll give a brief intro to the data structures, then consider all of the broad
+categories of functionality and methods in separate sections.
+
+.. _basics.series:
+
+Series
+------
+
+:class:`Series` is a one-dimensional labeled array capable of holding any data
+type (integers, strings, floating point numbers, Python objects, etc.). The axis
+labels are collectively referred to as the **index**. The basic method to create a :class:`Series` is to call:
+
+.. code-block:: python
+
+   s = pd.Series(data, index=index)
+
+Here, ``data`` can be many different things:
+
+* a Python dict
+* an ndarray
+* a scalar value (like 5)
+
+The passed **index** is a list of axis labels. The constructor's behavior
+depends on **data**'s type:
+
+**From ndarray**
+
+If ``data`` is an ndarray, **index** must be the same length as **data**. If no
+index is passed, one will be created having values ``[0, ..., len(data) - 1]``.
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   s
+   s.index
+
+   pd.Series(np.random.randn(5))
+
+.. note::
+
+    pandas supports non-unique index values. If an operation
+    that does not support duplicate index values is attempted, an exception
+    will be raised at that time.
+
+**From dict**
+
+:class:`Series` can be instantiated from dicts:
+
+.. ipython:: python
+
+   d = {"b": 1, "a": 0, "c": 2}
+   pd.Series(d)
+
+If an index is passed, the values in data corresponding to the labels in the
+index will be pulled out.
+
+.. ipython:: python
+
+   d = {"a": 0.0, "b": 1.0, "c": 2.0}
+   pd.Series(d)
+   pd.Series(d, index=["b", "c", "d", "a"])
+
+.. note::
+
+    NaN (not a number) is the standard missing data marker used in pandas.
+
+**From scalar value**
+
+If ``data`` is a scalar value, the value will be repeated to match
+the length of **index**.  If the **index** is not provided, it defaults
+to ``RangeIndex(1)``.
+
+.. ipython:: python
+
+   pd.Series(5.0, index=["a", "b", "c", "d", "e"])
+
+Series is ndarray-like
+~~~~~~~~~~~~~~~~~~~~~~
+
+:class:`Series` acts very similarly to a :class:`numpy.ndarray` and is a valid argument to most NumPy functions.
+However, operations such as slicing will also slice the index.
+
+.. ipython:: python
+
+    s.iloc[0]
+    s.iloc[:3]
+    s[s > s.median()]
+    s.iloc[[4, 3, 1]]
+    np.exp(s)
+
+.. note::
+
+   We will address array-based indexing like ``s.iloc[[4, 3, 1]]``
+   in the :ref:`section on indexing <indexing>`.
+
+Like a NumPy array, a pandas :class:`Series` has a single :attr:`~Series.dtype`.
+
+.. ipython:: python
+
+   s.dtype
+
+This is often a NumPy dtype. However, pandas and 3rd-party libraries
+extend NumPy's type system in a few places, in which case the dtype would
+be an :class:`~pandas.api.extensions.ExtensionDtype`. Some examples within
+pandas are :ref:`categorical` and :ref:`integer_na`. See :ref:`basics.dtypes`
+for more.
+
+If you need the actual array backing a :class:`Series`, use :attr:`Series.array`.
+
+.. ipython:: python
+
+   s.array
+
+Accessing the array can be useful when you need to do some operation without the
+index (to disable :ref:`automatic alignment <dsintro.alignment>`, for example).
+
+:attr:`Series.array` will always be an :class:`~pandas.api.extensions.ExtensionArray`.
+Briefly, an ExtensionArray is a thin wrapper around one or more *concrete* arrays like a
+:class:`numpy.ndarray`. pandas knows how to take an :class:`~pandas.api.extensions.ExtensionArray` and
+store it in a :class:`Series` or a column of a :class:`DataFrame`.
+See :ref:`basics.dtypes` for more.
+
+While :class:`Series` is ndarray-like, if you need an *actual* ndarray, then use
+:meth:`Series.to_numpy`.
+
+.. ipython:: python
+
+   s.to_numpy()
+
+Even if the :class:`Series` is backed by a :class:`~pandas.api.extensions.ExtensionArray`,
+:meth:`Series.to_numpy` will return a NumPy ndarray.
+
+Series is dict-like
+~~~~~~~~~~~~~~~~~~~
+
+A :class:`Series` is also like a fixed-size dict in that you can get and set values by index
+label:
+
+.. ipython:: python
+
+    s["a"]
+    s["e"] = 12.0
+    s
+    "e" in s
+    "f" in s
+
+If a label is not contained in the index, an exception is raised:
+
+.. ipython:: python
+    :okexcept:
+
+    s["f"]
+
+Using the :meth:`Series.get` method, a missing label will return None or specified default:
+
+.. ipython:: python
+
+   s.get("f")
+
+   s.get("f", np.nan)
+
+These labels can also be accessed by :ref:`attribute<indexing.attribute_access>`.
+
+Vectorized operations and label alignment with Series
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+When working with raw NumPy arrays, looping through value-by-value is usually
+not necessary. The same is true when working with :class:`Series` in pandas.
+:class:`Series` can also be passed into most NumPy methods expecting an ndarray.
+
+.. ipython:: python
+
+    s + s
+    s * 2
+    np.exp(s)
+
+A key difference between :class:`Series` and ndarray is that operations between :class:`Series`
+automatically align the data based on label. Thus, you can write computations
+without giving consideration to whether the :class:`Series` involved have the same
+labels.
+
+.. ipython:: python
+
+    s.iloc[1:] + s.iloc[:-1]
+
+The result of an operation between unaligned :class:`Series` will have the **union** of
+the indexes involved. If a label is not found in one :class:`Series` or the other, the
+result will be marked as missing ``NaN``. Being able to write code without doing
+any explicit data alignment grants immense freedom and flexibility in
+interactive data analysis and research. The integrated data alignment features
+of the pandas data structures set pandas apart from the majority of related
+tools for working with labeled data.
+
+.. note::
+
+    In general, we chose to make the default result of operations between
+    differently indexed objects yield the **union** of the indexes in order to
+    avoid loss of information. Having an index label, though the data is
+    missing, is typically important information as part of a computation. You
+    of course have the option of dropping labels with missing data via the
+    **dropna** function.
+
+Name attribute
+~~~~~~~~~~~~~~
+
+.. _dsintro.name_attribute:
+
+:class:`Series` also has a ``name`` attribute:
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(5), name="something")
+   s
+   s.name
+
+The :class:`Series` ``name`` can be assigned automatically in many cases, in particular,
+when selecting a single column from a :class:`DataFrame`, the ``name`` will be assigned
+the column label.
+
+You can rename a :class:`Series` with the :meth:`pandas.Series.rename` method.
+
+.. ipython:: python
+
+   s2 = s.rename("different")
+   s2.name
+
+Note that ``s`` and ``s2`` refer to different objects.
+
+.. _basics.dataframe:
+
+DataFrame
+---------
+
+:class:`DataFrame` is a 2-dimensional labeled data structure with columns of
+potentially different types. You can think of it like a spreadsheet or SQL
+table, or a dict of Series objects. It is generally the most commonly used
+pandas object. Like Series, DataFrame accepts many different kinds of input:
+
+* Dict of 1D ndarrays, lists, dicts, or :class:`Series`
+* 2-D numpy.ndarray
+* `Structured or record
+  <https://numpy.org/doc/stable/user/basics.rec.html>`__ ndarray
+* A :class:`Series`
+* Another :class:`DataFrame`
+
+Along with the data, you can optionally pass **index** (row labels) and
+**columns** (column labels) arguments. If you pass an index and / or columns,
+you are guaranteeing the index and / or columns of the resulting
+DataFrame. Thus, a dict of Series plus a specific index will discard all data
+not matching up to the passed index.
+
+If axis labels are not passed, they will be constructed from the input data
+based on common sense rules.
+
+From dict of Series or dicts
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The resulting **index** will be the **union** of the indexes of the various
+Series. If there are any nested dicts, these will first be converted to
+Series. If no columns are passed, the columns will be the ordered list of dict
+keys.
+
+.. ipython:: python
+
+    d = {
+        "one": pd.Series([1.0, 2.0, 3.0], index=["a", "b", "c"]),
+        "two": pd.Series([1.0, 2.0, 3.0, 4.0], index=["a", "b", "c", "d"]),
+    }
+    df = pd.DataFrame(d)
+    df
+
+    pd.DataFrame(d, index=["d", "b", "a"])
+    pd.DataFrame(d, index=["d", "b", "a"], columns=["two", "three"])
+
+The row and column labels can be accessed respectively by accessing the
+**index** and **columns** attributes:
+
+.. note::
+
+   When a particular set of columns is passed along with a dict of data, the
+   passed columns override the keys in the dict.
+
+.. ipython:: python
+
+   df.index
+   df.columns
+
+From dict of ndarrays / lists
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+All ndarrays must share the same length. If an index is passed, it must
+also be the same length as the arrays. If no index is passed, the
+result will be ``range(n)``, where ``n`` is the array length.
+
+.. ipython:: python
+
+   d = {"one": [1.0, 2.0, 3.0, 4.0], "two": [4.0, 3.0, 2.0, 1.0]}
+   pd.DataFrame(d)
+   pd.DataFrame(d, index=["a", "b", "c", "d"])
+
+From structured or record array
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+This case is handled identically to a dict of arrays.
+
+.. ipython:: python
+
+   data = np.zeros((2,), dtype=[("A", "i4"), ("B", "f4"), ("C", "S10")])
+   data[:] = [(1, 2.0, "Hello"), (2, 3.0, "World")]
+
+   pd.DataFrame(data)
+   pd.DataFrame(data, index=["first", "second"])
+   pd.DataFrame(data, columns=["C", "A", "B"])
+
+.. note::
+
+    DataFrame is not intended to work exactly like a 2-dimensional NumPy
+    ndarray.
+
+.. _basics.dataframe.from_list_of_dicts:
+
+From a list of dicts
+~~~~~~~~~~~~~~~~~~~~
+
+.. ipython:: python
+
+   data2 = [{"a": 1, "b": 2}, {"a": 5, "b": 10, "c": 20}]
+   pd.DataFrame(data2)
+   pd.DataFrame(data2, index=["first", "second"])
+   pd.DataFrame(data2, columns=["a", "b"])
+
+.. _basics.dataframe.from_dict_of_tuples:
+
+From a dict of tuples
+~~~~~~~~~~~~~~~~~~~~~
+
+You can automatically create a MultiIndexed frame by passing a tuples
+dictionary.
+
+.. ipython:: python
+
+   pd.DataFrame(
+       {
+           ("a", "b"): {("A", "B"): 1, ("A", "C"): 2},
+           ("a", "a"): {("A", "C"): 3, ("A", "B"): 4},
+           ("a", "c"): {("A", "B"): 5, ("A", "C"): 6},
+           ("b", "a"): {("A", "C"): 7, ("A", "B"): 8},
+           ("b", "b"): {("A", "D"): 9, ("A", "B"): 10},
+       }
+   )
+
+.. _basics.dataframe.from_series:
+
+From a Series
+~~~~~~~~~~~~~
+
+The result will be a DataFrame with the same index as the input Series, and
+with one column whose name is the original name of the Series (only if no other
+column name provided).
+
+.. ipython:: python
+
+   ser = pd.Series(range(3), index=list("abc"), name="ser")
+   pd.DataFrame(ser)
+
+.. _basics.dataframe.from_list_namedtuples:
+
+From a list of namedtuples
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The field names of the first ``namedtuple`` in the list determine the columns
+of the :class:`DataFrame`. The remaining namedtuples (or tuples) are simply unpacked
+and their values are fed into the rows of the :class:`DataFrame`. If any of those
+tuples is shorter than the first ``namedtuple`` then the later columns in the
+corresponding row are marked as missing values. If any are longer than the
+first ``namedtuple``, a ``ValueError`` is raised.
+
+.. ipython:: python
+
+    from collections import namedtuple
+
+    Point = namedtuple("Point", "x y")
+
+    pd.DataFrame([Point(0, 0), Point(0, 3), (2, 3)])
+
+    Point3D = namedtuple("Point3D", "x y z")
+
+    pd.DataFrame([Point3D(0, 0, 0), Point3D(0, 3, 5), Point(2, 3)])
+
+
+.. _basics.dataframe.from_list_dataclasses:
+
+From a list of dataclasses
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Data Classes as introduced in `PEP557 <https://www.python.org/dev/peps/pep-0557>`__,
+can be passed into the DataFrame constructor.
+Passing a list of dataclasses is equivalent to passing a list of dictionaries.
+
+Please be aware, that all values in the list should be dataclasses, mixing
+types in the list would result in a ``TypeError``.
+
+.. ipython:: python
+
+    from dataclasses import make_dataclass
+
+    Point = make_dataclass("Point", [("x", int), ("y", int)])
+
+    pd.DataFrame([Point(0, 0), Point(0, 3), Point(2, 3)])
+
+**Missing data**
+
+To construct a DataFrame with missing data, we use ``np.nan`` to
+represent missing values. Alternatively, you may pass a ``numpy.MaskedArray``
+as the data argument to the DataFrame constructor, and its masked entries will
+be considered missing. See :ref:`Missing data <missing_data>` for more.
+
+Alternate constructors
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. _basics.dataframe.from_dict:
+
+**DataFrame.from_dict**
+
+:meth:`DataFrame.from_dict` takes a dict of dicts or a dict of array-like sequences
+and returns a DataFrame. It operates like the :class:`DataFrame` constructor except
+for the ``orient`` parameter which is ``'columns'`` by default, but which can be
+set to ``'index'`` in order to use the dict keys as row labels.
+
+
+.. ipython:: python
+
+   pd.DataFrame.from_dict(dict([("A", [1, 2, 3]), ("B", [4, 5, 6])]))
+
+If you pass ``orient='index'``, the keys will be the row labels. In this
+case, you can also pass the desired column names:
+
+.. ipython:: python
+
+   pd.DataFrame.from_dict(
+       dict([("A", [1, 2, 3]), ("B", [4, 5, 6])]),
+       orient="index",
+       columns=["one", "two", "three"],
+   )
+
+.. _basics.dataframe.from_records:
+
+**DataFrame.from_records**
+
+:meth:`DataFrame.from_records` takes a list of tuples or an ndarray with structured
+dtype. It works analogously to the normal :class:`DataFrame` constructor, except that
+the resulting DataFrame index may be a specific field of the structured
+dtype.
+
+.. ipython:: python
+
+   data
+   pd.DataFrame.from_records(data, index="C")
+
+.. _basics.dataframe.sel_add_del:
+
+Column selection, addition, deletion
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can treat a :class:`DataFrame` semantically like a dict of like-indexed :class:`Series`
+objects. Getting, setting, and deleting columns works with the same syntax as
+the analogous dict operations:
+
+.. ipython:: python
+
+   df["one"]
+   df["three"] = df["one"] * df["two"]
+   df["flag"] = df["one"] > 2
+   df
+
+Columns can be deleted or popped like with a dict:
+
+.. ipython:: python
+
+   del df["two"]
+   three = df.pop("three")
+   df
+
+When inserting a scalar value, it will naturally be propagated to fill the
+column:
+
+.. ipython:: python
+
+   df["foo"] = "bar"
+   df
+
+When inserting a :class:`Series` that does not have the same index as the :class:`DataFrame`, it
+will be conformed to the DataFrame's index:
+
+.. ipython:: python
+
+   df["one_trunc"] = df["one"][:2]
+   df
+
+You can insert raw ndarrays but their length must match the length of the
+DataFrame's index.
+
+By default, columns get inserted at the end. :meth:`DataFrame.insert`
+inserts at a particular location in the columns:
+
+.. ipython:: python
+
+   df.insert(1, "bar", df["one"])
+   df
+
+.. _dsintro.chained_assignment:
+
+Assigning new columns in method chains
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Inspired by `dplyr's
+<https://dplyr.tidyverse.org/reference/mutate.html>`__
+``mutate`` verb, DataFrame has an :meth:`~pandas.DataFrame.assign`
+method that allows you to easily create new columns that are potentially
+derived from existing columns.
+
+.. ipython:: python
+
+   iris = pd.read_csv("data/iris.data")
+   iris.head()
+   iris.assign(sepal_ratio=iris["SepalWidth"] / iris["SepalLength"]).head()
+
+In the example above, we inserted a precomputed value. We can also pass in
+a function of one argument to be evaluated on the DataFrame being assigned to.
+
+.. ipython:: python
+
+   iris.assign(sepal_ratio=lambda x: (x["SepalWidth"] / x["SepalLength"])).head()
+
+:meth:`~pandas.DataFrame.assign` **always** returns a copy of the data, leaving the original
+DataFrame untouched.
+
+Passing a callable, as opposed to an actual value to be inserted, is
+useful when you don't have a reference to the DataFrame at hand. This is
+common when using :meth:`~pandas.DataFrame.assign` in a chain of operations. For example,
+we can limit the DataFrame to just those observations with a Sepal Length
+greater than 5, calculate the ratio, and plot:
+
+.. ipython:: python
+
+   @savefig basics_assign.png
+   (
+       iris.query("SepalLength > 5")
+       .assign(
+           SepalRatio=lambda x: x.SepalWidth / x.SepalLength,
+           PetalRatio=lambda x: x.PetalWidth / x.PetalLength,
+       )
+       .plot(kind="scatter", x="SepalRatio", y="PetalRatio")
+   )
+
+Since a function is passed in, the function is computed on the DataFrame
+being assigned to. Importantly, this is the DataFrame that's been filtered
+to those rows with sepal length greater than 5. The filtering happens first,
+and then the ratio calculations. This is an example where we didn't
+have a reference to the *filtered* DataFrame available.
+
+The function signature for :meth:`~pandas.DataFrame.assign` is simply ``**kwargs``. The keys
+are the column names for the new fields, and the values are either a value
+to be inserted (for example, a :class:`Series` or NumPy array), or a function
+of one argument to be called on the :class:`DataFrame`. A *copy* of the original
+:class:`DataFrame` is returned, with the new values inserted.
+
+The order of ``**kwargs`` is preserved. This allows
+for *dependent* assignment, where an expression later in ``**kwargs`` can refer
+to a column created earlier in the same :meth:`~DataFrame.assign`.
+
+.. ipython:: python
+
+   dfa = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+   dfa.assign(C=lambda x: x["A"] + x["B"], D=lambda x: x["A"] + x["C"])
+
+In the second expression, ``x['C']`` will refer to the newly created column,
+that's equal to ``dfa['A'] + dfa['B']``.
+
+
+Indexing / selection
+~~~~~~~~~~~~~~~~~~~~
+The basics of indexing are as follows:
+
+.. csv-table::
+    :header: "Operation", "Syntax", "Result"
+    :widths: 30, 20, 10
+
+    Select column, ``df[col]``, Series
+    Select row by label, ``df.loc[label]``, Series
+    Select row by integer location, ``df.iloc[loc]``, Series
+    Slice rows, ``df[5:10]``, DataFrame
+    Select rows by boolean vector, ``df[bool_vec]``, DataFrame
+
+Row selection, for example, returns a :class:`Series` whose index is the columns of the
+:class:`DataFrame`:
+
+.. ipython:: python
+
+   df.loc["b"]
+   df.iloc[2]
+
+For a more exhaustive treatment of sophisticated label-based indexing and
+slicing, see the :ref:`section on indexing <indexing>`. We will address the
+fundamentals of reindexing / conforming to new sets of labels in the
+:ref:`section on reindexing <basics.reindexing>`.
+
+.. _dsintro.alignment:
+
+Data alignment and arithmetic
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Data alignment between :class:`DataFrame` objects automatically align on **both the
+columns and the index (row labels)**. Again, the resulting object will have the
+union of the column and row labels.
+
+.. ipython:: python
+
+    df = pd.DataFrame(np.random.randn(10, 4), columns=["A", "B", "C", "D"])
+    df2 = pd.DataFrame(np.random.randn(7, 3), columns=["A", "B", "C"])
+    df + df2
+
+When doing an operation between :class:`DataFrame` and :class:`Series`, the default behavior is
+to align the :class:`Series` **index** on the :class:`DataFrame` **columns**, thus `broadcasting
+<https://numpy.org/doc/stable/user/basics.broadcasting.html>`__
+row-wise. For example:
+
+.. ipython:: python
+
+   df - df.iloc[0]
+
+For explicit control over the matching and broadcasting behavior, see the
+section on :ref:`flexible binary operations <basics.binop>`.
+
+Arithmetic operations with scalars operate element-wise:
+
+.. ipython:: python
+
+   df * 5 + 2
+   1 / df
+   df ** 4
+
+.. _dsintro.boolean:
+
+Boolean operators operate element-wise as well:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"a": [1, 0, 1], "b": [0, 1, 1]}, dtype=bool)
+   df2 = pd.DataFrame({"a": [0, 1, 1], "b": [1, 1, 0]}, dtype=bool)
+   df1 & df2
+   df1 | df2
+   df1 ^ df2
+   -df1
+
+Transposing
+~~~~~~~~~~~
+
+To transpose, access the ``T`` attribute or :meth:`DataFrame.transpose`,
+similar to an ndarray:
+
+.. ipython:: python
+
+   # only show the first 5 rows
+   df[:5].T
+
+.. _dsintro.numpy_interop:
+
+DataFrame interoperability with NumPy functions
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Most NumPy functions can be called directly on :class:`Series` and :class:`DataFrame`.
+
+.. ipython:: python
+
+   np.exp(df)
+   np.asarray(df)
+
+:class:`DataFrame` is not intended to be a drop-in replacement for ndarray as its
+indexing semantics and data model are quite different in places from an n-dimensional
+array.
+
+:class:`Series` implements ``__array_ufunc__``, which allows it to work with NumPy's
+`universal functions <https://numpy.org/doc/stable/reference/ufuncs.html>`_.
+
+The ufunc is applied to the underlying array in a :class:`Series`.
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3, 4])
+   np.exp(ser)
+
+When multiple :class:`Series` are passed to a ufunc, they are aligned before
+performing the operation.
+
+Like other parts of the library, pandas will automatically align labeled inputs
+as part of a ufunc with multiple inputs. For example, using :meth:`numpy.remainder`
+on two :class:`Series` with differently ordered labels will align before the operation.
+
+.. ipython:: python
+
+   ser1 = pd.Series([1, 2, 3], index=["a", "b", "c"])
+   ser2 = pd.Series([1, 3, 5], index=["b", "a", "c"])
+   ser1
+   ser2
+   np.remainder(ser1, ser2)
+
+As usual, the union of the two indices is taken, and non-overlapping values are filled
+with missing values.
+
+.. ipython:: python
+
+   ser3 = pd.Series([2, 4, 6], index=["b", "c", "d"])
+   ser3
+   np.remainder(ser1, ser3)
+
+When a binary ufunc is applied to a :class:`Series` and :class:`Index`, the :class:`Series`
+implementation takes precedence and a :class:`Series` is returned.
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3])
+   idx = pd.Index([4, 5, 6])
+
+   np.maximum(ser, idx)
+
+NumPy ufuncs are safe to apply to :class:`Series` backed by non-ndarray arrays,
+for example :class:`arrays.SparseArray` (see :ref:`sparse.calculation`). If possible,
+the ufunc is applied without converting the underlying data to an ndarray.
+
+Console display
+~~~~~~~~~~~~~~~
+
+A very large :class:`DataFrame` will be truncated to display them in the console.
+You can also get a summary using :meth:`~pandas.DataFrame.info`.
+(The **baseball** dataset is from the **plyr** R package):
+
+.. ipython:: python
+   :suppress:
+
+   # force a summary to be printed
+   pd.set_option("display.max_rows", 5)
+
+.. ipython:: python
+
+   baseball = pd.read_csv("data/baseball.csv")
+   print(baseball)
+   baseball.info()
+
+.. ipython:: python
+   :suppress:
+   :okwarning:
+
+   # restore GlobalPrintConfig
+   pd.reset_option(r"^display\.")
+
+However, using :meth:`DataFrame.to_string` will return a string representation of the
+:class:`DataFrame` in tabular form, though it won't always fit the console width:
+
+.. ipython:: python
+
+   print(baseball.iloc[-20:, :12].to_string())
+
+Wide DataFrames will be printed across multiple rows by
+default:
+
+.. ipython:: python
+
+   pd.DataFrame(np.random.randn(3, 12))
+
+You can change how much to print on a single row by setting the ``display.width``
+option:
+
+.. ipython:: python
+
+   pd.set_option("display.width", 40)  # default is 80
+
+   pd.DataFrame(np.random.randn(3, 12))
+
+You can adjust the max width of the individual columns by setting ``display.max_colwidth``
+
+.. ipython:: python
+
+   datafile = {
+       "filename": ["filename_01", "filename_02"],
+       "path": [
+           "media/user_name/storage/folder_01/filename_01",
+           "media/user_name/storage/folder_02/filename_02",
+       ],
+   }
+
+   pd.set_option("display.max_colwidth", 30)
+   pd.DataFrame(datafile)
+
+   pd.set_option("display.max_colwidth", 100)
+   pd.DataFrame(datafile)
+
+.. ipython:: python
+   :suppress:
+
+   pd.reset_option("display.width")
+   pd.reset_option("display.max_colwidth")
+
+You can also disable this feature via the ``expand_frame_repr`` option.
+This will print the table in one block.
+
+DataFrame column attribute access and IPython completion
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If a :class:`DataFrame` column label is a valid Python variable name, the column can be
+accessed like an attribute:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"foo1": np.random.randn(5), "foo2": np.random.randn(5)})
+   df
+   df.foo1
+
+The columns are also connected to the `IPython <https://ipython.org>`__
+completion mechanism so they can be tab-completed:
+
+.. code-block:: ipython
+
+    In [5]: df.foo<TAB>  # noqa: E225, E999
+    df.foo1  df.foo2
diff --git a/pandas/doc/source/user_guide/duplicates.rst b/pandas/doc/source/user_guide/duplicates.rst
new file mode 100644
index 0000000000000..7894789846ce8
--- /dev/null
+++ b/pandas/doc/source/user_guide/duplicates.rst
@@ -0,0 +1,206 @@
+.. _duplicates:
+
+****************
+Duplicate Labels
+****************
+
+:class:`Index` objects are not required to be unique; you can have duplicate row
+or column labels. This may be a bit confusing at first. If you're familiar with
+SQL, you know that row labels are similar to a primary key on a table, and you
+would never want duplicates in a SQL table. But one of pandas' roles is to clean
+messy, real-world data before it goes to some downstream system. And real-world
+data has duplicates, even in fields that are supposed to be unique.
+
+This section describes how duplicate labels change the behavior of certain
+operations, and how prevent duplicates from arising during operations, or to
+detect them if they do.
+
+.. ipython:: python
+
+   import pandas as pd
+   import numpy as np
+
+Consequences of Duplicate Labels
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some pandas methods (:meth:`Series.reindex` for example) just don't work with
+duplicates present. The output can't be determined, and so pandas raises.
+
+.. ipython:: python
+   :okexcept:
+   :okwarning:
+
+   s1 = pd.Series([0, 1, 2], index=["a", "b", "b"])
+   s1.reindex(["a", "b", "c"])
+
+Other methods, like indexing, can give very surprising results. Typically
+indexing with a scalar will *reduce dimensionality*. Slicing a ``DataFrame``
+with a scalar will return a ``Series``. Slicing a ``Series`` with a scalar will
+return a scalar. But with duplicates, this isn't the case.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame([[0, 1, 2], [3, 4, 5]], columns=["A", "A", "B"])
+   df1
+
+We have duplicates in the columns. If we slice ``'B'``, we get back a ``Series``
+
+.. ipython:: python
+
+   df1["B"]  # a series
+
+But slicing ``'A'`` returns a ``DataFrame``
+
+
+.. ipython:: python
+
+   df1["A"]  # a DataFrame
+
+This applies to row labels as well
+
+.. ipython:: python
+
+   df2 = pd.DataFrame({"A": [0, 1, 2]}, index=["a", "a", "b"])
+   df2
+   df2.loc["b", "A"]  # a scalar
+   df2.loc["a", "A"]  # a Series
+
+Duplicate Label Detection
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can check whether an :class:`Index` (storing the row or column labels) is
+unique with :attr:`Index.is_unique`:
+
+.. ipython:: python
+
+   df2
+   df2.index.is_unique
+   df2.columns.is_unique
+
+.. note::
+
+   Checking whether an index is unique is somewhat expensive for large datasets.
+   pandas does cache this result, so re-checking on the same index is very fast.
+
+:meth:`Index.duplicated` will return a boolean ndarray indicating whether a
+label is repeated.
+
+.. ipython:: python
+
+   df2.index.duplicated()
+
+Which can be used as a boolean filter to drop duplicate rows.
+
+.. ipython:: python
+
+   df2.loc[~df2.index.duplicated(), :]
+
+If you need additional logic to handle duplicate labels, rather than just
+dropping the repeats, using :meth:`~DataFrame.groupby` on the index is a common
+trick. For example, we'll resolve duplicates by taking the average of all rows
+with the same label.
+
+.. ipython:: python
+
+   df2.groupby(level=0).mean()
+
+.. _duplicates.disallow:
+
+Disallowing Duplicate Labels
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. versionadded:: 1.2.0
+
+As noted above, handling duplicates is an important feature when reading in raw
+data. That said, you may want to avoid introducing duplicates as part of a data
+processing pipeline (from methods like :meth:`pandas.concat`,
+:meth:`~DataFrame.rename`, etc.). Both :class:`Series` and :class:`DataFrame`
+*disallow* duplicate labels by calling ``.set_flags(allows_duplicate_labels=False)``.
+(the default is to allow them). If there are duplicate labels, an exception
+will be raised.
+
+.. ipython:: python
+   :okexcept:
+
+   pd.Series([0, 1, 2], index=["a", "b", "b"]).set_flags(allows_duplicate_labels=False)
+
+This applies to both row and column labels for a :class:`DataFrame`
+
+.. ipython:: python
+   :okexcept:
+
+   pd.DataFrame([[0, 1, 2], [3, 4, 5]], columns=["A", "B", "C"],).set_flags(
+       allows_duplicate_labels=False
+   )
+
+This attribute can be checked or set with :attr:`~DataFrame.flags.allows_duplicate_labels`,
+which indicates whether that object can have duplicate labels.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [0, 1, 2, 3]}, index=["x", "y", "X", "Y"]).set_flags(
+       allows_duplicate_labels=False
+   )
+   df
+   df.flags.allows_duplicate_labels
+
+:meth:`DataFrame.set_flags` can be used to return a new ``DataFrame`` with attributes
+like ``allows_duplicate_labels`` set to some value
+
+.. ipython:: python
+
+   df2 = df.set_flags(allows_duplicate_labels=True)
+   df2.flags.allows_duplicate_labels
+
+The new ``DataFrame`` returned is a view on the same data as the old ``DataFrame``.
+Or the property can just be set directly on the same object
+
+
+.. ipython:: python
+
+   df2.flags.allows_duplicate_labels = False
+   df2.flags.allows_duplicate_labels
+
+When processing raw, messy data you might initially read in the messy data
+(which potentially has duplicate labels), deduplicate, and then disallow duplicates
+going forward, to ensure that your data pipeline doesn't introduce duplicates.
+
+
+.. code-block:: python
+
+   >>> raw = pd.read_csv("...")
+   >>> deduplicated = raw.groupby(level=0).first()  # remove duplicates
+   >>> deduplicated.flags.allows_duplicate_labels = False  # disallow going forward
+
+Setting ``allows_duplicate_labels=False`` on a ``Series`` or ``DataFrame`` with duplicate
+labels or performing an operation that introduces duplicate labels on a ``Series`` or
+``DataFrame`` that disallows duplicates will raise an
+:class:`errors.DuplicateLabelError`.
+
+.. ipython:: python
+   :okexcept:
+
+   df.rename(str.upper)
+
+This error message contains the labels that are duplicated, and the numeric positions
+of all the duplicates (including the "original") in the ``Series`` or ``DataFrame``
+
+Duplicate Label Propagation
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In general, disallowing duplicates is "sticky". It's preserved through
+operations.
+
+.. ipython:: python
+   :okexcept:
+
+   s1 = pd.Series(0, index=["a", "b"]).set_flags(allows_duplicate_labels=False)
+   s1
+   s1.head().rename({"a": "b"})
+
+.. warning::
+
+   This is an experimental feature. Currently, many methods fail to
+   propagate the ``allows_duplicate_labels`` value. In future versions
+   it is expected that every method taking or returning one or more
+   DataFrame or Series objects will propagate ``allows_duplicate_labels``.
diff --git a/pandas/doc/source/user_guide/enhancingperf.rst b/pandas/doc/source/user_guide/enhancingperf.rst
new file mode 100644
index 0000000000000..e55a6cda47ac2
--- /dev/null
+++ b/pandas/doc/source/user_guide/enhancingperf.rst
@@ -0,0 +1,770 @@
+.. _enhancingperf:
+
+{{ header }}
+
+*********************
+Enhancing performance
+*********************
+
+In this part of the tutorial, we will investigate how to speed up certain
+functions operating on pandas :class:`DataFrame` using Cython, Numba and :func:`pandas.eval`.
+Generally, using Cython and Numba can offer a larger speedup than using :func:`pandas.eval`
+but will require a lot more code.
+
+.. note::
+
+   In addition to following the steps in this tutorial, users interested in enhancing
+   performance are highly encouraged to install the
+   :ref:`recommended dependencies<install.recommended_dependencies>` for pandas.
+   These dependencies are often not installed by default, but will offer speed
+   improvements if present.
+
+.. _enhancingperf.cython:
+
+Cython (writing C extensions for pandas)
+----------------------------------------
+
+For many use cases writing pandas in pure Python and NumPy is sufficient. In some
+computationally heavy applications however, it can be possible to achieve sizable
+speed-ups by offloading work to `cython <https://cython.org/>`__.
+
+This tutorial assumes you have refactored as much as possible in Python, for example
+by trying to remove for-loops and making use of NumPy vectorization. It's always worth
+optimising in Python first.
+
+This tutorial walks through a "typical" process of cythonizing a slow computation.
+We use an `example from the Cython documentation <https://docs.cython.org/en/latest/src/quickstart/cythonize.html>`__
+but in the context of pandas. Our final cythonized solution is around 100 times
+faster than the pure Python solution.
+
+.. _enhancingperf.pure:
+
+Pure Python
+~~~~~~~~~~~
+
+We have a :class:`DataFrame` to which we want to apply a function row-wise.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "a": np.random.randn(1000),
+           "b": np.random.randn(1000),
+           "N": np.random.randint(100, 1000, (1000)),
+           "x": "x",
+       }
+   )
+   df
+
+Here's the function in pure Python:
+
+.. ipython:: python
+
+   def f(x):
+       return x * (x - 1)
+
+
+   def integrate_f(a, b, N):
+       s = 0
+       dx = (b - a) / N
+       for i in range(N):
+           s += f(a + i * dx)
+       return s * dx
+
+We achieve our result by using :meth:`DataFrame.apply` (row-wise):
+
+.. ipython:: python
+
+   %timeit df.apply(lambda x: integrate_f(x["a"], x["b"], x["N"]), axis=1)
+
+Let's take a look and see where the time is spent during this operation
+using the `prun ipython magic function <https://ipython.readthedocs.io/en/stable/interactive/magics.html#magic-prun>`__:
+
+.. ipython:: python
+
+   # most time consuming 4 calls
+   %prun -l 4 df.apply(lambda x: integrate_f(x["a"], x["b"], x["N"]), axis=1)  # noqa E999
+
+By far the majority of time is spend inside either ``integrate_f`` or ``f``,
+hence we'll concentrate our efforts cythonizing these two functions.
+
+.. _enhancingperf.plain:
+
+Plain Cython
+~~~~~~~~~~~~
+
+First we're going to need to import the Cython magic function to IPython:
+
+.. ipython:: python
+   :okwarning:
+
+   %load_ext Cython
+
+
+Now, let's simply copy our functions over to Cython:
+
+.. ipython::
+
+   In [2]: %%cython
+      ...: def f_plain(x):
+      ...:     return x * (x - 1)
+      ...: def integrate_f_plain(a, b, N):
+      ...:     s = 0
+      ...:     dx = (b - a) / N
+      ...:     for i in range(N):
+      ...:         s += f_plain(a + i * dx)
+      ...:     return s * dx
+      ...:
+
+
+.. ipython:: python
+
+   %timeit df.apply(lambda x: integrate_f_plain(x["a"], x["b"], x["N"]), axis=1)
+
+This has improved the performance compared to the pure Python approach by one-third.
+
+.. _enhancingperf.type:
+
+Declaring C types
+~~~~~~~~~~~~~~~~~
+
+We can annotate the function variables and return types as well as use ``cdef``
+and ``cpdef`` to improve performance:
+
+.. ipython::
+
+   In [3]: %%cython
+      ...: cdef double f_typed(double x) except? -2:
+      ...:     return x * (x - 1)
+      ...: cpdef double integrate_f_typed(double a, double b, int N):
+      ...:     cdef int i
+      ...:     cdef double s, dx
+      ...:     s = 0
+      ...:     dx = (b - a) / N
+      ...:     for i in range(N):
+      ...:         s += f_typed(a + i * dx)
+      ...:     return s * dx
+      ...:
+
+.. ipython:: python
+
+   %timeit df.apply(lambda x: integrate_f_typed(x["a"], x["b"], x["N"]), axis=1)
+
+Annotating the functions with C types yields an over ten times performance improvement compared to
+the original Python implementation.
+
+.. _enhancingperf.ndarray:
+
+Using ndarray
+~~~~~~~~~~~~~
+
+When re-profiling, time is spent creating a :class:`Series` from each row, and calling ``__getitem__`` from both
+the index and the series (three times for each row). These Python function calls are expensive and
+can be improved by passing an ``np.ndarray``.
+
+.. ipython:: python
+
+   %prun -l 4 df.apply(lambda x: integrate_f_typed(x["a"], x["b"], x["N"]), axis=1)
+
+.. ipython::
+
+   In [4]: %%cython
+      ...: cimport numpy as np
+      ...: import numpy as np
+      ...: cdef double f_typed(double x) except? -2:
+      ...:     return x * (x - 1)
+      ...: cpdef double integrate_f_typed(double a, double b, int N):
+      ...:     cdef int i
+      ...:     cdef double s, dx
+      ...:     s = 0
+      ...:     dx = (b - a) / N
+      ...:     for i in range(N):
+      ...:         s += f_typed(a + i * dx)
+      ...:     return s * dx
+      ...: cpdef np.ndarray[double] apply_integrate_f(np.ndarray col_a, np.ndarray col_b,
+      ...:                                            np.ndarray col_N):
+      ...:     assert (col_a.dtype == np.float64
+      ...:             and col_b.dtype == np.float64 and col_N.dtype == np.dtype(int))
+      ...:     cdef Py_ssize_t i, n = len(col_N)
+      ...:     assert (len(col_a) == len(col_b) == n)
+      ...:     cdef np.ndarray[double] res = np.empty(n)
+      ...:     for i in range(len(col_a)):
+      ...:         res[i] = integrate_f_typed(col_a[i], col_b[i], col_N[i])
+      ...:     return res
+      ...:
+
+
+This implementation creates an array of zeros and inserts the result
+of ``integrate_f_typed`` applied over each row. Looping over an ``ndarray`` is faster
+in Cython than looping over a :class:`Series` object.
+
+Since ``apply_integrate_f`` is typed to accept an ``np.ndarray``, :meth:`Series.to_numpy`
+calls are needed to utilize this function.
+
+.. ipython:: python
+
+   %timeit apply_integrate_f(df["a"].to_numpy(), df["b"].to_numpy(), df["N"].to_numpy())
+
+Performance has improved from the prior implementation by almost ten times.
+
+.. _enhancingperf.boundswrap:
+
+Disabling compiler directives
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The majority of the time is now spent in ``apply_integrate_f``. Disabling Cython's ``boundscheck``
+and ``wraparound`` checks can yield more performance.
+
+.. ipython:: python
+
+   %prun -l 4 apply_integrate_f(df["a"].to_numpy(), df["b"].to_numpy(), df["N"].to_numpy())
+
+.. ipython::
+
+   In [5]: %%cython
+      ...: cimport cython
+      ...: cimport numpy as np
+      ...: import numpy as np
+      ...: cdef np.float64_t f_typed(np.float64_t x) except? -2:
+      ...:     return x * (x - 1)
+      ...: cpdef np.float64_t integrate_f_typed(np.float64_t a, np.float64_t b, np.int64_t N):
+      ...:     cdef np.int64_t i
+      ...:     cdef np.float64_t s = 0.0, dx
+      ...:     dx = (b - a) / N
+      ...:     for i in range(N):
+      ...:         s += f_typed(a + i * dx)
+      ...:     return s * dx
+      ...: @cython.boundscheck(False)
+      ...: @cython.wraparound(False)
+      ...: cpdef np.ndarray[np.float64_t] apply_integrate_f_wrap(
+      ...:     np.ndarray[np.float64_t] col_a,
+      ...:     np.ndarray[np.float64_t] col_b,
+      ...:     np.ndarray[np.int64_t] col_N
+      ...: ):
+      ...:     cdef np.int64_t i, n = len(col_N)
+      ...:     assert len(col_a) == len(col_b) == n
+      ...:     cdef np.ndarray[np.float64_t] res = np.empty(n, dtype=np.float64)
+      ...:     for i in range(n):
+      ...:         res[i] = integrate_f_typed(col_a[i], col_b[i], col_N[i])
+      ...:     return res
+      ...:
+
+.. ipython:: python
+
+   %timeit apply_integrate_f_wrap(df["a"].to_numpy(), df["b"].to_numpy(), df["N"].to_numpy())
+
+However, a loop indexer ``i`` accessing an invalid location in an array would cause a segfault because memory access isn't checked.
+For more about ``boundscheck`` and ``wraparound``, see the Cython docs on
+`compiler directives <https://cython.readthedocs.io/en/latest/src/userguide/source_files_and_compilation.html#compiler-directives>`__.
+
+.. _enhancingperf.numba:
+
+Numba (JIT compilation)
+-----------------------
+
+An alternative to statically compiling Cython code is to use a dynamic just-in-time (JIT) compiler with `Numba <https://numba.pydata.org/>`__.
+
+Numba allows you to write a pure Python function which can be JIT compiled to native machine instructions, similar in performance to C, C++ and Fortran,
+by decorating your function with ``@jit``.
+
+Numba works by generating optimized machine code using the LLVM compiler infrastructure at import time, runtime, or statically (using the included pycc tool).
+Numba supports compilation of Python to run on either CPU or GPU hardware and is designed to integrate with the Python scientific software stack.
+
+.. note::
+
+    The ``@jit`` compilation will add overhead to the runtime of the function, so performance benefits may not be realized especially when using small data sets.
+    Consider `caching <https://numba.readthedocs.io/en/stable/developer/caching.html>`__ your function to avoid compilation overhead each time your function is run.
+
+Numba can be used in 2 ways with pandas:
+
+#. Specify the ``engine="numba"`` keyword in select pandas methods
+#. Define your own Python function decorated with ``@jit`` and pass the underlying NumPy array of :class:`Series` or :class:`DataFrame` (using :meth:`Series.to_numpy`) into the function
+
+pandas Numba Engine
+~~~~~~~~~~~~~~~~~~~
+
+If Numba is installed, one can specify ``engine="numba"`` in select pandas methods to execute the method using Numba.
+Methods that support ``engine="numba"`` will also have an ``engine_kwargs`` keyword that accepts a dictionary that allows one to specify
+``"nogil"``, ``"nopython"`` and ``"parallel"`` keys with boolean values to pass into the ``@jit`` decorator.
+If ``engine_kwargs`` is not specified, it defaults to ``{"nogil": False, "nopython": True, "parallel": False}`` unless otherwise specified.
+
+.. note::
+
+   In terms of performance, **the first time a function is run using the Numba engine will be slow**
+   as Numba will have some function compilation overhead. However, the JIT compiled functions are cached,
+   and subsequent calls will be fast. In general, the Numba engine is performant with
+   a larger amount of data points (e.g. 1+ million).
+
+   .. code-block:: ipython
+
+      In [1]: data = pd.Series(range(1_000_000))  # noqa: E225
+
+      In [2]: roll = data.rolling(10)
+
+      In [3]: def f(x):
+         ...:     return np.sum(x) + 5
+      # Run the first time, compilation time will affect performance
+      In [4]: %timeit -r 1 -n 1 roll.apply(f, engine='numba', raw=True)
+      1.23 s ± 0 ns per loop (mean ± std. dev. of 1 run, 1 loop each)
+      # Function is cached and performance will improve
+      In [5]: %timeit roll.apply(f, engine='numba', raw=True)
+      188 ms ± 1.93 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+      In [6]: %timeit roll.apply(f, engine='cython', raw=True)
+      3.92 s ± 59 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+If your compute hardware contains multiple CPUs, the largest performance gain can be realized by setting ``parallel`` to ``True``
+to leverage more than 1 CPU. Internally, pandas leverages numba to parallelize computations over the columns of a :class:`DataFrame`;
+therefore, this performance benefit is only beneficial for a :class:`DataFrame` with a large number of columns.
+
+.. code-block:: ipython
+
+   In [1]: import numba
+
+   In [2]: numba.set_num_threads(1)
+
+   In [3]: df = pd.DataFrame(np.random.randn(10_000, 100))
+
+   In [4]: roll = df.rolling(100)
+
+   In [5]: %timeit roll.mean(engine="numba", engine_kwargs={"parallel": True})
+   347 ms ± 26 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [6]: numba.set_num_threads(2)
+
+   In [7]: %timeit roll.mean(engine="numba", engine_kwargs={"parallel": True})
+   201 ms ± 2.97 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+Custom Function Examples
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+A custom Python function decorated with ``@jit`` can be used with pandas objects by passing their NumPy array
+representations with :meth:`Series.to_numpy`.
+
+.. code-block:: python
+
+   import numba
+
+
+   @numba.jit
+   def f_plain(x):
+       return x * (x - 1)
+
+
+   @numba.jit
+   def integrate_f_numba(a, b, N):
+       s = 0
+       dx = (b - a) / N
+       for i in range(N):
+           s += f_plain(a + i * dx)
+       return s * dx
+
+
+   @numba.jit
+   def apply_integrate_f_numba(col_a, col_b, col_N):
+       n = len(col_N)
+       result = np.empty(n, dtype="float64")
+       assert len(col_a) == len(col_b) == n
+       for i in range(n):
+           result[i] = integrate_f_numba(col_a[i], col_b[i], col_N[i])
+       return result
+
+
+   def compute_numba(df):
+       result = apply_integrate_f_numba(
+           df["a"].to_numpy(), df["b"].to_numpy(), df["N"].to_numpy()
+       )
+       return pd.Series(result, index=df.index, name="result")
+
+
+.. code-block:: ipython
+
+   In [4]: %timeit compute_numba(df)
+   1000 loops, best of 3: 798 us per loop
+
+In this example, using Numba was faster than Cython.
+
+Numba can also be used to write vectorized functions that do not require the user to explicitly
+loop over the observations of a vector; a vectorized function will be applied to each row automatically.
+Consider the following example of doubling each observation:
+
+.. code-block:: python
+
+   import numba
+
+
+   def double_every_value_nonumba(x):
+       return x * 2
+
+
+   @numba.vectorize
+   def double_every_value_withnumba(x):  # noqa E501
+       return x * 2
+
+.. code-block:: ipython
+
+   # Custom function without numba
+   In [5]: %timeit df["col1_doubled"] = df["a"].apply(double_every_value_nonumba)  # noqa E501
+   1000 loops, best of 3: 797 us per loop
+
+   # Standard implementation (faster than a custom function)
+   In [6]: %timeit df["col1_doubled"] = df["a"] * 2
+   1000 loops, best of 3: 233 us per loop
+
+   # Custom function with numba
+   In [7]: %timeit df["col1_doubled"] = double_every_value_withnumba(df["a"].to_numpy())
+   1000 loops, best of 3: 145 us per loop
+
+Caveats
+~~~~~~~
+
+Numba is best at accelerating functions that apply numerical functions to NumPy
+arrays. If you try to ``@jit`` a function that contains unsupported `Python <https://numba.readthedocs.io/en/stable/reference/pysupported.html>`__
+or `NumPy <https://numba.readthedocs.io/en/stable/reference/numpysupported.html>`__
+code, compilation will revert `object mode <https://numba.readthedocs.io/en/stable/glossary.html#term-object-mode>`__ which
+will mostly likely not speed up your function. If you would
+prefer that Numba throw an error if it cannot compile a function in a way that
+speeds up your code, pass Numba the argument
+``nopython=True`` (e.g.  ``@jit(nopython=True)``). For more on
+troubleshooting Numba modes, see the `Numba troubleshooting page
+<https://numba.readthedocs.io/en/stable/user/troubleshoot.html>`__.
+
+Using ``parallel=True`` (e.g. ``@jit(parallel=True)``) may result in a ``SIGABRT`` if the threading layer leads to unsafe
+behavior. You can first `specify a safe threading layer <https://numba.readthedocs.io/en/stable/user/threading-layer.html#selecting-a-threading-layer-for-safe-parallel-execution>`__
+before running a JIT function with ``parallel=True``.
+
+Generally if the you encounter a segfault (``SIGSEGV``) while using Numba, please report the issue
+to the `Numba issue tracker. <https://github.com/numba/numba/issues/new/choose>`__
+
+.. _enhancingperf.eval:
+
+Expression evaluation via :func:`~pandas.eval`
+----------------------------------------------
+
+The top-level function :func:`pandas.eval` implements performant expression evaluation of
+:class:`~pandas.Series` and :class:`~pandas.DataFrame`. Expression evaluation allows operations
+to be expressed as strings and can potentially provide a performance improvement
+by evaluate arithmetic and boolean expression all at once for large :class:`~pandas.DataFrame`.
+
+.. note::
+
+   You should not use :func:`~pandas.eval` for simple
+   expressions or for expressions involving small DataFrames. In fact,
+   :func:`~pandas.eval` is many orders of magnitude slower for
+   smaller expressions or objects than plain Python. A good rule of thumb is
+   to only use :func:`~pandas.eval` when you have a
+   :class:`~pandas.core.frame.DataFrame` with more than 10,000 rows.
+
+Supported syntax
+~~~~~~~~~~~~~~~~
+
+These operations are supported by :func:`pandas.eval`:
+
+* Arithmetic operations except for the left shift (``<<``) and right shift
+  (``>>``) operators, e.g., ``df + 2 * pi / s ** 4 % 42 - the_golden_ratio``
+* Comparison operations, including chained comparisons, e.g., ``2 < df < df2``
+* Boolean operations, e.g., ``df < df2 and df3 < df4 or not df_bool``
+* ``list`` and ``tuple`` literals, e.g., ``[1, 2]`` or ``(1, 2)``
+* Attribute access, e.g., ``df.a``
+* Subscript expressions, e.g., ``df[0]``
+* Simple variable evaluation, e.g., ``pd.eval("df")`` (this is not very useful)
+* Math functions: ``sin``, ``cos``, ``exp``, ``log``, ``expm1``, ``log1p``,
+  ``sqrt``, ``sinh``, ``cosh``, ``tanh``, ``arcsin``, ``arccos``, ``arctan``, ``arccosh``,
+  ``arcsinh``, ``arctanh``, ``abs``, ``arctan2`` and ``log10``.
+
+The following Python syntax is **not** allowed:
+
+* Expressions
+
+    * Function calls other than math functions.
+    * ``is``/``is not`` operations
+    * ``if`` expressions
+    * ``lambda`` expressions
+    * ``list``/``set``/``dict`` comprehensions
+    * Literal ``dict`` and ``set`` expressions
+    * ``yield`` expressions
+    * Generator expressions
+    * Boolean expressions consisting of only scalar values
+
+* Statements
+
+    * Neither `simple <https://docs.python.org/3/reference/simple_stmts.html>`__
+      or `compound <https://docs.python.org/3/reference/compound_stmts.html>`__
+      statements are allowed. This includes ``for``, ``while``, and
+      ``if``.
+
+Local variables
+~~~~~~~~~~~~~~~
+
+You must *explicitly reference* any local variable that you want to use in an
+expression by placing the ``@`` character in front of the name. This mechanism is
+the same for both :meth:`DataFrame.query` and :meth:`DataFrame.eval`. For example,
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(5, 2), columns=list("ab"))
+   newcol = np.random.randn(len(df))
+   df.eval("b + @newcol")
+   df.query("b < @newcol")
+
+If you don't prefix the local variable with ``@``, pandas will raise an
+exception telling you the variable is undefined.
+
+When using :meth:`DataFrame.eval` and :meth:`DataFrame.query`, this allows you
+to have a local variable and a :class:`~pandas.DataFrame` column with the same
+name in an expression.
+
+
+.. ipython:: python
+
+   a = np.random.randn()
+   df.query("@a < a")
+   df.loc[a < df["a"]]  # same as the previous expression
+
+.. warning::
+
+   :func:`pandas.eval` will raise an exception if you cannot use the ``@`` prefix because it
+   isn't defined in that context.
+
+   .. ipython:: python
+      :okexcept:
+
+      a, b = 1, 2
+      pd.eval("@a + b")
+
+   In this case, you should simply refer to the variables like you would in
+   standard Python.
+
+   .. ipython:: python
+
+      pd.eval("a + b")
+
+
+:func:`pandas.eval` parsers
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+There are two different expression syntax parsers.
+
+The default ``'pandas'`` parser allows a more intuitive syntax for expressing
+query-like operations (comparisons, conjunctions and disjunctions). In
+particular, the precedence of the ``&`` and ``|`` operators is made equal to
+the precedence of the corresponding boolean operations ``and`` and ``or``.
+
+For example, the above conjunction can be written without parentheses.
+Alternatively, you can use the ``'python'`` parser to enforce strict Python
+semantics.
+
+.. ipython:: python
+
+   nrows, ncols = 20000, 100
+   df1, df2, df3, df4 = [pd.DataFrame(np.random.randn(nrows, ncols)) for _ in range(4)]
+
+   expr = "(df1 > 0) & (df2 > 0) & (df3 > 0) & (df4 > 0)"
+   x = pd.eval(expr, parser="python")
+   expr_no_parens = "df1 > 0 & df2 > 0 & df3 > 0 & df4 > 0"
+   y = pd.eval(expr_no_parens, parser="pandas")
+   np.all(x == y)
+
+
+The same expression can be "anded" together with the word :keyword:`and` as
+well:
+
+.. ipython:: python
+
+   expr = "(df1 > 0) & (df2 > 0) & (df3 > 0) & (df4 > 0)"
+   x = pd.eval(expr, parser="python")
+   expr_with_ands = "df1 > 0 and df2 > 0 and df3 > 0 and df4 > 0"
+   y = pd.eval(expr_with_ands, parser="pandas")
+   np.all(x == y)
+
+The :keyword:`and` and :keyword:`or` operators here have the same precedence that they would
+in Python.
+
+
+:func:`pandas.eval` engines
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+There are two different expression engines.
+
+The ``'numexpr'`` engine is the more performant engine that can yield performance improvements
+compared to standard Python syntax for large :class:`DataFrame`. This engine requires the
+optional dependency ``numexpr`` to be installed.
+
+The ``'python'`` engine is generally *not* useful except for testing
+other evaluation engines against it. You will achieve **no** performance
+benefits using :func:`~pandas.eval` with ``engine='python'`` and may
+incur a performance hit.
+
+.. ipython:: python
+
+   %timeit df1 + df2 + df3 + df4
+
+.. ipython:: python
+
+   %timeit pd.eval("df1 + df2 + df3 + df4", engine="python")
+
+
+The :meth:`DataFrame.eval` method
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In addition to the top level :func:`pandas.eval` function you can also
+evaluate an expression in the "context" of a :class:`~pandas.DataFrame`.
+
+.. ipython:: python
+   :suppress:
+
+   try:
+       del a
+   except NameError:
+       pass
+
+   try:
+       del b
+   except NameError:
+       pass
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(5, 2), columns=["a", "b"])
+   df.eval("a + b")
+
+Any expression that is a valid :func:`pandas.eval` expression is also a valid
+:meth:`DataFrame.eval` expression, with the added benefit that you don't have to
+prefix the name of the :class:`~pandas.DataFrame` to the column(s) you're
+interested in evaluating.
+
+In addition, you can perform assignment of columns within an expression.
+This allows for *formulaic evaluation*. The assignment target can be a
+new column name or an existing column name, and it must be a valid Python
+identifier.
+
+.. ipython:: python
+
+   df = pd.DataFrame(dict(a=range(5), b=range(5, 10)))
+   df = df.eval("c = a + b")
+   df = df.eval("d = a + b + c")
+   df = df.eval("a = 1")
+   df
+
+A copy of the :class:`DataFrame` with the
+new or modified columns is returned, and the original frame is unchanged.
+
+.. ipython:: python
+
+   df
+   df.eval("e = a - c")
+   df
+
+Multiple column assignments can be performed by using a multi-line string.
+
+.. ipython:: python
+
+   df.eval(
+       """
+   c = a + b
+   d = a + b + c
+   a = 1""",
+   )
+
+The equivalent in standard Python would be
+
+.. ipython:: python
+
+   df = pd.DataFrame(dict(a=range(5), b=range(5, 10)))
+   df["c"] = df["a"] + df["b"]
+   df["d"] = df["a"] + df["b"] + df["c"]
+   df["a"] = 1
+   df
+
+
+:func:`~pandas.eval` performance comparison
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:func:`pandas.eval` works well with expressions containing large arrays.
+
+.. ipython:: python
+
+   nrows, ncols = 20000, 100
+   df1, df2, df3, df4 = [pd.DataFrame(np.random.randn(nrows, ncols)) for _ in range(4)]
+
+
+:class:`DataFrame` arithmetic:
+
+.. ipython:: python
+
+   %timeit df1 + df2 + df3 + df4
+
+.. ipython:: python
+
+   %timeit pd.eval("df1 + df2 + df3 + df4")
+
+
+:class:`DataFrame` comparison:
+
+.. ipython:: python
+
+   %timeit (df1 > 0) & (df2 > 0) & (df3 > 0) & (df4 > 0)
+
+.. ipython:: python
+
+   %timeit pd.eval("(df1 > 0) & (df2 > 0) & (df3 > 0) & (df4 > 0)")
+
+
+:class:`DataFrame` arithmetic with unaligned axes.
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(50))
+   %timeit df1 + df2 + df3 + df4 + s
+
+.. ipython:: python
+
+   %timeit pd.eval("df1 + df2 + df3 + df4 + s")
+
+.. note::
+
+   Operations such as
+
+   .. code-block:: python
+
+      1 and 2  # would parse to 1 & 2, but should evaluate to 2
+      3 or 4  # would parse to 3 | 4, but should evaluate to 3
+      ~1  # this is okay, but slower when using eval
+
+   should be performed in Python. An exception will be raised if you try to
+   perform any boolean/bitwise operations with scalar operands that are not
+   of type ``bool`` or ``np.bool_``.
+
+Here is a plot showing the running time of
+:func:`pandas.eval` as function of the size of the frame involved in the
+computation. The two lines are two different engines.
+
+..
+    The eval-perf.png figure below was generated with /doc/scripts/eval_performance.py
+
+.. image:: ../_static/eval-perf.png
+
+You will only see the performance benefits of using the ``numexpr`` engine with :func:`pandas.eval` if your :class:`~pandas.DataFrame`
+has more than approximately 100,000 rows.
+
+This plot was created using a :class:`DataFrame` with 3 columns each containing
+floating point values generated using ``numpy.random.randn()``.
+
+Expression evaluation limitations with ``numexpr``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Expressions that would result in an object dtype or involve datetime operations
+because of ``NaT`` must be evaluated in Python space, but part of an expression
+can still be evaluated with ``numexpr``. For example:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"strings": np.repeat(list("cba"), 3), "nums": np.repeat(range(3), 3)}
+   )
+   df
+   df.query("strings == 'a' and nums == 1")
+
+The numeric part of the comparison (``nums == 1``) will be evaluated by
+``numexpr`` and the object part of the comparison (``"strings == 'a'``) will
+be evaluated by Python.
diff --git a/pandas/doc/source/user_guide/gotchas.rst b/pandas/doc/source/user_guide/gotchas.rst
new file mode 100644
index 0000000000000..e85eead4e0f09
--- /dev/null
+++ b/pandas/doc/source/user_guide/gotchas.rst
@@ -0,0 +1,376 @@
+.. _gotchas:
+
+{{ header }}
+
+********************************
+Frequently Asked Questions (FAQ)
+********************************
+
+.. _df-memory-usage:
+
+DataFrame memory usage
+----------------------
+The memory usage of a :class:`DataFrame` (including the index) is shown when calling
+the :meth:`~DataFrame.info`. A configuration option, ``display.memory_usage``
+(see :ref:`the list of options <options.available>`), specifies if the
+:class:`DataFrame` memory usage will be displayed when invoking the :meth:`~DataFrame.info`
+method.
+
+For example, the memory usage of the :class:`DataFrame` below is shown
+when calling :meth:`~DataFrame.info`:
+
+.. ipython:: python
+
+    dtypes = [
+        "int64",
+        "float64",
+        "datetime64[ns]",
+        "timedelta64[ns]",
+        "complex128",
+        "object",
+        "bool",
+    ]
+    n = 5000
+    data = {t: np.random.randint(100, size=n).astype(t) for t in dtypes}
+    df = pd.DataFrame(data)
+    df["categorical"] = df["object"].astype("category")
+
+    df.info()
+
+The ``+`` symbol indicates that the true memory usage could be higher, because
+pandas does not count the memory used by values in columns with
+``dtype=object``.
+
+Passing ``memory_usage='deep'`` will enable a more accurate memory usage report,
+accounting for the full usage of the contained objects. This is optional
+as it can be expensive to do this deeper introspection.
+
+.. ipython:: python
+
+   df.info(memory_usage="deep")
+
+By default the display option is set to ``True`` but can be explicitly
+overridden by passing the ``memory_usage`` argument when invoking :meth:`~DataFrame.info`.
+
+The memory usage of each column can be found by calling the
+:meth:`~DataFrame.memory_usage` method. This returns a :class:`Series` with an index
+represented by column names and memory usage of each column shown in bytes. For
+the :class:`DataFrame` above, the memory usage of each column and the total memory
+usage can be found with the :meth:`~DataFrame.memory_usage` method:
+
+.. ipython:: python
+
+    df.memory_usage()
+
+    # total memory usage of dataframe
+    df.memory_usage().sum()
+
+By default the memory usage of the :class:`DataFrame` index is shown in the
+returned :class:`Series`, the memory usage of the index can be suppressed by passing
+the ``index=False`` argument:
+
+.. ipython:: python
+
+    df.memory_usage(index=False)
+
+The memory usage displayed by the :meth:`~DataFrame.info` method utilizes the
+:meth:`~DataFrame.memory_usage` method to determine the memory usage of a
+:class:`DataFrame` while also formatting the output in human-readable units (base-2
+representation; i.e. 1KB = 1024 bytes).
+
+See also :ref:`Categorical Memory Usage <categorical.memory>`.
+
+.. _gotchas.truth:
+
+Using if/truth statements with pandas
+-------------------------------------
+
+pandas follows the NumPy convention of raising an error when you try to convert
+something to a ``bool``. This happens in an ``if``-statement or when using the
+boolean operations: ``and``, ``or``, and ``not``. It is not clear what the result
+of the following code should be:
+
+.. code-block:: python
+
+    >>> if pd.Series([False, True, False]):
+    ...     pass
+
+Should it be ``True`` because it's not zero-length, or ``False`` because there
+are ``False`` values? It is unclear, so instead, pandas raises a ``ValueError``:
+
+.. ipython:: python
+    :okexcept:
+
+    if pd.Series([False, True, False]):
+        print("I was true")
+
+You need to explicitly choose what you want to do with the :class:`DataFrame`, e.g.
+use :meth:`~DataFrame.any`, :meth:`~DataFrame.all` or :meth:`~DataFrame.empty`.
+Alternatively, you might want to compare if the pandas object is ``None``:
+
+.. ipython:: python
+
+    if pd.Series([False, True, False]) is not None:
+        print("I was not None")
+
+
+Below is how to check if any of the values are ``True``:
+
+.. ipython:: python
+
+    if pd.Series([False, True, False]).any():
+        print("I am any")
+
+Bitwise Boolean
+~~~~~~~~~~~~~~~
+
+Bitwise boolean operators like ``==`` and ``!=`` return a boolean :class:`Series`
+which performs an element-wise comparison when compared to a scalar.
+
+.. ipython:: python
+
+   s = pd.Series(range(5))
+   s == 4
+
+See :ref:`boolean comparisons<basics.compare>` for more examples.
+
+Using the ``in`` operator
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Using the Python ``in`` operator on a :class:`Series` tests for membership in the
+**index**, not membership among the values.
+
+.. ipython:: python
+
+    s = pd.Series(range(5), index=list("abcde"))
+    2 in s
+    'b' in s
+
+If this behavior is surprising, keep in mind that using ``in`` on a Python
+dictionary tests keys, not values, and :class:`Series` are dict-like.
+To test for membership in the values, use the method :meth:`~pandas.Series.isin`:
+
+.. ipython:: python
+
+    s.isin([2])
+    s.isin([2]).any()
+
+For :class:`DataFrame`, likewise, ``in`` applies to the column axis,
+testing for membership in the list of column names.
+
+.. _gotchas.udf-mutation:
+
+Mutating with User Defined Function (UDF) methods
+-------------------------------------------------
+
+This section applies to pandas methods that take a UDF. In particular, the methods
+:meth:`DataFrame.apply`, :meth:`DataFrame.aggregate`, :meth:`DataFrame.transform`, and
+:meth:`DataFrame.filter`.
+
+It is a general rule in programming that one should not mutate a container
+while it is being iterated over. Mutation will invalidate the iterator,
+causing unexpected behavior. Consider the example:
+
+.. ipython:: python
+
+   values = [0, 1, 2, 3, 4, 5]
+   n_removed = 0
+   for k, value in enumerate(values):
+       idx = k - n_removed
+       if value % 2 == 1:
+           del values[idx]
+           n_removed += 1
+       else:
+           values[idx] = value + 1
+   values
+
+One probably would have expected that the result would be ``[1, 3, 5]``.
+When using a pandas method that takes a UDF, internally pandas is often
+iterating over the
+:class:`DataFrame` or other pandas object. Therefore, if the UDF mutates (changes)
+the :class:`DataFrame`, unexpected behavior can arise.
+
+Here is a similar example with :meth:`DataFrame.apply`:
+
+.. ipython:: python
+   :okexcept:
+
+   def f(s):
+       s.pop("a")
+       return s
+
+   df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+   df.apply(f, axis="columns")
+
+To resolve this issue, one can make a copy so that the mutation does
+not apply to the container being iterated over.
+
+.. ipython:: python
+
+   values = [0, 1, 2, 3, 4, 5]
+   n_removed = 0
+   for k, value in enumerate(values.copy()):
+       idx = k - n_removed
+       if value % 2 == 1:
+           del values[idx]
+           n_removed += 1
+       else:
+           values[idx] = value + 1
+   values
+
+.. ipython:: python
+
+   def f(s):
+       s = s.copy()
+       s.pop("a")
+       return s
+
+   df = pd.DataFrame({"a": [1, 2, 3], 'b': [4, 5, 6]})
+   df.apply(f, axis="columns")
+
+Missing value representation for NumPy types
+--------------------------------------------
+
+``np.nan`` as the ``NA`` representation for NumPy types
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+For lack of ``NA`` (missing) support from the ground up in NumPy and Python in
+general, ``NA`` could have been represented with:
+
+* A *masked array* solution: an array of data and an array of boolean values
+  indicating whether a value is there or is missing.
+* Using a special sentinel value, bit pattern, or set of sentinel values to
+  denote ``NA`` across the dtypes.
+
+The special value ``np.nan`` (Not-A-Number) was chosen as the ``NA`` value for NumPy types, and there are API
+functions like :meth:`DataFrame.isna` and :meth:`DataFrame.notna` which can be used across the dtypes to
+detect NA values. However, this choice has a downside of coercing missing integer data as float types as
+shown in :ref:`gotchas.intna`.
+
+``NA`` type promotions for NumPy types
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+When introducing NAs into an existing :class:`Series` or :class:`DataFrame` via
+:meth:`~Series.reindex` or some other means, boolean and integer types will be
+promoted to a different dtype in order to store the NAs. The promotions are
+summarized in this table:
+
+.. csv-table::
+   :header: "Typeclass","Promotion dtype for storing NAs"
+   :widths: 40,60
+
+   ``floating``, no change
+   ``object``, no change
+   ``integer``, cast to ``float64``
+   ``boolean``, cast to ``object``
+
+.. _gotchas.intna:
+
+Support for integer ``NA``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In the absence of high performance ``NA`` support being built into NumPy from
+the ground up, the primary casualty is the ability to represent NAs in integer
+arrays. For example:
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3, 4, 5], index=list("abcde"))
+   s
+   s.dtype
+
+   s2 = s.reindex(["a", "b", "c", "f", "u"])
+   s2
+   s2.dtype
+
+This trade-off is made largely for memory and performance reasons, and also so
+that the resulting :class:`Series` continues to be "numeric".
+
+If you need to represent integers with possibly missing values, use one of
+the nullable-integer extension dtypes provided by pandas or pyarrow
+
+* :class:`Int8Dtype`
+* :class:`Int16Dtype`
+* :class:`Int32Dtype`
+* :class:`Int64Dtype`
+* :class:`ArrowDtype`
+
+.. ipython:: python
+
+   s_int = pd.Series([1, 2, 3, 4, 5], index=list("abcde"), dtype=pd.Int64Dtype())
+   s_int
+   s_int.dtype
+
+   s2_int = s_int.reindex(["a", "b", "c", "f", "u"])
+   s2_int
+   s2_int.dtype
+
+   s_int_pa = pd.Series([1, 2, None], dtype="int64[pyarrow]")
+   s_int_pa
+
+See :ref:`integer_na` and :ref:`pyarrow` for more.
+
+Why not make NumPy like R?
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Many people have suggested that NumPy should simply emulate the ``NA`` support
+present in the more domain-specific statistical programming language `R
+<https://www.r-project.org/>`__. Part of the reason is the
+`NumPy type hierarchy <https://numpy.org/doc/stable/user/basics.types.html>`__.
+
+The R language, by contrast, only has a handful of built-in data types:
+``integer``, ``numeric`` (floating-point), ``character``, and
+``boolean``. ``NA`` types are implemented by reserving special bit patterns for
+each type to be used as the missing value. While doing this with the full NumPy
+type hierarchy would be possible, it would be a more substantial trade-off
+(especially for the 8- and 16-bit data types) and implementation undertaking.
+
+However, R ``NA`` semantics are now available by using masked NumPy types such as :class:`Int64Dtype`
+or PyArrow types (:class:`ArrowDtype`).
+
+
+Differences with NumPy
+----------------------
+For :class:`Series` and :class:`DataFrame` objects, :meth:`~DataFrame.var` normalizes by
+``N-1`` to produce `unbiased estimates of the population variance <https://en.wikipedia.org/wiki/Bias_of_an_estimator>`__, while NumPy's
+:meth:`numpy.var` normalizes by N, which measures the variance of the sample. Note that
+:meth:`~DataFrame.cov` normalizes by ``N-1`` in both pandas and NumPy.
+
+.. _gotchas.thread-safety:
+
+Thread-safety
+-------------
+
+pandas is not 100% thread safe. The known issues relate to
+the :meth:`~DataFrame.copy` method. If you are doing a lot of copying of
+:class:`DataFrame` objects shared among threads, we recommend holding locks inside
+the threads where the data copying occurs.
+
+See `this link <https://stackoverflow.com/questions/13592618/python-pandas-dataframe-thread-safe>`__
+for more information.
+
+
+Byte-ordering issues
+--------------------
+Occasionally you may have to deal with data that were created on a machine with
+a different byte order than the one on which you are running Python. A common
+symptom of this issue is an error like::
+
+    Traceback
+        ...
+    ValueError: Big-endian buffer not supported on little-endian compiler
+
+To deal
+with this issue you should convert the underlying NumPy array to the native
+system byte order *before* passing it to :class:`Series` or :class:`DataFrame`
+constructors using something similar to the following:
+
+.. ipython:: python
+
+   x = np.array(list(range(10)), ">i4")  # big endian
+   newx = x.byteswap().view(x.dtype.newbyteorder())  # force native byteorder
+   s = pd.Series(newx)
+
+See `the NumPy documentation on byte order
+<https://numpy.org/doc/stable/user/byteswapping.html>`__ for more
+details.
diff --git a/pandas/doc/source/user_guide/groupby.rst b/pandas/doc/source/user_guide/groupby.rst
new file mode 100644
index 0000000000000..4ec34db6ed959
--- /dev/null
+++ b/pandas/doc/source/user_guide/groupby.rst
@@ -0,0 +1,1749 @@
+.. _groupby:
+
+{{ header }}
+
+*****************************
+Group by: split-apply-combine
+*****************************
+
+By "group by" we are referring to a process involving one or more of the following
+steps:
+
+* **Splitting** the data into groups based on some criteria.
+* **Applying** a function to each group independently.
+* **Combining** the results into a data structure.
+
+Out of these, the split step is the most straightforward. In the apply step, we
+might wish to do one of the following:
+
+* **Aggregation**: compute a summary statistic (or statistics) for each
+  group. Some examples:
+
+    * Compute group sums or means.
+    * Compute group sizes / counts.
+
+* **Transformation**: perform some group-specific computations and return a
+  like-indexed object. Some examples:
+
+    * Standardize data (zscore) within a group.
+    * Filling NAs within groups with a value derived from each group.
+
+* **Filtration**: discard some groups, according to a group-wise computation
+  that evaluates to True or False. Some examples:
+
+    * Discard data that belong to groups with only a few members.
+    * Filter out data based on the group sum or mean.
+
+Many of these operations are defined on GroupBy objects. These operations are similar
+to those of the :ref:`aggregating API <basics.aggregate>`,
+:ref:`window API <window.overview>`, and :ref:`resample API <timeseries.aggregate>`.
+
+It is possible that a given operation does not fall into one of these categories or
+is some combination of them. In such a case, it may be possible to compute the
+operation using GroupBy's ``apply`` method. This method will examine the results of the
+apply step and try to sensibly combine them into a single result if it doesn't fit into either
+of the above three categories.
+
+.. note::
+
+   An operation that is split into multiple steps using built-in GroupBy operations
+   will be more efficient than using the ``apply`` method with a user-defined Python
+   function.
+
+
+The name GroupBy should be quite familiar to those who have used
+a SQL-based tool (or ``itertools``), in which you can write code like:
+
+.. code-block:: sql
+
+   SELECT Column1, Column2, mean(Column3), sum(Column4)
+   FROM SomeTable
+   GROUP BY Column1, Column2
+
+We aim to make operations like this natural and easy to express using
+pandas. We'll address each area of GroupBy functionality, then provide some
+non-trivial examples / use cases.
+
+See the :ref:`cookbook<cookbook.grouping>` for some advanced strategies.
+
+.. _groupby.split:
+
+Splitting an object into groups
+-------------------------------
+
+The abstract definition of grouping is to provide a mapping of labels to
+group names. To create a GroupBy object (more on what the GroupBy object is
+later), you may do the following:
+
+.. ipython:: python
+
+    speeds = pd.DataFrame(
+        [
+            ("bird", "Falconiformes", 389.0),
+            ("bird", "Psittaciformes", 24.0),
+            ("mammal", "Carnivora", 80.2),
+            ("mammal", "Primates", np.nan),
+            ("mammal", "Carnivora", 58),
+        ],
+        index=["falcon", "parrot", "lion", "monkey", "leopard"],
+        columns=("class", "order", "max_speed"),
+    )
+    speeds
+
+    grouped = speeds.groupby("class")
+    grouped = speeds.groupby(["class", "order"])
+
+The mapping can be specified many different ways:
+
+* A Python function, to be called on each of the index labels.
+* A list or NumPy array of the same length as the index.
+* A dict or ``Series``, providing a ``label -> group name`` mapping.
+* For ``DataFrame`` objects, a string indicating either a column name or
+  an index level name to be used to group.
+* A list of any of the above things.
+
+Collectively we refer to the grouping objects as the **keys**. For example,
+consider the following ``DataFrame``:
+
+.. note::
+
+   A string passed to ``groupby`` may refer to either a column or an index level.
+   If a string matches both a column name and an index level name, a
+   ``ValueError`` will be raised.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+           "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+           "C": np.random.randn(8),
+           "D": np.random.randn(8),
+       }
+   )
+   df
+
+On a DataFrame, we obtain a GroupBy object by calling :meth:`~DataFrame.groupby`.
+This method returns a ``pandas.api.typing.DataFrameGroupBy`` instance.
+We could naturally group by either the ``A`` or ``B`` columns, or both:
+
+.. ipython:: python
+
+   grouped = df.groupby("A")
+   grouped = df.groupby("B")
+   grouped = df.groupby(["A", "B"])
+
+.. note::
+
+   ``df.groupby('A')`` is just syntactic sugar for ``df.groupby(df['A'])``.
+
+The above GroupBy will split the DataFrame on its index (rows). To split by columns, first do
+a transpose:
+
+.. ipython::
+
+    In [4]: def get_letter_type(letter):
+       ...:     if letter.lower() in 'aeiou':
+       ...:         return 'vowel'
+       ...:     else:
+       ...:         return 'consonant'
+       ...:
+
+    In [5]: grouped = df.T.groupby(get_letter_type)
+
+pandas :class:`~pandas.Index` objects support duplicate values. If a
+non-unique index is used as the group key in a groupby operation, all values
+for the same index value will be considered to be in one group and thus the
+output of aggregation functions will only contain unique index values:
+
+.. ipython:: python
+
+   index = [1, 2, 3, 1, 2, 3]
+
+   s = pd.Series([1, 2, 3, 10, 20, 30], index=index)
+
+   s
+
+   grouped = s.groupby(level=0)
+
+   grouped.first()
+
+   grouped.last()
+
+   grouped.sum()
+
+Note that **no splitting occurs** until it's needed. Creating the GroupBy object
+only verifies that you've passed a valid mapping.
+
+.. note::
+
+   Many kinds of complicated data manipulations can be expressed in terms of
+   GroupBy operations (though it can't be guaranteed to be the most efficient implementation).
+   You can get quite creative with the label mapping functions.
+
+.. _groupby.sorting:
+
+GroupBy sorting
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+By default the group keys are sorted during the ``groupby`` operation. You may however pass ``sort=False`` for potential speedups. With ``sort=False`` the order among group-keys follows the order of appearance of the keys in the original dataframe:
+
+.. ipython:: python
+
+   df2 = pd.DataFrame({"X": ["B", "B", "A", "A"], "Y": [1, 2, 3, 4]})
+   df2.groupby(["X"]).sum()
+   df2.groupby(["X"], sort=False).sum()
+
+
+Note that ``groupby`` will preserve the order in which *observations* are sorted *within* each group.
+For example, the groups created by ``groupby()`` below are in the order they appeared in the original ``DataFrame``:
+
+.. ipython:: python
+
+   df3 = pd.DataFrame({"X": ["A", "B", "A", "B"], "Y": [1, 4, 3, 2]})
+   df3.groupby("X").get_group("A")
+
+   df3.groupby(["X"]).get_group(("B",))
+
+
+.. _groupby.dropna:
+
+GroupBy dropna
+^^^^^^^^^^^^^^
+
+By default ``NA`` values are excluded from group keys during the ``groupby`` operation. However,
+in case you want to include ``NA`` values in group keys, you could pass ``dropna=False`` to achieve it.
+
+.. ipython:: python
+
+    df_list = [[1, 2, 3], [1, None, 4], [2, 1, 3], [1, 2, 2]]
+    df_dropna = pd.DataFrame(df_list, columns=["a", "b", "c"])
+
+    df_dropna
+
+.. ipython:: python
+
+    # Default ``dropna`` is set to True, which will exclude NaNs in keys
+    df_dropna.groupby(by=["b"], dropna=True).sum()
+
+    # In order to allow NaN in keys, set ``dropna`` to False
+    df_dropna.groupby(by=["b"], dropna=False).sum()
+
+The default setting of ``dropna`` argument is ``True`` which means ``NA`` are not included in group keys.
+
+
+.. _groupby.attributes:
+
+GroupBy object attributes
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The ``groups`` attribute is a dictionary whose keys are the computed unique groups
+and corresponding values are the index labels belonging to each group. In the
+above example we have:
+
+.. ipython:: python
+
+   df.groupby("A").groups
+   df.T.groupby(get_letter_type).groups
+
+Calling the standard Python ``len`` function on the GroupBy object returns
+the number of groups, which is the same as the length of the ``groups`` dictionary:
+
+.. ipython:: python
+
+   grouped = df.groupby(["A", "B"])
+   grouped.groups
+   len(grouped)
+
+
+.. _groupby.tabcompletion:
+
+``GroupBy`` will tab complete column names, GroupBy operations, and other attributes:
+
+.. ipython:: python
+
+   n = 10
+   weight = np.random.normal(166, 20, size=n)
+   height = np.random.normal(60, 10, size=n)
+   time = pd.date_range("1/1/2000", periods=n)
+   gender = np.random.choice(["male", "female"], size=n)
+   df = pd.DataFrame(
+       {"height": height, "weight": weight, "gender": gender}, index=time
+   )
+   df
+   gb = df.groupby("gender")
+
+
+.. ipython::
+
+   @verbatim
+   In [1]: gb.<TAB>  # noqa: E225, E999
+   gb.agg        gb.boxplot    gb.cummin     gb.describe   gb.filter     gb.get_group  gb.height     gb.last       gb.median     gb.ngroups    gb.plot       gb.rank       gb.std        gb.transform
+   gb.aggregate  gb.count      gb.cumprod    gb.dtype      gb.first      gb.groups     gb.hist       gb.max        gb.min        gb.nth        gb.prod       gb.resample   gb.sum        gb.var
+   gb.apply      gb.cummax     gb.cumsum     gb.gender     gb.head       gb.indices    gb.mean       gb.name       gb.ohlc       gb.quantile   gb.size       gb.tail       gb.weight
+
+.. _groupby.multiindex:
+
+GroupBy with MultiIndex
+~~~~~~~~~~~~~~~~~~~~~~~
+
+With :ref:`hierarchically-indexed data <advanced.hierarchical>`, it's quite
+natural to group by one of the levels of the hierarchy.
+
+Let's create a Series with a two-level ``MultiIndex``.
+
+.. ipython:: python
+
+
+   arrays = [
+       ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+       ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+   index = pd.MultiIndex.from_arrays(arrays, names=["first", "second"])
+   s = pd.Series(np.random.randn(8), index=index)
+   s
+
+We can then group by one of the levels in ``s``.
+
+.. ipython:: python
+
+   grouped = s.groupby(level=0)
+   grouped.sum()
+
+If the MultiIndex has names specified, these can be passed instead of the level
+number:
+
+.. ipython:: python
+
+   s.groupby(level="second").sum()
+
+Grouping with multiple levels is supported.
+
+.. ipython:: python
+
+   arrays = [
+       ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+       ["doo", "doo", "bee", "bee", "bop", "bop", "bop", "bop"],
+       ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+   index = pd.MultiIndex.from_arrays(arrays, names=["first", "second", "third"])
+   s = pd.Series(np.random.randn(8), index=index)
+   s
+   s.groupby(level=["first", "second"]).sum()
+
+Index level names may be supplied as keys.
+
+.. ipython:: python
+
+   s.groupby(["first", "second"]).sum()
+
+More on the ``sum`` function and aggregation later.
+
+Grouping DataFrame with Index levels and columns
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+A DataFrame may be grouped by a combination of columns and index levels. You
+can specify both column and index names, or use a :class:`Grouper`.
+
+Let's first create a DataFrame with a MultiIndex:
+
+.. ipython:: python
+
+   arrays = [
+       ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+       ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+
+   index = pd.MultiIndex.from_arrays(arrays, names=["first", "second"])
+
+   df = pd.DataFrame({"A": [1, 1, 1, 1, 2, 2, 3, 3], "B": np.arange(8)}, index=index)
+
+   df
+
+Then we group ``df`` by the ``second`` index level and the ``A`` column.
+
+.. ipython:: python
+
+   df.groupby([pd.Grouper(level=1), "A"]).sum()
+
+Index levels may also be specified by name.
+
+.. ipython:: python
+
+   df.groupby([pd.Grouper(level="second"), "A"]).sum()
+
+Index level names may be specified as keys directly to ``groupby``.
+
+.. ipython:: python
+
+   df.groupby(["second", "A"]).sum()
+
+DataFrame column selection in GroupBy
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Once you have created the GroupBy object from a DataFrame, you might want to do
+something different for each of the columns. Thus, by using ``[]`` on the GroupBy
+object in a similar way as the one used to get a column from a DataFrame, you can do:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+           "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+           "C": np.random.randn(8),
+           "D": np.random.randn(8),
+       }
+   )
+
+   df
+
+   grouped = df.groupby(["A"])
+   grouped_C = grouped["C"]
+   grouped_D = grouped["D"]
+
+This is mainly syntactic sugar for the alternative, which is much more verbose:
+
+.. ipython:: python
+
+   df["C"].groupby(df["A"])
+
+Additionally, this method avoids recomputing the internal grouping information
+derived from the passed key.
+
+You can also include the grouping columns if you want to operate on them.
+
+.. ipython:: python
+
+   grouped[["A", "B"]].sum()
+
+.. note::
+
+   The ``groupby`` operation in pandas drops the ``name`` field of the columns Index object
+   after the operation. This change ensures consistency in syntax between different
+   column selection methods within groupby operations.
+
+.. _groupby.iterating-label:
+
+Iterating through groups
+------------------------
+
+With the GroupBy object in hand, iterating through the grouped data is very
+natural and functions similarly to :py:func:`itertools.groupby`:
+
+.. ipython::
+
+   In [4]: grouped = df.groupby('A')
+
+   In [5]: for name, group in grouped:
+      ...:     print(name)
+      ...:     print(group)
+      ...:
+
+In the case of grouping by multiple keys, the group name will be a tuple:
+
+.. ipython::
+
+   In [5]: for name, group in df.groupby(['A', 'B']):
+      ...:     print(name)
+      ...:     print(group)
+      ...:
+
+See :ref:`timeseries.iterating-label`.
+
+Selecting a group
+-----------------
+
+A single group can be selected using
+:meth:`.DataFrameGroupBy.get_group`:
+
+.. ipython:: python
+
+   grouped.get_group("bar")
+
+Or for an object grouped on multiple columns:
+
+.. ipython:: python
+
+   df.groupby(["A", "B"]).get_group(("bar", "one"))
+
+.. _groupby.aggregate:
+
+Aggregation
+-----------
+
+An aggregation is a GroupBy operation that reduces the dimension of the grouping
+object. The result of an aggregation is, or at least is treated as,
+a scalar value for each column in a group. For example, producing the sum of each
+column in a group of values.
+
+.. ipython:: python
+
+   animals = pd.DataFrame(
+       {
+           "kind": ["cat", "dog", "cat", "dog"],
+           "height": [9.1, 6.0, 9.5, 34.0],
+           "weight": [7.9, 7.5, 9.9, 198.0],
+       }
+   )
+   animals
+   animals.groupby("kind").sum()
+
+In the result, the keys of the groups appear in the index by default. They can be
+instead included in the columns by passing ``as_index=False``.
+
+.. ipython:: python
+
+   animals.groupby("kind", as_index=False).sum()
+
+.. _groupby.aggregate.builtin:
+
+Built-in aggregation methods
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Many common aggregations are built-in to GroupBy objects as methods. Of the methods
+listed below, those with a ``*`` do *not* have an efficient, GroupBy-specific, implementation.
+
+.. csv-table::
+    :header: "Method", "Description"
+    :widths: 20, 80
+
+        :meth:`~.DataFrameGroupBy.any`,Compute whether any of the values in the groups are truthy
+        :meth:`~.DataFrameGroupBy.all`,Compute whether all of the values in the groups are truthy
+        :meth:`~.DataFrameGroupBy.count`,Compute the number of non-NA values in the groups
+        :meth:`~.DataFrameGroupBy.cov` * ,Compute the covariance of the groups
+        :meth:`~.DataFrameGroupBy.first`,Compute the first occurring value in each group
+        :meth:`~.DataFrameGroupBy.idxmax`,Compute the index of the maximum value in each group
+        :meth:`~.DataFrameGroupBy.idxmin`,Compute the index of the minimum value in each group
+        :meth:`~.DataFrameGroupBy.last`,Compute the last occurring value in each group
+        :meth:`~.DataFrameGroupBy.max`,Compute the maximum value in each group
+        :meth:`~.DataFrameGroupBy.mean`,Compute the mean of each group
+        :meth:`~.DataFrameGroupBy.median`,Compute the median of each group
+        :meth:`~.DataFrameGroupBy.min`,Compute the minimum value in each group
+        :meth:`~.DataFrameGroupBy.nunique`,Compute the number of unique values in each group
+        :meth:`~.DataFrameGroupBy.prod`,Compute the product of the values in each group
+        :meth:`~.DataFrameGroupBy.quantile`,Compute a given quantile of the values in each group
+        :meth:`~.DataFrameGroupBy.sem`,Compute the standard error of the mean of the values in each group
+        :meth:`~.DataFrameGroupBy.size`,Compute the number of values in each group
+        :meth:`~.DataFrameGroupBy.skew` * ,Compute the skew of the values in each group
+        :meth:`~.DataFrameGroupBy.std`,Compute the standard deviation of the values in each group
+        :meth:`~.DataFrameGroupBy.sum`,Compute the sum of the values in each group
+        :meth:`~.DataFrameGroupBy.var`,Compute the variance of the values in each group
+
+Some examples:
+
+.. ipython:: python
+
+   df.groupby("A")[["C", "D"]].max()
+   df.groupby(["A", "B"]).mean()
+
+Another aggregation example is to compute the size of each group.
+This is included in GroupBy as the ``size`` method. It returns a Series whose
+index consists of the group names and the values are the sizes of each group.
+
+.. ipython:: python
+
+   grouped = df.groupby(["A", "B"])
+   grouped.size()
+
+While the :meth:`.DataFrameGroupBy.describe` method is not itself a reducer, it
+can be used to conveniently produce a collection of summary statistics about each of
+the groups.
+
+.. ipython:: python
+
+   grouped.describe()
+
+Another aggregation example is to compute the number of unique values of each group.
+This is similar to the :meth:`.DataFrameGroupBy.value_counts` function, except that it only counts the
+number of unique values.
+
+.. ipython:: python
+
+   ll = [['foo', 1], ['foo', 2], ['foo', 2], ['bar', 1], ['bar', 1]]
+   df4 = pd.DataFrame(ll, columns=["A", "B"])
+   df4
+   df4.groupby("A")["B"].nunique()
+
+.. note::
+
+   Aggregation functions **will not** return the groups that you are aggregating over
+   as named *columns* when ``as_index=True``, the default. The grouped columns will
+   be the **indices** of the returned object.
+
+   Passing ``as_index=False`` **will** return the groups that you are aggregating over as
+   named columns, regardless if they are named **indices** or *columns* in the inputs.
+
+
+.. _groupby.aggregate.agg:
+
+The :meth:`~.DataFrameGroupBy.aggregate` method
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. note::
+    The :meth:`~.DataFrameGroupBy.aggregate` method can accept many different types of
+    inputs. This section details using string aliases for various GroupBy methods; other
+    inputs are detailed in the sections below.
+
+Any reduction method that pandas implements can be passed as a string to
+:meth:`~.DataFrameGroupBy.aggregate`. Users are encouraged to use the shorthand,
+``agg``. It will operate as if the corresponding method was called.
+
+.. ipython:: python
+
+   grouped = df.groupby("A")
+   grouped[["C", "D"]].aggregate("sum")
+
+   grouped = df.groupby(["A", "B"])
+   grouped.agg("sum")
+
+The result of the aggregation will have the group names as the
+new index. In the case of multiple keys, the result is a
+:ref:`MultiIndex <advanced.hierarchical>` by default. As mentioned above, this can be
+changed by using the ``as_index`` option:
+
+.. ipython:: python
+
+   grouped = df.groupby(["A", "B"], as_index=False)
+   grouped.agg("sum")
+
+   df.groupby("A", as_index=False)[["C", "D"]].agg("sum")
+
+Note that you could use the :meth:`DataFrame.reset_index` DataFrame function to achieve
+the same result as the column names are stored in the resulting ``MultiIndex``, although
+this will make an extra copy.
+
+.. ipython:: python
+
+   df.groupby(["A", "B"]).agg("sum").reset_index()
+
+.. _groupby.aggregate.udf:
+
+Aggregation with user-defined functions
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Users can also provide their own User-Defined Functions (UDFs) for custom aggregations.
+
+.. warning::
+
+    When aggregating with a UDF, the UDF should not mutate the
+    provided ``Series``. See :ref:`gotchas.udf-mutation` for more information.
+
+.. note::
+
+    Aggregating with a UDF is often less performant than using
+    the pandas built-in methods on GroupBy. Consider breaking up a complex operation
+    into a chain of operations that utilize the built-in methods.
+
+.. ipython:: python
+
+   animals
+   animals.groupby("kind")[["height"]].agg(lambda x: set(x))
+
+The resulting dtype will reflect that of the aggregating function. If the results from different groups have
+different dtypes, then a common dtype will be determined in the same way as ``DataFrame`` construction.
+
+.. ipython:: python
+
+   animals.groupby("kind")[["height"]].agg(lambda x: x.astype(int).sum())
+
+.. _groupby.aggregate.multifunc:
+
+Applying multiple functions at once
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+On a grouped ``Series``, you can pass a list or dict of functions to
+:meth:`SeriesGroupBy.agg`, outputting a DataFrame:
+
+.. ipython:: python
+
+   grouped = df.groupby("A")
+   grouped["C"].agg(["sum", "mean", "std"])
+
+On a grouped ``DataFrame``, you can pass a list of functions to
+:meth:`DataFrameGroupBy.agg` to aggregate each
+column, which produces an aggregated result with a hierarchical column index:
+
+.. ipython:: python
+
+   grouped[["C", "D"]].agg(["sum", "mean", "std"])
+
+
+The resulting aggregations are named after the functions themselves.
+
+For a ``Series``, if you need to rename, you can add in a chained operation like this:
+
+.. ipython:: python
+
+   (
+       grouped["C"]
+       .agg(["sum", "mean", "std"])
+       .rename(columns={"sum": "foo", "mean": "bar", "std": "baz"})
+   )
+
+Or, you can simply pass a list of tuples each with the name of the new column and the aggregate function:
+
+.. ipython:: python
+
+   (
+      grouped["C"]
+      .agg([("foo", "sum"), ("bar", "mean"), ("baz", "std")])
+   )
+
+For a grouped ``DataFrame``, you can rename in a similar manner:
+
+By chaining ``rename`` operation,
+
+.. ipython:: python
+
+   (
+       grouped[["C", "D"]].agg(["sum", "mean", "std"]).rename(
+           columns={"sum": "foo", "mean": "bar", "std": "baz"}
+       )
+   )
+
+Or, passing a list of tuples,
+
+.. ipython:: python
+
+   (
+      grouped[["C", "D"]].agg(
+         [("foo", "sum"), ("bar", "mean"), ("baz", "std")]
+      )
+   )
+
+.. note::
+
+   In general, the output column names should be unique, but pandas will allow
+   you apply to the same function (or two functions with the same name) to the same
+   column.
+
+   .. ipython:: python
+
+      grouped["C"].agg(["sum", "sum"])
+
+
+   pandas also allows you to provide multiple lambdas. In this case, pandas
+   will mangle the name of the (nameless) lambda functions, appending ``_<i>``
+   to each subsequent lambda.
+
+   .. ipython:: python
+
+      grouped["C"].agg([lambda x: x.max() - x.min(), lambda x: x.median() - x.mean()])
+
+
+.. _groupby.aggregate.named:
+
+Named aggregation
+~~~~~~~~~~~~~~~~~
+
+To support column-specific aggregation *with control over the output column names*, pandas
+accepts the special syntax in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg`, known as "named aggregation", where
+
+- The keywords are the *output* column names
+- The values are tuples whose first element is the column to select
+  and the second element is the aggregation to apply to that column. pandas
+  provides the :class:`NamedAgg` namedtuple with the fields ``['column', 'aggfunc']``
+  to make it clearer what the arguments are. As usual, the aggregation can
+  be a callable or a string alias.
+
+.. ipython:: python
+
+   animals
+
+   animals.groupby("kind").agg(
+       min_height=pd.NamedAgg(column="height", aggfunc="min"),
+       max_height=pd.NamedAgg(column="height", aggfunc="max"),
+       average_weight=pd.NamedAgg(column="weight", aggfunc="mean"),
+   )
+
+
+:class:`NamedAgg` is just a ``namedtuple``. Plain tuples are allowed as well.
+
+.. ipython:: python
+
+   animals.groupby("kind").agg(
+       min_height=("height", "min"),
+       max_height=("height", "max"),
+       average_weight=("weight", "mean"),
+   )
+
+
+If the column names you want are not valid Python keywords, construct a dictionary
+and unpack the keyword arguments
+
+.. ipython:: python
+
+   animals.groupby("kind").agg(
+       **{
+           "total weight": pd.NamedAgg(column="weight", aggfunc="sum")
+       }
+   )
+
+When using named aggregation, additional keyword arguments are not passed through
+to the aggregation functions; only pairs
+of ``(column, aggfunc)`` should be passed as ``**kwargs``. If your aggregation functions
+require additional arguments, apply them partially with :meth:`functools.partial`.
+
+Named aggregation is also valid for Series groupby aggregations. In this case there's
+no column selection, so the values are just the functions.
+
+.. ipython:: python
+
+   animals.groupby("kind").height.agg(
+       min_height="min",
+       max_height="max",
+   )
+
+Applying different functions to DataFrame columns
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+By passing a dict to ``aggregate`` you can apply a different aggregation to the
+columns of a DataFrame:
+
+.. ipython:: python
+
+   grouped.agg({"C": "sum", "D": lambda x: np.std(x, ddof=1)})
+
+The function names can also be strings. In order for a string to be valid it
+must be implemented on GroupBy:
+
+.. ipython:: python
+
+   grouped.agg({"C": "sum", "D": "std"})
+
+.. _groupby.transform:
+
+Transformation
+--------------
+
+A transformation is a GroupBy operation whose result is indexed the same
+as the one being grouped. Common examples include :meth:`~.DataFrameGroupBy.cumsum` and
+:meth:`~.DataFrameGroupBy.diff`.
+
+.. ipython:: python
+
+    speeds
+    grouped = speeds.groupby("class")["max_speed"]
+    grouped.cumsum()
+    grouped.diff()
+
+Unlike aggregations, the groupings that are used to split
+the original object are not included in the result.
+
+.. note::
+
+    Since transformations do not include the groupings that are used to split the result,
+    the arguments ``as_index`` and ``sort`` in :meth:`DataFrame.groupby` and
+    :meth:`Series.groupby` have no effect.
+
+A common use of a transformation is to add the result back into the original DataFrame.
+
+.. ipython:: python
+
+    result = speeds.copy()
+    result["cumsum"] = grouped.cumsum()
+    result["diff"] = grouped.diff()
+    result
+
+Built-in transformation methods
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The following methods on GroupBy act as transformations.
+
+.. csv-table::
+    :header: "Method", "Description"
+    :widths: 20, 80
+
+        :meth:`~.DataFrameGroupBy.bfill`,Back fill NA values within each group
+        :meth:`~.DataFrameGroupBy.cumcount`,Compute the cumulative count within each group
+        :meth:`~.DataFrameGroupBy.cummax`,Compute the cumulative max within each group
+        :meth:`~.DataFrameGroupBy.cummin`,Compute the cumulative min within each group
+        :meth:`~.DataFrameGroupBy.cumprod`,Compute the cumulative product within each group
+        :meth:`~.DataFrameGroupBy.cumsum`,Compute the cumulative sum within each group
+        :meth:`~.DataFrameGroupBy.diff`,Compute the difference between adjacent values within each group
+        :meth:`~.DataFrameGroupBy.ffill`,Forward fill NA values within each group
+        :meth:`~.DataFrameGroupBy.pct_change`,Compute the percent change between adjacent values within each group
+        :meth:`~.DataFrameGroupBy.rank`,Compute the rank of each value within each group
+        :meth:`~.DataFrameGroupBy.shift`,Shift values up or down within each group
+
+In addition, passing any built-in aggregation method as a string to
+:meth:`~.DataFrameGroupBy.transform` (see the next section) will broadcast the result
+across the group, producing a transformed result. If the aggregation method has an efficient
+implementation, this will be performant as well.
+
+.. _groupby.transformation.transform:
+
+The :meth:`~.DataFrameGroupBy.transform` method
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Similar to the :ref:`aggregation method <groupby.aggregate.agg>`, the
+:meth:`~.DataFrameGroupBy.transform` method can accept string aliases to the built-in
+transformation methods in the previous section. It can *also* accept string aliases to
+the built-in aggregation methods. When an aggregation method is provided, the result
+will be broadcast across the group.
+
+.. ipython:: python
+
+    speeds
+    grouped = speeds.groupby("class")[["max_speed"]]
+    grouped.transform("cumsum")
+    grouped.transform("sum")
+
+In addition to string aliases, the :meth:`~.DataFrameGroupBy.transform` method can
+also accept User-Defined Functions (UDFs). The UDF must:
+
+* Return a result that is either the same size as the group chunk or
+  broadcastable to the size of the group chunk (e.g., a scalar,
+  ``grouped.transform(lambda x: x.iloc[-1])``).
+* Operate column-by-column on the group chunk.  The transform is applied to
+  the first group chunk using chunk.apply.
+* Not perform in-place operations on the group chunk. Group chunks should
+  be treated as immutable, and changes to a group chunk may produce unexpected
+  results. See :ref:`gotchas.udf-mutation` for more information.
+* (Optionally) operates on all columns of the entire group chunk at once. If this is
+  supported, a fast path is used starting from the *second* chunk.
+
+.. note::
+
+    Transforming by supplying ``transform`` with a UDF is
+    often less performant than using the built-in methods on GroupBy.
+    Consider breaking up a complex operation into a chain of operations that utilize
+    the built-in methods.
+
+    All of the examples in this section can be made more performant by calling
+    built-in methods instead of using UDFs.
+    See :ref:`below for examples <groupby_efficient_transforms>`.
+
+.. versionchanged:: 2.0.0
+
+    When using ``.transform`` on a grouped DataFrame and the transformation function
+    returns a DataFrame, pandas now aligns the result's index
+    with the input's index. You can call ``.to_numpy()`` within the transformation
+    function to avoid alignment.
+
+Similar to :ref:`groupby.aggregate.agg`, the resulting dtype will reflect that of the
+transformation function. If the results from different groups have different dtypes, then
+a common dtype will be determined in the same way as ``DataFrame`` construction.
+
+Suppose we wish to standardize the data within each group:
+
+.. ipython:: python
+
+   index = pd.date_range("10/1/1999", periods=1100)
+   ts = pd.Series(np.random.normal(0.5, 2, 1100), index)
+   ts = ts.rolling(window=100, min_periods=100).mean().dropna()
+
+   ts.head()
+   ts.tail()
+
+   transformed = ts.groupby(lambda x: x.year).transform(
+       lambda x: (x - x.mean()) / x.std()
+   )
+
+
+We would expect the result to now have mean 0 and standard deviation 1 within
+each group (up to floating-point error), which we can easily check:
+
+.. ipython:: python
+
+   # Original Data
+   grouped = ts.groupby(lambda x: x.year)
+   grouped.mean()
+   grouped.std()
+
+   # Transformed Data
+   grouped_trans = transformed.groupby(lambda x: x.year)
+   grouped_trans.mean()
+   grouped_trans.std()
+
+We can also visually compare the original and transformed data sets.
+
+.. ipython:: python
+
+   compare = pd.DataFrame({"Original": ts, "Transformed": transformed})
+
+   @savefig groupby_transform_plot.png
+   compare.plot()
+
+Transformation functions that have lower dimension outputs are broadcast to
+match the shape of the input array.
+
+.. ipython:: python
+
+   ts.groupby(lambda x: x.year).transform(lambda x: x.max() - x.min())
+
+Another common data transform is to replace missing data with the group mean.
+
+.. ipython:: python
+
+   cols = ["A", "B", "C"]
+   values = np.random.randn(1000, 3)
+   values[np.random.randint(0, 1000, 100), 0] = np.nan
+   values[np.random.randint(0, 1000, 50), 1] = np.nan
+   values[np.random.randint(0, 1000, 200), 2] = np.nan
+   data_df = pd.DataFrame(values, columns=cols)
+   data_df
+
+   countries = np.array(["US", "UK", "GR", "JP"])
+   key = countries[np.random.randint(0, 4, 1000)]
+
+   grouped = data_df.groupby(key)
+
+   # Non-NA count in each group
+   grouped.count()
+
+   transformed = grouped.transform(lambda x: x.fillna(x.mean()))
+
+We can verify that the group means have not changed in the transformed data,
+and that the transformed data contains no NAs.
+
+.. ipython:: python
+
+   grouped_trans = transformed.groupby(key)
+
+   grouped.mean()  # original group means
+   grouped_trans.mean()  # transformation did not change group means
+
+   grouped.count()  # original has some missing data points
+   grouped_trans.count()  # counts after transformation
+   grouped_trans.size()  # Verify non-NA count equals group size
+
+.. _groupby_efficient_transforms:
+
+As mentioned in the note above, each of the examples in this section can be computed
+more efficiently using built-in methods. In the code below, the inefficient way
+using a UDF is commented out and the faster alternative appears below.
+
+.. ipython:: python
+
+    # result = ts.groupby(lambda x: x.year).transform(
+    #     lambda x: (x - x.mean()) / x.std()
+    # )
+    grouped = ts.groupby(lambda x: x.year)
+    result = (ts - grouped.transform("mean")) / grouped.transform("std")
+
+    # result = ts.groupby(lambda x: x.year).transform(lambda x: x.max() - x.min())
+    grouped = ts.groupby(lambda x: x.year)
+    result = grouped.transform("max") - grouped.transform("min")
+
+    # grouped = data_df.groupby(key)
+    # result = grouped.transform(lambda x: x.fillna(x.mean()))
+    grouped = data_df.groupby(key)
+    result = data_df.fillna(grouped.transform("mean"))
+
+.. _groupby.transform.window_resample:
+
+Window and resample operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+It is possible to use ``resample()``, ``expanding()`` and
+``rolling()`` as methods on groupbys.
+
+The example below will apply the ``rolling()`` method on the samples of
+the column B, based on the groups of column A.
+
+.. ipython:: python
+
+   df_re = pd.DataFrame({"A": [1] * 10 + [5] * 10, "B": np.arange(20)})
+   df_re
+
+   df_re.groupby("A").rolling(4).B.mean()
+
+
+The ``expanding()`` method will accumulate a given operation
+(``sum()`` in the example) for all the members of each particular
+group.
+
+.. ipython:: python
+
+   df_re.groupby("A").expanding().sum()
+
+
+Suppose you want to use the ``resample()`` method to get a daily
+frequency in each group of your dataframe, and wish to complete the
+missing values with the ``ffill()`` method.
+
+.. ipython:: python
+
+   df_re = pd.DataFrame(
+       {
+           "date": pd.date_range(start="2016-01-01", periods=4, freq="W"),
+           "group": [1, 1, 2, 2],
+           "val": [5, 6, 7, 8],
+       }
+   ).set_index("date")
+   df_re
+
+   df_re.groupby("group").resample("1D").ffill()
+
+.. _groupby.filter:
+
+Filtration
+----------
+
+A filtration is a GroupBy operation that subsets the original grouping object. It
+may either filter out entire groups, part of groups, or both. Filtrations return
+a filtered version of the calling object, including the grouping columns when provided.
+In the following example, ``class`` is included in the result.
+
+.. ipython:: python
+
+    speeds
+    speeds.groupby("class").nth(1)
+
+.. note::
+
+    Unlike aggregations, filtrations do not add the group keys to the index of the
+    result. Because of this, passing ``as_index=False`` or ``sort=True`` will not
+    affect these methods.
+
+Filtrations will respect subsetting the columns of the GroupBy object.
+
+.. ipython:: python
+
+    speeds.groupby("class")[["order", "max_speed"]].nth(1)
+
+Built-in filtrations
+~~~~~~~~~~~~~~~~~~~~
+
+The following methods on GroupBy act as filtrations. All these methods have an
+efficient, GroupBy-specific, implementation.
+
+.. csv-table::
+    :header: "Method", "Description"
+    :widths: 20, 80
+
+        :meth:`~.DataFrameGroupBy.head`,Select the top row(s) of each group
+        :meth:`~.DataFrameGroupBy.nth`,Select the nth row(s) of each group
+        :meth:`~.DataFrameGroupBy.tail`,Select the bottom row(s) of each group
+
+Users can also use transformations along with Boolean indexing to construct complex
+filtrations within groups. For example, suppose we are given groups of products and
+their volumes, and we wish to subset the data to only the largest products capturing no
+more than 90% of the total volume within each group.
+
+.. ipython:: python
+
+    product_volumes = pd.DataFrame(
+        {
+            "group": list("xxxxyyy"),
+            "product": list("abcdefg"),
+            "volume": [10, 30, 20, 15, 40, 10, 20],
+        }
+    )
+    product_volumes
+
+    # Sort by volume to select the largest products first
+    product_volumes = product_volumes.sort_values("volume", ascending=False)
+    grouped = product_volumes.groupby("group")["volume"]
+    cumpct = grouped.cumsum() / grouped.transform("sum")
+    cumpct
+    significant_products = product_volumes[cumpct <= 0.9]
+    significant_products.sort_values(["group", "product"])
+
+The :class:`~DataFrameGroupBy.filter` method
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. note::
+
+    Filtering by supplying ``filter`` with a User-Defined Function (UDF) is
+    often less performant than using the built-in methods on GroupBy.
+    Consider breaking up a complex operation into a chain of operations that utilize
+    the built-in methods.
+
+The ``filter`` method takes a User-Defined Function (UDF) that, when applied to
+an entire group, returns either ``True`` or ``False``. The result of the ``filter``
+method is then the subset of groups for which the UDF returned ``True``.
+
+Suppose we want to take only elements that belong to groups with a group sum greater
+than 2.
+
+.. ipython:: python
+
+   sf = pd.Series([1, 1, 2, 3, 3, 3])
+   sf.groupby(sf).filter(lambda x: x.sum() > 2)
+
+Another useful operation is filtering out elements that belong to groups
+with only a couple members.
+
+.. ipython:: python
+
+   dff = pd.DataFrame({"A": np.arange(8), "B": list("aabbbbcc")})
+   dff.groupby("B").filter(lambda x: len(x) > 2)
+
+Alternatively, instead of dropping the offending groups, we can return a
+like-indexed objects where the groups that do not pass the filter are filled
+with NaNs.
+
+.. ipython:: python
+
+   dff.groupby("B").filter(lambda x: len(x) > 2, dropna=False)
+
+For DataFrames with multiple columns, filters should explicitly specify a column as the filter criterion.
+
+.. ipython:: python
+
+   dff["C"] = np.arange(8)
+   dff.groupby("B").filter(lambda x: len(x["C"]) > 2)
+
+.. _groupby.apply:
+
+Flexible ``apply``
+------------------
+
+Some operations on the grouped data might not fit into the aggregation,
+transformation, or filtration categories. For these, you can use the ``apply``
+function.
+
+.. warning::
+
+   ``apply`` has to try to infer from the result whether it should act as a reducer,
+   transformer, *or* filter, depending on exactly what is passed to it. Thus the
+   grouped column(s) may be included in the output or not. While
+   it tries to intelligently guess how to behave, it can sometimes guess wrong.
+
+.. note::
+
+   All of the examples in this section can be more reliably, and more efficiently,
+   computed using other pandas functionality.
+
+.. ipython:: python
+
+   df
+   grouped = df.groupby("A")
+
+   # could also just call .describe()
+   grouped["C"].apply(lambda x: x.describe())
+
+The dimension of the returned result can also change:
+
+.. ipython:: python
+
+    grouped = df.groupby('A')['C']
+
+    def f(group):
+        return pd.DataFrame({'original': group,
+                             'demeaned': group - group.mean()})
+
+    grouped.apply(f)
+
+``apply`` on a Series can operate on a returned value from the applied function
+that is itself a series, and possibly upcast the result to a DataFrame:
+
+.. ipython:: python
+
+    def f(x):
+        return pd.Series([x, x ** 2], index=["x", "x^2"])
+
+
+    s = pd.Series(np.random.rand(5))
+    s
+    s.apply(f)
+
+Similar to :ref:`groupby.aggregate.agg`, the resulting dtype will reflect that of the
+apply function. If the results from different groups have different dtypes, then
+a common dtype will be determined in the same way as ``DataFrame`` construction.
+
+Control grouped column(s) placement with ``group_keys``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To control whether the grouped column(s) are included in the indices, you can use
+the argument ``group_keys`` which defaults to ``True``. Compare
+
+.. ipython:: python
+
+    df.groupby("A", group_keys=True).apply(lambda x: x)
+
+with
+
+.. ipython:: python
+
+    df.groupby("A", group_keys=False).apply(lambda x: x)
+
+
+Numba accelerated routines
+--------------------------
+
+.. versionadded:: 1.1
+
+If `Numba <https://numba.pydata.org/>`__ is installed as an optional dependency, the ``transform`` and
+``aggregate`` methods support ``engine='numba'`` and ``engine_kwargs`` arguments.
+See :ref:`enhancing performance with Numba <enhancingperf.numba>` for general usage of the arguments
+and performance considerations.
+
+The function signature must start with ``values, index`` **exactly** as the data belonging to each group
+will be passed into ``values``, and the group index will be passed into ``index``.
+
+.. warning::
+
+   When using ``engine='numba'``, there will be no "fall back" behavior internally. The group
+   data and group index will be passed as NumPy arrays to the JITed user defined function, and no
+   alternative execution attempts will be tried.
+
+Other useful features
+---------------------
+
+Exclusion of non-numeric columns
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Again consider the example DataFrame we've been looking at:
+
+.. ipython:: python
+
+   df
+
+Suppose we wish to compute the standard deviation grouped by the ``A``
+column. There is a slight problem, namely that we don't care about the data in
+column ``B`` because it is not numeric. You can avoid non-numeric columns by
+specifying ``numeric_only=True``:
+
+.. ipython:: python
+
+   df.groupby("A").std(numeric_only=True)
+
+Note that ``df.groupby('A').colname.std().`` is more efficient than
+``df.groupby('A').std().colname``. So if the result of an aggregation function
+is only needed over one column (here ``colname``), it may be filtered
+*before* applying the aggregation function.
+
+.. ipython:: python
+
+    from decimal import Decimal
+
+    df_dec = pd.DataFrame(
+        {
+            "id": [1, 2, 1, 2],
+            "int_column": [1, 2, 3, 4],
+            "dec_column": [
+                Decimal("0.50"),
+                Decimal("0.15"),
+                Decimal("0.25"),
+                Decimal("0.40"),
+            ],
+        }
+    )
+    df_dec.groupby(["id"])[["dec_column"]].sum()
+
+.. _groupby.observed:
+
+Handling of (un)observed Categorical values
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+When using a ``Categorical`` grouper (as a single grouper, or as part of multiple groupers), the ``observed`` keyword
+controls whether to return a cartesian product of all possible groupers values (``observed=False``) or only those
+that are observed groupers (``observed=True``).
+
+Show all values:
+
+.. ipython:: python
+
+   pd.Series([1, 1, 1]).groupby(
+       pd.Categorical(["a", "a", "a"], categories=["a", "b"]), observed=False
+   ).count()
+
+Show only the observed values:
+
+.. ipython:: python
+
+   pd.Series([1, 1, 1]).groupby(
+       pd.Categorical(["a", "a", "a"], categories=["a", "b"]), observed=True
+   ).count()
+
+The returned dtype of the grouped will *always* include *all* of the categories that were grouped.
+
+.. ipython:: python
+
+   s = (
+       pd.Series([1, 1, 1])
+       .groupby(pd.Categorical(["a", "a", "a"], categories=["a", "b"]), observed=True)
+       .count()
+   )
+   s.index.dtype
+
+.. _groupby.missing:
+
+NA group handling
+~~~~~~~~~~~~~~~~~
+
+By ``NA``, we are referring to any ``NA`` values, including
+:class:`NA`, ``NaN``, ``NaT``, and ``None``. If there are any ``NA`` values in the
+grouping key, by default these will be excluded. In other words, any
+"``NA`` group" will be dropped. You can include NA groups by specifying ``dropna=False``.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"key": [1.0, 1.0, np.nan, 2.0, np.nan], "A": [1, 2, 3, 4, 5]})
+   df
+
+   df.groupby("key", dropna=True).sum()
+
+   df.groupby("key", dropna=False).sum()
+
+Grouping with ordered factors
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Categorical variables represented as instances of pandas's ``Categorical`` class
+can be used as group keys. If so, the order of the levels will be preserved. When
+``observed=False`` and ``sort=False``, any unobserved categories will be at the
+end of the result in order.
+
+.. ipython:: python
+
+    days = pd.Categorical(
+        values=["Wed", "Mon", "Thu", "Mon", "Wed", "Sat"],
+        categories=["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"],
+    )
+    data = pd.DataFrame(
+       {
+           "day": days,
+           "workers": [3, 4, 1, 4, 2, 2],
+       }
+    )
+    data
+
+    data.groupby("day", observed=False, sort=True).sum()
+
+    data.groupby("day", observed=False, sort=False).sum()
+
+.. _groupby.specify:
+
+Grouping with a grouper specification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You may need to specify a bit more data to properly group. You can
+use the ``pd.Grouper`` to provide this local control.
+
+.. ipython:: python
+
+   import datetime
+
+   df = pd.DataFrame(
+       {
+           "Branch": "A A A A A A A B".split(),
+           "Buyer": "Carl Mark Carl Carl Joe Joe Joe Carl".split(),
+           "Quantity": [1, 3, 5, 1, 8, 1, 9, 3],
+           "Date": [
+               datetime.datetime(2013, 1, 1, 13, 0),
+               datetime.datetime(2013, 1, 1, 13, 5),
+               datetime.datetime(2013, 10, 1, 20, 0),
+               datetime.datetime(2013, 10, 2, 10, 0),
+               datetime.datetime(2013, 10, 1, 20, 0),
+               datetime.datetime(2013, 10, 2, 10, 0),
+               datetime.datetime(2013, 12, 2, 12, 0),
+               datetime.datetime(2013, 12, 2, 14, 0),
+           ],
+       }
+   )
+
+   df
+
+Groupby a specific column with the desired frequency. This is like resampling.
+
+.. ipython:: python
+
+   df.groupby([pd.Grouper(freq="1ME", key="Date"), "Buyer"])[["Quantity"]].sum()
+
+When ``freq`` is specified, the object returned by ``pd.Grouper`` will be an
+instance of ``pandas.api.typing.TimeGrouper``. When there is a column and index
+with the same name, you can use ``key`` to group by the column and ``level``
+to group by the index.
+
+.. ipython:: python
+
+   df = df.set_index("Date")
+   df["Date"] = df.index + pd.offsets.MonthEnd(2)
+   df.groupby([pd.Grouper(freq="6ME", key="Date"), "Buyer"])[["Quantity"]].sum()
+
+   df.groupby([pd.Grouper(freq="6ME", level="Date"), "Buyer"])[["Quantity"]].sum()
+
+
+Taking the first rows of each group
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Just like for a DataFrame or Series you can call head and tail on a groupby:
+
+.. ipython:: python
+
+   df = pd.DataFrame([[1, 2], [1, 4], [5, 6]], columns=["A", "B"])
+   df
+
+   g = df.groupby("A")
+   g.head(1)
+
+   g.tail(1)
+
+This shows the first or last n rows from each group.
+
+.. _groupby.nth:
+
+Taking the nth row of each group
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To select the nth item from each group, use :meth:`.DataFrameGroupBy.nth` or
+:meth:`.SeriesGroupBy.nth`. Arguments supplied can be any integer, lists of integers,
+slices, or lists of slices; see below for examples. When the nth element of a group
+does not exist an error is *not* raised; instead no corresponding rows are returned.
+
+In general this operation acts as a filtration. In certain cases it will also return
+one row per group, making it also a reduction. However because in general it can
+return zero or multiple rows per group, pandas treats it as a filtration in all cases.
+
+.. ipython:: python
+
+   df = pd.DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=["A", "B"])
+   g = df.groupby("A")
+
+   g.nth(0)
+   g.nth(-1)
+   g.nth(1)
+
+If the nth element of a group does not exist, then no corresponding row is included
+in the result. In particular, if the specified ``n`` is larger than any group, the
+result will be an empty DataFrame.
+
+.. ipython:: python
+
+   g.nth(5)
+
+If you want to select the nth not-null item, use the ``dropna`` kwarg. For a DataFrame this should be either ``'any'`` or ``'all'`` just like you would pass to dropna:
+
+.. ipython:: python
+
+   # nth(0) is the same as g.first()
+   g.nth(0, dropna="any")
+   g.first()
+
+   # nth(-1) is the same as g.last()
+   g.nth(-1, dropna="any")
+   g.last()
+
+   g.B.nth(0, dropna="all")
+
+You can also select multiple rows from each group by specifying multiple nth values as a list of ints.
+
+.. ipython:: python
+
+   business_dates = pd.date_range(start="4/1/2014", end="6/30/2014", freq="B")
+   df = pd.DataFrame(1, index=business_dates, columns=["a", "b"])
+   # get the first, 4th, and last date index for each month
+   df.groupby([df.index.year, df.index.month]).nth([0, 3, -1])
+
+You may also use slices or lists of slices.
+
+.. ipython:: python
+
+   df.groupby([df.index.year, df.index.month]).nth[1:]
+   df.groupby([df.index.year, df.index.month]).nth[1:, :-1]
+
+Enumerate group items
+~~~~~~~~~~~~~~~~~~~~~
+
+To see the order in which each row appears within its group, use the
+``cumcount`` method:
+
+.. ipython:: python
+
+   dfg = pd.DataFrame(list("aaabba"), columns=["A"])
+   dfg
+
+   dfg.groupby("A").cumcount()
+
+   dfg.groupby("A").cumcount(ascending=False)
+
+.. _groupby.ngroup:
+
+Enumerate groups
+~~~~~~~~~~~~~~~~
+
+To see the ordering of the groups (as opposed to the order of rows
+within a group given by ``cumcount``) you can use
+:meth:`.DataFrameGroupBy.ngroup`.
+
+
+
+Note that the numbers given to the groups match the order in which the
+groups would be seen when iterating over the groupby object, not the
+order they are first observed.
+
+.. ipython:: python
+
+   dfg = pd.DataFrame(list("aaabba"), columns=["A"])
+   dfg
+
+   dfg.groupby("A").ngroup()
+
+   dfg.groupby("A").ngroup(ascending=False)
+
+Plotting
+~~~~~~~~
+
+Groupby also works with some plotting methods.  In this case, suppose we
+suspect that the values in column 1 are 3 times higher on average in group "B".
+
+
+.. ipython:: python
+
+   np.random.seed(1234)
+   df = pd.DataFrame(np.random.randn(50, 2))
+   df["g"] = np.random.choice(["A", "B"], size=50)
+   df.loc[df["g"] == "B", 1] += 3
+
+We can easily visualize this with a boxplot:
+
+.. ipython:: python
+   :okwarning:
+
+   @savefig groupby_boxplot.png
+   df.groupby("g").boxplot()
+
+The result of calling ``boxplot`` is a dictionary whose keys are the values
+of our grouping column ``g`` ("A" and "B"). The values of the resulting dictionary
+can be controlled by the ``return_type`` keyword of ``boxplot``.
+See the :ref:`visualization documentation<visualization.box>` for more.
+
+.. warning::
+
+  For historical reasons, ``df.groupby("g").boxplot()`` is not equivalent
+  to ``df.boxplot(by="g")``. See :ref:`here<visualization.box.return>` for
+  an explanation.
+
+.. _groupby.pipe:
+
+Piping function calls
+~~~~~~~~~~~~~~~~~~~~~
+
+Similar to the functionality provided by ``DataFrame`` and ``Series``, functions
+that take ``GroupBy`` objects can be chained together using a ``pipe`` method to
+allow for a cleaner, more readable syntax. To read about ``.pipe`` in general terms,
+see :ref:`here <basics.pipe>`.
+
+Combining ``.groupby`` and ``.pipe`` is often useful when you need to reuse
+GroupBy objects.
+
+As an example, imagine having a DataFrame with columns for stores, products,
+revenue and quantity sold. We'd like to do a groupwise calculation of *prices*
+(i.e. revenue/quantity) per store and per product. We could do this in a
+multi-step operation, but expressing it in terms of piping can make the
+code more readable. First we set the data:
+
+.. ipython:: python
+
+   n = 1000
+   df = pd.DataFrame(
+       {
+           "Store": np.random.choice(["Store_1", "Store_2"], n),
+           "Product": np.random.choice(["Product_1", "Product_2"], n),
+           "Revenue": (np.random.random(n) * 50 + 10).round(2),
+           "Quantity": np.random.randint(1, 10, size=n),
+       }
+   )
+   df.head(2)
+
+We now find the prices per store/product.
+
+.. ipython:: python
+
+   (
+       df.groupby(["Store", "Product"])
+       .pipe(lambda grp: grp.Revenue.sum() / grp.Quantity.sum())
+       .unstack()
+       .round(2)
+   )
+
+Piping can also be expressive when you want to deliver a grouped object to some
+arbitrary function, for example:
+
+.. ipython:: python
+
+   def mean(groupby):
+       return groupby.mean()
+
+
+   df.groupby(["Store", "Product"]).pipe(mean)
+
+Here ``mean`` takes a GroupBy object and finds the mean of the Revenue and Quantity
+columns respectively for each Store-Product combination. The ``mean`` function can
+be any function that takes in a GroupBy object; the ``.pipe`` will pass the GroupBy
+object as a parameter into the function you specify.
+
+Examples
+--------
+
+.. _groupby.multicolumn_factorization:
+
+Multi-column factorization
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+By using :meth:`.DataFrameGroupBy.ngroup`, we can extract
+information about the groups in a way similar to :func:`factorize` (as described
+further in the :ref:`reshaping API <reshaping.factorize>`) but which applies
+naturally to multiple columns of mixed type and different
+sources. This can be useful as an intermediate categorical-like step
+in processing, when the relationships between the group rows are more
+important than their content, or as input to an algorithm which only
+accepts the integer encoding. (For more information about support in
+pandas for full categorical data, see the :ref:`Categorical
+introduction <categorical>` and the
+:ref:`API documentation <api.arrays.categorical>`.)
+
+.. ipython:: python
+
+    dfg = pd.DataFrame({"A": [1, 1, 2, 3, 2], "B": list("aaaba")})
+
+    dfg
+
+    dfg.groupby(["A", "B"]).ngroup()
+
+    dfg.groupby(["A", [0, 0, 0, 1, 1]]).ngroup()
+
+GroupBy by indexer to 'resample' data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Resampling produces new hypothetical samples (resamples) from already existing observed data or from a model that generates data. These new samples are similar to the pre-existing samples.
+
+In order for resample to work on indices that are non-datetimelike, the following procedure can be utilized.
+
+In the following examples, **df.index // 5** returns an integer array which is used to determine what gets selected for the groupby operation.
+
+.. note::
+
+   The example below shows how we can downsample by consolidation of samples into fewer ones.
+   Here by using **df.index // 5**, we are aggregating the samples in bins. By applying **std()**
+   function, we aggregate the information contained in many samples into a small subset of values
+   which is their standard deviation thereby reducing the number of samples.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 2))
+   df
+   df.index // 5
+   df.groupby(df.index // 5).std()
+
+Returning a Series to propagate names
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Group DataFrame columns, compute a set of metrics and return a named Series.
+The Series name is used as the name for the column index. This is especially
+useful in conjunction with reshaping operations such as stacking, in which the
+column index name will be used as the name of the inserted column:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "a": [0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2],
+           "b": [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1],
+           "c": [1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0],
+           "d": [0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1],
+       }
+   )
+
+   def compute_metrics(x):
+       result = {"b_sum": x["b"].sum(), "c_mean": x["c"].mean()}
+       return pd.Series(result, name="metrics")
+
+   result = df.groupby("a").apply(compute_metrics)
+
+   result
+
+   result.stack()
diff --git a/pandas/doc/source/user_guide/index.rst b/pandas/doc/source/user_guide/index.rst
new file mode 100644
index 0000000000000..f0d6a76f0de5b
--- /dev/null
+++ b/pandas/doc/source/user_guide/index.rst
@@ -0,0 +1,90 @@
+{{ header }}
+
+.. _user_guide:
+
+==========
+User Guide
+==========
+
+The User Guide covers all of pandas by topic area. Each of the subsections
+introduces a topic (such as "working with missing data"), and discusses how
+pandas approaches the problem, with many examples throughout.
+
+Users brand-new to pandas should start with :ref:`10min`.
+
+For a high level summary of the pandas fundamentals, see :ref:`dsintro` and :ref:`basics`.
+
+Further information on any specific method can be obtained in the
+:ref:`api`.
+
+How to read these guides
+------------------------
+In these guides you will see input code inside code blocks such as:
+
+::
+
+    import pandas as pd
+    pd.DataFrame({'A': [1, 2, 3]})
+
+
+or:
+
+.. ipython:: python
+
+    import pandas as pd
+    pd.DataFrame({'A': [1, 2, 3]})
+
+The first block is a standard python input, while in the second the ``In [1]:`` indicates the input is inside a `notebook <https://jupyter.org>`__. In Jupyter Notebooks the last line is printed and plots are shown inline.
+
+For example:
+
+.. ipython:: python
+
+    a = 1
+    a
+is equivalent to:
+
+::
+
+    a = 1
+    print(a)
+
+
+
+Guides
+-------
+
+.. If you update this toctree, also update the manual toctree in the
+   main index.rst.template
+
+.. toctree::
+    :maxdepth: 2
+
+    10min
+    dsintro
+    basics
+    io
+    pyarrow
+    indexing
+    advanced
+    copy_on_write
+    merging
+    reshaping
+    text
+    missing_data
+    duplicates
+    categorical
+    integer_na
+    boolean
+    visualization
+    style
+    groupby
+    window
+    timeseries
+    timedeltas
+    options
+    enhancingperf
+    scale
+    sparse
+    gotchas
+    cookbook
diff --git a/pandas/doc/source/user_guide/indexing.rst b/pandas/doc/source/user_guide/indexing.rst
new file mode 100644
index 0000000000000..ed5c7806b2e23
--- /dev/null
+++ b/pandas/doc/source/user_guide/indexing.rst
@@ -0,0 +1,1717 @@
+.. _indexing:
+
+{{ header }}
+
+***************************
+Indexing and selecting data
+***************************
+
+The axis labeling information in pandas objects serves many purposes:
+
+* Identifies data (i.e. provides *metadata*) using known indicators,
+  important for analysis, visualization, and interactive console display.
+* Enables automatic and explicit data alignment.
+* Allows intuitive getting and setting of subsets of the data set.
+
+In this section, we will focus on the final point: namely, how to slice, dice,
+and generally get and set subsets of pandas objects. The primary focus will be
+on Series and DataFrame as they have received more development attention in
+this area.
+
+.. note::
+
+   The Python and NumPy indexing operators ``[]`` and attribute operator ``.``
+   provide quick and easy access to pandas data structures across a wide range
+   of use cases. This makes interactive work intuitive, as there's little new
+   to learn if you already know how to deal with Python dictionaries and NumPy
+   arrays. However, since the type of the data to be accessed isn't known in
+   advance, directly using standard operators has some optimization limits. For
+   production code, we recommended that you take advantage of the optimized
+   pandas data access methods exposed in this chapter.
+
+See the :ref:`MultiIndex / Advanced Indexing <advanced>` for ``MultiIndex`` and more advanced indexing documentation.
+
+See the :ref:`cookbook<cookbook.selection>` for some advanced strategies.
+
+.. _indexing.choice:
+
+Different choices for indexing
+------------------------------
+
+Object selection has had a number of user-requested additions in order to
+support more explicit location based indexing. pandas now supports three types
+of multi-axis indexing.
+
+* ``.loc`` is primarily label based, but may also be used with a boolean array. ``.loc`` will raise ``KeyError`` when the items are not found. Allowed inputs are:
+
+    * A single label, e.g. ``5`` or ``'a'`` (Note that ``5`` is interpreted as a
+      *label* of the index. This use is **not** an integer position along the
+      index.).
+    * A list or array of labels ``['a', 'b', 'c']``.
+    * A slice object with labels ``'a':'f'`` (Note that contrary to usual Python
+      slices, **both** the start and the stop are included, when present in the
+      index! See :ref:`Slicing with labels <indexing.slicing_with_labels>`
+      and :ref:`Endpoints are inclusive <advanced.endpoints_are_inclusive>`.)
+    * A boolean array (any ``NA`` values will be treated as ``False``).
+    * A ``callable`` function with one argument (the calling Series or DataFrame) and
+      that returns valid output for indexing (one of the above).
+    * A tuple of row (and column) indices whose elements are one of the
+      above inputs.
+
+  See more at :ref:`Selection by Label <indexing.label>`.
+
+* ``.iloc`` is primarily integer position based (from ``0`` to
+  ``length-1`` of the axis), but may also be used with a boolean
+  array.  ``.iloc`` will raise ``IndexError`` if a requested
+  indexer is out-of-bounds, except *slice* indexers which allow
+  out-of-bounds indexing.  (this conforms with Python/NumPy *slice*
+  semantics).  Allowed inputs are:
+
+    * An integer e.g. ``5``.
+    * A list or array of integers ``[4, 3, 0]``.
+    * A slice object with ints ``1:7``.
+    * A boolean array (any ``NA`` values will be treated as ``False``).
+    * A ``callable`` function with one argument (the calling Series or DataFrame) and
+      that returns valid output for indexing (one of the above).
+    * A tuple of row (and column) indices whose elements are one of the
+      above inputs.
+
+  See more at :ref:`Selection by Position <indexing.integer>`,
+  :ref:`Advanced Indexing <advanced>` and :ref:`Advanced
+  Hierarchical <advanced.advanced_hierarchical>`.
+
+* ``.loc``, ``.iloc``, and also ``[]`` indexing can accept a ``callable`` as indexer. See more at :ref:`Selection By Callable <indexing.callable>`.
+
+  .. note::
+
+     Destructuring tuple keys into row (and column) indexes occurs
+     *before* callables are applied, so you cannot return a tuple from
+     a callable to index both rows and columns.
+
+Getting values from an object with multi-axes selection uses the following
+notation (using ``.loc`` as an example, but the following applies to ``.iloc`` as
+well). Any of the axes accessors may be the null slice ``:``. Axes left out of
+the specification are assumed to be ``:``, e.g. ``p.loc['a']`` is equivalent to
+``p.loc['a', :]``.
+
+
+.. ipython:: python
+
+   ser = pd.Series(range(5), index=list("abcde"))
+   ser.loc[["a", "c", "e"]]
+
+   df = pd.DataFrame(np.arange(25).reshape(5, 5), index=list("abcde"), columns=list("abcde"))
+   df.loc[["a", "c", "e"], ["b", "d"]]
+
+.. _indexing.basics:
+
+Basics
+------
+
+As mentioned when introducing the data structures in the :ref:`last section
+<basics>`, the primary function of indexing with ``[]`` (a.k.a. ``__getitem__``
+for those familiar with implementing class behavior in Python) is selecting out
+lower-dimensional slices. The following table shows return type values when
+indexing pandas objects with ``[]``:
+
+.. csv-table::
+    :header: "Object Type", "Selection", "Return Value Type"
+    :widths: 30, 30, 60
+
+    Series, ``series[label]``, scalar value
+    DataFrame, ``frame[colname]``, ``Series`` corresponding to colname
+
+Here we construct a simple time series data set to use for illustrating the
+indexing functionality:
+
+.. ipython:: python
+
+   dates = pd.date_range('1/1/2000', periods=8)
+   df = pd.DataFrame(np.random.randn(8, 4),
+                     index=dates, columns=['A', 'B', 'C', 'D'])
+   df
+
+.. note::
+
+   None of the indexing functionality is time series specific unless
+   specifically stated.
+
+Thus, as per above, we have the most basic indexing using ``[]``:
+
+.. ipython:: python
+
+   s = df['A']
+   s[dates[5]]
+
+You can pass a list of columns to ``[]`` to select columns in that order.
+If a column is not contained in the DataFrame, an exception will be
+raised. Multiple columns can also be set in this manner:
+
+.. ipython:: python
+
+   df
+   df[['B', 'A']] = df[['A', 'B']]
+   df
+
+You may find this useful for applying a transform (in-place) to a subset of the
+columns.
+
+.. warning::
+
+   pandas aligns all AXES when setting ``Series`` and ``DataFrame`` from ``.loc``.
+
+   This will **not** modify ``df`` because the column alignment is before value assignment.
+
+   .. ipython:: python
+
+      df[['A', 'B']]
+      df.loc[:, ['B', 'A']] = df[['A', 'B']]
+      df[['A', 'B']]
+
+   The correct way to swap column values is by using raw values:
+
+   .. ipython:: python
+
+      df.loc[:, ['B', 'A']] = df[['A', 'B']].to_numpy()
+      df[['A', 'B']]
+
+   However, pandas does not align AXES when setting ``Series`` and ``DataFrame`` from ``.iloc``
+   because ``.iloc`` operates by position.
+
+   This will modify ``df`` because the column alignment is not done before value assignment.
+
+   .. ipython:: python
+
+      df[['A', 'B']]
+      df.iloc[:, [1, 0]] = df[['A', 'B']]
+      df[['A','B']]
+
+
+Attribute access
+----------------
+
+.. _indexing.columns.multiple:
+
+.. _indexing.df_cols:
+
+.. _indexing.attribute_access:
+
+You may access an index on a ``Series`` or  column on a ``DataFrame`` directly
+as an attribute:
+
+.. ipython:: python
+
+   sa = pd.Series([1, 2, 3], index=list('abc'))
+   dfa = df.copy()
+
+.. ipython:: python
+
+   sa.b
+   dfa.A
+
+.. ipython:: python
+
+   sa.a = 5
+   sa
+   dfa.A = list(range(len(dfa.index)))  # ok if A already exists
+   dfa
+   dfa['A'] = list(range(len(dfa.index)))  # use this form to create a new column
+   dfa
+
+.. warning::
+
+   - You can use this access only if the index element is a valid Python identifier, e.g. ``s.1`` is not allowed.
+     See `here for an explanation of valid identifiers
+     <https://docs.python.org/3/reference/lexical_analysis.html#identifiers>`__.
+
+   - The attribute will not be available if it conflicts with an existing method name, e.g. ``s.min`` is not allowed, but ``s['min']`` is possible.
+
+   - Similarly, the attribute will not be available if it conflicts with any of the following list: ``index``,
+     ``major_axis``, ``minor_axis``, ``items``.
+
+   - In any of these cases, standard indexing will still work, e.g. ``s['1']``, ``s['min']``, and ``s['index']`` will
+     access the corresponding element or column.
+
+If you are using the IPython environment, you may also use tab-completion to
+see these accessible attributes.
+
+You can also assign a ``dict`` to a row of a ``DataFrame``:
+
+.. ipython:: python
+
+   x = pd.DataFrame({'x': [1, 2, 3], 'y': [3, 4, 5]})
+   x.iloc[1] = {'x': 9, 'y': 99}
+   x
+
+You can use attribute access to modify an existing element of a Series or column of a DataFrame, but be careful;
+if you try to use attribute access to create a new column, it creates a new attribute rather than a
+new column and will this raise a ``UserWarning``:
+
+.. ipython:: python
+   :okwarning:
+
+    df_new = pd.DataFrame({'one': [1., 2., 3.]})
+    df_new.two = [4, 5, 6]
+    df_new
+
+
+Slicing ranges
+--------------
+
+The most robust and consistent way of slicing ranges along arbitrary axes is
+described in the :ref:`Selection by Position <indexing.integer>` section
+detailing the ``.iloc`` method. For now, we explain the semantics of slicing using the ``[]`` operator.
+
+    .. note::
+
+        When the :class:`Series` has float indices, slicing will select by position.
+
+With Series, the syntax works exactly as with an ndarray, returning a slice of
+the values and the corresponding labels:
+
+.. ipython:: python
+
+   s[:5]
+   s[::2]
+   s[::-1]
+
+Note that setting works as well:
+
+.. ipython:: python
+
+   s2 = s.copy()
+   s2[:5] = 0
+   s2
+
+With DataFrame, slicing inside of ``[]`` **slices the rows**. This is provided
+largely as a convenience since it is such a common operation.
+
+.. ipython:: python
+
+   df[:3]
+   df[::-1]
+
+.. _indexing.label:
+
+Selection by label
+------------------
+
+.. warning::
+
+   ``.loc`` is strict when you present slicers that are not compatible (or convertible) with the index type. For example
+   using integers in a ``DatetimeIndex``. These will raise a ``TypeError``.
+
+   .. ipython:: python
+      :okexcept:
+
+      dfl = pd.DataFrame(np.random.randn(5, 4),
+                         columns=list('ABCD'),
+                         index=pd.date_range('20130101', periods=5))
+      dfl
+      dfl.loc[2:3]
+
+  String likes in slicing *can* be convertible to the type of the index and lead to natural slicing.
+
+  .. ipython:: python
+
+     dfl.loc['20130102':'20130104']
+
+pandas provides a suite of methods in order to have **purely label based indexing**. This is a strict inclusion based protocol.
+Every label asked for must be in the index, or a ``KeyError`` will be raised.
+When slicing, both the start bound **AND** the stop bound are *included*, if present in the index.
+Integers are valid labels, but they refer to the label **and not the position**.
+
+The ``.loc`` attribute is the primary access method. The following are valid inputs:
+
+* A single label, e.g. ``5`` or ``'a'`` (Note that ``5`` is interpreted as a *label* of the index. This use is **not** an integer position along the index.).
+* A list or array of labels ``['a', 'b', 'c']``.
+* A slice object with labels ``'a':'f'`` (Note that contrary to usual Python
+  slices, **both** the start and the stop are included, when present in the
+  index! See :ref:`Slicing with labels <indexing.slicing_with_labels>`.
+* A boolean array.
+* A ``callable``, see :ref:`Selection By Callable <indexing.callable>`.
+
+.. ipython:: python
+
+   s1 = pd.Series(np.random.randn(6), index=list('abcdef'))
+   s1
+   s1.loc['c':]
+   s1.loc['b']
+
+Note that setting works as well:
+
+.. ipython:: python
+
+   s1.loc['c':] = 0
+   s1
+
+With a DataFrame:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.random.randn(6, 4),
+                      index=list('abcdef'),
+                      columns=list('ABCD'))
+   df1
+   df1.loc[['a', 'b', 'd'], :]
+
+Accessing via label slices:
+
+.. ipython:: python
+
+   df1.loc['d':, 'A':'C']
+
+For getting a cross section using a label (equivalent to ``df.xs('a')``):
+
+.. ipython:: python
+
+   df1.loc['a']
+
+For getting values with a boolean array:
+
+.. ipython:: python
+
+   df1.loc['a'] > 0
+   df1.loc[:, df1.loc['a'] > 0]
+
+NA values in a boolean array propagate as ``False``:
+
+.. ipython:: python
+
+   mask = pd.array([True, False, True, False, pd.NA, False], dtype="boolean")
+   mask
+   df1[mask]
+
+For getting a value explicitly:
+
+.. ipython:: python
+
+   # this is also equivalent to ``df1.at['a','A']``
+   df1.loc['a', 'A']
+
+.. _indexing.slicing_with_labels:
+
+Slicing with labels
+~~~~~~~~~~~~~~~~~~~
+
+When using ``.loc`` with slices, if both the start and the stop labels are
+present in the index, then elements *located* between the two (including them)
+are returned:
+
+.. ipython:: python
+
+   s = pd.Series(list('abcde'), index=[0, 3, 2, 5, 4])
+   s.loc[3:5]
+
+If the index is sorted, and can be compared against start and stop labels,
+then slicing will still work as expected, by selecting labels which *rank*
+between the two:
+
+.. ipython:: python
+
+   s.sort_index()
+   s.sort_index().loc[1:6]
+
+However, if at least one of the two is absent *and* the index is not sorted, an
+error will be raised (since doing otherwise would be computationally expensive,
+as well as potentially ambiguous for mixed type indexes). For instance, in the
+above example, ``s.loc[1:6]`` would raise ``KeyError``.
+
+For the rationale behind this behavior, see
+:ref:`Endpoints are inclusive <advanced.endpoints_are_inclusive>`.
+
+.. ipython:: python
+
+   s = pd.Series(list('abcdef'), index=[0, 3, 2, 5, 4, 2])
+   s.loc[3:5]
+
+Also, if the index has duplicate labels *and* either the start or the stop label is duplicated,
+an error will be raised. For instance, in the above example, ``s.loc[2:5]`` would raise a ``KeyError``.
+
+For more information about duplicate labels, see
+:ref:`Duplicate Labels <duplicates>`.
+
+.. _indexing.integer:
+
+Selection by position
+---------------------
+
+pandas provides a suite of methods in order to get **purely integer based indexing**. The semantics follow closely Python and NumPy slicing. These are ``0-based`` indexing. When slicing, the start bound is *included*, while the upper bound is *excluded*. Trying to use a non-integer, even a **valid** label will raise an ``IndexError``.
+
+The ``.iloc`` attribute is the primary access method. The following are valid inputs:
+
+* An integer e.g. ``5``.
+* A list or array of integers ``[4, 3, 0]``.
+* A slice object with ints ``1:7``.
+* A boolean array.
+* A ``callable``, see :ref:`Selection By Callable <indexing.callable>`.
+* A tuple of row (and column) indexes, whose elements are one of the
+  above types.
+
+.. ipython:: python
+
+   s1 = pd.Series(np.random.randn(5), index=list(range(0, 10, 2)))
+   s1
+   s1.iloc[:3]
+   s1.iloc[3]
+
+Note that setting works as well:
+
+.. ipython:: python
+
+   s1.iloc[:3] = 0
+   s1
+
+With a DataFrame:
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.random.randn(6, 4),
+                      index=list(range(0, 12, 2)),
+                      columns=list(range(0, 8, 2)))
+   df1
+
+Select via integer slicing:
+
+.. ipython:: python
+
+   df1.iloc[:3]
+   df1.iloc[1:5, 2:4]
+
+Select via integer list:
+
+.. ipython:: python
+
+   df1.iloc[[1, 3, 5], [1, 3]]
+
+.. ipython:: python
+
+   df1.iloc[1:3, :]
+
+.. ipython:: python
+
+   df1.iloc[:, 1:3]
+
+.. ipython:: python
+
+   # this is also equivalent to ``df1.iat[1,1]``
+   df1.iloc[1, 1]
+
+For getting a cross section using an integer position (equiv to ``df.xs(1)``):
+
+.. ipython:: python
+
+   df1.iloc[1]
+
+Out of range slice indexes are handled gracefully just as in Python/NumPy.
+
+.. ipython:: python
+
+    # these are allowed in Python/NumPy.
+    x = list('abcdef')
+    x
+    x[4:10]
+    x[8:10]
+    s = pd.Series(x)
+    s
+    s.iloc[4:10]
+    s.iloc[8:10]
+
+Note that using slices that go out of bounds can result in
+an empty axis (e.g. an empty DataFrame being returned).
+
+.. ipython:: python
+
+   dfl = pd.DataFrame(np.random.randn(5, 2), columns=list('AB'))
+   dfl
+   dfl.iloc[:, 2:3]
+   dfl.iloc[:, 1:3]
+   dfl.iloc[4:6]
+
+A single indexer that is out of bounds will raise an ``IndexError``.
+A list of indexers where any element is out of bounds will raise an
+``IndexError``.
+
+.. ipython:: python
+   :okexcept:
+
+   dfl.iloc[[4, 5, 6]]
+
+.. ipython:: python
+   :okexcept:
+
+   dfl.iloc[:, 4]
+
+.. _indexing.callable:
+
+Selection by callable
+---------------------
+
+``.loc``, ``.iloc``, and also ``[]`` indexing can accept a ``callable`` as indexer.
+The ``callable`` must be a function with one argument (the calling Series or DataFrame) that returns valid output for indexing.
+
+.. note::
+
+   For ``.iloc`` indexing, returning a tuple from the callable is
+   not supported, since tuple destructuring for row and column indexes
+   occurs *before* applying callables.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.random.randn(6, 4),
+                      index=list('abcdef'),
+                      columns=list('ABCD'))
+   df1
+
+   df1.loc[lambda df: df['A'] > 0, :]
+   df1.loc[:, lambda df: ['A', 'B']]
+
+   df1.iloc[:, lambda df: [0, 1]]
+
+   df1[lambda df: df.columns[0]]
+
+
+You can use callable indexing in ``Series``.
+
+.. ipython:: python
+
+   df1['A'].loc[lambda s: s > 0]
+
+Using these methods / indexers, you can chain data selection operations
+without using a temporary variable.
+
+.. ipython:: python
+
+   bb = pd.read_csv('data/baseball.csv', index_col='id')
+   (bb.groupby(['year', 'team']).sum(numeric_only=True)
+      .loc[lambda df: df['r'] > 100])
+
+
+.. _combining_positional_and_label_based_indexing:
+
+Combining positional and label-based indexing
+---------------------------------------------
+
+If you wish to get the 0th and the 2nd elements from the index in the 'A' column, you can do:
+
+.. ipython:: python
+
+  dfd = pd.DataFrame({'A': [1, 2, 3],
+                      'B': [4, 5, 6]},
+                     index=list('abc'))
+  dfd
+  dfd.loc[dfd.index[[0, 2]], 'A']
+
+This can also be expressed using ``.iloc``, by explicitly getting locations on the indexers, and using
+*positional* indexing to select things.
+
+.. ipython:: python
+
+  dfd.iloc[[0, 2], dfd.columns.get_loc('A')]
+
+For getting *multiple* indexers, using ``.get_indexer``:
+
+.. ipython:: python
+
+  dfd.iloc[[0, 2], dfd.columns.get_indexer(['A', 'B'])]
+
+
+Reindexing
+~~~~~~~~~~
+
+The idiomatic way to achieve selecting potentially not-found elements is via ``.reindex()``. See also the section on :ref:`reindexing <basics.reindexing>`.
+
+.. ipython:: python
+
+  s = pd.Series([1, 2, 3])
+  s.reindex([1, 2, 3])
+
+Alternatively, if you want to select only *valid* keys, the following is idiomatic and efficient; it is guaranteed to preserve the dtype of the selection.
+
+.. ipython:: python
+
+   labels = [1, 2, 3]
+   s.loc[s.index.intersection(labels)]
+
+Having a duplicated index will raise for a ``.reindex()``:
+
+.. ipython:: python
+   :okexcept:
+
+   s = pd.Series(np.arange(4), index=['a', 'a', 'b', 'c'])
+   labels = ['c', 'd']
+   s.reindex(labels)
+
+Generally, you can intersect the desired labels with the current
+axis, and then reindex.
+
+.. ipython:: python
+
+   s.loc[s.index.intersection(labels)].reindex(labels)
+
+However, this would *still* raise if your resulting index is duplicated.
+
+.. ipython:: python
+   :okexcept:
+
+   labels = ['a', 'd']
+   s.loc[s.index.intersection(labels)].reindex(labels)
+
+
+.. _indexing.basics.partial_setting:
+
+Selecting random samples
+------------------------
+
+A random selection of rows or columns from a Series or DataFrame with the :meth:`~DataFrame.sample` method. The method will sample rows by default, and accepts a specific number of rows/columns to return, or a fraction of rows.
+
+.. ipython:: python
+
+    s = pd.Series([0, 1, 2, 3, 4, 5])
+
+    # When no arguments are passed, returns 1 row.
+    s.sample()
+
+    # One may specify either a number of rows:
+    s.sample(n=3)
+
+    # Or a fraction of the rows:
+    s.sample(frac=0.5)
+
+By default, ``sample`` will return each row at most once, but one can also sample with replacement
+using the ``replace`` option:
+
+.. ipython:: python
+
+    s = pd.Series([0, 1, 2, 3, 4, 5])
+
+    # Without replacement (default):
+    s.sample(n=6, replace=False)
+
+    # With replacement:
+    s.sample(n=6, replace=True)
+
+
+By default, each row has an equal probability of being selected, but if you want rows
+to have different probabilities, you can pass the ``sample`` function sampling weights as
+``weights``. These weights can be a list, a NumPy array, or a Series, but they must be of the same length as the object you are sampling. Missing values will be treated as a weight of zero, and inf values are not allowed. If weights do not sum to 1, they will be re-normalized by dividing all weights by the sum of the weights. For example:
+
+.. ipython:: python
+
+    s = pd.Series([0, 1, 2, 3, 4, 5])
+    example_weights = [0, 0, 0.2, 0.2, 0.2, 0.4]
+    s.sample(n=3, weights=example_weights)
+
+    # Weights will be re-normalized automatically
+    example_weights2 = [0.5, 0, 0, 0, 0, 0]
+    s.sample(n=1, weights=example_weights2)
+
+When applied to a DataFrame, you can use a column of the DataFrame as sampling weights
+(provided you are sampling rows and not columns) by simply passing the name of the column
+as a string.
+
+.. ipython:: python
+
+    df2 = pd.DataFrame({'col1': [9, 8, 7, 6],
+                        'weight_column': [0.5, 0.4, 0.1, 0]})
+    df2.sample(n=3, weights='weight_column')
+
+``sample`` also allows users to sample columns instead of rows using the ``axis`` argument.
+
+.. ipython:: python
+
+    df3 = pd.DataFrame({'col1': [1, 2, 3], 'col2': [2, 3, 4]})
+    df3.sample(n=1, axis=1)
+
+Finally, one can also set a seed for ``sample``'s random number generator using the ``random_state`` argument, which will accept either an integer (as a seed) or a NumPy RandomState object.
+
+.. ipython:: python
+
+    df4 = pd.DataFrame({'col1': [1, 2, 3], 'col2': [2, 3, 4]})
+
+    # With a given seed, the sample will always draw the same rows.
+    df4.sample(n=2, random_state=2)
+    df4.sample(n=2, random_state=2)
+
+
+
+Setting with enlargement
+------------------------
+
+The ``.loc/[]`` operations can perform enlargement when setting a non-existent key for that axis.
+
+In the ``Series`` case this is effectively an appending operation.
+
+.. ipython:: python
+
+   se = pd.Series([1, 2, 3])
+   se
+   se[5] = 5.
+   se
+
+A ``DataFrame`` can be enlarged on either axis via ``.loc``.
+
+.. ipython:: python
+
+   dfi = pd.DataFrame(np.arange(6).reshape(3, 2),
+                      columns=['A', 'B'])
+   dfi
+   dfi.loc[:, 'C'] = dfi.loc[:, 'A']
+   dfi
+
+This is like an ``append`` operation on the ``DataFrame``.
+
+.. ipython:: python
+
+   dfi.loc[3] = 5
+   dfi
+
+.. _indexing.basics.get_value:
+
+Fast scalar value getting and setting
+-------------------------------------
+
+Since indexing with ``[]`` must handle a lot of cases (single-label access,
+slicing, boolean indexing, etc.), it has a bit of overhead in order to figure
+out what you're asking for. If you only want to access a scalar value, the
+fastest way is to use the ``at`` and ``iat`` methods, which are implemented on
+all of the data structures.
+
+Similarly to ``loc``, ``at`` provides **label** based scalar lookups, while, ``iat`` provides **integer** based lookups analogously to ``iloc``
+
+.. ipython:: python
+
+   s.iat[5]
+   df.at[dates[5], 'A']
+   df.iat[3, 0]
+
+You can also set using these same indexers.
+
+.. ipython:: python
+
+   df.at[dates[5], 'E'] = 7
+   df.iat[3, 0] = 7
+
+``at`` may enlarge the object in-place as above if the indexer is missing.
+
+.. ipython:: python
+
+   df.at[dates[-1] + pd.Timedelta('1 day'), 0] = 7
+   df
+
+Boolean indexing
+----------------
+
+.. _indexing.boolean:
+
+Another common operation is the use of boolean vectors to filter the data.
+The operators are: ``|`` for ``or``, ``&`` for ``and``, and ``~`` for ``not``.
+These **must** be grouped by using parentheses, since by default Python will
+evaluate an expression such as ``df['A'] > 2 & df['B'] < 3`` as
+``df['A'] > (2 & df['B']) < 3``, while the desired evaluation order is
+``(df['A'] > 2) & (df['B'] < 3)``.
+
+Using a boolean vector to index a Series works exactly as in a NumPy ndarray:
+
+.. ipython:: python
+
+   s = pd.Series(range(-3, 4))
+   s
+   s[s > 0]
+   s[(s < -1) | (s > 0.5)]
+   s[~(s < 0)]
+
+You may select rows from a DataFrame using a boolean vector the same length as
+the DataFrame's index (for example, something derived from one of the columns
+of the DataFrame):
+
+.. ipython:: python
+
+   df[df['A'] > 0]
+
+List comprehensions and the ``map`` method of Series can also be used to produce
+more complex criteria:
+
+.. ipython:: python
+
+   df2 = pd.DataFrame({'a': ['one', 'one', 'two', 'three', 'two', 'one', 'six'],
+                       'b': ['x', 'y', 'y', 'x', 'y', 'x', 'x'],
+                       'c': np.random.randn(7)})
+
+   # only want 'two' or 'three'
+   criterion = df2['a'].map(lambda x: x.startswith('t'))
+
+   df2[criterion]
+
+   # equivalent but slower
+   df2[[x.startswith('t') for x in df2['a']]]
+
+   # Multiple criteria
+   df2[criterion & (df2['b'] == 'x')]
+
+With the choice methods :ref:`Selection by Label <indexing.label>`, :ref:`Selection by Position <indexing.integer>`,
+and :ref:`Advanced Indexing <advanced>` you may select along more than one axis using boolean vectors combined with other indexing expressions.
+
+.. ipython:: python
+
+   df2.loc[criterion & (df2['b'] == 'x'), 'b':'c']
+
+.. warning::
+
+   While ``loc`` supports two kinds of boolean indexing, ``iloc`` only supports indexing with a
+   boolean array. If the indexer is a boolean ``Series``, an error will be raised. For instance,
+   in the following example, ``df.iloc[s.values, 1]`` is ok. The boolean indexer is an array.
+   But ``df.iloc[s, 1]`` would raise ``ValueError``.
+
+   .. ipython:: python
+
+      df = pd.DataFrame([[1, 2], [3, 4], [5, 6]],
+                        index=list('abc'),
+                        columns=['A', 'B'])
+      s = (df['A'] > 2)
+      s
+
+      df.loc[s, 'B']
+
+      df.iloc[s.values, 1]
+
+.. _indexing.basics.indexing_isin:
+
+Indexing with isin
+------------------
+
+Consider the :meth:`~Series.isin` method of ``Series``, which returns a boolean
+vector that is true wherever the ``Series`` elements exist in the passed list.
+This allows you to select rows where one or more columns have values you want:
+
+.. ipython:: python
+
+   s = pd.Series(np.arange(5), index=np.arange(5)[::-1], dtype='int64')
+   s
+   s.isin([2, 4, 6])
+   s[s.isin([2, 4, 6])]
+
+The same method is available for ``Index`` objects and is useful for the cases
+when you don't know which of the sought labels are in fact present:
+
+.. ipython:: python
+
+   s[s.index.isin([2, 4, 6])]
+
+   # compare it to the following
+   s.reindex([2, 4, 6])
+
+In addition to that, ``MultiIndex`` allows selecting a separate level to use
+in the membership check:
+
+.. ipython:: python
+
+   s_mi = pd.Series(np.arange(6),
+                    index=pd.MultiIndex.from_product([[0, 1], ['a', 'b', 'c']]))
+   s_mi
+   s_mi.iloc[s_mi.index.isin([(1, 'a'), (2, 'b'), (0, 'c')])]
+   s_mi.iloc[s_mi.index.isin(['a', 'c', 'e'], level=1)]
+
+DataFrame also has an :meth:`~DataFrame.isin` method.  When calling ``isin``, pass a set of
+values as either an array or dict.  If values is an array, ``isin`` returns
+a DataFrame of booleans that is the same shape as the original DataFrame, with True
+wherever the element is in the sequence of values.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'vals': [1, 2, 3, 4], 'ids': ['a', 'b', 'f', 'n'],
+                      'ids2': ['a', 'n', 'c', 'n']})
+
+   values = ['a', 'b', 1, 3]
+
+   df.isin(values)
+
+Oftentimes you'll want to match certain values with certain columns.
+Just make values a ``dict`` where the key is the column, and the value is
+a list of items you want to check for.
+
+.. ipython:: python
+
+   values = {'ids': ['a', 'b'], 'vals': [1, 3]}
+
+   df.isin(values)
+
+To return the DataFrame of booleans where the values are *not* in the original DataFrame,
+use the ``~`` operator:
+
+.. ipython:: python
+
+   values = {'ids': ['a', 'b'], 'vals': [1, 3]}
+
+   ~df.isin(values)
+
+Combine DataFrame's ``isin`` with the ``any()`` and ``all()`` methods to
+quickly select subsets of your data that meet a given criteria.
+To select a row where each column meets its own criterion:
+
+.. ipython:: python
+
+  values = {'ids': ['a', 'b'], 'ids2': ['a', 'c'], 'vals': [1, 3]}
+
+  row_mask = df.isin(values).all(axis=1)
+
+  df[row_mask]
+
+.. _indexing.where_mask:
+
+The :meth:`~pandas.DataFrame.where` Method and Masking
+------------------------------------------------------
+
+Selecting values from a Series with a boolean vector generally returns a
+subset of the data. To guarantee that selection output has the same shape as
+the original data, you can use the ``where`` method in ``Series`` and ``DataFrame``.
+
+To return only the selected rows:
+
+.. ipython:: python
+
+   s[s > 0]
+
+To return a Series of the same shape as the original:
+
+.. ipython:: python
+
+   s.where(s > 0)
+
+Selecting values from a DataFrame with a boolean criterion now also preserves
+input data shape. ``where`` is used under the hood as the implementation.
+The code below is equivalent to ``df.where(df < 0)``.
+
+.. ipython:: python
+
+   dates = pd.date_range('1/1/2000', periods=8)
+   df = pd.DataFrame(np.random.randn(8, 4),
+                     index=dates, columns=['A', 'B', 'C', 'D'])
+   df[df < 0]
+
+In addition, ``where`` takes an optional ``other`` argument for replacement of
+values where the condition is False, in the returned copy.
+
+.. ipython:: python
+
+   df.where(df < 0, -df)
+
+You may wish to set values based on some boolean criteria.
+This can be done intuitively like so:
+
+.. ipython:: python
+
+   s2 = s.copy()
+   s2[s2 < 0] = 0
+   s2
+
+   df2 = df.copy()
+   df2[df2 < 0] = 0
+   df2
+
+``where`` returns a modified copy of the data.
+
+.. note::
+
+   The signature for :func:`DataFrame.where` differs from :func:`numpy.where`.
+   Roughly ``df1.where(m, df2)`` is equivalent to ``np.where(m, df1, df2)``.
+
+   .. ipython:: python
+
+      df.where(df < 0, -df) == np.where(df < 0, df, -df)
+
+**Alignment**
+
+Furthermore, ``where`` aligns the input boolean condition (ndarray or DataFrame),
+such that partial selection with setting is possible. This is analogous to
+partial setting via ``.loc`` (but on the contents rather than the axis labels).
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2[df2[1:4] > 0] = 3
+   df2
+
+Where can also accept ``axis`` and ``level`` parameters to align the input when
+performing the ``where``.
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2.where(df2 > 0, df2['A'], axis='index')
+
+This is equivalent to (but faster than) the following.
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df.apply(lambda x, y: x.where(x > 0, y), y=df['A'])
+
+``where`` can accept a callable as condition and ``other`` arguments. The function must
+be with one argument (the calling Series or DataFrame) and that returns valid output
+as condition and ``other`` argument.
+
+.. ipython:: python
+
+   df3 = pd.DataFrame({'A': [1, 2, 3],
+                       'B': [4, 5, 6],
+                       'C': [7, 8, 9]})
+   df3.where(lambda x: x > 4, lambda x: x + 10)
+
+Mask
+~~~~
+
+:meth:`~pandas.DataFrame.mask` is the inverse boolean operation of ``where``.
+
+.. ipython:: python
+
+   s.mask(s >= 0)
+   df.mask(df >= 0)
+
+.. _indexing.np_where:
+
+Setting with enlargement conditionally using :func:`numpy`
+----------------------------------------------------------
+
+An alternative to :meth:`~pandas.DataFrame.where` is to use :func:`numpy.where`.
+Combined with setting a new column, you can use it to enlarge a DataFrame where the
+values are determined conditionally.
+
+Consider you have two choices to choose from in the following DataFrame. And you want to
+set a new column color to 'green' when the second column has 'Z'.  You can do the
+following:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'col1': list('ABBC'), 'col2': list('ZZXY')})
+   df['color'] = np.where(df['col2'] == 'Z', 'green', 'red')
+   df
+
+If you have multiple conditions, you can use :func:`numpy.select` to achieve that.  Say
+corresponding to three conditions there are three choice of colors, with a fourth color
+as a fallback, you can do the following.
+
+.. ipython:: python
+
+   conditions = [
+       (df['col2'] == 'Z') & (df['col1'] == 'A'),
+       (df['col2'] == 'Z') & (df['col1'] == 'B'),
+       (df['col1'] == 'B')
+   ]
+   choices = ['yellow', 'blue', 'purple']
+   df['color'] = np.select(conditions, choices, default='black')
+   df
+
+.. _indexing.query:
+
+The :meth:`~pandas.DataFrame.query` Method
+------------------------------------------
+
+:class:`~pandas.DataFrame` objects have a :meth:`~pandas.DataFrame.query`
+method that allows selection using an expression.
+
+You can get the value of the frame where column ``b`` has values
+between the values of columns ``a`` and ``c``. For example:
+
+.. ipython:: python
+
+   n = 10
+   df = pd.DataFrame(np.random.rand(n, 3), columns=list('abc'))
+   df
+
+   # pure python
+   df[(df['a'] < df['b']) & (df['b'] < df['c'])]
+
+   # query
+   df.query('(a < b) & (b < c)')
+
+Do the same thing but fall back on a named index if there is no column
+with the name ``a``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randint(n / 2, size=(n, 2)), columns=list('bc'))
+   df.index.name = 'a'
+   df
+   df.query('a < b and b < c')
+
+If instead you don't want to or cannot name your index, you can use the name
+``index`` in your query expression:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randint(n, size=(n, 2)), columns=list('bc'))
+   df
+   df.query('index < b < c')
+
+.. note::
+
+   If the name of your index overlaps with a column name, the column name is
+   given precedence. For example,
+
+   .. ipython:: python
+
+      df = pd.DataFrame({'a': np.random.randint(5, size=5)})
+      df.index.name = 'a'
+      df.query('a > 2')  # uses the column 'a', not the index
+
+   You can still use the index in a query expression by using the special
+   identifier 'index':
+
+   .. ipython:: python
+
+      df.query('index > 2')
+
+   If for some reason you have a column named ``index``, then you can refer to
+   the index as ``ilevel_0`` as well, but at this point you should consider
+   renaming your columns to something less ambiguous.
+
+:class:`~pandas.MultiIndex` :meth:`~pandas.DataFrame.query` Syntax
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can also use the levels of a ``DataFrame`` with a
+:class:`~pandas.MultiIndex` as if they were columns in the frame:
+
+.. ipython:: python
+
+   n = 10
+   colors = np.random.choice(['red', 'green'], size=n)
+   foods = np.random.choice(['eggs', 'ham'], size=n)
+   colors
+   foods
+
+   index = pd.MultiIndex.from_arrays([colors, foods], names=['color', 'food'])
+   df = pd.DataFrame(np.random.randn(n, 2), index=index)
+   df
+   df.query('color == "red"')
+
+If the levels of the ``MultiIndex`` are unnamed, you can refer to them using
+special names:
+
+.. ipython:: python
+
+   df.index.names = [None, None]
+   df
+   df.query('ilevel_0 == "red"')
+
+
+The convention is ``ilevel_0``, which means "index level 0" for the 0th level
+of the ``index``.
+
+
+:meth:`~pandas.DataFrame.query` Use Cases
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A use case for :meth:`~pandas.DataFrame.query` is when you have a collection of
+:class:`~pandas.DataFrame` objects that have a subset of column names (or index
+levels/names) in common. You can pass the same query to both frames *without*
+having to specify which frame you're interested in querying
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.rand(n, 3), columns=list('abc'))
+   df
+   df2 = pd.DataFrame(np.random.rand(n + 2, 3), columns=df.columns)
+   df2
+   expr = '0.0 <= a <= c <= 0.5'
+   map(lambda frame: frame.query(expr), [df, df2])
+
+:meth:`~pandas.DataFrame.query` Python versus pandas Syntax Comparison
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Full numpy-like syntax:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randint(n, size=(n, 3)), columns=list('abc'))
+   df
+   df.query('(a < b) & (b < c)')
+   df[(df['a'] < df['b']) & (df['b'] < df['c'])]
+
+Slightly nicer by removing the parentheses (comparison operators bind tighter
+than ``&`` and ``|``):
+
+.. ipython:: python
+
+   df.query('a < b & b < c')
+
+Use English instead of symbols:
+
+.. ipython:: python
+
+   df.query('a < b and b < c')
+
+Pretty close to how you might write it on paper:
+
+.. ipython:: python
+
+   df.query('a < b < c')
+
+The ``in`` and ``not in`` operators
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`~pandas.DataFrame.query` also supports special use of Python's ``in`` and
+``not in`` comparison operators, providing a succinct syntax for calling the
+``isin`` method of a ``Series`` or ``DataFrame``.
+
+.. ipython:: python
+
+   # get all rows where columns "a" and "b" have overlapping values
+   df = pd.DataFrame({'a': list('aabbccddeeff'), 'b': list('aaaabbbbcccc'),
+                      'c': np.random.randint(5, size=12),
+                      'd': np.random.randint(9, size=12)})
+   df
+   df.query('a in b')
+
+   # How you'd do it in pure Python
+   df[df['a'].isin(df['b'])]
+
+   df.query('a not in b')
+
+   # pure Python
+   df[~df['a'].isin(df['b'])]
+
+
+You can combine this with other expressions for very succinct queries:
+
+
+.. ipython:: python
+
+   # rows where cols a and b have overlapping values
+   # and col c's values are less than col d's
+   df.query('a in b and c < d')
+
+   # pure Python
+   df[df['b'].isin(df['a']) & (df['c'] < df['d'])]
+
+
+.. note::
+
+   Note that ``in`` and ``not in`` are evaluated in Python, since ``numexpr``
+   has no equivalent of this operation. However, **only the** ``in``/``not in``
+   **expression itself** is evaluated in vanilla Python. For example, in the
+   expression
+
+   .. code-block:: python
+
+      df.query('a in b + c + d')
+
+   ``(b + c + d)`` is evaluated by ``numexpr`` and *then* the ``in``
+   operation is evaluated in plain Python. In general, any operations that can
+   be evaluated using ``numexpr`` will be.
+
+Special use of the ``==`` operator with ``list`` objects
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Comparing a ``list`` of values to a column using ``==``/``!=`` works similarly
+to ``in``/``not in``.
+
+.. ipython:: python
+
+   df.query('b == ["a", "b", "c"]')
+
+   # pure Python
+   df[df['b'].isin(["a", "b", "c"])]
+
+   df.query('c == [1, 2]')
+
+   df.query('c != [1, 2]')
+
+   # using in/not in
+   df.query('[1, 2] in c')
+
+   df.query('[1, 2] not in c')
+
+   # pure Python
+   df[df['c'].isin([1, 2])]
+
+
+Boolean operators
+~~~~~~~~~~~~~~~~~
+
+You can negate boolean expressions with the word ``not`` or the ``~`` operator.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.rand(n, 3), columns=list('abc'))
+   df['bools'] = np.random.rand(len(df)) > 0.5
+   df.query('~bools')
+   df.query('not bools')
+   df.query('not bools') == df[~df['bools']]
+
+Of course, expressions can be arbitrarily complex too:
+
+.. ipython:: python
+
+   # short query syntax
+   shorter = df.query('a < b < c and (not bools) or bools > 2')
+
+   # equivalent in pure Python
+   longer = df[(df['a'] < df['b'])
+               & (df['b'] < df['c'])
+               & (~df['bools'])
+               | (df['bools'] > 2)]
+
+   shorter
+   longer
+
+   shorter == longer
+
+
+Performance of :meth:`~pandas.DataFrame.query`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``DataFrame.query()`` using ``numexpr`` is slightly faster than Python for
+large frames.
+
+..
+    The eval-perf.png figure below was generated with /doc/scripts/eval_performance.py
+
+.. image:: ../_static/query-perf.png
+
+
+
+You will only see the performance benefits of using the ``numexpr`` engine
+with ``DataFrame.query()`` if your frame has more than approximately 100,000
+rows.
+
+
+
+This plot was created using a ``DataFrame`` with 3 columns each containing
+floating point values generated using ``numpy.random.randn()``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(8, 4),
+                     index=dates, columns=['A', 'B', 'C', 'D'])
+   df2 = df.copy()
+
+
+Duplicate data
+--------------
+
+.. _indexing.duplicate:
+
+If you want to identify and remove duplicate rows in a DataFrame,  there are
+two methods that will help: ``duplicated`` and ``drop_duplicates``. Each
+takes as an argument the columns to use to identify duplicated rows.
+
+* ``duplicated`` returns a boolean vector whose length is the number of rows, and which indicates whether a row is duplicated.
+* ``drop_duplicates`` removes duplicate rows.
+
+By default, the first observed row of a duplicate set is considered unique, but
+each method has a ``keep`` parameter to specify targets to be kept.
+
+* ``keep='first'`` (default): mark / drop duplicates except for the first occurrence.
+* ``keep='last'``: mark / drop duplicates except for the last occurrence.
+* ``keep=False``: mark  / drop all duplicates.
+
+.. ipython:: python
+
+   df2 = pd.DataFrame({'a': ['one', 'one', 'two', 'two', 'two', 'three', 'four'],
+                       'b': ['x', 'y', 'x', 'y', 'x', 'x', 'x'],
+                       'c': np.random.randn(7)})
+   df2
+   df2.duplicated('a')
+   df2.duplicated('a', keep='last')
+   df2.duplicated('a', keep=False)
+   df2.drop_duplicates('a')
+   df2.drop_duplicates('a', keep='last')
+   df2.drop_duplicates('a', keep=False)
+
+Also, you can pass a list of columns to identify duplications.
+
+.. ipython:: python
+
+   df2.duplicated(['a', 'b'])
+   df2.drop_duplicates(['a', 'b'])
+
+To drop duplicates by index value, use ``Index.duplicated`` then perform slicing.
+The same set of options are available for the ``keep`` parameter.
+
+.. ipython:: python
+
+   df3 = pd.DataFrame({'a': np.arange(6),
+                       'b': np.random.randn(6)},
+                      index=['a', 'a', 'b', 'c', 'b', 'a'])
+   df3
+   df3.index.duplicated()
+   df3[~df3.index.duplicated()]
+   df3[~df3.index.duplicated(keep='last')]
+   df3[~df3.index.duplicated(keep=False)]
+
+.. _indexing.dictionarylike:
+
+Dictionary-like :meth:`~pandas.DataFrame.get` method
+----------------------------------------------------
+
+Each of Series or DataFrame have a ``get`` method which can return a
+default value.
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=['a', 'b', 'c'])
+   s.get('a')  # equivalent to s['a']
+   s.get('x', default=-1)
+
+.. _indexing.lookup:
+
+Looking up values by index/column labels
+----------------------------------------
+
+Sometimes you want to extract a set of values given a sequence of row labels
+and column labels, this can be achieved by ``pandas.factorize``  and NumPy indexing.
+For instance:
+
+.. ipython:: python
+
+    df = pd.DataFrame({'col': ["A", "A", "B", "B"],
+                       'A': [80, 23, np.nan, 22],
+                       'B': [80, 55, 76, 67]})
+    df
+    idx, cols = pd.factorize(df['col'])
+    df.reindex(cols, axis=1).to_numpy()[np.arange(len(df)), idx]
+
+Formerly this could be achieved with the dedicated ``DataFrame.lookup`` method
+which was deprecated in version 1.2.0 and removed in version 2.0.0.
+
+.. _indexing.class:
+
+Index objects
+-------------
+
+The pandas :class:`~pandas.Index` class and its subclasses can be viewed as
+implementing an *ordered multiset*. Duplicates are allowed.
+
+:class:`~pandas.Index` also provides the infrastructure necessary for
+lookups, data alignment, and reindexing. The easiest way to create an
+:class:`~pandas.Index` directly is to pass a ``list`` or other sequence to
+:class:`~pandas.Index`:
+
+.. ipython:: python
+
+   index = pd.Index(['e', 'd', 'a', 'b'])
+   index
+   'd' in index
+
+or using numbers:
+
+.. ipython:: python
+
+   index = pd.Index([1, 5, 12])
+   index
+   5 in index
+
+If no dtype is given, ``Index`` tries to infer the dtype from the data.
+It is also possible to give an explicit dtype when instantiating an :class:`Index`:
+
+.. ipython:: python
+
+   index = pd.Index(['e', 'd', 'a', 'b'], dtype="string")
+   index
+   index = pd.Index([1, 5, 12], dtype="int8")
+   index
+   index = pd.Index([1, 5, 12], dtype="float32")
+   index
+
+You can also pass a ``name`` to be stored in the index:
+
+.. ipython:: python
+
+   index = pd.Index(['e', 'd', 'a', 'b'], name='something')
+   index.name
+
+The name, if set, will be shown in the console display:
+
+.. ipython:: python
+
+   index = pd.Index(list(range(5)), name='rows')
+   columns = pd.Index(['A', 'B', 'C'], name='cols')
+   df = pd.DataFrame(np.random.randn(5, 3), index=index, columns=columns)
+   df
+   df['A']
+
+.. _indexing.set_metadata:
+
+Setting metadata
+~~~~~~~~~~~~~~~~
+
+Indexes are "mostly immutable", but it is possible to set and change their
+``name`` attribute. You can use the ``rename``, ``set_names`` to set these attributes
+directly, and they default to returning a copy.
+
+See :ref:`Advanced Indexing <advanced>` for usage of MultiIndexes.
+
+.. ipython:: python
+
+  ind = pd.Index([1, 2, 3])
+  ind.rename("apple")
+  ind
+  ind = ind.set_names(["apple"])
+  ind.name = "bob"
+  ind
+
+``set_names``, ``set_levels``, and ``set_codes`` also take an optional
+``level`` argument
+
+.. ipython:: python
+
+  index = pd.MultiIndex.from_product([range(3), ['one', 'two']], names=['first', 'second'])
+  index
+  index.levels[1]
+  index.set_levels(["a", "b"], level=1)
+
+.. _indexing.set_ops:
+
+Set operations on Index objects
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The two main operations are ``union`` and ``intersection``.
+Difference is provided via the ``.difference()`` method.
+
+.. ipython:: python
+
+   a = pd.Index(['c', 'b', 'a'])
+   b = pd.Index(['c', 'e', 'd'])
+   a.difference(b)
+
+Also available is the ``symmetric_difference`` operation, which returns elements
+that appear in either ``idx1`` or ``idx2``, but not in both. This is
+equivalent to the Index created by ``idx1.difference(idx2).union(idx2.difference(idx1))``,
+with duplicates dropped.
+
+.. ipython:: python
+
+   idx1 = pd.Index([1, 2, 3, 4])
+   idx2 = pd.Index([2, 3, 4, 5])
+   idx1.symmetric_difference(idx2)
+
+.. note::
+
+   The resulting index from a set operation will be sorted in ascending order.
+
+When performing :meth:`Index.union` between indexes with different dtypes, the indexes
+must be cast to a common dtype. Typically, though not always, this is object dtype. The
+exception is when performing a union between integer and float data. In this case, the
+integer values are converted to float
+
+.. ipython:: python
+
+   idx1 = pd.Index([0, 1, 2])
+   idx2 = pd.Index([0.5, 1.5])
+   idx1.union(idx2)
+
+.. _indexing.missing:
+
+Missing values
+~~~~~~~~~~~~~~
+
+.. important::
+
+   Even though ``Index`` can hold missing values (``NaN``), it should be avoided
+   if you do not want any unexpected results. For example, some operations
+   exclude missing values implicitly.
+
+``Index.fillna`` fills missing values with specified scalar value.
+
+.. ipython:: python
+
+   idx1 = pd.Index([1, np.nan, 3, 4])
+   idx1
+   idx1.fillna(2)
+
+   idx2 = pd.DatetimeIndex([pd.Timestamp('2011-01-01'),
+                            pd.NaT,
+                            pd.Timestamp('2011-01-03')])
+   idx2
+   idx2.fillna(pd.Timestamp('2011-01-02'))
+
+Set / reset index
+-----------------
+
+Occasionally you will load or create a data set into a DataFrame and want to
+add an index after you've already done so. There are a couple of different
+ways.
+
+.. _indexing.set_index:
+
+Set an index
+~~~~~~~~~~~~
+
+DataFrame has a :meth:`~DataFrame.set_index` method which takes a column name
+(for a regular ``Index``) or a list of column names (for a ``MultiIndex``).
+To create a new, re-indexed DataFrame:
+
+.. ipython:: python
+
+   data = pd.DataFrame({'a': ['bar', 'bar', 'foo', 'foo'],
+                        'b': ['one', 'two', 'one', 'two'],
+                        'c': ['z', 'y', 'x', 'w'],
+                        'd': [1., 2., 3, 4]})
+   data
+   indexed1 = data.set_index('c')
+   indexed1
+   indexed2 = data.set_index(['a', 'b'])
+   indexed2
+
+The ``append`` keyword option allow you to keep the existing index and append
+the given columns to a MultiIndex:
+
+.. ipython:: python
+
+   frame = data.set_index('c', drop=False)
+   frame = frame.set_index(['a', 'b'], append=True)
+   frame
+
+Other options in ``set_index`` allow you not drop the index columns.
+
+.. ipython:: python
+
+   data.set_index('c', drop=False)
+
+Reset the index
+~~~~~~~~~~~~~~~
+
+As a convenience, there is a new function on DataFrame called
+:meth:`~DataFrame.reset_index` which transfers the index values into the
+DataFrame's columns and sets a simple integer index.
+This is the inverse operation of :meth:`~DataFrame.set_index`.
+
+
+.. ipython:: python
+
+   data
+   data.reset_index()
+
+The output is more similar to a SQL table or a record array. The names for the
+columns derived from the index are the ones stored in the ``names`` attribute.
+
+You can use the ``level`` keyword to remove only a portion of the index:
+
+.. ipython:: python
+
+   frame
+   frame.reset_index(level=1)
+
+
+``reset_index`` takes an optional parameter ``drop`` which if true simply
+discards the index, instead of putting index values in the DataFrame's columns.
+
+Adding an ad hoc index
+~~~~~~~~~~~~~~~~~~~~~~
+
+You can assign a custom index to the ``index`` attribute:
+
+.. ipython:: python
+
+   df_idx = pd.DataFrame(range(4))
+   df_idx.index = pd.Index([10, 20, 30, 40], name="a")
+   df_idx
+
+Why does assignment fail when using chained indexing?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:ref:`Copy-on-Write <copy_on_write>` is the new default with pandas 3.0.
+This means that chained indexing will never work.
+See :ref:`this section <copy_on_write_chained_assignment>`
+for more context.
diff --git a/pandas/doc/source/user_guide/integer_na.rst b/pandas/doc/source/user_guide/integer_na.rst
new file mode 100644
index 0000000000000..8d35d1583d3bd
--- /dev/null
+++ b/pandas/doc/source/user_guide/integer_na.rst
@@ -0,0 +1,160 @@
+.. currentmodule:: pandas
+
+{{ header }}
+
+.. _integer_na:
+
+**************************
+Nullable integer data type
+**************************
+
+.. note::
+
+   IntegerArray is currently experimental. Its API or implementation may
+   change without warning. Uses :attr:`pandas.NA` as the missing value.
+
+In :ref:`missing_data`, we saw that pandas primarily uses ``NaN`` to represent
+missing data. Because ``NaN`` is a float, this forces an array of integers with
+any missing values to become floating point. In some cases, this may not matter
+much. But if your integer column is, say, an identifier, casting to float can
+be problematic. Some integers cannot even be represented as floating point
+numbers.
+
+Construction
+------------
+
+pandas can represent integer data with possibly missing values using
+:class:`arrays.IntegerArray`. This is an :ref:`extension type <extending.extension-types>`
+implemented within pandas.
+
+.. ipython:: python
+
+   arr = pd.array([1, 2, None], dtype=pd.Int64Dtype())
+   arr
+
+Or the string alias ``"Int64"`` (note the capital ``"I"``) to differentiate from
+NumPy's ``'int64'`` dtype:
+
+.. ipython:: python
+
+   pd.array([1, 2, np.nan], dtype="Int64")
+
+All NA-like values are replaced with :attr:`pandas.NA`.
+
+.. ipython:: python
+
+   pd.array([1, 2, np.nan, None, pd.NA], dtype="Int64")
+
+This array can be stored in a :class:`DataFrame` or :class:`Series` like any
+NumPy array.
+
+.. ipython:: python
+
+   pd.Series(arr)
+
+You can also pass the list-like object to the :class:`Series` constructor
+with the dtype.
+
+.. warning::
+
+   Currently :meth:`pandas.array` and :meth:`pandas.Series` use different
+   rules for dtype inference. :meth:`pandas.array` will infer a
+   nullable-integer dtype
+
+   .. ipython:: python
+
+      pd.array([1, None])
+      pd.array([1, 2])
+
+   For backwards-compatibility, :class:`Series` infers these as either
+   integer or float dtype.
+
+   .. ipython:: python
+
+      pd.Series([1, None])
+      pd.Series([1, 2])
+
+   We recommend explicitly providing the dtype to avoid confusion.
+
+   .. ipython:: python
+
+      pd.array([1, None], dtype="Int64")
+      pd.Series([1, None], dtype="Int64")
+
+   In the future, we may provide an option for :class:`Series` to infer a
+   nullable-integer dtype.
+
+If you create a column of ``NA`` values (for example to fill them later)
+with ``df['new_col'] = pd.NA``, the ``dtype`` would be set to ``object`` in the
+new column. The performance on this column will be worse than with
+the appropriate type. It's better to use
+``df['new_col'] = pd.Series(pd.NA, dtype="Int64")``
+(or another ``dtype`` that supports ``NA``).
+
+.. ipython:: python
+
+   df = pd.DataFrame()
+   df['objects'] = pd.NA
+   df.dtypes
+
+Operations
+----------
+
+Operations involving an integer array will behave similar to NumPy arrays.
+Missing values will be propagated, and the data will be coerced to another
+dtype if needed.
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, None], dtype="Int64")
+
+   # arithmetic
+   s + 1
+
+   # comparison
+   s == 1
+
+   # slicing operation
+   s.iloc[1:3]
+
+   # operate with other dtypes
+   s + s.iloc[1:3].astype("Int8")
+
+   # coerce when needed
+   s + 0.01
+
+These dtypes can operate as part of a ``DataFrame``.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": s, "B": [1, 1, 3], "C": list("aab")})
+   df
+   df.dtypes
+
+
+These dtypes can be merged, reshaped & casted.
+
+.. ipython:: python
+
+   pd.concat([df[["A"]], df[["B", "C"]]], axis=1).dtypes
+   df["A"].astype(float)
+
+Reduction and groupby operations such as :meth:`~DataFrame.sum` work as well.
+
+.. ipython:: python
+
+   df.sum(numeric_only=True)
+   df.sum()
+   df.groupby("B").A.sum()
+
+Scalar NA value
+---------------
+
+:class:`arrays.IntegerArray` uses :attr:`pandas.NA` as its scalar
+missing value. Slicing a single element that's missing will return
+:attr:`pandas.NA`
+
+.. ipython:: python
+
+   a = pd.array([1, None], dtype="Int64")
+   a[1]
diff --git a/pandas/doc/source/user_guide/io.rst b/pandas/doc/source/user_guide/io.rst
new file mode 100644
index 0000000000000..07d06f61b3fd6
--- /dev/null
+++ b/pandas/doc/source/user_guide/io.rst
@@ -0,0 +1,6487 @@
+.. _io:
+
+.. currentmodule:: pandas
+
+
+===============================
+IO tools (text, CSV, HDF5, ...)
+===============================
+
+The pandas I/O API is a set of top level ``reader`` functions accessed like
+:func:`pandas.read_csv` that generally return a pandas object. The corresponding
+``writer`` functions are object methods that are accessed like
+:meth:`DataFrame.to_csv`. Below is a table containing available ``readers`` and
+``writers``.
+
+.. csv-table::
+    :header: "Format Type", "Data Description", "Reader", "Writer"
+    :widths: 30, 100, 60, 60
+
+    text,`CSV <https://en.wikipedia.org/wiki/Comma-separated_values>`__, :ref:`read_csv<io.read_csv_table>`, :ref:`to_csv<io.store_in_csv>`
+    text,Fixed-Width Text File, :ref:`read_fwf<io.fwf_reader>` , NA
+    text,`JSON <https://www.json.org/>`__, :ref:`read_json<io.json_reader>`, :ref:`to_json<io.json_writer>`
+    text,`HTML <https://en.wikipedia.org/wiki/HTML>`__, :ref:`read_html<io.read_html>`, :ref:`to_html<io.html>`
+    text,`LaTeX <https://en.wikipedia.org/wiki/LaTeX>`__, :ref:`Styler.to_latex<io.latex>` , NA
+    text,`XML <https://www.w3.org/standards/xml/core>`__, :ref:`read_xml<io.read_xml>`, :ref:`to_xml<io.xml>`
+    text, Local clipboard, :ref:`read_clipboard<io.clipboard>`, :ref:`to_clipboard<io.clipboard>`
+    binary,`MS Excel <https://en.wikipedia.org/wiki/Microsoft_Excel>`__ , :ref:`read_excel<io.excel_reader>`, :ref:`to_excel<io.excel_writer>`
+    binary,`OpenDocument <http://opendocumentformat.org>`__, :ref:`read_excel<io.ods>`, NA
+    binary,`HDF5 Format <https://support.hdfgroup.org/HDF5/whatishdf5.html>`__, :ref:`read_hdf<io.hdf5>`, :ref:`to_hdf<io.hdf5>`
+    binary,`Feather Format <https://github.com/wesm/feather>`__, :ref:`read_feather<io.feather>`, :ref:`to_feather<io.feather>`
+    binary,`Parquet Format <https://parquet.apache.org/>`__, :ref:`read_parquet<io.parquet>`, :ref:`to_parquet<io.parquet>`
+    binary,`ORC Format <https://orc.apache.org/>`__, :ref:`read_orc<io.orc>`, :ref:`to_orc<io.orc>`
+    binary,`Stata <https://en.wikipedia.org/wiki/Stata>`__, :ref:`read_stata<io.stata_reader>`, :ref:`to_stata<io.stata_writer>`
+    binary,`SAS <https://en.wikipedia.org/wiki/SAS_(software)>`__, :ref:`read_sas<io.sas_reader>` , NA
+    binary,`SPSS <https://en.wikipedia.org/wiki/SPSS>`__, :ref:`read_spss<io.spss_reader>` , NA
+    binary,`Python Pickle Format <https://docs.python.org/3/library/pickle.html>`__, :ref:`read_pickle<io.pickle>`, :ref:`to_pickle<io.pickle>`
+    SQL,`SQL <https://en.wikipedia.org/wiki/SQL>`__, :ref:`read_sql<io.sql>`,:ref:`to_sql<io.sql>`
+
+:ref:`Here <io.perf>` is an informal performance comparison for some of these IO methods.
+
+.. note::
+   For examples that use the ``StringIO`` class, make sure you import it
+   with ``from io import StringIO`` for Python 3.
+
+.. _io.read_csv_table:
+
+CSV & text files
+----------------
+
+The workhorse function for reading text files (a.k.a. flat files) is
+:func:`read_csv`. See the :ref:`cookbook<cookbook.csv>` for some advanced strategies.
+
+Parsing options
+'''''''''''''''
+
+:func:`read_csv` accepts the following common arguments:
+
+Basic
++++++
+
+filepath_or_buffer : various
+  Either a path to a file (a :class:`python:str`, :class:`python:pathlib.Path`)
+  URL (including http, ftp, and S3
+  locations), or any object with a ``read()`` method (such as an open file or
+  :class:`~python:io.StringIO`).
+sep : str, defaults to ``','`` for :func:`read_csv`, ``\t`` for :func:`read_table`
+  Delimiter to use. If sep is ``None``, the C engine cannot automatically detect
+  the separator, but the Python parsing engine can, meaning the latter will be
+  used and automatically detect the separator by Python's builtin sniffer tool,
+  :class:`python:csv.Sniffer`. In addition, separators longer than 1 character and
+  different from ``'\s+'`` will be interpreted as regular expressions and
+  will also force the use of the Python parsing engine. Note that regex
+  delimiters are prone to ignoring quoted data. Regex example: ``'\\r\\t'``.
+delimiter : str, default ``None``
+  Alternative argument name for sep.
+
+Column and index locations and names
+++++++++++++++++++++++++++++++++++++
+
+header : int or list of ints, default ``'infer'``
+  Row number(s) to use as the column names, and the start of the
+  data. Default behavior is to infer the column names: if no names are
+  passed the behavior is identical to ``header=0`` and column names
+  are inferred from the first line of the file, if column names are
+  passed explicitly then the behavior is identical to
+  ``header=None``. Explicitly pass ``header=0`` to be able to replace
+  existing names.
+
+  The header can be a list of ints that specify row locations
+  for a MultiIndex on the columns e.g. ``[0,1,3]``. Intervening rows
+  that are not specified will be skipped (e.g. 2 in this example is
+  skipped). Note that this parameter ignores commented lines and empty
+  lines if ``skip_blank_lines=True``, so header=0 denotes the first
+  line of data rather than the first line of the file.
+names : array-like, default ``None``
+  List of column names to use. If file contains no header row, then you should
+  explicitly pass ``header=None``. Duplicates in this list are not allowed.
+index_col : int, str, sequence of int / str, or False, optional, default ``None``
+  Column(s) to use as the row labels of the ``DataFrame``, either given as
+  string name or column index. If a sequence of int / str is given, a
+  MultiIndex is used.
+
+  .. note::
+     ``index_col=False`` can be used to force pandas to *not* use the first
+     column as the index, e.g. when you have a malformed file with delimiters at
+     the end of each line.
+
+  The default value of ``None`` instructs pandas to guess. If the number of
+  fields in the column header row is equal to the number of fields in the body
+  of the data file, then a default index is used.  If it is larger, then
+  the first columns are used as index so that the remaining number of fields in
+  the body are equal to the number of fields in the header.
+
+  The first row after the header is used to determine the number of columns,
+  which will go into the index. If the subsequent rows contain less columns
+  than the first row, they are filled with ``NaN``.
+
+  This can be avoided through ``usecols``. This ensures that the columns are
+  taken as is and the trailing data are ignored.
+usecols : list-like or callable, default ``None``
+  Return a subset of the columns. If list-like, all elements must either
+  be positional (i.e. integer indices into the document columns) or strings
+  that correspond to column names provided either by the user in ``names`` or
+  inferred from the document header row(s). If ``names`` are given, the document
+  header row(s) are not taken into account. For example, a valid list-like
+  ``usecols`` parameter would be ``[0, 1, 2]`` or ``['foo', 'bar', 'baz']``.
+
+  Element order is ignored, so ``usecols=[0, 1]`` is the same as ``[1, 0]``. To
+  instantiate a DataFrame from ``data`` with element order preserved use
+  ``pd.read_csv(data, usecols=['foo', 'bar'])[['foo', 'bar']]`` for columns
+  in ``['foo', 'bar']`` order or
+  ``pd.read_csv(data, usecols=['foo', 'bar'])[['bar', 'foo']]`` for
+  ``['bar', 'foo']`` order.
+
+  If callable, the callable function will be evaluated against the column names,
+  returning names where the callable function evaluates to True:
+
+  .. ipython:: python
+
+     import pandas as pd
+     from io import StringIO
+
+     data = "col1,col2,col3\na,b,1\na,b,2\nc,d,3"
+     pd.read_csv(StringIO(data))
+     pd.read_csv(StringIO(data), usecols=lambda x: x.upper() in ["COL1", "COL3"])
+
+  Using this parameter results in much faster parsing time and lower memory usage
+  when using the c engine. The Python engine loads the data first before deciding
+  which columns to drop.
+
+General parsing configuration
++++++++++++++++++++++++++++++
+
+dtype : Type name or dict of column -> type, default ``None``
+  Data type for data or columns. E.g. ``{'a': np.float64, 'b': np.int32, 'c': 'Int64'}``
+  Use ``str`` or ``object`` together with suitable ``na_values`` settings to preserve
+  and not interpret dtype. If converters are specified, they will be applied INSTEAD
+  of dtype conversion.
+
+  .. versionadded:: 1.5.0
+
+     Support for defaultdict was added. Specify a defaultdict as input where
+     the default determines the dtype of the columns which are not explicitly
+     listed.
+
+dtype_backend : {"numpy_nullable", "pyarrow"}, defaults to NumPy backed DataFrames
+  Which dtype_backend to use, e.g. whether a DataFrame should have NumPy
+  arrays, nullable dtypes are used for all dtypes that have a nullable
+  implementation when "numpy_nullable" is set, pyarrow is used for all
+  dtypes if "pyarrow" is set.
+
+  The dtype_backends are still experimental.
+
+  .. versionadded:: 2.0
+
+engine : {``'c'``, ``'python'``, ``'pyarrow'``}
+  Parser engine to use. The C and pyarrow engines are faster, while the python engine
+  is currently more feature-complete. Multithreading is currently only supported by
+  the pyarrow engine.
+
+  .. versionadded:: 1.4.0
+
+     The "pyarrow" engine was added as an *experimental* engine, and some features
+     are unsupported, or may not work correctly, with this engine.
+converters : dict, default ``None``
+  Dict of functions for converting values in certain columns. Keys can either be
+  integers or column labels.
+true_values : list, default ``None``
+  Values to consider as ``True``.
+false_values : list, default ``None``
+  Values to consider as ``False``.
+skipinitialspace : boolean, default ``False``
+  Skip spaces after delimiter.
+skiprows : list-like or integer, default ``None``
+  Line numbers to skip (0-indexed) or number of lines to skip (int) at the start
+  of the file.
+
+  If callable, the callable function will be evaluated against the row
+  indices, returning True if the row should be skipped and False otherwise:
+
+  .. ipython:: python
+
+     data = "col1,col2,col3\na,b,1\na,b,2\nc,d,3"
+     pd.read_csv(StringIO(data))
+     pd.read_csv(StringIO(data), skiprows=lambda x: x % 2 != 0)
+
+skipfooter : int, default ``0``
+  Number of lines at bottom of file to skip (unsupported with engine='c').
+
+nrows : int, default ``None``
+  Number of rows of file to read. Useful for reading pieces of large files.
+low_memory : boolean, default ``True``
+  Internally process the file in chunks, resulting in lower memory use
+  while parsing, but possibly mixed type inference.  To ensure no mixed
+  types either set ``False``, or specify the type with the ``dtype`` parameter.
+  Note that the entire file is read into a single ``DataFrame`` regardless,
+  use the ``chunksize`` or ``iterator`` parameter to return the data in chunks.
+  (Only valid with C parser)
+memory_map : boolean, default False
+  If a filepath is provided for ``filepath_or_buffer``, map the file object
+  directly onto memory and access the data directly from there. Using this
+  option can improve performance because there is no longer any I/O overhead.
+
+NA and missing data handling
+++++++++++++++++++++++++++++
+
+na_values : scalar, str, list-like, or dict, default ``None``
+  Additional strings to recognize as NA/NaN. If dict passed, specific per-column
+  NA values. See :ref:`na values const <io.navaluesconst>` below
+  for a list of the values interpreted as NaN by default.
+
+keep_default_na : boolean, default ``True``
+  Whether or not to include the default NaN values when parsing the data.
+  Depending on whether ``na_values`` is passed in, the behavior is as follows:
+
+  * If ``keep_default_na`` is ``True``, and ``na_values`` are specified, ``na_values``
+    is appended to the default NaN values used for parsing.
+  * If ``keep_default_na`` is ``True``, and ``na_values`` are not specified, only
+    the default NaN values are used for parsing.
+  * If ``keep_default_na`` is ``False``, and ``na_values`` are specified, only
+    the NaN values specified ``na_values`` are used for parsing.
+  * If ``keep_default_na`` is ``False``, and ``na_values`` are not specified, no
+    strings will be parsed as NaN.
+
+  Note that if ``na_filter`` is passed in as ``False``, the ``keep_default_na`` and
+  ``na_values`` parameters will be ignored.
+na_filter : boolean, default ``True``
+  Detect missing value markers (empty strings and the value of na_values). In
+  data without any NAs, passing ``na_filter=False`` can improve the performance
+  of reading a large file.
+verbose : boolean, default ``False``
+  Indicate number of NA values placed in non-numeric columns.
+skip_blank_lines : boolean, default ``True``
+  If ``True``, skip over blank lines rather than interpreting as NaN values.
+
+.. _io.read_csv_table.datetime:
+
+Datetime handling
++++++++++++++++++
+
+parse_dates : boolean or list of ints or names or list of lists or dict, default ``False``.
+  * If ``True`` -> try parsing the index.
+  * If ``[1, 2, 3]`` ->  try parsing columns 1, 2, 3 each as a separate date
+    column.
+
+  .. note::
+     A fast-path exists for iso8601-formatted dates.
+date_format : str or dict of column -> format, default ``None``
+   If used in conjunction with ``parse_dates``, will parse dates according to this
+   format. For anything more complex,
+   please read in as ``object`` and then apply :func:`to_datetime` as-needed.
+
+   .. versionadded:: 2.0.0
+dayfirst : boolean, default ``False``
+  DD/MM format dates, international and European format.
+cache_dates : boolean, default True
+  If True, use a cache of unique, converted dates to apply the datetime
+  conversion. May produce significant speed-up when parsing duplicate
+  date strings, especially ones with timezone offsets.
+
+Iteration
++++++++++
+
+iterator : boolean, default ``False``
+  Return ``TextFileReader`` object for iteration or getting chunks with
+  ``get_chunk()``.
+chunksize : int, default ``None``
+  Return ``TextFileReader`` object for iteration. See :ref:`iterating and chunking
+  <io.chunking>` below.
+
+Quoting, compression, and file format
++++++++++++++++++++++++++++++++++++++
+
+compression : {``'infer'``, ``'gzip'``, ``'bz2'``, ``'zip'``, ``'xz'``, ``'zstd'``, ``None``, ``dict``}, default ``'infer'``
+  For on-the-fly decompression of on-disk data. If 'infer', then use gzip,
+  bz2, zip, xz, or zstandard if ``filepath_or_buffer`` is path-like ending in '.gz', '.bz2',
+  '.zip', '.xz', '.zst', respectively, and no decompression otherwise. If using 'zip',
+  the ZIP file must contain only one data file to be read in.
+  Set to ``None`` for no decompression. Can also be a dict with key ``'method'``
+  set to one of {``'zip'``, ``'gzip'``, ``'bz2'``, ``'zstd'``} and other key-value pairs are
+  forwarded to ``zipfile.ZipFile``, ``gzip.GzipFile``, ``bz2.BZ2File``, or ``zstandard.ZstdDecompressor``.
+  As an example, the following could be passed for faster compression and to
+  create a reproducible gzip archive:
+  ``compression={'method': 'gzip', 'compresslevel': 1, 'mtime': 1}``.
+
+  .. versionchanged:: 1.2.0 Previous versions forwarded dict entries for 'gzip' to ``gzip.open``.
+thousands : str, default ``None``
+  Thousands separator.
+decimal : str, default ``'.'``
+  Character to recognize as decimal point. E.g. use ``','`` for European data.
+float_precision : string, default None
+  Specifies which converter the C engine should use for floating-point values.
+  The options are ``None`` for the ordinary converter, ``high`` for the
+  high-precision converter, and ``round_trip`` for the round-trip converter.
+lineterminator : str (length 1), default ``None``
+  Character to break file into lines. Only valid with C parser.
+quotechar : str (length 1)
+  The character used to denote the start and end of a quoted item. Quoted items
+  can include the delimiter and it will be ignored.
+quoting : int or ``csv.QUOTE_*`` instance, default ``0``
+  Control field quoting behavior per ``csv.QUOTE_*`` constants. Use one of
+  ``QUOTE_MINIMAL`` (0), ``QUOTE_ALL`` (1), ``QUOTE_NONNUMERIC`` (2) or
+  ``QUOTE_NONE`` (3).
+doublequote : boolean, default ``True``
+   When ``quotechar`` is specified and ``quoting`` is not ``QUOTE_NONE``,
+   indicate whether or not to interpret two consecutive ``quotechar`` elements
+   **inside** a field as a single ``quotechar`` element.
+escapechar : str (length 1), default ``None``
+  One-character string used to escape delimiter when quoting is ``QUOTE_NONE``.
+comment : str, default ``None``
+  Indicates remainder of line should not be parsed. If found at the beginning of
+  a line, the line will be ignored altogether. This parameter must be a single
+  character. Like empty lines (as long as ``skip_blank_lines=True``), fully
+  commented lines are ignored by the parameter ``header`` but not by ``skiprows``.
+  For example, if ``comment='#'``, parsing '#empty\\na,b,c\\n1,2,3' with
+  ``header=0`` will result in 'a,b,c' being treated as the header.
+encoding : str, default ``None``
+  Encoding to use for UTF when reading/writing (e.g. ``'utf-8'``). `List of
+  Python standard encodings
+  <https://docs.python.org/3/library/codecs.html#standard-encodings>`_.
+dialect : str or :class:`python:csv.Dialect` instance, default ``None``
+  If provided, this parameter will override values (default or not) for the
+  following parameters: ``delimiter``, ``doublequote``, ``escapechar``,
+  ``skipinitialspace``, ``quotechar``, and ``quoting``. If it is necessary to
+  override values, a ParserWarning will be issued. See :class:`python:csv.Dialect`
+  documentation for more details.
+
+Error handling
+++++++++++++++
+
+on_bad_lines : {{'error', 'warn', 'skip'}}, default 'error'
+    Specifies what to do upon encountering a bad line (a line with too many fields).
+    Allowed values are :
+
+    - 'error', raise an ParserError when a bad line is encountered.
+    - 'warn', print a warning when a bad line is encountered and skip that line.
+    - 'skip', skip bad lines without raising or warning when they are encountered.
+
+    .. versionadded:: 1.3.0
+
+.. _io.dtypes:
+
+Specifying column data types
+''''''''''''''''''''''''''''
+
+You can indicate the data type for the whole ``DataFrame`` or individual
+columns:
+
+.. ipython:: python
+
+    import numpy as np
+
+    data = "a,b,c,d\n1,2,3,4\n5,6,7,8\n9,10,11"
+    print(data)
+
+    df = pd.read_csv(StringIO(data), dtype=object)
+    df
+    df["a"][0]
+    df = pd.read_csv(StringIO(data), dtype={"b": object, "c": np.float64, "d": "Int64"})
+    df.dtypes
+
+Fortunately, pandas offers more than one way to ensure that your column(s)
+contain only one ``dtype``. If you're unfamiliar with these concepts, you can
+see :ref:`here<basics.dtypes>` to learn more about dtypes, and
+:ref:`here<basics.object_conversion>` to learn more about ``object`` conversion in
+pandas.
+
+
+For instance, you can use the ``converters`` argument
+of :func:`~pandas.read_csv`:
+
+.. ipython:: python
+
+    data = "col_1\n1\n2\n'A'\n4.22"
+    df = pd.read_csv(StringIO(data), converters={"col_1": str})
+    df
+    df["col_1"].apply(type).value_counts()
+
+Or you can use the :func:`~pandas.to_numeric` function to coerce the
+dtypes after reading in the data,
+
+.. ipython:: python
+
+    df2 = pd.read_csv(StringIO(data))
+    df2["col_1"] = pd.to_numeric(df2["col_1"], errors="coerce")
+    df2
+    df2["col_1"].apply(type).value_counts()
+
+which will convert all valid parsing to floats, leaving the invalid parsing
+as ``NaN``.
+
+Ultimately, how you deal with reading in columns containing mixed dtypes
+depends on your specific needs. In the case above, if you wanted to ``NaN`` out
+the data anomalies, then :func:`~pandas.to_numeric` is probably your best option.
+However, if you wanted for all the data to be coerced, no matter the type, then
+using the ``converters`` argument of :func:`~pandas.read_csv` would certainly be
+worth trying.
+
+.. note::
+   In some cases, reading in abnormal data with columns containing mixed dtypes
+   will result in an inconsistent dataset. If you rely on pandas to infer the
+   dtypes of your columns, the parsing engine will go and infer the dtypes for
+   different chunks of the data, rather than the whole dataset at once. Consequently,
+   you can end up with column(s) with mixed dtypes. For example,
+
+   .. ipython:: python
+        :okwarning:
+
+        col_1 = list(range(500000)) + ["a", "b"] + list(range(500000))
+        df = pd.DataFrame({"col_1": col_1})
+        df.to_csv("foo.csv")
+        mixed_df = pd.read_csv("foo.csv")
+        mixed_df["col_1"].apply(type).value_counts()
+        mixed_df["col_1"].dtype
+
+   will result with ``mixed_df`` containing an ``int`` dtype for certain chunks
+   of the column, and ``str`` for others due to the mixed dtypes from the
+   data that was read in. It is important to note that the overall column will be
+   marked with a ``dtype`` of ``object``, which is used for columns with mixed dtypes.
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("foo.csv")
+
+Setting ``dtype_backend="numpy_nullable"`` will result in nullable dtypes for every column.
+
+.. ipython:: python
+
+   data = """a,b,c,d,e,f,g,h,i,j
+   1,2.5,True,a,,,,,12-31-2019,
+   3,4.5,False,b,6,7.5,True,a,12-31-2019,
+   """
+
+   df = pd.read_csv(StringIO(data), dtype_backend="numpy_nullable", parse_dates=["i"])
+   df
+   df.dtypes
+
+.. _io.categorical:
+
+Specifying categorical dtype
+''''''''''''''''''''''''''''
+
+``Categorical`` columns can be parsed directly by specifying ``dtype='category'`` or
+``dtype=CategoricalDtype(categories, ordered)``.
+
+.. ipython:: python
+
+   data = "col1,col2,col3\na,b,1\na,b,2\nc,d,3"
+
+   pd.read_csv(StringIO(data))
+   pd.read_csv(StringIO(data)).dtypes
+   pd.read_csv(StringIO(data), dtype="category").dtypes
+
+Individual columns can be parsed as a ``Categorical`` using a dict
+specification:
+
+.. ipython:: python
+
+   pd.read_csv(StringIO(data), dtype={"col1": "category"}).dtypes
+
+Specifying ``dtype='category'`` will result in an unordered ``Categorical``
+whose ``categories`` are the unique values observed in the data. For more
+control on the categories and order, create a
+:class:`~pandas.api.types.CategoricalDtype` ahead of time, and pass that for
+that column's ``dtype``.
+
+.. ipython:: python
+
+   from pandas.api.types import CategoricalDtype
+
+   dtype = CategoricalDtype(["d", "c", "b", "a"], ordered=True)
+   pd.read_csv(StringIO(data), dtype={"col1": dtype}).dtypes
+
+When using ``dtype=CategoricalDtype``, "unexpected" values outside of
+``dtype.categories`` are treated as missing values.
+
+.. ipython:: python
+
+   dtype = CategoricalDtype(["a", "b", "d"])  # No 'c'
+   pd.read_csv(StringIO(data), dtype={"col1": dtype}).col1
+
+This matches the behavior of :meth:`Categorical.set_categories`.
+
+.. note::
+
+   With ``dtype='category'``, the resulting categories will always be parsed
+   as strings (object dtype). If the categories are numeric they can be
+   converted using the :func:`to_numeric` function, or as appropriate, another
+   converter such as :func:`to_datetime`.
+
+   When ``dtype`` is a ``CategoricalDtype`` with homogeneous ``categories`` (
+   all numeric, all datetimes, etc.), the conversion is done automatically.
+
+   .. ipython:: python
+
+      df = pd.read_csv(StringIO(data), dtype="category")
+      df.dtypes
+      df["col3"]
+      new_categories = pd.to_numeric(df["col3"].cat.categories)
+      df["col3"] = df["col3"].cat.rename_categories(new_categories)
+      df["col3"]
+
+
+Naming and using columns
+''''''''''''''''''''''''
+
+.. _io.headers:
+
+Handling column names
++++++++++++++++++++++
+
+A file may or may not have a header row. pandas assumes the first row should be
+used as the column names:
+
+.. ipython:: python
+
+    data = "a,b,c\n1,2,3\n4,5,6\n7,8,9"
+    print(data)
+    pd.read_csv(StringIO(data))
+
+By specifying the ``names`` argument in conjunction with ``header`` you can
+indicate other names to use and whether or not to throw away the header row (if
+any):
+
+.. ipython:: python
+
+    print(data)
+    pd.read_csv(StringIO(data), names=["foo", "bar", "baz"], header=0)
+    pd.read_csv(StringIO(data), names=["foo", "bar", "baz"], header=None)
+
+If the header is in a row other than the first, pass the row number to
+``header``. This will skip the preceding rows:
+
+.. ipython:: python
+
+    data = "skip this skip it\na,b,c\n1,2,3\n4,5,6\n7,8,9"
+    pd.read_csv(StringIO(data), header=1)
+
+.. note::
+
+  Default behavior is to infer the column names: if no names are
+  passed the behavior is identical to ``header=0`` and column names
+  are inferred from the first non-blank line of the file, if column
+  names are passed explicitly then the behavior is identical to
+  ``header=None``.
+
+.. _io.dupe_names:
+
+Duplicate names parsing
+'''''''''''''''''''''''
+
+If the file or header contains duplicate names, pandas will by default
+distinguish between them so as to prevent overwriting data:
+
+.. ipython:: python
+
+   data = "a,b,a\n0,1,2\n3,4,5"
+   pd.read_csv(StringIO(data))
+
+There is no more duplicate data because duplicate columns 'X', ..., 'X' become
+'X', 'X.1', ..., 'X.N'.
+
+.. _io.usecols:
+
+Filtering columns (``usecols``)
++++++++++++++++++++++++++++++++
+
+The ``usecols`` argument allows you to select any subset of the columns in a
+file, either using the column names, position numbers or a callable:
+
+.. ipython:: python
+
+    data = "a,b,c,d\n1,2,3,foo\n4,5,6,bar\n7,8,9,baz"
+    pd.read_csv(StringIO(data))
+    pd.read_csv(StringIO(data), usecols=["b", "d"])
+    pd.read_csv(StringIO(data), usecols=[0, 2, 3])
+    pd.read_csv(StringIO(data), usecols=lambda x: x.upper() in ["A", "C"])
+
+The ``usecols`` argument can also be used to specify which columns not to
+use in the final result:
+
+.. ipython:: python
+
+   pd.read_csv(StringIO(data), usecols=lambda x: x not in ["a", "c"])
+
+In this case, the callable is specifying that we exclude the "a" and "c"
+columns from the output.
+
+Comments and empty lines
+''''''''''''''''''''''''
+
+.. _io.skiplines:
+
+Ignoring line comments and empty lines
+++++++++++++++++++++++++++++++++++++++
+
+If the ``comment`` parameter is specified, then completely commented lines will
+be ignored. By default, completely blank lines will be ignored as well.
+
+.. ipython:: python
+
+   data = "\na,b,c\n  \n# commented line\n1,2,3\n\n4,5,6"
+   print(data)
+   pd.read_csv(StringIO(data), comment="#")
+
+If ``skip_blank_lines=False``, then ``read_csv`` will not ignore blank lines:
+
+.. ipython:: python
+
+   data = "a,b,c\n\n1,2,3\n\n\n4,5,6"
+   pd.read_csv(StringIO(data), skip_blank_lines=False)
+
+.. warning::
+
+   The presence of ignored lines might create ambiguities involving line numbers;
+   the parameter ``header`` uses row numbers (ignoring commented/empty
+   lines), while ``skiprows`` uses line numbers (including commented/empty lines):
+
+   .. ipython:: python
+
+      data = "#comment\na,b,c\nA,B,C\n1,2,3"
+      pd.read_csv(StringIO(data), comment="#", header=1)
+      data = "A,B,C\n#comment\na,b,c\n1,2,3"
+      pd.read_csv(StringIO(data), comment="#", skiprows=2)
+
+   If both ``header`` and ``skiprows`` are specified, ``header`` will be
+   relative to the end of ``skiprows``. For example:
+
+.. ipython:: python
+
+   data = (
+       "# empty\n"
+       "# second empty line\n"
+       "# third emptyline\n"
+       "X,Y,Z\n"
+       "1,2,3\n"
+       "A,B,C\n"
+       "1,2.,4.\n"
+       "5.,NaN,10.0\n"
+   )
+   print(data)
+   pd.read_csv(StringIO(data), comment="#", skiprows=4, header=1)
+
+.. _io.comments:
+
+Comments
+++++++++
+
+Sometimes comments or meta data may be included in a file:
+
+.. ipython:: python
+
+   data = (
+       "ID,level,category\n"
+       "Patient1,123000,x # really unpleasant\n"
+       "Patient2,23000,y # wouldn't take his medicine\n"
+       "Patient3,1234018,z # awesome"
+   )
+   with open("tmp.csv", "w") as fh:
+       fh.write(data)
+
+   print(open("tmp.csv").read())
+
+By default, the parser includes the comments in the output:
+
+.. ipython:: python
+
+   df = pd.read_csv("tmp.csv")
+   df
+
+We can suppress the comments using the ``comment`` keyword:
+
+.. ipython:: python
+
+   df = pd.read_csv("tmp.csv", comment="#")
+   df
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp.csv")
+
+.. _io.unicode:
+
+Dealing with Unicode data
+'''''''''''''''''''''''''
+
+The ``encoding`` argument should be used for encoded unicode data, which will
+result in byte strings being decoded to unicode in the result:
+
+.. ipython:: python
+
+   from io import BytesIO
+
+   data = b"word,length\n" b"Tr\xc3\xa4umen,7\n" b"Gr\xc3\xbc\xc3\x9fe,5"
+   data = data.decode("utf8").encode("latin-1")
+   df = pd.read_csv(BytesIO(data), encoding="latin-1")
+   df
+   df["word"][1]
+
+Some formats which encode all characters as multiple bytes, like UTF-16, won't
+parse correctly at all without specifying the encoding. `Full list of Python
+standard encodings
+<https://docs.python.org/3/library/codecs.html#standard-encodings>`_.
+
+.. _io.index_col:
+
+Index columns and trailing delimiters
+'''''''''''''''''''''''''''''''''''''
+
+If a file has one more column of data than the number of column names, the
+first column will be used as the ``DataFrame``'s row names:
+
+.. ipython:: python
+
+    data = "a,b,c\n4,apple,bat,5.7\n8,orange,cow,10"
+    pd.read_csv(StringIO(data))
+
+.. ipython:: python
+
+    data = "index,a,b,c\n4,apple,bat,5.7\n8,orange,cow,10"
+    pd.read_csv(StringIO(data), index_col=0)
+
+Ordinarily, you can achieve this behavior using the ``index_col`` option.
+
+There are some exception cases when a file has been prepared with delimiters at
+the end of each data line, confusing the parser. To explicitly disable the
+index column inference and discard the last column, pass ``index_col=False``:
+
+.. ipython:: python
+
+    data = "a,b,c\n4,apple,bat,\n8,orange,cow,"
+    print(data)
+    pd.read_csv(StringIO(data))
+    pd.read_csv(StringIO(data), index_col=False)
+
+If a subset of data is being parsed using the ``usecols`` option, the
+``index_col`` specification is based on that subset, not the original data.
+
+.. ipython:: python
+
+    data = "a,b,c\n4,apple,bat,\n8,orange,cow,"
+    print(data)
+    pd.read_csv(StringIO(data), usecols=["b", "c"])
+    pd.read_csv(StringIO(data), usecols=["b", "c"], index_col=0)
+
+.. _io.parse_dates:
+
+Date Handling
+'''''''''''''
+
+Specifying date columns
++++++++++++++++++++++++
+
+To better facilitate working with datetime data, :func:`read_csv`
+uses the keyword arguments ``parse_dates`` and ``date_format``
+to allow users to specify a variety of columns and date/time formats to turn the
+input text data into ``datetime`` objects.
+
+The simplest case is to just pass in ``parse_dates=True``:
+
+.. ipython:: python
+
+   with open("foo.csv", mode="w") as f:
+       f.write("date,A,B,C\n20090101,a,1,2\n20090102,b,3,4\n20090103,c,4,5")
+
+   # Use a column as an index, and parse it as dates.
+   df = pd.read_csv("foo.csv", index_col=0, parse_dates=True)
+   df
+
+   # These are Python datetime objects
+   df.index
+
+It is often the case that we may want to store date and time data separately,
+or store various date fields separately. the ``parse_dates`` keyword can be
+used to specify columns to parse the dates and/or times.
+
+
+.. note::
+   If a column or index contains an unparsable date, the entire column or
+   index will be returned unaltered as an object data type. For non-standard
+   datetime parsing, use :func:`to_datetime` after ``pd.read_csv``.
+
+
+.. note::
+   read_csv has a fast_path for parsing datetime strings in iso8601 format,
+   e.g "2000-01-01T00:01:02+00:00" and similar variations. If you can arrange
+   for your data to store datetimes in this format, load times will be
+   significantly faster, ~20x has been observed.
+
+
+Date parsing functions
+++++++++++++++++++++++
+
+Finally, the parser allows you to specify a custom ``date_format``.
+Performance-wise, you should try these methods of parsing dates in order:
+
+1. If you know the format, use ``date_format``, e.g.:
+   ``date_format="%d/%m/%Y"`` or ``date_format={column_name: "%d/%m/%Y"}``.
+
+2. If you different formats for different columns, or want to pass any extra options (such
+   as ``utc``) to ``to_datetime``, then you should read in your data as ``object`` dtype, and
+   then use ``to_datetime``.
+
+
+.. _io.csv.mixed_timezones:
+
+Parsing a CSV with mixed timezones
+++++++++++++++++++++++++++++++++++
+
+pandas cannot natively represent a column or index with mixed timezones. If your CSV
+file contains columns with a mixture of timezones, the default result will be
+an object-dtype column with strings, even with ``parse_dates``.
+To parse the mixed-timezone values as a datetime column, read in as ``object`` dtype and
+then call :func:`to_datetime` with ``utc=True``.
+
+
+.. ipython:: python
+
+   content = """\
+   a
+   2000-01-01T00:00:00+05:00
+   2000-01-01T00:00:00+06:00"""
+   df = pd.read_csv(StringIO(content))
+   df["a"] = pd.to_datetime(df["a"], utc=True)
+   df["a"]
+
+
+.. _io.dayfirst:
+
+
+Inferring datetime format
++++++++++++++++++++++++++
+
+Here are some examples of datetime strings that can be guessed (all
+representing December 30th, 2011 at 00:00:00):
+
+* "20111230"
+* "2011/12/30"
+* "20111230 00:00:00"
+* "12/30/2011 00:00:00"
+* "30/Dec/2011 00:00:00"
+* "30/December/2011 00:00:00"
+
+Note that format inference is sensitive to ``dayfirst``.  With
+``dayfirst=True``, it will guess "01/12/2011" to be December 1st. With
+``dayfirst=False`` (default) it will guess "01/12/2011" to be January 12th.
+
+If you try to parse a column of date strings, pandas will attempt to guess the format
+from the first non-NaN element, and will then parse the rest of the column with that
+format. If pandas fails to guess the format (for example if your first string is
+``'01 December US/Pacific 2000'``), then a warning will be raised and each
+row will be parsed individually by ``dateutil.parser.parse``. The safest
+way to parse dates is to explicitly set ``format=``.
+
+.. ipython:: python
+
+   df = pd.read_csv(
+       "foo.csv",
+       index_col=0,
+       parse_dates=True,
+   )
+   df
+
+In the case that you have mixed datetime formats within the same column, you can
+pass  ``format='mixed'``
+
+.. ipython:: python
+
+   data = StringIO("date\n12 Jan 2000\n2000-01-13\n")
+   df = pd.read_csv(data)
+   df['date'] = pd.to_datetime(df['date'], format='mixed')
+   df
+
+or, if your datetime formats are all ISO8601 (possibly not identically-formatted):
+
+.. ipython:: python
+
+   data = StringIO("date\n2020-01-01\n2020-01-01 03:00\n")
+   df = pd.read_csv(data)
+   df['date'] = pd.to_datetime(df['date'], format='ISO8601')
+   df
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("foo.csv")
+
+International date formats
+++++++++++++++++++++++++++
+
+While US date formats tend to be MM/DD/YYYY, many international formats use
+DD/MM/YYYY instead. For convenience, a ``dayfirst`` keyword is provided:
+
+.. ipython:: python
+
+   data = "date,value,cat\n1/6/2000,5,a\n2/6/2000,10,b\n3/6/2000,15,c"
+   print(data)
+   with open("tmp.csv", "w") as fh:
+       fh.write(data)
+
+   pd.read_csv("tmp.csv", parse_dates=[0])
+   pd.read_csv("tmp.csv", dayfirst=True, parse_dates=[0])
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp.csv")
+
+Writing CSVs to binary file objects
++++++++++++++++++++++++++++++++++++
+
+.. versionadded:: 1.2.0
+
+``df.to_csv(..., mode="wb")`` allows writing a CSV to a file object
+opened binary mode. In most cases, it is not necessary to specify
+``mode`` as pandas will auto-detect whether the file object is
+opened in text or binary mode.
+
+.. ipython:: python
+
+   import io
+
+   data = pd.DataFrame([0, 1, 2])
+   buffer = io.BytesIO()
+   data.to_csv(buffer, encoding="utf-8", compression="gzip")
+
+.. _io.float_precision:
+
+Specifying method for floating-point conversion
+'''''''''''''''''''''''''''''''''''''''''''''''
+
+The parameter ``float_precision`` can be specified in order to use
+a specific floating-point converter during parsing with the C engine.
+The options are the ordinary converter, the high-precision converter, and
+the round-trip converter (which is guaranteed to round-trip values after
+writing to a file). For example:
+
+.. ipython:: python
+
+   val = "0.3066101993807095471566981359501369297504425048828125"
+   data = "a,b,c\n1,2,{0}".format(val)
+   abs(
+       pd.read_csv(
+           StringIO(data),
+           engine="c",
+           float_precision=None,
+       )["c"][0] - float(val)
+   )
+   abs(
+       pd.read_csv(
+           StringIO(data),
+           engine="c",
+           float_precision="high",
+       )["c"][0] - float(val)
+   )
+   abs(
+       pd.read_csv(StringIO(data), engine="c", float_precision="round_trip")["c"][0]
+       - float(val)
+   )
+
+
+.. _io.thousands:
+
+Thousand separators
+'''''''''''''''''''
+
+For large numbers that have been written with a thousands separator, you can
+set the ``thousands`` keyword to a string of length 1 so that integers will be parsed
+correctly.
+
+By default, numbers with a thousands separator will be parsed as strings:
+
+.. ipython:: python
+
+   data = (
+       "ID|level|category\n"
+       "Patient1|123,000|x\n"
+       "Patient2|23,000|y\n"
+       "Patient3|1,234,018|z"
+   )
+
+   with open("tmp.csv", "w") as fh:
+       fh.write(data)
+
+   df = pd.read_csv("tmp.csv", sep="|")
+   df
+
+   df.level.dtype
+
+The ``thousands`` keyword allows integers to be parsed correctly:
+
+.. ipython:: python
+
+    df = pd.read_csv("tmp.csv", sep="|", thousands=",")
+    df
+
+    df.level.dtype
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp.csv")
+
+.. _io.na_values:
+
+NA values
+'''''''''
+
+To control which values are parsed as missing values (which are signified by
+``NaN``), specify a string in ``na_values``. If you specify a list of strings,
+then all values in it are considered to be missing values. If you specify a
+number (a ``float``, like ``5.0`` or an ``integer`` like ``5``), the
+corresponding equivalent values will also imply a missing value (in this case
+effectively ``[5.0, 5]`` are recognized as ``NaN``).
+
+To completely override the default values that are recognized as missing, specify ``keep_default_na=False``.
+
+.. _io.navaluesconst:
+
+The default ``NaN`` recognized values are ``['-1.#IND', '1.#QNAN', '1.#IND', '-1.#QNAN', '#N/A N/A', '#N/A', 'N/A',
+'n/a', 'NA', '<NA>', '#NA', 'NULL', 'null', 'NaN', '-NaN', 'nan', '-nan', 'None', '']``.
+
+Let us consider some examples:
+
+.. code-block:: python
+
+   pd.read_csv("path_to_file.csv", na_values=[5])
+
+In the example above ``5`` and ``5.0`` will be recognized as ``NaN``, in
+addition to the defaults. A string will first be interpreted as a numerical
+``5``, then as a ``NaN``.
+
+.. code-block:: python
+
+   pd.read_csv("path_to_file.csv", keep_default_na=False, na_values=[""])
+
+Above, only an empty field will be recognized as ``NaN``.
+
+.. code-block:: python
+
+   pd.read_csv("path_to_file.csv", keep_default_na=False, na_values=["NA", "0"])
+
+Above, both ``NA`` and ``0`` as strings are ``NaN``.
+
+.. code-block:: python
+
+   pd.read_csv("path_to_file.csv", na_values=["Nope"])
+
+The default values, in addition to the string ``"Nope"`` are recognized as
+``NaN``.
+
+.. _io.infinity:
+
+Infinity
+''''''''
+
+``inf`` like values will be parsed as ``np.inf`` (positive infinity), and ``-inf`` as ``-np.inf`` (negative infinity).
+These will ignore the case of the value, meaning ``Inf``, will also be parsed as ``np.inf``.
+
+.. _io.boolean:
+
+Boolean values
+''''''''''''''
+
+The common values ``True``, ``False``, ``TRUE``, and ``FALSE`` are all
+recognized as boolean. Occasionally you might want to recognize other values
+as being boolean. To do this, use the ``true_values`` and ``false_values``
+options as follows:
+
+.. ipython:: python
+
+    data = "a,b,c\n1,Yes,2\n3,No,4"
+    print(data)
+    pd.read_csv(StringIO(data))
+    pd.read_csv(StringIO(data), true_values=["Yes"], false_values=["No"])
+
+.. _io.bad_lines:
+
+Handling "bad" lines
+''''''''''''''''''''
+
+Some files may have malformed lines with too few fields or too many. Lines with
+too few fields will have NA values filled in the trailing fields. Lines with
+too many fields will raise an error by default:
+
+.. ipython:: python
+    :okexcept:
+
+    data = "a,b,c\n1,2,3\n4,5,6,7\n8,9,10"
+    pd.read_csv(StringIO(data))
+
+You can elect to skip bad lines:
+
+.. ipython:: python
+
+    data = "a,b,c\n1,2,3\n4,5,6,7\n8,9,10"
+    pd.read_csv(StringIO(data), on_bad_lines="skip")
+
+.. versionadded:: 1.4.0
+
+Or pass a callable function to handle the bad line if ``engine="python"``.
+The bad line will be a list of strings that was split by the ``sep``:
+
+.. ipython:: python
+
+    external_list = []
+    def bad_lines_func(line):
+        external_list.append(line)
+        return line[-3:]
+    pd.read_csv(StringIO(data), on_bad_lines=bad_lines_func, engine="python")
+    external_list
+
+.. note::
+
+   The callable function will handle only a line with too many fields.
+   Bad lines caused by other errors will be silently skipped.
+
+   .. ipython:: python
+
+      bad_lines_func = lambda line: print(line)
+
+      data = 'name,type\nname a,a is of type a\nname b,"b\" is of type b"'
+      data
+      pd.read_csv(StringIO(data), on_bad_lines=bad_lines_func, engine="python")
+
+   The line was not processed in this case, as a "bad line" here is caused by an escape character.
+
+You can also use the ``usecols`` parameter to eliminate extraneous column
+data that appear in some lines but not others:
+
+.. ipython:: python
+   :okexcept:
+
+   pd.read_csv(StringIO(data), usecols=[0, 1, 2])
+
+In case you want to keep all data including the lines with too many fields, you can
+specify a sufficient number of ``names``. This ensures that lines with not enough
+fields are filled with ``NaN``.
+
+.. ipython:: python
+
+   pd.read_csv(StringIO(data), names=['a', 'b', 'c', 'd'])
+
+.. _io.dialect:
+
+Dialect
+'''''''
+
+The ``dialect`` keyword gives greater flexibility in specifying the file format.
+By default it uses the Excel dialect but you can specify either the dialect name
+or a :class:`python:csv.Dialect` instance.
+
+Suppose you had data with unenclosed quotes:
+
+.. ipython:: python
+
+   data = "label1,label2,label3\n" 'index1,"a,c,e\n' "index2,b,d,f"
+   print(data)
+
+By default, ``read_csv`` uses the Excel dialect and treats the double quote as
+the quote character, which causes it to fail when it finds a newline before it
+finds the closing double quote.
+
+We can get around this using ``dialect``:
+
+.. ipython:: python
+   :okwarning:
+
+   import csv
+
+   dia = csv.excel()
+   dia.quoting = csv.QUOTE_NONE
+   pd.read_csv(StringIO(data), dialect=dia)
+
+All of the dialect options can be specified separately by keyword arguments:
+
+.. ipython:: python
+
+    data = "a,b,c~1,2,3~4,5,6"
+    pd.read_csv(StringIO(data), lineterminator="~")
+
+Another common dialect option is ``skipinitialspace``, to skip any whitespace
+after a delimiter:
+
+.. ipython:: python
+
+   data = "a, b, c\n1, 2, 3\n4, 5, 6"
+   print(data)
+   pd.read_csv(StringIO(data), skipinitialspace=True)
+
+The parsers make every attempt to "do the right thing" and not be fragile. Type
+inference is a pretty big deal. If a column can be coerced to integer dtype
+without altering the contents, the parser will do so. Any non-numeric
+columns will come through as object dtype as with the rest of pandas objects.
+
+.. _io.quoting:
+
+Quoting and Escape Characters
+'''''''''''''''''''''''''''''
+
+Quotes (and other escape characters) in embedded fields can be handled in any
+number of ways. One way is to use backslashes; to properly parse this data, you
+should pass the ``escapechar`` option:
+
+.. ipython:: python
+
+   data = 'a,b\n"hello, \\"Bob\\", nice to see you",5'
+   print(data)
+   pd.read_csv(StringIO(data), escapechar="\\")
+
+.. _io.fwf_reader:
+.. _io.fwf:
+
+Files with fixed width columns
+''''''''''''''''''''''''''''''
+
+While :func:`read_csv` reads delimited data, the :func:`read_fwf` function works
+with data files that have known and fixed column widths. The function parameters
+to ``read_fwf`` are largely the same as ``read_csv`` with two extra parameters, and
+a different usage of the ``delimiter`` parameter:
+
+* ``colspecs``: A list of pairs (tuples) giving the extents of the
+  fixed-width fields of each line as half-open intervals (i.e.,  [from, to[ ).
+  String value 'infer' can be used to instruct the parser to try detecting
+  the column specifications from the first 100 rows of the data. Default
+  behavior, if not specified, is to infer.
+* ``widths``: A list of field widths which can be used instead of 'colspecs'
+  if the intervals are contiguous.
+* ``delimiter``: Characters to consider as filler characters in the fixed-width file.
+  Can be used to specify the filler character of the fields
+  if it is not spaces (e.g., '~').
+
+Consider a typical fixed-width data file:
+
+.. ipython:: python
+
+   data1 = (
+       "id8141    360.242940   149.910199   11950.7\n"
+       "id1594    444.953632   166.985655   11788.4\n"
+       "id1849    364.136849   183.628767   11806.2\n"
+       "id1230    413.836124   184.375703   11916.8\n"
+       "id1948    502.953953   173.237159   12468.3"
+   )
+   with open("bar.csv", "w") as f:
+       f.write(data1)
+
+In order to parse this file into a ``DataFrame``, we simply need to supply the
+column specifications to the ``read_fwf`` function along with the file name:
+
+.. ipython:: python
+
+   # Column specifications are a list of half-intervals
+   colspecs = [(0, 6), (8, 20), (21, 33), (34, 43)]
+   df = pd.read_fwf("bar.csv", colspecs=colspecs, header=None, index_col=0)
+   df
+
+Note how the parser automatically picks column names X.<column number> when
+``header=None`` argument is specified. Alternatively, you can supply just the
+column widths for contiguous columns:
+
+.. ipython:: python
+
+   # Widths are a list of integers
+   widths = [6, 14, 13, 10]
+   df = pd.read_fwf("bar.csv", widths=widths, header=None)
+   df
+
+The parser will take care of extra white spaces around the columns
+so it's ok to have extra separation between the columns in the file.
+
+By default, ``read_fwf`` will try to infer the file's ``colspecs`` by using the
+first 100 rows of the file. It can do it only in cases when the columns are
+aligned and correctly separated by the provided ``delimiter`` (default delimiter
+is whitespace).
+
+.. ipython:: python
+
+   df = pd.read_fwf("bar.csv", header=None, index_col=0)
+   df
+
+``read_fwf`` supports the ``dtype`` parameter for specifying the types of
+parsed columns to be different from the inferred type.
+
+.. ipython:: python
+
+   pd.read_fwf("bar.csv", header=None, index_col=0).dtypes
+   pd.read_fwf("bar.csv", header=None, dtype={2: "object"}).dtypes
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("bar.csv")
+
+
+Indexes
+'''''''
+
+Files with an "implicit" index column
++++++++++++++++++++++++++++++++++++++
+
+Consider a file with one less entry in the header than the number of data
+column:
+
+.. ipython:: python
+
+   data = "A,B,C\n20090101,a,1,2\n20090102,b,3,4\n20090103,c,4,5"
+   print(data)
+   with open("foo.csv", "w") as f:
+       f.write(data)
+
+In this special case, ``read_csv`` assumes that the first column is to be used
+as the index of the ``DataFrame``:
+
+.. ipython:: python
+
+   pd.read_csv("foo.csv")
+
+Note that the dates weren't automatically parsed. In that case you would need
+to do as before:
+
+.. ipython:: python
+
+   df = pd.read_csv("foo.csv", parse_dates=True)
+   df.index
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("foo.csv")
+
+
+Reading an index with a ``MultiIndex``
+++++++++++++++++++++++++++++++++++++++
+
+.. _io.csv_multiindex:
+
+Suppose you have data indexed by two columns:
+
+.. ipython:: python
+
+   data = 'year,indiv,zit,xit\n1977,"A",1.2,.6\n1977,"B",1.5,.5'
+   print(data)
+   with open("mindex_ex.csv", mode="w") as f:
+       f.write(data)
+
+The ``index_col`` argument to ``read_csv`` can take a list of
+column numbers to turn multiple columns into a ``MultiIndex`` for the index of the
+returned object:
+
+.. ipython:: python
+
+   df = pd.read_csv("mindex_ex.csv", index_col=[0, 1])
+   df
+   df.loc[1977]
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("mindex_ex.csv")
+
+.. _io.multi_index_columns:
+
+Reading columns with a ``MultiIndex``
++++++++++++++++++++++++++++++++++++++
+
+By specifying list of row locations for the ``header`` argument, you
+can read in a ``MultiIndex`` for the columns. Specifying non-consecutive
+rows will skip the intervening rows.
+
+.. ipython:: python
+
+   mi_idx = pd.MultiIndex.from_arrays([[1, 2, 3, 4], list("abcd")], names=list("ab"))
+   mi_col = pd.MultiIndex.from_arrays([[1, 2], list("ab")], names=list("cd"))
+   df = pd.DataFrame(np.ones((4, 2)), index=mi_idx, columns=mi_col)
+   df.to_csv("mi.csv")
+   print(open("mi.csv").read())
+   pd.read_csv("mi.csv", header=[0, 1, 2, 3], index_col=[0, 1])
+
+``read_csv`` is also able to interpret a more common format
+of multi-columns indices.
+
+.. ipython:: python
+
+   data = ",a,a,a,b,c,c\n,q,r,s,t,u,v\none,1,2,3,4,5,6\ntwo,7,8,9,10,11,12"
+   print(data)
+   with open("mi2.csv", "w") as fh:
+       fh.write(data)
+
+   pd.read_csv("mi2.csv", header=[0, 1], index_col=0)
+
+.. note::
+   If an ``index_col`` is not specified (e.g. you don't have an index, or wrote it
+   with ``df.to_csv(..., index=False)``, then any ``names`` on the columns index will
+   be *lost*.
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("mi.csv")
+   os.remove("mi2.csv")
+
+.. _io.sniff:
+
+Automatically "sniffing" the delimiter
+''''''''''''''''''''''''''''''''''''''
+
+``read_csv`` is capable of inferring delimited (not necessarily
+comma-separated) files, as pandas uses the :class:`python:csv.Sniffer`
+class of the csv module. For this, you have to specify ``sep=None``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 4))
+   df.to_csv("tmp2.csv", sep=":", index=False)
+   pd.read_csv("tmp2.csv", sep=None, engine="python")
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp2.csv")
+
+.. _io.multiple_files:
+
+Reading multiple files to create a single DataFrame
+'''''''''''''''''''''''''''''''''''''''''''''''''''
+
+It's best to use :func:`~pandas.concat` to combine multiple files.
+See the :ref:`cookbook<cookbook.csv.multiple_files>` for an example.
+
+.. _io.chunking:
+
+Iterating through files chunk by chunk
+''''''''''''''''''''''''''''''''''''''
+
+Suppose you wish to iterate through a (potentially very large) file lazily
+rather than reading the entire file into memory, such as the following:
+
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 4))
+   df.to_csv("tmp.csv", index=False)
+   table = pd.read_csv("tmp.csv")
+   table
+
+
+By specifying a ``chunksize`` to ``read_csv``, the return
+value will be an iterable object of type ``TextFileReader``:
+
+.. ipython:: python
+
+   with pd.read_csv("tmp.csv", chunksize=4) as reader:
+       print(reader)
+       for chunk in reader:
+           print(chunk)
+
+.. versionchanged:: 1.2
+
+  ``read_csv/json/sas`` return a context-manager when iterating through a file.
+
+Specifying ``iterator=True`` will also return the ``TextFileReader`` object:
+
+.. ipython:: python
+
+   with pd.read_csv("tmp.csv", iterator=True) as reader:
+       print(reader.get_chunk(5))
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp.csv")
+
+Specifying the parser engine
+''''''''''''''''''''''''''''
+
+pandas currently supports three engines, the C engine, the python engine, and an experimental
+pyarrow engine (requires the ``pyarrow`` package). In general, the pyarrow engine is fastest
+on larger workloads and is equivalent in speed to the C engine on most other workloads.
+The python engine tends to be slower than the pyarrow and C engines on most workloads. However,
+the pyarrow engine is much less robust than the C engine, which lacks a few features compared to the
+Python engine.
+
+Where possible, pandas uses the C parser (specified as ``engine='c'``), but it may fall
+back to Python if C-unsupported options are specified.
+
+Currently, options unsupported by the C and pyarrow engines include:
+
+* ``sep`` other than a single character (e.g. regex separators)
+* ``skipfooter``
+
+Specifying any of the above options will produce a ``ParserWarning`` unless the
+python engine is selected explicitly using ``engine='python'``.
+
+Options that are unsupported by the pyarrow engine which are not covered by the list above include:
+
+* ``float_precision``
+* ``chunksize``
+* ``comment``
+* ``nrows``
+* ``thousands``
+* ``memory_map``
+* ``dialect``
+* ``on_bad_lines``
+* ``quoting``
+* ``lineterminator``
+* ``converters``
+* ``decimal``
+* ``iterator``
+* ``dayfirst``
+* ``verbose``
+* ``skipinitialspace``
+* ``low_memory``
+
+Specifying these options with ``engine='pyarrow'`` will raise a ``ValueError``.
+
+.. _io.remote:
+
+Reading/writing remote files
+''''''''''''''''''''''''''''
+
+You can pass in a URL to read or write remote files to many of pandas' IO
+functions - the following example shows reading a CSV file:
+
+.. code-block:: python
+
+   df = pd.read_csv("https://download.bls.gov/pub/time.series/cu/cu.item", sep="\t")
+
+.. versionadded:: 1.3.0
+
+A custom header can be sent alongside HTTP(s) requests by passing a dictionary
+of header key value mappings to the ``storage_options`` keyword argument as shown below:
+
+.. code-block:: python
+
+   headers = {"User-Agent": "pandas"}
+   df = pd.read_csv(
+       "https://download.bls.gov/pub/time.series/cu/cu.item",
+       sep="\t",
+       storage_options=headers
+   )
+
+All URLs which are not local files or HTTP(s) are handled by
+`fsspec`_, if installed, and its various filesystem implementations
+(including Amazon S3, Google Cloud, SSH, FTP, webHDFS...).
+Some of these implementations will require additional packages to be
+installed, for example
+S3 URLs require the `s3fs
+<https://pypi.org/project/s3fs/>`_ library:
+
+.. code-block:: python
+
+   df = pd.read_json("s3://pandas-test/adatafile.json")
+
+When dealing with remote storage systems, you might need
+extra configuration with environment variables or config files in
+special locations. For example, to access data in your S3 bucket,
+you will need to define credentials in one of the several ways listed in
+the `S3Fs documentation
+<https://s3fs.readthedocs.io/en/latest/#credentials>`_. The same is true
+for several of the storage backends, and you should follow the links
+at `fsimpl1`_ for implementations built into ``fsspec`` and `fsimpl2`_
+for those not included in the main ``fsspec``
+distribution.
+
+You can also pass parameters directly to the backend driver. Since ``fsspec`` does not
+utilize the ``AWS_S3_HOST`` environment variable, we can directly define a
+dictionary containing the endpoint_url and pass the object into the storage
+option parameter:
+
+.. code-block:: python
+
+   storage_options = {"client_kwargs": {"endpoint_url": "http://127.0.0.1:5555"}}
+   df = pd.read_json("s3://pandas-test/test-1", storage_options=storage_options)
+
+More sample configurations and documentation can be found at `S3Fs documentation
+<https://s3fs.readthedocs.io/en/latest/index.html?highlight=host#s3-compatible-storage>`__.
+
+If you do *not* have S3 credentials, you can still access public
+data by specifying an anonymous connection, such as
+
+.. versionadded:: 1.2.0
+
+.. code-block:: python
+
+   pd.read_csv(
+       "s3://ncei-wcsd-archive/data/processed/SH1305/18kHz/SaKe2013"
+       "-D20130523-T080854_to_SaKe2013-D20130523-T085643.csv",
+       storage_options={"anon": True},
+   )
+
+``fsspec`` also allows complex URLs, for accessing data in compressed
+archives, local caching of files, and more. To locally cache the above
+example, you would modify the call to
+
+.. code-block:: python
+
+   pd.read_csv(
+       "simplecache::s3://ncei-wcsd-archive/data/processed/SH1305/18kHz/"
+       "SaKe2013-D20130523-T080854_to_SaKe2013-D20130523-T085643.csv",
+       storage_options={"s3": {"anon": True}},
+   )
+
+where we specify that the "anon" parameter is meant for the "s3" part of
+the implementation, not to the caching implementation. Note that this caches to a temporary
+directory for the duration of the session only, but you can also specify
+a permanent store.
+
+.. _fsspec: https://filesystem-spec.readthedocs.io/en/latest/
+.. _fsimpl1: https://filesystem-spec.readthedocs.io/en/latest/api.html#built-in-implementations
+.. _fsimpl2: https://filesystem-spec.readthedocs.io/en/latest/api.html#other-known-implementations
+
+Writing out data
+''''''''''''''''
+
+.. _io.store_in_csv:
+
+Writing to CSV format
++++++++++++++++++++++
+
+The ``Series`` and ``DataFrame`` objects have an instance method ``to_csv`` which
+allows storing the contents of the object as a comma-separated-values file. The
+function takes a number of arguments. Only the first is required.
+
+* ``path_or_buf``: A string path to the file to write or a file object.  If a file object it must be opened with ``newline=''``
+* ``sep`` : Field delimiter for the output file (default ",")
+* ``na_rep``: A string representation of a missing value (default '')
+* ``float_format``: Format string for floating point numbers
+* ``columns``: Columns to write (default None)
+* ``header``: Whether to write out the column names (default True)
+* ``index``: whether to write row (index) names (default True)
+* ``index_label``: Column label(s) for index column(s) if desired. If None
+  (default), and ``header`` and ``index`` are True, then the index names are
+  used. (A sequence should be given if the ``DataFrame`` uses MultiIndex).
+* ``mode`` : Python write mode, default 'w'
+* ``encoding``: a string representing the encoding to use if the contents are
+  non-ASCII, for Python versions prior to 3
+* ``lineterminator``: Character sequence denoting line end (default ``os.linesep``)
+* ``quoting``: Set quoting rules as in csv module (default csv.QUOTE_MINIMAL). Note that if you have set a ``float_format`` then floats are converted to strings and csv.QUOTE_NONNUMERIC will treat them as non-numeric
+* ``quotechar``: Character used to quote fields (default '"')
+* ``doublequote``: Control quoting of ``quotechar`` in fields (default True)
+* ``escapechar``: Character used to escape ``sep`` and ``quotechar`` when
+  appropriate (default None)
+* ``chunksize``: Number of rows to write at a time
+* ``date_format``: Format string for datetime objects
+
+Writing a formatted string
+++++++++++++++++++++++++++
+
+.. _io.formatting:
+
+The ``DataFrame`` object has an instance method ``to_string`` which allows control
+over the string representation of the object. All arguments are optional:
+
+* ``buf`` default None, for example a StringIO object
+* ``columns`` default None, which columns to write
+* ``col_space`` default None, minimum width of each column.
+* ``na_rep`` default ``NaN``, representation of NA value
+* ``formatters`` default None, a dictionary (by column) of functions each of
+  which takes a single argument and returns a formatted string
+* ``float_format`` default None, a function which takes a single (float)
+  argument and returns a formatted string; to be applied to floats in the
+  ``DataFrame``.
+* ``sparsify`` default True, set to False for a ``DataFrame`` with a hierarchical
+  index to print every MultiIndex key at each row.
+* ``index_names`` default True, will print the names of the indices
+* ``index`` default True, will print the index (ie, row labels)
+* ``header`` default True, will print the column labels
+* ``justify`` default ``left``, will print column headers left- or
+  right-justified
+
+The ``Series`` object also has a ``to_string`` method, but with only the ``buf``,
+``na_rep``, ``float_format`` arguments. There is also a ``length`` argument
+which, if set to ``True``, will additionally output the length of the Series.
+
+.. _io.json:
+
+JSON
+----
+
+Read and write ``JSON`` format files and strings.
+
+.. _io.json_writer:
+
+Writing JSON
+''''''''''''
+
+A ``Series`` or ``DataFrame`` can be converted to a valid JSON string. Use ``to_json``
+with optional parameters:
+
+* ``path_or_buf`` : the pathname or buffer to write the output.
+  This can be ``None`` in which case a JSON string is returned.
+* ``orient`` :
+
+  ``Series``:
+      * default is ``index``
+      * allowed values are {``split``, ``records``, ``index``}
+
+  ``DataFrame``:
+      * default is ``columns``
+      * allowed values are {``split``, ``records``, ``index``, ``columns``, ``values``, ``table``}
+
+  The format of the JSON string
+
+  .. csv-table::
+     :widths: 20, 150
+
+     ``split``, dict like {index -> [index]; columns -> [columns]; data -> [values]}
+     ``records``, list like [{column -> value}; ... ]
+     ``index``, dict like {index -> {column -> value}}
+     ``columns``, dict like {column -> {index -> value}}
+     ``values``, just the values array
+     ``table``, adhering to the JSON `Table Schema`_
+
+* ``date_format`` : string, type of date conversion, 'epoch' for timestamp, 'iso' for ISO8601.
+* ``double_precision`` : The number of decimal places to use when encoding floating point values, default 10.
+* ``force_ascii`` : force encoded string to be ASCII, default True.
+* ``date_unit`` : The time unit to encode to, governs timestamp and ISO8601 precision. One of 's', 'ms', 'us' or 'ns' for seconds, milliseconds, microseconds and nanoseconds respectively. Default 'ms'.
+* ``default_handler`` : The handler to call if an object cannot otherwise be converted to a suitable format for JSON. Takes a single argument, which is the object to convert, and returns a serializable object.
+* ``lines`` : If ``records`` orient, then will write each record per line as json.
+* ``mode`` : string, writer mode when writing to path. 'w' for write, 'a' for append. Default 'w'
+
+Note ``NaN``'s, ``NaT``'s and ``None`` will be converted to ``null`` and ``datetime`` objects will be converted based on the ``date_format`` and ``date_unit`` parameters.
+
+.. ipython:: python
+
+   dfj = pd.DataFrame(np.random.randn(5, 2), columns=list("AB"))
+   json = dfj.to_json()
+   json
+
+Orient options
+++++++++++++++
+
+There are a number of different options for the format of the resulting JSON
+file / string. Consider the following ``DataFrame`` and ``Series``:
+
+.. ipython:: python
+
+  dfjo = pd.DataFrame(
+      dict(A=range(1, 4), B=range(4, 7), C=range(7, 10)),
+      columns=list("ABC"),
+      index=list("xyz"),
+  )
+  dfjo
+  sjo = pd.Series(dict(x=15, y=16, z=17), name="D")
+  sjo
+
+**Column oriented** (the default for ``DataFrame``) serializes the data as
+nested JSON objects with column labels acting as the primary index:
+
+.. ipython:: python
+
+  dfjo.to_json(orient="columns")
+  # Not available for Series
+
+**Index oriented** (the default for ``Series``) similar to column oriented
+but the index labels are now primary:
+
+.. ipython:: python
+
+  dfjo.to_json(orient="index")
+  sjo.to_json(orient="index")
+
+**Record oriented** serializes the data to a JSON array of column -> value records,
+index labels are not included. This is useful for passing ``DataFrame`` data to plotting
+libraries, for example the JavaScript library ``d3.js``:
+
+.. ipython:: python
+
+  dfjo.to_json(orient="records")
+  sjo.to_json(orient="records")
+
+**Value oriented** is a bare-bones option which serializes to nested JSON arrays of
+values only, column and index labels are not included:
+
+.. ipython:: python
+
+  dfjo.to_json(orient="values")
+  # Not available for Series
+
+**Split oriented** serializes to a JSON object containing separate entries for
+values, index and columns. Name is also included for ``Series``:
+
+.. ipython:: python
+
+  dfjo.to_json(orient="split")
+  sjo.to_json(orient="split")
+
+**Table oriented** serializes to the JSON `Table Schema`_, allowing for the
+preservation of metadata including but not limited to dtypes and index names.
+
+.. note::
+
+  Any orient option that encodes to a JSON object will not preserve the ordering of
+  index and column labels during round-trip serialization. If you wish to preserve
+  label ordering use the ``split`` option as it uses ordered containers.
+
+Date handling
++++++++++++++
+
+Writing in ISO date format:
+
+.. ipython:: python
+
+   dfd = pd.DataFrame(np.random.randn(5, 2), columns=list("AB"))
+   dfd["date"] = pd.Timestamp("20130101")
+   dfd = dfd.sort_index(axis=1, ascending=False)
+   json = dfd.to_json(date_format="iso")
+   json
+
+Writing in ISO date format, with microseconds:
+
+.. ipython:: python
+
+   json = dfd.to_json(date_format="iso", date_unit="us")
+   json
+
+Writing to a file, with a date index and a date column:
+
+.. ipython:: python
+
+   dfj2 = dfj.copy()
+   dfj2["date"] = pd.Timestamp("20130101")
+   dfj2["ints"] = list(range(5))
+   dfj2["bools"] = True
+   dfj2.index = pd.date_range("20130101", periods=5)
+   dfj2.to_json("test.json", date_format="iso")
+
+   with open("test.json") as fh:
+       print(fh.read())
+
+Fallback behavior
++++++++++++++++++
+
+If the JSON serializer cannot handle the container contents directly it will
+fall back in the following manner:
+
+* if the dtype is unsupported (e.g. ``np.complex_``) then the ``default_handler``, if provided, will be called
+  for each value, otherwise an exception is raised.
+
+* if an object is unsupported it will attempt the following:
+
+
+    - check if the object has defined a ``toDict`` method and call it.
+      A ``toDict`` method should return a ``dict`` which will then be JSON serialized.
+
+    - invoke the ``default_handler`` if one was provided.
+
+    - convert the object to a ``dict`` by traversing its contents. However this will often fail
+      with an ``OverflowError`` or give unexpected results.
+
+In general the best approach for unsupported objects or dtypes is to provide a ``default_handler``.
+For example:
+
+.. code-block:: python
+
+  >>> DataFrame([1.0, 2.0, complex(1.0, 2.0)]).to_json()  # raises
+  RuntimeError: Unhandled numpy dtype 15
+
+can be dealt with by specifying a simple ``default_handler``:
+
+.. ipython:: python
+
+   pd.DataFrame([1.0, 2.0, complex(1.0, 2.0)]).to_json(default_handler=str)
+
+.. _io.json_reader:
+
+Reading JSON
+''''''''''''
+
+Reading a JSON string to pandas object can take a number of parameters.
+The parser will try to parse a ``DataFrame`` if ``typ`` is not supplied or
+is ``None``. To explicitly force ``Series`` parsing, pass ``typ=series``
+
+* ``filepath_or_buffer`` : a **VALID** JSON string or file handle / StringIO. The string could be
+  a URL. Valid URL schemes include http, ftp, S3, and file. For file URLs, a host
+  is expected. For instance, a local file could be
+  file ://localhost/path/to/table.json
+* ``typ``    : type of object to recover (series or frame), default 'frame'
+* ``orient`` :
+
+  Series :
+      * default is ``index``
+      * allowed values are {``split``, ``records``, ``index``}
+
+  DataFrame
+      * default is ``columns``
+      * allowed values are {``split``, ``records``, ``index``, ``columns``, ``values``, ``table``}
+
+  The format of the JSON string
+
+  .. csv-table::
+     :widths: 20, 150
+
+     ``split``, dict like {index -> [index]; columns -> [columns]; data -> [values]}
+     ``records``, list like [{column -> value} ...]
+     ``index``, dict like {index -> {column -> value}}
+     ``columns``, dict like {column -> {index -> value}}
+     ``values``, just the values array
+     ``table``, adhering to the JSON `Table Schema`_
+
+
+* ``dtype`` : if True, infer dtypes, if a dict of column to dtype, then use those, if ``False``, then don't infer dtypes at all, default is True, apply only to the data.
+* ``convert_axes`` : boolean, try to convert the axes to the proper dtypes, default is ``True``
+* ``convert_dates`` : a list of columns to parse for dates; If ``True``, then try to parse date-like columns, default is ``True``.
+* ``keep_default_dates`` : boolean, default ``True``. If parsing dates, then parse the default date-like columns.
+* ``precise_float`` : boolean, default ``False``. Set to enable usage of higher precision (strtod) function when decoding string to double values. Default (``False``) is to use fast but less precise builtin functionality.
+* ``date_unit`` : string, the timestamp unit to detect if converting dates. Default
+  None. By default the timestamp precision will be detected, if this is not desired
+  then pass one of 's', 'ms', 'us' or 'ns' to force timestamp precision to
+  seconds, milliseconds, microseconds or nanoseconds respectively.
+* ``lines`` : reads file as one json object per line.
+* ``encoding`` : The encoding to use to decode py3 bytes.
+* ``chunksize`` : when used in combination with ``lines=True``, return a ``pandas.api.typing.JsonReader`` which reads in ``chunksize`` lines per iteration.
+* ``engine``: Either ``"ujson"``, the built-in JSON parser, or ``"pyarrow"`` which dispatches to pyarrow's ``pyarrow.json.read_json``.
+  The ``"pyarrow"`` is only available when ``lines=True``
+
+The parser will raise one of ``ValueError/TypeError/AssertionError`` if the JSON is not parseable.
+
+If a non-default ``orient`` was used when encoding to JSON be sure to pass the same
+option here so that decoding produces sensible results, see `Orient Options`_ for an
+overview.
+
+Data conversion
++++++++++++++++
+
+The default of ``convert_axes=True``, ``dtype=True``, and ``convert_dates=True``
+will try to parse the axes, and all of the data into appropriate types,
+including dates. If you need to override specific dtypes, pass a dict to
+``dtype``. ``convert_axes`` should only be set to ``False`` if you need to
+preserve string-like numbers (e.g. '1', '2') in an axes.
+
+.. note::
+
+  Large integer values may be converted to dates if ``convert_dates=True`` and the data and / or column labels appear 'date-like'. The exact threshold depends on the ``date_unit`` specified. 'date-like' means that the column label meets one of the following criteria:
+
+  * it ends with ``'_at'``
+  * it ends with ``'_time'``
+  * it begins with ``'timestamp'``
+  * it is ``'modified'``
+  * it is ``'date'``
+
+.. warning::
+
+   When reading JSON data, automatic coercing into dtypes has some quirks:
+
+   * an index can be reconstructed in a different order from serialization, that is, the returned order is not guaranteed to be the same as before serialization
+   * a column that was ``float`` data will be converted to ``integer`` if it can be done safely, e.g. a column of ``1.``
+   * bool columns will be converted to ``integer`` on reconstruction
+
+   Thus there are times where you may want to specify specific dtypes via the ``dtype`` keyword argument.
+
+Reading from a JSON string:
+
+.. ipython:: python
+
+   from io import StringIO
+   pd.read_json(StringIO(json))
+
+Reading from a file:
+
+.. ipython:: python
+
+   pd.read_json("test.json")
+
+Don't convert any data (but still convert axes and dates):
+
+.. ipython:: python
+
+   pd.read_json("test.json", dtype=object).dtypes
+
+Specify dtypes for conversion:
+
+.. ipython:: python
+
+   pd.read_json("test.json", dtype={"A": "float32", "bools": "int8"}).dtypes
+
+Preserve string indices:
+
+.. ipython:: python
+
+   from io import StringIO
+   si = pd.DataFrame(
+       np.zeros((4, 4)), columns=list(range(4)), index=[str(i) for i in range(4)]
+   )
+   si
+   si.index
+   si.columns
+   json = si.to_json()
+
+   sij = pd.read_json(StringIO(json), convert_axes=False)
+   sij
+   sij.index
+   sij.columns
+
+Dates written in nanoseconds need to be read back in nanoseconds:
+
+.. ipython:: python
+
+   from io import StringIO
+   json = dfj2.to_json(date_format="iso", date_unit="ns")
+
+   # Try to parse timestamps as milliseconds -> Won't Work
+   dfju = pd.read_json(StringIO(json), date_unit="ms")
+   dfju
+
+   # Let pandas detect the correct precision
+   dfju = pd.read_json(StringIO(json))
+   dfju
+
+   # Or specify that all timestamps are in nanoseconds
+   dfju = pd.read_json(StringIO(json), date_unit="ns")
+   dfju
+
+By setting the ``dtype_backend`` argument you can control the default dtypes used for the resulting DataFrame.
+
+.. ipython:: python
+
+    data = (
+     '{"a":{"0":1,"1":3},"b":{"0":2.5,"1":4.5},"c":{"0":true,"1":false},"d":{"0":"a","1":"b"},'
+     '"e":{"0":null,"1":6.0},"f":{"0":null,"1":7.5},"g":{"0":null,"1":true},"h":{"0":null,"1":"a"},'
+     '"i":{"0":"12-31-2019","1":"12-31-2019"},"j":{"0":null,"1":null}}'
+    )
+    df = pd.read_json(StringIO(data), dtype_backend="pyarrow")
+    df
+    df.dtypes
+
+.. _io.json_normalize:
+
+Normalization
+'''''''''''''
+
+pandas provides a utility function to take a dict or list of dicts and *normalize* this semi-structured data
+into a flat table.
+
+.. ipython:: python
+
+   data = [
+       {"id": 1, "name": {"first": "Coleen", "last": "Volk"}},
+       {"name": {"given": "Mark", "family": "Regner"}},
+       {"id": 2, "name": "Faye Raker"},
+   ]
+   pd.json_normalize(data)
+
+.. ipython:: python
+
+   data = [
+       {
+           "state": "Florida",
+           "shortname": "FL",
+           "info": {"governor": "Rick Scott"},
+           "county": [
+               {"name": "Dade", "population": 12345},
+               {"name": "Broward", "population": 40000},
+               {"name": "Palm Beach", "population": 60000},
+           ],
+       },
+       {
+           "state": "Ohio",
+           "shortname": "OH",
+           "info": {"governor": "John Kasich"},
+           "county": [
+               {"name": "Summit", "population": 1234},
+               {"name": "Cuyahoga", "population": 1337},
+           ],
+       },
+   ]
+
+   pd.json_normalize(data, "county", ["state", "shortname", ["info", "governor"]])
+
+The max_level parameter provides more control over which level to end normalization.
+With max_level=1 the following snippet normalizes until 1st nesting level of the provided dict.
+
+.. ipython:: python
+
+    data = [
+        {
+            "CreatedBy": {"Name": "User001"},
+            "Lookup": {
+                "TextField": "Some text",
+                "UserField": {"Id": "ID001", "Name": "Name001"},
+            },
+            "Image": {"a": "b"},
+        }
+    ]
+    pd.json_normalize(data, max_level=1)
+
+.. _io.jsonl:
+
+Line delimited json
+'''''''''''''''''''
+
+pandas is able to read and write line-delimited json files that are common in data processing pipelines
+using Hadoop or Spark.
+
+For line-delimited json files, pandas can also return an iterator which reads in ``chunksize`` lines at a time. This can be useful for large files or to read from a stream.
+
+.. ipython:: python
+
+  from io import StringIO
+  jsonl = """
+      {"a": 1, "b": 2}
+      {"a": 3, "b": 4}
+  """
+  df = pd.read_json(StringIO(jsonl), lines=True)
+  df
+  df.to_json(orient="records", lines=True)
+
+  # reader is an iterator that returns ``chunksize`` lines each iteration
+  with pd.read_json(StringIO(jsonl), lines=True, chunksize=1) as reader:
+      reader
+      for chunk in reader:
+          print(chunk)
+
+Line-limited json can also be read using the pyarrow reader by specifying ``engine="pyarrow"``.
+
+.. ipython:: python
+
+   from io import BytesIO
+   df = pd.read_json(BytesIO(jsonl.encode()), lines=True, engine="pyarrow")
+   df
+
+.. versionadded:: 2.0.0
+
+.. _io.table_schema:
+
+Table schema
+''''''''''''
+
+`Table Schema`_ is a spec for describing tabular datasets as a JSON
+object. The JSON includes information on the field names, types, and
+other attributes. You can use the orient ``table`` to build
+a JSON string with two fields, ``schema`` and ``data``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": [1, 2, 3],
+           "B": ["a", "b", "c"],
+           "C": pd.date_range("2016-01-01", freq="D", periods=3),
+       },
+       index=pd.Index(range(3), name="idx"),
+   )
+   df
+   df.to_json(orient="table", date_format="iso")
+
+The ``schema`` field contains the ``fields`` key, which itself contains
+a list of column name to type pairs, including the ``Index`` or ``MultiIndex``
+(see below for a list of types).
+The ``schema`` field also contains a ``primaryKey`` field if the (Multi)index
+is unique.
+
+The second field, ``data``, contains the serialized data with the ``records``
+orient.
+The index is included, and any datetimes are ISO 8601 formatted, as required
+by the Table Schema spec.
+
+The full list of types supported are described in the Table Schema
+spec. This table shows the mapping from pandas types:
+
+=============== =================
+pandas type     Table Schema type
+=============== =================
+int64           integer
+float64         number
+bool            boolean
+datetime64[ns]  datetime
+timedelta64[ns] duration
+categorical     any
+object          str
+=============== =================
+
+A few notes on the generated table schema:
+
+* The ``schema`` object contains a ``pandas_version`` field. This contains
+  the version of pandas' dialect of the schema, and will be incremented
+  with each revision.
+* All dates are converted to UTC when serializing. Even timezone naive values,
+  which are treated as UTC with an offset of 0.
+
+  .. ipython:: python
+
+     from pandas.io.json import build_table_schema
+
+     s = pd.Series(pd.date_range("2016", periods=4))
+     build_table_schema(s)
+
+* datetimes with a timezone (before serializing), include an additional field
+  ``tz`` with the time zone name (e.g. ``'US/Central'``).
+
+  .. ipython:: python
+
+     s_tz = pd.Series(pd.date_range("2016", periods=12, tz="US/Central"))
+     build_table_schema(s_tz)
+
+* Periods are converted to timestamps before serialization, and so have the
+  same behavior of being converted to UTC. In addition, periods will contain
+  and additional field ``freq`` with the period's frequency, e.g. ``'A-DEC'``.
+
+  .. ipython:: python
+
+     s_per = pd.Series(1, index=pd.period_range("2016", freq="Y-DEC", periods=4))
+     build_table_schema(s_per)
+
+* Categoricals use the ``any`` type and an ``enum`` constraint listing
+  the set of possible values. Additionally, an ``ordered`` field is included:
+
+  .. ipython:: python
+
+     s_cat = pd.Series(pd.Categorical(["a", "b", "a"]))
+     build_table_schema(s_cat)
+
+* A ``primaryKey`` field, containing an array of labels, is included
+  *if the index is unique*:
+
+  .. ipython:: python
+
+     s_dupe = pd.Series([1, 2], index=[1, 1])
+     build_table_schema(s_dupe)
+
+* The ``primaryKey`` behavior is the same with MultiIndexes, but in this
+  case the ``primaryKey`` is an array:
+
+  .. ipython:: python
+
+     s_multi = pd.Series(1, index=pd.MultiIndex.from_product([("a", "b"), (0, 1)]))
+     build_table_schema(s_multi)
+
+* The default naming roughly follows these rules:
+
+    - For series, the ``object.name`` is used. If that's none, then the
+      name is ``values``
+    - For ``DataFrames``, the stringified version of the column name is used
+    - For ``Index`` (not ``MultiIndex``), ``index.name`` is used, with a
+      fallback to ``index`` if that is None.
+    - For ``MultiIndex``, ``mi.names`` is used. If any level has no name,
+      then ``level_<i>`` is used.
+
+``read_json`` also accepts ``orient='table'`` as an argument. This allows for
+the preservation of metadata such as dtypes and index names in a
+round-trippable manner.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "foo": [1, 2, 3, 4],
+           "bar": ["a", "b", "c", "d"],
+           "baz": pd.date_range("2018-01-01", freq="D", periods=4),
+           "qux": pd.Categorical(["a", "b", "c", "c"]),
+       },
+       index=pd.Index(range(4), name="idx"),
+   )
+   df
+   df.dtypes
+
+   df.to_json("test.json", orient="table")
+   new_df = pd.read_json("test.json", orient="table")
+   new_df
+   new_df.dtypes
+
+Please note that the literal string 'index' as the name of an :class:`Index`
+is not round-trippable, nor are any names beginning with ``'level_'`` within a
+:class:`MultiIndex`. These are used by default in :func:`DataFrame.to_json` to
+indicate missing values and the subsequent read cannot distinguish the intent.
+
+.. ipython:: python
+   :okwarning:
+
+   df.index.name = "index"
+   df.to_json("test.json", orient="table")
+   new_df = pd.read_json("test.json", orient="table")
+   print(new_df.index.name)
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("test.json")
+
+When using ``orient='table'`` along with user-defined ``ExtensionArray``,
+the generated schema will contain an additional ``extDtype`` key in the respective
+``fields`` element. This extra key is not standard but does enable JSON roundtrips
+for extension types (e.g. ``read_json(df.to_json(orient="table"), orient="table")``).
+
+The ``extDtype`` key carries the name of the extension, if you have properly registered
+the ``ExtensionDtype``, pandas will use said name to perform a lookup into the registry
+and re-convert the serialized data into your custom dtype.
+
+.. _Table Schema: https://specs.frictionlessdata.io/table-schema/
+
+
+HTML
+----
+
+.. _io.read_html:
+
+Reading HTML content
+''''''''''''''''''''''
+
+.. warning::
+
+   We **highly encourage** you to read the :ref:`HTML Table Parsing gotchas <io.html.gotchas>`
+   below regarding the issues surrounding the BeautifulSoup4/html5lib/lxml parsers.
+
+The top-level :func:`~pandas.io.html.read_html` function can accept an HTML
+string/file/URL and will parse HTML tables into list of pandas ``DataFrames``.
+Let's look at a few examples.
+
+.. note::
+
+   ``read_html`` returns a ``list`` of ``DataFrame`` objects, even if there is
+   only a single table contained in the HTML content.
+
+Read a URL with no options:
+
+.. code-block:: ipython
+
+   In [320]: url = "https://www.fdic.gov/resources/resolutions/bank-failures/failed-bank-list"
+
+   In [321]: pd.read_html(url)
+   Out[321]:
+   [                         Bank NameBank           CityCity StateSt  ...              Acquiring InstitutionAI Closing DateClosing FundFund
+    0                    Almena State Bank             Almena      KS  ...                          Equity Bank    October 23, 2020    10538
+    1           First City Bank of Florida  Fort Walton Beach      FL  ...            United Fidelity Bank, fsb    October 16, 2020    10537
+    2                 The First State Bank      Barboursville      WV  ...                       MVB Bank, Inc.       April 3, 2020    10536
+    3                   Ericson State Bank            Ericson      NE  ...           Farmers and Merchants Bank   February 14, 2020    10535
+    4     City National Bank of New Jersey             Newark      NJ  ...                      Industrial Bank    November 1, 2019    10534
+    ..                                 ...                ...     ...  ...                                  ...                 ...      ...
+    558                 Superior Bank, FSB           Hinsdale      IL  ...                Superior Federal, FSB       July 27, 2001     6004
+    559                Malta National Bank              Malta      OH  ...                    North Valley Bank         May 3, 2001     4648
+    560    First Alliance Bank & Trust Co.         Manchester      NH  ...  Southern New Hampshire Bank & Trust    February 2, 2001     4647
+    561  National State Bank of Metropolis         Metropolis      IL  ...              Banterra Bank of Marion   December 14, 2000     4646
+    562                   Bank of Honolulu           Honolulu      HI  ...                   Bank of the Orient    October 13, 2000     4645
+
+    [563 rows x 7 columns]]
+
+.. note::
+
+   The data from the above URL changes every Monday so the resulting data above may be slightly different.
+
+Read a URL while passing headers alongside the HTTP request:
+
+.. code-block:: ipython
+
+   In [322]: url = 'https://www.sump.org/notes/request/' # HTTP request reflector
+
+   In [323]: pd.read_html(url)
+   Out[323]:
+   [                   0                    1
+    0     Remote Socket:  51.15.105.256:51760
+    1  Protocol Version:             HTTP/1.1
+    2    Request Method:                  GET
+    3       Request URI:      /notes/request/
+    4     Request Query:                  NaN,
+    0   Accept-Encoding:             identity
+    1              Host:         www.sump.org
+    2        User-Agent:    Python-urllib/3.8
+    3        Connection:                close]
+
+   In [324]: headers = {
+      .....:    'User-Agent':'Mozilla Firefox v14.0',
+      .....:    'Accept':'application/json',
+      .....:    'Connection':'keep-alive',
+      .....:    'Auth':'Bearer 2*/f3+fe68df*4'
+      .....: }
+
+   In [325]: pd.read_html(url, storage_options=headers)
+   Out[325]:
+   [                   0                    1
+    0     Remote Socket:  51.15.105.256:51760
+    1  Protocol Version:             HTTP/1.1
+    2    Request Method:                  GET
+    3       Request URI:      /notes/request/
+    4     Request Query:                  NaN,
+    0        User-Agent: Mozilla Firefox v14.0
+    1    AcceptEncoding:   gzip,  deflate,  br
+    2            Accept:      application/json
+    3        Connection:             keep-alive
+    4              Auth:  Bearer 2*/f3+fe68df*4]
+
+.. note::
+
+   We see above that the headers we passed are reflected in the HTTP request.
+
+Read in the content of the file from the above URL and pass it to ``read_html``
+as a string:
+
+.. ipython:: python
+
+   html_str = """
+            <table>
+                <tr>
+                    <th>A</th>
+                    <th colspan="1">B</th>
+                    <th rowspan="1">C</th>
+                </tr>
+                <tr>
+                    <td>a</td>
+                    <td>b</td>
+                    <td>c</td>
+                </tr>
+            </table>
+        """
+
+   with open("tmp.html", "w") as f:
+       f.write(html_str)
+   df = pd.read_html("tmp.html")
+   df[0]
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("tmp.html")
+
+You can even pass in an instance of ``StringIO`` if you so desire:
+
+.. ipython:: python
+
+   dfs = pd.read_html(StringIO(html_str))
+   dfs[0]
+
+.. note::
+
+   The following examples are not run by the IPython evaluator due to the fact
+   that having so many network-accessing functions slows down the documentation
+   build. If you spot an error or an example that doesn't run, please do not
+   hesitate to report it over on `pandas GitHub issues page
+   <https://github.com/pandas-dev/pandas/issues>`__.
+
+
+Read a URL and match a table that contains specific text:
+
+.. code-block:: python
+
+   match = "Metcalf Bank"
+   df_list = pd.read_html(url, match=match)
+
+Specify a header row (by default ``<th>`` or ``<td>`` elements located within a
+``<thead>`` are used to form the column index, if multiple rows are contained within
+``<thead>`` then a MultiIndex is created); if specified, the header row is taken
+from the data minus the parsed header elements (``<th>`` elements).
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, header=0)
+
+Specify an index column:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, index_col=0)
+
+Specify a number of rows to skip:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, skiprows=0)
+
+Specify a number of rows to skip using a list (``range`` works
+as well):
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, skiprows=range(2))
+
+Specify an HTML attribute:
+
+.. code-block:: python
+
+   dfs1 = pd.read_html(url, attrs={"id": "table"})
+   dfs2 = pd.read_html(url, attrs={"class": "sortable"})
+   print(np.array_equal(dfs1[0], dfs2[0]))  # Should be True
+
+Specify values that should be converted to NaN:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, na_values=["No Acquirer"])
+
+Specify whether to keep the default set of NaN values:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, keep_default_na=False)
+
+Specify converters for columns. This is useful for numerical text data that has
+leading zeros.  By default columns that are numerical are cast to numeric
+types and the leading zeros are lost. To avoid this, we can convert these
+columns to strings.
+
+.. code-block:: python
+
+   url_mcc = "https://en.wikipedia.org/wiki/Mobile_country_code?oldid=899173761"
+   dfs = pd.read_html(
+       url_mcc,
+       match="Telekom Albania",
+       header=0,
+       converters={"MNC": str},
+   )
+
+Use some combination of the above:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, match="Metcalf Bank", index_col=0)
+
+Read in pandas ``to_html`` output (with some loss of floating point precision):
+
+.. code-block:: python
+
+   df = pd.DataFrame(np.random.randn(2, 2))
+   s = df.to_html(float_format="{0:.40g}".format)
+   dfin = pd.read_html(s, index_col=0)
+
+The ``lxml`` backend will raise an error on a failed parse if that is the only
+parser you provide. If you only have a single parser you can provide just a
+string, but it is considered good practice to pass a list with one string if,
+for example, the function expects a sequence of strings. You may use:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, "Metcalf Bank", index_col=0, flavor=["lxml"])
+
+Or you could pass ``flavor='lxml'`` without a list:
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, "Metcalf Bank", index_col=0, flavor="lxml")
+
+However, if you have bs4 and html5lib installed and pass ``None`` or ``['lxml',
+'bs4']`` then the parse will most likely succeed. Note that *as soon as a parse
+succeeds, the function will return*.
+
+.. code-block:: python
+
+   dfs = pd.read_html(url, "Metcalf Bank", index_col=0, flavor=["lxml", "bs4"])
+
+Links can be extracted from cells along with the text using ``extract_links="all"``.
+
+.. ipython:: python
+
+    html_table = """
+    <table>
+      <tr>
+        <th>GitHub</th>
+      </tr>
+      <tr>
+        <td><a href="https://github.com/pandas-dev/pandas">pandas</a></td>
+      </tr>
+    </table>
+    """
+
+    df = pd.read_html(
+        StringIO(html_table),
+        extract_links="all"
+    )[0]
+    df
+    df[("GitHub", None)]
+    df[("GitHub", None)].str[1]
+
+.. versionadded:: 1.5.0
+
+.. _io.html:
+
+Writing to HTML files
+''''''''''''''''''''''
+
+``DataFrame`` objects have an instance method ``to_html`` which renders the
+contents of the ``DataFrame`` as an HTML table. The function arguments are as
+in the method ``to_string`` described above.
+
+.. note::
+
+   Not all of the possible options for ``DataFrame.to_html`` are shown here for
+   brevity's sake. See :func:`.DataFrame.to_html` for the
+   full set of options.
+
+.. note::
+
+   In an HTML-rendering supported environment like a Jupyter Notebook, ``display(HTML(...))```
+   will render the raw HTML into the environment.
+
+.. ipython:: python
+
+   from IPython.display import display, HTML
+
+   df = pd.DataFrame(np.random.randn(2, 2))
+   df
+   html = df.to_html()
+   print(html)  # raw html
+   display(HTML(html))
+
+The ``columns`` argument will limit the columns shown:
+
+.. ipython:: python
+
+   html = df.to_html(columns=[0])
+   print(html)
+   display(HTML(html))
+
+``float_format`` takes a Python callable to control the precision of floating
+point values:
+
+.. ipython:: python
+
+   html = df.to_html(float_format="{0:.10f}".format)
+   print(html)
+   display(HTML(html))
+
+
+``bold_rows`` will make the row labels bold by default, but you can turn that
+off:
+
+.. ipython:: python
+
+   html = df.to_html(bold_rows=False)
+   print(html)
+   display(HTML(html))
+
+
+The ``classes`` argument provides the ability to give the resulting HTML
+table CSS classes. Note that these classes are *appended* to the existing
+``'dataframe'`` class.
+
+.. ipython:: python
+
+   print(df.to_html(classes=["awesome_table_class", "even_more_awesome_class"]))
+
+The ``render_links`` argument provides the ability to add hyperlinks to cells
+that contain URLs.
+
+.. ipython:: python
+
+   url_df = pd.DataFrame(
+       {
+           "name": ["Python", "pandas"],
+           "url": ["https://www.python.org/", "https://pandas.pydata.org"],
+       }
+   )
+   html = url_df.to_html(render_links=True)
+   print(html)
+   display(HTML(html))
+
+Finally, the ``escape`` argument allows you to control whether the
+"<", ">" and "&" characters escaped in the resulting HTML (by default it is
+``True``). So to get the HTML without escaped characters pass ``escape=False``
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": list("&<>"), "b": np.random.randn(3)})
+
+Escaped:
+
+.. ipython:: python
+
+   html = df.to_html()
+   print(html)
+   display(HTML(html))
+
+Not escaped:
+
+.. ipython:: python
+
+   html = df.to_html(escape=False)
+   print(html)
+   display(HTML(html))
+
+.. note::
+
+   Some browsers may not show a difference in the rendering of the previous two
+   HTML tables.
+
+
+.. _io.html.gotchas:
+
+HTML Table Parsing Gotchas
+''''''''''''''''''''''''''
+
+There are some versioning issues surrounding the libraries that are used to
+parse HTML tables in the top-level pandas io function ``read_html``.
+
+**Issues with** |lxml|_
+
+* Benefits
+
+    - |lxml|_ is very fast.
+
+    - |lxml|_ requires Cython to install correctly.
+
+* Drawbacks
+
+    - |lxml|_ does *not* make any guarantees about the results of its parse
+      *unless* it is given |svm|_.
+
+    - In light of the above, we have chosen to allow you, the user, to use the
+      |lxml|_ backend, but **this backend will use** |html5lib|_ if |lxml|_
+      fails to parse
+
+    - It is therefore *highly recommended* that you install both
+      |BeautifulSoup4|_ and |html5lib|_, so that you will still get a valid
+      result (provided everything else is valid) even if |lxml|_ fails.
+
+**Issues with** |BeautifulSoup4|_ **using** |lxml|_ **as a backend**
+
+* The above issues hold here as well since |BeautifulSoup4|_ is essentially
+  just a wrapper around a parser backend.
+
+**Issues with** |BeautifulSoup4|_ **using** |html5lib|_ **as a backend**
+
+* Benefits
+
+    - |html5lib|_ is far more lenient than |lxml|_ and consequently deals
+      with *real-life markup* in a much saner way rather than just, e.g.,
+      dropping an element without notifying you.
+
+    - |html5lib|_ *generates valid HTML5 markup from invalid markup
+      automatically*. This is extremely important for parsing HTML tables,
+      since it guarantees a valid document. However, that does NOT mean that
+      it is "correct", since the process of fixing markup does not have a
+      single definition.
+
+    - |html5lib|_ is pure Python and requires no additional build steps beyond
+      its own installation.
+
+* Drawbacks
+
+    - The biggest drawback to using |html5lib|_ is that it is slow as
+      molasses.  However consider the fact that many tables on the web are not
+      big enough for the parsing algorithm runtime to matter. It is more
+      likely that the bottleneck will be in the process of reading the raw
+      text from the URL over the web, i.e., IO (input-output). For very large
+      tables, this might not be true.
+
+
+.. |svm| replace:: **strictly valid markup**
+.. _svm: https://validator.w3.org/docs/help.html#validation_basics
+
+.. |html5lib| replace:: **html5lib**
+.. _html5lib: https://github.com/html5lib/html5lib-python
+
+.. |BeautifulSoup4| replace:: **BeautifulSoup4**
+.. _BeautifulSoup4: https://www.crummy.com/software/BeautifulSoup
+
+.. |lxml| replace:: **lxml**
+.. _lxml: https://lxml.de
+
+.. _io.latex:
+
+LaTeX
+-----
+
+.. versionadded:: 1.3.0
+
+Currently there are no methods to read from LaTeX, only output methods.
+
+Writing to LaTeX files
+''''''''''''''''''''''
+
+.. note::
+
+   DataFrame *and* Styler objects currently have a ``to_latex`` method. We recommend
+   using the `Styler.to_latex() <../reference/api/pandas.io.formats.style.Styler.to_latex.rst>`__ method
+   over `DataFrame.to_latex() <../reference/api/pandas.DataFrame.to_latex.rst>`__ due to the former's greater flexibility with
+   conditional styling, and the latter's possible future deprecation.
+
+Review the documentation for `Styler.to_latex <../reference/api/pandas.io.formats.style.Styler.to_latex.rst>`__,
+which gives examples of conditional styling and explains the operation of its keyword
+arguments.
+
+For simple application the following pattern is sufficient.
+
+.. ipython:: python
+
+   df = pd.DataFrame([[1, 2], [3, 4]], index=["a", "b"], columns=["c", "d"])
+   print(df.style.to_latex())
+
+To format values before output, chain the `Styler.format <../reference/api/pandas.io.formats.style.Styler.format.rst>`__
+method.
+
+.. ipython:: python
+
+   print(df.style.format("€ {}").to_latex())
+
+XML
+---
+
+.. _io.read_xml:
+
+Reading XML
+'''''''''''
+
+.. versionadded:: 1.3.0
+
+The top-level :func:`~pandas.io.xml.read_xml` function can accept an XML
+string/file/URL and will parse nodes and attributes into a pandas ``DataFrame``.
+
+.. note::
+
+   Since there is no standard XML structure where design types can vary in
+   many ways, ``read_xml`` works best with flatter, shallow versions. If
+   an XML document is deeply nested, use the ``stylesheet`` feature to
+   transform XML into a flatter version.
+
+Let's look at a few examples.
+
+Read an XML string:
+
+.. ipython:: python
+
+    from io import StringIO
+   xml = """<?xml version="1.0" encoding="UTF-8"?>
+   <bookstore>
+     <book category="cooking">
+       <title lang="en">Everyday Italian</title>
+       <author>Giada De Laurentiis</author>
+       <year>2005</year>
+       <price>30.00</price>
+     </book>
+     <book category="children">
+       <title lang="en">Harry Potter</title>
+       <author>J K. Rowling</author>
+       <year>2005</year>
+       <price>29.99</price>
+     </book>
+     <book category="web">
+       <title lang="en">Learning XML</title>
+       <author>Erik T. Ray</author>
+       <year>2003</year>
+       <price>39.95</price>
+     </book>
+   </bookstore>"""
+
+   df = pd.read_xml(StringIO(xml))
+   df
+
+Read a URL with no options:
+
+.. ipython:: python
+
+   df = pd.read_xml("https://www.w3schools.com/xml/books.xml")
+   df
+
+Read in the content of the "books.xml" file and pass it to ``read_xml``
+as a string:
+
+.. ipython:: python
+
+   file_path = "books.xml"
+   with open(file_path, "w") as f:
+       f.write(xml)
+
+   with open(file_path, "r") as f:
+       df = pd.read_xml(StringIO(f.read()))
+   df
+
+Read in the content of the "books.xml" as instance of ``StringIO`` or
+``BytesIO`` and pass it to ``read_xml``:
+
+.. ipython:: python
+
+   with open(file_path, "r") as f:
+       sio = StringIO(f.read())
+
+   df = pd.read_xml(sio)
+   df
+
+.. ipython:: python
+
+   with open(file_path, "rb") as f:
+       bio = BytesIO(f.read())
+
+   df = pd.read_xml(bio)
+   df
+
+Even read XML from AWS S3 buckets such as NIH NCBI PMC Article Datasets providing
+Biomedical and Life Science Journals:
+
+.. code-block:: python
+
+   >>> df = pd.read_xml(
+   ...    "s3://pmc-oa-opendata/oa_comm/xml/all/PMC1236943.xml",
+   ...    xpath=".//journal-meta",
+   ...)
+   >>> df
+         journal-id  journal-title  issn  publisher
+   0 Cardiovasc Ultrasound Cardiovascular Ultrasound 1476-7120 NaN
+
+With `lxml`_ as default ``parser``, you access the full-featured XML library
+that extends Python's ElementTree API. One powerful tool is ability to query
+nodes selectively or conditionally with more expressive XPath:
+
+.. _lxml: https://lxml.de
+
+.. ipython:: python
+
+   df = pd.read_xml(file_path, xpath="//book[year=2005]")
+   df
+
+Specify only elements or only attributes to parse:
+
+.. ipython:: python
+
+   df = pd.read_xml(file_path, elems_only=True)
+   df
+
+.. ipython:: python
+
+   df = pd.read_xml(file_path, attrs_only=True)
+   df
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("books.xml")
+
+XML documents can have namespaces with prefixes and default namespaces without
+prefixes both of which are denoted with a special attribute ``xmlns``. In order
+to parse by node under a namespace context, ``xpath`` must reference a prefix.
+
+For example, below XML contains a namespace with prefix, ``doc``, and URI at
+``https://example.com``. In order to parse ``doc:row`` nodes,
+``namespaces`` must be used.
+
+.. ipython:: python
+
+   xml = """<?xml version='1.0' encoding='utf-8'?>
+   <doc:data xmlns:doc="https://example.com">
+     <doc:row>
+       <doc:shape>square</doc:shape>
+       <doc:degrees>360</doc:degrees>
+       <doc:sides>4.0</doc:sides>
+     </doc:row>
+     <doc:row>
+       <doc:shape>circle</doc:shape>
+       <doc:degrees>360</doc:degrees>
+       <doc:sides/>
+     </doc:row>
+     <doc:row>
+       <doc:shape>triangle</doc:shape>
+       <doc:degrees>180</doc:degrees>
+       <doc:sides>3.0</doc:sides>
+     </doc:row>
+   </doc:data>"""
+
+   df = pd.read_xml(StringIO(xml),
+                    xpath="//doc:row",
+                    namespaces={"doc": "https://example.com"})
+   df
+
+Similarly, an XML document can have a default namespace without prefix. Failing
+to assign a temporary prefix will return no nodes and raise a ``ValueError``.
+But assigning *any* temporary name to correct URI allows parsing by nodes.
+
+.. ipython:: python
+
+   xml = """<?xml version='1.0' encoding='utf-8'?>
+   <data xmlns="https://example.com">
+    <row>
+      <shape>square</shape>
+      <degrees>360</degrees>
+      <sides>4.0</sides>
+    </row>
+    <row>
+      <shape>circle</shape>
+      <degrees>360</degrees>
+      <sides/>
+    </row>
+    <row>
+      <shape>triangle</shape>
+      <degrees>180</degrees>
+      <sides>3.0</sides>
+    </row>
+   </data>"""
+
+   df = pd.read_xml(StringIO(xml),
+                    xpath="//pandas:row",
+                    namespaces={"pandas": "https://example.com"})
+   df
+
+However, if XPath does not reference node names such as default, ``/*``, then
+``namespaces`` is not required.
+
+.. note::
+
+   Since ``xpath`` identifies the parent of content to be parsed, only immediate
+   descendants which include child nodes or current attributes are parsed.
+   Therefore, ``read_xml`` will not parse the text of grandchildren or other
+   descendants and will not parse attributes of any descendant. To retrieve
+   lower level content, adjust xpath to lower level. For example,
+
+   .. ipython:: python
+        :okwarning:
+
+      xml = """
+      <data>
+        <row>
+          <shape sides="4">square</shape>
+          <degrees>360</degrees>
+        </row>
+        <row>
+          <shape sides="0">circle</shape>
+          <degrees>360</degrees>
+        </row>
+        <row>
+          <shape sides="3">triangle</shape>
+          <degrees>180</degrees>
+        </row>
+      </data>"""
+
+      df = pd.read_xml(StringIO(xml), xpath="./row")
+      df
+
+   shows the attribute ``sides`` on ``shape`` element was not parsed as
+   expected since this attribute resides on the child of ``row`` element
+   and not ``row`` element itself. In other words, ``sides`` attribute is a
+   grandchild level descendant of ``row`` element. However, the ``xpath``
+   targets ``row`` element which covers only its children and attributes.
+
+With `lxml`_ as parser, you can flatten nested XML documents with an XSLT
+script which also can be string/file/URL types. As background, `XSLT`_ is
+a special-purpose language written in a special XML file that can transform
+original XML documents into other XML, HTML, even text (CSV, JSON, etc.)
+using an XSLT processor.
+
+.. _lxml: https://lxml.de
+.. _XSLT: https://www.w3.org/TR/xslt/
+
+For example, consider this somewhat nested structure of Chicago "L" Rides
+where station and rides elements encapsulate data in their own sections.
+With below XSLT, ``lxml`` can transform original nested document into a flatter
+output (as shown below for demonstration) for easier parse into ``DataFrame``:
+
+.. ipython:: python
+
+   xml = """<?xml version='1.0' encoding='utf-8'?>
+    <response>
+     <row>
+       <station id="40850" name="Library"/>
+       <month>2020-09-01T00:00:00</month>
+       <rides>
+         <avg_weekday_rides>864.2</avg_weekday_rides>
+         <avg_saturday_rides>534</avg_saturday_rides>
+         <avg_sunday_holiday_rides>417.2</avg_sunday_holiday_rides>
+       </rides>
+     </row>
+     <row>
+       <station id="41700" name="Washington/Wabash"/>
+       <month>2020-09-01T00:00:00</month>
+       <rides>
+         <avg_weekday_rides>2707.4</avg_weekday_rides>
+         <avg_saturday_rides>1909.8</avg_saturday_rides>
+         <avg_sunday_holiday_rides>1438.6</avg_sunday_holiday_rides>
+       </rides>
+     </row>
+     <row>
+       <station id="40380" name="Clark/Lake"/>
+       <month>2020-09-01T00:00:00</month>
+       <rides>
+         <avg_weekday_rides>2949.6</avg_weekday_rides>
+         <avg_saturday_rides>1657</avg_saturday_rides>
+         <avg_sunday_holiday_rides>1453.8</avg_sunday_holiday_rides>
+       </rides>
+     </row>
+    </response>"""
+
+   xsl = """<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+      <xsl:output method="xml" omit-xml-declaration="no" indent="yes"/>
+      <xsl:strip-space elements="*"/>
+      <xsl:template match="/response">
+         <xsl:copy>
+           <xsl:apply-templates select="row"/>
+         </xsl:copy>
+      </xsl:template>
+      <xsl:template match="row">
+         <xsl:copy>
+           <station_id><xsl:value-of select="station/@id"/></station_id>
+           <station_name><xsl:value-of select="station/@name"/></station_name>
+           <xsl:copy-of select="month|rides/*"/>
+         </xsl:copy>
+      </xsl:template>
+    </xsl:stylesheet>"""
+
+   output = """<?xml version='1.0' encoding='utf-8'?>
+    <response>
+      <row>
+         <station_id>40850</station_id>
+         <station_name>Library</station_name>
+         <month>2020-09-01T00:00:00</month>
+         <avg_weekday_rides>864.2</avg_weekday_rides>
+         <avg_saturday_rides>534</avg_saturday_rides>
+         <avg_sunday_holiday_rides>417.2</avg_sunday_holiday_rides>
+      </row>
+      <row>
+         <station_id>41700</station_id>
+         <station_name>Washington/Wabash</station_name>
+         <month>2020-09-01T00:00:00</month>
+         <avg_weekday_rides>2707.4</avg_weekday_rides>
+         <avg_saturday_rides>1909.8</avg_saturday_rides>
+         <avg_sunday_holiday_rides>1438.6</avg_sunday_holiday_rides>
+      </row>
+      <row>
+         <station_id>40380</station_id>
+         <station_name>Clark/Lake</station_name>
+         <month>2020-09-01T00:00:00</month>
+         <avg_weekday_rides>2949.6</avg_weekday_rides>
+         <avg_saturday_rides>1657</avg_saturday_rides>
+         <avg_sunday_holiday_rides>1453.8</avg_sunday_holiday_rides>
+      </row>
+    </response>"""
+
+   df = pd.read_xml(StringIO(xml), stylesheet=StringIO(xsl))
+   df
+
+For very large XML files that can range in hundreds of megabytes to gigabytes, :func:`pandas.read_xml`
+supports parsing such sizeable files using `lxml's iterparse`_ and `etree's iterparse`_
+which are memory-efficient methods to iterate through an XML tree and extract specific elements and attributes.
+without holding entire tree in memory.
+
+.. versionadded:: 1.5.0
+
+.. _`lxml's iterparse`: https://lxml.de/3.2/parsing.html#iterparse-and-iterwalk
+.. _`etree's iterparse`: https://docs.python.org/3/library/xml.etree.elementtree.html#xml.etree.ElementTree.iterparse
+
+To use this feature, you must pass a physical XML file path into ``read_xml`` and use the ``iterparse`` argument.
+Files should not be compressed or point to online sources but stored on local disk. Also, ``iterparse`` should be
+a dictionary where the key is the repeating nodes in document (which become the rows) and the value is a list of
+any element or attribute that is a descendant (i.e., child, grandchild) of repeating node. Since XPath is not
+used in this method, descendants do not need to share same relationship with one another. Below shows example
+of reading in Wikipedia's very large (12 GB+) latest article data dump.
+
+.. code-block:: ipython
+
+    In [1]: df = pd.read_xml(
+    ...         "/path/to/downloaded/enwikisource-latest-pages-articles.xml",
+    ...         iterparse = {"page": ["title", "ns", "id"]}
+    ...     )
+    ...     df
+    Out[2]:
+                                                         title   ns        id
+    0                                       Gettysburg Address    0     21450
+    1                                                Main Page    0     42950
+    2                            Declaration by United Nations    0      8435
+    3             Constitution of the United States of America    0      8435
+    4                     Declaration of Independence (Israel)    0     17858
+    ...                                                    ...  ...       ...
+    3578760               Page:Black cat 1897 07 v2 n10.pdf/17  104    219649
+    3578761               Page:Black cat 1897 07 v2 n10.pdf/43  104    219649
+    3578762               Page:Black cat 1897 07 v2 n10.pdf/44  104    219649
+    3578763      The History of Tom Jones, a Foundling/Book IX    0  12084291
+    3578764  Page:Shakespeare of Stratford (1926) Yale.djvu/91  104     21450
+
+    [3578765 rows x 3 columns]
+
+.. _io.xml:
+
+Writing XML
+'''''''''''
+
+.. versionadded:: 1.3.0
+
+``DataFrame`` objects have an instance method ``to_xml`` which renders the
+contents of the ``DataFrame`` as an XML document.
+
+.. note::
+
+   This method does not support special properties of XML including DTD,
+   CData, XSD schemas, processing instructions, comments, and others.
+   Only namespaces at the root level is supported. However, ``stylesheet``
+   allows design changes after initial output.
+
+Let's look at a few examples.
+
+Write an XML without options:
+
+.. ipython:: python
+
+   geom_df = pd.DataFrame(
+       {
+           "shape": ["square", "circle", "triangle"],
+           "degrees": [360, 360, 180],
+           "sides": [4, np.nan, 3],
+       }
+   )
+
+   print(geom_df.to_xml())
+
+
+Write an XML with new root and row name:
+
+.. ipython:: python
+
+   print(geom_df.to_xml(root_name="geometry", row_name="objects"))
+
+Write an attribute-centric XML:
+
+.. ipython:: python
+
+   print(geom_df.to_xml(attr_cols=geom_df.columns.tolist()))
+
+Write a mix of elements and attributes:
+
+.. ipython:: python
+
+   print(
+       geom_df.to_xml(
+           index=False,
+           attr_cols=['shape'],
+           elem_cols=['degrees', 'sides'])
+   )
+
+Any ``DataFrames`` with hierarchical columns will be flattened for XML element names
+with levels delimited by underscores:
+
+.. ipython:: python
+
+   ext_geom_df = pd.DataFrame(
+       {
+           "type": ["polygon", "other", "polygon"],
+           "shape": ["square", "circle", "triangle"],
+           "degrees": [360, 360, 180],
+           "sides": [4, np.nan, 3],
+       }
+   )
+
+   pvt_df = ext_geom_df.pivot_table(index='shape',
+                                    columns='type',
+                                    values=['degrees', 'sides'],
+                                    aggfunc='sum')
+   pvt_df
+
+   print(pvt_df.to_xml())
+
+Write an XML with default namespace:
+
+.. ipython:: python
+
+   print(geom_df.to_xml(namespaces={"": "https://example.com"}))
+
+Write an XML with namespace prefix:
+
+.. ipython:: python
+
+   print(
+       geom_df.to_xml(namespaces={"doc": "https://example.com"},
+                      prefix="doc")
+   )
+
+Write an XML without declaration or pretty print:
+
+.. ipython:: python
+
+   print(
+       geom_df.to_xml(xml_declaration=False,
+                      pretty_print=False)
+   )
+
+Write an XML and transform with stylesheet:
+
+.. ipython:: python
+
+   xsl = """<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+      <xsl:output method="xml" omit-xml-declaration="no" indent="yes"/>
+      <xsl:strip-space elements="*"/>
+      <xsl:template match="/data">
+        <geometry>
+          <xsl:apply-templates select="row"/>
+        </geometry>
+      </xsl:template>
+      <xsl:template match="row">
+        <object index="{index}">
+          <xsl:if test="shape!='circle'">
+              <xsl:attribute name="type">polygon</xsl:attribute>
+          </xsl:if>
+          <xsl:copy-of select="shape"/>
+          <property>
+              <xsl:copy-of select="degrees|sides"/>
+          </property>
+        </object>
+      </xsl:template>
+    </xsl:stylesheet>"""
+
+   print(geom_df.to_xml(stylesheet=StringIO(xsl)))
+
+
+XML Final Notes
+'''''''''''''''
+
+* All XML documents adhere to `W3C specifications`_. Both ``etree`` and ``lxml``
+  parsers will fail to parse any markup document that is not well-formed or
+  follows XML syntax rules. Do be aware HTML is not an XML document unless it
+  follows XHTML specs. However, other popular markup types including KML, XAML,
+  RSS, MusicML, MathML are compliant `XML schemas`_.
+
+* For above reason, if your application builds XML prior to pandas operations,
+  use appropriate DOM libraries like ``etree`` and ``lxml`` to build the necessary
+  document and not by string concatenation or regex adjustments. Always remember
+  XML is a *special* text file with markup rules.
+
+* With very large XML files (several hundred MBs to GBs), XPath and XSLT
+  can become memory-intensive operations. Be sure to have enough available
+  RAM for reading and writing to large XML files (roughly about 5 times the
+  size of text).
+
+* Because XSLT is a programming language, use it with caution since such scripts
+  can pose a security risk in your environment and can run large or infinite
+  recursive operations. Always test scripts on small fragments before full run.
+
+* The `etree`_ parser supports all functionality of both ``read_xml`` and
+  ``to_xml`` except for complex XPath and any XSLT. Though limited in features,
+  ``etree`` is still a reliable and capable parser and tree builder. Its
+  performance may trail ``lxml`` to a certain degree for larger files but
+  relatively unnoticeable on small to medium size files.
+
+.. _`W3C specifications`: https://www.w3.org/TR/xml/
+.. _`XML schemas`: https://en.wikipedia.org/wiki/List_of_types_of_XML_schemas
+.. _`etree`: https://docs.python.org/3/library/xml.etree.elementtree.html
+
+
+
+.. _io.excel:
+
+Excel files
+-----------
+
+The :func:`~pandas.read_excel` method can read Excel 2007+ (``.xlsx``) files
+using the ``openpyxl`` Python module. Excel 2003 (``.xls``) files
+can be read using ``xlrd``. Binary Excel (``.xlsb``)
+files can be read using ``pyxlsb``. All formats can be read
+using :ref:`calamine<io.calamine>` engine.
+The :meth:`~DataFrame.to_excel` instance method is used for
+saving a ``DataFrame`` to Excel.  Generally the semantics are
+similar to working with :ref:`csv<io.read_csv_table>` data.
+See the :ref:`cookbook<cookbook.excel>` for some advanced strategies.
+
+.. note::
+
+   When ``engine=None``, the following logic will be used to determine the engine:
+
+   - If ``path_or_buffer`` is an OpenDocument format (.odf, .ods, .odt),
+     then `odf <https://pypi.org/project/odfpy/>`_ will be used.
+   - Otherwise if ``path_or_buffer`` is an xls format, ``xlrd`` will be used.
+   - Otherwise if ``path_or_buffer`` is in xlsb format, ``pyxlsb`` will be used.
+   - Otherwise ``openpyxl`` will be used.
+
+.. _io.excel_reader:
+
+Reading Excel files
+'''''''''''''''''''
+
+In the most basic use-case, ``read_excel`` takes a path to an Excel
+file, and the ``sheet_name`` indicating which sheet to parse.
+
+When using the ``engine_kwargs`` parameter, pandas will pass these arguments to the
+engine. For this, it is important to know which function pandas is
+using internally.
+
+* For the engine openpyxl, pandas is using :func:`openpyxl.load_workbook` to read in (``.xlsx``) and (``.xlsm``) files.
+
+* For the engine xlrd, pandas is using :func:`xlrd.open_workbook` to read in (``.xls``) files.
+
+* For the engine pyxlsb, pandas is using :func:`pyxlsb.open_workbook` to read in (``.xlsb``) files.
+
+* For the engine odf, pandas is using :func:`odf.opendocument.load` to read in (``.ods``) files.
+
+* For the engine calamine, pandas is using :func:`python_calamine.load_workbook`
+  to read in (``.xlsx``), (``.xlsm``), (``.xls``), (``.xlsb``), (``.ods``) files.
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xls", sheet_name="Sheet1")
+
+
+.. _io.excel.excelfile_class:
+
+``ExcelFile`` class
++++++++++++++++++++
+
+To facilitate working with multiple sheets from the same file, the ``ExcelFile``
+class can be used to wrap the file and can be passed into ``read_excel``
+There will be a performance benefit for reading multiple sheets as the file is
+read into memory only once.
+
+.. code-block:: python
+
+   xlsx = pd.ExcelFile("path_to_file.xls")
+   df = pd.read_excel(xlsx, "Sheet1")
+
+The ``ExcelFile`` class can also be used as a context manager.
+
+.. code-block:: python
+
+   with pd.ExcelFile("path_to_file.xls") as xls:
+       df1 = pd.read_excel(xls, "Sheet1")
+       df2 = pd.read_excel(xls, "Sheet2")
+
+The ``sheet_names`` property will generate
+a list of the sheet names in the file.
+
+The primary use-case for an ``ExcelFile`` is parsing multiple sheets with
+different parameters:
+
+.. code-block:: python
+
+    data = {}
+    # For when Sheet1's format differs from Sheet2
+    with pd.ExcelFile("path_to_file.xls") as xls:
+        data["Sheet1"] = pd.read_excel(xls, "Sheet1", index_col=None, na_values=["NA"])
+        data["Sheet2"] = pd.read_excel(xls, "Sheet2", index_col=1)
+
+Note that if the same parsing parameters are used for all sheets, a list
+of sheet names can simply be passed to ``read_excel`` with no loss in performance.
+
+.. code-block:: python
+
+    # using the ExcelFile class
+    data = {}
+    with pd.ExcelFile("path_to_file.xls") as xls:
+        data["Sheet1"] = pd.read_excel(xls, "Sheet1", index_col=None, na_values=["NA"])
+        data["Sheet2"] = pd.read_excel(xls, "Sheet2", index_col=None, na_values=["NA"])
+
+    # equivalent using the read_excel function
+    data = pd.read_excel(
+        "path_to_file.xls", ["Sheet1", "Sheet2"], index_col=None, na_values=["NA"]
+    )
+
+``ExcelFile`` can also be called with a ``xlrd.book.Book`` object
+as a parameter. This allows the user to control how the excel file is read.
+For example, sheets can be loaded on demand by calling ``xlrd.open_workbook()``
+with ``on_demand=True``.
+
+.. code-block:: python
+
+    import xlrd
+
+    xlrd_book = xlrd.open_workbook("path_to_file.xls", on_demand=True)
+    with pd.ExcelFile(xlrd_book) as xls:
+        df1 = pd.read_excel(xls, "Sheet1")
+        df2 = pd.read_excel(xls, "Sheet2")
+
+.. _io.excel.specifying_sheets:
+
+Specifying sheets
++++++++++++++++++
+
+.. note:: The second argument is ``sheet_name``, not to be confused with ``ExcelFile.sheet_names``.
+
+.. note:: An ExcelFile's attribute ``sheet_names`` provides access to a list of sheets.
+
+* The arguments ``sheet_name`` allows specifying the sheet or sheets to read.
+* The default value for ``sheet_name`` is 0, indicating to read the first sheet
+* Pass a string to refer to the name of a particular sheet in the workbook.
+* Pass an integer to refer to the index of a sheet. Indices follow Python
+  convention, beginning at 0.
+* Pass a list of either strings or integers, to return a dictionary of specified sheets.
+* Pass a ``None`` to return a dictionary of all available sheets.
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xls", "Sheet1", index_col=None, na_values=["NA"])
+
+Using the sheet index:
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xls", 0, index_col=None, na_values=["NA"])
+
+Using all default values:
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xls")
+
+Using None to get all sheets:
+
+.. code-block:: python
+
+   # Returns a dictionary of DataFrames
+   pd.read_excel("path_to_file.xls", sheet_name=None)
+
+Using a list to get multiple sheets:
+
+.. code-block:: python
+
+   # Returns the 1st and 4th sheet, as a dictionary of DataFrames.
+   pd.read_excel("path_to_file.xls", sheet_name=["Sheet1", 3])
+
+``read_excel`` can read more than one sheet, by setting ``sheet_name`` to either
+a list of sheet names, a list of sheet positions, or ``None`` to read all sheets.
+Sheets can be specified by sheet index or sheet name, using an integer or string,
+respectively.
+
+.. _io.excel.reading_multiindex:
+
+Reading a ``MultiIndex``
+++++++++++++++++++++++++
+
+``read_excel`` can read a ``MultiIndex`` index, by passing a list of columns to ``index_col``
+and a ``MultiIndex`` column by passing a list of rows to ``header``.  If either the ``index``
+or ``columns`` have serialized level names those will be read in as well by specifying
+the rows/columns that make up the levels.
+
+For example, to read in a ``MultiIndex`` index without names:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"a": [1, 2, 3, 4], "b": [5, 6, 7, 8]},
+       index=pd.MultiIndex.from_product([["a", "b"], ["c", "d"]]),
+   )
+   df.to_excel("path_to_file.xlsx")
+   df = pd.read_excel("path_to_file.xlsx", index_col=[0, 1])
+   df
+
+If the index has level names, they will be parsed as well, using the same
+parameters.
+
+.. ipython:: python
+
+   df.index = df.index.set_names(["lvl1", "lvl2"])
+   df.to_excel("path_to_file.xlsx")
+   df = pd.read_excel("path_to_file.xlsx", index_col=[0, 1])
+   df
+
+
+If the source file has both ``MultiIndex`` index and columns, lists specifying each
+should be passed to ``index_col`` and ``header``:
+
+.. ipython:: python
+
+   df.columns = pd.MultiIndex.from_product([["a"], ["b", "d"]], names=["c1", "c2"])
+   df.to_excel("path_to_file.xlsx")
+   df = pd.read_excel("path_to_file.xlsx", index_col=[0, 1], header=[0, 1])
+   df
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("path_to_file.xlsx")
+
+Missing values in columns specified in ``index_col`` will be forward filled to
+allow roundtripping with ``to_excel`` for ``merged_cells=True``. To avoid forward
+filling the missing values use ``set_index`` after reading the data instead of
+``index_col``.
+
+Parsing specific columns
+++++++++++++++++++++++++
+
+It is often the case that users will insert columns to do temporary computations
+in Excel and you may not want to read in those columns. ``read_excel`` takes
+a ``usecols`` keyword to allow you to specify a subset of columns to parse.
+
+You can specify a comma-delimited set of Excel columns and ranges as a string:
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xls", "Sheet1", usecols="A,C:E")
+
+If ``usecols`` is a list of integers, then it is assumed to be the file column
+indices to be parsed.
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xls", "Sheet1", usecols=[0, 2, 3])
+
+Element order is ignored, so ``usecols=[0, 1]`` is the same as ``[1, 0]``.
+
+If ``usecols`` is a list of strings, it is assumed that each string corresponds
+to a column name provided either by the user in ``names`` or inferred from the
+document header row(s). Those strings define which columns will be parsed:
+
+.. code-block:: python
+
+    pd.read_excel("path_to_file.xls", "Sheet1", usecols=["foo", "bar"])
+
+Element order is ignored, so ``usecols=['baz', 'joe']`` is the same as ``['joe', 'baz']``.
+
+If ``usecols`` is callable, the callable function will be evaluated against
+the column names, returning names where the callable function evaluates to ``True``.
+
+.. code-block:: python
+
+    pd.read_excel("path_to_file.xls", "Sheet1", usecols=lambda x: x.isalpha())
+
+Parsing dates
++++++++++++++
+
+Datetime-like values are normally automatically converted to the appropriate
+dtype when reading the excel file. But if you have a column of strings that
+*look* like dates (but are not actually formatted as dates in excel), you can
+use the ``parse_dates`` keyword to parse those strings to datetimes:
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xls", "Sheet1", parse_dates=["date_strings"])
+
+
+Cell converters
++++++++++++++++
+
+It is possible to transform the contents of Excel cells via the ``converters``
+option. For instance, to convert a column to boolean:
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xls", "Sheet1", converters={"MyBools": bool})
+
+This options handles missing values and treats exceptions in the converters
+as missing data. Transformations are applied cell by cell rather than to the
+column as a whole, so the array dtype is not guaranteed. For instance, a
+column of integers with missing values cannot be transformed to an array
+with integer dtype, because NaN is strictly a float. You can manually mask
+missing data to recover integer dtype:
+
+.. code-block:: python
+
+   def cfun(x):
+       return int(x) if x else -1
+
+
+   pd.read_excel("path_to_file.xls", "Sheet1", converters={"MyInts": cfun})
+
+Dtype specifications
+++++++++++++++++++++
+
+As an alternative to converters, the type for an entire column can
+be specified using the ``dtype`` keyword, which takes a dictionary
+mapping column names to types.  To interpret data with
+no type inference, use the type ``str`` or ``object``.
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xls", dtype={"MyInts": "int64", "MyText": str})
+
+.. _io.excel_writer:
+
+Writing Excel files
+'''''''''''''''''''
+
+Writing Excel files to disk
++++++++++++++++++++++++++++
+
+To write a ``DataFrame`` object to a sheet of an Excel file, you can use the
+``to_excel`` instance method.  The arguments are largely the same as ``to_csv``
+described above, the first argument being the name of the excel file, and the
+optional second argument the name of the sheet to which the ``DataFrame`` should be
+written. For example:
+
+.. code-block:: python
+
+   df.to_excel("path_to_file.xlsx", sheet_name="Sheet1")
+
+Files with a
+``.xlsx`` extension will be written using ``xlsxwriter`` (if available) or
+``openpyxl``.
+
+The ``DataFrame`` will be written in a way that tries to mimic the REPL output.
+The ``index_label`` will be placed in the second
+row instead of the first. You can place it in the first row by setting the
+``merge_cells`` option in ``to_excel()`` to ``False``:
+
+.. code-block:: python
+
+   df.to_excel("path_to_file.xlsx", index_label="label", merge_cells=False)
+
+In order to write separate ``DataFrames`` to separate sheets in a single Excel file,
+one can pass an :class:`~pandas.io.excel.ExcelWriter`.
+
+.. code-block:: python
+
+   with pd.ExcelWriter("path_to_file.xlsx") as writer:
+       df1.to_excel(writer, sheet_name="Sheet1")
+       df2.to_excel(writer, sheet_name="Sheet2")
+
+.. _io.excel_writing_buffer:
+
+When using the ``engine_kwargs`` parameter, pandas will pass these arguments to the
+engine. For this, it is important to know which function pandas is using internally.
+
+* For the engine openpyxl, pandas is using :func:`openpyxl.Workbook` to create a new sheet and :func:`openpyxl.load_workbook` to append data to an existing sheet. The openpyxl engine writes to (``.xlsx``) and (``.xlsm``) files.
+
+* For the engine xlsxwriter, pandas is using :func:`xlsxwriter.Workbook` to write to (``.xlsx``) files.
+
+* For the engine odf, pandas is using :func:`odf.opendocument.OpenDocumentSpreadsheet` to write to (``.ods``) files.
+
+Writing Excel files to memory
++++++++++++++++++++++++++++++
+
+pandas supports writing Excel files to buffer-like objects such as ``StringIO`` or
+``BytesIO`` using :class:`~pandas.io.excel.ExcelWriter`.
+
+.. code-block:: python
+
+   from io import BytesIO
+
+   bio = BytesIO()
+
+   # By setting the 'engine' in the ExcelWriter constructor.
+   writer = pd.ExcelWriter(bio, engine="xlsxwriter")
+   df.to_excel(writer, sheet_name="Sheet1")
+
+   # Save the workbook
+   writer.save()
+
+   # Seek to the beginning and read to copy the workbook to a variable in memory
+   bio.seek(0)
+   workbook = bio.read()
+
+.. note::
+
+    ``engine`` is optional but recommended.  Setting the engine determines
+    the version of workbook produced. Setting ``engine='xlrd'`` will produce an
+    Excel 2003-format workbook (xls).  Using either ``'openpyxl'`` or
+    ``'xlsxwriter'`` will produce an Excel 2007-format workbook (xlsx). If
+    omitted, an Excel 2007-formatted workbook is produced.
+
+
+.. _io.excel.writers:
+
+Excel writer engines
+''''''''''''''''''''
+
+pandas chooses an Excel writer via two methods:
+
+1. the ``engine`` keyword argument
+2. the filename extension (via the default specified in config options)
+
+By default, pandas uses the `XlsxWriter`_  for ``.xlsx``, `openpyxl`_
+for ``.xlsm``. If you have multiple
+engines installed, you can set the default engine through :ref:`setting the
+config options <options>` ``io.excel.xlsx.writer`` and
+``io.excel.xls.writer``. pandas will fall back on `openpyxl`_ for ``.xlsx``
+files if `Xlsxwriter`_ is not available.
+
+.. _XlsxWriter: https://xlsxwriter.readthedocs.io
+.. _openpyxl: https://openpyxl.readthedocs.io/
+
+To specify which writer you want to use, you can pass an engine keyword
+argument to ``to_excel`` and to ``ExcelWriter``. The built-in engines are:
+
+* ``openpyxl``: version 2.4 or higher is required
+* ``xlsxwriter``
+
+.. code-block:: python
+
+   # By setting the 'engine' in the DataFrame 'to_excel()' methods.
+   df.to_excel("path_to_file.xlsx", sheet_name="Sheet1", engine="xlsxwriter")
+
+   # By setting the 'engine' in the ExcelWriter constructor.
+   writer = pd.ExcelWriter("path_to_file.xlsx", engine="xlsxwriter")
+
+   # Or via pandas configuration.
+   from pandas import options  # noqa: E402
+
+   options.io.excel.xlsx.writer = "xlsxwriter"
+
+   df.to_excel("path_to_file.xlsx", sheet_name="Sheet1")
+
+.. _io.excel.style:
+
+Style and formatting
+''''''''''''''''''''
+
+The look and feel of Excel worksheets created from pandas can be modified using the following parameters on the ``DataFrame``'s ``to_excel`` method.
+
+* ``float_format`` : Format string for floating point numbers (default ``None``).
+* ``freeze_panes`` : A tuple of two integers representing the bottommost row and rightmost column to freeze. Each of these parameters is one-based, so (1, 1) will freeze the first row and first column (default ``None``).
+
+.. note::
+
+    As of pandas 3.0, by default spreadsheets created with the ``to_excel`` method
+    will not contain any styling. Users wishing to bold text, add bordered styles,
+    etc in a worksheet output by ``to_excel`` can do so by using :meth:`Styler.to_excel`
+    to create styled excel files. For documentation on styling spreadsheets, see
+    `here <https://pandas.pydata.org/docs/user_guide/style.html#Export-to-Excel>`__.
+
+
+.. code-block:: python
+
+    css = "border: 1px solid black; font-weight: bold;"
+    df.style.map_index(lambda x: css).map_index(lambda x: css, axis=1).to_excel("myfile.xlsx")
+
+Using the `Xlsxwriter`_ engine provides many options for controlling the
+format of an Excel worksheet created with the ``to_excel`` method.  Excellent examples can be found in the
+`Xlsxwriter`_ documentation here: https://xlsxwriter.readthedocs.io/working_with_pandas.html
+
+.. _io.ods:
+
+OpenDocument Spreadsheets
+-------------------------
+
+The io methods for `Excel files`_ also support reading and writing OpenDocument spreadsheets
+using the `odfpy <https://pypi.org/project/odfpy/>`__ module. The semantics and features for reading and writing
+OpenDocument spreadsheets match what can be done for `Excel files`_ using
+``engine='odf'``. The optional dependency 'odfpy' needs to be installed.
+
+The :func:`~pandas.read_excel` method can read OpenDocument spreadsheets
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.ods", engine="odf")
+
+Similarly, the :func:`~pandas.to_excel` method can write OpenDocument spreadsheets
+
+.. code-block:: python
+
+   # Writes DataFrame to a .ods file
+   df.to_excel("path_to_file.ods", engine="odf")
+
+.. _io.xlsb:
+
+Binary Excel (.xlsb) files
+--------------------------
+
+The :func:`~pandas.read_excel` method can also read binary Excel files
+using the ``pyxlsb`` module. The semantics and features for reading
+binary Excel files mostly match what can be done for `Excel files`_ using
+``engine='pyxlsb'``. ``pyxlsb`` does not recognize datetime types
+in files and will return floats instead (you can use :ref:`calamine<io.calamine>`
+if you need recognize datetime types).
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xlsb", engine="pyxlsb")
+
+.. note::
+
+   Currently pandas only supports *reading* binary Excel files. Writing
+   is not implemented.
+
+.. _io.calamine:
+
+Calamine (Excel and ODS files)
+------------------------------
+
+The :func:`~pandas.read_excel` method can read Excel file (``.xlsx``, ``.xlsm``, ``.xls``, ``.xlsb``)
+and OpenDocument spreadsheets (``.ods``) using the ``python-calamine`` module.
+This module is a binding for Rust library `calamine <https://crates.io/crates/calamine>`__
+and is faster than other engines in most cases. The optional dependency 'python-calamine' needs to be installed.
+
+.. code-block:: python
+
+   # Returns a DataFrame
+   pd.read_excel("path_to_file.xlsb", engine="calamine")
+
+.. _io.clipboard:
+
+Clipboard
+---------
+
+A handy way to grab data is to use the :meth:`~DataFrame.read_clipboard` method,
+which takes the contents of the clipboard buffer and passes them to the
+``read_csv`` method. For instance, you can copy the following text to the
+clipboard (CTRL-C on many operating systems):
+
+.. code-block:: console
+
+     A B C
+   x 1 4 p
+   y 2 5 q
+   z 3 6 r
+
+And then import the data directly to a ``DataFrame`` by calling:
+
+.. code-block:: python
+
+    >>> clipdf = pd.read_clipboard()
+    >>> clipdf
+      A B C
+    x 1 4 p
+    y 2 5 q
+    z 3 6 r
+
+The ``to_clipboard`` method can be used to write the contents of a ``DataFrame`` to
+the clipboard. Following which you can paste the clipboard contents into other
+applications (CTRL-V on many operating systems). Here we illustrate writing a
+``DataFrame`` into clipboard and reading it back.
+
+.. code-block:: python
+
+    >>> df = pd.DataFrame(
+    ...     {"A": [1, 2, 3], "B": [4, 5, 6], "C": ["p", "q", "r"]}, index=["x", "y", "z"]
+    ... )
+
+    >>> df
+      A B C
+    x 1 4 p
+    y 2 5 q
+    z 3 6 r
+    >>> df.to_clipboard()
+    >>> pd.read_clipboard()
+      A B C
+    x 1 4 p
+    y 2 5 q
+    z 3 6 r
+
+We can see that we got the same content back, which we had earlier written to the clipboard.
+
+.. note::
+
+   You may need to install xclip or xsel (with PyQt5, PyQt4 or qtpy) on Linux to use these methods.
+
+.. _io.pickle:
+
+Pickling
+--------
+
+All pandas objects are equipped with ``to_pickle`` methods which use Python's
+``cPickle`` module to save data structures to disk using the pickle format.
+
+.. ipython:: python
+
+   df
+   df.to_pickle("foo.pkl")
+
+The ``read_pickle`` function in the ``pandas`` namespace can be used to load
+any pickled pandas object (or any other pickled object) from file:
+
+
+.. ipython:: python
+
+   pd.read_pickle("foo.pkl")
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("foo.pkl")
+
+.. warning::
+
+   Loading pickled data received from untrusted sources can be unsafe.
+
+   See: https://docs.python.org/3/library/pickle.html
+
+.. warning::
+
+   :func:`read_pickle` is only guaranteed backwards compatible back to a few minor release.
+
+.. _io.pickle.compression:
+
+Compressed pickle files
+'''''''''''''''''''''''
+
+:func:`read_pickle`, :meth:`DataFrame.to_pickle` and :meth:`Series.to_pickle` can read
+and write compressed pickle files. The compression types of ``gzip``, ``bz2``, ``xz``, ``zstd`` are supported for reading and writing.
+The ``zip`` file format only supports reading and must contain only one data file
+to be read.
+
+The compression type can be an explicit parameter or be inferred from the file extension.
+If 'infer', then use ``gzip``, ``bz2``, ``zip``, ``xz``, ``zstd`` if filename ends in ``'.gz'``, ``'.bz2'``, ``'.zip'``,
+``'.xz'``, or ``'.zst'``, respectively.
+
+The compression parameter can also be a ``dict`` in order to pass options to the
+compression protocol. It must have a ``'method'`` key set to the name
+of the compression protocol, which must be one of
+{``'zip'``, ``'gzip'``, ``'bz2'``, ``'xz'``, ``'zstd'``}. All other key-value pairs are passed to
+the underlying compression library.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": np.random.randn(1000),
+           "B": "foo",
+           "C": pd.date_range("20130101", periods=1000, freq="s"),
+       }
+   )
+   df
+
+Using an explicit compression type:
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.compress", compression="gzip")
+   rt = pd.read_pickle("data.pkl.compress", compression="gzip")
+   rt
+
+Inferring compression type from the extension:
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.xz", compression="infer")
+   rt = pd.read_pickle("data.pkl.xz", compression="infer")
+   rt
+
+The default is to 'infer':
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.gz")
+   rt = pd.read_pickle("data.pkl.gz")
+   rt
+
+   df["A"].to_pickle("s1.pkl.bz2")
+   rt = pd.read_pickle("s1.pkl.bz2")
+   rt
+
+Passing options to the compression protocol in order to speed up compression:
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.gz", compression={"method": "gzip", "compresslevel": 1})
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("data.pkl.compress")
+   os.remove("data.pkl.xz")
+   os.remove("data.pkl.gz")
+   os.remove("s1.pkl.bz2")
+
+.. _io.msgpack:
+
+msgpack
+-------
+
+pandas support for ``msgpack`` has been removed in version 1.0.0. It is
+recommended to use :ref:`pickle <io.pickle>` instead.
+
+Alternatively, you can also the Arrow IPC serialization format for on-the-wire
+transmission of pandas objects. For documentation on pyarrow, see
+`here <https://arrow.apache.org/docs/python/ipc.html>`__.
+
+
+.. _io.hdf5:
+
+HDF5 (PyTables)
+---------------
+
+``HDFStore`` is a dict-like object which reads and writes pandas using
+the high performance HDF5 format using the excellent `PyTables
+<https://www.pytables.org/>`__ library. See the :ref:`cookbook <cookbook.hdf>`
+for some advanced strategies
+
+.. warning::
+
+   pandas uses PyTables for reading and writing HDF5 files, which allows
+   serializing object-dtype data with pickle. Loading pickled data received from
+   untrusted sources can be unsafe.
+
+   See: https://docs.python.org/3/library/pickle.html for more.
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   os.remove("store.h5")
+
+.. ipython:: python
+
+   store = pd.HDFStore("store.h5")
+   print(store)
+
+Objects can be written to the file just like adding key-value pairs to a
+dict:
+
+.. ipython:: python
+
+   index = pd.date_range("1/1/2000", periods=8)
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   df = pd.DataFrame(np.random.randn(8, 3), index=index, columns=["A", "B", "C"])
+
+   # store.put('s', s) is an equivalent method
+   store["s"] = s
+
+   store["df"] = df
+
+   store
+
+In a current or later Python session, you can retrieve stored objects:
+
+.. ipython:: python
+
+   # store.get('df') is an equivalent method
+   store["df"]
+
+   # dotted (attribute) access provides get as well
+   store.df
+
+Deletion of the object specified by the key:
+
+.. ipython:: python
+
+   # store.remove('df') is an equivalent method
+   del store["df"]
+
+   store
+
+Closing a Store and using a context manager:
+
+.. ipython:: python
+
+   store.close()
+   store
+   store.is_open
+
+   # Working with, and automatically closing the store using a context manager
+   with pd.HDFStore("store.h5") as store:
+       store.keys()
+
+.. ipython:: python
+   :suppress:
+
+   store.close()
+   os.remove("store.h5")
+
+
+
+Read/write API
+''''''''''''''
+
+``HDFStore`` supports a top-level API using  ``read_hdf`` for reading and ``to_hdf`` for writing,
+similar to how ``read_csv`` and ``to_csv`` work.
+
+.. ipython:: python
+
+   df_tl = pd.DataFrame({"A": list(range(5)), "B": list(range(5))})
+   df_tl.to_hdf("store_tl.h5", key="table", append=True)
+   pd.read_hdf("store_tl.h5", "table", where=["index>2"])
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   os.remove("store_tl.h5")
+
+
+HDFStore will by default not drop rows that are all missing. This behavior can be changed by setting ``dropna=True``.
+
+
+.. ipython:: python
+
+   df_with_missing = pd.DataFrame(
+       {
+           "col1": [0, np.nan, 2],
+           "col2": [1, np.nan, np.nan],
+       }
+   )
+   df_with_missing
+
+   df_with_missing.to_hdf("file.h5", key="df_with_missing", format="table", mode="w")
+
+   pd.read_hdf("file.h5", "df_with_missing")
+
+   df_with_missing.to_hdf(
+       "file.h5", key="df_with_missing", format="table", mode="w", dropna=True
+   )
+   pd.read_hdf("file.h5", "df_with_missing")
+
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("file.h5")
+
+
+.. _io.hdf5-fixed:
+
+Fixed format
+''''''''''''
+
+The examples above show storing using ``put``, which write the HDF5 to ``PyTables`` in a fixed array format, called
+the ``fixed`` format. These types of stores are **not** appendable once written (though you can simply
+remove them and rewrite). Nor are they **queryable**; they must be
+retrieved in their entirety. They also do not support dataframes with non-unique column names.
+The ``fixed`` format stores offer very fast writing and slightly faster reading than ``table`` stores.
+This format is specified by default when using ``put`` or ``to_hdf`` or by ``format='fixed'`` or ``format='f'``.
+
+.. warning::
+
+   A ``fixed`` format will raise a ``TypeError`` if you try to retrieve using a ``where``:
+
+   .. ipython:: python
+      :okexcept:
+
+      pd.DataFrame(np.random.randn(10, 2)).to_hdf("test_fixed.h5", key="df")
+      pd.read_hdf("test_fixed.h5", "df", where="index>5")
+
+   .. ipython:: python
+      :suppress:
+
+      os.remove("test_fixed.h5")
+
+
+.. _io.hdf5-table:
+
+Table format
+''''''''''''
+
+``HDFStore`` supports another ``PyTables`` format on disk, the ``table``
+format. Conceptually a ``table`` is shaped very much like a DataFrame,
+with rows and columns. A ``table`` may be appended to in the same or
+other sessions.  In addition, delete and query type operations are
+supported. This format is specified by ``format='table'`` or ``format='t'``
+to ``append`` or ``put`` or ``to_hdf``.
+
+This format can be set as an option as well ``pd.set_option('io.hdf.default_format','table')`` to
+enable ``put/append/to_hdf`` to by default store in the ``table`` format.
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   os.remove("store.h5")
+
+.. ipython:: python
+
+   store = pd.HDFStore("store.h5")
+   df1 = df[0:4]
+   df2 = df[4:]
+
+   # append data (creates a table automatically)
+   store.append("df", df1)
+   store.append("df", df2)
+   store
+
+   # select the entire object
+   store.select("df")
+
+   # the type of stored data
+   store.root.df._v_attrs.pandas_type
+
+.. note::
+
+   You can also create a ``table`` by passing ``format='table'`` or ``format='t'`` to a ``put`` operation.
+
+.. _io.hdf5-keys:
+
+Hierarchical keys
+'''''''''''''''''
+
+Keys to a store can be specified as a string. These can be in a
+hierarchical path-name like format (e.g. ``foo/bar/bah``), which will
+generate a hierarchy of sub-stores (or ``Groups`` in PyTables
+parlance). Keys can be specified without the leading '/' and are **always**
+absolute (e.g. 'foo' refers to '/foo'). Removal operations can remove
+everything in the sub-store and **below**, so be *careful*.
+
+.. ipython:: python
+
+   store.put("foo/bar/bah", df)
+   store.append("food/orange", df)
+   store.append("food/apple", df)
+   store
+
+   # a list of keys are returned
+   store.keys()
+
+   # remove all nodes under this level
+   store.remove("food")
+   store
+
+
+You can walk through the group hierarchy using the ``walk`` method which
+will yield a tuple for each group key along with the relative keys of its contents.
+
+.. ipython:: python
+
+   for (path, subgroups, subkeys) in store.walk():
+       for subgroup in subgroups:
+           print("GROUP: {}/{}".format(path, subgroup))
+       for subkey in subkeys:
+           key = "/".join([path, subkey])
+           print("KEY: {}".format(key))
+           print(store.get(key))
+
+
+
+.. warning::
+
+    Hierarchical keys cannot be retrieved as dotted (attribute) access as described above for items stored under the root node.
+
+    .. ipython:: python
+       :okexcept:
+
+       store.foo.bar.bah
+
+    .. ipython:: python
+
+       # you can directly access the actual PyTables node but using the root node
+       store.root.foo.bar.bah
+
+    Instead, use explicit string based keys:
+
+    .. ipython:: python
+
+       store["foo/bar/bah"]
+
+
+.. _io.hdf5-types:
+
+Storing types
+'''''''''''''
+
+Storing mixed types in a table
+++++++++++++++++++++++++++++++
+
+Storing mixed-dtype data is supported. Strings are stored as a
+fixed-width using the maximum size of the appended column. Subsequent attempts
+at appending longer strings will raise a ``ValueError``.
+
+Passing ``min_itemsize={`values`: size}`` as a parameter to append
+will set a larger minimum for the string columns. Storing ``floats,
+strings, ints, bools, datetime64`` are currently supported. For string
+columns, passing ``nan_rep = 'nan'`` to append will change the default
+nan representation on disk (which converts to/from ``np.nan``), this
+defaults to ``nan``.
+
+.. ipython:: python
+
+    df_mixed = pd.DataFrame(
+        {
+            "A": np.random.randn(8),
+            "B": np.random.randn(8),
+            "C": np.array(np.random.randn(8), dtype="float32"),
+            "string": "string",
+            "int": 1,
+            "bool": True,
+            "datetime64": pd.Timestamp("20010102"),
+        },
+        index=list(range(8)),
+    )
+    df_mixed.loc[df_mixed.index[3:5], ["A", "B", "string", "datetime64"]] = np.nan
+
+    store.append("df_mixed", df_mixed, min_itemsize={"values": 50})
+    df_mixed1 = store.select("df_mixed")
+    df_mixed1
+    df_mixed1.dtypes.value_counts()
+
+    # we have provided a minimum string column size
+    store.root.df_mixed.table
+
+Storing MultiIndex DataFrames
++++++++++++++++++++++++++++++
+
+Storing MultiIndex ``DataFrames`` as tables is very similar to
+storing/selecting from homogeneous index ``DataFrames``.
+
+.. ipython:: python
+
+   index = pd.MultiIndex(
+      levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],
+      codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+      names=["foo", "bar"],
+   )
+   df_mi = pd.DataFrame(np.random.randn(10, 3), index=index, columns=["A", "B", "C"])
+   df_mi
+
+   store.append("df_mi", df_mi)
+   store.select("df_mi")
+
+   # the levels are automatically included as data columns
+   store.select("df_mi", "foo=bar")
+
+.. note::
+   The ``index`` keyword is reserved and cannot be use as a level name.
+
+.. _io.hdf5-query:
+
+Querying
+''''''''
+
+Querying a table
+++++++++++++++++
+
+``select`` and ``delete`` operations have an optional criterion that can
+be specified to select/delete only a subset of the data. This allows one
+to have a very large on-disk table and retrieve only a portion of the
+data.
+
+A query is specified using the ``Term`` class under the hood, as a boolean expression.
+
+* ``index`` and ``columns`` are supported indexers of ``DataFrames``.
+* if ``data_columns`` are specified, these can be used as additional indexers.
+* level name in a MultiIndex, with default name  ``level_0``, ``level_1``, … if not provided.
+
+Valid comparison operators are:
+
+``=, ==, !=, >, >=, <, <=``
+
+Valid boolean expressions are combined with:
+
+* ``|`` : or
+* ``&`` : and
+* ``(`` and ``)`` : for grouping
+
+These rules are similar to how boolean expressions are used in pandas for indexing.
+
+.. note::
+
+   - ``=`` will be automatically expanded to the comparison operator ``==``
+   - ``~`` is the not operator, but can only be used in very limited
+     circumstances
+   - If a list/tuple of expressions is passed they will be combined via ``&``
+
+The following are valid expressions:
+
+* ``'index >= date'``
+* ``"columns = ['A', 'D']"``
+* ``"columns in ['A', 'D']"``
+* ``'columns = A'``
+* ``'columns == A'``
+* ``"~(columns = ['A', 'B'])"``
+* ``'index > df.index[3] & string = "bar"'``
+* ``'(index > df.index[3] & index <= df.index[6]) | string = "bar"'``
+* ``"ts >= Timestamp('2012-02-01')"``
+* ``"major_axis>=20130101"``
+
+The ``indexers`` are on the left-hand side of the sub-expression:
+
+``columns``, ``major_axis``, ``ts``
+
+The right-hand side of the sub-expression (after a comparison operator) can be:
+
+* functions that will be evaluated, e.g. ``Timestamp('2012-02-01')``
+* strings, e.g. ``"bar"``
+* date-like, e.g. ``20130101``, or ``"20130101"``
+* lists, e.g. ``"['A', 'B']"``
+* variables that are defined in the local names space, e.g. ``date``
+
+.. note::
+
+   Passing a string to a query by interpolating it into the query
+   expression is not recommended. Simply assign the string of interest to a
+   variable and use that variable in an expression. For example, do this
+
+   .. code-block:: python
+
+      string = "HolyMoly'"
+      store.select("df", "index == string")
+
+   instead of this
+
+   .. code-block:: python
+
+      string = "HolyMoly'"
+      store.select('df', f'index == {string}')
+
+   The latter will **not** work and will raise a ``SyntaxError``.Note that
+   there's a single quote followed by a double quote in the ``string``
+   variable.
+
+   If you *must* interpolate, use the ``'%r'`` format specifier
+
+   .. code-block:: python
+
+      store.select("df", "index == %r" % string)
+
+   which will quote ``string``.
+
+
+Here are some examples:
+
+.. ipython:: python
+
+    dfq = pd.DataFrame(
+        np.random.randn(10, 4),
+        columns=list("ABCD"),
+        index=pd.date_range("20130101", periods=10),
+    )
+    store.append("dfq", dfq, format="table", data_columns=True)
+
+Use boolean expressions, with in-line function evaluation.
+
+.. ipython:: python
+
+    store.select("dfq", "index>pd.Timestamp('20130104') & columns=['A', 'B']")
+
+Use inline column reference.
+
+.. ipython:: python
+
+   store.select("dfq", where="A>0 or C>0")
+
+The ``columns`` keyword can be supplied to select a list of columns to be
+returned, this is equivalent to passing a
+``'columns=list_of_columns_to_filter'``:
+
+.. ipython:: python
+
+   store.select("df", "columns=['A', 'B']")
+
+``start`` and ``stop`` parameters can be specified to limit the total search
+space. These are in terms of the total number of rows in a table.
+
+.. note::
+
+   ``select`` will raise a ``ValueError`` if the query expression has an unknown
+   variable reference. Usually this means that you are trying to select on a column
+   that is **not** a data_column.
+
+   ``select`` will raise a ``SyntaxError`` if the query expression is not valid.
+
+
+.. _io.hdf5-timedelta:
+
+Query timedelta64[ns]
++++++++++++++++++++++
+
+You can store and query using the ``timedelta64[ns]`` type. Terms can be
+specified in the format: ``<float>(<unit>)``, where float may be signed (and fractional), and unit can be
+``D,s,ms,us,ns`` for the timedelta. Here's an example:
+
+.. ipython:: python
+
+   from datetime import timedelta
+
+   dftd = pd.DataFrame(
+       {
+           "A": pd.Timestamp("20130101"),
+           "B": [
+               pd.Timestamp("20130101") + timedelta(days=i, seconds=10)
+               for i in range(10)
+           ],
+       }
+   )
+   dftd["C"] = dftd["A"] - dftd["B"]
+   dftd
+   store.append("dftd", dftd, data_columns=True)
+   store.select("dftd", "C<'-3.5D'")
+
+.. _io.query_multi:
+
+Query MultiIndex
+++++++++++++++++
+
+Selecting from a ``MultiIndex`` can be achieved by using the name of the level.
+
+.. ipython:: python
+
+   df_mi.index.names
+   store.select("df_mi", "foo=baz and bar=two")
+
+If the ``MultiIndex`` levels names are ``None``, the levels are automatically made available via
+the ``level_n`` keyword with ``n`` the level of the ``MultiIndex`` you want to select from.
+
+.. ipython:: python
+
+   index = pd.MultiIndex(
+       levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],
+       codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+   )
+   df_mi_2 = pd.DataFrame(np.random.randn(10, 3), index=index, columns=["A", "B", "C"])
+   df_mi_2
+
+   store.append("df_mi_2", df_mi_2)
+
+   # the levels are automatically included as data columns with keyword level_n
+   store.select("df_mi_2", "level_0=foo and level_1=two")
+
+
+Indexing
+++++++++
+
+You can create/modify an index for a table with ``create_table_index``
+after data is already in the table (after and ``append/put``
+operation). Creating a table index is **highly** encouraged. This will
+speed your queries a great deal when you use a ``select`` with the
+indexed dimension as the ``where``.
+
+.. note::
+
+   Indexes are automagically created on the indexables
+   and any data columns you specify. This behavior can be turned off by passing
+   ``index=False`` to ``append``.
+
+.. ipython:: python
+
+   # we have automagically already created an index (in the first section)
+   i = store.root.df.table.cols.index.index
+   i.optlevel, i.kind
+
+   # change an index by passing new parameters
+   store.create_table_index("df", optlevel=9, kind="full")
+   i = store.root.df.table.cols.index.index
+   i.optlevel, i.kind
+
+Oftentimes when appending large amounts of data to a store, it is useful to turn off index creation for each append, then recreate at the end.
+
+.. ipython:: python
+
+   df_1 = pd.DataFrame(np.random.randn(10, 2), columns=list("AB"))
+   df_2 = pd.DataFrame(np.random.randn(10, 2), columns=list("AB"))
+
+   st = pd.HDFStore("appends.h5", mode="w")
+   st.append("df", df_1, data_columns=["B"], index=False)
+   st.append("df", df_2, data_columns=["B"], index=False)
+   st.get_storer("df").table
+
+Then create the index when finished appending.
+
+.. ipython:: python
+
+   st.create_table_index("df", columns=["B"], optlevel=9, kind="full")
+   st.get_storer("df").table
+
+   st.close()
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   os.remove("appends.h5")
+
+See `here <https://stackoverflow.com/questions/17893370/ptrepack-sortby-needs-full-index>`__ for how to create a completely-sorted-index (CSI) on an existing store.
+
+.. _io.hdf5-query-data-columns:
+
+Query via data columns
+++++++++++++++++++++++
+
+You can designate (and index) certain columns that you want to be able
+to perform queries (other than the ``indexable`` columns, which you can
+always query). For instance say you want to perform this common
+operation, on-disk, and return just the frame that matches this
+query. You can specify ``data_columns = True`` to force all columns to
+be ``data_columns``.
+
+.. ipython:: python
+
+   df_dc = df.copy()
+   df_dc["string"] = "foo"
+   df_dc.loc[df_dc.index[4:6], "string"] = np.nan
+   df_dc.loc[df_dc.index[7:9], "string"] = "bar"
+   df_dc["string2"] = "cool"
+   df_dc.loc[df_dc.index[1:3], ["B", "C"]] = 1.0
+   df_dc
+
+   # on-disk operations
+   store.append("df_dc", df_dc, data_columns=["B", "C", "string", "string2"])
+   store.select("df_dc", where="B > 0")
+
+   # getting creative
+   store.select("df_dc", "B > 0 & C > 0 & string == foo")
+
+   # this is in-memory version of this type of selection
+   df_dc[(df_dc.B > 0) & (df_dc.C > 0) & (df_dc.string == "foo")]
+
+   # we have automagically created this index and the B/C/string/string2
+   # columns are stored separately as ``PyTables`` columns
+   store.root.df_dc.table
+
+There is some performance degradation by making lots of columns into
+``data columns``, so it is up to the user to designate these. In addition,
+you cannot change data columns (nor indexables) after the first
+append/put operation (Of course you can simply read in the data and
+create a new table!).
+
+Iterator
+++++++++
+
+You can pass ``iterator=True`` or ``chunksize=number_in_a_chunk``
+to ``select`` and ``select_as_multiple`` to return an iterator on the results.
+The default is 50,000 rows returned in a chunk.
+
+.. ipython:: python
+
+   for df in store.select("df", chunksize=3):
+       print(df)
+
+.. note::
+
+   You can also use the iterator with ``read_hdf`` which will open, then
+   automatically close the store when finished iterating.
+
+   .. code-block:: python
+
+      for df in pd.read_hdf("store.h5", "df", chunksize=3):
+          print(df)
+
+Note, that the chunksize keyword applies to the **source** rows. So if you
+are doing a query, then the chunksize will subdivide the total rows in the table
+and the query applied, returning an iterator on potentially unequal sized chunks.
+
+Here is a recipe for generating a query and using it to create equal sized return
+chunks.
+
+.. ipython:: python
+
+   dfeq = pd.DataFrame({"number": np.arange(1, 11)})
+   dfeq
+
+   store.append("dfeq", dfeq, data_columns=["number"])
+
+   def chunks(l, n):
+       return [l[i: i + n] for i in range(0, len(l), n)]
+
+   evens = [2, 4, 6, 8, 10]
+   coordinates = store.select_as_coordinates("dfeq", "number=evens")
+   for c in chunks(coordinates, 2):
+       print(store.select("dfeq", where=c))
+
+Advanced queries
+++++++++++++++++
+
+Select a single column
+^^^^^^^^^^^^^^^^^^^^^^
+
+To retrieve a single indexable or data column, use the
+method ``select_column``. This will, for example, enable you to get the index
+very quickly. These return a ``Series`` of the result, indexed by the row number.
+These do not currently accept the ``where`` selector.
+
+.. ipython:: python
+
+   store.select_column("df_dc", "index")
+   store.select_column("df_dc", "string")
+
+.. _io.hdf5-selecting_coordinates:
+
+Selecting coordinates
+^^^^^^^^^^^^^^^^^^^^^
+
+Sometimes you want to get the coordinates (a.k.a the index locations) of your query. This returns an
+``Index`` of the resulting locations. These coordinates can also be passed to subsequent
+``where`` operations.
+
+.. ipython:: python
+
+   df_coord = pd.DataFrame(
+       np.random.randn(1000, 2), index=pd.date_range("20000101", periods=1000)
+   )
+   store.append("df_coord", df_coord)
+   c = store.select_as_coordinates("df_coord", "index > 20020101")
+   c
+   store.select("df_coord", where=c)
+
+.. _io.hdf5-where_mask:
+
+Selecting using a where mask
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Sometime your query can involve creating a list of rows to select. Usually this ``mask`` would
+be a resulting ``index`` from an indexing operation. This example selects the months of
+a datetimeindex which are 5.
+
+.. ipython:: python
+
+   df_mask = pd.DataFrame(
+       np.random.randn(1000, 2), index=pd.date_range("20000101", periods=1000)
+   )
+   store.append("df_mask", df_mask)
+   c = store.select_column("df_mask", "index")
+   where = c[pd.DatetimeIndex(c).month == 5].index
+   store.select("df_mask", where=where)
+
+Storer object
+^^^^^^^^^^^^^
+
+If you want to inspect the stored object, retrieve via
+``get_storer``. You could use this programmatically to say get the number
+of rows in an object.
+
+.. ipython:: python
+
+   store.get_storer("df_dc").nrows
+
+
+Multiple table queries
+++++++++++++++++++++++
+
+The methods ``append_to_multiple`` and
+``select_as_multiple`` can perform appending/selecting from
+multiple tables at once. The idea is to have one table (call it the
+selector table) that you index most/all of the columns, and perform your
+queries. The other table(s) are data tables with an index matching the
+selector table's index. You can then perform a very fast query
+on the selector table, yet get lots of data back. This method is similar to
+having a very wide table, but enables more efficient queries.
+
+The ``append_to_multiple`` method splits a given single DataFrame
+into multiple tables according to ``d``, a dictionary that maps the
+table names to a list of 'columns' you want in that table. If ``None``
+is used in place of a list, that table will have the remaining
+unspecified columns of the given DataFrame. The argument ``selector``
+defines which table is the selector table (which you can make queries from).
+The argument ``dropna`` will drop rows from the input ``DataFrame`` to ensure
+tables are synchronized.  This means that if a row for one of the tables
+being written to is entirely ``np.nan``, that row will be dropped from all tables.
+
+If ``dropna`` is False, **THE USER IS RESPONSIBLE FOR SYNCHRONIZING THE TABLES**.
+Remember that entirely ``np.Nan`` rows are not written to the HDFStore, so if
+you choose to call ``dropna=False``, some tables may have more rows than others,
+and therefore ``select_as_multiple`` may not work or it may return unexpected
+results.
+
+.. ipython:: python
+
+   df_mt = pd.DataFrame(
+       np.random.randn(8, 6),
+       index=pd.date_range("1/1/2000", periods=8),
+       columns=["A", "B", "C", "D", "E", "F"],
+   )
+   df_mt["foo"] = "bar"
+   df_mt.loc[df_mt.index[1], ("A", "B")] = np.nan
+
+   # you can also create the tables individually
+   store.append_to_multiple(
+       {"df1_mt": ["A", "B"], "df2_mt": None}, df_mt, selector="df1_mt"
+   )
+   store
+
+   # individual tables were created
+   store.select("df1_mt")
+   store.select("df2_mt")
+
+   # as a multiple
+   store.select_as_multiple(
+       ["df1_mt", "df2_mt"],
+       where=["A>0", "B>0"],
+       selector="df1_mt",
+   )
+
+
+Delete from a table
+'''''''''''''''''''
+
+You can delete from a table selectively by specifying a ``where``. In
+deleting rows, it is important to understand the ``PyTables`` deletes
+rows by erasing the rows, then **moving** the following data. Thus
+deleting can potentially be a very expensive operation depending on the
+orientation of your data. To get optimal performance, it's
+worthwhile to have the dimension you are deleting be the first of the
+``indexables``.
+
+Data is ordered (on the disk) in terms of the ``indexables``. Here's a
+simple use case. You store panel-type data, with dates in the
+``major_axis`` and ids in the ``minor_axis``. The data is then
+interleaved like this:
+
+* date_1
+    * id_1
+    * id_2
+    *  .
+    * id_n
+* date_2
+    * id_1
+    *  .
+    * id_n
+
+It should be clear that a delete operation on the ``major_axis`` will be
+fairly quick, as one chunk is removed, then the following data moved. On
+the other hand a delete operation on the ``minor_axis`` will be very
+expensive. In this case it would almost certainly be faster to rewrite
+the table using a ``where`` that selects all but the missing data.
+
+.. warning::
+
+   Please note that HDF5 **DOES NOT RECLAIM SPACE** in the h5 files
+   automatically. Thus, repeatedly deleting (or removing nodes) and adding
+   again, **WILL TEND TO INCREASE THE FILE SIZE**.
+
+   To *repack and clean* the file, use :ref:`ptrepack <io.hdf5-ptrepack>`.
+
+.. _io.hdf5-notes:
+
+Notes & caveats
+'''''''''''''''
+
+
+Compression
++++++++++++
+
+``PyTables`` allows the stored data to be compressed. This applies to
+all kinds of stores, not just tables. Two parameters are used to
+control compression: ``complevel`` and ``complib``.
+
+* ``complevel`` specifies if and how hard data is to be compressed.
+  ``complevel=0`` and ``complevel=None`` disables compression and
+  ``0<complevel<10`` enables compression.
+
+* ``complib`` specifies which compression library to use.
+  If nothing is  specified the default library ``zlib`` is used. A
+  compression library usually optimizes for either good compression rates
+  or speed and the results will depend on the type of data. Which type of
+  compression to choose depends on your specific needs and data. The list
+  of supported compression libraries:
+
+  - `zlib <https://zlib.net/>`_: The default compression library.
+    A classic in terms of compression, achieves good compression
+    rates but is somewhat slow.
+  - `lzo <https://www.oberhumer.com/opensource/lzo/>`_: Fast
+    compression and decompression.
+  - `bzip2 <https://sourceware.org/bzip2/>`_: Good compression rates.
+  - `blosc <https://www.blosc.org/>`_: Fast compression and
+    decompression.
+
+    Support for alternative blosc compressors:
+
+    - `blosc:blosclz <https://www.blosc.org/>`_ This is the
+      default compressor for ``blosc``
+    - `blosc:lz4
+      <https://fastcompression.blogspot.com/p/lz4.html>`_:
+      A compact, very popular and fast compressor.
+    - `blosc:lz4hc
+      <https://fastcompression.blogspot.com/p/lz4.html>`_:
+      A tweaked version of LZ4, produces better
+      compression ratios at the expense of speed.
+    - `blosc:snappy <https://google.github.io/snappy/>`_:
+      A popular compressor used in many places.
+    - `blosc:zlib <https://zlib.net/>`_: A classic;
+      somewhat slower than the previous ones, but
+      achieving better compression ratios.
+    - `blosc:zstd <https://facebook.github.io/zstd/>`_: An
+      extremely well balanced codec; it provides the best
+      compression ratios among the others above, and at
+      reasonably fast speed.
+
+  If ``complib`` is defined as something other than the listed libraries a
+  ``ValueError`` exception is issued.
+
+.. note::
+
+   If the library specified with the ``complib`` option is missing on your platform,
+   compression defaults to ``zlib`` without further ado.
+
+Enable compression for all objects within the file:
+
+.. code-block:: python
+
+   store_compressed = pd.HDFStore(
+       "store_compressed.h5", complevel=9, complib="blosc:blosclz"
+   )
+
+Or on-the-fly compression (this only applies to tables) in stores where compression is not enabled:
+
+.. code-block:: python
+
+   store.append("df", df, complib="zlib", complevel=5)
+
+.. _io.hdf5-ptrepack:
+
+ptrepack
+++++++++
+
+``PyTables`` offers better write performance when tables are compressed after
+they are written, as opposed to turning on compression at the very
+beginning. You can use the supplied ``PyTables`` utility
+``ptrepack``. In addition, ``ptrepack`` can change compression levels
+after the fact.
+
+.. code-block:: console
+
+   ptrepack --chunkshape=auto --propindexes --complevel=9 --complib=blosc in.h5 out.h5
+
+Furthermore ``ptrepack in.h5 out.h5`` will *repack* the file to allow
+you to reuse previously deleted space. Alternatively, one can simply
+remove the file and write again, or use the ``copy`` method.
+
+.. _io.hdf5-caveats:
+
+Caveats
++++++++
+
+.. warning::
+
+   ``HDFStore`` is **not-threadsafe for writing**. The underlying
+   ``PyTables`` only supports concurrent reads (via threading or
+   processes). If you need reading and writing *at the same time*, you
+   need to serialize these operations in a single thread in a single
+   process. You will corrupt your data otherwise. See the (:issue:`2397`) for more information.
+
+* If you use locks to manage write access between multiple processes, you
+  may want to use :py:func:`~os.fsync` before releasing write locks. For
+  convenience you can use ``store.flush(fsync=True)`` to do this for you.
+* Once a ``table`` is created columns (DataFrame)
+  are fixed; only exactly the same columns can be appended
+* Be aware that timezones (e.g., ``zoneinfo.ZoneInfo('US/Eastern')``)
+  are not necessarily equal across timezone versions.  So if data is
+  localized to a specific timezone in the HDFStore using one version
+  of a timezone library and that data is updated with another version, the data
+  will be converted to UTC since these timezones are not considered
+  equal.  Either use the same version of timezone library or use ``tz_convert`` with
+  the updated timezone definition.
+
+.. warning::
+
+   ``PyTables`` will show a ``NaturalNameWarning`` if a column name
+   cannot be used as an attribute selector.
+   *Natural* identifiers contain only letters, numbers, and underscores,
+   and may not begin with a number.
+   Other identifiers cannot be used in a ``where`` clause
+   and are generally a bad idea.
+
+.. _io.hdf5-data_types:
+
+DataTypes
+'''''''''
+
+``HDFStore`` will map an object dtype to the ``PyTables`` underlying
+dtype. This means the following types are known to work:
+
+======================================================  =========================
+Type                                                    Represents missing values
+======================================================  =========================
+floating : ``float64, float32, float16``                ``np.nan``
+integer : ``int64, int32, int8, uint64,uint32, uint8``
+boolean
+``datetime64[ns]``                                      ``NaT``
+``timedelta64[ns]``                                     ``NaT``
+categorical : see the section below
+object : ``strings``                                    ``np.nan``
+======================================================  =========================
+
+``unicode`` columns are not supported, and **WILL FAIL**.
+
+.. _io.hdf5-categorical:
+
+Categorical data
+++++++++++++++++
+
+You can write data that contains ``category`` dtypes to a ``HDFStore``.
+Queries work the same as if it was an object array. However, the ``category`` dtyped data is
+stored in a more efficient manner.
+
+.. ipython:: python
+
+   dfcat = pd.DataFrame(
+       {"A": pd.Series(list("aabbcdba")).astype("category"), "B": np.random.randn(8)}
+   )
+   dfcat
+   dfcat.dtypes
+   cstore = pd.HDFStore("cats.h5", mode="w")
+   cstore.append("dfcat", dfcat, format="table", data_columns=["A"])
+   result = cstore.select("dfcat", where="A in ['b', 'c']")
+   result
+   result.dtypes
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   cstore.close()
+   os.remove("cats.h5")
+
+
+String columns
+++++++++++++++
+
+**min_itemsize**
+
+The underlying implementation of ``HDFStore`` uses a fixed column width (itemsize) for string columns.
+A string column itemsize is calculated as the maximum of the
+length of data (for that column) that is passed to the ``HDFStore``, **in the first append**. Subsequent appends,
+may introduce a string for a column **larger** than the column can hold, an Exception will be raised (otherwise you
+could have a silent truncation of these columns, leading to loss of information). In the future we may relax this and
+allow a user-specified truncation to occur.
+
+Pass ``min_itemsize`` on the first table creation to a-priori specify the minimum length of a particular string column.
+``min_itemsize`` can be an integer, or a dict mapping a column name to an integer. You can pass ``values`` as a key to
+allow all *indexables* or *data_columns* to have this min_itemsize.
+
+Passing a ``min_itemsize`` dict will cause all passed columns to be created as *data_columns* automatically.
+
+.. note::
+
+   If you are not passing any ``data_columns``, then the ``min_itemsize`` will be the maximum of the length of any string passed
+
+.. ipython:: python
+
+   dfs = pd.DataFrame({"A": "foo", "B": "bar"}, index=list(range(5)))
+   dfs
+
+   # A and B have a size of 30
+   store.append("dfs", dfs, min_itemsize=30)
+   store.get_storer("dfs").table
+
+   # A is created as a data_column with a size of 30
+   # B is size is calculated
+   store.append("dfs2", dfs, min_itemsize={"A": 30})
+   store.get_storer("dfs2").table
+
+**nan_rep**
+
+String columns will serialize a ``np.nan`` (a missing value) with the ``nan_rep`` string representation. This defaults to the string value ``nan``.
+You could inadvertently turn an actual ``nan`` value into a missing value.
+
+.. ipython:: python
+
+   dfss = pd.DataFrame({"A": ["foo", "bar", "nan"]})
+   dfss
+
+   store.append("dfss", dfss)
+   store.select("dfss")
+
+   # here you need to specify a different nan rep
+   store.append("dfss2", dfss, nan_rep="_nan_")
+   store.select("dfss2")
+
+
+Performance
+'''''''''''
+
+* ``tables`` format come with a writing performance penalty as compared to
+  ``fixed`` stores. The benefit is the ability to append/delete and
+  query (potentially very large amounts of data).  Write times are
+  generally longer as compared with regular stores. Query times can
+  be quite fast, especially on an indexed axis.
+* You can pass ``chunksize=<int>`` to ``append``, specifying the
+  write chunksize (default is 50000). This will significantly lower
+  your memory usage on writing.
+* You can pass ``expectedrows=<int>`` to the first ``append``,
+  to set the TOTAL number of rows that ``PyTables`` will expect.
+  This will optimize read/write performance.
+* Duplicate rows can be written to tables, but are filtered out in
+  selection (with the last items being selected; thus a table is
+  unique on major, minor pairs)
+* A ``PerformanceWarning`` will be raised if you are attempting to
+  store types that will be pickled by PyTables (rather than stored as
+  endemic types). See
+  `Here <https://stackoverflow.com/questions/14355151/how-to-make-pandas-hdfstore-put-operation-faster/14370190#14370190>`__
+  for more information and some solutions.
+
+
+.. ipython:: python
+   :suppress:
+
+   store.close()
+   os.remove("store.h5")
+
+
+.. _io.feather:
+
+Feather
+-------
+
+Feather provides binary columnar serialization for data frames. It is designed to make reading and writing data
+frames efficient, and to make sharing data across data analysis languages easy.
+
+Feather is designed to faithfully serialize and de-serialize DataFrames, supporting all of the pandas
+dtypes, including extension dtypes such as categorical and datetime with tz.
+
+Several caveats:
+
+* The format will NOT write an ``Index``, or ``MultiIndex`` for the
+  ``DataFrame`` and will raise an error if a non-default one is provided. You
+  can ``.reset_index()`` to store the index or ``.reset_index(drop=True)`` to
+  ignore it.
+* Duplicate column names and non-string columns names are not supported
+* Actual Python objects in object dtype columns are not supported. These will
+  raise a helpful error message on an attempt at serialization.
+
+See the `Full Documentation <https://github.com/wesm/feather>`__.
+
+.. ipython:: python
+
+   import pytz
+
+   df = pd.DataFrame(
+       {
+           "a": list("abc"),
+           "b": list(range(1, 4)),
+           "c": np.arange(3, 6).astype("u1"),
+           "d": np.arange(4.0, 7.0, dtype="float64"),
+           "e": [True, False, True],
+           "f": pd.Categorical(list("abc")),
+           "g": pd.date_range("20130101", periods=3),
+           "h": pd.date_range("20130101", periods=3, tz=pytz.timezone("US/Eastern")),
+           "i": pd.date_range("20130101", periods=3, freq="ns"),
+       }
+   )
+
+   df
+   df.dtypes
+
+Write to a feather file.
+
+.. ipython:: python
+   :okwarning:
+
+   df.to_feather("example.feather")
+
+Read from a feather file.
+
+.. ipython:: python
+   :okwarning:
+
+   result = pd.read_feather("example.feather")
+   result
+
+   # we preserve dtypes
+   result.dtypes
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("example.feather")
+
+
+.. _io.parquet:
+
+Parquet
+-------
+
+`Apache Parquet <https://parquet.apache.org/>`__ provides a partitioned binary columnar serialization for data frames. It is designed to
+make reading and writing data frames efficient, and to make sharing data across data analysis
+languages easy. Parquet can use a variety of compression techniques to shrink the file size as much as possible
+while still maintaining good read performance.
+
+Parquet is designed to faithfully serialize and de-serialize ``DataFrame`` s, supporting all of the pandas
+dtypes, including extension dtypes such as datetime with tz.
+
+Several caveats.
+
+* Duplicate column names and non-string columns names are not supported.
+* The ``pyarrow`` engine always writes the index to the output, but ``fastparquet`` only writes non-default
+  indexes. This extra column can cause problems for non-pandas consumers that are not expecting it. You can
+  force including or omitting indexes with the ``index`` argument, regardless of the underlying engine.
+* Index level names, if specified, must be strings.
+* In the ``pyarrow`` engine, categorical dtypes for non-string types can be serialized to parquet, but will de-serialize as their primitive dtype.
+* The ``pyarrow`` engine preserves the ``ordered`` flag of categorical dtypes with string types. ``fastparquet`` does not preserve the ``ordered`` flag.
+* Non supported types include ``Interval`` and actual Python object types. These will raise a helpful error message
+  on an attempt at serialization. ``Period`` type is supported with pyarrow >= 0.16.0.
+* The ``pyarrow`` engine preserves extension data types such as the nullable integer and string data
+  type (requiring pyarrow >= 0.16.0, and requiring the extension type to implement the needed protocols,
+  see the :ref:`extension types documentation <extending.extension.arrow>`).
+
+You can specify an ``engine`` to direct the serialization. This can be one of ``pyarrow``, or ``fastparquet``, or ``auto``.
+If the engine is NOT specified, then the ``pd.options.io.parquet.engine`` option is checked; if this is also ``auto``,
+then ``pyarrow`` is tried, and falling back to ``fastparquet``.
+
+See the documentation for `pyarrow <https://arrow.apache.org/docs/python/>`__ and `fastparquet <https://fastparquet.readthedocs.io/en/latest/>`__.
+
+.. note::
+
+   These engines are very similar and should read/write nearly identical parquet format files.
+   ``pyarrow>=8.0.0`` supports timedelta data, ``fastparquet>=0.1.4`` supports timezone aware datetimes.
+   These libraries differ by having different underlying dependencies (``fastparquet`` by using ``numba``, while ``pyarrow`` uses a c-library).
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "a": list("abc"),
+           "b": list(range(1, 4)),
+           "c": np.arange(3, 6).astype("u1"),
+           "d": np.arange(4.0, 7.0, dtype="float64"),
+           "e": [True, False, True],
+           "f": pd.date_range("20130101", periods=3),
+           "g": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+           "h": pd.Categorical(list("abc")),
+           "i": pd.Categorical(list("abc"), ordered=True),
+       }
+   )
+
+   df
+   df.dtypes
+
+Write to a parquet file.
+
+.. ipython:: python
+
+   df.to_parquet("example_pa.parquet", engine="pyarrow")
+   df.to_parquet("example_fp.parquet", engine="fastparquet")
+
+Read from a parquet file.
+
+.. ipython:: python
+
+   result = pd.read_parquet("example_fp.parquet", engine="fastparquet")
+   result = pd.read_parquet("example_pa.parquet", engine="pyarrow")
+
+   result.dtypes
+
+By setting the ``dtype_backend`` argument you can control the default dtypes used for the resulting DataFrame.
+
+.. ipython:: python
+
+   result = pd.read_parquet("example_pa.parquet", engine="pyarrow", dtype_backend="pyarrow")
+
+   result.dtypes
+
+.. note::
+
+   Note that this is not supported for ``fastparquet``.
+
+
+Read only certain columns of a parquet file.
+
+.. ipython:: python
+
+   result = pd.read_parquet(
+       "example_fp.parquet",
+       engine="fastparquet",
+       columns=["a", "b"],
+   )
+   result = pd.read_parquet(
+       "example_pa.parquet",
+       engine="pyarrow",
+       columns=["a", "b"],
+   )
+   result.dtypes
+
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("example_pa.parquet")
+   os.remove("example_fp.parquet")
+
+
+Handling indexes
+''''''''''''''''
+
+Serializing a ``DataFrame`` to parquet may include the implicit index as one or
+more columns in the output file. Thus, this code:
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 2], "b": [3, 4]})
+    df.to_parquet("test.parquet", engine="pyarrow")
+
+creates a parquet file with *three* columns if you use ``pyarrow`` for serialization:
+``a``, ``b``, and ``__index_level_0__``. If you're using ``fastparquet``, the
+index `may or may not <https://fastparquet.readthedocs.io/en/latest/api.html#fastparquet.write>`_
+be written to the file.
+
+This unexpected extra column causes some databases like Amazon Redshift to reject
+the file, because that column doesn't exist in the target table.
+
+If you want to omit a dataframe's indexes when writing, pass ``index=False`` to
+:func:`~pandas.DataFrame.to_parquet`:
+
+.. ipython:: python
+
+    df.to_parquet("test.parquet", index=False)
+
+This creates a parquet file with just the two expected columns, ``a`` and ``b``.
+If your ``DataFrame`` has a custom index, you won't get it back when you load
+this file into a ``DataFrame``.
+
+Passing ``index=True`` will *always* write the index, even if that's not the
+underlying engine's default behavior.
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("test.parquet")
+
+
+Partitioning Parquet files
+''''''''''''''''''''''''''
+
+Parquet supports partitioning of data based on the values of one or more columns.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [0, 0, 1, 1], "b": [0, 1, 0, 1]})
+    df.to_parquet(path="test", engine="pyarrow", partition_cols=["a"], compression=None)
+
+The ``path`` specifies the parent directory to which data will be saved.
+The ``partition_cols`` are the column names by which the dataset will be partitioned.
+Columns are partitioned in the order they are given. The partition splits are
+determined by the unique values in the partition columns.
+The above example creates a partitioned dataset that may look like:
+
+.. code-block:: text
+
+    test
+    ├── a=0
+    │   ├── 0bac803e32dc42ae83fddfd029cbdebc.parquet
+    │   └──  ...
+    └── a=1
+        ├── e6ab24a4f45147b49b54a662f0c412a3.parquet
+        └── ...
+
+.. ipython:: python
+   :suppress:
+
+   from shutil import rmtree
+
+   try:
+       rmtree("test")
+   except OSError:
+       pass
+
+.. _io.orc:
+
+ORC
+---
+
+Similar to the :ref:`parquet <io.parquet>` format, the `ORC Format <https://orc.apache.org/>`__ is a binary columnar serialization
+for data frames. It is designed to make reading data frames efficient. pandas provides both the reader and the writer for the
+ORC format, :func:`~pandas.read_orc` and :func:`~pandas.DataFrame.to_orc`. This requires the `pyarrow <https://arrow.apache.org/docs/python/>`__ library.
+
+.. warning::
+
+   * It is *highly recommended* to install pyarrow using conda due to some issues occurred by pyarrow.
+   * :func:`~pandas.DataFrame.to_orc` requires pyarrow>=7.0.0.
+   * :func:`~pandas.read_orc` and :func:`~pandas.DataFrame.to_orc` are not supported on Windows yet, you can find valid environments on :ref:`install optional dependencies <install.warn_orc>`.
+   * For supported dtypes please refer to `supported ORC features in Arrow <https://arrow.apache.org/docs/cpp/orc.html#data-types>`__.
+   * Currently timezones in datetime columns are not preserved when a dataframe is converted into ORC files.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "a": list("abc"),
+           "b": list(range(1, 4)),
+           "c": np.arange(4.0, 7.0, dtype="float64"),
+           "d": [True, False, True],
+           "e": pd.date_range("20130101", periods=3),
+       }
+   )
+
+   df
+   df.dtypes
+
+Write to an orc file.
+
+.. ipython:: python
+
+   df.to_orc("example_pa.orc", engine="pyarrow")
+
+Read from an orc file.
+
+.. ipython:: python
+
+   result = pd.read_orc("example_pa.orc")
+
+   result.dtypes
+
+Read only certain columns of an orc file.
+
+.. ipython:: python
+
+   result = pd.read_orc(
+       "example_pa.orc",
+       columns=["a", "b"],
+   )
+   result.dtypes
+
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("example_pa.orc")
+
+
+.. _io.sql:
+
+SQL queries
+-----------
+
+The :mod:`pandas.io.sql` module provides a collection of query wrappers to both
+facilitate data retrieval and to reduce dependency on DB-specific API.
+
+Where available, users may first want to opt for `Apache Arrow ADBC
+<https://arrow.apache.org/adbc/current/index.html>`_ drivers. These drivers
+should provide the best performance, null handling, and type detection.
+
+  .. versionadded:: 2.2.0
+
+     Added native support for ADBC drivers
+
+For a full list of ADBC drivers and their development status, see the `ADBC Driver
+Implementation Status <https://arrow.apache.org/adbc/current/driver/status.html>`_
+documentation.
+
+Where an ADBC driver is not available or may be missing functionality,
+users should opt for installing SQLAlchemy alongside their database driver library.
+Examples of such drivers are `psycopg2 <https://www.psycopg.org/>`__
+for PostgreSQL or `pymysql <https://github.com/PyMySQL/PyMySQL>`__ for MySQL.
+For `SQLite <https://docs.python.org/3/library/sqlite3.html>`__ this is
+included in Python's standard library by default.
+You can find an overview of supported drivers for each SQL dialect in the
+`SQLAlchemy docs <https://docs.sqlalchemy.org/en/latest/dialects/index.html>`__.
+
+If SQLAlchemy is not installed, you can use a :class:`sqlite3.Connection` in place of
+a SQLAlchemy engine, connection, or URI string.
+
+See also some :ref:`cookbook examples <cookbook.sql>` for some advanced strategies.
+
+The key functions are:
+
+.. autosummary::
+
+    read_sql_table
+    read_sql_query
+    read_sql
+    DataFrame.to_sql
+
+.. note::
+
+    The function :func:`~pandas.read_sql` is a convenience wrapper around
+    :func:`~pandas.read_sql_table` and :func:`~pandas.read_sql_query` (and for
+    backward compatibility) and will delegate to specific function depending on
+    the provided input (database table name or sql query).
+    Table names do not need to be quoted if they have special characters.
+
+In the following example, we use the `SQlite <https://www.sqlite.org/index.html>`__ SQL database
+engine. You can use a temporary SQLite database where data are stored in
+"memory".
+
+To connect using an ADBC driver you will want to install the ``adbc_driver_sqlite`` using your
+package manager. Once installed, you can use the DBAPI interface provided by the ADBC driver
+to connect to your database.
+
+.. code-block:: python
+
+   import adbc_driver_sqlite.dbapi as sqlite_dbapi
+
+   # Create the connection
+   with sqlite_dbapi.connect("sqlite:///:memory:") as conn:
+        df = pd.read_sql_table("data", conn)
+
+To connect with SQLAlchemy you use the :func:`create_engine` function to create an engine
+object from database URI. You only need to create the engine once per database you are
+connecting to.
+For more information on :func:`create_engine` and the URI formatting, see the examples
+below and the SQLAlchemy `documentation <https://docs.sqlalchemy.org/en/latest/core/engines.html>`__
+
+.. ipython:: python
+
+   from sqlalchemy import create_engine
+
+   # Create your engine.
+   engine = create_engine("sqlite:///:memory:")
+
+If you want to manage your own connections you can pass one of those instead. The example below opens a
+connection to the database using a Python context manager that automatically closes the connection after
+the block has completed.
+See the `SQLAlchemy docs <https://docs.sqlalchemy.org/en/latest/core/connections.html#basic-usage>`__
+for an explanation of how the database connection is handled.
+
+.. code-block:: python
+
+   with engine.connect() as conn, conn.begin():
+       data = pd.read_sql_table("data", conn)
+
+.. warning::
+
+        When you open a connection to a database you are also responsible for closing it.
+        Side effects of leaving a connection open may include locking the database or
+        other breaking behaviour.
+
+Writing DataFrames
+''''''''''''''''''
+
+Assuming the following data is in a ``DataFrame`` ``data``, we can insert it into
+the database using :func:`~pandas.DataFrame.to_sql`.
+
++-----+------------+-------+-------+-------+
+| id  |    Date    | Col_1 | Col_2 | Col_3 |
++=====+============+=======+=======+=======+
+| 26  | 2012-10-18 |   X   |  25.7 | True  |
++-----+------------+-------+-------+-------+
+| 42  | 2012-10-19 |   Y   | -12.4 | False |
++-----+------------+-------+-------+-------+
+| 63  | 2012-10-20 |   Z   |  5.73 | True  |
++-----+------------+-------+-------+-------+
+
+
+.. ipython:: python
+
+   import datetime
+
+   c = ["id", "Date", "Col_1", "Col_2", "Col_3"]
+   d = [
+       (26, datetime.datetime(2010, 10, 18), "X", 27.5, True),
+       (42, datetime.datetime(2010, 10, 19), "Y", -12.5, False),
+       (63, datetime.datetime(2010, 10, 20), "Z", 5.73, True),
+   ]
+
+   data = pd.DataFrame(d, columns=c)
+
+   data
+   data.to_sql("data", con=engine)
+
+With some databases, writing large DataFrames can result in errors due to
+packet size limitations being exceeded. This can be avoided by setting the
+``chunksize`` parameter when calling ``to_sql``.  For example, the following
+writes ``data`` to the database in batches of 1000 rows at a time:
+
+.. ipython:: python
+
+    data.to_sql("data_chunked", con=engine, chunksize=1000)
+
+SQL data types
+++++++++++++++
+
+Ensuring consistent data type management across SQL databases is challenging.
+Not every SQL database offers the same types, and even when they do the implementation
+of a given type can vary in ways that have subtle effects on how types can be
+preserved.
+
+For the best odds at preserving database types users are advised to use
+ADBC drivers when available. The Arrow type system offers a wider array of
+types that more closely match database types than the historical pandas/NumPy
+type system. To illustrate, note this (non-exhaustive) listing of types
+available in different databases and pandas backends:
+
++-----------------+-----------------------+----------------+---------+
+|numpy/pandas     |arrow                  |postgres        |sqlite   |
++=================+=======================+================+=========+
+|int16/Int16      |int16                  |SMALLINT        |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|int32/Int32      |int32                  |INTEGER         |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|int64/Int64      |int64                  |BIGINT          |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|float32          |float32                |REAL            |REAL     |
++-----------------+-----------------------+----------------+---------+
+|float64          |float64                |DOUBLE PRECISION|REAL     |
++-----------------+-----------------------+----------------+---------+
+|object           |string                 |TEXT            |TEXT     |
++-----------------+-----------------------+----------------+---------+
+|bool             |``bool_``              |BOOLEAN         |         |
++-----------------+-----------------------+----------------+---------+
+|datetime64[ns]   |timestamp(us)          |TIMESTAMP       |         |
++-----------------+-----------------------+----------------+---------+
+|datetime64[ns,tz]|timestamp(us,tz)       |TIMESTAMPTZ     |         |
++-----------------+-----------------------+----------------+---------+
+|                 |date32                 |DATE            |         |
++-----------------+-----------------------+----------------+---------+
+|                 |month_day_nano_interval|INTERVAL        |         |
++-----------------+-----------------------+----------------+---------+
+|                 |binary                 |BINARY          |BLOB     |
++-----------------+-----------------------+----------------+---------+
+|                 |decimal128             |DECIMAL [#f1]_  |         |
++-----------------+-----------------------+----------------+---------+
+|                 |list                   |ARRAY [#f1]_    |         |
++-----------------+-----------------------+----------------+---------+
+|                 |struct                 |COMPOSITE TYPE  |         |
+|                 |                       | [#f1]_         |         |
++-----------------+-----------------------+----------------+---------+
+
+.. rubric:: Footnotes
+
+.. [#f1] Not implemented as of writing, but theoretically possible
+
+If you are interested in preserving database types as best as possible
+throughout the lifecycle of your DataFrame, users are encouraged to
+leverage the ``dtype_backend="pyarrow"`` argument of :func:`~pandas.read_sql`
+
+.. code-block:: ipython
+
+   # for roundtripping
+   with pg_dbapi.connect(uri) as conn:
+       df2 = pd.read_sql("pandas_table", conn, dtype_backend="pyarrow")
+
+This will prevent your data from being converted to the traditional pandas/NumPy
+type system, which often converts SQL types in ways that make them impossible to
+round-trip.
+
+In case an ADBC driver is not available, :func:`~pandas.DataFrame.to_sql`
+will try to map your data to an appropriate SQL data type based on the dtype of
+the data. When you have columns of dtype ``object``, pandas will try to infer
+the data type.
+
+You can always override the default type by specifying the desired SQL type of
+any of the columns by using the ``dtype`` argument. This argument needs a
+dictionary mapping column names to SQLAlchemy types (or strings for the sqlite3
+fallback mode).
+For example, specifying to use the sqlalchemy ``String`` type instead of the
+default ``Text`` type for string columns:
+
+.. ipython:: python
+
+    from sqlalchemy.types import String
+
+    data.to_sql("data_dtype", con=engine, dtype={"Col_1": String})
+
+.. note::
+
+    Due to the limited support for timedelta's in the different database
+    flavors, columns with type ``timedelta64`` will be written as integer
+    values as nanoseconds to the database and a warning will be raised. The only
+    exception to this is when using the ADBC PostgreSQL driver in which case a
+    timedelta will be written to the database as an ``INTERVAL``
+
+.. note::
+
+    Columns of ``category`` dtype will be converted to the dense representation
+    as you would get with ``np.asarray(categorical)`` (e.g. for string categories
+    this gives an array of strings).
+    Because of this, reading the database table back in does **not** generate
+    a categorical.
+
+.. _io.sql_datetime_data:
+
+Datetime data types
+'''''''''''''''''''
+
+Using ADBC or SQLAlchemy, :func:`~pandas.DataFrame.to_sql` is capable of writing
+datetime data that is timezone naive or timezone aware. However, the resulting
+data stored in the database ultimately depends on the supported data type
+for datetime data of the database system being used.
+
+The following table lists supported data types for datetime data for some
+common databases. Other database dialects may have different data types for
+datetime data.
+
+===========   =============================================  ===================
+Database      SQL Datetime Types                             Timezone Support
+===========   =============================================  ===================
+SQLite        ``TEXT``                                       No
+MySQL         ``TIMESTAMP`` or ``DATETIME``                  No
+PostgreSQL    ``TIMESTAMP`` or ``TIMESTAMP WITH TIME ZONE``  Yes
+===========   =============================================  ===================
+
+When writing timezone aware data to databases that do not support timezones,
+the data will be written as timezone naive timestamps that are in local time
+with respect to the timezone.
+
+:func:`~pandas.read_sql_table` is also capable of reading datetime data that is
+timezone aware or naive. When reading ``TIMESTAMP WITH TIME ZONE`` types, pandas
+will convert the data to UTC.
+
+.. _io.sql.method:
+
+Insertion method
+++++++++++++++++
+
+The parameter ``method`` controls the SQL insertion clause used.
+Possible values are:
+
+- ``None``: Uses standard SQL ``INSERT`` clause (one per row).
+- ``'multi'``: Pass multiple values in a single ``INSERT`` clause.
+  It uses a *special* SQL syntax not supported by all backends.
+  This usually provides better performance for analytic databases
+  like *Presto* and *Redshift*, but has worse performance for
+  traditional SQL backend if the table contains many columns.
+  For more information check the SQLAlchemy `documentation
+  <https://docs.sqlalchemy.org/en/latest/core/dml.html#sqlalchemy.sql.expression.Insert.values.params.*args>`__.
+- callable with signature ``(pd_table, conn, keys, data_iter)``:
+  This can be used to implement a more performant insertion method based on
+  specific backend dialect features.
+
+Example of a callable using PostgreSQL `COPY clause
+<https://www.postgresql.org/docs/current/sql-copy.html>`__::
+
+  # Alternative to_sql() *method* for DBs that support COPY FROM
+  import csv
+  from io import StringIO
+
+  def psql_insert_copy(table, conn, keys, data_iter):
+      """
+      Execute SQL statement inserting data
+
+      Parameters
+      ----------
+      table : pandas.io.sql.SQLTable
+      conn : sqlalchemy.engine.Engine or sqlalchemy.engine.Connection
+      keys : list of str
+          Column names
+      data_iter : Iterable that iterates the values to be inserted
+      """
+      # gets a DBAPI connection that can provide a cursor
+      dbapi_conn = conn.connection
+      with dbapi_conn.cursor() as cur:
+          s_buf = StringIO()
+          writer = csv.writer(s_buf)
+          writer.writerows(data_iter)
+          s_buf.seek(0)
+
+          columns = ', '.join(['"{}"'.format(k) for k in keys])
+          if table.schema:
+              table_name = '{}.{}'.format(table.schema, table.name)
+          else:
+              table_name = table.name
+
+          sql = 'COPY {} ({}) FROM STDIN WITH CSV'.format(
+              table_name, columns)
+          cur.copy_expert(sql=sql, file=s_buf)
+
+Reading tables
+''''''''''''''
+
+:func:`~pandas.read_sql_table` will read a database table given the
+table name and optionally a subset of columns to read.
+
+.. note::
+
+    In order to use :func:`~pandas.read_sql_table`, you **must** have the
+    ADBC driver or SQLAlchemy optional dependency installed.
+
+.. ipython:: python
+
+   pd.read_sql_table("data", engine)
+
+.. note::
+
+  ADBC drivers will map database types directly back to arrow types. For other drivers
+  note that pandas infers column dtypes from query outputs, and not by looking
+  up data types in the physical database schema. For example, assume ``userid``
+  is an integer column in a table. Then, intuitively, ``select userid ...`` will
+  return integer-valued series, while ``select cast(userid as text) ...`` will
+  return object-valued (str) series. Accordingly, if the query output is empty,
+  then all resulting columns will be returned as object-valued (since they are
+  most general). If you foresee that your query will sometimes generate an empty
+  result, you may want to explicitly typecast afterwards to ensure dtype
+  integrity.
+
+You can also specify the name of the column as the ``DataFrame`` index,
+and specify a subset of columns to be read.
+
+.. ipython:: python
+
+   pd.read_sql_table("data", engine, index_col="id")
+   pd.read_sql_table("data", engine, columns=["Col_1", "Col_2"])
+
+And you can explicitly force columns to be parsed as dates:
+
+.. ipython:: python
+
+   pd.read_sql_table("data", engine, parse_dates=["Date"])
+
+If needed you can explicitly specify a format string, or a dict of arguments
+to pass to :func:`pandas.to_datetime`:
+
+.. code-block:: python
+
+   pd.read_sql_table("data", engine, parse_dates={"Date": "%Y-%m-%d"})
+   pd.read_sql_table(
+       "data",
+       engine,
+       parse_dates={"Date": {"format": "%Y-%m-%d %H:%M:%S"}},
+   )
+
+
+You can check if a table exists using :func:`~pandas.io.sql.has_table`
+
+Schema support
+''''''''''''''
+
+Reading from and writing to different schemas is supported through the ``schema``
+keyword in the :func:`~pandas.read_sql_table` and :func:`~pandas.DataFrame.to_sql`
+functions. Note however that this depends on the database flavor (sqlite does not
+have schemas). For example:
+
+.. code-block:: python
+
+   df.to_sql(name="table", con=engine, schema="other_schema")
+   pd.read_sql_table("table", engine, schema="other_schema")
+
+Querying
+''''''''
+
+You can query using raw SQL in the :func:`~pandas.read_sql_query` function.
+In this case you must use the SQL variant appropriate for your database.
+When using SQLAlchemy, you can also pass SQLAlchemy Expression language constructs,
+which are database-agnostic.
+
+.. ipython:: python
+
+   pd.read_sql_query("SELECT * FROM data", engine)
+
+Of course, you can specify a more "complex" query.
+
+.. ipython:: python
+
+   pd.read_sql_query("SELECT id, Col_1, Col_2 FROM data WHERE id = 42;", engine)
+
+The :func:`~pandas.read_sql_query` function supports a ``chunksize`` argument.
+Specifying this will return an iterator through chunks of the query result:
+
+.. ipython:: python
+
+    df = pd.DataFrame(np.random.randn(20, 3), columns=list("abc"))
+    df.to_sql(name="data_chunks", con=engine, index=False)
+
+.. ipython:: python
+
+    for chunk in pd.read_sql_query("SELECT * FROM data_chunks", engine, chunksize=5):
+        print(chunk)
+
+
+Engine connection examples
+''''''''''''''''''''''''''
+
+To connect with SQLAlchemy you use the :func:`create_engine` function to create an engine
+object from database URI. You only need to create the engine once per database you are
+connecting to.
+
+.. code-block:: python
+
+   from sqlalchemy import create_engine
+
+   engine = create_engine("postgresql://scott:tiger@localhost:5432/mydatabase")
+
+   engine = create_engine("mysql+mysqldb://scott:tiger@localhost/foo")
+
+   engine = create_engine("oracle://scott:tiger@127.0.0.1:1521/sidname")
+
+   engine = create_engine("mssql+pyodbc://mydsn")
+
+   # sqlite://<nohostname>/<path>
+   # where <path> is relative:
+   engine = create_engine("sqlite:///foo.db")
+
+   # or absolute, starting with a slash:
+   engine = create_engine("sqlite:////absolute/path/to/foo.db")
+
+For more information see the examples the SQLAlchemy `documentation <https://docs.sqlalchemy.org/en/latest/core/engines.html>`__
+
+
+Advanced SQLAlchemy queries
+'''''''''''''''''''''''''''
+
+You can use SQLAlchemy constructs to describe your query.
+
+Use :func:`sqlalchemy.text` to specify query parameters in a backend-neutral way
+
+.. ipython:: python
+
+   import sqlalchemy as sa
+
+   pd.read_sql(
+       sa.text("SELECT * FROM data where Col_1=:col1"), engine, params={"col1": "X"}
+   )
+
+If you have an SQLAlchemy description of your database you can express where conditions using SQLAlchemy expressions
+
+.. ipython:: python
+
+   metadata = sa.MetaData()
+   data_table = sa.Table(
+       "data",
+       metadata,
+       sa.Column("index", sa.Integer),
+       sa.Column("Date", sa.DateTime),
+       sa.Column("Col_1", sa.String),
+       sa.Column("Col_2", sa.Float),
+       sa.Column("Col_3", sa.Boolean),
+   )
+
+   pd.read_sql(sa.select(data_table).where(data_table.c.Col_3 is True), engine)
+
+You can combine SQLAlchemy expressions with parameters passed to :func:`read_sql` using :func:`sqlalchemy.bindparam`
+
+.. ipython:: python
+
+    import datetime as dt
+
+    expr = sa.select(data_table).where(data_table.c.Date > sa.bindparam("date"))
+    pd.read_sql(expr, engine, params={"date": dt.datetime(2010, 10, 18)})
+
+
+Sqlite fallback
+'''''''''''''''
+
+The use of sqlite is supported without using SQLAlchemy.
+This mode requires a Python database adapter which respect the `Python
+DB-API <https://www.python.org/dev/peps/pep-0249/>`__.
+
+You can create connections like so:
+
+.. code-block:: python
+
+   import sqlite3
+
+   con = sqlite3.connect(":memory:")
+
+And then issue the following queries:
+
+.. code-block:: python
+
+   data.to_sql("data", con)
+   pd.read_sql_query("SELECT * FROM data", con)
+
+
+.. _io.bigquery:
+
+Google BigQuery
+---------------
+
+The ``pandas-gbq`` package provides functionality to read/write from Google BigQuery.
+
+Full documentation can be found `here <https://pandas-gbq.readthedocs.io/en/latest/>`__.
+
+.. _io.stata:
+
+STATA format
+------------
+
+.. _io.stata_writer:
+
+Writing to stata format
+'''''''''''''''''''''''
+
+The method :func:`.DataFrame.to_stata` will write a DataFrame
+into a .dta file. The format version of this file is always 115 (Stata 12).
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 2), columns=list("AB"))
+   df.to_stata("stata.dta")
+
+*Stata* data files have limited data type support; only strings with
+244 or fewer characters, ``int8``, ``int16``, ``int32``, ``float32``
+and ``float64`` can be stored in ``.dta`` files.  Additionally,
+*Stata* reserves certain values to represent missing data. Exporting a
+non-missing value that is outside of the permitted range in Stata for
+a particular data type will retype the variable to the next larger
+size.  For example, ``int8`` values are restricted to lie between -127
+and 100 in Stata, and so variables with values above 100 will trigger
+a conversion to ``int16``. ``nan`` values in floating points data
+types are stored as the basic missing data type (``.`` in *Stata*).
+
+.. note::
+
+    It is not possible to export missing data values for integer data types.
+
+
+The *Stata* writer gracefully handles other data types including ``int64``,
+``bool``, ``uint8``, ``uint16``, ``uint32`` by casting to
+the smallest supported type that can represent the data.  For example, data
+with a type of ``uint8`` will be cast to ``int8`` if all values are less than
+100 (the upper bound for non-missing ``int8`` data in *Stata*), or, if values are
+outside of this range, the variable is cast to ``int16``.
+
+
+.. warning::
+
+   Conversion from ``int64`` to ``float64`` may result in a loss of precision
+   if ``int64`` values are larger than 2**53.
+
+.. warning::
+
+  :class:`~pandas.io.stata.StataWriter` and
+  :func:`.DataFrame.to_stata` only support fixed width
+  strings containing up to 244 characters, a limitation imposed by the version
+  115 dta file format. Attempting to write *Stata* dta files with strings
+  longer than 244 characters raises a ``ValueError``.
+
+.. _io.stata_reader:
+
+Reading from Stata format
+'''''''''''''''''''''''''
+
+The top-level function ``read_stata`` will read a dta file and return
+either a ``DataFrame`` or a :class:`pandas.api.typing.StataReader` that can
+be used to read the file incrementally.
+
+.. ipython:: python
+
+   pd.read_stata("stata.dta")
+
+Specifying a ``chunksize`` yields a
+:class:`pandas.api.typing.StataReader` instance that can be used to
+read ``chunksize`` lines from the file at a time.  The ``StataReader``
+object can be used as an iterator.
+
+.. ipython:: python
+
+  with pd.read_stata("stata.dta", chunksize=3) as reader:
+      for df in reader:
+          print(df.shape)
+
+For more fine-grained control, use ``iterator=True`` and specify
+``chunksize`` with each call to
+:func:`~pandas.io.stata.StataReader.read`.
+
+.. ipython:: python
+
+  with pd.read_stata("stata.dta", iterator=True) as reader:
+      chunk1 = reader.read(5)
+      chunk2 = reader.read(5)
+
+Currently the ``index`` is retrieved as a column.
+
+The parameter ``convert_categoricals`` indicates whether value labels should be
+read and used to create a ``Categorical`` variable from them. Value labels can
+also be retrieved by the function ``value_labels``, which requires :func:`~pandas.io.stata.StataReader.read`
+to be called before use.
+
+The parameter ``convert_missing`` indicates whether missing value
+representations in Stata should be preserved.  If ``False`` (the default),
+missing values are represented as ``np.nan``.  If ``True``, missing values are
+represented using ``StataMissingValue`` objects, and columns containing missing
+values will have ``object`` data type.
+
+.. note::
+
+   :func:`~pandas.read_stata` and
+   :class:`~pandas.io.stata.StataReader` support .dta formats 113-115
+   (Stata 10-12), 117 (Stata 13), and 118 (Stata 14).
+
+.. note::
+
+   Setting ``preserve_dtypes=False`` will upcast to the standard pandas data types:
+   ``int64`` for all integer types and ``float64`` for floating point data.  By default,
+   the Stata data types are preserved when importing.
+
+.. note::
+
+   All :class:`~pandas.io.stata.StataReader` objects, whether created by :func:`~pandas.read_stata`
+   (when using ``iterator=True`` or ``chunksize``) or instantiated by hand, must be used as context
+   managers (e.g. the ``with`` statement).
+   While the :meth:`~pandas.io.stata.StataReader.close` method is available, its use is unsupported.
+   It is not part of the public API and will be removed in with future without warning.
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("stata.dta")
+
+.. _io.stata-categorical:
+
+Categorical data
+++++++++++++++++
+
+``Categorical`` data can be exported to *Stata* data files as value labeled data.
+The exported data consists of the underlying category codes as integer data values
+and the categories as value labels.  *Stata* does not have an explicit equivalent
+to a ``Categorical`` and information about *whether* the variable is ordered
+is lost when exporting.
+
+.. warning::
+
+    *Stata* only supports string value labels, and so ``str`` is called on the
+    categories when exporting data.  Exporting ``Categorical`` variables with
+    non-string categories produces a warning, and can result a loss of
+    information if the ``str`` representations of the categories are not unique.
+
+Labeled data can similarly be imported from *Stata* data files as ``Categorical``
+variables using the keyword argument ``convert_categoricals`` (``True`` by default).
+The keyword argument ``order_categoricals`` (``True`` by default) determines
+whether imported ``Categorical`` variables are ordered.
+
+.. note::
+
+    When importing categorical data, the values of the variables in the *Stata*
+    data file are not preserved since ``Categorical`` variables always
+    use integer data types between ``-1`` and ``n-1`` where ``n`` is the number
+    of categories. If the original values in the *Stata* data file are required,
+    these can be imported by setting ``convert_categoricals=False``, which will
+    import original data (but not the variable labels). The original values can
+    be matched to the imported categorical data since there is a simple mapping
+    between the original *Stata* data values and the category codes of imported
+    Categorical variables: missing values are assigned code ``-1``, and the
+    smallest original value is assigned ``0``, the second smallest is assigned
+    ``1`` and so on until the largest original value is assigned the code ``n-1``.
+
+.. note::
+
+    *Stata* supports partially labeled series. These series have value labels for
+    some but not all data values. Importing a partially labeled series will produce
+    a ``Categorical`` with string categories for the values that are labeled and
+    numeric categories for values with no label.
+
+.. _io.sas:
+
+.. _io.sas_reader:
+
+SAS formats
+-----------
+
+The top-level function :func:`read_sas` can read (but not write) SAS
+XPORT (.xpt) and SAS7BDAT (.sas7bdat) format files.
+
+SAS files only contain two value types: ASCII text and floating point
+values (usually 8 bytes but sometimes truncated).  For xport files,
+there is no automatic type conversion to integers, dates, or
+categoricals.  For SAS7BDAT files, the format codes may allow date
+variables to be automatically converted to dates.  By default the
+whole file is read and returned as a ``DataFrame``.
+
+Specify a ``chunksize`` or use ``iterator=True`` to obtain reader
+objects (``XportReader`` or ``SAS7BDATReader``) for incrementally
+reading the file.  The reader objects also have attributes that
+contain additional information about the file and its variables.
+
+Read a SAS7BDAT file:
+
+.. code-block:: python
+
+    df = pd.read_sas("sas_data.sas7bdat")
+
+Obtain an iterator and read an XPORT file 100,000 lines at a time:
+
+.. code-block:: python
+
+    def do_something(chunk):
+        pass
+
+
+    with pd.read_sas("sas_xport.xpt", chunk=100000) as rdr:
+        for chunk in rdr:
+            do_something(chunk)
+
+The specification_ for the xport file format is available from the SAS
+web site.
+
+.. _specification: https://support.sas.com/content/dam/SAS/support/en/technical-papers/record-layout-of-a-sas-version-5-or-6-data-set-in-sas-transport-xport-format.pdf
+
+No official documentation is available for the SAS7BDAT format.
+
+.. _io.spss:
+
+.. _io.spss_reader:
+
+SPSS formats
+------------
+
+The top-level function :func:`read_spss` can read (but not write) SPSS
+SAV (.sav) and  ZSAV (.zsav) format files.
+
+SPSS files contain column names. By default the
+whole file is read, categorical columns are converted into ``pd.Categorical``,
+and a ``DataFrame`` with all columns is returned.
+
+Specify the ``usecols`` parameter to obtain a subset of columns. Specify ``convert_categoricals=False``
+to avoid converting categorical columns into ``pd.Categorical``.
+
+Read an SPSS file:
+
+.. code-block:: python
+
+    df = pd.read_spss("spss_data.sav")
+
+Extract a subset of columns contained in ``usecols`` from an SPSS file and
+avoid converting categorical columns into ``pd.Categorical``:
+
+.. code-block:: python
+
+    df = pd.read_spss(
+        "spss_data.sav",
+        usecols=["foo", "bar"],
+        convert_categoricals=False,
+    )
+
+More information about the SAV and ZSAV file formats is available here_.
+
+.. _here: https://www.ibm.com/docs/en/spss-statistics/22.0.0
+
+.. _io.other:
+
+Other file formats
+------------------
+
+pandas itself only supports IO with a limited set of file formats that map
+cleanly to its tabular data model. For reading and writing other file formats
+into and from pandas, we recommend these packages from the broader community.
+
+netCDF
+''''''
+
+xarray_ provides data structures inspired by the pandas ``DataFrame`` for working
+with multi-dimensional datasets, with a focus on the netCDF file format and
+easy conversion to and from pandas.
+
+.. _xarray: https://xarray.pydata.org/en/stable/
+
+.. _io.perf:
+
+Performance considerations
+--------------------------
+
+This is an informal comparison of various IO methods, using pandas
+0.24.2. Timings are machine dependent and small differences should be
+ignored.
+
+.. code-block:: ipython
+
+   In [1]: sz = 1000000
+   In [2]: df = pd.DataFrame({'A': np.random.randn(sz), 'B': [1] * sz})
+
+   In [3]: df.info()
+   <class 'pandas.DataFrame'>
+   RangeIndex: 1000000 entries, 0 to 999999
+   Data columns (total 2 columns):
+   A    1000000 non-null float64
+   B    1000000 non-null int64
+   dtypes: float64(1), int64(1)
+   memory usage: 15.3 MB
+
+The following test functions will be used below to compare the performance of several IO methods:
+
+.. code-block:: python
+
+
+
+   import numpy as np
+
+   import os
+
+   sz = 1000000
+   df = pd.DataFrame({"A": np.random.randn(sz), "B": [1] * sz})
+
+   sz = 1000000
+   np.random.seed(42)
+   df = pd.DataFrame({"A": np.random.randn(sz), "B": [1] * sz})
+
+
+   def test_sql_write(df):
+       if os.path.exists("test.sql"):
+           os.remove("test.sql")
+       sql_db = sqlite3.connect("test.sql")
+       df.to_sql(name="test_table", con=sql_db)
+       sql_db.close()
+
+
+   def test_sql_read():
+       sql_db = sqlite3.connect("test.sql")
+       pd.read_sql_query("select * from test_table", sql_db)
+       sql_db.close()
+
+
+   def test_hdf_fixed_write(df):
+       df.to_hdf("test_fixed.hdf", key="test", mode="w")
+
+
+   def test_hdf_fixed_read():
+       pd.read_hdf("test_fixed.hdf", "test")
+
+
+   def test_hdf_fixed_write_compress(df):
+       df.to_hdf("test_fixed_compress.hdf", key="test", mode="w", complib="blosc")
+
+
+   def test_hdf_fixed_read_compress():
+       pd.read_hdf("test_fixed_compress.hdf", "test")
+
+
+   def test_hdf_table_write(df):
+       df.to_hdf("test_table.hdf", key="test", mode="w", format="table")
+
+
+   def test_hdf_table_read():
+       pd.read_hdf("test_table.hdf", "test")
+
+
+   def test_hdf_table_write_compress(df):
+       df.to_hdf(
+           "test_table_compress.hdf", key="test", mode="w", complib="blosc", format="table"
+       )
+
+
+   def test_hdf_table_read_compress():
+       pd.read_hdf("test_table_compress.hdf", "test")
+
+
+   def test_csv_write(df):
+       df.to_csv("test.csv", mode="w")
+
+
+   def test_csv_read():
+       pd.read_csv("test.csv", index_col=0)
+
+
+   def test_feather_write(df):
+       df.to_feather("test.feather")
+
+
+   def test_feather_read():
+       pd.read_feather("test.feather")
+
+
+   def test_pickle_write(df):
+       df.to_pickle("test.pkl")
+
+
+   def test_pickle_read():
+       pd.read_pickle("test.pkl")
+
+
+   def test_pickle_write_compress(df):
+       df.to_pickle("test.pkl.compress", compression="xz")
+
+
+   def test_pickle_read_compress():
+       pd.read_pickle("test.pkl.compress", compression="xz")
+
+
+   def test_parquet_write(df):
+       df.to_parquet("test.parquet")
+
+
+   def test_parquet_read():
+       pd.read_parquet("test.parquet")
+
+When writing, the top three functions in terms of speed are ``test_feather_write``, ``test_hdf_fixed_write`` and ``test_hdf_fixed_write_compress``.
+
+.. code-block:: ipython
+
+   In [4]: %timeit test_sql_write(df)
+   3.29 s ± 43.2 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [5]: %timeit test_hdf_fixed_write(df)
+   19.4 ms ± 560 µs per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [6]: %timeit test_hdf_fixed_write_compress(df)
+   19.6 ms ± 308 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [7]: %timeit test_hdf_table_write(df)
+   449 ms ± 5.61 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [8]: %timeit test_hdf_table_write_compress(df)
+   448 ms ± 11.9 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [9]: %timeit test_csv_write(df)
+   3.66 s ± 26.2 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [10]: %timeit test_feather_write(df)
+   9.75 ms ± 117 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
+
+   In [11]: %timeit test_pickle_write(df)
+   30.1 ms ± 229 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [12]: %timeit test_pickle_write_compress(df)
+   4.29 s ± 15.9 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [13]: %timeit test_parquet_write(df)
+   67.6 ms ± 706 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+When reading, the top three functions in terms of speed are ``test_feather_read``, ``test_pickle_read`` and
+``test_hdf_fixed_read``.
+
+
+.. code-block:: ipython
+
+   In [14]: %timeit test_sql_read()
+   1.77 s ± 17.7 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [15]: %timeit test_hdf_fixed_read()
+   19.4 ms ± 436 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [16]: %timeit test_hdf_fixed_read_compress()
+   19.5 ms ± 222 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [17]: %timeit test_hdf_table_read()
+   38.6 ms ± 857 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [18]: %timeit test_hdf_table_read_compress()
+   38.8 ms ± 1.49 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+   In [19]: %timeit test_csv_read()
+   452 ms ± 9.04 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [20]: %timeit test_feather_read()
+   12.4 ms ± 99.7 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
+
+   In [21]: %timeit test_pickle_read()
+   18.4 ms ± 191 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
+
+   In [22]: %timeit test_pickle_read_compress()
+   915 ms ± 7.48 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
+
+   In [23]: %timeit test_parquet_read()
+   24.4 ms ± 146 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+
+The files ``test.pkl.compress``, ``test.parquet`` and ``test.feather`` took the least space on disk (in bytes).
+
+.. code-block:: none
+
+    29519500 Oct 10 06:45 test.csv
+    16000248 Oct 10 06:45 test.feather
+    8281983  Oct 10 06:49 test.parquet
+    16000857 Oct 10 06:47 test.pkl
+    7552144  Oct 10 06:48 test.pkl.compress
+    34816000 Oct 10 06:42 test.sql
+    24009288 Oct 10 06:43 test_fixed.hdf
+    24009288 Oct 10 06:43 test_fixed_compress.hdf
+    24458940 Oct 10 06:44 test_table.hdf
+    24458940 Oct 10 06:44 test_table_compress.hdf
diff --git a/pandas/doc/source/user_guide/merging.rst b/pandas/doc/source/user_guide/merging.rst
new file mode 100644
index 0000000000000..60a66f5e6f2a8
--- /dev/null
+++ b/pandas/doc/source/user_guide/merging.rst
@@ -0,0 +1,1095 @@
+.. _merging:
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from matplotlib import pyplot as plt
+   import pandas.util._doctools as doctools
+
+   p = doctools.TablePlotter()
+
+
+************************************
+Merge, join, concatenate and compare
+************************************
+
+pandas provides various methods for combining and comparing :class:`Series` or
+:class:`DataFrame`.
+
+* :func:`~pandas.concat`: Merge multiple :class:`Series` or :class:`DataFrame` objects along a shared index or column
+* :meth:`DataFrame.join`: Merge multiple :class:`DataFrame` objects along the columns
+* :meth:`DataFrame.combine_first`: Update missing values with non-missing values in the same location
+* :func:`~pandas.merge`: Combine two :class:`Series` or :class:`DataFrame` objects with SQL-style joining
+* :func:`~pandas.merge_ordered`: Combine two :class:`Series` or :class:`DataFrame` objects along an ordered axis
+* :func:`~pandas.merge_asof`: Combine two :class:`Series` or :class:`DataFrame` objects by near instead of exact matching keys
+* :meth:`Series.compare` and :meth:`DataFrame.compare`: Show differences in values between two :class:`Series` or :class:`DataFrame` objects
+
+.. _merging.concat:
+
+:func:`~pandas.concat`
+----------------------
+
+The :func:`~pandas.concat` function concatenates an arbitrary amount of
+:class:`Series` or :class:`DataFrame` objects along an axis while
+performing optional set logic (union or intersection) of the indexes on
+the other axes. Like ``numpy.concatenate``, :func:`~pandas.concat`
+takes a list or dict of homogeneously-typed objects and concatenates them.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(
+       {
+           "A": ["A0", "A1", "A2", "A3"],
+           "B": ["B0", "B1", "B2", "B3"],
+           "C": ["C0", "C1", "C2", "C3"],
+           "D": ["D0", "D1", "D2", "D3"],
+       },
+       index=[0, 1, 2, 3],
+   )
+
+   df2 = pd.DataFrame(
+       {
+           "A": ["A4", "A5", "A6", "A7"],
+           "B": ["B4", "B5", "B6", "B7"],
+           "C": ["C4", "C5", "C6", "C7"],
+           "D": ["D4", "D5", "D6", "D7"],
+       },
+       index=[4, 5, 6, 7],
+   )
+
+   df3 = pd.DataFrame(
+       {
+           "A": ["A8", "A9", "A10", "A11"],
+           "B": ["B8", "B9", "B10", "B11"],
+           "C": ["C8", "C9", "C10", "C11"],
+           "D": ["D8", "D9", "D10", "D11"],
+       },
+       index=[8, 9, 10, 11],
+   )
+
+   frames = [df1, df2, df3]
+   result = pd.concat(frames)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_basic.png
+   p.plot(frames, result, labels=["df1", "df2", "df3"], vertical=True);
+   plt.close("all");
+
+.. note::
+
+   :func:`~pandas.concat` makes a full copy of the data, and iteratively
+   reusing :func:`~pandas.concat` can create unnecessary copies. Collect all
+   :class:`DataFrame` or :class:`Series` objects in a list before using
+   :func:`~pandas.concat`.
+
+   .. code-block:: python
+
+      frames = [process_your_file(f) for f in files]
+      result = pd.concat(frames)
+
+.. note::
+
+   When concatenating :class:`DataFrame` with named axes, pandas will attempt to preserve
+   these index/column names whenever possible. In the case where all inputs share a
+   common name, this name will be assigned to the result. When the input names do
+   not all agree, the result will be unnamed. The same is true for :class:`MultiIndex`,
+   but the logic is applied separately on a level-by-level basis.
+
+
+Joining logic of the resulting axis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The ``join`` keyword specifies how to handle axis values that don't exist in the first
+:class:`DataFrame`.
+
+``join='outer'`` takes the union of all axis values
+
+.. ipython:: python
+
+   df4 = pd.DataFrame(
+       {
+           "B": ["B2", "B3", "B6", "B7"],
+           "D": ["D2", "D3", "D6", "D7"],
+           "F": ["F2", "F3", "F6", "F7"],
+       },
+       index=[2, 3, 6, 7],
+   )
+   result = pd.concat([df1, df4], axis=1)
+   result
+
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_axis1.png
+   p.plot([df1, df4], result, labels=["df1", "df4"], vertical=False);
+   plt.close("all");
+
+``join='inner'`` takes the intersection of the axis values
+
+.. ipython:: python
+
+   result = pd.concat([df1, df4], axis=1, join="inner")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_axis1_inner.png
+   p.plot([df1, df4], result, labels=["df1", "df4"], vertical=False);
+   plt.close("all");
+
+To perform an effective "left" join using the *exact index* from the original
+:class:`DataFrame`, result can be reindexed.
+
+.. ipython:: python
+
+   result = pd.concat([df1, df4], axis=1).reindex(df1.index)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_axis1_join_axes.png
+   p.plot([df1, df4], result, labels=["df1", "df4"], vertical=False);
+   plt.close("all");
+
+.. _merging.ignore_index:
+
+Ignoring indexes on the concatenation axis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+For :class:`DataFrame` objects which don't have a meaningful index, the ``ignore_index``
+ignores overlapping indexes.
+
+.. ipython:: python
+
+   result = pd.concat([df1, df4], ignore_index=True, sort=False)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_ignore_index.png
+   p.plot([df1, df4], result, labels=["df1", "df4"], vertical=True);
+   plt.close("all");
+
+.. _merging.mixed_ndims:
+
+Concatenating :class:`Series` and :class:`DataFrame` together
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can concatenate a mix of :class:`Series` and :class:`DataFrame` objects. The
+:class:`Series` will be transformed to :class:`DataFrame` with the column name as
+the name of the :class:`Series`.
+
+.. ipython:: python
+
+   s1 = pd.Series(["X0", "X1", "X2", "X3"], name="X")
+   result = pd.concat([df1, s1], axis=1)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_mixed_ndim.png
+   p.plot([df1, s1], result, labels=["df1", "s1"], vertical=False);
+   plt.close("all");
+
+Unnamed :class:`Series` will be numbered consecutively.
+
+.. ipython:: python
+
+   s2 = pd.Series(["_0", "_1", "_2", "_3"])
+   result = pd.concat([df1, s2, s2, s2], axis=1)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_unnamed_series.png
+   p.plot([df1, s2], result, labels=["df1", "s2"], vertical=False);
+   plt.close("all");
+
+``ignore_index=True`` will drop all name references.
+
+.. ipython:: python
+
+   result = pd.concat([df1, s1], axis=1, ignore_index=True)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_series_ignore_index.png
+   p.plot([df1, s1], result, labels=["df1", "s1"], vertical=False);
+   plt.close("all");
+
+Resulting ``keys``
+~~~~~~~~~~~~~~~~~~
+
+The ``keys`` argument adds another axis level to the resulting index or column (creating
+a :class:`MultiIndex`) associate specific keys with each original :class:`DataFrame`.
+
+.. ipython:: python
+
+   result = pd.concat(frames, keys=["x", "y", "z"])
+   result
+   result.loc["y"]
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_keys.png
+   p.plot(frames, result, labels=["df1", "df2", "df3"], vertical=True)
+   plt.close("all");
+
+The ``keys`` argument can override the column names
+when creating a new :class:`DataFrame` based on existing :class:`Series`.
+
+.. ipython:: python
+
+   s3 = pd.Series([0, 1, 2, 3], name="foo")
+   s4 = pd.Series([0, 1, 2, 3])
+   s5 = pd.Series([0, 1, 4, 5])
+
+   pd.concat([s3, s4, s5], axis=1)
+   pd.concat([s3, s4, s5], axis=1, keys=["red", "blue", "yellow"])
+
+You can also pass a dict to :func:`concat` in which case the dict keys will be used
+for the ``keys`` argument unless other ``keys`` argument is specified:
+
+.. ipython:: python
+
+   pieces = {"x": df1, "y": df2, "z": df3}
+   result = pd.concat(pieces)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_dict.png
+   p.plot([df1, df2, df3], result, labels=["df1", "df2", "df3"], vertical=True);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.concat(pieces, keys=["z", "y"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_dict_keys.png
+   p.plot([df1, df2, df3], result, labels=["df1", "df2", "df3"], vertical=True);
+   plt.close("all");
+
+The :class:`MultiIndex` created has levels that are constructed from the passed keys and
+the index of the :class:`DataFrame` pieces:
+
+.. ipython:: python
+
+   result.index.levels
+
+``levels`` argument allows specifying resulting levels associated with the ``keys``
+
+.. ipython:: python
+
+   result = pd.concat(
+       pieces, keys=["x", "y", "z"], levels=[["z", "y", "x", "w"]], names=["group_key"]
+   )
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_concat_dict_keys_names.png
+   p.plot([df1, df2, df3], result, labels=["df1", "df2", "df3"], vertical=True);
+   plt.close("all");
+
+.. ipython:: python
+
+   result.index.levels
+
+.. _merging.append.row:
+
+Appending rows to a :class:`DataFrame`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If you have a :class:`Series` that you want to append as a single row to a :class:`DataFrame`, you can convert the row into a
+:class:`DataFrame` and use :func:`concat`
+
+.. ipython:: python
+
+   s2 = pd.Series(["X0", "X1", "X2", "X3"], index=["A", "B", "C", "D"])
+   result = pd.concat([df1, s2.to_frame().T], ignore_index=True)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_append_series_as_row.png
+   p.plot([df1, s2], result, labels=["df1", "s2"], vertical=True);
+   plt.close("all");
+
+.. _merging.join:
+
+:func:`~pandas.merge`
+---------------------
+
+:func:`~pandas.merge` performs join operations similar to relational databases like SQL.
+Users who are familiar with SQL but new to pandas can reference a
+:ref:`comparison with SQL<compare_with_sql.join>`.
+
+Merge types
+~~~~~~~~~~~
+
+:func:`~pandas.merge` implements common SQL style joining operations.
+
+* **one-to-one**: joining two :class:`DataFrame` objects on
+  their indexes which must contain unique values.
+* **many-to-one**: joining a unique index to one or
+  more columns in a different :class:`DataFrame`.
+* **many-to-many** : joining columns on columns.
+
+.. note::
+
+   When joining columns on columns, potentially a many-to-many join, any
+   indexes on the passed :class:`DataFrame` objects **will be discarded**.
+
+
+For a **many-to-many** join, if a key combination appears
+more than once in both tables, the :class:`DataFrame` will have the **Cartesian
+product** of the associated data.
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+       {
+           "key": ["K0", "K1", "K2", "K3"],
+           "A": ["A0", "A1", "A2", "A3"],
+           "B": ["B0", "B1", "B2", "B3"],
+       }
+   )
+
+   right = pd.DataFrame(
+       {
+           "key": ["K0", "K1", "K2", "K3"],
+           "C": ["C0", "C1", "C2", "C3"],
+           "D": ["D0", "D1", "D2", "D3"],
+       }
+   )
+   result = pd.merge(left, right, on="key")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+The ``how`` argument to :func:`~pandas.merge` specifies which keys are
+included in the resulting table. If a key combination **does not appear** in
+either the left or right tables, the values in the joined table will be
+``NA``. Here is a summary of the ``how`` options and their SQL equivalent names:
+
+.. csv-table::
+    :header: "Merge method", "SQL Join Name", "Description"
+    :widths: 20, 20, 60
+
+    ``left``, ``LEFT OUTER JOIN``, Use keys from left frame only
+    ``right``, ``RIGHT OUTER JOIN``, Use keys from right frame only
+    ``outer``, ``FULL OUTER JOIN``, Use union of keys from both frames
+    ``inner``, ``INNER JOIN``, Use intersection of keys from both frames
+    ``cross``, ``CROSS JOIN``, Create the cartesian product of rows of both frames
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+      {
+         "key1": ["K0", "K0", "K1", "K2"],
+         "key2": ["K0", "K1", "K0", "K1"],
+         "A": ["A0", "A1", "A2", "A3"],
+         "B": ["B0", "B1", "B2", "B3"],
+      }
+   )
+   right = pd.DataFrame(
+      {
+         "key1": ["K0", "K1", "K1", "K2"],
+         "key2": ["K0", "K0", "K0", "K0"],
+         "C": ["C0", "C1", "C2", "C3"],
+         "D": ["D0", "D1", "D2", "D3"],
+      }
+   )
+   result = pd.merge(left, right, how="left", on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key_left.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.merge(left, right, how="right", on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key_right.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+
+.. ipython:: python
+
+   result = pd.merge(left, right, how="outer", on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key_outer.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.merge(left, right, how="inner", on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key_inner.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.merge(left, right, how="cross")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_cross.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+You can merge :class:`Series` and a :class:`DataFrame` with a :class:`MultiIndex` if the names of
+the :class:`MultiIndex` correspond to the columns from the :class:`DataFrame`. Transform
+the :class:`Series` to a :class:`DataFrame` using :meth:`Series.reset_index` before merging
+
+.. ipython:: python
+
+   df = pd.DataFrame({"Let": ["A", "B", "C"], "Num": [1, 2, 3]})
+   df
+
+   ser = pd.Series(
+       ["a", "b", "c", "d", "e", "f"],
+       index=pd.MultiIndex.from_arrays(
+           [["A", "B", "C"] * 2, [1, 2, 3, 4, 5, 6]], names=["Let", "Num"]
+       ),
+   )
+   ser
+
+   pd.merge(df, ser.reset_index(), on=["Let", "Num"])
+
+
+Performing an outer join with duplicate join keys in :class:`DataFrame`
+
+.. ipython:: python
+
+   left = pd.DataFrame({"A": [1, 2], "B": [2, 2]})
+
+   right = pd.DataFrame({"A": [4, 5, 6], "B": [2, 2, 2]})
+
+   result = pd.merge(left, right, on="B", how="outer")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_on_key_dup.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+
+.. warning::
+
+  Merging on duplicate keys significantly increase the dimensions of the result
+  and can cause a memory overflow.
+
+.. _merging.validation:
+
+Merge key uniqueness
+~~~~~~~~~~~~~~~~~~~~
+
+The ``validate`` argument checks whether the uniqueness of merge keys.
+Key uniqueness is checked before merge operations and can protect against memory overflows
+and unexpected key duplication.
+
+.. ipython:: python
+   :okexcept:
+
+   left = pd.DataFrame({"A": [1, 2], "B": [1, 2]})
+   right = pd.DataFrame({"A": [4, 5, 6], "B": [2, 2, 2]})
+   result = pd.merge(left, right, on="B", how="outer", validate="one_to_one")
+
+If the user is aware of the duplicates in the right :class:`DataFrame` but wants to
+ensure there are no duplicates in the left :class:`DataFrame`, one can use the
+``validate='one_to_many'`` argument instead, which will not raise an exception.
+
+.. ipython:: python
+
+   pd.merge(left, right, on="B", how="outer", validate="one_to_many")
+
+
+.. _merging.indicator:
+
+Merge result indicator
+~~~~~~~~~~~~~~~~~~~~~~
+
+:func:`~pandas.merge` accepts the argument ``indicator``. If ``True``, a
+Categorical-type column called ``_merge`` will be added to the output object
+that takes on values:
+
+  ===================================   ================
+  Observation Origin                    ``_merge`` value
+  ===================================   ================
+  Merge key only in ``'left'`` frame    ``left_only``
+  Merge key only in ``'right'`` frame   ``right_only``
+  Merge key in both frames              ``both``
+  ===================================   ================
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"col1": [0, 1], "col_left": ["a", "b"]})
+   df2 = pd.DataFrame({"col1": [1, 2, 2], "col_right": [2, 2, 2]})
+   pd.merge(df1, df2, on="col1", how="outer", indicator=True)
+
+A string argument to ``indicator`` will use the value as the name for the indicator column.
+
+.. ipython:: python
+
+   pd.merge(df1, df2, on="col1", how="outer", indicator="indicator_column")
+
+
+Overlapping value columns
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The merge ``suffixes`` argument takes a tuple or list of strings to append to
+overlapping column names in the input :class:`DataFrame` to disambiguate the result
+columns:
+
+.. ipython:: python
+
+   left = pd.DataFrame({"k": ["K0", "K1", "K2"], "v": [1, 2, 3]})
+   right = pd.DataFrame({"k": ["K0", "K0", "K3"], "v": [4, 5, 6]})
+
+   result = pd.merge(left, right, on="k")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_overlapped.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.merge(left, right, on="k", suffixes=("_l", "_r"))
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_overlapped_suffix.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+:meth:`DataFrame.join`
+----------------------
+
+:meth:`DataFrame.join` combines the columns of multiple,
+potentially differently-indexed :class:`DataFrame` into a single result
+:class:`DataFrame`.
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+       {"A": ["A0", "A1", "A2"], "B": ["B0", "B1", "B2"]}, index=["K0", "K1", "K2"]
+   )
+
+   right = pd.DataFrame(
+       {"C": ["C0", "C2", "C3"], "D": ["D0", "D2", "D3"]}, index=["K0", "K2", "K3"]
+   )
+
+   result = left.join(right)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = left.join(right, how="outer")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_outer.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = left.join(right, how="inner")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_inner.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+:meth:`DataFrame.join` takes an optional ``on`` argument which may be a column
+or multiple column names that the passed :class:`DataFrame` is to be
+aligned.
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+       {
+           "A": ["A0", "A1", "A2", "A3"],
+           "B": ["B0", "B1", "B2", "B3"],
+           "key": ["K0", "K1", "K0", "K1"],
+       }
+   )
+
+   right = pd.DataFrame({"C": ["C0", "C1"], "D": ["D0", "D1"]}, index=["K0", "K1"])
+
+   result = left.join(right, on="key")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_key_columns.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. ipython:: python
+
+   result = pd.merge(
+       left, right, left_on="key", right_index=True, how="left", sort=False
+   )
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_key_columns.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. _merging.multikey_join:
+
+To join on multiple keys, the passed :class:`DataFrame` must have a :class:`MultiIndex`:
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+       {
+           "A": ["A0", "A1", "A2", "A3"],
+           "B": ["B0", "B1", "B2", "B3"],
+           "key1": ["K0", "K0", "K1", "K2"],
+           "key2": ["K0", "K1", "K0", "K1"],
+       }
+   )
+
+   index = pd.MultiIndex.from_tuples(
+       [("K0", "K0"), ("K1", "K0"), ("K2", "K0"), ("K2", "K1")]
+   )
+   right = pd.DataFrame(
+       {"C": ["C0", "C1", "C2", "C3"], "D": ["D0", "D1", "D2", "D3"]}, index=index
+   )
+   result = left.join(right, on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_multikeys.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. _merging.df_inner_join:
+
+The default for :class:`DataFrame.join` is to perform a left join
+which uses only the keys found in the
+calling :class:`DataFrame`. Other join types can be specified with ``how``.
+
+.. ipython:: python
+
+   result = left.join(right, on=["key1", "key2"], how="inner")
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_multikeys_inner.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. _merging.join_on_mi:
+
+Joining a single Index to a MultiIndex
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can join a :class:`DataFrame` with a :class:`Index` to a :class:`DataFrame` with a :class:`MultiIndex` on a level.
+The ``name`` of the :class:`Index` will match the level name of the :class:`MultiIndex`.
+
+..  ipython:: python
+
+    left = pd.DataFrame(
+        {"A": ["A0", "A1", "A2"], "B": ["B0", "B1", "B2"]},
+        index=pd.Index(["K0", "K1", "K2"], name="key"),
+    )
+
+    index = pd.MultiIndex.from_tuples(
+        [("K0", "Y0"), ("K1", "Y1"), ("K2", "Y2"), ("K2", "Y3")],
+        names=["key", "Y"],
+    )
+    right = pd.DataFrame(
+        {"C": ["C0", "C1", "C2", "C3"], "D": ["D0", "D1", "D2", "D3"]},
+        index=index,
+    )
+
+    result = left.join(right, how="inner")
+    result
+
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_multiindex_inner.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. _merging.join_with_two_multi_indexes:
+
+Joining with two :class:`MultiIndex`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The :class:`MultiIndex` of the input argument must be completely used
+in the join and is a subset of the indices in the left argument.
+
+.. ipython:: python
+
+   leftindex = pd.MultiIndex.from_product(
+       [list("abc"), list("xy"), [1, 2]], names=["abc", "xy", "num"]
+   )
+   left = pd.DataFrame({"v1": range(12)}, index=leftindex)
+   left
+
+   rightindex = pd.MultiIndex.from_product(
+       [list("abc"), list("xy")], names=["abc", "xy"]
+   )
+   right = pd.DataFrame({"v2": [100 * i for i in range(1, 7)]}, index=rightindex)
+   right
+
+   left.join(right, on=["abc", "xy"], how="inner")
+
+.. ipython:: python
+
+   leftindex = pd.MultiIndex.from_tuples(
+       [("K0", "X0"), ("K0", "X1"), ("K1", "X2")], names=["key", "X"]
+   )
+   left = pd.DataFrame(
+       {"A": ["A0", "A1", "A2"], "B": ["B0", "B1", "B2"]}, index=leftindex
+   )
+
+   rightindex = pd.MultiIndex.from_tuples(
+       [("K0", "Y0"), ("K1", "Y1"), ("K2", "Y2"), ("K2", "Y3")], names=["key", "Y"]
+   )
+   right = pd.DataFrame(
+       {"C": ["C0", "C1", "C2", "C3"], "D": ["D0", "D1", "D2", "D3"]}, index=rightindex
+   )
+
+   result = pd.merge(
+       left.reset_index(), right.reset_index(), on=["key"], how="inner"
+   ).set_index(["key", "X", "Y"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_merge_two_multiindex.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. _merging.merge_on_columns_and_levels:
+
+Merging on a combination of columns and index levels
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Strings passed as the ``on``, ``left_on``, and ``right_on`` parameters
+may refer to either column names or index level names.  This enables merging
+:class:`DataFrame` instances on a combination of index levels and columns without
+resetting indexes.
+
+.. ipython:: python
+
+   left_index = pd.Index(["K0", "K0", "K1", "K2"], name="key1")
+
+   left = pd.DataFrame(
+       {
+           "A": ["A0", "A1", "A2", "A3"],
+           "B": ["B0", "B1", "B2", "B3"],
+           "key2": ["K0", "K1", "K0", "K1"],
+       },
+       index=left_index,
+   )
+
+   right_index = pd.Index(["K0", "K1", "K2", "K2"], name="key1")
+
+   right = pd.DataFrame(
+       {
+           "C": ["C0", "C1", "C2", "C3"],
+           "D": ["D0", "D1", "D2", "D3"],
+           "key2": ["K0", "K0", "K0", "K1"],
+       },
+       index=right_index,
+   )
+
+   result = left.merge(right, on=["key1", "key2"])
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merge_on_index_and_column.png
+   p.plot([left, right], result, labels=["left", "right"], vertical=False);
+   plt.close("all");
+
+.. note::
+
+   When :class:`DataFrame` are joined on a string that matches an index level in both
+   arguments, the index level is preserved as an index level in the resulting
+   :class:`DataFrame`.
+
+.. note::
+
+   When :class:`DataFrame` are joined using only some of the levels of a :class:`MultiIndex`,
+   the extra levels will be dropped from the resulting join. To
+   preserve those levels, use :meth:`DataFrame.reset_index` on those level
+   names to move those levels to columns prior to the join.
+
+.. _merging.multiple_join:
+
+Joining multiple :class:`DataFrame`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A list or tuple of :class:`DataFrame` can also be passed to :meth:`~DataFrame.join`
+to join them together on their indexes.
+
+.. ipython:: python
+
+   right2 = pd.DataFrame({"v": [7, 8, 9]}, index=["K1", "K1", "K2"])
+   result = left.join([right, right2])
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_join_multi_df.png
+   p.plot(
+       [left, right, right2],
+       result,
+       labels=["left", "right", "right2"],
+       vertical=False,
+   );
+   plt.close("all");
+
+.. _merging.combine_first.update:
+
+:meth:`DataFrame.combine_first`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`DataFrame.combine_first` update missing values from one :class:`DataFrame`
+with the non-missing values in another :class:`DataFrame` in the corresponding
+location.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(
+       [[np.nan, 3.0, 5.0], [-4.6, np.nan, np.nan], [np.nan, 7.0, np.nan]]
+   )
+   df2 = pd.DataFrame([[-42.6, np.nan, -8.2], [-5.0, 1.6, 4]], index=[1, 2])
+   result = df1.combine_first(df2)
+   result
+
+.. ipython:: python
+   :suppress:
+
+   @savefig merging_combine_first.png
+   p.plot([df1, df2], result, labels=["df1", "df2"], vertical=False);
+   plt.close("all");
+
+.. _merging.merge_ordered:
+
+:func:`merge_ordered`
+---------------------
+
+:func:`merge_ordered` combines order data such as numeric or time series data
+with optional filling of missing data with ``fill_method``.
+
+.. ipython:: python
+
+   left = pd.DataFrame(
+       {"k": ["K0", "K1", "K1", "K2"], "lv": [1, 2, 3, 4], "s": ["a", "b", "c", "d"]}
+   )
+
+   right = pd.DataFrame({"k": ["K1", "K2", "K4"], "rv": [1, 2, 3]})
+
+   pd.merge_ordered(left, right, fill_method="ffill", left_by="s")
+
+.. _merging.merge_asof:
+
+:func:`merge_asof`
+---------------------
+
+:func:`merge_asof` is similar to an ordered left-join except that matches are on the
+nearest key rather than equal keys. For each row in the ``left`` :class:`DataFrame`,
+the last row in the ``right`` :class:`DataFrame` are selected where the ``on`` key is less
+than the left's key. Both :class:`DataFrame` must be sorted by the key.
+
+Optionally :func:`merge_asof` can perform a group-wise merge by matching the
+``by`` key in addition to the nearest match on the ``on`` key.
+
+.. ipython:: python
+
+   trades = pd.DataFrame(
+       {
+           "time": pd.to_datetime(
+               [
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.038",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.048",
+               ]
+           ),
+           "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+           "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+           "quantity": [75, 155, 100, 100, 100],
+       },
+       columns=["time", "ticker", "price", "quantity"],
+   )
+
+   quotes = pd.DataFrame(
+       {
+           "time": pd.to_datetime(
+               [
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.030",
+                   "20160525 13:30:00.041",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.049",
+                   "20160525 13:30:00.072",
+                   "20160525 13:30:00.075",
+               ]
+           ),
+           "ticker": ["GOOG", "MSFT", "MSFT", "MSFT", "GOOG", "AAPL", "GOOG", "MSFT"],
+           "bid": [720.50, 51.95, 51.97, 51.99, 720.50, 97.99, 720.50, 52.01],
+           "ask": [720.93, 51.96, 51.98, 52.00, 720.93, 98.01, 720.88, 52.03],
+       },
+       columns=["time", "ticker", "bid", "ask"],
+   )
+   trades
+   quotes
+   pd.merge_asof(trades, quotes, on="time", by="ticker")
+
+:func:`merge_asof` within ``2ms`` between the quote time and the trade time.
+
+.. ipython:: python
+
+   pd.merge_asof(trades, quotes, on="time", by="ticker", tolerance=pd.Timedelta("2ms"))
+
+:func:`merge_asof` within ``10ms`` between the quote time and the trade time and
+exclude exact matches on time. Note that though we exclude the exact matches
+(of the quotes), prior quotes **do** propagate to that point in time.
+
+.. ipython:: python
+
+   pd.merge_asof(
+       trades,
+       quotes,
+       on="time",
+       by="ticker",
+       tolerance=pd.Timedelta("10ms"),
+       allow_exact_matches=False,
+   )
+
+.. _merging.compare:
+
+:meth:`~Series.compare`
+-----------------------
+
+The :meth:`Series.compare` and :meth:`DataFrame.compare` methods allow you to
+compare two :class:`DataFrame` or :class:`Series`, respectively, and summarize their differences.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "col1": ["a", "a", "b", "b", "a"],
+           "col2": [1.0, 2.0, 3.0, np.nan, 5.0],
+           "col3": [1.0, 2.0, 3.0, 4.0, 5.0],
+       },
+       columns=["col1", "col2", "col3"],
+   )
+   df
+   df2 = df.copy()
+   df2.loc[0, "col1"] = "c"
+   df2.loc[2, "col3"] = 4.0
+   df2
+   df.compare(df2)
+
+By default, if two corresponding values are equal, they will be shown as ``NaN``.
+Furthermore, if all values in an entire row / column are equal, that row / column will be
+omitted from the result. The remaining differences will be aligned on columns.
+
+Stack the differences on rows.
+
+.. ipython:: python
+
+   df.compare(df2, align_axis=0)
+
+Keep all original rows and columns with ``keep_shape=True``
+
+.. ipython:: python
+
+   df.compare(df2, keep_shape=True)
+
+Keep all the original values even if they are equal.
+
+.. ipython:: python
+
+   df.compare(df2, keep_shape=True, keep_equal=True)
diff --git a/pandas/doc/source/user_guide/missing_data.rst b/pandas/doc/source/user_guide/missing_data.rst
new file mode 100644
index 0000000000000..e15939eb49239
--- /dev/null
+++ b/pandas/doc/source/user_guide/missing_data.rst
@@ -0,0 +1,695 @@
+.. _missing_data:
+
+{{ header }}
+
+*************************
+Working with missing data
+*************************
+
+Values considered "missing"
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas uses different sentinel values to represent a missing (also referred to as NA)
+depending on the data type.
+
+``numpy.nan`` for NumPy data types. The disadvantage of using NumPy data types
+is that the original data type will be coerced to ``np.float64`` or ``object``.
+
+.. ipython:: python
+
+   pd.Series([1, 2], dtype=np.int64).reindex([0, 1, 2])
+   pd.Series([True, False], dtype=np.bool_).reindex([0, 1, 2])
+
+:class:`NaT` for NumPy ``np.datetime64``, ``np.timedelta64``, and :class:`PeriodDtype`. For typing applications,
+use :class:`api.typing.NaTType`.
+
+.. ipython:: python
+
+   pd.Series([1, 2], dtype=np.dtype("timedelta64[ns]")).reindex([0, 1, 2])
+   pd.Series([1, 2], dtype=np.dtype("datetime64[ns]")).reindex([0, 1, 2])
+   pd.Series(["2020", "2020"], dtype=pd.PeriodDtype("D")).reindex([0, 1, 2])
+
+:class:`NA` for :class:`StringDtype`, :class:`Int64Dtype` (and other bit widths),
+:class:`Float64Dtype` (and other bit widths), :class:`BooleanDtype` and :class:`ArrowDtype`.
+These types will maintain the original data type of the data.
+For typing applications, use :class:`api.typing.NAType`.
+
+.. ipython:: python
+
+   pd.Series([1, 2], dtype="Int64").reindex([0, 1, 2])
+   pd.Series([True, False], dtype="boolean[pyarrow]").reindex([0, 1, 2])
+
+To detect these missing value, use the :func:`isna` or :func:`notna` methods.
+
+.. ipython:: python
+
+   ser = pd.Series([pd.Timestamp("2020-01-01"), pd.NaT])
+   ser
+   pd.isna(ser)
+
+
+.. note::
+
+   :func:`isna` or :func:`notna` will also consider ``None`` a missing value.
+
+   .. ipython:: python
+
+      ser = pd.Series([1, None], dtype=object)
+      ser
+      pd.isna(ser)
+
+.. warning::
+
+   Equality comparisons between ``np.nan``, :class:`NaT`, and :class:`NA`
+   do not act like ``None``
+
+   .. ipython:: python
+
+      None == None  # noqa: E711
+      np.nan == np.nan
+      pd.NaT == pd.NaT
+      pd.NA == pd.NA
+
+   Therefore, an equality comparison between a :class:`DataFrame` or :class:`Series`
+   with one of these missing values does not provide the same information as
+   :func:`isna` or :func:`notna`.
+
+   .. ipython:: python
+
+      ser = pd.Series([True, None], dtype="boolean[pyarrow]")
+      ser == pd.NA
+      pd.isna(ser)
+
+
+.. _missing_data.NA:
+
+:class:`NA` semantics
+~~~~~~~~~~~~~~~~~~~~~
+
+.. warning::
+
+   Experimental: the behaviour of :class:`NA` can still change without warning.
+
+Starting from pandas 1.0, an experimental :class:`NA` value (singleton) is
+available to represent scalar missing values. The goal of :class:`NA` is provide a
+"missing" indicator that can be used consistently across data types
+(instead of ``np.nan``, ``None`` or ``pd.NaT`` depending on the data type).
+
+For example, when having missing values in a :class:`Series` with the nullable integer
+dtype, it will use :class:`NA`:
+
+.. ipython:: python
+
+    s = pd.Series([1, 2, None], dtype="Int64")
+    s
+    s[2]
+    s[2] is pd.NA
+
+Currently, pandas does not use those data types using :class:`NA` by default in
+a :class:`DataFrame` or :class:`Series`, so you need to specify
+the dtype explicitly. An easy way to convert to those dtypes is explained in the
+:ref:`conversion section <missing_data.NA.conversion>`.
+
+Propagation in arithmetic and comparison operations
+---------------------------------------------------
+
+In general, missing values *propagate* in operations involving :class:`NA`. When
+one of the operands is unknown, the outcome of the operation is also unknown.
+
+For example, :class:`NA` propagates in arithmetic operations, similarly to
+``np.nan``:
+
+.. ipython:: python
+
+   pd.NA + 1
+   "a" * pd.NA
+
+There are a few special cases when the result is known, even when one of the
+operands is ``NA``.
+
+.. ipython:: python
+
+   pd.NA ** 0
+   1 ** pd.NA
+
+In equality and comparison operations, :class:`NA` also propagates. This deviates
+from the behaviour of ``np.nan``, where comparisons with ``np.nan`` always
+return ``False``.
+
+.. ipython:: python
+
+   pd.NA == 1
+   pd.NA == pd.NA
+   pd.NA < 2.5
+
+To check if a value is equal to :class:`NA`, use :func:`isna`
+
+.. ipython:: python
+
+   pd.isna(pd.NA)
+
+
+.. note::
+
+   An exception on this basic propagation rule are *reductions* (such as the
+   mean or the minimum), where pandas defaults to skipping missing values. See the
+   :ref:`calculation section <missing_data.calculations>` for more.
+
+Logical operations
+------------------
+
+For logical operations, :class:`NA` follows the rules of the
+`three-valued logic <https://en.wikipedia.org/wiki/Three-valued_logic>`__ (or
+*Kleene logic*, similarly to R, SQL and Julia). This logic means to only
+propagate missing values when it is logically required.
+
+For example, for the logical "or" operation (``|``), if one of the operands
+is ``True``, we already know the result will be ``True``, regardless of the
+other value (so regardless the missing value would be ``True`` or ``False``).
+In this case, :class:`NA` does not propagate:
+
+.. ipython:: python
+
+   True | False
+   True | pd.NA
+   pd.NA | True
+
+On the other hand, if one of the operands is ``False``, the result depends
+on the value of the other operand. Therefore, in this case :class:`NA`
+propagates:
+
+.. ipython:: python
+
+   False | True
+   False | False
+   False | pd.NA
+
+The behaviour of the logical "and" operation (``&``) can be derived using
+similar logic (where now :class:`NA` will not propagate if one of the operands
+is already ``False``):
+
+.. ipython:: python
+
+   False & True
+   False & False
+   False & pd.NA
+
+.. ipython:: python
+
+   True & True
+   True & False
+   True & pd.NA
+
+
+``NA`` in a boolean context
+---------------------------
+
+Since the actual value of an NA is unknown, it is ambiguous to convert NA
+to a boolean value.
+
+.. ipython:: python
+   :okexcept:
+
+   bool(pd.NA)
+
+This also means that :class:`NA` cannot be used in a context where it is
+evaluated to a boolean, such as ``if condition: ...`` where ``condition`` can
+potentially be :class:`NA`. In such cases, :func:`isna` can be used to check
+for :class:`NA` or ``condition`` being :class:`NA` can be avoided, for example by
+filling missing values beforehand.
+
+A similar situation occurs when using :class:`Series` or :class:`DataFrame` objects in ``if``
+statements, see :ref:`gotchas.truth`.
+
+NumPy ufuncs
+------------
+
+:attr:`pandas.NA` implements NumPy's ``__array_ufunc__`` protocol. Most ufuncs
+work with ``NA``, and generally return ``NA``:
+
+.. ipython:: python
+
+   np.log(pd.NA)
+   np.add(pd.NA, 1)
+
+.. warning::
+
+   Currently, ufuncs involving an ndarray and ``NA`` will return an
+   object-dtype filled with NA values.
+
+   .. ipython:: python
+
+      a = np.array([1, 2, 3])
+      np.greater(a, pd.NA)
+
+   The return type here may change to return a different array type
+   in the future.
+
+See :ref:`dsintro.numpy_interop` for more on ufuncs.
+
+.. _missing_data.NA.conversion:
+
+Conversion
+^^^^^^^^^^
+
+If you have a :class:`DataFrame` or :class:`Series` using ``np.nan``,
+:meth:`DataFrame.convert_dtypes` and :meth:`Series.convert_dtypes`, respectively,
+will convert your data to use the nullable data types supporting :class:`NA`,
+such as :class:`Int64Dtype` or :class:`ArrowDtype`. This is especially helpful after reading
+in data sets from IO methods where data types were inferred.
+
+In this example, while the dtypes of all columns are changed, we show the results for
+the first 10 columns.
+
+.. ipython:: python
+
+   import io
+   data = io.StringIO("a,b\n,True\n2,")
+   df = pd.read_csv(data)
+   df.dtypes
+   df_conv = df.convert_dtypes()
+   df_conv
+   df_conv.dtypes
+
+.. _missing.inserting:
+
+Inserting missing data
+~~~~~~~~~~~~~~~~~~~~~~
+
+You can insert missing values by simply assigning to a :class:`Series` or :class:`DataFrame`.
+The missing value sentinel used will be chosen based on the dtype.
+
+.. ipython:: python
+
+   ser = pd.Series([1., 2., 3.])
+   ser.loc[0] = None
+   ser
+
+   ser = pd.Series([pd.Timestamp("2021"), pd.Timestamp("2021")])
+   ser.iloc[0] = np.nan
+   ser
+
+   ser = pd.Series([True, False], dtype="boolean[pyarrow]")
+   ser.iloc[0] = None
+   ser
+
+For ``object`` types, pandas will use the value given:
+
+.. ipython:: python
+
+   s = pd.Series(["a", "b", "c"], dtype=object)
+   s.loc[0] = None
+   s.loc[1] = np.nan
+   s
+
+.. _missing_data.calculations:
+
+Calculations with missing data
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Missing values propagate through arithmetic operations between pandas objects.
+
+.. ipython:: python
+
+   ser1 = pd.Series([np.nan, np.nan, 2, 3])
+   ser2 = pd.Series([np.nan, 1, np.nan, 4])
+   ser1
+   ser2
+   ser1 + ser2
+
+The descriptive statistics and computational methods discussed in the
+:ref:`data structure overview <basics.stats>` (and listed :ref:`here
+<api.series.stats>` and :ref:`here <api.dataframe.stats>`) all
+account for missing data.
+
+When summing data, NA values or empty data will be treated as zero.
+
+.. ipython:: python
+
+   pd.Series([np.nan]).sum()
+   pd.Series([], dtype="float64").sum()
+
+When taking the product, NA values or empty data will be treated as 1.
+
+.. ipython:: python
+
+   pd.Series([np.nan]).prod()
+   pd.Series([], dtype="float64").prod()
+
+Cumulative methods like :meth:`~DataFrame.cumsum` and :meth:`~DataFrame.cumprod`
+ignore NA values by default, but preserve them in the resulting array. To override
+this behaviour and include NA values in the calculation, use ``skipna=False``.
+
+
+.. ipython:: python
+
+   ser = pd.Series([1, np.nan, 3, np.nan])
+   ser
+   ser.cumsum()
+   ser.cumsum(skipna=False)
+
+.. _missing_data.dropna:
+
+Dropping missing data
+~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`~DataFrame.dropna` drops rows or columns with missing data.
+
+.. ipython:: python
+
+   df = pd.DataFrame([[np.nan, 1, 2], [1, 2, np.nan], [1, 2, 3]])
+   df
+   df.dropna()
+   df.dropna(axis=1)
+
+   ser = pd.Series([1, pd.NA], dtype="int64[pyarrow]")
+   ser.dropna()
+
+Filling missing data
+~~~~~~~~~~~~~~~~~~~~
+
+.. _missing_data.fillna:
+
+Filling by value
+----------------
+
+:meth:`~DataFrame.fillna` replaces NA values with non-NA data.
+
+Replace NA with a scalar value
+
+.. ipython:: python
+
+   data = {"np": [1.0, np.nan, np.nan, 2], "arrow": pd.array([1.0, pd.NA, pd.NA, 2], dtype="float64[pyarrow]")}
+   df = pd.DataFrame(data)
+   df
+   df.fillna(0)
+
+When the data has object dtype, you can control what type of NA values are present.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": [pd.NA, np.nan, None]}, dtype=object)
+   df
+   df.fillna(None)
+   df.fillna(np.nan)
+   df.fillna(pd.NA)
+
+However when the dtype is not object, these will all be replaced with the proper NA value for the dtype.
+
+.. ipython:: python
+
+   data = {"np": [1.0, np.nan, np.nan, 2], "arrow": pd.array([1.0, pd.NA, pd.NA, 2], dtype="float64[pyarrow]")}
+   df = pd.DataFrame(data)
+   df
+   df.fillna(None)
+   df.fillna(np.nan)
+   df.fillna(pd.NA)
+
+Fill gaps forward or backward
+
+.. ipython:: python
+
+   df.ffill()
+   df.bfill()
+
+.. _missing_data.fillna.limit:
+
+Limit the number of NA values filled
+
+.. ipython:: python
+
+   df.ffill(limit=1)
+
+NA values can be replaced with corresponding value from a :class:`Series` or :class:`DataFrame`
+where the index and column aligns between the original object and the filled object.
+
+.. ipython:: python
+
+   dff = pd.DataFrame(np.arange(30, dtype=np.float64).reshape(10, 3), columns=list("ABC"))
+   dff.iloc[3:5, 0] = np.nan
+   dff.iloc[4:6, 1] = np.nan
+   dff.iloc[5:8, 2] = np.nan
+   dff
+   dff.fillna(dff.mean())
+
+.. note::
+
+   :meth:`DataFrame.where` can also be used to fill NA values.Same result as above.
+
+   .. ipython:: python
+
+      dff.where(pd.notna(dff), dff.mean(), axis="columns")
+
+
+.. _missing_data.interpolate:
+
+Interpolation
+-------------
+
+:meth:`DataFrame.interpolate` and :meth:`Series.interpolate` fills NA values
+using various interpolation methods.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": [1, 2.1, np.nan, 4.7, 5.6, 6.8],
+           "B": [0.25, np.nan, np.nan, 4, 12.2, 14.4],
+       }
+   )
+   df
+   df.interpolate()
+
+   idx = pd.date_range("2020-01-01", periods=10, freq="D")
+   data = np.random.default_rng(2).integers(0, 10, 10).astype(np.float64)
+   ts = pd.Series(data, index=idx)
+   ts.iloc[[1, 2, 5, 6, 9]] = np.nan
+
+   ts
+   @savefig series_before_interpolate.png
+   ts.plot()
+
+.. ipython:: python
+
+   ts.interpolate()
+   @savefig series_interpolate.png
+   ts.interpolate().plot()
+
+Interpolation relative to a :class:`Timestamp` in the :class:`DatetimeIndex`
+is available by setting ``method="time"``
+
+.. ipython:: python
+
+   ts2 = ts.iloc[[0, 1, 3, 7, 9]]
+   ts2
+   ts2.interpolate()
+   ts2.interpolate(method="time")
+
+For a floating-point index, use ``method='values'``:
+
+.. ipython:: python
+
+   idx = [0.0, 1.0, 10.0]
+   ser = pd.Series([0.0, np.nan, 10.0], idx)
+   ser
+   ser.interpolate()
+   ser.interpolate(method="values")
+
+If you have scipy_ installed, you can pass the name of a 1-d interpolation routine to ``method``.
+as specified in the scipy interpolation documentation_ and reference guide_.
+The appropriate interpolation method will depend on the data type.
+
+.. tip::
+
+   If you are dealing with a time series that is growing at an increasing rate,
+   use ``method='barycentric'``.
+
+   If you have values approximating a cumulative distribution function,
+   use ``method='pchip'``.
+
+   To fill missing values with goal of smooth plotting use ``method='akima'``.
+
+   .. ipython:: python
+
+      df = pd.DataFrame(
+         {
+            "A": [1, 2.1, np.nan, 4.7, 5.6, 6.8],
+            "B": [0.25, np.nan, np.nan, 4, 12.2, 14.4],
+         }
+      )
+      df
+      df.interpolate(method="barycentric")
+      df.interpolate(method="pchip")
+      df.interpolate(method="akima")
+
+When interpolating via a polynomial or spline approximation, you must also specify
+the degree or order of the approximation:
+
+.. ipython:: python
+
+   df.interpolate(method="spline", order=2)
+   df.interpolate(method="polynomial", order=2)
+
+Comparing several methods.
+
+.. ipython:: python
+
+   np.random.seed(2)
+
+   ser = pd.Series(np.arange(1, 10.1, 0.25) ** 2 + np.random.randn(37))
+   missing = np.array([4, 13, 14, 15, 16, 17, 18, 20, 29])
+   ser.iloc[missing] = np.nan
+   methods = ["linear", "quadratic", "cubic"]
+
+   df = pd.DataFrame({m: ser.interpolate(method=m) for m in methods})
+   @savefig compare_interpolations.png
+   df.plot()
+
+Interpolating new observations from expanding data with :meth:`Series.reindex`.
+
+.. ipython:: python
+
+   ser = pd.Series(np.sort(np.random.uniform(size=100)))
+
+   # interpolate at new_index
+   new_index = ser.index.union(pd.Index([49.25, 49.5, 49.75, 50.25, 50.5, 50.75]))
+   interp_s = ser.reindex(new_index).interpolate(method="pchip")
+   interp_s.loc[49:51]
+
+.. _scipy: https://scipy.org/
+.. _documentation: https://docs.scipy.org/doc/scipy/reference/interpolate.html#univariate-interpolation
+.. _guide: https://docs.scipy.org/doc/scipy/tutorial/interpolate.html
+
+.. _missing_data.interp_limits:
+
+Interpolation limits
+^^^^^^^^^^^^^^^^^^^^
+
+:meth:`~DataFrame.interpolate` accepts a ``limit`` keyword
+argument to limit the number of consecutive ``NaN`` values
+filled since the last valid observation
+
+.. ipython:: python
+
+   ser = pd.Series([np.nan, np.nan, 5, np.nan, np.nan, np.nan, 13, np.nan, np.nan])
+   ser
+   ser.interpolate()
+   ser.interpolate(limit=1)
+
+By default, ``NaN`` values are filled in a ``forward`` direction. Use
+``limit_direction`` parameter to fill ``backward`` or from ``both`` directions.
+
+.. ipython:: python
+
+   ser.interpolate(limit=1, limit_direction="backward")
+   ser.interpolate(limit=1, limit_direction="both")
+   ser.interpolate(limit_direction="both")
+
+By default, ``NaN`` values are filled whether they are surrounded by
+existing valid values or outside existing valid values. The ``limit_area``
+parameter restricts filling to either inside or outside values.
+
+.. ipython:: python
+
+   # fill one consecutive inside value in both directions
+   ser.interpolate(limit_direction="both", limit_area="inside", limit=1)
+
+   # fill all consecutive outside values backward
+   ser.interpolate(limit_direction="backward", limit_area="outside")
+
+   # fill all consecutive outside values in both directions
+   ser.interpolate(limit_direction="both", limit_area="outside")
+
+.. _missing_data.replace:
+
+Replacing values
+----------------
+
+:meth:`Series.replace` and :meth:`DataFrame.replace` can be used similar to
+:meth:`Series.fillna` and :meth:`DataFrame.fillna` to replace or insert missing values.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.eye(3))
+   df
+   df_missing = df.replace(0, np.nan)
+   df_missing
+   df_filled = df_missing.replace(np.nan, 2)
+   df_filled
+
+Replacing more than one value is possible by passing a list.
+
+.. ipython:: python
+
+   df_filled.replace([1, 44], [2, 28])
+
+Replacing using a mapping dict.
+
+.. ipython:: python
+
+   df_filled.replace({1: 44, 2: 28})
+
+.. _missing_data.replace_expression:
+
+Regular expression replacement
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. note::
+
+   Python strings prefixed with the ``r`` character such as ``r'hello world'``
+   are `"raw" strings <https://docs.python.org/3/reference/lexical_analysis.html#string-and-bytes-literals>`_.
+   They have different semantics regarding backslashes than strings without this prefix.
+   Backslashes in raw strings  will be interpreted as an escaped backslash, e.g., ``r'\' == '\\'``.
+
+Replace the '.' with ``NaN``
+
+.. ipython:: python
+
+   d = {"a": list(range(4)), "b": list("ab.."), "c": ["a", "b", np.nan, "d"]}
+   df = pd.DataFrame(d)
+   df.replace(".", np.nan)
+
+Replace the '.' with ``NaN`` with regular expression that removes surrounding whitespace
+
+.. ipython:: python
+
+   df.replace(r"\s*\.\s*", np.nan, regex=True)
+
+Replace with a list of regexes.
+
+.. ipython:: python
+
+   df.replace([r"\.", r"(a)"], ["dot", r"\1stuff"], regex=True)
+
+Replace with a regex in a mapping dict.
+
+.. ipython:: python
+
+   df.replace({"b": r"\s*\.\s*"}, {"b": np.nan}, regex=True)
+
+Pass nested dictionaries of regular expressions that use the ``regex`` keyword.
+
+.. ipython:: python
+
+   df.replace({"b": {"b": r""}}, regex=True)
+   df.replace(regex={"b": {r"\s*\.\s*": np.nan}})
+   df.replace({"b": r"\s*(\.)\s*"}, {"b": r"\1ty"}, regex=True)
+
+Pass a list of regular expressions that will replace matches with a scalar.
+
+.. ipython:: python
+
+   df.replace([r"\s*\.\s*", r"a|b"], "placeholder", regex=True)
+
+All of the regular expression examples can also be passed with the
+``to_replace`` argument as the ``regex`` argument. In this case the ``value``
+argument must be passed explicitly by name or ``regex`` must be a nested
+dictionary.
+
+.. ipython:: python
+
+   df.replace(regex=[r"\s*\.\s*", r"a|b"], value="placeholder")
+
+.. note::
+
+   A regular expression object from ``re.compile`` is a valid input as well.
diff --git a/pandas/doc/source/user_guide/options.rst b/pandas/doc/source/user_guide/options.rst
new file mode 100644
index 0000000000000..7757d95c2bccd
--- /dev/null
+++ b/pandas/doc/source/user_guide/options.rst
@@ -0,0 +1,407 @@
+.. _options:
+
+{{ header }}
+
+********************
+Options and settings
+********************
+
+Overview
+--------
+pandas has an options API to configure and customize global behavior related to
+:class:`DataFrame` display, data behavior and more.
+
+Options have a full "dotted-style", case-insensitive name (e.g. ``display.max_rows``).
+You can get/set options directly as attributes of the top-level ``options`` attribute:
+
+.. ipython:: python
+
+   import pandas as pd
+
+   pd.options.display.max_rows
+   pd.options.display.max_rows = 999
+   pd.options.display.max_rows
+
+The API is composed of 5 relevant functions, available directly from the ``pandas``
+namespace:
+
+* :func:`~pandas.get_option` / :func:`~pandas.set_option` - get/set the value of a single option.
+* :func:`~pandas.reset_option` - reset one or more options to their default value.
+* :func:`~pandas.describe_option` - print the descriptions of one or more options.
+* :func:`~pandas.option_context` - execute a codeblock with a set of options
+  that revert to prior settings after execution.
+
+.. note::
+
+   Developers can check out `pandas/core/config_init.py <https://github.com/pandas-dev/pandas/blob/main/pandas/core/config_init.py>`_ for more information.
+
+All of the functions above accept a regexp pattern (``re.search`` style) as an argument,
+to match an unambiguous substring:
+
+.. ipython:: python
+
+   pd.get_option("display.chop_threshold")
+   pd.set_option("display.chop_threshold", 2)
+   pd.get_option("display.chop_threshold")
+   pd.set_option("chop", 4)
+   pd.get_option("display.chop_threshold")
+
+
+The following will **not work** because it matches multiple option names, e.g.
+``display.max_colwidth``, ``display.max_rows``, ``display.max_columns``:
+
+.. ipython:: python
+   :okexcept:
+
+   pd.get_option("max")
+
+
+.. warning::
+
+    Using this form of shorthand may cause your code to break if new options with similar names are added in future versions.
+
+
+.. ipython:: python
+   :suppress:
+   :okwarning:
+
+   pd.reset_option("all")
+
+.. _options.available:
+
+Available options
+-----------------
+
+You can get a list of available options and their descriptions with :func:`~pandas.describe_option`. When called
+with no argument :func:`~pandas.describe_option` will print out the descriptions for all available options.
+
+.. ipython:: python
+
+   pd.describe_option()
+
+Getting and setting options
+---------------------------
+
+As described above, :func:`~pandas.get_option` and :func:`~pandas.set_option`
+are available from the pandas namespace.  To change an option, call
+``set_option('option regex', new_value)``.
+
+.. ipython:: python
+
+   pd.get_option("mode.sim_interactive")
+   pd.set_option("mode.sim_interactive", True)
+   pd.get_option("mode.sim_interactive")
+
+.. note::
+
+   The option ``'mode.sim_interactive'`` is mostly used for debugging purposes.
+
+You can use :func:`~pandas.reset_option` to revert to a setting's default value
+
+.. ipython:: python
+   :suppress:
+
+   pd.reset_option("display.max_rows")
+
+.. ipython:: python
+
+   pd.get_option("display.max_rows")
+   pd.set_option("display.max_rows", 999)
+   pd.get_option("display.max_rows")
+   pd.reset_option("display.max_rows")
+   pd.get_option("display.max_rows")
+
+
+It's also possible to reset multiple options at once (using a regex):
+
+.. ipython:: python
+   :okwarning:
+
+   pd.reset_option("^display")
+
+
+:func:`~pandas.option_context` context manager has been exposed through
+the top-level API, allowing you to execute code with given option values. Option values
+are restored automatically when you exit the ``with`` block:
+
+.. ipython:: python
+
+   with pd.option_context("display.max_rows", 10, "display.max_columns", 5):
+       print(pd.get_option("display.max_rows"))
+       print(pd.get_option("display.max_columns"))
+   print(pd.get_option("display.max_rows"))
+   print(pd.get_option("display.max_columns"))
+
+
+Setting startup options in Python/IPython environment
+-----------------------------------------------------
+
+Using startup scripts for the Python/IPython environment to import pandas and set options makes working with pandas more efficient.
+To do this, create a ``.py`` or ``.ipy`` script in the startup directory of the desired profile.
+An example where the startup folder is in a default IPython profile can be found at:
+
+.. code-block:: none
+
+  $IPYTHONDIR/profile_default/startup
+
+More information can be found in the `IPython documentation
+<https://ipython.org/ipython-doc/stable/interactive/tutorial.html#startup-files>`__.  An example startup script for pandas is displayed below:
+
+.. code-block:: python
+
+  import pandas as pd
+
+  pd.set_option("display.max_rows", 999)
+  pd.set_option("display.precision", 5)
+
+.. _options.frequently_used:
+
+Frequently used options
+-----------------------
+The following is a demonstrates the more frequently used display options.
+
+``display.max_rows`` and ``display.max_columns`` sets the maximum number
+of rows and columns displayed when a frame is pretty-printed. Truncated
+lines are replaced by an ellipsis.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(7, 2))
+   pd.set_option("display.max_rows", 7)
+   df
+   pd.set_option("display.max_rows", 5)
+   df
+   pd.reset_option("display.max_rows")
+
+Once the ``display.max_rows`` is exceeded, the ``display.min_rows`` options
+determines how many rows are shown in the truncated repr.
+
+.. ipython:: python
+
+   pd.set_option("display.max_rows", 8)
+   pd.set_option("display.min_rows", 4)
+   # below max_rows -> all rows shown
+   df = pd.DataFrame(np.random.randn(7, 2))
+   df
+   # above max_rows -> only min_rows (4) rows shown
+   df = pd.DataFrame(np.random.randn(9, 2))
+   df
+   pd.reset_option("display.max_rows")
+   pd.reset_option("display.min_rows")
+
+``display.expand_frame_repr`` allows for the representation of a
+:class:`DataFrame` to stretch across pages, wrapped over the all the columns.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(5, 10))
+   pd.set_option("expand_frame_repr", True)
+   df
+   pd.set_option("expand_frame_repr", False)
+   df
+   pd.reset_option("expand_frame_repr")
+
+``display.large_repr`` displays a :class:`DataFrame` that exceed
+``max_columns`` or ``max_rows`` as a truncated frame or summary.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 10))
+   pd.set_option("display.max_rows", 5)
+   pd.set_option("large_repr", "truncate")
+   df
+   pd.set_option("large_repr", "info")
+   df
+   pd.reset_option("large_repr")
+   pd.reset_option("display.max_rows")
+
+``display.max_colwidth`` sets the maximum width of columns.  Cells
+of this length or longer will be truncated with an ellipsis.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.array(
+           [
+               ["foo", "bar", "bim", "uncomfortably long string"],
+               ["horse", "cow", "banana", "apple"],
+           ]
+       )
+   )
+   pd.set_option("max_colwidth", 40)
+   df
+   pd.set_option("max_colwidth", 6)
+   df
+   pd.reset_option("max_colwidth")
+
+``display.max_info_columns`` sets a threshold for the number of columns
+displayed when calling :meth:`~pandas.DataFrame.info`.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 10))
+   pd.set_option("max_info_columns", 11)
+   df.info()
+   pd.set_option("max_info_columns", 5)
+   df.info()
+   pd.reset_option("max_info_columns")
+
+``display.max_info_rows``: :meth:`~pandas.DataFrame.info` will usually show null-counts for each column.
+For a large :class:`DataFrame`, this can be quite slow. ``max_info_rows`` and ``max_info_cols``
+limit this null check to the specified rows and columns respectively. The :meth:`~pandas.DataFrame.info`
+keyword argument ``show_counts=True`` will override this.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.choice([0, 1, np.nan], size=(10, 10)))
+   df
+   pd.set_option("max_info_rows", 11)
+   df.info()
+   pd.set_option("max_info_rows", 5)
+   df.info()
+   pd.reset_option("max_info_rows")
+
+``display.precision`` sets the output display precision in terms of decimal places.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(5, 5))
+   pd.set_option("display.precision", 7)
+   df
+   pd.set_option("display.precision", 4)
+   df
+
+``display.chop_threshold`` sets the rounding threshold to zero when displaying a
+:class:`Series` or :class:`DataFrame`. This setting does not change the
+precision at which the number is stored.
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(6, 6))
+   pd.set_option("chop_threshold", 0)
+   df
+   pd.set_option("chop_threshold", 0.5)
+   df
+   pd.reset_option("chop_threshold")
+
+``display.colheader_justify`` controls the justification of the headers.
+The options are ``'right'``, and ``'left'``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.array([np.random.randn(6), np.random.randint(1, 9, 6) * 0.1, np.zeros(6)]).T,
+       columns=["A", "B", "C"],
+       dtype="float",
+   )
+   pd.set_option("colheader_justify", "right")
+   df
+   pd.set_option("colheader_justify", "left")
+   df
+   pd.reset_option("colheader_justify")
+
+
+.. _basics.console_output:
+
+Number formatting
+------------------
+
+pandas also allows you to set how numbers are displayed in the console.
+This option is not set through the ``set_options`` API.
+
+Use the ``set_eng_float_format`` function
+to alter the floating-point formatting of pandas objects to produce a particular
+format.
+
+.. ipython:: python
+
+   import numpy as np
+
+   pd.set_eng_float_format(accuracy=3, use_eng_prefix=True)
+   s = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])
+   s / 1.0e3
+   s / 1.0e6
+
+.. ipython:: python
+   :suppress:
+   :okwarning:
+
+   pd.reset_option("^display")
+
+Use :meth:`~pandas.DataFrame.round` to specifically control rounding of an individual :class:`DataFrame`
+
+.. _options.east_asian_width:
+
+Unicode formatting
+------------------
+
+.. warning::
+
+   Enabling this option will affect the performance for printing of DataFrame and Series (about 2 times slower).
+   Use only when it is actually required.
+
+Some East Asian countries use Unicode characters whose width corresponds to two Latin characters.
+If a DataFrame or Series contains these characters, the default output mode may not align them properly.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"国籍": ["UK", "日本"], "名前": ["Alice", "しのぶ"]})
+   df
+
+Enabling ``display.unicode.east_asian_width`` allows pandas to check each character's "East Asian Width" property.
+These characters can be aligned properly by setting this option to ``True``. However, this will result in longer render
+times than the standard ``len`` function.
+
+.. ipython:: python
+
+   pd.set_option("display.unicode.east_asian_width", True)
+   df
+
+In addition, Unicode characters whose width is "ambiguous" can either be 1 or 2 characters wide depending on the
+terminal setting or encoding. The option ``display.unicode.ambiguous_as_wide`` can be used to handle the ambiguity.
+
+By default, an "ambiguous" character's width, such as "¡" (inverted exclamation) in the example below, is taken to be 1.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": ["xxx", "¡¡"], "b": ["yyy", "¡¡"]})
+   df
+
+
+Enabling ``display.unicode.ambiguous_as_wide`` makes pandas interpret these characters' widths to be 2.
+(Note that this option will only be effective when ``display.unicode.east_asian_width`` is enabled.)
+
+However, setting this option incorrectly for your terminal will cause these characters to be aligned incorrectly:
+
+.. ipython:: python
+
+   pd.set_option("display.unicode.ambiguous_as_wide", True)
+   df
+
+
+.. ipython:: python
+   :suppress:
+
+   pd.set_option("display.unicode.east_asian_width", False)
+   pd.set_option("display.unicode.ambiguous_as_wide", False)
+
+.. _options.table_schema:
+
+Table schema display
+--------------------
+
+:class:`DataFrame` and :class:`Series` will publish a Table Schema representation
+by default. This can be enabled globally with the
+``display.html.table_schema`` option:
+
+.. ipython:: python
+
+  pd.set_option("display.html.table_schema", True)
+
+Only ``'display.max_rows'`` are serialized and published.
+
+
+.. ipython:: python
+    :suppress:
+
+    pd.reset_option("display.html.table_schema")
diff --git a/pandas/doc/source/user_guide/pyarrow.rst b/pandas/doc/source/user_guide/pyarrow.rst
new file mode 100644
index 0000000000000..1807341530e69
--- /dev/null
+++ b/pandas/doc/source/user_guide/pyarrow.rst
@@ -0,0 +1,196 @@
+.. _pyarrow:
+
+{{ header }}
+
+*********************
+PyArrow Functionality
+*********************
+
+pandas can utilize `PyArrow <https://arrow.apache.org/docs/python/index.html>`__ to extend functionality and improve the performance
+of various APIs. This includes:
+
+* More extensive `data types <https://arrow.apache.org/docs/python/api/datatypes.html>`__ compared to NumPy
+* Missing data support (NA) for all data types
+* Performant IO reader integration
+* Facilitate interoperability with other dataframe libraries based on the Apache Arrow specification (e.g. polars, cuDF)
+
+To use this functionality, please ensure you have :ref:`installed the minimum supported PyArrow version. <install.optional_dependencies>`
+
+
+Data Structure Integration
+--------------------------
+
+A :class:`Series`, :class:`Index`, or the columns of a :class:`DataFrame` can be directly backed by a :external+pyarrow:py:class:`pyarrow.ChunkedArray`
+which is similar to a NumPy array. To construct these from the main pandas data structures, you can pass in a string of the type followed by
+``[pyarrow]``, e.g. ``"int64[pyarrow]"`` into the ``dtype`` parameter
+
+.. ipython:: python
+
+   ser = pd.Series([-1.5, 0.2, None], dtype="float32[pyarrow]")
+   ser
+
+   idx = pd.Index([True, None], dtype="bool[pyarrow]")
+   idx
+
+   df = pd.DataFrame([[1, 2], [3, 4]], dtype="uint64[pyarrow]")
+   df
+
+.. note::
+
+   The string alias ``"string[pyarrow]"`` maps to ``pd.StringDtype("pyarrow")`` which is not equivalent to
+   specifying ``dtype=pd.ArrowDtype(pa.string())``. Generally, operations on the data will behave similarly
+   except ``pd.StringDtype("pyarrow")`` can return NumPy-backed nullable types while ``pd.ArrowDtype(pa.string())``
+   will return :class:`ArrowDtype`.
+
+   .. ipython:: python
+
+      import pyarrow as pa
+      data = list("abc")
+      ser_sd = pd.Series(data, dtype="string[pyarrow]")
+      ser_ad = pd.Series(data, dtype=pd.ArrowDtype(pa.string()))
+      ser_ad.dtype == ser_sd.dtype
+      ser_sd.str.contains("a")
+      ser_ad.str.contains("a")
+
+For PyArrow types that accept parameters, you can pass in a PyArrow type with those parameters
+into :class:`ArrowDtype` to use in the ``dtype`` parameter.
+
+.. ipython:: python
+
+   import pyarrow as pa
+   list_str_type = pa.list_(pa.string())
+   ser = pd.Series([["hello"], ["there"]], dtype=pd.ArrowDtype(list_str_type))
+   ser
+
+.. ipython:: python
+
+   from datetime import time
+   idx = pd.Index([time(12, 30), None], dtype=pd.ArrowDtype(pa.time64("us")))
+   idx
+
+.. ipython:: python
+
+   from decimal import Decimal
+   decimal_type = pd.ArrowDtype(pa.decimal128(3, scale=2))
+   data = [[Decimal("3.19"), None], [None, Decimal("-1.23")]]
+   df = pd.DataFrame(data, dtype=decimal_type)
+   df
+
+If you already have an :external+pyarrow:py:class:`pyarrow.Array` or :external+pyarrow:py:class:`pyarrow.ChunkedArray`,
+you can pass it into :class:`.arrays.ArrowExtensionArray` to construct the associated :class:`Series`, :class:`Index`
+or :class:`DataFrame` object.
+
+.. ipython:: python
+
+   pa_array = pa.array(
+       [{"1": "2"}, {"10": "20"}, None],
+       type=pa.map_(pa.string(), pa.string()),
+   )
+   ser = pd.Series(pd.arrays.ArrowExtensionArray(pa_array))
+   ser
+
+To retrieve a pyarrow :external+pyarrow:py:class:`pyarrow.ChunkedArray` from a :class:`Series` or :class:`Index`, you can call
+the pyarrow array constructor on the :class:`Series` or :class:`Index`.
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, None], dtype="uint8[pyarrow]")
+   pa.array(ser)
+
+   idx = pd.Index(ser)
+   pa.array(idx)
+
+To convert a :external+pyarrow:py:class:`pyarrow.Table` to a :class:`DataFrame`, you can call the
+:external+pyarrow:py:meth:`pyarrow.Table.to_pandas` method with ``types_mapper=pd.ArrowDtype``.
+
+.. ipython:: python
+
+   table = pa.table([pa.array([1, 2, 3], type=pa.int64())], names=["a"])
+
+   df = table.to_pandas(types_mapper=pd.ArrowDtype)
+   df
+   df.dtypes
+
+
+Operations
+----------
+
+PyArrow data structure integration is implemented through pandas' :class:`~pandas.api.extensions.ExtensionArray` :ref:`interface <extending.extension-types>`;
+therefore, supported functionality exists where this interface is integrated within the pandas API. Additionally, this functionality
+is accelerated with PyArrow `compute functions <https://arrow.apache.org/docs/python/api/compute.html>`__ where available. This includes:
+
+* Numeric aggregations
+* Numeric arithmetic
+* Numeric rounding
+* Logical and comparison functions
+* String functionality
+* Datetime functionality
+
+The following are just some examples of operations that are accelerated by native PyArrow compute functions.
+
+.. ipython:: python
+
+   import pyarrow as pa
+   ser = pd.Series([-1.545, 0.211, None], dtype="float32[pyarrow]")
+   ser.mean()
+   ser + ser
+   ser > (ser + 1)
+
+   ser.dropna()
+   ser.isna()
+   ser.fillna(0)
+
+.. ipython:: python
+
+   ser_str = pd.Series(["a", "b", None], dtype=pd.ArrowDtype(pa.string()))
+   ser_str.str.startswith("a")
+
+.. ipython:: python
+
+   from datetime import datetime
+   pa_type = pd.ArrowDtype(pa.timestamp("ns"))
+   ser_dt = pd.Series([datetime(2022, 1, 1), None], dtype=pa_type)
+   ser_dt.dt.strftime("%Y-%m")
+
+I/O Reading
+-----------
+
+PyArrow also provides IO reading functionality that has been integrated into several pandas IO readers. The following
+functions provide an ``engine`` keyword that can dispatch to PyArrow to accelerate reading from an IO source.
+
+* :func:`read_csv`
+* :func:`read_feather`
+* :func:`read_json`
+* :func:`read_orc`
+* :func:`read_parquet`
+* :func:`read_table` (experimental)
+
+.. ipython:: python
+
+   import io
+   data = io.StringIO("""a,b,c
+      1,2.5,True
+      3,4.5,False
+   """)
+   df = pd.read_csv(data, engine="pyarrow")
+   df
+
+By default, these functions and all other IO reader functions return NumPy-backed data. These readers can return
+PyArrow-backed data by specifying the parameter ``dtype_backend="pyarrow"``. A reader does not need to set
+``engine="pyarrow"`` to necessarily return PyArrow-backed data.
+
+.. ipython:: python
+
+    import io
+    data = io.StringIO("""a,b,c,d,e,f,g,h,i
+        1,2.5,True,a,,,,,
+        3,4.5,False,b,6,7.5,True,a,
+    """)
+    df_pyarrow = pd.read_csv(data, dtype_backend="pyarrow")
+    df_pyarrow.dtypes
+
+Several non-IO reader functions can also use the ``dtype_backend`` argument to return PyArrow-backed data including:
+
+* :func:`to_numeric`
+* :meth:`DataFrame.convert_dtypes`
+* :meth:`Series.convert_dtypes`
diff --git a/pandas/doc/source/user_guide/reshaping.rst b/pandas/doc/source/user_guide/reshaping.rst
new file mode 100644
index 0000000000000..8c5e98791a9ef
--- /dev/null
+++ b/pandas/doc/source/user_guide/reshaping.rst
@@ -0,0 +1,677 @@
+.. _reshaping:
+
+{{ header }}
+
+**************************
+Reshaping and pivot tables
+**************************
+
+.. _reshaping.reshaping:
+
+
+pandas provides methods for manipulating a :class:`Series` and :class:`DataFrame` to alter the
+representation of the data for further data processing or data summarization.
+
+* :func:`~pandas.pivot` and :func:`~pandas.pivot_table`: Group unique values within one or more discrete categories.
+* :meth:`~DataFrame.stack` and :meth:`~DataFrame.unstack`: Pivot a column or row level to the opposite axis respectively.
+* :func:`~pandas.melt` and :func:`~pandas.wide_to_long`: Unpivot a wide :class:`DataFrame` to a long format.
+* :func:`~pandas.get_dummies` and :func:`~pandas.from_dummies`: Conversions with indicator variables.
+* :meth:`~Series.explode`: Convert a column of list-like values to individual rows.
+* :func:`~pandas.crosstab`: Calculate a cross-tabulation of multiple 1 dimensional factor arrays.
+* :func:`~pandas.cut`: Transform continuous variables to discrete, categorical values
+* :func:`~pandas.factorize`: Encode 1 dimensional variables into integer labels.
+
+
+:func:`~pandas.pivot` and :func:`~pandas.pivot_table`
+-----------------------------------------------------
+
+.. image:: ../_static/reshaping_pivot.png
+
+:func:`~pandas.pivot`
+~~~~~~~~~~~~~~~~~~~~~
+
+Data is often stored in so-called "stacked" or "record" format. In a "record" or "wide" format,
+typically there is one row for each subject. In the "stacked" or "long" format there are
+multiple rows for each subject where applicable.
+
+.. ipython:: python
+
+   data = {
+      "value": range(12),
+      "variable": ["A"] * 3 + ["B"] * 3 + ["C"] * 3 + ["D"] * 3,
+      "date": pd.to_datetime(["2020-01-03", "2020-01-04", "2020-01-05"] * 4)
+   }
+   df = pd.DataFrame(data)
+
+To perform time series operations with each unique variable, a better
+representation would be where the ``columns`` are the unique variables and an
+``index`` of dates identifies individual observations. To reshape the data into
+this form, we use the :meth:`DataFrame.pivot` method (also implemented as a
+top level function :func:`~pandas.pivot`):
+
+.. ipython:: python
+
+   pivoted = df.pivot(index="date", columns="variable", values="value")
+   pivoted
+
+If the ``values`` argument is omitted, and the input :class:`DataFrame` has more than
+one column of values which are not used as column or index inputs to :meth:`~DataFrame.pivot`,
+then the resulting "pivoted" :class:`DataFrame` will have :ref:`hierarchical columns
+<advanced.hierarchical>` whose topmost level indicates the respective value
+column:
+
+.. ipython:: python
+
+   df["value2"] = df["value"] * 2
+   pivoted = df.pivot(index="date", columns="variable")
+   pivoted
+
+You can then select subsets from the pivoted :class:`DataFrame`:
+
+.. ipython:: python
+
+   pivoted["value2"]
+
+Note that this returns a view on the underlying data in the case where the data
+are homogeneously-typed.
+
+.. note::
+
+   :func:`~pandas.pivot` can only handle unique rows specified by ``index`` and ``columns``.
+   If you data contains duplicates, use :func:`~pandas.pivot_table`.
+
+
+.. _reshaping.pivot:
+
+:func:`~pandas.pivot_table`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+While :meth:`~DataFrame.pivot` provides general purpose pivoting with various
+data types, pandas also provides :func:`~pandas.pivot_table` or :meth:`~DataFrame.pivot_table`
+for pivoting with aggregation of numeric data.
+
+The function :func:`~pandas.pivot_table` can be used to create spreadsheet-style
+pivot tables. See the :ref:`cookbook<cookbook.pivot>` for some advanced
+strategies.
+
+.. ipython:: python
+
+   import datetime
+
+   df = pd.DataFrame(
+       {
+           "A": ["one", "one", "two", "three"] * 6,
+           "B": ["A", "B", "C"] * 8,
+           "C": ["foo", "foo", "foo", "bar", "bar", "bar"] * 4,
+           "D": np.random.randn(24),
+           "E": np.random.randn(24),
+           "F": [datetime.datetime(2013, i, 1) for i in range(1, 13)]
+           + [datetime.datetime(2013, i, 15) for i in range(1, 13)],
+       }
+   )
+   df
+   pd.pivot_table(df, values="D", index=["A", "B"], columns=["C"])
+   pd.pivot_table(
+       df, values=["D", "E"],
+       index=["B"],
+       columns=["A", "C"],
+       aggfunc="sum",
+   )
+   pd.pivot_table(
+       df, values="E",
+       index=["B", "C"],
+       columns=["A"],
+       aggfunc=["sum", "mean"],
+   )
+
+The result is a :class:`DataFrame` potentially having a :class:`MultiIndex` on the
+index or column. If the ``values`` column name is not given, the pivot table
+will include all of the data in an additional level of hierarchy in the columns:
+
+.. ipython:: python
+
+   pd.pivot_table(df[["A", "B", "C", "D", "E"]], index=["A", "B"], columns=["C"])
+
+Also, you can use :class:`Grouper` for ``index`` and ``columns`` keywords. For detail of :class:`Grouper`, see :ref:`Grouping with a Grouper specification <groupby.specify>`.
+
+.. ipython:: python
+
+   pd.pivot_table(df, values="D", index=pd.Grouper(freq="ME", key="F"), columns="C")
+
+.. _reshaping.pivot.margins:
+
+Adding margins
+^^^^^^^^^^^^^^
+
+Passing ``margins=True`` to :meth:`~DataFrame.pivot_table` will add a row and column with an
+``All`` label with partial group aggregates across the categories on the
+rows and columns:
+
+.. ipython:: python
+
+   table = df.pivot_table(
+       index=["A", "B"],
+       columns="C",
+       values=["D", "E"],
+       margins=True,
+       aggfunc="std"
+   )
+   table
+
+Additionally, you can call :meth:`DataFrame.stack` to display a pivoted DataFrame
+as having a multi-level index:
+
+.. ipython:: python
+
+    table.stack()
+
+.. _reshaping.stacking:
+
+:meth:`~DataFrame.stack` and :meth:`~DataFrame.unstack`
+-------------------------------------------------------
+
+.. image:: ../_static/reshaping_stack.png
+
+Closely related to the :meth:`~DataFrame.pivot` method are the related
+:meth:`~DataFrame.stack` and :meth:`~DataFrame.unstack` methods available on
+:class:`Series` and :class:`DataFrame`. These methods are designed to work together with
+:class:`MultiIndex` objects (see the section on :ref:`hierarchical indexing
+<advanced.hierarchical>`).
+
+* :meth:`~DataFrame.stack`: "pivot" a level of the (possibly hierarchical) column labels,
+  returning a :class:`DataFrame` with an index with a new inner-most level of row
+  labels.
+* :meth:`~DataFrame.unstack`: (inverse operation of :meth:`~DataFrame.stack`) "pivot" a level of the
+  (possibly hierarchical) row index to the column axis, producing a reshaped
+  :class:`DataFrame` with a new inner-most level of column labels.
+
+.. image:: ../_static/reshaping_unstack.png
+
+.. ipython:: python
+
+   tuples = [
+      ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+      ["one", "two", "one", "two", "one", "two", "one", "two"],
+   ]
+   index = pd.MultiIndex.from_arrays(tuples, names=["first", "second"])
+   df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=["A", "B"])
+   df2 = df[:4]
+   df2
+
+The :meth:`~DataFrame.stack` function "compresses" a level in the :class:`DataFrame` columns to
+produce either:
+
+* A :class:`Series`, in the case of a :class:`Index` in the columns.
+* A :class:`DataFrame`, in the case of a :class:`MultiIndex` in the columns.
+
+If the columns have a :class:`MultiIndex`, you can choose which level to stack. The
+stacked level becomes the new lowest level in a :class:`MultiIndex` on the columns:
+
+.. ipython:: python
+
+   stacked = df2.stack()
+   stacked
+
+With a "stacked" :class:`DataFrame` or :class:`Series` (having a :class:`MultiIndex` as the
+``index``), the inverse operation of :meth:`~DataFrame.stack` is :meth:`~DataFrame.unstack`, which by default
+unstacks the **last level**:
+
+.. ipython:: python
+
+   stacked.unstack()
+   stacked.unstack(1)
+   stacked.unstack(0)
+
+.. _reshaping.unstack_by_name:
+
+.. image:: ../_static/reshaping_unstack_1.png
+
+If the indexes have names, you can use the level names instead of specifying
+the level numbers:
+
+.. ipython:: python
+
+   stacked.unstack("second")
+
+
+.. image:: ../_static/reshaping_unstack_0.png
+
+Notice that the :meth:`~DataFrame.stack` and :meth:`~DataFrame.unstack` methods implicitly sort the index
+levels involved. Hence a call to :meth:`~DataFrame.stack` and then :meth:`~DataFrame.unstack`, or vice versa,
+will result in a **sorted** copy of the original :class:`DataFrame` or :class:`Series`:
+
+.. ipython:: python
+
+   index = pd.MultiIndex.from_product([[2, 1], ["a", "b"]])
+   df = pd.DataFrame(np.random.randn(4), index=index, columns=["A"])
+   df
+   all(df.unstack().stack() == df.sort_index())
+
+.. _reshaping.stack_multiple:
+
+Multiple levels
+~~~~~~~~~~~~~~~
+
+You may also stack or unstack more than one level at a time by passing a list
+of levels, in which case the end result is as if each level in the list were
+processed individually.
+
+.. ipython:: python
+
+    columns = pd.MultiIndex.from_tuples(
+        [
+            ("A", "cat", "long"),
+            ("B", "cat", "long"),
+            ("A", "dog", "short"),
+            ("B", "dog", "short"),
+        ],
+        names=["exp", "animal", "hair_length"],
+    )
+    df = pd.DataFrame(np.random.randn(4, 4), columns=columns)
+    df
+
+    df.stack(level=["animal", "hair_length"])
+
+The list of levels can contain either level names or level numbers but
+not a mixture of the two.
+
+.. ipython:: python
+
+    # df.stack(level=['animal', 'hair_length'])
+    # from above is equivalent to:
+    df.stack(level=[1, 2])
+
+Missing data
+~~~~~~~~~~~~
+
+Unstacking can result in missing values if subgroups do not have the same
+set of labels. By default, missing values will be replaced with the default
+fill value for that data type.
+
+.. ipython:: python
+
+   columns = pd.MultiIndex.from_tuples(
+       [
+           ("A", "cat"),
+           ("B", "dog"),
+           ("B", "cat"),
+           ("A", "dog"),
+       ],
+       names=["exp", "animal"],
+   )
+   index = pd.MultiIndex.from_product(
+       [("bar", "baz", "foo", "qux"), ("one", "two")], names=["first", "second"]
+   )
+   df = pd.DataFrame(np.random.randn(8, 4), index=index, columns=columns)
+   df3 = df.iloc[[0, 1, 4, 7], [1, 2]]
+   df3
+   df3.unstack()
+
+The missing value can be filled with a specific value with the ``fill_value`` argument.
+
+.. ipython:: python
+
+   df3.unstack(fill_value=-1e9)
+
+.. _reshaping.melt:
+
+:func:`~pandas.melt` and :func:`~pandas.wide_to_long`
+-----------------------------------------------------
+
+.. image:: ../_static/reshaping_melt.png
+
+The top-level :func:`~pandas.melt` function and the corresponding :meth:`DataFrame.melt`
+are useful to reshape a :class:`DataFrame` into a format where one or more columns
+are *identifier variables*, while all other columns, considered *measured
+variables*, are "unpivoted" to the row axis, leaving just two non-identifier
+columns, "variable" and "value". The names of those columns can be customized
+by supplying the ``var_name`` and ``value_name`` parameters.
+
+.. ipython:: python
+
+   cheese = pd.DataFrame(
+       {
+           "first": ["John", "Mary"],
+           "last": ["Doe", "Bo"],
+           "height": [5.5, 6.0],
+           "weight": [130, 150],
+       }
+   )
+   cheese
+   cheese.melt(id_vars=["first", "last"])
+   cheese.melt(id_vars=["first", "last"], var_name="quantity")
+
+When transforming a DataFrame using :func:`~pandas.melt`, the index will be ignored.
+The original index values can be kept by setting the ``ignore_index=False`` parameter to ``False`` (default is ``True``).
+``ignore_index=False`` will however duplicate index values.
+
+.. ipython:: python
+
+   index = pd.MultiIndex.from_tuples([("person", "A"), ("person", "B")])
+   cheese = pd.DataFrame(
+       {
+           "first": ["John", "Mary"],
+           "last": ["Doe", "Bo"],
+           "height": [5.5, 6.0],
+           "weight": [130, 150],
+       },
+       index=index,
+   )
+   cheese
+   cheese.melt(id_vars=["first", "last"])
+   cheese.melt(id_vars=["first", "last"], ignore_index=False)
+
+:func:`~pandas.wide_to_long` is similar to :func:`~pandas.melt` with more customization for
+column matching.
+
+.. ipython:: python
+
+  dft = pd.DataFrame(
+      {
+          "A1970": {0: "a", 1: "b", 2: "c"},
+          "A1980": {0: "d", 1: "e", 2: "f"},
+          "B1970": {0: 2.5, 1: 1.2, 2: 0.7},
+          "B1980": {0: 3.2, 1: 1.3, 2: 0.1},
+          "X": dict(zip(range(3), np.random.randn(3))),
+      }
+  )
+  dft["id"] = dft.index
+  dft
+  pd.wide_to_long(dft, ["A", "B"], i="id", j="year")
+
+.. _reshaping.dummies:
+
+:func:`~pandas.get_dummies` and :func:`~pandas.from_dummies`
+------------------------------------------------------------
+
+To convert categorical variables of a :class:`Series` into a "dummy" or "indicator",
+:func:`~pandas.get_dummies` creates a new :class:`DataFrame` with columns of the unique
+variables and the values representing the presence of those variables per row.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"key": list("bbacab"), "data1": range(6)})
+
+   pd.get_dummies(df["key"])
+   df["key"].str.get_dummies()
+
+``prefix`` adds a prefix to the the column names which is useful for merging the result
+with the original :class:`DataFrame`:
+
+.. ipython:: python
+
+   dummies = pd.get_dummies(df["key"], prefix="key")
+   dummies
+
+   df[["data1"]].join(dummies)
+
+This function is often used along with discretization functions like :func:`~pandas.cut`:
+
+.. ipython:: python
+
+   values = np.random.randn(10)
+   values
+
+   bins = [0, 0.2, 0.4, 0.6, 0.8, 1]
+
+   pd.get_dummies(pd.cut(values, bins))
+
+
+:func:`get_dummies` also accepts a :class:`DataFrame`. By default, ``object``, ``string``,
+or ``categorical`` type columns are encoded as dummy variables with other columns unaltered.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": ["a", "b", "a"], "B": ["c", "c", "b"], "C": [1, 2, 3]})
+    pd.get_dummies(df)
+
+Specifying the ``columns`` keyword will encode a column of any type.
+
+.. ipython:: python
+
+    pd.get_dummies(df, columns=["A"])
+
+As with the :class:`Series` version, you can pass values for the ``prefix`` and
+``prefix_sep``. By default the column name is used as the prefix and ``_`` as
+the prefix separator. You can specify ``prefix`` and ``prefix_sep`` in 3 ways:
+
+* string: Use the same value for ``prefix`` or ``prefix_sep`` for each column
+  to be encoded.
+* list: Must be the same length as the number of columns being encoded.
+* dict: Mapping column name to prefix.
+
+.. ipython:: python
+
+    simple = pd.get_dummies(df, prefix="new_prefix")
+    simple
+    from_list = pd.get_dummies(df, prefix=["from_A", "from_B"])
+    from_list
+    from_dict = pd.get_dummies(df, prefix={"B": "from_B", "A": "from_A"})
+    from_dict
+
+To avoid collinearity when feeding the result to statistical models,
+specify ``drop_first=True``.
+
+.. ipython:: python
+
+    s = pd.Series(list("abcaa"))
+
+    pd.get_dummies(s)
+
+    pd.get_dummies(s, drop_first=True)
+
+When a column contains only one level, it will be omitted in the result.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": list("aaaaa"), "B": list("ababc")})
+
+    pd.get_dummies(df)
+
+    pd.get_dummies(df, drop_first=True)
+
+The values can be cast to a different type using the ``dtype`` argument.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": list("abc"), "B": [1.1, 2.2, 3.3]})
+
+    pd.get_dummies(df, dtype=np.float32).dtypes
+
+.. versionadded:: 1.5.0
+
+:func:`~pandas.from_dummies` converts the output of :func:`~pandas.get_dummies` back into
+a :class:`Series` of categorical values from indicator values.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"prefix_a": [0, 1, 0], "prefix_b": [1, 0, 1]})
+   df
+
+   pd.from_dummies(df, sep="_")
+
+Dummy coded data only requires ``k - 1`` categories to be included, in this case
+the last category is the default category. The default category can be modified with
+``default_category``.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"prefix_a": [0, 1, 0]})
+   df
+
+   pd.from_dummies(df, sep="_", default_category="b")
+
+.. _reshaping.explode:
+
+:meth:`~Series.explode`
+-----------------------
+
+For a :class:`DataFrame` column with nested, list-like values, :meth:`~Series.explode` will transform
+each list-like value to a separate row. The resulting :class:`Index` will be duplicated corresponding
+to the index label from the original row:
+
+.. ipython:: python
+
+   keys = ["panda1", "panda2", "panda3"]
+   values = [["eats", "shoots"], ["shoots", "leaves"], ["eats", "leaves"]]
+   df = pd.DataFrame({"keys": keys, "values": values})
+   df
+   df["values"].explode()
+
+:class:`DataFrame.explode` can also explode the column in the :class:`DataFrame`.
+
+.. ipython:: python
+
+   df.explode("values")
+
+:meth:`Series.explode` will replace empty lists with a missing value indicator and preserve scalar entries.
+
+.. ipython:: python
+
+   s = pd.Series([[1, 2, 3], "foo", [], ["a", "b"]])
+   s
+   s.explode()
+
+A comma-separated string value can be split into individual values in a list and then exploded to a new row.
+
+.. ipython:: python
+
+    df = pd.DataFrame([{"var1": "a,b,c", "var2": 1}, {"var1": "d,e,f", "var2": 2}])
+    df.assign(var1=df.var1.str.split(",")).explode("var1")
+
+.. _reshaping.crosstabulations:
+
+:func:`~pandas.crosstab`
+------------------------
+
+Use :func:`~pandas.crosstab` to compute a cross-tabulation of two (or more)
+factors. By default :func:`~pandas.crosstab` computes a frequency table of the factors
+unless an array of values and an aggregation function are passed.
+
+Any :class:`Series` passed will have their name attributes used unless row or column
+names for the cross-tabulation are specified
+
+.. ipython:: python
+
+    a = np.array(["foo", "foo", "bar", "bar", "foo", "foo"], dtype=object)
+    b = np.array(["one", "one", "two", "one", "two", "one"], dtype=object)
+    c = np.array(["dull", "dull", "shiny", "dull", "dull", "shiny"], dtype=object)
+    pd.crosstab(a, [b, c], rownames=["a"], colnames=["b", "c"])
+
+
+If :func:`~pandas.crosstab` receives only two :class:`Series`, it will provide a frequency table.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {"A": [1, 2, 2, 2, 2], "B": [3, 3, 4, 4, 4], "C": [1, 1, np.nan, 1, 1]}
+    )
+    df
+
+    pd.crosstab(df["A"], df["B"])
+
+:func:`~pandas.crosstab` can also summarize to :class:`Categorical` data.
+
+.. ipython:: python
+
+    foo = pd.Categorical(["a", "b"], categories=["a", "b", "c"])
+    bar = pd.Categorical(["d", "e"], categories=["d", "e", "f"])
+    pd.crosstab(foo, bar)
+
+For :class:`Categorical` data, to include **all** of data categories even if the actual data does
+not contain any instances of a particular category, use ``dropna=False``.
+
+.. ipython:: python
+
+    pd.crosstab(foo, bar, dropna=False)
+
+Normalization
+~~~~~~~~~~~~~
+
+Frequency tables can also be normalized to show percentages rather than counts
+using the ``normalize`` argument:
+
+.. ipython:: python
+
+   pd.crosstab(df["A"], df["B"], normalize=True)
+
+``normalize`` can also normalize values within each row or within each column:
+
+.. ipython:: python
+
+   pd.crosstab(df["A"], df["B"], normalize="columns")
+
+:func:`~pandas.crosstab` can also accept a third :class:`Series` and an aggregation function
+(``aggfunc``) that will be applied to the values of the third :class:`Series` within
+each group defined by the first two :class:`Series`:
+
+.. ipython:: python
+
+   pd.crosstab(df["A"], df["B"], values=df["C"], aggfunc="sum")
+
+Adding margins
+~~~~~~~~~~~~~~
+
+``margins=True`` will add a row and column with an ``All`` label with partial group aggregates
+across the categories on the rows and columns:
+
+.. ipython:: python
+
+   pd.crosstab(
+       df["A"], df["B"], values=df["C"], aggfunc="sum", normalize=True, margins=True
+   )
+
+.. _reshaping.tile:
+.. _reshaping.tile.cut:
+
+:func:`~pandas.cut`
+-------------------
+
+The :func:`~pandas.cut` function computes groupings for the values of the input
+array and is often used to transform continuous variables to discrete or
+categorical variables:
+
+
+An integer ``bins`` will form equal-width bins.
+
+.. ipython:: python
+
+   ages = np.array([10, 15, 13, 12, 23, 25, 28, 59, 60])
+
+   pd.cut(ages, bins=3)
+
+A list of ordered bin edges will assign an interval for each variable.
+
+.. ipython:: python
+
+   pd.cut(ages, bins=[0, 18, 35, 70])
+
+If the ``bins`` keyword is an :class:`IntervalIndex`, then these will be
+used to bin the passed data.
+
+.. ipython:: python
+
+   pd.cut(ages, bins=pd.IntervalIndex.from_breaks([0, 40, 70]))
+
+.. _reshaping.factorize:
+
+:func:`~pandas.factorize`
+-------------------------
+
+:func:`~pandas.factorize` encodes 1 dimensional values into integer labels. Missing values
+are encoded as ``-1``.
+
+.. ipython:: python
+
+   x = pd.Series(["A", "A", np.nan, "B", 3.14, np.inf])
+   x
+   labels, uniques = pd.factorize(x)
+   labels
+   uniques
+
+:class:`Categorical` will similarly encode 1 dimensional values for further
+categorical operations
+
+.. ipython:: python
+
+   pd.Categorical(x)
diff --git a/pandas/doc/source/user_guide/scale.rst b/pandas/doc/source/user_guide/scale.rst
new file mode 100644
index 0000000000000..d12993f7ead4b
--- /dev/null
+++ b/pandas/doc/source/user_guide/scale.rst
@@ -0,0 +1,228 @@
+.. _scale:
+
+*************************
+Scaling to large datasets
+*************************
+
+pandas provides data structures for in-memory analytics, which makes using pandas
+to analyze datasets that are larger than memory somewhat tricky. Even datasets
+that are a sizable fraction of memory become unwieldy, as some pandas operations need
+to make intermediate copies.
+
+This document provides a few recommendations for scaling your analysis to larger datasets.
+It's a complement to :ref:`enhancingperf`, which focuses on speeding up analysis
+for datasets that fit in memory.
+
+Load less data
+--------------
+
+Suppose our raw dataset on disk has many columns.
+
+.. ipython:: python
+   :okwarning:
+
+   import pandas as pd
+   import numpy as np
+
+   def make_timeseries(start="2000-01-01", end="2000-12-31", freq="1D", seed=None):
+       index = pd.date_range(start=start, end=end, freq=freq, name="timestamp")
+       n = len(index)
+       state = np.random.RandomState(seed)
+       columns = {
+           "name": state.choice(["Alice", "Bob", "Charlie"], size=n),
+           "id": state.poisson(1000, size=n),
+           "x": state.rand(n) * 2 - 1,
+           "y": state.rand(n) * 2 - 1,
+       }
+       df = pd.DataFrame(columns, index=index, columns=sorted(columns))
+       if df.index[-1] == end:
+           df = df.iloc[:-1]
+       return df
+
+   timeseries = [
+       make_timeseries(freq="1min", seed=i).rename(columns=lambda x: f"{x}_{i}")
+       for i in range(10)
+   ]
+   ts_wide = pd.concat(timeseries, axis=1)
+   ts_wide.head()
+   ts_wide.to_parquet("timeseries_wide.parquet")
+
+To load the columns we want, we have two options.
+Option 1 loads in all the data and then filters to what we need.
+
+.. ipython:: python
+
+   columns = ["id_0", "name_0", "x_0", "y_0"]
+
+   pd.read_parquet("timeseries_wide.parquet")[columns]
+
+Option 2 only loads the columns we request.
+
+.. ipython:: python
+
+   pd.read_parquet("timeseries_wide.parquet", columns=columns)
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("timeseries_wide.parquet")
+
+If we were to measure the memory usage of the two calls, we'd see that specifying
+``columns`` uses about 1/10th the memory in this case.
+
+With :func:`pandas.read_csv`, you can specify ``usecols`` to limit the columns
+read into memory. Not all file formats that can be read by pandas provide an option
+to read a subset of columns.
+
+Use efficient datatypes
+-----------------------
+
+The default pandas data types are not the most memory efficient. This is
+especially true for text data columns with relatively few unique values (commonly
+referred to as "low-cardinality" data). By using more efficient data types, you
+can store larger datasets in memory.
+
+.. ipython:: python
+   :okwarning:
+
+   ts = make_timeseries(freq="30s", seed=0)
+   ts.to_parquet("timeseries.parquet")
+   ts = pd.read_parquet("timeseries.parquet")
+   ts
+
+.. ipython:: python
+   :suppress:
+
+   os.remove("timeseries.parquet")
+
+Now, let's inspect the data types and memory usage to see where we should focus our
+attention.
+
+.. ipython:: python
+
+   ts.dtypes
+
+.. ipython:: python
+
+   ts.memory_usage(deep=True)  # memory usage in bytes
+
+
+The ``name`` column is taking up much more memory than any other. It has just a
+few unique values, so it's a good candidate for converting to a
+:class:`pandas.Categorical`. With a :class:`pandas.Categorical`, we store each unique name once and use
+space-efficient integers to know which specific name is used in each row.
+
+
+.. ipython:: python
+
+   ts2 = ts.copy()
+   ts2["name"] = ts2["name"].astype("category")
+   ts2.memory_usage(deep=True)
+
+We can go a bit further and downcast the numeric columns to their smallest types
+using :func:`pandas.to_numeric`.
+
+.. ipython:: python
+
+   ts2["id"] = pd.to_numeric(ts2["id"], downcast="unsigned")
+   ts2[["x", "y"]] = ts2[["x", "y"]].apply(pd.to_numeric, downcast="float")
+   ts2.dtypes
+
+.. ipython:: python
+
+   ts2.memory_usage(deep=True)
+
+.. ipython:: python
+
+   reduction = ts2.memory_usage(deep=True).sum() / ts.memory_usage(deep=True).sum()
+   print(f"{reduction:0.2f}")
+
+In all, we've reduced the in-memory footprint of this dataset to 1/5 of its
+original size.
+
+See :ref:`categorical` for more on :class:`pandas.Categorical` and :ref:`basics.dtypes`
+for an overview of all of pandas' dtypes.
+
+Use chunking
+------------
+
+Some workloads can be achieved with chunking by splitting a large problem into a bunch of small problems. For example,
+converting an individual CSV file into a Parquet file and repeating that for each file in a directory. As long as each chunk
+fits in memory, you can work with datasets that are much larger than memory.
+
+.. note::
+
+   Chunking works well when the operation you're performing requires zero or minimal
+   coordination between chunks. For more complicated workflows, you're better off
+   :ref:`using other libraries <scale.other_libraries>`.
+
+Suppose we have an even larger "logical dataset" on disk that's a directory of parquet
+files. Each file in the directory represents a different year of the entire dataset.
+
+.. ipython:: python
+   :okwarning:
+
+   import pathlib
+
+   N = 12
+   starts = [f"20{i:>02d}-01-01" for i in range(N)]
+   ends = [f"20{i:>02d}-12-13" for i in range(N)]
+
+   pathlib.Path("data/timeseries").mkdir(exist_ok=True)
+
+   for i, (start, end) in enumerate(zip(starts, ends)):
+       ts = make_timeseries(start=start, end=end, freq="1min", seed=i)
+       ts.to_parquet(f"data/timeseries/ts-{i:0>2d}.parquet")
+
+
+::
+
+   data
+   └── timeseries
+       ├── ts-00.parquet
+       ├── ts-01.parquet
+       ├── ts-02.parquet
+       ├── ts-03.parquet
+       ├── ts-04.parquet
+       ├── ts-05.parquet
+       ├── ts-06.parquet
+       ├── ts-07.parquet
+       ├── ts-08.parquet
+       ├── ts-09.parquet
+       ├── ts-10.parquet
+       └── ts-11.parquet
+
+Now we'll implement an out-of-core :meth:`pandas.Series.value_counts`. The peak memory usage of this
+workflow is the single largest chunk, plus a small series storing the unique value
+counts up to this point. As long as each individual file fits in memory, this will
+work for arbitrary-sized datasets.
+
+.. ipython:: python
+
+   %%time
+   files = pathlib.Path("data/timeseries/").glob("ts*.parquet")
+   counts = pd.Series(dtype=int)
+   for path in files:
+       df = pd.read_parquet(path)
+       counts = counts.add(df["name"].value_counts(), fill_value=0)
+   counts.astype(int)
+
+Some readers, like :meth:`pandas.read_csv`, offer parameters to control the
+``chunksize`` when reading a single file.
+
+Manually chunking is an OK option for workflows that don't
+require too sophisticated of operations. Some operations, like :meth:`pandas.DataFrame.groupby`, are
+much harder to do chunkwise. In these cases, you may be better switching to a
+different library that implements these out-of-core algorithms for you.
+
+.. _scale.other_libraries:
+
+Use Other Libraries
+-------------------
+
+There are other libraries which provide similar APIs to pandas and work nicely with pandas DataFrame,
+and can give you the ability to scale your large dataset processing and analytics
+by parallel runtime, distributed memory, clustering, etc. You can find more information
+in `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html#out-of-core>`_.
diff --git a/pandas/doc/source/user_guide/sparse.rst b/pandas/doc/source/user_guide/sparse.rst
new file mode 100644
index 0000000000000..25bcb8bcc0c93
--- /dev/null
+++ b/pandas/doc/source/user_guide/sparse.rst
@@ -0,0 +1,273 @@
+.. _sparse:
+
+{{ header }}
+
+**********************
+Sparse data structures
+**********************
+
+pandas provides data structures for efficiently storing sparse data.
+These are not necessarily sparse in the typical "mostly 0". Rather, you can view these
+objects as being "compressed" where any data matching a specific value (``NaN`` / missing value, though any value
+can be chosen, including 0) is omitted. The compressed values are not actually stored in the array.
+
+.. ipython:: python
+
+   arr = np.random.randn(10)
+   arr[2:-2] = np.nan
+   ts = pd.Series(pd.arrays.SparseArray(arr))
+   ts
+
+Notice the dtype, ``Sparse[float64, nan]``. The ``nan`` means that elements in the
+array that are ``nan`` aren't actually stored, only the non-``nan`` elements are.
+Those non-``nan`` elements have a ``float64`` dtype.
+
+The sparse objects exist for memory efficiency reasons. Suppose you had a
+large, mostly NA :class:`DataFrame`:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10000, 4))
+   df.iloc[:9998] = np.nan
+   sdf = df.astype(pd.SparseDtype("float", np.nan))
+   sdf.head()
+   sdf.dtypes
+   sdf.sparse.density
+
+As you can see, the density (% of values that have not been "compressed") is
+extremely low. This sparse object takes up much less memory on disk (pickled)
+and in the Python interpreter.
+
+.. ipython:: python
+
+   'dense : {:0.2f} bytes'.format(df.memory_usage().sum() / 1e3)
+   'sparse: {:0.2f} bytes'.format(sdf.memory_usage().sum() / 1e3)
+
+Functionally, their behavior should be nearly
+identical to their dense counterparts.
+
+.. _sparse.array:
+
+SparseArray
+-----------
+
+:class:`arrays.SparseArray` is a :class:`~pandas.api.extensions.ExtensionArray`
+for storing an array of sparse values (see :ref:`basics.dtypes` for more
+on extension arrays). It is a 1-dimensional ndarray-like object storing
+only values distinct from the ``fill_value``:
+
+.. ipython:: python
+
+   arr = np.random.randn(10)
+   arr[2:5] = np.nan
+   arr[7:8] = np.nan
+   sparr = pd.arrays.SparseArray(arr)
+   sparr
+
+A sparse array can be converted to a regular (dense) ndarray with :meth:`numpy.asarray`
+
+.. ipython:: python
+
+   np.asarray(sparr)
+
+
+.. _sparse.dtype:
+
+SparseDtype
+-----------
+
+The :attr:`SparseArray.dtype` property stores two pieces of information
+
+1. The dtype of the non-sparse values
+2. The scalar fill value
+
+
+.. ipython:: python
+
+   sparr.dtype
+
+
+A :class:`SparseDtype` may be constructed by passing only a dtype
+
+.. ipython:: python
+
+   pd.SparseDtype(np.dtype('datetime64[ns]'))
+
+in which case a default fill value will be used (for NumPy dtypes this is often the
+"missing" value for that dtype). To override this default an explicit fill value may be
+passed instead
+
+.. ipython:: python
+
+   pd.SparseDtype(np.dtype('datetime64[ns]'),
+                  fill_value=pd.Timestamp('2017-01-01'))
+
+Finally, the string alias ``'Sparse[dtype]'`` may be used to specify a sparse dtype
+in many places
+
+.. ipython:: python
+
+   pd.array([1, 0, 0, 2], dtype='Sparse[int]')
+
+.. _sparse.accessor:
+
+Sparse accessor
+---------------
+
+pandas provides a ``.sparse`` accessor, similar to ``.str`` for string data, ``.cat``
+for categorical data, and ``.dt`` for datetime-like data. This namespace provides
+attributes and methods that are specific to sparse data.
+
+.. ipython:: python
+
+   s = pd.Series([0, 0, 1, 2], dtype="Sparse[int]")
+   s.sparse.density
+   s.sparse.fill_value
+
+This accessor is available only on data with ``SparseDtype``, and on the :class:`Series`
+class itself for creating a Series with sparse data from a scipy COO matrix with.
+
+A ``.sparse`` accessor has been added for :class:`DataFrame` as well.
+See :ref:`api.frame.sparse` for more.
+
+.. _sparse.calculation:
+
+Sparse calculation
+------------------
+
+You can apply NumPy `ufuncs <https://numpy.org/doc/stable/reference/ufuncs.html>`_
+to :class:`arrays.SparseArray` and get a :class:`arrays.SparseArray` as a result.
+
+.. ipython:: python
+
+   arr = pd.arrays.SparseArray([1., np.nan, np.nan, -2., np.nan])
+   np.abs(arr)
+
+
+The *ufunc* is also applied to ``fill_value``. This is needed to get
+the correct dense result.
+
+.. ipython:: python
+
+   arr = pd.arrays.SparseArray([1., -1, -1, -2., -1], fill_value=-1)
+   np.abs(arr)
+   np.abs(arr).to_dense()
+
+
+**Conversion**
+
+To convert data from sparse to dense, use the ``.sparse`` accessors
+
+.. ipython:: python
+
+   sdf.sparse.to_dense()
+
+From dense to sparse, use :meth:`DataFrame.astype` with a :class:`SparseDtype`.
+
+.. ipython:: python
+
+   dense = pd.DataFrame({"A": [1, 0, 0, 1]})
+   dtype = pd.SparseDtype(int, fill_value=0)
+   dense.astype(dtype)
+
+
+.. _sparse.scipysparse:
+
+Interaction with *scipy.sparse*
+-------------------------------
+
+Use :meth:`DataFrame.sparse.from_spmatrix` to create a :class:`DataFrame` with sparse values from a sparse matrix.
+
+.. ipython:: python
+
+   from scipy.sparse import csr_matrix
+
+   arr = np.random.random(size=(1000, 5))
+   arr[arr < .9] = 0
+
+   sp_arr = csr_matrix(arr)
+   sp_arr
+
+   sdf = pd.DataFrame.sparse.from_spmatrix(sp_arr)
+   sdf.head()
+   sdf.dtypes
+
+All sparse formats are supported, but matrices that are not in :mod:`COOrdinate <scipy.sparse>` format will be converted, copying data as needed.
+To convert back to sparse SciPy matrix in COO format, you can use the :meth:`DataFrame.sparse.to_coo` method:
+
+.. ipython:: python
+
+   sdf.sparse.to_coo()
+
+:meth:`Series.sparse.to_coo` is implemented for transforming a :class:`Series` with sparse values indexed by a :class:`MultiIndex` to a :class:`scipy.sparse.coo_matrix`.
+
+The method requires a :class:`MultiIndex` with two or more levels.
+
+.. ipython:: python
+
+   s = pd.Series([3.0, np.nan, 1.0, 3.0, np.nan, np.nan])
+   s.index = pd.MultiIndex.from_tuples(
+       [
+           (1, 2, "a", 0),
+           (1, 2, "a", 1),
+           (1, 1, "b", 0),
+           (1, 1, "b", 1),
+           (2, 1, "b", 0),
+           (2, 1, "b", 1),
+       ],
+       names=["A", "B", "C", "D"],
+   )
+   ss = s.astype('Sparse')
+   ss
+
+In the example below, we transform the :class:`Series` to a sparse representation of a 2-d array by specifying that the first and second ``MultiIndex`` levels define labels for the rows and the third and fourth levels define labels for the columns. We also specify that the column and row labels should be sorted in the final sparse representation.
+
+.. ipython:: python
+
+   A, rows, columns = ss.sparse.to_coo(
+       row_levels=["A", "B"], column_levels=["C", "D"], sort_labels=True
+   )
+
+   A
+   A.todense()
+   rows
+   columns
+
+Specifying different row and column labels (and not sorting them) yields a different sparse matrix:
+
+.. ipython:: python
+
+   A, rows, columns = ss.sparse.to_coo(
+       row_levels=["A", "B", "C"], column_levels=["D"], sort_labels=False
+   )
+
+   A
+   A.todense()
+   rows
+   columns
+
+A convenience method :meth:`Series.sparse.from_coo` is implemented for creating a :class:`Series` with sparse values from a ``scipy.sparse.coo_matrix``.
+
+.. ipython:: python
+
+   from scipy import sparse
+   A = sparse.coo_matrix(([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])), shape=(3, 4))
+   A
+   A.todense()
+
+The default behaviour (with ``dense_index=False``) simply returns a :class:`Series` containing
+only the non-null entries.
+
+.. ipython:: python
+
+   ss = pd.Series.sparse.from_coo(A)
+   ss
+
+Specifying ``dense_index=True`` will result in an index that is the Cartesian product of the
+row and columns coordinates of the matrix. Note that this will consume a significant amount of memory
+(relative to ``dense_index=False``) if the sparse matrix is large (and sparse) enough.
+
+.. ipython:: python
+
+   ss_dense = pd.Series.sparse.from_coo(A, dense_index=True)
+   ss_dense
diff --git a/pandas/doc/source/user_guide/style.ipynb b/pandas/doc/source/user_guide/style.ipynb
new file mode 100644
index 0000000000000..9cda1486eb48b
--- /dev/null
+++ b/pandas/doc/source/user_guide/style.ipynb
@@ -0,0 +1,2212 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Table Visualization\n",
+    "\n",
+    "This section demonstrates visualization of tabular data using the [Styler][styler]\n",
+    "class. For information on visualization with charting please see [Chart Visualization][viz]. This document is written as a Jupyter Notebook, and can be viewed or downloaded [here][download].\n",
+    "\n",
+    "## Styler Object and Customising the Display\n",
+    "Styling and output display customisation should be performed **after** the data in a DataFrame has been processed. The Styler is **not** dynamically updated if further changes to the DataFrame are made. The `DataFrame.style` attribute is a property that returns a [Styler][styler] object. It has a `_repr_html_` method defined on it so it is rendered automatically in Jupyter Notebook.\n",
+    "\n",
+    "The Styler, which can be used for large data but is primarily designed for small data, currently has the ability to output to these formats:\n",
+    "\n",
+    "  - HTML\n",
+    "  - LaTeX\n",
+    "  - String (and CSV by extension)\n",
+    "  - Excel\n",
+    "  - (JSON is not currently available)\n",
+    "\n",
+    "The first three of these have display customisation methods designed to format and customise the output. These include:\n",
+    "\n",
+    "  - Formatting values, the index and columns headers, using [.format()][formatfunc] and [.format_index()][formatfuncindex],\n",
+    "  - Renaming the index or column header labels, using [.relabel_index()][relabelfunc]\n",
+    "  - Hiding certain columns, the index and/or column headers, or index names, using [.hide()][hidefunc]\n",
+    "  - Concatenating similar DataFrames, using [.concat()][concatfunc]\n",
+    "  \n",
+    "[styler]: ../reference/api/pandas.io.formats.style.Styler.rst\n",
+    "[viz]: visualization.rst\n",
+    "[download]: https://nbviewer.org/github/pandas-dev/pandas/blob/main/doc/source/user_guide/style.ipynb\n",
+    "[format]: https://docs.python.org/3/library/string.html#format-specification-mini-language\n",
+    "[formatfunc]: ../reference/api/pandas.io.formats.style.Styler.format.rst\n",
+    "[formatfuncindex]: ../reference/api/pandas.io.formats.style.Styler.format_index.rst\n",
+    "[relabelfunc]: ../reference/api/pandas.io.formats.style.Styler.relabel_index.rst\n",
+    "[hidefunc]: ../reference/api/pandas.io.formats.style.Styler.hide.rst\n",
+    "[concatfunc]: ../reference/api/pandas.io.formats.style.Styler.concat.rst"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Formatting the Display\n",
+    "\n",
+    "### Formatting Values\n",
+    "\n",
+    "The [Styler][styler] distinguishes the *display* value from the *actual* value, in both data values and index or columns headers. To control the display value, the text is printed in each cell as a string, and we can use the [.format()][formatfunc] and [.format_index()][formatfuncindex] methods to manipulate this according to a [format spec string][format] or a callable that takes a single value and returns a string. It is possible to define this for the whole table, or index, or for individual columns, or MultiIndex levels. We can also overwrite index names.\n",
+    "\n",
+    "Additionally, the format function has a **precision** argument to specifically help format floats, as well as **decimal** and **thousands** separators to support other locales, an **na_rep** argument to display missing data, and an **escape** and **hyperlinks** arguments to help displaying safe-HTML or safe-LaTeX. The default formatter is configured to adopt pandas' global options such as `styler.format.precision` option, controllable using `with pd.option_context('format.precision', 2):`\n",
+    "\n",
+    "[styler]: ../reference/api/pandas.io.formats.style.Styler.rst\n",
+    "[format]: https://docs.python.org/3/library/string.html#format-specification-mini-language\n",
+    "[formatfunc]: ../reference/api/pandas.io.formats.style.Styler.format.rst\n",
+    "[formatfuncindex]: ../reference/api/pandas.io.formats.style.Styler.format_index.rst\n",
+    "[relabelfunc]: ../reference/api/pandas.io.formats.style.Styler.relabel_index.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "\n",
+    "df = pd.DataFrame(\n",
+    "    {\"strings\": [\"Adam\", \"Mike\"], \"ints\": [1, 3], \"floats\": [1.123, 1000.23]}\n",
+    ")\n",
+    "df.style.format(precision=3, thousands=\".\", decimal=\",\").format_index(\n",
+    "    str.upper, axis=1\n",
+    ").relabel_index([\"row 1\", \"row 2\"], axis=0)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Using Styler to manipulate the display is a useful feature because maintaining the indexing and data values for other purposes gives greater control. You do not have to overwrite your DataFrame to display it how you like. Here is a more comprehensive example of using the formatting functions whilst still relying on the underlying data for indexing and calculations."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "weather_df = pd.DataFrame(\n",
+    "    np.random.default_rng().random((10, 2)) * 5,\n",
+    "    index=pd.date_range(start=\"2021-01-01\", periods=10),\n",
+    "    columns=[\"Tokyo\", \"Beijing\"],\n",
+    ")\n",
+    "\n",
+    "\n",
+    "def rain_condition(v):\n",
+    "    if v < 1.75:\n",
+    "        return \"Dry\"\n",
+    "    elif v < 2.75:\n",
+    "        return \"Rain\"\n",
+    "    return \"Heavy Rain\"\n",
+    "\n",
+    "\n",
+    "def make_pretty(styler):\n",
+    "    styler.set_caption(\"Weather Conditions\")\n",
+    "    styler.format(rain_condition)\n",
+    "    styler.format_index(lambda v: v.strftime(\"%A\"))\n",
+    "    styler.background_gradient(axis=None, vmin=1, vmax=5, cmap=\"YlGnBu\")\n",
+    "    return styler\n",
+    "\n",
+    "\n",
+    "weather_df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "weather_df.loc[\"2021-01-04\":\"2021-01-08\"].style.pipe(make_pretty)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Hiding Data\n",
+    "\n",
+    "The index and column headers can be completely hidden, as well subselecting rows or columns that one wishes to exclude. Both these options are performed using the same methods.\n",
+    "\n",
+    "The index can be hidden from rendering by calling [.hide()][hideidx] without any arguments, which might be useful if your index is integer based. Similarly column headers can be hidden by calling [.hide(axis=\"columns\")][hideidx] without any further arguments.\n",
+    "\n",
+    "Specific rows or columns can be hidden from rendering by calling the same [.hide()][hideidx] method and passing in a row/column label, a list-like or a slice of row/column labels to for the ``subset`` argument.\n",
+    "\n",
+    "Hiding does not change the integer arrangement of CSS classes, e.g. hiding the first two columns of a DataFrame means the column class indexing will still start at `col2`, since `col0` and `col1` are simply ignored.\n",
+    "\n",
+    "[hideidx]: ../reference/api/pandas.io.formats.style.Styler.hide.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.DataFrame(np.random.default_rng().standard_normal((5, 5)))\n",
+    "df.style.hide(subset=[0, 2, 4], axis=0).hide(subset=[0, 2, 4], axis=1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "To invert the function to a **show** functionality it is best practice to compose a list of hidden items."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "show = [0, 2, 4]\n",
+    "df.style.hide([row for row in df.index if row not in show], axis=0).hide(\n",
+    "    [col for col in df.columns if col not in show], axis=1\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Concatenating DataFrame Outputs\n",
+    "\n",
+    "Two or more Stylers can be concatenated together provided they share the same columns. This is very useful for showing summary statistics for a DataFrame, and is often used in combination with DataFrame.agg.\n",
+    "\n",
+    "Since the objects concatenated are Stylers they can independently be styled as will be shown below and their concatenation preserves those styles."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "summary_styler = (\n",
+    "    df.agg([\"sum\", \"mean\"]).style.format(precision=3).relabel_index([\"Sum\", \"Average\"])\n",
+    ")\n",
+    "df.style.format(precision=1).concat(summary_styler)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Styler Object and HTML \n",
+    "\n",
+    "The [Styler][styler] was originally constructed to support the wide array of HTML formatting options. Its HTML output creates an HTML `<table>` and leverages CSS styling language to manipulate many parameters including colors, fonts, borders, background, etc. See [here][w3schools] for more information on styling HTML tables. This allows a lot of flexibility out of the box, and even enables web developers to integrate DataFrames into their existing user interface designs.\n",
+    "\n",
+    "Below we demonstrate the default output, which looks very similar to the standard DataFrame HTML representation. But the HTML here has already attached some CSS classes to each cell, even if we haven't yet created any styles. We can view these by calling the  [.to_html()][tohtml] method, which returns the raw HTML as string, which is useful for further processing or adding to a file - read on in [More about CSS and HTML](#More-About-CSS-and-HTML). This section will also provide a walkthrough for how to convert this default output to represent a DataFrame output that is more communicative. For example how we can build `s`:\n",
+    "\n",
+    "[tohtml]: ../reference/api/pandas.io.formats.style.Styler.to_html.rst\n",
+    "\n",
+    "[styler]: ../reference/api/pandas.io.formats.style.Styler.rst\n",
+    "[w3schools]: https://www.w3schools.com/html/html_tables.asp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "idx = pd.Index([\"Tumour (Positive)\", \"Non-Tumour (Negative)\"], name=\"Actual Label:\")\n",
+    "cols = pd.MultiIndex.from_product(\n",
+    "    [[\"Decision Tree\", \"Regression\", \"Random\"], [\"Tumour\", \"Non-Tumour\"]],\n",
+    "    names=[\"Model:\", \"Predicted:\"],\n",
+    ")\n",
+    "df = pd.DataFrame(\n",
+    "    [[38.0, 2.0, 18.0, 22.0, 21, np.nan], [19, 439, 6, 452, 226, 232]],\n",
+    "    index=idx,\n",
+    "    columns=cols,\n",
+    ")\n",
+    "df.style"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to just create the below example: code is covered throughout the guide.\n",
+    "s = (\n",
+    "    df.style.hide([(\"Random\", \"Tumour\"), (\"Random\", \"Non-Tumour\")], axis=\"columns\")\n",
+    "    .format(\"{:.0f}\")\n",
+    "    .set_table_styles(\n",
+    "        [\n",
+    "            {\"selector\": \"\", \"props\": \"border-collapse: separate;\"},\n",
+    "            {\"selector\": \"caption\", \"props\": \"caption-side: bottom; font-size:1.3em;\"},\n",
+    "            {\n",
+    "                \"selector\": \".index_name\",\n",
+    "                \"props\": \"font-style: italic; color: darkgrey; font-weight:normal;\",\n",
+    "            },\n",
+    "            {\n",
+    "                \"selector\": \"th:not(.index_name)\",\n",
+    "                \"props\": \"background-color: #000066; color: white;\",\n",
+    "            },\n",
+    "            {\"selector\": \"th.col_heading\", \"props\": \"text-align: center;\"},\n",
+    "            {\"selector\": \"th.col_heading.level0\", \"props\": \"font-size: 1.5em;\"},\n",
+    "            {\"selector\": \"th.col2\", \"props\": \"border-left: 1px solid white;\"},\n",
+    "            {\"selector\": \".col2\", \"props\": \"border-left: 1px solid #000066;\"},\n",
+    "            {\"selector\": \"td\", \"props\": \"text-align: center; font-weight:bold;\"},\n",
+    "            {\"selector\": \".true\", \"props\": \"background-color: #e6ffe6;\"},\n",
+    "            {\"selector\": \".false\", \"props\": \"background-color: #ffe6e6;\"},\n",
+    "            {\"selector\": \".border-red\", \"props\": \"border: 2px dashed red;\"},\n",
+    "            {\"selector\": \".border-green\", \"props\": \"border: 2px dashed green;\"},\n",
+    "            {\"selector\": \"td:hover\", \"props\": \"background-color: #ffffb3;\"},\n",
+    "        ]\n",
+    "    )\n",
+    "    .set_td_classes(\n",
+    "        pd.DataFrame(\n",
+    "            [\n",
+    "                [\"true border-green\", \"false\", \"true\", \"false border-red\", \"\", \"\"],\n",
+    "                [\"false\", \"true\", \"false\", \"true\", \"\", \"\"],\n",
+    "            ],\n",
+    "            index=df.index,\n",
+    "            columns=df.columns,\n",
+    "        )\n",
+    "    )\n",
+    "    .set_caption(\"Confusion matrix for multiple cancer prediction models.\")\n",
+    "    .set_tooltips(\n",
+    "        pd.DataFrame(\n",
+    "            [\n",
+    "                [\n",
+    "                    \"This model has a very strong true positive rate\",\n",
+    "                    \"\",\n",
+    "                    \"\",\n",
+    "                    \"This model's total number of false negatives is too high\",\n",
+    "                    \"\",\n",
+    "                    \"\",\n",
+    "                ],\n",
+    "                [\"\", \"\", \"\", \"\", \"\", \"\"],\n",
+    "            ],\n",
+    "            index=df.index,\n",
+    "            columns=df.columns,\n",
+    "        ),\n",
+    "        css_class=\"pd-tt\",\n",
+    "        props=\"visibility: hidden; \"\n",
+    "        \"position: absolute; z-index: 1; \"\n",
+    "        \"border: 1px solid #000066;\"\n",
+    "        \"background-color: white; color: #000066; font-size: 0.8em;\"\n",
+    "        \"transform: translate(0px, -24px); padding: 0.6em; border-radius: 0.5em;\",\n",
+    "    )\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The first step we have taken is the create the Styler object from the DataFrame and then select the range of interest by hiding unwanted columns with [.hide()][hideidx].\n",
+    "\n",
+    "[hideidx]: ../reference/api/pandas.io.formats.style.Styler.hide.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s = df.style.format(\"{:.0f}\").hide(\n",
+    "    [(\"Random\", \"Tumour\"), (\"Random\", \"Non-Tumour\")], axis=\"columns\"\n",
+    ")\n",
+    "s"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_hide\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Methods to Add Styles\n",
+    "\n",
+    "There are **3 primary methods of adding custom CSS styles** to [Styler][styler]:\n",
+    "\n",
+    "- Using [.set_table_styles()][table] to control broader areas of the table with specified internal CSS. Although table styles allow the flexibility to add CSS selectors and properties controlling all individual parts of the table, they are unwieldy for individual cell specifications. Also, note that table styles cannot be exported to Excel. \n",
+    "- Using [.set_td_classes()][td_class] to directly link either external CSS classes to your data cells or link the internal CSS classes created by [.set_table_styles()][table]. See [here](#Setting-Classes-and-Linking-to-External-CSS). These cannot be used on column header rows or indexes, and also won't export to Excel. \n",
+    "- Using the [.apply()][apply] and [.map()][map] functions to add direct internal CSS to specific data cells. See [here](#Styler-Functions). As of v1.4.0 there are also methods that work directly on column header rows or indexes: [.apply_index()][applyindex] and [.map_index()][mapindex]. Note that only these methods add styles that will export to Excel. These methods work in a similar way to [DataFrame.apply()][dfapply] and [DataFrame.map()][dfmap].\n",
+    "\n",
+    "[table]: ../reference/api/pandas.io.formats.style.Styler.set_table_styles.rst\n",
+    "[styler]: ../reference/api/pandas.io.formats.style.Styler.rst\n",
+    "[td_class]: ../reference/api/pandas.io.formats.style.Styler.set_td_classes.rst\n",
+    "[apply]: ../reference/api/pandas.io.formats.style.Styler.apply.rst\n",
+    "[map]: ../reference/api/pandas.io.formats.style.Styler.map.rst\n",
+    "[applyindex]: ../reference/api/pandas.io.formats.style.Styler.apply_index.rst\n",
+    "[mapindex]: ../reference/api/pandas.io.formats.style.Styler.map_index.rst\n",
+    "[dfapply]: ../reference/api/pandas.DataFrame.apply.rst\n",
+    "[dfmap]: ../reference/api/pandas.DataFrame.map.rst"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Table Styles\n",
+    "\n",
+    "Table styles are flexible enough to control all individual parts of the table, including column headers and indexes. \n",
+    "However, they can be unwieldy to type for individual data cells or for any kind of conditional formatting, so we recommend that table styles are used for broad styling, such as entire rows or columns at a time.\n",
+    "\n",
+    "Table styles are also used to control features which can apply to the whole table at once such as creating a generic hover functionality. The `:hover` pseudo-selector, as well as other pseudo-selectors, can only be used this way.\n",
+    "\n",
+    "To replicate the normal format of CSS selectors and properties (attribute value pairs), e.g. \n",
+    "\n",
+    "```\n",
+    "tr:hover {\n",
+    "  background-color: #ffff99;\n",
+    "}\n",
+    "```\n",
+    "\n",
+    "the necessary format to pass styles to [.set_table_styles()][table] is as a list of dicts, each with a CSS-selector tag and CSS-properties. Properties can either be a list of 2-tuples, or a regular CSS-string, for example:\n",
+    "\n",
+    "[table]: ../reference/api/pandas.io.formats.style.Styler.set_table_styles.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cell_hover = {  # for row hover use <tr> instead of <td>\n",
+    "    \"selector\": \"td:hover\",\n",
+    "    \"props\": [(\"background-color\", \"#ffffb3\")],\n",
+    "}\n",
+    "index_names = {\n",
+    "    \"selector\": \".index_name\",\n",
+    "    \"props\": \"font-style: italic; color: darkgrey; font-weight:normal;\",\n",
+    "}\n",
+    "headers = {\n",
+    "    \"selector\": \"th:not(.index_name)\",\n",
+    "    \"props\": \"background-color: #000066; color: white;\",\n",
+    "}\n",
+    "s.set_table_styles([cell_hover, index_names, headers])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_tab_styles1\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Next we just add a couple more styling artifacts targeting specific parts of the table. Be careful here, since we are *chaining methods* we need to explicitly instruct the method **not to** ``overwrite`` the existing styles."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s.set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \"th.col_heading\", \"props\": \"text-align: center;\"},\n",
+    "        {\"selector\": \"th.col_heading.level0\", \"props\": \"font-size: 1.5em;\"},\n",
+    "        {\"selector\": \"td\", \"props\": \"text-align: center; font-weight: bold;\"},\n",
+    "    ],\n",
+    "    overwrite=False,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_tab_styles2\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "As a convenience method (*since version 1.2.0*) we can also pass a **dict** to [.set_table_styles()][table] which contains row or column keys. Behind the scenes Styler just indexes the keys and adds relevant `.col<m>` or `.row<n>` classes as necessary to the given CSS selectors.\n",
+    "\n",
+    "[table]: ../reference/api/pandas.io.formats.style.Styler.set_table_styles.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s.set_table_styles(\n",
+    "    {\n",
+    "        (\"Regression\", \"Tumour\"): [\n",
+    "            {\"selector\": \"th\", \"props\": \"border-left: 1px solid white\"},\n",
+    "            {\"selector\": \"td\", \"props\": \"border-left: 1px solid #000066\"},\n",
+    "        ]\n",
+    "    },\n",
+    "    overwrite=False,\n",
+    "    axis=0,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"xyz01\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Setting Classes and Linking to External CSS\n",
+    "\n",
+    "If you have designed a website then it is likely you will already have an external CSS file that controls the styling of table and cell objects within it. You may want to use these native files rather than duplicate all the CSS in python (and duplicate any maintenance work).\n",
+    "\n",
+    "### Table Attributes\n",
+    "\n",
+    "It is very easy to add a `class` to the main `<table>` using [.set_table_attributes()][tableatt]. This method can also attach inline styles - read more in [CSS Hierarchies](#CSS-Hierarchies).\n",
+    "\n",
+    "[tableatt]: ../reference/api/pandas.io.formats.style.Styler.set_table_attributes.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "out = s.set_table_attributes('class=\"my-table-cls\"').to_html()\n",
+    "print(out[out.find(\"<table\") :][:109])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Data Cell CSS Classes\n",
+    "\n",
+    "*New in version 1.2.0*\n",
+    "\n",
+    "The [.set_td_classes()][tdclass] method accepts a DataFrame with matching indices and columns to the underlying [Styler][styler]'s DataFrame. That DataFrame will contain strings as css-classes to add to individual data cells: the `<td>` elements of the `<table>`. Rather than use external CSS we will create our classes internally and add them to table style. We will save adding the borders until the [section on tooltips](#Tooltips-and-Captions).\n",
+    "\n",
+    "[tdclass]: ../reference/api/pandas.io.formats.style.Styler.set_td_classes.rst\n",
+    "[styler]: ../reference/api/pandas.io.formats.style.Styler.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s.set_table_styles(\n",
+    "    [  # create internal CSS classes\n",
+    "        {\"selector\": \".true\", \"props\": \"background-color: #e6ffe6;\"},\n",
+    "        {\"selector\": \".false\", \"props\": \"background-color: #ffe6e6;\"},\n",
+    "    ],\n",
+    "    overwrite=False,\n",
+    ")\n",
+    "cell_color = pd.DataFrame(\n",
+    "    [[\"true \", \"false \", \"true \", \"false \"], [\"false \", \"true \", \"false \", \"true \"]],\n",
+    "    index=df.index,\n",
+    "    columns=df.columns[:4],\n",
+    ")\n",
+    "s.set_td_classes(cell_color)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_classes\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Styler Functions\n",
+    "\n",
+    "### Acting on Data\n",
+    "\n",
+    "We use the following methods to pass your style functions. Both of those methods take a function (and some other keyword arguments) and apply it to the DataFrame in a certain way, rendering CSS styles.\n",
+    "\n",
+    "- [.map()][map] (elementwise): accepts a function that takes a single value and returns a string with the CSS attribute-value pair.\n",
+    "- [.apply()][apply] (column-/row-/table-wise): accepts a function that takes a Series or DataFrame and returns a Series, DataFrame, or numpy array with an identical shape where each element is a string with a CSS attribute-value pair. This method passes each column or row of your DataFrame one-at-a-time or the entire table at once, depending on the `axis` keyword argument. For columnwise use `axis=0`, rowwise use `axis=1`, and for the entire table at once use `axis=None`.\n",
+    "\n",
+    "This method is powerful for applying multiple, complex logic to data cells. We create a new DataFrame to demonstrate this.\n",
+    "\n",
+    "[apply]: ../reference/api/pandas.io.formats.style.Styler.apply.rst\n",
+    "[map]: ../reference/api/pandas.io.formats.style.Styler.map.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2 = pd.DataFrame(\n",
+    "    np.random.default_rng(0).standard_normal((10, 4)), columns=[\"A\", \"B\", \"C\", \"D\"]\n",
+    ")\n",
+    "df2.style"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "For example we can build a function that colors text if it is negative, and chain this with a function that partially fades cells of negligible value. Since this looks at each element in turn we use ``map``."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def style_negative(v, props=\"\"):\n",
+    "    return props if v < 0 else None\n",
+    "\n",
+    "\n",
+    "s2 = df2.style.map(style_negative, props=\"color:red;\").map(\n",
+    "    lambda v: \"opacity: 20%;\" if (v < 0.3) and (v > -0.3) else None\n",
+    ")\n",
+    "s2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s2.set_uuid(\"after_applymap\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can also build a function that highlights the maximum value across rows, cols, and the DataFrame all at once. In this case we use ``apply``. Below we highlight the maximum in a column."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def highlight_max(s, props=\"\"):\n",
+    "    return np.where(s == np.nanmax(s.values), props, \"\")\n",
+    "\n",
+    "\n",
+    "s2.apply(highlight_max, props=\"color:white;background-color:darkblue\", axis=0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s2.set_uuid(\"after_apply\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can use the same function across the different axes, highlighting here the DataFrame maximum in purple, and row maximums in pink."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s2.apply(highlight_max, props=\"color:white;background-color:pink;\", axis=1).apply(\n",
+    "    highlight_max, props=\"color:white;background-color:purple\", axis=None\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s2.set_uuid(\"after_apply_again\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This last example shows how some styles have been overwritten by others. In general the most recent style applied is active but you can read more in the [section on CSS hierarchies](#CSS-Hierarchies). You can also apply these styles to more granular parts of the DataFrame - read more in section on [subset slicing](#Finer-Control-with-Slicing).\n",
+    "\n",
+    "It is possible to replicate some of this functionality using just classes but it can be more cumbersome. See [item 3)  of Optimization](#Optimization)\n",
+    "\n",
+    "<div class=\"alert alert-info\">\n",
+    "\n",
+    "*Debugging Tip*: If you're having trouble writing your style function, try just passing it into ``DataFrame.apply``. Internally, ``Styler.apply`` uses ``DataFrame.apply`` so the result should be the same, and with ``DataFrame.apply`` you will be able to inspect the CSS string output of your intended function in each cell.\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Acting on the Index and Column Headers\n",
+    "\n",
+    "Similar application is achieved for headers by using:\n",
+    "    \n",
+    "- [.map_index()][mapindex] (elementwise): accepts a function that takes a single value and returns a string with the CSS attribute-value pair.\n",
+    "- [.apply_index()][applyindex] (level-wise): accepts a function that takes a Series and returns a Series, or numpy array with an identical shape where each element is a string with a CSS attribute-value pair. This method passes each level of your Index one-at-a-time. To style the index use `axis=0` and to style the column headers use `axis=1`.\n",
+    "\n",
+    "You can select a `level` of a `MultiIndex` but currently no similar `subset` application is available for these methods.\n",
+    "\n",
+    "[applyindex]: ../reference/api/pandas.io.formats.style.Styler.apply_index.rst\n",
+    "[mapindex]: ../reference/api/pandas.io.formats.style.Styler.map_index.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s2.map_index(lambda v: \"color:pink;\" if v > 4 else \"color:darkblue;\", axis=0)\n",
+    "s2.apply_index(\n",
+    "    lambda s: np.where(s.isin([\"A\", \"B\"]), \"color:pink;\", \"color:darkblue;\"), axis=1\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Tooltips and Captions\n",
+    "\n",
+    "Table captions can be added with the [.set_caption()][caption] method. You can use table styles to control the CSS relevant to the caption.\n",
+    "\n",
+    "[caption]: ../reference/api/pandas.io.formats.style.Styler.set_caption.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s.set_caption(\n",
+    "    \"Confusion matrix for multiple cancer prediction models.\"\n",
+    ").set_table_styles(\n",
+    "    [{\"selector\": \"caption\", \"props\": \"caption-side: bottom; font-size:1.25em;\"}],\n",
+    "    overwrite=False,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_caption\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Adding tooltips (*since version 1.3.0*) can be done using the [.set_tooltips()][tooltips] method in the same way you can add CSS classes to data cells by providing a string based DataFrame with intersecting indices and columns. You don't have to specify a `css_class` name or any css `props` for the tooltips, since there are standard defaults, but the option is there if you want more visual control. \n",
+    "\n",
+    "[tooltips]: ../reference/api/pandas.io.formats.style.Styler.set_tooltips.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "tt = pd.DataFrame(\n",
+    "    [\n",
+    "        [\n",
+    "            \"This model has a very strong true positive rate\",\n",
+    "            \"This model's total number of false negatives is too high\",\n",
+    "        ]\n",
+    "    ],\n",
+    "    index=[\"Tumour (Positive)\"],\n",
+    "    columns=df.columns[[0, 3]],\n",
+    ")\n",
+    "s.set_tooltips(\n",
+    "    tt,\n",
+    "    props=\"visibility: hidden; position: absolute; z-index: 1; \"\n",
+    "    \"border: 1px solid #000066;\"\n",
+    "    \"background-color: white; color: #000066; font-size: 0.8em;\"\n",
+    "    \"transform: translate(0px, -24px); padding: 0.6em; \"\n",
+    "    \"border-radius: 0.5em;\",\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_tooltips\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The only thing left to do for our table is to add the highlighting borders to draw the audience attention to the tooltips. We will create internal CSS classes as before using table styles. **Setting classes always overwrites** so we need to make sure we add the previous classes."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "s.set_table_styles(\n",
+    "    [  # create internal CSS classes\n",
+    "        {\"selector\": \".border-red\", \"props\": \"border: 2px dashed red;\"},\n",
+    "        {\"selector\": \".border-green\", \"props\": \"border: 2px dashed green;\"},\n",
+    "    ],\n",
+    "    overwrite=False,\n",
+    ")\n",
+    "cell_border = pd.DataFrame(\n",
+    "    [[\"border-green \", \" \", \" \", \"border-red \"], [\" \", \" \", \" \", \" \"]],\n",
+    "    index=df.index,\n",
+    "    columns=df.columns[:4],\n",
+    ")\n",
+    "s.set_td_classes(cell_color + cell_border)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hidden cell to avoid CSS clashes and latter code upcoding previous formatting\n",
+    "s.set_uuid(\"after_borders\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Finer Control with Slicing\n",
+    "\n",
+    "The examples we have shown so far for the `Styler.apply` and `Styler.map` functions have not demonstrated the use of the ``subset`` argument. This is a useful argument which permits a lot of flexibility: it allows you to apply styles to specific rows or columns, without having to code that logic into your `style` function.\n",
+    "\n",
+    "The value passed to `subset` behaves similar to slicing a DataFrame;\n",
+    "\n",
+    "- A scalar is treated as a column label\n",
+    "- A list (or Series or NumPy array) is treated as multiple column labels\n",
+    "- A tuple is treated as `(row_indexer, column_indexer)`\n",
+    "\n",
+    "Consider using `pd.IndexSlice` to construct the tuple for the last one. We will create a MultiIndexed DataFrame to demonstrate the functionality."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df3 = pd.DataFrame(\n",
+    "    np.random.default_rng().standard_normal((4, 4)),\n",
+    "    pd.MultiIndex.from_product([[\"A\", \"B\"], [\"r1\", \"r2\"]]),\n",
+    "    columns=[\"c1\", \"c2\", \"c3\", \"c4\"],\n",
+    ")\n",
+    "df3"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We will use subset to highlight the maximum in the third and fourth columns with red text. We will highlight the subset sliced region in yellow."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "slice_ = [\"c3\", \"c4\"]\n",
+    "df3.style.apply(\n",
+    "    highlight_max, props=\"color:red;\", axis=0, subset=slice_\n",
+    ").set_properties(**{\"background-color\": \"#ffffb3\"}, subset=slice_)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "If combined with the ``IndexSlice`` as suggested then it can index across both dimensions with greater flexibility."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "idx = pd.IndexSlice\n",
+    "slice_ = idx[idx[:, \"r1\"], idx[\"c2\":\"c4\"]]\n",
+    "df3.style.apply(\n",
+    "    highlight_max, props=\"color:red;\", axis=0, subset=slice_\n",
+    ").set_properties(**{\"background-color\": \"#ffffb3\"}, subset=slice_)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This also provides the flexibility to sub select rows when used with the `axis=1`."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "slice_ = idx[idx[:, \"r2\"], :]\n",
+    "df3.style.apply(\n",
+    "    highlight_max, props=\"color:red;\", axis=1, subset=slice_\n",
+    ").set_properties(**{\"background-color\": \"#ffffb3\"}, subset=slice_)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "There is also scope to provide **conditional filtering**. \n",
+    "\n",
+    "Suppose we want to highlight the maximum across columns 2 and 4 only in the case that the sum of columns 1 and 3 is less than -2.0 *(essentially excluding rows* `(:,'r2')`*)*."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "slice_ = idx[idx[(df3[\"c1\"] + df3[\"c3\"]) < -2.0], [\"c2\", \"c4\"]]\n",
+    "df3.style.apply(\n",
+    "    highlight_max, props=\"color:red;\", axis=1, subset=slice_\n",
+    ").set_properties(**{\"background-color\": \"#ffffb3\"}, subset=slice_)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Only label-based slicing is supported right now, not positional, and not callables.\n",
+    "\n",
+    "If your style function uses a `subset` or `axis` keyword argument, consider wrapping your function in a `functools.partial`, partialing out that keyword.\n",
+    "\n",
+    "```python\n",
+    "my_func2 = functools.partial(my_func, subset=42)\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Optimization\n",
+    "\n",
+    "Generally, for smaller tables and most cases, the rendered HTML does not need to be optimized, and we don't really recommend it. There are two cases where it is worth considering:\n",
+    "\n",
+    " - If you are rendering and styling a very large HTML table, certain browsers have performance issues.\n",
+    " - If you are using ``Styler`` to dynamically create part of online user interfaces and want to improve network performance.\n",
+    " \n",
+    "Here we recommend the following steps to implement:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 1. Remove UUID and cell_ids\n",
+    "\n",
+    "Ignore the `uuid` and set `cell_ids` to `False`. This will prevent unnecessary HTML."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-warning\">\n",
+    "\n",
+    "<font color=red>This is sub-optimal:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4 = pd.DataFrame([[1, 2], [3, 4]])\n",
+    "s4 = df4.style"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-info\">\n",
+    "\n",
+    "<font color=green>This is better:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pandas.io.formats.style import Styler\n",
+    "\n",
+    "s4 = Styler(df4, uuid_len=0, cell_ids=False)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. Use table styles\n",
+    "\n",
+    "Use table styles where possible (e.g. for all cells or rows or columns at a time) since the CSS is nearly always more efficient than other formats."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-warning\">\n",
+    "\n",
+    "<font color=red>This is sub-optimal:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "props = 'font-family: \"Times New Roman\", Times, serif; color: #e83e8c; font-size:1.3em;'\n",
+    "df4.style.map(lambda x: props, subset=[1])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-info\">\n",
+    "\n",
+    "<font color=green>This is better:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.set_table_styles([{\"selector\": \"td.col1\", \"props\": props}])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 3. Set classes instead of using Styler functions\n",
+    "\n",
+    "For large DataFrames where the same style is applied to many cells it can be more efficient to declare the styles as classes and then apply those classes to data cells, rather than directly applying styles to cells. It is, however, probably still easier to use the Styler function api when you are not concerned about optimization."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-warning\">\n",
+    "\n",
+    "<font color=red>This is sub-optimal:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.style.apply(\n",
+    "    highlight_max, props=\"color:white;background-color:darkblue;\", axis=0\n",
+    ").apply(highlight_max, props=\"color:white;background-color:pink;\", axis=1).apply(\n",
+    "    highlight_max, props=\"color:white;background-color:purple\", axis=None\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div class=\"alert alert-info\">\n",
+    "\n",
+    "<font color=green>This is better:</font>\n",
+    "\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "build = lambda x: pd.DataFrame(x, index=df2.index, columns=df2.columns)\n",
+    "cls1 = build(df2.apply(highlight_max, props=\"cls-1 \", axis=0))\n",
+    "cls2 = build(\n",
+    "    df2.apply(highlight_max, props=\"cls-2 \", axis=1, result_type=\"expand\").values\n",
+    ")\n",
+    "cls3 = build(highlight_max(df2, props=\"cls-3 \"))\n",
+    "df2.style.set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \".cls-1\", \"props\": \"color:white;background-color:darkblue;\"},\n",
+    "        {\"selector\": \".cls-2\", \"props\": \"color:white;background-color:pink;\"},\n",
+    "        {\"selector\": \".cls-3\", \"props\": \"color:white;background-color:purple;\"},\n",
+    "    ]\n",
+    ").set_td_classes(cls1 + cls2 + cls3)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 4. Don't use tooltips\n",
+    "\n",
+    "Tooltips require `cell_ids` to work and they generate extra HTML elements for *every* data cell."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 5. If every byte counts use string replacement\n",
+    "\n",
+    "You can remove unnecessary HTML, or shorten the default class names by replacing the default css dict. You can read a little more about CSS [below](#More-About-CSS-and-HTML)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "my_css = {\n",
+    "    \"row_heading\": \"\",\n",
+    "    \"col_heading\": \"\",\n",
+    "    \"index_name\": \"\",\n",
+    "    \"col\": \"c\",\n",
+    "    \"row\": \"r\",\n",
+    "    \"col_trim\": \"\",\n",
+    "    \"row_trim\": \"\",\n",
+    "    \"level\": \"l\",\n",
+    "    \"data\": \"\",\n",
+    "    \"blank\": \"\",\n",
+    "}\n",
+    "html = Styler(df4, uuid_len=0, cell_ids=False)\n",
+    "html.set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \"td\", \"props\": props},\n",
+    "        {\"selector\": \".c1\", \"props\": \"color:green;\"},\n",
+    "        {\"selector\": \".l0\", \"props\": \"color:blue;\"},\n",
+    "    ],\n",
+    "    css_class_names=my_css,\n",
+    ")\n",
+    "print(html.to_html())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "html"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Builtin Styles"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Some styling functions are common enough that we've \"built them in\" to the `Styler`, so you don't have to write them and apply them yourself. The current list of such functions is:\n",
+    "\n",
+    " - [.highlight_null][nullfunc]: for use with identifying missing data. \n",
+    " - [.highlight_min][minfunc] and [.highlight_max][maxfunc]: for use with identifying extremities in data.\n",
+    " - [.highlight_between][betweenfunc] and [.highlight_quantile][quantilefunc]: for use with identifying classes within data.\n",
+    " - [.background_gradient][bgfunc]: a flexible method for highlighting cells based on their, or other, values on a numeric scale.\n",
+    " - [.text_gradient][textfunc]: similar method for highlighting text based on their, or other, values on a numeric scale.\n",
+    " - [.bar][barfunc]: to display mini-charts within cell backgrounds.\n",
+    " \n",
+    "The individual documentation on each function often gives more examples of their arguments.\n",
+    "\n",
+    "[nullfunc]: ../reference/api/pandas.io.formats.style.Styler.highlight_null.rst\n",
+    "[minfunc]: ../reference/api/pandas.io.formats.style.Styler.highlight_min.rst\n",
+    "[maxfunc]: ../reference/api/pandas.io.formats.style.Styler.highlight_max.rst\n",
+    "[betweenfunc]: ../reference/api/pandas.io.formats.style.Styler.highlight_between.rst\n",
+    "[quantilefunc]: ../reference/api/pandas.io.formats.style.Styler.highlight_quantile.rst\n",
+    "[bgfunc]: ../reference/api/pandas.io.formats.style.Styler.background_gradient.rst\n",
+    "[textfunc]: ../reference/api/pandas.io.formats.style.Styler.text_gradient.rst\n",
+    "[barfunc]: ../reference/api/pandas.io.formats.style.Styler.bar.rst"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Highlight Null"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.iloc[0, 2] = np.nan\n",
+    "df2.iloc[4, 3] = np.nan\n",
+    "df2.loc[:4].style.highlight_null(color=\"yellow\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Highlight Min or Max"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.loc[:4].style.highlight_max(\n",
+    "    axis=1, props=(\"color:white; font-weight:bold; background-color:darkblue;\")\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Highlight Between\n",
+    "This method accepts ranges as float, or NumPy arrays or Series provided the indexes match."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "left = pd.Series([1.0, 0.0, 1.0], index=[\"A\", \"B\", \"D\"])\n",
+    "df2.loc[:4].style.highlight_between(\n",
+    "    left=left, right=1.5, axis=1, props=\"color:white; background-color:purple;\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Highlight Quantile\n",
+    "Useful for detecting the highest or lowest percentile values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.loc[:4].style.highlight_quantile(q_left=0.85, axis=None, color=\"yellow\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Background Gradient and Text Gradient"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "You can create \"heatmaps\" with the `background_gradient` and `text_gradient` methods. These require matplotlib, and we'll use [Seaborn](https://seaborn.pydata.org/) to get a nice colormap."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import seaborn as sns\n",
+    "\n",
+    "cm = sns.light_palette(\"green\", as_cmap=True)\n",
+    "\n",
+    "df2.style.background_gradient(cmap=cm)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.style.text_gradient(cmap=cm)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "[.background_gradient][bgfunc] and [.text_gradient][textfunc] have a number of keyword arguments to customise the gradients and colors. See the documentation.\n",
+    "\n",
+    "[bgfunc]: ../reference/api/pandas.io.formats.style.Styler.background_gradient.rst\n",
+    "[textfunc]: ../reference/api/pandas.io.formats.style.Styler.text_gradient.rst"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Set properties\n",
+    "\n",
+    "Use `Styler.set_properties` when the style doesn't actually depend on the values. This is just a simple wrapper for `.map` where the function returns the same properties for all cells."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.loc[:4].style.set_properties(\n",
+    "    **{\"background-color\": \"black\", \"color\": \"lawngreen\", \"border-color\": \"white\"}\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Bar charts"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "You can include \"bar charts\" in your DataFrame."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.style.bar(subset=[\"A\", \"B\"], color=\"#d65f5f\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Additional keyword arguments give more control on centering and positioning, and you can pass a list of `[color_negative, color_positive]` to highlight lower and higher values or a matplotlib colormap.\n",
+    "\n",
+    "To showcase an example here's how you can change the above with the new `align` option, combined with setting `vmin` and `vmax` limits, the `width` of the figure, and underlying css `props` of cells, leaving space to display the text and the bars. We also use `text_gradient` to color the text the same as the bars using a matplotlib colormap (although in this case the visualization is probably better without this additional effect)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.style.format(\"{:.3f}\", na_rep=\"\").bar(\n",
+    "    align=0,\n",
+    "    vmin=-2.5,\n",
+    "    vmax=2.5,\n",
+    "    cmap=\"bwr\",\n",
+    "    height=50,\n",
+    "    width=60,\n",
+    "    props=\"width: 120px; border-right: 1px solid black;\",\n",
+    ").text_gradient(cmap=\"bwr\", vmin=-2.5, vmax=2.5)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The following example aims to give a highlight of the behavior of the new align options:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# Hide the construction of the display chart from the user\n",
+    "import pandas as pd\n",
+    "from IPython.display import HTML\n",
+    "\n",
+    "# Test series\n",
+    "test1 = pd.Series([-100, -60, -30, -20], name=\"All Negative\")\n",
+    "test2 = pd.Series([-10, -5, 0, 90], name=\"Both Pos and Neg\")\n",
+    "test3 = pd.Series([10, 20, 50, 100], name=\"All Positive\")\n",
+    "test4 = pd.Series([100, 103, 101, 102], name=\"Large Positive\")\n",
+    "\n",
+    "\n",
+    "head = \"\"\"\n",
+    "<table>\n",
+    "    <thead>\n",
+    "        <th>Align</th>\n",
+    "        <th>All Negative</th>\n",
+    "        <th>Both Neg and Pos</th>\n",
+    "        <th>All Positive</th>\n",
+    "        <th>Large Positive</th>\n",
+    "    </thead>\n",
+    "    </tbody>\n",
+    "\n",
+    "\"\"\"\n",
+    "\n",
+    "aligns = [\"left\", \"right\", \"zero\", \"mid\", \"mean\", 99]\n",
+    "for align in aligns:\n",
+    "    row = \"<tr><th>{}</th>\".format(align)\n",
+    "    for series in [test1, test2, test3, test4]:\n",
+    "        s = series.copy()\n",
+    "        s.name = \"\"\n",
+    "        row += \"<td>{}</td>\".format(\n",
+    "            s.to_frame()\n",
+    "            .style.hide(axis=\"index\")\n",
+    "            .bar(align=align, color=[\"#d65f5f\", \"#5fba7d\"], width=100)\n",
+    "            .to_html()\n",
+    "        )  # testn['width']\n",
+    "    row += \"</tr>\"\n",
+    "    head += row\n",
+    "\n",
+    "head += \"\"\"\n",
+    "</tbody>\n",
+    "</table>\"\"\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "HTML(head)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Sharing styles"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Say you have a lovely style built up for a DataFrame, and now you want to apply the same style to a second DataFrame. Export the style with `df1.style.export`, and import it on the second DataFrame with `df1.style.set`"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "style1 = (\n",
+    "    df2.style.map(style_negative, props=\"color:red;\")\n",
+    "    .map(lambda v: \"opacity: 20%;\" if (v < 0.3) and (v > -0.3) else None)\n",
+    "    .set_table_styles([{\"selector\": \"th\", \"props\": \"color: blue;\"}])\n",
+    "    .hide(axis=\"index\")\n",
+    ")\n",
+    "style1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "style2 = df3.style\n",
+    "style2.use(style1.export())\n",
+    "style2"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Notice that you're able to share the styles even though they're data aware. The styles are re-evaluated on the new DataFrame they've been `use`d upon."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Limitations\n",
+    "\n",
+    "- DataFrame only (use `Series.to_frame().style`)\n",
+    "- The index and columns do not need to be unique, but certain styling functions can only work with unique indexes.\n",
+    "- No large repr, and construction performance isn't great; although we have some [HTML optimizations](#Optimization)\n",
+    "- You can only apply styles, you can't insert new HTML entities, except via subclassing."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Other Fun and Useful Stuff\n",
+    "\n",
+    "Here are a few interesting examples."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Widgets\n",
+    "\n",
+    "`Styler` interacts pretty well with widgets. If you're viewing this online instead of running the notebook yourself, you're missing out on interactively adjusting the color palette."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from ipywidgets import widgets\n",
+    "\n",
+    "\n",
+    "@widgets.interact\n",
+    "def f(h_neg=(0, 359, 1), h_pos=(0, 359), s=(0.0, 99.9), l_post=(0.0, 99.9)):\n",
+    "    return df2.style.background_gradient(\n",
+    "        cmap=sns.palettes.diverging_palette(\n",
+    "            h_neg=h_neg, h_pos=h_pos, s=s, l=l_post, as_cmap=True\n",
+    "        )\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Magnify"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def magnify():\n",
+    "    return [\n",
+    "        {\"selector\": \"th\", \"props\": [(\"font-size\", \"4pt\")]},\n",
+    "        {\"selector\": \"td\", \"props\": [(\"padding\", \"0em 0em\")]},\n",
+    "        {\"selector\": \"th:hover\", \"props\": [(\"font-size\", \"12pt\")]},\n",
+    "        {\n",
+    "            \"selector\": \"tr:hover td:hover\",\n",
+    "            \"props\": [(\"max-width\", \"200px\"), (\"font-size\", \"12pt\")],\n",
+    "        },\n",
+    "    ]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cmap = sns.diverging_palette(5, 250, as_cmap=True)\n",
+    "bigdf = pd.DataFrame(np.random.default_rng(25).standard_normal((20, 25))).cumsum()\n",
+    "\n",
+    "bigdf.style.background_gradient(cmap, axis=1).set_properties(\n",
+    "    **{\"max-width\": \"80px\", \"font-size\": \"1pt\"}\n",
+    ").set_caption(\"Hover to magnify\").format(precision=2).set_table_styles(magnify())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Sticky Headers\n",
+    "\n",
+    "If you display a large matrix or DataFrame in a notebook, but you want to always see the column and row headers you can use the [.set_sticky][sticky] method which manipulates the table styles CSS.\n",
+    "\n",
+    "[sticky]: ../reference/api/pandas.io.formats.style.Styler.set_sticky.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "bigdf = pd.DataFrame(np.random.default_rng().standard_normal((16, 100)))\n",
+    "bigdf.style.set_sticky(axis=\"index\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "It is also possible to stick MultiIndexes and even only specific levels."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "bigdf.index = pd.MultiIndex.from_product([[\"A\", \"B\"], [0, 1], [0, 1, 2, 3]])\n",
+    "bigdf.style.set_sticky(axis=\"index\", pixel_size=18, levels=[1, 2])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### HTML Escaping\n",
+    "\n",
+    "Suppose you have to display HTML within HTML, that can be a bit of pain when the renderer can't distinguish. You can use the `escape` formatting option to handle this, and even use it within a formatter that contains HTML itself.\n",
+    "\n",
+    "Note that if you're using `Styler` on untrusted, user-provided input to serve HTML then you should escape the input to prevent security vulnerabilities. See the Jinja2 documentation for more."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4 = pd.DataFrame([[\"<div></div>\", '\"&other\"', \"<span></span>\"]])\n",
+    "df4.style"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.format(escape=\"html\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.format(\n",
+    "    '<a href=\"https://pandas.pydata.org\" target=\"_blank\">{}</a>', escape=\"html\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Export to Excel\n",
+    "\n",
+    "Some support (*since version 0.20.0*) is available for exporting styled `DataFrames` to Excel worksheets using the `OpenPyXL` or `XlsxWriter` engines. CSS2.2 properties handled include:\n",
+    "\n",
+    "- `background-color`\n",
+    "- `border-style` properties\n",
+    "- `border-width` properties\n",
+    "- `border-color` properties\n",
+    "- `color`\n",
+    "- `font-family`\n",
+    "- `font-style`\n",
+    "- `font-weight`\n",
+    "- `text-align`\n",
+    "- `text-decoration`\n",
+    "- `vertical-align`\n",
+    "- `white-space: nowrap`\n",
+    "\n",
+    "\n",
+    "- Shorthand and side-specific border properties are supported (e.g. `border-style` and `border-left-style`) as well as the `border` shorthands for all sides (`border: 1px solid green`) or specified sides (`border-left: 1px solid green`). Using a `border` shorthand will override any border properties set before it (See [CSS Working Group](https://drafts.csswg.org/css-backgrounds/#border-shorthands) for more details)\n",
+    "\n",
+    "\n",
+    "- Only CSS2 named colors and hex colors of the form `#rgb` or `#rrggbb` are currently supported.\n",
+    "- The following pseudo CSS properties are also available to set Excel specific style properties:\n",
+    "    - `number-format`\n",
+    "    - `border-style` (for Excel-specific styles: \"hair\", \"mediumDashDot\", \"dashDotDot\", \"mediumDashDotDot\", \"dashDot\", \"slantDashDot\", or \"mediumDashed\")\n",
+    "\n",
+    "Table level styles, and data cell CSS-classes are not included in the export to Excel: individual cells must have their properties mapped by the `Styler.apply` and/or `Styler.map` methods."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df2.style.map(style_negative, props=\"color:red;\").highlight_max(axis=0).to_excel(\n",
+    "    \"styled.xlsx\", engine=\"openpyxl\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "A screenshot of the output:\n",
+    "\n",
+    "![Excel spreadsheet with styled DataFrame](../_static/style-excel.png)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Export to LaTeX\n",
+    "\n",
+    "There is support (*since version 1.3.0*) to export `Styler` to LaTeX. The documentation for the [.to_latex][latex] method gives further detail and numerous examples.\n",
+    "\n",
+    "[latex]: ../reference/api/pandas.io.formats.style.Styler.to_latex.rst"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## More About CSS and HTML\n",
+    "\n",
+    "Cascading Style Sheet (CSS) language, which is designed to influence how a browser renders HTML elements, has its own peculiarities. It never reports errors: it just silently ignores them and doesn't render your objects how you intend so can sometimes be frustrating. Here is a very brief primer on how ``Styler`` creates HTML and interacts with CSS, with advice on common pitfalls to avoid."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### CSS Classes and Ids\n",
+    "\n",
+    "The precise structure of the CSS `class` attached to each cell is as follows.\n",
+    "\n",
+    "- Cells with Index and Column names include `index_name` and `level<k>` where `k` is its level in a MultiIndex\n",
+    "- Index label cells include\n",
+    "  + `row_heading`\n",
+    "  + `level<k>` where `k` is the level in a MultiIndex\n",
+    "  + `row<m>` where `m` is the numeric position of the row\n",
+    "- Column label cells include\n",
+    "  + `col_heading`\n",
+    "  + `level<k>` where `k` is the level in a MultiIndex\n",
+    "  + `col<n>` where `n` is the numeric position of the column\n",
+    "- Data cells include\n",
+    "  + `data`\n",
+    "  + `row<m>`, where `m` is the numeric position of the cell.\n",
+    "  + `col<n>`, where `n` is the numeric position of the cell.\n",
+    "- Blank cells include `blank`\n",
+    "- Trimmed cells include `col_trim` or `row_trim`\n",
+    "\n",
+    "The structure of the `id` is `T_uuid_level<k>_row<m>_col<n>` where `level<k>` is used only on headings, and headings will only have either `row<m>` or `col<n>` whichever is needed. By default we've also prepended each row/column identifier with a UUID unique to each DataFrame so that the style from one doesn't collide with the styling from another within the same notebook or page. You can read more about the use of UUIDs in [Optimization](#Optimization).\n",
+    "\n",
+    "We can see example of the HTML by calling the [.to_html()][tohtml] method.\n",
+    "\n",
+    "[tohtml]: ../reference/api/pandas.io.formats.style.Styler.to_html.rst"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(\n",
+    "    pd.DataFrame(\n",
+    "        [[1, 2], [3, 4]], index=[\"i1\", \"i2\"], columns=[\"c1\", \"c2\"]\n",
+    "    ).style.to_html()\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### CSS Hierarchies\n",
+    "\n",
+    "The examples have shown that when CSS styles overlap, the one that comes last in the HTML render, takes precedence. So the following yield different results:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4 = pd.DataFrame([[\"text\"]])\n",
+    "df4.style.map(lambda x: \"color:green;\").map(lambda x: \"color:red;\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.map(lambda x: \"color:red;\").map(lambda x: \"color:green;\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This is only true for CSS rules that are equivalent in hierarchy, or importance. You can read more about [CSS specificity here](https://www.w3schools.com/css/css_specificity.asp) but for our purposes it suffices to summarize the key points:\n",
+    "\n",
+    "A CSS importance score for each HTML element is derived by starting at zero and adding:\n",
+    "\n",
+    " - 1000 for an inline style attribute\n",
+    " - 100 for each ID\n",
+    " - 10 for each attribute, class or pseudo-class\n",
+    " - 1 for each element name or pseudo-element\n",
+    " \n",
+    "Let's use this to describe the action of the following configurations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.set_uuid(\"a_\").set_table_styles(\n",
+    "    [{\"selector\": \"td\", \"props\": \"color:red;\"}]\n",
+    ").map(lambda x: \"color:green;\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This text is red because the generated selector `#T_a_ td` is worth 101 (ID plus element), whereas `#T_a_row0_col0` is only worth 100 (ID), so is considered inferior even though in the HTML it comes after the previous."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.set_uuid(\"b_\").set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \"td\", \"props\": \"color:red;\"},\n",
+    "        {\"selector\": \".cls-1\", \"props\": \"color:blue;\"},\n",
+    "    ]\n",
+    ").map(lambda x: \"color:green;\").set_td_classes(pd.DataFrame([[\"cls-1\"]]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In the above case the text is blue because the selector `#T_b_ .cls-1` is worth 110 (ID plus class), which takes precedence."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.set_uuid(\"c_\").set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \"td\", \"props\": \"color:red;\"},\n",
+    "        {\"selector\": \".cls-1\", \"props\": \"color:blue;\"},\n",
+    "        {\"selector\": \"td.data\", \"props\": \"color:yellow;\"},\n",
+    "    ]\n",
+    ").map(lambda x: \"color:green;\").set_td_classes(pd.DataFrame([[\"cls-1\"]]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we have created another table style this time the selector `T_c_ td.data` (ID plus element plus class) gets bumped up to 111. \n",
+    "\n",
+    "If your style fails to be applied, and its really frustrating, try the `!important` trump card."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df4.style.set_uuid(\"d_\").set_table_styles(\n",
+    "    [\n",
+    "        {\"selector\": \"td\", \"props\": \"color:red;\"},\n",
+    "        {\"selector\": \".cls-1\", \"props\": \"color:blue;\"},\n",
+    "        {\"selector\": \"td.data\", \"props\": \"color:yellow;\"},\n",
+    "    ]\n",
+    ").map(lambda x: \"color:green !important;\").set_td_classes(pd.DataFrame([[\"cls-1\"]]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally got that green text after all!"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Extensibility\n",
+    "\n",
+    "The core of pandas is, and will remain, its \"high-performance, easy-to-use data structures\".\n",
+    "With that in mind, we hope that `DataFrame.style` accomplishes two goals\n",
+    "\n",
+    "- Provide an API that is pleasing to use interactively and is \"good enough\" for many tasks\n",
+    "- Provide the foundations for dedicated libraries to build on\n",
+    "\n",
+    "If you build a great library on top of this, let us know and we'll [link](https://pandas.pydata.org/community/ecosystem.html) to it.\n",
+    "\n",
+    "### Subclassing\n",
+    "\n",
+    "If the default template doesn't quite suit your needs, you can subclass Styler and extend or override the template.\n",
+    "We'll show an example of extending the default template to insert a custom header before each table."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from jinja2 import Environment, ChoiceLoader, FileSystemLoader\n",
+    "from IPython.display import HTML\n",
+    "from pandas.io.formats.style import Styler"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We'll use the following template:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "with open(\"templates/myhtml.tpl\") as f_html:\n",
+    "    print(f_html.read())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now that we've created a template, we need to set up a subclass of ``Styler`` that\n",
+    "knows about it."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class MyStyler(Styler):\n",
+    "    env = Environment(\n",
+    "        loader=ChoiceLoader(\n",
+    "            [\n",
+    "                FileSystemLoader(\"templates\"),  # contains ours\n",
+    "                Styler.loader,  # the default\n",
+    "            ]\n",
+    "        )\n",
+    "    )\n",
+    "    template_html_table = env.get_template(\"myhtml.tpl\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Notice that we include the original loader in our environment's loader.\n",
+    "That's because we extend the original template, so the Jinja environment needs\n",
+    "to be able to find it.\n",
+    "\n",
+    "Now we can use that custom styler. It's `__init__` takes a DataFrame."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "MyStyler(df3)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Our custom template accepts a `table_title` keyword. We can provide the value in the `.to_html` method."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "HTML(MyStyler(df3).to_html(table_title=\"Extending Example\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "For convenience, we provide the `Styler.from_custom_template` method that does the same as the custom subclass."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "EasyStyler = Styler.from_custom_template(\"templates\", \"myhtml.tpl\")\n",
+    "HTML(EasyStyler(df3).to_html(table_title=\"Another Title\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Template Structure\n",
+    "\n",
+    "Here's the template structure for the both the style generation template and the table generation template:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Style template:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "with open(\"templates/html_style_structure.html\") as f_sty:\n",
+    "    style_structure = f_sty.read()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "HTML(style_structure)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Table template:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "with open(\"templates/html_table_structure.html\") as f_table_struct:\n",
+    "    table_structure = f_table_struct.read()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "HTML(table_structure)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "See the template in the [GitHub repo](https://github.com/pandas-dev/pandas) for more details."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "nbsphinx": "hidden"
+   },
+   "outputs": [],
+   "source": [
+    "# # Hack to get the same style in the notebook as the\n",
+    "# # main site. This is hidden in the docs.\n",
+    "# from IPython.display import HTML\n",
+    "# with open(\"themes/nature_with_gtoc/static/nature.css_t\") as f:\n",
+    "#     css = f.read()\n",
+    "\n",
+    "# HTML('<style>{}</style>'.format(css))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/pandas/doc/source/user_guide/templates/html_style_structure.html b/pandas/doc/source/user_guide/templates/html_style_structure.html
new file mode 100644
index 0000000000000..dc0c03ac363a9
--- /dev/null
+++ b/pandas/doc/source/user_guide/templates/html_style_structure.html
@@ -0,0 +1,35 @@
+<!--
+     This is an HTML fragment that gets included into a notebook & rst document
+
+     Inspired by nbconvert
+
+     https://github.com/jupyter/nbconvert/blob/8ac591a0b8694147d0f34bf6392594c2811c1395/docs/source/template_structure.html
+
+
+   -->
+<style type="text/css">
+    /* Overrides of notebook CSS for static HTML export */
+    .template_block {
+        background-color: hsla(120, 60%, 70%, 0.2);
+        margin: 10px;
+        padding: 5px;
+        border: 1px solid hsla(120, 60%, 70%, 0.5);
+        border-left: 2px solid black;
+    }
+    .template_block pre {
+        background: transparent;
+        padding: 0;
+    }
+    .big_vertical_ellipsis {
+        font-size: 24pt;
+    }
+</style>
+
+<div class="template_block">before_style</div>
+<div class="template_block">style
+    <pre>&lt;style type=&quot;text/css&quot;&gt;</pre>
+    <div class="template_block">table_styles</div>
+    <div class="template_block">before_cellstyle</div>
+    <div class="template_block">cellstyle</div>
+    <pre>&lt;/style&gt;</pre>
+</div><!-- /style -->
diff --git a/pandas/doc/source/user_guide/templates/html_table_structure.html b/pandas/doc/source/user_guide/templates/html_table_structure.html
new file mode 100644
index 0000000000000..e03f9591d2a35
--- /dev/null
+++ b/pandas/doc/source/user_guide/templates/html_table_structure.html
@@ -0,0 +1,48 @@
+<!--
+     This is an HTML fragment that gets included into a notebook & rst document
+
+     Inspired by nbconvert
+
+     https://github.com/jupyter/nbconvert/blob/8ac591a0b8694147d0f34bf6392594c2811c1395/docs/source/template_structure.html
+
+
+   -->
+<style type="text/css">
+    /* Overrides of notebook CSS for static HTML export */
+    .template_block {
+        background-color: hsla(120, 60%, 70%, 0.2);
+        margin: 10px;
+        padding: 5px;
+        border: 1px solid hsla(120, 60%, 70%, 0.5);
+        border-left: 2px solid black;
+    }
+    .template_block pre {
+        background: transparent;
+        padding: 0;
+    }
+    .big_vertical_ellipsis {
+        font-size: 24pt;
+    }
+</style>
+
+<div class="template_block" >before_table</div>
+
+<div class="template_block" >table
+    <pre>&lt;table ...&gt;</pre>
+    <div class="template_block">caption</div>
+
+    <div class="template_block" >thead
+        <div class="template_block" >before_head_rows</div>
+        <div class="template_block">head_tr (loop over headers)</div>
+        <div class="template_block" >after_head_rows</div>
+    </div>
+
+    <div class="template_block" >tbody
+        <div class="template_block" >before_rows</div>
+        <div class="template_block">tr (loop over data rows)</div>
+        <div class="template_block" >after_rows</div>
+    </div>
+    <pre>&lt;/table&gt;</pre>
+</div><!-- /table -->
+
+<div class="template_block" >after_table</div>
diff --git a/pandas/doc/source/user_guide/templates/myhtml.tpl b/pandas/doc/source/user_guide/templates/myhtml.tpl
new file mode 100644
index 0000000000000..1e204d0bd4568
--- /dev/null
+++ b/pandas/doc/source/user_guide/templates/myhtml.tpl
@@ -0,0 +1,5 @@
+{% extends "html_table.tpl" %}
+{% block table %}
+<h1>{{ table_title|default("My Table") }}</h1>
+{{ super() }}
+{% endblock table %}
diff --git a/pandas/doc/source/user_guide/text.rst b/pandas/doc/source/user_guide/text.rst
new file mode 100644
index 0000000000000..e96faecd9a266
--- /dev/null
+++ b/pandas/doc/source/user_guide/text.rst
@@ -0,0 +1,781 @@
+.. _text:
+
+{{ header }}
+
+======================
+Working with text data
+======================
+
+.. _text.types:
+
+Text data types
+---------------
+
+There are two ways to store text data in pandas:
+
+1. ``object`` dtype NumPy array.
+2. :class:`StringDtype` extension type.
+
+We recommend using :class:`StringDtype` to store text data.
+
+Prior to pandas 1.0, ``object`` dtype was the only option. This was unfortunate
+for many reasons:
+
+1. You can accidentally store a *mixture* of strings and non-strings in an
+   ``object`` dtype array. It's better to have a dedicated dtype.
+2. ``object`` dtype breaks dtype-specific operations like :meth:`DataFrame.select_dtypes`.
+   There isn't a clear way to select *just* text while excluding non-text
+   but still object-dtype columns.
+3. When reading code, the contents of an ``object`` dtype array is less clear
+   than ``'string'``.
+
+Currently, the performance of ``object`` dtype arrays of strings and
+:class:`arrays.StringArray` are about the same. We expect future enhancements
+to significantly increase the performance and lower the memory overhead of
+:class:`~arrays.StringArray`.
+
+.. warning::
+
+   ``StringArray`` is currently considered experimental. The implementation
+   and parts of the API may change without warning.
+
+For backwards-compatibility, ``object`` dtype remains the default type we
+infer a list of strings to:
+
+.. ipython:: python
+
+   pd.Series(["a", "b", "c"])
+
+To explicitly request ``string`` dtype, specify the ``dtype``:
+
+.. ipython:: python
+
+   pd.Series(["a", "b", "c"], dtype="string")
+   pd.Series(["a", "b", "c"], dtype=pd.StringDtype())
+
+Or ``astype`` after the ``Series`` or ``DataFrame`` is created:
+
+.. ipython:: python
+
+   s = pd.Series(["a", "b", "c"])
+   s
+   s.astype("string")
+
+
+You can also use :class:`StringDtype`/``"string"`` as the dtype on non-string data and
+it will be converted to ``string`` dtype:
+
+.. ipython:: python
+
+   s = pd.Series(["a", 2, np.nan], dtype="string")
+   s
+   type(s[1])
+
+or convert from existing pandas data:
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, np.nan], dtype="Int64")
+   s1
+   s2 = s1.astype("string")
+   s2
+   type(s2[0])
+
+
+.. _text.differences:
+
+Behavior differences
+^^^^^^^^^^^^^^^^^^^^
+
+These are places where the behavior of ``StringDtype`` objects differ from
+``object`` dtype:
+
+l. For ``StringDtype``, :ref:`string accessor methods<api.series.str>`
+   that return **numeric** output will always return a nullable integer dtype,
+   rather than either int or float dtype, depending on the presence of NA values.
+   Methods returning **boolean** output will return a nullable boolean dtype.
+
+   .. ipython:: python
+
+      s = pd.Series(["a", None, "b"], dtype="string")
+      s
+      s.str.count("a")
+      s.dropna().str.count("a")
+
+   Both outputs are ``Int64`` dtype. Compare that with object-dtype:
+
+   .. ipython:: python
+
+      s2 = pd.Series(["a", None, "b"], dtype="object")
+      s2.str.count("a")
+      s2.dropna().str.count("a")
+
+   When NA values are present, the output dtype is float64. Similarly for
+   methods returning boolean values.
+
+   .. ipython:: python
+
+      s.str.isdigit()
+      s.str.match("a")
+
+2. Some string methods, like :meth:`Series.str.decode` are not available
+   on ``StringArray`` because ``StringArray`` only holds strings, not
+   bytes.
+3. In comparison operations, :class:`arrays.StringArray` and ``Series`` backed
+   by a ``StringArray`` will return an object with :class:`BooleanDtype`,
+   rather than a ``bool`` dtype object. Missing values in a ``StringArray``
+   will propagate in comparison operations, rather than always comparing
+   unequal like :attr:`numpy.nan`.
+
+Everything else that follows in the rest of this document applies equally to
+``string`` and ``object`` dtype.
+
+.. _text.string_methods:
+
+String methods
+--------------
+
+Series and Index are equipped with a set of string processing methods
+that make it easy to operate on each element of the array. Perhaps most
+importantly, these methods exclude missing/NA values automatically. These are
+accessed via the ``str`` attribute and generally have names matching
+the equivalent (scalar) built-in string methods:
+
+.. ipython:: python
+
+   s = pd.Series(
+       ["A", "B", "C", "Aaba", "Baca", np.nan, "CABA", "dog", "cat"], dtype="string"
+   )
+   s.str.lower()
+   s.str.upper()
+   s.str.len()
+
+.. ipython:: python
+
+   idx = pd.Index([" jack", "jill ", " jesse ", "frank"])
+   idx.str.strip()
+   idx.str.lstrip()
+   idx.str.rstrip()
+
+The string methods on Index are especially useful for cleaning up or
+transforming DataFrame columns. For instance, you may have columns with
+leading or trailing whitespace:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.random.randn(3, 2), columns=[" Column A ", " Column B "], index=range(3)
+   )
+   df
+
+Since ``df.columns`` is an Index object, we can use the ``.str`` accessor
+
+.. ipython:: python
+
+   df.columns.str.strip()
+   df.columns.str.lower()
+
+These string methods can then be used to clean up the columns as needed.
+Here we are removing leading and trailing whitespaces, lower casing all names,
+and replacing any remaining whitespaces with underscores:
+
+.. ipython:: python
+
+   df.columns = df.columns.str.strip().str.lower().str.replace(" ", "_")
+   df
+
+.. note::
+
+    If you have a ``Series`` where lots of elements are repeated
+    (i.e. the number of unique elements in the ``Series`` is a lot smaller than the length of the
+    ``Series``), it can be faster to convert the original ``Series`` to one of type
+    ``category`` and then use ``.str.<method>`` or ``.dt.<property>`` on that.
+    The performance difference comes from the fact that, for ``Series`` of type ``category``, the
+    string operations are done on the ``.categories`` and not on each element of the
+    ``Series``.
+
+    Please note that a ``Series`` of type ``category`` with string ``.categories`` has
+    some limitations in comparison to ``Series`` of type string (e.g. you can't add strings to
+    each other: ``s + " " + s`` won't work if ``s`` is a ``Series`` of type ``category``). Also,
+    ``.str`` methods which operate on elements of type ``list`` are not available on such a
+    ``Series``.
+
+.. _text.warn_types:
+
+.. warning::
+
+    The type of the Series is inferred and is one among the allowed types (i.e. strings).
+
+    Generally speaking, the ``.str`` accessor is intended to work only on strings. With very few
+    exceptions, other uses are not supported, and may be disabled at a later point.
+
+.. _text.split:
+
+Splitting and replacing strings
+-------------------------------
+
+Methods like ``split`` return a Series of lists:
+
+.. ipython:: python
+
+   s2 = pd.Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype="string")
+   s2.str.split("_")
+
+Elements in the split lists can be accessed using ``get`` or ``[]`` notation:
+
+.. ipython:: python
+
+   s2.str.split("_").str.get(1)
+   s2.str.split("_").str[1]
+
+It is easy to expand this to return a DataFrame using ``expand``.
+
+.. ipython:: python
+
+   s2.str.split("_", expand=True)
+
+When original ``Series`` has :class:`StringDtype`, the output columns will all
+be :class:`StringDtype` as well.
+
+It is also possible to limit the number of splits:
+
+.. ipython:: python
+
+   s2.str.split("_", expand=True, n=1)
+
+``rsplit`` is similar to ``split`` except it works in the reverse direction,
+i.e., from the end of the string to the beginning of the string:
+
+.. ipython:: python
+
+   s2.str.rsplit("_", expand=True, n=1)
+
+``replace`` optionally uses `regular expressions
+<https://docs.python.org/3/library/re.html>`__:
+
+.. ipython:: python
+
+   s3 = pd.Series(
+       ["A", "B", "C", "Aaba", "Baca", "", np.nan, "CABA", "dog", "cat"],
+       dtype="string",
+   )
+   s3
+   s3.str.replace("^.a|dog", "XX-XX ", case=False, regex=True)
+
+
+.. versionchanged:: 2.0
+
+Single character pattern with ``regex=True`` will also be treated as regular expressions:
+
+.. ipython:: python
+
+   s4 = pd.Series(["a.b", ".", "b", np.nan, ""], dtype="string")
+   s4
+   s4.str.replace(".", "a", regex=True)
+
+If you want literal replacement of a string (equivalent to :meth:`str.replace`), you
+can set the optional ``regex`` parameter to ``False``, rather than escaping each
+character. In this case both ``pat`` and ``repl`` must be strings:
+
+.. ipython:: python
+
+    dollars = pd.Series(["12", "-$10", "$10,000"], dtype="string")
+
+    # These lines are equivalent
+    dollars.str.replace(r"-\$", "-", regex=True)
+    dollars.str.replace("-$", "-", regex=False)
+
+The ``replace`` method can also take a callable as replacement. It is called
+on every ``pat`` using :func:`re.sub`. The callable should expect one
+positional argument (a regex object) and return a string.
+
+.. ipython:: python
+
+   # Reverse every lowercase alphabetic word
+   pat = r"[a-z]+"
+
+   def repl(m):
+       return m.group(0)[::-1]
+
+   pd.Series(["foo 123", "bar baz", np.nan], dtype="string").str.replace(
+       pat, repl, regex=True
+   )
+
+   # Using regex groups
+   pat = r"(?P<one>\w+) (?P<two>\w+) (?P<three>\w+)"
+
+   def repl(m):
+       return m.group("two").swapcase()
+
+   pd.Series(["Foo Bar Baz", np.nan], dtype="string").str.replace(
+       pat, repl, regex=True
+   )
+
+The ``replace`` method also accepts a compiled regular expression object
+from :func:`re.compile` as a pattern. All flags should be included in the
+compiled regular expression object.
+
+.. ipython:: python
+
+   import re
+
+   regex_pat = re.compile(r"^.a|dog", flags=re.IGNORECASE)
+   s3.str.replace(regex_pat, "XX-XX ", regex=True)
+
+Including a ``flags`` argument when calling ``replace`` with a compiled
+regular expression object will raise a ``ValueError``.
+
+.. ipython::
+
+    @verbatim
+    In [1]: s3.str.replace(regex_pat, 'XX-XX ', flags=re.IGNORECASE)
+    ---------------------------------------------------------------------------
+    ValueError: case and flags cannot be set when pat is a compiled regex
+
+``removeprefix`` and ``removesuffix`` have the same effect as ``str.removeprefix`` and ``str.removesuffix`` added in Python 3.9
+<https://docs.python.org/3/library/stdtypes.html#str.removeprefix>`__:
+
+.. versionadded:: 1.4.0
+
+.. ipython:: python
+
+   s = pd.Series(["str_foo", "str_bar", "no_prefix"])
+   s.str.removeprefix("str_")
+
+   s = pd.Series(["foo_str", "bar_str", "no_suffix"])
+   s.str.removesuffix("_str")
+
+.. _text.concatenate:
+
+Concatenation
+-------------
+
+There are several ways to concatenate a ``Series`` or ``Index``, either with itself or others, all based on :meth:`~Series.str.cat`,
+resp. ``Index.str.cat``.
+
+Concatenating a single Series into a string
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The content of a ``Series`` (or ``Index``) can be concatenated:
+
+.. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "d"], dtype="string")
+    s.str.cat(sep=",")
+
+If not specified, the keyword ``sep`` for the separator defaults to the empty string, ``sep=''``:
+
+.. ipython:: python
+
+    s.str.cat()
+
+By default, missing values are ignored. Using ``na_rep``, they can be given a representation:
+
+.. ipython:: python
+
+    t = pd.Series(["a", "b", np.nan, "d"], dtype="string")
+    t.str.cat(sep=",")
+    t.str.cat(sep=",", na_rep="-")
+
+Concatenating a Series and something list-like into a Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The first argument to :meth:`~Series.str.cat` can be a list-like object, provided that it matches the length of the calling ``Series`` (or ``Index``).
+
+.. ipython:: python
+
+    s.str.cat(["A", "B", "C", "D"])
+
+Missing values on either side will result in missing values in the result as well, *unless* ``na_rep`` is specified:
+
+.. ipython:: python
+
+    s.str.cat(t)
+    s.str.cat(t, na_rep="-")
+
+Concatenating a Series and something array-like into a Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The parameter ``others`` can also be two-dimensional. In this case, the number or rows must match the lengths of the calling ``Series`` (or ``Index``).
+
+.. ipython:: python
+
+    d = pd.concat([t, s], axis=1)
+    s
+    d
+    s.str.cat(d, na_rep="-")
+
+Concatenating a Series and an indexed object into a Series, with alignment
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+For concatenation with a ``Series`` or ``DataFrame``, it is possible to align the indexes before concatenation by setting
+the ``join``-keyword.
+
+.. ipython:: python
+   :okwarning:
+
+   u = pd.Series(["b", "d", "a", "c"], index=[1, 3, 0, 2], dtype="string")
+   s
+   u
+   s.str.cat(u)
+   s.str.cat(u, join="left")
+
+The usual options are available for ``join`` (one of ``'left', 'outer', 'inner', 'right'``).
+In particular, alignment also means that the different lengths do not need to coincide anymore.
+
+.. ipython:: python
+
+    v = pd.Series(["z", "a", "b", "d", "e"], index=[-1, 0, 1, 3, 4], dtype="string")
+    s
+    v
+    s.str.cat(v, join="left", na_rep="-")
+    s.str.cat(v, join="outer", na_rep="-")
+
+The same alignment can be used when ``others`` is a ``DataFrame``:
+
+.. ipython:: python
+
+    f = d.loc[[3, 2, 1, 0], :]
+    s
+    f
+    s.str.cat(f, join="left", na_rep="-")
+
+Concatenating a Series and many objects into a Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Several array-like items (specifically: ``Series``, ``Index``, and 1-dimensional variants of ``np.ndarray``)
+can be combined in a list-like container (including iterators, ``dict``-views, etc.).
+
+.. ipython:: python
+
+    s
+    u
+    s.str.cat([u, u.to_numpy()], join="left")
+
+All elements without an index (e.g. ``np.ndarray``) within the passed list-like must match in length to the calling ``Series`` (or ``Index``),
+but ``Series`` and ``Index`` may have arbitrary length (as long as alignment is not disabled with ``join=None``):
+
+.. ipython:: python
+
+    v
+    s.str.cat([v, u, u.to_numpy()], join="outer", na_rep="-")
+
+If using ``join='right'`` on a list-like of ``others`` that contains different indexes,
+the union of these indexes will be used as the basis for the final concatenation:
+
+.. ipython:: python
+
+    u.loc[[3]]
+    v.loc[[-1, 0]]
+    s.str.cat([u.loc[[3]], v.loc[[-1, 0]]], join="right", na_rep="-")
+
+Indexing with ``.str``
+----------------------
+
+.. _text.indexing:
+
+You can use ``[]`` notation to directly index by position locations. If you index past the end
+of the string, the result will be a ``NaN``.
+
+
+.. ipython:: python
+
+   s = pd.Series(
+       ["A", "B", "C", "Aaba", "Baca", np.nan, "CABA", "dog", "cat"], dtype="string"
+   )
+
+   s.str[0]
+   s.str[1]
+
+Extracting substrings
+---------------------
+
+.. _text.extract:
+
+Extract first match in each subject (extract)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``extract`` method accepts a `regular expression
+<https://docs.python.org/3/library/re.html>`__ with at least one
+capture group.
+
+Extracting a regular expression with more than one group returns a
+DataFrame with one column per group.
+
+.. ipython:: python
+
+   pd.Series(
+       ["a1", "b2", "c3"],
+       dtype="string",
+   ).str.extract(r"([ab])(\d)", expand=False)
+
+Elements that do not match return a row filled with ``NaN``. Thus, a
+Series of messy strings can be "converted" into a like-indexed Series
+or DataFrame of cleaned-up or more useful strings, without
+necessitating ``get()`` to access tuples or ``re.match`` objects. The
+dtype of the result is always object, even if no match is found and
+the result only contains ``NaN``.
+
+Named groups like
+
+.. ipython:: python
+
+   pd.Series(["a1", "b2", "c3"], dtype="string").str.extract(
+       r"(?P<letter>[ab])(?P<digit>\d)", expand=False
+   )
+
+and optional groups like
+
+.. ipython:: python
+
+   pd.Series(
+       ["a1", "b2", "3"],
+       dtype="string",
+   ).str.extract(r"([ab])?(\d)", expand=False)
+
+can also be used. Note that any capture group names in the regular
+expression will be used for column names; otherwise capture group
+numbers will be used.
+
+Extracting a regular expression with one group returns a ``DataFrame``
+with one column if ``expand=True``.
+
+.. ipython:: python
+
+   pd.Series(["a1", "b2", "c3"], dtype="string").str.extract(r"[ab](\d)", expand=True)
+
+It returns a Series if ``expand=False``.
+
+.. ipython:: python
+
+   pd.Series(["a1", "b2", "c3"], dtype="string").str.extract(r"[ab](\d)", expand=False)
+
+Calling on an ``Index`` with a regex with exactly one capture group
+returns a ``DataFrame`` with one column if ``expand=True``.
+
+.. ipython:: python
+
+   s = pd.Series(["a1", "b2", "c3"], ["A11", "B22", "C33"], dtype="string")
+   s
+   s.index.str.extract("(?P<letter>[a-zA-Z])", expand=True)
+
+It returns an ``Index`` if ``expand=False``.
+
+.. ipython:: python
+
+   s.index.str.extract("(?P<letter>[a-zA-Z])", expand=False)
+
+Calling on an ``Index`` with a regex with more than one capture group
+returns a ``DataFrame`` if ``expand=True``.
+
+.. ipython:: python
+
+   s.index.str.extract("(?P<letter>[a-zA-Z])([0-9]+)", expand=True)
+
+It raises ``ValueError`` if ``expand=False``.
+
+.. ipython:: python
+   :okexcept:
+
+    s.index.str.extract("(?P<letter>[a-zA-Z])([0-9]+)", expand=False)
+
+The table below summarizes the behavior of ``extract(expand=False)``
+(input subject in first column, number of groups in regex in
+first row)
+
++--------+---------+------------+
+|        | 1 group | >1 group   |
++--------+---------+------------+
+| Index  | Index   | ValueError |
++--------+---------+------------+
+| Series | Series  | DataFrame  |
++--------+---------+------------+
+
+Extract all matches in each subject (extractall)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. _text.extractall:
+
+Unlike ``extract`` (which returns only the first match),
+
+.. ipython:: python
+
+   s = pd.Series(["a1a2", "b1", "c1"], index=["A", "B", "C"], dtype="string")
+   s
+   two_groups = "(?P<letter>[a-z])(?P<digit>[0-9])"
+   s.str.extract(two_groups, expand=True)
+
+the ``extractall`` method returns every match. The result of
+``extractall`` is always a ``DataFrame`` with a ``MultiIndex`` on its
+rows. The last level of the ``MultiIndex`` is named ``match`` and
+indicates the order in the subject.
+
+.. ipython:: python
+
+   s.str.extractall(two_groups)
+
+When each subject string in the Series has exactly one match,
+
+.. ipython:: python
+
+   s = pd.Series(["a3", "b3", "c2"], dtype="string")
+   s
+
+then ``extractall(pat).xs(0, level='match')`` gives the same result as
+``extract(pat)``.
+
+.. ipython:: python
+
+   extract_result = s.str.extract(two_groups, expand=True)
+   extract_result
+   extractall_result = s.str.extractall(two_groups)
+   extractall_result
+   extractall_result.xs(0, level="match")
+
+``Index`` also supports ``.str.extractall``. It returns a ``DataFrame`` which has the
+same result as a ``Series.str.extractall`` with a default index (starts from 0).
+
+.. ipython:: python
+
+   pd.Index(["a1a2", "b1", "c1"]).str.extractall(two_groups)
+
+   pd.Series(["a1a2", "b1", "c1"], dtype="string").str.extractall(two_groups)
+
+
+Testing for strings that match or contain a pattern
+---------------------------------------------------
+
+You can check whether elements contain a pattern:
+
+.. ipython:: python
+
+   pattern = r"[0-9][a-z]"
+   pd.Series(
+       ["1", "2", "3a", "3b", "03c", "4dx"],
+       dtype="string",
+   ).str.contains(pattern)
+
+Or whether elements match a pattern:
+
+.. ipython:: python
+
+   pd.Series(
+       ["1", "2", "3a", "3b", "03c", "4dx"],
+       dtype="string",
+   ).str.match(pattern)
+
+.. ipython:: python
+
+   pd.Series(
+       ["1", "2", "3a", "3b", "03c", "4dx"],
+       dtype="string",
+   ).str.fullmatch(pattern)
+
+.. note::
+
+    The distinction between ``match``, ``fullmatch``, and ``contains`` is strictness:
+    ``fullmatch`` tests whether the entire string matches the regular expression;
+    ``match`` tests whether there is a match of the regular expression that begins
+    at the first character of the string; and ``contains`` tests whether there is
+    a match of the regular expression at any position within the string.
+
+    The corresponding functions in the ``re`` package for these three match modes are
+    `re.fullmatch <https://docs.python.org/3/library/re.html#re.fullmatch>`_,
+    `re.match <https://docs.python.org/3/library/re.html#re.match>`_, and
+    `re.search <https://docs.python.org/3/library/re.html#re.search>`_,
+    respectively.
+
+Methods like ``match``, ``fullmatch``, ``contains``, ``startswith``, and
+``endswith`` take an extra ``na`` argument so missing values can be considered
+True or False:
+
+.. ipython:: python
+
+   s4 = pd.Series(
+       ["A", "B", "C", "Aaba", "Baca", np.nan, "CABA", "dog", "cat"], dtype="string"
+   )
+   s4.str.contains("A", na=False)
+
+.. _text.indicator:
+
+Creating indicator variables
+----------------------------
+
+You can extract dummy variables from string columns.
+For example if they are separated by a ``'|'``:
+
+.. ipython:: python
+
+    s = pd.Series(["a", "a|b", np.nan, "a|c"], dtype="string")
+    s.str.get_dummies(sep="|")
+
+String ``Index`` also supports ``get_dummies`` which returns a ``MultiIndex``.
+
+.. ipython:: python
+
+    idx = pd.Index(["a", "a|b", np.nan, "a|c"])
+    idx.str.get_dummies(sep="|")
+
+See also :func:`~pandas.get_dummies`.
+
+Method summary
+--------------
+
+.. _text.summary:
+
+.. csv-table::
+    :header: "Method", "Description"
+    :widths: 20, 80
+
+    :meth:`~Series.str.cat`,Concatenate strings
+    :meth:`~Series.str.split`,Split strings on delimiter
+    :meth:`~Series.str.rsplit`,Split strings on delimiter working from the end of the string
+    :meth:`~Series.str.get`,Index into each element (retrieve i-th element)
+    :meth:`~Series.str.join`,Join strings in each element of the Series with passed separator
+    :meth:`~Series.str.get_dummies`,Split strings on the delimiter returning DataFrame of dummy variables
+    :meth:`~Series.str.contains`,Return boolean array if each string contains pattern/regex
+    :meth:`~Series.str.replace`,Replace occurrences of pattern/regex/string with some other string or the return value of a callable given the occurrence
+    :meth:`~Series.str.removeprefix`,Remove prefix from string i.e. only remove if string starts with prefix.
+    :meth:`~Series.str.removesuffix`,Remove suffix from string i.e. only remove if string ends with suffix.
+    :meth:`~Series.str.repeat`,Duplicate values (``s.str.repeat(3)`` equivalent to ``x * 3``)
+    :meth:`~Series.str.pad`,Add whitespace to the sides of strings
+    :meth:`~Series.str.center`,Equivalent to ``str.center``
+    :meth:`~Series.str.ljust`,Equivalent to ``str.ljust``
+    :meth:`~Series.str.rjust`,Equivalent to ``str.rjust``
+    :meth:`~Series.str.zfill`,Equivalent to ``str.zfill``
+    :meth:`~Series.str.wrap`,Split long strings into lines with length less than a given width
+    :meth:`~Series.str.slice`,Slice each string in the Series
+    :meth:`~Series.str.slice_replace`,Replace slice in each string with passed value
+    :meth:`~Series.str.count`,Count occurrences of pattern
+    :meth:`~Series.str.startswith`,Equivalent to ``str.startswith(pat)`` for each element
+    :meth:`~Series.str.endswith`,Equivalent to ``str.endswith(pat)`` for each element
+    :meth:`~Series.str.findall`,Compute list of all occurrences of pattern/regex for each string
+    :meth:`~Series.str.match`,Call ``re.match`` on each element returning matched groups as list
+    :meth:`~Series.str.extract`,Call ``re.search`` on each element returning DataFrame with one row for each element and one column for each regex capture group
+    :meth:`~Series.str.extractall`,Call ``re.findall`` on each element returning DataFrame with one row for each match and one column for each regex capture group
+    :meth:`~Series.str.len`,Compute string lengths
+    :meth:`~Series.str.strip`,Equivalent to ``str.strip``
+    :meth:`~Series.str.rstrip`,Equivalent to ``str.rstrip``
+    :meth:`~Series.str.lstrip`,Equivalent to ``str.lstrip``
+    :meth:`~Series.str.partition`,Equivalent to ``str.partition``
+    :meth:`~Series.str.rpartition`,Equivalent to ``str.rpartition``
+    :meth:`~Series.str.lower`,Equivalent to ``str.lower``
+    :meth:`~Series.str.casefold`,Equivalent to ``str.casefold``
+    :meth:`~Series.str.upper`,Equivalent to ``str.upper``
+    :meth:`~Series.str.find`,Equivalent to ``str.find``
+    :meth:`~Series.str.rfind`,Equivalent to ``str.rfind``
+    :meth:`~Series.str.index`,Equivalent to ``str.index``
+    :meth:`~Series.str.rindex`,Equivalent to ``str.rindex``
+    :meth:`~Series.str.capitalize`,Equivalent to ``str.capitalize``
+    :meth:`~Series.str.swapcase`,Equivalent to ``str.swapcase``
+    :meth:`~Series.str.normalize`,Return Unicode normal form. Equivalent to ``unicodedata.normalize``
+    :meth:`~Series.str.translate`,Equivalent to ``str.translate``
+    :meth:`~Series.str.isalnum`,Equivalent to ``str.isalnum``
+    :meth:`~Series.str.isalpha`,Equivalent to ``str.isalpha``
+    :meth:`~Series.str.isdigit`,Equivalent to ``str.isdigit``
+    :meth:`~Series.str.isspace`,Equivalent to ``str.isspace``
+    :meth:`~Series.str.islower`,Equivalent to ``str.islower``
+    :meth:`~Series.str.isupper`,Equivalent to ``str.isupper``
+    :meth:`~Series.str.istitle`,Equivalent to ``str.istitle``
+    :meth:`~Series.str.isnumeric`,Equivalent to ``str.isnumeric``
+    :meth:`~Series.str.isdecimal`,Equivalent to ``str.isdecimal``
diff --git a/pandas/doc/source/user_guide/timedeltas.rst b/pandas/doc/source/user_guide/timedeltas.rst
new file mode 100644
index 0000000000000..01df17bac5fd7
--- /dev/null
+++ b/pandas/doc/source/user_guide/timedeltas.rst
@@ -0,0 +1,489 @@
+.. _timedeltas:
+
+{{ header }}
+
+.. _timedeltas.timedeltas:
+
+***********
+Time deltas
+***********
+
+Timedeltas are differences in times, expressed in difference units, e.g. days, hours, minutes,
+seconds. They can be both positive and negative.
+
+``Timedelta`` is a subclass of ``datetime.timedelta``, and behaves in a similar manner,
+but allows compatibility with ``np.timedelta64`` types as well as a host of custom representation,
+parsing, and attributes.
+
+Parsing
+-------
+
+You can construct a ``Timedelta`` scalar through various arguments, including `ISO 8601 Duration`_ strings.
+
+.. ipython:: python
+
+   import datetime
+
+   # strings
+   pd.Timedelta("1 days")
+   pd.Timedelta("1 days 00:00:00")
+   pd.Timedelta("1 days 2 hours")
+   pd.Timedelta("-1 days 2 min 3us")
+
+   # like datetime.timedelta
+   # note: these MUST be specified as keyword arguments
+   pd.Timedelta(days=1, seconds=1)
+
+   # integers with a unit
+   pd.Timedelta(1, unit="D")
+
+   # from a datetime.timedelta/np.timedelta64
+   pd.Timedelta(datetime.timedelta(days=1, seconds=1))
+   pd.Timedelta(np.timedelta64(1, "ms"))
+
+   # negative Timedeltas have this string repr
+   # to be more consistent with datetime.timedelta conventions
+   pd.Timedelta("-1us")
+
+   # a NaT
+   pd.Timedelta("nan")
+   pd.Timedelta("nat")
+
+   # ISO 8601 Duration strings
+   pd.Timedelta("P0DT0H1M0S")
+   pd.Timedelta("P0DT0H0M0.000000123S")
+
+:ref:`DateOffsets<timeseries.offsets>` (``Day, Hour, Minute, Second, Milli, Micro, Nano``) can also be used in construction.
+
+.. ipython:: python
+
+   pd.Timedelta(pd.offsets.Second(2))
+
+Further, operations among the scalars yield another scalar ``Timedelta``.
+
+.. ipython:: python
+
+   pd.Timedelta(pd.offsets.Day(2)) + pd.Timedelta(pd.offsets.Second(2)) + pd.Timedelta(
+       "00:00:00.000123"
+   )
+
+to_timedelta
+~~~~~~~~~~~~
+
+Using the top-level ``pd.to_timedelta``, you can convert a scalar, array, list,
+or Series from a recognized timedelta format / value into a ``Timedelta`` type.
+It will construct Series if the input is a Series, a scalar if the input is
+scalar-like, otherwise it will output a ``TimedeltaIndex``.
+
+You can parse a single string to a Timedelta:
+
+.. ipython:: python
+
+   pd.to_timedelta("1 days 06:05:01.00003")
+   pd.to_timedelta("15.5us")
+
+or a list/array of strings:
+
+.. ipython:: python
+
+   pd.to_timedelta(["1 days 06:05:01.00003", "15.5us", "nan"])
+
+The ``unit`` keyword argument specifies the unit of the Timedelta if the input
+is numeric:
+
+.. ipython:: python
+
+   pd.to_timedelta(np.arange(5), unit="s")
+   pd.to_timedelta(np.arange(5), unit="D")
+
+.. warning::
+    If a string or array of strings is passed as an input then the ``unit`` keyword
+    argument will be ignored. If a string without units is passed then the default
+    unit of nanoseconds is assumed.
+
+.. _timedeltas.limitations:
+
+Timedelta limitations
+~~~~~~~~~~~~~~~~~~~~~
+
+pandas represents ``Timedeltas`` in nanosecond resolution using
+64 bit integers. As such, the 64 bit integer limits determine
+the ``Timedelta`` limits.
+
+.. ipython:: python
+
+   pd.Timedelta.min
+   pd.Timedelta.max
+
+.. _timedeltas.operations:
+
+Operations
+----------
+
+You can operate on Series/DataFrames and construct ``timedelta64[ns]`` Series through
+subtraction operations on ``datetime64[ns]`` Series, or ``Timestamps``.
+
+.. ipython:: python
+
+   s = pd.Series(pd.date_range("2012-1-1", periods=3, freq="D"))
+   td = pd.Series([pd.Timedelta(days=i) for i in range(3)])
+   df = pd.DataFrame({"A": s, "B": td})
+   df
+   df["C"] = df["A"] + df["B"]
+   df
+   df.dtypes
+
+   s - s.max()
+   s - datetime.datetime(2011, 1, 1, 3, 5)
+   s + datetime.timedelta(minutes=5)
+   s + pd.offsets.Minute(5)
+   s + pd.offsets.Minute(5) + pd.offsets.Milli(5)
+
+Operations with scalars from a ``timedelta64[ns]`` series:
+
+.. ipython:: python
+
+   y = s - s[0]
+   y
+
+Series of timedeltas with ``NaT`` values are supported:
+
+.. ipython:: python
+
+   y = s - s.shift()
+   y
+
+Elements can be set to ``NaT`` using ``np.nan`` analogously to datetimes:
+
+.. ipython:: python
+
+   y[1] = np.nan
+   y
+
+Operands can also appear in a reversed order (a singular object operated with a Series):
+
+.. ipython:: python
+
+   s.max() - s
+   datetime.datetime(2011, 1, 1, 3, 5) - s
+   datetime.timedelta(minutes=5) + s
+
+``min, max`` and the corresponding ``idxmin, idxmax`` operations are supported on frames:
+
+.. ipython:: python
+
+   A = s - pd.Timestamp("20120101") - pd.Timedelta("00:05:05")
+   B = s - pd.Series(pd.date_range("2012-1-2", periods=3, freq="D"))
+
+   df = pd.DataFrame({"A": A, "B": B})
+   df
+
+   df.min()
+   df.min(axis=1)
+
+   df.idxmin()
+   df.idxmax()
+
+``min, max, idxmin, idxmax`` operations are supported on Series as well. A scalar result will be a ``Timedelta``.
+
+.. ipython:: python
+
+   df.min().max()
+   df.min(axis=1).min()
+
+   df.min().idxmax()
+   df.min(axis=1).idxmin()
+
+You can fillna on timedeltas, passing a timedelta to get a particular value.
+
+.. ipython:: python
+
+   y.fillna(pd.Timedelta(0))
+   y.fillna(pd.Timedelta(10, unit="s"))
+   y.fillna(pd.Timedelta("-1 days, 00:00:05"))
+
+You can also negate, multiply and use ``abs`` on ``Timedeltas``:
+
+.. ipython:: python
+
+   td1 = pd.Timedelta("-1 days 2 hours 3 seconds")
+   td1
+   -1 * td1
+   -td1
+   abs(td1)
+
+.. _timedeltas.timedeltas_reductions:
+
+Reductions
+----------
+
+Numeric reduction operation for ``timedelta64[ns]`` will return ``Timedelta`` objects. As usual
+``NaT`` are skipped during evaluation.
+
+.. ipython:: python
+
+   y2 = pd.Series(
+       pd.to_timedelta(["-1 days +00:00:05", "nat", "-1 days +00:00:05", "1 days"])
+   )
+   y2
+   y2.mean()
+   y2.median()
+   y2.quantile(0.1)
+   y2.sum()
+
+.. _timedeltas.timedeltas_convert:
+
+Frequency conversion
+--------------------
+
+Timedelta Series and ``TimedeltaIndex``, and ``Timedelta`` can be converted to other frequencies by astyping to a specific timedelta dtype.
+
+.. ipython:: python
+
+   december = pd.Series(pd.date_range("20121201", periods=4))
+   january = pd.Series(pd.date_range("20130101", periods=4))
+   td = january - december
+
+   td[2] += datetime.timedelta(minutes=5, seconds=3)
+   td[3] = np.nan
+   td
+
+   # to seconds
+   td.astype("timedelta64[s]")
+
+For timedelta64 resolutions other than the supported "s", "ms", "us", "ns",
+an alternative is to divide by another timedelta object. Note that division by the NumPy scalar is true division, while astyping is equivalent of floor division.
+
+.. ipython:: python
+
+   # to days
+   td / np.timedelta64(1, "D")
+
+Dividing or multiplying a ``timedelta64[ns]`` Series by an integer or integer Series
+yields another ``timedelta64[ns]`` dtypes Series.
+
+.. ipython:: python
+
+   td * -1
+   td * pd.Series([1, 2, 3, 4])
+
+Rounded division (floor-division) of a ``timedelta64[ns]`` Series by a scalar
+``Timedelta`` gives a series of integers.
+
+.. ipython:: python
+
+   td // pd.Timedelta(days=3, hours=4)
+   pd.Timedelta(days=3, hours=4) // td
+
+.. _timedeltas.mod_divmod:
+
+The mod (%) and divmod operations are defined for ``Timedelta`` when operating with another timedelta-like or with a numeric argument.
+
+.. ipython:: python
+
+   pd.Timedelta(hours=37) % datetime.timedelta(hours=2)
+
+   # divmod against a timedelta-like returns a pair (int, Timedelta)
+   divmod(datetime.timedelta(hours=2), pd.Timedelta(minutes=11))
+
+   # divmod against a numeric returns a pair (Timedelta, Timedelta)
+   divmod(pd.Timedelta(hours=25), 86400000000000)
+
+Attributes
+----------
+
+You can access various components of the ``Timedelta`` or ``TimedeltaIndex`` directly using the attributes ``days,seconds,microseconds,nanoseconds``. These are identical to the values returned by ``datetime.timedelta``, in that, for example, the ``.seconds`` attribute represents the number of seconds >= 0 and < 1 day. These are signed according to whether the ``Timedelta`` is signed.
+
+These operations can also be directly accessed via the ``.dt`` property of the ``Series`` as well.
+
+.. note::
+
+   Note that the attributes are NOT the displayed values of the ``Timedelta``. Use ``.components`` to retrieve the displayed values.
+
+For a ``Series``:
+
+.. ipython:: python
+
+   td.dt.days
+   td.dt.seconds
+
+You can access the value of the fields for a scalar ``Timedelta`` directly.
+
+.. ipython:: python
+
+   tds = pd.Timedelta("31 days 5 min 3 sec")
+   tds.days
+   tds.seconds
+   (-tds).seconds
+
+You can use the ``.components`` property to access a reduced form of the timedelta. This returns a ``DataFrame`` indexed
+similarly to the ``Series``. These are the *displayed* values of the ``Timedelta``.
+
+.. ipython:: python
+
+   td.dt.components
+   td.dt.components.seconds
+
+.. _timedeltas.isoformat:
+
+You can convert a ``Timedelta`` to an `ISO 8601 Duration`_ string with the
+``.isoformat`` method
+
+.. ipython:: python
+
+    pd.Timedelta(
+        days=6, minutes=50, seconds=3, milliseconds=10, microseconds=10, nanoseconds=12
+    ).isoformat()
+
+.. _ISO 8601 Duration: https://en.wikipedia.org/wiki/ISO_8601#Durations
+
+.. _timedeltas.index:
+
+TimedeltaIndex
+--------------
+
+To generate an index with time delta, you can use either the :class:`TimedeltaIndex` or
+the :func:`timedelta_range` constructor.
+
+Using ``TimedeltaIndex`` you can pass string-like, ``Timedelta``, ``timedelta``,
+or ``np.timedelta64`` objects. Passing ``np.nan/pd.NaT/nat`` will represent missing values.
+
+.. ipython:: python
+
+   pd.TimedeltaIndex(
+       [
+           "1 days",
+           "1 days, 00:00:05",
+           np.timedelta64(2, "D"),
+           datetime.timedelta(days=2, seconds=2),
+       ]
+   )
+
+The string 'infer' can be passed in order to set the frequency of the index as the
+inferred frequency upon creation:
+
+.. ipython:: python
+
+   pd.TimedeltaIndex(["0 days", "10 days", "20 days"], freq="infer")
+
+Generating ranges of time deltas
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Similar to :func:`date_range`, you can construct regular ranges of a ``TimedeltaIndex``
+using :func:`timedelta_range`.  The default frequency for ``timedelta_range`` is
+calendar day:
+
+.. ipython:: python
+
+   pd.timedelta_range(start="1 days", periods=5)
+
+Various combinations of ``start``, ``end``, and ``periods`` can be used with
+``timedelta_range``:
+
+.. ipython:: python
+
+   pd.timedelta_range(start="1 days", end="5 days")
+
+   pd.timedelta_range(end="10 days", periods=4)
+
+The ``freq`` parameter can passed a variety of :ref:`frequency aliases <timeseries.offset_aliases>`:
+
+.. ipython:: python
+
+   pd.timedelta_range(start="1 days", end="2 days", freq="30min")
+
+   pd.timedelta_range(start="1 days", periods=5, freq="2D5h")
+
+
+Specifying ``start``, ``end``, and ``periods`` will generate a range of evenly spaced
+timedeltas from ``start`` to ``end`` inclusively, with ``periods`` number of elements
+in the resulting ``TimedeltaIndex``:
+
+.. ipython:: python
+
+   pd.timedelta_range("0 days", "4 days", periods=5)
+
+   pd.timedelta_range("0 days", "4 days", periods=10)
+
+Using the TimedeltaIndex
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Similarly to other of the datetime-like indices, ``DatetimeIndex`` and ``PeriodIndex``, you can use
+``TimedeltaIndex`` as the index of pandas objects.
+
+.. ipython:: python
+
+   s = pd.Series(
+       np.arange(100),
+       index=pd.timedelta_range("1 days", periods=100, freq="h"),
+   )
+   s
+
+Selections work similarly, with coercion on string-likes and slices:
+
+.. ipython:: python
+
+   s["1 day":"2 day"]
+   s["1 day 01:00:00"]
+   s[pd.Timedelta("1 day 1h")]
+
+Furthermore you can use partial string selection and the range will be inferred:
+
+.. ipython:: python
+
+   s["1 day":"1 day 5 hours"]
+
+Operations
+~~~~~~~~~~
+
+Finally, the combination of ``TimedeltaIndex`` with ``DatetimeIndex`` allow certain combination operations that are NaT preserving:
+
+.. ipython:: python
+
+   tdi = pd.TimedeltaIndex(["1 days", pd.NaT, "2 days"])
+   tdi.to_list()
+   dti = pd.date_range("20130101", periods=3)
+   dti.to_list()
+   (dti + tdi).to_list()
+   (dti - tdi).to_list()
+
+Conversions
+~~~~~~~~~~~
+
+Similarly to frequency conversion on a ``Series`` above, you can convert these indices to yield another Index.
+
+.. ipython:: python
+
+   tdi / np.timedelta64(1, "s")
+   tdi.astype("timedelta64[s]")
+
+Scalars type ops work as well. These can potentially return a *different* type of index.
+
+.. ipython:: python
+
+   # adding or timedelta and date -> datelike
+   tdi + pd.Timestamp("20130101")
+
+   # subtraction of a date and a timedelta -> datelike
+   # note that trying to subtract a date from a Timedelta will raise an exception
+   (pd.Timestamp("20130101") - tdi).to_list()
+
+   # timedelta + timedelta -> timedelta
+   tdi + pd.Timedelta("10 days")
+
+   # division can result in a Timedelta if the divisor is an integer
+   tdi / 2
+
+   # or a float64 Index if the divisor is a Timedelta
+   tdi / tdi[0]
+
+.. _timedeltas.resampling:
+
+Resampling
+----------
+
+Similar to :ref:`timeseries resampling <timeseries.resampling>`, we can resample with a ``TimedeltaIndex``.
+
+.. ipython:: python
+
+   s.resample("D").mean()
diff --git a/pandas/doc/source/user_guide/timeseries.rst b/pandas/doc/source/user_guide/timeseries.rst
new file mode 100644
index 0000000000000..d046d13f71daf
--- /dev/null
+++ b/pandas/doc/source/user_guide/timeseries.rst
@@ -0,0 +1,2698 @@
+.. _timeseries:
+
+{{ header }}
+
+********************************
+Time series / date functionality
+********************************
+
+pandas contains extensive capabilities and features for working with time series data for all domains.
+Using the NumPy ``datetime64`` and ``timedelta64`` dtypes, pandas has consolidated a large number of
+features from other Python libraries like ``scikits.timeseries`` as well as created
+a tremendous amount of new functionality for manipulating time series data.
+
+For example, pandas supports:
+
+Parsing time series information from various sources and formats
+
+.. ipython:: python
+
+   import datetime
+
+   dti = pd.to_datetime(
+       ["1/1/2018", np.datetime64("2018-01-01"), datetime.datetime(2018, 1, 1)]
+   )
+   dti
+
+Generate sequences of fixed-frequency dates and time spans
+
+.. ipython:: python
+
+   dti = pd.date_range("2018-01-01", periods=3, freq="h")
+   dti
+
+Manipulating and converting date times with timezone information
+
+.. ipython:: python
+
+   dti = dti.tz_localize("UTC")
+   dti
+   dti.tz_convert("US/Pacific")
+
+Resampling or converting a time series to a particular frequency
+
+.. ipython:: python
+
+   idx = pd.date_range("2018-01-01", periods=5, freq="h")
+   ts = pd.Series(range(len(idx)), index=idx)
+   ts
+   ts.resample("2h").mean()
+
+Performing date and time arithmetic with absolute or relative time increments
+
+.. ipython:: python
+
+    friday = pd.Timestamp("2018-01-05")
+    friday.day_name()
+    # Add 1 day
+    saturday = friday + pd.Timedelta("1 day")
+    saturday.day_name()
+    # Add 1 business day (Friday --> Monday)
+    monday = friday + pd.offsets.BDay()
+    monday.day_name()
+
+pandas provides a relatively compact and self-contained set of tools for
+performing the above tasks and more.
+
+
+.. _timeseries.overview:
+
+Overview
+--------
+
+pandas captures 4 general time related concepts:
+
+#. Date times: A specific date and time with timezone support. Similar to ``datetime.datetime`` from the standard library.
+#. Time deltas: An absolute time duration. Similar to ``datetime.timedelta`` from the standard library.
+#. Time spans: A span of time defined by a point in time and its associated frequency.
+#. Date offsets: A relative time duration that respects calendar arithmetic. Similar to ``dateutil.relativedelta.relativedelta`` from the ``dateutil`` package.
+
+=====================   =================  ===================   ============================================  ========================================
+Concept                 Scalar Class       Array Class           pandas Data Type                              Primary Creation Method
+=====================   =================  ===================   ============================================  ========================================
+Date times              ``Timestamp``      ``DatetimeIndex``     ``datetime64[ns]`` or ``datetime64[ns, tz]``  ``to_datetime`` or ``date_range``
+Time deltas             ``Timedelta``      ``TimedeltaIndex``    ``timedelta64[ns]``                           ``to_timedelta`` or ``timedelta_range``
+Time spans              ``Period``         ``PeriodIndex``       ``period[freq]``                              ``Period`` or ``period_range``
+Date offsets            ``DateOffset``     ``None``              ``None``                                      ``DateOffset``
+=====================   =================  ===================   ============================================  ========================================
+
+For time series data, it's conventional to represent the time component in the index of a :class:`Series` or :class:`DataFrame`
+so manipulations can be performed with respect to the time element.
+
+.. ipython:: python
+
+   pd.Series(range(3), index=pd.date_range("2000", freq="D", periods=3))
+
+However, :class:`Series` and :class:`DataFrame` can directly also support the time component as data itself.
+
+.. ipython:: python
+
+   pd.Series(pd.date_range("2000", freq="D", periods=3))
+
+:class:`Series` and :class:`DataFrame` have extended data type support and functionality for ``datetime``, ``timedelta``
+and ``Period`` data when passed into those constructors. ``DateOffset``
+data however will be stored as ``object`` data.
+
+.. ipython:: python
+
+   pd.Series(pd.period_range("1/1/2011", freq="M", periods=3))
+   pd.Series([pd.DateOffset(1), pd.DateOffset(2)])
+   pd.Series(pd.date_range("1/1/2011", freq="ME", periods=3))
+
+Lastly, pandas represents null date times, time deltas, and time spans as ``NaT`` which
+is useful for representing missing or null date like values and behaves similar
+as ``np.nan`` does for float data.
+
+.. ipython:: python
+
+   pd.Timestamp(pd.NaT)
+   pd.Timedelta(pd.NaT)
+   pd.Period(pd.NaT)
+   # Equality acts as np.nan would
+   pd.NaT == pd.NaT
+
+.. _timeseries.representation:
+
+Timestamps vs. time spans
+-------------------------
+
+Timestamped data is the most basic type of time series data that associates
+values with points in time. For pandas objects it means using the points in
+time.
+
+.. ipython:: python
+
+   import datetime
+
+   pd.Timestamp(datetime.datetime(2012, 5, 1))
+   pd.Timestamp("2012-05-01")
+   pd.Timestamp(2012, 5, 1)
+
+However, in many cases it is more natural to associate things like change
+variables with a time span instead. The span represented by ``Period`` can be
+specified explicitly, or inferred from datetime string format.
+
+For example:
+
+.. ipython:: python
+
+   pd.Period("2011-01")
+
+   pd.Period("2012-05", freq="D")
+
+:class:`Timestamp` and :class:`Period` can serve as an index. Lists of
+``Timestamp`` and ``Period`` are automatically coerced to :class:`DatetimeIndex`
+and :class:`PeriodIndex` respectively.
+
+.. ipython:: python
+
+   dates = [
+       pd.Timestamp("2012-05-01"),
+       pd.Timestamp("2012-05-02"),
+       pd.Timestamp("2012-05-03"),
+   ]
+   ts = pd.Series(np.random.randn(3), dates)
+
+   type(ts.index)
+   ts.index
+
+   ts
+
+   periods = [pd.Period("2012-01"), pd.Period("2012-02"), pd.Period("2012-03")]
+
+   ts = pd.Series(np.random.randn(3), periods)
+
+   type(ts.index)
+   ts.index
+
+   ts
+
+pandas allows you to capture both representations and
+convert between them. Under the hood, pandas represents timestamps using
+instances of ``Timestamp`` and sequences of timestamps using instances of
+``DatetimeIndex``. For regular time spans, pandas uses ``Period`` objects for
+scalar values and ``PeriodIndex`` for sequences of spans. Better support for
+irregular intervals with arbitrary start and end points are forth-coming in
+future releases.
+
+
+.. _timeseries.converting:
+
+Converting to timestamps
+------------------------
+
+To convert a :class:`Series` or list-like object of date-like objects e.g. strings,
+epochs, or a mixture, you can use the ``to_datetime`` function. When passed
+a ``Series``, this returns a ``Series`` (with the same index), while a list-like
+is converted to a ``DatetimeIndex``:
+
+.. ipython:: python
+
+    pd.to_datetime(pd.Series(["Jul 31, 2009", "Jan 10, 2010", None]))
+
+    pd.to_datetime(["2005/11/23", "2010/12/31"])
+
+If you use dates which start with the day first (i.e. European style),
+you can pass the ``dayfirst`` flag:
+
+.. ipython:: python
+    :okwarning:
+
+    pd.to_datetime(["04-01-2012 10:00"], dayfirst=True)
+
+    pd.to_datetime(["04-14-2012 10:00"], dayfirst=True)
+
+.. warning::
+
+   You see in the above example that ``dayfirst`` isn't strict. If a date
+   can't be parsed with the day being first it will be parsed as if
+   ``dayfirst`` were ``False`` and a warning will also be raised.
+
+If you pass a single string to ``to_datetime``, it returns a single ``Timestamp``.
+``Timestamp`` can also accept string input, but it doesn't accept string parsing
+options like ``dayfirst`` or ``format``, so use ``to_datetime`` if these are required.
+
+.. ipython:: python
+
+    pd.to_datetime("2010/11/12")
+
+    pd.Timestamp("2010/11/12")
+
+You can also use the ``DatetimeIndex`` constructor directly:
+
+.. ipython:: python
+
+    pd.DatetimeIndex(["2018-01-01", "2018-01-03", "2018-01-05"])
+
+The string 'infer' can be passed in order to set the frequency of the index as the
+inferred frequency upon creation:
+
+.. ipython:: python
+
+    pd.DatetimeIndex(["2018-01-01", "2018-01-03", "2018-01-05"], freq="infer")
+
+.. _timeseries.converting.format:
+
+Providing a format argument
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In addition to the required datetime string, a ``format`` argument can be passed to ensure specific parsing.
+This could also potentially speed up the conversion considerably.
+
+.. ipython:: python
+
+    pd.to_datetime("2010/11/12", format="%Y/%m/%d")
+
+    pd.to_datetime("12-11-2010 00:00", format="%d-%m-%Y %H:%M")
+
+For more information on the choices available when specifying the ``format``
+option, see the Python `datetime documentation`_.
+
+.. _datetime documentation: https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior
+
+Assembling datetime from multiple DataFrame columns
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can also pass a ``DataFrame`` of integer or string columns to assemble into a ``Series`` of ``Timestamps``.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"year": [2015, 2016], "month": [2, 3], "day": [4, 5], "hour": [2, 3]}
+   )
+   pd.to_datetime(df)
+
+
+You can pass only the columns that you need to assemble.
+
+.. ipython:: python
+
+   pd.to_datetime(df[["year", "month", "day"]])
+
+``pd.to_datetime`` looks for standard designations of the datetime component in the column names, including:
+
+* required: ``year``, ``month``, ``day``
+* optional: ``hour``, ``minute``, ``second``, ``millisecond``, ``microsecond``, ``nanosecond``
+
+Invalid data
+~~~~~~~~~~~~
+
+The default behavior, ``errors='raise'``, is to raise when unparsable:
+
+.. ipython:: python
+   :okexcept:
+
+   pd.to_datetime(['2009/07/31', 'asd'], errors='raise')
+
+Pass ``errors='coerce'`` to convert unparsable data to ``NaT`` (not a time):
+
+.. ipython:: python
+
+   pd.to_datetime(["2009/07/31", "asd"], errors="coerce")
+
+
+.. _timeseries.converting.epoch:
+
+Epoch timestamps
+~~~~~~~~~~~~~~~~
+
+pandas supports converting integer or float epoch times to ``Timestamp`` and
+``DatetimeIndex``. The default unit is nanoseconds, since that is how ``Timestamp``
+objects are stored internally. However, epochs are often stored in another ``unit``
+which can be specified. These are computed from the starting point specified by the
+``origin`` parameter.
+
+.. ipython:: python
+
+   pd.to_datetime(
+       [1349720105, 1349806505, 1349892905, 1349979305, 1350065705], unit="s"
+   )
+
+   pd.to_datetime(
+       [1349720105100, 1349720105200, 1349720105300, 1349720105400, 1349720105500],
+       unit="ms",
+   )
+
+.. note::
+
+   The ``unit`` parameter does not use the same strings as the ``format`` parameter
+   that was discussed :ref:`above<timeseries.converting.format>`. The
+   available units are listed on the documentation for :func:`pandas.to_datetime`.
+
+Constructing a :class:`Timestamp` or :class:`DatetimeIndex` with an epoch timestamp
+with the ``tz`` argument specified will raise a ValueError. If you have
+epochs in wall time in another timezone, you can read the epochs
+as timezone-naive timestamps and then localize to the appropriate timezone:
+
+.. ipython:: python
+
+   pd.Timestamp(1262347200000000000).tz_localize("US/Pacific")
+   pd.DatetimeIndex([1262347200000000000]).tz_localize("US/Pacific")
+
+.. note::
+
+   Epoch times will be rounded to the nearest nanosecond.
+
+.. warning::
+
+   Conversion of float epoch times can lead to inaccurate and unexpected results.
+   :ref:`Python floats <python:tut-fp-issues>` have about 15 digits precision in
+   decimal. Rounding during conversion from float to high precision ``Timestamp`` is
+   unavoidable. The only way to achieve exact precision is to use a fixed-width
+   types (e.g. an int64).
+
+   .. ipython:: python
+
+      pd.to_datetime([1490195805.433, 1490195805.433502912], unit="s")
+      pd.to_datetime(1490195805433502912, unit="ns")
+
+.. seealso::
+
+   :ref:`timeseries.origin`
+
+.. _timeseries.converting.epoch_inverse:
+
+From timestamps to epoch
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+To invert the operation from above, namely, to convert from a ``Timestamp`` to a 'unix' epoch:
+
+.. ipython:: python
+
+   stamps = pd.date_range("2012-10-08 18:15:05", periods=4, freq="D")
+   stamps
+
+We subtract the epoch (midnight at January 1, 1970 UTC) and then floor divide by the
+"unit" (1 second).
+
+.. ipython:: python
+
+   (stamps - pd.Timestamp("1970-01-01")) // pd.Timedelta("1s")
+
+.. _timeseries.origin:
+
+Using the ``origin`` parameter
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Using the ``origin`` parameter, one can specify an alternative starting point for creation
+of a ``DatetimeIndex``. For example, to use 1960-01-01 as the starting date:
+
+.. ipython:: python
+
+   pd.to_datetime([1, 2, 3], unit="D", origin=pd.Timestamp("1960-01-01"))
+
+The default is set at ``origin='unix'``, which defaults to ``1970-01-01 00:00:00``.
+Commonly called 'unix epoch' or POSIX time.
+
+.. ipython:: python
+
+   pd.to_datetime([1, 2, 3], unit="D")
+
+.. _timeseries.daterange:
+
+Generating ranges of timestamps
+-------------------------------
+
+To generate an index with timestamps, you can use either the ``DatetimeIndex`` or
+``Index`` constructor and pass in a list of datetime objects:
+
+.. ipython:: python
+
+   dates = [
+       datetime.datetime(2012, 5, 1),
+       datetime.datetime(2012, 5, 2),
+       datetime.datetime(2012, 5, 3),
+   ]
+
+   # Note the frequency information
+   index = pd.DatetimeIndex(dates)
+   index
+
+   # Automatically converted to DatetimeIndex
+   index = pd.Index(dates)
+   index
+
+In practice this becomes very cumbersome because we often need a very long
+index with a large number of timestamps. If we need timestamps on a regular
+frequency, we can use the :func:`date_range` and :func:`bdate_range` functions
+to create a ``DatetimeIndex``. The default frequency for ``date_range`` is a
+**calendar day** while the default for ``bdate_range`` is a **business day**:
+
+.. ipython:: python
+
+   start = datetime.datetime(2011, 1, 1)
+   end = datetime.datetime(2012, 1, 1)
+
+   index = pd.date_range(start, end)
+   index
+
+   index = pd.bdate_range(start, end)
+   index
+
+Convenience functions like ``date_range`` and ``bdate_range`` can utilize a
+variety of :ref:`frequency aliases <timeseries.offset_aliases>`:
+
+.. ipython:: python
+
+   pd.date_range(start, periods=1000, freq="ME")
+
+   pd.bdate_range(start, periods=250, freq="BQS")
+
+``date_range`` and ``bdate_range`` make it easy to generate a range of dates
+using various combinations of parameters like ``start``, ``end``, ``periods``,
+and ``freq``. The start and end dates are strictly inclusive, so dates outside
+of those specified will not be generated:
+
+.. ipython:: python
+
+   pd.date_range(start, end, freq="BME")
+
+   pd.date_range(start, end, freq="W")
+
+   pd.bdate_range(end=end, periods=20)
+
+   pd.bdate_range(start=start, periods=20)
+
+Specifying ``start``, ``end``, and ``periods`` will generate a range of evenly spaced
+dates from ``start`` to ``end`` inclusively, with ``periods`` number of elements in the
+resulting ``DatetimeIndex``:
+
+.. ipython:: python
+
+   pd.date_range("2018-01-01", "2018-01-05", periods=5)
+
+   pd.date_range("2018-01-01", "2018-01-05", periods=10)
+
+.. _timeseries.custom-freq-ranges:
+
+Custom frequency ranges
+~~~~~~~~~~~~~~~~~~~~~~~
+
+``bdate_range`` can also generate a range of custom frequency dates by using
+the ``weekmask`` and ``holidays`` parameters.  These parameters will only be
+used if a custom frequency string is passed.
+
+.. ipython:: python
+
+   weekmask = "Mon Wed Fri"
+
+   holidays = [datetime.datetime(2011, 1, 5), datetime.datetime(2011, 3, 14)]
+
+   pd.bdate_range(start, end, freq="C", weekmask=weekmask, holidays=holidays)
+
+   pd.bdate_range(start, end, freq="CBMS", weekmask=weekmask)
+
+.. seealso::
+
+   :ref:`timeseries.custombusinessdays`
+
+.. _timeseries.timestamp-limits:
+
+Timestamp limitations
+---------------------
+
+The limits of timestamp representation depend on the chosen resolution. For
+nanosecond resolution, the time span that
+can be represented using a 64-bit integer is limited to approximately 584 years:
+
+.. ipython:: python
+
+   pd.Timestamp.min
+   pd.Timestamp.max
+
+When choosing second-resolution, the available range grows to  ``+/- 2.9e11 years``.
+Different resolutions can be converted to each other through ``as_unit``.
+
+.. seealso::
+
+   :ref:`timeseries.oob`
+
+.. _timeseries.datetimeindex:
+
+Indexing
+--------
+
+One of the main uses for ``DatetimeIndex`` is as an index for pandas objects.
+The ``DatetimeIndex`` class contains many time series related optimizations:
+
+* A large range of dates for various offsets are pre-computed and cached
+  under the hood in order to make generating subsequent date ranges very fast
+  (just have to grab a slice).
+* Fast shifting using the ``shift`` method on pandas objects.
+* Unioning of overlapping ``DatetimeIndex`` objects with the same frequency is
+  very fast (important for fast data alignment).
+* Quick access to date fields via properties such as ``year``, ``month``, etc.
+* Regularization functions like ``snap`` and very fast ``asof`` logic.
+
+``DatetimeIndex`` objects have all the basic functionality of regular ``Index``
+objects, and a smorgasbord of advanced time series specific methods for easy
+frequency processing.
+
+.. seealso::
+    :ref:`Reindexing methods <basics.reindexing>`
+
+.. note::
+
+    While pandas does not force you to have a sorted date index, some of these
+    methods may have unexpected or incorrect behavior if the dates are unsorted.
+
+``DatetimeIndex`` can be used like a regular index and offers all of its
+intelligent functionality like selection, slicing, etc.
+
+.. ipython:: python
+
+   rng = pd.date_range(start, end, freq="BME")
+   ts = pd.Series(np.random.randn(len(rng)), index=rng)
+   ts.index
+   ts[:5].index
+   ts[::2].index
+
+.. _timeseries.partialindexing:
+
+Partial string indexing
+~~~~~~~~~~~~~~~~~~~~~~~
+
+Dates and strings that parse to timestamps can be passed as indexing parameters:
+
+.. ipython:: python
+
+   ts["1/31/2011"]
+
+   ts[datetime.datetime(2011, 12, 25):]
+
+   ts["10/31/2011":"12/31/2011"]
+
+To provide convenience for accessing longer time series, you can also pass in
+the year or year and month as strings:
+
+.. ipython:: python
+
+   ts["2011"]
+
+   ts["2011-6"]
+
+This type of slicing will work on a ``DataFrame`` with a ``DatetimeIndex`` as well. Since the
+partial string selection is a form of label slicing, the endpoints **will be** included. This
+would include matching times on an included date:
+
+.. warning::
+
+   Indexing ``DataFrame`` rows with a *single* string with getitem (e.g. ``frame[dtstring]``)
+   is deprecated starting with pandas 1.2.0 (given the ambiguity whether it is indexing
+   the rows or selecting a column) and will be removed in a future version. The equivalent
+   with ``.loc`` (e.g. ``frame.loc[dtstring]``) is still supported.
+
+.. ipython:: python
+
+   dft = pd.DataFrame(
+       np.random.randn(100000, 1),
+       columns=["A"],
+       index=pd.date_range("20130101", periods=100000, freq="min"),
+   )
+   dft
+   dft.loc["2013"]
+
+This starts on the very first time in the month, and includes the last date and
+time for the month:
+
+.. ipython:: python
+
+   dft["2013-1":"2013-2"]
+
+This specifies a stop time **that includes all of the times on the last day**:
+
+.. ipython:: python
+
+   dft["2013-1":"2013-2-28"]
+
+This specifies an **exact** stop time (and is not the same as the above):
+
+.. ipython:: python
+
+   dft["2013-1":"2013-2-28 00:00:00"]
+
+We are stopping on the included end-point as it is part of the index:
+
+.. ipython:: python
+
+   dft["2013-1-15":"2013-1-15 12:30:00"]
+
+``DatetimeIndex`` partial string indexing also works on a ``DataFrame`` with a ``MultiIndex``:
+
+.. ipython:: python
+
+   dft2 = pd.DataFrame(
+       np.random.randn(20, 1),
+       columns=["A"],
+       index=pd.MultiIndex.from_product(
+           [pd.date_range("20130101", periods=10, freq="12h"), ["a", "b"]]
+       ),
+   )
+   dft2
+   dft2.loc["2013-01-05"]
+   idx = pd.IndexSlice
+   dft2 = dft2.swaplevel(0, 1).sort_index()
+   dft2.loc[idx[:, "2013-01-05"], :]
+
+Slicing with string indexing also honors UTC offset.
+
+.. ipython:: python
+
+    df = pd.DataFrame([0], index=pd.DatetimeIndex(["2019-01-01"], tz="US/Pacific"))
+    df
+    df["2019-01-01 12:00:00+04:00":"2019-01-01 13:00:00+04:00"]
+
+.. _timeseries.slice_vs_exact_match:
+
+Slice vs. exact match
+~~~~~~~~~~~~~~~~~~~~~
+
+The same string used as an indexing parameter can be treated either as a slice or as an exact match depending on the resolution of the index. If the string is less accurate than the index, it will be treated as a slice, otherwise as an exact match.
+
+Consider a ``Series`` object with a minute resolution index:
+
+.. ipython:: python
+
+    series_minute = pd.Series(
+        [1, 2, 3],
+        pd.DatetimeIndex(
+            ["2011-12-31 23:59:00", "2012-01-01 00:00:00", "2012-01-01 00:02:00"]
+        ),
+    )
+    series_minute.index.resolution
+
+A timestamp string less accurate than a minute gives a ``Series`` object.
+
+.. ipython:: python
+
+    series_minute["2011-12-31 23"]
+
+A timestamp string with minute resolution (or more accurate), gives a scalar instead, i.e. it is not casted to a slice.
+
+.. ipython:: python
+
+    series_minute["2011-12-31 23:59"]
+    series_minute["2011-12-31 23:59:00"]
+
+If index resolution is second, then the minute-accurate timestamp gives a
+``Series``.
+
+.. ipython:: python
+
+    series_second = pd.Series(
+        [1, 2, 3],
+        pd.DatetimeIndex(
+            ["2011-12-31 23:59:59", "2012-01-01 00:00:00", "2012-01-01 00:00:01"]
+        ),
+    )
+    series_second.index.resolution
+    series_second["2011-12-31 23:59"]
+
+If the timestamp string is treated as a slice, it can be used to index ``DataFrame`` with ``.loc[]`` as well.
+
+.. ipython:: python
+
+    dft_minute = pd.DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6]}, index=series_minute.index
+    )
+    dft_minute.loc["2011-12-31 23"]
+
+
+.. warning::
+
+   However, if the string is treated as an exact match, the selection in ``DataFrame``'s ``[]`` will be column-wise and not row-wise, see :ref:`Indexing Basics <indexing.basics>`. For example ``dft_minute['2011-12-31 23:59']`` will raise ``KeyError`` as ``'2012-12-31 23:59'`` has the same resolution as the index and there is no column with such name:
+
+   To *always* have unambiguous selection, whether the row is treated as a slice or a single selection, use ``.loc``.
+
+   .. ipython:: python
+
+      dft_minute.loc["2011-12-31 23:59"]
+
+Note also that ``DatetimeIndex`` resolution cannot be less precise than day.
+
+.. ipython:: python
+
+    series_monthly = pd.Series(
+        [1, 2, 3], pd.DatetimeIndex(["2011-12", "2012-01", "2012-02"])
+    )
+    series_monthly.index.resolution
+    series_monthly["2011-12"]  # returns Series
+
+
+Exact indexing
+~~~~~~~~~~~~~~
+
+As discussed in previous section, indexing a ``DatetimeIndex`` with a partial string depends on the "accuracy" of the period, in other words how specific the interval is in relation to the resolution of the index. In contrast, indexing with ``Timestamp`` or ``datetime`` objects is exact, because the objects have exact meaning. These also follow the semantics of *including both endpoints*.
+
+These ``Timestamp`` and ``datetime`` objects have exact ``hours, minutes,`` and ``seconds``, even though they were not explicitly specified (they are ``0``).
+
+.. ipython:: python
+
+   dft[datetime.datetime(2013, 1, 1): datetime.datetime(2013, 2, 28)]
+
+With no defaults.
+
+.. ipython:: python
+
+   dft[
+       datetime.datetime(2013, 1, 1, 10, 12, 0): datetime.datetime(
+           2013, 2, 28, 10, 12, 0
+       )
+   ]
+
+Truncating & fancy indexing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A :meth:`~DataFrame.truncate` convenience function is provided that is similar
+to slicing. Note that ``truncate`` assumes a 0 value for any unspecified date
+component in a ``DatetimeIndex`` in contrast to slicing which returns any
+partially matching dates:
+
+.. ipython:: python
+
+   rng2 = pd.date_range("2011-01-01", "2012-01-01", freq="W")
+   ts2 = pd.Series(np.random.randn(len(rng2)), index=rng2)
+
+   ts2.truncate(before="2011-11", after="2011-12")
+   ts2["2011-11":"2011-12"]
+
+Even complicated fancy indexing that breaks the ``DatetimeIndex`` frequency
+regularity will result in a ``DatetimeIndex``, although frequency is lost:
+
+.. ipython:: python
+
+   ts2.iloc[[0, 2, 6]].index
+
+.. _timeseries.components:
+
+Time/date components
+--------------------
+
+There are several time/date properties that one can access from ``Timestamp`` or a collection of timestamps like a ``DatetimeIndex``.
+
+.. csv-table::
+    :header: "Property", "Description"
+    :widths: 15, 65
+
+    year, "The year of the datetime"
+    month,"The month of the datetime"
+    day,"The days of the datetime"
+    hour,"The hour of the datetime"
+    minute,"The minutes of the datetime"
+    second,"The seconds of the datetime"
+    microsecond,"The microseconds of the datetime"
+    nanosecond,"The nanoseconds of the datetime"
+    date,"Returns datetime.date (does not contain timezone information)"
+    time,"Returns datetime.time (does not contain timezone information)"
+    timetz,"Returns datetime.time as local time with timezone information"
+    dayofyear,"The ordinal day of year"
+    day_of_year,"The ordinal day of year"
+    dayofweek,"The number of the day of the week with Monday=0, Sunday=6"
+    day_of_week,"The number of the day of the week with Monday=0, Sunday=6"
+    weekday,"The number of the day of the week with Monday=0, Sunday=6"
+    quarter,"Quarter of the date: Jan-Mar = 1, Apr-Jun = 2, etc."
+    days_in_month,"The number of days in the month of the datetime"
+    is_month_start,"Logical indicating if first day of month (defined by frequency)"
+    is_month_end,"Logical indicating if last day of month (defined by frequency)"
+    is_quarter_start,"Logical indicating if first day of quarter (defined by frequency)"
+    is_quarter_end,"Logical indicating if last day of quarter (defined by frequency)"
+    is_year_start,"Logical indicating if first day of year (defined by frequency)"
+    is_year_end,"Logical indicating if last day of year (defined by frequency)"
+    is_leap_year,"Logical indicating if the date belongs to a leap year"
+
+.. note::
+
+   You can use ``DatetimeIndex.isocalendar().week`` to access week of year date information.
+
+Furthermore, if you have a ``Series`` with datetimelike values, then you can
+access these properties via the ``.dt`` accessor, as detailed in the section
+on :ref:`.dt accessors<basics.dt_accessors>`.
+
+You may obtain the year, week and day components of the ISO year from the ISO 8601 standard:
+
+.. ipython:: python
+
+   idx = pd.date_range(start="2019-12-29", freq="D", periods=4)
+   idx.isocalendar()
+   idx.to_series().dt.isocalendar()
+
+.. _timeseries.offsets:
+
+DateOffset objects
+------------------
+
+In the preceding examples, frequency strings (e.g. ``'D'``) were used to specify
+a frequency that defined:
+
+* how the date times in :class:`DatetimeIndex` were spaced when using :meth:`date_range`
+* the frequency of a :class:`Period` or :class:`PeriodIndex`
+
+These frequency strings map to a :class:`DateOffset` object and its subclasses. A :class:`DateOffset`
+is similar to a :class:`Timedelta` that represents a duration of time but follows specific calendar duration rules.
+For example, a :class:`Timedelta` day will always increment ``datetimes`` by 24 hours, while a :class:`DateOffset` day
+will increment ``datetimes`` to the same time the next day whether a day represents 23, 24 or 25 hours due to daylight
+savings time. However, all :class:`DateOffset` subclasses that are an hour or smaller
+(``Hour``, ``Minute``, ``Second``, ``Milli``, ``Micro``, ``Nano``) behave like
+:class:`Timedelta` and respect absolute time.
+
+The basic :class:`DateOffset` acts similar to ``dateutil.relativedelta`` (`relativedelta documentation`_)
+that shifts a date time by the corresponding calendar duration specified. The
+arithmetic operator (``+``) can be used to perform the shift.
+
+.. ipython:: python
+
+   # This particular day contains a day light savings time transition
+   ts = pd.Timestamp("2016-10-30 00:00:00", tz="Europe/Helsinki")
+   # Respects absolute time
+   ts + pd.Timedelta(days=1)
+   # Respects calendar time
+   ts + pd.DateOffset(days=1)
+   friday = pd.Timestamp("2018-01-05")
+   friday.day_name()
+   # Add 2 business days (Friday --> Tuesday)
+   two_business_days = 2 * pd.offsets.BDay()
+   friday + two_business_days
+   (friday + two_business_days).day_name()
+
+
+Most ``DateOffsets`` have associated frequencies strings, or offset aliases, that can be passed
+into ``freq`` keyword arguments. The available date offsets and associated frequency strings can be found below:
+
+.. csv-table::
+    :header: "Date Offset", "Frequency String", "Description"
+    :widths: 15, 15, 65
+
+    :class:`~pandas.tseries.offsets.DateOffset`, None, "Generic offset class, defaults to absolute 24 hours"
+    :class:`~pandas.tseries.offsets.BDay` or :class:`~pandas.tseries.offsets.BusinessDay`, ``'B'``,"business day (weekday)"
+    :class:`~pandas.tseries.offsets.CDay` or :class:`~pandas.tseries.offsets.CustomBusinessDay`, ``'C'``, "custom business day"
+    :class:`~pandas.tseries.offsets.Week`, ``'W'``, "one week, optionally anchored on a day of the week"
+    :class:`~pandas.tseries.offsets.WeekOfMonth`, ``'WOM'``, "the x-th day of the y-th week of each month"
+    :class:`~pandas.tseries.offsets.LastWeekOfMonth`, ``'LWOM'``, "the x-th day of the last week of each month"
+    :class:`~pandas.tseries.offsets.MonthEnd`, ``'ME'``, "calendar month end"
+    :class:`~pandas.tseries.offsets.MonthBegin`, ``'MS'``, "calendar month begin"
+    :class:`~pandas.tseries.offsets.BMonthEnd` or :class:`~pandas.tseries.offsets.BusinessMonthEnd`, ``'BME'``, "business month end"
+    :class:`~pandas.tseries.offsets.BMonthBegin` or :class:`~pandas.tseries.offsets.BusinessMonthBegin`, ``'BMS'``, "business month begin"
+    :class:`~pandas.tseries.offsets.CBMonthEnd` or :class:`~pandas.tseries.offsets.CustomBusinessMonthEnd`, ``'CBME'``, "custom business month end"
+    :class:`~pandas.tseries.offsets.CBMonthBegin` or :class:`~pandas.tseries.offsets.CustomBusinessMonthBegin`, ``'CBMS'``, "custom business month begin"
+    :class:`~pandas.tseries.offsets.SemiMonthEnd`, ``'SME'``, "15th (or other day_of_month) and calendar month end"
+    :class:`~pandas.tseries.offsets.SemiMonthBegin`, ``'SMS'``, "15th (or other day_of_month) and calendar month begin"
+    :class:`~pandas.tseries.offsets.QuarterEnd`, ``'QE'``, "calendar quarter end"
+    :class:`~pandas.tseries.offsets.QuarterBegin`, ``'QS'``, "calendar quarter begin"
+    :class:`~pandas.tseries.offsets.BQuarterEnd`, ``'BQE``, "business quarter end"
+    :class:`~pandas.tseries.offsets.BQuarterBegin`, ``'BQS'``, "business quarter begin"
+    :class:`~pandas.tseries.offsets.FY5253Quarter`, ``'REQ'``, "retail (aka 52-53 week) quarter"
+    :class:`~pandas.tseries.offsets.YearEnd`, ``'YE'``, "calendar year end"
+    :class:`~pandas.tseries.offsets.YearBegin`, ``'YS'`` or ``'BYS'``,"calendar year begin"
+    :class:`~pandas.tseries.offsets.BYearEnd`, ``'BYE'``, "business year end"
+    :class:`~pandas.tseries.offsets.BYearBegin`, ``'BYS'``, "business year begin"
+    :class:`~pandas.tseries.offsets.FY5253`, ``'RE'``, "retail (aka 52-53 week) year"
+    :class:`~pandas.tseries.offsets.Easter`, None, "Easter holiday"
+    :class:`~pandas.tseries.offsets.BusinessHour`, ``'bh'``, "business hour"
+    :class:`~pandas.tseries.offsets.CustomBusinessHour`, ``'cbh'``, "custom business hour"
+    :class:`~pandas.tseries.offsets.Day`, ``'D'``, "one absolute day"
+    :class:`~pandas.tseries.offsets.Hour`, ``'h'``, "one hour"
+    :class:`~pandas.tseries.offsets.Minute`, ``'min'``,"one minute"
+    :class:`~pandas.tseries.offsets.Second`, ``'s'``, "one second"
+    :class:`~pandas.tseries.offsets.Milli`, ``'ms'``, "one millisecond"
+    :class:`~pandas.tseries.offsets.Micro`, ``'us'``, "one microsecond"
+    :class:`~pandas.tseries.offsets.Nano`, ``'ns'``, "one nanosecond"
+
+``DateOffsets`` additionally have :meth:`rollforward` and :meth:`rollback`
+methods for moving a date forward or backward respectively to a valid offset
+date relative to the offset. For example, business offsets will roll dates
+that land on the weekends (Saturday and Sunday) forward to Monday since
+business offsets operate on the weekdays.
+
+.. ipython:: python
+
+   ts = pd.Timestamp("2018-01-06 00:00:00")
+   ts.day_name()
+   # BusinessHour's valid offset dates are Monday through Friday
+   offset = pd.offsets.BusinessHour(start="09:00")
+   # Bring the date to the closest offset date (Monday)
+   offset.rollforward(ts)
+   # Date is brought to the closest offset date first and then the hour is added
+   ts + offset
+
+These operations preserve time (hour, minute, etc) information by default.
+To reset time to midnight, use :meth:`normalize` before or after applying
+the operation (depending on whether you want the time information included
+in the operation).
+
+.. ipython:: python
+
+   ts = pd.Timestamp("2014-01-01 09:00")
+   day = pd.offsets.Day()
+   day + ts
+   (day + ts).normalize()
+
+   ts = pd.Timestamp("2014-01-01 22:00")
+   hour = pd.offsets.Hour()
+   hour + ts
+   (hour + ts).normalize()
+   (hour + pd.Timestamp("2014-01-01 23:30")).normalize()
+
+.. _relativedelta documentation: https://dateutil.readthedocs.io/en/stable/relativedelta.html
+
+
+Parametric offsets
+~~~~~~~~~~~~~~~~~~
+
+Some of the offsets can be "parameterized" when created to result in different
+behaviors. For example, the ``Week`` offset for generating weekly data accepts a
+``weekday`` parameter which results in the generated dates always lying on a
+particular day of the week:
+
+.. ipython:: python
+
+   d = datetime.datetime(2008, 8, 18, 9, 0)
+   d
+   d + pd.offsets.Week()
+   d + pd.offsets.Week(weekday=4)
+   (d + pd.offsets.Week(weekday=4)).weekday()
+
+   d - pd.offsets.Week()
+
+The ``normalize`` option will be effective for addition and subtraction.
+
+.. ipython:: python
+
+   d + pd.offsets.Week(normalize=True)
+   d - pd.offsets.Week(normalize=True)
+
+
+Another example is parameterizing ``YearEnd`` with the specific ending month:
+
+.. ipython:: python
+
+   d + pd.offsets.YearEnd()
+   d + pd.offsets.YearEnd(month=6)
+
+
+.. _timeseries.offsetseries:
+
+Using offsets with ``Series`` / ``DatetimeIndex``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Offsets can be used with either a ``Series`` or ``DatetimeIndex`` to
+apply the offset to each element.
+
+.. ipython:: python
+
+   rng = pd.date_range("2012-01-01", "2012-01-03")
+   s = pd.Series(rng)
+   rng
+   rng + pd.DateOffset(months=2)
+   s + pd.DateOffset(months=2)
+   s - pd.DateOffset(months=2)
+
+If the offset class maps directly to a ``Timedelta`` (``Day``, ``Hour``,
+``Minute``, ``Second``, ``Micro``, ``Milli``, ``Nano``) it can be
+used exactly like a ``Timedelta`` - see the
+:ref:`Timedelta section<timedeltas.operations>` for more examples.
+
+.. ipython:: python
+
+   s - pd.offsets.Day(2)
+   td = s - pd.Series(pd.date_range("2011-12-29", "2011-12-31"))
+   td
+   td + pd.offsets.Minute(15)
+
+Note that some offsets (such as ``BQuarterEnd``) do not have a
+vectorized implementation.  They can still be used but may
+calculate significantly slower and will show a ``PerformanceWarning``
+
+.. ipython:: python
+   :okwarning:
+
+   rng + pd.offsets.BQuarterEnd()
+
+
+.. _timeseries.custombusinessdays:
+
+Custom business days
+~~~~~~~~~~~~~~~~~~~~
+
+The ``CDay`` or ``CustomBusinessDay`` class provides a parametric
+``BusinessDay`` class which can be used to create customized business day
+calendars which account for local holidays and local weekend conventions.
+
+As an interesting example, let's look at Egypt where a Friday-Saturday weekend is observed.
+
+.. ipython:: python
+
+    weekmask_egypt = "Sun Mon Tue Wed Thu"
+
+    # They also observe International Workers' Day so let's
+    # add that for a couple of years
+
+    holidays = [
+        "2012-05-01",
+        datetime.datetime(2013, 5, 1),
+        np.datetime64("2014-05-01"),
+    ]
+    bday_egypt = pd.offsets.CustomBusinessDay(
+        holidays=holidays,
+        weekmask=weekmask_egypt,
+    )
+    dt = datetime.datetime(2013, 4, 30)
+    dt + 2 * bday_egypt
+
+Let's map to the weekday names:
+
+.. ipython:: python
+
+    dts = pd.date_range(dt, periods=5, freq=bday_egypt)
+
+    pd.Series(dts.weekday, dts).map(pd.Series("Mon Tue Wed Thu Fri Sat Sun".split()))
+
+Holiday calendars can be used to provide the list of holidays.  See the
+:ref:`holiday calendar<timeseries.holiday>` section for more information.
+
+.. ipython:: python
+
+    from pandas.tseries.holiday import USFederalHolidayCalendar
+
+    bday_us = pd.offsets.CustomBusinessDay(calendar=USFederalHolidayCalendar())
+
+    # Friday before MLK Day
+    dt = datetime.datetime(2014, 1, 17)
+
+    # Tuesday after MLK Day (Monday is skipped because it's a holiday)
+    dt + bday_us
+
+Monthly offsets that respect a certain holiday calendar can be defined
+in the usual way.
+
+.. ipython:: python
+
+    bmth_us = pd.offsets.CustomBusinessMonthBegin(calendar=USFederalHolidayCalendar())
+
+    # Skip new years
+    dt = datetime.datetime(2013, 12, 17)
+    dt + bmth_us
+
+    # Define date index with custom offset
+    pd.date_range(start="20100101", end="20120101", freq=bmth_us)
+
+.. note::
+
+    The frequency string 'C' is used to indicate that a CustomBusinessDay
+    DateOffset is used, it is important to note that since CustomBusinessDay is
+    a parameterised type, instances of CustomBusinessDay may differ and this is
+    not detectable from the 'C' frequency string. The user therefore needs to
+    ensure that the 'C' frequency string is used consistently within the user's
+    application.
+
+.. _timeseries.businesshour:
+
+Business hour
+~~~~~~~~~~~~~
+
+The ``BusinessHour`` class provides a business hour representation on ``BusinessDay``,
+allowing to use specific start and end times.
+
+By default, ``BusinessHour`` uses 9:00 - 17:00 as business hours.
+Adding ``BusinessHour`` will increment ``Timestamp`` by hourly frequency.
+If target ``Timestamp`` is out of business hours, move to the next business hour
+then increment it. If the result exceeds the business hours end, the remaining
+hours are added to the next business day.
+
+.. ipython:: python
+
+    bh = pd.offsets.BusinessHour()
+    bh
+
+    # 2014-08-01 is Friday
+    pd.Timestamp("2014-08-01 10:00").weekday()
+    pd.Timestamp("2014-08-01 10:00") + bh
+
+    # Below example is the same as: pd.Timestamp('2014-08-01 09:00') + bh
+    pd.Timestamp("2014-08-01 08:00") + bh
+
+    # If the results is on the end time, move to the next business day
+    pd.Timestamp("2014-08-01 16:00") + bh
+
+    # Remainings are added to the next day
+    pd.Timestamp("2014-08-01 16:30") + bh
+
+    # Adding 2 business hours
+    pd.Timestamp("2014-08-01 10:00") + pd.offsets.BusinessHour(2)
+
+    # Subtracting 3 business hours
+    pd.Timestamp("2014-08-01 10:00") + pd.offsets.BusinessHour(-3)
+
+You can also specify ``start`` and ``end`` time by keywords. The argument must
+be a ``str`` with an ``hour:minute`` representation or a ``datetime.time``
+instance. Specifying seconds, microseconds and nanoseconds as business hour
+results in ``ValueError``.
+
+.. ipython:: python
+
+    bh = pd.offsets.BusinessHour(start="11:00", end=datetime.time(20, 0))
+    bh
+
+    pd.Timestamp("2014-08-01 13:00") + bh
+    pd.Timestamp("2014-08-01 09:00") + bh
+    pd.Timestamp("2014-08-01 18:00") + bh
+
+Passing ``start`` time later than ``end`` represents midnight business hour.
+In this case, business hour exceeds midnight and overlap to the next day.
+Valid business hours are distinguished by whether it started from valid ``BusinessDay``.
+
+.. ipython:: python
+
+    bh = pd.offsets.BusinessHour(start="17:00", end="09:00")
+    bh
+
+    pd.Timestamp("2014-08-01 17:00") + bh
+    pd.Timestamp("2014-08-01 23:00") + bh
+
+    # Although 2014-08-02 is Saturday,
+    # it is valid because it starts from 08-01 (Friday).
+    pd.Timestamp("2014-08-02 04:00") + bh
+
+    # Although 2014-08-04 is Monday,
+    # it is out of business hours because it starts from 08-03 (Sunday).
+    pd.Timestamp("2014-08-04 04:00") + bh
+
+Applying ``BusinessHour.rollforward`` and ``rollback`` to out of business hours results in
+the next business hour start or previous day's end. Different from other offsets, ``BusinessHour.rollforward``
+may output different results from ``apply`` by definition.
+
+This is because one day's business hour end is equal to next day's business hour start. For example,
+under the default business hours (9:00 - 17:00), there is no gap (0 minutes) between ``2014-08-01 17:00`` and
+``2014-08-04 09:00``.
+
+.. ipython:: python
+
+    # This adjusts a Timestamp to business hour edge
+    pd.offsets.BusinessHour().rollback(pd.Timestamp("2014-08-02 15:00"))
+    pd.offsets.BusinessHour().rollforward(pd.Timestamp("2014-08-02 15:00"))
+
+    # It is the same as BusinessHour() + pd.Timestamp('2014-08-01 17:00').
+    # And it is the same as BusinessHour() + pd.Timestamp('2014-08-04 09:00')
+    pd.offsets.BusinessHour() + pd.Timestamp("2014-08-02 15:00")
+
+    # BusinessDay results (for reference)
+    pd.offsets.BusinessHour().rollforward(pd.Timestamp("2014-08-02"))
+
+    # It is the same as BusinessDay() + pd.Timestamp('2014-08-01')
+    # The result is the same as rollworward because BusinessDay never overlap.
+    pd.offsets.BusinessHour() + pd.Timestamp("2014-08-02")
+
+``BusinessHour`` regards Saturday and Sunday as holidays. To use arbitrary
+holidays, you can use ``CustomBusinessHour`` offset, as explained in the
+following subsection.
+
+.. _timeseries.custombusinesshour:
+
+Custom business hour
+~~~~~~~~~~~~~~~~~~~~
+
+The ``CustomBusinessHour`` is a mixture of ``BusinessHour`` and ``CustomBusinessDay`` which
+allows you to specify arbitrary holidays. ``CustomBusinessHour`` works as the same
+as ``BusinessHour`` except that it skips specified custom holidays.
+
+.. ipython:: python
+
+    from pandas.tseries.holiday import USFederalHolidayCalendar
+
+    bhour_us = pd.offsets.CustomBusinessHour(calendar=USFederalHolidayCalendar())
+    # Friday before MLK Day
+    dt = datetime.datetime(2014, 1, 17, 15)
+
+    dt + bhour_us
+
+    # Tuesday after MLK Day (Monday is skipped because it's a holiday)
+    dt + bhour_us * 2
+
+You can use keyword arguments supported by either ``BusinessHour`` and ``CustomBusinessDay``.
+
+.. ipython:: python
+
+    bhour_mon = pd.offsets.CustomBusinessHour(start="10:00", weekmask="Tue Wed Thu Fri")
+
+    # Monday is skipped because it's a holiday, business hour starts from 10:00
+    dt + bhour_mon * 2
+
+.. _timeseries.offset_aliases:
+
+Offset aliases
+~~~~~~~~~~~~~~
+
+A number of string aliases are given to useful common time series
+frequencies. We will refer to these aliases as *offset aliases*.
+
+.. csv-table::
+    :header: "Alias", "Description"
+    :widths: 15, 100
+
+    "B", "business day frequency"
+    "C", "custom business day frequency"
+    "D", "calendar day frequency"
+    "W", "weekly frequency"
+    "ME", "month end frequency"
+    "SME", "semi-month end frequency (15th and end of month)"
+    "BME", "business month end frequency"
+    "CBME", "custom business month end frequency"
+    "MS", "month start frequency"
+    "SMS", "semi-month start frequency (1st and 15th)"
+    "BMS", "business month start frequency"
+    "CBMS", "custom business month start frequency"
+    "QE", "quarter end frequency"
+    "BQE", "business quarter end frequency"
+    "QS", "quarter start frequency"
+    "BQS", "business quarter start frequency"
+    "YE", "year end frequency"
+    "BYE", "business year end frequency"
+    "YS", "year start frequency"
+    "BYS", "business year start frequency"
+    "h", "hourly frequency"
+    "bh", "business hour frequency"
+    "cbh", "custom business hour frequency"
+    "min", "minutely frequency"
+    "s", "secondly frequency"
+    "ms", "milliseconds"
+    "us", "microseconds"
+    "ns", "nanoseconds"
+
+.. deprecated:: 2.2.0
+
+   Aliases ``H``, ``BH``, ``CBH``, ``T``, ``S``, ``L``, ``U``, and ``N``
+   are deprecated in favour of the aliases ``h``, ``bh``, ``cbh``,
+   ``min``, ``s``, ``ms``, ``us``, and ``ns``.
+
+   Aliases ``Y``, ``M``, and ``Q`` are deprecated in favour of the aliases
+   ``YE``, ``ME``, ``QE``.
+
+
+.. note::
+
+    When using the offset aliases above, it should be noted that functions
+    such as :func:`date_range`, :func:`bdate_range`, will only return
+    timestamps that are in the interval defined by ``start_date`` and
+    ``end_date``. If the ``start_date`` does not correspond to the frequency,
+    the returned timestamps will start at the next valid timestamp, same for
+    ``end_date``, the returned timestamps will stop at the previous valid
+    timestamp.
+
+   For example, for the offset ``MS``, if the ``start_date`` is not the first
+   of the month, the returned timestamps will start with the first day of the
+   next month. If ``end_date`` is not the first day of a month, the last
+   returned timestamp will be the first day of the corresponding month.
+
+   .. ipython:: python
+
+       dates_lst_1 = pd.date_range("2020-01-06", "2020-04-03", freq="MS")
+       dates_lst_1
+
+       dates_lst_2 = pd.date_range("2020-01-01", "2020-04-01", freq="MS")
+       dates_lst_2
+
+   We can see in the above example :func:`date_range` and
+   :func:`bdate_range` will only return the valid timestamps between the
+   ``start_date`` and ``end_date``. If these are not valid timestamps for the
+   given frequency it will roll to the next value for ``start_date``
+   (respectively previous for the ``end_date``)
+
+.. _timeseries.period_aliases:
+
+Period aliases
+~~~~~~~~~~~~~~
+
+A number of string aliases are given to useful common time series
+frequencies. We will refer to these aliases as *period aliases*.
+
+.. csv-table::
+    :header: "Alias", "Description"
+    :widths: 15, 100
+
+    "B", "business day frequency"
+    "D", "calendar day frequency"
+    "W", "weekly frequency"
+    "M", "monthly frequency"
+    "Q", "quarterly frequency"
+    "Y", "yearly frequency"
+    "h", "hourly frequency"
+    "min", "minutely frequency"
+    "s", "secondly frequency"
+    "ms", "milliseconds"
+    "us", "microseconds"
+    "ns", "nanoseconds"
+
+.. deprecated:: 2.2.0
+
+   Aliases ``H``, ``T``, ``S``, ``L``, ``U``, and ``N`` are deprecated in favour of the aliases
+   ``h``, ``min``, ``s``, ``ms``, ``us``, and ``ns``.
+
+
+Combining aliases
+~~~~~~~~~~~~~~~~~
+
+As we have seen previously, the alias and the offset instance are fungible in
+most functions:
+
+.. ipython:: python
+
+   pd.date_range(start, periods=5, freq="B")
+
+   pd.date_range(start, periods=5, freq=pd.offsets.BDay())
+
+You can combine together day and intraday offsets:
+
+.. ipython:: python
+
+   pd.date_range(start, periods=10, freq="2h20min")
+
+   pd.date_range(start, periods=10, freq="1D10us")
+
+Anchored offsets
+~~~~~~~~~~~~~~~~
+
+For some frequencies you can specify an anchoring suffix:
+
+.. csv-table::
+    :header: "Alias", "Description"
+    :widths: 15, 100
+
+    "W\-SUN", "weekly frequency (Sundays). Same as 'W'"
+    "W\-MON", "weekly frequency (Mondays)"
+    "W\-TUE", "weekly frequency (Tuesdays)"
+    "W\-WED", "weekly frequency (Wednesdays)"
+    "W\-THU", "weekly frequency (Thursdays)"
+    "W\-FRI", "weekly frequency (Fridays)"
+    "W\-SAT", "weekly frequency (Saturdays)"
+    "(B)Q(E)(S)\-DEC", "quarterly frequency, year ends in December. Same as 'QE'"
+    "(B)Q(E)(S)\-JAN", "quarterly frequency, year ends in January"
+    "(B)Q(E)(S)\-FEB", "quarterly frequency, year ends in February"
+    "(B)Q(E)(S)\-MAR", "quarterly frequency, year ends in March"
+    "(B)Q(E)(S)\-APR", "quarterly frequency, year ends in April"
+    "(B)Q(E)(S)\-MAY", "quarterly frequency, year ends in May"
+    "(B)Q(E)(S)\-JUN", "quarterly frequency, year ends in June"
+    "(B)Q(E)(S)\-JUL", "quarterly frequency, year ends in July"
+    "(B)Q(E)(S)\-AUG", "quarterly frequency, year ends in August"
+    "(B)Q(E)(S)\-SEP", "quarterly frequency, year ends in September"
+    "(B)Q(E)(S)\-OCT", "quarterly frequency, year ends in October"
+    "(B)Q(E)(S)\-NOV", "quarterly frequency, year ends in November"
+    "(B)Y(E)(S)\-DEC", "annual frequency, anchored end of December. Same as 'YE'"
+    "(B)Y(E)(S)\-JAN", "annual frequency, anchored end of January"
+    "(B)Y(E)(S)\-FEB", "annual frequency, anchored end of February"
+    "(B)Y(E)(S)\-MAR", "annual frequency, anchored end of March"
+    "(B)Y(E)(S)\-APR", "annual frequency, anchored end of April"
+    "(B)Y(E)(S)\-MAY", "annual frequency, anchored end of May"
+    "(B)Y(E)(S)\-JUN", "annual frequency, anchored end of June"
+    "(B)Y(E)(S)\-JUL", "annual frequency, anchored end of July"
+    "(B)Y(E)(S)\-AUG", "annual frequency, anchored end of August"
+    "(B)Y(E)(S)\-SEP", "annual frequency, anchored end of September"
+    "(B)Y(E)(S)\-OCT", "annual frequency, anchored end of October"
+    "(B)Y(E)(S)\-NOV", "annual frequency, anchored end of November"
+
+These can be used as arguments to ``date_range``, ``bdate_range``, constructors
+for ``DatetimeIndex``, as well as various other timeseries-related functions
+in pandas.
+
+Anchored offset semantics
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+For those offsets that are anchored to the start or end of specific
+frequency (``MonthEnd``, ``MonthBegin``, ``WeekEnd``, etc), the following
+rules apply to rolling forward and backwards.
+
+When ``n`` is not 0, if the given date is not on an anchor point, it snapped to the next(previous)
+anchor point, and moved ``|n|-1`` additional steps forwards or backwards.
+
+.. ipython:: python
+
+   pd.Timestamp("2014-01-02") + pd.offsets.MonthBegin(n=1)
+   pd.Timestamp("2014-01-02") + pd.offsets.MonthEnd(n=1)
+
+   pd.Timestamp("2014-01-02") - pd.offsets.MonthBegin(n=1)
+   pd.Timestamp("2014-01-02") - pd.offsets.MonthEnd(n=1)
+
+   pd.Timestamp("2014-01-02") + pd.offsets.MonthBegin(n=4)
+   pd.Timestamp("2014-01-02") - pd.offsets.MonthBegin(n=4)
+
+If the given date *is* on an anchor point, it is moved ``|n|`` points forwards
+or backwards.
+
+.. ipython:: python
+
+   pd.Timestamp("2014-01-01") + pd.offsets.MonthBegin(n=1)
+   pd.Timestamp("2014-01-31") + pd.offsets.MonthEnd(n=1)
+
+   pd.Timestamp("2014-01-01") - pd.offsets.MonthBegin(n=1)
+   pd.Timestamp("2014-01-31") - pd.offsets.MonthEnd(n=1)
+
+   pd.Timestamp("2014-01-01") + pd.offsets.MonthBegin(n=4)
+   pd.Timestamp("2014-01-31") - pd.offsets.MonthBegin(n=4)
+
+For the case when ``n=0``, the date is not moved if on an anchor point, otherwise
+it is rolled forward to the next anchor point.
+
+.. ipython:: python
+
+   pd.Timestamp("2014-01-02") + pd.offsets.MonthBegin(n=0)
+   pd.Timestamp("2014-01-02") + pd.offsets.MonthEnd(n=0)
+
+   pd.Timestamp("2014-01-01") + pd.offsets.MonthBegin(n=0)
+   pd.Timestamp("2014-01-31") + pd.offsets.MonthEnd(n=0)
+
+.. _timeseries.holiday:
+
+Holidays / holiday calendars
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Holidays and calendars provide a simple way to define holiday rules to be used
+with ``CustomBusinessDay`` or in other analysis that requires a predefined
+set of holidays.  The ``AbstractHolidayCalendar`` class provides all the necessary
+methods to return a list of holidays and only ``rules`` need to be defined
+in a specific holiday calendar class. Furthermore, the ``start_date`` and ``end_date``
+class attributes determine over what date range holidays are generated.  These
+should be overwritten on the ``AbstractHolidayCalendar`` class to have the range
+apply to all calendar subclasses.  ``USFederalHolidayCalendar`` is the
+only calendar that exists and primarily serves as an example for developing
+other calendars.
+
+For holidays that occur on fixed dates (e.g., US Memorial Day or July 4th) an
+observance rule determines when that holiday is observed if it falls on a weekend
+or some other non-observed day.  Defined observance rules are:
+
+.. csv-table::
+    :header: "Rule", "Description"
+    :widths: 15, 70
+
+    "next_workday", "move Saturday and Sunday to Monday"
+    "previous_workday", "move Saturday and Sunday to Friday"
+    "nearest_workday", "move Saturday to Friday and Sunday to Monday"
+    "before_nearest_workday", "apply ``nearest_workday`` and then move to previous workday before that day"
+    "after_nearest_workday", "apply ``nearest_workday`` and then move to next workday after that day"
+    "sunday_to_monday", "move Sunday to following Monday"
+    "next_monday_or_tuesday", "move Saturday to Monday and Sunday/Monday to Tuesday"
+    "previous_friday", "move Saturday and Sunday to previous Friday"
+    "next_monday", "move Saturday and Sunday to following Monday"
+    "weekend_to_monday", "same as ``next_monday``"
+
+An example of how holidays and holiday calendars are defined:
+
+.. ipython:: python
+
+    from pandas.tseries.holiday import (
+        Holiday,
+        USMemorialDay,
+        AbstractHolidayCalendar,
+        nearest_workday,
+        MO,
+    )
+
+    class ExampleCalendar(AbstractHolidayCalendar):
+        rules = [
+            USMemorialDay,
+            Holiday("July 4th", month=7, day=4, observance=nearest_workday),
+            Holiday(
+                "Columbus Day",
+                month=10,
+                day=1,
+                offset=pd.DateOffset(weekday=MO(2)),
+            ),
+        ]
+
+    cal = ExampleCalendar()
+    cal.holidays(datetime.datetime(2012, 1, 1), datetime.datetime(2012, 12, 31))
+
+:hint:
+   **weekday=MO(2)** is same as **2 * Week(weekday=2)**
+
+Using this calendar, creating an index or doing offset arithmetic skips weekends
+and holidays (i.e., Memorial Day/July 4th).  For example, the below defines
+a custom business day offset using the ``ExampleCalendar``.  Like any other offset,
+it can be used to create a ``DatetimeIndex`` or added to ``datetime``
+or ``Timestamp`` objects.
+
+.. ipython:: python
+
+    pd.date_range(
+        start="7/1/2012", end="7/10/2012", freq=pd.offsets.CDay(calendar=cal)
+    ).to_pydatetime()
+    offset = pd.offsets.CustomBusinessDay(calendar=cal)
+    datetime.datetime(2012, 5, 25) + offset
+    datetime.datetime(2012, 7, 3) + offset
+    datetime.datetime(2012, 7, 3) + 2 * offset
+    datetime.datetime(2012, 7, 6) + offset
+
+Ranges are defined by the ``start_date`` and ``end_date`` class attributes
+of ``AbstractHolidayCalendar``.  The defaults are shown below.
+
+.. ipython:: python
+
+    AbstractHolidayCalendar.start_date
+    AbstractHolidayCalendar.end_date
+
+These dates can be overwritten by setting the attributes as
+datetime/Timestamp/string.
+
+.. ipython:: python
+
+    AbstractHolidayCalendar.start_date = datetime.datetime(2012, 1, 1)
+    AbstractHolidayCalendar.end_date = datetime.datetime(2012, 12, 31)
+    cal.holidays()
+
+Every calendar class is accessible by name using the ``get_calendar`` function
+which returns a holiday class instance.  Any imported calendar class will
+automatically be available by this function.  Also, ``HolidayCalendarFactory``
+provides an easy interface to create calendars that are combinations of calendars
+or calendars with additional rules.
+
+.. ipython:: python
+
+    from pandas.tseries.holiday import get_calendar, HolidayCalendarFactory, USLaborDay
+
+    cal = get_calendar("ExampleCalendar")
+    cal.rules
+    new_cal = HolidayCalendarFactory("NewExampleCalendar", cal, USLaborDay)
+    new_cal.rules
+
+.. _timeseries.advanced_datetime:
+
+Time Series-related instance methods
+------------------------------------
+
+Shifting / lagging
+~~~~~~~~~~~~~~~~~~
+
+One may want to *shift* or *lag* the values in a time series back and forward in
+time. The method for this is :meth:`~Series.shift`, which is available on all of
+the pandas objects.
+
+.. ipython:: python
+
+   ts = pd.Series(range(len(rng)), index=rng)
+   ts = ts[:5]
+   ts.shift(1)
+
+The ``shift`` method accepts a ``freq`` argument which can accept a
+``DateOffset`` class or other ``timedelta``-like object or also an
+:ref:`offset alias <timeseries.offset_aliases>`.
+
+When ``freq`` is specified, ``shift`` method changes all the dates in the index
+rather than changing the alignment of the data and the index:
+
+.. ipython:: python
+
+   ts.shift(5, freq="D")
+   ts.shift(5, freq=pd.offsets.BDay())
+   ts.shift(5, freq="BME")
+
+Note that with when ``freq`` is specified, the leading entry is no longer NaN
+because the data is not being realigned.
+
+Frequency conversion
+~~~~~~~~~~~~~~~~~~~~
+
+The primary function for changing frequencies is the :meth:`~Series.asfreq`
+method. For a ``DatetimeIndex``, this is basically just a thin, but convenient
+wrapper around :meth:`~Series.reindex`  which generates a ``date_range`` and
+calls ``reindex``.
+
+.. ipython:: python
+
+   dr = pd.date_range("1/1/2010", periods=3, freq=3 * pd.offsets.BDay())
+   ts = pd.Series(np.random.randn(3), index=dr)
+   ts
+   ts.asfreq(pd.offsets.BDay())
+
+``asfreq`` provides a further convenience so you can specify an interpolation
+method for any gaps that may appear after the frequency conversion.
+
+.. ipython:: python
+
+   ts.asfreq(pd.offsets.BDay(), method="pad")
+
+Filling forward / backward
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Related to ``asfreq`` and ``reindex`` is :meth:`~Series.fillna`, which is
+documented in the :ref:`missing data section <missing_data.fillna>`.
+
+Converting to Python datetimes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``DatetimeIndex`` can be converted to an array of Python native
+:py:class:`datetime.datetime` objects using the ``to_pydatetime`` method.
+
+.. _timeseries.resampling:
+
+Resampling
+----------
+
+pandas has a simple, powerful, and efficient functionality for performing
+resampling operations during frequency conversion (e.g., converting secondly
+data into 5-minutely data). This is extremely common in, but not limited to,
+financial applications.
+
+:meth:`~Series.resample` is a time-based groupby, followed by a reduction method
+on each of its groups. See some :ref:`cookbook examples <cookbook.resample>` for
+some advanced strategies.
+
+The ``resample()`` method can be used directly from ``DataFrameGroupBy`` objects,
+see the :ref:`groupby docs <groupby.transform.window_resample>`.
+
+Basics
+~~~~~~
+
+.. ipython:: python
+
+   rng = pd.date_range("1/1/2012", periods=100, freq="s")
+
+   ts = pd.Series(np.random.randint(0, 500, len(rng)), index=rng)
+
+   ts.resample("5Min").sum()
+
+The ``resample`` function is very flexible and allows you to specify many
+different parameters to control the frequency conversion and resampling
+operation.
+
+Any built-in method available via :ref:`GroupBy <api.groupby>` is available as
+a method of the returned object, including ``sum``, ``mean``, ``std``, ``sem``,
+``max``, ``min``, ``median``, ``first``, ``last``, ``ohlc``:
+
+.. ipython:: python
+
+   ts.resample("5Min").mean()
+
+   ts.resample("5Min").ohlc()
+
+   ts.resample("5Min").max()
+
+
+For downsampling, ``closed`` can be set to 'left' or 'right' to specify which
+end of the interval is closed:
+
+.. ipython:: python
+
+   ts.resample("5Min", closed="right").mean()
+
+   ts.resample("5Min", closed="left").mean()
+
+Parameters like ``label`` are used to manipulate the resulting labels.
+``label`` specifies whether the result is labeled with the beginning or
+the end of the interval.
+
+.. ipython:: python
+
+   ts.resample("5Min").mean()  # by default label='left'
+
+   ts.resample("5Min", label="left").mean()
+
+.. warning::
+
+    The default values for ``label`` and ``closed`` is '**left**' for all
+    frequency offsets except for 'ME', 'YE', 'QE', 'BME', 'BYE', 'BQE', and 'W'
+    which all have a default of 'right'.
+
+    This might unintendedly lead to looking ahead, where the value for a later
+    time is pulled back to a previous time as in the following example with
+    the :class:`~pandas.tseries.offsets.BusinessDay` frequency:
+
+    .. ipython:: python
+
+        s = pd.date_range("2000-01-01", "2000-01-05").to_series()
+        s.iloc[2] = pd.NaT
+        s.dt.day_name()
+
+        # default: label='left', closed='left'
+        s.resample("B").last().dt.day_name()
+
+    Notice how the value for Sunday got pulled back to the previous Friday.
+    To get the behavior where the value for Sunday is pushed to Monday, use
+    instead
+
+    .. ipython:: python
+
+        s.resample("B", label="right", closed="right").last().dt.day_name()
+
+The ``axis`` parameter can be set to 0 or 1 and allows you to resample the
+specified axis for a ``DataFrame``.
+
+``kind`` can be set to 'timestamp' or 'period' to convert the resulting index
+to/from timestamp and time span representations. By default ``resample``
+retains the input representation.
+
+``convention`` can be set to 'start' or 'end' when resampling period data
+(detail below). It specifies how low frequency periods are converted to higher
+frequency periods.
+
+
+Upsampling
+~~~~~~~~~~
+
+For upsampling, you can specify a way to upsample and the ``limit`` parameter to interpolate over the gaps that are created:
+
+.. ipython:: python
+
+   # from secondly to every 250 milliseconds
+
+   ts[:2].resample("250ms").asfreq()
+
+   ts[:2].resample("250ms").ffill()
+
+   ts[:2].resample("250ms").ffill(limit=2)
+
+Sparse resampling
+~~~~~~~~~~~~~~~~~
+
+Sparse timeseries are the ones where you have a lot fewer points relative
+to the amount of time you are looking to resample. Naively upsampling a sparse
+series can potentially generate lots of intermediate values. When you don't want
+to use a method to fill these values, e.g. ``fill_method`` is ``None``, then
+intermediate values will be filled with ``NaN``.
+
+Since ``resample`` is a time-based groupby, the following is a method to efficiently
+resample only the groups that are not all ``NaN``.
+
+.. ipython:: python
+
+    rng = pd.date_range("2014-1-1", periods=100, freq="D") + pd.Timedelta("1s")
+    ts = pd.Series(range(100), index=rng)
+
+If we want to resample to the full range of the series:
+
+.. ipython:: python
+
+    ts.resample("3min").sum()
+
+We can instead only resample those groups where we have points as follows:
+
+.. ipython:: python
+
+    from functools import partial
+    from pandas.tseries.frequencies import to_offset
+
+    def round(t, freq):
+        # round a Timestamp to a specified freq
+        freq = to_offset(freq)
+        td = pd.Timedelta(freq)
+        return pd.Timestamp((t.value // td.value) * td.value)
+
+    ts.groupby(partial(round, freq="3min")).sum()
+
+.. _timeseries.aggregate:
+
+Aggregation
+~~~~~~~~~~~
+
+The ``resample()`` method returns a ``pandas.api.typing.Resampler`` instance.  Similar to
+the :ref:`aggregating API <basics.aggregate>`, :ref:`groupby API <groupby.aggregate>`,
+and the :ref:`window API <window.overview>`, a ``Resampler`` can be selectively resampled.
+
+Resampling a ``DataFrame``, the default will be to act on all columns with the same function.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.random.randn(1000, 3),
+       index=pd.date_range("1/1/2012", freq="s", periods=1000),
+       columns=["A", "B", "C"],
+   )
+   r = df.resample("3min")
+   r.mean()
+
+We can select a specific column or columns using standard getitem.
+
+.. ipython:: python
+
+   r["A"].mean()
+
+   r[["A", "B"]].mean()
+
+You can pass a list or dict of functions to do aggregation with, outputting a ``DataFrame``:
+
+.. ipython:: python
+
+   r["A"].agg(["sum", "mean", "std"])
+
+On a resampled ``DataFrame``, you can pass a list of functions to apply to each
+column, which produces an aggregated result with a hierarchical index:
+
+.. ipython:: python
+
+   r.agg(["sum", "mean"])
+
+By passing a dict to ``aggregate`` you can apply a different aggregation to the
+columns of a ``DataFrame``:
+
+.. ipython:: python
+   :okexcept:
+
+   r.agg({"A": "sum", "B": lambda x: np.std(x, ddof=1)})
+
+The function names can also be strings. In order for a string to be valid it
+must be implemented on the resampled object:
+
+.. ipython:: python
+
+   r.agg({"A": "sum", "B": "std"})
+
+Furthermore, you can also specify multiple aggregation functions for each column separately.
+
+.. ipython:: python
+
+   r.agg({"A": ["sum", "std"], "B": ["mean", "std"]})
+
+
+If a ``DataFrame`` does not have a datetimelike index, but instead you want
+to resample based on datetimelike column in the frame, it can passed to the
+``on`` keyword.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"date": pd.date_range("2015-01-01", freq="W", periods=5), "a": np.arange(5)},
+       index=pd.MultiIndex.from_arrays(
+           [[1, 2, 3, 4, 5], pd.date_range("2015-01-01", freq="W", periods=5)],
+           names=["v", "d"],
+       ),
+   )
+   df
+   df.resample("MS", on="date")[["a"]].sum()
+
+Similarly, if you instead want to resample by a datetimelike
+level of ``MultiIndex``, its name or location can be passed to the
+``level`` keyword.
+
+.. ipython:: python
+
+   df.resample("MS", level="d")[["a"]].sum()
+
+.. _timeseries.iterating-label:
+
+Iterating through groups
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+With the ``Resampler`` object in hand, iterating through the grouped data is very
+natural and functions similarly to :py:func:`itertools.groupby`:
+
+.. ipython:: python
+
+   small = pd.Series(
+       range(6),
+       index=pd.to_datetime(
+           [
+               "2017-01-01T00:00:00",
+               "2017-01-01T00:30:00",
+               "2017-01-01T00:31:00",
+               "2017-01-01T01:00:00",
+               "2017-01-01T03:00:00",
+               "2017-01-01T03:05:00",
+           ]
+       ),
+   )
+   resampled = small.resample("h")
+
+   for name, group in resampled:
+       print("Group: ", name)
+       print("-" * 27)
+       print(group, end="\n\n")
+
+See :ref:`groupby.iterating-label` or :class:`Resampler.__iter__` for more.
+
+.. _timeseries.adjust-the-start-of-the-bins:
+
+Use ``origin`` or ``offset`` to adjust the start of the bins
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The bins of the grouping are adjusted based on the beginning of the day of the time series starting point. This works well with frequencies that are multiples of a day (like ``30D``) or that divide a day evenly (like ``90s`` or ``1min``). This can create inconsistencies with some frequencies that do not meet this criteria. To change this behavior you can specify a fixed Timestamp with the argument ``origin``.
+
+For example:
+
+.. ipython:: python
+
+    start, end = "2000-10-01 23:30:00", "2000-10-02 00:30:00"
+    middle = "2000-10-02 00:00:00"
+    rng = pd.date_range(start, end, freq="7min")
+    ts = pd.Series(np.arange(len(rng)) * 3, index=rng)
+    ts
+
+Here we can see that, when using ``origin`` with its default value (``'start_day'``), the result after ``'2000-10-02 00:00:00'`` are not identical depending on the start of time series:
+
+.. ipython:: python
+
+    ts.resample("17min", origin="start_day").sum()
+    ts[middle:end].resample("17min", origin="start_day").sum()
+
+
+Here we can see that, when setting ``origin`` to ``'epoch'``, the result after ``'2000-10-02 00:00:00'`` are identical depending on the start of time series:
+
+.. ipython:: python
+
+   ts.resample("17min", origin="epoch").sum()
+   ts[middle:end].resample("17min", origin="epoch").sum()
+
+
+If needed you can use a custom timestamp for ``origin``:
+
+.. ipython:: python
+
+   ts.resample("17min", origin="2001-01-01").sum()
+   ts[middle:end].resample("17min", origin=pd.Timestamp("2001-01-01")).sum()
+
+If needed you can just adjust the bins with an ``offset`` Timedelta that would be added to the default ``origin``.
+Those two examples are equivalent for this time series:
+
+.. ipython:: python
+
+    ts.resample("17min", origin="start").sum()
+    ts.resample("17min", offset="23h30min").sum()
+
+
+Note the use of ``'start'`` for ``origin`` on the last example. In that case, ``origin`` will be set to the first value of the timeseries.
+
+Backward resample
+~~~~~~~~~~~~~~~~~
+
+.. versionadded:: 1.3.0
+
+Instead of adjusting the beginning of bins, sometimes we need to fix the end of the bins to make a backward resample with a given ``freq``. The backward resample sets ``closed`` to ``'right'`` by default since the last value should be considered as the edge point for the last bin.
+
+We can set ``origin`` to ``'end'``. The value for a specific ``Timestamp`` index stands for the resample result from the current ``Timestamp`` minus ``freq`` to the current ``Timestamp`` with a right close.
+
+.. ipython:: python
+
+   ts.resample('17min', origin='end').sum()
+
+Besides, in contrast with the ``'start_day'`` option, ``end_day`` is supported. This will set the origin as the ceiling midnight of the largest ``Timestamp``.
+
+.. ipython:: python
+
+   ts.resample('17min', origin='end_day').sum()
+
+The above result uses ``2000-10-02 00:29:00`` as the last bin's right edge since the following computation.
+
+.. ipython:: python
+
+   ceil_mid = rng.max().ceil('D')
+   freq = pd.offsets.Minute(17)
+   bin_res = ceil_mid - freq * ((ceil_mid - rng.max()) // freq)
+   bin_res
+
+.. _timeseries.periods:
+
+Time span representation
+------------------------
+
+Regular intervals of time are represented by ``Period`` objects in pandas while
+sequences of ``Period`` objects are collected in a ``PeriodIndex``, which can
+be created with the convenience function ``period_range``.
+
+Period
+~~~~~~
+
+A ``Period`` represents a span of time (e.g., a day, a month, a quarter, etc).
+You can specify the span via ``freq`` keyword using a frequency alias like below.
+Because ``freq`` represents a span of ``Period``, it cannot be negative like "-3D".
+
+.. ipython:: python
+
+   pd.Period("2012", freq="Y-DEC")
+
+   pd.Period("2012-1-1", freq="D")
+
+   pd.Period("2012-1-1 19:00", freq="h")
+
+   pd.Period("2012-1-1 19:00", freq="5h")
+
+Adding and subtracting integers from periods shifts the period by its own
+frequency. Arithmetic is not allowed between ``Period`` with different ``freq`` (span).
+
+.. ipython:: python
+
+   p = pd.Period("2012", freq="Y-DEC")
+   p + 1
+   p - 3
+   p = pd.Period("2012-01", freq="2M")
+   p + 2
+   p - 1
+   p == pd.Period("2012-01", freq="3M")
+
+
+If ``Period`` freq is daily or higher (``D``, ``h``, ``min``, ``s``, ``ms``, ``us``, and ``ns``), ``offsets`` and ``timedelta``-like can be added if the result can have the same freq. Otherwise, ``ValueError`` will be raised.
+
+.. ipython:: python
+
+   p = pd.Period("2014-07-01 09:00", freq="h")
+   p + pd.offsets.Hour(2)
+   p + datetime.timedelta(minutes=120)
+   p + np.timedelta64(7200, "s")
+
+.. ipython:: python
+   :okexcept:
+
+   p + pd.offsets.Minute(5)
+
+
+If ``Period`` has other frequencies, only the same ``offsets`` can be added. Otherwise, ``ValueError`` will be raised.
+
+.. ipython:: python
+
+   p = pd.Period("2014-07", freq="M")
+   p + pd.offsets.MonthEnd(3)
+
+.. ipython:: python
+   :okexcept:
+
+   p + pd.offsets.MonthBegin(3)
+
+
+Taking the difference of ``Period`` instances with the same frequency will
+return the number of frequency units between them:
+
+.. ipython:: python
+
+   pd.Period("2012", freq="Y-DEC") - pd.Period("2002", freq="Y-DEC")
+
+PeriodIndex and period_range
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Regular sequences of ``Period`` objects can be collected in a ``PeriodIndex``,
+which can be constructed using the ``period_range`` convenience function:
+
+.. ipython:: python
+
+   prng = pd.period_range("1/1/2011", "1/1/2012", freq="M")
+   prng
+
+The ``PeriodIndex`` constructor can also be used directly:
+
+.. ipython:: python
+
+   pd.PeriodIndex(["2011-1", "2011-2", "2011-3"], freq="M")
+
+Passing multiplied frequency outputs a sequence of ``Period`` which
+has multiplied span.
+
+.. ipython:: python
+
+   pd.period_range(start="2014-01", freq="3M", periods=4)
+
+If ``start`` or ``end`` are ``Period`` objects, they will be used as anchor
+endpoints for a ``PeriodIndex`` with frequency matching that of the
+``PeriodIndex`` constructor.
+
+.. ipython:: python
+
+   pd.period_range(
+       start=pd.Period("2017Q1", freq="Q"), end=pd.Period("2017Q2", freq="Q"), freq="M"
+   )
+
+Just like ``DatetimeIndex``, a ``PeriodIndex`` can also be used to index pandas
+objects:
+
+.. ipython:: python
+
+   ps = pd.Series(np.random.randn(len(prng)), prng)
+   ps
+
+``PeriodIndex`` supports addition and subtraction with the same rule as ``Period``.
+
+.. ipython:: python
+
+   idx = pd.period_range("2014-07-01 09:00", periods=5, freq="h")
+   idx
+   idx + pd.offsets.Hour(2)
+
+   idx = pd.period_range("2014-07", periods=5, freq="M")
+   idx
+   idx + pd.offsets.MonthEnd(3)
+
+``PeriodIndex`` has its own dtype named ``period``, refer to :ref:`Period Dtypes <timeseries.period_dtype>`.
+
+.. _timeseries.period_dtype:
+
+Period dtypes
+~~~~~~~~~~~~~
+
+``PeriodIndex`` has a custom ``period`` dtype. This is a pandas extension
+dtype similar to the :ref:`timezone aware dtype <timeseries.timezone_series>` (``datetime64[ns, tz]``).
+
+The ``period`` dtype holds the ``freq`` attribute and is represented with
+``period[freq]`` like ``period[D]`` or ``period[M]``, using :ref:`frequency strings <timeseries.period_aliases>`.
+
+.. ipython:: python
+
+   pi = pd.period_range("2016-01-01", periods=3, freq="M")
+   pi
+   pi.dtype
+
+The ``period`` dtype can be used in ``.astype(...)``. It allows one to change the
+``freq`` of a ``PeriodIndex`` like ``.asfreq()`` and convert a
+``DatetimeIndex`` to ``PeriodIndex`` like ``to_period()``:
+
+.. ipython:: python
+
+   # change monthly freq to daily freq
+   pi.astype("period[D]")
+
+   # convert to DatetimeIndex
+   pi.astype("datetime64[ns]")
+
+   # convert to PeriodIndex
+   dti = pd.date_range("2011-01-01", freq="ME", periods=3)
+   dti
+   dti.astype("period[M]")
+
+PeriodIndex partial string indexing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+PeriodIndex now supports partial string slicing with non-monotonic indexes.
+
+You can pass in dates and strings to ``Series`` and ``DataFrame`` with ``PeriodIndex``, in the same manner as ``DatetimeIndex``. For details, refer to :ref:`DatetimeIndex Partial String Indexing <timeseries.partialindexing>`.
+
+.. ipython:: python
+
+   ps["2011-01"]
+
+   ps[datetime.datetime(2011, 12, 25):]
+
+   ps["10/31/2011":"12/31/2011"]
+
+Passing a string representing a lower frequency than ``PeriodIndex`` returns partial sliced data.
+
+.. ipython:: python
+
+   ps["2011"]
+
+   dfp = pd.DataFrame(
+       np.random.randn(600, 1),
+       columns=["A"],
+       index=pd.period_range("2013-01-01 9:00", periods=600, freq="min"),
+   )
+   dfp
+   dfp.loc["2013-01-01 10h"]
+
+As with ``DatetimeIndex``, the endpoints will be included in the result. The example below slices data starting from 10:00 to 11:59.
+
+.. ipython:: python
+
+   dfp["2013-01-01 10h":"2013-01-01 11h"]
+
+
+Frequency conversion and resampling with PeriodIndex
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+The frequency of ``Period`` and ``PeriodIndex`` can be converted via the ``asfreq``
+method. Let's start with the fiscal year 2011, ending in December:
+
+.. ipython:: python
+
+   p = pd.Period("2011", freq="Y-DEC")
+   p
+
+We can convert it to a monthly frequency. Using the ``how`` parameter, we can
+specify whether to return the starting or ending month:
+
+.. ipython:: python
+
+   p.asfreq("M", how="start")
+
+   p.asfreq("M", how="end")
+
+The shorthands 's' and 'e' are provided for convenience:
+
+.. ipython:: python
+
+   p.asfreq("M", "s")
+   p.asfreq("M", "e")
+
+Converting to a "super-period" (e.g., annual frequency is a super-period of
+quarterly frequency) automatically returns the super-period that includes the
+input period:
+
+.. ipython:: python
+
+   p = pd.Period("2011-12", freq="M")
+
+   p.asfreq("Y-NOV")
+
+Note that since we converted to an annual frequency that ends the year in
+November, the monthly period of December 2011 is actually in the 2012 Y-NOV
+period.
+
+.. _timeseries.quarterly:
+
+Period conversions with anchored frequencies are particularly useful for
+working with various quarterly data common to economics, business, and other
+fields. Many organizations define quarters relative to the month in which their
+fiscal year starts and ends. Thus, first quarter of 2011 could start in 2010 or
+a few months into 2011. Via anchored frequencies, pandas works for all quarterly
+frequencies ``Q-JAN`` through ``Q-DEC``.
+
+``Q-DEC`` define regular calendar quarters:
+
+.. ipython:: python
+
+   p = pd.Period("2012Q1", freq="Q-DEC")
+
+   p.asfreq("D", "s")
+
+   p.asfreq("D", "e")
+
+``Q-MAR`` defines fiscal year end in March:
+
+.. ipython:: python
+
+   p = pd.Period("2011Q4", freq="Q-MAR")
+
+   p.asfreq("D", "s")
+
+   p.asfreq("D", "e")
+
+.. _timeseries.interchange:
+
+Converting between representations
+----------------------------------
+
+Timestamped data can be converted to PeriodIndex-ed data using ``to_period``
+and vice-versa using ``to_timestamp``:
+
+.. ipython:: python
+
+   rng = pd.date_range("1/1/2012", periods=5, freq="ME")
+
+   ts = pd.Series(np.random.randn(len(rng)), index=rng)
+
+   ts
+
+   ps = ts.to_period()
+
+   ps
+
+   ps.to_timestamp()
+
+Remember that 's' and 'e' can be used to return the timestamps at the start or
+end of the period:
+
+.. ipython:: python
+
+   ps.to_timestamp("D", how="s")
+
+Converting between period and timestamp enables some convenient arithmetic
+functions to be used. In the following example, we convert a quarterly
+frequency with year ending in November to 9am of the end of the month following
+the quarter end:
+
+.. ipython:: python
+
+   prng = pd.period_range("1990Q1", "2000Q4", freq="Q-NOV")
+
+   ts = pd.Series(np.random.randn(len(prng)), prng)
+
+   ts.index = (prng.asfreq("M", "e") + 1).asfreq("h", "s") + 9
+
+   ts.head()
+
+.. _timeseries.oob:
+
+Representing out-of-bounds spans
+--------------------------------
+
+If you have data that is outside of the ``Timestamp`` bounds, see :ref:`Timestamp limitations <timeseries.timestamp-limits>`,
+then you can use a ``PeriodIndex`` and/or ``Series`` of ``Periods`` to do computations.
+
+.. ipython:: python
+
+   span = pd.period_range("1215-01-01", "1381-01-01", freq="D")
+   span
+
+To convert from an ``int64`` based YYYYMMDD representation.
+
+.. ipython:: python
+
+   s = pd.Series([20121231, 20141130, 99991231])
+   s
+
+   def conv(x):
+       return pd.Period(year=x // 10000, month=x // 100 % 100, day=x % 100, freq="D")
+
+   s.apply(conv)
+   s.apply(conv)[2]
+
+These can easily be converted to a ``PeriodIndex``:
+
+.. ipython:: python
+
+   span = pd.PeriodIndex(s.apply(conv))
+   span
+
+.. _timeseries.timezone:
+
+Time zone handling
+------------------
+
+pandas provides rich support for working with timestamps in different time
+zones using the ``zoneinfo``, ``pytz`` and ``dateutil`` libraries or :class:`datetime.timezone`
+objects from the standard library.
+
+
+Working with time zones
+~~~~~~~~~~~~~~~~~~~~~~~
+
+By default, pandas objects are time zone unaware:
+
+.. ipython:: python
+
+   rng = pd.date_range("3/6/2012 00:00", periods=15, freq="D")
+   rng.tz is None
+
+To localize these dates to a time zone (assign a particular time zone to a naive date),
+you can use the ``tz_localize`` method or the ``tz`` keyword argument in
+:func:`date_range`, :class:`Timestamp`, or :class:`DatetimeIndex`.
+You can either pass ``zoneinfo``, ``pytz`` or ``dateutil`` time zone objects or Olson time zone database strings.
+Olson time zone strings will return ``pytz`` time zone objects by default.
+To return ``dateutil`` time zone objects, append ``dateutil/`` before the string.
+
+* For ``zoneinfo``, a list of available timezones are available from :py:func:`zoneinfo.available_timezones`.
+* In ``pytz`` you can find a list of common (and less common) time zones using ``pytz.all_timezones``.
+* ``dateutil`` uses the OS time zones so there isn't a fixed list available. For
+  common zones, the names are the same as ``pytz`` and ``zoneinfo``.
+
+.. ipython:: python
+
+   import dateutil
+
+   # pytz
+   rng_pytz = pd.date_range("3/6/2012 00:00", periods=3, freq="D", tz="Europe/London")
+   rng_pytz.tz
+
+   # dateutil
+   rng_dateutil = pd.date_range("3/6/2012 00:00", periods=3, freq="D")
+   rng_dateutil = rng_dateutil.tz_localize("dateutil/Europe/London")
+   rng_dateutil.tz
+
+   # dateutil - utc special case
+   rng_utc = pd.date_range(
+       "3/6/2012 00:00",
+       periods=3,
+       freq="D",
+       tz=dateutil.tz.tzutc(),
+   )
+   rng_utc.tz
+
+.. ipython:: python
+
+   # datetime.timezone
+   rng_utc = pd.date_range(
+       "3/6/2012 00:00",
+       periods=3,
+       freq="D",
+       tz=datetime.timezone.utc,
+   )
+   rng_utc.tz
+
+Note that the ``UTC`` time zone is a special case in ``dateutil`` and should be constructed explicitly
+as an instance of ``dateutil.tz.tzutc``. You can also construct other time
+zones objects explicitly first.
+
+.. ipython:: python
+
+   import pytz
+
+   # pytz
+   tz_pytz = pytz.timezone("Europe/London")
+   rng_pytz = pd.date_range("3/6/2012 00:00", periods=3, freq="D")
+   rng_pytz = rng_pytz.tz_localize(tz_pytz)
+   rng_pytz.tz == tz_pytz
+
+   # dateutil
+   tz_dateutil = dateutil.tz.gettz("Europe/London")
+   rng_dateutil = pd.date_range("3/6/2012 00:00", periods=3, freq="D", tz=tz_dateutil)
+   rng_dateutil.tz == tz_dateutil
+
+To convert a time zone aware pandas object from one time zone to another,
+you can use the ``tz_convert`` method.
+
+.. ipython:: python
+
+   rng_pytz.tz_convert("US/Eastern")
+
+.. note::
+
+    When using ``pytz`` time zones, :class:`DatetimeIndex` will construct a different
+    time zone object than a :class:`Timestamp` for the same time zone input. A :class:`DatetimeIndex`
+    can hold a collection of :class:`Timestamp` objects that may have different UTC offsets and cannot be
+    succinctly represented by one ``pytz`` time zone instance while one :class:`Timestamp`
+    represents one point in time with a specific UTC offset.
+
+    .. ipython:: python
+
+       dti = pd.date_range("2019-01-01", periods=3, freq="D", tz="US/Pacific")
+       dti.tz
+       ts = pd.Timestamp("2019-01-01", tz="US/Pacific")
+       ts.tz
+
+.. warning::
+
+        Be wary of conversions between libraries. For some time zones, ``pytz`` and ``dateutil`` have different
+        definitions of the zone. This is more of a problem for unusual time zones than for
+        'standard' zones like ``US/Eastern``.
+
+.. warning::
+
+    Be aware that a time zone definition across versions of time zone libraries may not
+    be considered equal.  This may cause problems when working with stored data that
+    is localized using one version and operated on with a different version.
+    See :ref:`here<io.hdf5-notes>` for how to handle such a situation.
+
+.. warning::
+
+    For ``pytz`` time zones, it is incorrect to pass a time zone object directly into
+    the ``datetime.datetime`` constructor
+    (e.g., ``datetime.datetime(2011, 1, 1, tzinfo=pytz.timezone('US/Eastern'))``.
+    Instead, the datetime needs to be localized using the ``localize`` method
+    on the ``pytz`` time zone object.
+
+.. warning::
+
+    Be aware that for times in the future, correct conversion between time zones
+    (and UTC) cannot be guaranteed by any time zone library because a timezone's
+    offset from UTC may be changed by the respective government.
+
+.. warning::
+
+    If you are using dates beyond 2038-01-18 with ``pytz``, due to current deficiencies
+    in the underlying libraries caused by the year 2038 problem, daylight saving time (DST) adjustments
+    to timezone aware dates will not be applied. If and when the underlying libraries are fixed,
+    the DST transitions will be applied.
+
+    For example, for two dates that are in British Summer Time (and so would normally be GMT+1), both the following asserts evaluate as true:
+
+    .. ipython:: python
+
+      import pytz
+
+       d_2037 = "2037-03-31T010101"
+       d_2038 = "2038-03-31T010101"
+       DST = pytz.timezone("Europe/London")
+       assert pd.Timestamp(d_2037, tz=DST) != pd.Timestamp(d_2037, tz="GMT")
+       assert pd.Timestamp(d_2038, tz=DST) == pd.Timestamp(d_2038, tz="GMT")
+
+Under the hood, all timestamps are stored in UTC. Values from a time zone aware
+:class:`DatetimeIndex` or :class:`Timestamp` will have their fields (day, hour, minute, etc.)
+localized to the time zone. However, timestamps with the same UTC value are
+still considered to be equal even if they are in different time zones:
+
+.. ipython:: python
+
+   rng_eastern = rng_utc.tz_convert("US/Eastern")
+   rng_berlin = rng_utc.tz_convert("Europe/Berlin")
+
+   rng_eastern[2]
+   rng_berlin[2]
+   rng_eastern[2] == rng_berlin[2]
+
+Operations between :class:`Series` in different time zones will yield UTC
+:class:`Series`, aligning the data on the UTC timestamps:
+
+.. ipython:: python
+
+   ts_utc = pd.Series(range(3), pd.date_range("20130101", periods=3, tz="UTC"))
+   eastern = ts_utc.tz_convert("US/Eastern")
+   berlin = ts_utc.tz_convert("Europe/Berlin")
+   result = eastern + berlin
+   result
+   result.index
+
+To remove time zone information, use ``tz_localize(None)`` or ``tz_convert(None)``.
+``tz_localize(None)`` will remove the time zone yielding the local time representation.
+``tz_convert(None)`` will remove the time zone after converting to UTC time.
+
+.. ipython:: python
+
+   didx = pd.date_range(start="2014-08-01 09:00", freq="h", periods=3, tz="US/Eastern")
+   didx
+   didx.tz_localize(None)
+   didx.tz_convert(None)
+
+   # tz_convert(None) is identical to tz_convert('UTC').tz_localize(None)
+   didx.tz_convert("UTC").tz_localize(None)
+
+.. _timeseries.fold:
+
+Fold
+~~~~
+
+For ambiguous times, pandas supports explicitly specifying the keyword-only fold argument.
+Due to daylight saving time, one wall clock time can occur twice when shifting
+from summer to winter time; fold describes whether the datetime-like corresponds
+to the first (0) or the second time (1) the wall clock hits the ambiguous time.
+Fold is supported only for constructing from naive ``datetime.datetime``
+(see `datetime documentation <https://docs.python.org/3/library/datetime.html>`__ for details) or from :class:`Timestamp`
+or for constructing from components (see below). Only ``dateutil`` timezones are supported
+(see `dateutil documentation <https://dateutil.readthedocs.io/en/stable/tz.html#dateutil.tz.enfold>`__
+for ``dateutil`` methods that deal with ambiguous datetimes) as ``pytz``
+timezones do not support fold (see `pytz documentation <http://pytz.sourceforge.net/index.html>`__
+for details on how ``pytz`` deals with ambiguous datetimes). To localize an ambiguous datetime
+with ``pytz``, please use :meth:`Timestamp.tz_localize`. In general, we recommend to rely
+on :meth:`Timestamp.tz_localize` when localizing ambiguous datetimes if you need direct
+control over how they are handled.
+
+.. ipython:: python
+
+   pd.Timestamp(
+       datetime.datetime(2019, 10, 27, 1, 30, 0, 0),
+       tz="dateutil/Europe/London",
+       fold=0,
+   )
+   pd.Timestamp(
+       year=2019,
+       month=10,
+       day=27,
+       hour=1,
+       minute=30,
+       tz="dateutil/Europe/London",
+       fold=1,
+   )
+
+.. _timeseries.timezone_ambiguous:
+
+Ambiguous times when localizing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``tz_localize`` may not be able to determine the UTC offset of a timestamp
+because daylight savings time (DST) in a local time zone causes some times to occur
+twice within one day ("clocks fall back"). The following options are available:
+
+* ``'raise'``: Raises a ``ValueError`` (the default behavior)
+* ``'infer'``: Attempt to determine the correct offset based on the monotonicity of the timestamps
+* ``'NaT'``: Replaces ambiguous times with ``NaT``
+* ``bool``: ``True`` represents a DST time, ``False`` represents non-DST time. An array-like of ``bool`` values is supported for a sequence of times.
+
+.. ipython:: python
+
+   rng_hourly = pd.DatetimeIndex(
+       ["11/06/2011 00:00", "11/06/2011 01:00", "11/06/2011 01:00", "11/06/2011 02:00"]
+   )
+
+This will fail as there are ambiguous times (``'11/06/2011 01:00'``)
+
+.. ipython:: python
+   :okexcept:
+
+   rng_hourly.tz_localize('US/Eastern')
+
+Handle these ambiguous times by specifying the following.
+
+.. ipython:: python
+
+   rng_hourly.tz_localize("US/Eastern", ambiguous="infer")
+   rng_hourly.tz_localize("US/Eastern", ambiguous="NaT")
+   rng_hourly.tz_localize("US/Eastern", ambiguous=[True, True, False, False])
+
+.. _timeseries.timezone_nonexistent:
+
+Nonexistent times when localizing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A DST transition may also shift the local time ahead by 1 hour creating nonexistent
+local times ("clocks spring forward"). The behavior of localizing a timeseries with nonexistent times
+can be controlled by the ``nonexistent`` argument. The following options are available:
+
+* ``'raise'``: Raises a ``ValueError`` (the default behavior)
+* ``'NaT'``: Replaces nonexistent times with ``NaT``
+* ``'shift_forward'``: Shifts nonexistent times forward to the closest real time
+* ``'shift_backward'``: Shifts nonexistent times backward to the closest real time
+* timedelta object: Shifts nonexistent times by the timedelta duration
+
+.. ipython:: python
+
+    dti = pd.date_range(start="2015-03-29 02:30:00", periods=3, freq="h")
+    # 2:30 is a nonexistent time
+
+Localization of nonexistent times will raise an error by default.
+
+.. ipython:: python
+   :okexcept:
+
+   dti.tz_localize('Europe/Warsaw')
+
+Transform nonexistent times to ``NaT`` or shift the times.
+
+.. ipython:: python
+
+    dti
+    dti.tz_localize("Europe/Warsaw", nonexistent="shift_forward")
+    dti.tz_localize("Europe/Warsaw", nonexistent="shift_backward")
+    dti.tz_localize("Europe/Warsaw", nonexistent=pd.Timedelta(1, unit="h"))
+    dti.tz_localize("Europe/Warsaw", nonexistent="NaT")
+
+
+.. _timeseries.timezone_series:
+
+Time zone Series operations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A :class:`Series` with time zone **naive** values is
+represented with a dtype of ``datetime64[ns]``.
+
+.. ipython:: python
+
+   s_naive = pd.Series(pd.date_range("20130101", periods=3))
+   s_naive
+
+A :class:`Series` with a time zone **aware** values is
+represented with a dtype of ``datetime64[ns, tz]`` where ``tz`` is the time zone
+
+.. ipython:: python
+
+   s_aware = pd.Series(pd.date_range("20130101", periods=3, tz="US/Eastern"))
+   s_aware
+
+Both of these :class:`Series` time zone information
+can be manipulated via the ``.dt`` accessor, see :ref:`the dt accessor section <basics.dt_accessors>`.
+
+For example, to localize and convert a naive stamp to time zone aware.
+
+.. ipython:: python
+
+   s_naive.dt.tz_localize("UTC").dt.tz_convert("US/Eastern")
+
+Time zone information can also be manipulated using the ``astype`` method.
+This method can convert between different timezone-aware dtypes.
+
+.. ipython:: python
+
+   # convert to a new time zone
+   s_aware.astype("datetime64[ns, CET]")
+
+.. note::
+
+   Using :meth:`Series.to_numpy` on a ``Series``, returns a NumPy array of the data.
+   NumPy does not currently support time zones (even though it is *printing* in the local time zone!),
+   therefore an object array of Timestamps is returned for time zone aware data:
+
+   .. ipython:: python
+
+      s_naive.to_numpy()
+      s_aware.to_numpy()
+
+   By converting to an object array of Timestamps, it preserves the time zone
+   information. For example, when converting back to a Series:
+
+   .. ipython:: python
+
+      pd.Series(s_aware.to_numpy())
+
+   However, if you want an actual NumPy ``datetime64[ns]`` array (with the values
+   converted to UTC) instead of an array of objects, you can specify the
+   ``dtype`` argument:
+
+   .. ipython:: python
+
+      s_aware.to_numpy(dtype="datetime64[ns]")
diff --git a/pandas/doc/source/user_guide/visualization.rst b/pandas/doc/source/user_guide/visualization.rst
new file mode 100644
index 0000000000000..4b5cdca23103c
--- /dev/null
+++ b/pandas/doc/source/user_guide/visualization.rst
@@ -0,0 +1,1808 @@
+.. _visualization:
+
+{{ header }}
+
+*******************
+Chart visualization
+*******************
+
+
+.. note::
+
+   The examples below assume that you're using `Jupyter <https://jupyter.org/>`_.
+
+This section demonstrates visualization through charting. For information on
+visualization of tabular data please see the section on `Table Visualization <style.ipynb>`_.
+
+We use the standard convention for referencing the matplotlib API:
+
+.. ipython:: python
+
+   import matplotlib.pyplot as plt
+
+   plt.close("all")
+
+We provide the basics in pandas to easily create decent looking plots.
+See `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_ for visualization
+libraries that go beyond the basics documented here.
+
+.. note::
+
+   All calls to ``np.random`` are seeded with 123456.
+
+.. _visualization.basic:
+
+Basic plotting: ``plot``
+------------------------
+
+We will demonstrate the basics, see the :ref:`cookbook<cookbook.plotting>` for
+some advanced strategies.
+
+The ``plot`` method on Series and DataFrame is just a simple wrapper around
+:meth:`plt.plot() <matplotlib.axes.Axes.plot>`:
+
+.. ipython:: python
+
+   np.random.seed(123456)
+
+   ts = pd.Series(np.random.randn(1000), index=pd.date_range("1/1/2000", periods=1000))
+   ts = ts.cumsum()
+
+   @savefig series_plot_basic.png
+   ts.plot();
+
+If the index consists of dates, it calls :meth:`gcf().autofmt_xdate() <matplotlib.figure.Figure.autofmt_xdate>`
+to try to format the x-axis nicely as per above.
+
+On DataFrame, :meth:`~DataFrame.plot` is a convenience to plot all of the columns with labels:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list("ABCD"))
+   df = df.cumsum()
+
+   plt.figure();
+   @savefig frame_plot_basic.png
+   df.plot();
+
+You can plot one column versus another using the ``x`` and ``y`` keywords in
+:meth:`~DataFrame.plot`:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df3 = pd.DataFrame(np.random.randn(1000, 2), columns=["B", "C"]).cumsum()
+   df3["A"] = pd.Series(list(range(len(df))))
+
+   @savefig df_plot_xy.png
+   df3.plot(x="A", y="B");
+
+.. note::
+
+   For more formatting and styling options, see
+   :ref:`formatting <visualization.formatting>` below.
+
+.. ipython:: python
+    :suppress:
+
+    plt.close("all")
+
+.. _visualization.other:
+
+Other plots
+-----------
+
+Plotting methods allow for a handful of plot styles other than the
+default line plot. These methods can be provided as the ``kind``
+keyword argument to :meth:`~DataFrame.plot`, and include:
+
+* :ref:`'bar' <visualization.barplot>` or :ref:`'barh' <visualization.barplot>` for bar plots
+* :ref:`'hist' <visualization.hist>` for histogram
+* :ref:`'box' <visualization.box>` for boxplot
+* :ref:`'kde' <visualization.kde>` or :ref:`'density' <visualization.kde>` for density plots
+* :ref:`'area' <visualization.area_plot>` for area plots
+* :ref:`'scatter' <visualization.scatter>` for scatter plots
+* :ref:`'hexbin' <visualization.hexbin>` for hexagonal bin plots
+* :ref:`'pie' <visualization.pie>` for pie plots
+
+For example, a bar plot can be created the following way:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig bar_plot_ex.png
+   df.iloc[5].plot(kind="bar");
+
+You can also create these other plots using the methods ``DataFrame.plot.<kind>`` instead of providing the ``kind`` keyword argument. This makes it easier to discover plot methods and the specific arguments they use:
+
+.. ipython::
+    :verbatim:
+
+    In [14]: df = pd.DataFrame()
+
+    In [15]: df.plot.<TAB>  # noqa: E225, E999
+    df.plot.area     df.plot.barh     df.plot.density  df.plot.hist     df.plot.line     df.plot.scatter
+    df.plot.bar      df.plot.box      df.plot.hexbin   df.plot.kde      df.plot.pie
+
+In addition to these ``kind`` s, there are the :ref:`DataFrame.hist() <visualization.hist>`,
+and :ref:`DataFrame.boxplot() <visualization.box>` methods, which use a separate interface.
+
+Finally, there are several :ref:`plotting functions <visualization.tools>` in ``pandas.plotting``
+that take a :class:`Series` or :class:`DataFrame` as an argument. These
+include:
+
+* :ref:`Scatter Matrix <visualization.scatter_matrix>`
+* :ref:`Andrews Curves <visualization.andrews_curves>`
+* :ref:`Parallel Coordinates <visualization.parallel_coordinates>`
+* :ref:`Lag Plot <visualization.lag>`
+* :ref:`Autocorrelation Plot <visualization.autocorrelation>`
+* :ref:`Bootstrap Plot <visualization.bootstrap>`
+* :ref:`RadViz <visualization.radviz>`
+
+Plots may also be adorned with :ref:`errorbars <visualization.errorbars>`
+or :ref:`tables <visualization.table>`.
+
+.. _visualization.barplot:
+
+Bar plots
+~~~~~~~~~
+
+For labeled, non-time series data, you may wish to produce a bar plot:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig bar_plot_ex.png
+   df.iloc[5].plot.bar();
+   plt.axhline(0, color="k");
+
+Calling a DataFrame's :meth:`plot.bar() <DataFrame.plot.bar>` method produces a multiple
+bar plot:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df2 = pd.DataFrame(np.random.rand(10, 4), columns=["a", "b", "c", "d"])
+
+   @savefig bar_plot_multi_ex.png
+   df2.plot.bar();
+
+To produce a stacked bar plot, pass ``stacked=True``:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+
+.. ipython:: python
+
+   @savefig bar_plot_stacked_ex.png
+   df2.plot.bar(stacked=True);
+
+To get horizontal bar plots, use the ``barh`` method:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+
+.. ipython:: python
+
+   @savefig barh_plot_stacked_ex.png
+   df2.plot.barh(stacked=True);
+
+.. _visualization.hist:
+
+Histograms
+~~~~~~~~~~
+
+Histograms can be drawn by using the :meth:`DataFrame.plot.hist` and :meth:`Series.plot.hist` methods.
+
+.. ipython:: python
+
+   df4 = pd.DataFrame(
+       {
+           "a": np.random.randn(1000) + 1,
+           "b": np.random.randn(1000),
+           "c": np.random.randn(1000) - 1,
+       },
+       columns=["a", "b", "c"],
+   )
+
+   plt.figure();
+
+   @savefig hist_new.png
+   df4.plot.hist(alpha=0.5);
+
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+A histogram can be stacked using ``stacked=True``. Bin size can be changed
+using the ``bins`` keyword.
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig hist_new_stacked.png
+   df4.plot.hist(stacked=True, bins=20);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+You can pass other keywords supported by matplotlib ``hist``. For example,
+horizontal and cumulative histograms can be drawn by
+``orientation='horizontal'`` and ``cumulative=True``.
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig hist_new_kwargs.png
+   df4["a"].plot.hist(orientation="horizontal", cumulative=True);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+See the :meth:`hist <matplotlib.axes.Axes.hist>` method and the
+`matplotlib hist documentation <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.hist.html>`__ for more.
+
+
+The existing interface ``DataFrame.hist`` to plot histogram still can be used.
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig hist_plot_ex.png
+   df["A"].diff().hist();
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+:meth:`DataFrame.hist` plots the histograms of the columns on multiple
+subplots:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig frame_hist_ex.png
+   df.diff().hist(color="k", alpha=0.5, bins=50);
+
+
+The ``by`` keyword can be specified to plot grouped histograms:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   data = pd.Series(np.random.randn(1000))
+
+   @savefig grouped_hist.png
+   data.hist(by=np.random.randint(0, 4, 1000), figsize=(6, 4));
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+In addition, the ``by`` keyword can also be specified in :meth:`DataFrame.plot.hist`.
+
+.. versionchanged:: 1.4.0
+
+.. ipython:: python
+
+   data = pd.DataFrame(
+       {
+           "a": np.random.choice(["x", "y", "z"], 1000),
+           "b": np.random.choice(["e", "f", "g"], 1000),
+           "c": np.random.randn(1000),
+           "d": np.random.randn(1000) - 1,
+       },
+   )
+
+   @savefig grouped_hist_by.png
+   data.plot.hist(by=["a", "b"], figsize=(10, 5));
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.box:
+
+Box plots
+~~~~~~~~~
+
+Boxplot can be drawn calling :meth:`Series.plot.box` and :meth:`DataFrame.plot.box`,
+or :meth:`DataFrame.boxplot` to visualize the distribution of values within each column.
+
+For instance, here is a boxplot representing five trials of 10 observations of
+a uniform random variable on [0,1).
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.rand(10, 5), columns=["A", "B", "C", "D", "E"])
+
+   @savefig box_plot_new.png
+   df.plot.box();
+
+Boxplot can be colorized by passing ``color`` keyword. You can pass a ``dict``
+whose keys are ``boxes``, ``whiskers``, ``medians`` and ``caps``.
+If some keys are missing in the ``dict``, default colors are used
+for the corresponding artists. Also, boxplot has ``sym`` keyword to specify fliers style.
+
+When you pass other type of arguments via ``color`` keyword, it will be directly
+passed to matplotlib for all the ``boxes``, ``whiskers``, ``medians`` and ``caps``
+colorization.
+
+The colors are applied to every boxes to be drawn. If you want
+more complicated colorization, you can get each drawn artists by passing
+:ref:`return_type <visualization.box.return>`.
+
+.. ipython:: python
+
+   color = {
+       "boxes": "DarkGreen",
+       "whiskers": "DarkOrange",
+       "medians": "DarkBlue",
+       "caps": "Gray",
+   }
+
+   @savefig box_new_colorize.png
+   df.plot.box(color=color, sym="r+");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Also, you can pass other keywords supported by matplotlib ``boxplot``.
+For example, horizontal and custom-positioned boxplot can be drawn by
+``vert=False`` and ``positions`` keywords.
+
+.. ipython:: python
+
+   @savefig box_new_kwargs.png
+   df.plot.box(vert=False, positions=[1, 4, 5, 6, 8]);
+
+
+See the :meth:`boxplot <matplotlib.axes.Axes.boxplot>` method and the
+`matplotlib boxplot documentation <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.boxplot.html>`__ for more.
+
+
+The existing interface ``DataFrame.boxplot`` to plot boxplot still can be used.
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+   :okwarning:
+
+   df = pd.DataFrame(np.random.rand(10, 5))
+   plt.figure();
+
+   @savefig box_plot_ex.png
+   bp = df.boxplot()
+
+You can create a stratified boxplot using the ``by`` keyword argument to create
+groupings.  For instance,
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+   :okwarning:
+
+   df = pd.DataFrame(np.random.rand(10, 2), columns=["Col1", "Col2"])
+   df["X"] = pd.Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+
+   plt.figure();
+
+   @savefig box_plot_ex2.png
+   bp = df.boxplot(by="X")
+
+You can also pass a subset of columns to plot, as well as group by multiple
+columns:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+   :okwarning:
+
+   df = pd.DataFrame(np.random.rand(10, 3), columns=["Col1", "Col2", "Col3"])
+   df["X"] = pd.Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+   df["Y"] = pd.Series(["A", "B", "A", "B", "A", "B", "A", "B", "A", "B"])
+
+   plt.figure();
+
+   @savefig box_plot_ex3.png
+   bp = df.boxplot(column=["Col1", "Col2"], by=["X", "Y"])
+
+.. ipython:: python
+   :suppress:
+
+    plt.close("all")
+
+You could also create groupings with :meth:`DataFrame.plot.box`, for instance:
+
+.. versionchanged:: 1.4.0
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   np.random.seed(123456)
+
+.. ipython:: python
+   :okwarning:
+
+   df = pd.DataFrame(np.random.rand(10, 3), columns=["Col1", "Col2", "Col3"])
+   df["X"] = pd.Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+
+   plt.figure();
+
+   @savefig box_plot_ex4.png
+   bp = df.plot.box(column=["Col1", "Col2"], by="X")
+
+.. ipython:: python
+   :suppress:
+
+    plt.close("all")
+
+.. _visualization.box.return:
+
+In ``boxplot``, the return type can be controlled by the ``return_type``, keyword. The valid choices are ``{"axes", "dict", "both", None}``.
+Faceting, created by ``DataFrame.boxplot`` with the ``by``
+keyword, will affect the output type as well:
+
+================ ======= ==========================
+``return_type``  Faceted Output type
+================ ======= ==========================
+``None``         No      axes
+``None``         Yes     2-D ndarray of axes
+``'axes'``       No      axes
+``'axes'``       Yes     Series of axes
+``'dict'``       No      dict of artists
+``'dict'``       Yes     Series of dicts of artists
+``'both'``       No      namedtuple
+``'both'``       Yes     Series of namedtuples
+================ ======= ==========================
+
+``Groupby.boxplot`` always returns a ``Series`` of ``return_type``.
+
+.. ipython:: python
+   :okwarning:
+
+   np.random.seed(1234)
+   df_box = pd.DataFrame(np.random.randn(50, 2))
+   df_box["g"] = np.random.choice(["A", "B"], size=50)
+   df_box.loc[df_box["g"] == "B", 1] += 3
+
+   @savefig boxplot_groupby.png
+   bp = df_box.boxplot(by="g")
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+The subplots above are split by the numeric columns first, then the value of
+the ``g`` column. Below the subplots are first split by the value of ``g``,
+then by the numeric columns.
+
+.. ipython:: python
+   :okwarning:
+
+   @savefig groupby_boxplot_vis.png
+   bp = df_box.groupby("g").boxplot()
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.area_plot:
+
+Area plot
+~~~~~~~~~
+
+You can create area plots with :meth:`Series.plot.area` and :meth:`DataFrame.plot.area`.
+Area plots are stacked by default. To produce stacked area plot, each column must be either all positive or all negative values.
+
+When input data contains ``NaN``, it will be automatically filled by 0. If you want to drop or fill by different values, use :func:`dataframe.dropna` or :func:`dataframe.fillna` before calling ``plot``.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+   plt.figure()
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.rand(10, 4), columns=["a", "b", "c", "d"])
+
+   @savefig area_plot_stacked.png
+   df.plot.area();
+
+To produce an unstacked plot, pass ``stacked=False``. Alpha value is set to 0.5 unless otherwise specified:
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+
+.. ipython:: python
+
+   @savefig area_plot_unstacked.png
+   df.plot.area(stacked=False);
+
+.. _visualization.scatter:
+
+Scatter plot
+~~~~~~~~~~~~
+
+Scatter plot can be drawn by using the :meth:`DataFrame.plot.scatter` method.
+Scatter plot requires numeric columns for the x and y axes.
+These can be specified by the ``x`` and ``y`` keywords.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+   plt.close("all")
+   plt.figure()
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.rand(50, 4), columns=["a", "b", "c", "d"])
+   df["species"] = pd.Categorical(
+       ["setosa"] * 20 + ["versicolor"] * 20 + ["virginica"] * 10
+   )
+
+   @savefig scatter_plot.png
+   df.plot.scatter(x="a", y="b");
+
+To plot multiple column groups in a single axes, repeat ``plot`` method specifying target ``ax``.
+It is recommended to specify ``color`` and ``label`` keywords to distinguish each groups.
+
+.. ipython:: python
+   :okwarning:
+
+   ax = df.plot.scatter(x="a", y="b", color="DarkBlue", label="Group 1")
+   @savefig scatter_plot_repeated.png
+   df.plot.scatter(x="c", y="d", color="DarkGreen", label="Group 2", ax=ax);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+The keyword ``c`` may be given as the name of a column to provide colors for
+each point:
+
+.. ipython:: python
+
+   @savefig scatter_plot_colored.png
+   df.plot.scatter(x="a", y="b", c="c", s=50);
+
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+If a categorical column is passed to ``c``, then a discrete colorbar will be produced:
+
+.. versionadded:: 1.3.0
+
+.. ipython:: python
+
+   @savefig scatter_plot_categorical.png
+   df.plot.scatter(x="a", y="b", c="species", cmap="viridis", s=50);
+
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+You can pass other keywords supported by matplotlib
+:meth:`scatter <matplotlib.axes.Axes.scatter>`. The example  below shows a
+bubble chart using a column of the ``DataFrame`` as the bubble size.
+
+.. ipython:: python
+
+   @savefig scatter_plot_bubble.png
+   df.plot.scatter(x="a", y="b", s=df["c"] * 200);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+See the :meth:`scatter <matplotlib.axes.Axes.scatter>` method and the
+`matplotlib scatter documentation <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.scatter.html>`__ for more.
+
+.. _visualization.hexbin:
+
+Hexagonal bin plot
+~~~~~~~~~~~~~~~~~~
+
+You can create hexagonal bin plots with :meth:`DataFrame.plot.hexbin`.
+Hexbin plots can be a useful alternative to scatter plots if your data are
+too dense to plot each point individually.
+
+.. ipython:: python
+   :suppress:
+
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(1000, 2), columns=["a", "b"])
+   df["b"] = df["b"] + np.arange(1000)
+
+   @savefig hexbin_plot.png
+   df.plot.hexbin(x="a", y="b", gridsize=25);
+
+
+A useful keyword argument is ``gridsize``; it controls the number of hexagons
+in the x-direction, and defaults to 100. A larger ``gridsize`` means more, smaller
+bins.
+
+By default, a histogram of the counts around each ``(x, y)`` point is computed.
+You can specify alternative aggregations by passing values to the ``C`` and
+``reduce_C_function`` arguments. ``C`` specifies the value at each ``(x, y)`` point
+and ``reduce_C_function`` is a function of one argument that reduces all the
+values in a bin to a single number (e.g. ``mean``, ``max``, ``sum``, ``std``).  In this
+example the positions are given by columns ``a`` and ``b``, while the value is
+given by column ``z``. The bins are aggregated with NumPy's ``max`` function.
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(1000, 2), columns=["a", "b"])
+   df["b"] = df["b"] + np.arange(1000)
+   df["z"] = np.random.uniform(0, 3, 1000)
+
+   @savefig hexbin_plot_agg.png
+   df.plot.hexbin(x="a", y="b", C="z", reduce_C_function=np.max, gridsize=25);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+See the :meth:`hexbin <matplotlib.axes.Axes.hexbin>` method and the
+`matplotlib hexbin documentation <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.hexbin.html>`__ for more.
+
+.. _visualization.pie:
+
+Pie plot
+~~~~~~~~
+
+You can create a pie plot with :meth:`DataFrame.plot.pie` or :meth:`Series.plot.pie`.
+If your data includes any ``NaN``, they will be automatically filled with 0.
+A ``ValueError`` will be raised if there are any negative values in your data.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+   plt.figure()
+
+.. ipython:: python
+   :okwarning:
+
+   series = pd.Series(3 * np.random.rand(4), index=["a", "b", "c", "d"], name="series")
+
+   @savefig series_pie_plot.png
+   series.plot.pie(figsize=(6, 6));
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+For pie plots it's best to use square figures, i.e. a figure aspect ratio 1.
+You can create the figure with equal width and height, or force the aspect ratio
+to be equal after plotting by calling ``ax.set_aspect('equal')`` on the returned
+``axes`` object.
+
+Note that pie plot with :class:`DataFrame` requires that you either specify a
+target column by the ``y`` argument or ``subplots=True``. When ``y`` is
+specified, pie plot of selected column will be drawn. If ``subplots=True`` is
+specified, pie plots for each column are drawn as subplots. A legend will be
+drawn in each pie plots by default; specify ``legend=False`` to hide it.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+   plt.figure()
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       3 * np.random.rand(4, 2), index=["a", "b", "c", "d"], columns=["x", "y"]
+   )
+
+   @savefig df_pie_plot.png
+   df.plot.pie(subplots=True, figsize=(8, 4));
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+You can use the ``labels`` and ``colors`` keywords to specify the labels and colors of each wedge.
+
+.. warning::
+
+   Most pandas plots use the ``label`` and ``color`` arguments (note the lack of "s" on those).
+   To be consistent with :func:`matplotlib.pyplot.pie` you must use ``labels`` and ``colors``.
+
+If you want to hide wedge labels, specify ``labels=None``.
+If ``fontsize`` is specified, the value will be applied to wedge labels.
+Also, other keywords supported by :func:`matplotlib.pyplot.pie` can be used.
+
+
+.. ipython:: python
+   :suppress:
+
+   plt.figure()
+
+.. ipython:: python
+
+   @savefig series_pie_plot_options.png
+   series.plot.pie(
+       labels=["AA", "BB", "CC", "DD"],
+       colors=["r", "g", "b", "c"],
+       autopct="%.2f",
+       fontsize=20,
+       figsize=(6, 6),
+   );
+
+If you pass values whose sum total is less than 1.0 they will be rescaled so that they sum to 1.
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+   plt.figure()
+
+.. ipython:: python
+   :okwarning:
+
+   series = pd.Series([0.1] * 4, index=["a", "b", "c", "d"], name="series2")
+
+   @savefig series_pie_plot_semi.png
+   series.plot.pie(figsize=(6, 6));
+
+See the `matplotlib pie documentation <https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.pie.html>`__ for more.
+
+.. ipython:: python
+    :suppress:
+
+    plt.close("all")
+
+.. _visualization.missing_data:
+
+Plotting with missing data
+--------------------------
+
+pandas tries to be pragmatic about plotting ``DataFrames`` or ``Series``
+that contain missing data. Missing values are dropped, left out, or filled
+depending on the plot type.
+
++----------------+--------------------------------------+
+| Plot Type      | NaN Handling                         |
++================+======================================+
+| Line           | Leave gaps at NaNs                   |
++----------------+--------------------------------------+
+| Line (stacked) | Fill 0's                             |
++----------------+--------------------------------------+
+| Bar            | Fill 0's                             |
++----------------+--------------------------------------+
+| Scatter        | Drop NaNs                            |
++----------------+--------------------------------------+
+| Histogram      | Drop NaNs (column-wise)              |
++----------------+--------------------------------------+
+| Box            | Drop NaNs (column-wise)              |
++----------------+--------------------------------------+
+| Area           | Fill 0's                             |
++----------------+--------------------------------------+
+| KDE            | Drop NaNs (column-wise)              |
++----------------+--------------------------------------+
+| Hexbin         | Drop NaNs                            |
++----------------+--------------------------------------+
+| Pie            | Fill 0's                             |
++----------------+--------------------------------------+
+
+If any of these defaults are not what you want, or if you want to be
+explicit about how missing values are handled, consider using
+:meth:`~pandas.DataFrame.fillna` or :meth:`~pandas.DataFrame.dropna`
+before plotting.
+
+.. _visualization.tools:
+
+Plotting tools
+--------------
+
+These functions can be imported from ``pandas.plotting``
+and take a :class:`Series` or :class:`DataFrame` as an argument.
+
+.. _visualization.scatter_matrix:
+
+Scatter matrix plot
+~~~~~~~~~~~~~~~~~~~
+
+You can create a scatter plot matrix using the
+``scatter_matrix`` method in ``pandas.plotting``:
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   from pandas.plotting import scatter_matrix
+
+   df = pd.DataFrame(np.random.randn(1000, 4), columns=["a", "b", "c", "d"])
+
+   @savefig scatter_matrix_kde.png
+   scatter_matrix(df, alpha=0.2, figsize=(6, 6), diagonal="kde");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.kde:
+
+Density plot
+~~~~~~~~~~~~
+
+You can create density plots using the :meth:`Series.plot.kde` and :meth:`DataFrame.plot.kde` methods.
+
+.. ipython:: python
+   :suppress:
+
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   ser = pd.Series(np.random.randn(1000))
+
+   @savefig kde_plot.png
+   ser.plot.kde();
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.andrews_curves:
+
+Andrews curves
+~~~~~~~~~~~~~~
+
+Andrews curves allow one to plot multivariate data as a large number
+of curves that are created using the attributes of samples as coefficients
+for Fourier series, see the `Wikipedia entry <https://en.wikipedia.org/wiki/Andrews_plot>`__
+for more information. By coloring these curves differently for each class
+it is possible to visualize data clustering. Curves belonging to samples
+of the same class will usually be closer together and form larger structures.
+
+**Note**: The "Iris" dataset is available `here <https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/csv/iris.csv>`__.
+
+.. ipython:: python
+
+   from pandas.plotting import andrews_curves
+
+   data = pd.read_csv("data/iris.data")
+
+   plt.figure();
+
+   @savefig andrews_curves.png
+   andrews_curves(data, "Name");
+
+.. _visualization.parallel_coordinates:
+
+Parallel coordinates
+~~~~~~~~~~~~~~~~~~~~
+
+Parallel coordinates is a plotting technique for plotting multivariate data,
+see the `Wikipedia entry <https://en.wikipedia.org/wiki/Parallel_coordinates>`__
+for an introduction.
+Parallel coordinates allows one to see clusters in data and to estimate other statistics visually.
+Using parallel coordinates points are represented as connected line segments.
+Each vertical line represents one attribute. One set of connected line segments
+represents one data point. Points that tend to cluster will appear closer together.
+
+.. ipython:: python
+
+   from pandas.plotting import parallel_coordinates
+
+   data = pd.read_csv("data/iris.data")
+
+   plt.figure();
+
+   @savefig parallel_coordinates.png
+   parallel_coordinates(data, "Name");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.lag:
+
+Lag plot
+~~~~~~~~
+
+Lag plots are used to check if a data set or time series is random. Random
+data should not exhibit any structure in the lag plot. Non-random structure
+implies that the underlying data are not random. The ``lag`` argument may
+be passed, and when ``lag=1`` the plot is essentially ``data[:-1]`` vs.
+``data[1:]``.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   from pandas.plotting import lag_plot
+
+   plt.figure();
+
+   spacing = np.linspace(-99 * np.pi, 99 * np.pi, num=1000)
+   data = pd.Series(0.1 * np.random.rand(1000) + 0.9 * np.sin(spacing))
+
+   @savefig lag_plot.png
+   lag_plot(data);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.autocorrelation:
+
+Autocorrelation plot
+~~~~~~~~~~~~~~~~~~~~
+
+Autocorrelation plots are often used for checking randomness in time series.
+This is done by computing autocorrelations for data values at varying time lags.
+If time series is random, such autocorrelations should be near zero for any and
+all time-lag separations. If time series is non-random then one or more of the
+autocorrelations will be significantly non-zero. The horizontal lines displayed
+in the plot correspond to 95% and 99% confidence bands. The dashed line is 99%
+confidence band. See the
+`Wikipedia entry <https://en.wikipedia.org/wiki/Correlogram>`__ for more about
+autocorrelation plots.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   from pandas.plotting import autocorrelation_plot
+
+   plt.figure();
+
+   spacing = np.linspace(-9 * np.pi, 9 * np.pi, num=1000)
+   data = pd.Series(0.7 * np.random.rand(1000) + 0.3 * np.sin(spacing))
+
+   @savefig autocorrelation_plot.png
+   autocorrelation_plot(data);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.bootstrap:
+
+Bootstrap plot
+~~~~~~~~~~~~~~
+
+Bootstrap plots are used to visually assess the uncertainty of a statistic, such
+as mean, median, midrange, etc. A random subset of a specified size is selected
+from a data set, the statistic in question is computed for this subset and the
+process is repeated a specified number of times. Resulting plots and histograms
+are what constitutes the bootstrap plot.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   from pandas.plotting import bootstrap_plot
+
+   data = pd.Series(np.random.rand(1000))
+
+   @savefig bootstrap_plot.png
+   bootstrap_plot(data, size=50, samples=500, color="grey");
+
+.. ipython:: python
+   :suppress:
+
+    plt.close("all")
+
+.. _visualization.radviz:
+
+RadViz
+~~~~~~
+
+RadViz is a way of visualizing multi-variate data. It is based on a simple
+spring tension minimization algorithm. Basically you set up a bunch of points in
+a plane. In our case they are equally spaced on a unit circle. Each point
+represents a single attribute. You then pretend that each sample in the data set
+is attached to each of these points by a spring, the stiffness of which is
+proportional to the numerical value of that attribute (they are normalized to
+unit interval). The point in the plane, where our sample settles to (where the
+forces acting on our sample are at an equilibrium) is where a dot representing
+our sample will be drawn. Depending on which class that sample belongs it will
+be colored differently.
+See the R package `Radviz <https://cran.r-project.org/web/packages/Radviz/index.html>`__
+for more information.
+
+**Note**: The "Iris" dataset is available `here <https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/csv/iris.csv>`__.
+
+.. ipython:: python
+
+   from pandas.plotting import radviz
+
+   data = pd.read_csv("data/iris.data")
+
+   plt.figure();
+
+   @savefig radviz.png
+   radviz(data, "Name");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.formatting:
+
+Plot formatting
+---------------
+
+Setting the plot style
+~~~~~~~~~~~~~~~~~~~~~~
+
+From version 1.5 and up, matplotlib offers a range of pre-configured plotting styles. Setting the
+style can be used to easily give plots the general look that you want.
+Setting the style is as easy as calling ``matplotlib.style.use(my_plot_style)`` before
+creating your plot. For example you could write ``matplotlib.style.use('ggplot')`` for ggplot-style
+plots.
+
+You can see the various available style names at ``matplotlib.style.available`` and it's very
+easy to try them out.
+
+General plot style arguments
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Most plotting methods have a set of keyword arguments that control the
+layout and formatting of the returned plot:
+
+.. ipython:: python
+
+   plt.figure();
+   @savefig series_plot_basic2.png
+   ts.plot(style="k--", label="Series");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+For each kind of plot (e.g. ``line``, ``bar``, ``scatter``) any additional arguments
+keywords are passed along to the corresponding matplotlib function
+(:meth:`ax.plot() <matplotlib.axes.Axes.plot>`,
+:meth:`ax.bar() <matplotlib.axes.Axes.bar>`,
+:meth:`ax.scatter() <matplotlib.axes.Axes.scatter>`). These can be used
+to control additional styling, beyond what pandas provides.
+
+Controlling the legend
+~~~~~~~~~~~~~~~~~~~~~~
+
+You may set the ``legend`` argument to ``False`` to hide the legend, which is
+shown by default.
+
+.. ipython:: python
+   :suppress:
+
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list("ABCD"))
+   df = df.cumsum()
+
+   @savefig frame_plot_basic_noleg.png
+   df.plot(legend=False);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+
+Controlling the labels
+~~~~~~~~~~~~~~~~~~~~~~
+
+You may set the ``xlabel`` and ``ylabel`` arguments to give the plot custom labels
+for x and y axis. By default, pandas will pick up index name as xlabel, while leaving
+it empty for ylabel.
+
+.. ipython:: python
+
+   df.plot();
+
+   @savefig plot_xlabel_ylabel.png
+   df.plot(xlabel="new x", ylabel="new y");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+
+Scales
+~~~~~~
+
+You may pass ``logy`` to get a log-scale Y axis.
+
+.. ipython:: python
+   :suppress:
+
+   plt.figure()
+   np.random.seed(123456)
+
+.. ipython:: python
+
+   ts = pd.Series(np.random.randn(1000), index=pd.date_range("1/1/2000", periods=1000))
+   ts = np.exp(ts.cumsum())
+
+   @savefig series_plot_logy.png
+   ts.plot(logy=True);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+See also the ``logx`` and ``loglog`` keyword arguments.
+
+Plotting on a secondary y-axis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To plot data on a secondary y-axis, use the ``secondary_y`` keyword:
+
+.. ipython:: python
+   :suppress:
+
+   plt.figure()
+
+.. ipython:: python
+
+   df["A"].plot();
+
+   @savefig series_plot_secondary_y.png
+   df["B"].plot(secondary_y=True, style="g");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+To plot some columns in a ``DataFrame``, give the column names to the ``secondary_y``
+keyword:
+
+.. ipython:: python
+
+   plt.figure();
+   ax = df.plot(secondary_y=["A", "B"])
+   ax.set_ylabel("CD scale");
+   @savefig frame_plot_secondary_y.png
+   ax.right_ax.set_ylabel("AB scale");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Note that the columns plotted on the secondary y-axis is automatically marked
+with "(right)" in the legend. To turn off the automatic marking, use the
+``mark_right=False`` keyword:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig frame_plot_secondary_y_no_right.png
+   df.plot(secondary_y=["A", "B"], mark_right=False);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _plotting.formatters:
+
+Custom formatters for timeseries plots
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas provides custom formatters for timeseries plots. These change the
+formatting of the axis labels for dates and times. By default,
+the custom formatters are applied only to plots created by pandas with
+:meth:`DataFrame.plot` or :meth:`Series.plot`. To have them apply to all
+plots, including those made by matplotlib, set the option
+``pd.options.plotting.matplotlib.register_converters = True`` or use
+:meth:`pandas.plotting.register_matplotlib_converters`.
+
+Suppressing tick resolution adjustment
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas includes automatic tick resolution adjustment for regular frequency
+time-series data. For limited cases where pandas cannot infer the frequency
+information (e.g., in an externally created ``twinx``), you can choose to
+suppress this behavior for alignment purposes.
+
+Here is the default behavior, notice how the x-axis tick labeling is performed:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig ser_plot_suppress.png
+   df["A"].plot();
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Using the ``x_compat`` parameter, you can suppress this behavior:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig ser_plot_suppress_parm.png
+   df["A"].plot(x_compat=True);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+If you have more than one plot that needs to be suppressed, the ``use`` method
+in ``pandas.plotting.plot_params`` can be used in a ``with`` statement:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig ser_plot_suppress_context.png
+   with pd.plotting.plot_params.use("x_compat", True):
+       df["A"].plot(color="r")
+       df["B"].plot(color="g")
+       df["C"].plot(color="b")
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Automatic date tick adjustment
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+``TimedeltaIndex`` now uses the native matplotlib
+tick locator methods, it is useful to call the automatic
+date tick adjustment from matplotlib for figures whose ticklabels overlap.
+
+See the :meth:`autofmt_xdate <matplotlib.figure.autofmt_xdate>` method and the
+`matplotlib documentation <https://matplotlib.org/2.0.2/users/recipes.html#fixing-common-date-annoyances>`__ for more.
+
+Subplots
+~~~~~~~~
+
+Each ``Series`` in a ``DataFrame`` can be plotted on a different axis
+with the ``subplots`` keyword:
+
+.. ipython:: python
+
+   @savefig frame_plot_subplots.png
+   df.plot(subplots=True, figsize=(6, 6));
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Using layout and targeting multiple axes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The layout of subplots can be specified by the ``layout`` keyword. It can accept
+``(rows, columns)``. The ``layout`` keyword can be used in
+``hist`` and ``boxplot`` also. If the input is invalid, a ``ValueError`` will be raised.
+
+The number of axes which can be contained by rows x columns specified by ``layout`` must be
+larger than the number of required subplots. If layout can contain more axes than required,
+blank axes are not drawn. Similar to a NumPy array's ``reshape`` method, you
+can use ``-1`` for one dimension to automatically calculate the number of rows
+or columns needed, given the other.
+
+.. ipython:: python
+
+   @savefig frame_plot_subplots_layout.png
+   df.plot(subplots=True, layout=(2, 3), figsize=(6, 6), sharex=False);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+The above example is identical to using:
+
+.. ipython:: python
+
+   df.plot(subplots=True, layout=(2, -1), figsize=(6, 6), sharex=False);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+The required number of columns (3) is inferred from the number of series to plot
+and the given number of rows (2).
+
+You can pass multiple axes created beforehand as list-like via ``ax`` keyword.
+This allows more complicated layouts.
+The passed axes must be the same number as the subplots being drawn.
+
+When multiple axes are passed via the ``ax`` keyword, ``layout``, ``sharex`` and ``sharey`` keywords
+don't affect to the output. You should explicitly pass ``sharex=False`` and ``sharey=False``,
+otherwise you will see a warning.
+
+.. ipython:: python
+
+   fig, axes = plt.subplots(4, 4, figsize=(9, 9))
+   plt.subplots_adjust(wspace=0.5, hspace=0.5)
+   target1 = [axes[0][0], axes[1][1], axes[2][2], axes[3][3]]
+   target2 = [axes[3][0], axes[2][1], axes[1][2], axes[0][3]]
+
+   df.plot(subplots=True, ax=target1, legend=False, sharex=False, sharey=False);
+   @savefig frame_plot_subplots_multi_ax.png
+   (-df).plot(subplots=True, ax=target2, legend=False, sharex=False, sharey=False);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Another option is passing an ``ax`` argument to :meth:`Series.plot` to plot on a particular axis:
+
+.. ipython:: python
+
+   np.random.seed(123456)
+   ts = pd.Series(np.random.randn(1000), index=pd.date_range("1/1/2000", periods=1000))
+   ts = ts.cumsum()
+
+   df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list("ABCD"))
+   df = df.cumsum()
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. ipython:: python
+
+   fig, axes = plt.subplots(nrows=2, ncols=2)
+   plt.subplots_adjust(wspace=0.2, hspace=0.5)
+   df["A"].plot(ax=axes[0, 0]);
+   axes[0, 0].set_title("A");
+   df["B"].plot(ax=axes[0, 1]);
+   axes[0, 1].set_title("B");
+   df["C"].plot(ax=axes[1, 0]);
+   axes[1, 0].set_title("C");
+   df["D"].plot(ax=axes[1, 1]);
+   @savefig series_plot_multi.png
+   axes[1, 1].set_title("D");
+
+.. ipython:: python
+   :suppress:
+
+    plt.close("all")
+
+.. _visualization.errorbars:
+
+Plotting with error bars
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Plotting with error bars is supported in :meth:`DataFrame.plot` and :meth:`Series.plot`.
+
+Horizontal and vertical error bars can be supplied to the ``xerr`` and ``yerr`` keyword arguments to :meth:`~DataFrame.plot`. The error values can be specified using a variety of formats:
+
+* As a :class:`DataFrame` or ``dict`` of errors with column names matching the ``columns`` attribute of the plotting :class:`DataFrame` or matching the ``name`` attribute of the :class:`Series`.
+* As a ``str`` indicating which of the columns of plotting :class:`DataFrame` contain the error values.
+* As raw values (``list``, ``tuple``, or ``np.ndarray``). Must be the same length as the plotting :class:`DataFrame`/:class:`Series`.
+
+Here is an example of one way to easily plot group means with standard deviations from the raw data.
+
+.. ipython:: python
+
+   # Generate the data
+   ix3 = pd.MultiIndex.from_arrays(
+       [
+           ["a", "a", "a", "a", "a", "b", "b", "b", "b", "b"],
+           ["foo", "foo", "foo", "bar", "bar", "foo", "foo", "bar", "bar", "bar"],
+       ],
+       names=["letter", "word"],
+   )
+
+   df3 = pd.DataFrame(
+       {
+           "data1": [9, 3, 2, 4, 3, 2, 4, 6, 3, 2],
+           "data2": [9, 6, 5, 7, 5, 4, 5, 6, 5, 1],
+       },
+       index=ix3,
+   )
+
+   # Group by index labels and take the means and standard deviations
+   # for each group
+   gp3 = df3.groupby(level=("letter", "word"))
+   means = gp3.mean()
+   errors = gp3.std()
+   means
+   errors
+
+   # Plot
+   fig, ax = plt.subplots()
+   @savefig errorbar_example.png
+   means.plot.bar(yerr=errors, ax=ax, capsize=4, rot=0);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Asymmetrical error bars are also supported, however raw error values must be provided in this case. For a ``N`` length :class:`Series`, a ``2xN`` array should be provided indicating lower and upper (or left and right) errors. For a ``MxN`` :class:`DataFrame`, asymmetrical errors should be in a ``Mx2xN`` array.
+
+Here is an example of one way to plot the min/max range using asymmetrical error bars.
+
+.. ipython:: python
+
+   mins = gp3.min()
+   maxs = gp3.max()
+
+   # errors should be positive, and defined in the order of lower, upper
+   errors = [[means[c] - mins[c], maxs[c] - means[c]] for c in df3.columns]
+
+   # Plot
+   fig, ax = plt.subplots()
+   @savefig errorbar_asymmetrical_example.png
+   means.plot.bar(yerr=errors, ax=ax, capsize=4, rot=0);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+.. _visualization.table:
+
+Plotting tables
+~~~~~~~~~~~~~~~
+
+Plotting with matplotlib table is now supported in  :meth:`DataFrame.plot` and :meth:`Series.plot` with a ``table`` keyword. The ``table`` keyword can accept ``bool``, :class:`DataFrame` or :class:`Series`. The simple way to draw a table is to specify ``table=True``. Data will be transposed to meet matplotlib's default layout.
+
+.. ipython:: python
+
+   np.random.seed(123456)
+   fig, ax = plt.subplots(1, 1, figsize=(7, 6.5))
+   df = pd.DataFrame(np.random.rand(5, 3), columns=["a", "b", "c"])
+   ax.xaxis.tick_top()  # Display x-axis ticks on top.
+
+   @savefig line_plot_table_true.png
+   df.plot(table=True, ax=ax);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Also, you can pass a different :class:`DataFrame` or :class:`Series` to the
+``table`` keyword. The data will be drawn as displayed in print method
+(not transposed automatically). If required, it should be transposed manually
+as seen in the example below.
+
+.. ipython:: python
+
+   fig, ax = plt.subplots(1, 1, figsize=(7, 6.75))
+   ax.xaxis.tick_top()  # Display x-axis ticks on top.
+
+   @savefig line_plot_table_data.png
+   df.plot(table=np.round(df.T, 2), ax=ax);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+There also exists a helper function ``pandas.plotting.table``, which creates a
+table from :class:`DataFrame` or :class:`Series`, and adds it to an
+``matplotlib.Axes`` instance. This function can accept keywords which the
+matplotlib `table <https://matplotlib.org/stable/api/_as_gen/matplotlib.axes.Axes.table.html>`__ has.
+
+.. ipython:: python
+
+   from pandas.plotting import table
+
+   fig, ax = plt.subplots(1, 1)
+
+   table(ax, np.round(df.describe(), 2), loc="upper right", colWidths=[0.2, 0.2, 0.2]);
+
+   @savefig line_plot_table_describe.png
+   df.plot(ax=ax, ylim=(0, 2), legend=None);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+**Note**: You can get table instances on the axes using ``axes.tables`` property for further decorations. See the `matplotlib table documentation <https://matplotlib.org/api/axes_api.html#matplotlib.axes.Axes.table>`__ for more.
+
+.. _visualization.colormaps:
+
+Colormaps
+~~~~~~~~~
+
+A potential issue when plotting a large number of columns is that it can be
+difficult to distinguish some series due to repetition in the default colors. To
+remedy this, ``DataFrame`` plotting supports the use of the ``colormap`` argument,
+which accepts either a Matplotlib `colormap <https://matplotlib.org/api/cm_api.html>`__
+or a string that is a name of a colormap registered with Matplotlib. A
+visualization of the default matplotlib colormaps is available `here
+<https://matplotlib.org/stable/gallery/color/colormap_reference.html>`__.
+
+As matplotlib does not directly support colormaps for line-based plots, the
+colors are selected based on an even spacing determined by the number of columns
+in the ``DataFrame``. There is no consideration made for background color, so some
+colormaps will produce lines that are not easily visible.
+
+To use the cubehelix colormap, we can pass ``colormap='cubehelix'``.
+
+.. ipython:: python
+
+   np.random.seed(123456)
+   df = pd.DataFrame(np.random.randn(1000, 10), index=ts.index)
+   df = df.cumsum()
+
+   plt.figure();
+
+   @savefig cubehelix.png
+   df.plot(colormap="cubehelix");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Alternatively, we can pass the colormap itself:
+
+.. ipython:: python
+
+   from matplotlib import cm
+
+   plt.figure();
+
+   @savefig cubehelix_cm.png
+   df.plot(colormap=cm.cubehelix);
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Colormaps can also be used other plot types, like bar charts:
+
+.. ipython:: python
+
+   np.random.seed(123456)
+   dd = pd.DataFrame(np.random.randn(10, 10)).map(abs)
+   dd = dd.cumsum()
+
+   plt.figure();
+
+   @savefig greens.png
+   dd.plot.bar(colormap="Greens");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Parallel coordinates charts:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig parallel_gist_rainbow.png
+   parallel_coordinates(data, "Name", colormap="gist_rainbow");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Andrews curves charts:
+
+.. ipython:: python
+
+   plt.figure();
+
+   @savefig andrews_curve_winter.png
+   andrews_curves(data, "Name", colormap="winter");
+
+.. ipython:: python
+   :suppress:
+
+   plt.close("all")
+
+Plotting directly with Matplotlib
+---------------------------------
+
+In some situations it may still be preferable or necessary to prepare plots
+directly with matplotlib, for instance when a certain type of plot or
+customization is not (yet) supported by pandas. ``Series`` and ``DataFrame``
+objects behave like arrays and can therefore be passed directly to
+matplotlib functions without explicit casts.
+
+pandas also automatically registers formatters and locators that recognize date
+indices, thereby extending date and time support to practically all plot types
+available in matplotlib. Although this formatting does not provide the same
+level of refinement you would get when plotting via pandas, it can be faster
+when plotting a large number of points.
+
+.. ipython:: python
+
+   np.random.seed(123456)
+   price = pd.Series(
+       np.random.randn(150).cumsum(),
+       index=pd.date_range("2000-1-1", periods=150, freq="B"),
+   )
+   ma = price.rolling(20).mean()
+   mstd = price.rolling(20).std()
+
+   plt.figure();
+
+   plt.plot(price.index, price, "k");
+   plt.plot(ma.index, ma, "b");
+   @savefig bollinger.png
+   plt.fill_between(mstd.index, ma - 2 * mstd, ma + 2 * mstd, color="b", alpha=0.2);
+
+.. ipython:: python
+   :suppress:
+
+    plt.close("all")
+
+Plotting backends
+-----------------
+
+pandas can be extended with third-party plotting backends. The
+main idea is letting users select a plotting backend different than the provided
+one based on Matplotlib.
+
+This can be done by passing 'backend.module' as the argument ``backend`` in ``plot``
+function. For example:
+
+.. code-block:: python
+
+    >>> Series([1, 2, 3]).plot(backend="backend.module")
+
+Alternatively, you can also set this option globally, do you don't need to specify
+the keyword in each ``plot`` call. For example:
+
+.. code-block:: python
+
+    >>> pd.set_option("plotting.backend", "backend.module")
+    >>> pd.Series([1, 2, 3]).plot()
+
+Or:
+
+.. code-block:: python
+
+    >>> pd.options.plotting.backend = "backend.module"
+    >>> pd.Series([1, 2, 3]).plot()
+
+This would be more or less equivalent to:
+
+.. code-block:: python
+
+    >>> import backend.module
+    >>> backend.module.plot(pd.Series([1, 2, 3]))
+
+The backend module can then use other visualization tools (Bokeh, Altair, hvplot,...)
+to generate the plots. Some libraries implementing a backend for pandas are listed
+on `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_.
+
+Developers guide can be found at
+https://pandas.pydata.org/docs/dev/development/extending.html#plotting-backends
diff --git a/pandas/doc/source/user_guide/window.rst b/pandas/doc/source/user_guide/window.rst
new file mode 100644
index 0000000000000..5b27442c80bb8
--- /dev/null
+++ b/pandas/doc/source/user_guide/window.rst
@@ -0,0 +1,637 @@
+.. _window:
+
+{{ header }}
+
+********************
+Windowing operations
+********************
+
+pandas contains a compact set of APIs for performing windowing operations - an operation that performs
+an aggregation over a sliding partition of values. The API functions similarly to the ``groupby`` API
+in that :class:`Series` and :class:`DataFrame` call the windowing method with
+necessary parameters and then subsequently call the aggregation function.
+
+.. ipython:: python
+
+   s = pd.Series(range(5))
+   s.rolling(window=2).sum()
+
+The windows are comprised by looking back the length of the window from the current observation.
+The result above can be derived by taking the sum of the following windowed partitions of data:
+
+.. ipython:: python
+
+   for window in s.rolling(window=2):
+       print(window)
+
+
+.. _window.overview:
+
+Overview
+--------
+
+pandas supports 4 types of windowing operations:
+
+#. Rolling window: Generic fixed or variable sliding window over the values.
+#. Weighted window: Weighted, non-rectangular window supplied by the ``scipy.signal`` library.
+#. Expanding window: Accumulating window over the values.
+#. Exponentially Weighted window: Accumulating and exponentially weighted window over the values.
+
+=============================   =================  =============================================   ===========================  ========================  ===================================  ===========================
+Concept                         Method             Returned Object                                 Supports time-based windows  Supports chained groupby  Supports table method                Supports online operations
+=============================   =================  =============================================   ===========================  ========================  ===================================  ===========================
+Rolling window                  ``rolling``        ``pandas.typing.api.Rolling``                   Yes                          Yes                       Yes (as of version 1.3)              No
+Weighted window                 ``rolling``        ``pandas.typing.api.Window``                    No                           No                        No                                   No
+Expanding window                ``expanding``      ``pandas.typing.api.Expanding``                 No                           Yes                       Yes (as of version 1.3)              No
+Exponentially Weighted window   ``ewm``            ``pandas.typing.api.ExponentialMovingWindow``   No                           Yes (as of version 1.2)   No                                   Yes (as of version 1.3)
+=============================   =================  =============================================   ===========================  ========================  ===================================  ===========================
+
+As noted above, some operations support specifying a window based on a time offset:
+
+.. ipython:: python
+
+   s = pd.Series(range(5), index=pd.date_range('2020-01-01', periods=5, freq='1D'))
+   s.rolling(window='2D').sum()
+
+Additionally, some methods support chaining a ``groupby`` operation with a windowing operation
+which will first group the data by the specified keys and then perform a windowing operation per group.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': ['a', 'b', 'a', 'b', 'a'], 'B': range(5)})
+   df.groupby('A').expanding().sum()
+
+.. note::
+
+   Windowing operations currently only support numeric data (integer and float)
+   and will always return ``float64`` values.
+
+.. warning::
+
+    Some windowing aggregation, ``mean``, ``sum``, ``var`` and ``std`` methods may suffer from numerical
+    imprecision due to the underlying windowing algorithms accumulating sums. When values differ
+    with magnitude ``1/np.finfo(np.double).eps`` (approximately :math:`4.5 \times 10^{15}`),
+    this results in truncation. It must be
+    noted, that large values may have an impact on windows, which do not include these values. `Kahan summation
+    <https://en.wikipedia.org/wiki/Kahan_summation_algorithm>`__ is used
+    to compute the rolling sums to preserve accuracy as much as possible.
+
+
+.. versionadded:: 1.3.0
+
+Some windowing operations also support the ``method='table'`` option in the constructor which
+performs the windowing operation over an entire :class:`DataFrame` instead of a single column at a time.
+This can provide a useful performance benefit for a :class:`DataFrame` with many columns
+or the ability to utilize other columns during the windowing
+operation. The ``method='table'`` option can only be used if ``engine='numba'`` is specified
+in the corresponding method call.
+
+For example, a `weighted mean <https://en.wikipedia.org/wiki/Weighted_arithmetic_mean>`__ calculation can
+be calculated with :meth:`~Rolling.apply` by specifying a separate column of weights.
+
+.. ipython:: python
+   :okwarning:
+
+   def weighted_mean(x):
+       arr = np.ones((1, x.shape[1]))
+       arr[:, :2] = (x[:, :2] * x[:, 2]).sum(axis=0) / x[:, 2].sum()
+       return arr
+
+   df = pd.DataFrame([[1, 2, 0.6], [2, 3, 0.4], [3, 4, 0.2], [4, 5, 0.7]])
+   df.rolling(2, method="table", min_periods=0).apply(weighted_mean, raw=True, engine="numba")  # noqa: E501
+
+.. versionadded:: 1.3
+
+Some windowing operations also support an ``online`` method after constructing a windowing object
+which returns a new object that supports passing in new :class:`DataFrame` or :class:`Series` objects
+to continue the windowing calculation with the new values (i.e. online calculations).
+
+The methods on this new windowing objects must call the aggregation method first to "prime" the initial
+state of the online calculation. Then, new :class:`DataFrame` or :class:`Series` objects can be passed in
+the ``update`` argument to continue the windowing calculation.
+
+.. ipython:: python
+
+   df = pd.DataFrame([[1, 2, 0.6], [2, 3, 0.4], [3, 4, 0.2], [4, 5, 0.7]])
+   df.ewm(0.5).mean()
+
+.. ipython:: python
+   :okwarning:
+
+   online_ewm = df.head(2).ewm(0.5).online()
+   online_ewm.mean()
+   online_ewm.mean(update=df.tail(1))
+
+All windowing operations support a ``min_periods`` argument that dictates the minimum amount of
+non-``np.nan`` values a window must have; otherwise, the resulting value is ``np.nan``.
+``min_periods`` defaults to 1 for time-based windows and ``window`` for fixed windows
+
+.. ipython:: python
+
+   s = pd.Series([np.nan, 1, 2, np.nan, np.nan, 3])
+   s.rolling(window=3, min_periods=1).sum()
+   s.rolling(window=3, min_periods=2).sum()
+   # Equivalent to min_periods=3
+   s.rolling(window=3, min_periods=None).sum()
+
+
+Additionally, all windowing operations supports the ``aggregate`` method for returning a result
+of multiple aggregations applied to a window.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": range(5), "B": range(10, 15)})
+   df.expanding().agg(["sum", "mean", "std"])
+
+
+.. _window.generic:
+
+Rolling window
+--------------
+
+Generic rolling windows support specifying windows as a fixed number of observations or variable
+number of observations based on an offset. If a time based offset is provided, the corresponding
+time based index must be monotonic.
+
+.. ipython:: python
+
+   times = ['2020-01-01', '2020-01-03', '2020-01-04', '2020-01-05', '2020-01-29']
+   s = pd.Series(range(5), index=pd.DatetimeIndex(times))
+   s
+   # Window with 2 observations
+   s.rolling(window=2).sum()
+   # Window with 2 days worth of observations
+   s.rolling(window='2D').sum()
+
+For all supported aggregation functions, see :ref:`api.functions_rolling`.
+
+.. _window.center:
+
+Centering windows
+~~~~~~~~~~~~~~~~~
+
+By default the labels are set to the right edge of the window, but a
+``center`` keyword is available so the labels can be set at the center.
+
+.. ipython:: python
+
+   s = pd.Series(range(10))
+   s.rolling(window=5).mean()
+   s.rolling(window=5, center=True).mean()
+
+
+This can also be applied to datetime-like indices.
+
+.. versionadded:: 1.3.0
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {"A": [0, 1, 2, 3, 4]}, index=pd.date_range("2020", periods=5, freq="1D")
+    )
+    df
+    df.rolling("2D", center=False).mean()
+    df.rolling("2D", center=True).mean()
+
+
+.. _window.endpoints:
+
+Rolling window endpoints
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+The inclusion of the interval endpoints in rolling window calculations can be specified with the ``closed``
+parameter:
+
+=============  ====================
+Value          Behavior
+=============  ====================
+``'right'``     close right endpoint
+``'left'``     close left endpoint
+``'both'``     close both endpoints
+``'neither'``  open endpoints
+=============  ====================
+
+For example, having the right endpoint open is useful in many problems that require that there is no contamination
+from present information back to past information. This allows the rolling window to compute statistics
+"up to that point in time", but not including that point in time.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"x": 1},
+       index=[
+           pd.Timestamp("20130101 09:00:01"),
+           pd.Timestamp("20130101 09:00:02"),
+           pd.Timestamp("20130101 09:00:03"),
+           pd.Timestamp("20130101 09:00:04"),
+           pd.Timestamp("20130101 09:00:06"),
+       ],
+   )
+
+   df["right"] = df.rolling("2s", closed="right").x.sum()  # default
+   df["both"] = df.rolling("2s", closed="both").x.sum()
+   df["left"] = df.rolling("2s", closed="left").x.sum()
+   df["neither"] = df.rolling("2s", closed="neither").x.sum()
+
+   df
+
+.. _window.custom_rolling_window:
+
+Custom window rolling
+~~~~~~~~~~~~~~~~~~~~~
+
+In addition to accepting an integer or offset as a ``window`` argument, ``rolling`` also accepts
+a ``BaseIndexer`` subclass that allows a user to define a custom method for calculating window bounds.
+The ``BaseIndexer`` subclass will need to define a ``get_window_bounds`` method that returns
+a tuple of two arrays, the first being the starting indices of the windows and second being the
+ending indices of the windows. Additionally, ``num_values``, ``min_periods``, ``center``, ``closed``
+and ``step`` will automatically be passed to ``get_window_bounds`` and the defined method must
+always accept these arguments.
+
+For example, if we have the following :class:`DataFrame`
+
+.. ipython:: python
+
+   use_expanding = [True, False, True, False, True]
+   use_expanding
+   df = pd.DataFrame({"values": range(5)})
+   df
+
+and we want to use an expanding window where ``use_expanding`` is ``True`` otherwise a window of size
+1, we can create the following ``BaseIndexer`` subclass:
+
+.. ipython:: python
+
+   from pandas.api.indexers import BaseIndexer
+
+   class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            start = np.empty(num_values, dtype=np.int64)
+            end = np.empty(num_values, dtype=np.int64)
+            for i in range(num_values):
+                if self.use_expanding[i]:
+                    start[i] = 0
+                    end[i] = i + 1
+                else:
+                    start[i] = i
+                    end[i] = i + self.window_size
+            return start, end
+
+   indexer = CustomIndexer(window_size=1, use_expanding=use_expanding)
+
+   df.rolling(indexer).sum()
+
+You can view other examples of ``BaseIndexer`` subclasses `here <https://github.com/pandas-dev/pandas/blob/main/pandas/core/indexers/objects.py>`__
+
+One subclass of note within those examples is the ``VariableOffsetWindowIndexer`` that allows
+rolling operations over a non-fixed offset like a ``BusinessDay``.
+
+.. ipython:: python
+
+   from pandas.api.indexers import VariableOffsetWindowIndexer
+
+   df = pd.DataFrame(range(10), index=pd.date_range("2020", periods=10))
+   offset = pd.offsets.BDay(1)
+   indexer = VariableOffsetWindowIndexer(index=df.index, offset=offset)
+   df
+   df.rolling(indexer).sum()
+
+For some problems knowledge of the future is available for analysis. For example, this occurs when
+each data point is a full time series read from an experiment, and the task is to extract underlying
+conditions. In these cases it can be useful to perform forward-looking rolling window computations.
+:func:`FixedForwardWindowIndexer <pandas.api.indexers.FixedForwardWindowIndexer>` class is available for this purpose.
+This :func:`BaseIndexer <pandas.api.indexers.BaseIndexer>` subclass implements a closed fixed-width
+forward-looking rolling window, and we can use it as follows:
+
+.. ipython:: python
+
+   from pandas.api.indexers import FixedForwardWindowIndexer
+   indexer = FixedForwardWindowIndexer(window_size=2)
+   df.rolling(indexer, min_periods=1).sum()
+
+We can also achieve this by using slicing, applying rolling aggregation, and then flipping the result as shown in example below:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       data=[
+           [pd.Timestamp("2018-01-01 00:00:00"), 100],
+           [pd.Timestamp("2018-01-01 00:00:01"), 101],
+           [pd.Timestamp("2018-01-01 00:00:03"), 103],
+           [pd.Timestamp("2018-01-01 00:00:04"), 111],
+       ],
+       columns=["time", "value"],
+   ).set_index("time")
+   df
+
+   reversed_df = df[::-1].rolling("2s").sum()[::-1]
+   reversed_df
+
+.. _window.rolling_apply:
+
+Rolling apply
+~~~~~~~~~~~~~
+
+The :meth:`~Rolling.apply` function takes an extra ``func`` argument and performs
+generic rolling computations. The ``func`` argument should be a single function
+that produces a single value from an ndarray input. ``raw`` specifies whether
+the windows are cast as :class:`Series` objects (``raw=False``) or ndarray objects (``raw=True``).
+
+.. ipython:: python
+
+   def mad(x):
+       return np.fabs(x - x.mean()).mean()
+
+   s = pd.Series(range(10))
+   s.rolling(window=4).apply(mad, raw=True)
+
+.. _window.numba_engine:
+
+Numba engine
+~~~~~~~~~~~~
+
+Additionally, :meth:`~Rolling.apply` can leverage `Numba <https://numba.pydata.org/>`__
+if installed as an optional dependency. The apply aggregation can be executed using Numba by specifying
+``engine='numba'`` and ``engine_kwargs`` arguments (``raw`` must also be set to ``True``).
+See :ref:`enhancing performance with Numba <enhancingperf.numba>` for general usage of the arguments and performance considerations.
+
+Numba will be applied in potentially two routines:
+
+#. If ``func`` is a standard Python function, the engine will `JIT <https://numba.readthedocs.io/en/stable/user/overview.html>`__ the passed function. ``func`` can also be a JITed function in which case the engine will not JIT the function again.
+#. The engine will JIT the for loop where the apply function is applied to each window.
+
+The ``engine_kwargs`` argument is a dictionary of keyword arguments that will be passed into the
+`numba.jit decorator <https://numba.readthedocs.io/en/stable/user/jit.html>`__.
+These keyword arguments will be applied to *both* the passed function (if a standard Python function)
+and the apply for loop over each window.
+
+.. versionadded:: 1.3.0
+
+``mean``, ``median``, ``max``, ``min``, and ``sum`` also support the ``engine`` and ``engine_kwargs`` arguments.
+
+.. _window.cov_corr:
+
+Binary window functions
+~~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`~Rolling.cov` and :meth:`~Rolling.corr` can compute moving window statistics about
+two :class:`Series` or any combination of :class:`DataFrame`/:class:`Series` or
+:class:`DataFrame`/:class:`DataFrame`. Here is the behavior in each case:
+
+* two :class:`Series`: compute the statistic for the pairing.
+* :class:`DataFrame`/:class:`Series`: compute the statistics for each column of the DataFrame
+  with the passed Series, thus returning a DataFrame.
+* :class:`DataFrame`/:class:`DataFrame`: by default compute the statistic for matching column
+  names, returning a DataFrame. If the keyword argument ``pairwise=True`` is
+  passed then computes the statistic for each pair of columns, returning a :class:`DataFrame` with a
+  :class:`MultiIndex` whose values are the dates in question (see :ref:`the next section
+  <window.corr_pairwise>`).
+
+For example:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       np.random.randn(10, 4),
+       index=pd.date_range("2020-01-01", periods=10),
+       columns=["A", "B", "C", "D"],
+   )
+   df = df.cumsum()
+
+   df2 = df[:4]
+   df2.rolling(window=2).corr(df2["B"])
+
+.. _window.corr_pairwise:
+
+Computing rolling pairwise covariances and correlations
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In financial data analysis and other fields it's common to compute covariance
+and correlation matrices for a collection of time series. Often one is also
+interested in moving-window covariance and correlation matrices. This can be
+done by passing the ``pairwise`` keyword argument, which in the case of
+:class:`DataFrame` inputs will yield a MultiIndexed :class:`DataFrame` whose ``index`` are the dates in
+question. In the case of a single DataFrame argument the ``pairwise`` argument
+can even be omitted:
+
+.. note::
+
+    Missing values are ignored and each entry is computed using the pairwise
+    complete observations.
+
+    Assuming the missing data are missing at random this results in an estimate
+    for the covariance matrix which is unbiased. However, for many applications
+    this estimate may not be acceptable because the estimated covariance matrix
+    is not guaranteed to be positive semi-definite. This could lead to
+    estimated correlations having absolute values which are greater than one,
+    and/or a non-invertible covariance matrix. See `Estimation of covariance
+    matrices <https://en.wikipedia.org/w/index.php?title=Estimation_of_covariance_matrices>`_
+    for more details.
+
+.. ipython:: python
+
+   covs = (
+       df[["B", "C", "D"]]
+       .rolling(window=4)
+       .cov(df[["A", "B", "C"]], pairwise=True)
+   )
+   covs
+
+
+.. _window.weighted:
+
+Weighted window
+---------------
+
+The ``win_type`` argument in ``.rolling`` generates a weighted windows that are commonly used in filtering
+and spectral estimation. ``win_type`` must be string that corresponds to a `scipy.signal window function
+<https://docs.scipy.org/doc/scipy/reference/signal.windows.html#module-scipy.signal.windows>`__.
+Scipy must be installed in order to use these windows, and supplementary arguments
+that the Scipy window methods take must be specified in the aggregation function.
+
+
+.. ipython:: python
+
+   s = pd.Series(range(10))
+   s.rolling(window=5).mean()
+   s.rolling(window=5, win_type="triang").mean()
+   # Supplementary Scipy arguments passed in the aggregation function
+   s.rolling(window=5, win_type="gaussian").mean(std=0.1)
+
+For all supported aggregation functions, see :ref:`api.functions_window`.
+
+.. _window.expanding:
+
+Expanding window
+----------------
+
+An expanding window yields the value of an aggregation statistic with all the data available up to that
+point in time. Since these calculations are a special case of rolling statistics,
+they are implemented in pandas such that the following two calls are equivalent:
+
+.. ipython:: python
+
+   df = pd.DataFrame(range(5))
+   df.rolling(window=len(df), min_periods=1).mean()
+   df.expanding(min_periods=1).mean()
+
+For all supported aggregation functions, see :ref:`api.functions_expanding`.
+
+
+.. _window.exponentially_weighted:
+
+Exponentially weighted window
+-----------------------------
+
+An exponentially weighted window is similar to an expanding window but with each prior point
+being exponentially weighted down relative to the current point.
+
+In general, a weighted moving average is calculated as
+
+.. math::
+
+    y_t = \frac{\sum_{i=0}^t w_i x_{t-i}}{\sum_{i=0}^t w_i},
+
+where :math:`x_t` is the input, :math:`y_t` is the result and the :math:`w_i`
+are the weights.
+
+For all supported aggregation functions, see :ref:`api.functions_ewm`.
+
+The EW functions support two variants of exponential weights.
+The default, ``adjust=True``, uses the weights :math:`w_i = (1 - \alpha)^i`
+which gives
+
+.. math::
+
+    y_t = \frac{x_t + (1 - \alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ...
+    + (1 - \alpha)^t x_{0}}{1 + (1 - \alpha) + (1 - \alpha)^2 + ...
+    + (1 - \alpha)^t}
+
+When ``adjust=False`` is specified, moving averages are calculated as
+
+.. math::
+
+    y_0 &= x_0 \\
+    y_t &= (1 - \alpha) y_{t-1} + \alpha x_t,
+
+which is equivalent to using weights
+
+.. math::
+
+    w_i = \begin{cases}
+        \alpha (1 - \alpha)^i & \text{if } i < t \\
+        (1 - \alpha)^i        & \text{if } i = t.
+    \end{cases}
+
+.. note::
+
+   These equations are sometimes written in terms of :math:`\alpha' = 1 - \alpha`, e.g.
+
+   .. math::
+
+      y_t = \alpha' y_{t-1} + (1 - \alpha') x_t.
+
+The difference between the above two variants arises because we are
+dealing with series which have finite history. Consider a series of infinite
+history, with ``adjust=True``:
+
+.. math::
+
+    y_t = \frac{x_t + (1 - \alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ...}
+    {1 + (1 - \alpha) + (1 - \alpha)^2 + ...}
+
+Noting that the denominator is a geometric series with initial term equal to 1
+and a ratio of :math:`1 - \alpha` we have
+
+.. math::
+
+    y_t &= \frac{x_t + (1 - \alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ...}
+    {\frac{1}{1 - (1 - \alpha)}}\\
+    &= [x_t + (1 - \alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ...] \alpha \\
+    &= \alpha x_t + [(1-\alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ...]\alpha \\
+    &= \alpha x_t + (1 - \alpha)[x_{t-1} + (1 - \alpha) x_{t-2} + ...]\alpha\\
+    &= \alpha x_t + (1 - \alpha) y_{t-1}
+
+which is the same expression as ``adjust=False`` above and therefore
+shows the equivalence of the two variants for infinite series.
+When ``adjust=False``, we have :math:`y_0 = x_0` and
+:math:`y_t = \alpha x_t + (1 - \alpha) y_{t-1}`.
+Therefore, there is an assumption that :math:`x_0` is not an ordinary value
+but rather an exponentially weighted moment of the infinite series up to that
+point.
+
+One must have :math:`0 < \alpha \leq 1`, and while it is possible to pass
+:math:`\alpha` directly, it's often easier to think about either the
+**span**, **center  of mass (com)** or **half-life** of an EW moment:
+
+.. math::
+
+   \alpha =
+    \begin{cases}
+        \frac{2}{s + 1},            & \text{for span}\ s \geq 1\\
+        \frac{1}{1 + c},            & \text{for center of mass}\ c \geq 0\\
+        1 - e^{\frac{\log 0.5}{h}}, & \text{for half-life}\ h > 0
+    \end{cases}
+
+One must specify precisely one of **span**, **center of mass**, **half-life**
+and **alpha** to the EW functions:
+
+* **Span** corresponds to what is commonly called an "N-day EW moving average".
+* **Center of mass** has a more physical interpretation and can be thought of
+  in terms of span: :math:`c = (s - 1) / 2`.
+* **Half-life** is the period of time for the exponential weight to reduce to
+  one half.
+* **Alpha** specifies the smoothing factor directly.
+
+You can also specify ``halflife`` in terms of a timedelta convertible unit to specify the amount of
+time it takes for an observation to decay to half its value when also specifying a sequence
+of ``times``.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    df
+    times = ["2020-01-01", "2020-01-03", "2020-01-10", "2020-01-15", "2020-01-17"]
+    df.ewm(halflife="4 days", times=pd.DatetimeIndex(times)).mean()
+
+The following formula is used to compute exponentially weighted mean with an input vector of times:
+
+.. math::
+
+    y_t = \frac{\sum_{i=0}^t 0.5^\frac{t_{t} - t_{i}}{\lambda} x_{t-i}}{\sum_{i=0}^t 0.5^\frac{t_{t} - t_{i}}{\lambda}},
+
+
+ExponentialMovingWindow also has an ``ignore_na`` argument, which determines how
+intermediate null values affect the calculation of the weights.
+When ``ignore_na=False`` (the default), weights are calculated based on absolute
+positions, so that intermediate null values affect the result.
+When ``ignore_na=True``,
+weights are calculated by ignoring intermediate null values.
+For example, assuming ``adjust=True``, if ``ignore_na=False``, the weighted
+average of ``3, NaN, 5`` would be calculated as
+
+.. math::
+
+        \frac{(1-\alpha)^2 \cdot 3 + 1 \cdot 5}{(1-\alpha)^2 + 1}.
+
+Whereas if ``ignore_na=True``, the weighted average would be calculated as
+
+.. math::
+
+        \frac{(1-\alpha) \cdot 3 + 1 \cdot 5}{(1-\alpha) + 1}.
+
+The :meth:`~Ewm.var`, :meth:`~Ewm.std`, and :meth:`~Ewm.cov` functions have a ``bias`` argument,
+specifying whether the result should contain biased or unbiased statistics.
+For example, if ``bias=True``, ``ewmvar(x)`` is calculated as
+``ewmvar(x) = ewma(x**2) - ewma(x)**2``;
+whereas if ``bias=False`` (the default), the biased variance statistics
+are scaled by debiasing factors
+
+.. math::
+
+    \frac{\left(\sum_{i=0}^t w_i\right)^2}{\left(\sum_{i=0}^t w_i\right)^2 - \sum_{i=0}^t w_i^2}.
+
+(For :math:`w_i = 1`, this reduces to the usual :math:`N / (N - 1)` factor,
+with :math:`N = t + 1`.)
+See `Weighted Sample Variance <https://en.wikipedia.org/wiki/Weighted_arithmetic_mean#Weighted_sample_variance>`__
+on Wikipedia for further details.
diff --git a/pandas/doc/source/whatsnew/index.rst b/pandas/doc/source/whatsnew/index.rst
new file mode 100644
index 0000000000000..1dd6c5fabef04
--- /dev/null
+++ b/pandas/doc/source/whatsnew/index.rst
@@ -0,0 +1,341 @@
+.. _release:
+
+{{ header }}
+
+*************
+Release notes
+*************
+
+This is the list of changes to pandas between each release. For full details,
+see the `commit logs <https://github.com/pandas-dev/pandas/commits/>`_. For install and
+upgrade instructions, see :ref:`install`.
+
+Version 3.0
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v3.0.0
+
+Version 2.3
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v2.3.0
+
+Version 2.2
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v2.2.3
+   v2.2.2
+   v2.2.1
+   v2.2.0
+
+Version 2.1
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v2.1.4
+   v2.1.3
+   v2.1.2
+   v2.1.1
+   v2.1.0
+
+Version 2.0
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v2.0.3
+   v2.0.2
+   v2.0.1
+   v2.0.0
+
+Version 1.5
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.5.3
+   v1.5.2
+   v1.5.1
+   v1.5.0
+
+Version 1.4
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.4.4
+   v1.4.3
+   v1.4.2
+   v1.4.1
+   v1.4.0
+
+Version 1.3
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.3.5
+   v1.3.4
+   v1.3.3
+   v1.3.2
+   v1.3.1
+   v1.3.0
+
+Version 1.2
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.2.5
+   v1.2.4
+   v1.2.3
+   v1.2.2
+   v1.2.1
+   v1.2.0
+
+Version 1.1
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.1.5
+   v1.1.4
+   v1.1.3
+   v1.1.2
+   v1.1.1
+   v1.1.0
+
+Version 1.0
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v1.0.5
+   v1.0.4
+   v1.0.3
+   v1.0.2
+   v1.0.1
+   v1.0.0
+
+Version 0.25
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.25.3
+   v0.25.2
+   v0.25.1
+   v0.25.0
+
+Version 0.24
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.24.2
+   v0.24.1
+   v0.24.0
+
+Version 0.23
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.23.4
+   v0.23.3
+   v0.23.2
+   v0.23.1
+   v0.23.0
+
+Version 0.22
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.22.0
+
+Version 0.21
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.21.1
+   v0.21.0
+
+Version 0.20
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.20.3
+   v0.20.2
+   v0.20.0
+
+Version 0.19
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.19.2
+   v0.19.1
+   v0.19.0
+
+Version 0.18
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.18.1
+   v0.18.0
+
+Version 0.17
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.17.1
+   v0.17.0
+
+Version 0.16
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.16.2
+   v0.16.1
+   v0.16.0
+
+Version 0.15
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.15.2
+   v0.15.1
+   v0.15.0
+
+Version 0.14
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.14.1
+   v0.14.0
+
+Version 0.13
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.13.1
+   v0.13.0
+
+Version 0.12
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.12.0
+
+Version 0.11
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.11.0
+
+Version 0.10
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.10.1
+   v0.10.0
+
+Version 0.9
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.9.1
+   v0.9.0
+
+Version 0.8
+------------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.8.1
+   v0.8.0
+
+Version 0.7
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.7.3
+   v0.7.2
+   v0.7.1
+   v0.7.0
+
+Version 0.6
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.6.1
+   v0.6.0
+
+Version 0.5
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.5.0
+
+Version 0.4
+-----------
+
+.. toctree::
+   :maxdepth: 2
+
+   v0.4.x
diff --git a/pandas/doc/source/whatsnew/v0.10.0.rst b/pandas/doc/source/whatsnew/v0.10.0.rst
new file mode 100644
index 0000000000000..905583c708905
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.10.0.rst
@@ -0,0 +1,591 @@
+.. _whatsnew_0100:
+
+Version 0.10.0 (December 17, 2012)
+----------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.9.1 and includes many new features and
+enhancements along with a large number of bug fixes. There are also a number of
+important API changes that long-time pandas users should pay close attention
+to.
+
+File parsing new features
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The delimited file parsing engine (the guts of ``read_csv`` and ``read_table``)
+has been rewritten from the ground up and now uses a fraction the amount of
+memory while parsing, while being 40% or more faster in most use cases (in some
+cases much faster).
+
+There are also many new features:
+
+- Much-improved Unicode handling via the ``encoding`` option.
+- Column filtering (``usecols``)
+- Dtype specification (``dtype`` argument)
+- Ability to specify strings to be recognized as True/False
+- Ability to yield NumPy record arrays (``as_recarray``)
+- High performance ``delim_whitespace`` option
+- Decimal format (e.g. European format) specification
+- Easier CSV dialect options: ``escapechar``, ``lineterminator``,
+  ``quotechar``, etc.
+- More robust handling of many exceptional kinds of files observed in the wild
+
+API changes
+~~~~~~~~~~~
+
+**Deprecated DataFrame BINOP TimeSeries special case behavior**
+
+The default behavior of binary operations between a DataFrame and a Series has
+always been to align on the DataFrame's columns and broadcast down the rows,
+**except** in the special case that the DataFrame contains time series. Since
+there are now method for each binary operator enabling you to specify how you
+want to broadcast, we are phasing out this special case (Zen of Python:
+*Special cases aren't special enough to break the rules*). Here's what I'm
+talking about:
+
+.. ipython:: python
+   :okwarning:
+
+   import pandas as pd
+
+   df = pd.DataFrame(np.random.randn(6, 4), index=pd.date_range("1/1/2000", periods=6))
+   df
+   # deprecated now
+   df - df[0]
+   # Change your code to
+   df.sub(df[0], axis=0)  # align on axis 0 (rows)
+
+You will get a deprecation warning in the 0.10.x series, and the deprecated
+functionality will be removed in 0.11 or later.
+
+**Altered resample default behavior**
+
+The default time series ``resample`` binning behavior of daily ``D`` and
+*higher* frequencies has been changed to ``closed='left', label='left'``. Lower
+nfrequencies are unaffected. The prior defaults were causing a great deal of
+confusion for users, especially resampling data to daily frequency (which
+labeled the aggregated group with the end of the interval: the next day).
+
+.. code-block:: ipython
+
+   In [1]: dates = pd.date_range('1/1/2000', '1/5/2000', freq='4h')
+
+   In [2]: series = pd.Series(np.arange(len(dates)), index=dates)
+
+   In [3]: series
+   Out[3]:
+   2000-01-01 00:00:00     0
+   2000-01-01 04:00:00     1
+   2000-01-01 08:00:00     2
+   2000-01-01 12:00:00     3
+   2000-01-01 16:00:00     4
+   2000-01-01 20:00:00     5
+   2000-01-02 00:00:00     6
+   2000-01-02 04:00:00     7
+   2000-01-02 08:00:00     8
+   2000-01-02 12:00:00     9
+   2000-01-02 16:00:00    10
+   2000-01-02 20:00:00    11
+   2000-01-03 00:00:00    12
+   2000-01-03 04:00:00    13
+   2000-01-03 08:00:00    14
+   2000-01-03 12:00:00    15
+   2000-01-03 16:00:00    16
+   2000-01-03 20:00:00    17
+   2000-01-04 00:00:00    18
+   2000-01-04 04:00:00    19
+   2000-01-04 08:00:00    20
+   2000-01-04 12:00:00    21
+   2000-01-04 16:00:00    22
+   2000-01-04 20:00:00    23
+   2000-01-05 00:00:00    24
+   Freq: 4H, dtype: int64
+
+   In [4]: series.resample('D', how='sum')
+   Out[4]:
+   2000-01-01     15
+   2000-01-02     51
+   2000-01-03     87
+   2000-01-04    123
+   2000-01-05     24
+   Freq: D, dtype: int64
+
+   In [5]: # old behavior
+   In [6]: series.resample('D', how='sum', closed='right', label='right')
+   Out[6]:
+   2000-01-01      0
+   2000-01-02     21
+   2000-01-03     57
+   2000-01-04     93
+   2000-01-05    129
+   Freq: D, dtype: int64
+
+- Infinity and negative infinity are no longer treated as NA by ``isnull`` and
+  ``notnull``. That they ever were was a relic of early pandas. This behavior
+  can be re-enabled globally by the ``mode.use_inf_as_null`` option:
+
+.. code-block:: ipython
+
+    In [6]: s = pd.Series([1.5, np.inf, 3.4, -np.inf])
+
+    In [7]: pd.isnull(s)
+    Out[7]:
+    0    False
+    1    False
+    2    False
+    3    False
+    Length: 4, dtype: bool
+
+    In [8]: s.fillna(0)
+    Out[8]:
+    0    1.500000
+    1         inf
+    2    3.400000
+    3        -inf
+    Length: 4, dtype: float64
+
+    In [9]: pd.set_option('use_inf_as_null', True)
+
+    In [10]: pd.isnull(s)
+    Out[10]:
+    0    False
+    1     True
+    2    False
+    3     True
+    Length: 4, dtype: bool
+
+    In [11]: s.fillna(0)
+    Out[11]:
+    0    1.5
+    1    0.0
+    2    3.4
+    3    0.0
+    Length: 4, dtype: float64
+
+    In [12]: pd.reset_option('use_inf_as_null')
+
+- Methods with the ``inplace`` option now all return ``None`` instead of the
+  calling object. E.g. code written like ``df = df.fillna(0, inplace=True)``
+  may stop working. To fix, simply delete the unnecessary variable assignment.
+
+- ``pandas.merge`` no longer sorts the group keys (``sort=False``) by
+  default. This was done for performance reasons: the group-key sorting is
+  often one of the more expensive parts of the computation and is often
+  unnecessary.
+
+- The default column names for a file with no header have been changed to the
+  integers ``0`` through ``N - 1``. This is to create consistency with the
+  DataFrame constructor with no columns specified. The v0.9.0 behavior (names
+  ``X0``, ``X1``, ...) can be reproduced by specifying ``prefix='X'``:
+
+.. code-block:: ipython
+
+    In [6]: import io
+
+    In [7]: data = """
+      ...: a,b,c
+      ...: 1,Yes,2
+      ...: 3,No,4
+      ...: """
+      ...:
+
+    In [8]: print(data)
+
+        a,b,c
+        1,Yes,2
+        3,No,4
+
+    In [9]: pd.read_csv(io.StringIO(data), header=None)
+    Out[9]:
+           0    1  2
+    0      a    b  c
+    1      1  Yes  2
+    2      3   No  4
+
+    In [10]: pd.read_csv(io.StringIO(data), header=None, prefix="X")
+    Out[10]:
+            X0   X1 X2
+    0       a    b  c
+    1       1  Yes  2
+    2       3   No  4
+
+- Values like ``'Yes'`` and ``'No'`` are not interpreted as boolean by default,
+  though this can be controlled by new ``true_values`` and ``false_values``
+  arguments:
+
+.. code-block:: ipython
+
+    In [4]: print(data)
+
+        a,b,c
+        1,Yes,2
+        3,No,4
+
+    In [5]: pd.read_csv(io.StringIO(data))
+    Out[5]:
+           a    b  c
+    0      1  Yes  2
+    1      3   No  4
+
+    In [6]: pd.read_csv(io.StringIO(data), true_values=["Yes"], false_values=["No"])
+    Out[6]:
+           a      b  c
+    0      1   True  2
+    1      3  False  4
+
+- The file parsers will not recognize non-string values arising from a
+  converter function as NA if passed in the ``na_values`` argument. It's better
+  to do post-processing using the ``replace`` function instead.
+
+- Calling ``fillna`` on Series or DataFrame with no arguments is no longer
+  valid code. You must either specify a fill value or an interpolation method:
+
+.. code-block:: ipython
+
+    In [6]: s = pd.Series([np.nan, 1.0, 2.0, np.nan, 4])
+
+    In [7]: s
+    Out[7]:
+    0      NaN
+    1      1.0
+    2      2.0
+    3      NaN
+    4      4.0
+    dtype: float64
+
+    In [8]: s.fillna(0)
+    Out[8]:
+    0      0.0
+    1      1.0
+    2      2.0
+    3      0.0
+    4      4.0
+    dtype: float64
+
+    In [9]: s.fillna(method="pad")
+    Out[9]:
+    0      NaN
+    1      1.0
+    2      2.0
+    3      2.0
+    4      4.0
+    dtype: float64
+
+Convenience methods ``ffill`` and  ``bfill`` have been added:
+
+.. ipython:: python
+
+   s = pd.Series([np.nan, 1.0, 2.0, np.nan, 4])
+   s.ffill()
+
+
+- ``Series.apply`` will now operate on a returned value from the applied
+  function, that is itself a series, and possibly upcast the result to a
+  DataFrame
+
+  .. ipython:: python
+
+      def f(x):
+          return pd.Series([x, x ** 2], index=["x", "x^2"])
+
+
+      s = pd.Series(np.random.rand(5))
+      s
+      s.apply(f)
+
+- New API functions for working with pandas options (:issue:`2097`):
+
+  - ``get_option`` / ``set_option`` - get/set the value of an option. Partial
+    names are accepted.  - ``reset_option`` - reset one or more options to
+    their default value. Partial names are accepted.  - ``describe_option`` -
+    print a description of one or more options. When called with no
+    arguments. print all registered options.
+
+  Note: ``set_printoptions``/ ``reset_printoptions`` are now deprecated (but
+  functioning), the print options now live under "display.XYZ". For example:
+
+  .. ipython:: python
+
+     pd.get_option("display.max_rows")
+
+- to_string() methods now always return unicode strings  (:issue:`2224`).
+
+New features
+~~~~~~~~~~~~
+
+Wide DataFrame printing
+~~~~~~~~~~~~~~~~~~~~~~~
+
+Instead of printing the summary information, pandas now splits the string
+representation across multiple rows by default:
+
+.. ipython:: python
+
+   wide_frame = pd.DataFrame(np.random.randn(5, 16))
+
+   wide_frame
+
+The old behavior of printing out summary information can be achieved via the
+'expand_frame_repr' print option:
+
+.. ipython:: python
+
+   pd.set_option("expand_frame_repr", False)
+
+   wide_frame
+
+.. ipython:: python
+   :suppress:
+
+   pd.reset_option("expand_frame_repr")
+
+The width of each line can be changed via 'line_width' (80 by default):
+
+.. code-block:: python
+
+   pd.set_option("line_width", 40)
+
+   wide_frame
+
+
+Updated PyTables support
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+:ref:`Docs <io.hdf5>` for PyTables ``Table`` format & several enhancements to the api. Here is a taste of what to expect.
+
+.. code-block:: ipython
+
+    In [41]: store = pd.HDFStore('store.h5')
+
+    In [42]: df = pd.DataFrame(np.random.randn(8, 3),
+       ....:                   index=pd.date_range('1/1/2000', periods=8),
+       ....:                   columns=['A', 'B', 'C'])
+
+    In [43]: df
+    Out[43]:
+                       A         B         C
+    2000-01-01 -2.036047  0.000830 -0.955697
+    2000-01-02 -0.898872 -0.725411  0.059904
+    2000-01-03 -0.449644  1.082900 -1.221265
+    2000-01-04  0.361078  1.330704  0.855932
+    2000-01-05 -1.216718  1.488887  0.018993
+    2000-01-06 -0.877046  0.045976  0.437274
+    2000-01-07 -0.567182 -0.888657 -0.556383
+    2000-01-08  0.655457  1.117949 -2.782376
+
+    [8 rows x 3 columns]
+
+    # appending data frames
+    In [44]: df1 = df[0:4]
+
+    In [45]: df2 = df[4:]
+
+    In [46]: store.append('df', df1)
+
+    In [47]: store.append('df', df2)
+
+    In [48]: store
+    Out[48]:
+    <class 'pandas.io.pytables.HDFStore'>
+    File path: store.h5
+    /df            frame_table  (typ->appendable,nrows->8,ncols->3,indexers->[index])
+
+    # selecting the entire store
+    In [49]: store.select('df')
+    Out[49]:
+                       A         B         C
+    2000-01-01 -2.036047  0.000830 -0.955697
+    2000-01-02 -0.898872 -0.725411  0.059904
+    2000-01-03 -0.449644  1.082900 -1.221265
+    2000-01-04  0.361078  1.330704  0.855932
+    2000-01-05 -1.216718  1.488887  0.018993
+    2000-01-06 -0.877046  0.045976  0.437274
+    2000-01-07 -0.567182 -0.888657 -0.556383
+    2000-01-08  0.655457  1.117949 -2.782376
+
+    [8 rows x 3 columns]
+
+.. code-block:: ipython
+
+    In [50]: wp = pd.Panel(np.random.randn(2, 5, 4), items=['Item1', 'Item2'],
+       ....:               major_axis=pd.date_range('1/1/2000', periods=5),
+       ....:               minor_axis=['A', 'B', 'C', 'D'])
+
+    In [51]: wp
+    Out[51]:
+    <class 'pandas.core.panel.Panel'>
+    Dimensions: 2 (items) x 5 (major_axis) x 4 (minor_axis)
+    Items axis: Item1 to Item2
+    Major_axis axis: 2000-01-01 00:00:00 to 2000-01-05 00:00:00
+    Minor_axis axis: A to D
+
+    # storing a panel
+    In [52]: store.append('wp', wp)
+
+    # selecting via A QUERY
+    In [53]: store.select('wp', [pd.Term('major_axis>20000102'),
+       ....:                     pd.Term('minor_axis', '=', ['A', 'B'])])
+       ....:
+    Out[53]:
+    <class 'pandas.core.panel.Panel'>
+    Dimensions: 2 (items) x 3 (major_axis) x 2 (minor_axis)
+    Items axis: Item1 to Item2
+    Major_axis axis: 2000-01-03 00:00:00 to 2000-01-05 00:00:00
+    Minor_axis axis: A to B
+
+    # removing data from tables
+    In [54]: store.remove('wp', pd.Term('major_axis>20000103'))
+    Out[54]: 8
+
+    In [55]: store.select('wp')
+    Out[55]:
+    <class 'pandas.core.panel.Panel'>
+    Dimensions: 2 (items) x 3 (major_axis) x 4 (minor_axis)
+    Items axis: Item1 to Item2
+    Major_axis axis: 2000-01-01 00:00:00 to 2000-01-03 00:00:00
+    Minor_axis axis: A to D
+
+    # deleting a store
+    In [56]: del store['df']
+
+    In [57]: store
+    Out[57]:
+    <class 'pandas.io.pytables.HDFStore'>
+    File path: store.h5
+    /wp            wide_table   (typ->appendable,nrows->12,ncols->2,indexers->[major_axis,minor_axis])
+
+
+**Enhancements**
+
+- added ability to hierarchical keys
+
+   .. code-block:: ipython
+
+        In [58]: store.put('foo/bar/bah', df)
+
+        In [59]: store.append('food/orange', df)
+
+        In [60]: store.append('food/apple', df)
+
+        In [61]: store
+        Out[61]:
+        <class 'pandas.io.pytables.HDFStore'>
+        File path: store.h5
+        /foo/bar/bah            frame        (shape->[8,3])
+        /food/apple             frame_table  (typ->appendable,nrows->8,ncols->3,indexers->[index])
+        /food/orange            frame_table  (typ->appendable,nrows->8,ncols->3,indexers->[index])
+        /wp                     wide_table   (typ->appendable,nrows->12,ncols->2,indexers->[major_axis,minor_axis])
+
+        # remove all nodes under this level
+        In [62]: store.remove('food')
+
+        In [63]: store
+        Out[63]:
+        <class 'pandas.io.pytables.HDFStore'>
+        File path: store.h5
+        /foo/bar/bah            frame        (shape->[8,3])
+        /wp                     wide_table   (typ->appendable,nrows->12,ncols->2,indexers->[major_axis,minor_axis])
+
+- added mixed-dtype support!
+
+   .. code-block:: ipython
+
+        In [64]: df['string'] = 'string'
+
+        In [65]: df['int'] = 1
+
+        In [66]: store.append('df', df)
+
+        In [67]: df1 = store.select('df')
+
+        In [68]: df1
+        Out[68]:
+                           A         B         C  string  int
+        2000-01-01 -2.036047  0.000830 -0.955697  string    1
+        2000-01-02 -0.898872 -0.725411  0.059904  string    1
+        2000-01-03 -0.449644  1.082900 -1.221265  string    1
+        2000-01-04  0.361078  1.330704  0.855932  string    1
+        2000-01-05 -1.216718  1.488887  0.018993  string    1
+        2000-01-06 -0.877046  0.045976  0.437274  string    1
+        2000-01-07 -0.567182 -0.888657 -0.556383  string    1
+        2000-01-08  0.655457  1.117949 -2.782376  string    1
+
+        [8 rows x 5 columns]
+
+        In [69]: df1.get_dtype_counts()
+        Out[69]:
+        float64    3
+        int64      1
+        object     1
+        dtype: int64
+
+- performance improvements on table writing
+- support for arbitrarily indexed dimensions
+- ``SparseSeries`` now has a ``density`` property (:issue:`2384`)
+- enable ``Series.str.strip/lstrip/rstrip`` methods to take an input argument
+  to strip arbitrary characters (:issue:`2411`)
+- implement ``value_vars`` in ``melt`` to limit values to certain columns
+  and add ``melt`` to pandas namespace (:issue:`2412`)
+
+**Bug Fixes**
+
+- added ``Term`` method of specifying where conditions (:issue:`1996`).
+- ``del store['df']`` now call ``store.remove('df')`` for store deletion
+- deleting of consecutive rows is much faster than before
+- ``min_itemsize`` parameter can be specified in table creation to force a
+  minimum size for indexing columns (the previous implementation would set the
+  column size based on the first append)
+- indexing support via ``create_table_index`` (requires PyTables >= 2.3)
+  (:issue:`698`).
+- appending on a store would fail if the table was not first created via ``put``
+- fixed issue with missing attributes after loading a pickled dataframe (GH2431)
+- minor change to select and remove: require a table ONLY if where is also
+  provided (and not None)
+
+**Compatibility**
+
+0.10 of ``HDFStore`` is backwards compatible for reading tables created in a prior version of pandas,
+however, query terms using the prior (undocumented) methodology are unsupported. You must read in the entire
+file and write it out using the new format to take advantage of the updates.
+
+N dimensional panels (experimental)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Adding experimental support for Panel4D and factory functions to create n-dimensional named panels.
+Here is a taste of what to expect.
+
+.. code-block:: ipython
+
+  In [58]: p4d = Panel4D(np.random.randn(2, 2, 5, 4),
+    ....:       labels=['Label1','Label2'],
+    ....:       items=['Item1', 'Item2'],
+    ....:       major_axis=date_range('1/1/2000', periods=5),
+    ....:       minor_axis=['A', 'B', 'C', 'D'])
+    ....:
+
+  In [59]: p4d
+  Out[59]:
+  <class 'pandas.core.panelnd.Panel4D'>
+  Dimensions: 2 (labels) x 2 (items) x 5 (major_axis) x 4 (minor_axis)
+  Labels axis: Label1 to Label2
+  Items axis: Item1 to Item2
+  Major_axis axis: 2000-01-01 00:00:00 to 2000-01-05 00:00:00
+  Minor_axis axis: A to D
+
+
+
+
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+.. _whatsnew_0.10.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.9.0..v0.10.0
diff --git a/pandas/doc/source/whatsnew/v0.10.1.rst b/pandas/doc/source/whatsnew/v0.10.1.rst
new file mode 100644
index 0000000000000..611ac2021fcec
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.10.1.rst
@@ -0,0 +1,268 @@
+.. _whatsnew_0101:
+
+Version 0.10.1 (January 22, 2013)
+---------------------------------
+
+{{ header }}
+
+
+This is a minor release from 0.10.0 and includes new features, enhancements,
+and bug fixes. In particular, there is substantial new HDFStore functionality
+contributed by Jeff Reback.
+
+An undesired API breakage with functions taking the ``inplace`` option has been
+reverted and deprecation warnings added.
+
+API changes
+~~~~~~~~~~~
+
+- Functions taking an ``inplace`` option return the calling object as before. A
+  deprecation message has been added
+- Groupby aggregations Max/Min no longer exclude non-numeric data (:issue:`2700`)
+- Resampling an empty DataFrame now returns an empty DataFrame instead of
+  raising an exception (:issue:`2640`)
+- The file reader will now raise an exception when NA values are found in an
+  explicitly specified integer column instead of converting the column to float
+  (:issue:`2631`)
+- DatetimeIndex.unique now returns a DatetimeIndex with the same name and
+- timezone instead of an array (:issue:`2563`)
+
+New features
+~~~~~~~~~~~~
+
+- MySQL support for database (contribution from Dan Allan)
+
+HDFStore
+~~~~~~~~
+
+You may need to upgrade your existing data files. Please visit the
+**compatibility** section in the main docs.
+
+
+.. ipython:: python
+   :suppress:
+   :okexcept:
+
+   import os
+
+   os.remove("store.h5")
+
+You can designate (and index) certain columns that you want to be able to
+perform queries on a table, by passing a list to ``data_columns``
+
+.. ipython:: python
+
+   store = pd.HDFStore("store.h5")
+   df = pd.DataFrame(
+       np.random.randn(8, 3),
+       index=pd.date_range("1/1/2000", periods=8),
+       columns=["A", "B", "C"],
+   )
+   df["string"] = "foo"
+   df.loc[df.index[4:6], "string"] = np.nan
+   df.loc[df.index[7:9], "string"] = "bar"
+   df["string2"] = "cool"
+   df
+
+   # on-disk operations
+   store.append("df", df, data_columns=["B", "C", "string", "string2"])
+   store.select("df", "B>0 and string=='foo'")
+
+   # this is in-memory version of this type of selection
+   df[(df.B > 0) & (df.string == "foo")]
+
+Retrieving unique values in an indexable or data column.
+
+.. code-block:: python
+
+   # note that this is deprecated as of 0.14.0
+   # can be replicated by: store.select_column('df','index').unique()
+   store.unique("df", "index")
+   store.unique("df", "string")
+
+You can now store ``datetime64`` in data columns
+
+.. ipython:: python
+
+    df_mixed = df.copy()
+    df_mixed["datetime64"] = pd.Timestamp("20010102")
+    df_mixed.loc[df_mixed.index[3:4], ["A", "B"]] = np.nan
+
+    store.append("df_mixed", df_mixed)
+    df_mixed1 = store.select("df_mixed")
+    df_mixed1
+    df_mixed1.dtypes.value_counts()
+
+You can pass ``columns`` keyword to select to filter a list of the return
+columns, this is equivalent to passing a
+``Term('columns',list_of_columns_to_filter)``
+
+.. ipython:: python
+
+   store.select("df", columns=["A", "B"])
+
+``HDFStore`` now serializes MultiIndex dataframes when appending tables.
+
+.. code-block:: ipython
+
+    In [19]: index = pd.MultiIndex(levels=[['foo', 'bar', 'baz', 'qux'],
+       ....:                               ['one', 'two', 'three']],
+       ....:                       labels=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3],
+       ....:                               [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+       ....:                       names=['foo', 'bar'])
+       ....:
+
+    In [20]: df = pd.DataFrame(np.random.randn(10, 3), index=index,
+       ....:                   columns=['A', 'B', 'C'])
+       ....:
+
+    In [21]: df
+    Out[21]:
+                      A         B         C
+    foo bar
+    foo one   -0.116619  0.295575 -1.047704
+        two    1.640556  1.905836  2.772115
+        three  0.088787 -1.144197 -0.633372
+    bar one    0.925372 -0.006438 -0.820408
+        two   -0.600874 -1.039266  0.824758
+    baz two   -0.824095 -0.337730 -0.927764
+        three -0.840123  0.248505 -0.109250
+    qux one    0.431977 -0.460710  0.336505
+        two   -3.207595 -1.535854  0.409769
+        three -0.673145 -0.741113 -0.110891
+
+    In [22]: store.append('mi', df)
+
+    In [23]: store.select('mi')
+    Out[23]:
+                      A         B         C
+    foo bar
+    foo one   -0.116619  0.295575 -1.047704
+        two    1.640556  1.905836  2.772115
+        three  0.088787 -1.144197 -0.633372
+    bar one    0.925372 -0.006438 -0.820408
+        two   -0.600874 -1.039266  0.824758
+    baz two   -0.824095 -0.337730 -0.927764
+        three -0.840123  0.248505 -0.109250
+    qux one    0.431977 -0.460710  0.336505
+        two   -3.207595 -1.535854  0.409769
+        three -0.673145 -0.741113 -0.110891
+
+    # the levels are automatically included as data columns
+    In [24]: store.select('mi', "foo='bar'")
+    Out[24]:
+                    A         B         C
+    foo bar
+    bar one  0.925372 -0.006438 -0.820408
+        two -0.600874 -1.039266  0.824758
+
+Multi-table creation via ``append_to_multiple`` and selection via
+``select_as_multiple`` can create/select from multiple tables and return a
+combined result, by using ``where`` on a selector table.
+
+.. ipython:: python
+
+   df_mt = pd.DataFrame(
+       np.random.randn(8, 6),
+       index=pd.date_range("1/1/2000", periods=8),
+       columns=["A", "B", "C", "D", "E", "F"],
+   )
+   df_mt["foo"] = "bar"
+
+   # you can also create the tables individually
+   store.append_to_multiple(
+       {"df1_mt": ["A", "B"], "df2_mt": None}, df_mt, selector="df1_mt"
+   )
+   store
+
+   # individual tables were created
+   store.select("df1_mt")
+   store.select("df2_mt")
+
+   # as a multiple
+   store.select_as_multiple(
+       ["df1_mt", "df2_mt"], where=["A>0", "B>0"], selector="df1_mt"
+   )
+
+.. ipython:: python
+   :suppress:
+
+   store.close()
+   os.remove("store.h5")
+
+**Enhancements**
+
+- ``HDFStore`` now can read native PyTables table format tables
+
+- You can pass ``nan_rep = 'my_nan_rep'`` to append, to change the default nan
+  representation on disk (which converts to/from ``np.nan``), this defaults to
+  ``nan``.
+
+- You can pass ``index`` to ``append``. This defaults to ``True``. This will
+  automagically create indices on the *indexables* and *data columns* of the
+  table
+
+- You can pass ``chunksize=an integer`` to ``append``, to change the writing
+  chunksize (default is 50000). This will significantly lower your memory usage
+  on writing.
+
+- You can pass ``expectedrows=an integer`` to the first ``append``, to set the
+  TOTAL number of expected rows that ``PyTables`` will expected. This will
+  optimize read/write performance.
+
+- ``Select`` now supports passing ``start`` and ``stop`` to provide selection
+  space limiting in selection.
+
+- Greatly improved ISO8601 (e.g., yyyy-mm-dd) date parsing for file parsers (:issue:`2698`)
+- Allow ``DataFrame.merge`` to handle combinatorial sizes too large for 64-bit
+  integer (:issue:`2690`)
+- Series now has unary negation (-series) and inversion (~series) operators (:issue:`2686`)
+- DataFrame.plot now includes a ``logx`` parameter to change the x-axis to log scale (:issue:`2327`)
+- Series arithmetic operators can now handle constant and ndarray input (:issue:`2574`)
+- ExcelFile now takes a ``kind`` argument to specify the file type (:issue:`2613`)
+- A faster implementation for Series.str methods (:issue:`2602`)
+
+**Bug Fixes**
+
+- ``HDFStore`` tables can now store ``float32`` types correctly (cannot be
+  mixed with ``float64`` however)
+- Fixed Google Analytics prefix when specifying request segment (:issue:`2713`).
+- Function to reset Google Analytics token store so users can recover from
+  improperly setup client secrets (:issue:`2687`).
+- Fixed groupby bug resulting in segfault when passing in MultiIndex (:issue:`2706`)
+- Fixed bug where passing a Series with datetime64 values into ``to_datetime``
+  results in bogus output values (:issue:`2699`)
+- Fixed bug in ``pattern in HDFStore`` expressions when pattern is not a valid
+  regex (:issue:`2694`)
+- Fixed performance issues while aggregating boolean data (:issue:`2692`)
+- When given a boolean mask key and a Series of new values, Series __setitem__
+  will now align the incoming values with the original Series (:issue:`2686`)
+- Fixed MemoryError caused by performing counting sort on sorting MultiIndex
+  levels with a very large number of combinatorial values (:issue:`2684`)
+- Fixed bug that causes plotting to fail when the index is a DatetimeIndex with
+  a fixed-offset timezone (:issue:`2683`)
+- Corrected business day subtraction logic when the offset is more than 5 bdays
+  and the starting date is on a weekend (:issue:`2680`)
+- Fixed C file parser behavior when the file has more columns than data
+  (:issue:`2668`)
+- Fixed file reader bug that misaligned columns with data in the presence of an
+  implicit column and a specified ``usecols`` value
+- DataFrames with numerical or datetime indices are now sorted prior to
+  plotting (:issue:`2609`)
+- Fixed DataFrame.from_records error when passed columns, index, but empty
+  records (:issue:`2633`)
+- Several bug fixed for Series operations when dtype is datetime64 (:issue:`2689`,
+  :issue:`2629`, :issue:`2626`)
+
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+.. _whatsnew_0.10.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.10.0..v0.10.1
diff --git a/pandas/doc/source/whatsnew/v0.11.0.rst b/pandas/doc/source/whatsnew/v0.11.0.rst
new file mode 100644
index 0000000000000..dcb0d3229aa5d
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.11.0.rst
@@ -0,0 +1,476 @@
+.. _whatsnew_0110:
+
+Version 0.11.0 (April 22, 2013)
+-------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.10.1 and includes many new features and
+enhancements along with a large number of bug fixes. The methods of Selecting
+Data have had quite a number of additions, and Dtype support is now full-fledged.
+There are also a number of important API changes that long-time pandas users should
+pay close attention to.
+
+There is a new section in the documentation, :ref:`10 Minutes to pandas <10min>`,
+primarily geared to new users.
+
+There is a new section in the documentation, :ref:`Cookbook <cookbook>`, a collection
+of useful recipes in pandas (and that we want contributions!).
+
+There are several libraries that are now :ref:`Recommended Dependencies <install.recommended_dependencies>`
+
+Selection choices
+~~~~~~~~~~~~~~~~~
+
+Starting in 0.11.0, object selection has had a number of user-requested additions in
+order to support more explicit location based indexing. pandas now supports
+three types of multi-axis indexing.
+
+- ``.loc`` is strictly label based, will raise ``KeyError`` when the items are not found, allowed inputs are:
+
+  - A single label, e.g. ``5`` or ``'a'``, (note that ``5`` is interpreted as a *label* of the index. This use is **not** an integer position along the index)
+  - A list or array of labels ``['a', 'b', 'c']``
+  - A slice object with labels ``'a':'f'``, (note that contrary to usual python slices, **both** the start and the stop are included!)
+  - A boolean array
+
+  See more at :ref:`Selection by Label <indexing.label>`
+
+- ``.iloc`` is strictly integer position based (from ``0`` to ``length-1`` of the axis), will raise ``IndexError`` when the requested indices are out of bounds. Allowed inputs are:
+
+  - An integer e.g. ``5``
+  - A list or array of integers ``[4, 3, 0]``
+  - A slice object with ints ``1:7``
+  - A boolean array
+
+  See more at :ref:`Selection by Position <indexing.integer>`
+
+- ``.ix`` supports mixed integer and label based access. It is primarily label based, but will fallback to integer positional access. ``.ix`` is the most general and will support
+  any of the inputs to ``.loc`` and ``.iloc``, as well as support for floating point label schemes. ``.ix`` is especially useful when dealing with mixed positional and label
+  based hierarchical indexes.
+
+  As using integer slices with ``.ix`` have different behavior depending on whether the slice
+  is interpreted as position based or label based, it's usually better to be
+  explicit and use ``.iloc`` or ``.loc``.
+
+  See more at :ref:`Advanced Indexing <advanced>` and :ref:`Advanced Hierarchical <advanced.advanced_hierarchical>`.
+
+
+Selection deprecations
+~~~~~~~~~~~~~~~~~~~~~~
+
+Starting in version 0.11.0, these methods *may* be deprecated in future versions.
+
+- ``irow``
+- ``icol``
+- ``iget_value``
+
+See the section :ref:`Selection by Position <indexing.integer>` for substitutes.
+
+Dtypes
+~~~~~~
+
+Numeric dtypes will propagate and can coexist in DataFrames. If a dtype is passed (either directly via the ``dtype`` keyword, a passed ``ndarray``, or a passed ``Series``, then it will be preserved in DataFrame operations. Furthermore, different numeric dtypes will **NOT** be combined. The following example will give you a taste.
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.random.randn(8, 1), columns=['A'], dtype='float32')
+   df1
+   df1.dtypes
+   df2 = pd.DataFrame({'A': pd.Series(np.random.randn(8), dtype='float16'),
+                       'B': pd.Series(np.random.randn(8)),
+                       'C': pd.Series(range(8), dtype='uint8')})
+   df2
+   df2.dtypes
+
+   # here you get some upcasting
+   df3 = df1.reindex_like(df2).fillna(value=0.0) + df2
+   df3
+   df3.dtypes
+
+Dtype conversion
+~~~~~~~~~~~~~~~~
+
+This is lower-common-denominator upcasting, meaning you get the dtype which can accommodate all of the types
+
+.. ipython:: python
+
+   df3.values.dtype
+
+Conversion
+
+.. ipython:: python
+
+   df3.astype('float32').dtypes
+
+Mixed conversion
+
+.. code-block:: ipython
+
+    In [12]: df3['D'] = '1.'
+
+    In [13]: df3['E'] = '1'
+
+    In [14]: df3.convert_objects(convert_numeric=True).dtypes
+    Out[14]:
+    A    float32
+    B    float64
+    C    float64
+    D    float64
+    E      int64
+    dtype: object
+
+    # same, but specific dtype conversion
+    In [15]: df3['D'] = df3['D'].astype('float16')
+
+    In [16]: df3['E'] = df3['E'].astype('int32')
+
+    In [17]: df3.dtypes
+    Out[17]:
+    A    float32
+    B    float64
+    C    float64
+    D    float16
+    E      int32
+    dtype: object
+
+Forcing date coercion (and setting ``NaT`` when not datelike)
+
+.. code-block:: ipython
+
+    In [18]: import datetime
+
+    In [19]: s = pd.Series([datetime.datetime(2001, 1, 1, 0, 0), 'foo', 1.0, 1,
+       ....:                pd.Timestamp('20010104'), '20010105'], dtype='O')
+       ....:
+
+    In [20]: s.convert_objects(convert_dates='coerce')
+    Out[20]:
+    0   2001-01-01
+    1          NaT
+    2          NaT
+    3          NaT
+    4   2001-01-04
+    5   2001-01-05
+    dtype: datetime64[ns]
+
+Dtype gotchas
+~~~~~~~~~~~~~
+
+**Platform gotchas**
+
+Starting in 0.11.0, construction of DataFrame/Series will use default dtypes of ``int64`` and ``float64``,
+*regardless of platform*. This is not an apparent change from earlier versions of pandas. If you specify
+dtypes, they *WILL* be respected, however (:issue:`2837`)
+
+The following will all result in ``int64`` dtypes
+
+.. code-block:: ipython
+
+    In [21]: pd.DataFrame([1, 2], columns=['a']).dtypes
+    Out[21]:
+    a    int64
+    dtype: object
+
+    In [22]: pd.DataFrame({'a': [1, 2]}).dtypes
+    Out[22]:
+    a    int64
+    dtype: object
+
+    In [23]: pd.DataFrame({'a': 1}, index=range(2)).dtypes
+    Out[23]:
+    a    int64
+    dtype: object
+
+Keep in mind that ``DataFrame(np.array([1,2]))`` **WILL** result in ``int32`` on 32-bit platforms!
+
+
+**Upcasting gotchas**
+
+Performing indexing operations on integer type data can easily upcast the data.
+The dtype of the input data will be preserved in cases where ``nans`` are not introduced.
+
+.. code-block:: ipython
+
+    In [24]: dfi = df3.astype('int32')
+
+    In [25]: dfi['D'] = dfi['D'].astype('int64')
+
+    In [26]: dfi
+    Out[26]:
+      A  B  C  D  E
+    0  0  0  0  1  1
+    1 -2  0  1  1  1
+    2 -2  0  2  1  1
+    3  0 -1  3  1  1
+    4  1  0  4  1  1
+    5  0  0  5  1  1
+    6  0 -1  6  1  1
+    7  0  0  7  1  1
+
+    In [27]: dfi.dtypes
+    Out[27]:
+    A    int32
+    B    int32
+    C    int32
+    D    int64
+    E    int32
+    dtype: object
+
+    In [28]: casted = dfi[dfi > 0]
+
+    In [29]: casted
+    Out[29]:
+        A   B    C  D  E
+    0  NaN NaN  NaN  1  1
+    1  NaN NaN  1.0  1  1
+    2  NaN NaN  2.0  1  1
+    3  NaN NaN  3.0  1  1
+    4  1.0 NaN  4.0  1  1
+    5  NaN NaN  5.0  1  1
+    6  NaN NaN  6.0  1  1
+    7  NaN NaN  7.0  1  1
+
+    In [30]: casted.dtypes
+    Out[30]:
+    A    float64
+    B    float64
+    C    float64
+    D      int64
+    E      int32
+    dtype: object
+
+While float dtypes are unchanged.
+
+.. code-block:: ipython
+
+    In [31]: df4 = df3.copy()
+
+    In [32]: df4['A'] = df4['A'].astype('float32')
+
+    In [33]: df4.dtypes
+    Out[33]:
+    A    float32
+    B    float64
+    C    float64
+    D    float16
+    E      int32
+    dtype: object
+
+    In [34]: casted = df4[df4 > 0]
+
+    In [35]: casted
+    Out[35]:
+              A         B    C    D  E
+    0       NaN       NaN  NaN  1.0  1
+    1       NaN  0.567020  1.0  1.0  1
+    2       NaN  0.276232  2.0  1.0  1
+    3       NaN       NaN  3.0  1.0  1
+    4  1.933792       NaN  4.0  1.0  1
+    5       NaN  0.113648  5.0  1.0  1
+    6       NaN       NaN  6.0  1.0  1
+    7       NaN  0.524988  7.0  1.0  1
+
+    In [36]: casted.dtypes
+    Out[36]:
+    A    float32
+    B    float64
+    C    float64
+    D    float16
+    E      int32
+    dtype: object
+
+Datetimes conversion
+~~~~~~~~~~~~~~~~~~~~
+
+Datetime64[ns] columns in a DataFrame (or a Series) allow the use of ``np.nan`` to indicate a nan value,
+in addition to the traditional ``NaT``, or not-a-time. This allows convenient nan setting in a generic way.
+Furthermore ``datetime64[ns]`` columns are created by default, when passed datetimelike objects (*this change was introduced in 0.10.1*)
+(:issue:`2809`, :issue:`2810`)
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(6, 2), pd.date_range('20010102', periods=6),
+                     columns=['A', ' B'])
+   df['timestamp'] = pd.Timestamp('20010103')
+   df
+
+   # datetime64[ns] out of the box
+   df.dtypes.value_counts()
+
+   # use the traditional nan, which is mapped to NaT internally
+   df.loc[df.index[2:4], ['A', 'timestamp']] = np.nan
+   df
+
+Astype conversion on ``datetime64[ns]`` to ``object``, implicitly converts ``NaT`` to ``np.nan``
+
+.. ipython:: python
+
+   import datetime
+   s = pd.Series([datetime.datetime(2001, 1, 2, 0, 0) for i in range(3)])
+   s.dtype
+   s[1] = np.nan
+   s
+   s.dtype
+   s = s.astype('O')
+   s
+   s.dtype
+
+
+API changes
+~~~~~~~~~~~
+
+  - Added to_series() method to indices, to facilitate the creation of indexers
+    (:issue:`3275`)
+
+  - ``HDFStore``
+
+    - added the method ``select_column`` to select a single column from a table as a Series.
+    - deprecated the ``unique`` method, can be replicated by ``select_column(key,column).unique()``
+    - ``min_itemsize`` parameter to ``append`` will now automatically create data_columns for passed keys
+
+Enhancements
+~~~~~~~~~~~~
+
+  - Improved performance of df.to_csv() by up to 10x in some cases. (:issue:`3059`)
+
+  - Numexpr is now a :ref:`Recommended Dependencies <install.recommended_dependencies>`, to accelerate certain
+    types of numerical and boolean operations
+
+  - Bottleneck is now a :ref:`Recommended Dependencies <install.recommended_dependencies>`, to accelerate certain
+    types of ``nan`` operations
+
+  - ``HDFStore``
+
+    - support ``read_hdf/to_hdf`` API similar to ``read_csv/to_csv``
+
+      .. ipython:: python
+
+          df = pd.DataFrame({'A': range(5), 'B': range(5)})
+          df.to_hdf('store.h5', key='table', append=True)
+          pd.read_hdf('store.h5', 'table', where=['index > 2'])
+
+      .. ipython:: python
+          :suppress:
+          :okexcept:
+
+          import os
+
+          os.remove('store.h5')
+
+    - provide dotted attribute access to ``get`` from stores, e.g. ``store.df == store['df']``
+
+    - new keywords ``iterator=boolean``, and ``chunksize=number_in_a_chunk`` are
+      provided to support iteration on ``select`` and ``select_as_multiple`` (:issue:`3076`)
+
+  - You can now select timestamps from an *unordered* timeseries similarly to an *ordered* timeseries (:issue:`2437`)
+
+  - You can now select with a string from a DataFrame with a datelike index, in a similar way to a Series (:issue:`3070`)
+
+    .. code-block:: ipython
+
+     In [30]: idx = pd.date_range("2001-10-1", periods=5, freq='M')
+
+     In [31]: ts = pd.Series(np.random.rand(len(idx)), index=idx)
+
+     In [32]: ts['2001']
+     Out[32]:
+     2001-10-31    0.117967
+     2001-11-30    0.702184
+     2001-12-31    0.414034
+     Freq: M, dtype: float64
+
+     In [33]: df = pd.DataFrame({'A': ts})
+
+     In [34]: df['2001']
+     Out[34]:
+                        A
+     2001-10-31  0.117967
+     2001-11-30  0.702184
+     2001-12-31  0.414034
+
+  - ``Squeeze`` to possibly remove length 1 dimensions from an object.
+
+    .. code-block:: python
+
+       >>> p = pd.Panel(np.random.randn(3, 4, 4), items=['ItemA', 'ItemB', 'ItemC'],
+       ...              major_axis=pd.date_range('20010102', periods=4),
+       ...              minor_axis=['A', 'B', 'C', 'D'])
+       >>> p
+       <class 'pandas.core.panel.Panel'>
+       Dimensions: 3 (items) x 4 (major_axis) x 4 (minor_axis)
+       Items axis: ItemA to ItemC
+       Major_axis axis: 2001-01-02 00:00:00 to 2001-01-05 00:00:00
+       Minor_axis axis: A to D
+
+       >>> p.reindex(items=['ItemA']).squeeze()
+                          A         B         C         D
+       2001-01-02  0.926089 -2.026458  0.501277 -0.204683
+       2001-01-03 -0.076524  1.081161  1.141361  0.479243
+       2001-01-04  0.641817 -0.185352  1.824568  0.809152
+       2001-01-05  0.575237  0.669934  1.398014 -0.399338
+
+       >>> p.reindex(items=['ItemA'], minor=['B']).squeeze()
+       2001-01-02   -2.026458
+       2001-01-03    1.081161
+       2001-01-04   -0.185352
+       2001-01-05    0.669934
+       Freq: D, Name: B, dtype: float64
+
+  - In ``pd.io.data.Options``,
+
+    + Fix bug when trying to fetch data for the current month when already
+      past expiry.
+    + Now using lxml to scrape html instead of BeautifulSoup (lxml was faster).
+    + New instance variables for calls and puts are automatically created
+      when a method that creates them is called. This works for current month
+      where the instance variables are simply ``calls`` and ``puts``. Also
+      works for future expiry months and save the instance variable as
+      ``callsMMYY`` or ``putsMMYY``, where ``MMYY`` are, respectively, the
+      month and year of the option's expiry.
+    + ``Options.get_near_stock_price`` now allows the user to specify the
+      month for which to get relevant options data.
+    + ``Options.get_forward_data`` now has optional kwargs ``near`` and
+      ``above_below``. This allows the user to specify if they would like to
+      only return forward looking data for options near the current stock
+      price. This just obtains the data from Options.get_near_stock_price
+      instead of Options.get_xxx_data() (:issue:`2758`).
+
+  - Cursor coordinate information is now displayed in time-series plots.
+
+  - added option ``display.max_seq_items`` to control the number of
+    elements printed per sequence pprinting it.  (:issue:`2979`)
+
+  - added option ``display.chop_threshold`` to control display of small numerical
+    values. (:issue:`2739`)
+
+  - added option ``display.max_info_rows`` to prevent verbose_info from being
+    calculated for frames above 1M rows (configurable). (:issue:`2807`, :issue:`2918`)
+
+  - value_counts() now accepts a "normalize" argument, for normalized
+    histograms. (:issue:`2710`).
+
+  - DataFrame.from_records now accepts not only dicts but any instance of
+    the collections.Mapping ABC.
+
+  - added option ``display.mpl_style`` providing a sleeker visual style
+    for plots. Based on https://gist.github.com/huyng/816622 (:issue:`3075`).
+
+  - Treat boolean values as integers (values 1 and 0) for numeric
+    operations. (:issue:`2641`)
+
+  - to_html() now accepts an optional "escape" argument to control reserved
+    HTML character escaping (enabled by default) and escapes ``&``, in addition
+    to ``<`` and ``>``.  (:issue:`2919`)
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+.. _whatsnew_0.11.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.10.1..v0.11.0
diff --git a/pandas/doc/source/whatsnew/v0.12.0.rst b/pandas/doc/source/whatsnew/v0.12.0.rst
new file mode 100644
index 0000000000000..08d3a6b188322
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.12.0.rst
@@ -0,0 +1,564 @@
+.. _whatsnew_0120:
+
+Version 0.12.0 (July 24, 2013)
+------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.11.0 and includes several new features and
+enhancements along with a large number of bug fixes.
+
+Highlights include a consistent I/O API naming scheme, routines to read html,
+write MultiIndexes to csv files, read & write STATA data files, read & write JSON format
+files, Python 3 support for ``HDFStore``, filtering of groupby expressions via ``filter``, and a
+revamped ``replace`` routine that accepts regular expressions.
+
+API changes
+~~~~~~~~~~~
+
+  - The I/O API is now much more consistent with a set of top level ``reader`` functions
+    accessed like ``pd.read_csv()`` that generally return a ``pandas`` object.
+
+    * ``read_csv``
+    * ``read_excel``
+    * ``read_hdf``
+    * ``read_sql``
+    * ``read_json``
+    * ``read_html``
+    * ``read_stata``
+    * ``read_clipboard``
+
+    The corresponding ``writer`` functions are object methods that are accessed like ``df.to_csv()``
+
+    * ``to_csv``
+    * ``to_excel``
+    * ``to_hdf``
+    * ``to_sql``
+    * ``to_json``
+    * ``to_html``
+    * ``to_stata``
+    * ``to_clipboard``
+
+
+  - Fix modulo and integer division on Series,DataFrames to act similarly to ``float`` dtypes to return
+    ``np.nan`` or ``np.inf`` as appropriate (:issue:`3590`). This correct a numpy bug that treats ``integer``
+    and ``float`` dtypes differently.
+
+    .. ipython:: python
+
+        p = pd.DataFrame({"first": [4, 5, 8], "second": [0, 0, 3]})
+        p % 0
+        p % p
+        p / p
+        p / 0
+
+  - Add ``squeeze`` keyword to ``groupby`` to allow reduction from
+    DataFrame -> Series if groups are unique. This is a Regression from 0.10.1.
+    We are reverting back to the prior behavior. This means groupby will return the
+    same shaped objects whether the groups are unique or not. Revert this issue (:issue:`2893`)
+    with (:issue:`3596`).
+
+    .. code-block:: ipython
+
+        In [2]: df2 = pd.DataFrame([{"val1": 1, "val2": 20},
+           ...:                     {"val1": 1, "val2": 19},
+           ...:                     {"val1": 1, "val2": 27},
+           ...:                     {"val1": 1, "val2": 12}])
+
+        In [3]: def func(dataf):
+           ...:     return dataf["val2"] - dataf["val2"].mean()
+           ...:
+
+        In [4]: # squeezing the result frame to a series (because we have unique groups)
+           ...: df2.groupby("val1", squeeze=True).apply(func)
+        Out[4]:
+        0    0.5
+        1   -0.5
+        2    7.5
+        3   -7.5
+        Name: 1, dtype: float64
+
+        In [5]: # no squeezing (the default, and behavior in 0.10.1)
+           ...: df2.groupby("val1").apply(func)
+        Out[5]:
+        val2    0    1    2    3
+        val1
+        1     0.5 -0.5  7.5 -7.5
+
+  - Raise on ``iloc`` when boolean indexing with a label based indexer mask
+    e.g. a boolean Series, even with integer labels, will raise. Since ``iloc``
+    is purely positional based, the labels on the Series are not alignable (:issue:`3631`)
+
+    This case is rarely used, and there are plenty of alternatives. This preserves the
+    ``iloc`` API to be *purely* positional based.
+
+    .. ipython:: python
+
+       df = pd.DataFrame(range(5), index=list("ABCDE"), columns=["a"])
+       mask = df.a % 2 == 0
+       mask
+
+       # this is what you should use
+       df.loc[mask]
+
+       # this will work as well
+       df.iloc[mask.values]
+
+    ``df.iloc[mask]`` will raise a ``ValueError``
+
+  - The ``raise_on_error`` argument to plotting functions is removed. Instead,
+    plotting functions raise a ``TypeError`` when the ``dtype`` of the object
+    is ``object`` to remind you to avoid ``object`` arrays whenever possible
+    and thus you should cast to an appropriate numeric dtype if you need to
+    plot something.
+
+  - Add ``colormap`` keyword to DataFrame plotting methods. Accepts either a
+    matplotlib colormap object (ie, matplotlib.cm.jet) or a string name of such
+    an object (ie, 'jet'). The colormap is sampled to select the color for each
+    column. Please see :ref:`visualization.colormaps` for more information.
+    (:issue:`3860`)
+
+  - ``DataFrame.interpolate()`` is now deprecated. Please use
+    ``DataFrame.fillna()`` and ``DataFrame.replace()`` instead. (:issue:`3582`,
+    :issue:`3675`, :issue:`3676`)
+
+  - the ``method`` and ``axis`` arguments of ``DataFrame.replace()`` are
+    deprecated
+
+  - ``DataFrame.replace`` 's ``infer_types`` parameter is removed and now
+    performs conversion by default. (:issue:`3907`)
+
+  - Add the keyword ``allow_duplicates`` to ``DataFrame.insert`` to allow a duplicate column
+    to be inserted if ``True``, default is ``False`` (same as prior to 0.12) (:issue:`3679`)
+  - Implement ``__nonzero__`` for ``NDFrame`` objects (:issue:`3691`, :issue:`3696`)
+
+  - IO API
+
+    - Added top-level function ``read_excel`` to replace the following,
+      The original API is deprecated and will be removed in a future version
+
+      .. code-block:: python
+
+         from pandas.io.parsers import ExcelFile
+
+         xls = ExcelFile("path_to_file.xls")
+         xls.parse("Sheet1", index_col=None, na_values=["NA"])
+
+      With
+
+      .. code-block:: python
+
+         import pandas as pd
+
+         pd.read_excel("path_to_file.xls", "Sheet1", index_col=None, na_values=["NA"])
+
+    - Added top-level function ``read_sql`` that is equivalent to the following
+
+      .. code-block:: python
+
+         from pandas.io.sql import read_frame
+
+         read_frame(...)
+
+  - ``DataFrame.to_html`` and ``DataFrame.to_latex`` now accept a path for
+    their first argument (:issue:`3702`)
+
+  - Do not allow astypes on ``datetime64[ns]`` except to ``object``, and
+    ``timedelta64[ns]`` to ``object/int`` (:issue:`3425`)
+
+  - The behavior of ``datetime64`` dtypes has changed with respect to certain
+    so-called reduction operations (:issue:`3726`). The following operations now
+    raise a ``TypeError`` when performed on a ``Series`` and return an *empty*
+    ``Series`` when performed on a ``DataFrame`` similar to performing these
+    operations on, for example, a ``DataFrame`` of ``slice`` objects:
+
+    - sum, prod, mean, std, var, skew, kurt, corr, and cov
+
+  - ``read_html`` now defaults to ``None`` when reading, and falls back on
+    ``bs4`` + ``html5lib`` when lxml fails to parse. a list of parsers to try
+    until success is also valid
+
+  - The internal ``pandas`` class hierarchy has changed (slightly). The
+    previous ``PandasObject`` now is called ``PandasContainer`` and a new
+    ``PandasObject`` has become the base class for ``PandasContainer`` as well
+    as ``Index``, ``Categorical``, ``GroupBy``, ``SparseList``, and
+    ``SparseArray`` (+ their base classes). Currently, ``PandasObject``
+    provides string methods (from ``StringMixin``). (:issue:`4090`, :issue:`4092`)
+
+  - New ``StringMixin`` that, given a ``__unicode__`` method, gets python 2 and
+    python 3 compatible string methods (``__str__``, ``__bytes__``, and
+    ``__repr__``). Plus string safety throughout. Now employed in many places
+    throughout the pandas library. (:issue:`4090`, :issue:`4092`)
+
+IO enhancements
+~~~~~~~~~~~~~~~
+
+  - ``pd.read_html()`` can now parse HTML strings, files or urls and return
+    DataFrames, courtesy of @cpcloud. (:issue:`3477`, :issue:`3605`, :issue:`3606`, :issue:`3616`).
+    It works with a *single* parser backend: BeautifulSoup4 + html5lib :ref:`See the docs<io.html>`
+
+    You can use ``pd.read_html()`` to read the output from ``DataFrame.to_html()`` like so
+
+    .. ipython:: python
+
+        import io
+        df = pd.DataFrame({"a": range(3), "b": list("abc")})
+        print(df)
+        html = df.to_html()
+        alist = pd.read_html(io.StringIO(html), index_col=0)
+        print(df == alist[0])
+
+    Note that ``alist`` here is a Python ``list`` so ``pd.read_html()`` and
+    ``DataFrame.to_html()`` are not inverses.
+
+    - ``pd.read_html()`` no longer performs hard conversion of date strings
+      (:issue:`3656`).
+
+    .. warning::
+
+      You may have to install an older version of BeautifulSoup4,
+      :ref:`See the installation docs<install.optional_dependencies>`
+
+  - Added module for reading and writing Stata files: ``pandas.io.stata`` (:issue:`1512`)
+    accessible via ``read_stata`` top-level function for reading,
+    and ``to_stata`` DataFrame method for writing, :ref:`See the docs<io.stata>`
+
+  - Added module for reading and writing json format files: ``pandas.io.json``
+    accessible via ``read_json`` top-level function for reading,
+    and ``to_json`` DataFrame method for writing, :ref:`See the docs<io.json>`
+    various issues (:issue:`1226`, :issue:`3804`, :issue:`3876`, :issue:`3867`, :issue:`1305`)
+
+  - ``MultiIndex`` column support for reading and writing csv format files
+
+    - The ``header`` option in ``read_csv`` now accepts a
+      list of the rows from which to read the index.
+
+    - The option, ``tupleize_cols`` can now be specified in both ``to_csv`` and
+      ``read_csv``, to provide compatibility for the pre 0.12 behavior of
+      writing and reading ``MultIndex`` columns via a list of tuples. The default in
+      0.12 is to write lists of tuples and *not* interpret list of tuples as a
+      ``MultiIndex`` column.
+
+      Note: The default behavior in 0.12 remains unchanged from prior versions, but starting with 0.13,
+      the default *to* write and read ``MultiIndex`` columns will be in the new
+      format. (:issue:`3571`, :issue:`1651`, :issue:`3141`)
+
+    - If an ``index_col`` is not specified (e.g. you don't have an index, or wrote it
+      with ``df.to_csv(..., index=False``), then any ``names`` on the columns index will
+      be *lost*.
+
+      .. ipython:: python
+
+         mi_idx = pd.MultiIndex.from_arrays([[1, 2, 3, 4], list("abcd")], names=list("ab"))
+         mi_col = pd.MultiIndex.from_arrays([[1, 2], list("ab")], names=list("cd"))
+         df = pd.DataFrame(np.ones((4, 2)), index=mi_idx, columns=mi_col)
+         df.to_csv("mi.csv")
+         print(open("mi.csv").read())
+         pd.read_csv("mi.csv", header=[0, 1, 2, 3], index_col=[0, 1])
+
+      .. ipython:: python
+         :suppress:
+
+         import os
+
+         os.remove("mi.csv")
+
+  - Support for ``HDFStore`` (via ``PyTables 3.0.0``) on Python3
+
+  - Iterator support via ``read_hdf`` that automatically opens and closes the
+    store when iteration is finished. This is only for *tables*
+
+    .. code-block:: ipython
+
+        In [25]: path = 'store_iterator.h5'
+
+        In [26]: pd.DataFrame(np.random.randn(10, 2)).to_hdf(path, 'df', table=True)
+
+        In [27]: for df in pd.read_hdf(path, 'df', chunksize=3):
+           ....:     print(df)
+           ....:
+                  0         1
+        0  0.713216 -0.778461
+        1 -0.661062  0.862877
+        2  0.344342  0.149565
+                  0         1
+        3 -0.626968 -0.875772
+        4 -0.930687 -0.218983
+        5  0.949965 -0.442354
+                  0         1
+        6 -0.402985  1.111358
+        7 -0.241527 -0.670477
+        8  0.049355  0.632633
+                  0         1
+        9 -1.502767 -1.225492
+
+
+
+  - ``read_csv`` will now throw a more informative error message when a file
+    contains no columns, e.g., all newline characters
+
+Other enhancements
+~~~~~~~~~~~~~~~~~~
+
+  - ``DataFrame.replace()`` now allows regular expressions on contained
+    ``Series`` with object dtype. See the examples section in the regular docs
+    :ref:`Replacing via String Expression <missing_data.replace_expression>`
+
+    For example you can do
+
+    .. ipython:: python
+
+        df = pd.DataFrame({"a": list("ab.."), "b": [1, 2, 3, 4]})
+        df.replace(regex=r"\s*\.\s*", value=np.nan)
+
+    to replace all occurrences of the string ``'.'`` with zero or more
+    instances of surrounding white space with ``NaN``.
+
+    Regular string replacement still works as expected. For example, you can do
+
+    .. ipython:: python
+
+        df.replace(".", np.nan)
+
+    to replace all occurrences of the string ``'.'`` with ``NaN``.
+
+  - ``pd.melt()`` now accepts the optional parameters ``var_name`` and ``value_name``
+    to specify custom column names of the returned DataFrame.
+
+  - ``pd.set_option()`` now allows N option, value pairs (:issue:`3667`).
+
+    Let's say that we had an option ``'a.b'`` and another option ``'b.c'``.
+    We can set them at the same time:
+
+    .. code-block:: ipython
+
+        In [31]: pd.get_option('a.b')
+        Out[31]: 2
+
+        In [32]: pd.get_option('b.c')
+        Out[32]: 3
+
+        In [33]: pd.set_option('a.b', 1, 'b.c', 4)
+
+        In [34]: pd.get_option('a.b')
+        Out[34]: 1
+
+        In [35]: pd.get_option('b.c')
+        Out[35]: 4
+
+  - The ``filter`` method for group objects returns a subset of the original
+    object. Suppose we want to take only elements that belong to groups with a
+    group sum greater than 2.
+
+    .. ipython:: python
+
+       sf = pd.Series([1, 1, 2, 3, 3, 3])
+       sf.groupby(sf).filter(lambda x: x.sum() > 2)
+
+    The argument of ``filter`` must a function that, applied to the group as a
+    whole, returns ``True`` or ``False``.
+
+    Another useful operation is filtering out elements that belong to groups
+    with only a couple members.
+
+    .. ipython:: python
+
+       dff = pd.DataFrame({"A": np.arange(8), "B": list("aabbbbcc")})
+       dff.groupby("B").filter(lambda x: len(x) > 2)
+
+    Alternatively, instead of dropping the offending groups, we can return a
+    like-indexed objects where the groups that do not pass the filter are
+    filled with NaNs.
+
+    .. ipython:: python
+
+       dff.groupby("B").filter(lambda x: len(x) > 2, dropna=False)
+
+  - Series and DataFrame hist methods now take a ``figsize`` argument (:issue:`3834`)
+
+  - DatetimeIndexes no longer try to convert mixed-integer indexes during join
+    operations (:issue:`3877`)
+
+  - Timestamp.min and Timestamp.max now represent valid Timestamp instances instead
+    of the default datetime.min and datetime.max (respectively), thanks @SleepingPills
+
+  - ``read_html`` now raises when no tables are found and BeautifulSoup==4.2.0
+    is detected (:issue:`4214`)
+
+
+Experimental features
+~~~~~~~~~~~~~~~~~~~~~
+
+  - Added experimental ``CustomBusinessDay`` class to support ``DateOffsets``
+    with custom holiday calendars and custom weekmasks. (:issue:`2301`)
+
+    .. note::
+
+       This uses the ``numpy.busdaycalendar`` API introduced in Numpy 1.7 and
+       therefore requires Numpy 1.7.0 or newer.
+
+    .. ipython:: python
+
+      from pandas.tseries.offsets import CustomBusinessDay
+      from datetime import datetime
+
+      # As an interesting example, let's look at Egypt where
+      # a Friday-Saturday weekend is observed.
+      weekmask_egypt = "Sun Mon Tue Wed Thu"
+      # They also observe International Workers' Day so let's
+      # add that for a couple of years
+      holidays = ["2012-05-01", datetime(2013, 5, 1), np.datetime64("2014-05-01")]
+      bday_egypt = CustomBusinessDay(holidays=holidays, weekmask=weekmask_egypt)
+      dt = datetime(2013, 4, 30)
+      print(dt + 2 * bday_egypt)
+      dts = pd.date_range(dt, periods=5, freq=bday_egypt)
+      print(pd.Series(dts.weekday, dts).map(pd.Series("Mon Tue Wed Thu Fri Sat Sun".split())))
+
+Bug fixes
+~~~~~~~~~
+
+  - Plotting functions now raise a ``TypeError`` before trying to plot anything
+    if the associated objects have a dtype of ``object`` (:issue:`1818`,
+    :issue:`3572`, :issue:`3911`, :issue:`3912`), but they will try to convert object arrays to
+    numeric arrays if possible so that you can still plot, for example, an
+    object array with floats. This happens before any drawing takes place which
+    eliminates any spurious plots from showing up.
+
+  - ``fillna`` methods now raise a ``TypeError`` if the ``value`` parameter is
+    a list or tuple.
+
+  - ``Series.str`` now supports iteration (:issue:`3638`). You can iterate over the
+    individual elements of each string in the ``Series``. Each iteration yields
+    a ``Series`` with either a single character at each index of the original
+    ``Series`` or ``NaN``. For example,
+
+    .. code-block:: ipython
+
+        In [38]: strs = "go", "bow", "joe", "slow"
+
+        In [32]: ds = pd.Series(strs)
+
+        In [33]: for s in ds.str:
+            ...:     print(s)
+
+        0    g
+        1    b
+        2    j
+        3    s
+        dtype: object
+        0    o
+        1    o
+        2    o
+        3    l
+        dtype: object
+        0    NaN
+        1      w
+        2      e
+        3      o
+        dtype: object
+        0    NaN
+        1    NaN
+        2    NaN
+        3      w
+        dtype: object
+
+        In [41]: s
+        Out[41]:
+        0    NaN
+        1    NaN
+        2    NaN
+        3      w
+        dtype: object
+
+        In [42]: s.dropna().values.item() == "w"
+        Out[42]: True
+
+    The last element yielded by the iterator will be a ``Series`` containing
+    the last element of the longest string in the ``Series`` with all other
+    elements being ``NaN``. Here since ``'slow'`` is the longest string
+    and there are no other strings with the same length ``'w'`` is the only
+    non-null string in the yielded ``Series``.
+
+  - ``HDFStore``
+
+    - Will retain index attributes (freq,tz,name) on recreation (:issue:`3499`)
+    - Will warn with a ``AttributeConflictWarning`` if you are attempting to append
+      an index with a different frequency than the existing, or attempting
+      to append an index with a different name than the existing
+    - Support datelike columns with a timezone as data_columns (:issue:`2852`)
+
+  - Non-unique index support clarified (:issue:`3468`).
+
+    - Fix assigning a new index to a duplicate index in a DataFrame would fail (:issue:`3468`)
+    - Fix construction of a DataFrame with a duplicate index
+    - ref_locs support to allow duplicative indices across dtypes,
+      allows iget support to always find the index (even across dtypes) (:issue:`2194`)
+    - applymap on a DataFrame with a non-unique index now works
+      (removed warning) (:issue:`2786`), and fix (:issue:`3230`)
+    - Fix to_csv to handle non-unique columns (:issue:`3495`)
+    - Duplicate indexes with getitem will return items in the correct order (:issue:`3455`, :issue:`3457`)
+      and handle missing elements like unique indices (:issue:`3561`)
+    - Duplicate indexes with and empty DataFrame.from_records will return a correct frame (:issue:`3562`)
+    - Concat to produce a non-unique columns when duplicates are across dtypes is fixed (:issue:`3602`)
+    - Allow insert/delete to non-unique columns (:issue:`3679`)
+    - Non-unique indexing with a slice via ``loc`` and friends fixed (:issue:`3659`)
+    - Allow insert/delete to non-unique columns (:issue:`3679`)
+    - Extend ``reindex`` to correctly deal with non-unique indices (:issue:`3679`)
+    - ``DataFrame.itertuples()`` now works with frames with duplicate column
+      names (:issue:`3873`)
+    - Bug in non-unique indexing via ``iloc`` (:issue:`4017`); added ``takeable`` argument to
+      ``reindex`` for location-based taking
+    - Allow non-unique indexing in series via ``.ix/.loc`` and ``__getitem__`` (:issue:`4246`)
+    - Fixed non-unique indexing memory allocation issue with ``.ix/.loc`` (:issue:`4280`)
+
+  - ``DataFrame.from_records`` did not accept empty recarrays (:issue:`3682`)
+  - ``read_html`` now correctly skips tests (:issue:`3741`)
+  - Fixed a bug where ``DataFrame.replace`` with a compiled regular expression
+    in the ``to_replace`` argument wasn't working (:issue:`3907`)
+  - Improved ``network`` test decorator to catch ``IOError`` (and therefore
+    ``URLError`` as well). Added ``with_connectivity_check`` decorator to allow
+    explicitly checking a website as a proxy for seeing if there is network
+    connectivity. Plus, new ``optional_args`` decorator factory for decorators.
+    (:issue:`3910`, :issue:`3914`)
+  - Fixed testing issue where too many sockets where open thus leading to a
+    connection reset issue (:issue:`3982`, :issue:`3985`, :issue:`4028`,
+    :issue:`4054`)
+  - Fixed failing tests in test_yahoo, test_google where symbols were not
+    retrieved but were being accessed (:issue:`3982`, :issue:`3985`,
+    :issue:`4028`, :issue:`4054`)
+  - ``Series.hist`` will now take the figure from the current environment if
+    one is not passed
+  - Fixed bug where a 1xN DataFrame would barf on a 1xN mask (:issue:`4071`)
+  - Fixed running of ``tox`` under python3 where the pickle import was getting
+    rewritten in an incompatible way (:issue:`4062`, :issue:`4063`)
+  - Fixed bug where sharex and sharey were not being passed to grouped_hist
+    (:issue:`4089`)
+  - Fixed bug in ``DataFrame.replace`` where a nested dict wasn't being
+    iterated over when regex=False (:issue:`4115`)
+  - Fixed bug in the parsing of microseconds when using the ``format``
+    argument in ``to_datetime`` (:issue:`4152`)
+  - Fixed bug in ``PandasAutoDateLocator`` where ``invert_xaxis`` triggered
+    incorrectly ``MilliSecondLocator``  (:issue:`3990`)
+  - Fixed bug in plotting that wasn't raising on invalid colormap for
+    matplotlib 1.1.1 (:issue:`4215`)
+  - Fixed the legend displaying in ``DataFrame.plot(kind='kde')`` (:issue:`4216`)
+  - Fixed bug where Index slices weren't carrying the name attribute
+    (:issue:`4226`)
+  - Fixed bug in initializing ``DatetimeIndex`` with an array of strings
+    in a certain time zone (:issue:`4229`)
+  - Fixed bug where html5lib wasn't being properly skipped (:issue:`4265`)
+  - Fixed bug where get_data_famafrench wasn't using the correct file edges
+    (:issue:`4281`)
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+.. _whatsnew_0.12.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.11.0..v0.12.0
diff --git a/pandas/doc/source/whatsnew/v0.13.0.rst b/pandas/doc/source/whatsnew/v0.13.0.rst
new file mode 100644
index 0000000000000..8e323d8aac5e3
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.13.0.rst
@@ -0,0 +1,1386 @@
+.. _whatsnew_0130:
+
+Version 0.13.0 (January 3, 2014)
+--------------------------------
+
+{{ header }}
+
+
+
+This is a major release from 0.12.0 and includes a number of API changes, several new features and
+enhancements along with a large number of bug fixes.
+
+Highlights include:
+
+- support for a new index type ``Float64Index``, and other Indexing enhancements
+- ``HDFStore`` has a new string based syntax for query specification
+- support for new methods of interpolation
+- updated ``timedelta`` operations
+- a new string manipulation method ``extract``
+- Nanosecond support for Offsets
+- ``isin`` for DataFrames
+
+Several experimental features are added, including:
+
+- new ``eval/query`` methods for expression evaluation
+- support for ``msgpack`` serialization
+- an i/o interface to Google's ``BigQuery``
+
+Their are several new or updated docs sections including:
+
+- :ref:`Comparison with SQL<compare_with_sql>`, which should be useful for those familiar with SQL but still learning pandas.
+- :ref:`Comparison with R<compare_with_r>`, idiom translations from R to pandas.
+- :ref:`Enhancing Performance<enhancingperf>`, ways to enhance pandas performance with ``eval/query``.
+
+.. warning::
+
+   In 0.13.0 ``Series`` has internally been refactored to no longer sub-class ``ndarray``
+   but instead subclass ``NDFrame``, similar to the rest of the pandas containers. This should be
+   a transparent change with only very limited API implications. See :ref:`Internal Refactoring<whatsnew_0130.refactoring>`
+
+API changes
+~~~~~~~~~~~
+
+- ``read_excel`` now supports an integer in its ``sheetname`` argument giving
+  the index of the sheet to read in (:issue:`4301`).
+- Text parser now treats anything that reads like inf ("inf", "Inf", "-Inf",
+  "iNf", etc.) as infinity. (:issue:`4220`, :issue:`4219`), affecting
+  ``read_table``, ``read_csv``, etc.
+- ``pandas`` now is Python 2/3 compatible without the need for 2to3 thanks to
+  @jtratner. As a result, pandas now uses iterators more extensively. This
+  also led to the introduction of substantive parts of the Benjamin
+  Peterson's ``six`` library into compat. (:issue:`4384`, :issue:`4375`,
+  :issue:`4372`)
+- ``pandas.util.compat`` and ``pandas.util.py3compat`` have been merged into
+  ``pandas.compat``. ``pandas.compat`` now includes many functions allowing
+  2/3 compatibility. It contains both list and iterator versions of range,
+  filter, map and zip, plus other necessary elements for Python 3
+  compatibility. ``lmap``, ``lzip``, ``lrange`` and ``lfilter`` all produce
+  lists instead of iterators, for compatibility with ``numpy``, subscripting
+  and ``pandas`` constructors.(:issue:`4384`, :issue:`4375`, :issue:`4372`)
+- ``Series.get`` with negative indexers now returns the same as ``[]`` (:issue:`4390`)
+- Changes to how ``Index`` and ``MultiIndex`` handle metadata (``levels``,
+  ``labels``, and ``names``) (:issue:`4039`):
+
+  .. code-block:: python
+
+     # previously, you would have set levels or labels directly
+     >>> pd.index.levels = [[1, 2, 3, 4], [1, 2, 4, 4]]
+
+     # now, you use the set_levels or set_labels methods
+     >>> index = pd.index.set_levels([[1, 2, 3, 4], [1, 2, 4, 4]])
+
+     # similarly, for names, you can rename the object
+     # but setting names is not deprecated
+     >>> index = pd.index.set_names(["bob", "cranberry"])
+
+     # and all methods take an inplace kwarg - but return None
+     >>> pd.index.set_names(["bob", "cranberry"], inplace=True)
+
+- **All** division with ``NDFrame`` objects is now *truedivision*, regardless
+  of the future import. This means that operating on pandas objects will by default
+  use *floating point* division, and return a floating point dtype.
+  You can use ``//`` and ``floordiv`` to do integer division.
+
+  Integer division
+
+  .. code-block:: ipython
+
+     In [3]: arr = np.array([1, 2, 3, 4])
+
+     In [4]: arr2 = np.array([5, 3, 2, 1])
+
+     In [5]: arr / arr2
+     Out[5]: array([0, 0, 1, 4])
+
+     In [6]: pd.Series(arr) // pd.Series(arr2)
+     Out[6]:
+     0    0
+     1    0
+     2    1
+     3    4
+     dtype: int64
+
+  True Division
+
+  .. code-block:: ipython
+
+      In [7]: pd.Series(arr) / pd.Series(arr2)  # no future import required
+      Out[7]:
+      0    0.200000
+      1    0.666667
+      2    1.500000
+      3    4.000000
+      dtype: float64
+
+- Infer and downcast dtype if ``downcast='infer'`` is passed to ``fillna/ffill/bfill`` (:issue:`4604`)
+- ``__nonzero__`` for all NDFrame objects, will now raise a ``ValueError``, this reverts back to (:issue:`1073`, :issue:`4633`)
+  behavior. See :ref:`gotchas<gotchas.truth>` for a more detailed discussion.
+
+  This prevents doing boolean comparison on *entire* pandas objects, which is inherently ambiguous. These all will raise a ``ValueError``.
+
+  .. code-block:: python
+
+     >>> df = pd.DataFrame({'A': np.random.randn(10),
+     ...                    'B': np.random.randn(10),
+     ...                    'C': pd.date_range('20130101', periods=10)
+     ...                    })
+     ...
+     >>> if df:
+     ...     pass
+     ...
+     Traceback (most recent call last):
+         ...
+     ValueError: The truth value of a DataFrame is ambiguous.  Use a.empty,
+     a.bool(), a.item(), a.any() or a.all().
+
+     >>> df1 = df
+     >>> df2 = df
+     >>> df1 and df2
+     Traceback (most recent call last):
+         ...
+     ValueError: The truth value of a DataFrame is ambiguous.  Use a.empty,
+     a.bool(), a.item(), a.any() or a.all().
+
+     >>> d = [1, 2, 3]
+     >>> s1 = pd.Series(d)
+     >>> s2 = pd.Series(d)
+     >>> s1 and s2
+     Traceback (most recent call last):
+         ...
+     ValueError: The truth value of a DataFrame is ambiguous.  Use a.empty,
+     a.bool(), a.item(), a.any() or a.all().
+
+  Added the ``.bool()`` method to ``NDFrame`` objects to facilitate evaluating of single-element boolean Series:
+
+  .. code-block:: python
+
+     >>> pd.Series([True]).bool()
+      True
+     >>> pd.Series([False]).bool()
+      False
+     >>> pd.DataFrame([[True]]).bool()
+      True
+     >>> pd.DataFrame([[False]]).bool()
+      False
+
+- All non-Index NDFrames (``Series``, ``DataFrame``, ``Panel``, ``Panel4D``,
+  ``SparsePanel``, etc.), now support the entire set of arithmetic operators
+  and arithmetic flex methods (add, sub, mul, etc.). ``SparsePanel`` does not
+  support ``pow`` or ``mod`` with non-scalars. (:issue:`3765`)
+- ``Series`` and ``DataFrame`` now have a ``mode()`` method to calculate the
+  statistical mode(s) by axis/Series. (:issue:`5367`)
+
+- Chained assignment will now by default warn if the user is assigning to a copy. This can be changed
+  with the option ``mode.chained_assignment``, allowed options are ``raise/warn/None``.
+
+  .. ipython:: python
+
+     dfc = pd.DataFrame({'A': ['aaa', 'bbb', 'ccc'], 'B': [1, 2, 3]})
+     pd.set_option('chained_assignment', 'warn')
+
+  The following warning / exception will show if this is attempted.
+
+  .. ipython:: python
+     :okwarning:
+
+     dfc.loc[0]['A'] = 1111
+
+  ::
+
+     Traceback (most recent call last)
+        ...
+     SettingWithCopyWarning:
+        A value is trying to be set on a copy of a slice from a DataFrame.
+        Try using .loc[row_index,col_indexer] = value instead
+
+  Here is the correct method of assignment.
+
+  .. ipython:: python
+
+     dfc.loc[0, 'A'] = 11
+     dfc
+
+- ``Panel.reindex`` has the following call signature ``Panel.reindex(items=None, major_axis=None, minor_axis=None, **kwargs)``
+   to conform with other ``NDFrame`` objects. See :ref:`Internal Refactoring<whatsnew_0130.refactoring>` for more information.
+
+- ``Series.argmin`` and ``Series.argmax`` are now aliased to ``Series.idxmin`` and ``Series.idxmax``. These return the *index* of the
+   min or max element respectively. Prior to 0.13.0 these would return the position of the min / max element. (:issue:`6214`)
+
+Prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+These were announced changes in 0.12 or prior that are taking effect as of 0.13.0
+
+- Remove deprecated ``Factor`` (:issue:`3650`)
+- Remove deprecated ``set_printoptions/reset_printoptions`` (:issue:`3046`)
+- Remove deprecated ``_verbose_info`` (:issue:`3215`)
+- Remove deprecated ``read_clipboard/to_clipboard/ExcelFile/ExcelWriter`` from ``pandas.io.parsers`` (:issue:`3717`)
+  These are available as functions in the main pandas namespace (e.g. ``pd.read_clipboard``)
+- default for ``tupleize_cols`` is now ``False`` for both ``to_csv`` and ``read_csv``. Fair warning in 0.12 (:issue:`3604`)
+- default for ``display.max_seq_len`` is now 100 rather than ``None``. This activates
+  truncated display ("...") of long sequences in various places. (:issue:`3391`)
+
+Deprecations
+~~~~~~~~~~~~
+
+Deprecated in 0.13.0
+
+- deprecated ``iterkv``, which will be removed in a future release (this was
+  an alias of iteritems used to bypass ``2to3``'s changes).
+  (:issue:`4384`, :issue:`4375`, :issue:`4372`)
+- deprecated the string method ``match``, whose role is now performed more
+  idiomatically by ``extract``. In a future release, the default behavior
+  of ``match`` will change to become analogous to ``contains``, which returns
+  a boolean indexer. (Their
+  distinction is strictness: ``match`` relies on ``re.match`` while
+  ``contains`` relies on ``re.search``.) In this release, the deprecated
+  behavior is the default, but the new behavior is available through the
+  keyword argument ``as_indexer=True``.
+
+Indexing API changes
+~~~~~~~~~~~~~~~~~~~~
+
+Prior to 0.13, it was impossible to use a label indexer (``.loc/.ix``) to set a value that
+was not contained in the index of a particular axis. (:issue:`2578`). See :ref:`the docs<indexing.basics.partial_setting>`
+
+In the ``Series`` case this is effectively an appending operation
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3])
+   s
+   s[5] = 5.
+   s
+
+.. ipython:: python
+
+   dfi = pd.DataFrame(np.arange(6).reshape(3, 2),
+                      columns=['A', 'B'])
+   dfi
+
+This would previously ``KeyError``
+
+.. ipython:: python
+
+   dfi.loc[:, 'C'] = dfi.loc[:, 'A']
+   dfi
+
+This is like an ``append`` operation.
+
+.. ipython:: python
+
+   dfi.loc[3] = 5
+   dfi
+
+A Panel setting operation on an arbitrary axis aligns the input to the Panel
+
+.. code-block:: ipython
+
+   In [20]: p = pd.Panel(np.arange(16).reshape(2, 4, 2),
+      ....:              items=['Item1', 'Item2'],
+      ....:              major_axis=pd.date_range('2001/1/12', periods=4),
+      ....:              minor_axis=['A', 'B'], dtype='float64')
+      ....:
+
+   In [21]: p
+   Out[21]:
+   <class 'pandas.core.panel.Panel'>
+   Dimensions: 2 (items) x 4 (major_axis) x 2 (minor_axis)
+   Items axis: Item1 to Item2
+   Major_axis axis: 2001-01-12 00:00:00 to 2001-01-15 00:00:00
+   Minor_axis axis: A to B
+
+   In [22]: p.loc[:, :, 'C'] = pd.Series([30, 32], index=p.items)
+
+   In [23]: p
+   Out[23]:
+   <class 'pandas.core.panel.Panel'>
+   Dimensions: 2 (items) x 4 (major_axis) x 3 (minor_axis)
+   Items axis: Item1 to Item2
+   Major_axis axis: 2001-01-12 00:00:00 to 2001-01-15 00:00:00
+   Minor_axis axis: A to C
+
+   In [24]: p.loc[:, :, 'C']
+   Out[24]:
+               Item1  Item2
+   2001-01-12   30.0   32.0
+   2001-01-13   30.0   32.0
+   2001-01-14   30.0   32.0
+   2001-01-15   30.0   32.0
+
+Float64Index API change
+~~~~~~~~~~~~~~~~~~~~~~~
+
+- Added a new index type, ``Float64Index``. This will be automatically created when passing floating values in index creation.
+  This enables a pure label-based slicing paradigm that makes ``[],ix,loc`` for scalar indexing and slicing work exactly the
+  same. (:issue:`263`)
+
+  Construction is by default for floating type values.
+
+  .. ipython:: python
+
+     index = pd.Index([1.5, 2, 3, 4.5, 5])
+     index
+     s = pd.Series(range(5), index=index)
+     s
+
+  Scalar selection for ``[],.ix,.loc`` will always be label based. An integer will match an equal float index (e.g. ``3`` is equivalent to ``3.0``)
+
+  .. ipython:: python
+
+     s[3]
+     s.loc[3]
+
+  The only positional indexing is via ``iloc``
+
+  .. ipython:: python
+
+     s.iloc[3]
+
+  A scalar index that is not found will raise ``KeyError``
+
+  Slicing is ALWAYS on the values of the index, for ``[],ix,loc`` and ALWAYS positional with ``iloc``
+
+  .. ipython:: python
+     :okwarning:
+
+     s.loc[2:4]
+     s.iloc[2:4]
+
+  In float indexes, slicing using floats are allowed
+
+  .. ipython:: python
+
+     s[2.1:4.6]
+     s.loc[2.1:4.6]
+
+- Indexing on other index types are preserved (and positional fallback for ``[],ix``), with the exception, that floating point slicing
+  on indexes on non ``Float64Index`` will now raise a ``TypeError``.
+
+  .. code-block:: ipython
+
+     In [1]: pd.Series(range(5))[3.5]
+     TypeError: the label [3.5] is not a proper indexer for this index type (Int64Index)
+
+     In [1]: pd.Series(range(5))[3.5:4.5]
+     TypeError: the slice start [3.5] is not a proper indexer for this index type (Int64Index)
+
+  Using a scalar float indexer will be deprecated in a future version, but is allowed for now.
+
+  .. code-block:: ipython
+
+     In [3]: pd.Series(range(5))[3.0]
+     Out[3]: 3
+
+HDFStore API changes
+~~~~~~~~~~~~~~~~~~~~
+
+- Query Format Changes. A much more string-like query format is now supported. See :ref:`the docs<io.hdf5-query>`.
+
+  .. ipython:: python
+
+     path = 'test.h5'
+     dfq = pd.DataFrame(np.random.randn(10, 4),
+                        columns=list('ABCD'),
+                        index=pd.date_range('20130101', periods=10))
+     dfq.to_hdf(path, key='dfq', format='table', data_columns=True)
+
+  Use boolean expressions, with in-line function evaluation.
+
+  .. ipython:: python
+
+     pd.read_hdf(path, 'dfq',
+                 where="index>Timestamp('20130104') & columns=['A', 'B']")
+
+  Use an inline column reference
+
+  .. ipython:: python
+
+     pd.read_hdf(path, 'dfq',
+                 where="A>0 or C>0")
+
+  .. ipython:: python
+     :suppress:
+
+     import os
+     os.remove(path)
+
+- the ``format`` keyword now replaces the ``table`` keyword; allowed values are ``fixed(f)`` or ``table(t)``
+  the same defaults as prior < 0.13.0 remain, e.g. ``put`` implies ``fixed`` format and ``append`` implies
+  ``table`` format. This default format can be set as an option by setting ``io.hdf.default_format``.
+
+  .. ipython:: python
+
+     path = 'test.h5'
+     df = pd.DataFrame(np.random.randn(10, 2))
+     df.to_hdf(path, key='df_table', format='table')
+     df.to_hdf(path, key='df_table2', append=True)
+     df.to_hdf(path, key='df_fixed')
+     with pd.HDFStore(path) as store:
+         print(store)
+
+  .. ipython:: python
+     :suppress:
+
+     import os
+     os.remove(path)
+
+- Significant table writing performance improvements
+- handle a passed ``Series`` in table format (:issue:`4330`)
+- can now serialize a ``timedelta64[ns]`` dtype in a table (:issue:`3577`), See :ref:`the docs<io.hdf5-timedelta>`.
+- added an ``is_open`` property to indicate if the underlying file handle is_open;
+  a closed store will now report 'CLOSED' when viewing the store (rather than raising an error)
+  (:issue:`4409`)
+- a close of a ``HDFStore`` now will close that instance of the ``HDFStore``
+  but will only close the actual file if the ref count (by ``PyTables``) w.r.t. all of the open handles
+  are 0. Essentially you have a local instance of ``HDFStore`` referenced by a variable. Once you
+  close it, it will report closed. Other references (to the same file) will continue to operate
+  until they themselves are closed. Performing an action on a closed file will raise
+  ``ClosedFileError``
+
+  .. ipython:: python
+
+     path = 'test.h5'
+     df = pd.DataFrame(np.random.randn(10, 2))
+     store1 = pd.HDFStore(path)
+     store2 = pd.HDFStore(path)
+     store1.append('df', df)
+     store2.append('df2', df)
+
+     store1
+     store2
+     store1.close()
+     store2
+     store2.close()
+     store2
+
+  .. ipython:: python
+     :suppress:
+
+     import os
+     os.remove(path)
+
+- removed the ``_quiet`` attribute, replace by a ``DuplicateWarning`` if retrieving
+  duplicate rows from a table (:issue:`4367`)
+- removed the ``warn`` argument from ``open``. Instead a ``PossibleDataLossError`` exception will
+  be raised if you try to use ``mode='w'`` with an OPEN file handle (:issue:`4367`)
+- allow a passed locations array or mask as a ``where`` condition (:issue:`4467`).
+  See :ref:`the docs<io.hdf5-where_mask>` for an example.
+- add the keyword ``dropna=True`` to ``append`` to change whether ALL nan rows are not written
+  to the store (default is ``True``, ALL nan rows are NOT written), also settable
+  via the option ``io.hdf.dropna_table`` (:issue:`4625`)
+- pass through store creation arguments; can be used to support in-memory stores
+
+DataFrame repr changes
+~~~~~~~~~~~~~~~~~~~~~~
+
+The HTML and plain text representations of :class:`DataFrame` now show
+a truncated view of the table once it exceeds a certain size, rather
+than switching to the short info view (:issue:`4886`, :issue:`5550`).
+This makes the representation more consistent as small DataFrames get
+larger.
+
+.. image:: ../_static/df_repr_truncated.png
+   :alt: Truncated HTML representation of a DataFrame
+
+To get the info view, call :meth:`DataFrame.info`. If you prefer the
+info view as the repr for large DataFrames, you can set this by running
+``set_option('display.large_repr', 'info')``.
+
+Enhancements
+~~~~~~~~~~~~
+
+- ``df.to_clipboard()`` learned a new ``excel`` keyword that let's you
+  paste df data directly into excel (enabled by default). (:issue:`5070`).
+- ``read_html`` now raises a ``URLError`` instead of catching and raising a
+  ``ValueError`` (:issue:`4303`, :issue:`4305`)
+- Added a test for ``read_clipboard()`` and ``to_clipboard()`` (:issue:`4282`)
+- Clipboard functionality now works with PySide (:issue:`4282`)
+- Added a more informative error message when plot arguments contain
+  overlapping color and style arguments (:issue:`4402`)
+- ``to_dict`` now takes ``records`` as a possible out type.  Returns an array
+  of column-keyed dictionaries. (:issue:`4936`)
+
+- ``NaN`` handing in get_dummies (:issue:`4446`) with ``dummy_na``
+
+  .. ipython:: python
+
+     # previously, nan was erroneously counted as 2 here
+     # now it is not counted at all
+     pd.get_dummies([1, 2, np.nan])
+
+     # unless requested
+     pd.get_dummies([1, 2, np.nan], dummy_na=True)
+
+
+- ``timedelta64[ns]`` operations. See :ref:`the docs<timedeltas.timedeltas_convert>`.
+
+  .. warning::
+
+     Most of these operations require ``numpy >= 1.7``
+
+  Using the new top-level ``to_timedelta``, you can convert a scalar or array from the standard
+  timedelta format (produced by ``to_csv``) into a timedelta type (``np.timedelta64`` in ``nanoseconds``).
+
+  .. code-block:: ipython
+
+     In [53]: pd.to_timedelta('1 days 06:05:01.00003')
+     Out[53]: Timedelta('1 days 06:05:01.000030')
+
+     In [54]: pd.to_timedelta('15.5us')
+     Out[54]: Timedelta('0 days 00:00:00.000015500')
+
+     In [55]: pd.to_timedelta(['1 days 06:05:01.00003', '15.5us', 'nan'])
+     Out[55]: TimedeltaIndex(['1 days 06:05:01.000030', '0 days 00:00:00.000015500', NaT], dtype='timedelta64[ns]', freq=None)
+
+     In [56]: pd.to_timedelta(np.arange(5), unit='s')
+     Out[56]:
+     TimedeltaIndex(['0 days 00:00:00', '0 days 00:00:01', '0 days 00:00:02',
+                     '0 days 00:00:03', '0 days 00:00:04'],
+                    dtype='timedelta64[ns]', freq=None)
+
+     In [57]: pd.to_timedelta(np.arange(5), unit='d')
+     Out[57]: TimedeltaIndex(['0 days', '1 days', '2 days', '3 days', '4 days'], dtype='timedelta64[ns]', freq=None)
+
+  A Series of dtype ``timedelta64[ns]`` can now be divided by another
+  ``timedelta64[ns]`` object, or astyped to yield a ``float64`` dtyped Series. This
+  is frequency conversion. See :ref:`the docs<timedeltas.timedeltas_convert>` for the docs.
+
+  .. ipython:: python
+
+     import datetime
+     td = pd.Series(pd.date_range('20130101', periods=4)) - pd.Series(
+         pd.date_range('20121201', periods=4))
+     td[2] += np.timedelta64(datetime.timedelta(minutes=5, seconds=3))
+     td[3] = np.nan
+     td
+
+  .. code-block:: ipython
+
+     # to days
+     In [63]: td / np.timedelta64(1, 'D')
+     Out[63]:
+     0    31.000000
+     1    31.000000
+     2    31.003507
+     3          NaN
+     dtype: float64
+
+     In [64]: td.astype('timedelta64[D]')
+     Out[64]:
+     0    31.0
+     1    31.0
+     2    31.0
+     3     NaN
+     dtype: float64
+
+     # to seconds
+     In [65]: td / np.timedelta64(1, 's')
+     Out[65]:
+     0    2678400.0
+     1    2678400.0
+     2    2678703.0
+     3          NaN
+     dtype: float64
+
+     In [66]: td.astype('timedelta64[s]')
+     Out[66]:
+     0    2678400.0
+     1    2678400.0
+     2    2678703.0
+     3          NaN
+     dtype: float64
+
+  Dividing or multiplying a ``timedelta64[ns]`` Series by an integer or integer Series
+
+  .. ipython:: python
+
+     td * -1
+     td * pd.Series([1, 2, 3, 4])
+
+  Absolute ``DateOffset`` objects can act equivalently to ``timedeltas``
+
+  .. ipython:: python
+
+     from pandas import offsets
+     td + offsets.Minute(5) + offsets.Milli(5)
+
+  Fillna is now supported for timedeltas
+
+  .. ipython:: python
+
+     td.fillna(pd.Timedelta(0))
+     td.fillna(datetime.timedelta(days=1, seconds=5))
+
+  You can do numeric reduction operations on timedeltas.
+
+  .. ipython:: python
+
+     td.mean()
+     td.quantile(.1)
+
+- ``plot(kind='kde')`` now accepts the optional parameters ``bw_method`` and
+  ``ind``, passed to scipy.stats.gaussian_kde() (for scipy >= 0.11.0) to set
+  the bandwidth, and to gkde.evaluate() to specify the indices at which it
+  is evaluated, respectively. See scipy docs. (:issue:`4298`)
+
+- DataFrame constructor now accepts a numpy masked record array (:issue:`3478`)
+
+- The new vectorized string method ``extract`` return regular expression
+  matches more conveniently.
+
+  .. ipython:: python
+     :okwarning:
+
+     pd.Series(['a1', 'b2', 'c3']).str.extract('[ab](\\d)')
+
+  Elements that do not match return ``NaN``. Extracting a regular expression
+  with more than one group returns a DataFrame with one column per group.
+
+
+  .. ipython:: python
+     :okwarning:
+
+     pd.Series(['a1', 'b2', 'c3']).str.extract('([ab])(\\d)')
+
+  Elements that do not match return a row of ``NaN``.
+  Thus, a Series of messy strings can be *converted* into a
+  like-indexed Series or DataFrame of cleaned-up or more useful strings,
+  without necessitating ``get()`` to access tuples or ``re.match`` objects.
+
+  Named groups like
+
+  .. ipython:: python
+     :okwarning:
+
+     pd.Series(['a1', 'b2', 'c3']).str.extract(
+         '(?P<letter>[ab])(?P<digit>\\d)')
+
+  and optional groups can also be used.
+
+  .. ipython:: python
+     :okwarning:
+
+      pd.Series(['a1', 'b2', '3']).str.extract(
+          '(?P<letter>[ab])?(?P<digit>\\d)')
+
+- ``read_stata`` now accepts Stata 13 format (:issue:`4291`)
+
+- ``read_fwf`` now infers the column specifications from the first 100 rows of
+  the file if the data has correctly separated and properly aligned columns
+  using the delimiter provided to the function (:issue:`4488`).
+
+- support for nanosecond times as an offset
+
+  .. warning::
+
+     These operations require ``numpy >= 1.7``
+
+  Period conversions in the range of seconds and below were reworked and extended
+  up to nanoseconds. Periods in the nanosecond range are now available.
+
+  .. code-block:: python
+
+     In [79]: pd.date_range('2013-01-01', periods=5, freq='5N')
+     Out[79]:
+     DatetimeIndex([          '2013-01-01 00:00:00',
+                    '2013-01-01 00:00:00.000000005',
+                    '2013-01-01 00:00:00.000000010',
+                    '2013-01-01 00:00:00.000000015',
+                    '2013-01-01 00:00:00.000000020'],
+                   dtype='datetime64[ns]', freq='5N')
+
+  or with frequency as offset
+
+  .. ipython:: python
+
+     pd.date_range('2013-01-01', periods=5, freq=pd.offsets.Nano(5))
+
+  Timestamps can be modified in the nanosecond range
+
+  .. ipython:: python
+
+     t = pd.Timestamp('20130101 09:01:02')
+     t + pd.tseries.offsets.Nano(123)
+
+- A new method, ``isin`` for DataFrames, which plays nicely with boolean indexing. The argument to ``isin``, what we're comparing the DataFrame to, can be a DataFrame, Series, dict, or array of values. See :ref:`the docs<indexing.basics.indexing_isin>` for more.
+
+  To get the rows where any of the conditions are met:
+
+  .. ipython:: python
+
+     dfi = pd.DataFrame({'A': [1, 2, 3, 4], 'B': ['a', 'b', 'f', 'n']})
+     dfi
+     other = pd.DataFrame({'A': [1, 3, 3, 7], 'B': ['e', 'f', 'f', 'e']})
+     mask = dfi.isin(other)
+     mask
+     dfi[mask.any(axis=1)]
+
+- ``Series`` now supports a ``to_frame`` method to convert it to a single-column DataFrame (:issue:`5164`)
+
+- All R datasets listed here http://stat.ethz.ch/R-manual/R-devel/library/datasets/html/00Index.html can now be loaded into pandas objects
+
+  .. code-block:: python
+
+     # note that pandas.rpy was deprecated in v0.16.0
+     import pandas.rpy.common as com
+     com.load_data('Titanic')
+
+- ``tz_localize`` can infer a fall daylight savings transition based on the structure
+  of the unlocalized data (:issue:`4230`), see :ref:`the docs<timeseries.timezone>`
+
+- ``DatetimeIndex`` is now in the API documentation, see :ref:`the docs<api.datetimeindex>`
+
+- :meth:`~pandas.io.json.json_normalize` is a new method to allow you to create a flat table
+  from semi-structured JSON data. See :ref:`the docs<io.json_normalize>` (:issue:`1067`)
+
+- Added PySide support for the qtpandas DataFrameModel and DataFrameWidget.
+
+- Python csv parser now supports usecols (:issue:`4335`)
+
+- Frequencies gained several new offsets:
+
+  * ``LastWeekOfMonth`` (:issue:`4637`)
+  * ``FY5253``, and ``FY5253Quarter`` (:issue:`4511`)
+
+
+- DataFrame has a new ``interpolate`` method, similar to Series (:issue:`4434`, :issue:`1892`)
+
+  .. ipython:: python
+
+      df = pd.DataFrame({'A': [1, 2.1, np.nan, 4.7, 5.6, 6.8],
+                        'B': [.25, np.nan, np.nan, 4, 12.2, 14.4]})
+      df.interpolate()
+
+  Additionally, the ``method`` argument to ``interpolate`` has been expanded
+  to include ``'nearest', 'zero', 'slinear', 'quadratic', 'cubic',
+  'barycentric', 'krogh', 'piecewise_polynomial', 'pchip', 'polynomial', 'spline'``
+  The new methods require scipy_. Consult the Scipy reference guide_ and documentation_ for more information
+  about when the various methods are appropriate. See :ref:`the docs<missing_data.interpolate>`.
+
+  Interpolate now also accepts a ``limit`` keyword argument.
+  This works similar to ``fillna``'s limit:
+
+  .. ipython:: python
+
+    ser = pd.Series([1, 3, np.nan, np.nan, np.nan, 11])
+    ser.interpolate(limit=2)
+
+- Added ``wide_to_long`` panel data convenience function. See :ref:`the docs<reshaping.melt>`.
+
+  .. ipython:: python
+
+    np.random.seed(123)
+    df = pd.DataFrame({"A1970" : {0 : "a", 1 : "b", 2 : "c"},
+                       "A1980" : {0 : "d", 1 : "e", 2 : "f"},
+                       "B1970" : {0 : 2.5, 1 : 1.2, 2 : .7},
+                       "B1980" : {0 : 3.2, 1 : 1.3, 2 : .1},
+                       "X"     : dict(zip(range(3), np.random.randn(3)))
+                      })
+    df["id"] = df.index
+    df
+    pd.wide_to_long(df, ["A", "B"], i="id", j="year")
+
+.. _scipy: http://www.scipy.org
+.. _documentation: http://docs.scipy.org/doc/scipy/reference/interpolate.html#univariate-interpolation
+.. _guide: https://docs.scipy.org/doc/scipy/tutorial/interpolate.html
+
+- ``to_csv`` now takes a ``date_format`` keyword argument that specifies how
+  output datetime objects should be formatted. Datetimes encountered in the
+  index, columns, and values will all have this formatting applied. (:issue:`4313`)
+- ``DataFrame.plot`` will scatter plot x versus y by passing ``kind='scatter'`` (:issue:`2215`)
+- Added support for Google Analytics v3 API segment IDs that also supports v2 IDs. (:issue:`5271`)
+
+.. _whatsnew_0130.experimental:
+
+Experimental
+~~~~~~~~~~~~
+
+- The new :func:`~pandas.eval` function implements expression evaluation using
+  ``numexpr`` behind the scenes. This results in large speedups for
+  complicated expressions involving large DataFrames/Series. For example,
+
+  .. ipython:: python
+
+     nrows, ncols = 20000, 100
+     df1, df2, df3, df4 = [pd.DataFrame(np.random.randn(nrows, ncols))
+                           for _ in range(4)]
+
+  .. ipython:: python
+
+     # eval with NumExpr backend
+     %timeit pd.eval('df1 + df2 + df3 + df4')
+
+  .. ipython:: python
+
+     # pure Python evaluation
+     %timeit df1 + df2 + df3 + df4
+
+  For more details, see the :ref:`the docs<enhancingperf.eval>`
+
+- Similar to ``pandas.eval``, :class:`~pandas.DataFrame` has a new
+  ``DataFrame.eval`` method that evaluates an expression in the context of
+  the ``DataFrame``. For example,
+
+  .. ipython:: python
+     :suppress:
+
+     try:
+         del a  # noqa: F821
+     except NameError:
+         pass
+
+     try:
+         del b  # noqa: F821
+     except NameError:
+         pass
+
+  .. ipython:: python
+
+     df = pd.DataFrame(np.random.randn(10, 2), columns=['a', 'b'])
+     df.eval('a + b')
+
+- :meth:`~pandas.DataFrame.query` method has been added that allows
+  you to select elements of a ``DataFrame`` using a natural query syntax
+  nearly identical to Python syntax. For example,
+
+  .. ipython:: python
+     :suppress:
+
+     try:
+         del a  # noqa: F821
+     except NameError:
+         pass
+
+     try:
+         del b  # noqa: F821
+     except NameError:
+         pass
+
+     try:
+         del c  # noqa: F821
+     except NameError:
+         pass
+
+  .. ipython:: python
+
+     n = 20
+     df = pd.DataFrame(np.random.randint(n, size=(n, 3)), columns=['a', 'b', 'c'])
+     df.query('a < b < c')
+
+  selects all the rows of ``df`` where ``a < b < c`` evaluates to ``True``.
+  For more details see the :ref:`the docs<indexing.query>`.
+
+- ``pd.read_msgpack()`` and ``pd.to_msgpack()`` are now a supported method of serialization
+  of arbitrary pandas (and python objects) in a lightweight portable binary format. See :ref:`the docs<io.msgpack>`
+
+  .. warning::
+
+     Since this is an EXPERIMENTAL LIBRARY, the storage format may not be stable until a future release.
+
+  .. code-block:: python
+
+     df = pd.DataFrame(np.random.rand(5, 2), columns=list('AB'))
+     df.to_msgpack('foo.msg')
+     pd.read_msgpack('foo.msg')
+
+     s = pd.Series(np.random.rand(5), index=pd.date_range('20130101', periods=5))
+     pd.to_msgpack('foo.msg', df, s)
+     pd.read_msgpack('foo.msg')
+
+  You can pass ``iterator=True`` to iterator over the unpacked results
+
+  .. code-block:: python
+
+     for o in pd.read_msgpack('foo.msg', iterator=True):
+         print(o)
+
+  .. ipython:: python
+     :suppress:
+     :okexcept:
+
+     os.remove('foo.msg')
+
+- ``pandas.io.gbq`` provides a simple way to extract from, and load data into,
+  Google's BigQuery Data Sets by way of pandas DataFrames. BigQuery is a high
+  performance SQL-like database service, useful for performing ad-hoc queries
+  against extremely large datasets. :ref:`See the docs <io.bigquery>`
+
+  .. code-block:: python
+
+     from pandas.io import gbq
+
+     # A query to select the average monthly temperatures in the
+     # in the year 2000 across the USA. The dataset,
+     # publicata:samples.gsod, is available on all BigQuery accounts,
+     # and is based on NOAA gsod data.
+
+     query = """SELECT station_number as STATION,
+     month as MONTH, AVG(mean_temp) as MEAN_TEMP
+     FROM publicdata:samples.gsod
+     WHERE YEAR = 2000
+     GROUP BY STATION, MONTH
+     ORDER BY STATION, MONTH ASC"""
+
+     # Fetch the result set for this query
+
+     # Your Google BigQuery Project ID
+     # To find this, see your dashboard:
+     # https://console.developers.google.com/iam-admin/projects?authuser=0
+     projectid = 'xxxxxxxxx'
+     df = gbq.read_gbq(query, project_id=projectid)
+
+     # Use pandas to process and reshape the dataset
+
+     df2 = df.pivot(index='STATION', columns='MONTH', values='MEAN_TEMP')
+     df3 = pd.concat([df2.min(), df2.mean(), df2.max()],
+                     axis=1, keys=["Min Tem", "Mean Temp", "Max Temp"])
+
+  The resulting DataFrame is::
+
+     > df3
+                 Min Tem  Mean Temp    Max Temp
+      MONTH
+      1     -53.336667  39.827892   89.770968
+      2     -49.837500  43.685219   93.437932
+      3     -77.926087  48.708355   96.099998
+      4     -82.892858  55.070087   97.317240
+      5     -92.378261  61.428117  102.042856
+      6     -77.703334  65.858888  102.900000
+      7     -87.821428  68.169663  106.510714
+      8     -89.431999  68.614215  105.500000
+      9     -86.611112  63.436935  107.142856
+      10    -78.209677  56.880838   92.103333
+      11    -50.125000  48.861228   94.996428
+      12    -50.332258  42.286879   94.396774
+
+  .. warning::
+
+     To use this module, you will need a BigQuery account. See
+     <https://cloud.google.com/products/big-query> for details.
+
+     As of 10/10/13, there is a bug in Google's API preventing result sets
+     from being larger than 100,000 rows. A patch is scheduled for the week of
+     10/14/13.
+
+.. _whatsnew_0130.refactoring:
+
+Internal refactoring
+~~~~~~~~~~~~~~~~~~~~
+
+In 0.13.0 there is a major refactor primarily to subclass ``Series`` from
+``NDFrame``, which is the base class currently for ``DataFrame`` and ``Panel``,
+to unify methods and behaviors. Series formerly subclassed directly from
+``ndarray``. (:issue:`4080`, :issue:`3862`, :issue:`816`)
+
+.. warning::
+
+   There are two potential incompatibilities from < 0.13.0
+
+   - Using certain numpy functions would previously return a ``Series`` if passed a ``Series``
+     as an argument. This seems only to affect ``np.ones_like``, ``np.empty_like``,
+     ``np.diff`` and ``np.where``. These now return ``ndarrays``.
+
+     .. ipython:: python
+
+        s = pd.Series([1, 2, 3, 4])
+
+     Numpy Usage
+
+     .. ipython:: python
+
+        np.ones_like(s)
+        np.diff(s)
+        np.where(s > 1, s, np.nan)
+
+     Pandonic Usage
+
+     .. ipython:: python
+
+        pd.Series(1, index=s.index)
+        s.diff()
+        s.where(s > 1)
+
+   - Passing a ``Series`` directly to a cython function expecting an ``ndarray`` type will no
+     long work directly, you must pass ``Series.values``, See :ref:`Enhancing Performance<enhancingperf.ndarray>`
+
+   - ``Series(0.5)`` would previously return the scalar ``0.5``, instead this will return a 1-element ``Series``
+
+   - This change breaks ``rpy2<=2.3.8``. an Issue has been opened against rpy2 and a workaround
+     is detailed in :issue:`5698`. Thanks @JanSchulz.
+
+- Pickle compatibility is preserved for pickles created prior to 0.13. These must be unpickled with ``pd.read_pickle``, see :ref:`Pickling<io.pickle>`.
+
+- Refactor of series.py/frame.py/panel.py to move common code to generic.py
+
+  - added ``_setup_axes`` to created generic NDFrame structures
+  - moved methods
+
+    - ``from_axes,_wrap_array,axes,ix,loc,iloc,shape,empty,swapaxes,transpose,pop``
+    - ``__iter__,keys,__contains__,__len__,__neg__,__invert__``
+    - ``convert_objects,as_blocks,as_matrix,values``
+    - ``__getstate__,__setstate__`` (compat remains in frame/panel)
+    - ``__getattr__,__setattr__``
+    - ``_indexed_same,reindex_like,align,where,mask``
+    - ``fillna,replace`` (``Series`` replace is now consistent with ``DataFrame``)
+    - ``filter`` (also added axis argument to selectively filter on a different axis)
+    - ``reindex,reindex_axis,take``
+    - ``truncate`` (moved to become part of ``NDFrame``)
+
+- These are API changes which make ``Panel`` more consistent with ``DataFrame``
+
+  - ``swapaxes`` on a ``Panel`` with the same axes specified now return a copy
+  - support attribute access for setting
+  - filter supports the same API as the original ``DataFrame`` filter
+
+- Reindex called with no arguments will now return a copy of the input object
+
+- ``TimeSeries`` is now an alias for ``Series``. the property ``is_time_series``
+  can be used to distinguish (if desired)
+
+- Refactor of Sparse objects to use BlockManager
+
+  - Created a new block type in internals, ``SparseBlock``, which can hold multi-dtypes
+    and is non-consolidatable. ``SparseSeries`` and ``SparseDataFrame`` now inherit
+    more methods from there hierarchy (Series/DataFrame), and no longer inherit
+    from ``SparseArray`` (which instead is the object of the ``SparseBlock``)
+  - Sparse suite now supports integration with non-sparse data. Non-float sparse
+    data is supportable (partially implemented)
+  - Operations on sparse structures within DataFrames should preserve sparseness,
+    merging type operations will convert to dense (and back to sparse), so might
+    be somewhat inefficient
+  - enable setitem on ``SparseSeries`` for boolean/integer/slices
+  - ``SparsePanels`` implementation is unchanged (e.g. not using BlockManager, needs work)
+
+- added ``ftypes`` method to Series/DataFrame, similar to ``dtypes``, but indicates
+  if the underlying is sparse/dense (as well as the dtype)
+- All ``NDFrame`` objects can now use ``__finalize__()`` to specify various
+  values to propagate to new objects from an existing one (e.g. ``name`` in ``Series`` will
+  follow more automatically now)
+- Internal type checking is now done via a suite of generated classes, allowing ``isinstance(value, klass)``
+  without having to directly import the klass, courtesy of @jtratner
+- Bug in Series update where the parent frame is not updating its cache based on
+  changes (:issue:`4080`) or types (:issue:`3217`), fillna (:issue:`3386`)
+- Indexing with dtype conversions fixed (:issue:`4463`, :issue:`4204`)
+- Refactor ``Series.reindex`` to core/generic.py (:issue:`4604`, :issue:`4618`), allow ``method=`` in reindexing
+  on a Series to work
+- ``Series.copy`` no longer accepts the ``order`` parameter and is now consistent with ``NDFrame`` copy
+- Refactor ``rename`` methods to core/generic.py; fixes ``Series.rename`` for (:issue:`4605`), and adds ``rename``
+  with the same signature for ``Panel``
+- Refactor ``clip`` methods to core/generic.py (:issue:`4798`)
+- Refactor of ``_get_numeric_data/_get_bool_data`` to core/generic.py, allowing Series/Panel functionality
+- ``Series`` (for index) / ``Panel`` (for items) now allow attribute access to its elements  (:issue:`1903`)
+
+  .. ipython:: python
+
+     s = pd.Series([1, 2, 3], index=list('abc'))
+     s.b
+     s.a = 5
+     s
+
+.. _release.bug_fixes-0.13.0:
+
+Bug fixes
+~~~~~~~~~
+
+- ``HDFStore``
+
+  - raising an invalid ``TypeError`` rather than ``ValueError`` when
+    appending with a different block ordering (:issue:`4096`)
+  - ``read_hdf`` was not respecting as passed ``mode`` (:issue:`4504`)
+  - appending a 0-len table will work correctly (:issue:`4273`)
+  - ``to_hdf`` was raising when passing both arguments ``append`` and
+    ``table`` (:issue:`4584`)
+  - reading from a store with duplicate columns across dtypes would raise
+    (:issue:`4767`)
+  - Fixed a bug where ``ValueError`` wasn't correctly raised when column
+    names weren't strings (:issue:`4956`)
+  - A zero length series written in Fixed format not deserializing properly.
+    (:issue:`4708`)
+  - Fixed decoding perf issue on pyt3 (:issue:`5441`)
+  - Validate levels in a MultiIndex before storing (:issue:`5527`)
+  - Correctly handle ``data_columns`` with a Panel (:issue:`5717`)
+- Fixed bug in tslib.tz_convert(vals, tz1, tz2): it could raise IndexError
+  exception while trying to access trans[pos + 1] (:issue:`4496`)
+- The ``by`` argument now works correctly with the ``layout`` argument
+  (:issue:`4102`, :issue:`4014`) in ``*.hist`` plotting methods
+- Fixed bug in ``PeriodIndex.map`` where using ``str`` would return the str
+  representation of the index (:issue:`4136`)
+- Fixed test failure ``test_time_series_plot_color_with_empty_kwargs`` when
+  using custom matplotlib default colors (:issue:`4345`)
+- Fix running of stata IO tests. Now uses temporary files to write
+  (:issue:`4353`)
+- Fixed an issue where ``DataFrame.sum`` was slower than ``DataFrame.mean``
+  for integer valued frames (:issue:`4365`)
+- ``read_html`` tests now work with Python 2.6 (:issue:`4351`)
+- Fixed bug where ``network`` testing was throwing ``NameError`` because a
+  local variable was undefined (:issue:`4381`)
+- In ``to_json``, raise if a passed ``orient`` would cause loss of data
+  because of a duplicate index (:issue:`4359`)
+- In ``to_json``, fix date handling so milliseconds are the default timestamp
+  as the docstring says (:issue:`4362`).
+- ``as_index`` is no longer ignored when doing groupby apply (:issue:`4648`,
+  :issue:`3417`)
+- JSON NaT handling fixed, NaTs are now serialized to ``null`` (:issue:`4498`)
+- Fixed JSON handling of escapable characters in JSON object keys
+  (:issue:`4593`)
+- Fixed passing ``keep_default_na=False`` when ``na_values=None``
+  (:issue:`4318`)
+- Fixed bug with ``values`` raising an error on a DataFrame with duplicate
+  columns and mixed dtypes, surfaced in (:issue:`4377`)
+- Fixed bug with duplicate columns and type conversion in ``read_json`` when
+  ``orient='split'`` (:issue:`4377`)
+- Fixed JSON bug where locales with decimal separators other than '.' threw
+  exceptions when encoding / decoding certain values. (:issue:`4918`)
+- Fix ``.iat`` indexing with a ``PeriodIndex`` (:issue:`4390`)
+- Fixed an issue where ``PeriodIndex`` joining with self was returning a new
+  instance rather than the same instance (:issue:`4379`); also adds a test
+  for this for the other index types
+- Fixed a bug with all the dtypes being converted to object when using the
+  CSV cparser with the usecols parameter (:issue:`3192`)
+- Fix an issue in merging blocks where the resulting DataFrame had partially
+  set _ref_locs (:issue:`4403`)
+- Fixed an issue where hist subplots were being overwritten when they were
+  called using the top level matplotlib API (:issue:`4408`)
+- Fixed a bug where calling ``Series.astype(str)`` would truncate the string
+  (:issue:`4405`, :issue:`4437`)
+- Fixed a py3 compat issue where bytes were being repr'd as tuples
+  (:issue:`4455`)
+- Fixed Panel attribute naming conflict if item is named 'a'
+  (:issue:`3440`)
+- Fixed an issue where duplicate indexes were raising when plotting
+  (:issue:`4486`)
+- Fixed an issue where cumsum and cumprod didn't work with bool dtypes
+  (:issue:`4170`, :issue:`4440`)
+- Fixed Panel slicing issued in ``xs`` that was returning an incorrect dimmed
+  object (:issue:`4016`)
+- Fix resampling bug where custom reduce function not used if only one group
+  (:issue:`3849`, :issue:`4494`)
+- Fixed Panel assignment with a transposed frame (:issue:`3830`)
+- Raise on set indexing with a Panel and a Panel as a value which needs
+  alignment (:issue:`3777`)
+- frozenset objects now raise in the ``Series`` constructor (:issue:`4482`,
+  :issue:`4480`)
+- Fixed issue with sorting a duplicate MultiIndex that has multiple dtypes
+  (:issue:`4516`)
+- Fixed bug in ``DataFrame.set_values`` which was causing name attributes to
+  be lost when expanding the index. (:issue:`3742`, :issue:`4039`)
+- Fixed issue where individual ``names``, ``levels`` and ``labels`` could be
+  set on ``MultiIndex`` without validation (:issue:`3714`, :issue:`4039`)
+- Fixed (:issue:`3334`) in pivot_table. Margins did not compute if values is
+  the index.
+- Fix bug in having a rhs of ``np.timedelta64`` or ``np.offsets.DateOffset``
+  when operating with datetimes (:issue:`4532`)
+- Fix arithmetic with series/datetimeindex and ``np.timedelta64`` not working
+  the same (:issue:`4134`) and buggy timedelta in NumPy 1.6 (:issue:`4135`)
+- Fix bug in ``pd.read_clipboard`` on windows with PY3 (:issue:`4561`); not
+  decoding properly
+- ``tslib.get_period_field()`` and ``tslib.get_period_field_arr()`` now raise
+  if code argument out of range (:issue:`4519`, :issue:`4520`)
+- Fix boolean indexing on an empty series loses index names (:issue:`4235`),
+  infer_dtype works with empty arrays.
+- Fix reindexing with multiple axes; if an axes match was not replacing the
+  current axes, leading to a possible lazy frequency inference issue
+  (:issue:`3317`)
+- Fixed issue where ``DataFrame.apply`` was reraising exceptions incorrectly
+  (causing the original stack trace to be truncated).
+- Fix selection with ``ix/loc`` and non_unique selectors (:issue:`4619`)
+- Fix assignment with iloc/loc involving a dtype change in an existing column
+  (:issue:`4312`, :issue:`5702`) have internal setitem_with_indexer in core/indexing
+  to use Block.setitem
+- Fixed bug where thousands operator was not handled correctly for floating
+  point numbers in csv_import (:issue:`4322`)
+- Fix an issue with CacheableOffset not properly being used by many
+  DateOffset; this prevented the DateOffset from being cached (:issue:`4609`)
+- Fix boolean comparison with a DataFrame on the lhs, and a list/tuple on the
+  rhs (:issue:`4576`)
+- Fix error/dtype conversion with setitem of ``None`` on ``Series/DataFrame``
+  (:issue:`4667`)
+- Fix decoding based on a passed in non-default encoding in ``pd.read_stata``
+  (:issue:`4626`)
+- Fix ``DataFrame.from_records`` with a plain-vanilla ``ndarray``.
+  (:issue:`4727`)
+- Fix some inconsistencies with ``Index.rename`` and ``MultiIndex.rename``,
+  etc. (:issue:`4718`, :issue:`4628`)
+- Bug in using ``iloc/loc`` with a cross-sectional and duplicate indices
+  (:issue:`4726`)
+- Bug with using ``QUOTE_NONE`` with ``to_csv`` causing ``Exception``.
+  (:issue:`4328`)
+- Bug with Series indexing not raising an error when the right-hand-side has
+  an incorrect length (:issue:`2702`)
+- Bug in MultiIndexing with a partial string selection as one part of a
+  MultIndex (:issue:`4758`)
+- Bug with reindexing on the index with a non-unique index will now raise
+  ``ValueError`` (:issue:`4746`)
+- Bug in setting with ``loc/ix`` a single indexer with a MultiIndex axis and
+  a NumPy array, related to (:issue:`3777`)
+- Bug in concatenation with duplicate columns across dtypes not merging with
+  axis=0 (:issue:`4771`, :issue:`4975`)
+- Bug in ``iloc`` with a slice index failing (:issue:`4771`)
+- Incorrect error message with no colspecs or width in ``read_fwf``.
+  (:issue:`4774`)
+- Fix bugs in indexing in a Series with a duplicate index (:issue:`4548`,
+  :issue:`4550`)
+- Fixed bug with reading compressed files with ``read_fwf`` in Python 3.
+  (:issue:`3963`)
+- Fixed an issue with a duplicate index and assignment with a dtype change
+  (:issue:`4686`)
+- Fixed bug with reading compressed files in as ``bytes`` rather than ``str``
+  in Python 3. Simplifies bytes-producing file-handling in Python 3
+  (:issue:`3963`, :issue:`4785`).
+- Fixed an issue related to ticklocs/ticklabels with log scale bar plots
+  across different versions of matplotlib (:issue:`4789`)
+- Suppressed DeprecationWarning associated with internal calls issued by
+  repr() (:issue:`4391`)
+- Fixed an issue with a duplicate index and duplicate selector with ``.loc``
+  (:issue:`4825`)
+- Fixed an issue with ``DataFrame.sort_index`` where, when sorting by a
+  single column and passing a list for ``ascending``, the argument for
+  ``ascending`` was being interpreted as ``True`` (:issue:`4839`,
+  :issue:`4846`)
+- Fixed ``Panel.tshift`` not working. Added ``freq`` support to ``Panel.shift``
+  (:issue:`4853`)
+- Fix an issue in TextFileReader w/ Python engine (i.e. PythonParser)
+  with thousands != "," (:issue:`4596`)
+- Bug in getitem with a duplicate index when using where (:issue:`4879`)
+- Fix Type inference code coerces float column into datetime (:issue:`4601`)
+- Fixed ``_ensure_numeric`` does not check for complex numbers
+  (:issue:`4902`)
+- Fixed a bug in ``Series.hist`` where two figures were being created when
+  the ``by`` argument was passed (:issue:`4112`, :issue:`4113`).
+- Fixed a bug in ``convert_objects`` for > 2 ndims (:issue:`4937`)
+- Fixed a bug in DataFrame/Panel cache insertion and subsequent indexing
+  (:issue:`4939`, :issue:`5424`)
+- Fixed string methods for ``FrozenNDArray`` and ``FrozenList``
+  (:issue:`4929`)
+- Fixed a bug with setting invalid or out-of-range values in indexing
+  enlargement scenarios (:issue:`4940`)
+- Tests for fillna on empty Series (:issue:`4346`), thanks @immerrr
+- Fixed ``copy()`` to shallow copy axes/indices as well and thereby keep
+  separate metadata. (:issue:`4202`, :issue:`4830`)
+- Fixed skiprows option in Python parser for read_csv (:issue:`4382`)
+- Fixed bug preventing ``cut`` from working with ``np.inf`` levels without
+  explicitly passing labels (:issue:`3415`)
+- Fixed wrong check for overlapping in ``DatetimeIndex.union``
+  (:issue:`4564`)
+- Fixed conflict between thousands separator and date parser in csv_parser
+  (:issue:`4678`)
+- Fix appending when dtypes are not the same (error showing mixing
+  float/np.datetime64) (:issue:`4993`)
+- Fix repr for DateOffset. No longer show duplicate entries in kwds.
+  Removed unused offset fields. (:issue:`4638`)
+- Fixed wrong index name during read_csv if using usecols. Applies to c
+  parser only. (:issue:`4201`)
+- ``Timestamp`` objects can now appear in the left hand side of a comparison
+  operation with a ``Series`` or ``DataFrame`` object (:issue:`4982`).
+- Fix a bug when indexing with ``np.nan`` via ``iloc/loc`` (:issue:`5016`)
+- Fixed a bug where low memory c parser could create different types in
+  different chunks of the same file. Now coerces to numerical type or raises
+  warning. (:issue:`3866`)
+- Fix a bug where reshaping a ``Series`` to its own shape raised
+  ``TypeError`` (:issue:`4554`) and other reshaping issues.
+- Bug in setting with ``ix/loc`` and a mixed int/string index (:issue:`4544`)
+- Make sure series-series boolean comparisons are label based (:issue:`4947`)
+- Bug in multi-level indexing with a Timestamp partial indexer
+  (:issue:`4294`)
+- Tests/fix for MultiIndex construction of an all-nan frame (:issue:`4078`)
+- Fixed a bug where :func:`~pandas.read_html` wasn't correctly inferring
+  values of tables with commas (:issue:`5029`)
+- Fixed a bug where :func:`~pandas.read_html` wasn't providing a stable
+  ordering of returned tables (:issue:`4770`, :issue:`5029`).
+- Fixed a bug where :func:`~pandas.read_html` was incorrectly parsing when
+  passed ``index_col=0`` (:issue:`5066`).
+- Fixed a bug where :func:`~pandas.read_html` was incorrectly inferring the
+  type of headers (:issue:`5048`).
+- Fixed a bug where ``DatetimeIndex`` joins with ``PeriodIndex`` caused a
+  stack overflow (:issue:`3899`).
+- Fixed a bug where ``groupby`` objects didn't allow plots (:issue:`5102`).
+- Fixed a bug where ``groupby`` objects weren't tab-completing column names
+  (:issue:`5102`).
+- Fixed a bug where ``groupby.plot()`` and friends were duplicating figures
+  multiple times (:issue:`5102`).
+- Provide automatic conversion of ``object`` dtypes on fillna, related
+  (:issue:`5103`)
+- Fixed a bug where default options were being overwritten in the option
+  parser cleaning (:issue:`5121`).
+- Treat a list/ndarray identically for ``iloc`` indexing with list-like
+  (:issue:`5006`)
+- Fix ``MultiIndex.get_level_values()`` with missing values (:issue:`5074`)
+- Fix bound checking for Timestamp() with datetime64 input (:issue:`4065`)
+- Fix a bug where ``TestReadHtml`` wasn't calling the correct ``read_html()``
+  function (:issue:`5150`).
+- Fix a bug with ``NDFrame.replace()`` which made replacement appear as
+  though it was (incorrectly) using regular expressions (:issue:`5143`).
+- Fix better error message for to_datetime (:issue:`4928`)
+- Made sure different locales are tested on travis-ci (:issue:`4918`). Also
+  adds a couple of utilities for getting locales and setting locales with a
+  context manager.
+- Fixed segfault on ``isnull(MultiIndex)`` (now raises an error instead)
+  (:issue:`5123`, :issue:`5125`)
+- Allow duplicate indices when performing operations that align
+  (:issue:`5185`, :issue:`5639`)
+- Compound dtypes in a constructor raise ``NotImplementedError``
+  (:issue:`5191`)
+- Bug in comparing duplicate frames (:issue:`4421`) related
+- Bug in describe on duplicate frames
+- Bug in ``to_datetime`` with a format and ``coerce=True`` not raising
+  (:issue:`5195`)
+- Bug in ``loc`` setting with multiple indexers and a rhs of a Series that
+  needs broadcasting (:issue:`5206`)
+- Fixed bug where inplace setting of levels or labels on ``MultiIndex`` would
+  not clear cached ``values`` property and therefore return wrong ``values``.
+  (:issue:`5215`)
+- Fixed bug where filtering a grouped DataFrame or Series did not maintain
+  the original ordering (:issue:`4621`).
+- Fixed ``Period`` with a business date freq to always roll-forward if on a
+  non-business date. (:issue:`5203`)
+- Fixed bug in Excel writers where frames with duplicate column names weren't
+  written correctly. (:issue:`5235`)
+- Fixed issue with ``drop`` and a non-unique index on Series (:issue:`5248`)
+- Fixed segfault in C parser caused by passing more names than columns in
+  the file. (:issue:`5156`)
+- Fix ``Series.isin`` with date/time-like dtypes (:issue:`5021`)
+- C and Python Parser can now handle the more common MultiIndex column
+  format which doesn't have a row for index names (:issue:`4702`)
+- Bug when trying to use an out-of-bounds date as an object dtype
+  (:issue:`5312`)
+- Bug when trying to display an embedded PandasObject (:issue:`5324`)
+- Allows operating of Timestamps to return a datetime if the result is out-of-bounds
+  related (:issue:`5312`)
+- Fix return value/type signature of ``initObjToJSON()`` to be compatible
+  with numpy's ``import_array()`` (:issue:`5334`, :issue:`5326`)
+- Bug when renaming then set_index on a DataFrame (:issue:`5344`)
+- Test suite no longer leaves around temporary files when testing graphics. (:issue:`5347`)
+  (thanks for catching this @yarikoptic!)
+- Fixed html tests on win32. (:issue:`4580`)
+- Make sure that ``head/tail`` are ``iloc`` based, (:issue:`5370`)
+- Fixed bug for ``PeriodIndex`` string representation if there are 1 or 2
+  elements. (:issue:`5372`)
+- The GroupBy methods ``transform`` and ``filter`` can be used on Series
+  and DataFrames that have repeated (non-unique) indices. (:issue:`4620`)
+- Fix empty series not printing name in repr (:issue:`4651`)
+- Make tests create temp files in temp directory by default. (:issue:`5419`)
+- ``pd.to_timedelta`` of a scalar returns a scalar (:issue:`5410`)
+- ``pd.to_timedelta`` accepts ``NaN`` and ``NaT``, returning ``NaT`` instead of raising (:issue:`5437`)
+- performance improvements in ``isnull`` on larger size pandas objects
+- Fixed various setitem with 1d ndarray that does not have a matching
+  length to the indexer (:issue:`5508`)
+- Bug in getitem with a MultiIndex and ``iloc`` (:issue:`5528`)
+- Bug in delitem on a Series (:issue:`5542`)
+- Bug fix in apply when using custom function and objects are not mutated (:issue:`5545`)
+- Bug in selecting from a non-unique index with ``loc`` (:issue:`5553`)
+- Bug in groupby returning non-consistent types when user function returns a ``None``, (:issue:`5592`)
+- Work around regression in numpy 1.7.0 which erroneously raises IndexError from ``ndarray.item`` (:issue:`5666`)
+- Bug in repeated indexing of object with resultant non-unique index (:issue:`5678`)
+- Bug in fillna with Series and a passed series/dict (:issue:`5703`)
+- Bug in groupby transform with a datetime-like grouper (:issue:`5712`)
+- Bug in MultiIndex selection in PY3 when using certain keys (:issue:`5725`)
+- Row-wise concat of differing dtypes failing in certain cases (:issue:`5754`)
+
+.. _whatsnew_0.13.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.12.0..v0.13.0
diff --git a/pandas/doc/source/whatsnew/v0.13.1.rst b/pandas/doc/source/whatsnew/v0.13.1.rst
new file mode 100644
index 0000000000000..483dd15a8467a
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.13.1.rst
@@ -0,0 +1,481 @@
+.. _whatsnew_0131:
+
+Version 0.13.1 (February 3, 2014)
+---------------------------------
+
+{{ header }}
+
+
+
+This is a minor release from 0.13.0 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+Highlights include:
+
+- Added ``infer_datetime_format`` keyword to ``read_csv/to_datetime`` to allow speedups for homogeneously formatted datetimes.
+- Will intelligently limit display precision for datetime/timedelta formats.
+- Enhanced Panel :meth:`~pandas.Panel.apply` method.
+- Suggested tutorials in new :ref:`Tutorials<tutorials>` section.
+- Our pandas ecosystem is growing, We now feature related projects in a new `ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_ section.
+- Much work has been taking place on improving the docs, and a new :ref:`Contributing<contributing>` section has been added.
+- Even though it may only be of interest to devs, we <3 our new CI status page: `ScatterCI <http://scatterci.github.io/pydata/pandas>`__.
+
+.. warning::
+
+   0.13.1 fixes a bug that was caused by a combination of having numpy < 1.8, and doing
+   chained assignment on a string-like array.
+   Chained indexing can have unexpected results and should generally be avoided.
+
+   This would previously segfault:
+
+   .. code-block:: python
+
+      df = pd.DataFrame({"A": np.array(["foo", "bar", "bah", "foo", "bar"])})
+      df["A"].iloc[0] = np.nan
+
+   The recommended way to do this type of assignment is:
+
+   .. ipython:: python
+
+      df = pd.DataFrame({"A": np.array(["foo", "bar", "bah", "foo", "bar"])})
+      df.loc[0, "A"] = np.nan
+      df
+
+Output formatting enhancements
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- df.info() view now display dtype info per column (:issue:`5682`)
+
+- df.info() now honors the option ``max_info_rows``, to disable null counts for large frames (:issue:`5974`)
+
+  .. ipython:: python
+
+     max_info_rows = pd.get_option("max_info_rows")
+
+     df = pd.DataFrame(
+         {
+             "A": np.random.randn(10),
+             "B": np.random.randn(10),
+             "C": pd.date_range("20130101", periods=10),
+         }
+     )
+     df.iloc[3:6, [0, 2]] = np.nan
+
+  .. ipython:: python
+
+     # set to not display the null counts
+     pd.set_option("max_info_rows", 0)
+     df.info()
+
+  .. ipython:: python
+
+     # this is the default (same as in 0.13.0)
+     pd.set_option("max_info_rows", max_info_rows)
+     df.info()
+
+- Add ``show_dimensions`` display option for the new DataFrame repr to control whether the dimensions print.
+
+  .. ipython:: python
+
+      df = pd.DataFrame([[1, 2], [3, 4]])
+      pd.set_option("show_dimensions", False)
+      df
+
+      pd.set_option("show_dimensions", True)
+      df
+
+- The ``ArrayFormatter`` for ``datetime`` and ``timedelta64`` now intelligently
+  limit precision based on the values in the array (:issue:`3401`)
+
+  Previously output might look like:
+
+  ..   code-block:: text
+
+        age                 today               diff
+      0 2001-01-01 00:00:00 2013-04-19 00:00:00 4491 days, 00:00:00
+      1 2004-06-01 00:00:00 2013-04-19 00:00:00 3244 days, 00:00:00
+
+  Now the output looks like:
+
+  .. ipython:: python
+
+     df = pd.DataFrame(
+         [pd.Timestamp("20010101"), pd.Timestamp("20040601")], columns=["age"]
+     )
+     df["today"] = pd.Timestamp("20130419")
+     df["diff"] = df["today"] - df["age"]
+     df
+
+API changes
+~~~~~~~~~~~
+
+- Add ``-NaN`` and ``-nan`` to the default set of NA values (:issue:`5952`).
+  See :ref:`NA Values <io.na_values>`.
+
+- Added ``Series.str.get_dummies`` vectorized string method (:issue:`6021`), to extract
+  dummy/indicator variables for separated string columns:
+
+  .. ipython:: python
+
+      s = pd.Series(["a", "a|b", np.nan, "a|c"])
+      s.str.get_dummies(sep="|")
+
+- Added the ``NDFrame.equals()`` method to compare if two NDFrames are
+  equal have equal axes, dtypes, and values. Added the
+  ``array_equivalent`` function to compare if two ndarrays are
+  equal. NaNs in identical locations are treated as
+  equal. (:issue:`5283`) See also :ref:`the docs<basics.equals>` for a motivating example.
+
+  .. code-block:: python
+
+      df = pd.DataFrame({"col": ["foo", 0, np.nan]})
+      df2 = pd.DataFrame({"col": [np.nan, 0, "foo"]}, index=[2, 1, 0])
+      df.equals(df2)
+      df.equals(df2.sort_index())
+
+- ``DataFrame.apply`` will use the ``reduce`` argument to determine whether a
+  ``Series`` or a ``DataFrame`` should be returned when the ``DataFrame`` is
+  empty (:issue:`6007`).
+
+  Previously, calling ``DataFrame.apply`` an empty ``DataFrame`` would return
+  either a ``DataFrame`` if there were no columns, or the function being
+  applied would be called with an empty ``Series`` to guess whether a
+  ``Series`` or ``DataFrame`` should be returned:
+
+  .. code-block:: ipython
+
+    In [32]: def applied_func(col):
+      ....:    print("Apply function being called with: ", col)
+      ....:    return col.sum()
+      ....:
+
+    In [33]: empty = DataFrame(columns=['a', 'b'])
+
+    In [34]: empty.apply(applied_func)
+    Apply function being called with:  Series([], Length: 0, dtype: float64)
+    Out[34]:
+    a   NaN
+    b   NaN
+    Length: 2, dtype: float64
+
+  Now, when ``apply`` is called on an empty ``DataFrame``: if the ``reduce``
+  argument is ``True`` a ``Series`` will returned, if it is ``False`` a
+  ``DataFrame`` will be returned, and if it is ``None`` (the default) the
+  function being applied will be called with an empty series to try and guess
+  the return type.
+
+  .. code-block:: ipython
+
+    In [35]: empty.apply(applied_func, reduce=True)
+    Out[35]:
+    a   NaN
+    b   NaN
+    Length: 2, dtype: float64
+
+    In [36]: empty.apply(applied_func, reduce=False)
+    Out[36]:
+    Empty DataFrame
+    Columns: [a, b]
+    Index: []
+
+    [0 rows x 2 columns]
+
+
+Prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+There are no announced changes in 0.13 or prior that are taking effect as of 0.13.1
+
+Deprecations
+~~~~~~~~~~~~
+
+There are no deprecations of prior behavior in 0.13.1
+
+Enhancements
+~~~~~~~~~~~~
+
+- ``pd.read_csv`` and ``pd.to_datetime`` learned a new ``infer_datetime_format`` keyword which greatly
+  improves parsing perf in many cases. Thanks to @lexual for suggesting and @danbirken
+  for rapidly implementing. (:issue:`5490`, :issue:`6021`)
+
+  If ``parse_dates`` is enabled and this flag is set, pandas will attempt to
+  infer the format of the datetime strings in the columns, and if it can
+  be inferred, switch to a faster method of parsing them.  In some cases
+  this can increase the parsing speed by ~5-10x.
+
+  .. code-block:: python
+
+      # Try to infer the format for the index column
+      df = pd.read_csv(
+          "foo.csv", index_col=0, parse_dates=True, infer_datetime_format=True
+      )
+
+- ``date_format`` and ``datetime_format`` keywords can now be specified when writing to ``excel``
+  files (:issue:`4133`)
+
+- ``MultiIndex.from_product`` convenience function for creating a MultiIndex from
+  the cartesian product of a set of iterables (:issue:`6055`):
+
+  .. ipython:: python
+
+      shades = ["light", "dark"]
+      colors = ["red", "green", "blue"]
+
+      pd.MultiIndex.from_product([shades, colors], names=["shade", "color"])
+
+- Panel :meth:`~pandas.Panel.apply` will work on non-ufuncs. See :ref:`the docs<basics.apply>`.
+
+  .. code-block:: ipython
+
+      In [28]: import pandas._testing as tm
+
+      In [29]: panel = tm.makePanel(5)
+
+      In [30]: panel
+      Out[30]:
+      <class 'pandas.core.panel.Panel'>
+      Dimensions: 3 (items) x 5 (major_axis) x 4 (minor_axis)
+      Items axis: ItemA to ItemC
+      Major_axis axis: 2000-01-03 00:00:00 to 2000-01-07 00:00:00
+      Minor_axis axis: A to D
+
+      In [31]: panel['ItemA']
+      Out[31]:
+                         A         B         C         D
+      2000-01-03 -0.673690  0.577046 -1.344312 -1.469388
+      2000-01-04  0.113648 -1.715002  0.844885  0.357021
+      2000-01-05 -1.478427 -1.039268  1.075770 -0.674600
+      2000-01-06  0.524988 -0.370647 -0.109050 -1.776904
+      2000-01-07  0.404705 -1.157892  1.643563 -0.968914
+
+      [5 rows x 4 columns]
+
+  Specifying an ``apply`` that operates on a Series (to return a single element)
+
+  .. code-block:: ipython
+
+      In [32]: panel.apply(lambda x: x.dtype, axis='items')
+      Out[32]:
+                        A        B        C        D
+      2000-01-03  float64  float64  float64  float64
+      2000-01-04  float64  float64  float64  float64
+      2000-01-05  float64  float64  float64  float64
+      2000-01-06  float64  float64  float64  float64
+      2000-01-07  float64  float64  float64  float64
+
+      [5 rows x 4 columns]
+
+  A similar reduction type operation
+
+  .. code-block:: ipython
+
+      In [33]: panel.apply(lambda x: x.sum(), axis='major_axis')
+      Out[33]:
+            ItemA     ItemB     ItemC
+      A -1.108775 -1.090118 -2.984435
+      B -3.705764  0.409204  1.866240
+      C  2.110856  2.960500 -0.974967
+      D -4.532785  0.303202 -3.685193
+
+      [4 rows x 3 columns]
+
+  This is equivalent to
+
+  .. code-block:: ipython
+
+      In [34]: panel.sum('major_axis')
+      Out[34]:
+            ItemA     ItemB     ItemC
+      A -1.108775 -1.090118 -2.984435
+      B -3.705764  0.409204  1.866240
+      C  2.110856  2.960500 -0.974967
+      D -4.532785  0.303202 -3.685193
+
+      [4 rows x 3 columns]
+
+  A transformation operation that returns a Panel, but is computing
+  the z-score across the major_axis
+
+  .. code-block:: ipython
+
+      In [35]: result = panel.apply(lambda x: (x - x.mean()) / x.std(),
+        ....:                      axis='major_axis')
+        ....:
+
+      In [36]: result
+      Out[36]:
+      <class 'pandas.core.panel.Panel'>
+      Dimensions: 3 (items) x 5 (major_axis) x 4 (minor_axis)
+      Items axis: ItemA to ItemC
+      Major_axis axis: 2000-01-03 00:00:00 to 2000-01-07 00:00:00
+      Minor_axis axis: A to D
+
+      In [37]: result['ItemA']                           # noqa E999
+      Out[37]:
+                        A         B         C         D
+      2000-01-03 -0.535778  1.500802 -1.506416 -0.681456
+      2000-01-04  0.397628 -1.108752  0.360481  1.529895
+      2000-01-05 -1.489811 -0.339412  0.557374  0.280845
+      2000-01-06  0.885279  0.421830 -0.453013 -1.053785
+      2000-01-07  0.742682 -0.474468  1.041575 -0.075499
+
+      [5 rows x 4 columns]
+
+- Panel :meth:`~pandas.Panel.apply` operating on cross-sectional slabs. (:issue:`1148`)
+
+  .. code-block:: ipython
+
+      In [38]: def f(x):
+         ....:     return ((x.T - x.mean(1)) / x.std(1)).T
+         ....:
+
+      In [39]: result = panel.apply(f, axis=['items', 'major_axis'])
+
+      In [40]: result
+      Out[40]:
+      <class 'pandas.core.panel.Panel'>
+      Dimensions: 4 (items) x 5 (major_axis) x 3 (minor_axis)
+      Items axis: A to D
+      Major_axis axis: 2000-01-03 00:00:00 to 2000-01-07 00:00:00
+      Minor_axis axis: ItemA to ItemC
+
+      In [41]: result.loc[:, :, 'ItemA']
+      Out[41]:
+                         A         B         C         D
+      2000-01-03  0.012922 -0.030874 -0.629546 -0.757034
+      2000-01-04  0.392053 -1.071665  0.163228  0.548188
+      2000-01-05 -1.093650 -0.640898  0.385734 -1.154310
+      2000-01-06  1.005446 -1.154593 -0.595615 -0.809185
+      2000-01-07  0.783051 -0.198053  0.919339 -1.052721
+
+      [5 rows x 4 columns]
+
+  This is equivalent to the following
+
+  .. code-block:: ipython
+
+      In [42]: result = pd.Panel({ax: f(panel.loc[:, :, ax]) for ax in panel.minor_axis})
+
+      In [43]: result
+      Out[43]:
+      <class 'pandas.core.panel.Panel'>
+      Dimensions: 4 (items) x 5 (major_axis) x 3 (minor_axis)
+      Items axis: A to D
+      Major_axis axis: 2000-01-03 00:00:00 to 2000-01-07 00:00:00
+      Minor_axis axis: ItemA to ItemC
+
+      In [44]: result.loc[:, :, 'ItemA']
+      Out[44]:
+                         A         B         C         D
+      2000-01-03  0.012922 -0.030874 -0.629546 -0.757034
+      2000-01-04  0.392053 -1.071665  0.163228  0.548188
+      2000-01-05 -1.093650 -0.640898  0.385734 -1.154310
+      2000-01-06  1.005446 -1.154593 -0.595615 -0.809185
+      2000-01-07  0.783051 -0.198053  0.919339 -1.052721
+
+      [5 rows x 4 columns]
+
+Performance
+~~~~~~~~~~~
+
+Performance improvements for 0.13.1
+
+- Series datetime/timedelta binary operations (:issue:`5801`)
+- DataFrame ``count/dropna`` for ``axis=1``
+- Series.str.contains now has a ``regex=False`` keyword which can be faster for plain (non-regex) string patterns. (:issue:`5879`)
+- Series.str.extract (:issue:`5944`)
+- ``dtypes/ftypes`` methods (:issue:`5968`)
+- indexing with object dtypes (:issue:`5968`)
+- ``DataFrame.apply`` (:issue:`6013`)
+- Regression in JSON IO (:issue:`5765`)
+- Index construction from Series (:issue:`6150`)
+
+Experimental
+~~~~~~~~~~~~
+
+There are no experimental changes in 0.13.1
+
+.. _release.bug_fixes-0.13.1:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in ``io.wb.get_countries`` not including all countries (:issue:`6008`)
+- Bug in Series replace with timestamp dict (:issue:`5797`)
+- read_csv/read_table now respects the ``prefix`` kwarg (:issue:`5732`).
+- Bug in selection with missing values via ``.ix`` from a duplicate indexed DataFrame failing (:issue:`5835`)
+- Fix issue of boolean comparison on empty DataFrames (:issue:`5808`)
+- Bug in isnull handling ``NaT`` in an object array (:issue:`5443`)
+- Bug in ``to_datetime`` when passed a ``np.nan`` or integer datelike and a format string (:issue:`5863`)
+- Bug in groupby dtype conversion with datetimelike (:issue:`5869`)
+- Regression in handling of empty Series as indexers to Series  (:issue:`5877`)
+- Bug in internal caching, related to (:issue:`5727`)
+- Testing bug in reading JSON/msgpack from a non-filepath on windows under py3 (:issue:`5874`)
+- Bug when assigning to .ix[tuple(...)] (:issue:`5896`)
+- Bug in fully reindexing a Panel (:issue:`5905`)
+- Bug in idxmin/max with object dtypes (:issue:`5914`)
+- Bug in ``BusinessDay`` when adding n days to a date not on offset when n>5 and n%5==0 (:issue:`5890`)
+- Bug in assigning to chained series with a series via ix (:issue:`5928`)
+- Bug in creating an empty DataFrame, copying, then assigning (:issue:`5932`)
+- Bug in DataFrame.tail with empty frame (:issue:`5846`)
+- Bug in propagating metadata on ``resample`` (:issue:`5862`)
+- Fixed string-representation of ``NaT`` to be "NaT" (:issue:`5708`)
+- Fixed string-representation for Timestamp to show nanoseconds if present (:issue:`5912`)
+- ``pd.match`` not returning passed sentinel
+- ``Panel.to_frame()`` no longer fails when ``major_axis`` is a
+  ``MultiIndex`` (:issue:`5402`).
+- Bug in ``pd.read_msgpack`` with inferring a ``DateTimeIndex`` frequency
+  incorrectly (:issue:`5947`)
+- Fixed ``to_datetime`` for array with both Tz-aware datetimes and ``NaT``'s  (:issue:`5961`)
+- Bug in rolling skew/kurtosis when passed a Series with bad data (:issue:`5749`)
+- Bug in scipy ``interpolate`` methods with a datetime index (:issue:`5975`)
+- Bug in NaT comparison if a mixed datetime/np.datetime64 with NaT were passed (:issue:`5968`)
+- Fixed bug with ``pd.concat`` losing dtype information if all inputs are empty (:issue:`5742`)
+- Recent changes in IPython cause warnings to be emitted when using previous versions
+  of pandas in QTConsole, now fixed. If you're using an older version and
+  need to suppress the warnings, see (:issue:`5922`).
+- Bug in merging ``timedelta`` dtypes (:issue:`5695`)
+- Bug in plotting.scatter_matrix function. Wrong alignment among diagonal
+  and off-diagonal plots, see (:issue:`5497`).
+- Regression in Series with a MultiIndex via ix (:issue:`6018`)
+- Bug in Series.xs with a MultiIndex (:issue:`6018`)
+- Bug in Series construction of mixed type with datelike and an integer (which should result in
+  object type and not automatic conversion) (:issue:`6028`)
+- Possible segfault when chained indexing with an object array under NumPy 1.7.1 (:issue:`6026`, :issue:`6056`)
+- Bug in setting using fancy indexing a single element with a non-scalar (e.g. a list),
+  (:issue:`6043`)
+- ``to_sql`` did not respect ``if_exists`` (:issue:`4110` :issue:`4304`)
+- Regression in ``.get(None)`` indexing from 0.12 (:issue:`5652`)
+- Subtle ``iloc`` indexing bug, surfaced in (:issue:`6059`)
+- Bug with insert of strings into DatetimeIndex (:issue:`5818`)
+- Fixed unicode bug in to_html/HTML repr (:issue:`6098`)
+- Fixed missing arg validation in get_options_data (:issue:`6105`)
+- Bug in assignment with duplicate columns in a frame where the locations
+  are a slice (e.g. next to each other) (:issue:`6120`)
+- Bug in propagating _ref_locs during construction of a DataFrame with dups
+  index/columns (:issue:`6121`)
+- Bug in ``DataFrame.apply`` when using mixed datelike reductions (:issue:`6125`)
+- Bug in ``DataFrame.append`` when appending a row with different columns (:issue:`6129`)
+- Bug in DataFrame construction with recarray and non-ns datetime dtype (:issue:`6140`)
+- Bug in ``.loc`` setitem indexing with a dataframe on rhs, multiple item setting, and
+  a datetimelike (:issue:`6152`)
+- Fixed a bug in ``query``/``eval`` during lexicographic string comparisons (:issue:`6155`).
+- Fixed a bug in ``query`` where the index of a single-element ``Series`` was
+  being thrown away (:issue:`6148`).
+- Bug in ``HDFStore`` on appending a dataframe with MultiIndexed columns to
+  an existing table (:issue:`6167`)
+- Consistency with dtypes in setting an empty DataFrame (:issue:`6171`)
+- Bug in selecting on a MultiIndex ``HDFStore`` even in the presence of under
+  specified column spec (:issue:`6169`)
+- Bug in ``nanops.var`` with ``ddof=1`` and 1 elements would sometimes return ``inf``
+  rather than ``nan`` on some platforms (:issue:`6136`)
+- Bug in Series and DataFrame bar plots ignoring the ``use_index`` keyword (:issue:`6209`)
+- Bug in groupby with mixed str/int under python3 fixed; ``argsort`` was failing (:issue:`6212`)
+
+.. _whatsnew_0.13.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.13.0..v0.13.1
diff --git a/pandas/doc/source/whatsnew/v0.14.0.rst b/pandas/doc/source/whatsnew/v0.14.0.rst
new file mode 100644
index 0000000000000..05831197936eb
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.14.0.rst
@@ -0,0 +1,1143 @@
+.. _whatsnew_0140:
+
+Version 0.14.0 (May 31 , 2014)
+------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.13.1 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+- Highlights include:
+
+  - Officially support Python 3.4
+  - SQL interfaces updated to use ``sqlalchemy``, See :ref:`Here<whatsnew_0140.sql>`.
+  - Display interface changes, See :ref:`Here<whatsnew_0140.display>`
+  - MultiIndexing Using Slicers, See :ref:`Here<whatsnew_0140.slicers>`.
+  - Ability to join a singly-indexed DataFrame with a MultiIndexed DataFrame, see :ref:`Here <merging.join_on_mi>`
+  - More consistency in groupby results and more flexible groupby specifications, See :ref:`Here<whatsnew_0140.groupby>`
+  - Holiday calendars are now supported in ``CustomBusinessDay``, see :ref:`Here <timeseries.holiday>`
+  - Several improvements in plotting functions, including: hexbin, area and pie plots, see :ref:`Here<whatsnew_0140.plotting>`.
+  - Performance doc section on I/O operations, See :ref:`Here <io.perf>`
+
+- :ref:`Other Enhancements <whatsnew_0140.enhancements>`
+
+- :ref:`API Changes <whatsnew_0140.api>`
+
+- :ref:`Text Parsing API Changes <whatsnew_0140.parsing>`
+
+- :ref:`Groupby API Changes <whatsnew_0140.groupby>`
+
+- :ref:`Performance Improvements <whatsnew_0140.performance>`
+
+- :ref:`Prior Deprecations <whatsnew_0140.prior_deprecations>`
+
+- :ref:`Deprecations <whatsnew_0140.deprecations>`
+
+- :ref:`Known Issues <whatsnew_0140.knownissues>`
+
+- :ref:`Bug Fixes <whatsnew_0140.bug_fixes>`
+
+.. warning::
+
+   In 0.14.0 all ``NDFrame`` based containers have undergone significant internal refactoring. Before that each block of
+   homogeneous data had its own labels and extra care was necessary to keep those in sync with the parent container's labels.
+   This should not have any visible user/API behavior changes (:issue:`6745`)
+
+.. _whatsnew_0140.api:
+
+API changes
+~~~~~~~~~~~
+
+- ``read_excel`` uses 0 as the default sheet (:issue:`6573`)
+- ``iloc`` will now accept out-of-bounds indexers for slices, e.g. a value that exceeds the length of the object being
+  indexed. These will be excluded. This will make pandas conform more with python/numpy indexing of out-of-bounds
+  values. A single indexer that is out-of-bounds and drops the dimensions of the object will still raise
+  ``IndexError`` (:issue:`6296`, :issue:`6299`). This could result in an empty axis (e.g. an empty DataFrame being returned)
+
+  .. ipython:: python
+
+     dfl = pd.DataFrame(np.random.randn(5, 2), columns=list('AB'))
+     dfl
+     dfl.iloc[:, 2:3]
+     dfl.iloc[:, 1:3]
+     dfl.iloc[4:6]
+
+  These are out-of-bounds selections
+
+  .. code-block:: python
+
+     >>> dfl.iloc[[4, 5, 6]]
+     IndexError: positional indexers are out-of-bounds
+
+     >>> dfl.iloc[:, 4]
+     IndexError: single positional indexer is out-of-bounds
+
+- Slicing with negative start, stop & step values handles corner cases better (:issue:`6531`):
+
+  - ``df.iloc[:-len(df)]`` is now empty
+  - ``df.iloc[len(df)::-1]`` now enumerates all elements in reverse
+
+- The :meth:`DataFrame.interpolate` keyword ``downcast`` default has been changed from ``infer`` to
+  ``None``. This is to preserve the original dtype unless explicitly requested otherwise (:issue:`6290`).
+- When converting a dataframe to HTML it used to return ``Empty DataFrame``. This special case has
+  been removed, instead a header with the column names is returned (:issue:`6062`).
+- ``Series`` and ``Index`` now internally share more common operations, e.g. ``factorize(),nunique(),value_counts()`` are
+  now supported on ``Index`` types as well. The ``Series.weekday`` property from is removed
+  from Series for API consistency. Using a ``DatetimeIndex/PeriodIndex`` method on a Series will now raise a ``TypeError``.
+  (:issue:`4551`, :issue:`4056`, :issue:`5519`, :issue:`6380`, :issue:`7206`).
+
+- Add ``is_month_start``, ``is_month_end``, ``is_quarter_start``, ``is_quarter_end``, ``is_year_start``, ``is_year_end`` accessors for ``DateTimeIndex`` / ``Timestamp`` which return a boolean array of whether the timestamp(s) are at the start/end of the month/quarter/year defined by the frequency of the ``DateTimeIndex`` / ``Timestamp`` (:issue:`4565`, :issue:`6998`)
+
+- Local variable usage has changed in
+  :func:`pandas.eval`/:meth:`DataFrame.eval`/:meth:`DataFrame.query`
+  (:issue:`5987`). For the :class:`~pandas.DataFrame` methods, two things have
+  changed
+
+  - Column names are now given precedence over locals
+  - Local variables must be referred to explicitly. This means that even if
+    you have a local variable that is *not* a column you must still refer to
+    it with the ``'@'`` prefix.
+  - You can have an expression like ``df.query('@a < a')`` with no complaints
+    from ``pandas`` about ambiguity of the name ``a``.
+  - The top-level :func:`pandas.eval` function does not allow you use the
+    ``'@'`` prefix and provides you with an error message telling you so.
+  - ``NameResolutionError`` was removed because it isn't necessary anymore.
+
+- Define and document the order of column vs index names in query/eval (:issue:`6676`)
+- ``concat`` will now concatenate mixed Series and DataFrames using the Series name
+  or numbering columns as needed (:issue:`2385`). See :ref:`the docs <merging.mixed_ndims>`
+- Slicing and advanced/boolean indexing operations on ``Index`` classes as well
+  as :meth:`Index.delete` and :meth:`Index.drop` methods will no longer change the type of the
+  resulting index (:issue:`6440`, :issue:`7040`)
+
+  .. ipython:: python
+
+     i = pd.Index([1, 2, 3, 'a', 'b', 'c'])
+     i[[0, 1, 2]]
+     i.drop(['a', 'b', 'c'])
+
+  Previously, the above operation would return ``Int64Index``.  If you'd like
+  to do this manually, use :meth:`Index.astype`
+
+  .. ipython:: python
+
+     i[[0, 1, 2]].astype(np.int_)
+
+- ``set_index`` no longer converts MultiIndexes to an Index of tuples. For example,
+  the old behavior returned an Index in this case (:issue:`6459`):
+
+  .. ipython:: python
+     :suppress:
+
+     np.random.seed(1234)
+     from itertools import product
+     tuples = list(product(('a', 'b'), ('c', 'd')))
+     mi = pd.MultiIndex.from_tuples(tuples)
+     df_multi = pd.DataFrame(np.random.randn(4, 2), index=mi)
+     tuple_ind = pd.Index(tuples, tupleize_cols=False)
+     df_multi.index
+
+  .. ipython:: python
+
+     # Old behavior, casted MultiIndex to an Index
+     tuple_ind
+     df_multi.set_index(tuple_ind)
+
+     # New behavior
+     mi
+     df_multi.set_index(mi)
+
+  This also applies when passing multiple indices to ``set_index``:
+
+  .. ipython:: python
+
+    @suppress
+    df_multi.index = tuple_ind
+
+    # Old output, 2-level MultiIndex of tuples
+    df_multi.set_index([df_multi.index, df_multi.index])
+
+    @suppress
+    df_multi.index = mi
+
+    # New output, 4-level MultiIndex
+    df_multi.set_index([df_multi.index, df_multi.index])
+
+- ``pairwise`` keyword was added to the statistical moment functions
+  ``rolling_cov``, ``rolling_corr``, ``ewmcov``, ``ewmcorr``,
+  ``expanding_cov``, ``expanding_corr`` to allow the calculation of moving
+  window covariance and correlation matrices (:issue:`4950`). See
+  :ref:`Computing rolling pairwise covariances and correlations
+  <window.corr_pairwise>` in the docs.
+
+  .. code-block:: ipython
+
+     In [1]: df = pd.DataFrame(np.random.randn(10, 4), columns=list('ABCD'))
+
+     In [4]: covs = pd.rolling_cov(df[['A', 'B', 'C']],
+       ....:                       df[['B', 'C', 'D']],
+       ....:                       5,
+       ....:                       pairwise=True)
+
+
+     In [5]: covs[df.index[-1]]
+     Out[5]:
+               B         C         D
+     A  0.035310  0.326593 -0.505430
+     B  0.137748 -0.006888 -0.005383
+     C -0.006888  0.861040  0.020762
+
+- ``Series.iteritems()`` is now lazy (returns an iterator rather than a list). This was the documented behavior prior to 0.14. (:issue:`6760`)
+
+- Added ``nunique`` and ``value_counts`` functions to ``Index`` for counting unique elements. (:issue:`6734`)
+- ``stack`` and ``unstack`` now raise a ``ValueError`` when the ``level`` keyword refers
+  to a non-unique item in the ``Index`` (previously raised a ``KeyError``). (:issue:`6738`)
+- drop unused order argument from ``Series.sort``; args now are in the same order as ``Series.order``;
+  add ``na_position`` arg to conform to ``Series.order`` (:issue:`6847`)
+- default sorting algorithm for ``Series.order`` is now ``quicksort``, to conform with ``Series.sort``
+  (and numpy defaults)
+- add ``inplace`` keyword to ``Series.order/sort`` to make them inverses (:issue:`6859`)
+- ``DataFrame.sort`` now places NaNs at the beginning or end of the sort according to the ``na_position`` parameter. (:issue:`3917`)
+- accept ``TextFileReader`` in ``concat``, which was affecting a common user idiom (:issue:`6583`), this was a regression
+  from 0.13.1
+- Added ``factorize`` functions to ``Index`` and ``Series`` to get indexer and unique values (:issue:`7090`)
+- ``describe`` on a DataFrame with a mix of Timestamp and string like objects returns a different Index (:issue:`7088`).
+  Previously the index was unintentionally sorted.
+- Arithmetic operations with **only** ``bool`` dtypes now give a warning indicating
+  that they are evaluated in Python space for ``+``, ``-``,
+  and ``*`` operations and raise for all others (:issue:`7011`, :issue:`6762`,
+  :issue:`7015`, :issue:`7210`)
+
+  .. code-block:: python
+
+     >>> x = pd.Series(np.random.rand(10) > 0.5)
+     >>> y = True
+     >>> x + y  # warning generated: should do x | y instead
+     UserWarning: evaluating in Python space because the '+' operator is not
+     supported by numexpr for the bool dtype, use '|' instead
+     >>> x / y  # this raises because it doesn't make sense
+     NotImplementedError: operator '/' not implemented for bool dtypes
+
+- In ``HDFStore``, ``select_as_multiple`` will always raise a ``KeyError``, when a key or the selector is not found (:issue:`6177`)
+- ``df['col'] = value`` and ``df.loc[:,'col'] = value`` are now completely equivalent;
+  previously the ``.loc`` would not necessarily coerce the dtype of the resultant series (:issue:`6149`)
+- ``dtypes`` and ``ftypes`` now return a series with ``dtype=object`` on empty containers (:issue:`5740`)
+- ``df.to_csv`` will now return a string of the CSV data if neither a target path nor a buffer is provided
+  (:issue:`6061`)
+- ``pd.infer_freq()`` will now raise a ``TypeError`` if given an invalid ``Series/Index``
+  type (:issue:`6407`, :issue:`6463`)
+- A tuple passed to ``DataFame.sort_index`` will be interpreted as the levels of
+  the index, rather than requiring a list of tuple (:issue:`4370`)
+- all offset operations now return ``Timestamp`` types (rather than datetime), Business/Week frequencies were incorrect (:issue:`4069`)
+- ``to_excel`` now converts ``np.inf`` into a string representation,
+  customizable by the ``inf_rep`` keyword argument (Excel has no native inf
+  representation) (:issue:`6782`)
+- Replace ``pandas.compat.scipy.scoreatpercentile`` with ``numpy.percentile`` (:issue:`6810`)
+- ``.quantile`` on a ``datetime[ns]`` series now returns ``Timestamp`` instead
+  of ``np.datetime64`` objects (:issue:`6810`)
+- change ``AssertionError`` to ``TypeError`` for invalid types passed to ``concat`` (:issue:`6583`)
+- Raise a ``TypeError`` when ``DataFrame`` is passed an iterator as the
+  ``data`` argument (:issue:`5357`)
+
+
+.. _whatsnew_0140.display:
+
+Display changes
+~~~~~~~~~~~~~~~
+
+- The default way of printing large DataFrames has changed. DataFrames
+  exceeding ``max_rows`` and/or ``max_columns`` are now displayed in a
+  centrally truncated view, consistent with the printing of a
+  :class:`pandas.Series` (:issue:`5603`).
+
+  In previous versions, a DataFrame was truncated once the dimension
+  constraints were reached and an ellipse (...) signaled that part of
+  the data was cut off.
+
+  .. image:: ../_static/trunc_before.png
+      :alt: The previous look of truncate.
+
+  In the current version, large DataFrames are centrally truncated,
+  showing a preview of head and tail in both dimensions.
+
+  .. image:: ../_static/trunc_after.png
+     :alt: The new look.
+
+- allow option ``'truncate'`` for ``display.show_dimensions`` to only show the dimensions if the
+  frame is truncated (:issue:`6547`).
+
+  The default for ``display.show_dimensions`` will now be ``truncate``. This is consistent with
+  how Series display length.
+
+  .. ipython:: python
+
+     dfd = pd.DataFrame(np.arange(25).reshape(-1, 5),
+                        index=[0, 1, 2, 3, 4],
+                        columns=[0, 1, 2, 3, 4])
+
+     # show dimensions since this is truncated
+     with pd.option_context('display.max_rows', 2, 'display.max_columns', 2,
+                            'display.show_dimensions', 'truncate'):
+         print(dfd)
+
+     # will not show dimensions since it is not truncated
+     with pd.option_context('display.max_rows', 10, 'display.max_columns', 40,
+                            'display.show_dimensions', 'truncate'):
+         print(dfd)
+
+- Regression in the display of a MultiIndexed Series with ``display.max_rows`` is less than the
+  length of the series (:issue:`7101`)
+- Fixed a bug in the HTML repr of a truncated Series or DataFrame not showing the class name with the
+  ``large_repr`` set to 'info' (:issue:`7105`)
+- The ``verbose`` keyword in ``DataFrame.info()``, which controls whether to shorten the ``info``
+  representation, is now ``None`` by default. This will follow the global setting in
+  ``display.max_info_columns``. The global setting can be overridden with ``verbose=True`` or
+  ``verbose=False``.
+- Fixed a bug with the ``info`` repr not honoring the ``display.max_info_columns`` setting (:issue:`6939`)
+- Offset/freq info now in Timestamp __repr__ (:issue:`4553`)
+
+.. _whatsnew_0140.parsing:
+
+Text parsing API changes
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+:func:`read_csv`/:func:`read_table` will now be noisier w.r.t invalid options rather than falling back to the ``PythonParser``.
+
+- Raise ``ValueError`` when ``sep`` specified with
+  ``delim_whitespace=True`` in :func:`read_csv`/:func:`read_table`
+  (:issue:`6607`)
+- Raise ``ValueError`` when ``engine='c'`` specified with unsupported
+  options in :func:`read_csv`/:func:`read_table` (:issue:`6607`)
+- Raise ``ValueError`` when fallback to python parser causes options to be
+  ignored (:issue:`6607`)
+- Produce :class:`~pandas.io.parsers.ParserWarning` on fallback to python
+  parser when no options are ignored (:issue:`6607`)
+- Translate ``sep='\s+'`` to ``delim_whitespace=True`` in
+  :func:`read_csv`/:func:`read_table` if no other C-unsupported options
+  specified (:issue:`6607`)
+
+.. _whatsnew_0140.groupby:
+
+GroupBy API changes
+~~~~~~~~~~~~~~~~~~~
+
+More consistent behavior for some groupby methods:
+
+- groupby ``head`` and ``tail`` now act more like ``filter`` rather than an aggregation:
+
+  .. code-block:: ipython
+
+     In [1]: df = pd.DataFrame([[1, 2], [1, 4], [5, 6]], columns=['A', 'B'])
+
+     In [2]: g = df.groupby('A')
+
+     In [3]: g.head(1)  # filters DataFrame
+     Out[3]:
+        A  B
+     0  1  2
+     2  5  6
+
+     In [4]: g.apply(lambda x: x.head(1))  # used to simply fall-through
+     Out[4]:
+          A  B
+     A
+     1 0  1  2
+     5 2  5  6
+
+- groupby head and tail respect column selection:
+
+  .. code-block:: ipython
+
+     In [19]: g[['B']].head(1)
+     Out[19]:
+        B
+     0  2
+     2  6
+
+     [2 rows x 1 columns]
+
+- groupby ``nth`` now reduces by default; filtering can be achieved by passing ``as_index=False``. With an optional ``dropna`` argument to ignore
+  NaN. See :ref:`the docs <groupby.nth>`.
+
+  Reducing
+
+  .. ipython:: python
+
+     df = pd.DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=['A', 'B'])
+     g = df.groupby('A')
+     g.nth(0)
+
+     # this is equivalent to g.first()
+     g.nth(0, dropna='any')
+
+     # this is equivalent to g.last()
+     g.nth(-1, dropna='any')
+
+  Filtering
+
+  .. ipython:: python
+
+     gf = df.groupby('A', as_index=False)
+     gf.nth(0)
+     gf.nth(0, dropna='any')
+
+- groupby will now not return the grouped column for non-cython functions (:issue:`5610`, :issue:`5614`, :issue:`6732`),
+  as its already the index
+
+  .. ipython:: python
+
+     df = pd.DataFrame([[1, np.nan], [1, 4], [5, 6], [5, 8]], columns=['A', 'B'])
+     g = df.groupby('A')
+     g.count()
+     g.describe()
+
+- passing ``as_index`` will leave the grouped column in-place (this is not change in 0.14.0)
+
+  .. ipython:: python
+
+     df = pd.DataFrame([[1, np.nan], [1, 4], [5, 6], [5, 8]], columns=['A', 'B'])
+     g = df.groupby('A', as_index=False)
+     g.count()
+     g.describe()
+
+- Allow specification of a more complex groupby via ``pd.Grouper``, such as grouping
+  by a Time and a string field simultaneously. See :ref:`the docs <groupby.specify>`. (:issue:`3794`)
+
+- Better propagation/preservation of Series names when performing groupby
+  operations:
+
+  - ``SeriesGroupBy.agg`` will ensure that the name attribute of the original
+    series is propagated to the result (:issue:`6265`).
+  - If the function provided to ``GroupBy.apply`` returns a named series, the
+    name of the series will be kept as the name of the column index of the
+    DataFrame returned by ``GroupBy.apply`` (:issue:`6124`).  This facilitates
+    ``DataFrame.stack`` operations where the name of the column index is used as
+    the name of the inserted column containing the pivoted data.
+
+
+.. _whatsnew_0140.sql:
+
+SQL
+~~~
+
+The SQL reading and writing functions now support more database flavors
+through SQLAlchemy (:issue:`2717`, :issue:`4163`, :issue:`5950`, :issue:`6292`).
+All databases supported by SQLAlchemy can be used, such
+as PostgreSQL, MySQL, Oracle, Microsoft SQL server (see documentation of
+SQLAlchemy on `included dialects
+<https://sqlalchemy.readthedocs.io/en/latest/dialects/index.html>`_).
+
+The functionality of providing DBAPI connection objects will only be supported
+for sqlite3 in the future. The ``'mysql'`` flavor is deprecated.
+
+The new functions :func:`~pandas.read_sql_query` and :func:`~pandas.read_sql_table`
+are introduced. The function :func:`~pandas.read_sql` is kept as a convenience
+wrapper around the other two and will delegate to specific function depending on
+the provided input (database table name or sql query).
+
+In practice, you have to provide a SQLAlchemy ``engine`` to the sql functions.
+To connect with SQLAlchemy you use the :func:`create_engine` function to create an engine
+object from database URI. You only need to create the engine once per database you are
+connecting to. For an in-memory sqlite database:
+
+.. ipython:: python
+
+   from sqlalchemy import create_engine
+   # Create your connection.
+   engine = create_engine('sqlite:///:memory:')
+
+This ``engine`` can then be used to write or read data to/from this database:
+
+.. ipython:: python
+
+    df = pd.DataFrame({'A': [1, 2, 3], 'B': ['a', 'b', 'c']})
+    df.to_sql(name='db_table', con=engine, index=False)
+
+You can read data from a database by specifying the table name:
+
+.. ipython:: python
+
+   pd.read_sql_table('db_table', engine)
+
+or by specifying a sql query:
+
+.. ipython:: python
+
+   pd.read_sql_query('SELECT * FROM db_table', engine)
+
+Some other enhancements to the sql functions include:
+
+- support for writing the index. This can be controlled with the ``index``
+  keyword (default is True).
+- specify the column label to use when writing the index with ``index_label``.
+- specify string columns to parse as datetimes with the ``parse_dates``
+  keyword in :func:`~pandas.read_sql_query` and :func:`~pandas.read_sql_table`.
+
+.. warning::
+
+    Some of the existing functions or function aliases have been deprecated
+    and will be removed in future versions. This includes: ``tquery``, ``uquery``,
+    ``read_frame``, ``frame_query``, ``write_frame``.
+
+.. warning::
+
+    The support for the 'mysql' flavor when using DBAPI connection objects has been deprecated.
+    MySQL will be further supported with SQLAlchemy engines (:issue:`6900`).
+
+
+.. _whatsnew_0140.slicers:
+
+Multi-indexing using slicers
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In 0.14.0 we added a new way to slice MultiIndexed objects.
+You can slice a MultiIndex by providing multiple indexers.
+
+You can provide any of the selectors as if you are indexing by label, see :ref:`Selection by Label <indexing.label>`,
+including slices, lists of labels, labels, and boolean indexers.
+
+You can use ``slice(None)`` to select all the contents of *that* level. You do not need to specify all the
+*deeper* levels, they will be implied as ``slice(None)``.
+
+As usual, **both sides** of the slicers are included as this is label indexing.
+
+See :ref:`the docs<advanced.mi_slicers>`
+See also issues (:issue:`6134`, :issue:`4036`, :issue:`3057`, :issue:`2598`, :issue:`5641`, :issue:`7106`)
+
+.. warning::
+
+   You should specify all axes in the ``.loc`` specifier, meaning the indexer for the **index** and
+   for the **columns**. Their are some ambiguous cases where the passed indexer could be mis-interpreted
+   as indexing *both* axes, rather than into say the MultiIndex for the rows.
+
+   You should do this:
+
+  .. code-block:: python
+
+     >>> df.loc[(slice('A1', 'A3'), ...), :]  # noqa: E901
+
+   rather than this:
+
+  .. code-block:: python
+
+     >>> df.loc[(slice('A1', 'A3'), ...)]  # noqa: E901
+
+.. warning::
+
+   You will need to make sure that the selection axes are fully lexsorted!
+
+.. ipython:: python
+
+   def mklbl(prefix, n):
+       return ["%s%s" % (prefix, i) for i in range(n)]
+
+   index = pd.MultiIndex.from_product([mklbl('A', 4),
+                                       mklbl('B', 2),
+                                       mklbl('C', 4),
+                                       mklbl('D', 2)])
+   columns = pd.MultiIndex.from_tuples([('a', 'foo'), ('a', 'bar'),
+                                        ('b', 'foo'), ('b', 'bah')],
+                                       names=['lvl0', 'lvl1'])
+   df = pd.DataFrame(np.arange(len(index) * len(columns)).reshape((len(index),
+                     len(columns))),
+                     index=index,
+                     columns=columns).sort_index().sort_index(axis=1)
+   df
+
+Basic MultiIndex slicing using slices, lists, and labels.
+
+.. ipython:: python
+
+   df.loc[(slice('A1', 'A3'), slice(None), ['C1', 'C3']), :]
+
+You can use a ``pd.IndexSlice`` to shortcut the creation of these slices
+
+.. ipython:: python
+
+   idx = pd.IndexSlice
+   df.loc[idx[:, :, ['C1', 'C3']], idx[:, 'foo']]
+
+It is possible to perform quite complicated selections using this method on multiple
+axes at the same time.
+
+.. ipython:: python
+
+   df.loc['A1', (slice(None), 'foo')]
+   df.loc[idx[:, :, ['C1', 'C3']], idx[:, 'foo']]
+
+Using a boolean indexer you can provide selection related to the *values*.
+
+.. ipython:: python
+
+   mask = df[('a', 'foo')] > 200
+   df.loc[idx[mask, :, ['C1', 'C3']], idx[:, 'foo']]
+
+You can also specify the ``axis`` argument to ``.loc`` to interpret the passed
+slicers on a single axis.
+
+.. ipython:: python
+
+   df.loc(axis=0)[:, :, ['C1', 'C3']]
+
+Furthermore you can *set* the values using these methods
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2.loc(axis=0)[:, :, ['C1', 'C3']] = -10
+   df2
+
+You can use a right-hand-side of an alignable object as well.
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2.loc[idx[:, :, ['C1', 'C3']], :] = df2 * 1000
+   df2
+
+.. _whatsnew_0140.plotting:
+
+Plotting
+~~~~~~~~
+
+- Hexagonal bin plots from ``DataFrame.plot`` with ``kind='hexbin'`` (:issue:`5478`), See :ref:`the docs<visualization.hexbin>`.
+- ``DataFrame.plot`` and ``Series.plot`` now supports area plot with specifying ``kind='area'`` (:issue:`6656`), See :ref:`the docs<visualization.area_plot>`
+- Pie plots from ``Series.plot`` and ``DataFrame.plot`` with ``kind='pie'`` (:issue:`6976`), See :ref:`the docs<visualization.pie>`.
+- Plotting with Error Bars is now supported in the ``.plot`` method of ``DataFrame`` and ``Series`` objects (:issue:`3796`, :issue:`6834`), See :ref:`the docs<visualization.errorbars>`.
+- ``DataFrame.plot`` and ``Series.plot`` now support a ``table`` keyword for plotting ``matplotlib.Table``, See :ref:`the docs<visualization.table>`. The ``table`` keyword can receive the following values.
+
+  - ``False``: Do nothing (default).
+  - ``True``: Draw a table using the ``DataFrame`` or ``Series`` called ``plot`` method. Data will be transposed to meet matplotlib's default layout.
+  - ``DataFrame`` or ``Series``: Draw matplotlib.table using the passed data. The data will be drawn as displayed in print method (not transposed automatically).
+    Also, helper function ``pandas.tools.plotting.table`` is added to create a table from ``DataFrame`` and ``Series``, and add it to an ``matplotlib.Axes``.
+
+- ``plot(legend='reverse')`` will now reverse the order of legend labels for
+  most plot kinds. (:issue:`6014`)
+- Line plot and area plot can be stacked by ``stacked=True`` (:issue:`6656`)
+
+- Following keywords are now acceptable for :meth:`DataFrame.plot` with ``kind='bar'`` and ``kind='barh'``:
+
+  - ``width``: Specify the bar width. In previous versions, static value 0.5 was passed to matplotlib and it cannot be overwritten. (:issue:`6604`)
+  - ``align``: Specify the bar alignment. Default is ``center`` (different from matplotlib). In previous versions, pandas passes ``align='edge'`` to matplotlib and adjust the location to ``center`` by itself, and it results ``align`` keyword is not applied as expected. (:issue:`4525`)
+  - ``position``: Specify relative alignments for bar plot layout. From 0 (left/bottom-end) to 1(right/top-end). Default is 0.5 (center). (:issue:`6604`)
+
+  Because of the default ``align`` value changes, coordinates of bar plots are now located on integer values (0.0, 1.0, 2.0 ...). This is intended to make bar plot be located on the same coordinates as line plot. However, bar plot may differs unexpectedly when you manually adjust the bar location or drawing area, such as using ``set_xlim``, ``set_ylim``, etc. In this cases, please modify your script to meet with new coordinates.
+
+- The :func:`parallel_coordinates` function now takes argument ``color``
+  instead of ``colors``. A ``FutureWarning`` is raised to alert that
+  the old ``colors`` argument will not be supported in a future release. (:issue:`6956`)
+
+- The :func:`parallel_coordinates` and :func:`andrews_curves` functions now take
+  positional argument ``frame`` instead of ``data``. A ``FutureWarning`` is
+  raised if the old ``data`` argument is used by name. (:issue:`6956`)
+
+- :meth:`DataFrame.boxplot` now supports ``layout`` keyword (:issue:`6769`)
+- :meth:`DataFrame.boxplot` has a new keyword argument, ``return_type``. It accepts ``'dict'``,
+  ``'axes'``, or ``'both'``, in which case a namedtuple with the matplotlib
+  axes and a dict of matplotlib Lines is returned.
+
+
+.. _whatsnew_0140.prior_deprecations:
+
+Prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+There are prior version deprecations that are taking effect as of 0.14.0.
+
+- Remove :class:`DateRange` in favor of :class:`DatetimeIndex` (:issue:`6816`)
+- Remove ``column`` keyword from ``DataFrame.sort`` (:issue:`4370`)
+- Remove ``precision`` keyword from :func:`set_eng_float_format` (:issue:`395`)
+- Remove ``force_unicode`` keyword from :meth:`DataFrame.to_string`,
+  :meth:`DataFrame.to_latex`, and :meth:`DataFrame.to_html`; these function
+  encode in unicode by default (:issue:`2224`, :issue:`2225`)
+- Remove ``nanRep`` keyword from :meth:`DataFrame.to_csv` and
+  :meth:`DataFrame.to_string` (:issue:`275`)
+- Remove ``unique`` keyword from :meth:`HDFStore.select_column` (:issue:`3256`)
+- Remove ``inferTimeRule`` keyword from :func:`Timestamp.offset` (:issue:`391`)
+- Remove ``name`` keyword from :func:`get_data_yahoo` and
+  :func:`get_data_google` ( `commit b921d1a <https://github.com/pandas-dev/pandas/commit/b921d1a2>`__ )
+- Remove ``offset`` keyword from :class:`DatetimeIndex` constructor
+  ( `commit 3136390 <https://github.com/pandas-dev/pandas/commit/3136390>`__ )
+- Remove ``time_rule`` from several rolling-moment statistical functions, such
+  as :func:`rolling_sum` (:issue:`1042`)
+- Removed neg ``-`` boolean operations on numpy arrays in favor of inv ``~``, as this is going to
+  be deprecated in numpy 1.9 (:issue:`6960`)
+
+.. _whatsnew_0140.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- The :func:`pivot_table`/:meth:`DataFrame.pivot_table` and :func:`crosstab` functions
+  now take arguments ``index`` and ``columns`` instead of ``rows`` and ``cols``.  A
+  ``FutureWarning`` is raised to alert that the old ``rows`` and ``cols`` arguments
+  will not be supported in a future release (:issue:`5505`)
+
+- The :meth:`DataFrame.drop_duplicates` and :meth:`DataFrame.duplicated` methods
+  now take argument ``subset`` instead of ``cols`` to better align with
+  :meth:`DataFrame.dropna`.  A ``FutureWarning`` is raised to alert that the old
+  ``cols`` arguments will not be supported in a future release (:issue:`6680`)
+
+- The :meth:`DataFrame.to_csv` and :meth:`DataFrame.to_excel` functions
+  now takes argument ``columns`` instead of ``cols``.  A
+  ``FutureWarning`` is raised to alert that the old ``cols`` arguments
+  will not be supported in a future release (:issue:`6645`)
+
+- Indexers will warn ``FutureWarning`` when used with a scalar indexer and
+  a non-floating point Index (:issue:`4892`, :issue:`6960`)
+
+  .. code-block:: ipython
+
+     # non-floating point indexes can only be indexed by integers / labels
+     In [1]: pd.Series(1, np.arange(5))[3.0]
+             pandas/core/index.py:469: FutureWarning: scalar indexers for index type Int64Index should be integers and not floating point
+     Out[1]: 1
+
+     In [2]: pd.Series(1, np.arange(5)).iloc[3.0]
+             pandas/core/index.py:469: FutureWarning: scalar indexers for index type Int64Index should be integers and not floating point
+     Out[2]: 1
+
+     In [3]: pd.Series(1, np.arange(5)).iloc[3.0:4]
+             pandas/core/index.py:527: FutureWarning: slice indexers when using iloc should be integers and not floating point
+     Out[3]:
+             3    1
+             dtype: int64
+
+     # these are Float64Indexes, so integer or floating point is acceptable
+     In [4]: pd.Series(1, np.arange(5.))[3]
+     Out[4]: 1
+
+     In [5]: pd.Series(1, np.arange(5.))[3.0]
+     Out[6]: 1
+
+- Numpy 1.9 compat w.r.t. deprecation warnings (:issue:`6960`)
+
+- :meth:`Panel.shift` now has a function signature that matches :meth:`DataFrame.shift`.
+  The old positional argument ``lags`` has been changed to a keyword argument
+  ``periods`` with a default value of 1. A ``FutureWarning`` is raised if the
+  old argument ``lags`` is used by name. (:issue:`6910`)
+- The ``order`` keyword argument of :func:`factorize` will be removed. (:issue:`6926`).
+
+- Remove the ``copy`` keyword from :meth:`DataFrame.xs`, :meth:`Panel.major_xs`, :meth:`Panel.minor_xs`. A view will be
+  returned if possible, otherwise a copy will be made. Previously the user could think that ``copy=False`` would
+  ALWAYS return a view. (:issue:`6894`)
+
+- The :func:`parallel_coordinates` function now takes argument ``color``
+  instead of ``colors``. A ``FutureWarning`` is raised to alert that
+  the old ``colors`` argument will not be supported in a future release. (:issue:`6956`)
+
+- The :func:`parallel_coordinates` and :func:`andrews_curves` functions now take
+  positional argument ``frame`` instead of ``data``. A ``FutureWarning`` is
+  raised if the old ``data`` argument is used by name. (:issue:`6956`)
+
+- The support for the 'mysql' flavor when using DBAPI connection objects has been deprecated.
+  MySQL will be further supported with SQLAlchemy engines (:issue:`6900`).
+
+- The following ``io.sql`` functions have been deprecated: ``tquery``, ``uquery``, ``read_frame``, ``frame_query``, ``write_frame``.
+
+- The ``percentile_width`` keyword argument in :meth:`~DataFrame.describe` has been deprecated.
+  Use the ``percentiles`` keyword instead, which takes a list of percentiles to display. The
+  default output is unchanged.
+
+- The default return type of :func:`boxplot` will change from a dict to a matplotlib Axes
+  in a future release. You can use the future behavior now by passing ``return_type='axes'``
+  to boxplot.
+
+.. _whatsnew_0140.knownissues:
+
+Known issues
+~~~~~~~~~~~~
+
+- OpenPyXL 2.0.0 breaks backwards compatibility (:issue:`7169`)
+
+
+.. _whatsnew_0140.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+- DataFrame and Series will create a MultiIndex object if passed a tuples dict, See :ref:`the docs<basics.dataframe.from_dict_of_tuples>` (:issue:`3323`)
+
+  .. ipython:: python
+
+     pd.Series({('a', 'b'): 1, ('a', 'a'): 0,
+                ('a', 'c'): 2, ('b', 'a'): 3, ('b', 'b'): 4})
+     pd.DataFrame({('a', 'b'): {('A', 'B'): 1, ('A', 'C'): 2},
+                  ('a', 'a'): {('A', 'C'): 3, ('A', 'B'): 4},
+                  ('a', 'c'): {('A', 'B'): 5, ('A', 'C'): 6},
+                  ('b', 'a'): {('A', 'C'): 7, ('A', 'B'): 8},
+                  ('b', 'b'): {('A', 'D'): 9, ('A', 'B'): 10}})
+
+- Added the ``sym_diff`` method to ``Index`` (:issue:`5543`)
+- ``DataFrame.to_latex`` now takes a longtable keyword, which if True will return a table in a longtable environment. (:issue:`6617`)
+- Add option to turn off escaping in ``DataFrame.to_latex`` (:issue:`6472`)
+- ``pd.read_clipboard`` will, if the keyword ``sep`` is unspecified, try to detect data copied from a spreadsheet
+  and parse accordingly. (:issue:`6223`)
+- Joining a singly-indexed DataFrame with a MultiIndexed DataFrame (:issue:`3662`)
+
+  See :ref:`the docs<merging.join_on_mi>`. Joining MultiIndex DataFrames on both the left and right is not yet supported ATM.
+
+  .. ipython:: python
+
+     household = pd.DataFrame({'household_id': [1, 2, 3],
+                               'male': [0, 1, 0],
+                               'wealth': [196087.3, 316478.7, 294750]
+                               },
+                              columns=['household_id', 'male', 'wealth']
+                              ).set_index('household_id')
+     household
+     portfolio = pd.DataFrame({'household_id': [1, 2, 2, 3, 3, 3, 4],
+                               'asset_id': ["nl0000301109",
+                                            "nl0000289783",
+                                            "gb00b03mlx29",
+                                            "gb00b03mlx29",
+                                            "lu0197800237",
+                                            "nl0000289965",
+                                            np.nan],
+                               'name': ["ABN Amro",
+                                        "Robeco",
+                                        "Royal Dutch Shell",
+                                        "Royal Dutch Shell",
+                                        "AAB Eastern Europe Equity Fund",
+                                        "Postbank BioTech Fonds",
+                                        np.nan],
+                               'share': [1.0, 0.4, 0.6, 0.15, 0.6, 0.25, 1.0]
+                               },
+                              columns=['household_id', 'asset_id', 'name', 'share']
+                              ).set_index(['household_id', 'asset_id'])
+     portfolio
+
+     household.join(portfolio, how='inner')
+
+- ``quotechar``, ``doublequote``, and ``escapechar`` can now be specified when
+  using ``DataFrame.to_csv`` (:issue:`5414`, :issue:`4528`)
+- Partially sort by only the specified levels of a MultiIndex with the
+  ``sort_remaining`` boolean kwarg. (:issue:`3984`)
+- Added ``to_julian_date`` to ``TimeStamp`` and ``DatetimeIndex``.  The Julian
+  Date is used primarily in astronomy and represents the number of days from
+  noon, January 1, 4713 BC.  Because nanoseconds are used to define the time
+  in pandas the actual range of dates that you can use is 1678 AD to 2262 AD. (:issue:`4041`)
+- ``DataFrame.to_stata`` will now check data for compatibility with Stata data types
+  and will upcast when needed.  When it is not possible to losslessly upcast, a warning
+  is issued (:issue:`6327`)
+- ``DataFrame.to_stata`` and ``StataWriter`` will accept keyword arguments time_stamp
+  and data_label which allow the time stamp and dataset label to be set when creating a
+  file. (:issue:`6545`)
+- ``pandas.io.gbq`` now handles reading unicode strings properly. (:issue:`5940`)
+- :ref:`Holidays Calendars<timeseries.holiday>` are now available and can be used with the ``CustomBusinessDay`` offset (:issue:`6719`)
+- ``Float64Index`` is now backed by a ``float64`` dtype ndarray instead of an
+  ``object`` dtype array (:issue:`6471`).
+- Implemented ``Panel.pct_change`` (:issue:`6904`)
+- Added ``how`` option to rolling-moment functions to dictate how to handle resampling; :func:`rolling_max` defaults to max,
+  :func:`rolling_min` defaults to min, and all others default to mean (:issue:`6297`)
+- ``CustomBusinessMonthBegin`` and ``CustomBusinessMonthEnd`` are now available (:issue:`6866`)
+- :meth:`Series.quantile` and :meth:`DataFrame.quantile` now accept an array of
+  quantiles.
+- :meth:`~DataFrame.describe` now accepts an array of percentiles to include in the summary statistics (:issue:`4196`)
+- ``pivot_table`` can now accept ``Grouper`` by ``index`` and ``columns`` keywords (:issue:`6913`)
+
+  .. ipython:: python
+
+     import datetime
+     df = pd.DataFrame({
+         'Branch': 'A A A A A B'.split(),
+         'Buyer': 'Carl Mark Carl Carl Joe Joe'.split(),
+         'Quantity': [1, 3, 5, 1, 8, 1],
+         'Date': [datetime.datetime(2013, 11, 1, 13, 0),
+                  datetime.datetime(2013, 9, 1, 13, 5),
+                  datetime.datetime(2013, 10, 1, 20, 0),
+                  datetime.datetime(2013, 10, 2, 10, 0),
+                  datetime.datetime(2013, 11, 1, 20, 0),
+                  datetime.datetime(2013, 10, 2, 10, 0)],
+         'PayDay': [datetime.datetime(2013, 10, 4, 0, 0),
+                    datetime.datetime(2013, 10, 15, 13, 5),
+                    datetime.datetime(2013, 9, 5, 20, 0),
+                    datetime.datetime(2013, 11, 2, 10, 0),
+                    datetime.datetime(2013, 10, 7, 20, 0),
+                    datetime.datetime(2013, 9, 5, 10, 0)]})
+     df
+
+  .. code-block:: ipython
+
+     In [75]: df.pivot_table(values='Quantity',
+        ....:                index=pd.Grouper(freq='M', key='Date'),
+        ....:                columns=pd.Grouper(freq='M', key='PayDay'),
+        ....:                aggfunc="sum")
+     Out[75]:
+     PayDay      2013-09-30  2013-10-31  2013-11-30
+     Date
+     2013-09-30         NaN         3.0         NaN
+     2013-10-31         6.0         NaN         1.0
+     2013-11-30         NaN         9.0         NaN
+
+     [3 rows x 3 columns]
+
+- Arrays of strings can be wrapped to a specified width (``str.wrap``) (:issue:`6999`)
+- Add :meth:`~Series.nsmallest` and :meth:`Series.nlargest` methods to Series, See :ref:`the docs <basics.nsorted>` (:issue:`3960`)
+
+- ``PeriodIndex`` fully supports partial string indexing like ``DatetimeIndex`` (:issue:`7043`)
+
+  .. code-block:: ipython
+
+     In [76]: prng = pd.period_range('2013-01-01 09:00', periods=100, freq='H')
+
+     In [77]: ps = pd.Series(np.random.randn(len(prng)), index=prng)
+
+     In [78]: ps
+     Out[78]:
+     2013-01-01 09:00    0.015696
+     2013-01-01 10:00   -2.242685
+     2013-01-01 11:00    1.150036
+     2013-01-01 12:00    0.991946
+     2013-01-01 13:00    0.953324
+                           ...
+     2013-01-05 08:00    0.285296
+     2013-01-05 09:00    0.484288
+     2013-01-05 10:00    1.363482
+     2013-01-05 11:00   -0.781105
+     2013-01-05 12:00   -0.468018
+     Freq: H, Length: 100, dtype: float64
+
+     In [79]: ps['2013-01-02']
+     Out[79]:
+     2013-01-02 00:00    0.553439
+     2013-01-02 01:00    1.318152
+     2013-01-02 02:00   -0.469305
+     2013-01-02 03:00    0.675554
+     2013-01-02 04:00   -1.817027
+                           ...
+     2013-01-02 19:00    0.036142
+     2013-01-02 20:00   -2.074978
+     2013-01-02 21:00    0.247792
+     2013-01-02 22:00   -0.897157
+     2013-01-02 23:00   -0.136795
+     Freq: H, Length: 24, dtype: float64
+
+- ``read_excel`` can now read milliseconds in Excel dates and times with xlrd >= 0.9.3. (:issue:`5945`)
+- ``pd.stats.moments.rolling_var`` now uses Welford's method for increased numerical stability (:issue:`6817`)
+- pd.expanding_apply and pd.rolling_apply now take args and kwargs that are passed on to
+  the func (:issue:`6289`)
+- ``DataFrame.rank()`` now has a percentage rank option (:issue:`5971`)
+- ``Series.rank()`` now has a percentage rank option (:issue:`5971`)
+- ``Series.rank()`` and ``DataFrame.rank()`` now accept ``method='dense'`` for ranks without gaps (:issue:`6514`)
+- Support passing ``encoding`` with xlwt (:issue:`3710`)
+- Refactor Block classes removing ``Block.items`` attributes to avoid duplication
+  in item handling (:issue:`6745`, :issue:`6988`).
+- Testing statements updated to use specialized asserts (:issue:`6175`)
+
+
+
+.. _whatsnew_0140.performance:
+
+Performance
+~~~~~~~~~~~
+
+- Performance improvement when converting ``DatetimeIndex`` to floating ordinals
+  using ``DatetimeConverter`` (:issue:`6636`)
+- Performance improvement for  ``DataFrame.shift`` (:issue:`5609`)
+- Performance improvement in indexing into a MultiIndexed Series (:issue:`5567`)
+- Performance improvements in single-dtyped indexing (:issue:`6484`)
+- Improve performance of DataFrame construction with certain offsets, by removing faulty caching
+  (e.g. MonthEnd,BusinessMonthEnd), (:issue:`6479`)
+- Improve performance of ``CustomBusinessDay`` (:issue:`6584`)
+- improve performance of slice indexing on Series with string keys (:issue:`6341`, :issue:`6372`)
+- Performance improvement for ``DataFrame.from_records`` when reading a
+  specified number of rows from an iterable (:issue:`6700`)
+- Performance improvements in timedelta conversions for integer dtypes (:issue:`6754`)
+- Improved performance of compatible pickles (:issue:`6899`)
+- Improve performance in certain reindexing operations by optimizing ``take_2d`` (:issue:`6749`)
+- ``GroupBy.count()`` is now implemented in Cython and is much faster for large
+  numbers of groups (:issue:`7016`).
+
+Experimental
+~~~~~~~~~~~~
+
+There are no experimental changes in 0.14.0
+
+
+.. _whatsnew_0140.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in Series ValueError when index doesn't match data (:issue:`6532`)
+- Prevent segfault due to MultiIndex not being supported in HDFStore table
+  format (:issue:`1848`)
+- Bug in ``pd.DataFrame.sort_index`` where mergesort wasn't stable when ``ascending=False`` (:issue:`6399`)
+- Bug in ``pd.tseries.frequencies.to_offset`` when argument has leading zeros (:issue:`6391`)
+- Bug in version string gen. for dev versions with shallow clones / install from tarball (:issue:`6127`)
+- Inconsistent tz parsing ``Timestamp`` / ``to_datetime`` for current year (:issue:`5958`)
+- Indexing bugs with reordered indexes (:issue:`6252`, :issue:`6254`)
+- Bug in ``.xs`` with a Series multiindex (:issue:`6258`, :issue:`5684`)
+- Bug in conversion of a string types to a DatetimeIndex with a specified frequency (:issue:`6273`, :issue:`6274`)
+- Bug in ``eval`` where type-promotion failed for large expressions (:issue:`6205`)
+- Bug in interpolate with ``inplace=True`` (:issue:`6281`)
+- ``HDFStore.remove`` now handles start and stop (:issue:`6177`)
+- ``HDFStore.select_as_multiple`` handles start and stop the same way as ``select`` (:issue:`6177`)
+- ``HDFStore.select_as_coordinates`` and ``select_column`` works with a ``where`` clause that results in filters (:issue:`6177`)
+- Regression in join of non_unique_indexes (:issue:`6329`)
+- Issue with groupby ``agg`` with a single function and a mixed-type frame (:issue:`6337`)
+- Bug in ``DataFrame.replace()`` when passing a non- ``bool``
+  ``to_replace`` argument (:issue:`6332`)
+- Raise when trying to align on different levels of a MultiIndex assignment (:issue:`3738`)
+- Bug in setting complex dtypes via boolean indexing (:issue:`6345`)
+- Bug in TimeGrouper/resample when presented with a non-monotonic DatetimeIndex that would return invalid results. (:issue:`4161`)
+- Bug in index name propagation in TimeGrouper/resample (:issue:`4161`)
+- TimeGrouper has a more compatible API to the rest of the groupers (e.g. ``groups`` was missing) (:issue:`3881`)
+- Bug in multiple grouping with a TimeGrouper depending on target column order (:issue:`6764`)
+- Bug in ``pd.eval`` when parsing strings with possible tokens like ``'&'``
+  (:issue:`6351`)
+- Bug correctly handle placements of ``-inf`` in Panels when dividing by integer 0 (:issue:`6178`)
+- ``DataFrame.shift`` with ``axis=1`` was raising (:issue:`6371`)
+- Disabled clipboard tests until release time (run locally with ``nosetests -A disabled``) (:issue:`6048`).
+- Bug in ``DataFrame.replace()`` when passing a nested ``dict`` that contained
+  keys not in the values to be replaced (:issue:`6342`)
+- ``str.match`` ignored the na flag (:issue:`6609`).
+- Bug in take with duplicate columns that were not consolidated (:issue:`6240`)
+- Bug in interpolate changing dtypes (:issue:`6290`)
+- Bug in ``Series.get``, was using a buggy access method (:issue:`6383`)
+- Bug in hdfstore queries of the form ``where=[('date', '>=', datetime(2013,1,1)), ('date', '<=', datetime(2014,1,1))]`` (:issue:`6313`)
+- Bug in ``DataFrame.dropna`` with duplicate indices (:issue:`6355`)
+- Regression in chained getitem indexing with embedded list-like from 0.12 (:issue:`6394`)
+- ``Float64Index`` with nans not comparing correctly (:issue:`6401`)
+- ``eval``/``query`` expressions with strings containing the ``@`` character
+  will now work (:issue:`6366`).
+- Bug in ``Series.reindex`` when specifying a ``method`` with some nan values was inconsistent (noted on a resample) (:issue:`6418`)
+- Bug in :meth:`DataFrame.replace` where nested dicts were erroneously
+  depending on the order of dictionary keys and values (:issue:`5338`).
+- Performance issue in concatenating with empty objects (:issue:`3259`)
+- Clarify sorting of ``sym_diff`` on ``Index`` objects with ``NaN`` values (:issue:`6444`)
+- Regression in ``MultiIndex.from_product`` with a ``DatetimeIndex`` as input (:issue:`6439`)
+- Bug in ``str.extract`` when passed a non-default index (:issue:`6348`)
+- Bug in ``str.split`` when passed ``pat=None`` and ``n=1`` (:issue:`6466`)
+- Bug in ``io.data.DataReader`` when passed ``"F-F_Momentum_Factor"`` and ``data_source="famafrench"`` (:issue:`6460`)
+- Bug in ``sum`` of a ``timedelta64[ns]`` series (:issue:`6462`)
+- Bug in ``resample`` with a timezone and certain offsets (:issue:`6397`)
+- Bug in ``iat/iloc`` with duplicate indices on a Series (:issue:`6493`)
+- Bug in ``read_html`` where nan's were incorrectly being used to indicate
+  missing values in text. Should use the empty string for consistency with the
+  rest of pandas (:issue:`5129`).
+- Bug in ``read_html`` tests where redirected invalid URLs would make one test
+  fail (:issue:`6445`).
+- Bug in multi-axis indexing using ``.loc`` on non-unique indices (:issue:`6504`)
+- Bug that caused _ref_locs corruption when slice indexing across columns axis of a DataFrame (:issue:`6525`)
+- Regression from 0.13 in the treatment of numpy ``datetime64`` non-ns dtypes in Series creation (:issue:`6529`)
+- ``.names`` attribute of MultiIndexes passed to ``set_index`` are now preserved (:issue:`6459`).
+- Bug in setitem with a duplicate index and an alignable rhs (:issue:`6541`)
+- Bug in setitem with ``.loc`` on mixed integer Indexes (:issue:`6546`)
+- Bug in ``pd.read_stata`` which would use the wrong data types and missing values (:issue:`6327`)
+- Bug in ``DataFrame.to_stata`` that lead to data loss in certain cases, and could be exported using the
+  wrong data types and missing values (:issue:`6335`)
+- ``StataWriter`` replaces missing values in string columns by empty string (:issue:`6802`)
+- Inconsistent types in ``Timestamp`` addition/subtraction (:issue:`6543`)
+- Bug in preserving frequency across Timestamp addition/subtraction (:issue:`4547`)
+- Bug in empty list lookup caused ``IndexError`` exceptions (:issue:`6536`, :issue:`6551`)
+- ``Series.quantile`` raising on an ``object`` dtype (:issue:`6555`)
+- Bug in ``.xs`` with a ``nan`` in level when dropped (:issue:`6574`)
+- Bug in fillna with ``method='bfill/ffill'`` and ``datetime64[ns]`` dtype (:issue:`6587`)
+- Bug in sql writing with mixed dtypes possibly leading to data loss (:issue:`6509`)
+- Bug in ``Series.pop`` (:issue:`6600`)
+- Bug in ``iloc`` indexing when positional indexer matched ``Int64Index`` of the corresponding axis and no reordering happened (:issue:`6612`)
+- Bug in ``fillna`` with ``limit`` and ``value`` specified
+- Bug in ``DataFrame.to_stata`` when columns have non-string names (:issue:`4558`)
+- Bug in compat with ``np.compress``, surfaced in (:issue:`6658`)
+- Bug in binary operations with a rhs of a Series not aligning (:issue:`6681`)
+- Bug in ``DataFrame.to_stata`` which incorrectly handles nan values and ignores ``with_index`` keyword argument (:issue:`6685`)
+- Bug in resample with extra bins when using an evenly divisible frequency (:issue:`4076`)
+- Bug in consistency of groupby aggregation when passing a custom function (:issue:`6715`)
+- Bug in resample when ``how=None`` resample freq is the same as the axis frequency (:issue:`5955`)
+- Bug in downcasting inference with empty arrays (:issue:`6733`)
+- Bug in ``obj.blocks`` on sparse containers dropping all but the last items of same for dtype (:issue:`6748`)
+- Bug in unpickling ``NaT (NaTType)`` (:issue:`4606`)
+- Bug in ``DataFrame.replace()`` where regex meta characters were being treated
+  as regex even when ``regex=False`` (:issue:`6777`).
+- Bug in timedelta ops on 32-bit platforms (:issue:`6808`)
+- Bug in setting a tz-aware index directly via ``.index`` (:issue:`6785`)
+- Bug in expressions.py where numexpr would try to evaluate arithmetic ops
+  (:issue:`6762`).
+- Bug in Makefile where it didn't remove Cython generated C files with ``make
+  clean`` (:issue:`6768`)
+- Bug with numpy < 1.7.2 when reading long strings from ``HDFStore`` (:issue:`6166`)
+- Bug in ``DataFrame._reduce`` where non bool-like (0/1) integers were being
+  converted into bools. (:issue:`6806`)
+- Regression from 0.13 with ``fillna`` and a Series on datetime-like (:issue:`6344`)
+- Bug in adding ``np.timedelta64`` to ``DatetimeIndex`` with timezone outputs incorrect results (:issue:`6818`)
+- Bug in ``DataFrame.replace()`` where changing a dtype through replacement
+  would only replace the first occurrence of a value (:issue:`6689`)
+- Better error message when passing a frequency of 'MS' in ``Period`` construction (GH5332)
+- Bug in ``Series.__unicode__`` when ``max_rows=None`` and the Series has more than 1000 rows. (:issue:`6863`)
+- Bug in ``groupby.get_group`` where a datelike wasn't always accepted (:issue:`5267`)
+- Bug in ``groupBy.get_group`` created by ``TimeGrouper`` raises ``AttributeError`` (:issue:`6914`)
+- Bug in ``DatetimeIndex.tz_localize`` and ``DatetimeIndex.tz_convert`` converting ``NaT`` incorrectly (:issue:`5546`)
+- Bug in arithmetic operations affecting ``NaT`` (:issue:`6873`)
+- Bug in ``Series.str.extract`` where the resulting ``Series`` from a single
+  group match wasn't renamed to the group name
+- Bug in ``DataFrame.to_csv`` where setting ``index=False`` ignored the
+  ``header`` kwarg (:issue:`6186`)
+- Bug in ``DataFrame.plot`` and ``Series.plot``, where the legend behave inconsistently when plotting to the same axes repeatedly (:issue:`6678`)
+- Internal tests for patching ``__finalize__`` / bug in merge not finalizing (:issue:`6923`, :issue:`6927`)
+- accept ``TextFileReader`` in ``concat``, which was affecting a common user idiom (:issue:`6583`)
+- Bug in C parser with leading white space (:issue:`3374`)
+- Bug in C parser with ``delim_whitespace=True`` and ``\r``-delimited lines
+- Bug in python parser with explicit MultiIndex in row following column header (:issue:`6893`)
+- Bug in ``Series.rank`` and ``DataFrame.rank`` that caused small floats (<1e-13) to all receive the same rank (:issue:`6886`)
+- Bug in ``DataFrame.apply`` with functions that used ``*args`` or ``**kwargs`` and returned
+  an empty result (:issue:`6952`)
+- Bug in sum/mean on 32-bit platforms on overflows (:issue:`6915`)
+- Moved ``Panel.shift`` to ``NDFrame.slice_shift`` and fixed to respect multiple dtypes. (:issue:`6959`)
+- Bug in enabling ``subplots=True`` in ``DataFrame.plot`` only has single column raises ``TypeError``, and ``Series.plot`` raises ``AttributeError`` (:issue:`6951`)
+- Bug in ``DataFrame.plot`` draws unnecessary axes when enabling ``subplots`` and ``kind=scatter`` (:issue:`6951`)
+- Bug in ``read_csv`` from a filesystem with non-utf-8 encoding (:issue:`6807`)
+- Bug in ``iloc`` when setting / aligning (:issue:`6766`)
+- Bug causing UnicodeEncodeError when get_dummies called with unicode values and a prefix (:issue:`6885`)
+- Bug in timeseries-with-frequency plot cursor display (:issue:`5453`)
+- Bug surfaced in ``groupby.plot`` when using a ``Float64Index`` (:issue:`7025`)
+- Stopped tests from failing if options data isn't able to be downloaded from Yahoo (:issue:`7034`)
+- Bug in ``parallel_coordinates`` and ``radviz`` where reordering of class column
+  caused possible color/class mismatch (:issue:`6956`)
+- Bug in ``radviz`` and ``andrews_curves`` where multiple values of 'color'
+  were being passed to plotting method (:issue:`6956`)
+- Bug in ``Float64Index.isin()`` where containing ``nan`` s would make indices
+  claim that they contained all the things (:issue:`7066`).
+- Bug in ``DataFrame.boxplot`` where it failed to use the axis passed as the ``ax`` argument (:issue:`3578`)
+- Bug in the ``XlsxWriter`` and ``XlwtWriter`` implementations that resulted in datetime columns being formatted without the time (:issue:`7075`)
+  were being passed to plotting method
+- :func:`read_fwf` treats ``None`` in ``colspec`` like regular python slices. It now reads from the beginning
+  or until the end of the line when ``colspec`` contains a ``None`` (previously raised a ``TypeError``)
+- Bug in cache coherence with chained indexing and slicing; add ``_is_view`` property to ``NDFrame`` to correctly predict
+  views; mark ``is_copy`` on ``xs`` only if its an actual copy (and not a view) (:issue:`7084`)
+- Bug in DatetimeIndex creation from string ndarray with ``dayfirst=True`` (:issue:`5917`)
+- Bug in ``MultiIndex.from_arrays`` created from ``DatetimeIndex`` doesn't preserve ``freq`` and ``tz`` (:issue:`7090`)
+- Bug in ``unstack`` raises ``ValueError`` when ``MultiIndex`` contains ``PeriodIndex`` (:issue:`4342`)
+- Bug in ``boxplot`` and ``hist`` draws unnecessary axes (:issue:`6769`)
+- Regression in ``groupby.nth()`` for out-of-bounds indexers (:issue:`6621`)
+- Bug in ``quantile`` with datetime values (:issue:`6965`)
+- Bug in ``Dataframe.set_index``, ``reindex`` and ``pivot`` don't preserve ``DatetimeIndex`` and ``PeriodIndex`` attributes (:issue:`3950`, :issue:`5878`, :issue:`6631`)
+- Bug in ``MultiIndex.get_level_values`` doesn't preserve ``DatetimeIndex`` and ``PeriodIndex`` attributes (:issue:`7092`)
+- Bug in ``Groupby`` doesn't preserve ``tz`` (:issue:`3950`)
+- Bug in ``PeriodIndex`` partial string slicing (:issue:`6716`)
+- Bug in the HTML repr of a truncated Series or DataFrame not showing the class name with the ``large_repr`` set to 'info'
+  (:issue:`7105`)
+- Bug in ``DatetimeIndex`` specifying ``freq`` raises ``ValueError`` when passed value is too short (:issue:`7098`)
+- Fixed a bug with the ``info`` repr not honoring the ``display.max_info_columns`` setting (:issue:`6939`)
+- Bug ``PeriodIndex`` string slicing with out of bounds values (:issue:`5407`)
+- Fixed a memory error in the hashtable implementation/factorizer on resizing of large tables (:issue:`7157`)
+- Bug in ``isnull`` when applied to 0-dimensional object arrays (:issue:`7176`)
+- Bug in ``query``/``eval`` where global constants were not looked up correctly
+  (:issue:`7178`)
+- Bug in recognizing out-of-bounds positional list indexers with ``iloc`` and a multi-axis tuple indexer (:issue:`7189`)
+- Bug in setitem with a single value, MultiIndex and integer indices (:issue:`7190`, :issue:`7218`)
+- Bug in expressions evaluation with reversed ops, showing in series-dataframe ops (:issue:`7198`, :issue:`7192`)
+- Bug in multi-axis indexing with > 2 ndim and a MultiIndex (:issue:`7199`)
+- Fix a bug where invalid eval/query operations would blow the stack (:issue:`5198`)
+
+
+.. _whatsnew_0.14.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.13.1..v0.14.0
diff --git a/pandas/doc/source/whatsnew/v0.14.1.rst b/pandas/doc/source/whatsnew/v0.14.1.rst
new file mode 100644
index 0000000000000..a8f8955c3c1b9
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.14.1.rst
@@ -0,0 +1,284 @@
+.. _whatsnew_0141:
+
+Version 0.14.1 (July 11, 2014)
+------------------------------
+
+{{ header }}
+
+
+This is a minor release from 0.14.0 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+- Highlights include:
+
+  - New methods :meth:`~pandas.DataFrame.select_dtypes` to select columns
+    based on the dtype and :meth:`~pandas.Series.sem` to calculate the
+    standard error of the mean.
+  - Support for dateutil timezones (see :ref:`docs <timeseries.timezone>`).
+  - Support for ignoring full line comments in the :func:`~pandas.read_csv`
+    text parser.
+  - New documentation section on :ref:`Options and Settings <options>`.
+  - Lots of bug fixes.
+
+- :ref:`Enhancements <whatsnew_0141.enhancements>`
+- :ref:`API Changes <whatsnew_0141.api>`
+- :ref:`Performance Improvements <whatsnew_0141.performance>`
+- :ref:`Experimental Changes <whatsnew_0141.experimental>`
+- :ref:`Bug Fixes <whatsnew_0141.bug_fixes>`
+
+.. _whatsnew_0141.api:
+
+API changes
+~~~~~~~~~~~
+
+- Openpyxl now raises a ValueError on construction of the openpyxl writer
+  instead of warning on pandas import (:issue:`7284`).
+
+- For ``StringMethods.extract``, when no match is found, the result - only
+  containing ``NaN`` values - now also has ``dtype=object`` instead of
+  ``float`` (:issue:`7242`)
+
+- ``Period`` objects no longer raise a ``TypeError`` when compared using ``==``
+  with another object that *isn't* a ``Period``. Instead
+  when comparing a ``Period`` with another object using ``==`` if the other
+  object isn't a ``Period`` ``False`` is returned. (:issue:`7376`)
+
+- Previously, the behaviour on resetting the time or not in
+  ``offsets.apply``, ``rollforward`` and ``rollback`` operations differed
+  between offsets. With the support of the ``normalize`` keyword for all offsets(see
+  below) with a default value of False (preserve time), the behaviour changed for certain
+  offsets (BusinessMonthBegin, MonthEnd, BusinessMonthEnd, CustomBusinessMonthEnd,
+  BusinessYearBegin, LastWeekOfMonth, FY5253Quarter, LastWeekOfMonth, Easter):
+
+  .. code-block:: ipython
+
+    In [6]: from pandas.tseries import offsets
+
+    In [7]: d = pd.Timestamp('2014-01-01 09:00')
+
+    # old behaviour < 0.14.1
+    In [8]: d + offsets.MonthEnd()
+    Out[8]: pd.Timestamp('2014-01-31 00:00:00')
+
+  Starting from 0.14.1 all offsets preserve time by default. The old
+  behaviour can be obtained with ``normalize=True``
+
+  .. ipython:: python
+     :suppress:
+
+     import pandas.tseries.offsets as offsets
+
+     d = pd.Timestamp("2014-01-01 09:00")
+
+  .. ipython:: python
+
+     # new behaviour
+     d + offsets.MonthEnd()
+     d + offsets.MonthEnd(normalize=True)
+
+  Note that for the other offsets the default behaviour did not change.
+
+- Add back ``#N/A N/A`` as a default NA value in text parsing, (regression from 0.12) (:issue:`5521`)
+- Raise a ``TypeError`` on inplace-setting with a ``.where`` and a non ``np.nan`` value as this is inconsistent
+  with a set-item expression like ``df[mask] = None`` (:issue:`7656`)
+
+
+.. _whatsnew_0141.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+- Add ``dropna`` argument to ``value_counts`` and ``nunique`` (:issue:`5569`).
+- Add :meth:`~pandas.DataFrame.select_dtypes` method to allow selection of
+  columns based on dtype (:issue:`7316`). See :ref:`the docs <basics.selectdtypes>`.
+- All ``offsets`` supports the ``normalize`` keyword to specify whether
+  ``offsets.apply``, ``rollforward`` and ``rollback`` resets the time (hour,
+  minute, etc) or not (default ``False``, preserves time) (:issue:`7156`):
+
+  .. code-block:: python
+
+     import pandas.tseries.offsets as offsets
+
+     day = offsets.Day()
+     day.apply(pd.Timestamp("2014-01-01 09:00"))
+
+     day = offsets.Day(normalize=True)
+     day.apply(pd.Timestamp("2014-01-01 09:00"))
+
+- ``PeriodIndex`` is represented as the same format as ``DatetimeIndex`` (:issue:`7601`)
+- ``StringMethods`` now work on empty Series (:issue:`7242`)
+- The file parsers ``read_csv`` and ``read_table`` now ignore line comments provided by
+  the parameter ``comment``, which accepts only a single character for the C reader.
+  In particular, they allow for comments before file data begins (:issue:`2685`)
+- Add ``NotImplementedError`` for simultaneous use of ``chunksize`` and ``nrows``
+  for read_csv() (:issue:`6774`).
+- Tests for basic reading of public S3 buckets now exist (:issue:`7281`).
+- ``read_html`` now sports an ``encoding`` argument that is passed to the
+  underlying parser library. You can use this to read non-ascii encoded web
+  pages (:issue:`7323`).
+- ``read_excel`` now supports reading from URLs in the same way
+  that ``read_csv`` does.  (:issue:`6809`)
+- Support for dateutil timezones, which can now be used in the same way as
+  pytz timezones across pandas. (:issue:`4688`)
+
+  .. ipython:: python
+
+     rng = pd.date_range(
+         "3/6/2012 00:00", periods=10, freq="D", tz="dateutil/Europe/London"
+     )
+     rng.tz
+
+  See :ref:`the docs <timeseries.timezone>`.
+
+- Implemented ``sem`` (standard error of the mean) operation for ``Series``,
+  ``DataFrame``, ``Panel``, and ``Groupby`` (:issue:`6897`)
+- Add ``nlargest`` and ``nsmallest`` to the ``Series`` ``groupby`` allowlist,
+  which means you can now use these methods on a ``SeriesGroupBy`` object
+  (:issue:`7053`).
+- All offsets ``apply``, ``rollforward`` and ``rollback`` can now handle ``np.datetime64``, previously results in ``ApplyTypeError`` (:issue:`7452`)
+- ``Period`` and ``PeriodIndex`` can contain ``NaT`` in its values (:issue:`7485`)
+- Support pickling ``Series``, ``DataFrame`` and ``Panel`` objects with
+  non-unique labels along *item* axis (``index``, ``columns`` and ``items``
+  respectively) (:issue:`7370`).
+- Improved inference of datetime/timedelta with mixed null objects. Regression from 0.13.1 in interpretation of an object Index
+  with all null elements (:issue:`7431`)
+
+.. _whatsnew_0141.performance:
+
+Performance
+~~~~~~~~~~~
+- Improvements in dtype inference for numeric operations involving yielding performance gains for dtypes: ``int64``, ``timedelta64``, ``datetime64`` (:issue:`7223`)
+- Improvements in Series.transform for significant performance gains (:issue:`6496`)
+- Improvements in DataFrame.transform with ufuncs and built-in grouper functions for significant performance gains (:issue:`7383`)
+- Regression in groupby aggregation of datetime64 dtypes (:issue:`7555`)
+- Improvements in ``MultiIndex.from_product`` for large iterables (:issue:`7627`)
+
+
+.. _whatsnew_0141.experimental:
+
+Experimental
+~~~~~~~~~~~~
+
+- ``pandas.io.data.Options`` has a new method, ``get_all_data`` method, and now consistently returns a
+  MultiIndexed ``DataFrame`` (:issue:`5602`)
+- ``io.gbq.read_gbq`` and ``io.gbq.to_gbq`` were refactored to remove the
+  dependency on the Google ``bq.py`` command line client. This submodule
+  now uses ``httplib2`` and the Google ``apiclient`` and ``oauth2client`` API client
+  libraries which should be more stable and, therefore, reliable than
+  ``bq.py``. See :ref:`the docs <io.bigquery>`. (:issue:`6937`).
+
+
+.. _whatsnew_0141.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in ``DataFrame.where`` with a symmetric shaped frame and a passed other of a DataFrame (:issue:`7506`)
+- Bug in Panel indexing with a MultiIndex axis (:issue:`7516`)
+- Regression in datetimelike slice indexing with a duplicated index and non-exact end-points (:issue:`7523`)
+- Bug in setitem with list-of-lists and single vs mixed types (:issue:`7551`:)
+- Bug in time ops with non-aligned Series (:issue:`7500`)
+- Bug in timedelta inference when assigning an incomplete Series (:issue:`7592`)
+- Bug in groupby ``.nth`` with a Series and integer-like column name (:issue:`7559`)
+- Bug in ``Series.get`` with a boolean accessor (:issue:`7407`)
+- Bug in ``value_counts`` where ``NaT`` did not qualify as missing (``NaN``) (:issue:`7423`)
+- Bug in ``to_timedelta`` that accepted invalid units and misinterpreted 'm/h' (:issue:`7611`, :issue:`6423`)
+- Bug in line plot doesn't set correct ``xlim`` if ``secondary_y=True`` (:issue:`7459`)
+- Bug in grouped ``hist`` and ``scatter`` plots use old ``figsize`` default (:issue:`7394`)
+- Bug in plotting subplots with ``DataFrame.plot``, ``hist`` clears passed ``ax`` even if the number of subplots is one (:issue:`7391`).
+- Bug in plotting subplots with ``DataFrame.boxplot`` with ``by`` kw raises ``ValueError`` if the number of subplots exceeds 1 (:issue:`7391`).
+- Bug in subplots displays ``ticklabels`` and ``labels`` in different rule (:issue:`5897`)
+- Bug in ``Panel.apply`` with a MultiIndex as an axis (:issue:`7469`)
+- Bug in ``DatetimeIndex.insert`` doesn't preserve ``name`` and ``tz`` (:issue:`7299`)
+- Bug in ``DatetimeIndex.asobject`` doesn't preserve ``name`` (:issue:`7299`)
+- Bug in MultiIndex slicing with datetimelike ranges (strings and Timestamps), (:issue:`7429`)
+- Bug in ``Index.min`` and ``max`` doesn't handle ``nan`` and ``NaT`` properly (:issue:`7261`)
+- Bug in ``PeriodIndex.min/max`` results in ``int`` (:issue:`7609`)
+- Bug in ``resample`` where ``fill_method`` was ignored if you passed ``how`` (:issue:`2073`)
+- Bug in ``TimeGrouper`` doesn't exclude column specified by ``key`` (:issue:`7227`)
+- Bug in ``DataFrame`` and ``Series`` bar and barh plot raises ``TypeError`` when ``bottom``
+  and ``left`` keyword is specified (:issue:`7226`)
+- Bug in ``DataFrame.hist`` raises ``TypeError`` when it contains non numeric column (:issue:`7277`)
+- Bug in ``Index.delete`` does not preserve ``name`` and ``freq`` attributes (:issue:`7302`)
+- Bug in ``DataFrame.query()``/``eval`` where local string variables with the @
+  sign were being treated as temporaries attempting to be deleted
+  (:issue:`7300`).
+- Bug in ``Float64Index`` which didn't allow duplicates (:issue:`7149`).
+- Bug in ``DataFrame.replace()`` where truthy values were being replaced
+  (:issue:`7140`).
+- Bug in ``StringMethods.extract()`` where a single match group Series
+  would use the matcher's name instead of the group name (:issue:`7313`).
+- Bug in ``isnull()`` when ``mode.use_inf_as_null == True`` where isnull
+  wouldn't test ``True`` when it encountered an ``inf``/``-inf``
+  (:issue:`7315`).
+- Bug in inferred_freq results in None for eastern hemisphere timezones (:issue:`7310`)
+- Bug in ``Easter`` returns incorrect date when offset is negative (:issue:`7195`)
+- Bug in broadcasting with ``.div``, integer dtypes and divide-by-zero (:issue:`7325`)
+- Bug in ``CustomBusinessDay.apply`` raises ``NameError`` when ``np.datetime64`` object is passed (:issue:`7196`)
+- Bug in ``MultiIndex.append``, ``concat`` and ``pivot_table`` don't preserve timezone (:issue:`6606`)
+- Bug in ``.loc`` with a list of indexers on a single-multi index level (that is not nested) (:issue:`7349`)
+- Bug in ``Series.map`` when mapping a dict with tuple keys of different lengths (:issue:`7333`)
+- Bug all ``StringMethods`` now work on empty Series (:issue:`7242`)
+- Fix delegation of ``read_sql`` to ``read_sql_query`` when query does not contain 'select' (:issue:`7324`).
+- Bug where a string column name assignment to a ``DataFrame`` with a
+  ``Float64Index`` raised a ``TypeError`` during a call to ``np.isnan``
+  (:issue:`7366`).
+- Bug where ``NDFrame.replace()`` didn't correctly replace objects with
+  ``Period`` values (:issue:`7379`).
+- Bug in ``.ix`` getitem should always return a Series (:issue:`7150`)
+- Bug in MultiIndex slicing with incomplete indexers (:issue:`7399`)
+- Bug in MultiIndex slicing with a step in a sliced level (:issue:`7400`)
+- Bug where negative indexers in ``DatetimeIndex`` were not correctly sliced
+  (:issue:`7408`)
+- Bug where ``NaT`` wasn't repr'd correctly in a ``MultiIndex`` (:issue:`7406`,
+  :issue:`7409`).
+- Bug where bool objects were converted to ``nan`` in ``convert_objects``
+  (:issue:`7416`).
+- Bug in ``quantile`` ignoring the axis keyword argument (:issue:`7306`)
+- Bug where ``nanops._maybe_null_out`` doesn't work with complex numbers
+  (:issue:`7353`)
+- Bug in several ``nanops`` functions when ``axis==0`` for
+  1-dimensional ``nan`` arrays (:issue:`7354`)
+- Bug where ``nanops.nanmedian`` doesn't work when ``axis==None``
+  (:issue:`7352`)
+- Bug where ``nanops._has_infs`` doesn't work with many dtypes
+  (:issue:`7357`)
+- Bug in ``StataReader.data`` where reading a 0-observation dta failed (:issue:`7369`)
+- Bug in ``StataReader`` when reading Stata 13 (117) files containing fixed width strings (:issue:`7360`)
+- Bug in ``StataWriter`` where encoding was ignored (:issue:`7286`)
+- Bug in ``DatetimeIndex`` comparison doesn't handle ``NaT`` properly (:issue:`7529`)
+- Bug in passing input with ``tzinfo`` to some offsets ``apply``, ``rollforward`` or ``rollback`` resets ``tzinfo`` or raises ``ValueError`` (:issue:`7465`)
+- Bug in ``DatetimeIndex.to_period``, ``PeriodIndex.asobject``, ``PeriodIndex.to_timestamp`` doesn't preserve ``name`` (:issue:`7485`)
+- Bug in ``DatetimeIndex.to_period`` and ``PeriodIndex.to_timestamp`` handle ``NaT`` incorrectly (:issue:`7228`)
+- Bug in ``offsets.apply``, ``rollforward`` and ``rollback`` may return normal ``datetime`` (:issue:`7502`)
+- Bug in ``resample`` raises ``ValueError`` when target contains ``NaT`` (:issue:`7227`)
+- Bug in ``Timestamp.tz_localize`` resets ``nanosecond`` info (:issue:`7534`)
+- Bug in ``DatetimeIndex.asobject`` raises ``ValueError`` when it contains ``NaT`` (:issue:`7539`)
+- Bug in ``Timestamp.__new__`` doesn't preserve nanosecond properly (:issue:`7610`)
+- Bug in ``Index.astype(float)`` where it would return an ``object`` dtype
+  ``Index`` (:issue:`7464`).
+- Bug in ``DataFrame.reset_index`` loses ``tz`` (:issue:`3950`)
+- Bug in ``DatetimeIndex.freqstr`` raises ``AttributeError`` when ``freq`` is ``None`` (:issue:`7606`)
+- Bug in ``GroupBy.size`` created by ``TimeGrouper`` raises ``AttributeError`` (:issue:`7453`)
+- Bug in single column bar plot is misaligned (:issue:`7498`).
+- Bug in area plot with tz-aware time series raises ``ValueError`` (:issue:`7471`)
+- Bug in non-monotonic ``Index.union`` may preserve ``name`` incorrectly (:issue:`7458`)
+- Bug in ``DatetimeIndex.intersection`` doesn't preserve timezone (:issue:`4690`)
+- Bug in ``rolling_var`` where a window larger than the array would raise an error(:issue:`7297`)
+- Bug with last plotted timeseries dictating ``xlim`` (:issue:`2960`)
+- Bug with ``secondary_y`` axis not being considered for timeseries ``xlim`` (:issue:`3490`)
+- Bug in ``Float64Index`` assignment with a non scalar indexer (:issue:`7586`)
+- Bug in ``pandas.core.strings.str_contains`` does not properly match in a case insensitive fashion when ``regex=False`` and ``case=False`` (:issue:`7505`)
+- Bug in ``expanding_cov``, ``expanding_corr``, ``rolling_cov``, and ``rolling_corr`` for two arguments with mismatched index  (:issue:`7512`)
+- Bug in ``to_sql`` taking the boolean column as text column (:issue:`7678`)
+- Bug in grouped ``hist`` doesn't handle ``rot`` kw and ``sharex`` kw properly (:issue:`7234`)
+- Bug in ``.loc`` performing fallback integer indexing with ``object`` dtype indices (:issue:`7496`)
+- Bug (regression) in ``PeriodIndex`` constructor when passed ``Series`` objects (:issue:`7701`).
+
+
+.. _whatsnew_0.14.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.14.0..v0.14.1
diff --git a/pandas/doc/source/whatsnew/v0.15.0.rst b/pandas/doc/source/whatsnew/v0.15.0.rst
new file mode 100644
index 0000000000000..1ee7c5cbc6b9e
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.15.0.rst
@@ -0,0 +1,1303 @@
+.. _whatsnew_0150:
+
+Version 0.15.0 (October 18, 2014)
+---------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.14.1 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+.. warning::
+
+   pandas >= 0.15.0 will no longer support compatibility with NumPy versions <
+   1.7.0. If you want to use the latest versions of pandas, please upgrade to
+   NumPy >= 1.7.0 (:issue:`7711`)
+
+- Highlights include:
+
+  - The ``Categorical`` type was integrated as a first-class pandas type, see :ref:`here <whatsnew_0150.cat>`
+  - New scalar type ``Timedelta``, and a new index type ``TimedeltaIndex``, see :ref:`here <whatsnew_0150.timedeltaindex>`
+  - New datetimelike properties accessor ``.dt`` for Series, see :ref:`Datetimelike Properties <whatsnew_0150.dt>`
+  - New DataFrame default display for ``df.info()`` to include memory usage, see :ref:`Memory Usage <whatsnew_0150.memory>`
+  - ``read_csv`` will now by default ignore blank lines when parsing, see :ref:`here <whatsnew_0150.blanklines>`
+  - API change in using Indexes in set operations, see :ref:`here <whatsnew_0150.index_set_ops>`
+  - Enhancements in the handling of timezones, see :ref:`here <whatsnew_0150.tz>`
+  - A lot of improvements to the rolling and expanding moment functions, see :ref:`here <whatsnew_0150.roll>`
+  - Internal refactoring of the ``Index`` class to no longer sub-class ``ndarray``, see :ref:`Internal Refactoring <whatsnew_0150.refactoring>`
+  - dropping support for ``PyTables`` less than version 3.0.0, and ``numexpr`` less than version 2.1 (:issue:`7990`)
+  - Split indexing documentation into :ref:`Indexing and Selecting Data <indexing>` and :ref:`MultiIndex / Advanced Indexing <advanced>`
+  - Split out string methods documentation into :ref:`Working with Text Data <text>`
+
+- Check the :ref:`API Changes <whatsnew_0150.api>` and :ref:`deprecations <whatsnew_0150.deprecations>` before updating
+
+- :ref:`Other Enhancements <whatsnew_0150.enhancements>`
+
+- :ref:`Performance Improvements <whatsnew_0150.performance>`
+
+- :ref:`Bug Fixes <whatsnew_0150.bug_fixes>`
+
+.. warning::
+
+   In 0.15.0 ``Index`` has internally been refactored to no longer sub-class ``ndarray``
+   but instead subclass ``PandasObject``, similarly to the rest of the pandas objects. This change allows very easy sub-classing and creation of new index types. This should be
+   a transparent change with only very limited API implications (See the :ref:`Internal Refactoring <whatsnew_0150.refactoring>`)
+
+.. warning::
+
+   The refactoring in :class:`~pandas.Categorical` changed the two argument constructor from
+   "codes/labels and levels" to "values and levels (now called 'categories')". This can lead to subtle bugs. If you use
+   :class:`~pandas.Categorical` directly, please audit your code before updating to this pandas
+   version and change it to use the :meth:`~pandas.Categorical.from_codes` constructor. See more on ``Categorical`` :ref:`here <whatsnew_0150.cat>`
+
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0150.cat:
+
+Categoricals in Series/DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`~pandas.Categorical` can now be included in ``Series`` and ``DataFrames`` and gained new
+methods to manipulate. Thanks to Jan Schulz for much of this API/implementation. (:issue:`3943`, :issue:`5313`, :issue:`5314`,
+:issue:`7444`, :issue:`7839`, :issue:`7848`, :issue:`7864`, :issue:`7914`, :issue:`7768`, :issue:`8006`, :issue:`3678`,
+:issue:`8075`, :issue:`8076`, :issue:`8143`, :issue:`8453`, :issue:`8518`).
+
+For full docs, see the :ref:`categorical introduction <categorical>` and the
+:ref:`API documentation <api.arrays.categorical>`.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"id": [1, 2, 3, 4, 5, 6],
+                       "raw_grade": ['a', 'b', 'b', 'a', 'a', 'e']})
+
+    df["grade"] = df["raw_grade"].astype("category")
+    df["grade"]
+
+    # Rename the categories
+    df["grade"] = df["grade"].cat.rename_categories(["very good", "good", "very bad"])
+
+    # Reorder the categories and simultaneously add the missing categories
+    df["grade"] = df["grade"].cat.set_categories(["very bad", "bad",
+                                                  "medium", "good", "very good"])
+    df["grade"]
+    df.sort_values("grade")
+    df.groupby("grade", observed=False).size()
+
+- ``pandas.core.group_agg`` and ``pandas.core.factor_agg`` were removed. As an alternative, construct
+  a dataframe and use ``df.groupby(<group>).agg(<func>)``.
+
+- Supplying "codes/labels and levels" to the :class:`~pandas.Categorical` constructor is not
+  supported anymore. Supplying two arguments to the constructor is now interpreted as
+  "values and levels (now called 'categories')". Please change your code to use the :meth:`~pandas.Categorical.from_codes`
+  constructor.
+
+- The ``Categorical.labels`` attribute was renamed to ``Categorical.codes`` and is read
+  only. If you want to manipulate codes, please use one of the
+  :ref:`API methods on Categoricals <api.arrays.categorical>`.
+
+- The ``Categorical.levels`` attribute is renamed to ``Categorical.categories``.
+
+
+.. _whatsnew_0150.timedeltaindex:
+
+TimedeltaIndex/scalar
+^^^^^^^^^^^^^^^^^^^^^
+
+We introduce a new scalar type ``Timedelta``, which is a subclass of ``datetime.timedelta``, and behaves in a similar manner,
+but allows compatibility with ``np.timedelta64`` types as well as a host of custom representation, parsing, and attributes.
+This type is very similar to how ``Timestamp`` works for ``datetimes``. It is a nice-API box for the type. See the :ref:`docs <timedeltas.timedeltas>`.
+(:issue:`3009`, :issue:`4533`, :issue:`8209`, :issue:`8187`, :issue:`8190`, :issue:`7869`, :issue:`7661`, :issue:`8345`, :issue:`8471`)
+
+.. warning::
+
+   ``Timedelta`` scalars (and ``TimedeltaIndex``) component fields are *not the same* as the component fields on a ``datetime.timedelta`` object. For example, ``.seconds`` on a ``datetime.timedelta`` object returns the total number of seconds combined between ``hours``, ``minutes`` and ``seconds``. In contrast, the pandas ``Timedelta`` breaks out hours, minutes, microseconds and nanoseconds separately.
+
+   .. code-block:: ipython
+
+      # Timedelta accessor
+      In [9]: tds = pd.Timedelta('31 days 5 min 3 sec')
+
+      In [10]: tds.minutes
+      Out[10]: 5L
+
+      In [11]: tds.seconds
+      Out[11]: 3L
+
+      # datetime.timedelta accessor
+      # this is 5 minutes * 60 + 3 seconds
+      In [12]: tds.to_pytimedelta().seconds
+      Out[12]: 303
+
+   **Note**: this is no longer true starting from v0.16.0, where full
+   compatibility with ``datetime.timedelta`` is introduced. See the
+   :ref:`0.16.0 whatsnew entry <whatsnew_0160.api_breaking.timedelta>`
+
+.. warning::
+
+       Prior to 0.15.0 ``pd.to_timedelta`` would return a ``Series`` for list-like/Series input, and a ``np.timedelta64`` for scalar input.
+       It will now return a ``TimedeltaIndex`` for list-like input, ``Series`` for Series input, and ``Timedelta`` for scalar input.
+
+       The arguments to ``pd.to_timedelta`` are now ``(arg,unit='ns',box=True,coerce=False)``, previously were ``(arg,box=True,unit='ns')`` as these are more logical.
+
+Construct a scalar
+
+.. ipython:: python
+
+   pd.Timedelta('1 days 06:05:01.00003')
+   pd.Timedelta('15.5us')
+   pd.Timedelta('1 hour 15.5us')
+
+   # negative Timedeltas have this string repr
+   # to be more consistent with datetime.timedelta conventions
+   pd.Timedelta('-1us')
+
+   # a NaT
+   pd.Timedelta('nan')
+
+Access fields for a ``Timedelta``
+
+.. ipython:: python
+
+   td = pd.Timedelta('1 hour 3m 15.5us')
+   td.seconds
+   td.microseconds
+   td.nanoseconds
+
+Construct a ``TimedeltaIndex``
+
+.. ipython:: python
+   :suppress:
+
+   import datetime
+
+.. ipython:: python
+
+   pd.TimedeltaIndex(['1 days', '1 days, 00:00:05',
+                      np.timedelta64(2, 'D'),
+                      datetime.timedelta(days=2, seconds=2)])
+
+Constructing a ``TimedeltaIndex`` with a regular range
+
+.. ipython:: python
+
+   pd.timedelta_range('1 days', periods=5, freq='D')
+
+.. code-block:: python
+
+   In [20]: pd.timedelta_range(start='1 days', end='2 days', freq='30T')
+   Out[20]:
+   TimedeltaIndex(['1 days 00:00:00', '1 days 00:30:00', '1 days 01:00:00',
+                   '1 days 01:30:00', '1 days 02:00:00', '1 days 02:30:00',
+                   '1 days 03:00:00', '1 days 03:30:00', '1 days 04:00:00',
+                   '1 days 04:30:00', '1 days 05:00:00', '1 days 05:30:00',
+                   '1 days 06:00:00', '1 days 06:30:00', '1 days 07:00:00',
+                   '1 days 07:30:00', '1 days 08:00:00', '1 days 08:30:00',
+                   '1 days 09:00:00', '1 days 09:30:00', '1 days 10:00:00',
+                   '1 days 10:30:00', '1 days 11:00:00', '1 days 11:30:00',
+                   '1 days 12:00:00', '1 days 12:30:00', '1 days 13:00:00',
+                   '1 days 13:30:00', '1 days 14:00:00', '1 days 14:30:00',
+                   '1 days 15:00:00', '1 days 15:30:00', '1 days 16:00:00',
+                   '1 days 16:30:00', '1 days 17:00:00', '1 days 17:30:00',
+                   '1 days 18:00:00', '1 days 18:30:00', '1 days 19:00:00',
+                   '1 days 19:30:00', '1 days 20:00:00', '1 days 20:30:00',
+                   '1 days 21:00:00', '1 days 21:30:00', '1 days 22:00:00',
+                   '1 days 22:30:00', '1 days 23:00:00', '1 days 23:30:00',
+                   '2 days 00:00:00'],
+                  dtype='timedelta64[ns]', freq='30T')
+
+You can now use a ``TimedeltaIndex`` as the index of a pandas object
+
+.. ipython:: python
+
+   s = pd.Series(np.arange(5),
+                 index=pd.timedelta_range('1 days', periods=5, freq='s'))
+   s
+
+You can select with partial string selections
+
+.. ipython:: python
+
+   s['1 day 00:00:02']
+   s['1 day':'1 day 00:00:02']
+
+Finally, the combination of ``TimedeltaIndex`` with ``DatetimeIndex`` allow certain combination operations that are ``NaT`` preserving:
+
+.. ipython:: python
+
+   tdi = pd.TimedeltaIndex(['1 days', pd.NaT, '2 days'])
+   tdi.tolist()
+   dti = pd.date_range('20130101', periods=3)
+   dti.tolist()
+
+   (dti + tdi).tolist()
+   (dti - tdi).tolist()
+
+- iteration of a ``Series`` e.g. ``list(Series(...))`` of ``timedelta64[ns]`` would prior to v0.15.0 return ``np.timedelta64`` for each element. These will now be wrapped in ``Timedelta``.
+
+
+.. _whatsnew_0150.memory:
+
+Memory usage
+^^^^^^^^^^^^
+
+Implemented methods to find memory usage of a DataFrame. See the :ref:`FAQ <df-memory-usage>` for more. (:issue:`6852`).
+
+A new display option ``display.memory_usage`` (see :ref:`options`) sets the default behavior of the ``memory_usage`` argument in the ``df.info()`` method. By default ``display.memory_usage`` is ``True``.
+
+.. ipython:: python
+
+    dtypes = ['int64', 'float64', 'datetime64[ns]', 'timedelta64[ns]',
+              'complex128', 'object', 'bool']
+    n = 5000
+    data = {t: np.random.randint(100, size=n).astype(t) for t in dtypes}
+    df = pd.DataFrame(data)
+    df['categorical'] = df['object'].astype('category')
+
+    df.info()
+
+Additionally :meth:`~pandas.DataFrame.memory_usage` is an available method for a dataframe object which returns the memory usage of each column.
+
+.. ipython:: python
+
+    df.memory_usage(index=True)
+
+
+.. _whatsnew_0150.dt:
+
+Series.dt accessor
+^^^^^^^^^^^^^^^^^^
+
+``Series`` has gained an accessor to succinctly return datetime like properties for the *values* of the Series, if its a datetime/period like Series. (:issue:`7207`)
+This will return a Series, indexed like the existing Series. See the :ref:`docs <basics.dt_accessors>`
+
+.. ipython:: python
+
+   # datetime
+   s = pd.Series(pd.date_range('20130101 09:10:12', periods=4))
+   s
+   s.dt.hour
+   s.dt.second
+   s.dt.day
+   s.dt.freq
+
+This enables nice expressions like this:
+
+.. ipython:: python
+
+   s[s.dt.day == 2]
+
+You can easily produce tz aware transformations:
+
+.. ipython:: python
+
+   stz = s.dt.tz_localize('US/Eastern')
+   stz
+   stz.dt.tz
+
+You can also chain these types of operations:
+
+.. ipython:: python
+
+   s.dt.tz_localize('UTC').dt.tz_convert('US/Eastern')
+
+The ``.dt`` accessor works for period and timedelta dtypes.
+
+.. ipython:: python
+
+   # period
+   s = pd.Series(pd.period_range('20130101', periods=4, freq='D'))
+   s
+   s.dt.year
+   s.dt.day
+
+.. ipython:: python
+
+   # timedelta
+   s = pd.Series(pd.timedelta_range('1 day 00:00:05', periods=4, freq='s'))
+   s
+   s.dt.days
+   s.dt.seconds
+   s.dt.components
+
+
+.. _whatsnew_0150.tz:
+
+Timezone handling improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- ``tz_localize(None)`` for tz-aware ``Timestamp`` and ``DatetimeIndex`` now removes timezone holding local time,
+  previously this resulted in ``Exception`` or ``TypeError`` (:issue:`7812`)
+
+  .. code-block:: ipython
+
+     In [58]: ts = pd.Timestamp('2014-08-01 09:00', tz='US/Eastern')
+
+     In[59]: ts
+     Out[59]: Timestamp('2014-08-01 09:00:00-0400', tz='US/Eastern')
+
+     In [60]: ts.tz_localize(None)
+     Out[60]: Timestamp('2014-08-01 09:00:00')
+
+     In [61]: didx = pd.date_range(start='2014-08-01 09:00', freq='H',
+        ....:                      periods=10, tz='US/Eastern')
+        ....:
+
+     In [62]: didx
+     Out[62]:
+     DatetimeIndex(['2014-08-01 09:00:00-04:00', '2014-08-01 10:00:00-04:00',
+                    '2014-08-01 11:00:00-04:00', '2014-08-01 12:00:00-04:00',
+                    '2014-08-01 13:00:00-04:00', '2014-08-01 14:00:00-04:00',
+                    '2014-08-01 15:00:00-04:00', '2014-08-01 16:00:00-04:00',
+                    '2014-08-01 17:00:00-04:00', '2014-08-01 18:00:00-04:00'],
+                   dtype='datetime64[ns, US/Eastern]', freq='H')
+
+     In [63]: didx.tz_localize(None)
+     Out[63]:
+     DatetimeIndex(['2014-08-01 09:00:00', '2014-08-01 10:00:00',
+                    '2014-08-01 11:00:00', '2014-08-01 12:00:00',
+                    '2014-08-01 13:00:00', '2014-08-01 14:00:00',
+                    '2014-08-01 15:00:00', '2014-08-01 16:00:00',
+                    '2014-08-01 17:00:00', '2014-08-01 18:00:00'],
+                   dtype='datetime64[ns]', freq=None)
+
+- ``tz_localize`` now accepts the ``ambiguous`` keyword which allows for passing an array of bools
+  indicating whether the date belongs in DST or not, 'NaT' for setting transition times to NaT,
+  'infer' for inferring DST/non-DST, and 'raise' (default) for an ``AmbiguousTimeError`` to be raised. See :ref:`the docs<timeseries.timezone_ambiguous>` for more details (:issue:`7943`)
+
+- ``DataFrame.tz_localize`` and ``DataFrame.tz_convert`` now accepts an optional ``level`` argument
+  for localizing a specific level of a MultiIndex (:issue:`7846`)
+
+- ``Timestamp.tz_localize`` and ``Timestamp.tz_convert`` now raise ``TypeError`` in error cases, rather than ``Exception`` (:issue:`8025`)
+
+- a timeseries/index localized to UTC when inserted into a Series/DataFrame will preserve the UTC timezone (rather than being a naive ``datetime64[ns]``) as ``object`` dtype (:issue:`8411`)
+
+- ``Timestamp.__repr__`` displays ``dateutil.tz.tzoffset`` info (:issue:`7907`)
+
+
+.. _whatsnew_0150.roll:
+
+Rolling/expanding moments improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- :func:`rolling_min`, :func:`rolling_max`, :func:`rolling_cov`, and :func:`rolling_corr`
+  now return objects with all ``NaN`` when ``len(arg) < min_periods <= window`` rather
+  than raising. (This makes all rolling functions consistent in this behavior). (:issue:`7766`)
+
+  Prior to 0.15.0
+
+  .. ipython:: python
+
+     s = pd.Series([10, 11, 12, 13])
+
+  .. code-block:: ipython
+
+     In [15]: pd.rolling_min(s, window=10, min_periods=5)
+     ValueError: min_periods (5) must be <= window (4)
+
+  New behavior
+
+  .. code-block:: ipython
+
+     In [4]: pd.rolling_min(s, window=10, min_periods=5)
+     Out[4]:
+     0   NaN
+     1   NaN
+     2   NaN
+     3   NaN
+     dtype: float64
+
+- :func:`rolling_max`, :func:`rolling_min`, :func:`rolling_sum`, :func:`rolling_mean`, :func:`rolling_median`,
+  :func:`rolling_std`, :func:`rolling_var`, :func:`rolling_skew`, :func:`rolling_kurt`, :func:`rolling_quantile`,
+  :func:`rolling_cov`, :func:`rolling_corr`, :func:`rolling_corr_pairwise`,
+  :func:`rolling_window`, and :func:`rolling_apply` with ``center=True`` previously would return a result of the same
+  structure as the input ``arg`` with ``NaN`` in the final ``(window-1)/2`` entries.
+
+  Now the final ``(window-1)/2`` entries of the result are calculated as if the input ``arg`` were followed
+  by ``(window-1)/2`` ``NaN`` values (or with shrinking windows, in the case of :func:`rolling_apply`).
+  (:issue:`7925`, :issue:`8269`)
+
+  Prior behavior (note final value is ``NaN``):
+
+  .. code-block:: ipython
+
+    In [7]: pd.rolling_sum(Series(range(4)), window=3, min_periods=0, center=True)
+    Out[7]:
+    0     1
+    1     3
+    2     6
+    3   NaN
+    dtype: float64
+
+  New behavior (note final value is ``5 = sum([2, 3, NaN])``):
+
+  .. code-block:: ipython
+
+     In [7]: pd.rolling_sum(pd.Series(range(4)), window=3,
+       ....:                min_periods=0, center=True)
+     Out[7]:
+     0    1
+     1    3
+     2    6
+     3    5
+     dtype: float64
+
+- :func:`rolling_window` now normalizes the weights properly in rolling mean mode (``mean=True``) so that
+  the calculated weighted means (e.g. 'triang', 'gaussian') are distributed about the same means as those
+  calculated without weighting (i.e. 'boxcar'). See :ref:`the note on normalization <window.weighted>` for further details. (:issue:`7618`)
+
+  .. ipython:: python
+
+    s = pd.Series([10.5, 8.8, 11.4, 9.7, 9.3])
+
+  Behavior prior to 0.15.0:
+
+  .. code-block:: ipython
+
+     In [39]: pd.rolling_window(s, window=3, win_type='triang', center=True)
+     Out[39]:
+     0         NaN
+     1    6.583333
+     2    6.883333
+     3    6.683333
+     4         NaN
+     dtype: float64
+
+  New behavior
+
+  .. code-block:: ipython
+
+     In [10]: pd.rolling_window(s, window=3, win_type='triang', center=True)
+     Out[10]:
+     0       NaN
+     1     9.875
+     2    10.325
+     3    10.025
+     4       NaN
+     dtype: float64
+
+- Removed ``center`` argument from all :func:`expanding_ <expanding_apply>` functions (see :ref:`list <api.functions_expanding>`),
+  as the results produced when ``center=True`` did not make much sense. (:issue:`7925`)
+
+- Added optional ``ddof`` argument to :func:`expanding_cov` and :func:`rolling_cov`.
+  The default value of ``1`` is backwards-compatible. (:issue:`8279`)
+
+- Documented the ``ddof`` argument to :func:`expanding_var`, :func:`expanding_std`,
+  :func:`rolling_var`, and :func:`rolling_std`. These functions' support of a
+  ``ddof`` argument (with a default value of ``1``) was previously undocumented. (:issue:`8064`)
+
+- :func:`ewma`, :func:`ewmstd`, :func:`ewmvol`, :func:`ewmvar`, :func:`ewmcov`, and :func:`ewmcorr`
+  now interpret ``min_periods`` in the same manner that the :func:`rolling_*` and :func:`expanding_*` functions do:
+  a given result entry will be ``NaN`` if the (expanding, in this case) window does not contain
+  at least ``min_periods`` values. The previous behavior was to set to ``NaN`` the ``min_periods`` entries
+  starting with the first non- ``NaN`` value. (:issue:`7977`)
+
+  Prior behavior (note values start at index ``2``, which is ``min_periods`` after index ``0``
+  (the index of the first non-empty value)):
+
+  .. ipython:: python
+
+    s  = pd.Series([1, None, None, None, 2, 3])
+
+  .. code-block:: ipython
+
+        In [51]: pd.ewma(s, com=3., min_periods=2)
+        Out[51]:
+        0         NaN
+        1         NaN
+        2    1.000000
+        3    1.000000
+        4    1.571429
+        5    2.189189
+        dtype: float64
+
+  New behavior (note values start at index ``4``, the location of the 2nd (since ``min_periods=2``) non-empty value):
+
+  .. code-block:: ipython
+
+     In [2]: pd.ewma(s, com=3., min_periods=2)
+     Out[2]:
+     0         NaN
+     1         NaN
+     2         NaN
+     3         NaN
+     4    1.759644
+     5    2.383784
+     dtype: float64
+
+- :func:`ewmstd`, :func:`ewmvol`, :func:`ewmvar`, :func:`ewmcov`, and :func:`ewmcorr`
+  now have an optional ``adjust`` argument, just like :func:`ewma` does,
+  affecting how the weights are calculated.
+  The default value of ``adjust`` is ``True``, which is backwards-compatible.
+  See :ref:`Exponentially weighted moment functions <window.exponentially_weighted>` for details. (:issue:`7911`)
+
+- :func:`ewma`, :func:`ewmstd`, :func:`ewmvol`, :func:`ewmvar`, :func:`ewmcov`, and :func:`ewmcorr`
+  now have an optional ``ignore_na`` argument.
+  When ``ignore_na=False`` (the default), missing values are taken into account in the weights calculation.
+  When ``ignore_na=True`` (which reproduces the pre-0.15.0 behavior), missing values are ignored in the weights calculation.
+  (:issue:`7543`)
+
+  .. code-block:: ipython
+
+     In [7]: pd.ewma(pd.Series([None, 1., 8.]), com=2.)
+     Out[7]:
+     0    NaN
+     1    1.0
+     2    5.2
+     dtype: float64
+
+     In [8]: pd.ewma(pd.Series([1., None, 8.]), com=2.,
+       ....:         ignore_na=True)  # pre-0.15.0 behavior
+     Out[8]:
+     0    1.0
+     1    1.0
+     2    5.2
+     dtype: float64
+
+     In [9]: pd.ewma(pd.Series([1., None, 8.]), com=2.,
+       ....:         ignore_na=False)  # new default
+     Out[9]:
+     0    1.000000
+     1    1.000000
+     2    5.846154
+     dtype: float64
+
+  .. warning::
+
+     By default (``ignore_na=False``) the :func:`ewm*` functions' weights calculation
+     in the presence of missing values is different than in pre-0.15.0 versions.
+     To reproduce the pre-0.15.0 calculation of weights in the presence of missing values
+     one must specify explicitly ``ignore_na=True``.
+
+- Bug in :func:`expanding_cov`, :func:`expanding_corr`, :func:`rolling_cov`, :func:`rolling_cor`, :func:`ewmcov`, and :func:`ewmcorr`
+  returning results with columns sorted by name and producing an error for non-unique columns;
+  now handles non-unique columns and returns columns in original order
+  (except for the case of two DataFrames with ``pairwise=False``, where behavior is unchanged) (:issue:`7542`)
+- Bug in :func:`rolling_count` and :func:`expanding_*` functions unnecessarily producing error message for zero-length data (:issue:`8056`)
+- Bug in :func:`rolling_apply` and :func:`expanding_apply` interpreting ``min_periods=0`` as ``min_periods=1`` (:issue:`8080`)
+- Bug in :func:`expanding_std` and :func:`expanding_var` for a single value producing a confusing error message (:issue:`7900`)
+- Bug in :func:`rolling_std` and :func:`rolling_var` for a single value producing ``0`` rather than ``NaN`` (:issue:`7900`)
+
+- Bug in :func:`ewmstd`, :func:`ewmvol`, :func:`ewmvar`, and :func:`ewmcov`
+  calculation of de-biasing factors when ``bias=False`` (the default).
+  Previously an incorrect constant factor was used, based on ``adjust=True``, ``ignore_na=True``,
+  and an infinite number of observations.
+  Now a different factor is used for each entry, based on the actual weights
+  (analogous to the usual ``N/(N-1)`` factor).
+  In particular, for a single point a value of ``NaN`` is returned when ``bias=False``,
+  whereas previously a value of (approximately) ``0`` was returned.
+
+  For example, consider the following pre-0.15.0 results for ``ewmvar(..., bias=False)``,
+  and the corresponding debiasing factors:
+
+  .. ipython:: python
+
+     s = pd.Series([1., 2., 0., 4.])
+
+  .. code-block:: ipython
+
+         In [89]: pd.ewmvar(s, com=2., bias=False)
+         Out[89]:
+         0   -2.775558e-16
+         1    3.000000e-01
+         2    9.556787e-01
+         3    3.585799e+00
+         dtype: float64
+
+         In [90]: pd.ewmvar(s, com=2., bias=False) / pd.ewmvar(s, com=2., bias=True)
+         Out[90]:
+         0    1.25
+         1    1.25
+         2    1.25
+         3    1.25
+         dtype: float64
+
+  Note that entry ``0`` is approximately 0, and the debiasing factors are a constant 1.25.
+  By comparison, the following 0.15.0 results have a ``NaN`` for entry ``0``,
+  and the debiasing factors are decreasing (towards 1.25):
+
+  .. code-block:: ipython
+
+     In [14]: pd.ewmvar(s, com=2., bias=False)
+     Out[14]:
+     0         NaN
+     1    0.500000
+     2    1.210526
+     3    4.089069
+     dtype: float64
+
+     In [15]: pd.ewmvar(s, com=2., bias=False) / pd.ewmvar(s, com=2., bias=True)
+     Out[15]:
+     0         NaN
+     1    2.083333
+     2    1.583333
+     3    1.425439
+     dtype: float64
+
+  See :ref:`Exponentially weighted moment functions <window.exponentially_weighted>` for details. (:issue:`7912`)
+
+
+.. _whatsnew_0150.sql:
+
+Improvements in the SQL IO module
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Added support for a ``chunksize`` parameter to ``to_sql`` function. This allows DataFrame to be written in chunks and avoid packet-size overflow errors (:issue:`8062`).
+- Added support for a ``chunksize`` parameter to ``read_sql`` function. Specifying this argument will return an iterator through chunks of the query result (:issue:`2908`).
+- Added support for writing ``datetime.date`` and ``datetime.time`` object columns with ``to_sql`` (:issue:`6932`).
+- Added support for specifying a ``schema`` to read from/write to with ``read_sql_table`` and ``to_sql`` (:issue:`7441`, :issue:`7952`).
+  For example:
+
+  .. code-block:: python
+
+         df.to_sql('table', engine, schema='other_schema')  # noqa F821
+         pd.read_sql_table('table', engine, schema='other_schema')  # noqa F821
+
+- Added support for writing ``NaN`` values with ``to_sql`` (:issue:`2754`).
+- Added support for writing datetime64 columns with ``to_sql`` for all database flavors (:issue:`7103`).
+
+
+.. _whatsnew_0150.api:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0150.api_breaking:
+
+Breaking changes
+^^^^^^^^^^^^^^^^
+
+API changes related to ``Categorical`` (see :ref:`here <whatsnew_0150.cat>`
+for more details):
+
+- The ``Categorical`` constructor with two arguments changed from
+  "codes/labels and levels" to "values and levels (now called 'categories')".
+  This can lead to subtle bugs. If you use :class:`~pandas.Categorical` directly,
+  please audit your code by changing it to use the :meth:`~pandas.Categorical.from_codes`
+  constructor.
+
+  An old function call like (prior to 0.15.0):
+
+  .. code-block:: python
+
+    pd.Categorical([0,1,0,2,1], levels=['a', 'b', 'c'])
+
+  will have to adapted to the following to keep the same behaviour:
+
+  .. code-block:: ipython
+
+    In [2]: pd.Categorical.from_codes([0,1,0,2,1], categories=['a', 'b', 'c'])
+    Out[2]:
+    [a, b, a, c, b]
+    Categories (3, object): [a, b, c]
+
+API changes related to the introduction of the ``Timedelta`` scalar (see
+:ref:`above <whatsnew_0150.timedeltaindex>` for more details):
+
+- Prior to 0.15.0 :func:`to_timedelta` would return a ``Series`` for list-like/Series input,
+  and a ``np.timedelta64`` for scalar input. It will now return a ``TimedeltaIndex`` for
+  list-like input, ``Series`` for Series input, and ``Timedelta`` for scalar input.
+
+For API changes related to the rolling and expanding functions, see detailed overview :ref:`above <whatsnew_0150.roll>`.
+
+Other notable API changes:
+
+- Consistency when indexing with ``.loc`` and a list-like indexer when no values are found.
+
+  .. ipython:: python
+
+     df = pd.DataFrame([['a'], ['b']], index=[1, 2])
+     df
+
+  In prior versions there was a difference in these two constructs:
+
+  - ``df.loc[[3]]`` would return a frame reindexed by 3 (with all ``np.nan`` values)
+  - ``df.loc[[3],:]`` would raise ``KeyError``.
+
+  Both will now raise a ``KeyError``. The rule is that *at least 1* indexer must be found when using a list-like and ``.loc`` (:issue:`7999`)
+
+  Furthermore in prior versions these were also different:
+
+  - ``df.loc[[1,3]]`` would return a frame reindexed by [1,3]
+  - ``df.loc[[1,3],:]`` would raise ``KeyError``.
+
+  Both will now return a frame reindex by [1,3]. E.g.
+
+  .. code-block:: ipython
+
+     In [3]: df.loc[[1, 3]]
+     Out[3]:
+          0
+     1    a
+     3  NaN
+
+     In [4]: df.loc[[1, 3], :]
+     Out[4]:
+          0
+     1    a
+     3  NaN
+
+  This can also be seen in multi-axis indexing with a ``Panel``.
+
+  .. code-block:: python
+
+     >>> p = pd.Panel(np.arange(2 * 3 * 4).reshape(2, 3, 4),
+     ...              items=['ItemA', 'ItemB'],
+     ...              major_axis=[1, 2, 3],
+     ...              minor_axis=['A', 'B', 'C', 'D'])
+     >>> p
+     <class 'pandas.core.panel.Panel'>
+     Dimensions: 2 (items) x 3 (major_axis) x 4 (minor_axis)
+     Items axis: ItemA to ItemB
+     Major_axis axis: 1 to 3
+     Minor_axis axis: A to D
+
+
+  The following would raise ``KeyError`` prior to 0.15.0:
+
+  .. code-block:: ipython
+
+     In [5]:
+     Out[5]:
+        ItemA  ItemD
+     1      3    NaN
+     2      7    NaN
+     3     11    NaN
+
+  Furthermore, ``.loc`` will raise If no values are found in a MultiIndex with a list-like indexer:
+
+  .. ipython:: python
+     :okexcept:
+
+     s = pd.Series(np.arange(3, dtype='int64'),
+                   index=pd.MultiIndex.from_product([['A'],
+                                                    ['foo', 'bar', 'baz']],
+                                                    names=['one', 'two'])
+                   ).sort_index()
+     s
+     try:
+         s.loc[['D']]
+     except KeyError as e:
+         print("KeyError: " + str(e))
+
+- Assigning values to ``None`` now considers the dtype when choosing an 'empty' value (:issue:`7941`).
+
+  Previously, assigning to ``None`` in numeric containers changed the
+  dtype to object (or errored, depending on the call). It now uses
+  ``NaN``:
+
+  .. ipython:: python
+
+     s = pd.Series([1., 2., 3.])
+     s.loc[0] = None
+     s
+
+  ``NaT`` is now used similarly for datetime containers.
+
+  For object containers, we now preserve ``None`` values (previously these
+  were converted to ``NaN`` values).
+
+  .. ipython:: python
+
+     s = pd.Series(["a", "b", "c"])
+     s.loc[0] = None
+     s
+
+  To insert a ``NaN``, you must explicitly use ``np.nan``. See the :ref:`docs <missing.inserting>`.
+
+- In prior versions, updating a pandas object inplace would not reflect in other python references to this object. (:issue:`8511`, :issue:`5104`)
+
+  .. ipython:: python
+
+     s = pd.Series([1, 2, 3])
+     s2 = s
+     s += 1.5
+
+  Behavior prior to v0.15.0
+
+  .. code-block:: ipython
+
+
+     # the original object
+     In [5]: s
+     Out[5]:
+     0    2.5
+     1    3.5
+     2    4.5
+     dtype: float64
+
+
+     # a reference to the original object
+     In [7]: s2
+     Out[7]:
+     0    1
+     1    2
+     2    3
+     dtype: int64
+
+  This is now the correct behavior
+
+  .. ipython:: python
+
+     # the original object
+     s
+
+     # a reference to the original object
+     s2
+
+.. _whatsnew_0150.blanklines:
+
+- Made both the C-based and Python engines for ``read_csv`` and ``read_table`` ignore empty lines in input as well as
+  white space-filled lines, as long as ``sep`` is not white space. This is an API change
+  that can be controlled by the keyword parameter ``skip_blank_lines``.  See :ref:`the docs <io.skiplines>` (:issue:`4466`)
+
+- A timeseries/index localized to UTC when inserted into a Series/DataFrame will preserve the UTC timezone
+  and inserted as ``object`` dtype rather than being converted to a naive ``datetime64[ns]`` (:issue:`8411`).
+
+- Bug in passing a ``DatetimeIndex`` with a timezone that was not being retained in DataFrame construction from a dict (:issue:`7822`)
+
+  In prior versions this would drop the timezone, now it retains the timezone,
+  but gives a column of ``object`` dtype:
+
+  .. ipython:: python
+
+        i = pd.date_range('1/1/2011', periods=3, freq='10s', tz='US/Eastern')
+        i
+        df = pd.DataFrame({'a': i})
+        df
+        df.dtypes
+
+  Previously this would have yielded a column of ``datetime64`` dtype, but without timezone info.
+
+  The behaviour of assigning a column to an existing dataframe as ``df['a'] = i``
+  remains unchanged (this already returned an  ``object`` column with a timezone).
+
+- When passing multiple levels to :meth:`~pandas.DataFrame.stack`, it will now raise a ``ValueError`` when the
+  levels aren't all level names or all level numbers (:issue:`7660`). See
+  :ref:`Reshaping by stacking and unstacking <reshaping.stack_multiple>`.
+
+- Raise a ``ValueError`` in ``df.to_hdf`` with 'fixed' format, if ``df`` has non-unique columns as the resulting file will be broken (:issue:`7761`)
+
+- ``SettingWithCopy`` raise/warnings (according to the option ``mode.chained_assignment``) will now be issued when setting a value on a sliced mixed-dtype DataFrame using chained-assignment. (:issue:`7845`, :issue:`7950`)
+
+  .. code-block:: python
+
+     In [1]: df = pd.DataFrame(np.arange(0, 9), columns=['count'])
+
+     In [2]: df['group'] = 'b'
+
+     In [3]: df.iloc[0:5]['group'] = 'a'
+     /usr/local/bin/ipython:1: SettingWithCopyWarning:
+     A value is trying to be set on a copy of a slice from a DataFrame.
+     Try using .loc[row_indexer,col_indexer] = value instead
+
+     See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
+
+- ``merge``, ``DataFrame.merge``, and ``ordered_merge`` now return the same type
+  as the ``left`` argument (:issue:`7737`).
+
+- Previously an enlargement with a mixed-dtype frame would act unlike ``.append`` which will preserve dtypes (related :issue:`2578`, :issue:`8176`):
+
+  .. ipython:: python
+
+     df = pd.DataFrame([[True, 1], [False, 2]],
+                       columns=["female", "fitness"])
+     df
+     df.dtypes
+
+     # dtypes are now preserved
+     df.loc[2] = df.loc[1]
+     df
+     df.dtypes
+
+- ``Series.to_csv()`` now returns a string when ``path=None``, matching the behaviour of ``DataFrame.to_csv()`` (:issue:`8215`).
+
+- ``read_hdf`` now raises ``IOError`` when a file that doesn't exist is passed in. Previously, a new, empty file was created, and a ``KeyError`` raised (:issue:`7715`).
+
+- ``DataFrame.info()`` now ends its output with a newline character (:issue:`8114`)
+- Concatenating no objects will now raise a ``ValueError`` rather than a bare ``Exception``.
+- Merge errors will now be sub-classes of ``ValueError`` rather than raw ``Exception`` (:issue:`8501`)
+- ``DataFrame.plot`` and ``Series.plot`` keywords are now have consistent orders (:issue:`8037`)
+
+
+.. _whatsnew_0150.refactoring:
+
+Internal refactoring
+^^^^^^^^^^^^^^^^^^^^
+
+In 0.15.0 ``Index`` has internally been refactored to no longer sub-class ``ndarray``
+but instead subclass ``PandasObject``, similarly to the rest of the pandas objects. This
+change allows very easy sub-classing and creation of new index types. This should be
+a transparent change with only very limited API implications (:issue:`5080`, :issue:`7439`, :issue:`7796`, :issue:`8024`, :issue:`8367`, :issue:`7997`, :issue:`8522`):
+
+- you may need to unpickle pandas version < 0.15.0 pickles using ``pd.read_pickle`` rather than ``pickle.load``. See :ref:`pickle docs <io.pickle>`
+- when plotting with a ``PeriodIndex``, the matplotlib internal axes will now be arrays of ``Period`` rather than a ``PeriodIndex`` (this is similar to how a ``DatetimeIndex`` passes arrays of ``datetimes`` now)
+- MultiIndexes will now raise similarly to other pandas objects w.r.t. truth testing, see :ref:`here <gotchas.truth>` (:issue:`7897`).
+- When plotting a DatetimeIndex directly with matplotlib's ``plot`` function,
+  the axis labels will no longer be formatted as dates but as integers (the
+  internal representation of a ``datetime64``). **UPDATE** This is fixed
+  in 0.15.1, see :ref:`here <whatsnew_0151.datetime64_plotting>`.
+
+.. _whatsnew_0150.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- The attributes ``Categorical`` ``labels`` and ``levels`` attributes are
+  deprecated and renamed to ``codes`` and ``categories``.
+- The ``outtype`` argument to ``pd.DataFrame.to_dict`` has been deprecated in favor of ``orient``. (:issue:`7840`)
+- The ``convert_dummies`` method has been deprecated in favor of
+  ``get_dummies`` (:issue:`8140`)
+- The ``infer_dst`` argument in ``tz_localize`` will be deprecated in favor of
+  ``ambiguous`` to allow for more flexibility in dealing with DST transitions.
+  Replace ``infer_dst=True`` with ``ambiguous='infer'`` for the same behavior (:issue:`7943`).
+  See :ref:`the docs<timeseries.timezone_ambiguous>` for more details.
+- The top-level ``pd.value_range`` has been deprecated and can be replaced by ``.describe()`` (:issue:`8481`)
+
+.. _whatsnew_0150.index_set_ops:
+
+- The ``Index`` set operations ``+`` and ``-`` were deprecated in order to provide these for numeric type operations on certain index types. ``+`` can be replaced by ``.union()`` or ``|``, and ``-`` by ``.difference()``. Further the method name ``Index.diff()`` is deprecated and can be replaced by ``Index.difference()`` (:issue:`8226`)
+
+  .. code-block:: python
+
+     # +
+     pd.Index(['a', 'b', 'c']) + pd.Index(['b', 'c', 'd'])
+
+     # should be replaced by
+     pd.Index(['a', 'b', 'c']).union(pd.Index(['b', 'c', 'd']))
+
+  .. code-block:: python
+
+     # -
+     pd.Index(['a', 'b', 'c']) - pd.Index(['b', 'c', 'd'])
+
+     # should be replaced by
+     pd.Index(['a', 'b', 'c']).difference(pd.Index(['b', 'c', 'd']))
+
+- The ``infer_types`` argument to :func:`~pandas.read_html` now has no
+  effect and is deprecated (:issue:`7762`, :issue:`7032`).
+
+
+.. _whatsnew_0150.prior_deprecations:
+
+Removal of prior version deprecations/changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Remove ``DataFrame.delevel`` method in favor of ``DataFrame.reset_index``
+
+
+
+.. _whatsnew_0150.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+Enhancements in the importing/exporting of Stata files:
+
+- Added support for bool, uint8, uint16 and uint32 data types in ``to_stata`` (:issue:`7097`, :issue:`7365`)
+- Added conversion option when importing Stata files (:issue:`8527`)
+- ``DataFrame.to_stata`` and ``StataWriter`` check string length for
+  compatibility with limitations imposed in dta files where fixed-width
+  strings must contain 244 or fewer characters.  Attempting to write Stata
+  dta files with strings longer than 244 characters raises a ``ValueError``. (:issue:`7858`)
+- ``read_stata`` and ``StataReader`` can import missing data information into a
+  ``DataFrame`` by setting the argument ``convert_missing`` to ``True``. When
+  using this options, missing values are returned as ``StataMissingValue``
+  objects and columns containing missing values have ``object`` data type. (:issue:`8045`)
+
+Enhancements in the plotting functions:
+
+- Added ``layout`` keyword to ``DataFrame.plot``. You can pass a tuple of ``(rows, columns)``, one of which can be ``-1`` to automatically infer (:issue:`6667`, :issue:`8071`).
+- Allow to pass multiple axes to ``DataFrame.plot``, ``hist`` and ``boxplot`` (:issue:`5353`, :issue:`6970`, :issue:`7069`)
+- Added support for ``c``, ``colormap`` and ``colorbar`` arguments for ``DataFrame.plot`` with ``kind='scatter'`` (:issue:`7780`)
+- Histogram from ``DataFrame.plot`` with ``kind='hist'`` (:issue:`7809`), See :ref:`the docs<visualization.hist>`.
+- Boxplot from ``DataFrame.plot`` with ``kind='box'`` (:issue:`7998`), See :ref:`the docs<visualization.box>`.
+
+Other:
+
+- ``read_csv`` now has a keyword parameter ``float_precision`` which specifies which floating-point converter the C engine should use during parsing, see :ref:`here <io.float_precision>` (:issue:`8002`, :issue:`8044`)
+
+- Added ``searchsorted`` method to ``Series`` objects (:issue:`7447`)
+
+- :func:`describe` on mixed-types DataFrames is more flexible. Type-based column filtering is now possible via the ``include``/``exclude`` arguments.
+  See the :ref:`docs <basics.describe>` (:issue:`8164`).
+
+  .. ipython:: python
+
+    df = pd.DataFrame({'catA': ['foo', 'foo', 'bar'] * 8,
+                       'catB': ['a', 'b', 'c', 'd'] * 6,
+                       'numC': np.arange(24),
+                       'numD': np.arange(24.) + .5})
+    df.describe(include=["object"])
+    df.describe(include=["number", "object"], exclude=["float"])
+
+  Requesting all columns is possible with the shorthand 'all'
+
+  .. ipython:: python
+
+    df.describe(include='all')
+
+  Without those arguments, ``describe`` will behave as before, including only numerical columns or, if none are, only categorical columns. See also the :ref:`docs <basics.describe>`
+
+- Added ``split`` as an option to the ``orient`` argument in ``pd.DataFrame.to_dict``. (:issue:`7840`)
+
+- The ``get_dummies`` method can now be used on DataFrames. By default only
+  categorical columns are encoded as 0's and 1's, while other columns are
+  left untouched.
+
+  .. ipython:: python
+
+    df = pd.DataFrame({'A': ['a', 'b', 'a'], 'B': ['c', 'c', 'b'],
+                    'C': [1, 2, 3]})
+    pd.get_dummies(df)
+
+- ``PeriodIndex`` supports ``resolution`` as the same as ``DatetimeIndex`` (:issue:`7708`)
+- ``pandas.tseries.holiday`` has added support for additional holidays and ways to observe holidays (:issue:`7070`)
+- ``pandas.tseries.holiday.Holiday`` now supports a list of offsets in Python3 (:issue:`7070`)
+- ``pandas.tseries.holiday.Holiday`` now supports a days_of_week parameter (:issue:`7070`)
+- ``GroupBy.nth()`` now supports selecting multiple nth values (:issue:`7910`)
+
+  .. ipython:: python
+
+    business_dates = pd.date_range(start='4/1/2014', end='6/30/2014', freq='B')
+    df = pd.DataFrame(1, index=business_dates, columns=['a', 'b'])
+    # get the first, 4th, and last date index for each month
+    df.groupby([df.index.year, df.index.month]).nth([0, 3, -1])
+
+- ``Period`` and ``PeriodIndex`` supports addition/subtraction with ``timedelta``-likes (:issue:`7966`)
+
+  If ``Period`` freq is ``D``, ``H``, ``T``, ``S``, ``L``, ``U``, ``N``, ``Timedelta``-like can be added if the result can have same freq. Otherwise, only the same ``offsets`` can be added.
+
+  .. code-block:: ipython
+
+     In [104]: idx = pd.period_range('2014-07-01 09:00', periods=5, freq='H')
+
+     In [105]: idx
+     Out[105]:
+     PeriodIndex(['2014-07-01 09:00', '2014-07-01 10:00', '2014-07-01 11:00',
+                  '2014-07-01 12:00', '2014-07-01 13:00'],
+                 dtype='period[H]')
+
+     In [106]: idx + pd.offsets.Hour(2)
+     Out[106]:
+     PeriodIndex(['2014-07-01 11:00', '2014-07-01 12:00', '2014-07-01 13:00',
+                  '2014-07-01 14:00', '2014-07-01 15:00'],
+                 dtype='period[H]')
+
+     In [107]: idx + pd.Timedelta('120m')
+     Out[107]:
+     PeriodIndex(['2014-07-01 11:00', '2014-07-01 12:00', '2014-07-01 13:00',
+                  '2014-07-01 14:00', '2014-07-01 15:00'],
+                 dtype='period[H]')
+
+     In [108]: idx = pd.period_range('2014-07', periods=5, freq='M')
+
+     In [109]: idx
+     Out[109]: PeriodIndex(['2014-07', '2014-08', '2014-09', '2014-10', '2014-11'], dtype='period[M]')
+
+     In [110]: idx + pd.offsets.MonthEnd(3)
+     Out[110]: PeriodIndex(['2014-10', '2014-11', '2014-12', '2015-01', '2015-02'], dtype='period[M]')
+
+- Added experimental compatibility with ``openpyxl`` for versions >= 2.0. The ``DataFrame.to_excel``
+  method ``engine`` keyword now recognizes ``openpyxl1`` and ``openpyxl2``
+  which will explicitly require openpyxl v1 and v2 respectively, failing if
+  the requested version is not available. The ``openpyxl`` engine is a now a
+  meta-engine that automatically uses whichever version of openpyxl is
+  installed. (:issue:`7177`)
+
+- ``DataFrame.fillna`` can now accept a ``DataFrame`` as a fill value (:issue:`8377`)
+
+- Passing multiple levels to :meth:`~pandas.DataFrame.stack` will now work when multiple level
+  numbers are passed (:issue:`7660`). See
+  :ref:`Reshaping by stacking and unstacking <reshaping.stack_multiple>`.
+
+- :func:`set_names`, :func:`set_labels`, and :func:`set_levels` methods now take an optional ``level`` keyword argument to all modification of specific level(s) of a MultiIndex. Additionally :func:`set_names` now accepts a scalar string value when operating on an ``Index`` or on a specific level of a ``MultiIndex`` (:issue:`7792`)
+
+  .. ipython:: python
+
+      idx = pd.MultiIndex.from_product([['a'], range(3), list("pqr")],
+                                       names=['foo', 'bar', 'baz'])
+      idx.set_names('qux', level=0)
+      idx.set_names(['qux', 'corge'], level=[0, 1])
+      idx.set_levels(['a', 'b', 'c'], level='bar')
+      idx.set_levels([['a', 'b', 'c'], [1, 2, 3]], level=[1, 2])
+
+- ``Index.isin`` now supports a ``level`` argument to specify which index level
+  to use for membership tests (:issue:`7892`, :issue:`7890`)
+
+  .. code-block:: ipython
+
+     In [1]: idx = pd.MultiIndex.from_product([[0, 1], ['a', 'b', 'c']])
+
+     In [2]: idx.values
+     Out[2]: array([(0, 'a'), (0, 'b'), (0, 'c'), (1, 'a'), (1, 'b'), (1, 'c')], dtype=object)
+
+     In [3]: idx.isin(['a', 'c', 'e'], level=1)
+     Out[3]: array([ True, False,  True,  True, False,  True], dtype=bool)
+
+- ``Index`` now supports ``duplicated`` and ``drop_duplicates``. (:issue:`4060`)
+
+  .. ipython:: python
+
+     idx = pd.Index([1, 2, 3, 4, 1, 2])
+     idx
+     idx.duplicated()
+     idx.drop_duplicates()
+
+- add ``copy=True`` argument to ``pd.concat`` to enable pass through of complete blocks (:issue:`8252`)
+
+- Added support for numpy 1.8+ data types (``bool_``, ``int_``, ``float_``, ``string_``) for conversion to R dataframe  (:issue:`8400`)
+
+
+
+.. _whatsnew_0150.performance:
+
+Performance
+~~~~~~~~~~~
+
+- Performance improvements in ``DatetimeIndex.__iter__`` to allow faster iteration (:issue:`7683`)
+- Performance improvements in ``Period`` creation (and ``PeriodIndex`` setitem) (:issue:`5155`)
+- Improvements in Series.transform for significant performance gains (revised) (:issue:`6496`)
+- Performance improvements in ``StataReader`` when reading large files (:issue:`8040`, :issue:`8073`)
+- Performance improvements in ``StataWriter`` when writing large files (:issue:`8079`)
+- Performance and memory usage improvements in multi-key ``groupby`` (:issue:`8128`)
+- Performance improvements in groupby ``.agg`` and ``.apply`` where builtins max/min were not mapped to numpy/cythonized versions (:issue:`7722`)
+- Performance improvement in writing to sql (``to_sql``) of up to 50% (:issue:`8208`).
+- Performance benchmarking of groupby for large value of ngroups (:issue:`6787`)
+- Performance improvement in ``CustomBusinessDay``, ``CustomBusinessMonth`` (:issue:`8236`)
+- Performance improvement for ``MultiIndex.values`` for multi-level indexes containing datetimes (:issue:`8543`)
+
+
+
+.. _whatsnew_0150.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in pivot_table, when using margins and a dict aggfunc (:issue:`8349`)
+- Bug in ``read_csv`` where ``squeeze=True`` would return a view (:issue:`8217`)
+- Bug in checking of table name in ``read_sql`` in certain cases (:issue:`7826`).
+- Bug in ``DataFrame.groupby`` where ``Grouper`` does not recognize level when frequency is specified (:issue:`7885`)
+- Bug in multiindexes dtypes getting mixed up when DataFrame is saved to SQL table (:issue:`8021`)
+- Bug in ``Series`` 0-division with a float and integer operand dtypes  (:issue:`7785`)
+- Bug in ``Series.astype("unicode")`` not calling ``unicode`` on the values correctly (:issue:`7758`)
+- Bug in ``DataFrame.as_matrix()`` with mixed ``datetime64[ns]`` and ``timedelta64[ns]`` dtypes (:issue:`7778`)
+- Bug in ``HDFStore.select_column()`` not preserving UTC timezone info when selecting a ``DatetimeIndex`` (:issue:`7777`)
+- Bug in ``to_datetime`` when ``format='%Y%m%d'`` and ``coerce=True`` are specified, where previously an object array was returned (rather than
+  a coerced time-series with ``NaT``), (:issue:`7930`)
+- Bug in ``DatetimeIndex`` and ``PeriodIndex`` in-place addition and subtraction cause different result from normal one (:issue:`6527`)
+- Bug in adding and subtracting ``PeriodIndex`` with ``PeriodIndex`` raise ``TypeError`` (:issue:`7741`)
+- Bug in ``combine_first`` with ``PeriodIndex`` data raises ``TypeError`` (:issue:`3367`)
+- Bug in MultiIndex slicing with missing indexers (:issue:`7866`)
+- Bug in MultiIndex slicing with various edge cases (:issue:`8132`)
+- Regression in MultiIndex indexing with a non-scalar type object (:issue:`7914`)
+- Bug in ``Timestamp`` comparisons with ``==`` and ``int64`` dtype (:issue:`8058`)
+- Bug in pickles contains ``DateOffset`` may raise ``AttributeError`` when ``normalize`` attribute is referred internally (:issue:`7748`)
+- Bug in ``Panel`` when using ``major_xs`` and ``copy=False`` is passed (deprecation warning fails because of missing ``warnings``) (:issue:`8152`).
+- Bug in pickle deserialization that failed for pre-0.14.1 containers with dup items trying to avoid ambiguity
+  when matching block and manager items, when there's only one block there's no ambiguity (:issue:`7794`)
+- Bug in putting a ``PeriodIndex`` into a ``Series`` would convert to ``int64`` dtype, rather than ``object`` of ``Periods`` (:issue:`7932`)
+- Bug in ``HDFStore`` iteration when passing a where (:issue:`8014`)
+- Bug in ``DataFrameGroupby.transform`` when transforming with a passed non-sorted key (:issue:`8046`, :issue:`8430`)
+- Bug in repeated timeseries line and area plot may result in ``ValueError`` or incorrect kind (:issue:`7733`)
+- Bug in inference in a ``MultiIndex`` with ``datetime.date`` inputs (:issue:`7888`)
+- Bug in ``get`` where an ``IndexError`` would not cause the default value to be returned (:issue:`7725`)
+- Bug in ``offsets.apply``, ``rollforward`` and ``rollback`` may reset nanosecond (:issue:`7697`)
+- Bug in ``offsets.apply``, ``rollforward`` and ``rollback`` may raise ``AttributeError`` if ``Timestamp`` has ``dateutil`` tzinfo (:issue:`7697`)
+- Bug in sorting a MultiIndex frame with a ``Float64Index`` (:issue:`8017`)
+- Bug in inconsistent panel setitem with a rhs of a ``DataFrame`` for alignment (:issue:`7763`)
+- Bug in ``is_superperiod`` and ``is_subperiod`` cannot handle higher frequencies than ``S`` (:issue:`7760`, :issue:`7772`, :issue:`7803`)
+- Bug in 32-bit platforms with ``Series.shift`` (:issue:`8129`)
+- Bug in ``PeriodIndex.unique`` returns int64 ``np.ndarray`` (:issue:`7540`)
+- Bug in ``groupby.apply`` with a non-affecting mutation in the function (:issue:`8467`)
+- Bug in ``DataFrame.reset_index`` which has ``MultiIndex`` contains ``PeriodIndex`` or ``DatetimeIndex`` with tz raises ``ValueError`` (:issue:`7746`, :issue:`7793`)
+- Bug in ``DataFrame.plot`` with ``subplots=True`` may draw unnecessary minor xticks and yticks (:issue:`7801`)
+- Bug in ``StataReader`` which did not read variable labels in 117 files due to difference between Stata documentation and implementation (:issue:`7816`)
+- Bug in ``StataReader`` where strings were always converted to 244 characters-fixed width irrespective of underlying string size (:issue:`7858`)
+- Bug in ``DataFrame.plot`` and ``Series.plot`` may ignore ``rot`` and ``fontsize`` keywords (:issue:`7844`)
+- Bug in ``DatetimeIndex.value_counts`` doesn't preserve tz  (:issue:`7735`)
+- Bug in ``PeriodIndex.value_counts`` results in ``Int64Index`` (:issue:`7735`)
+- Bug in ``DataFrame.join`` when doing left join on index and there are multiple matches (:issue:`5391`)
+- Bug in ``GroupBy.transform()`` where int groups with a transform that
+  didn't preserve the index were incorrectly truncated (:issue:`7972`).
+- Bug in ``groupby`` where callable objects without name attributes would take the wrong path,
+  and produce a ``DataFrame`` instead of a ``Series`` (:issue:`7929`)
+- Bug in ``groupby`` error message when a DataFrame grouping column is duplicated (:issue:`7511`)
+- Bug in ``read_html`` where the ``infer_types`` argument forced coercion of
+  date-likes incorrectly (:issue:`7762`, :issue:`7032`).
+- Bug in ``Series.str.cat`` with an index which was filtered as to not include the first item (:issue:`7857`)
+- Bug in ``Timestamp`` cannot parse ``nanosecond`` from string (:issue:`7878`)
+- Bug in ``Timestamp`` with string offset and ``tz`` results incorrect (:issue:`7833`)
+- Bug in ``tslib.tz_convert`` and ``tslib.tz_convert_single`` may return different results (:issue:`7798`)
+- Bug in ``DatetimeIndex.intersection`` of non-overlapping timestamps with tz raises ``IndexError`` (:issue:`7880`)
+- Bug in alignment with TimeOps and non-unique indexes (:issue:`8363`)
+- Bug in ``GroupBy.filter()`` where fast path vs. slow path made the filter
+  return a non scalar value that appeared valid but wasn't (:issue:`7870`).
+- Bug in ``date_range()``/``DatetimeIndex()`` when the timezone was inferred from input dates yet incorrect
+  times were returned when crossing DST boundaries (:issue:`7835`, :issue:`7901`).
+- Bug in ``to_excel()`` where a negative sign was being prepended to positive infinity and was absent for negative infinity (:issue:`7949`)
+- Bug in area plot draws legend with incorrect ``alpha`` when ``stacked=True`` (:issue:`8027`)
+- ``Period`` and ``PeriodIndex`` addition/subtraction with ``np.timedelta64`` results in incorrect internal representations (:issue:`7740`)
+- Bug in ``Holiday`` with no offset or observance (:issue:`7987`)
+- Bug in ``DataFrame.to_latex`` formatting when columns or index is a ``MultiIndex`` (:issue:`7982`).
+- Bug in ``DateOffset`` around Daylight Savings Time produces unexpected results (:issue:`5175`).
+- Bug in ``DataFrame.shift`` where empty columns would throw ``ZeroDivisionError`` on numpy 1.7 (:issue:`8019`)
+- Bug in installation where ``html_encoding/*.html`` wasn't installed and
+  therefore some tests were not running correctly (:issue:`7927`).
+- Bug in ``read_html`` where ``bytes`` objects were not tested for in
+  ``_read`` (:issue:`7927`).
+- Bug in ``DataFrame.stack()`` when one of the column levels was a datelike (:issue:`8039`)
+- Bug in broadcasting numpy scalars with ``DataFrame`` (:issue:`8116`)
+- Bug in ``pivot_table`` performed with nameless ``index`` and ``columns`` raises ``KeyError`` (:issue:`8103`)
+- Bug in ``DataFrame.plot(kind='scatter')`` draws points and errorbars with different colors when the color is specified by ``c`` keyword (:issue:`8081`)
+- Bug in ``Float64Index`` where ``iat`` and ``at`` were not testing and were
+  failing (:issue:`8092`).
+- Bug in ``DataFrame.boxplot()`` where y-limits were not set correctly when
+  producing multiple axes (:issue:`7528`, :issue:`5517`).
+- Bug in ``read_csv`` where line comments were not handled correctly given
+  a custom line terminator or ``delim_whitespace=True`` (:issue:`8122`).
+- Bug in ``read_html`` where empty tables caused a ``StopIteration`` (:issue:`7575`)
+- Bug in casting when setting a column in a same-dtype block (:issue:`7704`)
+- Bug in accessing groups from a ``GroupBy`` when the original grouper
+  was a tuple (:issue:`8121`).
+- Bug in ``.at`` that would accept integer indexers on a non-integer index and do fallback (:issue:`7814`)
+- Bug with kde plot and NaNs (:issue:`8182`)
+- Bug in ``GroupBy.count`` with float32 data type were nan values were not excluded (:issue:`8169`).
+- Bug with stacked barplots and NaNs (:issue:`8175`).
+- Bug in resample with non evenly divisible offsets (e.g. '7s') (:issue:`8371`)
+- Bug in interpolation methods with the ``limit`` keyword when no values needed interpolating (:issue:`7173`).
+- Bug where ``col_space`` was ignored in ``DataFrame.to_string()`` when ``header=False`` (:issue:`8230`).
+- Bug with ``DatetimeIndex.asof`` incorrectly matching partial strings and returning the wrong date (:issue:`8245`).
+- Bug in plotting methods modifying the global matplotlib rcParams (:issue:`8242`).
+- Bug in ``DataFrame.__setitem__`` that caused errors when setting a dataframe column to a sparse array (:issue:`8131`)
+- Bug where ``Dataframe.boxplot()`` failed when entire column was empty (:issue:`8181`).
+- Bug with messed variables in ``radviz`` visualization (:issue:`8199`).
+- Bug in interpolation methods with the ``limit`` keyword when no values needed interpolating (:issue:`7173`).
+- Bug where ``col_space`` was ignored in ``DataFrame.to_string()`` when ``header=False`` (:issue:`8230`).
+- Bug in ``to_clipboard`` that would clip long column data (:issue:`8305`)
+- Bug in ``DataFrame`` terminal display: Setting max_column/max_rows to zero did not trigger auto-resizing of dfs to fit terminal width/height (:issue:`7180`).
+- Bug in OLS where running with "cluster" and "nw_lags" parameters did not work correctly, but also did not throw an error
+  (:issue:`5884`).
+- Bug in ``DataFrame.dropna`` that interpreted non-existent columns in the subset argument as the 'last column' (:issue:`8303`)
+- Bug in ``Index.intersection`` on non-monotonic non-unique indexes (:issue:`8362`).
+- Bug in masked series assignment where mismatching types would break alignment (:issue:`8387`)
+- Bug in ``NDFrame.equals`` gives false negatives with dtype=object (:issue:`8437`)
+- Bug in assignment with indexer where type diversity would break alignment (:issue:`8258`)
+- Bug in ``NDFrame.loc`` indexing when row/column names were lost when target was a list/ndarray (:issue:`6552`)
+- Regression in ``NDFrame.loc`` indexing when rows/columns were converted to Float64Index if target was an empty list/ndarray (:issue:`7774`)
+- Bug in ``Series`` that allows it to be indexed by a ``DataFrame`` which has unexpected results.  Such indexing is no longer permitted (:issue:`8444`)
+- Bug in item assignment of a ``DataFrame`` with MultiIndex columns where right-hand-side columns were not aligned (:issue:`7655`)
+- Suppress FutureWarning generated by NumPy when comparing object arrays containing NaN for equality (:issue:`7065`)
+- Bug in ``DataFrame.eval()`` where the dtype of the ``not`` operator (``~``)
+  was not correctly inferred as ``bool``.
+
+
+.. _whatsnew_0.15.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.14.1..v0.15.0
diff --git a/pandas/doc/source/whatsnew/v0.15.1.rst b/pandas/doc/source/whatsnew/v0.15.1.rst
new file mode 100644
index 0000000000000..f16c9b3f5d45b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.15.1.rst
@@ -0,0 +1,324 @@
+.. _whatsnew_0151:
+
+Version 0.15.1 (November 9, 2014)
+---------------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release from 0.15.0 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+- :ref:`Enhancements <whatsnew_0151.enhancements>`
+- :ref:`API Changes <whatsnew_0151.api>`
+- :ref:`Bug Fixes <whatsnew_0151.bug_fixes>`
+
+.. _whatsnew_0151.api:
+
+API changes
+~~~~~~~~~~~
+
+- ``s.dt.hour`` and other ``.dt`` accessors will now return ``np.nan`` for missing values (rather than previously -1), (:issue:`8689`)
+
+  .. ipython:: python
+
+     s = pd.Series(pd.date_range("20130101", periods=5, freq="D"))
+     s.iloc[2] = np.nan
+     s
+
+  previous behavior:
+
+  .. code-block:: ipython
+
+     In [6]: s.dt.hour
+     Out[6]:
+     0    0
+     1    0
+     2   -1
+     3    0
+     4    0
+     dtype: int64
+
+  current behavior:
+
+  .. ipython:: python
+
+     s.dt.hour
+
+- ``groupby`` with ``as_index=False`` will not add erroneous extra columns to
+  result (:issue:`8582`):
+
+  .. ipython:: python
+
+     np.random.seed(2718281)
+     df = pd.DataFrame(np.random.randint(0, 100, (10, 2)), columns=["jim", "joe"])
+     df.head()
+
+     ts = pd.Series(5 * np.random.randint(0, 3, 10))
+
+  previous behavior:
+
+  .. code-block:: ipython
+
+     In [4]: df.groupby(ts, as_index=False).max()
+     Out[4]:
+        NaN  jim  joe
+     0    0   72   83
+     1    5   77   84
+     2   10   96   65
+
+  current behavior:
+
+  .. code-block:: ipython
+
+     In [4]: df.groupby(ts, as_index=False).max()
+     Out[4]:
+        jim  joe
+     0   72   83
+     1   77   84
+     2   96   65
+
+- ``groupby`` will not erroneously exclude columns if the column name conflicts
+  with the grouper name (:issue:`8112`):
+
+  .. ipython:: python
+
+     df = pd.DataFrame({"jim": range(5), "joe": range(5, 10)})
+     df
+     gr = df.groupby(df["jim"] < 2)
+
+  previous behavior (excludes 1st column from output):
+
+  .. code-block:: ipython
+
+     In [4]: gr.apply("sum")
+     Out[4]:
+            joe
+     jim
+     False   24
+     True    11
+
+  current behavior:
+
+  .. ipython:: python
+
+     gr.apply("sum")
+
+- Support for slicing with monotonic decreasing indexes, even if ``start`` or ``stop`` is
+  not found in the index (:issue:`7860`):
+
+  .. ipython:: python
+
+    s = pd.Series(["a", "b", "c", "d"], [4, 3, 2, 1])
+    s
+
+  previous behavior:
+
+  .. code-block:: ipython
+
+     In [8]: s.loc[3.5:1.5]
+     KeyError: 3.5
+
+  current behavior:
+
+  .. ipython:: python
+
+     s.loc[3.5:1.5]
+
+- ``io.data.Options`` has been fixed for a change in the format of the Yahoo Options page (:issue:`8612`), (:issue:`8741`)
+
+  .. note::
+
+    As a result of a change in Yahoo's option page layout, when an expiry date is given,
+    ``Options`` methods now return data for a single expiry date.  Previously, methods returned all
+    data for the selected month.
+
+  The ``month`` and ``year`` parameters have been undeprecated and can be used to get all
+  options data for a given month.
+
+  If an expiry date that is not valid is given, data for the next expiry after the given
+  date is returned.
+
+  Option data frames are now saved on the instance as ``callsYYMMDD`` or ``putsYYMMDD``.  Previously
+  they were saved as ``callsMMYY`` and ``putsMMYY``.  The next expiry is saved as ``calls`` and ``puts``.
+
+  New features:
+
+  - The expiry parameter can now be a single date or a list-like object containing dates.
+
+  - A new property ``expiry_dates`` was added, which returns all available expiry dates.
+
+  Current behavior:
+
+  .. code-block:: ipython
+
+      In [17]: from pandas.io.data import Options
+
+      In [18]: aapl = Options('aapl', 'yahoo')
+
+      In [19]: aapl.get_call_data().iloc[0:5, 0:1]
+      Out[19]:
+                                                   Last
+      Strike Expiry     Type Symbol
+      80     2014-11-14 call AAPL141114C00080000  29.05
+      84     2014-11-14 call AAPL141114C00084000  24.80
+      85     2014-11-14 call AAPL141114C00085000  24.05
+      86     2014-11-14 call AAPL141114C00086000  22.76
+      87     2014-11-14 call AAPL141114C00087000  21.74
+
+      In [20]: aapl.expiry_dates
+      Out[20]:
+      [datetime.date(2014, 11, 14),
+       datetime.date(2014, 11, 22),
+       datetime.date(2014, 11, 28),
+       datetime.date(2014, 12, 5),
+       datetime.date(2014, 12, 12),
+       datetime.date(2014, 12, 20),
+       datetime.date(2015, 1, 17),
+       datetime.date(2015, 2, 20),
+       datetime.date(2015, 4, 17),
+       datetime.date(2015, 7, 17),
+       datetime.date(2016, 1, 15),
+       datetime.date(2017, 1, 20)]
+
+      In [21]: aapl.get_near_stock_price(expiry=aapl.expiry_dates[0:3]).iloc[0:5, 0:1]
+      Out[21]:
+                                                  Last
+      Strike Expiry     Type Symbol
+      109    2014-11-22 call AAPL141122C00109000  1.48
+             2014-11-28 call AAPL141128C00109000  1.79
+      110    2014-11-14 call AAPL141114C00110000  0.55
+             2014-11-22 call AAPL141122C00110000  1.02
+             2014-11-28 call AAPL141128C00110000  1.32
+
+.. _whatsnew_0151.datetime64_plotting:
+
+- pandas now also registers the ``datetime64`` dtype in matplotlib's units registry
+  to plot such values as datetimes. This is activated once pandas is imported. In
+  previous versions, plotting an array of ``datetime64`` values will have resulted
+  in plotted integer values. To keep the previous behaviour, you can do
+  ``del matplotlib.units.registry[np.datetime64]`` (:issue:`8614`).
+
+
+.. _whatsnew_0151.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+- ``concat`` permits a wider variety of iterables of pandas objects to be
+  passed as the first parameter (:issue:`8645`):
+
+  .. ipython:: python
+
+     from collections import deque
+
+     df1 = pd.DataFrame([1, 2, 3])
+     df2 = pd.DataFrame([4, 5, 6])
+
+  previous behavior:
+
+  .. code-block:: ipython
+
+     In [7]: pd.concat(deque((df1, df2)))
+     TypeError: first argument must be a list-like of pandas objects, you passed an object of type "deque"
+
+  current behavior:
+
+  .. ipython:: python
+
+     pd.concat(deque((df1, df2)))
+
+- Represent ``MultiIndex`` labels with a dtype that utilizes memory based on the level size. In prior versions, the memory usage was a constant 8 bytes per element in each level. In addition, in prior versions, the *reported* memory usage was incorrect as it didn't show the usage for the memory occupied by the underling data array. (:issue:`8456`)
+
+  .. ipython:: python
+
+     dfi = pd.DataFrame(
+         1, index=pd.MultiIndex.from_product([["a"], range(1000)]), columns=["A"]
+     )
+
+  previous behavior:
+
+  .. code-block:: ipython
+
+     # this was underreported in prior versions
+     In [1]: dfi.memory_usage(index=True)
+     Out[1]:
+     Index    8000 # took about 24008 bytes in < 0.15.1
+     A        8000
+     dtype: int64
+
+
+  current behavior:
+
+  .. ipython:: python
+
+     dfi.memory_usage(index=True)
+
+- Added Index properties ``is_monotonic_increasing`` and ``is_monotonic_decreasing`` (:issue:`8680`).
+
+- Added option to select columns when importing Stata files (:issue:`7935`)
+
+- Qualify memory usage in ``DataFrame.info()`` by adding ``+`` if it is a lower bound (:issue:`8578`)
+
+- Raise errors in certain aggregation cases where an argument such as ``numeric_only`` is not handled (:issue:`8592`).
+
+- Added support for 3-character ISO and non-standard country codes in :func:`io.wb.download` (:issue:`8482`)
+
+- World Bank data requests now will warn/raise based
+  on an ``errors`` argument, as well as a list of hard-coded country codes and
+  the World Bank's JSON response.  In prior versions, the error messages
+  didn't look at the World Bank's JSON response.  Problem-inducing input were
+  simply dropped prior to the request. The issue was that many good countries
+  were cropped in the hard-coded approach.  All countries will work now, but
+  some bad countries will raise exceptions because some edge cases break the
+  entire response. (:issue:`8482`)
+
+- Added option to ``Series.str.split()`` to return a ``DataFrame`` rather than a ``Series`` (:issue:`8428`)
+
+- Added option to ``df.info(null_counts=None|True|False)`` to override the default display options and force showing of the null-counts (:issue:`8701`)
+
+
+.. _whatsnew_0151.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in unpickling of a ``CustomBusinessDay`` object (:issue:`8591`)
+- Bug in coercing ``Categorical`` to a records array, e.g. ``df.to_records()`` (:issue:`8626`)
+- Bug in ``Categorical`` not created properly with ``Series.to_frame()`` (:issue:`8626`)
+- Bug in coercing in astype of a ``Categorical`` of a passed ``pd.Categorical`` (this now raises ``TypeError`` correctly), (:issue:`8626`)
+- Bug in ``cut``/``qcut`` when using ``Series`` and ``retbins=True`` (:issue:`8589`)
+- Bug in writing Categorical columns to an SQL database with ``to_sql`` (:issue:`8624`).
+- Bug in comparing ``Categorical`` of datetime raising when being compared to a scalar datetime (:issue:`8687`)
+- Bug in selecting from a ``Categorical`` with ``.iloc`` (:issue:`8623`)
+- Bug in groupby-transform with a Categorical (:issue:`8623`)
+- Bug in duplicated/drop_duplicates with a Categorical (:issue:`8623`)
+- Bug in ``Categorical`` reflected comparison operator raising if the first argument was a numpy array scalar (e.g. np.int64) (:issue:`8658`)
+- Bug in Panel indexing with a list-like (:issue:`8710`)
+- Compat issue is ``DataFrame.dtypes`` when ``options.mode.use_inf_as_null`` is True (:issue:`8722`)
+- Bug in ``read_csv``, ``dialect`` parameter would not take a string (:issue:`8703`)
+- Bug in slicing a MultiIndex level with an empty-list (:issue:`8737`)
+- Bug in numeric index operations of add/sub with Float/Index Index with numpy arrays (:issue:`8608`)
+- Bug in setitem with empty indexer and unwanted coercion of dtypes (:issue:`8669`)
+- Bug in ix/loc block splitting on setitem (manifests with integer-like dtypes, e.g. datetime64) (:issue:`8607`)
+- Bug when doing label based indexing with integers not found in the index for
+  non-unique but monotonic indexes (:issue:`8680`).
+- Bug when indexing a Float64Index with ``np.nan`` on numpy 1.7 (:issue:`8980`).
+- Fix ``shape`` attribute for ``MultiIndex`` (:issue:`8609`)
+- Bug in ``GroupBy`` where a name conflict between the grouper and columns
+  would break ``groupby`` operations (:issue:`7115`, :issue:`8112`)
+- Fixed a bug where plotting a column ``y`` and specifying a label would mutate the index name of the original DataFrame (:issue:`8494`)
+- Fix regression in plotting of a DatetimeIndex directly with matplotlib (:issue:`8614`).
+- Bug in ``date_range`` where partially-specified dates would incorporate current date (:issue:`6961`)
+- Bug in Setting by indexer to a scalar value with a mixed-dtype ``Panel4d`` was failing (:issue:`8702`)
+- Bug where ``DataReader``'s would fail if one of the symbols passed was invalid.  Now returns data for valid symbols and np.nan for invalid (:issue:`8494`)
+- Bug in ``get_quote_yahoo`` that wouldn't allow non-float return values (:issue:`5229`).
+
+
+.. _whatsnew_0.15.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.15.0..v0.15.1
diff --git a/pandas/doc/source/whatsnew/v0.15.2.rst b/pandas/doc/source/whatsnew/v0.15.2.rst
new file mode 100644
index 0000000000000..a9003710540d7
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.15.2.rst
@@ -0,0 +1,297 @@
+.. _whatsnew_0152:
+
+Version 0.15.2 (December 12, 2014)
+----------------------------------
+
+{{ header }}
+
+
+This is a minor release from 0.15.1 and includes a large number of bug fixes
+along with several new features, enhancements, and performance improvements.
+A small number of API changes were necessary to fix existing bugs.
+We recommend that all users upgrade to this version.
+
+- :ref:`Enhancements <whatsnew_0152.enhancements>`
+- :ref:`API Changes <whatsnew_0152.api>`
+- :ref:`Performance Improvements <whatsnew_0152.performance>`
+- :ref:`Bug Fixes <whatsnew_0152.bug_fixes>`
+
+.. _whatsnew_0152.api:
+
+API changes
+~~~~~~~~~~~
+
+- Indexing in ``MultiIndex`` beyond lex-sort depth is now supported, though
+  a lexically sorted index will have a better performance. (:issue:`2646`)
+
+  .. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({'jim':[0, 0, 1, 1],
+       ...:                    'joe':['x', 'x', 'z', 'y'],
+       ...:                    'jolie':np.random.rand(4)}).set_index(['jim', 'joe'])
+       ...:
+
+    In [2]: df
+    Out[2]:
+                jolie
+    jim joe
+    0   x    0.126970
+        x    0.966718
+    1   z    0.260476
+        y    0.897237
+
+    [4 rows x 1 columns]
+
+    In [3]: df.index.lexsort_depth
+    Out[3]: 1
+
+    # in prior versions this would raise a KeyError
+    # will now show a PerformanceWarning
+    In [4]: df.loc[(1, 'z')]
+    Out[4]:
+                jolie
+    jim joe
+    1   z    0.260476
+
+    [1 rows x 1 columns]
+
+    # lexically sorting
+    In [5]: df2 = df.sort_index()
+
+    In [6]: df2
+    Out[6]:
+                jolie
+    jim joe
+    0   x    0.126970
+        x    0.966718
+    1   y    0.897237
+        z    0.260476
+
+    [4 rows x 1 columns]
+
+    In [7]: df2.index.lexsort_depth
+    Out[7]: 2
+
+    In [8]: df2.loc[(1,'z')]
+    Out[8]:
+                jolie
+    jim joe
+    1   z    0.260476
+
+    [1 rows x 1 columns]
+
+- Bug in unique of Series with ``category`` dtype, which returned all categories regardless
+  whether they were "used" or not (see :issue:`8559` for the discussion).
+  Previous behaviour was to return all categories:
+
+  .. code-block:: ipython
+
+    In [3]: cat = pd.Categorical(['a', 'b', 'a'], categories=['a', 'b', 'c'])
+
+    In [4]: cat
+    Out[4]:
+    [a, b, a]
+    Categories (3, object): [a < b < c]
+
+    In [5]: cat.unique()
+    Out[5]: array(['a', 'b', 'c'], dtype=object)
+
+  Now, only the categories that do effectively occur in the array are returned:
+
+  .. ipython:: python
+
+    cat = pd.Categorical(['a', 'b', 'a'], categories=['a', 'b', 'c'])
+    cat.unique()
+
+- ``Series.all`` and ``Series.any`` now support the ``level`` and ``skipna`` parameters. ``Series.all``, ``Series.any``, ``Index.all``, and ``Index.any`` no longer support the ``out`` and ``keepdims`` parameters, which existed for compatibility with ndarray. Various index types no longer support the ``all`` and ``any`` aggregation functions and will now raise ``TypeError``. (:issue:`8302`).
+
+- Allow equality comparisons of Series with a categorical dtype and object dtype; previously these would raise ``TypeError`` (:issue:`8938`)
+
+- Bug in ``NDFrame``: conflicting attribute/column names now behave consistently between getting and setting. Previously, when both a column and attribute named ``y`` existed, ``data.y`` would return the attribute, while ``data.y = z`` would update the column (:issue:`8994`)
+
+  .. ipython:: python
+
+     data = pd.DataFrame({'x': [1, 2, 3]})
+     data.y = 2
+     data['y'] = [2, 4, 6]
+     data
+
+     # this assignment was inconsistent
+     data.y = 5
+
+  Old behavior:
+
+  .. code-block:: ipython
+
+     In [6]: data.y
+     Out[6]: 2
+
+     In [7]: data['y'].values
+     Out[7]: array([5, 5, 5])
+
+  New behavior:
+
+  .. ipython:: python
+
+     data.y
+     data['y'].values
+
+- ``Timestamp('now')`` is now equivalent to ``Timestamp.now()`` in that it returns the local time rather than UTC. Also, ``Timestamp('today')`` is now equivalent to ``Timestamp.today()`` and both have ``tz`` as a possible argument. (:issue:`9000`)
+
+- Fix negative step support for label-based slices (:issue:`8753`)
+
+  Old behavior:
+
+  .. code-block:: ipython
+
+     In [1]: s = pd.Series(np.arange(3), ['a', 'b', 'c'])
+     Out[1]:
+     a    0
+     b    1
+     c    2
+     dtype: int64
+
+     In [2]: s.loc['c':'a':-1]
+     Out[2]:
+     c    2
+     dtype: int64
+
+  New behavior:
+
+  .. ipython:: python
+
+     s = pd.Series(np.arange(3), ['a', 'b', 'c'])
+     s.loc['c':'a':-1]
+
+
+.. _whatsnew_0152.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+``Categorical`` enhancements:
+
+- Added ability to export Categorical data to Stata (:issue:`8633`).  See :ref:`here <io.stata-categorical>` for limitations of categorical variables exported to Stata data files.
+- Added flag ``order_categoricals`` to ``StataReader`` and ``read_stata`` to select whether to order imported categorical data (:issue:`8836`).  See :ref:`here <io.stata-categorical>` for more information on importing categorical variables from Stata data files.
+- Added ability to export Categorical data to/from HDF5 (:issue:`7621`). Queries work the same as if it was an object array. However, the ``category`` dtyped data is stored in a more efficient manner. See :ref:`here <io.hdf5-categorical>` for an example and caveats w.r.t. prior versions of pandas.
+- Added support for ``searchsorted()`` on ``Categorical`` class (:issue:`8420`).
+
+Other enhancements:
+
+- Added the ability to specify the SQL type of columns when writing a DataFrame
+  to a database (:issue:`8778`).
+  For example, specifying to use the sqlalchemy ``String`` type instead of the
+  default ``Text`` type for string columns:
+
+  .. code-block:: python
+
+     from sqlalchemy.types import String
+     data.to_sql('data_dtype', engine, dtype={'Col_1': String})  # noqa F821
+
+- ``Series.all`` and ``Series.any`` now support the ``level`` and ``skipna`` parameters (:issue:`8302`):
+
+  .. code-block:: python
+
+     >>> s = pd.Series([False, True, False], index=[0, 0, 1])
+     >>> s.any(level=0)
+     0     True
+     1    False
+     dtype: bool
+
+- ``Panel`` now supports the ``all`` and ``any`` aggregation functions. (:issue:`8302`):
+
+  .. code-block:: python
+
+     >>> p = pd.Panel(np.random.rand(2, 5, 4) > 0.1)
+     >>> p.all()
+            0      1      2     3
+     0   True   True   True  True
+     1   True  False   True  True
+     2   True   True   True  True
+     3  False   True  False  True
+     4   True   True   True  True
+
+- Added support for ``utcfromtimestamp()``, ``fromtimestamp()``, and ``combine()`` on ``Timestamp`` class (:issue:`5351`).
+- Added Google Analytics (``pandas.io.ga``) basic documentation (:issue:`8835`). See `here <https://pandas.pydata.org/pandas-docs/version/0.15.2/remote_data.html#remote-data-ga>`__.
+- ``Timedelta`` arithmetic returns ``NotImplemented`` in unknown cases, allowing extensions by custom classes (:issue:`8813`).
+- ``Timedelta`` now supports arithmetic with ``numpy.ndarray`` objects of the appropriate dtype (numpy 1.8 or newer only) (:issue:`8884`).
+- Added ``Timedelta.to_timedelta64()`` method to the public API (:issue:`8884`).
+- Added ``gbq.generate_bq_schema()`` function to the gbq module (:issue:`8325`).
+- ``Series`` now works with map objects the same way as generators (:issue:`8909`).
+- Added context manager to ``HDFStore`` for automatic closing (:issue:`8791`).
+- ``to_datetime`` gains an ``exact`` keyword to allow for a format to not require an exact match for a provided format string (if its ``False``). ``exact`` defaults to ``True`` (meaning that exact matching is still the default)  (:issue:`8904`)
+- Added ``axvlines`` boolean option to parallel_coordinates plot function, determines whether vertical lines will be printed, default is True
+- Added ability to read table footers to read_html (:issue:`8552`)
+- ``to_sql`` now infers data types of non-NA values for columns that contain NA values and have dtype ``object`` (:issue:`8778`).
+
+
+.. _whatsnew_0152.performance:
+
+Performance
+~~~~~~~~~~~
+
+- Reduce memory usage when skiprows is an integer in read_csv (:issue:`8681`)
+- Performance boost for ``to_datetime`` conversions with a passed ``format=``, and the ``exact=False`` (:issue:`8904`)
+
+
+.. _whatsnew_0152.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in concat of Series with ``category`` dtype which were coercing to ``object``. (:issue:`8641`)
+- Bug in Timestamp-Timestamp not returning a Timedelta type and datelike-datelike ops with timezones (:issue:`8865`)
+- Made consistent a timezone mismatch exception (either tz operated with None or incompatible timezone), will now return ``TypeError`` rather than ``ValueError`` (a couple of edge cases only), (:issue:`8865`)
+- Bug in using a ``pd.Grouper(key=...)`` with no level/axis or level only (:issue:`8795`, :issue:`8866`)
+- Report a ``TypeError`` when invalid/no parameters are passed in a groupby (:issue:`8015`)
+- Bug in packaging pandas with ``py2app/cx_Freeze`` (:issue:`8602`, :issue:`8831`)
+- Bug in ``groupby`` signatures that didn't include \*args or \*\*kwargs (:issue:`8733`).
+- ``io.data.Options`` now raises ``RemoteDataError`` when no expiry dates are available from Yahoo and when it receives no data from Yahoo (:issue:`8761`), (:issue:`8783`).
+- Unclear error message in csv parsing when passing dtype and names and the parsed data is a different data type (:issue:`8833`)
+- Bug in slicing a MultiIndex with an empty list and at least one boolean indexer (:issue:`8781`)
+- ``io.data.Options`` now raises ``RemoteDataError`` when no expiry dates are available from Yahoo (:issue:`8761`).
+- ``Timedelta`` kwargs may now be numpy ints and floats (:issue:`8757`).
+- Fixed several outstanding bugs for ``Timedelta`` arithmetic and comparisons (:issue:`8813`, :issue:`5963`, :issue:`5436`).
+- ``sql_schema`` now generates dialect appropriate ``CREATE TABLE`` statements (:issue:`8697`)
+- ``slice`` string method now takes step into account (:issue:`8754`)
+- Bug in ``BlockManager`` where setting values with different type would break block integrity (:issue:`8850`)
+- Bug in ``DatetimeIndex`` when using ``time`` object as key (:issue:`8667`)
+- Bug in ``merge`` where ``how='left'`` and ``sort=False`` would not preserve left frame order (:issue:`7331`)
+- Bug in ``MultiIndex.reindex`` where reindexing at level would not reorder labels (:issue:`4088`)
+- Bug in certain operations with dateutil timezones, manifesting with dateutil 2.3 (:issue:`8639`)
+- Regression in DatetimeIndex iteration with a Fixed/Local offset timezone (:issue:`8890`)
+- Bug in ``to_datetime`` when parsing a nanoseconds using the ``%f`` format (:issue:`8989`)
+- ``io.data.Options`` now raises ``RemoteDataError`` when no expiry dates are available from Yahoo and when it receives no data from Yahoo (:issue:`8761`), (:issue:`8783`).
+- Fix: The font size was only set on x axis if vertical or the y axis if horizontal. (:issue:`8765`)
+- Fixed division by 0 when reading big csv files in python 3 (:issue:`8621`)
+- Bug in outputting a MultiIndex with ``to_html,index=False`` which would add an extra column (:issue:`8452`)
+- Imported categorical variables from Stata files retain the ordinal information in the underlying data (:issue:`8836`).
+- Defined ``.size`` attribute across ``NDFrame`` objects to provide compat with numpy >= 1.9.1; buggy with ``np.array_split`` (:issue:`8846`)
+- Skip testing of histogram plots for matplotlib <= 1.2 (:issue:`8648`).
+- Bug where ``get_data_google`` returned object dtypes (:issue:`3995`)
+- Bug in ``DataFrame.stack(..., dropna=False)`` when the DataFrame's ``columns`` is a ``MultiIndex``
+  whose ``labels`` do not reference all its ``levels``. (:issue:`8844`)
+- Bug in that Option context applied on ``__enter__`` (:issue:`8514`)
+- Bug in resample that causes a ValueError when resampling across multiple days
+  and the last offset is not calculated from the start of the range (:issue:`8683`)
+- Bug where ``DataFrame.plot(kind='scatter')`` fails when checking if an np.array is in the DataFrame (:issue:`8852`)
+- Bug in ``pd.infer_freq/DataFrame.inferred_freq`` that prevented proper sub-daily frequency inference when the index contained DST days (:issue:`8772`).
+- Bug where index name was still used when plotting a series with ``use_index=False`` (:issue:`8558`).
+- Bugs when trying to stack multiple columns, when some (or all) of the level names are numbers (:issue:`8584`).
+- Bug in ``MultiIndex`` where ``__contains__`` returns wrong result if index is not lexically sorted or unique (:issue:`7724`)
+- BUG CSV: fix problem with trailing white space in skipped rows, (:issue:`8679`), (:issue:`8661`), (:issue:`8983`)
+- Regression in ``Timestamp`` does not parse 'Z' zone designator for UTC (:issue:`8771`)
+- Bug in ``StataWriter`` the produces writes strings with 244 characters irrespective of actual size (:issue:`8969`)
+- Fixed ValueError raised by cummin/cummax when datetime64 Series contains NaT. (:issue:`8965`)
+- Bug in DataReader returns object dtype if there are missing values (:issue:`8980`)
+- Bug in plotting if sharex was enabled and index was a timeseries, would show labels on multiple axes (:issue:`3964`).
+- Bug where passing a unit to the TimedeltaIndex constructor applied the to nano-second conversion twice. (:issue:`9011`).
+- Bug in plotting of a period-like array (:issue:`9012`)
+
+
+.. _whatsnew_0.15.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.15.1..v0.15.2
diff --git a/pandas/doc/source/whatsnew/v0.16.0.rst b/pandas/doc/source/whatsnew/v0.16.0.rst
new file mode 100644
index 0000000000000..d53ea095bb96c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.16.0.rst
@@ -0,0 +1,690 @@
+.. _whatsnew_0160:
+
+Version 0.16.0 (March 22, 2015)
+-------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.15.2 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+Highlights include:
+
+- ``DataFrame.assign`` method, see :ref:`here <whatsnew_0160.enhancements.assign>`
+- ``Series.to_coo/from_coo`` methods to interact with ``scipy.sparse``, see :ref:`here <whatsnew_0160.enhancements.sparse>`
+- Backwards incompatible change to ``Timedelta`` to conform the ``.seconds`` attribute with ``datetime.timedelta``, see :ref:`here <whatsnew_0160.api_breaking.timedelta>`
+- Changes to the ``.loc`` slicing API to conform with the behavior of ``.ix`` see :ref:`here <whatsnew_0160.api_breaking.indexing>`
+- Changes to the default for ordering in the ``Categorical`` constructor, see :ref:`here <whatsnew_0160.api_breaking.categorical>`
+-  Enhancement to the ``.str`` accessor to make string operations easier, see :ref:`here <whatsnew_0160.enhancements.string>`
+- The ``pandas.tools.rplot``, ``pandas.sandbox.qtpandas`` and ``pandas.rpy``
+  modules are deprecated. We refer users to external packages like
+  `seaborn <http://stanford.edu/~mwaskom/software/seaborn/>`_,
+  `pandas-qt <https://github.com/datalyze-solutions/pandas-qt>`_ and
+  `rpy2 <http://rpy2.bitbucket.org/>`_ for similar or equivalent
+  functionality, see :ref:`here <whatsnew_0160.deprecations>`
+
+Check the :ref:`API Changes <whatsnew_0160.api>` and :ref:`deprecations <whatsnew_0160.deprecations>` before updating.
+
+.. contents:: What's new in v0.16.0
+    :local:
+    :backlinks: none
+
+
+.. _whatsnew_0160.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0160.enhancements.assign:
+
+DataFrame assign
+^^^^^^^^^^^^^^^^
+
+Inspired by `dplyr's
+<https://dplyr.tidyverse.org/articles/dplyr.html#mutating-operations>`__ ``mutate`` verb, DataFrame has a new
+:meth:`~pandas.DataFrame.assign` method.
+The function signature for ``assign`` is simply ``**kwargs``. The keys
+are the column names for the new fields, and the values are either a value
+to be inserted (for example, a ``Series`` or NumPy array), or a function
+of one argument to be called on the ``DataFrame``. The new values are inserted,
+and the entire DataFrame (with all original and new columns) is returned.
+
+.. ipython:: python
+
+   iris = pd.read_csv('data/iris.data')
+   iris.head()
+
+   iris.assign(sepal_ratio=iris['SepalWidth'] / iris['SepalLength']).head()
+
+Above was an example of inserting a precomputed value. We can also pass in
+a function to be evaluated.
+
+.. ipython:: python
+
+    iris.assign(sepal_ratio=lambda x: (x['SepalWidth']
+                                       / x['SepalLength'])).head()
+
+The power of ``assign`` comes when used in chains of operations. For example,
+we can limit the DataFrame to just those with a Sepal Length greater than 5,
+calculate the ratio, and plot
+
+.. ipython:: python
+
+   iris = pd.read_csv('data/iris.data')
+   (iris.query('SepalLength > 5')
+        .assign(SepalRatio=lambda x: x.SepalWidth / x.SepalLength,
+                PetalRatio=lambda x: x.PetalWidth / x.PetalLength)
+        .plot(kind='scatter', x='SepalRatio', y='PetalRatio'))
+
+.. image:: ../_static/whatsnew_assign.png
+  :scale: 50 %
+
+See the :ref:`documentation <dsintro.chained_assignment>` for more. (:issue:`9229`)
+
+
+.. _whatsnew_0160.enhancements.sparse:
+
+Interaction with scipy.sparse
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Added :meth:`SparseSeries.to_coo` and :meth:`SparseSeries.from_coo` methods (:issue:`8048`) for converting to and from ``scipy.sparse.coo_matrix`` instances (see :ref:`here <sparse.scipysparse>`). For example, given a SparseSeries with MultiIndex we can convert to a ``scipy.sparse.coo_matrix`` by specifying the row and column labels as index levels:
+
+.. code-block:: python
+
+   s = pd.Series([3.0, np.nan, 1.0, 3.0, np.nan, np.nan])
+   s.index = pd.MultiIndex.from_tuples([(1, 2, 'a', 0),
+                                        (1, 2, 'a', 1),
+                                        (1, 1, 'b', 0),
+                                        (1, 1, 'b', 1),
+                                        (2, 1, 'b', 0),
+                                        (2, 1, 'b', 1)],
+                                       names=['A', 'B', 'C', 'D'])
+
+   s
+
+   # SparseSeries
+   ss = s.to_sparse()
+   ss
+
+   A, rows, columns = ss.to_coo(row_levels=['A', 'B'],
+                                column_levels=['C', 'D'],
+                                sort_labels=False)
+
+   A
+   A.todense()
+   rows
+   columns
+
+The from_coo method is a convenience method for creating a ``SparseSeries``
+from a ``scipy.sparse.coo_matrix``:
+
+.. code-block:: python
+
+   from scipy import sparse
+   A = sparse.coo_matrix(([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])),
+                         shape=(3, 4))
+   A
+   A.todense()
+
+   ss = pd.SparseSeries.from_coo(A)
+   ss
+
+.. _whatsnew_0160.enhancements.string:
+
+String methods enhancements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Following new methods are accessible via ``.str`` accessor to apply the function to each values. This is intended to make it more consistent with standard methods on strings. (:issue:`9282`, :issue:`9352`, :issue:`9386`, :issue:`9387`, :issue:`9439`)
+
+  =============  =============  =============  ===============    ===============
+  ..             ..             Methods        ..                 ..
+  =============  =============  =============  ===============    ===============
+  ``isalnum()``  ``isalpha()``  ``isdigit()``  ``isdigit()``      ``isspace()``
+  ``islower()``  ``isupper()``  ``istitle()``  ``isnumeric()``    ``isdecimal()``
+  ``find()``     ``rfind()``    ``ljust()``    ``rjust()``        ``zfill()``
+  =============  =============  =============  ===============    ===============
+
+  .. ipython:: python
+
+     s = pd.Series(['abcd', '3456', 'EFGH'])
+     s.str.isalpha()
+     s.str.find('ab')
+
+- :meth:`Series.str.pad` and :meth:`Series.str.center` now accept ``fillchar`` option to specify filling character (:issue:`9352`)
+
+  .. ipython:: python
+
+     s = pd.Series(['12', '300', '25'])
+     s.str.pad(5, fillchar='_')
+
+- Added :meth:`Series.str.slice_replace`, which previously raised ``NotImplementedError`` (:issue:`8888`)
+
+  .. ipython:: python
+
+     s = pd.Series(['ABCD', 'EFGH', 'IJK'])
+     s.str.slice_replace(1, 3, 'X')
+     # replaced with empty char
+     s.str.slice_replace(0, 1)
+
+.. _whatsnew_0160.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Reindex now supports ``method='nearest'`` for frames or series with a monotonic increasing or decreasing index (:issue:`9258`):
+
+  .. ipython:: python
+
+     df = pd.DataFrame({'x': range(5)})
+     df.reindex([0.2, 1.8, 3.5], method='nearest')
+
+  This method is also exposed by the lower level ``Index.get_indexer`` and ``Index.get_loc`` methods.
+
+- The ``read_excel()`` function's :ref:`sheetname <io.excel.specifying_sheets>` argument now accepts a list and ``None``, to get multiple or all sheets respectively.  If more than one sheet is specified, a dictionary is returned. (:issue:`9450`)
+
+  .. code-block:: python
+
+     # Returns the 1st and 4th sheet, as a dictionary of DataFrames.
+     pd.read_excel('path_to_file.xls', sheetname=['Sheet1', 3])
+
+
+- Allow Stata files to be read incrementally with an iterator; support for long strings in Stata files. See the docs :ref:`here<io.stata_reader>` (:issue:`9493`:).
+- Paths beginning with ~ will now be expanded to begin with the user's home directory (:issue:`9066`)
+- Added time interval selection in ``get_data_yahoo`` (:issue:`9071`)
+- Added ``Timestamp.to_datetime64()`` to complement ``Timedelta.to_timedelta64()`` (:issue:`9255`)
+- ``tseries.frequencies.to_offset()`` now accepts ``Timedelta`` as input (:issue:`9064`)
+- Lag parameter was added to the autocorrelation method of ``Series``, defaults to lag-1 autocorrelation (:issue:`9192`)
+- ``Timedelta`` will now accept ``nanoseconds`` keyword in constructor (:issue:`9273`)
+- SQL code now safely escapes table and column names (:issue:`8986`)
+- Added auto-complete for ``Series.str.<tab>``, ``Series.dt.<tab>`` and ``Series.cat.<tab>`` (:issue:`9322`)
+- ``Index.get_indexer`` now supports ``method='pad'`` and ``method='backfill'`` even for any target array, not just monotonic targets. These methods also work for monotonic decreasing as well as monotonic increasing indexes (:issue:`9258`).
+- ``Index.asof`` now works on all index types (:issue:`9258`).
+- A ``verbose`` argument has been augmented in ``io.read_excel()``, defaults to False. Set to True to print sheet names as they are parsed. (:issue:`9450`)
+- Added ``days_in_month`` (compatibility alias ``daysinmonth``) property to ``Timestamp``, ``DatetimeIndex``, ``Period``, ``PeriodIndex``, and ``Series.dt`` (:issue:`9572`)
+- Added ``decimal`` option in ``to_csv`` to provide formatting for non-'.' decimal separators (:issue:`781`)
+- Added ``normalize`` option for ``Timestamp`` to normalized to midnight (:issue:`8794`)
+- Added example for ``DataFrame`` import to R using HDF5 file and ``rhdf5``
+  library. See the documentation for more
+  (:issue:`9636`).
+
+.. _whatsnew_0160.api:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0160.api_breaking:
+
+.. _whatsnew_0160.api_breaking.timedelta:
+
+Changes in timedelta
+^^^^^^^^^^^^^^^^^^^^
+
+In v0.15.0 a new scalar type ``Timedelta`` was introduced, that is a
+sub-class of ``datetime.timedelta``. Mentioned :ref:`here <whatsnew_0150.timedeltaindex>` was a notice of an API change w.r.t. the ``.seconds`` accessor. The intent was to provide a user-friendly set of accessors that give the 'natural' value for that unit, e.g. if you had a ``Timedelta('1 day, 10:11:12')``, then ``.seconds`` would return 12. However, this is at odds with the definition of ``datetime.timedelta``, which defines ``.seconds`` as ``10 * 3600 + 11 * 60 + 12 == 36672``.
+
+So in v0.16.0, we are restoring the API to match that of ``datetime.timedelta``. Further, the component values are still available through the ``.components`` accessor. This affects the ``.seconds`` and ``.microseconds`` accessors, and removes the ``.hours``, ``.minutes``, ``.milliseconds`` accessors. These changes affect ``TimedeltaIndex`` and the Series ``.dt`` accessor as well. (:issue:`9185`, :issue:`9139`)
+
+Previous behavior
+
+.. code-block:: ipython
+
+   In [2]: t = pd.Timedelta('1 day, 10:11:12.100123')
+
+   In [3]: t.days
+   Out[3]: 1
+
+   In [4]: t.seconds
+   Out[4]: 12
+
+   In [5]: t.microseconds
+   Out[5]: 123
+
+New behavior
+
+.. ipython:: python
+
+   t = pd.Timedelta('1 day, 10:11:12.100123')
+   t.days
+   t.seconds
+   t.microseconds
+
+Using ``.components`` allows the full component access
+
+.. ipython:: python
+
+   t.components
+   t.components.seconds
+
+.. _whatsnew_0160.api_breaking.indexing:
+
+Indexing changes
+^^^^^^^^^^^^^^^^
+
+The behavior of a small sub-set of edge cases for using ``.loc`` have changed (:issue:`8613`). Furthermore we have improved the content of the error messages that are raised:
+
+- Slicing with ``.loc`` where the start and/or stop bound is not found in the index is now allowed; this previously would raise a ``KeyError``. This makes the behavior the same as ``.ix`` in this case. This change is only for slicing, not when indexing with a single label.
+
+  .. ipython:: python
+
+     df = pd.DataFrame(np.random.randn(5, 4),
+                       columns=list('ABCD'),
+                       index=pd.date_range('20130101', periods=5))
+     df
+     s = pd.Series(range(5), [-2, -1, 1, 2, 3])
+     s
+
+  Previous behavior
+
+  .. code-block:: ipython
+
+     In [4]: df.loc['2013-01-02':'2013-01-10']
+     KeyError: 'stop bound [2013-01-10] is not in the [index]'
+
+     In [6]: s.loc[-10:3]
+     KeyError: 'start bound [-10] is not the [index]'
+
+  New behavior
+
+  .. ipython:: python
+
+     df.loc['2013-01-02':'2013-01-10']
+     s.loc[-10:3]
+
+- Allow slicing with float-like values on an integer index for ``.ix``. Previously this was only enabled for ``.loc``:
+
+  Previous behavior
+
+  .. code-block:: ipython
+
+     In [8]: s.ix[-1.0:2]
+     TypeError: the slice start value [-1.0] is not a proper indexer for this index type (Int64Index)
+
+  New behavior
+
+  .. code-block:: python
+
+     In [2]: s.ix[-1.0:2]
+     Out[2]:
+     -1    1
+      1    2
+      2    3
+     dtype: int64
+
+- Provide a useful exception for indexing with an invalid type for that index when using ``.loc``. For example trying to use ``.loc`` on an index of type ``DatetimeIndex`` or ``PeriodIndex`` or ``TimedeltaIndex``, with an integer (or a float).
+
+  Previous behavior
+
+  .. code-block:: python
+
+     In [4]: df.loc[2:3]
+     KeyError: 'start bound [2] is not the [index]'
+
+  New behavior
+
+  .. code-block:: ipython
+
+     In [4]: df.loc[2:3]
+     TypeError: Cannot do slice indexing on <class 'pandas.tseries.index.DatetimeIndex'> with <type 'int'> keys
+
+
+.. _whatsnew_0160.api_breaking.categorical:
+
+Categorical changes
+^^^^^^^^^^^^^^^^^^^
+
+In prior versions, ``Categoricals`` that had an unspecified ordering (meaning no ``ordered`` keyword was passed) were defaulted as ``ordered`` Categoricals. Going forward, the ``ordered`` keyword in the ``Categorical`` constructor will default to ``False``. Ordering must now be explicit.
+
+Furthermore, previously you *could* change the ``ordered`` attribute of a Categorical by just setting the attribute, e.g. ``cat.ordered=True``; This is now deprecated and you should use ``cat.as_ordered()`` or ``cat.as_unordered()``. These will by default return a **new** object and not modify the existing object. (:issue:`9347`, :issue:`9190`)
+
+Previous behavior
+
+.. code-block:: ipython
+
+   In [3]: s = pd.Series([0, 1, 2], dtype='category')
+
+   In [4]: s
+   Out[4]:
+   0    0
+   1    1
+   2    2
+   dtype: category
+   Categories (3, int64): [0 < 1 < 2]
+
+   In [5]: s.cat.ordered
+   Out[5]: True
+
+   In [6]: s.cat.ordered = False
+
+   In [7]: s
+   Out[7]:
+   0    0
+   1    1
+   2    2
+   dtype: category
+   Categories (3, int64): [0, 1, 2]
+
+New behavior
+
+.. ipython:: python
+
+   s = pd.Series([0, 1, 2], dtype='category')
+   s
+   s.cat.ordered
+   s = s.cat.as_ordered()
+   s
+   s.cat.ordered
+
+   # you can set in the constructor of the Categorical
+   s = pd.Series(pd.Categorical([0, 1, 2], ordered=True))
+   s
+   s.cat.ordered
+
+For ease of creation of series of categorical data, we have added the ability to pass keywords when calling ``.astype()``. These are passed directly to the constructor.
+
+.. code-block:: python
+
+    In [54]: s = pd.Series(["a", "b", "c", "a"]).astype('category', ordered=True)
+
+    In [55]: s
+    Out[55]:
+    0    a
+    1    b
+    2    c
+    3    a
+    dtype: category
+    Categories (3, object): [a < b < c]
+
+    In [56]: s = (pd.Series(["a", "b", "c", "a"])
+       ....:        .astype('category', categories=list('abcdef'), ordered=False))
+
+    In [57]: s
+    Out[57]:
+    0    a
+    1    b
+    2    c
+    3    a
+    dtype: category
+    Categories (6, object): [a, b, c, d, e, f]
+
+
+.. _whatsnew_0160.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- ``Index.duplicated`` now returns ``np.array(dtype=bool)`` rather than ``Index(dtype=object)`` containing ``bool`` values. (:issue:`8875`)
+- ``DataFrame.to_json`` now returns accurate type serialisation for each column for frames of mixed dtype (:issue:`9037`)
+
+  Previously data was coerced to a common dtype before serialisation, which for
+  example resulted in integers being serialised to floats:
+
+  .. code-block:: ipython
+
+    In [2]: pd.DataFrame({'i': [1,2], 'f': [3.0, 4.2]}).to_json()
+    Out[2]: '{"f":{"0":3.0,"1":4.2},"i":{"0":1.0,"1":2.0}}'
+
+  Now each column is serialised using its correct dtype:
+
+  .. code-block:: ipython
+
+    In [2]:  pd.DataFrame({'i': [1,2], 'f': [3.0, 4.2]}).to_json()
+    Out[2]: '{"f":{"0":3.0,"1":4.2},"i":{"0":1,"1":2}}'
+
+- ``DatetimeIndex``, ``PeriodIndex`` and ``TimedeltaIndex.summary`` now output the same format. (:issue:`9116`)
+- ``TimedeltaIndex.freqstr`` now output the same string format as ``DatetimeIndex``. (:issue:`9116`)
+
+- Bar and horizontal bar plots no longer add a dashed line along the info axis. The prior style can be achieved with matplotlib's ``axhline`` or ``axvline`` methods (:issue:`9088`).
+
+- ``Series`` accessors ``.dt``, ``.cat`` and ``.str`` now raise ``AttributeError`` instead of ``TypeError`` if the series does not contain the appropriate type of data (:issue:`9617`). This follows Python's built-in exception hierarchy more closely and ensures that tests like ``hasattr(s, 'cat')`` are consistent on both Python 2 and 3.
+
+- ``Series`` now supports bitwise operation for integral types (:issue:`9016`). Previously even if the input dtypes were integral, the output dtype was coerced to ``bool``.
+
+  Previous behavior
+
+  .. code-block:: ipython
+
+     In [2]: pd.Series([0, 1, 2, 3], list('abcd')) | pd.Series([4, 4, 4, 4], list('abcd'))
+     Out[2]:
+     a    True
+     b    True
+     c    True
+     d    True
+     dtype: bool
+
+  New behavior. If the input dtypes are integral, the output dtype is also integral and the output
+  values are the result of the bitwise operation.
+
+  .. code-block:: ipython
+
+     In [2]: pd.Series([0, 1, 2, 3], list('abcd')) | pd.Series([4, 4, 4, 4], list('abcd'))
+     Out[2]:
+     a    4
+     b    5
+     c    6
+     d    7
+     dtype: int64
+
+
+- During division involving a ``Series`` or ``DataFrame``, ``0/0`` and ``0//0`` now give ``np.nan`` instead of ``np.inf``. (:issue:`9144`, :issue:`8445`)
+
+  Previous behavior
+
+  .. code-block:: ipython
+
+        In [2]: p = pd.Series([0, 1])
+
+        In [3]: p / 0
+        Out[3]:
+        0    inf
+        1    inf
+        dtype: float64
+
+        In [4]: p // 0
+        Out[4]:
+        0    inf
+        1    inf
+        dtype: float64
+
+
+
+  New behavior
+
+  .. ipython:: python
+
+     p = pd.Series([0, 1])
+     p / 0
+     p // 0
+
+- ``Series.values_counts`` and ``Series.describe`` for categorical data will now put ``NaN`` entries at the end. (:issue:`9443`)
+- ``Series.describe`` for categorical data will now give counts and frequencies of 0, not ``NaN``, for unused categories (:issue:`9443`)
+
+- Due to a bug fix, looking up a partial string label with ``DatetimeIndex.asof`` now includes values that match the string, even if they are after the start of the partial string label (:issue:`9258`).
+
+  Old behavior:
+
+  .. code-block:: ipython
+
+    In [4]: pd.to_datetime(['2000-01-31', '2000-02-28']).asof('2000-02')
+    Out[4]: Timestamp('2000-01-31 00:00:00')
+
+  Fixed behavior:
+
+  .. ipython:: python
+
+    pd.to_datetime(['2000-01-31', '2000-02-28']).asof('2000-02')
+
+  To reproduce the old behavior, simply add more precision to the label (e.g., use ``2000-02-01`` instead of ``2000-02``).
+
+
+.. _whatsnew_0160.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- The ``rplot`` trellis plotting interface is deprecated and will be removed
+  in a future version. We refer to external packages like
+  `seaborn <http://stanford.edu/~mwaskom/software/seaborn/>`_ for similar
+  but more refined functionality (:issue:`3445`).
+  The documentation includes some examples how to convert your existing code
+  from ``rplot`` to seaborn `here <https://pandas.pydata.org/pandas-docs/version/0.18.1/visualization.html#trellis-plotting-interface>`__.
+
+- The ``pandas.sandbox.qtpandas`` interface is deprecated and will be removed in a future version.
+  We refer users to the external package `pandas-qt <https://github.com/datalyze-solutions/pandas-qt>`_. (:issue:`9615`)
+
+- The ``pandas.rpy`` interface is deprecated and will be removed in a future version.
+  Similar functionality can be accessed through the `rpy2 <http://rpy2.bitbucket.org/>`_ project (:issue:`9602`)
+
+- Adding ``DatetimeIndex/PeriodIndex`` to another ``DatetimeIndex/PeriodIndex`` is being deprecated as a set-operation. This will be changed to a ``TypeError`` in a future version. ``.union()`` should be used for the union set operation. (:issue:`9094`)
+- Subtracting ``DatetimeIndex/PeriodIndex`` from another ``DatetimeIndex/PeriodIndex`` is being deprecated as a set-operation. This will be changed to an actual numeric subtraction yielding a ``TimeDeltaIndex`` in a future version. ``.difference()`` should be used for the differencing set operation. (:issue:`9094`)
+
+
+.. _whatsnew_0160.prior_deprecations:
+
+Removal of prior version deprecations/changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- ``DataFrame.pivot_table`` and ``crosstab``'s ``rows`` and ``cols`` keyword arguments were removed in favor
+  of ``index`` and ``columns`` (:issue:`6581`)
+- ``DataFrame.to_excel`` and ``DataFrame.to_csv`` ``cols`` keyword argument was removed in favor of ``columns`` (:issue:`6581`)
+- Removed ``convert_dummies`` in favor of ``get_dummies`` (:issue:`6581`)
+- Removed ``value_range`` in favor of ``describe`` (:issue:`6581`)
+
+.. _whatsnew_0160.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Fixed a performance regression for ``.loc`` indexing with an array or list-like (:issue:`9126`:).
+- ``DataFrame.to_json`` 30x performance improvement for mixed dtype frames. (:issue:`9037`)
+- Performance improvements in ``MultiIndex.duplicated`` by working with labels instead of values (:issue:`9125`)
+- Improved the speed of ``nunique`` by calling ``unique`` instead of ``value_counts`` (:issue:`9129`, :issue:`7771`)
+- Performance improvement of up to 10x in ``DataFrame.count`` and ``DataFrame.dropna`` by taking advantage of homogeneous/heterogeneous dtypes appropriately (:issue:`9136`)
+- Performance improvement of up to 20x in ``DataFrame.count`` when using a ``MultiIndex`` and the ``level`` keyword argument  (:issue:`9163`)
+- Performance and memory usage improvements in ``merge`` when key space exceeds ``int64`` bounds (:issue:`9151`)
+- Performance improvements in multi-key ``groupby`` (:issue:`9429`)
+- Performance improvements in ``MultiIndex.sortlevel`` (:issue:`9445`)
+- Performance and memory usage improvements in ``DataFrame.duplicated`` (:issue:`9398`)
+- Cythonized ``Period`` (:issue:`9440`)
+- Decreased memory usage on ``to_hdf`` (:issue:`9648`)
+
+.. _whatsnew_0160.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Changed ``.to_html`` to remove leading/trailing spaces in table body (:issue:`4987`)
+- Fixed issue using ``read_csv`` on s3 with Python 3 (:issue:`9452`)
+- Fixed compatibility issue in ``DatetimeIndex`` affecting architectures where ``numpy.int_`` defaults to ``numpy.int32`` (:issue:`8943`)
+- Bug in Panel indexing with an object-like (:issue:`9140`)
+- Bug in the returned ``Series.dt.components`` index was reset to the default index (:issue:`9247`)
+- Bug in ``Categorical.__getitem__/__setitem__`` with listlike input getting incorrect results from indexer coercion (:issue:`9469`)
+- Bug in partial setting with a DatetimeIndex (:issue:`9478`)
+- Bug in groupby for integer and datetime64 columns when applying an aggregator that caused the value to be
+  changed when the number was sufficiently large (:issue:`9311`, :issue:`6620`)
+- Fixed bug in ``to_sql`` when mapping a ``Timestamp`` object column (datetime
+  column with timezone info) to the appropriate sqlalchemy type (:issue:`9085`).
+- Fixed bug in ``to_sql`` ``dtype`` argument not accepting an instantiated
+  SQLAlchemy type  (:issue:`9083`).
+- Bug in ``.loc`` partial setting with a ``np.datetime64`` (:issue:`9516`)
+- Incorrect dtypes inferred on datetimelike looking ``Series`` & on ``.xs`` slices (:issue:`9477`)
+- Items in ``Categorical.unique()`` (and ``s.unique()`` if ``s`` is of dtype ``category``) now appear in the order in which they are originally found, not in sorted order (:issue:`9331`). This is now consistent with the behavior for other dtypes in pandas.
+- Fixed bug on big endian platforms which produced incorrect results in ``StataReader`` (:issue:`8688`).
+- Bug in ``MultiIndex.has_duplicates`` when having many levels causes an indexer overflow (:issue:`9075`, :issue:`5873`)
+- Bug in ``pivot`` and ``unstack`` where ``nan`` values would break index alignment (:issue:`4862`, :issue:`7401`, :issue:`7403`, :issue:`7405`, :issue:`7466`, :issue:`9497`)
+- Bug in left ``join`` on MultiIndex with ``sort=True`` or null values (:issue:`9210`).
+- Bug in ``MultiIndex`` where inserting new keys would fail (:issue:`9250`).
+- Bug in ``groupby`` when key space exceeds ``int64`` bounds (:issue:`9096`).
+- Bug in ``unstack`` with ``TimedeltaIndex`` or ``DatetimeIndex`` and nulls (:issue:`9491`).
+- Bug in ``rank`` where comparing floats with tolerance will cause inconsistent behaviour (:issue:`8365`).
+- Fixed character encoding bug in ``read_stata`` and ``StataReader`` when loading data from a URL (:issue:`9231`).
+- Bug in adding ``offsets.Nano`` to other offsets raises ``TypeError`` (:issue:`9284`)
+- Bug in ``DatetimeIndex`` iteration, related to (:issue:`8890`), fixed in (:issue:`9100`)
+- Bugs in ``resample`` around DST transitions. This required fixing offset classes so they behave correctly on DST transitions. (:issue:`5172`, :issue:`8744`, :issue:`8653`, :issue:`9173`, :issue:`9468`).
+- Bug in binary operator method (eg ``.mul()``) alignment with integer levels (:issue:`9463`).
+- Bug in boxplot, scatter and hexbin plot may show an unnecessary warning (:issue:`8877`)
+- Bug in subplot with ``layout`` kw may show unnecessary warning (:issue:`9464`)
+- Bug in using grouper functions that need passed through arguments (e.g. axis), when using wrapped function (e.g. ``fillna``), (:issue:`9221`)
+- ``DataFrame`` now properly supports simultaneous ``copy`` and ``dtype`` arguments in constructor (:issue:`9099`)
+- Bug in ``read_csv`` when using skiprows on a file with CR line endings with the c engine. (:issue:`9079`)
+- ``isnull`` now detects ``NaT`` in ``PeriodIndex`` (:issue:`9129`)
+- Bug in groupby ``.nth()`` with a multiple column groupby (:issue:`8979`)
+- Bug in ``DataFrame.where`` and ``Series.where`` coerce numerics to string incorrectly (:issue:`9280`)
+- Bug in ``DataFrame.where`` and ``Series.where`` raise ``ValueError`` when string list-like is passed. (:issue:`9280`)
+- Accessing ``Series.str`` methods on with non-string values now raises ``TypeError`` instead of producing incorrect results (:issue:`9184`)
+- Bug in ``DatetimeIndex.__contains__`` when index has duplicates and is not monotonic increasing (:issue:`9512`)
+- Fixed division by zero error for ``Series.kurt()`` when all values are equal (:issue:`9197`)
+- Fixed issue in the ``xlsxwriter`` engine where it added a default 'General' format to cells if no other format was applied. This prevented other row or column formatting being applied. (:issue:`9167`)
+- Fixes issue with ``index_col=False`` when ``usecols`` is also specified in ``read_csv``. (:issue:`9082`)
+- Bug where ``wide_to_long`` would modify the input stub names list (:issue:`9204`)
+- Bug in ``to_sql`` not storing float64 values using double precision. (:issue:`9009`)
+- ``SparseSeries`` and ``SparsePanel`` now accept zero argument constructors (same as their non-sparse counterparts) (:issue:`9272`).
+- Regression in merging ``Categorical`` and ``object`` dtypes (:issue:`9426`)
+- Bug in ``read_csv`` with buffer overflows with certain malformed input files (:issue:`9205`)
+- Bug in groupby MultiIndex with missing pair (:issue:`9049`, :issue:`9344`)
+- Fixed bug in ``Series.groupby`` where grouping on ``MultiIndex`` levels would ignore the sort argument (:issue:`9444`)
+- Fix bug in ``DataFrame.Groupby`` where ``sort=False`` is ignored in the case of Categorical columns. (:issue:`8868`)
+- Fixed bug with reading CSV files from Amazon S3 on python 3 raising a TypeError (:issue:`9452`)
+- Bug in the Google BigQuery reader where the 'jobComplete' key may be present but False in the query results (:issue:`8728`)
+- Bug in ``Series.values_counts`` with excluding ``NaN`` for categorical type ``Series`` with ``dropna=True`` (:issue:`9443`)
+- Fixed missing numeric_only option for ``DataFrame.std/var/sem`` (:issue:`9201`)
+- Support constructing ``Panel`` or ``Panel4D`` with scalar data (:issue:`8285`)
+- ``Series`` text representation disconnected from ``max_rows``/``max_columns`` (:issue:`7508`).
+
+\
+
+- ``Series`` number formatting inconsistent when truncated (:issue:`8532`).
+
+  Previous behavior
+
+  .. code-block:: python
+
+    In [2]: pd.options.display.max_rows = 10
+    In [3]: s = pd.Series([1,1,1,1,1,1,1,1,1,1,0.9999,1,1]*10)
+    In [4]: s
+    Out[4]:
+    0    1
+    1    1
+    2    1
+    ...
+    127    0.9999
+    128    1.0000
+    129    1.0000
+    Length: 130, dtype: float64
+
+  New behavior
+
+  .. code-block:: python
+
+    0      1.0000
+    1      1.0000
+    2      1.0000
+    3      1.0000
+    4      1.0000
+    ...
+    125    1.0000
+    126    1.0000
+    127    0.9999
+    128    1.0000
+    129    1.0000
+    dtype: float64
+
+- A Spurious ``SettingWithCopy`` Warning was generated when setting a new item in a frame in some cases (:issue:`8730`)
+
+  The following would previously report a ``SettingWithCopy`` Warning.
+
+  .. ipython:: python
+
+     df1 = pd.DataFrame({'x': pd.Series(['a', 'b', 'c']),
+                         'y': pd.Series(['d', 'e', 'f'])})
+     df2 = df1[['x']]
+     df2['y'] = ['g', 'h', 'i']
+
+
+.. _whatsnew_0.16.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.15.2..v0.16.0
diff --git a/pandas/doc/source/whatsnew/v0.16.1.rst b/pandas/doc/source/whatsnew/v0.16.1.rst
new file mode 100644
index 0000000000000..cbf5b7703bd79
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.16.1.rst
@@ -0,0 +1,480 @@
+.. _whatsnew_0161:
+
+Version 0.16.1 (May 11, 2015)
+-----------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release from 0.16.0 and includes a large number of
+bug fixes along several new features, enhancements, and performance improvements.
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- Support for a ``CategoricalIndex``, a category based index, see :ref:`here <whatsnew_0161.enhancements.categoricalindex>`
+- New section on how-to-contribute to *pandas*, see :ref:`here <contributing>`
+- Revised "Merge, join, and concatenate" documentation, including graphical examples to make it easier to understand each operations, see :ref:`here <merging>`
+- New method ``sample`` for drawing random samples from Series, DataFrames and Panels. See :ref:`here <whatsnew_0161.enhancements.sample>`
+- The default ``Index`` printing has changed to a more uniform format, see :ref:`here <whatsnew_0161.index_repr>`
+- ``BusinessHour`` datetime-offset is now supported, see :ref:`here <timeseries.businesshour>`
+
+-  Further enhancement to the ``.str`` accessor to make string operations easier, see :ref:`here <whatsnew_0161.enhancements.string>`
+
+.. contents:: What's new in v0.16.1
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0161.enhancements:
+
+.. warning::
+
+   In pandas 0.17.0, the sub-package ``pandas.io.data`` will be removed in favor of a separately installable package (:issue:`8961`).
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_0161.enhancements.categoricalindex:
+
+CategoricalIndex
+^^^^^^^^^^^^^^^^
+
+We introduce a ``CategoricalIndex``, a new type of index object that is useful for supporting
+indexing with duplicates. This is a container around a ``Categorical`` (introduced in v0.15.0)
+and allows efficient indexing and storage of an index with a large number of duplicated elements. Prior to 0.16.1,
+setting the index of a ``DataFrame/Series`` with a ``category`` dtype would convert this to regular object-based ``Index``.
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({'A': np.arange(6),
+       ...:                    'B': pd.Series(list('aabbca'))
+       ...:                           .astype('category', categories=list('cab'))
+       ...:                    })
+       ...:
+
+    In [2]: df
+    Out[2]:
+       A  B
+    0  0  a
+    1  1  a
+    2  2  b
+    3  3  b
+    4  4  c
+    5  5  a
+
+    In [3]: df.dtypes
+    Out[3]:
+    A       int64
+    B    category
+    dtype: object
+
+    In [4]: df.B.cat.categories
+    Out[4]: Index(['c', 'a', 'b'], dtype='object')
+
+
+setting the index, will create a ``CategoricalIndex``
+
+.. code-block:: ipython
+
+    In [5]: df2 = df.set_index('B')
+
+    In [6]: df2.index
+    Out[6]: CategoricalIndex(['a', 'a', 'b', 'b', 'c', 'a'], categories=['c', 'a', 'b'], ordered=False, name='B', dtype='category')
+
+indexing with ``__getitem__/.iloc/.loc/.ix`` works similarly to an Index with duplicates.
+The indexers MUST be in the category or the operation will raise.
+
+.. code-block:: ipython
+
+    In [7]: df2.loc['a']
+    Out[7]:
+       A
+    B
+    a  0
+    a  1
+    a  5
+
+and preserves the ``CategoricalIndex``
+
+.. code-block:: ipython
+
+    In [8]: df2.loc['a'].index
+    Out[8]: CategoricalIndex(['a', 'a', 'a'], categories=['c', 'a', 'b'], ordered=False, name='B', dtype='category')
+
+
+sorting will order by the order of the categories
+
+.. code-block:: ipython
+
+    In [9]: df2.sort_index()
+    Out[9]:
+       A
+    B
+    c  4
+    a  0
+    a  1
+    a  5
+    b  2
+    b  3
+
+groupby operations on the index will preserve the index nature as well
+
+.. code-block:: ipython
+
+    In [10]: df2.groupby(level=0).sum()
+    Out[10]:
+       A
+    B
+    c  4
+    a  6
+    b  5
+
+    In [11]: df2.groupby(level=0).sum().index
+    Out[11]: CategoricalIndex(['c', 'a', 'b'], categories=['c', 'a', 'b'], ordered=False, name='B', dtype='category')
+
+
+reindexing operations, will return a resulting index based on the type of the passed
+indexer, meaning that passing a list will return a plain-old-``Index``; indexing with
+a ``Categorical`` will return a ``CategoricalIndex``, indexed according to the categories
+of the PASSED ``Categorical`` dtype. This allows one to arbitrarily index these even with
+values NOT in the categories, similarly to how you can reindex ANY pandas index.
+
+.. code-block:: ipython
+
+    In [12]: df2.reindex(['a', 'e'])
+    Out[12]:
+         A
+    B
+    a  0.0
+    a  1.0
+    a  5.0
+    e  NaN
+
+    In [13]: df2.reindex(['a', 'e']).index
+    Out[13]: pd.Index(['a', 'a', 'a', 'e'], dtype='object', name='B')
+
+    In [14]: df2.reindex(pd.Categorical(['a', 'e'], categories=list('abcde')))
+    Out[14]:
+         A
+    B
+    a  0.0
+    a  1.0
+    a  5.0
+    e  NaN
+
+    In [15]: df2.reindex(pd.Categorical(['a', 'e'], categories=list('abcde'))).index
+    Out[15]: pd.CategoricalIndex(['a', 'a', 'a', 'e'],
+                                 categories=['a', 'b', 'c', 'd', 'e'],
+                                 ordered=False, name='B',
+                                 dtype='category')
+
+See the :ref:`documentation <advanced.categoricalindex>` for more. (:issue:`7629`, :issue:`10038`, :issue:`10039`)
+
+.. _whatsnew_0161.enhancements.sample:
+
+Sample
+^^^^^^
+
+Series, DataFrames, and Panels now have a new method: :meth:`~pandas.DataFrame.sample`.
+The method accepts a specific number of rows or columns to return, or a fraction of the
+total number or rows or columns. It also has options for sampling with or without replacement,
+for passing in a column for weights for non-uniform sampling, and for setting seed values to
+facilitate replication. (:issue:`2419`)
+
+.. ipython:: python
+
+   example_series = pd.Series([0, 1, 2, 3, 4, 5])
+
+   # When no arguments are passed, returns 1
+   example_series.sample()
+
+   # One may specify either a number of rows:
+   example_series.sample(n=3)
+
+   # Or a fraction of the rows:
+   example_series.sample(frac=0.5)
+
+   # weights are accepted.
+   example_weights = [0, 0, 0.2, 0.2, 0.2, 0.4]
+   example_series.sample(n=3, weights=example_weights)
+
+   # weights will also be normalized if they do not sum to one,
+   # and missing values will be treated as zeros.
+   example_weights2 = [0.5, 0, 0, 0, None, np.nan]
+   example_series.sample(n=1, weights=example_weights2)
+
+
+When applied to a DataFrame, one may pass the name of a column to specify sampling weights
+when sampling from rows.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"col1": [9, 8, 7, 6], "weight_column": [0.5, 0.4, 0.1, 0]})
+   df.sample(n=3, weights="weight_column")
+
+
+.. _whatsnew_0161.enhancements.string:
+
+String methods enhancements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:ref:`Continuing from v0.16.0 <whatsnew_0160.enhancements.string>`, the following
+enhancements make string operations easier and more consistent with standard python string operations.
+
+
+- Added ``StringMethods`` (``.str`` accessor) to ``Index`` (:issue:`9068`)
+
+  The ``.str`` accessor is now available for both ``Series`` and ``Index``.
+
+  .. ipython:: python
+
+     idx = pd.Index([" jack", "jill ", " jesse ", "frank"])
+     idx.str.strip()
+
+  One special case for the ``.str`` accessor on ``Index`` is that if a string method returns ``bool``, the ``.str`` accessor
+  will return a ``np.array`` instead of a boolean ``Index`` (:issue:`8875`). This enables the following expression
+  to work naturally:
+
+  .. ipython:: python
+
+     idx = pd.Index(["a1", "a2", "b1", "b2"])
+     s = pd.Series(range(4), index=idx)
+     s
+     idx.str.startswith("a")
+     s[s.index.str.startswith("a")]
+
+- The following new methods are accessible via ``.str`` accessor to apply the function to each values. (:issue:`9766`, :issue:`9773`, :issue:`10031`, :issue:`10045`, :issue:`10052`)
+
+  ================  ===============  ===============  ===============  ================
+  ..                ..               Methods          ..               ..
+  ================  ===============  ===============  ===============  ================
+  ``capitalize()``  ``swapcase()``   ``normalize()``  ``partition()``  ``rpartition()``
+  ``index()``       ``rindex()``     ``translate()``
+  ================  ===============  ===============  ===============  ================
+
+- ``split`` now takes ``expand`` keyword to specify whether to expand dimensionality. ``return_type`` is deprecated. (:issue:`9847`)
+
+  .. ipython:: python
+
+     s = pd.Series(["a,b", "a,c", "b,c"])
+
+     # return Series
+     s.str.split(",")
+
+     # return DataFrame
+     s.str.split(",", expand=True)
+
+     idx = pd.Index(["a,b", "a,c", "b,c"])
+
+     # return Index
+     idx.str.split(",")
+
+     # return MultiIndex
+     idx.str.split(",", expand=True)
+
+
+- Improved ``extract`` and ``get_dummies`` methods for ``Index.str`` (:issue:`9980`)
+
+
+.. _whatsnew_0161.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- ``BusinessHour`` offset is now supported, which represents business hours starting from 09:00 - 17:00 on ``BusinessDay`` by default. See :ref:`Here <timeseries.businesshour>` for details. (:issue:`7905`)
+
+  .. ipython:: python
+
+     pd.Timestamp("2014-08-01 09:00") + pd.tseries.offsets.BusinessHour()
+     pd.Timestamp("2014-08-01 07:00") + pd.tseries.offsets.BusinessHour()
+     pd.Timestamp("2014-08-01 16:30") + pd.tseries.offsets.BusinessHour()
+
+- ``DataFrame.diff`` now takes an ``axis`` parameter that determines the direction of differencing (:issue:`9727`)
+
+- Allow ``clip``, ``clip_lower``, and ``clip_upper`` to accept array-like arguments as thresholds (This is a regression from 0.11.0). These methods now have an ``axis`` parameter which determines how the Series or DataFrame will be aligned with the threshold(s). (:issue:`6966`)
+
+- ``DataFrame.mask()`` and ``Series.mask()`` now support same keywords as ``where`` (:issue:`8801`)
+
+- ``drop`` function can now accept ``errors`` keyword to suppress ``ValueError`` raised when any of label does not exist in the target data. (:issue:`6736`)
+
+  .. ipython:: python
+
+    df = pd.DataFrame(np.random.randn(3, 3), columns=["A", "B", "C"])
+    df.drop(["A", "X"], axis=1, errors="ignore")
+
+- Add support for separating years and quarters using dashes, for
+  example 2014-Q1.  (:issue:`9688`)
+
+- Allow conversion of values with dtype ``datetime64`` or ``timedelta64`` to strings using ``astype(str)`` (:issue:`9757`)
+- ``get_dummies`` function now accepts ``sparse`` keyword.  If set to ``True``, the return ``DataFrame`` is sparse, e.g. ``SparseDataFrame``. (:issue:`8823`)
+- ``Period`` now accepts ``datetime64`` as value input. (:issue:`9054`)
+
+- Allow timedelta string conversion when leading zero is missing from time definition, ie ``0:00:00`` vs ``00:00:00``. (:issue:`9570`)
+- Allow ``Panel.shift`` with ``axis='items'`` (:issue:`9890`)
+
+- Trying to write an excel file now raises ``NotImplementedError`` if the ``DataFrame`` has a ``MultiIndex`` instead of writing a broken Excel file. (:issue:`9794`)
+- Allow ``Categorical.add_categories`` to accept ``Series`` or ``np.array``. (:issue:`9927`)
+
+- Add/delete ``str/dt/cat`` accessors dynamically from ``__dir__``. (:issue:`9910`)
+- Add ``normalize`` as a ``dt`` accessor method. (:issue:`10047`)
+
+- ``DataFrame`` and ``Series`` now have ``_constructor_expanddim`` property as overridable constructor for one higher dimensionality data. This should be used only when it is really needed, see :ref:`here <extending.subclassing-pandas>`
+
+- ``pd.lib.infer_dtype`` now returns ``'bytes'`` in Python 3 where appropriate. (:issue:`10032`)
+
+
+.. _whatsnew_0161.api:
+
+API changes
+~~~~~~~~~~~
+
+- When passing in an ax to ``df.plot( ..., ax=ax)``, the ``sharex`` kwarg will now default to ``False``.
+  The result is that the visibility of xlabels and xticklabels will not anymore be changed. You
+  have to do that by yourself for the right axes in your figure or set ``sharex=True`` explicitly
+  (but this changes the visible for all axes in the figure, not only the one which is passed in!).
+  If pandas creates the subplots itself (e.g. no passed in ``ax`` kwarg), then the
+  default is still ``sharex=True`` and the visibility changes are applied.
+
+- :meth:`~pandas.DataFrame.assign` now inserts new columns in alphabetical order. Previously
+  the order was arbitrary. (:issue:`9777`)
+
+- By default, ``read_csv`` and ``read_table`` will now try to infer the compression type based on the file extension. Set ``compression=None`` to restore the previous behavior (no decompression). (:issue:`9770`)
+
+.. _whatsnew_0161.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- ``Series.str.split``'s ``return_type`` keyword was removed in favor of ``expand`` (:issue:`9847`)
+
+
+.. _whatsnew_0161.index_repr:
+
+Index representation
+~~~~~~~~~~~~~~~~~~~~
+
+The string representation of ``Index`` and its sub-classes have now been unified. These will show a single-line display if there are few values; a wrapped multi-line display for a lot of values (but less than ``display.max_seq_items``; if lots of items (> ``display.max_seq_items``) will show a truncated display (the head and tail of the data). The formatting for ``MultiIndex`` is unchanged (a multi-line wrapped display). The display width responds to the option ``display.max_seq_items``, which is defaulted to 100. (:issue:`6482`)
+
+Previous behavior
+
+.. code-block:: ipython
+
+   In [2]: pd.Index(range(4), name='foo')
+   Out[2]: Int64Index([0, 1, 2, 3], dtype='int64')
+
+   In [3]: pd.Index(range(104), name='foo')
+   Out[3]: Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, ...], dtype='int64')
+
+   In [4]: pd.date_range('20130101', periods=4, name='foo', tz='US/Eastern')
+   Out[4]:
+   <class 'pandas.tseries.index.DatetimeIndex'>
+   [2013-01-01 00:00:00-05:00, ..., 2013-01-04 00:00:00-05:00]
+   Length: 4, Freq: D, Timezone: US/Eastern
+
+   In [5]: pd.date_range('20130101', periods=104, name='foo', tz='US/Eastern')
+   Out[5]:
+   <class 'pandas.tseries.index.DatetimeIndex'>
+   [2013-01-01 00:00:00-05:00, ..., 2013-04-14 00:00:00-04:00]
+   Length: 104, Freq: D, Timezone: US/Eastern
+
+New behavior
+
+.. ipython:: python
+
+   pd.set_option("display.width", 80)
+   pd.Index(range(4), name="foo")
+   pd.Index(range(30), name="foo")
+   pd.Index(range(104), name="foo")
+   pd.CategoricalIndex(["a", "bb", "ccc", "dddd"], ordered=True, name="foobar")
+   pd.CategoricalIndex(["a", "bb", "ccc", "dddd"] * 10, ordered=True, name="foobar")
+   pd.CategoricalIndex(["a", "bb", "ccc", "dddd"] * 100, ordered=True, name="foobar")
+   pd.date_range("20130101", periods=4, name="foo", tz="US/Eastern")
+   pd.date_range("20130101", periods=25, freq="D")
+   pd.date_range("20130101", periods=104, name="foo", tz="US/Eastern")
+
+
+.. _whatsnew_0161.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved csv write performance with mixed dtypes, including datetimes by up to 5x (:issue:`9940`)
+- Improved csv write performance generally by 2x (:issue:`9940`)
+- Improved the performance of ``pd.lib.max_len_string_array`` by 5-7x (:issue:`10024`)
+
+
+.. _whatsnew_0161.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug where labels did not appear properly in the legend of ``DataFrame.plot()``, passing ``label=`` arguments works, and Series indices are no longer mutated. (:issue:`9542`)
+- Bug in json serialization causing a segfault when a frame had zero length. (:issue:`9805`)
+- Bug in ``read_csv`` where missing trailing delimiters would cause segfault. (:issue:`5664`)
+- Bug in retaining index name on appending (:issue:`9862`)
+- Bug in ``scatter_matrix`` draws unexpected axis ticklabels (:issue:`5662`)
+- Fixed bug in ``StataWriter`` resulting in changes to input ``DataFrame`` upon save (:issue:`9795`).
+- Bug in ``transform`` causing length mismatch when null entries were present and a fast aggregator was being used (:issue:`9697`)
+- Bug in ``equals`` causing false negatives when block order differed (:issue:`9330`)
+- Bug in grouping with multiple ``pd.Grouper`` where one is non-time based (:issue:`10063`)
+- Bug in ``read_sql_table`` error when reading postgres table with timezone (:issue:`7139`)
+- Bug in ``DataFrame`` slicing may not retain metadata (:issue:`9776`)
+- Bug where ``TimdeltaIndex`` were not properly serialized in fixed ``HDFStore`` (:issue:`9635`)
+- Bug with ``TimedeltaIndex`` constructor ignoring ``name`` when given another ``TimedeltaIndex`` as data (:issue:`10025`).
+- Bug in ``DataFrameFormatter._get_formatted_index`` with not applying ``max_colwidth`` to the ``DataFrame`` index (:issue:`7856`)
+- Bug in ``.loc`` with a read-only ndarray data source (:issue:`10043`)
+- Bug in ``groupby.apply()`` that would raise if a passed user defined function either returned only ``None`` (for all input). (:issue:`9685`)
+- Always use temporary files in pytables tests (:issue:`9992`)
+- Bug in plotting continuously using ``secondary_y`` may not show legend properly. (:issue:`9610`, :issue:`9779`)
+- Bug in ``DataFrame.plot(kind="hist")`` results in ``TypeError`` when ``DataFrame`` contains non-numeric columns  (:issue:`9853`)
+- Bug where repeated plotting of ``DataFrame`` with a ``DatetimeIndex`` may raise ``TypeError`` (:issue:`9852`)
+- Bug in ``setup.py`` that would allow an incompat cython version to build (:issue:`9827`)
+- Bug in plotting ``secondary_y`` incorrectly attaches ``right_ax`` property to secondary axes specifying itself recursively. (:issue:`9861`)
+- Bug in ``Series.quantile`` on empty Series of type ``Datetime`` or ``Timedelta`` (:issue:`9675`)
+- Bug in ``where`` causing incorrect results when upcasting was required (:issue:`9731`)
+- Bug in ``FloatArrayFormatter`` where decision boundary for displaying "small" floats in decimal format is off by one order of magnitude for a given display.precision (:issue:`9764`)
+- Fixed bug where ``DataFrame.plot()`` raised an error when both ``color`` and ``style`` keywords were passed and there was no color symbol in the style strings (:issue:`9671`)
+- Not showing a ``DeprecationWarning`` on combining list-likes with an ``Index`` (:issue:`10083`)
+- Bug in ``read_csv`` and ``read_table`` when using ``skip_rows`` parameter if blank lines are present. (:issue:`9832`)
+- Bug in ``read_csv()`` interprets ``index_col=True`` as ``1`` (:issue:`9798`)
+- Bug in index equality comparisons using ``==`` failing on Index/MultiIndex type incompatibility (:issue:`9785`)
+- Bug in which ``SparseDataFrame`` could not take ``nan`` as a column name (:issue:`8822`)
+- Bug in ``to_msgpack`` and ``read_msgpack`` zlib and blosc compression support (:issue:`9783`)
+- Bug ``GroupBy.size`` doesn't attach index name properly if grouped by ``TimeGrouper`` (:issue:`9925`)
+- Bug causing an exception in slice assignments because ``length_of_indexer`` returns wrong results (:issue:`9995`)
+- Bug in csv parser causing lines with initial white space plus one non-space character to be skipped. (:issue:`9710`)
+- Bug in C csv parser causing spurious NaNs when data started with newline followed by white space. (:issue:`10022`)
+- Bug causing elements with a null group to spill into the final group when grouping by a ``Categorical`` (:issue:`9603`)
+- Bug where .iloc and .loc behavior is not consistent on empty dataframes (:issue:`9964`)
+- Bug in invalid attribute access on a ``TimedeltaIndex`` incorrectly raised ``ValueError`` instead of ``AttributeError`` (:issue:`9680`)
+- Bug in unequal comparisons between categorical data and a scalar, which was not in the categories (e.g. ``Series(Categorical(list("abc"), ordered=True)) > "d"``. This returned ``False`` for all elements, but now raises a ``TypeError``. Equality comparisons also now return ``False`` for ``==`` and ``True`` for ``!=``. (:issue:`9848`)
+- Bug in DataFrame ``__setitem__`` when right hand side is a dictionary (:issue:`9874`)
+- Bug in ``where`` when dtype is ``datetime64/timedelta64``, but dtype of other is not (:issue:`9804`)
+- Bug in ``MultiIndex.sortlevel()`` results in unicode level name breaks (:issue:`9856`)
+- Bug in which ``groupby.transform`` incorrectly enforced output dtypes to match input dtypes. (:issue:`9807`)
+- Bug in ``DataFrame`` constructor when ``columns`` parameter is set, and ``data`` is an empty list (:issue:`9939`)
+- Bug in bar plot with ``log=True`` raises ``TypeError`` if all values are less than 1 (:issue:`9905`)
+- Bug in horizontal bar plot ignores ``log=True`` (:issue:`9905`)
+- Bug in PyTables queries that did not return proper results using the index (:issue:`8265`, :issue:`9676`)
+- Bug where dividing a dataframe containing values of type ``Decimal`` by another ``Decimal`` would raise. (:issue:`9787`)
+- Bug where using DataFrames asfreq would remove the name of the index. (:issue:`9885`)
+- Bug causing extra index point when resample BM/BQ (:issue:`9756`)
+- Changed caching in ``AbstractHolidayCalendar`` to be at the instance level rather than at the class level as the latter can result in unexpected behaviour. (:issue:`9552`)
+- Fixed latex output for MultiIndexed dataframes (:issue:`9778`)
+- Bug causing an exception when setting an empty range using ``DataFrame.loc`` (:issue:`9596`)
+- Bug in hiding ticklabels with subplots and shared axes when adding a new plot to an existing grid of axes (:issue:`9158`)
+- Bug in ``transform`` and ``filter`` when grouping on a categorical variable (:issue:`9921`)
+- Bug in ``transform`` when groups are equal in number and dtype to the input index (:issue:`9700`)
+- Google BigQuery connector now imports dependencies on a per-method basis.(:issue:`9713`)
+- Updated BigQuery connector to no longer use deprecated ``oauth2client.tools.run()`` (:issue:`8327`)
+- Bug in subclassed ``DataFrame``. It may not return the correct class, when slicing or subsetting it. (:issue:`9632`)
+- Bug in ``.median()`` where non-float null values are not handled correctly (:issue:`10040`)
+- Bug in Series.fillna() where it raises if a numerically convertible string is given (:issue:`10092`)
+
+
+.. _whatsnew_0.16.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.16.0..v0.16.1
diff --git a/pandas/doc/source/whatsnew/v0.16.2.rst b/pandas/doc/source/whatsnew/v0.16.2.rst
new file mode 100644
index 0000000000000..ef73c4b092fc1
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.16.2.rst
@@ -0,0 +1,215 @@
+.. _whatsnew_0162:
+
+Version 0.16.2 (June 12, 2015)
+------------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release from 0.16.1 and includes a large number of
+bug fixes along some new features (:meth:`~DataFrame.pipe` method), enhancements, and performance improvements.
+
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- A new ``pipe`` method, see :ref:`here <whatsnew_0162.enhancements.pipe>`
+- Documentation on how to use numba_ with *pandas*, see :ref:`here <enhancingperf.numba>`
+
+
+.. contents:: What's new in v0.16.2
+    :local:
+    :backlinks: none
+
+.. _numba: http://numba.pydata.org
+
+.. _whatsnew_0162.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0162.enhancements.pipe:
+
+Pipe
+^^^^
+
+We've introduced a new method :meth:`DataFrame.pipe`. As suggested by the name, ``pipe``
+should be used to pipe data through a chain of function calls.
+The goal is to avoid confusing nested function calls like
+
+.. code-block:: python
+
+   # df is a DataFrame
+   # f, g, and h are functions that take and return DataFrames
+   f(g(h(df), arg1=1), arg2=2, arg3=3)  # noqa F821
+
+The logic flows from inside out, and function names are separated from their keyword arguments.
+This can be rewritten as
+
+.. code-block:: python
+
+   (
+       df.pipe(h)  # noqa F821
+       .pipe(g, arg1=1)  # noqa F821
+       .pipe(f, arg2=2, arg3=3)  # noqa F821
+   )
+
+Now both the code and the logic flow from top to bottom. Keyword arguments are next to
+their functions. Overall the code is much more readable.
+
+In the example above, the functions ``f``, ``g``, and ``h`` each expected the DataFrame as the first positional argument.
+When the function you wish to apply takes its data anywhere other than the first argument, pass a tuple
+of ``(function, keyword)`` indicating where the DataFrame should flow. For example:
+
+.. code-block:: ipython
+
+    In [1]: import statsmodels.formula.api as sm
+
+    In [2]: bb = pd.read_csv("data/baseball.csv", index_col="id")
+
+    # sm.ols takes (formula, data)
+    In [3]: (
+    ...:     bb.query("h > 0")
+    ...:     .assign(ln_h=lambda df: np.log(df.h))
+    ...:     .pipe((sm.ols, "data"), "hr ~ ln_h + year + g + C(lg)")
+    ...:     .fit()
+    ...:     .summary()
+    ...: )
+    ...:
+    Out[3]:
+    <class 'statsmodels.iolib.summary.Summary'>
+    """
+                                OLS Regression Results
+    ==============================================================================
+    Dep. Variable:                     hr   R-squared:                       0.685
+    Model:                            OLS   Adj. R-squared:                  0.665
+    Method:                 Least Squares   F-statistic:                     34.28
+    Date:                Tue, 22 Nov 2022   Prob (F-statistic):           3.48e-15
+    Time:                        05:35:23   Log-Likelihood:                -205.92
+    No. Observations:                  68   AIC:                             421.8
+    Df Residuals:                      63   BIC:                             432.9
+    Df Model:                           4
+    Covariance Type:            nonrobust
+    ===============================================================================
+                    coef    std err          t      P>|t|      [0.025      0.975]
+    -------------------------------------------------------------------------------
+    Intercept   -8484.7720   4664.146     -1.819      0.074   -1.78e+04     835.780
+    C(lg)[T.NL]    -2.2736      1.325     -1.716      0.091      -4.922       0.375
+    ln_h           -1.3542      0.875     -1.547      0.127      -3.103       0.395
+    year            4.2277      2.324      1.819      0.074      -0.417       8.872
+    g               0.1841      0.029      6.258      0.000       0.125       0.243
+    ==============================================================================
+    Omnibus:                       10.875   Durbin-Watson:                   1.999
+    Prob(Omnibus):                  0.004   Jarque-Bera (JB):               17.298
+    Skew:                           0.537   Prob(JB):                     0.000175
+    Kurtosis:                       5.225   Cond. No.                     1.49e+07
+    ==============================================================================
+
+    Notes:
+    [1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+    [2] The condition number is large, 1.49e+07. This might indicate that there are
+    strong multicollinearity or other numerical problems.
+    """
+
+The pipe method is inspired by unix pipes, which stream text through
+processes. More recently dplyr_ and magrittr_ have introduced the
+popular ``(%>%)`` pipe operator for R_.
+
+See the :ref:`documentation <basics.pipe>` for more. (:issue:`10129`)
+
+.. _dplyr: https://github.com/tidyverse/dplyr
+.. _magrittr: https://github.com/smbache/magrittr
+.. _R: http://www.r-project.org
+
+.. _whatsnew_0162.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Added ``rsplit`` to Index/Series StringMethods (:issue:`10303`)
+
+- Removed the hard-coded size limits on the ``DataFrame`` HTML representation
+  in the IPython notebook, and leave this to IPython itself (only for IPython
+  v3.0 or greater). This eliminates the duplicate scroll bars that appeared in
+  the notebook with large frames (:issue:`10231`).
+
+  Note that the notebook has a ``toggle output scrolling`` feature to limit the
+  display of very large frames (by clicking left of the output).
+  You can also configure the way DataFrames are displayed using the pandas
+  options, see here :ref:`here <options.frequently_used>`.
+
+- ``axis`` parameter of ``DataFrame.quantile`` now accepts also ``index`` and ``column``. (:issue:`9543`)
+
+.. _whatsnew_0162.api:
+
+API changes
+~~~~~~~~~~~
+
+- ``Holiday`` now raises ``NotImplementedError`` if both ``offset`` and ``observance`` are used in the constructor instead of returning an incorrect result (:issue:`10217`).
+
+
+.. _whatsnew_0162.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved ``Series.resample`` performance with ``dtype=datetime64[ns]`` (:issue:`7754`)
+- Increase performance of ``str.split`` when ``expand=True`` (:issue:`10081`)
+
+.. _whatsnew_0162.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in ``Series.hist`` raises an error when a one row ``Series`` was given (:issue:`10214`)
+- Bug where ``HDFStore.select`` modifies the passed columns list (:issue:`7212`)
+- Bug in ``Categorical`` repr with ``display.width`` of ``None`` in Python 3 (:issue:`10087`)
+- Bug in ``to_json`` with certain orients and a ``CategoricalIndex`` would segfault (:issue:`10317`)
+- Bug where some of the nan functions do not have consistent return dtypes (:issue:`10251`)
+- Bug in ``DataFrame.quantile`` on checking that a valid axis was passed (:issue:`9543`)
+- Bug in ``groupby.apply`` aggregation for ``Categorical`` not preserving categories (:issue:`10138`)
+- Bug in ``to_csv`` where ``date_format`` is ignored if the ``datetime`` is fractional (:issue:`10209`)
+- Bug in ``DataFrame.to_json`` with mixed data types (:issue:`10289`)
+- Bug in cache updating when consolidating (:issue:`10264`)
+- Bug in ``mean()`` where integer dtypes can overflow (:issue:`10172`)
+- Bug where ``Panel.from_dict`` does not set dtype when specified (:issue:`10058`)
+- Bug in ``Index.union`` raises ``AttributeError`` when passing array-likes. (:issue:`10149`)
+- Bug in ``Timestamp``'s' ``microsecond``, ``quarter``, ``dayofyear``, ``week`` and ``daysinmonth`` properties return ``np.int`` type, not built-in ``int``. (:issue:`10050`)
+- Bug in ``NaT`` raises ``AttributeError`` when accessing to ``daysinmonth``, ``dayofweek`` properties. (:issue:`10096`)
+- Bug in Index repr when using the ``max_seq_items=None`` setting (:issue:`10182`).
+- Bug in getting timezone data with ``dateutil`` on various platforms ( :issue:`9059`, :issue:`8639`, :issue:`9663`, :issue:`10121`)
+- Bug in displaying datetimes with mixed frequencies; display 'ms' datetimes to the proper precision. (:issue:`10170`)
+- Bug in ``setitem`` where type promotion is applied to the entire block (:issue:`10280`)
+- Bug in ``Series`` arithmetic methods may incorrectly hold names (:issue:`10068`)
+- Bug in ``GroupBy.get_group`` when grouping on multiple keys, one of which is categorical. (:issue:`10132`)
+- Bug in ``DatetimeIndex`` and ``TimedeltaIndex`` names are lost after timedelta arithmetic ( :issue:`9926`)
+- Bug in ``DataFrame`` construction from nested ``dict`` with ``datetime64`` (:issue:`10160`)
+- Bug in ``Series`` construction from ``dict`` with ``datetime64`` keys (:issue:`9456`)
+- Bug in ``Series.plot(label="LABEL")`` not correctly setting the label (:issue:`10119`)
+- Bug in ``plot`` not defaulting to matplotlib ``axes.grid`` setting (:issue:`9792`)
+- Bug causing strings containing an exponent, but no decimal to be parsed as ``int`` instead of ``float`` in ``engine='python'`` for the ``read_csv`` parser (:issue:`9565`)
+- Bug in ``Series.align`` resets ``name`` when ``fill_value`` is specified (:issue:`10067`)
+- Bug in ``read_csv`` causing index name not to be set on an empty DataFrame (:issue:`10184`)
+- Bug in ``SparseSeries.abs`` resets ``name`` (:issue:`10241`)
+- Bug in ``TimedeltaIndex`` slicing may reset freq (:issue:`10292`)
+- Bug in ``GroupBy.get_group`` raises ``ValueError`` when group key contains ``NaT`` (:issue:`6992`)
+- Bug in ``SparseSeries`` constructor ignores input data name (:issue:`10258`)
+- Bug in ``Categorical.remove_categories`` causing a ``ValueError`` when removing the ``NaN`` category if underlying dtype is floating-point (:issue:`10156`)
+- Bug where infer_freq infers time rule (WOM-5XXX) unsupported by to_offset (:issue:`9425`)
+- Bug in ``DataFrame.to_hdf()`` where table format would raise a seemingly unrelated error for invalid (non-string) column names. This is now explicitly forbidden. (:issue:`9057`)
+- Bug to handle masking empty ``DataFrame`` (:issue:`10126`).
+- Bug where MySQL interface could not handle numeric table/column names (:issue:`10255`)
+- Bug in ``read_csv`` with a ``date_parser`` that returned a ``datetime64`` array of other time resolution than ``[ns]`` (:issue:`10245`)
+- Bug in ``Panel.apply`` when the result has ndim=0 (:issue:`10332`)
+- Bug in ``read_hdf`` where ``auto_close`` could not be passed (:issue:`9327`).
+- Bug in ``read_hdf`` where open stores could not be used (:issue:`10330`).
+- Bug in adding empty ``DataFrames``, now results in a ``DataFrame`` that ``.equals`` an empty ``DataFrame`` (:issue:`10181`).
+- Bug in ``to_hdf`` and ``HDFStore`` which did not check that complib choices were valid (:issue:`4582`, :issue:`8874`).
+
+
+.. _whatsnew_0.16.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.16.1..v0.16.2
diff --git a/pandas/doc/source/whatsnew/v0.17.0.rst b/pandas/doc/source/whatsnew/v0.17.0.rst
new file mode 100644
index 0000000000000..fb71ec60a22f0
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.17.0.rst
@@ -0,0 +1,1178 @@
+.. _whatsnew_0170:
+
+Version 0.17.0 (October 9, 2015)
+--------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.16.2 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+.. warning::
+
+   pandas >= 0.17.0 will no longer support compatibility with Python version 3.2 (:issue:`9118`)
+
+.. warning::
+
+   The ``pandas.io.data`` package is deprecated and will be replaced by the
+   `pandas-datareader package <https://github.com/pydata/pandas-datareader>`_.
+   This will allow the data modules to be independently updated to your pandas
+   installation. The API for ``pandas-datareader v0.1.1`` is exactly the same
+   as in ``pandas v0.17.0`` (:issue:`8961`, :issue:`10861`).
+
+   After installing pandas-datareader, you can easily change your imports:
+
+   .. code-block:: python
+
+     from pandas.io import data, wb
+
+   becomes
+
+   .. code-block:: python
+
+     from pandas_datareader import data, wb
+
+Highlights include:
+
+- Release the Global Interpreter Lock (GIL) on some cython operations, see :ref:`here <whatsnew_0170.gil>`
+- Plotting methods are now available as attributes of the ``.plot`` accessor, see :ref:`here <whatsnew_0170.plot>`
+- The sorting API has been revamped to remove some long-time inconsistencies, see :ref:`here <whatsnew_0170.api_breaking.sorting>`
+- Support for a ``datetime64[ns]`` with timezones as a first-class dtype, see :ref:`here <whatsnew_0170.tz>`
+- The default for ``to_datetime`` will now be to ``raise`` when presented with unparsable formats,
+  previously this would return the original input. Also, date parse
+  functions now return consistent results. See :ref:`here <whatsnew_0170.api_breaking.to_datetime>`
+- The default for ``dropna`` in ``HDFStore`` has changed to ``False``, to store by default all rows even
+  if they are all ``NaN``, see :ref:`here <whatsnew_0170.api_breaking.hdf_dropna>`
+- Datetime accessor (``dt``) now supports ``Series.dt.strftime`` to generate formatted strings for datetime-likes, and ``Series.dt.total_seconds`` to generate each duration of the timedelta in seconds. See :ref:`here <whatsnew_0170.strftime>`
+- ``Period`` and ``PeriodIndex`` can handle multiplied freq like ``3D``, which corresponding to 3 days span. See :ref:`here <whatsnew_0170.periodfreq>`
+- Development installed versions of pandas will now have ``PEP440`` compliant version strings (:issue:`9518`)
+- Development support for benchmarking with the `Air Speed Velocity library <https://github.com/spacetelescope/asv/>`_ (:issue:`8361`)
+- Support for reading SAS xport files, see :ref:`here <whatsnew_0170.enhancements.sas_xport>`
+- Documentation comparing SAS to *pandas*, see :ref:`here <compare_with_sas>`
+- Removal of the automatic TimeSeries broadcasting, deprecated since 0.8.0, see :ref:`here <whatsnew_0170.prior_deprecations>`
+- Display format with plain text can optionally align with Unicode East Asian Width, see :ref:`here <whatsnew_0170.east_asian_width>`
+- Compatibility with Python 3.5 (:issue:`11097`)
+- Compatibility with matplotlib 1.5.0 (:issue:`11111`)
+
+Check the :ref:`API Changes <whatsnew_0170.api>` and :ref:`deprecations <whatsnew_0170.deprecations>` before updating.
+
+.. contents:: What's new in v0.17.0
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0170.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0170.tz:
+
+Datetime with TZ
+^^^^^^^^^^^^^^^^
+
+We are adding an implementation that natively supports datetime with timezones. A ``Series`` or a ``DataFrame`` column previously
+*could* be assigned a datetime with timezones, and would work as an ``object`` dtype. This had performance issues with a large
+number rows. See the :ref:`docs <timeseries.timezone_series>` for more details. (:issue:`8260`, :issue:`10763`, :issue:`11034`).
+
+The new implementation allows for having a single-timezone across all rows, with operations in a performant manner.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "A": pd.date_range("20130101", periods=3),
+           "B": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+           "C": pd.date_range("20130101", periods=3, tz="CET"),
+       }
+   )
+   df
+   df.dtypes
+
+.. ipython:: python
+
+   df.B
+   df.B.dt.tz_localize(None)
+
+This uses a new-dtype representation as well, that is very similar in look-and-feel to its numpy cousin ``datetime64[ns]``
+
+.. ipython:: python
+
+   df["B"].dtype
+   type(df["B"].dtype)
+
+.. note::
+
+   There is a slightly different string repr for the underlying ``DatetimeIndex`` as a result of the dtype changes, but
+   functionally these are the same.
+
+   Previous behavior:
+
+   .. code-block:: ipython
+
+      In [1]: pd.date_range('20130101', periods=3, tz='US/Eastern')
+      Out[1]: DatetimeIndex(['2013-01-01 00:00:00-05:00', '2013-01-02 00:00:00-05:00',
+                             '2013-01-03 00:00:00-05:00'],
+                            dtype='datetime64[ns]', freq='D', tz='US/Eastern')
+
+      In [2]: pd.date_range('20130101', periods=3, tz='US/Eastern').dtype
+      Out[2]: dtype('<M8[ns]')
+
+   New behavior:
+
+   .. ipython:: python
+
+      pd.date_range("20130101", periods=3, tz="US/Eastern")
+      pd.date_range("20130101", periods=3, tz="US/Eastern").dtype
+
+.. _whatsnew_0170.gil:
+
+Releasing the GIL
+^^^^^^^^^^^^^^^^^
+
+We are releasing the global-interpreter-lock (GIL) on some cython operations.
+This will allow other threads to run simultaneously during computation, potentially allowing performance improvements
+from multi-threading. Notably ``groupby``, ``nsmallest``, ``value_counts`` and some indexing operations benefit from this. (:issue:`8882`)
+
+For example the groupby expression in the following code will have the GIL released during the factorization step, e.g. ``df.groupby('key')``
+as well as the ``.sum()`` operation.
+
+.. code-block:: python
+
+   N = 1000000
+   ngroups = 10
+   df = DataFrame(
+       {"key": np.random.randint(0, ngroups, size=N), "data": np.random.randn(N)}
+   )
+   df.groupby("key")["data"].sum()
+
+Releasing of the GIL could benefit an application that uses threads for user interactions (e.g. QT_), or performing multi-threaded computations. A nice example of a library that can handle these types of computation-in-parallel is the dask_ library.
+
+.. _dask: https://dask.readthedocs.io/en/latest/
+.. _QT: https://wiki.python.org/moin/PyQt
+
+.. _whatsnew_0170.plot:
+
+Plot submethods
+^^^^^^^^^^^^^^^
+
+The Series and DataFrame ``.plot()`` method allows for customizing :ref:`plot types<visualization.other>` by supplying the ``kind`` keyword arguments. Unfortunately, many of these kinds of plots use different required and optional keyword arguments, which makes it difficult to discover what any given plot kind uses out of the dozens of possible arguments.
+
+To alleviate this issue, we have added a new, optional plotting interface, which exposes each kind of plot as a method of the ``.plot`` attribute. Instead of writing ``series.plot(kind=<kind>, ...)``, you can now also use ``series.plot.<kind>(...)``:
+
+.. ipython::
+    :verbatim:
+
+    In [13]: df = pd.DataFrame(np.random.rand(10, 2), columns=['a', 'b'])
+
+    In [14]: df.plot.bar()
+
+.. image:: ../_static/whatsnew_plot_submethods.png
+
+As a result of this change, these methods are now all discoverable via tab-completion:
+
+.. ipython::
+    :verbatim:
+
+    In [15]: df.plot.<TAB>  # noqa: E225, E999
+    df.plot.area     df.plot.barh     df.plot.density  df.plot.hist     df.plot.line     df.plot.scatter
+    df.plot.bar      df.plot.box      df.plot.hexbin   df.plot.kde      df.plot.pie
+
+Each method signature only includes relevant arguments. Currently, these are limited to required arguments, but in the future these will include optional arguments, as well. For an overview, see the new :ref:`api.dataframe.plotting` API documentation.
+
+.. _whatsnew_0170.strftime:
+
+Additional methods for ``dt`` accessor
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Series.dt.strftime
+""""""""""""""""""
+
+We are now supporting a ``Series.dt.strftime`` method for datetime-likes to generate a formatted string (:issue:`10110`). Examples:
+
+.. ipython:: python
+
+   # DatetimeIndex
+   s = pd.Series(pd.date_range("20130101", periods=4))
+   s
+   s.dt.strftime("%Y/%m/%d")
+
+.. ipython:: python
+
+   # PeriodIndex
+   s = pd.Series(pd.period_range("20130101", periods=4))
+   s
+   s.dt.strftime("%Y/%m/%d")
+
+The string format is as the python standard library and details can be found `here <https://docs.python.org/2/library/datetime.html#strftime-and-strptime-behavior>`_
+
+Series.dt.total_seconds
+"""""""""""""""""""""""
+
+``pd.Series`` of type ``timedelta64`` has new method ``.dt.total_seconds()`` returning the duration of the timedelta in seconds (:issue:`10817`)
+
+.. ipython:: python
+
+   # TimedeltaIndex
+   s = pd.Series(pd.timedelta_range("1 minutes", periods=4))
+   s
+   s.dt.total_seconds()
+
+.. _whatsnew_0170.periodfreq:
+
+Period frequency enhancement
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Period``, ``PeriodIndex`` and ``period_range`` can now accept multiplied freq. Also, ``Period.freq`` and ``PeriodIndex.freq`` are now stored as a ``DateOffset`` instance like ``DatetimeIndex``, and not as ``str`` (:issue:`7811`)
+
+A multiplied freq represents a span of corresponding length. The example below creates a period of 3 days. Addition and subtraction will shift the period by its span.
+
+.. ipython:: python
+
+   p = pd.Period("2015-08-01", freq="3D")
+   p
+   p + 1
+   p - 2
+   p.to_timestamp()
+   p.to_timestamp(how="E")
+
+You can use the multiplied freq in ``PeriodIndex`` and ``period_range``.
+
+.. ipython:: python
+
+   idx = pd.period_range("2015-08-01", periods=4, freq="2D")
+   idx
+   idx + 1
+
+.. _whatsnew_0170.enhancements.sas_xport:
+
+Support for SAS XPORT files
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`~pandas.io.read_sas` provides support for reading *SAS XPORT* format files. (:issue:`4052`).
+
+.. code-block:: python
+
+    df = pd.read_sas("sas_xport.xpt")
+
+It is also possible to obtain an iterator and read an XPORT file
+incrementally.
+
+.. code-block:: python
+
+    for df in pd.read_sas("sas_xport.xpt", chunksize=10000):
+        do_something(df)
+
+See the :ref:`docs <io.sas>` for more details.
+
+.. _whatsnew_0170.matheval:
+
+Support for math functions in .eval()
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`~pandas.eval` now supports calling math functions (:issue:`4893`)
+
+.. code-block:: python
+
+    df = pd.DataFrame({"a": np.random.randn(10)})
+    df.eval("b = sin(a)")
+
+The support math functions are ``sin``, ``cos``, ``exp``, ``log``, ``expm1``, ``log1p``,
+``sqrt``, ``sinh``, ``cosh``, ``tanh``, ``arcsin``, ``arccos``, ``arctan``, ``arccosh``,
+``arcsinh``, ``arctanh``, ``abs`` and ``arctan2``.
+
+These functions map to the intrinsics for the ``NumExpr`` engine.  For the Python
+engine, they are mapped to ``NumPy`` calls.
+
+Changes to Excel with ``MultiIndex``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In version 0.16.2 a ``DataFrame`` with ``MultiIndex`` columns could not be written to Excel via ``to_excel``.
+That functionality has been added (:issue:`10564`), along with updating  ``read_excel`` so that the data can
+be read back with, no loss of information, by specifying which columns/rows make up the ``MultiIndex``
+in the ``header`` and ``index_col`` parameters (:issue:`4679`)
+
+See the :ref:`documentation <io.excel>` for more details.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       [[1, 2, 3, 4], [5, 6, 7, 8]],
+       columns=pd.MultiIndex.from_product(
+           [["foo", "bar"], ["a", "b"]], names=["col1", "col2"]
+       ),
+       index=pd.MultiIndex.from_product([["j"], ["l", "k"]], names=["i1", "i2"]),
+   )
+
+   df
+   df.to_excel("test.xlsx")
+
+   df = pd.read_excel("test.xlsx", header=[0, 1], index_col=[0, 1])
+   df
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("test.xlsx")
+
+Previously, it was necessary to specify the ``has_index_names`` argument in ``read_excel``,
+if the serialized data had index names.  For version 0.17.0 the output format of ``to_excel``
+has been changed to make this keyword unnecessary - the change is shown below.
+
+**Old**
+
+.. image:: ../_static/old-excel-index.png
+
+**New**
+
+.. image:: ../_static/new-excel-index.png
+
+.. warning::
+
+   Excel files saved in version 0.16.2 or prior that had index names will still able to be read in,
+   but the ``has_index_names`` argument must specified to ``True``.
+
+.. _whatsnew_0170.gbq:
+
+Google BigQuery enhancements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+- Added ability to automatically create a table/dataset using the :func:`pandas.io.gbq.to_gbq` function if the destination table/dataset does not exist. (:issue:`8325`, :issue:`11121`).
+- Added ability to replace an existing table and schema when calling the :func:`pandas.io.gbq.to_gbq` function via the ``if_exists`` argument. See the `docs <https://pandas-gbq.readthedocs.io/en/latest/writing.html>`__ for more details (:issue:`8325`).
+- ``InvalidColumnOrder`` and ``InvalidPageToken`` in the gbq module will raise ``ValueError`` instead of ``IOError``.
+- The ``generate_bq_schema()`` function is now deprecated and will be removed in a future version (:issue:`11121`)
+- The gbq module will now support Python 3 (:issue:`11094`).
+
+.. _whatsnew_0170.east_asian_width:
+
+Display alignment with Unicode East Asian width
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. warning::
+
+   Enabling this option will affect the performance for printing of ``DataFrame`` and ``Series`` (about 2 times slower).
+   Use only when it is actually required.
+
+Some East Asian countries use Unicode characters its width is corresponding to 2 alphabets. If a ``DataFrame`` or ``Series`` contains these characters, the default output cannot be aligned properly. The following options are added to enable precise handling for these characters.
+
+- ``display.unicode.east_asian_width``: Whether to use the Unicode East Asian Width to calculate the display text width. (:issue:`2612`)
+- ``display.unicode.ambiguous_as_wide``: Whether to handle Unicode characters belong to Ambiguous as Wide. (:issue:`11102`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({u"国籍": ["UK", u"日本"], u"名前": ["Alice", u"しのぶ"]})
+   df
+
+.. ipython:: python
+
+   pd.set_option("display.unicode.east_asian_width", True)
+   df
+
+For further details, see :ref:`here <options.east_asian_width>`
+
+.. ipython:: python
+   :suppress:
+
+   pd.set_option("display.unicode.east_asian_width", False)
+
+.. _whatsnew_0170.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Support for ``openpyxl`` >= 2.2. The API for style support is now stable (:issue:`10125`)
+- ``merge`` now accepts the argument ``indicator`` which adds a Categorical-type column (by default called ``_merge``) to the output object that takes on the values (:issue:`8790`)
+
+  ===================================   ================
+  Observation Origin                    ``_merge`` value
+  ===================================   ================
+  Merge key only in ``'left'`` frame    ``left_only``
+  Merge key only in ``'right'`` frame   ``right_only``
+  Merge key in both frames              ``both``
+  ===================================   ================
+
+  .. ipython:: python
+
+    df1 = pd.DataFrame({"col1": [0, 1], "col_left": ["a", "b"]})
+    df2 = pd.DataFrame({"col1": [1, 2, 2], "col_right": [2, 2, 2]})
+    pd.merge(df1, df2, on="col1", how="outer", indicator=True)
+
+  For more, see the :ref:`updated docs <merging.indicator>`
+
+- ``pd.to_numeric`` is a new function to coerce strings to numbers (possibly with coercion) (:issue:`11133`)
+
+- ``pd.merge`` will now allow duplicate column names if they are not merged upon (:issue:`10639`).
+
+- ``pd.pivot`` will now allow passing index as ``None`` (:issue:`3962`).
+
+- ``pd.concat`` will now use existing Series names if provided (:issue:`10698`).
+
+  .. ipython:: python
+
+     foo = pd.Series([1, 2], name="foo")
+     bar = pd.Series([1, 2])
+     baz = pd.Series([4, 5])
+
+  Previous behavior:
+
+  .. code-block:: ipython
+
+     In [1]: pd.concat([foo, bar, baz], axis=1)
+     Out[1]:
+           0  1  2
+        0  1  1  4
+        1  2  2  5
+
+  New behavior:
+
+  .. ipython:: python
+
+    pd.concat([foo, bar, baz], axis=1)
+
+- ``DataFrame`` has gained the ``nlargest`` and ``nsmallest`` methods (:issue:`10393`)
+
+- Add a ``limit_direction`` keyword argument that works with ``limit`` to enable ``interpolate`` to fill ``NaN`` values forward, backward, or both (:issue:`9218`, :issue:`10420`, :issue:`11115`)
+
+  .. ipython:: python
+
+     ser = pd.Series([np.nan, np.nan, 5, np.nan, np.nan, np.nan, 13])
+     ser.interpolate(limit=1, limit_direction="both")
+
+- Added a ``DataFrame.round`` method to round the values to a variable number of decimal places (:issue:`10568`).
+
+  .. ipython:: python
+
+     df = pd.DataFrame(
+         np.random.random([3, 3]),
+         columns=["A", "B", "C"],
+         index=["first", "second", "third"],
+     )
+     df
+     df.round(2)
+     df.round({"A": 0, "C": 2})
+
+- ``drop_duplicates`` and ``duplicated`` now accept a ``keep`` keyword to target first, last, and all duplicates. The ``take_last`` keyword is deprecated, see :ref:`here <whatsnew_0170.deprecations>` (:issue:`6511`, :issue:`8505`)
+
+  .. ipython:: python
+
+     s = pd.Series(["A", "B", "C", "A", "B", "D"])
+     s.drop_duplicates()
+     s.drop_duplicates(keep="last")
+     s.drop_duplicates(keep=False)
+
+- Reindex now has a ``tolerance`` argument that allows for finer control of :ref:`basics.limits_on_reindex_fill` (:issue:`10411`):
+
+  .. ipython:: python
+
+     df = pd.DataFrame({"x": range(5), "t": pd.date_range("2000-01-01", periods=5)})
+     df.reindex([0.1, 1.9, 3.5], method="nearest", tolerance=0.2)
+
+  When used on a ``DatetimeIndex``, ``TimedeltaIndex`` or ``PeriodIndex``, ``tolerance`` will coerced into a ``Timedelta`` if possible. This allows you to specify tolerance with a string:
+
+  .. ipython:: python
+
+     df = df.set_index("t")
+     df.reindex(pd.to_datetime(["1999-12-31"]), method="nearest", tolerance="1 day")
+
+  ``tolerance`` is also exposed by the lower level ``Index.get_indexer`` and ``Index.get_loc`` methods.
+
+- Added functionality to use the ``base`` argument when resampling a ``TimeDeltaIndex`` (:issue:`10530`)
+
+- ``DatetimeIndex`` can be instantiated using strings contains ``NaT`` (:issue:`7599`)
+
+- ``to_datetime`` can now accept the ``yearfirst`` keyword (:issue:`7599`)
+
+- ``pandas.tseries.offsets`` larger than the ``Day`` offset can now be used with a ``Series`` for addition/subtraction (:issue:`10699`).  See the :ref:`docs <timeseries.offsetseries>` for more details.
+
+- ``pd.Timedelta.total_seconds()`` now returns Timedelta duration to ns precision (previously microsecond precision) (:issue:`10939`)
+
+- ``PeriodIndex`` now supports arithmetic with ``np.ndarray`` (:issue:`10638`)
+
+- Support pickling of ``Period`` objects (:issue:`10439`)
+
+- ``.as_blocks`` will now take a ``copy`` optional argument to return a copy of the data, default is to copy (no change in behavior from prior versions), (:issue:`9607`)
+
+- ``regex`` argument to ``DataFrame.filter`` now handles numeric column names instead of raising ``ValueError`` (:issue:`10384`).
+
+- Enable reading gzip compressed files via URL, either by explicitly setting the compression parameter or by inferring from the presence of the HTTP Content-Encoding header in the response (:issue:`8685`)
+
+- Enable writing Excel files in :ref:`memory <io.excel_writing_buffer>` using StringIO/BytesIO (:issue:`7074`)
+
+- Enable serialization of lists and dicts to strings in ``ExcelWriter`` (:issue:`8188`)
+
+- SQL io functions now accept a SQLAlchemy connectable. (:issue:`7877`)
+
+- ``pd.read_sql`` and ``to_sql`` can accept database URI as ``con`` parameter (:issue:`10214`)
+
+- ``read_sql_table`` will now allow reading from views (:issue:`10750`).
+
+- Enable writing complex values to ``HDFStores`` when using the ``table`` format (:issue:`10447`)
+
+- Enable ``pd.read_hdf`` to be used without specifying a key when the HDF file contains a single dataset (:issue:`10443`)
+
+- ``pd.read_stata`` will now read Stata 118 type files. (:issue:`9882`)
+
+- ``msgpack`` submodule has been updated to 0.4.6 with backward compatibility (:issue:`10581`)
+
+- ``DataFrame.to_dict`` now accepts ``orient='index'`` keyword argument (:issue:`10844`).
+
+- ``DataFrame.apply`` will return a Series of dicts if the passed function returns a dict and ``reduce=True`` (:issue:`8735`).
+
+- Allow passing ``kwargs`` to the interpolation methods (:issue:`10378`).
+
+- Improved error message when concatenating an empty iterable of ``Dataframe`` objects (:issue:`9157`)
+
+- ``pd.read_csv`` can now read bz2-compressed files incrementally, and the C parser can read bz2-compressed files from AWS S3 (:issue:`11070`, :issue:`11072`).
+
+- In ``pd.read_csv``, recognize ``s3n://`` and ``s3a://`` URLs as designating S3 file storage (:issue:`11070`, :issue:`11071`).
+
+- Read CSV files from AWS S3 incrementally, instead of first downloading the entire file. (Full file download still required for compressed files in Python 2.)  (:issue:`11070`, :issue:`11073`)
+
+- ``pd.read_csv`` is now able to infer compression type for files read from AWS S3 storage (:issue:`11070`, :issue:`11074`).
+
+
+.. _whatsnew_0170.api:
+
+.. _whatsnew_0170.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0170.api_breaking.sorting:
+
+Changes to sorting API
+^^^^^^^^^^^^^^^^^^^^^^
+
+The sorting API has had some longtime inconsistencies. (:issue:`9816`, :issue:`8239`).
+
+Here is a summary of the API **PRIOR** to 0.17.0:
+
+- ``Series.sort`` is **INPLACE** while ``DataFrame.sort`` returns a new object.
+- ``Series.order`` returns a new object
+- It was possible to use ``Series/DataFrame.sort_index`` to sort by **values** by passing the ``by`` keyword.
+- ``Series/DataFrame.sortlevel`` worked only on a ``MultiIndex`` for sorting by index.
+
+To address these issues, we have revamped the API:
+
+- We have introduced a new method, :meth:`DataFrame.sort_values`, which is the merger of ``DataFrame.sort()``, ``Series.sort()``,
+  and ``Series.order()``, to handle sorting of **values**.
+- The existing methods ``Series.sort()``, ``Series.order()``, and ``DataFrame.sort()`` have been deprecated and will be removed in a
+  future version.
+- The ``by`` argument of ``DataFrame.sort_index()`` has been deprecated and will be removed in a future version.
+- The existing method ``.sort_index()`` will gain the ``level`` keyword to enable level sorting.
+
+We now have two distinct and non-overlapping methods of sorting. A ``*`` marks items that
+will show a ``FutureWarning``.
+
+To sort by the **values**:
+
+==================================    ====================================
+Previous                              Replacement
+==================================    ====================================
+\* ``Series.order()``                 ``Series.sort_values()``
+\* ``Series.sort()``                  ``Series.sort_values(inplace=True)``
+\* ``DataFrame.sort(columns=...)``    ``DataFrame.sort_values(by=...)``
+==================================    ====================================
+
+To sort by the **index**:
+
+==================================    ====================================
+Previous                              Replacement
+==================================    ====================================
+``Series.sort_index()``               ``Series.sort_index()``
+``Series.sortlevel(level=...)``       ``Series.sort_index(level=...``)
+``DataFrame.sort_index()``            ``DataFrame.sort_index()``
+``DataFrame.sortlevel(level=...)``    ``DataFrame.sort_index(level=...)``
+\* ``DataFrame.sort()``                 ``DataFrame.sort_index()``
+==================================    ====================================
+
+We have also deprecated and changed similar methods in two Series-like classes, ``Index`` and ``Categorical``.
+
+==================================    ====================================
+Previous                              Replacement
+==================================    ====================================
+\* ``Index.order()``                  ``Index.sort_values()``
+\* ``Categorical.order()``            ``Categorical.sort_values()``
+==================================    ====================================
+
+.. _whatsnew_0170.api_breaking.to_datetime:
+
+Changes to to_datetime and to_timedelta
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Error handling
+""""""""""""""
+
+The default for ``pd.to_datetime`` error handling has changed to ``errors='raise'``.
+In prior versions it was ``errors='ignore'``. Furthermore, the ``coerce`` argument
+has been deprecated in favor of ``errors='coerce'``. This means that invalid parsing
+will raise rather that return the original input as in previous versions. (:issue:`10636`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: pd.to_datetime(['2009-07-31', 'asd'])
+   Out[2]: array(['2009-07-31', 'asd'], dtype=object)
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [3]: pd.to_datetime(['2009-07-31', 'asd'])
+   ValueError: Unknown string format
+
+Of course you can coerce this as well.
+
+.. ipython:: python
+
+   pd.to_datetime(["2009-07-31", "asd"], errors="coerce")
+
+To keep the previous behavior, you can use ``errors='ignore'``:
+
+.. code-block:: ipython
+
+   In [4]: pd.to_datetime(["2009-07-31", "asd"], errors="ignore")
+   Out[4]: Index(['2009-07-31', 'asd'], dtype='object')
+
+Furthermore, ``pd.to_timedelta`` has gained a similar API, of ``errors='raise'|'ignore'|'coerce'``, and the ``coerce`` keyword
+has been deprecated in favor of ``errors='coerce'``.
+
+Consistent parsing
+""""""""""""""""""
+
+The string parsing of ``to_datetime``, ``Timestamp`` and ``DatetimeIndex`` has
+been made consistent. (:issue:`7599`)
+
+Prior to v0.17.0, ``Timestamp`` and ``to_datetime`` may parse year-only datetime-string incorrectly using today's date, otherwise ``DatetimeIndex``
+uses the beginning of the year. ``Timestamp`` and ``to_datetime`` may raise ``ValueError`` in some types of datetime-string which ``DatetimeIndex``
+can parse, such as a quarterly string.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [1]: pd.Timestamp('2012Q2')
+   Traceback
+      ...
+   ValueError: Unable to parse 2012Q2
+
+   # Results in today's date.
+   In [2]: pd.Timestamp('2014')
+   Out [2]: 2014-08-12 00:00:00
+
+v0.17.0 can parse them as below. It works on ``DatetimeIndex`` also.
+
+New behavior:
+
+.. ipython:: python
+
+   pd.Timestamp("2012Q2")
+   pd.Timestamp("2014")
+   pd.DatetimeIndex(["2012Q2", "2014"])
+
+.. note::
+
+   If you want to perform calculations based on today's date, use ``Timestamp.now()`` and ``pandas.tseries.offsets``.
+
+   .. ipython:: python
+
+      import pandas.tseries.offsets as offsets
+
+      pd.Timestamp.now()
+      pd.Timestamp.now() + offsets.DateOffset(years=1)
+
+Changes to Index comparisons
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Operator equal on ``Index`` should behavior similarly to ``Series`` (:issue:`9947`, :issue:`10637`)
+
+Starting in v0.17.0, comparing ``Index`` objects of different lengths will raise
+a ``ValueError``. This is to be consistent with the behavior of ``Series``.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: pd.Index([1, 2, 3]) == pd.Index([1, 4, 5])
+   Out[2]: array([ True, False, False], dtype=bool)
+
+   In [3]: pd.Index([1, 2, 3]) == pd.Index([2])
+   Out[3]: array([False,  True, False], dtype=bool)
+
+   In [4]: pd.Index([1, 2, 3]) == pd.Index([1, 2])
+   Out[4]: False
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [8]: pd.Index([1, 2, 3]) == pd.Index([1, 4, 5])
+   Out[8]: array([ True, False, False], dtype=bool)
+
+   In [9]: pd.Index([1, 2, 3]) == pd.Index([2])
+   ValueError: Lengths must match to compare
+
+   In [10]: pd.Index([1, 2, 3]) == pd.Index([1, 2])
+   ValueError: Lengths must match to compare
+
+Note that this is different from the ``numpy`` behavior where a comparison can
+be broadcast:
+
+.. ipython:: python
+
+   np.array([1, 2, 3]) == np.array([1])
+
+or it can return False if broadcasting can not be done:
+
+.. code-block:: ipython
+
+   In [11]: np.array([1, 2, 3]) == np.array([1, 2])
+   Out[11]: False
+
+Changes to boolean comparisons vs. None
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Boolean comparisons of a ``Series`` vs ``None`` will now be equivalent to comparing with ``np.nan``, rather than raise ``TypeError``. (:issue:`1079`).
+
+.. ipython:: python
+
+   s = pd.Series(range(3), dtype="float")
+   s.iloc[1] = None
+   s
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [5]: s == None
+   TypeError: Could not compare <type 'NoneType'> type with Series
+
+New behavior:
+
+.. ipython:: python
+
+   s == None
+
+Usually you simply want to know which values are null.
+
+.. ipython:: python
+
+   s.isnull()
+
+.. warning::
+
+   You generally will want to use ``isnull/notnull`` for these types of comparisons, as ``isnull/notnull`` tells you which elements are null. One has to be
+   mindful that ``nan's`` don't compare equal, but ``None's`` do. Note that pandas/numpy uses the fact that ``np.nan != np.nan``, and treats ``None`` like ``np.nan``.
+
+   .. ipython:: python
+
+      None == None
+      np.nan == np.nan
+
+.. _whatsnew_0170.api_breaking.hdf_dropna:
+
+HDFStore dropna behavior
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+The default behavior for HDFStore write functions with ``format='table'`` is now to keep rows that are all missing. Previously, the behavior was to drop rows that were all missing save the index. The previous behavior can be replicated using the ``dropna=True`` option. (:issue:`9382`)
+
+Previous behavior:
+
+.. ipython:: python
+
+   df_with_missing = pd.DataFrame(
+       {"col1": [0, np.nan, 2], "col2": [1, np.nan, np.nan]}
+   )
+
+   df_with_missing
+
+
+.. code-block:: ipython
+
+   In [27]:
+   df_with_missing.to_hdf('file.h5',
+                          key='df_with_missing',
+                          format='table',
+                          mode='w')
+
+   In [28]: pd.read_hdf('file.h5', 'df_with_missing')
+
+   Out [28]:
+         col1  col2
+     0     0     1
+     2     2   NaN
+
+
+New behavior:
+
+.. ipython:: python
+
+   df_with_missing.to_hdf("file.h5", key="df_with_missing", format="table", mode="w")
+
+   pd.read_hdf("file.h5", "df_with_missing")
+
+.. ipython:: python
+   :suppress:
+
+   import os
+
+   os.remove("file.h5")
+
+See the :ref:`docs <io.hdf5>` for more details.
+
+.. _whatsnew_0170.api_breaking.display_precision:
+
+Changes to ``display.precision`` option
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``display.precision`` option has been clarified to refer to decimal places (:issue:`10451`).
+
+Earlier versions of pandas would format floating point numbers to have one less decimal place than the value in
+``display.precision``.
+
+.. code-block:: ipython
+
+  In [1]: pd.set_option('display.precision', 2)
+
+  In [2]: pd.DataFrame({'x': [123.456789]})
+  Out[2]:
+         x
+  0  123.5
+
+If interpreting precision as "significant figures" this did work for scientific notation but that same interpretation
+did not work for values with standard formatting. It was also out of step with how numpy handles formatting.
+
+Going forward the value of ``display.precision`` will directly control the number of places after the decimal, for
+regular formatting as well as scientific notation, similar to how numpy's ``precision`` print option works.
+
+.. ipython:: python
+
+  pd.set_option("display.precision", 2)
+  pd.DataFrame({"x": [123.456789]})
+
+To preserve output behavior with prior versions the default value of ``display.precision`` has been reduced to ``6``
+from ``7``.
+
+.. ipython:: python
+  :suppress:
+
+  pd.set_option("display.precision", 6)
+
+.. _whatsnew_0170.api_breaking.categorical_unique:
+
+Changes to ``Categorical.unique``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Categorical.unique`` now returns new ``Categoricals`` with ``categories`` and ``codes`` that are unique, rather than returning ``np.array`` (:issue:`10508`)
+
+- unordered category: values and categories are sorted by appearance order.
+- ordered category: values are sorted by appearance order, categories keep existing order.
+
+.. ipython:: python
+
+   cat = pd.Categorical(["C", "A", "B", "C"], categories=["A", "B", "C"], ordered=True)
+   cat
+   cat.unique()
+
+   cat = pd.Categorical(["C", "A", "B", "C"], categories=["A", "B", "C"])
+   cat
+   cat.unique()
+
+Changes to ``bool`` passed as ``header`` in parsers
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In earlier versions of pandas, if a bool was passed the ``header`` argument of
+``read_csv``, ``read_excel``, or ``read_html`` it was implicitly converted to
+an integer, resulting in ``header=0`` for ``False`` and ``header=1`` for ``True``
+(:issue:`6113`)
+
+A ``bool`` input to ``header`` will now raise a ``TypeError``
+
+.. code-block:: ipython
+
+   In [29]: df = pd.read_csv('data.csv', header=False)
+   TypeError: Passing a bool to header is invalid. Use header=None for no header or
+   header=int or list-like of ints to specify the row(s) making up the column names
+
+
+.. _whatsnew_0170.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- Line and kde plot with ``subplots=True`` now uses default colors, not all black. Specify ``color='k'`` to draw all lines in black (:issue:`9894`)
+- Calling the ``.value_counts()`` method on a Series with a ``categorical`` dtype now returns a Series with a ``CategoricalIndex`` (:issue:`10704`)
+- The metadata properties of subclasses of pandas objects will now be serialized (:issue:`10553`).
+- ``groupby`` using ``Categorical`` follows the same rule as ``Categorical.unique`` described above  (:issue:`10508`)
+- When constructing ``DataFrame`` with an array of ``complex64`` dtype previously meant the corresponding column
+  was automatically promoted to the ``complex128`` dtype. pandas will now preserve the itemsize of the input for complex data (:issue:`10952`)
+- some numeric reduction operators would return ``ValueError``, rather than ``TypeError`` on object types that includes strings and numbers (:issue:`11131`)
+- Passing currently unsupported ``chunksize`` argument to ``read_excel`` or ``ExcelFile.parse`` will now raise ``NotImplementedError`` (:issue:`8011`)
+- Allow an ``ExcelFile`` object to be passed into ``read_excel`` (:issue:`11198`)
+- ``DatetimeIndex.union`` does not infer ``freq`` if ``self`` and the input have ``None`` as ``freq`` (:issue:`11086`)
+- ``NaT``'s methods now either raise ``ValueError``, or return ``np.nan`` or ``NaT`` (:issue:`9513`)
+
+  ===============================     ===============================================================
+  Behavior                            Methods
+  ===============================     ===============================================================
+  return ``np.nan``                   ``weekday``, ``isoweekday``
+  return ``NaT``                      ``date``, ``now``, ``replace``, ``to_datetime``, ``today``
+  return ``np.datetime64('NaT')``     ``to_datetime64`` (unchanged)
+  raise ``ValueError``                All other public methods (names not beginning with underscores)
+  ===============================     ===============================================================
+
+.. _whatsnew_0170.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- For ``Series`` the following indexing functions are deprecated (:issue:`10177`).
+
+  =====================  =================================
+  Deprecated Function    Replacement
+  =====================  =================================
+  ``.irow(i)``           ``.iloc[i]`` or ``.iat[i]``
+  ``.iget(i)``           ``.iloc[i]`` or ``.iat[i]``
+  ``.iget_value(i)``     ``.iloc[i]`` or ``.iat[i]``
+  =====================  =================================
+
+- For ``DataFrame`` the following indexing functions are deprecated (:issue:`10177`).
+
+  =====================  =================================
+  Deprecated Function    Replacement
+  =====================  =================================
+  ``.irow(i)``           ``.iloc[i]``
+  ``.iget_value(i, j)``  ``.iloc[i, j]`` or ``.iat[i, j]``
+  ``.icol(j)``           ``.iloc[:, j]``
+  =====================  =================================
+
+.. note:: These indexing function have been deprecated in the documentation since 0.11.0.
+
+- ``Categorical.name`` was deprecated to make ``Categorical`` more ``numpy.ndarray`` like. Use ``Series(cat, name="whatever")`` instead (:issue:`10482`).
+- Setting missing values (NaN) in a ``Categorical``'s ``categories`` will issue a warning (:issue:`10748`). You can still have missing values in the ``values``.
+- ``drop_duplicates`` and ``duplicated``'s ``take_last`` keyword was deprecated in favor of ``keep``. (:issue:`6511`, :issue:`8505`)
+- ``Series.nsmallest`` and ``nlargest``'s ``take_last`` keyword was deprecated in favor of ``keep``. (:issue:`10792`)
+- ``DataFrame.combineAdd`` and ``DataFrame.combineMult`` are deprecated. They
+  can easily be replaced by using the ``add`` and ``mul`` methods:
+  ``DataFrame.add(other, fill_value=0)`` and ``DataFrame.mul(other, fill_value=1.)``
+  (:issue:`10735`).
+- ``TimeSeries`` deprecated in favor of ``Series`` (note that this has been an alias since 0.13.0), (:issue:`10890`)
+- ``SparsePanel`` deprecated and will be removed in a future version (:issue:`11157`).
+- ``Series.is_time_series`` deprecated in favor of ``Series.index.is_all_dates`` (:issue:`11135`)
+- Legacy offsets (like ``'A@JAN'``) are deprecated (note that this has been alias since 0.8.0) (:issue:`10878`)
+- ``WidePanel`` deprecated in favor of ``Panel``, ``LongPanel`` in favor of ``DataFrame`` (note these have been aliases since < 0.11.0), (:issue:`10892`)
+- ``DataFrame.convert_objects`` has been deprecated in favor of type-specific functions ``pd.to_datetime``, ``pd.to_timestamp`` and ``pd.to_numeric`` (new in 0.17.0) (:issue:`11133`).
+
+.. _whatsnew_0170.prior_deprecations:
+
+Removal of prior version deprecations/changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Removal of ``na_last`` parameters from ``Series.order()`` and ``Series.sort()``, in favor of ``na_position``. (:issue:`5231`)
+- Remove of ``percentile_width`` from ``.describe()``, in favor of ``percentiles``. (:issue:`7088`)
+- Removal of ``colSpace`` parameter from ``DataFrame.to_string()``, in favor of ``col_space``, circa 0.8.0 version.
+- Removal of automatic time-series broadcasting (:issue:`2304`)
+
+  .. ipython:: python
+
+     np.random.seed(1234)
+     df = pd.DataFrame(
+         np.random.randn(5, 2),
+         columns=list("AB"),
+         index=pd.date_range("2013-01-01", periods=5),
+     )
+     df
+
+  Previously
+
+  .. code-block:: ipython
+
+     In [3]: df + df.A
+     FutureWarning: TimeSeries broadcasting along DataFrame index by default is deprecated.
+     Please use DataFrame.<op> to explicitly broadcast arithmetic operations along the index
+
+     Out[3]:
+                         A         B
+     2013-01-01  0.942870 -0.719541
+     2013-01-02  2.865414  1.120055
+     2013-01-03 -1.441177  0.166574
+     2013-01-04  1.719177  0.223065
+     2013-01-05  0.031393 -2.226989
+
+  Current
+
+  .. ipython:: python
+
+     df.add(df.A, axis="index")
+
+
+- Remove ``table`` keyword in ``HDFStore.put/append``, in favor of using ``format=`` (:issue:`4645`)
+- Remove ``kind`` in ``read_excel/ExcelFile`` as its unused (:issue:`4712`)
+- Remove ``infer_type`` keyword from ``pd.read_html`` as its unused (:issue:`4770`, :issue:`7032`)
+- Remove ``offset`` and ``timeRule`` keywords from ``Series.tshift/shift``, in favor of ``freq`` (:issue:`4853`, :issue:`4864`)
+- Remove ``pd.load/pd.save`` aliases in favor of ``pd.to_pickle/pd.read_pickle`` (:issue:`3787`)
+
+.. _whatsnew_0170.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Development support for benchmarking with the `Air Speed Velocity library <https://github.com/spacetelescope/asv/>`_ (:issue:`8361`)
+- Added vbench benchmarks for alternative ExcelWriter engines and reading Excel files (:issue:`7171`)
+- Performance improvements in ``Categorical.value_counts`` (:issue:`10804`)
+- Performance improvements in ``SeriesGroupBy.nunique`` and ``SeriesGroupBy.value_counts`` and ``SeriesGroupby.transform`` (:issue:`10820`, :issue:`11077`)
+- Performance improvements in ``DataFrame.drop_duplicates`` with integer dtypes (:issue:`10917`)
+- Performance improvements in ``DataFrame.duplicated`` with wide frames. (:issue:`10161`, :issue:`11180`)
+- 4x improvement in ``timedelta`` string parsing (:issue:`6755`, :issue:`10426`)
+- 8x improvement in ``timedelta64`` and ``datetime64`` ops (:issue:`6755`)
+- Significantly improved performance of indexing ``MultiIndex`` with slicers (:issue:`10287`)
+- 8x improvement in ``iloc`` using list-like input (:issue:`10791`)
+- Improved performance of ``Series.isin`` for datetimelike/integer Series (:issue:`10287`)
+- 20x improvement in ``concat`` of Categoricals when categories are identical (:issue:`10587`)
+- Improved performance of ``to_datetime`` when specified format string is ISO8601 (:issue:`10178`)
+- 2x improvement of ``Series.value_counts`` for float dtype (:issue:`10821`)
+- Enable ``infer_datetime_format`` in ``to_datetime`` when date components do not have 0 padding (:issue:`11142`)
+- Regression from 0.16.1 in constructing ``DataFrame`` from nested dictionary (:issue:`11084`)
+- Performance improvements in addition/subtraction operations for ``DateOffset`` with ``Series`` or ``DatetimeIndex``  (:issue:`10744`, :issue:`11205`)
+
+.. _whatsnew_0170.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in incorrect computation of ``.mean()`` on ``timedelta64[ns]`` because of overflow (:issue:`9442`)
+- Bug in  ``.isin`` on older numpies (:issue:`11232`)
+- Bug in ``DataFrame.to_html(index=False)`` renders unnecessary ``name`` row (:issue:`10344`)
+- Bug in ``DataFrame.to_latex()`` the ``column_format`` argument could not be passed (:issue:`9402`)
+- Bug in ``DatetimeIndex`` when localizing with ``NaT`` (:issue:`10477`)
+- Bug in ``Series.dt`` ops in preserving meta-data (:issue:`10477`)
+- Bug in preserving ``NaT`` when passed in an otherwise invalid ``to_datetime`` construction (:issue:`10477`)
+- Bug in ``DataFrame.apply`` when function returns categorical series. (:issue:`9573`)
+- Bug in ``to_datetime`` with invalid dates and formats supplied (:issue:`10154`)
+- Bug in ``Index.drop_duplicates`` dropping name(s) (:issue:`10115`)
+- Bug in ``Series.quantile`` dropping name (:issue:`10881`)
+- Bug in ``pd.Series`` when setting a value on an empty ``Series`` whose index has a frequency. (:issue:`10193`)
+- Bug in ``pd.Series.interpolate`` with invalid ``order`` keyword values. (:issue:`10633`)
+- Bug in ``DataFrame.plot`` raises ``ValueError`` when color name is specified by multiple characters (:issue:`10387`)
+- Bug in ``Index`` construction with a mixed list of tuples (:issue:`10697`)
+- Bug in ``DataFrame.reset_index`` when index contains ``NaT``. (:issue:`10388`)
+- Bug in ``ExcelReader`` when worksheet is empty (:issue:`6403`)
+- Bug in ``BinGrouper.group_info`` where returned values are not compatible with base class (:issue:`10914`)
+- Bug in clearing the cache on ``DataFrame.pop`` and a subsequent inplace op (:issue:`10912`)
+- Bug in indexing with a mixed-integer ``Index`` causing an ``ImportError`` (:issue:`10610`)
+- Bug in ``Series.count`` when index has nulls (:issue:`10946`)
+- Bug in pickling of a non-regular freq ``DatetimeIndex`` (:issue:`11002`)
+- Bug causing ``DataFrame.where`` to not respect the ``axis`` parameter when the frame has a symmetric shape. (:issue:`9736`)
+- Bug in ``Table.select_column`` where name is not preserved (:issue:`10392`)
+- Bug in ``offsets.generate_range`` where ``start`` and ``end`` have finer precision than ``offset`` (:issue:`9907`)
+- Bug in ``pd.rolling_*`` where ``Series.name`` would be lost in the output (:issue:`10565`)
+- Bug in ``stack`` when index or columns are not unique. (:issue:`10417`)
+- Bug in setting a ``Panel`` when an axis has a MultiIndex (:issue:`10360`)
+- Bug in ``USFederalHolidayCalendar`` where ``USMemorialDay`` and ``USMartinLutherKingJr`` were incorrect (:issue:`10278` and :issue:`9760` )
+- Bug in ``.sample()`` where returned object, if set, gives unnecessary ``SettingWithCopyWarning`` (:issue:`10738`)
+- Bug in ``.sample()`` where weights passed as ``Series`` were not aligned along axis before being treated positionally, potentially causing problems if weight indices were not aligned with sampled object. (:issue:`10738`)
+
+- Regression fixed in (:issue:`9311`, :issue:`6620`, :issue:`9345`), where groupby with a datetime-like converting to float with certain aggregators (:issue:`10979`)
+
+- Bug in ``DataFrame.interpolate`` with ``axis=1`` and ``inplace=True`` (:issue:`10395`)
+- Bug in ``io.sql.get_schema`` when specifying multiple columns as primary
+  key (:issue:`10385`).
+
+- Bug in ``groupby(sort=False)`` with datetime-like ``Categorical`` raises ``ValueError`` (:issue:`10505`)
+- Bug in ``groupby(axis=1)`` with ``filter()`` throws ``IndexError`` (:issue:`11041`)
+- Bug in ``test_categorical`` on big-endian builds (:issue:`10425`)
+- Bug in ``Series.shift`` and ``DataFrame.shift`` not supporting categorical data (:issue:`9416`)
+- Bug in ``Series.map`` using categorical ``Series`` raises ``AttributeError`` (:issue:`10324`)
+- Bug in ``MultiIndex.get_level_values`` including ``Categorical`` raises ``AttributeError`` (:issue:`10460`)
+- Bug in ``pd.get_dummies`` with ``sparse=True`` not returning ``SparseDataFrame`` (:issue:`10531`)
+- Bug in ``Index`` subtypes (such as ``PeriodIndex``) not returning their own type for ``.drop`` and ``.insert`` methods (:issue:`10620`)
+- Bug in ``algos.outer_join_indexer`` when ``right`` array is empty (:issue:`10618`)
+
+- Bug in ``filter`` (regression from 0.16.0) and ``transform`` when grouping on multiple keys, one of which is datetime-like (:issue:`10114`)
+
+
+- Bug in ``to_datetime`` and ``to_timedelta`` causing ``Index`` name to be lost (:issue:`10875`)
+- Bug in ``len(DataFrame.groupby)`` causing ``IndexError`` when there's a column containing only NaNs (:issue:`11016`)
+
+- Bug that caused segfault when resampling an empty Series (:issue:`10228`)
+- Bug in ``DatetimeIndex`` and ``PeriodIndex.value_counts`` resets name from its result, but retains in result's ``Index``. (:issue:`10150`)
+- Bug in ``pd.eval`` using ``numexpr`` engine coerces 1 element numpy array to scalar (:issue:`10546`)
+- Bug in ``pd.concat`` with ``axis=0`` when column is of dtype ``category`` (:issue:`10177`)
+- Bug in ``read_msgpack`` where input type is not always checked (:issue:`10369`, :issue:`10630`)
+- Bug in ``pd.read_csv`` with kwargs ``index_col=False``, ``index_col=['a', 'b']`` or ``dtype``
+  (:issue:`10413`, :issue:`10467`, :issue:`10577`)
+- Bug in ``Series.from_csv`` with ``header`` kwarg not setting the ``Series.name`` or the ``Series.index.name`` (:issue:`10483`)
+- Bug in ``groupby.var`` which caused variance to be inaccurate for small float values (:issue:`10448`)
+- Bug in ``Series.plot(kind='hist')`` Y Label not informative (:issue:`10485`)
+- Bug in ``read_csv`` when using a converter which generates a ``uint8`` type (:issue:`9266`)
+
+- Bug causes memory leak in time-series line and area plot (:issue:`9003`)
+
+- Bug when setting a ``Panel`` sliced along the major or minor axes when the right-hand side is a ``DataFrame`` (:issue:`11014`)
+- Bug that returns ``None`` and does not raise ``NotImplementedError`` when operator functions (e.g. ``.add``) of ``Panel`` are not implemented (:issue:`7692`)
+
+- Bug in line and kde plot cannot accept multiple colors when ``subplots=True`` (:issue:`9894`)
+- Bug in ``DataFrame.plot`` raises ``ValueError`` when color name is specified by multiple characters (:issue:`10387`)
+
+- Bug in left and right ``align`` of ``Series`` with ``MultiIndex`` may be inverted (:issue:`10665`)
+- Bug in left and right ``join`` of with ``MultiIndex`` may be inverted (:issue:`10741`)
+
+- Bug in ``read_stata`` when reading a file with a different order set in ``columns`` (:issue:`10757`)
+- Bug in ``Categorical`` may not representing properly when category contains ``tz`` or ``Period`` (:issue:`10713`)
+- Bug in ``Categorical.__iter__`` may not returning correct ``datetime`` and ``Period`` (:issue:`10713`)
+- Bug in indexing with a ``PeriodIndex`` on an object with a ``PeriodIndex`` (:issue:`4125`)
+- Bug in ``read_csv`` with ``engine='c'``: EOF preceded by a comment, blank line, etc. was not handled correctly (:issue:`10728`, :issue:`10548`)
+
+- Reading "famafrench" data via ``DataReader`` results in HTTP 404 error because of the website url is changed (:issue:`10591`).
+- Bug in ``read_msgpack`` where DataFrame to decode has duplicate column names (:issue:`9618`)
+- Bug in ``io.common.get_filepath_or_buffer`` which caused reading of valid S3 files to fail if the bucket also contained keys for which the user does not have read permission (:issue:`10604`)
+- Bug in vectorised setting of timestamp columns with python ``datetime.date`` and numpy ``datetime64`` (:issue:`10408`, :issue:`10412`)
+- Bug in ``Index.take`` may add unnecessary ``freq`` attribute (:issue:`10791`)
+- Bug in ``merge`` with empty ``DataFrame`` may raise ``IndexError`` (:issue:`10824`)
+- Bug in ``to_latex`` where unexpected keyword argument for some documented arguments (:issue:`10888`)
+- Bug in indexing of large ``DataFrame`` where ``IndexError`` is uncaught (:issue:`10645` and :issue:`10692`)
+- Bug in ``read_csv`` when using the ``nrows`` or ``chunksize`` parameters if file contains only a header line (:issue:`9535`)
+- Bug in serialization of ``category`` types in HDF5 in presence of alternate encodings. (:issue:`10366`)
+- Bug in ``pd.DataFrame`` when constructing an empty DataFrame with a string dtype (:issue:`9428`)
+- Bug in ``pd.DataFrame.diff`` when DataFrame is not consolidated (:issue:`10907`)
+- Bug in ``pd.unique`` for arrays with the ``datetime64`` or ``timedelta64`` dtype that meant an array with object dtype was returned instead the original dtype (:issue:`9431`)
+- Bug in ``Timedelta`` raising error when slicing from 0s (:issue:`10583`)
+- Bug in ``DatetimeIndex.take`` and ``TimedeltaIndex.take`` may not raise ``IndexError`` against invalid index (:issue:`10295`)
+- Bug in ``Series([np.nan]).astype('M8[ms]')``, which now returns ``Series([pd.NaT])`` (:issue:`10747`)
+- Bug in ``PeriodIndex.order`` reset freq (:issue:`10295`)
+- Bug in ``date_range`` when ``freq`` divides ``end`` as nanos (:issue:`10885`)
+- Bug in ``iloc`` allowing memory outside bounds of a Series to be accessed with negative integers (:issue:`10779`)
+- Bug in ``read_msgpack`` where encoding is not respected (:issue:`10581`)
+- Bug preventing access to the first index when using ``iloc`` with a list containing the appropriate negative integer (:issue:`10547`, :issue:`10779`)
+- Bug in ``TimedeltaIndex`` formatter causing error while trying to save ``DataFrame`` with ``TimedeltaIndex`` using ``to_csv`` (:issue:`10833`)
+- Bug in ``DataFrame.where`` when handling Series slicing (:issue:`10218`, :issue:`9558`)
+- Bug where ``pd.read_gbq`` throws ``ValueError`` when Bigquery returns zero rows (:issue:`10273`)
+- Bug in ``to_json`` which was causing segmentation fault when serializing 0-rank ndarray (:issue:`9576`)
+- Bug in plotting functions may raise ``IndexError`` when plotted on ``GridSpec`` (:issue:`10819`)
+- Bug in plot result may show unnecessary minor ticklabels (:issue:`10657`)
+- Bug in ``groupby`` incorrect computation for aggregation on ``DataFrame`` with ``NaT`` (E.g ``first``, ``last``, ``min``). (:issue:`10590`, :issue:`11010`)
+- Bug when constructing ``DataFrame`` where passing a dictionary with only scalar values and specifying columns did not raise an error (:issue:`10856`)
+- Bug in ``.var()`` causing roundoff errors for highly similar values (:issue:`10242`)
+- Bug in ``DataFrame.plot(subplots=True)`` with duplicated columns outputs incorrect result (:issue:`10962`)
+- Bug in ``Index`` arithmetic may result in incorrect class (:issue:`10638`)
+- Bug in ``date_range`` results in empty if freq is negative annually, quarterly and monthly (:issue:`11018`)
+- Bug in ``DatetimeIndex`` cannot infer negative freq (:issue:`11018`)
+- Remove use of some deprecated numpy comparison operations, mainly in tests. (:issue:`10569`)
+- Bug in ``Index`` dtype may not applied properly (:issue:`11017`)
+- Bug in ``io.gbq`` when testing for minimum google api client version (:issue:`10652`)
+- Bug in ``DataFrame`` construction from nested ``dict`` with ``timedelta`` keys (:issue:`11129`)
+- Bug in ``.fillna`` against may raise ``TypeError`` when data contains datetime dtype (:issue:`7095`, :issue:`11153`)
+- Bug in ``.groupby`` when number of keys to group by is same as length of index (:issue:`11185`)
+- Bug in ``convert_objects`` where converted values might not be returned if all null and ``coerce`` (:issue:`9589`)
+- Bug in ``convert_objects`` where ``copy`` keyword was not respected (:issue:`9589`)
+
+
+.. _whatsnew_0.17.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.16.2..v0.17.0
diff --git a/pandas/doc/source/whatsnew/v0.17.1.rst b/pandas/doc/source/whatsnew/v0.17.1.rst
new file mode 100644
index 0000000000000..93de83875746b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.17.1.rst
@@ -0,0 +1,213 @@
+.. _whatsnew_0171:
+
+Version 0.17.1 (November 21, 2015)
+----------------------------------
+
+{{ header }}
+
+
+.. note::
+
+   We are proud to announce that *pandas* has become a sponsored project of the (`NumFOCUS organization`_). This will help ensure the success of development of *pandas* as a world-class open-source project.
+
+.. _numfocus organization: http://www.numfocus.org/blog/numfocus-announces-new-fiscally-sponsored-project-pandas
+
+This is a minor bug-fix release from 0.17.0 and includes a large number of
+bug fixes along several new features, enhancements, and performance improvements.
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- Support for Conditional HTML Formatting, see :ref:`here <whatsnew_0171.style>`
+- Releasing the GIL on the csv reader & other ops, see :ref:`here <whatsnew_0171.performance>`
+- Fixed regression in ``DataFrame.drop_duplicates`` from 0.16.2, causing incorrect results on integer values (:issue:`11376`)
+
+.. contents:: What's new in v0.17.1
+    :local:
+    :backlinks: none
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0171.style:
+
+Conditional HTML formatting
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. warning::
+    This is a new feature and is under active development.
+    We'll be adding features an  possibly making breaking changes in future
+    releases. Feedback is welcome in :issue:`11610`
+
+We've added *experimental* support for conditional HTML formatting:
+the visual styling of a DataFrame based on the data.
+The styling is accomplished with HTML and CSS.
+Accesses the styler class with the :attr:`pandas.DataFrame.style`, attribute,
+an instance of :class:`.Styler` with your data attached.
+
+Here's a quick example:
+
+  .. ipython:: python
+
+    np.random.seed(123)
+    df = pd.DataFrame(np.random.randn(10, 5), columns=list("abcde"))
+    html = df.style.background_gradient(cmap="viridis", low=0.5)
+
+We can render the HTML to get the following table.
+
+.. raw:: html
+   :file: whatsnew_0171_html_table.html
+
+:class:`.Styler` interacts nicely with the Jupyter Notebook.
+See the :ref:`documentation </user_guide/style.ipynb>` for more.
+
+.. _whatsnew_0171.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+- ``DatetimeIndex`` now supports conversion to strings with ``astype(str)`` (:issue:`10442`)
+- Support for ``compression`` (gzip/bz2) in :meth:`pandas.DataFrame.to_csv` (:issue:`7615`)
+- ``pd.read_*`` functions can now also accept :class:`python:pathlib.Path`, or :class:`py:py._path.local.LocalPath`
+  objects for the ``filepath_or_buffer`` argument. (:issue:`11033`)
+  - The ``DataFrame`` and ``Series`` functions ``.to_csv()``, ``.to_html()`` and ``.to_latex()`` can now handle paths beginning with tildes (e.g. ``~/Documents/``) (:issue:`11438`)
+- ``DataFrame`` now uses the fields of a ``namedtuple`` as columns, if columns are not supplied (:issue:`11181`)
+- ``DataFrame.itertuples()`` now returns ``namedtuple`` objects, when possible. (:issue:`11269`, :issue:`11625`)
+- Added ``axvlines_kwds`` to parallel coordinates plot (:issue:`10709`)
+- Option to ``.info()`` and ``.memory_usage()`` to provide for deep introspection of memory consumption. Note that this can be expensive to compute and therefore is an optional parameter. (:issue:`11595`)
+
+  .. ipython:: python
+
+     df = pd.DataFrame({"A": ["foo"] * 1000})  # noqa: F821
+     df["B"] = df["A"].astype("category")
+
+     # shows the '+' as we have object dtypes
+     df.info()
+
+     # we have an accurate memory assessment (but can be expensive to compute this)
+     df.info(memory_usage="deep")
+
+- ``Index`` now has a ``fillna`` method (:issue:`10089`)
+
+  .. ipython:: python
+
+     pd.Index([1, np.nan, 3]).fillna(2)
+
+- Series of type ``category`` now make ``.str.<...>`` and ``.dt.<...>`` accessor methods / properties available, if the categories are of that type. (:issue:`10661`)
+
+  .. ipython:: python
+
+     s = pd.Series(list("aabb")).astype("category")
+     s
+     s.str.contains("a")
+
+     date = pd.Series(pd.date_range("1/1/2015", periods=5)).astype("category")
+     date
+     date.dt.day
+
+- ``pivot_table`` now has a ``margins_name`` argument so you can use something other than the default of 'All' (:issue:`3335`)
+- Implement export of ``datetime64[ns, tz]`` dtypes with a fixed HDF5 store (:issue:`11411`)
+- Pretty printing sets (e.g. in DataFrame cells) now uses set literal syntax (``{x, y}``) instead of
+  Legacy Python syntax (``set([x, y])``) (:issue:`11215`)
+- Improve the error message in :func:`pandas.io.gbq.to_gbq` when a streaming insert fails (:issue:`11285`)
+  and when the DataFrame does not match the schema of the destination table (:issue:`11359`)
+
+.. _whatsnew_0171.api:
+
+API changes
+~~~~~~~~~~~
+
+- raise ``NotImplementedError`` in ``Index.shift`` for non-supported index types (:issue:`8038`)
+- ``min`` and ``max`` reductions on ``datetime64`` and ``timedelta64`` dtyped series now
+  result in ``NaT`` and not ``nan`` (:issue:`11245`).
+- Indexing with a null key will raise a ``TypeError``, instead of a ``ValueError`` (:issue:`11356`)
+- ``Series.ptp`` will now ignore missing values by default (:issue:`11163`)
+
+.. _whatsnew_0171.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- The ``pandas.io.ga`` module which implements ``google-analytics`` support is deprecated and will be removed in a future version (:issue:`11308`)
+- Deprecate the ``engine`` keyword in ``.to_csv()``, which will be removed in a future version (:issue:`11274`)
+
+.. _whatsnew_0171.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Checking monotonic-ness before sorting on an index (:issue:`11080`)
+- ``Series.dropna`` performance improvement when its dtype can't contain ``NaN`` (:issue:`11159`)
+- Release the GIL on most datetime field operations (e.g. ``DatetimeIndex.year``, ``Series.dt.year``), normalization, and conversion to and from ``Period``, ``DatetimeIndex.to_period`` and ``PeriodIndex.to_timestamp`` (:issue:`11263`)
+- Release the GIL on some rolling algos: ``rolling_median``, ``rolling_mean``, ``rolling_max``, ``rolling_min``, ``rolling_var``, ``rolling_kurt``, ``rolling_skew`` (:issue:`11450`)
+- Release the GIL when reading and parsing text files in ``read_csv``, ``read_table`` (:issue:`11272`)
+- Improved performance of ``rolling_median`` (:issue:`11450`)
+- Improved performance of ``to_excel`` (:issue:`11352`)
+- Performance bug in repr of ``Categorical`` categories, which was rendering the strings before chopping them for display (:issue:`11305`)
+- Performance improvement in ``Categorical.remove_unused_categories``, (:issue:`11643`).
+- Improved performance of ``Series`` constructor with no data and ``DatetimeIndex`` (:issue:`11433`)
+- Improved performance of ``shift``, ``cumprod``, and ``cumsum`` with groupby (:issue:`4095`)
+
+.. _whatsnew_0171.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- ``SparseArray.__iter__()`` now does not cause ``PendingDeprecationWarning`` in Python 3.5 (:issue:`11622`)
+- Regression from 0.16.2 for output formatting of long floats/nan, restored in (:issue:`11302`)
+- ``Series.sort_index()`` now correctly handles the ``inplace`` option (:issue:`11402`)
+- Incorrectly distributed .c file in the build on ``PyPi`` when reading a csv of floats and passing ``na_values=<a scalar>`` would show an exception (:issue:`11374`)
+- Bug in ``.to_latex()`` output broken when the index has a name (:issue:`10660`)
+- Bug in ``HDFStore.append`` with strings whose encoded length exceeded the max unencoded length (:issue:`11234`)
+- Bug in merging ``datetime64[ns, tz]`` dtypes (:issue:`11405`)
+- Bug in ``HDFStore.select`` when comparing with a numpy scalar in a where clause (:issue:`11283`)
+- Bug in using ``DataFrame.ix`` with a MultiIndex indexer (:issue:`11372`)
+- Bug in ``date_range`` with ambiguous endpoints (:issue:`11626`)
+- Prevent adding new attributes to the accessors ``.str``, ``.dt`` and ``.cat``. Retrieving such
+  a value was not possible, so error out on setting it. (:issue:`10673`)
+- Bug in tz-conversions with an ambiguous time and ``.dt`` accessors (:issue:`11295`)
+- Bug in output formatting when using an index of ambiguous times (:issue:`11619`)
+- Bug in comparisons of Series vs list-likes (:issue:`11339`)
+- Bug in ``DataFrame.replace`` with a ``datetime64[ns, tz]`` and a non-compat to_replace (:issue:`11326`, :issue:`11153`)
+- Bug in ``isnull`` where ``numpy.datetime64('NaT')`` in a ``numpy.array`` was not determined to be null(:issue:`11206`)
+- Bug in list-like indexing with a mixed-integer Index (:issue:`11320`)
+- Bug in ``pivot_table`` with ``margins=True`` when indexes are of ``Categorical`` dtype (:issue:`10993`)
+- Bug in ``DataFrame.plot`` cannot use hex strings colors (:issue:`10299`)
+- Regression in ``DataFrame.drop_duplicates`` from 0.16.2, causing incorrect results on integer values (:issue:`11376`)
+- Bug in ``pd.eval`` where unary ops in a list error (:issue:`11235`)
+- Bug in ``squeeze()`` with zero length arrays (:issue:`11230`, :issue:`8999`)
+- Bug in ``describe()`` dropping column names for hierarchical indexes (:issue:`11517`)
+- Bug in ``DataFrame.pct_change()`` not propagating ``axis`` keyword on ``.fillna`` method (:issue:`11150`)
+- Bug in ``.to_csv()`` when a mix of integer and string column names are passed as the ``columns`` parameter (:issue:`11637`)
+- Bug in indexing with a ``range``, (:issue:`11652`)
+- Bug in inference of numpy scalars and preserving dtype when setting columns (:issue:`11638`)
+- Bug in ``to_sql`` using unicode column names giving UnicodeEncodeError with (:issue:`11431`).
+- Fix regression in setting of ``xticks`` in ``plot`` (:issue:`11529`).
+- Bug in ``holiday.dates`` where observance rules could not be applied to holiday and doc enhancement (:issue:`11477`, :issue:`11533`)
+- Fix plotting issues when having plain ``Axes`` instances instead of ``SubplotAxes`` (:issue:`11520`, :issue:`11556`).
+- Bug in ``DataFrame.to_latex()`` produces an extra rule when ``header=False`` (:issue:`7124`)
+- Bug in ``df.groupby(...).apply(func)`` when a func returns a ``Series`` containing a new datetimelike column (:issue:`11324`)
+- Bug in ``pandas.json`` when file to load is big (:issue:`11344`)
+- Bugs in ``to_excel`` with duplicate columns (:issue:`11007`, :issue:`10982`, :issue:`10970`)
+- Fixed a bug that prevented the construction of an empty series of dtype ``datetime64[ns, tz]`` (:issue:`11245`).
+- Bug in ``read_excel`` with MultiIndex containing integers (:issue:`11317`)
+- Bug in ``to_excel`` with openpyxl 2.2+ and merging (:issue:`11408`)
+- Bug in ``DataFrame.to_dict()`` produces a ``np.datetime64`` object instead of ``Timestamp`` when only datetime is present in data (:issue:`11327`)
+- Bug in ``DataFrame.corr()`` raises exception when computes Kendall correlation for DataFrames with boolean and not boolean columns (:issue:`11560`)
+- Bug in the link-time error caused by C ``inline`` functions on FreeBSD 10+ (with ``clang``) (:issue:`10510`)
+- Bug in ``DataFrame.to_csv`` in passing through arguments for formatting ``MultiIndexes``, including ``date_format`` (:issue:`7791`)
+- Bug in ``DataFrame.join()`` with ``how='right'`` producing a ``TypeError`` (:issue:`11519`)
+- Bug in ``Series.quantile`` with empty list results has ``Index`` with ``object`` dtype (:issue:`11588`)
+- Bug in ``pd.merge`` results in empty ``Int64Index`` rather than ``Index(dtype=object)`` when the merge result is empty (:issue:`11588`)
+- Bug in ``Categorical.remove_unused_categories`` when having ``NaN`` values (:issue:`11599`)
+- Bug in ``DataFrame.to_sparse()`` loses column names for MultiIndexes (:issue:`11600`)
+- Bug in ``DataFrame.round()`` with non-unique column index producing a Fatal Python error (:issue:`11611`)
+- Bug in ``DataFrame.round()`` with ``decimals`` being a non-unique indexed Series producing extra columns (:issue:`11618`)
+
+
+.. _whatsnew_0.17.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.17.0..v0.17.1
diff --git a/pandas/doc/source/whatsnew/v0.18.0.rst b/pandas/doc/source/whatsnew/v0.18.0.rst
new file mode 100644
index 0000000000000..563035e0e2940
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.18.0.rst
@@ -0,0 +1,1346 @@
+.. _whatsnew_0180:
+
+Version 0.18.0 (March 13, 2016)
+-------------------------------
+
+{{ header }}
+
+
+This is a major release from 0.17.1 and includes a small number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+.. warning::
+
+   pandas >= 0.18.0 no longer supports compatibility with Python version 2.6
+   and 3.3 (:issue:`7718`, :issue:`11273`)
+
+.. warning::
+
+   ``numexpr`` version 2.4.4 will now show a warning and not be used as a computation back-end for pandas because of some buggy behavior. This does not affect other versions (>= 2.1 and >= 2.4.6). (:issue:`12489`)
+
+Highlights include:
+
+- Moving and expanding window functions are now methods on Series and DataFrame,
+  similar to ``.groupby``, see :ref:`here <whatsnew_0180.enhancements.moments>`.
+- Adding support for a ``RangeIndex`` as a specialized form of the ``Int64Index``
+  for memory savings, see :ref:`here <whatsnew_0180.enhancements.rangeindex>`.
+- API breaking change to the ``.resample`` method to make it more ``.groupby``
+  like, see :ref:`here <whatsnew_0180.breaking.resample>`.
+- Removal of support for positional indexing with floats, which was deprecated
+  since 0.14.0. This will now raise a ``TypeError``, see :ref:`here <whatsnew_0180.float_indexers>`.
+- The ``.to_xarray()`` function has been added for compatibility with the
+  `xarray package <http://xarray.pydata.org/en/stable/>`__, see :ref:`here <whatsnew_0180.enhancements.xarray>`.
+- The ``read_sas`` function has been enhanced to read ``sas7bdat`` files, see :ref:`here <whatsnew_0180.enhancements.sas>`.
+- Addition of the :ref:`.str.extractall() method <whatsnew_0180.enhancements.extract>`,
+  and API changes to the :ref:`.str.extract() method <whatsnew_0180.enhancements.extract>`
+  and :ref:`.str.cat() method <whatsnew_0180.enhancements.strcat>`.
+- ``pd.test()`` top-level nose test runner is available (:issue:`4327`).
+
+Check the :ref:`API Changes <whatsnew_0180.api_breaking>` and :ref:`deprecations <whatsnew_0180.deprecations>` before updating.
+
+.. contents:: What's new in v0.18.0
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0180.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0180.enhancements.moments:
+
+Window functions are now methods
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Window functions have been refactored to be methods on ``Series/DataFrame`` objects, rather than top-level functions, which are now deprecated. This allows these window-type functions, to have a similar API to that of ``.groupby``. See the full documentation :ref:`here <window.overview>` (:issue:`11603`, :issue:`12373`)
+
+
+.. ipython:: python
+
+   np.random.seed(1234)
+   df = pd.DataFrame({'A': range(10), 'B': np.random.randn(10)})
+   df
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [8]: pd.rolling_mean(df, window=3)
+           FutureWarning: pd.rolling_mean is deprecated for DataFrame and will be removed in a future version, replace with
+                          DataFrame.rolling(window=3,center=False).mean()
+   Out[8]:
+       A         B
+   0 NaN       NaN
+   1 NaN       NaN
+   2   1  0.237722
+   3   2 -0.023640
+   4   3  0.133155
+   5   4 -0.048693
+   6   5  0.342054
+   7   6  0.370076
+   8   7  0.079587
+   9   8 -0.954504
+
+New behavior:
+
+.. ipython:: python
+
+   r = df.rolling(window=3)
+
+These show a descriptive repr
+
+.. ipython:: python
+
+   r
+with tab-completion of available methods and properties.
+
+.. code-block:: ipython
+
+   In [9]: r.<TAB>  # noqa E225, E999
+   r.A           r.agg         r.apply       r.count       r.exclusions  r.max         r.median      r.name        r.skew        r.sum
+   r.B           r.aggregate   r.corr        r.cov         r.kurt        r.mean        r.min         r.quantile    r.std         r.var
+
+The methods operate on the ``Rolling`` object itself
+
+.. ipython:: python
+
+   r.mean()
+
+They provide getitem accessors
+
+.. ipython:: python
+
+   r['A'].mean()
+
+And multiple aggregations
+
+.. ipython:: python
+
+   r.agg({'A': ['mean', 'std'],
+          'B': ['mean', 'std']})
+
+.. _whatsnew_0180.enhancements.rename:
+
+Changes to rename
+^^^^^^^^^^^^^^^^^
+
+``Series.rename`` and ``NDFrame.rename_axis`` can now take a scalar or list-like
+argument for altering the Series or axis *name*, in addition to their old behaviors of altering labels. (:issue:`9494`, :issue:`11965`)
+
+.. ipython:: python
+
+   s = pd.Series(np.random.randn(5))
+   s.rename('newname')
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(5, 2))
+   (df.rename_axis("indexname")
+      .rename_axis("columns_name", axis="columns"))
+
+The new functionality works well in method chains. Previously these methods only accepted functions or dicts mapping a *label* to a new label.
+This continues to work as before for function or dict-like values.
+
+
+.. _whatsnew_0180.enhancements.rangeindex:
+
+Range Index
+^^^^^^^^^^^
+
+A ``RangeIndex`` has been added to the ``Int64Index`` sub-classes to support a memory saving alternative for common use cases. This has a similar implementation to the python ``range`` object (``xrange`` in python 2), in that it only stores the start, stop, and step values for the index. It will transparently interact with the user API, converting to ``Int64Index`` if needed.
+
+This will now be the default constructed index for ``NDFrame`` objects, rather than previous an ``Int64Index``. (:issue:`939`, :issue:`12070`, :issue:`12071`, :issue:`12109`, :issue:`12888`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [3]: s = pd.Series(range(1000))
+
+   In [4]: s.index
+   Out[4]:
+   Int64Index([  0,   1,   2,   3,   4,   5,   6,   7,   8,   9,
+               ...
+               990, 991, 992, 993, 994, 995, 996, 997, 998, 999], dtype='int64', length=1000)
+
+   In [6]: s.index.nbytes
+   Out[6]: 8000
+
+
+New behavior:
+
+.. ipython:: python
+
+   s = pd.Series(range(1000))
+   s.index
+   s.index.nbytes
+
+.. _whatsnew_0180.enhancements.extract:
+
+Changes to str.extract
+^^^^^^^^^^^^^^^^^^^^^^
+
+The :ref:`.str.extract <text.extract>` method takes a regular
+expression with capture groups, finds the first match in each subject
+string, and returns the contents of the capture groups
+(:issue:`11386`).
+
+In v0.18.0, the ``expand`` argument was added to
+``extract``.
+
+- ``expand=False``: it returns a ``Series``, ``Index``, or ``DataFrame``, depending on the subject and regular expression pattern (same behavior as pre-0.18.0).
+- ``expand=True``: it always returns a ``DataFrame``, which is more consistent and less confusing from the perspective of a user.
+
+Currently the default is ``expand=None`` which gives a ``FutureWarning`` and uses ``expand=False``. To avoid this warning, please explicitly specify ``expand``.
+
+.. code-block:: ipython
+
+   In [1]: pd.Series(['a1', 'b2', 'c3']).str.extract(r'[ab](\d)', expand=None)
+   FutureWarning: currently extract(expand=None) means expand=False (return Index/Series/DataFrame)
+   but in a future version of pandas this will be changed to expand=True (return DataFrame)
+
+   Out[1]:
+   0      1
+   1      2
+   2    NaN
+   dtype: object
+
+Extracting a regular expression with one group returns a Series if
+``expand=False``.
+
+.. ipython:: python
+
+   pd.Series(['a1', 'b2', 'c3']).str.extract(r'[ab](\d)', expand=False)
+
+It returns a ``DataFrame`` with one column if ``expand=True``.
+
+.. ipython:: python
+
+   pd.Series(['a1', 'b2', 'c3']).str.extract(r'[ab](\d)', expand=True)
+
+Calling on an ``Index`` with a regex with exactly one capture group
+returns an ``Index`` if ``expand=False``.
+
+.. ipython:: python
+
+   s = pd.Series(["a1", "b2", "c3"], ["A11", "B22", "C33"])
+   s.index
+   s.index.str.extract("(?P<letter>[a-zA-Z])", expand=False)
+
+It returns a ``DataFrame`` with one column if ``expand=True``.
+
+.. ipython:: python
+
+   s.index.str.extract("(?P<letter>[a-zA-Z])", expand=True)
+
+Calling on an ``Index`` with a regex with more than one capture group
+raises ``ValueError`` if ``expand=False``.
+
+.. code-block:: python
+
+    >>> s.index.str.extract("(?P<letter>[a-zA-Z])([0-9]+)", expand=False)
+    ValueError: only one regex group is supported with Index
+
+It returns a ``DataFrame`` if ``expand=True``.
+
+.. ipython:: python
+
+   s.index.str.extract("(?P<letter>[a-zA-Z])([0-9]+)", expand=True)
+
+In summary, ``extract(expand=True)`` always returns a ``DataFrame``
+with a row for every subject string, and a column for every capture
+group.
+
+.. _whatsnew_0180.enhancements.extractall:
+
+Addition of str.extractall
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :ref:`.str.extractall <text.extractall>` method was added
+(:issue:`11386`).  Unlike ``extract``, which returns only the first
+match.
+
+.. ipython:: python
+
+   s = pd.Series(["a1a2", "b1", "c1"], ["A", "B", "C"])
+   s
+   s.str.extract(r"(?P<letter>[ab])(?P<digit>\d)", expand=False)
+
+The ``extractall`` method returns all matches.
+
+.. ipython:: python
+
+   s.str.extractall(r"(?P<letter>[ab])(?P<digit>\d)")
+
+.. _whatsnew_0180.enhancements.strcat:
+
+Changes to str.cat
+^^^^^^^^^^^^^^^^^^
+
+The method ``.str.cat()`` concatenates the members of a ``Series``. Before, if ``NaN`` values were present in the Series, calling ``.str.cat()`` on it would return ``NaN``, unlike the rest of the ``Series.str.*`` API. This behavior has been amended to ignore ``NaN`` values by default. (:issue:`11435`).
+
+A new, friendlier ``ValueError`` is added to protect against the mistake of supplying the ``sep`` as an arg, rather than as a kwarg. (:issue:`11334`).
+
+.. ipython:: python
+
+    pd.Series(['a', 'b', np.nan, 'c']).str.cat(sep=' ')
+    pd.Series(['a', 'b', np.nan, 'c']).str.cat(sep=' ', na_rep='?')
+
+.. code-block:: ipython
+
+    In [2]: pd.Series(['a', 'b', np.nan, 'c']).str.cat(' ')
+    ValueError: Did you mean to supply a ``sep`` keyword?
+
+
+.. _whatsnew_0180.enhancements.rounding:
+
+Datetimelike rounding
+^^^^^^^^^^^^^^^^^^^^^
+
+``DatetimeIndex``, ``Timestamp``, ``TimedeltaIndex``, ``Timedelta`` have gained the ``.round()``, ``.floor()`` and ``.ceil()`` method for datetimelike rounding, flooring and ceiling. (:issue:`4314`, :issue:`11963`)
+
+Naive datetimes
+
+.. ipython:: python
+
+   dr = pd.date_range('20130101 09:12:56.1234', periods=3)
+   dr
+   dr.round('s')
+
+   # Timestamp scalar
+   dr[0]
+   dr[0].round('10s')
+
+Tz-aware are rounded, floored and ceiled in local times
+
+.. ipython:: python
+
+   dr = dr.tz_localize('US/Eastern')
+   dr
+   dr.round('s')
+
+Timedeltas
+
+.. code-block:: ipython
+
+   In [37]: t = pd.timedelta_range('1 days 2 hr 13 min 45 us', periods=3, freq='d')
+
+   In [38]: t
+   Out[38]:
+   TimedeltaIndex(['1 days 02:13:00.000045', '2 days 02:13:00.000045',
+                   '3 days 02:13:00.000045'],
+                  dtype='timedelta64[ns]', freq='D')
+
+   In [39]: t.round('10min')
+   Out[39]:
+   TimedeltaIndex(['1 days 02:10:00', '2 days 02:10:00',
+                   '3 days 02:10:00'],
+                  dtype='timedelta64[ns]', freq=None)
+
+   # Timedelta scalar
+   In [40]: t[0]
+   Out[40]: Timedelta('1 days 02:13:00.000045')
+
+   In [41]: t[0].round('2h')
+   Out[41]: Timedelta('1 days 02:00:00')
+
+
+In addition, ``.round()``, ``.floor()`` and ``.ceil()`` will be available through the ``.dt`` accessor of ``Series``.
+
+.. ipython:: python
+
+   s = pd.Series(dr)
+   s
+   s.dt.round('D')
+
+Formatting of integers in FloatIndex
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Integers in ``FloatIndex``, e.g. 1., are now formatted with a decimal point and a ``0`` digit, e.g. ``1.0`` (:issue:`11713`)
+This change not only affects the display to the console, but also the output of IO methods like ``.to_csv`` or ``.to_html``.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: s = pd.Series([1, 2, 3], index=np.arange(3.))
+
+   In [3]: s
+   Out[3]:
+   0    1
+   1    2
+   2    3
+   dtype: int64
+
+   In [4]: s.index
+   Out[4]: Float64Index([0.0, 1.0, 2.0], dtype='float64')
+
+   In [5]: print(s.to_csv(path=None))
+   0,1
+   1,2
+   2,3
+
+
+New behavior:
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=np.arange(3.))
+   s
+   s.index
+   print(s.to_csv(path_or_buf=None, header=False))
+
+Changes to dtype assignment behaviors
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When a DataFrame's slice is updated with a new slice of the same dtype, the dtype of the DataFrame will now remain the same. (:issue:`10503`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [5]: df = pd.DataFrame({'a': [0, 1, 1],
+                              'b': pd.Series([100, 200, 300], dtype='uint32')})
+
+   In [7]: df.dtypes
+   Out[7]:
+   a     int64
+   b    uint32
+   dtype: object
+
+   In [8]: ix = df['a'] == 1
+
+   In [9]: df.loc[ix, 'b'] = df.loc[ix, 'b']
+
+   In [11]: df.dtypes
+   Out[11]:
+   a    int64
+   b    int64
+   dtype: object
+
+New behavior:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': [0, 1, 1],
+                      'b': pd.Series([100, 200, 300], dtype='uint32')})
+   df.dtypes
+   ix = df['a'] == 1
+   df.loc[ix, 'b'] = df.loc[ix, 'b']
+   df.dtypes
+
+When a DataFrame's integer slice is partially updated with a new slice of floats that could potentially be down-casted to integer without losing precision, the dtype of the slice will be set to float instead of integer.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [4]: df = pd.DataFrame(np.array(range(1,10)).reshape(3,3),
+                             columns=list('abc'),
+                             index=[[4,4,8], [8,10,12]])
+
+   In [5]: df
+   Out[5]:
+         a  b  c
+   4 8   1  2  3
+     10  4  5  6
+   8 12  7  8  9
+
+   In [7]: df.ix[4, 'c'] = np.array([0., 1.])
+
+   In [8]: df
+   Out[8]:
+         a  b  c
+   4 8   1  2  0
+     10  4  5  1
+   8 12  7  8  9
+
+New behavior:
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.array(range(1,10)).reshape(3,3),
+                     columns=list('abc'),
+                     index=[[4,4,8], [8,10,12]])
+   df
+   df.loc[4, 'c'] = np.array([0., 1.])
+   df
+
+.. _whatsnew_0180.enhancements.xarray:
+
+Method to_xarray
+^^^^^^^^^^^^^^^^
+
+In a future version of pandas, we will be deprecating ``Panel`` and other > 2 ndim objects. In order to provide for continuity,
+all ``NDFrame`` objects have gained the ``.to_xarray()`` method in order to convert to ``xarray`` objects, which has
+a pandas-like interface for > 2 ndim. (:issue:`11972`)
+
+See the `xarray full-documentation here <http://xarray.pydata.org/en/stable/>`__.
+
+.. code-block:: ipython
+
+   In [1]: p = Panel(np.arange(2*3*4).reshape(2,3,4))
+
+   In [2]: p.to_xarray()
+   Out[2]:
+   <xarray.DataArray (items: 2, major_axis: 3, minor_axis: 4)>
+   array([[[ 0,  1,  2,  3],
+           [ 4,  5,  6,  7],
+           [ 8,  9, 10, 11]],
+
+          [[12, 13, 14, 15],
+           [16, 17, 18, 19],
+           [20, 21, 22, 23]]])
+   Coordinates:
+     * items       (items) int64 0 1
+     * major_axis  (major_axis) int64 0 1 2
+     * minor_axis  (minor_axis) int64 0 1 2 3
+
+Latex representation
+^^^^^^^^^^^^^^^^^^^^
+
+``DataFrame`` has gained a ``._repr_latex_()`` method in order to allow for conversion to latex in a ipython/jupyter notebook using nbconvert. (:issue:`11778`)
+
+Note that this must be activated by setting the option ``pd.display.latex.repr=True`` (:issue:`12182`)
+
+For example, if you have a jupyter notebook you plan to convert to latex using nbconvert, place the statement ``pd.display.latex.repr=True`` in the first cell to have the contained DataFrame output also stored as latex.
+
+The options ``display.latex.escape`` and ``display.latex.longtable`` have also been added to the configuration and are used automatically by the ``to_latex``
+method. See the :ref:`available options docs <options.available>` for more info.
+
+.. _whatsnew_0180.enhancements.sas:
+
+``pd.read_sas()`` changes
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``read_sas`` has gained the ability to read SAS7BDAT files, including compressed files.  The files can be read in entirety, or incrementally.  For full details see :ref:`here <io.sas>`. (:issue:`4052`)
+
+.. _whatsnew_0180.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Handle truncated floats in SAS xport files (:issue:`11713`)
+- Added option to hide index in ``Series.to_string`` (:issue:`11729`)
+- ``read_excel`` now supports s3 urls of the format ``s3://bucketname/filename`` (:issue:`11447`)
+- add support for ``AWS_S3_HOST`` env variable when reading from s3 (:issue:`12198`)
+- A simple version of ``Panel.round()`` is now implemented (:issue:`11763`)
+- For Python 3.x, ``round(DataFrame)``, ``round(Series)``, ``round(Panel)`` will work (:issue:`11763`)
+- ``sys.getsizeof(obj)`` returns the memory usage of a pandas object, including the
+  values it contains (:issue:`11597`)
+- ``Series`` gained an ``is_unique`` attribute (:issue:`11946`)
+- ``DataFrame.quantile`` and ``Series.quantile`` now accept ``interpolation`` keyword (:issue:`10174`).
+- Added ``DataFrame.style.format`` for more flexible formatting of cell values (:issue:`11692`)
+- ``DataFrame.select_dtypes`` now allows the ``np.float16`` type code (:issue:`11990`)
+- ``pivot_table()`` now accepts most iterables for the ``values`` parameter (:issue:`12017`)
+- Added Google ``BigQuery`` service account authentication support, which enables authentication on remote servers. (:issue:`11881`, :issue:`12572`). For further details see `here <https://pandas-gbq.readthedocs.io/en/latest/intro.html>`__
+- ``HDFStore`` is now iterable: ``for k in store`` is equivalent to ``for k in store.keys()`` (:issue:`12221`).
+- Add missing methods/fields to ``.dt`` for ``Period`` (:issue:`8848`)
+- The entire code base has been ``PEP``-ified (:issue:`12096`)
+
+.. _whatsnew_0180.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- the leading white spaces have been removed from the output of ``.to_string(index=False)`` method (:issue:`11833`)
+- the ``out`` parameter has been removed from the ``Series.round()`` method. (:issue:`11763`)
+- ``DataFrame.round()`` leaves non-numeric columns unchanged in its return, rather than raises. (:issue:`11885`)
+- ``DataFrame.head(0)`` and ``DataFrame.tail(0)`` return empty frames, rather than ``self``.  (:issue:`11937`)
+- ``Series.head(0)`` and ``Series.tail(0)`` return empty series, rather than ``self``.  (:issue:`11937`)
+- ``to_msgpack`` and ``read_msgpack`` encoding now defaults to ``'utf-8'``. (:issue:`12170`)
+- the order of keyword arguments to text file parsing functions (``.read_csv()``, ``.read_table()``, ``.read_fwf()``) changed to group related arguments. (:issue:`11555`)
+- ``NaTType.isoformat`` now returns the string ``'NaT`` to allow the result to
+  be passed to the constructor of ``Timestamp``. (:issue:`12300`)
+
+NaT and Timedelta operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``NaT`` and ``Timedelta`` have expanded arithmetic operations, which are extended to ``Series``
+arithmetic where applicable.  Operations defined for ``datetime64[ns]`` or ``timedelta64[ns]``
+are now also defined for ``NaT`` (:issue:`11564`).
+
+``NaT`` now supports arithmetic operations with integers and floats.
+
+.. ipython:: python
+
+   pd.NaT * 1
+   pd.NaT * 1.5
+   pd.NaT / 2
+   pd.NaT * np.nan
+
+``NaT`` defines more arithmetic operations with ``datetime64[ns]`` and ``timedelta64[ns]``.
+
+.. ipython:: python
+
+   pd.NaT / pd.NaT
+   pd.Timedelta('1s') / pd.NaT
+
+``NaT`` may represent either a ``datetime64[ns]`` null or a ``timedelta64[ns]`` null.
+Given the ambiguity, it is treated as a ``timedelta64[ns]``, which allows more operations
+to succeed.
+
+.. ipython:: python
+
+   pd.NaT + pd.NaT
+
+   # same as
+   pd.Timedelta('1s') + pd.Timedelta('1s')
+
+as opposed to
+
+.. code-block:: ipython
+
+   In [3]: pd.Timestamp('19900315') + pd.Timestamp('19900315')
+   TypeError: unsupported operand type(s) for +: 'Timestamp' and 'Timestamp'
+
+However, when wrapped in a ``Series`` whose ``dtype`` is ``datetime64[ns]`` or ``timedelta64[ns]``,
+the ``dtype`` information is respected.
+
+.. code-block:: ipython
+
+   In [1]: pd.Series([pd.NaT], dtype='<M8[ns]') + pd.Series([pd.NaT], dtype='<M8[ns]')
+   TypeError: can only operate on a datetimes for subtraction,
+              but the operator [__add__] was passed
+
+.. ipython:: python
+
+   pd.Series([pd.NaT], dtype='<m8[ns]') + pd.Series([pd.NaT], dtype='<m8[ns]')
+
+``Timedelta`` division by ``floats`` now works.
+
+.. ipython:: python
+
+   pd.Timedelta('1s') / 2.0
+
+Subtraction by ``Timedelta`` in a ``Series`` by a ``Timestamp`` works (:issue:`11925`)
+
+.. ipython:: python
+
+   ser = pd.Series(pd.timedelta_range('1 day', periods=3))
+   ser
+   pd.Timestamp('2012-01-01') - ser
+
+
+``NaT.isoformat()`` now returns ``'NaT'``. This change allows
+``pd.Timestamp`` to rehydrate any timestamp like object from its isoformat
+(:issue:`12300`).
+
+Changes to msgpack
+^^^^^^^^^^^^^^^^^^
+
+Forward incompatible changes in ``msgpack`` writing format were made over 0.17.0 and 0.18.0; older versions of pandas cannot read files packed by newer versions (:issue:`12129`, :issue:`10527`)
+
+Bugs in ``to_msgpack`` and ``read_msgpack`` introduced in 0.17.0 and fixed in 0.18.0, caused files packed in Python 2 unreadable by Python 3 (:issue:`12142`). The following table describes the backward and forward compat of msgpacks.
+
+.. warning::
+
+   +----------------------+------------------------+
+   | Packed with          | Can be unpacked with   |
+   +======================+========================+
+   | pre-0.17 / Python 2  | any                    |
+   +----------------------+------------------------+
+   | pre-0.17 / Python 3  | any                    |
+   +----------------------+------------------------+
+   | 0.17 / Python 2      | - ==0.17 / Python 2    |
+   |                      | - >=0.18 / any Python  |
+   +----------------------+------------------------+
+   | 0.17 / Python 3      | >=0.18 / any Python    |
+   +----------------------+------------------------+
+   | 0.18                 | >= 0.18                |
+   +----------------------+------------------------+
+
+
+   0.18.0 is backward-compatible for reading files packed by older versions, except for files packed with 0.17 in Python 2, in which case only they can only be unpacked in Python 2.
+
+Signature change for .rank
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Series.rank`` and ``DataFrame.rank`` now have the same signature (:issue:`11759`)
+
+Previous signature
+
+.. code-block:: ipython
+
+   In [3]: pd.Series([0,1]).rank(method='average', na_option='keep',
+                                 ascending=True, pct=False)
+   Out[3]:
+   0    1
+   1    2
+   dtype: float64
+
+   In [4]: pd.DataFrame([0,1]).rank(axis=0, numeric_only=None,
+                                    method='average', na_option='keep',
+                                    ascending=True, pct=False)
+   Out[4]:
+      0
+   0  1
+   1  2
+
+New signature
+
+.. ipython:: python
+
+   pd.Series([0,1]).rank(axis=0, method='average', numeric_only=False,
+                         na_option='keep', ascending=True, pct=False)
+   pd.DataFrame([0,1]).rank(axis=0, method='average', numeric_only=False,
+                            na_option='keep', ascending=True, pct=False)
+
+
+Bug in QuarterBegin with n=0
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions, the behavior of the QuarterBegin offset was inconsistent
+depending on the date when the ``n`` parameter was 0. (:issue:`11406`)
+
+The general semantics of anchored offsets for ``n=0`` is to not move the date
+when it is an anchor point (e.g., a quarter start date), and otherwise roll
+forward to the next anchor point.
+
+.. ipython:: python
+
+   d = pd.Timestamp('2014-02-01')
+   d
+   d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
+   d + pd.offsets.QuarterBegin(n=0, startingMonth=1)
+
+For the ``QuarterBegin`` offset in previous versions, the date would be rolled
+*backwards* if date was in the same month as the quarter start date.
+
+.. code-block:: ipython
+
+   In [3]: d = pd.Timestamp('2014-02-15')
+
+   In [4]: d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
+   Out[4]: Timestamp('2014-02-01 00:00:00')
+
+This behavior has been corrected in version 0.18.0, which is consistent with
+other anchored offsets like ``MonthBegin`` and ``YearBegin``.
+
+.. ipython:: python
+
+   d = pd.Timestamp('2014-02-15')
+   d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
+
+.. _whatsnew_0180.breaking.resample:
+
+Resample API
+^^^^^^^^^^^^
+
+Like the change in the window functions API :ref:`above <whatsnew_0180.enhancements.moments>`, ``.resample(...)`` is changing to have a more groupby-like API. (:issue:`11732`, :issue:`12702`, :issue:`12202`, :issue:`12332`, :issue:`12334`, :issue:`12348`, :issue:`12448`).
+
+.. ipython:: python
+
+   np.random.seed(1234)
+   df = pd.DataFrame(np.random.rand(10,4),
+                     columns=list('ABCD'),
+                     index=pd.date_range('2010-01-01 09:00:00',
+                                         periods=10, freq='s'))
+   df
+
+
+**Previous API**:
+
+You would write a resampling operation that immediately evaluates. If a ``how`` parameter was not provided, it
+would default to ``how='mean'``.
+
+.. code-block:: ipython
+
+   In [6]: df.resample('2s')
+   Out[6]:
+                            A         B         C         D
+   2010-01-01 09:00:00  0.485748  0.447351  0.357096  0.793615
+   2010-01-01 09:00:02  0.820801  0.794317  0.364034  0.531096
+   2010-01-01 09:00:04  0.433985  0.314582  0.424104  0.625733
+   2010-01-01 09:00:06  0.624988  0.609738  0.633165  0.612452
+   2010-01-01 09:00:08  0.510470  0.534317  0.573201  0.806949
+
+You could also specify a ``how`` directly
+
+.. code-block:: ipython
+
+   In [7]: df.resample('2s', how='sum')
+   Out[7]:
+                            A         B         C         D
+   2010-01-01 09:00:00  0.971495  0.894701  0.714192  1.587231
+   2010-01-01 09:00:02  1.641602  1.588635  0.728068  1.062191
+   2010-01-01 09:00:04  0.867969  0.629165  0.848208  1.251465
+   2010-01-01 09:00:06  1.249976  1.219477  1.266330  1.224904
+   2010-01-01 09:00:08  1.020940  1.068634  1.146402  1.613897
+
+**New API**:
+
+Now, you can write ``.resample(..)`` as a 2-stage operation like ``.groupby(...)``, which
+yields a ``Resampler``.
+
+.. ipython:: python
+   :okwarning:
+
+   r = df.resample('2s')
+   r
+
+Downsampling
+""""""""""""
+
+You can then use this object to perform operations.
+These are downsampling operations (going from a higher frequency to a lower one).
+
+.. ipython:: python
+
+   r.mean()
+
+.. ipython:: python
+
+   r.sum()
+
+Furthermore, resample now supports ``getitem`` operations to perform the resample on specific columns.
+
+.. ipython:: python
+
+   r[['A','C']].mean()
+
+and ``.aggregate`` type operations.
+
+.. ipython:: python
+
+   r.agg({'A' : 'mean', 'B' : 'sum'})
+
+These accessors can of course, be combined
+
+.. ipython:: python
+
+   r[['A','B']].agg(['mean','sum'])
+
+Upsampling
+""""""""""
+
+.. currentmodule:: pandas.tseries.resample
+
+Upsampling operations take you from a lower frequency to a higher frequency. These are now
+performed with the ``Resampler`` objects with :meth:`~Resampler.backfill`,
+:meth:`~Resampler.ffill`, :meth:`~Resampler.fillna` and :meth:`~Resampler.asfreq` methods.
+
+.. code-block:: ipython
+
+   In [89]: s = pd.Series(np.arange(5, dtype='int64'),
+                 index=pd.date_range('2010-01-01', periods=5, freq='Q'))
+
+   In [90]: s
+   Out[90]:
+   2010-03-31    0
+   2010-06-30    1
+   2010-09-30    2
+   2010-12-31    3
+   2011-03-31    4
+   Freq: Q-DEC, Length: 5, dtype: int64
+
+Previously
+
+.. code-block:: ipython
+
+   In [6]: s.resample('M', fill_method='ffill')
+   Out[6]:
+   2010-03-31    0
+   2010-04-30    0
+   2010-05-31    0
+   2010-06-30    1
+   2010-07-31    1
+   2010-08-31    1
+   2010-09-30    2
+   2010-10-31    2
+   2010-11-30    2
+   2010-12-31    3
+   2011-01-31    3
+   2011-02-28    3
+   2011-03-31    4
+   Freq: M, dtype: int64
+
+New API
+
+.. code-block:: ipython
+
+   In [91]: s.resample('M').ffill()
+   Out[91]:
+   2010-03-31    0
+   2010-04-30    0
+   2010-05-31    0
+   2010-06-30    1
+   2010-07-31    1
+   2010-08-31    1
+   2010-09-30    2
+   2010-10-31    2
+   2010-11-30    2
+   2010-12-31    3
+   2011-01-31    3
+   2011-02-28    3
+   2011-03-31    4
+   Freq: M, Length: 13, dtype: int64
+
+.. note::
+
+   In the new API, you can either downsample OR upsample. The prior implementation would allow you to pass an aggregator function (like ``mean``) even though you were upsampling, providing a bit of confusion.
+
+Previous API will work but with deprecations
+""""""""""""""""""""""""""""""""""""""""""""
+
+.. warning::
+
+   This new API for resample includes some internal changes for the prior-to-0.18.0 API, to work with a deprecation warning in most cases, as the resample operation returns a deferred object. We can intercept operations and just do what the (pre 0.18.0) API did (with a warning). Here is a typical use case:
+
+   .. code-block:: ipython
+
+      In [4]: r = df.resample('2s')
+
+      In [6]: r*10
+      pandas/tseries/resample.py:80: FutureWarning: .resample() is now a deferred operation
+      use .resample(...).mean() instead of .resample(...)
+
+      Out[6]:
+                            A         B         C         D
+      2010-01-01 09:00:00  4.857476  4.473507  3.570960  7.936154
+      2010-01-01 09:00:02  8.208011  7.943173  3.640340  5.310957
+      2010-01-01 09:00:04  4.339846  3.145823  4.241039  6.257326
+      2010-01-01 09:00:06  6.249881  6.097384  6.331650  6.124518
+      2010-01-01 09:00:08  5.104699  5.343172  5.732009  8.069486
+
+   However, getting and assignment operations directly on a ``Resampler`` will raise a ``ValueError``:
+
+   .. code-block:: ipython
+
+      In [7]: r.iloc[0] = 5
+      ValueError: .resample() is now a deferred operation
+      use .resample(...).mean() instead of .resample(...)
+
+   There is a situation where the new API can not perform all the operations when using original code.
+   This code is intending to resample every 2s, take the ``mean`` AND then take the ``min`` of those results.
+
+   .. code-block:: ipython
+
+      In [4]: df.resample('2s').min()
+      Out[4]:
+      A    0.433985
+      B    0.314582
+      C    0.357096
+      D    0.531096
+      dtype: float64
+
+   The new API will:
+
+   .. ipython:: python
+
+      df.resample('2s').min()
+
+   The good news is the return dimensions will differ between the new API and the old API, so this should loudly raise
+   an exception.
+
+   To replicate the original operation
+
+   .. ipython:: python
+
+      df.resample('2s').mean().min()
+
+Changes to eval
+^^^^^^^^^^^^^^^
+
+In prior versions, new columns assignments in an ``eval`` expression resulted
+in an inplace change to the ``DataFrame``. (:issue:`9297`, :issue:`8664`, :issue:`10486`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': np.linspace(0, 10, 5), 'b': range(5)})
+   df
+
+.. ipython:: python
+   :suppress:
+
+   df.eval('c = a + b', inplace=True)
+
+.. code-block:: ipython
+
+   In [12]: df.eval('c = a + b')
+   FutureWarning: eval expressions containing an assignment currentlydefault to operating inplace.
+   This will change in a future version of pandas, use inplace=True to avoid this warning.
+
+   In [13]: df
+   Out[13]:
+         a  b     c
+   0   0.0  0   0.0
+   1   2.5  1   3.5
+   2   5.0  2   7.0
+   3   7.5  3  10.5
+   4  10.0  4  14.0
+
+In version 0.18.0, a new ``inplace`` keyword was added to choose whether the
+assignment should be done inplace or return a copy.
+
+.. ipython:: python
+
+   df
+   df.eval('d = c - b', inplace=False)
+   df
+   df.eval('d = c - b', inplace=True)
+   df
+
+.. warning::
+
+   For backwards compatibility, ``inplace`` defaults to ``True`` if not specified.
+   This will change in a future version of pandas. If your code depends on an
+   inplace assignment you should update to explicitly set ``inplace=True``
+
+The ``inplace`` keyword parameter was also added the ``query`` method.
+
+.. ipython:: python
+
+   df.query('a > 5')
+   df.query('a > 5', inplace=True)
+   df
+
+.. warning::
+
+   Note that the default value for ``inplace`` in a ``query``
+   is ``False``, which is consistent with prior versions.
+
+``eval`` has also been updated to allow multi-line expressions for multiple
+assignments.  These expressions will be evaluated one at a time in order.  Only
+assignments are valid for multi-line expressions.
+
+.. ipython:: python
+
+   df
+   df.eval("""
+   e = d + a
+   f = e - 22
+   g = f / 2.0""", inplace=True)
+   df
+
+
+.. _whatsnew_0180.api:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- ``DataFrame.between_time`` and ``Series.between_time`` now only parse a fixed set of time strings. Parsing of date strings is no longer supported and raises a ``ValueError``. (:issue:`11818`)
+
+  .. code-block:: ipython
+
+     In [107]: s = pd.Series(range(10), pd.date_range('2015-01-01', freq='H', periods=10))
+
+     In [108]: s.between_time("7:00am", "9:00am")
+     Out[108]:
+     2015-01-01 07:00:00    7
+     2015-01-01 08:00:00    8
+     2015-01-01 09:00:00    9
+     Freq: H, Length: 3, dtype: int64
+
+  This will now raise.
+
+  .. code-block:: ipython
+
+     In [2]: s.between_time('20150101 07:00:00','20150101 09:00:00')
+     ValueError: Cannot convert arg ['20150101 07:00:00'] to a time.
+
+- ``.memory_usage()`` now includes values in the index, as does memory_usage in ``.info()`` (:issue:`11597`)
+- ``DataFrame.to_latex()`` now supports non-ascii encodings (eg ``utf-8``) in Python 2 with the parameter ``encoding`` (:issue:`7061`)
+- ``pandas.merge()`` and ``DataFrame.merge()`` will show a specific error message when trying to merge with an object that is not of type ``DataFrame`` or a subclass (:issue:`12081`)
+- ``DataFrame.unstack`` and ``Series.unstack`` now take ``fill_value`` keyword to allow direct replacement of missing values when an unstack results in missing values in the resulting ``DataFrame``. As an added benefit, specifying ``fill_value`` will preserve the data type of the original stacked data.  (:issue:`9746`)
+- As part of the new API for :ref:`window functions <whatsnew_0180.enhancements.moments>` and :ref:`resampling <whatsnew_0180.breaking.resample>`, aggregation functions have been clarified, raising more informative error messages on invalid aggregations. (:issue:`9052`). A full set of examples are presented in :ref:`groupby <groupby.aggregate>`.
+- Statistical functions for ``NDFrame`` objects (like ``sum(), mean(), min()``) will now raise if non-numpy-compatible arguments are passed in for ``**kwargs`` (:issue:`12301`)
+- ``.to_latex`` and ``.to_html`` gain a ``decimal`` parameter like ``.to_csv``; the default is ``'.'`` (:issue:`12031`)
+- More helpful error message when constructing a ``DataFrame`` with empty data but with indices (:issue:`8020`)
+- ``.describe()`` will now properly handle bool dtype as a categorical (:issue:`6625`)
+- More helpful error message with an invalid ``.transform`` with user defined input (:issue:`10165`)
+- Exponentially weighted functions now allow specifying alpha directly (:issue:`10789`) and raise ``ValueError`` if parameters violate ``0 < alpha <= 1`` (:issue:`12492`)
+
+.. _whatsnew_0180.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+.. _whatsnew_0180.window_deprecations:
+
+- The functions ``pd.rolling_*``, ``pd.expanding_*``, and ``pd.ewm*`` are deprecated and replaced by the corresponding method call. Note that
+  the new suggested syntax includes all of the arguments (even if default) (:issue:`11603`)
+
+  .. code-block:: ipython
+
+     In [1]: s = pd.Series(range(3))
+
+     In [2]: pd.rolling_mean(s,window=2,min_periods=1)
+             FutureWarning: pd.rolling_mean is deprecated for Series and
+                  will be removed in a future version, replace with
+                  Series.rolling(min_periods=1,window=2,center=False).mean()
+     Out[2]:
+             0    0.0
+             1    0.5
+             2    1.5
+             dtype: float64
+
+     In [3]: pd.rolling_cov(s, s, window=2)
+             FutureWarning: pd.rolling_cov is deprecated for Series and
+                  will be removed in a future version, replace with
+                  Series.rolling(window=2).cov(other=<Series>)
+     Out[3]:
+             0    NaN
+             1    0.5
+             2    0.5
+             dtype: float64
+
+- The ``freq`` and ``how`` arguments to the ``.rolling``, ``.expanding``, and ``.ewm`` (new) functions are deprecated, and will be removed in a future version. You can simply resample the input prior to creating a window function. (:issue:`11603`).
+
+  For example, instead of ``s.rolling(window=5,freq='D').max()`` to get the max value on a rolling 5 Day window, one could use ``s.resample('D').mean().rolling(window=5).max()``, which first resamples the data to daily data, then provides a rolling 5 day window.
+
+- ``pd.tseries.frequencies.get_offset_name`` function is deprecated. Use offset's ``.freqstr`` property as alternative (:issue:`11192`)
+- ``pandas.stats.fama_macbeth`` routines are deprecated and will be removed in a future version (:issue:`6077`)
+- ``pandas.stats.ols``, ``pandas.stats.plm`` and ``pandas.stats.var`` routines are deprecated and will be removed in a future version (:issue:`6077`)
+- show a ``FutureWarning`` rather than a ``DeprecationWarning`` on using long-time deprecated syntax in ``HDFStore.select``, where the ``where`` clause is not a string-like (:issue:`12027`)
+
+- The ``pandas.options.display.mpl_style`` configuration has been deprecated
+  and will be removed in a future version of pandas. This functionality
+  is better handled by matplotlib's `style sheets`_ (:issue:`11783`).
+
+
+.. _style sheets: http://matplotlib.org/users/style_sheets.html
+
+.. _whatsnew_0180.float_indexers:
+
+Removal of deprecated float indexers
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In :issue:`4892` indexing with floating point numbers on a non-``Float64Index`` was deprecated (in version 0.14.0).
+In 0.18.0, this deprecation warning is removed and these will now raise a ``TypeError``. (:issue:`12165`, :issue:`12333`)
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=[4, 5, 6])
+   s
+   s2 = pd.Series([1, 2, 3], index=list('abc'))
+   s2
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   # this is label indexing
+   In [2]: s[5.0]
+   FutureWarning: scalar indexers for index type Int64Index should be integers and not floating point
+   Out[2]: 2
+
+   # this is positional indexing
+   In [3]: s.iloc[1.0]
+   FutureWarning: scalar indexers for index type Int64Index should be integers and not floating point
+   Out[3]: 2
+
+   # this is label indexing
+   In [4]: s.loc[5.0]
+   FutureWarning: scalar indexers for index type Int64Index should be integers and not floating point
+   Out[4]: 2
+
+   # .ix would coerce 1.0 to the positional 1, and index
+   In [5]: s2.ix[1.0] = 10
+   FutureWarning: scalar indexers for index type Index should be integers and not floating point
+
+   In [6]: s2
+   Out[6]:
+   a     1
+   b    10
+   c     3
+   dtype: int64
+
+New behavior:
+
+For iloc, getting & setting via a float scalar will always raise.
+
+.. code-block:: ipython
+
+   In [3]: s.iloc[2.0]
+   TypeError: cannot do label indexing on <class 'pandas.indexes.numeric.Int64Index'> with these indexers [2.0] of <type 'float'>
+
+Other indexers will coerce to a like integer for both getting and setting. The ``FutureWarning`` has been dropped for ``.loc``, ``.ix`` and ``[]``.
+
+.. ipython:: python
+
+   s[5.0]
+   s.loc[5.0]
+
+and setting
+
+.. ipython:: python
+
+   s_copy = s.copy()
+   s_copy[5.0] = 10
+   s_copy
+   s_copy = s.copy()
+   s_copy.loc[5.0] = 10
+   s_copy
+
+Positional setting with ``.ix`` and a float indexer will ADD this value to the index, rather than previously setting the value by position.
+
+.. code-block:: ipython
+
+   In [3]: s2.ix[1.0] = 10
+   In [4]: s2
+   Out[4]:
+   a       1
+   b       2
+   c       3
+   1.0    10
+   dtype: int64
+
+Slicing will also coerce integer-like floats to integers for a non-``Float64Index``.
+
+.. ipython:: python
+
+   s.loc[5.0:6]
+
+Note that for floats that are NOT coercible to ints, the label based bounds will be excluded
+
+.. ipython:: python
+
+   s.loc[5.1:6]
+
+Float indexing on a ``Float64Index`` is unchanged.
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=np.arange(3.))
+   s[1.0]
+   s[1.0:2.5]
+
+.. _whatsnew_0180.prior_deprecations:
+
+Removal of prior version deprecations/changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Removal of ``rolling_corr_pairwise`` in favor of ``.rolling().corr(pairwise=True)`` (:issue:`4950`)
+- Removal of ``expanding_corr_pairwise`` in favor of ``.expanding().corr(pairwise=True)`` (:issue:`4950`)
+- Removal of ``DataMatrix`` module. This was not imported into the pandas namespace in any event (:issue:`12111`)
+- Removal of ``cols`` keyword in favor of ``subset`` in ``DataFrame.duplicated()`` and ``DataFrame.drop_duplicates()`` (:issue:`6680`)
+- Removal of the ``read_frame`` and ``frame_query`` (both aliases for ``pd.read_sql``)
+  and ``write_frame`` (alias of ``to_sql``) functions in the ``pd.io.sql`` namespace,
+  deprecated since 0.14.0 (:issue:`6292`).
+- Removal of the ``order`` keyword from ``.factorize()`` (:issue:`6930`)
+
+.. _whatsnew_0180.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of ``andrews_curves`` (:issue:`11534`)
+- Improved huge ``DatetimeIndex``, ``PeriodIndex`` and ``TimedeltaIndex``'s ops performance including ``NaT`` (:issue:`10277`)
+- Improved performance of ``pandas.concat`` (:issue:`11958`)
+- Improved performance of ``StataReader`` (:issue:`11591`)
+- Improved performance in construction of ``Categoricals`` with ``Series`` of datetimes containing ``NaT`` (:issue:`12077`)
+
+
+- Improved performance of ISO 8601 date parsing for dates without separators (:issue:`11899`), leading zeros (:issue:`11871`) and with white space preceding the time zone (:issue:`9714`)
+
+
+
+
+.. _whatsnew_0180.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in ``GroupBy.size`` when data-frame is empty. (:issue:`11699`)
+- Bug in ``Period.end_time`` when a multiple of time period is requested (:issue:`11738`)
+- Regression in ``.clip`` with tz-aware datetimes (:issue:`11838`)
+- Bug in ``date_range`` when the boundaries fell on the frequency (:issue:`11804`, :issue:`12409`)
+- Bug in consistency of passing nested dicts to ``.groupby(...).agg(...)`` (:issue:`9052`)
+- Accept unicode in ``Timedelta`` constructor (:issue:`11995`)
+- Bug in value label reading for ``StataReader`` when reading incrementally (:issue:`12014`)
+- Bug in vectorized ``DateOffset`` when ``n`` parameter is ``0`` (:issue:`11370`)
+- Compat for numpy 1.11 w.r.t. ``NaT`` comparison changes (:issue:`12049`)
+- Bug in ``read_csv`` when reading from a ``StringIO`` in threads (:issue:`11790`)
+- Bug in not treating ``NaT`` as a missing value in datetimelikes when factorizing & with ``Categoricals`` (:issue:`12077`)
+- Bug in getitem when the values of a ``Series`` were tz-aware (:issue:`12089`)
+- Bug in ``Series.str.get_dummies`` when one of the variables was 'name' (:issue:`12180`)
+- Bug in ``pd.concat`` while concatenating tz-aware NaT series. (:issue:`11693`, :issue:`11755`, :issue:`12217`)
+- Bug in ``pd.read_stata`` with version <= 108 files (:issue:`12232`)
+- Bug in ``Series.resample`` using a frequency of ``Nano`` when the index is a ``DatetimeIndex`` and contains non-zero nanosecond parts (:issue:`12037`)
+- Bug in resampling with ``.nunique`` and a sparse index (:issue:`12352`)
+- Removed some compiler warnings (:issue:`12471`)
+- Work around compat issues with ``boto`` in python 3.5 (:issue:`11915`)
+- Bug in ``NaT`` subtraction from ``Timestamp`` or ``DatetimeIndex`` with timezones (:issue:`11718`)
+- Bug in subtraction of ``Series`` of a single tz-aware ``Timestamp`` (:issue:`12290`)
+- Use compat iterators in PY2 to support ``.next()`` (:issue:`12299`)
+- Bug in ``Timedelta.round`` with negative values (:issue:`11690`)
+- Bug in ``.loc`` against ``CategoricalIndex`` may result in normal ``Index`` (:issue:`11586`)
+- Bug in ``DataFrame.info`` when duplicated column names exist (:issue:`11761`)
+- Bug in ``.copy`` of datetime tz-aware objects (:issue:`11794`)
+- Bug in ``Series.apply`` and ``Series.map`` where ``timedelta64`` was not boxed (:issue:`11349`)
+- Bug in ``DataFrame.set_index()`` with tz-aware ``Series`` (:issue:`12358`)
+
+
+
+- Bug in subclasses of ``DataFrame`` where ``AttributeError`` did not propagate (:issue:`11808`)
+- Bug groupby on tz-aware data where selection not returning ``Timestamp`` (:issue:`11616`)
+- Bug in ``pd.read_clipboard`` and ``pd.to_clipboard`` functions not supporting Unicode; upgrade included ``pyperclip`` to v1.5.15 (:issue:`9263`)
+- Bug in ``DataFrame.query`` containing an assignment (:issue:`8664`)
+
+- Bug in ``from_msgpack`` where ``__contains__()`` fails for columns of the unpacked ``DataFrame``, if the ``DataFrame`` has object columns. (:issue:`11880`)
+- Bug in ``.resample`` on categorical data with ``TimedeltaIndex`` (:issue:`12169`)
+
+
+- Bug in timezone info lost when broadcasting scalar datetime to ``DataFrame`` (:issue:`11682`)
+- Bug in ``Index`` creation from ``Timestamp`` with mixed tz coerces to UTC (:issue:`11488`)
+- Bug in ``to_numeric`` where it does not raise if input is more than one dimension (:issue:`11776`)
+- Bug in parsing timezone offset strings with non-zero minutes (:issue:`11708`)
+- Bug in ``df.plot`` using incorrect colors for bar plots under matplotlib 1.5+ (:issue:`11614`)
+- Bug in the ``groupby`` ``plot`` method when using keyword arguments (:issue:`11805`).
+- Bug in ``DataFrame.duplicated`` and ``drop_duplicates`` causing spurious matches when setting ``keep=False`` (:issue:`11864`)
+- Bug in ``.loc`` result with duplicated key may have ``Index`` with incorrect dtype (:issue:`11497`)
+- Bug in ``pd.rolling_median`` where memory allocation failed even with sufficient memory (:issue:`11696`)
+- Bug in ``DataFrame.style`` with spurious zeros (:issue:`12134`)
+- Bug in ``DataFrame.style`` with integer columns not starting at 0 (:issue:`12125`)
+- Bug in ``.style.bar`` may not rendered properly using specific browser (:issue:`11678`)
+- Bug in rich comparison of ``Timedelta`` with a ``numpy.array`` of ``Timedelta`` that caused an infinite recursion (:issue:`11835`)
+- Bug in ``DataFrame.round`` dropping column index name (:issue:`11986`)
+- Bug in ``df.replace`` while replacing value in mixed dtype ``Dataframe`` (:issue:`11698`)
+- Bug in ``Index`` prevents copying name of passed ``Index``, when a new name is not provided (:issue:`11193`)
+- Bug in ``read_excel`` failing to read any non-empty sheets when empty sheets exist and ``sheetname=None`` (:issue:`11711`)
+- Bug in ``read_excel`` failing to raise ``NotImplemented`` error when keywords ``parse_dates`` and ``date_parser`` are provided (:issue:`11544`)
+- Bug in ``read_sql`` with ``pymysql`` connections failing to return chunked data (:issue:`11522`)
+- Bug in ``.to_csv`` ignoring formatting parameters ``decimal``, ``na_rep``, ``float_format`` for float indexes (:issue:`11553`)
+- Bug in ``Int64Index`` and ``Float64Index`` preventing the use of the modulo operator (:issue:`9244`)
+- Bug in ``MultiIndex.drop`` for not lexsorted MultiIndexes (:issue:`12078`)
+
+- Bug in ``DataFrame`` when masking an empty ``DataFrame`` (:issue:`11859`)
+
+
+- Bug in ``.plot`` potentially modifying the ``colors`` input when the number of columns didn't match the number of series provided (:issue:`12039`).
+- Bug in ``Series.plot`` failing when index has a ``CustomBusinessDay`` frequency (:issue:`7222`).
+- Bug in ``.to_sql`` for ``datetime.time`` values with sqlite fallback (:issue:`8341`)
+- Bug in ``read_excel`` failing to read data with one column when ``squeeze=True`` (:issue:`12157`)
+- Bug in ``read_excel`` failing to read one empty column (:issue:`12292`, :issue:`9002`)
+- Bug in ``.groupby`` where a ``KeyError`` was not raised for a wrong column if there was only one row in the dataframe (:issue:`11741`)
+- Bug in ``.read_csv`` with dtype specified on empty data producing an error (:issue:`12048`)
+- Bug in ``.read_csv`` where strings like ``'2E'`` are treated as valid floats (:issue:`12237`)
+- Bug in building *pandas* with debugging symbols (:issue:`12123`)
+
+
+- Removed ``millisecond`` property of ``DatetimeIndex``. This would always raise a ``ValueError`` (:issue:`12019`).
+- Bug in ``Series`` constructor with read-only data (:issue:`11502`)
+- Removed ``pandas._testing.choice()``.  Should use ``np.random.choice()``, instead. (:issue:`12386`)
+- Bug in ``.loc`` setitem indexer preventing the use of a TZ-aware DatetimeIndex (:issue:`12050`)
+- Bug in ``.style`` indexes and MultiIndexes not appearing (:issue:`11655`)
+- Bug in ``to_msgpack`` and ``from_msgpack`` which did not correctly serialize or deserialize ``NaT`` (:issue:`12307`).
+- Bug in ``.skew`` and ``.kurt`` due to roundoff error for highly similar values (:issue:`11974`)
+- Bug in ``Timestamp`` constructor where microsecond resolution was lost if HHMMSS were not separated with ':' (:issue:`10041`)
+- Bug in ``buffer_rd_bytes`` src->buffer could be freed more than once if reading failed, causing a segfault (:issue:`12098`)
+
+- Bug in ``crosstab`` where arguments with non-overlapping indexes would return a ``KeyError`` (:issue:`10291`)
+
+- Bug in ``DataFrame.apply`` in which reduction was not being prevented for cases in which ``dtype`` was not a numpy dtype (:issue:`12244`)
+- Bug when initializing categorical series with a scalar value. (:issue:`12336`)
+- Bug when specifying a UTC ``DatetimeIndex`` by setting ``utc=True`` in ``.to_datetime`` (:issue:`11934`)
+- Bug when increasing the buffer size of CSV reader in ``read_csv`` (:issue:`12494`)
+- Bug when setting columns of a ``DataFrame`` with duplicate column names (:issue:`12344`)
+
+
+.. _whatsnew_0.18.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.17.1..v0.18.0
diff --git a/pandas/doc/source/whatsnew/v0.18.1.rst b/pandas/doc/source/whatsnew/v0.18.1.rst
new file mode 100644
index 0000000000000..85e0e63016729
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.18.1.rst
@@ -0,0 +1,844 @@
+.. _whatsnew_0181:
+
+Version 0.18.1 (May 3, 2016)
+----------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release from 0.18.0 and includes a large number of
+bug fixes along with several new features, enhancements, and performance improvements.
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- ``.groupby(...)`` has been enhanced to provide convenient syntax when working with ``.rolling(..)``, ``.expanding(..)`` and ``.resample(..)`` per group, see :ref:`here <whatsnew_0181.deferred_ops>`
+- ``pd.to_datetime()`` has gained the ability to assemble dates from a ``DataFrame``, see :ref:`here <whatsnew_0181.enhancements.assembling>`
+- Method chaining improvements, see :ref:`here <whatsnew_0181.enhancements.method_chain>`.
+- Custom business hour offset, see :ref:`here <whatsnew_0181.enhancements.custombusinesshour>`.
+- Many bug fixes in the handling of ``sparse``, see :ref:`here <whatsnew_0181.sparse>`
+- Expanded the :ref:`Tutorials section <tutorial-modern>` with a feature on modern pandas, courtesy of `@TomAugsburger <https://twitter.com/TomAugspurger>`__. (:issue:`13045`).
+
+
+.. contents:: What's new in v0.18.1
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0181.new_features:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0181.enhancements.custombusinesshour:
+
+Custom business hour
+^^^^^^^^^^^^^^^^^^^^
+
+The ``CustomBusinessHour`` is a mixture of ``BusinessHour`` and ``CustomBusinessDay`` which
+allows you to specify arbitrary holidays. For details,
+see :ref:`Custom Business Hour <timeseries.custombusinesshour>` (:issue:`11514`)
+
+.. ipython:: python
+
+    from pandas.tseries.offsets import CustomBusinessHour
+    from pandas.tseries.holiday import USFederalHolidayCalendar
+
+    bhour_us = CustomBusinessHour(calendar=USFederalHolidayCalendar())
+
+Friday before MLK Day
+
+.. ipython:: python
+
+    import datetime
+
+    dt = datetime.datetime(2014, 1, 17, 15)
+
+    dt + bhour_us
+
+Tuesday after MLK Day (Monday is skipped because it's a holiday)
+
+.. ipython:: python
+
+    dt + bhour_us * 2
+
+.. _whatsnew_0181.deferred_ops:
+
+Method ``.groupby(..)`` syntax with window and resample operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``.groupby(...)`` has been enhanced to provide convenient syntax when working with ``.rolling(..)``, ``.expanding(..)`` and ``.resample(..)`` per group, see (:issue:`12486`, :issue:`12738`).
+
+You can now use ``.rolling(..)`` and ``.expanding(..)`` as methods on groupbys. These return another deferred object (similar to what ``.rolling()`` and ``.expanding()`` do on ungrouped pandas objects). You can then operate on these ``RollingGroupby`` objects in a similar manner.
+
+Previously you would have to do this to get a rolling window mean per-group:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)})
+   df
+
+.. code-block:: ipython
+
+   In [1]: df.groupby("A").apply(lambda x: x.rolling(4).B.mean())
+   Out[1]:
+   A
+   1  0      NaN
+      1      NaN
+      2      NaN
+      3      1.5
+      4      2.5
+      5      3.5
+      6      4.5
+      7      5.5
+      8      6.5
+      9      7.5
+      10     8.5
+      11     9.5
+      12    10.5
+      13    11.5
+      14    12.5
+      15    13.5
+      16    14.5
+      17    15.5
+      18    16.5
+      19    17.5
+   2  20     NaN
+      21     NaN
+      22     NaN
+      23    21.5
+      24    22.5
+      25    23.5
+      26    24.5
+      27    25.5
+      28    26.5
+      29    27.5
+      30    28.5
+      31    29.5
+   3  32     NaN
+      33     NaN
+      34     NaN
+      35    33.5
+      36    34.5
+      37    35.5
+      38    36.5
+      39    37.5
+   Name: B, dtype: float64
+
+Now you can do:
+
+.. ipython:: python
+
+   df.groupby("A").rolling(4).B.mean()
+
+For ``.resample(..)`` type of operations, previously you would have to:
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "date": pd.date_range(start="2016-01-01", periods=4, freq="W"),
+           "group": [1, 1, 2, 2],
+           "val": [5, 6, 7, 8],
+       }
+   ).set_index("date")
+
+   df
+
+.. code-block:: ipython
+
+   In[1]: df.groupby("group").apply(lambda x: x.resample("1D").ffill())
+   Out[1]:
+                     group  val
+   group date
+   1     2016-01-03      1    5
+         2016-01-04      1    5
+         2016-01-05      1    5
+         2016-01-06      1    5
+         2016-01-07      1    5
+         2016-01-08      1    5
+         2016-01-09      1    5
+         2016-01-10      1    6
+   2     2016-01-17      2    7
+         2016-01-18      2    7
+         2016-01-19      2    7
+         2016-01-20      2    7
+         2016-01-21      2    7
+         2016-01-22      2    7
+         2016-01-23      2    7
+         2016-01-24      2    8
+
+Now you can do:
+
+.. code-block:: ipython
+
+   In[1]: df.groupby("group").resample("1D").ffill()
+   Out[1]:
+                     group  val
+   group date
+   1     2016-01-03      1    5
+         2016-01-04      1    5
+         2016-01-05      1    5
+         2016-01-06      1    5
+         2016-01-07      1    5
+         2016-01-08      1    5
+         2016-01-09      1    5
+         2016-01-10      1    6
+   2     2016-01-17      2    7
+         2016-01-18      2    7
+         2016-01-19      2    7
+         2016-01-20      2    7
+         2016-01-21      2    7
+         2016-01-22      2    7
+         2016-01-23      2    7
+         2016-01-24      2    8
+
+.. _whatsnew_0181.enhancements.method_chain:
+
+Method chaining improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The following methods / indexers now accept a ``callable``. It is intended to make
+these more useful in method chains, see the :ref:`documentation <indexing.callable>`.
+(:issue:`11485`, :issue:`12533`)
+
+- ``.where()`` and ``.mask()``
+- ``.loc[]``, ``iloc[]`` and ``.ix[]``
+- ``[]`` indexing
+
+Methods ``.where()`` and ``.mask()``
+""""""""""""""""""""""""""""""""""""
+
+These can accept a callable for the condition and ``other``
+arguments.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+   df.where(lambda x: x > 4, lambda x: x + 10)
+
+Methods ``.loc[]``, ``.iloc[]``, ``.ix[]``
+""""""""""""""""""""""""""""""""""""""""""
+
+These can accept a callable, and a tuple of callable as a slicer. The callable
+can return a valid boolean indexer or anything which is valid for these indexer's input.
+
+.. ipython:: python
+
+   # callable returns bool indexer
+   df.loc[lambda x: x.A >= 2, lambda x: x.sum() > 10]
+
+   # callable returns list of labels
+   df.loc[lambda x: [1, 2], lambda x: ["A", "B"]]
+
+Indexing with ``[]``
+""""""""""""""""""""
+
+Finally, you can use a callable in ``[]`` indexing of Series, DataFrame and Panel.
+The callable must return a valid input for ``[]`` indexing depending on its
+class and index type.
+
+.. ipython:: python
+
+   df[lambda x: "A"]
+
+Using these methods / indexers, you can chain data selection operations
+without using temporary variable.
+
+.. ipython:: python
+
+   bb = pd.read_csv("data/baseball.csv", index_col="id")
+   (bb.groupby(["year", "team"]).sum(numeric_only=True).loc[lambda df: df.r > 100])
+
+.. _whatsnew_0181.partial_string_indexing:
+
+Partial string indexing on ``DatetimeIndex`` when part of a ``MultiIndex``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Partial string indexing now matches on ``DateTimeIndex`` when part of a ``MultiIndex`` (:issue:`10331`)
+
+.. code-block:: ipython
+
+   In [20]: dft2 = pd.DataFrame(
+      ....:     np.random.randn(20, 1),
+      ....:     columns=["A"],
+      ....:     index=pd.MultiIndex.from_product(
+      ....:         [pd.date_range("20130101", periods=10, freq="12H"), ["a", "b"]]
+      ....:     ),
+      ....: )
+      ....:
+
+   In [21]: dft2
+   Out[21]:
+                                 A
+   2013-01-01 00:00:00 a  0.469112
+                       b -0.282863
+   2013-01-01 12:00:00 a -1.509059
+                       b -1.135632
+   2013-01-02 00:00:00 a  1.212112
+   ...                         ...
+   2013-01-04 12:00:00 b  0.271860
+   2013-01-05 00:00:00 a -0.424972
+                       b  0.567020
+   2013-01-05 12:00:00 a  0.276232
+                       b -1.087401
+
+   [20 rows x 1 columns]
+
+   In [22]: dft2.loc["2013-01-05"]
+   Out[22]:
+                                 A
+   2013-01-05 00:00:00 a -0.424972
+                       b  0.567020
+   2013-01-05 12:00:00 a  0.276232
+                       b -1.087401
+
+   [4 rows x 1 columns]
+
+On other levels
+
+.. code-block:: ipython
+
+   In [26]: idx = pd.IndexSlice
+
+   In [27]: dft2 = dft2.swaplevel(0, 1).sort_index()
+
+   In [28]: dft2
+   Out[28]:
+                                 A
+   a 2013-01-01 00:00:00  0.469112
+     2013-01-01 12:00:00 -1.509059
+     2013-01-02 00:00:00  1.212112
+     2013-01-02 12:00:00  0.119209
+     2013-01-03 00:00:00 -0.861849
+   ...                         ...
+   b 2013-01-03 12:00:00  1.071804
+     2013-01-04 00:00:00 -0.706771
+     2013-01-04 12:00:00  0.271860
+     2013-01-05 00:00:00  0.567020
+     2013-01-05 12:00:00 -1.087401
+
+   [20 rows x 1 columns]
+
+   In [29]: dft2.loc[idx[:, "2013-01-05"], :]
+   Out[29]:
+                                 A
+   a 2013-01-05 00:00:00 -0.424972
+     2013-01-05 12:00:00  0.276232
+   b 2013-01-05 00:00:00  0.567020
+     2013-01-05 12:00:00 -1.087401
+
+   [4 rows x 1 columns]
+
+.. _whatsnew_0181.enhancements.assembling:
+
+Assembling datetimes
+^^^^^^^^^^^^^^^^^^^^
+
+``pd.to_datetime()`` has gained the ability to assemble datetimes from a passed in ``DataFrame`` or a dict. (:issue:`8158`).
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {"year": [2015, 2016], "month": [2, 3], "day": [4, 5], "hour": [2, 3]}
+   )
+   df
+
+Assembling using the passed frame.
+
+.. ipython:: python
+
+   pd.to_datetime(df)
+
+You can pass only the columns that you need to assemble.
+
+.. ipython:: python
+
+   pd.to_datetime(df[["year", "month", "day"]])
+
+.. _whatsnew_0181.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- ``pd.read_csv()`` now supports ``delim_whitespace=True`` for the Python engine (:issue:`12958`)
+- ``pd.read_csv()`` now supports opening ZIP files that contains a single CSV, via extension inference or explicit ``compression='zip'`` (:issue:`12175`)
+- ``pd.read_csv()`` now supports opening files using xz compression, via extension inference or explicit ``compression='xz'`` is specified; ``xz`` compressions is also supported by ``DataFrame.to_csv`` in the same way (:issue:`11852`)
+- ``pd.read_msgpack()`` now always gives writeable ndarrays even when compression is used (:issue:`12359`).
+- ``pd.read_msgpack()`` now supports serializing and de-serializing categoricals with msgpack (:issue:`12573`)
+- ``.to_json()`` now supports ``NDFrames`` that contain categorical and sparse data (:issue:`10778`)
+- ``interpolate()`` now supports ``method='akima'`` (:issue:`7588`).
+- ``pd.read_excel()`` now accepts path objects (e.g. ``pathlib.Path``, ``py.path.local``) for the file path, in line with other ``read_*`` functions (:issue:`12655`)
+- Added ``.weekday_name`` property as a component to ``DatetimeIndex`` and the ``.dt`` accessor. (:issue:`11128`)
+
+- ``Index.take`` now handles ``allow_fill`` and ``fill_value`` consistently (:issue:`12631`)
+
+  .. ipython:: python
+
+     idx = pd.Index([1.0, 2.0, 3.0, 4.0], dtype="float")
+
+     # default, allow_fill=True, fill_value=None
+     idx.take([2, -1])
+     idx.take([2, -1], fill_value=True)
+
+- ``Index`` now supports ``.str.get_dummies()`` which returns ``MultiIndex``, see :ref:`Creating Indicator Variables <text.indicator>` (:issue:`10008`, :issue:`10103`)
+
+  .. ipython:: python
+
+     idx = pd.Index(["a|b", "a|c", "b|c"])
+     idx.str.get_dummies("|")
+
+
+- ``pd.crosstab()`` has gained a ``normalize`` argument for normalizing frequency tables (:issue:`12569`). Examples in the updated docs :ref:`here <reshaping.crosstabulations>`.
+- ``.resample(..).interpolate()`` is now supported (:issue:`12925`)
+- ``.isin()`` now accepts passed ``sets`` (:issue:`12988`)
+
+.. _whatsnew_0181.sparse:
+
+Sparse changes
+~~~~~~~~~~~~~~
+
+These changes conform sparse handling to return the correct types and work to make a smoother experience with indexing.
+
+``SparseArray.take`` now returns a scalar for scalar input, ``SparseArray`` for others. Furthermore, it handles a negative indexer with the same rule as ``Index`` (:issue:`10560`, :issue:`12796`)
+
+.. code-block:: python
+
+   s = pd.SparseArray([np.nan, np.nan, 1, 2, 3, np.nan, 4, 5, np.nan, 6])
+   s.take(0)
+   s.take([1, 2, 3])
+
+- Bug in ``SparseSeries[]`` indexing with ``Ellipsis`` raises ``KeyError`` (:issue:`9467`)
+- Bug in ``SparseArray[]`` indexing with tuples are not handled properly (:issue:`12966`)
+- Bug in ``SparseSeries.loc[]`` with list-like input raises ``TypeError`` (:issue:`10560`)
+- Bug in ``SparseSeries.iloc[]`` with scalar input may raise ``IndexError`` (:issue:`10560`)
+- Bug in ``SparseSeries.loc[]``, ``.iloc[]`` with ``slice`` returns ``SparseArray``, rather than ``SparseSeries`` (:issue:`10560`)
+- Bug in ``SparseDataFrame.loc[]``, ``.iloc[]`` may results in dense ``Series``, rather than ``SparseSeries`` (:issue:`12787`)
+- Bug in ``SparseArray`` addition ignores ``fill_value`` of right hand side (:issue:`12910`)
+- Bug in ``SparseArray`` mod raises ``AttributeError`` (:issue:`12910`)
+- Bug in ``SparseArray`` pow calculates ``1 ** np.nan`` as ``np.nan`` which must be 1 (:issue:`12910`)
+- Bug in ``SparseArray`` comparison output may incorrect result or raise ``ValueError`` (:issue:`12971`)
+- Bug in ``SparseSeries.__repr__`` raises ``TypeError`` when it is longer than ``max_rows`` (:issue:`10560`)
+- Bug in ``SparseSeries.shape`` ignores ``fill_value`` (:issue:`10452`)
+- Bug in ``SparseSeries`` and ``SparseArray`` may have different ``dtype`` from its dense values (:issue:`12908`)
+- Bug in ``SparseSeries.reindex`` incorrectly handle ``fill_value`` (:issue:`12797`)
+- Bug in ``SparseArray.to_frame()`` results in ``DataFrame``, rather than ``SparseDataFrame`` (:issue:`9850`)
+- Bug in ``SparseSeries.value_counts()`` does not count ``fill_value`` (:issue:`6749`)
+- Bug in ``SparseArray.to_dense()`` does not preserve ``dtype`` (:issue:`10648`)
+- Bug in ``SparseArray.to_dense()`` incorrectly handle ``fill_value`` (:issue:`12797`)
+- Bug in ``pd.concat()`` of ``SparseSeries`` results in dense (:issue:`10536`)
+- Bug in ``pd.concat()`` of ``SparseDataFrame`` incorrectly handle ``fill_value`` (:issue:`9765`)
+- Bug in ``pd.concat()`` of ``SparseDataFrame`` may raise ``AttributeError`` (:issue:`12174`)
+- Bug in ``SparseArray.shift()`` may raise ``NameError`` or ``TypeError`` (:issue:`12908`)
+
+.. _whatsnew_0181.api:
+
+API changes
+~~~~~~~~~~~
+
+.. _whatsnew_0181.api.groubynth:
+
+Method ``.groupby(..).nth()`` changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The index in ``.groupby(..).nth()`` output is now more consistent when the ``as_index`` argument is passed (:issue:`11039`):
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": ["a", "b", "a"], "B": [1, 2, 3]})
+   df
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [3]: df.groupby('A', as_index=True)['B'].nth(0)
+   Out[3]:
+   0    1
+   1    2
+   Name: B, dtype: int64
+
+   In [4]: df.groupby('A', as_index=False)['B'].nth(0)
+   Out[4]:
+   0    1
+   1    2
+   Name: B, dtype: int64
+
+New behavior:
+
+.. ipython:: python
+
+    df.groupby("A", as_index=True)["B"].nth(0)
+    df.groupby("A", as_index=False)["B"].nth(0)
+
+Furthermore, previously, a ``.groupby`` would always sort, regardless if ``sort=False`` was passed with ``.nth()``.
+
+.. ipython:: python
+
+   np.random.seed(1234)
+   df = pd.DataFrame(np.random.randn(100, 2), columns=["a", "b"])
+   df["c"] = np.random.randint(0, 4, 100)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [4]: df.groupby('c', sort=True).nth(1)
+   Out[4]:
+             a         b
+   c
+   0 -0.334077  0.002118
+   1  0.036142 -2.074978
+   2 -0.720589  0.887163
+   3  0.859588 -0.636524
+
+   In [5]: df.groupby('c', sort=False).nth(1)
+   Out[5]:
+             a         b
+   c
+   0 -0.334077  0.002118
+   1  0.036142 -2.074978
+   2 -0.720589  0.887163
+   3  0.859588 -0.636524
+
+New behavior:
+
+.. ipython:: python
+
+   df.groupby("c", sort=True).nth(1)
+   df.groupby("c", sort=False).nth(1)
+
+
+.. _whatsnew_0181.numpy_compatibility:
+
+NumPy function compatibility
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Compatibility between pandas array-like methods (e.g. ``sum`` and ``take``) and their ``numpy``
+counterparts has been greatly increased by augmenting the signatures of the ``pandas`` methods so
+as to accept arguments that can be passed in from ``numpy``, even if they are not necessarily
+used in the ``pandas`` implementation (:issue:`12644`, :issue:`12638`, :issue:`12687`)
+
+- ``.searchsorted()`` for ``Index`` and ``TimedeltaIndex`` now accept a ``sorter`` argument to maintain compatibility with numpy's ``searchsorted`` function (:issue:`12238`)
+- Bug in numpy compatibility of ``np.round()`` on a ``Series`` (:issue:`12600`)
+
+An example of this signature augmentation is illustrated below:
+
+.. code-block:: python
+
+   sp = pd.SparseDataFrame([1, 2, 3])
+   sp
+
+Previous behaviour:
+
+.. code-block:: ipython
+
+   In [2]: np.cumsum(sp, axis=0)
+   ...
+   TypeError: cumsum() takes at most 2 arguments (4 given)
+
+New behaviour:
+
+.. code-block:: python
+
+   np.cumsum(sp, axis=0)
+
+.. _whatsnew_0181.apply_resample:
+
+Using ``.apply`` on GroupBy resampling
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Using ``apply`` on resampling groupby operations (using a ``pd.TimeGrouper``) now has the same output types as similar ``apply`` calls on other groupby operations. (:issue:`11742`).
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {"date": pd.to_datetime(["10/10/2000", "11/10/2000"]), "value": [10, 13]}
+    )
+    df
+
+Previous behavior:
+
+.. code-block:: ipython
+
+    In [1]: df.groupby(pd.TimeGrouper(key='date',
+       ...:                           freq='M')).apply(lambda x: x.value.sum())
+    Out[1]:
+    ...
+    TypeError: cannot concatenate a non-NDFrame object
+
+    # Output is a Series
+    In [2]: df.groupby(pd.TimeGrouper(key='date',
+       ...:                           freq='M')).apply(lambda x: x[['value']].sum())
+    Out[2]:
+    date
+    2000-10-31  value    10
+    2000-11-30  value    13
+    dtype: int64
+
+New behavior:
+
+.. code-block:: ipython
+
+    # Output is a Series
+    In [55]: df.groupby(pd.TimeGrouper(key='date',
+        ...:                           freq='M')).apply(lambda x: x.value.sum())
+    Out[55]:
+    date
+    2000-10-31    10
+    2000-11-30    13
+    Freq: M, dtype: int64
+
+    # Output is a DataFrame
+    In [56]: df.groupby(pd.TimeGrouper(key='date',
+        ...:                           freq='M')).apply(lambda x: x[['value']].sum())
+    Out[56]:
+                value
+    date
+    2000-10-31     10
+    2000-11-30     13
+
+.. _whatsnew_0181.read_csv_exceptions:
+
+Changes in ``read_csv`` exceptions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+
+In order to standardize the ``read_csv`` API for both the ``c`` and ``python`` engines, both will now raise an
+``EmptyDataError``, a subclass of ``ValueError``, in response to empty columns or header (:issue:`12493`, :issue:`12506`)
+
+Previous behaviour:
+
+.. code-block:: ipython
+
+   In [1]: import io
+
+   In [2]: df = pd.read_csv(io.StringIO(''), engine='c')
+   ...
+   ValueError: No columns to parse from file
+
+   In [3]: df = pd.read_csv(io.StringIO(''), engine='python')
+   ...
+   StopIteration
+
+New behaviour:
+
+.. code-block:: ipython
+
+   In [1]: df = pd.read_csv(io.StringIO(''), engine='c')
+   ...
+   pandas.io.common.EmptyDataError: No columns to parse from file
+
+   In [2]: df = pd.read_csv(io.StringIO(''), engine='python')
+   ...
+   pandas.io.common.EmptyDataError: No columns to parse from file
+
+In addition to this error change, several others have been made as well:
+
+- ``CParserError`` now sub-classes ``ValueError`` instead of just a ``Exception`` (:issue:`12551`)
+- A ``CParserError`` is now raised instead of a generic ``Exception`` in ``read_csv`` when the ``c`` engine cannot parse a column (:issue:`12506`)
+- A ``ValueError`` is now raised instead of a generic ``Exception`` in ``read_csv`` when the ``c`` engine encounters a ``NaN`` value in an integer column (:issue:`12506`)
+- A ``ValueError`` is now raised instead of a generic ``Exception`` in ``read_csv`` when ``true_values`` is specified, and the ``c`` engine encounters an element in a column containing unencodable bytes (:issue:`12506`)
+- ``pandas.parser.OverflowError`` exception has been removed and has been replaced with Python's built-in ``OverflowError`` exception (:issue:`12506`)
+- ``pd.read_csv()`` no longer allows a combination of strings and integers for the ``usecols`` parameter (:issue:`12678`)
+
+
+.. _whatsnew_0181.api.to_datetime:
+
+Method ``to_datetime`` error changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Bugs in ``pd.to_datetime()`` when passing a ``unit`` with convertible entries and ``errors='coerce'`` or non-convertible with ``errors='ignore'``. Furthermore, an ``OutOfBoundsDateime`` exception will be raised when an out-of-range value is encountered for that unit when ``errors='raise'``. (:issue:`11758`, :issue:`13052`, :issue:`13059`)
+
+Previous behaviour:
+
+.. code-block:: ipython
+
+   In [27]: pd.to_datetime(1420043460, unit='s', errors='coerce')
+   Out[27]: NaT
+
+   In [28]: pd.to_datetime(11111111, unit='D', errors='ignore')
+   OverflowError: Python int too large to convert to C long
+
+   In [29]: pd.to_datetime(11111111, unit='D', errors='raise')
+   OverflowError: Python int too large to convert to C long
+
+New behaviour:
+
+.. code-block:: ipython
+
+   In [2]: pd.to_datetime(1420043460, unit='s', errors='coerce')
+   Out[2]: Timestamp('2014-12-31 16:31:00')
+
+   In [3]: pd.to_datetime(11111111, unit='D', errors='ignore')
+   Out[3]: 11111111
+
+   In [4]: pd.to_datetime(11111111, unit='D', errors='raise')
+   OutOfBoundsDatetime: cannot convert input with unit 'D'
+
+.. _whatsnew_0181.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- ``.swaplevel()`` for ``Series``, ``DataFrame``, ``Panel``, and ``MultiIndex`` now features defaults for its first two parameters ``i`` and ``j`` that swap the two innermost levels of the index. (:issue:`12934`)
+- ``.searchsorted()`` for ``Index`` and ``TimedeltaIndex`` now accept a ``sorter`` argument to maintain compatibility with numpy's ``searchsorted`` function (:issue:`12238`)
+- ``Period`` and ``PeriodIndex`` now raises ``IncompatibleFrequency`` error which inherits ``ValueError`` rather than raw ``ValueError`` (:issue:`12615`)
+- ``Series.apply`` for category dtype now applies the passed function to each of the ``.categories`` (and not the ``.codes``), and returns a ``category`` dtype if possible (:issue:`12473`)
+- ``read_csv`` will now raise a ``TypeError`` if ``parse_dates`` is neither a boolean, list, or dictionary (matches the doc-string) (:issue:`5636`)
+- The default for ``.query()/.eval()`` is now ``engine=None``, which will use ``numexpr`` if it's installed; otherwise it will fallback to the ``python`` engine. This mimics the pre-0.18.1 behavior if ``numexpr`` is installed (and which, previously, if numexpr was not installed, ``.query()/.eval()`` would raise). (:issue:`12749`)
+- ``pd.show_versions()`` now includes ``pandas_datareader`` version (:issue:`12740`)
+- Provide a proper ``__name__`` and ``__qualname__`` attributes for generic functions (:issue:`12021`)
+- ``pd.concat(ignore_index=True)`` now uses ``RangeIndex`` as default (:issue:`12695`)
+- ``pd.merge()`` and ``DataFrame.join()`` will show a ``UserWarning`` when merging/joining a single- with a multi-leveled dataframe (:issue:`9455`, :issue:`12219`)
+- Compat with ``scipy`` > 0.17 for deprecated ``piecewise_polynomial`` interpolation method; support for the replacement ``from_derivatives`` method (:issue:`12887`)
+
+.. _whatsnew_0181.deprecations:
+
+Deprecations
+^^^^^^^^^^^^
+
+- The method name ``Index.sym_diff()`` is deprecated and can be replaced by ``Index.symmetric_difference()`` (:issue:`12591`)
+- The method name ``Categorical.sort()`` is deprecated in favor of ``Categorical.sort_values()`` (:issue:`12882`)
+
+
+
+
+
+
+
+
+.. _whatsnew_0181.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved speed of SAS reader (:issue:`12656`, :issue:`12961`)
+- Performance improvements in ``.groupby(..).cumcount()`` (:issue:`11039`)
+- Improved memory usage in ``pd.read_csv()`` when using ``skiprows=an_integer`` (:issue:`13005`)
+- Improved performance of ``DataFrame.to_sql`` when checking case sensitivity for tables. Now only checks if table has been created correctly when table name is not lower case. (:issue:`12876`)
+- Improved performance of ``Period`` construction and time series plotting (:issue:`12903`, :issue:`11831`).
+- Improved performance of ``.str.encode()`` and ``.str.decode()`` methods (:issue:`13008`)
+- Improved performance of ``to_numeric`` if input is numeric dtype (:issue:`12777`)
+- Improved performance of sparse arithmetic with ``IntIndex`` (:issue:`13036`)
+
+
+
+
+
+
+
+
+.. _whatsnew_0181.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- ``usecols`` parameter in ``pd.read_csv`` is now respected even when the lines of a CSV file are not even (:issue:`12203`)
+- Bug in ``groupby.transform(..)`` when ``axis=1`` is specified with a non-monotonic ordered index (:issue:`12713`)
+- Bug in ``Period`` and ``PeriodIndex`` creation raises ``KeyError`` if ``freq="Minute"`` is specified. Note that "Minute" freq is deprecated in v0.17.0, and recommended to use ``freq="T"`` instead (:issue:`11854`)
+- Bug in ``.resample(...).count()`` with a ``PeriodIndex`` always raising a ``TypeError`` (:issue:`12774`)
+- Bug in ``.resample(...)`` with a ``PeriodIndex`` casting to a ``DatetimeIndex`` when empty (:issue:`12868`)
+- Bug in ``.resample(...)`` with a ``PeriodIndex`` when resampling to an existing frequency (:issue:`12770`)
+- Bug in printing data which contains ``Period`` with different ``freq`` raises ``ValueError`` (:issue:`12615`)
+- Bug in ``Series`` construction with ``Categorical`` and ``dtype='category'`` is specified (:issue:`12574`)
+- Bugs in concatenation with a coercible dtype was too aggressive, resulting in different dtypes in output formatting when an object was longer than ``display.max_rows`` (:issue:`12411`, :issue:`12045`, :issue:`11594`, :issue:`10571`, :issue:`12211`)
+- Bug in ``float_format`` option with option not being validated as a callable. (:issue:`12706`)
+- Bug in ``GroupBy.filter`` when ``dropna=False`` and no groups fulfilled the criteria (:issue:`12768`)
+- Bug in ``__name__`` of ``.cum*`` functions (:issue:`12021`)
+- Bug in ``.astype()`` of a ``Float64Inde/Int64Index`` to an ``Int64Index`` (:issue:`12881`)
+- Bug in round tripping an integer based index in ``.to_json()/.read_json()`` when ``orient='index'`` (the default) (:issue:`12866`)
+- Bug in plotting ``Categorical`` dtypes cause error when attempting stacked bar plot (:issue:`13019`)
+- Compat with >= ``numpy`` 1.11 for ``NaT`` comparisons (:issue:`12969`)
+- Bug in ``.drop()`` with a non-unique ``MultiIndex``. (:issue:`12701`)
+- Bug in ``.concat`` of datetime tz-aware and naive DataFrames (:issue:`12467`)
+- Bug in correctly raising a ``ValueError`` in ``.resample(..).fillna(..)`` when passing a non-string (:issue:`12952`)
+- Bug fixes in various encoding and header processing issues in ``pd.read_sas()`` (:issue:`12659`, :issue:`12654`, :issue:`12647`, :issue:`12809`)
+- Bug in ``pd.crosstab()`` where would silently ignore ``aggfunc`` if ``values=None`` (:issue:`12569`).
+- Potential segfault in ``DataFrame.to_json`` when serialising ``datetime.time`` (:issue:`11473`).
+- Potential segfault in ``DataFrame.to_json`` when attempting to serialise 0d array (:issue:`11299`).
+- Segfault in ``to_json`` when attempting to serialise a ``DataFrame`` or ``Series`` with non-ndarray values; now supports serialization of ``category``, ``sparse``, and ``datetime64[ns, tz]`` dtypes (:issue:`10778`).
+- Bug in ``DataFrame.to_json`` with unsupported dtype not passed to default handler (:issue:`12554`).
+- Bug in ``.align`` not returning the sub-class (:issue:`12983`)
+- Bug in aligning a ``Series`` with a ``DataFrame`` (:issue:`13037`)
+- Bug in ``ABCPanel`` in which ``Panel4D`` was not being considered as a valid instance of this generic type (:issue:`12810`)
+
+
+- Bug in consistency of ``.name`` on ``.groupby(..).apply(..)`` cases (:issue:`12363`)
+
+- Bug in ``Timestamp.__repr__`` that caused ``pprint`` to fail in nested structures (:issue:`12622`)
+- Bug in ``Timedelta.min`` and ``Timedelta.max``, the properties now report the true minimum/maximum ``timedeltas`` as recognized by pandas. See the :ref:`documentation <timedeltas.limitations>`. (:issue:`12727`)
+- Bug in ``.quantile()`` with interpolation may coerce to ``float`` unexpectedly (:issue:`12772`)
+- Bug in ``.quantile()`` with empty ``Series`` may return scalar rather than empty ``Series`` (:issue:`12772`)
+
+
+- Bug in ``.loc`` with out-of-bounds in a large indexer would raise ``IndexError`` rather than ``KeyError`` (:issue:`12527`)
+- Bug in resampling when using a ``TimedeltaIndex`` and ``.asfreq()``, would previously not include the final fencepost (:issue:`12926`)
+
+- Bug in equality testing with a ``Categorical`` in a ``DataFrame`` (:issue:`12564`)
+- Bug in ``GroupBy.first()``, ``.last()`` returns incorrect row when ``TimeGrouper`` is used (:issue:`7453`)
+
+
+
+- Bug in ``pd.read_csv()`` with the ``c`` engine when specifying ``skiprows`` with newlines in quoted items (:issue:`10911`, :issue:`12775`)
+- Bug in ``DataFrame`` timezone lost when assigning tz-aware datetime ``Series`` with alignment (:issue:`12981`)
+
+
+
+
+- Bug in ``.value_counts()`` when ``normalize=True`` and ``dropna=True`` where nulls still contributed to the normalized count (:issue:`12558`)
+- Bug in ``Series.value_counts()`` loses name if its dtype is ``category`` (:issue:`12835`)
+- Bug in ``Series.value_counts()`` loses timezone info (:issue:`12835`)
+- Bug in ``Series.value_counts(normalize=True)`` with ``Categorical`` raises ``UnboundLocalError`` (:issue:`12835`)
+- Bug in ``Panel.fillna()`` ignoring ``inplace=True`` (:issue:`12633`)
+- Bug in ``pd.read_csv()`` when specifying ``names``, ``usecols``, and ``parse_dates`` simultaneously with the ``c`` engine (:issue:`9755`)
+- Bug in ``pd.read_csv()`` when specifying ``delim_whitespace=True`` and ``lineterminator`` simultaneously with the ``c`` engine (:issue:`12912`)
+- Bug in ``Series.rename``, ``DataFrame.rename`` and ``DataFrame.rename_axis`` not treating ``Series`` as mappings to relabel (:issue:`12623`).
+- Clean in ``.rolling.min`` and ``.rolling.max`` to enhance dtype handling (:issue:`12373`)
+- Bug in ``groupby`` where complex types are coerced to float (:issue:`12902`)
+- Bug in ``Series.map`` raises ``TypeError`` if its dtype is ``category`` or tz-aware ``datetime`` (:issue:`12473`)
+
+- Bugs on 32bit platforms for some test comparisons (:issue:`12972`)
+- Bug in index coercion when falling back from ``RangeIndex`` construction (:issue:`12893`)
+- Better error message in window functions when invalid argument (e.g. a float window) is passed (:issue:`12669`)
+
+- Bug in slicing subclassed ``DataFrame`` defined to return subclassed ``Series`` may return normal ``Series`` (:issue:`11559`)
+
+
+- Bug in ``.str`` accessor methods may raise ``ValueError`` if input has ``name`` and the result is ``DataFrame`` or ``MultiIndex`` (:issue:`12617`)
+- Bug in ``DataFrame.last_valid_index()`` and ``DataFrame.first_valid_index()`` on empty frames (:issue:`12800`)
+
+
+- Bug in ``CategoricalIndex.get_loc`` returns different result from regular ``Index`` (:issue:`12531`)
+- Bug in ``PeriodIndex.resample`` where name not propagated (:issue:`12769`)
+
+- Bug in ``date_range`` ``closed`` keyword and timezones (:issue:`12684`).
+
+- Bug in ``pd.concat`` raises ``AttributeError`` when input data contains tz-aware datetime and timedelta (:issue:`12620`)
+- Bug in ``pd.concat`` did not handle empty ``Series`` properly (:issue:`11082`)
+
+- Bug in ``.plot.bar`` alignment when ``width`` is specified with ``int`` (:issue:`12979`)
+
+
+- Bug in ``fill_value`` is ignored if the argument to a binary operator is a constant (:issue:`12723`)
+
+- Bug in ``pd.read_html()`` when using bs4 flavor and parsing table with a header and only one column (:issue:`9178`)
+
+- Bug in ``.pivot_table`` when ``margins=True`` and ``dropna=True`` where nulls still contributed to margin count (:issue:`12577`)
+- Bug in ``.pivot_table`` when ``dropna=False`` where table index/column names disappear (:issue:`12133`)
+- Bug in ``pd.crosstab()`` when ``margins=True`` and ``dropna=False`` which raised (:issue:`12642`)
+
+- Bug in ``Series.name`` when ``name`` attribute can be a hashable type (:issue:`12610`)
+
+- Bug in ``.describe()`` resets categorical columns information (:issue:`11558`)
+- Bug where ``loffset`` argument was not applied when calling ``resample().count()`` on a timeseries (:issue:`12725`)
+- ``pd.read_excel()`` now accepts column names associated with keyword argument ``names`` (:issue:`12870`)
+- Bug in ``pd.to_numeric()`` with ``Index`` returns ``np.ndarray``, rather than ``Index`` (:issue:`12777`)
+- Bug in ``pd.to_numeric()`` with datetime-like may raise ``TypeError`` (:issue:`12777`)
+- Bug in ``pd.to_numeric()`` with scalar raises ``ValueError`` (:issue:`12777`)
+
+
+.. _whatsnew_0.18.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.18.0..v0.18.1
diff --git a/pandas/doc/source/whatsnew/v0.19.0.rst b/pandas/doc/source/whatsnew/v0.19.0.rst
new file mode 100644
index 0000000000000..1ae711113773f
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.19.0.rst
@@ -0,0 +1,1625 @@
+.. _whatsnew_0190:
+
+Version 0.19.0 (October 2, 2016)
+--------------------------------
+
+{{ header }}
+
+This is a major release from 0.18.1 and includes number of API changes, several new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+Highlights include:
+
+- :func:`merge_asof` for asof-style time-series joining, see :ref:`here <whatsnew_0190.enhancements.asof_merge>`
+- ``.rolling()`` is now time-series aware, see :ref:`here <whatsnew_0190.enhancements.rolling_ts>`
+- :func:`read_csv` now supports parsing ``Categorical`` data, see :ref:`here <whatsnew_0190.enhancements.read_csv_categorical>`
+- A function :func:`union_categorical` has been added for combining categoricals, see :ref:`here <whatsnew_0190.enhancements.union_categoricals>`
+- ``PeriodIndex`` now has its own ``period`` dtype, and changed to be more consistent with other ``Index`` classes. See :ref:`here <whatsnew_0190.api.period>`
+- Sparse data structures gained enhanced support of ``int`` and ``bool`` dtypes, see :ref:`here <whatsnew_0190.sparse>`
+- Comparison operations with ``Series`` no longer ignores the index, see :ref:`here <whatsnew_0190.api.series_ops>` for an overview of the API changes.
+- Introduction of a pandas development API for utility functions, see :ref:`here <whatsnew_0190.dev_api>`.
+- Deprecation of ``Panel4D`` and ``PanelND``. We recommend to represent these types of n-dimensional data with the `xarray package <http://xarray.pydata.org/en/stable/>`__.
+- Removal of the previously deprecated modules ``pandas.io.data``, ``pandas.io.wb``, ``pandas.tools.rplot``.
+
+.. warning::
+
+    pandas >= 0.19.0 will no longer silence numpy ufunc warnings upon import, see :ref:`here <whatsnew_0190.errstate>`.
+
+.. contents:: What's new in v0.19.0
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0190.new_features:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0190.enhancements.asof_merge:
+
+Function ``merge_asof`` for asof-style time-series joining
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A long-time requested feature has been added through the :func:`merge_asof` function, to
+support asof style joining of time-series (:issue:`1870`, :issue:`13695`, :issue:`13709`, :issue:`13902`). Full documentation is
+:ref:`here <merging.merge_asof>`.
+
+The :func:`merge_asof` performs an asof merge, which is similar to a left-join
+except that we match on nearest key rather than equal keys.
+
+.. ipython:: python
+
+   left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+   right = pd.DataFrame({"a": [1, 2, 3, 6, 7], "right_val": [1, 2, 3, 6, 7]})
+
+   left
+   right
+
+We typically want to match exactly when possible, and use the most
+recent value otherwise.
+
+.. ipython:: python
+
+   pd.merge_asof(left, right, on="a")
+
+We can also match rows ONLY with prior data, and not an exact match.
+
+.. ipython:: python
+
+   pd.merge_asof(left, right, on="a", allow_exact_matches=False)
+
+
+In a typical time-series example, we have ``trades`` and ``quotes`` and we want to ``asof-join`` them.
+This also illustrates using the ``by`` parameter to group data before merging.
+
+.. ipython:: python
+
+   trades = pd.DataFrame(
+       {
+           "time": pd.to_datetime(
+               [
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.038",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.048",
+               ]
+           ),
+           "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+           "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+           "quantity": [75, 155, 100, 100, 100],
+       },
+       columns=["time", "ticker", "price", "quantity"],
+   )
+
+   quotes = pd.DataFrame(
+       {
+           "time": pd.to_datetime(
+               [
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.023",
+                   "20160525 13:30:00.030",
+                   "20160525 13:30:00.041",
+                   "20160525 13:30:00.048",
+                   "20160525 13:30:00.049",
+                   "20160525 13:30:00.072",
+                   "20160525 13:30:00.075",
+               ]
+           ),
+           "ticker": ["GOOG", "MSFT", "MSFT", "MSFT", "GOOG", "AAPL", "GOOG", "MSFT"],
+           "bid": [720.50, 51.95, 51.97, 51.99, 720.50, 97.99, 720.50, 52.01],
+           "ask": [720.93, 51.96, 51.98, 52.00, 720.93, 98.01, 720.88, 52.03],
+       },
+       columns=["time", "ticker", "bid", "ask"],
+   )
+
+.. ipython:: python
+
+   trades
+   quotes
+
+An asof merge joins on the ``on``, typically a datetimelike field, which is ordered, and
+in this case we are using a grouper in the ``by`` field. This is like a left-outer join, except
+that forward filling happens automatically taking the most recent non-NaN value.
+
+.. ipython:: python
+
+   pd.merge_asof(trades, quotes, on="time", by="ticker")
+
+This returns a merged DataFrame with the entries in the same order as the original left
+passed DataFrame (``trades`` in this case), with the fields of the ``quotes`` merged.
+
+.. _whatsnew_0190.enhancements.rolling_ts:
+
+Method ``.rolling()`` is now time-series aware
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``.rolling()`` objects are now time-series aware and can accept a time-series offset (or convertible) for the ``window`` argument (:issue:`13327`, :issue:`12995`).
+See the full documentation :ref:`here <window.generic>`.
+
+.. ipython:: python
+
+   dft = pd.DataFrame(
+       {"B": [0, 1, 2, np.nan, 4]},
+       index=pd.date_range("20130101 09:00:00", periods=5, freq="s"),
+   )
+   dft
+
+This is a regular frequency index. Using an integer window parameter works to roll along the window frequency.
+
+.. ipython:: python
+
+   dft.rolling(2).sum()
+   dft.rolling(2, min_periods=1).sum()
+
+Specifying an offset allows a more intuitive specification of the rolling frequency.
+
+.. ipython:: python
+
+   dft.rolling("2s").sum()
+
+Using a non-regular, but still monotonic index, rolling with an integer window does not impart any special calculation.
+
+.. ipython:: python
+
+
+   dft = pd.DataFrame(
+       {"B": [0, 1, 2, np.nan, 4]},
+       index=pd.Index(
+           [
+               pd.Timestamp("20130101 09:00:00"),
+               pd.Timestamp("20130101 09:00:02"),
+               pd.Timestamp("20130101 09:00:03"),
+               pd.Timestamp("20130101 09:00:05"),
+               pd.Timestamp("20130101 09:00:06"),
+           ],
+           name="foo",
+       ),
+   )
+
+   dft
+   dft.rolling(2).sum()
+
+Using the time-specification generates variable windows for this sparse data.
+
+.. ipython:: python
+
+   dft.rolling("2s").sum()
+
+Furthermore, we now allow an optional ``on`` parameter to specify a column (rather than the
+default of the index) in a DataFrame.
+
+.. ipython:: python
+
+   dft = dft.reset_index()
+   dft
+   dft.rolling("2s", on="foo").sum()
+
+.. _whatsnew_0190.enhancements.read_csv_dupe_col_names_support:
+
+Method ``read_csv`` has improved support for duplicate column names
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. ipython:: python
+   :suppress:
+
+   from io import StringIO
+
+:ref:`Duplicate column names <io.dupe_names>` are now supported in :func:`read_csv` whether
+they are in the file or passed in as the ``names`` parameter (:issue:`7160`, :issue:`9424`)
+
+.. ipython:: python
+
+   data = "0,1,2\n3,4,5"
+   names = ["a", "b", "a"]
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [2]: pd.read_csv(StringIO(data), names=names)
+   Out[2]:
+      a  b  a
+   0  2  1  2
+   1  5  4  5
+
+The first ``a`` column contained the same data as the second ``a`` column, when it should have
+contained the values ``[0, 3]``.
+
+**New behavior**:
+
+.. ipython:: python
+   :okexcept:
+
+   pd.read_csv(StringIO(data), names=names)
+
+
+.. _whatsnew_0190.enhancements.read_csv_categorical:
+
+Method ``read_csv`` supports parsing ``Categorical`` directly
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :func:`read_csv` function now supports parsing a ``Categorical`` column when
+specified as a dtype (:issue:`10153`).  Depending on the structure of the data,
+this can result in a faster parse time and lower memory usage compared to
+converting to ``Categorical`` after parsing.  See the io :ref:`docs here <io.categorical>`.
+
+.. ipython:: python
+
+   data = """
+   col1,col2,col3
+   a,b,1
+   a,b,2
+   c,d,3
+   """
+
+   pd.read_csv(StringIO(data))
+   pd.read_csv(StringIO(data)).dtypes
+   pd.read_csv(StringIO(data), dtype="category").dtypes
+
+Individual columns can be parsed as a ``Categorical`` using a dict specification
+
+.. ipython:: python
+
+   pd.read_csv(StringIO(data), dtype={"col1": "category"}).dtypes
+
+.. note::
+
+   The resulting categories will always be parsed as strings (object dtype).
+   If the categories are numeric they can be converted using the
+   :func:`to_numeric` function, or as appropriate, another converter
+   such as :func:`to_datetime`.
+
+   .. ipython:: python
+
+      df = pd.read_csv(StringIO(data), dtype="category")
+      df.dtypes
+      df["col3"]
+      new_categories = pd.to_numeric(df["col3"].cat.categories)
+      df["col3"] = df["col3"].cat.rename_categories(new_categories)
+      df["col3"]
+
+.. _whatsnew_0190.enhancements.union_categoricals:
+
+Categorical concatenation
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- A function :func:`union_categoricals` has been added for combining categoricals, see :ref:`Unioning Categoricals<categorical.union>` (:issue:`13361`, :issue:`13763`, :issue:`13846`, :issue:`14173`)
+
+  .. ipython:: python
+
+     from pandas.api.types import union_categoricals
+
+     a = pd.Categorical(["b", "c"])
+     b = pd.Categorical(["a", "b"])
+     union_categoricals([a, b])
+
+- ``concat`` and ``append`` now can concat ``category`` dtypes with different ``categories`` as ``object`` dtype (:issue:`13524`)
+
+  .. ipython:: python
+
+     s1 = pd.Series(["a", "b"], dtype="category")
+     s2 = pd.Series(["b", "c"], dtype="category")
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [1]: pd.concat([s1, s2])
+   ValueError: incompatible categories in categorical concat
+
+**New behavior**:
+
+.. ipython:: python
+
+   pd.concat([s1, s2])
+
+.. _whatsnew_0190.enhancements.semi_month_offsets:
+
+Semi-month offsets
+^^^^^^^^^^^^^^^^^^
+
+pandas has gained new frequency offsets, ``SemiMonthEnd`` ('SM') and ``SemiMonthBegin`` ('SMS').
+These provide date offsets anchored (by default) to the 15th and end of month, and 15th and 1st of month respectively.
+(:issue:`1543`)
+
+.. ipython:: python
+
+   from pandas.tseries.offsets import SemiMonthEnd, SemiMonthBegin
+
+**SemiMonthEnd**:
+
+.. code-block:: python
+
+   In [46]: pd.Timestamp("2016-01-01") + SemiMonthEnd()
+   Out[46]: Timestamp('2016-01-15 00:00:00')
+
+   In [47]: pd.date_range("2015-01-01", freq="SM", periods=4)
+   Out[47]: DatetimeIndex(['2015-01-15', '2015-01-31', '2015-02-15', '2015-02-28'], dtype='datetime64[ns]', freq='SM-15')
+
+**SemiMonthBegin**:
+
+.. ipython:: python
+
+   pd.Timestamp("2016-01-01") + SemiMonthBegin()
+
+   pd.date_range("2015-01-01", freq="SMS", periods=4)
+
+Using the anchoring suffix, you can also specify the day of month to use instead of the 15th.
+
+.. code-block:: python
+
+   In [50]: pd.date_range("2015-01-01", freq="SMS-16", periods=4)
+   Out[50]: DatetimeIndex(['2015-01-01', '2015-01-16', '2015-02-01', '2015-02-16'], dtype='datetime64[ns]', freq='SMS-16')
+
+   In [51]: pd.date_range("2015-01-01", freq="SM-14", periods=4)
+   Out[51]: DatetimeIndex(['2015-01-14', '2015-01-31', '2015-02-14', '2015-02-28'], dtype='datetime64[ns]', freq='SM-14')
+
+.. _whatsnew_0190.enhancements.index:
+
+New Index methods
+^^^^^^^^^^^^^^^^^
+
+The following methods and options are added to ``Index``, to be more consistent with the ``Series`` and ``DataFrame`` API.
+
+``Index`` now supports the ``.where()`` function for same shape indexing (:issue:`13170`)
+
+.. ipython:: python
+
+   idx = pd.Index(["a", "b", "c"])
+   idx.where([True, False, True])
+
+
+``Index`` now supports ``.dropna()`` to exclude missing values (:issue:`6194`)
+
+.. ipython:: python
+
+   idx = pd.Index([1, 2, np.nan, 4])
+   idx.dropna()
+
+For ``MultiIndex``, values are dropped if any level is missing by default. Specifying
+``how='all'`` only drops values where all levels are missing.
+
+.. ipython:: python
+
+   midx = pd.MultiIndex.from_arrays([[1, 2, np.nan, 4], [1, 2, np.nan, np.nan]])
+   midx
+   midx.dropna()
+   midx.dropna(how="all")
+
+``Index`` now supports ``.str.extractall()`` which returns a ``DataFrame``, see the :ref:`docs here <text.extractall>` (:issue:`10008`, :issue:`13156`)
+
+.. ipython:: python
+
+   idx = pd.Index(["a1a2", "b1", "c1"])
+   idx.str.extractall(r"[ab](?P<digit>\d)")
+
+``Index.astype()`` now accepts an optional boolean argument ``copy``, which allows optional copying if the requirements on dtype are satisfied (:issue:`13209`)
+
+.. _whatsnew_0190.gbq:
+
+Google BigQuery enhancements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- The :func:`read_gbq` method has gained the ``dialect`` argument to allow users to specify whether to use BigQuery's legacy SQL or BigQuery's standard SQL. See the `docs <https://pandas-gbq.readthedocs.io/en/latest/reading.html>`__ for more details (:issue:`13615`).
+- The :func:`~DataFrame.to_gbq` method now allows the DataFrame column order to differ from the destination table schema (:issue:`11359`).
+
+.. _whatsnew_0190.errstate:
+
+Fine-grained NumPy errstate
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previous versions of pandas would permanently silence numpy's ufunc error handling when ``pandas`` was imported. pandas did this in order to silence the warnings that would arise from using numpy ufuncs on missing data, which are usually represented as ``NaN`` s. Unfortunately, this silenced legitimate warnings arising in non-pandas code in the application. Starting with 0.19.0, pandas will use the ``numpy.errstate`` context manager to silence these warnings in a more fine-grained manner, only around where these operations are actually used in the pandas code base. (:issue:`13109`, :issue:`13145`)
+
+After upgrading pandas, you may see *new* ``RuntimeWarnings`` being issued from your code. These are likely legitimate, and the underlying cause likely existed in the code when using previous versions of pandas that simply silenced the warning. Use `numpy.errstate <https://numpy.org/doc/stable/reference/generated/numpy.errstate.html>`__ around the source of the ``RuntimeWarning`` to control how these conditions are handled.
+
+.. _whatsnew_0190.get_dummies_dtypes:
+
+Method ``get_dummies`` now returns integer dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``pd.get_dummies`` function now returns dummy-encoded columns as small integers, rather than floats (:issue:`8725`). This should provide an improved memory footprint.
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [1]: pd.get_dummies(['a', 'b', 'a', 'c']).dtypes
+
+   Out[1]:
+   a    float64
+   b    float64
+   c    float64
+   dtype: object
+
+**New behavior**:
+
+.. ipython:: python
+
+   pd.get_dummies(["a", "b", "a", "c"]).dtypes
+
+
+.. _whatsnew_0190.enhancements.to_numeric_downcast:
+
+Downcast values to smallest possible dtype in ``to_numeric``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``pd.to_numeric()`` now accepts a ``downcast`` parameter, which will downcast the data if possible to smallest specified numerical dtype (:issue:`13352`)
+
+.. ipython:: python
+
+   s = ["1", 2, 3]
+   pd.to_numeric(s, downcast="unsigned")
+   pd.to_numeric(s, downcast="integer")
+
+.. _whatsnew_0190.dev_api:
+
+pandas development API
+^^^^^^^^^^^^^^^^^^^^^^
+
+As part of making pandas API more uniform and accessible in the future, we have created a standard
+sub-package of pandas, ``pandas.api`` to hold public API's. We are starting by exposing type
+introspection functions in ``pandas.api.types``. More sub-packages and officially sanctioned API's
+will be published in future versions of pandas (:issue:`13147`, :issue:`13634`)
+
+The following are now part of this API:
+
+.. ipython:: python
+
+   import pprint
+   from pandas.api import types
+
+   funcs = [f for f in dir(types) if not f.startswith("_")]
+   pprint.pprint(funcs)
+
+.. note::
+
+   Calling these functions from the internal module ``pandas.core.common`` will now show a ``DeprecationWarning`` (:issue:`13990`)
+
+
+.. _whatsnew_0190.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- ``Timestamp`` can now accept positional and keyword parameters similar to :func:`datetime.datetime` (:issue:`10758`, :issue:`11630`)
+
+  .. ipython:: python
+
+     pd.Timestamp(2012, 1, 1)
+
+     pd.Timestamp(year=2012, month=1, day=1, hour=8, minute=30)
+
+- The ``.resample()`` function now accepts a ``on=`` or ``level=`` parameter for resampling on a datetimelike column or ``MultiIndex`` level (:issue:`13500`)
+
+  .. ipython:: python
+
+     df = pd.DataFrame(
+         {"date": pd.date_range("2015-01-01", freq="W", periods=5), "a": np.arange(5)},
+         index=pd.MultiIndex.from_arrays(
+             [[1, 2, 3, 4, 5], pd.date_range("2015-01-01", freq="W", periods=5)],
+             names=["v", "d"],
+         ),
+     )
+     df
+
+  .. code-block:: ipython
+
+     In [74]: df.resample("M", on="date")[["a"]].sum()
+     Out[74]:
+                 a
+     date
+     2015-01-31  6
+     2015-02-28  4
+
+     [2 rows x 1 columns]
+
+     In [75]: df.resample("M", level="d")[["a"]].sum()
+     Out[75]:
+                 a
+     d
+     2015-01-31  6
+     2015-02-28  4
+
+     [2 rows x 1 columns]
+
+- The ``.get_credentials()`` method of ``GbqConnector`` can now first try to fetch `the application default credentials <https://developers.google.com/identity/protocols/application-default-credentials>`__. See the docs for more details (:issue:`13577`).
+- The ``.tz_localize()`` method of ``DatetimeIndex`` and ``Timestamp`` has gained the ``errors`` keyword, so you can potentially coerce nonexistent timestamps to ``NaT``. The default behavior remains to raising a ``NonExistentTimeError`` (:issue:`13057`)
+- ``.to_hdf/read_hdf()`` now accept path objects (e.g. ``pathlib.Path``, ``py.path.local``) for the file path (:issue:`11773`)
+- The ``pd.read_csv()`` with ``engine='python'`` has gained support for the
+  ``decimal`` (:issue:`12933`), ``na_filter`` (:issue:`13321`) and the ``memory_map`` option (:issue:`13381`).
+- Consistent with the Python API, ``pd.read_csv()`` will now interpret ``+inf`` as positive infinity (:issue:`13274`)
+- The ``pd.read_html()`` has gained support for the ``na_values``, ``converters``, ``keep_default_na``  options (:issue:`13461`)
+- ``Categorical.astype()`` now accepts an optional boolean argument ``copy``, effective when dtype is categorical (:issue:`13209`)
+- ``DataFrame`` has gained the ``.asof()`` method to return the last non-NaN values according to the selected subset (:issue:`13358`)
+- The ``DataFrame`` constructor will now respect key ordering if a list of ``OrderedDict`` objects are passed in (:issue:`13304`)
+- ``pd.read_html()`` has gained support for the ``decimal`` option (:issue:`12907`)
+- ``Series`` has gained the properties ``.is_monotonic``, ``.is_monotonic_increasing``, ``.is_monotonic_decreasing``, similar to ``Index`` (:issue:`13336`)
+- ``DataFrame.to_sql()`` now allows a single value as the SQL type for all columns (:issue:`11886`).
+- ``Series.append`` now supports the ``ignore_index`` option (:issue:`13677`)
+- ``.to_stata()`` and ``StataWriter`` can now write variable labels to Stata dta files using a dictionary to make column names to labels (:issue:`13535`, :issue:`13536`)
+- ``.to_stata()`` and ``StataWriter`` will automatically convert ``datetime64[ns]`` columns to Stata format ``%tc``, rather than raising a ``ValueError`` (:issue:`12259`)
+- ``read_stata()`` and ``StataReader`` raise with a more explicit error message when reading Stata files with repeated value labels when ``convert_categoricals=True`` (:issue:`13923`)
+- ``DataFrame.style`` will now render sparsified MultiIndexes (:issue:`11655`)
+- ``DataFrame.style`` will now show column level names (e.g. ``DataFrame.columns.names``) (:issue:`13775`)
+- ``DataFrame`` has gained support to re-order the columns based on the values
+  in a row using ``df.sort_values(by='...', axis=1)`` (:issue:`10806`)
+
+  .. ipython:: python
+
+     df = pd.DataFrame({"A": [2, 7], "B": [3, 5], "C": [4, 8]}, index=["row1", "row2"])
+     df
+     df.sort_values(by="row2", axis=1)
+
+- Added documentation to :ref:`I/O<io.dtypes>` regarding the perils of reading in columns with mixed dtypes and how to handle it (:issue:`13746`)
+- :meth:`~DataFrame.to_html` now has a ``border`` argument to control the value in the opening ``<table>`` tag. The default is the value of the ``html.border`` option, which defaults to 1. This also affects the notebook HTML repr, but since Jupyter's CSS includes a border-width attribute, the visual effect is the same. (:issue:`11563`).
+- Raise ``ImportError`` in the sql functions when ``sqlalchemy`` is not installed and a connection string is used (:issue:`11920`).
+- Compatibility with matplotlib 2.0. Older versions of pandas should also work with matplotlib 2.0 (:issue:`13333`)
+- ``Timestamp``, ``Period``, ``DatetimeIndex``, ``PeriodIndex`` and ``.dt`` accessor have gained a ``.is_leap_year`` property to check whether the date belongs to a leap year. (:issue:`13727`)
+- ``astype()`` will now accept a dict of column name to data types mapping as the ``dtype`` argument. (:issue:`12086`)
+- The ``pd.read_json`` and ``DataFrame.to_json`` has gained support for reading and writing json lines with ``lines`` option see :ref:`Line delimited json <io.jsonl>` (:issue:`9180`)
+- :func:`read_excel` now supports the true_values and false_values keyword arguments (:issue:`13347`)
+- ``groupby()`` will now accept a scalar and a single-element list for specifying ``level`` on a non-``MultiIndex`` grouper. (:issue:`13907`)
+- Non-convertible dates in an excel date column will be returned without conversion and the column will be ``object`` dtype, rather than raising an exception (:issue:`10001`).
+- ``pd.Timedelta(None)`` is now accepted and will return ``NaT``, mirroring ``pd.Timestamp`` (:issue:`13687`)
+- ``pd.read_stata()`` can now handle some format 111 files, which are produced by SAS when generating Stata dta files (:issue:`11526`)
+- ``Series`` and ``Index`` now support ``divmod`` which will return a tuple of
+  series or indices. This behaves like a standard binary operator with regards
+  to broadcasting rules (:issue:`14208`).
+
+
+.. _whatsnew_0190.api:
+
+API changes
+~~~~~~~~~~~
+
+``Series.tolist()`` will now return Python types
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Series.tolist()`` will now return Python types in the output, mimicking NumPy ``.tolist()`` behavior (:issue:`10904`)
+
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3])
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [7]: type(s.tolist()[0])
+   Out[7]:
+    <class 'numpy.int64'>
+
+**New behavior**:
+
+.. ipython:: python
+
+   type(s.tolist()[0])
+
+.. _whatsnew_0190.api.series_ops:
+
+``Series`` operators for different indexes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Following ``Series`` operators have been changed to make all operators consistent,
+including ``DataFrame`` (:issue:`1134`, :issue:`4581`, :issue:`13538`)
+
+- ``Series`` comparison operators now raise ``ValueError`` when ``index`` are different.
+- ``Series`` logical operators align both ``index`` of left and right hand side.
+
+.. warning::
+   Until 0.18.1, comparing ``Series`` with the same length, would succeed even if
+   the ``.index`` are different (the result ignores ``.index``). As of 0.19.0, this will raises ``ValueError`` to be more strict. This section also describes how to keep previous behavior or align different indexes, using the flexible comparison methods like ``.eq``.
+
+
+As a result, ``Series`` and ``DataFrame`` operators behave as below:
+
+Arithmetic operators
+""""""""""""""""""""
+
+Arithmetic operators align both ``index`` (no changes).
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, 3], index=list("ABC"))
+   s2 = pd.Series([2, 2, 2], index=list("ABD"))
+   s1 + s2
+
+   df1 = pd.DataFrame([1, 2, 3], index=list("ABC"))
+   df2 = pd.DataFrame([2, 2, 2], index=list("ABD"))
+   df1 + df2
+
+Comparison operators
+""""""""""""""""""""
+
+Comparison operators raise ``ValueError`` when ``.index`` are different.
+
+**Previous behavior** (``Series``):
+
+``Series`` compared values ignoring the ``.index`` as long as both had the same length:
+
+.. code-block:: ipython
+
+   In [1]: s1 == s2
+   Out[1]:
+   A    False
+   B     True
+   C    False
+   dtype: bool
+
+**New behavior** (``Series``):
+
+.. code-block:: ipython
+
+   In [2]: s1 == s2
+   Out[2]:
+   ValueError: Can only compare identically-labeled Series objects
+
+.. note::
+
+   To achieve the same result as previous versions (compare values based on locations ignoring ``.index``), compare both ``.values``.
+
+   .. ipython:: python
+
+      s1.values == s2.values
+
+   If you want to compare ``Series`` aligning its ``.index``, see flexible comparison methods section below:
+
+   .. ipython:: python
+
+      s1.eq(s2)
+
+**Current behavior** (``DataFrame``, no change):
+
+.. code-block:: ipython
+
+   In [3]: df1 == df2
+   Out[3]:
+   ValueError: Can only compare identically-labeled DataFrame objects
+
+Logical operators
+"""""""""""""""""
+
+Logical operators align both ``.index`` of left and right hand side.
+
+**Previous behavior** (``Series``), only left hand side ``index`` was kept:
+
+.. code-block:: ipython
+
+   In [4]: s1 = pd.Series([True, False, True], index=list('ABC'))
+   In [5]: s2 = pd.Series([True, True, True], index=list('ABD'))
+   In [6]: s1 & s2
+   Out[6]:
+   A     True
+   B    False
+   C    False
+   dtype: bool
+
+**New behavior** (``Series``):
+
+.. ipython:: python
+
+   s1 = pd.Series([True, False, True], index=list("ABC"))
+   s2 = pd.Series([True, True, True], index=list("ABD"))
+   s1 & s2
+
+.. note::
+   ``Series`` logical operators fill a ``NaN`` result with ``False``.
+
+.. note::
+   To achieve the same result as previous versions (compare values based on only left hand side index), you can use ``reindex_like``:
+
+   .. ipython:: python
+
+      s1 & s2.reindex_like(s1)
+
+**Current behavior** (``DataFrame``, no change):
+
+.. ipython:: python
+
+   df1 = pd.DataFrame([True, False, True], index=list("ABC"))
+   df2 = pd.DataFrame([True, True, True], index=list("ABD"))
+   df1 & df2
+
+Flexible comparison methods
+"""""""""""""""""""""""""""
+
+``Series`` flexible comparison methods like ``eq``, ``ne``, ``le``, ``lt``, ``ge`` and ``gt`` now align both ``index``. Use these operators if you want to compare two ``Series``
+which has the different ``index``.
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, 3], index=["a", "b", "c"])
+   s2 = pd.Series([2, 2, 2], index=["b", "c", "d"])
+   s1.eq(s2)
+   s1.ge(s2)
+
+Previously, this worked the same as comparison operators (see above).
+
+.. _whatsnew_0190.api.promote:
+
+``Series`` type promotion on assignment
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A ``Series`` will now correctly promote its dtype for assignment with incompat values to the current dtype (:issue:`13234`)
+
+
+.. ipython:: python
+   :okwarning:
+
+   s = pd.Series()
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [2]: s["a"] = pd.Timestamp("2016-01-01")
+
+   In [3]: s["b"] = 3.0
+   TypeError: invalid type promotion
+
+**New behavior**:
+
+.. ipython:: python
+
+   s["a"] = pd.Timestamp("2016-01-01")
+   s["b"] = 3.0
+   s
+   s.dtype
+
+.. _whatsnew_0190.api.to_datetime_coerce:
+
+Function ``.to_datetime()`` changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously if ``.to_datetime()`` encountered mixed integers/floats and strings, but no datetimes with ``errors='coerce'`` it would convert all to ``NaT``.
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [2]: pd.to_datetime([1, 'foo'], errors='coerce')
+   Out[2]: DatetimeIndex(['NaT', 'NaT'], dtype='datetime64[ns]', freq=None)
+
+**Current behavior**:
+
+This will now convert integers/floats with the default unit of ``ns``.
+
+.. ipython:: python
+
+   pd.to_datetime([1, "foo"], errors="coerce")
+
+Bug fixes related to ``.to_datetime()``:
+
+- Bug in ``pd.to_datetime()`` when passing integers or floats, and no ``unit`` and ``errors='coerce'`` (:issue:`13180`).
+- Bug in ``pd.to_datetime()`` when passing invalid data types (e.g. bool); will now respect the ``errors`` keyword (:issue:`13176`)
+- Bug in ``pd.to_datetime()`` which overflowed on ``int8``, and ``int16`` dtypes (:issue:`13451`)
+- Bug in ``pd.to_datetime()`` raise ``AttributeError`` with ``NaN`` and the other string is not valid when ``errors='ignore'`` (:issue:`12424`)
+- Bug in ``pd.to_datetime()`` did not cast floats correctly when ``unit`` was specified, resulting in truncated datetime (:issue:`13834`)
+
+.. _whatsnew_0190.api.merging:
+
+Merging changes
+^^^^^^^^^^^^^^^
+
+Merging will now preserve the dtype of the join keys (:issue:`8596`)
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"key": [1], "v1": [10]})
+   df1
+   df2 = pd.DataFrame({"key": [1, 2], "v1": [20, 30]})
+   df2
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [5]: pd.merge(df1, df2, how='outer')
+   Out[5]:
+      key    v1
+   0  1.0  10.0
+   1  1.0  20.0
+   2  2.0  30.0
+
+   In [6]: pd.merge(df1, df2, how='outer').dtypes
+   Out[6]:
+   key    float64
+   v1     float64
+   dtype: object
+
+**New behavior**:
+
+We are able to preserve the join keys
+
+.. ipython:: python
+
+   pd.merge(df1, df2, how="outer")
+   pd.merge(df1, df2, how="outer").dtypes
+
+Of course if you have missing values that are introduced, then the
+resulting dtype will be upcast, which is unchanged from previous.
+
+.. ipython:: python
+
+   pd.merge(df1, df2, how="outer", on="key")
+   pd.merge(df1, df2, how="outer", on="key").dtypes
+
+.. _whatsnew_0190.api.describe:
+
+Method ``.describe()`` changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Percentile identifiers in the index of a ``.describe()`` output will now be rounded to the least precision that keeps them distinct (:issue:`13104`)
+
+.. ipython:: python
+
+   s = pd.Series([0, 1, 2, 3, 4])
+   df = pd.DataFrame([0, 1, 2, 3, 4])
+
+**Previous behavior**:
+
+The percentiles were rounded to at most one decimal place, which could raise ``ValueError`` for a data frame if the percentiles were duplicated.
+
+.. code-block:: ipython
+
+   In [3]: s.describe(percentiles=[0.0001, 0.0005, 0.001, 0.999, 0.9995, 0.9999])
+   Out[3]:
+   count     5.000000
+   mean      2.000000
+   std       1.581139
+   min       0.000000
+   0.0%      0.000400
+   0.1%      0.002000
+   0.1%      0.004000
+   50%       2.000000
+   99.9%     3.996000
+   100.0%    3.998000
+   100.0%    3.999600
+   max       4.000000
+   dtype: float64
+
+   In [4]: df.describe(percentiles=[0.0001, 0.0005, 0.001, 0.999, 0.9995, 0.9999])
+   Out[4]:
+   ...
+   ValueError: cannot reindex from a duplicate axis
+
+**New behavior**:
+
+.. ipython:: python
+
+   s.describe(percentiles=[0.0001, 0.0005, 0.001, 0.999, 0.9995, 0.9999])
+   df.describe(percentiles=[0.0001, 0.0005, 0.001, 0.999, 0.9995, 0.9999])
+
+Furthermore:
+
+- Passing duplicated ``percentiles`` will now raise a ``ValueError``.
+- Bug in ``.describe()`` on a DataFrame with a mixed-dtype column index, which would previously raise a ``TypeError`` (:issue:`13288`)
+
+.. _whatsnew_0190.api.period:
+
+``Period`` changes
+^^^^^^^^^^^^^^^^^^
+
+The ``PeriodIndex`` now has ``period`` dtype
+""""""""""""""""""""""""""""""""""""""""""""
+
+``PeriodIndex`` now has its own ``period`` dtype. The ``period`` dtype is a
+pandas extension dtype like ``category`` or the :ref:`timezone aware dtype <timeseries.timezone_series>` (``datetime64[ns, tz]``) (:issue:`13941`).
+As a consequence of this change, ``PeriodIndex`` no longer has an integer dtype:
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [1]: pi = pd.PeriodIndex(['2016-08-01'], freq='D')
+
+   In [2]: pi
+   Out[2]: PeriodIndex(['2016-08-01'], dtype='int64', freq='D')
+
+   In [3]: pd.api.types.is_integer_dtype(pi)
+   Out[3]: True
+
+   In [4]: pi.dtype
+   Out[4]: dtype('int64')
+
+**New behavior**:
+
+.. ipython:: python
+   :okwarning:
+
+   pi = pd.PeriodIndex(["2016-08-01"], freq="D")
+   pi
+   pd.api.types.is_integer_dtype(pi)
+   pd.api.types.is_period_dtype(pi)
+   pi.dtype
+   type(pi.dtype)
+
+.. _whatsnew_0190.api.periodnat:
+
+``Period('NaT')`` now returns ``pd.NaT``
+""""""""""""""""""""""""""""""""""""""""
+
+Previously, ``Period`` has its own ``Period('NaT')`` representation different from ``pd.NaT``. Now ``Period('NaT')`` has been changed to return ``pd.NaT``. (:issue:`12759`, :issue:`13582`)
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [5]: pd.Period('NaT', freq='D')
+   Out[5]: Period('NaT', 'D')
+
+**New behavior**:
+
+These result in ``pd.NaT`` without providing ``freq`` option.
+
+.. ipython:: python
+
+   pd.Period("NaT")
+   pd.Period(None)
+
+
+To be compatible with ``Period`` addition and subtraction, ``pd.NaT`` now supports addition and subtraction with ``int``. Previously it raised ``ValueError``.
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [5]: pd.NaT + 1
+   ...
+   ValueError: Cannot add integral value to Timestamp without freq.
+
+**New behavior**:
+
+.. ipython:: python
+
+   pd.NaT + 1
+   pd.NaT - 1
+
+``PeriodIndex.values`` now returns array of ``Period`` object
+"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
+
+``.values`` is changed to return an array of ``Period`` objects, rather than an array
+of integers (:issue:`13988`).
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [6]: pi = pd.PeriodIndex(['2011-01', '2011-02'], freq='M')
+   In [7]: pi.values
+   Out[7]: array([492, 493])
+
+**New behavior**:
+
+.. ipython:: python
+
+   pi = pd.PeriodIndex(["2011-01", "2011-02"], freq="M")
+   pi.values
+
+
+.. _whatsnew_0190.api.setops:
+
+Index ``+`` / ``-`` no longer used for set operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Addition and subtraction of the base Index type and of DatetimeIndex
+(not the numeric index types)
+previously performed set operations (set union and difference). This
+behavior was already deprecated since 0.15.0 (in favor using the specific
+``.union()`` and ``.difference()`` methods), and is now disabled. When
+possible, ``+`` and ``-`` are now used for element-wise operations, for
+example for concatenating strings or subtracting datetimes
+(:issue:`8227`, :issue:`14127`).
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [1]: pd.Index(['a', 'b']) + pd.Index(['a', 'c'])
+   FutureWarning: using '+' to provide set union with Indexes is deprecated, use '|' or .union()
+   Out[1]: Index(['a', 'b', 'c'], dtype='object')
+
+**New behavior**: the same operation will now perform element-wise addition:
+
+.. ipython:: python
+
+   pd.Index(["a", "b"]) + pd.Index(["a", "c"])
+
+Note that numeric Index objects already performed element-wise operations.
+For example, the behavior of adding two integer Indexes is unchanged.
+The base ``Index`` is now made consistent with this behavior.
+
+.. ipython:: python
+
+   pd.Index([1, 2, 3]) + pd.Index([2, 3, 4])
+
+Further, because of this change, it is now possible to subtract two
+DatetimeIndex objects resulting in a TimedeltaIndex:
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+    In [1]: (pd.DatetimeIndex(['2016-01-01', '2016-01-02'])
+       ...:  - pd.DatetimeIndex(['2016-01-02', '2016-01-03']))
+    FutureWarning: using '-' to provide set differences with datetimelike Indexes is deprecated, use .difference()
+    Out[1]: DatetimeIndex(['2016-01-01'], dtype='datetime64[ns]', freq=None)
+
+**New behavior**:
+
+.. ipython:: python
+
+    (
+        pd.DatetimeIndex(["2016-01-01", "2016-01-02"])
+        - pd.DatetimeIndex(["2016-01-02", "2016-01-03"])
+    )
+
+
+.. _whatsnew_0190.api.difference:
+
+``Index.difference`` and ``.symmetric_difference`` changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Index.difference`` and ``Index.symmetric_difference`` will now, more consistently, treat ``NaN`` values as any other values. (:issue:`13514`)
+
+.. ipython:: python
+
+   idx1 = pd.Index([1, 2, 3, np.nan])
+   idx2 = pd.Index([0, 1, np.nan])
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [3]: idx1.difference(idx2)
+   Out[3]: Float64Index([nan, 2.0, 3.0], dtype='float64')
+
+   In [4]: idx1.symmetric_difference(idx2)
+   Out[4]: Float64Index([0.0, nan, 2.0, 3.0], dtype='float64')
+
+**New behavior**:
+
+.. ipython:: python
+
+   idx1.difference(idx2)
+   idx1.symmetric_difference(idx2)
+
+.. _whatsnew_0190.api.unique_index:
+
+``Index.unique`` consistently returns ``Index``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Index.unique()`` now returns unique values as an
+``Index`` of the appropriate ``dtype``. (:issue:`13395`).
+Previously, most ``Index`` classes returned ``np.ndarray``, and ``DatetimeIndex``,
+``TimedeltaIndex`` and ``PeriodIndex`` returned ``Index`` to keep metadata like timezone.
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [1]: pd.Index([1, 2, 3]).unique()
+   Out[1]: array([1, 2, 3])
+
+   In [2]: pd.DatetimeIndex(['2011-01-01', '2011-01-02',
+      ...:                   '2011-01-03'], tz='Asia/Tokyo').unique()
+   Out[2]:
+   DatetimeIndex(['2011-01-01 00:00:00+09:00', '2011-01-02 00:00:00+09:00',
+                  '2011-01-03 00:00:00+09:00'],
+                 dtype='datetime64[ns, Asia/Tokyo]', freq=None)
+
+**New behavior**:
+
+.. ipython:: python
+
+   pd.Index([1, 2, 3]).unique()
+   pd.DatetimeIndex(
+       ["2011-01-01", "2011-01-02", "2011-01-03"], tz="Asia/Tokyo"
+   ).unique()
+
+.. _whatsnew_0190.api.multiindex:
+
+``MultiIndex`` constructors, ``groupby`` and ``set_index`` preserve categorical dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``MultiIndex.from_arrays`` and ``MultiIndex.from_product`` will now preserve categorical dtype
+in ``MultiIndex`` levels (:issue:`13743`, :issue:`13854`).
+
+.. ipython:: python
+
+   cat = pd.Categorical(["a", "b"], categories=list("bac"))
+   lvl1 = ["foo", "bar"]
+   midx = pd.MultiIndex.from_arrays([cat, lvl1])
+   midx
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [4]: midx.levels[0]
+   Out[4]: Index(['b', 'a', 'c'], dtype='object')
+
+   In [5]: midx.get_level_values[0]
+   Out[5]: Index(['a', 'b'], dtype='object')
+
+**New behavior**: the single level is now a ``CategoricalIndex``:
+
+.. ipython:: python
+
+   midx.levels[0]
+   midx.get_level_values(0)
+
+An analogous change has been made to ``MultiIndex.from_product``.
+As a consequence, ``groupby`` and ``set_index`` also preserve categorical dtypes in indexes
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [0, 1], "B": [10, 11], "C": cat})
+   df_grouped = df.groupby(by=["A", "C"], observed=False).first()
+   df_set_idx = df.set_index(["A", "C"])
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [11]: df_grouped.index.levels[1]
+   Out[11]: Index(['b', 'a', 'c'], dtype='object', name='C')
+   In [12]: df_grouped.reset_index().dtypes
+   Out[12]:
+   A      int64
+   C     object
+   B    float64
+   dtype: object
+
+   In [13]: df_set_idx.index.levels[1]
+   Out[13]: Index(['b', 'a', 'c'], dtype='object', name='C')
+   In [14]: df_set_idx.reset_index().dtypes
+   Out[14]:
+   A      int64
+   C     object
+   B      int64
+   dtype: object
+
+**New behavior**:
+
+.. ipython:: python
+
+   df_grouped.index.levels[1]
+   df_grouped.reset_index().dtypes
+
+   df_set_idx.index.levels[1]
+   df_set_idx.reset_index().dtypes
+
+.. _whatsnew_0190.api.autogenerated_chunksize_index:
+
+Function ``read_csv`` will progressively enumerate chunks
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When :func:`read_csv` is called with ``chunksize=n`` and without specifying an index,
+each chunk used to have an independently generated index from ``0`` to ``n-1``.
+They are now given instead a progressive index, starting from ``0`` for the first chunk,
+from ``n`` for the second, and so on, so that, when concatenated, they are identical to
+the result of calling :func:`read_csv` without the ``chunksize=`` argument
+(:issue:`12185`).
+
+.. ipython:: python
+
+   data = "A,B\n0,1\n2,3\n4,5\n6,7"
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [2]: pd.concat(pd.read_csv(StringIO(data), chunksize=2))
+   Out[2]:
+      A  B
+   0  0  1
+   1  2  3
+   0  4  5
+   1  6  7
+
+**New behavior**:
+
+.. ipython:: python
+
+   pd.concat(pd.read_csv(StringIO(data), chunksize=2))
+
+.. _whatsnew_0190.sparse:
+
+Sparse changes
+^^^^^^^^^^^^^^
+
+These changes allow pandas to handle sparse data with more dtypes, and for work to make a smoother experience with data handling.
+
+Types ``int64`` and ``bool`` support enhancements
+"""""""""""""""""""""""""""""""""""""""""""""""""
+
+Sparse data structures now gained enhanced support of ``int64`` and ``bool`` ``dtype`` (:issue:`667`, :issue:`13849`).
+
+Previously, sparse data were ``float64`` dtype by default, even if all inputs were of ``int`` or ``bool`` dtype. You had to specify ``dtype`` explicitly to create sparse data with ``int64`` dtype. Also, ``fill_value`` had to be specified explicitly because the default was ``np.nan`` which doesn't appear in ``int64`` or ``bool`` data.
+
+.. code-block:: ipython
+
+   In [1]: pd.SparseArray([1, 2, 0, 0])
+   Out[1]:
+   [1.0, 2.0, 0.0, 0.0]
+   Fill: nan
+   IntIndex
+   Indices: array([0, 1, 2, 3], dtype=int32)
+
+   # specifying int64 dtype, but all values are stored in sp_values because
+   # fill_value default is np.nan
+   In [2]: pd.SparseArray([1, 2, 0, 0], dtype=np.int64)
+   Out[2]:
+   [1, 2, 0, 0]
+   Fill: nan
+   IntIndex
+   Indices: array([0, 1, 2, 3], dtype=int32)
+
+   In [3]: pd.SparseArray([1, 2, 0, 0], dtype=np.int64, fill_value=0)
+   Out[3]:
+   [1, 2, 0, 0]
+   Fill: 0
+   IntIndex
+   Indices: array([0, 1], dtype=int32)
+
+As of v0.19.0, sparse data keeps the input dtype, and uses more appropriate ``fill_value`` defaults (``0`` for ``int64`` dtype, ``False`` for ``bool`` dtype).
+
+.. ipython:: python
+
+   pd.arrays.SparseArray([1, 2, 0, 0], dtype=np.int64)
+   pd.arrays.SparseArray([True, False, False, False])
+
+See the :ref:`docs <sparse.dtype>` for more details.
+
+Operators now preserve dtypes
+"""""""""""""""""""""""""""""
+
+- Sparse data structure now can preserve ``dtype`` after arithmetic ops (:issue:`13848`)
+
+.. code-block:: python
+
+   s = pd.SparseSeries([0, 2, 0, 1], fill_value=0, dtype=np.int64)
+   s.dtype
+
+   s + 1
+
+- Sparse data structure now support ``astype`` to convert internal ``dtype`` (:issue:`13900`)
+
+.. code-block:: python
+
+   s = pd.SparseSeries([1.0, 0.0, 2.0, 0.0], fill_value=0)
+   s
+   s.astype(np.int64)
+
+``astype`` fails if data contains values which cannot be converted to specified ``dtype``.
+Note that the limitation is applied to ``fill_value`` which default is ``np.nan``.
+
+.. code-block:: ipython
+
+   In [7]: pd.SparseSeries([1., np.nan, 2., np.nan], fill_value=np.nan).astype(np.int64)
+   Out[7]:
+   ValueError: unable to coerce current fill_value nan to int64 dtype
+
+Other sparse fixes
+""""""""""""""""""
+
+- Subclassed ``SparseDataFrame`` and ``SparseSeries`` now preserve class types when slicing or transposing. (:issue:`13787`)
+- ``SparseArray`` with ``bool`` dtype now supports logical (bool) operators (:issue:`14000`)
+- Bug in ``SparseSeries`` with ``MultiIndex`` ``[]`` indexing may raise ``IndexError`` (:issue:`13144`)
+- Bug in ``SparseSeries`` with ``MultiIndex`` ``[]`` indexing result may have normal ``Index`` (:issue:`13144`)
+- Bug in ``SparseDataFrame`` in which ``axis=None`` did not default to ``axis=0`` (:issue:`13048`)
+- Bug in ``SparseSeries`` and ``SparseDataFrame`` creation with ``object`` dtype may raise ``TypeError`` (:issue:`11633`)
+- Bug in ``SparseDataFrame`` doesn't respect passed ``SparseArray`` or ``SparseSeries`` 's dtype and ``fill_value``  (:issue:`13866`)
+- Bug in ``SparseArray`` and ``SparseSeries`` don't apply ufunc to ``fill_value`` (:issue:`13853`)
+- Bug in ``SparseSeries.abs`` incorrectly keeps negative ``fill_value`` (:issue:`13853`)
+- Bug in single row slicing on multi-type ``SparseDataFrame`` s, types were previously forced to float (:issue:`13917`)
+- Bug in ``SparseSeries`` slicing changes integer dtype to float (:issue:`8292`)
+- Bug in ``SparseDataFarme`` comparison ops may raise ``TypeError`` (:issue:`13001`)
+- Bug in ``SparseDataFarme.isnull`` raises ``ValueError`` (:issue:`8276`)
+- Bug in ``SparseSeries`` representation with ``bool`` dtype may raise ``IndexError`` (:issue:`13110`)
+- Bug in ``SparseSeries`` and ``SparseDataFrame`` of ``bool`` or ``int64`` dtype may display its values like ``float64`` dtype (:issue:`13110`)
+- Bug in sparse indexing using ``SparseArray`` with ``bool`` dtype may return incorrect result  (:issue:`13985`)
+- Bug in ``SparseArray`` created from ``SparseSeries`` may lose ``dtype`` (:issue:`13999`)
+- Bug in ``SparseSeries`` comparison with dense returns normal ``Series`` rather than ``SparseSeries`` (:issue:`13999`)
+
+
+.. _whatsnew_0190.indexer_dtype:
+
+Indexer dtype changes
+^^^^^^^^^^^^^^^^^^^^^
+
+.. note::
+
+   This change only affects 64 bit python running on Windows, and only affects relatively advanced
+   indexing operations
+
+Methods such as ``Index.get_indexer`` that return an indexer array, coerce that array to a "platform int", so that it can be
+directly used in 3rd party library operations like ``numpy.take``.  Previously, a platform int was defined as ``np.int_``
+which corresponds to a C integer, but the correct type, and what is being used now, is ``np.intp``, which corresponds
+to the C integer size that can hold a pointer (:issue:`3033`, :issue:`13972`).
+
+These types are the same on many platform, but for 64 bit python on Windows,
+``np.int_`` is 32 bits, and ``np.intp`` is 64 bits.  Changing this behavior improves performance for many
+operations on that platform.
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [1]: i = pd.Index(['a', 'b', 'c'])
+
+   In [2]: i.get_indexer(['b', 'b', 'c']).dtype
+   Out[2]: dtype('int32')
+
+**New behavior**:
+
+.. code-block:: ipython
+
+   In [1]: i = pd.Index(['a', 'b', 'c'])
+
+   In [2]: i.get_indexer(['b', 'b', 'c']).dtype
+   Out[2]: dtype('int64')
+
+
+.. _whatsnew_0190.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- ``Timestamp.to_pydatetime`` will issue a ``UserWarning`` when ``warn=True``, and the instance has a non-zero number of nanoseconds, previously this would print a message to stdout (:issue:`14101`).
+- ``Series.unique()`` with datetime and timezone now returns return array of ``Timestamp`` with timezone (:issue:`13565`).
+- ``Panel.to_sparse()`` will raise a ``NotImplementedError`` exception when called (:issue:`13778`).
+- ``Index.reshape()`` will raise a ``NotImplementedError`` exception when called (:issue:`12882`).
+- ``.filter()`` enforces mutual exclusion of the keyword arguments (:issue:`12399`).
+- ``eval``'s upcasting rules for ``float32`` types have been updated to be more consistent with NumPy's rules.  New behavior will not upcast to ``float64`` if you multiply a pandas ``float32`` object by a scalar float64 (:issue:`12388`).
+- An ``UnsupportedFunctionCall`` error is now raised if NumPy ufuncs like ``np.mean`` are called on groupby or resample objects (:issue:`12811`).
+- ``__setitem__`` will no longer apply a callable rhs as a function instead of storing it. Call ``where`` directly to get the previous behavior (:issue:`13299`).
+- Calls to ``.sample()`` will respect the random seed set via ``numpy.random.seed(n)`` (:issue:`13161`)
+- ``Styler.apply`` is now more strict about the outputs your function must return. For ``axis=0`` or ``axis=1``, the output shape must be identical. For ``axis=None``, the output must be a DataFrame with identical columns and index labels (:issue:`13222`).
+- ``Float64Index.astype(int)`` will now raise ``ValueError`` if ``Float64Index`` contains ``NaN`` values (:issue:`13149`)
+- ``TimedeltaIndex.astype(int)`` and ``DatetimeIndex.astype(int)`` will now return ``Int64Index`` instead of ``np.array`` (:issue:`13209`)
+- Passing ``Period`` with multiple frequencies to normal ``Index`` now returns ``Index`` with ``object`` dtype (:issue:`13664`)
+- ``PeriodIndex.fillna`` with ``Period`` has different freq now coerces to ``object`` dtype (:issue:`13664`)
+- Faceted boxplots from ``DataFrame.boxplot(by=col)`` now return a ``Series`` when ``return_type`` is not None. Previously these returned an ``OrderedDict``. Note that when ``return_type=None``, the default, these still return a 2-D NumPy array (:issue:`12216`, :issue:`7096`).
+- ``pd.read_hdf`` will now raise a ``ValueError`` instead of ``KeyError``, if a mode other than ``r``, ``r+`` and ``a`` is supplied. (:issue:`13623`)
+- ``pd.read_csv()``, ``pd.read_table()``, and ``pd.read_hdf()`` raise the builtin ``FileNotFoundError`` exception for Python 3.x when called on a nonexistent file; this is back-ported as ``IOError`` in Python 2.x (:issue:`14086`)
+- More informative exceptions are passed through the csv parser. The exception type would now be the original exception type instead of ``CParserError`` (:issue:`13652`).
+- ``pd.read_csv()`` in the C engine will now issue a ``ParserWarning`` or raise a ``ValueError`` when ``sep`` encoded is more than one character long (:issue:`14065`)
+- ``DataFrame.values`` will now return ``float64`` with a ``DataFrame`` of mixed ``int64`` and ``uint64`` dtypes, conforming to ``np.find_common_type`` (:issue:`10364`, :issue:`13917`)
+- ``.groupby.groups`` will now return a dictionary of ``Index`` objects, rather than a dictionary of ``np.ndarray`` or ``lists`` (:issue:`14293`)
+
+.. _whatsnew_0190.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+- ``Series.reshape`` and ``Categorical.reshape`` have been deprecated and will be removed in a subsequent release (:issue:`12882`, :issue:`12882`)
+- ``PeriodIndex.to_datetime`` has been deprecated in favor of ``PeriodIndex.to_timestamp`` (:issue:`8254`)
+- ``Timestamp.to_datetime`` has been deprecated in favor of ``Timestamp.to_pydatetime`` (:issue:`8254`)
+- ``Index.to_datetime`` and ``DatetimeIndex.to_datetime`` have been deprecated in favor of ``pd.to_datetime`` (:issue:`8254`)
+- ``pandas.core.datetools`` module has been deprecated and will be removed in a subsequent release (:issue:`14094`)
+- ``SparseList`` has been deprecated and will be removed in a future version (:issue:`13784`)
+- ``DataFrame.to_html()`` and ``DataFrame.to_latex()`` have dropped the ``colSpace`` parameter in favor of ``col_space`` (:issue:`13857`)
+- ``DataFrame.to_sql()`` has deprecated the ``flavor`` parameter, as it is superfluous when SQLAlchemy is not installed (:issue:`13611`)
+- Deprecated ``read_csv`` keywords:
+
+  - ``compact_ints`` and ``use_unsigned`` have been deprecated and will be removed in a future version (:issue:`13320`)
+  - ``buffer_lines`` has been deprecated and will be removed in a future version (:issue:`13360`)
+  - ``as_recarray`` has been deprecated and will be removed in a future version (:issue:`13373`)
+  - ``skip_footer`` has been deprecated in favor of ``skipfooter`` and will be removed in a future version (:issue:`13349`)
+
+- top-level ``pd.ordered_merge()`` has been renamed to ``pd.merge_ordered()`` and the original name will be removed in a future version (:issue:`13358`)
+- ``Timestamp.offset`` property (and named arg in the constructor), has been deprecated in favor of ``freq`` (:issue:`12160`)
+- ``pd.tseries.util.pivot_annual`` is deprecated. Use ``pivot_table`` as alternative, an example is :ref:`here <cookbook.pivot>` (:issue:`736`)
+- ``pd.tseries.util.isleapyear`` has been deprecated and will be removed in a subsequent release. Datetime-likes now have a ``.is_leap_year`` property (:issue:`13727`)
+- ``Panel4D`` and ``PanelND`` constructors are deprecated and will be removed in a future version. The recommended way to represent these types of n-dimensional data are with the `xarray package <http://xarray.pydata.org/en/stable/>`__. pandas provides a :meth:`~Panel4D.to_xarray` method to automate this conversion (:issue:`13564`).
+- ``pandas.tseries.frequencies.get_standard_freq`` is deprecated. Use  ``pandas.tseries.frequencies.to_offset(freq).rule_code`` instead (:issue:`13874`)
+- ``pandas.tseries.frequencies.to_offset``'s ``freqstr`` keyword is deprecated in favor of ``freq`` (:issue:`13874`)
+- ``Categorical.from_array`` has been deprecated and will be removed in a future version (:issue:`13854`)
+
+.. _whatsnew_0190.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- The ``SparsePanel`` class has been removed (:issue:`13778`)
+- The ``pd.sandbox`` module has been removed in favor of the external library ``pandas-qt`` (:issue:`13670`)
+- The ``pandas.io.data`` and ``pandas.io.wb`` modules are removed in favor of
+  the `pandas-datareader package <https://github.com/pydata/pandas-datareader>`__ (:issue:`13724`).
+- The ``pandas.tools.rplot`` module has been removed in favor of
+  the `seaborn package <https://github.com/mwaskom/seaborn>`__ (:issue:`13855`)
+- ``DataFrame.to_csv()`` has dropped the ``engine`` parameter, as was deprecated in 0.17.1 (:issue:`11274`, :issue:`13419`)
+- ``DataFrame.to_dict()`` has dropped the ``outtype`` parameter in favor of ``orient`` (:issue:`13627`, :issue:`8486`)
+- ``pd.Categorical`` has dropped setting of the ``ordered`` attribute directly in favor of the ``set_ordered`` method (:issue:`13671`)
+- ``pd.Categorical`` has dropped the ``levels`` attribute in favor of ``categories`` (:issue:`8376`)
+- ``DataFrame.to_sql()`` has dropped the ``mysql`` option for the ``flavor`` parameter (:issue:`13611`)
+- ``Panel.shift()`` has dropped the ``lags`` parameter in favor of ``periods`` (:issue:`14041`)
+- ``pd.Index`` has dropped the ``diff`` method in favor of ``difference`` (:issue:`13669`)
+- ``pd.DataFrame`` has dropped the ``to_wide`` method in favor of ``to_panel`` (:issue:`14039`)
+- ``Series.to_csv`` has dropped the ``nanRep`` parameter in favor of ``na_rep`` (:issue:`13804`)
+- ``Series.xs``, ``DataFrame.xs``, ``Panel.xs``, ``Panel.major_xs``, and ``Panel.minor_xs`` have dropped the ``copy`` parameter (:issue:`13781`)
+- ``str.split`` has dropped the ``return_type`` parameter in favor of ``expand`` (:issue:`13701`)
+- Removal of the legacy time rules (offset aliases), deprecated since 0.17.0 (this has been alias since 0.8.0) (:issue:`13590`, :issue:`13868`). Now legacy time rules raises ``ValueError``. For the list of currently supported offsets, see :ref:`here <timeseries.offset_aliases>`.
+- The default value for the ``return_type`` parameter for ``DataFrame.plot.box`` and ``DataFrame.boxplot`` changed from ``None`` to ``"axes"``. These methods will now return a matplotlib axes by default instead of a dictionary of artists. See :ref:`here <visualization.box.return>` (:issue:`6581`).
+- The ``tquery`` and ``uquery`` functions in the ``pandas.io.sql`` module are removed (:issue:`5950`).
+
+
+.. _whatsnew_0190.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of sparse ``IntIndex.intersect`` (:issue:`13082`)
+- Improved performance of sparse arithmetic with ``BlockIndex`` when the number of blocks are large, though recommended to use ``IntIndex`` in such cases (:issue:`13082`)
+- Improved performance of ``DataFrame.quantile()`` as it now operates per-block (:issue:`11623`)
+- Improved performance of float64 hash table operations, fixing some very slow indexing and groupby operations in python 3 (:issue:`13166`, :issue:`13334`)
+- Improved performance of ``DataFrameGroupBy.transform`` (:issue:`12737`)
+- Improved performance of ``Index`` and ``Series`` ``.duplicated`` (:issue:`10235`)
+- Improved performance of ``Index.difference`` (:issue:`12044`)
+- Improved performance of ``RangeIndex.is_monotonic_increasing`` and ``is_monotonic_decreasing`` (:issue:`13749`)
+- Improved performance of datetime string parsing in ``DatetimeIndex`` (:issue:`13692`)
+- Improved performance of hashing ``Period`` (:issue:`12817`)
+- Improved performance of ``factorize`` of datetime with timezone (:issue:`13750`)
+- Improved performance of by lazily creating indexing hashtables on larger Indexes (:issue:`14266`)
+- Improved performance of ``groupby.groups`` (:issue:`14293`)
+- Unnecessary materializing of a MultiIndex when introspecting for memory usage (:issue:`14308`)
+
+.. _whatsnew_0190.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in ``groupby().shift()``, which could cause a segfault or corruption in rare circumstances when grouping by columns with missing values (:issue:`13813`)
+- Bug in ``groupby().cumsum()`` calculating ``cumprod`` when ``axis=1``. (:issue:`13994`)
+- Bug in ``pd.to_timedelta()`` in which the ``errors`` parameter was not being respected (:issue:`13613`)
+- Bug in ``io.json.json_normalize()``, where non-ascii keys raised an exception (:issue:`13213`)
+- Bug when passing a not-default-indexed ``Series`` as ``xerr`` or ``yerr`` in ``.plot()`` (:issue:`11858`)
+- Bug in area plot draws legend incorrectly if subplot is enabled or legend is moved after plot (matplotlib 1.5.0 is required to draw area plot legend properly) (:issue:`9161`, :issue:`13544`)
+- Bug in ``DataFrame`` assignment with an object-dtyped ``Index`` where the resultant column is mutable to the original object. (:issue:`13522`)
+- Bug in matplotlib ``AutoDataFormatter``; this restores the second scaled formatting and re-adds micro-second scaled formatting (:issue:`13131`)
+- Bug in selection from a ``HDFStore`` with a fixed format and ``start`` and/or ``stop`` specified will now return the selected range (:issue:`8287`)
+- Bug in ``Categorical.from_codes()`` where an unhelpful error was raised when an invalid ``ordered`` parameter was passed in (:issue:`14058`)
+- Bug in ``Series`` construction from a tuple of integers on windows not returning default dtype (int64) (:issue:`13646`)
+- Bug in ``TimedeltaIndex`` addition with a Datetime-like object where addition overflow was not being caught (:issue:`14068`)
+- Bug in ``.groupby(..).resample(..)`` when the same object is called multiple times (:issue:`13174`)
+- Bug in ``.to_records()`` when index name is a unicode string (:issue:`13172`)
+- Bug in calling ``.memory_usage()`` on object which doesn't implement (:issue:`12924`)
+- Regression in ``Series.quantile`` with nans (also shows up in ``.median()`` and ``.describe()`` ); furthermore now names the ``Series`` with the quantile (:issue:`13098`, :issue:`13146`)
+- Bug in ``SeriesGroupBy.transform`` with datetime values and missing groups (:issue:`13191`)
+- Bug where empty ``Series`` were incorrectly coerced in datetime-like numeric operations (:issue:`13844`)
+- Bug in ``Categorical`` constructor when passed a ``Categorical`` containing datetimes with timezones (:issue:`14190`)
+- Bug in ``Series.str.extractall()`` with ``str`` index raises ``ValueError``  (:issue:`13156`)
+- Bug in ``Series.str.extractall()`` with single group and quantifier  (:issue:`13382`)
+- Bug in ``DatetimeIndex`` and ``Period`` subtraction raises ``ValueError`` or ``AttributeError`` rather than ``TypeError`` (:issue:`13078`)
+- Bug in ``Index`` and ``Series`` created with ``NaN`` and ``NaT`` mixed data may not have ``datetime64`` dtype  (:issue:`13324`)
+- Bug in ``Index`` and ``Series`` may ignore ``np.datetime64('nat')`` and ``np.timdelta64('nat')`` to infer dtype (:issue:`13324`)
+- Bug in ``PeriodIndex`` and ``Period`` subtraction raises ``AttributeError`` (:issue:`13071`)
+- Bug in ``PeriodIndex`` construction returning a ``float64`` index in some circumstances (:issue:`13067`)
+- Bug in ``.resample(..)`` with a ``PeriodIndex`` not changing its ``freq`` appropriately when empty (:issue:`13067`)
+- Bug in ``.resample(..)`` with a ``PeriodIndex`` not retaining its type or name with an empty ``DataFrame`` appropriately when empty (:issue:`13212`)
+- Bug in ``groupby(..).apply(..)`` when the passed function returns scalar values per group (:issue:`13468`).
+- Bug in ``groupby(..).resample(..)`` where passing some keywords would raise an exception (:issue:`13235`)
+- Bug in ``.tz_convert`` on a tz-aware ``DateTimeIndex`` that relied on index being sorted for correct results (:issue:`13306`)
+- Bug in ``.tz_localize`` with ``dateutil.tz.tzlocal`` may return incorrect result (:issue:`13583`)
+- Bug in ``DatetimeTZDtype`` dtype with ``dateutil.tz.tzlocal`` cannot be regarded as valid dtype (:issue:`13583`)
+- Bug in ``pd.read_hdf()`` where attempting to load an HDF file with a single dataset, that had one or more categorical columns, failed unless the key argument was set to the name of the dataset. (:issue:`13231`)
+- Bug in ``.rolling()`` that allowed a negative integer window in construction of the ``Rolling()`` object, but would later fail on aggregation (:issue:`13383`)
+- Bug in ``Series`` indexing with tuple-valued data and a numeric index (:issue:`13509`)
+- Bug in printing ``pd.DataFrame`` where unusual elements with the ``object`` dtype were causing segfaults (:issue:`13717`)
+- Bug in ranking ``Series`` which could result in segfaults (:issue:`13445`)
+- Bug in various index types, which did not propagate the name of passed index (:issue:`12309`)
+- Bug in ``DatetimeIndex``, which did not honour the ``copy=True`` (:issue:`13205`)
+- Bug in ``DatetimeIndex.is_normalized`` returns incorrectly for normalized date_range in case of local timezones (:issue:`13459`)
+- Bug in ``pd.concat`` and ``.append`` may coerces ``datetime64`` and ``timedelta`` to ``object`` dtype containing python built-in ``datetime`` or ``timedelta`` rather than ``Timestamp`` or ``Timedelta`` (:issue:`13626`)
+- Bug in ``PeriodIndex.append`` may raises ``AttributeError`` when the result is ``object`` dtype (:issue:`13221`)
+- Bug in ``CategoricalIndex.append`` may accept normal ``list`` (:issue:`13626`)
+- Bug in ``pd.concat`` and ``.append`` with the same timezone get reset to UTC (:issue:`7795`)
+- Bug in ``Series`` and ``DataFrame`` ``.append`` raises ``AmbiguousTimeError`` if data contains datetime near DST boundary (:issue:`13626`)
+- Bug in ``DataFrame.to_csv()`` in which float values were being quoted even though quotations were specified for non-numeric values only (:issue:`12922`, :issue:`13259`)
+- Bug in ``DataFrame.describe()`` raising ``ValueError`` with only boolean columns (:issue:`13898`)
+- Bug in ``MultiIndex`` slicing where extra elements were returned when level is non-unique (:issue:`12896`)
+- Bug in ``.str.replace`` does not raise ``TypeError`` for invalid replacement (:issue:`13438`)
+- Bug in ``MultiIndex.from_arrays`` which didn't check for input array lengths matching (:issue:`13599`)
+- Bug in ``cartesian_product`` and ``MultiIndex.from_product`` which may raise with empty input arrays (:issue:`12258`)
+- Bug in ``pd.read_csv()`` which may cause a segfault or corruption when iterating in large chunks over a stream/file under rare circumstances (:issue:`13703`)
+- Bug in ``pd.read_csv()`` which caused errors to be raised when a dictionary containing scalars is passed in for ``na_values`` (:issue:`12224`)
+- Bug in ``pd.read_csv()`` which caused BOM files to be incorrectly parsed by not ignoring the BOM (:issue:`4793`)
+- Bug in ``pd.read_csv()`` with ``engine='python'`` which raised errors when a numpy array was passed in for ``usecols`` (:issue:`12546`)
+- Bug in ``pd.read_csv()`` where the index columns were being incorrectly parsed when parsed as dates with a ``thousands`` parameter (:issue:`14066`)
+- Bug in ``pd.read_csv()`` with ``engine='python'`` in which ``NaN`` values weren't being detected after data was converted to numeric values (:issue:`13314`)
+- Bug in ``pd.read_csv()`` in which the ``nrows`` argument was not properly validated for both engines (:issue:`10476`)
+- Bug in ``pd.read_csv()`` with ``engine='python'`` in which infinities of mixed-case forms were not being interpreted properly (:issue:`13274`)
+- Bug in ``pd.read_csv()`` with ``engine='python'`` in which trailing ``NaN`` values were not being parsed (:issue:`13320`)
+- Bug in ``pd.read_csv()`` with ``engine='python'`` when reading from a ``tempfile.TemporaryFile`` on Windows with Python 3 (:issue:`13398`)
+- Bug in ``pd.read_csv()`` that prevents ``usecols`` kwarg from accepting single-byte unicode strings (:issue:`13219`)
+- Bug in ``pd.read_csv()`` that prevents ``usecols`` from being an empty set (:issue:`13402`)
+- Bug in ``pd.read_csv()`` in the C engine where the NULL character was not being parsed as NULL (:issue:`14012`)
+- Bug in ``pd.read_csv()`` with ``engine='c'`` in which NULL ``quotechar`` was not accepted even though ``quoting`` was specified as ``None`` (:issue:`13411`)
+- Bug in ``pd.read_csv()`` with ``engine='c'`` in which fields were not properly cast to float when quoting was specified as non-numeric (:issue:`13411`)
+- Bug in ``pd.read_csv()`` in Python 2.x with non-UTF8 encoded, multi-character separated data (:issue:`3404`)
+- Bug in ``pd.read_csv()``, where aliases for utf-xx (e.g. UTF-xx, UTF_xx, utf_xx) raised UnicodeDecodeError (:issue:`13549`)
+- Bug in ``pd.read_csv``, ``pd.read_table``, ``pd.read_fwf``, ``pd.read_stata`` and ``pd.read_sas`` where files were opened by parsers but not closed if both ``chunksize`` and ``iterator`` were ``None``. (:issue:`13940`)
+- Bug in ``StataReader``, ``StataWriter``, ``XportReader`` and ``SAS7BDATReader`` where a file was not properly closed when an error was raised. (:issue:`13940`)
+- Bug in ``pd.pivot_table()`` where ``margins_name`` is ignored when ``aggfunc`` is a list (:issue:`13354`)
+- Bug in ``pd.Series.str.zfill``, ``center``, ``ljust``, ``rjust``, and ``pad`` when passing non-integers, did not raise ``TypeError`` (:issue:`13598`)
+- Bug in checking for any null objects in a ``TimedeltaIndex``, which always returned ``True`` (:issue:`13603`)
+- Bug in ``Series`` arithmetic raises ``TypeError`` if it contains datetime-like as ``object`` dtype (:issue:`13043`)
+- Bug ``Series.isnull()`` and ``Series.notnull()`` ignore ``Period('NaT')``  (:issue:`13737`)
+- Bug ``Series.fillna()`` and ``Series.dropna()`` don't affect to ``Period('NaT')``  (:issue:`13737`
+- Bug in ``.fillna(value=np.nan)`` incorrectly raises ``KeyError`` on a ``category`` dtyped ``Series`` (:issue:`14021`)
+- Bug in extension dtype creation where the created types were not is/identical (:issue:`13285`)
+- Bug in ``.resample(..)`` where incorrect warnings were triggered by IPython introspection (:issue:`13618`)
+- Bug in ``NaT`` - ``Period`` raises ``AttributeError`` (:issue:`13071`)
+- Bug in ``Series`` comparison may output incorrect result if rhs contains ``NaT`` (:issue:`9005`)
+- Bug in ``Series`` and ``Index`` comparison may output incorrect result if it contains ``NaT`` with ``object`` dtype (:issue:`13592`)
+- Bug in ``Period`` addition raises ``TypeError`` if ``Period`` is on right hand side (:issue:`13069`)
+- Bug in ``Period`` and ``Series`` or ``Index`` comparison raises ``TypeError`` (:issue:`13200`)
+- Bug in ``pd.set_eng_float_format()`` that would prevent NaN and Inf from formatting (:issue:`11981`)
+- Bug in ``.unstack`` with ``Categorical`` dtype resets ``.ordered`` to ``True`` (:issue:`13249`)
+- Clean some compile time warnings in datetime parsing (:issue:`13607`)
+- Bug in ``factorize`` raises ``AmbiguousTimeError`` if data contains datetime near DST boundary (:issue:`13750`)
+- Bug in ``.set_index`` raises ``AmbiguousTimeError`` if new index contains DST boundary and multi levels (:issue:`12920`)
+- Bug in ``.shift`` raises ``AmbiguousTimeError`` if data contains datetime near DST boundary (:issue:`13926`)
+- Bug in ``pd.read_hdf()`` returns incorrect result when a ``DataFrame`` with a ``categorical`` column and a query which doesn't match any values (:issue:`13792`)
+- Bug in ``.iloc`` when indexing with a non lexsorted MultiIndex (:issue:`13797`)
+- Bug in ``.loc`` when indexing with date strings in a reverse sorted ``DatetimeIndex`` (:issue:`14316`)
+- Bug in ``Series`` comparison operators when dealing with zero dim NumPy arrays (:issue:`13006`)
+- Bug in ``.combine_first`` may return incorrect ``dtype`` (:issue:`7630`, :issue:`10567`)
+- Bug in ``groupby`` where ``apply`` returns different result depending on whether first result is ``None`` or not (:issue:`12824`)
+- Bug in ``groupby(..).nth()`` where the group key is included inconsistently if called after ``.head()/.tail()`` (:issue:`12839`)
+- Bug in ``.to_html``, ``.to_latex`` and ``.to_string`` silently ignore custom datetime formatter passed through the ``formatters`` key word (:issue:`10690`)
+- Bug in ``DataFrame.iterrows()``, not yielding a ``Series`` subclasse if defined (:issue:`13977`)
+- Bug in ``pd.to_numeric`` when ``errors='coerce'`` and input contains non-hashable objects (:issue:`13324`)
+- Bug in invalid ``Timedelta`` arithmetic and comparison may raise ``ValueError`` rather than ``TypeError`` (:issue:`13624`)
+- Bug in invalid datetime parsing in ``to_datetime`` and ``DatetimeIndex`` may raise ``TypeError`` rather than ``ValueError`` (:issue:`11169`, :issue:`11287`)
+- Bug in ``Index`` created with tz-aware ``Timestamp`` and mismatched ``tz`` option incorrectly coerces timezone (:issue:`13692`)
+- Bug in ``DatetimeIndex`` with nanosecond frequency does not include timestamp specified with ``end`` (:issue:`13672`)
+- Bug in ``Series`` when setting a slice with a ``np.timedelta64`` (:issue:`14155`)
+- Bug in ``Index`` raises ``OutOfBoundsDatetime`` if ``datetime`` exceeds ``datetime64[ns]`` bounds, rather than coercing to ``object`` dtype (:issue:`13663`)
+- Bug in ``Index`` may ignore specified ``datetime64`` or ``timedelta64`` passed as ``dtype``  (:issue:`13981`)
+- Bug in ``RangeIndex`` can be created without no arguments rather than raises ``TypeError`` (:issue:`13793`)
+- Bug in ``.value_counts()`` raises ``OutOfBoundsDatetime`` if data exceeds ``datetime64[ns]`` bounds (:issue:`13663`)
+- Bug in ``DatetimeIndex`` may raise ``OutOfBoundsDatetime`` if input ``np.datetime64`` has other unit than ``ns`` (:issue:`9114`)
+- Bug in ``Series`` creation with ``np.datetime64`` which has other unit than ``ns`` as ``object`` dtype results in incorrect values (:issue:`13876`)
+- Bug in ``resample`` with timedelta data where data was casted to float (:issue:`13119`).
+- Bug in ``pd.isnull()`` ``pd.notnull()`` raise ``TypeError`` if input datetime-like has other unit than ``ns`` (:issue:`13389`)
+- Bug in ``pd.merge()`` may raise ``TypeError`` if input datetime-like has other unit than ``ns`` (:issue:`13389`)
+- Bug in ``HDFStore``/``read_hdf()`` discarded ``DatetimeIndex.name`` if ``tz`` was set (:issue:`13884`)
+- Bug in ``Categorical.remove_unused_categories()`` changes ``.codes`` dtype to platform int (:issue:`13261`)
+- Bug in ``groupby`` with ``as_index=False`` returns all NaN's when grouping on multiple columns including a categorical one (:issue:`13204`)
+- Bug in ``df.groupby(...)[...]`` where getitem with ``Int64Index`` raised an error (:issue:`13731`)
+- Bug in the CSS classes assigned to ``DataFrame.style`` for index names. Previously they were assigned ``"col_heading level<n> col<c>"`` where ``n`` was the number of levels + 1. Now they are assigned ``"index_name level<n>"``, where ``n`` is the correct level for that MultiIndex.
+- Bug where ``pd.read_gbq()`` could throw ``ImportError: No module named discovery`` as a result of a naming conflict with another python package called apiclient  (:issue:`13454`)
+- Bug in ``Index.union`` returns an incorrect result with a named empty index (:issue:`13432`)
+- Bugs in ``Index.difference`` and ``DataFrame.join`` raise in Python3 when using mixed-integer indexes (:issue:`13432`, :issue:`12814`)
+- Bug in subtract tz-aware ``datetime.datetime`` from tz-aware ``datetime64`` series (:issue:`14088`)
+- Bug in ``.to_excel()`` when DataFrame contains a MultiIndex which contains a label with a NaN value (:issue:`13511`)
+- Bug in invalid frequency offset string like "D1", "-2-3H" may not raise ``ValueError`` (:issue:`13930`)
+- Bug in ``concat`` and ``groupby`` for hierarchical frames with ``RangeIndex`` levels (:issue:`13542`).
+- Bug in ``Series.str.contains()`` for Series containing only ``NaN`` values of ``object`` dtype (:issue:`14171`)
+- Bug in ``agg()`` function on groupby dataframe changes dtype of ``datetime64[ns]`` column to ``float64`` (:issue:`12821`)
+- Bug in using NumPy ufunc with ``PeriodIndex`` to add or subtract integer raise ``IncompatibleFrequency``. Note that using standard operator like ``+`` or ``-`` is recommended, because standard operators use more efficient path (:issue:`13980`)
+- Bug in operations on ``NaT`` returning ``float`` instead of ``datetime64[ns]`` (:issue:`12941`)
+- Bug in ``Series`` flexible arithmetic methods (like ``.add()``) raises ``ValueError`` when ``axis=None`` (:issue:`13894`)
+- Bug in ``DataFrame.to_csv()`` with ``MultiIndex`` columns in which a stray empty line was added (:issue:`6618`)
+- Bug in ``DatetimeIndex``, ``TimedeltaIndex`` and ``PeriodIndex.equals()`` may return ``True`` when input isn't ``Index`` but contains the same values (:issue:`13107`)
+- Bug in assignment against datetime with timezone may not work if it contains datetime near DST boundary (:issue:`14146`)
+- Bug in ``pd.eval()`` and ``HDFStore`` query truncating long float literals with python 2 (:issue:`14241`)
+- Bug in ``Index`` raises ``KeyError`` displaying incorrect column when column is not in the df and columns contains duplicate values (:issue:`13822`)
+- Bug in ``Period`` and ``PeriodIndex`` creating wrong dates when frequency has combined offset aliases (:issue:`13874`)
+- Bug in ``.to_string()`` when called with an integer ``line_width`` and ``index=False`` raises an UnboundLocalError exception because ``idx`` referenced before assignment.
+- Bug in ``eval()`` where the ``resolvers`` argument would not accept a list (:issue:`14095`)
+- Bugs in ``stack``, ``get_dummies``, ``make_axis_dummies`` which don't preserve categorical dtypes in (multi)indexes (:issue:`13854`)
+- ``PeriodIndex`` can now accept ``list`` and ``array`` which contains ``pd.NaT`` (:issue:`13430`)
+- Bug in ``df.groupby`` where ``.median()`` returns arbitrary values if grouped dataframe contains empty bins (:issue:`13629`)
+- Bug in ``Index.copy()`` where ``name`` parameter was ignored (:issue:`14302`)
+
+
+.. _whatsnew_0.19.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.18.1..v0.19.0
diff --git a/pandas/doc/source/whatsnew/v0.19.1.rst b/pandas/doc/source/whatsnew/v0.19.1.rst
new file mode 100644
index 0000000000000..6ff3fb6900a99
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.19.1.rst
@@ -0,0 +1,77 @@
+.. _whatsnew_0191:
+
+Version 0.19.1 (November 3, 2016)
+---------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a minor bug-fix release from 0.19.0 and includes some small regression fixes,
+bug fixes and performance improvements.
+We recommend that all users upgrade to this version.
+
+.. contents:: What's new in v0.19.1
+    :local:
+    :backlinks: none
+
+
+.. _whatsnew_0191.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Fixed performance regression in factorization of ``Period`` data (:issue:`14338`)
+- Fixed performance regression in ``Series.asof(where)`` when ``where`` is a scalar (:issue:`14461`)
+- Improved performance in ``DataFrame.asof(where)`` when ``where`` is a scalar (:issue:`14461`)
+- Improved performance in ``.to_json()`` when ``lines=True`` (:issue:`14408`)
+- Improved performance in certain types of ``loc`` indexing with a MultiIndex (:issue:`14551`).
+
+
+.. _whatsnew_0191.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Source installs from PyPI will now again work without ``cython`` installed, as in previous versions (:issue:`14204`)
+- Compat with Cython 0.25 for building (:issue:`14496`)
+- Fixed regression where user-provided file handles were closed in ``read_csv`` (c engine) (:issue:`14418`).
+- Fixed regression in ``DataFrame.quantile`` when missing values where present in some columns (:issue:`14357`).
+- Fixed regression in ``Index.difference`` where the ``freq`` of a ``DatetimeIndex`` was incorrectly set (:issue:`14323`)
+- Added back ``pandas.core.common.array_equivalent`` with a deprecation warning (:issue:`14555`).
+- Bug in ``pd.read_csv`` for the C engine in which quotation marks were improperly parsed in skipped rows (:issue:`14459`)
+- Bug in ``pd.read_csv`` for Python 2.x in which Unicode quote characters were no longer being respected (:issue:`14477`)
+- Fixed regression in ``Index.append`` when categorical indices were appended (:issue:`14545`).
+- Fixed regression in ``pd.DataFrame`` where constructor fails when given dict with ``None`` value (:issue:`14381`)
+- Fixed regression in ``DatetimeIndex._maybe_cast_slice_bound`` when index is empty (:issue:`14354`).
+- Bug in localizing an ambiguous timezone when a boolean is passed (:issue:`14402`)
+- Bug in ``TimedeltaIndex`` addition with a Datetime-like object where addition overflow in the negative direction was not being caught (:issue:`14068`, :issue:`14453`)
+- Bug in string indexing against data with ``object`` ``Index`` may raise ``AttributeError`` (:issue:`14424`)
+- Correctly raise ``ValueError`` on empty input to ``pd.eval()`` and ``df.query()`` (:issue:`13139`)
+- Bug in ``RangeIndex.intersection`` when result is a empty set (:issue:`14364`).
+- Bug in groupby-transform broadcasting that could cause incorrect dtype coercion (:issue:`14457`)
+- Bug in ``Series.__setitem__`` which allowed mutating read-only arrays (:issue:`14359`).
+- Bug in ``DataFrame.insert`` where multiple calls with duplicate columns can fail (:issue:`14291`)
+- ``pd.merge()`` will raise ``ValueError`` with non-boolean parameters in passed boolean type arguments (:issue:`14434`)
+- Bug in ``Timestamp`` where dates very near the minimum (1677-09) could underflow on creation (:issue:`14415`)
+- Bug in ``pd.concat`` where names of the ``keys`` were not propagated to the resulting ``MultiIndex`` (:issue:`14252`)
+- Bug in ``pd.concat`` where ``axis`` cannot take string parameters ``'rows'`` or ``'columns'`` (:issue:`14369`)
+- Bug in ``pd.concat`` with dataframes heterogeneous in length and tuple ``keys`` (:issue:`14438`)
+- Bug in ``MultiIndex.set_levels`` where illegal level values were still set after raising an error (:issue:`13754`)
+- Bug in ``DataFrame.to_json`` where ``lines=True`` and a value contained a ``}`` character (:issue:`14391`)
+- Bug in ``df.groupby`` causing an ``AttributeError`` when grouping a single index frame by a column and the index level (:issue:`14327`)
+- Bug in ``df.groupby`` where ``TypeError`` raised when ``pd.Grouper(key=...)`` is passed in a list (:issue:`14334`)
+- Bug in ``pd.pivot_table`` may raise ``TypeError`` or ``ValueError`` when ``index`` or ``columns``
+  is not scalar and ``values`` is not specified (:issue:`14380`)
+
+
+.. _whatsnew_0.19.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.19.0..v0.19.1
diff --git a/pandas/doc/source/whatsnew/v0.19.2.rst b/pandas/doc/source/whatsnew/v0.19.2.rst
new file mode 100644
index 0000000000000..db9d9e65c923d
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.19.2.rst
@@ -0,0 +1,98 @@
+.. _whatsnew_0192:
+
+Version 0.19.2 (December 24, 2016)
+----------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a minor bug-fix release in the 0.19.x series and includes some small regression fixes,
+bug fixes and performance improvements.
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- Compatibility with Python 3.6
+- Added a `Pandas Cheat Sheet <https://github.com/pandas-dev/pandas/tree/main/doc/cheatsheet/Pandas_Cheat_Sheet.pdf>`__. (:issue:`13202`).
+
+
+.. contents:: What's new in v0.19.2
+    :local:
+    :backlinks: none
+
+
+.. _whatsnew_0192.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+The ``pd.merge_asof()``, added in 0.19.0, gained some improvements:
+
+- ``pd.merge_asof()`` gained ``left_index``/``right_index`` and ``left_by``/``right_by`` arguments (:issue:`14253`)
+- ``pd.merge_asof()`` can take multiple columns in ``by`` parameter and has specialized dtypes for better performance (:issue:`13936`)
+
+
+.. _whatsnew_0192.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Performance regression with ``PeriodIndex`` (:issue:`14822`)
+- Performance regression in indexing with getitem (:issue:`14930`)
+- Improved performance of ``.replace()`` (:issue:`12745`)
+- Improved performance ``Series`` creation with a datetime index and dictionary data (:issue:`14894`)
+
+
+.. _whatsnew_0192.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Compat with python 3.6 for pickling of some offsets (:issue:`14685`)
+- Compat with python 3.6 for some indexing exception types (:issue:`14684`, :issue:`14689`)
+- Compat with python 3.6 for deprecation warnings in the test suite (:issue:`14681`)
+- Compat with python 3.6 for Timestamp pickles (:issue:`14689`)
+- Compat with ``dateutil==2.6.0``; segfault reported in the testing suite (:issue:`14621`)
+- Allow ``nanoseconds`` in ``Timestamp.replace`` as a kwarg (:issue:`14621`)
+- Bug in ``pd.read_csv`` in which aliasing was being done for ``na_values`` when passed in as a dictionary (:issue:`14203`)
+- Bug in ``pd.read_csv`` in which column indices for a dict-like ``na_values`` were not being respected (:issue:`14203`)
+- Bug in ``pd.read_csv`` where reading files fails, if the number of headers is equal to the number of lines in the file (:issue:`14515`)
+- Bug in ``pd.read_csv`` for the Python engine in which an unhelpful error message was being raised when multi-char delimiters were not being respected with quotes (:issue:`14582`)
+- Fix bugs (:issue:`14734`, :issue:`13654`) in ``pd.read_sas`` and ``pandas.io.sas.sas7bdat.SAS7BDATReader`` that caused problems when reading a SAS file incrementally.
+- Bug in ``pd.read_csv`` for the Python engine in which an unhelpful error message was being raised when ``skipfooter`` was not being respected by Python's CSV library (:issue:`13879`)
+- Bug in ``.fillna()`` in which timezone aware datetime64 values were incorrectly rounded (:issue:`14872`)
+- Bug in ``.groupby(..., sort=True)`` of a non-lexsorted MultiIndex when grouping with multiple levels (:issue:`14776`)
+- Bug in ``pd.cut`` with negative values and a single bin (:issue:`14652`)
+- Bug in ``pd.to_numeric`` where a 0 was not unsigned on a ``downcast='unsigned'`` argument (:issue:`14401`)
+- Bug in plotting regular and irregular timeseries using shared axes
+  (``sharex=True`` or ``ax.twinx()``) (:issue:`13341`, :issue:`14322`).
+- Bug in not propagating exceptions in parsing invalid datetimes, noted in python 3.6 (:issue:`14561`)
+- Bug in resampling a ``DatetimeIndex`` in local TZ, covering a DST change, which would raise ``AmbiguousTimeError`` (:issue:`14682`)
+- Bug in indexing that transformed ``RecursionError`` into ``KeyError`` or ``IndexingError`` (:issue:`14554`)
+- Bug in ``HDFStore`` when writing a ``MultiIndex`` when using ``data_columns=True`` (:issue:`14435`)
+- Bug in ``HDFStore.append()`` when writing a ``Series`` and passing a ``min_itemsize`` argument containing a value for the ``index`` (:issue:`11412`)
+- Bug when writing to a ``HDFStore`` in ``table`` format with a ``min_itemsize`` value for the ``index`` and without asking to append (:issue:`10381`)
+- Bug in ``Series.groupby.nunique()`` raising an ``IndexError`` for an empty ``Series`` (:issue:`12553`)
+- Bug in ``DataFrame.nlargest`` and ``DataFrame.nsmallest`` when the index had duplicate values (:issue:`13412`)
+- Bug in clipboard functions on linux with python2 with unicode and separators (:issue:`13747`)
+- Bug in clipboard functions on Windows 10 and python 3 (:issue:`14362`, :issue:`12807`)
+- Bug in ``.to_clipboard()`` and Excel compat (:issue:`12529`)
+- Bug in ``DataFrame.combine_first()`` for integer columns (:issue:`14687`).
+- Bug in ``pd.read_csv()`` in which the ``dtype`` parameter was not being respected for empty data (:issue:`14712`)
+- Bug in ``pd.read_csv()`` in which the ``nrows`` parameter was not being respected for large input when using the C engine for parsing (:issue:`7626`)
+- Bug in ``pd.merge_asof()`` could not handle timezone-aware DatetimeIndex when a tolerance was specified (:issue:`14844`)
+- Explicit check in ``to_stata`` and ``StataWriter`` for out-of-range values when writing doubles (:issue:`14618`)
+- Bug in ``.plot(kind='kde')`` which did not drop missing values to generate the KDE Plot, instead generating an empty plot. (:issue:`14821`)
+- Bug in ``unstack()`` if called with a list of column(s) as an argument, regardless of the dtypes of all columns, they get coerced to ``object`` (:issue:`11847`)
+
+
+.. _whatsnew_0.19.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.19.1..v0.19.2
diff --git a/pandas/doc/source/whatsnew/v0.20.0.rst b/pandas/doc/source/whatsnew/v0.20.0.rst
new file mode 100644
index 0000000000000..d6d1d96ccc878
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.20.0.rst
@@ -0,0 +1,1839 @@
+.. _whatsnew_0200:
+
+Version 0.20.1 (May 5, 2017)
+----------------------------
+
+{{ header }}
+
+This is a major release from 0.19.2 and includes a number of API changes, deprecations, new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+Highlights include:
+
+- New ``.agg()`` API for Series/DataFrame similar to the groupby-rolling-resample API's, see :ref:`here <whatsnew_0200.enhancements.agg>`
+- Integration with the ``feather-format``, including a new top-level ``pd.read_feather()`` and ``DataFrame.to_feather()`` method, see :ref:`here <io.feather>`.
+- The ``.ix`` indexer has been deprecated, see :ref:`here <whatsnew_0200.api_breaking.deprecate_ix>`
+- ``Panel`` has been deprecated, see :ref:`here <whatsnew_0200.api_breaking.deprecate_panel>`
+- Addition of an ``IntervalIndex`` and ``Interval`` scalar type, see :ref:`here <whatsnew_0200.enhancements.intervalindex>`
+- Improved user API when grouping by index levels in ``.groupby()``, see :ref:`here <whatsnew_0200.enhancements.groupby_access>`
+- Improved support for ``UInt64`` dtypes, see :ref:`here <whatsnew_0200.enhancements.uint64_support>`
+- A new orient for JSON serialization, ``orient='table'``, that uses the Table Schema spec and that gives the possibility for a more interactive repr in the Jupyter Notebook, see :ref:`here <whatsnew_0200.enhancements.table_schema>`
+- Experimental support for exporting styled DataFrames (``DataFrame.style``) to Excel, see :ref:`here <whatsnew_0200.enhancements.style_excel>`
+- Window binary corr/cov operations now return a MultiIndexed ``DataFrame`` rather than a ``Panel``, as ``Panel`` is now deprecated, see :ref:`here <whatsnew_0200.api_breaking.rolling_pairwise>`
+- Support for S3 handling now uses ``s3fs``, see :ref:`here <whatsnew_0200.api_breaking.s3>`
+- Google BigQuery support now uses the ``pandas-gbq`` library, see :ref:`here <whatsnew_0200.api_breaking.gbq>`
+
+.. warning::
+
+  pandas has changed the internal structure and layout of the code base.
+  This can affect imports that are not from the top-level ``pandas.*`` namespace, please see the changes :ref:`here <whatsnew_0200.privacy>`.
+
+Check the :ref:`API Changes <whatsnew_0200.api_breaking>` and :ref:`deprecations <whatsnew_0200.deprecations>` before updating.
+
+.. note::
+
+   This is a combined release for 0.20.0 and 0.20.1.
+   Version 0.20.1 contains one additional change for backwards-compatibility with downstream projects using pandas' ``utils`` routines. (:issue:`16250`)
+
+.. contents:: What's new in v0.20.0
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0200.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0200.enhancements.agg:
+
+Method ``agg`` API for DataFrame/Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Series & DataFrame have been enhanced to support the aggregation API. This is a familiar API
+from groupby, window operations, and resampling. This allows aggregation operations in a concise way
+by using :meth:`~DataFrame.agg` and :meth:`~DataFrame.transform`. The full documentation
+is :ref:`here <basics.aggregate>` (:issue:`1623`).
+
+Here is a sample
+
+.. ipython:: python
+
+   df = pd.DataFrame(np.random.randn(10, 3), columns=['A', 'B', 'C'],
+                     index=pd.date_range('1/1/2000', periods=10))
+   df.iloc[3:7] = np.nan
+   df
+
+One can operate using string function names, callables, lists, or dictionaries of these.
+
+Using a single function is equivalent to ``.apply``.
+
+.. ipython:: python
+
+   df.agg('sum')
+
+Multiple aggregations with a list of functions.
+
+.. ipython:: python
+
+   df.agg(['sum', 'min'])
+
+Using a dict provides the ability to apply specific aggregations per column.
+You will get a matrix-like output of all of the aggregators. The output has one column
+per unique function. Those functions applied to a particular column will be ``NaN``:
+
+.. ipython:: python
+
+   df.agg({'A': ['sum', 'min'], 'B': ['min', 'max']})
+
+The API also supports a ``.transform()`` function for broadcasting results.
+
+.. ipython:: python
+   :okwarning:
+
+   df.transform(['abs', lambda x: x - x.min()])
+
+When presented with mixed dtypes that cannot be aggregated, ``.agg()`` will only take the valid
+aggregations. This is similar to how groupby ``.agg()`` works. (:issue:`15015`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 2, 3],
+                      'B': [1., 2., 3.],
+                      'C': ['foo', 'bar', 'baz'],
+                      'D': pd.date_range('20130101', periods=3)})
+   df.dtypes
+
+.. code-block:: python
+
+   In [10]: df.agg(['min', 'sum'])
+   Out[10]:
+        A    B          C          D
+   min  1  1.0        bar 2013-01-01
+   sum  6  6.0  foobarbaz        NaT
+
+.. _whatsnew_0200.enhancements.dataio_dtype:
+
+Keyword argument ``dtype`` for data IO
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``'python'`` engine for :func:`read_csv`, as well as the :func:`read_fwf` function for parsing
+fixed-width text files and :func:`read_excel` for parsing Excel files, now accept the ``dtype`` keyword argument for specifying the types of specific columns (:issue:`14295`). See the :ref:`io docs <io.dtypes>` for more information.
+
+.. ipython:: python
+   :suppress:
+
+   from io import StringIO
+
+.. ipython:: python
+
+   data = "a  b\n1  2\n3  4"
+   pd.read_fwf(StringIO(data)).dtypes
+   pd.read_fwf(StringIO(data), dtype={'a': 'float64', 'b': 'object'}).dtypes
+
+.. _whatsnew_0120.enhancements.datetime_origin:
+
+Method ``.to_datetime()`` has gained an ``origin`` parameter
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`to_datetime` has gained a new parameter, ``origin``, to define a reference date
+from where to compute the resulting timestamps when parsing numerical values with a specific ``unit`` specified. (:issue:`11276`, :issue:`11745`)
+
+For example, with 1960-01-01 as the starting date:
+
+.. ipython:: python
+
+   pd.to_datetime([1, 2, 3], unit='D', origin=pd.Timestamp('1960-01-01'))
+
+The default is set at ``origin='unix'``, which defaults to ``1970-01-01 00:00:00``, which is
+commonly called 'unix epoch' or POSIX time. This was the previous default, so this is a backward compatible change.
+
+.. ipython:: python
+
+   pd.to_datetime([1, 2, 3], unit='D')
+
+
+.. _whatsnew_0200.enhancements.groupby_access:
+
+GroupBy enhancements
+^^^^^^^^^^^^^^^^^^^^
+
+Strings passed to ``DataFrame.groupby()`` as the ``by`` parameter may now reference either column names or index level names. Previously, only column names could be referenced. This allows to easily group by a column and index level at the same time. (:issue:`5677`)
+
+.. ipython:: python
+
+   arrays = [['bar', 'bar', 'baz', 'baz', 'foo', 'foo', 'qux', 'qux'],
+             ['one', 'two', 'one', 'two', 'one', 'two', 'one', 'two']]
+
+   index = pd.MultiIndex.from_arrays(arrays, names=['first', 'second'])
+
+   df = pd.DataFrame({'A': [1, 1, 1, 1, 2, 2, 3, 3],
+                      'B': np.arange(8)},
+                     index=index)
+   df
+
+   df.groupby(['second', 'A']).sum()
+
+
+.. _whatsnew_0200.enhancements.compressed_urls:
+
+Better support for compressed URLs in ``read_csv``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The compression code was refactored (:issue:`12688`). As a result, reading
+dataframes from URLs in :func:`read_csv` or :func:`read_table` now supports
+additional compression methods: ``xz``, ``bz2``, and ``zip`` (:issue:`14570`).
+Previously, only ``gzip`` compression was supported. By default, compression of
+URLs and paths are now inferred using their file extensions. Additionally,
+support for bz2 compression in the python 2 C-engine improved (:issue:`14874`).
+
+.. ipython:: python
+
+   url = ('https://github.com/{repo}/raw/{branch}/{path}'
+          .format(repo='pandas-dev/pandas',
+                  branch='main',
+                  path='pandas/tests/io/parser/data/salaries.csv.bz2'))
+   # default, infer compression
+   df = pd.read_csv(url, sep='\t', compression='infer')
+   # explicitly specify compression
+   df = pd.read_csv(url, sep='\t', compression='bz2')
+   df.head(2)
+
+.. _whatsnew_0200.enhancements.pickle_compression:
+
+Pickle file IO now supports compression
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`read_pickle`, :meth:`DataFrame.to_pickle` and :meth:`Series.to_pickle`
+can now read from and write to compressed pickle files. Compression methods
+can be an explicit parameter or be inferred from the file extension.
+See :ref:`the docs here. <io.pickle.compression>`
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': np.random.randn(1000),
+                      'B': 'foo',
+                      'C': pd.date_range('20130101', periods=1000, freq='s')})
+
+Using an explicit compression type
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.compress", compression="gzip")
+   rt = pd.read_pickle("data.pkl.compress", compression="gzip")
+   rt.head()
+
+The default is to infer the compression type from the extension (``compression='infer'``):
+
+.. ipython:: python
+
+   df.to_pickle("data.pkl.gz")
+   rt = pd.read_pickle("data.pkl.gz")
+   rt.head()
+   df["A"].to_pickle("s1.pkl.bz2")
+   rt = pd.read_pickle("s1.pkl.bz2")
+   rt.head()
+
+.. ipython:: python
+   :suppress:
+
+   import os
+   os.remove("data.pkl.compress")
+   os.remove("data.pkl.gz")
+   os.remove("s1.pkl.bz2")
+
+.. _whatsnew_0200.enhancements.uint64_support:
+
+UInt64 support improved
+^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has significantly improved support for operations involving unsigned,
+or purely non-negative, integers. Previously, handling these integers would
+result in improper rounding or data-type casting, leading to incorrect results.
+Notably, a new numerical index, ``UInt64Index``, has been created (:issue:`14937`)
+
+.. code-block:: ipython
+
+   In [1]: idx = pd.UInt64Index([1, 2, 3])
+   In [2]: df = pd.DataFrame({'A': ['a', 'b', 'c']}, index=idx)
+   In [3]: df.index
+   Out[3]: UInt64Index([1, 2, 3], dtype='uint64')
+
+- Bug in converting object elements of array-like objects to unsigned 64-bit integers (:issue:`4471`, :issue:`14982`)
+- Bug in ``Series.unique()`` in which unsigned 64-bit integers were causing overflow (:issue:`14721`)
+- Bug in ``DataFrame`` construction in which unsigned 64-bit integer elements were being converted to objects (:issue:`14881`)
+- Bug in ``pd.read_csv()`` in which unsigned 64-bit integer elements were being improperly converted to the wrong data types (:issue:`14983`)
+- Bug in ``pd.unique()`` in which unsigned 64-bit integers were causing overflow (:issue:`14915`)
+- Bug in ``pd.value_counts()`` in which unsigned 64-bit integers were being erroneously truncated in the output (:issue:`14934`)
+
+.. _whatsnew_0200.enhancements.groupy_categorical:
+
+GroupBy on categoricals
+^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions, ``.groupby(..., sort=False)`` would fail with a ``ValueError`` when grouping on a categorical series with some categories not appearing in the data. (:issue:`13179`)
+
+.. ipython:: python
+
+   chromosomes = np.r_[np.arange(1, 23).astype(str), ['X', 'Y']]
+   df = pd.DataFrame({
+       'A': np.random.randint(100),
+       'B': np.random.randint(100),
+       'C': np.random.randint(100),
+       'chromosomes': pd.Categorical(np.random.choice(chromosomes, 100),
+                                     categories=chromosomes,
+                                     ordered=True)})
+   df
+
+**Previous behavior**:
+
+.. code-block:: ipython
+
+   In [3]: df[df.chromosomes != '1'].groupby('chromosomes', observed=False, sort=False).sum()
+   ---------------------------------------------------------------------------
+   ValueError: items in new_categories are not the same as in old categories
+
+**New behavior**:
+
+.. ipython:: python
+
+   df[df.chromosomes != '1'].groupby('chromosomes', observed=False, sort=False).sum()
+
+.. _whatsnew_0200.enhancements.table_schema:
+
+Table schema output
+^^^^^^^^^^^^^^^^^^^
+
+The new orient ``'table'`` for :meth:`DataFrame.to_json`
+will generate a `Table Schema`_ compatible string representation of
+the data.
+
+.. code-block:: ipython
+
+   In [38]: df = pd.DataFrame(
+      ....: {'A': [1, 2, 3],
+      ....:  'B': ['a', 'b', 'c'],
+      ....:  'C': pd.date_range('2016-01-01', freq='d', periods=3)},
+      ....: index=pd.Index(range(3), name='idx'))
+   In [39]: df
+   Out[39]:
+        A  B          C
+   idx
+   0    1  a 2016-01-01
+   1    2  b 2016-01-02
+   2    3  c 2016-01-03
+
+   [3 rows x 3 columns]
+
+   In [40]: df.to_json(orient='table')
+   Out[40]:
+   '{"schema":{"fields":[{"name":"idx","type":"integer"},{"name":"A","type":"integer"},{"name":"B","type":"string"},{"name":"C","type":"datetime"}],"primaryKey":["idx"],"pandas_version":"1.4.0"},"data":[{"idx":0,"A":1,"B":"a","C":"2016-01-01T00:00:00.000"},{"idx":1,"A":2,"B":"b","C":"2016-01-02T00:00:00.000"},{"idx":2,"A":3,"B":"c","C":"2016-01-03T00:00:00.000"}]}'
+
+
+See :ref:`IO: Table Schema for more information <io.table_schema>`.
+
+Additionally, the repr for ``DataFrame`` and ``Series`` can now publish
+this JSON Table schema representation of the Series or DataFrame if you are
+using IPython (or another frontend like `nteract`_ using the Jupyter messaging
+protocol).
+This gives frontends like the Jupyter notebook and `nteract`_
+more flexibility in how they display pandas objects, since they have
+more information about the data.
+You must enable this by setting the ``display.html.table_schema`` option to ``True``.
+
+.. _Table Schema: http://specs.frictionlessdata.io/json-table-schema/
+.. _nteract: https://nteract.io/
+
+.. _whatsnew_0200.enhancements.scipy_sparse:
+
+SciPy sparse matrix from/to SparseDataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas now supports creating sparse dataframes directly from ``scipy.sparse.spmatrix`` instances.
+See the :ref:`documentation <sparse.scipysparse>` for more information. (:issue:`4343`)
+
+All sparse formats are supported, but matrices that are not in :mod:`COOrdinate <scipy.sparse>` format will be converted, copying data as needed.
+
+.. code-block:: python
+
+   from scipy.sparse import csr_matrix
+   arr = np.random.random(size=(1000, 5))
+   arr[arr < .9] = 0
+   sp_arr = csr_matrix(arr)
+   sp_arr
+   sdf = pd.SparseDataFrame(sp_arr)
+   sdf
+
+To convert a ``SparseDataFrame`` back to sparse SciPy matrix in COO format, you can use:
+
+.. code-block:: python
+
+   sdf.to_coo()
+
+.. _whatsnew_0200.enhancements.style_excel:
+
+Excel output for styled DataFrames
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Experimental support has been added to export ``DataFrame.style`` formats to Excel using the ``openpyxl`` engine. (:issue:`15530`)
+
+For example, after running the following, ``styled.xlsx`` renders as below:
+
+.. ipython:: python
+
+   np.random.seed(24)
+   df = pd.DataFrame({'A': np.linspace(1, 10, 10)})
+   df = pd.concat([df, pd.DataFrame(np.random.RandomState(24).randn(10, 4),
+                                    columns=list('BCDE'))],
+                  axis=1)
+   df.iloc[0, 2] = np.nan
+   df
+   styled = (df.style
+             .map(lambda val: 'color:red;' if val < 0 else 'color:black;')
+             .highlight_max())
+   styled.to_excel('styled.xlsx', engine='openpyxl')
+
+.. image:: ../_static/style-excel.png
+
+.. ipython:: python
+   :suppress:
+
+   import os
+   os.remove('styled.xlsx')
+
+See the :ref:`Style documentation </user_guide/style.ipynb#Export-to-Excel>` for more detail.
+
+.. _whatsnew_0200.enhancements.intervalindex:
+
+IntervalIndex
+^^^^^^^^^^^^^
+
+pandas has gained an ``IntervalIndex`` with its own dtype, ``interval`` as well as the ``Interval`` scalar type. These allow first-class support for interval
+notation, specifically as a return type for the categories in :func:`cut` and :func:`qcut`. The ``IntervalIndex`` allows some unique indexing, see the
+:ref:`docs <advanced.intervalindex>`. (:issue:`7640`, :issue:`8625`)
+
+.. warning::
+
+   These indexing behaviors of the IntervalIndex are provisional and may change in a future version of pandas. Feedback on usage is welcome.
+
+
+Previous behavior:
+
+The returned categories were strings, representing Intervals
+
+.. code-block:: ipython
+
+   In [1]: c = pd.cut(range(4), bins=2)
+
+   In [2]: c
+   Out[2]:
+   [(-0.003, 1.5], (-0.003, 1.5], (1.5, 3], (1.5, 3]]
+   Categories (2, object): [(-0.003, 1.5] < (1.5, 3]]
+
+   In [3]: c.categories
+   Out[3]: Index(['(-0.003, 1.5]', '(1.5, 3]'], dtype='object')
+
+New behavior:
+
+.. ipython:: python
+
+   c = pd.cut(range(4), bins=2)
+   c
+   c.categories
+
+Furthermore, this allows one to bin *other* data with these same bins, with ``NaN`` representing a missing
+value similar to other dtypes.
+
+.. ipython:: python
+
+   pd.cut([0, 3, 5, 1], bins=c.categories)
+
+An ``IntervalIndex`` can also be used in ``Series`` and ``DataFrame`` as the index.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': range(4),
+                      'B': pd.cut([0, 3, 1, 1], bins=c.categories)
+                      }).set_index('B')
+   df
+
+Selecting via a specific interval:
+
+.. ipython:: python
+
+   df.loc[pd.Interval(1.5, 3.0)]
+
+Selecting via a scalar value that is contained *in* the intervals.
+
+.. ipython:: python
+
+   df.loc[0]
+
+.. _whatsnew_0200.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- ``DataFrame.rolling()`` now accepts the parameter ``closed='right'|'left'|'both'|'neither'`` to choose the rolling window-endpoint closedness. See the :ref:`documentation <window.endpoints>` (:issue:`13965`)
+- Integration with the ``feather-format``, including a new top-level ``pd.read_feather()`` and ``DataFrame.to_feather()`` method, see :ref:`here <io.feather>`.
+- ``Series.str.replace()`` now accepts a callable, as replacement, which is passed to ``re.sub`` (:issue:`15055`)
+- ``Series.str.replace()`` now accepts a compiled regular expression as a pattern (:issue:`15446`)
+- ``Series.sort_index`` accepts parameters ``kind`` and ``na_position`` (:issue:`13589`, :issue:`14444`)
+- ``DataFrame`` and ``DataFrame.groupby()``  have gained a ``nunique()`` method to count the distinct values over an axis (:issue:`14336`, :issue:`15197`).
+- ``DataFrame`` has gained a ``melt()`` method, equivalent to ``pd.melt()``, for unpivoting from a wide to long format (:issue:`12640`).
+- ``pd.read_excel()`` now preserves sheet order when using ``sheetname=None`` (:issue:`9930`)
+- Multiple offset aliases with decimal points are now supported (e.g. ``0.5min`` is parsed as ``30s``) (:issue:`8419`)
+- ``.isnull()`` and ``.notnull()`` have been added to ``Index`` object to make them more consistent with the ``Series`` API (:issue:`15300`)
+- New ``UnsortedIndexError`` (subclass of ``KeyError``) raised when indexing/slicing into an
+  unsorted MultiIndex (:issue:`11897`). This allows differentiation between errors due to lack
+  of sorting or an incorrect key. See :ref:`here <advanced.unsorted>`
+- ``MultiIndex`` has gained a ``.to_frame()`` method to convert to a ``DataFrame`` (:issue:`12397`)
+- ``pd.cut`` and ``pd.qcut`` now support datetime64 and timedelta64 dtypes (:issue:`14714`, :issue:`14798`)
+- ``pd.qcut`` has gained the ``duplicates='raise'|'drop'`` option to control whether to raise on duplicated edges (:issue:`7751`)
+- ``Series`` provides a ``to_excel`` method to output Excel files (:issue:`8825`)
+- The ``usecols`` argument in ``pd.read_csv()`` now accepts a callable function as a value  (:issue:`14154`)
+- The ``skiprows`` argument in ``pd.read_csv()`` now accepts a callable function as a value  (:issue:`10882`)
+- The ``nrows`` and ``chunksize`` arguments in ``pd.read_csv()`` are supported if both are passed (:issue:`6774`, :issue:`15755`)
+- ``DataFrame.plot`` now prints a title above each subplot if ``suplots=True`` and ``title`` is a list of strings (:issue:`14753`)
+- ``DataFrame.plot`` can pass the matplotlib 2.0 default color cycle as a single string as color parameter, see `here <http://matplotlib.org/2.0.0/users/colors.html#cn-color-selection>`__. (:issue:`15516`)
+- ``Series.interpolate()`` now supports timedelta as an index type with ``method='time'`` (:issue:`6424`)
+- Addition of a ``level`` keyword to ``DataFrame/Series.rename`` to rename
+  labels in the specified level of a MultiIndex (:issue:`4160`).
+- ``DataFrame.reset_index()`` will now interpret a tuple ``index.name`` as a key spanning across levels of ``columns``, if this is a ``MultiIndex`` (:issue:`16164`)
+- ``Timedelta.isoformat`` method added for formatting Timedeltas as an `ISO 8601 duration`_. See the :ref:`Timedelta docs <timedeltas.isoformat>` (:issue:`15136`)
+- ``.select_dtypes()`` now allows the string ``datetimetz`` to generically select datetimes with tz (:issue:`14910`)
+- The ``.to_latex()`` method will now accept ``multicolumn`` and ``multirow`` arguments to use the accompanying LaTeX enhancements
+- ``pd.merge_asof()`` gained the option ``direction='backward'|'forward'|'nearest'`` (:issue:`14887`)
+- ``Series/DataFrame.asfreq()`` have gained a ``fill_value`` parameter, to fill missing values (:issue:`3715`).
+- ``Series/DataFrame.resample.asfreq`` have gained a ``fill_value`` parameter, to fill missing values during resampling (:issue:`3715`).
+- :func:`pandas.util.hash_pandas_object` has gained the ability to hash a ``MultiIndex`` (:issue:`15224`)
+- ``Series/DataFrame.squeeze()`` have gained the ``axis`` parameter. (:issue:`15339`)
+- ``DataFrame.to_excel()`` has a new ``freeze_panes`` parameter to turn on Freeze Panes when exporting to Excel (:issue:`15160`)
+- ``pd.read_html()`` will parse multiple header rows, creating a MultiIndex header. (:issue:`13434`).
+- HTML table output skips ``colspan`` or ``rowspan`` attribute if equal to 1. (:issue:`15403`)
+- :class:`pandas.io.formats.style.Styler` template now has blocks for easier extension, see the :ref:`example notebook </user_guide/style.ipynb#Subclassing>` (:issue:`15649`)
+- :meth:`Styler.render() <pandas.io.formats.style.Styler.render>` now accepts ``**kwargs`` to allow user-defined variables in the template (:issue:`15649`)
+- Compatibility with Jupyter notebook 5.0; MultiIndex column labels are left-aligned and MultiIndex row-labels are top-aligned (:issue:`15379`)
+- ``TimedeltaIndex`` now has a custom date-tick formatter specifically designed for nanosecond level precision (:issue:`8711`)
+- ``pd.api.types.union_categoricals`` gained the ``ignore_ordered`` argument to allow ignoring the ordered attribute of unioned categoricals (:issue:`13410`). See the :ref:`categorical union docs <categorical.union>` for more information.
+- ``DataFrame.to_latex()`` and ``DataFrame.to_string()`` now allow optional header aliases. (:issue:`15536`)
+- Re-enable the ``parse_dates`` keyword of ``pd.read_excel()`` to parse string columns as dates (:issue:`14326`)
+- Added ``.empty`` property to subclasses of ``Index``. (:issue:`15270`)
+- Enabled floor division for ``Timedelta`` and ``TimedeltaIndex`` (:issue:`15828`)
+- ``pandas.io.json.json_normalize()`` gained the option ``errors='ignore'|'raise'``; the default is ``errors='raise'`` which is backward compatible. (:issue:`14583`)
+- ``pandas.io.json.json_normalize()`` with an empty ``list`` will return an empty ``DataFrame`` (:issue:`15534`)
+- ``pandas.io.json.json_normalize()`` has gained a ``sep`` option that accepts ``str`` to separate joined fields; the default is ".", which is backward compatible. (:issue:`14883`)
+- :meth:`MultiIndex.remove_unused_levels` has been added to facilitate :ref:`removing unused levels <advanced.shown_levels>`. (:issue:`15694`)
+- ``pd.read_csv()`` will now raise a ``ParserError`` error whenever any parsing error occurs (:issue:`15913`, :issue:`15925`)
+- ``pd.read_csv()`` now supports the ``error_bad_lines`` and ``warn_bad_lines`` arguments for the Python parser (:issue:`15925`)
+- The ``display.show_dimensions`` option can now also be used to specify
+  whether the length of a ``Series`` should be shown in its repr (:issue:`7117`).
+- ``parallel_coordinates()`` has gained a ``sort_labels`` keyword argument that sorts class labels and the colors assigned to them (:issue:`15908`)
+- Options added to allow one to turn on/off using ``bottleneck`` and ``numexpr``, see :ref:`here <basics.accelerate>` (:issue:`16157`)
+- ``DataFrame.style.bar()`` now accepts two more options to further customize the bar chart. Bar alignment is set with ``align='left'|'mid'|'zero'``, the default is "left", which is backward compatible; You can now pass a list of ``color=[color_negative, color_positive]``. (:issue:`14757`)
+
+.. _ISO 8601 duration: https://en.wikipedia.org/wiki/ISO_8601#Durations
+
+
+.. _whatsnew_0200.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew.api_breaking.io_compat:
+
+Possible incompatibility for HDF5 formats created with pandas < 0.13.0
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``pd.TimeSeries`` was deprecated officially in 0.17.0, though has already been an alias since 0.13.0. It has
+been dropped in favor of ``pd.Series``. (:issue:`15098`).
+
+This *may* cause HDF5 files that were created in prior versions to become unreadable if ``pd.TimeSeries``
+was used. This is most likely to be for pandas < 0.13.0. If you find yourself in this situation.
+You can use a recent prior version of pandas to read in your HDF5 files,
+then write them out again after applying the procedure below.
+
+.. code-block:: ipython
+
+   In [2]: s = pd.TimeSeries([1, 2, 3], index=pd.date_range('20130101', periods=3))
+
+   In [3]: s
+   Out[3]:
+   2013-01-01    1
+   2013-01-02    2
+   2013-01-03    3
+   Freq: D, dtype: int64
+
+   In [4]: type(s)
+   Out[4]: pandas.core.series.TimeSeries
+
+   In [5]: s = pd.Series(s)
+
+   In [6]: s
+   Out[6]:
+   2013-01-01    1
+   2013-01-02    2
+   2013-01-03    3
+   Freq: D, dtype: int64
+
+   In [7]: type(s)
+   Out[7]: pandas.core.series.Series
+
+
+.. _whatsnew_0200.api_breaking.index_map:
+
+Map on Index types now return other Index types
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``map`` on an ``Index`` now returns an ``Index``, not a numpy array (:issue:`12766`)
+
+.. ipython:: python
+
+   idx = pd.Index([1, 2])
+   idx
+   mi = pd.MultiIndex.from_tuples([(1, 2), (2, 4)])
+   mi
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [5]: idx.map(lambda x: x * 2)
+   Out[5]: array([2, 4])
+
+   In [6]: idx.map(lambda x: (x, x * 2))
+   Out[6]: array([(1, 2), (2, 4)], dtype=object)
+
+   In [7]: mi.map(lambda x: x)
+   Out[7]: array([(1, 2), (2, 4)], dtype=object)
+
+   In [8]: mi.map(lambda x: x[0])
+   Out[8]: array([1, 2])
+
+New behavior:
+
+.. ipython:: python
+
+   idx.map(lambda x: x * 2)
+   idx.map(lambda x: (x, x * 2))
+
+   mi.map(lambda x: x)
+
+   mi.map(lambda x: x[0])
+
+
+``map`` on a ``Series`` with ``datetime64`` values may return ``int64`` dtypes rather than ``int32``
+
+.. code-block:: ipython
+
+   In [64]: s = pd.Series(pd.date_range('2011-01-02T00:00', '2011-01-02T02:00', freq='H')
+      ....:               .tz_localize('Asia/Tokyo'))
+      ....:
+
+   In [65]: s
+   Out[65]:
+   0   2011-01-02 00:00:00+09:00
+   1   2011-01-02 01:00:00+09:00
+   2   2011-01-02 02:00:00+09:00
+   Length: 3, dtype: datetime64[ns, Asia/Tokyo]
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [9]: s.map(lambda x: x.hour)
+   Out[9]:
+   0    0
+   1    1
+   2    2
+   dtype: int32
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [66]: s.map(lambda x: x.hour)
+   Out[66]:
+   0    0
+   1    1
+   2    2
+   Length: 3, dtype: int64
+
+
+.. _whatsnew_0200.api_breaking.index_dt_field:
+
+Accessing datetime fields of Index now return Index
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The datetime-related attributes (see :ref:`here <timeseries.components>`
+for an overview) of ``DatetimeIndex``, ``PeriodIndex`` and ``TimedeltaIndex`` previously
+returned numpy arrays. They will now return a new ``Index`` object, except
+in the case of a boolean field, where the result will still be a boolean ndarray. (:issue:`15022`)
+
+Previous behaviour:
+
+.. code-block:: ipython
+
+   In [1]: idx = pd.date_range("2015-01-01", periods=5, freq='10H')
+
+   In [2]: idx.hour
+   Out[2]: array([ 0, 10, 20,  6, 16], dtype=int32)
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [67]: idx = pd.date_range("2015-01-01", periods=5, freq='10H')
+
+   In [68]: idx.hour
+   Out[68]: Index([0, 10, 20, 6, 16], dtype='int32')
+
+This has the advantage that specific ``Index`` methods are still available on the
+result. On the other hand, this might have backward incompatibilities: e.g.
+compared to numpy arrays, ``Index`` objects are not mutable. To get the original
+ndarray, you can always convert explicitly using ``np.asarray(idx.hour)``.
+
+.. _whatsnew_0200.api_breaking.unique:
+
+pd.unique will now be consistent with extension types
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In prior versions, using :meth:`Series.unique` and :func:`pandas.unique` on ``Categorical`` and tz-aware
+data-types would yield different return types. These are now made consistent. (:issue:`15903`)
+
+- Datetime tz-aware
+
+  Previous behaviour:
+
+  .. code-block:: ipython
+
+     # Series
+     In [5]: pd.Series([pd.Timestamp('20160101', tz='US/Eastern'),
+        ...:            pd.Timestamp('20160101', tz='US/Eastern')]).unique()
+     Out[5]: array([Timestamp('2016-01-01 00:00:00-0500', tz='US/Eastern')], dtype=object)
+
+     In [6]: pd.unique(pd.Series([pd.Timestamp('20160101', tz='US/Eastern'),
+        ...:                      pd.Timestamp('20160101', tz='US/Eastern')]))
+     Out[6]: array(['2016-01-01T05:00:00.000000000'], dtype='datetime64[ns]')
+
+     # Index
+     In [7]: pd.Index([pd.Timestamp('20160101', tz='US/Eastern'),
+        ...:           pd.Timestamp('20160101', tz='US/Eastern')]).unique()
+     Out[7]: DatetimeIndex(['2016-01-01 00:00:00-05:00'], dtype='datetime64[ns, US/Eastern]', freq=None)
+
+     In [8]: pd.unique([pd.Timestamp('20160101', tz='US/Eastern'),
+        ...:            pd.Timestamp('20160101', tz='US/Eastern')])
+     Out[8]: array(['2016-01-01T05:00:00.000000000'], dtype='datetime64[ns]')
+
+  New behavior:
+
+  .. ipython:: python
+
+     # Series, returns an array of Timestamp tz-aware
+     pd.Series([pd.Timestamp(r'20160101', tz=r'US/Eastern'),
+                pd.Timestamp(r'20160101', tz=r'US/Eastern')]).unique()
+     pd.unique(pd.Series([pd.Timestamp('20160101', tz='US/Eastern'),
+               pd.Timestamp('20160101', tz='US/Eastern')]))
+
+     # Index, returns a DatetimeIndex
+     pd.Index([pd.Timestamp('20160101', tz='US/Eastern'),
+               pd.Timestamp('20160101', tz='US/Eastern')]).unique()
+     pd.unique(pd.Index([pd.Timestamp('20160101', tz='US/Eastern'),
+                         pd.Timestamp('20160101', tz='US/Eastern')]))
+
+- Categoricals
+
+  Previous behaviour:
+
+  .. code-block:: ipython
+
+     In [1]: pd.Series(list('baabc'), dtype='category').unique()
+     Out[1]:
+     [b, a, c]
+     Categories (3, object): [b, a, c]
+
+     In [2]: pd.unique(pd.Series(list('baabc'), dtype='category'))
+     Out[2]: array(['b', 'a', 'c'], dtype=object)
+
+  New behavior:
+
+  .. ipython:: python
+
+     # returns a Categorical
+     pd.Series(list('baabc'), dtype='category').unique()
+     pd.unique(pd.Series(list('baabc'), dtype='category'))
+
+.. _whatsnew_0200.api_breaking.s3:
+
+S3 file handling
+^^^^^^^^^^^^^^^^
+
+pandas now uses `s3fs <http://s3fs.readthedocs.io/>`_ for handling S3 connections. This shouldn't break
+any code. However, since ``s3fs`` is not a required dependency, you will need to install it separately, like ``boto``
+in prior versions of pandas. (:issue:`11915`).
+
+.. _whatsnew_0200.api_breaking.partial_string_indexing:
+
+Partial string indexing changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:ref:`DatetimeIndex Partial String Indexing <timeseries.partialindexing>` now works as an exact match, provided that string resolution coincides with index resolution, including a case when both are seconds (:issue:`14826`). See :ref:`Slice vs. Exact Match <timeseries.slice_vs_exact_match>` for details.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': [1, 2, 3]}, pd.DatetimeIndex(['2011-12-31 23:59:59',
+                                                         '2012-01-01 00:00:00',
+                                                         '2012-01-01 00:00:01']))
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [4]: df['2011-12-31 23:59:59']
+   Out[4]:
+                          a
+   2011-12-31 23:59:59  1
+
+   In [5]: df['a']['2011-12-31 23:59:59']
+   Out[5]:
+   2011-12-31 23:59:59    1
+   Name: a, dtype: int64
+
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [4]: df['2011-12-31 23:59:59']
+   KeyError: '2011-12-31 23:59:59'
+
+   In [5]: df['a']['2011-12-31 23:59:59']
+   Out[5]: 1
+
+.. _whatsnew_0200.api_breaking.concat_dtypes:
+
+Concat of different float dtypes will not automatically upcast
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, ``concat`` of multiple objects with different ``float`` dtypes would automatically upcast results to a dtype of ``float64``.
+Now the smallest acceptable dtype will be used (:issue:`13247`)
+
+.. ipython:: python
+
+   df1 = pd.DataFrame(np.array([1.0], dtype=np.float32, ndmin=2))
+   df1.dtypes
+
+   df2 = pd.DataFrame(np.array([np.nan], dtype=np.float32, ndmin=2))
+   df2.dtypes
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [7]: pd.concat([df1, df2]).dtypes
+   Out[7]:
+   0    float64
+   dtype: object
+
+New behavior:
+
+.. ipython:: python
+
+   pd.concat([df1, df2]).dtypes
+
+.. _whatsnew_0200.api_breaking.gbq:
+
+pandas Google BigQuery support has moved
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has split off Google BigQuery support into a separate package ``pandas-gbq``. You can ``conda install pandas-gbq -c conda-forge`` or
+``pip install pandas-gbq`` to get it. The functionality of :func:`read_gbq` and :meth:`DataFrame.to_gbq` remain the same with the
+currently released version of ``pandas-gbq=0.1.4``. Documentation is now hosted `here <https://pandas-gbq.readthedocs.io/>`__  (:issue:`15347`)
+
+.. _whatsnew_0200.api_breaking.memory_usage:
+
+Memory usage for Index is more accurate
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions, showing ``.memory_usage()`` on a pandas structure that has an index, would only include actual index values and not include structures that facilitated fast indexing. This will generally be different for ``Index`` and ``MultiIndex`` and less-so for other index types. (:issue:`15237`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [8]: index = pd.Index(['foo', 'bar', 'baz'])
+
+   In [9]: index.memory_usage(deep=True)
+   Out[9]: 180
+
+   In [10]: index.get_loc('foo')
+   Out[10]: 0
+
+   In [11]: index.memory_usage(deep=True)
+   Out[11]: 180
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [8]: index = pd.Index(['foo', 'bar', 'baz'])
+
+   In [9]: index.memory_usage(deep=True)
+   Out[9]: 180
+
+   In [10]: index.get_loc('foo')
+   Out[10]: 0
+
+   In [11]: index.memory_usage(deep=True)
+   Out[11]: 260
+
+.. _whatsnew_0200.api_breaking.sort_index:
+
+DataFrame.sort_index changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In certain cases, calling ``.sort_index()`` on a MultiIndexed DataFrame would return the *same* DataFrame without seeming to sort.
+This would happen with a ``lexsorted``, but non-monotonic levels. (:issue:`15622`, :issue:`15687`, :issue:`14015`, :issue:`13431`, :issue:`15797`)
+
+This is *unchanged* from prior versions, but shown for illustration purposes:
+
+.. code-block:: python
+
+   In [81]: df = pd.DataFrame(np.arange(6), columns=['value'],
+      ....:                   index=pd.MultiIndex.from_product([list('BA'), range(3)]))
+      ....:
+   In [82]: df
+
+   Out[82]:
+        value
+   B 0      0
+     1      1
+     2      2
+   A 0      3
+     1      4
+     2      5
+
+   [6 rows x 1 columns]
+
+.. code-block:: python
+
+   In [87]: df.index.is_lexsorted()
+   Out[87]: False
+
+   In [88]: df.index.is_monotonic
+   Out[88]: False
+
+Sorting works as expected
+
+.. ipython:: python
+
+   df.sort_index()
+
+.. code-block:: python
+
+   In [90]: df.sort_index().index.is_lexsorted()
+   Out[90]: True
+
+   In [91]: df.sort_index().index.is_monotonic
+   Out[91]: True
+
+However, this example, which has a non-monotonic 2nd level,
+doesn't behave as desired.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'value': [1, 2, 3, 4]},
+                     index=pd.MultiIndex([['a', 'b'], ['bb', 'aa']],
+                                         [[0, 0, 1, 1], [0, 1, 0, 1]]))
+   df
+
+Previous behavior:
+
+.. code-block:: python
+
+   In [11]: df.sort_index()
+   Out[11]:
+         value
+   a bb      1
+     aa      2
+   b bb      3
+     aa      4
+
+   In [14]: df.sort_index().index.is_lexsorted()
+   Out[14]: True
+
+   In [15]: df.sort_index().index.is_monotonic
+   Out[15]: False
+
+New behavior:
+
+.. code-block:: python
+
+   In [94]: df.sort_index()
+   Out[94]:
+         value
+   a aa      2
+     bb      1
+   b aa      4
+     bb      3
+
+   [4 rows x 1 columns]
+
+   In [95]: df.sort_index().index.is_lexsorted()
+   Out[95]: True
+
+   In [96]: df.sort_index().index.is_monotonic
+   Out[96]: True
+
+
+.. _whatsnew_0200.api_breaking.groupby_describe:
+
+GroupBy describe formatting
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The output formatting of ``groupby.describe()`` now labels the ``describe()`` metrics in the columns instead of the index.
+This format is consistent with ``groupby.agg()`` when applying multiple functions at once. (:issue:`4792`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame({'A': [1, 1, 2, 2], 'B': [1, 2, 3, 4]})
+
+   In [2]: df.groupby('A').describe()
+   Out[2]:
+                   B
+   A
+   1 count  2.000000
+     mean   1.500000
+     std    0.707107
+     min    1.000000
+     25%    1.250000
+     50%    1.500000
+     75%    1.750000
+     max    2.000000
+   2 count  2.000000
+     mean   3.500000
+     std    0.707107
+     min    3.000000
+     25%    3.250000
+     50%    3.500000
+     75%    3.750000
+     max    4.000000
+
+   In [3]: df.groupby('A').agg(["mean", "std", "min", "max"])
+   Out[3]:
+        B
+     mean       std amin amax
+   A
+   1  1.5  0.707107    1    2
+   2  3.5  0.707107    3    4
+
+New behavior:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 1, 2, 2], 'B': [1, 2, 3, 4]})
+
+   df.groupby('A').describe()
+
+   df.groupby('A').agg(["mean", "std", "min", "max"])
+
+.. _whatsnew_0200.api_breaking.rolling_pairwise:
+
+Window binary corr/cov operations return a MultiIndex DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A binary window operation, like ``.corr()`` or ``.cov()``, when operating on a ``.rolling(..)``, ``.expanding(..)``, or ``.ewm(..)`` object,
+will now return a 2-level ``MultiIndexed DataFrame`` rather than a ``Panel``, as ``Panel`` is now deprecated,
+see :ref:`here <whatsnew_0200.api_breaking.deprecate_panel>`. These are equivalent in function,
+but a MultiIndexed ``DataFrame`` enjoys more support in pandas.
+See the section on :ref:`Windowed Binary Operations <window.cov_corr>` for more information. (:issue:`15677`)
+
+.. ipython:: python
+
+   np.random.seed(1234)
+   df = pd.DataFrame(np.random.rand(100, 2),
+                     columns=pd.Index(['A', 'B'], name='bar'),
+                     index=pd.date_range('20160101',
+                                         periods=100, freq='D', name='foo'))
+   df.tail()
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: df.rolling(12).corr()
+   Out[2]:
+   <class 'pandas.core.panel.Panel'>
+   Dimensions: 100 (items) x 2 (major_axis) x 2 (minor_axis)
+   Items axis: 2016-01-01 00:00:00 to 2016-04-09 00:00:00
+   Major_axis axis: A to B
+   Minor_axis axis: A to B
+
+New behavior:
+
+.. ipython:: python
+
+   res = df.rolling(12).corr()
+   res.tail()
+
+Retrieving a correlation matrix for a cross-section
+
+.. ipython:: python
+
+   df.rolling(12).corr().loc['2016-04-07']
+
+.. _whatsnew_0200.api_breaking.hdfstore_where:
+
+HDFStore where string comparison
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions most types could be compared to string column in a ``HDFStore``
+usually resulting in an invalid comparison, returning an empty result frame. These comparisons will now raise a
+``TypeError`` (:issue:`15492`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'unparsed_date': ['2014-01-01', '2014-01-01']})
+   df.to_hdf('store.h5', key='key', format='table', data_columns=True)
+   df.dtypes
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [4]: pd.read_hdf('store.h5', 'key', where='unparsed_date > ts')
+   File "<string>", line 1
+     (unparsed_date > 1970-01-01 00:00:01.388552400)
+                           ^
+   SyntaxError: invalid token
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [18]: ts = pd.Timestamp('2014-01-01')
+
+   In [19]: pd.read_hdf('store.h5', 'key', where='unparsed_date > ts')
+   TypeError: Cannot compare 2014-01-01 00:00:00 of
+   type <class 'pandas.tslib.Timestamp'> to string column
+
+.. ipython:: python
+   :suppress:
+
+   import os
+   os.remove('store.h5')
+
+.. _whatsnew_0200.api_breaking.index_order:
+
+Index.intersection and inner join now preserve the order of the left Index
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`Index.intersection` now preserves the order of the calling ``Index`` (left)
+instead of the other ``Index`` (right) (:issue:`15582`). This affects inner
+joins, :meth:`DataFrame.join` and :func:`merge`, and the ``.align`` method.
+
+- ``Index.intersection``
+
+  .. ipython:: python
+
+     left = pd.Index([2, 1, 0])
+     left
+     right = pd.Index([1, 2, 3])
+     right
+
+  Previous behavior:
+
+  .. code-block:: ipython
+
+     In [4]: left.intersection(right)
+     Out[4]: Int64Index([1, 2], dtype='int64')
+
+  New behavior:
+
+  .. ipython:: python
+
+     left.intersection(right)
+
+- ``DataFrame.join`` and ``pd.merge``
+
+  .. ipython:: python
+
+     left = pd.DataFrame({'a': [20, 10, 0]}, index=[2, 1, 0])
+     left
+     right = pd.DataFrame({'b': [100, 200, 300]}, index=[1, 2, 3])
+     right
+
+  Previous behavior:
+
+  .. code-block:: ipython
+
+     In [4]: left.join(right, how='inner')
+     Out[4]:
+        a    b
+     1  10  100
+     2  20  200
+
+  New behavior:
+
+  .. ipython:: python
+
+     left.join(right, how='inner')
+
+.. _whatsnew_0200.api_breaking.pivot_table:
+
+Pivot table always returns a DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The documentation for :meth:`pivot_table` states that a ``DataFrame`` is *always* returned. Here a bug
+is fixed that allowed this to return a ``Series`` under certain circumstance. (:issue:`4386`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'col1': [3, 4, 5],
+                      'col2': ['C', 'D', 'E'],
+                      'col3': [1, 3, 9]})
+   df
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: df.pivot_table('col1', index=['col3', 'col2'], aggfunc="sum")
+   Out[2]:
+   col3  col2
+   1     C       3
+   3     D       4
+   9     E       5
+   Name: col1, dtype: int64
+
+New behavior:
+
+.. ipython:: python
+
+   df.pivot_table('col1', index=['col3', 'col2'], aggfunc="sum")
+
+.. _whatsnew_0200.api:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- ``numexpr`` version is now required to be >= 2.4.6 and it will not be used at all if this requisite is not fulfilled (:issue:`15213`).
+- ``CParserError`` has been renamed to ``ParserError`` in ``pd.read_csv()`` and will be removed in the future (:issue:`12665`)
+- ``SparseArray.cumsum()`` and ``SparseSeries.cumsum()`` will now always return ``SparseArray`` and ``SparseSeries`` respectively (:issue:`12855`)
+- ``DataFrame.applymap()`` with an empty ``DataFrame`` will return a copy of the empty ``DataFrame`` instead of a ``Series`` (:issue:`8222`)
+- ``Series.map()`` now respects default values of dictionary subclasses with a ``__missing__`` method, such as ``collections.Counter`` (:issue:`15999`)
+- ``.loc`` has compat with ``.ix`` for accepting iterators, and NamedTuples (:issue:`15120`)
+- ``interpolate()`` and ``fillna()`` will raise a ``ValueError`` if the ``limit`` keyword argument is not greater than 0. (:issue:`9217`)
+- ``pd.read_csv()`` will now issue a ``ParserWarning`` whenever there are conflicting values provided by the ``dialect`` parameter and the user (:issue:`14898`)
+- ``pd.read_csv()`` will now raise a ``ValueError`` for the C engine if the quote character is larger than one byte (:issue:`11592`)
+- ``inplace`` arguments now require a boolean value, else a ``ValueError`` is thrown (:issue:`14189`)
+- ``pandas.api.types.is_datetime64_ns_dtype`` will now report ``True`` on a tz-aware dtype, similar to ``pandas.api.types.is_datetime64_any_dtype``
+- ``DataFrame.asof()`` will return a null filled ``Series`` instead the scalar ``NaN`` if a match is not found (:issue:`15118`)
+- Specific support for ``copy.copy()`` and ``copy.deepcopy()`` functions on NDFrame objects (:issue:`15444`)
+- ``Series.sort_values()`` accepts a one element list of bool for consistency with the behavior of ``DataFrame.sort_values()`` (:issue:`15604`)
+- ``.merge()`` and ``.join()`` on ``category`` dtype columns will now preserve the category dtype when possible (:issue:`10409`)
+- ``SparseDataFrame.default_fill_value`` will be 0, previously was ``nan`` in the return from ``pd.get_dummies(..., sparse=True)`` (:issue:`15594`)
+- The default behaviour of ``Series.str.match`` has changed from extracting
+  groups to matching the pattern. The extracting behaviour was deprecated
+  since pandas version 0.13.0 and can be done with the ``Series.str.extract``
+  method (:issue:`5224`). As a consequence, the ``as_indexer`` keyword is
+  ignored (no longer needed to specify the new behaviour) and is deprecated.
+- ``NaT`` will now correctly report ``False`` for datetimelike boolean operations such as ``is_month_start`` (:issue:`15781`)
+- ``NaT`` will now correctly return ``np.nan`` for ``Timedelta`` and ``Period`` accessors such as ``days`` and ``quarter`` (:issue:`15782`)
+- ``NaT`` will now returns ``NaT`` for ``tz_localize`` and ``tz_convert``
+  methods (:issue:`15830`)
+- ``DataFrame`` and ``Panel`` constructors with invalid input will now raise ``ValueError`` rather than ``PandasError``, if called with scalar inputs and not axes (:issue:`15541`)
+- ``DataFrame`` and ``Panel`` constructors with invalid input will now raise ``ValueError`` rather than ``pandas.core.common.PandasError``, if called with scalar inputs and not axes; The exception ``PandasError`` is removed as well. (:issue:`15541`)
+- The exception ``pandas.core.common.AmbiguousIndexError`` is removed as it is not referenced (:issue:`15541`)
+
+
+.. _whatsnew_0200.privacy:
+
+Reorganization of the library: privacy changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0200.privacy.extensions:
+
+Modules privacy has changed
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Some formerly public python/c/c++/cython extension modules have been moved and/or renamed. These are all removed from the public API.
+Furthermore, the ``pandas.core``, ``pandas.compat``, and ``pandas.util`` top-level modules are now considered to be PRIVATE.
+If indicated, a deprecation warning will be issued if you reference these modules. (:issue:`12588`)
+
+.. csv-table::
+    :header: "Previous Location", "New Location", "Deprecated"
+    :widths: 30, 30, 4
+
+    "pandas.lib", "pandas._libs.lib", "X"
+    "pandas.tslib", "pandas._libs.tslib", "X"
+    "pandas.computation", "pandas.core.computation", "X"
+    "pandas.msgpack", "pandas.io.msgpack", ""
+    "pandas.index", "pandas._libs.index", ""
+    "pandas.algos", "pandas._libs.algos", ""
+    "pandas.hashtable", "pandas._libs.hashtable", ""
+    "pandas.indexes", "pandas.core.indexes", ""
+    "pandas.json", "pandas._libs.json / pandas.io.json", "X"
+    "pandas.parser", "pandas._libs.parsers", "X"
+    "pandas.formats", "pandas.io.formats", ""
+    "pandas.sparse", "pandas.core.sparse", ""
+    "pandas.tools", "pandas.core.reshape", "X"
+    "pandas.types", "pandas.core.dtypes", "X"
+    "pandas.io.sas.saslib", "pandas.io.sas._sas", ""
+    "pandas._join", "pandas._libs.join", ""
+    "pandas._hash", "pandas._libs.hashing", ""
+    "pandas._period", "pandas._libs.period", ""
+    "pandas._sparse", "pandas._libs.sparse", ""
+    "pandas._testing", "pandas._libs.testing", ""
+    "pandas._window", "pandas._libs.window", ""
+
+
+Some new subpackages are created with public functionality that is not directly
+exposed in the top-level namespace: ``pandas.errors``, ``pandas.plotting`` and
+``pandas.testing`` (more details below). Together with ``pandas.api.types`` and
+certain functions in the ``pandas.io`` and ``pandas.tseries`` submodules,
+these are now the public subpackages.
+
+Further changes:
+
+- The function :func:`~pandas.api.types.union_categoricals` is now importable from ``pandas.api.types``, formerly from ``pandas.types.concat`` (:issue:`15998`)
+- The type import ``pandas.tslib.NaTType`` is deprecated and can be replaced by using ``type(pandas.NaT)`` (:issue:`16146`)
+- The public functions in ``pandas.tools.hashing`` deprecated from that locations, but are now importable from ``pandas.util`` (:issue:`16223`)
+- The modules in ``pandas.util``: ``decorators``, ``print_versions``, ``doctools``, ``validators``, ``depr_module`` are now private. Only the functions exposed in ``pandas.util`` itself are public (:issue:`16223`)
+
+.. _whatsnew_0200.privacy.errors:
+
+``pandas.errors``
+^^^^^^^^^^^^^^^^^
+
+We are adding a standard public module for all pandas exceptions & warnings ``pandas.errors``. (:issue:`14800`). Previously
+these exceptions & warnings could be imported from ``pandas.core.common`` or ``pandas.io.common``. These exceptions and warnings
+will be removed from the ``*.common`` locations in a future release. (:issue:`15541`)
+
+The following are now part of this API:
+
+.. code-block:: python
+
+   ['DtypeWarning',
+    'EmptyDataError',
+    'OutOfBoundsDatetime',
+    'ParserError',
+    'ParserWarning',
+    'PerformanceWarning',
+    'UnsortedIndexError',
+    'UnsupportedFunctionCall']
+
+
+.. _whatsnew_0200.privacy.testing:
+
+``pandas.testing``
+^^^^^^^^^^^^^^^^^^
+
+We are adding a standard module that exposes the public testing functions in ``pandas.testing`` (:issue:`9895`). Those functions can be used when writing tests for functionality using pandas objects.
+
+The following testing functions are now part of this API:
+
+- :func:`testing.assert_frame_equal`
+- :func:`testing.assert_series_equal`
+- :func:`testing.assert_index_equal`
+
+
+.. _whatsnew_0200.privacy.plotting:
+
+``pandas.plotting``
+^^^^^^^^^^^^^^^^^^^
+
+A new public ``pandas.plotting`` module has been added that holds plotting functionality that was previously in either ``pandas.tools.plotting`` or in the top-level namespace. See the :ref:`deprecations sections <whatsnew_0200.privacy.deprecate_plotting>` for more details.
+
+.. _whatsnew_0200.privacy.development:
+
+Other development changes
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Building pandas for development now requires ``cython >= 0.23`` (:issue:`14831`)
+- Require at least 0.23 version of cython to avoid problems with character encodings (:issue:`14699`)
+- Switched the test framework to use `pytest <http://doc.pytest.org/en/latest>`__ (:issue:`13097`)
+- Reorganization of tests directory layout (:issue:`14854`, :issue:`15707`).
+
+
+.. _whatsnew_0200.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+.. _whatsnew_0200.api_breaking.deprecate_ix:
+
+Deprecate ``.ix``
+^^^^^^^^^^^^^^^^^
+
+The ``.ix`` indexer is deprecated, in favor of the more strict ``.iloc`` and ``.loc`` indexers. ``.ix`` offers a lot of magic on the inference of what the user wants to do. More specifically, ``.ix`` can decide to index *positionally* OR via *labels*, depending on the data type of the index. This has caused quite a bit of user confusion over the years. The full indexing documentation is :ref:`here <indexing>`. (:issue:`14218`)
+
+The recommended methods of indexing are:
+
+- ``.loc`` if you want to *label* index
+- ``.iloc`` if you want to *positionally* index.
+
+Using ``.ix`` will now show a ``DeprecationWarning`` with a link to some examples of how to convert code `here <https://pandas.pydata.org/pandas-docs/version/1.0/user_guide/indexing.html#ix-indexer-is-deprecated>`__.
+
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 2, 3],
+                      'B': [4, 5, 6]},
+                     index=list('abc'))
+
+   df
+
+Previous behavior, where you wish to get the 0th and the 2nd elements from the index in the 'A' column.
+
+.. code-block:: ipython
+
+   In [3]: df.ix[[0, 2], 'A']
+   Out[3]:
+   a    1
+   c    3
+   Name: A, dtype: int64
+
+Using ``.loc``. Here we will select the appropriate indexes from the index, then use *label* indexing.
+
+.. ipython:: python
+
+   df.loc[df.index[[0, 2]], 'A']
+
+Using ``.iloc``. Here we will get the location of the 'A' column, then use *positional* indexing to select things.
+
+.. ipython:: python
+
+   df.iloc[[0, 2], df.columns.get_loc('A')]
+
+
+.. _whatsnew_0200.api_breaking.deprecate_panel:
+
+Deprecate Panel
+^^^^^^^^^^^^^^^
+
+``Panel`` is deprecated and will be removed in a future version. The recommended way to represent 3-D data are
+with a ``MultiIndex`` on a ``DataFrame`` via the :meth:`~Panel.to_frame` or with the `xarray package <http://xarray.pydata.org/en/stable/>`__. pandas
+provides a :meth:`~Panel.to_xarray` method to automate this conversion (:issue:`13563`).
+
+.. code-block:: ipython
+
+    In [133]: import pandas._testing as tm
+
+    In [134]: p = tm.makePanel()
+
+    In [135]: p
+    Out[135]:
+    <class 'pandas.core.panel.Panel'>
+    Dimensions: 3 (items) x 3 (major_axis) x 4 (minor_axis)
+    Items axis: ItemA to ItemC
+    Major_axis axis: 2000-01-03 00:00:00 to 2000-01-05 00:00:00
+    Minor_axis axis: A to D
+
+Convert to a MultiIndex DataFrame
+
+.. code-block:: ipython
+
+    In [136]: p.to_frame()
+    Out[136]:
+                         ItemA     ItemB     ItemC
+    major      minor
+    2000-01-03 A      0.628776 -1.409432  0.209395
+               B      0.988138 -1.347533 -0.896581
+               C     -0.938153  1.272395 -0.161137
+               D     -0.223019 -0.591863 -1.051539
+    2000-01-04 A      0.186494  1.422986 -0.592886
+               B     -0.072608  0.363565  1.104352
+               C     -1.239072 -1.449567  0.889157
+               D      2.123692 -0.414505 -0.319561
+    2000-01-05 A      0.952478 -2.147855 -1.473116
+               B     -0.550603 -0.014752 -0.431550
+               C      0.139683 -1.195524  0.288377
+               D      0.122273 -1.425795 -0.619993
+
+    [12 rows x 3 columns]
+
+Convert to an xarray DataArray
+
+.. code-block:: ipython
+
+    In [137]: p.to_xarray()
+    Out[137]:
+    <xarray.DataArray (items: 3, major_axis: 3, minor_axis: 4)>
+    array([[[ 0.628776,  0.988138, -0.938153, -0.223019],
+            [ 0.186494, -0.072608, -1.239072,  2.123692],
+            [ 0.952478, -0.550603,  0.139683,  0.122273]],
+
+           [[-1.409432, -1.347533,  1.272395, -0.591863],
+            [ 1.422986,  0.363565, -1.449567, -0.414505],
+            [-2.147855, -0.014752, -1.195524, -1.425795]],
+
+           [[ 0.209395, -0.896581, -0.161137, -1.051539],
+            [-0.592886,  1.104352,  0.889157, -0.319561],
+            [-1.473116, -0.43155 ,  0.288377, -0.619993]]])
+    Coordinates:
+      * items       (items) object 'ItemA' 'ItemB' 'ItemC'
+      * major_axis  (major_axis) datetime64[ns] 2000-01-03 2000-01-04 2000-01-05
+      * minor_axis  (minor_axis) object 'A' 'B' 'C' 'D'
+
+.. _whatsnew_0200.api_breaking.deprecate_group_agg_dict:
+
+Deprecate groupby.agg() with a dictionary when renaming
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``.groupby(..).agg(..)``, ``.rolling(..).agg(..)``, and ``.resample(..).agg(..)``  syntax can accept a variable of inputs, including scalars,
+list, and a dict of column names to scalars or lists. This provides a useful syntax for constructing multiple
+(potentially different) aggregations.
+
+However, ``.agg(..)`` can *also* accept a dict that allows 'renaming' of the result columns. This is a complicated and confusing syntax, as well as not consistent
+between ``Series`` and ``DataFrame``. We are deprecating this 'renaming' functionality.
+
+- We are deprecating passing a dict to a grouped/rolled/resampled ``Series``. This allowed
+  one to ``rename`` the resulting aggregation, but this had a completely different
+  meaning than passing a dictionary to a grouped ``DataFrame``, which accepts column-to-aggregations.
+- We are deprecating passing a dict-of-dicts to a grouped/rolled/resampled ``DataFrame`` in a similar manner.
+
+This is an illustrative example:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 1, 1, 2, 2],
+                      'B': range(5),
+                      'C': range(5)})
+   df
+
+Here is a typical useful syntax for computing different aggregations for different columns. This
+is a natural, and useful syntax. We aggregate from the dict-to-list by taking the specified
+columns and applying the list of functions. This returns a ``MultiIndex`` for the columns (this is *not* deprecated).
+
+.. ipython:: python
+
+   df.groupby('A').agg({'B': 'sum', 'C': 'min'})
+
+Here's an example of the first deprecation, passing a dict to a grouped ``Series``. This
+is a combination aggregation & renaming:
+
+.. code-block:: ipython
+
+   In [6]: df.groupby('A').B.agg({'foo': 'count'})
+   FutureWarning: using a dict on a Series for aggregation
+   is deprecated and will be removed in a future version
+
+   Out[6]:
+      foo
+   A
+   1    3
+   2    2
+
+You can accomplish the same operation, more idiomatically by:
+
+.. ipython:: python
+
+   df.groupby('A').B.agg(['count']).rename(columns={'count': 'foo'})
+
+
+Here's an example of the second deprecation, passing a dict-of-dict to a grouped ``DataFrame``:
+
+.. code-block:: python
+
+   In [23]: (df.groupby('A')
+       ...:    .agg({'B': {'foo': 'sum'}, 'C': {'bar': 'min'}})
+       ...:  )
+   FutureWarning: using a dict with renaming is deprecated and
+   will be removed in a future version
+
+   Out[23]:
+        B   C
+      foo bar
+   A
+   1   3   0
+   2   7   3
+
+
+You can accomplish nearly the same by:
+
+.. ipython:: python
+
+   (df.groupby('A')
+      .agg({'B': 'sum', 'C': 'min'})
+      .rename(columns={'B': 'foo', 'C': 'bar'})
+    )
+
+
+
+.. _whatsnew_0200.privacy.deprecate_plotting:
+
+Deprecate .plotting
+^^^^^^^^^^^^^^^^^^^
+
+The ``pandas.tools.plotting`` module has been deprecated,  in favor of the top level ``pandas.plotting`` module. All the public plotting functions are now available
+from ``pandas.plotting`` (:issue:`12548`).
+
+Furthermore, the top-level ``pandas.scatter_matrix`` and ``pandas.plot_params`` are deprecated.
+Users can import these from ``pandas.plotting`` as well.
+
+Previous script:
+
+.. code-block:: python
+
+   pd.tools.plotting.scatter_matrix(df)
+   pd.scatter_matrix(df)
+
+Should be changed to:
+
+.. code-block:: python
+
+   pd.plotting.scatter_matrix(df)
+
+
+
+.. _whatsnew_0200.deprecations.other:
+
+Other deprecations
+^^^^^^^^^^^^^^^^^^
+
+- ``SparseArray.to_dense()`` has deprecated the ``fill`` parameter, as that parameter was not being respected (:issue:`14647`)
+- ``SparseSeries.to_dense()`` has deprecated the ``sparse_only`` parameter (:issue:`14647`)
+- ``Series.repeat()`` has deprecated the ``reps`` parameter in favor of ``repeats`` (:issue:`12662`)
+- The ``Series`` constructor and ``.astype`` method have deprecated accepting timestamp dtypes without a frequency (e.g. ``np.datetime64``) for the ``dtype`` parameter (:issue:`15524`)
+- ``Index.repeat()`` and ``MultiIndex.repeat()`` have deprecated the ``n`` parameter in favor of ``repeats`` (:issue:`12662`)
+- ``Categorical.searchsorted()`` and ``Series.searchsorted()`` have deprecated the ``v`` parameter in favor of ``value`` (:issue:`12662`)
+- ``TimedeltaIndex.searchsorted()``, ``DatetimeIndex.searchsorted()``, and ``PeriodIndex.searchsorted()`` have deprecated the ``key`` parameter in favor of ``value`` (:issue:`12662`)
+- ``DataFrame.astype()`` has deprecated the ``raise_on_error`` parameter in favor of ``errors`` (:issue:`14878`)
+- ``Series.sortlevel`` and ``DataFrame.sortlevel`` have been deprecated in favor of ``Series.sort_index`` and ``DataFrame.sort_index`` (:issue:`15099`)
+- importing ``concat`` from ``pandas.tools.merge`` has been deprecated in favor of imports from the ``pandas`` namespace. This should only affect explicit imports (:issue:`15358`)
+- ``Series/DataFrame/Panel.consolidate()`` been deprecated as a public method. (:issue:`15483`)
+- The ``as_indexer`` keyword of ``Series.str.match()`` has been deprecated (ignored keyword) (:issue:`15257`).
+- The following top-level pandas functions have been deprecated and will be removed in a future version (:issue:`13790`, :issue:`15940`)
+
+  * ``pd.pnow()``, replaced by ``Period.now()``
+  * ``pd.Term``, is removed, as it is not applicable to user code. Instead use in-line string expressions in the where clause when searching in HDFStore
+  * ``pd.Expr``, is removed, as it is not applicable to user code.
+  * ``pd.match()``, is removed.
+  * ``pd.groupby()``, replaced by using the ``.groupby()`` method directly on a ``Series/DataFrame``
+  * ``pd.get_store()``, replaced by a direct call to ``pd.HDFStore(...)``
+- ``is_any_int_dtype``, ``is_floating_dtype``, and ``is_sequence`` are deprecated from ``pandas.api.types`` (:issue:`16042`)
+
+.. _whatsnew_0200.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- The ``pandas.rpy`` module is removed. Similar functionality can be accessed
+  through the `rpy2 <https://rpy2.readthedocs.io/>`__ project.
+  See the `R interfacing docs <https://pandas.pydata.org/pandas-docs/version/0.20/r_interface.html>`__ for more details.
+- The ``pandas.io.ga`` module with a ``google-analytics`` interface is removed (:issue:`11308`).
+  Similar functionality can be found in the `Google2Pandas <https://github.com/panalysis/Google2Pandas>`__ package.
+- ``pd.to_datetime`` and ``pd.to_timedelta`` have dropped the ``coerce`` parameter in favor of ``errors`` (:issue:`13602`)
+- ``pandas.stats.fama_macbeth``, ``pandas.stats.ols``, ``pandas.stats.plm`` and ``pandas.stats.var``, as well as the top-level ``pandas.fama_macbeth`` and ``pandas.ols`` routines are removed. Similar functionality can be found in the `statsmodels <https://www.statsmodels.org/dev/>`__ package. (:issue:`11898`)
+- The ``TimeSeries`` and ``SparseTimeSeries`` classes, aliases of ``Series``
+  and ``SparseSeries``, are removed (:issue:`10890`, :issue:`15098`).
+- ``Series.is_time_series`` is dropped in favor of ``Series.index.is_all_dates`` (:issue:`15098`)
+- The deprecated ``irow``, ``icol``, ``iget`` and ``iget_value`` methods are removed
+  in favor of ``iloc`` and ``iat`` as explained :ref:`here <whatsnew_0170.deprecations>` (:issue:`10711`).
+- The deprecated ``DataFrame.iterkv()`` has been removed in favor of ``DataFrame.iteritems()`` (:issue:`10711`)
+- The ``Categorical`` constructor has dropped the ``name`` parameter (:issue:`10632`)
+- ``Categorical`` has dropped support for ``NaN`` categories (:issue:`10748`)
+- The ``take_last`` parameter has been dropped from ``duplicated()``, ``drop_duplicates()``, ``nlargest()``, and ``nsmallest()`` methods (:issue:`10236`, :issue:`10792`, :issue:`10920`)
+- ``Series``, ``Index``, and ``DataFrame`` have dropped the ``sort`` and ``order`` methods (:issue:`10726`)
+- Where clauses in ``pytables`` are only accepted as strings and expressions types and not other data-types (:issue:`12027`)
+- ``DataFrame`` has dropped the ``combineAdd`` and ``combineMult`` methods in favor of ``add`` and ``mul`` respectively (:issue:`10735`)
+
+.. _whatsnew_0200.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of ``pd.wide_to_long()`` (:issue:`14779`)
+- Improved performance of ``pd.factorize()`` by releasing the GIL with ``object`` dtype when inferred as strings (:issue:`14859`, :issue:`16057`)
+- Improved performance of timeseries plotting with an irregular DatetimeIndex
+  (or with ``compat_x=True``) (:issue:`15073`).
+- Improved performance of ``groupby().cummin()`` and ``groupby().cummax()`` (:issue:`15048`, :issue:`15109`, :issue:`15561`, :issue:`15635`)
+- Improved performance and reduced memory when indexing with a ``MultiIndex`` (:issue:`15245`)
+- When reading buffer object in ``read_sas()`` method without specified format, filepath string is inferred rather than buffer object. (:issue:`14947`)
+- Improved performance of ``.rank()`` for categorical data (:issue:`15498`)
+- Improved performance when using ``.unstack()`` (:issue:`15503`)
+- Improved performance of merge/join on ``category`` columns (:issue:`10409`)
+- Improved performance of ``drop_duplicates()`` on ``bool`` columns (:issue:`12963`)
+- Improve performance of ``pd.core.groupby.GroupBy.apply`` when the applied
+  function used the ``.name`` attribute of the group DataFrame (:issue:`15062`).
+- Improved performance of ``iloc`` indexing with a list or array (:issue:`15504`).
+- Improved performance of ``Series.sort_index()`` with a monotonic index (:issue:`15694`)
+- Improved performance in ``pd.read_csv()`` on some platforms with buffered reads (:issue:`16039`)
+
+.. _whatsnew_0200.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Conversion
+^^^^^^^^^^
+
+- Bug in ``Timestamp.replace`` now raises ``TypeError`` when incorrect argument names are given; previously this raised ``ValueError`` (:issue:`15240`)
+- Bug in ``Timestamp.replace`` with compat for passing long integers (:issue:`15030`)
+- Bug in ``Timestamp`` returning UTC based time/date attributes when a timezone was provided (:issue:`13303`, :issue:`6538`)
+- Bug in ``Timestamp`` incorrectly localizing timezones during construction (:issue:`11481`, :issue:`15777`)
+- Bug in ``TimedeltaIndex`` addition where overflow was being allowed without error (:issue:`14816`)
+- Bug in ``TimedeltaIndex`` raising a ``ValueError`` when boolean indexing with ``loc`` (:issue:`14946`)
+- Bug in catching an overflow in ``Timestamp`` + ``Timedelta/Offset`` operations (:issue:`15126`)
+- Bug in ``DatetimeIndex.round()`` and ``Timestamp.round()`` floating point accuracy when rounding by milliseconds or less (:issue:`14440`, :issue:`15578`)
+- Bug in ``astype()`` where ``inf`` values were incorrectly converted to integers. Now raises error now with ``astype()`` for Series and DataFrames (:issue:`14265`)
+- Bug in ``DataFrame(..).apply(to_numeric)`` when values are of type decimal.Decimal. (:issue:`14827`)
+- Bug in ``describe()`` when passing a numpy array which does not contain the median to the ``percentiles`` keyword argument (:issue:`14908`)
+- Cleaned up ``PeriodIndex`` constructor, including raising on floats more consistently (:issue:`13277`)
+- Bug in using ``__deepcopy__`` on empty NDFrame objects (:issue:`15370`)
+- Bug in ``.replace()`` may result in incorrect dtypes. (:issue:`12747`, :issue:`15765`)
+- Bug in ``Series.replace`` and ``DataFrame.replace`` which failed on empty replacement dicts (:issue:`15289`)
+- Bug in ``Series.replace`` which replaced a numeric by string (:issue:`15743`)
+- Bug in ``Index`` construction with ``NaN`` elements and integer dtype specified (:issue:`15187`)
+- Bug in ``Series`` construction with a datetimetz (:issue:`14928`)
+- Bug in ``Series.dt.round()`` inconsistent behaviour on ``NaT`` 's with different arguments (:issue:`14940`)
+- Bug in ``Series`` constructor when both ``copy=True`` and ``dtype`` arguments are provided (:issue:`15125`)
+- Incorrect dtyped ``Series`` was returned by comparison methods (e.g., ``lt``, ``gt``, ...) against a constant for an empty ``DataFrame`` (:issue:`15077`)
+- Bug in ``Series.ffill()`` with mixed dtypes containing tz-aware datetimes. (:issue:`14956`)
+- Bug in ``DataFrame.fillna()`` where the argument ``downcast`` was ignored when fillna value was of type ``dict`` (:issue:`15277`)
+- Bug in ``.asfreq()``, where frequency was not set for empty ``Series`` (:issue:`14320`)
+- Bug in ``DataFrame`` construction with nulls and datetimes in a list-like (:issue:`15869`)
+- Bug in ``DataFrame.fillna()`` with tz-aware datetimes (:issue:`15855`)
+- Bug in ``is_string_dtype``, ``is_timedelta64_ns_dtype``, and ``is_string_like_dtype`` in which an error was raised when ``None`` was passed in (:issue:`15941`)
+- Bug in the return type of ``pd.unique`` on a ``Categorical``, which was returning an ndarray and not a ``Categorical`` (:issue:`15903`)
+- Bug in ``Index.to_series()`` where the index was not copied (and so mutating later would change the original), (:issue:`15949`)
+- Bug in indexing with partial string indexing with a len-1 DataFrame (:issue:`16071`)
+- Bug in ``Series`` construction where passing invalid dtype didn't raise an error. (:issue:`15520`)
+
+Indexing
+^^^^^^^^
+
+- Bug in ``Index`` power operations with reversed operands (:issue:`14973`)
+- Bug in ``DataFrame.sort_values()`` when sorting by multiple columns where one column is of type ``int64`` and contains ``NaT`` (:issue:`14922`)
+- Bug in ``DataFrame.reindex()`` in which ``method`` was ignored when passing ``columns`` (:issue:`14992`)
+- Bug in ``DataFrame.loc`` with indexing a ``MultiIndex`` with a ``Series`` indexer (:issue:`14730`, :issue:`15424`)
+- Bug in ``DataFrame.loc`` with indexing a ``MultiIndex`` with a numpy array (:issue:`15434`)
+- Bug in ``Series.asof`` which raised if the series contained all ``np.nan`` (:issue:`15713`)
+- Bug in ``.at`` when selecting from a tz-aware column (:issue:`15822`)
+- Bug in ``Series.where()`` and ``DataFrame.where()`` where array-like conditionals were being rejected (:issue:`15414`)
+- Bug in ``Series.where()`` where TZ-aware data was converted to float representation (:issue:`15701`)
+- Bug in ``.loc`` that would not return the correct dtype for scalar access for a DataFrame (:issue:`11617`)
+- Bug in output formatting of a ``MultiIndex`` when names are integers (:issue:`12223`, :issue:`15262`)
+- Bug in ``Categorical.searchsorted()`` where alphabetical instead of the provided categorical order was used (:issue:`14522`)
+- Bug in ``Series.iloc`` where a ``Categorical`` object for list-like indexes input was returned, where a ``Series`` was expected. (:issue:`14580`)
+- Bug in ``DataFrame.isin`` comparing datetimelike to empty frame (:issue:`15473`)
+- Bug in ``.reset_index()`` when an all ``NaN`` level of a ``MultiIndex`` would fail (:issue:`6322`)
+- Bug in ``.reset_index()`` when raising error for index name already present in ``MultiIndex`` columns (:issue:`16120`)
+- Bug in creating a ``MultiIndex`` with tuples and not passing a list of names; this will now raise ``ValueError`` (:issue:`15110`)
+- Bug in the HTML display with a ``MultiIndex`` and truncation (:issue:`14882`)
+- Bug in the display of ``.info()`` where a qualifier (+) would always be displayed with a ``MultiIndex`` that contains only non-strings (:issue:`15245`)
+- Bug in ``pd.concat()`` where the names of ``MultiIndex`` of resulting ``DataFrame`` are not handled correctly when ``None`` is presented in the names of ``MultiIndex`` of input ``DataFrame`` (:issue:`15787`)
+- Bug in ``DataFrame.sort_index()`` and ``Series.sort_index()`` where ``na_position`` doesn't work with a ``MultiIndex`` (:issue:`14784`, :issue:`16604`)
+- Bug in ``pd.concat()`` when combining objects with a ``CategoricalIndex`` (:issue:`16111`)
+- Bug in indexing with a scalar and a ``CategoricalIndex`` (:issue:`16123`)
+
+IO
+^^
+
+- Bug in ``pd.to_numeric()`` in which float and unsigned integer elements were being improperly casted (:issue:`14941`, :issue:`15005`)
+- Bug in ``pd.read_fwf()`` where the skiprows parameter was not being respected during column width inference (:issue:`11256`)
+- Bug in ``pd.read_csv()`` in which the ``dialect`` parameter was not being verified before processing (:issue:`14898`)
+- Bug in ``pd.read_csv()`` in which missing data was being improperly handled with ``usecols`` (:issue:`6710`)
+- Bug in ``pd.read_csv()`` in which a file containing a row with many columns followed by rows with fewer columns would cause a crash (:issue:`14125`)
+- Bug in ``pd.read_csv()`` for the C engine where ``usecols`` were being indexed incorrectly with ``parse_dates`` (:issue:`14792`)
+- Bug in ``pd.read_csv()`` with ``parse_dates`` when multi-line headers are specified (:issue:`15376`)
+- Bug in ``pd.read_csv()`` with ``float_precision='round_trip'`` which caused a segfault when a text entry is parsed (:issue:`15140`)
+- Bug in ``pd.read_csv()`` when an index was specified and no values were specified as null values (:issue:`15835`)
+- Bug in ``pd.read_csv()`` in which certain invalid file objects caused the Python interpreter to crash (:issue:`15337`)
+- Bug in ``pd.read_csv()`` in which invalid values for ``nrows`` and ``chunksize`` were allowed (:issue:`15767`)
+- Bug in ``pd.read_csv()`` for the Python engine in which unhelpful error messages were being raised when parsing errors occurred (:issue:`15910`)
+- Bug in ``pd.read_csv()`` in which the ``skipfooter`` parameter was not being properly validated (:issue:`15925`)
+- Bug in ``pd.to_csv()`` in which there was numeric overflow when a timestamp index was being written (:issue:`15982`)
+- Bug in ``pd.util.hashing.hash_pandas_object()`` in which hashing of categoricals depended on the ordering of categories, instead of just their values. (:issue:`15143`)
+- Bug in ``.to_json()`` where ``lines=True`` and contents (keys or values) contain escaped characters (:issue:`15096`)
+- Bug in ``.to_json()`` causing single byte ascii characters to be expanded to four byte unicode (:issue:`15344`)
+- Bug in ``.to_json()`` for the C engine where rollover was not correctly handled for case where frac is odd and diff is exactly 0.5 (:issue:`15716`, :issue:`15864`)
+- Bug in ``pd.read_json()`` for Python 2 where ``lines=True`` and contents contain non-ascii unicode characters (:issue:`15132`)
+- Bug in ``pd.read_msgpack()`` in which ``Series`` categoricals were being improperly processed (:issue:`14901`)
+- Bug in ``pd.read_msgpack()`` which did not allow loading of a dataframe with an index of type ``CategoricalIndex`` (:issue:`15487`)
+- Bug in ``pd.read_msgpack()`` when deserializing a ``CategoricalIndex`` (:issue:`15487`)
+- Bug in ``DataFrame.to_records()`` with converting a ``DatetimeIndex`` with a timezone (:issue:`13937`)
+- Bug in ``DataFrame.to_records()`` which failed with unicode characters in column names (:issue:`11879`)
+- Bug in ``.to_sql()`` when writing a DataFrame with numeric index names (:issue:`15404`).
+- Bug in ``DataFrame.to_html()`` with ``index=False`` and ``max_rows`` raising in ``IndexError`` (:issue:`14998`)
+- Bug in ``pd.read_hdf()`` passing a ``Timestamp`` to the ``where`` parameter with a non date column (:issue:`15492`)
+- Bug in ``DataFrame.to_stata()`` and ``StataWriter`` which produces incorrectly formatted files to be produced for some locales (:issue:`13856`)
+- Bug in ``StataReader`` and ``StataWriter`` which allows invalid encodings (:issue:`15723`)
+- Bug in the ``Series`` repr not showing the length when the output was truncated (:issue:`15962`).
+
+Plotting
+^^^^^^^^
+
+- Bug in ``DataFrame.hist`` where ``plt.tight_layout`` caused an ``AttributeError``  (use ``matplotlib >= 2.0.1``) (:issue:`9351`)
+- Bug in ``DataFrame.boxplot`` where ``fontsize`` was not applied to the tick labels on both axes (:issue:`15108`)
+- Bug in the date and time converters pandas registers with matplotlib not handling multiple dimensions (:issue:`16026`)
+- Bug in ``pd.scatter_matrix()`` could accept either ``color`` or ``c``, but not both (:issue:`14855`)
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in ``.groupby(..).resample()`` when passed the ``on=`` kwarg. (:issue:`15021`)
+- Properly set ``__name__`` and ``__qualname__`` for ``Groupby.*`` functions (:issue:`14620`)
+- Bug in ``GroupBy.get_group()`` failing with a categorical grouper (:issue:`15155`)
+- Bug in ``.groupby(...).rolling(...)`` when ``on`` is specified and using a ``DatetimeIndex`` (:issue:`15130`, :issue:`13966`)
+- Bug in groupby operations with ``timedelta64`` when passing ``numeric_only=False`` (:issue:`5724`)
+- Bug in ``groupby.apply()`` coercing ``object`` dtypes to numeric types, when not all values were numeric (:issue:`14423`, :issue:`15421`, :issue:`15670`)
+- Bug in ``resample``, where a non-string ``loffset`` argument would not be applied when resampling a timeseries (:issue:`13218`)
+- Bug in ``DataFrame.groupby().describe()`` when grouping on ``Index`` containing tuples (:issue:`14848`)
+- Bug in ``groupby().nunique()`` with a datetimelike-grouper where bins counts were incorrect (:issue:`13453`)
+- Bug in ``groupby.transform()`` that would coerce the resultant dtypes back to the original (:issue:`10972`, :issue:`11444`)
+- Bug in ``groupby.agg()`` incorrectly localizing timezone on ``datetime`` (:issue:`15426`, :issue:`10668`, :issue:`13046`)
+- Bug in ``.rolling/expanding()`` functions where ``count()`` was not counting ``np.Inf``, nor handling ``object`` dtypes (:issue:`12541`)
+- Bug in ``.rolling()`` where ``pd.Timedelta`` or ``datetime.timedelta`` was not accepted as a ``window`` argument (:issue:`15440`)
+- Bug in ``Rolling.quantile`` function that caused a segmentation fault when called with a quantile value outside of the range [0, 1] (:issue:`15463`)
+- Bug in ``DataFrame.resample().median()`` if duplicate column names are present (:issue:`14233`)
+
+Sparse
+^^^^^^
+
+- Bug in ``SparseSeries.reindex`` on single level with list of length 1 (:issue:`15447`)
+- Bug in repr-formatting a ``SparseDataFrame`` after a value was set on (a copy of) one of its series (:issue:`15488`)
+- Bug in ``SparseDataFrame`` construction with lists not coercing to dtype (:issue:`15682`)
+- Bug in sparse array indexing in which indices were not being validated (:issue:`15863`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in ``pd.merge_asof()`` where ``left_index`` or ``right_index`` caused a failure when multiple ``by`` was specified (:issue:`15676`)
+- Bug in ``pd.merge_asof()`` where ``left_index``/``right_index`` together caused a failure when ``tolerance`` was specified (:issue:`15135`)
+- Bug in ``DataFrame.pivot_table()`` where ``dropna=True`` would not drop all-NaN columns when the columns was a ``category`` dtype (:issue:`15193`)
+- Bug in ``pd.melt()`` where passing a tuple value for ``value_vars`` caused a ``TypeError`` (:issue:`15348`)
+- Bug in ``pd.pivot_table()`` where no error was raised when values argument was not in the columns (:issue:`14938`)
+- Bug in ``pd.concat()`` in which concatenating with an empty dataframe with ``join='inner'`` was being improperly handled (:issue:`15328`)
+- Bug with ``sort=True`` in ``DataFrame.join`` and ``pd.merge`` when joining on indexes (:issue:`15582`)
+- Bug in ``DataFrame.nsmallest`` and ``DataFrame.nlargest`` where identical values resulted in duplicated rows (:issue:`15297`)
+- Bug in :func:`pandas.pivot_table` incorrectly raising ``UnicodeError`` when passing unicode input for ``margins`` keyword (:issue:`13292`)
+
+Numeric
+^^^^^^^
+
+- Bug in ``.rank()`` which incorrectly ranks ordered categories (:issue:`15420`)
+- Bug in ``.corr()`` and ``.cov()`` where the column and index were the same object (:issue:`14617`)
+- Bug in ``.mode()`` where ``mode`` was not returned if was only a single value (:issue:`15714`)
+- Bug in ``pd.cut()`` with a single bin on an all 0s array (:issue:`15428`)
+- Bug in ``pd.qcut()`` with a single quantile and an array with identical values (:issue:`15431`)
+- Bug in ``pandas.tools.utils.cartesian_product()`` with large input can cause overflow on windows (:issue:`15265`)
+- Bug in ``.eval()`` which caused multi-line evals to fail with local variables not on the first line (:issue:`15342`)
+
+Other
+^^^^^
+
+- Compat with SciPy 0.19.0 for testing on ``.interpolate()`` (:issue:`15662`)
+- Compat for 32-bit platforms for ``.qcut/cut``; bins will now be ``int64`` dtype (:issue:`14866`)
+- Bug in interactions with ``Qt`` when a ``QtApplication`` already exists (:issue:`14372`)
+- Avoid use of ``np.finfo()`` during ``import pandas`` removed to mitigate deadlock on Python GIL misuse (:issue:`14641`)
+
+
+.. _whatsnew_0.20.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.19.2..v0.20.0
diff --git a/pandas/doc/source/whatsnew/v0.20.2.rst b/pandas/doc/source/whatsnew/v0.20.2.rst
new file mode 100644
index 0000000000000..6945b360a97b0
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.20.2.rst
@@ -0,0 +1,143 @@
+.. _whatsnew_0202:
+
+Version 0.20.2 (June 4, 2017)
+-----------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a minor bug-fix release in the 0.20.x series and includes some small regression fixes,
+bug fixes and performance improvements.
+We recommend that all users upgrade to this version.
+
+.. contents:: What's new in v0.20.2
+    :local:
+    :backlinks: none
+
+
+.. _whatsnew_0202.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+- Unblocked access to additional compression types supported in pytables: 'blosc:blosclz, 'blosc:lz4', 'blosc:lz4hc', 'blosc:snappy', 'blosc:zlib', 'blosc:zstd' (:issue:`14478`)
+- ``Series`` provides a ``to_latex`` method (:issue:`16180`)
+
+- A new groupby method :meth:`.GroupBy.ngroup`,
+  parallel to the existing :meth:`.GroupBy.cumcount`,
+  has been added to return the group order (:issue:`11642`); see
+  :ref:`here <groupby.ngroup>`.
+
+.. _whatsnew_0202.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Performance regression fix when indexing with a list-like (:issue:`16285`)
+- Performance regression fix for MultiIndexes (:issue:`16319`, :issue:`16346`)
+- Improved performance of ``.clip()`` with scalar arguments (:issue:`15400`)
+- Improved performance of groupby with categorical groupers (:issue:`16413`)
+- Improved performance of ``MultiIndex.remove_unused_levels()`` (:issue:`16556`)
+
+.. _whatsnew_0202.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Silenced a warning on some Windows environments about "tput: terminal attributes: No such device or address" when
+  detecting the terminal size. This fix only applies to python 3 (:issue:`16496`)
+- Bug in using ``pathlib.Path`` or ``py.path.local`` objects with io functions (:issue:`16291`)
+- Bug in ``Index.symmetric_difference()`` on two equal MultiIndex's, results in a ``TypeError`` (:issue:`13490`)
+- Bug in ``DataFrame.update()`` with ``overwrite=False`` and ``NaN values`` (:issue:`15593`)
+- Passing an invalid engine to :func:`read_csv` now raises an informative
+  ``ValueError`` rather than ``UnboundLocalError``. (:issue:`16511`)
+- Bug in :func:`unique` on an array of tuples (:issue:`16519`)
+- Bug in :func:`cut` when ``labels`` are set, resulting in incorrect label ordering (:issue:`16459`)
+- Fixed a compatibility issue with IPython 6.0's tab completion showing deprecation warnings on ``Categoricals`` (:issue:`16409`)
+
+Conversion
+^^^^^^^^^^
+
+- Bug in :func:`to_numeric` in which empty data inputs were causing a segfault of the interpreter (:issue:`16302`)
+- Silence numpy warnings when broadcasting ``DataFrame`` to ``Series`` with comparison ops (:issue:`16378`, :issue:`16306`)
+
+
+Indexing
+^^^^^^^^
+
+- Bug in ``DataFrame.reset_index(level=)`` with single level index (:issue:`16263`)
+- Bug in partial string indexing with a monotonic, but not strictly-monotonic, index incorrectly reversing the slice bounds (:issue:`16515`)
+- Bug in ``MultiIndex.remove_unused_levels()`` that would not return a ``MultiIndex`` equal to the original. (:issue:`16556`)
+
+IO
+^^
+
+- Bug in :func:`read_csv` when ``comment`` is passed in a space delimited text file (:issue:`16472`)
+- Bug in :func:`read_csv` not raising an exception with nonexistent columns in ``usecols`` when it had the correct length (:issue:`14671`)
+- Bug that would force importing of the clipboard routines unnecessarily, potentially causing an import error on startup (:issue:`16288`)
+- Bug that raised ``IndexError`` when HTML-rendering an empty ``DataFrame`` (:issue:`15953`)
+- Bug in :func:`read_csv` in which tarfile object inputs were raising an error in Python 2.x for the C engine (:issue:`16530`)
+- Bug where ``DataFrame.to_html()`` ignored the ``index_names`` parameter (:issue:`16493`)
+- Bug where ``pd.read_hdf()`` returns numpy strings for index names (:issue:`13492`)
+
+- Bug in ``HDFStore.select_as_multiple()`` where start/stop arguments were not respected (:issue:`16209`)
+
+Plotting
+^^^^^^^^
+
+- Bug in ``DataFrame.plot`` with a single column and a list-like ``color`` (:issue:`3486`)
+- Bug in ``plot`` where ``NaT`` in ``DatetimeIndex`` results in ``Timestamp.min`` (:issue:`12405`)
+- Bug in ``DataFrame.boxplot`` where ``figsize`` keyword was not respected for non-grouped boxplots (:issue:`11959`)
+
+
+
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in creating a time-based rolling window on an empty ``DataFrame`` (:issue:`15819`)
+- Bug in ``rolling.cov()`` with offset window (:issue:`16058`)
+- Bug in ``.resample()`` and ``.groupby()`` when aggregating on integers (:issue:`16361`)
+
+
+Sparse
+^^^^^^
+
+- Bug in construction of ``SparseDataFrame`` from ``scipy.sparse.dok_matrix`` (:issue:`16179`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in ``DataFrame.stack`` with unsorted levels in ``MultiIndex`` columns (:issue:`16323`)
+- Bug in ``pd.wide_to_long()`` where no error was raised when ``i`` was not a unique identifier (:issue:`16382`)
+- Bug in ``Series.isin(..)`` with a list of tuples (:issue:`16394`)
+- Bug in construction of a ``DataFrame`` with mixed dtypes including an all-NaT column. (:issue:`16395`)
+- Bug in ``DataFrame.agg()`` and ``Series.agg()`` with aggregating on non-callable attributes (:issue:`16405`)
+
+
+Numeric
+^^^^^^^
+- Bug in ``.interpolate()``, where ``limit_direction`` was not respected when ``limit=None`` (default) was passed (:issue:`16282`)
+
+Categorical
+^^^^^^^^^^^
+
+- Fixed comparison operations considering the order of the categories when both categoricals are unordered (:issue:`16014`)
+
+Other
+^^^^^
+
+- Bug in ``DataFrame.drop()`` with an empty-list with non-unique indices (:issue:`16270`)
+
+
+.. _whatsnew_0.20.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.20.0..v0.20.2
diff --git a/pandas/doc/source/whatsnew/v0.20.3.rst b/pandas/doc/source/whatsnew/v0.20.3.rst
new file mode 100644
index 0000000000000..ff28f6830783e
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.20.3.rst
@@ -0,0 +1,76 @@
+.. _whatsnew_0203:
+
+Version 0.20.3 (July 7, 2017)
+-----------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a minor bug-fix release in the 0.20.x series and includes some small regression fixes
+and bug fixes. We recommend that all users upgrade to this version.
+
+.. contents:: What's new in v0.20.3
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0203.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Fixed a bug in failing to compute rolling computations of a column-MultiIndexed ``DataFrame`` (:issue:`16789`, :issue:`16825`)
+- Fixed a pytest marker failing downstream packages' tests suites (:issue:`16680`)
+
+Conversion
+^^^^^^^^^^
+
+- Bug in pickle compat prior to the v0.20.x series, when ``UTC`` is a timezone in a Series/DataFrame/Index (:issue:`16608`)
+- Bug in ``Series`` construction when passing a ``Series`` with ``dtype='category'`` (:issue:`16524`).
+- Bug in :meth:`DataFrame.astype` when passing a ``Series`` as the ``dtype`` kwarg. (:issue:`16717`).
+
+Indexing
+^^^^^^^^
+
+- Bug in ``Float64Index`` causing an empty array instead of ``None`` to be returned from ``.get(np.nan)`` on a Series whose index did not contain any ``NaN`` s (:issue:`8569`)
+- Bug in ``MultiIndex.isin`` causing an error when passing an empty iterable (:issue:`16777`)
+- Fixed a bug in a slicing DataFrame/Series that have a  ``TimedeltaIndex`` (:issue:`16637`)
+
+IO
+^^
+
+- Bug in :func:`read_csv` in which files weren't opened as binary files by the C engine on Windows, causing EOF characters mid-field, which would fail (:issue:`16039`, :issue:`16559`, :issue:`16675`)
+- Bug in :func:`read_hdf` in which reading a ``Series`` saved to an HDF file in 'fixed' format fails when an explicit ``mode='r'`` argument is supplied (:issue:`16583`)
+- Bug in :meth:`DataFrame.to_latex` where ``bold_rows`` was wrongly specified to be ``True`` by default, whereas in reality row labels remained non-bold whatever parameter provided. (:issue:`16707`)
+- Fixed an issue with :meth:`DataFrame.style` where generated element ids were not unique (:issue:`16780`)
+- Fixed loading a ``DataFrame`` with a ``PeriodIndex``, from a ``format='fixed'`` HDFStore, in Python 3, that was written in Python 2 (:issue:`16781`)
+
+Plotting
+^^^^^^^^
+
+- Fixed regression that prevented RGB and RGBA tuples from being used as color arguments (:issue:`16233`)
+- Fixed an issue with :meth:`DataFrame.plot.scatter` that incorrectly raised a ``KeyError`` when categorical data is used for plotting (:issue:`16199`)
+
+Reshaping
+^^^^^^^^^
+
+- ``PeriodIndex`` / ``TimedeltaIndex.join`` was missing the ``sort=`` kwarg (:issue:`16541`)
+- Bug in joining on a ``MultiIndex`` with a ``category`` dtype for a level (:issue:`16627`).
+- Bug in :func:`merge` when merging/joining with multiple categorical columns (:issue:`16767`)
+
+Categorical
+^^^^^^^^^^^
+
+- Bug in ``DataFrame.sort_values`` not respecting the ``kind`` parameter with categorical data (:issue:`16793`)
+
+
+.. _whatsnew_0.20.3.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.20.2..v0.20.3
diff --git a/pandas/doc/source/whatsnew/v0.21.0.rst b/pandas/doc/source/whatsnew/v0.21.0.rst
new file mode 100644
index 0000000000000..43719cd53b0ff
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.21.0.rst
@@ -0,0 +1,1232 @@
+.. _whatsnew_0210:
+
+Version 0.21.0 (October 27, 2017)
+---------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import * # noqa F401, F403
+
+
+This is a major release from 0.20.3 and includes a number of API changes, deprecations, new features,
+enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
+users upgrade to this version.
+
+Highlights include:
+
+- Integration with `Apache Parquet <https://parquet.apache.org/>`__, including a new top-level :func:`read_parquet` function and :meth:`DataFrame.to_parquet` method, see :ref:`here <whatsnew_0210.enhancements.parquet>`.
+- New user-facing :class:`pandas.api.types.CategoricalDtype` for specifying
+  categoricals independent of the data, see :ref:`here <whatsnew_0210.enhancements.categorical_dtype>`.
+- The behavior of ``sum`` and ``prod`` on all-NaN Series/DataFrames is now consistent and no longer depends on whether `bottleneck <https://bottleneck.readthedocs.io>`__ is installed, and ``sum`` and ``prod`` on empty Series now return NaN instead of 0, see :ref:`here <whatsnew_0210.api_breaking.bottleneck>`.
+- Compatibility fixes for pypy, see :ref:`here <whatsnew_0210.pypy>`.
+- Additions to the ``drop``, ``reindex`` and ``rename`` API to make them more consistent, see :ref:`here <whatsnew_0210.enhancements.drop_api>`.
+- Addition of the new methods ``DataFrame.infer_objects`` (see :ref:`here <whatsnew_0210.enhancements.infer_objects>`) and ``GroupBy.pipe`` (see :ref:`here <whatsnew_0210.enhancements.GroupBy_pipe>`).
+- Indexing with a list of labels, where one or more of the labels is missing, is deprecated and will raise a KeyError in a future version, see :ref:`here <whatsnew_0210.api_breaking.loc>`.
+
+Check the :ref:`API Changes <whatsnew_0210.api_breaking>` and :ref:`deprecations <whatsnew_0210.deprecations>` before updating.
+
+.. contents:: What's new in v0.21.0
+    :local:
+    :backlinks: none
+    :depth: 2
+
+.. _whatsnew_0210.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0210.enhancements.parquet:
+
+Integration with Apache Parquet file format
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Integration with `Apache Parquet <https://parquet.apache.org/>`__, including a new top-level :func:`read_parquet` and :func:`DataFrame.to_parquet` method, see :ref:`here <io.parquet>` (:issue:`15838`, :issue:`17438`).
+
+`Apache Parquet <https://parquet.apache.org/>`__ provides a cross-language, binary file format for reading and writing data frames efficiently.
+Parquet is designed to faithfully serialize and de-serialize ``DataFrame`` s, supporting all of the pandas
+dtypes, including extension dtypes such as datetime with timezones.
+
+This functionality depends on either the `pyarrow <http://arrow.apache.org/docs/python/>`__ or `fastparquet <https://fastparquet.readthedocs.io/en/latest/>`__ library.
+For more details, see :ref:`the IO docs on Parquet <io.parquet>`.
+
+
+.. _whatsnew_0210.enhancements.infer_objects:
+
+Method ``infer_objects`` type conversion
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`DataFrame.infer_objects` and :meth:`Series.infer_objects`
+methods have been added to perform dtype inference on object columns, replacing
+some of the functionality of the deprecated ``convert_objects``
+method. See the documentation :ref:`here <basics.object_conversion>`
+for more details. (:issue:`11221`)
+
+This method only performs soft conversions on object columns, converting Python objects
+to native types, but not any coercive conversions. For example:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 2, 3],
+                      'B': np.array([1, 2, 3], dtype='object'),
+                      'C': ['1', '2', '3']})
+   df.dtypes
+   df.infer_objects().dtypes
+
+Note that column ``'C'`` was not converted - only scalar numeric types
+will be converted to a new type.  Other types of conversion should be accomplished
+using the :func:`to_numeric` function (or :func:`to_datetime`, :func:`to_timedelta`).
+
+.. ipython:: python
+
+   df = df.infer_objects()
+   df['C'] = pd.to_numeric(df['C'], errors='coerce')
+   df.dtypes
+
+.. _whatsnew_0210.enhancements.attribute_access:
+
+Improved warnings when attempting to create columns
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+New users are often puzzled by the relationship between column operations and
+attribute access on ``DataFrame`` instances (:issue:`7175`). One specific
+instance of this confusion is attempting to create a new column by setting an
+attribute on the ``DataFrame``:
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame({'one': [1., 2., 3.]})
+   In [2]: df.two = [4, 5, 6]
+
+This does not raise any obvious exceptions, but also does not create a new column:
+
+.. code-block:: ipython
+
+   In [3]: df
+   Out[3]:
+       one
+   0  1.0
+   1  2.0
+   2  3.0
+
+Setting a list-like data structure into a new attribute now raises a ``UserWarning`` about the potential for unexpected behavior. See :ref:`Attribute Access <indexing.attribute_access>`.
+
+.. _whatsnew_0210.enhancements.drop_api:
+
+Method ``drop`` now also accepts index/columns keywords
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`~DataFrame.drop` method has gained ``index``/``columns`` keywords as an
+alternative to specifying the ``axis``. This is similar to the behavior of ``reindex``
+(:issue:`12392`).
+
+For example:
+
+.. ipython:: python
+
+    df = pd.DataFrame(np.arange(8).reshape(2, 4),
+                      columns=['A', 'B', 'C', 'D'])
+    df
+    df.drop(['B', 'C'], axis=1)
+    # the following is now equivalent
+    df.drop(columns=['B', 'C'])
+
+.. _whatsnew_0210.enhancements.rename_reindex_axis:
+
+Methods ``rename``, ``reindex`` now also accept axis keyword
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`DataFrame.rename` and :meth:`DataFrame.reindex` methods have gained
+the ``axis`` keyword to specify the axis to target with the operation
+(:issue:`12392`).
+
+Here's ``rename``:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+   df.rename(str.lower, axis='columns')
+   df.rename(id, axis='index')
+
+And ``reindex``:
+
+.. ipython:: python
+
+   df.reindex(['A', 'B', 'C'], axis='columns')
+   df.reindex([0, 1, 3], axis='index')
+
+The "index, columns" style continues to work as before.
+
+.. ipython:: python
+
+   df.rename(index=id, columns=str.lower)
+   df.reindex(index=[0, 1, 3], columns=['A', 'B', 'C'])
+
+We *highly* encourage using named arguments to avoid confusion when using either
+style.
+
+.. _whatsnew_0210.enhancements.categorical_dtype:
+
+``CategoricalDtype`` for specifying categoricals
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`pandas.api.types.CategoricalDtype` has been added to the public API and
+expanded to include the ``categories`` and ``ordered`` attributes. A
+``CategoricalDtype`` can be used to specify the set of categories and
+orderedness of an array, independent of the data. This can be useful for example,
+when converting string data to a ``Categorical`` (:issue:`14711`,
+:issue:`15078`, :issue:`16015`, :issue:`17643`):
+
+.. ipython:: python
+
+   from pandas.api.types import CategoricalDtype
+
+   s = pd.Series(['a', 'b', 'c', 'a'])  # strings
+   dtype = CategoricalDtype(categories=['a', 'b', 'c', 'd'], ordered=True)
+   s.astype(dtype)
+
+One place that deserves special mention is in :meth:`read_csv`. Previously, with
+``dtype={'col': 'category'}``, the returned values and categories would always
+be strings.
+
+.. ipython:: python
+   :suppress:
+
+   from io import StringIO
+
+.. ipython:: python
+
+   data = 'A,B\na,1\nb,2\nc,3'
+   pd.read_csv(StringIO(data), dtype={'B': 'category'}).B.cat.categories
+
+Notice the "object" dtype.
+
+With a ``CategoricalDtype`` of all numerics, datetimes, or
+timedeltas, we can automatically convert to the correct type
+
+.. ipython:: python
+
+   dtype = {'B': CategoricalDtype([1, 2, 3])}
+   pd.read_csv(StringIO(data), dtype=dtype).B.cat.categories
+
+The values have been correctly interpreted as integers.
+
+The ``.dtype`` property of a ``Categorical``, ``CategoricalIndex`` or a
+``Series`` with categorical type will now return an instance of
+``CategoricalDtype``. While the repr has changed, ``str(CategoricalDtype())`` is
+still the string ``'category'``. We'll take this moment to remind users that the
+*preferred* way to detect categorical data is to use
+:func:`pandas.api.types.is_categorical_dtype`, and not ``str(dtype) == 'category'``.
+
+See the :ref:`CategoricalDtype docs <categorical.categoricaldtype>` for more.
+
+.. _whatsnew_0210.enhancements.GroupBy_pipe:
+
+``GroupBy`` objects now have a ``pipe`` method
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``GroupBy`` objects now have a ``pipe`` method, similar to the one on
+``DataFrame`` and ``Series``, that allow for functions that take a
+``GroupBy`` to be composed in a clean, readable syntax. (:issue:`17871`)
+
+For a concrete example on combining ``.groupby`` and ``.pipe`` , imagine having a
+DataFrame with columns for stores, products, revenue and sold quantity. We'd like to
+do a groupwise calculation of *prices* (i.e. revenue/quantity) per store and per product.
+We could do this in a multi-step operation, but expressing it in terms of piping can make the
+code more readable.
+
+First we set the data:
+
+.. ipython:: python
+
+   import numpy as np
+   n = 1000
+   df = pd.DataFrame({'Store': np.random.choice(['Store_1', 'Store_2'], n),
+                      'Product': np.random.choice(['Product_1',
+                                                   'Product_2',
+                                                   'Product_3'
+                                                   ], n),
+                      'Revenue': (np.random.random(n) * 50 + 10).round(2),
+                      'Quantity': np.random.randint(1, 10, size=n)})
+   df.head(2)
+
+Now, to find prices per store/product, we can simply do:
+
+.. ipython:: python
+
+   (df.groupby(['Store', 'Product'])
+      .pipe(lambda grp: grp.Revenue.sum() / grp.Quantity.sum())
+      .unstack().round(2))
+
+See the :ref:`documentation <groupby.pipe>` for more.
+
+
+.. _whatsnew_0210.enhancements.rename_categories:
+
+``Categorical.rename_categories`` accepts a dict-like
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`~Series.cat.rename_categories` now accepts a dict-like argument for
+``new_categories``. The previous categories are looked up in the dictionary's
+keys and replaced if found. The behavior of missing and extra keys is the same
+as in :meth:`DataFrame.rename`.
+
+.. ipython:: python
+
+   c = pd.Categorical(['a', 'a', 'b'])
+   c.rename_categories({"a": "eh", "b": "bee"})
+
+.. warning::
+
+    To assist with upgrading pandas, ``rename_categories`` treats ``Series`` as
+    list-like. Typically, Series are considered to be dict-like (e.g. in
+    ``.rename``, ``.map``). In a future version of pandas ``rename_categories``
+    will change to treat them as dict-like. Follow the warning message's
+    recommendations for writing future-proof code.
+
+    .. code-block:: ipython
+
+        In [33]: c.rename_categories(pd.Series([0, 1], index=['a', 'c']))
+        FutureWarning: Treating Series 'new_categories' as a list-like and using the values.
+        In a future version, 'rename_categories' will treat Series like a dictionary.
+        For dict-like, use 'new_categories.to_dict()'
+        For list-like, use 'new_categories.values'.
+        Out[33]:
+        [0, 0, 1]
+        Categories (2, int64): [0, 1]
+
+
+.. _whatsnew_0210.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+New functions or methods
+""""""""""""""""""""""""
+
+- :meth:`.Resampler.nearest` is added to support nearest-neighbor upsampling (:issue:`17496`).
+- :class:`~pandas.Index` has added support for a ``to_frame`` method (:issue:`15230`).
+
+New keywords
+""""""""""""
+
+- Added a ``skipna`` parameter to :func:`~pandas.api.types.infer_dtype` to
+  support type inference in the presence of missing values (:issue:`17059`).
+- :func:`Series.to_dict` and :func:`DataFrame.to_dict` now support an ``into`` keyword which allows you to specify the ``collections.Mapping`` subclass that you would like returned.  The default is ``dict``, which is backwards compatible. (:issue:`16122`)
+- :func:`Series.set_axis` and :func:`DataFrame.set_axis` now support the ``inplace`` parameter. (:issue:`14636`)
+- :func:`Series.to_pickle` and :func:`DataFrame.to_pickle` have gained a ``protocol`` parameter (:issue:`16252`). By default, this parameter is set to `HIGHEST_PROTOCOL <https://docs.python.org/3/library/pickle.html#data-stream-format>`__
+- :func:`read_feather` has gained the ``nthreads`` parameter for multi-threaded operations (:issue:`16359`)
+- :func:`DataFrame.clip` and :func:`Series.clip` have gained an ``inplace`` argument. (:issue:`15388`)
+- :func:`crosstab` has gained a ``margins_name`` parameter to define the name of the row / column that will contain the totals when ``margins=True``. (:issue:`15972`)
+- :func:`read_json` now accepts a ``chunksize`` parameter that can be used when ``lines=True``. If ``chunksize`` is passed, read_json now returns an iterator which reads in ``chunksize`` lines with each iteration. (:issue:`17048`)
+- :func:`read_json` and :func:`~DataFrame.to_json` now accept a ``compression`` argument which allows them to transparently handle compressed files. (:issue:`17798`)
+
+Various enhancements
+""""""""""""""""""""
+
+- Improved the import time of pandas by about 2.25x.  (:issue:`16764`)
+- Support for `PEP 519 -- Adding a file system path protocol
+  <https://www.python.org/dev/peps/pep-0519/>`_ on most readers (e.g.
+  :func:`read_csv`) and writers (e.g. :meth:`DataFrame.to_csv`) (:issue:`13823`).
+- Added a ``__fspath__`` method to ``pd.HDFStore``, ``pd.ExcelFile``,
+  and ``pd.ExcelWriter`` to work properly with the file system path protocol (:issue:`13823`).
+- The ``validate`` argument for :func:`merge` now checks whether a merge is one-to-one, one-to-many, many-to-one, or many-to-many. If a merge is found to not be an example of specified merge type, an exception of type ``MergeError`` will be raised. For more, see :ref:`here <merging.validation>` (:issue:`16270`)
+- Added support for `PEP 518 <https://www.python.org/dev/peps/pep-0518/>`_ (``pyproject.toml``) to the build system (:issue:`16745`)
+- :func:`RangeIndex.append` now returns a ``RangeIndex`` object when possible (:issue:`16212`)
+- :func:`Series.rename_axis` and :func:`DataFrame.rename_axis` with ``inplace=True`` now return ``None`` while renaming the axis inplace. (:issue:`15704`)
+- :func:`api.types.infer_dtype` now infers decimals. (:issue:`15690`)
+- :func:`DataFrame.select_dtypes` now accepts scalar values for include/exclude as well as list-like. (:issue:`16855`)
+- :func:`date_range` now accepts 'YS' in addition to 'AS' as an alias for start of year. (:issue:`9313`)
+- :func:`date_range` now accepts 'Y' in addition to 'A' as an alias for end of year. (:issue:`9313`)
+- :func:`DataFrame.add_prefix` and :func:`DataFrame.add_suffix` now accept strings containing the '%' character. (:issue:`17151`)
+- Read/write methods that infer compression (:func:`read_csv`, :func:`read_table`, :func:`read_pickle`, and :meth:`~DataFrame.to_pickle`) can now infer from path-like objects, such as ``pathlib.Path``. (:issue:`17206`)
+- :func:`read_sas` now recognizes much more of the most frequently used date (datetime) formats in SAS7BDAT files. (:issue:`15871`)
+- :func:`DataFrame.items` and :func:`Series.items` are now present in both Python 2 and 3 and is lazy in all cases. (:issue:`13918`, :issue:`17213`)
+- :meth:`pandas.io.formats.style.Styler.where` has been implemented as a convenience for :meth:`pandas.io.formats.style.Styler.applymap`. (:issue:`17474`)
+- :func:`MultiIndex.is_monotonic_decreasing` has been implemented.  Previously returned ``False`` in all cases. (:issue:`16554`)
+- :func:`read_excel` raises ``ImportError`` with a better message if ``xlrd`` is not installed. (:issue:`17613`)
+- :meth:`DataFrame.assign` will preserve the original order of ``**kwargs`` for Python 3.6+ users instead of sorting the column names. (:issue:`14207`)
+- :func:`Series.reindex`, :func:`DataFrame.reindex`, :func:`Index.get_indexer` now support list-like argument for ``tolerance``. (:issue:`17367`)
+
+.. _whatsnew_0210.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0210.api_breaking.deps:
+
+Dependencies have increased minimum versions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We have updated our minimum supported versions of dependencies (:issue:`15206`, :issue:`15543`, :issue:`15214`).
+If installed, we now require:
+
+   +--------------+-----------------+----------+
+   | Package      | Minimum Version | Required |
+   +==============+=================+==========+
+   | Numpy        | 1.9.0           |    X     |
+   +--------------+-----------------+----------+
+   | Matplotlib   | 1.4.3           |          |
+   +--------------+-----------------+----------+
+   | Scipy        | 0.14.0          |          |
+   +--------------+-----------------+----------+
+   | Bottleneck   | 1.0.0           |          |
+   +--------------+-----------------+----------+
+
+Additionally, support has been dropped for Python 3.4 (:issue:`15251`).
+
+
+.. _whatsnew_0210.api_breaking.bottleneck:
+
+Sum/prod of all-NaN or empty Series/DataFrames is now consistently NaN
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. note::
+
+   The changes described here have been partially reverted. See
+   the :ref:`v0.22.0 Whatsnew <whatsnew_0220>` for more.
+
+
+The behavior of ``sum`` and ``prod`` on all-NaN Series/DataFrames no longer depends on
+whether `bottleneck <https://bottleneck.readthedocs.io>`__ is installed, and return value of ``sum`` and ``prod`` on an empty Series has changed (:issue:`9422`, :issue:`15507`).
+
+Calling ``sum`` or ``prod`` on an empty or all-``NaN`` ``Series``, or columns of a ``DataFrame``, will result in ``NaN``. See the :ref:`docs <missing_data.calculations>`.
+
+.. ipython:: python
+
+   s = pd.Series([np.nan])
+
+Previously WITHOUT ``bottleneck`` installed:
+
+.. code-block:: ipython
+
+   In [2]: s.sum()
+   Out[2]: np.nan
+
+Previously WITH ``bottleneck``:
+
+.. code-block:: ipython
+
+   In [2]: s.sum()
+   Out[2]: 0.0
+
+New behavior, without regard to the bottleneck installation:
+
+.. ipython:: python
+
+   s.sum()
+
+Note that this also changes the sum of an empty ``Series``. Previously this always returned 0 regardless of a ``bottleneck`` installation:
+
+.. code-block:: ipython
+
+   In [1]: pd.Series([]).sum()
+   Out[1]: 0
+
+but for consistency with the all-NaN case, this was changed to return 0 as well:
+
+.. code-block:: ipython
+
+   In [2]: pd.Series([]).sum()
+   Out[2]: 0
+
+.. _whatsnew_0210.api_breaking.loc:
+
+Indexing with a list with missing labels is deprecated
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, selecting with a list of labels, where one or more labels were missing would always succeed, returning ``NaN`` for missing labels.
+This will now show a ``FutureWarning``. In the future this will raise a ``KeyError`` (:issue:`15747`).
+This warning will trigger on a ``DataFrame`` or a ``Series`` for using ``.loc[]``  or ``[[]]`` when passing a list-of-labels with at least 1 missing label.
+
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3])
+   s
+
+Previous behavior
+
+.. code-block:: ipython
+
+   In [4]: s.loc[[1, 2, 3]]
+   Out[4]:
+   1    2.0
+   2    3.0
+   3    NaN
+   dtype: float64
+
+
+Current behavior
+
+.. code-block:: ipython
+
+   In [4]: s.loc[[1, 2, 3]]
+   Passing list-likes to .loc or [] with any missing label will raise
+   KeyError in the future, you can use .reindex() as an alternative.
+
+   See the documentation here:
+   https://pandas.pydata.org/pandas-docs/stable/indexing.html#deprecate-loc-reindex-listlike
+
+   Out[4]:
+   1    2.0
+   2    3.0
+   3    NaN
+   dtype: float64
+
+The idiomatic way to achieve selecting potentially not-found elements is via ``.reindex()``
+
+.. ipython:: python
+
+   s.reindex([1, 2, 3])
+
+Selection with all keys found is unchanged.
+
+.. ipython:: python
+
+   s.loc[[1, 2]]
+
+
+.. _whatsnew_0210.api.na_changes:
+
+NA naming changes
+^^^^^^^^^^^^^^^^^
+
+In order to promote more consistency among the pandas API, we have added additional top-level
+functions :func:`isna` and :func:`notna` that are aliases for :func:`isnull` and :func:`notnull`.
+The naming scheme is now more consistent with methods like ``.dropna()`` and ``.fillna()``. Furthermore
+in all cases where ``.isnull()`` and ``.notnull()`` methods are defined, these have additional methods
+named ``.isna()`` and ``.notna()``, these are included for classes ``Categorical``,
+``Index``, ``Series``, and ``DataFrame``. (:issue:`15001`).
+
+The configuration option ``pd.options.mode.use_inf_as_null`` is deprecated, and ``pd.options.mode.use_inf_as_na`` is added as a replacement.
+
+
+.. _whatsnew_0210.api_breaking.iteration_scalars:
+
+Iteration of Series/Index will now return Python scalars
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, when using certain iteration methods for a ``Series`` with dtype ``int`` or ``float``, you would receive a ``numpy`` scalar, e.g. a ``np.int64``, rather than a Python ``int``. Issue (:issue:`10904`) corrected this for ``Series.tolist()`` and ``list(Series)``. This change makes all iteration methods consistent, in particular, for ``__iter__()`` and ``.map()``; note that this only affects int/float dtypes. (:issue:`13236`, :issue:`13258`, :issue:`14216`).
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3])
+   s
+
+Previously:
+
+.. code-block:: ipython
+
+   In [2]: type(list(s)[0])
+   Out[2]: numpy.int64
+
+New behavior:
+
+.. ipython:: python
+
+   type(list(s)[0])
+
+Furthermore this will now correctly box the results of iteration for :func:`DataFrame.to_dict` as well.
+
+.. ipython:: python
+
+   d = {'a': [1], 'b': ['b']}
+   df = pd.DataFrame(d)
+
+Previously:
+
+.. code-block:: ipython
+
+   In [8]: type(df.to_dict()['a'][0])
+   Out[8]: numpy.int64
+
+New behavior:
+
+.. ipython:: python
+
+   type(df.to_dict()['a'][0])
+
+
+.. _whatsnew_0210.api_breaking.loc_with_index:
+
+Indexing with a Boolean Index
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously when passing a boolean ``Index`` to ``.loc``, if the index of the ``Series/DataFrame`` had ``boolean`` labels,
+you would get a label based selection, potentially duplicating result labels, rather than a boolean indexing selection
+(where ``True`` selects elements), this was inconsistent how a boolean numpy array indexed. The new behavior is to
+act like a boolean numpy array indexer. (:issue:`17738`)
+
+Previous behavior:
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=[False, True, False])
+   s
+
+.. code-block:: ipython
+
+   In [59]: s.loc[pd.Index([True, False, True])]
+   Out[59]:
+   True     2
+   False    1
+   False    3
+   True     2
+   dtype: int64
+
+Current behavior
+
+.. ipython:: python
+
+   s.loc[pd.Index([True, False, True])]
+
+
+Furthermore, previously if you had an index that was non-numeric (e.g. strings), then a boolean Index would raise a ``KeyError``.
+This will now be treated as a boolean indexer.
+
+Previously behavior:
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3], index=['a', 'b', 'c'])
+   s
+
+.. code-block:: ipython
+
+   In [39]: s.loc[pd.Index([True, False, True])]
+   KeyError: "None of [Index([True, False, True], dtype='object')] are in the [index]"
+
+Current behavior
+
+.. ipython:: python
+
+   s.loc[pd.Index([True, False, True])]
+
+
+.. _whatsnew_0210.api_breaking.period_index_resampling:
+
+``PeriodIndex`` resampling
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, resampling a ``Series``/``DataFrame`` indexed by a ``PeriodIndex`` returned a ``DatetimeIndex`` in some cases (:issue:`12884`). Resampling to a multiplied frequency now returns a ``PeriodIndex`` (:issue:`15944`). As a minor enhancement, resampling a ``PeriodIndex`` can now handle ``NaT`` values (:issue:`13224`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [1]: pi = pd.period_range('2017-01', periods=12, freq='M')
+
+   In [2]: s = pd.Series(np.arange(12), index=pi)
+
+   In [3]: resampled = s.resample('2Q').mean()
+
+   In [4]: resampled
+   Out[4]:
+   2017-03-31     1.0
+   2017-09-30     5.5
+   2018-03-31    10.0
+   Freq: 2Q-DEC, dtype: float64
+
+   In [5]: resampled.index
+   Out[5]: DatetimeIndex(['2017-03-31', '2017-09-30', '2018-03-31'], dtype='datetime64[ns]', freq='2Q-DEC')
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [1]: pi = pd.period_range('2017-01', periods=12, freq='M')
+
+   In [2]: s = pd.Series(np.arange(12), index=pi)
+
+   In [3]: resampled = s.resample('2Q').mean()
+
+   In [4]: resampled
+   Out[4]:
+   2017Q1    2.5
+   2017Q3    8.5
+   Freq: 2Q-DEC, dtype: float64
+
+   In [5]: resampled.index
+   Out[5]: PeriodIndex(['2017Q1', '2017Q3'], dtype='period[2Q-DEC]')
+
+Upsampling and calling ``.ohlc()`` previously returned a ``Series``, basically identical to calling ``.asfreq()``. OHLC upsampling now returns a DataFrame with columns ``open``, ``high``, ``low`` and ``close`` (:issue:`13083`). This is consistent with downsampling and ``DatetimeIndex`` behavior.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [1]: pi = pd.period_range(start='2000-01-01', freq='D', periods=10)
+
+   In [2]: s = pd.Series(np.arange(10), index=pi)
+
+   In [3]: s.resample('H').ohlc()
+   Out[3]:
+   2000-01-01 00:00    0.0
+                   ...
+   2000-01-10 23:00    NaN
+   Freq: H, Length: 240, dtype: float64
+
+   In [4]: s.resample('M').ohlc()
+   Out[4]:
+            open  high  low  close
+   2000-01     0     9    0      9
+
+New behavior:
+
+.. code-block:: ipython
+
+   In [56]: pi = pd.period_range(start='2000-01-01', freq='D', periods=10)
+
+   In [57]: s = pd.Series(np.arange(10), index=pi)
+
+   In [58]: s.resample('H').ohlc()
+   Out[58]:
+                     open  high  low  close
+   2000-01-01 00:00   0.0   0.0  0.0    0.0
+   2000-01-01 01:00   NaN   NaN  NaN    NaN
+   2000-01-01 02:00   NaN   NaN  NaN    NaN
+   2000-01-01 03:00   NaN   NaN  NaN    NaN
+   2000-01-01 04:00   NaN   NaN  NaN    NaN
+   ...                ...   ...  ...    ...
+   2000-01-10 19:00   NaN   NaN  NaN    NaN
+   2000-01-10 20:00   NaN   NaN  NaN    NaN
+   2000-01-10 21:00   NaN   NaN  NaN    NaN
+   2000-01-10 22:00   NaN   NaN  NaN    NaN
+   2000-01-10 23:00   NaN   NaN  NaN    NaN
+
+   [240 rows x 4 columns]
+
+   In [59]: s.resample('M').ohlc()
+   Out[59]:
+            open  high  low  close
+   2000-01     0     9    0      9
+
+   [1 rows x 4 columns]
+
+
+.. _whatsnew_0210.api_breaking.pandas_eval:
+
+Improved error handling during item assignment in pd.eval
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`eval` will now raise a ``ValueError`` when item assignment malfunctions, or
+inplace operations are specified, but there is no item assignment in the expression (:issue:`16732`)
+
+.. ipython:: python
+
+   arr = np.array([1, 2, 3])
+
+Previously, if you attempted the following expression, you would get a not very helpful error message:
+
+.. code-block:: ipython
+
+   In [3]: pd.eval("a = 1 + 2", target=arr, inplace=True)
+   ...
+   IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`)
+   and integer or boolean arrays are valid indices
+
+This is a very long way of saying numpy arrays don't support string-item indexing. With this
+change, the error message is now this:
+
+.. code-block:: python
+
+   In [3]: pd.eval("a = 1 + 2", target=arr, inplace=True)
+   ...
+   ValueError: Cannot assign expression output to target
+
+It also used to be possible to evaluate expressions inplace, even if there was no item assignment:
+
+.. code-block:: ipython
+
+   In [4]: pd.eval("1 + 2", target=arr, inplace=True)
+   Out[4]: 3
+
+However, this input does not make much sense because the output is not being assigned to
+the target. Now, a ``ValueError`` will be raised when such an input is passed in:
+
+.. code-block:: ipython
+
+   In [4]: pd.eval("1 + 2", target=arr, inplace=True)
+   ...
+   ValueError: Cannot operate inplace if there is no assignment
+
+
+.. _whatsnew_0210.api_breaking.dtype_conversions:
+
+Dtype conversions
+^^^^^^^^^^^^^^^^^
+
+Previously assignments, ``.where()`` and ``.fillna()`` with a ``bool`` assignment, would coerce to same the type (e.g. int / float), or raise for datetimelikes. These will now preserve the bools with ``object`` dtypes. (:issue:`16821`).
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, 3])
+
+.. code-block:: python
+
+   In [5]: s[1] = True
+
+   In [6]: s
+   Out[6]:
+   0    1
+   1    1
+   2    3
+   dtype: int64
+
+New behavior
+
+.. code-block:: ipython
+
+   In [7]: s[1] = True
+
+   In [8]: s
+   Out[8]:
+   0       1
+   1    True
+   2       3
+   Length: 3, dtype: object
+
+Previously, as assignment to a datetimelike with a non-datetimelike would coerce the
+non-datetime-like item being assigned (:issue:`14145`).
+
+.. ipython:: python
+
+   s = pd.Series([pd.Timestamp('2011-01-01'), pd.Timestamp('2012-01-01')])
+
+.. code-block:: python
+
+   In [1]: s[1] = 1
+
+   In [2]: s
+   Out[2]:
+   0   2011-01-01 00:00:00.000000000
+   1   1970-01-01 00:00:00.000000001
+   dtype: datetime64[ns]
+
+These now coerce to ``object`` dtype.
+
+.. code-block:: python
+
+   In [1]: s[1] = 1
+
+   In [2]: s
+   Out[2]:
+   0    2011-01-01 00:00:00
+   1                      1
+   dtype: object
+
+- Inconsistent behavior in ``.where()`` with datetimelikes which would raise rather than coerce to ``object`` (:issue:`16402`)
+- Bug in assignment against ``int64`` data with ``np.ndarray`` with ``float64`` dtype may keep ``int64`` dtype (:issue:`14001`)
+
+
+.. _whatsnew_210.api.multiindex_single:
+
+MultiIndex constructor with a single level
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``MultiIndex`` constructors no longer squeezes a MultiIndex with all
+length-one levels down to a regular ``Index``. This affects all the
+``MultiIndex`` constructors. (:issue:`17178`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: pd.MultiIndex.from_tuples([('a',), ('b',)])
+   Out[2]: Index(['a', 'b'], dtype='object')
+
+Length 1 levels are no longer special-cased. They behave exactly as if you had
+length 2+ levels, so a :class:`MultiIndex` is always returned from all of the
+``MultiIndex`` constructors:
+
+.. ipython:: python
+
+   pd.MultiIndex.from_tuples([('a',), ('b',)])
+
+.. _whatsnew_0210.api.utc_localization_with_series:
+
+UTC localization with Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, :func:`to_datetime` did not localize datetime ``Series`` data when ``utc=True`` was passed. Now, :func:`to_datetime` will correctly localize ``Series`` with a ``datetime64[ns, UTC]`` dtype to be consistent with how list-like and ``Index`` data are handled. (:issue:`6415`).
+
+Previous behavior
+
+.. ipython:: python
+
+   s = pd.Series(['20130101 00:00:00'] * 3)
+
+.. code-block:: ipython
+
+   In [12]: pd.to_datetime(s, utc=True)
+   Out[12]:
+   0   2013-01-01
+   1   2013-01-01
+   2   2013-01-01
+   dtype: datetime64[ns]
+
+New behavior
+
+.. ipython:: python
+
+   pd.to_datetime(s, utc=True)
+
+Additionally, DataFrames with datetime columns that were parsed by :func:`read_sql_table` and :func:`read_sql_query` will also be localized to UTC only if the original SQL columns were timezone aware datetime columns.
+
+.. _whatsnew_0210.api.consistency_of_range_functions:
+
+Consistency of range functions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions, there were some inconsistencies between the various range functions: :func:`date_range`, :func:`bdate_range`, :func:`period_range`, :func:`timedelta_range`, and :func:`interval_range`. (:issue:`17471`).
+
+One of the inconsistent behaviors occurred when the ``start``, ``end`` and ``period`` parameters were all specified, potentially leading to ambiguous ranges.  When all three parameters were passed, ``interval_range`` ignored the ``period`` parameter, ``period_range`` ignored the ``end`` parameter, and the other range functions raised.  To promote consistency among the range functions, and avoid potentially ambiguous ranges, ``interval_range`` and ``period_range`` will now raise when all three parameters are passed.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [2]: pd.interval_range(start=0, end=4, periods=6)
+   Out[2]:
+   IntervalIndex([(0, 1], (1, 2], (2, 3]]
+                 closed='right',
+                 dtype='interval[int64]')
+
+  In [3]: pd.period_range(start='2017Q1', end='2017Q4', periods=6, freq='Q')
+  Out[3]: PeriodIndex(['2017Q1', '2017Q2', '2017Q3', '2017Q4', '2018Q1', '2018Q2'], dtype='period[Q-DEC]', freq='Q-DEC')
+
+New behavior:
+
+.. code-block:: ipython
+
+  In [2]: pd.interval_range(start=0, end=4, periods=6)
+  ---------------------------------------------------------------------------
+  ValueError: Of the three parameters: start, end, and periods, exactly two must be specified
+
+  In [3]: pd.period_range(start='2017Q1', end='2017Q4', periods=6, freq='Q')
+  ---------------------------------------------------------------------------
+  ValueError: Of the three parameters: start, end, and periods, exactly two must be specified
+
+Additionally, the endpoint parameter ``end`` was not included in the intervals produced by ``interval_range``.  However, all other range functions include ``end`` in their output.  To promote consistency among the range functions, ``interval_range`` will now include ``end`` as the right endpoint of the final interval, except if ``freq`` is specified in a way which skips ``end``.
+
+Previous behavior:
+
+.. code-block:: ipython
+
+   In [4]: pd.interval_range(start=0, end=4)
+   Out[4]:
+   IntervalIndex([(0, 1], (1, 2], (2, 3]]
+                 closed='right',
+                 dtype='interval[int64]')
+
+
+New behavior:
+
+.. ipython:: python
+
+   pd.interval_range(start=0, end=4)
+
+.. _whatsnew_0210.api.mpl_converters:
+
+No automatic Matplotlib converters
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas no longer registers our ``date``, ``time``, ``datetime``,
+``datetime64``, and ``Period`` converters with matplotlib when pandas is
+imported. Matplotlib plot methods (``plt.plot``, ``ax.plot``, ...), will not
+nicely format the x-axis for ``DatetimeIndex`` or ``PeriodIndex`` values. You
+must explicitly register these methods:
+
+pandas built-in ``Series.plot`` and ``DataFrame.plot`` *will* register these
+converters on first-use (:issue:`17710`).
+
+.. note::
+
+  This change has been temporarily reverted in pandas 0.21.1,
+  for more details see :ref:`here <whatsnew_0211.converters>`.
+
+.. _whatsnew_0210.api:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- The Categorical constructor no longer accepts a scalar for the ``categories`` keyword. (:issue:`16022`)
+- Accessing a non-existent attribute on a closed :class:`~pandas.HDFStore` will now
+  raise an ``AttributeError`` rather than a ``ClosedFileError`` (:issue:`16301`)
+- :func:`read_csv` now issues a ``UserWarning`` if the ``names`` parameter contains duplicates (:issue:`17095`)
+- :func:`read_csv` now treats ``'null'`` and ``'n/a'`` strings as missing values by default (:issue:`16471`, :issue:`16078`)
+- :class:`pandas.HDFStore`'s string representation is now faster and less detailed. For the previous behavior, use ``pandas.HDFStore.info()``. (:issue:`16503`).
+- Compression defaults in HDF stores now follow pytables standards. Default is no compression and if ``complib`` is missing and ``complevel`` > 0 ``zlib`` is used (:issue:`15943`)
+- ``Index.get_indexer_non_unique()`` now returns a ndarray indexer rather than an ``Index``; this is consistent with ``Index.get_indexer()`` (:issue:`16819`)
+- Removed the ``@slow`` decorator from ``pandas._testing``, which caused issues for some downstream packages' test suites. Use ``@pytest.mark.slow`` instead, which achieves the same thing (:issue:`16850`)
+- Moved definition of ``MergeError`` to the ``pandas.errors`` module.
+- The signature of :func:`Series.set_axis` and :func:`DataFrame.set_axis` has been changed from ``set_axis(axis, labels)`` to ``set_axis(labels, axis=0)``, for consistency with the rest of the API. The old signature is deprecated and will show a ``FutureWarning`` (:issue:`14636`)
+- :func:`Series.argmin` and :func:`Series.argmax` will now raise a ``TypeError`` when used with ``object`` dtypes, instead of a ``ValueError`` (:issue:`13595`)
+- :class:`Period` is now immutable, and will now raise an ``AttributeError`` when a user tries to assign a new value to the ``ordinal`` or ``freq`` attributes (:issue:`17116`).
+- :func:`to_datetime` when passed a tz-aware ``origin=`` kwarg will now raise a more informative ``ValueError`` rather than a ``TypeError`` (:issue:`16842`)
+- :func:`to_datetime` now raises a ``ValueError`` when format includes ``%W`` or ``%U`` without also including day of the week and calendar year (:issue:`16774`)
+- Renamed non-functional ``index`` to ``index_col`` in :func:`read_stata` to improve API consistency (:issue:`16342`)
+- Bug in :func:`DataFrame.drop` caused boolean labels ``False`` and ``True`` to be treated as labels 0 and 1 respectively when dropping indices from a numeric index. This will now raise a ValueError (:issue:`16877`)
+- Restricted DateOffset keyword arguments.  Previously, ``DateOffset`` subclasses allowed arbitrary keyword arguments which could lead to unexpected behavior.  Now, only valid arguments will be accepted. (:issue:`17176`).
+
+.. _whatsnew_0210.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- :meth:`DataFrame.from_csv` and :meth:`Series.from_csv` have been deprecated in favor of :func:`read_csv` (:issue:`4191`)
+- :func:`read_excel` has deprecated ``sheetname`` in favor of ``sheet_name`` for consistency with ``.to_excel()`` (:issue:`10559`).
+- :func:`read_excel` has deprecated ``parse_cols`` in favor of ``usecols`` for consistency with :func:`read_csv` (:issue:`4988`)
+- :func:`read_csv` has deprecated the ``tupleize_cols`` argument. Column tuples will always be converted to a ``MultiIndex`` (:issue:`17060`)
+- :meth:`DataFrame.to_csv` has deprecated the ``tupleize_cols`` argument. MultiIndex columns will be always written as rows in the CSV file (:issue:`17060`)
+- The ``convert`` parameter has been deprecated in the ``.take()`` method, as it was not being respected (:issue:`16948`)
+- ``pd.options.html.border`` has been deprecated in favor of ``pd.options.display.html.border`` (:issue:`15793`).
+- :func:`SeriesGroupBy.nth` has deprecated ``True`` in favor of ``'all'`` for its kwarg ``dropna`` (:issue:`11038`).
+- :func:`DataFrame.as_blocks` is deprecated, as this is exposing the internal implementation (:issue:`17302`)
+- ``pd.TimeGrouper`` is deprecated in favor of :class:`pandas.Grouper` (:issue:`16747`)
+- ``cdate_range`` has been deprecated in favor of :func:`bdate_range`, which has gained ``weekmask`` and ``holidays`` parameters for building custom frequency date ranges. See the :ref:`documentation <timeseries.custom-freq-ranges>` for more details (:issue:`17596`)
+- passing ``categories`` or ``ordered`` kwargs to :func:`Series.astype` is deprecated, in favor of passing a :ref:`CategoricalDtype <whatsnew_0210.enhancements.categorical_dtype>` (:issue:`17636`)
+- ``.get_value`` and ``.set_value`` on ``Series``, ``DataFrame``, ``Panel``, ``SparseSeries``, and ``SparseDataFrame`` are deprecated in favor of using ``.iat[]`` or ``.at[]`` accessors (:issue:`15269`)
+- Passing a non-existent column in ``.to_excel(..., columns=)`` is deprecated and will raise a ``KeyError`` in the future (:issue:`17295`)
+- ``raise_on_error`` parameter to :func:`Series.where`, :func:`Series.mask`, :func:`DataFrame.where`, :func:`DataFrame.mask` is deprecated, in favor of ``errors=`` (:issue:`14968`)
+- Using :meth:`DataFrame.rename_axis` and :meth:`Series.rename_axis` to alter index or column *labels* is now deprecated in favor of using ``.rename``. ``rename_axis`` may still be used to alter the name of the index or columns (:issue:`17833`).
+- :meth:`~DataFrame.reindex_axis` has been deprecated in favor of :meth:`~DataFrame.reindex`. See :ref:`here <whatsnew_0210.enhancements.rename_reindex_axis>` for more (:issue:`17833`).
+
+.. _whatsnew_0210.deprecations.select:
+
+Series.select and DataFrame.select
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`Series.select` and :meth:`DataFrame.select` methods are deprecated in favor of using ``df.loc[labels.map(crit)]`` (:issue:`12401`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': [1, 2, 3]}, index=['foo', 'bar', 'baz'])
+
+.. code-block:: ipython
+
+   In [3]: df.select(lambda x: x in ['bar', 'baz'])
+   FutureWarning: select is deprecated and will be removed in a future release. You can use .loc[crit] as a replacement
+   Out[3]:
+        A
+   bar  2
+   baz  3
+
+.. ipython:: python
+
+   df.loc[df.index.map(lambda x: x in ['bar', 'baz'])]
+
+
+.. _whatsnew_0210.deprecations.argmin_min:
+
+Series.argmax and Series.argmin
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The behavior of :func:`Series.argmax` and :func:`Series.argmin` have been deprecated in favor of :func:`Series.idxmax` and :func:`Series.idxmin`, respectively (:issue:`16830`).
+
+For compatibility with NumPy arrays, ``pd.Series`` implements ``argmax`` and
+``argmin``. Since pandas 0.13.0, ``argmax`` has been an alias for
+:meth:`pandas.Series.idxmax`, and ``argmin`` has been an alias for
+:meth:`pandas.Series.idxmin`. They return the *label* of the maximum or minimum,
+rather than the *position*.
+
+We've deprecated the current behavior of ``Series.argmax`` and
+``Series.argmin``. Using either of these will emit a ``FutureWarning``. Use
+:meth:`Series.idxmax` if you want the label of the maximum. Use
+``Series.values.argmax()`` if you want the position of the maximum. Likewise for
+the minimum. In a future release ``Series.argmax`` and ``Series.argmin`` will
+return the position of the maximum or minimum.
+
+.. _whatsnew_0210.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- :func:`read_excel` has dropped the ``has_index_names`` parameter (:issue:`10967`)
+- The ``pd.options.display.height`` configuration has been dropped (:issue:`3663`)
+- The ``pd.options.display.line_width`` configuration has been dropped (:issue:`2881`)
+- The ``pd.options.display.mpl_style`` configuration has been dropped (:issue:`12190`)
+- ``Index`` has dropped the ``.sym_diff()`` method in favor of ``.symmetric_difference()`` (:issue:`12591`)
+- ``Categorical`` has dropped the ``.order()`` and ``.sort()`` methods in favor of ``.sort_values()`` (:issue:`12882`)
+- :func:`eval` and :func:`DataFrame.eval` have changed the default of ``inplace`` from ``None`` to ``False`` (:issue:`11149`)
+- The function ``get_offset_name`` has been dropped in favor of the ``.freqstr`` attribute for an offset (:issue:`11834`)
+- pandas no longer tests for compatibility with hdf5-files created with pandas < 0.11 (:issue:`17404`).
+
+
+
+.. _whatsnew_0210.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of instantiating :class:`SparseDataFrame` (:issue:`16773`)
+- :attr:`Series.dt` no longer performs frequency inference, yielding a large speedup when accessing the attribute (:issue:`17210`)
+- Improved performance of :meth:`~Series.cat.set_categories` by not materializing the values (:issue:`17508`)
+- :attr:`Timestamp.microsecond` no longer re-computes on attribute access (:issue:`17331`)
+- Improved performance of the :class:`CategoricalIndex` for data that is already categorical dtype (:issue:`17513`)
+- Improved performance of :meth:`RangeIndex.min` and :meth:`RangeIndex.max` by using ``RangeIndex`` properties to perform the computations (:issue:`17607`)
+
+.. _whatsnew_0210.docs:
+
+Documentation changes
+~~~~~~~~~~~~~~~~~~~~~
+
+- Several ``NaT`` method docstrings (e.g. :func:`NaT.ctime`) were incorrect (:issue:`17327`)
+- The documentation has had references to versions < v0.17 removed and cleaned up (:issue:`17442`, :issue:`17442`, :issue:`17404` & :issue:`17504`)
+
+.. _whatsnew_0210.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Conversion
+^^^^^^^^^^
+
+- Bug in assignment against datetime-like data with ``int`` may incorrectly convert to datetime-like (:issue:`14145`)
+- Bug in assignment against ``int64`` data with ``np.ndarray`` with ``float64`` dtype may keep ``int64`` dtype (:issue:`14001`)
+- Fixed the return type of ``IntervalIndex.is_non_overlapping_monotonic`` to be a Python ``bool`` for consistency with similar attributes/methods.  Previously returned a ``numpy.bool_``. (:issue:`17237`)
+- Bug in ``IntervalIndex.is_non_overlapping_monotonic`` when intervals are closed on both sides and overlap at a point (:issue:`16560`)
+- Bug in :func:`Series.fillna` returns frame when ``inplace=True`` and ``value`` is dict (:issue:`16156`)
+- Bug in :attr:`Timestamp.weekday_name` returning a UTC-based weekday name when localized to a timezone (:issue:`17354`)
+- Bug in ``Timestamp.replace`` when replacing ``tzinfo`` around DST changes (:issue:`15683`)
+- Bug in ``Timedelta`` construction and arithmetic that would not propagate the ``Overflow`` exception (:issue:`17367`)
+- Bug in :meth:`~DataFrame.astype` converting to object dtype when passed extension type classes (``DatetimeTZDtype``, ``CategoricalDtype``) rather than instances. Now a ``TypeError`` is raised when a class is passed (:issue:`17780`).
+- Bug in :meth:`to_numeric` in which elements were not always being coerced to numeric when ``errors='coerce'`` (:issue:`17007`, :issue:`17125`)
+- Bug in ``DataFrame`` and ``Series`` constructors where ``range`` objects are converted to ``int32`` dtype on Windows instead of ``int64`` (:issue:`16804`)
+
+Indexing
+^^^^^^^^
+
+- When called with a null slice (e.g. ``df.iloc[:]``), the ``.iloc`` and ``.loc`` indexers return a shallow copy of the original object. Previously they returned the original object. (:issue:`13873`).
+- When called on an unsorted ``MultiIndex``, the ``loc`` indexer now will raise ``UnsortedIndexError`` only if proper slicing is used on non-sorted levels (:issue:`16734`).
+- Fixes regression in 0.20.3 when indexing with a string on a ``TimedeltaIndex`` (:issue:`16896`).
+- Fixed :func:`TimedeltaIndex.get_loc` handling of ``np.timedelta64`` inputs (:issue:`16909`).
+- Fix :func:`MultiIndex.sort_index` ordering when ``ascending`` argument is a list, but not all levels are specified, or are in a different order (:issue:`16934`).
+- Fixes bug where indexing with ``np.inf`` caused an ``OverflowError`` to be raised (:issue:`16957`)
+- Bug in reindexing on an empty ``CategoricalIndex`` (:issue:`16770`)
+- Fixes ``DataFrame.loc`` for setting with alignment and tz-aware ``DatetimeIndex`` (:issue:`16889`)
+- Avoids ``IndexError`` when passing an Index or Series to ``.iloc`` with older numpy (:issue:`17193`)
+- Allow unicode empty strings as placeholders in multilevel columns in Python 2 (:issue:`17099`)
+- Bug in ``.iloc`` when used with inplace addition or assignment and an int indexer on a ``MultiIndex`` causing the wrong indexes to be read from and written to (:issue:`17148`)
+- Bug in ``.isin()`` in which checking membership in empty ``Series`` objects raised an error (:issue:`16991`)
+- Bug in ``CategoricalIndex`` reindexing in which specified indices containing duplicates were not being respected (:issue:`17323`)
+- Bug in intersection of ``RangeIndex`` with negative step (:issue:`17296`)
+- Bug in ``IntervalIndex`` where performing a scalar lookup fails for included right endpoints of non-overlapping monotonic decreasing indexes (:issue:`16417`, :issue:`17271`)
+- Bug in :meth:`DataFrame.first_valid_index` and :meth:`DataFrame.last_valid_index` when no valid entry (:issue:`17400`)
+- Bug in :func:`Series.rename` when called with a callable, incorrectly alters the name of the ``Series``, rather than the name of the ``Index``. (:issue:`17407`)
+- Bug in :func:`String.str_get` raises ``IndexError`` instead of inserting NaNs when using a negative index. (:issue:`17704`)
+
+IO
+^^
+
+- Bug in :func:`read_hdf` when reading a timezone aware index from ``fixed`` format HDFStore (:issue:`17618`)
+- Bug in :func:`read_csv` in which columns were not being thoroughly de-duplicated (:issue:`17060`)
+- Bug in :func:`read_csv` in which specified column names were not being thoroughly de-duplicated (:issue:`17095`)
+- Bug in :func:`read_csv` in which non integer values for the header argument generated an unhelpful / unrelated error message (:issue:`16338`)
+- Bug in :func:`read_csv` in which memory management issues in exception handling, under certain conditions, would cause the interpreter to segfault (:issue:`14696`, :issue:`16798`).
+- Bug in :func:`read_csv` when called with ``low_memory=False`` in which a CSV with at least one column > 2GB in size would incorrectly raise a ``MemoryError`` (:issue:`16798`).
+- Bug in :func:`read_csv` when called with a single-element list ``header`` would return a ``DataFrame`` of all NaN values (:issue:`7757`)
+- Bug in :meth:`DataFrame.to_csv` defaulting to 'ascii' encoding in Python 3, instead of 'utf-8' (:issue:`17097`)
+- Bug in :func:`read_stata` where value labels could not be read when using an iterator (:issue:`16923`)
+- Bug in :func:`read_stata` where the index was not set (:issue:`16342`)
+- Bug in :func:`read_html` where import check fails when run in multiple threads (:issue:`16928`)
+- Bug in :func:`read_csv` where automatic delimiter detection caused a ``TypeError`` to be thrown when a bad line was encountered rather than the correct error message (:issue:`13374`)
+- Bug in :meth:`DataFrame.to_html` with ``notebook=True`` where DataFrames with named indices or non-MultiIndex indices had undesired horizontal or vertical alignment for column or row labels, respectively (:issue:`16792`)
+- Bug in :meth:`DataFrame.to_html` in which there was no validation of the ``justify`` parameter (:issue:`17527`)
+- Bug in :func:`HDFStore.select` when reading a contiguous mixed-data table featuring VLArray (:issue:`17021`)
+- Bug in :func:`to_json` where several conditions (including objects with unprintable symbols, objects with deep recursion, overlong labels) caused segfaults instead of raising the appropriate exception (:issue:`14256`)
+
+Plotting
+^^^^^^^^
+- Bug in plotting methods using ``secondary_y`` and ``fontsize`` not setting secondary axis font size (:issue:`12565`)
+- Bug when plotting ``timedelta`` and ``datetime`` dtypes on y-axis (:issue:`16953`)
+- Line plots no longer assume monotonic x data when calculating xlims, they show the entire lines now even for unsorted x data. (:issue:`11310`, :issue:`11471`)
+- With matplotlib 2.0.0 and above, calculation of x limits for line plots is left to matplotlib, so that its new default settings are applied. (:issue:`15495`)
+- Bug in ``Series.plot.bar`` or ``DataFrame.plot.bar`` with ``y`` not respecting user-passed ``color`` (:issue:`16822`)
+- Bug causing ``plotting.parallel_coordinates`` to reset the random seed when using random colors (:issue:`17525`)
+
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in ``DataFrame.resample(...).size()`` where an empty ``DataFrame`` did not return a ``Series`` (:issue:`14962`)
+- Bug in :func:`infer_freq` causing indices with 2-day gaps during the working week to be wrongly inferred as business daily (:issue:`16624`)
+- Bug in ``.rolling(...).quantile()`` which incorrectly used different defaults than :func:`Series.quantile` and :func:`DataFrame.quantile` (:issue:`9413`, :issue:`16211`)
+- Bug in ``groupby.transform()`` that would coerce boolean dtypes back to float (:issue:`16875`)
+- Bug in ``Series.resample(...).apply()`` where an empty ``Series`` modified the source index and did not return the name of a ``Series`` (:issue:`14313`)
+- Bug in ``.rolling(...).apply(...)`` with a ``DataFrame`` with a ``DatetimeIndex``, a ``window`` of a timedelta-convertible and ``min_periods >= 1`` (:issue:`15305`)
+- Bug in ``DataFrame.groupby`` where index and column keys were not recognized correctly when the number of keys equaled the number of elements on the groupby axis (:issue:`16859`)
+- Bug in ``groupby.nunique()`` with ``TimeGrouper`` which cannot handle ``NaT`` correctly (:issue:`17575`)
+- Bug in ``DataFrame.groupby`` where a single level selection from a ``MultiIndex`` unexpectedly sorts (:issue:`17537`)
+- Bug in ``DataFrame.groupby`` where spurious warning is raised when ``Grouper`` object is used to override ambiguous column name (:issue:`17383`)
+- Bug in ``TimeGrouper`` differs when passes as a list and as a scalar (:issue:`17530`)
+
+Sparse
+^^^^^^
+
+- Bug in ``SparseSeries`` raises ``AttributeError`` when a dictionary is passed in as data (:issue:`16905`)
+- Bug in :func:`SparseDataFrame.fillna` not filling all NaNs when frame was instantiated from SciPy sparse matrix (:issue:`16112`)
+- Bug in :func:`SparseSeries.unstack` and :func:`SparseDataFrame.stack` (:issue:`16614`, :issue:`15045`)
+- Bug in :func:`make_sparse` treating two numeric/boolean data, which have same bits, as same when array ``dtype`` is ``object`` (:issue:`17574`)
+- :func:`SparseArray.all` and :func:`SparseArray.any` are now implemented to handle ``SparseArray``, these were used but not implemented (:issue:`17570`)
+
+Reshaping
+^^^^^^^^^
+- Joining/Merging with a non unique ``PeriodIndex`` raised a ``TypeError`` (:issue:`16871`)
+- Bug in :func:`crosstab` where non-aligned series of integers were casted to float (:issue:`17005`)
+- Bug in merging with categorical dtypes with datetimelikes incorrectly raised a ``TypeError`` (:issue:`16900`)
+- Bug when using :func:`isin` on a large object series and large comparison array (:issue:`16012`)
+- Fixes regression from 0.20, :func:`Series.aggregate` and :func:`DataFrame.aggregate` allow dictionaries as return values again (:issue:`16741`)
+- Fixes dtype of result with integer dtype input, from :func:`pivot_table` when called with ``margins=True`` (:issue:`17013`)
+- Bug in :func:`crosstab` where passing two ``Series`` with the same name raised a ``KeyError`` (:issue:`13279`)
+- :func:`Series.argmin`, :func:`Series.argmax`, and their counterparts on ``DataFrame`` and groupby objects work correctly with floating point data that contains infinite values (:issue:`13595`).
+- Bug in :func:`unique` where checking a tuple of strings raised a ``TypeError`` (:issue:`17108`)
+- Bug in :func:`concat` where order of result index was unpredictable if it contained non-comparable elements (:issue:`17344`)
+- Fixes regression when sorting by multiple columns on a ``datetime64`` dtype ``Series`` with ``NaT`` values (:issue:`16836`)
+- Bug in :func:`pivot_table` where the result's columns did not preserve the categorical dtype of ``columns`` when ``dropna`` was ``False`` (:issue:`17842`)
+- Bug in ``DataFrame.drop_duplicates`` where dropping with non-unique column names raised a ``ValueError`` (:issue:`17836`)
+- Bug in :func:`unstack` which, when called on a list of levels, would discard the ``fillna`` argument (:issue:`13971`)
+- Bug in the alignment of ``range`` objects and other list-likes with ``DataFrame`` leading to operations being performed row-wise instead of column-wise (:issue:`17901`)
+
+Numeric
+^^^^^^^
+- Bug in ``.clip()`` with ``axis=1`` and a list-like for ``threshold`` is passed; previously this raised ``ValueError`` (:issue:`15390`)
+- :func:`Series.clip` and :func:`DataFrame.clip` now treat NA values for upper and lower arguments as ``None`` instead of raising ``ValueError`` (:issue:`17276`).
+
+
+Categorical
+^^^^^^^^^^^
+- Bug in :func:`Series.isin` when called with a categorical (:issue:`16639`)
+- Bug in the categorical constructor with empty values and categories causing the ``.categories`` to be an empty ``Float64Index`` rather than an empty ``Index`` with object dtype (:issue:`17248`)
+- Bug in categorical operations with :ref:`Series.cat <categorical.cat>` not preserving the original Series' name (:issue:`17509`)
+- Bug in :func:`DataFrame.merge` failing for categorical columns with boolean/int data types (:issue:`17187`)
+- Bug in constructing a ``Categorical``/``CategoricalDtype`` when the specified ``categories`` are of categorical type (:issue:`17884`).
+
+.. _whatsnew_0210.pypy:
+
+PyPy
+^^^^
+
+- Compatibility with PyPy in :func:`read_csv` with ``usecols=[<unsorted ints>]`` and
+  :func:`read_json` (:issue:`17351`)
+- Split tests into cases for CPython and PyPy where needed, which highlights the fragility
+  of index matching with ``float('nan')``, ``np.nan`` and ``NAT`` (:issue:`17351`)
+- Fix :func:`DataFrame.memory_usage` to support PyPy. Objects on PyPy do not have a fixed size,
+  so an approximation is used instead (:issue:`17228`)
+
+Other
+^^^^^
+- Bug where some inplace operators were not being wrapped and produced a copy when invoked (:issue:`12962`)
+- Bug in :func:`eval` where the ``inplace`` parameter was being incorrectly handled (:issue:`16732`)
+
+
+
+.. _whatsnew_0.21.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.20.3..v0.21.0
diff --git a/pandas/doc/source/whatsnew/v0.21.1.rst b/pandas/doc/source/whatsnew/v0.21.1.rst
new file mode 100644
index 0000000000000..a8f9f9c9e0840
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.21.1.rst
@@ -0,0 +1,187 @@
+.. _whatsnew_0211:
+
+Version 0.21.1 (December 12, 2017)
+----------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a minor bug-fix release in the 0.21.x series and includes some small regression fixes,
+bug fixes and performance improvements.
+We recommend that all users upgrade to this version.
+
+Highlights include:
+
+- Temporarily restore matplotlib datetime plotting functionality. This should
+  resolve issues for users who implicitly relied on pandas to plot datetimes
+  with matplotlib. See :ref:`here <whatsnew_0211.converters>`.
+- Improvements to the Parquet IO functions introduced in 0.21.0. See
+  :ref:`here <whatsnew_0211.enhancements.parquet>`.
+
+
+.. contents:: What's new in v0.21.1
+    :local:
+    :backlinks: none
+
+
+.. _whatsnew_0211.converters:
+
+Restore Matplotlib datetime converter registration
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas implements some matplotlib converters for nicely formatting the axis
+labels on plots with ``datetime`` or ``Period`` values. Prior to pandas 0.21.0,
+these were implicitly registered with matplotlib, as a side effect of ``import
+pandas``.
+
+In pandas 0.21.0, we required users to explicitly register the
+converter. This caused problems for some users who relied on those converters
+being present for regular ``matplotlib.pyplot`` plotting methods, so we're
+temporarily reverting that change; pandas 0.21.1 again registers the converters on
+import, just like before 0.21.0.
+
+We've added a new option to control the converters:
+``pd.options.plotting.matplotlib.register_converters``. By default, they are
+registered. Toggling this to ``False`` removes pandas' formatters and restore
+any converters we overwrote when registering them (:issue:`18301`).
+
+We're working with the matplotlib developers to make this easier. We're trying
+to balance user convenience (automatically registering the converters) with
+import performance and best practices (importing pandas shouldn't have the side
+effect of overwriting any custom converters you've already set). In the future
+we hope to have most of the datetime formatting functionality in matplotlib,
+with just the pandas-specific converters in pandas. We'll then gracefully
+deprecate the automatic registration of converters in favor of users explicitly
+registering them when they want them.
+
+.. _whatsnew_0211.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0211.enhancements.parquet:
+
+Improvements to the Parquet IO functionality
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- :func:`DataFrame.to_parquet` will now write non-default indexes when the
+  underlying engine supports it. The indexes will be preserved when reading
+  back in with :func:`read_parquet` (:issue:`18581`).
+- :func:`read_parquet` now allows to specify the columns to read from a parquet file (:issue:`18154`)
+- :func:`read_parquet` now allows to specify kwargs which are passed to the respective engine (:issue:`18216`)
+
+.. _whatsnew_0211.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- :meth:`Timestamp.timestamp` is now available in Python 2.7. (:issue:`17329`)
+- :class:`Grouper` and :class:`TimeGrouper` now have a friendly repr output (:issue:`18203`).
+
+.. _whatsnew_0211.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- ``pandas.tseries.register`` has been renamed to
+  :func:`pandas.plotting.register_matplotlib_converters` (:issue:`18301`)
+
+.. _whatsnew_0211.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of plotting large series/dataframes (:issue:`18236`).
+
+.. _whatsnew_0211.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Conversion
+^^^^^^^^^^
+
+- Bug in :class:`TimedeltaIndex` subtraction could incorrectly overflow when ``NaT`` is present (:issue:`17791`)
+- Bug in :class:`DatetimeIndex` subtracting datetimelike from DatetimeIndex could fail to overflow (:issue:`18020`)
+- Bug in :meth:`IntervalIndex.copy` when copying and ``IntervalIndex`` with non-default ``closed`` (:issue:`18339`)
+- Bug in :func:`DataFrame.to_dict` where columns of datetime that are tz-aware were not converted to required arrays when used with ``orient='records'``, raising ``TypeError`` (:issue:`18372`)
+- Bug in :class:`DateTimeIndex` and :meth:`date_range` where mismatching tz-aware ``start`` and ``end`` timezones would not raise an err if ``end.tzinfo`` is None (:issue:`18431`)
+- Bug in :meth:`Series.fillna` which raised when passed a long integer on Python 2 (:issue:`18159`).
+
+Indexing
+^^^^^^^^
+
+- Bug in a boolean comparison of a ``datetime.datetime`` and a ``datetime64[ns]`` dtype Series (:issue:`17965`)
+- Bug where a ``MultiIndex`` with more than a million records was not raising ``AttributeError`` when trying to access a missing attribute (:issue:`18165`)
+- Bug in :class:`IntervalIndex` constructor when a list of intervals is passed with non-default ``closed`` (:issue:`18334`)
+- Bug in ``Index.putmask`` when an invalid mask passed (:issue:`18368`)
+- Bug in masked assignment of a ``timedelta64[ns]`` dtype ``Series``, incorrectly coerced to float (:issue:`18493`)
+
+IO
+^^
+
+- Bug in :class:`~pandas.io.stata.StataReader` not converting date/time columns with display formatting addressed (:issue:`17990`). Previously columns with display formatting were normally left as ordinal numbers and not converted to datetime objects.
+- Bug in :func:`read_csv` when reading a compressed UTF-16 encoded file (:issue:`18071`)
+- Bug in :func:`read_csv` for handling null values in index columns when specifying ``na_filter=False`` (:issue:`5239`)
+- Bug in :func:`read_csv` when reading numeric category fields with high cardinality (:issue:`18186`)
+- Bug in :meth:`DataFrame.to_csv` when the table had ``MultiIndex`` columns, and a list of strings was passed in for ``header`` (:issue:`5539`)
+- Bug in parsing integer datetime-like columns with specified format in ``read_sql`` (:issue:`17855`).
+- Bug in :meth:`DataFrame.to_msgpack` when serializing data of the ``numpy.bool_`` datatype (:issue:`18390`)
+- Bug in :func:`read_json` not decoding when reading line delimited JSON from S3 (:issue:`17200`)
+- Bug in :func:`pandas.io.json.json_normalize` to avoid modification of ``meta`` (:issue:`18610`)
+- Bug in :func:`to_latex` where repeated MultiIndex values were not printed even though a higher level index differed from the previous row (:issue:`14484`)
+- Bug when reading NaN-only categorical columns in :class:`HDFStore` (:issue:`18413`)
+- Bug in :meth:`DataFrame.to_latex` with ``longtable=True`` where a latex multicolumn always spanned over three columns (:issue:`17959`)
+
+Plotting
+^^^^^^^^
+
+- Bug in ``DataFrame.plot()`` and ``Series.plot()`` with :class:`DatetimeIndex` where a figure generated by them is not picklable in Python 3 (:issue:`18439`)
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in ``DataFrame.resample(...).apply(...)`` when there is a callable that returns different columns (:issue:`15169`)
+- Bug in ``DataFrame.resample(...)`` when there is a time change (DST) and resampling frequency is 12h or higher (:issue:`15549`)
+- Bug in ``pd.DataFrameGroupBy.count()`` when counting over a datetimelike column (:issue:`13393`)
+- Bug in ``rolling.var`` where calculation is inaccurate with a zero-valued array (:issue:`18430`)
+
+Reshaping
+^^^^^^^^^
+
+- Error message in ``pd.merge_asof()`` for key datatype mismatch now includes datatype of left and right key (:issue:`18068`)
+- Bug in ``pd.concat`` when empty and non-empty DataFrames or Series are concatenated (:issue:`18178` :issue:`18187`)
+- Bug in ``DataFrame.filter(...)`` when :class:`unicode` is passed as a condition in Python 2 (:issue:`13101`)
+- Bug when merging empty DataFrames when ``np.seterr(divide='raise')`` is set (:issue:`17776`)
+
+Numeric
+^^^^^^^
+
+- Bug in ``pd.Series.rolling.skew()`` and ``rolling.kurt()`` with all equal values has floating issue (:issue:`18044`)
+
+Categorical
+^^^^^^^^^^^
+
+- Bug in :meth:`DataFrame.astype` where casting to 'category' on an empty ``DataFrame`` causes a segmentation fault (:issue:`18004`)
+- Error messages in the testing module have been improved when items have different ``CategoricalDtype`` (:issue:`18069`)
+- ``CategoricalIndex`` can now correctly take a ``pd.api.types.CategoricalDtype`` as its dtype (:issue:`18116`)
+- Bug in ``Categorical.unique()`` returning read-only ``codes``  array when all categories were ``NaN`` (:issue:`18051`)
+- Bug in ``DataFrame.groupby(axis=1)`` with a ``CategoricalIndex`` (:issue:`18432`)
+
+String
+^^^^^^
+
+- :meth:`Series.str.split` will now propagate ``NaN`` values across all expanded columns instead of ``None`` (:issue:`18450`)
+
+
+.. _whatsnew_0.21.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.21.0..v0.21.1
diff --git a/pandas/doc/source/whatsnew/v0.22.0.rst b/pandas/doc/source/whatsnew/v0.22.0.rst
new file mode 100644
index 0000000000000..8a9227ac37b67
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.22.0.rst
@@ -0,0 +1,283 @@
+.. _whatsnew_0220:
+
+Version 0.22.0 (December 29, 2017)
+----------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import *  # noqa F401, F403
+
+
+This is a major release from 0.21.1 and includes a single, API-breaking change.
+We recommend that all users upgrade to this version after carefully reading the
+release note (singular!).
+
+.. _whatsnew_0220.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas 0.22.0 changes the handling of empty and all-*NA* sums and products. The
+summary is that
+
+* The sum of an empty or all-*NA* ``Series`` is now ``0``
+* The product of an empty or all-*NA* ``Series`` is now ``1``
+* We've added a ``min_count`` parameter to ``.sum()`` and ``.prod()`` controlling
+  the minimum number of valid values for the result to be valid. If fewer than
+  ``min_count`` non-*NA* values are present, the result is *NA*. The default is
+  ``0``. To return ``NaN``, the 0.21 behavior, use ``min_count=1``.
+
+Some background: In pandas 0.21, we fixed a long-standing inconsistency
+in the return value of all-*NA* series depending on whether or not bottleneck
+was installed. See :ref:`whatsnew_0210.api_breaking.bottleneck`. At the same
+time, we changed the sum and prod of an empty ``Series`` to also be ``NaN``.
+
+Based on feedback, we've partially reverted those changes.
+
+Arithmetic operations
+^^^^^^^^^^^^^^^^^^^^^
+
+The default sum for empty or all-*NA* ``Series`` is now ``0``.
+
+*pandas 0.21.x*
+
+.. code-block:: ipython
+
+   In [1]: pd.Series([]).sum()
+   Out[1]: nan
+
+   In [2]: pd.Series([np.nan]).sum()
+   Out[2]: nan
+
+*pandas 0.22.0*
+
+.. ipython:: python
+   :okwarning:
+
+   pd.Series([]).sum()
+   pd.Series([np.nan]).sum()
+
+The default behavior is the same as pandas 0.20.3 with bottleneck installed. It
+also matches the behavior of NumPy's ``np.nansum`` on empty and all-*NA* arrays.
+
+To have the sum of an empty series return ``NaN`` (the default behavior of
+pandas 0.20.3 without bottleneck, or pandas 0.21.x), use the ``min_count``
+keyword.
+
+.. ipython:: python
+   :okwarning:
+
+   pd.Series([]).sum(min_count=1)
+
+Thanks to the ``skipna`` parameter, the ``.sum`` on an all-*NA*
+series is conceptually the same as the ``.sum`` of an empty one with
+``skipna=True`` (the default).
+
+.. ipython:: python
+
+   pd.Series([np.nan]).sum(min_count=1)  # skipna=True by default
+
+The ``min_count`` parameter refers to the minimum number of *non-null* values
+required for a non-NA sum or product.
+
+:meth:`Series.prod` has been updated to behave the same as :meth:`Series.sum`,
+returning ``1`` instead.
+
+.. ipython:: python
+   :okwarning:
+
+   pd.Series([]).prod()
+   pd.Series([np.nan]).prod()
+   pd.Series([]).prod(min_count=1)
+
+These changes affect :meth:`DataFrame.sum` and :meth:`DataFrame.prod` as well.
+Finally, a few less obvious places in pandas are affected by this change.
+
+Grouping by a Categorical
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Grouping by a ``Categorical`` and summing now returns ``0`` instead of
+``NaN`` for categories with no observations. The product now returns ``1``
+instead of ``NaN``.
+
+*pandas 0.21.x*
+
+.. code-block:: ipython
+
+   In [8]: grouper = pd.Categorical(['a', 'a'], categories=['a', 'b'])
+
+   In [9]: pd.Series([1, 2]).groupby(grouper, observed=False).sum()
+   Out[9]:
+   a    3.0
+   b    NaN
+   dtype: float64
+
+*pandas 0.22*
+
+.. ipython:: python
+
+   grouper = pd.Categorical(["a", "a"], categories=["a", "b"])
+   pd.Series([1, 2]).groupby(grouper, observed=False).sum()
+
+To restore the 0.21 behavior of returning ``NaN`` for unobserved groups,
+use ``min_count>=1``.
+
+.. ipython:: python
+
+   pd.Series([1, 2]).groupby(grouper, observed=False).sum(min_count=1)
+
+Resample
+^^^^^^^^
+
+The sum and product of all-*NA* bins has changed from ``NaN`` to ``0`` for
+sum and ``1`` for product.
+
+*pandas 0.21.x*
+
+.. code-block:: ipython
+
+   In [11]: s = pd.Series([1, 1, np.nan, np.nan],
+      ....:               index=pd.date_range('2017', periods=4))
+      ....: s
+   Out[11]:
+   2017-01-01    1.0
+   2017-01-02    1.0
+   2017-01-03    NaN
+   2017-01-04    NaN
+   Freq: D, dtype: float64
+
+   In [12]: s.resample('2d').sum()
+   Out[12]:
+   2017-01-01    2.0
+   2017-01-03    NaN
+   Freq: 2D, dtype: float64
+
+*pandas 0.22.0*
+
+.. code-block:: ipython
+
+   In [11]: s = pd.Series([1, 1, np.nan, np.nan],
+      ....:               index=pd.date_range("2017", periods=4))
+
+   In [12]: s.resample("2d").sum()
+   Out[12]:
+   2017-01-01    2.0
+   2017-01-03    0.0
+   Freq: 2D, Length: 2, dtype: float64
+
+To restore the 0.21 behavior of returning ``NaN``, use ``min_count>=1``.
+
+.. code-block:: ipython
+
+   In [13]: s.resample("2d").sum(min_count=1)
+   Out[13]:
+   2017-01-01    2.0
+   2017-01-03    NaN
+   Freq: 2D, Length: 2, dtype: float64
+
+
+In particular, upsampling and taking the sum or product is affected, as
+upsampling introduces missing values even if the original series was
+entirely valid.
+
+*pandas 0.21.x*
+
+.. code-block:: ipython
+
+   In [14]: idx = pd.DatetimeIndex(['2017-01-01', '2017-01-02'])
+
+   In [15]: pd.Series([1, 2], index=idx).resample('12H').sum()
+   Out[15]:
+   2017-01-01 00:00:00    1.0
+   2017-01-01 12:00:00    NaN
+   2017-01-02 00:00:00    2.0
+   Freq: 12H, dtype: float64
+
+*pandas 0.22.0*
+
+.. code-block:: ipython
+
+   In [14]: idx = pd.DatetimeIndex(["2017-01-01", "2017-01-02"])
+   In [15]: pd.Series([1, 2], index=idx).resample("12H").sum()
+   Out[15]:
+   2017-01-01 00:00:00    1
+   2017-01-01 12:00:00    0
+   2017-01-02 00:00:00    2
+   Freq: 12H, Length: 3, dtype: int64
+
+Once again, the ``min_count`` keyword is available to restore the 0.21 behavior.
+
+.. code-block:: ipython
+
+   In [16]: pd.Series([1, 2], index=idx).resample("12H").sum(min_count=1)
+   Out[16]:
+   2017-01-01 00:00:00    1.0
+   2017-01-01 12:00:00    NaN
+   2017-01-02 00:00:00    2.0
+   Freq: 12H, Length: 3, dtype: float64
+
+
+Rolling and expanding
+^^^^^^^^^^^^^^^^^^^^^
+
+Rolling and expanding already have a ``min_periods`` keyword that behaves
+similar to ``min_count``. The only case that changes is when doing a rolling
+or expanding sum with ``min_periods=0``. Previously this returned ``NaN``,
+when fewer than ``min_periods`` non-*NA* values were in the window. Now it
+returns ``0``.
+
+*pandas 0.21.1*
+
+.. code-block:: ipython
+
+   In [17]: s = pd.Series([np.nan, np.nan])
+
+   In [18]: s.rolling(2, min_periods=0).sum()
+   Out[18]:
+   0   NaN
+   1   NaN
+   dtype: float64
+
+*pandas 0.22.0*
+
+.. ipython:: python
+
+   s = pd.Series([np.nan, np.nan])
+   s.rolling(2, min_periods=0).sum()
+
+The default behavior of ``min_periods=None``, implying that ``min_periods``
+equals the window size, is unchanged.
+
+Compatibility
+~~~~~~~~~~~~~
+
+If you maintain a library that should work across pandas versions, it
+may be easiest to exclude pandas 0.21 from your requirements. Otherwise, all your
+``sum()`` calls would need to check if the ``Series`` is empty before summing.
+
+With setuptools, in your ``setup.py`` use::
+
+    install_requires=['pandas!=0.21.*', ...]
+
+With conda, use
+
+.. code-block:: yaml
+
+    requirements:
+      run:
+        - pandas !=0.21.0,!=0.21.1
+
+Note that the inconsistency in the return value for all-*NA* series is still
+there for pandas 0.20.3 and earlier. Avoiding pandas 0.21 will only help with
+the empty case.
+
+
+.. _whatsnew_0.22.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.21.1..v0.22.0
diff --git a/pandas/doc/source/whatsnew/v0.23.0.rst b/pandas/doc/source/whatsnew/v0.23.0.rst
new file mode 100644
index 0000000000000..7f7609edc27b6
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.23.0.rst
@@ -0,0 +1,1535 @@
+.. _whatsnew_0230:
+
+What's new in 0.23.0 (May 15, 2018)
+-----------------------------------
+
+{{ header }}
+
+.. ipython:: python
+   :suppress:
+
+   from pandas import * # noqa F401, F403
+
+
+This is a major release from 0.22.0 and includes a number of API changes,
+deprecations, new features, enhancements, and performance improvements along
+with a large number of bug fixes. We recommend that all users upgrade to this
+version.
+
+Highlights include:
+
+- :ref:`Round-trippable JSON format with 'table' orient <whatsnew_0230.enhancements.round-trippable_json>`.
+- :ref:`Instantiation from dicts respects order for Python 3.6+ <whatsnew_0230.api_breaking.dict_insertion_order>`.
+- :ref:`Dependent column arguments for assign <whatsnew_0230.enhancements.assign_dependent>`.
+- :ref:`Merging / sorting on a combination of columns and index levels <whatsnew_0230.enhancements.merge_on_columns_and_levels>`.
+- :ref:`Extending pandas with custom types <whatsnew_023.enhancements.extension>`.
+- :ref:`Excluding unobserved categories from groupby <whatsnew_0230.enhancements.categorical_grouping>`.
+- :ref:`Changes to make output shape of DataFrame.apply consistent <whatsnew_0230.api_breaking.apply>`.
+
+Check the :ref:`API Changes <whatsnew_0230.api_breaking>` and :ref:`deprecations <whatsnew_0230.deprecations>` before updating.
+
+.. warning::
+
+   Starting January 1, 2019, pandas feature releases will support Python 3 only.
+   See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+.. contents:: What's new in v0.23.0
+    :local:
+    :backlinks: none
+    :depth: 2
+
+.. _whatsnew_0230.enhancements:
+
+New features
+~~~~~~~~~~~~
+
+.. _whatsnew_0230.enhancements.round-trippable_json:
+
+JSON read/write round-trippable with ``orient='table'``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A ``DataFrame`` can now be written to and subsequently read back via JSON while preserving metadata through usage of the ``orient='table'`` argument (see :issue:`18912` and :issue:`9146`). Previously, none of the available ``orient`` values guaranteed the preservation of dtypes and index names, amongst other metadata.
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame({'foo': [1, 2, 3, 4],
+      ...:                    'bar': ['a', 'b', 'c', 'd'],
+      ...:                    'baz': pd.date_range('2018-01-01', freq='d', periods=4),
+      ...:                    'qux': pd.Categorical(['a', 'b', 'c', 'c'])},
+      ...:                   index=pd.Index(range(4), name='idx'))
+
+   In [2]: df
+   Out[2]:
+        foo bar        baz qux
+   idx
+   0      1   a 2018-01-01   a
+   1      2   b 2018-01-02   b
+   2      3   c 2018-01-03   c
+   3      4   d 2018-01-04   c
+
+   [4 rows x 4 columns]
+
+   In [3]: df.dtypes
+   Out[3]:
+   foo             int64
+   bar            object
+   baz    datetime64[ns]
+   qux          category
+   Length: 4, dtype: object
+
+   In [4]: df.to_json('test.json', orient='table')
+
+   In [5]: new_df = pd.read_json('test.json', orient='table')
+
+   In [6]: new_df
+   Out[6]:
+        foo bar        baz qux
+   idx
+   0      1   a 2018-01-01   a
+   1      2   b 2018-01-02   b
+   2      3   c 2018-01-03   c
+   3      4   d 2018-01-04   c
+
+   [4 rows x 4 columns]
+
+   In [7]: new_df.dtypes
+   Out[7]:
+   foo             int64
+   bar            object
+   baz    datetime64[ns]
+   qux          category
+   Length: 4, dtype: object
+
+Please note that the string ``index`` is not supported with the round trip format, as it is used by default in ``write_json`` to indicate a missing index name.
+
+.. ipython:: python
+   :okwarning:
+
+   df.index.name = 'index'
+
+   df.to_json('test.json', orient='table')
+   new_df = pd.read_json('test.json', orient='table')
+   new_df
+   new_df.dtypes
+
+.. ipython:: python
+   :suppress:
+
+   import os
+   os.remove('test.json')
+
+
+.. _whatsnew_0230.enhancements.assign_dependent:
+
+
+Method ``.assign()`` accepts dependent arguments
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :func:`DataFrame.assign` now accepts dependent keyword arguments for python version later than 3.6 (see also `PEP 468
+<https://www.python.org/dev/peps/pep-0468/>`_). Later keyword arguments may now refer to earlier ones if the argument is a callable. See the
+:ref:`documentation here <dsintro.chained_assignment>` (:issue:`14207`)
+
+.. ipython:: python
+
+    df = pd.DataFrame({'A': [1, 2, 3]})
+    df
+    df.assign(B=df.A, C=lambda x: x['A'] + x['B'])
+
+.. warning::
+
+  This may subtly change the behavior of your code when you're
+  using ``.assign()`` to update an existing column. Previously, callables
+  referring to other variables being updated would get the "old" values
+
+  Previous behavior:
+
+  .. code-block:: ipython
+
+      In [2]: df = pd.DataFrame({"A": [1, 2, 3]})
+
+      In [3]: df.assign(A=lambda df: df.A + 1, C=lambda df: df.A * -1)
+      Out[3]:
+         A  C
+      0  2 -1
+      1  3 -2
+      2  4 -3
+
+  New behavior:
+
+  .. ipython:: python
+
+      df.assign(A=df.A + 1, C=lambda df: df.A * -1)
+
+
+
+.. _whatsnew_0230.enhancements.merge_on_columns_and_levels:
+
+Merging on a combination of columns and index levels
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Strings passed to :meth:`DataFrame.merge` as the ``on``, ``left_on``, and ``right_on``
+parameters may now refer to either column names or index level names.
+This enables merging ``DataFrame`` instances on a combination of index levels
+and columns without resetting indexes. See the :ref:`Merge on columns and
+levels <merging.merge_on_columns_and_levels>` documentation section.
+(:issue:`14355`)
+
+.. ipython:: python
+
+   left_index = pd.Index(['K0', 'K0', 'K1', 'K2'], name='key1')
+
+   left = pd.DataFrame({'A': ['A0', 'A1', 'A2', 'A3'],
+                        'B': ['B0', 'B1', 'B2', 'B3'],
+                        'key2': ['K0', 'K1', 'K0', 'K1']},
+                       index=left_index)
+
+   right_index = pd.Index(['K0', 'K1', 'K2', 'K2'], name='key1')
+
+   right = pd.DataFrame({'C': ['C0', 'C1', 'C2', 'C3'],
+                         'D': ['D0', 'D1', 'D2', 'D3'],
+                         'key2': ['K0', 'K0', 'K0', 'K1']},
+                        index=right_index)
+
+   left.merge(right, on=['key1', 'key2'])
+
+.. _whatsnew_0230.enhancements.sort_by_columns_and_levels:
+
+Sorting by a combination of columns and index levels
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Strings passed to :meth:`DataFrame.sort_values` as the ``by`` parameter may
+now refer to either column names or index level names.  This enables sorting
+``DataFrame`` instances by a combination of index levels and columns without
+resetting indexes. See the :ref:`Sorting by Indexes and Values
+<basics.sort_indexes_and_values>` documentation section.
+(:issue:`14353`)
+
+.. ipython:: python
+
+   # Build MultiIndex
+   idx = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('a', 2),
+                                    ('b', 2), ('b', 1), ('b', 1)])
+   idx.names = ['first', 'second']
+
+   # Build DataFrame
+   df_multi = pd.DataFrame({'A': np.arange(6, 0, -1)},
+                           index=idx)
+   df_multi
+
+   # Sort by 'second' (index) and 'A' (column)
+   df_multi.sort_values(by=['second', 'A'])
+
+
+.. _whatsnew_023.enhancements.extension:
+
+Extending pandas with custom types (experimental)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas now supports storing array-like objects that aren't necessarily 1-D NumPy
+arrays as columns in a DataFrame or values in a Series. This allows third-party
+libraries to implement extensions to NumPy's types, similar to how pandas
+implemented categoricals, datetimes with timezones, periods, and intervals.
+
+As a demonstration, we'll use cyberpandas_, which provides an ``IPArray`` type
+for storing ip addresses.
+
+.. code-block:: ipython
+
+   In [1]: from cyberpandas import IPArray
+
+   In [2]: values = IPArray([
+      ...:     0,
+      ...:     3232235777,
+      ...:     42540766452641154071740215577757643572
+      ...: ])
+      ...:
+      ...:
+
+``IPArray`` isn't a normal 1-D NumPy array, but because it's a pandas
+:class:`~pandas.api.extensions.ExtensionArray`, it can be stored properly inside pandas' containers.
+
+.. code-block:: ipython
+
+   In [3]: ser = pd.Series(values)
+
+   In [4]: ser
+   Out[4]:
+   0                         0.0.0.0
+   1                     192.168.1.1
+   2    2001:db8:85a3::8a2e:370:7334
+   dtype: ip
+
+Notice that the dtype is ``ip``. The missing value semantics of the underlying
+array are respected:
+
+.. code-block:: ipython
+
+   In [5]: ser.isna()
+   Out[5]:
+   0     True
+   1    False
+   2    False
+   dtype: bool
+
+For more, see the :ref:`extension types <extending.extension-types>`
+documentation. If you build an extension array, publicize it on `the ecosystem page <https://pandas.pydata.org/community/ecosystem.html>`_.
+
+.. _cyberpandas: https://cyberpandas.readthedocs.io/en/latest/
+
+
+.. _whatsnew_0230.enhancements.categorical_grouping:
+
+New ``observed`` keyword for excluding unobserved categories in ``GroupBy``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Grouping by a categorical includes the unobserved categories in the output.
+When grouping by multiple categorical columns, this means you get the cartesian product of all the
+categories, including combinations where there are no observations, which can result in a large
+number of groups. We have added a keyword ``observed`` to control this behavior, it defaults to
+``observed=False`` for backward-compatibility. (:issue:`14942`, :issue:`8138`, :issue:`15217`, :issue:`17594`, :issue:`8669`, :issue:`20583`, :issue:`20902`)
+
+.. ipython:: python
+
+   cat1 = pd.Categorical(["a", "a", "b", "b"],
+                         categories=["a", "b", "z"], ordered=True)
+   cat2 = pd.Categorical(["c", "d", "c", "d"],
+                         categories=["c", "d", "y"], ordered=True)
+   df = pd.DataFrame({"A": cat1, "B": cat2, "values": [1, 2, 3, 4]})
+   df['C'] = ['foo', 'bar'] * 2
+   df
+
+To show all values, the previous behavior:
+
+.. ipython:: python
+
+   df.groupby(['A', 'B', 'C'], observed=False).count()
+
+
+To show only observed values:
+
+.. ipython:: python
+
+   df.groupby(['A', 'B', 'C'], observed=True).count()
+
+For pivoting operations, this behavior is *already* controlled by the ``dropna`` keyword:
+
+.. ipython:: python
+
+   cat1 = pd.Categorical(["a", "a", "b", "b"],
+                         categories=["a", "b", "z"], ordered=True)
+   cat2 = pd.Categorical(["c", "d", "c", "d"],
+                         categories=["c", "d", "y"], ordered=True)
+   df = pd.DataFrame({"A": cat1, "B": cat2, "values": [1, 2, 3, 4]})
+   df
+
+
+.. code-block:: ipython
+
+    In [1]: pd.pivot_table(df, values='values', index=['A', 'B'], dropna=True)
+
+    Out[1]:
+         values
+    A B
+    a c     1.0
+      d     2.0
+    b c     3.0
+      d     4.0
+
+    In [2]: pd.pivot_table(df, values='values', index=['A', 'B'], dropna=False)
+
+    Out[2]:
+         values
+    A B
+    a c     1.0
+      d     2.0
+      y     NaN
+    b c     3.0
+      d     4.0
+      y     NaN
+    z c     NaN
+      d     NaN
+      y     NaN
+
+
+.. _whatsnew_0230.enhancements.window_raw:
+
+Rolling/Expanding.apply() accepts ``raw=False`` to pass a ``Series`` to the function
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`Series.rolling().apply() <.Rolling.apply>`, :func:`DataFrame.rolling().apply() <.Rolling.apply>`,
+:func:`Series.expanding().apply() <.Expanding.apply>`, and :func:`DataFrame.expanding().apply() <.Expanding.apply>` have gained a ``raw=None`` parameter.
+This is similar to :func:`DataFame.apply`. This parameter, if ``True`` allows one to send a ``np.ndarray`` to the applied function. If ``False`` a ``Series`` will be passed. The
+default is ``None``, which preserves backward compatibility, so this will default to ``True``, sending an ``np.ndarray``.
+In a future version the default will be changed to ``False``, sending a ``Series``. (:issue:`5071`, :issue:`20584`)
+
+.. ipython:: python
+
+   s = pd.Series(np.arange(5), np.arange(5) + 1)
+   s
+
+Pass a ``Series``:
+
+.. ipython:: python
+
+   s.rolling(2, min_periods=1).apply(lambda x: x.iloc[-1], raw=False)
+
+Mimic the original behavior of passing a ndarray:
+
+.. ipython:: python
+
+   s.rolling(2, min_periods=1).apply(lambda x: x[-1], raw=True)
+
+
+.. _whatsnew_0210.enhancements.limit_area:
+
+``DataFrame.interpolate`` has gained the ``limit_area`` kwarg
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.interpolate` has gained a ``limit_area`` parameter to allow further control of which ``NaN`` s are replaced.
+Use ``limit_area='inside'`` to fill only NaNs surrounded by valid values or use ``limit_area='outside'`` to fill only ``NaN`` s
+outside the existing valid values while preserving those inside.  (:issue:`16284`) See the :ref:`full documentation here <missing_data.interp_limits>`.
+
+
+.. ipython:: python
+
+   ser = pd.Series([np.nan, np.nan, 5, np.nan, np.nan,
+                    np.nan, 13, np.nan, np.nan])
+   ser
+
+Fill one consecutive inside value in both directions
+
+.. ipython:: python
+
+   ser.interpolate(limit_direction='both', limit_area='inside', limit=1)
+
+Fill all consecutive outside values backward
+
+.. ipython:: python
+
+   ser.interpolate(limit_direction='backward', limit_area='outside')
+
+Fill all consecutive outside values in both directions
+
+.. ipython:: python
+
+   ser.interpolate(limit_direction='both', limit_area='outside')
+
+.. _whatsnew_0210.enhancements.get_dummies_dtype:
+
+Function ``get_dummies`` now supports ``dtype`` argument
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :func:`get_dummies` now accepts a ``dtype`` argument, which specifies a dtype for the new columns. The default remains uint8. (:issue:`18330`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': [1, 2], 'b': [3, 4], 'c': [5, 6]})
+   pd.get_dummies(df, columns=['c']).dtypes
+   pd.get_dummies(df, columns=['c'], dtype=bool).dtypes
+
+
+.. _whatsnew_0230.enhancements.timedelta_mod:
+
+Timedelta mod method
+^^^^^^^^^^^^^^^^^^^^
+
+``mod`` (%) and ``divmod`` operations are now defined on ``Timedelta`` objects
+when operating with either timedelta-like or with numeric arguments.
+See the :ref:`documentation here <timedeltas.mod_divmod>`. (:issue:`19365`)
+
+.. ipython:: python
+
+    td = pd.Timedelta(hours=37)
+    td % pd.Timedelta(minutes=45)
+
+.. _whatsnew_0230.enhancements.ran_inf:
+
+Method ``.rank()`` handles ``inf`` values when ``NaN`` are present
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions, ``.rank()`` would assign ``inf`` elements ``NaN`` as their ranks. Now ranks are calculated properly. (:issue:`6945`)
+
+.. ipython:: python
+
+    s = pd.Series([-np.inf, 0, 1, np.nan, np.inf])
+    s
+
+Previous behavior:
+
+.. code-block:: ipython
+
+    In [11]: s.rank()
+    Out[11]:
+    0    1.0
+    1    2.0
+    2    3.0
+    3    NaN
+    4    NaN
+    dtype: float64
+
+Current behavior:
+
+.. ipython:: python
+
+    s.rank()
+
+Furthermore, previously if you rank ``inf`` or ``-inf`` values together with ``NaN`` values, the calculation won't distinguish ``NaN`` from infinity when using 'top' or 'bottom' argument.
+
+.. ipython:: python
+
+    s = pd.Series([np.nan, np.nan, -np.inf, -np.inf])
+    s
+
+Previous behavior:
+
+.. code-block:: ipython
+
+    In [15]: s.rank(na_option='top')
+    Out[15]:
+    0    2.5
+    1    2.5
+    2    2.5
+    3    2.5
+    dtype: float64
+
+Current behavior:
+
+.. ipython:: python
+
+    s.rank(na_option='top')
+
+These bugs were squashed:
+
+- Bug in :meth:`DataFrame.rank` and :meth:`Series.rank` when ``method='dense'`` and ``pct=True`` in which percentile ranks were not being used with the number of distinct observations (:issue:`15630`)
+- Bug in :meth:`Series.rank` and :meth:`DataFrame.rank` when ``ascending='False'`` failed to return correct ranks for infinity if ``NaN`` were present (:issue:`19538`)
+- Bug in :func:`DataFrameGroupBy.rank` where ranks were incorrect when both infinity and ``NaN`` were present (:issue:`20561`)
+
+
+.. _whatsnew_0230.enhancements.str_cat_align:
+
+``Series.str.cat`` has gained the ``join`` kwarg
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, :meth:`Series.str.cat` did not -- in contrast to most of ``pandas`` -- align :class:`Series` on their index before concatenation (see :issue:`18657`).
+The method has now gained a keyword ``join`` to control the manner of alignment, see examples below and :ref:`here <text.concatenate>`.
+
+In v.0.23 ``join`` will default to None (meaning no alignment), but this default will change to ``'left'`` in a future version of pandas.
+
+.. ipython:: python
+   :okwarning:
+
+    s = pd.Series(['a', 'b', 'c', 'd'])
+    t = pd.Series(['b', 'd', 'e', 'c'], index=[1, 3, 4, 2])
+    s.str.cat(t)
+    s.str.cat(t, join='left', na_rep='-')
+
+Furthermore, :meth:`Series.str.cat` now works for ``CategoricalIndex`` as well (previously raised a ``ValueError``; see :issue:`20842`).
+
+.. _whatsnew_0230.enhancements.astype_category:
+
+``DataFrame.astype`` performs column-wise conversion to ``Categorical``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.astype` can now perform column-wise conversion to ``Categorical`` by supplying the string ``'category'`` or
+a :class:`~pandas.api.types.CategoricalDtype`. Previously, attempting this would raise a ``NotImplementedError``. See the
+:ref:`categorical.objectcreation` section of the documentation for more details and examples. (:issue:`12860`, :issue:`18099`)
+
+Supplying the string ``'category'`` performs column-wise conversion, with only labels appearing in a given column set as categories:
+
+.. ipython:: python
+
+    df = pd.DataFrame({'A': list('abca'), 'B': list('bccd')})
+    df = df.astype('category')
+    df['A'].dtype
+    df['B'].dtype
+
+
+Supplying a ``CategoricalDtype`` will make the categories in each column consistent with the supplied dtype:
+
+.. ipython:: python
+
+    from pandas.api.types import CategoricalDtype
+    df = pd.DataFrame({'A': list('abca'), 'B': list('bccd')})
+    cdt = CategoricalDtype(categories=list('abcd'), ordered=True)
+    df = df.astype(cdt)
+    df['A'].dtype
+    df['B'].dtype
+
+
+.. _whatsnew_0230.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Unary ``+`` now permitted for ``Series`` and ``DataFrame`` as  numeric operator (:issue:`16073`)
+- Better support for :meth:`~pandas.io.formats.style.Styler.to_excel` output with the ``xlsxwriter`` engine. (:issue:`16149`)
+- :func:`pandas.tseries.frequencies.to_offset` now accepts leading '+' signs e.g. '+1h'. (:issue:`18171`)
+- :func:`MultiIndex.unique` now supports the ``level=`` argument, to get unique values from a specific index level (:issue:`17896`)
+- :class:`pandas.io.formats.style.Styler` now has method ``hide_index()`` to determine whether the index will be rendered in output (:issue:`14194`)
+- :class:`pandas.io.formats.style.Styler` now has method ``hide_columns()`` to determine whether columns will be hidden in output (:issue:`14194`)
+- Improved wording of ``ValueError`` raised in :func:`to_datetime` when ``unit=`` is passed with a non-convertible value (:issue:`14350`)
+- :func:`Series.fillna` now accepts a Series or a dict as a ``value`` for a categorical dtype (:issue:`17033`)
+- :func:`pandas.read_clipboard` updated to use qtpy, falling back to PyQt5 and then PyQt4, adding compatibility with Python3 and multiple python-qt bindings (:issue:`17722`)
+- Improved wording of ``ValueError`` raised in :func:`read_csv` when the ``usecols`` argument cannot match all columns. (:issue:`17301`)
+- :func:`DataFrame.corrwith` now silently drops non-numeric columns when passed a Series. Before, an exception was raised (:issue:`18570`).
+- :class:`IntervalIndex` now supports time zone aware ``Interval`` objects (:issue:`18537`, :issue:`18538`)
+- :func:`Series` / :func:`DataFrame` tab completion also returns identifiers in the first level of a :func:`MultiIndex`. (:issue:`16326`)
+- :func:`read_excel` has gained the ``nrows`` parameter (:issue:`16645`)
+- :meth:`DataFrame.append` can now in more cases preserve the type of the calling dataframe's columns (e.g. if both are ``CategoricalIndex``) (:issue:`18359`)
+- :meth:`DataFrame.to_json` and :meth:`Series.to_json` now accept an ``index`` argument which allows the user to exclude the index from the JSON output (:issue:`17394`)
+- ``IntervalIndex.to_tuples()`` has gained the ``na_tuple`` parameter to control whether NA is returned as a tuple of NA, or NA itself (:issue:`18756`)
+- ``Categorical.rename_categories``, ``CategoricalIndex.rename_categories`` and :attr:`Series.cat.rename_categories`
+  can now take a callable as their argument (:issue:`18862`)
+- :class:`Interval` and :class:`IntervalIndex` have gained a ``length`` attribute (:issue:`18789`)
+- ``Resampler`` objects now have a functioning :attr:`.Resampler.pipe` method.
+  Previously, calls to ``pipe`` were diverted to  the ``mean`` method (:issue:`17905`).
+- :func:`~pandas.api.types.is_scalar` now returns ``True`` for ``DateOffset`` objects (:issue:`18943`).
+- :func:`DataFrame.pivot` now accepts a list for the ``values=`` kwarg (:issue:`17160`).
+- Added :func:`pandas.api.extensions.register_dataframe_accessor`,
+  :func:`pandas.api.extensions.register_series_accessor`, and
+  :func:`pandas.api.extensions.register_index_accessor`, accessor for libraries downstream of pandas
+  to register custom accessors like ``.cat`` on pandas objects. See
+  :ref:`Registering Custom Accessors <extending.register-accessors>` for more (:issue:`14781`).
+
+- ``IntervalIndex.astype`` now supports conversions between subtypes when passed an ``IntervalDtype`` (:issue:`19197`)
+- :class:`IntervalIndex` and its associated constructor methods (``from_arrays``, ``from_breaks``, ``from_tuples``) have gained a ``dtype`` parameter (:issue:`19262`)
+- Added :func:`.SeriesGroupBy.is_monotonic_increasing` and :func:`.SeriesGroupBy.is_monotonic_decreasing` (:issue:`17015`)
+- For subclassed ``DataFrames``, :func:`DataFrame.apply` will now preserve the ``Series`` subclass (if defined) when passing the data to the applied function (:issue:`19822`)
+- :func:`DataFrame.from_dict` now accepts a ``columns`` argument that can be used to specify the column names when ``orient='index'`` is used (:issue:`18529`)
+- Added option ``display.html.use_mathjax`` so `MathJax <https://www.mathjax.org/>`_ can be disabled when rendering tables in ``Jupyter`` notebooks (:issue:`19856`, :issue:`19824`)
+- :func:`DataFrame.replace` now supports the ``method`` parameter, which can be used to specify the replacement method when ``to_replace`` is a scalar, list or tuple and ``value`` is ``None`` (:issue:`19632`)
+- :meth:`Timestamp.month_name`, :meth:`DatetimeIndex.month_name`, and :meth:`Series.dt.month_name` are now available (:issue:`12805`)
+- :meth:`Timestamp.day_name` and :meth:`DatetimeIndex.day_name` are now available to return day names with a specified locale (:issue:`12806`)
+- :meth:`DataFrame.to_sql` now performs a multi-value insert if the underlying connection supports itk rather than inserting row by row.
+  ``SQLAlchemy`` dialects supporting multi-value inserts include: ``mysql``, ``postgresql``, ``sqlite`` and any dialect with ``supports_multivalues_insert``. (:issue:`14315`, :issue:`8953`)
+- :func:`read_html` now accepts a ``displayed_only`` keyword argument to controls whether or not hidden elements are parsed (``True`` by default) (:issue:`20027`)
+- :func:`read_html` now reads all ``<tbody>`` elements in a ``<table>``, not just the first. (:issue:`20690`)
+- :meth:`.Rolling.quantile` and :meth:`.Expanding.quantile` now accept the ``interpolation`` keyword, ``linear`` by default (:issue:`20497`)
+- zip compression is supported via ``compression=zip`` in :func:`DataFrame.to_pickle`, :func:`Series.to_pickle`, :func:`DataFrame.to_csv`, :func:`Series.to_csv`, :func:`DataFrame.to_json`, :func:`Series.to_json`. (:issue:`17778`)
+- :class:`~pandas.tseries.offsets.WeekOfMonth` constructor now supports ``n=0`` (:issue:`20517`).
+- :class:`DataFrame` and :class:`Series` now support matrix multiplication (``@``) operator (:issue:`10259`) for Python>=3.5
+- Updated :meth:`DataFrame.to_gbq` and :meth:`pandas.read_gbq` signature and documentation to reflect changes from
+  the pandas-gbq library version 0.4.0. Adds intersphinx mapping to pandas-gbq
+  library. (:issue:`20564`)
+- Added new writer for exporting Stata dta files in version 117, ``StataWriter117``.  This format supports exporting strings with lengths up to 2,000,000 characters (:issue:`16450`)
+- :func:`to_hdf` and :func:`read_hdf` now accept an ``errors`` keyword argument to control encoding error handling (:issue:`20835`)
+- :func:`cut` has gained the ``duplicates='raise'|'drop'`` option to control whether to raise on duplicated edges (:issue:`20947`)
+- :func:`date_range`, :func:`timedelta_range`, and :func:`interval_range` now return a linearly spaced index if ``start``, ``stop``, and ``periods`` are specified, but ``freq`` is not. (:issue:`20808`, :issue:`20983`, :issue:`20976`)
+
+.. _whatsnew_0230.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0230.api_breaking.deps:
+
+Dependencies have increased minimum versions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We have updated our minimum supported versions of dependencies (:issue:`15184`).
+If installed, we now require:
+
++-----------------+-----------------+----------+---------------+
+| Package         | Minimum Version | Required |     Issue     |
++=================+=================+==========+===============+
+| python-dateutil | 2.5.0           |    X     | :issue:`15184`|
++-----------------+-----------------+----------+---------------+
+| openpyxl        | 2.4.0           |          | :issue:`15184`|
++-----------------+-----------------+----------+---------------+
+| beautifulsoup4  | 4.2.1           |          | :issue:`20082`|
++-----------------+-----------------+----------+---------------+
+| setuptools      | 24.2.0          |          | :issue:`20698`|
++-----------------+-----------------+----------+---------------+
+
+.. _whatsnew_0230.api_breaking.dict_insertion_order:
+
+Instantiation from dicts preserves dict insertion order for Python 3.6+
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Until Python 3.6, dicts in Python had no formally defined ordering. For Python
+version 3.6 and later, dicts are ordered by insertion order, see
+`PEP 468 <https://www.python.org/dev/peps/pep-0468/>`_.
+pandas will use the dict's insertion order, when creating a ``Series`` or
+``DataFrame`` from a dict and you're using Python version 3.6 or
+higher. (:issue:`19884`)
+
+Previous behavior (and current behavior if on Python < 3.6):
+
+.. code-block:: ipython
+
+    In [16]: pd.Series({'Income': 2000,
+       ....:            'Expenses': -1500,
+       ....:            'Taxes': -200,
+       ....:            'Net result': 300})
+    Out[16]:
+    Expenses     -1500
+    Income        2000
+    Net result     300
+    Taxes         -200
+    dtype: int64
+
+Note the Series above is ordered alphabetically by the index values.
+
+New behavior (for Python >= 3.6):
+
+.. ipython:: python
+
+    pd.Series({'Income': 2000,
+               'Expenses': -1500,
+               'Taxes': -200,
+               'Net result': 300})
+
+Notice that the Series is now ordered by insertion order. This new behavior is
+used for all relevant pandas types (``Series``, ``DataFrame``, ``SparseSeries``
+and ``SparseDataFrame``).
+
+If you wish to retain the old behavior while using Python >= 3.6, you can use
+``.sort_index()``:
+
+.. ipython:: python
+
+    pd.Series({'Income': 2000,
+               'Expenses': -1500,
+               'Taxes': -200,
+               'Net result': 300}).sort_index()
+
+.. _whatsnew_0230.api_breaking.deprecate_panel:
+
+Deprecate Panel
+^^^^^^^^^^^^^^^
+
+``Panel`` was deprecated in the 0.20.x release, showing as a ``DeprecationWarning``. Using ``Panel`` will now show a ``FutureWarning``. The recommended way to represent 3-D data are
+with a ``MultiIndex`` on a ``DataFrame`` via the :meth:`~Panel.to_frame` or with the `xarray package <http://xarray.pydata.org/en/stable/>`__. pandas
+provides a :meth:`~Panel.to_xarray` method to automate this conversion (:issue:`13563`, :issue:`18324`).
+
+.. code-block:: ipython
+
+    In [75]: import pandas._testing as tm
+
+    In [76]: p = tm.makePanel()
+
+    In [77]: p
+    Out[77]:
+    <class 'pandas.core.panel.Panel'>
+    Dimensions: 3 (items) x 3 (major_axis) x 4 (minor_axis)
+    Items axis: ItemA to ItemC
+    Major_axis axis: 2000-01-03 00:00:00 to 2000-01-05 00:00:00
+    Minor_axis axis: A to D
+
+Convert to a MultiIndex DataFrame
+
+.. code-block:: ipython
+
+    In [78]: p.to_frame()
+    Out[78]:
+                         ItemA     ItemB     ItemC
+    major      minor
+    2000-01-03 A      0.469112  0.721555  0.404705
+               B     -1.135632  0.271860 -1.039268
+               C      0.119209  0.276232 -1.344312
+               D     -2.104569  0.113648 -0.109050
+    2000-01-04 A     -0.282863 -0.706771  0.577046
+               B      1.212112 -0.424972 -0.370647
+               C     -1.044236 -1.087401  0.844885
+               D     -0.494929 -1.478427  1.643563
+    2000-01-05 A     -1.509059 -1.039575 -1.715002
+               B     -0.173215  0.567020 -1.157892
+               C     -0.861849 -0.673690  1.075770
+               D      1.071804  0.524988 -1.469388
+
+    [12 rows x 3 columns]
+
+Convert to an xarray DataArray
+
+.. code-block:: ipython
+
+    In [79]: p.to_xarray()
+    Out[79]:
+    <xarray.DataArray (items: 3, major_axis: 3, minor_axis: 4)>
+    array([[[ 0.469112, -1.135632,  0.119209, -2.104569],
+            [-0.282863,  1.212112, -1.044236, -0.494929],
+            [-1.509059, -0.173215, -0.861849,  1.071804]],
+
+           [[ 0.721555,  0.27186 ,  0.276232,  0.113648],
+            [-0.706771, -0.424972, -1.087401, -1.478427],
+            [-1.039575,  0.56702 , -0.67369 ,  0.524988]],
+
+           [[ 0.404705, -1.039268, -1.344312, -0.10905 ],
+            [ 0.577046, -0.370647,  0.844885,  1.643563],
+            [-1.715002, -1.157892,  1.07577 , -1.469388]]])
+    Coordinates:
+      * items       (items) object 'ItemA' 'ItemB' 'ItemC'
+      * major_axis  (major_axis) datetime64[ns] 2000-01-03 2000-01-04 2000-01-05
+      * minor_axis  (minor_axis) object 'A' 'B' 'C' 'D'
+
+
+.. _whatsnew_0230.api_breaking.core_common:
+
+pandas.core.common removals
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The following error & warning messages are removed from ``pandas.core.common`` (:issue:`13634`, :issue:`19769`):
+
+- ``PerformanceWarning``
+- ``UnsupportedFunctionCall``
+- ``UnsortedIndexError``
+- ``AbstractMethodError``
+
+These are available from import from ``pandas.errors`` (since 0.19.0).
+
+
+.. _whatsnew_0230.api_breaking.apply:
+
+Changes to make output of ``DataFrame.apply`` consistent
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`DataFrame.apply` was inconsistent when applying an arbitrary user-defined-function that returned a list-like with ``axis=1``. Several bugs and inconsistencies
+are resolved. If the applied function returns a Series, then pandas will return a DataFrame; otherwise a Series will be returned, this includes the case
+where a list-like (e.g. ``tuple`` or ``list`` is returned) (:issue:`16353`, :issue:`17437`, :issue:`17970`, :issue:`17348`, :issue:`17892`, :issue:`18573`,
+:issue:`17602`, :issue:`18775`, :issue:`18901`, :issue:`18919`).
+
+.. ipython:: python
+
+    df = pd.DataFrame(np.tile(np.arange(3), 6).reshape(6, -1) + 1,
+                      columns=['A', 'B', 'C'])
+    df
+
+Previous behavior: if the returned shape happened to match the length of original columns, this would return a ``DataFrame``.
+If the return shape did not match, a ``Series`` with lists was returned.
+
+.. code-block:: python
+
+   In [3]: df.apply(lambda x: [1, 2, 3], axis=1)
+   Out[3]:
+      A  B  C
+   0  1  2  3
+   1  1  2  3
+   2  1  2  3
+   3  1  2  3
+   4  1  2  3
+   5  1  2  3
+
+   In [4]: df.apply(lambda x: [1, 2], axis=1)
+   Out[4]:
+   0    [1, 2]
+   1    [1, 2]
+   2    [1, 2]
+   3    [1, 2]
+   4    [1, 2]
+   5    [1, 2]
+   dtype: object
+
+
+New behavior: When the applied function returns a list-like, this will now *always* return a ``Series``.
+
+.. ipython:: python
+
+    df.apply(lambda x: [1, 2, 3], axis=1)
+    df.apply(lambda x: [1, 2], axis=1)
+
+To have expanded columns, you can use ``result_type='expand'``
+
+.. ipython:: python
+
+    df.apply(lambda x: [1, 2, 3], axis=1, result_type='expand')
+
+To broadcast the result across the original columns (the old behaviour for
+list-likes of the correct length), you can use ``result_type='broadcast'``.
+The shape must match the original columns.
+
+.. ipython:: python
+
+    df.apply(lambda x: [1, 2, 3], axis=1, result_type='broadcast')
+
+Returning a ``Series`` allows one to control the exact return structure and column names:
+
+.. ipython:: python
+
+    df.apply(lambda x: pd.Series([1, 2, 3], index=['D', 'E', 'F']), axis=1)
+
+.. _whatsnew_0230.api_breaking.concat:
+
+Concatenation will no longer sort
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In a future version of pandas :func:`pandas.concat` will no longer sort the non-concatenation axis when it is not already aligned.
+The current behavior is the same as the previous (sorting), but now a warning is issued when ``sort`` is not specified and the non-concatenation axis is not aligned (:issue:`4588`).
+
+.. ipython:: python
+   :okwarning:
+
+   df1 = pd.DataFrame({"a": [1, 2], "b": [1, 2]}, columns=['b', 'a'])
+   df2 = pd.DataFrame({"a": [4, 5]})
+
+   pd.concat([df1, df2])
+
+To keep the previous behavior (sorting) and silence the warning, pass ``sort=True``
+
+.. ipython:: python
+
+   pd.concat([df1, df2], sort=True)
+
+To accept the future behavior (no sorting), pass ``sort=False``
+
+.. ipython
+
+   pd.concat([df1, df2], sort=False)
+
+Note that this change also applies to :meth:`DataFrame.append`, which has also received a ``sort`` keyword for controlling this behavior.
+
+
+.. _whatsnew_0230.api_breaking.build_changes:
+
+Build changes
+^^^^^^^^^^^^^
+
+- Building pandas for development now requires ``cython >= 0.24`` (:issue:`18613`)
+- Building from source now explicitly requires ``setuptools`` in ``setup.py`` (:issue:`18113`)
+- Updated conda recipe to be in compliance with conda-build 3.0+ (:issue:`18002`)
+
+.. _whatsnew_0230.api_breaking.index_division_by_zero:
+
+Index division by zero fills correctly
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Division operations on ``Index`` and subclasses will now fill division of positive numbers by zero with ``np.inf``, division of negative numbers by zero with ``-np.inf`` and ``0 / 0`` with ``np.nan``.  This matches existing ``Series`` behavior. (:issue:`19322`, :issue:`19347`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+    In [6]: index = pd.Int64Index([-1, 0, 1])
+
+    In [7]: index / 0
+    Out[7]: Int64Index([0, 0, 0], dtype='int64')
+
+    # Previous behavior yielded different results depending on the type of zero in the divisor
+    In [8]: index / 0.0
+    Out[8]: Float64Index([-inf, nan, inf], dtype='float64')
+
+    In [9]: index = pd.UInt64Index([0, 1])
+
+    In [10]: index / np.array([0, 0], dtype=np.uint64)
+    Out[10]: UInt64Index([0, 0], dtype='uint64')
+
+    In [11]: pd.RangeIndex(1, 5) / 0
+    ZeroDivisionError: integer division or modulo by zero
+
+Current behavior:
+
+.. code-block:: ipython
+
+    In [12]: index = pd.Int64Index([-1, 0, 1])
+    # division by zero gives -infinity where negative,
+    # +infinity where positive, and NaN for 0 / 0
+    In [13]: index / 0
+
+    # The result of division by zero should not depend on
+    # whether the zero is int or float
+    In [14]: index / 0.0
+
+    In [15]: index = pd.UInt64Index([0, 1])
+    In [16]: index / np.array([0, 0], dtype=np.uint64)
+
+    In [17]: pd.RangeIndex(1, 5) / 0
+
+.. _whatsnew_0230.api_breaking.extract:
+
+Extraction of matching patterns from strings
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+By default, extracting matching patterns from strings with :func:`str.extract` used to return a
+``Series`` if a single group was being extracted (a ``DataFrame`` if more than one group was
+extracted). As of pandas 0.23.0 :func:`str.extract` always returns a ``DataFrame``, unless
+``expand`` is set to ``False``. Finally, ``None`` was an accepted value for
+the ``expand`` parameter (which was equivalent to ``False``), but now raises a ``ValueError``. (:issue:`11386`)
+
+Previous behavior:
+
+.. code-block:: ipython
+
+    In [1]: s = pd.Series(['number 10', '12 eggs'])
+
+    In [2]: extracted = s.str.extract(r'.*(\d\d).*')
+
+    In [3]: extracted
+    Out [3]:
+    0    10
+    1    12
+    dtype: object
+
+    In [4]: type(extracted)
+    Out [4]:
+    pandas.core.series.Series
+
+New behavior:
+
+.. ipython:: python
+
+    s = pd.Series(['number 10', '12 eggs'])
+    extracted = s.str.extract(r'.*(\d\d).*')
+    extracted
+    type(extracted)
+
+To restore previous behavior, simply set ``expand`` to ``False``:
+
+.. ipython:: python
+
+    s = pd.Series(['number 10', '12 eggs'])
+    extracted = s.str.extract(r'.*(\d\d).*', expand=False)
+    extracted
+    type(extracted)
+
+.. _whatsnew_0230.api_breaking.cdt_ordered:
+
+Default value for the ``ordered`` parameter of ``CategoricalDtype``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The default value of the ``ordered`` parameter for :class:`~pandas.api.types.CategoricalDtype` has changed from ``False`` to ``None`` to allow updating of ``categories`` without impacting ``ordered``.  Behavior should remain consistent for downstream objects, such as :class:`Categorical` (:issue:`18790`)
+
+In previous versions, the default value for the ``ordered`` parameter was ``False``.  This could potentially lead to the ``ordered`` parameter unintentionally being changed from ``True`` to ``False`` when users attempt to update ``categories`` if ``ordered`` is not explicitly specified, as it would silently default to ``False``.  The new behavior for ``ordered=None`` is to retain the existing value of ``ordered``.
+
+New behavior:
+
+.. code-block:: ipython
+
+    In [2]: from pandas.api.types import CategoricalDtype
+
+    In [3]: cat = pd.Categorical(list('abcaba'), ordered=True, categories=list('cba'))
+
+    In [4]: cat
+    Out[4]:
+    [a, b, c, a, b, a]
+    Categories (3, object): [c < b < a]
+
+    In [5]: cdt = CategoricalDtype(categories=list('cbad'))
+
+    In [6]: cat.astype(cdt)
+    Out[6]:
+    [a, b, c, a, b, a]
+    Categories (4, object): [c < b < a < d]
+
+Notice in the example above that the converted ``Categorical`` has retained ``ordered=True``.  Had the default value for ``ordered`` remained as ``False``, the converted ``Categorical`` would have become unordered, despite ``ordered=False`` never being explicitly specified.  To change the value of ``ordered``, explicitly pass it to the new dtype, e.g. ``CategoricalDtype(categories=list('cbad'), ordered=False)``.
+
+Note that the unintentional conversion of ``ordered`` discussed above did not arise in previous versions due to separate bugs that prevented ``astype`` from doing any type of category to category conversion (:issue:`10696`, :issue:`18593`).  These bugs have been fixed in this release, and motivated changing the default value of ``ordered``.
+
+.. _whatsnew_0230.api_breaking.pretty_printing:
+
+Better pretty-printing of DataFrames in a terminal
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Previously, the default value for the maximum number of columns was
+``pd.options.display.max_columns=20``. This meant that relatively wide data
+frames would not fit within the terminal width, and pandas would introduce line
+breaks to display these 20 columns. This resulted in an output that was
+relatively difficult to read:
+
+.. image:: ../_static/print_df_old.png
+
+If Python runs in a terminal, the maximum number of columns is now determined
+automatically so that the printed data frame fits within the current terminal
+width (``pd.options.display.max_columns=0``) (:issue:`17023`). If Python runs
+as a Jupyter kernel (such as the Jupyter QtConsole or a Jupyter notebook, as
+well as in many IDEs), this value cannot be inferred automatically and is thus
+set to ``20`` as in previous versions. In a terminal, this results in a much
+nicer output:
+
+.. image:: ../_static/print_df_new.png
+
+Note that if you don't like the new default, you can always set this option
+yourself. To revert to the old setting, you can run this line:
+
+.. code-block:: python
+
+  pd.options.display.max_columns = 20
+
+.. _whatsnew_0230.api.datetimelike:
+
+Datetimelike API changes
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- The default ``Timedelta`` constructor now accepts an ``ISO 8601 Duration`` string as an argument (:issue:`19040`)
+- Subtracting ``NaT`` from a :class:`Series` with ``dtype='datetime64[ns]'`` returns a ``Series`` with ``dtype='timedelta64[ns]'`` instead of ``dtype='datetime64[ns]'`` (:issue:`18808`)
+- Addition or subtraction of ``NaT`` from :class:`TimedeltaIndex` will return ``TimedeltaIndex`` instead of ``DatetimeIndex`` (:issue:`19124`)
+- :func:`DatetimeIndex.shift` and :func:`TimedeltaIndex.shift` will now raise ``NullFrequencyError`` (which subclasses ``ValueError``, which was raised in older versions) when the index object frequency is ``None`` (:issue:`19147`)
+- Addition and subtraction of ``NaN`` from a :class:`Series` with ``dtype='timedelta64[ns]'`` will raise a ``TypeError`` instead of treating the ``NaN`` as ``NaT`` (:issue:`19274`)
+- ``NaT`` division with :class:`datetime.timedelta` will now return ``NaN`` instead of raising (:issue:`17876`)
+- Operations between a :class:`Series` with dtype ``dtype='datetime64[ns]'`` and a :class:`PeriodIndex` will correctly raises ``TypeError`` (:issue:`18850`)
+- Subtraction of :class:`Series` with timezone-aware ``dtype='datetime64[ns]'`` with mismatched timezones will raise ``TypeError`` instead of ``ValueError`` (:issue:`18817`)
+- :class:`Timestamp` will no longer silently ignore unused or invalid ``tz`` or ``tzinfo`` keyword arguments (:issue:`17690`)
+- :class:`Timestamp` will no longer silently ignore invalid ``freq`` arguments (:issue:`5168`)
+- :class:`CacheableOffset` and :class:`WeekDay` are no longer available in the ``pandas.tseries.offsets`` module (:issue:`17830`)
+- ``pandas.tseries.frequencies.get_freq_group()`` and ``pandas.tseries.frequencies.DAYS`` are removed from the public API (:issue:`18034`)
+- :func:`Series.truncate` and :func:`DataFrame.truncate` will raise a ``ValueError`` if the index is not sorted instead of an unhelpful ``KeyError`` (:issue:`17935`)
+- :attr:`Series.first` and :attr:`DataFrame.first` will now raise a ``TypeError``
+  rather than ``NotImplementedError`` when index is not a :class:`DatetimeIndex` (:issue:`20725`).
+- :attr:`Series.last` and :attr:`DataFrame.last` will now raise a ``TypeError``
+  rather than ``NotImplementedError`` when index is not a :class:`DatetimeIndex` (:issue:`20725`).
+- Restricted ``DateOffset`` keyword arguments. Previously, ``DateOffset`` subclasses allowed arbitrary keyword arguments which could lead to unexpected behavior. Now, only valid arguments will be accepted. (:issue:`17176`, :issue:`18226`).
+- :func:`pandas.merge` provides a more informative error message when trying to merge on timezone-aware and timezone-naive columns (:issue:`15800`)
+- For :class:`DatetimeIndex` and :class:`TimedeltaIndex` with ``freq=None``, addition or subtraction of integer-dtyped array or ``Index`` will raise ``NullFrequencyError`` instead of ``TypeError`` (:issue:`19895`)
+- :class:`Timestamp` constructor now accepts a ``nanosecond`` keyword or positional argument (:issue:`18898`)
+- :class:`DatetimeIndex` will now raise an ``AttributeError`` when the ``tz`` attribute is set after instantiation (:issue:`3746`)
+- :class:`DatetimeIndex` with a ``pytz`` timezone will now return a consistent ``pytz`` timezone (:issue:`18595`)
+
+.. _whatsnew_0230.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- :func:`Series.astype` and :func:`Index.astype` with an incompatible dtype will now raise a ``TypeError`` rather than a ``ValueError`` (:issue:`18231`)
+- ``Series`` construction with an ``object`` dtyped tz-aware datetime and ``dtype=object`` specified, will now return an ``object`` dtyped ``Series``, previously this would infer the datetime dtype (:issue:`18231`)
+- A :class:`Series` of ``dtype=category`` constructed from an empty ``dict`` will now have categories of ``dtype=object`` rather than ``dtype=float64``, consistently with the case in which an empty list is passed (:issue:`18515`)
+- All-NaN levels in a ``MultiIndex`` are now assigned ``float`` rather than ``object`` dtype, promoting consistency with ``Index`` (:issue:`17929`).
+- Levels names of a ``MultiIndex`` (when not None) are now required to be unique: trying to create a ``MultiIndex`` with repeated names will raise a ``ValueError`` (:issue:`18872`)
+- Both construction and renaming of ``Index``/``MultiIndex`` with non-hashable ``name``/``names`` will now raise ``TypeError`` (:issue:`20527`)
+- :func:`Index.map` can now accept ``Series`` and dictionary input objects (:issue:`12756`, :issue:`18482`, :issue:`18509`).
+- :func:`DataFrame.unstack` will now default to filling with ``np.nan`` for ``object`` columns. (:issue:`12815`)
+- :class:`IntervalIndex` constructor will raise if the ``closed`` parameter conflicts with how the input data is inferred to be closed (:issue:`18421`)
+- Inserting missing values into indexes will work for all types of indexes and automatically insert the correct type of missing value (``NaN``, ``NaT``, etc.) regardless of the type passed in (:issue:`18295`)
+- When created with duplicate labels, ``MultiIndex`` now raises a ``ValueError``. (:issue:`17464`)
+- :func:`Series.fillna` now raises a ``TypeError`` instead of a ``ValueError`` when passed a list, tuple or DataFrame as a ``value`` (:issue:`18293`)
+- :func:`pandas.DataFrame.merge` no longer casts a ``float`` column to ``object`` when merging on ``int`` and ``float`` columns (:issue:`16572`)
+- :func:`pandas.merge` now raises a ``ValueError`` when trying to merge on incompatible data types (:issue:`9780`)
+- The default NA value for :class:`UInt64Index` has changed from 0 to ``NaN``, which impacts methods that mask with NA, such as ``UInt64Index.where()`` (:issue:`18398`)
+- Refactored ``setup.py`` to use ``find_packages`` instead of explicitly listing out all subpackages (:issue:`18535`)
+- Rearranged the order of keyword arguments in :func:`read_excel` to align with :func:`read_csv` (:issue:`16672`)
+- :func:`wide_to_long` previously kept numeric-like suffixes as ``object`` dtype. Now they are cast to numeric if possible (:issue:`17627`)
+- In :func:`read_excel`, the ``comment`` argument is now exposed as a named parameter (:issue:`18735`)
+- Rearranged the order of keyword arguments in :func:`read_excel` to align with :func:`read_csv` (:issue:`16672`)
+- The options ``html.border`` and ``mode.use_inf_as_null`` were deprecated in prior versions, these will now show ``FutureWarning`` rather than a ``DeprecationWarning`` (:issue:`19003`)
+- :class:`IntervalIndex` and ``IntervalDtype`` no longer support categorical, object, and string subtypes (:issue:`19016`)
+- ``IntervalDtype`` now returns ``True`` when compared against ``'interval'`` regardless of subtype, and ``IntervalDtype.name`` now returns ``'interval'`` regardless of subtype (:issue:`18980`)
+- ``KeyError`` now raises instead of ``ValueError`` in :meth:`~DataFrame.drop`, :meth:`~Panel.drop`, :meth:`~Series.drop`, :meth:`~Index.drop` when dropping a non-existent element in an axis with duplicates (:issue:`19186`)
+- :func:`Series.to_csv` now accepts a ``compression`` argument that works in the same way as the ``compression`` argument in :func:`DataFrame.to_csv` (:issue:`18958`)
+- Set operations (union, difference...) on :class:`IntervalIndex` with incompatible index types will now raise a ``TypeError`` rather than a ``ValueError`` (:issue:`19329`)
+- :class:`DateOffset` objects render more simply, e.g. ``<DateOffset: days=1>`` instead of ``<DateOffset: kwds={'days': 1}>`` (:issue:`19403`)
+- ``Categorical.fillna`` now validates its ``value`` and ``method`` keyword arguments. It now raises when both or none are specified, matching the behavior of :meth:`Series.fillna` (:issue:`19682`)
+- ``pd.to_datetime('today')`` now returns a datetime, consistent with ``pd.Timestamp('today')``; previously ``pd.to_datetime('today')`` returned a ``.normalized()`` datetime (:issue:`19935`)
+- :func:`Series.str.replace` now takes an optional ``regex`` keyword which, when set to ``False``, uses literal string replacement rather than regex replacement (:issue:`16808`)
+- :func:`DatetimeIndex.strftime` and :func:`PeriodIndex.strftime` now return an ``Index`` instead of a numpy array to be consistent with similar accessors (:issue:`20127`)
+- Constructing a Series from a list of length 1 no longer broadcasts this list when a longer index is specified (:issue:`19714`, :issue:`20391`).
+- :func:`DataFrame.to_dict` with ``orient='index'`` no longer casts int columns to float for a DataFrame with only int and float columns (:issue:`18580`)
+- A user-defined-function that is passed to :func:`Series.rolling().aggregate() <.Rolling.aggregate>`, :func:`DataFrame.rolling().aggregate() <.Rolling.aggregate>`, or its expanding cousins, will now *always* be passed a ``Series``, rather than a ``np.array``; ``.apply()`` only has the ``raw`` keyword, see :ref:`here <whatsnew_0230.enhancements.window_raw>`. This is consistent with the signatures of ``.aggregate()`` across pandas (:issue:`20584`)
+- Rolling and Expanding types raise ``NotImplementedError`` upon iteration (:issue:`11704`).
+
+.. _whatsnew_0230.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- ``Series.from_array`` and ``SparseSeries.from_array`` are deprecated. Use the normal constructor ``Series(..)`` and ``SparseSeries(..)`` instead (:issue:`18213`).
+- ``DataFrame.as_matrix`` is deprecated. Use ``DataFrame.values`` instead (:issue:`18458`).
+- ``Series.asobject``, ``DatetimeIndex.asobject``, ``PeriodIndex.asobject`` and ``TimeDeltaIndex.asobject`` have been deprecated. Use ``.astype(object)`` instead (:issue:`18572`)
+- Grouping by a tuple of keys now emits a ``FutureWarning`` and is deprecated.
+  In the future, a tuple passed to ``'by'`` will always refer to a single key
+  that is the actual tuple, instead of treating the tuple as multiple keys. To
+  retain the previous behavior, use a list instead of a tuple (:issue:`18314`)
+- ``Series.valid`` is deprecated. Use :meth:`Series.dropna` instead (:issue:`18800`).
+- :func:`read_excel` has deprecated the ``skip_footer`` parameter. Use ``skipfooter`` instead (:issue:`18836`)
+- :meth:`ExcelFile.parse` has deprecated ``sheetname`` in favor of ``sheet_name`` for consistency with :func:`read_excel` (:issue:`20920`).
+- The ``is_copy`` attribute is deprecated and will be removed in a future version (:issue:`18801`).
+- ``IntervalIndex.from_intervals`` is deprecated in favor of the :class:`IntervalIndex` constructor (:issue:`19263`)
+- ``DataFrame.from_items`` is deprecated. Use :func:`DataFrame.from_dict` instead, or ``DataFrame.from_dict(OrderedDict())`` if you wish to preserve the key order (:issue:`17320`, :issue:`17312`)
+- Indexing a :class:`MultiIndex` or a :class:`FloatIndex` with a list containing some missing keys will now show a :class:`FutureWarning`, which is consistent with other types of indexes (:issue:`17758`).
+
+- The ``broadcast`` parameter of ``.apply()`` is deprecated in favor of ``result_type='broadcast'`` (:issue:`18577`)
+- The ``reduce`` parameter of ``.apply()`` is deprecated in favor of ``result_type='reduce'`` (:issue:`18577`)
+- The ``order`` parameter of :func:`factorize` is deprecated and will be removed in a future release (:issue:`19727`)
+- :attr:`Timestamp.weekday_name`, :attr:`DatetimeIndex.weekday_name`, and :attr:`Series.dt.weekday_name` are deprecated in favor of :meth:`Timestamp.day_name`, :meth:`DatetimeIndex.day_name`, and :meth:`Series.dt.day_name` (:issue:`12806`)
+
+- ``pandas.tseries.plotting.tsplot`` is deprecated. Use :func:`Series.plot` instead (:issue:`18627`)
+- ``Index.summary()`` is deprecated and will be removed in a future version (:issue:`18217`)
+- ``NDFrame.get_ftype_counts()`` is deprecated and will be removed in a future version (:issue:`18243`)
+- The ``convert_datetime64`` parameter in :func:`DataFrame.to_records` has been deprecated and will be removed in a future version. The NumPy bug motivating this parameter has been resolved. The default value for this parameter has also changed from ``True`` to ``None`` (:issue:`18160`).
+- :func:`Series.rolling().apply() <.Rolling.apply>`, :func:`DataFrame.rolling().apply() <.Rolling.apply>`, :func:`Series.expanding().apply() <.Expanding.apply>`, and :func:`DataFrame.expanding().apply() <.Expanding.apply>` have deprecated passing an ``np.array`` by default. One will need to pass the new ``raw`` parameter to be explicit about what is passed (:issue:`20584`)
+- The ``data``, ``base``, ``strides``, ``flags`` and ``itemsize`` properties
+  of the ``Series`` and ``Index`` classes have been deprecated and will be
+  removed in a future version (:issue:`20419`).
+- ``DatetimeIndex.offset`` is deprecated. Use ``DatetimeIndex.freq`` instead (:issue:`20716`)
+- Floor division between an integer ndarray and a :class:`Timedelta` is deprecated. Divide by :attr:`Timedelta.value` instead (:issue:`19761`)
+- Setting ``PeriodIndex.freq`` (which was not guaranteed to work correctly) is deprecated. Use :meth:`PeriodIndex.asfreq` instead (:issue:`20678`)
+- ``Index.get_duplicates()`` is deprecated and will be removed in a future version (:issue:`20239`)
+- The previous default behavior of negative indices in ``Categorical.take`` is deprecated. In a future version it will change from meaning missing values to meaning positional indices from the right. The future behavior is consistent with :meth:`Series.take` (:issue:`20664`).
+- Passing multiple axes to the ``axis`` parameter in :func:`DataFrame.dropna` has been deprecated and will be removed in a future version (:issue:`20987`)
+
+
+.. _whatsnew_0230.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Warnings against the obsolete usage ``Categorical(codes, categories)``, which were emitted for instance when the first two arguments to ``Categorical()`` had different dtypes, and recommended the use of ``Categorical.from_codes``, have now been removed (:issue:`8074`)
+- The ``levels`` and ``labels`` attributes of a ``MultiIndex`` can no longer be set directly (:issue:`4039`).
+- ``pd.tseries.util.pivot_annual`` has been removed (deprecated since v0.19). Use ``pivot_table`` instead (:issue:`18370`)
+- ``pd.tseries.util.isleapyear`` has been removed (deprecated since v0.19). Use ``.is_leap_year`` property in Datetime-likes instead (:issue:`18370`)
+- ``pd.ordered_merge`` has been removed (deprecated since v0.19). Use ``pd.merge_ordered`` instead (:issue:`18459`)
+- The ``SparseList`` class has been removed (:issue:`14007`)
+- The ``pandas.io.wb`` and ``pandas.io.data`` stub modules have been removed (:issue:`13735`)
+- ``Categorical.from_array`` has been removed (:issue:`13854`)
+- The ``freq`` and ``how`` parameters have been removed from the ``rolling``/``expanding``/``ewm`` methods of DataFrame
+  and Series (deprecated since v0.18). Instead, resample before calling the methods. (:issue:`18601` & :issue:`18668`)
+- ``DatetimeIndex.to_datetime``, ``Timestamp.to_datetime``, ``PeriodIndex.to_datetime``, and ``Index.to_datetime`` have been removed (:issue:`8254`, :issue:`14096`, :issue:`14113`)
+- :func:`read_csv` has dropped the ``skip_footer`` parameter (:issue:`13386`)
+- :func:`read_csv` has dropped the ``as_recarray`` parameter (:issue:`13373`)
+- :func:`read_csv` has dropped the ``buffer_lines`` parameter (:issue:`13360`)
+- :func:`read_csv` has dropped the ``compact_ints`` and ``use_unsigned`` parameters (:issue:`13323`)
+- The ``Timestamp`` class has dropped the ``offset`` attribute in favor of ``freq`` (:issue:`13593`)
+- The ``Series``, ``Categorical``, and ``Index`` classes have dropped the ``reshape`` method (:issue:`13012`)
+- ``pandas.tseries.frequencies.get_standard_freq`` has been removed in favor of ``pandas.tseries.frequencies.to_offset(freq).rule_code`` (:issue:`13874`)
+- The ``freqstr`` keyword has been removed from ``pandas.tseries.frequencies.to_offset`` in favor of ``freq`` (:issue:`13874`)
+- The ``Panel4D`` and ``PanelND`` classes have been removed (:issue:`13776`)
+- The ``Panel`` class has dropped the ``to_long`` and ``toLong`` methods (:issue:`19077`)
+- The options ``display.line_with`` and ``display.height`` are removed in favor of ``display.width`` and ``display.max_rows`` respectively (:issue:`4391`, :issue:`19107`)
+- The ``labels`` attribute of the ``Categorical`` class has been removed in favor of :attr:`Categorical.codes` (:issue:`7768`)
+- The ``flavor`` parameter have been removed from :func:`to_sql` method (:issue:`13611`)
+- The modules ``pandas.tools.hashing`` and ``pandas.util.hashing`` have been removed (:issue:`16223`)
+- The top-level functions ``pd.rolling_*``, ``pd.expanding_*`` and ``pd.ewm*`` have been removed (Deprecated since v0.18).
+  Instead, use the DataFrame/Series methods :attr:`~DataFrame.rolling`, :attr:`~DataFrame.expanding` and :attr:`~DataFrame.ewm` (:issue:`18723`)
+- Imports from ``pandas.core.common`` for functions such as ``is_datetime64_dtype`` are now removed. These are located in ``pandas.api.types``. (:issue:`13634`, :issue:`19769`)
+- The ``infer_dst`` keyword in :meth:`Series.tz_localize`, :meth:`DatetimeIndex.tz_localize`
+  and :class:`DatetimeIndex` have been removed. ``infer_dst=True`` is equivalent to
+  ``ambiguous='infer'``, and ``infer_dst=False`` to ``ambiguous='raise'`` (:issue:`7963`).
+- When ``.resample()`` was changed from an eager to a lazy operation, like ``.groupby()`` in v0.18.0, we put in place compatibility (with a ``FutureWarning``),
+  so operations would continue to work. This is now fully removed, so a ``Resampler`` will no longer forward compat operations (:issue:`20554`)
+- Remove long deprecated ``axis=None`` parameter from ``.replace()`` (:issue:`20271`)
+
+.. _whatsnew_0230.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Indexers on ``Series`` or ``DataFrame`` no longer create a reference cycle (:issue:`17956`)
+- Added a keyword argument, ``cache``, to :func:`to_datetime` that improved the performance of converting duplicate datetime arguments (:issue:`11665`)
+- :class:`DateOffset` arithmetic performance is improved (:issue:`18218`)
+- Converting a ``Series`` of ``Timedelta`` objects to days, seconds, etc... sped up through vectorization of underlying methods (:issue:`18092`)
+- Improved performance of ``.map()`` with a ``Series/dict`` input (:issue:`15081`)
+- The overridden ``Timedelta`` properties of days, seconds and microseconds have been removed, leveraging their built-in Python versions instead (:issue:`18242`)
+- ``Series`` construction will reduce the number of copies made of the input data in certain cases (:issue:`17449`)
+- Improved performance of :func:`Series.dt.date` and :func:`DatetimeIndex.date` (:issue:`18058`)
+- Improved performance of :func:`Series.dt.time` and :func:`DatetimeIndex.time` (:issue:`18461`)
+- Improved performance of :func:`IntervalIndex.symmetric_difference` (:issue:`18475`)
+- Improved performance of ``DatetimeIndex`` and ``Series`` arithmetic operations with Business-Month and Business-Quarter frequencies (:issue:`18489`)
+- :func:`Series` / :func:`DataFrame` tab completion limits to 100 values, for better performance. (:issue:`18587`)
+- Improved performance of :func:`DataFrame.median` with ``axis=1`` when bottleneck is not installed (:issue:`16468`)
+- Improved performance of :func:`MultiIndex.get_loc` for large indexes, at the cost of a reduction in performance for small ones (:issue:`18519`)
+- Improved performance of :func:`MultiIndex.remove_unused_levels` when there are no unused levels, at the cost of a reduction in performance when there are (:issue:`19289`)
+- Improved performance of :func:`Index.get_loc` for non-unique indexes (:issue:`19478`)
+- Improved performance of pairwise ``.rolling()`` and ``.expanding()`` with ``.cov()`` and ``.corr()`` operations (:issue:`17917`)
+- Improved performance of :func:`.GroupBy.rank` (:issue:`15779`)
+- Improved performance of variable ``.rolling()`` on ``.min()`` and ``.max()`` (:issue:`19521`)
+- Improved performance of :func:`.GroupBy.ffill` and :func:`.GroupBy.bfill` (:issue:`11296`)
+- Improved performance of :func:`.GroupBy.any` and :func:`.GroupBy.all` (:issue:`15435`)
+- Improved performance of :func:`.GroupBy.pct_change` (:issue:`19165`)
+- Improved performance of :func:`Series.isin` in the case of categorical dtypes (:issue:`20003`)
+- Improved performance of ``getattr(Series, attr)`` when the Series has certain index types. This manifested in slow printing of large Series with a ``DatetimeIndex`` (:issue:`19764`)
+- Fixed a performance regression for :func:`GroupBy.nth` and :func:`GroupBy.last` with some object columns (:issue:`19283`)
+- Improved performance of :func:`.Categorical.from_codes` (:issue:`18501`)
+
+.. _whatsnew_0230.docs:
+
+Documentation changes
+~~~~~~~~~~~~~~~~~~~~~
+
+Thanks to all of the contributors who participated in the pandas Documentation
+Sprint, which took place on March 10th. We had about 500 participants from over
+30 locations across the world. You should notice that many of the
+:ref:`API docstrings <api>` have greatly improved.
+
+There were too many simultaneous contributions to include a release note for each
+improvement, but this `GitHub search`_ should give you an idea of how many docstrings
+were improved.
+
+Special thanks to `Marc Garcia`_ for organizing the sprint. For more information,
+read the `NumFOCUS blogpost`_ recapping the sprint.
+
+.. _GitHub search: https://github.com/pandas-dev/pandas/pulls?utf8=%E2%9C%93&q=is%3Apr+label%3ADocs+created%3A2018-03-10..2018-03-15+
+.. _NumFOCUS blogpost: https://www.numfocus.org/blog/worldwide-pandas-sprint/
+.. _Marc Garcia: https://github.com/datapythonista
+
+- Changed spelling of "numpy" to "NumPy", and "python" to "Python". (:issue:`19017`)
+- Consistency when introducing code samples, using either colon or period.
+  Rewrote some sentences for greater clarity, added more dynamic references
+  to functions, methods and classes.
+  (:issue:`18941`, :issue:`18948`, :issue:`18973`, :issue:`19017`)
+- Added a reference to :func:`DataFrame.assign` in the concatenate section of the merging documentation (:issue:`18665`)
+
+.. _whatsnew_0230.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+
+.. warning::
+
+   A class of bugs were introduced in pandas 0.21 with ``CategoricalDtype`` that
+   affects the correctness of operations like ``merge``, ``concat``, and
+   indexing when comparing multiple unordered ``Categorical`` arrays that have
+   the same categories, but in a different order. We highly recommend upgrading
+   or manually aligning your categories before doing these operations.
+
+- Bug in ``Categorical.equals`` returning the wrong result when comparing two
+  unordered ``Categorical`` arrays with the same categories, but in a different
+  order (:issue:`16603`)
+- Bug in :func:`pandas.api.types.union_categoricals` returning the wrong result
+  when for unordered categoricals with the categories in a different order.
+  This affected :func:`pandas.concat` with Categorical data (:issue:`19096`).
+- Bug in :func:`pandas.merge` returning the wrong result when joining on an
+  unordered ``Categorical`` that had the same categories but in a different
+  order (:issue:`19551`)
+- Bug in :meth:`CategoricalIndex.get_indexer` returning the wrong result when
+  ``target`` was an unordered ``Categorical`` that had the same categories as
+  ``self`` but in a different order (:issue:`19551`)
+- Bug in :meth:`Index.astype` with a categorical dtype where the resultant index is not converted to a :class:`CategoricalIndex` for all types of index (:issue:`18630`)
+- Bug in :meth:`Series.astype` and ``Categorical.astype()`` where an existing categorical data does not get updated (:issue:`10696`, :issue:`18593`)
+- Bug in :meth:`Series.str.split` with ``expand=True`` incorrectly raising an IndexError on empty strings (:issue:`20002`).
+- Bug in :class:`Index` constructor with ``dtype=CategoricalDtype(...)`` where ``categories`` and ``ordered`` are not maintained (:issue:`19032`)
+- Bug in :class:`Series` constructor with scalar and ``dtype=CategoricalDtype(...)`` where ``categories`` and ``ordered`` are not maintained (:issue:`19565`)
+- Bug in ``Categorical.__iter__`` not converting to Python types (:issue:`19909`)
+- Bug in :func:`pandas.factorize` returning the unique codes for the ``uniques``. This now returns a ``Categorical`` with the same dtype as the input (:issue:`19721`)
+- Bug in :func:`pandas.factorize` including an item for missing values in the ``uniques`` return value (:issue:`19721`)
+- Bug in :meth:`Series.take` with categorical data interpreting ``-1`` in ``indices`` as missing value markers, rather than the last element of the Series (:issue:`20664`)
+
+Datetimelike
+^^^^^^^^^^^^
+
+- Bug in :func:`Series.__sub__` subtracting a non-nanosecond ``np.datetime64`` object from a ``Series`` gave incorrect results (:issue:`7996`)
+- Bug in :class:`DatetimeIndex`, :class:`TimedeltaIndex` addition and subtraction of zero-dimensional integer arrays gave incorrect results (:issue:`19012`)
+- Bug in :class:`DatetimeIndex` and :class:`TimedeltaIndex` where adding or subtracting an array-like of ``DateOffset`` objects either raised (``np.array``, ``pd.Index``) or broadcast incorrectly (``pd.Series``) (:issue:`18849`)
+- Bug in :func:`Series.__add__` adding Series with dtype ``timedelta64[ns]`` to a timezone-aware ``DatetimeIndex`` incorrectly dropped timezone information (:issue:`13905`)
+- Adding a ``Period`` object to a ``datetime`` or ``Timestamp`` object will now correctly raise a ``TypeError`` (:issue:`17983`)
+- Bug in :class:`Timestamp` where comparison with an array of ``Timestamp`` objects would result in a ``RecursionError`` (:issue:`15183`)
+- Bug in :class:`Series` floor-division where operating on a scalar ``timedelta`` raises an exception (:issue:`18846`)
+- Bug in :class:`DatetimeIndex` where the repr was not showing high-precision time values at the end of a day (e.g., 23:59:59.999999999) (:issue:`19030`)
+- Bug in ``.astype()`` to non-ns timedelta units would hold the incorrect dtype (:issue:`19176`, :issue:`19223`, :issue:`12425`)
+- Bug in subtracting :class:`Series` from ``NaT`` incorrectly returning ``NaT`` (:issue:`19158`)
+- Bug in :func:`Series.truncate` which raises ``TypeError`` with a monotonic ``PeriodIndex`` (:issue:`17717`)
+- Bug in :func:`~DataFrame.pct_change` using ``periods`` and ``freq`` returned different length outputs (:issue:`7292`)
+- Bug in comparison of :class:`DatetimeIndex` against ``None`` or ``datetime.date`` objects raising ``TypeError`` for ``==`` and ``!=`` comparisons instead of all-``False`` and all-``True``, respectively (:issue:`19301`)
+- Bug in :class:`Timestamp` and :func:`to_datetime` where a string representing a barely out-of-bounds timestamp would be incorrectly rounded down instead of raising ``OutOfBoundsDatetime`` (:issue:`19382`)
+- Bug in :func:`Timestamp.floor` :func:`DatetimeIndex.floor` where time stamps far in the future and past were not rounded correctly (:issue:`19206`)
+- Bug in :func:`to_datetime` where passing an out-of-bounds datetime with ``errors='coerce'`` and ``utc=True`` would raise ``OutOfBoundsDatetime`` instead of parsing to ``NaT`` (:issue:`19612`)
+- Bug in :class:`DatetimeIndex` and :class:`TimedeltaIndex` addition and subtraction where name of the returned object was not always set consistently. (:issue:`19744`)
+- Bug in :class:`DatetimeIndex` and :class:`TimedeltaIndex` addition and subtraction where operations with numpy arrays raised ``TypeError`` (:issue:`19847`)
+- Bug in :class:`DatetimeIndex` and :class:`TimedeltaIndex` where setting the ``freq`` attribute was not fully supported (:issue:`20678`)
+
+Timedelta
+^^^^^^^^^
+
+- Bug in :func:`Timedelta.__mul__` where multiplying by ``NaT`` returned ``NaT`` instead of raising a ``TypeError`` (:issue:`19819`)
+- Bug in :class:`Series` with ``dtype='timedelta64[ns]'`` where addition or subtraction of ``TimedeltaIndex`` had results cast to ``dtype='int64'`` (:issue:`17250`)
+- Bug in :class:`Series` with ``dtype='timedelta64[ns]'`` where addition or subtraction of ``TimedeltaIndex`` could return a ``Series`` with an incorrect name (:issue:`19043`)
+- Bug in :func:`Timedelta.__floordiv__` and :func:`Timedelta.__rfloordiv__` dividing by many incompatible numpy objects was incorrectly allowed (:issue:`18846`)
+- Bug where dividing a scalar timedelta-like object with :class:`TimedeltaIndex` performed the reciprocal operation (:issue:`19125`)
+- Bug in :class:`TimedeltaIndex` where division by a ``Series`` would return a ``TimedeltaIndex`` instead of a ``Series`` (:issue:`19042`)
+- Bug in :func:`Timedelta.__add__`, :func:`Timedelta.__sub__` where adding or subtracting a ``np.timedelta64`` object would return another ``np.timedelta64`` instead of a ``Timedelta`` (:issue:`19738`)
+- Bug in :func:`Timedelta.__floordiv__`, :func:`Timedelta.__rfloordiv__` where operating with a ``Tick`` object would raise a ``TypeError`` instead of returning a numeric value (:issue:`19738`)
+- Bug in :func:`Period.asfreq` where periods near ``datetime(1, 1, 1)`` could be converted incorrectly (:issue:`19643`, :issue:`19834`)
+- Bug in :func:`Timedelta.total_seconds` causing precision errors, for example ``Timedelta('30S').total_seconds()==30.000000000000004`` (:issue:`19458`)
+- Bug in :func:`Timedelta.__rmod__` where operating with a ``numpy.timedelta64`` returned a ``timedelta64`` object instead of a ``Timedelta`` (:issue:`19820`)
+- Multiplication of :class:`TimedeltaIndex` by ``TimedeltaIndex`` will now raise ``TypeError`` instead of raising ``ValueError`` in cases of length mismatch (:issue:`19333`)
+- Bug in indexing a :class:`TimedeltaIndex` with a ``np.timedelta64`` object which was raising a ``TypeError`` (:issue:`20393`)
+
+
+Timezones
+^^^^^^^^^
+
+- Bug in creating a ``Series`` from an array that contains both tz-naive and tz-aware values will result in a ``Series`` whose dtype is tz-aware instead of object (:issue:`16406`)
+- Bug in comparison of timezone-aware :class:`DatetimeIndex` against ``NaT`` incorrectly raising ``TypeError`` (:issue:`19276`)
+- Bug in :meth:`DatetimeIndex.astype` when converting between timezone aware dtypes, and converting from timezone aware to naive (:issue:`18951`)
+- Bug in comparing :class:`DatetimeIndex`, which failed to raise ``TypeError`` when attempting to compare timezone-aware and timezone-naive datetimelike objects (:issue:`18162`)
+- Bug in localization of a naive, datetime string in a ``Series`` constructor with a ``datetime64[ns, tz]`` dtype (:issue:`174151`)
+- :func:`Timestamp.replace` will now handle Daylight Savings transitions gracefully (:issue:`18319`)
+- Bug in tz-aware :class:`DatetimeIndex` where addition/subtraction with a :class:`TimedeltaIndex` or array with ``dtype='timedelta64[ns]'`` was incorrect (:issue:`17558`)
+- Bug in :func:`DatetimeIndex.insert` where inserting ``NaT`` into a timezone-aware index incorrectly raised (:issue:`16357`)
+- Bug in :class:`DataFrame` constructor, where tz-aware Datetimeindex and a given column name will result in an empty ``DataFrame`` (:issue:`19157`)
+- Bug in :func:`Timestamp.tz_localize` where localizing a timestamp near the minimum or maximum valid values could overflow and return a timestamp with an incorrect nanosecond value (:issue:`12677`)
+- Bug when iterating over :class:`DatetimeIndex` that was localized with fixed timezone offset that rounded nanosecond precision to microseconds (:issue:`19603`)
+- Bug in :func:`DataFrame.diff` that raised an ``IndexError`` with tz-aware values (:issue:`18578`)
+- Bug in :func:`melt` that converted tz-aware dtypes to tz-naive (:issue:`15785`)
+- Bug in :func:`Dataframe.count` that raised an ``ValueError``, if :func:`Dataframe.dropna` was called for a single column with timezone-aware values. (:issue:`13407`)
+
+Offsets
+^^^^^^^
+
+- Bug in :class:`WeekOfMonth` and :class:`Week` where addition and subtraction did not roll correctly (:issue:`18510`, :issue:`18672`, :issue:`18864`)
+- Bug in :class:`WeekOfMonth` and :class:`LastWeekOfMonth` where default keyword arguments for constructor raised ``ValueError`` (:issue:`19142`)
+- Bug in :class:`FY5253Quarter`, :class:`LastWeekOfMonth` where rollback and rollforward behavior was inconsistent with addition and subtraction behavior (:issue:`18854`)
+- Bug in :class:`FY5253` where ``datetime`` addition and subtraction incremented incorrectly for dates on the year-end but not normalized to midnight (:issue:`18854`)
+- Bug in :class:`FY5253` where date offsets could incorrectly raise an ``AssertionError`` in arithmetic operations (:issue:`14774`)
+
+Numeric
+^^^^^^^
+- Bug in :class:`Series` constructor with an int or float list where specifying ``dtype=str``, ``dtype='str'`` or ``dtype='U'`` failed to convert the data elements to strings (:issue:`16605`)
+- Bug in :class:`Index` multiplication and division methods where operating with a ``Series`` would return an ``Index`` object instead of a ``Series`` object (:issue:`19042`)
+- Bug in the :class:`DataFrame` constructor in which data containing very large positive or very large negative numbers was causing ``OverflowError`` (:issue:`18584`)
+- Bug in :class:`Index` constructor with ``dtype='uint64'`` where int-like floats were not coerced to :class:`UInt64Index` (:issue:`18400`)
+- Bug in :class:`DataFrame` flex arithmetic (e.g. ``df.add(other, fill_value=foo)``) with a ``fill_value`` other than ``None`` failed to raise ``NotImplementedError`` in corner cases where either the frame or ``other`` has length zero (:issue:`19522`)
+- Multiplication and division of numeric-dtyped :class:`Index` objects with timedelta-like scalars returns ``TimedeltaIndex`` instead of raising ``TypeError`` (:issue:`19333`)
+- Bug where ``NaN`` was returned instead of 0 by :func:`Series.pct_change` and :func:`DataFrame.pct_change` when ``fill_method`` is not ``None`` (:issue:`19873`)
+
+Strings
+^^^^^^^
+- Bug in :func:`Series.str.get` with a dictionary in the values and the index not in the keys, raising ``KeyError`` (:issue:`20671`)
+
+
+Indexing
+^^^^^^^^
+
+- Bug in :class:`Index` construction from list of mixed type tuples (:issue:`18505`)
+- Bug in :func:`Index.drop` when passing a list of both tuples and non-tuples (:issue:`18304`)
+- Bug in :func:`DataFrame.drop`, :meth:`Panel.drop`, :meth:`Series.drop`, :meth:`Index.drop` where no ``KeyError`` is raised when dropping a non-existent element from an axis that contains duplicates (:issue:`19186`)
+- Bug in indexing a datetimelike ``Index`` that raised ``ValueError`` instead of ``IndexError`` (:issue:`18386`).
+- :func:`Index.to_series` now accepts ``index`` and ``name`` kwargs (:issue:`18699`)
+- :func:`DatetimeIndex.to_series` now accepts ``index`` and ``name`` kwargs (:issue:`18699`)
+- Bug in indexing non-scalar value from ``Series`` having non-unique ``Index`` will return value flattened (:issue:`17610`)
+- Bug in indexing with iterator containing only missing keys, which raised no error (:issue:`20748`)
+- Fixed inconsistency in ``.ix`` between list and scalar keys when the index has integer dtype and does not include the desired keys (:issue:`20753`)
+- Bug in ``__setitem__`` when indexing a :class:`DataFrame` with a 2-d boolean ndarray (:issue:`18582`)
+- Bug in ``str.extractall`` when there were no matches empty :class:`Index` was returned instead of appropriate :class:`MultiIndex` (:issue:`19034`)
+- Bug in :class:`IntervalIndex` where empty and purely NA data was constructed inconsistently depending on the construction method (:issue:`18421`)
+- Bug in :func:`IntervalIndex.symmetric_difference` where the symmetric difference with a non-``IntervalIndex`` did not raise (:issue:`18475`)
+- Bug in :class:`IntervalIndex` where set operations that returned an empty ``IntervalIndex`` had the wrong dtype (:issue:`19101`)
+- Bug in :meth:`DataFrame.drop_duplicates` where no ``KeyError`` is raised when passing in columns that don't exist on the ``DataFrame`` (:issue:`19726`)
+- Bug in ``Index`` subclasses constructors that ignore unexpected keyword arguments (:issue:`19348`)
+- Bug in :meth:`Index.difference` when taking difference of an ``Index`` with itself (:issue:`20040`)
+- Bug in :meth:`DataFrame.first_valid_index` and :meth:`DataFrame.last_valid_index` in presence of entire rows of NaNs in the middle of values (:issue:`20499`).
+- Bug in :class:`IntervalIndex` where some indexing operations were not supported for overlapping or non-monotonic ``uint64`` data (:issue:`20636`)
+- Bug in ``Series.is_unique`` where extraneous output in stderr is shown if Series contains objects with ``__ne__`` defined (:issue:`20661`)
+- Bug in ``.loc`` assignment with a single-element list-like incorrectly assigns as a list (:issue:`19474`)
+- Bug in partial string indexing on a ``Series/DataFrame`` with a monotonic decreasing ``DatetimeIndex`` (:issue:`19362`)
+- Bug in performing in-place operations on a ``DataFrame`` with a duplicate ``Index`` (:issue:`17105`)
+- Bug in :meth:`IntervalIndex.get_loc` and :meth:`IntervalIndex.get_indexer` when used with an :class:`IntervalIndex` containing a single interval (:issue:`17284`, :issue:`20921`)
+- Bug in ``.loc`` with a ``uint64`` indexer (:issue:`20722`)
+
+MultiIndex
+^^^^^^^^^^
+
+- Bug in :func:`MultiIndex.__contains__` where non-tuple keys would return ``True`` even if they had been dropped (:issue:`19027`)
+- Bug in :func:`MultiIndex.set_labels` which would cause casting (and potentially clipping) of the new labels if the ``level`` argument is not 0 or a list like [0, 1, ... ]  (:issue:`19057`)
+- Bug in :func:`MultiIndex.get_level_values` which would return an invalid index on level of ints with missing values (:issue:`17924`)
+- Bug in :func:`MultiIndex.unique` when called on empty :class:`MultiIndex` (:issue:`20568`)
+- Bug in :func:`MultiIndex.unique` which would not preserve level names (:issue:`20570`)
+- Bug in :func:`MultiIndex.remove_unused_levels` which would fill nan values (:issue:`18417`)
+- Bug in :func:`MultiIndex.from_tuples` which would fail to take zipped tuples in python3 (:issue:`18434`)
+- Bug in :func:`MultiIndex.get_loc` which would fail to automatically cast values between float and int (:issue:`18818`, :issue:`15994`)
+- Bug in :func:`MultiIndex.get_loc` which would cast boolean to integer labels (:issue:`19086`)
+- Bug in :func:`MultiIndex.get_loc` which would fail to locate keys containing ``NaN`` (:issue:`18485`)
+- Bug in :func:`MultiIndex.get_loc` in large :class:`MultiIndex`, would fail when levels had different dtypes (:issue:`18520`)
+- Bug in indexing where nested indexers having only numpy arrays are handled incorrectly (:issue:`19686`)
+
+
+IO
+^^
+
+- :func:`read_html` now rewinds seekable IO objects after parse failure, before attempting to parse with a new parser. If a parser errors and the object is non-seekable, an informative error is raised suggesting the use of a different parser (:issue:`17975`)
+- :meth:`DataFrame.to_html` now has an option to add an id to the leading ``<table>`` tag (:issue:`8496`)
+- Bug in :func:`read_msgpack` with a non existent file is passed in Python 2 (:issue:`15296`)
+- Bug in :func:`read_csv` where a ``MultiIndex`` with duplicate columns was not being mangled appropriately (:issue:`18062`)
+- Bug in :func:`read_csv` where missing values were not being handled properly when ``keep_default_na=False`` with dictionary ``na_values`` (:issue:`19227`)
+- Bug in :func:`read_csv` causing heap corruption on 32-bit, big-endian architectures (:issue:`20785`)
+- Bug in :func:`read_sas` where a file with 0 variables gave an ``AttributeError`` incorrectly. Now it gives an ``EmptyDataError`` (:issue:`18184`)
+- Bug in :func:`DataFrame.to_latex` where pairs of braces meant to serve as invisible placeholders were escaped (:issue:`18667`)
+- Bug in :func:`DataFrame.to_latex` where a ``NaN`` in a ``MultiIndex`` would cause an ``IndexError`` or incorrect output (:issue:`14249`)
+- Bug in :func:`DataFrame.to_latex` where a non-string index-level name would result in an ``AttributeError`` (:issue:`19981`)
+- Bug in :func:`DataFrame.to_latex` where the combination of an index name and the ``index_names=False`` option would result in incorrect output (:issue:`18326`)
+- Bug in :func:`DataFrame.to_latex` where a ``MultiIndex`` with an empty string as its name would result in incorrect output (:issue:`18669`)
+- Bug in :func:`DataFrame.to_latex` where missing space characters caused wrong escaping and produced non-valid latex in some cases (:issue:`20859`)
+- Bug in :func:`read_json` where large numeric values were causing an ``OverflowError`` (:issue:`18842`)
+- Bug in :func:`DataFrame.to_parquet` where an exception was raised if the write destination is S3 (:issue:`19134`)
+- :class:`Interval` now supported in :func:`DataFrame.to_excel` for all Excel file types (:issue:`19242`)
+- :class:`Timedelta` now supported in :func:`DataFrame.to_excel` for all Excel file types (:issue:`19242`, :issue:`9155`, :issue:`19900`)
+- Bug in :meth:`pandas.io.stata.StataReader.value_labels` raising an ``AttributeError`` when called on very old files. Now returns an empty dict (:issue:`19417`)
+- Bug in :func:`read_pickle` when unpickling objects with :class:`TimedeltaIndex` or :class:`Float64Index` created with pandas prior to version 0.20 (:issue:`19939`)
+- Bug in :meth:`pandas.io.json.json_normalize` where sub-records are not properly normalized if any sub-records values are NoneType (:issue:`20030`)
+- Bug in ``usecols`` parameter in :func:`read_csv` where error is not raised correctly when passing a string. (:issue:`20529`)
+- Bug in :func:`HDFStore.keys` when reading a file with a soft link causes exception (:issue:`20523`)
+- Bug in :func:`HDFStore.select_column` where a key which is not a valid store raised an ``AttributeError`` instead of a ``KeyError`` (:issue:`17912`)
+
+Plotting
+^^^^^^^^
+
+- Better error message when attempting to plot but matplotlib is not installed (:issue:`19810`).
+- :func:`DataFrame.plot` now raises a ``ValueError`` when the ``x`` or ``y`` argument is improperly formed (:issue:`18671`)
+- Bug in :func:`DataFrame.plot` when ``x`` and ``y`` arguments given as positions caused incorrect referenced columns for line, bar and area plots (:issue:`20056`)
+- Bug in formatting tick labels with ``datetime.time()`` and fractional seconds (:issue:`18478`).
+- :meth:`Series.plot.kde` has exposed the args ``ind`` and ``bw_method`` in the docstring (:issue:`18461`). The argument ``ind`` may now also be an integer (number of sample points).
+- :func:`DataFrame.plot` now supports multiple columns to the ``y`` argument (:issue:`19699`)
+
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug when grouping by a single column and aggregating with a class like ``list`` or ``tuple`` (:issue:`18079`)
+- Fixed regression in :func:`DataFrame.groupby` which would not emit an error when called with a tuple key not in the index (:issue:`18798`)
+- Bug in :func:`DataFrame.resample` which silently ignored unsupported (or mistyped) options for ``label``, ``closed`` and ``convention`` (:issue:`19303`)
+- Bug in :func:`DataFrame.groupby` where tuples were interpreted as lists of keys rather than as keys (:issue:`17979`, :issue:`18249`)
+- Bug in :func:`DataFrame.groupby` where aggregation by ``first``/``last``/``min``/``max`` was causing timestamps to lose precision (:issue:`19526`)
+- Bug in :func:`DataFrame.transform` where particular aggregation functions were being incorrectly cast to match the dtype(s) of the grouped data (:issue:`19200`)
+- Bug in :func:`DataFrame.groupby` passing the ``on=`` kwarg, and subsequently using ``.apply()`` (:issue:`17813`)
+- Bug in :func:`DataFrame.resample().aggregate <.Resampler.aggregate>` not raising a ``KeyError`` when aggregating a non-existent column (:issue:`16766`, :issue:`19566`)
+- Bug in :func:`DataFrameGroupBy.cumsum` and :func:`DataFrameGroupBy.cumprod` when ``skipna`` was passed (:issue:`19806`)
+- Bug in :func:`DataFrame.resample` that dropped timezone information (:issue:`13238`)
+- Bug in :func:`DataFrame.groupby` where transformations using ``np.all`` and ``np.any`` were raising a ``ValueError`` (:issue:`20653`)
+- Bug in :func:`DataFrame.resample` where ``ffill``, ``bfill``, ``pad``, ``backfill``, ``fillna``, ``interpolate``, and ``asfreq`` were ignoring ``loffset``. (:issue:`20744`)
+- Bug in :func:`DataFrame.groupby` when applying a function that has mixed data types and the user supplied function can fail on the grouping column (:issue:`20949`)
+- Bug in :func:`DataFrameGroupBy.rolling().apply() <.Rolling.apply>` where operations performed against the associated :class:`DataFrameGroupBy` object could impact the inclusion of the grouped item(s) in the result (:issue:`14013`)
+
+Sparse
+^^^^^^
+
+- Bug in which creating a :class:`SparseDataFrame` from a dense ``Series`` or an unsupported type raised an uncontrolled exception (:issue:`19374`)
+- Bug in :class:`SparseDataFrame.to_csv` causing exception (:issue:`19384`)
+- Bug in :class:`SparseSeries.memory_usage` which caused segfault by accessing non sparse elements (:issue:`19368`)
+- Bug in constructing a :class:`SparseArray`: if ``data`` is a scalar and ``index`` is defined it will coerce to ``float64`` regardless of scalar's dtype. (:issue:`19163`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in :func:`DataFrame.merge` where referencing a ``CategoricalIndex`` by name, where the ``by`` kwarg would ``KeyError`` (:issue:`20777`)
+- Bug in :func:`DataFrame.stack` which fails trying to sort mixed type levels under Python 3 (:issue:`18310`)
+- Bug in :func:`DataFrame.unstack` which casts int to float if ``columns`` is a ``MultiIndex`` with unused levels (:issue:`17845`)
+- Bug in :func:`DataFrame.unstack` which raises an error if ``index`` is a ``MultiIndex`` with unused labels on the unstacked level (:issue:`18562`)
+- Fixed construction of a :class:`Series` from a ``dict`` containing ``NaN`` as key (:issue:`18480`)
+- Fixed construction of a :class:`DataFrame` from a ``dict`` containing ``NaN`` as key (:issue:`18455`)
+- Disabled construction of a :class:`Series` where len(index) > len(data) = 1, which previously would broadcast the data item, and now raises a ``ValueError`` (:issue:`18819`)
+- Suppressed error in the construction of a :class:`DataFrame` from a ``dict`` containing scalar values when the corresponding keys are not included in the passed index (:issue:`18600`)
+
+- Fixed (changed from ``object`` to ``float64``) dtype of :class:`DataFrame` initialized with axes, no data, and ``dtype=int`` (:issue:`19646`)
+- Bug in :func:`Series.rank` where ``Series`` containing ``NaT`` modifies the ``Series`` inplace (:issue:`18521`)
+- Bug in :func:`cut` which fails when using readonly arrays (:issue:`18773`)
+- Bug in :func:`DataFrame.pivot_table` which fails when the ``aggfunc`` arg is of type string.  The behavior is now consistent with other methods like ``agg`` and ``apply`` (:issue:`18713`)
+- Bug in :func:`DataFrame.merge` in which merging using ``Index`` objects as vectors raised an Exception (:issue:`19038`)
+- Bug in :func:`DataFrame.stack`, :func:`DataFrame.unstack`, :func:`Series.unstack` which were not returning subclasses (:issue:`15563`)
+- Bug in timezone comparisons, manifesting as a conversion of the index to UTC in ``.concat()`` (:issue:`18523`)
+- Bug in :func:`concat` when concatenating sparse and dense series it returns only a ``SparseDataFrame``. Should be a ``DataFrame``. (:issue:`18914`, :issue:`18686`, and :issue:`16874`)
+- Improved error message for :func:`DataFrame.merge` when there is no common merge key (:issue:`19427`)
+- Bug in :func:`DataFrame.join` which does an ``outer`` instead of a ``left`` join when being called with multiple DataFrames and some have non-unique indices (:issue:`19624`)
+- :func:`Series.rename` now accepts ``axis`` as a kwarg (:issue:`18589`)
+- Bug in :func:`~DataFrame.rename` where an Index of same-length tuples was converted to a MultiIndex (:issue:`19497`)
+- Comparisons between :class:`Series` and :class:`Index` would return a ``Series`` with an incorrect name, ignoring the ``Index``'s name attribute (:issue:`19582`)
+- Bug in :func:`qcut` where datetime and timedelta data with ``NaT`` present raised a ``ValueError`` (:issue:`19768`)
+- Bug in :func:`DataFrame.iterrows`, which would infers strings not compliant to `ISO8601 <https://en.wikipedia.org/wiki/ISO_8601>`_ to datetimes (:issue:`19671`)
+- Bug in :class:`Series` constructor with ``Categorical`` where a ``ValueError`` is not raised when an index of different length is given (:issue:`19342`)
+- Bug in :meth:`DataFrame.astype` where column metadata is lost when converting to categorical or a dictionary of dtypes (:issue:`19920`)
+- Bug in :func:`cut` and :func:`qcut` where timezone information was dropped (:issue:`19872`)
+- Bug in :class:`Series` constructor with a ``dtype=str``, previously raised in some cases (:issue:`19853`)
+- Bug in :func:`get_dummies`, and :func:`select_dtypes`, where duplicate column names caused incorrect behavior (:issue:`20848`)
+- Bug in :func:`isna`, which cannot handle ambiguous typed lists (:issue:`20675`)
+- Bug in :func:`concat` which raises an error when concatenating TZ-aware dataframes and all-NaT dataframes (:issue:`12396`)
+- Bug in :func:`concat` which raises an error when concatenating empty TZ-aware series (:issue:`18447`)
+
+Other
+^^^^^
+
+- Improved error message when attempting to use a Python keyword as an identifier in a ``numexpr`` backed query (:issue:`18221`)
+- Bug in accessing a :func:`pandas.get_option`, which raised ``KeyError`` rather than ``OptionError`` when looking up a non-existent option key in some cases (:issue:`19789`)
+- Bug in :func:`testing.assert_series_equal` and :func:`testing.assert_frame_equal` for Series or DataFrames with differing unicode data (:issue:`20503`)
+
+.. _whatsnew_0.23.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.22.0..v0.23.0
diff --git a/pandas/doc/source/whatsnew/v0.23.1.rst b/pandas/doc/source/whatsnew/v0.23.1.rst
new file mode 100644
index 0000000000000..a98933e7f5969
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.23.1.rst
@@ -0,0 +1,151 @@
+.. _whatsnew_0231:
+
+What's new in 0.23.1 (June 12, 2018)
+------------------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release in the 0.23.x series and includes some small regression fixes
+and bug fixes. We recommend that all users upgrade to this version.
+
+.. warning::
+
+   Starting January 1, 2019, pandas feature releases will support Python 3 only.
+   See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+.. contents:: What's new in v0.23.1
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0231.fixed_regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+**Comparing Series with datetime.date**
+
+We've reverted a 0.23.0 change to comparing a :class:`Series` holding datetimes and a ``datetime.date`` object (:issue:`21152`).
+In pandas 0.22 and earlier, comparing a Series holding datetimes and ``datetime.date`` objects would coerce the ``datetime.date`` to a datetime before comparing.
+This was inconsistent with Python, NumPy, and :class:`DatetimeIndex`, which never consider a datetime and ``datetime.date`` equal.
+
+In 0.23.0, we unified operations between DatetimeIndex and Series, and in the process changed comparisons between a Series of datetimes and ``datetime.date`` without warning.
+
+We've temporarily restored the 0.22.0 behavior, so datetimes and dates may again compare equal, but restore the 0.23.0 behavior in a future release.
+
+To summarize, here's the behavior in 0.22.0, 0.23.0, 0.23.1:
+
+.. code-block:: python
+
+   # 0.22.0... Silently coerce the datetime.date
+   >>> import datetime
+   >>> pd.Series(pd.date_range('2017', periods=2)) == datetime.date(2017, 1, 1)
+   0     True
+   1    False
+   dtype: bool
+
+   # 0.23.0... Do not coerce the datetime.date
+   >>> pd.Series(pd.date_range('2017', periods=2)) == datetime.date(2017, 1, 1)
+   0    False
+   1    False
+   dtype: bool
+
+   # 0.23.1... Coerce the datetime.date with a warning
+   >>> pd.Series(pd.date_range('2017', periods=2)) == datetime.date(2017, 1, 1)
+   /bin/python:1: FutureWarning: Comparing Series of datetimes with 'datetime.date'.  Currently, the
+   'datetime.date' is coerced to a datetime. In the future pandas will
+   not coerce, and the values not compare equal to the 'datetime.date'.
+   To retain the current behavior, convert the 'datetime.date' to a
+   datetime with 'pd.Timestamp'.
+     #!/bin/python3
+   0     True
+   1    False
+   dtype: bool
+
+In addition, ordering comparisons will raise a ``TypeError`` in the future.
+
+**Other fixes**
+
+- Reverted the ability of :func:`~DataFrame.to_sql` to perform multivalue
+  inserts as this caused regression in certain cases (:issue:`21103`).
+  In the future this will be made configurable.
+- Fixed regression in the :attr:`DatetimeIndex.date` and :attr:`DatetimeIndex.time`
+  attributes in case of timezone-aware data: :attr:`DatetimeIndex.time` returned
+  a tz-aware time instead of tz-naive (:issue:`21267`) and :attr:`DatetimeIndex.date`
+  returned incorrect date when the input date has a non-UTC timezone (:issue:`21230`).
+- Fixed regression in :meth:`pandas.io.json.json_normalize` when called with ``None`` values
+  in nested levels in JSON, and to not drop keys with value as ``None`` (:issue:`21158`, :issue:`21356`).
+- Bug in :meth:`~DataFrame.to_csv` causes encoding error when compression and encoding are specified (:issue:`21241`, :issue:`21118`)
+- Bug preventing pandas from being importable with -OO optimization (:issue:`21071`)
+- Bug in :meth:`Categorical.fillna` incorrectly raising a ``TypeError`` when ``value`` the individual categories are iterable and ``value`` is an iterable (:issue:`21097`, :issue:`19788`)
+- Fixed regression in constructors coercing NA values like ``None`` to strings when passing ``dtype=str`` (:issue:`21083`)
+- Regression in :func:`pivot_table` where an ordered ``Categorical`` with missing
+  values for the pivot's ``index`` would give a mis-aligned result (:issue:`21133`)
+- Fixed regression in merging on boolean index/columns (:issue:`21119`).
+
+.. _whatsnew_0231.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improved performance of :meth:`CategoricalIndex.is_monotonic_increasing`, :meth:`CategoricalIndex.is_monotonic_decreasing` and :meth:`CategoricalIndex.is_monotonic` (:issue:`21025`)
+- Improved performance of :meth:`CategoricalIndex.is_unique` (:issue:`21107`)
+
+
+.. _whatsnew_0231.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Groupby/resample/rolling**
+
+- Bug in :func:`DataFrame.agg` where applying multiple aggregation functions to a :class:`DataFrame` with duplicated column names would cause a stack overflow (:issue:`21063`)
+- Bug in :func:`.GroupBy.ffill` and :func:`.GroupBy.bfill` where the fill within a grouping would not always be applied as intended due to the implementations' use of a non-stable sort (:issue:`21207`)
+- Bug in :func:`.GroupBy.rank` where results did not scale to 100% when specifying ``method='dense'`` and ``pct=True``
+- Bug in :func:`pandas.DataFrame.rolling` and :func:`pandas.Series.rolling` which incorrectly accepted a 0 window size rather than raising (:issue:`21286`)
+
+**Data-type specific**
+
+- Bug in :meth:`Series.str.replace` where the method throws ``TypeError`` on Python 3.5.2 (:issue:`21078`)
+- Bug in :class:`Timedelta` where passing a float with a unit would prematurely round the float precision (:issue:`14156`)
+- Bug in :func:`pandas.testing.assert_index_equal` which raised ``AssertionError`` incorrectly, when comparing two :class:`CategoricalIndex` objects with param ``check_categorical=False`` (:issue:`19776`)
+
+**Sparse**
+
+- Bug in :attr:`SparseArray.shape` which previously only returned the shape :attr:`SparseArray.sp_values` (:issue:`21126`)
+
+**Indexing**
+
+- Bug in :meth:`Series.reset_index` where appropriate error was not raised with an invalid level name (:issue:`20925`)
+- Bug in :func:`interval_range` when ``start``/``periods`` or ``end``/``periods`` are specified with float ``start`` or ``end`` (:issue:`21161`)
+- Bug in :meth:`MultiIndex.set_names` where error raised for a ``MultiIndex`` with ``nlevels == 1`` (:issue:`21149`)
+- Bug in :class:`IntervalIndex` constructors where creating an ``IntervalIndex`` from categorical data was not fully supported (:issue:`21243`, :issue:`21253`)
+- Bug in :meth:`MultiIndex.sort_index` which was not guaranteed to sort correctly with ``level=1``; this was also causing data misalignment in particular :meth:`DataFrame.stack` operations (:issue:`20994`, :issue:`20945`, :issue:`21052`)
+
+**Plotting**
+
+- New keywords (sharex, sharey) to turn on/off sharing of x/y-axis by subplots generated with pandas.DataFrame().groupby().boxplot() (:issue:`20968`)
+
+**I/O**
+
+- Bug in IO methods specifying ``compression='zip'`` which produced uncompressed zip archives (:issue:`17778`, :issue:`21144`)
+- Bug in :meth:`DataFrame.to_stata` which prevented exporting DataFrames to buffers and most file-like objects (:issue:`21041`)
+- Bug in :meth:`read_stata` and :class:`StataReader` which did not correctly decode utf-8 strings on Python 3 from Stata 14 files (dta version 118) (:issue:`21244`)
+- Bug in IO JSON :func:`read_json` reading empty JSON schema with ``orient='table'`` back to :class:`DataFrame` caused an error (:issue:`21287`)
+
+**Reshaping**
+
+- Bug in :func:`concat` where error was raised in concatenating :class:`Series` with numpy scalar and tuple names (:issue:`21015`)
+- Bug in :func:`concat` warning message providing the wrong guidance for future behavior (:issue:`21101`)
+
+**Other**
+
+- Tab completion on :class:`Index` in IPython no longer outputs deprecation warnings (:issue:`21125`)
+- Bug preventing pandas being used on Windows without C++ redistributable installed (:issue:`21106`)
+
+.. _whatsnew_0.23.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.23.0..v0.23.1
diff --git a/pandas/doc/source/whatsnew/v0.23.2.rst b/pandas/doc/source/whatsnew/v0.23.2.rst
new file mode 100644
index 0000000000000..99650e8291d3d
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.23.2.rst
@@ -0,0 +1,123 @@
+.. _whatsnew_0232:
+
+What's new in 0.23.2 (July 5, 2018)
+-----------------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release in the 0.23.x series and includes some small regression fixes
+and bug fixes. We recommend that all users upgrade to this version.
+
+.. note::
+
+   pandas 0.23.2 is first pandas release that's compatible with
+   Python 3.7 (:issue:`20552`)
+
+.. warning::
+
+   Starting January 1, 2019, pandas feature releases will support Python 3 only.
+   See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+.. contents:: What's new in v0.23.2
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0232.enhancements:
+
+Logical reductions over entire DataFrame
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:meth:`DataFrame.all` and :meth:`DataFrame.any` now accept ``axis=None`` to reduce over all axes to a scalar (:issue:`19976`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2], "B": [True, False]})
+   df.all(axis=None)
+
+
+This also provides compatibility with NumPy 1.15, which now dispatches to ``DataFrame.all``.
+With NumPy 1.15 and pandas 0.23.1 or earlier, :func:`numpy.all` will no longer reduce over every axis:
+
+.. code-block:: python
+
+   >>> # NumPy 1.15, pandas 0.23.1
+   >>> np.any(pd.DataFrame({"A": [False], "B": [False]}))
+   A    False
+   B    False
+   dtype: bool
+
+With pandas 0.23.2, that will correctly return False, as it did with NumPy < 1.15.
+
+.. ipython:: python
+
+   np.any(pd.DataFrame({"A": [False], "B": [False]}))
+
+
+.. _whatsnew_0232.fixed_regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`to_csv` when handling file-like object incorrectly (:issue:`21471`)
+- Re-allowed duplicate level names of a ``MultiIndex``. Accessing a level that has a duplicate name by name still raises an error (:issue:`19029`).
+- Bug in both :meth:`DataFrame.first_valid_index` and :meth:`Series.first_valid_index` raised for a row index having duplicate values (:issue:`21441`)
+- Fixed printing of DataFrames with hierarchical columns with long names (:issue:`21180`)
+- Fixed regression in :meth:`~DataFrame.reindex` and :meth:`~DataFrame.groupby`
+  with a MultiIndex or multiple keys that contains categorical datetime-like values (:issue:`21390`).
+- Fixed regression in unary negative operations with object dtype (:issue:`21380`)
+- Bug in :meth:`Timestamp.ceil` and :meth:`Timestamp.floor` when timestamp is a multiple of the rounding frequency (:issue:`21262`)
+- Fixed regression in :func:`to_clipboard` that defaulted to copying dataframes with space delimited instead of tab delimited (:issue:`21104`)
+
+
+Build changes
+~~~~~~~~~~~~~
+
+- The source and binary distributions no longer include test data files, resulting in smaller download sizes. Tests relying on these data files will be skipped when using ``pandas.test()``. (:issue:`19320`)
+
+.. _whatsnew_0232.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Conversion**
+
+- Bug in constructing :class:`Index` with an iterator or generator (:issue:`21470`)
+- Bug in :meth:`Series.nlargest` for signed and unsigned integer dtypes when the minimum value is present (:issue:`21426`)
+
+**Indexing**
+
+- Bug in :meth:`Index.get_indexer_non_unique` with categorical key (:issue:`21448`)
+- Bug in comparison operations for :class:`MultiIndex` where error was raised on equality / inequality comparison involving a MultiIndex with ``nlevels == 1`` (:issue:`21149`)
+- Bug in :meth:`DataFrame.drop` behaviour is not consistent for unique and non-unique indexes (:issue:`21494`)
+- Bug in :func:`DataFrame.duplicated` with a large number of columns causing a 'maximum recursion depth exceeded' (:issue:`21524`).
+
+**I/O**
+
+- Bug in :func:`read_csv` that caused it to incorrectly raise an error when ``nrows=0``, ``low_memory=True``, and ``index_col`` was not ``None`` (:issue:`21141`)
+- Bug in :func:`json_normalize` when formatting the ``record_prefix`` with integer columns (:issue:`21536`)
+
+**Categorical**
+
+- Bug in rendering :class:`Series` with ``Categorical`` dtype in rare conditions under Python 2.7 (:issue:`21002`)
+
+**Timezones**
+
+- Bug in :class:`Timestamp` and :class:`DatetimeIndex` where passing a :class:`Timestamp` localized after a DST transition would return a datetime before the DST transition (:issue:`20854`)
+- Bug in comparing :class:`DataFrame` with tz-aware :class:`DatetimeIndex` columns with a DST transition that raised a ``KeyError`` (:issue:`19970`)
+- Bug in :meth:`DatetimeIndex.shift` where an ``AssertionError`` would raise when shifting across DST (:issue:`8616`)
+- Bug in :class:`Timestamp` constructor where passing an invalid timezone offset designator (``Z``) would not raise a ``ValueError`` (:issue:`8910`)
+- Bug in :meth:`Timestamp.replace` where replacing at a DST boundary would retain an incorrect offset (:issue:`7825`)
+- Bug in :meth:`DatetimeIndex.reindex` when reindexing a tz-naive and tz-aware :class:`DatetimeIndex` (:issue:`8306`)
+- Bug in :meth:`DatetimeIndex.resample` when downsampling across a DST boundary (:issue:`8531`)
+
+**Timedelta**
+
+- Bug in :class:`Timedelta` where non-zero timedeltas shorter than 1 microsecond were considered False (:issue:`21484`)
+
+.. _whatsnew_0.23.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.23.1..v0.23.2
diff --git a/pandas/doc/source/whatsnew/v0.23.3.rst b/pandas/doc/source/whatsnew/v0.23.3.rst
new file mode 100644
index 0000000000000..bb8862a89b003
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.23.3.rst
@@ -0,0 +1,16 @@
+.. _whatsnew_0233:
+
+What's new in 0.23.3 (July 7, 2018)
+-----------------------------------
+
+{{ header }}
+
+This release fixes a build issue with the sdist for Python 3.7 (:issue:`21785`)
+There are no other changes.
+
+.. _whatsnew_0.23.3.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.23.2..v0.23.3
diff --git a/pandas/doc/source/whatsnew/v0.23.4.rst b/pandas/doc/source/whatsnew/v0.23.4.rst
new file mode 100644
index 0000000000000..eadac6f569926
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.23.4.rst
@@ -0,0 +1,47 @@
+.. _whatsnew_0234:
+
+What's new in 0.23.4 (August 3, 2018)
+-------------------------------------
+
+{{ header }}
+
+
+This is a minor bug-fix release in the 0.23.x series and includes some small regression fixes
+and bug fixes. We recommend that all users upgrade to this version.
+
+.. warning::
+
+   Starting January 1, 2019, pandas feature releases will support Python 3 only.
+   See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+.. contents:: What's new in v0.23.4
+    :local:
+    :backlinks: none
+
+.. _whatsnew_0234.fixed_regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Python 3.7 with Windows gave all missing values for rolling variance calculations (:issue:`21813`)
+
+.. _whatsnew_0234.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Groupby/resample/rolling**
+
+- Bug where calling :func:`DataFrameGroupBy.agg` with a list of functions including ``ohlc`` as the non-initial element would raise a ``ValueError`` (:issue:`21716`)
+- Bug in ``roll_quantile`` caused a memory leak when calling ``.rolling(...).quantile(q)`` with ``q`` in (0,1) (:issue:`21965`)
+
+**Missing**
+
+- Bug in :func:`Series.clip` and :func:`DataFrame.clip` cannot accept list-like threshold containing ``NaN`` (:issue:`19992`)
+
+.. _whatsnew_0.23.4.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.23.3..v0.23.4
diff --git a/pandas/doc/source/whatsnew/v0.24.0.rst b/pandas/doc/source/whatsnew/v0.24.0.rst
new file mode 100644
index 0000000000000..60e77a8c5d8c5
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.24.0.rst
@@ -0,0 +1,1963 @@
+.. _whatsnew_0240:
+
+What's new in 0.24.0 (January 25, 2019)
+---------------------------------------
+
+.. warning::
+
+   The 0.24.x series of releases will be the last to support Python 2. Future feature
+   releases will support Python 3 only. See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more
+   details.
+
+{{ header }}
+
+This is a major release from 0.23.4 and includes a number of API changes, new
+features, enhancements, and performance improvements along with a large number
+of bug fixes.
+
+Highlights include:
+
+* :ref:`Optional Integer NA Support <whatsnew_0240.enhancements.intna>`
+* :ref:`New APIs for accessing the array backing a Series or Index <whatsnew_0240.values_api>`
+* :ref:`A new top-level method for creating arrays <whatsnew_0240.enhancements.array>`
+* :ref:`Store Interval and Period data in a Series or DataFrame <whatsnew_0240.enhancements.interval>`
+* :ref:`Support for joining on two MultiIndexes <whatsnew_0240.enhancements.join_with_two_multiindexes>`
+
+
+Check the :ref:`API Changes <whatsnew_0240.api_breaking>` and :ref:`deprecations <whatsnew_0240.deprecations>` before updating.
+
+These are the changes in pandas 0.24.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_0240.enhancements.intna:
+
+Optional integer NA support
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has gained the ability to hold integer dtypes with missing values. This long requested feature is enabled through the use of :ref:`extension types <extending.extension-types>`.
+
+.. note::
+
+   IntegerArray is currently experimental. Its API or implementation may
+   change without warning.
+
+We can construct a ``Series`` with the specified dtype. The dtype string ``Int64`` is a pandas ``ExtensionDtype``. Specifying a list or array using the traditional missing value
+marker of ``np.nan`` will infer to integer dtype. The display of the ``Series`` will also use the ``NaN`` to indicate missing values in string outputs. (:issue:`20700`, :issue:`20747`, :issue:`22441`, :issue:`21789`, :issue:`22346`)
+
+.. ipython:: python
+
+   s = pd.Series([1, 2, np.nan], dtype='Int64')
+   s
+
+
+Operations on these dtypes will propagate ``NaN`` as other pandas operations.
+
+.. ipython:: python
+
+   # arithmetic
+   s + 1
+
+   # comparison
+   s == 1
+
+   # indexing
+   s.iloc[1:3]
+
+   # operate with other dtypes
+   s + s.iloc[1:3].astype('Int8')
+
+   # coerce when needed
+   s + 0.01
+
+These dtypes can operate as part of a ``DataFrame``.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'A': s, 'B': [1, 1, 3], 'C': list('aab')})
+   df
+   df.dtypes
+
+
+These dtypes can be merged, reshaped, and casted.
+
+.. ipython:: python
+
+   pd.concat([df[['A']], df[['B', 'C']]], axis=1).dtypes
+   df['A'].astype(float)
+
+Reduction and groupby operations such as ``sum`` work.
+
+.. ipython:: python
+
+   df.sum()
+   df.groupby('B').A.sum()
+
+.. warning::
+
+   The Integer NA support currently uses the capitalized dtype version, e.g. ``Int8`` as compared to the traditional ``int8``. This may be changed at a future date.
+
+See :ref:`integer_na` for more.
+
+
+.. _whatsnew_0240.values_api:
+
+Accessing the values in a Series or Index
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:attr:`Series.array` and :attr:`Index.array` have been added for extracting the array backing a
+``Series`` or ``Index``. (:issue:`19954`, :issue:`23623`)
+
+.. ipython:: python
+
+   idx = pd.period_range('2000', periods=4)
+   idx.array
+   pd.Series(idx).array
+
+Historically, this would have been done with ``series.values``, but with
+``.values`` it was unclear whether the returned value would be the actual array,
+some transformation of it, or one of pandas custom arrays (like
+``Categorical``). For example, with :class:`PeriodIndex`, ``.values`` generates
+a new ndarray of period objects each time.
+
+.. ipython:: python
+
+   idx.values
+   id(idx.values)
+   id(idx.values)
+
+If you need an actual NumPy array, use :meth:`Series.to_numpy` or :meth:`Index.to_numpy`.
+
+.. ipython:: python
+
+   idx.to_numpy()
+   pd.Series(idx).to_numpy()
+
+For Series and Indexes backed by normal NumPy arrays, :attr:`Series.array` will return a
+new :class:`arrays.PandasArray`, which is a thin (no-copy) wrapper around a
+:class:`numpy.ndarray`. :class:`~arrays.PandasArray` isn't especially useful on its own,
+but it does provide the same interface as any extension array defined in pandas or by
+a third-party library.
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3])
+   ser.array
+   ser.to_numpy()
+
+We haven't removed or deprecated :attr:`Series.values` or :attr:`DataFrame.values`, but we
+highly recommend and using ``.array`` or ``.to_numpy()`` instead.
+
+See :ref:`Dtypes <basics.dtypes>` and :ref:`Attributes and Underlying Data <basics.attrs>` for more.
+
+
+.. _whatsnew_0240.enhancements.array:
+
+``pandas.array``: a new top-level method for creating arrays
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A new top-level method :func:`array` has been added for creating 1-dimensional arrays (:issue:`22860`).
+This can be used to create any :ref:`extension array <extending.extension-types>`, including
+extension arrays registered by `3rd party libraries <https://pandas.pydata.org/community/ecosystem.html>`_.
+See the :ref:`dtypes docs <basics.dtypes>` for more on extension arrays.
+
+.. ipython:: python
+
+   pd.array([1, 2, np.nan], dtype='Int64')
+   pd.array(['a', 'b', 'c'], dtype='category')
+
+Passing data for which there isn't dedicated extension type (e.g. float, integer, etc.)
+will return a new :class:`arrays.PandasArray`, which is just a thin (no-copy)
+wrapper around a :class:`numpy.ndarray` that satisfies the pandas extension array interface.
+
+.. ipython:: python
+
+   pd.array([1, 2, 3])
+
+On their own, a :class:`~arrays.PandasArray` isn't a very useful object.
+But if you need write low-level code that works generically for any
+:class:`~pandas.api.extensions.ExtensionArray`, :class:`~arrays.PandasArray`
+satisfies that need.
+
+Notice that by default, if no ``dtype`` is specified, the dtype of the returned
+array is inferred from the data. In particular, note that the first example of
+``[1, 2, np.nan]`` would have returned a floating-point array, since ``NaN``
+is a float.
+
+.. ipython:: python
+
+   pd.array([1, 2, np.nan])
+
+
+.. _whatsnew_0240.enhancements.interval:
+
+Storing Interval and Period data in Series and DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Interval` and :class:`Period` data may now be stored in a :class:`Series` or :class:`DataFrame`, in addition to an
+:class:`IntervalIndex` and :class:`PeriodIndex` like previously (:issue:`19453`, :issue:`22862`).
+
+.. ipython:: python
+
+   ser = pd.Series(pd.interval_range(0, 5))
+   ser
+   ser.dtype
+
+For periods:
+
+.. ipython:: python
+
+   pser = pd.Series(pd.period_range("2000", freq="D", periods=5))
+   pser
+   pser.dtype
+
+Previously, these would be cast to a NumPy array with object dtype. In general,
+this should result in better performance when storing an array of intervals or periods
+in a :class:`Series` or column of a :class:`DataFrame`.
+
+Use :attr:`Series.array` to extract the underlying array of intervals or periods
+from the ``Series``:
+
+.. ipython:: python
+
+   ser.array
+   pser.array
+
+These return an instance of :class:`arrays.IntervalArray` or :class:`arrays.PeriodArray`,
+the new extension arrays that back interval and period data.
+
+.. warning::
+
+   For backwards compatibility, :attr:`Series.values` continues to return
+   a NumPy array of objects for Interval and Period data. We recommend
+   using :attr:`Series.array` when you need the array of data stored in the
+   ``Series``, and :meth:`Series.to_numpy` when you know you need a NumPy array.
+
+   See :ref:`Dtypes <basics.dtypes>` and :ref:`Attributes and Underlying Data <basics.attrs>`
+   for more.
+
+
+.. _whatsnew_0240.enhancements.join_with_two_multiindexes:
+
+Joining with two multi-indexes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`DataFrame.merge` and :func:`DataFrame.join` can now be used to join multi-indexed ``Dataframe`` instances on the overlapping index levels (:issue:`6360`)
+
+See the :ref:`Merge, join, and concatenate
+<merging.Join_with_two_multi_indexes>` documentation section.
+
+.. ipython:: python
+
+   index_left = pd.MultiIndex.from_tuples([('K0', 'X0'), ('K0', 'X1'),
+                                          ('K1', 'X2')],
+                                          names=['key', 'X'])
+
+   left = pd.DataFrame({'A': ['A0', 'A1', 'A2'],
+                        'B': ['B0', 'B1', 'B2']}, index=index_left)
+
+   index_right = pd.MultiIndex.from_tuples([('K0', 'Y0'), ('K1', 'Y1'),
+                                           ('K2', 'Y2'), ('K2', 'Y3')],
+                                           names=['key', 'Y'])
+
+   right = pd.DataFrame({'C': ['C0', 'C1', 'C2', 'C3'],
+                         'D': ['D0', 'D1', 'D2', 'D3']}, index=index_right)
+
+   left.join(right)
+
+For earlier versions this can be done using the following.
+
+.. ipython:: python
+
+   pd.merge(left.reset_index(), right.reset_index(),
+            on=['key'], how='inner').set_index(['key', 'X', 'Y'])
+
+.. _whatsnew_0240.enhancements.read_html:
+
+Function ``read_html`` enhancements
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`read_html` previously ignored ``colspan`` and ``rowspan`` attributes.
+Now it understands them, treating them as sequences of cells with the same
+value. (:issue:`17054`)
+
+.. ipython:: python
+
+    from io import StringIO
+    result = pd.read_html(StringIO("""
+      <table>
+        <thead>
+          <tr>
+            <th>A</th><th>B</th><th>C</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr>
+            <td colspan="2">1</td><td>2</td>
+          </tr>
+        </tbody>
+      </table>"""))
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [13]: result
+    Out [13]:
+    [   A  B   C
+     0  1  2 NaN]
+
+*New behavior*:
+
+.. ipython:: python
+
+    result
+
+
+New ``Styler.pipe()`` method
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+The :class:`~pandas.io.formats.style.Styler` class has gained a
+:meth:`~pandas.io.formats.style.Styler.pipe` method.  This provides a
+convenient way to apply users' predefined styling functions, and can help reduce
+"boilerplate" when using DataFrame styling functionality repeatedly within a notebook. (:issue:`23229`)
+
+.. ipython:: python
+
+    df = pd.DataFrame({'N': [1250, 1500, 1750], 'X': [0.25, 0.35, 0.50]})
+
+    def format_and_align(styler):
+        return (styler.format({'N': '{:,}', 'X': '{:.1%}'})
+                      .set_properties(**{'text-align': 'right'}))
+
+    df.style.pipe(format_and_align).set_caption('Summary of results.')
+
+Similar methods already exist for other classes in pandas, including :meth:`DataFrame.pipe`,
+:meth:`GroupBy.pipe() <.GroupBy.pipe>`, and :meth:`Resampler.pipe() <.Resampler.pipe>`.
+
+.. _whatsnew_0240.enhancements.rename_axis:
+
+Renaming names in a MultiIndex
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`DataFrame.rename_axis` now supports ``index`` and ``columns`` arguments
+and :func:`Series.rename_axis` supports ``index`` argument (:issue:`19978`).
+
+This change allows a dictionary to be passed so that some of the names
+of a ``MultiIndex`` can be changed.
+
+Example:
+
+.. ipython:: python
+
+    mi = pd.MultiIndex.from_product([list('AB'), list('CD'), list('EF')],
+                                    names=['AB', 'CD', 'EF'])
+    df = pd.DataFrame(list(range(len(mi))), index=mi, columns=['N'])
+    df
+    df.rename_axis(index={'CD': 'New'})
+
+See the :ref:`Advanced documentation on renaming<advanced.index_names>` for more details.
+
+.. _whatsnew_0240.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- :func:`merge` now directly allows merge between objects of type ``DataFrame`` and named ``Series``, without the need to convert the ``Series`` object into a ``DataFrame`` beforehand (:issue:`21220`)
+- ``ExcelWriter`` now accepts ``mode`` as a keyword argument, enabling append to existing workbooks when using the ``openpyxl`` engine (:issue:`3441`)
+- ``FrozenList`` has gained the ``.union()`` and ``.difference()`` methods. This functionality greatly simplifies groupby's that rely on explicitly excluding certain columns. See :ref:`Splitting an object into groups <groupby.split>` for more information (:issue:`15475`, :issue:`15506`).
+- :func:`DataFrame.to_parquet` now accepts ``index`` as an argument, allowing
+  the user to override the engine's default behavior to include or omit the
+  dataframe's indexes from the resulting Parquet file. (:issue:`20768`)
+- :func:`read_feather` now accepts ``columns`` as an argument, allowing the user to specify which columns should be read. (:issue:`24025`)
+- :meth:`DataFrame.corr` and :meth:`Series.corr` now accept a callable for generic calculation methods of correlation, e.g. histogram intersection (:issue:`22684`)
+- :func:`DataFrame.to_string` now accepts ``decimal`` as an argument, allowing the user to specify which decimal separator should be used in the output. (:issue:`23614`)
+- :func:`DataFrame.to_html` now accepts ``render_links`` as an argument, allowing the user to generate HTML with links to any URLs that appear in the DataFrame.
+  See the :ref:`section on writing HTML <io.html>` in the IO docs for example usage. (:issue:`2679`)
+- :func:`pandas.read_csv` now supports pandas extension types as an argument to ``dtype``, allowing the user to use pandas extension types when reading CSVs. (:issue:`23228`)
+- The :meth:`~DataFrame.shift` method now accepts ``fill_value`` as an argument, allowing the user to specify a value which will be used instead of NA/NaT in the empty periods. (:issue:`15486`)
+- :func:`to_datetime` now supports the ``%Z`` and ``%z`` directive when passed into ``format`` (:issue:`13486`)
+- :func:`Series.mode` and :func:`DataFrame.mode` now support the ``dropna`` parameter which can be used to specify whether ``NaN``/``NaT`` values should be considered (:issue:`17534`)
+- :func:`DataFrame.to_csv` and :func:`Series.to_csv` now support the ``compression`` keyword when a file handle is passed. (:issue:`21227`)
+- :meth:`Index.droplevel` is now implemented also for flat indexes, for compatibility with :class:`MultiIndex` (:issue:`21115`)
+- :meth:`Series.droplevel` and :meth:`DataFrame.droplevel` are now implemented (:issue:`20342`)
+- Added support for reading from/writing to Google Cloud Storage via the ``gcsfs`` library (:issue:`19454`, :issue:`23094`)
+- :func:`DataFrame.to_gbq` and :func:`read_gbq` signature and documentation updated to
+  reflect changes from the `pandas-gbq library version 0.8.0
+  <https://pandas-gbq.readthedocs.io/en/latest/changelog.html#changelog-0-8-0>`__.
+  Adds a ``credentials`` argument, which enables the use of any kind of
+  `google-auth credentials
+  <https://google-auth.readthedocs.io/en/latest/>`__. (:issue:`21627`,
+  :issue:`22557`, :issue:`23662`)
+- New method :meth:`HDFStore.walk` will recursively walk the group hierarchy of an HDF5 file (:issue:`10932`)
+- :func:`read_html` copies cell data across ``colspan`` and ``rowspan``, and it treats all-``th`` table rows as headers if ``header`` kwarg is not given and there is no ``thead`` (:issue:`17054`)
+- :meth:`Series.nlargest`, :meth:`Series.nsmallest`, :meth:`DataFrame.nlargest`, and :meth:`DataFrame.nsmallest` now accept the value ``"all"`` for the ``keep`` argument. This keeps all ties for the nth largest/smallest value (:issue:`16818`)
+- :class:`IntervalIndex` has gained the :meth:`~IntervalIndex.set_closed` method to change the existing ``closed`` value (:issue:`21670`)
+- :func:`~DataFrame.to_csv`, :func:`~Series.to_csv`, :func:`~DataFrame.to_json`, and :func:`~Series.to_json` now support ``compression='infer'`` to infer compression based on filename extension (:issue:`15008`).
+  The default compression for ``to_csv``, ``to_json``, and ``to_pickle`` methods has been updated to ``'infer'`` (:issue:`22004`).
+- :meth:`DataFrame.to_sql` now supports writing ``TIMESTAMP WITH TIME ZONE`` types for supported databases. For databases that don't support timezones, datetime data will be stored as timezone unaware local timestamps. See the :ref:`io.sql_datetime_data` for implications (:issue:`9086`).
+- :func:`to_timedelta` now supports iso-formatted timedelta strings (:issue:`21877`)
+- :class:`Series` and :class:`DataFrame` now support :class:`Iterable` objects in the constructor (:issue:`2193`)
+- :class:`DatetimeIndex` has gained the :attr:`DatetimeIndex.timetz` attribute. This returns the local time with timezone information. (:issue:`21358`)
+- :meth:`~Timestamp.round`, :meth:`~Timestamp.ceil`, and :meth:`~Timestamp.floor` for :class:`DatetimeIndex` and :class:`Timestamp`
+  now support an ``ambiguous`` argument for handling datetimes that are rounded to ambiguous times (:issue:`18946`)
+  and a ``nonexistent`` argument for handling datetimes that are rounded to nonexistent times. See :ref:`timeseries.timezone_nonexistent` (:issue:`22647`)
+- The result of :meth:`~DataFrame.resample` is now iterable similar to ``groupby()`` (:issue:`15314`).
+- :meth:`Series.resample` and :meth:`DataFrame.resample` have gained the :meth:`.Resampler.quantile` (:issue:`15023`).
+- :meth:`DataFrame.resample` and :meth:`Series.resample` with a :class:`PeriodIndex` will now respect the ``base`` argument in the same fashion as with a :class:`DatetimeIndex`. (:issue:`23882`)
+- :meth:`pandas.api.types.is_list_like` has gained a keyword ``allow_sets`` which is ``True`` by default; if ``False``,
+  all instances of ``set`` will not be considered "list-like" anymore (:issue:`23061`)
+- :meth:`Index.to_frame` now supports overriding column name(s) (:issue:`22580`).
+- :meth:`Categorical.from_codes` now can take a ``dtype`` parameter as an alternative to passing ``categories`` and ``ordered`` (:issue:`24398`).
+- New attribute ``__git_version__`` will return git commit sha of current build (:issue:`21295`).
+- Compatibility with Matplotlib 3.0 (:issue:`22790`).
+- Added :meth:`Interval.overlaps`, :meth:`arrays.IntervalArray.overlaps`, and :meth:`IntervalIndex.overlaps` for determining overlaps between interval-like objects (:issue:`21998`)
+- :func:`read_fwf` now accepts keyword ``infer_nrows`` (:issue:`15138`).
+- :func:`~DataFrame.to_parquet` now supports writing a ``DataFrame`` as a directory of parquet files partitioned by a subset of the columns when ``engine = 'pyarrow'`` (:issue:`23283`)
+- :meth:`Timestamp.tz_localize`, :meth:`DatetimeIndex.tz_localize`, and :meth:`Series.tz_localize` have gained the ``nonexistent`` argument for alternative handling of nonexistent times. See :ref:`timeseries.timezone_nonexistent` (:issue:`8917`, :issue:`24466`)
+- :meth:`Index.difference`, :meth:`Index.intersection`, :meth:`Index.union`, and :meth:`Index.symmetric_difference` now have an optional ``sort`` parameter to control whether the results should be sorted if possible (:issue:`17839`, :issue:`24471`)
+- :meth:`read_excel` now accepts ``usecols`` as a list of column names or callable (:issue:`18273`)
+- :meth:`MultiIndex.to_flat_index` has been added to flatten multiple levels into a single-level :class:`Index` object.
+- :meth:`DataFrame.to_stata` and :class:`pandas.io.stata.StataWriter117` can write mixed string columns to Stata strl format (:issue:`23633`)
+- :meth:`DataFrame.between_time` and :meth:`DataFrame.at_time` have gained the ``axis`` parameter (:issue:`8839`)
+- :meth:`DataFrame.to_records` now accepts ``index_dtypes`` and ``column_dtypes`` parameters to allow different data types in stored column and index records (:issue:`18146`)
+- :class:`IntervalIndex` has gained the :attr:`~IntervalIndex.is_overlapping` attribute to indicate if the ``IntervalIndex`` contains any overlapping intervals (:issue:`23309`)
+- :func:`pandas.DataFrame.to_sql` has gained the ``method`` argument to control SQL insertion clause. See the :ref:`insertion method <io.sql.method>` section in the documentation. (:issue:`8953`)
+- :meth:`DataFrame.corrwith` now supports Spearman's rank correlation, Kendall's tau as well as callable correlation methods. (:issue:`21925`)
+- :meth:`DataFrame.to_json`, :meth:`DataFrame.to_csv`, :meth:`DataFrame.to_pickle`, and other export methods now support tilde(~) in path argument. (:issue:`23473`)
+
+.. _whatsnew_0240.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas 0.24.0 includes a number of API breaking changes.
+
+
+.. _whatsnew_0240.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We have updated our minimum supported versions of dependencies (:issue:`21242`, :issue:`18742`, :issue:`23774`, :issue:`24767`).
+If installed, we now require:
+
++-----------------+-----------------+----------+
+| Package         | Minimum Version | Required |
++=================+=================+==========+
+| numpy           | 1.12.0          |    X     |
++-----------------+-----------------+----------+
+| bottleneck      | 1.2.0           |          |
++-----------------+-----------------+----------+
+| fastparquet     | 0.2.1           |          |
++-----------------+-----------------+----------+
+| matplotlib      | 2.0.0           |          |
++-----------------+-----------------+----------+
+| numexpr         | 2.6.1           |          |
++-----------------+-----------------+----------+
+| pandas-gbq      | 0.8.0           |          |
++-----------------+-----------------+----------+
+| pyarrow         | 0.9.0           |          |
++-----------------+-----------------+----------+
+| pytables        | 3.4.2           |          |
++-----------------+-----------------+----------+
+| scipy           | 0.18.1          |          |
++-----------------+-----------------+----------+
+| xlrd            | 1.0.0           |          |
++-----------------+-----------------+----------+
+| pytest (dev)    | 3.6             |          |
++-----------------+-----------------+----------+
+
+Additionally we no longer depend on ``feather-format`` for feather based storage
+and replaced it with references to ``pyarrow`` (:issue:`21639` and :issue:`23053`).
+
+.. _whatsnew_0240.api_breaking.csv_line_terminator:
+
+``os.linesep`` is used for ``line_terminator`` of ``DataFrame.to_csv``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`DataFrame.to_csv` now uses :func:`os.linesep` rather than ``'\n'``
+for the default line terminator (:issue:`20353`).
+This change only affects when running on Windows, where ``'\r\n'`` was used for line terminator
+even when ``'\n'`` was passed in ``line_terminator``.
+
+*Previous behavior* on Windows:
+
+.. code-block:: ipython
+
+    In [1]: data = pd.DataFrame({"string_with_lf": ["a\nbc"],
+       ...:                      "string_with_crlf": ["a\r\nbc"]})
+
+    In [2]: # When passing file PATH to to_csv,
+       ...: # line_terminator does not work, and csv is saved with '\r\n'.
+       ...: # Also, this converts all '\n's in the data to '\r\n'.
+       ...: data.to_csv("test.csv", index=False, line_terminator='\n')
+
+    In [3]: with open("test.csv", mode='rb') as f:
+       ...:     print(f.read())
+    Out[3]: b'string_with_lf,string_with_crlf\r\n"a\r\nbc","a\r\r\nbc"\r\n'
+
+    In [4]: # When passing file OBJECT with newline option to
+       ...: # to_csv, line_terminator works.
+       ...: with open("test2.csv", mode='w', newline='\n') as f:
+       ...:     data.to_csv(f, index=False, line_terminator='\n')
+
+    In [5]: with open("test2.csv", mode='rb') as f:
+       ...:     print(f.read())
+    Out[5]: b'string_with_lf,string_with_crlf\n"a\nbc","a\r\nbc"\n'
+
+
+*New behavior* on Windows:
+
+Passing ``line_terminator`` explicitly, set the ``line terminator`` to that character.
+
+.. code-block:: ipython
+
+   In [1]: data = pd.DataFrame({"string_with_lf": ["a\nbc"],
+      ...:                      "string_with_crlf": ["a\r\nbc"]})
+
+   In [2]: data.to_csv("test.csv", index=False, line_terminator='\n')
+
+   In [3]: with open("test.csv", mode='rb') as f:
+      ...:     print(f.read())
+   Out[3]: b'string_with_lf,string_with_crlf\n"a\nbc","a\r\nbc"\n'
+
+
+On Windows, the value of ``os.linesep`` is ``'\r\n'``, so if ``line_terminator`` is not
+set, ``'\r\n'`` is used for line terminator.
+
+.. code-block:: ipython
+
+   In [1]: data = pd.DataFrame({"string_with_lf": ["a\nbc"],
+      ...:                      "string_with_crlf": ["a\r\nbc"]})
+
+   In [2]: data.to_csv("test.csv", index=False)
+
+   In [3]: with open("test.csv", mode='rb') as f:
+      ...:     print(f.read())
+   Out[3]: b'string_with_lf,string_with_crlf\r\n"a\nbc","a\r\nbc"\r\n'
+
+
+For file objects, specifying ``newline`` is not sufficient to set the line terminator.
+You must pass in the ``line_terminator`` explicitly, even in this case.
+
+.. code-block:: ipython
+
+   In [1]: data = pd.DataFrame({"string_with_lf": ["a\nbc"],
+      ...:                      "string_with_crlf": ["a\r\nbc"]})
+
+   In [2]: with open("test2.csv", mode='w', newline='\n') as f:
+      ...:     data.to_csv(f, index=False)
+
+   In [3]: with open("test2.csv", mode='rb') as f:
+      ...:     print(f.read())
+   Out[3]: b'string_with_lf,string_with_crlf\r\n"a\nbc","a\r\nbc"\r\n'
+
+.. _whatsnew_0240.bug_fixes.nan_with_str_dtype:
+
+Proper handling of ``np.nan`` in a string data-typed column with the Python engine
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+There was bug in :func:`read_excel` and :func:`read_csv` with the Python
+engine, where missing values turned to ``'nan'`` with ``dtype=str`` and
+``na_filter=True``. Now, these missing values are converted to the string
+missing indicator, ``np.nan``. (:issue:`20377`)
+
+.. ipython:: python
+   :suppress:
+
+   from io import StringIO
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [5]: data = 'a,b,c\n1,,3\n4,5,6'
+   In [6]: df = pd.read_csv(StringIO(data), engine='python', dtype=str, na_filter=True)
+   In [7]: df.loc[0, 'b']
+   Out[7]:
+   'nan'
+
+*New behavior*:
+
+.. ipython:: python
+
+   data = 'a,b,c\n1,,3\n4,5,6'
+   df = pd.read_csv(StringIO(data), engine='python', dtype=str, na_filter=True)
+   df.loc[0, 'b']
+
+Notice how we now instead output ``np.nan`` itself instead of a stringified form of it.
+
+.. _whatsnew_0240.api.timezone_offset_parsing:
+
+Parsing datetime strings with timezone offsets
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, parsing datetime strings with UTC offsets with :func:`to_datetime`
+or :class:`DatetimeIndex` would automatically convert the datetime to UTC
+without timezone localization. This is inconsistent from parsing the same
+datetime string with :class:`Timestamp` which would preserve the UTC
+offset in the ``tz`` attribute. Now, :func:`to_datetime` preserves the UTC
+offset in the ``tz`` attribute when all the datetime strings have the same
+UTC offset (:issue:`17697`, :issue:`11736`, :issue:`22457`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [2]: pd.to_datetime("2015-11-18 15:30:00+05:30")
+    Out[2]: Timestamp('2015-11-18 10:00:00')
+
+    In [3]: pd.Timestamp("2015-11-18 15:30:00+05:30")
+    Out[3]: Timestamp('2015-11-18 15:30:00+0530', tz='pytz.FixedOffset(330)')
+
+    # Different UTC offsets would automatically convert the datetimes to UTC (without a UTC timezone)
+    In [4]: pd.to_datetime(["2015-11-18 15:30:00+05:30", "2015-11-18 16:30:00+06:30"])
+    Out[4]: DatetimeIndex(['2015-11-18 10:00:00', '2015-11-18 10:00:00'], dtype='datetime64[ns]', freq=None)
+
+*New behavior*:
+
+.. ipython:: python
+
+    pd.to_datetime("2015-11-18 15:30:00+05:30")
+    pd.Timestamp("2015-11-18 15:30:00+05:30")
+
+Parsing datetime strings with the same UTC offset will preserve the UTC offset in the ``tz``
+
+.. ipython:: python
+
+    pd.to_datetime(["2015-11-18 15:30:00+05:30"] * 2)
+
+Parsing datetime strings with different UTC offsets will now create an Index of
+``datetime.datetime`` objects with different UTC offsets
+
+.. code-block:: ipython
+
+    In [59]: idx = pd.to_datetime(["2015-11-18 15:30:00+05:30",
+                                   "2015-11-18 16:30:00+06:30"])
+
+    In[60]: idx
+    Out[60]: Index([2015-11-18 15:30:00+05:30, 2015-11-18 16:30:00+06:30], dtype='object')
+
+    In[61]: idx[0]
+    Out[61]: Timestamp('2015-11-18 15:30:00+0530', tz='UTC+05:30')
+
+    In[62]: idx[1]
+    Out[62]: Timestamp('2015-11-18 16:30:00+0630', tz='UTC+06:30')
+
+Passing ``utc=True`` will mimic the previous behavior but will correctly indicate
+that the dates have been converted to UTC
+
+.. ipython:: python
+
+    pd.to_datetime(["2015-11-18 15:30:00+05:30",
+                    "2015-11-18 16:30:00+06:30"], utc=True)
+
+
+.. _whatsnew_0240.api_breaking.read_csv_mixed_tz:
+
+Parsing mixed-timezones with :func:`read_csv`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`read_csv` no longer silently converts mixed-timezone columns to UTC (:issue:`24987`).
+
+*Previous behavior*
+
+.. code-block:: python
+
+   >>> import io
+   >>> content = """\
+   ... a
+   ... 2000-01-01T00:00:00+05:00
+   ... 2000-01-01T00:00:00+06:00"""
+   >>> df = pd.read_csv(io.StringIO(content), parse_dates=['a'])
+   >>> df.a
+   0   1999-12-31 19:00:00
+   1   1999-12-31 18:00:00
+   Name: a, dtype: datetime64[ns]
+
+*New behavior*
+
+.. code-block:: ipython
+
+   In[64]: import io
+
+   In[65]: content = """\
+      ...: a
+      ...: 2000-01-01T00:00:00+05:00
+      ...: 2000-01-01T00:00:00+06:00"""
+
+   In[66]: df = pd.read_csv(io.StringIO(content), parse_dates=['a'])
+
+   In[67]: df.a
+   Out[67]:
+   0   2000-01-01 00:00:00+05:00
+   1   2000-01-01 00:00:00+06:00
+   Name: a, Length: 2, dtype: object
+
+As can be seen, the ``dtype`` is object; each value in the column is a string.
+To convert the strings to an array of datetimes, the ``date_parser`` argument
+
+.. code-block:: ipython
+
+   In [3]: df = pd.read_csv(
+      ...:     io.StringIO(content),
+      ...:     parse_dates=['a'],
+      ...:     date_parser=lambda col: pd.to_datetime(col, utc=True),
+      ...: )
+
+   In [4]: df.a
+   Out[4]:
+   0   1999-12-31 19:00:00+00:00
+   1   1999-12-31 18:00:00+00:00
+   Name: a, dtype: datetime64[ns, UTC]
+
+See :ref:`whatsnew_0240.api.timezone_offset_parsing` for more.
+
+.. _whatsnew_0240.api_breaking.period_end_time:
+
+Time values in ``dt.end_time`` and ``to_timestamp(how='end')``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The time values in :class:`Period` and :class:`PeriodIndex` objects are now set
+to '23:59:59.999999999' when calling :attr:`Series.dt.end_time`, :attr:`Period.end_time`,
+:attr:`PeriodIndex.end_time`, :func:`Period.to_timestamp` with ``how='end'``,
+or :func:`PeriodIndex.to_timestamp` with ``how='end'`` (:issue:`17157`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [2]: p = pd.Period('2017-01-01', 'D')
+   In [3]: pi = pd.PeriodIndex([p])
+
+   In [4]: pd.Series(pi).dt.end_time[0]
+   Out[4]: Timestamp(2017-01-01 00:00:00)
+
+   In [5]: p.end_time
+   Out[5]: Timestamp(2017-01-01 23:59:59.999999999)
+
+*New behavior*:
+
+Calling :attr:`Series.dt.end_time` will now result in a time of '23:59:59.999999999' as
+is the case with :attr:`Period.end_time`, for example
+
+.. ipython:: python
+
+   p = pd.Period('2017-01-01', 'D')
+   pi = pd.PeriodIndex([p])
+
+   pd.Series(pi).dt.end_time[0]
+
+   p.end_time
+
+.. _whatsnew_0240.api_breaking.datetime_unique:
+
+Series.unique for timezone-aware data
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The return type of :meth:`Series.unique` for datetime with timezone values has changed
+from an :class:`numpy.ndarray` of :class:`Timestamp` objects to a :class:`arrays.DatetimeArray` (:issue:`24024`).
+
+.. ipython:: python
+
+   ser = pd.Series([pd.Timestamp('2000', tz='UTC'),
+                    pd.Timestamp('2000', tz='UTC')])
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [3]: ser.unique()
+   Out[3]: array([Timestamp('2000-01-01 00:00:00+0000', tz='UTC')], dtype=object)
+
+
+*New behavior*:
+
+.. ipython:: python
+
+   ser.unique()
+
+
+.. _whatsnew_0240.api_breaking.sparse_values:
+
+Sparse data structure refactor
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``SparseArray``, the array backing ``SparseSeries`` and the columns in a ``SparseDataFrame``,
+is now an extension array (:issue:`21978`, :issue:`19056`, :issue:`22835`).
+To conform to this interface and for consistency with the rest of pandas, some API breaking
+changes were made:
+
+- ``SparseArray`` is no longer a subclass of :class:`numpy.ndarray`. To convert a ``SparseArray`` to a NumPy array, use :func:`numpy.asarray`.
+- ``SparseArray.dtype`` and ``SparseSeries.dtype`` are now instances of :class:`SparseDtype`, rather than ``np.dtype``. Access the underlying dtype with ``SparseDtype.subtype``.
+- ``numpy.asarray(sparse_array)`` now returns a dense array with all the values, not just the non-fill-value values (:issue:`14167`)
+- ``SparseArray.take`` now matches the API of :meth:`pandas.api.extensions.ExtensionArray.take` (:issue:`19506`):
+
+  * The default value of ``allow_fill`` has changed from ``False`` to ``True``.
+  * The ``out`` and ``mode`` parameters are now longer accepted (previously, this raised if they were specified).
+  * Passing a scalar for ``indices`` is no longer allowed.
+
+- The result of :func:`concat` with a mix of sparse and dense Series is a Series with sparse values, rather than a ``SparseSeries``.
+- ``SparseDataFrame.combine`` and ``DataFrame.combine_first`` no longer supports combining a sparse column with a dense column while preserving the sparse subtype. The result will be an object-dtype SparseArray.
+- Setting :attr:`SparseArray.fill_value` to a fill value with a different dtype is now allowed.
+- ``DataFrame[column]`` is now a :class:`Series` with sparse values, rather than a :class:`SparseSeries`, when slicing a single column with sparse values (:issue:`23559`).
+- The result of :meth:`Series.where` is now a ``Series`` with sparse values, like with other extension arrays (:issue:`24077`)
+
+Some new warnings are issued for operations that require or are likely to materialize a large dense array:
+
+- A :class:`errors.PerformanceWarning` is issued when using fillna with a ``method``, as a dense array is constructed to create the filled array. Filling with a ``value`` is the efficient way to fill a sparse array.
+- A :class:`errors.PerformanceWarning` is now issued when concatenating sparse Series with differing fill values. The fill value from the first sparse array continues to be used.
+
+In addition to these API breaking changes, many :ref:`Performance Improvements and Bug Fixes have been made <whatsnew_0240.bug_fixes.sparse>`.
+
+Finally, a ``Series.sparse`` accessor was added to provide sparse-specific methods like :meth:`Series.sparse.from_coo`.
+
+.. ipython:: python
+
+   s = pd.Series([0, 0, 1, 1, 1], dtype='Sparse[int]')
+   s.sparse.density
+
+.. _whatsnew_0240.api_breaking.get_dummies:
+
+:meth:`get_dummies` always returns a DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, when ``sparse=True`` was passed to :func:`get_dummies`, the return value could be either
+a :class:`DataFrame` or a :class:`SparseDataFrame`, depending on whether all or a just a subset
+of the columns were dummy-encoded. Now, a :class:`DataFrame` is always returned (:issue:`24284`).
+
+*Previous behavior*
+
+The first :func:`get_dummies` returns a :class:`DataFrame` because the column ``A``
+is not dummy encoded. When just ``["B", "C"]`` are passed to ``get_dummies``,
+then all the columns are dummy-encoded, and a :class:`SparseDataFrame` was returned.
+
+.. code-block:: ipython
+
+   In [2]: df = pd.DataFrame({"A": [1, 2], "B": ['a', 'b'], "C": ['a', 'a']})
+
+   In [3]: type(pd.get_dummies(df, sparse=True))
+   Out[3]: pandas.DataFrame
+
+   In [4]: type(pd.get_dummies(df[['B', 'C']], sparse=True))
+   Out[4]: pandas.core.sparse.frame.SparseDataFrame
+
+.. ipython:: python
+   :suppress:
+
+   df = pd.DataFrame({"A": [1, 2], "B": ['a', 'b'], "C": ['a', 'a']})
+
+*New behavior*
+
+Now, the return type is consistently a :class:`DataFrame`.
+
+.. ipython:: python
+
+   type(pd.get_dummies(df, sparse=True))
+   type(pd.get_dummies(df[['B', 'C']], sparse=True))
+
+.. note::
+
+   There's no difference in memory usage between a :class:`SparseDataFrame`
+   and a :class:`DataFrame` with sparse values. The memory usage will
+   be the same as in the previous version of pandas.
+
+.. _whatsnew_0240.api_breaking.frame_to_dict_index_orient:
+
+Raise ValueError in ``DataFrame.to_dict(orient='index')``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Bug in :func:`DataFrame.to_dict` raises ``ValueError`` when used with
+``orient='index'`` and a non-unique index instead of losing data (:issue:`22801`)
+
+.. ipython:: python
+    :okexcept:
+
+    df = pd.DataFrame({'a': [1, 2], 'b': [0.5, 0.75]}, index=['A', 'A'])
+    df
+
+    df.to_dict(orient='index')
+
+.. _whatsnew_0240.api.datetimelike.normalize:
+
+Tick DateOffset normalize restrictions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Creating a ``Tick`` object (:class:`Day`, :class:`Hour`, :class:`Minute`,
+:class:`Second`, :class:`Milli`, :class:`Micro`, :class:`Nano`) with
+``normalize=True`` is no longer supported.  This prevents unexpected behavior
+where addition could fail to be monotone or associative.  (:issue:`21427`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+
+   In [2]: ts = pd.Timestamp('2018-06-11 18:01:14')
+
+   In [3]: ts
+   Out[3]: Timestamp('2018-06-11 18:01:14')
+
+   In [4]: tic = pd.offsets.Hour(n=2, normalize=True)
+      ...:
+
+   In [5]: tic
+   Out[5]: <2 * Hours>
+
+   In [6]: ts + tic
+   Out[6]: Timestamp('2018-06-11 00:00:00')
+
+   In [7]: ts + tic + tic + tic == ts + (tic + tic + tic)
+   Out[7]: False
+
+*New behavior*:
+
+.. ipython:: python
+
+    ts = pd.Timestamp('2018-06-11 18:01:14')
+    tic = pd.offsets.Hour(n=2)
+    ts + tic + tic + tic == ts + (tic + tic + tic)
+
+
+.. _whatsnew_0240.api.datetimelike:
+
+
+.. _whatsnew_0240.api.period_subtraction:
+
+Period subtraction
+^^^^^^^^^^^^^^^^^^
+
+Subtraction of a ``Period`` from another ``Period`` will give a ``DateOffset``.
+instead of an integer (:issue:`21314`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [2]: june = pd.Period('June 2018')
+
+    In [3]: april = pd.Period('April 2018')
+
+    In [4]: june - april
+    Out [4]: 2
+
+*New behavior*:
+
+.. ipython:: python
+
+    june = pd.Period('June 2018')
+    april = pd.Period('April 2018')
+    june - april
+
+Similarly, subtraction of a ``Period`` from a ``PeriodIndex`` will now return
+an ``Index`` of ``DateOffset`` objects instead of an ``Int64Index``
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [2]: pi = pd.period_range('June 2018', freq='M', periods=3)
+
+    In [3]: pi - pi[0]
+    Out[3]: Int64Index([0, 1, 2], dtype='int64')
+
+*New behavior*:
+
+.. ipython:: python
+
+    pi = pd.period_range('June 2018', freq='M', periods=3)
+    pi - pi[0]
+
+
+.. _whatsnew_0240.api.timedelta64_subtract_nan:
+
+Addition/subtraction of ``NaN`` from :class:`DataFrame`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Adding or subtracting ``NaN`` from a :class:`DataFrame` column with
+``timedelta64[ns]`` dtype will now raise a ``TypeError`` instead of returning
+all-``NaT``.  This is for compatibility with ``TimedeltaIndex`` and
+``Series`` behavior (:issue:`22163`)
+
+.. ipython:: python
+
+   df = pd.DataFrame([pd.Timedelta(days=1)])
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [4]: df = pd.DataFrame([pd.Timedelta(days=1)])
+
+    In [5]: df - np.nan
+    Out[5]:
+        0
+    0 NaT
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [2]: df - np.nan
+    ...
+    TypeError: unsupported operand type(s) for -: 'TimedeltaIndex' and 'float'
+
+.. _whatsnew_0240.api.dataframe_cmp_broadcasting:
+
+DataFrame comparison operations broadcasting changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Previously, the broadcasting behavior of :class:`DataFrame` comparison
+operations (``==``, ``!=``, ...) was inconsistent with the behavior of
+arithmetic operations (``+``, ``-``, ...).  The behavior of the comparison
+operations has been changed to match the arithmetic operations in these cases.
+(:issue:`22880`)
+
+The affected cases are:
+
+- operating against a 2-dimensional ``np.ndarray`` with either 1 row or 1 column will now broadcast the same way a ``np.ndarray`` would (:issue:`23000`).
+- a list or tuple with length matching the number of rows in the :class:`DataFrame` will now raise ``ValueError`` instead of operating column-by-column (:issue:`22880`.
+- a list or tuple with length matching the number of columns in the :class:`DataFrame` will now operate row-by-row instead of raising ``ValueError`` (:issue:`22880`).
+
+.. ipython:: python
+
+   arr = np.arange(6).reshape(3, 2)
+   df = pd.DataFrame(arr)
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [5]: df == arr[[0], :]
+       ...: # comparison previously broadcast where arithmetic would raise
+   Out[5]:
+          0      1
+   0   True   True
+   1  False  False
+   2  False  False
+   In [6]: df + arr[[0], :]
+   ...
+   ValueError: Unable to coerce to DataFrame, shape must be (3, 2): given (1, 2)
+
+   In [7]: df == (1, 2)
+       ...: # length matches number of columns;
+       ...: # comparison previously raised where arithmetic would broadcast
+   ...
+   ValueError: Invalid broadcasting comparison [(1, 2)] with block values
+   In [8]: df + (1, 2)
+   Out[8]:
+      0  1
+   0  1  3
+   1  3  5
+   2  5  7
+
+   In [9]: df == (1, 2, 3)
+       ...:  # length matches number of rows
+       ...:  # comparison previously broadcast where arithmetic would raise
+   Out[9]:
+          0      1
+   0  False   True
+   1   True  False
+   2  False  False
+   In [10]: df + (1, 2, 3)
+   ...
+   ValueError: Unable to coerce to Series, length must be 2: given 3
+
+*New behavior*:
+
+.. ipython:: python
+
+   # Comparison operations and arithmetic operations both broadcast.
+   df == arr[[0], :]
+   df + arr[[0], :]
+
+.. ipython:: python
+
+   # Comparison operations and arithmetic operations both broadcast.
+   df == (1, 2)
+   df + (1, 2)
+
+.. code-block:: ipython
+
+   # Comparison operations and arithmetic operations both raise ValueError.
+   In [6]: df == (1, 2, 3)
+   ...
+   ValueError: Unable to coerce to Series, length must be 2: given 3
+
+   In [7]: df + (1, 2, 3)
+   ...
+   ValueError: Unable to coerce to Series, length must be 2: given 3
+
+.. _whatsnew_0240.api.dataframe_arithmetic_broadcasting:
+
+DataFrame arithmetic operations broadcasting changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`DataFrame` arithmetic operations when operating with 2-dimensional
+``np.ndarray`` objects now broadcast in the same way as ``np.ndarray``
+broadcast.  (:issue:`23000`)
+
+.. ipython:: python
+
+   arr = np.arange(6).reshape(3, 2)
+   df = pd.DataFrame(arr)
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [5]: df + arr[[0], :]   # 1 row, 2 columns
+   ...
+   ValueError: Unable to coerce to DataFrame, shape must be (3, 2): given (1, 2)
+   In [6]: df + arr[:, [1]]   # 1 column, 3 rows
+   ...
+   ValueError: Unable to coerce to DataFrame, shape must be (3, 2): given (3, 1)
+
+*New behavior*:
+
+.. ipython:: python
+
+   df + arr[[0], :]   # 1 row, 2 columns
+   df + arr[:, [1]]   # 1 column, 3 rows
+
+.. _whatsnew_0240.api.incompatibilities:
+
+Series and Index data-dtype incompatibilities
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``Series`` and ``Index`` constructors now raise when the
+data is incompatible with a passed ``dtype=`` (:issue:`15832`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [4]: pd.Series([-1], dtype="uint64")
+    Out [4]:
+    0    18446744073709551615
+    dtype: uint64
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [4]: pd.Series([-1], dtype="uint64")
+    Out [4]:
+    ...
+    OverflowError: Trying to coerce negative values to unsigned integers
+
+.. _whatsnew_0240.api.concat_categorical:
+
+Concatenation changes
+^^^^^^^^^^^^^^^^^^^^^
+
+Calling :func:`pandas.concat` on a ``Categorical`` of ints with NA values now
+causes them to be processed as objects when concatenating with anything
+other than another ``Categorical`` of ints (:issue:`19214`)
+
+.. ipython:: python
+
+    s = pd.Series([0, 1, np.nan])
+    c = pd.Series([0, 1, np.nan], dtype="category")
+
+*Previous behavior*
+
+.. code-block:: ipython
+
+    In [3]: pd.concat([s, c])
+    Out[3]:
+    0    0.0
+    1    1.0
+    2    NaN
+    0    0.0
+    1    1.0
+    2    NaN
+    dtype: float64
+
+*New behavior*
+
+.. ipython:: python
+
+    pd.concat([s, c])
+
+Datetimelike API changes
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- For :class:`DatetimeIndex` and :class:`TimedeltaIndex` with non-``None`` ``freq`` attribute, addition or subtraction of integer-dtyped array or ``Index`` will return an object of the same class (:issue:`19959`)
+- :class:`DateOffset` objects are now immutable. Attempting to alter one of these will now raise ``AttributeError`` (:issue:`21341`)
+- :class:`PeriodIndex` subtraction of another ``PeriodIndex`` will now return an object-dtype :class:`Index` of :class:`DateOffset` objects instead of raising a ``TypeError`` (:issue:`20049`)
+- :func:`cut` and :func:`qcut` now returns a :class:`DatetimeIndex` or :class:`TimedeltaIndex` bins when the input is datetime or timedelta dtype respectively and ``retbins=True`` (:issue:`19891`)
+- :meth:`DatetimeIndex.to_period` and :meth:`Timestamp.to_period` will issue a warning when timezone information will be lost (:issue:`21333`)
+- :meth:`PeriodIndex.tz_convert` and :meth:`PeriodIndex.tz_localize` have been removed (:issue:`21781`)
+
+.. _whatsnew_0240.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- A newly constructed empty :class:`DataFrame` with integer as the ``dtype`` will now only be cast to ``float64`` if ``index`` is specified (:issue:`22858`)
+- :meth:`Series.str.cat` will now raise if ``others`` is a ``set`` (:issue:`23009`)
+- Passing scalar values to :class:`DatetimeIndex` or :class:`TimedeltaIndex` will now raise ``TypeError`` instead of ``ValueError`` (:issue:`23539`)
+- ``max_rows`` and ``max_cols`` parameters removed from :class:`HTMLFormatter` since truncation is handled by :class:`DataFrameFormatter` (:issue:`23818`)
+- :func:`read_csv` will now raise a ``ValueError`` if a column with missing values is declared as having dtype ``bool`` (:issue:`20591`)
+- The column order of the resultant :class:`DataFrame` from :meth:`MultiIndex.to_frame` is now guaranteed to match the :attr:`MultiIndex.names` order. (:issue:`22420`)
+- Incorrectly passing a :class:`DatetimeIndex` to :meth:`MultiIndex.from_tuples`, rather than a sequence of tuples, now raises a ``TypeError`` rather than a ``ValueError`` (:issue:`24024`)
+- :func:`pd.offsets.generate_range` argument ``time_rule`` has been removed; use ``offset`` instead (:issue:`24157`)
+- In 0.23.x, pandas would raise a ``ValueError`` on a merge of a numeric column (e.g. ``int`` dtyped column) and an ``object`` dtyped column (:issue:`9780`). We have re-enabled the ability to merge ``object`` and other dtypes; pandas will still raise on a merge between a numeric and an ``object`` dtyped column that is composed only of strings (:issue:`21681`)
+- Accessing a level of a ``MultiIndex`` with a duplicate name (e.g. in
+  :meth:`~MultiIndex.get_level_values`) now raises a ``ValueError`` instead of a ``KeyError`` (:issue:`21678`).
+- Invalid construction of ``IntervalDtype`` will now always raise a ``TypeError`` rather than a ``ValueError`` if the subdtype is invalid (:issue:`21185`)
+- Trying to reindex a ``DataFrame`` with a non unique ``MultiIndex`` now raises a ``ValueError`` instead of an ``Exception`` (:issue:`21770`)
+- :class:`Index` subtraction will attempt to operate element-wise instead of raising ``TypeError`` (:issue:`19369`)
+- :class:`pandas.io.formats.style.Styler` supports a ``number-format`` property when using :meth:`~pandas.io.formats.style.Styler.to_excel` (:issue:`22015`)
+- :meth:`DataFrame.corr` and :meth:`Series.corr` now raise a ``ValueError`` along with a helpful error message instead of a ``KeyError`` when supplied with an invalid method (:issue:`22298`)
+- :meth:`shift` will now always return a copy, instead of the previous behaviour of returning self when shifting by 0 (:issue:`22397`)
+- :meth:`DataFrame.set_index` now gives a better (and less frequent) KeyError, raises a ``ValueError`` for incorrect types,
+  and will not fail on duplicate column names with ``drop=True``. (:issue:`22484`)
+- Slicing a single row of a DataFrame with multiple ExtensionArrays of the same type now preserves the dtype, rather than coercing to object (:issue:`22784`)
+- :class:`DateOffset` attribute ``_cacheable`` and method ``_should_cache`` have been removed (:issue:`23118`)
+- :meth:`Series.searchsorted`, when supplied a scalar value to search for, now returns a scalar instead of an array (:issue:`23801`).
+- :meth:`Categorical.searchsorted`, when supplied a scalar value to search for, now returns a scalar instead of an array (:issue:`23466`).
+- :meth:`Categorical.searchsorted` now raises a ``KeyError`` rather that a ``ValueError``, if a searched for key is not found in its categories (:issue:`23466`).
+- :meth:`Index.hasnans` and :meth:`Series.hasnans` now always return a python boolean. Previously, a python or a numpy boolean could be returned, depending on circumstances (:issue:`23294`).
+- The order of the arguments of :func:`DataFrame.to_html` and :func:`DataFrame.to_string` is rearranged to be consistent with each other. (:issue:`23614`)
+- :meth:`CategoricalIndex.reindex` now raises a ``ValueError`` if the target index is non-unique and not equal to the current index. It previously only raised if the target index was not of a categorical dtype (:issue:`23963`).
+- :func:`Series.to_list` and :func:`Index.to_list` are now aliases of ``Series.tolist`` respectively ``Index.tolist`` (:issue:`8826`)
+- The result of ``SparseSeries.unstack`` is now a :class:`DataFrame` with sparse values, rather than a :class:`SparseDataFrame` (:issue:`24372`).
+- :class:`DatetimeIndex` and :class:`TimedeltaIndex` no longer ignore the dtype precision. Passing a non-nanosecond resolution dtype will raise a ``ValueError`` (:issue:`24753`)
+
+
+.. _whatsnew_0240.api.extension:
+
+Extension type changes
+~~~~~~~~~~~~~~~~~~~~~~
+
+**Equality and hashability**
+
+pandas now requires that extension dtypes be hashable (i.e. the respective
+``ExtensionDtype`` objects; hashability is not a requirement for the values
+of the corresponding ``ExtensionArray``). The base class implements
+a default ``__eq__`` and ``__hash__``. If you have a parametrized dtype, you should
+update the ``ExtensionDtype._metadata`` tuple to match the signature of your
+``__init__`` method. See :class:`pandas.api.extensions.ExtensionDtype` for more (:issue:`22476`).
+
+**New and changed methods**
+
+- :meth:`~pandas.api.types.ExtensionArray.dropna` has been added (:issue:`21185`)
+- :meth:`~pandas.api.types.ExtensionArray.repeat` has been added (:issue:`24349`)
+- The ``ExtensionArray`` constructor, ``_from_sequence`` now take the keyword arg ``copy=False`` (:issue:`21185`)
+- :meth:`pandas.api.extensions.ExtensionArray.shift` added as part of the basic ``ExtensionArray`` interface (:issue:`22387`).
+- :meth:`~pandas.api.types.ExtensionArray.searchsorted` has been added (:issue:`24350`)
+- Support for reduction operations such as ``sum``, ``mean`` via opt-in base class method override (:issue:`22762`)
+- :func:`ExtensionArray.isna` is allowed to return an ``ExtensionArray`` (:issue:`22325`).
+
+**Dtype changes**
+
+- ``ExtensionDtype`` has gained the ability to instantiate from string dtypes, e.g. ``decimal`` would instantiate a registered ``DecimalDtype``; furthermore
+  the ``ExtensionDtype`` has gained the method ``construct_array_type`` (:issue:`21185`)
+- Added ``ExtensionDtype._is_numeric`` for controlling whether an extension dtype is considered numeric (:issue:`22290`).
+- Added :meth:`pandas.api.types.register_extension_dtype` to register an extension type with pandas (:issue:`22664`)
+- Updated the ``.type`` attribute for ``PeriodDtype``, ``DatetimeTZDtype``, and ``IntervalDtype`` to be instances of the dtype (``Period``, ``Timestamp``, and ``Interval`` respectively) (:issue:`22938`)
+
+.. _whatsnew_0240.enhancements.extension_array_operators:
+
+**Operator support**
+
+A ``Series`` based on an ``ExtensionArray`` now supports arithmetic and comparison
+operators (:issue:`19577`). There are two approaches for providing operator support for an ``ExtensionArray``:
+
+1. Define each of the operators on your ``ExtensionArray`` subclass.
+2. Use an operator implementation from pandas that depends on operators that are already defined
+   on the underlying elements (scalars) of the ``ExtensionArray``.
+
+See the :ref:`ExtensionArray Operator Support
+<extending.extension.operator>` documentation section for details on both
+ways of adding operator support.
+
+**Other changes**
+
+- A default repr for :class:`pandas.api.extensions.ExtensionArray` is now provided (:issue:`23601`).
+- :meth:`ExtensionArray._formatting_values` is deprecated. Use :attr:`ExtensionArray._formatter` instead. (:issue:`23601`)
+- An ``ExtensionArray`` with a boolean dtype now works correctly as a boolean indexer. :meth:`pandas.api.types.is_bool_dtype` now properly considers them boolean (:issue:`22326`)
+
+**Bug fixes**
+
+- Bug in :meth:`Series.get` for ``Series`` using ``ExtensionArray`` and integer index (:issue:`21257`)
+- :meth:`~Series.shift` now dispatches to :meth:`ExtensionArray.shift` (:issue:`22386`)
+- :meth:`Series.combine` works correctly with :class:`~pandas.api.extensions.ExtensionArray` inside of :class:`Series` (:issue:`20825`)
+- :meth:`Series.combine` with scalar argument now works for any function type (:issue:`21248`)
+- :meth:`Series.astype` and :meth:`DataFrame.astype` now dispatch to :meth:`ExtensionArray.astype` (:issue:`21185`).
+- Slicing a single row of a ``DataFrame`` with multiple ExtensionArrays of the same type now preserves the dtype, rather than coercing to object (:issue:`22784`)
+- Bug when concatenating multiple ``Series`` with different extension dtypes not casting to object dtype (:issue:`22994`)
+- Series backed by an ``ExtensionArray`` now work with :func:`util.hash_pandas_object` (:issue:`23066`)
+- :meth:`DataFrame.stack` no longer converts to object dtype for DataFrames where each column has the same extension dtype. The output Series will have the same dtype as the columns (:issue:`23077`).
+- :meth:`Series.unstack` and :meth:`DataFrame.unstack` no longer convert extension arrays to object-dtype ndarrays. Each column in the output ``DataFrame`` will now have the same dtype as the input (:issue:`23077`).
+- Bug when grouping :meth:`Dataframe.groupby` and aggregating on ``ExtensionArray`` it was not returning the actual ``ExtensionArray`` dtype (:issue:`23227`).
+- Bug in :func:`pandas.merge` when merging on an extension array-backed column (:issue:`23020`).
+
+
+.. _whatsnew_0240.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- :attr:`MultiIndex.labels` has been deprecated and replaced by :attr:`MultiIndex.codes`.
+  The functionality is unchanged. The new name better reflects the natures of
+  these codes and makes the ``MultiIndex`` API more similar to the API for :class:`CategoricalIndex` (:issue:`13443`).
+  As a consequence, other uses of the name ``labels`` in ``MultiIndex`` have also been deprecated and replaced with ``codes``:
+
+  - You should initialize a ``MultiIndex`` instance using a parameter named ``codes`` rather than ``labels``.
+  - ``MultiIndex.set_labels`` has been deprecated in favor of :meth:`MultiIndex.set_codes`.
+  - For method :meth:`MultiIndex.copy`, the ``labels`` parameter has been deprecated and replaced by a ``codes`` parameter.
+- :meth:`DataFrame.to_stata`, :meth:`read_stata`, :class:`StataReader` and :class:`StataWriter` have deprecated the ``encoding`` argument. The encoding of a Stata dta file is determined by the file type and cannot be changed (:issue:`21244`)
+- :meth:`MultiIndex.to_hierarchical` is deprecated and will be removed in a future version (:issue:`21613`)
+- :meth:`Series.ptp` is deprecated. Use ``numpy.ptp`` instead (:issue:`21614`)
+- :meth:`Series.compress` is deprecated. Use ``Series[condition]`` instead (:issue:`18262`)
+- The signature of :meth:`Series.to_csv` has been uniformed to that of :meth:`DataFrame.to_csv`: the name of the first argument is now ``path_or_buf``, the order of subsequent arguments has changed, the ``header`` argument now defaults to ``True``. (:issue:`19715`)
+- :meth:`Categorical.from_codes` has deprecated providing float values for the ``codes`` argument. (:issue:`21767`)
+- :func:`pandas.read_table` is deprecated. Instead, use :func:`read_csv` passing ``sep='\t'`` if necessary. This deprecation has been removed in 0.25.0. (:issue:`21948`)
+- :meth:`Series.str.cat` has deprecated using arbitrary list-likes *within* list-likes. A list-like container may still contain
+  many ``Series``, ``Index`` or 1-dimensional ``np.ndarray``, or alternatively, only scalar values. (:issue:`21950`)
+- :meth:`FrozenNDArray.searchsorted` has deprecated the ``v`` parameter in favor of ``value`` (:issue:`14645`)
+- :func:`DatetimeIndex.shift` and :func:`PeriodIndex.shift` now accept ``periods`` argument instead of ``n`` for consistency with :func:`Index.shift` and :func:`Series.shift`. Using ``n`` throws a deprecation warning (:issue:`22458`, :issue:`22912`)
+- The ``fastpath`` keyword of the different Index constructors is deprecated (:issue:`23110`).
+- :meth:`Timestamp.tz_localize`, :meth:`DatetimeIndex.tz_localize`, and :meth:`Series.tz_localize` have deprecated the ``errors`` argument in favor of the ``nonexistent`` argument (:issue:`8917`)
+- The class ``FrozenNDArray`` has been deprecated. When unpickling, ``FrozenNDArray`` will be unpickled to ``np.ndarray`` once this class is removed (:issue:`9031`)
+- The methods :meth:`DataFrame.update` and :meth:`Panel.update` have deprecated the ``raise_conflict=False|True`` keyword in favor of ``errors='ignore'|'raise'`` (:issue:`23585`)
+- The methods :meth:`Series.str.partition` and :meth:`Series.str.rpartition` have deprecated the ``pat`` keyword in favor of ``sep`` (:issue:`22676`)
+- Deprecated the ``nthreads`` keyword of :func:`pandas.read_feather` in favor of ``use_threads`` to reflect the changes in ``pyarrow>=0.11.0``. (:issue:`23053`)
+- :func:`pandas.read_excel` has deprecated accepting ``usecols`` as an integer. Please pass in a list of ints from 0 to ``usecols`` inclusive instead (:issue:`23527`)
+- Constructing a :class:`TimedeltaIndex` from data with ``datetime64``-dtyped data is deprecated, will raise ``TypeError`` in a future version (:issue:`23539`)
+- Constructing a :class:`DatetimeIndex` from data with ``timedelta64``-dtyped data is deprecated, will raise ``TypeError`` in a future version (:issue:`23675`)
+- The ``keep_tz=False`` option (the default) of the ``keep_tz`` keyword of
+  :meth:`DatetimeIndex.to_series` is deprecated (:issue:`17832`).
+- Timezone converting a tz-aware ``datetime.datetime`` or :class:`Timestamp` with :class:`Timestamp` and the ``tz`` argument is now deprecated. Instead, use :meth:`Timestamp.tz_convert` (:issue:`23579`)
+- :func:`pandas.api.types.is_period` is deprecated in favor of ``pandas.api.types.is_period_dtype`` (:issue:`23917`)
+- :func:`pandas.api.types.is_datetimetz` is deprecated in favor of ``pandas.api.types.is_datetime64tz`` (:issue:`23917`)
+- Creating a :class:`TimedeltaIndex`, :class:`DatetimeIndex`, or :class:`PeriodIndex` by passing range arguments ``start``, ``end``, and ``periods`` is deprecated in favor of :func:`timedelta_range`, :func:`date_range`, or :func:`period_range` (:issue:`23919`)
+- Passing a string alias like ``'datetime64[ns, UTC]'`` as the ``unit`` parameter to :class:`DatetimeTZDtype` is deprecated. Use :class:`DatetimeTZDtype.construct_from_string` instead (:issue:`23990`).
+- The ``skipna`` parameter of :meth:`~pandas.api.types.infer_dtype` will switch to ``True`` by default in a future version of pandas (:issue:`17066`, :issue:`24050`)
+- In :meth:`Series.where` with Categorical data, providing an ``other`` that is not present in the categories is deprecated. Convert the categorical to a different dtype or add the ``other`` to the categories first (:issue:`24077`).
+- :meth:`Series.clip_lower`, :meth:`Series.clip_upper`, :meth:`DataFrame.clip_lower` and :meth:`DataFrame.clip_upper` are deprecated and will be removed in a future version. Use ``Series.clip(lower=threshold)``, ``Series.clip(upper=threshold)`` and the equivalent ``DataFrame`` methods (:issue:`24203`)
+- :meth:`Series.nonzero` is deprecated and will be removed in a future version (:issue:`18262`)
+- Passing an integer to :meth:`Series.fillna` and :meth:`DataFrame.fillna` with ``timedelta64[ns]`` dtypes is deprecated, will raise ``TypeError`` in a future version.  Use ``obj.fillna(pd.Timedelta(...))`` instead (:issue:`24694`)
+- ``Series.cat.categorical``, ``Series.cat.name`` and ``Series.cat.index`` have been deprecated. Use the attributes on ``Series.cat`` or ``Series`` directly. (:issue:`24751`).
+- Passing a dtype without a precision like ``np.dtype('datetime64')`` or ``timedelta64`` to :class:`Index`, :class:`DatetimeIndex` and :class:`TimedeltaIndex` is now deprecated. Use the nanosecond-precision dtype instead (:issue:`24753`).
+
+.. _whatsnew_0240.deprecations.datetimelike_int_ops:
+
+Integer addition/subtraction with datetimes and timedeltas is deprecated
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In the past, users could—in some cases—add or subtract integers or integer-dtype
+arrays from :class:`Timestamp`, :class:`DatetimeIndex` and :class:`TimedeltaIndex`.
+
+This usage is now deprecated.  Instead add or subtract integer multiples of
+the object's ``freq`` attribute (:issue:`21939`, :issue:`23878`).
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: ts = pd.Timestamp('1994-05-06 12:15:16', freq=pd.offsets.Hour())
+    In [6]: ts + 2
+    Out[6]: Timestamp('1994-05-06 14:15:16', freq='H')
+
+    In [7]: tdi = pd.timedelta_range('1D', periods=2)
+    In [8]: tdi - np.array([2, 1])
+    Out[8]: TimedeltaIndex(['-1 days', '1 days'], dtype='timedelta64[ns]', freq=None)
+
+    In [9]: dti = pd.date_range('2001-01-01', periods=2, freq='7D')
+    In [10]: dti + pd.Index([1, 2])
+    Out[10]: DatetimeIndex(['2001-01-08', '2001-01-22'], dtype='datetime64[ns]', freq=None)
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [108]: ts = pd.Timestamp('1994-05-06 12:15:16', freq=pd.offsets.Hour())
+
+    In[109]: ts + 2 * ts.freq
+    Out[109]: Timestamp('1994-05-06 14:15:16', freq='H')
+
+    In [110]: tdi = pd.timedelta_range('1D', periods=2)
+
+    In [111]: tdi - np.array([2 * tdi.freq, 1 * tdi.freq])
+    Out[111]: TimedeltaIndex(['-1 days', '1 days'], dtype='timedelta64[ns]', freq=None)
+
+    In [112]: dti = pd.date_range('2001-01-01', periods=2, freq='7D')
+
+    In [113]: dti + pd.Index([1 * dti.freq, 2 * dti.freq])
+    Out[113]: DatetimeIndex(['2001-01-08', '2001-01-22'], dtype='datetime64[ns]', freq=None)
+
+
+.. _whatsnew_0240.deprecations.integer_tz:
+
+Passing integer data and a timezone to DatetimeIndex
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The behavior of :class:`DatetimeIndex` when passed integer data and
+a timezone is changing in a future version of pandas. Previously, these
+were interpreted as wall times in the desired timezone. In the future,
+these will be interpreted as wall times in UTC, which are then converted
+to the desired timezone (:issue:`24559`).
+
+The default behavior remains the same, but issues a warning:
+
+.. code-block:: ipython
+
+   In [3]: pd.DatetimeIndex([946684800000000000], tz="US/Central")
+   /bin/ipython:1: FutureWarning:
+       Passing integer-dtype data and a timezone to DatetimeIndex. Integer values
+       will be interpreted differently in a future version of pandas. Previously,
+       these were viewed as datetime64[ns] values representing the wall time
+       *in the specified timezone*. In the future, these will be viewed as
+       datetime64[ns] values representing the wall time *in UTC*. This is similar
+       to a nanosecond-precision UNIX epoch. To accept the future behavior, use
+
+           pd.to_datetime(integer_data, utc=True).tz_convert(tz)
+
+       To keep the previous behavior, use
+
+           pd.to_datetime(integer_data).tz_localize(tz)
+
+    #!/bin/python3
+    Out[3]: DatetimeIndex(['2000-01-01 00:00:00-06:00'], dtype='datetime64[ns, US/Central]', freq=None)
+
+As the warning message explains, opt in to the future behavior by specifying that
+the integer values are UTC, and then converting to the final timezone:
+
+.. ipython:: python
+
+   pd.to_datetime([946684800000000000], utc=True).tz_convert('US/Central')
+
+The old behavior can be retained with by localizing directly to the final timezone:
+
+.. ipython:: python
+
+   pd.to_datetime([946684800000000000]).tz_localize('US/Central')
+
+.. _whatsnew_0240.deprecations.tz_aware_array:
+
+Converting timezone-aware Series and Index to NumPy arrays
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The conversion from a :class:`Series` or :class:`Index` with timezone-aware
+datetime data will change to preserve timezones by default (:issue:`23569`).
+
+NumPy doesn't have a dedicated dtype for timezone-aware datetimes.
+In the past, converting a :class:`Series` or :class:`DatetimeIndex` with
+timezone-aware datatimes would convert to a NumPy array by
+
+1. converting the tz-aware data to UTC
+2. dropping the timezone-info
+3. returning a :class:`numpy.ndarray` with ``datetime64[ns]`` dtype
+
+Future versions of pandas will preserve the timezone information by returning an
+object-dtype NumPy array where each value is a :class:`Timestamp` with the correct
+timezone attached
+
+.. ipython:: python
+
+   ser = pd.Series(pd.date_range('2000', periods=2, tz="CET"))
+   ser
+
+The default behavior remains the same, but issues a warning
+
+.. code-block:: python
+
+   In [8]: np.asarray(ser)
+   /bin/ipython:1: FutureWarning: Converting timezone-aware DatetimeArray to timezone-naive
+         ndarray with 'datetime64[ns]' dtype. In the future, this will return an ndarray
+         with 'object' dtype where each element is a 'pandas.Timestamp' with the correct 'tz'.
+
+           To accept the future behavior, pass 'dtype=object'.
+           To keep the old behavior, pass 'dtype="datetime64[ns]"'.
+     #!/bin/python3
+   Out[8]:
+   array(['1999-12-31T23:00:00.000000000', '2000-01-01T23:00:00.000000000'],
+         dtype='datetime64[ns]')
+
+The previous or future behavior can be obtained, without any warnings, by specifying
+the ``dtype``
+
+*Previous behavior*
+
+.. ipython:: python
+
+   np.asarray(ser, dtype='datetime64[ns]')
+
+*Future behavior*
+
+.. ipython:: python
+
+   # New behavior
+   np.asarray(ser, dtype=object)
+
+
+Or by using :meth:`Series.to_numpy`
+
+.. ipython:: python
+
+   ser.to_numpy()
+   ser.to_numpy(dtype="datetime64[ns]")
+
+All the above applies to a :class:`DatetimeIndex` with tz-aware values as well.
+
+.. _whatsnew_0240.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- The ``LongPanel`` and ``WidePanel`` classes have been removed (:issue:`10892`)
+- :meth:`Series.repeat` has renamed the ``reps`` argument to ``repeats`` (:issue:`14645`)
+- Several private functions were removed from the (non-public) module ``pandas.core.common`` (:issue:`22001`)
+- Removal of the previously deprecated module ``pandas.core.datetools`` (:issue:`14105`, :issue:`14094`)
+- Strings passed into :meth:`DataFrame.groupby` that refer to both column and index levels will raise a ``ValueError`` (:issue:`14432`)
+- :meth:`Index.repeat` and :meth:`MultiIndex.repeat` have renamed the ``n`` argument to ``repeats`` (:issue:`14645`)
+- The ``Series`` constructor and ``.astype`` method will now raise a ``ValueError`` if timestamp dtypes are passed in without a unit (e.g. ``np.datetime64``) for the ``dtype`` parameter (:issue:`15987`)
+- Removal of the previously deprecated ``as_indexer`` keyword completely from ``str.match()`` (:issue:`22356`, :issue:`6581`)
+- The modules ``pandas.types``, ``pandas.computation``, and ``pandas.util.decorators`` have been removed (:issue:`16157`, :issue:`16250`)
+- Removed the ``pandas.formats.style`` shim for :class:`pandas.io.formats.style.Styler` (:issue:`16059`)
+- ``pandas.pnow``, ``pandas.match``, ``pandas.groupby``, ``pd.get_store``, ``pd.Expr``, and ``pd.Term`` have been removed (:issue:`15538`, :issue:`15940`)
+- :meth:`Categorical.searchsorted` and :meth:`Series.searchsorted` have renamed the ``v`` argument to ``value`` (:issue:`14645`)
+- ``pandas.parser``, ``pandas.lib``, and ``pandas.tslib`` have been removed (:issue:`15537`)
+- :meth:`Index.searchsorted` have renamed the ``key`` argument to ``value`` (:issue:`14645`)
+- ``DataFrame.consolidate`` and ``Series.consolidate`` have been removed (:issue:`15501`)
+- Removal of the previously deprecated module ``pandas.json`` (:issue:`19944`)
+- The module ``pandas.tools`` has been removed (:issue:`15358`, :issue:`16005`)
+- :meth:`SparseArray.get_values` and :meth:`SparseArray.to_dense` have dropped the ``fill`` parameter (:issue:`14686`)
+- ``DataFrame.sortlevel`` and ``Series.sortlevel`` have been removed (:issue:`15099`)
+- :meth:`SparseSeries.to_dense` has dropped the ``sparse_only`` parameter (:issue:`14686`)
+- :meth:`DataFrame.astype` and :meth:`Series.astype` have renamed the ``raise_on_error`` argument to ``errors`` (:issue:`14967`)
+- ``is_sequence``, ``is_any_int_dtype``, and ``is_floating_dtype`` have been removed from ``pandas.api.types`` (:issue:`16163`, :issue:`16189`)
+
+.. _whatsnew_0240.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Slicing Series and DataFrames with an monotonically increasing :class:`CategoricalIndex`
+  is now very fast and has speed comparable to slicing with an ``Int64Index``.
+  The speed increase is both when indexing by label (using .loc) and position(.iloc) (:issue:`20395`)
+  Slicing a monotonically increasing :class:`CategoricalIndex` itself (i.e. ``ci[1000:2000]``)
+  shows similar speed improvements as above (:issue:`21659`)
+- Improved performance of :meth:`CategoricalIndex.equals` when comparing to another :class:`CategoricalIndex` (:issue:`24023`)
+- Improved performance of :func:`Series.describe` in case of numeric dtpyes (:issue:`21274`)
+- Improved performance of :func:`.GroupBy.rank` when dealing with tied rankings (:issue:`21237`)
+- Improved performance of :func:`DataFrame.set_index` with columns consisting of :class:`Period` objects (:issue:`21582`, :issue:`21606`)
+- Improved performance of :meth:`Series.at` and :meth:`Index.get_value` for Extension Arrays values (e.g. :class:`Categorical`) (:issue:`24204`)
+- Improved performance of membership checks in :class:`Categorical` and :class:`CategoricalIndex`
+  (i.e. ``x in cat``-style checks are much faster). :meth:`CategoricalIndex.contains`
+  is likewise much faster (:issue:`21369`, :issue:`21508`)
+- Improved performance of :meth:`HDFStore.groups` (and dependent functions like
+  :meth:`HDFStore.keys`.  (i.e. ``x in store`` checks are much faster)
+  (:issue:`21372`)
+- Improved the performance of :func:`pandas.get_dummies` with ``sparse=True`` (:issue:`21997`)
+- Improved performance of :func:`IndexEngine.get_indexer_non_unique` for sorted, non-unique indexes (:issue:`9466`)
+- Improved performance of :func:`PeriodIndex.unique` (:issue:`23083`)
+- Improved performance of :func:`concat` for ``Series`` objects (:issue:`23404`)
+- Improved performance of :meth:`DatetimeIndex.normalize` and :meth:`Timestamp.normalize` for timezone naive or UTC datetimes (:issue:`23634`)
+- Improved performance of :meth:`DatetimeIndex.tz_localize` and various ``DatetimeIndex`` attributes with dateutil UTC timezone (:issue:`23772`)
+- Fixed a performance regression on Windows with Python 3.7 of :func:`read_csv` (:issue:`23516`)
+- Improved performance of :class:`Categorical` constructor for ``Series`` objects (:issue:`23814`)
+- Improved performance of :meth:`~DataFrame.where` for Categorical data (:issue:`24077`)
+- Improved performance of iterating over a :class:`Series`. Using :meth:`DataFrame.itertuples` now creates iterators
+  without internally allocating lists of all elements (:issue:`20783`)
+- Improved performance of :class:`Period` constructor, additionally benefitting ``PeriodArray`` and ``PeriodIndex`` creation (:issue:`24084`, :issue:`24118`)
+- Improved performance of tz-aware :class:`DatetimeArray` binary operations (:issue:`24491`)
+
+.. _whatsnew_0240.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+
+- Bug in :meth:`Categorical.from_codes` where ``NaN`` values in ``codes`` were silently converted to ``0`` (:issue:`21767`). In the future this will raise a ``ValueError``. Also changes the behavior of ``.from_codes([1.1, 2.0])``.
+- Bug in :meth:`Categorical.sort_values` where ``NaN`` values were always positioned in front regardless of ``na_position`` value. (:issue:`22556`).
+- Bug when indexing with a boolean-valued ``Categorical``. Now a boolean-valued ``Categorical`` is treated as a boolean mask (:issue:`22665`)
+- Constructing a :class:`CategoricalIndex` with empty values and boolean categories was raising a ``ValueError`` after a change to dtype coercion (:issue:`22702`).
+- Bug in :meth:`Categorical.take` with a user-provided ``fill_value`` not encoding the ``fill_value``, which could result in a ``ValueError``, incorrect results, or a segmentation fault (:issue:`23296`).
+- In :meth:`Series.unstack`, specifying a ``fill_value`` not present in the categories now raises a ``TypeError`` rather than ignoring the ``fill_value`` (:issue:`23284`)
+- Bug when resampling :meth:`DataFrame.resample` and aggregating on categorical data, the categorical dtype was getting lost. (:issue:`23227`)
+- Bug in many methods of the ``.str``-accessor, which always failed on calling the ``CategoricalIndex.str`` constructor (:issue:`23555`, :issue:`23556`)
+- Bug in :meth:`Series.where` losing the categorical dtype for categorical data (:issue:`24077`)
+- Bug in :meth:`Categorical.apply` where ``NaN`` values could be handled unpredictably. They now remain unchanged (:issue:`24241`)
+- Bug in :class:`Categorical` comparison methods incorrectly raising ``ValueError`` when operating against a :class:`DataFrame` (:issue:`24630`)
+- Bug in :meth:`Categorical.set_categories` where setting fewer new categories with ``rename=True`` caused a segmentation fault (:issue:`24675`)
+
+Datetimelike
+^^^^^^^^^^^^
+
+- Fixed bug where two :class:`DateOffset` objects with different ``normalize`` attributes could evaluate as equal (:issue:`21404`)
+- Fixed bug where :meth:`Timestamp.resolution` incorrectly returned 1-microsecond ``timedelta`` instead of 1-nanosecond :class:`Timedelta` (:issue:`21336`, :issue:`21365`)
+- Bug in :func:`to_datetime` that did not consistently return an :class:`Index` when ``box=True`` was specified (:issue:`21864`)
+- Bug in :class:`DatetimeIndex` comparisons where string comparisons incorrectly raises ``TypeError`` (:issue:`22074`)
+- Bug in :class:`DatetimeIndex` comparisons when comparing against ``timedelta64[ns]`` dtyped arrays; in some cases ``TypeError`` was incorrectly raised, in others it incorrectly failed to raise (:issue:`22074`)
+- Bug in :class:`DatetimeIndex` comparisons when comparing against object-dtyped arrays (:issue:`22074`)
+- Bug in :class:`DataFrame` with ``datetime64[ns]`` dtype addition and subtraction with ``Timedelta``-like objects (:issue:`22005`, :issue:`22163`)
+- Bug in :class:`DataFrame` with ``datetime64[ns]`` dtype addition and subtraction with ``DateOffset`` objects returning an ``object`` dtype instead of ``datetime64[ns]`` dtype (:issue:`21610`, :issue:`22163`)
+- Bug in :class:`DataFrame` with ``datetime64[ns]`` dtype comparing against ``NaT`` incorrectly (:issue:`22242`, :issue:`22163`)
+- Bug in :class:`DataFrame` with ``datetime64[ns]`` dtype subtracting ``Timestamp``-like object incorrectly returned ``datetime64[ns]`` dtype instead of ``timedelta64[ns]`` dtype (:issue:`8554`, :issue:`22163`)
+- Bug in :class:`DataFrame` with ``datetime64[ns]`` dtype subtracting ``np.datetime64`` object with non-nanosecond unit failing to convert to nanoseconds (:issue:`18874`, :issue:`22163`)
+- Bug in :class:`DataFrame` comparisons against ``Timestamp``-like objects failing to raise ``TypeError`` for inequality checks with mismatched types (:issue:`8932`, :issue:`22163`)
+- Bug in :class:`DataFrame` with mixed dtypes including ``datetime64[ns]`` incorrectly raising ``TypeError`` on equality comparisons (:issue:`13128`, :issue:`22163`)
+- Bug in :attr:`DataFrame.values` returning a :class:`DatetimeIndex` for a single-column ``DataFrame`` with tz-aware datetime values. Now a 2-D :class:`numpy.ndarray` of :class:`Timestamp` objects is returned (:issue:`24024`)
+- Bug in :meth:`DataFrame.eq` comparison against ``NaT`` incorrectly returning ``True`` or ``NaN`` (:issue:`15697`, :issue:`22163`)
+- Bug in :class:`DatetimeIndex` subtraction that incorrectly failed to raise ``OverflowError`` (:issue:`22492`, :issue:`22508`)
+- Bug in :class:`DatetimeIndex` incorrectly allowing indexing with ``Timedelta`` object (:issue:`20464`)
+- Bug in :class:`DatetimeIndex` where frequency was being set if original frequency was ``None`` (:issue:`22150`)
+- Bug in rounding methods of :class:`DatetimeIndex` (:meth:`~DatetimeIndex.round`, :meth:`~DatetimeIndex.ceil`, :meth:`~DatetimeIndex.floor`) and :class:`Timestamp` (:meth:`~Timestamp.round`, :meth:`~Timestamp.ceil`, :meth:`~Timestamp.floor`) could give rise to loss of precision (:issue:`22591`)
+- Bug in :func:`to_datetime` with an :class:`Index` argument that would drop the ``name`` from the result (:issue:`21697`)
+- Bug in :class:`PeriodIndex` where adding or subtracting a :class:`timedelta` or :class:`Tick` object produced incorrect results (:issue:`22988`)
+- Bug in the :class:`Series` repr with period-dtype data missing a space before the data (:issue:`23601`)
+- Bug in :func:`date_range` when decrementing a start date to a past end date by a negative frequency (:issue:`23270`)
+- Bug in :meth:`Series.min` which would return ``NaN`` instead of ``NaT`` when called on a series of ``NaT`` (:issue:`23282`)
+- Bug in :meth:`Series.combine_first` not properly aligning categoricals, so that missing values in ``self`` where not filled by valid values from ``other`` (:issue:`24147`)
+- Bug in :func:`DataFrame.combine` with datetimelike values raising a TypeError (:issue:`23079`)
+- Bug in :func:`date_range` with frequency of ``Day`` or higher where dates sufficiently far in the future could wrap around to the past instead of raising ``OutOfBoundsDatetime`` (:issue:`14187`)
+- Bug in :func:`period_range` ignoring the frequency of ``start`` and ``end`` when those are provided as :class:`Period` objects (:issue:`20535`).
+- Bug in :class:`PeriodIndex` with attribute ``freq.n`` greater than 1 where adding a :class:`DateOffset` object would return incorrect results (:issue:`23215`)
+- Bug in :class:`Series` that interpreted string indices as lists of characters when setting datetimelike values (:issue:`23451`)
+- Bug in :class:`DataFrame` when creating a new column from an ndarray of :class:`Timestamp` objects with timezones creating an object-dtype column, rather than datetime with timezone (:issue:`23932`)
+- Bug in :class:`Timestamp` constructor which would drop the frequency of an input :class:`Timestamp` (:issue:`22311`)
+- Bug in :class:`DatetimeIndex` where calling ``np.array(dtindex, dtype=object)`` would incorrectly return an array of ``long`` objects (:issue:`23524`)
+- Bug in :class:`Index` where passing a timezone-aware :class:`DatetimeIndex` and ``dtype=object`` would incorrectly raise a ``ValueError`` (:issue:`23524`)
+- Bug in :class:`Index` where calling ``np.array(dtindex, dtype=object)`` on a timezone-naive :class:`DatetimeIndex` would return an array of ``datetime`` objects instead of :class:`Timestamp` objects, potentially losing nanosecond portions of the timestamps (:issue:`23524`)
+- Bug in :class:`Categorical.__setitem__` not allowing setting with another ``Categorical`` when both are unordered and have the same categories, but in a different order (:issue:`24142`)
+- Bug in :func:`date_range` where using dates with millisecond resolution or higher could return incorrect values or the wrong number of values in the index (:issue:`24110`)
+- Bug in :class:`DatetimeIndex` where constructing a :class:`DatetimeIndex` from a :class:`Categorical` or :class:`CategoricalIndex` would incorrectly drop timezone information (:issue:`18664`)
+- Bug in :class:`DatetimeIndex` and :class:`TimedeltaIndex` where indexing with ``Ellipsis`` would incorrectly lose the index's ``freq`` attribute (:issue:`21282`)
+- Clarified error message produced when passing an incorrect ``freq`` argument to :class:`DatetimeIndex` with ``NaT`` as the first entry in the passed data (:issue:`11587`)
+- Bug in :func:`to_datetime` where ``box`` and ``utc`` arguments were ignored when passing a :class:`DataFrame` or ``dict`` of unit mappings (:issue:`23760`)
+- Bug in :attr:`Series.dt` where the cache would not update properly after an in-place operation (:issue:`24408`)
+- Bug in :class:`PeriodIndex` where comparisons against an array-like object with length 1 failed to raise ``ValueError`` (:issue:`23078`)
+- Bug in :meth:`DatetimeIndex.astype`, :meth:`PeriodIndex.astype` and :meth:`TimedeltaIndex.astype` ignoring the sign of the ``dtype`` for unsigned integer dtypes (:issue:`24405`).
+- Fixed bug in :meth:`Series.max` with ``datetime64[ns]``-dtype failing to return ``NaT`` when nulls are present and ``skipna=False`` is passed (:issue:`24265`)
+- Bug in :func:`to_datetime` where arrays of ``datetime`` objects containing both timezone-aware and timezone-naive ``datetimes`` would fail to raise ``ValueError`` (:issue:`24569`)
+- Bug in :func:`to_datetime` with invalid datetime format doesn't coerce input to ``NaT`` even if ``errors='coerce'`` (:issue:`24763`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :class:`DataFrame` with ``timedelta64[ns]`` dtype division by ``Timedelta``-like scalar incorrectly returning ``timedelta64[ns]`` dtype instead of ``float64`` dtype (:issue:`20088`, :issue:`22163`)
+- Bug in adding a :class:`Index` with object dtype to a :class:`Series` with ``timedelta64[ns]`` dtype incorrectly raising (:issue:`22390`)
+- Bug in multiplying a :class:`Series` with numeric dtype against a ``timedelta`` object (:issue:`22390`)
+- Bug in :class:`Series` with numeric dtype when adding or subtracting an array or ``Series`` with ``timedelta64`` dtype (:issue:`22390`)
+- Bug in :class:`Index` with numeric dtype when multiplying or dividing an array with dtype ``timedelta64`` (:issue:`22390`)
+- Bug in :class:`TimedeltaIndex` incorrectly allowing indexing with ``Timestamp`` object (:issue:`20464`)
+- Fixed bug where subtracting :class:`Timedelta` from an object-dtyped array would raise ``TypeError`` (:issue:`21980`)
+- Fixed bug in adding a :class:`DataFrame` with all-``timedelta64[ns]`` dtypes to a :class:`DataFrame` with all-integer dtypes returning incorrect results instead of raising ``TypeError`` (:issue:`22696`)
+- Bug in :class:`TimedeltaIndex` where adding a timezone-aware datetime scalar incorrectly returned a timezone-naive :class:`DatetimeIndex` (:issue:`23215`)
+- Bug in :class:`TimedeltaIndex` where adding ``np.timedelta64('NaT')`` incorrectly returned an all-``NaT`` :class:`DatetimeIndex` instead of an all-``NaT`` :class:`TimedeltaIndex` (:issue:`23215`)
+- Bug in :class:`Timedelta` and :func:`to_timedelta` have inconsistencies in supported unit string (:issue:`21762`)
+- Bug in :class:`TimedeltaIndex` division where dividing by another :class:`TimedeltaIndex` raised ``TypeError`` instead of returning a :class:`Float64Index` (:issue:`23829`, :issue:`22631`)
+- Bug in :class:`TimedeltaIndex` comparison operations where comparing against non-``Timedelta``-like objects would raise ``TypeError`` instead of returning all-``False`` for ``__eq__`` and all-``True`` for ``__ne__`` (:issue:`24056`)
+- Bug in :class:`Timedelta` comparisons when comparing with a ``Tick`` object incorrectly raising ``TypeError`` (:issue:`24710`)
+
+Timezones
+^^^^^^^^^
+
+- Bug in :meth:`Index.shift` where an ``AssertionError`` would raise when shifting across DST (:issue:`8616`)
+- Bug in :class:`Timestamp` constructor where passing an invalid timezone offset designator (``Z``) would not raise a ``ValueError`` (:issue:`8910`)
+- Bug in :meth:`Timestamp.replace` where replacing at a DST boundary would retain an incorrect offset (:issue:`7825`)
+- Bug in :meth:`Series.replace` with ``datetime64[ns, tz]`` data when replacing ``NaT`` (:issue:`11792`)
+- Bug in :class:`Timestamp` when passing different string date formats with a timezone offset would produce different timezone offsets (:issue:`12064`)
+- Bug when comparing a tz-naive :class:`Timestamp` to a tz-aware :class:`DatetimeIndex` which would coerce the :class:`DatetimeIndex` to tz-naive (:issue:`12601`)
+- Bug in :meth:`Series.truncate` with a tz-aware :class:`DatetimeIndex` which would cause a core dump (:issue:`9243`)
+- Bug in :class:`Series` constructor which would coerce tz-aware and tz-naive :class:`Timestamp` to tz-aware (:issue:`13051`)
+- Bug in :class:`Index` with ``datetime64[ns, tz]`` dtype that did not localize integer data correctly (:issue:`20964`)
+- Bug in :class:`DatetimeIndex` where constructing with an integer and tz would not localize correctly (:issue:`12619`)
+- Fixed bug where :meth:`DataFrame.describe` and :meth:`Series.describe` on tz-aware datetimes did not show ``first`` and ``last`` result (:issue:`21328`)
+- Bug in :class:`DatetimeIndex` comparisons failing to raise ``TypeError`` when comparing timezone-aware ``DatetimeIndex`` against ``np.datetime64`` (:issue:`22074`)
+- Bug in ``DataFrame`` assignment with a timezone-aware scalar (:issue:`19843`)
+- Bug in :func:`DataFrame.asof` that raised a ``TypeError`` when attempting to compare tz-naive and tz-aware timestamps (:issue:`21194`)
+- Bug when constructing a :class:`DatetimeIndex` with :class:`Timestamp` constructed with the ``replace`` method across DST (:issue:`18785`)
+- Bug when setting a new value with :meth:`DataFrame.loc` with a :class:`DatetimeIndex` with a DST transition (:issue:`18308`, :issue:`20724`)
+- Bug in :meth:`Index.unique` that did not re-localize tz-aware dates correctly (:issue:`21737`)
+- Bug when indexing a :class:`Series` with a DST transition (:issue:`21846`)
+- Bug in :meth:`DataFrame.resample` and :meth:`Series.resample` where an ``AmbiguousTimeError`` or ``NonExistentTimeError`` would raise if a timezone aware timeseries ended on a DST transition (:issue:`19375`, :issue:`10117`)
+- Bug in :meth:`DataFrame.drop` and :meth:`Series.drop` when specifying a tz-aware Timestamp key to drop from a :class:`DatetimeIndex` with a DST transition (:issue:`21761`)
+- Bug in :class:`DatetimeIndex` constructor where ``NaT`` and ``dateutil.tz.tzlocal`` would raise an ``OutOfBoundsDatetime`` error (:issue:`23807`)
+- Bug in :meth:`DatetimeIndex.tz_localize` and :meth:`Timestamp.tz_localize` with ``dateutil.tz.tzlocal`` near a DST transition that would return an incorrectly localized datetime (:issue:`23807`)
+- Bug in :class:`Timestamp` constructor where a ``dateutil.tz.tzutc`` timezone passed with a ``datetime.datetime`` argument would be converted to a ``pytz.UTC`` timezone (:issue:`23807`)
+- Bug in :func:`to_datetime` where ``utc=True`` was not respected when specifying a ``unit`` and ``errors='ignore'`` (:issue:`23758`)
+- Bug in :func:`to_datetime` where ``utc=True`` was not respected when passing a :class:`Timestamp` (:issue:`24415`)
+- Bug in :meth:`DataFrame.any` returns wrong value when ``axis=1`` and the data is of datetimelike type (:issue:`23070`)
+- Bug in :meth:`DatetimeIndex.to_period` where a timezone aware index was converted to UTC first before creating :class:`PeriodIndex` (:issue:`22905`)
+- Bug in :meth:`DataFrame.tz_localize`, :meth:`DataFrame.tz_convert`, :meth:`Series.tz_localize`, and :meth:`Series.tz_convert` where ``copy=False`` would mutate the original argument inplace (:issue:`6326`)
+- Bug in :meth:`DataFrame.max` and :meth:`DataFrame.min` with ``axis=1`` where a :class:`Series` with ``NaN`` would be returned when all columns contained the same timezone (:issue:`10390`)
+
+Offsets
+^^^^^^^
+
+- Bug in :class:`FY5253` where date offsets could incorrectly raise an ``AssertionError`` in arithmetic operations (:issue:`14774`)
+- Bug in :class:`DateOffset` where keyword arguments ``week`` and ``milliseconds`` were accepted and ignored.  Passing these will now raise ``ValueError`` (:issue:`19398`)
+- Bug in adding :class:`DateOffset` with :class:`DataFrame` or :class:`PeriodIndex` incorrectly raising ``TypeError`` (:issue:`23215`)
+- Bug in comparing :class:`DateOffset` objects with non-DateOffset objects, particularly strings, raising ``ValueError`` instead of returning ``False`` for equality checks and ``True`` for not-equal checks (:issue:`23524`)
+
+Numeric
+^^^^^^^
+
+- Bug in :class:`Series` ``__rmatmul__`` doesn't support matrix vector multiplication (:issue:`21530`)
+- Bug in :func:`factorize` fails with read-only array (:issue:`12813`)
+- Fixed bug in :func:`unique` handled signed zeros inconsistently: for some inputs 0.0 and -0.0 were treated as equal and for some inputs as different. Now they are treated as equal for all inputs (:issue:`21866`)
+- Bug in :meth:`DataFrame.agg`, :meth:`DataFrame.transform` and :meth:`DataFrame.apply` where,
+  when supplied with a list of functions and ``axis=1`` (e.g. ``df.apply(['sum', 'mean'], axis=1)``),
+  a ``TypeError`` was wrongly raised. For all three methods such calculation are now done correctly. (:issue:`16679`).
+- Bug in :class:`Series` comparison against datetime-like scalars and arrays (:issue:`22074`)
+- Bug in :class:`DataFrame` multiplication between boolean dtype and integer returning ``object`` dtype instead of integer dtype (:issue:`22047`, :issue:`22163`)
+- Bug in :meth:`DataFrame.apply` where, when supplied with a string argument and additional positional or keyword arguments (e.g. ``df.apply('sum', min_count=1)``), a ``TypeError`` was wrongly raised (:issue:`22376`)
+- Bug in :meth:`DataFrame.astype` to extension dtype may raise ``AttributeError`` (:issue:`22578`)
+- Bug in :class:`DataFrame` with ``timedelta64[ns]`` dtype arithmetic operations with ``ndarray`` with integer dtype incorrectly treating the narray as ``timedelta64[ns]`` dtype (:issue:`23114`)
+- Bug in :meth:`Series.rpow` with object dtype ``NaN`` for ``1 ** NA`` instead of ``1`` (:issue:`22922`).
+- :meth:`Series.agg` can now handle numpy NaN-aware methods like :func:`numpy.nansum` (:issue:`19629`)
+- Bug in :meth:`Series.rank` and :meth:`DataFrame.rank` when ``pct=True`` and more than 2\ :sup:`24` rows are present resulted in percentages greater than 1.0 (:issue:`18271`)
+- Calls such as :meth:`DataFrame.round` with a non-unique :meth:`CategoricalIndex` now return expected data. Previously, data would be improperly duplicated (:issue:`21809`).
+- Added ``log10``, ``floor`` and ``ceil`` to the list of supported functions in :meth:`DataFrame.eval` (:issue:`24139`, :issue:`24353`)
+- Logical operations ``&, |, ^`` between :class:`Series` and :class:`Index` will no longer raise ``ValueError`` (:issue:`22092`)
+- Checking PEP 3141 numbers in :func:`~pandas.api.types.is_scalar` function returns ``True`` (:issue:`22903`)
+- Reduction methods like :meth:`Series.sum` now accept the default value of ``keepdims=False`` when called from a NumPy ufunc, rather than raising a ``TypeError``. Full support for ``keepdims`` has not been implemented (:issue:`24356`).
+
+Conversion
+^^^^^^^^^^
+
+- Bug in :meth:`DataFrame.combine_first` in which column types were unexpectedly converted to float (:issue:`20699`)
+- Bug in :meth:`DataFrame.clip` in which column types are not preserved and casted to float (:issue:`24162`)
+- Bug in :meth:`DataFrame.clip` when order of columns of dataframes doesn't match, result observed is wrong in numeric values (:issue:`20911`)
+- Bug in :meth:`DataFrame.astype` where converting to an extension dtype when duplicate column names are present causes a ``RecursionError`` (:issue:`24704`)
+
+Strings
+^^^^^^^
+
+- Bug in :meth:`Index.str.partition` was not nan-safe (:issue:`23558`).
+- Bug in :meth:`Index.str.split` was not nan-safe (:issue:`23677`).
+- Bug :func:`Series.str.contains` not respecting the ``na`` argument for a ``Categorical`` dtype ``Series`` (:issue:`22158`)
+- Bug in :meth:`Index.str.cat` when the result contained only ``NaN`` (:issue:`24044`)
+
+Interval
+^^^^^^^^
+
+- Bug in the :class:`IntervalIndex` constructor where the ``closed`` parameter did not always override the inferred ``closed`` (:issue:`19370`)
+- Bug in the ``IntervalIndex`` repr where a trailing comma was missing after the list of intervals (:issue:`20611`)
+- Bug in :class:`Interval` where scalar arithmetic operations did not retain the ``closed`` value (:issue:`22313`)
+- Bug in :class:`IntervalIndex` where indexing with datetime-like values raised a ``KeyError`` (:issue:`20636`)
+- Bug in ``IntervalTree`` where data containing ``NaN`` triggered a warning and resulted in incorrect indexing queries with :class:`IntervalIndex` (:issue:`23352`)
+
+Indexing
+^^^^^^^^
+
+- Bug in :meth:`DataFrame.ne` fails if columns contain column name "dtype" (:issue:`22383`)
+- The traceback from a ``KeyError`` when asking ``.loc`` for a single missing label is now shorter and more clear (:issue:`21557`)
+- :class:`PeriodIndex` now emits a ``KeyError`` when a malformed string is looked up, which is consistent with the behavior of :class:`DatetimeIndex` (:issue:`22803`)
+- When ``.ix`` is asked for a missing integer label in a :class:`MultiIndex` with a first level of integer type, it now raises a ``KeyError``, consistently with the case of a flat :class:`Int64Index`, rather than falling back to positional indexing (:issue:`21593`)
+- Bug in :meth:`Index.reindex` when reindexing a tz-naive and tz-aware :class:`DatetimeIndex` (:issue:`8306`)
+- Bug in :meth:`Series.reindex` when reindexing an empty series with a ``datetime64[ns, tz]`` dtype (:issue:`20869`)
+- Bug in :class:`DataFrame` when setting values with ``.loc`` and a timezone aware :class:`DatetimeIndex` (:issue:`11365`)
+- ``DataFrame.__getitem__`` now accepts dictionaries and dictionary keys as list-likes of labels, consistently with ``Series.__getitem__`` (:issue:`21294`)
+- Fixed ``DataFrame[np.nan]`` when columns are non-unique (:issue:`21428`)
+- Bug when indexing :class:`DatetimeIndex` with nanosecond resolution dates and timezones (:issue:`11679`)
+- Bug where indexing with a Numpy array containing negative values would mutate the indexer (:issue:`21867`)
+- Bug where mixed indexes wouldn't allow integers for ``.at`` (:issue:`19860`)
+- ``Float64Index.get_loc`` now raises ``KeyError`` when boolean key passed. (:issue:`19087`)
+- Bug in :meth:`DataFrame.loc` when indexing with an :class:`IntervalIndex` (:issue:`19977`)
+- :class:`Index` no longer mangles ``None``, ``NaN`` and ``NaT``, i.e. they are treated as three different keys. However, for numeric Index all three are still coerced to a ``NaN`` (:issue:`22332`)
+- Bug in ``scalar in Index`` if scalar is a float while the ``Index`` is of integer dtype (:issue:`22085`)
+- Bug in :func:`MultiIndex.set_levels` when levels value is not subscriptable (:issue:`23273`)
+- Bug where setting a timedelta column by ``Index`` causes it to be casted to double, and therefore lose precision (:issue:`23511`)
+- Bug in :func:`Index.union` and :func:`Index.intersection` where name of the ``Index`` of the result was not computed correctly for certain cases (:issue:`9943`, :issue:`9862`)
+- Bug in :class:`Index` slicing with boolean :class:`Index` may raise ``TypeError`` (:issue:`22533`)
+- Bug in ``PeriodArray.__setitem__`` when accepting slice and list-like value (:issue:`23978`)
+- Bug in :class:`DatetimeIndex`, :class:`TimedeltaIndex` where indexing with ``Ellipsis`` would lose their ``freq`` attribute (:issue:`21282`)
+- Bug in ``iat`` where using it to assign an incompatible value would create a new column (:issue:`23236`)
+
+Missing
+^^^^^^^
+
+- Bug in :func:`DataFrame.fillna` where a ``ValueError`` would raise when one column contained a ``datetime64[ns, tz]`` dtype (:issue:`15522`)
+- Bug in :func:`Series.hasnans` that could be incorrectly cached and return incorrect answers if null elements are introduced after an initial call (:issue:`19700`)
+- :func:`Series.isin` now treats all NaN-floats as equal also for ``np.object_``-dtype. This behavior is consistent with the behavior for float64 (:issue:`22119`)
+- :func:`unique` no longer mangles NaN-floats and the ``NaT``-object for ``np.object_``-dtype, i.e. ``NaT`` is no longer coerced to a NaN-value and is treated as a different entity. (:issue:`22295`)
+- :class:`DataFrame` and :class:`Series` now properly handle numpy masked arrays with hardened masks. Previously, constructing a DataFrame or Series from a masked array with a hard mask would create a pandas object containing the underlying value, rather than the expected NaN. (:issue:`24574`)
+- Bug in :class:`DataFrame` constructor where ``dtype`` argument was not honored when handling numpy masked record arrays. (:issue:`24874`)
+
+MultiIndex
+^^^^^^^^^^
+
+- Bug in :func:`io.formats.style.Styler.applymap` where ``subset=`` with :class:`MultiIndex` slice would reduce to :class:`Series` (:issue:`19861`)
+- Removed compatibility for :class:`MultiIndex` pickles prior to version 0.8.0; compatibility with :class:`MultiIndex` pickles from version 0.13 forward is maintained (:issue:`21654`)
+- :meth:`MultiIndex.get_loc_level` (and as a consequence, ``.loc`` on a ``Series`` or ``DataFrame`` with a :class:`MultiIndex` index) will now raise a ``KeyError``, rather than returning an empty ``slice``, if asked a label which is present in the ``levels`` but is unused (:issue:`22221`)
+- :class:`MultiIndex` has gained the :meth:`MultiIndex.from_frame`, it allows constructing a :class:`MultiIndex` object from a :class:`DataFrame` (:issue:`22420`)
+- Fix ``TypeError`` in Python 3 when creating :class:`MultiIndex` in which some levels have mixed types, e.g. when some labels are tuples (:issue:`15457`)
+
+IO
+^^
+
+- Bug in :func:`read_csv` in which a column specified with ``CategoricalDtype`` of boolean categories was not being correctly coerced from string values to booleans (:issue:`20498`)
+- Bug in :func:`read_csv` in which unicode column names were not being properly recognized with Python 2.x (:issue:`13253`)
+- Bug in :meth:`DataFrame.to_sql` when writing timezone aware data (``datetime64[ns, tz]`` dtype) would raise a ``TypeError`` (:issue:`9086`)
+- Bug in :meth:`DataFrame.to_sql` where a naive :class:`DatetimeIndex` would be written as ``TIMESTAMP WITH TIMEZONE`` type in supported databases, e.g. PostgreSQL (:issue:`23510`)
+- Bug in :meth:`read_excel` when ``parse_cols`` is specified with an empty dataset (:issue:`9208`)
+- :func:`read_html` no longer ignores all-whitespace ``<tr>`` within ``<thead>`` when considering the ``skiprows`` and ``header`` arguments. Previously, users had to decrease their ``header`` and ``skiprows`` values on such tables to work around the issue. (:issue:`21641`)
+- :func:`read_excel` will correctly show the deprecation warning for previously deprecated ``sheetname`` (:issue:`17994`)
+- :func:`read_csv` and :func:`read_table` will throw ``UnicodeError`` and not coredump on badly encoded strings (:issue:`22748`)
+- :func:`read_csv` will correctly parse timezone-aware datetimes (:issue:`22256`)
+- Bug in :func:`read_csv` in which memory management was prematurely optimized for the C engine when the data was being read in chunks (:issue:`23509`)
+- Bug in :func:`read_csv` in unnamed columns were being improperly identified when extracting a multi-index (:issue:`23687`)
+- :func:`read_sas` will parse numbers in sas7bdat-files that have width less than 8 bytes correctly. (:issue:`21616`)
+- :func:`read_sas` will correctly parse sas7bdat files with many columns (:issue:`22628`)
+- :func:`read_sas` will correctly parse sas7bdat files with data page types having also bit 7 set (so page type is 128 + 256 = 384) (:issue:`16615`)
+- Bug in :func:`read_sas` in which an incorrect error was raised on an invalid file format. (:issue:`24548`)
+- Bug in :meth:`detect_client_encoding` where potential ``IOError`` goes unhandled when importing in a mod_wsgi process due to restricted access to stdout. (:issue:`21552`)
+- Bug in :func:`DataFrame.to_html` with ``index=False`` misses truncation indicators (...) on truncated DataFrame (:issue:`15019`, :issue:`22783`)
+- Bug in :func:`DataFrame.to_html` with ``index=False`` when both columns and row index are ``MultiIndex`` (:issue:`22579`)
+- Bug in :func:`DataFrame.to_html` with ``index_names=False`` displaying index name (:issue:`22747`)
+- Bug in :func:`DataFrame.to_html` with ``header=False`` not displaying row index names (:issue:`23788`)
+- Bug in :func:`DataFrame.to_html` with ``sparsify=False`` that caused it to raise ``TypeError`` (:issue:`22887`)
+- Bug in :func:`DataFrame.to_string` that broke column alignment when ``index=False`` and width of first column's values is greater than the width of first column's header (:issue:`16839`, :issue:`13032`)
+- Bug in :func:`DataFrame.to_string` that caused representations of :class:`DataFrame` to not take up the whole window (:issue:`22984`)
+- Bug in :func:`DataFrame.to_csv` where a single level MultiIndex incorrectly wrote a tuple. Now just the value of the index is written (:issue:`19589`).
+- :class:`HDFStore` will raise ``ValueError`` when the ``format`` kwarg is passed to the constructor (:issue:`13291`)
+- Bug in :meth:`HDFStore.append` when appending a :class:`DataFrame` with an empty string column and ``min_itemsize`` < 8 (:issue:`12242`)
+- Bug in :func:`read_csv` in which memory leaks occurred in the C engine when parsing ``NaN`` values due to insufficient cleanup on completion or error (:issue:`21353`)
+- Bug in :func:`read_csv` in which incorrect error messages were being raised when ``skipfooter`` was passed in along with ``nrows``, ``iterator``, or ``chunksize`` (:issue:`23711`)
+- Bug in :func:`read_csv` in which :class:`MultiIndex` index names were being improperly handled in the cases when they were not provided (:issue:`23484`)
+- Bug in :func:`read_csv` in which unnecessary warnings were being raised when the dialect's values conflicted with the default arguments (:issue:`23761`)
+- Bug in :func:`read_html` in which the error message was not displaying the valid flavors when an invalid one was provided (:issue:`23549`)
+- Bug in :meth:`read_excel` in which extraneous header names were extracted, even though none were specified (:issue:`11733`)
+- Bug in :meth:`read_excel` in which column names were not being properly converted to string sometimes in Python 2.x (:issue:`23874`)
+- Bug in :meth:`read_excel` in which ``index_col=None`` was not being respected and parsing index columns anyway (:issue:`18792`, :issue:`20480`)
+- Bug in :meth:`read_excel` in which ``usecols`` was not being validated for proper column names when passed in as a string (:issue:`20480`)
+- Bug in :meth:`DataFrame.to_dict` when the resulting dict contains non-Python scalars in the case of numeric data (:issue:`23753`)
+- :func:`DataFrame.to_string`, :func:`DataFrame.to_html`, :func:`DataFrame.to_latex` will correctly format output when a string is passed as the ``float_format`` argument (:issue:`21625`, :issue:`22270`)
+- Bug in :func:`read_csv` that caused it to raise ``OverflowError`` when trying to use 'inf' as ``na_value`` with integer index column (:issue:`17128`)
+- Bug in :func:`read_csv` that caused the C engine on Python 3.6+ on Windows to improperly read CSV filenames with accented or special characters (:issue:`15086`)
+- Bug in :func:`read_fwf` in which the compression type of a file was not being properly inferred (:issue:`22199`)
+- Bug in :func:`pandas.io.json.json_normalize` that caused it to raise ``TypeError`` when two consecutive elements of ``record_path`` are dicts (:issue:`22706`)
+- Bug in :meth:`DataFrame.to_stata`, :class:`pandas.io.stata.StataWriter` and :class:`pandas.io.stata.StataWriter117` where a exception would leave a partially written and invalid dta file (:issue:`23573`)
+- Bug in :meth:`DataFrame.to_stata` and :class:`pandas.io.stata.StataWriter117` that produced invalid files when using strLs with non-ASCII characters (:issue:`23573`)
+- Bug in :class:`HDFStore` that caused it to raise ``ValueError`` when reading a Dataframe in Python 3 from fixed format written in Python 2 (:issue:`24510`)
+- Bug in :func:`DataFrame.to_string` and more generally in the floating ``repr`` formatter. Zeros were not trimmed if ``inf`` was present in a columns while it was the case with NA values. Zeros are now trimmed as in the presence of NA (:issue:`24861`).
+- Bug in the ``repr`` when truncating the number of columns and having a wide last column (:issue:`24849`).
+
+Plotting
+^^^^^^^^
+
+- Bug in :func:`DataFrame.plot.scatter` and :func:`DataFrame.plot.hexbin` caused x-axis label and ticklabels to disappear when colorbar was on in IPython inline backend (:issue:`10611`, :issue:`10678`, and :issue:`20455`)
+- Bug in plotting a Series with datetimes using :func:`matplotlib.axes.Axes.scatter` (:issue:`22039`)
+- Bug in :func:`DataFrame.plot.bar` caused bars to use multiple colors instead of a single one (:issue:`20585`)
+- Bug in validating color parameter caused extra color to be appended to the given color array. This happened to multiple plotting functions using matplotlib. (:issue:`20726`)
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :func:`.Rolling.min` and :func:`.Rolling.max` with ``closed='left'``, a datetime-like index and only one entry in the series leading to segfault (:issue:`24718`)
+- Bug in :func:`.GroupBy.first` and :func:`.GroupBy.last` with ``as_index=False`` leading to the loss of timezone information (:issue:`15884`)
+- Bug in :meth:`DateFrame.resample` when downsampling across a DST boundary (:issue:`8531`)
+- Bug in date anchoring for :meth:`DateFrame.resample` with offset :class:`Day` when n > 1 (:issue:`24127`)
+- Bug where ``ValueError`` is wrongly raised when calling :func:`.SeriesGroupBy.count` method of a
+  ``SeriesGroupBy`` when the grouping variable only contains NaNs and numpy version < 1.13 (:issue:`21956`).
+- Multiple bugs in :func:`.Rolling.min` with ``closed='left'`` and a
+  datetime-like index leading to incorrect results and also segfault. (:issue:`21704`)
+- Bug in :meth:`.Resampler.apply` when passing positional arguments to applied func (:issue:`14615`).
+- Bug in :meth:`Series.resample` when passing ``numpy.timedelta64`` to ``loffset`` kwarg (:issue:`7687`).
+- Bug in :meth:`.Resampler.asfreq` when frequency of ``TimedeltaIndex`` is a subperiod of a new frequency (:issue:`13022`).
+- Bug in :meth:`.SeriesGroupBy.mean` when values were integral but could not fit inside of int64, overflowing instead. (:issue:`22487`)
+- :func:`.RollingGroupby.agg` and :func:`.ExpandingGroupby.agg` now support multiple aggregation functions as parameters (:issue:`15072`)
+- Bug in :meth:`DataFrame.resample` and :meth:`Series.resample` when resampling by a weekly offset (``'W'``) across a DST transition (:issue:`9119`, :issue:`21459`)
+- Bug in :meth:`DataFrame.expanding` in which the ``axis`` argument was not being respected during aggregations (:issue:`23372`)
+- Bug in :meth:`.GroupBy.transform` which caused missing values when the input function can accept a :class:`DataFrame` but renames it (:issue:`23455`).
+- Bug in :func:`.GroupBy.nth` where column order was not always preserved (:issue:`20760`)
+- Bug in :meth:`.GroupBy.rank` with ``method='dense'`` and ``pct=True`` when a group has only one member would raise a ``ZeroDivisionError`` (:issue:`23666`).
+- Calling :meth:`.GroupBy.rank` with empty groups and ``pct=True`` was raising a ``ZeroDivisionError`` (:issue:`22519`)
+- Bug in :meth:`DataFrame.resample` when resampling ``NaT`` in ``TimeDeltaIndex`` (:issue:`13223`).
+- Bug in :meth:`DataFrame.groupby` did not respect the ``observed`` argument when selecting a column and instead always used ``observed=False`` (:issue:`23970`)
+- Bug in :func:`.SeriesGroupBy.pct_change` or :func:`.DataFrameGroupBy.pct_change` would previously work across groups when calculating the percent change, where it now correctly works per group (:issue:`21200`, :issue:`21235`).
+- Bug preventing hash table creation with very large number (2^32) of rows (:issue:`22805`)
+- Bug in groupby when grouping on categorical causes ``ValueError`` and incorrect grouping if ``observed=True`` and ``nan`` is present in categorical column (:issue:`24740`, :issue:`21151`).
+
+Reshaping
+^^^^^^^^^
+
+- Bug in :func:`pandas.concat` when joining resampled DataFrames with timezone aware index (:issue:`13783`)
+- Bug in :func:`pandas.concat` when joining only ``Series`` the ``names`` argument of ``concat`` is no longer ignored (:issue:`23490`)
+- Bug in :meth:`Series.combine_first` with ``datetime64[ns, tz]`` dtype which would return tz-naive result (:issue:`21469`)
+- Bug in :meth:`Series.where` and :meth:`DataFrame.where` with ``datetime64[ns, tz]`` dtype (:issue:`21546`)
+- Bug in :meth:`DataFrame.where` with an empty DataFrame and empty ``cond`` having non-bool dtype (:issue:`21947`)
+- Bug in :meth:`Series.mask` and :meth:`DataFrame.mask` with ``list`` conditionals (:issue:`21891`)
+- Bug in :meth:`DataFrame.replace` raises RecursionError when converting OutOfBounds ``datetime64[ns, tz]`` (:issue:`20380`)
+- :func:`.GroupBy.rank` now raises a ``ValueError`` when an invalid value is passed for argument ``na_option`` (:issue:`22124`)
+- Bug in :func:`get_dummies` with Unicode attributes in Python 2 (:issue:`22084`)
+- Bug in :meth:`DataFrame.replace` raises ``RecursionError`` when replacing empty lists (:issue:`22083`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` when dict is used as the ``to_replace`` value and one key in the dict is another key's value, the results were inconsistent between using integer key and using string key (:issue:`20656`)
+- Bug in :meth:`DataFrame.drop_duplicates` for empty ``DataFrame`` which incorrectly raises an error (:issue:`20516`)
+- Bug in :func:`pandas.wide_to_long` when a string is passed to the stubnames argument and a column name is a substring of that stubname (:issue:`22468`)
+- Bug in :func:`merge` when merging ``datetime64[ns, tz]`` data that contained a DST transition (:issue:`18885`)
+- Bug in :func:`merge_asof` when merging on float values within defined tolerance (:issue:`22981`)
+- Bug in :func:`pandas.concat` when concatenating a multicolumn DataFrame with tz-aware data against a DataFrame with a different number of columns (:issue:`22796`)
+- Bug in :func:`merge_asof` where confusing error message raised when attempting to merge with missing values (:issue:`23189`)
+- Bug in :meth:`DataFrame.nsmallest` and :meth:`DataFrame.nlargest` for dataframes that have a :class:`MultiIndex` for columns (:issue:`23033`).
+- Bug in :func:`pandas.melt` when passing column names that are not present in ``DataFrame`` (:issue:`23575`)
+- Bug in :meth:`DataFrame.append` with a :class:`Series` with a dateutil timezone would raise a ``TypeError`` (:issue:`23682`)
+- Bug in :class:`Series` construction when passing no data and ``dtype=str`` (:issue:`22477`)
+- Bug in :func:`cut` with ``bins`` as an overlapping ``IntervalIndex`` where multiple bins were returned per item instead of raising a ``ValueError`` (:issue:`23980`)
+- Bug in :func:`pandas.concat` when joining ``Series`` datetimetz with ``Series`` category would lose timezone (:issue:`23816`)
+- Bug in :meth:`DataFrame.join` when joining on partial MultiIndex would drop names (:issue:`20452`).
+- :meth:`DataFrame.nlargest` and :meth:`DataFrame.nsmallest` now returns the correct n values when keep != 'all' also when tied on the first columns (:issue:`22752`)
+- Constructing a DataFrame with an index argument that wasn't already an instance of :class:`.Index` was broken (:issue:`22227`).
+- Bug in :class:`DataFrame` prevented list subclasses to be used to construction (:issue:`21226`)
+- Bug in :func:`DataFrame.unstack` and :func:`DataFrame.pivot_table` returning a misleading error message when the resulting DataFrame has more elements than int32 can handle. Now, the error message is improved, pointing towards the actual problem (:issue:`20601`)
+- Bug in :func:`DataFrame.unstack` where a ``ValueError`` was raised when unstacking timezone aware values (:issue:`18338`)
+- Bug in :func:`DataFrame.stack` where timezone aware values were converted to timezone naive values (:issue:`19420`)
+- Bug in :func:`merge_asof` where a ``TypeError`` was raised when ``by_col`` were timezone aware values (:issue:`21184`)
+- Bug showing an incorrect shape when throwing error during ``DataFrame`` construction. (:issue:`20742`)
+
+.. _whatsnew_0240.bug_fixes.sparse:
+
+Sparse
+^^^^^^
+
+- Updating a boolean, datetime, or timedelta column to be Sparse now works (:issue:`22367`)
+- Bug in :meth:`Series.to_sparse` with Series already holding sparse data not constructing properly (:issue:`22389`)
+- Providing a ``sparse_index`` to the SparseArray constructor no longer defaults the na-value to ``np.nan`` for all dtypes. The correct na_value for ``data.dtype`` is now used.
+- Bug in ``SparseArray.nbytes`` under-reporting its memory usage by not including the size of its sparse index.
+- Improved performance of :meth:`Series.shift` for non-NA ``fill_value``, as values are no longer converted to a dense array.
+- Bug in ``DataFrame.groupby`` not including ``fill_value`` in the groups for non-NA ``fill_value`` when grouping by a sparse column (:issue:`5078`)
+- Bug in unary inversion operator (``~``) on a ``SparseSeries`` with boolean values. The performance of this has also been improved (:issue:`22835`)
+- Bug in :meth:`SparseArary.unique` not returning the unique values (:issue:`19595`)
+- Bug in :meth:`SparseArray.nonzero` and :meth:`SparseDataFrame.dropna` returning shifted/incorrect results (:issue:`21172`)
+- Bug in :meth:`DataFrame.apply` where dtypes would lose sparseness (:issue:`23744`)
+- Bug in :func:`concat` when concatenating a list of :class:`Series` with all-sparse values changing the ``fill_value`` and converting to a dense Series (:issue:`24371`)
+
+Style
+^^^^^
+
+- :meth:`~pandas.io.formats.style.Styler.background_gradient` now takes a ``text_color_threshold`` parameter to automatically lighten the text color based on the luminance of the background color. This improves readability with dark background colors without the need to limit the background colormap range. (:issue:`21258`)
+- :meth:`~pandas.io.formats.style.Styler.background_gradient` now also supports tablewise application (in addition to rowwise and columnwise) with ``axis=None`` (:issue:`15204`)
+- :meth:`~pandas.io.formats.style.Styler.bar` now also supports tablewise application (in addition to rowwise and columnwise) with ``axis=None`` and setting clipping range with ``vmin`` and ``vmax`` (:issue:`21548` and :issue:`21526`). ``NaN`` values are also handled properly.
+
+Build changes
+^^^^^^^^^^^^^
+
+- Building pandas for development now requires ``cython >= 0.28.2`` (:issue:`21688`)
+- Testing pandas now requires ``hypothesis>=3.58``.  You can find `the Hypothesis docs here <https://hypothesis.readthedocs.io/en/latest/index.html>`_, and a pandas-specific introduction :ref:`in the contributing guide <using-hypothesis>`. (:issue:`22280`)
+- Building pandas on macOS now targets minimum macOS 10.9 if run on macOS 10.9 or above (:issue:`23424`)
+
+Other
+^^^^^
+
+- Bug where C variables were declared with external linkage causing import errors if certain other C libraries were imported before pandas. (:issue:`24113`)
+
+
+.. _whatsnew_0.24.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.23.4..v0.24.0
diff --git a/pandas/doc/source/whatsnew/v0.24.1.rst b/pandas/doc/source/whatsnew/v0.24.1.rst
new file mode 100644
index 0000000000000..dd859dabc9c64
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.24.1.rst
@@ -0,0 +1,94 @@
+.. _whatsnew_0241:
+
+What's new in 0.24.1 (February 3, 2019)
+---------------------------------------
+
+.. warning::
+
+   The 0.24.x series of releases will be the last to support Python 2. Future feature
+   releases will support Python 3 only. See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+{{ header }}
+
+These are the changes in pandas 0.24.1. See :ref:`release` for a full changelog
+including other versions of pandas. See :ref:`whatsnew_0240` for the 0.24.0 changelog.
+
+.. _whatsnew_0241.api:
+
+API changes
+~~~~~~~~~~~
+
+Changing the ``sort`` parameter for :class:`Index` set operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The default ``sort`` value for :meth:`Index.union` has changed from ``True`` to ``None`` (:issue:`24959`).
+The default *behavior*, however, remains the same: the result is sorted, unless
+
+1. ``self`` and ``other`` are identical
+2. ``self`` or ``other`` is empty
+3. ``self`` or ``other`` contain values that can not be compared (a ``RuntimeWarning`` is raised).
+
+This change will allow ``sort=True`` to mean "always sort" in a future release.
+
+The same change applies to :meth:`Index.difference` and :meth:`Index.symmetric_difference`, which
+would not sort the result when the values could not be compared.
+
+The ``sort`` option for :meth:`Index.intersection` has changed in three ways.
+
+1. The default has changed from ``True`` to ``False``, to restore the
+   pandas 0.23.4 and earlier behavior of not sorting by default.
+2. The behavior of ``sort=True`` can now be obtained with ``sort=None``.
+   This will sort the result only if the values in ``self`` and ``other``
+   are not identical.
+3. The value ``sort=True`` is no longer allowed. A future version of pandas
+   will properly support ``sort=True`` meaning "always sort".
+
+.. _whatsnew_0241.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`DataFrame.to_dict` with ``records`` orient raising an
+  ``AttributeError`` when the ``DataFrame`` contained more than 255 columns, or
+  wrongly converting column names that were not valid python identifiers (:issue:`24939`, :issue:`24940`).
+- Fixed regression in :func:`read_sql` when passing certain queries with MySQL/pymysql (:issue:`24988`).
+- Fixed regression in :class:`Index.intersection` incorrectly sorting the values by default (:issue:`24959`).
+- Fixed regression in :func:`merge` when merging an empty ``DataFrame`` with multiple timezone-aware columns on one of the timezone-aware columns (:issue:`25014`).
+- Fixed regression in :meth:`Series.rename_axis` and :meth:`DataFrame.rename_axis` where passing ``None`` failed to remove the axis name (:issue:`25034`)
+- Fixed regression in :func:`to_timedelta` with ``box=False`` incorrectly returning a ``datetime64`` object instead of a ``timedelta64`` object (:issue:`24961`)
+- Fixed regression where custom hashable types could not be used as column keys in :meth:`DataFrame.set_index` (:issue:`24969`)
+
+.. _whatsnew_0241.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Reshaping**
+
+- Bug in :meth:`DataFrame.groupby` with :class:`Grouper` when there is a time change (DST) and grouping frequency is ``'1d'`` (:issue:`24972`)
+
+**Visualization**
+
+- Fixed the warning for implicitly registered matplotlib converters not showing. See :ref:`whatsnew_0211.converters` for more (:issue:`24963`).
+
+**Other**
+
+- Fixed AttributeError when printing a DataFrame's HTML repr after accessing the IPython config object (:issue:`25036`)
+
+.. _whatsnew_0.241.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. Including the contributors hardcoded for this release, as backporting with
+   MeeseeksDev loses the commit authors
+
+A total of 7 people contributed patches to this release. People with a "+" by their names contributed a patch for the first time.
+
+* Alex Buchkovsky
+* Roman Yurchak
+* h-vetinari
+* jbrockmendel
+* Jeremy Schendel
+* Joris Van den Bossche
+* Tom Augspurger
diff --git a/pandas/doc/source/whatsnew/v0.24.2.rst b/pandas/doc/source/whatsnew/v0.24.2.rst
new file mode 100644
index 0000000000000..d8f2e17cb9e4f
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.24.2.rst
@@ -0,0 +1,107 @@
+.. _whatsnew_0242:
+
+What's new in 0.24.2 (March 12, 2019)
+-------------------------------------
+
+.. warning::
+
+   The 0.24.x series of releases will be the last to support Python 2. Future feature
+   releases will support Python 3 only. See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more.
+
+{{ header }}
+
+These are the changes in pandas 0.24.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+.. _whatsnew_0242.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`DataFrame.all` and :meth:`DataFrame.any` where ``bool_only=True`` was ignored (:issue:`25101`)
+- Fixed issue in ``DataFrame`` construction with passing a mixed list of mixed types could segfault. (:issue:`25075`)
+- Fixed regression in :meth:`DataFrame.apply` causing ``RecursionError`` when ``dict``-like classes were passed as argument. (:issue:`25196`)
+- Fixed regression in :meth:`DataFrame.replace` where ``regex=True`` was only replacing patterns matching the start of the string (:issue:`25259`)
+- Fixed regression in :meth:`DataFrame.duplicated`, where empty dataframe was not returning a boolean dtyped Series. (:issue:`25184`)
+- Fixed regression in :meth:`Series.min` and :meth:`Series.max` where ``numeric_only=True`` was ignored when the ``Series`` contained ``Categorical`` data (:issue:`25299`)
+- Fixed regression in subtraction between :class:`Series` objects with ``datetime64[ns]`` dtype incorrectly raising ``OverflowError`` when the ``Series`` on the right contains null values (:issue:`25317`)
+- Fixed regression in :class:`TimedeltaIndex` where ``np.sum(index)`` incorrectly returned a zero-dimensional object instead of a scalar (:issue:`25282`)
+- Fixed regression in ``IntervalDtype`` construction where passing an incorrect string with 'Interval' as a prefix could result in a ``RecursionError``. (:issue:`25338`)
+- Fixed regression in creating a period-dtype array from a read-only NumPy array of period objects. (:issue:`25403`)
+- Fixed regression in :class:`Categorical`, where constructing it from a categorical ``Series`` and an explicit ``categories=`` that differed from that in the ``Series`` created an invalid object which could trigger segfaults. (:issue:`25318`)
+- Fixed regression in :func:`to_timedelta` losing precision when converting floating data to ``Timedelta`` data (:issue:`25077`).
+- Fixed pip installing from source into an environment without NumPy (:issue:`25193`)
+- Fixed regression in :meth:`DataFrame.replace` where large strings of numbers would be coerced into ``int64``, causing an ``OverflowError`` (:issue:`25616`)
+- Fixed regression in :func:`factorize` when passing a custom ``na_sentinel`` value with ``sort=True`` (:issue:`25409`).
+- Fixed regression in :meth:`DataFrame.to_csv` writing duplicate line endings with gzip compress (:issue:`25311`)
+
+.. _whatsnew_0242.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**I/O**
+
+- Better handling of terminal printing when the terminal dimensions are not known (:issue:`25080`)
+- Bug in reading a HDF5 table-format ``DataFrame`` created in Python 2, in Python 3 (:issue:`24925`)
+- Bug in reading a JSON with ``orient='table'`` generated by :meth:`DataFrame.to_json` with ``index=False`` (:issue:`25170`)
+- Bug where float indexes could have misaligned values when printing (:issue:`25061`)
+
+**Categorical**
+
+- Bug where calling :meth:`Series.replace` on categorical data could return a ``Series`` with incorrect dimensions (:issue:`24971`)
+-
+
+**Reshaping**
+
+- Bug in :meth:`.GroupBy.transform` where applying a function to a timezone aware column would return a timezone naive result (:issue:`24198`)
+- Bug in :func:`DataFrame.join` when joining on a timezone aware :class:`DatetimeIndex` (:issue:`23931`)
+
+**Visualization**
+
+- Bug in :meth:`Series.plot` where a secondary y axis could not be set to log scale (:issue:`25545`)
+
+**Other**
+
+- Bug in :meth:`Series.is_unique` where single occurrences of ``NaN`` were not considered unique (:issue:`25180`)
+- Bug in :func:`merge` when merging an empty ``DataFrame`` with an ``Int64`` column or a non-empty ``DataFrame`` with an ``Int64`` column that is all ``NaN`` (:issue:`25183`)
+- Bug in ``IntervalTree`` where a ``RecursionError`` occurs upon construction due to an overflow when adding endpoints, which also causes :class:`IntervalIndex` to crash during indexing operations (:issue:`25485`)
+- Bug in :attr:`Series.size` raising for some extension-array-backed ``Series``, rather than returning the size (:issue:`25580`)
+- Bug in resampling raising for nullable integer-dtype columns (:issue:`25580`)
+
+.. _whatsnew_0242.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. Including the contributors hardcoded for this release, as backporting with
+   MeeseeksDev loses the commit authors
+
+A total of 25 people contributed patches to this release. People with a "+" by their names contributed a patch for the first time.
+
+* Albert Villanova del Moral
+* Arno Veenstra +
+* chris-b1
+* Devin Petersohn +
+* EternalLearner42 +
+* Flavien Lambert +
+* gfyoung
+* Gioia Ballin
+* jbrockmendel
+* Jeff Reback
+* Jeremy Schendel
+* Johan von Forstner +
+* Joris Van den Bossche
+* Josh
+* Justin Zheng
+* Kendall Masse
+* Matthew Roeschke
+* Max Bolingbroke +
+* rbenes +
+* Sterling Paramore +
+* Tao He +
+* Thomas A Caswell
+* Tom Augspurger
+* Vibhu Agarwal +
+* William Ayd
+* Zach Angell
diff --git a/pandas/doc/source/whatsnew/v0.25.0.rst b/pandas/doc/source/whatsnew/v0.25.0.rst
new file mode 100644
index 0000000000000..bddb47cd3f629
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.25.0.rst
@@ -0,0 +1,1275 @@
+.. _whatsnew_0250:
+
+What's new in 0.25.0 (July 18, 2019)
+------------------------------------
+
+.. warning::
+
+   Starting with the 0.25.x series of releases, pandas only supports Python 3.5.3 and higher.
+   See `Dropping Python 2.7 <https://pandas.pydata.org/pandas-docs/version/0.24/install.html#install-dropping-27>`_ for more details.
+
+.. warning::
+
+   The minimum supported Python version will be bumped to 3.6 in a future release.
+
+.. warning::
+
+   ``Panel`` has been fully removed. For N-D labeled data structures, please
+   use `xarray <https://xarray.pydata.org/en/stable/>`_
+
+.. warning::
+
+   :func:`read_pickle` and :func:`read_msgpack` are only guaranteed backwards compatible back to
+   pandas version 0.20.3 (:issue:`27082`)
+
+{{ header }}
+
+These are the changes in pandas 0.25.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_0250.enhancements.agg_relabel:
+
+GroupBy aggregation with relabeling
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has added special groupby behavior, known as "named aggregation", for naming the
+output columns when applying multiple aggregation functions to specific columns (:issue:`18366`, :issue:`26512`).
+
+.. ipython:: python
+
+   animals = pd.DataFrame({'kind': ['cat', 'dog', 'cat', 'dog'],
+                           'height': [9.1, 6.0, 9.5, 34.0],
+                           'weight': [7.9, 7.5, 9.9, 198.0]})
+   animals
+   animals.groupby("kind").agg(
+       min_height=pd.NamedAgg(column='height', aggfunc='min'),
+       max_height=pd.NamedAgg(column='height', aggfunc='max'),
+       average_weight=pd.NamedAgg(column='weight', aggfunc="mean"),
+   )
+
+Pass the desired columns names as the ``**kwargs`` to ``.agg``. The values of ``**kwargs``
+should be tuples where the first element is the column selection, and the second element is the
+aggregation function to apply. pandas provides the ``pandas.NamedAgg`` namedtuple to make it clearer
+what the arguments to the function are, but plain tuples are accepted as well.
+
+.. ipython:: python
+
+   animals.groupby("kind").agg(
+       min_height=('height', 'min'),
+       max_height=('height', 'max'),
+       average_weight=('weight', 'mean'),
+   )
+
+Named aggregation is the recommended replacement for the deprecated "dict-of-dicts"
+approach to naming the output of column-specific aggregations (:ref:`whatsnew_0200.api_breaking.deprecate_group_agg_dict`).
+
+A similar approach is now available for Series groupby objects as well. Because there's no need for
+column selection, the values can just be the functions to apply
+
+.. ipython:: python
+
+   animals.groupby("kind").height.agg(
+       min_height="min",
+       max_height="max",
+   )
+
+
+This type of aggregation is the recommended alternative to the deprecated behavior when passing
+a dict to a Series groupby aggregation (:ref:`whatsnew_0200.api_breaking.deprecate_group_agg_dict`).
+
+See :ref:`groupby.aggregate.named` for more.
+
+.. _whatsnew_0250.enhancements.multiple_lambdas:
+
+GroupBy aggregation with multiple lambdas
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+You can now provide multiple lambda functions to a list-like aggregation in
+:class:`.GroupBy.agg` (:issue:`26430`).
+
+.. ipython:: python
+
+   animals.groupby('kind').height.agg([
+       lambda x: x.iloc[0], lambda x: x.iloc[-1]
+   ])
+
+   animals.groupby('kind').agg([
+       lambda x: x.iloc[0] - x.iloc[1],
+       lambda x: x.iloc[0] + x.iloc[1]
+   ])
+
+Previously, these raised a ``SpecificationError``.
+
+.. _whatsnew_0250.enhancements.multi_index_repr:
+
+Better repr for MultiIndex
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Printing of :class:`MultiIndex` instances now shows tuples of each row and ensures
+that the tuple items are vertically aligned, so it's now easier to understand
+the structure of the ``MultiIndex``. (:issue:`13480`):
+
+The repr now looks like this:
+
+.. ipython:: python
+
+   pd.MultiIndex.from_product([['a', 'abc'], range(500)])
+
+Previously, outputting a :class:`MultiIndex` printed all the ``levels`` and
+``codes`` of the ``MultiIndex``, which was visually unappealing and made
+the output more difficult to navigate. For example (limiting the range to 5):
+
+.. code-block:: ipython
+
+   In [1]: pd.MultiIndex.from_product([['a', 'abc'], range(5)])
+   Out[1]: MultiIndex(levels=[['a', 'abc'], [0, 1, 2, 3]],
+      ...:            codes=[[0, 0, 0, 0, 1, 1, 1, 1], [0, 1, 2, 3, 0, 1, 2, 3]])
+
+In the new repr, all values will be shown, if the number of rows is smaller
+than :attr:`options.display.max_seq_items` (default: 100 items). Horizontally,
+the output will truncate, if it's wider than :attr:`options.display.width`
+(default: 80 characters).
+
+.. _whatsnew_0250.enhancements.shorter_truncated_repr:
+
+Shorter truncated repr for Series and DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Currently, the default display options of pandas ensure that when a Series
+or DataFrame has more than 60 rows, its repr gets truncated to this maximum
+of 60 rows (the ``display.max_rows`` option). However, this still gives
+a repr that takes up a large part of the vertical screen estate. Therefore,
+a new option ``display.min_rows`` is introduced with a default of 10 which
+determines the number of rows showed in the truncated repr:
+
+- For small Series or DataFrames, up to ``max_rows`` number of rows is shown
+  (default: 60).
+- For larger Series of DataFrame with a length above ``max_rows``, only
+  ``min_rows`` number of rows is shown (default: 10, i.e. the first and last
+  5 rows).
+
+This dual option allows to still see the full content of relatively small
+objects (e.g. ``df.head(20)`` shows all 20 rows), while giving a brief repr
+for large objects.
+
+To restore the previous behaviour of a single threshold, set
+``pd.options.display.min_rows = None``.
+
+.. _whatsnew_0250.enhancements.json_normalize_with_max_level:
+
+JSON normalize with max_level param support
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`json_normalize` normalizes the provided input dict to all
+nested levels. The new max_level parameter provides more control over
+which level to end normalization (:issue:`23843`):
+
+The repr now looks like this:
+
+.. code-block:: ipython
+
+    from pandas.io.json import json_normalize
+    data = [{
+        'CreatedBy': {'Name': 'User001'},
+        'Lookup': {'TextField': 'Some text',
+                   'UserField': {'Id': 'ID001', 'Name': 'Name001'}},
+        'Image': {'a': 'b'}
+    }]
+    json_normalize(data, max_level=1)
+
+
+.. _whatsnew_0250.enhancements.explode:
+
+Series.explode to split list-like values to rows
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Series` and :class:`DataFrame` have gained the :meth:`DataFrame.explode` methods to transform list-likes to individual rows. See :ref:`section on Exploding list-like column <reshaping.explode>` in docs for more information (:issue:`16538`, :issue:`10511`)
+
+
+Here is a typical usecase. You have comma separated string in a column.
+
+.. ipython:: python
+
+    df = pd.DataFrame([{'var1': 'a,b,c', 'var2': 1},
+                       {'var1': 'd,e,f', 'var2': 2}])
+    df
+
+Creating a long form ``DataFrame`` is now straightforward using chained operations
+
+.. ipython:: python
+
+    df.assign(var1=df.var1.str.split(',')).explode('var1')
+
+.. _whatsnew_0250.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- :func:`DataFrame.plot` keywords ``logy``, ``logx`` and ``loglog`` can now accept the value ``'sym'`` for symlog scaling. (:issue:`24867`)
+- Added support for ISO week year format ('%G-%V-%u') when parsing datetimes using :meth:`to_datetime` (:issue:`16607`)
+- Indexing of ``DataFrame`` and ``Series`` now accepts zerodim ``np.ndarray`` (:issue:`24919`)
+- :meth:`Timestamp.replace` now supports the ``fold`` argument to disambiguate DST transition times (:issue:`25017`)
+- :meth:`DataFrame.at_time` and :meth:`Series.at_time` now support :class:`datetime.time` objects with timezones (:issue:`24043`)
+- :meth:`DataFrame.pivot_table` now accepts an ``observed`` parameter which is passed to underlying calls to :meth:`DataFrame.groupby` to speed up grouping categorical data. (:issue:`24923`)
+- ``Series.str`` has gained :meth:`Series.str.casefold` method to removes all case distinctions present in a string (:issue:`25405`)
+- :meth:`DataFrame.set_index` now works for instances of ``abc.Iterator``, provided their output is of the same length as the calling frame (:issue:`22484`, :issue:`24984`)
+- :meth:`DatetimeIndex.union` now supports the ``sort`` argument. The behavior of the sort parameter matches that of :meth:`Index.union` (:issue:`24994`)
+- :meth:`RangeIndex.union` now supports the ``sort`` argument. If ``sort=False`` an unsorted ``Int64Index`` is always returned. ``sort=None`` is the default and returns a monotonically increasing ``RangeIndex`` if possible or a sorted ``Int64Index`` if not (:issue:`24471`)
+- :meth:`TimedeltaIndex.intersection` now also supports the ``sort`` keyword (:issue:`24471`)
+- :meth:`DataFrame.rename` now supports the ``errors`` argument to raise errors when attempting to rename nonexistent keys (:issue:`13473`)
+- Added :ref:`api.frame.sparse` for working with a ``DataFrame`` whose values are sparse (:issue:`25681`)
+- :class:`RangeIndex` has gained :attr:`~RangeIndex.start`, :attr:`~RangeIndex.stop`, and :attr:`~RangeIndex.step` attributes (:issue:`25710`)
+- :class:`datetime.timezone` objects are now supported as arguments to timezone methods and constructors (:issue:`25065`)
+- :meth:`DataFrame.query` and :meth:`DataFrame.eval` now supports quoting column names with backticks to refer to names with spaces (:issue:`6508`)
+- :func:`merge_asof` now gives a more clear error message when merge keys are categoricals that are not equal (:issue:`26136`)
+- :meth:`.Rolling` supports exponential (or Poisson) window type (:issue:`21303`)
+- Error message for missing required imports now includes the original import error's text (:issue:`23868`)
+- :class:`DatetimeIndex` and :class:`TimedeltaIndex` now have a ``mean`` method (:issue:`24757`)
+- :meth:`DataFrame.describe` now formats integer percentiles without decimal point (:issue:`26660`)
+- Added support for reading SPSS .sav files using :func:`read_spss` (:issue:`26537`)
+- Added new option ``plotting.backend`` to be able to select a plotting backend different than the existing ``matplotlib`` one. Use ``pandas.set_option('plotting.backend', '<backend-module>')`` where ``<backend-module`` is a library implementing the pandas plotting API (:issue:`14130`)
+- :class:`pandas.offsets.BusinessHour` supports multiple opening hours intervals (:issue:`15481`)
+- :func:`read_excel` can now use ``openpyxl`` to read Excel files via the ``engine='openpyxl'`` argument. This will become the default in a future release (:issue:`11499`)
+- :func:`pandas.io.excel.read_excel` supports reading OpenDocument tables. Specify ``engine='odf'`` to enable. Consult the :ref:`IO User Guide <io.ods>` for more details (:issue:`9070`)
+- :class:`Interval`, :class:`IntervalIndex`, and :class:`~arrays.IntervalArray` have gained an :attr:`~Interval.is_empty` attribute denoting if the given interval(s) are empty (:issue:`27219`)
+
+.. _whatsnew_0250.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_0250.api_breaking.utc_offset_indexing:
+
+
+Indexing with date strings with UTC offsets
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Indexing a :class:`DataFrame` or :class:`Series` with a :class:`DatetimeIndex` with a
+date string with a UTC offset would previously ignore the UTC offset. Now, the UTC offset
+is respected in indexing. (:issue:`24076`, :issue:`16785`)
+
+.. ipython:: python
+
+    df = pd.DataFrame([0], index=pd.DatetimeIndex(['2019-01-01'], tz='US/Pacific'))
+    df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df['2019-01-01 00:00:00+04:00':'2019-01-01 01:00:00+04:00']
+    Out[3]:
+                               0
+    2019-01-01 00:00:00-08:00  0
+
+*New behavior*:
+
+.. ipython:: python
+
+    df['2019-01-01 12:00:00+04:00':'2019-01-01 13:00:00+04:00']
+
+
+.. _whatsnew_0250.api_breaking.multi_indexing:
+
+
+``MultiIndex`` constructed from levels and codes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Constructing a :class:`MultiIndex` with ``NaN`` levels or codes value < -1 was allowed previously.
+Now, construction with codes value < -1 is not allowed and ``NaN`` levels' corresponding codes
+would be reassigned as -1. (:issue:`19387`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [1]: pd.MultiIndex(levels=[[np.nan, None, pd.NaT, 128, 2]],
+       ...:               codes=[[0, -1, 1, 2, 3, 4]])
+       ...:
+    Out[1]: MultiIndex(levels=[[nan, None, NaT, 128, 2]],
+                       codes=[[0, -1, 1, 2, 3, 4]])
+
+    In [2]: pd.MultiIndex(levels=[[1, 2]], codes=[[0, -2]])
+    Out[2]: MultiIndex(levels=[[1, 2]],
+                       codes=[[0, -2]])
+
+*New behavior*:
+
+.. ipython:: python
+    :okexcept:
+
+    pd.MultiIndex(levels=[[np.nan, None, pd.NaT, 128, 2]],
+                  codes=[[0, -1, 1, 2, 3, 4]])
+    pd.MultiIndex(levels=[[1, 2]], codes=[[0, -2]])
+
+
+.. _whatsnew_0250.api_breaking.groupby_apply_first_group_once:
+
+``GroupBy.apply`` on ``DataFrame`` evaluates first group only once
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The implementation of :meth:`.DataFrameGroupBy.apply`
+previously evaluated the supplied function consistently twice on the first group
+to infer if it is safe to use a fast code path. Particularly for functions with
+side effects, this was an undesired behavior and may have led to surprises. (:issue:`2936`, :issue:`2656`, :issue:`7739`, :issue:`10519`, :issue:`12155`, :issue:`20084`, :issue:`21417`)
+
+Now every group is evaluated only a single time.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": ["x", "y"], "b": [1, 2]})
+    df
+
+    def func(group):
+        print(group.name)
+        return group
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [3]: df.groupby('a').apply(func)
+   x
+   x
+   y
+   Out[3]:
+      a  b
+   0  x  1
+   1  y  2
+
+*New behavior*:
+
+.. code-block:: python
+
+   In [3]: df.groupby('a').apply(func)
+   x
+   y
+   Out[3]:
+      a  b
+   0  x  1
+   1  y  2
+
+Concatenating sparse values
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When passed DataFrames whose values are sparse, :func:`concat` will now return a
+:class:`Series` or :class:`DataFrame` with sparse values, rather than a :class:`SparseDataFrame` (:issue:`25702`).
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 1])})
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [2]: type(pd.concat([df, df]))
+   pandas.core.sparse.frame.SparseDataFrame
+
+*New behavior*:
+
+.. ipython:: python
+
+   type(pd.concat([df, df]))
+
+
+This now matches the existing behavior of :class:`concat` on ``Series`` with sparse values.
+:func:`concat` will continue to return a ``SparseDataFrame`` when all the values
+are instances of ``SparseDataFrame``.
+
+This change also affects routines using :func:`concat` internally, like :func:`get_dummies`,
+which now returns a :class:`DataFrame` in all cases (previously a ``SparseDataFrame`` was
+returned if all the columns were dummy encoded, and a :class:`DataFrame` otherwise).
+
+Providing any ``SparseSeries`` or ``SparseDataFrame`` to :func:`concat` will
+cause a ``SparseSeries`` or ``SparseDataFrame`` to be returned, as before.
+
+The ``.str``-accessor performs stricter type checks
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Due to the lack of more fine-grained dtypes, :attr:`Series.str` so far only checked whether the data was
+of ``object`` dtype. :attr:`Series.str` will now infer the dtype data *within* the Series; in particular,
+``'bytes'``-only data will raise an exception (except for :meth:`Series.str.decode`, :meth:`Series.str.get`,
+:meth:`Series.str.len`, :meth:`Series.str.slice`), see :issue:`23163`, :issue:`23011`, :issue:`23551`.
+
+*Previous behavior*:
+
+.. code-block:: python
+
+    In [1]: s = pd.Series(np.array(['a', 'ba', 'cba'], 'S'), dtype=object)
+
+    In [2]: s
+    Out[2]:
+    0      b'a'
+    1     b'ba'
+    2    b'cba'
+    dtype: object
+
+    In [3]: s.str.startswith(b'a')
+    Out[3]:
+    0     True
+    1    False
+    2    False
+    dtype: bool
+
+*New behavior*:
+
+.. ipython:: python
+    :okexcept:
+
+    s = pd.Series(np.array(['a', 'ba', 'cba'], 'S'), dtype=object)
+    s
+    s.str.startswith(b'a')
+
+.. _whatsnew_0250.api_breaking.groupby_categorical:
+
+Categorical dtypes are preserved during GroupBy
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, columns that were categorical, but not the groupby key(s) would be converted to ``object`` dtype during groupby operations. pandas now will preserve these dtypes. (:issue:`18502`)
+
+.. ipython:: python
+
+   cat = pd.Categorical(["foo", "bar", "bar", "qux"], ordered=True)
+   df = pd.DataFrame({'payload': [-1, -2, -1, -2], 'col': cat})
+   df
+   df.dtypes
+
+*Previous Behavior*:
+
+.. code-block:: python
+
+   In [5]: df.groupby('payload').first().col.dtype
+   Out[5]: dtype('O')
+
+*New Behavior*:
+
+.. ipython:: python
+
+   df.groupby('payload').first().col.dtype
+
+
+.. _whatsnew_0250.api_breaking.incompatible_index_unions:
+
+Incompatible Index type unions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When performing :func:`Index.union` operations between objects of incompatible dtypes,
+the result will be a base :class:`Index` of dtype ``object``. This behavior holds true for
+unions between :class:`Index` objects that previously would have been prohibited. The dtype
+of empty :class:`Index` objects will now be evaluated before performing union operations
+rather than simply returning the other :class:`Index` object. :func:`Index.union` can now be
+considered commutative, such that ``A.union(B) == B.union(A)`` (:issue:`23525`).
+
+*Previous behavior*:
+
+.. code-block:: python
+
+    In [1]: pd.period_range('19910905', periods=2).union(pd.Int64Index([1, 2, 3]))
+    ...
+    ValueError: can only call with other PeriodIndex-ed objects
+
+    In [2]: pd.Index([], dtype=object).union(pd.Index([1, 2, 3]))
+    Out[2]: Int64Index([1, 2, 3], dtype='int64')
+
+*New behavior*:
+
+.. code-block:: python
+
+    In [3]: pd.period_range('19910905', periods=2).union(pd.Int64Index([1, 2, 3]))
+    Out[3]: Index([1991-09-05, 1991-09-06, 1, 2, 3], dtype='object')
+    In [4]: pd.Index([], dtype=object).union(pd.Index([1, 2, 3]))
+    Out[4]: Index([1, 2, 3], dtype='object')
+
+Note that integer- and floating-dtype indexes are considered "compatible". The integer
+values are coerced to floating point, which may result in loss of precision. See
+:ref:`indexing.set_ops` for more.
+
+
+``DataFrame`` GroupBy ffill/bfill no longer return group labels
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The methods ``ffill``, ``bfill``, ``pad`` and ``backfill`` of
+:class:`.DataFrameGroupBy`
+previously included the group labels in the return value, which was
+inconsistent with other groupby transforms. Now only the filled values
+are returned. (:issue:`21521`)
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": ["x", "y"], "b": [1, 2]})
+    df
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [3]: df.groupby("a").ffill()
+   Out[3]:
+      a  b
+   0  x  1
+   1  y  2
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby("a").ffill()
+
+``DataFrame`` describe on an empty Categorical / object column will return top and freq
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When calling :meth:`DataFrame.describe` with an empty categorical / object
+column, the 'top' and 'freq' columns were previously omitted, which was inconsistent with
+the output for non-empty columns. Now the 'top' and 'freq' columns will always be included,
+with :attr:`numpy.nan` in the case of an empty :class:`DataFrame` (:issue:`26397`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({"empty_col": pd.Categorical([])})
+   df
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [3]: df.describe()
+   Out[3]:
+           empty_col
+   count           0
+   unique          0
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.describe()
+
+``__str__`` methods now call ``__repr__`` rather than vice versa
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has until now mostly defined string representations in a pandas objects'
+``__str__``/``__unicode__``/``__bytes__`` methods, and called ``__str__`` from the ``__repr__``
+method, if a specific ``__repr__`` method is not found. This is not needed for Python3.
+In pandas 0.25, the string representations of pandas objects are now generally
+defined in ``__repr__``, and calls to ``__str__`` in general now pass the call on to
+the ``__repr__``, if a specific ``__str__`` method doesn't exist, as is standard for Python.
+This change is backward compatible for direct usage of pandas, but if you subclass
+pandas objects *and* give your subclasses specific ``__str__``/``__repr__`` methods,
+you may have to adjust your ``__str__``/``__repr__`` methods (:issue:`26495`).
+
+.. _whatsnew_0250.api_breaking.interval_indexing:
+
+
+Indexing an ``IntervalIndex`` with ``Interval`` objects
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Indexing methods for :class:`IntervalIndex` have been modified to require exact matches only for :class:`Interval` queries.
+``IntervalIndex`` methods previously matched on any overlapping ``Interval``.  Behavior with scalar points, e.g. querying
+with an integer, is unchanged (:issue:`16316`).
+
+.. ipython:: python
+
+   ii = pd.IntervalIndex.from_tuples([(0, 4), (1, 5), (5, 8)])
+   ii
+
+The ``in`` operator (``__contains__``) now only returns ``True`` for exact matches to ``Intervals`` in the ``IntervalIndex``, whereas
+this would previously return ``True`` for any ``Interval`` overlapping an ``Interval`` in the ``IntervalIndex``.
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [4]: pd.Interval(1, 2, closed='neither') in ii
+   Out[4]: True
+
+   In [5]: pd.Interval(-10, 10, closed='both') in ii
+   Out[5]: True
+
+*New behavior*:
+
+.. ipython:: python
+
+   pd.Interval(1, 2, closed='neither') in ii
+   pd.Interval(-10, 10, closed='both') in ii
+
+The :meth:`~IntervalIndex.get_loc` method now only returns locations for exact matches to ``Interval`` queries, as opposed to the previous behavior of
+returning locations for overlapping matches.  A ``KeyError`` will be raised if an exact match is not found.
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [6]: ii.get_loc(pd.Interval(1, 5))
+   Out[6]: array([0, 1])
+
+   In [7]: ii.get_loc(pd.Interval(2, 6))
+   Out[7]: array([0, 1, 2])
+
+*New behavior*:
+
+.. code-block:: python
+
+   In [6]: ii.get_loc(pd.Interval(1, 5))
+   Out[6]: 1
+
+   In [7]: ii.get_loc(pd.Interval(2, 6))
+   ---------------------------------------------------------------------------
+   KeyError: Interval(2, 6, closed='right')
+
+Likewise, :meth:`~IntervalIndex.get_indexer` and :meth:`~IntervalIndex.get_indexer_non_unique` will also only return locations for exact matches
+to ``Interval`` queries, with ``-1`` denoting that an exact match was not found.
+
+These indexing changes extend to querying a :class:`Series` or :class:`DataFrame` with an ``IntervalIndex`` index.
+
+.. ipython:: python
+
+   s = pd.Series(list('abc'), index=ii)
+   s
+
+Selecting from a ``Series`` or ``DataFrame`` using ``[]`` (``__getitem__``) or ``loc`` now only returns exact matches for ``Interval`` queries.
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [8]: s[pd.Interval(1, 5)]
+   Out[8]:
+   (0, 4]    a
+   (1, 5]    b
+   dtype: object
+
+   In [9]: s.loc[pd.Interval(1, 5)]
+   Out[9]:
+   (0, 4]    a
+   (1, 5]    b
+   dtype: object
+
+*New behavior*:
+
+.. ipython:: python
+
+   s[pd.Interval(1, 5)]
+   s.loc[pd.Interval(1, 5)]
+
+Similarly, a ``KeyError`` will be raised for non-exact matches instead of returning overlapping matches.
+
+*Previous behavior*:
+
+.. code-block:: python
+
+   In [9]: s[pd.Interval(2, 3)]
+   Out[9]:
+   (0, 4]    a
+   (1, 5]    b
+   dtype: object
+
+   In [10]: s.loc[pd.Interval(2, 3)]
+   Out[10]:
+   (0, 4]    a
+   (1, 5]    b
+   dtype: object
+
+*New behavior*:
+
+.. code-block:: python
+
+   In [6]: s[pd.Interval(2, 3)]
+   ---------------------------------------------------------------------------
+   KeyError: Interval(2, 3, closed='right')
+
+   In [7]: s.loc[pd.Interval(2, 3)]
+   ---------------------------------------------------------------------------
+   KeyError: Interval(2, 3, closed='right')
+
+The :meth:`~IntervalIndex.overlaps` method can be used to create a boolean indexer that replicates the
+previous behavior of returning overlapping matches.
+
+*New behavior*:
+
+.. ipython:: python
+
+   idxr = s.index.overlaps(pd.Interval(2, 3))
+   idxr
+   s[idxr]
+   s.loc[idxr]
+
+
+.. _whatsnew_0250.api_breaking.ufunc:
+
+Binary ufuncs on Series now align
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Applying a binary ufunc like :func:`numpy.power` now aligns the inputs
+when both are :class:`Series` (:issue:`23293`).
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, 3], index=['a', 'b', 'c'])
+   s2 = pd.Series([3, 4, 5], index=['d', 'c', 'b'])
+   s1
+   s2
+
+*Previous behavior*
+
+.. code-block:: ipython
+
+   In [5]: np.power(s1, s2)
+   Out[5]:
+   a      1
+   b     16
+   c    243
+   dtype: int64
+
+*New behavior*
+
+.. ipython:: python
+
+   np.power(s1, s2)
+
+This matches the behavior of other binary operations in pandas, like :meth:`Series.add`.
+To retain the previous behavior, convert the other ``Series`` to an array before
+applying the ufunc.
+
+.. ipython:: python
+
+   np.power(s1, s2.array)
+
+Categorical.argsort now places missing values at the end
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`Categorical.argsort` now places missing values at the end of the array, making it
+consistent with NumPy and the rest of pandas (:issue:`21801`).
+
+.. ipython:: python
+
+   cat = pd.Categorical(['b', None, 'a'], categories=['a', 'b'], ordered=True)
+
+*Previous behavior*
+
+.. code-block:: ipython
+
+   In [2]: cat = pd.Categorical(['b', None, 'a'], categories=['a', 'b'], ordered=True)
+
+   In [3]: cat.argsort()
+   Out[3]: array([1, 2, 0])
+
+   In [4]: cat[cat.argsort()]
+   Out[4]:
+   [NaN, a, b]
+   categories (2, object): [a < b]
+
+*New behavior*
+
+.. ipython:: python
+
+   cat.argsort()
+   cat[cat.argsort()]
+
+.. _whatsnew_0250.api_breaking.list_of_dict:
+
+Column order is preserved when passing a list of dicts to DataFrame
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Starting with Python 3.7 the key-order of ``dict`` is `guaranteed <https://mail.python.org/pipermail/python-dev/2017-December/151283.html>`_. In practice, this has been true since
+Python 3.6. The :class:`DataFrame` constructor now treats a list of dicts in the same way as
+it does a list of ``OrderedDict``, i.e. preserving the order of the dicts.
+This change applies only when pandas is running on Python>=3.6 (:issue:`27309`).
+
+.. ipython:: python
+
+   data = [
+       {'name': 'Joe', 'state': 'NY', 'age': 18},
+       {'name': 'Jane', 'state': 'KY', 'age': 19, 'hobby': 'Minecraft'},
+       {'name': 'Jean', 'state': 'OK', 'age': 20, 'finances': 'good'}
+   ]
+
+*Previous Behavior*:
+
+The columns were lexicographically sorted previously,
+
+.. code-block:: python
+
+   In [1]: pd.DataFrame(data)
+   Out[1]:
+      age finances      hobby  name state
+   0   18      NaN        NaN   Joe    NY
+   1   19      NaN  Minecraft  Jane    KY
+   2   20     good        NaN  Jean    OK
+
+*New Behavior*:
+
+The column order now matches the insertion-order of the keys in the ``dict``,
+considering all the records from top to bottom. As a consequence, the column
+order of the resulting DataFrame has changed compared to previous pandas versions.
+
+.. ipython:: python
+
+   pd.DataFrame(data)
+
+.. _whatsnew_0250.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Due to dropping support for Python 2.7, a number of optional dependencies have updated minimum versions (:issue:`25725`, :issue:`24942`, :issue:`25752`).
+Independently, some minimum supported versions of dependencies were updated (:issue:`23519`, :issue:`25554`).
+If installed, we now require:
+
++-----------------+-----------------+----------+
+| Package         | Minimum Version | Required |
++=================+=================+==========+
+| numpy           | 1.13.3          |    X     |
++-----------------+-----------------+----------+
+| pytz            | 2015.4          |    X     |
++-----------------+-----------------+----------+
+| python-dateutil | 2.6.1           |    X     |
++-----------------+-----------------+----------+
+| bottleneck      | 1.2.1           |          |
++-----------------+-----------------+----------+
+| numexpr         | 2.6.2           |          |
++-----------------+-----------------+----------+
+| pytest (dev)    | 4.0.2           |          |
++-----------------+-----------------+----------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+
+| Package         | Minimum Version |
++=================+=================+
+| beautifulsoup4  | 4.6.0           |
++-----------------+-----------------+
+| fastparquet     | 0.2.1           |
++-----------------+-----------------+
+| gcsfs           | 0.2.2           |
++-----------------+-----------------+
+| lxml            | 3.8.0           |
++-----------------+-----------------+
+| matplotlib      | 2.2.2           |
++-----------------+-----------------+
+| openpyxl        | 2.4.8           |
++-----------------+-----------------+
+| pyarrow         | 0.9.0           |
++-----------------+-----------------+
+| pymysql         | 0.7.1           |
++-----------------+-----------------+
+| pytables        | 3.4.2           |
++-----------------+-----------------+
+| scipy           | 0.19.0          |
++-----------------+-----------------+
+| sqlalchemy      | 1.1.4           |
++-----------------+-----------------+
+| xarray          | 0.8.2           |
++-----------------+-----------------+
+| xlrd            | 1.1.0           |
++-----------------+-----------------+
+| xlsxwriter      | 0.9.8           |
++-----------------+-----------------+
+| xlwt            | 1.2.0           |
++-----------------+-----------------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_0250.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- :class:`DatetimeTZDtype` will now standardize pytz timezones to a common timezone instance (:issue:`24713`)
+- :class:`Timestamp` and :class:`Timedelta` scalars now implement the :meth:`to_numpy` method as aliases to :meth:`Timestamp.to_datetime64` and :meth:`Timedelta.to_timedelta64`, respectively. (:issue:`24653`)
+- :meth:`Timestamp.strptime` will now rise a ``NotImplementedError`` (:issue:`25016`)
+- Comparing :class:`Timestamp` with unsupported objects now returns :py:obj:`NotImplemented` instead of raising ``TypeError``. This implies that unsupported rich comparisons are delegated to the other object, and are now consistent with Python 3 behavior for ``datetime`` objects (:issue:`24011`)
+- Bug in :meth:`DatetimeIndex.snap` which didn't preserving the ``name`` of the input :class:`Index` (:issue:`25575`)
+- The ``arg`` argument in :meth:`.DataFrameGroupBy.agg` has been renamed to ``func`` (:issue:`26089`)
+- The ``arg`` argument in :meth:`.Window.aggregate` has been renamed to ``func`` (:issue:`26372`)
+- Most pandas classes had a ``__bytes__`` method, which was used for getting a python2-style bytestring representation of the object. This method has been removed as a part of dropping Python2 (:issue:`26447`)
+- The ``.str``-accessor has been disabled for 1-level :class:`MultiIndex`, use :meth:`MultiIndex.to_flat_index` if necessary (:issue:`23679`)
+- Removed support of gtk package for clipboards (:issue:`26563`)
+- Using an unsupported version of Beautiful Soup 4 will now raise an ``ImportError`` instead of a ``ValueError`` (:issue:`27063`)
+- :meth:`Series.to_excel` and :meth:`DataFrame.to_excel` will now raise a ``ValueError`` when saving timezone aware data. (:issue:`27008`, :issue:`7056`)
+- :meth:`ExtensionArray.argsort` places NA values at the end of the sorted array. (:issue:`21801`)
+- :meth:`DataFrame.to_hdf` and :meth:`Series.to_hdf` will now raise a ``NotImplementedError`` when saving a :class:`MultiIndex` with extension data types for a ``fixed`` format. (:issue:`7775`)
+- Passing duplicate ``names`` in :meth:`read_csv` will now raise a ``ValueError`` (:issue:`17346`)
+
+.. _whatsnew_0250.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+Sparse subclasses
+^^^^^^^^^^^^^^^^^
+
+The ``SparseSeries`` and ``SparseDataFrame`` subclasses are deprecated. Their functionality is better-provided
+by a ``Series`` or ``DataFrame`` with sparse values.
+
+**Previous way**
+
+.. code-block:: python
+
+   df = pd.SparseDataFrame({"A": [0, 0, 1, 2]})
+   df.dtypes
+
+**New way**
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 0, 1, 2])})
+   df.dtypes
+
+The memory usage of the two approaches is identical (:issue:`19239`).
+
+msgpack format
+^^^^^^^^^^^^^^
+
+The msgpack format is deprecated as of 0.25 and will be removed in a future version. It is recommended to use pyarrow for on-the-wire transmission of pandas objects. (:issue:`27084`)
+
+
+Other deprecations
+^^^^^^^^^^^^^^^^^^
+
+- The deprecated ``.ix[]`` indexer now raises a more visible ``FutureWarning`` instead of ``DeprecationWarning`` (:issue:`26438`).
+- Deprecated the ``units=M`` (months) and ``units=Y`` (year) parameters for ``units`` of :func:`pandas.to_timedelta`, :func:`pandas.Timedelta` and :func:`pandas.TimedeltaIndex` (:issue:`16344`)
+- :meth:`pandas.concat` has deprecated the ``join_axes``-keyword. Instead, use :meth:`DataFrame.reindex` or :meth:`DataFrame.reindex_like` on the result or on the inputs (:issue:`21951`)
+- The :attr:`SparseArray.values` attribute is deprecated. You can use ``np.asarray(...)`` or
+  the :meth:`SparseArray.to_dense` method instead (:issue:`26421`).
+- The functions :func:`pandas.to_datetime` and :func:`pandas.to_timedelta` have deprecated the ``box`` keyword. Instead, use :meth:`to_numpy` or :meth:`Timestamp.to_datetime64` or :meth:`Timedelta.to_timedelta64`. (:issue:`24416`)
+- The :meth:`DataFrame.compound` and :meth:`Series.compound` methods are deprecated and will be removed in a future version (:issue:`26405`).
+- The internal attributes ``_start``, ``_stop`` and ``_step`` attributes of :class:`RangeIndex` have been deprecated.
+  Use the public attributes :attr:`~RangeIndex.start`, :attr:`~RangeIndex.stop` and :attr:`~RangeIndex.step` instead (:issue:`26581`).
+- The :meth:`Series.ftype`, :meth:`Series.ftypes` and :meth:`DataFrame.ftypes` methods are deprecated and will be removed in a future version.
+  Instead, use :meth:`Series.dtype` and :meth:`DataFrame.dtypes` (:issue:`26705`).
+- The :meth:`Series.get_values`, :meth:`DataFrame.get_values`, :meth:`Index.get_values`,
+  :meth:`SparseArray.get_values` and :meth:`Categorical.get_values` methods are deprecated.
+  One of ``np.asarray(..)`` or :meth:`~Series.to_numpy` can be used instead (:issue:`19617`).
+- The 'outer' method on NumPy ufuncs, e.g. ``np.subtract.outer`` has been deprecated on :class:`Series` objects. Convert the input to an array with :attr:`Series.array` first (:issue:`27186`)
+- :meth:`Timedelta.resolution` is deprecated and replaced with :meth:`Timedelta.resolution_string`.  In a future version, :meth:`Timedelta.resolution` will be changed to behave like the standard library :attr:`datetime.timedelta.resolution` (:issue:`21344`)
+- :func:`read_table` has been undeprecated. (:issue:`25220`)
+- :attr:`Index.dtype_str` is deprecated. (:issue:`18262`)
+- :attr:`Series.imag` and :attr:`Series.real` are deprecated. (:issue:`18262`)
+- :meth:`Series.put` is deprecated. (:issue:`18262`)
+- :meth:`Index.item` and :meth:`Series.item` is deprecated. (:issue:`18262`)
+- The default value ``ordered=None`` in :class:`~pandas.api.types.CategoricalDtype` has been deprecated in favor of ``ordered=False``. When converting between categorical types ``ordered=True`` must be explicitly passed in order to be preserved. (:issue:`26336`)
+- :meth:`Index.contains` is deprecated. Use ``key in index`` (``__contains__``) instead (:issue:`17753`).
+- :meth:`DataFrame.get_dtype_counts` is deprecated. (:issue:`18262`)
+- :meth:`Categorical.ravel` will return a :class:`Categorical` instead of a ``np.ndarray`` (:issue:`27199`)
+
+
+.. _whatsnew_0250.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+- Removed ``Panel`` (:issue:`25047`, :issue:`25191`, :issue:`25231`)
+- Removed the previously deprecated ``sheetname`` keyword in :func:`read_excel` (:issue:`16442`, :issue:`20938`)
+- Removed the previously deprecated ``TimeGrouper`` (:issue:`16942`)
+- Removed the previously deprecated ``parse_cols`` keyword in :func:`read_excel` (:issue:`16488`)
+- Removed the previously deprecated ``pd.options.html.border`` (:issue:`16970`)
+- Removed the previously deprecated ``convert_objects`` (:issue:`11221`)
+- Removed the previously deprecated ``select`` method of ``DataFrame`` and ``Series`` (:issue:`17633`)
+- Removed the previously deprecated behavior of :class:`Series` treated as list-like in :meth:`~Series.cat.rename_categories` (:issue:`17982`)
+- Removed the previously deprecated ``DataFrame.reindex_axis`` and ``Series.reindex_axis`` (:issue:`17842`)
+- Removed the previously deprecated behavior of altering column or index labels with :meth:`Series.rename_axis` or :meth:`DataFrame.rename_axis` (:issue:`17842`)
+- Removed the previously deprecated ``tupleize_cols`` keyword argument in :meth:`read_html`, :meth:`read_csv`, and :meth:`DataFrame.to_csv` (:issue:`17877`, :issue:`17820`)
+- Removed the previously deprecated ``DataFrame.from.csv`` and ``Series.from_csv`` (:issue:`17812`)
+- Removed the previously deprecated ``raise_on_error`` keyword argument in :meth:`DataFrame.where` and :meth:`DataFrame.mask` (:issue:`17744`)
+- Removed the previously deprecated ``ordered`` and ``categories`` keyword arguments in ``astype`` (:issue:`17742`)
+- Removed the previously deprecated ``cdate_range`` (:issue:`17691`)
+- Removed the previously deprecated ``True`` option for the ``dropna`` keyword argument in :func:`SeriesGroupBy.nth` (:issue:`17493`)
+- Removed the previously deprecated ``convert`` keyword argument in :meth:`Series.take` and :meth:`DataFrame.take` (:issue:`17352`)
+- Removed the previously deprecated behavior of arithmetic operations with ``datetime.date`` objects (:issue:`21152`)
+
+.. _whatsnew_0250.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Significant speedup in :class:`SparseArray` initialization that benefits most operations, fixing performance regression introduced in v0.20.0 (:issue:`24985`)
+- :meth:`DataFrame.to_stata` is now faster when outputting data with any string or non-native endian columns (:issue:`25045`)
+- Improved performance of :meth:`Series.searchsorted`. The speedup is especially large when the dtype is
+  int8/int16/int32 and the searched key is within the integer bounds for the dtype (:issue:`22034`)
+- Improved performance of :meth:`.GroupBy.quantile` (:issue:`20405`)
+- Improved performance of slicing and other selected operation on a :class:`RangeIndex` (:issue:`26565`, :issue:`26617`, :issue:`26722`)
+- :class:`RangeIndex` now performs standard lookup without instantiating an actual hashtable, hence saving memory (:issue:`16685`)
+- Improved performance of :meth:`read_csv` by faster tokenizing and faster parsing of small float numbers (:issue:`25784`)
+- Improved performance of :meth:`read_csv` by faster parsing of N/A and boolean values (:issue:`25804`)
+- Improved performance of :attr:`IntervalIndex.is_monotonic`, :attr:`IntervalIndex.is_monotonic_increasing` and :attr:`IntervalIndex.is_monotonic_decreasing` by removing conversion to :class:`MultiIndex` (:issue:`24813`)
+- Improved performance of :meth:`DataFrame.to_csv` when writing datetime dtypes (:issue:`25708`)
+- Improved performance of :meth:`read_csv` by much faster parsing of ``MM/YYYY`` and ``DD/MM/YYYY`` datetime formats (:issue:`25922`)
+- Improved performance of nanops for dtypes that cannot store NaNs. Speedup is particularly prominent for :meth:`Series.all` and :meth:`Series.any` (:issue:`25070`)
+- Improved performance of :meth:`Series.map` for dictionary mappers on categorical series by mapping the categories instead of mapping all values (:issue:`23785`)
+- Improved performance of :meth:`IntervalIndex.intersection` (:issue:`24813`)
+- Improved performance of :meth:`read_csv` by faster concatenating date columns without extra conversion to string for integer/float zero and float ``NaN``; by faster checking the string for the possibility of being a date (:issue:`25754`)
+- Improved performance of :attr:`IntervalIndex.is_unique` by removing conversion to ``MultiIndex`` (:issue:`24813`)
+- Restored performance of :meth:`DatetimeIndex.__iter__` by re-enabling specialized code path (:issue:`26702`)
+- Improved performance when building :class:`MultiIndex` with at least one :class:`CategoricalIndex` level (:issue:`22044`)
+- Improved performance by removing the need for a garbage collect when checking for ``SettingWithCopyWarning`` (:issue:`27031`)
+- For :meth:`to_datetime` changed default value of cache parameter to ``True`` (:issue:`26043`)
+- Improved performance of :class:`DatetimeIndex` and :class:`PeriodIndex` slicing given non-unique, monotonic data (:issue:`27136`).
+- Improved performance of :meth:`pd.read_json` for index-oriented data. (:issue:`26773`)
+- Improved performance of :meth:`MultiIndex.shape` (:issue:`27384`).
+
+.. _whatsnew_0250.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+
+Categorical
+^^^^^^^^^^^
+
+- Bug in :func:`DataFrame.at` and :func:`Series.at` that would raise exception if the index was a :class:`CategoricalIndex` (:issue:`20629`)
+- Fixed bug in comparison of ordered :class:`Categorical` that contained missing values with a scalar which sometimes incorrectly resulted in ``True`` (:issue:`26504`)
+- Bug in :meth:`DataFrame.dropna` when the :class:`DataFrame` has a :class:`CategoricalIndex` containing :class:`Interval` objects incorrectly raised a ``TypeError`` (:issue:`25087`)
+
+Datetimelike
+^^^^^^^^^^^^
+
+- Bug in :func:`to_datetime` which would raise an (incorrect) ``ValueError`` when called with a date far into the future and the ``format`` argument specified instead of raising ``OutOfBoundsDatetime`` (:issue:`23830`)
+- Bug in :func:`to_datetime` which would raise ``InvalidIndexError: Reindexing only valid with uniquely valued Index objects`` when called with ``cache=True``, with ``arg`` including at least two different elements from the set ``{None, numpy.nan, pandas.NaT}`` (:issue:`22305`)
+- Bug in :class:`DataFrame` and :class:`Series` where timezone aware data with ``dtype='datetime64[ns]`` was not cast to naive (:issue:`25843`)
+- Improved :class:`Timestamp` type checking in various datetime functions to prevent exceptions when using a subclassed ``datetime`` (:issue:`25851`)
+- Bug in :class:`Series` and :class:`DataFrame` repr where ``np.datetime64('NaT')`` and ``np.timedelta64('NaT')`` with ``dtype=object`` would be represented as ``NaN`` (:issue:`25445`)
+- Bug in :func:`to_datetime` which does not replace the invalid argument with ``NaT`` when error is set to coerce (:issue:`26122`)
+- Bug in adding :class:`DateOffset` with nonzero month to :class:`DatetimeIndex` would raise ``ValueError`` (:issue:`26258`)
+- Bug in :func:`to_datetime` which raises unhandled ``OverflowError`` when called with mix of invalid dates and ``NaN`` values with ``format='%Y%m%d'`` and ``error='coerce'`` (:issue:`25512`)
+- Bug in :meth:`isin` for datetimelike indexes; :class:`DatetimeIndex`, :class:`TimedeltaIndex` and :class:`PeriodIndex` where the ``levels`` parameter was ignored. (:issue:`26675`)
+- Bug in :func:`to_datetime` which raises ``TypeError`` for ``format='%Y%m%d'`` when called for invalid integer dates with length >= 6 digits with ``errors='ignore'``
+- Bug when comparing a :class:`PeriodIndex` against a zero-dimensional numpy array (:issue:`26689`)
+- Bug in constructing a ``Series`` or ``DataFrame`` from a numpy ``datetime64`` array with a non-ns unit and out-of-bound timestamps generating rubbish data, which will now correctly raise an ``OutOfBoundsDatetime`` error (:issue:`26206`).
+- Bug in :func:`date_range` with unnecessary ``OverflowError`` being raised for very large or very small dates (:issue:`26651`)
+- Bug where adding :class:`Timestamp` to a ``np.timedelta64`` object would raise instead of returning a :class:`Timestamp` (:issue:`24775`)
+- Bug where comparing a zero-dimensional numpy array containing a ``np.datetime64`` object to a :class:`Timestamp` would incorrect raise ``TypeError`` (:issue:`26916`)
+- Bug in :func:`to_datetime` which would raise ``ValueError: Tz-aware datetime.datetime cannot be converted to datetime64 unless utc=True`` when called with ``cache=True``, with ``arg`` including datetime strings with different offset (:issue:`26097`)
+-
+
+Timedelta
+^^^^^^^^^
+
+- Bug in :func:`TimedeltaIndex.intersection` where for non-monotonic indices in some cases an empty ``Index`` was returned when in fact an intersection existed (:issue:`25913`)
+- Bug with comparisons between :class:`Timedelta` and ``NaT`` raising ``TypeError`` (:issue:`26039`)
+- Bug when adding or subtracting a :class:`BusinessHour` to a :class:`Timestamp` with the resulting time landing in a following or prior day respectively (:issue:`26381`)
+- Bug when comparing a :class:`TimedeltaIndex` against a zero-dimensional numpy array (:issue:`26689`)
+
+Timezones
+^^^^^^^^^
+
+- Bug in :func:`DatetimeIndex.to_frame` where timezone aware data would be converted to timezone naive data (:issue:`25809`)
+- Bug in :func:`to_datetime` with ``utc=True`` and datetime strings that would apply previously parsed UTC offsets to subsequent arguments (:issue:`24992`)
+- Bug in :func:`Timestamp.tz_localize` and :func:`Timestamp.tz_convert` does not propagate ``freq`` (:issue:`25241`)
+- Bug in :func:`Series.at` where setting :class:`Timestamp` with timezone raises ``TypeError`` (:issue:`25506`)
+- Bug in :func:`DataFrame.update` when updating with timezone aware data would return timezone naive data (:issue:`25807`)
+- Bug in :func:`to_datetime` where an uninformative ``RuntimeError`` was raised when passing a naive :class:`Timestamp` with datetime strings with mixed UTC offsets (:issue:`25978`)
+- Bug in :func:`to_datetime` with ``unit='ns'`` would drop timezone information from the parsed argument (:issue:`26168`)
+- Bug in :func:`DataFrame.join` where joining a timezone aware index with a timezone aware column would result in a column of ``NaN`` (:issue:`26335`)
+- Bug in :func:`date_range` where ambiguous or nonexistent start or end times were not handled by the ``ambiguous`` or ``nonexistent`` keywords respectively (:issue:`27088`)
+- Bug in :meth:`DatetimeIndex.union` when combining a timezone aware and timezone unaware :class:`DatetimeIndex` (:issue:`21671`)
+- Bug when applying a numpy reduction function (e.g. :meth:`numpy.minimum`) to a timezone aware :class:`Series` (:issue:`15552`)
+
+Numeric
+^^^^^^^
+
+- Bug in :meth:`to_numeric` in which large negative numbers were being improperly handled (:issue:`24910`)
+- Bug in :meth:`to_numeric` in which numbers were being coerced to float, even though ``errors`` was not ``coerce`` (:issue:`24910`)
+- Bug in :meth:`to_numeric` in which invalid values for ``errors`` were being allowed (:issue:`26466`)
+- Bug in :class:`format` in which floating point complex numbers were not being formatted to proper display precision and trimming (:issue:`25514`)
+- Bug in error messages in :meth:`DataFrame.corr` and :meth:`Series.corr`. Added the possibility of using a callable. (:issue:`25729`)
+- Bug in :meth:`Series.divmod` and :meth:`Series.rdivmod` which would raise an (incorrect) ``ValueError`` rather than return a pair of :class:`Series` objects as result (:issue:`25557`)
+- Raises a helpful exception when a non-numeric index is sent to :meth:`interpolate` with methods which require numeric index. (:issue:`21662`)
+- Bug in :meth:`~pandas.eval` when comparing floats with scalar operators, for example: ``x < -0.1`` (:issue:`25928`)
+- Fixed bug where casting all-boolean array to integer extension array failed (:issue:`25211`)
+- Bug in ``divmod`` with a :class:`Series` object containing zeros incorrectly raising ``AttributeError`` (:issue:`26987`)
+- Inconsistency in :class:`Series` floor-division (``//``) and ``divmod`` filling positive//zero with ``NaN`` instead of ``Inf`` (:issue:`27321`)
+-
+
+Conversion
+^^^^^^^^^^
+
+- Bug in :func:`DataFrame.astype` when passing a dict of columns and types the ``errors`` parameter was ignored. (:issue:`25905`)
+-
+
+Strings
+^^^^^^^
+
+- Bug in the ``__name__`` attribute of several methods of :class:`Series.str`, which were set incorrectly (:issue:`23551`)
+- Improved error message when passing :class:`Series` of wrong dtype to :meth:`Series.str.cat` (:issue:`22722`)
+-
+
+
+Interval
+^^^^^^^^
+
+- Construction of :class:`Interval` is restricted to numeric, :class:`Timestamp` and :class:`Timedelta` endpoints (:issue:`23013`)
+- Fixed bug in :class:`Series`/:class:`DataFrame` not displaying ``NaN`` in :class:`IntervalIndex` with missing values (:issue:`25984`)
+- Bug in :meth:`IntervalIndex.get_loc` where a ``KeyError`` would be incorrectly raised for a decreasing :class:`IntervalIndex` (:issue:`25860`)
+- Bug in :class:`Index` constructor where passing mixed closed :class:`Interval` objects would result in a ``ValueError`` instead of an ``object`` dtype ``Index`` (:issue:`27172`)
+
+Indexing
+^^^^^^^^
+
+- Improved exception message when calling :meth:`DataFrame.iloc` with a list of non-numeric objects (:issue:`25753`).
+- Improved exception message when calling ``.iloc`` or ``.loc`` with a boolean indexer with different length (:issue:`26658`).
+- Bug in ``KeyError`` exception message when indexing a :class:`MultiIndex` with a non-existent key not displaying the original key (:issue:`27250`).
+- Bug in ``.iloc`` and ``.loc`` with a boolean indexer not raising an ``IndexError`` when too few items are passed (:issue:`26658`).
+- Bug in :meth:`DataFrame.loc` and :meth:`Series.loc` where ``KeyError`` was not raised for a ``MultiIndex`` when the key was less than or equal to the number of levels in the :class:`MultiIndex` (:issue:`14885`).
+- Bug in which :meth:`DataFrame.append` produced an erroneous warning indicating that a ``KeyError`` will be thrown in the future when the data to be appended contains new columns (:issue:`22252`).
+- Bug in which :meth:`DataFrame.to_csv` caused a segfault for a reindexed data frame, when the indices were single-level :class:`MultiIndex` (:issue:`26303`).
+- Fixed bug where assigning a :class:`arrays.PandasArray` to a :class:`.DataFrame` would raise error (:issue:`26390`)
+- Allow keyword arguments for callable local reference used in the :meth:`DataFrame.query` string (:issue:`26426`)
+- Fixed a ``KeyError`` when indexing a :class:`MultiIndex` level with a list containing exactly one label, which is missing (:issue:`27148`)
+- Bug which produced ``AttributeError`` on partial matching :class:`Timestamp` in a :class:`MultiIndex`  (:issue:`26944`)
+- Bug in :class:`Categorical` and  :class:`CategoricalIndex` with :class:`Interval` values when using the ``in`` operator (``__contains``) with objects that are not comparable to the values in the ``Interval`` (:issue:`23705`)
+- Bug in :meth:`DataFrame.loc` and :meth:`DataFrame.iloc` on a :class:`DataFrame` with a single timezone-aware datetime64[ns] column incorrectly returning a scalar instead of a :class:`Series` (:issue:`27110`)
+- Bug in :class:`CategoricalIndex` and :class:`Categorical` incorrectly raising ``ValueError`` instead of ``TypeError`` when a list is passed using the ``in`` operator (``__contains__``) (:issue:`21729`)
+- Bug in setting a new value in a :class:`Series` with a :class:`Timedelta` object incorrectly casting the value to an integer (:issue:`22717`)
+- Bug in :class:`Series` setting a new key (``__setitem__``) with a timezone-aware datetime incorrectly raising ``ValueError`` (:issue:`12862`)
+- Bug in :meth:`DataFrame.iloc` when indexing with a read-only indexer (:issue:`17192`)
+- Bug in :class:`Series` setting an existing tuple key (``__setitem__``) with timezone-aware datetime values incorrectly raising ``TypeError`` (:issue:`20441`)
+
+Missing
+^^^^^^^
+
+- Fixed misleading exception message in :meth:`Series.interpolate` if argument ``order`` is required, but omitted (:issue:`10633`, :issue:`24014`).
+- Fixed class type displayed in exception message in :meth:`DataFrame.dropna` if invalid ``axis`` parameter passed (:issue:`25555`)
+- A ``ValueError`` will now be thrown by :meth:`DataFrame.fillna` when ``limit`` is not a positive integer (:issue:`27042`)
+-
+
+MultiIndex
+^^^^^^^^^^
+
+- Bug in which incorrect exception raised by :class:`Timedelta` when testing the membership of :class:`MultiIndex` (:issue:`24570`)
+-
+
+IO
+^^
+
+- Bug in :func:`DataFrame.to_html` where values were truncated using display options instead of outputting the full content (:issue:`17004`)
+- Fixed bug in missing text when using :meth:`to_clipboard` if copying utf-16 characters in Python 3 on Windows (:issue:`25040`)
+- Bug in :func:`read_json` for ``orient='table'`` when it tries to infer dtypes by default, which is not applicable as dtypes are already defined in the JSON schema (:issue:`21345`)
+- Bug in :func:`read_json` for ``orient='table'`` and float index, as it infers index dtype by default, which is not applicable because index dtype is already defined in the JSON schema (:issue:`25433`)
+- Bug in :func:`read_json` for ``orient='table'`` and string of float column names, as it makes a column name type conversion to :class:`Timestamp`, which is not applicable because column names are already defined in the JSON schema (:issue:`25435`)
+- Bug in :func:`json_normalize` for ``errors='ignore'`` where missing values in the input data, were filled in resulting ``DataFrame`` with the string ``"nan"`` instead of ``numpy.nan`` (:issue:`25468`)
+- :meth:`DataFrame.to_html` now raises ``TypeError`` when using an invalid type for the ``classes`` parameter instead of ``AssertionError`` (:issue:`25608`)
+- Bug in :meth:`DataFrame.to_string` and :meth:`DataFrame.to_latex` that would lead to incorrect output when the ``header`` keyword is used (:issue:`16718`)
+- Bug in :func:`read_csv` not properly interpreting the UTF8 encoded filenames on Windows on Python 3.6+ (:issue:`15086`)
+- Improved performance in :meth:`pandas.read_stata` and :class:`pandas.io.stata.StataReader` when converting columns that have missing values (:issue:`25772`)
+- Bug in :meth:`DataFrame.to_html` where header numbers would ignore display options when rounding (:issue:`17280`)
+- Bug in :func:`read_hdf` where reading a table from an HDF5 file written directly with PyTables fails with a ``ValueError`` when using a sub-selection via the ``start`` or ``stop`` arguments (:issue:`11188`)
+- Bug in :func:`read_hdf` not properly closing store after a ``KeyError`` is raised (:issue:`25766`)
+- Improved the explanation for the failure when value labels are repeated in Stata dta files and suggested workarounds (:issue:`25772`)
+- Improved :meth:`pandas.read_stata` and :class:`pandas.io.stata.StataReader` to read incorrectly formatted 118 format files saved by Stata (:issue:`25960`)
+- Improved the ``col_space`` parameter in :meth:`DataFrame.to_html` to accept a string so CSS length values can be set correctly (:issue:`25941`)
+- Fixed bug in loading objects from S3 that contain ``#`` characters in the URL (:issue:`25945`)
+- Adds ``use_bqstorage_api`` parameter to :func:`read_gbq` to speed up downloads of large data frames. This feature requires version 0.10.0 of the ``pandas-gbq`` library as well as the ``google-cloud-bigquery-storage`` and ``fastavro`` libraries. (:issue:`26104`)
+- Fixed memory leak in :meth:`DataFrame.to_json` when dealing with numeric data (:issue:`24889`)
+- Bug in :func:`read_json` where date strings with ``Z`` were not converted to a UTC timezone (:issue:`26168`)
+- Added ``cache_dates=True`` parameter to :meth:`read_csv`, which allows to cache unique dates when they are parsed (:issue:`25990`)
+- :meth:`DataFrame.to_excel` now raises a ``ValueError`` when the caller's dimensions exceed the limitations of Excel (:issue:`26051`)
+- Fixed bug in :func:`pandas.read_csv` where a BOM would result in incorrect parsing using engine='python' (:issue:`26545`)
+- :func:`read_excel` now raises a ``ValueError`` when input is of type :class:`pandas.io.excel.ExcelFile` and ``engine`` param is passed since :class:`pandas.io.excel.ExcelFile` has an engine defined (:issue:`26566`)
+- Bug while selecting from :class:`HDFStore` with ``where=''`` specified (:issue:`26610`).
+- Fixed bug in :func:`DataFrame.to_excel` where custom objects (i.e. ``PeriodIndex``) inside merged cells were not being converted into types safe for the Excel writer (:issue:`27006`)
+- Bug in :meth:`read_hdf` where reading a timezone aware :class:`DatetimeIndex` would raise a ``TypeError`` (:issue:`11926`)
+- Bug in :meth:`to_msgpack` and :meth:`read_msgpack` which would raise a ``ValueError`` rather than a ``FileNotFoundError`` for an invalid path (:issue:`27160`)
+- Fixed bug in :meth:`DataFrame.to_parquet` which would raise a ``ValueError`` when the dataframe had no columns (:issue:`27339`)
+- Allow parsing of :class:`PeriodDtype` columns when using :func:`read_csv` (:issue:`26934`)
+
+Plotting
+^^^^^^^^
+
+- Fixed bug where :class:`api.extensions.ExtensionArray` could not be used in matplotlib plotting (:issue:`25587`)
+- Bug in an error message in :meth:`DataFrame.plot`. Improved the error message if non-numerics are passed to :meth:`DataFrame.plot` (:issue:`25481`)
+- Bug in incorrect ticklabel positions when plotting an index that are non-numeric / non-datetime (:issue:`7612`, :issue:`15912`, :issue:`22334`)
+- Fixed bug causing plots of :class:`PeriodIndex` timeseries to fail if the frequency is a multiple of the frequency rule code (:issue:`14763`)
+- Fixed bug when plotting a :class:`DatetimeIndex` with ``datetime.timezone.utc`` timezone (:issue:`17173`)
+-
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :meth:`.Resampler.agg` with a timezone aware index where ``OverflowError`` would raise when passing a list of functions (:issue:`22660`)
+- Bug in :meth:`.DataFrameGroupBy.nunique` in which the names of column levels were lost (:issue:`23222`)
+- Bug in :func:`.GroupBy.agg` when applying an aggregation function to timezone aware data (:issue:`23683`)
+- Bug in :func:`.GroupBy.first` and :func:`.GroupBy.last` where timezone information would be dropped (:issue:`21603`)
+- Bug in :func:`.GroupBy.size` when grouping only NA values (:issue:`23050`)
+- Bug in :func:`Series.groupby` where ``observed`` kwarg was previously ignored (:issue:`24880`)
+- Bug in :func:`Series.groupby` where using ``groupby`` with a :class:`MultiIndex` Series with a list of labels equal to the length of the series caused incorrect grouping (:issue:`25704`)
+- Ensured that ordering of outputs in ``groupby`` aggregation functions is consistent across all versions of Python (:issue:`25692`)
+- Ensured that result group order is correct when grouping on an ordered ``Categorical`` and specifying ``observed=True`` (:issue:`25871`, :issue:`25167`)
+- Bug in :meth:`.Rolling.min` and :meth:`.Rolling.max` that caused a memory leak (:issue:`25893`)
+- Bug in :meth:`.Rolling.count` and ``.Expanding.count`` was previously ignoring the ``axis`` keyword (:issue:`13503`)
+- Bug in :meth:`.GroupBy.idxmax` and :meth:`.GroupBy.idxmin` with datetime column would return incorrect dtype (:issue:`25444`, :issue:`15306`)
+- Bug in :meth:`.GroupBy.cumsum`, :meth:`.GroupBy.cumprod`, :meth:`.GroupBy.cummin` and :meth:`.GroupBy.cummax` with categorical column having absent categories, would return incorrect result or segfault (:issue:`16771`)
+- Bug in :meth:`.GroupBy.nth` where NA values in the grouping would return incorrect results (:issue:`26011`)
+- Bug in :meth:`.SeriesGroupBy.transform` where transforming an empty group would raise a ``ValueError`` (:issue:`26208`)
+- Bug in :meth:`.DataFrame.groupby` where passing a :class:`.Grouper` would return incorrect groups when using the ``.groups`` accessor (:issue:`26326`)
+- Bug in :meth:`.GroupBy.agg` where incorrect results are returned for uint64 columns. (:issue:`26310`)
+- Bug in :meth:`.Rolling.median` and :meth:`.Rolling.quantile` where MemoryError is raised with empty window (:issue:`26005`)
+- Bug in :meth:`.Rolling.median` and :meth:`.Rolling.quantile` where incorrect results are returned with ``closed='left'`` and ``closed='neither'`` (:issue:`26005`)
+- Improved :class:`.Rolling`, :class:`.Window` and :class:`.ExponentialMovingWindow` functions to exclude nuisance columns from results instead of raising errors and raise a ``DataError`` only if all columns are nuisance (:issue:`12537`)
+- Bug in :meth:`.Rolling.max` and :meth:`.Rolling.min` where incorrect results are returned with an empty variable window (:issue:`26005`)
+- Raise a helpful exception when an unsupported weighted window function is used as an argument of :meth:`.Window.aggregate` (:issue:`26597`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in :func:`pandas.merge` adds a string of ``None``, if ``None`` is assigned in suffixes instead of remain the column name as-is (:issue:`24782`).
+- Bug in :func:`merge` when merging by index name would sometimes result in an incorrectly numbered index (missing index values are now assigned NA) (:issue:`24212`, :issue:`25009`)
+- :func:`to_records` now accepts dtypes to its ``column_dtypes`` parameter (:issue:`24895`)
+- Bug in :func:`concat` where order of ``OrderedDict`` (and ``dict`` in Python 3.6+) is not respected, when passed in as  ``objs`` argument (:issue:`21510`)
+- Bug in :func:`pivot_table` where columns with ``NaN`` values are dropped even if ``dropna`` argument is ``False``, when the ``aggfunc`` argument contains a ``list`` (:issue:`22159`)
+- Bug in :func:`concat` where the resulting ``freq`` of two :class:`DatetimeIndex` with the same ``freq`` would be dropped (:issue:`3232`).
+- Bug in :func:`merge` where merging with equivalent Categorical dtypes was raising an error (:issue:`22501`)
+- bug in :class:`DataFrame` instantiating with a dict of iterators or generators (e.g. ``pd.DataFrame({'A': reversed(range(3))})``) raised an error (:issue:`26349`).
+- Bug in :class:`DataFrame` instantiating with a ``range`` (e.g. ``pd.DataFrame(range(3))``) raised an error (:issue:`26342`).
+- Bug in :class:`DataFrame` constructor when passing non-empty tuples would cause a segmentation fault (:issue:`25691`)
+- Bug in :func:`Series.apply` failed when the series is a timezone aware :class:`DatetimeIndex` (:issue:`25959`)
+- Bug in :func:`pandas.cut` where large bins could incorrectly raise an error due to an integer overflow (:issue:`26045`)
+- Bug in :func:`DataFrame.sort_index` where an error is thrown when a multi-indexed ``DataFrame`` is sorted on all levels with the initial level sorted last (:issue:`26053`)
+- Bug in :meth:`Series.nlargest` treats ``True`` as smaller than ``False`` (:issue:`26154`)
+- Bug in :func:`DataFrame.pivot_table` with a :class:`IntervalIndex` as pivot index would raise ``TypeError`` (:issue:`25814`)
+- Bug in which :meth:`DataFrame.from_dict` ignored order of ``OrderedDict`` when ``orient='index'`` (:issue:`8425`).
+- Bug in :meth:`DataFrame.transpose` where transposing a DataFrame with a timezone-aware datetime column would incorrectly raise ``ValueError`` (:issue:`26825`)
+- Bug in :func:`pivot_table` when pivoting a timezone aware column as the ``values`` would remove timezone information (:issue:`14948`)
+- Bug in :func:`merge_asof` when specifying multiple ``by`` columns where one is ``datetime64[ns, tz]`` dtype (:issue:`26649`)
+
+Sparse
+^^^^^^
+
+- Significant speedup in :class:`SparseArray` initialization that benefits most operations, fixing performance regression introduced in v0.20.0 (:issue:`24985`)
+- Bug in :class:`SparseFrame` constructor where passing ``None`` as the data would cause ``default_fill_value`` to be ignored (:issue:`16807`)
+- Bug in :class:`SparseDataFrame` when adding a column in which the length of values does not match length of index, ``AssertionError`` is raised instead of raising ``ValueError`` (:issue:`25484`)
+- Introduce a better error message in :meth:`Series.sparse.from_coo` so it returns a ``TypeError`` for inputs that are not coo matrices (:issue:`26554`)
+- Bug in :func:`numpy.modf` on a :class:`SparseArray`. Now a tuple of :class:`SparseArray` is returned (:issue:`26946`).
+
+
+Build changes
+^^^^^^^^^^^^^
+
+- Fix install error with PyPy on macOS (:issue:`26536`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+
+- Bug in :func:`factorize` when passing an ``ExtensionArray`` with a custom ``na_sentinel`` (:issue:`25696`).
+- :meth:`Series.count` miscounts NA values in ExtensionArrays (:issue:`26835`)
+- Added ``Series.__array_ufunc__`` to better handle NumPy ufuncs applied to Series backed by extension arrays (:issue:`23293`).
+- Keyword argument ``deep`` has been removed from :meth:`ExtensionArray.copy` (:issue:`27083`)
+
+Other
+^^^^^
+
+- Removed unused C functions from vendored UltraJSON implementation (:issue:`26198`)
+- Allow :class:`Index` and :class:`RangeIndex` to be passed to numpy ``min`` and ``max`` functions (:issue:`26125`)
+- Use actual class name in repr of empty objects of a ``Series`` subclass (:issue:`27001`).
+- Bug in :class:`DataFrame` where passing an object array of timezone-aware ``datetime`` objects would incorrectly raise ``ValueError`` (:issue:`13287`)
+
+.. _whatsnew_0.250.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.24.2..v0.25.0
diff --git a/pandas/doc/source/whatsnew/v0.25.1.rst b/pandas/doc/source/whatsnew/v0.25.1.rst
new file mode 100644
index 0000000000000..67017d7c9fb29
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.25.1.rst
@@ -0,0 +1,119 @@
+.. _whatsnew_0251:
+
+What's new in 0.25.1 (August 21, 2019)
+--------------------------------------
+
+These are the changes in pandas 0.25.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+IO and LZMA
+~~~~~~~~~~~
+
+Some users may unknowingly have an incomplete Python installation lacking the ``lzma`` module from the standard library. In this case, ``import pandas`` failed due to an ``ImportError`` (:issue:`27575`).
+pandas will now warn, rather than raising an ``ImportError`` if the ``lzma`` module is not present. Any subsequent attempt to use ``lzma`` methods will raise a ``RuntimeError``.
+A possible fix for the lack of the ``lzma`` module is to ensure you have the necessary libraries and then re-install Python.
+For example, on MacOS installing Python with ``pyenv`` may lead to an incomplete Python installation due to unmet system dependencies at compilation time (like ``xz``). Compilation will succeed, but Python might fail at run time. The issue can be solved by installing the necessary dependencies and then re-installing Python.
+
+.. _whatsnew_0251.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+
+- Bug in :meth:`Categorical.fillna` that would replace all values, not just those that are ``NaN`` (:issue:`26215`)
+
+Datetimelike
+^^^^^^^^^^^^
+
+- Bug in :func:`to_datetime` where passing a timezone-naive :class:`DatetimeArray` or :class:`DatetimeIndex` and ``utc=True`` would incorrectly return a timezone-naive result (:issue:`27733`)
+- Bug in :meth:`Period.to_timestamp` where a :class:`Period` outside the :class:`Timestamp` implementation bounds (roughly 1677-09-21 to 2262-04-11) would return an incorrect :class:`Timestamp` instead of raising ``OutOfBoundsDatetime`` (:issue:`19643`)
+- Bug in iterating over :class:`DatetimeIndex` when the underlying data is read-only (:issue:`28055`)
+
+Timezones
+^^^^^^^^^
+
+- Bug in :class:`Index` where a numpy object array with a timezone aware :class:`Timestamp` and ``np.nan`` would not return a :class:`DatetimeIndex` (:issue:`27011`)
+
+Numeric
+^^^^^^^
+
+- Bug in :meth:`Series.interpolate` when using a timezone aware :class:`DatetimeIndex` (:issue:`27548`)
+- Bug when printing negative floating point complex numbers would raise an ``IndexError`` (:issue:`27484`)
+- Bug where :class:`DataFrame` arithmetic operators such as :meth:`DataFrame.mul` with a :class:`Series` with axis=1 would raise an ``AttributeError`` on :class:`DataFrame` larger than the minimum threshold to invoke numexpr (:issue:`27636`)
+- Bug in :class:`DataFrame` arithmetic where missing values in results were incorrectly masked with ``NaN`` instead of ``Inf`` (:issue:`27464`)
+
+Conversion
+^^^^^^^^^^
+
+- Improved the warnings for the deprecated methods :meth:`Series.real` and :meth:`Series.imag` (:issue:`27610`)
+
+Interval
+^^^^^^^^
+
+- Bug in :class:`IntervalIndex` where ``dir(obj)`` would raise ``ValueError`` (:issue:`27571`)
+
+Indexing
+^^^^^^^^
+
+- Bug in partial-string indexing returning a NumPy array rather than a ``Series`` when indexing with a scalar like ``.loc['2015']`` (:issue:`27516`)
+- Break reference cycle involving :class:`Index` and other index classes to allow garbage collection of index objects without running the GC. (:issue:`27585`, :issue:`27840`)
+- Fix regression in assigning values to a single column of a DataFrame with a ``MultiIndex`` columns (:issue:`27841`).
+- Fix regression in ``.ix`` fallback with an ``IntervalIndex`` (:issue:`27865`).
+
+Missing
+^^^^^^^
+
+- Bug in :func:`pandas.isnull` or :func:`pandas.isna` when the input is a type e.g. ``type(pandas.Series())`` (:issue:`27482`)
+
+IO
+^^
+
+- Avoid calling ``S3File.s3`` when reading parquet, as this was removed in s3fs version 0.3.0 (:issue:`27756`)
+- Better error message when a negative header is passed in :func:`pandas.read_csv` (:issue:`27779`)
+- Follow the ``min_rows`` display option (introduced in v0.25.0) correctly in the HTML repr in the notebook (:issue:`27991`).
+
+Plotting
+^^^^^^^^
+
+- Added a ``pandas_plotting_backends`` entrypoint group for registering plot backends. See :ref:`extending.plotting-backends` for more (:issue:`26747`).
+- Fixed the re-instatement of Matplotlib datetime converters after calling
+  :meth:`pandas.plotting.deregister_matplotlib_converters` (:issue:`27481`).
+- Fix compatibility issue with matplotlib when passing a pandas ``Index`` to a plot call (:issue:`27775`).
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Fixed regression in :meth:`pands.core.groupby.DataFrameGroupBy.quantile` raising when multiple quantiles are given (:issue:`27526`)
+- Bug in :meth:`.DataFrameGroupBy.transform` where applying a timezone conversion lambda function would drop timezone information (:issue:`27496`)
+- Bug in :meth:`.GroupBy.nth` where ``observed=False`` was being ignored for Categorical groupers (:issue:`26385`)
+- Bug in windowing over read-only arrays (:issue:`27766`)
+- Fixed segfault in ``.DataFrameGroupBy.quantile`` when an invalid quantile was passed (:issue:`27470`)
+
+Reshaping
+^^^^^^^^^
+
+- A ``KeyError`` is now raised if ``.unstack()`` is called on a :class:`Series` or :class:`DataFrame` with a flat :class:`Index` passing a name which is not the correct one (:issue:`18303`)
+- Bug :meth:`merge_asof` could not merge :class:`Timedelta` objects when passing ``tolerance`` kwarg (:issue:`27642`)
+- Bug in :meth:`DataFrame.crosstab` when ``margins`` set to ``True`` and ``normalize`` is not ``False``, an error is raised. (:issue:`27500`)
+- :meth:`DataFrame.join` now suppresses the ``FutureWarning`` when the sort parameter is specified (:issue:`21952`)
+- Bug in :meth:`DataFrame.join` raising with readonly arrays (:issue:`27943`)
+
+Sparse
+^^^^^^
+
+- Bug in reductions for :class:`Series` with Sparse dtypes (:issue:`27080`)
+
+Other
+^^^^^
+
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` when replacing timezone-aware timestamps using a dict-like replacer (:issue:`27720`)
+- Bug in :meth:`Series.rename` when using a custom type indexer. Now any value that isn't callable or dict-like is treated as a scalar. (:issue:`27814`)
+
+.. _whatsnew_0.251.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.25.0..v0.25.1
diff --git a/pandas/doc/source/whatsnew/v0.25.2.rst b/pandas/doc/source/whatsnew/v0.25.2.rst
new file mode 100644
index 0000000000000..5bd00f17bd3c5
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.25.2.rst
@@ -0,0 +1,49 @@
+.. _whatsnew_0252:
+
+What's new in 0.25.2 (October 15, 2019)
+---------------------------------------
+
+These are the changes in pandas 0.25.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+.. note::
+
+    pandas 0.25.2 adds compatibility for Python 3.8 (:issue:`28147`).
+
+.. _whatsnew_0252.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Indexing
+^^^^^^^^
+
+- Fix regression in :meth:`DataFrame.reindex` not following the ``limit`` argument (:issue:`28631`).
+- Fix regression in :meth:`RangeIndex.get_indexer` for decreasing :class:`RangeIndex` where target values may be improperly identified as missing/present (:issue:`28678`)
+
+IO
+^^
+
+- Fix regression in notebook display where ``<th>`` tags were missing for :attr:`DataFrame.index` values (:issue:`28204`).
+- Regression in :meth:`~DataFrame.to_csv` where writing a :class:`Series` or :class:`DataFrame` indexed by an :class:`IntervalIndex` would incorrectly raise a ``TypeError`` (:issue:`28210`)
+- Fix :meth:`~DataFrame.to_csv` with ``ExtensionArray`` with list-like values (:issue:`28840`).
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug incorrectly raising an ``IndexError`` when passing a list of quantiles to :meth:`.DataFrameGroupBy.quantile` (:issue:`28113`).
+- Bug in :meth:`.GroupBy.shift`, :meth:`.GroupBy.bfill` and :meth:`.GroupBy.ffill` where timezone information would be dropped (:issue:`19995`, :issue:`27992`)
+
+Other
+^^^^^
+
+- Compatibility with Python 3.8 in :meth:`DataFrame.query` (:issue:`27261`)
+- Fix to ensure that tab-completion in an IPython console does not raise
+  warnings for deprecated attributes (:issue:`27900`).
+
+.. _whatsnew_0.252.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.25.1..v0.25.2
diff --git a/pandas/doc/source/whatsnew/v0.25.3.rst b/pandas/doc/source/whatsnew/v0.25.3.rst
new file mode 100644
index 0000000000000..e028c08e1e85c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.25.3.rst
@@ -0,0 +1,22 @@
+.. _whatsnew_0253:
+
+What's new in 0.25.3 (October 31, 2019)
+---------------------------------------
+
+These are the changes in pandas 0.25.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+.. _whatsnew_0253.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :meth:`DataFrameGroupBy.quantile` where NA values in the grouping could cause segfaults or incorrect results (:issue:`28882`)
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.25.2..v0.25.3
diff --git a/pandas/doc/source/whatsnew/v0.4.x.rst b/pandas/doc/source/whatsnew/v0.4.x.rst
new file mode 100644
index 0000000000000..83f6a6907f33c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.4.x.rst
@@ -0,0 +1,68 @@
+.. _whatsnew_04x:
+
+Versions 0.4.1 through 0.4.3 (September 25 - October 9, 2011)
+-------------------------------------------------------------
+
+{{ header }}
+
+New features
+~~~~~~~~~~~~
+
+- Added Python 3 support using 2to3 (:issue:`200`)
+- :ref:`Added <dsintro.name_attribute>` ``name`` attribute to ``Series``, now
+  prints as part of ``Series.__repr__``
+- :meth:`Series.isnull`` and :meth:`Series.notnull` (:issue:`209`, :issue:`203`)
+- :ref:`Added <basics.align>` ``Series.align`` method for aligning two series
+  with choice of join method (ENH56_)
+- :ref:`Added <advanced.get_level_values>` method ``get_level_values`` to
+  ``MultiIndex`` (:issue:`188`)
+- Set values in mixed-type ``DataFrame`` objects via ``.ix`` indexing attribute (:issue:`135`)
+- Added new ``DataFrame`` :ref:`methods <basics.dtypes>`
+  ``get_dtype_counts`` and property ``dtypes`` (ENHdc_)
+- Added :ref:`ignore_index <merging.ignore_index>` option to
+  ``DataFrame.append`` to stack DataFrames (ENH1b_)
+- ``read_csv`` tries to :ref:`sniff <io.sniff>` delimiters using
+  ``csv.Sniffer`` (:issue:`146`)
+- ``read_csv`` can :ref:`read <io.csv_multiindex>` multiple columns into a
+  ``MultiIndex``; DataFrame's ``to_csv`` method writes out a corresponding
+  ``MultiIndex`` (:issue:`151`)
+- ``DataFrame.rename`` has a new ``copy`` parameter to :ref:`rename
+  <basics.rename>` a DataFrame in place (ENHed_)
+- :ref:`Enable <reshaping.unstack_by_name>` unstacking by name (:issue:`142`)
+- :ref:`Enable <advanced.sortlevel_byname>` ``sortlevel`` to work by level (:issue:`141`)
+
+Performance enhancements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Altered binary operations on differently-indexed SparseSeries objects
+  to use the integer-based (dense) alignment logic which is faster with a
+  larger number of blocks (:issue:`205`)
+- Wrote faster Cython data alignment / merging routines resulting in
+  substantial speed increases
+- Improved performance of ``isnull`` and ``notnull``, a regression from v0.3.0
+  (:issue:`187`)
+- Refactored code related to ``DataFrame.join`` so that intermediate aligned
+  copies of the data in each ``DataFrame`` argument do not need to be created.
+  Substantial performance increases result (:issue:`176`)
+- Substantially improved performance of generic ``Index.intersection`` and
+  ``Index.union``
+- Implemented ``BlockManager.take`` resulting in significantly faster ``take``
+  performance on mixed-type ``DataFrame`` objects (:issue:`104`)
+- Improved performance of ``Series.sort_index``
+- Significant groupby performance enhancement: removed unnecessary integrity
+  checks in DataFrame internals that were slowing down slicing operations to
+  retrieve groups
+- Optimized ``_ensure_index`` function resulting in performance savings in
+  type-checking Index objects
+- Wrote fast time series merging / joining methods in Cython. Will be
+  integrated later into DataFrame.join and related functions
+
+.. _ENH1b: https://github.com/pandas-dev/pandas/commit/1ba56251f0013ff7cd8834e9486cef2b10098371
+.. _ENHdc: https://github.com/pandas-dev/pandas/commit/dca3c5c5a6a3769ee01465baca04cfdfa66a4f76
+.. _ENHed: https://github.com/pandas-dev/pandas/commit/edd9f1945fc010a57fa0ae3b3444d1fffe592591
+.. _ENH56: https://github.com/pandas-dev/pandas/commit/56e0c9ffafac79ce262b55a6a13e1b10a88fbe93
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.4.1..v0.4.3
diff --git a/pandas/doc/source/whatsnew/v0.5.0.rst b/pandas/doc/source/whatsnew/v0.5.0.rst
new file mode 100644
index 0000000000000..129b86dc1ce5b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.5.0.rst
@@ -0,0 +1,52 @@
+
+.. _whatsnew_050:
+
+Version 0.5.0 (October 24, 2011)
+--------------------------------
+
+{{ header }}
+
+New features
+~~~~~~~~~~~~
+
+- :ref:`Added <basics.df_join>` ``DataFrame.align`` method with standard join options
+- :ref:`Added <io.parse_dates>` ``parse_dates`` option to ``read_csv`` and ``read_table`` methods to optionally try to parse dates in the index columns
+- :ref:`Added <io.parse_dates>` ``nrows``, ``chunksize``, and ``iterator`` arguments to ``read_csv`` and ``read_table``. The last two return a new ``TextParser`` class capable of lazily iterating through chunks of a flat file (:issue:`242`)
+- :ref:`Added <merging.multikey_join>` ability to join on multiple columns in ``DataFrame.join`` (:issue:`214`)
+- Added private ``_get_duplicates`` function to ``Index`` for identifying duplicate values more easily (ENH5c_)
+- :ref:`Added <indexing.df_cols>` column attribute access to DataFrame.
+- :ref:`Added <indexing.df_cols>` Python tab completion hook for DataFrame columns. (:issue:`233`, :issue:`230`)
+- :ref:`Implemented <basics.describe>` ``Series.describe`` for Series containing objects (:issue:`241`)
+- :ref:`Added <merging.df_inner_join>` inner join option to ``DataFrame.join`` when joining on key(s) (:issue:`248`)
+- :ref:`Implemented <indexing.df_cols>` selecting DataFrame columns by passing a list to ``__getitem__`` (:issue:`253`)
+- :ref:`Implemented <indexing.set_ops>` & and | to intersect / union Index objects, respectively (:issue:`261`)
+- :ref:`Added<reshaping.pivot>` ``pivot_table`` convenience function to pandas namespace (:issue:`234`)
+- :ref:`Implemented <basics.rename_axis>` ``Panel.rename_axis`` function (:issue:`243`)
+- DataFrame will show index level names in console output (:issue:`334`)
+- :ref:`Implemented <advanced.take>` ``Panel.take``
+- :ref:`Added<basics.console_output>` ``set_eng_float_format`` for alternate DataFrame floating point string formatting (ENH61_)
+- :ref:`Added <indexing.set_index>` convenience ``set_index`` function for creating a DataFrame index from its existing columns
+- :ref:`Implemented <groupby.multiindex>` ``groupby`` hierarchical index level name  (:issue:`223`)
+- :ref:`Added <io.store_in_csv>` support for different delimiters in ``DataFrame.to_csv`` (:issue:`244`)
+
+Performance enhancements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- VBENCH Major performance improvements in file parsing functions ``read_csv`` and ``read_table``
+- VBENCH Added Cython function for converting tuples to ndarray very fast. Speeds up many MultiIndex-related operations
+- VBENCH Refactored merging / joining code into a tidy class and disabled unnecessary computations in the float/object case, thus getting about 10% better performance (:issue:`211`)
+- VBENCH Improved speed of ``DataFrame.xs`` on mixed-type DataFrame objects by about 5x, regression from 0.3.0 (:issue:`215`)
+- VBENCH With new ``DataFrame.align`` method, speeding up binary operations between differently-indexed DataFrame objects by 10-25%.
+- VBENCH Significantly sped up conversion of nested dict into DataFrame (:issue:`212`)
+- VBENCH Significantly speed up DataFrame ``__repr__`` and ``count`` on large mixed-type DataFrame objects
+
+.. _ENH61: https://github.com/pandas-dev/pandas/commit/6141961
+.. _ENH5c: https://github.com/pandas-dev/pandas/commit/5ca6ff5d822ee4ddef1ec0d87b6d83d8b4bbd3eb
+
+
+.. _whatsnew_0.5.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.4.0..v0.5.0
diff --git a/pandas/doc/source/whatsnew/v0.6.0.rst b/pandas/doc/source/whatsnew/v0.6.0.rst
new file mode 100644
index 0000000000000..5ddcd5d90e65c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.6.0.rst
@@ -0,0 +1,66 @@
+.. _whatsnew_060:
+
+Version 0.6.0 (November 25, 2011)
+---------------------------------
+
+{{ header }}
+
+New features
+~~~~~~~~~~~~
+- :ref:`Added <reshaping.melt>` ``melt`` function to ``pandas.core.reshape``
+- :ref:`Added <groupby.multiindex>` ``level`` parameter to group by level in Series and DataFrame descriptive statistics (:issue:`313`)
+- :ref:`Added <basics.head_tail>` ``head`` and ``tail`` methods to Series, analogous to DataFrame (:issue:`296`)
+- :ref:`Added <indexing.boolean>` ``Series.isin`` function which checks if each value is contained in a passed sequence (:issue:`289`)
+- :ref:`Added <io.formatting>` ``float_format`` option to ``Series.to_string``
+- :ref:`Added <io.parse_dates>` ``skip_footer`` (:issue:`291`) and ``converters`` (:issue:`343`) options to ``read_csv`` and ``read_table``
+- :ref:`Added <indexing.duplicate>` ``drop_duplicates`` and ``duplicated`` functions for removing duplicate DataFrame rows and checking for duplicate rows, respectively (:issue:`319`)
+- :ref:`Implemented <dsintro.boolean>` operators '&', '|', '^', '-' on DataFrame (:issue:`347`)
+- :ref:`Added <basics.stats>` ``Series.mad``, mean absolute deviation
+- :ref:`Added <timeseries.offsets>` ``QuarterEnd`` DateOffset (:issue:`321`)
+- :ref:`Added <dsintro.numpy_interop>` ``dot`` to DataFrame (:issue:`65`)
+- Added ``orient`` option to ``Panel.from_dict`` (:issue:`359`, :issue:`301`)
+- :ref:`Added <basics.dataframe.from_dict>` ``orient`` option to ``DataFrame.from_dict``
+- :ref:`Added <basics.dataframe.from_records>` passing list of tuples or list of lists to ``DataFrame.from_records`` (:issue:`357`)
+- :ref:`Added <groupby.multiindex>` multiple levels to groupby (:issue:`103`)
+- :ref:`Allow <basics.sorting>` multiple columns in ``by`` argument of ``DataFrame.sort_index`` (:issue:`92`, :issue:`362`)
+- :ref:`Added <indexing.basics.get_value>` fast ``get_value`` and ``put_value`` methods to DataFrame (:issue:`360`)
+- Added ``cov`` instance methods to Series and DataFrame (:issue:`194`, :issue:`362`)
+- :ref:`Added <visualization.barplot>` ``kind='bar'`` option to ``DataFrame.plot`` (:issue:`348`)
+- :ref:`Added <basics.idxmin>` ``idxmin`` and ``idxmax`` to Series and DataFrame (:issue:`286`)
+- :ref:`Added <io.clipboard>` ``read_clipboard`` function to parse DataFrame from clipboard (:issue:`300`)
+- :ref:`Added <basics.stats>` ``nunique`` function to Series for counting unique elements (:issue:`297`)
+- :ref:`Made <basics.dataframe>` DataFrame constructor use Series name if no columns passed (:issue:`373`)
+- :ref:`Support <io.parse_dates>` regular expressions in read_table/read_csv (:issue:`364`)
+- :ref:`Added <io.html>` ``DataFrame.to_html`` for writing DataFrame to HTML (:issue:`387`)
+- :ref:`Added <basics.dataframe>` support for MaskedArray data in DataFrame, masked values converted to NaN (:issue:`396`)
+- :ref:`Added <visualization.box>` ``DataFrame.boxplot`` function (:issue:`368`)
+- :ref:`Can <basics.apply>` pass extra args, kwds to DataFrame.apply (:issue:`376`)
+- :ref:`Implement <merging.multikey_join>` ``DataFrame.join`` with vector ``on`` argument (:issue:`312`)
+- :ref:`Added <visualization.basic>` ``legend`` boolean flag to ``DataFrame.plot`` (:issue:`324`)
+- :ref:`Can <reshaping.stacking>` pass multiple levels to ``stack`` and ``unstack`` (:issue:`370`)
+- :ref:`Can <reshaping.pivot>` pass multiple values columns to ``pivot_table`` (:issue:`381`)
+- :ref:`Use <groupby.multiindex>` Series name in GroupBy for result index (:issue:`363`)
+- :ref:`Added <basics.apply>` ``raw`` option to ``DataFrame.apply`` for performance if only need ndarray (:issue:`309`)
+- Added proper, tested weighted least squares to standard and panel OLS (:issue:`303`)
+
+Performance enhancements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- VBENCH Cythonized ``cache_readonly``, resulting in substantial micro-performance enhancements throughout the code base (:issue:`361`)
+- VBENCH Special Cython matrix iterator for applying arbitrary reduction operations with 3-5x better performance than ``np.apply_along_axis`` (:issue:`309`)
+- VBENCH Improved performance of ``MultiIndex.from_tuples``
+- VBENCH Special Cython matrix iterator for applying arbitrary reduction operations
+- VBENCH + DOCUMENT Add ``raw`` option to ``DataFrame.apply`` for getting better performance when
+- VBENCH Faster cythonized count by level in Series and DataFrame (:issue:`341`)
+- VBENCH? Significant GroupBy performance enhancement with multiple keys with many "empty" combinations
+- VBENCH New Cython vectorized function ``map_infer`` speeds up ``Series.apply`` and ``Series.map`` significantly when passed elementwise Python function, motivated by (:issue:`355`)
+- VBENCH Significantly improved performance of ``Series.order``, which also makes np.unique called on a Series faster (:issue:`327`)
+- VBENCH Vastly improved performance of GroupBy on axes with a MultiIndex (:issue:`299`)
+
+
+
+.. _whatsnew_0.6.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.5.0..v0.6.0
diff --git a/pandas/doc/source/whatsnew/v0.6.1.rst b/pandas/doc/source/whatsnew/v0.6.1.rst
new file mode 100644
index 0000000000000..58a7d1ee13278
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.6.1.rst
@@ -0,0 +1,58 @@
+
+.. _whatsnew_061:
+
+Version 0.6.1 (December 13, 2011)
+---------------------------------
+
+New features
+~~~~~~~~~~~~
+- Can append single rows (as Series) to a DataFrame
+- Add Spearman and Kendall rank correlation
+  options to Series.corr and DataFrame.corr (:issue:`428`)
+- :ref:`Added <indexing.basics.get_value>` ``get_value`` and ``set_value`` methods to
+  Series, DataFrame, and Panel for very low-overhead access (>2x faster in many
+  cases) to scalar elements (:issue:`437`, :issue:`438`). ``set_value`` is capable of
+  producing an enlarged object.
+- Add PyQt table widget to sandbox (:issue:`435`)
+- DataFrame.align can :ref:`accept Series arguments <basics.align.frame.series>`
+  and an :ref:`axis option <basics.df_join>` (:issue:`461`)
+- Implement new :ref:`SparseArray <sparse.array>` and ``SparseList``
+  data structures. SparseSeries now derives from SparseArray (:issue:`463`)
+- :ref:`Better console printing options <basics.console_output>` (:issue:`453`)
+- Implement fast data ranking for Series and
+  DataFrame, fast versions of scipy.stats.rankdata (:issue:`428`)
+- Implement ``DataFrame.from_items`` alternate
+  constructor (:issue:`444`)
+- DataFrame.convert_objects method for :ref:`inferring better dtypes <basics.cast>`
+  for object columns (:issue:`302`)
+- Add :ref:`rolling_corr_pairwise <window.corr_pairwise>` function for
+  computing Panel of correlation matrices (:issue:`189`)
+- Add :ref:`margins <reshaping.pivot.margins>` option to :ref:`pivot_table
+  <reshaping.pivot>` for computing subgroup aggregates (:issue:`114`)
+- Add ``Series.from_csv`` function (:issue:`482`)
+- :ref:`Can pass <window.cov_corr>` DataFrame/DataFrame and
+  DataFrame/Series to rolling_corr/rolling_cov (GH #462)
+- MultiIndex.get_level_values can :ref:`accept the level name <advanced.get_level_values>`
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Improve memory usage of ``DataFrame.describe`` (do not copy data
+  unnecessarily) (PR #425)
+
+- Optimize scalar value lookups in the general case by 25% or more in Series
+  and DataFrame
+
+- Fix performance regression in cross-sectional count in DataFrame, affecting
+  DataFrame.dropna speed
+- Column deletion in DataFrame copies no data (computes views on blocks) (GH
+  #158)
+
+
+
+.. _whatsnew_0.6.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.6.0..v0.6.1
diff --git a/pandas/doc/source/whatsnew/v0.7.0.rst b/pandas/doc/source/whatsnew/v0.7.0.rst
new file mode 100644
index 0000000000000..2336ccaeac820
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.7.0.rst
@@ -0,0 +1,384 @@
+.. _whatsnew_0700:
+
+Version 0.7.0 (February 9, 2012)
+--------------------------------
+
+{{ header }}
+
+
+New features
+~~~~~~~~~~~~
+
+- New unified :ref:`merge function <merging.join>` for efficiently performing
+  full gamut of database / relational-algebra operations. Refactored existing
+  join methods to use the new infrastructure, resulting in substantial
+  performance gains (:issue:`220`, :issue:`249`, :issue:`267`)
+
+- New :ref:`unified concatenation function <merging.concat>` for concatenating
+  Series, DataFrame or Panel objects along an axis. Can form union or
+  intersection of the other axes. Improves performance of ``Series.append`` and
+  ``DataFrame.append`` (:issue:`468`, :issue:`479`, :issue:`273`)
+
+- Can pass multiple DataFrames to
+  ``DataFrame.append`` to concatenate (stack) and multiple Series to
+  ``Series.append`` too
+
+- :ref:`Can<basics.dataframe.from_list_of_dicts>` pass list of dicts (e.g., a
+  list of JSON objects) to DataFrame constructor (:issue:`526`)
+
+- You can now :ref:`set multiple columns <indexing.columns.multiple>` in a
+  DataFrame via ``__getitem__``, useful for transformation (:issue:`342`)
+
+- Handle differently-indexed output values in ``DataFrame.apply`` (:issue:`498`)
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame(np.random.randn(10, 4))
+   In [2]: df.apply(lambda x: x.describe())
+   Out[2]:
+                  0          1          2          3
+   count  10.000000  10.000000  10.000000  10.000000
+   mean    0.190912  -0.395125  -0.731920  -0.403130
+   std     0.730951   0.813266   1.112016   0.961912
+   min    -0.861849  -2.104569  -1.776904  -1.469388
+   25%    -0.411391  -0.698728  -1.501401  -1.076610
+   50%     0.380863  -0.228039  -1.191943  -1.004091
+   75%     0.658444   0.057974  -0.034326   0.461706
+   max     1.212112   0.577046   1.643563   1.071804
+
+   [8 rows x 4 columns]
+
+- :ref:`Add<advanced.reorderlevels>` ``reorder_levels`` method to Series and
+  DataFrame (:issue:`534`)
+
+- :ref:`Add<indexing.dictionarylike>` dict-like ``get`` function to DataFrame
+  and Panel (:issue:`521`)
+
+- :ref:`Add<basics.iterrows>` ``DataFrame.iterrows`` method for efficiently
+  iterating through the rows of a DataFrame
+
+- Add ``DataFrame.to_panel`` with code adapted from
+  ``LongPanel.to_long``
+
+- :ref:`Add <basics.reindexing>` ``reindex_axis`` method added to DataFrame
+
+- :ref:`Add <basics.stats>` ``level`` option to binary arithmetic functions on
+  ``DataFrame`` and ``Series``
+
+- :ref:`Add <advanced.advanced_reindex>` ``level`` option to the ``reindex``
+  and ``align`` methods on Series and DataFrame for broadcasting values across
+  a level (:issue:`542`, :issue:`552`, others)
+
+- Add attribute-based item access to
+  ``Panel`` and add IPython completion (:issue:`563`)
+
+- :ref:`Add <visualization.basic>` ``logy`` option to ``Series.plot`` for
+  log-scaling on the Y axis
+
+- :ref:`Add <io.formatting>` ``index`` and ``header`` options to
+  ``DataFrame.to_string``
+
+- :ref:`Can <merging.multiple_join>` pass multiple DataFrames to
+  ``DataFrame.join`` to join on index (:issue:`115`)
+
+- :ref:`Can <merging.multiple_join>` pass multiple Panels to ``Panel.join``
+  (:issue:`115`)
+
+- :ref:`Added <io.formatting>` ``justify`` argument to ``DataFrame.to_string``
+  to allow different alignment of column headers
+
+- :ref:`Add <groupby.attributes>` ``sort`` option to GroupBy to allow disabling
+  sorting of the group keys for potential speedups (:issue:`595`)
+
+- :ref:`Can <basics.dataframe.from_series>` pass MaskedArray to Series
+  constructor (:issue:`563`)
+
+- Add Panel item access via attributes
+  and IPython completion (:issue:`554`)
+
+- Implement ``DataFrame.lookup``, fancy-indexing analogue for retrieving values
+  given a sequence of row and column labels (:issue:`338`)
+
+- Can pass a :ref:`list of functions <groupby.aggregate.multifunc>` to
+  aggregate with groupby on a DataFrame, yielding an aggregated result with
+  hierarchical columns (:issue:`166`)
+
+- Can call ``cummin`` and ``cummax`` on Series and DataFrame to get cumulative
+  minimum and maximum, respectively (:issue:`647`)
+
+- ``value_range`` added as utility function to get min and max of a dataframe
+  (:issue:`288`)
+
+- Added ``encoding`` argument to ``read_csv``, ``read_table``, ``to_csv`` and
+  ``from_csv`` for non-ascii text (:issue:`717`)
+
+- :ref:`Added <basics.stats>` ``abs`` method to pandas objects
+
+- :ref:`Added <reshaping.pivot>` ``crosstab`` function for easily computing frequency tables
+
+- :ref:`Added <indexing.set_ops>` ``isin`` method to index objects
+
+- :ref:`Added <advanced.xs>` ``level`` argument to ``xs`` method of DataFrame.
+
+
+API changes to integer indexing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+One of the potentially riskiest API changes in 0.7.0, but also one of the most
+important, was a complete review of how **integer indexes** are handled with
+regard to label-based indexing. Here is an example:
+
+.. code-block:: ipython
+
+    In [3]: s = pd.Series(np.random.randn(10), index=range(0, 20, 2))
+    In [4]: s
+    Out[4]:
+    0    -1.294524
+    2     0.413738
+    4     0.276662
+    6    -0.472035
+    8    -0.013960
+    10   -0.362543
+    12   -0.006154
+    14   -0.923061
+    16    0.895717
+    18    0.805244
+    Length: 10, dtype: float64
+
+    In [5]: s[0]
+    Out[5]: -1.2945235902555294
+
+    In [6]: s[2]
+    Out[6]: 0.41373810535784006
+
+    In [7]: s[4]
+    Out[7]: 0.2766617129497566
+
+This is all exactly identical to the behavior before. However, if you ask for a
+key **not** contained in the Series, in versions 0.6.1 and prior, Series would
+*fall back* on a location-based lookup. This now raises a ``KeyError``:
+
+.. code-block:: ipython
+
+   In [2]: s[1]
+   KeyError: 1
+
+This change also has the same impact on DataFrame:
+
+.. code-block:: ipython
+
+   In [3]: df = pd.DataFrame(np.random.randn(8, 4), index=range(0, 16, 2))
+
+   In [4]: df
+       0        1       2       3
+   0   0.88427  0.3363 -0.1787  0.03162
+   2   0.14451 -0.1415  0.2504  0.58374
+   4  -1.44779 -0.9186 -1.4996  0.27163
+   6  -0.26598 -2.4184 -0.2658  0.11503
+   8  -0.58776  0.3144 -0.8566  0.61941
+   10  0.10940 -0.7175 -1.0108  0.47990
+   12 -1.16919 -0.3087 -0.6049 -0.43544
+   14 -0.07337  0.3410  0.0424 -0.16037
+
+   In [5]: df.ix[3]
+   KeyError: 3
+
+In order to support purely integer-based indexing, the following methods have
+been added:
+
+.. csv-table::
+    :header: "Method","Description"
+    :widths: 40,60
+
+        ``Series.iget_value(i)``, Retrieve value stored at location ``i``
+        ``Series.iget(i)``, Alias for ``iget_value``
+        ``DataFrame.irow(i)``, Retrieve the ``i``-th row
+        ``DataFrame.icol(j)``, Retrieve the ``j``-th column
+        "``DataFrame.iget_value(i, j)``", Retrieve the value at row ``i`` and column ``j``
+
+API tweaks regarding label-based slicing
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Label-based slicing using ``ix`` now requires that the index be sorted
+(monotonic) **unless** both the start and endpoint are contained in the index:
+
+.. code-block:: python
+
+   In [1]: s = pd.Series(np.random.randn(6), index=list('gmkaec'))
+
+   In [2]: s
+   Out[2]:
+   g   -1.182230
+   m   -0.276183
+   k   -0.243550
+   a    1.628992
+   e    0.073308
+   c   -0.539890
+   dtype: float64
+
+Then this is OK:
+
+.. code-block:: python
+
+   In [3]: s.ix['k':'e']
+   Out[3]:
+   k   -0.243550
+   a    1.628992
+   e    0.073308
+   dtype: float64
+
+But this is not:
+
+.. code-block:: ipython
+
+   In [12]: s.ix['b':'h']
+   KeyError 'b'
+
+If the index had been sorted, the "range selection" would have been possible:
+
+.. code-block:: python
+
+   In [4]: s2 = s.sort_index()
+
+   In [5]: s2
+   Out[5]:
+   a    1.628992
+   c   -0.539890
+   e    0.073308
+   g   -1.182230
+   k   -0.243550
+   m   -0.276183
+   dtype: float64
+
+   In [6]: s2.ix['b':'h']
+   Out[6]:
+   c   -0.539890
+   e    0.073308
+   g   -1.182230
+   dtype: float64
+
+Changes to Series ``[]`` operator
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+As as notational convenience, you can pass a sequence of labels or a label
+slice to a Series when getting and setting values via ``[]`` (i.e. the
+``__getitem__`` and ``__setitem__`` methods). The behavior will be the same as
+passing similar input to ``ix`` **except in the case of integer indexing**:
+
+.. code-block:: ipython
+
+  In [8]: s = pd.Series(np.random.randn(6), index=list('acegkm'))
+
+  In [9]: s
+  Out[9]:
+  a   -1.206412
+  c    2.565646
+  e    1.431256
+  g    1.340309
+  k   -1.170299
+  m   -0.226169
+  Length: 6, dtype: float64
+
+  In [10]: s[['m', 'a', 'c', 'e']]
+  Out[10]:
+  m   -0.226169
+  a   -1.206412
+  c    2.565646
+  e    1.431256
+  Length: 4, dtype: float64
+
+  In [11]: s['b':'l']
+  Out[11]:
+  c    2.565646
+  e    1.431256
+  g    1.340309
+  k   -1.170299
+  Length: 4, dtype: float64
+
+  In [12]: s['c':'k']
+  Out[12]:
+  c    2.565646
+  e    1.431256
+  g    1.340309
+  k   -1.170299
+  Length: 4, dtype: float64
+
+In the case of integer indexes, the behavior will be exactly as before
+(shadowing ``ndarray``):
+
+.. code-block:: ipython
+
+  In [13]: s = pd.Series(np.random.randn(6), index=range(0, 12, 2))
+
+  In [14]: s[[4, 0, 2]]
+  Out[14]:
+  4    0.132003
+  0    0.410835
+  2    0.813850
+  Length: 3, dtype: float64
+
+  In [15]: s[1:5]
+  Out[15]:
+  2    0.813850
+  4    0.132003
+  6   -0.827317
+  8   -0.076467
+  Length: 4, dtype: float64
+
+If you wish to do indexing with sequences and slicing on an integer index with
+label semantics, use ``ix``.
+
+Other API changes
+~~~~~~~~~~~~~~~~~
+
+- The deprecated ``LongPanel`` class has been completely removed
+
+- If ``Series.sort`` is called on a column of a DataFrame, an exception will
+  now be raised. Before it was possible to accidentally mutate a DataFrame's
+  column by doing ``df[col].sort()`` instead of the side-effect free method
+  ``df[col].order()`` (:issue:`316`)
+
+- Miscellaneous renames and deprecations which will (harmlessly) raise
+  ``FutureWarning``
+
+- ``drop`` added as an optional parameter to ``DataFrame.reset_index`` (:issue:`699`)
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- :ref:`Cythonized GroupBy aggregations <groupby.aggregate.builtin>` no longer
+  presort the data, thus achieving a significant speedup (:issue:`93`).  GroupBy
+  aggregations with Python functions significantly sped up by clever
+  manipulation of the ndarray data type in Cython (:issue:`496`).
+- Better error message in DataFrame constructor when passed column labels
+  don't match data (:issue:`497`)
+- Substantially improve performance of multi-GroupBy aggregation when a
+  Python function is passed, reuse ndarray object in Cython (:issue:`496`)
+- Can store objects indexed by tuples and floats in HDFStore (:issue:`492`)
+- Don't print length by default in Series.to_string, add ``length`` option (:issue:`489`)
+- Improve Cython code for multi-groupby to aggregate without having to sort
+  the data (:issue:`93`)
+- Improve MultiIndex reindexing speed by storing tuples in the MultiIndex,
+  test for backwards unpickling compatibility
+- Improve column reindexing performance by using specialized Cython take
+  function
+- Further performance tweaking of Series.__getitem__ for standard use cases
+- Avoid Index dict creation in some cases (i.e. when getting slices, etc.),
+  regression from prior versions
+- Friendlier error message in setup.py if NumPy not installed
+- Use common set of NA-handling operations (sum, mean, etc.) in Panel class
+  also (:issue:`536`)
+- Default name assignment when calling ``reset_index`` on DataFrame with a
+  regular (non-hierarchical) index (:issue:`476`)
+- Use Cythonized groupers when possible in Series/DataFrame stat ops with
+  ``level`` parameter passed (:issue:`545`)
+- Ported skiplist data structure to C to speed up ``rolling_median`` by about
+  5-10x in most typical use cases (:issue:`374`)
+
+
+.. _whatsnew_0.7.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.6.1..v0.7.0
diff --git a/pandas/doc/source/whatsnew/v0.7.1.rst b/pandas/doc/source/whatsnew/v0.7.1.rst
new file mode 100644
index 0000000000000..7082ef8ed2882
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.7.1.rst
@@ -0,0 +1,41 @@
+.. _whatsnew_0701:
+
+Version 0.7.1 (February 29, 2012)
+---------------------------------
+
+{{ header }}
+
+
+This release includes a few new features and addresses over a dozen bugs in
+0.7.0.
+
+New features
+~~~~~~~~~~~~
+
+  - Add ``to_clipboard`` function to pandas namespace for writing objects to
+    the system clipboard (:issue:`774`)
+  - Add ``itertuples`` method to DataFrame for iterating through the rows of a
+    dataframe as tuples (:issue:`818`)
+  - Add ability to pass fill_value and method to DataFrame and Series align
+    method (:issue:`806`, :issue:`807`)
+  - Add fill_value option to reindex, align methods (:issue:`784`)
+  - Enable concat to produce DataFrame from Series (:issue:`787`)
+  - Add ``between`` method to Series (:issue:`802`)
+  - Add HTML representation hook to DataFrame for the IPython HTML notebook
+    (:issue:`773`)
+  - Support for reading Excel 2007 XML documents using openpyxl
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+  - Improve performance and memory usage of fillna on DataFrame
+  - Can concatenate a list of Series along axis=1 to obtain a DataFrame (:issue:`787`)
+
+
+
+.. _whatsnew_0.7.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.7.0..v0.7.1
diff --git a/pandas/doc/source/whatsnew/v0.7.2.rst b/pandas/doc/source/whatsnew/v0.7.2.rst
new file mode 100644
index 0000000000000..e10a7b499549b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.7.2.rst
@@ -0,0 +1,38 @@
+.. _whatsnew_0702:
+
+Version 0.7.2 (March 16, 2012)
+------------------------------
+
+{{ header }}
+
+
+This release targets bugs in 0.7.1, and adds a few minor features.
+
+New features
+~~~~~~~~~~~~
+
+  - Add additional tie-breaking methods in DataFrame.rank (:issue:`874`)
+  - Add ascending parameter to rank in Series, DataFrame (:issue:`875`)
+  - Add coerce_float option to DataFrame.from_records (:issue:`893`)
+  - Add sort_columns parameter to allow unsorted plots (:issue:`918`)
+  - Enable column access via attributes on GroupBy (:issue:`882`)
+  - Can pass dict of values to DataFrame.fillna (:issue:`661`)
+  - Can select multiple hierarchical groups by passing list of values in .ix
+    (:issue:`134`)
+  - Add ``axis`` option to DataFrame.fillna (:issue:`174`)
+  - Add level keyword to ``drop`` for dropping values from a level (:issue:`159`)
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+  - Use khash for Series.value_counts, add raw function to algorithms.py (:issue:`861`)
+  - Intercept __builtin__.sum in groupby (:issue:`885`)
+
+
+
+.. _whatsnew_0.7.2.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.7.1..v0.7.2
diff --git a/pandas/doc/source/whatsnew/v0.7.3.rst b/pandas/doc/source/whatsnew/v0.7.3.rst
new file mode 100644
index 0000000000000..5da6bef0c4f03
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.7.3.rst
@@ -0,0 +1,151 @@
+.. _whatsnew_0703:
+
+Version 0.7.3 (April 12, 2012)
+------------------------------
+
+{{ header }}
+
+
+This is a minor release from 0.7.2 and fixes many minor bugs and adds a number
+of nice new features. There are also a couple of API changes to note; these
+should not affect very many users, and we are inclined to call them "bug fixes"
+even though they do constitute a change in behavior. See the :ref:`full release
+notes <release>` or issue
+tracker on GitHub for a complete list.
+
+New features
+~~~~~~~~~~~~
+
+- New :ref:`fixed width file reader <io.fwf>`, ``read_fwf``
+- New :ref:`scatter_matrix <visualization.scatter_matrix>` function for making
+  a scatter plot matrix
+
+.. code-block:: python
+
+   from pandas.tools.plotting import scatter_matrix
+
+   scatter_matrix(df, alpha=0.2)  # noqa F821
+
+
+- Add ``stacked`` argument to Series and DataFrame's ``plot`` method for
+  :ref:`stacked bar plots <visualization.barplot>`.
+
+.. code-block:: python
+
+   df.plot(kind="bar", stacked=True)  # noqa F821
+
+
+.. code-block:: python
+
+   df.plot(kind="barh", stacked=True)  # noqa F821
+
+
+- Add log x and y :ref:`scaling options <visualization.basic>` to
+  ``DataFrame.plot`` and ``Series.plot``
+- Add ``kurt`` methods to Series and DataFrame for computing kurtosis
+
+
+NA boolean comparison API change
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Reverted some changes to how NA values (represented typically as ``NaN`` or
+``None``) are handled in non-numeric Series:
+
+.. code-block:: ipython
+
+   In [1]: series = pd.Series(["Steve", np.nan, "Joe"])
+
+   In [2]: series == "Steve"
+   Out[2]:
+   0     True
+   1    False
+   2    False
+   Length: 3, dtype: bool
+
+   In [3]: series != "Steve"
+   Out[3]:
+   0    False
+   1     True
+   2     True
+   Length: 3, dtype: bool
+
+In comparisons, NA / NaN will always come through as ``False`` except with
+``!=`` which is ``True``. *Be very careful* with boolean arithmetic, especially
+negation, in the presence of NA data. You may wish to add an explicit NA
+filter into boolean array operations if you are worried about this:
+
+.. code-block:: ipython
+
+   In [4]: mask = series == "Steve"
+
+   In [5]: series[mask & series.notnull()]
+   Out[5]:
+   0    Steve
+   Length: 1, dtype: object
+
+While propagating NA in comparisons may seem like the right behavior to some
+users (and you could argue on purely technical grounds that this is the right
+thing to do), the evaluation was made that propagating NA everywhere, including
+in numerical arrays, would cause a large amount of problems for users. Thus, a
+"practicality beats purity" approach was taken. This issue may be revisited at
+some point in the future.
+
+Other API changes
+~~~~~~~~~~~~~~~~~
+
+When calling ``apply`` on a grouped Series, the return value will also be a
+Series, to be more consistent with the ``groupby`` behavior with DataFrame:
+
+.. code-block:: ipython
+
+      In [6]: df = pd.DataFrame(
+         ...:     {
+         ...:         "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+         ...:         "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+         ...:         "C": np.random.randn(8),
+         ...:         "D": np.random.randn(8),
+         ...:     }
+         ...: )
+         ...:
+
+      In [7]: df
+      Out[7]:
+         A      B         C         D
+      0  foo    one  0.469112 -0.861849
+      1  bar    one -0.282863 -2.104569
+      2  foo    two -1.509059 -0.494929
+      3  bar  three -1.135632  1.071804
+      4  foo    two  1.212112  0.721555
+      5  bar    two -0.173215 -0.706771
+      6  foo    one  0.119209 -1.039575
+      7  foo  three -1.044236  0.271860
+
+      [8 rows x 4 columns]
+
+      In [8]: grouped = df.groupby("A")["C"]
+
+      In [9]: grouped.describe()
+      Out[9]:
+         count      mean       std       min       25%       50%       75%       max
+      A
+      bar    3.0 -0.530570  0.526860 -1.135632 -0.709248 -0.282863 -0.228039 -0.173215
+      foo    5.0 -0.150572  1.113308 -1.509059 -1.044236  0.119209  0.469112  1.212112
+
+      [2 rows x 8 columns]
+
+      In [10]: grouped.apply(lambda x: x.sort_values()[-2:])  # top 2 values
+      Out[10]:
+      A
+      bar  1   -0.282863
+           5   -0.173215
+      foo  0    0.469112
+           4    1.212112
+      Name: C, Length: 4, dtype: float64
+
+
+.. _whatsnew_0.7.3.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.7.2..v0.7.3
diff --git a/pandas/doc/source/whatsnew/v0.8.0.rst b/pandas/doc/source/whatsnew/v0.8.0.rst
new file mode 100644
index 0000000000000..ce02525a69ace
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.8.0.rst
@@ -0,0 +1,284 @@
+.. _whatsnew_080:
+
+Version 0.8.0 (June 29, 2012)
+-----------------------------
+
+{{ header }}
+
+
+This is a major release from 0.7.3 and includes extensive work on the time
+series handling and processing infrastructure as well as a great deal of new
+functionality throughout the library. It includes over 700 commits from more
+than 20 distinct authors. Most pandas 0.7.3 and earlier users should not
+experience any issues upgrading, but due to the migration to the NumPy
+datetime64 dtype, there may be a number of bugs and incompatibilities
+lurking. Lingering incompatibilities will be fixed ASAP in a 0.8.1 release if
+necessary. See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+Support for non-unique indexes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+All objects can now work with non-unique indexes. Data alignment / join
+operations work according to SQL join semantics (including, if application,
+index duplication in many-to-many joins)
+
+NumPy datetime64 dtype and 1.6 dependency
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Time series data are now represented using NumPy's datetime64 dtype; thus,
+pandas 0.8.0 now requires at least NumPy 1.6. It has been tested and verified
+to work with the development version (1.7+) of NumPy as well which includes
+some significant user-facing API changes. NumPy 1.6 also has a number of bugs
+having to do with nanosecond resolution data, so I recommend that you steer
+clear of NumPy 1.6's datetime64 API functions (though limited as they are) and
+only interact with this data using the interface that pandas provides.
+
+See the end of the 0.8.0 section for a "porting" guide listing potential issues
+for users migrating legacy code bases from pandas 0.7 or earlier to 0.8.0.
+
+Bug fixes to the 0.7.x series for legacy NumPy < 1.6 users will be provided as
+they arise. There will be no more further development in 0.7.x beyond bug
+fixes.
+
+Time Series changes and improvements
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. note::
+
+    With this release, legacy scikits.timeseries users should be able to port
+    their code to use pandas.
+
+.. note::
+
+    See :ref:`documentation <timeseries>` for overview of pandas timeseries API.
+
+- New datetime64 representation **speeds up join operations and data
+  alignment**, **reduces memory usage**, and improve serialization /
+  deserialization performance significantly over datetime.datetime
+- High performance and flexible **resample** method for converting from
+  high-to-low and low-to-high frequency. Supports interpolation, user-defined
+  aggregation functions, and control over how the intervals and result labeling
+  are defined. A suite of high performance Cython/C-based resampling functions
+  (including Open-High-Low-Close) have also been implemented.
+- Revamp of :ref:`frequency aliases <timeseries.offset_aliases>` and support for
+  **frequency shortcuts** like '15min', or '1h30min'
+- New :ref:`DatetimeIndex class <timeseries.datetimeindex>` supports both fixed
+  frequency and irregular time
+  series. Replaces now deprecated DateRange class
+- New ``PeriodIndex`` and ``Period`` classes for representing
+  :ref:`time spans <timeseries.periods>` and performing **calendar logic**,
+  including the ``12 fiscal quarterly frequencies <timeseries.quarterly>``.
+  This is a partial port of, and a substantial enhancement to,
+  elements of the scikits.timeseries code base. Support for conversion between
+  PeriodIndex and DatetimeIndex
+- New Timestamp data type subclasses ``datetime.datetime``, providing the same
+  interface while enabling working with nanosecond-resolution data. Also
+  provides :ref:`easy time zone conversions <timeseries.timezone>`.
+- Enhanced support for :ref:`time zones <timeseries.timezone>`. Add
+  ``tz_convert`` and ``tz_localize`` methods to TimeSeries and DataFrame. All
+  timestamps are stored as UTC; Timestamps from DatetimeIndex objects with time
+  zone set will be localized to local time. Time zone conversions are therefore
+  essentially free. User needs to know very little about pytz library now; only
+  time zone names as strings are required. Time zone-aware timestamps are
+  equal if and only if their UTC timestamps match. Operations between time
+  zone-aware time series with different time zones will result in a UTC-indexed
+  time series.
+- Time series **string indexing conveniences** / shortcuts: slice years, year
+  and month, and index values with strings
+- Enhanced time series **plotting**; adaptation of scikits.timeseries
+  matplotlib-based plotting code
+- New ``date_range``, ``bdate_range``, and ``period_range`` :ref:`factory
+  functions <timeseries.daterange>`
+- Robust **frequency inference** function ``infer_freq`` and ``inferred_freq``
+  property of DatetimeIndex, with option to infer frequency on construction of
+  DatetimeIndex
+- to_datetime function efficiently **parses array of strings** to
+  DatetimeIndex. DatetimeIndex will parse array or list of strings to
+  datetime64
+- **Optimized** support for datetime64-dtype data in Series and DataFrame
+  columns
+- New NaT (Not-a-Time) type to represent **NA** in timestamp arrays
+- Optimize Series.asof for looking up **"as of" values** for arrays of
+  timestamps
+- Milli, Micro, Nano date offset objects
+- Can index time series with datetime.time objects to select all data at
+  particular **time of day** (``TimeSeries.at_time``) or **between two times**
+  (``TimeSeries.between_time``)
+- Add :ref:`tshift <timeseries.advanced_datetime>` method for leading/lagging
+  using the frequency (if any) of the index, as opposed to a naive lead/lag
+  using shift
+
+Other new features
+~~~~~~~~~~~~~~~~~~
+
+- New :ref:`cut <reshaping.tile.cut>` and ``qcut`` functions (like R's cut
+  function) for computing a categorical variable from a continuous variable by
+  binning values either into value-based (``cut``) or quantile-based (``qcut``)
+  bins
+- Rename ``Factor`` to ``Categorical`` and add a number of usability features
+- Add :ref:`limit <missing_data.fillna.limit>` argument to fillna/reindex
+- More flexible multiple function application in GroupBy, and can pass list
+  (name, function) tuples to get result in particular order with given names
+- Add flexible :ref:`replace <missing_data.replace>` method for efficiently
+  substituting values
+- Enhanced :ref:`read_csv/read_table <io.parse_dates>` for reading time series
+  data and converting multiple columns to dates
+- Add :ref:`comments <io.comments>` option to parser functions: read_csv, etc.
+- Add :ref:`dayfirst <io.dayfirst>` option to parser functions for parsing
+  international DD/MM/YYYY dates
+- Allow the user to specify the CSV reader :ref:`dialect <io.dialect>` to
+  control quoting etc.
+- Handling :ref:`thousands <io.thousands>` separators in read_csv to improve
+  integer parsing.
+- Enable unstacking of multiple levels in one shot. Alleviate ``pivot_table``
+  bugs (empty columns being introduced)
+- Move to klib-based hash tables for indexing; better performance and less
+  memory usage than Python's dict
+- Add first, last, min, max, and prod optimized GroupBy functions
+- New :ref:`ordered_merge <merging.merge_ordered>` function
+- Add flexible :ref:`comparison <basics.binop>` instance methods eq, ne, lt,
+  gt, etc. to DataFrame, Series
+- Improve :ref:`scatter_matrix <visualization.scatter_matrix>` plotting
+  function and add histogram or kernel density estimates to diagonal
+- Add :ref:`'kde' <visualization.kde>` plot option for density plots
+- Support for converting DataFrame to R data.frame through rpy2
+- Improved support for complex numbers in Series and DataFrame
+- Add ``pct_change`` method to all data structures
+- Add max_colwidth configuration option for DataFrame console output
+- :ref:`Interpolate <missing_data.interpolate>` Series values using index values
+- Can select multiple columns from GroupBy
+- Add :ref:`update <merging.combine_first.update>` methods to Series/DataFrame
+  for updating values in place
+- Add ``any`` and ``all`` method to DataFrame
+
+New plotting methods
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: python
+
+   import pandas as pd
+
+   fx = pd.read_pickle("data/fx_prices")
+   import matplotlib.pyplot as plt
+
+``Series.plot`` now supports a ``secondary_y`` option:
+
+.. code-block:: python
+
+   plt.figure()
+
+   fx["FR"].plot(style="g")
+
+   fx["IT"].plot(style="k--", secondary_y=True)
+
+Vytautas Jancauskas, the 2012 GSOC participant, has added many new plot
+types. For example, ``'kde'`` is a new option:
+
+.. code-block:: python
+
+   s = pd.Series(
+       np.concatenate((np.random.randn(1000), np.random.randn(1000) * 0.5 + 3))
+   )
+   plt.figure()
+   s.hist(density=True, alpha=0.2)
+   s.plot(kind="kde")
+
+See :ref:`the plotting page <visualization.other>` for much more.
+
+Other API changes
+~~~~~~~~~~~~~~~~~
+
+- Deprecation of ``offset``, ``time_rule``, and ``timeRule`` arguments names in
+  time series functions. Warnings will be printed until pandas 0.9 or 1.0.
+
+Potential porting issues for pandas <= 0.7.3 users
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The major change that may affect you in pandas 0.8.0 is that time series
+indexes use NumPy's ``datetime64`` data type instead of ``dtype=object`` arrays
+of Python's built-in ``datetime.datetime`` objects. ``DateRange`` has been
+replaced by ``DatetimeIndex`` but otherwise behaved identically. But, if you
+have code that converts ``DateRange`` or ``Index`` objects that used to contain
+``datetime.datetime`` values to plain NumPy arrays, you may have bugs lurking
+with code using scalar values because you are handing control over to NumPy:
+
+.. ipython:: python
+
+   import datetime
+
+   rng = pd.date_range("1/1/2000", periods=10)
+   rng[5]
+   isinstance(rng[5], datetime.datetime)
+   rng_asarray = np.asarray(rng)
+   scalar_val = rng_asarray[5]
+   type(scalar_val)
+
+pandas's ``Timestamp`` object is a subclass of ``datetime.datetime`` that has
+nanosecond support (the ``nanosecond`` field store the nanosecond value between
+0 and 999). It should substitute directly into any code that used
+``datetime.datetime`` values before. Thus, I recommend not casting
+``DatetimeIndex`` to regular NumPy arrays.
+
+If you have code that requires an array of ``datetime.datetime`` objects, you
+have a couple of options. First, the ``astype(object)`` method of ``DatetimeIndex``
+produces an array of ``Timestamp`` objects:
+
+.. ipython:: python
+
+   stamp_array = rng.astype(object)
+   stamp_array
+   stamp_array[5]
+
+To get an array of proper ``datetime.datetime`` objects, use the
+``to_pydatetime`` method:
+
+.. ipython:: python
+
+   dt_array = rng.to_pydatetime()
+   dt_array
+   dt_array[5]
+
+matplotlib knows how to handle ``datetime.datetime`` but not Timestamp
+objects. While I recommend that you plot time series using ``TimeSeries.plot``,
+you can either use ``to_pydatetime`` or register a converter for the Timestamp
+type. See `matplotlib documentation
+<http://matplotlib.org/api/units_api.html>`__ for more on this.
+
+.. warning::
+
+    There are bugs in the user-facing API with the nanosecond datetime64 unit
+    in NumPy 1.6. In particular, the string version of the array shows garbage
+    values, and conversion to ``dtype=object`` is similarly broken.
+
+    .. ipython:: python
+
+       rng = pd.date_range("1/1/2000", periods=10)
+       rng
+       np.asarray(rng)
+       converted = np.asarray(rng, dtype=object)
+       converted[5]
+
+    **Trust me: don't panic**. If you are using NumPy 1.6 and restrict your
+    interaction with ``datetime64`` values to pandas's API you will be just
+    fine. There is nothing wrong with the data-type (a 64-bit integer
+    internally); all of the important data processing happens in pandas and is
+    heavily tested. I strongly recommend that you **do not work directly with
+    datetime64 arrays in NumPy 1.6** and only use the pandas API.
+
+
+**Support for non-unique indexes**: In the latter case, you may have code
+inside a ``try:... catch:`` block that failed due to the index not being
+unique. In many cases it will no longer fail (some method like ``append`` still
+check for uniqueness unless disabled). However, all is not lost: you can
+inspect ``index.is_unique`` and raise an exception explicitly if it is
+``False`` or go to a different code branch.
+
+
+.. _whatsnew_0.8.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.7.3..v0.8.0
diff --git a/pandas/doc/source/whatsnew/v0.8.1.rst b/pandas/doc/source/whatsnew/v0.8.1.rst
new file mode 100644
index 0000000000000..a00a57a0a1cdb
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.8.1.rst
@@ -0,0 +1,47 @@
+.. _whatsnew_0801:
+
+Version 0.8.1 (July 22, 2012)
+-----------------------------
+
+{{ header }}
+
+
+This release includes a few new features, performance enhancements, and over 30
+bug fixes from 0.8.0.  New features include notably NA friendly string
+processing functionality and a series of new plot types and options.
+
+New features
+~~~~~~~~~~~~
+
+  - Add :ref:`vectorized string processing methods <text.string_methods>`
+    accessible via Series.str (:issue:`620`)
+  - Add option to disable adjustment in EWMA (:issue:`1584`)
+  - :ref:`Radviz plot <visualization.radviz>` (:issue:`1566`)
+  - :ref:`Parallel coordinates plot <visualization.parallel_coordinates>`
+  - :ref:`Bootstrap plot <visualization.bootstrap>`
+  - Per column styles and secondary y-axis plotting (:issue:`1559`)
+  - New datetime converters millisecond plotting  (:issue:`1599`)
+  - Add option to disable "sparse" display of hierarchical indexes (:issue:`1538`)
+  - Series/DataFrame's ``set_index`` method can :ref:`append levels
+    <indexing.set_index>` to an existing Index/MultiIndex (:issue:`1569`, :issue:`1577`)
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+  - Improved implementation of rolling min and max (thanks to `Bottleneck
+    <https://bottleneck.readthedocs.io>`__ !)
+  - Add accelerated ``'median'`` GroupBy option (:issue:`1358`)
+  - Significantly improve the performance of parsing ISO8601-format date
+    strings with ``DatetimeIndex`` or ``to_datetime`` (:issue:`1571`)
+  - Improve the performance of GroupBy on single-key aggregations and use with
+    Categorical types
+  - Significant datetime parsing performance improvements
+
+
+
+.. _whatsnew_0.8.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.8.0..v0.8.1
diff --git a/pandas/doc/source/whatsnew/v0.9.0.rst b/pandas/doc/source/whatsnew/v0.9.0.rst
new file mode 100644
index 0000000000000..44ded51e31fda
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.9.0.rst
@@ -0,0 +1,109 @@
+.. _whatsnew_0900:
+
+{{ header }}
+
+
+Version 0.9.0 (October 7, 2012)
+-------------------------------
+
+This is a major release from 0.8.1 and includes several new features and
+enhancements along with a large number of bug fixes. New features include
+vectorized unicode encoding/decoding for ``Series.str``, ``to_latex`` method to
+DataFrame, more flexible parsing of boolean values, and enabling the download of
+options data from Yahoo! Finance.
+
+New features
+~~~~~~~~~~~~
+
+  - Add ``encode`` and ``decode`` for unicode handling to :ref:`vectorized
+    string processing methods <text.string_methods>` in Series.str  (:issue:`1706`)
+  - Add ``DataFrame.to_latex`` method (:issue:`1735`)
+  - Add convenient expanding window equivalents of all rolling_* ops (:issue:`1785`)
+  - Add Options class to pandas.io.data for fetching options data from Yahoo!
+    Finance (:issue:`1748`, :issue:`1739`)
+  - More flexible parsing of boolean values (Yes, No, TRUE, FALSE, etc)
+    (:issue:`1691`, :issue:`1295`)
+  - Add ``level`` parameter to ``Series.reset_index``
+  - ``TimeSeries.between_time`` can now select times across midnight (:issue:`1871`)
+  - Series constructor can now handle generator as input (:issue:`1679`)
+  - ``DataFrame.dropna`` can now take multiple axes (tuple/list) as input
+    (:issue:`924`)
+  - Enable ``skip_footer`` parameter in ``ExcelFile.parse`` (:issue:`1843`)
+
+API changes
+~~~~~~~~~~~
+
+  - The default column names when ``header=None`` and no columns names passed to
+    functions like ``read_csv`` has changed to be more Pythonic and amenable to
+    attribute access:
+
+.. ipython:: python
+
+   import io
+
+   data = """
+   0,0,1
+   1,1,0
+   0,1,0
+   """
+   df = pd.read_csv(io.StringIO(data), header=None)
+   df
+
+
+- Creating a Series from another Series, passing an index, will cause reindexing
+  to happen inside rather than treating the Series like an ndarray. Technically
+  improper usages like ``Series(df[col1], index=df[col2])`` that worked before
+  "by accident" (this was never intended) will lead to all NA Series in some
+  cases. To be perfectly clear:
+
+.. ipython:: python
+
+   s1 = pd.Series([1, 2, 3])
+   s1
+
+   s2 = pd.Series(s1, index=["foo", "bar", "baz"])
+   s2
+
+- Deprecated ``day_of_year`` API removed from PeriodIndex, use ``dayofyear``
+  (:issue:`1723`)
+
+- Don't modify NumPy suppress printoption to True at import time
+
+- The internal HDF5 data arrangement for DataFrames has been transposed.  Legacy
+  files will still be readable by HDFStore (:issue:`1834`, :issue:`1824`)
+
+- Legacy cruft removed: pandas.stats.misc.quantileTS
+
+- Use ISO8601 format for Period repr: monthly, daily, and on down (:issue:`1776`)
+
+- Empty DataFrame columns are now created as object dtype. This will prevent a
+  class of TypeErrors that was occurring in code where the dtype of a column
+  would depend on the presence of data or not (e.g. a SQL query having results)
+  (:issue:`1783`)
+
+- Setting parts of DataFrame/Panel using ix now aligns input Series/DataFrame
+  (:issue:`1630`)
+
+- ``first`` and ``last`` methods in ``GroupBy`` no longer drop non-numeric
+  columns (:issue:`1809`)
+
+- Resolved inconsistencies in specifying custom NA values in text parser.
+  ``na_values`` of type dict no longer override default NAs unless
+  ``keep_default_na`` is set to false explicitly (:issue:`1657`)
+
+- ``DataFrame.dot`` will not do data alignment, and also work with Series
+  (:issue:`1915`)
+
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+
+.. _whatsnew_0.9.0.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.8.1..v0.9.0
diff --git a/pandas/doc/source/whatsnew/v0.9.1.rst b/pandas/doc/source/whatsnew/v0.9.1.rst
new file mode 100644
index 0000000000000..cdc0671feeeb2
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v0.9.1.rst
@@ -0,0 +1,171 @@
+.. _whatsnew_0901:
+
+Version 0.9.1 (November 14, 2012)
+---------------------------------
+
+{{ header }}
+
+
+This is a bug fix release from 0.9.0 and includes several new features and
+enhancements along with a large number of bug fixes. The new features include
+by-column sort order for DataFrame and Series, improved NA handling for the rank
+method, masking functions for DataFrame, and intraday time-series filtering for
+DataFrame.
+
+New features
+~~~~~~~~~~~~
+
+  - ``Series.sort``, ``DataFrame.sort``, and ``DataFrame.sort_index`` can now be
+    specified in a per-column manner to support multiple sort orders (:issue:`928`)
+
+    .. code-block:: ipython
+
+       In [2]: df = pd.DataFrame(np.random.randint(0, 2, (6, 3)),
+          ...:                   columns=['A', 'B', 'C'])
+
+       In [3]: df.sort(['A', 'B'], ascending=[1, 0])
+
+       Out[3]:
+          A  B  C
+       3  0  1  1
+       4  0  1  1
+       2  0  0  1
+       0  1  0  0
+       1  1  0  0
+       5  1  0  0
+
+  - ``DataFrame.rank`` now supports additional argument values for the
+    ``na_option`` parameter so missing values can be assigned either the largest
+    or the smallest rank (:issue:`1508`, :issue:`2159`)
+
+    .. ipython:: python
+
+        df = pd.DataFrame(np.random.randn(6, 3), columns=['A', 'B', 'C'])
+
+        df.loc[2:4] = np.nan
+
+        df.rank()
+
+        df.rank(na_option='top')
+
+        df.rank(na_option='bottom')
+
+
+  - DataFrame has new ``where`` and ``mask`` methods to select values according to a
+    given boolean mask (:issue:`2109`, :issue:`2151`)
+
+        DataFrame currently supports slicing via a boolean vector the same length as the DataFrame (inside the ``[]``).
+        The returned DataFrame has the same number of columns as the original, but is sliced on its index.
+
+        .. ipython:: python
+
+            df = pd.DataFrame(np.random.randn(5, 3), columns=['A', 'B', 'C'])
+
+            df
+
+            df[df['A'] > 0]
+
+        If a DataFrame is sliced with a DataFrame based boolean condition (with the same size as the original DataFrame),
+        then a DataFrame the same size (index and columns) as the original is returned, with
+        elements that do not meet the boolean condition as ``NaN``. This is accomplished via
+        the new method ``DataFrame.where``. In addition, ``where`` takes an optional ``other`` argument for replacement.
+
+        .. ipython:: python
+
+           df[df > 0]
+
+           df.where(df > 0)
+
+           df.where(df > 0, -df)
+
+        Furthermore, ``where`` now aligns the input boolean condition (ndarray or DataFrame), such that partial selection
+        with setting is possible. This is analogous to partial setting via ``.ix`` (but on the contents rather than the axis labels)
+
+        .. ipython:: python
+
+           df2 = df.copy()
+           df2[df2[1:4] > 0] = 3
+           df2
+
+        ``DataFrame.mask`` is the inverse boolean operation of ``where``.
+
+        .. ipython:: python
+
+           df.mask(df <= 0)
+
+  - Enable referencing of Excel columns by their column names (:issue:`1936`)
+
+    .. code-block:: ipython
+
+       In [1]: xl = pd.ExcelFile('data/test.xls')
+
+       In [2]: xl.parse('Sheet1', index_col=0, parse_dates=True,
+                        parse_cols='A:D')
+
+
+  - Added option to disable pandas-style tick locators and formatters
+    using ``series.plot(x_compat=True)`` or ``pandas.plot_params['x_compat'] =
+    True`` (:issue:`2205`)
+  - Existing TimeSeries methods ``at_time`` and ``between_time`` were added to
+    DataFrame (:issue:`2149`)
+  - DataFrame.dot can now accept ndarrays (:issue:`2042`)
+  - DataFrame.drop now supports non-unique indexes (:issue:`2101`)
+  - Panel.shift now supports negative periods (:issue:`2164`)
+  - DataFrame now support unary ~ operator (:issue:`2110`)
+
+API changes
+~~~~~~~~~~~
+
+  - Upsampling data with a PeriodIndex will result in a higher frequency
+    TimeSeries that spans the original time window
+
+    .. code-block:: ipython
+
+       In [1]: prng = pd.period_range('2012Q1', periods=2, freq='Q')
+
+       In [2]: s = pd.Series(np.random.randn(len(prng)), prng)
+
+       In [4]: s.resample('M')
+       Out[4]:
+       2012-01   -1.471992
+       2012-02         NaN
+       2012-03         NaN
+       2012-04   -0.493593
+       2012-05         NaN
+       2012-06         NaN
+       Freq: M, dtype: float64
+
+  - Period.end_time now returns the last nanosecond in the time interval
+    (:issue:`2124`, :issue:`2125`, :issue:`1764`)
+
+    .. ipython:: python
+
+        p = pd.Period('2012')
+
+        p.end_time
+
+
+  - File parsers no longer coerce to float or bool for columns that have custom
+    converters specified (:issue:`2184`)
+
+    .. ipython:: python
+
+        import io
+
+        data = ('A,B,C\n'
+                '00001,001,5\n'
+                '00002,002,6')
+        pd.read_csv(io.StringIO(data), converters={'A': lambda x: x.strip()})
+
+
+See the :ref:`full release notes
+<release>` or issue tracker
+on GitHub for a complete list.
+
+
+.. _whatsnew_0.9.1.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.9.0..v0.9.1
diff --git a/pandas/doc/source/whatsnew/v1.0.0.rst b/pandas/doc/source/whatsnew/v1.0.0.rst
new file mode 100755
index 0000000000000..98cb9c4ad7b45
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.0.rst
@@ -0,0 +1,1302 @@
+.. _whatsnew_100:
+
+What's new in 1.0.0 (January 29, 2020)
+--------------------------------------
+
+These are the changes in pandas 1.0.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+.. note::
+
+    The pandas 1.0 release removed a lot of functionality that was deprecated
+    in previous releases (see :ref:`below <whatsnew_100.prior_deprecations>`
+    for an overview). It is recommended to first upgrade to pandas 0.25 and to
+    ensure your code is working without warnings, before upgrading to pandas
+    1.0.
+
+
+New deprecation policy
+~~~~~~~~~~~~~~~~~~~~~~
+
+Starting with pandas 1.0.0, pandas will adopt a variant of `SemVer`_ to
+version releases. Briefly,
+
+* Deprecations will be introduced in minor releases (e.g. 1.1.0, 1.2.0, 2.1.0, ...)
+* Deprecations will be enforced in major releases (e.g. 1.0.0, 2.0.0, 3.0.0, ...)
+* API-breaking changes will be made only in major releases (except for experimental features)
+
+See :ref:`policies.version` for more.
+
+.. _2019 Pandas User Survey: https://pandas.pydata.org/community/blog/2019-user-survey.html
+.. _SemVer: https://semver.org
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_100.numba_rolling_apply:
+
+Using Numba in ``rolling.apply`` and ``expanding.apply``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added an ``engine`` keyword to :meth:`~core.window.rolling.Rolling.apply` and :meth:`~core.window.expanding.Expanding.apply`
+that allows the user to execute the routine using `Numba <https://numba.pydata.org/>`__ instead of Cython.
+Using the Numba engine can yield significant performance gains if the apply function can operate on numpy arrays and
+the data set is larger (1 million rows or greater). For more details, see
+:ref:`rolling apply documentation <window.numba_engine>` (:issue:`28987`, :issue:`30936`)
+
+.. _whatsnew_100.custom_window:
+
+Defining custom windows for rolling operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added a :func:`pandas.api.indexers.BaseIndexer` class that allows users to define how
+window bounds are created during ``rolling`` operations. Users can define their own ``get_window_bounds``
+method on a :func:`pandas.api.indexers.BaseIndexer` subclass that will generate the start and end
+indices used for each window during the rolling aggregation. For more details and example usage, see
+the :ref:`custom window rolling documentation <window.custom_rolling_window>`
+
+.. _whatsnew_100.to_markdown:
+
+Converting to markdown
+^^^^^^^^^^^^^^^^^^^^^^
+
+We've added :meth:`~DataFrame.to_markdown` for creating a markdown table (:issue:`11052`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2, 3], "B": [1, 2, 3]}, index=['a', 'a', 'b'])
+   print(df.to_markdown())
+
+Experimental new features
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_100.NA:
+
+Experimental ``NA`` scalar to denote missing values
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A new ``pd.NA`` value (singleton) is introduced to represent scalar missing
+values. Up to now, pandas used several values to represent missing data: ``np.nan`` is used for this for float data, ``np.nan`` or
+``None`` for object-dtype data and ``pd.NaT`` for datetime-like data. The
+goal of ``pd.NA`` is to provide a "missing" indicator that can be used
+consistently across data types. ``pd.NA`` is currently used by the nullable integer and boolean
+data types and the new string data type (:issue:`28095`).
+
+.. warning::
+
+   Experimental: the behaviour of ``pd.NA`` can still change without warning.
+
+For example, creating a Series using the nullable integer dtype:
+
+.. ipython:: python
+
+    s = pd.Series([1, 2, None], dtype="Int64")
+    s
+    s[2]
+
+Compared to ``np.nan``, ``pd.NA`` behaves differently in certain operations.
+In addition to arithmetic operations, ``pd.NA`` also propagates as "missing"
+or "unknown" in comparison operations:
+
+.. ipython:: python
+
+    np.nan > 1
+    pd.NA > 1
+
+For logical operations, ``pd.NA`` follows the rules of the
+`three-valued logic <https://en.wikipedia.org/wiki/Three-valued_logic>`__ (or
+*Kleene logic*). For example:
+
+.. ipython:: python
+
+    pd.NA | True
+
+For more, see :ref:`NA section <missing_data.NA>` in the user guide on missing
+data.
+
+
+.. _whatsnew_100.string:
+
+Dedicated string data type
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added :class:`StringDtype`, an extension type dedicated to string data.
+Previously, strings were typically stored in object-dtype NumPy arrays. (:issue:`29975`)
+
+.. warning::
+
+   ``StringDtype`` is currently considered experimental. The implementation
+   and parts of the API may change without warning.
+
+The ``'string'`` extension type solves several issues with object-dtype NumPy arrays:
+
+1. You can accidentally store a *mixture* of strings and non-strings in an
+   ``object`` dtype array. A ``StringArray`` can only store strings.
+2. ``object`` dtype breaks dtype-specific operations like :meth:`DataFrame.select_dtypes`.
+   There isn't a clear way to select *just* text while excluding non-text,
+   but still object-dtype columns.
+3. When reading code, the contents of an ``object`` dtype array is less clear
+   than ``string``.
+
+
+.. ipython:: python
+
+   pd.Series(['abc', None, 'def'], dtype=pd.StringDtype())
+
+You can use the alias ``"string"`` as well.
+
+.. ipython:: python
+
+   s = pd.Series(['abc', None, 'def'], dtype="string")
+   s
+
+The usual string accessor methods work. Where appropriate, the return type
+of the Series or columns of a DataFrame will also have string dtype.
+
+.. ipython:: python
+
+   s.str.upper()
+   s.str.split('b', expand=True).dtypes
+
+String accessor methods returning integers will return a value with :class:`Int64Dtype`
+
+.. ipython:: python
+
+   s.str.count("a")
+
+We recommend explicitly using the ``string`` data type when working with strings.
+See :ref:`text.types` for more.
+
+.. _whatsnew_100.boolean:
+
+Boolean data type with missing values support
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added :class:`BooleanDtype` / :class:`~arrays.BooleanArray`, an extension
+type dedicated to boolean data that can hold missing values. The default
+``bool`` data type based on a bool-dtype NumPy array, the column can only hold
+``True`` or ``False``, and not missing values. This new :class:`~arrays.BooleanArray`
+can store missing values as well by keeping track of this in a separate mask.
+(:issue:`29555`, :issue:`30095`, :issue:`31131`)
+
+.. ipython:: python
+
+   pd.Series([True, False, None], dtype=pd.BooleanDtype())
+
+You can use the alias ``"boolean"`` as well.
+
+.. ipython:: python
+
+   s = pd.Series([True, False, None], dtype="boolean")
+   s
+
+.. _whatsnew_100.convert_dtypes:
+
+Method ``convert_dtypes`` to ease use of supported extension dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In order to encourage use of the extension dtypes ``StringDtype``,
+``BooleanDtype``, ``Int64Dtype``, ``Int32Dtype``, etc., that support ``pd.NA``, the
+methods :meth:`DataFrame.convert_dtypes` and :meth:`Series.convert_dtypes`
+have been introduced. (:issue:`29752`) (:issue:`30929`)
+
+Example:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'x': ['abc', None, 'def'],
+                      'y': [1, 2, np.nan],
+                      'z': [True, False, True]})
+   df
+   df.dtypes
+
+.. ipython:: python
+
+   converted = df.convert_dtypes()
+   converted
+   converted.dtypes
+
+This is especially useful after reading in data using readers such as :func:`read_csv`
+and :func:`read_excel`.
+See :ref:`here <missing_data.NA.conversion>` for a description.
+
+
+.. _whatsnew_100.enhancements.other:
+
+Other enhancements
+~~~~~~~~~~~~~~~~~~
+
+- :meth:`DataFrame.to_string` added the ``max_colwidth`` parameter to control when wide columns are truncated (:issue:`9784`)
+- Added the ``na_value`` argument to :meth:`Series.to_numpy`, :meth:`Index.to_numpy` and :meth:`DataFrame.to_numpy` to control the value used for missing data (:issue:`30322`)
+- :meth:`MultiIndex.from_product` infers level names from inputs if not explicitly provided (:issue:`27292`)
+- :meth:`DataFrame.to_latex` now accepts ``caption`` and ``label`` arguments (:issue:`25436`)
+- DataFrames with :ref:`nullable integer <integer_na>`, the :ref:`new string dtype <text.types>`
+  and period data type can now be converted to ``pyarrow`` (>=0.15.0), which means that it is
+  supported in writing to the Parquet file format when using the ``pyarrow`` engine (:issue:`28368`).
+  Full roundtrip to parquet (writing and reading back in with :meth:`~DataFrame.to_parquet` / :func:`read_parquet`)
+  is supported starting with pyarrow >= 0.16 (:issue:`20612`).
+- :func:`to_parquet` now appropriately handles the ``schema`` argument for user defined schemas in the pyarrow engine. (:issue:`30270`)
+- :meth:`DataFrame.to_json` now accepts an ``indent`` integer argument to enable pretty printing of JSON output (:issue:`12004`)
+- :meth:`read_stata` can read Stata 119 dta files. (:issue:`28250`)
+- Implemented :meth:`.Window.var` and :meth:`.Window.std` functions (:issue:`26597`)
+- Added ``encoding`` argument to :meth:`DataFrame.to_string` for non-ascii text (:issue:`28766`)
+- Added ``encoding`` argument to :func:`DataFrame.to_html` for non-ascii text (:issue:`28663`)
+- :meth:`Styler.background_gradient` now accepts ``vmin`` and ``vmax`` arguments (:issue:`12145`)
+- :meth:`Styler.format` added the ``na_rep`` parameter to help format the missing values (:issue:`21527`, :issue:`28358`)
+- :func:`read_excel` now can read binary Excel (``.xlsb``) files by passing ``engine='pyxlsb'``. For more details and example usage, see the :ref:`Binary Excel files documentation <io.xlsb>`. Closes :issue:`8540`.
+- The ``partition_cols`` argument in :meth:`DataFrame.to_parquet` now accepts a string (:issue:`27117`)
+- :func:`pandas.read_json` now parses ``NaN``, ``Infinity`` and ``-Infinity`` (:issue:`12213`)
+- DataFrame constructor preserve ``ExtensionArray`` dtype with ``ExtensionArray`` (:issue:`11363`)
+- :meth:`DataFrame.sort_values` and :meth:`Series.sort_values` have gained ``ignore_index`` keyword to be able to reset index after sorting (:issue:`30114`)
+- :meth:`DataFrame.sort_index` and :meth:`Series.sort_index` have gained ``ignore_index`` keyword to reset index (:issue:`30114`)
+- :meth:`DataFrame.drop_duplicates` has gained ``ignore_index`` keyword to reset index (:issue:`30114`)
+- Added new writer for exporting Stata dta files in versions 118 and 119, ``StataWriterUTF8``.  These files formats support exporting strings containing Unicode characters. Format 119 supports data sets with more than 32,767 variables (:issue:`23573`, :issue:`30959`)
+- :meth:`Series.map` now accepts ``collections.abc.Mapping`` subclasses as a mapper (:issue:`29733`)
+- Added an experimental :attr:`~DataFrame.attrs` for storing global metadata about a dataset (:issue:`29062`)
+- :meth:`Timestamp.fromisocalendar` is now compatible with python 3.8 and above (:issue:`28115`)
+- :meth:`DataFrame.to_pickle` and :func:`read_pickle` now accept URL (:issue:`30163`)
+
+
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_100.api_breaking.MultiIndex._names:
+
+Avoid using names from ``MultiIndex.levels``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+As part of a larger refactor to :class:`MultiIndex` the level names are now
+stored separately from the levels (:issue:`27242`). We recommend using
+:attr:`MultiIndex.names` to access the names, and :meth:`Index.set_names`
+to update the names.
+
+For backwards compatibility, you can still *access* the names via the levels.
+
+.. ipython:: python
+
+   mi = pd.MultiIndex.from_product([[1, 2], ['a', 'b']], names=['x', 'y'])
+   mi.levels[0].name
+
+However, it is no longer possible to *update* the names of the ``MultiIndex``
+via the level.
+
+.. ipython:: python
+   :okexcept:
+
+   mi.levels[0].name = "new name"
+   mi.names
+
+To update, use ``MultiIndex.set_names``, which returns a new ``MultiIndex``.
+
+.. ipython:: python
+
+   mi2 = mi.set_names("new name", level=0)
+   mi2.names
+
+New repr for :class:`~pandas.arrays.IntervalArray`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`pandas.arrays.IntervalArray` adopts a new ``__repr__`` in accordance with other array classes (:issue:`25022`)
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: pd.arrays.IntervalArray.from_tuples([(0, 1), (2, 3)])
+   Out[2]:
+   IntervalArray([(0, 1], (2, 3]],
+                 closed='right',
+                 dtype='interval[int64]')
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   pd.arrays.IntervalArray.from_tuples([(0, 1), (2, 3)])
+
+``DataFrame.rename`` now only accepts one positional argument
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.rename` would previously accept positional arguments that would lead
+to ambiguous or undefined behavior. From pandas 1.0, only the very first argument, which
+maps labels to their new names along the default axis, is allowed to be passed by position
+(:issue:`29136`).
+
+.. ipython:: python
+   :suppress:
+
+   df = pd.DataFrame([[1]])
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame([[1]])
+   In [2]: df.rename({0: 1}, {0: 2})
+   Out[2]:
+   FutureWarning: ...Use named arguments to resolve ambiguity...
+      2
+   1  1
+
+*pandas 1.0.0*
+
+.. code-block:: ipython
+
+   In [3]: df.rename({0: 1}, {0: 2})
+   Traceback (most recent call last):
+   ...
+   TypeError: rename() takes from 1 to 2 positional arguments but 3 were given
+
+Note that errors will now be raised when conflicting or potentially ambiguous arguments are provided.
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [4]: df.rename({0: 1}, index={0: 2})
+   Out[4]:
+      0
+   1  1
+
+   In [5]: df.rename(mapper={0: 1}, index={0: 2})
+   Out[5]:
+      0
+   2  1
+
+*pandas 1.0.0*
+
+.. code-block:: ipython
+
+   In [6]: df.rename({0: 1}, index={0: 2})
+   Traceback (most recent call last):
+   ...
+   TypeError: Cannot specify both 'mapper' and any of 'index' or 'columns'
+
+   In [7]: df.rename(mapper={0: 1}, index={0: 2})
+   Traceback (most recent call last):
+   ...
+   TypeError: Cannot specify both 'mapper' and any of 'index' or 'columns'
+
+You can still change the axis along which the first positional argument is applied by
+supplying the ``axis`` keyword argument.
+
+.. ipython:: python
+
+   df.rename({0: 1})
+   df.rename({0: 1}, axis=1)
+
+If you would like to update both the index and column labels, be sure to use the respective
+keywords.
+
+.. ipython:: python
+
+   df.rename(index={0: 1}, columns={0: 2})
+
+Extended verbose info output for :class:`~pandas.DataFrame`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.info` now shows line numbers for the columns summary (:issue:`17304`)
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: df = pd.DataFrame({"int_col": [1, 2, 3],
+   ...                    "text_col": ["a", "b", "c"],
+   ...                    "float_col": [0.0, 0.1, 0.2]})
+   In [2]: df.info(verbose=True)
+   <class 'pandas.DataFrame'>
+   RangeIndex: 3 entries, 0 to 2
+   Data columns (total 3 columns):
+   int_col      3 non-null int64
+   text_col     3 non-null object
+   float_col    3 non-null float64
+   dtypes: float64(1), int64(1), object(1)
+   memory usage: 152.0+ bytes
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   df = pd.DataFrame({"int_col": [1, 2, 3],
+                      "text_col": ["a", "b", "c"],
+                      "float_col": [0.0, 0.1, 0.2]})
+   df.info(verbose=True)
+
+:meth:`pandas.array` inference changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`pandas.array` now infers pandas' new extension types in several cases (:issue:`29791`):
+
+1. String data (including missing values) now returns a :class:`arrays.StringArray`.
+2. Integer data (including missing values) now returns a :class:`arrays.IntegerArray`.
+3. Boolean data (including missing values) now returns the new :class:`arrays.BooleanArray`
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: pd.array(["a", None])
+   Out[1]:
+   <PandasArray>
+   ['a', None]
+   Length: 2, dtype: object
+
+   In [2]: pd.array([1, None])
+   Out[2]:
+   <PandasArray>
+   [1, None]
+   Length: 2, dtype: object
+
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   pd.array(["a", None])
+   pd.array([1, None])
+
+As a reminder, you can specify the ``dtype`` to disable all inference.
+
+:class:`arrays.IntegerArray` now uses :attr:`pandas.NA`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`arrays.IntegerArray` now uses :attr:`pandas.NA` rather than
+:attr:`numpy.nan` as its missing value marker (:issue:`29964`).
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: a = pd.array([1, 2, None], dtype="Int64")
+   In [2]: a
+   Out[2]:
+   <IntegerArray>
+   [1, 2, NaN]
+   Length: 3, dtype: Int64
+
+   In [3]: a[2]
+   Out[3]:
+   nan
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   a = pd.array([1, 2, None], dtype="Int64")
+   a
+   a[2]
+
+This has a few API-breaking consequences.
+
+**Converting to a NumPy ndarray**
+
+When converting to a NumPy array missing values will be ``pd.NA``, which cannot
+be converted to a float. So calling ``np.asarray(integer_array, dtype="float")``
+will now raise.
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+    In [1]: np.asarray(a, dtype="float")
+    Out[1]:
+    array([ 1.,  2., nan])
+
+*pandas 1.0.0*
+
+.. ipython:: python
+   :okexcept:
+
+   np.asarray(a, dtype="float")
+
+Use :meth:`arrays.IntegerArray.to_numpy` with an explicit ``na_value`` instead.
+
+.. ipython:: python
+
+   a.to_numpy(dtype="float", na_value=np.nan)
+
+**Reductions can return** ``pd.NA``
+
+When performing a reduction such as a sum with ``skipna=False``, the result
+will now be ``pd.NA`` instead of ``np.nan`` in presence of missing values
+(:issue:`30958`).
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+    In [1]: pd.Series(a).sum(skipna=False)
+    Out[1]:
+    nan
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   pd.Series(a).sum(skipna=False)
+
+**value_counts returns a nullable integer dtype**
+
+:meth:`Series.value_counts` with a nullable integer dtype now returns a nullable
+integer dtype for the values.
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: pd.Series([2, 1, 1, None], dtype="Int64").value_counts().dtype
+   Out[1]:
+   dtype('int64')
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   pd.Series([2, 1, 1, None], dtype="Int64").value_counts().dtype
+
+See :ref:`missing_data.NA` for more on the differences between :attr:`pandas.NA`
+and :attr:`numpy.nan`.
+
+:class:`arrays.IntegerArray` comparisons return :class:`arrays.BooleanArray`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Comparison operations on a :class:`arrays.IntegerArray` now returns a
+:class:`arrays.BooleanArray` rather than a NumPy array (:issue:`29964`).
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: a = pd.array([1, 2, None], dtype="Int64")
+   In [2]: a
+   Out[2]:
+   <IntegerArray>
+   [1, 2, NaN]
+   Length: 3, dtype: Int64
+
+   In [3]: a > 1
+   Out[3]:
+   array([False,  True, False])
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   a = pd.array([1, 2, None], dtype="Int64")
+   a > 1
+
+Note that missing values now propagate, rather than always comparing unequal
+like :attr:`numpy.nan`. See :ref:`missing_data.NA` for more.
+
+By default :meth:`Categorical.min` now returns the minimum instead of np.nan
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When :class:`Categorical` contains ``np.nan``,
+:meth:`Categorical.min` no longer return ``np.nan`` by default (skipna=True) (:issue:`25303`)
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]: pd.Categorical([1, 2, np.nan], ordered=True).min()
+   Out[1]: nan
+
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   pd.Categorical([1, 2, np.nan], ordered=True).min()
+
+
+Default dtype of empty :class:`pandas.Series`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Initialising an empty :class:`pandas.Series` without specifying a dtype will raise a ``DeprecationWarning`` now
+(:issue:`17261`). The default dtype will change from ``float64`` to ``object`` in future releases so that it is
+consistent with the behaviour of :class:`DataFrame` and :class:`Index`.
+
+*pandas 1.0.0*
+
+.. code-block:: ipython
+
+   In [1]: pd.Series()
+   Out[2]:
+   DeprecationWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.
+   Series([], dtype: float64)
+
+Result dtype inference changes for resample operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The rules for the result dtype in :meth:`DataFrame.resample` aggregations have changed for extension types (:issue:`31359`).
+Previously, pandas would attempt to convert the result back to the original dtype, falling back to the usual
+inference rules if that was not possible. Now, pandas will only return a result of the original dtype if the
+scalar values in the result are instances of the extension dtype's scalar type.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": ['a', 'b']}, dtype='category',
+                     index=pd.date_range('2000', periods=2))
+   df
+
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1]> df.resample("2D").agg(lambda x: 'a').A.dtype
+   Out[1]:
+   CategoricalDtype(categories=['a', 'b'], ordered=False)
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   df.resample("2D").agg(lambda x: 'a').A.dtype
+
+This fixes an inconsistency between ``resample`` and ``groupby``.
+This also fixes a potential bug, where the **values** of the result might change
+depending on how the results are cast back to the original dtype.
+
+*pandas 0.25.x*
+
+.. code-block:: ipython
+
+   In [1] df.resample("2D").agg(lambda x: 'c')
+   Out[1]:
+
+        A
+   0  NaN
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   df.resample("2D").agg(lambda x: 'c')
+
+
+.. _whatsnew_100.api_breaking.python:
+
+Increased minimum version for Python
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 1.0.0 supports Python 3.6.1 and higher (:issue:`29212`).
+
+.. _whatsnew_100.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Some minimum supported versions of dependencies were updated (:issue:`29766`, :issue:`29723`).
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.13.3          |    X     |         |
++-----------------+-----------------+----------+---------+
+| pytz            | 2015.4          |    X     |         |
++-----------------+-----------------+----------+---------+
+| python-dateutil | 2.6.1           |    X     |         |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.2.1           |          |         |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.6.2           |          |         |
++-----------------+-----------------+----------+---------+
+| pytest (dev)    | 4.0.2           |          |         |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  | 4.6.0           |         |
++-----------------+-----------------+---------+
+| fastparquet     | 0.3.2           |    X    |
++-----------------+-----------------+---------+
+| gcsfs           | 0.2.2           |         |
++-----------------+-----------------+---------+
+| lxml            | 3.8.0           |         |
++-----------------+-----------------+---------+
+| matplotlib      | 2.2.2           |         |
++-----------------+-----------------+---------+
+| numba           | 0.46.0          |    X    |
++-----------------+-----------------+---------+
+| openpyxl        | 2.5.7           |    X    |
++-----------------+-----------------+---------+
+| pyarrow         | 0.13.0          |    X    |
++-----------------+-----------------+---------+
+| pymysql         | 0.7.1           |         |
++-----------------+-----------------+---------+
+| pytables        | 3.4.2           |         |
++-----------------+-----------------+---------+
+| s3fs            | 0.3.0           |    X    |
++-----------------+-----------------+---------+
+| scipy           | 0.19.0          |         |
++-----------------+-----------------+---------+
+| sqlalchemy      | 1.1.4           |         |
++-----------------+-----------------+---------+
+| xarray          | 0.8.2           |         |
++-----------------+-----------------+---------+
+| xlrd            | 1.1.0           |         |
++-----------------+-----------------+---------+
+| xlsxwriter      | 0.9.8           |         |
++-----------------+-----------------+---------+
+| xlwt            | 1.2.0           |         |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+Build changes
+^^^^^^^^^^^^^
+
+pandas has added a `pyproject.toml <https://www.python.org/dev/peps/pep-0517/>`_ file and will no longer include
+cythonized files in the source distribution uploaded to PyPI (:issue:`28341`, :issue:`20775`). If you're installing
+a built distribution (wheel) or via conda, this shouldn't have any effect on you. If you're building pandas from
+source, you should no longer need to install Cython into your build environment before calling ``pip install pandas``.
+
+
+.. _whatsnew_100.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` now raises on invalid operation names (:issue:`27489`)
+- :meth:`pandas.api.types.infer_dtype` will now return "integer-na" for integer and ``np.nan`` mix (:issue:`27283`)
+- :meth:`MultiIndex.from_arrays` will no longer infer names from arrays if ``names=None`` is explicitly provided (:issue:`27292`)
+- In order to improve tab-completion, pandas does not include most deprecated attributes when introspecting a pandas object using ``dir`` (e.g. ``dir(df)``).
+  To see which attributes are excluded, see an object's ``_deprecations`` attribute, for example ``pd.DataFrame._deprecations`` (:issue:`28805`).
+- The returned dtype of :func:`unique` now matches the input dtype. (:issue:`27874`)
+- Changed the default configuration value for ``options.matplotlib.register_converters`` from ``True`` to ``"auto"`` (:issue:`18720`).
+  Now, pandas custom formatters will only be applied to plots created by pandas, through :meth:`~DataFrame.plot`.
+  Previously, pandas' formatters would be applied to all plots created *after* a :meth:`~DataFrame.plot`.
+  See :ref:`units registration <whatsnew_100.matplotlib_units>` for more.
+- :meth:`Series.dropna` has dropped its ``**kwargs`` argument in favor of a single ``how`` parameter.
+  Supplying anything else than ``how`` to ``**kwargs`` raised a ``TypeError`` previously (:issue:`29388`)
+- When testing pandas, the new minimum required version of pytest is 5.0.1 (:issue:`29664`)
+- :meth:`Series.str.__iter__` was deprecated and will be removed in future releases (:issue:`28277`).
+- Added ``<NA>`` to the list of default NA values for :meth:`read_csv` (:issue:`30821`)
+
+.. _whatsnew_100.api.documentation:
+
+Documentation improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Added new section on :ref:`scale` (:issue:`28315`).
+- Added sub-section on :ref:`io.query_multi` for HDF5 datasets (:issue:`28791`).
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- :meth:`Series.item` and :meth:`Index.item` have been _undeprecated_ (:issue:`29250`)
+- ``Index.set_value`` has been deprecated. For a given index ``idx``, array ``arr``,
+  value in ``idx`` of ``idx_val`` and a new value of ``val``, ``idx.set_value(arr, idx_val, val)``
+  is equivalent to ``arr[idx.get_loc(idx_val)] = val``, which should be used instead (:issue:`28621`).
+- :func:`is_extension_type` is deprecated, :func:`is_extension_array_dtype` should be used instead (:issue:`29457`)
+- :func:`eval` keyword argument "truediv" is deprecated and will be removed in a future version (:issue:`29812`)
+- :meth:`DateOffset.isAnchored` and :meth:`DatetOffset.onOffset` are deprecated and will be removed in a future version, use :meth:`DateOffset.is_anchored` and :meth:`DateOffset.is_on_offset` instead (:issue:`30340`)
+- ``pandas.tseries.frequencies.get_offset`` is deprecated and will be removed in a future version, use ``pandas.tseries.frequencies.to_offset`` instead (:issue:`4205`)
+- :meth:`Categorical.take_nd` and :meth:`CategoricalIndex.take_nd` are deprecated, use :meth:`Categorical.take` and :meth:`CategoricalIndex.take` instead (:issue:`27745`)
+- The parameter ``numeric_only`` of :meth:`Categorical.min` and :meth:`Categorical.max` is deprecated and replaced with ``skipna`` (:issue:`25303`)
+- The parameter ``label`` in :func:`lreshape` has been deprecated and will be removed in a future version (:issue:`29742`)
+- ``pandas.core.index`` has been deprecated and will be removed in a future version, the public classes are available in the top-level namespace (:issue:`19711`)
+- :func:`pandas.json_normalize` is now exposed in the top-level namespace.
+  Usage of ``json_normalize`` as ``pandas.io.json.json_normalize`` is now deprecated and
+  it is recommended to use ``json_normalize`` as :func:`pandas.json_normalize` instead (:issue:`27586`).
+- The ``numpy`` argument of :meth:`pandas.read_json` is deprecated (:issue:`28512`).
+- :meth:`DataFrame.to_stata`, :meth:`DataFrame.to_feather`, and :meth:`DataFrame.to_parquet` argument "fname" is deprecated, use "path" instead (:issue:`23574`)
+- The deprecated internal attributes ``_start``, ``_stop`` and ``_step`` of :class:`RangeIndex` now raise a ``FutureWarning`` instead of a ``DeprecationWarning`` (:issue:`26581`)
+- The ``pandas.util.testing`` module has been deprecated. Use the public API in ``pandas.testing`` documented at :ref:`api.general.testing` (:issue:`16232`).
+- ``pandas.SparseArray`` has been deprecated.  Use ``pandas.arrays.SparseArray`` (:class:`arrays.SparseArray`) instead. (:issue:`30642`)
+- The parameter ``is_copy`` of :meth:`Series.take` and :meth:`DataFrame.take` has been deprecated and will be removed in a future version. (:issue:`27357`)
+- Support for multi-dimensional indexing (e.g. ``index[:, None]``) on a :class:`Index` is deprecated and will be removed in a future version, convert to a numpy array before indexing instead (:issue:`30588`)
+- The ``pandas.np`` submodule is now deprecated. Import numpy directly instead (:issue:`30296`)
+- The ``pandas.datetime`` class is now deprecated. Import from ``datetime`` instead (:issue:`30610`)
+- :class:`~DataFrame.diff` will raise a ``TypeError`` rather than implicitly losing the dtype of extension types in the future. Convert to the correct dtype before calling ``diff`` instead (:issue:`31025`)
+
+**Selecting Columns from a Grouped DataFrame**
+
+When selecting columns from a :class:`DataFrameGroupBy` object, passing individual keys (or a tuple of keys) inside single brackets is deprecated,
+a list of items should be used instead. (:issue:`23566`) For example:
+
+.. code-block:: ipython
+
+    df = pd.DataFrame({
+        "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+        "B": np.random.randn(8),
+        "C": np.random.randn(8),
+    })
+    g = df.groupby('A')
+
+    # single key, returns SeriesGroupBy
+    g['B']
+
+    # tuple of single key, returns SeriesGroupBy
+    g[('B',)]
+
+    # tuple of multiple keys, returns DataFrameGroupBy, raises FutureWarning
+    g[('B', 'C')]
+
+    # multiple keys passed directly, returns DataFrameGroupBy, raises FutureWarning
+    # (implicitly converts the passed strings into a single tuple)
+    g['B', 'C']
+
+    # proper way, returns DataFrameGroupBy
+    g[['B', 'C']]
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+**Removed SparseSeries and SparseDataFrame**
+
+``SparseSeries``, ``SparseDataFrame`` and the ``DataFrame.to_sparse`` method
+have been removed (:issue:`28425`). We recommend using a ``Series`` or
+``DataFrame`` with sparse values instead.
+
+.. _whatsnew_100.matplotlib_units:
+
+**Matplotlib unit registration**
+
+Previously, pandas would register converters with matplotlib as a side effect of importing pandas (:issue:`18720`).
+This changed the output of plots made via matplotlib plots after pandas was imported, even if you were using
+matplotlib directly rather than :meth:`~DataFrame.plot`.
+
+To use pandas formatters with a matplotlib plot, specify
+
+.. code-block:: ipython
+
+   In [1]: import pandas as pd
+   In [2]: pd.options.plotting.matplotlib.register_converters = True
+
+Note that plots created by :meth:`DataFrame.plot` and :meth:`Series.plot` *do* register the converters
+automatically. The only behavior change is when plotting a date-like object via ``matplotlib.pyplot.plot``
+or ``matplotlib.Axes.plot``. See :ref:`plotting.formatters` for more.
+
+**Other removals**
+
+- Removed the previously deprecated keyword "index" from :func:`read_stata`, :class:`StataReader`, and :meth:`StataReader.read`, use "index_col" instead (:issue:`17328`)
+- Removed ``StataReader.data`` method, use :meth:`StataReader.read` instead (:issue:`9493`)
+- Removed ``pandas.plotting._matplotlib.tsplot``, use :meth:`Series.plot` instead (:issue:`19980`)
+- ``pandas.tseries.converter.register`` has been moved to :func:`pandas.plotting.register_matplotlib_converters` (:issue:`18307`)
+- :meth:`Series.plot` no longer accepts positional arguments, pass keyword arguments instead (:issue:`30003`)
+- :meth:`DataFrame.hist` and :meth:`Series.hist` no longer allows ``figsize="default"``, specify figure size by passing a tuple instead (:issue:`30003`)
+- Floordiv of integer-dtyped array by :class:`Timedelta` now raises ``TypeError`` (:issue:`21036`)
+- :class:`TimedeltaIndex` and :class:`DatetimeIndex` no longer accept non-nanosecond dtype strings like "timedelta64" or "datetime64", use "timedelta64[ns]" and "datetime64[ns]" instead (:issue:`24806`)
+- Changed the default "skipna" argument in :func:`pandas.api.types.infer_dtype` from ``False`` to ``True`` (:issue:`24050`)
+- Removed ``Series.ix`` and ``DataFrame.ix`` (:issue:`26438`)
+- Removed ``Index.summary`` (:issue:`18217`)
+- Removed the previously deprecated keyword "fastpath" from the :class:`Index` constructor (:issue:`23110`)
+- Removed ``Series.get_value``, ``Series.set_value``, ``DataFrame.get_value``, ``DataFrame.set_value`` (:issue:`17739`)
+- Removed ``Series.compound`` and ``DataFrame.compound`` (:issue:`26405`)
+- Changed the default "inplace" argument in :meth:`DataFrame.set_index` and :meth:`Series.set_axis` from ``None`` to ``False`` (:issue:`27600`)
+- Removed ``Series.cat.categorical``, ``Series.cat.index``, ``Series.cat.name`` (:issue:`24751`)
+- Removed the previously deprecated keyword "box" from :func:`to_datetime` and :func:`to_timedelta`; in addition these now always returns :class:`DatetimeIndex`, :class:`TimedeltaIndex`, :class:`Index`, :class:`Series`, or :class:`DataFrame` (:issue:`24486`)
+- :func:`to_timedelta`, :class:`Timedelta`, and :class:`TimedeltaIndex` no longer allow "M", "y", or "Y" for the "unit" argument (:issue:`23264`)
+- Removed the previously deprecated keyword "time_rule" from (non-public) ``offsets.generate_range``, which has been moved to :func:`core.arrays._ranges.generate_range` (:issue:`24157`)
+- :meth:`DataFrame.loc` or :meth:`Series.loc` with listlike indexers and missing labels will no longer reindex (:issue:`17295`)
+- :meth:`DataFrame.to_excel` and :meth:`Series.to_excel` with non-existent columns will no longer reindex (:issue:`17295`)
+- Removed the previously deprecated keyword "join_axes" from :func:`concat`; use ``reindex_like`` on the result instead (:issue:`22318`)
+- Removed the previously deprecated keyword "by" from :meth:`DataFrame.sort_index`, use :meth:`DataFrame.sort_values` instead (:issue:`10726`)
+- Removed support for nested renaming in :meth:`DataFrame.aggregate`, :meth:`Series.aggregate`, :meth:`core.groupby.DataFrameGroupBy.aggregate`, :meth:`core.groupby.SeriesGroupBy.aggregate`, :meth:`core.window.rolling.Rolling.aggregate` (:issue:`18529`)
+- Passing ``datetime64`` data to :class:`TimedeltaIndex` or ``timedelta64`` data to ``DatetimeIndex`` now raises ``TypeError`` (:issue:`23539`, :issue:`23937`)
+- Passing ``int64`` values to :class:`DatetimeIndex` and a timezone now interprets the values as nanosecond timestamps in UTC, not wall times in the given timezone (:issue:`24559`)
+- A tuple passed to :meth:`DataFrame.groupby` is now exclusively treated as a single key (:issue:`18314`)
+- Removed ``Index.contains``, use ``key in index`` instead (:issue:`30103`)
+- Addition and subtraction of ``int`` or integer-arrays is no longer allowed in :class:`Timestamp`, :class:`DatetimeIndex`, :class:`TimedeltaIndex`, use ``obj + n * obj.freq`` instead of ``obj + n`` (:issue:`22535`)
+- Removed ``Series.ptp`` (:issue:`21614`)
+- Removed ``Series.from_array`` (:issue:`18258`)
+- Removed ``DataFrame.from_items`` (:issue:`18458`)
+- Removed ``DataFrame.as_matrix``, ``Series.as_matrix`` (:issue:`18458`)
+- Removed ``Series.asobject`` (:issue:`18477`)
+- Removed ``DataFrame.as_blocks``, ``Series.as_blocks``, ``DataFrame.blocks``, ``Series.blocks`` (:issue:`17656`)
+- :meth:`pandas.Series.str.cat` now defaults to aligning ``others``, using ``join='left'`` (:issue:`27611`)
+- :meth:`pandas.Series.str.cat` does not accept list-likes *within* list-likes anymore (:issue:`27611`)
+- :meth:`Series.where` with ``Categorical`` dtype (or :meth:`DataFrame.where` with ``Categorical`` column) no longer allows setting new categories (:issue:`24114`)
+- Removed the previously deprecated keywords "start", "end", and "periods" from the :class:`DatetimeIndex`, :class:`TimedeltaIndex`, and :class:`PeriodIndex` constructors; use :func:`date_range`, :func:`timedelta_range`, and :func:`period_range` instead (:issue:`23919`)
+- Removed the previously deprecated keyword "verify_integrity" from the :class:`DatetimeIndex` and :class:`TimedeltaIndex` constructors (:issue:`23919`)
+- Removed the previously deprecated keyword "fastpath" from ``pandas.core.internals.blocks.make_block`` (:issue:`19265`)
+- Removed the previously deprecated keyword "dtype" from :meth:`Block.make_block_same_class` (:issue:`19434`)
+- Removed ``ExtensionArray._formatting_values``. Use :attr:`ExtensionArray._formatter` instead. (:issue:`23601`)
+- Removed ``MultiIndex.to_hierarchical`` (:issue:`21613`)
+- Removed ``MultiIndex.labels``, use :attr:`MultiIndex.codes` instead (:issue:`23752`)
+- Removed the previously deprecated keyword "labels" from the :class:`MultiIndex` constructor, use "codes" instead (:issue:`23752`)
+- Removed ``MultiIndex.set_labels``, use :meth:`MultiIndex.set_codes` instead (:issue:`23752`)
+- Removed the previously deprecated keyword "labels" from :meth:`MultiIndex.set_codes`, :meth:`MultiIndex.copy`, :meth:`MultiIndex.drop`, use "codes" instead (:issue:`23752`)
+- Removed support for legacy HDF5 formats (:issue:`29787`)
+- Passing a dtype alias (e.g. 'datetime64[ns, UTC]') to :class:`DatetimeTZDtype` is no longer allowed, use :meth:`DatetimeTZDtype.construct_from_string` instead (:issue:`23990`)
+- Removed the previously deprecated keyword "skip_footer" from :func:`read_excel`; use "skipfooter" instead (:issue:`18836`)
+- :func:`read_excel` no longer allows an integer value for the parameter ``usecols``, instead pass a list of integers from 0 to ``usecols`` inclusive (:issue:`23635`)
+- Removed the previously deprecated keyword "convert_datetime64" from :meth:`DataFrame.to_records` (:issue:`18902`)
+- Removed ``IntervalIndex.from_intervals`` in favor of the :class:`IntervalIndex` constructor (:issue:`19263`)
+- Changed the default "keep_tz" argument in :meth:`DatetimeIndex.to_series` from ``None`` to ``True`` (:issue:`23739`)
+- Removed ``api.types.is_period`` and ``api.types.is_datetimetz`` (:issue:`23917`)
+- Ability to read pickles containing :class:`Categorical` instances created with pre-0.16 version of pandas has been removed (:issue:`27538`)
+- Removed ``pandas.tseries.plotting.tsplot`` (:issue:`18627`)
+- Removed the previously deprecated keywords "reduce" and "broadcast" from :meth:`DataFrame.apply` (:issue:`18577`)
+- Removed the previously deprecated ``assert_raises_regex`` function in ``pandas._testing`` (:issue:`29174`)
+- Removed the previously deprecated ``FrozenNDArray`` class in ``pandas.core.indexes.frozen`` (:issue:`29335`)
+- Removed the previously deprecated keyword "nthreads" from :func:`read_feather`, use "use_threads" instead (:issue:`23053`)
+- Removed ``Index.is_lexsorted_for_tuple`` (:issue:`29305`)
+- Removed support for nested renaming in :meth:`DataFrame.aggregate`, :meth:`Series.aggregate`, :meth:`core.groupby.DataFrameGroupBy.aggregate`, :meth:`core.groupby.SeriesGroupBy.aggregate`, :meth:`core.window.rolling.Rolling.aggregate` (:issue:`29608`)
+- Removed ``Series.valid``; use :meth:`Series.dropna` instead (:issue:`18800`)
+- Removed ``DataFrame.is_copy``, ``Series.is_copy`` (:issue:`18812`)
+- Removed ``DataFrame.get_ftype_counts``, ``Series.get_ftype_counts`` (:issue:`18243`)
+- Removed ``DataFrame.ftypes``, ``Series.ftypes``, ``Series.ftype`` (:issue:`26744`)
+- Removed ``Index.get_duplicates``, use ``idx[idx.duplicated()].unique()`` instead (:issue:`20239`)
+- Removed ``Series.clip_upper``, ``Series.clip_lower``, ``DataFrame.clip_upper``, ``DataFrame.clip_lower`` (:issue:`24203`)
+- Removed the ability to alter :attr:`DatetimeIndex.freq`, :attr:`TimedeltaIndex.freq`, or :attr:`PeriodIndex.freq` (:issue:`20772`)
+- Removed ``DatetimeIndex.offset`` (:issue:`20730`)
+- Removed ``DatetimeIndex.asobject``, ``TimedeltaIndex.asobject``, ``PeriodIndex.asobject``, use ``astype(object)`` instead (:issue:`29801`)
+- Removed the previously deprecated keyword "order" from :func:`factorize` (:issue:`19751`)
+- Removed the previously deprecated keyword "encoding" from :func:`read_stata` and :meth:`DataFrame.to_stata` (:issue:`21400`)
+- Changed the default "sort" argument in :func:`concat` from ``None`` to ``False`` (:issue:`20613`)
+- Removed the previously deprecated keyword "raise_conflict" from :meth:`DataFrame.update`, use "errors" instead (:issue:`23585`)
+- Removed the previously deprecated keyword "n" from :meth:`DatetimeIndex.shift`, :meth:`TimedeltaIndex.shift`, :meth:`PeriodIndex.shift`, use "periods" instead (:issue:`22458`)
+- Removed the previously deprecated keywords "how", "fill_method", and "limit" from :meth:`DataFrame.resample` (:issue:`30139`)
+- Passing an integer to :meth:`Series.fillna` or :meth:`DataFrame.fillna` with ``timedelta64[ns]`` dtype now raises ``TypeError`` (:issue:`24694`)
+- Passing multiple axes to :meth:`DataFrame.dropna` is no longer supported (:issue:`20995`)
+- Removed ``Series.nonzero``, use ``to_numpy().nonzero()`` instead (:issue:`24048`)
+- Passing floating dtype ``codes`` to :meth:`Categorical.from_codes` is no longer supported, pass ``codes.astype(np.int64)`` instead (:issue:`21775`)
+- Removed the previously deprecated keyword "pat" from :meth:`Series.str.partition` and :meth:`Series.str.rpartition`, use "sep" instead (:issue:`23767`)
+- Removed ``Series.put`` (:issue:`27106`)
+- Removed ``Series.real``, ``Series.imag`` (:issue:`27106`)
+- Removed ``Series.to_dense``, ``DataFrame.to_dense`` (:issue:`26684`)
+- Removed ``Index.dtype_str``, use ``str(index.dtype)`` instead (:issue:`27106`)
+- :meth:`Categorical.ravel` returns a :class:`Categorical` instead of a ``ndarray`` (:issue:`27199`)
+- The 'outer' method on Numpy ufuncs, e.g. ``np.subtract.outer`` operating on :class:`Series` objects is no longer supported, and will raise ``NotImplementedError`` (:issue:`27198`)
+- Removed ``Series.get_dtype_counts`` and ``DataFrame.get_dtype_counts`` (:issue:`27145`)
+- Changed the default "fill_value" argument in :meth:`Categorical.take` from ``True`` to ``False`` (:issue:`20841`)
+- Changed the default value for the ``raw`` argument in :func:`Series.rolling().apply() <.Rolling.apply>`, :func:`DataFrame.rolling().apply() <.Rolling.apply>`, :func:`Series.expanding().apply() <.Expanding.apply>`, and :func:`DataFrame.expanding().apply() <.Expanding.apply>` from ``None`` to ``False`` (:issue:`20584`)
+- Removed deprecated behavior of :meth:`Series.argmin` and :meth:`Series.argmax`, use :meth:`Series.idxmin` and :meth:`Series.idxmax` for the old behavior (:issue:`16955`)
+- Passing a tz-aware ``datetime.datetime`` or :class:`Timestamp` into the :class:`Timestamp` constructor with the ``tz`` argument now raises a ``ValueError`` (:issue:`23621`)
+- Removed ``Series.base``, ``Index.base``, ``Categorical.base``, ``Series.flags``, ``Index.flags``, ``PeriodArray.flags``, ``Series.strides``, ``Index.strides``, ``Series.itemsize``, ``Index.itemsize``, ``Series.data``, ``Index.data`` (:issue:`20721`)
+- Changed :meth:`Timedelta.resolution` to match the behavior of the standard library ``datetime.timedelta.resolution``, for the old behavior, use :meth:`Timedelta.resolution_string` (:issue:`26839`)
+- Removed ``Timestamp.weekday_name``, ``DatetimeIndex.weekday_name``, and ``Series.dt.weekday_name`` (:issue:`18164`)
+- Removed the previously deprecated keyword "errors" in :meth:`Timestamp.tz_localize`, :meth:`DatetimeIndex.tz_localize`, and :meth:`Series.tz_localize` (:issue:`22644`)
+- Changed the default "ordered" argument in :class:`CategoricalDtype` from ``None`` to ``False`` (:issue:`26336`)
+- :meth:`Series.set_axis` and :meth:`DataFrame.set_axis` now require "labels" as the first argument and "axis" as an optional named parameter (:issue:`30089`)
+- Removed ``to_msgpack``, ``read_msgpack``, ``DataFrame.to_msgpack``, ``Series.to_msgpack`` (:issue:`27103`)
+- Removed ``Series.compress`` (:issue:`21930`)
+- Removed the previously deprecated keyword "fill_value" from :meth:`Categorical.fillna`, use "value" instead (:issue:`19269`)
+- Removed the previously deprecated keyword "data" from :func:`andrews_curves`, use "frame" instead (:issue:`6956`)
+- Removed the previously deprecated keyword "data" from :func:`parallel_coordinates`, use "frame" instead (:issue:`6956`)
+- Removed the previously deprecated keyword "colors" from :func:`parallel_coordinates`, use "color" instead (:issue:`6956`)
+- Removed the previously deprecated keywords "verbose" and "private_key" from :func:`read_gbq` (:issue:`30200`)
+- Calling ``np.array`` and ``np.asarray`` on tz-aware :class:`Series` and :class:`DatetimeIndex` will now return an object array of tz-aware :class:`Timestamp` (:issue:`24596`)
+-
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Performance improvement in :class:`DataFrame` arithmetic and comparison operations with scalars (:issue:`24990`, :issue:`29853`)
+- Performance improvement in indexing with a non-unique :class:`IntervalIndex` (:issue:`27489`)
+- Performance improvement in :attr:`MultiIndex.is_monotonic` (:issue:`27495`)
+- Performance improvement in :func:`cut` when ``bins`` is an :class:`IntervalIndex` (:issue:`27668`)
+- Performance improvement when initializing a :class:`DataFrame` using a ``range`` (:issue:`30171`)
+- Performance improvement in :meth:`DataFrame.corr` when ``method`` is ``"spearman"`` (:issue:`28139`)
+- Performance improvement in :meth:`DataFrame.replace` when provided a list of values to replace (:issue:`28099`)
+- Performance improvement in :meth:`DataFrame.select_dtypes` by using vectorization instead of iterating over a loop (:issue:`28317`)
+- Performance improvement in :meth:`Categorical.searchsorted` and  :meth:`CategoricalIndex.searchsorted` (:issue:`28795`)
+- Performance improvement when comparing a :class:`Categorical` with a scalar and the scalar is not found in the categories (:issue:`29750`)
+- Performance improvement when checking if values in a :class:`Categorical` are equal, equal or larger or larger than a given scalar.
+  The improvement is not present if checking if the :class:`Categorical` is less than or less than or equal than the scalar (:issue:`29820`)
+- Performance improvement in :meth:`Index.equals` and  :meth:`MultiIndex.equals` (:issue:`29134`)
+- Performance improvement in :func:`~pandas.api.types.infer_dtype` when ``skipna`` is ``True`` (:issue:`28814`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+
+Categorical
+^^^^^^^^^^^
+
+- Added test to assert the :func:`fillna` raises the correct ``ValueError`` message when the value isn't a value from categories (:issue:`13628`)
+- Bug in :meth:`Categorical.astype` where ``NaN`` values were handled incorrectly when casting to int (:issue:`28406`)
+- :meth:`DataFrame.reindex` with a :class:`CategoricalIndex` would fail when the targets contained duplicates, and wouldn't fail if the source contained duplicates (:issue:`28107`)
+- Bug in :meth:`Categorical.astype` not allowing for casting to extension dtypes (:issue:`28668`)
+- Bug where :func:`merge` was unable to join on categorical and extension dtype columns (:issue:`28668`)
+- :meth:`Categorical.searchsorted` and :meth:`CategoricalIndex.searchsorted` now work on unordered categoricals also (:issue:`21667`)
+- Added test to assert roundtripping to parquet with :func:`DataFrame.to_parquet` or :func:`read_parquet` will preserve Categorical dtypes for string types (:issue:`27955`)
+- Changed the error message in :meth:`Categorical.remove_categories` to always show the invalid removals as a set (:issue:`28669`)
+- Using date accessors on a categorical dtyped :class:`Series` of datetimes was not returning an object of the
+  same type as if one used the :meth:`.str.` / :meth:`.dt.` on a :class:`Series` of that type. E.g. when accessing :meth:`Series.dt.tz_localize` on a
+  :class:`Categorical` with duplicate entries, the accessor was skipping duplicates (:issue:`27952`)
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` that would give incorrect results on categorical data (:issue:`26988`)
+- Bug where calling :meth:`Categorical.min` or :meth:`Categorical.max` on an empty Categorical would raise a numpy exception (:issue:`30227`)
+- The following methods now also correctly output values for unobserved categories when called through ``groupby(..., observed=False)`` (:issue:`17605`)
+  * :meth:`core.groupby.SeriesGroupBy.count`
+  * :meth:`core.groupby.SeriesGroupBy.size`
+  * :meth:`core.groupby.SeriesGroupBy.nunique`
+  * :meth:`core.groupby.SeriesGroupBy.nth`
+
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :meth:`Series.__setitem__` incorrectly casting ``np.timedelta64("NaT")`` to ``np.datetime64("NaT")`` when inserting into a :class:`Series` with datetime64 dtype (:issue:`27311`)
+- Bug in :meth:`Series.dt` property lookups when the underlying data is read-only (:issue:`27529`)
+- Bug in ``HDFStore.__getitem__`` incorrectly reading tz attribute created in Python 2 (:issue:`26443`)
+- Bug in :func:`to_datetime` where passing arrays of malformed ``str`` with errors="coerce" could incorrectly lead to raising ``ValueError`` (:issue:`28299`)
+- Bug in :meth:`core.groupby.SeriesGroupBy.nunique` where ``NaT`` values were interfering with the count of unique values (:issue:`27951`)
+- Bug in :class:`Timestamp` subtraction when subtracting a :class:`Timestamp` from a ``np.datetime64`` object incorrectly raising ``TypeError`` (:issue:`28286`)
+- Addition and subtraction of integer or integer-dtype arrays with :class:`Timestamp` will now raise ``NullFrequencyError`` instead of ``ValueError`` (:issue:`28268`)
+- Bug in :class:`Series` and :class:`DataFrame` with integer dtype failing to raise ``TypeError`` when adding or subtracting a ``np.datetime64`` object (:issue:`28080`)
+- Bug in :meth:`Series.astype`, :meth:`Index.astype`, and :meth:`DataFrame.astype` failing to handle ``NaT`` when casting to an integer dtype (:issue:`28492`)
+- Bug in :class:`Week` with ``weekday`` incorrectly raising ``AttributeError`` instead of ``TypeError`` when adding or subtracting an invalid type (:issue:`28530`)
+- Bug in :class:`DataFrame` arithmetic operations when operating with a :class:`Series` with dtype ``'timedelta64[ns]'`` (:issue:`28049`)
+- Bug in :func:`core.groupby.generic.SeriesGroupBy.apply` raising ``ValueError`` when a column in the original DataFrame is a datetime and the column labels are not standard integers (:issue:`28247`)
+- Bug in :func:`pandas._config.localization.get_locales` where the ``locales -a`` encodes the locales list as windows-1252 (:issue:`23638`, :issue:`24760`, :issue:`27368`)
+- Bug in :meth:`Series.var` failing to raise ``TypeError`` when called with ``timedelta64[ns]`` dtype (:issue:`28289`)
+- Bug in :meth:`DatetimeIndex.strftime` and :meth:`Series.dt.strftime` where ``NaT`` was converted to the string ``'NaT'`` instead of ``np.nan`` (:issue:`29578`)
+- Bug in masking datetime-like arrays with a boolean mask of an incorrect length not raising an ``IndexError`` (:issue:`30308`)
+- Bug in :attr:`Timestamp.resolution` being a property instead of a class attribute (:issue:`29910`)
+- Bug in :func:`pandas.to_datetime` when called with ``None`` raising ``TypeError`` instead of returning ``NaT`` (:issue:`30011`)
+- Bug in :func:`pandas.to_datetime` failing for ``deque`` objects when using ``cache=True`` (the default) (:issue:`29403`)
+- Bug in :meth:`Series.item` with ``datetime64`` or ``timedelta64`` dtype, :meth:`DatetimeIndex.item`, and :meth:`TimedeltaIndex.item` returning an integer instead of a :class:`Timestamp` or :class:`Timedelta` (:issue:`30175`)
+- Bug in :class:`DatetimeIndex` addition when adding a non-optimized :class:`DateOffset` incorrectly dropping timezone information (:issue:`30336`)
+- Bug in :meth:`DataFrame.drop` where attempting to drop non-existent values from a DatetimeIndex would yield a confusing error message (:issue:`30399`)
+- Bug in :meth:`DataFrame.append` would remove the timezone-awareness of new data (:issue:`30238`)
+- Bug in :meth:`Series.cummin` and :meth:`Series.cummax` with timezone-aware dtype incorrectly dropping its timezone (:issue:`15553`)
+- Bug in :class:`DatetimeArray`, :class:`TimedeltaArray`, and :class:`PeriodArray` where inplace addition and subtraction did not actually operate inplace (:issue:`24115`)
+- Bug in :func:`pandas.to_datetime` when called with ``Series`` storing ``IntegerArray`` raising ``TypeError`` instead of returning ``Series`` (:issue:`30050`)
+- Bug in :func:`date_range` with custom business hours as ``freq`` and given number of ``periods`` (:issue:`30593`)
+- Bug in :class:`PeriodIndex` comparisons with incorrectly casting integers to :class:`Period` objects, inconsistent with the :class:`Period` comparison behavior (:issue:`30722`)
+- Bug in :meth:`DatetimeIndex.insert` raising a ``ValueError`` instead of a ``TypeError`` when trying to insert a timezone-aware :class:`Timestamp` into a timezone-naive :class:`DatetimeIndex`, or vice-versa (:issue:`30806`)
+
+Timedelta
+^^^^^^^^^
+- Bug in subtracting a :class:`TimedeltaIndex` or :class:`TimedeltaArray` from a ``np.datetime64`` object (:issue:`29558`)
+-
+
+Timezones
+^^^^^^^^^
+
+-
+
+
+Numeric
+^^^^^^^
+- Bug in :meth:`DataFrame.quantile` with zero-column :class:`DataFrame` incorrectly raising (:issue:`23925`)
+- :class:`DataFrame` flex inequality comparisons methods (:meth:`DataFrame.lt`, :meth:`DataFrame.le`, :meth:`DataFrame.gt`, :meth:`DataFrame.ge`) with object-dtype and ``complex`` entries failing to raise ``TypeError`` like their :class:`Series` counterparts (:issue:`28079`)
+- Bug in :class:`DataFrame` logical operations (``&``, ``|``, ``^``) not matching :class:`Series` behavior by filling NA values (:issue:`28741`)
+- Bug in :meth:`DataFrame.interpolate` where specifying axis by name references variable before it is assigned (:issue:`29142`)
+- Bug in :meth:`Series.var` not computing the right value with a nullable integer dtype series not passing through ddof argument (:issue:`29128`)
+- Improved error message when using ``frac`` > 1 and ``replace`` = False (:issue:`27451`)
+- Bug in numeric indexes resulted in it being possible to instantiate an :class:`Int64Index`, :class:`UInt64Index`, or :class:`Float64Index` with an invalid dtype (e.g. datetime-like) (:issue:`29539`)
+- Bug in :class:`UInt64Index` precision loss while constructing from a list with values in the ``np.uint64`` range (:issue:`29526`)
+- Bug in :class:`NumericIndex` construction that caused indexing to fail when integers in the ``np.uint64`` range were used (:issue:`28023`)
+- Bug in :class:`NumericIndex` construction that caused :class:`UInt64Index` to be casted to :class:`Float64Index` when integers in the ``np.uint64`` range were used to index a :class:`DataFrame` (:issue:`28279`)
+- Bug in :meth:`Series.interpolate` when using method=`index` with an unsorted index, would previously return incorrect results. (:issue:`21037`)
+- Bug in :meth:`DataFrame.round` where a :class:`DataFrame` with a :class:`CategoricalIndex` of :class:`IntervalIndex` columns would incorrectly raise a ``TypeError`` (:issue:`30063`)
+- Bug in :meth:`Series.pct_change` and :meth:`DataFrame.pct_change` when there are duplicated indices (:issue:`30463`)
+- Bug in :class:`DataFrame` cumulative operations (e.g. cumsum, cummax) incorrect casting to object-dtype (:issue:`19296`)
+- Bug in :class:`~DataFrame.diff` losing the dtype for extension types (:issue:`30889`)
+- Bug in :class:`DataFrame.diff` raising an ``IndexError`` when one of the columns was a nullable integer dtype (:issue:`30967`)
+
+Conversion
+^^^^^^^^^^
+
+-
+
+Strings
+^^^^^^^
+
+- Calling :meth:`Series.str.isalnum` (and other "ismethods") on an empty ``Series`` would return an ``object`` dtype instead of ``bool`` (:issue:`29624`)
+-
+
+
+Interval
+^^^^^^^^
+
+- Bug in :meth:`IntervalIndex.get_indexer` where a :class:`Categorical` or :class:`CategoricalIndex` ``target`` would incorrectly raise a ``TypeError`` (:issue:`30063`)
+- Bug in ``pandas.core.dtypes.cast.infer_dtype_from_scalar`` where passing ``pandas_dtype=True`` did not infer :class:`IntervalDtype` (:issue:`30337`)
+- Bug in :class:`Series` constructor where constructing a ``Series`` from a ``list`` of :class:`Interval` objects resulted in ``object`` dtype instead of :class:`IntervalDtype` (:issue:`23563`)
+- Bug in :class:`IntervalDtype` where the ``kind`` attribute was incorrectly set as ``None`` instead of ``"O"`` (:issue:`30568`)
+- Bug in :class:`IntervalIndex`, :class:`~arrays.IntervalArray`, and :class:`Series` with interval data where equality comparisons were incorrect (:issue:`24112`)
+
+Indexing
+^^^^^^^^
+
+- Bug in assignment using a reverse slicer (:issue:`26939`)
+- Bug in :meth:`DataFrame.explode` would duplicate frame in the presence of duplicates in the index (:issue:`28010`)
+- Bug in reindexing a :meth:`PeriodIndex` with another type of index that contained a ``Period`` (:issue:`28323`) (:issue:`28337`)
+- Fix assignment of column via ``.loc`` with numpy non-ns datetime type (:issue:`27395`)
+- Bug in :meth:`Float64Index.astype` where ``np.inf`` was not handled properly when casting to an integer dtype (:issue:`28475`)
+- :meth:`Index.union` could fail when the left contained duplicates (:issue:`28257`)
+- Bug when indexing with ``.loc`` where the index was a :class:`CategoricalIndex` with non-string categories didn't work (:issue:`17569`, :issue:`30225`)
+- :meth:`Index.get_indexer_non_unique` could fail with ``TypeError`` in some cases, such as when searching for ints in a string index (:issue:`28257`)
+- Bug in :meth:`Float64Index.get_loc` incorrectly raising ``TypeError`` instead of ``KeyError`` (:issue:`29189`)
+- Bug in :meth:`DataFrame.loc` with incorrect dtype when setting Categorical value in 1-row DataFrame (:issue:`25495`)
+- :meth:`MultiIndex.get_loc` can't find missing values when input includes missing values (:issue:`19132`)
+- Bug in :meth:`Series.__setitem__` incorrectly assigning values with boolean indexer when the length of new data matches the number of ``True`` values and new data is not a ``Series`` or an ``np.array`` (:issue:`30567`)
+- Bug in indexing with a :class:`PeriodIndex` incorrectly accepting integers representing years, use e.g. ``ser.loc["2007"]`` instead of ``ser.loc[2007]`` (:issue:`30763`)
+
+Missing
+^^^^^^^
+
+-
+
+MultiIndex
+^^^^^^^^^^
+
+- Constructor for :class:`MultiIndex` verifies that the given ``sortorder`` is compatible with the actual ``lexsort_depth``  if ``verify_integrity`` parameter is ``True`` (the default) (:issue:`28735`)
+- Series and MultiIndex ``.drop`` with ``MultiIndex`` raise exception if labels not in given in level (:issue:`8594`)
+-
+
+IO
+^^
+
+- :meth:`read_csv` now accepts binary mode file buffers when using the Python csv engine (:issue:`23779`)
+- Bug in :meth:`DataFrame.to_json` where using a Tuple as a column or index value and using ``orient="columns"`` or ``orient="index"`` would produce invalid JSON (:issue:`20500`)
+- Improve infinity parsing. :meth:`read_csv` now interprets ``Infinity``, ``+Infinity``, ``-Infinity`` as floating point values (:issue:`10065`)
+- Bug in :meth:`DataFrame.to_csv` where values were truncated when the length of ``na_rep`` was shorter than the text input data. (:issue:`25099`)
+- Bug in :func:`DataFrame.to_string` where values were truncated using display options instead of outputting the full content (:issue:`9784`)
+- Bug in :meth:`DataFrame.to_json` where a datetime column label would not be written out in ISO format with ``orient="table"`` (:issue:`28130`)
+- Bug in :func:`DataFrame.to_parquet` where writing to GCS would fail with ``engine='fastparquet'`` if the file did not already exist (:issue:`28326`)
+- Bug in :func:`read_hdf` closing stores that it didn't open when Exceptions are raised (:issue:`28699`)
+- Bug in :meth:`DataFrame.read_json` where using ``orient="index"`` would not maintain the order (:issue:`28557`)
+- Bug in :meth:`DataFrame.to_html` where the length of the ``formatters`` argument was not verified (:issue:`28469`)
+- Bug in :meth:`DataFrame.read_excel` with ``engine='ods'`` when ``sheet_name`` argument references a non-existent sheet (:issue:`27676`)
+- Bug in :meth:`pandas.io.formats.style.Styler` formatting for floating values not displaying decimals correctly (:issue:`13257`)
+- Bug in :meth:`DataFrame.to_html` when using ``formatters=<list>`` and ``max_cols`` together. (:issue:`25955`)
+- Bug in :meth:`Styler.background_gradient` not able to work with dtype ``Int64`` (:issue:`28869`)
+- Bug in :meth:`DataFrame.to_clipboard` which did not work reliably in ipython (:issue:`22707`)
+- Bug in :func:`read_json` where default encoding was not set to ``utf-8`` (:issue:`29565`)
+- Bug in :class:`PythonParser` where str and bytes were being mixed when dealing with the decimal field (:issue:`29650`)
+- :meth:`read_gbq` now accepts ``progress_bar_type`` to display progress bar while the data downloads. (:issue:`29857`)
+- Bug in :func:`pandas.io.json.json_normalize` where a missing value in the location specified by ``record_path`` would raise a ``TypeError`` (:issue:`30148`)
+- :func:`read_excel` now accepts binary data (:issue:`15914`)
+- Bug in :meth:`read_csv` in which encoding handling was limited to just the string ``utf-16`` for the C engine (:issue:`24130`)
+
+Plotting
+^^^^^^^^
+
+- Bug in :meth:`Series.plot` not able to plot boolean values (:issue:`23719`)
+- Bug in :meth:`DataFrame.plot` not able to plot when no rows (:issue:`27758`)
+- Bug in :meth:`DataFrame.plot` producing incorrect legend markers when plotting multiple series on the same axis (:issue:`18222`)
+- Bug in :meth:`DataFrame.plot` when ``kind='box'`` and data contains datetime or timedelta data. These types are now automatically dropped (:issue:`22799`)
+- Bug in :meth:`DataFrame.plot.line` and :meth:`DataFrame.plot.area` produce wrong xlim in x-axis (:issue:`27686`, :issue:`25160`, :issue:`24784`)
+- Bug where :meth:`DataFrame.boxplot` would not accept a ``color`` parameter like :meth:`DataFrame.plot.box` (:issue:`26214`)
+- Bug in the ``xticks`` argument being ignored for :meth:`DataFrame.plot.bar` (:issue:`14119`)
+- :func:`set_option` now validates that the plot backend provided to ``'plotting.backend'`` implements the backend when the option is set, rather than when a plot is created (:issue:`28163`)
+- :meth:`DataFrame.plot` now allow a ``backend`` keyword argument to allow changing between backends in one session (:issue:`28619`).
+- Bug in color validation incorrectly raising for non-color styles (:issue:`29122`).
+- Allow :meth:`DataFrame.plot.scatter` to plot ``objects`` and ``datetime`` type data (:issue:`18755`, :issue:`30391`)
+- Bug in :meth:`DataFrame.hist`, ``xrot=0`` does not work with ``by`` and subplots (:issue:`30288`).
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :meth:`core.groupby.DataFrameGroupBy.apply` only showing output from a single group when function returns an :class:`Index` (:issue:`28652`)
+- Bug in :meth:`DataFrame.groupby` with multiple groups where an ``IndexError`` would be raised if any group contained all NA values (:issue:`20519`)
+- Bug in :meth:`.Resampler.size` and :meth:`.Resampler.count` returning wrong dtype when used with an empty :class:`Series` or :class:`DataFrame` (:issue:`28427`)
+- Bug in :meth:`DataFrame.rolling` not allowing for rolling over datetimes when ``axis=1`` (:issue:`28192`)
+- Bug in :meth:`DataFrame.rolling` not allowing rolling over multi-index levels (:issue:`15584`).
+- Bug in :meth:`DataFrame.rolling` not allowing rolling on monotonic decreasing time indexes (:issue:`19248`).
+- Bug in :meth:`DataFrame.groupby` not offering selection by column name when ``axis=1`` (:issue:`27614`)
+- Bug in :meth:`core.groupby.DataFrameGroupby.agg` not able to use lambda function with named aggregation (:issue:`27519`)
+- Bug in :meth:`DataFrame.groupby` losing column name information when grouping by a categorical column (:issue:`28787`)
+- Remove error raised due to duplicated input functions in named aggregation in :meth:`DataFrame.groupby` and :meth:`Series.groupby`. Previously error will be raised if the same function is applied on the same column and now it is allowed if new assigned names are different. (:issue:`28426`)
+- :meth:`core.groupby.SeriesGroupBy.value_counts` will be able to handle the case even when the :class:`Grouper` makes empty groups (:issue:`28479`)
+- Bug in :meth:`core.window.rolling.Rolling.quantile` ignoring ``interpolation`` keyword argument when used within a groupby (:issue:`28779`)
+- Bug in :meth:`DataFrame.groupby` where ``any``, ``all``, ``nunique`` and transform functions would incorrectly handle duplicate column labels (:issue:`21668`)
+- Bug in :meth:`core.groupby.DataFrameGroupBy.agg` with timezone-aware datetime64 column incorrectly casting results to the original dtype (:issue:`29641`)
+- Bug in :meth:`DataFrame.groupby` when using axis=1 and having a single level columns index (:issue:`30208`)
+- Bug in :meth:`DataFrame.groupby` when using nunique on axis=1 (:issue:`30253`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` with multiple list-like q value and integer column names (:issue:`30289`)
+- Bug in :meth:`.DataFrameGroupBy.pct_change` and :meth:`.SeriesGroupBy.pct_change` causes ``TypeError`` when ``fill_method`` is ``None`` (:issue:`30463`)
+- Bug in :meth:`Rolling.count` and :meth:`Expanding.count` argument where ``min_periods`` was ignored (:issue:`26996`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in :meth:`DataFrame.apply` that caused incorrect output with empty :class:`DataFrame` (:issue:`28202`, :issue:`21959`)
+- Bug in :meth:`DataFrame.stack` not handling non-unique indexes correctly when creating MultiIndex (:issue:`28301`)
+- Bug in :meth:`pivot_table` not returning correct type ``float`` when ``margins=True`` and ``aggfunc='mean'`` (:issue:`24893`)
+- Bug :func:`merge_asof` could not use :class:`datetime.timedelta` for ``tolerance`` kwarg (:issue:`28098`)
+- Bug in :func:`merge`, did not append suffixes correctly with MultiIndex (:issue:`28518`)
+- :func:`qcut` and :func:`cut` now handle boolean input (:issue:`20303`)
+- Fix to ensure all int dtypes can be used in :func:`merge_asof` when using a tolerance value. Previously every non-int64 type would raise an erroneous ``MergeError`` (:issue:`28870`).
+- Better error message in :func:`get_dummies` when ``columns`` isn't a list-like value (:issue:`28383`)
+- Bug in :meth:`Index.join` that caused infinite recursion error for mismatched ``MultiIndex`` name orders. (:issue:`25760`, :issue:`28956`)
+- Bug :meth:`Series.pct_change` where supplying an anchored frequency would throw a ``ValueError`` (:issue:`28664`)
+- Bug where :meth:`DataFrame.equals` returned True incorrectly in some cases when two DataFrames had the same columns in different orders (:issue:`28839`)
+- Bug in :meth:`DataFrame.replace` that caused non-numeric replacer's dtype not respected (:issue:`26632`)
+- Bug in :func:`melt` where supplying mixed strings and numeric values for ``id_vars`` or ``value_vars`` would incorrectly raise a ``ValueError`` (:issue:`29718`)
+- Dtypes are now preserved when transposing a ``DataFrame`` where each column is the same extension dtype (:issue:`30091`)
+- Bug in :func:`merge_asof` merging on a tz-aware ``left_index`` and ``right_on`` a tz-aware column (:issue:`29864`)
+- Improved error message and docstring in :func:`cut` and :func:`qcut` when ``labels=True`` (:issue:`13318`)
+- Bug in missing ``fill_na`` parameter to :meth:`DataFrame.unstack` with list of levels (:issue:`30740`)
+
+Sparse
+^^^^^^
+- Bug in :class:`SparseDataFrame` arithmetic operations incorrectly casting inputs to float (:issue:`28107`)
+- Bug in ``DataFrame.sparse`` returning a ``Series`` when there was a column named ``sparse`` rather than the accessor (:issue:`30758`)
+- Fixed :meth:`operator.xor` with a boolean-dtype ``SparseArray``. Now returns a sparse result, rather than object dtype (:issue:`31025`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+
+- Bug in :class:`arrays.PandasArray` when setting a scalar string (:issue:`28118`, :issue:`28150`).
+- Bug where nullable integers could not be compared to strings (:issue:`28930`)
+- Bug where :class:`DataFrame` constructor raised ``ValueError`` with list-like data and ``dtype`` specified (:issue:`30280`)
+
+Other
+^^^^^
+- Trying to set the ``display.precision``, ``display.max_rows`` or ``display.max_columns`` using :meth:`set_option` to anything but a ``None`` or a positive int will raise a ``ValueError`` (:issue:`23348`)
+- Using :meth:`DataFrame.replace` with overlapping keys in a nested dictionary will no longer raise, now matching the behavior of a flat dictionary (:issue:`27660`)
+- :meth:`DataFrame.to_csv` and :meth:`Series.to_csv` now support dicts as ``compression`` argument with key ``'method'`` being the compression method and others as additional compression options when the compression method is ``'zip'``. (:issue:`26023`)
+- Bug in :meth:`Series.diff` where a boolean series would incorrectly raise a ``TypeError`` (:issue:`17294`)
+- :meth:`Series.append` will no longer raise a ``TypeError`` when passed a tuple of ``Series`` (:issue:`28410`)
+- Fix corrupted error message when calling ``pandas.libs._json.encode()`` on a 0d array (:issue:`18878`)
+- Backtick quoting in :meth:`DataFrame.query` and :meth:`DataFrame.eval` can now also be used to use invalid identifiers like names that start with a digit, are python keywords, or are using single character operators. (:issue:`27017`)
+- Bug in ``pd.core.util.hashing.hash_pandas_object`` where arrays containing tuples were incorrectly treated as non-hashable (:issue:`28969`)
+- Bug in :meth:`DataFrame.append` that raised ``IndexError`` when appending with empty list (:issue:`28769`)
+- Fix :class:`AbstractHolidayCalendar` to return correct results for
+  years after 2030 (now goes up to 2200) (:issue:`27790`)
+- Fixed :class:`~arrays.IntegerArray` returning ``inf`` rather than ``NaN`` for operations dividing by ``0`` (:issue:`27398`)
+- Fixed ``pow`` operations for :class:`~arrays.IntegerArray` when the other value is ``0`` or ``1`` (:issue:`29997`)
+- Bug in :meth:`Series.count` raises if use_inf_as_na is enabled (:issue:`29478`)
+- Bug in :class:`Index` where a non-hashable name could be set without raising ``TypeError`` (:issue:`29069`)
+- Bug in :class:`DataFrame` constructor when passing a 2D ``ndarray`` and an extension dtype (:issue:`12513`)
+- Bug in :meth:`DataFrame.to_csv` when supplied a series with a ``dtype="string"`` and a ``na_rep``, the ``na_rep`` was being truncated to 2 characters. (:issue:`29975`)
+- Bug where :meth:`DataFrame.itertuples` would incorrectly determine whether or not namedtuples could be used for dataframes of 255 columns (:issue:`28282`)
+- Handle nested NumPy ``object`` arrays in :func:`testing.assert_series_equal` for ExtensionArray implementations (:issue:`30841`)
+- Bug in :class:`Index` constructor incorrectly allowing 2-dimensional input arrays (:issue:`13601`, :issue:`27125`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_100.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v0.25.3..v1.0.0
diff --git a/pandas/doc/source/whatsnew/v1.0.1.rst b/pandas/doc/source/whatsnew/v1.0.1.rst
new file mode 100644
index 0000000000000..bceac69b4914a
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.1.rst
@@ -0,0 +1,79 @@
+.. _whatsnew_101:
+
+What's new in 1.0.1 (February 5, 2020)
+--------------------------------------
+
+These are the changes in pandas 1.0.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_101.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :class:`DataFrame` setting values with a slice (e.g. ``df[-4:] = 1``) indexing by label instead of position (:issue:`31469`)
+- Fixed regression when indexing a ``Series`` or ``DataFrame`` indexed by ``DatetimeIndex`` with a slice containing a :class:`datetime.date` (:issue:`31501`)
+- Fixed regression in ``DataFrame.__setitem__`` raising an ``AttributeError`` with a :class:`MultiIndex` and a non-monotonic indexer (:issue:`31449`)
+- Fixed regression in :class:`Series` multiplication when multiplying a numeric :class:`Series` with >10000 elements with a timedelta-like scalar (:issue:`31457`)
+- Fixed regression in ``.groupby().agg()`` raising an ``AssertionError`` for some reductions like ``min`` on object-dtype columns (:issue:`31522`)
+- Fixed regression in ``.groupby()`` aggregations with categorical dtype using Cythonized reduction functions (e.g. ``first``) (:issue:`31450`)
+- Fixed regression in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` if called with a function which returned a non-pandas non-scalar object (e.g. a list or numpy array) (:issue:`31441`)
+- Fixed regression in :meth:`DataFrame.groupby` whereby taking the minimum or maximum of a column with period dtype would raise a ``TypeError``. (:issue:`31471`)
+- Fixed regression in :meth:`DataFrame.groupby` with an empty DataFrame grouping by a level of a MultiIndex (:issue:`31670`).
+- Fixed regression in :meth:`DataFrame.apply` with object dtype and non-reducing function (:issue:`31505`)
+- Fixed regression in :meth:`to_datetime` when parsing non-nanosecond resolution datetimes (:issue:`31491`)
+- Fixed regression in :meth:`~DataFrame.to_csv` where specifying an ``na_rep`` might truncate the values written (:issue:`31447`)
+- Fixed regression in :class:`Categorical` construction with ``numpy.str_`` categories (:issue:`31499`)
+- Fixed regression in :meth:`DataFrame.loc` and :meth:`DataFrame.iloc` when selecting a row containing a single ``datetime64`` or ``timedelta64`` column (:issue:`31649`)
+- Fixed regression where setting :attr:`pd.options.display.max_colwidth` was not accepting negative integer. In addition, this behavior has been deprecated in favor of using ``None`` (:issue:`31532`)
+- Fixed regression in objTOJSON.c fix return-type warning (:issue:`31463`)
+- Fixed regression in :meth:`qcut` when passed a nullable integer. (:issue:`31389`)
+- Fixed regression in assigning to a :class:`Series` using a nullable integer dtype (:issue:`31446`)
+- Fixed performance regression when indexing a ``DataFrame`` or ``Series`` with a :class:`MultiIndex` for the index using a list of labels (:issue:`31648`)
+- Fixed regression in :meth:`read_csv` used in file like object ``RawIOBase`` is not recognize ``encoding`` option (:issue:`31575`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_101.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- Support for negative integer for :attr:`pd.options.display.max_colwidth` is deprecated in favor of using ``None`` (:issue:`31532`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_101.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Datetimelike**
+
+- Fixed bug in :meth:`to_datetime` raising when ``cache=True`` and out-of-bound values are present (:issue:`31491`)
+
+**Numeric**
+
+- Bug in dtypes being lost in ``DataFrame.__invert__`` (``~`` operator) with mixed dtypes (:issue:`31183`)
+  and for extension-array backed ``Series`` and ``DataFrame`` (:issue:`23087`)
+
+**Plotting**
+
+- Plotting tz-aware timeseries no longer gives UserWarning (:issue:`31205`)
+
+**Interval**
+
+- Bug in :meth:`Series.shift` with ``interval`` dtype raising a ``TypeError`` when shifting an interval array of integers or datetimes (:issue:`34195`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_101.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.0..v1.0.1|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.0.2.rst b/pandas/doc/source/whatsnew/v1.0.2.rst
new file mode 100644
index 0000000000000..7354e2bafacc0
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.2.rst
@@ -0,0 +1,125 @@
+.. _whatsnew_102:
+
+What's new in 1.0.2 (March 12, 2020)
+------------------------------------
+
+These are the changes in pandas 1.0.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_102.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+**Groupby**
+
+- Fixed regression in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` which were failing on frames with :class:`MultiIndex` columns and a custom function (:issue:`31777`)
+- Fixed regression in ``groupby(..).rolling(..).apply()`` (``RollingGroupby``) where the ``raw`` parameter was ignored (:issue:`31754`)
+- Fixed regression in :meth:`rolling(..).corr() <.Rolling.corr>` when using a time offset (:issue:`31789`)
+- Fixed regression in :meth:`groupby(..).nunique() <.DataFrameGroupBy.nunique>` which was modifying the original values if ``NaN`` values were present (:issue:`31950`)
+- Fixed regression in ``DataFrame.groupby`` raising a ``ValueError`` from an internal operation (:issue:`31802`)
+- Fixed regression in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` calling a user-provided function an extra time on an empty input (:issue:`31760`)
+
+**I/O**
+
+- Fixed regression in :meth:`read_csv` in which the ``encoding`` option was not recognized with certain file-like objects (:issue:`31819`)
+- Fixed regression in :meth:`DataFrame.to_excel` when the ``columns`` keyword argument is passed (:issue:`31677`)
+- Fixed regression in :class:`ExcelFile` where the stream passed into the function was closed by the destructor. (:issue:`31467`)
+- Fixed regression where :func:`read_pickle` raised a ``UnicodeDecodeError`` when reading a py27 pickle with :class:`MultiIndex` column (:issue:`31988`).
+
+**Reindexing/alignment**
+
+- Fixed regression in :meth:`Series.align` when ``other`` is a :class:`DataFrame` and ``method`` is not ``None`` (:issue:`31785`)
+- Fixed regression in :meth:`DataFrame.reindex` and :meth:`Series.reindex` when reindexing with (tz-aware) index and ``method=nearest`` (:issue:`26683`)
+- Fixed regression in :meth:`DataFrame.reindex_like` on a :class:`DataFrame` subclass raised an  ``AssertionError`` (:issue:`31925`)
+- Fixed regression in :class:`DataFrame` arithmetic operations with mis-matched columns (:issue:`31623`)
+
+**Other**
+
+- Fixed regression in joining on :class:`DatetimeIndex` or :class:`TimedeltaIndex` to preserve ``freq`` in simple cases (:issue:`32166`)
+- Fixed regression in :meth:`Series.shift` with ``datetime64`` dtype when passing an integer ``fill_value`` (:issue:`32591`)
+- Fixed regression in the repr of an object-dtype :class:`Index` with bools and missing values (:issue:`32146`)
+
+
+.. ---------------------------------------------------------------------------
+
+Indexing with nullable Boolean arrays
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Previously indexing with a nullable Boolean array containing ``NA`` would raise a ``ValueError``, however this is now permitted with ``NA`` being treated as ``False``. (:issue:`31503`)
+
+.. ipython:: python
+
+    s = pd.Series([1, 2, 3, 4])
+    mask = pd.array([True, True, False, None], dtype="boolean")
+    s
+    mask
+
+*pandas 1.0.0-1.0.1*
+
+.. code-block:: python
+
+    >>> s[mask]
+    Traceback (most recent call last):
+    ...
+    ValueError: cannot mask with array containing NA / NaN values
+
+*pandas 1.0.2*
+
+.. ipython:: python
+
+    s[mask]
+
+.. _whatsnew_102.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+**Datetimelike**
+
+- Bug in :meth:`Series.astype` not copying for tz-naive and tz-aware ``datetime64`` dtype (:issue:`32490`)
+- Bug where :func:`to_datetime` would raise when passed ``pd.NA`` (:issue:`32213`)
+- Improved error message when subtracting two :class:`Timestamp` that result in an out-of-bounds :class:`Timedelta` (:issue:`31774`)
+
+**Categorical**
+
+- Fixed bug where :meth:`Categorical.from_codes` improperly raised a ``ValueError`` when passed nullable integer codes. (:issue:`31779`)
+- Fixed bug where :meth:`Categorical` constructor would raise a ``TypeError`` when given a numpy array containing ``pd.NA``. (:issue:`31927`)
+- Bug in :class:`Categorical` that would ignore or crash when calling :meth:`Series.replace` with a list-like ``to_replace`` (:issue:`31720`)
+
+**I/O**
+
+- Using ``pd.NA`` with :meth:`DataFrame.to_json` now correctly outputs a null value instead of an empty object (:issue:`31615`)
+- Bug in :meth:`pandas.json_normalize` when value in meta path is not iterable (:issue:`31507`)
+- Fixed pickling of ``pandas.NA``. Previously a new object was returned, which broke computations relying on ``NA`` being a singleton (:issue:`31847`)
+- Fixed bug in parquet roundtrip with nullable unsigned integer dtypes (:issue:`31896`).
+
+**Experimental dtypes**
+
+- Fixed bug in :meth:`DataFrame.convert_dtypes` for columns that were already using the ``"string"`` dtype (:issue:`31731`).
+- Fixed bug in :meth:`DataFrame.convert_dtypes` for series with mix of integers and strings (:issue:`32117`)
+- Fixed bug in :meth:`DataFrame.convert_dtypes` where ``BooleanDtype`` columns were converted to ``Int64`` (:issue:`32287`)
+- Fixed bug in setting values using a slice indexer with string dtype (:issue:`31772`)
+- Fixed bug where :meth:`.DataFrameGroupBy.first`, :meth:`.SeriesGroupBy.first`, :meth:`.DataFrameGroupBy.last`, and :meth:`.SeriesGroupBy.last` would raise a ``TypeError`` when groups contained ``pd.NA`` in a column of object dtype (:issue:`32123`)
+- Fixed bug where :meth:`DataFrameGroupBy.mean`, :meth:`DataFrameGroupBy.median`, :meth:`DataFrameGroupBy.var`, and :meth:`DataFrameGroupBy.std` would raise a ``TypeError`` on ``Int64`` dtype columns (:issue:`32219`)
+
+**Strings**
+
+- Using ``pd.NA`` with :meth:`Series.str.repeat` now correctly outputs a null value instead of raising error for vector inputs (:issue:`31632`)
+
+**Rolling**
+
+- Fixed rolling operations with variable window (defined by time duration) on decreasing time index (:issue:`32385`).
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_102.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.1..v1.0.2
diff --git a/pandas/doc/source/whatsnew/v1.0.3.rst b/pandas/doc/source/whatsnew/v1.0.3.rst
new file mode 100644
index 0000000000000..62e6ae5b1c5cc
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.3.rst
@@ -0,0 +1,29 @@
+
+.. _whatsnew_103:
+
+What's new in 1.0.3 (March 17, 2020)
+------------------------------------
+
+These are the changes in pandas 1.0.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_103.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in ``resample.agg`` when the underlying data is non-writeable (:issue:`31710`)
+- Fixed regression in :class:`DataFrame` exponentiation with reindexing (:issue:`32685`)
+
+.. _whatsnew_103.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.2..v1.0.3
diff --git a/pandas/doc/source/whatsnew/v1.0.4.rst b/pandas/doc/source/whatsnew/v1.0.4.rst
new file mode 100644
index 0000000000000..794990b027216
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.4.rst
@@ -0,0 +1,48 @@
+
+.. _whatsnew_104:
+
+What's new in 1.0.4 (May 28, 2020)
+------------------------------------
+
+These are the changes in pandas 1.0.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_104.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fix regression where :meth:`Series.isna` and :meth:`DataFrame.isna` would raise for categorical dtype when ``pandas.options.mode.use_inf_as_na`` was set to ``True`` (:issue:`33594`)
+- Fix regression in :meth:`.DataFrameGroupBy.first`, :meth:`.SeriesGroupBy.first`, :meth:`.DataFrameGroupBy.last`, and :meth:`.SeriesGroupBy.last` where None is not preserved in object dtype (:issue:`32800`)
+- Fix regression in DataFrame reductions using ``numeric_only=True`` and ExtensionArrays (:issue:`33256`).
+- Fix performance regression in ``memory_usage(deep=True)`` for object dtype (:issue:`33012`)
+- Fix regression where :meth:`Categorical.replace` would replace with ``NaN`` whenever the new value and replacement value were equal (:issue:`33288`)
+- Fix regression where an ordered :class:`Categorical` containing only ``NaN`` values would raise rather than returning ``NaN`` when taking the minimum or maximum  (:issue:`33450`)
+- Fix regression in :meth:`DataFrameGroupBy.agg` with dictionary input losing ``ExtensionArray`` dtypes (:issue:`32194`)
+- Fix to preserve the ability to index with the "nearest" method with xarray's CFTimeIndex, an :class:`Index` subclass (`pydata/xarray#3751 <https://github.com/pydata/xarray/issues/3751>`_, :issue:`32905`).
+- Fix regression in :meth:`DataFrame.describe` raising ``TypeError: unhashable type: 'dict'`` (:issue:`32409`)
+- Fix regression in :meth:`DataFrame.replace` casts columns to ``object`` dtype if items in ``to_replace`` not in values (:issue:`32988`)
+- Fix regression in :meth:`Series.groupby` would raise ``ValueError`` when grouping by :class:`PeriodIndex` level (:issue:`34010`)
+- Fix regression in :meth:`DataFrameGroupBy.rolling.apply` and :meth:`SeriesGroupBy.rolling.apply` ignoring args and kwargs parameters (:issue:`33433`)
+- Fix regression in error message with ``np.min`` or ``np.max`` on unordered :class:`Categorical` (:issue:`33115`)
+- Fix regression in :meth:`DataFrame.loc` and :meth:`Series.loc` throwing an error when a ``datetime64[ns, tz]`` value is provided (:issue:`32395`)
+
+.. _whatsnew_104.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :meth:`SeriesGroupBy.first`, :meth:`SeriesGroupBy.last`, :meth:`SeriesGroupBy.min`, and :meth:`SeriesGroupBy.max` returning floats when applied to nullable Booleans (:issue:`33071`)
+- Bug in :meth:`Rolling.min` and :meth:`Rolling.max`: Growing memory usage after multiple calls when using a fixed window (:issue:`30726`)
+- Bug in :meth:`~DataFrame.to_parquet` was not raising ``PermissionError`` when writing to a private s3 bucket with invalid creds. (:issue:`27679`)
+- Bug in :meth:`~DataFrame.to_csv` was silently failing when writing to an invalid s3 bucket. (:issue:`32486`)
+- Bug in :meth:`read_parquet` was raising a ``FileNotFoundError`` when passed an s3 directory path. (:issue:`26388`)
+- Bug in :meth:`~DataFrame.to_parquet` was throwing an ``AttributeError`` when writing a partitioned parquet file to s3 (:issue:`27596`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` causes the quantiles to be shifted when the ``by`` axis contains ``NaN`` (:issue:`33200`, :issue:`33569`)
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.3..v1.0.4
diff --git a/pandas/doc/source/whatsnew/v1.0.5.rst b/pandas/doc/source/whatsnew/v1.0.5.rst
new file mode 100644
index 0000000000000..9a5128a07bbfd
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.0.5.rst
@@ -0,0 +1,39 @@
+
+.. _whatsnew_105:
+
+What's new in 1.0.5 (June 17, 2020)
+-----------------------------------
+
+These are the changes in pandas 1.0.5. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_105.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fix regression in :meth:`read_parquet` when reading from file-like objects
+  (:issue:`34467`).
+- Fix regression in reading from public S3 buckets (:issue:`34626`).
+
+Note this disables the ability to read Parquet files from directories on S3
+again (:issue:`26388`, :issue:`34632`), which was added in the 1.0.4 release,
+but is now targeted for pandas 1.1.0.
+
+- Fixed regression in :meth:`~DataFrame.replace` raising an ``AssertionError`` when replacing values in an extension dtype with values of a different dtype (:issue:`34530`)
+
+.. _whatsnew_105.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Fixed building from source with Python 3.8 fetching the wrong version of NumPy (:issue:`34666`)
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.4..v1.0.5|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.1.0.rst b/pandas/doc/source/whatsnew/v1.1.0.rst
new file mode 100644
index 0000000000000..b199b113d26f2
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.0.rst
@@ -0,0 +1,1231 @@
+.. _whatsnew_110:
+
+What's new in 1.1.0 (July 28, 2020)
+-----------------------------------
+
+These are the changes in pandas 1.1.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_110.specify_missing_labels:
+
+KeyErrors raised by loc specify missing labels
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Previously, if labels were missing for a ``.loc`` call, a KeyError was raised stating that this was no longer supported.
+
+Now the error message also includes a list of the missing labels (max 10 items, display width 80 characters). See :issue:`34272`.
+
+
+.. _whatsnew_110.astype_string:
+
+All dtypes can now be converted to ``StringDtype``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, declaring or converting to :class:`StringDtype` was in general only possible if the data was already only ``str`` or nan-like (:issue:`31204`).
+:class:`StringDtype` now works in all situations where ``astype(str)`` or ``dtype=str`` work:
+
+For example, the below now works:
+
+.. ipython:: python
+
+   ser = pd.Series([1, "abc", np.nan], dtype="string")
+   ser
+   ser[0]
+   pd.Series([1, 2, np.nan], dtype="Int64").astype("string")
+
+
+.. _whatsnew_110.period_index_partial_string_slicing:
+
+Non-monotonic PeriodIndex partial string slicing
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`PeriodIndex` now supports partial string slicing for non-monotonic indexes, mirroring :class:`DatetimeIndex` behavior (:issue:`31096`)
+
+For example:
+
+.. ipython:: python
+
+   dti = pd.date_range("2014-01-01", periods=30, freq="30D")
+   pi = dti.to_period("D")
+   ser_monotonic = pd.Series(np.arange(30), index=pi)
+   shuffler = list(range(0, 30, 2)) + list(range(1, 31, 2))
+   ser = ser_monotonic.iloc[shuffler]
+   ser
+
+.. ipython:: python
+
+   ser["2014"]
+   ser.loc["May 2015"]
+
+
+.. _whatsnew_110.dataframe_or_series_comparing:
+
+Comparing two ``DataFrame`` or two ``Series`` and summarizing the differences
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added :meth:`DataFrame.compare` and :meth:`Series.compare` for comparing two ``DataFrame`` or two ``Series`` (:issue:`30429`)
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "col1": ["a", "a", "b", "b", "a"],
+           "col2": [1.0, 2.0, 3.0, np.nan, 5.0],
+           "col3": [1.0, 2.0, 3.0, 4.0, 5.0]
+       },
+       columns=["col1", "col2", "col3"],
+   )
+   df
+
+.. ipython:: python
+
+   df2 = df.copy()
+   df2.loc[0, 'col1'] = 'c'
+   df2.loc[2, 'col3'] = 4.0
+   df2
+
+.. ipython:: python
+
+   df.compare(df2)
+
+See :ref:`User Guide <merging.compare>` for more details.
+
+
+.. _whatsnew_110.groupby_key:
+
+Allow NA in groupby key
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+With :ref:`groupby <groupby.dropna>` , we've added a ``dropna`` keyword to :meth:`DataFrame.groupby` and :meth:`Series.groupby` in order to
+allow ``NA`` values in group keys. Users can define ``dropna`` to ``False`` if they want to include
+``NA`` values in groupby keys. The default is set to ``True`` for ``dropna`` to keep backwards
+compatibility (:issue:`3729`)
+
+.. ipython:: python
+
+    df_list = [[1, 2, 3], [1, None, 4], [2, 1, 3], [1, 2, 2]]
+    df_dropna = pd.DataFrame(df_list, columns=["a", "b", "c"])
+
+    df_dropna
+
+.. ipython:: python
+
+    # Default ``dropna`` is set to True, which will exclude NaNs in keys
+    df_dropna.groupby(by=["b"], dropna=True).sum()
+
+    # In order to allow NaN in keys, set ``dropna`` to False
+    df_dropna.groupby(by=["b"], dropna=False).sum()
+
+The default setting of ``dropna`` argument is ``True`` which means ``NA`` are not included in group keys.
+
+
+.. _whatsnew_110.key_sorting:
+
+Sorting with keys
+^^^^^^^^^^^^^^^^^
+
+We've added a ``key`` argument to the :class:`DataFrame` and :class:`Series` sorting methods, including
+:meth:`DataFrame.sort_values`, :meth:`DataFrame.sort_index`, :meth:`Series.sort_values`,
+and :meth:`Series.sort_index`. The ``key`` can be any callable function which is applied
+column-by-column to each column used for sorting, before sorting is performed (:issue:`27237`).
+See :ref:`sort_values with keys <basics.sort_value_key>` and :ref:`sort_index with keys
+<basics.sort_index_key>` for more information.
+
+.. ipython:: python
+
+   s = pd.Series(['C', 'a', 'B'])
+   s
+
+.. ipython:: python
+
+   s.sort_values()
+
+
+Note how this is sorted with capital letters first. If we apply the :meth:`Series.str.lower`
+method, we get
+
+.. ipython:: python
+
+   s.sort_values(key=lambda x: x.str.lower())
+
+
+When applied to a ``DataFrame``, they key is applied per-column to all columns or a subset if
+``by`` is specified, e.g.
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': ['C', 'C', 'a', 'a', 'B', 'B'],
+                      'b': [1, 2, 3, 4, 5, 6]})
+   df
+
+.. ipython:: python
+
+   df.sort_values(by=['a'], key=lambda col: col.str.lower())
+
+
+For more details, see examples and documentation in :meth:`DataFrame.sort_values`,
+:meth:`Series.sort_values`, and :meth:`~DataFrame.sort_index`.
+
+.. _whatsnew_110.timestamp_fold_support:
+
+Fold argument support in Timestamp constructor
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Timestamp:` now supports the keyword-only fold argument according to `PEP 495 <https://www.python.org/dev/peps/pep-0495/#the-fold-attribute>`_ similar to parent ``datetime.datetime`` class. It supports both accepting fold as an initialization argument and inferring fold from other constructor arguments (:issue:`25057`, :issue:`31338`). Support is limited to ``dateutil`` timezones as ``pytz`` doesn't support fold.
+
+For example:
+
+.. ipython:: python
+
+    ts = pd.Timestamp("2019-10-27 01:30:00+00:00")
+    ts.fold
+
+.. ipython:: python
+
+    ts = pd.Timestamp(year=2019, month=10, day=27, hour=1, minute=30,
+                      tz="dateutil/Europe/London", fold=1)
+    ts
+
+For more on working with fold, see :ref:`Fold subsection <timeseries.fold>` in the user guide.
+
+.. _whatsnew_110.to_datetime_multiple_tzname_tzoffset_support:
+
+Parsing timezone-aware format with different timezones in to_datetime
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`to_datetime` now supports parsing formats containing timezone names (``%Z``) and UTC offsets (``%z``) from different timezones then converting them to UTC by setting ``utc=True``. This would return a :class:`DatetimeIndex` with timezone at UTC as opposed to an :class:`Index` with ``object`` dtype if ``utc=True`` is not set (:issue:`32792`).
+
+For example:
+
+.. ipython:: python
+
+    tz_strs = ["2010-01-01 12:00:00 +0100", "2010-01-01 12:00:00 -0100",
+               "2010-01-01 12:00:00 +0300", "2010-01-01 12:00:00 +0400"]
+    pd.to_datetime(tz_strs, format='%Y-%m-%d %H:%M:%S %z', utc=True)
+
+.. code-block:: ipython
+
+   In[37]: pd.to_datetime(tz_strs, format='%Y-%m-%d %H:%M:%S %z')
+   Out[37]:
+   Index([2010-01-01 12:00:00+01:00, 2010-01-01 12:00:00-01:00,
+          2010-01-01 12:00:00+03:00, 2010-01-01 12:00:00+04:00],
+         dtype='object')
+
+.. _whatsnew_110.grouper_resample_origin:
+
+Grouper and resample now supports the arguments origin and offset
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Grouper` and :meth:`DataFrame.resample` now supports the arguments ``origin`` and ``offset``. It let the user control the timestamp on which to adjust the grouping. (:issue:`31809`)
+
+The bins of the grouping are adjusted based on the beginning of the day of the time series starting point. This works well with frequencies that are multiples of a day (like ``30D``) or that divides a day (like ``90s`` or ``1min``). But it can create inconsistencies with some frequencies that do not meet this criteria. To change this behavior you can now specify a fixed timestamp with the argument ``origin``.
+
+Two arguments are now deprecated (more information in the documentation of :meth:`DataFrame.resample`):
+
+- ``base`` should be replaced by ``offset``.
+- ``loffset`` should be replaced by directly adding an offset to the index :class:`DataFrame` after being resampled.
+
+Small example of the use of ``origin``:
+
+.. ipython:: python
+
+    start, end = '2000-10-01 23:30:00', '2000-10-02 00:30:00'
+    middle = '2000-10-02 00:00:00'
+    rng = pd.date_range(start, end, freq='7min')
+    ts = pd.Series(np.arange(len(rng)) * 3, index=rng)
+    ts
+
+Resample with the default behavior ``'start_day'`` (origin is ``2000-10-01 00:00:00``):
+
+.. ipython:: python
+
+    ts.resample('17min').sum()
+    ts.resample('17min', origin='start_day').sum()
+
+Resample using a fixed origin:
+
+.. ipython:: python
+
+    ts.resample('17min', origin='epoch').sum()
+    ts.resample('17min', origin='2000-01-01').sum()
+
+If needed you can adjust the bins with the argument ``offset`` (a :class:`Timedelta`) that would be added to the default ``origin``.
+
+For a full example, see: :ref:`timeseries.adjust-the-start-of-the-bins`.
+
+fsspec now used for filesystem handling
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+For reading and writing to filesystems other than local and reading from HTTP(S),
+the optional dependency ``fsspec`` will be used to dispatch operations (:issue:`33452`).
+This will give unchanged
+functionality for S3 and GCS storage, which were already supported, but also add
+support for several other storage implementations such as `Azure Datalake and Blob`_,
+SSH, FTP, dropbox and github. For docs and capabilities, see the `fsspec docs`_.
+
+The existing capability to interface with S3 and GCS will be unaffected by this
+change, as ``fsspec`` will still bring in the same packages as before.
+
+.. _Azure Datalake and Blob: https://github.com/fsspec/adlfs
+
+.. _fsspec docs: https://filesystem-spec.readthedocs.io/en/latest/
+
+.. _whatsnew_110.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- Compatibility with matplotlib 3.3.0 (:issue:`34850`)
+- :meth:`IntegerArray.astype` now supports ``datetime64`` dtype (:issue:`32538`)
+- :class:`IntegerArray` now implements the ``sum`` operation (:issue:`33172`)
+- Added :class:`pandas.errors.InvalidIndexError` (:issue:`34570`).
+- Added :meth:`DataFrame.value_counts` (:issue:`5377`)
+- Added a :func:`pandas.api.indexers.FixedForwardWindowIndexer` class to support forward-looking windows during ``rolling`` operations.
+- Added a :func:`pandas.api.indexers.VariableOffsetWindowIndexer` class to support ``rolling`` operations with non-fixed offsets (:issue:`34994`)
+- :meth:`~DataFrame.describe` now includes a ``datetime_is_numeric`` keyword to control how datetime columns are summarized (:issue:`30164`, :issue:`34798`)
+- :class:`~pandas.io.formats.style.Styler` may now render CSS more efficiently where multiple cells have the same styling (:issue:`30876`)
+- :meth:`~pandas.io.formats.style.Styler.highlight_null` now accepts ``subset`` argument (:issue:`31345`)
+- When writing directly to a sqlite connection :meth:`DataFrame.to_sql` now supports the ``multi`` method (:issue:`29921`)
+- :class:`pandas.errors.OptionError` is now exposed in ``pandas.errors`` (:issue:`27553`)
+- Added :meth:`api.extensions.ExtensionArray.argmax` and :meth:`api.extensions.ExtensionArray.argmin` (:issue:`24382`)
+- :func:`timedelta_range` will now infer a frequency when passed ``start``, ``stop``, and ``periods`` (:issue:`32377`)
+- Positional slicing on a :class:`IntervalIndex` now supports slices with ``step > 1`` (:issue:`31658`)
+- :class:`Series.str` now has a ``fullmatch`` method that matches a regular expression against the entire string in each row of the :class:`Series`, similar to ``re.fullmatch`` (:issue:`32806`).
+- :meth:`DataFrame.sample` will now also allow array-like and BitGenerator objects to be passed to ``random_state`` as seeds (:issue:`32503`)
+- :meth:`Index.union` will now raise ``RuntimeWarning`` for :class:`MultiIndex` objects if the object inside are unsortable. Pass ``sort=False`` to suppress this warning (:issue:`33015`)
+- Added :meth:`Series.dt.isocalendar` and :meth:`DatetimeIndex.isocalendar` that returns a :class:`DataFrame` with year, week, and day calculated according to the ISO 8601 calendar (:issue:`33206`, :issue:`34392`).
+- The :meth:`DataFrame.to_feather` method now supports additional keyword
+  arguments (e.g. to set the compression) that are added in pyarrow 0.17
+  (:issue:`33422`).
+- The :func:`cut` will now accept parameter ``ordered`` with default ``ordered=True``. If ``ordered=False`` and no labels are provided, an error will be raised (:issue:`33141`)
+- :meth:`DataFrame.to_csv`, :meth:`DataFrame.to_pickle`,
+  and :meth:`DataFrame.to_json` now support passing a dict of
+  compression arguments when using the ``gzip`` and ``bz2`` protocols.
+  This can be used to set a custom compression level, e.g.,
+  ``df.to_csv(path, compression={'method': 'gzip', 'compresslevel': 1}``
+  (:issue:`33196`)
+- :meth:`melt` has gained an ``ignore_index`` (default ``True``) argument that, if set to ``False``, prevents the method from dropping the index (:issue:`17440`).
+- :meth:`Series.update` now accepts objects that can be coerced to a :class:`Series`,
+  such as ``dict`` and ``list``, mirroring the behavior of :meth:`DataFrame.update` (:issue:`33215`)
+- :meth:`.DataFrameGroupBy.transform` and :meth:`.DataFrameGroupBy.aggregate` have gained ``engine`` and ``engine_kwargs`` arguments that support executing functions with ``Numba`` (:issue:`32854`, :issue:`33388`)
+- :meth:`.Resampler.interpolate` now supports SciPy interpolation method :class:`scipy.interpolate.CubicSpline` as method ``cubicspline`` (:issue:`33670`)
+- :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` now implement the ``sample`` method for doing random sampling within groups (:issue:`31775`)
+- :meth:`DataFrame.to_numpy` now supports the ``na_value`` keyword to control the NA sentinel in the output array (:issue:`33820`)
+- Added :class:`api.extension.ExtensionArray.equals` to the extension array interface, similar to :meth:`Series.equals` (:issue:`27081`)
+- The minimum supported dta version has increased to 105 in :func:`read_stata` and :class:`~pandas.io.stata.StataReader`  (:issue:`26667`).
+- :meth:`~DataFrame.to_stata` supports compression using the ``compression``
+  keyword argument. Compression can either be inferred or explicitly set using a string or a
+  dictionary containing both the method and any additional arguments that are passed to the
+  compression library. Compression was also added to the low-level Stata-file writers
+  :class:`~pandas.io.stata.StataWriter`, :class:`~pandas.io.stata.StataWriter117`,
+  and :class:`~pandas.io.stata.StataWriterUTF8` (:issue:`26599`).
+- :meth:`HDFStore.put` now accepts a ``track_times`` parameter. This parameter is passed to the ``create_table`` method of ``PyTables`` (:issue:`32682`).
+- :meth:`Series.plot` and :meth:`DataFrame.plot` now accepts ``xlabel`` and ``ylabel`` parameters to present labels on x and y axis (:issue:`9093`).
+- Made :class:`.Rolling` and :class:`.Expanding` iterable（:issue:`11704`)
+- Made ``option_context`` a :class:`contextlib.ContextDecorator`, which allows it to be used as a decorator over an entire function (:issue:`34253`).
+- :meth:`DataFrame.to_csv` and :meth:`Series.to_csv` now accept an ``errors`` argument (:issue:`22610`)
+- :meth:`.DataFrameGroupBy.groupby.transform` now allows ``func`` to be ``pad``, ``backfill`` and ``cumcount`` (:issue:`31269`).
+- :func:`read_json` now accepts an ``nrows`` parameter. (:issue:`33916`).
+- :meth:`DataFrame.hist`, :meth:`Series.hist`, :meth:`core.groupby.DataFrameGroupBy.hist`, and :meth:`core.groupby.SeriesGroupBy.hist` have gained the ``legend`` argument. Set to True to show a legend in the histogram. (:issue:`6279`)
+- :func:`concat` and :meth:`~DataFrame.append` now preserve extension dtypes, for example
+  combining a nullable integer column with a numpy integer column will no longer
+  result in object dtype but preserve the integer dtype (:issue:`33607`, :issue:`34339`, :issue:`34095`).
+- :func:`read_gbq` now allows to disable progress bar (:issue:`33360`).
+- :func:`read_gbq` now supports the ``max_results`` kwarg from ``pandas-gbq`` (:issue:`34639`).
+- :meth:`DataFrame.cov` and :meth:`Series.cov` now support a new parameter ``ddof`` to support delta degrees of freedom as in the corresponding numpy methods (:issue:`34611`).
+- :meth:`DataFrame.to_html` and :meth:`DataFrame.to_string`'s ``col_space`` parameter now accepts a list or dict to change only some specific columns' width (:issue:`28917`).
+- :meth:`DataFrame.to_excel` can now also write OpenOffice spreadsheet (.ods) files (:issue:`27222`)
+- :meth:`~Series.explode` now accepts ``ignore_index`` to reset the index, similar to :meth:`pd.concat` or :meth:`DataFrame.sort_values` (:issue:`34932`).
+- :meth:`DataFrame.to_markdown` and :meth:`Series.to_markdown` now accept ``index`` argument as an alias for tabulate's ``showindex`` (:issue:`32667`)
+- :meth:`read_csv` now accepts string values like "0", "0.0", "1", "1.0" as convertible to the nullable Boolean dtype (:issue:`34859`)
+- :class:`.ExponentialMovingWindow` now supports a ``times`` argument that allows ``mean`` to be calculated with observations spaced by the timestamps in ``times`` (:issue:`34839`)
+- :meth:`DataFrame.agg` and :meth:`Series.agg` now accept named aggregation for renaming the output columns/indexes. (:issue:`26513`)
+- ``compute.use_numba`` now exists as a configuration option that utilizes the numba engine when available (:issue:`33966`, :issue:`35374`)
+- :meth:`Series.plot` now supports asymmetric error bars. Previously, if :meth:`Series.plot` received a "2xN" array with error values for ``yerr`` and/or ``xerr``, the left/lower values (first row) were mirrored, while the right/upper values (second row) were ignored. Now, the first row represents the left/lower error values and the second row the right/upper error values. (:issue:`9536`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_110.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+``MultiIndex.get_indexer`` interprets ``method`` argument correctly
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+This restores the behavior of :meth:`MultiIndex.get_indexer` with ``method='backfill'`` or ``method='pad'`` to the behavior before pandas 0.23.0. In particular, MultiIndexes are treated as a list of tuples and padding or backfilling is done with respect to the ordering of these lists of tuples (:issue:`29896`).
+
+As an example of this, given:
+
+.. ipython:: python
+
+        df = pd.DataFrame({
+            'a': [0, 0, 0, 0],
+            'b': [0, 2, 3, 4],
+            'c': ['A', 'B', 'C', 'D'],
+        }).set_index(['a', 'b'])
+        mi_2 = pd.MultiIndex.from_product([[0], [-1, 0, 1, 3, 4, 5]])
+
+The differences in reindexing ``df`` with ``mi_2`` and using ``method='backfill'`` can be seen here:
+
+*pandas >= 0.23, < 1.1.0*:
+
+.. code-block:: ipython
+
+    In [1]: df.reindex(mi_2, method='backfill')
+    Out[1]:
+          c
+    0 -1  A
+       0  A
+       1  D
+       3  A
+       4  A
+       5  C
+
+*pandas <0.23, >= 1.1.0*
+
+.. ipython:: python
+
+        df.reindex(mi_2, method='backfill')
+
+And the differences in reindexing ``df`` with ``mi_2`` and using ``method='pad'`` can be seen here:
+
+*pandas >= 0.23, < 1.1.0*
+
+.. code-block:: ipython
+
+    In [1]: df.reindex(mi_2, method='pad')
+    Out[1]:
+            c
+    0 -1  NaN
+       0  NaN
+       1    D
+       3  NaN
+       4    A
+       5    C
+
+*pandas < 0.23, >= 1.1.0*
+
+.. ipython:: python
+
+        df.reindex(mi_2, method='pad')
+
+.. _whatsnew_110.notable_bug_fixes.indexing_raises_key_errors:
+
+Failed label-based lookups always raise KeyError
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Label lookups ``series[key]``, ``series.loc[key]`` and ``frame.loc[key]``
+used to raise either ``KeyError`` or ``TypeError`` depending on the type of
+key and type of :class:`Index`.  These now consistently raise ``KeyError`` (:issue:`31867`)
+
+.. ipython:: python
+
+    ser1 = pd.Series(range(3), index=[0, 1, 2])
+    ser2 = pd.Series(range(3), index=pd.date_range("2020-02-01", periods=3))
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: ser1[1.5]
+    ...
+    TypeError: cannot do label indexing on Int64Index with these indexers [1.5] of type float
+
+    In [4] ser1["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [5]: ser1.loc[1.5]
+    ...
+    TypeError: cannot do label indexing on Int64Index with these indexers [1.5] of type float
+
+    In [6]: ser1.loc["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [7]: ser2.loc[1]
+    ...
+    TypeError: cannot do label indexing on DatetimeIndex with these indexers [1] of type int
+
+    In [8]: ser2.loc[pd.Timestamp(0)]
+    ...
+    KeyError: Timestamp('1970-01-01 00:00:00')
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [3]: ser1[1.5]
+    ...
+    KeyError: 1.5
+
+    In [4] ser1["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [5]: ser1.loc[1.5]
+    ...
+    KeyError: 1.5
+
+    In [6]: ser1.loc["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [7]: ser2.loc[1]
+    ...
+    KeyError: 1
+
+    In [8]: ser2.loc[pd.Timestamp(0)]
+    ...
+    KeyError: Timestamp('1970-01-01 00:00:00')
+
+
+Similarly, :meth:`DataFrame.at` and :meth:`Series.at` will raise a ``TypeError`` instead of a ``ValueError`` if an incompatible key is passed, and ``KeyError`` if a missing key is passed, matching the behavior of ``.loc[]`` (:issue:`31722`)
+
+.. _whatsnew_110.notable_bug_fixes.indexing_int_multiindex_raises_key_errors:
+
+Failed Integer Lookups on MultiIndex Raise KeyError
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Indexing with integers with a :class:`MultiIndex` that has an integer-dtype
+first level incorrectly failed to raise ``KeyError`` when one or more of
+those integer keys is not present in the first level of the index (:issue:`33539`)
+
+.. ipython:: python
+
+    idx = pd.Index(range(4))
+    dti = pd.date_range("2000-01-03", periods=3)
+    mi = pd.MultiIndex.from_product([idx, dti])
+    ser = pd.Series(range(len(mi)), index=mi)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: ser[[5]]
+    Out[5]: Series([], dtype: int64)
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [5]: ser[[5]]
+    ...
+    KeyError: '[5] not in index'
+
+:meth:`DataFrame.merge` preserves right frame's row order
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+:meth:`DataFrame.merge` now preserves the right frame's row order when executing a right merge (:issue:`27453`)
+
+.. ipython:: python
+
+    left_df = pd.DataFrame({'animal': ['dog', 'pig'],
+                           'max_speed': [40, 11]})
+    right_df = pd.DataFrame({'animal': ['quetzal', 'pig'],
+                            'max_speed': [80, 11]})
+    left_df
+    right_df
+
+*Previous behavior*:
+
+.. code-block:: python
+
+    >>> left_df.merge(right_df, on=['animal', 'max_speed'], how="right")
+        animal  max_speed
+    0      pig         11
+    1  quetzal         80
+
+*New behavior*:
+
+.. ipython:: python
+
+    left_df.merge(right_df, on=['animal', 'max_speed'], how="right")
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_110.notable_bug_fixes.assignment_to_multiple_columns:
+
+Assignment to multiple columns of a DataFrame when some columns do not exist
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Assignment to multiple columns of a :class:`DataFrame` when some of the columns do not exist would previously assign the values to the last column. Now, new columns will be constructed with the right values. (:issue:`13658`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({'a': [0, 1, 2], 'b': [3, 4, 5]})
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [3]: df[['a', 'c']] = 1
+   In [4]: df
+   Out[4]:
+      a  b
+   0  1  1
+   1  1  1
+   2  1  1
+
+*New behavior*:
+
+.. ipython:: python
+
+   df[['a', 'c']] = 1
+   df
+
+.. _whatsnew_110.notable_bug_fixes.groupby_consistency:
+
+Consistency across groupby reductions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Using :meth:`DataFrame.groupby` with ``as_index=True`` and the aggregation ``nunique`` would include the grouping column(s) in the columns of the result. Now the grouping column(s) only appear in the index, consistent with other reductions. (:issue:`32579`)
+
+.. ipython:: python
+
+   df = pd.DataFrame({"a": ["x", "x", "y", "y"], "b": [1, 1, 2, 3]})
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [3]: df.groupby("a", as_index=True).nunique()
+   Out[4]:
+      a  b
+   a
+   x  1  1
+   y  1  2
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.groupby("a", as_index=True).nunique()
+
+Using :meth:`DataFrame.groupby` with ``as_index=False`` and the function ``idxmax``, ``idxmin``, ``mad``, ``nunique``, ``sem``, ``skew``, or ``std`` would modify the grouping column. Now the grouping column remains unchanged, consistent with other reductions. (:issue:`21090`, :issue:`10355`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [3]: df.groupby("a", as_index=False).nunique()
+   Out[4]:
+      a  b
+   0  1  1
+   1  1  2
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.groupby("a", as_index=False).nunique()
+
+The method :meth:`.DataFrameGroupBy.size` would previously ignore ``as_index=False``. Now the grouping columns are returned as columns, making the result a :class:`DataFrame` instead of a :class:`Series`. (:issue:`32599`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [3]: df.groupby("a", as_index=False).size()
+   Out[4]:
+   a
+   x    2
+   y    2
+   dtype: int64
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.groupby("a", as_index=False).size()
+
+.. _whatsnew_110.api_breaking.groupby_results_lost_as_index_false:
+
+:meth:`.DataFrameGroupby.agg` lost results with ``as_index=False`` when relabeling columns
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously :meth:`.DataFrameGroupby.agg` lost the result columns, when the ``as_index`` option was
+set to ``False`` and the result columns were relabeled. In this case the result values were replaced with
+the previous index (:issue:`32240`).
+
+.. ipython:: python
+
+   df = pd.DataFrame({"key": ["x", "y", "z", "x", "y", "z"],
+                      "val": [1.0, 0.8, 2.0, 3.0, 3.6, 0.75]})
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [2]: grouped = df.groupby("key", as_index=False)
+   In [3]: result = grouped.agg(min_val=pd.NamedAgg(column="val", aggfunc="min"))
+   In [4]: result
+   Out[4]:
+        min_val
+    0   x
+    1   y
+    2   z
+
+*New behavior*:
+
+.. ipython:: python
+
+   grouped = df.groupby("key", as_index=False)
+   result = grouped.agg(min_val=pd.NamedAgg(column="val", aggfunc="min"))
+   result
+
+
+.. _whatsnew_110.notable_bug_fixes.apply_applymap_first_once:
+
+apply and applymap on ``DataFrame`` evaluates first row/column only once
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. ipython:: python
+
+    df = pd.DataFrame({'a': [1, 2], 'b': [3, 6]})
+
+    def func(row):
+        print(row)
+        return row
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [4]: df.apply(func, axis=1)
+    a    1
+    b    3
+    Name: 0, dtype: int64
+    a    1
+    b    3
+    Name: 0, dtype: int64
+    a    2
+    b    6
+    Name: 1, dtype: int64
+    Out[4]:
+       a  b
+    0  1  3
+    1  2  6
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.apply(func, axis=1)
+
+.. _whatsnew_110.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_110.api_breaking.testing.check_freq:
+
+Added ``check_freq`` argument to ``testing.assert_frame_equal`` and ``testing.assert_series_equal``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``check_freq`` argument was added to :func:`testing.assert_frame_equal` and :func:`testing.assert_series_equal` in pandas 1.1.0 and defaults to ``True``. :func:`testing.assert_frame_equal` and :func:`testing.assert_series_equal` now raise ``AssertionError`` if the indexes do not have the same frequency. Before pandas 1.1.0, the index frequency was not checked.
+
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Some minimum supported versions of dependencies were updated (:issue:`33718`, :issue:`29766`, :issue:`29723`, pytables >= 3.4.3).
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.15.4          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| pytz            | 2015.4          |    X     |         |
++-----------------+-----------------+----------+---------+
+| python-dateutil | 2.7.3           |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.2.1           |          |         |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.6.2           |          |         |
++-----------------+-----------------+----------+---------+
+| pytest (dev)    | 4.0.2           |          |         |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  | 4.6.0           |         |
++-----------------+-----------------+---------+
+| fastparquet     | 0.3.2           |         |
++-----------------+-----------------+---------+
+| fsspec          | 0.7.4           |         |
++-----------------+-----------------+---------+
+| gcsfs           | 0.6.0           |    X    |
++-----------------+-----------------+---------+
+| lxml            | 3.8.0           |         |
++-----------------+-----------------+---------+
+| matplotlib      | 2.2.2           |         |
++-----------------+-----------------+---------+
+| numba           | 0.46.0          |         |
++-----------------+-----------------+---------+
+| openpyxl        | 2.5.7           |         |
++-----------------+-----------------+---------+
+| pyarrow         | 0.13.0          |         |
++-----------------+-----------------+---------+
+| pymysql         | 0.7.1           |         |
++-----------------+-----------------+---------+
+| pytables        | 3.4.3           |    X    |
++-----------------+-----------------+---------+
+| s3fs            | 0.4.0           |    X    |
++-----------------+-----------------+---------+
+| scipy           | 1.2.0           |    X    |
++-----------------+-----------------+---------+
+| sqlalchemy      | 1.1.4           |         |
++-----------------+-----------------+---------+
+| xarray          | 0.8.2           |         |
++-----------------+-----------------+---------+
+| xlrd            | 1.1.0           |         |
++-----------------+-----------------+---------+
+| xlsxwriter      | 0.9.8           |         |
++-----------------+-----------------+---------+
+| xlwt            | 1.2.0           |         |
++-----------------+-----------------+---------+
+| pandas-gbq      | 1.2.0           |    X    |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+Development changes
+^^^^^^^^^^^^^^^^^^^
+
+- The minimum version of Cython is now the most recent bug-fix version (0.29.16) (:issue:`33334`).
+
+
+.. _whatsnew_110.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- Lookups on a :class:`Series` with a single-item list containing a slice (e.g. ``ser[[slice(0, 4)]]``) are deprecated and will raise in a future version.  Either convert the list to a tuple, or pass the slice directly instead (:issue:`31333`)
+
+- :meth:`DataFrame.mean` and :meth:`DataFrame.median` with ``numeric_only=None`` will include ``datetime64`` and ``datetime64tz`` columns in a future version (:issue:`29941`)
+- Setting values with ``.loc`` using a positional slice is deprecated and will raise in a future version.  Use ``.loc`` with labels or ``.iloc`` with positions instead (:issue:`31840`)
+- :meth:`DataFrame.to_dict` has deprecated accepting short names for ``orient`` and will raise in a future version (:issue:`32515`)
+- :meth:`Categorical.to_dense` is deprecated and will be removed in a future version, use ``np.asarray(cat)`` instead (:issue:`32639`)
+- The ``fastpath`` keyword in the ``SingleBlockManager`` constructor is deprecated and will be removed in a future version (:issue:`33092`)
+- Providing ``suffixes`` as a ``set`` in :func:`pandas.merge` is deprecated. Provide a tuple instead (:issue:`33740`, :issue:`34741`).
+- Indexing a :class:`Series` with a multi-dimensional indexer like ``[:, None]`` to return an ``ndarray`` now raises a ``FutureWarning``. Convert to a NumPy array before indexing instead (:issue:`27837`)
+- :meth:`Index.is_mixed` is deprecated and will be removed in a future version, check ``index.inferred_type`` directly instead (:issue:`32922`)
+
+- Passing any arguments but the first one to :func:`read_html` as
+  positional arguments is deprecated. All other
+  arguments should be given as keyword arguments (:issue:`27573`).
+
+- Passing any arguments but ``path_or_buf`` (the first one) to
+  :func:`read_json` as positional arguments is deprecated. All
+  other arguments should be given as keyword arguments (:issue:`27573`).
+
+- Passing any arguments but the first two to :func:`read_excel` as
+  positional arguments is deprecated. All other
+  arguments should be given as keyword arguments (:issue:`27573`).
+
+- :func:`pandas.api.types.is_categorical` is deprecated and will be removed in a future version; use :func:`pandas.api.types.is_categorical_dtype` instead (:issue:`33385`)
+- :meth:`Index.get_value` is deprecated and will be removed in a future version (:issue:`19728`)
+- :meth:`Series.dt.week` and :meth:`Series.dt.weekofyear` are deprecated and will be removed in a future version, use :meth:`Series.dt.isocalendar().week` instead (:issue:`33595`)
+- :meth:`DatetimeIndex.week` and ``DatetimeIndex.weekofyear`` are deprecated and will be removed in a future version, use ``DatetimeIndex.isocalendar().week`` instead (:issue:`33595`)
+- :meth:`DatetimeArray.week` and ``DatetimeArray.weekofyear`` are deprecated and will be removed in a future version, use ``DatetimeArray.isocalendar().week`` instead (:issue:`33595`)
+- :meth:`DateOffset.__call__` is deprecated and will be removed in a future version, use ``offset + other`` instead (:issue:`34171`)
+- :meth:`~pandas.tseries.offsets.BusinessDay.apply_index` is deprecated and will be removed in a future version. Use ``offset + other`` instead (:issue:`34580`)
+- :meth:`DataFrame.tshift` and :meth:`Series.tshift` are deprecated and will be removed in a future version, use :meth:`DataFrame.shift` and :meth:`Series.shift` instead (:issue:`11631`)
+- Indexing an :class:`Index` object with a float key is deprecated, and will
+  raise an ``IndexError`` in the future. You can manually convert to an integer key
+  instead (:issue:`34191`).
+- The ``squeeze`` keyword in :meth:`~DataFrame.groupby` is deprecated and will be removed in a future version (:issue:`32380`)
+- The ``tz`` keyword in :meth:`Period.to_timestamp` is deprecated and will be removed in a future version; use ``per.to_timestamp(...).tz_localize(tz)`` instead (:issue:`34522`)
+- :meth:`DatetimeIndex.to_perioddelta` is deprecated and will be removed in a future version.  Use ``index - index.to_period(freq).to_timestamp()`` instead (:issue:`34853`)
+- :meth:`DataFrame.melt` accepting a ``value_name`` that already exists is deprecated, and will be removed in a future version (:issue:`34731`)
+- The ``center`` keyword in the :meth:`DataFrame.expanding` function is deprecated and will be removed in a future version (:issue:`20647`)
+
+
+
+.. ---------------------------------------------------------------------------
+
+
+.. _whatsnew_110.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Performance improvement in :class:`Timedelta` constructor (:issue:`30543`)
+- Performance improvement in :class:`Timestamp` constructor (:issue:`30543`)
+- Performance improvement in flex arithmetic ops between :class:`DataFrame` and :class:`Series` with ``axis=0`` (:issue:`31296`)
+- Performance improvement in arithmetic ops between :class:`DataFrame` and :class:`Series` with ``axis=1`` (:issue:`33600`)
+- The internal index method :meth:`~Index._shallow_copy` now copies cached attributes over to the new index,
+  avoiding creating these again on the new index. This can speed up many operations that depend on creating copies of
+  existing indexes (:issue:`28584`, :issue:`32640`, :issue:`32669`)
+- Significant performance improvement when creating a :class:`DataFrame` with
+  sparse values from ``scipy.sparse`` matrices using the
+  :meth:`DataFrame.sparse.from_spmatrix` constructor (:issue:`32821`,
+  :issue:`32825`,  :issue:`32826`, :issue:`32856`, :issue:`32858`).
+- Performance improvement for groupby methods :meth:`.Groupby.first`
+  and :meth:`.Groupby.last` (:issue:`34178`)
+- Performance improvement in :func:`factorize` for nullable (integer and Boolean) dtypes (:issue:`33064`).
+- Performance improvement when constructing :class:`Categorical` objects (:issue:`33921`)
+- Fixed performance regression in :func:`pandas.qcut` and :func:`pandas.cut` (:issue:`33921`)
+- Performance improvement in reductions (``sum``, ``prod``, ``min``, ``max``) for nullable (integer and Boolean) dtypes (:issue:`30982`, :issue:`33261`, :issue:`33442`).
+- Performance improvement in arithmetic operations between two :class:`DataFrame` objects (:issue:`32779`)
+- Performance improvement in :class:`.RollingGroupby` (:issue:`34052`)
+- Performance improvement in arithmetic operations (``sub``, ``add``, ``mul``, ``div``) for :class:`MultiIndex` (:issue:`34297`)
+- Performance improvement in ``DataFrame[bool_indexer]`` when ``bool_indexer`` is a ``list`` (:issue:`33924`)
+- Significant performance improvement of :meth:`io.formats.style.Styler.render` with styles added with various ways such as :meth:`io.formats.style.Styler.apply`, :meth:`io.formats.style.Styler.applymap` or :meth:`io.formats.style.Styler.bar` (:issue:`19917`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_110.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+
+Categorical
+^^^^^^^^^^^
+
+- Passing an invalid ``fill_value`` to :meth:`Categorical.take` raises a ``ValueError`` instead of ``TypeError`` (:issue:`33660`)
+- Combining a :class:`Categorical` with integer categories and which contains missing values with a float dtype column in operations such as :func:`concat` or :meth:`~DataFrame.append` will now result in a float column instead of an object dtype column (:issue:`33607`)
+- Bug where :func:`merge` was unable to join on non-unique categorical indices (:issue:`28189`)
+- Bug when passing categorical data to :class:`Index` constructor along with ``dtype=object`` incorrectly returning a :class:`CategoricalIndex` instead of object-dtype :class:`Index` (:issue:`32167`)
+- Bug where :class:`Categorical` comparison operator ``__ne__`` would incorrectly evaluate to ``False`` when either element was missing (:issue:`32276`)
+- :meth:`Categorical.fillna` now accepts :class:`Categorical` ``other`` argument (:issue:`32420`)
+- Repr of :class:`Categorical` was not distinguishing between ``int`` and ``str`` (:issue:`33676`)
+
+Datetimelike
+^^^^^^^^^^^^
+
+- Passing an integer dtype other than ``int64`` to ``np.array(period_index, dtype=...)`` will now raise ``TypeError`` instead of incorrectly using ``int64`` (:issue:`32255`)
+- :meth:`Series.to_timestamp` now raises a ``TypeError`` if the axis is not a :class:`PeriodIndex`. Previously an ``AttributeError`` was raised (:issue:`33327`)
+- :meth:`Series.to_period` now raises a ``TypeError`` if the axis is not a :class:`DatetimeIndex`. Previously an ``AttributeError`` was raised (:issue:`33327`)
+- :class:`Period` no longer accepts tuples for the ``freq`` argument (:issue:`34658`)
+- Bug in :class:`Timestamp` where constructing a :class:`Timestamp` from ambiguous epoch time and calling constructor again changed the :meth:`Timestamp.value` property (:issue:`24329`)
+- :meth:`DatetimeArray.searchsorted`, :meth:`TimedeltaArray.searchsorted`, :meth:`PeriodArray.searchsorted` not recognizing non-pandas scalars and incorrectly raising ``ValueError`` instead of ``TypeError`` (:issue:`30950`)
+- Bug in :class:`Timestamp` where constructing :class:`Timestamp` with dateutil timezone less than 128 nanoseconds before daylight saving time switch from winter to summer would result in nonexistent time (:issue:`31043`)
+- Bug in :meth:`Period.to_timestamp`, :meth:`Period.start_time` with microsecond frequency returning a timestamp one nanosecond earlier than the correct time (:issue:`31475`)
+- :class:`Timestamp` raised a confusing error message when year, month or day is missing (:issue:`31200`)
+- Bug in :class:`DatetimeIndex` constructor incorrectly accepting ``bool``-dtype inputs (:issue:`32668`)
+- Bug in :meth:`DatetimeIndex.searchsorted` not accepting a ``list`` or :class:`Series` as its argument (:issue:`32762`)
+- Bug where :meth:`PeriodIndex` raised when passed a :class:`Series` of strings (:issue:`26109`)
+- Bug in :class:`Timestamp` arithmetic when adding or subtracting an ``np.ndarray`` with ``timedelta64`` dtype (:issue:`33296`)
+- Bug in :meth:`DatetimeIndex.to_period` not inferring the frequency when called with no arguments (:issue:`33358`)
+- Bug in :meth:`DatetimeIndex.tz_localize` incorrectly retaining ``freq`` in some cases where the original ``freq`` is no longer valid (:issue:`30511`)
+- Bug in :meth:`DatetimeIndex.intersection` losing ``freq`` and timezone in some cases (:issue:`33604`)
+- Bug in :meth:`DatetimeIndex.get_indexer` where incorrect output would be returned for mixed datetime-like targets (:issue:`33741`)
+- Bug in :class:`DatetimeIndex` addition and subtraction with some types of :class:`DateOffset` objects incorrectly retaining an invalid ``freq`` attribute (:issue:`33779`)
+- Bug in :class:`DatetimeIndex` where setting the ``freq`` attribute on an index could silently change the ``freq`` attribute on another index viewing the same data (:issue:`33552`)
+- :meth:`DataFrame.min` and :meth:`DataFrame.max` were not returning consistent results with :meth:`Series.min` and :meth:`Series.max` when called on objects initialized with empty :func:`pd.to_datetime`
+- Bug in :meth:`DatetimeIndex.intersection` and :meth:`TimedeltaIndex.intersection` with results not having the correct ``name`` attribute (:issue:`33904`)
+- Bug in :meth:`DatetimeArray.__setitem__`, :meth:`TimedeltaArray.__setitem__`, :meth:`PeriodArray.__setitem__` incorrectly allowing values with ``int64`` dtype to be silently cast (:issue:`33717`)
+- Bug in subtracting :class:`TimedeltaIndex` from :class:`Period` incorrectly raising ``TypeError`` in some cases where it should succeed and ``IncompatibleFrequency`` in some cases where it should raise ``TypeError`` (:issue:`33883`)
+- Bug in constructing a :class:`Series` or :class:`Index` from a read-only NumPy array with non-ns
+  resolution which converted to object dtype instead of coercing to ``datetime64[ns]``
+  dtype when within the timestamp bounds (:issue:`34843`).
+- The ``freq`` keyword in :class:`Period`, :func:`date_range`, :func:`period_range`, :func:`pd.tseries.frequencies.to_offset` no longer allows tuples, pass as string instead (:issue:`34703`)
+- Bug in :meth:`DataFrame.append` when appending a :class:`Series` containing a scalar tz-aware :class:`Timestamp` to an empty :class:`DataFrame` resulted in an object column instead of ``datetime64[ns, tz]`` dtype (:issue:`35038`)
+- ``OutOfBoundsDatetime`` issues an improved error message when timestamp is out of implementation bounds. (:issue:`32967`)
+- Bug in :meth:`AbstractHolidayCalendar.holidays` when no rules were defined (:issue:`31415`)
+- Bug in :class:`Tick` comparisons raising ``TypeError`` when comparing against timedelta-like objects (:issue:`34088`)
+- Bug in :class:`Tick` multiplication raising ``TypeError`` when multiplying by a float (:issue:`34486`)
+
+Timedelta
+^^^^^^^^^
+
+- Bug in constructing a :class:`Timedelta` with a high precision integer that would round the :class:`Timedelta` components (:issue:`31354`)
+- Bug in dividing ``np.nan`` or ``None`` by :class:`Timedelta` incorrectly returning ``NaT`` (:issue:`31869`)
+- :class:`Timedelta` now understands ``µs`` as an identifier for microsecond (:issue:`32899`)
+- :class:`Timedelta` string representation now includes nanoseconds, when nanoseconds are non-zero (:issue:`9309`)
+- Bug in comparing a :class:`Timedelta` object against an ``np.ndarray`` with ``timedelta64`` dtype incorrectly viewing all entries as unequal (:issue:`33441`)
+- Bug in :func:`timedelta_range` that produced an extra point on a edge case (:issue:`30353`, :issue:`33498`)
+- Bug in :meth:`DataFrame.resample` that produced an extra point on a edge case (:issue:`30353`, :issue:`13022`, :issue:`33498`)
+- Bug in :meth:`DataFrame.resample` that ignored the ``loffset`` argument when dealing with timedelta (:issue:`7687`, :issue:`33498`)
+- Bug in :class:`Timedelta` and :func:`pandas.to_timedelta` that ignored the ``unit`` argument for string input (:issue:`12136`)
+
+Timezones
+^^^^^^^^^
+
+- Bug in :func:`to_datetime` with ``infer_datetime_format=True`` where timezone names (e.g. ``UTC``) would not be parsed correctly (:issue:`33133`)
+
+
+Numeric
+^^^^^^^
+- Bug in :meth:`DataFrame.floordiv` with ``axis=0`` not treating division-by-zero like :meth:`Series.floordiv` (:issue:`31271`)
+- Bug in :func:`to_numeric` with string argument ``"uint64"`` and ``errors="coerce"`` silently fails (:issue:`32394`)
+- Bug in :func:`to_numeric` with ``downcast="unsigned"`` fails for empty data (:issue:`32493`)
+- Bug in :meth:`DataFrame.mean` with ``numeric_only=False`` and either ``datetime64`` dtype or ``PeriodDtype`` column incorrectly raising ``TypeError`` (:issue:`32426`)
+- Bug in :meth:`DataFrame.count` with ``level="foo"`` and index level ``"foo"`` containing NaNs causes segmentation fault (:issue:`21824`)
+- Bug in :meth:`DataFrame.diff` with ``axis=1`` returning incorrect results with mixed dtypes (:issue:`32995`)
+- Bug in :meth:`DataFrame.corr` and :meth:`DataFrame.cov` raising when handling nullable integer columns with ``pandas.NA`` (:issue:`33803`)
+- Bug in arithmetic operations between :class:`DataFrame` objects with non-overlapping columns with duplicate labels causing an infinite loop (:issue:`35194`)
+- Bug in :class:`DataFrame` and :class:`Series` addition and subtraction between object-dtype objects and ``datetime64`` dtype objects (:issue:`33824`)
+- Bug in :meth:`Index.difference` giving incorrect results when comparing a :class:`Float64Index` and object :class:`Index` (:issue:`35217`)
+- Bug in :class:`DataFrame` reductions (e.g. ``df.min()``, ``df.max()``) with ``ExtensionArray`` dtypes (:issue:`34520`, :issue:`32651`)
+- :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` now raise a ValueError if ``limit_direction`` is ``'forward'`` or ``'both'`` and ``method`` is ``'backfill'`` or ``'bfill'`` or ``limit_direction`` is ``'backward'`` or ``'both'`` and ``method`` is ``'pad'`` or ``'ffill'`` (:issue:`34746`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :class:`Series` construction from NumPy array with big-endian ``datetime64`` dtype (:issue:`29684`)
+- Bug in :class:`Timedelta` construction with large nanoseconds keyword value (:issue:`32402`)
+- Bug in :class:`DataFrame` construction where sets would be duplicated rather than raising (:issue:`32582`)
+- The :class:`DataFrame` constructor no longer accepts a list of :class:`DataFrame` objects. Because of changes to NumPy, :class:`DataFrame` objects are now consistently treated as 2D objects, so a list of :class:`DataFrame` objects is considered 3D, and no longer acceptable for the :class:`DataFrame` constructor (:issue:`32289`).
+- Bug in :class:`DataFrame` when initiating a frame with lists and assign ``columns`` with nested list for ``MultiIndex`` (:issue:`32173`)
+- Improved error message for invalid construction of list when creating a new index (:issue:`35190`)
+
+
+Strings
+^^^^^^^
+
+- Bug in the :meth:`~Series.astype` method when converting "string" dtype data to nullable integer dtype (:issue:`32450`).
+- Fixed issue where taking ``min`` or ``max`` of a ``StringArray`` or ``Series`` with ``StringDtype`` type would raise. (:issue:`31746`)
+- Bug in :meth:`Series.str.cat` returning ``NaN`` output when other had :class:`Index` type (:issue:`33425`)
+- :func:`pandas.api.dtypes.is_string_dtype` no longer incorrectly identifies categorical series as string.
+
+Interval
+^^^^^^^^
+- Bug in :class:`IntervalArray` incorrectly allowing the underlying data to be changed when setting values (:issue:`32782`)
+
+Indexing
+^^^^^^^^
+
+- :meth:`DataFrame.xs` now raises a  ``TypeError`` if a ``level`` keyword is supplied and the axis is not a :class:`MultiIndex`. Previously an ``AttributeError`` was raised (:issue:`33610`)
+- Bug in slicing on a :class:`DatetimeIndex` with a partial-timestamp dropping high-resolution indices near the end of a year, quarter, or month (:issue:`31064`)
+- Bug in :meth:`PeriodIndex.get_loc` treating higher-resolution strings differently from :meth:`PeriodIndex.get_value` (:issue:`31172`)
+- Bug in :meth:`Series.at` and :meth:`DataFrame.at` not matching ``.loc`` behavior when looking up an integer in a :class:`Float64Index` (:issue:`31329`)
+- Bug in :meth:`PeriodIndex.is_monotonic` incorrectly returning ``True`` when containing leading ``NaT`` entries (:issue:`31437`)
+- Bug in :meth:`DatetimeIndex.get_loc` raising ``KeyError`` with converted-integer key instead of the user-passed key (:issue:`31425`)
+- Bug in :meth:`Series.xs` incorrectly returning ``Timestamp`` instead of ``datetime64`` in some object-dtype cases (:issue:`31630`)
+- Bug in :meth:`DataFrame.iat` incorrectly returning ``Timestamp`` instead of ``datetime`` in some object-dtype cases (:issue:`32809`)
+- Bug in :meth:`DataFrame.at` when either columns or index is non-unique (:issue:`33041`)
+- Bug in :meth:`Series.loc` and :meth:`DataFrame.loc` when indexing with an integer key on a object-dtype :class:`Index` that is not all-integers (:issue:`31905`)
+- Bug in :meth:`DataFrame.iloc.__setitem__` on a :class:`DataFrame` with duplicate columns incorrectly setting values for all matching columns (:issue:`15686`, :issue:`22036`)
+- Bug in :meth:`DataFrame.loc` and :meth:`Series.loc` with a :class:`DatetimeIndex`, :class:`TimedeltaIndex`, or :class:`PeriodIndex` incorrectly allowing lookups of non-matching datetime-like dtypes (:issue:`32650`)
+- Bug in :meth:`Series.__getitem__` indexing with non-standard scalars, e.g. ``np.dtype`` (:issue:`32684`)
+- Bug in :class:`Index` constructor where an unhelpful error message was raised for NumPy scalars (:issue:`33017`)
+- Bug in :meth:`DataFrame.lookup` incorrectly raising an ``AttributeError`` when ``frame.index`` or ``frame.columns`` is not unique; this will now raise a ``ValueError`` with a helpful error message (:issue:`33041`)
+- Bug in :class:`Interval` where a :class:`Timedelta` could not be added or subtracted from a :class:`Timestamp` interval (:issue:`32023`)
+- Bug in :meth:`DataFrame.copy` not invalidating _item_cache after copy caused post-copy value updates to not be reflected (:issue:`31784`)
+- Fixed regression in :meth:`DataFrame.loc` and :meth:`Series.loc` throwing an error when a ``datetime64[ns, tz]`` value is provided (:issue:`32395`)
+- Bug in :meth:`Series.__getitem__` with an integer key and a :class:`MultiIndex` with leading integer level failing to raise ``KeyError`` if the key is not present in the first level (:issue:`33355`)
+- Bug in :meth:`DataFrame.iloc` when slicing a single column :class:`DataFrame` with ``ExtensionDtype`` (e.g. ``df.iloc[:, :1]``) returning an invalid result (:issue:`32957`)
+- Bug in :meth:`DatetimeIndex.insert` and :meth:`TimedeltaIndex.insert` causing index ``freq`` to be lost when setting an element into an empty :class:`Series` (:issue:`33573`)
+- Bug in :meth:`Series.__setitem__` with an :class:`IntervalIndex` and a list-like key of integers (:issue:`33473`)
+- Bug in :meth:`Series.__getitem__` allowing missing labels with ``np.ndarray``, :class:`Index`, :class:`Series` indexers but not ``list``, these now all raise ``KeyError`` (:issue:`33646`)
+- Bug in :meth:`DataFrame.truncate` and :meth:`Series.truncate` where index was assumed to be monotone increasing (:issue:`33756`)
+- Indexing with a list of strings representing datetimes failed on :class:`DatetimeIndex` or :class:`PeriodIndex` (:issue:`11278`)
+- Bug in :meth:`Series.at` when used with a :class:`MultiIndex` would raise an exception on valid inputs (:issue:`26989`)
+- Bug in :meth:`DataFrame.loc` with dictionary of values changes columns with dtype of ``int`` to ``float`` (:issue:`34573`)
+- Bug in :meth:`Series.loc` when used with a :class:`MultiIndex` would raise an ``IndexingError`` when accessing a ``None`` value (:issue:`34318`)
+- Bug in :meth:`DataFrame.reset_index` and :meth:`Series.reset_index` would not preserve data types on an empty :class:`DataFrame` or :class:`Series` with a :class:`MultiIndex` (:issue:`19602`)
+- Bug in :class:`Series` and :class:`DataFrame` indexing with a ``time`` key on a :class:`DatetimeIndex` with ``NaT`` entries (:issue:`35114`)
+
+Missing
+^^^^^^^
+- Calling :meth:`fillna` on an empty :class:`Series` now correctly returns a shallow copied object. The behaviour is now consistent with :class:`Index`, :class:`DataFrame` and a non-empty :class:`Series` (:issue:`32543`).
+- Bug in :meth:`Series.replace` when argument ``to_replace`` is of type dict/list and is used on a :class:`Series` containing ``<NA>`` was raising a ``TypeError``. The method now handles this by ignoring ``<NA>`` values when doing the comparison for the replacement (:issue:`32621`)
+- Bug in :meth:`~Series.any` and :meth:`~Series.all` incorrectly returning ``<NA>`` for all ``False`` or all ``True`` values using the nulllable Boolean dtype and with ``skipna=False`` (:issue:`33253`)
+- Clarified documentation on interpolate with ``method=akima``. The ``der`` parameter must be scalar or ``None`` (:issue:`33426`)
+- :meth:`DataFrame.interpolate` uses the correct axis convention now. Previously interpolating along columns lead to interpolation along indices and vice versa. Furthermore interpolating with methods ``pad``, ``ffill``, ``bfill`` and ``backfill`` are identical to using these methods with :meth:`DataFrame.fillna` (:issue:`12918`, :issue:`29146`)
+- Bug in :meth:`DataFrame.interpolate` when called on a :class:`DataFrame` with column names of string type was throwing a ValueError. The method is now independent of the type of the column names (:issue:`33956`)
+- Passing :class:`NA` into a format string using format specs will now work. For example ``"{:.1f}".format(pd.NA)`` would previously raise a ``ValueError``, but will now return the string ``"<NA>"`` (:issue:`34740`)
+- Bug in :meth:`Series.map` not raising on invalid ``na_action`` (:issue:`32815`)
+
+MultiIndex
+^^^^^^^^^^
+
+- :meth:`DataFrame.swaplevels` now raises a ``TypeError`` if the axis is not a :class:`MultiIndex`. Previously an ``AttributeError`` was raised (:issue:`31126`)
+- Bug in :meth:`Dataframe.loc` when used with a :class:`MultiIndex`. The returned values were not in the same order as the given inputs (:issue:`22797`)
+
+.. ipython:: python
+
+        df = pd.DataFrame(np.arange(4),
+                          index=[["a", "a", "b", "b"], [1, 2, 1, 2]])
+        # Rows are now ordered as the requested keys
+        df.loc[(['b', 'a'], [2, 1]), :]
+
+- Bug in :meth:`MultiIndex.intersection` was not guaranteed to preserve order when ``sort=False``. (:issue:`31325`)
+- Bug in :meth:`DataFrame.truncate` was dropping :class:`MultiIndex` names. (:issue:`34564`)
+
+.. ipython:: python
+
+        left = pd.MultiIndex.from_arrays([["b", "a"], [2, 1]])
+        right = pd.MultiIndex.from_arrays([["a", "b", "c"], [1, 2, 3]])
+        # Common elements are now guaranteed to be ordered by the left side
+        left.intersection(right, sort=False)
+
+- Bug when joining two :class:`MultiIndex` without specifying level with different columns. Return-indexers parameter was ignored. (:issue:`34074`)
+
+IO
+^^
+- Passing a ``set`` as ``names`` argument to :func:`pandas.read_csv`, :func:`pandas.read_table`, or :func:`pandas.read_fwf` will raise ``ValueError: Names should be an ordered collection.`` (:issue:`34946`)
+- Bug in print-out when ``display.precision`` is zero. (:issue:`20359`)
+- Bug in :func:`read_json` where integer overflow was occurring when json contains big number strings. (:issue:`30320`)
+- :func:`read_csv` will now raise a ``ValueError`` when the arguments ``header`` and ``prefix`` both are not ``None``. (:issue:`27394`)
+- Bug in :meth:`DataFrame.to_json` was raising ``NotFoundError`` when ``path_or_buf`` was an S3 URI (:issue:`28375`)
+- Bug in :meth:`DataFrame.to_parquet` overwriting pyarrow's default for
+  ``coerce_timestamps``; following pyarrow's default allows writing nanosecond
+  timestamps with ``version="2.0"`` (:issue:`31652`).
+- Bug in :func:`read_csv` was raising ``TypeError`` when ``sep=None`` was used in combination with ``comment`` keyword (:issue:`31396`)
+- Bug in :class:`HDFStore` that caused it to set to ``int64`` the dtype of a ``datetime64`` column when reading a :class:`DataFrame` in Python 3 from fixed format written in Python 2 (:issue:`31750`)
+- :func:`read_sas` now handles dates and datetimes larger than :attr:`Timestamp.max` returning them as :class:`datetime.datetime` objects (:issue:`20927`)
+- Bug in :meth:`DataFrame.to_json` where ``Timedelta`` objects would not be serialized correctly with ``date_format="iso"`` (:issue:`28256`)
+- :func:`read_csv` will raise a ``ValueError`` when the column names passed in ``parse_dates`` are missing in the :class:`Dataframe` (:issue:`31251`)
+- Bug in :func:`read_excel` where a UTF-8 string with a high surrogate would cause a segmentation violation (:issue:`23809`)
+- Bug in :func:`read_csv` was causing a file descriptor leak on an empty file (:issue:`31488`)
+- Bug in :func:`read_csv` was causing a segfault when there were blank lines between the header and data rows (:issue:`28071`)
+- Bug in :func:`read_csv` was raising a misleading exception on a permissions issue (:issue:`23784`)
+- Bug in :func:`read_csv` was raising an ``IndexError`` when ``header=None`` and two extra data columns
+- Bug in :func:`read_sas` was raising an ``AttributeError`` when reading files from Google Cloud Storage (:issue:`33069`)
+- Bug in :meth:`DataFrame.to_sql` where an ``AttributeError`` was raised when saving an out of bounds date (:issue:`26761`)
+- Bug in :func:`read_excel` did not correctly handle multiple embedded spaces in OpenDocument text cells. (:issue:`32207`)
+- Bug in :func:`read_json` was raising ``TypeError`` when reading a ``list`` of Booleans into a :class:`Series`. (:issue:`31464`)
+- Bug in :func:`pandas.io.json.json_normalize` where location specified by ``record_path`` doesn't point to an array. (:issue:`26284`)
+- :func:`pandas.read_hdf` has a more explicit error message when loading an
+  unsupported HDF file (:issue:`9539`)
+- Bug in :meth:`~DataFrame.read_feather` was raising an ``ArrowIOError`` when reading an s3 or http file path (:issue:`29055`)
+- Bug in :meth:`~DataFrame.to_excel` could not handle the column name ``render`` and was raising an ``KeyError`` (:issue:`34331`)
+- Bug in :meth:`~SQLDatabase.execute` was raising a ``ProgrammingError`` for some DB-API drivers when the SQL statement contained the ``%`` character and no parameters were present (:issue:`34211`)
+- Bug in :meth:`~pandas.io.stata.StataReader` which resulted in categorical variables with different dtypes when reading data using an iterator. (:issue:`31544`)
+- :meth:`HDFStore.keys` has now an optional ``include`` parameter that allows the retrieval of all native HDF5 table names (:issue:`29916`)
+- ``TypeError`` exceptions raised by :func:`read_csv` and :func:`read_table` were showing as ``parser_f`` when an unexpected keyword argument was passed (:issue:`25648`)
+- Bug in :func:`read_excel` for ODS files removes 0.0 values (:issue:`27222`)
+- Bug in :func:`ujson.encode` was raising an ``OverflowError`` with numbers larger than ``sys.maxsize`` (:issue:`34395`)
+- Bug in :meth:`HDFStore.append_to_multiple` was raising a ``ValueError`` when the ``min_itemsize`` parameter is set (:issue:`11238`)
+- Bug in :meth:`~HDFStore.create_table` now raises an error when ``column`` argument was not specified in ``data_columns`` on input (:issue:`28156`)
+- :func:`read_json` now could read line-delimited json file from a file url while ``lines`` and ``chunksize`` are set.
+- Bug in :meth:`DataFrame.to_sql` when reading DataFrames with ``-np.inf`` entries with MySQL now has a more explicit ``ValueError`` (:issue:`34431`)
+- Bug where capitalised files extensions were not decompressed by read_* functions (:issue:`35164`)
+- Bug in :meth:`read_excel` that was raising a ``TypeError`` when ``header=None`` and ``index_col`` is given as a ``list`` (:issue:`31783`)
+- Bug in :func:`read_excel` where datetime values are used in the header in a :class:`MultiIndex` (:issue:`34748`)
+- :func:`read_excel` no longer takes ``**kwds`` arguments. This means that passing in the keyword argument ``chunksize`` now raises a ``TypeError`` (previously raised a ``NotImplementedError``), while passing in the keyword argument ``encoding`` now raises a ``TypeError`` (:issue:`34464`)
+- Bug in :meth:`DataFrame.to_records` was incorrectly losing timezone information in timezone-aware ``datetime64`` columns (:issue:`32535`)
+
+Plotting
+^^^^^^^^
+
+- :meth:`DataFrame.plot` for line/bar now accepts color by dictionary (:issue:`8193`).
+- Bug in :meth:`DataFrame.plot.hist` where weights are not working for multiple columns (:issue:`33173`)
+- Bug in :meth:`DataFrame.boxplot` and :meth:`DataFrame.plot.boxplot` lost color attributes of ``medianprops``, ``whiskerprops``, ``capprops`` and ``boxprops`` (:issue:`30346`)
+- Bug in :meth:`DataFrame.hist` where the order of ``column`` argument was ignored (:issue:`29235`)
+- Bug in :meth:`DataFrame.plot.scatter` that when adding multiple plots with different ``cmap``, colorbars always use the first ``cmap`` (:issue:`33389`)
+- Bug in :meth:`DataFrame.plot.scatter` was adding a colorbar to the plot even if the argument ``c`` was assigned to a column containing color names (:issue:`34316`)
+- Bug in :meth:`pandas.plotting.bootstrap_plot` was causing cluttered axes and overlapping labels (:issue:`34905`)
+- Bug in :meth:`DataFrame.plot.scatter` caused an error when plotting variable marker sizes (:issue:`32904`)
+
+GroupBy/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Using a :class:`pandas.api.indexers.BaseIndexer` with ``count``, ``min``, ``max``, ``median``, ``skew``,  ``cov``, ``corr`` will now return correct results for any monotonic :class:`pandas.api.indexers.BaseIndexer` descendant (:issue:`32865`)
+- :meth:`DataFrameGroupby.mean` and :meth:`SeriesGroupby.mean` (and similarly for :meth:`~DataFrameGroupby.median`, :meth:`~DataFrameGroupby.std` and :meth:`~DataFrameGroupby.var`) now raise a ``TypeError`` if a non-accepted keyword argument is passed into it. Previously an ``UnsupportedFunctionCall`` was raised (``AssertionError`` if ``min_count`` passed into :meth:`~DataFrameGroupby.median`) (:issue:`31485`)
+- Bug in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` raising ``ValueError`` when the ``by`` axis is not sorted, has duplicates, and the applied ``func`` does not mutate passed in objects (:issue:`30667`)
+- Bug in :meth:`DataFrameGroupBy.transform` produces an incorrect result with transformation functions (:issue:`30918`)
+- Bug in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` were returning the wrong result when grouping by multiple keys of which some were categorical and others not (:issue:`32494`)
+- Bug in :meth:`.DataFrameGroupBy.count` and :meth:`.SeriesGroupBy.count` causing segmentation fault when grouped-by columns contain NaNs (:issue:`32841`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` produces inconsistent type when aggregating Boolean :class:`Series` (:issue:`32894`)
+- Bug in :meth:`DataFrameGroupBy.sum` and :meth:`SeriesGroupBy.sum` where a large negative number would be returned when the number of non-null values was below ``min_count`` for nullable integer dtypes (:issue:`32861`)
+- Bug in :meth:`SeriesGroupBy.quantile` was raising on nullable integers (:issue:`33136`)
+- Bug in :meth:`DataFrame.resample` where an ``AmbiguousTimeError`` would be raised when the resulting timezone aware :class:`DatetimeIndex` had a DST transition at midnight (:issue:`25758`)
+- Bug in :meth:`DataFrame.groupby` where a ``ValueError`` would be raised when grouping by a categorical column with read-only categories and ``sort=False`` (:issue:`33410`)
+- Bug in :meth:`.DataFrameGroupBy.agg`, :meth:`.SeriesGroupBy.agg`, :meth:`.DataFrameGroupBy.transform`, :meth:`.SeriesGroupBy.transform`, :meth:`.DataFrameGroupBy.resample`, and :meth:`.SeriesGroupBy.resample` where subclasses are not preserved (:issue:`28330`)
+- Bug in :meth:`SeriesGroupBy.agg` where any column name was accepted in the named aggregation of :class:`SeriesGroupBy` previously. The behaviour now allows only ``str`` and callables else would raise ``TypeError``. (:issue:`34422`)
+- Bug in :meth:`DataFrame.groupby` lost the name of the :class:`Index` when one of the ``agg`` keys referenced an empty list (:issue:`32580`)
+- Bug in :meth:`Rolling.apply` where ``center=True`` was ignored when ``engine='numba'`` was specified (:issue:`34784`)
+- Bug in :meth:`DataFrame.ewm.cov` was throwing ``AssertionError`` for :class:`MultiIndex` inputs (:issue:`34440`)
+- Bug in :meth:`core.groupby.DataFrameGroupBy.quantile` raised ``TypeError`` for non-numeric types rather than dropping the columns (:issue:`27892`)
+- Bug in :meth:`core.groupby.DataFrameGroupBy.transform` when ``func='nunique'`` and columns are of type ``datetime64``, the result would also be of type ``datetime64`` instead of ``int64`` (:issue:`35109`)
+- Bug in :meth:`DataFrame.groupby` raising an ``AttributeError`` when selecting a column and aggregating with ``as_index=False`` (:issue:`35246`).
+- Bug in :meth:`DataFrameGroupBy.first` and :meth:`DataFrameGroupBy.last` that would raise an unnecessary ``ValueError`` when grouping on multiple ``Categoricals`` (:issue:`34951`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug effecting all numeric and Boolean reduction methods not returning subclassed data type. (:issue:`25596`)
+- Bug in :meth:`DataFrame.pivot_table` when only :class:`MultiIndexed` columns is set (:issue:`17038`)
+- Bug in :meth:`DataFrame.unstack` and :meth:`Series.unstack` can take tuple names in :class:`MultiIndexed` data (:issue:`19966`)
+- Bug in :meth:`DataFrame.pivot_table` when ``margin`` is ``True`` and only ``column`` is defined (:issue:`31016`)
+- Fixed incorrect error message in :meth:`DataFrame.pivot` when ``columns`` is set to ``None``. (:issue:`30924`)
+- Bug in :func:`crosstab` when inputs are two :class:`Series` and have tuple names, the output will keep a dummy :class:`MultiIndex` as columns. (:issue:`18321`)
+- :meth:`DataFrame.pivot` can now take lists for ``index`` and ``columns`` arguments (:issue:`21425`)
+- Bug in :func:`concat` where the resulting indices are not copied when ``copy=True`` (:issue:`29879`)
+- Bug in :meth:`SeriesGroupBy.aggregate` was resulting in aggregations being overwritten when they shared the same name (:issue:`30880`)
+- Bug where :meth:`Index.astype` would lose the :attr:`name` attribute when converting from ``Float64Index`` to ``Int64Index``, or when casting to an ``ExtensionArray`` dtype (:issue:`32013`)
+- :meth:`Series.append` will now raise a ``TypeError`` when passed a :class:`DataFrame` or a sequence containing :class:`DataFrame` (:issue:`31413`)
+- :meth:`DataFrame.replace` and :meth:`Series.replace` will raise a ``TypeError`` if ``to_replace`` is not an expected type. Previously the ``replace`` would fail silently (:issue:`18634`)
+- Bug on inplace operation of a :class:`Series` that was adding a column to the :class:`DataFrame` from where it was originally dropped from (using ``inplace=True``) (:issue:`30484`)
+- Bug in :meth:`DataFrame.apply` where callback was called with :class:`Series` parameter even though ``raw=True`` requested. (:issue:`32423`)
+- Bug in :meth:`DataFrame.pivot_table` losing timezone information when creating a :class:`MultiIndex` level from a column with timezone-aware dtype (:issue:`32558`)
+- Bug in :func:`concat` where when passing a non-dict mapping as ``objs`` would raise a ``TypeError`` (:issue:`32863`)
+- :meth:`DataFrame.agg` now provides more descriptive ``SpecificationError`` message when attempting to aggregate a non-existent column (:issue:`32755`)
+- Bug in :meth:`DataFrame.unstack` when :class:`MultiIndex` columns and :class:`MultiIndex` rows were used (:issue:`32624`, :issue:`24729` and :issue:`28306`)
+- Appending a dictionary to a :class:`DataFrame` without passing ``ignore_index=True`` will raise ``TypeError: Can only append a dict if ignore_index=True`` instead of ``TypeError: Can only append a :class:`Series` if ignore_index=True or if the :class:`Series` has a name`` (:issue:`30871`)
+- Bug in :meth:`DataFrame.corrwith`, :meth:`DataFrame.memory_usage`, :meth:`DataFrame.dot`,
+  :meth:`DataFrame.idxmin`, :meth:`DataFrame.idxmax`, :meth:`DataFrame.duplicated`, :meth:`DataFrame.isin`,
+  :meth:`DataFrame.count`, :meth:`Series.explode`, :meth:`Series.asof` and :meth:`DataFrame.asof` not
+  returning subclassed types. (:issue:`31331`)
+- Bug in :func:`concat` was not allowing for concatenation of :class:`DataFrame` and :class:`Series` with duplicate keys (:issue:`33654`)
+- Bug in :func:`cut` raised an error when the argument ``labels`` contains duplicates (:issue:`33141`)
+- Ensure only named functions can be used in :func:`eval` (:issue:`32460`)
+- Bug in :meth:`Dataframe.aggregate` and :meth:`Series.aggregate` was causing a recursive loop in some cases (:issue:`34224`)
+- Fixed bug in :func:`melt` where melting :class:`MultiIndex` columns with ``col_level > 0`` would raise a ``KeyError`` on ``id_vars`` (:issue:`34129`)
+- Bug in :meth:`Series.where` with an empty :class:`Series` and empty ``cond`` having non-bool dtype (:issue:`34592`)
+- Fixed regression where :meth:`DataFrame.apply` would raise ``ValueError`` for elements with ``S`` dtype (:issue:`34529`)
+
+Sparse
+^^^^^^
+- Creating a :class:`SparseArray` from timezone-aware dtype will issue a warning before dropping timezone information, instead of doing so silently (:issue:`32501`)
+- Bug in :meth:`arrays.SparseArray.from_spmatrix` wrongly read scipy sparse matrix (:issue:`31991`)
+- Bug in :meth:`Series.sum` with ``SparseArray`` raised a ``TypeError`` (:issue:`25777`)
+- Bug where :class:`DataFrame` containing an all-sparse :class:`SparseArray` filled with ``NaN`` when indexed by a list-like (:issue:`27781`, :issue:`29563`)
+- The repr of :class:`SparseDtype` now includes the repr of its ``fill_value`` attribute. Previously it used ``fill_value``'s  string representation (:issue:`34352`)
+- Bug where empty :class:`DataFrame` could not be cast to :class:`SparseDtype` (:issue:`33113`)
+- Bug in :meth:`arrays.SparseArray` was returning the incorrect type when indexing a sparse dataframe with an iterable (:issue:`34526`, :issue:`34540`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+
+- Fixed bug where :meth:`Series.value_counts` would raise on empty input of ``Int64`` dtype (:issue:`33317`)
+- Fixed bug in :func:`concat` when concatenating :class:`DataFrame` objects with non-overlapping columns resulting in object-dtype columns rather than preserving the extension dtype (:issue:`27692`, :issue:`33027`)
+- Fixed bug where :meth:`StringArray.isna` would return ``False`` for NA values when ``pandas.options.mode.use_inf_as_na`` was set to ``True`` (:issue:`33655`)
+- Fixed bug in :class:`Series` construction with EA dtype and index but no data or scalar data fails (:issue:`26469`)
+- Fixed bug that caused :meth:`Series.__repr__` to crash for extension types whose elements are multidimensional arrays (:issue:`33770`).
+- Fixed bug where :meth:`Series.update` would raise a ``ValueError`` for ``ExtensionArray`` dtypes with missing values (:issue:`33980`)
+- Fixed bug where :meth:`StringArray.memory_usage` was not implemented (:issue:`33963`)
+- Fixed bug where :meth:`DataFrameGroupBy` would ignore the ``min_count`` argument for aggregations on nullable Boolean dtypes (:issue:`34051`)
+- Fixed bug where the constructor of :class:`DataFrame` with ``dtype='string'`` would fail (:issue:`27953`, :issue:`33623`)
+- Bug where :class:`DataFrame` column set to scalar extension type was considered an object type rather than the extension type (:issue:`34832`)
+- Fixed bug in :meth:`IntegerArray.astype` to correctly copy the mask as well (:issue:`34931`).
+
+Other
+^^^^^
+
+- Set operations on an object-dtype :class:`Index` now always return object-dtype results (:issue:`31401`)
+- Fixed :func:`pandas.testing.assert_series_equal` to correctly raise if the ``left`` argument is a different subclass with ``check_series_type=True`` (:issue:`32670`).
+- Getting a missing attribute in a :meth:`DataFrame.query` or :meth:`DataFrame.eval` string raises the correct ``AttributeError`` (:issue:`32408`)
+- Fixed bug in :func:`pandas.testing.assert_series_equal` where dtypes were checked for ``Interval`` and ``ExtensionArray`` operands when ``check_dtype`` was ``False`` (:issue:`32747`)
+- Bug in :meth:`DataFrame.__dir__` caused a segfault when using unicode surrogates in a column name (:issue:`25509`)
+- Bug in :meth:`DataFrame.equals` and :meth:`Series.equals` in allowing subclasses to be equal (:issue:`34402`).
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_110.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.0.5..v1.1.0|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.1.1.rst b/pandas/doc/source/whatsnew/v1.1.1.rst
new file mode 100644
index 0000000000000..176a8b85e76b4
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.1.rst
@@ -0,0 +1,56 @@
+.. _whatsnew_111:
+
+What's new in 1.1.1 (August 20, 2020)
+-------------------------------------
+
+These are the changes in pandas 1.1.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_111.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`CategoricalIndex.format` where, when stringified scalars had different lengths, the shorter string would be right-filled with spaces, so it had the same length as the longest string (:issue:`35439`)
+- Fixed regression in :meth:`Series.truncate` when trying to truncate a single-element series (:issue:`35544`)
+- Fixed regression where :meth:`DataFrame.to_numpy` would raise a ``RuntimeError`` for mixed dtypes when converting to ``str`` (:issue:`35455`)
+- Fixed regression where :func:`read_csv` would raise a ``ValueError`` when ``pandas.options.mode.use_inf_as_na`` was set to ``True`` (:issue:`35493`)
+- Fixed regression where :func:`pandas.testing.assert_series_equal` would raise an error when non-numeric dtypes were passed with ``check_exact=True`` (:issue:`35446`)
+- Fixed regression in ``.groupby(..).rolling(..)`` where column selection was ignored (:issue:`35486`)
+- Fixed regression where :meth:`DataFrame.interpolate` would raise a ``TypeError`` when the :class:`DataFrame` was empty (:issue:`35598`)
+- Fixed regression in :meth:`DataFrame.shift` with ``axis=1`` and heterogeneous dtypes (:issue:`35488`)
+- Fixed regression in :meth:`DataFrame.diff` with read-only data (:issue:`35559`)
+- Fixed regression in ``.groupby(..).rolling(..)`` where a segfault would occur with ``center=True`` and an odd number of values (:issue:`35552`)
+- Fixed regression in :meth:`DataFrame.apply` where functions that altered the input in-place only operated on a single row (:issue:`35462`)
+- Fixed regression in :meth:`DataFrame.reset_index` would raise a ``ValueError`` on empty :class:`DataFrame` with a :class:`MultiIndex` with a ``datetime64`` dtype level (:issue:`35606`, :issue:`35657`)
+- Fixed regression where :func:`pandas.merge_asof` would raise a ``UnboundLocalError`` when ``left_index``, ``right_index`` and ``tolerance`` were set (:issue:`35558`)
+- Fixed regression in ``.groupby(..).rolling(..)`` where a custom ``BaseIndexer`` would be ignored (:issue:`35557`)
+- Fixed regression in :meth:`DataFrame.replace` and :meth:`Series.replace` where compiled regular expressions would be ignored during replacement (:issue:`35680`)
+- Fixed regression in :meth:`.DataFrameGroupBy.aggregate` where a list of functions would produce the wrong results if at least one of the functions did not aggregate (:issue:`35490`)
+- Fixed memory usage issue when instantiating large :class:`pandas.arrays.StringArray` (:issue:`35499`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_111.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in :class:`~pandas.io.formats.style.Styler` whereby ``cell_ids`` argument had no effect due to other recent changes (:issue:`35588`) (:issue:`35663`)
+- Bug in :func:`pandas.testing.assert_series_equal` and :func:`pandas.testing.assert_frame_equal` where extension dtypes were not ignored when ``check_dtypes`` was set to ``False`` (:issue:`35715`)
+- Bug in :meth:`to_timedelta` fails when ``arg`` is a :class:`Series` with ``Int64`` dtype containing null values (:issue:`35574`)
+- Bug in ``.groupby(..).rolling(..)`` where passing ``closed`` with column selection would raise a ``ValueError`` (:issue:`35549`)
+- Bug in :class:`DataFrame` constructor failing to raise ``ValueError`` in some cases when ``data`` and ``index`` have mismatched lengths (:issue:`33437`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_111.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.0..v1.1.1
diff --git a/pandas/doc/source/whatsnew/v1.1.2.rst b/pandas/doc/source/whatsnew/v1.1.2.rst
new file mode 100644
index 0000000000000..81b8e7df11625
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.2.rst
@@ -0,0 +1,64 @@
+.. _whatsnew_112:
+
+What's new in 1.1.2 (September 8, 2020)
+---------------------------------------
+
+These are the changes in pandas 1.1.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_112.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Regression in :meth:`DatetimeIndex.intersection` incorrectly raising ``AssertionError`` when intersecting against a list (:issue:`35876`)
+- Fix regression in updating a column inplace (e.g. using ``df['col'].fillna(.., inplace=True)``) (:issue:`35731`)
+- Fix regression in :meth:`DataFrame.append` mixing tz-aware and tz-naive datetime columns (:issue:`35460`)
+- Performance regression for :meth:`RangeIndex.format` (:issue:`35712`)
+- Regression where :meth:`MultiIndex.get_loc` would return a slice spanning the full index when passed an empty list (:issue:`35878`)
+- Fix regression in invalid cache after an indexing operation; this can manifest when setting which does not update the data (:issue:`35521`)
+- Regression in :meth:`DataFrame.replace` where a ``TypeError`` would be raised when attempting to replace elements of type :class:`Interval` (:issue:`35931`)
+- Fix regression in pickle roundtrip of the ``closed`` attribute of :class:`IntervalIndex` (:issue:`35658`)
+- Fixed regression in :meth:`DataFrameGroupBy.agg` where a ``ValueError: buffer source array is read-only`` would be raised when the underlying array is read-only (:issue:`36014`)
+- Fixed regression in :meth:`Series.groupby.rolling` number of levels of :class:`MultiIndex` in input was compressed to one (:issue:`36018`)
+- Fixed regression in :class:`DataFrameGroupBy` on an empty :class:`DataFrame` (:issue:`36197`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_112.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :meth:`DataFrame.eval` with ``object`` dtype column binary operations (:issue:`35794`)
+- Bug in :class:`Series` constructor raising a ``TypeError`` when constructing sparse datetime64 dtypes (:issue:`35762`)
+- Bug in :meth:`DataFrame.apply` with ``result_type="reduce"`` returning with incorrect index (:issue:`35683`)
+- Bug in :meth:`Series.astype` and :meth:`DataFrame.astype` not respecting the ``errors`` argument when set to ``"ignore"`` for extension dtypes (:issue:`35471`)
+- Bug in :meth:`DateTimeIndex.format` and :meth:`PeriodIndex.format` with ``name=True`` setting the first item to ``"None"`` where it should be ``""`` (:issue:`35712`)
+- Bug in :meth:`Float64Index.__contains__` incorrectly raising ``TypeError`` instead of returning ``False`` (:issue:`35788`)
+- Bug in :class:`Series` constructor incorrectly raising a ``TypeError`` when passed an ordered set (:issue:`36044`)
+- Bug in :meth:`Series.dt.isocalendar` and :meth:`DatetimeIndex.isocalendar` that returned incorrect year for certain dates (:issue:`36032`)
+- Bug in :class:`DataFrame` indexing returning an incorrect :class:`Series` in some cases when the series has been altered and a cache not invalidated (:issue:`33675`)
+- Bug in :meth:`DataFrame.corr` causing subsequent indexing lookups to be incorrect (:issue:`35882`)
+- Bug in :meth:`import_optional_dependency` returning incorrect package names in cases where package name is different from import name (:issue:`35948`)
+- Bug when setting empty :class:`DataFrame` column to a :class:`Series` in preserving name of index in frame (:issue:`31368`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_112.other:
+
+Other
+~~~~~
+- :meth:`factorize` now supports ``na_sentinel=None`` to include NaN in the uniques of the values and remove ``dropna`` keyword which was unintentionally exposed to public facing API in 1.1 version from :meth:`factorize` (:issue:`35667`)
+- :meth:`DataFrame.plot` and :meth:`Series.plot` raise ``UserWarning`` about usage of ``FixedFormatter`` and ``FixedLocator`` (:issue:`35684` and :issue:`35945`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_112.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.1..v1.1.2
diff --git a/pandas/doc/source/whatsnew/v1.1.3.rst b/pandas/doc/source/whatsnew/v1.1.3.rst
new file mode 100644
index 0000000000000..e752eb54d0c15
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.3.rst
@@ -0,0 +1,78 @@
+.. _whatsnew_113:
+
+What's new in 1.1.3 (October 5, 2020)
+-------------------------------------
+
+These are the changes in pandas 1.1.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+Enhancements
+~~~~~~~~~~~~
+
+Added support for new Python version
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 1.1.3 now supports Python 3.9 (:issue:`36296`).
+
+Development Changes
+^^^^^^^^^^^^^^^^^^^
+
+- The minimum version of Cython is now the most recent bug-fix version (0.29.21) (:issue:`36296`).
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_113.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.agg`, :meth:`DataFrame.apply`, :meth:`Series.agg`, and :meth:`Series.apply` where internal suffix is exposed to the users when no relabelling is applied (:issue:`36189`)
+- Fixed regression in :class:`IntegerArray` unary plus and minus operations raising a ``TypeError`` (:issue:`36063`)
+- Fixed regression when adding a :meth:`timedelta_range` to a :class:`Timestamp` raised a ``ValueError`` (:issue:`35897`)
+- Fixed regression in :meth:`Series.__getitem__` incorrectly raising when the input was a tuple (:issue:`35534`)
+- Fixed regression in :meth:`Series.__getitem__` incorrectly raising when the input was a frozenset (:issue:`35747`)
+- Fixed regression in modulo of :class:`Index`, :class:`Series` and :class:`DataFrame` using ``numexpr`` using C not Python semantics (:issue:`36047`, :issue:`36526`)
+- Fixed regression in :meth:`read_excel` with ``engine="odf"`` caused ``UnboundLocalError`` in some cases where cells had nested child nodes (:issue:`36122`, :issue:`35802`)
+- Fixed regression in :meth:`DataFrame.replace` inconsistent replace when using a float in the replace method (:issue:`35376`)
+- Fixed regression in :meth:`Series.loc` on a :class:`Series` with a :class:`MultiIndex` containing :class:`Timestamp` raising ``InvalidIndexError`` (:issue:`35858`)
+- Fixed regression in :class:`DataFrame` and :class:`Series` comparisons between numeric arrays and strings (:issue:`35700`, :issue:`36377`)
+- Fixed regression in :meth:`DataFrame.apply` with ``raw=True`` and user-function returning string (:issue:`35940`)
+- Fixed regression when setting empty :class:`DataFrame` column to a :class:`Series` in preserving name of index in frame (:issue:`36527`)
+- Fixed regression in :class:`Period` incorrect value for ordinal over the maximum timestamp (:issue:`36430`)
+- Fixed regression in :func:`read_table` raised ``ValueError`` when ``delim_whitespace`` was set to ``True`` (:issue:`35958`)
+- Fixed regression in :meth:`Series.dt.normalize` when normalizing pre-epoch dates the result was shifted one day (:issue:`36294`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_113.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :func:`read_spss` where passing a ``pathlib.Path`` as ``path`` would raise a ``TypeError`` (:issue:`33666`)
+- Bug in :meth:`Series.str.startswith` and :meth:`Series.str.endswith` with ``category`` dtype not propagating ``na`` parameter (:issue:`36241`)
+- Bug in :class:`Series` constructor where integer overflow would occur for sufficiently large scalar inputs when an index was provided (:issue:`36291`)
+- Bug in :meth:`DataFrame.sort_values` raising an ``AttributeError`` when sorting on a key that casts column to categorical dtype (:issue:`36383`)
+- Bug in :meth:`DataFrame.stack` raising a ``ValueError`` when stacking :class:`MultiIndex` columns based on position when the levels had duplicate names (:issue:`36353`)
+- Bug in :meth:`Series.astype` showing too much precision when casting from ``np.float32`` to string dtype (:issue:`36451`)
+- Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` when using ``NaN`` and a row length above 1,000,000 (:issue:`22205`)
+- Bug in :func:`cut` raising a ``ValueError`` when passed a :class:`Series` of labels with ``ordered=False`` (:issue:`36603`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_113.other:
+
+Other
+~~~~~
+- Reverted enhancement added in pandas-1.1.0 where :func:`timedelta_range` infers a frequency when passed ``start``, ``stop``, and ``periods`` (:issue:`32377`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_113.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.2..v1.1.3
diff --git a/pandas/doc/source/whatsnew/v1.1.4.rst b/pandas/doc/source/whatsnew/v1.1.4.rst
new file mode 100644
index 0000000000000..c6fda04070240
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.4.rst
@@ -0,0 +1,55 @@
+.. _whatsnew_114:
+
+What's new in 1.1.4 (October 30, 2020)
+--------------------------------------
+
+These are the changes in pandas 1.1.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_114.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :func:`read_csv` raising a ``ValueError`` when ``names`` was of type ``dict_keys`` (:issue:`36928`)
+- Fixed regression in :func:`read_csv` with more than 1M rows and specifying a ``index_col`` argument (:issue:`37094`)
+- Fixed regression where attempting to mutate a :class:`DateOffset` object would no longer raise an ``AttributeError`` (:issue:`36940`)
+- Fixed regression where :meth:`DataFrame.agg` would fail with :exc:`TypeError` when passed positional arguments to be passed on to the aggregation function (:issue:`36948`).
+- Fixed regression in :class:`RollingGroupby` with ``sort=False`` not being respected (:issue:`36889`)
+- Fixed regression in :meth:`Series.astype` converting ``None`` to ``"nan"`` when casting to string (:issue:`36904`)
+- Fixed regression in :meth:`Series.rank` method failing for read-only data (:issue:`37290`)
+- Fixed regression in :class:`RollingGroupby` causing a segmentation fault with Index of dtype object (:issue:`36727`)
+- Fixed regression in :meth:`DataFrame.resample(...).apply(...)` raised ``AttributeError`` when input was a :class:`DataFrame` and only a :class:`Series` was evaluated (:issue:`36951`)
+- Fixed regression in ``DataFrame.groupby(..).std()`` with nullable integer dtype (:issue:`37415`)
+- Fixed regression in :class:`PeriodDtype` comparing both equal and unequal to its string representation (:issue:`37265`)
+- Fixed regression where slicing :class:`DatetimeIndex` raised :exc:`AssertionError` on irregular time series with ``pd.NaT`` or on unsorted indices (:issue:`36953` and :issue:`35509`)
+- Fixed regression in certain offsets (:meth:`pd.offsets.Day() <pandas.tseries.offsets.Day>` and below) no longer being hashable (:issue:`37267`)
+- Fixed regression in :class:`StataReader` which required ``chunksize`` to be manually set when using an iterator to read a dataset (:issue:`37280`)
+- Fixed regression in setitem with :meth:`DataFrame.iloc` which raised error when trying to set a value while filtering with a boolean list (:issue:`36741`)
+- Fixed regression in setitem with a Series getting aligned before setting the values (:issue:`37427`)
+- Fixed regression in :attr:`MultiIndex.is_monotonic_increasing` returning wrong results with ``NaN`` in at least one of the levels (:issue:`37220`)
+- Fixed regression in inplace arithmetic operation (``+=``) on a Series not updating the parent DataFrame/Series (:issue:`36373`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_114.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
+- Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` raising a ``ValueError`` when the target was read-only (:issue:`37174`)
+- Bug in :meth:`.DataFrameGroupBy.fillna` and :meth:`.SeriesGroupBy.fillna` that introduced a performance regression after 1.0.5 (:issue:`36757`)
+- Bug in :meth:`DataFrame.info` was raising a ``KeyError`` when the DataFrame has integer column names (:issue:`37245`)
+- Bug in :meth:`DataFrameGroupby.apply` would drop a :class:`CategoricalIndex` when grouped on (:issue:`35792`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_114.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.3..v1.1.4
diff --git a/pandas/doc/source/whatsnew/v1.1.5.rst b/pandas/doc/source/whatsnew/v1.1.5.rst
new file mode 100644
index 0000000000000..4c822ae049936
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.1.5.rst
@@ -0,0 +1,56 @@
+.. _whatsnew_115:
+
+What's new in 1.1.5 (December 07, 2020)
+---------------------------------------
+
+These are the changes in pandas 1.1.5. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_115.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in addition of a timedelta-like scalar to a :class:`DatetimeIndex` raising incorrectly (:issue:`37295`)
+- Fixed regression in :meth:`Series.groupby` raising when the :class:`Index` of the :class:`Series` had a tuple as its name (:issue:`37755`)
+- Fixed regression in :meth:`DataFrame.loc` and :meth:`Series.loc` for ``__setitem__`` when one-dimensional tuple was given to select from :class:`MultiIndex` (:issue:`37711`)
+- Fixed regression in inplace operations on :class:`Series` with ``ExtensionDtype`` with NumPy dtyped operand (:issue:`37910`)
+- Fixed regression in metadata propagation for ``groupby`` iterator (:issue:`37343`)
+- Fixed regression in :class:`MultiIndex` constructed from a :class:`DatetimeIndex` not retaining frequency (:issue:`35563`)
+- Fixed regression in :class:`Index` constructor raising a ``AttributeError`` when passed a :class:`SparseArray` with datetime64 values (:issue:`35843`)
+- Fixed regression in :meth:`DataFrame.unstack` with columns with integer dtype (:issue:`37115`)
+- Fixed regression in indexing on a :class:`Series` with ``CategoricalDtype`` after unpickling (:issue:`37631`)
+- Fixed regression in :meth:`DataFrame.groupby` aggregation with out-of-bounds datetime objects in an object-dtype column (:issue:`36003`)
+- Fixed regression in ``df.groupby(..).rolling(..)`` with the resulting :class:`MultiIndex` when grouping by a label that is in the index (:issue:`37641`)
+- Fixed regression in :meth:`DataFrame.fillna` not filling ``NaN`` after other operations such as :meth:`DataFrame.pivot` (:issue:`36495`).
+- Fixed performance regression in ``df.groupby(..).rolling(..)`` (:issue:`38038`)
+- Fixed regression in :meth:`MultiIndex.intersection` returning duplicates when at least one of the indexes had duplicates (:issue:`36915`)
+- Fixed regression in :meth:`.DataFrameGroupBy.first`, :meth:`.SeriesGroupBy.first`, :meth:`.DataFrameGroupBy.last`, and :meth:`.SeriesGroupBy.last` where ``None`` was considered a non-NA value (:issue:`38286`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_115.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in pytables methods in python 3.9 (:issue:`38041`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_115.other:
+
+Other
+~~~~~
+- Only set ``-Werror`` as a compiler flag in the CI jobs (:issue:`33315`, :issue:`33314`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_115.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.4..v1.1.5|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.2.0.rst b/pandas/doc/source/whatsnew/v1.2.0.rst
new file mode 100644
index 0000000000000..12ab4f27d1e62
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.0.rst
@@ -0,0 +1,883 @@
+.. _whatsnew_120:
+
+What's new in 1.2.0 (December 26, 2020)
+---------------------------------------
+
+These are the changes in pandas 1.2.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. warning::
+
+   The `xlwt <https://xlwt.readthedocs.io/en/latest/>`_ package for writing old-style ``.xls``
+   excel files is no longer maintained.
+   The `xlrd <https://xlrd.readthedocs.io/en/latest/>`_ package is now only for reading
+   old-style ``.xls`` files.
+
+   Previously, the default argument ``engine=None`` to :func:`~pandas.read_excel`
+   would result in using the ``xlrd`` engine in many cases, including new
+   Excel 2007+ (``.xlsx``) files.
+   If `openpyxl <https://openpyxl.readthedocs.io/en/stable/>`_  is installed,
+   many of these cases will now default to using the ``openpyxl`` engine.
+   See the :func:`read_excel` documentation for more details.
+
+   Thus, it is strongly encouraged to install ``openpyxl`` to read Excel 2007+
+   (``.xlsx``) files.
+   **Please do not report issues when using ``xlrd`` to read ``.xlsx`` files.**
+   This is no longer supported, switch to using ``openpyxl`` instead.
+
+   Attempting to use the ``xlwt`` engine will raise a ``FutureWarning``
+   unless the option :attr:`io.excel.xls.writer` is set to ``"xlwt"``.
+   While this option is now deprecated and will also raise a ``FutureWarning``,
+   it can be globally set and the warning suppressed. Users are recommended to
+   write ``.xlsx`` files using the ``openpyxl`` engine instead.
+
+.. ---------------------------------------------------------------------------
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_120.duplicate_labels:
+
+Optionally disallow duplicate labels
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Series` and :class:`DataFrame` can now be created with ``allows_duplicate_labels=False`` flag to
+control whether the index or columns can contain duplicate labels (:issue:`28394`). This can be used to
+prevent accidental introduction of duplicate labels, which can affect downstream operations.
+
+By default, duplicates continue to be allowed.
+
+.. code-block:: ipython
+
+    In [1]: pd.Series([1, 2], index=['a', 'a'])
+    Out[1]:
+    a    1
+    a    2
+    Length: 2, dtype: int64
+
+    In [2]: pd.Series([1, 2], index=['a', 'a']).set_flags(allows_duplicate_labels=False)
+    ...
+    DuplicateLabelError: Index has duplicates.
+          positions
+    label
+    a        [0, 1]
+
+pandas will propagate the ``allows_duplicate_labels`` property through many operations.
+
+.. code-block:: ipython
+
+    In [3]: a = (
+       ...:     pd.Series([1, 2], index=['a', 'b'])
+       ...:       .set_flags(allows_duplicate_labels=False)
+       ...: )
+
+    In [4]: a
+    Out[4]:
+    a    1
+    b    2
+    Length: 2, dtype: int64
+
+    # An operation introducing duplicates
+    In [5]: a.reindex(['a', 'b', 'a'])
+    ...
+    DuplicateLabelError: Index has duplicates.
+          positions
+    label
+    a        [0, 2]
+
+    [1 rows x 1 columns]
+
+.. warning::
+
+   This is an experimental feature. Currently, many methods fail to
+   propagate the ``allows_duplicate_labels`` value. In future versions
+   it is expected that every method taking or returning one or more
+   DataFrame or Series objects will propagate ``allows_duplicate_labels``.
+
+See :ref:`duplicates` for more.
+
+The ``allows_duplicate_labels`` flag is stored in the new :attr:`DataFrame.flags`
+attribute. This stores global attributes that apply to the *pandas object*. This
+differs from :attr:`DataFrame.attrs`, which stores information that applies to
+the dataset.
+
+Passing arguments to fsspec backends
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Many read/write functions have acquired the ``storage_options`` optional argument,
+to pass a dictionary of parameters to the storage backend. This allows, for
+example, for passing credentials to S3 and GCS storage. The details of what
+parameters can be passed to which backends can be found in the documentation
+of the individual storage backends (detailed from the fsspec docs for
+`builtin implementations`_ and linked to `external ones`_). See
+Section :ref:`io.remote`.
+
+:issue:`35655` added fsspec support (including ``storage_options``)
+for reading excel files.
+
+.. _builtin implementations: https://filesystem-spec.readthedocs.io/en/latest/api.html#built-in-implementations
+.. _external ones: https://filesystem-spec.readthedocs.io/en/latest/api.html#other-known-implementations
+
+.. _whatsnew_120.binary_handle_to_csv:
+
+Support for binary file handles in ``to_csv``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`to_csv` supports file handles in binary mode (:issue:`19827` and :issue:`35058`)
+with ``encoding`` (:issue:`13068` and :issue:`23854`) and ``compression`` (:issue:`22555`).
+If pandas does not automatically detect whether the file handle is opened in binary or text mode,
+it is necessary to provide ``mode="wb"``.
+
+For example:
+
+.. ipython:: python
+
+   import io
+
+   data = pd.DataFrame([0, 1, 2])
+   buffer = io.BytesIO()
+   data.to_csv(buffer, encoding="utf-8", compression="gzip")
+
+Support for short caption and table position in ``to_latex``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.to_latex` now allows one to specify
+a floating table position (:issue:`35281`)
+and a short caption (:issue:`36267`).
+
+The keyword ``position`` has been added to set the position.
+
+.. ipython:: python
+   :okwarning:
+
+   data = pd.DataFrame({'a': [1, 2], 'b': [3, 4]})
+   table = data.to_latex(position='ht')
+   print(table)
+
+Usage of the keyword ``caption`` has been extended.
+Besides taking a single string as an argument,
+one can optionally provide a tuple ``(full_caption, short_caption)``
+to add a short caption macro.
+
+.. ipython:: python
+   :okwarning:
+
+   data = pd.DataFrame({'a': [1, 2], 'b': [3, 4]})
+   table = data.to_latex(caption=('the full long caption', 'short caption'))
+   print(table)
+
+.. _whatsnew_120.read_csv_table_precision_default:
+
+Change in default floating precision for ``read_csv`` and ``read_table``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+For the C parsing engine, the methods :meth:`read_csv` and :meth:`read_table` previously defaulted to a parser that
+could read floating point numbers slightly incorrectly with respect to the last bit in precision.
+The option ``floating_precision="high"`` has always been available to avoid this issue.
+Beginning with this version, the default is now to use the more accurate parser by making
+``floating_precision=None`` correspond to the high precision parser, and the new option
+``floating_precision="legacy"`` to use the legacy parser. The change to using the higher precision
+parser by default should have no impact on performance. (:issue:`17154`)
+
+.. _whatsnew_120.floating:
+
+Experimental nullable data types for float data
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've added :class:`Float32Dtype` / :class:`Float64Dtype` and :class:`~arrays.FloatingArray`.
+These are extension data types dedicated to floating point data that can hold the
+``pd.NA`` missing value indicator (:issue:`32265`, :issue:`34307`).
+
+While the default float data type already supports missing values using ``np.nan``,
+these new data types use ``pd.NA`` (and its corresponding behavior) as the missing
+value indicator, in line with the already existing nullable :ref:`integer <integer_na>`
+and :ref:`boolean <boolean>` data types.
+
+One example where the behavior of ``np.nan`` and ``pd.NA`` is different is
+comparison operations:
+
+.. ipython:: python
+
+  # the default NumPy float64 dtype
+  s1 = pd.Series([1.5, None])
+  s1
+  s1 > 1
+
+.. ipython:: python
+
+  # the new nullable float64 dtype
+  s2 = pd.Series([1.5, None], dtype="Float64")
+  s2
+  s2 > 1
+
+See the :ref:`missing_data.NA` doc section for more details on the behavior
+when using the ``pd.NA`` missing value indicator.
+
+As shown above, the dtype can be specified using the "Float64" or "Float32"
+string (capitalized to distinguish it from the default "float64" data type).
+Alternatively, you can also use the dtype object:
+
+.. ipython:: python
+
+   pd.Series([1.5, None], dtype=pd.Float32Dtype())
+
+Operations with the existing integer or boolean nullable data types that
+give float results will now also use the nullable floating data types (:issue:`38178`).
+
+.. warning::
+
+   Experimental: the new floating data types are currently experimental, and their
+   behavior or API may still change without warning. Especially the behavior
+   regarding NaN (distinct from NA missing values) is subject to change.
+
+.. _whatsnew_120.index_name_preservation:
+
+Index/column name preservation when aggregating
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When aggregating using :meth:`concat` or the :class:`DataFrame` constructor, pandas
+will now attempt to preserve index and column names whenever possible (:issue:`35847`).
+In the case where all inputs share a common name, this name will be assigned to the
+result. When the input names do not all agree, the result will be unnamed. Here is an
+example where the index name is preserved:
+
+.. ipython:: python
+
+    idx = pd.Index(range(5), name='abc')
+    ser = pd.Series(range(5, 10), index=idx)
+    pd.concat({'x': ser[1:], 'y': ser[:-1]}, axis=1)
+
+The same is true for :class:`MultiIndex`, but the logic is applied separately on a
+level-by-level basis.
+
+.. _whatsnew_120.groupby_ewm:
+
+GroupBy supports EWM operations directly
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`.DataFrameGroupBy` now supports exponentially weighted window operations directly (:issue:`16037`).
+
+.. ipython:: python
+
+    df = pd.DataFrame({'A': ['a', 'b', 'a', 'b'], 'B': range(4)})
+    df
+    df.groupby('A').ewm(com=1.0).mean()
+
+Additionally ``mean`` supports execution via `Numba <https://numba.pydata.org/>`__ with
+the  ``engine`` and ``engine_kwargs`` arguments. Numba must be installed as an optional dependency
+to use this feature.
+
+.. _whatsnew_120.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- Added ``day_of_week`` (compatibility alias ``dayofweek``) property to :class:`Timestamp`, :class:`.DatetimeIndex`, :class:`Period`, :class:`PeriodIndex` (:issue:`9605`)
+- Added ``day_of_year`` (compatibility alias ``dayofyear``) property to :class:`Timestamp`, :class:`.DatetimeIndex`, :class:`Period`, :class:`PeriodIndex` (:issue:`9605`)
+- Added :meth:`~DataFrame.set_flags` for setting table-wide flags on a Series or DataFrame (:issue:`28394`)
+- :meth:`DataFrame.applymap` now supports ``na_action`` (:issue:`23803`)
+- :class:`Index` with object dtype supports division and multiplication (:issue:`34160`)
+- :meth:`io.sql.get_schema` now supports a ``schema`` keyword argument that will add a schema into the create table statement (:issue:`28486`)
+- :meth:`DataFrame.explode` and :meth:`Series.explode` now support exploding of sets (:issue:`35614`)
+- :meth:`DataFrame.hist` now supports time series (datetime) data (:issue:`32590`)
+- :meth:`.Styler.set_table_styles` now allows the direct styling of rows and columns and can be chained (:issue:`35607`)
+- :class:`.Styler` now allows direct CSS class name addition to individual data cells (:issue:`36159`)
+- :meth:`.Rolling.mean` and :meth:`.Rolling.sum` use Kahan summation to calculate the mean to avoid numerical problems (:issue:`10319`, :issue:`11645`, :issue:`13254`, :issue:`32761`, :issue:`36031`)
+- :meth:`.DatetimeIndex.searchsorted`, :meth:`.TimedeltaIndex.searchsorted`, :meth:`PeriodIndex.searchsorted`, and :meth:`Series.searchsorted` with datetime-like dtypes will now try to cast string arguments (list-like and scalar) to the matching datetime-like type (:issue:`36346`)
+- Added methods :meth:`IntegerArray.prod`, :meth:`IntegerArray.min`, and :meth:`IntegerArray.max` (:issue:`33790`)
+- Calling a NumPy ufunc on a ``DataFrame`` with extension types now preserves the extension types when possible (:issue:`23743`)
+- Calling a binary-input NumPy ufunc on multiple ``DataFrame`` objects now aligns, matching the behavior of binary operations and ufuncs on ``Series`` (:issue:`23743`).
+  This change has been reverted in pandas 1.2.1, and the behaviour to not align DataFrames
+  is deprecated instead, see the :ref:`the 1.2.1 release notes <whatsnew_121.ufunc_deprecation>`.
+- Where possible :meth:`RangeIndex.difference` and :meth:`RangeIndex.symmetric_difference` will return :class:`RangeIndex` instead of :class:`Int64Index` (:issue:`36564`)
+- :meth:`DataFrame.to_parquet` now supports :class:`MultiIndex` for columns in parquet format (:issue:`34777`)
+- :func:`read_parquet` gained a ``use_nullable_dtypes=True`` option to use nullable dtypes that use ``pd.NA`` as missing value indicator where possible for the resulting DataFrame (default is ``False``, and only applicable for ``engine="pyarrow"``) (:issue:`31242`)
+- Added :meth:`.Rolling.sem` and :meth:`Expanding.sem` to compute the standard error of the mean (:issue:`26476`)
+- :meth:`.Rolling.var` and :meth:`.Rolling.std` use Kahan summation and Welford's Method to avoid numerical issues (:issue:`37051`)
+- :meth:`DataFrame.corr` and :meth:`DataFrame.cov` use Welford's Method to avoid numerical issues (:issue:`37448`)
+- :meth:`DataFrame.plot` now recognizes ``xlabel`` and ``ylabel`` arguments for plots of type ``scatter`` and ``hexbin`` (:issue:`37001`)
+- :class:`DataFrame` now supports the ``divmod`` operation (:issue:`37165`)
+- :meth:`DataFrame.to_parquet` now returns a ``bytes`` object when no ``path`` argument is passed (:issue:`37105`)
+- :class:`.Rolling` now supports the ``closed`` argument for fixed windows (:issue:`34315`)
+- :class:`.DatetimeIndex` and :class:`Series` with ``datetime64`` or ``datetime64tz`` dtypes now support ``std`` (:issue:`37436`)
+- :class:`Window` now supports all Scipy window types in ``win_type`` with flexible keyword argument support (:issue:`34556`)
+- :meth:`testing.assert_index_equal` now has a ``check_order`` parameter that allows indexes to be checked in an order-insensitive manner (:issue:`37478`)
+- :func:`read_csv` supports memory-mapping for compressed files (:issue:`37621`)
+- Add support for ``min_count`` keyword for :meth:`DataFrame.groupby` and :meth:`DataFrame.resample` for functions ``min``, ``max``, ``first`` and ``last`` (:issue:`37821`, :issue:`37768`)
+- Improve error reporting for :meth:`DataFrame.merge` when invalid merge column definitions were given (:issue:`16228`)
+- Improve numerical stability for :meth:`.Rolling.skew`, :meth:`.Rolling.kurt`, :meth:`Expanding.skew` and :meth:`Expanding.kurt` through implementation of Kahan summation (:issue:`6929`)
+- Improved error reporting for subsetting columns of a :class:`.DataFrameGroupBy` with ``axis=1`` (:issue:`37725`)
+- Implement method ``cross`` for :meth:`DataFrame.merge` and :meth:`DataFrame.join` (:issue:`5401`)
+- When :func:`read_csv`, :func:`read_sas` and :func:`read_json` are called with ``chunksize``/``iterator`` they can be used in a ``with`` statement as they return context-managers (:issue:`38225`)
+- Augmented the list of named colors available for styling Excel exports, enabling all of CSS4 colors (:issue:`38247`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_120.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+Consistency of DataFrame Reductions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+:meth:`DataFrame.any` and :meth:`DataFrame.all` with ``bool_only=True`` now
+determines whether to exclude object-dtype columns on a column-by-column basis,
+instead of checking if *all* object-dtype columns can be considered boolean.
+
+This prevents pathological behavior where applying the reduction on a subset
+of columns could result in a larger Series result. See (:issue:`37799`).
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": ["foo", "bar"], "B": [True, False]}, dtype=object)
+    df["C"] = pd.Series([True, True])
+
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.all(bool_only=True)
+    Out[5]:
+    C    True
+    dtype: bool
+
+    In [6]: df[["B", "C"]].all(bool_only=True)
+    Out[6]:
+    B    False
+    C    True
+    dtype: bool
+
+*New behavior*:
+
+.. ipython:: python
+   :okwarning:
+
+    In [5]: df.all(bool_only=True)
+
+    In [6]: df[["B", "C"]].all(bool_only=True)
+
+
+Other DataFrame reductions with ``numeric_only=None`` will also avoid
+this pathological behavior (:issue:`37827`):
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": [0, 1, 2], "B": ["a", "b", "c"]}, dtype=object)
+
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.mean()
+    Out[3]: Series([], dtype: float64)
+
+    In [4]: df[["A"]].mean()
+    Out[4]:
+    A    1.0
+    dtype: float64
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.mean()
+    Out[3]:
+    A    1.0
+    dtype: float64
+
+    In [4]: df[["A"]].mean()
+    Out[4]:
+    A    1.0
+    dtype: float64
+
+Moreover, DataFrame reductions with ``numeric_only=None`` will now be
+consistent with their Series counterparts.  In particular, for
+reductions where the Series method raises ``TypeError``, the
+DataFrame reduction will now consider that column non-numeric
+instead of casting to a NumPy array which may have different semantics (:issue:`36076`,
+:issue:`28949`, :issue:`21020`).
+
+.. ipython:: python
+   :okwarning:
+
+    ser = pd.Series([0, 1], dtype="category", name="A")
+    df = ser.to_frame()
+
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.any()
+    Out[5]:
+    A    True
+    dtype: bool
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.any()
+    Out[5]: Series([], dtype: bool)
+
+
+.. _whatsnew_120.api_breaking.python:
+
+Increased minimum version for Python
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 1.2.0 supports Python 3.7.1 and higher (:issue:`35214`).
+
+.. _whatsnew_120.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Some minimum supported versions of dependencies were updated (:issue:`35214`).
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.16.5          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| pytz            | 2017.3          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| python-dateutil | 2.7.3           |    X     |         |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.2.1           |          |         |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.6.8           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pytest (dev)    | 5.0.1           |          |    X    |
++-----------------+-----------------+----------+---------+
+| mypy (dev)      | 0.782           |          |    X    |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  | 4.6.0           |         |
++-----------------+-----------------+---------+
+| fastparquet     | 0.3.2           |         |
++-----------------+-----------------+---------+
+| fsspec          | 0.7.4           |         |
++-----------------+-----------------+---------+
+| gcsfs           | 0.6.0           |         |
++-----------------+-----------------+---------+
+| lxml            | 4.3.0           |    X    |
++-----------------+-----------------+---------+
+| matplotlib      | 2.2.3           |    X    |
++-----------------+-----------------+---------+
+| numba           | 0.46.0          |         |
++-----------------+-----------------+---------+
+| openpyxl        | 2.6.0           |    X    |
++-----------------+-----------------+---------+
+| pyarrow         | 0.15.0          |    X    |
++-----------------+-----------------+---------+
+| pymysql         | 0.7.11          |    X    |
++-----------------+-----------------+---------+
+| pytables        | 3.5.1           |    X    |
++-----------------+-----------------+---------+
+| s3fs            | 0.4.0           |         |
++-----------------+-----------------+---------+
+| scipy           | 1.2.0           |         |
++-----------------+-----------------+---------+
+| sqlalchemy      | 1.2.8           |    X    |
++-----------------+-----------------+---------+
+| xarray          | 0.12.3          |    X    |
++-----------------+-----------------+---------+
+| xlrd            | 1.2.0           |    X    |
++-----------------+-----------------+---------+
+| xlsxwriter      | 1.0.2           |    X    |
++-----------------+-----------------+---------+
+| xlwt            | 1.3.0           |    X    |
++-----------------+-----------------+---------+
+| pandas-gbq      | 0.12.0          |         |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_120.api.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- Sorting in descending order is now stable for :meth:`Series.sort_values` and :meth:`Index.sort_values` for Datetime-like :class:`Index` subclasses. This will affect sort order when sorting a DataFrame on multiple columns, sorting with a key function that produces duplicates, or requesting the sorting index when using :meth:`Index.sort_values`. When using :meth:`Series.value_counts`, the count of missing values is no longer necessarily last in the list of duplicate counts. Instead, its position corresponds to the position in the original Series. When using :meth:`Index.sort_values` for Datetime-like :class:`Index` subclasses, NaTs ignored the ``na_position`` argument and were sorted to the beginning. Now they respect ``na_position``, the default being ``last``, same as other :class:`Index` subclasses (:issue:`35992`)
+- Passing an invalid ``fill_value`` to :meth:`Categorical.take`, :meth:`.DatetimeArray.take`, :meth:`TimedeltaArray.take`, or :meth:`PeriodArray.take` now raises a ``TypeError`` instead of a ``ValueError`` (:issue:`37733`)
+- Passing an invalid ``fill_value`` to :meth:`Series.shift` with a ``CategoricalDtype`` now raises a ``TypeError`` instead of a ``ValueError`` (:issue:`37733`)
+- Passing an invalid value to :meth:`IntervalIndex.insert` or :meth:`CategoricalIndex.insert` now raises a ``TypeError`` instead of a ``ValueError`` (:issue:`37733`)
+- Attempting to reindex a Series with a :class:`CategoricalIndex` with an invalid ``fill_value`` now raises a ``TypeError`` instead of a ``ValueError`` (:issue:`37733`)
+- :meth:`CategoricalIndex.append` with an index that contains non-category values will now cast instead of raising ``TypeError`` (:issue:`38098`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_120.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+- Deprecated parameter ``inplace`` in :meth:`MultiIndex.set_codes` and :meth:`MultiIndex.set_levels` (:issue:`35626`)
+- Deprecated parameter ``dtype`` of method :meth:`~Index.copy` for all :class:`Index` subclasses. Use the :meth:`~Index.astype` method instead for changing dtype (:issue:`35853`)
+- Deprecated parameters ``levels`` and ``codes`` in :meth:`MultiIndex.copy`. Use the :meth:`~MultiIndex.set_levels` and :meth:`~MultiIndex.set_codes` methods instead (:issue:`36685`)
+- Date parser functions :func:`~pandas.io.date_converters.parse_date_time`, :func:`~pandas.io.date_converters.parse_date_fields`, :func:`~pandas.io.date_converters.parse_all_fields` and :func:`~pandas.io.date_converters.generic_parser` from ``pandas.io.date_converters`` are deprecated and will be removed in a future version; use :func:`to_datetime` instead (:issue:`35741`)
+- :meth:`DataFrame.lookup` is deprecated and will be removed in a future version, use :meth:`DataFrame.melt` and :meth:`DataFrame.loc` instead (:issue:`35224`)
+- The method :meth:`Index.to_native_types` is deprecated. Use ``.astype(str)`` instead (:issue:`28867`)
+- Deprecated indexing :class:`DataFrame` rows with a single datetime-like string as ``df[string]`` (given the ambiguity whether it is indexing the rows or selecting a column), use ``df.loc[string]`` instead (:issue:`36179`)
+- Deprecated :meth:`Index.is_all_dates` (:issue:`27744`)
+- The default value of ``regex`` for :meth:`Series.str.replace` will change from ``True`` to ``False`` in a future release. In addition, single character regular expressions will *not* be treated as literal strings when ``regex=True`` is set (:issue:`24804`)
+- Deprecated automatic alignment on comparison operations between :class:`DataFrame` and :class:`Series`, do ``frame, ser = frame.align(ser, axis=1, copy=False)`` before e.g. ``frame == ser`` (:issue:`28759`)
+- :meth:`Rolling.count` with ``min_periods=None`` will default to the size of the window in a future version (:issue:`31302`)
+- Using "outer" ufuncs on DataFrames to return 4d ndarray is now deprecated. Convert to an ndarray first (:issue:`23743`)
+- Deprecated slice-indexing on tz-aware :class:`DatetimeIndex` with naive ``datetime`` objects, to match scalar indexing behavior (:issue:`36148`)
+- :meth:`Index.ravel` returning a ``np.ndarray`` is deprecated, in the future this will return a view on the same index (:issue:`19956`)
+- Deprecate use of strings denoting units with 'M', 'Y' or 'y' in :func:`~pandas.to_timedelta` (:issue:`36666`)
+- :class:`Index` methods ``&``, ``|``, and ``^`` behaving as the set operations :meth:`Index.intersection`, :meth:`Index.union`, and :meth:`Index.symmetric_difference`, respectively, are deprecated and in the future will behave as pointwise boolean operations matching :class:`Series` behavior.  Use the named set methods instead (:issue:`36758`)
+- :meth:`Categorical.is_dtype_equal` and :meth:`CategoricalIndex.is_dtype_equal` are deprecated, will be removed in a future version (:issue:`37545`)
+- :meth:`Series.slice_shift` and :meth:`DataFrame.slice_shift` are deprecated, use :meth:`Series.shift` or :meth:`DataFrame.shift` instead (:issue:`37601`)
+- Partial slicing on unordered :class:`.DatetimeIndex` objects with keys that are not in the index is deprecated and will be removed in a future version (:issue:`18531`)
+- The ``how`` keyword in :meth:`PeriodIndex.astype` is deprecated and will be removed in a future version, use ``index.to_timestamp(how=how)`` instead (:issue:`37982`)
+- Deprecated :meth:`Index.asi8` for :class:`Index` subclasses other than :class:`.DatetimeIndex`, :class:`.TimedeltaIndex`, and :class:`PeriodIndex` (:issue:`37877`)
+- The ``inplace`` parameter of :meth:`Categorical.remove_unused_categories` is deprecated and will be removed in a future version (:issue:`37643`)
+- The ``null_counts`` parameter of :meth:`DataFrame.info` is deprecated and replaced by ``show_counts``. It will be removed in a future version (:issue:`37999`)
+
+**Calling NumPy ufuncs on non-aligned DataFrames**
+
+Calling NumPy ufuncs on non-aligned DataFrames changed behaviour in pandas
+1.2.0 (to align the inputs before calling the ufunc), but this change is
+reverted in pandas 1.2.1. The behaviour to not align is now deprecated instead,
+see the :ref:`the 1.2.1 release notes <whatsnew_121.ufunc_deprecation>` for
+more details.
+
+.. ---------------------------------------------------------------------------
+
+
+.. _whatsnew_120.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Performance improvements when creating DataFrame or Series with dtype ``str`` or :class:`StringDtype` from array with many string elements (:issue:`36304`, :issue:`36317`, :issue:`36325`, :issue:`36432`, :issue:`37371`)
+- Performance improvement in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` with the ``numba`` engine (:issue:`35759`)
+- Performance improvements when creating :meth:`Series.map` from a huge dictionary (:issue:`34717`)
+- Performance improvement in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` with the ``numba`` engine (:issue:`36240`)
+- :class:`.Styler` uuid method altered to compress data transmission over web whilst maintaining reasonably low table collision probability (:issue:`36345`)
+- Performance improvement in :func:`to_datetime` with non-ns time unit for ``float`` ``dtype`` columns (:issue:`20445`)
+- Performance improvement in setting values on an :class:`IntervalArray` (:issue:`36310`)
+- The internal index method :meth:`~Index._shallow_copy` now makes the new index and original index share cached attributes, avoiding creating these again, if created on either. This can speed up operations that depend on creating copies of existing indexes (:issue:`36840`)
+- Performance improvement in :meth:`.RollingGroupby.count` (:issue:`35625`)
+- Small performance decrease to :meth:`.Rolling.min` and :meth:`.Rolling.max` for fixed windows (:issue:`36567`)
+- Reduced peak memory usage in :meth:`DataFrame.to_pickle` when using ``protocol=5`` in python 3.8+ (:issue:`34244`)
+- Faster ``dir`` calls when the object has many index labels, e.g. ``dir(ser)`` (:issue:`37450`)
+- Performance improvement in :class:`ExpandingGroupby` (:issue:`37064`)
+- Performance improvement in :meth:`Series.astype` and :meth:`DataFrame.astype` for :class:`Categorical` (:issue:`8628`)
+- Performance improvement in :meth:`DataFrame.groupby` for ``float`` ``dtype`` (:issue:`28303`), changes of the underlying hash-function can lead to changes in float based indexes sort ordering for ties (e.g. :meth:`Index.value_counts`)
+- Performance improvement in :meth:`pd.isin` for inputs with more than 1e6 elements (:issue:`36611`)
+- Performance improvement for :meth:`DataFrame.__setitem__` with list-like indexers (:issue:`37954`)
+- :meth:`read_json` now avoids reading entire file into memory when chunksize is specified (:issue:`34548`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_120.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- :meth:`Categorical.fillna` will always return a copy, validate a passed fill value regardless of whether there are any NAs to fill, and disallow an ``NaT`` as a fill value for numeric categories (:issue:`36530`)
+- Bug in :meth:`Categorical.__setitem__` that incorrectly raised when trying to set a tuple value (:issue:`20439`)
+- Bug in :meth:`CategoricalIndex.equals` incorrectly casting non-category entries to ``np.nan`` (:issue:`37667`)
+- Bug in :meth:`CategoricalIndex.where` incorrectly setting non-category entries to ``np.nan`` instead of raising ``TypeError`` (:issue:`37977`)
+- Bug in :meth:`Categorical.to_numpy` and ``np.array(categorical)`` with tz-aware ``datetime64`` categories incorrectly dropping the time zone information instead of casting to object dtype (:issue:`38136`)
+
+Datetime-like
+^^^^^^^^^^^^^
+- Bug in :meth:`DataFrame.combine_first` that would convert datetime-like column on other :class:`DataFrame` to integer when the column is not present in original :class:`DataFrame` (:issue:`28481`)
+- Bug in :attr:`.DatetimeArray.date` where a ``ValueError`` would be raised with a read-only backing array (:issue:`33530`)
+- Bug in ``NaT`` comparisons failing to raise ``TypeError`` on invalid inequality comparisons (:issue:`35046`)
+- Bug in :class:`.DateOffset` where attributes reconstructed from pickle files differ from original objects when input values exceed normal ranges (e.g. months=12) (:issue:`34511`)
+- Bug in :meth:`.DatetimeIndex.get_slice_bound` where ``datetime.date`` objects were not accepted or naive :class:`Timestamp` with a tz-aware :class:`.DatetimeIndex` (:issue:`35690`)
+- Bug in :meth:`.DatetimeIndex.slice_locs` where ``datetime.date`` objects were not accepted (:issue:`34077`)
+- Bug in :meth:`.DatetimeIndex.searchsorted`, :meth:`.TimedeltaIndex.searchsorted`, :meth:`PeriodIndex.searchsorted`, and :meth:`Series.searchsorted` with ``datetime64``, ``timedelta64`` or :class:`Period` dtype placement of ``NaT`` values being inconsistent with NumPy (:issue:`36176`, :issue:`36254`)
+- Inconsistency in :class:`.DatetimeArray`, :class:`.TimedeltaArray`, and :class:`.PeriodArray` method ``__setitem__`` casting arrays of strings to datetime-like scalars but not scalar strings (:issue:`36261`)
+- Bug in :meth:`.DatetimeArray.take` incorrectly allowing ``fill_value`` with a mismatched time zone (:issue:`37356`)
+- Bug in :class:`.DatetimeIndex.shift` incorrectly raising when shifting empty indexes (:issue:`14811`)
+- :class:`Timestamp` and :class:`.DatetimeIndex` comparisons between tz-aware and tz-naive objects now follow the standard library ``datetime`` behavior, returning ``True``/``False`` for ``!=``/``==`` and raising for inequality comparisons (:issue:`28507`)
+- Bug in :meth:`.DatetimeIndex.equals` and :meth:`.TimedeltaIndex.equals` incorrectly considering ``int64`` indexes as equal (:issue:`36744`)
+- :meth:`Series.to_json`, :meth:`DataFrame.to_json`, and :meth:`read_json` now implement time zone parsing when orient structure is ``table`` (:issue:`35973`)
+- :meth:`astype` now attempts to convert to ``datetime64[ns, tz]`` directly from ``object`` with inferred time zone from string (:issue:`35973`)
+- Bug in :meth:`.TimedeltaIndex.sum` and :meth:`Series.sum` with ``timedelta64`` dtype on an empty index or series returning ``NaT`` instead of ``Timedelta(0)`` (:issue:`31751`)
+- Bug in :meth:`.DatetimeArray.shift` incorrectly allowing ``fill_value`` with a mismatched time zone (:issue:`37299`)
+- Bug in adding a :class:`.BusinessDay` with nonzero ``offset`` to a non-scalar other (:issue:`37457`)
+- Bug in :func:`to_datetime` with a read-only array incorrectly raising (:issue:`34857`)
+- Bug in :meth:`Series.isin` with ``datetime64[ns]`` dtype and :meth:`.DatetimeIndex.isin` incorrectly casting integers to datetimes (:issue:`36621`)
+- Bug in :meth:`Series.isin` with ``datetime64[ns]`` dtype and :meth:`.DatetimeIndex.isin` failing to consider tz-aware and tz-naive datetimes as always different (:issue:`35728`)
+- Bug in :meth:`Series.isin` with ``PeriodDtype`` dtype and :meth:`PeriodIndex.isin` failing to consider arguments with different ``PeriodDtype`` as always different (:issue:`37528`)
+- Bug in :class:`Period` constructor now correctly handles nanoseconds in the ``value`` argument (:issue:`34621` and :issue:`17053`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :class:`.TimedeltaIndex`, :class:`Series`, and :class:`DataFrame` floor-division with ``timedelta64`` dtypes and ``NaT`` in the denominator (:issue:`35529`)
+- Bug in parsing of ISO 8601 durations in :class:`Timedelta` and :func:`to_datetime` (:issue:`29773`, :issue:`36204`)
+- Bug in :func:`to_timedelta` with a read-only array incorrectly raising (:issue:`34857`)
+- Bug in :class:`Timedelta` incorrectly truncating to sub-second portion of a string input when it has precision higher than nanoseconds (:issue:`36738`)
+
+Timezones
+^^^^^^^^^
+
+- Bug in :func:`date_range` was raising ``AmbiguousTimeError`` for valid input with ``ambiguous=False`` (:issue:`35297`)
+- Bug in :meth:`Timestamp.replace` was losing fold information (:issue:`37610`)
+
+
+Numeric
+^^^^^^^
+- Bug in :func:`to_numeric` where float precision was incorrect (:issue:`31364`)
+- Bug in :meth:`DataFrame.any` with ``axis=1`` and ``bool_only=True`` ignoring the ``bool_only`` keyword (:issue:`32432`)
+- Bug in :meth:`Series.equals` where a ``ValueError`` was raised when NumPy arrays were compared to scalars (:issue:`35267`)
+- Bug in :class:`Series` where two Series each have a :class:`.DatetimeIndex` with different time zones having those indexes incorrectly changed when performing arithmetic operations (:issue:`33671`)
+- Bug in :mod:`pandas.testing` module functions when used with ``check_exact=False`` on complex numeric types (:issue:`28235`)
+- Bug in :meth:`DataFrame.__rmatmul__` error handling reporting transposed shapes (:issue:`21581`)
+- Bug in :class:`Series` flex arithmetic methods where the result when operating with a ``list``, ``tuple`` or ``np.ndarray`` would have an incorrect name (:issue:`36760`)
+- Bug in :class:`.IntegerArray` multiplication with ``timedelta`` and ``np.timedelta64`` objects (:issue:`36870`)
+- Bug in :class:`MultiIndex` comparison with tuple incorrectly treating tuple as array-like (:issue:`21517`)
+- Bug in :meth:`DataFrame.diff` with ``datetime64`` dtypes including ``NaT`` values failing to fill ``NaT`` results correctly (:issue:`32441`)
+- Bug in :class:`DataFrame` arithmetic ops incorrectly accepting keyword arguments (:issue:`36843`)
+- Bug in :class:`.IntervalArray` comparisons with :class:`Series` not returning Series (:issue:`36908`)
+- Bug in :class:`DataFrame` allowing arithmetic operations with list of array-likes with undefined results. Behavior changed to raising ``ValueError`` (:issue:`36702`)
+- Bug in :meth:`DataFrame.std` with ``timedelta64`` dtype and ``skipna=False`` (:issue:`37392`)
+- Bug in :meth:`DataFrame.min` and :meth:`DataFrame.max` with ``datetime64`` dtype and ``skipna=False`` (:issue:`36907`)
+- Bug in :meth:`DataFrame.idxmax` and :meth:`DataFrame.idxmin` with mixed dtypes incorrectly raising ``TypeError`` (:issue:`38195`)
+
+Conversion
+^^^^^^^^^^
+
+- Bug in :meth:`DataFrame.to_dict` with ``orient='records'`` now returns python native datetime objects for datetime-like columns (:issue:`21256`)
+- Bug in :meth:`Series.astype` conversion from ``string`` to ``float`` raised in presence of ``pd.NA`` values (:issue:`37626`)
+
+Strings
+^^^^^^^
+- Bug in :meth:`Series.to_string`, :meth:`DataFrame.to_string`, and :meth:`DataFrame.to_latex` adding a leading space when ``index=False`` (:issue:`24980`)
+- Bug in :func:`to_numeric` raising a ``TypeError`` when attempting to convert a string dtype Series containing only numeric strings and ``NA`` (:issue:`37262`)
+
+Interval
+^^^^^^^^
+
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` where :class:`Interval` dtypes would be converted to object dtypes (:issue:`34871`)
+- Bug in :meth:`IntervalIndex.take` with negative indices and ``fill_value=None`` (:issue:`37330`)
+- Bug in :meth:`IntervalIndex.putmask` with datetime-like dtype incorrectly casting to object dtype (:issue:`37968`)
+- Bug in :meth:`IntervalArray.astype` incorrectly dropping dtype information with a :class:`CategoricalDtype` object (:issue:`37984`)
+
+Indexing
+^^^^^^^^
+
+- Bug in :meth:`PeriodIndex.get_loc` incorrectly raising ``ValueError`` on non-datelike strings instead of ``KeyError``, causing similar errors in :meth:`Series.__getitem__`, :meth:`Series.__contains__`, and :meth:`Series.loc.__getitem__` (:issue:`34240`)
+- Bug in :meth:`Index.sort_values` where, when empty values were passed, the method would break by trying to compare missing values instead of pushing them to the end of the sort order (:issue:`35584`)
+- Bug in :meth:`Index.get_indexer` and :meth:`Index.get_indexer_non_unique` where ``int64`` arrays are returned instead of ``intp`` (:issue:`36359`)
+- Bug in :meth:`DataFrame.sort_index` where parameter ascending passed as a list on a single level index gives wrong result (:issue:`32334`)
+- Bug in :meth:`DataFrame.reset_index` was incorrectly raising a ``ValueError`` for input with a :class:`MultiIndex` with missing values in a level with ``Categorical`` dtype (:issue:`24206`)
+- Bug in indexing with boolean masks on datetime-like values sometimes returning a view instead of a copy (:issue:`36210`)
+- Bug in :meth:`DataFrame.__getitem__` and :meth:`DataFrame.loc.__getitem__` with :class:`IntervalIndex` columns and a numeric indexer (:issue:`26490`)
+- Bug in :meth:`Series.loc.__getitem__` with a non-unique :class:`MultiIndex` and an empty-list indexer (:issue:`13691`)
+- Bug in indexing on a :class:`Series` or :class:`DataFrame` with a :class:`MultiIndex` and a level named ``"0"`` (:issue:`37194`)
+- Bug in :meth:`Series.__getitem__` when using an unsigned integer array as an indexer giving incorrect results or segfaulting instead of raising ``KeyError`` (:issue:`37218`)
+- Bug in :meth:`Index.where` incorrectly casting numeric values to strings (:issue:`37591`)
+- Bug in :meth:`DataFrame.loc` returning empty result when indexer is a slice with negative step size (:issue:`38071`)
+- Bug in :meth:`Series.loc` and :meth:`DataFrame.loc` raises when the index was of ``object`` dtype and the given numeric label was in the index (:issue:`26491`)
+- Bug in :meth:`DataFrame.loc` returned requested key plus missing values when ``loc`` was applied to single level from a :class:`MultiIndex` (:issue:`27104`)
+- Bug in indexing on a :class:`Series` or :class:`DataFrame` with a :class:`CategoricalIndex` using a list-like indexer containing NA values (:issue:`37722`)
+- Bug in :meth:`DataFrame.loc.__setitem__` expanding an empty :class:`DataFrame` with mixed dtypes (:issue:`37932`)
+- Bug in :meth:`DataFrame.xs` ignored ``droplevel=False`` for columns (:issue:`19056`)
+- Bug in :meth:`DataFrame.reindex` raising ``IndexingError`` wrongly for empty DataFrame with ``tolerance`` not ``None`` or ``method="nearest"`` (:issue:`27315`)
+- Bug in indexing on a :class:`Series` or :class:`DataFrame` with a :class:`CategoricalIndex` using list-like indexer that contains elements that are in the index's ``categories`` but not in the index itself failing to raise ``KeyError`` (:issue:`37901`)
+- Bug on inserting a boolean label into a :class:`DataFrame` with a numeric :class:`Index` columns incorrectly casting to integer (:issue:`36319`)
+- Bug in :meth:`DataFrame.iloc` and :meth:`Series.iloc` aligning objects in ``__setitem__`` (:issue:`22046`)
+- Bug in :meth:`MultiIndex.drop` does not raise if labels are partially found (:issue:`37820`)
+- Bug in :meth:`DataFrame.loc` did not raise ``KeyError`` when missing combination was given with ``slice(None)`` for remaining levels (:issue:`19556`)
+- Bug in :meth:`DataFrame.loc` raising ``TypeError`` when non-integer slice was given to select values from :class:`MultiIndex` (:issue:`25165`, :issue:`24263`)
+- Bug in :meth:`Series.at` returning :class:`Series` with one element instead of scalar when index is a :class:`MultiIndex` with one level (:issue:`38053`)
+- Bug in :meth:`DataFrame.loc` returning and assigning elements in wrong order when indexer is differently ordered than the :class:`MultiIndex` to filter (:issue:`31330`, :issue:`34603`)
+- Bug in :meth:`DataFrame.loc` and :meth:`DataFrame.__getitem__`  raising ``KeyError`` when columns were :class:`MultiIndex` with only one level (:issue:`29749`)
+- Bug in :meth:`Series.__getitem__` and :meth:`DataFrame.__getitem__` raising blank ``KeyError`` without missing keys for :class:`IntervalIndex` (:issue:`27365`)
+- Bug in setting a new label on a :class:`DataFrame` or :class:`Series` with a :class:`CategoricalIndex` incorrectly raising ``TypeError`` when the new label is not among the index's categories (:issue:`38098`)
+- Bug in :meth:`Series.loc` and :meth:`Series.iloc` raising ``ValueError`` when inserting a list-like ``np.array``, ``list`` or ``tuple`` in an ``object`` Series of equal length (:issue:`37748`, :issue:`37486`)
+- Bug in :meth:`Series.loc` and :meth:`Series.iloc` setting all the values of an ``object`` Series with those of a list-like ``ExtensionArray`` instead of inserting it (:issue:`38271`)
+
+Missing
+^^^^^^^
+
+- Bug in :meth:`.SeriesGroupBy.transform` now correctly handles missing values for ``dropna=False`` (:issue:`35014`)
+- Bug in :meth:`Series.nunique` with ``dropna=True`` was returning incorrect results when both ``NA`` and ``None`` missing values were present (:issue:`37566`)
+- Bug in :meth:`Series.interpolate` where kwarg ``limit_area`` and ``limit_direction`` had no effect when using methods ``pad`` and ``backfill`` (:issue:`31048`)
+
+MultiIndex
+^^^^^^^^^^
+
+- Bug in :meth:`DataFrame.xs` when used with :class:`IndexSlice` raises ``TypeError`` with message ``"Expected label or tuple of labels"`` (:issue:`35301`)
+- Bug in :meth:`DataFrame.reset_index` with ``NaT`` values in index raises ``ValueError`` with message ``"cannot convert float NaN to integer"`` (:issue:`36541`)
+- Bug in :meth:`DataFrame.combine_first` when used with :class:`MultiIndex` containing string and ``NaN`` values raises ``TypeError`` (:issue:`36562`)
+- Bug in :meth:`MultiIndex.drop` dropped ``NaN`` values when non existing key was given as input (:issue:`18853`)
+- Bug in :meth:`MultiIndex.drop` dropping more values than expected when index has duplicates and is not sorted (:issue:`33494`)
+
+I/O
+^^^
+
+- :func:`read_sas` no longer leaks resources on failure (:issue:`35566`)
+- Bug in :meth:`DataFrame.to_csv` and :meth:`Series.to_csv` caused a ``ValueError`` when it was called with a filename in combination with ``mode`` containing a ``b`` (:issue:`35058`)
+- Bug in :meth:`read_csv` with ``float_precision='round_trip'`` did not handle ``decimal`` and ``thousands`` parameters (:issue:`35365`)
+- :meth:`to_pickle` and :meth:`read_pickle` were closing user-provided file objects (:issue:`35679`)
+- :meth:`to_csv` passes compression arguments for ``'gzip'`` always to ``gzip.GzipFile`` (:issue:`28103`)
+- :meth:`to_csv` did not support zip compression for binary file object not having a filename (:issue:`35058`)
+- :meth:`to_csv` and :meth:`read_csv` did not honor ``compression`` and ``encoding`` for path-like objects that are internally converted to file-like objects (:issue:`35677`, :issue:`26124`, :issue:`32392`)
+- :meth:`DataFrame.to_pickle`, :meth:`Series.to_pickle`, and :meth:`read_pickle` did not support compression for file-objects (:issue:`26237`, :issue:`29054`, :issue:`29570`)
+- Bug in :func:`LongTableBuilder.middle_separator` was duplicating LaTeX longtable entries in the List of Tables of a LaTeX document (:issue:`34360`)
+- Bug in :meth:`read_csv` with ``engine='python'`` truncating data if multiple items present in first row and first element started with BOM (:issue:`36343`)
+- Removed ``private_key`` and ``verbose`` from :func:`read_gbq` as they are no longer supported in ``pandas-gbq`` (:issue:`34654`, :issue:`30200`)
+- Bumped minimum pytables version to 3.5.1 to avoid a ``ValueError`` in :meth:`read_hdf` (:issue:`24839`)
+- Bug in :func:`read_table` and :func:`read_csv` when ``delim_whitespace=True`` and ``sep=default`` (:issue:`36583`)
+- Bug in :meth:`DataFrame.to_json` and :meth:`Series.to_json` when used with ``lines=True`` and ``orient='records'`` the last line of the record is not appended with 'new line character' (:issue:`36888`)
+- Bug in :meth:`read_parquet` with fixed offset time zones. String representation of time zones was not recognized (:issue:`35997`, :issue:`36004`)
+- Bug in :meth:`DataFrame.to_html`, :meth:`DataFrame.to_string`, and :meth:`DataFrame.to_latex` ignoring the ``na_rep`` argument when ``float_format`` was also specified (:issue:`9046`, :issue:`13828`)
+- Bug in output rendering of complex numbers showing too many trailing zeros (:issue:`36799`)
+- Bug in :class:`HDFStore` threw a ``TypeError`` when exporting an empty DataFrame with ``datetime64[ns, tz]`` dtypes with a fixed HDF5 store (:issue:`20594`)
+- Bug in :class:`HDFStore` was dropping time zone information when exporting a Series with ``datetime64[ns, tz]`` dtypes with a fixed HDF5 store (:issue:`20594`)
+- :func:`read_csv` was closing user-provided binary file handles when ``engine="c"`` and an ``encoding`` was requested (:issue:`36980`)
+- Bug in :meth:`DataFrame.to_hdf` was not dropping missing rows with ``dropna=True`` (:issue:`35719`)
+- Bug in :func:`read_html` was raising a ``TypeError`` when supplying a ``pathlib.Path`` argument to the ``io`` parameter (:issue:`37705`)
+- :meth:`DataFrame.to_excel`, :meth:`Series.to_excel`, :meth:`DataFrame.to_markdown`, and :meth:`Series.to_markdown` now support writing to fsspec URLs such as S3 and Google Cloud Storage (:issue:`33987`)
+- Bug in :func:`read_fwf` with ``skip_blank_lines=True`` was not skipping blank lines (:issue:`37758`)
+- Parse missing values using :func:`read_json` with ``dtype=False`` to ``NaN`` instead of ``None`` (:issue:`28501`)
+- :meth:`read_fwf` was inferring compression with ``compression=None`` which was not consistent with the other ``read_*`` functions (:issue:`37909`)
+- :meth:`DataFrame.to_html` was ignoring ``formatters`` argument for ``ExtensionDtype`` columns (:issue:`36525`)
+- Bumped minimum xarray version to 0.12.3 to avoid reference to the removed ``Panel`` class (:issue:`27101`, :issue:`37983`)
+- :meth:`DataFrame.to_csv` was re-opening file-like handles that also implement ``os.PathLike`` (:issue:`38125`)
+- Bug in the conversion of a sliced ``pyarrow.Table`` with missing values to a DataFrame (:issue:`38525`)
+- Bug in :func:`read_sql_table` raising a ``sqlalchemy.exc.OperationalError`` when column names contained a percentage sign (:issue:`37517`)
+
+Period
+^^^^^^
+
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` where :class:`Period` dtypes would be converted to object dtypes (:issue:`34871`)
+
+Plotting
+^^^^^^^^
+
+- Bug in :meth:`DataFrame.plot` was rotating xticklabels when ``subplots=True``, even if the x-axis wasn't an irregular time series (:issue:`29460`)
+- Bug in :meth:`DataFrame.plot` where a marker letter in the ``style`` keyword sometimes caused a ``ValueError`` (:issue:`21003`)
+- Bug in :meth:`DataFrame.plot.bar` and :meth:`Series.plot.bar` where ticks positions were assigned by value order instead of using the actual value for numeric or a smart ordering for string (:issue:`26186`, :issue:`11465`). This fix has been reverted in pandas 1.2.1, see :doc:`v1.2.1`
+- Twinned axes were losing their tick labels which should only happen to all but the last row or column of 'externally' shared axes (:issue:`33819`)
+- Bug in :meth:`Series.plot` and :meth:`DataFrame.plot` was throwing a :exc:`ValueError` when the Series or DataFrame was
+  indexed by a :class:`.TimedeltaIndex` with a fixed frequency and the x-axis lower limit was greater than the upper limit (:issue:`37454`)
+- Bug in :meth:`.DataFrameGroupBy.boxplot` when ``subplots=False`` would raise a ``KeyError`` (:issue:`16748`)
+- Bug in :meth:`DataFrame.plot` and :meth:`Series.plot` was overwriting matplotlib's shared y axes behavior when no ``sharey`` parameter was passed (:issue:`37942`)
+- Bug in :meth:`DataFrame.plot` was raising a ``TypeError`` with ``ExtensionDtype`` columns (:issue:`32073`)
+
+Styler
+^^^^^^
+
+- Bug in :meth:`Styler.render` HTML was generated incorrectly because of formatting error in ``rowspan`` attribute, it now matches with w3 syntax (:issue:`38234`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :meth:`.DataFrameGroupBy.count` and :meth:`SeriesGroupBy.sum` returning ``NaN`` for missing categories when grouped on multiple ``Categoricals``. Now returning ``0`` (:issue:`35028`)
+- Bug in :meth:`.DataFrameGroupBy.apply` that would sometimes throw an erroneous ``ValueError`` if the grouping axis had duplicate entries (:issue:`16646`)
+- Bug in :meth:`DataFrame.resample` that would throw a ``ValueError`` when resampling from ``"D"`` to ``"24H"`` over a transition into daylight savings time (DST) (:issue:`35219`)
+- Bug when combining methods :meth:`DataFrame.groupby` with :meth:`DataFrame.resample` and :meth:`DataFrame.interpolate` raising a ``TypeError`` (:issue:`35325`)
+- Bug in :meth:`.DataFrameGroupBy.apply` where a non-nuisance grouping column would be dropped from the output columns if another groupby method was called before ``.apply`` (:issue:`34656`)
+- Bug when subsetting columns on a :class:`.DataFrameGroupBy` (e.g. ``df.groupby('a')[['b']])``) would reset the attributes ``axis``, ``dropna``, ``group_keys``, ``level``, ``mutated``, ``sort``, and ``squeeze`` to their default values (:issue:`9959`)
+- Bug in :meth:`.DataFrameGroupBy.tshift` failing to raise ``ValueError`` when a frequency cannot be inferred for the index of a group (:issue:`35937`)
+- Bug in :meth:`DataFrame.groupby` does not always maintain column index name for ``any``, ``all``, ``bfill``, ``ffill``, ``shift`` (:issue:`29764`)
+- Bug in :meth:`.DataFrameGroupBy.apply` raising error with ``np.nan`` group(s) when ``dropna=False`` (:issue:`35889`)
+- Bug in :meth:`.Rolling.sum` returned wrong values when dtypes where mixed between float and integer and ``axis=1`` (:issue:`20649`, :issue:`35596`)
+- Bug in :meth:`.Rolling.count` returned ``np.nan`` with :class:`~pandas.api.indexers.FixedForwardWindowIndexer` as window, ``min_periods=0`` and only missing values in the window (:issue:`35579`)
+- Bug where :class:`.Rolling` produces incorrect window sizes when using a ``PeriodIndex`` (:issue:`34225`)
+- Bug in :meth:`.DataFrameGroupBy.ffill` and :meth:`.DataFrameGroupBy.bfill` where a ``NaN`` group would return filled values instead of ``NaN`` when ``dropna=True`` (:issue:`34725`)
+- Bug in :meth:`.RollingGroupby.count` where a ``ValueError`` was raised when specifying the ``closed`` parameter (:issue:`35869`)
+- Bug in :meth:`.DataFrameGroupBy.rolling` returning wrong values with partial centered window (:issue:`36040`)
+- Bug in :meth:`.DataFrameGroupBy.rolling` returned wrong values with time aware window containing ``NaN``. Raises ``ValueError`` because windows are not monotonic now (:issue:`34617`)
+- Bug in :meth:`.Rolling.__iter__` where a ``ValueError`` was not raised when ``min_periods`` was larger than ``window`` (:issue:`37156`)
+- Using :meth:`.Rolling.var` instead of :meth:`.Rolling.std` avoids numerical issues for :meth:`.Rolling.corr` when :meth:`.Rolling.var` is still within floating point precision while :meth:`.Rolling.std` is not (:issue:`31286`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.Resampler.quantile` raised ``TypeError`` when values were of type ``Timedelta`` (:issue:`29485`)
+- Bug in :meth:`.Rolling.median` and :meth:`.Rolling.quantile` returned wrong values for :class:`.BaseIndexer` subclasses with non-monotonic starting or ending points for windows (:issue:`37153`)
+- Bug in :meth:`DataFrame.groupby` dropped ``nan`` groups from result with ``dropna=False`` when grouping over a single column (:issue:`35646`, :issue:`35542`)
+- Bug in :meth:`.DataFrameGroupBy.head`, :meth:`DataFrameGroupBy.tail`, :meth:`SeriesGroupBy.head`, and :meth:`SeriesGroupBy.tail` would raise when used with ``axis=1`` (:issue:`9772`)
+- Bug in :meth:`.DataFrameGroupBy.transform` would raise when used with ``axis=1`` and a transformation kernel (e.g. "shift") (:issue:`36308`)
+- Bug in :meth:`.DataFrameGroupBy.resample` using ``.agg`` with sum produced different result than just calling ``.sum`` (:issue:`33548`)
+- Bug in :meth:`.DataFrameGroupBy.apply` dropped values on ``nan`` group when returning the same axes with the original frame (:issue:`38227`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` couldn't handle with arraylike ``q`` when grouping by columns (:issue:`33795`)
+- Bug in :meth:`DataFrameGroupBy.rank` with ``datetime64tz`` or period dtype incorrectly casting results to those dtypes instead of returning ``float64`` dtype (:issue:`38187`)
+
+Reshaping
+^^^^^^^^^
+
+- Bug in :meth:`DataFrame.crosstab` was returning incorrect results on inputs with duplicate row names, duplicate column names or duplicate names between row and column labels (:issue:`22529`)
+- Bug in :meth:`DataFrame.pivot_table` with ``aggfunc='count'`` or ``aggfunc='sum'`` returning ``NaN`` for missing categories when pivoted on a ``Categorical``. Now returning ``0`` (:issue:`31422`)
+- Bug in :func:`concat` and :class:`DataFrame` constructor where input index names are not preserved in some cases (:issue:`13475`)
+- Bug in func :meth:`crosstab` when using multiple columns with ``margins=True`` and ``normalize=True`` (:issue:`35144`)
+- Bug in :meth:`DataFrame.stack` where an empty DataFrame.stack would raise an error (:issue:`36113`). Now returning an empty Series with empty MultiIndex.
+- Bug in :meth:`Series.unstack`. Now a Series with single level of Index trying to unstack would raise a ``ValueError`` (:issue:`36113`)
+- Bug in :meth:`DataFrame.agg` with ``func={'name':<FUNC>}`` incorrectly raising ``TypeError`` when ``DataFrame.columns==['Name']`` (:issue:`36212`)
+- Bug in :meth:`Series.transform` would give incorrect results or raise when the argument ``func`` was a dictionary (:issue:`35811`)
+- Bug in :meth:`DataFrame.pivot` did not preserve :class:`MultiIndex` level names for columns when rows and columns are both multiindexed (:issue:`36360`)
+- Bug in :meth:`DataFrame.pivot` modified ``index`` argument when ``columns`` was passed but ``values`` was not (:issue:`37635`)
+- Bug in :meth:`DataFrame.join` returned a non deterministic level-order for the resulting :class:`MultiIndex` (:issue:`36910`)
+- Bug in :meth:`DataFrame.combine_first` caused wrong alignment with dtype ``string`` and one level of ``MultiIndex`` containing only ``NA`` (:issue:`37591`)
+- Fixed regression in :func:`merge` on merging :class:`.DatetimeIndex` with empty DataFrame (:issue:`36895`)
+- Bug in :meth:`DataFrame.apply` not setting index of return value when ``func`` return type is ``dict`` (:issue:`37544`)
+- Bug in :meth:`DataFrame.merge` and :meth:`pandas.merge` returning inconsistent ordering in result for ``how=right`` and ``how=left`` (:issue:`35382`)
+- Bug in :func:`merge_ordered` couldn't handle list-like ``left_by`` or ``right_by`` (:issue:`35269`)
+- Bug in :func:`merge_ordered` returned wrong join result when length of ``left_by`` or ``right_by`` equals to the rows of ``left`` or ``right`` (:issue:`38166`)
+- Bug in :func:`merge_ordered` didn't raise when elements in ``left_by`` or ``right_by`` not exist in ``left`` columns or ``right`` columns (:issue:`38167`)
+- Bug in :func:`DataFrame.drop_duplicates` not validating bool dtype for ``ignore_index`` keyword (:issue:`38274`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+
+- Fixed bug where :class:`DataFrame` column set to scalar extension type via a dict instantiation was considered an object type rather than the extension type (:issue:`35965`)
+- Fixed bug where ``astype()`` with equal dtype and ``copy=False`` would return a new object (:issue:`28488`)
+- Fixed bug when applying a NumPy ufunc with multiple outputs to an :class:`.IntegerArray` returning ``None`` (:issue:`36913`)
+- Fixed an inconsistency in :class:`.PeriodArray`'s ``__init__`` signature to those of :class:`.DatetimeArray` and :class:`.TimedeltaArray` (:issue:`37289`)
+- Reductions for :class:`.BooleanArray`, :class:`.Categorical`, :class:`.DatetimeArray`, :class:`.FloatingArray`, :class:`.IntegerArray`, :class:`.PeriodArray`, :class:`.TimedeltaArray`, and :class:`.PandasArray` are now keyword-only methods (:issue:`37541`)
+- Fixed a bug where a  ``TypeError`` was wrongly raised if a membership check was made on an ``ExtensionArray`` containing nan-like values (:issue:`37867`)
+
+Other
+^^^^^
+
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` incorrectly raising an ``AssertionError`` instead of a ``ValueError`` when invalid parameter combinations are passed (:issue:`36045`)
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` with numeric values and string ``to_replace`` (:issue:`34789`)
+- Fixed metadata propagation in :meth:`Series.abs` and ufuncs called on Series and DataFrames (:issue:`28283`)
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` incorrectly casting from ``PeriodDtype`` to object dtype (:issue:`34871`)
+- Fixed bug in metadata propagation incorrectly copying DataFrame columns as metadata when the column name overlaps with the metadata name (:issue:`37037`)
+- Fixed metadata propagation in the :class:`Series.dt`, :class:`Series.str` accessors, :class:`DataFrame.duplicated`, :class:`DataFrame.stack`, :class:`DataFrame.unstack`, :class:`DataFrame.pivot`, :class:`DataFrame.append`, :class:`DataFrame.diff`, :class:`DataFrame.applymap` and :class:`DataFrame.update` methods (:issue:`28283`, :issue:`37381`)
+- Fixed metadata propagation when selecting columns with ``DataFrame.__getitem__`` (:issue:`28283`)
+- Bug in :meth:`Index.intersection` with non-:class:`Index` failing to set the correct name on the returned :class:`Index` (:issue:`38111`)
+- Bug in :meth:`RangeIndex.intersection` failing to set the correct name on the returned :class:`Index` in some corner cases (:issue:`38197`)
+- Bug in :meth:`Index.difference` failing to set the correct name on the returned :class:`Index` in some corner cases (:issue:`38268`)
+- Bug in :meth:`Index.union` behaving differently depending on whether operand is an :class:`Index` or other list-like (:issue:`36384`)
+- Bug in :meth:`Index.intersection` with non-matching numeric dtypes casting to ``object`` dtype instead of minimal common dtype (:issue:`38122`)
+- Bug in :meth:`IntervalIndex.union` returning an incorrectly-typed :class:`Index` when empty (:issue:`38282`)
+- Passing an array with 2 or more dimensions to the :class:`Series` constructor now raises the more specific ``ValueError`` rather than a bare ``Exception`` (:issue:`35744`)
+- Bug in ``dir`` where ``dir(obj)`` wouldn't show attributes defined on the instance for pandas objects (:issue:`37173`)
+- Bug in :meth:`Index.drop` raising ``InvalidIndexError`` when index has duplicates (:issue:`38051`)
+- Bug in :meth:`RangeIndex.difference` returning :class:`Int64Index` in some cases where it should return :class:`RangeIndex` (:issue:`38028`)
+- Fixed bug in :func:`assert_series_equal` when comparing a datetime-like array with an equivalent non extension dtype array (:issue:`37609`)
+- Bug in :func:`.is_bool_dtype` would raise when passed a valid string such as ``"boolean"`` (:issue:`38386`)
+- Fixed regression in logical operators raising ``ValueError`` when columns of :class:`DataFrame` are a :class:`CategoricalIndex` with unused categories (:issue:`38367`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_120.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.1.5..v1.2.0
diff --git a/pandas/doc/source/whatsnew/v1.2.1.rst b/pandas/doc/source/whatsnew/v1.2.1.rst
new file mode 100644
index 0000000000000..cb344cc728566
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.1.rst
@@ -0,0 +1,153 @@
+.. _whatsnew_121:
+
+What's new in 1.2.1 (January 20, 2021)
+--------------------------------------
+
+These are the changes in pandas 1.2.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_121.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`~DataFrame.to_csv` that created corrupted zip files when there were more rows than ``chunksize`` (:issue:`38714`)
+- Fixed regression in :meth:`~DataFrame.to_csv` opening ``codecs.StreamReaderWriter`` in binary mode instead of in text mode (:issue:`39247`)
+- Fixed regression in :meth:`read_csv` and other read functions were the encoding error policy (``errors``) did not default to ``"replace"`` when no encoding was specified (:issue:`38989`)
+- Fixed regression in :func:`read_excel` with non-rawbyte file handles (:issue:`38788`)
+- Fixed regression in :meth:`DataFrame.to_stata` not removing the created file when an error occurred (:issue:`39202`)
+- Fixed regression in ``DataFrame.__setitem__`` raising ``ValueError`` when expanding :class:`DataFrame` and new column is from type ``"0 - name"`` (:issue:`39010`)
+- Fixed regression in setting with :meth:`DataFrame.loc`  raising ``ValueError`` when :class:`DataFrame` has unsorted :class:`MultiIndex` columns and indexer is a scalar (:issue:`38601`)
+- Fixed regression in setting with :meth:`DataFrame.loc` raising ``KeyError`` with :class:`MultiIndex` and list-like columns indexer enlarging :class:`DataFrame` (:issue:`39147`)
+- Fixed regression in :meth:`~DataFrame.groupby` with :class:`Categorical` grouping column not showing unused categories for ``grouped.indices`` (:issue:`38642`)
+- Fixed regression in :meth:`.DataFrameGroupBy.sem` and :meth:`.SeriesGroupBy.sem` where the presence of non-numeric columns would cause an error instead of being dropped (:issue:`38774`)
+- Fixed regression in :meth:`.DataFrameGroupBy.diff` raising for ``int8`` and ``int16`` columns (:issue:`39050`)
+- Fixed regression in :meth:`DataFrame.groupby` when aggregating an ``ExtensionDType`` that could fail for non-numeric values (:issue:`38980`)
+- Fixed regression in :meth:`.Rolling.skew` and :meth:`.Rolling.kurt` modifying the object inplace (:issue:`38908`)
+- Fixed regression in :meth:`DataFrame.any` and :meth:`DataFrame.all` not returning a result for tz-aware ``datetime64`` columns (:issue:`38723`)
+- Fixed regression in :meth:`DataFrame.apply` with ``axis=1`` using str accessor in apply function (:issue:`38979`)
+- Fixed regression in :meth:`DataFrame.replace` raising ``ValueError`` when :class:`DataFrame` has dtype ``bytes`` (:issue:`38900`)
+- Fixed regression in :meth:`Series.fillna` that raised ``RecursionError`` with ``datetime64[ns, UTC]`` dtype (:issue:`38851`)
+- Fixed regression in comparisons between ``NaT`` and ``datetime.date`` objects incorrectly returning ``True`` (:issue:`39151`)
+- Fixed regression in calling NumPy :func:`~numpy.ufunc.accumulate` ufuncs on DataFrames, e.g. ``np.maximum.accumulate(df)`` (:issue:`39259`)
+- Fixed regression in repr of float-like strings of an ``object`` dtype having trailing 0's truncated after the decimal (:issue:`38708`)
+- Fixed regression that raised ``AttributeError`` with PyArrow versions [0.16.0, 1.0.0) (:issue:`38801`)
+- Fixed regression in :func:`pandas.testing.assert_frame_equal` raising ``TypeError`` with ``check_like=True`` when :class:`Index` or columns have mixed dtype (:issue:`39168`)
+
+We have reverted a commit that resulted in several plotting related regressions in pandas 1.2.0 (:issue:`38969`, :issue:`38736`, :issue:`38865`, :issue:`38947` and :issue:`39126`).
+As a result, bugs reported as fixed in pandas 1.2.0 related to inconsistent tick labeling in bar plots are again present (:issue:`26186` and :issue:`11465`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_121.ufunc_deprecation:
+
+Calling NumPy ufuncs on non-aligned DataFrames
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Before pandas 1.2.0, calling a NumPy ufunc on non-aligned DataFrames (or
+DataFrame / Series combination) would ignore the indices, only match
+the inputs by shape, and use the index/columns of the first DataFrame for
+the result:
+
+.. code-block:: ipython
+
+    In [1]: df1 = pd.DataFrame({"a": [1, 2], "b": [3, 4]}, index=[0, 1])
+    In [2]: df2 = pd.DataFrame({"a": [1, 2], "b": [3, 4]}, index=[1, 2])
+    In [3]: df1
+    Out[3]:
+       a  b
+    0  1  3
+    1  2  4
+    In [4]: df2
+    Out[4]:
+       a  b
+    1  1  3
+    2  2  4
+
+    In [5]: np.add(df1, df2)
+    Out[5]:
+       a  b
+    0  2  6
+    1  4  8
+
+This contrasts with how other pandas operations work, which first align
+the inputs:
+
+.. code-block:: ipython
+
+    In [6]: df1 + df2
+    Out[6]:
+         a    b
+    0  NaN  NaN
+    1  3.0  7.0
+    2  NaN  NaN
+
+In pandas 1.2.0, we refactored how NumPy ufuncs are called on DataFrames, and
+this started to align the inputs first (:issue:`39184`), as happens in other
+pandas operations and as it happens for ufuncs called on Series objects.
+
+For pandas 1.2.1, we restored the previous behaviour to avoid a breaking
+change, but the above example of ``np.add(df1, df2)`` with non-aligned inputs
+will now to raise a warning, and a future pandas 2.0 release will start
+aligning the inputs first (:issue:`39184`). Calling a NumPy ufunc on Series
+objects (eg ``np.add(s1, s2)``) already aligns and continues to do so.
+
+To avoid the warning and keep the current behaviour of ignoring the indices,
+convert one of the arguments to a NumPy array:
+
+.. code-block:: ipython
+
+    In [7]: np.add(df1, np.asarray(df2))
+    Out[7]:
+       a  b
+    0  2  6
+    1  4  8
+
+To obtain the future behaviour and silence the warning, you can align manually
+before passing the arguments to the ufunc:
+
+.. code-block:: ipython
+
+    In [8]: df1, df2 = df1.align(df2)
+    In [9]: np.add(df1, df2)
+    Out[9]:
+         a    b
+    0  NaN  NaN
+    1  3.0  7.0
+    2  NaN  NaN
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_121.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- Bug in :meth:`read_csv` with ``float_precision="high"`` caused segfault or wrong parsing of long exponent strings. This resulted in a regression in some cases as the default for ``float_precision`` was changed in pandas 1.2.0 (:issue:`38753`)
+- Bug in :func:`read_csv` not closing an opened file handle when a ``csv.Error`` or ``UnicodeDecodeError`` occurred while initializing (:issue:`39024`)
+- Bug in :func:`pandas.testing.assert_index_equal` raising ``TypeError`` with ``check_order=False`` when :class:`Index` has mixed dtype (:issue:`39168`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_121.other:
+
+Other
+~~~~~
+
+- The deprecated attributes ``_AXIS_NAMES`` and ``_AXIS_NUMBERS`` of :class:`DataFrame` and :class:`Series` will no longer show up in ``dir`` or ``inspect.getmembers`` calls (:issue:`38740`)
+- Bumped minimum fastparquet version to 0.4.0 to avoid ``AttributeError`` from numba (:issue:`38344`)
+- Bumped minimum pymysql version to 0.8.1 to avoid test failures (:issue:`38344`)
+- Fixed build failure on MacOS 11 in Python 3.9.1 (:issue:`38766`)
+- Added reference to backwards incompatible ``check_freq`` arg of :func:`testing.assert_frame_equal` and :func:`testing.assert_series_equal` in :ref:`pandas 1.1.0 what's new <whatsnew_110.api_breaking.testing.check_freq>` (:issue:`34050`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_121.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.0..v1.2.1
diff --git a/pandas/doc/source/whatsnew/v1.2.2.rst b/pandas/doc/source/whatsnew/v1.2.2.rst
new file mode 100644
index 0000000000000..1a9204bc82986
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.2.rst
@@ -0,0 +1,49 @@
+.. _whatsnew_122:
+
+What's new in 1.2.2 (February 09, 2021)
+---------------------------------------
+
+These are the changes in pandas 1.2.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_122.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :func:`read_excel` that caused it to raise ``AttributeError`` when checking version of older xlrd versions (:issue:`38955`)
+- Fixed regression in :class:`DataFrame` constructor reordering element when construction from datetime ndarray with dtype not ``"datetime64[ns]"`` (:issue:`39422`)
+- Fixed regression in :meth:`DataFrame.astype` and :meth:`Series.astype` not casting to bytes dtype (:issue:`39474`)
+- Fixed regression in :meth:`~DataFrame.to_pickle` failing to create bz2/xz compressed pickle files with ``protocol=5`` (:issue:`39002`)
+- Fixed regression in :func:`pandas.testing.assert_series_equal` and :func:`pandas.testing.assert_frame_equal` always raising ``AssertionError`` when comparing extension dtypes (:issue:`39410`)
+- Fixed regression in :meth:`~DataFrame.to_csv` opening ``codecs.StreamWriter`` in binary mode instead of in text mode and ignoring user-provided ``mode`` (:issue:`39247`)
+- Fixed regression in :meth:`Categorical.astype` casting to incorrect dtype when ``np.int32`` is passed to dtype argument (:issue:`39402`)
+- Fixed regression in :meth:`~DataFrame.to_excel` creating corrupt files when appending (``mode="a"``) to an existing file (:issue:`39576`)
+- Fixed regression in :meth:`DataFrame.transform` failing in case of an empty DataFrame or Series (:issue:`39636`)
+- Fixed regression in :meth:`~DataFrame.groupby` or :meth:`~DataFrame.resample` when aggregating an all-NaN or numeric object dtype column (:issue:`39329`)
+- Fixed regression in :meth:`.Rolling.count` where the ``min_periods`` argument would be set to ``0`` after the operation (:issue:`39554`)
+- Fixed regression in :func:`read_excel` that incorrectly raised when the argument ``io`` was a non-path and non-buffer and the ``engine`` argument was specified (:issue:`39528`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_122.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+- :func:`pandas.read_excel` error message when a specified ``sheetname`` does not exist is now uniform across engines (:issue:`39250`)
+- Fixed bug in :func:`pandas.read_excel` producing incorrect results when the engine ``openpyxl`` is used and the excel file is missing or has incorrect dimension information; the fix requires ``openpyxl`` >= 3.0.0, prior versions may still fail (:issue:`38956`, :issue:`39001`)
+- Fixed bug in :func:`pandas.read_excel` sometimes producing a ``DataFrame`` with trailing rows of ``np.nan`` when the engine ``openpyxl`` is used (:issue:`39181`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_122.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.1..v1.2.2
diff --git a/pandas/doc/source/whatsnew/v1.2.3.rst b/pandas/doc/source/whatsnew/v1.2.3.rst
new file mode 100644
index 0000000000000..dec2d061504b4
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.3.rst
@@ -0,0 +1,32 @@
+.. _whatsnew_123:
+
+What's new in 1.2.3 (March 02, 2021)
+------------------------------------
+
+These are the changes in pandas 1.2.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_123.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`~DataFrame.to_excel` raising ``KeyError`` when giving duplicate columns with ``columns`` attribute (:issue:`39695`)
+- Fixed regression in nullable integer unary ops propagating mask on assignment (:issue:`39943`)
+- Fixed regression in :meth:`DataFrame.__setitem__` not aligning :class:`DataFrame` on right-hand side for boolean indexer (:issue:`39931`)
+- Fixed regression in :meth:`~DataFrame.to_json` failing to use ``compression`` with URL-like paths that are internally opened in binary mode or with user-provided file objects that are opened in binary mode (:issue:`39985`)
+- Fixed regression in :meth:`Series.sort_index` and :meth:`DataFrame.sort_index`, which exited with an ungraceful error when having kwarg ``ascending=None`` passed. Passing ``ascending=None`` is still considered invalid, and the improved error message suggests a proper usage (``ascending`` must be a boolean or a list-like of boolean) (:issue:`39434`)
+- Fixed regression in :meth:`DataFrame.transform` and :meth:`Series.transform` giving incorrect column labels when passed a dictionary with a mix of list and non-list values (:issue:`40018`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_123.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.2..v1.2.3
diff --git a/pandas/doc/source/whatsnew/v1.2.4.rst b/pandas/doc/source/whatsnew/v1.2.4.rst
new file mode 100644
index 0000000000000..433ee37508e66
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.4.rst
@@ -0,0 +1,33 @@
+.. _whatsnew_124:
+
+What's new in 1.2.4 (April 12, 2021)
+------------------------------------
+
+These are the changes in pandas 1.2.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_124.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+
+- Fixed regression in :meth:`DataFrame.sum` when ``min_count`` greater than the :class:`DataFrame` shape was passed resulted in a ``ValueError`` (:issue:`39738`)
+- Fixed regression in :meth:`DataFrame.to_json` raising ``AttributeError`` when run on PyPy (:issue:`39837`)
+- Fixed regression in (in)equality comparison of ``pd.NaT`` with a non-datetimelike numpy array returning a scalar instead of an array (:issue:`40722`)
+- Fixed regression in :meth:`DataFrame.where` not returning a copy in the case of an all True condition (:issue:`39595`)
+- Fixed regression in :meth:`DataFrame.replace` raising ``IndexError`` when ``regex`` was a multi-key dictionary (:issue:`39338`)
+- Fixed regression in repr of floats in an ``object`` column not respecting ``float_format`` when printed in the console or outputted through :meth:`DataFrame.to_string`, :meth:`DataFrame.to_html`, and :meth:`DataFrame.to_latex` (:issue:`40024`)
+- Fixed regression in NumPy ufuncs such as ``np.add`` not passing through all arguments for :class:`DataFrame` (:issue:`40662`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_124.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.3..v1.2.4
diff --git a/pandas/doc/source/whatsnew/v1.2.5.rst b/pandas/doc/source/whatsnew/v1.2.5.rst
new file mode 100644
index 0000000000000..d3ceb2b919b5d
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.2.5.rst
@@ -0,0 +1,31 @@
+.. _whatsnew_125:
+
+What's new in 1.2.5 (June 22, 2021)
+-----------------------------------
+
+These are the changes in pandas 1.2.5. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_125.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :func:`concat` between two :class:`DataFrame` where one has an :class:`Index` that is all-None and the other is :class:`DatetimeIndex` incorrectly raising (:issue:`40841`)
+- Fixed regression in :meth:`DataFrame.sum` and :meth:`DataFrame.prod` when ``min_count`` and ``numeric_only`` are both given (:issue:`41074`)
+- Fixed regression in :func:`read_csv` when using ``memory_map=True`` with an non-UTF8 encoding (:issue:`40986`)
+- Fixed regression in :meth:`DataFrame.replace` and :meth:`Series.replace` when the values to replace is a NumPy float array (:issue:`40371`)
+- Fixed regression in :func:`ExcelFile` when a corrupt file is opened but not closed (:issue:`41778`)
+- Fixed regression in :meth:`DataFrame.astype` with ``dtype=str`` failing to convert ``NaN`` in categorical columns (:issue:`41797`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_125.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.4..v1.2.5|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.3.0.rst b/pandas/doc/source/whatsnew/v1.3.0.rst
new file mode 100644
index 0000000000000..0e2d487a89ff5
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.0.rst
@@ -0,0 +1,1249 @@
+.. _whatsnew_130:
+
+What's new in 1.3.0 (July 2, 2021)
+----------------------------------
+
+These are the changes in pandas 1.3.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. warning::
+
+   When reading new Excel 2007+ (``.xlsx``) files, the default argument
+   ``engine=None`` to :func:`read_excel` will now result in using the
+   `openpyxl <https://openpyxl.readthedocs.io/en/stable/>`_ engine in all cases
+   when the option :attr:`io.excel.xlsx.reader` is set to ``"auto"``.
+   Previously, some cases would use the
+   `xlrd <https://xlrd.readthedocs.io/en/latest/>`_ engine instead. See
+   :ref:`What's new 1.2.0 <whatsnew_120>` for background on this change.
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_130.enhancements.read_csv_json_http_headers:
+
+Custom HTTP(s) headers when reading csv or json files
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When reading from a remote URL that is not handled by fsspec (e.g. HTTP and
+HTTPS) the dictionary passed to ``storage_options`` will be used to create the
+headers included in the request.  This can be used to control the User-Agent
+header or send other custom headers (:issue:`36688`).
+For example:
+
+.. code-block:: ipython
+
+    In [1]: headers = {"User-Agent": "pandas"}
+    In [2]: df = pd.read_csv(
+       ...:     "https://download.bls.gov/pub/time.series/cu/cu.item",
+       ...:     sep="\t",
+       ...:     storage_options=headers
+       ...: )
+
+.. _whatsnew_130.enhancements.read_to_xml:
+
+Read and write XML documents
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We added I/O support to read and render shallow versions of `XML`_ documents with
+:func:`read_xml` and :meth:`DataFrame.to_xml`. Using `lxml`_ as parser,
+both XPath 1.0 and XSLT 1.0 are available. (:issue:`27554`)
+
+.. _XML: https://www.w3.org/standards/xml/core
+.. _lxml: https://lxml.de
+
+.. code-block:: ipython
+
+    In [1]: xml = """<?xml version='1.0' encoding='utf-8'?>
+       ...: <data>
+       ...:  <row>
+       ...:     <shape>square</shape>
+       ...:     <degrees>360</degrees>
+       ...:     <sides>4.0</sides>
+       ...:  </row>
+       ...:  <row>
+       ...:     <shape>circle</shape>
+       ...:     <degrees>360</degrees>
+       ...:     <sides/>
+       ...:  </row>
+       ...:  <row>
+       ...:     <shape>triangle</shape>
+       ...:     <degrees>180</degrees>
+       ...:     <sides>3.0</sides>
+       ...:  </row>
+       ...:  </data>"""
+
+    In [2]: df = pd.read_xml(xml)
+    In [3]: df
+    Out[3]:
+          shape  degrees  sides
+    0    square      360    4.0
+    1    circle      360    NaN
+    2  triangle      180    3.0
+
+    In [4]: df.to_xml()
+    Out[4]:
+    <?xml version='1.0' encoding='utf-8'?>
+    <data>
+      <row>
+        <index>0</index>
+        <shape>square</shape>
+        <degrees>360</degrees>
+        <sides>4.0</sides>
+      </row>
+      <row>
+        <index>1</index>
+        <shape>circle</shape>
+        <degrees>360</degrees>
+        <sides/>
+      </row>
+      <row>
+        <index>2</index>
+        <shape>triangle</shape>
+        <degrees>180</degrees>
+        <sides>3.0</sides>
+      </row>
+    </data>
+
+For more, see :ref:`io.xml` in the user guide on IO tools.
+
+.. _whatsnew_130.enhancements.styler:
+
+Styler enhancements
+^^^^^^^^^^^^^^^^^^^
+
+We provided some focused development on :class:`.Styler`. See also the `Styler documentation <../user_guide/style.ipynb>`_
+which has been revised and improved (:issue:`39720`, :issue:`39317`, :issue:`40493`).
+
+ - The method :meth:`.Styler.set_table_styles` can now accept more natural CSS language for arguments, such as ``'color:red;'`` instead of ``[('color', 'red')]`` (:issue:`39563`)
+ - The methods :meth:`.Styler.highlight_null`, :meth:`.Styler.highlight_min`, and :meth:`.Styler.highlight_max` now allow custom CSS highlighting instead of the default background coloring (:issue:`40242`)
+ - :meth:`.Styler.apply` now accepts functions that return an ``ndarray`` when ``axis=None``, making it now consistent with the ``axis=0`` and ``axis=1`` behavior (:issue:`39359`)
+ - When incorrectly formatted CSS is given via :meth:`.Styler.apply` or :meth:`.Styler.applymap`, an error is now raised upon rendering (:issue:`39660`)
+ - :meth:`.Styler.format` now accepts the keyword argument ``escape`` for optional HTML and LaTeX escaping (:issue:`40388`, :issue:`41619`)
+ - :meth:`.Styler.background_gradient` has gained the argument ``gmap`` to supply a specific gradient map for shading (:issue:`22727`)
+ - :meth:`.Styler.clear` now clears :attr:`Styler.hidden_index` and :attr:`Styler.hidden_columns` as well (:issue:`40484`)
+ - Added the method :meth:`.Styler.highlight_between` (:issue:`39821`)
+ - Added the method :meth:`.Styler.highlight_quantile` (:issue:`40926`)
+ - Added the method :meth:`.Styler.text_gradient` (:issue:`41098`)
+ - Added the method :meth:`.Styler.set_tooltips` to allow hover tooltips; this can be used enhance interactive displays (:issue:`21266`, :issue:`40284`)
+ - Added the parameter ``precision`` to the method :meth:`.Styler.format` to control the display of floating point numbers (:issue:`40134`)
+ - :class:`.Styler` rendered HTML output now follows the `w3 HTML Style Guide <https://www.w3schools.com/html/html5_syntax.asp>`_ (:issue:`39626`)
+ - Many features of the :class:`.Styler` class are now either partially or fully usable on a DataFrame with a non-unique indexes or columns (:issue:`41143`)
+ - One has greater control of the display through separate sparsification of the index or columns using the :ref:`new styler options <options.available>`, which are also usable via :func:`option_context` (:issue:`41142`)
+ - Added the option ``styler.render.max_elements`` to avoid browser overload when styling large DataFrames (:issue:`40712`)
+ - Added the method :meth:`.Styler.to_latex` (:issue:`21673`, :issue:`42320`), which also allows some limited CSS conversion (:issue:`40731`)
+ - Added the method :meth:`.Styler.to_html` (:issue:`13379`)
+ - Added the method :meth:`.Styler.set_sticky` to make index and column headers permanently visible in scrolling HTML frames (:issue:`29072`)
+
+.. _whatsnew_130.enhancements.dataframe_honors_copy_with_dict:
+
+DataFrame constructor honors ``copy=False`` with dict
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When passing a dictionary to :class:`DataFrame` with ``copy=False``,
+a copy will no longer be made (:issue:`32960`).
+
+.. ipython:: python
+
+    arr = np.array([1, 2, 3])
+    df = pd.DataFrame({"A": arr, "B": arr.copy()}, copy=False)
+    df
+
+``df["A"]`` remains a view on ``arr``:
+
+.. ipython:: python
+
+    arr[0] = 0
+    assert df.iloc[0, 0] == 0
+
+The default behavior when not passing ``copy`` will remain unchanged, i.e.
+a copy will be made.
+
+.. _whatsnew_130.enhancements.arrow_string:
+
+PyArrow backed string data type
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We've enhanced the :class:`StringDtype`, an extension type dedicated to string data.
+(:issue:`39908`)
+
+It is now possible to specify a ``storage`` keyword option to :class:`StringDtype`. Use
+pandas options or specify the dtype using ``dtype='string[pyarrow]'`` to allow the
+StringArray to be backed by a PyArrow array instead of a NumPy array of Python objects.
+
+The PyArrow backed StringArray requires pyarrow 1.0.0 or greater to be installed.
+
+.. warning::
+
+   ``string[pyarrow]`` is currently considered experimental. The implementation
+   and parts of the API may change without warning.
+
+.. ipython:: python
+
+   pd.Series(['abc', None, 'def'], dtype=pd.StringDtype(storage="pyarrow"))
+
+You can use the alias ``"string[pyarrow]"`` as well.
+
+.. ipython:: python
+
+   s = pd.Series(['abc', None, 'def'], dtype="string[pyarrow]")
+   s
+
+You can also create a PyArrow backed string array using pandas options.
+
+.. ipython:: python
+
+    with pd.option_context("string_storage", "pyarrow"):
+        s = pd.Series(['abc', None, 'def'], dtype="string")
+    s
+
+The usual string accessor methods work. Where appropriate, the return type of the Series
+or columns of a DataFrame will also have string dtype.
+
+.. ipython:: python
+
+   s.str.upper()
+   s.str.split('b', expand=True).dtypes
+
+String accessor methods returning integers will return a value with :class:`Int64Dtype`
+
+.. ipython:: python
+
+   s.str.count("a")
+
+.. _whatsnew_130.enhancements.centered_datetimelike_rolling_window:
+
+Centered datetime-like rolling windows
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When performing rolling calculations on DataFrame and Series
+objects with a datetime-like index, a centered datetime-like window can now be
+used (:issue:`38780`).
+For example:
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {"A": [0, 1, 2, 3, 4]}, index=pd.date_range("2020", periods=5, freq="1D")
+    )
+    df
+    df.rolling("2D", center=True).mean()
+
+
+.. _whatsnew_130.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- :meth:`DataFrame.rolling`, :meth:`Series.rolling`, :meth:`DataFrame.expanding`, and :meth:`Series.expanding` now support a ``method`` argument with a ``'table'`` option that performs the windowing operation over an entire :class:`DataFrame`. See :ref:`Window Overview <window.overview>` for performance and functional benefits (:issue:`15095`, :issue:`38995`)
+- :class:`.ExponentialMovingWindow` now support a ``online`` method that can perform ``mean`` calculations in an online fashion. See :ref:`Window Overview <window.overview>` (:issue:`41673`)
+- Added :meth:`MultiIndex.dtypes` (:issue:`37062`)
+- Added ``end`` and ``end_day`` options for the ``origin`` argument in :meth:`DataFrame.resample` (:issue:`37804`)
+- Improved error message when ``usecols`` and ``names`` do not match for :func:`read_csv` and ``engine="c"`` (:issue:`29042`)
+- Improved consistency of error messages when passing an invalid ``win_type`` argument in :ref:`Window methods <api.window>` (:issue:`15969`)
+- :func:`read_sql_query` now accepts a ``dtype`` argument to cast the columnar data from the SQL database based on user input (:issue:`10285`)
+- :func:`read_csv` now raising ``ParserWarning`` if length of header or given names does not match length of data when ``usecols`` is not specified (:issue:`21768`)
+- Improved integer type mapping from pandas to SQLAlchemy when using :meth:`DataFrame.to_sql` (:issue:`35076`)
+- :func:`to_numeric` now supports downcasting of nullable ``ExtensionDtype`` objects (:issue:`33013`)
+- Added support for dict-like names in :class:`MultiIndex.set_names` and :class:`MultiIndex.rename` (:issue:`20421`)
+- :func:`read_excel` can now auto-detect .xlsb files and older .xls files (:issue:`35416`, :issue:`41225`)
+- :class:`ExcelWriter` now accepts an ``if_sheet_exists`` parameter to control the behavior of append mode when writing to existing sheets (:issue:`40230`)
+- :meth:`.Rolling.sum`, :meth:`.Expanding.sum`, :meth:`.Rolling.mean`, :meth:`.Expanding.mean`, :meth:`.ExponentialMovingWindow.mean`, :meth:`.Rolling.median`, :meth:`.Expanding.median`, :meth:`.Rolling.max`, :meth:`.Expanding.max`, :meth:`.Rolling.min`, and :meth:`.Expanding.min` now support `Numba <http://numba.pydata.org/>`_ execution with the ``engine`` keyword (:issue:`38895`, :issue:`41267`)
+- :meth:`DataFrame.apply` can now accept NumPy unary operators as strings, e.g. ``df.apply("sqrt")``, which was already the case for :meth:`Series.apply` (:issue:`39116`)
+- :meth:`DataFrame.apply` can now accept non-callable DataFrame properties as strings, e.g. ``df.apply("size")``, which was already the case for :meth:`Series.apply` (:issue:`39116`)
+- :meth:`DataFrame.applymap` can now accept kwargs to pass on to the user-provided ``func`` (:issue:`39987`)
+- Passing a :class:`DataFrame` indexer to ``iloc`` is now disallowed for :meth:`Series.__getitem__` and :meth:`DataFrame.__getitem__` (:issue:`39004`)
+- :meth:`Series.apply` can now accept list-like or dictionary-like arguments that aren't lists or dictionaries, e.g. ``ser.apply(np.array(["sum", "mean"]))``, which was already the case for :meth:`DataFrame.apply` (:issue:`39140`)
+- :meth:`DataFrame.plot.scatter` can now accept a categorical column for the argument ``c`` (:issue:`12380`, :issue:`31357`)
+- :meth:`Series.loc` now raises a helpful error message when the Series has a :class:`MultiIndex` and the indexer has too many dimensions (:issue:`35349`)
+- :func:`read_stata` now supports reading data from compressed files (:issue:`26599`)
+- Added support for parsing ``ISO 8601``-like timestamps with negative signs to :class:`Timedelta` (:issue:`37172`)
+- Added support for unary operators in :class:`FloatingArray` (:issue:`38749`)
+- :class:`RangeIndex` can now be constructed by passing a ``range`` object directly e.g. ``pd.RangeIndex(range(3))`` (:issue:`12067`)
+- :meth:`Series.round` and :meth:`DataFrame.round` now work with nullable integer and floating dtypes (:issue:`38844`)
+- :meth:`read_csv` and :meth:`read_json` expose the argument ``encoding_errors`` to control how encoding errors are handled (:issue:`39450`)
+- :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.any`, :meth:`.DataFrameGroupBy.all`, and :meth:`.SeriesGroupBy.all` use Kleene logic with nullable data types (:issue:`37506`)
+- :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.any`, :meth:`.DataFrameGroupBy.all`, and :meth:`.SeriesGroupBy.all` return a ``BooleanDtype`` for columns with nullable data types (:issue:`33449`)
+- :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.any`, :meth:`.DataFrameGroupBy.all`, and :meth:`.SeriesGroupBy.all` raising with ``object`` data containing ``pd.NA`` even when ``skipna=True`` (:issue:`37501`)
+- :meth:`.DataFrameGroupBy.rank` and :meth:`.SeriesGroupBy.rank` now supports object-dtype data (:issue:`38278`)
+- Constructing a :class:`DataFrame` or :class:`Series` with the ``data`` argument being a Python iterable that is *not* a NumPy ``ndarray`` consisting of NumPy scalars will now result in a dtype with a precision the maximum of the NumPy scalars; this was already the case when ``data`` is a NumPy ``ndarray`` (:issue:`40908`)
+- Add keyword ``sort`` to :func:`pivot_table` to allow non-sorting of the result (:issue:`39143`)
+- Add keyword ``dropna`` to :meth:`DataFrame.value_counts` to allow counting rows that include ``NA`` values (:issue:`41325`)
+- :meth:`Series.replace` will now cast results to ``PeriodDtype`` where possible instead of ``object`` dtype (:issue:`41526`)
+- Improved error message in ``corr`` and ``cov`` methods on :class:`.Rolling`, :class:`.Expanding`, and :class:`.ExponentialMovingWindow` when ``other`` is not a :class:`DataFrame` or :class:`Series` (:issue:`41741`)
+- :meth:`Series.between` can now accept ``left`` or ``right`` as arguments to ``inclusive`` to include only the left or right boundary (:issue:`40245`)
+- :meth:`DataFrame.explode` now supports exploding multiple columns. Its ``column`` argument now also accepts a list of str or tuples for exploding on multiple columns at the same time (:issue:`39240`)
+- :meth:`DataFrame.sample` now accepts the ``ignore_index`` argument to reset the index after sampling, similar to :meth:`DataFrame.drop_duplicates` and :meth:`DataFrame.sort_values` (:issue:`38581`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_130.notable_bug_fixes.categorical_unique_maintains_dtype:
+
+``Categorical.unique`` now always maintains same dtype as original
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, when calling :meth:`Categorical.unique` with categorical data, unused categories in the new array
+would be removed, making the dtype of the new array different than the
+original (:issue:`18291`)
+
+As an example of this, given:
+
+.. ipython:: python
+
+        dtype = pd.CategoricalDtype(['bad', 'neutral', 'good'], ordered=True)
+        cat = pd.Categorical(['good', 'good', 'bad', 'bad'], dtype=dtype)
+        original = pd.Series(cat)
+        unique = original.unique()
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [1]: unique
+    ['good', 'bad']
+    Categories (2, object): ['bad' < 'good']
+    In [2]: original.dtype == unique.dtype
+    False
+
+*New behavior*:
+
+.. ipython:: python
+
+        unique
+        original.dtype == unique.dtype
+
+.. _whatsnew_130.notable_bug_fixes.combine_first_preserves_dtype:
+
+Preserve dtypes in :meth:`DataFrame.combine_first`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.combine_first` will now preserve dtypes (:issue:`7509`)
+
+.. ipython:: python
+
+   df1 = pd.DataFrame({"A": [1, 2, 3], "B": [1, 2, 3]}, index=[0, 1, 2])
+   df1
+   df2 = pd.DataFrame({"B": [4, 5, 6], "C": [1, 2, 3]}, index=[2, 3, 4])
+   df2
+   combined = df1.combine_first(df2)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [1]: combined.dtypes
+   Out[2]:
+   A    float64
+   B    float64
+   C    float64
+   dtype: object
+
+*New behavior*:
+
+.. ipython:: python
+
+   combined.dtypes
+
+.. _whatsnew_130.notable_bug_fixes.groupby_preserves_dtype:
+
+Groupby methods agg and transform no longer changes return dtype for callables
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously the methods :meth:`.DataFrameGroupBy.aggregate`,
+:meth:`.SeriesGroupBy.aggregate`, :meth:`.DataFrameGroupBy.transform`, and
+:meth:`.SeriesGroupBy.transform` might cast the result dtype when the argument ``func``
+is callable, possibly leading to undesirable results (:issue:`21240`). The cast would
+occur if the result is numeric and casting back to the input dtype does not change any
+values as measured by ``np.allclose``. Now no such casting occurs.
+
+.. ipython:: python
+
+    df = pd.DataFrame({'key': [1, 1], 'a': [True, False], 'b': [True, True]})
+    df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.groupby('key').agg(lambda x: x.sum())
+    Out[5]:
+            a  b
+    key
+    1    True  2
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby('key').agg(lambda x: x.sum())
+
+.. _whatsnew_130.notable_bug_fixes.groupby_reductions_float_result:
+
+``float`` result for :meth:`.DataFrameGroupBy.mean`, :meth:`.DataFrameGroupBy.median`, and :meth:`.GDataFrameGroupBy.var`, :meth:`.SeriesGroupBy.mean`, :meth:`.SeriesGroupBy.median`, and :meth:`.SeriesGroupBy.var`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, these methods could result in different dtypes depending on the input values.
+Now, these methods will always return a float dtype. (:issue:`41137`)
+
+.. ipython:: python
+
+    df = pd.DataFrame({'a': [True], 'b': [1], 'c': [1.0]})
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.groupby(df.index).mean()
+    Out[5]:
+            a  b    c
+    0    True  1  1.0
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby(df.index).mean()
+
+.. _whatsnew_130.notable_bug_fixes.setitem_column_try_inplace:
+
+Try operating inplace when setting values with ``loc`` and ``iloc``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When setting an entire column using ``loc`` or ``iloc``, pandas will try to
+insert the values into the existing data rather than create an entirely new array.
+
+.. ipython:: python
+
+   df = pd.DataFrame(range(3), columns=["A"], dtype="float64")
+   values = df.values
+   new = np.array([5, 6, 7], dtype="int64")
+   df.loc[[0, 1, 2], "A"] = new
+
+In both the new and old behavior, the data in ``values`` is overwritten, but in
+the old behavior the dtype of ``df["A"]`` changed to ``int64``.
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [1]: df.dtypes
+   Out[1]:
+   A    int64
+   dtype: object
+   In [2]: np.shares_memory(df["A"].values, new)
+   Out[2]: False
+   In [3]: np.shares_memory(df["A"].values, values)
+   Out[3]: False
+
+In pandas 1.3.0, ``df`` continues to share data with ``values``
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.dtypes
+   np.shares_memory(df["A"], new)
+   np.shares_memory(df["A"], values)
+
+
+.. _whatsnew_130.notable_bug_fixes.setitem_never_inplace:
+
+Never operate inplace when setting ``frame[keys] = values``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When setting multiple columns using ``frame[keys] = values`` new arrays will
+replace pre-existing arrays for these keys, which will *not* be over-written
+(:issue:`39510`).  As a result, the columns will retain the dtype(s) of ``values``,
+never casting to the dtypes of the existing arrays.
+
+.. ipython:: python
+
+   df = pd.DataFrame(range(3), columns=["A"], dtype="float64")
+   df[["A"]] = 5
+
+In the old behavior, ``5`` was cast to ``float64`` and inserted into the existing
+array backing ``df``:
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [1]: df.dtypes
+   Out[1]:
+   A    float64
+
+In the new behavior, we get a new array, and retain an integer-dtyped ``5``:
+
+*New behavior*:
+
+.. ipython:: python
+
+   df.dtypes
+
+
+.. _whatsnew_130.notable_bug_fixes.setitem_with_bool_casting:
+
+Consistent casting with setting into Boolean Series
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Setting non-boolean values into a :class:`Series` with ``dtype=bool`` now consistently
+casts to ``dtype=object`` (:issue:`38709`)
+
+.. code-block:: ipython
+
+   In [1]: orig = pd.Series([True, False])
+
+   In [2]: ser = orig.copy()
+
+   In [3]: ser.iloc[1] = np.nan
+
+   In [4]: ser2 = orig.copy()
+
+   In [5]: ser2.iloc[1] = 2.0
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [1]: ser
+   Out [1]:
+   0    1.0
+   1    NaN
+   dtype: float64
+
+   In [2]:ser2
+   Out [2]:
+   0    True
+   1     2.0
+   dtype: object
+
+*New behavior*:
+
+.. code-block:: ipython
+
+   In [1]: ser
+   Out [1]:
+   0    True
+   1     NaN
+   dtype: object
+
+   In [2]:ser2
+   Out [2]:
+   0    True
+   1     2.0
+   dtype: object
+
+
+.. _whatsnew_130.notable_bug_fixes.rolling_groupby_column:
+
+DataFrameGroupBy.rolling and SeriesGroupBy.rolling no longer return grouped-by column in values
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The group-by column will now be dropped from the result of a
+``groupby.rolling`` operation (:issue:`32262`)
+
+.. ipython:: python
+
+    df = pd.DataFrame({"A": [1, 1, 2, 3], "B": [0, 1, 2, 3]})
+    df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [1]: df.groupby("A").rolling(2).sum()
+    Out[1]:
+           A    B
+    A
+    1 0  NaN  NaN
+    1    2.0  1.0
+    2 2  NaN  NaN
+    3 3  NaN  NaN
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby("A").rolling(2).sum()
+
+.. _whatsnew_130.notable_bug_fixes.rolling_var_precision:
+
+Removed artificial truncation in rolling variance and standard deviation
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`.Rolling.std` and :meth:`.Rolling.var` will no longer
+artificially truncate results that are less than ``~1e-8`` and ``~1e-15`` respectively to
+zero (:issue:`37051`, :issue:`40448`, :issue:`39872`).
+
+However, floating point artifacts may now exist in the results when rolling over larger values.
+
+.. ipython:: python
+
+   s = pd.Series([7, 5, 5, 5])
+   s.rolling(3).var()
+
+.. _whatsnew_130.notable_bug_fixes.rolling_groupby_multiindex:
+
+DataFrameGroupBy.rolling and SeriesGroupBy.rolling with MultiIndex no longer drop levels in the result
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrameGroupBy.rolling` and :meth:`SeriesGroupBy.rolling` will no longer drop levels of a :class:`DataFrame`
+with a :class:`MultiIndex` in the result. This can lead to a perceived duplication of levels in the resulting
+:class:`MultiIndex`, but this change restores the behavior that was present in version 1.1.3 (:issue:`38787`, :issue:`38523`).
+
+
+.. ipython:: python
+
+   index = pd.MultiIndex.from_tuples([('idx1', 'idx2')], names=['label1', 'label2'])
+   df = pd.DataFrame({'a': [1], 'b': [2]}, index=index)
+   df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [1]: df.groupby('label1').rolling(1).sum()
+    Out[1]:
+              a    b
+    label1
+    idx1    1.0  2.0
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby('label1').rolling(1).sum()
+
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_130.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.17.3          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| pytz            | 2017.3          |    X     |         |
++-----------------+-----------------+----------+---------+
+| python-dateutil | 2.7.3           |    X     |         |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.2.1           |          |         |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.7.0           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pytest (dev)    | 6.0             |          |    X    |
++-----------------+-----------------+----------+---------+
+| mypy (dev)      | 0.812           |          |    X    |
++-----------------+-----------------+----------+---------+
+| setuptools      | 38.6.0          |          |    X    |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  | 4.6.0           |         |
++-----------------+-----------------+---------+
+| fastparquet     | 0.4.0           |    X    |
++-----------------+-----------------+---------+
+| fsspec          | 0.7.4           |         |
++-----------------+-----------------+---------+
+| gcsfs           | 0.6.0           |         |
++-----------------+-----------------+---------+
+| lxml            | 4.3.0           |         |
++-----------------+-----------------+---------+
+| matplotlib      | 2.2.3           |         |
++-----------------+-----------------+---------+
+| numba           | 0.46.0          |         |
++-----------------+-----------------+---------+
+| openpyxl        | 3.0.0           |    X    |
++-----------------+-----------------+---------+
+| pyarrow         | 0.17.0          |    X    |
++-----------------+-----------------+---------+
+| pymysql         | 0.8.1           |    X    |
++-----------------+-----------------+---------+
+| pytables        | 3.5.1           |         |
++-----------------+-----------------+---------+
+| s3fs            | 0.4.0           |         |
++-----------------+-----------------+---------+
+| scipy           | 1.2.0           |         |
++-----------------+-----------------+---------+
+| sqlalchemy      | 1.3.0           |    X    |
++-----------------+-----------------+---------+
+| tabulate        | 0.8.7           |    X    |
++-----------------+-----------------+---------+
+| xarray          | 0.12.0          |         |
++-----------------+-----------------+---------+
+| xlrd            | 1.2.0           |         |
++-----------------+-----------------+---------+
+| xlsxwriter      | 1.0.2           |         |
++-----------------+-----------------+---------+
+| xlwt            | 1.3.0           |         |
++-----------------+-----------------+---------+
+| pandas-gbq      | 0.12.0          |         |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_130.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- Partially initialized :class:`CategoricalDtype` objects (i.e. those with ``categories=None``) will no longer compare as equal to fully initialized dtype objects (:issue:`38516`)
+- Accessing ``_constructor_expanddim`` on a :class:`DataFrame` and ``_constructor_sliced`` on a :class:`Series` now raise an ``AttributeError``. Previously a ``NotImplementedError`` was raised (:issue:`38782`)
+- Added new ``engine`` and ``**engine_kwargs`` parameters to :meth:`DataFrame.to_sql` to support other future "SQL engines". Currently we still only use ``SQLAlchemy`` under the hood, but more engines are planned to be supported such as `turbodbc <https://turbodbc.readthedocs.io/en/latest/>`_ (:issue:`36893`)
+- Removed redundant ``freq`` from :class:`PeriodIndex` string representation (:issue:`41653`)
+- :meth:`ExtensionDtype.construct_array_type` is now a required method instead of an optional one for :class:`ExtensionDtype` subclasses (:issue:`24860`)
+- Calling ``hash`` on non-hashable pandas objects will now raise ``TypeError`` with the built-in error message (e.g. ``unhashable type: 'Series'``). Previously it would raise a custom message such as ``'Series' objects are mutable, thus they cannot be hashed``. Furthermore, ``isinstance(<Series>, abc.collections.Hashable)`` will now return ``False`` (:issue:`40013`)
+- :meth:`.Styler.from_custom_template` now has two new arguments for template names, and removed the old ``name``, due to template inheritance having been introducing for better parsing (:issue:`42053`). Subclassing modifications to Styler attributes are also needed.
+
+.. _whatsnew_130.api_breaking.build:
+
+Build
+^^^^^
+- Documentation in ``.pptx`` and ``.pdf`` formats are no longer included in wheels or source distributions. (:issue:`30741`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+.. _whatsnew_130.deprecations.nuisance_columns:
+
+Deprecated dropping nuisance columns in DataFrame reductions and DataFrameGroupBy operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Calling a reduction (e.g. ``.min``, ``.max``, ``.sum``) on a :class:`DataFrame` with
+``numeric_only=None`` (the default), columns where the reduction raises a ``TypeError``
+are silently ignored and dropped from the result.
+
+This behavior is deprecated. In a future version, the ``TypeError`` will be raised,
+and users will need to select only valid columns before calling the function.
+
+For example:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2, 3, 4], "B": pd.date_range("2016-01-01", periods=4)})
+   df
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.prod()
+    Out[3]:
+    Out[3]:
+    A    24
+    dtype: int64
+
+*Future behavior*:
+
+.. code-block:: ipython
+
+    In [4]: df.prod()
+    ...
+    TypeError: 'DatetimeArray' does not implement reduction 'prod'
+
+    In [5]: df[["A"]].prod()
+    Out[5]:
+    A    24
+    dtype: int64
+
+
+Similarly, when applying a function to :class:`DataFrameGroupBy`, columns on which
+the function raises ``TypeError`` are currently silently ignored and dropped
+from the result.
+
+This behavior is deprecated.  In a future version, the ``TypeError``
+will be raised, and users will need to select only valid columns before calling
+the function.
+
+For example:
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": [1, 2, 3, 4], "B": pd.date_range("2016-01-01", periods=4)})
+   gb = df.groupby([1, 1, 2, 2])
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+    In [4]: gb.prod(numeric_only=False)
+    Out[4]:
+    A
+    1   2
+    2  12
+
+*Future behavior*:
+
+.. code-block:: ipython
+
+    In [5]: gb.prod(numeric_only=False)
+    ...
+    TypeError: datetime64 type does not support prod operations
+
+    In [6]: gb[["A"]].prod(numeric_only=False)
+    Out[6]:
+        A
+    1   2
+    2  12
+
+.. _whatsnew_130.deprecations.other:
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+- Deprecated allowing scalars to be passed to the :class:`Categorical` constructor (:issue:`38433`)
+- Deprecated constructing :class:`CategoricalIndex` without passing list-like data (:issue:`38944`)
+- Deprecated allowing subclass-specific keyword arguments in the :class:`Index` constructor, use the specific subclass directly instead (:issue:`14093`, :issue:`21311`, :issue:`22315`, :issue:`26974`)
+- Deprecated the :meth:`astype` method of datetimelike (``timedelta64[ns]``, ``datetime64[ns]``, ``Datetime64TZDtype``, ``PeriodDtype``) to convert to integer dtypes, use ``values.view(...)`` instead (:issue:`38544`). This deprecation was later reverted in pandas 1.4.0.
+- Deprecated :meth:`MultiIndex.is_lexsorted` and :meth:`MultiIndex.lexsort_depth`, use :meth:`MultiIndex.is_monotonic_increasing` instead (:issue:`32259`)
+- Deprecated keyword ``try_cast`` in :meth:`Series.where`, :meth:`Series.mask`, :meth:`DataFrame.where`, :meth:`DataFrame.mask`; cast results manually if desired (:issue:`38836`)
+- Deprecated comparison of :class:`Timestamp` objects with ``datetime.date`` objects.  Instead of e.g. ``ts <= mydate`` use ``ts <= pd.Timestamp(mydate)`` or ``ts.date() <= mydate`` (:issue:`36131`)
+- Deprecated :attr:`Rolling.win_type` returning ``"freq"`` (:issue:`38963`)
+- Deprecated :attr:`Rolling.is_datetimelike` (:issue:`38963`)
+- Deprecated :class:`DataFrame` indexer for :meth:`Series.__setitem__` and :meth:`DataFrame.__setitem__` (:issue:`39004`)
+- Deprecated :meth:`ExponentialMovingWindow.vol` (:issue:`39220`)
+- Using ``.astype`` to convert between ``datetime64[ns]`` dtype and :class:`DatetimeTZDtype` is deprecated and will raise in a future version, use ``obj.tz_localize`` or ``obj.dt.tz_localize`` instead (:issue:`38622`)
+- Deprecated casting ``datetime.date`` objects to ``datetime64`` when used as ``fill_value`` in :meth:`DataFrame.unstack`, :meth:`DataFrame.shift`, :meth:`Series.shift`, and :meth:`DataFrame.reindex`, pass ``pd.Timestamp(dateobj)`` instead (:issue:`39767`)
+- Deprecated :meth:`.Styler.set_na_rep` and :meth:`.Styler.set_precision` in favor of :meth:`.Styler.format` with ``na_rep`` and ``precision`` as existing and new input arguments respectively (:issue:`40134`, :issue:`40425`)
+- Deprecated :meth:`.Styler.where` in favor of using an alternative formulation with :meth:`Styler.applymap` (:issue:`40821`)
+- Deprecated allowing partial failure in :meth:`Series.transform` and :meth:`DataFrame.transform` when ``func`` is list-like or dict-like and raises anything but ``TypeError``; ``func`` raising anything but a ``TypeError`` will raise in a future version (:issue:`40211`)
+- Deprecated arguments ``error_bad_lines`` and ``warn_bad_lines`` in :meth:`read_csv` and :meth:`read_table` in favor of argument ``on_bad_lines`` (:issue:`15122`)
+- Deprecated support for ``np.ma.mrecords.MaskedRecords`` in the :class:`DataFrame` constructor, pass ``{name: data[name] for name in data.dtype.names}`` instead (:issue:`40363`)
+- Deprecated using :func:`merge`, :meth:`DataFrame.merge`, and :meth:`DataFrame.join` on a different number of levels (:issue:`34862`)
+- Deprecated the use of ``**kwargs`` in :class:`.ExcelWriter`; use the keyword argument ``engine_kwargs`` instead (:issue:`40430`)
+- Deprecated the ``level`` keyword for :class:`DataFrame` and :class:`Series` aggregations; use groupby instead (:issue:`39983`)
+- Deprecated the ``inplace`` parameter of :meth:`Categorical.remove_categories`, :meth:`Categorical.add_categories`, :meth:`Categorical.reorder_categories`, :meth:`Categorical.rename_categories`, :meth:`Categorical.set_categories` and will be removed in a future version (:issue:`37643`)
+- Deprecated :func:`merge` producing duplicated columns through the ``suffixes`` keyword  and already existing columns (:issue:`22818`)
+- Deprecated setting :attr:`Categorical._codes`, create a new :class:`Categorical` with the desired codes instead (:issue:`40606`)
+- Deprecated the ``convert_float`` optional argument in :func:`read_excel` and :meth:`ExcelFile.parse` (:issue:`41127`)
+- Deprecated behavior of :meth:`DatetimeIndex.union` with mixed timezones; in a future version both will be cast to UTC instead of object dtype (:issue:`39328`)
+- Deprecated using ``usecols`` with out of bounds indices for :func:`read_csv` with ``engine="c"`` (:issue:`25623`)
+- Deprecated special treatment of lists with first element a Categorical in the :class:`DataFrame` constructor; pass as ``pd.DataFrame({col: categorical, ...})`` instead (:issue:`38845`)
+- Deprecated behavior of :class:`DataFrame` constructor when a ``dtype`` is passed and the data cannot be cast to that dtype. In a future version, this will raise instead of being silently ignored (:issue:`24435`)
+- Deprecated the :attr:`Timestamp.freq` attribute.  For the properties that use it (``is_month_start``, ``is_month_end``, ``is_quarter_start``, ``is_quarter_end``, ``is_year_start``, ``is_year_end``), when you have a ``freq``, use e.g. ``freq.is_month_start(ts)`` (:issue:`15146`)
+- Deprecated construction of :class:`Series` or :class:`DataFrame` with ``DatetimeTZDtype`` data and ``datetime64[ns]`` dtype.  Use ``Series(data).dt.tz_localize(None)`` instead (:issue:`41555`, :issue:`33401`)
+- Deprecated behavior of :class:`Series` construction with large-integer values and small-integer dtype silently overflowing; use ``Series(data).astype(dtype)`` instead (:issue:`41734`)
+- Deprecated behavior of :class:`DataFrame` construction with floating data and integer dtype casting even when lossy; in a future version this will remain floating, matching :class:`Series` behavior (:issue:`41770`)
+- Deprecated inference of ``timedelta64[ns]``, ``datetime64[ns]``, or ``DatetimeTZDtype`` dtypes in :class:`Series` construction when data containing strings is passed and no ``dtype`` is passed (:issue:`33558`)
+- In a future version, constructing :class:`Series` or :class:`DataFrame` with ``datetime64[ns]`` data and ``DatetimeTZDtype`` will treat the data as wall-times instead of as UTC times (matching DatetimeIndex behavior). To treat the data as UTC times, use ``pd.Series(data).dt.tz_localize("UTC").dt.tz_convert(dtype.tz)`` or ``pd.Series(data.view("int64"), dtype=dtype)`` (:issue:`33401`)
+- Deprecated passing lists as ``key`` to :meth:`DataFrame.xs` and :meth:`Series.xs` (:issue:`41760`)
+- Deprecated boolean arguments of ``inclusive`` in :meth:`Series.between` to have ``{"left", "right", "neither", "both"}`` as standard argument values (:issue:`40628`)
+- Deprecated passing arguments as positional for all of the following, with exceptions noted (:issue:`41485`):
+
+  - :func:`concat` (other than ``objs``)
+  - :func:`read_csv` (other than ``filepath_or_buffer``)
+  - :func:`read_table` (other than ``filepath_or_buffer``)
+  - :meth:`DataFrame.clip` and :meth:`Series.clip` (other than ``upper`` and ``lower``)
+  - :meth:`DataFrame.drop_duplicates` (except for ``subset``), :meth:`Series.drop_duplicates`, :meth:`Index.drop_duplicates` and :meth:`MultiIndex.drop_duplicates`
+  - :meth:`DataFrame.drop` (other than ``labels``) and :meth:`Series.drop`
+  - :meth:`DataFrame.dropna` and :meth:`Series.dropna`
+  - :meth:`DataFrame.ffill`, :meth:`Series.ffill`, :meth:`DataFrame.bfill`, and :meth:`Series.bfill`
+  - :meth:`DataFrame.fillna` and :meth:`Series.fillna` (apart from ``value``)
+  - :meth:`DataFrame.interpolate` and :meth:`Series.interpolate` (other than ``method``)
+  - :meth:`DataFrame.mask` and :meth:`Series.mask` (other than ``cond`` and ``other``)
+  - :meth:`DataFrame.reset_index` (other than ``level``) and :meth:`Series.reset_index`
+  - :meth:`DataFrame.set_axis` and :meth:`Series.set_axis` (other than ``labels``)
+  - :meth:`DataFrame.set_index` (other than ``keys``)
+  - :meth:`DataFrame.sort_index` and :meth:`Series.sort_index`
+  - :meth:`DataFrame.sort_values` (other than ``by``) and :meth:`Series.sort_values`
+  - :meth:`DataFrame.where` and :meth:`Series.where` (other than ``cond`` and ``other``)
+  - :meth:`Index.set_names` and :meth:`MultiIndex.set_names` (except for ``names``)
+  - :meth:`MultiIndex.codes` (except for ``codes``)
+  - :meth:`MultiIndex.set_levels` (except for ``levels``)
+  - :meth:`Resampler.interpolate` (other than ``method``)
+
+
+.. ---------------------------------------------------------------------------
+
+
+.. _whatsnew_130.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :meth:`IntervalIndex.isin` (:issue:`38353`)
+- Performance improvement in :meth:`Series.mean` for nullable data types (:issue:`34814`)
+- Performance improvement in :meth:`Series.isin` for nullable data types (:issue:`38340`)
+- Performance improvement in :meth:`DataFrame.fillna` with ``method="pad"`` or ``method="backfill"`` for nullable floating and nullable integer dtypes (:issue:`39953`)
+- Performance improvement in :meth:`DataFrame.corr` for ``method=kendall`` (:issue:`28329`)
+- Performance improvement in :meth:`DataFrame.corr` for ``method=spearman`` (:issue:`40956`, :issue:`41885`)
+- Performance improvement in :meth:`.Rolling.corr` and :meth:`.Rolling.cov` (:issue:`39388`)
+- Performance improvement in :meth:`.RollingGroupby.corr`, :meth:`.ExpandingGroupby.corr`, :meth:`.ExpandingGroupby.corr` and :meth:`.ExpandingGroupby.cov` (:issue:`39591`)
+- Performance improvement in :func:`unique` for object data type (:issue:`37615`)
+- Performance improvement in :func:`json_normalize` for basic cases (including separators) (:issue:`40035` :issue:`15621`)
+- Performance improvement in :class:`.ExpandingGroupby` aggregation methods (:issue:`39664`)
+- Performance improvement in :class:`.Styler` where render times are more than 50% reduced and now matches :meth:`DataFrame.to_html` (:issue:`39972` :issue:`39952`, :issue:`40425`)
+- The method :meth:`.Styler.set_td_classes` is now as performant as :meth:`.Styler.apply` and :meth:`.Styler.applymap`, and even more so in some cases (:issue:`40453`)
+- Performance improvement in :meth:`.ExponentialMovingWindow.mean` with ``times`` (:issue:`39784`)
+- Performance improvement in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` when requiring the Python fallback implementation (:issue:`40176`)
+- Performance improvement in the conversion of a PyArrow Boolean array to a pandas nullable Boolean array (:issue:`41051`)
+- Performance improvement for concatenation of data with type :class:`CategoricalDtype` (:issue:`40193`)
+- Performance improvement in :meth:`.DataFrameGroupBy.cummin`, :meth:`.SeriesGroupBy.cummin`, :meth:`.DataFrameGroupBy.cummax`, and :meth:`.SeriesGroupBy.cummax` with nullable data types (:issue:`37493`)
+- Performance improvement in :meth:`Series.nunique` with nan values (:issue:`40865`)
+- Performance improvement in :meth:`DataFrame.transpose`, :meth:`Series.unstack` with ``DatetimeTZDtype`` (:issue:`40149`)
+- Performance improvement in :meth:`Series.plot` and :meth:`DataFrame.plot` with entry point lazy loading (:issue:`41492`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in :class:`CategoricalIndex` incorrectly failing to raise ``TypeError`` when scalar data is passed (:issue:`38614`)
+- Bug in ``CategoricalIndex.reindex`` failed when the :class:`Index` passed was not categorical but whose values were all labels in the category (:issue:`28690`)
+- Bug where constructing a :class:`Categorical` from an object-dtype array of ``date`` objects did not round-trip correctly with ``astype`` (:issue:`38552`)
+- Bug in constructing a :class:`DataFrame` from an ``ndarray`` and a :class:`CategoricalDtype` (:issue:`38857`)
+- Bug in setting categorical values into an object-dtype column in a :class:`DataFrame` (:issue:`39136`)
+- Bug in :meth:`DataFrame.reindex` was raising an ``IndexError`` when the new index contained duplicates and the old index was a :class:`CategoricalIndex` (:issue:`38906`)
+- Bug in :meth:`Categorical.fillna` with a tuple-like category raising ``NotImplementedError`` instead of ``ValueError`` when filling with a non-category tuple (:issue:`41914`)
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :class:`DataFrame` and :class:`Series` constructors sometimes dropping nanoseconds from :class:`Timestamp` (resp. :class:`Timedelta`) ``data``, with ``dtype=datetime64[ns]`` (resp. ``timedelta64[ns]``) (:issue:`38032`)
+- Bug in :meth:`DataFrame.first` and :meth:`Series.first` with an offset of one month returning an incorrect result when the first day is the last day of a month (:issue:`29623`)
+- Bug in constructing a :class:`DataFrame` or :class:`Series` with mismatched ``datetime64`` data and ``timedelta64`` dtype, or vice-versa, failing to raise a ``TypeError`` (:issue:`38575`, :issue:`38764`, :issue:`38792`)
+- Bug in constructing a :class:`Series` or :class:`DataFrame` with a ``datetime`` object out of bounds for ``datetime64[ns]`` dtype or a ``timedelta`` object out of bounds for ``timedelta64[ns]`` dtype (:issue:`38792`, :issue:`38965`)
+- Bug in :meth:`DatetimeIndex.intersection`, :meth:`DatetimeIndex.symmetric_difference`, :meth:`PeriodIndex.intersection`, :meth:`PeriodIndex.symmetric_difference` always returning object-dtype when operating with :class:`CategoricalIndex` (:issue:`38741`)
+- Bug in :meth:`DatetimeIndex.intersection` giving incorrect results with non-Tick frequencies with ``n != 1`` (:issue:`42104`)
+- Bug in :meth:`Series.where` incorrectly casting ``datetime64`` values to ``int64`` (:issue:`37682`)
+- Bug in :class:`Categorical` incorrectly typecasting ``datetime`` object to ``Timestamp`` (:issue:`38878`)
+- Bug in comparisons between :class:`Timestamp` object and ``datetime64`` objects just outside the implementation bounds for nanosecond ``datetime64`` (:issue:`39221`)
+- Bug in :meth:`Timestamp.round`, :meth:`Timestamp.floor`, :meth:`Timestamp.ceil` for values near the implementation bounds of :class:`Timestamp` (:issue:`39244`)
+- Bug in :meth:`Timedelta.round`, :meth:`Timedelta.floor`, :meth:`Timedelta.ceil` for values near the implementation bounds of :class:`Timedelta` (:issue:`38964`)
+- Bug in :func:`date_range` incorrectly creating :class:`DatetimeIndex` containing ``NaT`` instead of raising ``OutOfBoundsDatetime`` in corner cases (:issue:`24124`)
+- Bug in :func:`infer_freq` incorrectly fails to infer 'H' frequency of :class:`DatetimeIndex` if the latter has a timezone and crosses DST boundaries (:issue:`39556`)
+- Bug in :class:`Series` backed by :class:`DatetimeArray` or :class:`TimedeltaArray` sometimes failing to set the array's ``freq`` to ``None`` (:issue:`41425`)
+
+Timedelta
+^^^^^^^^^
+- Bug in constructing :class:`Timedelta` from ``np.timedelta64`` objects with non-nanosecond units that are out of bounds for ``timedelta64[ns]`` (:issue:`38965`)
+- Bug in constructing a :class:`TimedeltaIndex` incorrectly accepting ``np.datetime64("NaT")`` objects (:issue:`39462`)
+- Bug in constructing :class:`Timedelta` from an input string with only symbols and no digits failed to raise an error (:issue:`39710`)
+- Bug in :class:`TimedeltaIndex` and :func:`to_timedelta` failing to raise when passed non-nanosecond ``timedelta64`` arrays that overflow when converting to ``timedelta64[ns]`` (:issue:`40008`)
+
+Timezones
+^^^^^^^^^
+- Bug in different ``tzinfo`` objects representing UTC not being treated as equivalent (:issue:`39216`)
+- Bug in ``dateutil.tz.gettz("UTC")`` not being recognized as equivalent to other UTC-representing tzinfos (:issue:`39276`)
+
+Numeric
+^^^^^^^
+- Bug in :meth:`DataFrame.quantile`, :meth:`DataFrame.sort_values` causing incorrect subsequent indexing behavior (:issue:`38351`)
+- Bug in :meth:`DataFrame.sort_values` raising an :class:`IndexError` for empty ``by`` (:issue:`40258`)
+- Bug in :meth:`DataFrame.select_dtypes` with ``include=np.number`` would drop numeric ``ExtensionDtype`` columns (:issue:`35340`)
+- Bug in :meth:`DataFrame.mode` and :meth:`Series.mode` not keeping consistent integer :class:`Index` for empty input (:issue:`33321`)
+- Bug in :meth:`DataFrame.rank` when the DataFrame contained ``np.inf`` (:issue:`32593`)
+- Bug in :meth:`DataFrame.rank` with ``axis=0`` and columns holding incomparable types raising an ``IndexError`` (:issue:`38932`)
+- Bug in :meth:`Series.rank`, :meth:`DataFrame.rank`, :meth:`.DataFrameGroupBy.rank`, and :meth:`.SeriesGroupBy.rank` treating the most negative ``int64`` value as missing (:issue:`32859`)
+- Bug in :meth:`DataFrame.select_dtypes` different behavior between Windows and Linux with ``include="int"`` (:issue:`36596`)
+- Bug in :meth:`DataFrame.apply` and :meth:`DataFrame.agg` when passed the argument ``func="size"`` would operate on the entire ``DataFrame`` instead of rows or columns (:issue:`39934`)
+- Bug in :meth:`DataFrame.transform` would raise a ``SpecificationError`` when passed a dictionary and columns were missing; will now raise a ``KeyError`` instead (:issue:`40004`)
+- Bug in :meth:`.DataFrameGroupBy.rank` and :meth:`.SeriesGroupBy.rank` giving incorrect results with ``pct=True`` and equal values between consecutive groups (:issue:`40518`)
+- Bug in :meth:`Series.count` would result in an ``int32`` result on 32-bit platforms when argument ``level=None`` (:issue:`40908`)
+- Bug in :class:`Series` and :class:`DataFrame` reductions with methods ``any`` and ``all`` not returning Boolean results for object data (:issue:`12863`, :issue:`35450`, :issue:`27709`)
+- Bug in :meth:`Series.clip` would fail if the Series contains NA values and has nullable int or float as a data type (:issue:`40851`)
+- Bug in :meth:`UInt64Index.where` and :meth:`UInt64Index.putmask` with an ``np.int64`` dtype ``other`` incorrectly raising ``TypeError`` (:issue:`41974`)
+- Bug in :meth:`DataFrame.agg` not sorting the aggregated axis in the order of the provided aggregation functions when one or more aggregation function fails to produce results (:issue:`33634`)
+- Bug in :meth:`DataFrame.clip` not interpreting missing values as no threshold (:issue:`40420`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :meth:`Series.to_dict` with ``orient='records'`` now returns Python native types (:issue:`25969`)
+- Bug in :meth:`Series.view` and :meth:`Index.view` when converting between datetime-like (``datetime64[ns]``, ``datetime64[ns, tz]``, ``timedelta64``, ``period``) dtypes (:issue:`39788`)
+- Bug in creating a :class:`DataFrame` from an empty ``np.recarray`` not retaining the original dtypes (:issue:`40121`)
+- Bug in :class:`DataFrame` failing to raise a ``TypeError`` when constructing from a ``frozenset`` (:issue:`40163`)
+- Bug in :class:`Index` construction silently ignoring a passed ``dtype`` when the data cannot be cast to that dtype (:issue:`21311`)
+- Bug in :meth:`StringArray.astype` falling back to NumPy and raising when converting to ``dtype='categorical'`` (:issue:`40450`)
+- Bug in :func:`factorize` where, when given an array with a numeric NumPy dtype lower than int64, uint64 and float64, the unique values did not keep their original dtype (:issue:`41132`)
+- Bug in :class:`DataFrame` construction with a dictionary containing an array-like with ``ExtensionDtype`` and ``copy=True`` failing to make a copy (:issue:`38939`)
+- Bug in :meth:`qcut` raising error when taking ``Float64DType`` as input (:issue:`40730`)
+- Bug in :class:`DataFrame` and :class:`Series` construction with ``datetime64[ns]`` data and ``dtype=object`` resulting in ``datetime`` objects instead of :class:`Timestamp` objects (:issue:`41599`)
+- Bug in :class:`DataFrame` and :class:`Series` construction with ``timedelta64[ns]`` data and ``dtype=object`` resulting in ``np.timedelta64`` objects instead of :class:`Timedelta` objects (:issue:`41599`)
+- Bug in :class:`DataFrame` construction when given a two-dimensional object-dtype ``np.ndarray`` of :class:`Period` or :class:`Interval` objects failing to cast to :class:`PeriodDtype` or :class:`IntervalDtype`, respectively (:issue:`41812`)
+- Bug in constructing a :class:`Series` from a list and a :class:`PandasDtype` (:issue:`39357`)
+- Bug in creating a :class:`Series` from a ``range`` object that does not fit in the bounds of ``int64`` dtype (:issue:`30173`)
+- Bug in creating a :class:`Series` from a ``dict`` with all-tuple keys and an :class:`Index` that requires reindexing (:issue:`41707`)
+- Bug in :func:`.infer_dtype` not recognizing Series, Index, or array with a Period dtype (:issue:`23553`)
+- Bug in :func:`.infer_dtype` raising an error for general :class:`.ExtensionArray` objects. It will now return ``"unknown-array"`` instead of raising (:issue:`37367`)
+- Bug in :meth:`DataFrame.convert_dtypes` incorrectly raised a ``ValueError`` when called on an empty DataFrame (:issue:`40393`)
+
+Strings
+^^^^^^^
+- Bug in the conversion from ``pyarrow.ChunkedArray`` to :class:`~arrays.StringArray` when the original had zero chunks (:issue:`41040`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` ignoring replacements with ``regex=True`` for ``StringDType`` data (:issue:`41333`, :issue:`35977`)
+- Bug in :meth:`Series.str.extract` with :class:`~arrays.StringArray` returning object dtype for an empty :class:`DataFrame` (:issue:`41441`)
+- Bug in :meth:`Series.str.replace` where the ``case`` argument was ignored when ``regex=False`` (:issue:`41602`)
+
+Interval
+^^^^^^^^
+- Bug in :meth:`IntervalIndex.intersection` and :meth:`IntervalIndex.symmetric_difference` always returning object-dtype when operating with :class:`CategoricalIndex` (:issue:`38653`, :issue:`38741`)
+- Bug in :meth:`IntervalIndex.intersection` returning duplicates when at least one of the :class:`Index` objects have duplicates which are present in the other (:issue:`38743`)
+- :meth:`IntervalIndex.union`, :meth:`IntervalIndex.intersection`, :meth:`IntervalIndex.difference`, and :meth:`IntervalIndex.symmetric_difference` now cast to the appropriate dtype instead of raising a ``TypeError`` when operating with another :class:`IntervalIndex` with incompatible dtype (:issue:`39267`)
+- :meth:`PeriodIndex.union`, :meth:`PeriodIndex.intersection`, :meth:`PeriodIndex.symmetric_difference`, :meth:`PeriodIndex.difference` now cast to object dtype instead of raising ``IncompatibleFrequency`` when operating with another :class:`PeriodIndex` with incompatible dtype (:issue:`39306`)
+- Bug in :meth:`IntervalIndex.is_monotonic`, :meth:`IntervalIndex.get_loc`, :meth:`IntervalIndex.get_indexer_for`, and :meth:`IntervalIndex.__contains__` when NA values are present (:issue:`41831`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`Index.union` and :meth:`MultiIndex.union` dropping duplicate ``Index`` values when ``Index`` was not monotonic or ``sort`` was set to ``False`` (:issue:`36289`, :issue:`31326`, :issue:`40862`)
+- Bug in :meth:`CategoricalIndex.get_indexer` failing to raise ``InvalidIndexError`` when non-unique (:issue:`38372`)
+- Bug in :meth:`IntervalIndex.get_indexer` when ``target`` has ``CategoricalDtype`` and both the index and the target contain NA values (:issue:`41934`)
+- Bug in :meth:`Series.loc` raising a ``ValueError`` when input was filtered with a Boolean list and values to set were a list with lower dimension (:issue:`20438`)
+- Bug in inserting many new columns into a :class:`DataFrame` causing incorrect subsequent indexing behavior (:issue:`38380`)
+- Bug in :meth:`DataFrame.__setitem__` raising a ``ValueError`` when setting multiple values to duplicate columns (:issue:`15695`)
+- Bug in :meth:`DataFrame.loc`, :meth:`Series.loc`, :meth:`DataFrame.__getitem__` and :meth:`Series.__getitem__` returning incorrect elements for non-monotonic :class:`DatetimeIndex` for string slices (:issue:`33146`)
+- Bug in :meth:`DataFrame.reindex` and :meth:`Series.reindex` with timezone aware indexes raising a ``TypeError`` for ``method="ffill"`` and ``method="bfill"`` and specified ``tolerance`` (:issue:`38566`)
+- Bug in :meth:`DataFrame.reindex` with ``datetime64[ns]`` or ``timedelta64[ns]`` incorrectly casting to integers when the ``fill_value`` requires casting to object dtype (:issue:`39755`)
+- Bug in :meth:`DataFrame.__setitem__` raising a ``ValueError`` when setting on an empty :class:`DataFrame` using specified columns and a nonempty :class:`DataFrame` value (:issue:`38831`)
+- Bug in :meth:`DataFrame.loc.__setitem__` raising a ``ValueError`` when operating on a unique column when the :class:`DataFrame` has duplicate columns (:issue:`38521`)
+- Bug in :meth:`DataFrame.iloc.__setitem__` and :meth:`DataFrame.loc.__setitem__` with mixed dtypes when setting with a dictionary value (:issue:`38335`)
+- Bug in :meth:`Series.loc.__setitem__` and :meth:`DataFrame.loc.__setitem__` raising ``KeyError`` when provided a Boolean generator (:issue:`39614`)
+- Bug in :meth:`Series.iloc` and :meth:`DataFrame.iloc` raising a ``KeyError`` when provided a generator (:issue:`39614`)
+- Bug in :meth:`DataFrame.__setitem__` not raising a ``ValueError`` when the right hand side is a :class:`DataFrame` with wrong number of columns (:issue:`38604`)
+- Bug in :meth:`Series.__setitem__` raising a ``ValueError`` when setting a :class:`Series` with a scalar indexer (:issue:`38303`)
+- Bug in :meth:`DataFrame.loc` dropping levels of a :class:`MultiIndex` when the :class:`DataFrame` used as input has only one row (:issue:`10521`)
+- Bug in :meth:`DataFrame.__getitem__` and :meth:`Series.__getitem__` always raising ``KeyError`` when slicing with existing strings where the :class:`Index` has milliseconds (:issue:`33589`)
+- Bug in setting ``timedelta64`` or ``datetime64`` values into numeric :class:`Series` failing to cast to object dtype (:issue:`39086`, :issue:`39619`)
+- Bug in setting :class:`Interval` values into a :class:`Series` or :class:`DataFrame` with mismatched :class:`IntervalDtype` incorrectly casting the new values to the existing dtype (:issue:`39120`)
+- Bug in setting ``datetime64`` values into a :class:`Series` with integer-dtype incorrectly casting the datetime64 values to integers (:issue:`39266`)
+- Bug in setting ``np.datetime64("NaT")`` into a :class:`Series` with :class:`Datetime64TZDtype` incorrectly treating the timezone-naive value as timezone-aware (:issue:`39769`)
+- Bug in :meth:`Index.get_loc` not raising ``KeyError`` when ``key=NaN`` and ``method`` is specified but ``NaN`` is not in the :class:`Index` (:issue:`39382`)
+- Bug in :meth:`DatetimeIndex.insert` when inserting ``np.datetime64("NaT")`` into a timezone-aware index incorrectly treating the timezone-naive value as timezone-aware (:issue:`39769`)
+- Bug in incorrectly raising in :meth:`Index.insert`, when setting a new column that cannot be held in the existing ``frame.columns``, or in :meth:`Series.reset_index` or :meth:`DataFrame.reset_index` instead of casting to a compatible dtype (:issue:`39068`)
+- Bug in :meth:`RangeIndex.append` where a single object of length 1 was concatenated incorrectly (:issue:`39401`)
+- Bug in :meth:`RangeIndex.astype` where when converting to :class:`CategoricalIndex`, the categories became a :class:`Int64Index` instead of a :class:`RangeIndex` (:issue:`41263`)
+- Bug in setting ``numpy.timedelta64`` values into an object-dtype :class:`Series` using a Boolean indexer (:issue:`39488`)
+- Bug in setting numeric values into a into a boolean-dtypes :class:`Series` using ``at`` or ``iat`` failing to cast to object-dtype (:issue:`39582`)
+- Bug in :meth:`DataFrame.__setitem__` and :meth:`DataFrame.iloc.__setitem__` raising ``ValueError`` when trying to index with a row-slice and setting a list as values (:issue:`40440`)
+- Bug in :meth:`DataFrame.loc` not raising ``KeyError`` when the key was not found in :class:`MultiIndex` and the levels were not fully specified (:issue:`41170`)
+- Bug in :meth:`DataFrame.loc.__setitem__` when setting-with-expansion incorrectly raising when the index in the expanding axis contained duplicates (:issue:`40096`)
+- Bug in :meth:`DataFrame.loc.__getitem__` with :class:`MultiIndex` casting to float when at least one index column has float dtype and we retrieve a scalar (:issue:`41369`)
+- Bug in :meth:`DataFrame.loc` incorrectly matching non-Boolean index elements (:issue:`20432`)
+- Bug in indexing with ``np.nan`` on a :class:`Series` or :class:`DataFrame` with a :class:`CategoricalIndex` incorrectly raising ``KeyError`` when ``np.nan`` keys are present (:issue:`41933`)
+- Bug in :meth:`Series.__delitem__` with ``ExtensionDtype`` incorrectly casting to ``ndarray`` (:issue:`40386`)
+- Bug in :meth:`DataFrame.at` with a :class:`CategoricalIndex` returning incorrect results when passed integer keys (:issue:`41846`)
+- Bug in :meth:`DataFrame.loc` returning a :class:`MultiIndex` in the wrong order if an indexer has duplicates (:issue:`40978`)
+- Bug in :meth:`DataFrame.__setitem__` raising a ``TypeError`` when using a ``str`` subclass as the column name with a :class:`DatetimeIndex` (:issue:`37366`)
+- Bug in :meth:`PeriodIndex.get_loc` failing to raise a ``KeyError`` when given a :class:`Period` with a mismatched ``freq`` (:issue:`41670`)
+- Bug ``.loc.__getitem__`` with a :class:`UInt64Index` and negative-integer keys raising ``OverflowError`` instead of ``KeyError`` in some cases, wrapping around to positive integers in others (:issue:`41777`)
+- Bug in :meth:`Index.get_indexer` failing to raise ``ValueError`` in some cases with invalid ``method``, ``limit``, or ``tolerance`` arguments (:issue:`41918`)
+- Bug when slicing a :class:`Series` or :class:`DataFrame` with a :class:`TimedeltaIndex` when passing an invalid string raising ``ValueError`` instead of a ``TypeError`` (:issue:`41821`)
+- Bug in :class:`Index` constructor sometimes silently ignoring a specified ``dtype`` (:issue:`38879`)
+- :meth:`Index.where` behavior now mirrors :meth:`Index.putmask` behavior, i.e. ``index.where(mask, other)`` matches ``index.putmask(~mask, other)`` (:issue:`39412`)
+
+Missing
+^^^^^^^
+- Bug in :class:`Grouper` did not correctly propagate the ``dropna`` argument; :meth:`.DataFrameGroupBy.transform` now correctly handles missing values for ``dropna=True`` (:issue:`35612`)
+- Bug in :func:`isna`, :meth:`Series.isna`, :meth:`Index.isna`, :meth:`DataFrame.isna`, and the corresponding ``notna`` functions not recognizing ``Decimal("NaN")`` objects (:issue:`39409`)
+- Bug in :meth:`DataFrame.fillna` not accepting a dictionary for the ``downcast`` keyword (:issue:`40809`)
+- Bug in :func:`isna` not returning a copy of the mask for nullable types, causing any subsequent mask modification to change the original array (:issue:`40935`)
+- Bug in :class:`DataFrame` construction with float data containing ``NaN`` and an integer ``dtype`` casting instead of retaining the ``NaN`` (:issue:`26919`)
+- Bug in :meth:`Series.isin` and :meth:`MultiIndex.isin` didn't treat all nans as equivalent if they were in tuples (:issue:`41836`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`DataFrame.drop` raising a ``TypeError`` when the :class:`MultiIndex` is non-unique and ``level`` is not provided (:issue:`36293`)
+- Bug in :meth:`MultiIndex.intersection` duplicating ``NaN`` in the result (:issue:`38623`)
+- Bug in :meth:`MultiIndex.equals` incorrectly returning ``True`` when the :class:`MultiIndex` contained ``NaN`` even when they are differently ordered (:issue:`38439`)
+- Bug in :meth:`MultiIndex.intersection` always returning an empty result when intersecting with :class:`CategoricalIndex` (:issue:`38653`)
+- Bug in :meth:`MultiIndex.difference` incorrectly raising ``TypeError`` when indexes contain non-sortable entries (:issue:`41915`)
+- Bug in :meth:`MultiIndex.reindex` raising a ``ValueError`` when used on an empty :class:`MultiIndex` and indexing only a specific level (:issue:`41170`)
+- Bug in :meth:`MultiIndex.reindex` raising ``TypeError`` when reindexing against a flat :class:`Index` (:issue:`41707`)
+
+I/O
+^^^
+- Bug in :meth:`Index.__repr__` when ``display.max_seq_items=1`` (:issue:`38415`)
+- Bug in :func:`read_csv` not recognizing scientific notation if the argument ``decimal`` is set and ``engine="python"`` (:issue:`31920`)
+- Bug in :func:`read_csv` interpreting ``NA`` value as comment, when ``NA`` does contain the comment string fixed for ``engine="python"`` (:issue:`34002`)
+- Bug in :func:`read_csv` raising an ``IndexError`` with multiple header columns and ``index_col`` is specified when the file has no data rows (:issue:`38292`)
+- Bug in :func:`read_csv` not accepting ``usecols`` with a different length than ``names`` for ``engine="python"`` (:issue:`16469`)
+- Bug in :meth:`read_csv` returning object dtype when ``delimiter=","`` with ``usecols`` and ``parse_dates`` specified for ``engine="python"`` (:issue:`35873`)
+- Bug in :func:`read_csv` raising a ``TypeError`` when ``names`` and ``parse_dates`` is specified for ``engine="c"`` (:issue:`33699`)
+- Bug in :func:`read_clipboard` and :func:`DataFrame.to_clipboard` not working in WSL (:issue:`38527`)
+- Allow custom error values for the ``parse_dates`` argument of :func:`read_sql`, :func:`read_sql_query` and :func:`read_sql_table` (:issue:`35185`)
+- Bug in :meth:`DataFrame.to_hdf` and :meth:`Series.to_hdf` raising a ``KeyError`` when trying to apply for subclasses of ``DataFrame`` or ``Series`` (:issue:`33748`)
+- Bug in :meth:`.HDFStore.put` raising a wrong ``TypeError`` when saving a DataFrame with non-string dtype (:issue:`34274`)
+- Bug in :func:`json_normalize` resulting in the first element of a generator object not being included in the returned DataFrame (:issue:`35923`)
+- Bug in :func:`read_csv` applying the thousands separator to date columns when the column should be parsed for dates and ``usecols`` is specified for ``engine="python"`` (:issue:`39365`)
+- Bug in :func:`read_excel` forward filling :class:`MultiIndex` names when multiple header and index columns are specified (:issue:`34673`)
+- Bug in :func:`read_excel` not respecting :func:`set_option` (:issue:`34252`)
+- Bug in :func:`read_csv` not switching ``true_values`` and ``false_values`` for nullable Boolean dtype (:issue:`34655`)
+- Bug in :func:`read_json` when ``orient="split"`` not maintaining a numeric string index (:issue:`28556`)
+- :meth:`read_sql` returned an empty generator if ``chunksize`` was non-zero and the query returned no results. Now returns a generator with a single empty DataFrame (:issue:`34411`)
+- Bug in :func:`read_hdf` returning unexpected records when filtering on categorical string columns using the ``where`` parameter (:issue:`39189`)
+- Bug in :func:`read_sas` raising a ``ValueError`` when ``datetimes`` were null (:issue:`39725`)
+- Bug in :func:`read_excel` dropping empty values from single-column spreadsheets (:issue:`39808`)
+- Bug in :func:`read_excel` loading trailing empty rows/columns for some filetypes (:issue:`41167`)
+- Bug in :func:`read_excel` raising an ``AttributeError`` when the excel file had a ``MultiIndex`` header followed by two empty rows and no index (:issue:`40442`)
+- Bug in :func:`read_excel`, :func:`read_csv`, :func:`read_table`, :func:`read_fwf`, and :func:`read_clipboard` where one blank row after a ``MultiIndex`` header with no index would be dropped (:issue:`40442`)
+- Bug in :meth:`DataFrame.to_string` misplacing the truncation column when ``index=False`` (:issue:`40904`)
+- Bug in :meth:`DataFrame.to_string` adding an extra dot and misaligning the truncation row when ``index=False`` (:issue:`40904`)
+- Bug in :func:`read_orc` always raising an ``AttributeError`` (:issue:`40918`)
+- Bug in :func:`read_csv` and :func:`read_table` silently ignoring ``prefix`` if ``names`` and ``prefix`` are defined, now raising a ``ValueError`` (:issue:`39123`)
+- Bug in :func:`read_csv` and :func:`read_excel` not respecting the dtype for a duplicated column name when ``mangle_dupe_cols`` is set to ``True`` (:issue:`35211`)
+- Bug in :func:`read_csv` silently ignoring ``sep`` if ``delimiter`` and ``sep`` are defined, now raising a ``ValueError`` (:issue:`39823`)
+- Bug in :func:`read_csv` and :func:`read_table` misinterpreting arguments when ``sys.setprofile`` had been previously called (:issue:`41069`)
+- Bug in the conversion from PyArrow to pandas (e.g. for reading Parquet) with nullable dtypes and a PyArrow array whose data buffer size is not a multiple of the dtype size (:issue:`40896`)
+- Bug in :func:`read_excel` would raise an error when pandas could not determine the file type even though the user specified the ``engine`` argument (:issue:`41225`)
+- Bug in :func:`read_clipboard` copying from an excel file shifts values into the wrong column if there are null values in first column (:issue:`41108`)
+- Bug in :meth:`DataFrame.to_hdf` and :meth:`Series.to_hdf` raising a ``TypeError`` when trying to append a string column to an incompatible column (:issue:`41897`)
+
+Period
+^^^^^^
+- Comparisons of :class:`Period` objects or :class:`Index`, :class:`Series`, or :class:`DataFrame` with mismatched ``PeriodDtype`` now behave like other mismatched-type comparisons, returning ``False`` for equals, ``True`` for not-equal, and raising ``TypeError`` for inequality checks (:issue:`39274`)
+
+Plotting
+^^^^^^^^
+- Bug in :func:`plotting.scatter_matrix` raising when 2d ``ax`` argument passed (:issue:`16253`)
+- Prevent warnings when Matplotlib's ``constrained_layout`` is enabled (:issue:`25261`)
+- Bug in :func:`DataFrame.plot` was showing the wrong colors in the legend if the function was called repeatedly and some calls used ``yerr`` while others didn't (:issue:`39522`)
+- Bug in :func:`DataFrame.plot` was showing the wrong colors in the legend if the function was called repeatedly and some calls used ``secondary_y`` and others use ``legend=False`` (:issue:`40044`)
+- Bug in :meth:`DataFrame.plot.box` when ``dark_background`` theme was selected, caps or min/max markers for the plot were not visible (:issue:`40769`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` with :class:`PeriodDtype` columns incorrectly casting results too aggressively (:issue:`38254`)
+- Bug in :meth:`.SeriesGroupBy.value_counts` where unobserved categories in a grouped categorical Series were not tallied (:issue:`38672`)
+- Bug in :meth:`.SeriesGroupBy.value_counts` where an error was raised on an empty Series (:issue:`39172`)
+- Bug in :meth:`.GroupBy.indices` would contain non-existent indices when null values were present in the groupby keys (:issue:`9304`)
+- Fixed bug in :meth:`.DataFrameGroupBy.sum` and :meth:`.SeriesGroupBy.sum` causing a loss of precision by now using Kahan summation (:issue:`38778`)
+- Fixed bug in :meth:`.DataFrameGroupBy.cumsum`, :meth:`.SeriesGroupBy.cumsum`, :meth:`.DataFrameGroupBy.mean`, and :meth:`.SeriesGroupBy.mean` causing loss of precision through using Kahan summation (:issue:`38934`)
+- Bug in :meth:`.Resampler.aggregate` and :meth:`DataFrame.transform` raising a ``TypeError`` instead of ``SpecificationError`` when missing keys had mixed dtypes (:issue:`39025`)
+- Bug in :meth:`.DataFrameGroupBy.idxmin` and :meth:`.DataFrameGroupBy.idxmax` with ``ExtensionDtype`` columns (:issue:`38733`)
+- Bug in :meth:`Series.resample` would raise when the index was a :class:`PeriodIndex` consisting of ``NaT`` (:issue:`39227`)
+- Bug in :meth:`.RollingGroupby.corr` and :meth:`.ExpandingGroupby.corr` where the groupby column would return ``0`` instead of ``np.nan`` when providing ``other`` that was longer than each group (:issue:`39591`)
+- Bug in :meth:`.ExpandingGroupby.corr` and :meth:`.ExpandingGroupby.cov` where ``1`` would be returned instead of ``np.nan`` when providing ``other`` that was longer than each group (:issue:`39591`)
+- Bug in :meth:`.DataFrameGroupBy.mean`, :meth:`.SeriesGroupBy.mean`, :meth:`.DataFrameGroupBy.median`, :meth:`.SeriesGroupBy.median`, and :meth:`DataFrame.pivot_table` not propagating metadata (:issue:`28283`)
+- Bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` not calculating window bounds correctly when window is an offset and dates are in descending order (:issue:`40002`)
+- Bug in :meth:`Series.groupby` and :meth:`DataFrame.groupby` on an empty ``Series`` or ``DataFrame`` would lose index, columns, and/or data types when directly using the methods ``idxmax``, ``idxmin``, ``mad``, ``min``, ``max``, ``sum``, ``prod``, and ``skew`` or using them through ``apply``, ``aggregate``, or ``resample`` (:issue:`26411`)
+- Bug in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` where a :class:`MultiIndex` would be created instead of an :class:`Index` when used on a :class:`.RollingGroupby` object (:issue:`39732`)
+- Bug in :meth:`.DataFrameGroupBy.sample` where an error was raised when ``weights`` was specified and the index was an :class:`Int64Index` (:issue:`39927`)
+- Bug in :meth:`.DataFrameGroupBy.aggregate` and :meth:`.Resampler.aggregate` would sometimes raise a ``SpecificationError`` when passed a dictionary and columns were missing; will now always raise a ``KeyError`` instead (:issue:`40004`)
+- Bug in :meth:`.DataFrameGroupBy.sample` where column selection was not applied before computing the result (:issue:`39928`)
+- Bug in :class:`.ExponentialMovingWindow` when calling ``__getitem__`` would incorrectly raise a ``ValueError`` when providing ``times`` (:issue:`40164`)
+- Bug in :class:`.ExponentialMovingWindow` when calling ``__getitem__`` would not retain ``com``, ``span``, ``alpha`` or ``halflife`` attributes  (:issue:`40164`)
+- :class:`.ExponentialMovingWindow` now raises a ``NotImplementedError`` when specifying ``times`` with ``adjust=False`` due to an incorrect calculation (:issue:`40098`)
+- Bug in :meth:`.ExponentialMovingWindowGroupby.mean` where the ``times`` argument was ignored when ``engine='numba'`` (:issue:`40951`)
+- Bug in :meth:`.ExponentialMovingWindowGroupby.mean` where the wrong times were used the in case of multiple groups (:issue:`40951`)
+- Bug in :class:`.ExponentialMovingWindowGroupby` where the times vector and values became out of sync for non-trivial groups (:issue:`40951`)
+- Bug in :meth:`Series.asfreq` and :meth:`DataFrame.asfreq` dropping rows when the index was not sorted (:issue:`39805`)
+- Bug in aggregation functions for :class:`DataFrame` not respecting ``numeric_only`` argument when ``level`` keyword was given (:issue:`40660`)
+- Bug in :meth:`.SeriesGroupBy.aggregate` where using a user-defined function to aggregate a Series with an object-typed :class:`Index` causes an incorrect :class:`Index` shape (:issue:`40014`)
+- Bug in :class:`.RollingGroupby` where ``as_index=False`` argument in ``groupby`` was ignored (:issue:`39433`)
+- Bug in :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.any`, :meth:`.DataFrameGroupBy.all` and :meth:`.SeriesGroupBy.all` raising a ``ValueError`` when using with nullable type columns holding ``NA`` even with ``skipna=True`` (:issue:`40585`)
+- Bug in :meth:`.DataFrameGroupBy.cummin`, :meth:`.SeriesGroupBy.cummin`, :meth:`.DataFrameGroupBy.cummax` and :meth:`.SeriesGroupBy.cummax` incorrectly rounding integer values near the ``int64`` implementations bounds (:issue:`40767`)
+- Bug in :meth:`.DataFrameGroupBy.rank` and :meth:`.SeriesGroupBy.rank` with nullable dtypes incorrectly raising a ``TypeError`` (:issue:`41010`)
+- Bug in :meth:`.DataFrameGroupBy.cummin`, :meth:`.SeriesGroupBy.cummin`, :meth:`.DataFrameGroupBy.cummax` and :meth:`.SeriesGroupBy.cummax` computing wrong result with nullable data types too large to roundtrip when casting to float (:issue:`37493`)
+- Bug in :meth:`DataFrame.rolling` returning mean zero for all ``NaN`` window with ``min_periods=0`` if calculation is not numerical stable (:issue:`41053`)
+- Bug in :meth:`DataFrame.rolling` returning sum not zero for all ``NaN`` window with ``min_periods=0`` if calculation is not numerical stable (:issue:`41053`)
+- Bug in :meth:`.SeriesGroupBy.agg` failing to retain ordered :class:`CategoricalDtype` on order-preserving aggregations (:issue:`41147`)
+- Bug in :meth:`.DataFrameGroupBy.min`, :meth:`.SeriesGroupBy.min`, :meth:`.DataFrameGroupBy.max` and :meth:`.SeriesGroupBy.max` with multiple object-dtype columns and ``numeric_only=False`` incorrectly raising a ``ValueError`` (:issue:`41111`)
+- Bug in :meth:`.DataFrameGroupBy.rank` with the GroupBy object's ``axis=0`` and the ``rank`` method's keyword ``axis=1`` (:issue:`41320`)
+- Bug in :meth:`DataFrameGroupBy.__getitem__` with non-unique columns incorrectly returning a malformed :class:`SeriesGroupBy` instead of :class:`DataFrameGroupBy` (:issue:`41427`)
+- Bug in :meth:`.DataFrameGroupBy.transform` with non-unique columns incorrectly raising an ``AttributeError`` (:issue:`41427`)
+- Bug in :meth:`.Resampler.apply` with non-unique columns incorrectly dropping duplicated columns (:issue:`41445`)
+- Bug in :meth:`Series.groupby` aggregations incorrectly returning empty :class:`Series` instead of raising ``TypeError`` on aggregations that are invalid for its dtype, e.g. ``.prod`` with ``datetime64[ns]`` dtype (:issue:`41342`)
+- Bug in :class:`DataFrameGroupBy` aggregations incorrectly failing to drop columns with invalid dtypes for that aggregation when there are no valid columns (:issue:`41291`)
+- Bug in :meth:`DataFrame.rolling.__iter__` where ``on`` was not assigned to the index of the resulting objects (:issue:`40373`)
+- Bug in :meth:`.DataFrameGroupBy.transform` and :meth:`.DataFrameGroupBy.agg` with ``engine="numba"`` where ``*args`` were being cached with the user passed function (:issue:`41647`)
+- Bug in :class:`DataFrameGroupBy` methods ``agg``, ``transform``, ``sum``, ``bfill``, ``ffill``, ``pad``, ``pct_change``, ``shift``, ``ohlc`` dropping ``.columns.names`` (:issue:`41497`)
+
+
+Reshaping
+^^^^^^^^^
+- Bug in :func:`merge` raising error when performing an inner join with partial index and ``right_index=True`` when there was no overlap between indices (:issue:`33814`)
+- Bug in :meth:`DataFrame.unstack` with missing levels led to incorrect index names (:issue:`37510`)
+- Bug in :func:`merge_asof` propagating the right Index with ``left_index=True`` and ``right_on`` specification instead of left Index (:issue:`33463`)
+- Bug in :meth:`DataFrame.join` on a DataFrame with a :class:`MultiIndex` returned the wrong result when one of both indexes had only one level (:issue:`36909`)
+- :func:`merge_asof` now raises a ``ValueError`` instead of a cryptic ``TypeError`` in case of non-numerical merge columns (:issue:`29130`)
+- Bug in :meth:`DataFrame.join` not assigning values correctly when the DataFrame had a :class:`MultiIndex` where at least one dimension had dtype ``Categorical`` with non-alphabetically sorted categories (:issue:`38502`)
+- :meth:`Series.value_counts` and :meth:`Series.mode` now return consistent keys in original order (:issue:`12679`, :issue:`11227` and :issue:`39007`)
+- Bug in :meth:`DataFrame.stack` not handling ``NaN`` in :class:`MultiIndex` columns correctly (:issue:`39481`)
+- Bug in :meth:`DataFrame.apply` would give incorrect results when the argument ``func`` was a string, ``axis=1``, and the axis argument was not supported; now raises a ``ValueError`` instead (:issue:`39211`)
+- Bug in :meth:`DataFrame.sort_values` not reshaping the index correctly after sorting on columns when ``ignore_index=True`` (:issue:`39464`)
+- Bug in :meth:`DataFrame.append` returning incorrect dtypes with combinations of ``ExtensionDtype`` dtypes (:issue:`39454`)
+- Bug in :meth:`DataFrame.append` returning incorrect dtypes when used with combinations of ``datetime64`` and ``timedelta64`` dtypes (:issue:`39574`)
+- Bug in :meth:`DataFrame.append` with a :class:`DataFrame` with a :class:`MultiIndex` and appending a :class:`Series` whose :class:`Index` is not a :class:`MultiIndex` (:issue:`41707`)
+- Bug in :meth:`DataFrame.pivot_table` returning a :class:`MultiIndex` for a single value when operating on an empty DataFrame (:issue:`13483`)
+- :class:`Index` can now be passed to the :func:`numpy.all` function (:issue:`40180`)
+- Bug in :meth:`DataFrame.stack` not preserving ``CategoricalDtype`` in a :class:`MultiIndex` (:issue:`36991`)
+- Bug in :func:`to_datetime` raising an error when the input sequence contained unhashable items (:issue:`39756`)
+- Bug in :meth:`Series.explode` preserving the index when ``ignore_index`` was ``True`` and values were scalars (:issue:`40487`)
+- Bug in :func:`to_datetime` raising a ``ValueError`` when :class:`Series` contains ``None`` and ``NaT`` and has more than 50 elements (:issue:`39882`)
+- Bug in :meth:`Series.unstack` and :meth:`DataFrame.unstack` with object-dtype values containing timezone-aware datetime objects incorrectly raising ``TypeError`` (:issue:`41875`)
+- Bug in :meth:`DataFrame.melt` raising ``InvalidIndexError`` when :class:`DataFrame` has duplicate columns used as ``value_vars`` (:issue:`41951`)
+
+Sparse
+^^^^^^
+- Bug in :meth:`DataFrame.sparse.to_coo` raising a ``KeyError`` with columns that are a numeric :class:`Index` without a ``0`` (:issue:`18414`)
+- Bug in :meth:`SparseArray.astype` with ``copy=False`` producing incorrect results when going from integer dtype to floating dtype (:issue:`34456`)
+- Bug in :meth:`SparseArray.max` and :meth:`SparseArray.min` would always return an empty result (:issue:`40921`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :meth:`DataFrame.where` when ``other`` is a Series with an :class:`ExtensionDtype` (:issue:`38729`)
+- Fixed bug where :meth:`Series.idxmax`, :meth:`Series.idxmin`, :meth:`Series.argmax`, and :meth:`Series.argmin` would fail when the underlying data is an :class:`ExtensionArray` (:issue:`32749`, :issue:`33719`, :issue:`36566`)
+- Fixed bug where some properties of subclasses of :class:`PandasExtensionDtype` where improperly cached (:issue:`40329`)
+- Bug in :meth:`DataFrame.mask` where masking a DataFrame with an :class:`ExtensionDtype` raises a ``ValueError`` (:issue:`40941`)
+
+Styler
+^^^^^^
+- Bug in :class:`.Styler` where the ``subset`` argument in methods raised an error for some valid MultiIndex slices (:issue:`33562`)
+- :class:`.Styler` rendered HTML output has seen minor alterations to support w3 good code standards (:issue:`39626`)
+- Bug in :class:`.Styler` where rendered HTML was missing a column class identifier for certain header cells (:issue:`39716`)
+- Bug in :meth:`.Styler.background_gradient` where text-color was not determined correctly (:issue:`39888`)
+- Bug in :meth:`.Styler.set_table_styles` where multiple elements in CSS-selectors of the ``table_styles`` argument were not correctly added (:issue:`34061`)
+- Bug in :class:`.Styler` where copying from Jupyter dropped the top left cell and misaligned headers (:issue:`12147`)
+- Bug in :class:`Styler.where` where ``kwargs`` were not passed to the applicable callable (:issue:`40845`)
+- Bug in :class:`.Styler` causing CSS to duplicate on multiple renders (:issue:`39395`, :issue:`40334`)
+
+Other
+^^^^^
+- ``inspect.getmembers(Series)`` no longer raises an ``AbstractMethodError`` (:issue:`38782`)
+- Bug in :meth:`Series.where` with numeric dtype and ``other=None`` not casting to ``nan`` (:issue:`39761`)
+- Bug in :func:`.assert_series_equal`, :func:`.assert_frame_equal`, :func:`.assert_index_equal` and :func:`.assert_extension_array_equal` incorrectly raising when an attribute has an unrecognized NA type (:issue:`39461`)
+- Bug in :func:`.assert_index_equal` with ``exact=True`` not raising when comparing :class:`CategoricalIndex` instances with ``Int64Index`` and ``RangeIndex`` categories (:issue:`41263`)
+- Bug in :meth:`DataFrame.equals`, :meth:`Series.equals`, and :meth:`Index.equals` with object-dtype containing ``np.datetime64("NaT")`` or ``np.timedelta64("NaT")`` (:issue:`39650`)
+- Bug in :func:`show_versions` where console JSON output was not proper JSON (:issue:`39701`)
+- pandas can now compile on z/OS when using `xlc <https://www.ibm.com/products/xl-cpp-compiler-zos>`_ (:issue:`35826`)
+- Bug in :func:`pandas.util.hash_pandas_object` not recognizing ``hash_key``, ``encoding`` and ``categorize`` when the input object type is a :class:`DataFrame` (:issue:`41404`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_130.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.2.5..v1.3.0
diff --git a/pandas/doc/source/whatsnew/v1.3.1.rst b/pandas/doc/source/whatsnew/v1.3.1.rst
new file mode 100644
index 0000000000000..f3b21554e668c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.1.rst
@@ -0,0 +1,51 @@
+.. _whatsnew_131:
+
+What's new in 1.3.1 (July 25, 2021)
+-----------------------------------
+
+These are the changes in pandas 1.3.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_131.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- pandas could not be built on PyPy (:issue:`42355`)
+- :class:`DataFrame` constructed with an older version of pandas could not be unpickled (:issue:`42345`)
+- Performance regression in constructing a :class:`DataFrame` from a dictionary of dictionaries (:issue:`42248`)
+- Fixed regression in :meth:`DataFrame.agg` dropping values when the DataFrame had an Extension Array dtype, a duplicate index, and ``axis=1`` (:issue:`42380`)
+- Fixed regression in :meth:`DataFrame.astype` changing the order of noncontiguous data (:issue:`42396`)
+- Performance regression in :class:`DataFrame` in reduction operations requiring casting such as :meth:`DataFrame.mean` on integer data (:issue:`38592`)
+- Performance regression in :meth:`DataFrame.to_dict` and :meth:`Series.to_dict` when ``orient`` argument one of "records", "dict", or "split" (:issue:`42352`)
+- Fixed regression in indexing with a ``list`` subclass incorrectly raising ``TypeError`` (:issue:`42433`, :issue:`42461`)
+- Fixed regression in :meth:`DataFrame.isin` and :meth:`Series.isin` raising ``TypeError`` with nullable data containing at least one missing value (:issue:`42405`)
+- Regression in :func:`concat` between objects with bool dtype and integer dtype casting to object instead of to integer (:issue:`42092`)
+- Bug in :class:`Series` constructor not accepting a ``dask.Array`` (:issue:`38645`)
+- Fixed regression for ``SettingWithCopyWarning`` displaying incorrect stacklevel (:issue:`42570`)
+- Fixed regression for :func:`merge_asof` raising ``KeyError`` when one of the ``by`` columns is in the index (:issue:`34488`)
+- Fixed regression in :func:`to_datetime` returning pd.NaT for inputs that produce duplicated values, when ``cache=True`` (:issue:`42259`)
+- Fixed regression in :meth:`SeriesGroupBy.value_counts` that resulted in an ``IndexError`` when called on a Series with one row (:issue:`42618`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_131.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug in :meth:`DataFrame.transpose` dropping values when the DataFrame had an Extension Array dtype and a duplicate index (:issue:`42380`)
+- Fixed bug in :meth:`DataFrame.to_xml` raising ``KeyError`` when called with ``index=False`` and an offset index (:issue:`42458`)
+- Fixed bug in :meth:`.Styler.set_sticky` not handling index names correctly for single index columns case (:issue:`42537`)
+- Fixed bug in :meth:`DataFrame.copy` failing to consolidate blocks in the result (:issue:`42579`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_131.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.0..v1.3.1
diff --git a/pandas/doc/source/whatsnew/v1.3.2.rst b/pandas/doc/source/whatsnew/v1.3.2.rst
new file mode 100644
index 0000000000000..e3c6268547dd2
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.2.rst
@@ -0,0 +1,51 @@
+.. _whatsnew_132:
+
+What's new in 1.3.2 (August 15, 2021)
+-------------------------------------
+
+These are the changes in pandas 1.3.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_132.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Performance regression in :meth:`DataFrame.isin` and :meth:`Series.isin` for nullable data types (:issue:`42714`)
+- Regression in updating values of :class:`Series` using boolean index, created by using :meth:`DataFrame.pop` (:issue:`42530`)
+- Regression in :meth:`DataFrame.from_records` with empty records (:issue:`42456`)
+- Fixed regression in :meth:`DataFrame.shift` where ``TypeError`` occurred when shifting DataFrame created by concatenation of slices and fills with values (:issue:`42719`)
+- Regression in :meth:`DataFrame.agg` when the ``func`` argument returned lists and ``axis=1`` (:issue:`42727`)
+- Regression in :meth:`DataFrame.drop` does nothing if :class:`MultiIndex` has duplicates and indexer is a tuple or list of tuples (:issue:`42771`)
+- Fixed regression where :func:`read_csv` raised a ``ValueError`` when parameters ``names`` and ``prefix`` were both set to ``None`` (:issue:`42387`)
+- Fixed regression in comparisons between :class:`Timestamp` object and ``datetime64`` objects outside the implementation bounds for nanosecond ``datetime64`` (:issue:`42794`)
+- Fixed regression in :meth:`.Styler.highlight_min` and :meth:`.Styler.highlight_max` where ``pandas.NA`` was not successfully ignored (:issue:`42650`)
+- Fixed regression in :func:`concat` where ``copy=False`` was not honored in ``axis=1`` Series concatenation (:issue:`42501`)
+- Regression in :meth:`Series.nlargest` and :meth:`Series.nsmallest` with nullable integer or float dtype (:issue:`42816`)
+- Fixed regression in :meth:`Series.quantile` with :class:`Int64Dtype` (:issue:`42626`)
+- Fixed regression in :meth:`Series.groupby` and :meth:`DataFrame.groupby` where supplying the ``by`` argument with a Series named with a tuple would incorrectly raise (:issue:`42731`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_132.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :func:`read_excel` modifies the dtypes dictionary when reading a file with duplicate columns (:issue:`42462`)
+- 1D slices over extension types turn into N-dimensional slices over ExtensionArrays (:issue:`42430`)
+- Fixed bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` not calculating window bounds correctly for the first row when ``center=True`` and ``window`` is an offset that covers all the rows (:issue:`42753`)
+- :meth:`.Styler.hide_columns` now hides the index name header row as well as column headers (:issue:`42101`)
+- :meth:`.Styler.set_sticky` has amended CSS to control the column/index names and ensure the correct sticky positions (:issue:`42537`)
+- Bug in de-serializing datetime indexes in PYTHONOPTIMIZED mode (:issue:`42866`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_132.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.1..v1.3.2
diff --git a/pandas/doc/source/whatsnew/v1.3.3.rst b/pandas/doc/source/whatsnew/v1.3.3.rst
new file mode 100644
index 0000000000000..5bb7bcd1372c7
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.3.rst
@@ -0,0 +1,57 @@
+.. _whatsnew_133:
+
+What's new in 1.3.3 (September 12, 2021)
+----------------------------------------
+
+These are the changes in pandas 1.3.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_133.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :class:`DataFrame` constructor failing to broadcast for defined :class:`Index` and len one list of :class:`Timestamp` (:issue:`42810`)
+- Fixed regression in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` incorrectly raising in some cases (:issue:`42390`)
+- Fixed regression in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` where ``nan`` values were dropped even with ``dropna=False`` (:issue:`43205`)
+- Fixed regression in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` which were failing with ``pandas.NA`` (:issue:`42849`)
+- Fixed regression in :meth:`merge` where ``on`` columns with ``ExtensionDtype`` or ``bool`` data types were cast to ``object`` in ``right`` and ``outer`` merge (:issue:`40073`)
+- Fixed regression in :meth:`RangeIndex.where` and :meth:`RangeIndex.putmask` raising ``AssertionError`` when result did not represent a :class:`RangeIndex` (:issue:`43240`)
+- Fixed regression in :meth:`read_parquet` where the ``fastparquet`` engine would not work properly with fastparquet 0.7.0 (:issue:`43075`)
+- Fixed regression in :meth:`DataFrame.loc.__setitem__` raising ``ValueError`` when setting array as cell value (:issue:`43422`)
+- Fixed regression in :func:`is_list_like` where objects with ``__iter__`` set to ``None`` would be identified as iterable (:issue:`43373`)
+- Fixed regression in :meth:`DataFrame.__getitem__` raising error for slice of :class:`DatetimeIndex` when index is non monotonic (:issue:`43223`)
+- Fixed regression in :meth:`.Resampler.aggregate` when used after column selection would raise if ``func`` is a list of aggregation functions (:issue:`42905`)
+- Fixed regression in :meth:`DataFrame.corr` where Kendall correlation would produce incorrect results for columns with repeated values (:issue:`43401`)
+- Fixed regression in :meth:`DataFrame.groupby` where aggregation on columns with object types dropped results on those columns (:issue:`42395`, :issue:`43108`)
+- Fixed regression in :meth:`Series.fillna` raising ``TypeError`` when filling ``float`` ``Series`` with list-like fill value having a dtype which couldn't cast lostlessly (like ``float32`` filled with ``float64``) (:issue:`43424`)
+- Fixed regression in :func:`read_csv` raising ``AttributeError`` when the file handle is an ``tempfile.SpooledTemporaryFile`` object (:issue:`43439`)
+- Fixed performance regression in :meth:`core.window.ewm.ExponentialMovingWindow.mean` (:issue:`42333`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_133.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement for :meth:`DataFrame.__setitem__` when the key or value is not a :class:`DataFrame`, or key is not list-like (:issue:`43274`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_133.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug in :meth:`.DataFrameGroupBy.agg` and :meth:`.DataFrameGroupBy.transform` with ``engine="numba"`` where ``index`` data was not being correctly passed into ``func`` (:issue:`43133`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_133.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.2..v1.3.3
diff --git a/pandas/doc/source/whatsnew/v1.3.4.rst b/pandas/doc/source/whatsnew/v1.3.4.rst
new file mode 100644
index 0000000000000..4fc53b3f39dea
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.4.rst
@@ -0,0 +1,57 @@
+.. _whatsnew_134:
+
+What's new in 1.3.4 (October 17, 2021)
+--------------------------------------
+
+These are the changes in pandas 1.3.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_134.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.convert_dtypes` incorrectly converts byte strings to strings (:issue:`43183`)
+- Fixed regression in :meth:`.DataFrameGroupBy.agg` and :meth:`.SeriesGroupBy.agg` were failing silently with mixed data types along ``axis=1`` and :class:`MultiIndex` (:issue:`43209`)
+- Fixed regression in :func:`merge` with integer and ``NaN`` keys failing with ``outer`` merge (:issue:`43550`)
+- Fixed regression in :meth:`DataFrame.corr` raising ``ValueError`` with ``method="spearman"`` on 32-bit platforms (:issue:`43588`)
+- Fixed performance regression in :meth:`MultiIndex.equals` (:issue:`43549`)
+- Fixed performance regression in :meth:`.DataFrameGroupBy.first`, :meth:`.SeriesGroupBy.first`, :meth:`.DataFrameGroupBy.last`, and :meth:`.SeriesGroupBy.last` with :class:`StringDtype` (:issue:`41596`)
+- Fixed regression in :meth:`Series.cat.reorder_categories` failing to update the categories on the ``Series`` (:issue:`43232`)
+- Fixed regression in :meth:`Series.cat.categories` setter failing to update the categories on the ``Series`` (:issue:`43334`)
+- Fixed regression in :func:`read_csv` raising ``UnicodeDecodeError`` exception when ``memory_map=True`` (:issue:`43540`)
+- Fixed regression in :meth:`DataFrame.explode` raising ``AssertionError`` when ``column`` is any scalar which is not a string (:issue:`43314`)
+- Fixed regression in :meth:`Series.aggregate` attempting to pass ``args`` and ``kwargs`` multiple times to the user supplied ``func`` in certain cases (:issue:`43357`)
+- Fixed regression when iterating over a :class:`DataFrame.groupby.rolling` object causing the resulting DataFrames to have an incorrect index if the input groupings were not sorted (:issue:`43386`)
+- Fixed regression in :meth:`DataFrame.groupby.rolling.cov` and :meth:`DataFrame.groupby.rolling.corr` computing incorrect results if the input groupings were not sorted (:issue:`43386`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_134.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug in :meth:`pandas.DataFrame.groupby.rolling` and :class:`pandas.api.indexers.FixedForwardWindowIndexer` leading to segfaults and window endpoints being mixed across groups (:issue:`43267`)
+- Fixed bug in :meth:`.DataFrameGroupBy.mean` and :meth:`.SeriesGroupBy.mean` with datetimelike values including ``NaT`` values returning incorrect results (:issue:`43132`)
+- Fixed bug in :meth:`Series.aggregate` not passing the first ``args`` to the user supplied ``func`` in certain cases (:issue:`43357`)
+- Fixed memory leaks in :meth:`Series.rolling.quantile` and :meth:`Series.rolling.median` (:issue:`43339`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_134.other:
+
+Other
+~~~~~
+- The minimum version of Cython needed to compile pandas is now ``0.29.24`` (:issue:`43729`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_134.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.3..v1.3.4
diff --git a/pandas/doc/source/whatsnew/v1.3.5.rst b/pandas/doc/source/whatsnew/v1.3.5.rst
new file mode 100644
index 0000000000000..067d2214b1a5b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.3.5.rst
@@ -0,0 +1,34 @@
+.. _whatsnew_135:
+
+What's new in 1.3.5 (December 12, 2021)
+---------------------------------------
+
+These are the changes in pandas 1.3.5. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_135.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`Series.equals` when comparing floats with dtype object to None (:issue:`44190`)
+- Fixed regression in :func:`merge_asof` raising error when array was supplied as join key (:issue:`42844`)
+- Fixed regression when resampling :class:`DataFrame` with :class:`DateTimeIndex` with empty groups and ``uint8``, ``uint16`` or ``uint32`` columns incorrectly raising ``RuntimeError`` (:issue:`43329`)
+- Fixed regression in creating a :class:`DataFrame` from a timezone-aware :class:`Timestamp` scalar near a Daylight Savings Time transition (:issue:`42505`)
+- Fixed performance regression in :func:`read_csv` (:issue:`44106`)
+- Fixed regression in :meth:`Series.duplicated` and :meth:`Series.drop_duplicates` when Series has :class:`Categorical` dtype with boolean categories (:issue:`44351`)
+- Fixed regression in :meth:`.DataFrameGroupBy.sum` and :meth:`.SeriesGroupBy.sum` with ``timedelta64[ns]`` dtype containing ``NaT`` failing to treat that value as NA (:issue:`42659`)
+- Fixed regression in :meth:`.RollingGroupby.cov` and :meth:`.RollingGroupby.corr` when ``other`` had the same shape as each group would incorrectly return superfluous groups in the result (:issue:`42915`)
+
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_135.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.4..v1.3.5|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.4.0.rst b/pandas/doc/source/whatsnew/v1.4.0.rst
new file mode 100644
index 0000000000000..7b1aef07e5f00
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.4.0.rst
@@ -0,0 +1,1112 @@
+.. _whatsnew_140:
+
+What's new in 1.4.0 (January 22, 2022)
+--------------------------------------
+
+These are the changes in pandas 1.4.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_140.enhancements.warning_lineno:
+
+Improved warning messages
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, warning messages may have pointed to lines within the pandas
+library. Running the script ``setting_with_copy_warning.py``
+
+.. code-block:: python
+
+    import pandas as pd
+
+    df = pd.DataFrame({'a': [1, 2, 3]})
+    df[:2].loc[:, 'a'] = 5
+
+with pandas 1.3 resulted in::
+
+    .../site-packages/pandas/core/indexing.py:1951: SettingWithCopyWarning:
+    A value is trying to be set on a copy of a slice from a DataFrame.
+
+This made it difficult to determine where the warning was being generated from.
+Now pandas will inspect the call stack, reporting the first line outside of the
+pandas library that gave rise to the warning. The output of the above script is
+now::
+
+    setting_with_copy_warning.py:4: SettingWithCopyWarning:
+    A value is trying to be set on a copy of a slice from a DataFrame.
+
+
+
+
+.. _whatsnew_140.enhancements.ExtensionIndex:
+
+Index can hold arbitrary ExtensionArrays
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Until now, passing a custom :class:`ExtensionArray` to ``pd.Index`` would cast
+the array to ``object`` dtype. Now :class:`Index` can directly hold arbitrary
+ExtensionArrays (:issue:`43930`).
+
+*Previous behavior*:
+
+.. ipython:: python
+
+   arr = pd.array([1, 2, pd.NA])
+   idx = pd.Index(arr)
+
+In the old behavior, ``idx`` would be object-dtype:
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [1]: idx
+   Out[1]: Index([1, 2, <NA>], dtype='object')
+
+With the new behavior, we keep the original dtype:
+
+*New behavior*:
+
+.. ipython:: python
+
+   idx
+
+One exception to this is ``SparseArray``, which will continue to cast to numpy
+dtype until pandas 2.0. At that point it will retain its dtype like other
+ExtensionArrays.
+
+.. _whatsnew_140.enhancements.styler:
+
+Styler
+^^^^^^
+
+:class:`.Styler` has been further developed in 1.4.0. The following general enhancements have been made:
+
+  - Styling and formatting of indexes has been added, with :meth:`.Styler.apply_index`, :meth:`.Styler.applymap_index` and :meth:`.Styler.format_index`. These mirror the signature of the methods already used to style and format data values, and work with both HTML, LaTeX and Excel format (:issue:`41893`, :issue:`43101`, :issue:`41993`, :issue:`41995`)
+  - The new method :meth:`.Styler.hide` deprecates :meth:`.Styler.hide_index` and :meth:`.Styler.hide_columns` (:issue:`43758`)
+  - The keyword arguments ``level`` and ``names`` have been added to :meth:`.Styler.hide` (and implicitly to the deprecated methods :meth:`.Styler.hide_index` and :meth:`.Styler.hide_columns`) for additional control of visibility of MultiIndexes and of Index names (:issue:`25475`, :issue:`43404`, :issue:`43346`)
+  - The :meth:`.Styler.export` and :meth:`.Styler.use` have been updated to address all of the added functionality from v1.2.0 and v1.3.0 (:issue:`40675`)
+  - Global options under the category ``pd.options.styler`` have been extended to configure default ``Styler`` properties which address formatting, encoding, and HTML and LaTeX rendering. Note that formerly ``Styler`` relied on ``display.html.use_mathjax``, which has now been replaced by ``styler.html.mathjax`` (:issue:`41395`)
+  - Validation of certain keyword arguments, e.g. ``caption`` (:issue:`43368`)
+  - Various bug fixes as recorded below
+
+Additionally there are specific enhancements to the HTML specific rendering:
+
+  - :meth:`.Styler.bar` introduces additional arguments to control alignment and display (:issue:`26070`, :issue:`36419`), and it also validates the input arguments ``width`` and ``height`` (:issue:`42511`)
+  - :meth:`.Styler.to_html` introduces keyword arguments ``sparse_index``, ``sparse_columns``, ``bold_headers``, ``caption``, ``max_rows`` and ``max_columns`` (:issue:`41946`, :issue:`43149`, :issue:`42972`)
+  - :meth:`.Styler.to_html` omits CSSStyle rules for hidden table elements as a performance enhancement (:issue:`43619`)
+  - Custom CSS classes can now be directly specified without string replacement (:issue:`43686`)
+  - Ability to render hyperlinks automatically via a new ``hyperlinks`` formatting keyword argument (:issue:`45058`)
+
+There are also some LaTeX specific enhancements:
+
+  - :meth:`.Styler.to_latex` introduces keyword argument ``environment``, which also allows a specific "longtable" entry through a separate jinja2 template (:issue:`41866`)
+  - Naive sparsification is now possible for LaTeX without the necessity of including the multirow package (:issue:`43369`)
+  - *cline* support has been added for :class:`MultiIndex` row sparsification through a keyword argument (:issue:`45138`)
+
+.. _whatsnew_140.enhancements.pyarrow_csv_engine:
+
+Multi-threaded CSV reading with a new CSV Engine based on pyarrow
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`pandas.read_csv` now accepts ``engine="pyarrow"`` (requires at least
+``pyarrow`` 1.0.1) as an argument, allowing for faster csv parsing on multicore
+machines with pyarrow installed. See the :doc:`I/O docs </user_guide/io>` for
+more info. (:issue:`23697`, :issue:`43706`)
+
+.. _whatsnew_140.enhancements.window_rank:
+
+Rank function for rolling and expanding windows
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Added ``rank`` function to :class:`Rolling` and :class:`Expanding`. The new
+function supports the ``method``, ``ascending``, and ``pct`` flags of
+:meth:`DataFrame.rank`. The ``method`` argument supports ``min``, ``max``, and
+``average`` ranking methods.
+Example:
+
+.. ipython:: python
+
+    s = pd.Series([1, 4, 2, 3, 5, 3])
+    s.rolling(3).rank()
+
+    s.rolling(3).rank(method="max")
+
+.. _whatsnew_140.enhancements.groupby_indexing:
+
+Groupby positional indexing
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+It is now possible to specify positional ranges relative to the ends of each
+group.
+
+Negative arguments for :meth:`.DataFrameGroupBy.head`, :meth:`.SeriesGroupBy.head`, :meth:`.DataFrameGroupBy.tail`, and :meth:`.SeriesGroupBy.tail` now work
+correctly and result in ranges relative to the end and start of each group,
+respectively. Previously, negative arguments returned empty frames.
+
+.. ipython:: python
+
+    df = pd.DataFrame([["g", "g0"], ["g", "g1"], ["g", "g2"], ["g", "g3"],
+                       ["h", "h0"], ["h", "h1"]], columns=["A", "B"])
+    df.groupby("A").head(-1)
+
+
+:meth:`.DataFrameGroupBy.nth` and :meth:`.SeriesGroupBy.nth` now accept a slice or list of integers and slices.
+
+.. ipython:: python
+
+    df.groupby("A").nth(slice(1, -1))
+    df.groupby("A").nth([slice(None, 1), slice(-1, None)])
+
+:meth:`.DataFrameGroupBy.nth` and :meth:`.SeriesGroupBy.nth` now accept index notation.
+
+.. ipython:: python
+
+    df.groupby("A").nth[1, -1]
+    df.groupby("A").nth[1:-1]
+    df.groupby("A").nth[:1, -1:]
+
+.. _whatsnew_140.dict_tight:
+
+DataFrame.from_dict and DataFrame.to_dict have new ``'tight'`` option
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A new ``'tight'`` dictionary format that preserves :class:`MultiIndex` entries
+and names is now available with the :meth:`DataFrame.from_dict` and
+:meth:`DataFrame.to_dict` methods and can be used with the standard ``json``
+library to produce a tight representation of :class:`DataFrame` objects
+(:issue:`4889`).
+
+.. ipython:: python
+
+    df = pd.DataFrame.from_records(
+        [[1, 3], [2, 4]],
+        index=pd.MultiIndex.from_tuples([("a", "b"), ("a", "c")],
+                                        names=["n1", "n2"]),
+        columns=pd.MultiIndex.from_tuples([("x", 1), ("y", 2)],
+                                          names=["z1", "z2"]),
+    )
+    df
+    df.to_dict(orient='tight')
+
+.. _whatsnew_140.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- :meth:`concat` will preserve the ``attrs`` when it is the same for all objects and discard the ``attrs`` when they are different (:issue:`41828`)
+- :class:`DataFrameGroupBy` operations with ``as_index=False`` now correctly retain ``ExtensionDtype`` dtypes for columns being grouped on (:issue:`41373`)
+- Add support for assigning values to ``by`` argument in :meth:`DataFrame.plot.hist` and :meth:`DataFrame.plot.box` (:issue:`15079`)
+- :meth:`Series.sample`, :meth:`DataFrame.sample`, :meth:`.DataFrameGroupBy.sample`, and :meth:`.SeriesGroupBy.sample` now accept a ``np.random.Generator`` as input to ``random_state``. A generator will be more performant, especially with ``replace=False`` (:issue:`38100`)
+- :meth:`Series.ewm` and :meth:`DataFrame.ewm` now support a ``method`` argument with a ``'table'`` option that performs the windowing operation over an entire :class:`DataFrame`. See :ref:`Window Overview <window.overview>` for performance and functional benefits (:issue:`42273`)
+- :meth:`.DataFrameGroupBy.cummin`, :meth:`.SeriesGroupBy.cummin`, :meth:`.DataFrameGroupBy.cummax`, and :meth:`.SeriesGroupBy.cummax` now support the argument ``skipna`` (:issue:`34047`)
+- :meth:`read_table` now supports the argument ``storage_options`` (:issue:`39167`)
+- :meth:`DataFrame.to_stata` and :meth:`StataWriter` now accept the keyword only argument ``value_labels`` to save labels for non-categorical columns (:issue:`38454`)
+- Methods that relied on hashmap based algos such as :meth:`DataFrameGroupBy.value_counts`, :meth:`DataFrameGroupBy.count` and :func:`factorize` ignored imaginary component for complex numbers (:issue:`17927`)
+- Add :meth:`Series.str.removeprefix` and :meth:`Series.str.removesuffix` introduced in Python 3.9 to remove pre-/suffixes from string-type :class:`Series` (:issue:`36944`)
+- Attempting to write into a file in missing parent directory with :meth:`DataFrame.to_csv`, :meth:`DataFrame.to_html`, :meth:`DataFrame.to_excel`, :meth:`DataFrame.to_feather`, :meth:`DataFrame.to_parquet`, :meth:`DataFrame.to_stata`, :meth:`DataFrame.to_json`, :meth:`DataFrame.to_pickle`, and :meth:`DataFrame.to_xml` now explicitly mentions missing parent directory, the same is true for :class:`Series` counterparts (:issue:`24306`)
+- Indexing with ``.loc`` and ``.iloc`` now supports ``Ellipsis`` (:issue:`37750`)
+- :meth:`IntegerArray.all` , :meth:`IntegerArray.any`, :meth:`FloatingArray.any`, and :meth:`FloatingArray.all` use Kleene logic (:issue:`41967`)
+- Added support for nullable boolean and integer types in :meth:`DataFrame.to_stata`, :class:`~pandas.io.stata.StataWriter`, :class:`~pandas.io.stata.StataWriter117`, and :class:`~pandas.io.stata.StataWriterUTF8` (:issue:`40855`)
+- :meth:`DataFrame.__pos__` and :meth:`DataFrame.__neg__` now retain ``ExtensionDtype`` dtypes (:issue:`43883`)
+- The error raised when an optional dependency can't be imported now includes the original exception, for easier investigation (:issue:`43882`)
+- Added :meth:`.ExponentialMovingWindow.sum` (:issue:`13297`)
+- :meth:`Series.str.split` now supports a ``regex`` argument that explicitly specifies whether the pattern is a regular expression. Default is ``None`` (:issue:`43563`, :issue:`32835`, :issue:`25549`)
+- :meth:`DataFrame.dropna` now accepts a single label as ``subset`` along with array-like (:issue:`41021`)
+- Added :meth:`DataFrameGroupBy.value_counts` (:issue:`43564`)
+- :func:`read_csv` now accepts a ``callable`` function in ``on_bad_lines`` when ``engine="python"`` for custom handling of bad lines (:issue:`5686`)
+- :class:`ExcelWriter` argument ``if_sheet_exists="overlay"`` option added (:issue:`40231`)
+- :meth:`read_excel` now accepts a ``decimal`` argument that allow the user to specify the decimal point when parsing string columns to numeric (:issue:`14403`)
+- :meth:`.DataFrameGroupBy.mean`, :meth:`.SeriesGroupBy.mean`, :meth:`.DataFrameGroupBy.std`, :meth:`.SeriesGroupBy.std`, :meth:`.DataFrameGroupBy.var`, :meth:`.SeriesGroupBy.var`, :meth:`.DataFrameGroupBy.sum`, and :meth:`.SeriesGroupBy.sum` now support `Numba <http://numba.pydata.org/>`_ execution with the ``engine`` keyword (:issue:`43731`, :issue:`44862`, :issue:`44939`)
+- :meth:`Timestamp.isoformat` now handles the ``timespec`` argument from the base ``datetime`` class (:issue:`26131`)
+- :meth:`NaT.to_numpy` ``dtype`` argument is now respected, so ``np.timedelta64`` can be returned (:issue:`44460`)
+- New option ``display.max_dir_items`` customizes the number of columns added to :meth:`Dataframe.__dir__` and suggested for tab completion (:issue:`37996`)
+- Added "Juneteenth National Independence Day" to ``USFederalHolidayCalendar`` (:issue:`44574`)
+- :meth:`.Rolling.var`, :meth:`.Expanding.var`, :meth:`.Rolling.std`, and :meth:`.Expanding.std` now support `Numba <http://numba.pydata.org/>`_ execution with the ``engine`` keyword (:issue:`44461`)
+- :meth:`Series.info` has been added, for compatibility with :meth:`DataFrame.info` (:issue:`5167`)
+- Implemented :meth:`IntervalArray.min` and :meth:`IntervalArray.max`, as a result of which ``min`` and ``max`` now work for :class:`IntervalIndex`, :class:`Series` and :class:`DataFrame` with ``IntervalDtype`` (:issue:`44746`)
+- :meth:`UInt64Index.map` now retains ``dtype`` where possible (:issue:`44609`)
+- :meth:`read_json` can now parse unsigned long long integers (:issue:`26068`)
+- :meth:`DataFrame.take` now raises a ``TypeError`` when passed a scalar for the indexer (:issue:`42875`)
+- :meth:`is_list_like` now identifies duck-arrays as list-like unless ``.ndim == 0`` (:issue:`35131`)
+- :class:`ExtensionDtype` and :class:`ExtensionArray` are now (de)serialized when exporting a :class:`DataFrame` with :meth:`DataFrame.to_json` using ``orient='table'`` (:issue:`20612`, :issue:`44705`)
+- Add support for `Zstandard <http://facebook.github.io/zstd/>`_ compression to :meth:`DataFrame.to_pickle`/:meth:`read_pickle` and friends (:issue:`43925`)
+- :meth:`DataFrame.to_sql` now returns an ``int`` of the number of written rows (:issue:`23998`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_140.notable_bug_fixes.inconsistent_date_string_parsing:
+
+Inconsistent date string parsing
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``dayfirst`` option of :func:`to_datetime` isn't strict, and this can lead
+to surprising behavior:
+
+.. ipython:: python
+    :okwarning:
+
+    pd.to_datetime(["31-12-2021"], dayfirst=False)
+
+Now, a warning will be raised if a date string cannot be parsed accordance to
+the given ``dayfirst`` value when the value is a delimited date string (e.g.
+``31-12-2012``).
+
+.. _whatsnew_140.notable_bug_fixes.concat_with_empty_or_all_na:
+
+Ignoring dtypes in concat with empty or all-NA columns
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. note::
+    This behaviour change has been reverted in pandas 1.4.3.
+
+When using :func:`concat` to concatenate two or more :class:`DataFrame` objects,
+if one of the DataFrames was empty or had all-NA values, its dtype was
+*sometimes* ignored when finding the concatenated dtype.  These are now
+consistently *not* ignored (:issue:`43507`).
+
+.. code-block:: ipython
+
+    In [3]: df1 = pd.DataFrame({"bar": [pd.Timestamp("2013-01-01")]}, index=range(1))
+    In [4]: df2 = pd.DataFrame({"bar": np.nan}, index=range(1, 2))
+    In [5]: res = pd.concat([df1, df2])
+
+Previously, the float-dtype in ``df2`` would be ignored so the result dtype
+would be ``datetime64[ns]``. As a result, the ``np.nan`` would be cast to
+``NaT``.
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [6]: res
+    Out[6]:
+             bar
+    0 2013-01-01
+    1        NaT
+
+Now the float-dtype is respected. Since the common dtype for these DataFrames is
+object, the ``np.nan`` is retained.
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [6]: res
+    Out[6]:
+                       bar
+    0  2013-01-01 00:00:00
+    1                  NaN
+
+
+
+.. _whatsnew_140.notable_bug_fixes.value_counts_and_mode_do_not_coerce_to_nan:
+
+Null-values are no longer coerced to NaN-value in value_counts and mode
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`Series.value_counts` and :meth:`Series.mode` no longer coerce ``None``,
+``NaT`` and other null-values to a NaN-value for ``np.object_``-dtype. This
+behavior is now consistent with ``unique``, ``isin`` and others
+(:issue:`42688`).
+
+.. ipython:: python
+
+    s = pd.Series([True, None, pd.NaT, None, pd.NaT, None])
+    res = s.value_counts(dropna=False)
+
+Previously, all null-values were replaced by a NaN-value.
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: res
+    Out[3]:
+    NaN     5
+    True    1
+    dtype: int64
+
+Now null-values are no longer mangled.
+
+*New behavior*:
+
+.. ipython:: python
+
+    res
+
+.. _whatsnew_140.notable_bug_fixes.read_csv_mangle_dup_cols:
+
+mangle_dupe_cols in read_csv no longer renames unique columns conflicting with target names
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`read_csv` no longer renames unique column labels which conflict with the target
+names of duplicated columns. Already existing columns are skipped, i.e. the next
+available index is used for the target column name (:issue:`14704`).
+
+.. ipython:: python
+
+    import io
+
+    data = "a,a,a.1\n1,2,3"
+    res = pd.read_csv(io.StringIO(data))
+
+Previously, the second column was called ``a.1``, while the third column was
+also renamed to ``a.1.1``.
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: res
+    Out[3]:
+        a  a.1  a.1.1
+    0   1    2      3
+
+Now the renaming checks if ``a.1`` already exists when changing the name of the
+second column and jumps this index. The second column is instead renamed to
+``a.2``.
+
+*New behavior*:
+
+.. ipython:: python
+
+    res
+
+.. _whatsnew_140.notable_bug_fixes.unstack_pivot_int32_limit:
+
+unstack and pivot_table no longer raises ValueError for result that would exceed int32 limit
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously :meth:`DataFrame.pivot_table` and :meth:`DataFrame.unstack` would
+raise a ``ValueError`` if the operation could produce a result with more than
+``2**31 - 1`` elements. This operation now raises a
+:class:`errors.PerformanceWarning` instead (:issue:`26314`).
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df = DataFrame({"ind1": np.arange(2 ** 16), "ind2": np.arange(2 ** 16), "count": 0})
+    In [4]: df.pivot_table(index="ind1", columns="ind2", values="count", aggfunc="count")
+    ValueError: Unstacked DataFrame is too big, causing int32 overflow
+
+*New behavior*:
+
+.. code-block:: python
+
+    In [4]: df.pivot_table(index="ind1", columns="ind2", values="count", aggfunc="count")
+    PerformanceWarning: The following operation may generate 4294967296 cells in the resulting pandas object.
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.notable_bug_fixes.groupby_apply_mutation:
+
+groupby.apply consistent transform detection
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` are designed to be flexible, allowing users to perform
+aggregations, transformations, filters, and use it with user-defined functions
+that might not fall into any of these categories. As part of this, apply will
+attempt to detect when an operation is a transform, and in such a case, the
+result will have the same index as the input. In order to determine if the
+operation is a transform, pandas compares the input's index to the result's and
+determines if it has been mutated. Previously in pandas 1.3, different code
+paths used different definitions of "mutated": some would use Python's ``is``
+whereas others would test only up to equality.
+
+This inconsistency has been removed, pandas now tests up to equality.
+
+.. ipython:: python
+
+    def func(x):
+        return x.copy()
+
+    df = pd.DataFrame({'a': [1, 2], 'b': [3, 4], 'c': [5, 6]})
+    df
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.groupby(['a']).apply(func)
+    Out[3]:
+         a  b  c
+    a
+    1 0  1  3  5
+    2 1  2  4  6
+
+    In [4]: df.set_index(['a', 'b']).groupby(['a']).apply(func)
+    Out[4]:
+         c
+    a b
+    1 3  5
+    2 4  6
+
+In the examples above, the first uses a code path where pandas uses ``is`` and
+determines that ``func`` is not a transform whereas the second tests up to
+equality and determines that ``func`` is a transform. In the first case, the
+result's index is not the same as the input's.
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [5]: df.groupby(['a']).apply(func)
+    Out[5]:
+       a  b  c
+    0  1  3  5
+    1  2  4  6
+
+    In [6]: df.set_index(['a', 'b']).groupby(['a']).apply(func)
+    Out[6]:
+         c
+    a b
+    1 3  5
+    2 4  6
+
+Now in both cases it is determined that ``func`` is a transform. In each case,
+the result has the same index as the input.
+
+.. _whatsnew_140.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_140.api_breaking.python:
+
+Increased minimum version for Python
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 1.4.0 supports Python 3.8 and higher.
+
+.. _whatsnew_140.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.18.5          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| pytz            | 2020.1          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| python-dateutil | 2.8.1           |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.3.1           |          |    X    |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.7.1           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pytest (dev)    | 6.0             |          |         |
++-----------------+-----------------+----------+---------+
+| mypy (dev)      | 0.930           |          |    X    |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries
+<https://pandas.pydata.org/docs/getting_started/install.html>`_ the general
+recommendation is to use the latest version. The following table lists the
+lowest version per library that is currently being tested throughout the
+development of pandas. Optional libraries below the lowest tested version may
+still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  | 4.8.2           |    X    |
++-----------------+-----------------+---------+
+| fastparquet     | 0.4.0           |         |
++-----------------+-----------------+---------+
+| fsspec          | 0.7.4           |         |
++-----------------+-----------------+---------+
+| gcsfs           | 0.6.0           |         |
++-----------------+-----------------+---------+
+| lxml            | 4.5.0           |    X    |
++-----------------+-----------------+---------+
+| matplotlib      | 3.3.2           |    X    |
++-----------------+-----------------+---------+
+| numba           | 0.50.1          |    X    |
++-----------------+-----------------+---------+
+| openpyxl        | 3.0.3           |    X    |
++-----------------+-----------------+---------+
+| pandas-gbq      | 0.14.0          |    X    |
++-----------------+-----------------+---------+
+| pyarrow         | 1.0.1           |    X    |
++-----------------+-----------------+---------+
+| pymysql         | 0.10.1          |    X    |
++-----------------+-----------------+---------+
+| pytables        | 3.6.1           |    X    |
++-----------------+-----------------+---------+
+| s3fs            | 0.4.0           |         |
++-----------------+-----------------+---------+
+| scipy           | 1.4.1           |    X    |
++-----------------+-----------------+---------+
+| sqlalchemy      | 1.4.0           |    X    |
++-----------------+-----------------+---------+
+| tabulate        | 0.8.7           |         |
++-----------------+-----------------+---------+
+| xarray          | 0.15.1          |    X    |
++-----------------+-----------------+---------+
+| xlrd            | 2.0.1           |    X    |
++-----------------+-----------------+---------+
+| xlsxwriter      | 1.2.2           |    X    |
++-----------------+-----------------+---------+
+| xlwt            | 1.3.0           |         |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_140.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- :meth:`Index.get_indexer_for` no longer accepts keyword arguments (other than ``target``); in the past these would be silently ignored if the index was not unique (:issue:`42310`)
+- Change in the position of the ``min_rows`` argument in :meth:`DataFrame.to_string` due to change in the docstring (:issue:`44304`)
+- Reduction operations for :class:`DataFrame` or :class:`Series` now raising a ``ValueError`` when ``None`` is passed for ``skipna`` (:issue:`44178`)
+- :func:`read_csv` and :func:`read_html` no longer raising an error when one of the header rows consists only of ``Unnamed:`` columns (:issue:`13054`)
+- Changed the ``name`` attribute of several holidays in
+  ``USFederalHolidayCalendar`` to match `official federal holiday
+  names <https://www.opm.gov/policy-data-oversight/pay-leave/federal-holidays/>`_
+  specifically:
+
+   - "New Year's Day" gains the possessive apostrophe
+   - "Presidents Day" becomes "Washington's Birthday"
+   - "Martin Luther King Jr. Day" is now "Birthday of Martin Luther King, Jr."
+   - "July 4th" is now "Independence Day"
+   - "Thanksgiving" is now "Thanksgiving Day"
+   - "Christmas" is now "Christmas Day"
+   - Added "Juneteenth National Independence Day"
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+.. _whatsnew_140.deprecations.int64_uint64_float64index:
+
+Deprecated Int64Index, UInt64Index & Float64Index
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:class:`Int64Index`, :class:`UInt64Index` and :class:`Float64Index` have been
+deprecated in favor of the base :class:`Index` class and will be removed in
+pandas 2.0 (:issue:`43028`).
+
+For constructing a numeric index, you can use the base :class:`Index` class
+instead specifying the data type (which will also work on older pandas
+releases):
+
+.. code-block:: python
+
+    # replace
+    pd.Int64Index([1, 2, 3])
+    # with
+    pd.Index([1, 2, 3], dtype="int64")
+
+For checking the data type of an index object, you can replace ``isinstance``
+checks with checking the ``dtype``:
+
+.. code-block:: python
+
+    # replace
+    isinstance(idx, pd.Int64Index)
+    # with
+    idx.dtype == "int64"
+
+Currently, in order to maintain backward compatibility, calls to :class:`Index`
+will continue to return :class:`Int64Index`, :class:`UInt64Index` and
+:class:`Float64Index` when given numeric data, but in the future, an
+:class:`Index` will be returned.
+
+*Current behavior*:
+
+.. code-block:: ipython
+
+    In [1]: pd.Index([1, 2, 3], dtype="int32")
+    Out [1]: Int64Index([1, 2, 3], dtype='int64')
+    In [1]: pd.Index([1, 2, 3], dtype="uint64")
+    Out [1]: UInt64Index([1, 2, 3], dtype='uint64')
+
+*Future behavior*:
+
+.. code-block:: ipython
+
+    In [3]: pd.Index([1, 2, 3], dtype="int32")
+    Out [3]: Index([1, 2, 3], dtype='int32')
+    In [4]: pd.Index([1, 2, 3], dtype="uint64")
+    Out [4]: Index([1, 2, 3], dtype='uint64')
+
+
+.. _whatsnew_140.deprecations.frame_series_append:
+
+Deprecated DataFrame.append and Series.append
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.append` and :meth:`Series.append` have been deprecated and will
+be removed in a future version. Use :func:`pandas.concat` instead (:issue:`35407`).
+
+*Deprecated syntax*
+
+.. code-block:: ipython
+
+    In [1]: pd.Series([1, 2]).append(pd.Series([3, 4])
+    Out [1]:
+    <stdin>:1: FutureWarning: The series.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+    0    1
+    1    2
+    0    3
+    1    4
+    dtype: int64
+
+    In [2]: df1 = pd.DataFrame([[1, 2], [3, 4]], columns=list('AB'))
+    In [3]: df2 = pd.DataFrame([[5, 6], [7, 8]], columns=list('AB'))
+    In [4]: df1.append(df2)
+    Out [4]:
+    <stdin>:1: FutureWarning: The series.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+       A  B
+    0  1  2
+    1  3  4
+    0  5  6
+    1  7  8
+
+*Recommended syntax*
+
+.. ipython:: python
+
+    pd.concat([pd.Series([1, 2]), pd.Series([3, 4])])
+
+    df1 = pd.DataFrame([[1, 2], [3, 4]], columns=list('AB'))
+    df2 = pd.DataFrame([[5, 6], [7, 8]], columns=list('AB'))
+    pd.concat([df1, df2])
+
+
+.. _whatsnew_140.deprecations.other:
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+- Deprecated :meth:`Index.is_type_compatible` (:issue:`42113`)
+- Deprecated ``method`` argument in :meth:`Index.get_loc`, use ``index.get_indexer([label], method=...)`` instead (:issue:`42269`)
+- Deprecated treating integer keys in :meth:`Series.__setitem__` as positional when the index is a :class:`Float64Index` not containing the key, a :class:`IntervalIndex` with no entries containing the key, or a :class:`MultiIndex` with leading :class:`Float64Index` level not containing the key (:issue:`33469`)
+- Deprecated treating ``numpy.datetime64`` objects as UTC times when passed to the :class:`Timestamp` constructor along with a timezone. In a future version, these will be treated as wall-times. To retain the old behavior, use ``Timestamp(dt64).tz_localize("UTC").tz_convert(tz)`` (:issue:`24559`)
+- Deprecated ignoring missing labels when indexing with a sequence of labels on a level of a :class:`MultiIndex` (:issue:`42351`)
+- Creating an empty :class:`Series` without a ``dtype`` will now raise a more visible ``FutureWarning`` instead of a ``DeprecationWarning`` (:issue:`30017`)
+- Deprecated the ``kind`` argument in :meth:`Index.get_slice_bound`, :meth:`Index.slice_indexer`, and :meth:`Index.slice_locs`; in a future version passing ``kind`` will raise (:issue:`42857`)
+- Deprecated dropping of nuisance columns in :class:`Rolling`, :class:`Expanding`, and :class:`EWM` aggregations (:issue:`42738`)
+- Deprecated :meth:`Index.reindex` with a non-unique :class:`Index` (:issue:`42568`)
+- Deprecated :meth:`.Styler.render` in favor of :meth:`.Styler.to_html` (:issue:`42140`)
+- Deprecated :meth:`.Styler.hide_index` and :meth:`.Styler.hide_columns` in favor of :meth:`.Styler.hide` (:issue:`43758`)
+- Deprecated passing in a string column label into ``times`` in :meth:`DataFrame.ewm` (:issue:`43265`)
+- Deprecated the ``include_start`` and ``include_end`` arguments in :meth:`DataFrame.between_time`; in a future version passing ``include_start`` or ``include_end`` will raise (:issue:`40245`)
+- Deprecated the ``squeeze`` argument to :meth:`read_csv`, :meth:`read_table`, and :meth:`read_excel`. Users should squeeze the :class:`DataFrame` afterwards with ``.squeeze("columns")`` instead (:issue:`43242`)
+- Deprecated the ``index`` argument to :class:`SparseArray` construction (:issue:`23089`)
+- Deprecated the ``closed`` argument in :meth:`date_range` and :meth:`bdate_range` in favor of ``inclusive`` argument; In a future version passing ``closed`` will raise (:issue:`40245`)
+- Deprecated :meth:`.Rolling.validate`, :meth:`.Expanding.validate`, and :meth:`.ExponentialMovingWindow.validate` (:issue:`43665`)
+- Deprecated silent dropping of columns that raised a ``TypeError`` in :class:`Series.transform` and :class:`DataFrame.transform` when used with a dictionary (:issue:`43740`)
+- Deprecated silent dropping of columns that raised a ``TypeError``, ``DataError``, and some cases of ``ValueError`` in :meth:`Series.aggregate`, :meth:`DataFrame.aggregate`, :meth:`Series.groupby.aggregate`, and :meth:`DataFrame.groupby.aggregate` when used with a list (:issue:`43740`)
+- Deprecated casting behavior when setting timezone-aware value(s) into a timezone-aware :class:`Series` or :class:`DataFrame` column when the timezones do not match. Previously this cast to object dtype. In a future version, the values being inserted will be converted to the series or column's existing timezone (:issue:`37605`)
+- Deprecated casting behavior when passing an item with mismatched-timezone to :meth:`DatetimeIndex.insert`, :meth:`DatetimeIndex.putmask`, :meth:`DatetimeIndex.where` :meth:`DatetimeIndex.fillna`, :meth:`Series.mask`, :meth:`Series.where`, :meth:`Series.fillna`, :meth:`Series.shift`, :meth:`Series.replace`, :meth:`Series.reindex` (and :class:`DataFrame` column analogues). In the past this has cast to object ``dtype``. In a future version, these will cast the passed item to the index or series's timezone (:issue:`37605`, :issue:`44940`)
+- Deprecated the ``prefix`` keyword argument in :func:`read_csv` and :func:`read_table`, in a future version the argument will be removed (:issue:`43396`)
+- Deprecated passing non boolean argument to ``sort`` in :func:`concat` (:issue:`41518`)
+- Deprecated passing arguments as positional for :func:`read_fwf` other than ``filepath_or_buffer`` (:issue:`41485`)
+- Deprecated passing arguments as positional for :func:`read_xml` other than ``path_or_buffer`` (:issue:`45133`)
+- Deprecated passing ``skipna=None`` for :meth:`DataFrame.mad` and :meth:`Series.mad`, pass ``skipna=True`` instead (:issue:`44580`)
+- Deprecated the behavior of :func:`to_datetime` with the string "now" with ``utc=False``; in a future version this will match ``Timestamp("now")``, which in turn matches :meth:`Timestamp.now` returning the local time (:issue:`18705`)
+- Deprecated :meth:`DateOffset.apply`, use ``offset + other`` instead (:issue:`44522`)
+- Deprecated parameter ``names`` in :meth:`Index.copy` (:issue:`44916`)
+- A deprecation warning is now shown for :meth:`DataFrame.to_latex` indicating the arguments signature may change and emulate more the arguments to :meth:`.Styler.to_latex` in future versions (:issue:`44411`)
+- Deprecated behavior of :func:`concat` between objects with bool-dtype and numeric-dtypes; in a future version these will cast to object dtype instead of coercing bools to numeric values (:issue:`39817`)
+- Deprecated :meth:`Categorical.replace`, use :meth:`Series.replace` instead (:issue:`44929`)
+- Deprecated passing ``set`` or ``dict`` as indexer for :meth:`DataFrame.loc.__setitem__`, :meth:`DataFrame.loc.__getitem__`, :meth:`Series.loc.__setitem__`, :meth:`Series.loc.__getitem__`, :meth:`DataFrame.__getitem__`, :meth:`Series.__getitem__` and :meth:`Series.__setitem__` (:issue:`42825`)
+- Deprecated :meth:`Index.__getitem__` with a bool key; use ``index.values[key]`` to get the old behavior (:issue:`44051`)
+- Deprecated downcasting column-by-column in :meth:`DataFrame.where` with integer-dtypes (:issue:`44597`)
+- Deprecated :meth:`DatetimeIndex.union_many`, use :meth:`DatetimeIndex.union` instead (:issue:`44091`)
+- Deprecated :meth:`.Groupby.pad` in favor of :meth:`.Groupby.ffill` (:issue:`33396`)
+- Deprecated :meth:`.Groupby.backfill` in favor of :meth:`.Groupby.bfill` (:issue:`33396`)
+- Deprecated :meth:`.Resample.pad` in favor of :meth:`.Resample.ffill` (:issue:`33396`)
+- Deprecated :meth:`.Resample.backfill` in favor of :meth:`.Resample.bfill` (:issue:`33396`)
+- Deprecated ``numeric_only=None`` in :meth:`DataFrame.rank`; in a future version ``numeric_only`` must be either ``True`` or ``False`` (the default) (:issue:`45036`)
+- Deprecated the behavior of :meth:`Timestamp.utcfromtimestamp`, in the future it will return a timezone-aware UTC :class:`Timestamp` (:issue:`22451`)
+- Deprecated :meth:`NaT.freq` (:issue:`45071`)
+- Deprecated behavior of :class:`Series` and :class:`DataFrame` construction when passed float-dtype data containing ``NaN`` and an integer dtype ignoring the dtype argument; in a future version this will raise (:issue:`40110`)
+- Deprecated the behaviour of :meth:`Series.to_frame` and :meth:`Index.to_frame` to ignore the ``name`` argument when ``name=None``. Currently, this means to preserve the existing name, but in the future explicitly passing ``name=None`` will set ``None`` as the name of the column in the resulting DataFrame (:issue:`44212`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :meth:`.DataFrameGroupBy.sample` and :meth:`.SeriesGroupBy.sample`, especially when ``weights`` argument provided (:issue:`34483`)
+- Performance improvement when converting non-string arrays to string arrays (:issue:`34483`)
+- Performance improvement in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` for user-defined functions (:issue:`41598`)
+- Performance improvement in constructing :class:`DataFrame` objects (:issue:`42631`, :issue:`43142`, :issue:`43147`, :issue:`43307`, :issue:`43144`, :issue:`44826`)
+- Performance improvement in :meth:`.DataFrameGroupBy.shift` and :meth:`.SeriesGroupBy.shift` when ``fill_value`` argument is provided (:issue:`26615`)
+- Performance improvement in :meth:`DataFrame.corr` for ``method=pearson`` on data without missing values (:issue:`40956`)
+- Performance improvement in some :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` operations (:issue:`42992`, :issue:`43578`)
+- Performance improvement in :func:`read_stata` (:issue:`43059`, :issue:`43227`)
+- Performance improvement in :func:`read_sas` (:issue:`43333`)
+- Performance improvement in :meth:`to_datetime` with ``uint`` dtypes (:issue:`42606`)
+- Performance improvement in :meth:`to_datetime` with ``infer_datetime_format`` set to ``True`` (:issue:`43901`)
+- Performance improvement in :meth:`Series.sparse.to_coo` (:issue:`42880`)
+- Performance improvement in indexing with a :class:`UInt64Index` (:issue:`43862`)
+- Performance improvement in indexing with a :class:`Float64Index` (:issue:`43705`)
+- Performance improvement in indexing with a non-unique :class:`Index` (:issue:`43792`)
+- Performance improvement in indexing with a listlike indexer on a :class:`MultiIndex` (:issue:`43370`)
+- Performance improvement in indexing with a :class:`MultiIndex` indexer on another :class:`MultiIndex` (:issue:`43370`)
+- Performance improvement in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` (:issue:`43469`, :issue:`43725`)
+- Performance improvement in :meth:`.DataFrameGroupBy.count` and :meth:`.SeriesGroupBy.count` (:issue:`43730`, :issue:`43694`)
+- Performance improvement in :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.any`, :meth:`.DataFrameGroupBy.all`, and :meth:`.SeriesGroupBy.all` (:issue:`43675`, :issue:`42841`)
+- Performance improvement in :meth:`.DataFrameGroupBy.std` and :meth:`.SeriesGroupBy.std` (:issue:`43115`, :issue:`43576`)
+- Performance improvement in :meth:`.DataFrameGroupBy.cumsum` and :meth:`.SeriesGroupBy.cumsum` (:issue:`43309`)
+- :meth:`SparseArray.min` and :meth:`SparseArray.max` no longer require converting to a dense array (:issue:`43526`)
+- Indexing into a :class:`SparseArray` with a ``slice`` with ``step=1`` no longer requires converting to a dense array (:issue:`43777`)
+- Performance improvement in :meth:`SparseArray.take` with ``allow_fill=False`` (:issue:`43654`)
+- Performance improvement in :meth:`.Rolling.mean`, :meth:`.Expanding.mean`, :meth:`.Rolling.sum`, :meth:`.Expanding.sum`, :meth:`.Rolling.max`, :meth:`.Expanding.max`, :meth:`.Rolling.min` and :meth:`.Expanding.min` with ``engine="numba"`` (:issue:`43612`, :issue:`44176`, :issue:`45170`)
+- Improved performance of :meth:`pandas.read_csv` with ``memory_map=True`` when file encoding is UTF-8 (:issue:`43787`)
+- Performance improvement in :meth:`RangeIndex.sort_values` overriding :meth:`Index.sort_values` (:issue:`43666`)
+- Performance improvement in :meth:`RangeIndex.insert` (:issue:`43988`)
+- Performance improvement in :meth:`Index.insert` (:issue:`43953`)
+- Performance improvement in :meth:`DatetimeIndex.tolist` (:issue:`43823`)
+- Performance improvement in :meth:`DatetimeIndex.union` (:issue:`42353`)
+- Performance improvement in :meth:`Series.nsmallest` (:issue:`43696`)
+- Performance improvement in :meth:`DataFrame.insert` (:issue:`42998`)
+- Performance improvement in :meth:`DataFrame.dropna` (:issue:`43683`)
+- Performance improvement in :meth:`DataFrame.fillna` (:issue:`43316`)
+- Performance improvement in :meth:`DataFrame.values` (:issue:`43160`)
+- Performance improvement in :meth:`DataFrame.select_dtypes` (:issue:`42611`)
+- Performance improvement in :class:`DataFrame` reductions (:issue:`43185`, :issue:`43243`, :issue:`43311`, :issue:`43609`)
+- Performance improvement in :meth:`Series.unstack` and :meth:`DataFrame.unstack` (:issue:`43335`, :issue:`43352`, :issue:`42704`, :issue:`43025`)
+- Performance improvement in :meth:`Series.to_frame` (:issue:`43558`)
+- Performance improvement in :meth:`Series.mad` (:issue:`43010`)
+- Performance improvement in :func:`merge` (:issue:`43332`)
+- Performance improvement in :func:`to_csv` when index column is a datetime and is formatted (:issue:`39413`)
+- Performance improvement in :func:`to_csv` when :class:`MultiIndex` contains a lot of unused levels (:issue:`37484`)
+- Performance improvement in :func:`read_csv` when ``index_col`` was set with a numeric column (:issue:`44158`)
+- Performance improvement in :func:`concat` (:issue:`43354`)
+- Performance improvement in :meth:`SparseArray.__getitem__` (:issue:`23122`)
+- Performance improvement in constructing a :class:`DataFrame` from array-like objects like a ``Pytorch`` tensor (:issue:`44616`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in setting dtype-incompatible values into a :class:`Categorical` (or ``Series`` or ``DataFrame`` backed by ``Categorical``) raising ``ValueError`` instead of ``TypeError`` (:issue:`41919`)
+- Bug in :meth:`Categorical.searchsorted` when passing a dtype-incompatible value raising ``KeyError`` instead of ``TypeError`` (:issue:`41919`)
+- Bug in :meth:`Categorical.astype` casting datetimes and :class:`Timestamp` to int for dtype ``object`` (:issue:`44930`)
+- Bug in :meth:`Series.where` with ``CategoricalDtype`` when passing a dtype-incompatible value raising ``ValueError`` instead of ``TypeError`` (:issue:`41919`)
+- Bug in :meth:`Categorical.fillna` when passing a dtype-incompatible value raising ``ValueError`` instead of ``TypeError`` (:issue:`41919`)
+- Bug in :meth:`Categorical.fillna` with a tuple-like category raising ``ValueError`` instead of ``TypeError`` when filling with a non-category tuple (:issue:`41919`)
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :class:`DataFrame` constructor unnecessarily copying non-datetimelike 2D object arrays (:issue:`39272`)
+- Bug in :func:`to_datetime` with ``format`` and ``pandas.NA`` was raising ``ValueError`` (:issue:`42957`)
+- :func:`to_datetime` would silently swap ``MM/DD/YYYY`` and ``DD/MM/YYYY`` formats if the given ``dayfirst`` option could not be respected - now, a warning is raised in the case of delimited date strings (e.g. ``31-12-2012``) (:issue:`12585`)
+- Bug in :meth:`date_range` and :meth:`bdate_range` do not return right bound when ``start`` = ``end`` and set is closed on one side (:issue:`43394`)
+- Bug in inplace addition and subtraction of :class:`DatetimeIndex` or :class:`TimedeltaIndex` with :class:`DatetimeArray` or :class:`TimedeltaArray` (:issue:`43904`)
+- Bug in calling ``np.isnan``, ``np.isfinite``, or ``np.isinf`` on a timezone-aware :class:`DatetimeIndex` incorrectly raising ``TypeError`` (:issue:`43917`)
+- Bug in constructing a :class:`Series` from datetime-like strings with mixed timezones incorrectly partially-inferring datetime values (:issue:`40111`)
+- Bug in addition of a :class:`Tick` object and a ``np.timedelta64`` object incorrectly raising instead of returning :class:`Timedelta` (:issue:`44474`)
+- ``np.maximum.reduce`` and ``np.minimum.reduce`` now correctly return :class:`Timestamp` and :class:`Timedelta` objects when operating on :class:`Series`, :class:`DataFrame`, or :class:`Index` with ``datetime64[ns]`` or ``timedelta64[ns]`` dtype (:issue:`43923`)
+- Bug in adding a ``np.timedelta64`` object to a :class:`BusinessDay` or :class:`CustomBusinessDay` object incorrectly raising (:issue:`44532`)
+- Bug in :meth:`Index.insert` for inserting ``np.datetime64``, ``np.timedelta64`` or ``tuple`` into :class:`Index` with ``dtype='object'`` with negative loc adding ``None`` and replacing existing value (:issue:`44509`)
+- Bug in :meth:`Timestamp.to_pydatetime` failing to retain the ``fold`` attribute (:issue:`45087`)
+- Bug in :meth:`Series.mode` with ``DatetimeTZDtype`` incorrectly returning timezone-naive and ``PeriodDtype`` incorrectly raising (:issue:`41927`)
+- Fixed regression in :meth:`~Series.reindex` raising an error when using an incompatible fill value with a datetime-like dtype (or not raising a deprecation warning for using a ``datetime.date`` as fill value) (:issue:`42921`)
+- Bug in :class:`DateOffset` addition with :class:`Timestamp` where ``offset.nanoseconds`` would not be included in the result (:issue:`43968`, :issue:`36589`)
+- Bug in :meth:`Timestamp.fromtimestamp` not supporting the ``tz`` argument (:issue:`45083`)
+- Bug in :class:`DataFrame` construction from dict of :class:`Series` with mismatched index dtypes sometimes raising depending on the ordering of the passed dict (:issue:`44091`)
+- Bug in :class:`Timestamp` hashing during some DST transitions caused a segmentation fault (:issue:`33931` and :issue:`40817`)
+
+Timedelta
+^^^^^^^^^
+- Bug in division of all-``NaT`` :class:`TimeDeltaIndex`, :class:`Series` or :class:`DataFrame` column with object-dtype array like of numbers failing to infer the result as timedelta64-dtype (:issue:`39750`)
+- Bug in floor division of ``timedelta64[ns]`` data with a scalar returning garbage values (:issue:`44466`)
+- Bug in :class:`Timedelta` now properly taking into account any nanoseconds contribution of any kwarg (:issue:`43764`, :issue:`45227`)
+
+Time Zones
+^^^^^^^^^^
+- Bug in :func:`to_datetime` with ``infer_datetime_format=True`` failing to parse zero UTC offset (``Z``) correctly (:issue:`41047`)
+- Bug in :meth:`Series.dt.tz_convert` resetting index in a :class:`Series` with :class:`CategoricalIndex` (:issue:`43080`)
+- Bug in ``Timestamp`` and ``DatetimeIndex`` incorrectly raising a ``TypeError`` when subtracting two timezone-aware objects with mismatched timezones (:issue:`31793`)
+
+Numeric
+^^^^^^^
+- Bug in floor-dividing a list or tuple of integers by a :class:`Series` incorrectly raising (:issue:`44674`)
+- Bug in :meth:`DataFrame.rank` raising ``ValueError`` with ``object`` columns and ``method="first"`` (:issue:`41931`)
+- Bug in :meth:`DataFrame.rank` treating missing values and extreme values as equal (for example ``np.nan`` and ``np.inf``), causing incorrect results when ``na_option="bottom"`` or ``na_option="top`` used (:issue:`41931`)
+- Bug in ``numexpr`` engine still being used when the option ``compute.use_numexpr`` is set to ``False`` (:issue:`32556`)
+- Bug in :class:`DataFrame` arithmetic ops with a subclass whose :meth:`_constructor` attribute is a callable other than the subclass itself (:issue:`43201`)
+- Bug in arithmetic operations involving :class:`RangeIndex` where the result would have the incorrect ``name`` (:issue:`43962`)
+- Bug in arithmetic operations involving :class:`Series` where the result could have the incorrect ``name`` when the operands having matching NA or matching tuple names (:issue:`44459`)
+- Bug in division with ``IntegerDtype`` or ``BooleanDtype`` array and NA scalar incorrectly raising (:issue:`44685`)
+- Bug in multiplying a :class:`Series` with ``FloatingDtype`` with a timedelta-like scalar incorrectly raising (:issue:`44772`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :class:`UInt64Index` constructor when passing a list containing both positive integers small enough to cast to int64 and integers too large to hold in int64 (:issue:`42201`)
+- Bug in :class:`Series` constructor returning 0 for missing values with dtype ``int64`` and ``False`` for dtype ``bool`` (:issue:`43017`, :issue:`43018`)
+- Bug in constructing a :class:`DataFrame` from a :class:`PandasArray` containing :class:`Series` objects behaving differently than an equivalent ``np.ndarray`` (:issue:`43986`)
+- Bug in :class:`IntegerDtype` not allowing coercion from string dtype (:issue:`25472`)
+- Bug in :func:`to_datetime` with ``arg:xr.DataArray`` and ``unit="ns"`` specified raises ``TypeError`` (:issue:`44053`)
+- Bug in :meth:`DataFrame.convert_dtypes` not returning the correct type when a subclass does not overload :meth:`_constructor_sliced` (:issue:`43201`)
+- Bug in :meth:`DataFrame.astype` not propagating ``attrs`` from the original :class:`DataFrame` (:issue:`44414`)
+- Bug in :meth:`DataFrame.convert_dtypes` result losing ``columns.names`` (:issue:`41435`)
+- Bug in constructing a ``IntegerArray`` from pyarrow data failing to validate dtypes (:issue:`44891`)
+- Bug in :meth:`Series.astype` not allowing converting from a ``PeriodDtype`` to ``datetime64`` dtype, inconsistent with the :class:`PeriodIndex` behavior (:issue:`45038`)
+
+Strings
+^^^^^^^
+- Bug in checking for ``string[pyarrow]`` dtype incorrectly raising an ``ImportError`` when pyarrow is not installed (:issue:`44276`)
+
+Interval
+^^^^^^^^
+- Bug in :meth:`Series.where` with ``IntervalDtype`` incorrectly raising when the ``where`` call should not replace anything (:issue:`44181`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`Series.rename` with :class:`MultiIndex` and ``level`` is provided (:issue:`43659`)
+- Bug in :meth:`DataFrame.truncate` and :meth:`Series.truncate` when the object's :class:`Index` has a length greater than one but only one unique value (:issue:`42365`)
+- Bug in :meth:`Series.loc` and :meth:`DataFrame.loc` with a :class:`MultiIndex` when indexing with a tuple in which one of the levels is also a tuple (:issue:`27591`)
+- Bug in :meth:`Series.loc` with a :class:`MultiIndex` whose first level contains only ``np.nan`` values (:issue:`42055`)
+- Bug in indexing on a :class:`Series` or :class:`DataFrame` with a :class:`DatetimeIndex` when passing a string, the return type depended on whether the index was monotonic (:issue:`24892`)
+- Bug in indexing on a :class:`MultiIndex` failing to drop scalar levels when the indexer is a tuple containing a datetime-like string (:issue:`42476`)
+- Bug in :meth:`DataFrame.sort_values` and :meth:`Series.sort_values` when passing an ascending value, failed to raise or incorrectly raising ``ValueError`` (:issue:`41634`)
+- Bug in updating values of :class:`pandas.Series` using boolean index, created by using :meth:`pandas.DataFrame.pop` (:issue:`42530`)
+- Bug in :meth:`Index.get_indexer_non_unique` when index contains multiple ``np.nan`` (:issue:`35392`)
+- Bug in :meth:`DataFrame.query` did not handle the degree sign in a backticked column name, such as \`Temp(°C)\`, used in an expression to query a :class:`DataFrame` (:issue:`42826`)
+- Bug in :meth:`DataFrame.drop` where the error message did not show missing labels with commas when raising ``KeyError`` (:issue:`42881`)
+- Bug in :meth:`DataFrame.query` where method calls in query strings led to errors when the ``numexpr`` package was installed (:issue:`22435`)
+- Bug in :meth:`DataFrame.nlargest` and :meth:`Series.nlargest` where sorted result did not count indexes containing ``np.nan`` (:issue:`28984`)
+- Bug in indexing on a non-unique object-dtype :class:`Index` with an NA scalar (e.g. ``np.nan``) (:issue:`43711`)
+- Bug in :meth:`DataFrame.__setitem__` incorrectly writing into an existing column's array rather than setting a new array when the new dtype and the old dtype match (:issue:`43406`)
+- Bug in setting floating-dtype values into a :class:`Series` with integer dtype failing to set inplace when those values can be losslessly converted to integers (:issue:`44316`)
+- Bug in :meth:`Series.__setitem__` with object dtype when setting an array with matching size and dtype='datetime64[ns]' or dtype='timedelta64[ns]' incorrectly converting the datetime/timedeltas to integers (:issue:`43868`)
+- Bug in :meth:`DataFrame.sort_index` where ``ignore_index=True`` was not being respected when the index was already sorted (:issue:`43591`)
+- Bug in :meth:`Index.get_indexer_non_unique` when index contains multiple ``np.datetime64("NaT")`` and ``np.timedelta64("NaT")`` (:issue:`43869`)
+- Bug in setting a scalar :class:`Interval` value into a :class:`Series` with ``IntervalDtype`` when the scalar's sides are floats and the values' sides are integers (:issue:`44201`)
+- Bug when setting string-backed :class:`Categorical` values that can be parsed to datetimes into a :class:`DatetimeArray` or :class:`Series` or :class:`DataFrame` column backed by :class:`DatetimeArray` failing to parse these strings (:issue:`44236`)
+- Bug in :meth:`Series.__setitem__` with an integer dtype other than ``int64`` setting with a ``range`` object unnecessarily upcasting to ``int64`` (:issue:`44261`)
+- Bug in :meth:`Series.__setitem__` with a boolean mask indexer setting a listlike value of length 1 incorrectly broadcasting that value (:issue:`44265`)
+- Bug in :meth:`Series.reset_index` not ignoring ``name`` argument when ``drop`` and ``inplace`` are set to ``True`` (:issue:`44575`)
+- Bug in :meth:`DataFrame.loc.__setitem__` and :meth:`DataFrame.iloc.__setitem__` with mixed dtypes sometimes failing to operate in-place (:issue:`44345`)
+- Bug in :meth:`DataFrame.loc.__getitem__` incorrectly raising ``KeyError`` when selecting a single column with a boolean key (:issue:`44322`).
+- Bug in setting :meth:`DataFrame.iloc` with a single ``ExtensionDtype`` column and setting 2D values e.g. ``df.iloc[:] = df.values`` incorrectly raising (:issue:`44514`)
+- Bug in setting values with :meth:`DataFrame.iloc` with a single ``ExtensionDtype`` column and a tuple of arrays as the indexer (:issue:`44703`)
+- Bug in indexing on columns with ``loc`` or ``iloc`` using a slice with a negative step with ``ExtensionDtype`` columns incorrectly raising (:issue:`44551`)
+- Bug in :meth:`DataFrame.loc.__setitem__` changing dtype when indexer was completely ``False`` (:issue:`37550`)
+- Bug in :meth:`IntervalIndex.get_indexer_non_unique` returning boolean mask instead of array of integers for a non unique and non monotonic index (:issue:`44084`)
+- Bug in :meth:`IntervalIndex.get_indexer_non_unique` not handling targets of ``dtype`` 'object' with NaNs correctly (:issue:`44482`)
+- Fixed regression where a single column ``np.matrix`` was no longer coerced to a 1d ``np.ndarray`` when added to a :class:`DataFrame` (:issue:`42376`)
+- Bug in :meth:`Series.__getitem__` with a :class:`CategoricalIndex` of integers treating lists of integers as positional indexers, inconsistent with the behavior with a single scalar integer (:issue:`15470`, :issue:`14865`)
+- Bug in :meth:`Series.__setitem__` when setting floats or integers into integer-dtype :class:`Series` failing to upcast when necessary to retain precision (:issue:`45121`)
+- Bug in :meth:`DataFrame.iloc.__setitem__` ignores axis argument (:issue:`45032`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`DataFrame.fillna` with ``limit`` and no ``method`` ignores ``axis='columns'`` or ``axis = 1`` (:issue:`40989`, :issue:`17399`)
+- Bug in :meth:`DataFrame.fillna` not replacing missing values when using a dict-like ``value`` and duplicate column names (:issue:`43476`)
+- Bug in constructing a :class:`DataFrame` with a dictionary ``np.datetime64`` as a value and ``dtype='timedelta64[ns]'``, or vice-versa, incorrectly casting instead of raising (:issue:`44428`)
+- Bug in :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` with ``inplace=True`` not writing to the underlying array(s) in-place (:issue:`44749`)
+- Bug in :meth:`Index.fillna` incorrectly returning an unfilled :class:`Index` when NA values are present and ``downcast`` argument is specified. This now raises ``NotImplementedError`` instead; do not pass ``downcast`` argument (:issue:`44873`)
+- Bug in :meth:`DataFrame.dropna` changing :class:`Index` even if no entries were dropped (:issue:`41965`)
+- Bug in :meth:`Series.fillna` with an object-dtype incorrectly ignoring ``downcast="infer"`` (:issue:`44241`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`MultiIndex.get_loc` where the first level is a :class:`DatetimeIndex` and a string key is passed (:issue:`42465`)
+- Bug in :meth:`MultiIndex.reindex` when passing a ``level`` that corresponds to an ``ExtensionDtype`` level (:issue:`42043`)
+- Bug in :meth:`MultiIndex.get_loc` raising ``TypeError`` instead of ``KeyError`` on nested tuple (:issue:`42440`)
+- Bug in :meth:`MultiIndex.union` setting wrong ``sortorder`` causing errors in subsequent indexing operations with slices (:issue:`44752`)
+- Bug in :meth:`MultiIndex.putmask` where the other value was also a :class:`MultiIndex` (:issue:`43212`)
+- Bug in :meth:`MultiIndex.dtypes` duplicate level names returned only one dtype per name (:issue:`45174`)
+
+I/O
+^^^
+- Bug in :func:`read_excel` attempting to read chart sheets from .xlsx files (:issue:`41448`)
+- Bug in :func:`json_normalize` where ``errors=ignore`` could fail to ignore missing values of ``meta`` when ``record_path`` has a length greater than one (:issue:`41876`)
+- Bug in :func:`read_csv` with multi-header input and arguments referencing column names as tuples (:issue:`42446`)
+- Bug in :func:`read_fwf`, where difference in lengths of ``colspecs`` and ``names`` was not raising ``ValueError`` (:issue:`40830`)
+- Bug in :func:`Series.to_json` and :func:`DataFrame.to_json` where some attributes were skipped when serializing plain Python objects to JSON (:issue:`42768`, :issue:`33043`)
+- Column headers are dropped when constructing a :class:`DataFrame` from a sqlalchemy's ``Row`` object (:issue:`40682`)
+- Bug in unpickling an :class:`Index` with object dtype incorrectly inferring numeric dtypes (:issue:`43188`)
+- Bug in :func:`read_csv` where reading multi-header input with unequal lengths incorrectly raised ``IndexError`` (:issue:`43102`)
+- Bug in :func:`read_csv` raising ``ParserError`` when reading file in chunks and some chunk blocks have fewer columns than header for ``engine="c"`` (:issue:`21211`)
+- Bug in :func:`read_csv`, changed exception class when expecting a file path name or file-like object from ``OSError`` to ``TypeError`` (:issue:`43366`)
+- Bug in :func:`read_csv` and :func:`read_fwf` ignoring all ``skiprows`` except first when ``nrows`` is specified for ``engine='python'`` (:issue:`44021`, :issue:`10261`)
+- Bug in :func:`read_csv` keeping the original column in object format when ``keep_date_col=True`` is set (:issue:`13378`)
+- Bug in :func:`read_json` not handling non-numpy dtypes correctly (especially ``category``) (:issue:`21892`, :issue:`33205`)
+- Bug in :func:`json_normalize` where multi-character ``sep`` parameter is incorrectly prefixed to every key (:issue:`43831`)
+- Bug in :func:`json_normalize` where reading data with missing multi-level metadata would not respect ``errors="ignore"`` (:issue:`44312`)
+- Bug in :func:`read_csv` used second row to guess implicit index if ``header`` was set to ``None`` for ``engine="python"`` (:issue:`22144`)
+- Bug in :func:`read_csv` not recognizing bad lines when ``names`` were given for ``engine="c"`` (:issue:`22144`)
+- Bug in :func:`read_csv` with :code:`float_precision="round_trip"` which did not skip initial/trailing whitespace (:issue:`43713`)
+- Bug when Python is built without the lzma module: a warning was raised at the pandas import time, even if the lzma capability isn't used (:issue:`43495`)
+- Bug in :func:`read_csv` not applying dtype for ``index_col`` (:issue:`9435`)
+- Bug in dumping/loading a :class:`DataFrame` with ``yaml.dump(frame)`` (:issue:`42748`)
+- Bug in :func:`read_csv` raising ``ValueError`` when ``names`` was longer than ``header`` but equal to data rows for ``engine="python"`` (:issue:`38453`)
+- Bug in :class:`ExcelWriter`, where ``engine_kwargs`` were not passed through to all engines (:issue:`43442`)
+- Bug in :func:`read_csv` raising ``ValueError`` when ``parse_dates`` was used with :class:`MultiIndex` columns (:issue:`8991`)
+- Bug in :func:`read_csv` not raising an ``ValueError`` when ``\n`` was specified as ``delimiter`` or ``sep`` which conflicts with ``lineterminator`` (:issue:`43528`)
+- Bug in :func:`to_csv` converting datetimes in categorical :class:`Series` to integers (:issue:`40754`)
+- Bug in :func:`read_csv` converting columns to numeric after date parsing failed (:issue:`11019`)
+- Bug in :func:`read_csv` not replacing ``NaN`` values with ``np.nan`` before attempting date conversion (:issue:`26203`)
+- Bug in :func:`read_csv` raising ``AttributeError`` when attempting to read a .csv file and infer index column dtype from an nullable integer type (:issue:`44079`)
+- Bug in :func:`to_csv` always coercing datetime columns with different formats to the same format (:issue:`21734`)
+- :meth:`DataFrame.to_csv` and :meth:`Series.to_csv` with ``compression`` set to ``'zip'`` no longer create a zip file containing a file ending with ".zip". Instead, they try to infer the inner file name more smartly (:issue:`39465`)
+- Bug in :func:`read_csv` where reading a mixed column of booleans and missing values to a float type results in the missing values becoming 1.0 rather than NaN (:issue:`42808`, :issue:`34120`)
+- Bug in :func:`to_xml` raising error for ``pd.NA`` with extension array dtype (:issue:`43903`)
+- Bug in :func:`read_csv` when passing simultaneously a parser in ``date_parser`` and ``parse_dates=False``, the parsing was still called (:issue:`44366`)
+- Bug in :func:`read_csv` not setting name of :class:`MultiIndex` columns correctly when ``index_col`` is not the first column (:issue:`38549`)
+- Bug in :func:`read_csv` silently ignoring errors when failing to create a memory-mapped file (:issue:`44766`)
+- Bug in :func:`read_csv` when passing a ``tempfile.SpooledTemporaryFile`` opened in binary mode (:issue:`44748`)
+- Bug in :func:`read_json` raising ``ValueError`` when attempting to parse json strings containing "://" (:issue:`36271`)
+- Bug in :func:`read_csv` when ``engine="c"`` and ``encoding_errors=None`` which caused a segfault (:issue:`45180`)
+- Bug in :func:`read_csv` an invalid value of ``usecols`` leading to an unclosed file handle (:issue:`45384`)
+- Bug in :meth:`DataFrame.to_json` fix memory leak (:issue:`43877`)
+
+Period
+^^^^^^
+- Bug in adding a :class:`Period` object to a ``np.timedelta64`` object incorrectly raising ``TypeError`` (:issue:`44182`)
+- Bug in :meth:`PeriodIndex.to_timestamp` when the index has ``freq="B"`` inferring ``freq="D"`` for its result instead of ``freq="B"`` (:issue:`44105`)
+- Bug in :class:`Period` constructor incorrectly allowing ``np.timedelta64("NaT")`` (:issue:`44507`)
+- Bug in :meth:`PeriodIndex.to_timestamp` giving incorrect values for indexes with non-contiguous data (:issue:`44100`)
+- Bug in :meth:`Series.where` with ``PeriodDtype`` incorrectly raising when the ``where`` call should not replace anything (:issue:`45135`)
+
+Plotting
+^^^^^^^^
+- When given non-numeric data, :meth:`DataFrame.boxplot` now raises a ``ValueError`` rather than a cryptic ``KeyError`` or ``ZeroDivisionError``, in line with other plotting functions like :meth:`DataFrame.hist` (:issue:`43480`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`SeriesGroupBy.apply` where passing an unrecognized string argument failed to raise ``TypeError`` when the underlying ``Series`` is empty (:issue:`42021`)
+- Bug in :meth:`Series.rolling.apply`, :meth:`DataFrame.rolling.apply`, :meth:`Series.expanding.apply` and :meth:`DataFrame.expanding.apply` with ``engine="numba"`` where ``*args`` were being cached with the user passed function (:issue:`42287`)
+- Bug in :meth:`.DataFrameGroupBy.max`, :meth:`.SeriesGroupBy.max`, :meth:`.DataFrameGroupBy.min`, and :meth:`.SeriesGroupBy.min` with nullable integer dtypes losing precision (:issue:`41743`)
+- Bug in :meth:`DataFrame.groupby.rolling.var` would calculate the rolling variance only on the first group (:issue:`42442`)
+- Bug in :meth:`.DataFrameGroupBy.shift` and :meth:`.SeriesGroupBy.shift` that would return the grouping columns if ``fill_value`` was not ``None`` (:issue:`41556`)
+- Bug in :meth:`SeriesGroupBy.nlargest` and :meth:`SeriesGroupBy.nsmallest` would have an inconsistent index when the input :class:`Series` was sorted and ``n`` was greater than or equal to all group sizes (:issue:`15272`, :issue:`16345`, :issue:`29129`)
+- Bug in :meth:`pandas.DataFrame.ewm`, where non-float64 dtypes were silently failing (:issue:`42452`)
+- Bug in :meth:`pandas.DataFrame.rolling` operation along rows (``axis=1``) incorrectly omits columns containing ``float16`` and ``float32`` (:issue:`41779`)
+- Bug in :meth:`Resampler.aggregate` did not allow the use of Named Aggregation (:issue:`32803`)
+- Bug in :meth:`Series.rolling` when the :class:`Series` ``dtype`` was ``Int64`` (:issue:`43016`)
+- Bug in :meth:`DataFrame.rolling.corr` when the :class:`DataFrame` columns was a :class:`MultiIndex` (:issue:`21157`)
+- Bug in :meth:`DataFrame.groupby.rolling` when specifying ``on`` and calling ``__getitem__`` would subsequently return incorrect results (:issue:`43355`)
+- Bug in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` with time-based :class:`Grouper` objects incorrectly raising ``ValueError`` in corner cases where the grouping vector contains a ``NaT`` (:issue:`43500`, :issue:`43515`)
+- Bug in :meth:`.DataFrameGroupBy.mean` and :meth:`.SeriesGroupBy.mean` failing with ``complex`` dtype (:issue:`43701`)
+- Bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` not calculating window bounds correctly for the first row when ``center=True`` and index is decreasing (:issue:`43927`)
+- Bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` for centered datetimelike windows with uneven nanosecond (:issue:`43997`)
+- Bug in :meth:`.DataFrameGroupBy.mean` and :meth:`.SeriesGroupBy.mean` raising ``KeyError`` when column was selected at least twice (:issue:`44924`)
+- Bug in :meth:`.DataFrameGroupBy.nth` and :meth:`.SeriesGroupBy.nth` failing on ``axis=1`` (:issue:`43926`)
+- Bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` not respecting right bound on centered datetime-like windows, if the index contain duplicates (:issue:`3944`)
+- Bug in :meth:`Series.rolling` and :meth:`DataFrame.rolling` when using a :class:`pandas.api.indexers.BaseIndexer` subclass that returned unequal start and end arrays would segfault instead of raising a ``ValueError`` (:issue:`44470`)
+- Bug in :meth:`Groupby.nunique` not respecting ``observed=True`` for ``categorical`` grouping columns (:issue:`45128`)
+- Bug in :meth:`.DataFrameGroupBy.head`, :meth:`.SeriesGroupBy.head`, :meth:`.DataFrameGroupBy.tail`, and :meth:`.SeriesGroupBy.tail` not dropping groups with ``NaN`` when ``dropna=True`` (:issue:`45089`)
+- Bug in :meth:`GroupBy.__iter__` after selecting a subset of columns in a :class:`GroupBy` object, which returned all columns instead of the chosen subset (:issue:`44821`)
+- Bug in :meth:`Groupby.rolling` when non-monotonic data passed, fails to correctly raise ``ValueError`` (:issue:`43909`)
+- Bug where grouping by a :class:`Series` that has a ``categorical`` data type and length unequal to the axis of grouping raised ``ValueError`` (:issue:`44179`)
+
+Reshaping
+^^^^^^^^^
+- Improved error message when creating a :class:`DataFrame` column from a multi-dimensional :class:`numpy.ndarray` (:issue:`42463`)
+- Bug in :func:`concat` creating :class:`MultiIndex` with duplicate level entries when concatenating a :class:`DataFrame` with duplicates in :class:`Index` and multiple keys (:issue:`42651`)
+- Bug in :meth:`pandas.cut` on :class:`Series` with duplicate indices and non-exact :meth:`pandas.CategoricalIndex` (:issue:`42185`, :issue:`42425`)
+- Bug in :meth:`DataFrame.append` failing to retain dtypes when appended columns do not match (:issue:`43392`)
+- Bug in :func:`concat` of ``bool`` and ``boolean`` dtypes resulting in ``object`` dtype instead of ``boolean`` dtype (:issue:`42800`)
+- Bug in :func:`crosstab` when inputs are categorical :class:`Series`, there are categories that are not present in one or both of the :class:`Series`, and ``margins=True``. Previously the margin value for missing categories was ``NaN``. It is now correctly reported as 0 (:issue:`43505`)
+- Bug in :func:`concat` would fail when the ``objs`` argument all had the same index and the ``keys`` argument contained duplicates (:issue:`43595`)
+- Bug in :func:`concat` which ignored the ``sort`` parameter (:issue:`43375`)
+- Bug in :func:`merge` with :class:`MultiIndex` as column index for the ``on`` argument returning an error when assigning a column internally (:issue:`43734`)
+- Bug in :func:`crosstab` would fail when inputs are lists or tuples (:issue:`44076`)
+- Bug in :meth:`DataFrame.append` failing to retain ``index.name`` when appending a list of :class:`Series` objects (:issue:`44109`)
+- Fixed metadata propagation in :meth:`Dataframe.apply` method, consequently fixing the same issue for :meth:`Dataframe.transform`, :meth:`Dataframe.nunique` and :meth:`Dataframe.mode` (:issue:`28283`)
+- Bug in :func:`concat` casting levels of :class:`MultiIndex` to float if all levels only consist of missing values (:issue:`44900`)
+- Bug in :meth:`DataFrame.stack` with ``ExtensionDtype`` columns incorrectly raising (:issue:`43561`)
+- Bug in :func:`merge` raising ``KeyError`` when joining over differently named indexes with on keywords (:issue:`45094`)
+- Bug in :meth:`Series.unstack` with object doing unwanted type inference on resulting columns (:issue:`44595`)
+- Bug in :meth:`MultiIndex.join` with overlapping ``IntervalIndex`` levels (:issue:`44096`)
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` results is different ``dtype`` based on ``regex`` parameter (:issue:`44864`)
+- Bug in :meth:`DataFrame.pivot` with ``index=None`` when the :class:`DataFrame` index was a :class:`MultiIndex` (:issue:`23955`)
+
+Sparse
+^^^^^^
+- Bug in :meth:`DataFrame.sparse.to_coo` raising ``AttributeError`` when column names are not unique (:issue:`29564`)
+- Bug in :meth:`SparseArray.max` and :meth:`SparseArray.min` raising ``ValueError`` for arrays with 0 non-null elements (:issue:`43527`)
+- Bug in :meth:`DataFrame.sparse.to_coo` silently converting non-zero fill values to zero (:issue:`24817`)
+- Bug in :class:`SparseArray` comparison methods with an array-like operand of mismatched length raising ``AssertionError`` or unclear ``ValueError`` depending on the input (:issue:`43863`)
+- Bug in :class:`SparseArray` arithmetic methods ``floordiv`` and ``mod`` behaviors when dividing by zero not matching the non-sparse :class:`Series` behavior (:issue:`38172`)
+- Bug in :class:`SparseArray` unary methods as well as :meth:`SparseArray.isna` doesn't recalculate indexes (:issue:`44955`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :func:`array` failing to preserve :class:`PandasArray` (:issue:`43887`)
+- NumPy ufuncs ``np.abs``, ``np.positive``, ``np.negative`` now correctly preserve dtype when called on ExtensionArrays that implement ``__abs__, __pos__, __neg__``, respectively. In particular this is fixed for :class:`TimedeltaArray` (:issue:`43899`, :issue:`23316`)
+- NumPy ufuncs ``np.minimum.reduce`` ``np.maximum.reduce``, ``np.add.reduce``, and ``np.prod.reduce`` now work correctly instead of raising ``NotImplementedError`` on :class:`Series` with ``IntegerDtype`` or ``FloatDtype`` (:issue:`43923`, :issue:`44793`)
+- NumPy ufuncs with ``out`` keyword are now supported by arrays with ``IntegerDtype`` and ``FloatingDtype`` (:issue:`45122`)
+- Avoid raising ``PerformanceWarning`` about fragmented :class:`DataFrame` when using many columns with an extension dtype (:issue:`44098`)
+- Bug in :class:`IntegerArray` and :class:`FloatingArray` construction incorrectly coercing mismatched NA values (e.g. ``np.timedelta64("NaT")``) to numeric NA (:issue:`44514`)
+- Bug in :meth:`BooleanArray.__eq__` and :meth:`BooleanArray.__ne__` raising ``TypeError`` on comparison with an incompatible type (like a string). This caused :meth:`DataFrame.replace` to sometimes raise a ``TypeError`` if a nullable boolean column was included (:issue:`44499`)
+- Bug in :func:`array` incorrectly raising when passed a ``ndarray`` with ``float16`` dtype (:issue:`44715`)
+- Bug in calling ``np.sqrt`` on :class:`BooleanArray` returning a malformed :class:`FloatingArray` (:issue:`44715`)
+- Bug in :meth:`Series.where` with ``ExtensionDtype`` when ``other`` is a NA scalar incompatible with the :class:`Series` dtype (e.g. ``NaT`` with a numeric dtype) incorrectly casting to a compatible NA value (:issue:`44697`)
+- Bug in :meth:`Series.replace` where explicitly passing ``value=None`` is treated as if no ``value`` was passed, and ``None`` not being in the result (:issue:`36984`, :issue:`19998`)
+- Bug in :meth:`Series.replace` with unwanted downcasting being done in no-op replacements (:issue:`44498`)
+- Bug in :meth:`Series.replace` with ``FloatDtype``, ``string[python]``, or ``string[pyarrow]`` dtype not being preserved when possible (:issue:`33484`, :issue:`40732`, :issue:`31644`, :issue:`41215`, :issue:`25438`)
+
+Styler
+^^^^^^
+- Bug in :class:`.Styler` where the ``uuid`` at initialization maintained a floating underscore (:issue:`43037`)
+- Bug in :meth:`.Styler.to_html` where the ``Styler`` object was updated if the ``to_html`` method was called with some args (:issue:`43034`)
+- Bug in :meth:`.Styler.copy` where ``uuid`` was not previously copied (:issue:`40675`)
+- Bug in :meth:`Styler.apply` where functions which returned :class:`Series` objects were not correctly handled in terms of aligning their index labels (:issue:`13657`, :issue:`42014`)
+- Bug when rendering an empty :class:`DataFrame` with a named :class:`Index` (:issue:`43305`)
+- Bug when rendering a single level :class:`MultiIndex` (:issue:`43383`)
+- Bug when combining non-sparse rendering and :meth:`.Styler.hide_columns` or :meth:`.Styler.hide_index` (:issue:`43464`)
+- Bug setting a table style when using multiple selectors in :class:`.Styler` (:issue:`44011`)
+- Bugs where row trimming and column trimming failed to reflect hidden rows (:issue:`43703`, :issue:`44247`)
+
+Other
+^^^^^
+- Bug in :meth:`DataFrame.astype` with non-unique columns and a :class:`Series` ``dtype`` argument (:issue:`44417`)
+- Bug in :meth:`CustomBusinessMonthBegin.__add__` (:meth:`CustomBusinessMonthEnd.__add__`) not applying the extra ``offset`` parameter when beginning (end) of the target month is already a business day (:issue:`41356`)
+- Bug in :meth:`RangeIndex.union` with another ``RangeIndex`` with matching (even) ``step`` and starts differing by strictly less than ``step / 2`` (:issue:`44019`)
+- Bug in :meth:`RangeIndex.difference` with ``sort=None`` and ``step<0`` failing to sort (:issue:`44085`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` with ``value=None`` and ExtensionDtypes (:issue:`44270`, :issue:`37899`)
+- Bug in :meth:`FloatingArray.equals` failing to consider two arrays equal if they contain ``np.nan`` values (:issue:`44382`)
+- Bug in :meth:`DataFrame.shift` with ``axis=1`` and ``ExtensionDtype`` columns incorrectly raising when an incompatible ``fill_value`` is passed (:issue:`44564`)
+- Bug in :meth:`DataFrame.shift` with ``axis=1`` and ``periods`` larger than ``len(frame.columns)`` producing an invalid :class:`DataFrame` (:issue:`44978`)
+- Bug in :meth:`DataFrame.diff` when passing a NumPy integer object instead of an ``int`` object (:issue:`44572`)
+- Bug in :meth:`Series.replace` raising ``ValueError`` when using ``regex=True`` with a :class:`Series` containing ``np.nan`` values (:issue:`43344`)
+- Bug in :meth:`DataFrame.to_records` where an incorrect ``n`` was used when missing names were replaced by ``level_n`` (:issue:`44818`)
+- Bug in :meth:`DataFrame.eval` where ``resolvers`` argument was overriding the default resolvers (:issue:`34966`)
+- :meth:`Series.__repr__` and :meth:`DataFrame.__repr__` no longer replace all null-values in indexes with "NaN" but use their real string-representations. "NaN" is used only for ``float("nan")`` (:issue:`45263`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_140.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.3.5..v1.4.0
diff --git a/pandas/doc/source/whatsnew/v1.4.1.rst b/pandas/doc/source/whatsnew/v1.4.1.rst
new file mode 100644
index 0000000000000..dd2002bb87648
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.4.1.rst
@@ -0,0 +1,56 @@
+.. _whatsnew_141:
+
+What's new in 1.4.1 (February 12, 2022)
+---------------------------------------
+
+These are the changes in pandas 1.4.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_141.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Regression in :meth:`Series.mask` with ``inplace=True`` and ``PeriodDtype`` and an incompatible ``other`` coercing to a common dtype instead of raising (:issue:`45546`)
+- Regression in :func:`.assert_frame_equal` not respecting ``check_flags=False`` (:issue:`45554`)
+- Regression in :meth:`DataFrame.loc` raising ``ValueError`` when indexing (getting values) on a :class:`MultiIndex` with one level (:issue:`45779`)
+- Regression in :meth:`Series.fillna` with ``downcast=False`` incorrectly downcasting ``object`` dtype (:issue:`45603`)
+- Regression in :func:`api.types.is_bool_dtype` raising an ``AttributeError`` when evaluating a categorical :class:`Series` (:issue:`45615`)
+- Regression in :meth:`DataFrame.iat` setting values leading to not propagating correctly in subsequent lookups (:issue:`45684`)
+- Regression when setting values with :meth:`DataFrame.loc` losing :class:`Index` name if :class:`DataFrame` was empty before (:issue:`45621`)
+- Regression in :meth:`~Index.join` with overlapping :class:`IntervalIndex` raising an ``InvalidIndexError`` (:issue:`45661`)
+- Regression when setting values with :meth:`Series.loc` raising with all ``False`` indexer and :class:`Series` on the right hand side (:issue:`45778`)
+- Regression in :func:`read_sql` with a DBAPI2 connection that is not an instance of ``sqlite3.Connection`` incorrectly requiring SQLAlchemy be installed (:issue:`45660`)
+- Regression in :class:`DateOffset` when constructing with an integer argument with no keywords (e.g. ``pd.DateOffset(n)``) would behave like ``datetime.timedelta(days=0)`` (:issue:`45643`, :issue:`45890`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_141.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed segfault in :meth:`DataFrame.to_json` when dumping tz-aware datetimes in Python 3.10 (:issue:`42130`)
+- Stopped emitting unnecessary ``FutureWarning`` in :meth:`DataFrame.sort_values` with sparse columns (:issue:`45618`)
+- Fixed window aggregations in :meth:`DataFrame.rolling` and :meth:`Series.rolling` to skip over unused elements (:issue:`45647`)
+- Fixed builtin highlighters in :class:`.Styler` to be responsive to ``NA`` with nullable dtypes (:issue:`45804`)
+- Bug in :meth:`~Rolling.apply` with ``axis=1`` raising an erroneous ``ValueError`` (:issue:`45912`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_141.other:
+
+Other
+~~~~~
+- Reverted performance speedup of :meth:`DataFrame.corr` for ``method=pearson`` to fix precision regression (:issue:`45640`, :issue:`42761`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_141.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.4.0..v1.4.1
diff --git a/pandas/doc/source/whatsnew/v1.4.2.rst b/pandas/doc/source/whatsnew/v1.4.2.rst
new file mode 100644
index 0000000000000..64d7e683c9d27
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.4.2.rst
@@ -0,0 +1,45 @@
+.. _whatsnew_142:
+
+What's new in 1.4.2 (April 2, 2022)
+-----------------------------------
+
+These are the changes in pandas 1.4.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_142.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.drop` and :meth:`Series.drop` when :class:`Index` had extension dtype and duplicates (:issue:`45860`)
+- Fixed regression in :func:`read_csv` killing python process when invalid file input was given for ``engine="c"`` (:issue:`45957`)
+- Fixed memory performance regression in :meth:`Series.fillna` when called on a :class:`DataFrame` column with ``inplace=True`` (:issue:`46149`)
+- Provided an alternative solution for passing custom Excel formats in :meth:`.Styler.to_excel`, which was a regression based on stricter CSS validation. Examples available in the documentation for :meth:`.Styler.format` (:issue:`46152`)
+- Fixed regression in :meth:`DataFrame.replace` when a replacement value was also a target for replacement (:issue:`46306`)
+- Fixed regression in :meth:`DataFrame.replace` when the replacement value was explicitly ``None`` when passed in a dictionary to ``to_replace`` (:issue:`45601`, :issue:`45836`)
+- Fixed regression when setting values with :meth:`DataFrame.loc` losing :class:`MultiIndex` names if :class:`DataFrame`  was empty before (:issue:`46317`)
+- Fixed regression when rendering boolean datatype columns with :meth:`.Styler` (:issue:`46384`)
+- Fixed regression in :meth:`Groupby.rolling` with a frequency window that would raise a ``ValueError`` even if the datetimes within each group were monotonic (:issue:`46061`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_142.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fix some cases for subclasses that define their ``_constructor`` properties as general callables (:issue:`46018`)
+- Fixed "longtable" formatting in :meth:`.Styler.to_latex` when ``column_format`` is given in extended format (:issue:`46037`)
+- Fixed incorrect rendering in :meth:`.Styler.format` with ``hyperlinks="html"`` when the url contains a colon or other special characters (:issue:`46389`)
+- Improved error message in :class:`.Rolling` when ``window`` is a frequency and ``NaT`` is in the rolling axis (:issue:`46087`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_142.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.4.1..v1.4.2
diff --git a/pandas/doc/source/whatsnew/v1.4.3.rst b/pandas/doc/source/whatsnew/v1.4.3.rst
new file mode 100644
index 0000000000000..70b451a231453
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.4.3.rst
@@ -0,0 +1,72 @@
+.. _whatsnew_143:
+
+What's new in 1.4.3 (June 23, 2022)
+-----------------------------------
+
+These are the changes in pandas 1.4.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_143.concat:
+
+Behavior of ``concat`` with empty or all-NA DataFrame columns
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The behavior change in version 1.4.0 to stop ignoring the data type
+of empty or all-NA columns with float or object dtype in :func:`concat`
+(:ref:`whatsnew_140.notable_bug_fixes.concat_with_empty_or_all_na`) has been
+reverted (:issue:`45637`).
+
+
+.. _whatsnew_143.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.replace` when the replacement value was explicitly ``None`` when passed in a dictionary to ``to_replace`` also casting other columns to object dtype even when there were no values to replace (:issue:`46634`)
+- Fixed regression in :meth:`DataFrame.to_csv` raising error when :class:`DataFrame` contains extension dtype categorical column (:issue:`46297`, :issue:`46812`)
+- Fixed regression in representation of ``dtypes`` attribute of :class:`MultiIndex` (:issue:`46900`)
+- Fixed regression when setting values with :meth:`DataFrame.loc` updating :class:`RangeIndex` when index was set as new column and column was updated afterwards (:issue:`47128`)
+- Fixed regression in :meth:`DataFrame.fillna` and :meth:`DataFrame.update` creating a copy when updating inplace (:issue:`47188`)
+- Fixed regression in :meth:`DataFrame.nsmallest` led to wrong results when the sorting column has ``np.nan`` values (:issue:`46589`)
+- Fixed regression in :func:`read_fwf` raising ``ValueError`` when ``widths`` was specified with ``usecols`` (:issue:`46580`)
+- Fixed regression in :func:`concat` not sorting columns for mixed column names (:issue:`47127`)
+- Fixed regression in :meth:`.Groupby.transform` and :meth:`.Groupby.agg` failing with ``engine="numba"`` when the index was a :class:`MultiIndex` (:issue:`46867`)
+- Fixed regression in ``NaN`` comparison for :class:`Index` operations where the same object was compared (:issue:`47105`)
+- Fixed regression is :meth:`.Styler.to_latex` and :meth:`.Styler.to_html` where ``buf`` failed in combination with ``encoding`` (:issue:`47053`)
+- Fixed regression in :func:`read_csv` with ``index_col=False`` identifying first row as index names when ``header=None`` (:issue:`46955`)
+- Fixed regression in :meth:`.DataFrameGroupBy.agg` when used with list-likes or dict-likes and ``axis=1`` that would give incorrect results; now raises ``NotImplementedError`` (:issue:`46995`)
+- Fixed regression in :meth:`DataFrame.resample` and :meth:`DataFrame.rolling` when used with list-likes or dict-likes and ``axis=1`` that would raise an unintuitive error message; now raises ``NotImplementedError`` (:issue:`46904`)
+- Fixed regression in :func:`testing.assert_index_equal` when ``check_order=False`` and :class:`Index` has extension or object dtype (:issue:`47207`)
+- Fixed regression in :func:`read_excel` returning ints as floats on certain input sheets (:issue:`46988`)
+- Fixed regression in :meth:`DataFrame.shift` when ``axis`` is ``columns`` and ``fill_value`` is absent, ``freq`` is ignored (:issue:`47039`)
+- Fixed regression in :meth:`DataFrame.to_json` causing a segmentation violation when :class:`DataFrame` is created with an ``index`` parameter of the type :class:`PeriodIndex` (:issue:`46683`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_143.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :func:`pandas.eval`, :meth:`DataFrame.eval` and :meth:`DataFrame.query` where passing empty ``local_dict`` or ``global_dict`` was treated as passing ``None`` (:issue:`47084`)
+- Most I/O methods no longer suppress ``OSError`` and ``ValueError`` when closing file handles (:issue:`47136`)
+- Improving error message raised by :meth:`DataFrame.from_dict` when passing an invalid ``orient`` parameter (:issue:`47450`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_143.other:
+
+Other
+~~~~~
+- The minimum version of Cython needed to compile pandas is now ``0.29.30`` (:issue:`41935`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_143.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.4.2..v1.4.3
diff --git a/pandas/doc/source/whatsnew/v1.4.4.rst b/pandas/doc/source/whatsnew/v1.4.4.rst
new file mode 100644
index 0000000000000..56b1254d8a359
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.4.4.rst
@@ -0,0 +1,65 @@
+.. _whatsnew_144:
+
+What's new in 1.4.4 (August 31, 2022)
+-------------------------------------
+
+These are the changes in pandas 1.4.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_144.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.fillna` not working on a :class:`DataFrame` with a :class:`MultiIndex` (:issue:`47649`)
+- Fixed regression in taking NULL :class:`objects` from a :class:`DataFrame` causing a segmentation violation. These NULL values are created by :meth:`numpy.empty_like` (:issue:`46848`)
+- Fixed regression in :func:`concat` materializing the :class:`Index` during sorting even if the :class:`Index` was already sorted (:issue:`47501`)
+- Fixed regression in :func:`concat` or :func:`merge` handling of all-NaN ExtensionArrays with custom attributes (:issue:`47762`)
+- Fixed regression in calling bitwise numpy ufuncs (for example, ``np.bitwise_and``) on Index objects (:issue:`46769`)
+- Fixed regression in :func:`cut` when using a ``datetime64`` IntervalIndex as bins (:issue:`46218`)
+- Fixed regression in :meth:`DataFrame.select_dtypes` where ``include="number"`` included :class:`BooleanDtype` (:issue:`46870`)
+- Fixed regression in :meth:`DataFrame.loc` raising error when indexing with a ``NamedTuple`` (:issue:`48124`)
+- Fixed regression in :meth:`DataFrame.loc` not updating the cache correctly after values were set (:issue:`47867`)
+- Fixed regression in :meth:`DataFrame.loc` not aligning index in some cases when setting a :class:`DataFrame` (:issue:`47578`)
+- Fixed regression in :meth:`DataFrame.loc` setting a length-1 array like value to a single value in the DataFrame (:issue:`46268`)
+- Fixed regression when slicing with :meth:`DataFrame.loc` with :class:`DatetimeIndex` with a :class:`.DateOffset` object for its ``freq`` (:issue:`46671`)
+- Fixed regression in setting ``None`` or non-string value into a ``string``-dtype Series using a mask (:issue:`47628`)
+- Fixed regression in updating a DataFrame column through Series ``__setitem__`` (using chained assignment) not updating column values inplace and using too much memory (:issue:`47172`)
+- Fixed regression in :meth:`DataFrame.select_dtypes` returning a view on the original DataFrame (:issue:`48090`)
+- Fixed regression using custom Index subclasses (for example, used in xarray) with :meth:`~DataFrame.reset_index` or :meth:`Index.insert` (:issue:`47071`)
+- Fixed regression in :meth:`~Index.intersection` when the :class:`DatetimeIndex` has dates crossing daylight savings time (:issue:`46702`)
+- Fixed regression in :func:`merge` throwing an error when passing a :class:`Series` with a multi-level name (:issue:`47946`)
+- Fixed regression in :meth:`DataFrame.eval` creating a copy when updating inplace (:issue:`47449`)
+- Fixed regression where getting a row using :meth:`DataFrame.iloc` with :class:`SparseDtype` would raise (:issue:`46406`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_144.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- The ``FutureWarning`` raised when passing arguments (other than ``filepath_or_buffer``) as positional in :func:`read_csv` is now raised at the correct stacklevel (:issue:`47385`)
+- Bug in :meth:`DataFrame.to_sql` when ``method`` was a ``callable`` that did not return an ``int`` and would raise a ``TypeError`` (:issue:`46891`)
+- Bug in :meth:`.DataFrameGroupBy.value_counts` where ``subset`` had no effect (:issue:`46383`)
+- Bug when getting values with :meth:`DataFrame.loc` with a list of keys causing an internal inconsistency that could lead to a disconnect between ``frame.at[x, y]`` vs ``frame[y].loc[x]`` (:issue:`22372`)
+- Bug in the :meth:`Series.dt.strftime` accessor return a float instead of object dtype Series for all-NaT input, which also causes a spurious deprecation warning (:issue:`45858`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_144.other:
+
+Other
+~~~~~
+- The minimum version of Cython needed to compile pandas is now ``0.29.32`` (:issue:`47978`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_144.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.4.3..v1.4.4|HEAD
diff --git a/pandas/doc/source/whatsnew/v1.5.0.rst b/pandas/doc/source/whatsnew/v1.5.0.rst
new file mode 100644
index 0000000000000..43aa63c284f38
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.5.0.rst
@@ -0,0 +1,1313 @@
+.. _whatsnew_150:
+
+What's new in 1.5.0 (September 19, 2022)
+----------------------------------------
+
+These are the changes in pandas 1.5.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_150.enhancements.pandas-stubs:
+
+``pandas-stubs``
+^^^^^^^^^^^^^^^^
+
+The ``pandas-stubs`` library is now supported by the pandas development team, providing type stubs for the pandas API. Please visit
+https://github.com/pandas-dev/pandas-stubs for more information.
+
+We thank VirtusLab and Microsoft for their initial, significant contributions to ``pandas-stubs``
+
+.. _whatsnew_150.enhancements.arrow:
+
+Native PyArrow-backed ExtensionArray
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+With `Pyarrow <https://arrow.apache.org/docs/python/index.html>`__ installed, users can now create pandas objects
+that are backed by a ``pyarrow.ChunkedArray`` and ``pyarrow.DataType``.
+
+The ``dtype`` argument can accept a string of a `pyarrow data type <https://arrow.apache.org/docs/python/api/datatypes.html>`__
+with ``pyarrow`` in brackets e.g. ``"int64[pyarrow]"`` or, for pyarrow data types that take parameters, a :class:`ArrowDtype`
+initialized with a ``pyarrow.DataType``.
+
+.. ipython:: python
+
+    import pyarrow as pa
+    ser_float = pd.Series([1.0, 2.0, None], dtype="float32[pyarrow]")
+    ser_float
+
+    list_of_int_type = pd.ArrowDtype(pa.list_(pa.int64()))
+    ser_list = pd.Series([[1, 2], [3, None]], dtype=list_of_int_type)
+    ser_list
+
+    ser_list.take([1, 0])
+    ser_float * 5
+    ser_float.mean()
+    ser_float.dropna()
+
+Most operations are supported and have been implemented using `pyarrow compute <https://arrow.apache.org/docs/python/api/compute.html>`__ functions.
+We recommend installing the latest version of PyArrow to access the most recently implemented compute functions.
+
+.. warning::
+
+    This feature is experimental, and the API can change in a future release without warning.
+
+.. _whatsnew_150.enhancements.dataframe_interchange:
+
+DataFrame interchange protocol implementation
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas now implement the DataFrame interchange API spec.
+See the full details on the API at https://data-apis.org/dataframe-protocol/latest/index.html
+
+The protocol consists of two parts:
+
+- New method :meth:`DataFrame.__dataframe__` which produces the interchange object.
+  It effectively "exports" the pandas dataframe as an interchange object so
+  any other library which has the protocol implemented can "import" that dataframe
+  without knowing anything about the producer except that it makes an interchange object.
+- New function :func:`pandas.api.interchange.from_dataframe` which can take
+  an arbitrary interchange object from any conformant library and construct a
+  pandas DataFrame out of it.
+
+.. _whatsnew_150.enhancements.styler:
+
+Styler
+^^^^^^
+
+The most notable development is the new method :meth:`.Styler.concat` which
+allows adding customised footer rows to visualise additional calculations on the data,
+e.g. totals and counts etc. (:issue:`43875`, :issue:`46186`)
+
+Additionally there is an alternative output method :meth:`.Styler.to_string`,
+which allows using the Styler's formatting methods to create, for example, CSVs (:issue:`44502`).
+
+A new feature :meth:`.Styler.relabel_index` is also made available to provide full customisation of the display of
+index or column headers (:issue:`47864`)
+
+Minor feature improvements are:
+
+  - Adding the ability to render ``border`` and ``border-{side}`` CSS properties in Excel (:issue:`42276`)
+  - Making keyword arguments consist: :meth:`.Styler.highlight_null` now accepts ``color`` and deprecates ``null_color`` although this remains backwards compatible (:issue:`45907`)
+
+.. _whatsnew_150.enhancements.resample_group_keys:
+
+Control of index with ``group_keys`` in :meth:`DataFrame.resample`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The argument ``group_keys`` has been added to the method :meth:`DataFrame.resample`.
+As with :meth:`DataFrame.groupby`, this argument controls the whether each group is added
+to the index in the resample when :meth:`.Resampler.apply` is used.
+
+.. warning::
+   Not specifying the ``group_keys`` argument will retain the
+   previous behavior and emit a warning if the result will change
+   by specifying ``group_keys=False``. In a future version
+   of pandas, not specifying ``group_keys`` will default to
+   the same behavior as ``group_keys=False``.
+
+.. code-block:: ipython
+
+    In [11]: df = pd.DataFrame(
+       ....:     {'a': range(6)},
+       ....:     index=pd.date_range("2021-01-01", periods=6, freq="8H")
+       ....: )
+       ....:
+
+    In [12]: df.resample("D", group_keys=True).apply(lambda x: x)
+    Out[12]:
+                                    a
+    2021-01-01 2021-01-01 00:00:00  0
+               2021-01-01 08:00:00  1
+               2021-01-01 16:00:00  2
+    2021-01-02 2021-01-02 00:00:00  3
+               2021-01-02 08:00:00  4
+               2021-01-02 16:00:00  5
+
+    In [13]: df.resample("D", group_keys=False).apply(lambda x: x)
+    Out[13]:
+                         a
+    2021-01-01 00:00:00  0
+    2021-01-01 08:00:00  1
+    2021-01-01 16:00:00  2
+    2021-01-02 00:00:00  3
+    2021-01-02 08:00:00  4
+    2021-01-02 16:00:00  5
+
+Previously, the resulting index would depend upon the values returned by ``apply``,
+as seen in the following example.
+
+.. code-block:: ipython
+
+    In [1]: # pandas 1.3
+    In [2]: df.resample("D").apply(lambda x: x)
+    Out[2]:
+                         a
+    2021-01-01 00:00:00  0
+    2021-01-01 08:00:00  1
+    2021-01-01 16:00:00  2
+    2021-01-02 00:00:00  3
+    2021-01-02 08:00:00  4
+    2021-01-02 16:00:00  5
+
+    In [3]: df.resample("D").apply(lambda x: x.reset_index())
+    Out[3]:
+                               index  a
+    2021-01-01 0 2021-01-01 00:00:00  0
+               1 2021-01-01 08:00:00  1
+               2 2021-01-01 16:00:00  2
+    2021-01-02 0 2021-01-02 00:00:00  3
+               1 2021-01-02 08:00:00  4
+               2 2021-01-02 16:00:00  5
+
+.. _whatsnew_150.enhancements.from_dummies:
+
+from_dummies
+^^^^^^^^^^^^
+
+Added new function :func:`~pandas.from_dummies` to convert a dummy coded :class:`DataFrame` into a categorical :class:`DataFrame`.
+
+.. ipython:: python
+
+    import pandas as pd
+
+    df = pd.DataFrame({"col1_a": [1, 0, 1], "col1_b": [0, 1, 0],
+                       "col2_a": [0, 1, 0], "col2_b": [1, 0, 0],
+                       "col2_c": [0, 0, 1]})
+
+    pd.from_dummies(df, sep="_")
+
+.. _whatsnew_150.enhancements.orc:
+
+Writing to ORC files
+^^^^^^^^^^^^^^^^^^^^
+
+The new method :meth:`DataFrame.to_orc` allows writing to ORC files (:issue:`43864`).
+
+This functionality depends the `pyarrow <http://arrow.apache.org/docs/python/>`__ library. For more details, see :ref:`the IO docs on ORC <io.orc>`.
+
+.. warning::
+
+   * It is *highly recommended* to install pyarrow using conda due to some issues occurred by pyarrow.
+   * :func:`~pandas.DataFrame.to_orc` requires pyarrow>=7.0.0.
+   * :func:`~pandas.DataFrame.to_orc` is not supported on Windows yet, you can find valid environments on :ref:`install optional dependencies <install.warn_orc>`.
+   * For supported dtypes please refer to `supported ORC features in Arrow <https://arrow.apache.org/docs/cpp/orc.html#data-types>`__.
+   * Currently timezones in datetime columns are not preserved when a dataframe is converted into ORC files.
+
+.. code-block:: python
+
+    df = pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
+    df.to_orc("./out.orc")
+
+.. _whatsnew_150.enhancements.tar:
+
+Reading directly from TAR archives
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+I/O methods like :func:`read_csv` or :meth:`DataFrame.to_json` now allow reading and writing
+directly on TAR archives (:issue:`44787`).
+
+.. code-block:: python
+
+   df = pd.read_csv("./movement.tar.gz")
+   # ...
+   df.to_csv("./out.tar.gz")
+
+This supports ``.tar``, ``.tar.gz``, ``.tar.bz`` and ``.tar.xz2`` archives.
+The used compression method is inferred from the filename.
+If the compression method cannot be inferred, use the ``compression`` argument:
+
+.. code-block:: python
+
+   df = pd.read_csv(some_file_obj, compression={"method": "tar", "mode": "r:gz"}) # noqa F821
+
+(``mode`` being one of ``tarfile.open``'s modes: https://docs.python.org/3/library/tarfile.html#tarfile.open)
+
+
+.. _whatsnew_150.enhancements.read_xml_dtypes:
+
+read_xml now supports ``dtype``, ``converters``, and ``parse_dates``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Similar to other IO methods, :func:`pandas.read_xml` now supports assigning specific dtypes to columns,
+apply converter methods, and parse dates (:issue:`43567`).
+
+.. ipython:: python
+
+    from io import StringIO
+    xml_dates = """<?xml version='1.0' encoding='utf-8'?>
+    <data>
+      <row>
+        <shape>square</shape>
+        <degrees>00360</degrees>
+        <sides>4.0</sides>
+        <date>2020-01-01</date>
+       </row>
+      <row>
+        <shape>circle</shape>
+        <degrees>00360</degrees>
+        <sides/>
+        <date>2021-01-01</date>
+      </row>
+      <row>
+        <shape>triangle</shape>
+        <degrees>00180</degrees>
+        <sides>3.0</sides>
+        <date>2022-01-01</date>
+      </row>
+    </data>"""
+
+    df = pd.read_xml(
+        StringIO(xml_dates),
+        dtype={'sides': 'Int64'},
+        converters={'degrees': str},
+        parse_dates=['date']
+    )
+    df
+    df.dtypes
+
+
+.. _whatsnew_150.enhancements.read_xml_iterparse:
+
+read_xml now supports large XML using ``iterparse``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+For very large XML files that can range in hundreds of megabytes to gigabytes, :func:`pandas.read_xml`
+now supports parsing such sizeable files using `lxml's iterparse`_ and `etree's iterparse`_
+which are memory-efficient methods to iterate through XML trees and extract specific elements
+and attributes without holding entire tree in memory (:issue:`45442`).
+
+.. code-block:: ipython
+
+    In [1]: df = pd.read_xml(
+    ...      "/path/to/downloaded/enwikisource-latest-pages-articles.xml",
+    ...      iterparse = {"page": ["title", "ns", "id"]})
+    ...  )
+    df
+    Out[2]:
+                                                         title   ns        id
+    0                                       Gettysburg Address    0     21450
+    1                                                Main Page    0     42950
+    2                            Declaration by United Nations    0      8435
+    3             Constitution of the United States of America    0      8435
+    4                     Declaration of Independence (Israel)    0     17858
+    ...                                                    ...  ...       ...
+    3578760               Page:Black cat 1897 07 v2 n10.pdf/17  104    219649
+    3578761               Page:Black cat 1897 07 v2 n10.pdf/43  104    219649
+    3578762               Page:Black cat 1897 07 v2 n10.pdf/44  104    219649
+    3578763      The History of Tom Jones, a Foundling/Book IX    0  12084291
+    3578764  Page:Shakespeare of Stratford (1926) Yale.djvu/91  104     21450
+
+    [3578765 rows x 3 columns]
+
+
+.. _`lxml's iterparse`: https://lxml.de/3.2/parsing.html#iterparse-and-iterwalk
+.. _`etree's iterparse`: https://docs.python.org/3/library/xml.etree.elementtree.html#xml.etree.ElementTree.iterparse
+
+.. _whatsnew_150.enhancements.copy_on_write:
+
+Copy on Write
+^^^^^^^^^^^^^
+
+A new feature ``copy_on_write`` was added (:issue:`46958`). Copy on write ensures that
+any DataFrame or Series derived from another in any way always behaves as a copy.
+Copy on write disallows updating any other object than the object the method
+was applied to.
+
+Copy on write can be enabled through:
+
+.. code-block:: python
+
+    pd.set_option("mode.copy_on_write", True)
+    pd.options.mode.copy_on_write = True
+
+Alternatively, copy on write can be enabled locally through:
+
+.. code-block:: python
+
+    with pd.option_context("mode.copy_on_write", True):
+        ...
+
+Without copy on write, the parent :class:`DataFrame` is updated when updating a child
+:class:`DataFrame` that was derived from this :class:`DataFrame`.
+
+.. ipython:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": 1})
+    view = df["foo"]
+    view.iloc[0]
+    df
+
+With copy on write enabled, df won't be updated anymore:
+
+.. ipython:: python
+
+    with pd.option_context("mode.copy_on_write", True):
+        df = pd.DataFrame({"foo": [1, 2, 3], "bar": 1})
+        view = df["foo"]
+        view.iloc[0]
+        df
+
+A more detailed explanation can be found `here <https://phofl.github.io/cow-introduction.html>`_.
+
+.. _whatsnew_150.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- :meth:`Series.map` now raises when ``arg`` is dict but ``na_action`` is not either ``None`` or ``'ignore'`` (:issue:`46588`)
+- :meth:`MultiIndex.to_frame` now supports the argument ``allow_duplicates`` and raises on duplicate labels if it is missing or False (:issue:`45245`)
+- :class:`.StringArray` now accepts array-likes containing nan-likes (``None``, ``np.nan``) for the ``values`` parameter in its constructor in addition to strings and :attr:`pandas.NA`. (:issue:`40839`)
+- Improved the rendering of ``categories`` in :class:`CategoricalIndex` (:issue:`45218`)
+- :meth:`DataFrame.plot` will now allow the ``subplots`` parameter to be a list of iterables specifying column groups, so that columns may be grouped together in the same subplot (:issue:`29688`).
+- :meth:`to_numeric` now preserves float64 arrays when downcasting would generate values not representable in float32 (:issue:`43693`)
+- :meth:`Series.reset_index` and :meth:`DataFrame.reset_index` now support the argument ``allow_duplicates`` (:issue:`44410`)
+- :meth:`.DataFrameGroupBy.min`, :meth:`.SeriesGroupBy.min`, :meth:`.DataFrameGroupBy.max`, and :meth:`.SeriesGroupBy.max` now supports `Numba <https://numba.pydata.org/>`_ execution with the ``engine`` keyword (:issue:`45428`)
+- :func:`read_csv` now supports ``defaultdict`` as a ``dtype`` parameter (:issue:`41574`)
+- :meth:`DataFrame.rolling` and :meth:`Series.rolling` now support a ``step`` parameter with fixed-length windows (:issue:`15354`)
+- Implemented a ``bool``-dtype :class:`Index`, passing a bool-dtype array-like to ``pd.Index`` will now retain ``bool`` dtype instead of casting to ``object`` (:issue:`45061`)
+- Implemented a complex-dtype :class:`Index`, passing a complex-dtype array-like to ``pd.Index`` will now retain complex dtype instead of casting to ``object`` (:issue:`45845`)
+- :class:`Series` and :class:`DataFrame` with :class:`IntegerDtype` now supports bitwise operations (:issue:`34463`)
+- Add ``milliseconds`` field support for :class:`.DateOffset` (:issue:`43371`)
+- :meth:`DataFrame.where` tries to maintain dtype of :class:`DataFrame` if fill value can be cast without loss of precision (:issue:`45582`)
+- :meth:`DataFrame.reset_index` now accepts a ``names`` argument which renames the index names (:issue:`6878`)
+- :func:`concat` now raises when ``levels`` is given but ``keys`` is None (:issue:`46653`)
+- :func:`concat` now raises when ``levels`` contains duplicate values (:issue:`46653`)
+- Added ``numeric_only`` argument to :meth:`DataFrame.corr`, :meth:`DataFrame.corrwith`, :meth:`DataFrame.cov`, :meth:`DataFrame.idxmin`, :meth:`DataFrame.idxmax`, :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.DataFrameGroupBy.var`, :meth:`.SeriesGroupBy.var`, :meth:`.DataFrameGroupBy.std`, :meth:`.SeriesGroupBy.std`, :meth:`.DataFrameGroupBy.sem`, :meth:`.SeriesGroupBy.sem`, and :meth:`.DataFrameGroupBy.quantile` (:issue:`46560`)
+- A :class:`errors.PerformanceWarning` is now thrown when using ``string[pyarrow]`` dtype with methods that don't dispatch to ``pyarrow.compute`` methods (:issue:`42613`, :issue:`46725`)
+- Added ``validate`` argument to :meth:`DataFrame.join` (:issue:`46622`)
+- Added ``numeric_only`` argument to :meth:`.Resampler.sum`, :meth:`.Resampler.prod`, :meth:`.Resampler.min`, :meth:`.Resampler.max`, :meth:`.Resampler.first`, and :meth:`.Resampler.last` (:issue:`46442`)
+- ``times`` argument in :class:`.ExponentialMovingWindow` now accepts ``np.timedelta64`` (:issue:`47003`)
+- :class:`.DataError`, :class:`.SpecificationError`, ``SettingWithCopyError``, ``SettingWithCopyWarning``, :class:`.NumExprClobberingError`, :class:`.UndefinedVariableError`, :class:`.IndexingError`, :class:`.PyperclipException`, :class:`.PyperclipWindowsException`, :class:`.CSSWarning`, :class:`.PossibleDataLossError`, :class:`.ClosedFileError`, :class:`.IncompatibilityWarning`, :class:`.AttributeConflictWarning`, :class:`.DatabaseError`, :class:`.PossiblePrecisionLoss`, :class:`.ValueLabelTypeMismatch`, :class:`.InvalidColumnName`, and :class:`.CategoricalConversionWarning` are now exposed in ``pandas.errors`` (:issue:`27656`)
+- Added ``check_like`` argument to :func:`testing.assert_series_equal` (:issue:`47247`)
+- Add support for :meth:`.DataFrameGroupBy.ohlc` and :meth:`.SeriesGroupBy.ohlc` for extension array dtypes (:issue:`37493`)
+- Allow reading compressed SAS files with :func:`read_sas` (e.g., ``.sas7bdat.gz`` files)
+- :func:`pandas.read_html` now supports extracting links from table cells (:issue:`13141`)
+- :meth:`DatetimeIndex.astype` now supports casting timezone-naive indexes to ``datetime64[s]``, ``datetime64[ms]``, and ``datetime64[us]``, and timezone-aware indexes to the corresponding ``datetime64[unit, tzname]`` dtypes (:issue:`47579`)
+- :class:`Series` reducers (e.g. ``min``, ``max``, ``sum``, ``mean``) will now successfully operate when the dtype is numeric and ``numeric_only=True`` is provided; previously this would raise a ``NotImplementedError`` (:issue:`47500`)
+- :meth:`RangeIndex.union` now can return a :class:`RangeIndex` instead of a :class:`Int64Index` if the resulting values are equally spaced (:issue:`47557`, :issue:`43885`)
+- :meth:`DataFrame.compare` now accepts an argument ``result_names`` to allow the user to specify the result's names of both left and right DataFrame which are being compared. This is by default ``'self'`` and ``'other'`` (:issue:`44354`)
+- :meth:`DataFrame.quantile` gained a ``method`` argument that can accept ``table`` to evaluate multi-column quantiles (:issue:`43881`)
+- :class:`Interval` now supports checking whether one interval is contained by another interval (:issue:`46613`)
+- Added ``copy`` keyword to :meth:`Series.set_axis` and :meth:`DataFrame.set_axis` to allow user to set axis on a new object without necessarily copying the underlying data (:issue:`47932`)
+- The method :meth:`.ExtensionArray.factorize` accepts ``use_na_sentinel=False`` for determining how null values are to be treated (:issue:`46601`)
+- The ``Dockerfile`` now installs a dedicated ``pandas-dev`` virtual environment for pandas development instead of using the ``base`` environment (:issue:`48427`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_150.notable_bug_fixes.groupby_transform_dropna:
+
+Using ``dropna=True`` with ``groupby`` transforms
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A transform is an operation whose result has the same size as its input. When the
+result is a :class:`DataFrame` or :class:`Series`, it is also required that the
+index of the result matches that of the input. In pandas 1.4, using
+:meth:`.DataFrameGroupBy.transform` or :meth:`.SeriesGroupBy.transform` with null
+values in the groups and ``dropna=True`` gave incorrect results. Demonstrated by the
+examples below, the incorrect results either contained incorrect values, or the result
+did not have the same index as the input.
+
+.. ipython:: python
+
+    df = pd.DataFrame({'a': [1, 1, np.nan], 'b': [2, 3, 4]})
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+    In [3]: # Value in the last row should be np.nan
+            df.groupby('a', dropna=True).transform('sum')
+    Out[3]:
+       b
+    0  5
+    1  5
+    2  5
+
+    In [3]: # Should have one additional row with the value np.nan
+            df.groupby('a', dropna=True).transform(lambda x: x.sum())
+    Out[3]:
+       b
+    0  5
+    1  5
+
+    In [3]: # The value in the last row is np.nan interpreted as an integer
+            df.groupby('a', dropna=True).transform('ffill')
+    Out[3]:
+                         b
+    0                    2
+    1                    3
+    2 -9223372036854775808
+
+    In [3]: # Should have one additional row with the value np.nan
+            df.groupby('a', dropna=True).transform(lambda x: x)
+    Out[3]:
+       b
+    0  2
+    1  3
+
+*New behavior*:
+
+.. ipython:: python
+
+    df.groupby('a', dropna=True).transform('sum')
+    df.groupby('a', dropna=True).transform(lambda x: x.sum())
+    df.groupby('a', dropna=True).transform('ffill')
+    df.groupby('a', dropna=True).transform(lambda x: x)
+
+.. _whatsnew_150.notable_bug_fixes.to_json_incorrectly_localizing_naive_timestamps:
+
+Serializing tz-naive Timestamps with to_json() with ``iso_dates=True``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:meth:`DataFrame.to_json`, :meth:`Series.to_json`, and :meth:`Index.to_json`
+would incorrectly localize DatetimeArrays/DatetimeIndexes with tz-naive Timestamps
+to UTC. (:issue:`38760`)
+
+Note that this patch does not fix the localization of tz-aware Timestamps to UTC
+upon serialization. (Related issue :issue:`12997`)
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [32]: index = pd.date_range(
+       ....:     start='2020-12-28 00:00:00',
+       ....:     end='2020-12-28 02:00:00',
+       ....:     freq='1H',
+       ....: )
+       ....:
+
+    In [33]: a = pd.Series(
+       ....:     data=range(3),
+       ....:     index=index,
+       ....: )
+       ....:
+
+    In [4]: from io import StringIO
+
+    In [5]: a.to_json(date_format='iso')
+    Out[5]: '{"2020-12-28T00:00:00.000Z":0,"2020-12-28T01:00:00.000Z":1,"2020-12-28T02:00:00.000Z":2}'
+
+    In [6]: pd.read_json(StringIO(a.to_json(date_format='iso')), typ="series").index == a.index
+    Out[6]: array([False, False, False])
+
+*New Behavior*
+
+.. code-block:: ipython
+
+    In [34]: from io import StringIO
+
+    In [35]: a.to_json(date_format='iso')
+    Out[35]: '{"2020-12-28T00:00:00.000Z":0,"2020-12-28T01:00:00.000Z":1,"2020-12-28T02:00:00.000Z":2}'
+
+    # Roundtripping now works
+    In [36]: pd.read_json(StringIO(a.to_json(date_format='iso')), typ="series").index == a.index
+    Out[36]: array([ True,  True,  True])
+
+.. _whatsnew_150.notable_bug_fixes.groupby_value_counts_categorical:
+
+DataFrameGroupBy.value_counts with non-grouping categorical columns and ``observed=True``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Calling :meth:`.DataFrameGroupBy.value_counts` with ``observed=True`` would incorrectly drop non-observed categories of non-grouping columns (:issue:`46357`).
+
+.. code-block:: ipython
+
+    In [6]: df = pd.DataFrame(["a", "b", "c"], dtype="category").iloc[0:2]
+    In [7]: df
+    Out[7]:
+       0
+    0  a
+    1  b
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [8]: df.groupby(level=0, observed=True).value_counts()
+    Out[8]:
+    0  a    1
+    1  b    1
+    dtype: int64
+
+
+*New Behavior*
+
+.. code-block:: ipython
+
+    In [9]: df.groupby(level=0, observed=True).value_counts()
+    Out[9]:
+    0  a    1
+    1  a    0
+       b    1
+    0  b    0
+       c    0
+    1  c    0
+    dtype: int64
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_150.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.20.3          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+| mypy (dev)      | 0.971           |          |    X    |
++-----------------+-----------------+----------+---------+
+| beautifulsoup4  | 4.9.3           |          |    X    |
++-----------------+-----------------+----------+---------+
+| blosc           | 1.21.0          |          |    X    |
++-----------------+-----------------+----------+---------+
+| bottleneck      | 1.3.2           |          |    X    |
++-----------------+-----------------+----------+---------+
+| fsspec          | 2021.07.0       |          |    X    |
++-----------------+-----------------+----------+---------+
+| hypothesis      | 6.13.0          |          |    X    |
++-----------------+-----------------+----------+---------+
+| gcsfs           | 2021.07.0       |          |    X    |
++-----------------+-----------------+----------+---------+
+| jinja2          | 3.0.0           |          |    X    |
++-----------------+-----------------+----------+---------+
+| lxml            | 4.6.3           |          |    X    |
++-----------------+-----------------+----------+---------+
+| numba           | 0.53.1          |          |    X    |
++-----------------+-----------------+----------+---------+
+| numexpr         | 2.7.3           |          |    X    |
++-----------------+-----------------+----------+---------+
+| openpyxl        | 3.0.7           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pandas-gbq      | 0.15.0          |          |    X    |
++-----------------+-----------------+----------+---------+
+| psycopg2        | 2.8.6           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pymysql         | 1.0.2           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pyreadstat      | 1.1.2           |          |    X    |
++-----------------+-----------------+----------+---------+
+| pyxlsb          | 1.0.8           |          |    X    |
++-----------------+-----------------+----------+---------+
+| s3fs            | 2021.08.0       |          |    X    |
++-----------------+-----------------+----------+---------+
+| scipy           | 1.7.1           |          |    X    |
++-----------------+-----------------+----------+---------+
+| sqlalchemy      | 1.4.16          |          |    X    |
++-----------------+-----------------+----------+---------+
+| tabulate        | 0.8.9           |          |    X    |
++-----------------+-----------------+----------+---------+
+| xarray          | 0.19.0          |          |    X    |
++-----------------+-----------------+----------+---------+
+| xlsxwriter      | 1.4.3           |          |    X    |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| beautifulsoup4  |4.9.3            |    X    |
++-----------------+-----------------+---------+
+| blosc           |1.21.0           |    X    |
++-----------------+-----------------+---------+
+| bottleneck      |1.3.2            |    X    |
++-----------------+-----------------+---------+
+| brotlipy        |0.7.0            |         |
++-----------------+-----------------+---------+
+| fastparquet     |0.4.0            |         |
++-----------------+-----------------+---------+
+| fsspec          |2021.08.0        |    X    |
++-----------------+-----------------+---------+
+| html5lib        |1.1              |         |
++-----------------+-----------------+---------+
+| hypothesis      |6.13.0           |    X    |
++-----------------+-----------------+---------+
+| gcsfs           |2021.08.0        |    X    |
++-----------------+-----------------+---------+
+| jinja2          |3.0.0            |    X    |
++-----------------+-----------------+---------+
+| lxml            |4.6.3            |    X    |
++-----------------+-----------------+---------+
+| matplotlib      |3.3.2            |         |
++-----------------+-----------------+---------+
+| numba           |0.53.1           |    X    |
++-----------------+-----------------+---------+
+| numexpr         |2.7.3            |    X    |
++-----------------+-----------------+---------+
+| odfpy           |1.4.1            |         |
++-----------------+-----------------+---------+
+| openpyxl        |3.0.7            |    X    |
++-----------------+-----------------+---------+
+| pandas-gbq      |0.15.0           |    X    |
++-----------------+-----------------+---------+
+| psycopg2        |2.8.6            |    X    |
++-----------------+-----------------+---------+
+| pyarrow         |1.0.1            |         |
++-----------------+-----------------+---------+
+| pymysql         |1.0.2            |    X    |
++-----------------+-----------------+---------+
+| pyreadstat      |1.1.2            |    X    |
++-----------------+-----------------+---------+
+| pytables        |3.6.1            |         |
++-----------------+-----------------+---------+
+| python-snappy   |0.6.0            |         |
++-----------------+-----------------+---------+
+| pyxlsb          |1.0.8            |    X    |
++-----------------+-----------------+---------+
+| s3fs            |2021.08.0        |    X    |
++-----------------+-----------------+---------+
+| scipy           |1.7.1            |    X    |
++-----------------+-----------------+---------+
+| sqlalchemy      |1.4.16           |    X    |
++-----------------+-----------------+---------+
+| tabulate        |0.8.9            |    X    |
++-----------------+-----------------+---------+
+| tzdata          |2022a            |         |
++-----------------+-----------------+---------+
+| xarray          |0.19.0           |    X    |
++-----------------+-----------------+---------+
+| xlrd            |2.0.1            |         |
++-----------------+-----------------+---------+
+| xlsxwriter      |1.4.3            |    X    |
++-----------------+-----------------+---------+
+| xlwt            |1.3.0            |         |
++-----------------+-----------------+---------+
+| zstandard       |0.15.2           |         |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_150.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+
+- BigQuery I/O methods :func:`read_gbq` and :meth:`DataFrame.to_gbq` default to
+  ``auth_local_webserver = True``. Google has deprecated the
+  ``auth_local_webserver = False`` `"out of band" (copy-paste) flow
+  <https://developers.googleblog.com/2022/02/making-oauth-flows-safer.html?m=1#disallowed-oob>`_.
+  The ``auth_local_webserver = False`` option is planned to stop working in
+  October 2022. (:issue:`46312`)
+- :func:`read_json` now raises ``FileNotFoundError`` (previously ``ValueError``) when input is a string ending in ``.json``, ``.json.gz``, ``.json.bz2``, etc. but no such file exists. (:issue:`29102`)
+- Operations with :class:`Timestamp` or :class:`Timedelta` that would previously raise ``OverflowError`` instead raise ``OutOfBoundsDatetime`` or ``OutOfBoundsTimedelta`` where appropriate (:issue:`47268`)
+- When :func:`read_sas` previously returned ``None``, it now returns an empty :class:`DataFrame` (:issue:`47410`)
+- :class:`DataFrame` constructor raises if ``index`` or ``columns`` arguments are sets (:issue:`47215`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+.. warning::
+
+    In the next major version release, 2.0, several larger API changes are being considered without a formal deprecation such as
+    making the standard library `zoneinfo <https://docs.python.org/3/library/zoneinfo.html>`_ the default timezone implementation instead of ``pytz``,
+    having the :class:`Index` support all data types instead of having multiple subclasses (:class:`CategoricalIndex`, :class:`Int64Index`, etc.), and more.
+    The changes under consideration are logged in `this GitHub issue <https://github.com/pandas-dev/pandas/issues/44823>`_, and any
+    feedback or concerns are welcome.
+
+.. _whatsnew_150.deprecations.int_slicing_series:
+
+Label-based integer slicing on a Series with an Int64Index or RangeIndex
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In a future version, integer slicing on a :class:`Series` with a :class:`Int64Index` or :class:`RangeIndex` will be treated as *label-based*, not positional. This will make the behavior consistent with other :meth:`Series.__getitem__` and :meth:`Series.__setitem__` behaviors (:issue:`45162`).
+
+For example:
+
+.. ipython:: python
+
+   ser = pd.Series([1, 2, 3, 4, 5], index=[2, 3, 5, 7, 11])
+
+In the old behavior, ``ser[2:4]`` treats the slice as positional:
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+    In [3]: ser[2:4]
+    Out[3]:
+    5    3
+    7    4
+    dtype: int64
+
+In a future version, this will be treated as label-based:
+
+*Future behavior*:
+
+.. code-block:: ipython
+
+    In [4]: ser.loc[2:4]
+    Out[4]:
+    2    1
+    3    2
+    dtype: int64
+
+To retain the old behavior, use ``series.iloc[i:j]``. To get the future behavior,
+use ``series.loc[i:j]``.
+
+Slicing on a :class:`DataFrame` will not be affected.
+
+.. _whatsnew_150.deprecations.excel_writer_attributes:
+
+:class:`ExcelWriter` attributes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+All attributes of :class:`ExcelWriter` were previously documented as not
+public. However some third party Excel engines documented accessing
+``ExcelWriter.book`` or ``ExcelWriter.sheets``, and users were utilizing these
+and possibly other attributes. Previously these attributes were not safe to use;
+e.g. modifications to ``ExcelWriter.book`` would not update ``ExcelWriter.sheets``
+and conversely. In order to support this, pandas has made some attributes public
+and improved their implementations so that they may now be safely used. (:issue:`45572`)
+
+The following attributes are now public and considered safe to access.
+
+ - ``book``
+ - ``check_extension``
+ - ``close``
+ - ``date_format``
+ - ``datetime_format``
+ - ``engine``
+ - ``if_sheet_exists``
+ - ``sheets``
+ - ``supported_extensions``
+
+The following attributes have been deprecated. They now raise a ``FutureWarning``
+when accessed and will be removed in a future version. Users should be aware
+that their usage is considered unsafe, and can lead to unexpected results.
+
+ - ``cur_sheet``
+ - ``handles``
+ - ``path``
+ - ``save``
+ - ``write_cells``
+
+See the documentation of :class:`ExcelWriter` for further details.
+
+.. _whatsnew_150.deprecations.group_keys_in_apply:
+
+Using ``group_keys`` with transformers in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, if it was inferred that the function passed to
+:meth:`.DataFrameGroupBy.apply` or :meth:`.SeriesGroupBy.apply` was a transformer (i.e. the resulting index was equal to
+the input index), the ``group_keys`` argument of :meth:`DataFrame.groupby` and
+:meth:`Series.groupby` was ignored and the group keys would never be added to
+the index of the result. In the future, the group keys will be added to the index
+when the user specifies ``group_keys=True``.
+
+As ``group_keys=True`` is the default value of :meth:`DataFrame.groupby` and
+:meth:`Series.groupby`, not specifying ``group_keys`` with a transformer will
+raise a ``FutureWarning``. This can be silenced and the previous behavior
+retained by specifying ``group_keys=False``.
+
+.. _whatsnew_150.deprecations.setitem_column_try_inplace:
+   _ see also _whatsnew_130.notable_bug_fixes.setitem_column_try_inplace
+
+Inplace operation when setting values with ``loc`` and ``iloc``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Most of the time setting values with :meth:`DataFrame.iloc` attempts to set values
+inplace, only falling back to inserting a new array if necessary. There are
+some cases where this rule is not followed, for example when setting an entire
+column from an array with different dtype:
+
+.. ipython:: python
+
+   df = pd.DataFrame({'price': [11.1, 12.2]}, index=['book1', 'book2'])
+   original_prices = df['price']
+   new_prices = np.array([98, 99])
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.iloc[:, 0] = new_prices
+    In [4]: df.iloc[:, 0]
+    Out[4]:
+    book1    98
+    book2    99
+    Name: price, dtype: int64
+    In [5]: original_prices
+    Out[5]:
+    book1    11.1
+    book2    12.2
+    Name: price, float: 64
+
+This behavior is deprecated. In a future version, setting an entire column with
+iloc will attempt to operate inplace.
+
+*Future behavior*:
+
+.. code-block:: ipython
+
+    In [3]: df.iloc[:, 0] = new_prices
+    In [4]: df.iloc[:, 0]
+    Out[4]:
+    book1    98.0
+    book2    99.0
+    Name: price, dtype: float64
+    In [5]: original_prices
+    Out[5]:
+    book1    98.0
+    book2    99.0
+    Name: price, dtype: float64
+
+To get the old behavior, use :meth:`DataFrame.__setitem__` directly:
+
+.. code-block:: ipython
+
+    In [3]: df[df.columns[0]] = new_prices
+    In [4]: df.iloc[:, 0]
+    Out[4]
+    book1    98
+    book2    99
+    Name: price, dtype: int64
+    In [5]: original_prices
+    Out[5]:
+    book1    11.1
+    book2    12.2
+    Name: price, dtype: float64
+
+To get the old behaviour when ``df.columns`` is not unique and you want to
+change a single column by index, you can use :meth:`DataFrame.isetitem`, which
+has been added in pandas 1.5:
+
+.. code-block:: ipython
+
+    In [3]: df_with_duplicated_cols = pd.concat([df, df], axis='columns')
+    In [3]: df_with_duplicated_cols.isetitem(0, new_prices)
+    In [4]: df_with_duplicated_cols.iloc[:, 0]
+    Out[4]:
+    book1    98
+    book2    99
+    Name: price, dtype: int64
+    In [5]: original_prices
+    Out[5]:
+    book1    11.1
+    book2    12.2
+    Name: 0, dtype: float64
+
+.. _whatsnew_150.deprecations.numeric_only_default:
+
+``numeric_only`` default value
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Across the :class:`DataFrame`, :class:`.DataFrameGroupBy`, and :class:`.Resampler` operations such as
+``min``, ``sum``, and ``idxmax``, the default
+value of the ``numeric_only`` argument, if it exists at all, was inconsistent.
+Furthermore, operations with the default value ``None`` can lead to surprising
+results. (:issue:`46560`)
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"a": [1, 2], "b": ["x", "y"]})
+
+    In [2]: # Reading the next line without knowing the contents of df, one would
+            # expect the result to contain the products for both columns a and b.
+            df[["a", "b"]].prod()
+    Out[2]:
+    a    2
+    dtype: int64
+
+To avoid this behavior, the specifying the value ``numeric_only=None`` has been
+deprecated, and will be removed in a future version of pandas. In the future,
+all operations with a ``numeric_only`` argument will default to ``False``. Users
+should either call the operation only with columns that can be operated on, or
+specify ``numeric_only=True`` to operate only on Boolean, integer, and float columns.
+
+In order to support the transition to the new behavior, the following methods have
+gained the ``numeric_only`` argument.
+
+- :meth:`DataFrame.corr`
+- :meth:`DataFrame.corrwith`
+- :meth:`DataFrame.cov`
+- :meth:`DataFrame.idxmin`
+- :meth:`DataFrame.idxmax`
+- :meth:`.DataFrameGroupBy.cummin`
+- :meth:`.DataFrameGroupBy.cummax`
+- :meth:`.DataFrameGroupBy.idxmin`
+- :meth:`.DataFrameGroupBy.idxmax`
+- :meth:`.DataFrameGroupBy.var`
+- :meth:`.DataFrameGroupBy.std`
+- :meth:`.DataFrameGroupBy.sem`
+- :meth:`.DataFrameGroupBy.quantile`
+- :meth:`.Resampler.mean`
+- :meth:`.Resampler.median`
+- :meth:`.Resampler.sem`
+- :meth:`.Resampler.std`
+- :meth:`.Resampler.var`
+- :meth:`DataFrame.rolling` operations
+- :meth:`DataFrame.expanding` operations
+- :meth:`DataFrame.ewm` operations
+
+.. _whatsnew_150.deprecations.other:
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+- Deprecated the keyword ``line_terminator`` in :meth:`DataFrame.to_csv` and :meth:`Series.to_csv`, use ``lineterminator`` instead; this is for consistency with :func:`read_csv` and the standard library 'csv' module (:issue:`9568`)
+- Deprecated behavior of :meth:`SparseArray.astype`, :meth:`Series.astype`, and :meth:`DataFrame.astype` with :class:`SparseDtype` when passing a non-sparse ``dtype``. In a future version, this will cast to that non-sparse dtype instead of wrapping it in a :class:`SparseDtype` (:issue:`34457`)
+- Deprecated behavior of :meth:`DatetimeIndex.intersection` and :meth:`DatetimeIndex.symmetric_difference` (``union`` behavior was already deprecated in version 1.3.0) with mixed time zones; in a future version both will be cast to UTC instead of object dtype (:issue:`39328`, :issue:`45357`)
+- Deprecated :meth:`DataFrame.iteritems`, :meth:`Series.iteritems`, :meth:`HDFStore.iteritems` in favor of :meth:`DataFrame.items`, :meth:`Series.items`, :meth:`HDFStore.items`  (:issue:`45321`)
+- Deprecated :meth:`Series.is_monotonic` and :meth:`Index.is_monotonic` in favor of :meth:`Series.is_monotonic_increasing` and :meth:`Index.is_monotonic_increasing` (:issue:`45422`, :issue:`21335`)
+- Deprecated behavior of :meth:`DatetimeIndex.astype`, :meth:`TimedeltaIndex.astype`, :meth:`PeriodIndex.astype` when converting to an integer dtype other than ``int64``. In a future version, these will convert to exactly the specified dtype (instead of always ``int64``) and will raise if the conversion overflows (:issue:`45034`)
+- Deprecated the ``__array_wrap__`` method of DataFrame and Series, rely on standard numpy ufuncs instead (:issue:`45451`)
+- Deprecated treating float-dtype data as wall-times when passed with a timezone to :class:`Series` or :class:`DatetimeIndex` (:issue:`45573`)
+- Deprecated the behavior of :meth:`Series.fillna` and :meth:`DataFrame.fillna` with ``timedelta64[ns]`` dtype and incompatible fill value; in a future version this will cast to a common dtype (usually object) instead of raising, matching the behavior of other dtypes (:issue:`45746`)
+- Deprecated the ``warn`` parameter in :func:`infer_freq` (:issue:`45947`)
+- Deprecated allowing non-keyword arguments in :meth:`.ExtensionArray.argsort` (:issue:`46134`)
+- Deprecated treating all-bool ``object``-dtype columns as bool-like in :meth:`DataFrame.any` and :meth:`DataFrame.all` with ``bool_only=True``, explicitly cast to bool instead (:issue:`46188`)
+- Deprecated behavior of method :meth:`DataFrame.quantile`, attribute ``numeric_only`` will default False. Including datetime/timedelta columns in the result (:issue:`7308`).
+- Deprecated :attr:`Timedelta.freq` and :attr:`Timedelta.is_populated` (:issue:`46430`)
+- Deprecated :attr:`Timedelta.delta` (:issue:`46476`)
+- Deprecated passing arguments as positional in :meth:`DataFrame.any` and :meth:`Series.any` (:issue:`44802`)
+- Deprecated passing positional arguments to :meth:`DataFrame.pivot` and :func:`pivot` except ``data`` (:issue:`30228`)
+- Deprecated the methods :meth:`DataFrame.mad`, :meth:`Series.mad`, and the corresponding groupby methods (:issue:`11787`)
+- Deprecated positional arguments to :meth:`Index.join` except for ``other``, use keyword-only arguments instead of positional arguments (:issue:`46518`)
+- Deprecated positional arguments to :meth:`StringMethods.rsplit` and :meth:`StringMethods.split` except for ``pat``, use keyword-only arguments instead of positional arguments (:issue:`47423`)
+- Deprecated indexing on a timezone-naive :class:`DatetimeIndex` using a string representing a timezone-aware datetime (:issue:`46903`, :issue:`36148`)
+- Deprecated allowing ``unit="M"`` or ``unit="Y"`` in :class:`Timestamp` constructor with a non-round float value (:issue:`47267`)
+- Deprecated the ``display.column_space`` global configuration option (:issue:`7576`)
+- Deprecated the argument ``na_sentinel`` in :func:`factorize`, :meth:`Index.factorize`, and :meth:`.ExtensionArray.factorize`; pass ``use_na_sentinel=True`` instead to use the sentinel ``-1`` for NaN values and ``use_na_sentinel=False`` instead of ``na_sentinel=None`` to encode NaN values (:issue:`46910`)
+- Deprecated :meth:`.DataFrameGroupBy.transform` not aligning the result when the UDF returned DataFrame (:issue:`45648`)
+- Clarified warning from :func:`to_datetime` when delimited dates can't be parsed in accordance to specified ``dayfirst`` argument (:issue:`46210`)
+- Emit warning from :func:`to_datetime` when delimited dates can't be parsed in accordance to specified ``dayfirst`` argument even for dates where leading zero is omitted (e.g. ``31/1/2001``) (:issue:`47880`)
+- Deprecated :class:`Series` and :class:`Resampler` reducers (e.g. ``min``, ``max``, ``sum``, ``mean``) raising a ``NotImplementedError`` when the dtype is non-numric and ``numeric_only=True`` is provided; this will raise a ``TypeError`` in a future version (:issue:`47500`)
+- Deprecated :meth:`Series.rank` returning an empty result when the dtype is non-numeric and ``numeric_only=True`` is provided; this will raise a ``TypeError`` in a future version (:issue:`47500`)
+- Deprecated argument ``errors`` for :meth:`Series.mask`, :meth:`Series.where`, :meth:`DataFrame.mask`, and :meth:`DataFrame.where` as ``errors`` had no effect on this methods (:issue:`47728`)
+- Deprecated arguments ``*args`` and ``**kwargs`` in :class:`Rolling`, :class:`Expanding`, and :class:`ExponentialMovingWindow` ops. (:issue:`47836`)
+- Deprecated the ``inplace`` keyword in :meth:`Categorical.set_ordered`, :meth:`Categorical.as_ordered`, and :meth:`Categorical.as_unordered` (:issue:`37643`)
+- Deprecated setting a categorical's categories with ``cat.categories = ['a', 'b', 'c']``, use :meth:`Categorical.rename_categories` instead (:issue:`37643`)
+- Deprecated unused arguments ``encoding`` and ``verbose`` in :meth:`Series.to_excel` and :meth:`DataFrame.to_excel` (:issue:`47912`)
+- Deprecated the ``inplace`` keyword in :meth:`DataFrame.set_axis` and :meth:`Series.set_axis`, use ``obj = obj.set_axis(..., copy=False)`` instead (:issue:`48130`)
+- Deprecated producing a single element when iterating over a :class:`DataFrameGroupBy` or a :class:`SeriesGroupBy` that has been grouped by a list of length 1; A tuple of length one will be returned instead (:issue:`42795`)
+- Fixed up warning message of deprecation of :meth:`MultiIndex.lesort_depth` as public method, as the message previously referred to :meth:`MultiIndex.is_lexsorted` instead (:issue:`38701`)
+- Deprecated the ``sort_columns`` argument in :meth:`DataFrame.plot` and :meth:`Series.plot` (:issue:`47563`).
+- Deprecated positional arguments for all but the first argument of :meth:`DataFrame.to_stata` and :func:`read_stata`, use keyword arguments instead (:issue:`48128`).
+- Deprecated the ``mangle_dupe_cols`` argument in :func:`read_csv`, :func:`read_fwf`, :func:`read_table` and :func:`read_excel`. The argument was never implemented, and a new argument where the renaming pattern can be specified will be added instead (:issue:`47718`)
+- Deprecated allowing ``dtype='datetime64'`` or ``dtype=np.datetime64`` in :meth:`Series.astype`, use "datetime64[ns]" instead (:issue:`47844`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :meth:`DataFrame.corrwith` for column-wise (axis=0) Pearson and Spearman correlation when other is a :class:`Series` (:issue:`46174`)
+- Performance improvement in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` for some user-defined DataFrame -> Series functions (:issue:`45387`)
+- Performance improvement in :meth:`DataFrame.duplicated` when subset consists of only one column (:issue:`45236`)
+- Performance improvement in :meth:`.DataFrameGroupBy.diff` and :meth:`.SeriesGroupBy.diff` (:issue:`16706`)
+- Performance improvement in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` when broadcasting values for user-defined functions (:issue:`45708`)
+- Performance improvement in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` for user-defined functions when only a single group exists (:issue:`44977`)
+- Performance improvement in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` when grouping on a non-unique unsorted index (:issue:`46527`)
+- Performance improvement in :meth:`DataFrame.loc` and :meth:`Series.loc` for tuple-based indexing of a :class:`MultiIndex` (:issue:`45681`, :issue:`46040`, :issue:`46330`)
+- Performance improvement in :meth:`.DataFrameGroupBy.var` and :meth:`.SeriesGroupBy.var` with ``ddof`` other than one (:issue:`48152`)
+- Performance improvement in :meth:`DataFrame.to_records` when the index is a :class:`MultiIndex` (:issue:`47263`)
+- Performance improvement in :attr:`MultiIndex.values` when the MultiIndex contains levels of type DatetimeIndex, TimedeltaIndex or ExtensionDtypes (:issue:`46288`)
+- Performance improvement in :func:`merge` when left and/or right are empty (:issue:`45838`)
+- Performance improvement in :meth:`DataFrame.join` when left and/or right are empty (:issue:`46015`)
+- Performance improvement in :meth:`DataFrame.reindex` and :meth:`Series.reindex` when target is a :class:`MultiIndex` (:issue:`46235`)
+- Performance improvement when setting values in a pyarrow backed string array (:issue:`46400`)
+- Performance improvement in :func:`factorize` (:issue:`46109`)
+- Performance improvement in :class:`DataFrame` and :class:`Series` constructors for extension dtype scalars (:issue:`45854`)
+- Performance improvement in :func:`read_excel` when ``nrows`` argument provided (:issue:`32727`)
+- Performance improvement in :meth:`.Styler.to_excel` when applying repeated CSS formats (:issue:`47371`)
+- Performance improvement in :meth:`MultiIndex.is_monotonic_increasing`  (:issue:`47458`)
+- Performance improvement in :class:`BusinessHour` ``str`` and ``repr`` (:issue:`44764`)
+- Performance improvement in datetime arrays string formatting when one of the default strftime formats ``"%Y-%m-%d %H:%M:%S"`` or ``"%Y-%m-%d %H:%M:%S.%f"`` is used. (:issue:`44764`)
+- Performance improvement in :meth:`Series.to_sql` and :meth:`DataFrame.to_sql` (:class:`SQLiteTable`) when processing time arrays. (:issue:`44764`)
+- Performance improvement to :func:`read_sas` (:issue:`47404`)
+- Performance improvement in ``argmax`` and ``argmin`` for :class:`arrays.SparseArray` (:issue:`34197`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in :meth:`.Categorical.view` not accepting integer dtypes (:issue:`25464`)
+- Bug in :meth:`.CategoricalIndex.union` when the index's categories are integer-dtype and the index contains ``NaN`` values incorrectly raising instead of casting to ``float64`` (:issue:`45362`)
+- Bug in :meth:`concat` when concatenating two (or more) unordered :class:`CategoricalIndex` variables, whose categories are permutations, yields incorrect index values (:issue:`24845`)
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :meth:`DataFrame.quantile` with datetime-like dtypes and no rows incorrectly returning ``float64`` dtype instead of retaining datetime-like dtype (:issue:`41544`)
+- Bug in :func:`to_datetime` with sequences of ``np.str_`` objects incorrectly raising (:issue:`32264`)
+- Bug in :class:`Timestamp` construction when passing datetime components as positional arguments and ``tzinfo`` as a keyword argument incorrectly raising (:issue:`31929`)
+- Bug in :meth:`Index.astype` when casting from object dtype to ``timedelta64[ns]`` dtype incorrectly casting ``np.datetime64("NaT")`` values to ``np.timedelta64("NaT")`` instead of raising (:issue:`45722`)
+- Bug in :meth:`.SeriesGroupBy.value_counts` index when passing categorical column (:issue:`44324`)
+- Bug in :meth:`DatetimeIndex.tz_localize` localizing to UTC failing to make a copy of the underlying data (:issue:`46460`)
+- Bug in :meth:`DatetimeIndex.resolution` incorrectly returning "day" instead of "nanosecond" for nanosecond-resolution indexes (:issue:`46903`)
+- Bug in :class:`Timestamp` with an integer or float value and ``unit="Y"`` or ``unit="M"`` giving slightly-wrong results (:issue:`47266`)
+- Bug in :class:`.DatetimeArray` construction when passed another :class:`.DatetimeArray` and ``freq=None`` incorrectly inferring the freq from the given array (:issue:`47296`)
+- Bug in :func:`to_datetime` where ``OutOfBoundsDatetime`` would be thrown even if ``errors=coerce`` if there were more than 50 rows (:issue:`45319`)
+- Bug when adding a :class:`DateOffset` to a :class:`Series` would not add the ``nanoseconds`` field (:issue:`47856`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :func:`astype_nansafe` astype("timedelta64[ns]") fails when np.nan is included (:issue:`45798`)
+- Bug in constructing a :class:`Timedelta` with a ``np.timedelta64`` object and a ``unit`` sometimes silently overflowing and returning incorrect results instead of raising ``OutOfBoundsTimedelta`` (:issue:`46827`)
+- Bug in constructing a :class:`Timedelta` from a large integer or float with ``unit="W"`` silently overflowing and returning incorrect results instead of raising ``OutOfBoundsTimedelta`` (:issue:`47268`)
+
+Time Zones
+^^^^^^^^^^
+- Bug in :class:`Timestamp` constructor raising when passed a ``ZoneInfo`` tzinfo object (:issue:`46425`)
+
+Numeric
+^^^^^^^
+- Bug in operations with array-likes with ``dtype="boolean"`` and :attr:`NA` incorrectly altering the array in-place (:issue:`45421`)
+- Bug in arithmetic operations with nullable types without :attr:`NA` values not matching the same operation with non-nullable types (:issue:`48223`)
+- Bug in ``floordiv`` when dividing by ``IntegerDtype`` ``0`` would return ``0`` instead of ``inf`` (:issue:`48223`)
+- Bug in division, ``pow`` and ``mod`` operations on array-likes with ``dtype="boolean"`` not being like their ``np.bool_`` counterparts (:issue:`46063`)
+- Bug in multiplying a :class:`Series` with ``IntegerDtype`` or ``FloatingDtype`` by an array-like with ``timedelta64[ns]`` dtype incorrectly raising (:issue:`45622`)
+- Bug in :meth:`mean` where the optional dependency ``bottleneck`` causes precision loss linear in the length of the array. ``bottleneck`` has been disabled for :meth:`mean` improving the loss to log-linear but may result in a performance decrease. (:issue:`42878`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :meth:`DataFrame.astype` not preserving subclasses (:issue:`40810`)
+- Bug in constructing a :class:`Series` from a float-containing list or a floating-dtype ndarray-like (e.g. ``dask.Array``) and an integer dtype raising instead of casting like we would with an ``np.ndarray`` (:issue:`40110`)
+- Bug in :meth:`Float64Index.astype` to unsigned integer dtype incorrectly casting to ``np.int64`` dtype (:issue:`45309`)
+- Bug in :meth:`Series.astype` and :meth:`DataFrame.astype` from floating dtype to unsigned integer dtype failing to raise in the presence of negative values (:issue:`45151`)
+- Bug in :func:`array` with ``FloatingDtype`` and values containing float-castable strings incorrectly raising (:issue:`45424`)
+- Bug when comparing string and datetime64ns objects causing ``OverflowError`` exception. (:issue:`45506`)
+- Bug in metaclass of generic abstract dtypes causing :meth:`DataFrame.apply` and :meth:`Series.apply` to raise for the built-in function ``type`` (:issue:`46684`)
+- Bug in :meth:`DataFrame.to_records` returning inconsistent numpy types if the index was a :class:`MultiIndex` (:issue:`47263`)
+- Bug in :meth:`DataFrame.to_dict` for ``orient="list"`` or ``orient="index"`` was not returning native types (:issue:`46751`)
+- Bug in :meth:`DataFrame.apply` that returns a :class:`DataFrame` instead of a :class:`Series` when applied to an empty :class:`DataFrame` and ``axis=1`` (:issue:`39111`)
+- Bug when inferring the dtype from an iterable that is *not* a NumPy ``ndarray`` consisting of all NumPy unsigned integer scalars did not result in an unsigned integer dtype (:issue:`47294`)
+- Bug in :meth:`DataFrame.eval` when pandas objects (e.g. ``'Timestamp'``) were column names (:issue:`44603`)
+
+Strings
+^^^^^^^
+- Bug in :meth:`str.startswith` and :meth:`str.endswith` when using other series as parameter _pat_. Now raises ``TypeError`` (:issue:`3485`)
+- Bug in :meth:`Series.str.zfill` when strings contain leading signs, padding '0' before the sign character rather than after as ``str.zfill`` from standard library (:issue:`20868`)
+
+Interval
+^^^^^^^^
+- Bug in :meth:`IntervalArray.__setitem__` when setting ``np.nan`` into an integer-backed array raising ``ValueError`` instead of ``TypeError`` (:issue:`45484`)
+- Bug in :class:`IntervalDtype` when using datetime64[ns, tz] as a dtype string (:issue:`46999`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`DataFrame.iloc` where indexing a single row on a :class:`DataFrame` with a single ExtensionDtype column gave a copy instead of a view on the underlying data (:issue:`45241`)
+- Bug in :meth:`DataFrame.__getitem__` returning copy when :class:`DataFrame` has duplicated columns even if a unique column is selected (:issue:`45316`, :issue:`41062`)
+- Bug in :meth:`Series.align` does not create :class:`MultiIndex` with union of levels when both MultiIndexes intersections are identical (:issue:`45224`)
+- Bug in setting a NA value (``None`` or ``np.nan``) into a :class:`Series` with int-based :class:`IntervalDtype` incorrectly casting to object dtype instead of a float-based :class:`IntervalDtype` (:issue:`45568`)
+- Bug in indexing setting values into an ``ExtensionDtype`` column with ``df.iloc[:, i] = values`` with ``values`` having the same dtype as ``df.iloc[:, i]`` incorrectly inserting a new array instead of setting in-place (:issue:`33457`)
+- Bug in :meth:`Series.__setitem__` with a non-integer :class:`Index` when using an integer key to set a value that cannot be set inplace where a ``ValueError`` was raised instead of casting to a common dtype (:issue:`45070`)
+- Bug in :meth:`DataFrame.loc` not casting ``None`` to ``NA`` when setting value as a list into :class:`DataFrame` (:issue:`47987`)
+- Bug in :meth:`Series.__setitem__` when setting incompatible values into a ``PeriodDtype`` or ``IntervalDtype`` :class:`Series` raising when indexing with a boolean mask but coercing when indexing with otherwise-equivalent indexers; these now consistently coerce, along with :meth:`Series.mask` and :meth:`Series.where` (:issue:`45768`)
+- Bug in :meth:`DataFrame.where` with multiple columns with datetime-like dtypes failing to downcast results consistent with other dtypes (:issue:`45837`)
+- Bug in :func:`isin` upcasting to ``float64`` with unsigned integer dtype and list-like argument without a dtype (:issue:`46485`)
+- Bug in :meth:`Series.loc.__setitem__` and :meth:`Series.loc.__getitem__` not raising when using multiple keys without using a :class:`MultiIndex` (:issue:`13831`)
+- Bug in :meth:`Index.reindex` raising ``AssertionError`` when ``level`` was specified but no :class:`MultiIndex` was given; level is ignored now (:issue:`35132`)
+- Bug when setting a value too large for a :class:`Series` dtype failing to coerce to a common type (:issue:`26049`, :issue:`32878`)
+- Bug in :meth:`loc.__setitem__` treating ``range`` keys as positional instead of label-based (:issue:`45479`)
+- Bug in :meth:`DataFrame.__setitem__` casting extension array dtypes to object when setting with a scalar key and :class:`DataFrame` as value (:issue:`46896`)
+- Bug in :meth:`Series.__setitem__` when setting a scalar to a nullable pandas dtype would not raise a ``TypeError`` if the scalar could not be cast (losslessly) to the nullable type (:issue:`45404`)
+- Bug in :meth:`Series.__setitem__` when setting ``boolean`` dtype values containing ``NA`` incorrectly raising instead of casting to ``boolean`` dtype (:issue:`45462`)
+- Bug in :meth:`Series.loc` raising with boolean indexer containing ``NA`` when :class:`Index` did not match (:issue:`46551`)
+- Bug in :meth:`Series.__setitem__` where setting :attr:`NA` into a numeric-dtype :class:`Series` would incorrectly upcast to object-dtype rather than treating the value as ``np.nan`` (:issue:`44199`)
+- Bug in :meth:`DataFrame.loc` when setting values to a column and right hand side is a dictionary (:issue:`47216`)
+- Bug in :meth:`Series.__setitem__` with ``datetime64[ns]`` dtype, an all-``False`` boolean mask, and an incompatible value incorrectly casting to ``object`` instead of retaining ``datetime64[ns]`` dtype (:issue:`45967`)
+- Bug in :meth:`Index.__getitem__`  raising ``ValueError`` when indexer is from boolean dtype with ``NA`` (:issue:`45806`)
+- Bug in :meth:`Series.__setitem__` losing precision when enlarging :class:`Series` with scalar (:issue:`32346`)
+- Bug in :meth:`Series.mask` with ``inplace=True`` or setting values with a boolean mask with small integer dtypes incorrectly raising (:issue:`45750`)
+- Bug in :meth:`DataFrame.mask` with ``inplace=True`` and ``ExtensionDtype`` columns incorrectly raising (:issue:`45577`)
+- Bug in getting a column from a DataFrame with an object-dtype row index with datetime-like values: the resulting Series now preserves the exact object-dtype Index from the parent DataFrame (:issue:`42950`)
+- Bug in :meth:`DataFrame.__getattribute__` raising ``AttributeError`` if columns have ``"string"`` dtype (:issue:`46185`)
+- Bug in :meth:`DataFrame.compare` returning all ``NaN`` column when comparing extension array dtype and numpy dtype (:issue:`44014`)
+- Bug in :meth:`DataFrame.where` setting wrong values with ``"boolean"`` mask for numpy dtype (:issue:`44014`)
+- Bug in indexing on a :class:`DatetimeIndex` with a ``np.str_`` key incorrectly raising (:issue:`45580`)
+- Bug in :meth:`CategoricalIndex.get_indexer` when index contains ``NaN`` values, resulting in elements that are in target but not present in the index to be mapped to the index of the NaN element, instead of -1 (:issue:`45361`)
+- Bug in setting large integer values into :class:`Series` with ``float32`` or ``float16`` dtype incorrectly altering these values instead of coercing to ``float64`` dtype (:issue:`45844`)
+- Bug in :meth:`Series.asof` and :meth:`DataFrame.asof` incorrectly casting bool-dtype results to ``float64`` dtype (:issue:`16063`)
+- Bug in :meth:`NDFrame.xs`, :meth:`DataFrame.iterrows`, :meth:`DataFrame.loc` and :meth:`DataFrame.iloc` not always propagating metadata (:issue:`28283`)
+- Bug in :meth:`DataFrame.sum` min_count changes dtype if input contains NaNs (:issue:`46947`)
+- Bug in :class:`IntervalTree` that lead to an infinite recursion. (:issue:`46658`)
+- Bug in :class:`PeriodIndex` raising ``AttributeError`` when indexing on ``NA``, rather than putting ``NaT`` in its place. (:issue:`46673`)
+- Bug in :meth:`DataFrame.at` would allow the modification of multiple columns (:issue:`48296`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`Series.fillna` and :meth:`DataFrame.fillna` with ``downcast`` keyword not being respected in some cases where there are no NA values present (:issue:`45423`)
+- Bug in :meth:`Series.fillna` and :meth:`DataFrame.fillna` with :class:`IntervalDtype` and incompatible value raising instead of casting to a common (usually object) dtype (:issue:`45796`)
+- Bug in :meth:`Series.map` not respecting ``na_action`` argument if mapper is a ``dict`` or :class:`Series` (:issue:`47527`)
+- Bug in :meth:`DataFrame.interpolate` with object-dtype column not returning a copy with ``inplace=False`` (:issue:`45791`)
+- Bug in :meth:`DataFrame.dropna` allows to set both ``how`` and ``thresh`` incompatible arguments (:issue:`46575`)
+- Bug in :meth:`DataFrame.fillna` ignored ``axis`` when :class:`DataFrame` is single block (:issue:`47713`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`DataFrame.loc` returning empty result when slicing a :class:`MultiIndex` with a negative step size and non-null start/stop values (:issue:`46156`)
+- Bug in :meth:`DataFrame.loc` raising when slicing a :class:`MultiIndex` with a negative step size other than -1 (:issue:`46156`)
+- Bug in :meth:`DataFrame.loc` raising when slicing a :class:`MultiIndex` with a negative step size and slicing a non-int labeled index level (:issue:`46156`)
+- Bug in :meth:`Series.to_numpy` where multiindexed Series could not be converted to numpy arrays when an ``na_value`` was supplied (:issue:`45774`)
+- Bug in :class:`MultiIndex.equals` not commutative when only one side has extension array dtype (:issue:`46026`)
+- Bug in :meth:`MultiIndex.from_tuples` cannot construct Index of empty tuples (:issue:`45608`)
+
+I/O
+^^^
+- Bug in :meth:`DataFrame.to_stata` where no error is raised if the :class:`DataFrame` contains ``-np.inf`` (:issue:`45350`)
+- Bug in :func:`read_excel` results in an infinite loop with certain ``skiprows`` callables (:issue:`45585`)
+- Bug in :meth:`DataFrame.info` where a new line at the end of the output is omitted when called on an empty :class:`DataFrame` (:issue:`45494`)
+- Bug in :func:`read_csv` not recognizing line break for ``on_bad_lines="warn"`` for ``engine="c"`` (:issue:`41710`)
+- Bug in :meth:`DataFrame.to_csv` not respecting ``float_format`` for ``Float64`` dtype (:issue:`45991`)
+- Bug in :func:`read_csv` not respecting a specified converter to index columns in all cases (:issue:`40589`)
+- Bug in :func:`read_csv` interpreting second row as :class:`Index` names even when ``index_col=False`` (:issue:`46569`)
+- Bug in :func:`read_parquet` when ``engine="pyarrow"`` which caused partial write to disk when column of unsupported datatype was passed (:issue:`44914`)
+- Bug in :func:`DataFrame.to_excel` and :class:`ExcelWriter` would raise when writing an empty DataFrame to a ``.ods`` file (:issue:`45793`)
+- Bug in :func:`read_csv` ignoring non-existing header row for ``engine="python"`` (:issue:`47400`)
+- Bug in :func:`read_excel` raising uncontrolled ``IndexError`` when ``header`` references non-existing rows (:issue:`43143`)
+- Bug in :func:`read_html` where elements surrounding ``<br>`` were joined without a space between them (:issue:`29528`)
+- Bug in :func:`read_csv` when data is longer than header leading to issues with callables in ``usecols`` expecting strings (:issue:`46997`)
+- Bug in Parquet roundtrip for Interval dtype with ``datetime64[ns]`` subtype (:issue:`45881`)
+- Bug in :func:`read_excel` when reading a ``.ods`` file with newlines between xml elements (:issue:`45598`)
+- Bug in :func:`read_parquet` when ``engine="fastparquet"`` where the file was not closed on error (:issue:`46555`)
+- :meth:`DataFrame.to_html` now excludes the ``border`` attribute from ``<table>`` elements when ``border`` keyword is set to ``False``.
+- Bug in :func:`read_sas` with certain types of compressed SAS7BDAT files (:issue:`35545`)
+- Bug in :func:`read_excel` not forward filling :class:`MultiIndex` when no names were given (:issue:`47487`)
+- Bug in :func:`read_sas` returned ``None`` rather than an empty DataFrame for SAS7BDAT files with zero rows (:issue:`18198`)
+- Bug in :meth:`DataFrame.to_string` using wrong missing value with extension arrays in :class:`MultiIndex` (:issue:`47986`)
+- Bug in :class:`StataWriter` where value labels were always written with default encoding (:issue:`46750`)
+- Bug in :class:`StataWriterUTF8` where some valid characters were removed from variable names (:issue:`47276`)
+- Bug in :meth:`DataFrame.to_excel` when writing an empty dataframe with :class:`MultiIndex` (:issue:`19543`)
+- Bug in :func:`read_sas` with RLE-compressed SAS7BDAT files that contain 0x40 control bytes (:issue:`31243`)
+- Bug in :func:`read_sas` that scrambled column names (:issue:`31243`)
+- Bug in :func:`read_sas` with RLE-compressed SAS7BDAT files that contain 0x00 control bytes (:issue:`47099`)
+- Bug in :func:`read_parquet` with ``use_nullable_dtypes=True`` where ``float64`` dtype was returned instead of nullable ``Float64`` dtype (:issue:`45694`)
+- Bug in :meth:`DataFrame.to_json` where ``PeriodDtype`` would not make the serialization roundtrip when read back with :meth:`read_json` (:issue:`44720`)
+- Bug in :func:`read_xml` when reading XML files with Chinese character tags and would raise ``XMLSyntaxError`` (:issue:`47902`)
+
+Period
+^^^^^^
+- Bug in subtraction of :class:`Period` from :class:`.PeriodArray` returning wrong results (:issue:`45999`)
+- Bug in :meth:`Period.strftime` and :meth:`PeriodIndex.strftime`, directives ``%l`` and ``%u`` were giving wrong results (:issue:`46252`)
+- Bug in inferring an incorrect ``freq`` when passing a string to :class:`Period` microseconds that are a multiple of 1000 (:issue:`46811`)
+- Bug in constructing a :class:`Period` from a :class:`Timestamp` or ``np.datetime64`` object with non-zero nanoseconds and ``freq="ns"`` incorrectly truncating the nanoseconds (:issue:`46811`)
+- Bug in adding ``np.timedelta64("NaT", "ns")`` to a :class:`Period` with a timedelta-like freq incorrectly raising ``IncompatibleFrequency`` instead of returning ``NaT`` (:issue:`47196`)
+- Bug in adding an array of integers to an array with :class:`PeriodDtype` giving incorrect results when ``dtype.freq.n > 1`` (:issue:`47209`)
+- Bug in subtracting a :class:`Period` from an array with :class:`PeriodDtype` returning incorrect results instead of raising ``OverflowError`` when the operation overflows (:issue:`47538`)
+
+Plotting
+^^^^^^^^
+- Bug in :meth:`DataFrame.plot.barh` that prevented labeling the x-axis and ``xlabel`` updating the y-axis label (:issue:`45144`)
+- Bug in :meth:`DataFrame.plot.box` that prevented labeling the x-axis (:issue:`45463`)
+- Bug in :meth:`DataFrame.boxplot` that prevented passing in ``xlabel`` and ``ylabel`` (:issue:`45463`)
+- Bug in :meth:`DataFrame.boxplot` that prevented specifying ``vert=False`` (:issue:`36918`)
+- Bug in :meth:`DataFrame.plot.scatter` that prevented specifying ``norm`` (:issue:`45809`)
+- Fix showing "None" as ylabel in :meth:`Series.plot` when not setting ylabel (:issue:`46129`)
+- Bug in :meth:`DataFrame.plot` that led to xticks and vertical grids being improperly placed when plotting a quarterly series (:issue:`47602`)
+- Bug in :meth:`DataFrame.plot` that prevented setting y-axis label, limits and ticks for a secondary y-axis (:issue:`47753`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`DataFrame.resample` ignoring ``closed="right"`` on :class:`TimedeltaIndex` (:issue:`45414`)
+- Bug in :meth:`.DataFrameGroupBy.transform` fails when ``func="size"`` and the input DataFrame has multiple columns (:issue:`27469`)
+- Bug in :meth:`.DataFrameGroupBy.size` and :meth:`.DataFrameGroupBy.transform` with ``func="size"`` produced incorrect results when ``axis=1`` (:issue:`45715`)
+- Bug in :meth:`.ExponentialMovingWindow.mean` with ``axis=1`` and ``engine='numba'`` when the :class:`DataFrame` has more columns than rows (:issue:`46086`)
+- Bug when using ``engine="numba"`` would return the same jitted function when modifying ``engine_kwargs`` (:issue:`46086`)
+- Bug in :meth:`.DataFrameGroupBy.transform` fails when ``axis=1`` and ``func`` is ``"first"`` or ``"last"`` (:issue:`45986`)
+- Bug in :meth:`.DataFrameGroupBy.cumsum` with ``skipna=False`` giving incorrect results (:issue:`46216`)
+- Bug in :meth:`.DataFrameGroupBy.sum`, :meth:`.SeriesGroupBy.sum`, :meth:`.DataFrameGroupBy.prod`, :meth:`.SeriesGroupBy.prod, :meth:`.DataFrameGroupBy.cumsum`, and :meth:`.SeriesGroupBy.cumsum` with integer dtypes losing precision (:issue:`37493`)
+- Bug in :meth:`.DataFrameGroupBy.cumsum` and :meth:`.SeriesGroupBy.cumsum` with ``timedelta64[ns]`` dtype failing to recognize ``NaT`` as a null value (:issue:`46216`)
+- Bug in :meth:`.DataFrameGroupBy.cumsum` and :meth:`.SeriesGroupBy.cumsum` with integer dtypes causing overflows when sum was bigger than maximum of dtype (:issue:`37493`)
+- Bug in :meth:`.DataFrameGroupBy.cummin`, :meth:`.SeriesGroupBy.cummin`, :meth:`.DataFrameGroupBy.cummax` and :meth:`.SeriesGroupBy.cummax` with nullable dtypes incorrectly altering the original data in place (:issue:`46220`)
+- Bug in :meth:`DataFrame.groupby` raising error when ``None`` is in first level of :class:`MultiIndex` (:issue:`47348`)
+- Bug in :meth:`.DataFrameGroupBy.cummax` and :meth:`.SeriesGroupBy.cummax` with ``int64`` dtype with leading value being the smallest possible int64 (:issue:`46382`)
+- Bug in :meth:`.DataFrameGroupBy.cumprod` and :meth:`.SeriesGroupBy.cumprod` ``NaN`` influences calculation in different columns with ``skipna=False`` (:issue:`48064`)
+- Bug in :meth:`.DataFrameGroupBy.max` and :meth:`.SeriesGroupBy.max` with empty groups and ``uint64`` dtype incorrectly raising ``RuntimeError`` (:issue:`46408`)
+- Bug in :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` would fail when ``func`` was a string and args or kwargs were supplied (:issue:`46479`)
+- Bug in :meth:`SeriesGroupBy.apply` would incorrectly name its result when there was a unique group (:issue:`46369`)
+- Bug in :meth:`.Rolling.sum` and :meth:`.Rolling.mean` would give incorrect result with window of same values (:issue:`42064`, :issue:`46431`)
+- Bug in :meth:`.Rolling.var` and :meth:`.Rolling.std` would give non-zero result with window of same values (:issue:`42064`)
+- Bug in :meth:`.Rolling.skew` and :meth:`.Rolling.kurt` would give NaN with window of same values (:issue:`30993`)
+- Bug in :meth:`.Rolling.var` would segfault calculating weighted variance when window size was larger than data size (:issue:`46760`)
+- Bug in :meth:`Grouper.__repr__` where ``dropna`` was not included. Now it is (:issue:`46754`)
+- Bug in :meth:`DataFrame.rolling` gives ValueError when center=True, axis=1 and win_type is specified (:issue:`46135`)
+- Bug in :meth:`.DataFrameGroupBy.describe` and :meth:`.SeriesGroupBy.describe` produces inconsistent results for empty datasets (:issue:`41575`)
+- Bug in :meth:`DataFrame.resample` reduction methods when used with ``on`` would attempt to aggregate the provided column (:issue:`47079`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` would not respect ``dropna=False`` when the input DataFrame/Series had a NaN values in a :class:`MultiIndex` (:issue:`46783`)
+- Bug in :meth:`DataFrameGroupBy.resample` raises ``KeyError`` when getting the result from a key list which misses the resample key (:issue:`47362`)
+- Bug in :meth:`DataFrame.groupby` would lose index columns when the DataFrame is empty for transforms, like fillna (:issue:`47787`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` with ``dropna=False`` and ``sort=False`` would put any null groups at the end instead the order that they are encountered (:issue:`46584`)
+
+Reshaping
+^^^^^^^^^
+- Bug in :func:`concat` between a :class:`Series` with integer dtype and another with :class:`CategoricalDtype` with integer categories and containing ``NaN`` values casting to object dtype instead of ``float64`` (:issue:`45359`)
+- Bug in :func:`get_dummies` that selected object and categorical dtypes but not string (:issue:`44965`)
+- Bug in :meth:`DataFrame.align` when aligning a :class:`MultiIndex` to a :class:`Series` with another :class:`MultiIndex` (:issue:`46001`)
+- Bug in concatenation with ``IntegerDtype``, or ``FloatingDtype`` arrays where the resulting dtype did not mirror the behavior of the non-nullable dtypes (:issue:`46379`)
+- Bug in :func:`concat` losing dtype of columns when ``join="outer"`` and ``sort=True`` (:issue:`47329`)
+- Bug in :func:`concat` not sorting the column names when ``None`` is included (:issue:`47331`)
+- Bug in :func:`concat` with identical key leads to error when indexing :class:`MultiIndex` (:issue:`46519`)
+- Bug in :func:`pivot_table` raising ``TypeError`` when ``dropna=True`` and aggregation column has extension array dtype (:issue:`47477`)
+- Bug in :func:`merge` raising error for ``how="cross"`` when using ``FIPS`` mode in ssl library (:issue:`48024`)
+- Bug in :meth:`DataFrame.join` with a list when using suffixes to join DataFrames with duplicate column names (:issue:`46396`)
+- Bug in :meth:`DataFrame.pivot_table` with ``sort=False`` results in sorted index (:issue:`17041`)
+- Bug in :meth:`concat` when ``axis=1`` and ``sort=False`` where the resulting Index was a :class:`Int64Index` instead of a :class:`RangeIndex` (:issue:`46675`)
+- Bug in :meth:`wide_to_long` raises when ``stubnames`` is missing in columns and ``i`` contains string dtype column (:issue:`46044`)
+- Bug in :meth:`DataFrame.join` with categorical index results in unexpected reordering (:issue:`47812`)
+
+Sparse
+^^^^^^
+- Bug in :meth:`Series.where` and :meth:`DataFrame.where` with ``SparseDtype`` failing to retain the array's ``fill_value`` (:issue:`45691`)
+- Bug in :meth:`SparseArray.unique` fails to keep original elements order (:issue:`47809`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :meth:`IntegerArray.searchsorted` and :meth:`FloatingArray.searchsorted` returning inconsistent results when acting on ``np.nan`` (:issue:`45255`)
+
+Styler
+^^^^^^
+- Bug when attempting to apply styling functions to an empty DataFrame subset (:issue:`45313`)
+- Bug in :class:`CSSToExcelConverter` leading to ``TypeError`` when border color provided without border style for ``xlsxwriter`` engine (:issue:`42276`)
+- Bug in :meth:`Styler.set_sticky` leading to white text on white background in dark mode (:issue:`46984`)
+- Bug in :meth:`Styler.to_latex` causing ``UnboundLocalError`` when ``clines="all;data"`` and the ``DataFrame`` has no rows. (:issue:`47203`)
+- Bug in :meth:`Styler.to_excel` when using ``vertical-align: middle;`` with ``xlsxwriter`` engine (:issue:`30107`)
+- Bug when applying styles to a DataFrame with boolean column labels (:issue:`47838`)
+
+Metadata
+^^^^^^^^
+- Fixed metadata propagation in :meth:`DataFrame.melt` (:issue:`28283`)
+- Fixed metadata propagation in :meth:`DataFrame.explode` (:issue:`28283`)
+
+Other
+^^^^^
+
+.. ***DO NOT USE THIS SECTION***
+
+- Bug in :func:`.assert_index_equal` with ``names=True`` and ``check_order=False`` not checking names (:issue:`47328`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_150.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.4.4..v1.5.0
diff --git a/pandas/doc/source/whatsnew/v1.5.1.rst b/pandas/doc/source/whatsnew/v1.5.1.rst
new file mode 100644
index 0000000000000..bcd8ddb9cbc0b
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.5.1.rst
@@ -0,0 +1,122 @@
+.. _whatsnew_151:
+
+What's new in 1.5.1 (October 19, 2022)
+--------------------------------------
+
+These are the changes in pandas 1.5.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_151.groupby_categorical_regr:
+
+Behavior of ``groupby`` with categorical groupers (:issue:`48645`)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In versions of pandas prior to 1.5, ``groupby`` with ``dropna=False`` would still drop
+NA values when the grouper was a categorical dtype. A fix for this was attempted in
+1.5, however it introduced a regression where passing ``observed=False`` and
+``dropna=False`` to ``groupby`` would result in only observed categories. It was found
+that the patch fixing the ``dropna=False`` bug is incompatible with ``observed=False``,
+and decided that the best resolution is to restore the correct ``observed=False``
+behavior at the cost of reintroducing the ``dropna=False`` bug.
+
+.. ipython:: python
+
+   df = pd.DataFrame(
+       {
+           "x": pd.Categorical([1, None], categories=[1, 2, 3]),
+           "y": [3, 4],
+       }
+   )
+   df
+
+*1.5.0 behavior*:
+
+.. code-block:: ipython
+
+   In [3]: # Correct behavior, NA values are not dropped
+           df.groupby("x", observed=True, dropna=False).sum()
+   Out[3]:
+        y
+   x
+   1    3
+   NaN  4
+
+
+   In [4]: # Incorrect behavior, only observed categories present
+           df.groupby("x", observed=False, dropna=False).sum()
+   Out[4]:
+        y
+   x
+   1    3
+   NaN  4
+
+
+*1.5.1 behavior*:
+
+.. ipython:: python
+
+   # Incorrect behavior, NA values are dropped
+   df.groupby("x", observed=True, dropna=False).sum()
+
+   # Correct behavior, unobserved categories present (NA values still dropped)
+   df.groupby("x", observed=False, dropna=False).sum()
+
+.. _whatsnew_151.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed Regression in :meth:`Series.__setitem__` casting ``None`` to ``NaN`` for object dtype (:issue:`48665`)
+- Fixed Regression in :meth:`DataFrame.loc` when setting values as a :class:`DataFrame` with all ``True`` indexer (:issue:`48701`)
+- Regression in :func:`.read_csv` causing an ``EmptyDataError`` when using an UTF-8 file handle that was already read from (:issue:`48646`)
+- Regression in :func:`to_datetime` when ``utc=True`` and ``arg`` contained timezone naive and aware arguments raised a ``ValueError`` (:issue:`48678`)
+- Fixed regression in :meth:`DataFrame.loc` raising ``FutureWarning`` when setting an empty :class:`DataFrame` (:issue:`48480`)
+- Fixed regression in :meth:`DataFrame.describe` raising ``TypeError`` when result contains ``NA`` (:issue:`48778`)
+- Fixed regression in :meth:`DataFrame.plot` ignoring invalid ``colormap`` for ``kind="scatter"`` (:issue:`48726`)
+- Fixed regression in :meth:`MultiIndex.values` resetting ``freq`` attribute of underlying :class:`Index` object (:issue:`49054`)
+- Fixed performance regression in :func:`factorize` when ``na_sentinel`` is not ``None`` and ``sort=False`` (:issue:`48620`)
+- Fixed regression causing an ``AttributeError`` during warning emitted if the provided table name in :meth:`DataFrame.to_sql` and the table name actually used in the database do not match (:issue:`48733`)
+- Fixed regression in :func:`to_datetime` when ``arg`` was a date string with nanosecond and ``format`` contained ``%f`` would raise a ``ValueError`` (:issue:`48767`)
+- Fixed regression in :func:`testing.assert_frame_equal` raising for :class:`MultiIndex` with :class:`Categorical` and ``check_like=True`` (:issue:`48975`)
+- Fixed regression in :meth:`DataFrame.fillna` replacing wrong values for ``datetime64[ns]`` dtype and ``inplace=True`` (:issue:`48863`)
+- Fixed :meth:`.DataFrameGroupBy.size` not returning a Series when ``axis=1`` (:issue:`48738`)
+- Fixed Regression in :meth:`.DataFrameGroupBy.apply` when user defined function is called on an empty dataframe (:issue:`47985`)
+- Fixed regression in :meth:`DataFrame.apply` when passing non-zero ``axis`` via keyword argument (:issue:`48656`)
+- Fixed regression in :meth:`Series.groupby` and :meth:`DataFrame.groupby` when the grouper is a nullable data type (e.g. :class:`Int64`) or a PyArrow-backed string array, contains null values, and ``dropna=False`` (:issue:`48794`)
+- Fixed performance regression in :meth:`Series.isin` with mismatching dtypes (:issue:`49162`)
+- Fixed regression in :meth:`DataFrame.to_parquet` raising when file name was specified as ``bytes`` (:issue:`48944`)
+- Fixed regression in :class:`ExcelWriter` where the ``book`` attribute could no longer be set; however setting this attribute is now deprecated and this ability will be removed in a future version of pandas (:issue:`48780`)
+- Fixed regression in :meth:`DataFrame.corrwith` when computing correlation on tied data with ``method="spearman"`` (:issue:`48826`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_151.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :meth:`Series.__getitem__` not falling back to positional for integer keys and boolean :class:`Index` (:issue:`48653`)
+- Bug in :meth:`DataFrame.to_hdf` raising ``AssertionError`` with boolean index (:issue:`48667`)
+- Bug in :func:`testing.assert_index_equal` for extension arrays with non matching ``NA`` raising ``ValueError`` (:issue:`48608`)
+- Bug in :meth:`DataFrame.pivot_table` raising unexpected ``FutureWarning`` when setting datetime column as index (:issue:`48683`)
+- Bug in :meth:`DataFrame.sort_values` emitting unnecessary ``FutureWarning`` when called on :class:`DataFrame` with boolean sparse columns (:issue:`48784`)
+- Bug in :class:`.arrays.ArrowExtensionArray` with a comparison operator to an invalid object would not raise a ``NotImplementedError`` (:issue:`48833`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_151.other:
+
+Other
+~~~~~
+- Avoid showing deprecated signatures when introspecting functions with warnings about arguments becoming keyword-only (:issue:`48692`)
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_151.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.5.0..v1.5.1
diff --git a/pandas/doc/source/whatsnew/v1.5.2.rst b/pandas/doc/source/whatsnew/v1.5.2.rst
new file mode 100644
index 0000000000000..efc8ca6afa342
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.5.2.rst
@@ -0,0 +1,46 @@
+.. _whatsnew_152:
+
+What's new in 1.5.2 (November 21, 2022)
+---------------------------------------
+
+These are the changes in pandas 1.5.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_152.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`MultiIndex.join` for extension array dtypes (:issue:`49277`)
+- Fixed regression in :meth:`Series.replace` raising ``RecursionError`` with numeric dtype and when specifying ``value=None`` (:issue:`45725`)
+- Fixed regression in arithmetic operations for :class:`DataFrame` with :class:`MultiIndex` columns with different dtypes (:issue:`49769`)
+- Fixed regression in :meth:`DataFrame.plot` preventing :class:`~matplotlib.colors.Colormap` instance
+  from being passed using the ``colormap`` argument if Matplotlib 3.6+ is used (:issue:`49374`)
+- Fixed regression in :func:`date_range` returning an invalid set of periods for ``CustomBusinessDay`` frequency and ``start`` date with timezone (:issue:`49441`)
+- Fixed performance regression in groupby operations (:issue:`49676`)
+- Fixed regression in :class:`Timedelta` constructor returning object of wrong type when subclassing ``Timedelta`` (:issue:`49579`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_152.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in the Copy-on-Write implementation losing track of views in certain chained indexing cases (:issue:`48996`)
+- Fixed memory leak in :meth:`.Styler.to_excel` (:issue:`49751`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_152.other:
+
+Other
+~~~~~
+- Reverted ``color`` as an alias for ``c`` and ``size`` as an alias for ``s`` in function :meth:`DataFrame.plot.scatter` (:issue:`49732`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_152.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.5.1..v1.5.2
diff --git a/pandas/doc/source/whatsnew/v1.5.3.rst b/pandas/doc/source/whatsnew/v1.5.3.rst
new file mode 100644
index 0000000000000..5358d45048af8
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v1.5.3.rst
@@ -0,0 +1,58 @@
+.. _whatsnew_153:
+
+What's new in 1.5.3 (January 18, 2023)
+--------------------------------------
+
+These are the changes in pandas 1.5.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_153.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed performance regression in :meth:`Series.isin` when ``values`` is empty (:issue:`49839`)
+- Fixed regression in :meth:`DataFrame.memory_usage` showing unnecessary ``FutureWarning`` when :class:`DataFrame` is empty (:issue:`50066`)
+- Fixed regression in :meth:`.DataFrameGroupBy.transform` when used with ``as_index=False`` (:issue:`49834`)
+- Enforced reversion of ``color`` as an alias for ``c`` and ``size`` as an alias for ``s`` in function :meth:`DataFrame.plot.scatter` (:issue:`49732`)
+- Fixed regression in :meth:`.SeriesGroupBy.apply` setting a ``name`` attribute on the result if the result was a :class:`DataFrame` (:issue:`49907`)
+- Fixed performance regression in setting with the :meth:`~DataFrame.at` indexer (:issue:`49771`)
+- Fixed regression in :func:`to_datetime` raising ``ValueError`` when parsing array of ``float`` containing ``np.nan`` (:issue:`50237`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_153.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in the Copy-on-Write implementation losing track of views when indexing a :class:`DataFrame` with another :class:`DataFrame` (:issue:`50630`)
+- Bug in :meth:`.Styler.to_excel` leading to error when unrecognized ``border-style`` (e.g. ``"hair"``) provided to Excel writers (:issue:`48649`)
+- Bug in :meth:`Series.quantile` emitting warning from NumPy when :class:`Series` has only ``NA`` values (:issue:`50681`)
+- Bug when chaining several :meth:`.Styler.concat` calls, only the last styler was concatenated (:issue:`49207`)
+- Fixed bug when instantiating a :class:`DataFrame` subclass inheriting from ``typing.Generic`` that triggered a ``UserWarning`` on python 3.11 (:issue:`49649`)
+- Bug in :func:`pivot_table` with NumPy 1.24 or greater when the :class:`DataFrame` columns has nested elements (:issue:`50342`)
+- Bug in :func:`pandas.testing.assert_series_equal` (and equivalent ``assert_`` functions) when having nested data and using numpy >= 1.25 (:issue:`50360`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_153.other:
+
+Other
+~~~~~
+
+.. note::
+
+    If you are using :meth:`DataFrame.to_sql`, :func:`read_sql`, :func:`read_sql_table`, or :func:`read_sql_query` with SQLAlchemy 1.4.46 or greater,
+    you may see a ``sqlalchemy.exc.RemovedIn20Warning``. These warnings can be safely ignored for the SQLAlchemy 1.4.x releases
+    as pandas works toward compatibility with SQLAlchemy 2.0.
+
+- Reverted deprecation (:issue:`45324`) of behavior of :meth:`Series.__getitem__` and :meth:`Series.__setitem__` slicing with an integer :class:`Index`; this will remain positional (:issue:`49612`)
+- A ``FutureWarning`` raised when attempting to set values inplace with :meth:`DataFrame.loc` or :meth:`DataFrame.iloc` has been changed to a ``DeprecationWarning`` (:issue:`48673`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_153.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.5.2..v1.5.3
diff --git a/pandas/doc/source/whatsnew/v2.0.0.rst b/pandas/doc/source/whatsnew/v2.0.0.rst
new file mode 100644
index 0000000000000..ddcd69c3fd962
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.0.0.rst
@@ -0,0 +1,1416 @@
+.. _whatsnew_200:
+
+What's new in 2.0.0 (April 3, 2023)
+-----------------------------------
+
+These are the changes in pandas 2.0.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_200.enhancements.optional_dependency_management_pip:
+
+Installing optional dependencies with pip extras
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+When installing pandas using pip, sets of optional dependencies can also be installed by specifying extras.
+
+.. code-block:: bash
+
+  pip install "pandas[performance, aws]>=2.0.0"
+
+The available extras, found in the :ref:`installation guide<install.dependencies>`, are
+``[all, performance, computation, fss, aws, gcp, excel, parquet, feather, hdf5, spss, postgresql, mysql,
+sql-other, html, xml, plot, output_formatting, clipboard, compression, test]`` (:issue:`39164`).
+
+.. _whatsnew_200.enhancements.index_can_hold_numpy_numeric_dtypes:
+
+:class:`Index` can now hold numpy numeric dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+It is now possible to use any numpy numeric dtype in a :class:`Index` (:issue:`42717`).
+
+Previously it was only possible to use ``int64``, ``uint64`` & ``float64`` dtypes:
+
+.. code-block:: ipython
+
+    In [1]: pd.Index([1, 2, 3], dtype=np.int8)
+    Out[1]: Int64Index([1, 2, 3], dtype="int64")
+    In [2]: pd.Index([1, 2, 3], dtype=np.uint16)
+    Out[2]: UInt64Index([1, 2, 3], dtype="uint64")
+    In [3]: pd.Index([1, 2, 3], dtype=np.float32)
+    Out[3]: Float64Index([1.0, 2.0, 3.0], dtype="float64")
+
+:class:`Int64Index`, :class:`UInt64Index` & :class:`Float64Index` were deprecated in pandas
+version 1.4 and have now been removed. Instead :class:`Index` should be used directly, and
+can it now take all numpy numeric dtypes, i.e.
+``int8``/ ``int16``/``int32``/``int64``/``uint8``/``uint16``/``uint32``/``uint64``/``float32``/``float64`` dtypes:
+
+.. ipython:: python
+
+    pd.Index([1, 2, 3], dtype=np.int8)
+    pd.Index([1, 2, 3], dtype=np.uint16)
+    pd.Index([1, 2, 3], dtype=np.float32)
+
+The ability for :class:`Index` to hold the numpy numeric dtypes has meant some changes in pandas
+functionality. In particular, operations that previously were forced to create 64-bit indexes,
+can now create indexes with lower bit sizes, e.g. 32-bit indexes.
+
+Below is a possibly non-exhaustive list of changes:
+
+1. Instantiating using a numpy numeric array now follows the dtype of the numpy array.
+   Previously, all indexes created from numpy numeric arrays were forced to 64-bit. Now,
+   for example, ``Index(np.array([1, 2, 3]))`` will be ``int32`` on 32-bit systems, where
+   it previously would have been ``int64`` even on 32-bit systems.
+   Instantiating :class:`Index` using a list of numbers will still return 64bit dtypes,
+   e.g. ``Index([1, 2, 3])`` will have a ``int64`` dtype, which is the same as previously.
+2. The various numeric datetime attributes of :class:`DatetimeIndex` (:attr:`~DatetimeIndex.day`,
+   :attr:`~DatetimeIndex.month`, :attr:`~DatetimeIndex.year` etc.) were previously in of
+   dtype ``int64``, while they were ``int32`` for :class:`arrays.DatetimeArray`. They are now
+   ``int32`` on :class:`DatetimeIndex` also:
+
+   .. ipython:: python
+
+       idx = pd.date_range(start='1/1/2018', periods=3, freq='ME')
+       idx.array.year
+       idx.year
+
+3. Level dtypes on Indexes from :meth:`Series.sparse.from_coo` are now of dtype ``int32``,
+   the same as they are on the ``rows``/``cols`` on a scipy sparse matrix. Previously they
+   were of dtype ``int64``.
+
+   .. ipython:: python
+
+       from scipy import sparse
+       A = sparse.coo_matrix(
+           ([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])), shape=(3, 4)
+       )
+       ser = pd.Series.sparse.from_coo(A)
+       ser.index.dtypes
+
+4. :class:`Index` cannot be instantiated using a float16 dtype. Previously instantiating
+   an :class:`Index` using dtype ``float16`` resulted in a :class:`Float64Index` with a
+   ``float64`` dtype. It now raises a ``NotImplementedError``:
+
+   .. ipython:: python
+       :okexcept:
+
+       pd.Index([1, 2, 3], dtype=np.float16)
+
+
+.. _whatsnew_200.enhancements.io_dtype_backend:
+
+Argument ``dtype_backend``, to return pyarrow-backed or numpy-backed nullable dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The following functions gained a new keyword ``dtype_backend`` (:issue:`36712`)
+
+* :func:`read_csv`
+* :func:`read_clipboard`
+* :func:`read_fwf`
+* :func:`read_excel`
+* :func:`read_html`
+* :func:`read_xml`
+* :func:`read_json`
+* :func:`read_sql`
+* :func:`read_sql_query`
+* :func:`read_sql_table`
+* :func:`read_parquet`
+* :func:`read_orc`
+* :func:`read_feather`
+* :func:`read_spss`
+* :func:`to_numeric`
+* :meth:`DataFrame.convert_dtypes`
+* :meth:`Series.convert_dtypes`
+
+When this option is set to ``"numpy_nullable"`` it will return a :class:`DataFrame` that is
+backed by nullable dtypes.
+
+When this keyword is set to ``"pyarrow"``, then these functions will return pyarrow-backed nullable :class:`ArrowDtype` DataFrames (:issue:`48957`, :issue:`49997`):
+
+* :func:`read_csv`
+* :func:`read_clipboard`
+* :func:`read_fwf`
+* :func:`read_excel`
+* :func:`read_html`
+* :func:`read_xml`
+* :func:`read_json`
+* :func:`read_sql`
+* :func:`read_sql_query`
+* :func:`read_sql_table`
+* :func:`read_parquet`
+* :func:`read_orc`
+* :func:`read_feather`
+* :func:`read_spss`
+* :func:`to_numeric`
+* :meth:`DataFrame.convert_dtypes`
+* :meth:`Series.convert_dtypes`
+
+.. ipython:: python
+
+    import io
+    data = io.StringIO("""a,b,c,d,e,f,g,h,i
+        1,2.5,True,a,,,,,
+        3,4.5,False,b,6,7.5,True,a,
+    """)
+    df = pd.read_csv(data, dtype_backend="pyarrow")
+    df.dtypes
+
+    data.seek(0)
+    df_pyarrow = pd.read_csv(data, dtype_backend="pyarrow", engine="pyarrow")
+    df_pyarrow.dtypes
+
+Copy-on-Write improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- A new lazy copy mechanism that defers the copy until the object in question is modified
+  was added to the methods listed in
+  :ref:`Copy-on-Write optimizations <copy_on_write.optimizations>`.
+  These methods return views when Copy-on-Write is enabled, which provides a significant
+  performance improvement compared to the regular execution (:issue:`49473`).
+
+- Accessing a single column of a DataFrame as a Series (e.g. ``df["col"]``) now always
+  returns a new object every time it is constructed when Copy-on-Write is enabled (not
+  returning multiple times an identical, cached Series object). This ensures that those
+  Series objects correctly follow the Copy-on-Write rules (:issue:`49450`)
+
+- The :class:`Series` constructor will now create a lazy copy (deferring the copy until
+  a modification to the data happens) when constructing a Series from an existing
+  Series with the default of ``copy=False`` (:issue:`50471`)
+
+- The :class:`DataFrame` constructor will now create a lazy copy (deferring the copy until
+  a modification to the data happens) when constructing from an existing
+  :class:`DataFrame` with the default of ``copy=False`` (:issue:`51239`)
+
+- The :class:`DataFrame` constructor, when constructing a DataFrame from a dictionary
+  of Series objects and specifying ``copy=False``, will now use a lazy copy
+  of those Series objects for the columns of the DataFrame (:issue:`50777`)
+
+- The :class:`DataFrame` constructor, when constructing a DataFrame from a
+  :class:`Series` or :class:`Index` and specifying ``copy=False``, will
+  now respect Copy-on-Write.
+
+- The :class:`DataFrame` and :class:`Series` constructors, when constructing from
+  a NumPy array, will now copy the array by default to avoid mutating
+  the :class:`DataFrame` / :class:`Series`
+  when mutating the array. Specify ``copy=False`` to get the old behavior.
+  When setting ``copy=False`` pandas does not guarantee correct Copy-on-Write
+  behavior when the NumPy array is modified after creation of the
+  :class:`DataFrame` / :class:`Series`.
+
+- The :meth:`DataFrame.from_records` will now respect Copy-on-Write when called
+  with a :class:`DataFrame`.
+
+- Trying to set values using chained assignment (for example, ``df["a"][1:3] = 0``)
+  will now always raise a warning when Copy-on-Write is enabled. In this mode,
+  chained assignment can never work because we are always setting into a temporary
+  object that is the result of an indexing operation (getitem), which under
+  Copy-on-Write always behaves as a copy. Thus, assigning through a chain
+  can never update the original Series or DataFrame. Therefore, an informative
+  warning is raised to the user to avoid silently doing nothing (:issue:`49467`)
+
+- :meth:`DataFrame.replace` will now respect the Copy-on-Write mechanism
+  when ``inplace=True``.
+
+- :meth:`DataFrame.transpose` will now respect the Copy-on-Write mechanism.
+
+- Arithmetic operations that can be inplace, e.g. ``ser *= 2`` will now respect the
+  Copy-on-Write mechanism.
+
+- :meth:`DataFrame.__getitem__` will now respect the Copy-on-Write mechanism when the
+  :class:`DataFrame` has :class:`MultiIndex` columns.
+
+- :meth:`Series.__getitem__` will now respect the Copy-on-Write mechanism when the
+   :class:`Series` has a :class:`MultiIndex`.
+
+- :meth:`Series.view` will now respect the Copy-on-Write mechanism.
+
+Copy-on-Write can be enabled through one of
+
+.. code-block:: python
+
+    pd.set_option("mode.copy_on_write", True)
+
+
+.. code-block:: python
+
+    pd.options.mode.copy_on_write = True
+
+Alternatively, copy on write can be enabled locally through:
+
+.. code-block:: python
+
+    with pd.option_context("mode.copy_on_write", True):
+        ...
+
+.. _whatsnew_200.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- Added support for ``str`` accessor methods when using :class:`ArrowDtype`  with a ``pyarrow.string`` type (:issue:`50325`)
+- Added support for ``dt`` accessor methods when using :class:`ArrowDtype` with a ``pyarrow.timestamp`` type (:issue:`50954`)
+- :func:`read_sas` now supports using ``encoding='infer'`` to correctly read and use the encoding specified by the sas file. (:issue:`48048`)
+- :meth:`.DataFrameGroupBy.quantile`, :meth:`.SeriesGroupBy.quantile` and :meth:`.DataFrameGroupBy.std` now preserve nullable dtypes instead of casting to numpy dtypes (:issue:`37493`)
+- :meth:`.DataFrameGroupBy.std`, :meth:`.SeriesGroupBy.std` now support datetime64, timedelta64, and :class:`DatetimeTZDtype` dtypes (:issue:`48481`)
+- :meth:`Series.add_suffix`, :meth:`DataFrame.add_suffix`, :meth:`Series.add_prefix` and :meth:`DataFrame.add_prefix` support an ``axis`` argument. If ``axis`` is set, the default behaviour of which axis to consider can be overwritten (:issue:`47819`)
+- :func:`.testing.assert_frame_equal` now shows the first element where the DataFrames differ, analogously to ``pytest``'s output (:issue:`47910`)
+- Added ``index`` parameter to :meth:`DataFrame.to_dict` (:issue:`46398`)
+- Added support for extension array dtypes in :func:`merge` (:issue:`44240`)
+- Added metadata propagation for binary operators on :class:`DataFrame` (:issue:`28283`)
+- Added ``cumsum``, ``cumprod``, ``cummin`` and ``cummax`` to the ``ExtensionArray`` interface via ``_accumulate`` (:issue:`28385`)
+- :class:`.CategoricalConversionWarning`, :class:`.InvalidComparison`, :class:`.InvalidVersion`, :class:`.LossySetitemError`, and :class:`.NoBufferPresent` are now exposed in ``pandas.errors`` (:issue:`27656`)
+- Fix ``test`` optional_extra by adding missing test package ``pytest-asyncio`` (:issue:`48361`)
+- :func:`DataFrame.astype` exception message thrown improved to include column name when type conversion is not possible. (:issue:`47571`)
+- :func:`date_range` now supports a ``unit`` keyword ("s", "ms", "us", or "ns") to specify the desired resolution of the output index (:issue:`49106`)
+- :func:`timedelta_range` now supports a ``unit`` keyword ("s", "ms", "us", or "ns") to specify the desired resolution of the output index (:issue:`49824`)
+- :meth:`DataFrame.to_json` now supports a ``mode`` keyword with supported inputs 'w' and 'a'. Defaulting to 'w', 'a' can be used when lines=True and orient='records' to append record oriented json lines to an existing json file. (:issue:`35849`)
+- Added ``name`` parameter to :meth:`IntervalIndex.from_breaks`, :meth:`IntervalIndex.from_arrays` and :meth:`IntervalIndex.from_tuples` (:issue:`48911`)
+- Improve exception message when using :func:`.testing.assert_frame_equal` on a :class:`DataFrame` to include the column that is compared (:issue:`50323`)
+- Improved error message for :func:`merge_asof` when join-columns were duplicated (:issue:`50102`)
+- Added support for extension array dtypes to :func:`get_dummies` (:issue:`32430`)
+- Added :meth:`Index.infer_objects` analogous to :meth:`Series.infer_objects` (:issue:`50034`)
+- Added ``copy`` parameter to :meth:`Series.infer_objects` and :meth:`DataFrame.infer_objects`, passing ``False`` will avoid making copies for series or columns that are already non-object or where no better dtype can be inferred (:issue:`50096`)
+- :meth:`DataFrame.plot.hist` now recognizes ``xlabel`` and ``ylabel`` arguments (:issue:`49793`)
+- :meth:`Series.drop_duplicates` has gained ``ignore_index`` keyword to reset index (:issue:`48304`)
+- :meth:`Series.dropna` and :meth:`DataFrame.dropna` has gained ``ignore_index`` keyword to reset index (:issue:`31725`)
+- Improved error message in :func:`to_datetime` for non-ISO8601 formats, informing users about the position of the first error (:issue:`50361`)
+- Improved error message when trying to align :class:`DataFrame` objects (for example, in :func:`DataFrame.compare`) to clarify that "identically labelled" refers to both index and columns (:issue:`50083`)
+- Added support for :meth:`Index.min` and :meth:`Index.max` for pyarrow string dtypes (:issue:`51397`)
+- Added :meth:`DatetimeIndex.as_unit` and :meth:`TimedeltaIndex.as_unit` to convert to different resolutions; supported resolutions are "s", "ms", "us", and "ns" (:issue:`50616`)
+- Added :meth:`Series.dt.unit` and :meth:`Series.dt.as_unit` to convert to different resolutions; supported resolutions are "s", "ms", "us", and "ns" (:issue:`51223`)
+- Added new argument ``dtype`` to :func:`read_sql` to be consistent with :func:`read_sql_query` (:issue:`50797`)
+- :func:`read_csv`, :func:`read_table`, :func:`read_fwf` and :func:`read_excel` now accept ``date_format`` (:issue:`50601`)
+- :func:`to_datetime` now accepts ``"ISO8601"`` as an argument to ``format``, which will match any ISO8601 string (but possibly not identically-formatted) (:issue:`50411`)
+- :func:`to_datetime` now accepts ``"mixed"`` as an argument to ``format``, which will infer the format for each element individually (:issue:`50972`)
+- Added new argument ``engine`` to :func:`read_json` to support parsing JSON with pyarrow by specifying ``engine="pyarrow"`` (:issue:`48893`)
+- Added support for SQLAlchemy 2.0 (:issue:`40686`)
+- Added support for ``decimal`` parameter when ``engine="pyarrow"`` in :func:`read_csv` (:issue:`51302`)
+- :class:`Index` set operations :meth:`Index.union`, :meth:`Index.intersection`, :meth:`Index.difference`, and :meth:`Index.symmetric_difference` now support ``sort=True``, which will always return a sorted result, unlike the default ``sort=None`` which does not sort in some cases (:issue:`25151`)
+- Added new escape mode "latex-math" to avoid escaping "$" in formatter (:issue:`50040`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_200.notable_bug_fixes.cumsum_cumprod_overflow:
+
+:meth:`.DataFrameGroupBy.cumsum` and :meth:`.DataFrameGroupBy.cumprod` overflow instead of lossy casting to float
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions we cast to float when applying ``cumsum`` and ``cumprod`` which
+lead to incorrect results even if the result could be hold by ``int64`` dtype.
+Additionally, the aggregation overflows consistent with numpy and the regular
+:meth:`DataFrame.cumprod` and :meth:`DataFrame.cumsum` methods when the limit of
+``int64`` is reached (:issue:`37493`).
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"key": ["b"] * 7, "value": 625})
+    In [2]: df.groupby("key")["value"].cumprod()[5]
+    Out[2]: 5.960464477539062e+16
+
+We return incorrect results with the 6th value.
+
+*New Behavior*
+
+.. ipython:: python
+
+    df = pd.DataFrame({"key": ["b"] * 7, "value": 625})
+    df.groupby("key")["value"].cumprod()
+
+We overflow with the 7th value, but the 6th value is still correct.
+
+.. _whatsnew_200.notable_bug_fixes.groupby_nth_filter:
+
+:meth:`.DataFrameGroupBy.nth` and :meth:`.SeriesGroupBy.nth` now behave as filtrations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, :meth:`.DataFrameGroupBy.nth` and
+:meth:`.SeriesGroupBy.nth` acted as if they were aggregations. However, for most
+inputs ``n``, they may return either zero or multiple rows per group. This means
+that they are filtrations, similar to e.g. :meth:`.DataFrameGroupBy.head`. pandas
+now treats them as filtrations (:issue:`13666`).
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 1, 2, 1, 2], "b": [np.nan, 2.0, 3.0, 4.0, 5.0]})
+    gb = df.groupby("a")
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [5]: gb.nth(n=1)
+    Out[5]:
+       A    B
+    1  1  2.0
+    4  2  5.0
+
+*New Behavior*
+
+.. ipython:: python
+
+    gb.nth(n=1)
+
+In particular, the index of the result is derived from the input by selecting
+the appropriate rows. Also, when ``n`` is larger than the group, no rows instead of
+``NaN`` is returned.
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [5]: gb.nth(n=3, dropna="any")
+    Out[5]:
+        B
+    A
+    1 NaN
+    2 NaN
+
+*New Behavior*
+
+.. ipython:: python
+
+    gb.nth(n=3, dropna="any")
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_200.api_breaking.unsupported_datetimelike_dtype_arg:
+
+Construction with datetime64 or timedelta64 dtype with unsupported resolution
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+In past versions, when constructing a :class:`Series` or :class:`DataFrame` and
+passing a "datetime64" or "timedelta64" dtype with unsupported resolution
+(i.e. anything other than "ns"), pandas would silently replace the given dtype
+with its nanosecond analogue:
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [5]: pd.Series(["2016-01-01"], dtype="datetime64[s]")
+   Out[5]:
+   0   2016-01-01
+   dtype: datetime64[ns]
+
+   In [6] pd.Series(["2016-01-01"], dtype="datetime64[D]")
+   Out[6]:
+   0   2016-01-01
+   dtype: datetime64[ns]
+
+In pandas 2.0 we support resolutions "s", "ms", "us", and "ns". When passing
+a supported dtype (e.g. "datetime64[s]"), the result now has exactly
+the requested dtype:
+
+*New behavior*:
+
+.. ipython:: python
+
+   pd.Series(["2016-01-01"], dtype="datetime64[s]")
+
+With an un-supported dtype, pandas now raises instead of silently swapping in
+a supported dtype:
+
+*New behavior*:
+
+.. ipython:: python
+   :okexcept:
+
+   pd.Series(["2016-01-01"], dtype="datetime64[D]")
+
+.. _whatsnew_200.api_breaking.value_counts:
+
+Value counts sets the resulting name to ``count``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+In past versions, when running :meth:`Series.value_counts`, the result would inherit
+the original object's name, and the result index would be nameless. This would cause
+confusion when resetting the index, and the column names would not correspond with the
+column values.
+Now, the result name will be ``'count'`` (or ``'proportion'`` if ``normalize=True`` was passed),
+and the index will be named after the original object (:issue:`49497`).
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [8]: pd.Series(['quetzal', 'quetzal', 'elk'], name='animal').value_counts()
+
+    Out[2]:
+    quetzal    2
+    elk        1
+    Name: animal, dtype: int64
+
+*New behavior*:
+
+.. ipython:: python
+
+    pd.Series(['quetzal', 'quetzal', 'elk'], name='animal').value_counts()
+
+Likewise for other ``value_counts`` methods (for example, :meth:`DataFrame.value_counts`).
+
+.. _whatsnew_200.api_breaking.astype_to_unsupported_datetimelike:
+
+Disallow astype conversion to non-supported datetime64/timedelta64 dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+In previous versions, converting a :class:`Series` or :class:`DataFrame`
+from ``datetime64[ns]`` to a different ``datetime64[X]`` dtype would return
+with ``datetime64[ns]`` dtype instead of the requested dtype. In pandas 2.0,
+support is added for "datetime64[s]", "datetime64[ms]", and "datetime64[us]" dtypes,
+so converting to those dtypes gives exactly the requested dtype:
+
+*Previous behavior*:
+
+.. ipython:: python
+
+   idx = pd.date_range("2016-01-01", periods=3)
+   ser = pd.Series(idx)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [4]: ser.astype("datetime64[s]")
+   Out[4]:
+   0   2016-01-01
+   1   2016-01-02
+   2   2016-01-03
+   dtype: datetime64[ns]
+
+With the new behavior, we get exactly the requested dtype:
+
+*New behavior*:
+
+.. ipython:: python
+
+   ser.astype("datetime64[s]")
+
+For non-supported resolutions e.g. "datetime64[D]", we raise instead of silently
+ignoring the requested dtype:
+
+*New behavior*:
+
+.. ipython:: python
+   :okexcept:
+
+   ser.astype("datetime64[D]")
+
+For conversion from ``timedelta64[ns]`` dtypes, the old behavior converted
+to a floating point format.
+
+*Previous behavior*:
+
+.. ipython:: python
+
+   idx = pd.timedelta_range("1 Day", periods=3)
+   ser = pd.Series(idx)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [7]: ser.astype("timedelta64[s]")
+   Out[7]:
+   0     86400.0
+   1    172800.0
+   2    259200.0
+   dtype: float64
+
+   In [8]: ser.astype("timedelta64[D]")
+   Out[8]:
+   0    1.0
+   1    2.0
+   2    3.0
+   dtype: float64
+
+The new behavior, as for datetime64, either gives exactly the requested dtype or raises:
+
+*New behavior*:
+
+.. ipython:: python
+   :okexcept:
+
+   ser.astype("timedelta64[s]")
+   ser.astype("timedelta64[D]")
+
+.. _whatsnew_200.api_breaking.default_to_stdlib_tzinfos:
+
+UTC and fixed-offset timezones default to standard-library tzinfo objects
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+In previous versions, the default ``tzinfo`` object used to represent UTC
+was ``pytz.UTC``. In pandas 2.0, we default to ``datetime.timezone.utc`` instead.
+Similarly, for timezones represent fixed UTC offsets, we use ``datetime.timezone``
+objects instead of ``pytz.FixedOffset`` objects. See (:issue:`34916`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [2]: ts = pd.Timestamp("2016-01-01", tz="UTC")
+   In [3]: type(ts.tzinfo)
+   Out[3]: pytz.UTC
+
+   In [4]: ts2 = pd.Timestamp("2016-01-01 04:05:06-07:00")
+   In [3]: type(ts2.tzinfo)
+   Out[5]: pytz._FixedOffset
+
+*New behavior*:
+
+.. ipython:: python
+
+   ts = pd.Timestamp("2016-01-01", tz="UTC")
+   type(ts.tzinfo)
+
+   ts2 = pd.Timestamp("2016-01-01 04:05:06-07:00")
+   type(ts2.tzinfo)
+
+For timezones that are neither UTC nor fixed offsets, e.g. "US/Pacific", we
+continue to default to ``pytz`` objects.
+
+.. _whatsnew_200.api_breaking.zero_len_indexes:
+
+Empty DataFrames/Series will now default to have a ``RangeIndex``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Before, constructing an empty (where ``data`` is ``None`` or an empty list-like argument) :class:`Series` or :class:`DataFrame` without
+specifying the axes (``index=None``, ``columns=None``) would return the axes as empty :class:`Index` with object dtype.
+
+Now, the axes return an empty :class:`RangeIndex` (:issue:`49572`).
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+   In [8]: pd.Series().index
+   Out[8]:
+   Index([], dtype='object')
+
+   In [9] pd.DataFrame().axes
+   Out[9]:
+   [Index([], dtype='object'), Index([], dtype='object')]
+
+*New behavior*:
+
+.. ipython:: python
+
+   pd.Series().index
+   pd.DataFrame().axes
+
+.. _whatsnew_200.api_breaking.to_latex:
+
+DataFrame to LaTeX has a new render engine
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The existing :meth:`DataFrame.to_latex` has been restructured to utilise the
+extended implementation previously available under :meth:`.Styler.to_latex`.
+The arguments signature is similar, albeit ``col_space`` has been removed since
+it is ignored by LaTeX engines. This render engine also requires ``jinja2`` as a
+dependency which needs to be installed, since rendering is based upon jinja2 templates.
+
+The pandas latex options below are no longer used and have been removed. The generic
+max rows and columns arguments remain but for this functionality should be replaced
+by the Styler equivalents.
+The alternative options giving similar functionality are indicated below:
+
+- ``display.latex.escape``: replaced with ``styler.format.escape``,
+- ``display.latex.longtable``: replaced with ``styler.latex.environment``,
+- ``display.latex.multicolumn``, ``display.latex.multicolumn_format`` and
+  ``display.latex.multirow``: replaced with ``styler.sparse.rows``,
+  ``styler.sparse.columns``, ``styler.latex.multirow_align`` and
+  ``styler.latex.multicol_align``,
+- ``display.latex.repr``: replaced with ``styler.render.repr``,
+- ``display.max_rows`` and ``display.max_columns``: replace with
+  ``styler.render.max_rows``, ``styler.render.max_columns`` and
+  ``styler.render.max_elements``.
+
+Note that due to this change some defaults have also changed:
+
+- ``multirow`` now defaults to *True*.
+- ``multirow_align`` defaults to *"r"* instead of *"l"*.
+- ``multicol_align`` defaults to *"r"* instead of *"l"*.
+- ``escape`` now defaults to *False*.
+
+Note that the behaviour of ``_repr_latex_`` is also changed. Previously
+setting ``display.latex.repr`` would generate LaTeX only when using nbconvert for a
+JupyterNotebook, and not when the user is running the notebook. Now the
+``styler.render.repr`` option allows control of the specific output
+within JupyterNotebooks for operations (not just on nbconvert). See :issue:`39911`.
+
+.. _whatsnew_200.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++-------------------+-----------------+----------+---------+
+| Package           | Minimum Version | Required | Changed |
++===================+=================+==========+=========+
+| mypy (dev)        | 1.0             |          |    X    |
++-------------------+-----------------+----------+---------+
+| pytest (dev)      | 7.0.0           |          |    X    |
++-------------------+-----------------+----------+---------+
+| pytest-xdist (dev)| 2.2.0           |          |    X    |
++-------------------+-----------------+----------+---------+
+| hypothesis (dev)  | 6.34.2          |          |    X    |
++-------------------+-----------------+----------+---------+
+| python-dateutil   | 2.8.2           |    X     |    X    |
++-------------------+-----------------+----------+---------+
+| tzdata            | 2022.1          |    X     |    X    |
++-------------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++-----------------+-----------------+---------+
+| Package         | Minimum Version | Changed |
++=================+=================+=========+
+| pyarrow         | 7.0.0           |    X    |
++-----------------+-----------------+---------+
+| matplotlib      | 3.6.1           |    X    |
++-----------------+-----------------+---------+
+| fastparquet     | 0.6.3           |    X    |
++-----------------+-----------------+---------+
+| xarray          | 0.21.0          |    X    |
++-----------------+-----------------+---------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+Datetimes are now parsed with a consistent format
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In the past, :func:`to_datetime` guessed the format for each element independently. This was appropriate for some cases where elements had mixed date formats - however, it would regularly cause problems when users expected a consistent format but the function would switch formats between elements. As of version 2.0.0, parsing will use a consistent format, determined by the first non-NA value (unless the user specifies a format, in which case that is used).
+
+*Old behavior*:
+
+.. code-block:: ipython
+
+   In [1]: ser = pd.Series(['13-01-2000', '12-01-2000'])
+   In [2]: pd.to_datetime(ser)
+   Out[2]:
+   0   2000-01-13
+   1   2000-12-01
+   dtype: datetime64[ns]
+
+*New behavior*:
+
+.. ipython:: python
+    :okwarning:
+
+     ser = pd.Series(['13-01-2000', '12-01-2000'])
+     pd.to_datetime(ser)
+
+Note that this affects :func:`read_csv` as well.
+
+If you still need to parse dates with inconsistent formats, you can use
+``format='mixed'`` (possibly alongside ``dayfirst``) ::
+
+     ser = pd.Series(['13-01-2000', '12 January 2000'])
+     pd.to_datetime(ser, format='mixed', dayfirst=True)
+
+or, if your formats are all ISO8601 (but possibly not identically-formatted) ::
+
+     ser = pd.Series(['2020-01-01', '2020-01-01 03:00'])
+     pd.to_datetime(ser, format='ISO8601')
+
+.. _whatsnew_200.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- The ``tz``, ``nanosecond``, and ``unit`` keywords in the :class:`Timestamp` constructor are now keyword-only (:issue:`45307`, :issue:`32526`)
+- Passing ``nanoseconds`` greater than 999 or less than 0 in :class:`Timestamp` now raises a ``ValueError`` (:issue:`48538`, :issue:`48255`)
+- :func:`read_csv`: specifying an incorrect number of columns with ``index_col`` of now raises ``ParserError`` instead of ``IndexError`` when using the c parser.
+- Default value of ``dtype`` in :func:`get_dummies` is changed to ``bool`` from ``uint8`` (:issue:`45848`)
+- :meth:`DataFrame.astype`, :meth:`Series.astype`, and :meth:`DatetimeIndex.astype` casting datetime64 data to any of "datetime64[s]", "datetime64[ms]", "datetime64[us]" will return an object with the given resolution instead of coercing back to "datetime64[ns]" (:issue:`48928`)
+- :meth:`DataFrame.astype`, :meth:`Series.astype`, and :meth:`DatetimeIndex.astype` casting timedelta64 data to any of "timedelta64[s]", "timedelta64[ms]", "timedelta64[us]" will return an object with the given resolution instead of coercing to "float64" dtype (:issue:`48963`)
+- :meth:`DatetimeIndex.astype`, :meth:`TimedeltaIndex.astype`, :meth:`PeriodIndex.astype` :meth:`Series.astype`, :meth:`DataFrame.astype` with ``datetime64``, ``timedelta64`` or :class:`PeriodDtype` dtypes no longer allow converting to integer dtypes other than "int64", do ``obj.astype('int64', copy=False).astype(dtype)`` instead (:issue:`49715`)
+- :meth:`Index.astype` now allows casting from ``float64`` dtype to datetime-like dtypes, matching :class:`Series` behavior (:issue:`49660`)
+- Passing data with dtype of "timedelta64[s]", "timedelta64[ms]", or "timedelta64[us]" to :class:`TimedeltaIndex`, :class:`Series`, or :class:`DataFrame` constructors will now retain that dtype instead of casting to "timedelta64[ns]"; timedelta64 data with lower resolution will be cast to the lowest supported resolution "timedelta64[s]" (:issue:`49014`)
+- Passing ``dtype`` of "timedelta64[s]", "timedelta64[ms]", or "timedelta64[us]" to :class:`TimedeltaIndex`, :class:`Series`, or :class:`DataFrame` constructors will now retain that dtype instead of casting to "timedelta64[ns]"; passing a dtype with lower resolution for :class:`Series` or :class:`DataFrame` will be cast to the lowest supported resolution "timedelta64[s]" (:issue:`49014`)
+- Passing a ``np.datetime64`` object with non-nanosecond resolution to :class:`Timestamp` will retain the input resolution if it is "s", "ms", "us", or "ns"; otherwise it will be cast to the closest supported resolution (:issue:`49008`)
+- Passing ``datetime64`` values with resolution other than nanosecond to :func:`to_datetime` will retain the input resolution if it is "s", "ms", "us", or "ns"; otherwise it will be cast to the closest supported resolution (:issue:`50369`)
+- Passing integer values and a non-nanosecond datetime64 dtype (e.g. "datetime64[s]") :class:`DataFrame`, :class:`Series`, or :class:`Index` will treat the values as multiples of the dtype's unit, matching the behavior of e.g. ``Series(np.array(values, dtype="M8[s]"))`` (:issue:`51092`)
+- Passing a string in ISO-8601 format to :class:`Timestamp` will retain the resolution of the parsed input if it is "s", "ms", "us", or "ns"; otherwise it will be cast to the closest supported resolution (:issue:`49737`)
+- The ``other`` argument in :meth:`DataFrame.mask` and :meth:`Series.mask` now defaults to ``no_default`` instead of ``np.nan`` consistent with :meth:`DataFrame.where` and :meth:`Series.where`. Entries will be filled with the corresponding NULL value (``np.nan`` for numpy dtypes, ``pd.NA`` for extension dtypes). (:issue:`49111`)
+- Changed behavior of :meth:`Series.quantile` and :meth:`DataFrame.quantile` with :class:`SparseDtype` to retain sparse dtype (:issue:`49583`)
+- When creating a :class:`Series` with a object-dtype :class:`Index` of datetime objects, pandas no longer silently converts the index to a :class:`DatetimeIndex` (:issue:`39307`, :issue:`23598`)
+- :func:`pandas.testing.assert_index_equal` with parameter ``exact="equiv"`` now considers two indexes equal when both are either a :class:`RangeIndex` or :class:`Index` with an ``int64`` dtype. Previously it meant either a :class:`RangeIndex` or a :class:`Int64Index` (:issue:`51098`)
+- :meth:`Series.unique` with dtype "timedelta64[ns]" or "datetime64[ns]" now returns :class:`TimedeltaArray` or :class:`DatetimeArray` instead of ``numpy.ndarray`` (:issue:`49176`)
+- :func:`to_datetime` and :class:`DatetimeIndex` now allow sequences containing both ``datetime`` objects and numeric entries, matching :class:`Series` behavior (:issue:`49037`, :issue:`50453`)
+- :func:`pandas.api.types.is_string_dtype` now only returns ``True`` for array-likes with ``dtype=object`` when the elements are inferred to be strings (:issue:`15585`)
+- Passing a sequence containing ``datetime`` objects and ``date`` objects to :class:`Series` constructor will return with ``object`` dtype instead of ``datetime64[ns]`` dtype, consistent with :class:`Index` behavior (:issue:`49341`)
+- Passing strings that cannot be parsed as datetimes to :class:`Series` or :class:`DataFrame` with ``dtype="datetime64[ns]"`` will raise instead of silently ignoring the keyword and returning ``object`` dtype (:issue:`24435`)
+- Passing a sequence containing a type that cannot be converted to :class:`Timedelta` to :func:`to_timedelta` or to the :class:`Series` or :class:`DataFrame` constructor with ``dtype="timedelta64[ns]"`` or to :class:`TimedeltaIndex` now raises ``TypeError`` instead of ``ValueError`` (:issue:`49525`)
+- Changed behavior of :class:`Index` constructor with sequence containing at least one ``NaT`` and everything else either ``None`` or ``NaN`` to infer ``datetime64[ns]`` dtype instead of ``object``, matching :class:`Series` behavior (:issue:`49340`)
+- :func:`read_stata` with parameter ``index_col`` set to ``None`` (the default) will now set the index on the returned :class:`DataFrame` to a :class:`RangeIndex` instead of a :class:`Int64Index` (:issue:`49745`)
+- Changed behavior of :class:`Index`, :class:`Series`, and :class:`DataFrame` arithmetic methods when working with object-dtypes, the results no longer do type inference on the result of the array operations, use ``result.infer_objects(copy=False)`` to do type inference on the result (:issue:`49999`, :issue:`49714`)
+- Changed behavior of :class:`Index` constructor with an object-dtype ``numpy.ndarray`` containing all-``bool`` values or all-complex values, this will now retain object dtype, consistent with the :class:`Series` behavior (:issue:`49594`)
+- Changed behavior of :meth:`Series.astype` from object-dtype containing ``bytes`` objects to string dtypes; this now does ``val.decode()`` on bytes objects instead of ``str(val)``, matching :meth:`Index.astype` behavior (:issue:`45326`)
+- Added ``"None"`` to default ``na_values`` in :func:`read_csv` (:issue:`50286`)
+- Changed behavior of :class:`Series` and :class:`DataFrame` constructors when given an integer dtype and floating-point data that is not round numbers, this now raises ``ValueError`` instead of silently retaining the float dtype; do ``Series(data)`` or ``DataFrame(data)`` to get the old behavior, and ``Series(data).astype(dtype)`` or ``DataFrame(data).astype(dtype)`` to get the specified dtype (:issue:`49599`)
+- Changed behavior of :meth:`DataFrame.shift` with ``axis=1``, an integer ``fill_value``, and homogeneous datetime-like dtype, this now fills new columns with integer dtypes instead of casting to datetimelike (:issue:`49842`)
+- Files are now closed when encountering an exception in :func:`read_json` (:issue:`49921`)
+- Changed behavior of :func:`read_csv`, :func:`read_json` & :func:`read_fwf`, where the index will now always be a :class:`RangeIndex`, when no index is specified. Previously the index would be a :class:`Index` with dtype ``object`` if the new DataFrame/Series has length 0 (:issue:`49572`)
+- :meth:`DataFrame.values`, :meth:`DataFrame.to_numpy`, :meth:`DataFrame.xs`, :meth:`DataFrame.reindex`, :meth:`DataFrame.fillna`, and :meth:`DataFrame.replace` no longer silently consolidate the underlying arrays; do ``df = df.copy()`` to ensure consolidation (:issue:`49356`)
+- Creating a new DataFrame using a full slice on both axes with :attr:`~DataFrame.loc`
+  or :attr:`~DataFrame.iloc` (thus, ``df.loc[:, :]`` or ``df.iloc[:, :]``) now returns a
+  new DataFrame (shallow copy) instead of the original DataFrame, consistent with other
+  methods to get a full slice (for example ``df.loc[:]`` or ``df[:]``) (:issue:`49469`)
+- The :class:`Series` and :class:`DataFrame` constructors will now return a shallow copy
+  (i.e. share data, but not attributes) when passed a Series and DataFrame,
+  respectively, and with the default of ``copy=False`` (and if no other keyword triggers
+  a copy). Previously, the new Series or DataFrame would share the index attribute (e.g.
+  ``df.index = ...`` would also update the index of the parent or child) (:issue:`49523`)
+- Disallow computing ``cumprod`` for :class:`Timedelta` object; previously this returned incorrect values (:issue:`50246`)
+- :class:`DataFrame` objects read from a :class:`HDFStore` file without an index now have a :class:`RangeIndex` instead of an ``int64`` index (:issue:`51076`)
+- Instantiating an :class:`Index` with an numeric numpy dtype with data containing :class:`NA` and/or :class:`NaT` now raises a ``ValueError``. Previously a ``TypeError`` was raised (:issue:`51050`)
+- Loading a JSON file with duplicate columns using ``read_json(orient='split')`` renames columns to avoid duplicates, as :func:`read_csv` and the other readers do (:issue:`50370`)
+- The levels of the index of the :class:`Series` returned from ``Series.sparse.from_coo`` now always have dtype ``int32``. Previously they had dtype ``int64`` (:issue:`50926`)
+- :func:`to_datetime` with ``unit`` of either "Y" or "M" will now raise if a sequence contains a non-round ``float`` value, matching the ``Timestamp`` behavior (:issue:`50301`)
+- The methods :meth:`Series.round`, :meth:`DataFrame.__invert__`, :meth:`Series.__invert__`, :meth:`DataFrame.swapaxes`, :meth:`DataFrame.first`, :meth:`DataFrame.last`, :meth:`Series.first`, :meth:`Series.last` and :meth:`DataFrame.align` will now always return new objects (:issue:`51032`)
+- :class:`DataFrame` and :class:`DataFrameGroupBy` aggregations (e.g. "sum") with object-dtype columns no longer infer non-object dtypes for their results, explicitly call ``result.infer_objects(copy=False)`` on the result to obtain the old behavior (:issue:`51205`, :issue:`49603`)
+- Division by zero with :class:`ArrowDtype` dtypes returns ``-inf``, ``nan``, or ``inf`` depending on the numerator, instead of raising (:issue:`51541`)
+- Added :func:`pandas.api.types.is_any_real_numeric_dtype` to check for real numeric dtypes (:issue:`51152`)
+- :meth:`~arrays.ArrowExtensionArray.value_counts` now returns data with :class:`ArrowDtype` with ``pyarrow.int64`` type instead of ``"Int64"`` type (:issue:`51462`)
+- :func:`factorize` and :func:`unique` preserve the original dtype when passed numpy timedelta64 or datetime64 with non-nanosecond resolution (:issue:`48670`)
+
+.. note::
+
+    A current PDEP proposes the deprecation and removal of the keywords ``inplace`` and ``copy``
+    for all but a small subset of methods from the pandas API. The current discussion takes place
+    at `here <https://github.com/pandas-dev/pandas/pull/51466>`_. The keywords won't be necessary
+    anymore in the context of Copy-on-Write. If this proposal is accepted, both
+    keywords would be deprecated in the next release of pandas and removed in pandas 3.0.
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+- Deprecated parsing datetime strings with system-local timezone to ``tzlocal``, pass a ``tz`` keyword or explicitly call ``tz_localize`` instead (:issue:`50791`)
+- Deprecated argument ``infer_datetime_format`` in :func:`to_datetime` and :func:`read_csv`, as a strict version of it is now the default (:issue:`48621`)
+- Deprecated behavior of :func:`to_datetime` with ``unit`` when parsing strings, in a future version these will be parsed as datetimes (matching unit-less behavior) instead of cast to floats. To retain the old behavior, cast strings to numeric types before calling :func:`to_datetime` (:issue:`50735`)
+- Deprecated :func:`pandas.io.sql.execute` (:issue:`50185`)
+- :meth:`Index.is_boolean` has been deprecated. Use :func:`pandas.api.types.is_bool_dtype` instead (:issue:`50042`)
+- :meth:`Index.is_integer` has been deprecated. Use :func:`pandas.api.types.is_integer_dtype` instead (:issue:`50042`)
+- :meth:`Index.is_floating` has been deprecated. Use :func:`pandas.api.types.is_float_dtype` instead (:issue:`50042`)
+- :meth:`Index.holds_integer` has been deprecated. Use :func:`pandas.api.types.infer_dtype` instead (:issue:`50243`)
+- :meth:`Index.is_numeric` has been deprecated. Use :func:`pandas.api.types.is_any_real_numeric_dtype` instead (:issue:`50042`,:issue:`51152`)
+- :meth:`Index.is_categorical` has been deprecated. Use :func:`pandas.api.types.is_categorical_dtype` instead (:issue:`50042`)
+- :meth:`Index.is_object` has been deprecated. Use :func:`pandas.api.types.is_object_dtype` instead (:issue:`50042`)
+- :meth:`Index.is_interval` has been deprecated. Use :func:`pandas.api.types.is_interval_dtype` instead (:issue:`50042`)
+- Deprecated argument ``date_parser`` in :func:`read_csv`, :func:`read_table`, :func:`read_fwf`, and :func:`read_excel` in favour of ``date_format`` (:issue:`50601`)
+- Deprecated ``all`` and ``any`` reductions with ``datetime64`` and :class:`DatetimeTZDtype` dtypes, use e.g. ``(obj != pd.Timestamp(0), tz=obj.tz).all()`` instead (:issue:`34479`)
+- Deprecated unused arguments ``*args`` and ``**kwargs`` in :class:`Resampler` (:issue:`50977`)
+- Deprecated calling ``float`` or ``int`` on a single element :class:`Series` to return a ``float`` or ``int`` respectively. Extract the element before calling ``float`` or ``int`` instead (:issue:`51101`)
+- Deprecated :meth:`Grouper.groups`, use :meth:`Groupby.groups` instead (:issue:`51182`)
+- Deprecated :meth:`Grouper.grouper`, use :meth:`Groupby.grouper` instead (:issue:`51182`)
+- Deprecated :meth:`Grouper.obj`, use :meth:`Groupby.obj` instead (:issue:`51206`)
+- Deprecated :meth:`Grouper.indexer`, use :meth:`Resampler.indexer` instead (:issue:`51206`)
+- Deprecated :meth:`Grouper.ax`, use :meth:`Resampler.ax` instead (:issue:`51206`)
+- Deprecated keyword ``use_nullable_dtypes`` in :func:`read_parquet`, use ``dtype_backend`` instead (:issue:`51853`)
+- Deprecated :meth:`Series.pad` in favor of :meth:`Series.ffill` (:issue:`33396`)
+- Deprecated :meth:`Series.backfill` in favor of :meth:`Series.bfill` (:issue:`33396`)
+- Deprecated :meth:`DataFrame.pad` in favor of :meth:`DataFrame.ffill` (:issue:`33396`)
+- Deprecated :meth:`DataFrame.backfill` in favor of :meth:`DataFrame.bfill` (:issue:`33396`)
+- Deprecated :meth:`~pandas.io.stata.StataReader.close`. Use :class:`~pandas.io.stata.StataReader` as a context manager instead (:issue:`49228`)
+- Deprecated producing a scalar when iterating over a :class:`.DataFrameGroupBy` or a :class:`.SeriesGroupBy` that has been grouped by a ``level`` parameter that is a list of length 1; a tuple of length one will be returned instead (:issue:`51583`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+- Removed :class:`Int64Index`, :class:`UInt64Index` and :class:`Float64Index`. See also :ref:`here <whatsnew_200.enhancements.index_can_hold_numpy_numeric_dtypes>` for more information (:issue:`42717`)
+- Removed deprecated :attr:`Timestamp.freq`, :attr:`Timestamp.freqstr` and argument ``freq`` from the :class:`Timestamp` constructor and :meth:`Timestamp.fromordinal` (:issue:`14146`)
+- Removed deprecated :class:`CategoricalBlock`, :meth:`Block.is_categorical`, require datetime64 and timedelta64 values to be wrapped in :class:`DatetimeArray` or :class:`TimedeltaArray` before passing to :meth:`Block.make_block_same_class`, require ``DatetimeTZBlock.values`` to have the correct ndim when passing to the :class:`BlockManager` constructor, and removed the "fastpath" keyword from the :class:`SingleBlockManager` constructor (:issue:`40226`, :issue:`40571`)
+- Removed deprecated global option ``use_inf_as_null`` in favor of ``use_inf_as_na`` (:issue:`17126`)
+- Removed deprecated module ``pandas.core.index`` (:issue:`30193`)
+- Removed deprecated alias ``pandas.core.tools.datetimes.to_time``, import the function directly from ``pandas.core.tools.times`` instead (:issue:`34145`)
+- Removed deprecated alias ``pandas.io.json.json_normalize``, import the function directly from ``pandas.json_normalize`` instead (:issue:`27615`)
+- Removed deprecated :meth:`Categorical.to_dense`, use ``np.asarray(cat)`` instead (:issue:`32639`)
+- Removed deprecated :meth:`Categorical.take_nd` (:issue:`27745`)
+- Removed deprecated :meth:`Categorical.mode`, use ``Series(cat).mode()`` instead (:issue:`45033`)
+- Removed deprecated :meth:`Categorical.is_dtype_equal` and :meth:`CategoricalIndex.is_dtype_equal` (:issue:`37545`)
+- Removed deprecated :meth:`CategoricalIndex.take_nd` (:issue:`30702`)
+- Removed deprecated :meth:`Index.is_type_compatible` (:issue:`42113`)
+- Removed deprecated :meth:`Index.is_mixed`, check ``index.inferred_type`` directly instead (:issue:`32922`)
+- Removed deprecated :func:`pandas.api.types.is_categorical`; use :func:`pandas.api.types.is_categorical_dtype` instead  (:issue:`33385`)
+- Removed deprecated :meth:`Index.asi8` (:issue:`37877`)
+- Enforced deprecation changing behavior when passing ``datetime64[ns]`` dtype data and timezone-aware dtype to :class:`Series`, interpreting the values as wall-times instead of UTC times, matching :class:`DatetimeIndex` behavior (:issue:`41662`)
+- Enforced deprecation changing behavior when applying a numpy ufunc on multiple non-aligned (on the index or columns) :class:`DataFrame` that will now align the inputs first (:issue:`39239`)
+- Removed deprecated :meth:`DataFrame._AXIS_NUMBERS`, :meth:`DataFrame._AXIS_NAMES`, :meth:`Series._AXIS_NUMBERS`, :meth:`Series._AXIS_NAMES` (:issue:`33637`)
+- Removed deprecated :meth:`Index.to_native_types`, use ``obj.astype(str)`` instead (:issue:`36418`)
+- Removed deprecated :meth:`Series.iteritems`, :meth:`DataFrame.iteritems`, use ``obj.items`` instead (:issue:`45321`)
+- Removed deprecated :meth:`DataFrame.lookup` (:issue:`35224`)
+- Removed deprecated :meth:`Series.append`, :meth:`DataFrame.append`, use :func:`concat` instead (:issue:`35407`)
+- Removed deprecated :meth:`Series.iteritems`, :meth:`DataFrame.iteritems` and :meth:`HDFStore.iteritems` use ``obj.items`` instead (:issue:`45321`)
+- Removed deprecated :meth:`DatetimeIndex.union_many` (:issue:`45018`)
+- Removed deprecated ``weekofyear`` and ``week`` attributes of :class:`DatetimeArray`, :class:`DatetimeIndex` and ``dt`` accessor in favor of ``isocalendar().week`` (:issue:`33595`)
+- Removed deprecated :meth:`RangeIndex._start`, :meth:`RangeIndex._stop`, :meth:`RangeIndex._step`, use ``start``, ``stop``, ``step`` instead (:issue:`30482`)
+- Removed deprecated :meth:`DatetimeIndex.to_perioddelta`, Use ``dtindex - dtindex.to_period(freq).to_timestamp()`` instead (:issue:`34853`)
+- Removed deprecated :meth:`.Styler.hide_index` and :meth:`.Styler.hide_columns` (:issue:`49397`)
+- Removed deprecated :meth:`.Styler.set_na_rep` and :meth:`.Styler.set_precision` (:issue:`49397`)
+- Removed deprecated :meth:`.Styler.where` (:issue:`49397`)
+- Removed deprecated :meth:`.Styler.render` (:issue:`49397`)
+- Removed deprecated argument ``col_space`` in :meth:`DataFrame.to_latex` (:issue:`47970`)
+- Removed deprecated argument ``null_color`` in :meth:`.Styler.highlight_null` (:issue:`49397`)
+- Removed deprecated argument ``check_less_precise`` in :meth:`.testing.assert_frame_equal`, :meth:`.testing.assert_extension_array_equal`, :meth:`.testing.assert_series_equal`,  :meth:`.testing.assert_index_equal` (:issue:`30562`)
+- Removed deprecated ``null_counts`` argument in :meth:`DataFrame.info`. Use ``show_counts`` instead (:issue:`37999`)
+- Removed deprecated :meth:`Index.is_monotonic`, and :meth:`Series.is_monotonic`; use ``obj.is_monotonic_increasing`` instead (:issue:`45422`)
+- Removed deprecated :meth:`Index.is_all_dates` (:issue:`36697`)
+- Enforced deprecation disallowing passing a timezone-aware :class:`Timestamp` and ``dtype="datetime64[ns]"`` to :class:`Series` or :class:`DataFrame` constructors (:issue:`41555`)
+- Enforced deprecation disallowing passing a sequence of timezone-aware values and ``dtype="datetime64[ns]"`` to to :class:`Series` or :class:`DataFrame` constructors (:issue:`41555`)
+- Enforced deprecation disallowing ``numpy.ma.mrecords.MaskedRecords`` in the :class:`DataFrame` constructor; pass ``"{name: data[name] for name in data.dtype.names}`` instead (:issue:`40363`)
+- Enforced deprecation disallowing unit-less "datetime64" dtype in :meth:`Series.astype` and :meth:`DataFrame.astype` (:issue:`47844`)
+- Enforced deprecation disallowing using ``.astype`` to convert a ``datetime64[ns]`` :class:`Series`, :class:`DataFrame`, or :class:`DatetimeIndex` to timezone-aware dtype, use ``obj.tz_localize`` or ``ser.dt.tz_localize`` instead (:issue:`39258`)
+- Enforced deprecation disallowing using ``.astype`` to convert a timezone-aware :class:`Series`, :class:`DataFrame`, or :class:`DatetimeIndex` to timezone-naive ``datetime64[ns]`` dtype, use ``obj.tz_localize(None)`` or ``obj.tz_convert("UTC").tz_localize(None)`` instead (:issue:`39258`)
+- Enforced deprecation disallowing passing non boolean argument to sort in :func:`concat` (:issue:`44629`)
+- Removed Date parser functions :func:`~pandas.io.date_converters.parse_date_time`,
+  :func:`~pandas.io.date_converters.parse_date_fields`, :func:`~pandas.io.date_converters.parse_all_fields`
+  and :func:`~pandas.io.date_converters.generic_parser` (:issue:`24518`)
+- Removed argument ``index`` from the :class:`core.arrays.SparseArray` constructor (:issue:`43523`)
+- Remove argument ``squeeze`` from :meth:`DataFrame.groupby` and :meth:`Series.groupby` (:issue:`32380`)
+- Removed deprecated ``apply``, ``apply_index``, ``__call__``, ``onOffset``, and ``isAnchored`` attributes from :class:`DateOffset` (:issue:`34171`)
+- Removed ``keep_tz`` argument in :meth:`DatetimeIndex.to_series` (:issue:`29731`)
+- Remove arguments ``names`` and ``dtype`` from :meth:`Index.copy` and ``levels`` and ``codes`` from :meth:`MultiIndex.copy` (:issue:`35853`, :issue:`36685`)
+- Remove argument ``inplace`` from :meth:`MultiIndex.set_levels` and :meth:`MultiIndex.set_codes` (:issue:`35626`)
+- Removed arguments ``verbose`` and ``encoding`` from :meth:`DataFrame.to_excel` and :meth:`Series.to_excel` (:issue:`47912`)
+- Removed argument ``line_terminator`` from :meth:`DataFrame.to_csv` and :meth:`Series.to_csv`, use ``lineterminator`` instead (:issue:`45302`)
+- Removed argument ``inplace`` from :meth:`DataFrame.set_axis` and :meth:`Series.set_axis`, use ``obj = obj.set_axis(..., copy=False)`` instead (:issue:`48130`)
+- Disallow passing positional arguments to :meth:`MultiIndex.set_levels` and :meth:`MultiIndex.set_codes` (:issue:`41485`)
+- Disallow parsing to Timedelta strings with components with units "Y", "y", or "M", as these do not represent unambiguous durations (:issue:`36838`)
+- Removed :meth:`MultiIndex.is_lexsorted` and :meth:`MultiIndex.lexsort_depth` (:issue:`38701`)
+- Removed argument ``how`` from :meth:`PeriodIndex.astype`, use :meth:`PeriodIndex.to_timestamp` instead (:issue:`37982`)
+- Removed argument ``try_cast`` from :meth:`DataFrame.mask`, :meth:`DataFrame.where`, :meth:`Series.mask` and :meth:`Series.where` (:issue:`38836`)
+- Removed argument ``tz`` from :meth:`Period.to_timestamp`, use ``obj.to_timestamp(...).tz_localize(tz)`` instead (:issue:`34522`)
+- Removed argument ``sort_columns`` in :meth:`DataFrame.plot` and :meth:`Series.plot` (:issue:`47563`)
+- Removed argument ``is_copy`` from :meth:`DataFrame.take` and :meth:`Series.take` (:issue:`30615`)
+- Removed argument ``kind`` from :meth:`Index.get_slice_bound`, :meth:`Index.slice_indexer` and :meth:`Index.slice_locs` (:issue:`41378`)
+- Removed arguments ``prefix``, ``squeeze``, ``error_bad_lines`` and ``warn_bad_lines`` from :func:`read_csv` (:issue:`40413`, :issue:`43427`)
+- Removed arguments ``squeeze`` from :func:`read_excel` (:issue:`43427`)
+- Removed argument ``datetime_is_numeric`` from :meth:`DataFrame.describe` and :meth:`Series.describe` as datetime data will always be summarized as numeric data (:issue:`34798`)
+- Disallow passing list ``key`` to :meth:`Series.xs` and :meth:`DataFrame.xs`, pass a tuple instead (:issue:`41789`)
+- Disallow subclass-specific keywords (e.g. "freq", "tz", "names", "closed") in the :class:`Index` constructor (:issue:`38597`)
+- Removed argument ``inplace`` from :meth:`Categorical.remove_unused_categories` (:issue:`37918`)
+- Disallow passing non-round floats to :class:`Timestamp` with ``unit="M"`` or ``unit="Y"`` (:issue:`47266`)
+- Remove keywords ``convert_float`` and ``mangle_dupe_cols`` from :func:`read_excel` (:issue:`41176`)
+- Remove keyword ``mangle_dupe_cols`` from :func:`read_csv` and :func:`read_table` (:issue:`48137`)
+- Removed ``errors`` keyword from :meth:`DataFrame.where`, :meth:`Series.where`, :meth:`DataFrame.mask` and :meth:`Series.mask` (:issue:`47728`)
+- Disallow passing non-keyword arguments to :func:`read_excel` except ``io`` and ``sheet_name`` (:issue:`34418`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.drop` and :meth:`Series.drop` except ``labels`` (:issue:`41486`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.fillna` and :meth:`Series.fillna` except ``value`` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :meth:`StringMethods.split` and :meth:`StringMethods.rsplit` except for ``pat`` (:issue:`47448`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.set_index` except ``keys`` (:issue:`41495`)
+- Disallow passing non-keyword arguments to :meth:`Resampler.interpolate` except ``method`` (:issue:`41699`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.reset_index` and :meth:`Series.reset_index` except ``level`` (:issue:`41496`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.dropna` and :meth:`Series.dropna` (:issue:`41504`)
+- Disallow passing non-keyword arguments to :meth:`ExtensionArray.argsort` (:issue:`46134`)
+- Disallow passing non-keyword arguments to :meth:`Categorical.sort_values` (:issue:`47618`)
+- Disallow passing non-keyword arguments to :meth:`Index.drop_duplicates` and :meth:`Series.drop_duplicates` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.drop_duplicates` except for ``subset`` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.sort_index` and :meth:`Series.sort_index` (:issue:`41506`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.interpolate` and :meth:`Series.interpolate` except for ``method`` (:issue:`41510`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.any` and :meth:`Series.any` (:issue:`44896`)
+- Disallow passing non-keyword arguments to :meth:`Index.set_names` except for ``names`` (:issue:`41551`)
+- Disallow passing non-keyword arguments to :meth:`Index.join` except for ``other`` (:issue:`46518`)
+- Disallow passing non-keyword arguments to :func:`concat` except for ``objs`` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :func:`pivot` except for ``data`` (:issue:`48301`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.pivot` (:issue:`48301`)
+- Disallow passing non-keyword arguments to :func:`read_html` except for ``io`` (:issue:`27573`)
+- Disallow passing non-keyword arguments to :func:`read_json` except for ``path_or_buf`` (:issue:`27573`)
+- Disallow passing non-keyword arguments to :func:`read_sas` except for ``filepath_or_buffer`` (:issue:`47154`)
+- Disallow passing non-keyword arguments to :func:`read_stata` except for ``filepath_or_buffer`` (:issue:`48128`)
+- Disallow passing non-keyword arguments to :func:`read_csv` except ``filepath_or_buffer`` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :func:`read_table` except ``filepath_or_buffer`` (:issue:`41485`)
+- Disallow passing non-keyword arguments to :func:`read_fwf` except ``filepath_or_buffer`` (:issue:`44710`)
+- Disallow passing non-keyword arguments to :func:`read_xml` except for ``path_or_buffer`` (:issue:`45133`)
+- Disallow passing non-keyword arguments to :meth:`Series.mask` and :meth:`DataFrame.mask` except ``cond`` and ``other`` (:issue:`41580`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.to_stata` except for ``path`` (:issue:`48128`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.where` and :meth:`Series.where` except for ``cond`` and ``other`` (:issue:`41523`)
+- Disallow passing non-keyword arguments to :meth:`Series.set_axis` and :meth:`DataFrame.set_axis` except for ``labels`` (:issue:`41491`)
+- Disallow passing non-keyword arguments to :meth:`Series.rename_axis` and :meth:`DataFrame.rename_axis` except for ``mapper`` (:issue:`47587`)
+- Disallow passing non-keyword arguments to :meth:`Series.clip` and :meth:`DataFrame.clip` except ``lower`` and ``upper`` (:issue:`41511`)
+- Disallow passing non-keyword arguments to :meth:`Series.bfill`, :meth:`Series.ffill`, :meth:`DataFrame.bfill` and :meth:`DataFrame.ffill` (:issue:`41508`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.replace`, :meth:`Series.replace` except for ``to_replace`` and ``value`` (:issue:`47587`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.sort_values` except for ``by`` (:issue:`41505`)
+- Disallow passing non-keyword arguments to :meth:`Series.sort_values` (:issue:`41505`)
+- Disallow passing non-keyword arguments to :meth:`DataFrame.reindex` except for ``labels`` (:issue:`17966`)
+- Disallow :meth:`Index.reindex` with non-unique :class:`Index` objects (:issue:`42568`)
+- Disallowed constructing :class:`Categorical` with scalar ``data`` (:issue:`38433`)
+- Disallowed constructing :class:`CategoricalIndex` without passing ``data`` (:issue:`38944`)
+- Removed :meth:`.Rolling.validate`, :meth:`.Expanding.validate`, and :meth:`.ExponentialMovingWindow.validate` (:issue:`43665`)
+- Removed :attr:`Rolling.win_type` returning ``"freq"`` (:issue:`38963`)
+- Removed :attr:`Rolling.is_datetimelike` (:issue:`38963`)
+- Removed the ``level`` keyword in :class:`DataFrame` and :class:`Series` aggregations; use ``groupby`` instead (:issue:`39983`)
+- Removed deprecated :meth:`Timedelta.delta`, :meth:`Timedelta.is_populated`, and :attr:`Timedelta.freq` (:issue:`46430`, :issue:`46476`)
+- Removed deprecated :attr:`NaT.freq` (:issue:`45071`)
+- Removed deprecated :meth:`Categorical.replace`, use :meth:`Series.replace` instead (:issue:`44929`)
+- Removed the ``numeric_only`` keyword from :meth:`Categorical.min` and :meth:`Categorical.max` in favor of ``skipna`` (:issue:`48821`)
+- Changed behavior of :meth:`DataFrame.median` and :meth:`DataFrame.mean` with ``numeric_only=None`` to not exclude datetime-like columns THIS NOTE WILL BE IRRELEVANT ONCE ``numeric_only=None`` DEPRECATION IS ENFORCED (:issue:`29941`)
+- Removed :func:`is_extension_type` in favor of :func:`is_extension_array_dtype` (:issue:`29457`)
+- Removed ``.ExponentialMovingWindow.vol`` (:issue:`39220`)
+- Removed :meth:`Index.get_value` and :meth:`Index.set_value` (:issue:`33907`, :issue:`28621`)
+- Removed :meth:`Series.slice_shift` and :meth:`DataFrame.slice_shift` (:issue:`37601`)
+- Remove :meth:`DataFrameGroupBy.pad` and :meth:`DataFrameGroupBy.backfill` (:issue:`45076`)
+- Remove ``numpy`` argument from :func:`read_json` (:issue:`30636`)
+- Disallow passing abbreviations for ``orient`` in :meth:`DataFrame.to_dict` (:issue:`32516`)
+- Disallow partial slicing on an non-monotonic :class:`DatetimeIndex` with keys which are not in Index. This now raises a ``KeyError`` (:issue:`18531`)
+- Removed ``get_offset`` in favor of :func:`to_offset` (:issue:`30340`)
+- Removed the ``warn`` keyword in :func:`infer_freq` (:issue:`45947`)
+- Removed the ``include_start`` and ``include_end`` arguments in :meth:`DataFrame.between_time` in favor of ``inclusive`` (:issue:`43248`)
+- Removed the ``closed`` argument in :meth:`date_range` and :meth:`bdate_range` in favor of ``inclusive`` argument (:issue:`40245`)
+- Removed the ``center`` keyword in :meth:`DataFrame.expanding` (:issue:`20647`)
+- Removed the ``truediv`` keyword from :func:`eval` (:issue:`29812`)
+- Removed the ``method`` and ``tolerance`` arguments in :meth:`Index.get_loc`. Use ``index.get_indexer([label], method=..., tolerance=...)`` instead (:issue:`42269`)
+- Removed the ``pandas.datetime`` submodule (:issue:`30489`)
+- Removed the ``pandas.np`` submodule (:issue:`30296`)
+- Removed ``pandas.util.testing`` in favor of ``pandas.testing`` (:issue:`30745`)
+- Removed :meth:`Series.str.__iter__` (:issue:`28277`)
+- Removed ``pandas.SparseArray`` in favor of :class:`arrays.SparseArray` (:issue:`30642`)
+- Removed ``pandas.SparseSeries`` and ``pandas.SparseDataFrame``, including pickle support. (:issue:`30642`)
+- Enforced disallowing passing an integer ``fill_value`` to :meth:`DataFrame.shift` and :meth:`Series.shift`` with datetime64, timedelta64, or period dtypes (:issue:`32591`)
+- Enforced disallowing a string column label into ``times`` in :meth:`DataFrame.ewm` (:issue:`43265`)
+- Enforced disallowing passing ``True`` and ``False`` into ``inclusive`` in :meth:`Series.between` in favor of ``"both"`` and ``"neither"`` respectively (:issue:`40628`)
+- Enforced disallowing using ``usecols`` with out of bounds indices for ``read_csv`` with ``engine="c"`` (:issue:`25623`)
+- Enforced disallowing the use of ``**kwargs`` in :class:`.ExcelWriter`; use the keyword argument ``engine_kwargs`` instead (:issue:`40430`)
+- Enforced disallowing a tuple of column labels into :meth:`.DataFrameGroupBy.__getitem__` (:issue:`30546`)
+- Enforced disallowing missing labels when indexing with a sequence of labels on a level of a :class:`MultiIndex`. This now raises a ``KeyError`` (:issue:`42351`)
+- Enforced disallowing setting values with ``.loc`` using a positional slice. Use ``.loc`` with labels or ``.iloc`` with positions instead (:issue:`31840`)
+- Enforced disallowing positional indexing with a ``float`` key even if that key is a round number, manually cast to integer instead (:issue:`34193`)
+- Enforced disallowing using a :class:`DataFrame` indexer with ``.iloc``, use ``.loc`` instead for automatic alignment (:issue:`39022`)
+- Enforced disallowing ``set`` or ``dict`` indexers in ``__getitem__`` and ``__setitem__`` methods (:issue:`42825`)
+- Enforced disallowing indexing on a :class:`Index` or positional indexing on a :class:`Series` producing multi-dimensional objects e.g. ``obj[:, None]``, convert to numpy before indexing instead (:issue:`35141`)
+- Enforced disallowing ``dict`` or ``set`` objects in ``suffixes`` in :func:`merge` (:issue:`34810`)
+- Enforced disallowing :func:`merge` to produce duplicated columns through the ``suffixes`` keyword and already existing columns (:issue:`22818`)
+- Enforced disallowing using :func:`merge` or :func:`join` on a different number of levels (:issue:`34862`)
+- Enforced disallowing ``value_name`` argument in :func:`DataFrame.melt` to match an element in the :class:`DataFrame` columns (:issue:`35003`)
+- Enforced disallowing passing ``showindex`` into ``**kwargs`` in :func:`DataFrame.to_markdown` and :func:`Series.to_markdown` in favor of ``index`` (:issue:`33091`)
+- Removed setting Categorical._codes directly (:issue:`41429`)
+- Removed setting Categorical.categories directly (:issue:`47834`)
+- Removed argument ``inplace`` from :meth:`Categorical.add_categories`, :meth:`Categorical.remove_categories`, :meth:`Categorical.set_categories`, :meth:`Categorical.rename_categories`, :meth:`Categorical.reorder_categories`, :meth:`Categorical.set_ordered`, :meth:`Categorical.as_ordered`, :meth:`Categorical.as_unordered` (:issue:`37981`, :issue:`41118`, :issue:`41133`, :issue:`47834`)
+- Enforced :meth:`Rolling.count` with ``min_periods=None`` to default to the size of the window (:issue:`31302`)
+- Renamed ``fname`` to ``path`` in :meth:`DataFrame.to_parquet`, :meth:`DataFrame.to_stata` and :meth:`DataFrame.to_feather` (:issue:`30338`)
+- Enforced disallowing indexing a :class:`Series` with a single item list with a slice (e.g. ``ser[[slice(0, 2)]]``). Either convert the list to tuple, or pass the slice directly instead (:issue:`31333`)
+- Changed behavior indexing on a :class:`DataFrame` with a :class:`DatetimeIndex` index using a string indexer, previously this operated as a slice on rows, now it operates like any other column key; use ``frame.loc[key]`` for the old behavior (:issue:`36179`)
+- Enforced the ``display.max_colwidth`` option to not accept negative integers (:issue:`31569`)
+- Removed the ``display.column_space`` option in favor of ``df.to_string(col_space=...)`` (:issue:`47280`)
+- Removed the deprecated method ``mad`` from pandas classes (:issue:`11787`)
+- Removed the deprecated method ``tshift`` from pandas classes (:issue:`11631`)
+- Changed behavior of empty data passed into :class:`Series`; the default dtype will be ``object`` instead of ``float64`` (:issue:`29405`)
+- Changed the behavior of :meth:`DatetimeIndex.union`, :meth:`DatetimeIndex.intersection`, and :meth:`DatetimeIndex.symmetric_difference` with mismatched timezones to convert to UTC instead of casting to object dtype (:issue:`39328`)
+- Changed the behavior of :func:`to_datetime` with argument "now" with ``utc=False`` to match ``Timestamp("now")`` (:issue:`18705`)
+- Changed the behavior of indexing on a timezone-aware :class:`DatetimeIndex` with a timezone-naive ``datetime`` object or vice-versa; these now behave like any other non-comparable type by raising ``KeyError`` (:issue:`36148`)
+- Changed the behavior of :meth:`Index.reindex`, :meth:`Series.reindex`, and :meth:`DataFrame.reindex` with a ``datetime64`` dtype and a ``datetime.date`` object for ``fill_value``; these are no longer considered equivalent to ``datetime.datetime`` objects so the reindex casts to object dtype (:issue:`39767`)
+- Changed behavior of :meth:`SparseArray.astype` when given a dtype that is not explicitly ``SparseDtype``, cast to the exact requested dtype rather than silently using a ``SparseDtype`` instead (:issue:`34457`)
+- Changed behavior of :meth:`Index.ravel` to return a view on the original :class:`Index` instead of a ``np.ndarray`` (:issue:`36900`)
+- Changed behavior of :meth:`Series.to_frame` and :meth:`Index.to_frame` with explicit ``name=None`` to use ``None`` for the column name instead of the index's name or default ``0`` (:issue:`45523`)
+- Changed behavior of :func:`concat` with one array of ``bool``-dtype and another of integer dtype, this now returns ``object`` dtype instead of integer dtype; explicitly cast the bool object to integer before concatenating to get the old behavior (:issue:`45101`)
+- Changed behavior of :class:`DataFrame` constructor given floating-point ``data`` and an integer ``dtype``, when the data cannot be cast losslessly, the floating point dtype is retained, matching :class:`Series` behavior (:issue:`41170`)
+- Changed behavior of :class:`Index` constructor when given a ``np.ndarray`` with object-dtype containing numeric entries; this now retains object dtype rather than inferring a numeric dtype, consistent with :class:`Series` behavior (:issue:`42870`)
+- Changed behavior of :meth:`Index.__and__`, :meth:`Index.__or__` and :meth:`Index.__xor__` to behave as logical operations (matching :class:`Series` behavior) instead of aliases for set operations (:issue:`37374`)
+- Changed behavior of :class:`DataFrame` constructor when passed a list whose first element is a :class:`Categorical`, this now treats the elements as rows casting to ``object`` dtype, consistent with behavior for other types (:issue:`38845`)
+- Changed behavior of :class:`DataFrame` constructor when passed a ``dtype`` (other than int) that the data cannot be cast to; it now raises instead of silently ignoring the dtype (:issue:`41733`)
+- Changed the behavior of :class:`Series` constructor, it will no longer infer a datetime64 or timedelta64 dtype from string entries (:issue:`41731`)
+- Changed behavior of :class:`Timestamp` constructor with a ``np.datetime64`` object and a ``tz`` passed to interpret the input as a wall-time as opposed to a UTC time (:issue:`42288`)
+- Changed behavior of :meth:`Timestamp.utcfromtimestamp` to return a timezone-aware object satisfying ``Timestamp.utcfromtimestamp(val).timestamp() == val`` (:issue:`45083`)
+- Changed behavior of :class:`Index` constructor when passed a ``SparseArray`` or ``SparseDtype`` to retain that dtype instead of casting to ``numpy.ndarray`` (:issue:`43930`)
+- Changed behavior of setitem-like operations (``__setitem__``, ``fillna``, ``where``, ``mask``, ``replace``, ``insert``, fill_value for ``shift``) on an object with :class:`DatetimeTZDtype` when using a value with a non-matching timezone, the value will be cast to the object's timezone instead of casting both to object-dtype (:issue:`44243`)
+- Changed behavior of :class:`Index`, :class:`Series`, :class:`DataFrame` constructors with floating-dtype data and a :class:`DatetimeTZDtype`, the data are now interpreted as UTC-times instead of wall-times, consistent with how integer-dtype data are treated (:issue:`45573`)
+- Changed behavior of :class:`Series` and :class:`DataFrame` constructors with integer dtype and floating-point data containing ``NaN``, this now raises ``IntCastingNaNError`` (:issue:`40110`)
+- Changed behavior of :class:`Series` and :class:`DataFrame` constructors with an integer ``dtype`` and values that are too large to losslessly cast to this dtype, this now raises ``ValueError`` (:issue:`41734`)
+- Changed behavior of :class:`Series` and :class:`DataFrame` constructors with an integer ``dtype`` and values having either ``datetime64`` or ``timedelta64`` dtypes, this now raises ``TypeError``, use ``values.view("int64")`` instead (:issue:`41770`)
+- Removed the deprecated ``base`` and ``loffset`` arguments from :meth:`pandas.DataFrame.resample`, :meth:`pandas.Series.resample` and :class:`pandas.Grouper`. Use ``offset`` or ``origin`` instead (:issue:`31809`)
+- Changed behavior of :meth:`Series.fillna` and :meth:`DataFrame.fillna` with ``timedelta64[ns]`` dtype and an incompatible ``fill_value``; this now casts to ``object`` dtype instead of raising, consistent with the behavior with other dtypes (:issue:`45746`)
+- Change the default argument of ``regex`` for :meth:`Series.str.replace` from ``True`` to ``False``. Additionally, a single character ``pat`` with ``regex=True`` is now treated as a regular expression instead of a string literal. (:issue:`36695`, :issue:`24804`)
+- Changed behavior of :meth:`DataFrame.any` and :meth:`DataFrame.all` with ``bool_only=True``; object-dtype columns with all-bool values will no longer be included, manually cast to ``bool`` dtype first (:issue:`46188`)
+- Changed behavior of :meth:`DataFrame.max`, :class:`DataFrame.min`, :class:`DataFrame.mean`, :class:`DataFrame.median`, :class:`DataFrame.skew`, :class:`DataFrame.kurt` with ``axis=None`` to return a scalar applying the aggregation across both axes (:issue:`45072`)
+- Changed behavior of comparison of a :class:`Timestamp` with a ``datetime.date`` object; these now compare as un-equal and raise on inequality comparisons, matching the ``datetime.datetime`` behavior (:issue:`36131`)
+- Changed behavior of comparison of ``NaT`` with a ``datetime.date`` object; these now raise on inequality comparisons (:issue:`39196`)
+- Enforced deprecation of silently dropping columns that raised a ``TypeError`` in :class:`Series.transform` and :class:`DataFrame.transform` when used with a list or dictionary (:issue:`43740`)
+- Changed behavior of :meth:`DataFrame.apply` with list-like so that any partial failure will raise an error (:issue:`43740`)
+- Changed behaviour of :meth:`DataFrame.to_latex` to now use the Styler implementation via :meth:`.Styler.to_latex` (:issue:`47970`)
+- Changed behavior of :meth:`Series.__setitem__` with an integer key and a :class:`Float64Index` when the key is not present in the index; previously we treated the key as positional (behaving like ``series.iloc[key] = val``), now we treat it is a label (behaving like ``series.loc[key] = val``), consistent with :meth:`Series.__getitem__`` behavior (:issue:`33469`)
+- Removed ``na_sentinel`` argument from :func:`factorize`, :meth:`.Index.factorize`, and :meth:`.ExtensionArray.factorize` (:issue:`47157`)
+- Changed behavior of :meth:`Series.diff` and :meth:`DataFrame.diff` with :class:`ExtensionDtype` dtypes whose arrays do not implement ``diff``, these now raise ``TypeError`` rather than casting to numpy (:issue:`31025`)
+- Enforced deprecation of calling numpy "ufunc"s on :class:`DataFrame` with ``method="outer"``; this now raises ``NotImplementedError`` (:issue:`36955`)
+- Enforced deprecation disallowing passing ``numeric_only=True`` to :class:`Series` reductions (``rank``, ``any``, ``all``, ...) with non-numeric dtype (:issue:`47500`)
+- Changed behavior of :meth:`.DataFrameGroupBy.apply` and :meth:`.SeriesGroupBy.apply` so that ``group_keys`` is respected even if a transformer is detected (:issue:`34998`)
+- Comparisons between a :class:`DataFrame` and a :class:`Series` where the frame's columns do not match the series's index raise ``ValueError`` instead of automatically aligning, do ``left, right = left.align(right, axis=1, copy=False)`` before comparing (:issue:`36795`)
+- Enforced deprecation ``numeric_only=None`` (the default) in DataFrame reductions that would silently drop columns that raised; ``numeric_only`` now defaults to ``False`` (:issue:`41480`)
+- Changed default of ``numeric_only`` to ``False`` in all DataFrame methods with that argument (:issue:`46096`, :issue:`46906`)
+- Changed default of ``numeric_only`` to ``False`` in :meth:`Series.rank` (:issue:`47561`)
+- Enforced deprecation of silently dropping nuisance columns in groupby and resample operations when ``numeric_only=False`` (:issue:`41475`)
+- Enforced deprecation of silently dropping nuisance columns in :class:`Rolling`, :class:`Expanding`, and :class:`ExponentialMovingWindow` ops. This will now raise a :class:`.errors.DataError` (:issue:`42834`)
+- Changed behavior in setting values with ``df.loc[:, foo] = bar`` or ``df.iloc[:, foo] = bar``, these now always attempt to set values inplace before falling back to casting (:issue:`45333`)
+- Changed default of ``numeric_only`` in various :class:`.DataFrameGroupBy` methods; all methods now default to ``numeric_only=False`` (:issue:`46072`)
+- Changed default of ``numeric_only`` to ``False`` in :class:`.Resampler` methods (:issue:`47177`)
+- Using the method :meth:`.DataFrameGroupBy.transform` with a callable that returns DataFrames will align to the input's index (:issue:`47244`)
+- When providing a list of columns of length one to :meth:`DataFrame.groupby`, the keys that are returned by iterating over the resulting :class:`DataFrameGroupBy` object will now be tuples of length one (:issue:`47761`)
+- Removed deprecated methods :meth:`ExcelWriter.write_cells`, :meth:`ExcelWriter.save`, :meth:`ExcelWriter.cur_sheet`, :meth:`ExcelWriter.handles`, :meth:`ExcelWriter.path` (:issue:`45795`)
+- The :class:`ExcelWriter` attribute ``book`` can no longer be set; it is still available to be accessed and mutated (:issue:`48943`)
+- Removed unused ``*args`` and ``**kwargs`` in :class:`Rolling`, :class:`Expanding`, and :class:`ExponentialMovingWindow` ops (:issue:`47851`)
+- Removed the deprecated argument ``line_terminator`` from :meth:`DataFrame.to_csv` (:issue:`45302`)
+- Removed the deprecated argument ``label`` from :func:`lreshape` (:issue:`30219`)
+- Arguments after ``expr`` in :meth:`DataFrame.eval` and :meth:`DataFrame.query` are keyword-only (:issue:`47587`)
+- Removed :meth:`Index._get_attributes_dict` (:issue:`50648`)
+- Removed :meth:`Series.__array_wrap__` (:issue:`50648`)
+- Changed behavior of :meth:`.DataFrame.value_counts` to return a :class:`Series` with :class:`MultiIndex` for any list-like(one element or not) but an :class:`Index` for a single label (:issue:`50829`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :meth:`.DataFrameGroupBy.median` and :meth:`.SeriesGroupBy.median` and :meth:`.DataFrameGroupBy.cumprod` for nullable dtypes (:issue:`37493`)
+- Performance improvement in :meth:`.DataFrameGroupBy.all`, :meth:`.DataFrameGroupBy.any`, :meth:`.SeriesGroupBy.all`, and :meth:`.SeriesGroupBy.any` for object dtype (:issue:`50623`)
+- Performance improvement in :meth:`MultiIndex.argsort` and :meth:`MultiIndex.sort_values` (:issue:`48406`)
+- Performance improvement in :meth:`MultiIndex.size` (:issue:`48723`)
+- Performance improvement in :meth:`MultiIndex.union` without missing values and without duplicates (:issue:`48505`, :issue:`48752`)
+- Performance improvement in :meth:`MultiIndex.difference` (:issue:`48606`)
+- Performance improvement in :class:`MultiIndex` set operations with sort=None (:issue:`49010`)
+- Performance improvement in :meth:`.DataFrameGroupBy.mean`, :meth:`.SeriesGroupBy.mean`, :meth:`.DataFrameGroupBy.var`, and :meth:`.SeriesGroupBy.var` for extension array dtypes (:issue:`37493`)
+- Performance improvement in :meth:`MultiIndex.isin` when ``level=None`` (:issue:`48622`, :issue:`49577`)
+- Performance improvement in :meth:`MultiIndex.putmask` (:issue:`49830`)
+- Performance improvement in :meth:`Index.union` and :meth:`MultiIndex.union` when index contains duplicates (:issue:`48900`)
+- Performance improvement in :meth:`Series.rank` for pyarrow-backed dtypes (:issue:`50264`)
+- Performance improvement in :meth:`Series.searchsorted` for pyarrow-backed dtypes (:issue:`50447`)
+- Performance improvement in :meth:`Series.fillna` for extension array dtypes (:issue:`49722`, :issue:`50078`)
+- Performance improvement in :meth:`Index.join`, :meth:`Index.intersection` and :meth:`Index.union` for masked and arrow dtypes when :class:`Index` is monotonic (:issue:`50310`, :issue:`51365`)
+- Performance improvement for :meth:`Series.value_counts` with nullable dtype (:issue:`48338`)
+- Performance improvement for :class:`Series` constructor passing integer numpy array with nullable dtype (:issue:`48338`)
+- Performance improvement for :class:`DatetimeIndex` constructor passing a list (:issue:`48609`)
+- Performance improvement in :func:`merge` and :meth:`DataFrame.join` when joining on a sorted :class:`MultiIndex` (:issue:`48504`)
+- Performance improvement in :func:`to_datetime` when parsing strings with timezone offsets (:issue:`50107`)
+- Performance improvement in :meth:`DataFrame.loc` and :meth:`Series.loc` for tuple-based indexing of a :class:`MultiIndex` (:issue:`48384`)
+- Performance improvement for :meth:`Series.replace` with categorical dtype (:issue:`49404`)
+- Performance improvement for :meth:`MultiIndex.unique` (:issue:`48335`)
+- Performance improvement for indexing operations with nullable and arrow dtypes (:issue:`49420`, :issue:`51316`)
+- Performance improvement for :func:`concat` with extension array backed indexes (:issue:`49128`, :issue:`49178`)
+- Performance improvement for :func:`api.types.infer_dtype` (:issue:`51054`)
+- Reduce memory usage of :meth:`DataFrame.to_pickle`/:meth:`Series.to_pickle` when using BZ2 or LZMA (:issue:`49068`)
+- Performance improvement for :class:`~arrays.StringArray` constructor passing a numpy array with type ``np.str_`` (:issue:`49109`)
+- Performance improvement in :meth:`~arrays.IntervalArray.from_tuples` (:issue:`50620`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.factorize` (:issue:`49177`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.__setitem__` (:issue:`50248`, :issue:`50632`)
+- Performance improvement in :class:`~arrays.ArrowExtensionArray` comparison methods when array contains NA (:issue:`50524`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.to_numpy` (:issue:`49973`, :issue:`51227`)
+- Performance improvement when parsing strings to :class:`BooleanDtype` (:issue:`50613`)
+- Performance improvement in :meth:`DataFrame.join` when joining on a subset of a :class:`MultiIndex` (:issue:`48611`)
+- Performance improvement for :meth:`MultiIndex.intersection` (:issue:`48604`)
+- Performance improvement in :meth:`DataFrame.__setitem__` (:issue:`46267`)
+- Performance improvement in ``var`` and ``std`` for nullable dtypes (:issue:`48379`).
+- Performance improvement when iterating over pyarrow and nullable dtypes (:issue:`49825`, :issue:`49851`)
+- Performance improvements to :func:`read_sas` (:issue:`47403`, :issue:`47405`, :issue:`47656`, :issue:`48502`)
+- Memory improvement in :meth:`RangeIndex.sort_values` (:issue:`48801`)
+- Performance improvement in :meth:`Series.to_numpy` if ``copy=True`` by avoiding copying twice (:issue:`24345`)
+- Performance improvement in :meth:`Series.rename` with :class:`MultiIndex` (:issue:`21055`)
+- Performance improvement in :class:`DataFrameGroupBy` and :class:`SeriesGroupBy` when ``by`` is a categorical type and ``sort=False`` (:issue:`48976`)
+- Performance improvement in :class:`DataFrameGroupBy` and :class:`SeriesGroupBy` when ``by`` is a categorical type and ``observed=False`` (:issue:`49596`)
+- Performance improvement in :func:`read_stata` with parameter ``index_col`` set to ``None`` (the default). Now the index will be a :class:`RangeIndex` instead of :class:`Int64Index` (:issue:`49745`)
+- Performance improvement in :func:`merge` when not merging on the index - the new index will now be :class:`RangeIndex` instead of :class:`Int64Index` (:issue:`49478`)
+- Performance improvement in :meth:`DataFrame.to_dict` and :meth:`Series.to_dict` when using any non-object dtypes (:issue:`46470`)
+- Performance improvement in :func:`read_html` when there are multiple tables (:issue:`49929`)
+- Performance improvement in :class:`Period` constructor when constructing from a string or integer (:issue:`38312`)
+- Performance improvement in :func:`to_datetime` when using ``'%Y%m%d'`` format (:issue:`17410`)
+- Performance improvement in :func:`to_datetime` when format is given or can be inferred (:issue:`50465`)
+- Performance improvement in :meth:`Series.median` for nullable dtypes (:issue:`50838`)
+- Performance improvement in :func:`read_csv` when passing :func:`to_datetime` lambda-function to ``date_parser`` and inputs have mixed timezone offsets (:issue:`35296`)
+- Performance improvement in :func:`isna` and :func:`isnull` (:issue:`50658`)
+- Performance improvement in :meth:`.SeriesGroupBy.value_counts` with categorical dtype (:issue:`46202`)
+- Fixed a reference leak in :func:`read_hdf` (:issue:`37441`)
+- Fixed a memory leak in :meth:`DataFrame.to_json` and :meth:`Series.to_json` when serializing datetimes and timedeltas (:issue:`40443`)
+- Decreased memory usage in many :class:`DataFrameGroupBy` methods (:issue:`51090`)
+- Performance improvement in :meth:`DataFrame.round` for an integer ``decimal`` parameter (:issue:`17254`)
+- Performance improvement in :meth:`DataFrame.replace` and :meth:`Series.replace` when using a large dict for ``to_replace`` (:issue:`6697`)
+- Memory improvement in :class:`StataReader` when reading seekable files (:issue:`48922`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in :meth:`Categorical.set_categories` losing dtype information (:issue:`48812`)
+- Bug in :meth:`Series.replace` with categorical dtype when ``to_replace`` values overlap with new values (:issue:`49404`)
+- Bug in :meth:`Series.replace` with categorical dtype losing nullable dtypes of underlying categories (:issue:`49404`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` would reorder categories when used as a grouper (:issue:`48749`)
+- Bug in :class:`Categorical` constructor when constructing from a :class:`Categorical` object and ``dtype="category"`` losing ordered-ness (:issue:`49309`)
+- Bug in :meth:`.SeriesGroupBy.min`, :meth:`.SeriesGroupBy.max`, :meth:`.DataFrameGroupBy.min`, and :meth:`.DataFrameGroupBy.max` with unordered :class:`CategoricalDtype` with no groups failing to raise ``TypeError`` (:issue:`51034`)
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :func:`pandas.infer_freq`, raising ``TypeError`` when inferred on :class:`RangeIndex` (:issue:`47084`)
+- Bug in :func:`to_datetime` incorrectly raising ``OverflowError`` with string arguments corresponding to large integers (:issue:`50533`)
+- Bug in :func:`to_datetime` was raising on invalid offsets with ``errors='coerce'`` and ``infer_datetime_format=True`` (:issue:`48633`)
+- Bug in :class:`DatetimeIndex` constructor failing to raise when ``tz=None`` is explicitly specified in conjunction with timezone-aware ``dtype`` or data (:issue:`48659`)
+- Bug in subtracting a ``datetime`` scalar from :class:`DatetimeIndex` failing to retain the original ``freq`` attribute (:issue:`48818`)
+- Bug in ``pandas.tseries.holiday.Holiday`` where a half-open date interval causes inconsistent return types from :meth:`USFederalHolidayCalendar.holidays` (:issue:`49075`)
+- Bug in rendering :class:`DatetimeIndex` and :class:`Series` and :class:`DataFrame` with timezone-aware dtypes with ``dateutil`` or ``zoneinfo`` timezones near daylight-savings transitions (:issue:`49684`)
+- Bug in :func:`to_datetime` was raising ``ValueError`` when parsing :class:`Timestamp`, ``datetime.datetime``, ``datetime.date``, or ``np.datetime64`` objects when non-ISO8601 ``format`` was passed (:issue:`49298`, :issue:`50036`)
+- Bug in :func:`to_datetime` was raising ``ValueError`` when parsing empty string and non-ISO8601 format was passed. Now, empty strings will be parsed as :class:`NaT`, for compatibility with how is done for ISO8601 formats (:issue:`50251`)
+- Bug in :class:`Timestamp` was showing ``UserWarning``, which was not actionable by users, when parsing non-ISO8601 delimited date strings (:issue:`50232`)
+- Bug in :func:`to_datetime` was showing misleading ``ValueError`` when parsing dates with format containing ISO week directive and ISO weekday directive (:issue:`50308`)
+- Bug in :meth:`Timestamp.round` when the ``freq`` argument has zero-duration (e.g. "0ns") returning incorrect results instead of raising (:issue:`49737`)
+- Bug in :func:`to_datetime` was not raising ``ValueError`` when invalid format was passed and ``errors`` was ``'ignore'`` or ``'coerce'`` (:issue:`50266`)
+- Bug in :class:`DateOffset` was throwing ``TypeError`` when constructing with milliseconds and another super-daily argument (:issue:`49897`)
+- Bug in :func:`to_datetime` was not raising ``ValueError`` when parsing string with decimal date with format ``'%Y%m%d'`` (:issue:`50051`)
+- Bug in :func:`to_datetime` was not converting ``None`` to ``NaT`` when parsing mixed-offset date strings with ISO8601 format (:issue:`50071`)
+- Bug in :func:`to_datetime` was not returning input when parsing out-of-bounds date string with ``errors='ignore'`` and ``format='%Y%m%d'`` (:issue:`14487`)
+- Bug in :func:`to_datetime` was converting timezone-naive ``datetime.datetime`` to timezone-aware when parsing with timezone-aware strings, ISO8601 format, and ``utc=False`` (:issue:`50254`)
+- Bug in :func:`to_datetime` was throwing ``ValueError`` when parsing dates with ISO8601 format where some values were not zero-padded (:issue:`21422`)
+- Bug in :func:`to_datetime` was giving incorrect results when using ``format='%Y%m%d'`` and ``errors='ignore'`` (:issue:`26493`)
+- Bug in :func:`to_datetime` was failing to parse date strings ``'today'`` and ``'now'`` if ``format`` was not ISO8601 (:issue:`50359`)
+- Bug in :func:`Timestamp.utctimetuple` raising a ``TypeError`` (:issue:`32174`)
+- Bug in :func:`to_datetime` was raising ``ValueError`` when parsing mixed-offset :class:`Timestamp` with ``errors='ignore'`` (:issue:`50585`)
+- Bug in :func:`to_datetime` was incorrectly handling floating-point inputs within 1 ``unit`` of the overflow boundaries (:issue:`50183`)
+- Bug in :func:`to_datetime` with unit of "Y" or "M" giving incorrect results, not matching pointwise :class:`Timestamp` results (:issue:`50870`)
+- Bug in :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` with datetime or timedelta dtypes incorrectly raising ``ValueError`` (:issue:`11312`)
+- Bug in :func:`to_datetime` was not returning input with ``errors='ignore'`` when input was out-of-bounds (:issue:`50587`)
+- Bug in :func:`DataFrame.from_records` when given a :class:`DataFrame` input with timezone-aware datetime64 columns incorrectly dropping the timezone-awareness (:issue:`51162`)
+- Bug in :func:`to_datetime` was raising ``decimal.InvalidOperation`` when parsing date strings with ``errors='coerce'`` (:issue:`51084`)
+- Bug in :func:`to_datetime` with both ``unit`` and ``origin`` specified returning incorrect results (:issue:`42624`)
+- Bug in :meth:`Series.astype` and :meth:`DataFrame.astype` when converting an object-dtype object containing timezone-aware datetimes or strings to ``datetime64[ns]`` incorrectly localizing as UTC instead of raising ``TypeError`` (:issue:`50140`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` with datetime or timedelta dtypes giving incorrect results for groups containing ``NaT`` (:issue:`51373`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` incorrectly raising with :class:`PeriodDtype` or :class:`DatetimeTZDtype` (:issue:`51373`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :func:`to_timedelta` raising error when input has nullable dtype ``Float64`` (:issue:`48796`)
+- Bug in :class:`Timedelta` constructor incorrectly raising instead of returning ``NaT`` when given a ``np.timedelta64("nat")`` (:issue:`48898`)
+- Bug in :class:`Timedelta` constructor failing to raise when passed both a :class:`Timedelta` object and keywords (e.g. days, seconds) (:issue:`48898`)
+- Bug in :class:`Timedelta` comparisons with very large ``datetime.timedelta`` objects incorrect raising ``OutOfBoundsTimedelta`` (:issue:`49021`)
+
+Timezones
+^^^^^^^^^
+- Bug in :meth:`Series.astype` and :meth:`DataFrame.astype` with object-dtype containing multiple timezone-aware ``datetime`` objects with heterogeneous timezones to a :class:`DatetimeTZDtype` incorrectly raising (:issue:`32581`)
+- Bug in :func:`to_datetime` was failing to parse date strings with timezone name when ``format`` was specified with ``%Z`` (:issue:`49748`)
+- Better error message when passing invalid values to ``ambiguous`` parameter in :meth:`Timestamp.tz_localize` (:issue:`49565`)
+- Bug in string parsing incorrectly allowing a :class:`Timestamp` to be constructed with an invalid timezone, which would raise when trying to print (:issue:`50668`)
+- Corrected TypeError message in :func:`objects_to_datetime64ns` to inform that DatetimeIndex has mixed timezones (:issue:`50974`)
+
+Numeric
+^^^^^^^
+- Bug in :meth:`DataFrame.add` cannot apply ufunc when inputs contain mixed DataFrame type and Series type (:issue:`39853`)
+- Bug in arithmetic operations on :class:`Series` not propagating mask when combining masked dtypes and numpy dtypes (:issue:`45810`, :issue:`42630`)
+- Bug in :meth:`DataFrame.sem` and :meth:`Series.sem` where an erroneous ``TypeError`` would always raise when using data backed by an :class:`ArrowDtype` (:issue:`49759`)
+- Bug in :meth:`Series.__add__` casting to object for list and masked :class:`Series` (:issue:`22962`)
+- Bug in :meth:`~arrays.ArrowExtensionArray.mode` where ``dropna=False`` was not respected when there was ``NA`` values (:issue:`50982`)
+- Bug in :meth:`DataFrame.query` with ``engine="numexpr"`` and column names are ``min`` or ``max`` would raise a ``TypeError`` (:issue:`50937`)
+- Bug in :meth:`DataFrame.min` and :meth:`DataFrame.max` with tz-aware data containing ``pd.NaT`` and ``axis=1`` would return incorrect results (:issue:`51242`)
+
+Conversion
+^^^^^^^^^^
+- Bug in constructing :class:`Series` with ``int64`` dtype from a string list raising instead of casting (:issue:`44923`)
+- Bug in constructing :class:`Series` with masked dtype and boolean values with ``NA`` raising (:issue:`42137`)
+- Bug in :meth:`DataFrame.eval` incorrectly raising an ``AttributeError`` when there are negative values in function call (:issue:`46471`)
+- Bug in :meth:`Series.convert_dtypes` not converting dtype to nullable dtype when :class:`Series` contains ``NA`` and has dtype ``object`` (:issue:`48791`)
+- Bug where any :class:`ExtensionDtype` subclass with ``kind="M"`` would be interpreted as a timezone type (:issue:`34986`)
+- Bug in :class:`.arrays.ArrowExtensionArray` that would raise ``NotImplementedError`` when passed a sequence of strings or binary (:issue:`49172`)
+- Bug in :meth:`Series.astype` raising ``pyarrow.ArrowInvalid`` when converting from a non-pyarrow string dtype to a pyarrow numeric type (:issue:`50430`)
+- Bug in :meth:`DataFrame.astype` modifying input array inplace when converting to ``string`` and ``copy=False`` (:issue:`51073`)
+- Bug in :meth:`Series.to_numpy` converting to NumPy array before applying ``na_value`` (:issue:`48951`)
+- Bug in :meth:`DataFrame.astype` not copying data when converting to pyarrow dtype (:issue:`50984`)
+- Bug in :func:`to_datetime` was not respecting ``exact`` argument when ``format`` was an ISO8601 format (:issue:`12649`)
+- Bug in :meth:`TimedeltaArray.astype` raising ``TypeError`` when converting to a pyarrow duration type (:issue:`49795`)
+- Bug in :meth:`DataFrame.eval` and :meth:`DataFrame.query` raising for extension array dtypes (:issue:`29618`, :issue:`50261`, :issue:`31913`)
+- Bug in :meth:`Series` not copying data when created from :class:`Index` and ``dtype`` is equal to ``dtype`` from :class:`Index` (:issue:`52008`)
+
+Strings
+^^^^^^^
+- Bug in :func:`pandas.api.types.is_string_dtype` that would not return ``True`` for :class:`StringDtype` or :class:`ArrowDtype` with ``pyarrow.string()`` (:issue:`15585`)
+- Bug in converting string dtypes to "datetime64[ns]" or "timedelta64[ns]" incorrectly raising ``TypeError`` (:issue:`36153`)
+- Bug in setting values in a string-dtype column with an array, mutating the array as side effect when it contains missing values (:issue:`51299`)
+
+Interval
+^^^^^^^^
+- Bug in :meth:`IntervalIndex.is_overlapping` incorrect output if interval has duplicate left boundaries (:issue:`49581`)
+- Bug in :meth:`Series.infer_objects` failing to infer :class:`IntervalDtype` for an object series of :class:`Interval` objects (:issue:`50090`)
+- Bug in :meth:`Series.shift` with :class:`IntervalDtype` and invalid null ``fill_value`` failing to raise ``TypeError`` (:issue:`51258`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`DataFrame.__setitem__` raising when indexer is a :class:`DataFrame` with ``boolean`` dtype (:issue:`47125`)
+- Bug in :meth:`DataFrame.reindex` filling with wrong values when indexing columns and index for ``uint`` dtypes (:issue:`48184`)
+- Bug in :meth:`DataFrame.loc` when setting :class:`DataFrame` with different dtypes coercing values to single dtype (:issue:`50467`)
+- Bug in :meth:`DataFrame.sort_values` where ``None`` was not returned when ``by`` is empty list and ``inplace=True`` (:issue:`50643`)
+- Bug in :meth:`DataFrame.loc` coercing dtypes when setting values with a list indexer (:issue:`49159`)
+- Bug in :meth:`Series.loc` raising error for out of bounds end of slice indexer (:issue:`50161`)
+- Bug in :meth:`DataFrame.loc` raising ``ValueError`` with all ``False`` ``bool`` indexer and empty object (:issue:`51450`)
+- Bug in :meth:`DataFrame.loc` raising ``ValueError`` with ``bool`` indexer and :class:`MultiIndex` (:issue:`47687`)
+- Bug in :meth:`DataFrame.loc` raising ``IndexError`` when setting values for a pyarrow-backed column with a non-scalar indexer (:issue:`50085`)
+- Bug in :meth:`DataFrame.__getitem__`, :meth:`Series.__getitem__`, :meth:`DataFrame.__setitem__` and :meth:`Series.__setitem__`
+  when indexing on indexes with extension float dtypes (:class:`Float64` & :class:`Float64`) or complex dtypes using integers (:issue:`51053`)
+- Bug in :meth:`DataFrame.loc` modifying object when setting incompatible value with an empty indexer (:issue:`45981`)
+- Bug in :meth:`DataFrame.__setitem__` raising ``ValueError`` when right hand side is :class:`DataFrame` with :class:`MultiIndex` columns (:issue:`49121`)
+- Bug in :meth:`DataFrame.reindex` casting dtype to ``object`` when :class:`DataFrame` has single extension array column when re-indexing ``columns`` and ``index`` (:issue:`48190`)
+- Bug in :meth:`DataFrame.iloc` raising ``IndexError`` when indexer is a :class:`Series` with numeric extension array dtype (:issue:`49521`)
+- Bug in :func:`~DataFrame.describe` when formatting percentiles in the resulting index showed more decimals than needed (:issue:`46362`)
+- Bug in :meth:`DataFrame.compare` does not recognize differences when comparing ``NA`` with value in nullable dtypes (:issue:`48939`)
+- Bug in :meth:`Series.rename` with :class:`MultiIndex` losing extension array dtypes (:issue:`21055`)
+- Bug in :meth:`DataFrame.isetitem` coercing extension array dtypes in :class:`DataFrame` to object (:issue:`49922`)
+- Bug in :meth:`Series.__getitem__` returning corrupt object when selecting from an empty pyarrow backed object (:issue:`51734`)
+- Bug in :class:`BusinessHour` would cause creation of :class:`DatetimeIndex` to fail when no opening hour was included in the index (:issue:`49835`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`Index.equals` raising ``TypeError`` when :class:`Index` consists of tuples that contain ``NA`` (:issue:`48446`)
+- Bug in :meth:`Series.map` caused incorrect result when data has NaNs and defaultdict mapping was used (:issue:`48813`)
+- Bug in :class:`NA` raising a ``TypeError`` instead of return :class:`NA` when performing a binary operation with a ``bytes`` object (:issue:`49108`)
+- Bug in :meth:`DataFrame.update` with ``overwrite=False`` raising ``TypeError`` when ``self`` has column with ``NaT`` values and column not present in ``other`` (:issue:`16713`)
+- Bug in :meth:`Series.replace` raising ``RecursionError`` when replacing value in object-dtype :class:`Series` containing ``NA`` (:issue:`47480`)
+- Bug in :meth:`Series.replace` raising ``RecursionError`` when replacing value in numeric :class:`Series` with ``NA`` (:issue:`50758`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`MultiIndex.get_indexer` not matching ``NaN`` values (:issue:`29252`, :issue:`37222`, :issue:`38623`, :issue:`42883`, :issue:`43222`, :issue:`46173`, :issue:`48905`)
+- Bug in :meth:`MultiIndex.argsort` raising ``TypeError`` when index contains :attr:`NA` (:issue:`48495`)
+- Bug in :meth:`MultiIndex.difference` losing extension array dtype (:issue:`48606`)
+- Bug in :class:`MultiIndex.set_levels` raising ``IndexError`` when setting empty level (:issue:`48636`)
+- Bug in :meth:`MultiIndex.unique` losing extension array dtype (:issue:`48335`)
+- Bug in :meth:`MultiIndex.intersection` losing extension array (:issue:`48604`)
+- Bug in :meth:`MultiIndex.union` losing extension array (:issue:`48498`, :issue:`48505`, :issue:`48900`)
+- Bug in :meth:`MultiIndex.union` not sorting when sort=None and index contains missing values (:issue:`49010`)
+- Bug in :meth:`MultiIndex.append` not checking names for equality (:issue:`48288`)
+- Bug in :meth:`MultiIndex.symmetric_difference` losing extension array (:issue:`48607`)
+- Bug in :meth:`MultiIndex.join` losing dtypes when :class:`MultiIndex` has duplicates (:issue:`49830`)
+- Bug in :meth:`MultiIndex.putmask` losing extension array (:issue:`49830`)
+- Bug in :meth:`MultiIndex.value_counts` returning a :class:`Series` indexed by flat index of tuples instead of a :class:`MultiIndex` (:issue:`49558`)
+
+I/O
+^^^
+- Bug in :func:`read_sas` caused fragmentation of :class:`DataFrame` and raised :class:`.errors.PerformanceWarning` (:issue:`48595`)
+- Improved error message in :func:`read_excel` by including the offending sheet name when an exception is raised while reading a file (:issue:`48706`)
+- Bug when a pickling a subset PyArrow-backed data that would serialize the entire data instead of the subset (:issue:`42600`)
+- Bug in :func:`read_sql_query` ignoring ``dtype`` argument when ``chunksize`` is specified and result is empty (:issue:`50245`)
+- Bug in :func:`read_csv` for a single-line csv with fewer columns than ``names`` raised :class:`.errors.ParserError` with ``engine="c"`` (:issue:`47566`)
+- Bug in :func:`read_json` raising with ``orient="table"`` and ``NA`` value (:issue:`40255`)
+- Bug in displaying ``string`` dtypes not showing storage option (:issue:`50099`)
+- Bug in :meth:`DataFrame.to_string` with ``header=False`` that printed the index name on the same line as the first row of the data (:issue:`49230`)
+- Bug in :meth:`DataFrame.to_string` ignoring float formatter for extension arrays (:issue:`39336`)
+- Fixed memory leak which stemmed from the initialization of the internal JSON module (:issue:`49222`)
+- Fixed issue where :func:`json_normalize` would incorrectly remove leading characters from column names that matched the ``sep`` argument (:issue:`49861`)
+- Bug in :func:`read_csv` unnecessarily overflowing for extension array dtype when containing ``NA`` (:issue:`32134`)
+- Bug in :meth:`DataFrame.to_dict` not converting ``NA`` to ``None`` (:issue:`50795`)
+- Bug in :meth:`DataFrame.to_json` where it would segfault when failing to encode a string (:issue:`50307`)
+- Bug in :meth:`DataFrame.to_html` with ``na_rep`` set when the :class:`DataFrame` contains non-scalar data (:issue:`47103`)
+- Bug in :func:`read_xml` where file-like objects failed when iterparse is used (:issue:`50641`)
+- Bug in :func:`read_csv` when ``engine="pyarrow"`` where ``encoding`` parameter was not handled correctly (:issue:`51302`)
+- Bug in :func:`read_xml` ignored repeated elements when iterparse is used (:issue:`51183`)
+- Bug in :class:`ExcelWriter` leaving file handles open if an exception occurred during instantiation (:issue:`51443`)
+- Bug in :meth:`DataFrame.to_parquet` where non-string index or columns were raising a ``ValueError`` when ``engine="pyarrow"`` (:issue:`52036`)
+
+Period
+^^^^^^
+- Bug in :meth:`Period.strftime` and :meth:`PeriodIndex.strftime`, raising ``UnicodeDecodeError`` when a locale-specific directive was passed (:issue:`46319`)
+- Bug in adding a :class:`Period` object to an array of :class:`DateOffset` objects incorrectly raising ``TypeError`` (:issue:`50162`)
+- Bug in :class:`Period` where passing a string with finer resolution than nanosecond would result in a ``KeyError`` instead of dropping the extra precision (:issue:`50417`)
+- Bug in parsing strings representing Week-periods e.g. "2017-01-23/2017-01-29" as minute-frequency instead of week-frequency (:issue:`50803`)
+- Bug in :meth:`.DataFrameGroupBy.sum`, :meth:`.DataFrameGroupByGroupBy.cumsum`, :meth:`.DataFrameGroupByGroupBy.prod`, :meth:`.DataFrameGroupByGroupBy.cumprod` with :class:`PeriodDtype` failing to raise ``TypeError`` (:issue:`51040`)
+- Bug in parsing empty string with :class:`Period` incorrectly raising ``ValueError`` instead of returning ``NaT`` (:issue:`51349`)
+
+Plotting
+^^^^^^^^
+- Bug in :meth:`DataFrame.plot.hist`, not dropping elements of ``weights`` corresponding to ``NaN`` values in ``data`` (:issue:`48884`)
+- ``ax.set_xlim`` was sometimes raising ``UserWarning`` which users couldn't address due to ``set_xlim`` not accepting parsing arguments - the converter now uses :func:`Timestamp` instead (:issue:`49148`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :class:`.ExponentialMovingWindow` with ``online`` not raising a ``NotImplementedError`` for unsupported operations (:issue:`48834`)
+- Bug in :meth:`.DataFrameGroupBy.sample` raises ``ValueError`` when the object is empty (:issue:`48459`)
+- Bug in :meth:`Series.groupby` raises ``ValueError`` when an entry of the index is equal to the name of the index (:issue:`48567`)
+- Bug in :meth:`.DataFrameGroupBy.resample` produces inconsistent results when passing empty DataFrame (:issue:`47705`)
+- Bug in :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` would not include unobserved categories in result when grouping by categorical indexes (:issue:`49354`)
+- Bug in :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` would change result order depending on the input index when grouping by categoricals (:issue:`49223`)
+- Bug in :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` when grouping on categorical data would sort result values even when used with ``sort=False`` (:issue:`42482`)
+- Bug in :meth:`.DataFrameGroupBy.apply` and :class:`.SeriesGroupBy.apply` with ``as_index=False`` would not attempt the computation without using the grouping keys when using them failed with a ``TypeError`` (:issue:`49256`)
+- Bug in :meth:`.DataFrameGroupBy.describe` would describe the group keys (:issue:`49256`)
+- Bug in :meth:`.SeriesGroupBy.describe` with ``as_index=False`` would have the incorrect shape (:issue:`49256`)
+- Bug in :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` with ``dropna=False`` would drop NA values when the grouper was categorical (:issue:`36327`)
+- Bug in :meth:`.SeriesGroupBy.nunique` would incorrectly raise when the grouper was an empty categorical and ``observed=True`` (:issue:`21334`)
+- Bug in :meth:`.SeriesGroupBy.nth` would raise when grouper contained NA values after subsetting from a :class:`DataFrameGroupBy` (:issue:`26454`)
+- Bug in :meth:`DataFrame.groupby` would not include a :class:`.Grouper` specified by ``key`` in the result when ``as_index=False`` (:issue:`50413`)
+- Bug in :meth:`.DataFrameGroupBy.value_counts` would raise when used with a :class:`.TimeGrouper` (:issue:`50486`)
+- Bug in :meth:`.Resampler.size` caused a wide :class:`DataFrame` to be returned instead of a :class:`Series` with :class:`MultiIndex` (:issue:`46826`)
+- Bug in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` would raise incorrectly when grouper had ``axis=1`` for ``"idxmin"`` and ``"idxmax"`` arguments (:issue:`45986`)
+- Bug in :class:`.DataFrameGroupBy` would raise when used with an empty DataFrame, categorical grouper, and ``dropna=False`` (:issue:`50634`)
+- Bug in :meth:`.SeriesGroupBy.value_counts` did not respect ``sort=False`` (:issue:`50482`)
+- Bug in :meth:`.DataFrameGroupBy.resample` raises ``KeyError`` when getting the result from a key list when resampling on time index (:issue:`50840`)
+- Bug in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` would raise incorrectly when grouper had ``axis=1`` for ``"ngroup"`` argument (:issue:`45986`)
+- Bug in :meth:`.DataFrameGroupBy.describe` produced incorrect results when data had duplicate columns (:issue:`50806`)
+- Bug in :meth:`.DataFrameGroupBy.agg` with ``engine="numba"`` failing to respect ``as_index=False`` (:issue:`51228`)
+- Bug in :meth:`.DataFrameGroupBy.agg`, :meth:`.SeriesGroupBy.agg`, and :meth:`.Resampler.agg` would ignore arguments when passed a list of functions (:issue:`50863`)
+- Bug in :meth:`.DataFrameGroupBy.ohlc` ignoring ``as_index=False`` (:issue:`51413`)
+- Bug in :meth:`DataFrameGroupBy.agg` after subsetting columns (e.g. ``.groupby(...)[["a", "b"]]``) would not include groupings in the result (:issue:`51186`)
+
+Reshaping
+^^^^^^^^^
+- Bug in :meth:`DataFrame.pivot_table` raising ``TypeError`` for nullable dtype and ``margins=True`` (:issue:`48681`)
+- Bug in :meth:`DataFrame.unstack` and :meth:`Series.unstack` unstacking wrong level of :class:`MultiIndex` when :class:`MultiIndex` has mixed names (:issue:`48763`)
+- Bug in :meth:`DataFrame.melt` losing extension array dtype (:issue:`41570`)
+- Bug in :meth:`DataFrame.pivot` not respecting ``None`` as column name (:issue:`48293`)
+- Bug in :meth:`DataFrame.join` when ``left_on`` or ``right_on`` is or includes a :class:`CategoricalIndex` incorrectly raising ``AttributeError`` (:issue:`48464`)
+- Bug in :meth:`DataFrame.pivot_table` raising ``ValueError`` with parameter ``margins=True`` when result is an empty :class:`DataFrame` (:issue:`49240`)
+- Clarified error message in :func:`merge` when passing invalid ``validate`` option (:issue:`49417`)
+- Bug in :meth:`DataFrame.explode` raising ``ValueError`` on multiple columns with ``NaN`` values or empty lists (:issue:`46084`)
+- Bug in :meth:`DataFrame.transpose` with ``IntervalDtype`` column with ``timedelta64[ns]`` endpoints (:issue:`44917`)
+- Bug in :meth:`DataFrame.agg` and :meth:`Series.agg` would ignore arguments when passed a list of functions (:issue:`50863`)
+
+Sparse
+^^^^^^
+- Bug in :meth:`Series.astype` when converting a ``SparseDtype`` with ``datetime64[ns]`` subtype to ``int64`` dtype raising, inconsistent with the non-sparse behavior (:issue:`49631`,:issue:`50087`)
+- Bug in :meth:`Series.astype` when converting a from ``datetime64[ns]`` to ``Sparse[datetime64[ns]]`` incorrectly raising (:issue:`50082`)
+- Bug in :meth:`Series.sparse.to_coo` raising ``SystemError`` when :class:`MultiIndex` contains a ``ExtensionArray`` (:issue:`50996`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :meth:`Series.mean` overflowing unnecessarily with nullable integers (:issue:`48378`)
+- Bug in :meth:`Series.tolist` for nullable dtypes returning numpy scalars instead of python scalars (:issue:`49890`)
+- Bug in :meth:`Series.round` for pyarrow-backed dtypes raising ``AttributeError`` (:issue:`50437`)
+- Bug when concatenating an empty DataFrame with an ExtensionDtype to another DataFrame with the same ExtensionDtype, the resulting dtype turned into object (:issue:`48510`)
+- Bug in :meth:`array.PandasArray.to_numpy` raising with ``NA`` value when ``na_value`` is specified (:issue:`40638`)
+- Bug in :meth:`api.types.is_numeric_dtype` where a custom :class:`ExtensionDtype` would not return ``True`` if ``_is_numeric`` returned ``True`` (:issue:`50563`)
+- Bug in :meth:`api.types.is_integer_dtype`, :meth:`api.types.is_unsigned_integer_dtype`, :meth:`api.types.is_signed_integer_dtype`, :meth:`api.types.is_float_dtype` where a custom :class:`ExtensionDtype` would not return ``True`` if ``kind`` returned the corresponding NumPy type (:issue:`50667`)
+- Bug in :class:`Series` constructor unnecessarily overflowing for nullable unsigned integer dtypes (:issue:`38798`, :issue:`25880`)
+- Bug in setting non-string value into ``StringArray`` raising ``ValueError`` instead of ``TypeError`` (:issue:`49632`)
+- Bug in :meth:`DataFrame.reindex` not honoring the default ``copy=True`` keyword in case of columns with ExtensionDtype (and as a result also selecting multiple columns with getitem (``[]``) didn't correctly result in a copy) (:issue:`51197`)
+- Bug in :class:`~arrays.ArrowExtensionArray` logical operations ``&`` and ``|`` raising ``KeyError`` (:issue:`51688`)
+
+Styler
+^^^^^^
+- Fix :meth:`~pandas.io.formats.style.Styler.background_gradient` for nullable dtype :class:`Series` with ``NA`` values (:issue:`50712`)
+
+Metadata
+^^^^^^^^
+- Fixed metadata propagation in :meth:`DataFrame.corr` and :meth:`DataFrame.cov` (:issue:`28283`)
+
+Other
+^^^^^
+- Bug in incorrectly accepting dtype strings containing "[pyarrow]" more than once (:issue:`51548`)
+- Bug in :meth:`Series.searchsorted` inconsistent behavior when accepting :class:`DataFrame` as parameter ``value`` (:issue:`49620`)
+- Bug in :func:`array` failing to raise on :class:`DataFrame` inputs (:issue:`51167`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_200.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v1.5.0rc0..v2.0.0
diff --git a/pandas/doc/source/whatsnew/v2.0.1.rst b/pandas/doc/source/whatsnew/v2.0.1.rst
new file mode 100644
index 0000000000000..f027ec9107816
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.0.1.rst
@@ -0,0 +1,68 @@
+.. _whatsnew_201:
+
+What's new in 2.0.1 (April 24, 2023)
+------------------------------------
+
+These are the changes in pandas 2.0.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_201.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression for subclassed Series when constructing from a dictionary (:issue:`52445`)
+- Fixed regression in :meth:`.SeriesGroupBy.agg` failing when grouping with categorical data, multiple groupings, ``as_index=False``, and a list of aggregations (:issue:`52760`)
+- Fixed regression in :meth:`DataFrame.pivot` changing :class:`Index` name of input object (:issue:`52629`)
+- Fixed regression in :meth:`DataFrame.resample` raising on a DataFrame with no columns (:issue:`52484`)
+- Fixed regression in :meth:`DataFrame.sort_values` not resetting index when :class:`DataFrame` is already sorted and ``ignore_index=True`` (:issue:`52553`)
+- Fixed regression in :meth:`MultiIndex.isin` raising ``TypeError`` for ``Generator`` (:issue:`52568`)
+- Fixed regression in :meth:`Series.describe` showing ``RuntimeWarning`` for extension dtype :class:`Series` with one element (:issue:`52515`)
+- Fixed regression when adding a new column to a :class:`DataFrame` when the :attr:`DataFrame.columns` was a :class:`RangeIndex` and the new key was hashable but not a scalar (:issue:`52652`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_201.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :attr:`Series.dt.days` that would overflow ``int32`` number of days (:issue:`52391`)
+- Bug in :class:`arrays.DatetimeArray` constructor returning an incorrect unit when passed a non-nanosecond numpy datetime array (:issue:`52555`)
+- Bug in :class:`~arrays.ArrowExtensionArray` with duration dtype overflowing when constructed from data containing numpy ``NaT`` (:issue:`52843`)
+- Bug in :func:`Series.dt.round` when passing a ``freq`` of equal or higher resolution compared to the :class:`Series` would raise a ``ZeroDivisionError`` (:issue:`52761`)
+- Bug in :func:`Series.median` with :class:`ArrowDtype` returning an approximate median (:issue:`52679`)
+- Bug in :func:`api.interchange.from_dataframe` was unnecessarily raising on categorical dtypes (:issue:`49889`)
+- Bug in :func:`api.interchange.from_dataframe` was unnecessarily raising on large string dtypes (:issue:`52795`)
+- Bug in :func:`pandas.testing.assert_series_equal` where ``check_dtype=False`` would still raise for datetime or timedelta types with different resolutions (:issue:`52449`)
+- Bug in :func:`read_csv` casting PyArrow datetimes to NumPy when ``dtype_backend="pyarrow"`` and ``parse_dates`` is set causing a performance bottleneck in the process (:issue:`52546`)
+- Bug in :func:`to_datetime` and :func:`to_timedelta` when trying to convert numeric data with a :class:`ArrowDtype` (:issue:`52425`)
+- Bug in :func:`to_numeric` with ``errors='coerce'`` and ``dtype_backend='pyarrow'`` with :class:`ArrowDtype` data (:issue:`52588`)
+- Bug in :meth:`ArrowDtype.__from_arrow__` not respecting if dtype is explicitly given (:issue:`52533`)
+- Bug in :meth:`DataFrame.describe` not respecting ``ArrowDtype`` in ``include`` and ``exclude`` (:issue:`52570`)
+- Bug in :meth:`DataFrame.max` and related casting different :class:`Timestamp` resolutions always to nanoseconds (:issue:`52524`)
+- Bug in :meth:`Series.describe` not returning :class:`ArrowDtype` with ``pyarrow.float64`` type with numeric data (:issue:`52427`)
+- Bug in :meth:`Series.dt.tz_localize` incorrectly localizing timestamps with :class:`ArrowDtype` (:issue:`52677`)
+- Bug in arithmetic between ``np.datetime64`` and ``np.timedelta64`` ``NaT`` scalars with units always returning nanosecond resolution (:issue:`52295`)
+- Bug in logical and comparison operations between :class:`ArrowDtype` and numpy masked types (e.g. ``"boolean"``) (:issue:`52625`)
+- Fixed bug in :func:`merge` when merging with ``ArrowDtype`` one one and a NumPy dtype on the other side (:issue:`52406`)
+- Fixed segfault in :meth:`Series.to_numpy` with ``null[pyarrow]`` dtype (:issue:`52443`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_201.other:
+
+Other
+~~~~~
+- :class:`DataFrame` created from empty dicts had :attr:`~DataFrame.columns`  of dtype ``object``. It is now a :class:`RangeIndex` (:issue:`52404`)
+- :class:`Series` created from empty dicts had :attr:`~Series.index`  of dtype ``object``. It is now a :class:`RangeIndex` (:issue:`52404`)
+- Implemented :meth:`Series.str.split` and :meth:`Series.str.rsplit` for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`52401`)
+- Implemented most ``str`` accessor methods for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`52401`)
+- Supplying a non-integer hashable key that tests ``False`` in :func:`api.types.is_scalar` now raises a ``KeyError`` for :meth:`RangeIndex.get_loc`, like it does for :meth:`Index.get_loc`. Previously it raised an ``InvalidIndexError`` (:issue:`52652`).
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_201.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.0.0..v2.0.1
diff --git a/pandas/doc/source/whatsnew/v2.0.2.rst b/pandas/doc/source/whatsnew/v2.0.2.rst
new file mode 100644
index 0000000000000..559078d501a00
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.0.2.rst
@@ -0,0 +1,57 @@
+.. _whatsnew_202:
+
+What's new in 2.0.2 (May 29, 2023)
+-----------------------------------
+
+These are the changes in pandas 2.0.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_202.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed performance regression in :meth:`GroupBy.apply` (:issue:`53195`)
+- Fixed regression in :func:`merge` on Windows when dtype is ``np.intc`` (:issue:`52451`)
+- Fixed regression in :func:`read_sql` dropping columns with duplicated column names (:issue:`53117`)
+- Fixed regression in :meth:`DataFrame.loc` losing :class:`MultiIndex` name when enlarging object (:issue:`53053`)
+- Fixed regression in :meth:`DataFrame.to_string` printing a backslash at the end of the first row of data, instead of headers, when the DataFrame doesn't fit the line width (:issue:`53054`)
+- Fixed regression in :meth:`MultiIndex.join` returning levels in wrong order (:issue:`53093`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_202.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :class:`.arrays.ArrowExtensionArray` incorrectly assigning ``dict`` instead of ``list`` for ``.type`` with ``pyarrow.map_`` and raising a ``NotImplementedError`` with ``pyarrow.struct`` (:issue:`53328`)
+- Bug in :func:`api.interchange.from_dataframe` was raising ``IndexError`` on empty categorical data (:issue:`53077`)
+- Bug in :func:`api.interchange.from_dataframe` was returning :class:`DataFrame`'s of incorrect sizes when called on slices (:issue:`52824`)
+- Bug in :func:`api.interchange.from_dataframe` was unnecessarily raising on bitmasks (:issue:`49888`)
+- Bug in :func:`merge` when merging on datetime columns on different resolutions (:issue:`53200`)
+- Bug in :func:`read_csv` raising ``OverflowError`` for ``engine="pyarrow"`` and ``parse_dates`` set (:issue:`53295`)
+- Bug in :func:`to_datetime` was inferring format to contain ``"%H"`` instead of ``"%I"`` if date contained "AM" / "PM" tokens (:issue:`53147`)
+- Bug in :func:`to_timedelta` was raising ``ValueError`` with ``pandas.NA`` (:issue:`52909`)
+- Bug in :meth:`DataFrame.__getitem__` not preserving dtypes for :class:`MultiIndex` partial keys (:issue:`51895`)
+- Bug in :meth:`DataFrame.convert_dtypes` ignores ``convert_*`` keywords when set to False ``dtype_backend="pyarrow"`` (:issue:`52872`)
+- Bug in :meth:`DataFrame.convert_dtypes` losing timezone for tz-aware dtypes and ``dtype_backend="pyarrow"`` (:issue:`53382`)
+- Bug in :meth:`DataFrame.sort_values` raising for PyArrow ``dictionary`` dtype (:issue:`53232`)
+- Bug in :meth:`Series.describe` treating pyarrow-backed timestamps and timedeltas as categorical data (:issue:`53001`)
+- Bug in :meth:`Series.rename` not making a lazy copy when Copy-on-Write is enabled when a scalar is passed to it (:issue:`52450`)
+- Bug in :meth:`pd.array` raising for ``NumPy`` array and ``pa.large_string`` or ``pa.large_binary`` (:issue:`52590`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_202.other:
+
+Other
+~~~~~
+- Raised a better error message when calling :func:`Series.dt.to_pydatetime` with :class:`ArrowDtype` with ``pyarrow.date32`` or ``pyarrow.date64`` type (:issue:`52812`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_202.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.0.1..v2.0.2
diff --git a/pandas/doc/source/whatsnew/v2.0.3.rst b/pandas/doc/source/whatsnew/v2.0.3.rst
new file mode 100644
index 0000000000000..26e34e0c823ce
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.0.3.rst
@@ -0,0 +1,46 @@
+.. _whatsnew_203:
+
+What's new in 2.0.3 (June 28, 2023)
+-----------------------------------
+
+These are the changes in pandas 2.0.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_203.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Bug in :meth:`Timestamp.weekday`` was returning incorrect results before ``'0000-02-29'`` (:issue:`53738`)
+- Fixed performance regression in merging on datetime-like columns (:issue:`53231`)
+- Fixed regression when :meth:`DataFrame.to_string` creates extra space for string dtypes (:issue:`52690`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_203.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :func:`DataFrame.convert_dtype` and :func:`Series.convert_dtype` when trying to convert :class:`ArrowDtype` with ``dtype_backend="nullable_numpy"`` (:issue:`53648`)
+- Bug in :func:`RangeIndex.union` when using ``sort=True`` with another :class:`RangeIndex` (:issue:`53490`)
+- Bug in :func:`Series.reindex` when expanding a non-nanosecond datetime or timedelta :class:`Series` would not fill with ``NaT`` correctly (:issue:`53497`)
+- Bug in :func:`read_csv` when defining ``dtype`` with ``bool[pyarrow]`` for the ``"c"`` and ``"python"`` engines (:issue:`53390`)
+- Bug in :meth:`Series.str.split` and :meth:`Series.str.rsplit` with ``expand=True`` for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`53532`)
+- Bug in indexing methods (e.g. :meth:`DataFrame.__getitem__`) where taking the entire :class:`DataFrame`/:class:`Series` would raise an ``OverflowError`` when Copy on Write was enabled and the length of the array was over the maximum size a 32-bit integer can hold (:issue:`53616`)
+- Bug when constructing a :class:`DataFrame` with columns of an :class:`ArrowDtype` with a ``pyarrow.dictionary`` type that reindexes the data (:issue:`53617`)
+- Bug when indexing a :class:`DataFrame` or :class:`Series` with an :class:`Index` with a timestamp :class:`ArrowDtype` would raise an ``AttributeError`` (:issue:`53644`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_203.other:
+
+Other
+~~~~~
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_203.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.0.2..v2.0.3
diff --git a/pandas/doc/source/whatsnew/v2.1.0.rst b/pandas/doc/source/whatsnew/v2.1.0.rst
new file mode 100644
index 0000000000000..495c8244142f9
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.1.0.rst
@@ -0,0 +1,895 @@
+.. _whatsnew_210:
+
+What's new in 2.1.0 (Aug 30, 2023)
+--------------------------------------
+
+These are the changes in pandas 2.1.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_210.enhancements.pyarrow_dependency:
+
+PyArrow will become a required dependency with pandas 3.0
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+`PyArrow <https://arrow.apache.org/docs/python/index.html>`_ will become a required
+dependency of pandas starting with pandas 3.0. This decision was made based on
+`PDEP 10 <https://pandas.pydata.org/pdeps/0010-required-pyarrow-dependency.html>`_.
+
+This will enable more changes that are hugely beneficial to pandas users, including
+but not limited to:
+
+- inferring strings as PyArrow backed strings by default enabling a significant
+  reduction of the memory footprint and huge performance improvements.
+- inferring more complex dtypes with PyArrow by default, like ``Decimal``, ``lists``,
+  ``bytes``, ``structured data`` and more.
+- Better interoperability with other libraries that depend on Apache Arrow.
+
+We are collecting feedback on this decision `here <https://github.com/pandas-dev/pandas/issues/54466>`_.
+
+.. _whatsnew_210.enhancements.infer_strings:
+
+Avoid NumPy object dtype for strings by default
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, all strings were stored in columns with NumPy object dtype by default.
+This release introduces an option ``future.infer_string`` that infers all
+strings as PyArrow backed strings with dtype ``"string[pyarrow_numpy]"`` instead.
+This is a new string dtype implementation that follows NumPy semantics in comparison
+operations and will return ``np.nan`` as the missing value indicator.
+Setting the option will also infer the dtype ``"string"`` as a :class:`StringDtype` with
+storage set to ``"pyarrow_numpy"``, ignoring the value behind the option
+``mode.string_storage``.
+
+This option only works if PyArrow is installed. PyArrow backed strings have a
+significantly reduced memory footprint and provide a big performance improvement
+compared to NumPy object (:issue:`54430`).
+
+The option can be enabled with:
+
+.. code-block:: python
+
+    pd.options.future.infer_string = True
+
+This behavior will become the default with pandas 3.0.
+
+.. _whatsnew_210.enhancements.reduction_extension_dtypes:
+
+DataFrame reductions preserve extension dtypes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, the results of DataFrame reductions
+(:meth:`DataFrame.sum` :meth:`DataFrame.mean` etc.) had NumPy dtypes, even when the DataFrames
+were of extension dtypes. pandas can now keep the dtypes when doing reductions over DataFrame
+columns with a common dtype (:issue:`52788`).
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [1]: df = pd.DataFrame({"a": [1, 1, 2, 1], "b": [np.nan, 2.0, 3.0, 4.0]}, dtype="Int64")
+    In [2]: df.sum()
+    Out[2]:
+    a    5
+    b    9
+    dtype: int64
+    In [3]: df = df.astype("int64[pyarrow]")
+    In [4]: df.sum()
+    Out[4]:
+    a    5
+    b    9
+    dtype: int64
+
+*New Behavior*
+
+.. ipython:: python
+
+    df = pd.DataFrame({"a": [1, 1, 2, 1], "b": [np.nan, 2.0, 3.0, 4.0]}, dtype="Int64")
+    df.sum()
+    df = df.astype("int64[pyarrow]")
+    df.sum()
+
+Notice that the dtype is now a masked dtype and PyArrow dtype, respectively, while previously it was a NumPy integer dtype.
+
+To allow DataFrame reductions to preserve extension dtypes, :meth:`.ExtensionArray._reduce` has gotten a new keyword parameter ``keepdims``. Calling :meth:`.ExtensionArray._reduce` with ``keepdims=True`` should return an array of length 1 along the reduction axis. In order to maintain backward compatibility, the parameter is not required, but will it become required in the future. If the parameter is not found in the signature, DataFrame reductions can not preserve extension dtypes. Also, if the parameter is not found, a ``FutureWarning`` will be emitted and type checkers like mypy may complain about the signature not being compatible with :meth:`.ExtensionArray._reduce`.
+
+.. _whatsnew_210.enhancements.cow:
+
+Copy-on-Write improvements
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- :meth:`Series.transform` not respecting Copy-on-Write when ``func`` modifies :class:`Series` inplace (:issue:`53747`)
+- Calling :meth:`Index.values` will now return a read-only NumPy array (:issue:`53704`)
+- Setting a :class:`Series` into a :class:`DataFrame` now creates a lazy instead of a deep copy (:issue:`53142`)
+- The :class:`DataFrame` constructor, when constructing a DataFrame from a dictionary
+  of Index objects and specifying ``copy=False``, will now use a lazy copy
+  of those Index objects for the columns of the DataFrame (:issue:`52947`)
+- A shallow copy of a Series or DataFrame (``df.copy(deep=False)``) will now also return
+  a shallow copy of the rows/columns :class:`Index` objects instead of only a shallow copy of
+  the data, i.e. the index of the result is no longer identical
+  (``df.copy(deep=False).index is df.index`` is no longer True) (:issue:`53721`)
+- :meth:`DataFrame.head` and :meth:`DataFrame.tail` will now return deep copies (:issue:`54011`)
+- Add lazy copy mechanism to :meth:`DataFrame.eval` (:issue:`53746`)
+
+- Trying to operate inplace on a temporary column selection
+  (for example, ``df["a"].fillna(100, inplace=True)``)
+  will now always raise a warning when Copy-on-Write is enabled. In this mode,
+  operating inplace like this will never work, since the selection behaves
+  as a temporary copy. This holds true for:
+
+  - DataFrame.update / Series.update
+  - DataFrame.fillna / Series.fillna
+  - DataFrame.replace / Series.replace
+  - DataFrame.clip / Series.clip
+  - DataFrame.where / Series.where
+  - DataFrame.mask / Series.mask
+  - DataFrame.interpolate / Series.interpolate
+  - DataFrame.ffill / Series.ffill
+  - DataFrame.bfill / Series.bfill
+
+.. _whatsnew_210.enhancements.map_na_action:
+
+New :meth:`DataFrame.map` method and support for ExtensionArrays
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`DataFrame.map` been added and :meth:`DataFrame.applymap` has been deprecated. :meth:`DataFrame.map` has the same functionality as :meth:`DataFrame.applymap`, but the new name better communicates that this is the :class:`DataFrame` version of :meth:`Series.map` (:issue:`52353`).
+
+When given a callable, :meth:`Series.map` applies the callable to all elements of the :class:`Series`.
+Similarly, :meth:`DataFrame.map` applies the callable to all elements of the :class:`DataFrame`,
+while :meth:`Index.map` applies the callable to all elements of the :class:`Index`.
+
+Frequently, it is not desirable to apply the callable to nan-like values of the array and to avoid doing
+that, the ``map`` method could be called with ``na_action="ignore"``, i.e. ``ser.map(func, na_action="ignore")``.
+However, ``na_action="ignore"`` was not implemented for many :class:`.ExtensionArray` and ``Index`` types
+and ``na_action="ignore"`` did not work correctly for any :class:`.ExtensionArray` subclass except the nullable numeric ones (i.e. with dtype :class:`Int64` etc.).
+
+``na_action="ignore"`` now works for all array types (:issue:`52219`, :issue:`51645`, :issue:`51809`, :issue:`51936`, :issue:`52033`; :issue:`52096`).
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [1]: ser = pd.Series(["a", "b", np.nan], dtype="category")
+    In [2]: ser.map(str.upper, na_action="ignore")
+    NotImplementedError
+    In [3]: df = pd.DataFrame(ser)
+    In [4]: df.applymap(str.upper, na_action="ignore")  # worked for DataFrame
+         0
+    0    A
+    1    B
+    2  NaN
+    In [5]: idx = pd.Index(ser)
+    In [6]: idx.map(str.upper, na_action="ignore")
+    TypeError: CategoricalIndex.map() got an unexpected keyword argument 'na_action'
+
+*New behavior*:
+
+.. ipython:: python
+
+    ser = pd.Series(["a", "b", np.nan], dtype="category")
+    ser.map(str.upper, na_action="ignore")
+    df = pd.DataFrame(ser)
+    df.map(str.upper, na_action="ignore")
+    idx = pd.Index(ser)
+    idx.map(str.upper, na_action="ignore")
+
+Also, note that :meth:`Categorical.map` implicitly has had its ``na_action`` set to ``"ignore"`` by default.
+This has been deprecated and the default for :meth:`Categorical.map` will change
+to ``na_action=None``, consistent with all the other array types.
+
+.. _whatsnew_210.enhancements.new_stack:
+
+New implementation of :meth:`DataFrame.stack`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas has reimplemented :meth:`DataFrame.stack`. To use the new implementation, pass the argument ``future_stack=True``. This will become the only option in pandas 3.0.
+
+The previous implementation had two main behavioral downsides.
+
+1. The previous implementation would unnecessarily introduce NA values into the result. The user could have NA values automatically removed by passing ``dropna=True`` (the default), but doing this could also remove NA values from the result that existed in the input. See the examples below.
+2. The previous implementation with ``sort=True`` (the default) would sometimes sort part of the resulting index, and sometimes not. If the input's columns are *not* a :class:`MultiIndex`, then the resulting index would never be sorted. If the columns are a :class:`MultiIndex`, then in most cases the level(s) in the resulting index that come from stacking the column level(s) would be sorted. In rare cases such level(s) would be sorted in a non-standard order, depending on how the columns were created.
+
+The new implementation (``future_stack=True``) will no longer unnecessarily introduce NA values when stacking multiple levels and will never sort. As such, the arguments ``dropna`` and ``sort`` are not utilized and must remain unspecified when using ``future_stack=True``. These arguments will be removed in the next major release.
+
+.. ipython:: python
+
+    columns = pd.MultiIndex.from_tuples([("B", "d"), ("A", "c")])
+    df = pd.DataFrame([[0, 2], [1, 3]], index=["z", "y"], columns=columns)
+    df
+
+In the previous version (``future_stack=False``), the default of ``dropna=True`` would remove unnecessarily introduced NA values but still coerce the dtype to ``float64`` in the process. In the new version, no NAs are introduced and so there is no coercion of the dtype.
+
+.. ipython:: python
+    :okwarning:
+
+    df.stack([0, 1], future_stack=False, dropna=True)
+    df.stack([0, 1], future_stack=True)
+
+If the input contains NA values, the previous version would drop those as well with ``dropna=True`` or introduce new NA values with ``dropna=False``. The new version persists all values from the input.
+
+.. ipython:: python
+    :okwarning:
+
+    df = pd.DataFrame([[0, 2], [np.nan, np.nan]], columns=columns)
+    df
+    df.stack([0, 1], future_stack=False, dropna=True)
+    df.stack([0, 1], future_stack=False, dropna=False)
+    df.stack([0, 1], future_stack=True)
+
+.. _whatsnew_210.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- :meth:`Series.ffill` and :meth:`Series.bfill` are now supported for objects with :class:`IntervalDtype` (:issue:`54247`)
+- Added ``filters`` parameter to :func:`read_parquet` to filter out data, compatible with both ``engines`` (:issue:`53212`)
+- :meth:`.Categorical.map` and :meth:`CategoricalIndex.map` now have a ``na_action`` parameter.
+  :meth:`.Categorical.map` implicitly had a default value of ``"ignore"`` for ``na_action``. This has formally been deprecated and will be changed to ``None`` in the future.
+  Also notice that :meth:`Series.map` has default ``na_action=None`` and calls to series with categorical data will now use ``na_action=None`` unless explicitly set otherwise (:issue:`44279`)
+- :class:`api.extensions.ExtensionArray` now has a :meth:`~api.extensions.ExtensionArray.map` method (:issue:`51809`)
+- :meth:`DataFrame.applymap` now uses the :meth:`~api.extensions.ExtensionArray.map` method of underlying :class:`api.extensions.ExtensionArray` instances (:issue:`52219`)
+- :meth:`MultiIndex.sort_values` now supports ``na_position`` (:issue:`51612`)
+- :meth:`MultiIndex.sortlevel` and :meth:`Index.sortlevel` gained a new keyword ``na_position`` (:issue:`51612`)
+- :meth:`arrays.DatetimeArray.map`, :meth:`arrays.TimedeltaArray.map` and :meth:`arrays.PeriodArray.map` can now take a ``na_action`` argument (:issue:`51644`)
+- :meth:`arrays.SparseArray.map` now supports ``na_action`` (:issue:`52096`).
+- :meth:`pandas.read_html` now supports the ``storage_options`` keyword when used with a URL, allowing users to add headers to the outbound HTTP request (:issue:`49944`)
+- Add :meth:`Index.diff` and :meth:`Index.round` (:issue:`19708`)
+- Add ``"latex-math"`` as an option to the ``escape`` argument of :class:`.Styler` which will not escape all characters between ``"\("`` and ``"\)"`` during formatting (:issue:`51903`)
+- Add dtype of categories to ``repr`` information of :class:`CategoricalDtype` (:issue:`52179`)
+- Adding ``engine_kwargs`` parameter to :func:`read_excel` (:issue:`52214`)
+- Classes that are useful for type-hinting have been added to the public API in the new submodule ``pandas.api.typing`` (:issue:`48577`)
+- Implemented :attr:`Series.dt.is_month_start`, :attr:`Series.dt.is_month_end`, :attr:`Series.dt.is_year_start`, :attr:`Series.dt.is_year_end`, :attr:`Series.dt.is_quarter_start`, :attr:`Series.dt.is_quarter_end`, :attr:`Series.dt.days_in_month`, :attr:`Series.dt.unit`, :attr:`Series.dt.normalize`, :meth:`Series.dt.day_name`, :meth:`Series.dt.month_name`, :meth:`Series.dt.tz_convert` for :class:`ArrowDtype` with ``pyarrow.timestamp`` (:issue:`52388`, :issue:`51718`)
+- :meth:`.DataFrameGroupBy.agg` and :meth:`.DataFrameGroupBy.transform` now support grouping by multiple keys when the index is not a :class:`MultiIndex` for ``engine="numba"`` (:issue:`53486`)
+- :meth:`.SeriesGroupBy.agg` and :meth:`.DataFrameGroupBy.agg` now support passing in multiple functions for ``engine="numba"`` (:issue:`53486`)
+- :meth:`.SeriesGroupBy.transform` and :meth:`.DataFrameGroupBy.transform` now support passing in a string as the function for ``engine="numba"`` (:issue:`53579`)
+- :meth:`DataFrame.stack` gained the ``sort`` keyword to dictate whether the resulting :class:`MultiIndex` levels are sorted (:issue:`15105`)
+- :meth:`DataFrame.unstack` gained the ``sort`` keyword to dictate whether the resulting :class:`MultiIndex` levels are sorted (:issue:`15105`)
+- :meth:`Series.explode` now supports PyArrow-backed list types (:issue:`53602`)
+- :meth:`Series.str.join` now supports ``ArrowDtype(pa.string())`` (:issue:`53646`)
+- Add ``validate`` parameter to :meth:`Categorical.from_codes` (:issue:`50975`)
+- Added :meth:`.ExtensionArray.interpolate` used by :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` (:issue:`53659`)
+- Added ``engine_kwargs`` parameter to :meth:`DataFrame.to_excel` (:issue:`53220`)
+- Implemented :func:`api.interchange.from_dataframe` for :class:`DatetimeTZDtype` (:issue:`54239`)
+- Implemented ``__from_arrow__`` on :class:`DatetimeTZDtype` (:issue:`52201`)
+- Implemented ``__pandas_priority__`` to allow custom types to take precedence over :class:`DataFrame`, :class:`Series`, :class:`Index`, or :class:`.ExtensionArray` for arithmetic operations, :ref:`see the developer guide <extending.pandas_priority>` (:issue:`48347`)
+- Improve error message when having incompatible columns using :meth:`DataFrame.merge` (:issue:`51861`)
+- Improve error message when setting :class:`DataFrame` with wrong number of columns through :meth:`DataFrame.isetitem` (:issue:`51701`)
+- Improved error handling when using :meth:`DataFrame.to_json` with incompatible ``index`` and ``orient`` arguments (:issue:`52143`)
+- Improved error message when creating a DataFrame with empty data (0 rows), no index and an incorrect number of columns (:issue:`52084`)
+- Improved error message when providing an invalid ``index`` or ``offset`` argument to :class:`.VariableOffsetWindowIndexer` (:issue:`54379`)
+- Let :meth:`DataFrame.to_feather` accept a non-default :class:`Index` and non-string column names (:issue:`51787`)
+- Added a new parameter ``by_row`` to :meth:`Series.apply` and :meth:`DataFrame.apply`. When set to ``False`` the supplied callables will always operate on the whole Series or DataFrame (:issue:`53400`, :issue:`53601`).
+- :meth:`DataFrame.shift` and :meth:`Series.shift` now allow shifting by multiple periods by supplying a list of periods (:issue:`44424`)
+- Groupby aggregations with ``numba`` (such as :meth:`.DataFrameGroupBy.sum`) now can preserve the dtype of the input instead of casting to ``float64`` (:issue:`44952`)
+- Improved error message when :meth:`.DataFrameGroupBy.agg` failed (:issue:`52930`)
+- Many read/to_* functions, such as :meth:`DataFrame.to_pickle` and :func:`read_csv`, support forwarding compression arguments to ``lzma.LZMAFile`` (:issue:`52979`)
+- Reductions :meth:`Series.argmax`, :meth:`Series.argmin`, :meth:`Series.idxmax`, :meth:`Series.idxmin`, :meth:`Index.argmax`, :meth:`Index.argmin`, :meth:`DataFrame.idxmax`, :meth:`DataFrame.idxmin` are now supported for object-dtype (:issue:`4279`, :issue:`18021`, :issue:`40685`, :issue:`43697`)
+- :meth:`DataFrame.to_parquet` and :func:`read_parquet` will now write and read ``attrs`` respectively (:issue:`54346`)
+- :meth:`Index.all` and :meth:`Index.any` with floating dtypes and timedelta64 dtypes no longer raise ``TypeError``, matching the :meth:`Series.all` and :meth:`Series.any` behavior (:issue:`54566`)
+- :meth:`Series.cummax`, :meth:`Series.cummin` and :meth:`Series.cumprod` are now supported for pyarrow dtypes with pyarrow version 13.0 and above (:issue:`52085`)
+- Added support for the DataFrame Consortium Standard (:issue:`54383`)
+- Performance improvement in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` (:issue:`51722`)
+- PyArrow-backed integer dtypes now support bitwise operations (:issue:`54495`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_210.api_breaking.deps:
+
+Increased minimum version for Python
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 2.1.0 supports Python 3.9 and higher.
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++----------------------+-----------------+----------+---------+
+| Package              | Minimum Version | Required | Changed |
++======================+=================+==========+=========+
+| numpy                | 1.22.4          |    X     |    X    |
++----------------------+-----------------+----------+---------+
+| mypy (dev)           | 1.4.1           |          |    X    |
++----------------------+-----------------+----------+---------+
+| beautifulsoup4       | 4.11.1          |          |    X    |
++----------------------+-----------------+----------+---------+
+| bottleneck           | 1.3.4           |          |    X    |
++----------------------+-----------------+----------+---------+
+| dataframe-api-compat | 0.1.7           |          |    X    |
++----------------------+-----------------+----------+---------+
+| fastparquet          | 0.8.1           |          |    X    |
++----------------------+-----------------+----------+---------+
+| fsspec               | 2022.05.0       |          |    X    |
++----------------------+-----------------+----------+---------+
+| hypothesis           | 6.46.1          |          |    X    |
++----------------------+-----------------+----------+---------+
+| gcsfs                | 2022.05.0       |          |    X    |
++----------------------+-----------------+----------+---------+
+| jinja2               | 3.1.2           |          |    X    |
++----------------------+-----------------+----------+---------+
+| lxml                 | 4.8.0           |          |    X    |
++----------------------+-----------------+----------+---------+
+| numba                | 0.55.2          |          |    X    |
++----------------------+-----------------+----------+---------+
+| numexpr              | 2.8.0           |          |    X    |
++----------------------+-----------------+----------+---------+
+| openpyxl             | 3.0.10          |          |    X    |
++----------------------+-----------------+----------+---------+
+| pandas-gbq           | 0.17.5          |          |    X    |
++----------------------+-----------------+----------+---------+
+| psycopg2             | 2.9.3           |          |    X    |
++----------------------+-----------------+----------+---------+
+| pyreadstat           | 1.1.5           |          |    X    |
++----------------------+-----------------+----------+---------+
+| pyqt5                | 5.15.6          |          |    X    |
++----------------------+-----------------+----------+---------+
+| pytables             | 3.7.0           |          |    X    |
++----------------------+-----------------+----------+---------+
+| pytest               | 7.3.2           |          |    X    |
++----------------------+-----------------+----------+---------+
+| python-snappy        | 0.6.1           |          |    X    |
++----------------------+-----------------+----------+---------+
+| pyxlsb               | 1.0.9           |          |    X    |
++----------------------+-----------------+----------+---------+
+| s3fs                 | 2022.05.0       |          |    X    |
++----------------------+-----------------+----------+---------+
+| scipy                | 1.8.1           |          |    X    |
++----------------------+-----------------+----------+---------+
+| sqlalchemy           | 1.4.36          |          |    X    |
++----------------------+-----------------+----------+---------+
+| tabulate             | 0.8.10          |          |    X    |
++----------------------+-----------------+----------+---------+
+| xarray               | 2022.03.0       |          |    X    |
++----------------------+-----------------+----------+---------+
+| xlsxwriter           | 3.0.3           |          |    X    |
++----------------------+-----------------+----------+---------+
+| zstandard            | 0.17.0          |          |    X    |
++----------------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_210.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- :class:`arrays.PandasArray` has been renamed :class:`.NumpyExtensionArray` and the attached dtype name changed from ``PandasDtype`` to ``NumpyEADtype``; importing ``PandasArray`` still works until the next major version (:issue:`53694`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+Deprecated silent upcasting in setitem-like Series operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+PDEP-6: https://pandas.pydata.org/pdeps/0006-ban-upcasting.html
+
+Setitem-like operations on Series (or DataFrame columns) which silently upcast the dtype are
+deprecated and show a warning. Examples of affected operations are:
+
+- ``ser.fillna('foo', inplace=True)``
+- ``ser.where(ser.isna(), 'foo', inplace=True)``
+- ``ser.iloc[indexer] = 'foo'``
+- ``ser.loc[indexer] = 'foo'``
+- ``df.iloc[indexer, 0] = 'foo'``
+- ``df.loc[indexer, 'a'] = 'foo'``
+- ``ser[indexer] = 'foo'``
+
+where ``ser`` is a :class:`Series`, ``df`` is a :class:`DataFrame`, and ``indexer``
+could be a slice, a mask, a single value, a list or array of values, or any other
+allowed indexer.
+
+In a future version, these will raise an error and you should cast to a common dtype first.
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+  In [1]: ser = pd.Series([1, 2, 3])
+
+  In [2]: ser
+  Out[2]:
+  0    1
+  1    2
+  2    3
+  dtype: int64
+
+  In [3]: ser[0] = 'not an int64'
+
+  In [4]: ser
+  Out[4]:
+  0    not an int64
+  1               2
+  2               3
+  dtype: object
+
+*New behavior*:
+
+.. code-block:: ipython
+
+  In [1]: ser = pd.Series([1, 2, 3])
+
+  In [2]: ser
+  Out[2]:
+  0    1
+  1    2
+  2    3
+  dtype: int64
+
+  In [3]: ser[0] = 'not an int64'
+  FutureWarning:
+    Setting an item of incompatible dtype is deprecated and will raise an error in a future version of pandas.
+    Value 'not an int64' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
+
+  In [4]: ser
+  Out[4]:
+  0    not an int64
+  1               2
+  2               3
+  dtype: object
+
+To retain the current behaviour, in the case above you could cast ``ser`` to ``object`` dtype first:
+
+.. ipython:: python
+
+  ser = pd.Series([1, 2, 3])
+  ser = ser.astype('object')
+  ser[0] = 'not an int64'
+  ser
+
+Depending on the use-case, it might be more appropriate to cast to a different dtype.
+In the following, for example, we cast to ``float64``:
+
+.. ipython:: python
+
+  ser = pd.Series([1, 2, 3])
+  ser = ser.astype('float64')
+  ser[0] = 1.1
+  ser
+
+For further reading, please see https://pandas.pydata.org/pdeps/0006-ban-upcasting.html.
+
+Deprecated parsing datetimes with mixed time zones
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Parsing datetimes with mixed time zones is deprecated and shows a warning unless user passes ``utc=True`` to :func:`to_datetime` (:issue:`50887`)
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+  In [7]: data = ["2020-01-01 00:00:00+06:00", "2020-01-01 00:00:00+01:00"]
+
+  In [8]:  pd.to_datetime(data, utc=False)
+  Out[8]:
+  Index([2020-01-01 00:00:00+06:00, 2020-01-01 00:00:00+01:00], dtype='object')
+
+*New behavior*:
+
+.. code-block:: ipython
+
+  In [9]: pd.to_datetime(data, utc=False)
+  FutureWarning:
+    In a future version of pandas, parsing datetimes with mixed time zones will raise
+    a warning unless `utc=True`. Please specify `utc=True` to opt in to the new behaviour
+    and silence this warning. To create a `Series` with mixed offsets and `object` dtype,
+    please use `apply` and `datetime.datetime.strptime`.
+  Index([2020-01-01 00:00:00+06:00, 2020-01-01 00:00:00+01:00], dtype='object')
+
+In order to silence this warning and avoid an error in a future version of pandas,
+please specify ``utc=True``:
+
+.. ipython:: python
+
+    data = ["2020-01-01 00:00:00+06:00", "2020-01-01 00:00:00+01:00"]
+    pd.to_datetime(data, utc=True)
+
+To create a ``Series`` with mixed offsets and ``object`` dtype, please use ``apply``
+and ``datetime.datetime.strptime``:
+
+.. ipython:: python
+
+    import datetime as dt
+
+    data = ["2020-01-01 00:00:00+06:00", "2020-01-01 00:00:00+01:00"]
+    pd.Series(data).apply(lambda x: dt.datetime.strptime(x, '%Y-%m-%d %H:%M:%S%z'))
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+- Deprecated :attr:`.DataFrameGroupBy.dtypes`, check ``dtypes`` on the underlying object instead (:issue:`51045`)
+- Deprecated :attr:`DataFrame._data` and :attr:`Series._data`, use public APIs instead (:issue:`33333`)
+- Deprecated :func:`concat` behavior when any of the objects being concatenated have length 0; in the past the dtypes of empty objects were ignored when determining the resulting dtype, in a future version they will not (:issue:`39122`)
+- Deprecated :meth:`.Categorical.to_list`, use ``obj.tolist()`` instead (:issue:`51254`)
+- Deprecated :meth:`.DataFrameGroupBy.all` and :meth:`.DataFrameGroupBy.any` with datetime64 or :class:`PeriodDtype` values, matching the :class:`Series` and :class:`DataFrame` deprecations (:issue:`34479`)
+- Deprecated ``axis=1`` in :meth:`DataFrame.ewm`, :meth:`DataFrame.rolling`, :meth:`DataFrame.expanding`, transpose before calling the method instead (:issue:`51778`)
+- Deprecated ``axis=1`` in :meth:`DataFrame.groupby` and in :class:`Grouper` constructor, do ``frame.T.groupby(...)`` instead (:issue:`51203`)
+- Deprecated ``broadcast_axis`` keyword in :meth:`Series.align` and :meth:`DataFrame.align`, upcast before calling ``align`` with ``left = DataFrame({col: left for col in right.columns}, index=right.index)`` (:issue:`51856`)
+- Deprecated ``downcast`` keyword in :meth:`Index.fillna` (:issue:`53956`)
+- Deprecated ``fill_method`` and ``limit`` keywords in :meth:`DataFrame.pct_change`, :meth:`Series.pct_change`, :meth:`.DataFrameGroupBy.pct_change`, and :meth:`.SeriesGroupBy.pct_change`, explicitly call e.g. :meth:`DataFrame.ffill` or :meth:`DataFrame.bfill` before calling ``pct_change`` instead (:issue:`53491`)
+- Deprecated ``method``, ``limit``, and ``fill_axis`` keywords in :meth:`DataFrame.align` and :meth:`Series.align`, explicitly call :meth:`DataFrame.fillna` or :meth:`Series.fillna` on the alignment results instead (:issue:`51856`)
+- Deprecated ``quantile`` keyword in :meth:`.Rolling.quantile` and :meth:`.Expanding.quantile`, renamed to ``q`` instead (:issue:`52550`)
+- Deprecated accepting slices in :meth:`DataFrame.take`, call ``obj[slicer]`` or pass a sequence of integers instead (:issue:`51539`)
+- Deprecated behavior of :meth:`DataFrame.idxmax`, :meth:`DataFrame.idxmin`, :meth:`Series.idxmax`, :meth:`Series.idxmin` in with all-NA entries or any-NA and ``skipna=False``; in a future version these will raise ``ValueError`` (:issue:`51276`)
+- Deprecated explicit support for subclassing :class:`Index` (:issue:`45289`)
+- Deprecated making functions given to :meth:`Series.agg` attempt to operate on each element in the :class:`Series` and only operate on the whole :class:`Series` if the elementwise operations failed. In the future, functions given to :meth:`Series.agg` will always operate on the whole :class:`Series` only. To keep the current behavior, use :meth:`Series.transform` instead (:issue:`53325`)
+- Deprecated making the functions in a list of functions given to :meth:`DataFrame.agg` attempt to operate on each element in the :class:`DataFrame` and only operate on the columns of the :class:`DataFrame` if the elementwise operations failed. To keep the current behavior, use :meth:`DataFrame.transform` instead (:issue:`53325`)
+- Deprecated passing a :class:`DataFrame` to :meth:`DataFrame.from_records`, use :meth:`DataFrame.set_index` or :meth:`DataFrame.drop` instead (:issue:`51353`)
+- Deprecated silently dropping unrecognized timezones when parsing strings to datetimes (:issue:`18702`)
+- Deprecated the ``axis`` keyword in :meth:`DataFrame.ewm`, :meth:`Series.ewm`, :meth:`DataFrame.rolling`, :meth:`Series.rolling`, :meth:`DataFrame.expanding`, :meth:`Series.expanding` (:issue:`51778`)
+- Deprecated the ``axis`` keyword in :meth:`DataFrame.resample`, :meth:`Series.resample` (:issue:`51778`)
+- Deprecated the ``downcast`` keyword in :meth:`Series.interpolate`, :meth:`DataFrame.interpolate`, :meth:`Series.fillna`, :meth:`DataFrame.fillna`, :meth:`Series.ffill`, :meth:`DataFrame.ffill`, :meth:`Series.bfill`, :meth:`DataFrame.bfill` (:issue:`40988`)
+- Deprecated the behavior of :func:`concat` with both ``len(keys) != len(objs)``, in a future version this will raise instead of truncating to the shorter of the two sequences (:issue:`43485`)
+- Deprecated the behavior of :meth:`Series.argsort` in the presence of NA values; in a future version these will be sorted at the end instead of giving -1 (:issue:`54219`)
+- Deprecated the default of ``observed=False`` in :meth:`DataFrame.groupby` and :meth:`Series.groupby`; this will default to ``True`` in a future version (:issue:`43999`)
+- Deprecating pinning ``group.name`` to each group in :meth:`.SeriesGroupBy.aggregate` aggregations; if your operation requires utilizing the groupby keys, iterate over the groupby object instead (:issue:`41090`)
+- Deprecated the ``axis`` keyword in :meth:`.DataFrameGroupBy.idxmax`, :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.fillna`, :meth:`.DataFrameGroupBy.take`, :meth:`.DataFrameGroupBy.skew`, :meth:`.DataFrameGroupBy.rank`, :meth:`.DataFrameGroupBy.cumprod`, :meth:`.DataFrameGroupBy.cumsum`, :meth:`.DataFrameGroupBy.cummax`, :meth:`.DataFrameGroupBy.cummin`, :meth:`.DataFrameGroupBy.pct_change`, :meth:`.DataFrameGroupBy.diff`, :meth:`.DataFrameGroupBy.shift`, and :meth:`.DataFrameGroupBy.corrwith`; for ``axis=1`` operate on the underlying :class:`DataFrame` instead (:issue:`50405`, :issue:`51046`)
+- Deprecated :class:`.DataFrameGroupBy` with ``as_index=False`` not including groupings in the result when they are not columns of the DataFrame (:issue:`49519`)
+- Deprecated :func:`is_categorical_dtype`, use ``isinstance(obj.dtype, pd.CategoricalDtype)`` instead (:issue:`52527`)
+- Deprecated :func:`is_datetime64tz_dtype`, check ``isinstance(dtype, pd.DatetimeTZDtype)`` instead (:issue:`52607`)
+- Deprecated :func:`is_int64_dtype`, check ``dtype == np.dtype(np.int64)`` instead (:issue:`52564`)
+- Deprecated :func:`is_interval_dtype`, check ``isinstance(dtype, pd.IntervalDtype)`` instead (:issue:`52607`)
+- Deprecated :func:`is_period_dtype`, check ``isinstance(dtype, pd.PeriodDtype)`` instead (:issue:`52642`)
+- Deprecated :func:`is_sparse`, check ``isinstance(dtype, pd.SparseDtype)`` instead (:issue:`52642`)
+- Deprecated :meth:`.Styler.applymap_index`. Use the new :meth:`.Styler.map_index` method instead (:issue:`52708`)
+- Deprecated :meth:`.Styler.applymap`. Use the new :meth:`.Styler.map` method instead (:issue:`52708`)
+- Deprecated :meth:`DataFrame.applymap`. Use the new :meth:`DataFrame.map` method instead (:issue:`52353`)
+- Deprecated :meth:`DataFrame.swapaxes` and :meth:`Series.swapaxes`, use :meth:`DataFrame.transpose` or :meth:`Series.transpose` instead (:issue:`51946`)
+- Deprecated ``freq`` parameter in :class:`.PeriodArray` constructor, pass ``dtype`` instead (:issue:`52462`)
+- Deprecated allowing non-standard inputs in :func:`take`, pass either a ``numpy.ndarray``, :class:`.ExtensionArray`, :class:`Index`, or :class:`Series` (:issue:`52981`)
+- Deprecated allowing non-standard sequences for :func:`isin`, :func:`value_counts`, :func:`unique`, :func:`factorize`, case to one of ``numpy.ndarray``, :class:`Index`, :class:`.ExtensionArray`, or :class:`Series` before calling (:issue:`52986`)
+- Deprecated behavior of :class:`DataFrame` reductions ``sum``, ``prod``, ``std``, ``var``, ``sem`` with ``axis=None``, in a future version this will operate over both axes returning a scalar instead of behaving like ``axis=0``; note this also affects numpy functions e.g. ``np.sum(df)`` (:issue:`21597`)
+- Deprecated behavior of :func:`concat` when :class:`DataFrame` has columns that are all-NA, in a future version these will not be discarded when determining the resulting dtype (:issue:`40893`)
+- Deprecated behavior of :meth:`Series.dt.to_pydatetime`, in a future version this will return a :class:`Series` containing python ``datetime`` objects instead of an ``ndarray`` of datetimes; this matches the behavior of other :attr:`Series.dt` properties (:issue:`20306`)
+- Deprecated logical operations (``|``, ``&``, ``^``) between pandas objects and dtype-less sequences (e.g. ``list``, ``tuple``), wrap a sequence in a :class:`Series` or NumPy array before operating instead (:issue:`51521`)
+- Deprecated parameter ``convert_type`` in :meth:`Series.apply` (:issue:`52140`)
+- Deprecated passing a dictionary to :meth:`.SeriesGroupBy.agg`; pass a list of aggregations instead (:issue:`50684`)
+- Deprecated the ``fastpath`` keyword in :class:`Categorical` constructor, use :meth:`Categorical.from_codes` instead (:issue:`20110`)
+- Deprecated the behavior of :func:`is_bool_dtype` returning ``True`` for object-dtype :class:`Index` of bool objects (:issue:`52680`)
+- Deprecated the methods :meth:`Series.bool` and :meth:`DataFrame.bool` (:issue:`51749`)
+- Deprecated unused ``closed`` and ``normalize`` keywords in the :class:`DatetimeIndex` constructor (:issue:`52628`)
+- Deprecated unused ``closed`` keyword in the :class:`TimedeltaIndex` constructor (:issue:`52628`)
+- Deprecated logical operation between two non boolean :class:`Series` with different indexes always coercing the result to bool dtype. In a future version, this will maintain the return type of the inputs (:issue:`52500`, :issue:`52538`)
+- Deprecated :class:`Period` and :class:`PeriodDtype` with ``BDay`` freq, use a :class:`DatetimeIndex` with ``BDay`` freq instead (:issue:`53446`)
+- Deprecated :func:`value_counts`, use ``pd.Series(obj).value_counts()`` instead (:issue:`47862`)
+- Deprecated :meth:`Series.first` and :meth:`DataFrame.first`; create a mask and filter using ``.loc`` instead (:issue:`45908`)
+- Deprecated :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` for object-dtype (:issue:`53631`)
+- Deprecated :meth:`Series.last` and :meth:`DataFrame.last`; create a mask and filter using ``.loc`` instead (:issue:`53692`)
+- Deprecated allowing arbitrary ``fill_value`` in :class:`SparseDtype`, in a future version the ``fill_value`` will need to be compatible with the ``dtype.subtype``, either a scalar that can be held by that subtype or ``NaN`` for integer or bool subtypes (:issue:`23124`)
+- Deprecated allowing bool dtype in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile`, consistent with the :meth:`Series.quantile` and :meth:`DataFrame.quantile` behavior (:issue:`51424`)
+- Deprecated behavior of :func:`.testing.assert_series_equal` and :func:`.testing.assert_frame_equal` considering NA-like values (e.g. ``NaN`` vs ``None`` as equivalent) (:issue:`52081`)
+- Deprecated bytes input to :func:`read_excel`. To read a file path, use a string or path-like object (:issue:`53767`)
+- Deprecated constructing :class:`.SparseArray` from scalar data, pass a sequence instead (:issue:`53039`)
+- Deprecated falling back to filling when ``value`` is not specified in :meth:`DataFrame.replace` and :meth:`Series.replace` with non-dict-like ``to_replace`` (:issue:`33302`)
+- Deprecated literal json input to :func:`read_json`. Wrap literal json string input in ``io.StringIO`` instead (:issue:`53409`)
+- Deprecated literal string input to :func:`read_xml`. Wrap literal string/bytes input in ``io.StringIO`` / ``io.BytesIO`` instead (:issue:`53767`)
+- Deprecated literal string/bytes input to :func:`read_html`. Wrap literal string/bytes input in ``io.StringIO`` / ``io.BytesIO`` instead (:issue:`53767`)
+- Deprecated option ``mode.use_inf_as_na``, convert inf entries to ``NaN`` before instead (:issue:`51684`)
+- Deprecated parameter ``obj`` in :meth:`.DataFrameGroupBy.get_group` (:issue:`53545`)
+- Deprecated positional indexing on :class:`Series` with :meth:`Series.__getitem__` and :meth:`Series.__setitem__`, in a future version ``ser[item]`` will *always* interpret ``item`` as a label, not a position (:issue:`50617`)
+- Deprecated replacing builtin and NumPy functions in ``.agg``, ``.apply``, and ``.transform``; use the corresponding string alias (e.g. ``"sum"`` for ``sum`` or ``np.sum``) instead (:issue:`53425`)
+- Deprecated strings ``T``, ``t``, ``L`` and ``l`` denoting units in :func:`to_timedelta` (:issue:`52536`)
+- Deprecated the "method" and "limit" keywords in ``.ExtensionArray.fillna``, implement ``_pad_or_backfill`` instead (:issue:`53621`)
+- Deprecated the ``method`` and ``limit`` keywords in :meth:`DataFrame.replace` and :meth:`Series.replace` (:issue:`33302`)
+- Deprecated the ``method`` and ``limit`` keywords on :meth:`Series.fillna`, :meth:`DataFrame.fillna`, :meth:`.SeriesGroupBy.fillna`, :meth:`.DataFrameGroupBy.fillna`, and :meth:`.Resampler.fillna`, use ``obj.bfill()`` or ``obj.ffill()`` instead (:issue:`53394`)
+- Deprecated the behavior of :meth:`Series.__getitem__`, :meth:`Series.__setitem__`, :meth:`DataFrame.__getitem__`, :meth:`DataFrame.__setitem__` with an integer slice on objects with a floating-dtype index, in a future version this will be treated as *positional* indexing (:issue:`49612`)
+- Deprecated the use of non-supported datetime64 and timedelta64 resolutions with :func:`pandas.array`. Supported resolutions are: "s", "ms", "us", "ns" resolutions (:issue:`53058`)
+- Deprecated values ``"pad"``, ``"ffill"``, ``"bfill"``, ``"backfill"`` for :meth:`Series.interpolate` and :meth:`DataFrame.interpolate`, use ``obj.ffill()`` or ``obj.bfill()`` instead (:issue:`53581`)
+- Deprecated the behavior of :meth:`Index.argmax`, :meth:`Index.argmin`, :meth:`Series.argmax`, :meth:`Series.argmin` with either all-NAs and ``skipna=True`` or any-NAs and ``skipna=False`` returning -1; in a future version this will raise ``ValueError`` (:issue:`33941`, :issue:`33942`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_sql` except ``name`` and ``con`` (:issue:`54229`)
+- Deprecated silently ignoring ``fill_value`` when passing both ``freq`` and ``fill_value`` to :meth:`DataFrame.shift`, :meth:`Series.shift` and :meth:`.DataFrameGroupBy.shift`; in a future version this will raise ``ValueError`` (:issue:`53832`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :func:`concat` with homogeneous ``np.float64`` or ``np.float32`` dtypes (:issue:`52685`)
+- Performance improvement in :func:`factorize` for object columns not containing strings (:issue:`51921`)
+- Performance improvement in :func:`read_orc` when reading a remote URI file path (:issue:`51609`)
+- Performance improvement in :func:`read_parquet` and :meth:`DataFrame.to_parquet` when reading a remote file with ``engine="pyarrow"`` (:issue:`51609`)
+- Performance improvement in :func:`read_parquet` on string columns when using ``use_nullable_dtypes=True`` (:issue:`47345`)
+- Performance improvement in :meth:`DataFrame.clip` and :meth:`Series.clip` (:issue:`51472`)
+- Performance improvement in :meth:`DataFrame.filter` when ``items`` is given (:issue:`52941`)
+- Performance improvement in :meth:`DataFrame.first_valid_index` and :meth:`DataFrame.last_valid_index` for extension array dtypes (:issue:`51549`)
+- Performance improvement in :meth:`DataFrame.where` when ``cond`` is backed by an extension dtype (:issue:`51574`)
+- Performance improvement in :meth:`MultiIndex.set_levels` and :meth:`MultiIndex.set_codes` when ``verify_integrity=True`` (:issue:`51873`)
+- Performance improvement in :meth:`MultiIndex.sortlevel` when ``ascending`` is a list (:issue:`51612`)
+- Performance improvement in :meth:`Series.combine_first` (:issue:`51777`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.fillna` when array does not contain nulls (:issue:`51635`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.isna` when array has zero nulls or is all nulls (:issue:`51630`)
+- Performance improvement when parsing strings to ``boolean[pyarrow]`` dtype (:issue:`51730`)
+- Performance improvement when searching an :class:`Index` sliced from other indexes (:issue:`51738`)
+- Performance improvement in :func:`concat` (:issue:`52291`, :issue:`52290`)
+- :class:`Period`'s default formatter (``period_format``) is now significantly (~twice) faster. This improves performance of ``str(Period)``, ``repr(Period)``, and :meth:`.Period.strftime(fmt=None)`, as well as ``.PeriodArray.strftime(fmt=None)``, ``.PeriodIndex.strftime(fmt=None)`` and ``.PeriodIndex.format(fmt=None)``. ``to_csv`` operations involving :class:`.PeriodArray` or :class:`PeriodIndex` with default ``date_format`` are also significantly accelerated (:issue:`51459`)
+- Performance improvement accessing :attr:`arrays.IntegerArrays.dtype` & :attr:`arrays.FloatingArray.dtype` (:issue:`52998`)
+- Performance improvement for :class:`.DataFrameGroupBy`/:class:`.SeriesGroupBy` aggregations (e.g. :meth:`.DataFrameGroupBy.sum`) with ``engine="numba"`` (:issue:`53731`)
+- Performance improvement in :class:`DataFrame` reductions with ``axis=1`` and extension dtypes (:issue:`54341`)
+- Performance improvement in :class:`DataFrame` reductions with ``axis=None`` and extension dtypes (:issue:`54308`)
+- Performance improvement in :class:`MultiIndex` and multi-column operations (e.g. :meth:`DataFrame.sort_values`, :meth:`DataFrame.groupby`, :meth:`Series.unstack`) when index/column values are already sorted (:issue:`53806`)
+- Performance improvement in :class:`Series` reductions (:issue:`52341`)
+- Performance improvement in :func:`concat` when ``axis=1`` and objects have different indexes (:issue:`52541`)
+- Performance improvement in :func:`concat` when the concatenation axis is a :class:`MultiIndex` (:issue:`53574`)
+- Performance improvement in :func:`merge` for PyArrow backed strings (:issue:`54443`)
+- Performance improvement in :func:`read_csv` with ``engine="c"`` (:issue:`52632`)
+- Performance improvement in :meth:`.ArrowExtensionArray.to_numpy` (:issue:`52525`)
+- Performance improvement in :meth:`.DataFrameGroupBy.groups` (:issue:`53088`)
+- Performance improvement in :meth:`DataFrame.astype` when ``dtype`` is an extension dtype (:issue:`54299`)
+- Performance improvement in :meth:`DataFrame.iloc` when input is an single integer and dataframe is backed by extension dtypes (:issue:`54508`)
+- Performance improvement in :meth:`DataFrame.isin` for extension dtypes (:issue:`53514`)
+- Performance improvement in :meth:`DataFrame.loc` when selecting rows and columns (:issue:`53014`)
+- Performance improvement in :meth:`DataFrame.transpose` when transposing a DataFrame with a single PyArrow dtype (:issue:`54224`)
+- Performance improvement in :meth:`DataFrame.transpose` when transposing a DataFrame with a single masked dtype, e.g. :class:`Int64` (:issue:`52836`)
+- Performance improvement in :meth:`Series.add` for PyArrow string and binary dtypes (:issue:`53150`)
+- Performance improvement in :meth:`Series.corr` and :meth:`Series.cov` for extension dtypes (:issue:`52502`)
+- Performance improvement in :meth:`Series.drop_duplicates` for ``ArrowDtype`` (:issue:`54667`).
+- Performance improvement in :meth:`Series.ffill`, :meth:`Series.bfill`, :meth:`DataFrame.ffill`, :meth:`DataFrame.bfill` with PyArrow dtypes (:issue:`53950`)
+- Performance improvement in :meth:`Series.str.get_dummies` for PyArrow-backed strings (:issue:`53655`)
+- Performance improvement in :meth:`Series.str.get` for PyArrow-backed strings (:issue:`53152`)
+- Performance improvement in :meth:`Series.str.split` with ``expand=True`` for PyArrow-backed strings (:issue:`53585`)
+- Performance improvement in :meth:`Series.to_numpy` when dtype is a NumPy float dtype and ``na_value`` is ``np.nan`` (:issue:`52430`)
+- Performance improvement in :meth:`~arrays.ArrowExtensionArray.astype` when converting from a PyArrow timestamp or duration dtype to NumPy (:issue:`53326`)
+- Performance improvement in various :class:`MultiIndex` set and indexing operations (:issue:`53955`)
+- Performance improvement when doing various reshaping operations on :class:`arrays.IntegerArray` & :class:`arrays.FloatingArray` by avoiding doing unnecessary validation (:issue:`53013`)
+- Performance improvement when indexing with PyArrow timestamp and duration dtypes (:issue:`53368`)
+- Performance improvement when passing an array to :meth:`RangeIndex.take`, :meth:`DataFrame.loc`, or :meth:`DataFrame.iloc` and the DataFrame is using a RangeIndex (:issue:`53387`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in :meth:`CategoricalIndex.remove_categories` where ordered categories would not be maintained (:issue:`53935`).
+- Bug in :meth:`Series.astype` with ``dtype="category"`` for nullable arrays with read-only null value masks (:issue:`53658`)
+- Bug in :meth:`Series.map` , where the value of the ``na_action`` parameter was not used if the series held a :class:`Categorical` (:issue:`22527`).
+
+Datetimelike
+^^^^^^^^^^^^
+- :meth:`DatetimeIndex.map` with ``na_action="ignore"`` now works as expected (:issue:`51644`)
+- :meth:`DatetimeIndex.slice_indexer` now raises ``KeyError`` for non-monotonic indexes if either of the slice bounds is not in the index; this behaviour was previously deprecated but inconsistently handled (:issue:`53983`)
+- Bug in :class:`DateOffset` which had inconsistent behavior when multiplying a :class:`DateOffset` object by a constant (:issue:`47953`)
+- Bug in :func:`date_range` when ``freq`` was a :class:`DateOffset` with ``nanoseconds`` (:issue:`46877`)
+- Bug in :func:`to_datetime` converting :class:`Series` or :class:`DataFrame` containing :class:`arrays.ArrowExtensionArray` of PyArrow timestamps to numpy datetimes (:issue:`52545`)
+- Bug in :meth:`.DatetimeArray.map` and :meth:`DatetimeIndex.map`, where the supplied callable operated array-wise instead of element-wise (:issue:`51977`)
+- Bug in :meth:`DataFrame.to_sql` raising ``ValueError`` for PyArrow-backed date like dtypes (:issue:`53854`)
+- Bug in :meth:`Timestamp.date`, :meth:`Timestamp.isocalendar`, :meth:`Timestamp.timetuple`, and :meth:`Timestamp.toordinal` were returning incorrect results for inputs outside those supported by the Python standard library's datetime module (:issue:`53668`)
+- Bug in :meth:`Timestamp.round` with values close to the implementation bounds returning incorrect results instead of raising ``OutOfBoundsDatetime`` (:issue:`51494`)
+- Bug in constructing a :class:`Series` or :class:`DataFrame` from a datetime or timedelta scalar always inferring nanosecond resolution instead of inferring from the input (:issue:`52212`)
+- Bug in constructing a :class:`Timestamp` from a string representing a time without a date inferring an incorrect unit (:issue:`54097`)
+- Bug in constructing a :class:`Timestamp` with ``ts_input=pd.NA`` raising ``TypeError`` (:issue:`45481`)
+- Bug in parsing datetime strings with weekday but no day e.g. "2023 Sept Thu" incorrectly raising ``AttributeError`` instead of ``ValueError`` (:issue:`52659`)
+- Bug in the repr for :class:`Series` when dtype is a timezone aware datetime with non-nanosecond resolution raising ``OutOfBoundsDatetime`` (:issue:`54623`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :class:`TimedeltaIndex` division or multiplication leading to ``.freq`` of "0 Days" instead of ``None`` (:issue:`51575`)
+- Bug in :class:`Timedelta` with NumPy ``timedelta64`` objects not properly raising ``ValueError`` (:issue:`52806`)
+- Bug in :func:`to_timedelta` converting :class:`Series` or :class:`DataFrame` containing :class:`ArrowDtype` of ``pyarrow.duration`` to NumPy ``timedelta64`` (:issue:`54298`)
+- Bug in :meth:`Timedelta.__hash__`, raising an ``OutOfBoundsTimedelta`` on certain large values of second resolution (:issue:`54037`)
+- Bug in :meth:`Timedelta.round` with values close to the implementation bounds returning incorrect results instead of raising ``OutOfBoundsTimedelta`` (:issue:`51494`)
+- Bug in :meth:`TimedeltaIndex.map` with ``na_action="ignore"`` (:issue:`51644`)
+- Bug in :meth:`arrays.TimedeltaArray.map` and :meth:`TimedeltaIndex.map`, where the supplied callable operated array-wise instead of element-wise (:issue:`51977`)
+
+Timezones
+^^^^^^^^^
+- Bug in :func:`infer_freq` that raises ``TypeError`` for ``Series`` of timezone-aware timestamps (:issue:`52456`)
+- Bug in :meth:`DatetimeTZDtype.base` that always returns a NumPy dtype with nanosecond resolution (:issue:`52705`)
+
+Numeric
+^^^^^^^
+- Bug in :class:`RangeIndex` setting ``step`` incorrectly when being the subtrahend with minuend a numeric value (:issue:`53255`)
+- Bug in :meth:`Series.corr` and :meth:`Series.cov` raising ``AttributeError`` for masked dtypes (:issue:`51422`)
+- Bug when calling :meth:`Series.kurt` and :meth:`Series.skew` on NumPy data of all zero returning a Python type instead of a NumPy type (:issue:`53482`)
+- Bug in :meth:`Series.mean`, :meth:`DataFrame.mean` with object-dtype values containing strings that can be converted to numbers (e.g. "2") returning incorrect numeric results; these now raise ``TypeError`` (:issue:`36703`, :issue:`44008`)
+- Bug in :meth:`DataFrame.corrwith` raising ``NotImplementedError`` for PyArrow-backed dtypes (:issue:`52314`)
+- Bug in :meth:`DataFrame.size` and :meth:`Series.size` returning 64-bit integer instead of a Python int (:issue:`52897`)
+- Bug in :meth:`DateFrame.dot` returning ``object`` dtype for :class:`ArrowDtype` data (:issue:`53979`)
+- Bug in :meth:`Series.any`, :meth:`Series.all`, :meth:`DataFrame.any`, and :meth:`DataFrame.all` had the default value of ``bool_only`` set to ``None`` instead of ``False``; this change should have no impact on users (:issue:`53258`)
+- Bug in :meth:`Series.corr` and :meth:`Series.cov` raising ``AttributeError`` for masked dtypes (:issue:`51422`)
+- Bug in :meth:`Series.median` and :meth:`DataFrame.median` with object-dtype values containing strings that can be converted to numbers (e.g. "2") returning incorrect numeric results; these now raise ``TypeError`` (:issue:`34671`)
+- Bug in :meth:`Series.sum` converting dtype ``uint64`` to ``int64`` (:issue:`53401`)
+
+
+Conversion
+^^^^^^^^^^
+- Bug in :func:`DataFrame.style.to_latex` and :func:`DataFrame.style.to_html` if the DataFrame contains integers with more digits than can be represented by floating point double precision (:issue:`52272`)
+- Bug in :func:`array`  when given a ``datetime64`` or ``timedelta64`` dtype with unit of "s", "us", or "ms" returning :class:`.NumpyExtensionArray` instead of :class:`.DatetimeArray` or :class:`.TimedeltaArray` (:issue:`52859`)
+- Bug in :func:`array`  when given an empty list and no dtype returning :class:`.NumpyExtensionArray` instead of :class:`.FloatingArray` (:issue:`54371`)
+- Bug in :meth:`.ArrowDtype.numpy_dtype` returning nanosecond units for non-nanosecond ``pyarrow.timestamp`` and ``pyarrow.duration`` types (:issue:`51800`)
+- Bug in :meth:`DataFrame.__repr__` incorrectly raising a ``TypeError`` when the dtype of a column is ``np.record`` (:issue:`48526`)
+- Bug in :meth:`DataFrame.info` raising  ``ValueError`` when ``use_numba`` is set (:issue:`51922`)
+- Bug in :meth:`DataFrame.insert` raising ``TypeError`` if ``loc`` is ``np.int64`` (:issue:`53193`)
+- Bug in :meth:`HDFStore.select` loses precision of large int when stored and retrieved (:issue:`54186`)
+- Bug in :meth:`Series.astype` not supporting ``object_`` (:issue:`54251`)
+
+Strings
+^^^^^^^
+- Bug in :meth:`Series.str` that did not raise a  ``TypeError`` when iterated (:issue:`54173`)
+- Bug in ``repr`` for :class:`DataFrame`` with string-dtype columns (:issue:`54797`)
+
+Interval
+^^^^^^^^
+- :meth:`IntervalIndex.get_indexer` and :meth:`IntervalIndex.get_indexer_nonunique` raising if ``target`` is read-only array (:issue:`53703`)
+- Bug in :class:`IntervalDtype` where the object could be kept alive when deleted (:issue:`54184`)
+- Bug in :func:`interval_range` where a float ``step`` would produce incorrect intervals from floating point artifacts (:issue:`54477`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`DataFrame.__setitem__` losing dtype when setting a :class:`DataFrame` into duplicated columns (:issue:`53143`)
+- Bug in :meth:`DataFrame.__setitem__` with a boolean mask and :meth:`DataFrame.putmask` with mixed non-numeric dtypes and a value other than ``NaN`` incorrectly raising ``TypeError`` (:issue:`53291`)
+- Bug in :meth:`DataFrame.iloc` when using ``nan`` as the only element (:issue:`52234`)
+- Bug in :meth:`Series.loc` casting :class:`Series` to ``np.dnarray`` when assigning :class:`Series` at predefined index of ``object`` dtype :class:`Series` (:issue:`48933`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`DataFrame.interpolate` failing to fill across data when ``method`` is ``"pad"``, ``"ffill"``, ``"bfill"``, or ``"backfill"`` (:issue:`53898`)
+- Bug in :meth:`DataFrame.interpolate` ignoring ``inplace`` when :class:`DataFrame` is empty (:issue:`53199`)
+- Bug in :meth:`Series.idxmin`, :meth:`Series.idxmax`, :meth:`DataFrame.idxmin`, :meth:`DataFrame.idxmax` with a :class:`DatetimeIndex` index containing ``NaT`` incorrectly returning ``NaN`` instead of ``NaT`` (:issue:`43587`)
+- Bug in :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` failing to raise on invalid ``downcast`` keyword, which can be only ``None`` or ``"infer"`` (:issue:`53103`)
+- Bug in :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` with complex dtype incorrectly failing to fill ``NaN`` entries (:issue:`53635`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`MultiIndex.set_levels` not preserving dtypes for :class:`Categorical` (:issue:`52125`)
+- Bug in displaying a :class:`MultiIndex` with a long element (:issue:`52960`)
+
+I/O
+^^^
+- :meth:`DataFrame.to_orc` now raising ``ValueError`` when non-default :class:`Index` is given (:issue:`51828`)
+- :meth:`DataFrame.to_sql` now raising ``ValueError`` when the name param is left empty while using SQLAlchemy to connect (:issue:`52675`)
+- Bug in :func:`json_normalize` could not parse metadata fields list type (:issue:`37782`)
+- Bug in :func:`read_csv` where it would error when ``parse_dates`` was set to a list or dictionary with ``engine="pyarrow"`` (:issue:`47961`)
+- Bug in :func:`read_csv` with ``engine="pyarrow"`` raising when specifying a ``dtype`` with ``index_col`` (:issue:`53229`)
+- Bug in :func:`read_hdf` not properly closing store after an ``IndexError`` is raised (:issue:`52781`)
+- Bug in :func:`read_html` where style elements were read into DataFrames (:issue:`52197`)
+- Bug in :func:`read_html` where tail texts were removed together with elements containing ``display:none`` style (:issue:`51629`)
+- Bug in :func:`read_sql_table` raising an exception when reading a view (:issue:`52969`)
+- Bug in :func:`read_sql` when reading multiple timezone aware columns with the same column name (:issue:`44421`)
+- Bug in :func:`read_xml` stripping whitespace in string data (:issue:`53811`)
+- Bug in :meth:`DataFrame.to_html` where ``colspace`` was incorrectly applied in case of multi index columns (:issue:`53885`)
+- Bug in :meth:`DataFrame.to_html` where conversion for an empty :class:`DataFrame` with complex dtype raised a ``ValueError`` (:issue:`54167`)
+- Bug in :meth:`DataFrame.to_json` where :class:`.DateTimeArray`/:class:`.DateTimeIndex` with non nanosecond precision could not be serialized correctly (:issue:`53686`)
+- Bug when writing and reading empty Stata dta files where dtype information was lost (:issue:`46240`)
+- Bug where ``bz2`` was treated as a hard requirement (:issue:`53857`)
+
+Period
+^^^^^^
+- Bug in :class:`PeriodDtype` constructor failing to raise ``TypeError`` when no argument is passed or when ``None`` is passed (:issue:`27388`)
+- Bug in :class:`PeriodDtype` constructor incorrectly returning the same ``normalize`` for different :class:`DateOffset` ``freq`` inputs (:issue:`24121`)
+- Bug in :class:`PeriodDtype` constructor raising ``ValueError`` instead of ``TypeError`` when an invalid type is passed (:issue:`51790`)
+- Bug in :class:`PeriodDtype` where the object could be kept alive when deleted (:issue:`54184`)
+- Bug in :func:`read_csv` not processing empty strings as a null value, with ``engine="pyarrow"`` (:issue:`52087`)
+- Bug in :func:`read_csv` returning ``object`` dtype columns instead of ``float64`` dtype columns with ``engine="pyarrow"`` for columns that are all null with ``engine="pyarrow"`` (:issue:`52087`)
+- Bug in :meth:`Period.now` not accepting the ``freq`` parameter as a keyword argument (:issue:`53369`)
+- Bug in :meth:`PeriodIndex.map` with ``na_action="ignore"`` (:issue:`51644`)
+- Bug in :meth:`arrays.PeriodArray.map` and :meth:`PeriodIndex.map`, where the supplied callable operated array-wise instead of element-wise (:issue:`51977`)
+- Bug in incorrectly allowing construction of :class:`Period` or :class:`PeriodDtype` with :class:`CustomBusinessDay` freq; use :class:`BusinessDay` instead (:issue:`52534`)
+
+Plotting
+^^^^^^^^
+- Bug in :meth:`Series.plot` when invoked with ``color=None`` (:issue:`51953`)
+- Fixed UserWarning in :meth:`DataFrame.plot.scatter` when invoked with ``c="b"`` (:issue:`53908`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmax` returns wrong dtype when used on an empty DataFrameGroupBy or SeriesGroupBy (:issue:`51423`)
+- Bug in :meth:`DataFrame.groupby.rank` on nullable datatypes when passing ``na_option="bottom"`` or ``na_option="top"`` (:issue:`54206`)
+- Bug in :meth:`DataFrame.resample` and :meth:`Series.resample` in incorrectly allowing non-fixed ``freq`` when resampling on a :class:`TimedeltaIndex` (:issue:`51896`)
+- Bug in :meth:`DataFrame.resample` and :meth:`Series.resample` losing time zone when resampling empty data (:issue:`53664`)
+- Bug in :meth:`DataFrame.resample` and :meth:`Series.resample` where ``origin`` has no effect in resample when values are outside of axis  (:issue:`53662`)
+- Bug in weighted rolling aggregations when specifying ``min_periods=0`` (:issue:`51449`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` where, when the index of the
+  grouped :class:`Series` or :class:`DataFrame` was a :class:`DatetimeIndex`, :class:`TimedeltaIndex`
+  or :class:`PeriodIndex`, and the ``groupby`` method was given a function as its first argument,
+  the function operated on the whole index rather than each element of the index (:issue:`51979`)
+- Bug in :meth:`.DataFrameGroupBy.agg` with lists not respecting ``as_index=False`` (:issue:`52849`)
+- Bug in :meth:`.DataFrameGroupBy.apply` causing an error to be raised when the input :class:`DataFrame` was subset as a :class:`DataFrame` after groupby (``[['a']]`` and not ``['a']``) and the given callable returned :class:`Series` that were not all indexed the same (:issue:`52444`)
+- Bug in :meth:`.DataFrameGroupBy.apply` raising a ``TypeError`` when selecting multiple columns and providing a function that returns ``np.ndarray`` results (:issue:`18930`)
+- Bug in :meth:`.DataFrameGroupBy.groups` and :meth:`.SeriesGroupBy.groups` with a datetime key in conjunction with another key produced an incorrect number of group keys (:issue:`51158`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` may implicitly sort the result index with ``sort=False`` (:issue:`53009`)
+- Bug in :meth:`.SeriesGroupBy.size` where the dtype would be ``np.int64`` for data with :class:`ArrowDtype` or masked dtypes (e.g. ``Int64``) (:issue:`53831`)
+- Bug in :meth:`DataFrame.groupby` with column selection on the resulting groupby object not returning names as tuples when grouping by a list consisting of a single element (:issue:`53500`)
+- Bug in :meth:`.DataFrameGroupBy.var` and :meth:`.SeriesGroupBy.var` failing to raise ``TypeError`` when called with datetime64, timedelta64 or :class:`PeriodDtype` values (:issue:`52128`, :issue:`53045`)
+- Bug in :meth:`.DataFrameGroupBy.resample` with ``kind="period"`` raising ``AttributeError`` (:issue:`24103`)
+- Bug in :meth:`.Resampler.ohlc` with empty object returning a :class:`Series` instead of empty :class:`DataFrame` (:issue:`42902`)
+- Bug in :meth:`.SeriesGroupBy.count` and :meth:`.DataFrameGroupBy.count` where the dtype would be ``np.int64`` for data with :class:`ArrowDtype` or masked dtypes (e.g. ``Int64``) (:issue:`53831`)
+- Bug in :meth:`.SeriesGroupBy.nth` and :meth:`.DataFrameGroupBy.nth` after performing column selection when using ``dropna="any"`` or ``dropna="all"`` would not subset columns (:issue:`53518`)
+- Bug in :meth:`.SeriesGroupBy.nth` and :meth:`.DataFrameGroupBy.nth` raised after performing column selection when using ``dropna="any"`` or ``dropna="all"`` resulted in rows being dropped (:issue:`53518`)
+- Bug in :meth:`.SeriesGroupBy.sum` and :meth:`.DataFrameGroupBy.sum` summing ``np.inf + np.inf`` and ``(-np.inf) + (-np.inf)`` to ``np.nan`` instead of ``np.inf`` and ``-np.inf`` respectively (:issue:`53606`)
+- Bug in :meth:`Series.groupby` raising an error when grouped :class:`Series` has a :class:`DatetimeIndex` index and a :class:`Series` with a name that is a month is given to the ``by`` argument (:issue:`48509`)
+
+Reshaping
+^^^^^^^^^
+- Bug in :func:`concat` coercing to ``object`` dtype when one column has ``pa.null()`` dtype (:issue:`53702`)
+- Bug in :func:`crosstab` when ``dropna=False`` would not keep ``np.nan`` in the result (:issue:`10772`)
+- Bug in :func:`melt` where the ``variable`` column would lose extension dtypes (:issue:`54297`)
+- Bug in :func:`merge_asof` raising ``KeyError`` for extension dtypes (:issue:`52904`)
+- Bug in :func:`merge_asof` raising ``ValueError`` for data backed by read-only ndarrays (:issue:`53513`)
+- Bug in :func:`merge_asof` with ``left_index=True`` or ``right_index=True`` with mismatched index dtypes giving incorrect results in some cases instead of raising ``MergeError`` (:issue:`53870`)
+- Bug in :func:`merge` when merging on integer ``ExtensionDtype`` and float NumPy dtype raising ``TypeError`` (:issue:`46178`)
+- Bug in :meth:`DataFrame.agg` and :meth:`Series.agg` on non-unique columns would return incorrect type when dist-like argument passed in (:issue:`51099`)
+- Bug in :meth:`DataFrame.combine_first` ignoring other's columns if ``other`` is empty (:issue:`53792`)
+- Bug in :meth:`DataFrame.idxmin` and :meth:`DataFrame.idxmax`, where the axis dtype would be lost for empty frames (:issue:`53265`)
+- Bug in :meth:`DataFrame.merge` not merging correctly when having ``MultiIndex`` with single level (:issue:`52331`)
+- Bug in :meth:`DataFrame.stack` losing extension dtypes when columns is a :class:`MultiIndex` and frame contains mixed dtypes (:issue:`45740`)
+- Bug in :meth:`DataFrame.stack` sorting columns lexicographically (:issue:`53786`)
+- Bug in :meth:`DataFrame.transpose` inferring dtype for object column (:issue:`51546`)
+- Bug in :meth:`Series.combine_first` converting ``int64`` dtype to ``float64`` and losing precision on very large integers (:issue:`51764`)
+- Bug when joining empty :class:`DataFrame` objects, where the joined index would be a :class:`RangeIndex` instead of the joined index type (:issue:`52777`)
+
+Sparse
+^^^^^^
+- Bug in :class:`SparseDtype` constructor failing to raise ``TypeError`` when given an incompatible ``dtype`` for its subtype, which must be a NumPy dtype (:issue:`53160`)
+- Bug in :meth:`arrays.SparseArray.map` allowed the fill value to be included in the sparse values (:issue:`52095`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :class:`.ArrowStringArray` constructor raises ``ValueError`` with dictionary types of strings (:issue:`54074`)
+- Bug in :class:`DataFrame` constructor not copying :class:`Series` with extension dtype when given in dict (:issue:`53744`)
+- Bug in :class:`~arrays.ArrowExtensionArray` converting pandas non-nanosecond temporal objects from non-zero values to zero values (:issue:`53171`)
+- Bug in :meth:`Series.quantile` for PyArrow temporal types raising ``ArrowInvalid`` (:issue:`52678`)
+- Bug in :meth:`Series.rank` returning wrong order for small values with ``Float64`` dtype (:issue:`52471`)
+- Bug in :meth:`Series.unique` for boolean ``ArrowDtype`` with ``NA`` values (:issue:`54667`)
+- Bug in :meth:`~arrays.ArrowExtensionArray.__iter__` and :meth:`~arrays.ArrowExtensionArray.__getitem__` returning python datetime and timedelta objects for non-nano dtypes (:issue:`53326`)
+- Bug in :meth:`~arrays.ArrowExtensionArray.factorize` returning incorrect uniques for a ``pyarrow.dictionary`` type ``pyarrow.chunked_array`` with more than one chunk (:issue:`54844`)
+- Bug when passing an :class:`ExtensionArray` subclass to ``dtype`` keywords. This will now raise a ``UserWarning`` to encourage passing an instance instead (:issue:`31356`, :issue:`54592`)
+- Bug where the :class:`DataFrame` repr would not work when a column had an :class:`ArrowDtype` with a ``pyarrow.ExtensionDtype`` (:issue:`54063`)
+- Bug where the ``__from_arrow__`` method of masked ExtensionDtypes (e.g. :class:`Float64Dtype`, :class:`BooleanDtype`) would not accept PyArrow arrays of type ``pyarrow.null()`` (:issue:`52223`)
+
+Styler
+^^^^^^
+- Bug in :meth:`.Styler._copy` calling overridden methods in subclasses of :class:`.Styler` (:issue:`52728`)
+
+Metadata
+^^^^^^^^
+- Fixed metadata propagation in :meth:`DataFrame.max`, :meth:`DataFrame.min`, :meth:`DataFrame.prod`, :meth:`DataFrame.mean`, :meth:`Series.mode`, :meth:`DataFrame.median`, :meth:`DataFrame.sem`, :meth:`DataFrame.skew`, :meth:`DataFrame.kurt` (:issue:`28283`)
+- Fixed metadata propagation in :meth:`DataFrame.squeeze`, and :meth:`DataFrame.describe` (:issue:`28283`)
+- Fixed metadata propagation in :meth:`DataFrame.std` (:issue:`28283`)
+
+Other
+^^^^^
+- Bug in :class:`.FloatingArray.__contains__` with ``NaN`` item incorrectly returning ``False`` when ``NaN`` values are present (:issue:`52840`)
+- Bug in :class:`DataFrame` and :class:`Series` raising for data of complex dtype when ``NaN`` values are present (:issue:`53627`)
+- Bug in :class:`DatetimeIndex` where ``repr`` of index passed with time does not print time is midnight and non-day based freq(:issue:`53470`)
+- Bug in :func:`.testing.assert_frame_equal` and :func:`.testing.assert_series_equal` now throw assertion error for two unequal sets (:issue:`51727`)
+- Bug in :func:`.testing.assert_frame_equal` checks category dtypes even when asked not to check index type (:issue:`52126`)
+- Bug in :func:`api.interchange.from_dataframe` was not respecting ``allow_copy`` argument (:issue:`54322`)
+- Bug in :func:`api.interchange.from_dataframe` was raising during interchanging from non-pandas tz-aware data containing null values (:issue:`54287`)
+- Bug in :func:`api.interchange.from_dataframe` when converting an empty DataFrame object (:issue:`53155`)
+- Bug in :func:`from_dummies` where the resulting :class:`Index` did not match the original :class:`Index` (:issue:`54300`)
+- Bug in :func:`from_dummies` where the resulting data would always be ``object`` dtype instead of the dtype of the columns (:issue:`54300`)
+- Bug in :meth:`.DataFrameGroupBy.first`, :meth:`.DataFrameGroupBy.last`, :meth:`.SeriesGroupBy.first`, and :meth:`.SeriesGroupBy.last` where an empty group would return ``np.nan`` instead of the corresponding :class:`.ExtensionArray` NA value (:issue:`39098`)
+- Bug in :meth:`DataFrame.pivot_table` with casting the mean of ints back to an int (:issue:`16676`)
+- Bug in :meth:`DataFrame.reindex` with a ``fill_value`` that should be inferred with a :class:`ExtensionDtype` incorrectly inferring ``object`` dtype (:issue:`52586`)
+- Bug in :meth:`DataFrame.shift` with ``axis=1`` on a :class:`DataFrame` with a single :class:`ExtensionDtype` column giving incorrect results (:issue:`53832`)
+- Bug in :meth:`Index.sort_values` when a ``key`` is passed (:issue:`52764`)
+- Bug in :meth:`Series.align`, :meth:`DataFrame.align`, :meth:`Series.reindex`, :meth:`DataFrame.reindex`, :meth:`Series.interpolate`, :meth:`DataFrame.interpolate`, incorrectly failing to raise with method="asfreq" (:issue:`53620`)
+- Bug in :meth:`Series.argsort` failing to raise when an invalid ``axis`` is passed (:issue:`54257`)
+- Bug in :meth:`Series.map` when giving a callable to an empty series, the returned series had ``object`` dtype. It now keeps the original dtype (:issue:`52384`)
+- Bug in :meth:`Series.memory_usage` when ``deep=True`` throw an error with Series of objects and the returned value is incorrect, as it does not take into account GC corrections (:issue:`51858`)
+- Bug in :meth:`period_range` the default behavior when freq was not passed as an argument was incorrect(:issue:`53687`)
+- Fixed incorrect ``__name__`` attribute of ``pandas._libs.json`` (:issue:`52898`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_210.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.0.3..v2.1.0
diff --git a/pandas/doc/source/whatsnew/v2.1.1.rst b/pandas/doc/source/whatsnew/v2.1.1.rst
new file mode 100644
index 0000000000000..6101333ae760c
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.1.1.rst
@@ -0,0 +1,55 @@
+.. _whatsnew_211:
+
+What's new in 2.1.1 (September 20, 2023)
+----------------------------------------
+
+These are the changes in pandas 2.1.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_211.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :func:`concat` when :class:`DataFrame` 's have two different extension dtypes (:issue:`54848`)
+- Fixed regression in :func:`merge` when merging over a PyArrow string index (:issue:`54894`)
+- Fixed regression in :func:`read_csv` when ``usecols`` is given and ``dtypes`` is a dict for ``engine="python"`` (:issue:`54868`)
+- Fixed regression in :func:`read_csv` when ``delim_whitespace`` is True (:issue:`54918`, :issue:`54931`)
+- Fixed regression in :meth:`.GroupBy.get_group` raising for ``axis=1`` (:issue:`54858`)
+- Fixed regression in :meth:`DataFrame.__setitem__` raising ``AssertionError`` when setting a :class:`Series` with a partial :class:`MultiIndex` (:issue:`54875`)
+- Fixed regression in :meth:`DataFrame.filter` not respecting the order of elements for ``filter`` (:issue:`54980`)
+- Fixed regression in :meth:`DataFrame.to_sql` not roundtripping datetime columns correctly for sqlite (:issue:`54877`)
+- Fixed regression in :meth:`DataFrameGroupBy.agg` when aggregating a DataFrame with duplicate column names using a dictionary (:issue:`55006`)
+- Fixed regression in :meth:`MultiIndex.append` raising when appending overlapping :class:`IntervalIndex` levels (:issue:`54934`)
+- Fixed regression in :meth:`Series.drop_duplicates` for PyArrow strings (:issue:`54904`)
+- Fixed regression in :meth:`Series.interpolate` raising when ``fill_value`` was given (:issue:`54920`)
+- Fixed regression in :meth:`Series.value_counts` raising for numeric data if ``bins`` was specified (:issue:`54857`)
+- Fixed regression in comparison operations for PyArrow backed columns not propagating exceptions correctly (:issue:`54944`)
+- Fixed regression when comparing a :class:`Series` with ``datetime64`` dtype with ``None`` (:issue:`54870`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_211.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug for :class:`ArrowDtype` raising ``NotImplementedError`` for fixed-size list (:issue:`55000`)
+- Fixed bug in :meth:`DataFrame.stack` with ``future_stack=True`` and columns a non-:class:`MultiIndex` consisting of tuples (:issue:`54948`)
+- Fixed bug in :meth:`Series.dt.tz` with :class:`ArrowDtype` where a string was returned instead of a ``tzinfo`` object (:issue:`55003`)
+- Fixed bug in :meth:`Series.pct_change` and :meth:`DataFrame.pct_change` showing unnecessary ``FutureWarning`` (:issue:`54981`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_211.other:
+
+Other
+~~~~~
+- Reverted the deprecation that disallowed :meth:`Series.apply` returning a :class:`DataFrame` when the passed-in callable returns a :class:`Series` object (:issue:`52116`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_211.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.1.0..v2.1.1
diff --git a/pandas/doc/source/whatsnew/v2.1.2.rst b/pandas/doc/source/whatsnew/v2.1.2.rst
new file mode 100644
index 0000000000000..a13a273b01257
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.1.2.rst
@@ -0,0 +1,70 @@
+.. _whatsnew_212:
+
+What's new in 2.1.2 (October 26, 2023)
+---------------------------------------
+
+These are the changes in pandas 2.1.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_212.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+- Reverted deprecation of ``fill_method=None`` in :meth:`DataFrame.pct_change`, :meth:`Series.pct_change`, :meth:`DataFrameGroupBy.pct_change`, and :meth:`SeriesGroupBy.pct_change`; the values ``'backfill'``, ``'bfill'``, ``'pad'``, and ``'ffill'`` are still deprecated (:issue:`53491`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_212.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression in :meth:`DataFrame.join` where result has missing values and dtype is arrow backed string (:issue:`55348`)
+- Fixed regression in :meth:`~DataFrame.rolling` where non-nanosecond index or ``on`` column would produce incorrect results (:issue:`55026`, :issue:`55106`, :issue:`55299`)
+- Fixed regression in :meth:`DataFrame.resample` which was extrapolating back to ``origin`` when ``origin`` was outside its bounds (:issue:`55064`)
+- Fixed regression in :meth:`DataFrame.sort_index` which was not sorting correctly when the index was a sliced :class:`MultiIndex` (:issue:`55379`)
+- Fixed regression in :meth:`DataFrameGroupBy.agg` and :meth:`SeriesGroupBy.agg` where if the option ``compute.use_numba`` was set to True, groupby methods not supported by the numba engine would raise a ``TypeError`` (:issue:`55520`)
+- Fixed performance regression with wide DataFrames, typically involving methods where all columns were accessed individually (:issue:`55256`, :issue:`55245`)
+- Fixed regression in :func:`merge_asof` raising ``TypeError`` for ``by`` with datetime and timedelta dtypes (:issue:`55453`)
+- Fixed regression in :func:`read_parquet` when reading a file with a string column consisting of more than 2 GB of string data and using the ``"string"`` dtype (:issue:`55606`)
+- Fixed regression in :meth:`DataFrame.to_sql` not roundtripping datetime columns correctly for sqlite when using ``detect_types`` (:issue:`55554`)
+- Fixed regression in construction of certain DataFrame or Series subclasses (:issue:`54922`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_212.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug in :class:`.DataFrameGroupBy` reductions not preserving object dtype when ``infer_string`` is set (:issue:`55620`)
+- Fixed bug in :meth:`.SeriesGroupBy.value_counts` returning incorrect dtype for string columns (:issue:`55627`)
+- Fixed bug in :meth:`Categorical.equals` if other has arrow backed string dtype (:issue:`55364`)
+- Fixed bug in :meth:`DataFrame.__setitem__` not inferring string dtype for zero-dimensional array with ``infer_string=True`` (:issue:`55366`)
+- Fixed bug in :meth:`DataFrame.idxmin` and :meth:`DataFrame.idxmax` raising for arrow dtypes (:issue:`55368`)
+- Fixed bug in :meth:`DataFrame.interpolate` raising incorrect error message (:issue:`55347`)
+- Fixed bug in :meth:`Index.insert` raising when inserting ``None`` into :class:`Index` with ``dtype="string[pyarrow_numpy]"`` (:issue:`55365`)
+- Fixed bug in :meth:`Series.all`  and :meth:`Series.any` not treating missing values correctly for ``dtype="string[pyarrow_numpy]"`` (:issue:`55367`)
+- Fixed bug in :meth:`Series.floordiv` for :class:`ArrowDtype` (:issue:`55561`)
+- Fixed bug in :meth:`Series.mode` not sorting values for arrow backed string dtype (:issue:`55621`)
+- Fixed bug in :meth:`Series.rank` for ``string[pyarrow_numpy]`` dtype (:issue:`55362`)
+- Fixed bug in :meth:`Series.str.extractall` for :class:`ArrowDtype` dtype being converted to object (:issue:`53846`)
+- Fixed bug where PDEP-6 warning about setting an item of an incompatible dtype was being shown when creating a new conditional column (:issue:`55025`)
+- Silence ``Period[B]`` warnings introduced by :issue:`53446` during normal plotting activity (:issue:`55138`)
+- Fixed bug in :class:`Series` constructor not inferring string dtype when ``NA`` is the first value and ``infer_string`` is set (:issue:` 55655`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_212.other:
+
+Other
+~~~~~
+- Fixed non-working installation of optional dependency group ``output_formatting``. Replacing underscore ``_`` with a dash ``-`` fixes broken dependency resolution. A correct way to use now is ``pip install pandas[output-formatting]``.
+-
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_212.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.1.1..v2.1.2
diff --git a/pandas/doc/source/whatsnew/v2.1.3.rst b/pandas/doc/source/whatsnew/v2.1.3.rst
new file mode 100644
index 0000000000000..af626895a9e0e
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.1.3.rst
@@ -0,0 +1,33 @@
+.. _whatsnew_213:
+
+What's new in 2.1.3 (November 10, 2023)
+---------------------------------------
+
+These are the changes in pandas 2.1.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_213.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed infinite recursion from operations that return a new object on some DataFrame subclasses (:issue:`55763`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_213.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :meth:`DatetimeIndex.diff` raising ``TypeError`` (:issue:`55080`)
+- Bug in :meth:`Index.isin` raising for Arrow backed string and ``None`` value (:issue:`55821`)
+- Fix :func:`read_parquet` and :func:`read_feather` for `CVE-2023-47248 <https://www.cve.org/CVERecord?id=CVE-2023-47248>`__ (:issue:`55894`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_213.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.1.2..v2.1.3|HEAD
diff --git a/pandas/doc/source/whatsnew/v2.1.4.rst b/pandas/doc/source/whatsnew/v2.1.4.rst
new file mode 100644
index 0000000000000..73b1103c1bd37
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.1.4.rst
@@ -0,0 +1,45 @@
+.. _whatsnew_214:
+
+What's new in 2.1.4 (December 8, 2023)
+---------------------------------------
+
+These are the changes in pandas 2.1.4. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_214.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed regression when trying to read a pickled pandas :class:`DataFrame` from pandas 1.3 (:issue:`55137`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_214.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :class:`Series` constructor raising DeprecationWarning when ``index`` is a list of :class:`Series` (:issue:`55228`)
+- Bug in :class:`Series` when trying to cast date-like string inputs to :class:`ArrowDtype` of ``pyarrow.timestamp`` (:issue:`56266`)
+- Bug in :class:`Timestamp` construction with ``ts_input="now"`` or ``ts_input="today"`` giving a different unit from :meth:`Timestamp.now` or :meth:`Timestamp.today` (:issue:`55879`)
+- Bug in :meth:`Index.__getitem__` returning wrong result for Arrow dtypes and negative stepsize (:issue:`55832`)
+- Fixed bug in :func:`read_csv` not respecting object dtype when ``infer_string`` option is set (:issue:`56047`)
+- Fixed bug in :func:`to_numeric` converting to extension dtype for ``string[pyarrow_numpy]`` dtype (:issue:`56179`)
+- Fixed bug in :meth:`.DataFrameGroupBy.min` and :meth:`.DataFrameGroupBy.max` not preserving extension dtype for empty object (:issue:`55619`)
+- Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
+- Fixed bug in :meth:`DataFrame.to_hdf` raising when columns have ``StringDtype`` (:issue:`55088`)
+- Fixed bug in :meth:`Index.insert` casting object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
+- Fixed bug in :meth:`Series.__ne__` resulting in False for comparison between ``NA`` and string value for ``dtype="string[pyarrow_numpy]"`` (:issue:`56122`)
+- Fixed bug in :meth:`Series.mode` not keeping object dtype when ``infer_string`` is set (:issue:`56183`)
+- Fixed bug in :meth:`Series.reset_index` not preserving object dtype when ``infer_string`` is set (:issue:`56160`)
+- Fixed bug in :meth:`Series.str.split` and :meth:`Series.str.rsplit` when ``pat=None`` for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`56271`)
+- Fixed bug in :meth:`Series.str.translate` losing object dtype when string option is set (:issue:`56152`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_214.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.1.3..v2.1.4
diff --git a/pandas/doc/source/whatsnew/v2.2.0.rst b/pandas/doc/source/whatsnew/v2.2.0.rst
new file mode 100644
index 0000000000000..329ef2859f56f
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.2.0.rst
@@ -0,0 +1,951 @@
+.. _whatsnew_220:
+
+What's new in 2.2.0 (January 19, 2024)
+--------------------------------------
+
+These are the changes in pandas 2.2.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_220.upcoming_changes:
+
+Upcoming changes in pandas 3.0
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+pandas 3.0 will bring two bigger changes to the default behavior of pandas.
+
+Copy-on-Write
+^^^^^^^^^^^^^
+
+The currently optional mode Copy-on-Write will be enabled by default in pandas 3.0. There
+won't be an option to keep the current behavior enabled. The new behavioral semantics are
+explained in the :ref:`user guide about Copy-on-Write <copy_on_write>`.
+
+The new behavior can be enabled since pandas 2.0 with the following option:
+
+.. code-block:: ipython
+
+   pd.options.mode.copy_on_write = True
+
+This change brings different changes in behavior in how pandas operates with respect to
+copies and views. Some of these changes allow a clear deprecation, like the changes in
+chained assignment. Other changes are more subtle and thus, the warnings are hidden behind
+an option that can be enabled in pandas 2.2.
+
+.. code-block:: ipython
+
+   pd.options.mode.copy_on_write = "warn"
+
+This mode will warn in many different scenarios that aren't actually relevant to
+most queries. We recommend exploring this mode, but it is not necessary to get rid
+of all of these warnings. The :ref:`migration guide <copy_on_write.migration_guide>`
+explains the upgrade process in more detail.
+
+Dedicated string data type (backed by Arrow) by default
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Historically, pandas represented string columns with NumPy object data type. This
+representation has numerous problems, including slow performance and a large memory
+footprint. This will change in pandas 3.0. pandas will start inferring string columns
+as a new ``string`` data type, backed by Arrow, which represents strings contiguous in memory. This brings
+a huge performance and memory improvement.
+
+Old behavior:
+
+.. code-block:: ipython
+
+    In [1]: ser = pd.Series(["a", "b"])
+    Out[1]:
+    0    a
+    1    b
+    dtype: object
+
+New behavior:
+
+
+.. code-block:: ipython
+
+    In [1]: ser = pd.Series(["a", "b"])
+    Out[1]:
+    0    a
+    1    b
+    dtype: string
+
+The string data type that is used in these scenarios will mostly behave as NumPy
+object would, including missing value semantics and general operations on these
+columns.
+
+This change includes a few additional changes across the API:
+
+- Currently, specifying ``dtype="string"`` creates a dtype that is backed by Python strings
+  which are stored in a NumPy array. This will change in pandas 3.0, this dtype
+  will create an Arrow backed string column.
+- The column names and the Index will also be backed by Arrow strings.
+- PyArrow will become a required dependency with pandas 3.0 to accommodate this change.
+
+This future dtype inference logic can be enabled with:
+
+.. code-block:: ipython
+
+   pd.options.future.infer_string = True
+
+.. _whatsnew_220.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_220.enhancements.adbc_support:
+
+ADBC Driver support in to_sql and read_sql
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+:func:`read_sql` and :meth:`~DataFrame.to_sql` now work with `Apache Arrow ADBC
+<https://arrow.apache.org/adbc/current/index.html>`_ drivers. Compared to
+traditional drivers used via SQLAlchemy, ADBC drivers should provide
+significant performance improvements, better type support and cleaner
+nullability handling.
+
+.. code-block:: ipython
+
+   import adbc_driver_postgresql.dbapi as pg_dbapi
+
+   df = pd.DataFrame(
+       [
+           [1, 2, 3],
+           [4, 5, 6],
+       ],
+       columns=['a', 'b', 'c']
+   )
+   uri = "postgresql://postgres:postgres@localhost/postgres"
+   with pg_dbapi.connect(uri) as conn:
+       df.to_sql("pandas_table", conn, index=False)
+
+   # for round-tripping
+   with pg_dbapi.connect(uri) as conn:
+       df2 = pd.read_sql("pandas_table", conn)
+
+The Arrow type system offers a wider array of types that can more closely match
+what databases like PostgreSQL can offer. To illustrate, note this (non-exhaustive)
+listing of types available in different databases and pandas backends:
+
++-----------------+-----------------------+----------------+---------+
+|numpy/pandas     |arrow                  |postgres        |sqlite   |
++=================+=======================+================+=========+
+|int16/Int16      |int16                  |SMALLINT        |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|int32/Int32      |int32                  |INTEGER         |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|int64/Int64      |int64                  |BIGINT          |INTEGER  |
++-----------------+-----------------------+----------------+---------+
+|float32          |float32                |REAL            |REAL     |
++-----------------+-----------------------+----------------+---------+
+|float64          |float64                |DOUBLE PRECISION|REAL     |
++-----------------+-----------------------+----------------+---------+
+|object           |string                 |TEXT            |TEXT     |
++-----------------+-----------------------+----------------+---------+
+|bool             |``bool_``              |BOOLEAN         |         |
++-----------------+-----------------------+----------------+---------+
+|datetime64[ns]   |timestamp(us)          |TIMESTAMP       |         |
++-----------------+-----------------------+----------------+---------+
+|datetime64[ns,tz]|timestamp(us,tz)       |TIMESTAMPTZ     |         |
++-----------------+-----------------------+----------------+---------+
+|                 |date32                 |DATE            |         |
++-----------------+-----------------------+----------------+---------+
+|                 |month_day_nano_interval|INTERVAL        |         |
++-----------------+-----------------------+----------------+---------+
+|                 |binary                 |BINARY          |BLOB     |
++-----------------+-----------------------+----------------+---------+
+|                 |decimal128             |DECIMAL [#f1]_  |         |
++-----------------+-----------------------+----------------+---------+
+|                 |list                   |ARRAY [#f1]_    |         |
++-----------------+-----------------------+----------------+---------+
+|                 |struct                 |COMPOSITE TYPE  |         |
+|                 |                       | [#f1]_         |         |
++-----------------+-----------------------+----------------+---------+
+
+.. rubric:: Footnotes
+
+.. [#f1] Not implemented as of writing, but theoretically possible
+
+If you are interested in preserving database types as best as possible
+throughout the lifecycle of your DataFrame, users are encouraged to
+leverage the ``dtype_backend="pyarrow"`` argument of :func:`~pandas.read_sql`
+
+.. code-block:: ipython
+
+   # for round-tripping
+   with pg_dbapi.connect(uri) as conn:
+       df2 = pd.read_sql("pandas_table", conn, dtype_backend="pyarrow")
+
+This will prevent your data from being converted to the traditional pandas/NumPy
+type system, which often converts SQL types in ways that make them impossible to
+round-trip.
+
+For a full list of ADBC drivers and their development status, see the `ADBC Driver
+Implementation Status <https://arrow.apache.org/adbc/current/driver/status.html>`_
+documentation.
+
+.. _whatsnew_220.enhancements.case_when:
+
+Create a pandas Series based on one or more conditions
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :meth:`Series.case_when` function has been added to create a Series object based on one or more conditions. (:issue:`39154`)
+
+.. ipython:: python
+
+   import pandas as pd
+
+   df = pd.DataFrame(dict(a=[1, 2, 3], b=[4, 5, 6]))
+   default=pd.Series('default', index=df.index)
+   default.case_when(
+        caselist=[
+            (df.a == 1, 'first'),                              # condition, replacement
+            (df.a.gt(1) & df.b.eq(5), 'second'),  # condition, replacement
+        ],
+   )
+
+.. _whatsnew_220.enhancements.to_numpy_ea:
+
+``to_numpy`` for NumPy nullable and Arrow types converts to suitable NumPy dtype
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``to_numpy`` for NumPy nullable and Arrow types will now convert to a
+suitable NumPy dtype instead of ``object`` dtype for nullable and PyArrow backed extension dtypes.
+
+*Old behavior:*
+
+.. code-block:: ipython
+
+    In [1]: ser = pd.Series([1, 2, 3], dtype="Int64")
+    In [2]: ser.to_numpy()
+    Out[2]: array([1, 2, 3], dtype=object)
+
+*New behavior:*
+
+.. ipython:: python
+
+    ser = pd.Series([1, 2, 3], dtype="Int64")
+    ser.to_numpy()
+
+    ser = pd.Series([1, 2, 3], dtype="timestamp[ns][pyarrow]")
+    ser.to_numpy()
+
+The default NumPy dtype (without any arguments) is determined as follows:
+
+- float dtypes are cast to NumPy floats
+- integer dtypes without missing values are cast to NumPy integer dtypes
+- integer dtypes with missing values are cast to NumPy float dtypes and ``NaN`` is used as missing value indicator
+- boolean dtypes without missing values are cast to NumPy bool dtype
+- boolean dtypes with missing values keep object dtype
+- datetime and timedelta types are cast to Numpy datetime64 and timedelta64 types respectively and ``NaT`` is used as missing value indicator
+
+.. _whatsnew_220.enhancements.struct_accessor:
+
+Series.struct accessor for PyArrow structured data
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``Series.struct`` accessor provides attributes and methods for processing
+data with ``struct[pyarrow]`` dtype Series. For example,
+:meth:`Series.struct.explode` converts PyArrow structured data to a pandas
+DataFrame. (:issue:`54938`)
+
+.. ipython:: python
+
+    import pyarrow as pa
+    series = pd.Series(
+        [
+            {"project": "pandas", "version": "2.2.0"},
+            {"project": "numpy", "version": "1.25.2"},
+            {"project": "pyarrow", "version": "13.0.0"},
+        ],
+        dtype=pd.ArrowDtype(
+            pa.struct([
+                ("project", pa.string()),
+                ("version", pa.string()),
+            ])
+        ),
+    )
+    series.struct.explode()
+
+Use :meth:`Series.struct.field` to index into a (possible nested)
+struct field.
+
+
+.. ipython:: python
+
+    series.struct.field("project")
+
+.. _whatsnew_220.enhancements.list_accessor:
+
+Series.list accessor for PyArrow list data
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``Series.list`` accessor provides attributes and methods for processing
+data with ``list[pyarrow]`` dtype Series. For example,
+:meth:`Series.list.__getitem__` allows indexing pyarrow lists in
+a Series. (:issue:`55323`)
+
+.. ipython:: python
+
+    import pyarrow as pa
+    series = pd.Series(
+        [
+            [1, 2, 3],
+            [4, 5],
+            [6],
+        ],
+        dtype=pd.ArrowDtype(
+            pa.list_(pa.int64())
+        ),
+    )
+    series.list[0]
+
+.. _whatsnew_220.enhancements.calamine:
+
+Calamine engine for :func:`read_excel`
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``calamine`` engine was added to :func:`read_excel`.
+It uses ``python-calamine``, which provides Python bindings for the Rust library `calamine <https://crates.io/crates/calamine>`__.
+This engine supports Excel files (``.xlsx``, ``.xlsm``, ``.xls``, ``.xlsb``) and OpenDocument spreadsheets (``.ods``) (:issue:`50395`).
+
+There are two advantages of this engine:
+
+1. Calamine is often faster than other engines, some benchmarks show results up to 5x faster than 'openpyxl', 20x - 'odf', 4x - 'pyxlsb', and 1.5x - 'xlrd'.
+   But, 'openpyxl' and 'pyxlsb' are faster in reading a few rows from large files because of lazy iteration over rows.
+2. Calamine supports the recognition of datetime in ``.xlsb`` files, unlike 'pyxlsb' which is the only other engine in pandas that can read ``.xlsb`` files.
+
+.. code-block:: python
+
+   pd.read_excel("path_to_file.xlsb", engine="calamine")
+
+
+For more, see :ref:`io.calamine` in the user guide on IO tools.
+
+.. _whatsnew_220.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- :meth:`~DataFrame.to_sql` with method parameter set to ``multi`` works with Oracle on the backend
+- :attr:`Series.attrs` / :attr:`DataFrame.attrs` now uses a deepcopy for propagating ``attrs`` (:issue:`54134`).
+- :func:`get_dummies` now returning  extension dtypes ``boolean`` or ``bool[pyarrow]`` that are compatible with the input dtype (:issue:`56273`)
+- :func:`read_csv` now supports ``on_bad_lines`` parameter with ``engine="pyarrow"`` (:issue:`54480`)
+- :func:`read_sas` returns ``datetime64`` dtypes with resolutions better matching those stored natively in SAS, and avoids returning object-dtype in cases that cannot be stored with ``datetime64[ns]`` dtype (:issue:`56127`)
+- :func:`read_spss` now returns a :class:`DataFrame` that stores the metadata in :attr:`DataFrame.attrs` (:issue:`54264`)
+- :func:`tseries.api.guess_datetime_format` is now part of the public API (:issue:`54727`)
+- :meth:`DataFrame.apply` now allows the usage of numba (via ``engine="numba"``) to JIT compile the passed function, allowing for potential speedups (:issue:`54666`)
+- :meth:`ExtensionArray._explode` interface method added to allow extension type implementations of the ``explode`` method (:issue:`54833`)
+- :meth:`ExtensionArray.duplicated` added to allow extension type implementations of the ``duplicated`` method (:issue:`55255`)
+- :meth:`Series.ffill`, :meth:`Series.bfill`, :meth:`DataFrame.ffill`, and :meth:`DataFrame.bfill` have gained the argument ``limit_area``; 3rd party :class:`.ExtensionArray` authors need to add this argument to the method ``_pad_or_backfill`` (:issue:`56492`)
+- Allow passing ``read_only``, ``data_only`` and ``keep_links`` arguments to openpyxl using ``engine_kwargs`` of :func:`read_excel` (:issue:`55027`)
+- Implement :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` for :class:`ArrowDtype` and masked dtypes (:issue:`56267`)
+- Implement masked algorithms for :meth:`Series.value_counts` (:issue:`54984`)
+- Implemented :meth:`Series.dt` methods and attributes for :class:`ArrowDtype` with ``pyarrow.duration`` type (:issue:`52284`)
+- Implemented :meth:`Series.str.extract` for :class:`ArrowDtype` (:issue:`56268`)
+- Improved error message that appears in :meth:`DatetimeIndex.to_period` with frequencies which are not supported as period frequencies, such as ``"BMS"`` (:issue:`56243`)
+- Improved error message when constructing :class:`Period` with invalid offsets such as ``"QS"`` (:issue:`55785`)
+- The dtypes ``string[pyarrow]`` and ``string[pyarrow_numpy]`` now both utilize the ``large_string`` type from PyArrow to avoid overflow for long columns (:issue:`56259`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_220.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_220.notable_bug_fixes.merge_sort_behavior:
+
+:func:`merge` and :meth:`DataFrame.join` now consistently follow documented sort behavior
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, :func:`merge` and :meth:`DataFrame.join` did not
+always return a result that followed the documented sort behavior. pandas now
+follows the documented sort behavior in merge and join operations (:issue:`54611`, :issue:`56426`, :issue:`56443`).
+
+As documented, ``sort=True`` sorts the join keys lexicographically in the resulting
+:class:`DataFrame`. With ``sort=False``, the order of the join keys depends on the
+join type (``how`` keyword):
+
+- ``how="left"``: preserve the order of the left keys
+- ``how="right"``: preserve the order of the right keys
+- ``how="inner"``: preserve the order of the left keys
+- ``how="outer"``: sort keys lexicographically
+
+One example with changing behavior is inner joins with non-unique left join keys
+and ``sort=False``:
+
+.. ipython:: python
+
+    left = pd.DataFrame({"a": [1, 2, 1]})
+    right = pd.DataFrame({"a": [1, 2]})
+    result = pd.merge(left, right, how="inner", on="a", sort=False)
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [5]: result
+    Out[5]:
+       a
+    0  1
+    1  1
+    2  2
+
+*New Behavior*
+
+.. ipython:: python
+
+    result
+
+.. _whatsnew_220.notable_bug_fixes.multiindex_join_different_levels:
+
+:func:`merge` and :meth:`DataFrame.join` no longer reorder levels when levels differ
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, :func:`merge` and :meth:`DataFrame.join` would reorder
+index levels when joining on two indexes with different levels (:issue:`34133`).
+
+.. ipython:: python
+
+    left = pd.DataFrame({"left": 1}, index=pd.MultiIndex.from_tuples([("x", 1), ("x", 2)], names=["A", "B"]))
+    right = pd.DataFrame({"right": 2}, index=pd.MultiIndex.from_tuples([(1, 1), (2, 2)], names=["B", "C"]))
+    left
+    right
+    result = left.join(right)
+
+*Old Behavior*
+
+.. code-block:: ipython
+
+    In [5]: result
+    Out[5]:
+           left  right
+    B A C
+    1 x 1     1      2
+    2 x 2     1      2
+
+*New Behavior*
+
+.. ipython:: python
+
+    result
+
+.. _whatsnew_220.api_breaking.deps:
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+For `optional dependencies <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+Optional dependencies below the lowest tested version may still work but are not considered supported.
+The following table lists the optional dependencies that have had their minimum tested version increased.
+
++-----------------+---------------------+
+| Package         | New Minimum Version |
++=================+=====================+
+| beautifulsoup4  | 4.11.2              |
++-----------------+---------------------+
+| blosc           | 1.21.3              |
++-----------------+---------------------+
+| bottleneck      | 1.3.6               |
++-----------------+---------------------+
+| fastparquet     | 2022.12.0           |
++-----------------+---------------------+
+| fsspec          | 2022.11.0           |
++-----------------+---------------------+
+| gcsfs           | 2022.11.0           |
++-----------------+---------------------+
+| lxml            | 4.9.2               |
++-----------------+---------------------+
+| matplotlib      | 3.6.3               |
++-----------------+---------------------+
+| numba           | 0.56.4              |
++-----------------+---------------------+
+| numexpr         | 2.8.4               |
++-----------------+---------------------+
+| qtpy            | 2.3.0               |
++-----------------+---------------------+
+| openpyxl        | 3.1.0               |
++-----------------+---------------------+
+| psycopg2        | 2.9.6               |
++-----------------+---------------------+
+| pyreadstat      | 1.2.0               |
++-----------------+---------------------+
+| pytables        | 3.8.0               |
++-----------------+---------------------+
+| pyxlsb          | 1.0.10              |
++-----------------+---------------------+
+| s3fs            | 2022.11.0           |
++-----------------+---------------------+
+| scipy           | 1.10.0              |
++-----------------+---------------------+
+| sqlalchemy      | 2.0.0               |
++-----------------+---------------------+
+| tabulate        | 0.9.0               |
++-----------------+---------------------+
+| xarray          | 2022.12.0           |
++-----------------+---------------------+
+| xlsxwriter      | 3.0.5               |
++-----------------+---------------------+
+| zstandard       | 0.19.0              |
++-----------------+---------------------+
+| pyqt5           | 5.15.8              |
++-----------------+---------------------+
+| tzdata          | 2022.7              |
++-----------------+---------------------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_220.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- The hash values of nullable extension dtypes changed to improve the performance of the hashing operation (:issue:`56507`)
+- ``check_exact`` now only takes effect for floating-point dtypes in :func:`testing.assert_frame_equal` and :func:`testing.assert_series_equal`. In particular, integer dtypes are always checked exactly (:issue:`55882`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_220.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+Chained assignment
+^^^^^^^^^^^^^^^^^^
+
+In preparation of larger upcoming changes to the copy / view behaviour in pandas 3.0
+(:ref:`copy_on_write`, PDEP-7), we started deprecating *chained assignment*.
+
+Chained assignment occurs when you try to update a pandas DataFrame or Series through
+two subsequent indexing operations. Depending on the type and order of those operations
+this currently does or does not work.
+
+A typical example is as follows:
+
+.. code-block:: python
+
+    df = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
+
+    # first selecting rows with a mask, then assigning values to a column
+    # -> this has never worked and raises a SettingWithCopyWarning
+    df[df["bar"] > 5]["foo"] = 100
+
+    # first selecting the column, and then assigning to a subset of that column
+    # -> this currently works
+    df["foo"][df["bar"] > 5] = 100
+
+This second example of chained assignment currently works to update the original ``df``.
+This will no longer work in pandas 3.0, and therefore we started deprecating this:
+
+.. code-block:: python
+
+    >>> df["foo"][df["bar"] > 5] = 100
+    FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!
+    You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.
+    A typical example is when you are setting values in a column of a DataFrame, like:
+
+    df["col"][row_indexer] = value
+
+    Use `df.loc[row_indexer, "col"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.
+
+    See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
+
+You can fix this warning and ensure your code is ready for pandas 3.0 by removing
+the usage of chained assignment. Typically, this can be done by doing the assignment
+in a single step using for example ``.loc``. For the example above, we can do:
+
+.. code-block:: python
+
+    df.loc[df["bar"] > 5, "foo"] = 100
+
+The same deprecation applies to inplace methods that are done in a chained manner, such as:
+
+.. code-block:: python
+
+    >>> df["foo"].fillna(0, inplace=True)
+    FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
+    The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.
+
+    For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.
+
+When the goal is to update the column in the DataFrame ``df``, the alternative here is
+to call the method on ``df`` itself, such as ``df.fillna({"foo": 0}, inplace=True)``.
+
+See more details in the :ref:`migration guide <copy_on_write.migration_guide>`.
+
+
+Deprecate aliases ``M``, ``Q``, ``Y``, etc. in favour of ``ME``, ``QE``, ``YE``, etc. for offsets
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Deprecated the following frequency aliases (:issue:`9586`):
+
++-------------------------------+------------------+------------------+
+|offsets                        |deprecated aliases|new aliases       |
++===============================+==================+==================+
+|:class:`MonthEnd`              |      ``M``       |     ``ME``       |
++-------------------------------+------------------+------------------+
+|:class:`BusinessMonthEnd`      |      ``BM``      |     ``BME``      |
++-------------------------------+------------------+------------------+
+|:class:`SemiMonthEnd`          |      ``SM``      |     ``SME``      |
++-------------------------------+------------------+------------------+
+|:class:`CustomBusinessMonthEnd`|      ``CBM``     |     ``CBME``     |
++-------------------------------+------------------+------------------+
+|:class:`QuarterEnd`            |      ``Q``       |     ``QE``       |
++-------------------------------+------------------+------------------+
+|:class:`BQuarterEnd`           |      ``BQ``      |     ``BQE``      |
++-------------------------------+------------------+------------------+
+|:class:`YearEnd`               |      ``Y``       |     ``YE``       |
++-------------------------------+------------------+------------------+
+|:class:`BYearEnd`              |      ``BY``      |     ``BYE``      |
++-------------------------------+------------------+------------------+
+
+For example:
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [8]: pd.date_range('2020-01-01', periods=3, freq='Q-NOV')
+    Out[8]:
+    DatetimeIndex(['2020-02-29', '2020-05-31', '2020-08-31'],
+                  dtype='datetime64[ns]', freq='Q-NOV')
+
+*Future behavior*:
+
+.. ipython:: python
+
+    pd.date_range('2020-01-01', periods=3, freq='QE-NOV')
+
+.. _whatsnew_220.silent_downcasting:
+
+Deprecated automatic downcasting
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Deprecated the automatic downcasting of object dtype results in a number of
+methods. These would silently change the dtype in a hard to predict manner since the
+behavior was value dependent. Additionally, pandas is moving away from silent dtype
+changes (:issue:`54710`, :issue:`54261`).
+
+These methods are:
+
+- :meth:`Series.replace` and :meth:`DataFrame.replace`
+- :meth:`DataFrame.fillna`, :meth:`Series.fillna`
+- :meth:`DataFrame.ffill`, :meth:`Series.ffill`
+- :meth:`DataFrame.bfill`, :meth:`Series.bfill`
+- :meth:`DataFrame.mask`, :meth:`Series.mask`
+- :meth:`DataFrame.where`, :meth:`Series.where`
+- :meth:`DataFrame.clip`, :meth:`Series.clip`
+
+Explicitly call :meth:`DataFrame.infer_objects` to replicate the current behavior in the future.
+
+.. code-block:: ipython
+
+    result = result.infer_objects(copy=False)
+
+Or explicitly cast all-round floats to ints using ``astype``.
+
+Set the following option to opt into the future behavior:
+
+.. code-block:: ipython
+
+    In [9]: pd.set_option("future.no_silent_downcasting", True)
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+- Changed :meth:`Timedelta.resolution_string` to return ``h``, ``min``, ``s``, ``ms``, ``us``, and ``ns`` instead of ``H``, ``T``, ``S``, ``L``, ``U``, and ``N``, for compatibility with respective deprecations in frequency aliases (:issue:`52536`)
+- Deprecated :attr:`offsets.Day.delta`, :attr:`offsets.Hour.delta`, :attr:`offsets.Minute.delta`, :attr:`offsets.Second.delta`, :attr:`offsets.Milli.delta`, :attr:`offsets.Micro.delta`, :attr:`offsets.Nano.delta`, use ``pd.Timedelta(obj)`` instead (:issue:`55498`)
+- Deprecated :func:`pandas.api.types.is_interval` and :func:`pandas.api.types.is_period`, use ``isinstance(obj, pd.Interval)`` and ``isinstance(obj, pd.Period)`` instead (:issue:`55264`)
+- Deprecated :func:`read_gbq` and :meth:`DataFrame.to_gbq`. Use ``pandas_gbq.read_gbq`` and ``pandas_gbq.to_gbq`` instead https://pandas-gbq.readthedocs.io/en/latest/api.html (:issue:`55525`)
+- Deprecated :meth:`.DataFrameGroupBy.fillna` and :meth:`.SeriesGroupBy.fillna`; use :meth:`.DataFrameGroupBy.ffill`, :meth:`.DataFrameGroupBy.bfill` for forward and backward filling or :meth:`.DataFrame.fillna` to fill with a single value (or the Series equivalents) (:issue:`55718`)
+- Deprecated :meth:`DateOffset.is_anchored`, use ``obj.n == 1`` for non-Tick subclasses (for Tick this was always False) (:issue:`55388`)
+- Deprecated :meth:`DatetimeArray.__init__` and :meth:`TimedeltaArray.__init__`, use :func:`array` instead (:issue:`55623`)
+- Deprecated :meth:`Index.format`, use ``index.astype(str)`` or ``index.map(formatter)`` instead (:issue:`55413`)
+- Deprecated :meth:`Series.ravel`, the underlying array is already 1D, so ravel is not necessary (:issue:`52511`)
+- Deprecated :meth:`Series.resample` and :meth:`DataFrame.resample` with a :class:`PeriodIndex` (and the 'convention' keyword), convert to :class:`DatetimeIndex` (with ``.to_timestamp()``) before resampling instead (:issue:`53481`)
+- Deprecated :meth:`Series.view`, use :meth:`Series.astype` instead to change the dtype (:issue:`20251`)
+- Deprecated :meth:`offsets.Tick.is_anchored`, use ``False`` instead (:issue:`55388`)
+- Deprecated ``core.internals`` members ``Block``, ``ExtensionBlock``, and ``DatetimeTZBlock``, use public APIs instead (:issue:`55139`)
+- Deprecated ``year``, ``month``, ``quarter``, ``day``, ``hour``, ``minute``, and ``second`` keywords in the :class:`PeriodIndex` constructor, use :meth:`PeriodIndex.from_fields` instead (:issue:`55960`)
+- Deprecated accepting a type as an argument in :meth:`Index.view`, call without any arguments instead (:issue:`55709`)
+- Deprecated allowing non-integer ``periods`` argument in :func:`date_range`, :func:`timedelta_range`, :func:`period_range`, and :func:`interval_range` (:issue:`56036`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_clipboard` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_csv` except ``path_or_buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_dict` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_excel` except ``excel_writer`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_gbq` except ``destination_table`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_hdf` except ``path_or_buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_html` except ``buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_json` except ``path_or_buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_latex` except ``buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_markdown` except ``buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_parquet` except ``path`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_pickle` except ``path`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_string` except ``buf`` (:issue:`54229`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.to_xml` except ``path_or_buffer`` (:issue:`54229`)
+- Deprecated allowing passing :class:`BlockManager` objects to :class:`DataFrame` or :class:`SingleBlockManager` objects to :class:`Series` (:issue:`52419`)
+- Deprecated behavior of :meth:`Index.insert` with an object-dtype index silently performing type inference on the result, explicitly call ``result.infer_objects(copy=False)`` for the old behavior instead (:issue:`51363`)
+- Deprecated casting non-datetimelike values (mainly strings) in :meth:`Series.isin` and :meth:`Index.isin` with ``datetime64``, ``timedelta64``, and :class:`PeriodDtype` dtypes (:issue:`53111`)
+- Deprecated dtype inference in :class:`Index`, :class:`Series` and :class:`DataFrame` constructors when giving a pandas input, call ``.infer_objects`` on the input to keep the current behavior (:issue:`56012`)
+- Deprecated dtype inference when setting a :class:`Index` into a :class:`DataFrame`, cast explicitly instead (:issue:`56102`)
+- Deprecated including the groups in computations when using :meth:`.DataFrameGroupBy.apply` and :meth:`.DataFrameGroupBy.resample`; pass ``include_groups=False`` to exclude the groups (:issue:`7155`)
+- Deprecated indexing an :class:`Index`  with a boolean indexer of length zero (:issue:`55820`)
+- Deprecated not passing a tuple to :class:`.DataFrameGroupBy.get_group` or :class:`.SeriesGroupBy.get_group` when grouping by a length-1 list-like (:issue:`25971`)
+- Deprecated string ``AS`` denoting frequency in :class:`YearBegin` and strings ``AS-DEC``, ``AS-JAN``, etc. denoting annual frequencies with various fiscal year starts (:issue:`54275`)
+- Deprecated string ``A`` denoting frequency in :class:`YearEnd` and strings ``A-DEC``, ``A-JAN``, etc. denoting annual frequencies with various fiscal year ends (:issue:`54275`)
+- Deprecated string ``BAS`` denoting frequency in :class:`BYearBegin` and strings ``BAS-DEC``, ``BAS-JAN``, etc. denoting annual frequencies with various fiscal year starts (:issue:`54275`)
+- Deprecated string ``BA`` denoting frequency in :class:`BYearEnd` and strings ``BA-DEC``, ``BA-JAN``, etc. denoting annual frequencies with various fiscal year ends (:issue:`54275`)
+- Deprecated strings ``H``, ``BH``, and ``CBH`` denoting frequencies in :class:`Hour`, :class:`BusinessHour`, :class:`CustomBusinessHour` (:issue:`52536`)
+- Deprecated strings ``H``, ``S``, ``U``, and ``N`` denoting units in :func:`to_timedelta` (:issue:`52536`)
+- Deprecated strings ``H``, ``T``, ``S``, ``L``, ``U``, and ``N`` denoting units in :class:`Timedelta` (:issue:`52536`)
+- Deprecated strings ``T``, ``S``, ``L``, ``U``, and ``N`` denoting frequencies in :class:`Minute`, :class:`Second`, :class:`Milli`, :class:`Micro`, :class:`Nano` (:issue:`52536`)
+- Deprecated support for combining parsed datetime columns in :func:`read_csv` along with the ``keep_date_col`` keyword (:issue:`55569`)
+- Deprecated the :attr:`.DataFrameGroupBy.grouper` and :attr:`SeriesGroupBy.grouper`; these attributes will be removed in a future version of pandas (:issue:`56521`)
+- Deprecated the :class:`.Grouping` attributes ``group_index``, ``result_index``, and ``group_arraylike``; these will be removed in a future version of pandas (:issue:`56148`)
+- Deprecated the ``delim_whitespace`` keyword in :func:`read_csv` and :func:`read_table`, use ``sep="\\s+"`` instead (:issue:`55569`)
+- Deprecated the ``errors="ignore"`` option in :func:`to_datetime`, :func:`to_timedelta`, and :func:`to_numeric`; explicitly catch exceptions instead (:issue:`54467`)
+- Deprecated the ``fastpath`` keyword in the :class:`Series` constructor (:issue:`20110`)
+- Deprecated the ``kind`` keyword in :meth:`Series.resample` and :meth:`DataFrame.resample`, explicitly cast the object's ``index`` instead (:issue:`55895`)
+- Deprecated the ``ordinal`` keyword in :class:`PeriodIndex`, use :meth:`PeriodIndex.from_ordinals` instead (:issue:`55960`)
+- Deprecated the ``unit`` keyword in :class:`TimedeltaIndex` construction, use :func:`to_timedelta` instead (:issue:`55499`)
+- Deprecated the ``verbose`` keyword in :func:`read_csv` and :func:`read_table` (:issue:`55569`)
+- Deprecated the behavior of :meth:`DataFrame.replace` and :meth:`Series.replace` with :class:`CategoricalDtype`; in a future version replace will change the values while preserving the categories. To change the categories, use ``ser.cat.rename_categories`` instead (:issue:`55147`)
+- Deprecated the behavior of :meth:`Series.value_counts` and :meth:`Index.value_counts` with object dtype; in a future version these will not perform dtype inference on the resulting :class:`Index`, do ``result.index = result.index.infer_objects()`` to retain the old behavior (:issue:`56161`)
+- Deprecated the default of ``observed=False`` in :meth:`DataFrame.pivot_table`; will be ``True`` in a future version (:issue:`56236`)
+- Deprecated the extension test classes ``BaseNoReduceTests``, ``BaseBooleanReduceTests``, and ``BaseNumericReduceTests``, use ``BaseReduceTests`` instead (:issue:`54663`)
+- Deprecated the option ``mode.data_manager`` and the ``ArrayManager``; only the ``BlockManager`` will be available in future versions (:issue:`55043`)
+- Deprecated the previous implementation of :class:`DataFrame.stack`; specify ``future_stack=True`` to adopt the future version (:issue:`53515`)
+-
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_220.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Performance improvement in :func:`.testing.assert_frame_equal` and :func:`.testing.assert_series_equal` (:issue:`55949`, :issue:`55971`)
+- Performance improvement in :func:`concat` with ``axis=1`` and objects with unaligned indexes (:issue:`55084`)
+- Performance improvement in :func:`get_dummies` (:issue:`56089`)
+- Performance improvement in :func:`merge` and :func:`merge_ordered` when joining on sorted ascending keys (:issue:`56115`)
+- Performance improvement in :func:`merge_asof` when ``by`` is not ``None`` (:issue:`55580`, :issue:`55678`)
+- Performance improvement in :func:`read_stata` for files with many variables (:issue:`55515`)
+- Performance improvement in :meth:`DataFrame.groupby` when aggregating pyarrow timestamp and duration dtypes (:issue:`55031`)
+- Performance improvement in :meth:`DataFrame.join` when joining on unordered categorical indexes (:issue:`56345`)
+- Performance improvement in :meth:`DataFrame.loc` and :meth:`Series.loc` when indexing with a :class:`MultiIndex` (:issue:`56062`)
+- Performance improvement in :meth:`DataFrame.sort_index` and :meth:`Series.sort_index` when indexed by a :class:`MultiIndex` (:issue:`54835`)
+- Performance improvement in :meth:`DataFrame.to_dict` on converting DataFrame to dictionary (:issue:`50990`)
+- Performance improvement in :meth:`Index.difference` (:issue:`55108`)
+- Performance improvement in :meth:`Index.sort_values` when index is already sorted (:issue:`56128`)
+- Performance improvement in :meth:`MultiIndex.get_indexer` when ``method`` is not ``None`` (:issue:`55839`)
+- Performance improvement in :meth:`Series.duplicated` for pyarrow dtypes (:issue:`55255`)
+- Performance improvement in :meth:`Series.str.get_dummies` when dtype is ``"string[pyarrow]"`` or ``"string[pyarrow_numpy]"`` (:issue:`56110`)
+- Performance improvement in :meth:`Series.str` methods (:issue:`55736`)
+- Performance improvement in :meth:`Series.value_counts` and :meth:`Series.mode` for masked dtypes (:issue:`54984`, :issue:`55340`)
+- Performance improvement in :meth:`.DataFrameGroupBy.nunique` and :meth:`.SeriesGroupBy.nunique` (:issue:`55972`)
+- Performance improvement in :meth:`.SeriesGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.DataFrameGroupBy.idxmin` (:issue:`54234`)
+- Performance improvement when hashing a nullable extension array (:issue:`56507`)
+- Performance improvement when indexing into a non-unique index (:issue:`55816`)
+- Performance improvement when indexing with more than 4 keys (:issue:`54550`)
+- Performance improvement when localizing time to UTC (:issue:`55241`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_220.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- :meth:`Categorical.isin` raising ``InvalidIndexError`` for categorical containing overlapping :class:`Interval` values (:issue:`34974`)
+- Bug in :meth:`CategoricalDtype.__eq__` returning ``False`` for unordered categorical data with mixed types (:issue:`55468`)
+- Bug when casting ``pa.dictionary`` to :class:`CategoricalDtype` using a ``pa.DictionaryArray`` as categories (:issue:`56672`)
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :class:`DatetimeIndex` construction when passing both a ``tz`` and either ``dayfirst`` or ``yearfirst`` ignoring dayfirst/yearfirst (:issue:`55813`)
+- Bug in :class:`DatetimeIndex` when passing an object-dtype ndarray of float objects and a ``tz`` incorrectly localizing the result (:issue:`55780`)
+- Bug in :func:`Series.isin` with :class:`DatetimeTZDtype` dtype and comparison values that are all ``NaT`` incorrectly returning all-``False`` even if the series contains ``NaT`` entries (:issue:`56427`)
+- Bug in :func:`concat` raising ``AttributeError`` when concatenating all-NA DataFrame with :class:`DatetimeTZDtype` dtype DataFrame (:issue:`52093`)
+- Bug in :func:`testing.assert_extension_array_equal` that could use the wrong unit when comparing resolutions (:issue:`55730`)
+- Bug in :func:`to_datetime` and :class:`DatetimeIndex` when passing a list of mixed-string-and-numeric types incorrectly raising (:issue:`55780`)
+- Bug in :func:`to_datetime` and :class:`DatetimeIndex` when passing mixed-type objects with a mix of timezones or mix of timezone-awareness failing to raise ``ValueError`` (:issue:`55693`)
+- Bug in :meth:`.Tick.delta` with very large ticks raising ``OverflowError`` instead of ``OutOfBoundsTimedelta`` (:issue:`55503`)
+- Bug in :meth:`DatetimeIndex.shift` with non-nanosecond resolution incorrectly returning with nanosecond resolution (:issue:`56117`)
+- Bug in :meth:`DatetimeIndex.union` returning object dtype for tz-aware indexes with the same timezone but different units (:issue:`55238`)
+- Bug in :meth:`Index.is_monotonic_increasing` and :meth:`Index.is_monotonic_decreasing` always caching :meth:`Index.is_unique` as ``True`` when first value in index is ``NaT`` (:issue:`55755`)
+- Bug in :meth:`Index.view` to a datetime64 dtype with non-supported resolution incorrectly raising (:issue:`55710`)
+- Bug in :meth:`Series.dt.round` with non-nanosecond resolution and ``NaT`` entries incorrectly raising ``OverflowError`` (:issue:`56158`)
+- Bug in :meth:`Series.fillna` with non-nanosecond resolution dtypes and higher-resolution vector values returning incorrect (internally-corrupted) results (:issue:`56410`)
+- Bug in :meth:`Timestamp.unit` being inferred incorrectly from an ISO8601 format string with minute or hour resolution and a timezone offset (:issue:`56208`)
+- Bug in ``.astype`` converting from a higher-resolution ``datetime64`` dtype to a lower-resolution ``datetime64`` dtype (e.g. ``datetime64[us]->datetime64[ms]``) silently overflowing with values near the lower implementation bound (:issue:`55979`)
+- Bug in adding or subtracting a :class:`Week` offset to a ``datetime64`` :class:`Series`, :class:`Index`, or :class:`DataFrame` column with non-nanosecond resolution returning incorrect results (:issue:`55583`)
+- Bug in addition or subtraction of :class:`BusinessDay` offset with ``offset`` attribute to non-nanosecond :class:`Index`, :class:`Series`, or :class:`DataFrame` column giving incorrect results (:issue:`55608`)
+- Bug in addition or subtraction of :class:`DateOffset` objects with microsecond components to ``datetime64`` :class:`Index`, :class:`Series`, or :class:`DataFrame` columns with non-nanosecond resolution (:issue:`55595`)
+- Bug in addition or subtraction of very large :class:`.Tick` objects with :class:`Timestamp` or :class:`Timedelta` objects raising ``OverflowError`` instead of ``OutOfBoundsTimedelta`` (:issue:`55503`)
+- Bug in creating a :class:`Index`, :class:`Series`, or :class:`DataFrame` with a non-nanosecond :class:`DatetimeTZDtype` and inputs that would be out of bounds with nanosecond resolution incorrectly raising ``OutOfBoundsDatetime`` (:issue:`54620`)
+- Bug in creating a :class:`Index`, :class:`Series`, or :class:`DataFrame` with a non-nanosecond ``datetime64`` (or :class:`DatetimeTZDtype`) from mixed-numeric inputs treating those as nanoseconds instead of as multiples of the dtype's unit (which would happen with non-mixed numeric inputs) (:issue:`56004`)
+- Bug in creating a :class:`Index`, :class:`Series`, or :class:`DataFrame` with a non-nanosecond ``datetime64`` dtype and inputs that would be out of bounds for a ``datetime64[ns]`` incorrectly raising ``OutOfBoundsDatetime`` (:issue:`55756`)
+- Bug in parsing datetime strings with nanosecond resolution with non-ISO8601 formats incorrectly truncating sub-microsecond components (:issue:`56051`)
+- Bug in parsing datetime strings with sub-second resolution and trailing zeros incorrectly inferring second or millisecond resolution (:issue:`55737`)
+- Bug in the results of :func:`to_datetime` with an floating-dtype argument with ``unit`` not matching the pointwise results of :class:`Timestamp` (:issue:`56037`)
+- Fixed regression where :func:`concat` would raise an error when concatenating ``datetime64`` columns with differing resolutions (:issue:`53641`)
+
+Timedelta
+^^^^^^^^^
+- Bug in :class:`Timedelta` construction raising ``OverflowError`` instead of ``OutOfBoundsTimedelta`` (:issue:`55503`)
+- Bug in rendering (``__repr__``) of :class:`TimedeltaIndex` and :class:`Series` with timedelta64 values with non-nanosecond resolution entries that are all multiples of 24 hours failing to use the compact representation used in the nanosecond cases (:issue:`55405`)
+
+Timezones
+^^^^^^^^^
+- Bug in :class:`AbstractHolidayCalendar` where timezone data was not propagated when computing holiday observances (:issue:`54580`)
+- Bug in :class:`Timestamp` construction with an ambiguous value and a ``pytz`` timezone failing to raise ``pytz.AmbiguousTimeError`` (:issue:`55657`)
+- Bug in :meth:`Timestamp.tz_localize` with ``nonexistent="shift_forward`` around UTC+0 during DST (:issue:`51501`)
+
+Numeric
+^^^^^^^
+- Bug in :func:`read_csv` with ``engine="pyarrow"`` causing rounding errors for large integers (:issue:`52505`)
+- Bug in :meth:`Series.__floordiv__` and :meth:`Series.__truediv__` for :class:`ArrowDtype` with integral dtypes raising for large divisors (:issue:`56706`)
+- Bug in :meth:`Series.__floordiv__` for :class:`ArrowDtype` with integral dtypes raising for large values (:issue:`56645`)
+- Bug in :meth:`Series.pow` not filling missing values correctly (:issue:`55512`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` matching float ``0.0`` with ``False`` and vice versa (:issue:`55398`)
+- Bug in :meth:`Series.round` raising for nullable boolean dtype (:issue:`55936`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :meth:`DataFrame.astype` when called with ``str`` on unpickled array - the array might change in-place (:issue:`54654`)
+- Bug in :meth:`DataFrame.astype` where ``errors="ignore"`` had no effect for extension types (:issue:`54654`)
+- Bug in :meth:`Series.convert_dtypes` not converting all NA column to ``null[pyarrow]`` (:issue:`55346`)
+- Bug in :meth:``DataFrame.loc`` was not throwing "incompatible dtype warning" (see `PDEP6 <https://pandas.pydata.org/pdeps/0006-ban-upcasting.html>`_) when assigning a ``Series`` with a different dtype using a full column setter (e.g. ``df.loc[:, 'a'] = incompatible_value``) (:issue:`39584`)
+
+Strings
+^^^^^^^
+- Bug in :func:`pandas.api.types.is_string_dtype` while checking object array with no elements is of the string dtype (:issue:`54661`)
+- Bug in :meth:`DataFrame.apply` failing when ``engine="numba"`` and columns or index have ``StringDtype`` (:issue:`56189`)
+- Bug in :meth:`DataFrame.reindex` not matching :class:`Index` with ``string[pyarrow_numpy]`` dtype (:issue:`56106`)
+- Bug in :meth:`Index.str.cat` always casting result to object dtype (:issue:`56157`)
+- Bug in :meth:`Series.__mul__` for :class:`ArrowDtype` with ``pyarrow.string`` dtype and ``string[pyarrow]`` for the pyarrow backend (:issue:`51970`)
+- Bug in :meth:`Series.str.find` when ``start < 0`` for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`56411`)
+- Bug in :meth:`Series.str.fullmatch` when ``dtype=pandas.ArrowDtype(pyarrow.string()))`` allows partial matches when regex ends in literal //$ (:issue:`56652`)
+- Bug in :meth:`Series.str.replace` when ``n < 0`` for :class:`ArrowDtype` with ``pyarrow.string`` (:issue:`56404`)
+- Bug in :meth:`Series.str.startswith` and :meth:`Series.str.endswith` with arguments of type ``tuple[str, ...]`` for :class:`ArrowDtype` with ``pyarrow.string`` dtype (:issue:`56579`)
+- Bug in :meth:`Series.str.startswith` and :meth:`Series.str.endswith` with arguments of type ``tuple[str, ...]`` for ``string[pyarrow]`` (:issue:`54942`)
+- Bug in comparison operations for ``dtype="string[pyarrow_numpy]"`` raising if dtypes can't be compared (:issue:`56008`)
+
+Interval
+^^^^^^^^
+- Bug in :class:`Interval` ``__repr__`` not displaying UTC offsets for :class:`Timestamp` bounds. Additionally the hour, minute and second components will now be shown (:issue:`55015`)
+- Bug in :meth:`IntervalIndex.factorize` and :meth:`Series.factorize` with :class:`IntervalDtype` with datetime64 or timedelta64 intervals not preserving non-nanosecond units (:issue:`56099`)
+- Bug in :meth:`IntervalIndex.from_arrays` when passed ``datetime64`` or ``timedelta64`` arrays with mismatched resolutions constructing an invalid ``IntervalArray`` object (:issue:`55714`)
+- Bug in :meth:`IntervalIndex.from_tuples` raising if subtype is a nullable extension dtype (:issue:`56765`)
+- Bug in :meth:`IntervalIndex.get_indexer` with datetime or timedelta intervals incorrectly matching on integer targets (:issue:`47772`)
+- Bug in :meth:`IntervalIndex.get_indexer` with timezone-aware datetime intervals incorrectly matching on a sequence of timezone-naive targets (:issue:`47772`)
+- Bug in setting values on a :class:`Series` with an :class:`IntervalIndex` using a slice incorrectly raising (:issue:`54722`)
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`DataFrame.loc` mutating a boolean indexer when :class:`DataFrame` has a :class:`MultiIndex` (:issue:`56635`)
+- Bug in :meth:`DataFrame.loc` when setting :class:`Series` with extension dtype into NumPy dtype (:issue:`55604`)
+- Bug in :meth:`Index.difference` not returning a unique set of values when ``other`` is empty or ``other`` is considered non-comparable (:issue:`55113`)
+- Bug in setting :class:`Categorical` values into a :class:`DataFrame` with numpy dtypes raising ``RecursionError`` (:issue:`52927`)
+- Fixed bug when creating new column with missing values when setting a single string value (:issue:`56204`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`DataFrame.update` wasn't updating in-place for tz-aware datetime64 dtypes (:issue:`56227`)
+
+MultiIndex
+^^^^^^^^^^
+- Bug in :meth:`MultiIndex.get_indexer` not raising ``ValueError`` when ``method`` provided and index is non-monotonic (:issue:`53452`)
+
+I/O
+^^^
+- Bug in :func:`read_csv` where ``engine="python"`` did not respect ``chunksize`` arg when ``skiprows`` was specified (:issue:`56323`)
+- Bug in :func:`read_csv` where ``engine="python"`` was causing a ``TypeError`` when a callable ``skiprows`` and a chunk size was specified (:issue:`55677`)
+- Bug in :func:`read_csv` where ``on_bad_lines="warn"`` would write to ``stderr`` instead of raising a Python warning; this now yields a :class:`.errors.ParserWarning` (:issue:`54296`)
+- Bug in :func:`read_csv` with ``engine="pyarrow"`` where ``quotechar`` was ignored (:issue:`52266`)
+- Bug in :func:`read_csv` with ``engine="pyarrow"`` where ``usecols`` wasn't working with a CSV with no headers (:issue:`54459`)
+- Bug in :func:`read_excel`, with ``engine="xlrd"`` (``xls`` files) erroring when the file contains ``NaN`` or ``Inf`` (:issue:`54564`)
+- Bug in :func:`read_json` not handling dtype conversion properly if ``infer_string`` is set (:issue:`56195`)
+- Bug in :meth:`DataFrame.to_excel`, with ``OdsWriter`` (``ods`` files) writing Boolean/string value (:issue:`54994`)
+- Bug in :meth:`DataFrame.to_hdf` and :func:`read_hdf` with ``datetime64`` dtypes with non-nanosecond resolution failing to round-trip correctly (:issue:`55622`)
+- Bug in :meth:`DataFrame.to_stata` raising for extension dtypes (:issue:`54671`)
+- Bug in :meth:`~pandas.read_excel` with ``engine="odf"`` (``ods`` files) when a string cell contains an annotation (:issue:`55200`)
+- Bug in :meth:`~pandas.read_excel` with an ODS file without cached formatted cell for float values (:issue:`55219`)
+- Bug where :meth:`DataFrame.to_json` would raise an ``OverflowError`` instead of a ``TypeError`` with unsupported NumPy types (:issue:`55403`)
+
+Period
+^^^^^^
+- Bug in :class:`PeriodIndex` construction when more than one of ``data``, ``ordinal`` and ``**fields`` are passed failing to raise ``ValueError`` (:issue:`55961`)
+- Bug in :class:`Period` addition silently wrapping around instead of raising ``OverflowError`` (:issue:`55503`)
+- Bug in casting from :class:`PeriodDtype` with ``astype`` to ``datetime64`` or :class:`DatetimeTZDtype` with non-nanosecond unit incorrectly returning with nanosecond unit (:issue:`55958`)
+
+Plotting
+^^^^^^^^
+- Bug in :meth:`DataFrame.plot.box` with ``vert=False`` and a Matplotlib ``Axes`` created with ``sharey=True`` (:issue:`54941`)
+- Bug in :meth:`DataFrame.plot.scatter` discarding string columns (:issue:`56142`)
+- Bug in :meth:`Series.plot` when reusing an ``ax`` object failing to raise when a ``how`` keyword is passed (:issue:`55953`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, and :meth:`.SeriesGroupBy.idxmax` would not retain :class:`.Categorical` dtype when the index was a :class:`.CategoricalIndex` that contained NA values (:issue:`54234`)
+- Bug in :meth:`.DataFrameGroupBy.transform` and :meth:`.SeriesGroupBy.transform` when ``observed=False`` and ``f="idxmin"`` or ``f="idxmax"`` would incorrectly raise on unobserved categories (:issue:`54234`)
+- Bug in :meth:`.DataFrameGroupBy.value_counts` and :meth:`.SeriesGroupBy.value_counts` could result in incorrect sorting if the columns of the DataFrame or name of the Series are integers (:issue:`55951`)
+- Bug in :meth:`.DataFrameGroupBy.value_counts` and :meth:`.SeriesGroupBy.value_counts` would not respect ``sort=False`` in :meth:`DataFrame.groupby` and :meth:`Series.groupby` (:issue:`55951`)
+- Bug in :meth:`.DataFrameGroupBy.value_counts` and :meth:`.SeriesGroupBy.value_counts` would sort by proportions rather than frequencies when ``sort=True`` and ``normalize=True`` (:issue:`55951`)
+- Bug in :meth:`DataFrame.asfreq` and :meth:`Series.asfreq` with a :class:`DatetimeIndex` with non-nanosecond resolution incorrectly converting to nanosecond resolution (:issue:`55958`)
+- Bug in :meth:`DataFrame.ewm` when passed ``times`` with non-nanosecond ``datetime64`` or :class:`DatetimeTZDtype` dtype (:issue:`56262`)
+- Bug in :meth:`DataFrame.groupby` and :meth:`Series.groupby` where grouping by a combination of ``Decimal`` and NA values would fail when ``sort=True`` (:issue:`54847`)
+- Bug in :meth:`DataFrame.groupby` for DataFrame subclasses when selecting a subset of columns to apply the function to (:issue:`56761`)
+- Bug in :meth:`DataFrame.resample` not respecting ``closed`` and ``label`` arguments for :class:`~pandas.tseries.offsets.BusinessDay` (:issue:`55282`)
+- Bug in :meth:`DataFrame.resample` when resampling on a :class:`ArrowDtype` of ``pyarrow.timestamp`` or ``pyarrow.duration`` type (:issue:`55989`)
+- Bug in :meth:`DataFrame.resample` where bin edges were not correct for :class:`~pandas.tseries.offsets.BusinessDay` (:issue:`55281`)
+- Bug in :meth:`DataFrame.resample` where bin edges were not correct for :class:`~pandas.tseries.offsets.MonthBegin` (:issue:`55271`)
+- Bug in :meth:`DataFrame.rolling` and :meth:`Series.rolling` where duplicate datetimelike indexes are treated as consecutive rather than equal with ``closed='left'`` and ``closed='neither'`` (:issue:`20712`)
+- Bug in :meth:`DataFrame.rolling` and :meth:`Series.rolling` where either the ``index`` or ``on`` column was :class:`ArrowDtype` with ``pyarrow.timestamp`` type (:issue:`55849`)
+
+Reshaping
+^^^^^^^^^
+- Bug in :func:`concat` ignoring ``sort`` parameter when passed :class:`DatetimeIndex` indexes (:issue:`54769`)
+- Bug in :func:`concat` renaming :class:`Series` when ``ignore_index=False`` (:issue:`15047`)
+- Bug in :func:`merge_asof` raising ``TypeError`` when ``by`` dtype is not ``object``, ``int64``, or ``uint64`` (:issue:`22794`)
+- Bug in :func:`merge_asof` raising incorrect error for string dtype (:issue:`56444`)
+- Bug in :func:`merge_asof` when using a :class:`Timedelta` tolerance on a :class:`ArrowDtype` column (:issue:`56486`)
+- Bug in :func:`merge` not raising when merging datetime columns with timedelta columns (:issue:`56455`)
+- Bug in :func:`merge` not raising when merging string columns with numeric columns (:issue:`56441`)
+- Bug in :func:`merge` not sorting for new string dtype (:issue:`56442`)
+- Bug in :func:`merge` returning columns in incorrect order when left and/or right is empty (:issue:`51929`)
+- Bug in :meth:`DataFrame.melt` where an exception was raised if ``var_name`` was not a string (:issue:`55948`)
+- Bug in :meth:`DataFrame.melt` where it would not preserve the datetime (:issue:`55254`)
+- Bug in :meth:`DataFrame.pivot_table` where the row margin is incorrect when the columns have numeric names (:issue:`26568`)
+- Bug in :meth:`DataFrame.pivot` with numeric columns and extension dtype for data (:issue:`56528`)
+- Bug in :meth:`DataFrame.stack` with ``future_stack=True`` would not preserve NA values in the index (:issue:`56573`)
+
+Sparse
+^^^^^^
+- Bug in :meth:`arrays.SparseArray.take` when using a different fill value than the array's fill value (:issue:`55181`)
+
+Other
+^^^^^
+- :meth:`DataFrame.__dataframe__` did not support pyarrow large strings (:issue:`56702`)
+- Bug in :func:`DataFrame.describe` when formatting percentiles in the resulting percentile 99.999% is rounded to 100% (:issue:`55765`)
+- Bug in :func:`api.interchange.from_dataframe` where it raised  ``NotImplementedError`` when handling empty string columns (:issue:`56703`)
+- Bug in :func:`cut` and :func:`qcut` with ``datetime64`` dtype values with non-nanosecond units incorrectly returning nanosecond-unit bins (:issue:`56101`)
+- Bug in :func:`cut` incorrectly allowing cutting of timezone-aware datetimes with timezone-naive bins (:issue:`54964`)
+- Bug in :func:`infer_freq` and :meth:`DatetimeIndex.inferred_freq` with weekly frequencies and non-nanosecond resolutions (:issue:`55609`)
+- Bug in :meth:`DataFrame.apply` where passing ``raw=True`` ignored ``args`` passed to the applied function (:issue:`55009`)
+- Bug in :meth:`DataFrame.from_dict` which would always sort the rows of the created :class:`DataFrame`.  (:issue:`55683`)
+- Bug in :meth:`DataFrame.sort_index` when passing ``axis="columns"`` and ``ignore_index=True`` raising a ``ValueError`` (:issue:`56478`)
+- Bug in rendering ``inf`` values inside a :class:`DataFrame` with the ``use_inf_as_na`` option enabled (:issue:`55483`)
+- Bug in rendering a :class:`Series` with a :class:`MultiIndex` when one of the index level's names is 0 not having that name displayed (:issue:`55415`)
+- Bug in the error message when assigning an empty :class:`DataFrame` to a column (:issue:`55956`)
+- Bug when time-like strings were being cast to :class:`ArrowDtype` with ``pyarrow.time64`` type (:issue:`56463`)
+- Fixed a spurious deprecation warning from ``numba`` >= 0.58.0 when passing a numpy ufunc in :class:`core.window.Rolling.apply` with ``engine="numba"`` (:issue:`55247`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_220.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.1.4..v2.2.0
diff --git a/pandas/doc/source/whatsnew/v2.2.1.rst b/pandas/doc/source/whatsnew/v2.2.1.rst
new file mode 100644
index 0000000000000..4db0069ec4b95
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.2.1.rst
@@ -0,0 +1,90 @@
+.. _whatsnew_221:
+
+What's new in 2.2.1 (February 22, 2024)
+---------------------------------------
+
+These are the changes in pandas 2.2.1. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_221.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+- Added ``pyarrow`` pip extra so users can install pandas and pyarrow with pip with ``pip install pandas[pyarrow]`` (:issue:`54466`)
+
+.. _whatsnew_221.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- Fixed memory leak in :func:`read_csv` (:issue:`57039`)
+- Fixed performance regression in :meth:`Series.combine_first` (:issue:`55845`)
+- Fixed regression causing overflow for near-minimum timestamps (:issue:`57150`)
+- Fixed regression in :func:`concat` changing long-standing behavior that always sorted the non-concatenation axis when the axis was a :class:`DatetimeIndex` (:issue:`57006`)
+- Fixed regression in :func:`merge_ordered` raising ``TypeError`` for ``fill_method="ffill"`` and ``how="left"`` (:issue:`57010`)
+- Fixed regression in :func:`pandas.testing.assert_series_equal` defaulting to ``check_exact=True`` when checking the :class:`Index` (:issue:`57067`)
+- Fixed regression in :func:`read_json` where an :class:`Index` would be returned instead of a :class:`RangeIndex` (:issue:`57429`)
+- Fixed regression in :func:`wide_to_long` raising an ``AttributeError`` for string columns (:issue:`57066`)
+- Fixed regression in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.SeriesGroupBy.idxmax` ignoring the ``skipna`` argument (:issue:`57040`)
+- Fixed regression in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.SeriesGroupBy.idxmax` where values containing the minimum or maximum value for the dtype could produce incorrect results (:issue:`57040`)
+- Fixed regression in :meth:`CategoricalIndex.difference` raising ``KeyError`` when other contains null values other than NaN (:issue:`57318`)
+- Fixed regression in :meth:`DataFrame.groupby` raising ``ValueError`` when grouping by a :class:`Series` in some cases (:issue:`57276`)
+- Fixed regression in :meth:`DataFrame.loc` raising ``IndexError`` for non-unique, masked dtype indexes where result has more than 10,000 rows (:issue:`57027`)
+- Fixed regression in :meth:`DataFrame.loc` which was unnecessarily throwing "incompatible dtype warning" when expanding with partial row indexer and multiple columns (see `PDEP6 <https://pandas.pydata.org/pdeps/0006-ban-upcasting.html>`_) (:issue:`56503`)
+- Fixed regression in :meth:`DataFrame.map` with ``na_action="ignore"`` not being respected for NumPy nullable and :class:`ArrowDtypes` (:issue:`57316`)
+- Fixed regression in :meth:`DataFrame.merge` raising ``ValueError`` for certain types of 3rd-party extension arrays (:issue:`57316`)
+- Fixed regression in :meth:`DataFrame.query` with all ``NaT`` column with object dtype (:issue:`57068`)
+- Fixed regression in :meth:`DataFrame.shift` raising ``AssertionError`` for ``axis=1`` and empty :class:`DataFrame` (:issue:`57301`)
+- Fixed regression in :meth:`DataFrame.sort_index` not producing a stable sort for a index with duplicates (:issue:`57151`)
+- Fixed regression in :meth:`DataFrame.to_dict` with ``orient='list'`` and datetime or timedelta types returning integers (:issue:`54824`)
+- Fixed regression in :meth:`DataFrame.to_json` converting nullable integers to floats (:issue:`57224`)
+- Fixed regression in :meth:`DataFrame.to_sql` when ``method="multi"`` is passed and the dialect type is not Oracle (:issue:`57310`)
+- Fixed regression in :meth:`DataFrame.transpose` with nullable extension dtypes not having F-contiguous data potentially causing exceptions when used (:issue:`57315`)
+- Fixed regression in :meth:`DataFrame.update` emitting incorrect warnings about downcasting (:issue:`57124`)
+- Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` ignoring the ``skipna`` argument (:issue:`57040`)
+- Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` where values containing the minimum or maximum value for the dtype could produce incorrect results (:issue:`57040`)
+- Fixed regression in :meth:`ExtensionArray.to_numpy` raising for non-numeric masked dtypes (:issue:`56991`)
+- Fixed regression in :meth:`Index.join` raising ``TypeError`` when joining an empty index to a non-empty index containing mixed dtype values (:issue:`57048`)
+- Fixed regression in :meth:`Series.astype` introducing decimals when converting from integer with missing values to string dtype (:issue:`57418`)
+- Fixed regression in :meth:`Series.pct_change` raising a ``ValueError`` for an empty :class:`Series` (:issue:`57056`)
+- Fixed regression in :meth:`Series.to_numpy` when dtype is given as float and the data contains NaNs (:issue:`57121`)
+- Fixed regression in addition or subtraction of :class:`DateOffset` objects with millisecond components to ``datetime64`` :class:`Index`, :class:`Series`, or :class:`DataFrame` (:issue:`57529`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_221.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Fixed bug in :func:`pandas.api.interchange.from_dataframe` which was raising for Nullable integers (:issue:`55069`)
+- Fixed bug in :func:`pandas.api.interchange.from_dataframe` which was raising for empty inputs (:issue:`56700`)
+- Fixed bug in :func:`pandas.api.interchange.from_dataframe` which wasn't converting columns names to strings (:issue:`55069`)
+- Fixed bug in :meth:`DataFrame.__getitem__` for empty :class:`DataFrame` with Copy-on-Write enabled (:issue:`57130`)
+- Fixed bug in :meth:`PeriodIndex.asfreq` which was silently converting frequencies which are not supported as period frequencies instead of raising an error (:issue:`56945`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_221.other:
+
+Other
+~~~~~
+
+.. note::
+
+    The ``DeprecationWarning`` that was raised when pandas was imported without PyArrow being
+    installed has been removed. This decision was made because the warning was too noisy for too
+    many users and a lot of feedback was collected about the decision to make PyArrow a required
+    dependency. Pandas is currently considering the decision whether or not PyArrow should be added
+    as a hard dependency in 3.0. Interested users can follow the discussion
+    `here <https://github.com/pandas-dev/pandas/issues/57073>`_.
+
+- Added the argument ``skipna`` to :meth:`DataFrameGroupBy.first`, :meth:`DataFrameGroupBy.last`, :meth:`SeriesGroupBy.first`, and :meth:`SeriesGroupBy.last`; achieving ``skipna=False`` used to be available via :meth:`DataFrameGroupBy.nth`, but the behavior was changed in pandas 2.0.0 (:issue:`57019`)
+- Added the argument ``skipna`` to :meth:`Resampler.first`, :meth:`Resampler.last` (:issue:`57019`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_221.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.2.0..v2.2.1
diff --git a/pandas/doc/source/whatsnew/v2.2.2.rst b/pandas/doc/source/whatsnew/v2.2.2.rst
new file mode 100644
index 0000000000000..fbe5e9b4febb5
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.2.2.rst
@@ -0,0 +1,59 @@
+.. _whatsnew_222:
+
+What's new in 2.2.2 (April 10, 2024)
+---------------------------------------
+
+These are the changes in pandas 2.2.2. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_220.np2_compat:
+
+Pandas 2.2.2 is now compatible with numpy 2.0
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Pandas 2.2.2 is the first version of pandas that is generally compatible with the upcoming
+numpy 2.0 release, and wheels for pandas 2.2.2 will work with both numpy 1.x and 2.x.
+
+One major caveat is that arrays created with numpy 2.0's new ``StringDtype`` will convert
+to ``object`` dtyped arrays upon :class:`Series`/:class:`DataFrame` creation.
+Full support for numpy 2.0's StringDtype is expected to land in pandas 3.0.
+
+As usual please report any bugs discovered to our `issue tracker <https://github.com/pandas-dev/pandas/issues/new/choose>`_
+
+.. _whatsnew_222.regressions:
+
+Fixed regressions
+~~~~~~~~~~~~~~~~~
+- :meth:`DataFrame.__dataframe__` was producing incorrect data buffers when the a column's type was a pandas nullable on with missing values (:issue:`56702`)
+- :meth:`DataFrame.__dataframe__` was producing incorrect data buffers when the a column's type was a pyarrow nullable on with missing values (:issue:`57664`)
+- Avoid issuing a spurious ``DeprecationWarning`` when a custom :class:`DataFrame` or :class:`Series` subclass method is called (:issue:`57553`)
+- Fixed regression in precision of :func:`to_datetime` with string and ``unit`` input (:issue:`57051`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_222.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- :meth:`DataFrame.__dataframe__` was producing incorrect data buffers when the column's type was nullable boolean (:issue:`55332`)
+- :meth:`DataFrame.__dataframe__` was showing bytemask instead of bitmask for ``'string[pyarrow]'`` validity buffer (:issue:`57762`)
+- :meth:`DataFrame.__dataframe__` was showing non-null validity buffer (instead of ``None``) ``'string[pyarrow]'`` without missing values (:issue:`57761`)
+- :meth:`DataFrame.to_sql` was failing to find the right table when using the schema argument (:issue:`57539`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_222.other:
+
+Other
+~~~~~
+-
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_222.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.2.1..v2.2.2
diff --git a/pandas/doc/source/whatsnew/v2.2.3.rst b/pandas/doc/source/whatsnew/v2.2.3.rst
new file mode 100644
index 0000000000000..1696a7b6449af
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.2.3.rst
@@ -0,0 +1,45 @@
+.. _whatsnew_223:
+
+What's new in 2.2.3 (September 20, 2024)
+----------------------------------------
+
+These are the changes in pandas 2.2.3. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_220.py13_compat:
+
+Pandas 2.2.3 is now compatible with Python 3.13
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Pandas 2.2.3 is the first version of pandas that is generally compatible with the upcoming
+Python 3.13, and both wheels for free-threaded and normal Python 3.13 will be uploaded for
+this release.
+
+As usual please report any bugs discovered to our `issue tracker <https://github.com/pandas-dev/pandas/issues/new/choose>`_
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_223.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+- Bug in :func:`eval` on :class:`complex` including division ``/`` discards imaginary part. (:issue:`21374`)
+- Minor fixes for numpy 2.1 compatibility. (:issue:`59444`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_223.other:
+
+Other
+~~~~~
+- Missing licenses for 3rd party dependencies were added back into the wheels. (:issue:`58632`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_223.contributors:
+
+Contributors
+~~~~~~~~~~~~
+
+.. contributors:: v2.2.2..v2.2.3|HEAD
diff --git a/pandas/doc/source/whatsnew/v2.3.0.rst b/pandas/doc/source/whatsnew/v2.3.0.rst
new file mode 100644
index 0000000000000..09134763977c3
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v2.3.0.rst
@@ -0,0 +1,197 @@
+.. _whatsnew_230:
+
+What's new in 2.3.0 (Month XX, 2024)
+------------------------------------
+
+These are the changes in pandas 2.3.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+
+.. _whatsnew_230.upcoming_changes:
+
+Upcoming changes in pandas 3.0
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+
+.. _whatsnew_230.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_230.enhancements.enhancement1:
+
+enhancement1
+^^^^^^^^^^^^
+
+
+.. _whatsnew_230.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+
+- The semantics for the ``copy`` keyword in ``__array__`` methods (i.e. called
+  when using ``np.array()`` or ``np.asarray()`` on pandas objects) has been
+  updated to work correctly with NumPy >= 2 (:issue:`57739`)
+- :meth:`Series.str.decode` result now has ``StringDtype`` when ``future.infer_string`` is True (:issue:`60709`)
+- :meth:`~Series.to_hdf` and :meth:`~DataFrame.to_hdf` now round-trip with ``StringDtype``  (:issue:`60663`)
+- The :meth:`Series.str.decode` has gained the argument ``dtype`` to control the dtype of the result (:issue:`60940`)
+- The :meth:`~Series.cumsum`, :meth:`~Series.cummin`, and :meth:`~Series.cummax` reductions are now implemented for ``StringDtype`` columns (:issue:`60633`)
+- The :meth:`~Series.sum` reduction is now implemented for ``StringDtype`` columns (:issue:`59853`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_230.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_230.notable_bug_fixes.notable_bug_fix1:
+
+notable_bug_fix1
+^^^^^^^^^^^^^^^^
+
+.. _whatsnew_230.api_changes:
+
+API changes
+~~~~~~~~~~~
+
+- When enabling the ``future.infer_string`` option: Index set operations (like
+  union or intersection) will now ignore the dtype of an empty ``RangeIndex`` or
+  empty ``Index`` with object dtype when determining the dtype of the resulting
+  Index (:issue:`60797`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_230.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+- Deprecated allowing non-``bool`` values for ``na`` in :meth:`.str.contains`, :meth:`.str.startswith`, and :meth:`.str.endswith` for dtypes that do not already disallow these (:issue:`59615`)
+- Deprecated the ``"pyarrow_numpy"`` storage option for :class:`StringDtype` (:issue:`60152`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_230.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+-
+-
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_230.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+-
+-
+
+Datetimelike
+^^^^^^^^^^^^
+-
+-
+
+Timedelta
+^^^^^^^^^
+-
+-
+
+Timezones
+^^^^^^^^^
+-
+-
+
+Numeric
+^^^^^^^
+- Enabled :class:`Series.mode` and :class:`DataFrame.mode` with ``dropna=False`` to sort the result for all dtypes in the presence of NA values; previously only certain dtypes would sort (:issue:`60702`)
+-
+
+Conversion
+^^^^^^^^^^
+-
+-
+
+Strings
+^^^^^^^
+- Bug in :meth:`Series.__pos__` and :meth:`DataFrame.__pos__` did not raise for :class:`StringDtype` with ``storage="pyarrow"`` (:issue:`60710`)
+- Bug in :meth:`Series.rank` for :class:`StringDtype` with ``storage="pyarrow"`` incorrectly returning integer results in case of ``method="average"`` and raising an error if it would truncate results (:issue:`59768`)
+- Bug in :meth:`Series.replace` with :class:`StringDtype` when replacing with a non-string value was not upcasting to ``object`` dtype (:issue:`60282`)
+- Bug in :meth:`Series.str.replace` when ``n < 0`` for :class:`StringDtype` with ``storage="pyarrow"`` (:issue:`59628`)
+- Bug in ``ser.str.slice`` with negative ``step`` with :class:`ArrowDtype` and :class:`StringDtype` with ``storage="pyarrow"`` giving incorrect results (:issue:`59710`)
+- Bug in the ``center`` method on :class:`Series` and :class:`Index` object ``str`` accessors with pyarrow-backed dtype not matching the python behavior in corner cases with an odd number of fill characters (:issue:`54792`)
+
+Interval
+^^^^^^^^
+-
+-
+
+Indexing
+^^^^^^^^
+- Fixed bug in :meth:`Index.get_indexer` round-tripping through string dtype when ``infer_string`` is enabled (:issue:`55834`)
+-
+
+Missing
+^^^^^^^
+-
+-
+
+MultiIndex
+^^^^^^^^^^
+-
+-
+
+I/O
+^^^
+- :meth:`DataFrame.to_excel` was storing decimals as strings instead of numbers (:issue:`49598`)
+-
+
+Period
+^^^^^^
+-
+-
+
+Plotting
+^^^^^^^^
+-
+-
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+-
+-
+
+Reshaping
+^^^^^^^^^
+-
+-
+
+Sparse
+^^^^^^
+-
+-
+
+ExtensionArray
+^^^^^^^^^^^^^^
+-
+-
+
+Styler
+^^^^^^
+-
+-
+
+Other
+^^^^^
+- Fixed usage of ``inspect`` when the optional dependencies ``pyarrow`` or ``jinja2``
+  are not installed (:issue:`60196`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_230.contributors:
+
+Contributors
+~~~~~~~~~~~~
diff --git a/pandas/doc/source/whatsnew/v3.0.0.rst b/pandas/doc/source/whatsnew/v3.0.0.rst
new file mode 100644
index 0000000000000..7ebcf18a36a96
--- /dev/null
+++ b/pandas/doc/source/whatsnew/v3.0.0.rst
@@ -0,0 +1,847 @@
+.. _whatsnew_300:
+
+What's new in 3.0.0 (Month XX, 2024)
+------------------------------------
+
+These are the changes in pandas 3.0.0. See :ref:`release` for a full changelog
+including other versions of pandas.
+
+{{ header }}
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.enhancements:
+
+Enhancements
+~~~~~~~~~~~~
+
+.. _whatsnew_300.enhancements.enhancement1:
+
+Enhancement1
+^^^^^^^^^^^^
+
+.. _whatsnew_300.enhancements.enhancement2:
+
+Enhancement2
+^^^^^^^^^^^^
+
+.. _whatsnew_300.enhancements.other:
+
+Other enhancements
+^^^^^^^^^^^^^^^^^^
+- :class:`pandas.api.typing.FrozenList` is available for typing the outputs of :attr:`MultiIndex.names`, :attr:`MultiIndex.codes` and :attr:`MultiIndex.levels` (:issue:`58237`)
+- :class:`pandas.api.typing.SASReader` is available for typing the output of :func:`read_sas` (:issue:`55689`)
+- :meth:`pandas.api.interchange.from_dataframe` now uses the `PyCapsule Interface <https://arrow.apache.org/docs/format/CDataInterface/PyCapsuleInterface.html>`_ if available, only falling back to the Dataframe Interchange Protocol if that fails (:issue:`60739`)
+- Added :meth:`.Styler.to_typst` to write Styler objects to file, buffer or string in Typst format (:issue:`57617`)
+- Added missing :meth:`pandas.Series.info` to API reference (:issue:`60926`)
+- :class:`pandas.api.typing.NoDefault` is available for typing ``no_default``
+- :func:`DataFrame.to_excel` now raises an ``UserWarning`` when the character count in a cell exceeds Excel's limitation of 32767 characters (:issue:`56954`)
+- :func:`pandas.merge` now validates the ``how`` parameter input (merge type) (:issue:`59435`)
+- :func:`pandas.merge`, :meth:`DataFrame.merge` and :meth:`DataFrame.join` now support anti joins (``left_anti`` and ``right_anti``) in the ``how`` parameter (:issue:`42916`)
+- :func:`read_spss` now supports kwargs to be passed to pyreadstat (:issue:`56356`)
+- :func:`read_stata` now returns ``datetime64`` resolutions better matching those natively stored in the stata format (:issue:`55642`)
+- :meth:`DataFrame.agg` called with ``axis=1`` and a ``func`` which relabels the result index now raises a ``NotImplementedError`` (:issue:`58807`).
+- :meth:`Index.get_loc` now accepts also subclasses of ``tuple`` as keys (:issue:`57922`)
+- :meth:`Styler.set_tooltips` provides alternative method to storing tooltips by using title attribute of td elements. (:issue:`56981`)
+- Added missing parameter ``weights`` in :meth:`DataFrame.plot.kde` for the estimation of the PDF (:issue:`59337`)
+- Allow dictionaries to be passed to :meth:`pandas.Series.str.replace` via ``pat`` parameter (:issue:`51748`)
+- Support passing a :class:`Series` input to :func:`json_normalize` that retains the :class:`Series` :class:`Index` (:issue:`51452`)
+- Support reading value labels from Stata 108-format (Stata 6) and earlier files (:issue:`58154`)
+- Users can globally disable any ``PerformanceWarning`` by setting the option ``mode.performance_warnings`` to ``False`` (:issue:`56920`)
+- :meth:`Styler.format_index_names` can now be used to format the index and column names (:issue:`48936` and :issue:`47489`)
+- :class:`.errors.DtypeWarning` improved to include column names when mixed data types are detected (:issue:`58174`)
+- :class:`Rolling` and :class:`Expanding` now support ``pipe`` method (:issue:`57076`)
+- :class:`Series` now supports the Arrow PyCapsule Interface for export (:issue:`59518`)
+- :func:`DataFrame.to_excel` argument ``merge_cells`` now accepts a value of ``"columns"`` to only merge :class:`MultiIndex` column header header cells (:issue:`35384`)
+- :meth:`DataFrame.corrwith` now accepts ``min_periods`` as optional arguments, as in :meth:`DataFrame.corr` and :meth:`Series.corr` (:issue:`9490`)
+- :meth:`DataFrame.cummin`, :meth:`DataFrame.cummax`, :meth:`DataFrame.cumprod` and :meth:`DataFrame.cumsum` methods now have a ``numeric_only`` parameter (:issue:`53072`)
+- :meth:`DataFrame.ewm` now allows ``adjust=False`` when ``times`` is provided (:issue:`54328`)
+- :meth:`DataFrame.fillna` and :meth:`Series.fillna` can now accept ``value=None``; for non-object dtype the corresponding NA value will be used (:issue:`57723`)
+- :meth:`DataFrame.pivot_table` and :func:`pivot_table` now allow the passing of keyword arguments to ``aggfunc`` through ``**kwargs`` (:issue:`57884`)
+- :meth:`DataFrame.to_json` now encodes ``Decimal`` as strings instead of floats (:issue:`60698`)
+- :meth:`Series.cummin` and :meth:`Series.cummax` now supports :class:`CategoricalDtype` (:issue:`52335`)
+- :meth:`Series.plot` now correctly handle the ``ylabel`` parameter for pie charts, allowing for explicit control over the y-axis label (:issue:`58239`)
+- :meth:`DataFrame.plot.scatter` argument ``c`` now accepts a column of strings, where rows with the same string are colored identically (:issue:`16827` and :issue:`16485`)
+- :class:`DataFrameGroupBy` and :class:`SeriesGroupBy` methods ``sum``, ``mean``, ``median``, ``prod``, ``min``, ``max``, ``std``, ``var`` and ``sem`` now accept ``skipna`` parameter (:issue:`15675`)
+- :class:`Rolling` and :class:`Expanding` now support aggregations ``first`` and ``last`` (:issue:`33155`)
+- :func:`read_parquet` accepts ``to_pandas_kwargs`` which are forwarded to :meth:`pyarrow.Table.to_pandas` which enables passing additional keywords to customize the conversion to pandas, such as ``maps_as_pydicts`` to read the Parquet map data type as python dictionaries (:issue:`56842`)
+- :meth:`.DataFrameGroupBy.transform`, :meth:`.SeriesGroupBy.transform`, :meth:`.DataFrameGroupBy.agg`, :meth:`.SeriesGroupBy.agg`, :meth:`.SeriesGroupBy.apply`, :meth:`.DataFrameGroupBy.apply` now support ``kurt`` (:issue:`40139`)
+- :meth:`DataFrameGroupBy.transform`, :meth:`SeriesGroupBy.transform`, :meth:`DataFrameGroupBy.agg`, :meth:`SeriesGroupBy.agg`, :meth:`RollingGroupby.apply`, :meth:`ExpandingGroupby.apply`, :meth:`Rolling.apply`, :meth:`Expanding.apply`, :meth:`DataFrame.apply` with ``engine="numba"`` now supports positional arguments passed as kwargs (:issue:`58995`)
+- :meth:`Rolling.agg`, :meth:`Expanding.agg` and :meth:`ExponentialMovingWindow.agg` now accept :class:`NamedAgg` aggregations through ``**kwargs`` (:issue:`28333`)
+- :meth:`Series.map` can now accept kwargs to pass on to func (:issue:`59814`)
+- :meth:`Series.str.get_dummies` now accepts a  ``dtype`` parameter to specify the dtype of the resulting DataFrame (:issue:`47872`)
+- :meth:`pandas.concat` will raise a ``ValueError`` when ``ignore_index=True`` and ``keys`` is not ``None`` (:issue:`59274`)
+- :py:class:`frozenset` elements in pandas objects are now natively printed (:issue:`60690`)
+- Add ``"delete_rows"`` option to ``if_exists`` argument in :meth:`DataFrame.to_sql` deleting all records of the table before inserting data (:issue:`37210`).
+- Errors occurring during SQL I/O will now throw a generic :class:`.DatabaseError` instead of the raw Exception type from the underlying driver manager library (:issue:`60748`)
+- Implemented :meth:`Series.str.isascii` and :meth:`Series.str.isascii` (:issue:`59091`)
+- Multiplying two :class:`DateOffset` objects will now raise a ``TypeError`` instead of a ``RecursionError`` (:issue:`59442`)
+- Restore support for reading Stata 104-format and enable reading 103-format dta files (:issue:`58554`)
+- Support passing a :class:`Iterable[Hashable]` input to :meth:`DataFrame.drop_duplicates` (:issue:`59237`)
+- Support reading Stata 102-format (Stata 1) dta files (:issue:`58978`)
+- Support reading Stata 110-format (Stata 7) dta files (:issue:`47176`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.notable_bug_fixes:
+
+Notable bug fixes
+~~~~~~~~~~~~~~~~~
+
+These are bug fixes that might have notable behavior changes.
+
+.. _whatsnew_300.notable_bug_fixes.groupby_unobs_and_na:
+
+Improved behavior in groupby for ``observed=False``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+A number of bugs have been fixed due to improved handling of unobserved groups (:issue:`55738`). All remarks in this section equally impact :class:`.SeriesGroupBy`.
+
+In previous versions of pandas, a single grouping with :meth:`.DataFrameGroupBy.apply` or :meth:`.DataFrameGroupBy.agg` would pass the unobserved groups to the provided function, resulting in ``0`` below.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {
+            "key1": pd.Categorical(list("aabb"), categories=list("abc")),
+            "key2": [1, 1, 1, 2],
+            "values": [1, 2, 3, 4],
+        }
+    )
+    df
+    gb = df.groupby("key1", observed=False)
+    gb[["values"]].apply(lambda x: x.sum())
+
+However this was not the case when using multiple groupings, resulting in ``NaN`` below.
+
+.. code-block:: ipython
+
+    In [1]: gb = df.groupby(["key1", "key2"], observed=False)
+    In [2]: gb[["values"]].apply(lambda x: x.sum())
+    Out[2]:
+               values
+    key1 key2
+    a    1        3.0
+         2        NaN
+    b    1        3.0
+         2        4.0
+    c    1        NaN
+         2        NaN
+
+Now using multiple groupings will also pass the unobserved groups to the provided function.
+
+.. ipython:: python
+
+    gb = df.groupby(["key1", "key2"], observed=False)
+    gb[["values"]].apply(lambda x: x.sum())
+
+Similarly:
+
+- In previous versions of pandas the method :meth:`.DataFrameGroupBy.sum` would result in ``0`` for unobserved groups, but :meth:`.DataFrameGroupBy.prod`, :meth:`.DataFrameGroupBy.all`, and :meth:`.DataFrameGroupBy.any` would all result in NA values. Now these methods result in ``1``, ``True``, and ``False`` respectively.
+- :meth:`.DataFrameGroupBy.groups` did not include unobserved groups and now does.
+
+These improvements also fixed certain bugs in groupby:
+
+- :meth:`.DataFrameGroupBy.agg` would fail when there are multiple groupings, unobserved groups, and ``as_index=False`` (:issue:`36698`)
+- :meth:`.DataFrameGroupBy.groups` with ``sort=False`` would sort groups; they now occur in the order they are observed (:issue:`56966`)
+- :meth:`.DataFrameGroupBy.nunique` would fail when there are multiple groupings, unobserved groups, and ``as_index=False`` (:issue:`52848`)
+- :meth:`.DataFrameGroupBy.sum` would have incorrect values when there are multiple groupings, unobserved groups, and non-numeric data (:issue:`43891`)
+- :meth:`.DataFrameGroupBy.value_counts` would produce incorrect results when used with some categorical and some non-categorical groupings and ``observed=False`` (:issue:`56016`)
+
+.. _whatsnew_300.notable_bug_fixes.notable_bug_fix2:
+
+notable_bug_fix2
+^^^^^^^^^^^^^^^^
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.api_breaking:
+
+Backwards incompatible API changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. _whatsnew_300.api_breaking.datetime_resolution_inference:
+
+Datetime resolution inference
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Converting a sequence of strings, ``datetime`` objects, or ``np.datetime64`` objects to
+a ``datetime64`` dtype now performs inference on the appropriate resolution (AKA unit) for the output dtype. This affects :class:`Series`, :class:`DataFrame`, :class:`Index`, :class:`DatetimeIndex`, and :func:`to_datetime`.
+
+Previously, these would always give nanosecond resolution:
+
+.. code-block:: ipython
+
+    In [1]: dt = pd.Timestamp("2024-03-22 11:36").to_pydatetime()
+    In [2]: pd.to_datetime([dt]).dtype
+    Out[2]: dtype('<M8[ns]')
+    In [3]: pd.Index([dt]).dtype
+    Out[3]: dtype('<M8[ns]')
+    In [4]: pd.DatetimeIndex([dt]).dtype
+    Out[4]: dtype('<M8[ns]')
+    In [5]: pd.Series([dt]).dtype
+    Out[5]: dtype('<M8[ns]')
+
+This now infers the unit microsecond unit "us" from the pydatetime object, matching the scalar :class:`Timestamp` behavior.
+
+.. ipython:: python
+
+    In [1]: dt = pd.Timestamp("2024-03-22 11:36").to_pydatetime()
+    In [2]: pd.to_datetime([dt]).dtype
+    In [3]: pd.Index([dt]).dtype
+    In [4]: pd.DatetimeIndex([dt]).dtype
+    In [5]: pd.Series([dt]).dtype
+
+Similar when passed a sequence of ``np.datetime64`` objects, the resolution of the passed objects will be retained (or for lower-than-second resolution, second resolution will be used).
+
+When passing strings, the resolution will depend on the precision of the string, again matching the :class:`Timestamp` behavior. Previously:
+
+.. code-block:: ipython
+
+    In [2]: pd.to_datetime(["2024-03-22 11:43:01"]).dtype
+    Out[2]: dtype('<M8[ns]')
+    In [3]: pd.to_datetime(["2024-03-22 11:43:01.002"]).dtype
+    Out[3]: dtype('<M8[ns]')
+    In [4]: pd.to_datetime(["2024-03-22 11:43:01.002003"]).dtype
+    Out[4]: dtype('<M8[ns]')
+    In [5]: pd.to_datetime(["2024-03-22 11:43:01.002003004"]).dtype
+    Out[5]: dtype('<M8[ns]')
+
+The inferred resolution now matches that of the input strings:
+
+.. ipython:: python
+
+    In [2]: pd.to_datetime(["2024-03-22 11:43:01"]).dtype
+    In [3]: pd.to_datetime(["2024-03-22 11:43:01.002"]).dtype
+    In [4]: pd.to_datetime(["2024-03-22 11:43:01.002003"]).dtype
+    In [5]: pd.to_datetime(["2024-03-22 11:43:01.002003004"]).dtype
+
+In cases with mixed-resolution inputs, the highest resolution is used:
+
+.. code-block:: ipython
+
+    In [2]: pd.to_datetime([pd.Timestamp("2024-03-22 11:43:01"), "2024-03-22 11:43:01.002"]).dtype
+    Out[2]: dtype('<M8[ns]')
+
+.. _whatsnew_300.api_breaking.value_counts_sorting:
+
+Changed behavior in :meth:`DataFrame.value_counts` and :meth:`DataFrameGroupBy.value_counts` when ``sort=False``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In previous versions of pandas, :meth:`DataFrame.value_counts` with ``sort=False`` would sort the result by row labels (as was documented). This was nonintuitive and inconsistent with :meth:`Series.value_counts` which would maintain the order of the input. Now :meth:`DataFrame.value_counts` will maintain the order of the input.
+
+.. ipython:: python
+
+    df = pd.DataFrame(
+        {
+            "a": [2, 2, 2, 2, 1, 1, 1, 1],
+            "b": [2, 1, 3, 1, 2, 3, 1, 1],
+        }
+    )
+    df
+
+*Old behavior*
+
+.. code-block:: ipython
+
+    In [3]: df.value_counts(sort=False)
+    Out[3]:
+    a  b
+    1  1    2
+       2    1
+       3    1
+    2  1    2
+       2    1
+       3    1
+    Name: count, dtype: int64
+
+*New behavior*
+
+.. ipython:: python
+
+    df.value_counts(sort=False)
+
+This change also applies to :meth:`.DataFrameGroupBy.value_counts`. Here, there are two options for sorting: one ``sort`` passed to :meth:`DataFrame.groupby` and one passed directly to :meth:`.DataFrameGroupBy.value_counts`. The former will determine whether to sort the groups, the latter whether to sort the counts. All non-grouping columns will maintain the order of the input *within groups*.
+
+*Old behavior*
+
+.. code-block:: ipython
+
+    In [5]: df.groupby("a", sort=True).value_counts(sort=False)
+    Out[5]:
+    a  b
+    1  1    2
+       2    1
+       3    1
+    2  1    2
+       2    1
+       3    1
+    dtype: int64
+
+*New behavior*
+
+.. ipython:: python
+
+    df.groupby("a", sort=True).value_counts(sort=False)
+
+.. _whatsnew_300.api_breaking.deps:
+
+Increased minimum version for Python
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas 3.0.0 supports Python 3.10 and higher.
+
+Increased minimum versions for dependencies
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Some minimum supported versions of dependencies were updated.
+If installed, we now require:
+
++-----------------+-----------------+----------+---------+
+| Package         | Minimum Version | Required | Changed |
++=================+=================+==========+=========+
+| numpy           | 1.23.5          |    X     |    X    |
++-----------------+-----------------+----------+---------+
+
+For `optional libraries <https://pandas.pydata.org/docs/getting_started/install.html>`_ the general recommendation is to use the latest version.
+The following table lists the lowest version per library that is currently being tested throughout the development of pandas.
+Optional libraries below the lowest tested version may still work, but are not considered supported.
+
++------------------------+---------------------+
+| Package                | New Minimum Version |
++========================+=====================+
+| pytz                   | 2023.4              |
++------------------------+---------------------+
+| fastparquet            | 2023.10.0           |
++------------------------+---------------------+
+| adbc-driver-postgresql | 0.10.0              |
++------------------------+---------------------+
+| mypy (dev)             | 1.9.0               |
++------------------------+---------------------+
+
+See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
+
+.. _whatsnew_300.api_breaking.pytz:
+
+``pytz`` now an optional dependency
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+pandas now uses :py:mod:`zoneinfo` from the standard library as the default timezone implementation when passing a timezone
+string to various methods. (:issue:`34916`)
+
+*Old behavior:*
+
+.. code-block:: ipython
+
+    In [1]: ts = pd.Timestamp(2024, 1, 1).tz_localize("US/Pacific")
+    In [2]: ts.tz
+    <DstTzInfo 'US/Pacific' LMT-1 day, 16:07:00 STD>
+
+*New behavior:*
+
+.. ipython:: python
+
+    ts = pd.Timestamp(2024, 1, 1).tz_localize("US/Pacific")
+    ts.tz
+
+``pytz`` timezone objects are still supported when passed directly, but they will no longer be returned by default
+from string inputs. Moreover, ``pytz`` is no longer a required dependency of pandas, but can be installed
+with the pip extra ``pip install pandas[timezone]``.
+
+
+Additionally, pandas no longer throws ``pytz`` exceptions for timezone operations leading to ambiguous or nonexistent
+times. These cases will now raise a ``ValueError``.
+
+.. _whatsnew_300.api_breaking.other:
+
+Other API changes
+^^^^^^^^^^^^^^^^^
+- 3rd party ``py.path`` objects are no longer explicitly supported in IO methods. Use :py:class:`pathlib.Path` objects instead (:issue:`57091`)
+- :func:`read_table`'s ``parse_dates`` argument defaults to ``None`` to improve consistency with :func:`read_csv` (:issue:`57476`)
+- All classes inheriting from builtin ``tuple`` (including types created with :func:`collections.namedtuple`) are now hashed and compared as builtin ``tuple`` during indexing operations (:issue:`57922`)
+- Made ``dtype`` a required argument in :meth:`ExtensionArray._from_sequence_of_strings` (:issue:`56519`)
+- Passing a :class:`Series` input to :func:`json_normalize` will now retain the :class:`Series` :class:`Index`, previously output had a new :class:`RangeIndex` (:issue:`51452`)
+- Removed :meth:`Index.sort` which always raised a ``TypeError``. This attribute is not defined and will raise an ``AttributeError`` (:issue:`59283`)
+- Unused ``dtype`` argument has been removed from the :class:`MultiIndex` constructor (:issue:`60962`)
+- Updated :meth:`DataFrame.to_excel` so that the output spreadsheet has no styling. Custom styling can still be done using :meth:`Styler.to_excel` (:issue:`54154`)
+- pickle and HDF (``.h5``) files created with Python 2 are no longer explicitly supported (:issue:`57387`)
+- pickled objects from pandas version less than ``1.0.0`` are no longer supported (:issue:`57155`)
+- when comparing the indexes in :func:`testing.assert_series_equal`, check_exact defaults to True if an :class:`Index` is of integer dtypes. (:issue:`57386`)
+- Index set operations (like union or intersection) will now ignore the dtype of
+  an empty ``RangeIndex`` or empty ``Index`` with object dtype when determining
+  the dtype of the resulting Index (:issue:`60797`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.deprecations:
+
+Deprecations
+~~~~~~~~~~~~
+
+Copy keyword
+^^^^^^^^^^^^
+
+The ``copy`` keyword argument in the following methods is deprecated and
+will be removed in a future version:
+
+- :meth:`DataFrame.truncate` / :meth:`Series.truncate`
+- :meth:`DataFrame.tz_convert` / :meth:`Series.tz_convert`
+- :meth:`DataFrame.tz_localize` / :meth:`Series.tz_localize`
+- :meth:`DataFrame.infer_objects` / :meth:`Series.infer_objects`
+- :meth:`DataFrame.align` / :meth:`Series.align`
+- :meth:`DataFrame.astype` / :meth:`Series.astype`
+- :meth:`DataFrame.reindex` / :meth:`Series.reindex`
+- :meth:`DataFrame.reindex_like` / :meth:`Series.reindex_like`
+- :meth:`DataFrame.set_axis` / :meth:`Series.set_axis`
+- :meth:`DataFrame.to_period` / :meth:`Series.to_period`
+- :meth:`DataFrame.to_timestamp` / :meth:`Series.to_timestamp`
+- :meth:`DataFrame.rename` / :meth:`Series.rename`
+- :meth:`DataFrame.transpose`
+- :meth:`DataFrame.swaplevel`
+- :meth:`DataFrame.merge` / :func:`pd.merge`
+
+Copy-on-Write utilizes a lazy copy mechanism that defers copying the data until
+necessary. Use ``.copy`` to trigger an eager copy. The copy keyword has no effect
+starting with 3.0, so it can be safely removed from your code.
+
+Other Deprecations
+^^^^^^^^^^^^^^^^^^
+
+- Deprecated :func:`core.internals.api.make_block`, use public APIs instead (:issue:`56815`)
+- Deprecated :meth:`.DataFrameGroupby.corrwith` (:issue:`57158`)
+- Deprecated :meth:`Timestamp.utcfromtimestamp`, use ``Timestamp.fromtimestamp(ts, "UTC")`` instead (:issue:`56680`)
+- Deprecated :meth:`Timestamp.utcnow`, use ``Timestamp.now("UTC")`` instead (:issue:`56680`)
+- Deprecated allowing non-keyword arguments in :meth:`DataFrame.all`, :meth:`DataFrame.min`, :meth:`DataFrame.max`, :meth:`DataFrame.sum`, :meth:`DataFrame.prod`, :meth:`DataFrame.mean`, :meth:`DataFrame.median`, :meth:`DataFrame.sem`, :meth:`DataFrame.var`, :meth:`DataFrame.std`, :meth:`DataFrame.skew`, :meth:`DataFrame.kurt`, :meth:`Series.all`,  :meth:`Series.min`, :meth:`Series.max`, :meth:`Series.sum`, :meth:`Series.prod`, :meth:`Series.mean`, :meth:`Series.median`, :meth:`Series.sem`, :meth:`Series.var`, :meth:`Series.std`, :meth:`Series.skew`, and :meth:`Series.kurt`. (:issue:`57087`)
+- Deprecated allowing non-keyword arguments in :meth:`Series.to_markdown` except ``buf``. (:issue:`57280`)
+- Deprecated allowing non-keyword arguments in :meth:`Series.to_string` except ``buf``. (:issue:`57280`)
+- Deprecated behavior of :meth:`.DataFrameGroupBy.groups` and :meth:`.SeriesGroupBy.groups`, in a future version ``groups`` by one element list will return tuple instead of scalar. (:issue:`58858`)
+- Deprecated behavior of :meth:`Series.dt.to_pytimedelta`, in a future version this will return a :class:`Series` containing python ``datetime.timedelta`` objects instead of an ``ndarray`` of timedelta; this matches the behavior of other :meth:`Series.dt` properties. (:issue:`57463`)
+- Deprecated lowercase strings ``d``, ``b`` and ``c`` denoting frequencies in :class:`Day`, :class:`BusinessDay` and :class:`CustomBusinessDay` in favour of ``D``, ``B`` and ``C`` (:issue:`58998`)
+- Deprecated lowercase strings ``w``, ``w-mon``, ``w-tue``, etc. denoting frequencies in :class:`Week` in favour of ``W``, ``W-MON``, ``W-TUE``, etc. (:issue:`58998`)
+- Deprecated parameter ``method`` in :meth:`DataFrame.reindex_like` / :meth:`Series.reindex_like` (:issue:`58667`)
+- Deprecated strings ``w``, ``d``, ``MIN``, ``MS``, ``US`` and ``NS`` denoting units in :class:`Timedelta` in favour of ``W``, ``D``, ``min``, ``ms``, ``us`` and ``ns`` (:issue:`59051`)
+- Deprecated using ``epoch`` date format in :meth:`DataFrame.to_json` and :meth:`Series.to_json`, use ``iso`` instead. (:issue:`57063`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.prior_deprecations:
+
+Removal of prior version deprecations/changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Enforced deprecation of aliases ``M``, ``Q``, ``Y``, etc. in favour of ``ME``, ``QE``, ``YE``, etc. for offsets
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Renamed the following offset aliases (:issue:`57986`):
+
++-------------------------------+------------------+------------------+
+| offset                        | removed alias    | new alias        |
++===============================+==================+==================+
+|:class:`MonthEnd`              |      ``M``       |     ``ME``       |
++-------------------------------+------------------+------------------+
+|:class:`BusinessMonthEnd`      |      ``BM``      |     ``BME``      |
++-------------------------------+------------------+------------------+
+|:class:`SemiMonthEnd`          |      ``SM``      |     ``SME``      |
++-------------------------------+------------------+------------------+
+|:class:`CustomBusinessMonthEnd`|      ``CBM``     |     ``CBME``     |
++-------------------------------+------------------+------------------+
+|:class:`QuarterEnd`            |      ``Q``       |     ``QE``       |
++-------------------------------+------------------+------------------+
+|:class:`BQuarterEnd`           |      ``BQ``      |     ``BQE``      |
++-------------------------------+------------------+------------------+
+|:class:`YearEnd`               |      ``Y``       |     ``YE``       |
++-------------------------------+------------------+------------------+
+|:class:`BYearEnd`              |      ``BY``      |     ``BYE``      |
++-------------------------------+------------------+------------------+
+
+Other Removals
+^^^^^^^^^^^^^^
+- :class:`.DataFrameGroupBy.idxmin`, :class:`.DataFrameGroupBy.idxmax`, :class:`.SeriesGroupBy.idxmin`, and :class:`.SeriesGroupBy.idxmax` will now raise a ``ValueError`` when used with ``skipna=False`` and an NA value is encountered (:issue:`10694`)
+- :func:`concat` no longer ignores empty objects when determining output dtypes (:issue:`39122`)
+- :func:`concat` with all-NA entries no longer ignores the dtype of those entries when determining the result dtype (:issue:`40893`)
+- :func:`read_excel`, :func:`read_json`, :func:`read_html`, and :func:`read_xml` no longer accept raw string or byte representation of the data. That type of data must be wrapped in a :py:class:`StringIO` or :py:class:`BytesIO` (:issue:`53767`)
+- :func:`to_datetime` with a ``unit`` specified no longer parses strings into floats, instead parses them the same way as without ``unit`` (:issue:`50735`)
+- :meth:`DataFrame.groupby` with ``as_index=False`` and aggregation methods will no longer exclude from the result the groupings that do not arise from the input (:issue:`49519`)
+- :meth:`ExtensionArray._reduce` now requires a ``keepdims: bool = False`` parameter in the signature (:issue:`52788`)
+- :meth:`Series.dt.to_pydatetime` now returns a :class:`Series` of :py:class:`datetime.datetime` objects (:issue:`52459`)
+- :meth:`SeriesGroupBy.agg` no longer pins the name of the group to the input passed to the provided ``func`` (:issue:`51703`)
+- All arguments except ``name`` in :meth:`Index.rename` are now keyword only (:issue:`56493`)
+- All arguments except the first ``path``-like argument in IO writers are now keyword only (:issue:`54229`)
+- Changed behavior of :meth:`Series.__getitem__` and :meth:`Series.__setitem__` to always treat integer keys as labels, never as positional, consistent with :class:`DataFrame` behavior (:issue:`50617`)
+- Changed behavior of :meth:`Series.__getitem__`, :meth:`Series.__setitem__`, :meth:`DataFrame.__getitem__`, :meth:`DataFrame.__setitem__` with an integer slice on objects with a floating-dtype index. This is now treated as *positional* indexing (:issue:`49612`)
+- Disallow a callable argument to :meth:`Series.iloc` to return a ``tuple`` (:issue:`53769`)
+- Disallow allowing logical operations (``||``, ``&``, ``^``) between pandas objects and dtype-less sequences (e.g. ``list``, ``tuple``); wrap the objects in :class:`Series`, :class:`Index`, or ``np.array`` first instead (:issue:`52264`)
+- Disallow automatic casting to object in :class:`Series` logical operations (``&``, ``^``, ``||``) between series with mismatched indexes and dtypes other than ``object`` or ``bool`` (:issue:`52538`)
+- Disallow calling :meth:`Series.replace` or :meth:`DataFrame.replace` without a ``value`` and with non-dict-like ``to_replace`` (:issue:`33302`)
+- Disallow constructing a :class:`arrays.SparseArray` with scalar data (:issue:`53039`)
+- Disallow indexing an :class:`Index` with a boolean indexer of length zero, it now raises ``ValueError`` (:issue:`55820`)
+- Disallow non-standard (``np.ndarray``, :class:`Index`, :class:`ExtensionArray`, or :class:`Series`) to :func:`isin`, :func:`unique`, :func:`factorize` (:issue:`52986`)
+- Disallow passing a pandas type to :meth:`Index.view` (:issue:`55709`)
+- Disallow units other than "s", "ms", "us", "ns" for datetime64 and timedelta64 dtypes in :func:`array` (:issue:`53817`)
+- Removed "freq" keyword from :class:`PeriodArray` constructor, use "dtype" instead (:issue:`52462`)
+- Removed 'fastpath' keyword in :class:`Categorical` constructor (:issue:`20110`)
+- Removed 'kind' keyword in :meth:`Series.resample` and :meth:`DataFrame.resample` (:issue:`58125`)
+- Removed ``Block``, ``DatetimeTZBlock``, ``ExtensionBlock``, ``create_block_manager_from_blocks`` from ``pandas.core.internals`` and ``pandas.core.internals.api`` (:issue:`55139`)
+- Removed alias :class:`arrays.PandasArray` for :class:`arrays.NumpyExtensionArray` (:issue:`53694`)
+- Removed deprecated "method" and "limit" keywords from :meth:`Series.replace` and :meth:`DataFrame.replace` (:issue:`53492`)
+- Removed extension test classes ``BaseNoReduceTests``, ``BaseNumericReduceTests``, ``BaseBooleanReduceTests`` (:issue:`54663`)
+- Removed the "closed" and "normalize" keywords in :meth:`DatetimeIndex.__new__` (:issue:`52628`)
+- Removed the deprecated ``delim_whitespace`` keyword in :func:`read_csv` and :func:`read_table`, use ``sep=r"\s+"`` instead (:issue:`55569`)
+- Require :meth:`SparseDtype.fill_value` to be a valid value for the :meth:`SparseDtype.subtype` (:issue:`53043`)
+- Stopped automatically casting non-datetimelike values (mainly strings) in :meth:`Series.isin` and :meth:`Index.isin` with ``datetime64``, ``timedelta64``, and :class:`PeriodDtype` dtypes (:issue:`53111`)
+- Stopped performing dtype inference in :class:`Index`, :class:`Series` and :class:`DataFrame` constructors when given a pandas object (:class:`Series`, :class:`Index`, :class:`ExtensionArray`), call ``.infer_objects`` on the input to keep the current behavior (:issue:`56012`)
+- Stopped performing dtype inference when setting a :class:`Index` into a :class:`DataFrame` (:issue:`56102`)
+- Stopped performing dtype inference with in :meth:`Index.insert` with object-dtype index; this often affects the index/columns that result when setting new entries into an empty :class:`Series` or :class:`DataFrame` (:issue:`51363`)
+- Removed the "closed" and "unit" keywords in :meth:`TimedeltaIndex.__new__` (:issue:`52628`, :issue:`55499`)
+- All arguments in :meth:`Index.sort_values` are now keyword only (:issue:`56493`)
+- All arguments in :meth:`Series.to_dict` are now keyword only (:issue:`56493`)
+- Changed the default value of ``na_action`` in :meth:`Categorical.map` to ``None`` (:issue:`51645`)
+- Changed the default value of ``observed`` in :meth:`DataFrame.groupby` and :meth:`Series.groupby` to ``True`` (:issue:`51811`)
+- Enforce deprecation in :func:`testing.assert_series_equal` and :func:`testing.assert_frame_equal` with object dtype and mismatched null-like values, which are now considered not-equal (:issue:`18463`)
+- Enforce banning of upcasting in in-place setitem-like operations (:issue:`59007`) (see `PDEP6 <https://pandas.pydata.org/pdeps/0006-ban-upcasting.html>`_)
+- Enforced deprecation ``all`` and ``any`` reductions with ``datetime64``, :class:`DatetimeTZDtype`, and :class:`PeriodDtype` dtypes (:issue:`58029`)
+- Enforced deprecation disallowing ``float`` "periods" in :func:`date_range`, :func:`period_range`, :func:`timedelta_range`, :func:`interval_range`,  (:issue:`56036`)
+- Enforced deprecation disallowing parsing datetimes with mixed time zones unless user passes ``utc=True`` to :func:`to_datetime` (:issue:`57275`)
+- Enforced deprecation in :meth:`Series.value_counts` and :meth:`Index.value_counts` with object dtype performing dtype inference on the ``.index`` of the result (:issue:`56161`)
+- Enforced deprecation of :meth:`.DataFrameGroupBy.get_group` and :meth:`.SeriesGroupBy.get_group` allowing the ``name`` argument to be a non-tuple when grouping by a list of length 1 (:issue:`54155`)
+- Enforced deprecation of :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` for object-dtype (:issue:`57820`)
+- Enforced deprecation of :meth:`offsets.Tick.delta`, use ``pd.Timedelta(obj)`` instead (:issue:`55498`)
+- Enforced deprecation of ``axis=None`` acting the same as ``axis=0`` in the DataFrame reductions ``sum``, ``prod``, ``std``, ``var``, and ``sem``, passing ``axis=None`` will now reduce over both axes; this is particularly the case when doing e.g. ``numpy.sum(df)`` (:issue:`21597`)
+- Enforced deprecation of ``core.internals`` member ``DatetimeTZBlock`` (:issue:`58467`)
+- Enforced deprecation of ``date_parser`` in :func:`read_csv`, :func:`read_table`, :func:`read_fwf`, and :func:`read_excel` in favour of ``date_format`` (:issue:`50601`)
+- Enforced deprecation of ``keep_date_col`` keyword in :func:`read_csv` (:issue:`55569`)
+- Enforced deprecation of ``quantile`` keyword in :meth:`.Rolling.quantile` and :meth:`.Expanding.quantile`, renamed to ``q`` instead. (:issue:`52550`)
+- Enforced deprecation of argument ``infer_datetime_format`` in :func:`read_csv`, as a strict version of it is now the default (:issue:`48621`)
+- Enforced deprecation of combining parsed datetime columns in :func:`read_csv` in ``parse_dates`` (:issue:`55569`)
+- Enforced deprecation of non-standard (``np.ndarray``, :class:`ExtensionArray`, :class:`Index`, or :class:`Series`) argument to :func:`api.extensions.take` (:issue:`52981`)
+- Enforced deprecation of parsing system timezone strings to ``tzlocal``, which depended on system timezone, pass the 'tz' keyword instead (:issue:`50791`)
+- Enforced deprecation of passing a dictionary to :meth:`SeriesGroupBy.agg` (:issue:`52268`)
+- Enforced deprecation of string ``AS`` denoting frequency in :class:`YearBegin` and strings ``AS-DEC``, ``AS-JAN``, etc. denoting annual frequencies with various fiscal year starts (:issue:`57793`)
+- Enforced deprecation of string ``A`` denoting frequency in :class:`YearEnd` and strings ``A-DEC``, ``A-JAN``, etc. denoting annual frequencies with various fiscal year ends (:issue:`57699`)
+- Enforced deprecation of string ``BAS`` denoting frequency in :class:`BYearBegin` and strings ``BAS-DEC``, ``BAS-JAN``, etc. denoting annual frequencies with various fiscal year starts (:issue:`57793`)
+- Enforced deprecation of string ``BA`` denoting frequency in :class:`BYearEnd` and strings ``BA-DEC``, ``BA-JAN``, etc. denoting annual frequencies with various fiscal year ends (:issue:`57793`)
+- Enforced deprecation of strings ``H``, ``BH``, and ``CBH`` denoting frequencies in :class:`Hour`, :class:`BusinessHour`, :class:`CustomBusinessHour` (:issue:`59143`)
+- Enforced deprecation of strings ``H``, ``BH``, and ``CBH`` denoting units in :class:`Timedelta` (:issue:`59143`)
+- Enforced deprecation of strings ``T``, ``L``, ``U``, and ``N`` denoting frequencies in :class:`Minute`, :class:`Milli`, :class:`Micro`, :class:`Nano` (:issue:`57627`)
+- Enforced deprecation of strings ``T``, ``L``, ``U``, and ``N`` denoting units in :class:`Timedelta` (:issue:`57627`)
+- Enforced deprecation of the behavior of :func:`concat` when ``len(keys) != len(objs)`` would truncate to the shorter of the two. Now this raises a ``ValueError`` (:issue:`43485`)
+- Enforced deprecation of the behavior of :meth:`DataFrame.replace` and :meth:`Series.replace` with :class:`CategoricalDtype` that would introduce new categories. (:issue:`58270`)
+- Enforced deprecation of the behavior of :meth:`Series.argsort` in the presence of NA values (:issue:`58232`)
+- Enforced deprecation of values "pad", "ffill", "bfill", and "backfill" for :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` (:issue:`57869`)
+- Enforced deprecation removing :meth:`Categorical.to_list`, use ``obj.tolist()`` instead (:issue:`51254`)
+- Enforced silent-downcasting deprecation for :ref:`all relevant methods <whatsnew_220.silent_downcasting>` (:issue:`54710`)
+- In :meth:`DataFrame.stack`, the default value of ``future_stack`` is now ``True``; specifying ``False`` will raise a ``FutureWarning`` (:issue:`55448`)
+- Iterating over a :class:`.DataFrameGroupBy` or :class:`.SeriesGroupBy` will return tuples of length 1 for the groups when grouping by ``level`` a list of length 1 (:issue:`50064`)
+- Methods ``apply``, ``agg``, and ``transform`` will no longer replace NumPy functions (e.g. ``np.sum``) and built-in functions (e.g. ``min``) with the equivalent pandas implementation; use string aliases (e.g. ``"sum"`` and ``"min"``) if you desire to use the pandas implementation (:issue:`53974`)
+- Passing both ``freq`` and ``fill_value`` in :meth:`DataFrame.shift` and :meth:`Series.shift` and :meth:`.DataFrameGroupBy.shift` now raises a ``ValueError`` (:issue:`54818`)
+- Removed :meth:`.DataFrameGroupBy.quantile` and :meth:`.SeriesGroupBy.quantile` supporting bool dtype (:issue:`53975`)
+- Removed :meth:`DateOffset.is_anchored` and :meth:`offsets.Tick.is_anchored` (:issue:`56594`)
+- Removed ``DataFrame.applymap``, ``Styler.applymap`` and ``Styler.applymap_index`` (:issue:`52364`)
+- Removed ``DataFrame.bool`` and ``Series.bool`` (:issue:`51756`)
+- Removed ``DataFrame.first`` and ``DataFrame.last`` (:issue:`53710`)
+- Removed ``DataFrame.swapaxes`` and ``Series.swapaxes`` (:issue:`51946`)
+- Removed ``DataFrameGroupBy.grouper`` and ``SeriesGroupBy.grouper`` (:issue:`56521`)
+- Removed ``DataFrameGroupby.fillna`` and ``SeriesGroupBy.fillna``` (:issue:`55719`)
+- Removed ``Index.format``, use :meth:`Index.astype` with ``str`` or :meth:`Index.map` with a ``formatter`` function instead (:issue:`55439`)
+- Removed ``Resample.fillna`` (:issue:`55719`)
+- Removed ``Series.__int__`` and ``Series.__float__``. Call ``int(Series.iloc[0])`` or ``float(Series.iloc[0])`` instead. (:issue:`51131`)
+- Removed ``Series.ravel`` (:issue:`56053`)
+- Removed ``Series.view`` (:issue:`56054`)
+- Removed ``StataReader.close`` (:issue:`49228`)
+- Removed ``_data`` from :class:`DataFrame`, :class:`Series`, :class:`.arrays.ArrowExtensionArray` (:issue:`52003`)
+- Removed ``axis`` argument from :meth:`DataFrame.groupby`, :meth:`Series.groupby`, :meth:`DataFrame.rolling`, :meth:`Series.rolling`, :meth:`DataFrame.resample`, and :meth:`Series.resample` (:issue:`51203`)
+- Removed ``axis`` argument from all groupby operations (:issue:`50405`)
+- Removed ``convert_dtype`` from :meth:`Series.apply` (:issue:`52257`)
+- Removed ``method``, ``limit`` ``fill_axis`` and ``broadcast_axis`` keywords from :meth:`DataFrame.align` (:issue:`51968`)
+- Removed ``pandas.api.types.is_interval`` and ``pandas.api.types.is_period``, use ``isinstance(obj, pd.Interval)`` and ``isinstance(obj, pd.Period)`` instead (:issue:`55264`)
+- Removed ``pandas.io.sql.execute`` (:issue:`50185`)
+- Removed ``pandas.value_counts``, use :meth:`Series.value_counts` instead (:issue:`53493`)
+- Removed ``read_gbq`` and ``DataFrame.to_gbq``. Use ``pandas_gbq.read_gbq`` and ``pandas_gbq.to_gbq`` instead https://pandas-gbq.readthedocs.io/en/latest/api.html (:issue:`55525`)
+- Removed ``use_nullable_dtypes`` from :func:`read_parquet` (:issue:`51853`)
+- Removed ``year``, ``month``, ``quarter``, ``day``, ``hour``, ``minute``, and ``second`` keywords in the :class:`PeriodIndex` constructor, use :meth:`PeriodIndex.from_fields` instead (:issue:`55960`)
+- Removed argument ``limit`` from :meth:`DataFrame.pct_change`, :meth:`Series.pct_change`, :meth:`.DataFrameGroupBy.pct_change`, and :meth:`.SeriesGroupBy.pct_change`; the argument ``method`` must be set to ``None`` and will be removed in a future version of pandas (:issue:`53520`)
+- Removed deprecated argument ``obj`` in :meth:`.DataFrameGroupBy.get_group` and :meth:`.SeriesGroupBy.get_group` (:issue:`53545`)
+- Removed deprecated behavior of :meth:`Series.agg` using :meth:`Series.apply` (:issue:`53325`)
+- Removed deprecated keyword ``method`` on :meth:`Series.fillna`, :meth:`DataFrame.fillna` (:issue:`57760`)
+- Removed option ``mode.use_inf_as_na``, convert inf entries to ``NaN`` before instead (:issue:`51684`)
+- Removed support for :class:`DataFrame` in :meth:`DataFrame.from_records`(:issue:`51697`)
+- Removed support for ``errors="ignore"`` in :func:`to_datetime`, :func:`to_timedelta` and :func:`to_numeric` (:issue:`55734`)
+- Removed support for ``slice`` in :meth:`DataFrame.take` (:issue:`51539`)
+- Removed the ``ArrayManager`` (:issue:`55043`)
+- Removed the ``fastpath`` argument from the :class:`Series` constructor (:issue:`55466`)
+- Removed the ``is_boolean``, ``is_integer``, ``is_floating``, ``holds_integer``, ``is_numeric``, ``is_categorical``, ``is_object``, and ``is_interval`` attributes of :class:`Index` (:issue:`50042`)
+- Removed the ``ordinal`` keyword in :class:`PeriodIndex`, use :meth:`PeriodIndex.from_ordinals` instead (:issue:`55960`)
+- Removed unused arguments ``*args`` and ``**kwargs`` in :class:`Resampler` methods (:issue:`50977`)
+- Unrecognized timezones when parsing strings to datetimes now raises a ``ValueError`` (:issue:`51477`)
+- Removed the :class:`Grouper` attributes ``ax``, ``groups``, ``indexer``, and ``obj`` (:issue:`51206`, :issue:`51182`)
+- Removed deprecated keyword ``verbose`` on :func:`read_csv` and :func:`read_table` (:issue:`56556`)
+- Removed the ``method`` keyword in ``ExtensionArray.fillna``, implement ``ExtensionArray._pad_or_backfill`` instead (:issue:`53621`)
+- Removed the attribute ``dtypes`` from :class:`.DataFrameGroupBy` (:issue:`51997`)
+- Enforced deprecation of ``argmin``, ``argmax``, ``idxmin``, and ``idxmax`` returning a result when ``skipna=False`` and an NA value is encountered or all values are NA values; these operations will now raise in such cases (:issue:`33941`, :issue:`51276`)
+- Removed specifying ``include_groups=True`` in :class:`.DataFrameGroupBy.apply` and :class:`.Resampler.apply` (:issue:`7155`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.performance:
+
+Performance improvements
+~~~~~~~~~~~~~~~~~~~~~~~~
+- Eliminated circular reference in to original pandas object in accessor attributes (e.g. :attr:`Series.str`). However, accessor instantiation is no longer cached (:issue:`47667`, :issue:`41357`)
+- :attr:`Categorical.categories` returns a :class:`RangeIndex` columns instead of an :class:`Index` if the constructed ``values`` was a ``range``. (:issue:`57787`)
+- :class:`DataFrame` returns a :class:`RangeIndex` columns when possible when ``data`` is a ``dict`` (:issue:`57943`)
+- :class:`Series` returns a :class:`RangeIndex` index when possible when ``data`` is a ``dict`` (:issue:`58118`)
+- :func:`concat` returns a :class:`RangeIndex` column when possible when ``objs`` contains :class:`Series` and :class:`DataFrame` and ``axis=0`` (:issue:`58119`)
+- :func:`concat` returns a :class:`RangeIndex` level in the :class:`MultiIndex` result when ``keys`` is a ``range`` or :class:`RangeIndex` (:issue:`57542`)
+- :meth:`RangeIndex.append` returns a :class:`RangeIndex` instead of a :class:`Index` when appending values that could continue the :class:`RangeIndex` (:issue:`57467`)
+- :meth:`Series.str.extract` returns a :class:`RangeIndex` columns instead of an :class:`Index` column when possible (:issue:`57542`)
+- :meth:`Series.str.partition` with :class:`ArrowDtype` returns a :class:`RangeIndex` columns instead of an :class:`Index` column when possible (:issue:`57768`)
+- Performance improvement in :class:`DataFrame` when ``data`` is a ``dict`` and ``columns`` is specified (:issue:`24368`)
+- Performance improvement in :class:`MultiIndex` when setting :attr:`MultiIndex.names` doesn't invalidate all cached operations (:issue:`59578`)
+- Performance improvement in :meth:`DataFrame.join` for sorted but non-unique indexes (:issue:`56941`)
+- Performance improvement in :meth:`DataFrame.join` when left and/or right are non-unique and ``how`` is ``"left"``, ``"right"``, or ``"inner"`` (:issue:`56817`)
+- Performance improvement in :meth:`DataFrame.join` with ``how="left"`` or ``how="right"`` and ``sort=True`` (:issue:`56919`)
+- Performance improvement in :meth:`DataFrame.to_csv` when ``index=False`` (:issue:`59312`)
+- Performance improvement in :meth:`DataFrameGroupBy.ffill`, :meth:`DataFrameGroupBy.bfill`, :meth:`SeriesGroupBy.ffill`, and :meth:`SeriesGroupBy.bfill` (:issue:`56902`)
+- Performance improvement in :meth:`Index.join` by propagating cached attributes in cases where the result matches one of the inputs (:issue:`57023`)
+- Performance improvement in :meth:`Index.take` when ``indices`` is a full range indexer from zero to length of index (:issue:`56806`)
+- Performance improvement in :meth:`Index.to_frame` returning a :class:`RangeIndex` columns of a :class:`Index` when possible. (:issue:`58018`)
+- Performance improvement in :meth:`MultiIndex._engine` to use smaller dtypes if possible (:issue:`58411`)
+- Performance improvement in :meth:`MultiIndex.equals` for equal length indexes (:issue:`56990`)
+- Performance improvement in :meth:`MultiIndex.memory_usage` to ignore the index engine when it isn't already cached. (:issue:`58385`)
+- Performance improvement in :meth:`RangeIndex.__getitem__` with a boolean mask or integers returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57588`)
+- Performance improvement in :meth:`RangeIndex.append` when appending the same index (:issue:`57252`)
+- Performance improvement in :meth:`RangeIndex.argmin` and :meth:`RangeIndex.argmax` (:issue:`57823`)
+- Performance improvement in :meth:`RangeIndex.insert` returning a :class:`RangeIndex` instead of a :class:`Index` when the :class:`RangeIndex` is empty. (:issue:`57833`)
+- Performance improvement in :meth:`RangeIndex.round` returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57824`)
+- Performance improvement in :meth:`RangeIndex.searchsorted` (:issue:`58376`)
+- Performance improvement in :meth:`RangeIndex.to_numpy` when specifying an ``na_value`` (:issue:`58376`)
+- Performance improvement in :meth:`RangeIndex.value_counts` (:issue:`58376`)
+- Performance improvement in :meth:`RangeIndex.join` returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57651`, :issue:`57752`)
+- Performance improvement in :meth:`RangeIndex.reindex` returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57647`, :issue:`57752`)
+- Performance improvement in :meth:`RangeIndex.take` returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57445`, :issue:`57752`)
+- Performance improvement in :func:`merge` if hash-join can be used (:issue:`57970`)
+- Performance improvement in :meth:`CategoricalDtype.update_dtype` when ``dtype`` is a :class:`CategoricalDtype` with non ``None`` categories and ordered (:issue:`59647`)
+- Performance improvement in :meth:`DataFrame.astype` when converting to extension floating dtypes, e.g. "Float64" (:issue:`60066`)
+- Performance improvement in :meth:`to_hdf` avoid unnecessary reopenings of the HDF5 file to speedup data addition to files with a very large number of groups . (:issue:`58248`)
+- Performance improvement in ``DataFrameGroupBy.__len__`` and ``SeriesGroupBy.__len__`` (:issue:`57595`)
+- Performance improvement in indexing operations for string dtypes (:issue:`56997`)
+- Performance improvement in unary methods on a :class:`RangeIndex` returning a :class:`RangeIndex` instead of a :class:`Index` when possible. (:issue:`57825`)
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.bug_fixes:
+
+Bug fixes
+~~~~~~~~~
+
+Categorical
+^^^^^^^^^^^
+- Bug in :func:`Series.apply` where ``nan`` was ignored for :class:`CategoricalDtype` (:issue:`59938`)
+-
+
+Datetimelike
+^^^^^^^^^^^^
+- Bug in :attr:`is_year_start` where a DateTimeIndex constructed via a date_range with frequency 'MS' wouldn't have the correct year or quarter start attributes (:issue:`57377`)
+- Bug in :class:`DataFrame` raising ``ValueError`` when ``dtype`` is ``timedelta64`` and ``data`` is a list containing ``None`` (:issue:`60064`)
+- Bug in :class:`Timestamp` constructor failing to raise when ``tz=None`` is explicitly specified in conjunction with timezone-aware ``tzinfo`` or data (:issue:`48688`)
+- Bug in :func:`date_range` where the last valid timestamp would sometimes not be produced (:issue:`56134`)
+- Bug in :func:`date_range` where using a negative frequency value would not include all points between the start and end values (:issue:`56147`)
+- Bug in :func:`tseries.api.guess_datetime_format` would fail to infer time format when "%Y" == "%H%M" (:issue:`57452`)
+- Bug in :func:`tseries.frequencies.to_offset` would fail to parse frequency strings starting with "LWOM" (:issue:`59218`)
+- Bug in :meth:`DataFrame.min` and :meth:`DataFrame.max` casting ``datetime64`` and ``timedelta64`` columns to ``float64`` and losing precision (:issue:`60850`)
+- Bug in :meth:`Dataframe.agg` with df with missing values resulting in IndexError (:issue:`58810`)
+- Bug in :meth:`DatetimeIndex.is_year_start` and :meth:`DatetimeIndex.is_quarter_start` does not raise on Custom business days frequencies bigger then "1C" (:issue:`58664`)
+- Bug in :meth:`DatetimeIndex.is_year_start` and :meth:`DatetimeIndex.is_quarter_start` returning ``False`` on double-digit frequencies (:issue:`58523`)
+- Bug in :meth:`DatetimeIndex.union` and :meth:`DatetimeIndex.intersection` when ``unit`` was non-nanosecond (:issue:`59036`)
+- Bug in :meth:`Series.dt.microsecond` producing incorrect results for pyarrow backed :class:`Series`. (:issue:`59154`)
+- Bug in :meth:`to_datetime` not respecting dayfirst if an uncommon date string was passed. (:issue:`58859`)
+- Bug in :meth:`to_datetime` on float32 df with year, month, day etc. columns leads to precision issues and incorrect result. (:issue:`60506`)
+- Bug in :meth:`to_datetime` reports incorrect index in case of any failure scenario. (:issue:`58298`)
+- Bug in :meth:`to_datetime` wrongly converts when ``arg`` is a ``np.datetime64`` object with unit of ``ps``. (:issue:`60341`)
+- Bug in setting scalar values with mismatched resolution into arrays with non-nanosecond ``datetime64``, ``timedelta64`` or :class:`DatetimeTZDtype` incorrectly truncating those scalars (:issue:`56410`)
+
+Timedelta
+^^^^^^^^^
+- Accuracy improvement in :meth:`Timedelta.to_pytimedelta` to round microseconds consistently for large nanosecond based Timedelta (:issue:`57841`)
+- Bug in :meth:`DataFrame.cumsum` which was raising ``IndexError`` if dtype is ``timedelta64[ns]`` (:issue:`57956`)
+
+Timezones
+^^^^^^^^^
+-
+-
+
+Numeric
+^^^^^^^
+- Bug in :meth:`DataFrame.quantile` where the column type was not preserved when ``numeric_only=True`` with a list-like ``q`` produced an empty result (:issue:`59035`)
+- Bug in ``np.matmul`` with :class:`Index` inputs raising a ``TypeError`` (:issue:`57079`)
+
+Conversion
+^^^^^^^^^^
+- Bug in :meth:`DataFrame.astype` not casting ``values`` for Arrow-based dictionary dtype correctly (:issue:`58479`)
+- Bug in :meth:`DataFrame.update` bool dtype being converted to object (:issue:`55509`)
+- Bug in :meth:`Series.astype` might modify read-only array inplace when casting to a string dtype (:issue:`57212`)
+- Bug in :meth:`Series.convert_dtypes` and :meth:`DataFrame.convert_dtypes` removing timezone information for objects with :class:`ArrowDtype` (:issue:`60237`)
+- Bug in :meth:`Series.reindex` not maintaining ``float32`` type when a ``reindex`` introduces a missing value (:issue:`45857`)
+
+Strings
+^^^^^^^
+- Bug in :meth:`Series.value_counts` would not respect ``sort=False`` for series having ``string`` dtype (:issue:`55224`)
+-
+
+Interval
+^^^^^^^^
+- :meth:`Index.is_monotonic_decreasing`, :meth:`Index.is_monotonic_increasing`, and :meth:`Index.is_unique` could incorrectly be ``False`` for an ``Index`` created from a slice of another ``Index``. (:issue:`57911`)
+- Bug in :func:`interval_range` where start and end numeric types were always cast to 64 bit (:issue:`57268`)
+-
+
+Indexing
+^^^^^^^^
+- Bug in :meth:`DataFrame.__getitem__` returning modified columns when called with ``slice`` in Python 3.12 (:issue:`57500`)
+- Bug in :meth:`DataFrame.from_records` throwing a ``ValueError`` when passed an empty list in ``index`` (:issue:`58594`)
+- Bug in :meth:`MultiIndex.insert` when a new value inserted to a datetime-like level gets cast to ``NaT`` and fails indexing (:issue:`60388`)
+- Bug in printing :attr:`Index.names` and :attr:`MultiIndex.levels` would not escape single quotes (:issue:`60190`)
+
+Missing
+^^^^^^^
+- Bug in :meth:`DataFrame.fillna` and :meth:`Series.fillna` that would ignore the ``limit`` argument on :class:`.ExtensionArray` dtypes (:issue:`58001`)
+-
+
+MultiIndex
+^^^^^^^^^^
+- :func:`DataFrame.loc` with ``axis=0``  and :class:`MultiIndex` when setting a value adds extra columns (:issue:`58116`)
+- :meth:`DataFrame.melt` would not accept multiple names in ``var_name`` when the columns were a :class:`MultiIndex` (:issue:`58033`)
+- :meth:`MultiIndex.insert` would not insert NA value correctly at unified location of index -1 (:issue:`59003`)
+- :func:`MultiIndex.get_level_values` accessing a :class:`DatetimeIndex` does not carry the frequency attribute along (:issue:`58327`, :issue:`57949`)
+- Bug in :class:`DataFrame` arithmetic operations in case of unaligned MultiIndex columns (:issue:`60498`)
+-
+
+I/O
+^^^
+- Bug in :class:`DataFrame` and :class:`Series` ``repr`` of :py:class:`collections.abc.Mapping`` elements. (:issue:`57915`)
+- Bug in :meth:`.DataFrame.to_json` when ``"index"`` was a value in the :attr:`DataFrame.column` and :attr:`Index.name` was ``None``. Now, this will fail with a ``ValueError`` (:issue:`58925`)
+- Bug in :meth:`DataFrame._repr_html_` which ignored the ``"display.float_format"`` option (:issue:`59876`)
+- Bug in :meth:`DataFrame.from_records` where ``columns`` parameter with numpy structured array was not reordering and filtering out the columns (:issue:`59717`)
+- Bug in :meth:`DataFrame.to_dict` raises unnecessary ``UserWarning`` when columns are not unique and ``orient='tight'``. (:issue:`58281`)
+- Bug in :meth:`DataFrame.to_excel` when writing empty :class:`DataFrame` with :class:`MultiIndex` on both axes (:issue:`57696`)
+- Bug in :meth:`DataFrame.to_excel` where the :class:`MultiIndex` index with a period level was not a date (:issue:`60099`)
+- Bug in :meth:`DataFrame.to_stata` when writing :class:`DataFrame` and ``byteorder=`big```. (:issue:`58969`)
+- Bug in :meth:`DataFrame.to_stata` when writing more than 32,000 value labels. (:issue:`60107`)
+- Bug in :meth:`DataFrame.to_string` that raised ``StopIteration`` with nested DataFrames. (:issue:`16098`)
+- Bug in :meth:`HDFStore.get` was failing to save data of dtype datetime64[s] correctly (:issue:`59004`)
+- Bug in :meth:`read_csv` causing segmentation fault when ``encoding_errors`` is not a string. (:issue:`59059`)
+- Bug in :meth:`read_csv` raising ``TypeError`` when ``index_col`` is specified and ``na_values`` is a dict containing the key ``None``. (:issue:`57547`)
+- Bug in :meth:`read_csv` raising ``TypeError`` when ``nrows`` and ``iterator`` are specified without specifying a ``chunksize``. (:issue:`59079`)
+- Bug in :meth:`read_csv` where the order of the ``na_values`` makes an inconsistency when ``na_values`` is a list non-string values. (:issue:`59303`)
+- Bug in :meth:`read_excel` raising ``ValueError`` when passing array of boolean values when ``dtype="boolean"``. (:issue:`58159`)
+- Bug in :meth:`read_html` where ``rowspan`` in header row causes incorrect conversion to ``DataFrame``. (:issue:`60210`)
+- Bug in :meth:`read_json` not validating the ``typ`` argument to not be exactly ``"frame"`` or ``"series"`` (:issue:`59124`)
+- Bug in :meth:`read_json` where extreme value integers in string format were incorrectly parsed as a different integer number (:issue:`20608`)
+- Bug in :meth:`read_stata` raising ``KeyError`` when input file is stored in big-endian format and contains strL data. (:issue:`58638`)
+- Bug in :meth:`read_stata` where extreme value integers were incorrectly interpreted as missing for format versions 111 and prior (:issue:`58130`)
+- Bug in :meth:`read_stata` where the missing code for double was not recognised for format versions 105 and prior (:issue:`58149`)
+- Bug in :meth:`set_option` where setting the pandas option ``display.html.use_mathjax`` to ``False`` has no effect (:issue:`59884`)
+- Bug in :meth:`to_excel` where :class:`MultiIndex` columns would be merged to a single row when ``merge_cells=False`` is passed (:issue:`60274`)
+
+Period
+^^^^^^
+- Fixed error message when passing invalid period alias to :meth:`PeriodIndex.to_timestamp` (:issue:`58974`)
+-
+
+Plotting
+^^^^^^^^
+- Bug in :meth:`.DataFrameGroupBy.boxplot` failed when there were multiple groupings (:issue:`14701`)
+- Bug in :meth:`DataFrame.plot.bar` with ``stacked=True`` where labels on stacked bars with zero-height segments were incorrectly positioned at the base instead of the label position of the previous segment (:issue:`59429`)
+- Bug in :meth:`DataFrame.plot.line` raising ``ValueError`` when set both color and a ``dict`` style (:issue:`59461`)
+- Bug in :meth:`DataFrame.plot` that causes a shift to the right when the frequency multiplier is greater than one. (:issue:`57587`)
+- Bug in :meth:`Series.plot` with ``kind="pie"`` with :class:`ArrowDtype` (:issue:`59192`)
+
+Groupby/resample/rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+- Bug in :meth:`.DataFrameGroupBy.__len__` and :meth:`.SeriesGroupBy.__len__` would raise when the grouping contained NA values and ``dropna=False`` (:issue:`58644`)
+- Bug in :meth:`.DataFrameGroupBy.any` that returned True for groups where all Timedelta values are NaT. (:issue:`59712`)
+- Bug in :meth:`.DataFrameGroupBy.groups` and :meth:`.SeriesGroupby.groups` that would not respect groupby argument ``dropna`` (:issue:`55919`)
+- Bug in :meth:`.DataFrameGroupBy.median` where nat values gave an incorrect result. (:issue:`57926`)
+- Bug in :meth:`.DataFrameGroupBy.quantile` when ``interpolation="nearest"`` is inconsistent with :meth:`DataFrame.quantile` (:issue:`47942`)
+- Bug in :meth:`.Resampler.interpolate` on a :class:`DataFrame` with non-uniform sampling and/or indices not aligning with the resulting resampled index would result in wrong interpolation (:issue:`21351`)
+- Bug in :meth:`DataFrame.ewm` and :meth:`Series.ewm` when passed ``times`` and aggregation functions other than mean (:issue:`51695`)
+- Bug in :meth:`DataFrameGroupBy.agg` that raises ``AttributeError`` when there is dictionary input and duplicated columns, instead of returning a DataFrame with the aggregation of all duplicate columns. (:issue:`55041`)
+- Bug in :meth:`DataFrameGroupBy.apply` and :meth:`SeriesGroupBy.apply` for empty data frame with ``group_keys=False`` still creating output index using group keys. (:issue:`60471`)
+- Bug in :meth:`DataFrameGroupBy.apply` that was returning a completely empty DataFrame when all return values of ``func`` were ``None`` instead of returning an empty DataFrame with the original columns and dtypes. (:issue:`57775`)
+- Bug in :meth:`DataFrameGroupBy.apply` with ``as_index=False`` that was returning :class:`MultiIndex` instead of returning :class:`Index`. (:issue:`58291`)
+- Bug in :meth:`DataFrameGroupBy.cumsum` and :meth:`DataFrameGroupBy.cumprod` where ``numeric_only`` parameter was passed indirectly through kwargs instead of passing directly. (:issue:`58811`)
+- Bug in :meth:`DataFrameGroupBy.cumsum` where it did not return the correct dtype when the label contained ``None``. (:issue:`58811`)
+- Bug in :meth:`DataFrameGroupby.transform` and :meth:`SeriesGroupby.transform` with a reducer and ``observed=False`` that coerces dtype to float when there are unobserved categories. (:issue:`55326`)
+- Bug in :meth:`Rolling.apply` for ``method="table"`` where column order was not being respected due to the columns getting sorted by default. (:issue:`59666`)
+- Bug in :meth:`Rolling.apply` where the applied function could be called on fewer than ``min_period`` periods if ``method="table"``. (:issue:`58868`)
+- Bug in :meth:`Series.resample` could raise when the the date range ended shortly before a non-existent time. (:issue:`58380`)
+
+Reshaping
+^^^^^^^^^
+- Bug in :func:`qcut` where values at the quantile boundaries could be incorrectly assigned (:issue:`59355`)
+- Bug in :meth:`DataFrame.combine_first` not preserving the column order (:issue:`60427`)
+- Bug in :meth:`DataFrame.join` inconsistently setting result index name (:issue:`55815`)
+- Bug in :meth:`DataFrame.join` when a :class:`DataFrame` with a :class:`MultiIndex` would raise an ``AssertionError`` when :attr:`MultiIndex.names` contained ``None``. (:issue:`58721`)
+- Bug in :meth:`DataFrame.merge` where merging on a column containing only ``NaN`` values resulted in an out-of-bounds array access (:issue:`59421`)
+- Bug in :meth:`DataFrame.unstack` producing incorrect results when ``sort=False`` (:issue:`54987`, :issue:`55516`)
+- Bug in :meth:`DataFrame.merge` when merging two :class:`DataFrame` on ``intc`` or ``uintc`` types on Windows (:issue:`60091`, :issue:`58713`)
+- Bug in :meth:`DataFrame.pivot_table` incorrectly subaggregating results when called without an ``index`` argument (:issue:`58722`)
+- Bug in :meth:`DataFrame.stack` with the new implementation where ``ValueError`` is raised when ``level=[]`` (:issue:`60740`)
+- Bug in :meth:`DataFrame.unstack` producing incorrect results when manipulating empty :class:`DataFrame` with an :class:`ExtentionDtype` (:issue:`59123`)
+
+Sparse
+^^^^^^
+- Bug in :class:`SparseDtype` for equal comparison with na fill value. (:issue:`54770`)
+- Bug in :meth:`DataFrame.sparse.from_spmatrix` which hard coded an invalid ``fill_value`` for certain subtypes. (:issue:`59063`)
+- Bug in :meth:`DataFrame.sparse.to_dense` which ignored subclassing and always returned an instance of :class:`DataFrame` (:issue:`59913`)
+
+ExtensionArray
+^^^^^^^^^^^^^^
+- Bug in :class:`Categorical` when constructing with an :class:`Index` with :class:`ArrowDtype` (:issue:`60563`)
+- Bug in :meth:`.arrays.ArrowExtensionArray.__setitem__` which caused wrong behavior when using an integer array with repeated values as a key (:issue:`58530`)
+- Bug in :meth:`ArrowExtensionArray.factorize` where NA values were dropped when input was dictionary-encoded even when dropna was set to False(:issue:`60567`)
+- Bug in :meth:`api.types.is_datetime64_any_dtype` where a custom :class:`ExtensionDtype` would return ``False`` for array-likes (:issue:`57055`)
+- Bug in comparison between object with :class:`ArrowDtype` and incompatible-dtyped (e.g. string vs bool) incorrectly raising instead of returning all-``False`` (for ``==``) or all-``True`` (for ``!=``) (:issue:`59505`)
+- Bug in constructing pandas data structures when passing into ``dtype`` a string of the type followed by ``[pyarrow]`` while PyArrow is not installed would raise ``NameError`` rather than ``ImportError`` (:issue:`57928`)
+- Bug in various :class:`DataFrame` reductions for pyarrow temporal dtypes returning incorrect dtype when result was null (:issue:`59234`)
+
+Styler
+^^^^^^
+- Bug in :meth:`Styler.to_latex` where styling column headers when combined with a hidden index or hidden index-levels is fixed.
+
+Other
+^^^^^
+- Bug in :class:`DataFrame` when passing a ``dict`` with a NA scalar and ``columns`` that would always return ``np.nan`` (:issue:`57205`)
+- Bug in :class:`Series` ignoring errors when trying to convert :class:`Series` input data to the given ``dtype`` (:issue:`60728`)
+- Bug in :func:`eval` on :class:`ExtensionArray` on including division ``/`` failed with a ``TypeError``. (:issue:`58748`)
+- Bug in :func:`eval` where the names of the :class:`Series` were not preserved when using ``engine="numexpr"``. (:issue:`10239`)
+- Bug in :func:`eval` with ``engine="numexpr"`` returning unexpected result for float division. (:issue:`59736`)
+- Bug in :func:`to_numeric` raising ``TypeError`` when ``arg`` is a :class:`Timedelta` or :class:`Timestamp` scalar. (:issue:`59944`)
+- Bug in :func:`unique` on :class:`Index` not always returning :class:`Index` (:issue:`57043`)
+- Bug in :meth:`DataFrame.apply` where passing ``engine="numba"`` ignored ``args`` passed to the applied function (:issue:`58712`)
+- Bug in :meth:`DataFrame.eval` and :meth:`DataFrame.query` which caused an exception when using NumPy attributes via ``@`` notation, e.g., ``df.eval("@np.floor(a)")``. (:issue:`58041`)
+- Bug in :meth:`DataFrame.eval` and :meth:`DataFrame.query` which did not allow to use ``tan`` function. (:issue:`55091`)
+- Bug in :meth:`DataFrame.query` where using duplicate column names led to a ``TypeError``. (:issue:`59950`)
+- Bug in :meth:`DataFrame.query` which raised an exception or produced incorrect results when expressions contained backtick-quoted column names containing the hash character ``#``, backticks, or characters that fall outside the ASCII range (U+0001..U+007F). (:issue:`59285`) (:issue:`49633`)
+- Bug in :meth:`DataFrame.shift` where passing a ``freq`` on a DataFrame with no columns did not shift the index correctly. (:issue:`60102`)
+- Bug in :meth:`DataFrame.sort_index` when passing ``axis="columns"`` and ``ignore_index=True`` and ``ascending=False`` not returning a :class:`RangeIndex` columns (:issue:`57293`)
+- Bug in :meth:`DataFrame.transform` that was returning the wrong order unless the index was monotonically increasing. (:issue:`57069`)
+- Bug in :meth:`DataFrame.where` where using a non-bool type array in the function would return a ``ValueError`` instead of a ``TypeError`` (:issue:`56330`)
+- Bug in :meth:`Index.sort_values` when passing a key function that turns values into tuples, e.g. ``key=natsort.natsort_key``, would raise ``TypeError`` (:issue:`56081`)
+- Bug in :meth:`MultiIndex.fillna` error message was referring to ``isna`` instead of ``fillna`` (:issue:`60974`)
+- Bug in :meth:`Series.diff` allowing non-integer values for the ``periods`` argument. (:issue:`56607`)
+- Bug in :meth:`Series.dt` methods in :class:`ArrowDtype` that were returning incorrect values. (:issue:`57355`)
+- Bug in :meth:`Series.isin` raising ``TypeError`` when series is large (>10**6) and ``values`` contains NA (:issue:`60678`)
+- Bug in :meth:`Series.rank` that doesn't preserve missing values for nullable integers when ``na_option='keep'``. (:issue:`56976`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` inconsistently replacing matching instances when ``regex=True`` and missing values are present. (:issue:`56599`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` throwing ``ValueError`` when ``regex=True`` and all NA values. (:issue:`60688`)
+- Bug in :meth:`Series.to_string` when series contains complex floats with exponents (:issue:`60405`)
+- Bug in :meth:`read_csv` where chained fsspec TAR file and ``compression="infer"`` fails with ``tarfile.ReadError`` (:issue:`60028`)
+- Bug in Dataframe Interchange Protocol implementation was returning incorrect results for data buffers' associated dtype, for string and datetime columns (:issue:`54781`)
+- Bug in ``Series.list`` methods not preserving the original :class:`Index`. (:issue:`58425`)
+- Bug in ``Series.list`` methods not preserving the original name. (:issue:`60522`)
+- Bug in printing a :class:`DataFrame` with a :class:`DataFrame` stored in :attr:`DataFrame.attrs` raised a ``ValueError`` (:issue:`60455`)
+- Bug in printing a :class:`Series` with a :class:`DataFrame` stored in :attr:`Series.attrs` raised a ``ValueError`` (:issue:`60568`)
+- Fixed regression in :meth:`DataFrame.from_records` not initializing subclasses properly (:issue:`57008`)
+
+.. ***DO NOT USE THIS SECTION***
+
+-
+-
+
+.. ---------------------------------------------------------------------------
+.. _whatsnew_300.contributors:
+
+Contributors
+~~~~~~~~~~~~
diff --git a/pandas/doc/source/whatsnew/whatsnew_0171_html_table.html b/pandas/doc/source/whatsnew/whatsnew_0171_html_table.html
new file mode 100644
index 0000000000000..a76d6207bba58
--- /dev/null
+++ b/pandas/doc/source/whatsnew/whatsnew_0171_html_table.html
@@ -0,0 +1,872 @@
+
+        <style  type="text/css" >
+
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col0 {
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col1 {
+
+                background-color:  #89d548;
+
+                background-color:  #89d548;
+
+                background-color:  #89d548;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col2 {
+
+                background-color:  #3aba76;
+
+                background-color:  #3aba76;
+
+                background-color:  #3aba76;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col3 {
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col4 {
+
+                background-color:  #24878e;
+
+                background-color:  #24878e;
+
+                background-color:  #24878e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col0 {
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col1 {
+
+                background-color:  #2c738e;
+
+                background-color:  #2c738e;
+
+                background-color:  #2c738e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col2 {
+
+                background-color:  #1f9f88;
+
+                background-color:  #1f9f88;
+
+                background-color:  #1f9f88;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col3 {
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col4 {
+
+                background-color:  #297a8e;
+
+                background-color:  #297a8e;
+
+                background-color:  #297a8e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col0 {
+
+                background-color:  #27808e;
+
+                background-color:  #27808e;
+
+                background-color:  #27808e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col1 {
+
+                background-color:  #3bbb75;
+
+                background-color:  #3bbb75;
+
+                background-color:  #3bbb75;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col2 {
+
+                background-color:  #b2dd2d;
+
+                background-color:  #b2dd2d;
+
+                background-color:  #b2dd2d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col3 {
+
+                background-color:  #1f9a8a;
+
+                background-color:  #1f9a8a;
+
+                background-color:  #1f9a8a;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col4 {
+
+                background-color:  #228d8d;
+
+                background-color:  #228d8d;
+
+                background-color:  #228d8d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col0 {
+
+                background-color:  #218e8d;
+
+                background-color:  #218e8d;
+
+                background-color:  #218e8d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col1 {
+
+                background-color:  #efe51c;
+
+                background-color:  #efe51c;
+
+                background-color:  #efe51c;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col2 {
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col3 {
+
+                background-color:  #d5e21a;
+
+                background-color:  #d5e21a;
+
+                background-color:  #d5e21a;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col4 {
+
+                background-color:  #2eb37c;
+
+                background-color:  #2eb37c;
+
+                background-color:  #2eb37c;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col0 {
+
+                background-color:  #6ece58;
+
+                background-color:  #6ece58;
+
+                background-color:  #6ece58;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col1 {
+
+                background-color:  #b2dd2d;
+
+                background-color:  #b2dd2d;
+
+                background-color:  #b2dd2d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col2 {
+
+                background-color:  #238a8d;
+
+                background-color:  #238a8d;
+
+                background-color:  #238a8d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col3 {
+
+                background-color:  #f1e51d;
+
+                background-color:  #f1e51d;
+
+                background-color:  #f1e51d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col4 {
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col0 {
+
+                background-color:  #26828e;
+
+                background-color:  #26828e;
+
+                background-color:  #26828e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col1 {
+
+                background-color:  #81d34d;
+
+                background-color:  #81d34d;
+
+                background-color:  #81d34d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col2 {
+
+                background-color:  #2a768e;
+
+                background-color:  #2a768e;
+
+                background-color:  #2a768e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col3 {
+
+                background-color:  #34b679;
+
+                background-color:  #34b679;
+
+                background-color:  #34b679;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col4 {
+
+                background-color:  #297a8e;
+
+                background-color:  #297a8e;
+
+                background-color:  #297a8e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col0 {
+
+                background-color:  #1f998a;
+
+                background-color:  #1f998a;
+
+                background-color:  #1f998a;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col1 {
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col2 {
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+                background-color:  #31688e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col3 {
+
+                background-color:  #1f968b;
+
+                background-color:  #1f968b;
+
+                background-color:  #1f968b;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col4 {
+
+                background-color:  #5cc863;
+
+                background-color:  #5cc863;
+
+                background-color:  #5cc863;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col0 {
+
+                background-color:  #1f9e89;
+
+                background-color:  #1f9e89;
+
+                background-color:  #1f9e89;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col1 {
+
+                background-color:  #40bd72;
+
+                background-color:  #40bd72;
+
+                background-color:  #40bd72;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col2 {
+
+                background-color:  #5cc863;
+
+                background-color:  #5cc863;
+
+                background-color:  #5cc863;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col3 {
+
+                background-color:  #228c8d;
+
+                background-color:  #228c8d;
+
+                background-color:  #228c8d;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col4 {
+
+                background-color:  #28ae80;
+
+                background-color:  #28ae80;
+
+                background-color:  #28ae80;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col0 {
+
+                background-color:  #2a788e;
+
+                background-color:  #2a788e;
+
+                background-color:  #2a788e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col1 {
+
+                background-color:  #23898e;
+
+                background-color:  #23898e;
+
+                background-color:  #23898e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col2 {
+
+                background-color:  #1fa088;
+
+                background-color:  #1fa088;
+
+                background-color:  #1fa088;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col3 {
+
+                background-color:  #90d743;
+
+                background-color:  #90d743;
+
+                background-color:  #90d743;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col4 {
+
+                background-color:  #2cb17e;
+
+                background-color:  #2cb17e;
+
+                background-color:  #2cb17e;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col0 {
+
+                background-color:  #22a785;
+
+                background-color:  #22a785;
+
+                background-color:  #22a785;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col1 {
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col2 {
+
+                background-color:  #44bf70;
+
+                background-color:  #44bf70;
+
+                background-color:  #44bf70;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col3 {
+
+                background-color:  #d2e21b;
+
+                background-color:  #d2e21b;
+
+                background-color:  #d2e21b;
+
+            }
+
+            #T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col4 {
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+                background-color:  #fde725;
+
+            }
+
+        </style>
+
+        <table id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb">
+
+
+        <thead>
+
+            <tr>
+
+                <th class="blank">
+
+                <th class="col_heading level0 col0">a
+
+                <th class="col_heading level0 col1">b
+
+                <th class="col_heading level0 col2">c
+
+                <th class="col_heading level0 col3">d
+
+                <th class="col_heading level0 col4">e
+
+            </tr>
+
+        </thead>
+        <tbody>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row0">
+
+                        0
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col0" class="data row0 col0">
+
+                        -1.085631
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col1" class="data row0 col1">
+
+                        0.997345
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col2" class="data row0 col2">
+
+                        0.282978
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col3" class="data row0 col3">
+
+                        -1.506295
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow0_col4" class="data row0 col4">
+
+                        -0.5786
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row1">
+
+                        1
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col0" class="data row1 col0">
+
+                        1.651437
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col1" class="data row1 col1">
+
+                        -2.426679
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col2" class="data row1 col2">
+
+                        -0.428913
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col3" class="data row1 col3">
+
+                        1.265936
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow1_col4" class="data row1 col4">
+
+                        -0.86674
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row2">
+
+                        2
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col0" class="data row2 col0">
+
+                        -0.678886
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col1" class="data row2 col1">
+
+                        -0.094709
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col2" class="data row2 col2">
+
+                        1.49139
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col3" class="data row2 col3">
+
+                        -0.638902
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow2_col4" class="data row2 col4">
+
+                        -0.443982
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row3">
+
+                        3
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col0" class="data row3 col0">
+
+                        -0.434351
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col1" class="data row3 col1">
+
+                        2.20593
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col2" class="data row3 col2">
+
+                        2.186786
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col3" class="data row3 col3">
+
+                        1.004054
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow3_col4" class="data row3 col4">
+
+                        0.386186
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row4">
+
+                        4
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col0" class="data row4 col0">
+
+                        0.737369
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col1" class="data row4 col1">
+
+                        1.490732
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col2" class="data row4 col2">
+
+                        -0.935834
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col3" class="data row4 col3">
+
+                        1.175829
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow4_col4" class="data row4 col4">
+
+                        -1.253881
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row5">
+
+                        5
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col0" class="data row5 col0">
+
+                        -0.637752
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col1" class="data row5 col1">
+
+                        0.907105
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col2" class="data row5 col2">
+
+                        -1.428681
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col3" class="data row5 col3">
+
+                        -0.140069
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow5_col4" class="data row5 col4">
+
+                        -0.861755
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row6">
+
+                        6
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col0" class="data row6 col0">
+
+                        -0.255619
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col1" class="data row6 col1">
+
+                        -2.798589
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col2" class="data row6 col2">
+
+                        -1.771533
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col3" class="data row6 col3">
+
+                        -0.699877
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow6_col4" class="data row6 col4">
+
+                        0.927462
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row7">
+
+                        7
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col0" class="data row7 col0">
+
+                        -0.173636
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col1" class="data row7 col1">
+
+                        0.002846
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col2" class="data row7 col2">
+
+                        0.688223
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col3" class="data row7 col3">
+
+                        -0.879536
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow7_col4" class="data row7 col4">
+
+                        0.283627
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row8">
+
+                        8
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col0" class="data row8 col0">
+
+                        -0.805367
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col1" class="data row8 col1">
+
+                        -1.727669
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col2" class="data row8 col2">
+
+                        -0.3909
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col3" class="data row8 col3">
+
+                        0.573806
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow8_col4" class="data row8 col4">
+
+                        0.338589
+
+
+            </tr>
+
+            <tr>
+
+                <th id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fb" class="row_heading level4 row9">
+
+                        9
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col0" class="data row9 col0">
+
+                        -0.01183
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col1" class="data row9 col1">
+
+                        2.392365
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col2" class="data row9 col2">
+
+                        0.412912
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col3" class="data row9 col3">
+
+                        0.978736
+
+
+                <td id="T_8d7c2fec_8d9e_11e5_8e2c_a45e60bd97fbrow9_col4" class="data row9 col4">
+
+                        2.238143
+
+
+            </tr>
+
+        </tbody>
+        </table>
diff --git a/pandas/doc/sphinxext/README.rst b/pandas/doc/sphinxext/README.rst
new file mode 100644
index 0000000000000..ef52433e5e869
--- /dev/null
+++ b/pandas/doc/sphinxext/README.rst
@@ -0,0 +1,5 @@
+sphinxext
+=========
+
+This directory contains custom sphinx extensions in use in the pandas
+documentation.
diff --git a/pandas/doc/sphinxext/announce.py b/pandas/doc/sphinxext/announce.py
new file mode 100755
index 0000000000000..66e999e251c5e
--- /dev/null
+++ b/pandas/doc/sphinxext/announce.py
@@ -0,0 +1,176 @@
+#!/usr/bin/env python3
+"""
+Script to generate contributor and pull request lists
+
+This script generates contributor and pull request lists for release
+announcements using GitHub v3 protocol. Use requires an authentication token in
+order to have sufficient bandwidth, you can get one following the directions at
+`<https://help.github.com/articles/creating-an-access-token-for-command-line-use/>_
+Don't add any scope, as the default is read access to public information. The
+token may be stored in an environment variable as you only get one chance to
+see it.
+
+Usage::
+
+    $ ./scripts/announce.py <token> <revision range>
+
+The output is utf8 rst.
+
+Dependencies
+------------
+
+- gitpython
+- pygithub
+
+Some code was copied from scipy `tools/gh_lists.py` and `tools/authors.py`.
+
+Examples
+--------
+
+From the bash command line with $GITHUB token.
+
+    $ ./scripts/announce.py $GITHUB v1.11.0..v1.11.1 > announce.rst
+
+"""
+
+import codecs
+import os
+import re
+import textwrap
+
+from git import Repo
+
+# Contributors to be renamed.
+CONTRIBUTOR_MAPPING = {"znkjnffrezna": "znetbgcubravk"}
+
+UTF8Writer = codecs.getwriter("utf8")
+this_repo = Repo(os.path.join(os.path.dirname(__file__), "..", ".."))
+
+author_msg = """\
+A total of %d people contributed patches to this release.  People with a
+"+" by their names contributed a patch for the first time.
+"""
+
+pull_request_msg = """\
+A total of %d pull requests were merged for this release.
+"""
+
+
+def get_authors(revision_range):
+    pat = "^.*\\t(.*)$"
+    lst_release, cur_release = (r.strip() for r in revision_range.split(".."))
+
+    if "|" in cur_release:
+        # e.g. v1.0.1|HEAD
+        maybe_tag, head = cur_release.split("|")
+        assert head == "HEAD"
+        if maybe_tag in this_repo.tags:
+            cur_release = maybe_tag
+        else:
+            cur_release = head
+        revision_range = f"{lst_release}..{cur_release}"
+
+    # authors, in current release and previous to current release.
+    # We need two passes over the log for cur and prev, one to get the
+    # "Co-authored by" commits, which come from backports by the bot,
+    # and one for regular commits.
+    xpr = re.compile(r"Co-authored-by: (?P<name>[^<]+) ")
+    cur = set(
+        xpr.findall(
+            this_repo.git.log("--grep=Co-authored", "--pretty=%b", revision_range)
+        )
+    )
+    cur |= set(re.findall(pat, this_repo.git.shortlog("-s", revision_range), re.M))
+
+    pre = set(
+        xpr.findall(this_repo.git.log("--grep=Co-authored", "--pretty=%b", lst_release))
+    )
+    pre |= set(re.findall(pat, this_repo.git.shortlog("-s", lst_release), re.M))
+
+    # Homu is the author of auto merges, clean him out.
+    cur.discard("Homu")
+    pre.discard("Homu")
+
+    # Rename contributors according to mapping.
+    for old_name, new_name in CONTRIBUTOR_MAPPING.items():
+        old_name_decoded = codecs.decode(old_name, "rot13")
+        new_name_decoded = codecs.decode(new_name, "rot13")
+        if old_name_decoded in pre:
+            pre.discard(old_name_decoded)
+            pre.add(new_name_decoded)
+        if old_name_decoded in cur:
+            cur.discard(old_name_decoded)
+            cur.add(new_name_decoded)
+
+    # Append '+' to new authors.
+    authors = [s + " +" for s in cur - pre] + list(cur & pre)
+    authors.sort()
+    return authors
+
+
+def get_pull_requests(repo, revision_range):
+    prnums = []
+
+    # From regular merges
+    merges = this_repo.git.log("--oneline", "--merges", revision_range)
+    issues = re.findall("Merge pull request \\#(\\d*)", merges)
+    prnums.extend(int(s) for s in issues)
+
+    # From Homu merges (Auto merges)
+    issues = re.findall("Auto merge of \\#(\\d*)", merges)
+    prnums.extend(int(s) for s in issues)
+
+    # From fast forward squash-merges
+    commits = this_repo.git.log(
+        "--oneline", "--no-merges", "--first-parent", revision_range
+    )
+    issues = re.findall("^.*\\(\\#(\\d+)\\)$", commits, re.M)
+    prnums.extend(int(s) for s in issues)
+
+    # get PR data from GitHub repo
+    prnums.sort()
+    prs = [repo.get_pull(n) for n in prnums]
+    return prs
+
+
+def build_components(revision_range, heading="Contributors"):
+    lst_release, cur_release = (r.strip() for r in revision_range.split(".."))
+    authors = get_authors(revision_range)
+
+    return {
+        "heading": heading,
+        "author_message": author_msg % len(authors),
+        "authors": authors,
+    }
+
+
+def build_string(revision_range, heading="Contributors"):
+    components = build_components(revision_range, heading=heading)
+    components["uline"] = "=" * len(components["heading"])
+    components["authors"] = "* " + "\n* ".join(components["authors"])
+
+    # Don't change this to an fstring. It breaks the formatting.
+    tpl = textwrap.dedent(
+        """\
+    {heading}
+    {uline}
+
+    {author_message}
+    {authors}"""
+    ).format(**components)
+    return tpl
+
+
+def main(revision_range):
+    # document authors
+    text = build_string(revision_range)
+    print(text)
+
+
+if __name__ == "__main__":
+    from argparse import ArgumentParser
+
+    parser = ArgumentParser(description="Generate author lists for release")
+    parser.add_argument("revision_range", help="<revision>..<revision>")
+    args = parser.parse_args()
+    main(args.revision_range)
diff --git a/pandas/doc/sphinxext/contributors.py b/pandas/doc/sphinxext/contributors.py
new file mode 100644
index 0000000000000..06f205b5cc3ce
--- /dev/null
+++ b/pandas/doc/sphinxext/contributors.py
@@ -0,0 +1,58 @@
+"""Sphinx extension for listing code contributors to a release.
+
+Usage::
+
+   .. contributors:: v0.23.0..v0.23.1
+
+This will be replaced with a message indicating the number of
+code contributors and commits, and then list each contributor
+individually. For development versions (before a tag is available)
+use::
+
+    .. contributors:: v0.23.0..v0.23.1|HEAD
+
+While the v0.23.1 tag does not exist, that will use the HEAD of the
+branch as the end of the revision range.
+"""
+
+from announce import build_components
+from docutils import nodes
+from docutils.parsers.rst import Directive
+import git
+
+
+class ContributorsDirective(Directive):
+    required_arguments = 1
+    name = "contributors"
+
+    def run(self):
+        range_ = self.arguments[0]
+        if range_.endswith("x..HEAD"):
+            return [nodes.paragraph(), nodes.bullet_list()]
+        try:
+            components = build_components(range_)
+        except git.GitCommandError as exc:
+            return [
+                self.state.document.reporter.warning(
+                    f"Cannot find contributors for range {repr(range_)}: {exc}",
+                    line=self.lineno,
+                )
+            ]
+        else:
+            message = nodes.paragraph()
+            message += nodes.Text(components["author_message"])
+
+            listnode = nodes.bullet_list()
+
+            for author in components["authors"]:
+                para = nodes.paragraph()
+                para += nodes.Text(author)
+                listnode += nodes.list_item("", para)
+
+        return [message, listnode]
+
+
+def setup(app):
+    app.add_directive("contributors", ContributorsDirective)
+
+    return {"version": "0.1", "parallel_read_safe": True, "parallel_write_safe": True}
diff --git a/pandas/environment.yml b/pandas/environment.yml
new file mode 100644
index 0000000000000..69647a436e3ad
--- /dev/null
+++ b/pandas/environment.yml
@@ -0,0 +1,123 @@
+# Local development dependencies including docs building, website upload, ASV benchmark
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.10
+  - pip
+
+  # build dependencies
+  - versioneer
+  - cython~=3.0.5
+  - meson=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - pytest-qt>=4.4.0
+  - pytest-localserver
+  - pyqt>=5.15.9
+  - coverage
+
+  # required dependencies
+  - python-dateutil
+  - numpy<2
+
+  # optional dependencies
+  - beautifulsoup4>=4.11.2
+  - blosc
+  - bottleneck>=1.3.6
+  - fastparquet>=2023.10.0
+  - fsspec>=2022.11.0
+  - html5lib>=1.1
+  - hypothesis>=6.84.0
+  - gcsfs>=2022.11.0
+  - ipython
+  - pickleshare  # Needed for IPython Sphinx directive in the docs GH#60429
+  - jinja2>=3.1.2
+  - lxml>=4.9.2
+  - matplotlib>=3.6.3
+  - numba>=0.56.4
+  - numexpr>=2.8.4
+  - openpyxl>=3.1.0
+  - odfpy>=1.4.1
+  - py
+  - psycopg2>=2.9.6
+  - pyarrow>=10.0.1
+  - pymysql>=1.0.2
+  - pyreadstat>=1.2.0
+  - pytables>=3.8.0
+  - python-calamine>=0.1.7
+  - pytz>=2023.4
+  - pyxlsb>=1.0.10
+  - s3fs>=2022.11.0
+  - scipy>=1.10.0
+  - sqlalchemy>=2.0.0
+  - tabulate>=0.9.0
+  - xarray>=2022.12.0, <=2024.9.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=3.0.5
+  - zstandard>=0.19.0
+
+  # downstream packages
+  - dask-core
+  - seaborn-base
+
+  # local testing dependencies
+  - moto
+  - flask
+
+  # benchmarks
+  - asv>=0.6.1
+
+  ## The compiler packages are meta-packages and install the correct compiler (activation) packages on the respective platforms.
+  - c-compiler
+  - cxx-compiler
+
+  # code checks
+  - flake8=7.1.0  # run in subprocess over docstring examples
+  - mypy=1.13.0  # pre-commit uses locally installed mypy
+  - tokenize-rt  # scripts/check_for_inconsistent_pandas_namespace.py
+  - pre-commit>=4.0.1
+
+  # documentation
+  - gitpython  # obtain contributors from git for whatsnew
+  - gitdb
+  - google-auth
+  - natsort  # DataFrame.sort_values doctest
+  - numpydoc
+  - pydata-sphinx-theme=0.16
+  - pytest-cython  # doctest
+  - sphinx
+  - sphinx-design
+  - sphinx-copybutton
+  - types-python-dateutil
+  - types-PyMySQL
+  - types-pytz
+  - types-PyYAML
+  - types-setuptools
+
+  # documentation (jupyter notebooks)
+  - nbconvert>=7.11.0
+  - nbsphinx
+  - pandoc
+  - ipywidgets
+  - nbformat
+  - notebook>=7.0.6
+  - ipykernel
+
+  # web
+  # - jinja2  # already listed in optional dependencies, but documented here for reference
+  - markdown
+  - feedparser
+  - pyyaml
+  - requests
+  - pygments # Code highlighting
+
+  - pip:
+      - adbc-driver-postgresql>=0.10.0
+      - adbc-driver-sqlite>=0.8.0
+      - typing_extensions; python_version<"3.11"
+      - tzdata>=2022.7
diff --git a/pandas/generate_pxi.py b/pandas/generate_pxi.py
new file mode 100644
index 0000000000000..e7e85900aa6e9
--- /dev/null
+++ b/pandas/generate_pxi.py
@@ -0,0 +1,33 @@
+import argparse
+import os
+
+from Cython import Tempita
+
+
+def process_tempita(pxifile, outfile) -> None:
+    with open(pxifile, encoding="utf-8") as f:
+        tmpl = f.read()
+    pyxcontent = Tempita.sub(tmpl)
+
+    with open(outfile, "w", encoding="utf-8") as f:
+        f.write(pyxcontent)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("infile", type=str, help="Path to the input file")
+    parser.add_argument("-o", "--outdir", type=str, help="Path to the output directory")
+    args = parser.parse_args()
+
+    if not args.infile.endswith(".in"):
+        raise ValueError(f"Unexpected extension: {args.infile}")
+
+    outdir_abs = os.path.join(os.getcwd(), args.outdir)
+    outfile = os.path.join(
+        outdir_abs, os.path.splitext(os.path.split(args.infile)[1])[0]
+    )
+
+    process_tempita(args.infile, outfile)
+
+
+main()
diff --git a/pandas/generate_version.py b/pandas/generate_version.py
new file mode 100644
index 0000000000000..7dc1bdcf329d0
--- /dev/null
+++ b/pandas/generate_version.py
@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+
+# Note: This file has to live next to setup.py or versioneer will not work
+import argparse
+import os
+import sys
+
+import versioneer
+
+sys.path.insert(0, "")
+
+
+def write_version_info(path) -> None:
+    version = None
+    git_version = None
+
+    try:
+        import _version_meson
+
+        version = _version_meson.__version__
+        git_version = _version_meson.__git_version__
+    except ImportError:
+        version = versioneer.get_version()
+        git_version = versioneer.get_versions()["full-revisionid"]
+    if os.environ.get("MESON_DIST_ROOT"):
+        path = os.path.join(os.environ.get("MESON_DIST_ROOT"), path)
+    with open(path, "w", encoding="utf-8") as file:
+        file.write(f'__version__="{version}"\n')
+        file.write(f'__git_version__="{git_version}"\n')
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "-o",
+        "--outfile",
+        type=str,
+        help="Path to write version info to",
+        required=False,
+    )
+    parser.add_argument(
+        "--print",
+        default=False,
+        action="store_true",
+        help="Whether to print out the version",
+        required=False,
+    )
+    args = parser.parse_args()
+
+    if args.outfile:
+        if not args.outfile.endswith(".py"):
+            raise ValueError(
+                f"Output file must be a Python file. "
+                f"Got: {args.outfile} as filename instead"
+            )
+
+        write_version_info(args.outfile)
+
+    if args.print:
+        try:
+            import _version_meson
+
+            version = _version_meson.__version__
+        except ImportError:
+            version = versioneer.get_version()
+        print(version)
+
+
+main()
diff --git a/pandas/gitpod/Dockerfile b/pandas/gitpod/Dockerfile
new file mode 100644
index 0000000000000..dd4ddf64d16b4
--- /dev/null
+++ b/pandas/gitpod/Dockerfile
@@ -0,0 +1,100 @@
+#
+# Dockerfile for pandas development
+#
+# Usage:
+# -------
+#
+# To make a local build of the container, from the 'Docker-dev' directory:
+# docker build  --rm -f "Dockerfile" -t <build-tag> "."
+#
+# To use the container use the following command. It assumes that you are in
+# the root folder of the pandas git repository, making it available as
+# /home/pandas in the container. Whatever changes you make to that directory
+# are visible in the host and container.
+# The docker image is retrieved from the pandas dockerhub repository
+#
+# docker run --rm -it -v $(pwd):/home/pandas pandas/pandas-dev:<image-tag>
+#
+# By default the container will activate the conda environment pandas-dev
+# which contains all the dependencies needed for pandas development
+#
+# To build and install pandas run:
+#               python setup.py build_ext -j 4
+#               python -m pip install -e . --no-build-isolation
+#
+# This image is based on: Ubuntu 20.04 (focal)
+# https://hub.docker.com/_/ubuntu/?tab=tags&name=focal
+# OS/ARCH: linux/amd64
+FROM gitpod/workspace-base:latest
+
+ARG MAMBAFORGE_VERSION="23.1.0-3"
+ARG CONDA_ENV=pandas-dev
+ARG PANDAS_HOME="/home/pandas"
+
+
+# ---- Configure environment ----
+ENV CONDA_DIR=/home/gitpod/mambaforge3 \
+    SHELL=/bin/bash
+ENV PATH=${CONDA_DIR}/bin:$PATH \
+    WORKSPACE=/workspace/pandas
+
+# -----------------------------------------------------------------------------
+# ---- Creating as root - note: make sure to change to gitpod in the end ----
+USER root
+
+# Avoid warnings by switching to noninteractive
+ENV DEBIAN_FRONTEND=noninteractive
+
+# Configure apt and install packages
+RUN apt-get update \
+    && apt-get -y install --no-install-recommends apt-utils dialog 2>&1 \
+    #
+    # Install tzdata and configure timezone (fix for tests which try to read from "/etc/localtime")
+    && apt-get -y install tzdata \
+    && ln -fs /usr/share/zoneinfo/Etc/UTC /etc/localtime \
+    && dpkg-reconfigure -f noninteractive tzdata \
+    #
+    # Verify git, process tools, lsb-release (common in install instructions for CLIs) installed
+    && apt-get -y install git iproute2 procps iproute2 lsb-release \
+    #
+    # cleanup
+    && apt-get autoremove -y \
+    && apt-get clean -y \
+    && rm -rf /var/lib/apt/lists/*
+
+# Switch back to dialog for any ad-hoc use of apt-get
+ENV DEBIAN_FRONTEND=dialog
+
+# Allows this Dockerfile to activate conda environments
+SHELL ["/bin/bash", "--login", "-o", "pipefail", "-c"]
+
+# -----------------------------------------------------------------------------
+# ---- Installing mamba  ----
+RUN wget -q -O mambaforge3.sh \
+    "https://github.com/conda-forge/miniforge/releases/download/$MAMBAFORGE_VERSION/Mambaforge-$MAMBAFORGE_VERSION-Linux-x86_64.sh" && \
+    bash mambaforge3.sh -p ${CONDA_DIR} -b && \
+    rm mambaforge3.sh
+
+# -----------------------------------------------------------------------------
+# ---- Copy needed files ----
+# basic workspace configurations
+COPY ./gitpod/workspace_config /usr/local/bin/workspace_config
+
+RUN chmod a+rx /usr/local/bin/workspace_config && \
+    workspace_config
+
+# the container to create a conda environment from it
+COPY environment.yml /tmp/environment.yml
+
+# ---- Create conda environment ----
+RUN mamba env create -f /tmp/environment.yml && \
+    conda activate $CONDA_ENV && \
+    mamba install ccache -y && \
+    # needed for docs rendering later on
+    python -m pip install --no-cache-dir sphinx-autobuild && \
+    conda clean --all -f -y && \
+    rm -rf /tmp/*
+
+# -----------------------------------------------------------------------------
+# Always make sure we are not root
+USER gitpod
diff --git a/pandas/gitpod/gitpod.Dockerfile b/pandas/gitpod/gitpod.Dockerfile
new file mode 100644
index 0000000000000..ab3f25b231e67
--- /dev/null
+++ b/pandas/gitpod/gitpod.Dockerfile
@@ -0,0 +1,48 @@
+# Doing a local shallow clone - keeps the container secure
+# and much slimmer than using COPY directly or making a
+# remote clone
+ARG BASE_CONTAINER="pandas/pandas-dev:latest"
+FROM gitpod/workspace-base:latest as clone
+
+# the clone should be deep enough for versioneer to work
+RUN git clone https://github.com/pandas-dev/pandas --depth 12 /tmp/pandas
+
+# -----------------------------------------------------------------------------
+# Using the pandas-dev Docker image as a base
+# This way, we ensure we have all the needed compilers and dependencies
+# while reducing the build time
+FROM ${BASE_CONTAINER} as build
+
+# -----------------------------------------------------------------------------
+USER root
+
+# -----------------------------------------------------------------------------
+# ---- ENV variables ----
+# ---- Directories needed ----
+ENV WORKSPACE=/workspace/pandas/ \
+    CONDA_ENV=pandas-dev
+
+# Allows this micromamba.Dockerfile to activate conda environments
+SHELL ["/bin/bash", "--login", "-o", "pipefail", "-c"]
+
+# Copy over the shallow clone
+COPY --from=clone --chown=gitpod /tmp/pandas ${WORKSPACE}
+
+# Everything happens in the /workspace/pandas directory
+WORKDIR ${WORKSPACE}
+
+# Build pandas to populate the cache used by ccache
+RUN git config --global --add safe.directory /workspace/pandas
+RUN conda activate ${CONDA_ENV} && \
+    python -m pip install -e . --no-build-isolation && \
+    python setup.py build_ext --inplace && \
+    ccache -s
+
+# Gitpod will load the repository into /workspace/pandas. We remove the
+# directory from the image to prevent conflicts
+RUN rm -rf ${WORKSPACE}
+
+# -----------------------------------------------------------------------------
+# Always return to non privileged user
+RUN chown -R gitpod:gitpod /home/gitpod/.cache/
+USER gitpod
diff --git a/pandas/gitpod/settings.json b/pandas/gitpod/settings.json
new file mode 100644
index 0000000000000..2c2c3b551e1d1
--- /dev/null
+++ b/pandas/gitpod/settings.json
@@ -0,0 +1,5 @@
+{
+    "esbonio.server.pythonPath": "/usr/local/bin/python",
+    "restructuredtext.linter.disabledLinters": ["doc8","rst-lint", "rstcheck"],
+    "python.defaultInterpreterPath": "/usr/local/bin/python"
+}
diff --git a/pandas/gitpod/workspace_config b/pandas/gitpod/workspace_config
new file mode 100644
index 0000000000000..d49c93ec83db9
--- /dev/null
+++ b/pandas/gitpod/workspace_config
@@ -0,0 +1,54 @@
+#!/bin/bash
+# Basic configurations for the workspace
+
+set -e
+
+# gitpod/workspace-base needs at least one file here
+touch /home/gitpod/.bashrc.d/empty
+
+# Add git aliases
+git config --global alias.co checkout
+git config --global alias.ci commit
+git config --global alias.st status
+git config --global alias.br branch
+git config --global alias.hist "log --pretty=format:'%h %ad | %s%d [%an]' --graph --date=short"
+git config --global alias.type 'cat-file -t'
+git config --global alias.dump 'cat-file -p'
+
+# Enable basic vim defaults in ~/.vimrc
+echo "filetype plugin indent on" >>~/.vimrc
+echo "set colorcolumn=80" >>~/.vimrc
+echo "set number" >>~/.vimrc
+echo "syntax enable" >>~/.vimrc
+
+# Vanity custom bash prompt - makes it more legible
+echo "PS1='\[\e]0;\u \w\a\]\[\033[01;36m\]\u\[\033[m\] > \[\033[38;5;141m\]\w\[\033[m\] \\$ '" >>~/.bashrc
+
+# Enable prompt color in the skeleton .bashrc
+# hadolint ignore=SC2016
+sed -i 's/^#force_color_prompt=yes/force_color_prompt=yes/' /etc/skel/.bashrc
+
+# .gitpod.yml is configured to install pandas from /workspace/pandas
+echo "export PYTHONPATH=${WORKSPACE}" >>~/.bashrc
+
+# make conda activate command available from /bin/bash (login and interactive)
+if [[ ! -f "/etc/profile.d/conda.sh" ]]; then
+    ln -s ${CONDA_DIR}/etc/profile.d/conda.sh /etc/profile.d/conda.sh
+fi
+echo ". ${CONDA_DIR}/etc/profile.d/conda.sh" >>~/.bashrc
+echo "conda activate pandas-dev" >>~/.bashrc
+
+# Enable prompt color in the skeleton .bashrc
+# hadolint ignore=SC2016
+sed -i 's/^#force_color_prompt=yes/force_color_prompt=yes/' /etc/skel/.bashrc
+
+# .gitpod.yml is configured to install pandas from /workspace/pandas
+echo "export PYTHONPATH=/workspace/pandas" >>~/.bashrc
+
+# Set up ccache for compilers for this Dockerfile
+# REF: https://github.com/conda-forge/compilers-feedstock/issues/31
+echo "conda activate pandas-dev" >>~/.startuprc
+echo "export CC=\"ccache \$CC\"" >>~/.startuprc
+echo "export CXX=\"ccache \$CXX\"" >>~/.startuprc
+echo "source ~/.startuprc" >>~/.profile
+echo "source ~/.startuprc" >>~/.bashrc
diff --git a/pandas/io/clipboard/__init__.py b/pandas/io/clipboard/__init__.py
index 6491849925e86..a1ada378c0a94 100644
--- a/pandas/io/clipboard/__init__.py
+++ b/pandas/io/clipboard/__init__.py
@@ -735,11 +735,11 @@ def waitForNewPaste(timeout=None):
 
 __all__ = [
     "copy",
+    "determine_clipboard",
     "paste",
-    "waitForPaste",
-    "waitForNewPaste",
     "set_clipboard",
-    "determine_clipboard",
+    "waitForNewPaste",
+    "waitForPaste",
 ]
 
 # pandas aliases
diff --git a/pandas/pandas/__init__.py b/pandas/pandas/__init__.py
new file mode 100644
index 0000000000000..c570fb8d70204
--- /dev/null
+++ b/pandas/pandas/__init__.py
@@ -0,0 +1,350 @@
+from __future__ import annotations
+
+__docformat__ = "restructuredtext"
+
+# Let users know if they're missing any of our hard dependencies
+_hard_dependencies = ("numpy", "dateutil")
+_missing_dependencies = []
+
+for _dependency in _hard_dependencies:
+    try:
+        __import__(_dependency)
+    except ImportError as _e:  # pragma: no cover
+        _missing_dependencies.append(f"{_dependency}: {_e}")
+
+if _missing_dependencies:  # pragma: no cover
+    raise ImportError(
+        "Unable to import required dependencies:\n" + "\n".join(_missing_dependencies)
+    )
+del _hard_dependencies, _dependency, _missing_dependencies
+
+try:
+    # numpy compat
+    from pandas.compat import (
+        is_numpy_dev as _is_numpy_dev,  # pyright: ignore[reportUnusedImport] # noqa: F401
+    )
+except ImportError as _err:  # pragma: no cover
+    _module = _err.name
+    raise ImportError(
+        f"C extension: {_module} not built. If you want to import "
+        "pandas from the source directory, you may need to run "
+        "'python -m pip install -ve . --no-build-isolation -Ceditable-verbose=true' "
+        "to build the C extensions first."
+    ) from _err
+
+from pandas._config import (
+    get_option,
+    set_option,
+    reset_option,
+    describe_option,
+    option_context,
+    options,
+)
+
+# let init-time option registration happen
+import pandas.core.config_init  # pyright: ignore[reportUnusedImport] # noqa: F401
+
+from pandas.core.api import (
+    # dtype
+    ArrowDtype,
+    Int8Dtype,
+    Int16Dtype,
+    Int32Dtype,
+    Int64Dtype,
+    UInt8Dtype,
+    UInt16Dtype,
+    UInt32Dtype,
+    UInt64Dtype,
+    Float32Dtype,
+    Float64Dtype,
+    CategoricalDtype,
+    PeriodDtype,
+    IntervalDtype,
+    DatetimeTZDtype,
+    StringDtype,
+    BooleanDtype,
+    # missing
+    NA,
+    isna,
+    isnull,
+    notna,
+    notnull,
+    # indexes
+    Index,
+    CategoricalIndex,
+    RangeIndex,
+    MultiIndex,
+    IntervalIndex,
+    TimedeltaIndex,
+    DatetimeIndex,
+    PeriodIndex,
+    IndexSlice,
+    # tseries
+    NaT,
+    Period,
+    period_range,
+    Timedelta,
+    timedelta_range,
+    Timestamp,
+    date_range,
+    bdate_range,
+    Interval,
+    interval_range,
+    DateOffset,
+    # conversion
+    to_numeric,
+    to_datetime,
+    to_timedelta,
+    # misc
+    Flags,
+    Grouper,
+    factorize,
+    unique,
+    NamedAgg,
+    array,
+    Categorical,
+    set_eng_float_format,
+    Series,
+    DataFrame,
+)
+
+from pandas.core.dtypes.dtypes import SparseDtype
+
+from pandas.tseries.api import infer_freq
+from pandas.tseries import offsets
+
+from pandas.core.computation.api import eval
+
+from pandas.core.reshape.api import (
+    concat,
+    lreshape,
+    melt,
+    wide_to_long,
+    merge,
+    merge_asof,
+    merge_ordered,
+    crosstab,
+    pivot,
+    pivot_table,
+    get_dummies,
+    from_dummies,
+    cut,
+    qcut,
+)
+
+from pandas import api, arrays, errors, io, plotting, tseries
+from pandas import testing
+from pandas.util._print_versions import show_versions
+
+from pandas.io.api import (
+    # excel
+    ExcelFile,
+    ExcelWriter,
+    read_excel,
+    # parsers
+    read_csv,
+    read_fwf,
+    read_table,
+    # pickle
+    read_pickle,
+    to_pickle,
+    # pytables
+    HDFStore,
+    read_hdf,
+    # sql
+    read_sql,
+    read_sql_query,
+    read_sql_table,
+    # misc
+    read_clipboard,
+    read_parquet,
+    read_orc,
+    read_feather,
+    read_html,
+    read_xml,
+    read_json,
+    read_stata,
+    read_sas,
+    read_spss,
+)
+
+from pandas.io.json._normalize import json_normalize
+
+from pandas.util._tester import test
+
+# use the closest tagged version if possible
+_built_with_meson = False
+try:
+    from pandas._version_meson import (  # pyright: ignore [reportMissingImports]
+        __version__,
+        __git_version__,
+    )
+
+    _built_with_meson = True
+except ImportError:
+    from pandas._version import get_versions
+
+    v = get_versions()
+    __version__ = v.get("closest-tag", v["version"])
+    __git_version__ = v.get("full-revisionid")
+    del get_versions, v
+
+
+# module level doc-string
+__doc__ = """
+pandas - a powerful data analysis and manipulation library for Python
+=====================================================================
+
+**pandas** is a Python package providing fast, flexible, and expressive data
+structures designed to make working with "relational" or "labeled" data both
+easy and intuitive. It aims to be the fundamental high-level building block for
+doing practical, **real world** data analysis in Python. Additionally, it has
+the broader goal of becoming **the most powerful and flexible open source data
+analysis / manipulation tool available in any language**. It is already well on
+its way toward this goal.
+
+Main Features
+-------------
+Here are just a few of the things that pandas does well:
+
+  - Easy handling of missing data in floating point as well as non-floating
+    point data.
+  - Size mutability: columns can be inserted and deleted from DataFrame and
+    higher dimensional objects
+  - Automatic and explicit data alignment: objects can be explicitly aligned
+    to a set of labels, or the user can simply ignore the labels and let
+    `Series`, `DataFrame`, etc. automatically align the data for you in
+    computations.
+  - Powerful, flexible group by functionality to perform split-apply-combine
+    operations on data sets, for both aggregating and transforming data.
+  - Make it easy to convert ragged, differently-indexed data in other Python
+    and NumPy data structures into DataFrame objects.
+  - Intelligent label-based slicing, fancy indexing, and subsetting of large
+    data sets.
+  - Intuitive merging and joining data sets.
+  - Flexible reshaping and pivoting of data sets.
+  - Hierarchical labeling of axes (possible to have multiple labels per tick).
+  - Robust IO tools for loading data from flat files (CSV and delimited),
+    Excel files, databases, and saving/loading data from the ultrafast HDF5
+    format.
+  - Time series-specific functionality: date range generation and frequency
+    conversion, moving window statistics, date shifting and lagging.
+"""
+
+# Use __all__ to let type checkers know what is part of the public API.
+# Pandas is not (yet) a py.typed library: the public API is determined
+# based on the documentation.
+__all__ = [
+    "NA",
+    "ArrowDtype",
+    "BooleanDtype",
+    "Categorical",
+    "CategoricalDtype",
+    "CategoricalIndex",
+    "DataFrame",
+    "DateOffset",
+    "DatetimeIndex",
+    "DatetimeTZDtype",
+    "ExcelFile",
+    "ExcelWriter",
+    "Flags",
+    "Float32Dtype",
+    "Float64Dtype",
+    "Grouper",
+    "HDFStore",
+    "Index",
+    "IndexSlice",
+    "Int8Dtype",
+    "Int16Dtype",
+    "Int32Dtype",
+    "Int64Dtype",
+    "Interval",
+    "IntervalDtype",
+    "IntervalIndex",
+    "MultiIndex",
+    "NaT",
+    "NamedAgg",
+    "Period",
+    "PeriodDtype",
+    "PeriodIndex",
+    "RangeIndex",
+    "Series",
+    "SparseDtype",
+    "StringDtype",
+    "Timedelta",
+    "TimedeltaIndex",
+    "Timestamp",
+    "UInt8Dtype",
+    "UInt16Dtype",
+    "UInt32Dtype",
+    "UInt64Dtype",
+    "api",
+    "array",
+    "arrays",
+    "bdate_range",
+    "concat",
+    "crosstab",
+    "cut",
+    "date_range",
+    "describe_option",
+    "errors",
+    "eval",
+    "factorize",
+    "from_dummies",
+    "get_dummies",
+    "get_option",
+    "infer_freq",
+    "interval_range",
+    "io",
+    "isna",
+    "isnull",
+    "json_normalize",
+    "lreshape",
+    "melt",
+    "merge",
+    "merge_asof",
+    "merge_ordered",
+    "notna",
+    "notnull",
+    "offsets",
+    "option_context",
+    "options",
+    "period_range",
+    "pivot",
+    "pivot_table",
+    "plotting",
+    "qcut",
+    "read_clipboard",
+    "read_csv",
+    "read_excel",
+    "read_feather",
+    "read_fwf",
+    "read_hdf",
+    "read_html",
+    "read_json",
+    "read_orc",
+    "read_parquet",
+    "read_pickle",
+    "read_sas",
+    "read_spss",
+    "read_sql",
+    "read_sql_query",
+    "read_sql_table",
+    "read_stata",
+    "read_table",
+    "read_xml",
+    "reset_option",
+    "set_eng_float_format",
+    "set_option",
+    "show_versions",
+    "test",
+    "testing",
+    "timedelta_range",
+    "to_datetime",
+    "to_numeric",
+    "to_pickle",
+    "to_timedelta",
+    "tseries",
+    "unique",
+    "wide_to_long",
+]
diff --git a/pandas/pandas/_config/__init__.py b/pandas/pandas/_config/__init__.py
new file mode 100644
index 0000000000000..463e8af7cc561
--- /dev/null
+++ b/pandas/pandas/_config/__init__.py
@@ -0,0 +1,35 @@
+"""
+pandas._config is considered explicitly upstream of everything else in pandas,
+should have no intra-pandas dependencies.
+
+importing `dates` and `display` ensures that keys needed by _libs
+are initialized.
+"""
+
+__all__ = [
+    "config",
+    "describe_option",
+    "detect_console_encoding",
+    "get_option",
+    "option_context",
+    "options",
+    "reset_option",
+    "set_option",
+]
+from pandas._config import config
+from pandas._config import dates  # pyright: ignore[reportUnusedImport]  # noqa: F401
+from pandas._config.config import (
+    _global_config,
+    describe_option,
+    get_option,
+    option_context,
+    options,
+    reset_option,
+    set_option,
+)
+from pandas._config.display import detect_console_encoding
+
+
+def using_string_dtype() -> bool:
+    _mode_options = _global_config["future"]
+    return _mode_options["infer_string"]
diff --git a/pandas/pandas/_config/config.py b/pandas/pandas/_config/config.py
new file mode 100644
index 0000000000000..ce53e05608ba7
--- /dev/null
+++ b/pandas/pandas/_config/config.py
@@ -0,0 +1,910 @@
+"""
+The config module holds package-wide configurables and provides
+a uniform API for working with them.
+
+Overview
+========
+
+This module supports the following requirements:
+- options are referenced using keys in dot.notation, e.g. "x.y.option - z".
+- keys are case-insensitive.
+- functions should accept partial/regex keys, when unambiguous.
+- options can be registered by modules at import time.
+- options can be registered at init-time (via core.config_init)
+- options have a default value, and (optionally) a description and
+  validation function associated with them.
+- options can be deprecated, in which case referencing them
+  should produce a warning.
+- deprecated options can optionally be rerouted to a replacement
+  so that accessing a deprecated option reroutes to a differently
+  named option.
+- options can be reset to their default value.
+- all option can be reset to their default value at once.
+- all options in a certain sub - namespace can be reset at once.
+- the user can set / get / reset or ask for the description of an option.
+- a developer can register and mark an option as deprecated.
+- you can register a callback to be invoked when the option value
+  is set or reset. Changing the stored value is considered misuse, but
+  is not verboten.
+
+Implementation
+==============
+
+- Data is stored using nested dictionaries, and should be accessed
+  through the provided API.
+
+- "Registered options" and "Deprecated options" have metadata associated
+  with them, which are stored in auxiliary dictionaries keyed on the
+  fully-qualified key, e.g. "x.y.z.option".
+
+- the config_init module is imported by the package's __init__.py file.
+  placing any register_option() calls there will ensure those options
+  are available as soon as pandas is loaded. If you use register_option
+  in a module, it will only be available after that module is imported,
+  which you should be aware of.
+
+- `config_prefix` is a context_manager (for use with the `with` keyword)
+  which can save developers some typing, see the docstring.
+
+"""
+
+from __future__ import annotations
+
+from contextlib import contextmanager
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    NamedTuple,
+    cast,
+)
+import warnings
+
+from pandas._typing import F
+from pandas.util._exceptions import find_stack_level
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Generator,
+        Sequence,
+    )
+
+
+class DeprecatedOption(NamedTuple):
+    key: str
+    msg: str | None
+    rkey: str | None
+    removal_ver: str | None
+
+
+class RegisteredOption(NamedTuple):
+    key: str
+    defval: Any
+    doc: str
+    validator: Callable[[object], Any] | None
+    cb: Callable[[str], Any] | None
+
+
+# holds deprecated option metadata
+_deprecated_options: dict[str, DeprecatedOption] = {}
+
+# holds registered option metadata
+_registered_options: dict[str, RegisteredOption] = {}
+
+# holds the current values for registered options
+_global_config: dict[str, Any] = {}
+
+# keys which have a special meaning
+_reserved_keys: list[str] = ["all"]
+
+
+class OptionError(AttributeError, KeyError):
+    """
+    Exception raised for pandas.options.
+
+    Backwards compatible with KeyError checks.
+
+    See Also
+    --------
+    options : Access and modify global pandas settings.
+
+    Examples
+    --------
+    >>> pd.options.context
+    Traceback (most recent call last):
+    OptionError: No such option
+    """
+
+
+#
+# User API
+
+
+def _get_single_key(pat: str) -> str:
+    keys = _select_options(pat)
+    if len(keys) == 0:
+        _warn_if_deprecated(pat)
+        raise OptionError(f"No such keys(s): {pat!r}")
+    if len(keys) > 1:
+        raise OptionError("Pattern matched multiple keys")
+    key = keys[0]
+
+    _warn_if_deprecated(key)
+
+    key = _translate_key(key)
+
+    return key
+
+
+def get_option(pat: str) -> Any:
+    """
+    Retrieve the value of the specified option.
+
+    This method allows users to query the current value of a given option
+    in the pandas configuration system. Options control various display,
+    performance, and behavior-related settings within pandas.
+
+    Parameters
+    ----------
+    pat : str
+        Regexp which should match a single option.
+
+        .. warning::
+
+            Partial matches are supported for convenience, but unless you use the
+            full option name (e.g. x.y.z.option_name), your code may break in future
+            versions if new options with similar names are introduced.
+
+    Returns
+    -------
+    Any
+        The value of the option.
+
+    Raises
+    ------
+    OptionError : if no such option exists
+
+    See Also
+    --------
+    set_option : Set the value of the specified option or options.
+    reset_option : Reset one or more options to their default value.
+    describe_option : Print the description for one or more registered options.
+
+    Notes
+    -----
+    For all available options, please view the :ref:`User Guide <options.available>`
+    or use ``pandas.describe_option()``.
+
+    Examples
+    --------
+    >>> pd.get_option("display.max_columns")  # doctest: +SKIP
+    4
+    """
+    key = _get_single_key(pat)
+
+    # walk the nested dict
+    root, k = _get_root(key)
+    return root[k]
+
+
+def set_option(*args) -> None:
+    """
+    Set the value of the specified option or options.
+
+    This method allows fine-grained control over the behavior and display settings
+    of pandas. Options affect various functionalities such as output formatting,
+    display limits, and operational behavior. Settings can be modified at runtime
+    without requiring changes to global configurations or environment variables.
+
+    Parameters
+    ----------
+    *args : str | object
+        Arguments provided in pairs, which will be interpreted as (pattern, value)
+        pairs.
+        pattern: str
+        Regexp which should match a single option
+        value: object
+        New value of option
+
+        .. warning::
+
+            Partial pattern matches are supported for convenience, but unless you
+            use the full option name (e.g. x.y.z.option_name), your code may break in
+            future versions if new options with similar names are introduced.
+
+    Returns
+    -------
+    None
+        No return value.
+
+    Raises
+    ------
+    ValueError if odd numbers of non-keyword arguments are provided
+    TypeError if keyword arguments are provided
+    OptionError if no such option exists
+
+    See Also
+    --------
+    get_option : Retrieve the value of the specified option.
+    reset_option : Reset one or more options to their default value.
+    describe_option : Print the description for one or more registered options.
+    option_context : Context manager to temporarily set options in a ``with``
+        statement.
+
+    Notes
+    -----
+    For all available options, please view the :ref:`User Guide <options.available>`
+    or use ``pandas.describe_option()``.
+
+    Examples
+    --------
+    >>> pd.set_option("display.max_columns", 4)
+    >>> df = pd.DataFrame([[1, 2, 3, 4, 5], [6, 7, 8, 9, 10]])
+    >>> df
+    0  1  ...  3   4
+    0  1  2  ...  4   5
+    1  6  7  ...  9  10
+    [2 rows x 5 columns]
+    >>> pd.reset_option("display.max_columns")
+    """
+    # must at least 1 arg deal with constraints later
+    nargs = len(args)
+    if not nargs or nargs % 2 != 0:
+        raise ValueError("Must provide an even number of non-keyword arguments")
+
+    for k, v in zip(args[::2], args[1::2]):
+        key = _get_single_key(k)
+
+        opt = _get_registered_option(key)
+        if opt and opt.validator:
+            opt.validator(v)
+
+        # walk the nested dict
+        root, k_root = _get_root(key)
+        root[k_root] = v
+
+        if opt.cb:
+            opt.cb(key)
+
+
+def describe_option(pat: str = "", _print_desc: bool = True) -> str | None:
+    """
+    Print the description for one or more registered options.
+
+    Call with no arguments to get a listing for all registered options.
+
+    Parameters
+    ----------
+    pat : str, default ""
+        String or string regexp pattern.
+        Empty string will return all options.
+        For regexp strings, all matching keys will have their description displayed.
+    _print_desc : bool, default True
+        If True (default) the description(s) will be printed to stdout.
+        Otherwise, the description(s) will be returned as a string
+        (for testing).
+
+    Returns
+    -------
+    None
+        If ``_print_desc=True``.
+    str
+        If the description(s) as a string if ``_print_desc=False``.
+
+    See Also
+    --------
+    get_option : Retrieve the value of the specified option.
+    set_option : Set the value of the specified option or options.
+    reset_option : Reset one or more options to their default value.
+
+    Notes
+    -----
+    For all available options, please view the
+    :ref:`User Guide <options.available>`.
+
+    Examples
+    --------
+    >>> pd.describe_option("display.max_columns")  # doctest: +SKIP
+    display.max_columns : int
+        If max_cols is exceeded, switch to truncate view...
+    """
+    keys = _select_options(pat)
+    if len(keys) == 0:
+        raise OptionError(f"No such keys(s) for {pat=}")
+
+    s = "\n".join([_build_option_description(k) for k in keys])
+
+    if _print_desc:
+        print(s)
+        return None
+    return s
+
+
+def reset_option(pat: str) -> None:
+    """
+    Reset one or more options to their default value.
+
+    This method resets the specified pandas option(s) back to their default
+    values. It allows partial string matching for convenience, but users should
+    exercise caution to avoid unintended resets due to changes in option names
+    in future versions.
+
+    Parameters
+    ----------
+    pat : str/regex
+        If specified only options matching ``pat*`` will be reset.
+        Pass ``"all"`` as argument to reset all options.
+
+        .. warning::
+
+            Partial matches are supported for convenience, but unless you
+            use the full option name (e.g. x.y.z.option_name), your code may break
+            in future versions if new options with similar names are introduced.
+
+    Returns
+    -------
+    None
+        No return value.
+
+    See Also
+    --------
+    get_option : Retrieve the value of the specified option.
+    set_option : Set the value of the specified option or options.
+    describe_option : Print the description for one or more registered options.
+
+    Notes
+    -----
+    For all available options, please view the
+    :ref:`User Guide <options.available>`.
+
+    Examples
+    --------
+    >>> pd.reset_option("display.max_columns")  # doctest: +SKIP
+    """
+    keys = _select_options(pat)
+
+    if len(keys) == 0:
+        raise OptionError(f"No such keys(s) for {pat=}")
+
+    if len(keys) > 1 and len(pat) < 4 and pat != "all":
+        raise ValueError(
+            "You must specify at least 4 characters when "
+            "resetting multiple keys, use the special keyword "
+            '"all" to reset all the options to their default value'
+        )
+
+    for k in keys:
+        set_option(k, _registered_options[k].defval)
+
+
+def get_default_val(pat: str):
+    key = _get_single_key(pat)
+    return _get_registered_option(key).defval
+
+
+class DictWrapper:
+    """provide attribute-style access to a nested dict"""
+
+    d: dict[str, Any]
+
+    def __init__(self, d: dict[str, Any], prefix: str = "") -> None:
+        object.__setattr__(self, "d", d)
+        object.__setattr__(self, "prefix", prefix)
+
+    def __setattr__(self, key: str, val: Any) -> None:
+        prefix = object.__getattribute__(self, "prefix")
+        if prefix:
+            prefix += "."
+        prefix += key
+        # you can't set new keys
+        # can you can't overwrite subtrees
+        if key in self.d and not isinstance(self.d[key], dict):
+            set_option(prefix, val)
+        else:
+            raise OptionError("You can only set the value of existing options")
+
+    def __getattr__(self, key: str):
+        prefix = object.__getattribute__(self, "prefix")
+        if prefix:
+            prefix += "."
+        prefix += key
+        try:
+            v = object.__getattribute__(self, "d")[key]
+        except KeyError as err:
+            raise OptionError("No such option") from err
+        if isinstance(v, dict):
+            return DictWrapper(v, prefix)
+        else:
+            return get_option(prefix)
+
+    def __dir__(self) -> list[str]:
+        return list(self.d.keys())
+
+
+options = DictWrapper(_global_config)
+
+#
+# Functions for use by pandas developers, in addition to User - api
+
+
+@contextmanager
+def option_context(*args) -> Generator[None]:
+    """
+    Context manager to temporarily set options in a ``with`` statement.
+
+    This method allows users to set one or more pandas options temporarily
+    within a controlled block. The previous options' values are restored
+    once the block is exited. This is useful when making temporary adjustments
+    to pandas' behavior without affecting the global state.
+
+    Parameters
+    ----------
+    *args : str | object
+        An even amount of arguments provided in pairs which will be
+        interpreted as (pattern, value) pairs.
+
+    Returns
+    -------
+    None
+        No return value.
+
+    Yields
+    ------
+    None
+        No yield value.
+
+    See Also
+    --------
+    get_option : Retrieve the value of the specified option.
+    set_option : Set the value of the specified option.
+    reset_option : Reset one or more options to their default value.
+    describe_option : Print the description for one or more registered options.
+
+    Notes
+    -----
+    For all available options, please view the :ref:`User Guide <options.available>`
+    or use ``pandas.describe_option()``.
+
+    Examples
+    --------
+    >>> from pandas import option_context
+    >>> with option_context("display.max_rows", 10, "display.max_columns", 5):
+    ...     pass
+    """
+    if len(args) % 2 != 0 or len(args) < 2:
+        raise ValueError(
+            "Provide an even amount of arguments as "
+            "option_context(pat, val, pat, val...)."
+        )
+
+    ops = tuple(zip(args[::2], args[1::2]))
+    try:
+        undo = tuple((pat, get_option(pat)) for pat, val in ops)
+        for pat, val in ops:
+            set_option(pat, val)
+        yield
+    finally:
+        for pat, val in undo:
+            set_option(pat, val)
+
+
+def register_option(
+    key: str,
+    defval: object,
+    doc: str = "",
+    validator: Callable[[object], Any] | None = None,
+    cb: Callable[[str], Any] | None = None,
+) -> None:
+    """
+    Register an option in the package-wide pandas config object
+
+    Parameters
+    ----------
+    key : str
+        Fully-qualified key, e.g. "x.y.option - z".
+    defval : object
+        Default value of the option.
+    doc : str
+        Description of the option.
+    validator : Callable, optional
+        Function of a single argument, should raise `ValueError` if
+        called with a value which is not a legal value for the option.
+    cb
+        a function of a single argument "key", which is called
+        immediately after an option value is set/reset. key is
+        the full name of the option.
+
+    Raises
+    ------
+    ValueError if `validator` is specified and `defval` is not a valid value.
+
+    """
+    import keyword
+    import tokenize
+
+    key = key.lower()
+
+    if key in _registered_options:
+        raise OptionError(f"Option '{key}' has already been registered")
+    if key in _reserved_keys:
+        raise OptionError(f"Option '{key}' is a reserved key")
+
+    # the default value should be legal
+    if validator:
+        validator(defval)
+
+    # walk the nested dict, creating dicts as needed along the path
+    path = key.split(".")
+
+    for k in path:
+        if not re.match("^" + tokenize.Name + "$", k):
+            raise ValueError(f"{k} is not a valid identifier")
+        if keyword.iskeyword(k):
+            raise ValueError(f"{k} is a python keyword")
+
+    cursor = _global_config
+    msg = "Path prefix to option '{option}' is already an option"
+
+    for i, p in enumerate(path[:-1]):
+        if not isinstance(cursor, dict):
+            raise OptionError(msg.format(option=".".join(path[:i])))
+        if p not in cursor:
+            cursor[p] = {}
+        cursor = cursor[p]
+
+    if not isinstance(cursor, dict):
+        raise OptionError(msg.format(option=".".join(path[:-1])))
+
+    cursor[path[-1]] = defval  # initialize
+
+    # save the option metadata
+    _registered_options[key] = RegisteredOption(
+        key=key, defval=defval, doc=doc, validator=validator, cb=cb
+    )
+
+
+def deprecate_option(
+    key: str,
+    msg: str | None = None,
+    rkey: str | None = None,
+    removal_ver: str | None = None,
+) -> None:
+    """
+    Mark option `key` as deprecated, if code attempts to access this option,
+    a warning will be produced, using `msg` if given, or a default message
+    if not.
+    if `rkey` is given, any access to the key will be re-routed to `rkey`.
+
+    Neither the existence of `key` nor that if `rkey` is checked. If they
+    do not exist, any subsequence access will fail as usual, after the
+    deprecation warning is given.
+
+    Parameters
+    ----------
+    key : str
+        Name of the option to be deprecated.
+        must be a fully-qualified option name (e.g "x.y.z.rkey").
+    msg : str, optional
+        Warning message to output when the key is referenced.
+        if no message is given a default message will be emitted.
+    rkey : str, optional
+        Name of an option to reroute access to.
+        If specified, any referenced `key` will be
+        re-routed to `rkey` including set/get/reset.
+        rkey must be a fully-qualified option name (e.g "x.y.z.rkey").
+        used by the default message if no `msg` is specified.
+    removal_ver : str, optional
+        Specifies the version in which this option will
+        be removed. used by the default message if no `msg` is specified.
+
+    Raises
+    ------
+    OptionError
+        If the specified key has already been deprecated.
+    """
+    key = key.lower()
+
+    if key in _deprecated_options:
+        raise OptionError(f"Option '{key}' has already been defined as deprecated.")
+
+    _deprecated_options[key] = DeprecatedOption(key, msg, rkey, removal_ver)
+
+
+#
+# functions internal to the module
+
+
+def _select_options(pat: str) -> list[str]:
+    """
+    returns a list of keys matching `pat`
+
+    if pat=="all", returns all registered options
+    """
+    # short-circuit for exact key
+    if pat in _registered_options:
+        return [pat]
+
+    # else look through all of them
+    keys = sorted(_registered_options.keys())
+    if pat == "all":  # reserved key
+        return keys
+
+    return [k for k in keys if re.search(pat, k, re.I)]
+
+
+def _get_root(key: str) -> tuple[dict[str, Any], str]:
+    path = key.split(".")
+    cursor = _global_config
+    for p in path[:-1]:
+        cursor = cursor[p]
+    return cursor, path[-1]
+
+
+def _get_deprecated_option(key: str):
+    """
+    Retrieves the metadata for a deprecated option, if `key` is deprecated.
+
+    Returns
+    -------
+    DeprecatedOption (namedtuple) if key is deprecated, None otherwise
+    """
+    try:
+        d = _deprecated_options[key]
+    except KeyError:
+        return None
+    else:
+        return d
+
+
+def _get_registered_option(key: str):
+    """
+    Retrieves the option metadata if `key` is a registered option.
+
+    Returns
+    -------
+    RegisteredOption (namedtuple) if key is deprecated, None otherwise
+    """
+    return _registered_options.get(key)
+
+
+def _translate_key(key: str) -> str:
+    """
+    if key id deprecated and a replacement key defined, will return the
+    replacement key, otherwise returns `key` as - is
+    """
+    d = _get_deprecated_option(key)
+    if d:
+        return d.rkey or key
+    else:
+        return key
+
+
+def _warn_if_deprecated(key: str) -> bool:
+    """
+    Checks if `key` is a deprecated option and if so, prints a warning.
+
+    Returns
+    -------
+    bool - True if `key` is deprecated, False otherwise.
+    """
+    d = _get_deprecated_option(key)
+    if d:
+        if d.msg:
+            warnings.warn(
+                d.msg,
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        else:
+            msg = f"'{key}' is deprecated"
+            if d.removal_ver:
+                msg += f" and will be removed in {d.removal_ver}"
+            if d.rkey:
+                msg += f", please use '{d.rkey}' instead."
+            else:
+                msg += ", please refrain from using it."
+
+            warnings.warn(msg, FutureWarning, stacklevel=find_stack_level())
+        return True
+    return False
+
+
+def _build_option_description(k: str) -> str:
+    """Builds a formatted description of a registered option and prints it"""
+    o = _get_registered_option(k)
+    d = _get_deprecated_option(k)
+
+    s = f"{k} "
+
+    if o.doc:
+        s += "\n".join(o.doc.strip().split("\n"))
+    else:
+        s += "No description available."
+
+    if o:
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore", FutureWarning)
+            warnings.simplefilter("ignore", DeprecationWarning)
+            s += f"\n    [default: {o.defval}] [currently: {get_option(k)}]"
+
+    if d:
+        rkey = d.rkey or ""
+        s += "\n    (Deprecated"
+        s += f", use `{rkey}` instead."
+        s += ")"
+
+    return s
+
+
+# helpers
+
+
+@contextmanager
+def config_prefix(prefix: str) -> Generator[None]:
+    """
+    contextmanager for multiple invocations of API with a common prefix
+
+    supported API functions: (register / get / set )__option
+
+    Warning: This is not thread - safe, and won't work properly if you import
+    the API functions into your module using the "from x import y" construct.
+
+    Example
+    -------
+    import pandas._config.config as cf
+    with cf.config_prefix("display.font"):
+        cf.register_option("color", "red")
+        cf.register_option("size", " 5 pt")
+        cf.set_option(size, " 6 pt")
+        cf.get_option(size)
+        ...
+
+        etc'
+
+    will register options "display.font.color", "display.font.size", set the
+    value of "display.font.size"... and so on.
+    """
+    # Note: reset_option relies on set_option, and on key directly
+    # it does not fit in to this monkey-patching scheme
+
+    global register_option, get_option, set_option
+
+    def wrap(func: F) -> F:
+        def inner(key: str, *args, **kwds):
+            pkey = f"{prefix}.{key}"
+            return func(pkey, *args, **kwds)
+
+        return cast(F, inner)
+
+    _register_option = register_option
+    _get_option = get_option
+    _set_option = set_option
+    set_option = wrap(set_option)
+    get_option = wrap(get_option)
+    register_option = wrap(register_option)
+    try:
+        yield
+    finally:
+        set_option = _set_option
+        get_option = _get_option
+        register_option = _register_option
+
+
+# These factories and methods are handy for use as the validator
+# arg in register_option
+
+
+def is_type_factory(_type: type[Any]) -> Callable[[Any], None]:
+    """
+
+    Parameters
+    ----------
+    `_type` - a type to be compared against (e.g. type(x) == `_type`)
+
+    Returns
+    -------
+    validator - a function of a single argument x , which raises
+                ValueError if type(x) is not equal to `_type`
+
+    """
+
+    def inner(x) -> None:
+        if type(x) != _type:
+            raise ValueError(f"Value must have type '{_type}'")
+
+    return inner
+
+
+def is_instance_factory(_type: type | tuple[type, ...]) -> Callable[[Any], None]:
+    """
+
+    Parameters
+    ----------
+    `_type` - the type to be checked against
+
+    Returns
+    -------
+    validator - a function of a single argument x , which raises
+                ValueError if x is not an instance of `_type`
+
+    """
+    if isinstance(_type, tuple):
+        type_repr = "|".join(map(str, _type))
+    else:
+        type_repr = f"'{_type}'"
+
+    def inner(x) -> None:
+        if not isinstance(x, _type):
+            raise ValueError(f"Value must be an instance of {type_repr}")
+
+    return inner
+
+
+def is_one_of_factory(legal_values: Sequence) -> Callable[[Any], None]:
+    callables = [c for c in legal_values if callable(c)]
+    legal_values = [c for c in legal_values if not callable(c)]
+
+    def inner(x) -> None:
+        if x not in legal_values:
+            if not any(c(x) for c in callables):
+                uvals = [str(lval) for lval in legal_values]
+                pp_values = "|".join(uvals)
+                msg = f"Value must be one of {pp_values}"
+                if len(callables):
+                    msg += " or a callable"
+                raise ValueError(msg)
+
+    return inner
+
+
+def is_nonnegative_int(value: object) -> None:
+    """
+    Verify that value is None or a positive int.
+
+    Parameters
+    ----------
+    value : None or int
+            The `value` to be checked.
+
+    Raises
+    ------
+    ValueError
+        When the value is not None or is a negative integer
+    """
+    if value is None:
+        return
+
+    elif isinstance(value, int):
+        if value >= 0:
+            return
+
+    msg = "Value must be a nonnegative integer or None"
+    raise ValueError(msg)
+
+
+# common type validators, for convenience
+# usage: register_option(... , validator = is_int)
+is_int = is_type_factory(int)
+is_bool = is_type_factory(bool)
+is_float = is_type_factory(float)
+is_str = is_type_factory(str)
+is_text = is_instance_factory((str, bytes))
+
+
+def is_callable(obj: object) -> bool:
+    """
+
+    Parameters
+    ----------
+    `obj` - the object to be checked
+
+    Returns
+    -------
+    validator - returns True if object is callable
+        raises ValueError otherwise.
+
+    """
+    if not callable(obj):
+        raise ValueError("Value must be a callable")
+    return True
diff --git a/pandas/pandas/_config/dates.py b/pandas/pandas/_config/dates.py
new file mode 100644
index 0000000000000..2d9f5d390dc9c
--- /dev/null
+++ b/pandas/pandas/_config/dates.py
@@ -0,0 +1,26 @@
+"""
+config for datetime formatting
+"""
+
+from __future__ import annotations
+
+from pandas._config import config as cf
+
+pc_date_dayfirst_doc = """
+: boolean
+    When True, prints and parses dates with the day first, eg 20/01/2005
+"""
+
+pc_date_yearfirst_doc = """
+: boolean
+    When True, prints and parses dates with the year first, eg 2005/01/20
+"""
+
+with cf.config_prefix("display"):
+    # Needed upstream of `_libs` because these are used in tslibs.parsing
+    cf.register_option(
+        "date_dayfirst", False, pc_date_dayfirst_doc, validator=cf.is_bool
+    )
+    cf.register_option(
+        "date_yearfirst", False, pc_date_yearfirst_doc, validator=cf.is_bool
+    )
diff --git a/pandas/pandas/_config/display.py b/pandas/pandas/_config/display.py
new file mode 100644
index 0000000000000..df2c3ad36c855
--- /dev/null
+++ b/pandas/pandas/_config/display.py
@@ -0,0 +1,62 @@
+"""
+Unopinionated display configuration.
+"""
+
+from __future__ import annotations
+
+import locale
+import sys
+
+from pandas._config import config as cf
+
+# -----------------------------------------------------------------------------
+# Global formatting options
+_initial_defencoding: str | None = None
+
+
+def detect_console_encoding() -> str:
+    """
+    Try to find the most capable encoding supported by the console.
+    slightly modified from the way IPython handles the same issue.
+    """
+    global _initial_defencoding
+
+    encoding = None
+    try:
+        encoding = sys.stdout.encoding or sys.stdin.encoding
+    except (AttributeError, OSError):
+        pass
+
+    # try again for something better
+    if not encoding or "ascii" in encoding.lower():
+        try:
+            encoding = locale.getpreferredencoding()
+        except locale.Error:
+            # can be raised by locale.setlocale(), which is
+            #  called by getpreferredencoding
+            #  (on some systems, see stdlib locale docs)
+            pass
+
+    # when all else fails. this will usually be "ascii"
+    if not encoding or "ascii" in encoding.lower():
+        encoding = sys.getdefaultencoding()
+
+    # GH#3360, save the reported defencoding at import time
+    # MPL backends may change it. Make available for debugging.
+    if not _initial_defencoding:
+        _initial_defencoding = sys.getdefaultencoding()
+
+    return encoding
+
+
+pc_encoding_doc = """
+: str/unicode
+    Defaults to the detected encoding of the console.
+    Specifies the encoding to be used for strings returned by to_string,
+    these are generally strings meant to be displayed on the console.
+"""
+
+with cf.config_prefix("display"):
+    cf.register_option(
+        "encoding", detect_console_encoding(), pc_encoding_doc, validator=cf.is_text
+    )
diff --git a/pandas/pandas/_config/localization.py b/pandas/pandas/_config/localization.py
new file mode 100644
index 0000000000000..6602633f20399
--- /dev/null
+++ b/pandas/pandas/_config/localization.py
@@ -0,0 +1,176 @@
+"""
+Helpers for configuring locale settings.
+
+Name `localization` is chosen to avoid overlap with builtin `locale` module.
+"""
+
+from __future__ import annotations
+
+from contextlib import contextmanager
+import locale
+import platform
+import re
+import subprocess
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+from pandas._config.config import options
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+
+@contextmanager
+def set_locale(
+    new_locale: str | tuple[str, str], lc_var: int = locale.LC_ALL
+) -> Generator[str | tuple[str, str]]:
+    """
+    Context manager for temporarily setting a locale.
+
+    Parameters
+    ----------
+    new_locale : str or tuple
+        A string of the form <language_country>.<encoding>. For example to set
+        the current locale to US English with a UTF8 encoding, you would pass
+        "en_US.UTF-8".
+    lc_var : int, default `locale.LC_ALL`
+        The category of the locale being set.
+
+    Notes
+    -----
+    This is useful when you want to run a particular block of code under a
+    particular locale, without globally setting the locale. This probably isn't
+    thread-safe.
+    """
+    # getlocale is not always compliant with setlocale, use setlocale. GH#46595
+    current_locale = locale.setlocale(lc_var)
+
+    try:
+        locale.setlocale(lc_var, new_locale)
+        normalized_code, normalized_encoding = locale.getlocale()
+        if normalized_code is not None and normalized_encoding is not None:
+            yield f"{normalized_code}.{normalized_encoding}"
+        else:
+            yield new_locale
+    finally:
+        locale.setlocale(lc_var, current_locale)
+
+
+def can_set_locale(lc: str, lc_var: int = locale.LC_ALL) -> bool:
+    """
+    Check to see if we can set a locale, and subsequently get the locale,
+    without raising an Exception.
+
+    Parameters
+    ----------
+    lc : str
+        The locale to attempt to set.
+    lc_var : int, default `locale.LC_ALL`
+        The category of the locale being set.
+
+    Returns
+    -------
+    bool
+        Whether the passed locale can be set
+    """
+    try:
+        with set_locale(lc, lc_var=lc_var):
+            pass
+    except (ValueError, locale.Error):
+        # horrible name for a Exception subclass
+        return False
+    else:
+        return True
+
+
+def _valid_locales(locales: list[str] | str, normalize: bool) -> list[str]:
+    """
+    Return a list of normalized locales that do not throw an ``Exception``
+    when set.
+
+    Parameters
+    ----------
+    locales : str
+        A string where each locale is separated by a newline.
+    normalize : bool
+        Whether to call ``locale.normalize`` on each locale.
+
+    Returns
+    -------
+    valid_locales : list
+        A list of valid locales.
+    """
+    return [
+        loc
+        for loc in (
+            locale.normalize(loc.strip()) if normalize else loc.strip()
+            for loc in locales
+        )
+        if can_set_locale(loc)
+    ]
+
+
+def get_locales(
+    prefix: str | None = None,
+    normalize: bool = True,
+) -> list[str]:
+    """
+    Get all the locales that are available on the system.
+
+    Parameters
+    ----------
+    prefix : str
+        If not ``None`` then return only those locales with the prefix
+        provided. For example to get all English language locales (those that
+        start with ``"en"``), pass ``prefix="en"``.
+    normalize : bool
+        Call ``locale.normalize`` on the resulting list of available locales.
+        If ``True``, only locales that can be set without throwing an
+        ``Exception`` are returned.
+
+    Returns
+    -------
+    locales : list of strings
+        A list of locale strings that can be set with ``locale.setlocale()``.
+        For example::
+
+            locale.setlocale(locale.LC_ALL, locale_string)
+
+    On error will return an empty list (no locale available, e.g. Windows)
+
+    """
+    if platform.system() in ("Linux", "Darwin"):
+        raw_locales = subprocess.check_output(["locale", "-a"])
+    else:
+        # Other platforms e.g. windows platforms don't define "locale -a"
+        #  Note: is_platform_windows causes circular import here
+        return []
+
+    try:
+        # raw_locales is "\n" separated list of locales
+        # it may contain non-decodable parts, so split
+        # extract what we can and then rejoin.
+        split_raw_locales = raw_locales.split(b"\n")
+        out_locales = []
+        for x in split_raw_locales:
+            try:
+                out_locales.append(str(x, encoding=cast(str, options.display.encoding)))
+            except UnicodeError:
+                # 'locale -a' is used to populated 'raw_locales' and on
+                # Redhat 7 Linux (and maybe others) prints locale names
+                # using windows-1252 encoding.  Bug only triggered by
+                # a few special characters and when there is an
+                # extensive list of installed locales.
+                out_locales.append(str(x, encoding="windows-1252"))
+
+    except TypeError:
+        pass
+
+    if prefix is None:
+        return _valid_locales(out_locales, normalize)
+
+    pattern = re.compile(f"{prefix}.*")
+    found = pattern.findall("\n".join(out_locales))
+    return _valid_locales(found, normalize)
diff --git a/pandas/pandas/_libs/__init__.py b/pandas/pandas/_libs/__init__.py
new file mode 100644
index 0000000000000..d499f9a6cd75e
--- /dev/null
+++ b/pandas/pandas/_libs/__init__.py
@@ -0,0 +1,27 @@
+__all__ = [
+    "Interval",
+    "NaT",
+    "NaTType",
+    "OutOfBoundsDatetime",
+    "Period",
+    "Timedelta",
+    "Timestamp",
+    "iNaT",
+]
+
+
+# Below imports needs to happen first to ensure pandas top level
+# module gets monkeypatched with the pandas_datetime_CAPI
+# see pandas_datetime_exec in pd_datetime.c
+import pandas._libs.pandas_parser  # isort: skip # type: ignore[reportUnusedImport]
+import pandas._libs.pandas_datetime  # noqa: F401 # isort: skip # type: ignore[reportUnusedImport]
+from pandas._libs.interval import Interval
+from pandas._libs.tslibs import (
+    NaT,
+    NaTType,
+    OutOfBoundsDatetime,
+    Period,
+    Timedelta,
+    Timestamp,
+    iNaT,
+)
diff --git a/pandas/pandas/_libs/algos.pxd b/pandas/pandas/_libs/algos.pxd
new file mode 100644
index 0000000000000..3f9ddeff99059
--- /dev/null
+++ b/pandas/pandas/_libs/algos.pxd
@@ -0,0 +1,22 @@
+from pandas._libs.dtypes cimport (
+    numeric_object_t,
+    numeric_t,
+)
+
+
+cdef numeric_t kth_smallest_c(numeric_t* arr, Py_ssize_t k, Py_ssize_t n) noexcept nogil
+
+cdef enum TiebreakEnumType:
+    TIEBREAK_AVERAGE
+    TIEBREAK_MIN,
+    TIEBREAK_MAX
+    TIEBREAK_FIRST
+    TIEBREAK_FIRST_DESCENDING
+    TIEBREAK_DENSE
+
+
+cdef numeric_object_t get_rank_nan_fill_val(
+    bint rank_nans_highest,
+    numeric_object_t val,
+    bint is_datetimelike=*,
+)
diff --git a/pandas/pandas/_libs/algos.pyi b/pandas/pandas/_libs/algos.pyi
new file mode 100644
index 0000000000000..0a6be851e1efd
--- /dev/null
+++ b/pandas/pandas/_libs/algos.pyi
@@ -0,0 +1,443 @@
+from typing import Any
+
+import numpy as np
+
+from pandas._typing import npt
+
+class Infinity:
+    def __eq__(self, other) -> bool: ...
+    def __ne__(self, other) -> bool: ...
+    def __lt__(self, other) -> bool: ...
+    def __le__(self, other) -> bool: ...
+    def __gt__(self, other) -> bool: ...
+    def __ge__(self, other) -> bool: ...
+
+class NegInfinity:
+    def __eq__(self, other) -> bool: ...
+    def __ne__(self, other) -> bool: ...
+    def __lt__(self, other) -> bool: ...
+    def __le__(self, other) -> bool: ...
+    def __gt__(self, other) -> bool: ...
+    def __ge__(self, other) -> bool: ...
+
+def unique_deltas(
+    arr: np.ndarray,  # const int64_t[:]
+) -> np.ndarray: ...  # np.ndarray[np.int64, ndim=1]
+def is_lexsorted(list_of_arrays: list[npt.NDArray[np.int64]]) -> bool: ...
+def groupsort_indexer(
+    index: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+) -> tuple[
+    np.ndarray,  # ndarray[int64_t, ndim=1]
+    np.ndarray,  # ndarray[int64_t, ndim=1]
+]: ...
+def kth_smallest(
+    arr: np.ndarray,  # numeric[:]
+    k: int,
+) -> Any: ...  # numeric
+
+# ----------------------------------------------------------------------
+# Pairwise correlation/covariance
+
+def nancorr(
+    mat: npt.NDArray[np.float64],  # const float64_t[:, :]
+    cov: bool = ...,
+    minp: int | None = ...,
+) -> npt.NDArray[np.float64]: ...  # ndarray[float64_t, ndim=2]
+def nancorr_spearman(
+    mat: npt.NDArray[np.float64],  # ndarray[float64_t, ndim=2]
+    minp: int = ...,
+) -> npt.NDArray[np.float64]: ...  # ndarray[float64_t, ndim=2]
+
+# ----------------------------------------------------------------------
+
+def validate_limit(nobs: int | None, limit=...) -> int: ...
+def get_fill_indexer(
+    mask: npt.NDArray[np.bool_],
+    limit: int | None = None,
+) -> npt.NDArray[np.intp]: ...
+def pad(
+    old: np.ndarray,  # ndarray[numeric_object_t]
+    new: np.ndarray,  # ndarray[numeric_object_t]
+    limit=...,
+) -> npt.NDArray[np.intp]: ...  # np.ndarray[np.intp, ndim=1]
+def pad_inplace(
+    values: np.ndarray,  # numeric_object_t[:]
+    mask: np.ndarray,  # uint8_t[:]
+    limit=...,
+) -> None: ...
+def pad_2d_inplace(
+    values: np.ndarray,  # numeric_object_t[:, :]
+    mask: np.ndarray,  # const uint8_t[:, :]
+    limit=...,
+) -> None: ...
+def backfill(
+    old: np.ndarray,  # ndarray[numeric_object_t]
+    new: np.ndarray,  # ndarray[numeric_object_t]
+    limit=...,
+) -> npt.NDArray[np.intp]: ...  # np.ndarray[np.intp, ndim=1]
+def backfill_inplace(
+    values: np.ndarray,  # numeric_object_t[:]
+    mask: np.ndarray,  # uint8_t[:]
+    limit=...,
+) -> None: ...
+def backfill_2d_inplace(
+    values: np.ndarray,  # numeric_object_t[:, :]
+    mask: np.ndarray,  # const uint8_t[:, :]
+    limit=...,
+) -> None: ...
+def is_monotonic(
+    arr: np.ndarray,  # ndarray[numeric_object_t, ndim=1]
+    timelike: bool,
+) -> tuple[bool, bool, bool]: ...
+
+# ----------------------------------------------------------------------
+# rank_1d, rank_2d
+# ----------------------------------------------------------------------
+
+def rank_1d(
+    values: np.ndarray,  # ndarray[numeric_object_t, ndim=1]
+    labels: np.ndarray | None = ...,  # const int64_t[:]=None
+    is_datetimelike: bool = ...,
+    ties_method=...,
+    ascending: bool = ...,
+    pct: bool = ...,
+    na_option=...,
+    mask: npt.NDArray[np.bool_] | None = ...,
+) -> np.ndarray: ...  # np.ndarray[float64_t, ndim=1]
+def rank_2d(
+    in_arr: np.ndarray,  # ndarray[numeric_object_t, ndim=2]
+    axis: int = ...,
+    is_datetimelike: bool = ...,
+    ties_method=...,
+    ascending: bool = ...,
+    na_option=...,
+    pct: bool = ...,
+) -> np.ndarray: ...  # np.ndarray[float64_t, ndim=1]
+def diff_2d(
+    arr: np.ndarray,  # ndarray[diff_t, ndim=2]
+    out: np.ndarray,  # ndarray[out_t, ndim=2]
+    periods: int,
+    axis: int,
+    datetimelike: bool = ...,
+) -> None: ...
+def ensure_platform_int(arr: object) -> npt.NDArray[np.intp]: ...
+def ensure_object(arr: object) -> npt.NDArray[np.object_]: ...
+def ensure_float64(arr: object) -> npt.NDArray[np.float64]: ...
+def ensure_int8(arr: object) -> npt.NDArray[np.int8]: ...
+def ensure_int16(arr: object) -> npt.NDArray[np.int16]: ...
+def ensure_int32(arr: object) -> npt.NDArray[np.int32]: ...
+def ensure_int64(arr: object) -> npt.NDArray[np.int64]: ...
+def ensure_uint64(arr: object) -> npt.NDArray[np.uint64]: ...
+def take_1d_int8_int8(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int8_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int8_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int8_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int16_int16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int16_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int16_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int16_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int32_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int32_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int64_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_uint16_uint16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_uint32_uint32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_uint64_uint64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_int64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_float32_float32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_float32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_float64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_object_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_bool_bool(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_1d_bool_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int8_int8(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int8_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int8_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int8_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int16_int16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int16_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int16_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int16_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int32_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int32_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int64_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_int64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_uint16_uint16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_uint32_uint32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_uint64_uint64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_float32_float32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_float32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_float64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_object_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_bool_bool(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis0_bool_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int8_int8(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int8_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int8_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int8_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int16_int16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int16_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int16_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int16_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int32_int32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int32_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int64_int64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_uint16_uint16(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_uint32_uint32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_uint64_uint64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_int64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_float32_float32(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_float32_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_float64_float64(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_object_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_bool_bool(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_axis1_bool_object(
+    values: np.ndarray, indexer: npt.NDArray[np.intp], out: np.ndarray, fill_value=...
+) -> None: ...
+def take_2d_multi_int8_int8(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int8_int32(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int8_int64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int8_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int16_int16(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int16_int32(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int16_int64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int16_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int32_int32(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int32_int64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int32_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int64_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_float32_float32(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_float32_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_float64_float64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_object_object(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_bool_bool(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_bool_object(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
+def take_2d_multi_int64_int64(
+    values: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value=...,
+) -> None: ...
diff --git a/pandas/pandas/_libs/algos.pyx b/pandas/pandas/_libs/algos.pyx
new file mode 100644
index 0000000000000..60ee73ef6b43f
--- /dev/null
+++ b/pandas/pandas/_libs/algos.pyx
@@ -0,0 +1,1451 @@
+cimport cython
+from cython cimport Py_ssize_t
+from libc.math cimport (
+    fabs,
+    sqrt,
+)
+from libc.stdlib cimport (
+    free,
+    malloc,
+)
+from libc.string cimport memmove
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    NPY_FLOAT64,
+    NPY_INT8,
+    NPY_INT16,
+    NPY_INT32,
+    NPY_INT64,
+    NPY_OBJECT,
+    NPY_UINT64,
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    intp_t,
+    ndarray,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+
+cnp.import_array()
+
+cimport pandas._libs.util as util
+from pandas._libs.dtypes cimport (
+    numeric_object_t,
+    numeric_t,
+)
+from pandas._libs.khash cimport (
+    kh_destroy_int64,
+    kh_get_int64,
+    kh_init_int64,
+    kh_int64_t,
+    kh_put_int64,
+    kh_resize_int64,
+    khiter_t,
+)
+from pandas._libs.missing cimport (
+    checknull,
+    isnaobj,
+)
+from pandas._libs.util cimport get_nat
+
+cdef:
+    float64_t FP_ERR = 1e-13
+    float64_t NaN = <float64_t>np.nan
+    int64_t NPY_NAT = get_nat()
+
+
+tiebreakers = {
+    "average": TIEBREAK_AVERAGE,
+    "min": TIEBREAK_MIN,
+    "max": TIEBREAK_MAX,
+    "first": TIEBREAK_FIRST,
+    "dense": TIEBREAK_DENSE,
+}
+
+
+cdef bint are_diff(object left, object right):
+    try:
+        return fabs(left - right) > FP_ERR
+    except TypeError:
+        return left != right
+
+
+class Infinity:
+    """
+    Provide a positive Infinity comparison method for ranking.
+    """
+    def __lt__(self, other):
+        return False
+
+    def __le__(self, other):
+        return isinstance(other, Infinity)
+
+    def __eq__(self, other):
+        return isinstance(other, Infinity)
+
+    def __ne__(self, other):
+        return not isinstance(other, Infinity)
+
+    def __gt__(self, other):
+        return (not isinstance(other, Infinity) and
+                not checknull(other))
+
+    def __ge__(self, other):
+        return not checknull(other)
+
+
+class NegInfinity:
+    """
+    Provide a negative Infinity comparison method for ranking.
+    """
+    def __lt__(self, other):
+        return  (not isinstance(other, NegInfinity) and
+                 not checknull(other))
+
+    def __le__(self, other):
+        return not checknull(other)
+
+    def __eq__(self, other):
+        return isinstance(other, NegInfinity)
+
+    def __ne__(self, other):
+        return not isinstance(other, NegInfinity)
+
+    def __gt__(self, other):
+        return False
+
+    def __ge__(self, other):
+        return isinstance(other, NegInfinity)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef ndarray[int64_t, ndim=1] unique_deltas(const int64_t[:] arr):
+    """
+    Efficiently find the unique first-differences of the given array.
+
+    Parameters
+    ----------
+    arr : ndarray[int64_t]
+
+    Returns
+    -------
+    ndarray[int64_t]
+        An ordered ndarray[int64_t]
+    """
+    cdef:
+        Py_ssize_t i, n = len(arr)
+        int64_t val
+        khiter_t k
+        kh_int64_t *table
+        int ret = 0
+        list uniques = []
+        ndarray[int64_t, ndim=1] result
+
+    table = kh_init_int64()
+    kh_resize_int64(table, 10)
+    for i in range(n - 1):
+        val = arr[i + 1] - arr[i]
+        k = kh_get_int64(table, val)
+        if k == table.n_buckets:
+            kh_put_int64(table, val, &ret)
+            uniques.append(val)
+    kh_destroy_int64(table)
+
+    result = np.array(uniques, dtype=np.int64)
+    result.sort()
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_lexsorted(list_of_arrays: list) -> bool:
+    cdef:
+        Py_ssize_t i
+        Py_ssize_t n, nlevels
+        int64_t k, cur, pre
+        ndarray arr
+        bint result = True
+
+    nlevels = len(list_of_arrays)
+    n = len(list_of_arrays[0])
+
+    cdef int64_t **vecs = <int64_t**>malloc(nlevels * sizeof(int64_t*))
+    if vecs is NULL:
+        raise MemoryError()
+    for i in range(nlevels):
+        arr = list_of_arrays[i]
+        assert arr.dtype.name == "int64"
+        vecs[i] = <int64_t*>cnp.PyArray_DATA(arr)
+
+    # Assume uniqueness??
+    with nogil:
+        for i in range(1, n):
+            for k in range(nlevels):
+                cur = vecs[k][i]
+                pre = vecs[k][i -1]
+                if cur == pre:
+                    continue
+                elif cur > pre:
+                    break
+                else:
+                    result = False
+                    break
+            if not result:
+                break
+    free(vecs)
+    return result
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def groupsort_indexer(const intp_t[:] index, Py_ssize_t ngroups):
+    """
+    Compute a 1-d indexer.
+
+    The indexer is an ordering of the passed index,
+    ordered by the groups.
+
+    Parameters
+    ----------
+    index: np.ndarray[np.intp]
+        Mappings from group -> position.
+    ngroups: int64
+        Number of groups.
+
+    Returns
+    -------
+    ndarray[intp_t, ndim=1]
+        Indexer
+    ndarray[intp_t, ndim=1]
+        Group Counts
+
+    Notes
+    -----
+    This is a reverse of the label factorization process.
+    """
+    cdef:
+        Py_ssize_t i, label, n
+        intp_t[::1] indexer, where, counts
+
+    counts = np.zeros(ngroups + 1, dtype=np.intp)
+    n = len(index)
+    indexer = np.zeros(n, dtype=np.intp)
+    where = np.zeros(ngroups + 1, dtype=np.intp)
+
+    with nogil:
+
+        # count group sizes, location 0 for NA
+        for i in range(n):
+            counts[index[i] + 1] += 1
+
+        # mark the start of each contiguous group of like-indexed data
+        for i in range(1, ngroups + 1):
+            where[i] = where[i - 1] + counts[i - 1]
+
+        # this is our indexer
+        for i in range(n):
+            label = index[i] + 1
+            indexer[where[label]] = i
+            where[label] += 1
+
+    return indexer.base, counts.base
+
+
+cdef Py_ssize_t swap(numeric_t *a, numeric_t *b) noexcept nogil:
+    cdef:
+        numeric_t t
+
+    # cython doesn't allow pointer dereference so use array syntax
+    t = a[0]
+    a[0] = b[0]
+    b[0] = t
+    return 0
+
+
+cdef numeric_t kth_smallest_c(numeric_t* arr,
+                              Py_ssize_t k, Py_ssize_t n) noexcept nogil:
+    """
+    See kth_smallest.__doc__. The additional parameter n specifies the maximum
+    number of elements considered in arr, needed for compatibility with usage
+    in groupby.pyx
+    """
+    cdef:
+        Py_ssize_t i, j, left, m
+        numeric_t x
+
+    left = 0
+    m = n - 1
+
+    while left < m:
+        x = arr[k]
+        i = left
+        j = m
+
+        while 1:
+            while arr[i] < x:
+                i += 1
+            while x < arr[j]:
+                j -= 1
+            if i <= j:
+                swap(&arr[i], &arr[j])
+                i += 1
+                j -= 1
+
+            if i > j:
+                break
+
+        if j < k:
+            left = i
+        if k < i:
+            m = j
+    return arr[k]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def kth_smallest(numeric_t[::1] arr, Py_ssize_t k) -> numeric_t:
+    """
+    Compute the kth smallest value in arr. Note that the input
+    array will be modified.
+
+    Parameters
+    ----------
+    arr : numeric[::1]
+        Array to compute the kth smallest value for, must be
+        contiguous
+    k : Py_ssize_t
+
+    Returns
+    -------
+    numeric
+        The kth smallest value in arr
+    """
+    cdef:
+        numeric_t result
+
+    with nogil:
+        result = kth_smallest_c(&arr[0], k, arr.shape[0])
+
+    return result
+
+
+# ----------------------------------------------------------------------
+# Pairwise correlation/covariance
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+@cython.cdivision(True)
+def nancorr(const float64_t[:, :] mat, bint cov=False, minp=None):
+    cdef:
+        Py_ssize_t i, xi, yi, N, K
+        int64_t minpv
+        float64_t[:, ::1] result
+        uint8_t[:, :] mask
+        int64_t nobs = 0
+        float64_t vx, vy, dx, dy, meanx, meany, divisor, ssqdmx, ssqdmy, covxy
+
+    N, K = (<object>mat).shape
+
+    if minp is None:
+        minpv = 1
+    else:
+        minpv = <int64_t>minp
+
+    result = np.empty((K, K), dtype=np.float64)
+    mask = np.isfinite(mat).view(np.uint8)
+
+    with nogil:
+        for xi in range(K):
+            for yi in range(xi + 1):
+                # Welford's method for the variance-calculation
+                # https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+                nobs = ssqdmx = ssqdmy = covxy = meanx = meany = 0
+                for i in range(N):
+                    if mask[i, xi] and mask[i, yi]:
+                        vx = mat[i, xi]
+                        vy = mat[i, yi]
+                        nobs += 1
+                        dx = vx - meanx
+                        dy = vy - meany
+                        meanx += 1. / nobs * dx
+                        meany += 1. / nobs * dy
+                        ssqdmx += (vx - meanx) * dx
+                        ssqdmy += (vy - meany) * dy
+                        covxy += (vx - meanx) * dy
+
+                if nobs < minpv:
+                    result[xi, yi] = result[yi, xi] = NaN
+                else:
+                    divisor = (nobs - 1.0) if cov else sqrt(ssqdmx * ssqdmy)
+
+                    if divisor != 0:
+                        result[xi, yi] = result[yi, xi] = covxy / divisor
+                    else:
+                        result[xi, yi] = result[yi, xi] = NaN
+
+    return result.base
+
+# ----------------------------------------------------------------------
+# Pairwise Spearman correlation
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def nancorr_spearman(ndarray[float64_t, ndim=2] mat, Py_ssize_t minp=1) -> ndarray:
+    cdef:
+        Py_ssize_t i, xi, yi, N, K
+        ndarray[float64_t, ndim=2] result
+        ndarray[float64_t, ndim=2] ranked_mat
+        float64_t[::1] rankedx, rankedy
+        float64_t[::1] maskedx, maskedy
+        ndarray[uint8_t, ndim=2] mask
+        int64_t nobs = 0
+        bint no_nans
+        float64_t vx, vy, sumx, sumxx, sumyy, mean, divisor
+
+    N, K = (<object>mat).shape
+
+    # Handle the edge case where we know all results will be nan
+    # to keep conditional logic inside loop simpler
+    if N < minp:
+        result = np.full((K, K), np.nan, dtype=np.float64)
+        return result
+
+    result = np.empty((K, K), dtype=np.float64)
+    mask = np.isfinite(mat).view(np.uint8)
+    no_nans = mask.all()
+
+    ranked_mat = np.empty((N, K), dtype=np.float64)
+
+    # Note: we index into maskedx, maskedy in loops up to nobs, but using N is safe
+    # here since N >= nobs and values are stored contiguously
+    maskedx = np.empty(N, dtype=np.float64)
+    maskedy = np.empty(N, dtype=np.float64)
+    for i in range(K):
+        ranked_mat[:, i] = rank_1d(mat[:, i])
+
+    with nogil:
+        for xi in range(K):
+            for yi in range(xi + 1):
+                sumx = sumxx = sumyy = 0
+
+                # Fastpath for data with no nans/infs, allows avoiding mask checks
+                # and array reassignments
+                if no_nans:
+                    mean = (N + 1) / 2.
+
+                    # now the cov numerator
+                    for i in range(N):
+                        vx = ranked_mat[i, xi] - mean
+                        vy = ranked_mat[i, yi] - mean
+
+                        sumx += vx * vy
+                        sumxx += vx * vx
+                        sumyy += vy * vy
+                else:
+                    nobs = 0
+                    # Keep track of whether we need to recompute ranks
+                    all_ranks = True
+                    for i in range(N):
+                        all_ranks &= not (mask[i, xi] ^ mask[i, yi])
+                        if mask[i, xi] and mask[i, yi]:
+                            maskedx[nobs] = ranked_mat[i, xi]
+                            maskedy[nobs] = ranked_mat[i, yi]
+                            nobs += 1
+
+                    if nobs < minp:
+                        result[xi, yi] = result[yi, xi] = NaN
+                        continue
+                    else:
+                        if not all_ranks:
+                            with gil:
+                                # We need to slice back to nobs because rank_1d will
+                                # require arrays of nobs length
+                                rankedx = rank_1d(np.asarray(maskedx)[:nobs])
+                                rankedy = rank_1d(np.asarray(maskedy)[:nobs])
+                            for i in range(nobs):
+                                maskedx[i] = rankedx[i]
+                                maskedy[i] = rankedy[i]
+
+                        mean = (nobs + 1) / 2.
+
+                        # now the cov numerator
+                        for i in range(nobs):
+                            vx = maskedx[i] - mean
+                            vy = maskedy[i] - mean
+
+                            sumx += vx * vy
+                            sumxx += vx * vx
+                            sumyy += vy * vy
+
+                divisor = sqrt(sumxx * sumyy)
+
+                if divisor != 0:
+                    result[xi, yi] = result[yi, xi] = sumx / divisor
+                else:
+                    result[xi, yi] = result[yi, xi] = NaN
+
+    return result
+
+
+# ----------------------------------------------------------------------
+
+def validate_limit(nobs: int | None, limit=None) -> int:
+    """
+    Check that the `limit` argument is a positive integer.
+
+    Parameters
+    ----------
+    nobs : int
+    limit : object
+
+    Returns
+    -------
+    int
+        The limit.
+    """
+    if limit is None:
+        lim = nobs
+    else:
+        if not util.is_integer_object(limit):
+            raise ValueError("Limit must be an integer")
+        if limit < 1:
+            raise ValueError("Limit must be greater than 0")
+        lim = limit
+
+    return lim
+
+
+# TODO: overlap with libgroupby.group_fillna_indexer?
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def get_fill_indexer(const uint8_t[:] mask, limit=None):
+    """
+    Find an indexer to use for ffill to `take` on the array being filled.
+    """
+    cdef:
+        ndarray[intp_t, ndim=1] indexer
+        Py_ssize_t i, N = len(mask), last_valid
+        int lim
+
+        # fill_count is the number of consecutive NAs we have seen.
+        #  If it exceeds the given limit, we stop padding.
+        int fill_count = 0
+
+    lim = validate_limit(N, limit)
+    indexer = np.empty(N, dtype=np.intp)
+
+    last_valid = -1  # haven't yet seen anything non-NA
+
+    for i in range(N):
+        if not mask[i]:
+            indexer[i] = i
+            last_valid = i
+            fill_count = 0
+        else:
+            if fill_count < lim:
+                indexer[i] = last_valid
+            else:
+                indexer[i] = -1
+            fill_count += 1
+
+    return indexer
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def pad(
+    const numeric_object_t[:] old,
+    const numeric_object_t[:] new,
+    limit=None
+) -> ndarray:
+    # -> ndarray[intp_t, ndim=1]
+    cdef:
+        Py_ssize_t i, j, nleft, nright
+        ndarray[intp_t, ndim=1] indexer
+        numeric_object_t cur, next_val
+        int lim, fill_count = 0
+
+    nleft = len(old)
+    nright = len(new)
+    indexer = np.empty(nright, dtype=np.intp)
+    indexer[:] = -1
+
+    lim = validate_limit(nright, limit)
+
+    if nleft == 0 or nright == 0 or new[nright - 1] < old[0]:
+        return indexer
+
+    i = j = 0
+
+    cur = old[0]
+
+    while j <= nright - 1 and new[j] < cur:
+        j += 1
+
+    while True:
+        if j == nright:
+            break
+
+        if i == nleft - 1:
+            while j < nright:
+                if new[j] == cur:
+                    indexer[j] = i
+                elif new[j] > cur and fill_count < lim:
+                    indexer[j] = i
+                    fill_count += 1
+                j += 1
+            break
+
+        next_val = old[i + 1]
+
+        while j < nright and cur <= new[j] < next_val:
+            if new[j] == cur:
+                indexer[j] = i
+            elif fill_count < lim:
+                indexer[j] = i
+                fill_count += 1
+            j += 1
+
+        fill_count = 0
+        i += 1
+        cur = next_val
+
+    return indexer
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def pad_inplace(numeric_object_t[:] values, uint8_t[:] mask, limit=None):
+    cdef:
+        Py_ssize_t i, N
+        numeric_object_t val
+        uint8_t prev_mask
+        int lim, fill_count = 0
+
+    N = len(values)
+
+    # GH#2778
+    if N == 0:
+        return
+
+    lim = validate_limit(N, limit)
+
+    val = values[0]
+    prev_mask = mask[0]
+    for i in range(N):
+        if mask[i]:
+            if fill_count >= lim:
+                continue
+            fill_count += 1
+            values[i] = val
+            mask[i] = prev_mask
+        else:
+            fill_count = 0
+            val = values[i]
+            prev_mask = mask[i]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def pad_2d_inplace(numeric_object_t[:, :] values, uint8_t[:, :] mask, limit=None):
+    cdef:
+        Py_ssize_t i, j, N, K
+        numeric_object_t val
+        int lim, fill_count = 0
+
+    K, N = (<object>values).shape
+
+    # GH#2778
+    if N == 0:
+        return
+
+    lim = validate_limit(N, limit)
+
+    for j in range(K):
+        fill_count = 0
+        val = values[j, 0]
+        for i in range(N):
+            if mask[j, i]:
+                if fill_count >= lim or i == 0:
+                    continue
+                fill_count += 1
+                values[j, i] = val
+                mask[j, i] = False
+            else:
+                fill_count = 0
+                val = values[j, i]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def backfill(
+    const numeric_object_t[:] old,
+    const numeric_object_t[:] new,
+    limit=None
+) -> ndarray:  # -> ndarray[intp_t, ndim=1]
+    """
+    Backfilling logic for generating fill vector
+
+    Diagram of what's going on
+
+    Old      New    Fill vector    Mask
+            .        0               1
+            .        0               1
+            .        0               1
+    A        A        0               1
+            .        1               1
+            .        1               1
+            .        1               1
+            .        1               1
+            .        1               1
+    B        B        1               1
+            .        2               1
+            .        2               1
+            .        2               1
+    C        C        2               1
+            .                        0
+            .                        0
+    D
+    """
+    cdef:
+        Py_ssize_t i, j, nleft, nright
+        ndarray[intp_t, ndim=1] indexer
+        numeric_object_t cur, prev
+        int lim, fill_count = 0
+
+    nleft = len(old)
+    nright = len(new)
+    indexer = np.empty(nright, dtype=np.intp)
+    indexer[:] = -1
+
+    lim = validate_limit(nright, limit)
+
+    if nleft == 0 or nright == 0 or new[0] > old[nleft - 1]:
+        return indexer
+
+    i = nleft - 1
+    j = nright - 1
+
+    cur = old[nleft - 1]
+
+    while j >= 0 and new[j] > cur:
+        j -= 1
+
+    while True:
+        if j < 0:
+            break
+
+        if i == 0:
+            while j >= 0:
+                if new[j] == cur:
+                    indexer[j] = i
+                elif new[j] < cur and fill_count < lim:
+                    indexer[j] = i
+                    fill_count += 1
+                j -= 1
+            break
+
+        prev = old[i - 1]
+
+        while j >= 0 and prev < new[j] <= cur:
+            if new[j] == cur:
+                indexer[j] = i
+            elif new[j] < cur and fill_count < lim:
+                indexer[j] = i
+                fill_count += 1
+            j -= 1
+
+        fill_count = 0
+        i -= 1
+        cur = prev
+
+    return indexer
+
+
+def backfill_inplace(numeric_object_t[:] values, uint8_t[:] mask, limit=None):
+    pad_inplace(values[::-1], mask[::-1], limit=limit)
+
+
+def backfill_2d_inplace(numeric_object_t[:, :] values,
+                        uint8_t[:, :] mask,
+                        limit=None):
+    pad_2d_inplace(values[:, ::-1], mask[:, ::-1], limit)
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def is_monotonic(const numeric_object_t[:] arr, bint timelike):
+    """
+    Returns
+    -------
+    tuple
+        is_monotonic_inc : bool
+        is_monotonic_dec : bool
+        is_strict_monotonic : bool
+    """
+    cdef:
+        Py_ssize_t i, n
+        numeric_object_t prev, cur
+        bint is_monotonic_inc = 1
+        bint is_monotonic_dec = 1
+        bint is_unique = 1
+        bint is_strict_monotonic = 1
+
+    n = len(arr)
+
+    if n == 1:
+        if arr[0] != arr[0] or (numeric_object_t is int64_t and timelike and
+                                arr[0] == NPY_NAT):
+            # single value is NaN
+            return False, False, True
+        else:
+            return True, True, True
+    elif n < 2:
+        return True, True, True
+
+    if timelike and <int64_t>arr[0] == NPY_NAT:
+        return False, False, False
+
+    with nogil(numeric_object_t is not object):
+        prev = arr[0]
+        for i in range(1, n):
+            cur = arr[i]
+            if timelike and <int64_t>cur == NPY_NAT:
+                is_monotonic_inc = 0
+                is_monotonic_dec = 0
+                break
+            if cur < prev:
+                is_monotonic_inc = 0
+            elif cur > prev:
+                is_monotonic_dec = 0
+            elif cur == prev:
+                is_unique = 0
+            else:
+                # cur or prev is NaN
+                is_monotonic_inc = 0
+                is_monotonic_dec = 0
+                break
+            if not is_monotonic_inc and not is_monotonic_dec:
+                is_monotonic_inc = 0
+                is_monotonic_dec = 0
+                break
+            prev = cur
+
+    is_strict_monotonic = is_unique and (is_monotonic_inc or is_monotonic_dec)
+    return is_monotonic_inc, is_monotonic_dec, is_strict_monotonic
+
+
+# ----------------------------------------------------------------------
+# rank_1d, rank_2d
+# ----------------------------------------------------------------------
+
+cdef numeric_object_t get_rank_nan_fill_val(
+    bint rank_nans_highest,
+    numeric_object_t val,
+    bint is_datetimelike=False,
+):
+    """
+    Return the value we'll use to represent missing values when sorting depending
+    on if we'd like missing values to end up at the top/bottom. (The second parameter
+    is unused, but needed for fused type specialization)
+    """
+    if numeric_object_t is int64_t and is_datetimelike and not rank_nans_highest:
+        return NPY_NAT + 1
+
+    if rank_nans_highest:
+        if numeric_object_t is object:
+            return Infinity()
+        elif numeric_object_t is int64_t:
+            return util.INT64_MAX
+        elif numeric_object_t is int32_t:
+            return util.INT32_MAX
+        elif numeric_object_t is int16_t:
+            return util.INT16_MAX
+        elif numeric_object_t is int8_t:
+            return util.INT8_MAX
+        elif numeric_object_t is uint64_t:
+            return util.UINT64_MAX
+        elif numeric_object_t is uint32_t:
+            return util.UINT32_MAX
+        elif numeric_object_t is uint16_t:
+            return util.UINT16_MAX
+        elif numeric_object_t is uint8_t:
+            return util.UINT8_MAX
+        else:
+            return np.inf
+    else:
+        if numeric_object_t is object:
+            return NegInfinity()
+        elif numeric_object_t is int64_t:
+            # Note(jbrockmendel) 2022-03-15 for reasons unknown, using util.INT64_MIN
+            #  instead of NPY_NAT here causes build warnings and failure in
+            #  test_cummax_i8_at_implementation_bound
+            return NPY_NAT
+        elif numeric_object_t is int32_t:
+            return util.INT32_MIN
+        elif numeric_object_t is int16_t:
+            return util.INT16_MIN
+        elif numeric_object_t is int8_t:
+            return util.INT8_MIN
+        elif numeric_object_t is uint64_t:
+            return 0
+        elif numeric_object_t is uint32_t:
+            return 0
+        elif numeric_object_t is uint16_t:
+            return 0
+        elif numeric_object_t is uint8_t:
+            return 0
+        else:
+            return -np.inf
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def rank_1d(
+    ndarray[numeric_object_t, ndim=1] values,
+    const intp_t[:] labels=None,
+    bint is_datetimelike=False,
+    ties_method="average",
+    bint ascending=True,
+    bint pct=False,
+    na_option="keep",
+    const uint8_t[:] mask=None,
+):
+    """
+    Fast NaN-friendly version of ``scipy.stats.rankdata``.
+
+    Parameters
+    ----------
+    values : array of numeric_object_t values to be ranked
+    labels : np.ndarray[np.intp] or None
+        Array containing unique label for each group, with its ordering
+        matching up to the corresponding record in `values`. If not called
+        from a groupby operation, will be None.
+    is_datetimelike : bool, default False
+        True if `values` contains datetime-like entries.
+    ties_method : {'average', 'min', 'max', 'first', 'dense'}, default
+        'average'
+        * average: average rank of group
+        * min: lowest rank in group
+        * max: highest rank in group
+        * first: ranks assigned in order they appear in the array
+        * dense: like 'min', but rank always increases by 1 between groups
+    ascending : bool, default True
+        False for ranks by high (1) to low (N)
+        na_option : {'keep', 'top', 'bottom'}, default 'keep'
+    pct : bool, default False
+        Compute percentage rank of data within each group
+    na_option : {'keep', 'top', 'bottom'}, default 'keep'
+        * keep: leave NA values where they are
+        * top: smallest rank if ascending
+        * bottom: smallest rank if descending
+    mask : np.ndarray[bool], optional, default None
+        Specify locations to be treated as NA, for e.g. Categorical.
+    """
+    cdef:
+        TiebreakEnumType tiebreak
+        Py_ssize_t N
+        int64_t[::1] grp_sizes
+        intp_t[:] lexsort_indexer
+        float64_t[::1] out
+        ndarray[numeric_object_t, ndim=1] masked_vals
+        numeric_object_t[:] masked_vals_memview
+        bint keep_na, nans_rank_highest, check_labels, check_mask
+        numeric_object_t nan_fill_val
+
+    tiebreak = tiebreakers[ties_method]
+    if tiebreak == TIEBREAK_FIRST:
+        if not ascending:
+            tiebreak = TIEBREAK_FIRST_DESCENDING
+
+    keep_na = na_option == "keep"
+
+    N = len(values)
+    if labels is not None:
+        assert len(labels) == N
+    out = np.empty(N)
+    grp_sizes = np.ones(N, dtype=np.int64)
+
+    # If we don't care about labels, can short-circuit later label
+    # comparisons
+    check_labels = labels is not None
+
+    # For cases where a mask is not possible, we can avoid mask checks
+    check_mask = (
+        numeric_object_t is float32_t
+        or numeric_object_t is float64_t
+        or numeric_object_t is object
+        or (numeric_object_t is int64_t and is_datetimelike)
+    )
+    check_mask = check_mask or mask is not None
+
+    # Copy values into new array in order to fill missing data
+    # with mask, without obfuscating location of missing data
+    # in values array
+    if numeric_object_t is object and values.dtype != np.object_:
+        masked_vals = values.astype("O")
+    else:
+        masked_vals = values.copy()
+
+    if mask is not None:
+        pass
+    elif numeric_object_t is object:
+        mask = isnaobj(masked_vals)
+    elif numeric_object_t is int64_t and is_datetimelike:
+        mask = (masked_vals == NPY_NAT).astype(np.uint8)
+    elif numeric_object_t is float64_t or numeric_object_t is float32_t:
+        mask = np.isnan(masked_vals).astype(np.uint8)
+    else:
+        mask = np.zeros(shape=len(masked_vals), dtype=np.uint8)
+
+    # If `na_option == 'top'`, we want to assign the lowest rank
+    # to NaN regardless of ascending/descending. So if ascending,
+    # fill with lowest value of type to end up with lowest rank.
+    # If descending, fill with highest value since descending
+    # will flip the ordering to still end up with lowest rank.
+    # Symmetric logic applies to `na_option == 'bottom'`
+    nans_rank_highest = ascending ^ (na_option == "top")
+    nan_fill_val = get_rank_nan_fill_val(nans_rank_highest, <numeric_object_t>0)
+    if nans_rank_highest:
+        order = [masked_vals, mask]
+    else:
+        order = [masked_vals, ~(np.asarray(mask))]
+
+    if check_labels:
+        order.append(labels)
+
+    np.putmask(masked_vals, mask, nan_fill_val)
+    # putmask doesn't accept a memoryview, so we assign as a separate step
+    masked_vals_memview = masked_vals
+
+    # lexsort using labels, then mask, then actual values
+    # each label corresponds to a different group value,
+    # the mask helps you differentiate missing values before
+    # performing sort on the actual values
+    lexsort_indexer = np.lexsort(order).astype(np.intp, copy=False)
+
+    if not ascending:
+        lexsort_indexer = lexsort_indexer[::-1]
+
+    with nogil:
+        rank_sorted_1d(
+            out,
+            grp_sizes,
+            lexsort_indexer,
+            masked_vals_memview,
+            mask,
+            check_mask=check_mask,
+            N=N,
+            tiebreak=tiebreak,
+            keep_na=keep_na,
+            pct=pct,
+            labels=labels,
+        )
+
+    return np.asarray(out)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef void rank_sorted_1d(
+    float64_t[::1] out,
+    int64_t[::1] grp_sizes,
+    const intp_t[:] sort_indexer,
+    const numeric_object_t[:] masked_vals,
+    const uint8_t[:] mask,
+    bint check_mask,
+    Py_ssize_t N,
+    TiebreakEnumType tiebreak=TIEBREAK_AVERAGE,
+    bint keep_na=True,
+    bint pct=False,
+    # https://github.com/cython/cython/issues/1630, only trailing arguments can
+    # currently be omitted for cdef functions, which is why we keep this at the end
+    const intp_t[:] labels=None,
+) noexcept nogil:
+    """
+    See rank_1d.__doc__. Handles only actual ranking, so sorting and masking should
+    be handled in the caller. Note that `out` and `grp_sizes` are modified inplace.
+
+    Parameters
+    ----------
+    out : float64_t[::1]
+        Array to store computed ranks
+    grp_sizes : int64_t[::1]
+        Array to store group counts, only used if pct=True. Should only be None
+        if labels is None.
+    sort_indexer : intp_t[:]
+        Array of indices which sorts masked_vals
+    masked_vals : numeric_object_t[:]
+        The values input to rank_1d, with missing values replaced by fill values
+    mask : uint8_t[:]
+        Array where entries are True if the value is missing, False otherwise.
+    check_mask : bool
+        If False, assumes the mask is all False to skip mask indexing
+    N : Py_ssize_t
+        The number of elements to rank. Note: it is not always true that
+        N == len(out) or N == len(masked_vals) (see `nancorr_spearman` usage for why)
+    tiebreak : TiebreakEnumType, default TIEBREAK_AVERAGE
+        See rank_1d.__doc__ for the different modes
+    keep_na : bool, default True
+        Whether or not to keep nulls
+    pct : bool, default False
+        Compute percentage rank of data within each group
+    labels : See rank_1d.__doc__, default None. None implies all labels are the same.
+    """
+
+    cdef:
+        Py_ssize_t i, j, dups=0, sum_ranks=0,
+        Py_ssize_t grp_start=0, grp_vals_seen=1, grp_na_count=0
+        bint at_end, next_val_diff, group_changed, check_labels
+        int64_t grp_size
+
+    check_labels = labels is not None
+
+    # Loop over the length of the value array
+    # each incremental i value can be looked up in the lexsort_indexer
+    # array that we sorted previously, which gives us the location of
+    # that sorted value for retrieval back from the original
+    # values / masked_vals arrays
+    # TODO(cython3): de-duplicate once cython supports conditional nogil
+    with gil(numeric_object_t is object):
+        for i in range(N):
+            at_end = i == N - 1
+
+            # dups and sum_ranks will be incremented each loop where
+            # the value / group remains the same, and should be reset
+            # when either of those change. Used to calculate tiebreakers
+            dups += 1
+            sum_ranks += i - grp_start + 1
+
+            if numeric_object_t is object:
+                next_val_diff = at_end or are_diff(masked_vals[sort_indexer[i]],
+                                                   masked_vals[sort_indexer[i+1]])
+            else:
+                next_val_diff = at_end or (masked_vals[sort_indexer[i]]
+                                           != masked_vals[sort_indexer[i+1]])
+
+            # We'll need this check later anyway to determine group size, so just
+            # compute it here since shortcircuiting won't help
+            group_changed = at_end or (check_labels and
+                                       (labels[sort_indexer[i]]
+                                        != labels[sort_indexer[i+1]]))
+
+            # Update out only when there is a transition of values or labels.
+            # When a new value or group is encountered, go back #dups steps(
+            # the number of occurrence of current value) and assign the ranks
+            # based on the starting index of the current group (grp_start)
+            # and the current index
+            if (next_val_diff or group_changed or (check_mask and
+                                                   (mask[sort_indexer[i]]
+                                                    ^ mask[sort_indexer[i+1]]))):
+                # If keep_na, check for missing values and assign back
+                # to the result where appropriate
+                if keep_na and check_mask and mask[sort_indexer[i]]:
+                    grp_na_count = dups
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = NaN
+                elif tiebreak == TIEBREAK_AVERAGE:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = sum_ranks / <float64_t>dups
+                elif tiebreak == TIEBREAK_MIN:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = i - grp_start - dups + 2
+                elif tiebreak == TIEBREAK_MAX:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = i - grp_start + 1
+
+                # With n as the previous rank in the group and m as the number
+                # of duplicates in this stretch, if TIEBREAK_FIRST and ascending,
+                # then rankings should be n + 1, n + 2 ... n + m
+                elif tiebreak == TIEBREAK_FIRST:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = j + 1 - grp_start
+
+                # If TIEBREAK_FIRST and descending, the ranking should be
+                # n + m, n + (m - 1) ... n + 1. This is equivalent to
+                # (i - dups + 1) + (i - j + 1) - grp_start
+                elif tiebreak == TIEBREAK_FIRST_DESCENDING:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = 2 * i - j - dups + 2 - grp_start
+                elif tiebreak == TIEBREAK_DENSE:
+                    for j in range(i - dups + 1, i + 1):
+                        out[sort_indexer[j]] = grp_vals_seen
+
+                # Look forward to the next value (using the sorting in
+                # lexsort_indexer). If the value does not equal the current
+                # value then we need to reset the dups and sum_ranks, knowing
+                # that a new value is coming up. The conditional also needs
+                # to handle nan equality and the end of iteration. If group
+                # changes we do not record seeing a new value in the group
+                if not group_changed and (next_val_diff
+                                          or (check_mask and
+                                              (mask[sort_indexer[i]]
+                                               ^ mask[sort_indexer[i+1]]))):
+                    dups = sum_ranks = 0
+                    grp_vals_seen += 1
+
+                # Similar to the previous conditional, check now if we are
+                # moving to a new group. If so, keep track of the index where
+                # the new group occurs, so the tiebreaker calculations can
+                # decrement that from their position. Fill in the size of each
+                # group encountered (used by pct calculations later). Also be
+                # sure to reset any of the items helping to calculate dups
+                if group_changed:
+
+                    # If not dense tiebreak, group size used to compute
+                    # percentile will be # of non-null elements in group
+                    if tiebreak != TIEBREAK_DENSE:
+                        grp_size = i - grp_start + 1 - grp_na_count
+
+                    # Otherwise, it will be the number of distinct values
+                    # in the group, subtracting 1 if NaNs are present
+                    # since that is a distinct value we shouldn't count
+                    else:
+                        grp_size = grp_vals_seen - (grp_na_count > 0)
+
+                    for j in range(grp_start, i + 1):
+                        grp_sizes[sort_indexer[j]] = grp_size
+
+                    dups = sum_ranks = 0
+                    grp_na_count = 0
+                    grp_start = i + 1
+                    grp_vals_seen = 1
+
+    if pct:
+        for i in range(N):
+            if grp_sizes[i] != 0:
+                out[i] = out[i] / grp_sizes[i]
+
+
+def rank_2d(
+    ndarray[numeric_object_t, ndim=2] in_arr,
+    int axis=0,
+    bint is_datetimelike=False,
+    ties_method="average",
+    bint ascending=True,
+    na_option="keep",
+    bint pct=False,
+):
+    """
+    Fast NaN-friendly version of ``scipy.stats.rankdata``.
+    """
+    cdef:
+        Py_ssize_t k, n, col
+        float64_t[::1, :] out  # Column-major so columns are contiguous
+        int64_t[::1] grp_sizes
+        ndarray[numeric_object_t, ndim=2] values
+        numeric_object_t[:, :] masked_vals
+        intp_t[:, :] sort_indexer
+        uint8_t[:, :] mask
+        TiebreakEnumType tiebreak
+        bint check_mask, keep_na, nans_rank_highest
+        numeric_object_t nan_fill_val
+
+    tiebreak = tiebreakers[ties_method]
+    if tiebreak == TIEBREAK_FIRST:
+        if not ascending:
+            tiebreak = TIEBREAK_FIRST_DESCENDING
+
+    keep_na = na_option == "keep"
+
+    # For cases where a mask is not possible, we can avoid mask checks
+    check_mask = (
+        numeric_object_t is float32_t
+        or numeric_object_t is float64_t
+        or numeric_object_t is object
+        or (numeric_object_t is int64_t and is_datetimelike)
+    )
+
+    if axis == 1:
+        values = np.asarray(in_arr).T.copy()
+    else:
+        values = np.asarray(in_arr).copy()
+
+    if numeric_object_t is object:
+        if values.dtype != np.object_:
+            values = values.astype("O")
+
+    nans_rank_highest = ascending ^ (na_option == "top")
+    if check_mask:
+        nan_fill_val = get_rank_nan_fill_val(nans_rank_highest, <numeric_object_t>0)
+
+        if numeric_object_t is object:
+            mask = isnaobj(values).view(np.uint8)
+        elif numeric_object_t is float64_t or numeric_object_t is float32_t:
+            mask = np.isnan(values).view(np.uint8)
+        else:
+            # i.e. int64 and datetimelike
+            mask = (values == NPY_NAT).view(np.uint8)
+        np.putmask(values, mask, nan_fill_val)
+    else:
+        mask = np.zeros_like(values, dtype=np.uint8)
+
+    if nans_rank_highest:
+        order = (values, mask)
+    else:
+        order = (values, ~np.asarray(mask))
+
+    n, k = (<object>values).shape
+    out = np.empty((n, k), dtype="f8", order="F")
+    grp_sizes = np.ones(n, dtype=np.int64)
+
+    # lexsort is slower, so only use if we need to worry about the mask
+    if check_mask:
+        sort_indexer = np.lexsort(order, axis=0).astype(np.intp, copy=False)
+    else:
+        kind = "stable" if ties_method == "first" else None
+        sort_indexer = values.argsort(axis=0, kind=kind).astype(np.intp, copy=False)
+
+    if not ascending:
+        sort_indexer = sort_indexer[::-1, :]
+
+    # putmask doesn't accept a memoryview, so we assign in a separate step
+    masked_vals = values
+    with nogil:
+        for col in range(k):
+            rank_sorted_1d(
+                out[:, col],
+                grp_sizes,
+                sort_indexer[:, col],
+                masked_vals[:, col],
+                mask[:, col],
+                check_mask=check_mask,
+                N=n,
+                tiebreak=tiebreak,
+                keep_na=keep_na,
+                pct=pct,
+            )
+
+    if axis == 1:
+        return np.asarray(out.T)
+    else:
+        return np.asarray(out)
+
+
+ctypedef fused diff_t:
+    float64_t
+    float32_t
+    int8_t
+    int16_t
+    int32_t
+    int64_t
+
+ctypedef fused out_t:
+    float32_t
+    float64_t
+    int64_t
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def diff_2d(
+    const diff_t[:, :] arr,
+    ndarray[out_t, ndim=2] out,
+    Py_ssize_t periods,
+    int axis,
+    bint datetimelike=False,
+):
+    cdef:
+        Py_ssize_t i, j, sx, sy, start, stop
+        bint f_contig = arr.is_f_contig()
+        diff_t left, right
+
+    # Disable for unsupported dtype combinations,
+    #  see https://github.com/cython/cython/issues/2646
+    if (out_t is float32_t
+            and not (diff_t is float32_t or diff_t is int8_t or diff_t is int16_t)):
+        raise NotImplementedError  # pragma: no cover
+    elif (out_t is float64_t
+          and (diff_t is float32_t or diff_t is int8_t or diff_t is int16_t)):
+        raise NotImplementedError  # pragma: no cover
+    elif out_t is int64_t and diff_t is not int64_t:
+        # We only have out_t of int64_t if we have datetimelike
+        raise NotImplementedError  # pragma: no cover
+    else:
+        # We put this inside an indented else block to avoid cython build
+        #  warnings about unreachable code
+        sx, sy = (<object>arr).shape
+        with nogil:
+            if f_contig:
+                if axis == 0:
+                    if periods >= 0:
+                        start, stop = periods, sx
+                    else:
+                        start, stop = 0, sx + periods
+                    for j in range(sy):
+                        for i in range(start, stop):
+                            left = arr[i, j]
+                            right = arr[i - periods, j]
+                            if out_t is int64_t and datetimelike:
+                                if left == NPY_NAT or right == NPY_NAT:
+                                    out[i, j] = NPY_NAT
+                                else:
+                                    out[i, j] = left - right
+                            else:
+                                out[i, j] = left - right
+                else:
+                    if periods >= 0:
+                        start, stop = periods, sy
+                    else:
+                        start, stop = 0, sy + periods
+                    for j in range(start, stop):
+                        for i in range(sx):
+                            left = arr[i, j]
+                            right = arr[i, j - periods]
+                            if out_t is int64_t and datetimelike:
+                                if left == NPY_NAT or right == NPY_NAT:
+                                    out[i, j] = NPY_NAT
+                                else:
+                                    out[i, j] = left - right
+                            else:
+                                out[i, j] = left - right
+            else:
+                if axis == 0:
+                    if periods >= 0:
+                        start, stop = periods, sx
+                    else:
+                        start, stop = 0, sx + periods
+                    for i in range(start, stop):
+                        for j in range(sy):
+                            left = arr[i, j]
+                            right = arr[i - periods, j]
+                            if out_t is int64_t and datetimelike:
+                                if left == NPY_NAT or right == NPY_NAT:
+                                    out[i, j] = NPY_NAT
+                                else:
+                                    out[i, j] = left - right
+                            else:
+                                out[i, j] = left - right
+                else:
+                    if periods >= 0:
+                        start, stop = periods, sy
+                    else:
+                        start, stop = 0, sy + periods
+                    for i in range(sx):
+                        for j in range(start, stop):
+                            left = arr[i, j]
+                            right = arr[i, j - periods]
+                            if out_t is int64_t and datetimelike:
+                                if left == NPY_NAT or right == NPY_NAT:
+                                    out[i, j] = NPY_NAT
+                                else:
+                                    out[i, j] = left - right
+                            else:
+                                out[i, j] = left - right
+
+
+# generated from template
+include "algos_common_helper.pxi"
+include "algos_take_helper.pxi"
diff --git a/pandas/pandas/_libs/algos_common_helper.pxi.in b/pandas/pandas/_libs/algos_common_helper.pxi.in
new file mode 100644
index 0000000000000..ee815b8bbf7a6
--- /dev/null
+++ b/pandas/pandas/_libs/algos_common_helper.pxi.in
@@ -0,0 +1,73 @@
+"""
+Template for each `dtype` helper function using 1-d template
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+# ----------------------------------------------------------------------
+# ensure_dtype
+# ----------------------------------------------------------------------
+
+
+def ensure_platform_int(object arr):
+    # GH3033, GH1392
+    # platform int is the size of the int pointer, e.g. np.intp
+    if util.is_array(arr):
+        if (<ndarray>arr).descr.type_num == cnp.NPY_INTP:
+            return arr
+        else:
+            # equiv: arr.astype(np.intp)
+            return cnp.PyArray_Cast(<ndarray>arr, cnp.NPY_INTP)
+    else:
+        return np.array(arr, dtype=np.intp)
+
+
+def ensure_object(object arr):
+    if util.is_array(arr):
+        if (<ndarray>arr).descr.type_num == NPY_OBJECT:
+            return arr
+        else:
+            # equiv: arr.astype(object)
+            return cnp.PyArray_Cast(<ndarray>arr, NPY_OBJECT)
+    else:
+        return np.array(arr, dtype=np.object_)
+
+{{py:
+
+# name, c_type, dtype
+dtypes = [('float64', 'FLOAT64', 'float64'),
+          # ('float32', 'FLOAT32', 'float32'),  # disabling bc unused
+          ('int8', 'INT8', 'int8'),
+          ('int16', 'INT16', 'int16'),
+          ('int32', 'INT32', 'int32'),
+          ('int64', 'INT64', 'int64'),
+          ('uint64', 'UINT64', 'uint64'),
+          # Disabling uint and complex dtypes because we do not use them
+          #  (and compiling them increases wheel size) (except uint64)
+          # ('uint8', 'UINT8', 'uint8'),
+          # ('uint16', 'UINT16', 'uint16'),
+          # ('uint32', 'UINT32', 'uint32'),
+          # ('complex64', 'COMPLEX64', 'complex64'),
+          # ('complex128', 'COMPLEX128', 'complex128')
+]
+
+def get_dispatch(dtypes):
+
+    for name, c_type, dtype in dtypes:
+        yield name, c_type, dtype
+}}
+
+{{for name, c_type, dtype in get_dispatch(dtypes)}}
+
+
+def ensure_{{name}}(object arr):
+    if util.is_array(arr):
+        if (<ndarray>arr).descr.type_num == NPY_{{c_type}}:
+            return arr
+        else:
+            # equiv: arr.astype(np.{{dtype}})
+            return cnp.PyArray_Cast(<ndarray>arr, cnp.NPY_{{c_type}})
+    else:
+        return np.asarray(arr, dtype=np.{{dtype}})
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/algos_take_helper.pxi.in b/pandas/pandas/_libs/algos_take_helper.pxi.in
new file mode 100644
index 0000000000000..e6b39896280b9
--- /dev/null
+++ b/pandas/pandas/_libs/algos_take_helper.pxi.in
@@ -0,0 +1,247 @@
+"""
+Template for each `dtype` helper function for take
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+# ----------------------------------------------------------------------
+# take_1d, take_2d
+# ----------------------------------------------------------------------
+
+
+{{py:
+
+# c_type_in, c_type_out
+dtypes = [
+    ('uint8_t', 'uint8_t'),
+    ('uint8_t', 'object'),
+    ('uint16_t', 'uint16_t'),
+    ('uint32_t', 'uint32_t'),
+    ('uint64_t', 'uint64_t'),
+    ('int8_t', 'int8_t'),
+    ('int8_t', 'int32_t'),
+    ('int8_t', 'int64_t'),
+    ('int8_t', 'float64_t'),
+    ('int16_t', 'int16_t'),
+    ('int16_t', 'int32_t'),
+    ('int16_t', 'int64_t'),
+    ('int16_t', 'float64_t'),
+    ('int32_t', 'int32_t'),
+    ('int32_t', 'int64_t'),
+    ('int32_t', 'float64_t'),
+    ('int64_t', 'int64_t'),
+    ('int64_t', 'float64_t'),
+    ('float32_t', 'float32_t'),
+    ('float32_t', 'float64_t'),
+    ('float64_t', 'float64_t'),
+    ('object', 'object'),
+]
+
+
+def get_dispatch(dtypes):
+
+    for (c_type_in, c_type_out) in dtypes:
+
+        def get_name(dtype_name):
+            if dtype_name == "object":
+                return "object"
+            if dtype_name == "uint8_t":
+                return "bool"
+            return dtype_name[:-2]
+
+        name = get_name(c_type_in)
+        dest = get_name(c_type_out)
+
+        args = dict(name=name, dest=dest, c_type_in=c_type_in,
+                    c_type_out=c_type_out)
+
+        yield (name, dest, c_type_in, c_type_out)
+
+}}
+
+
+{{for name, dest, c_type_in, c_type_out in get_dispatch(dtypes)}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if c_type_in != "object"}}
+def take_1d_{{name}}_{{dest}}(const {{c_type_in}}[:] values,
+{{else}}
+def take_1d_{{name}}_{{dest}}(ndarray[{{c_type_in}}, ndim=1] values,
+{{endif}}
+                              const intp_t[:] indexer,
+                              {{c_type_out}}[:] out,
+                              fill_value=np.nan):
+
+    cdef:
+        Py_ssize_t i, n, idx
+        {{c_type_out}} fv
+
+    n = indexer.shape[0]
+
+    fv = fill_value
+
+    {{if c_type_out != "object"}}
+    with nogil:
+    {{else}}
+    if True:
+    {{endif}}
+        for i in range(n):
+            idx = indexer[i]
+            if idx == -1:
+                out[i] = fv
+            else:
+                {{if c_type_in == "uint8_t" and c_type_out == "object"}}
+                out[i] = True if values[idx] > 0 else False
+                {{else}}
+                out[i] = values[idx]
+                {{endif}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if c_type_in != "object"}}
+def take_2d_axis0_{{name}}_{{dest}}(const {{c_type_in}}[:, :] values,
+{{else}}
+def take_2d_axis0_{{name}}_{{dest}}(ndarray[{{c_type_in}}, ndim=2] values,
+{{endif}}
+                                    ndarray[intp_t, ndim=1] indexer,
+                                    {{c_type_out}}[:, :] out,
+                                    fill_value=np.nan):
+    cdef:
+        Py_ssize_t i, j, k, n, idx
+        {{c_type_out}} fv
+        {{if c_type_in == c_type_out != "object"}}
+        const {{c_type_out}} *v
+        {{c_type_out}} *o
+        {{endif}}
+
+    n = len(indexer)
+    k = values.shape[1]
+
+    fv = fill_value
+
+    {{if c_type_in == c_type_out != "object"}}
+    # GH#3130
+    with nogil:
+        if (values.strides[1] == out.strides[1] and
+            values.strides[1] == sizeof({{c_type_out}}) and
+            sizeof({{c_type_out}}) * n >= 256):
+
+            for i in range(n):
+                idx = indexer[i]
+                if idx == -1:
+                    for j in range(k):
+                        out[i, j] = fv
+                else:
+                    v = &values[idx, 0]
+                    o = &out[i, 0]
+                    memmove(o, v, <size_t>(sizeof({{c_type_out}}) * k))
+        else:
+            for i in range(n):
+                idx = indexer[i]
+                if idx == -1:
+                    for j in range(k):
+                        out[i, j] = fv
+                else:
+                    for j in range(k):
+                        out[i, j] = values[idx, j]
+    {{else}}
+
+    for i in range(n):
+        idx = indexer[i]
+        if idx == -1:
+            for j in range(k):
+                out[i, j] = fv
+        else:
+            for j in range(k):
+                {{if c_type_in == "uint8_t" and c_type_out == "object"}}
+                out[i, j] = True if values[idx, j] > 0 else False
+                {{else}}
+                out[i, j] = values[idx, j]
+                {{endif}}
+    {{endif}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if c_type_in != "object"}}
+def take_2d_axis1_{{name}}_{{dest}}(const {{c_type_in}}[:, :] values,
+{{else}}
+def take_2d_axis1_{{name}}_{{dest}}(ndarray[{{c_type_in}}, ndim=2] values,
+{{endif}}
+                                    ndarray[intp_t, ndim=1] indexer,
+                                    {{c_type_out}}[:, :] out,
+                                    fill_value=np.nan):
+
+    cdef:
+        Py_ssize_t i, j, k, n, idx
+        {{c_type_out}} fv
+
+    n = len(values)
+    k = len(indexer)
+
+    if n == 0 or k == 0:
+        return
+
+    fv = fill_value
+
+    {{if c_type_in == c_type_out != "object"}}
+    with nogil:
+        for i in range(n):
+            for j in range(k):
+                idx = indexer[j]
+                if idx == -1:
+                    out[i, j] = fv
+                else:
+                    out[i, j] = values[i, idx]
+
+    {{else}}
+    for i in range(n):
+        for j in range(k):
+            idx = indexer[j]
+            if idx == -1:
+                out[i, j] = fv
+            else:
+                {{if c_type_in == "uint8_t" and c_type_out == "object"}}
+                out[i, j] = True if values[i, idx] > 0 else False
+                {{else}}
+                out[i, j] = values[i, idx]
+                {{endif}}
+    {{endif}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def take_2d_multi_{{name}}_{{dest}}(ndarray[{{c_type_in}}, ndim=2] values,
+                                    indexer,
+                                    ndarray[{{c_type_out}}, ndim=2] out,
+                                    fill_value=np.nan):
+    cdef:
+        Py_ssize_t i, j, k, n, idx
+        ndarray[intp_t, ndim=1] idx0 = indexer[0]
+        ndarray[intp_t, ndim=1] idx1 = indexer[1]
+        {{c_type_out}} fv
+
+    n = len(idx0)
+    k = len(idx1)
+
+    fv = fill_value
+    for i in range(n):
+        idx = idx0[i]
+        if idx == -1:
+            for j in range(k):
+                out[i, j] = fv
+        else:
+            for j in range(k):
+                if idx1[j] == -1:
+                    out[i, j] = fv
+                else:
+                    {{if c_type_in == "uint8_t" and c_type_out == "object"}}
+                    out[i, j] = True if values[idx, idx1[j]] > 0 else False
+                    {{else}}
+                    out[i, j] = values[idx, idx1[j]]
+                    {{endif}}
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/arrays.pxd b/pandas/pandas/_libs/arrays.pxd
new file mode 100644
index 0000000000000..737da29da46a4
--- /dev/null
+++ b/pandas/pandas/_libs/arrays.pxd
@@ -0,0 +1,11 @@
+
+from numpy cimport ndarray
+
+
+cdef class NDArrayBacked:
+    cdef:
+        readonly ndarray _ndarray
+        readonly object _dtype
+
+    cpdef NDArrayBacked _from_backing_data(self, ndarray values)
+    cpdef __setstate__(self, state)
diff --git a/pandas/pandas/_libs/arrays.pyi b/pandas/pandas/_libs/arrays.pyi
new file mode 100644
index 0000000000000..dda23d9dec98b
--- /dev/null
+++ b/pandas/pandas/_libs/arrays.pyi
@@ -0,0 +1,40 @@
+from typing import Sequence
+
+import numpy as np
+
+from pandas._typing import (
+    AxisInt,
+    DtypeObj,
+    Self,
+    Shape,
+)
+
+class NDArrayBacked:
+    _dtype: DtypeObj
+    _ndarray: np.ndarray
+    def __init__(self, values: np.ndarray, dtype: DtypeObj) -> None: ...
+    @classmethod
+    def _simple_new(cls, values: np.ndarray, dtype: DtypeObj) -> Self: ...
+    def _from_backing_data(self, values: np.ndarray) -> Self: ...
+    def __setstate__(self, state) -> None: ...
+    def __len__(self) -> int: ...
+    @property
+    def shape(self) -> Shape: ...
+    @property
+    def ndim(self) -> int: ...
+    @property
+    def size(self) -> int: ...
+    @property
+    def nbytes(self) -> int: ...
+    def copy(self, order=...) -> Self: ...
+    def delete(self, loc, axis=...) -> Self: ...
+    def swapaxes(self, axis1, axis2) -> Self: ...
+    def repeat(self, repeats: int | Sequence[int], axis: int | None = ...) -> Self: ...
+    def reshape(self, *args, **kwargs) -> Self: ...
+    def ravel(self, order=...) -> Self: ...
+    @property
+    def T(self) -> Self: ...
+    @classmethod
+    def _concat_same_type(
+        cls, to_concat: Sequence[Self], axis: AxisInt = ...
+    ) -> Self: ...
diff --git a/pandas/pandas/_libs/arrays.pyx b/pandas/pandas/_libs/arrays.pyx
new file mode 100644
index 0000000000000..2932f3ff56396
--- /dev/null
+++ b/pandas/pandas/_libs/arrays.pyx
@@ -0,0 +1,194 @@
+"""
+Cython implementations for internal ExtensionArrays.
+"""
+cimport cython
+
+import numpy as np
+
+cimport numpy as cnp
+from cpython cimport PyErr_Clear
+from numpy cimport ndarray
+
+cnp.import_array()
+
+
+@cython.freelist(16)
+cdef class NDArrayBacked:
+    """
+    Implementing these methods in cython improves performance quite a bit.
+
+    import pandas as pd
+
+    from pandas._libs.arrays import NDArrayBacked as cls
+
+    dti = pd.date_range("2016-01-01", periods=3)
+    dta = dti._data
+    arr = dta._ndarray
+
+    obj = cls._simple_new(arr, arr.dtype)
+
+    # for foo in [arr, dta, obj]: ...
+
+    %timeit foo.copy()
+    299 ns ± 30 ns per loop     # <-- arr underlying ndarray (for reference)
+    530 ns ± 9.24 ns per loop   # <-- dta with cython NDArrayBacked
+    1.66 µs ± 46.3 ns per loop  # <-- dta without cython NDArrayBacked
+    328 ns ± 5.29 ns per loop   # <-- obj with NDArrayBacked.__cinit__
+    371 ns ± 6.97 ns per loop   # <-- obj with NDArrayBacked._simple_new
+
+    %timeit foo.T
+    125 ns ± 6.27 ns per loop   # <-- arr underlying ndarray (for reference)
+    226 ns ± 7.66 ns per loop   # <-- dta with cython NDArrayBacked
+    911 ns ± 16.6 ns per loop   # <-- dta without cython NDArrayBacked
+    215 ns ± 4.54 ns per loop   # <-- obj with NDArrayBacked._simple_new
+
+    """
+    # TODO: implement take in terms of cnp.PyArray_TakeFrom
+    # TODO: implement concat_same_type in terms of cnp.PyArray_Concatenate
+
+    # cdef:
+    #    readonly ndarray _ndarray
+    #    readonly object _dtype
+
+    def __init__(self, ndarray values, object dtype):
+        self._ndarray = values
+        self._dtype = dtype
+
+    @classmethod
+    def _simple_new(cls, ndarray values, object dtype):
+        cdef:
+            NDArrayBacked obj
+        obj = NDArrayBacked.__new__(cls)
+        obj._ndarray = values
+        obj._dtype = dtype
+        return obj
+
+    cpdef NDArrayBacked _from_backing_data(self, ndarray values):
+        """
+        Construct a new ExtensionArray `new_array` with `arr` as its _ndarray.
+
+        The returned array has the same dtype as self.
+
+        Caller is responsible for ensuring `values.dtype == self._ndarray.dtype`.
+
+        This should round-trip:
+            self == self._from_backing_data(self._ndarray)
+        """
+        # TODO: re-reuse simple_new if/when it can be cpdef
+        cdef:
+            NDArrayBacked obj
+        obj = NDArrayBacked.__new__(type(self))
+        obj._ndarray = values
+        obj._dtype = self._dtype
+        return obj
+
+    cpdef __setstate__(self, state):
+        if isinstance(state, dict):
+            if "_data" in state:
+                data = state.pop("_data")
+            elif "_ndarray" in state:
+                data = state.pop("_ndarray")
+            else:
+                raise ValueError  # pragma: no cover
+            self._ndarray = data
+            self._dtype = state.pop("_dtype")
+
+            for key, val in state.items():
+                setattr(self, key, val)
+        elif isinstance(state, tuple):
+            if len(state) != 3:
+                if len(state) == 1 and isinstance(state[0], dict):
+                    self.__setstate__(state[0])
+                    return
+                raise NotImplementedError(state)  # pragma: no cover
+
+            data, dtype = state[:2]
+            if isinstance(dtype, np.ndarray):
+                dtype, data = data, dtype
+            self._ndarray = data
+            self._dtype = dtype
+
+            if isinstance(state[2], dict):
+                for key, val in state[2].items():
+                    setattr(self, key, val)
+            else:
+                raise NotImplementedError(state)  # pragma: no cover
+        else:
+            raise NotImplementedError(state)  # pragma: no cover
+
+    def __len__(self) -> int:
+        return len(self._ndarray)
+
+    @property
+    def shape(self):
+        # object cast bc _ndarray.shape is npy_intp*
+        return (<object>(self._ndarray)).shape
+
+    @property
+    def ndim(self) -> int:
+        return self._ndarray.ndim
+
+    @property
+    def size(self) -> int:
+        return self._ndarray.size
+
+    @property
+    def nbytes(self) -> int:
+        return cnp.PyArray_NBYTES(self._ndarray)
+
+    def copy(self, order="C"):
+        cdef:
+            cnp.NPY_ORDER order_code
+            int success
+
+        success = cnp.PyArray_OrderConverter(order, &order_code)
+        if not success:
+            # clear exception so that we don't get a SystemError
+            PyErr_Clear()
+            # same message used by numpy
+            msg = f"order must be one of 'C', 'F', 'A', or 'K' (got '{order}')"
+            raise ValueError(msg)
+
+        res_values = cnp.PyArray_NewCopy(self._ndarray, order_code)
+        return self._from_backing_data(res_values)
+
+    def delete(self, loc, axis=0):
+        res_values = np.delete(self._ndarray, loc, axis=axis)
+        return self._from_backing_data(res_values)
+
+    def swapaxes(self, axis1, axis2):
+        res_values = cnp.PyArray_SwapAxes(self._ndarray, axis1, axis2)
+        return self._from_backing_data(res_values)
+
+    # TODO: pass NPY_MAXDIMS equiv to axis=None?
+    def repeat(self, repeats, axis: int | np.integer = 0):
+        if axis is None:
+            axis = 0
+        res_values = cnp.PyArray_Repeat(self._ndarray, repeats, <int>axis)
+        return self._from_backing_data(res_values)
+
+    def reshape(self, *args, **kwargs):
+        res_values = self._ndarray.reshape(*args, **kwargs)
+        return self._from_backing_data(res_values)
+
+    def ravel(self, order="C"):
+        # cnp.PyArray_OrderConverter(PyObject* obj, NPY_ORDER* order)
+        # res_values = cnp.PyArray_Ravel(self._ndarray, order)
+        res_values = self._ndarray.ravel(order)
+        return self._from_backing_data(res_values)
+
+    @property
+    def T(self):
+        res_values = self._ndarray.T
+        return self._from_backing_data(res_values)
+
+    def transpose(self, *axes):
+        res_values = self._ndarray.transpose(*axes)
+        return self._from_backing_data(res_values)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat, axis=0):
+        # NB: We are assuming at this point that dtypes all match
+        new_values = [obj._ndarray for obj in to_concat]
+        new_arr = cnp.PyArray_Concatenate(new_values, axis)
+        return to_concat[0]._from_backing_data(new_arr)
diff --git a/pandas/pandas/_libs/byteswap.pyi b/pandas/pandas/_libs/byteswap.pyi
new file mode 100644
index 0000000000000..bb0dbfc6a50b1
--- /dev/null
+++ b/pandas/pandas/_libs/byteswap.pyi
@@ -0,0 +1,5 @@
+def read_float_with_byteswap(data: bytes, offset: int, byteswap: bool) -> float: ...
+def read_double_with_byteswap(data: bytes, offset: int, byteswap: bool) -> float: ...
+def read_uint16_with_byteswap(data: bytes, offset: int, byteswap: bool) -> int: ...
+def read_uint32_with_byteswap(data: bytes, offset: int, byteswap: bool) -> int: ...
+def read_uint64_with_byteswap(data: bytes, offset: int, byteswap: bool) -> int: ...
diff --git a/pandas/pandas/_libs/byteswap.pyx b/pandas/pandas/_libs/byteswap.pyx
new file mode 100644
index 0000000000000..7a8a9fc5a9139
--- /dev/null
+++ b/pandas/pandas/_libs/byteswap.pyx
@@ -0,0 +1,85 @@
+"""
+The following are faster versions of struct.unpack that avoid the overhead of Python
+function calls.
+
+In the SAS7BDAT parser, they may be called up to (n_rows * n_cols) times.
+"""
+from cython cimport Py_ssize_t
+from libc.stdint cimport (
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+from libc.string cimport memcpy
+
+
+def read_float_with_byteswap(bytes data, Py_ssize_t offset, bint byteswap):
+    cdef uint32_t value
+    assert offset + <Py_ssize_t>sizeof(value) < len(data)
+    cdef const void *ptr = <unsigned char*>(data) + offset
+    memcpy(&value, ptr, sizeof(value))
+    if byteswap:
+        value = _byteswap4(value)
+
+    cdef float res
+    memcpy(&res, &value, sizeof(res))
+    return res
+
+
+def read_double_with_byteswap(bytes data, Py_ssize_t offset, bint byteswap):
+    cdef uint64_t value
+    assert offset + <Py_ssize_t>sizeof(value) < len(data)
+    cdef const void *ptr = <unsigned char*>(data) + offset
+    memcpy(&value, ptr, sizeof(value))
+    if byteswap:
+        value = _byteswap8(value)
+
+    cdef double res
+    memcpy(&res, &value, sizeof(res))
+    return res
+
+
+def read_uint16_with_byteswap(bytes data, Py_ssize_t offset, bint byteswap):
+    cdef uint16_t res
+    assert offset + <Py_ssize_t>sizeof(res) < len(data)
+    memcpy(&res, <const unsigned char*>(data) + offset, sizeof(res))
+    if byteswap:
+        res = _byteswap2(res)
+    return res
+
+
+def read_uint32_with_byteswap(bytes data, Py_ssize_t offset, bint byteswap):
+    cdef uint32_t res
+    assert offset + <Py_ssize_t>sizeof(res) < len(data)
+    memcpy(&res, <const unsigned char*>(data) + offset, sizeof(res))
+    if byteswap:
+        res = _byteswap4(res)
+    return res
+
+
+def read_uint64_with_byteswap(bytes data, Py_ssize_t offset, bint byteswap):
+    cdef uint64_t res
+    assert offset + <Py_ssize_t>sizeof(res) < len(data)
+    memcpy(&res, <const unsigned char*>(data) + offset, sizeof(res))
+    if byteswap:
+        res = _byteswap8(res)
+    return res
+
+
+# Byteswapping
+
+cdef extern from *:
+    """
+    #ifdef _MSC_VER
+        #define _byteswap2 _byteswap_ushort
+        #define _byteswap4 _byteswap_ulong
+        #define _byteswap8 _byteswap_uint64
+    #else
+        #define _byteswap2 __builtin_bswap16
+        #define _byteswap4 __builtin_bswap32
+        #define _byteswap8 __builtin_bswap64
+    #endif
+    """
+    uint16_t _byteswap2(uint16_t)
+    uint32_t _byteswap4(uint32_t)
+    uint64_t _byteswap8(uint64_t)
diff --git a/pandas/pandas/_libs/dtypes.pxd b/pandas/pandas/_libs/dtypes.pxd
new file mode 100644
index 0000000000000..ccfb2d2ef4a23
--- /dev/null
+++ b/pandas/pandas/_libs/dtypes.pxd
@@ -0,0 +1,36 @@
+"""
+Common location for shared fused types
+"""
+
+from numpy cimport (
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+
+# All numeric types except complex
+ctypedef fused numeric_t:
+    int8_t
+    int16_t
+    int32_t
+    int64_t
+
+    uint8_t
+    uint16_t
+    uint32_t
+    uint64_t
+
+    float32_t
+    float64_t
+
+# All numeric types + object, doesn't include complex
+ctypedef fused numeric_object_t:
+    numeric_t
+    object
diff --git a/pandas/pandas/_libs/free_threading_config.pxi.in b/pandas/pandas/_libs/free_threading_config.pxi.in
new file mode 100644
index 0000000000000..fe7ca399389ab
--- /dev/null
+++ b/pandas/pandas/_libs/free_threading_config.pxi.in
@@ -0,0 +1,3 @@
+# Autogenerated file containing Cython compile-time defines
+
+DEF CYTHON_COMPATIBLE_WITH_FREE_THREADING = @freethreading_compatible@
diff --git a/pandas/pandas/_libs/groupby.pyi b/pandas/pandas/_libs/groupby.pyi
new file mode 100644
index 0000000000000..163fc23535022
--- /dev/null
+++ b/pandas/pandas/_libs/groupby.pyi
@@ -0,0 +1,233 @@
+from typing import Literal
+
+import numpy as np
+
+from pandas._typing import npt
+
+def group_median_float64(
+    out: np.ndarray,  # ndarray[float64_t, ndim=2]
+    counts: npt.NDArray[np.int64],
+    values: np.ndarray,  # ndarray[float64_t, ndim=2]
+    labels: npt.NDArray[np.int64],
+    min_count: int = ...,  # Py_ssize_t
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    is_datetimelike: bool = ...,  # bint
+    skipna: bool = ...,
+) -> None: ...
+def group_cumprod(
+    out: np.ndarray,  # float64_t[:, ::1]
+    values: np.ndarray,  # const float64_t[:, :]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    is_datetimelike: bool,
+    skipna: bool = ...,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+) -> None: ...
+def group_cumsum(
+    out: np.ndarray,  # int64float_t[:, ::1]
+    values: np.ndarray,  # ndarray[int64float_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    is_datetimelike: bool,
+    skipna: bool = ...,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+) -> None: ...
+def group_shift_indexer(
+    out: np.ndarray,  # int64_t[::1]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    periods: int,
+) -> None: ...
+def group_fillna_indexer(
+    out: np.ndarray,  # ndarray[intp_t]
+    labels: np.ndarray,  # ndarray[int64_t]
+    mask: npt.NDArray[np.uint8],
+    limit: int,  # int64_t
+    compute_ffill: bool,
+    ngroups: int,
+) -> None: ...
+def group_any_all(
+    out: np.ndarray,  # uint8_t[::1]
+    values: np.ndarray,  # const uint8_t[::1]
+    labels: np.ndarray,  # const int64_t[:]
+    mask: np.ndarray,  # const uint8_t[::1]
+    val_test: Literal["any", "all"],
+    skipna: bool,
+    result_mask: np.ndarray | None,
+) -> None: ...
+def group_sum(
+    out: np.ndarray,  # complexfloatingintuint_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[complexfloatingintuint_t, ndim=2]
+    labels: np.ndarray,  # const intp_t[:]
+    mask: np.ndarray | None,
+    result_mask: np.ndarray | None = ...,
+    min_count: int = ...,
+    is_datetimelike: bool = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_prod(
+    out: np.ndarray,  # int64float_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[int64float_t, ndim=2]
+    labels: np.ndarray,  # const intp_t[:]
+    mask: np.ndarray | None,
+    result_mask: np.ndarray | None = ...,
+    min_count: int = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_var(
+    out: np.ndarray,  # floating[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[floating, ndim=2]
+    labels: np.ndarray,  # const intp_t[:]
+    min_count: int = ...,  # Py_ssize_t
+    ddof: int = ...,  # int64_t
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    is_datetimelike: bool = ...,
+    name: str = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_skew(
+    out: np.ndarray,  # float64_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[float64_T, ndim=2]
+    labels: np.ndarray,  # const intp_t[::1]
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_kurt(
+    out: np.ndarray,  # float64_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[float64_T, ndim=2]
+    labels: np.ndarray,  # const intp_t[::1]
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_mean(
+    out: np.ndarray,  # floating[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[floating, ndim=2]
+    labels: np.ndarray,  # const intp_t[:]
+    min_count: int = ...,  # Py_ssize_t
+    is_datetimelike: bool = ...,  # bint
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_ohlc(
+    out: np.ndarray,  # floatingintuint_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[floatingintuint_t, ndim=2]
+    labels: np.ndarray,  # const intp_t[:]
+    min_count: int = ...,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+) -> None: ...
+def group_quantile(
+    out: npt.NDArray[np.float64],
+    values: np.ndarray,  # ndarray[numeric, ndim=1]
+    labels: npt.NDArray[np.intp],
+    mask: npt.NDArray[np.uint8],
+    qs: npt.NDArray[np.float64],  # const
+    starts: npt.NDArray[np.int64],
+    ends: npt.NDArray[np.int64],
+    interpolation: Literal["linear", "lower", "higher", "nearest", "midpoint"],
+    result_mask: np.ndarray | None,
+    is_datetimelike: bool,
+) -> None: ...
+def group_last(
+    out: np.ndarray,  # rank_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[rank_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    mask: npt.NDArray[np.bool_] | None,
+    result_mask: npt.NDArray[np.bool_] | None = ...,
+    min_count: int = ...,  # Py_ssize_t
+    is_datetimelike: bool = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_nth(
+    out: np.ndarray,  # rank_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[rank_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    mask: npt.NDArray[np.bool_] | None,
+    result_mask: npt.NDArray[np.bool_] | None = ...,
+    min_count: int = ...,  # int64_t
+    rank: int = ...,  # int64_t
+    is_datetimelike: bool = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_rank(
+    out: np.ndarray,  # float64_t[:, ::1]
+    values: np.ndarray,  # ndarray[rank_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    is_datetimelike: bool,
+    ties_method: Literal["average", "min", "max", "first", "dense"] = ...,
+    ascending: bool = ...,
+    pct: bool = ...,
+    na_option: Literal["keep", "top", "bottom"] = ...,
+    mask: npt.NDArray[np.bool_] | None = ...,
+) -> None: ...
+def group_max(
+    out: np.ndarray,  # groupby_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[groupby_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    min_count: int = ...,
+    is_datetimelike: bool = ...,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_min(
+    out: np.ndarray,  # groupby_t[:, ::1]
+    counts: np.ndarray,  # int64_t[::1]
+    values: np.ndarray,  # ndarray[groupby_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    min_count: int = ...,
+    is_datetimelike: bool = ...,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_idxmin_idxmax(
+    out: npt.NDArray[np.intp],
+    counts: npt.NDArray[np.int64],
+    values: np.ndarray,  # ndarray[groupby_t, ndim=2]
+    labels: npt.NDArray[np.intp],
+    min_count: int = ...,
+    is_datetimelike: bool = ...,
+    mask: np.ndarray | None = ...,
+    name: str = ...,
+    skipna: bool = ...,
+    result_mask: np.ndarray | None = ...,
+) -> None: ...
+def group_cummin(
+    out: np.ndarray,  # groupby_t[:, ::1]
+    values: np.ndarray,  # ndarray[groupby_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    is_datetimelike: bool,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
+def group_cummax(
+    out: np.ndarray,  # groupby_t[:, ::1]
+    values: np.ndarray,  # ndarray[groupby_t, ndim=2]
+    labels: np.ndarray,  # const int64_t[:]
+    ngroups: int,
+    is_datetimelike: bool,
+    mask: np.ndarray | None = ...,
+    result_mask: np.ndarray | None = ...,
+    skipna: bool = ...,
+) -> None: ...
diff --git a/pandas/pandas/_libs/groupby.pyx b/pandas/pandas/_libs/groupby.pyx
new file mode 100644
index 0000000000000..f65fa2368967a
--- /dev/null
+++ b/pandas/pandas/_libs/groupby.pyx
@@ -0,0 +1,2302 @@
+cimport cython
+from cython cimport (
+    Py_ssize_t,
+    floating,
+)
+from libc.math cimport (
+    NAN,
+    sqrt,
+)
+from libc.stdlib cimport (
+    free,
+    malloc,
+)
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    complex64_t,
+    complex128_t,
+    float32_t,
+    float64_t,
+    int8_t,
+    int64_t,
+    intp_t,
+    ndarray,
+    uint8_t,
+    uint64_t,
+)
+
+cnp.import_array()
+
+from pandas._libs cimport util
+from pandas._libs.algos cimport (
+    get_rank_nan_fill_val,
+    kth_smallest_c,
+)
+
+from pandas._libs.algos import (
+    groupsort_indexer,
+    rank_1d,
+    take_2d_axis1_bool_bool,
+    take_2d_axis1_float64_float64,
+)
+
+from pandas._libs.dtypes cimport (
+    numeric_object_t,
+    numeric_t,
+)
+from pandas._libs.missing cimport checknull
+
+
+cdef int64_t NPY_NAT = util.get_nat()
+
+cdef float64_t NaN = <float64_t>np.nan
+
+cdef enum InterpolationEnumType:
+    INTERPOLATION_LINEAR,
+    INTERPOLATION_LOWER,
+    INTERPOLATION_HIGHER,
+    INTERPOLATION_NEAREST,
+    INTERPOLATION_MIDPOINT
+
+
+cdef float64_t median_linear_mask(
+    float64_t* a,
+    int n,
+    uint8_t* mask,
+    bint skipna=True
+) noexcept nogil:
+    cdef:
+        int i, j, na_count = 0
+        float64_t* tmp
+        float64_t result
+
+    if n == 0:
+        return NaN
+
+    # count NAs
+    for i in range(n):
+        if mask[i]:
+            na_count += 1
+
+    if na_count:
+        if na_count == n or not skipna:
+            return NaN
+
+        tmp = <float64_t*>malloc((n - na_count) * sizeof(float64_t))
+        if tmp is NULL:
+            raise MemoryError()
+
+        j = 0
+        for i in range(n):
+            if not mask[i]:
+                tmp[j] = a[i]
+                j += 1
+
+        a = tmp
+        n -= na_count
+
+    result = calc_median_linear(a, n)
+
+    if na_count:
+        free(a)
+
+    return result
+
+
+cdef float64_t median_linear(
+    float64_t* a,
+    int n,
+    bint is_datetimelike=False,
+    bint skipna=True,
+) noexcept nogil:
+    cdef:
+        int i, j, na_count = 0
+        float64_t* tmp
+        float64_t result
+
+    if n == 0:
+        return NaN
+
+    # count NAs
+    if is_datetimelike:
+        for i in range(n):
+            if a[i] == NPY_NAT:
+                na_count += 1
+    else:
+        for i in range(n):
+            if a[i] != a[i]:
+                na_count += 1
+
+    if na_count:
+        if na_count == n or not skipna:
+            return NaN
+
+        tmp = <float64_t*>malloc((n - na_count) * sizeof(float64_t))
+        if tmp is NULL:
+            raise MemoryError()
+
+        j = 0
+        if is_datetimelike:
+            for i in range(n):
+                if a[i] != NPY_NAT:
+                    tmp[j] = a[i]
+                    j += 1
+        else:
+            for i in range(n):
+                if a[i] == a[i]:
+                    tmp[j] = a[i]
+                    j += 1
+
+        a = tmp
+        n -= na_count
+
+    result = calc_median_linear(a, n)
+
+    if na_count:
+        free(a)
+
+    return result
+
+
+cdef float64_t calc_median_linear(float64_t* a, int n) noexcept nogil:
+    cdef:
+        float64_t result
+
+    if n % 2:
+        result = kth_smallest_c(a, n // 2, n)
+    else:
+        result = (kth_smallest_c(a, n // 2, n) +
+                  kth_smallest_c(a, n // 2 - 1, n)) / 2
+
+    return result
+
+
+ctypedef fused int64float_t:
+    int64_t
+    uint64_t
+    float32_t
+    float64_t
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_median_float64(
+    ndarray[float64_t, ndim=2] out,
+    ndarray[int64_t] counts,
+    ndarray[float64_t, ndim=2] values,
+    ndarray[intp_t] labels,
+    Py_ssize_t min_count=-1,
+    const uint8_t[:, :] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint is_datetimelike=False,
+    bint skipna=True,
+) -> None:
+    """
+    Only aggregates on axis=0
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, ngroups, size
+        ndarray[intp_t] _counts
+        ndarray[float64_t, ndim=2] data
+        ndarray[uint8_t, ndim=2] data_mask
+        ndarray[intp_t] indexer
+        float64_t* ptr
+        uint8_t* ptr_mask
+        float64_t result
+        bint uses_mask = mask is not None
+
+    assert min_count == -1, "'min_count' only used in sum and prod"
+
+    ngroups = len(counts)
+    N, K = (<object>values).shape
+
+    indexer, _counts = groupsort_indexer(labels, ngroups)
+    counts[:] = _counts[1:]
+
+    data = np.empty((K, N), dtype=np.float64)
+    ptr = <float64_t*>cnp.PyArray_DATA(data)
+
+    take_2d_axis1_float64_float64(values.T, indexer, out=data)
+
+    if uses_mask:
+        data_mask = np.empty((K, N), dtype=np.uint8)
+        ptr_mask = <uint8_t *>cnp.PyArray_DATA(data_mask)
+
+        take_2d_axis1_bool_bool(mask.T, indexer, out=data_mask, fill_value=1)
+
+        with nogil:
+
+            for i in range(K):
+                # exclude NA group
+                ptr += _counts[0]
+                ptr_mask += _counts[0]
+
+                for j in range(ngroups):
+                    size = _counts[j + 1]
+                    result = median_linear_mask(ptr, size, ptr_mask, skipna)
+                    out[j, i] = result
+
+                    if result != result:
+                        result_mask[j, i] = 1
+                    ptr += size
+                    ptr_mask += size
+
+    else:
+        with nogil:
+            for i in range(K):
+                # exclude NA group
+                ptr += _counts[0]
+                for j in range(ngroups):
+                    size = _counts[j + 1]
+                    out[j, i] = median_linear(ptr, size, is_datetimelike, skipna)
+                    ptr += size
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_cumprod(
+    int64float_t[:, ::1] out,
+    ndarray[int64float_t, ndim=2] values,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    bint skipna=True,
+    const uint8_t[:, :] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+) -> None:
+    """
+    Cumulative product of columns of `values`, in row groups `labels`.
+
+    Parameters
+    ----------
+    out : np.ndarray[np.float64, ndim=2]
+        Array to store cumprod in.
+    values : np.ndarray[np.float64, ndim=2]
+        Values to take cumprod of.
+    labels : np.ndarray[np.intp]
+        Labels to group by.
+    ngroups : int
+        Number of groups, larger than all entries of `labels`.
+    is_datetimelike : bool
+        Always false, `values` is never datetime-like.
+    skipna : bool
+        If true, ignore nans in `values`.
+    mask : np.ndarray[uint8], optional
+        Mask of values
+    result_mask : np.ndarray[int8], optional
+        Mask of out array
+
+    Notes
+    -----
+    This method modifies the `out` parameter, rather than returning an object.
+    """
+    cdef:
+        Py_ssize_t i, j, N, K
+        int64float_t val, na_val
+        int64float_t[:, ::1] accum
+        intp_t lab
+        uint8_t[:, ::1] accum_mask
+        bint isna_entry, isna_prev = False
+        bint uses_mask = mask is not None
+
+    N, K = (<object>values).shape
+    accum = np.ones((ngroups, K), dtype=(<object>values).dtype)
+    na_val = _get_na_val(<int64float_t>0, is_datetimelike)
+    accum_mask = np.zeros((ngroups, K), dtype="uint8")
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+
+            if lab < 0:
+                continue
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, False)
+
+                if not isna_entry:
+                    isna_prev = accum_mask[lab, j]
+                    if isna_prev:
+                        out[i, j] = na_val
+                        if uses_mask:
+                            result_mask[i, j] = True
+
+                    else:
+                        accum[lab, j] *= val
+                        out[i, j] = accum[lab, j]
+
+                else:
+                    if uses_mask:
+                        result_mask[i, j] = True
+                        out[i, j] = 0
+                    else:
+                        out[i, j] = na_val
+
+                    if not skipna:
+                        accum[lab, j] = na_val
+                        accum_mask[lab, j] = True
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_cumsum(
+    int64float_t[:, ::1] out,
+    ndarray[int64float_t, ndim=2] values,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    bint skipna=True,
+    const uint8_t[:, :] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+) -> None:
+    """
+    Cumulative sum of columns of `values`, in row groups `labels`.
+
+    Parameters
+    ----------
+    out : np.ndarray[ndim=2]
+        Array to store cumsum in.
+    values : np.ndarray[ndim=2]
+        Values to take cumsum of.
+    labels : np.ndarray[np.intp]
+        Labels to group by.
+    ngroups : int
+        Number of groups, larger than all entries of `labels`.
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    skipna : bool
+        If true, ignore nans in `values`.
+    mask : np.ndarray[uint8], optional
+        Mask of values
+    result_mask : np.ndarray[int8], optional
+        Mask of out array
+
+    Notes
+    -----
+    This method modifies the `out` parameter, rather than returning an object.
+    """
+    cdef:
+        Py_ssize_t i, j, N, K
+        int64float_t val, y, t, na_val
+        int64float_t[:, ::1] accum, compensation
+        uint8_t[:, ::1] accum_mask
+        intp_t lab
+        bint isna_entry, isna_prev = False
+        bint uses_mask = mask is not None
+
+    N, K = (<object>values).shape
+
+    if uses_mask:
+        accum_mask = np.zeros((ngroups, K), dtype="uint8")
+
+    accum = np.zeros((ngroups, K), dtype=np.asarray(values).dtype)
+    compensation = np.zeros((ngroups, K), dtype=np.asarray(values).dtype)
+
+    na_val = _get_na_val(<int64float_t>0, is_datetimelike)
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+
+            if uses_mask and lab < 0:
+                # GH#58811
+                result_mask[i, :] = True
+                out[i, :] = 0
+                continue
+            elif lab < 0:
+                continue
+
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_prev = accum_mask[lab, j]
+                    else:
+                        isna_prev = _treat_as_na(accum[lab, j], is_datetimelike)
+
+                    if isna_prev:
+                        if uses_mask:
+                            result_mask[i, j] = True
+                            # Be deterministic, out was initialized as empty
+                            out[i, j] = 0
+                        else:
+                            out[i, j] = na_val
+                        continue
+
+                if isna_entry:
+
+                    if uses_mask:
+                        result_mask[i, j] = True
+                        # Be deterministic, out was initialized as empty
+                        out[i, j] = 0
+                    else:
+                        out[i, j] = na_val
+
+                    if not skipna:
+                        if uses_mask:
+                            accum_mask[lab, j] = True
+                        else:
+                            accum[lab, j] = na_val
+
+                else:
+                    # For floats, use Kahan summation to reduce floating-point
+                    # error (https://en.wikipedia.org/wiki/Kahan_summation_algorithm)
+                    if int64float_t == float32_t or int64float_t == float64_t:
+                        y = val - compensation[lab, j]
+                        t = accum[lab, j] + y
+                        compensation[lab, j] = t - accum[lab, j] - y
+                    else:
+                        t = val + accum[lab, j]
+
+                    accum[lab, j] = t
+                    out[i, j] = t
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_shift_indexer(
+    int64_t[::1] out,
+    const intp_t[::1] labels,
+    int ngroups,
+    int periods,
+) -> None:
+    cdef:
+        Py_ssize_t N, i, ii, lab
+        int offset = 0, sign
+        int64_t idxer, idxer_slot
+        int64_t[::1] label_seen = np.zeros(ngroups, dtype=np.int64)
+        int64_t[:, ::1] label_indexer
+
+    N, = (<object>labels).shape
+
+    if periods < 0:
+        periods = -periods
+        offset = N - 1
+        sign = -1
+    elif periods > 0:
+        offset = 0
+        sign = 1
+
+    if periods == 0:
+        with nogil:
+            for i in range(N):
+                out[i] = i
+    else:
+        # array of each previous indexer seen
+        label_indexer = np.zeros((ngroups, periods), dtype=np.int64)
+        with nogil:
+            for i in range(N):
+                # reverse iterator if shifting backwards
+                ii = offset + sign * i
+                lab = labels[ii]
+
+                # Skip null keys
+                if lab == -1:
+                    out[ii] = -1
+                    continue
+
+                label_seen[lab] += 1
+
+                idxer_slot = label_seen[lab] % periods
+                idxer = label_indexer[lab, idxer_slot]
+
+                if label_seen[lab] > periods:
+                    out[ii] = idxer
+                else:
+                    out[ii] = -1
+
+                label_indexer[lab, idxer_slot] = ii
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_fillna_indexer(
+    Py_ssize_t[::1] out,
+    const intp_t[::1] labels,
+    const uint8_t[:] mask,
+    int64_t limit,
+    bint compute_ffill,
+    int ngroups,
+) -> None:
+    """
+    Indexes how to fill values forwards or backwards within a group.
+
+    Parameters
+    ----------
+    out : np.ndarray[np.intp]
+        Values into which this method will write its results.
+    labels : np.ndarray[np.intp]
+        Array containing unique label for each group, with its ordering
+        matching up to the corresponding record in `values`.
+    mask : np.ndarray[np.uint8]
+        Indicating whether a value is na or not.
+    limit : int64_t
+        Consecutive values to fill before stopping, or -1 for no limit.
+    compute_ffill : bint
+        Whether to compute ffill or bfill.
+    ngroups : int
+        Number of groups, larger than all entries of `labels`.
+
+    Notes
+    -----
+    This method modifies the `out` parameter rather than returning an object
+    """
+    cdef:
+        Py_ssize_t idx, N = len(out)
+        intp_t label
+        intp_t[::1] last = -1 * np.ones(ngroups, dtype=np.intp)
+        intp_t[::1] fill_count = np.zeros(ngroups, dtype=np.intp)
+
+    # Make sure all arrays are the same size
+    assert N == len(labels) == len(mask)
+
+    with nogil:
+        # Can't use for loop with +/- step
+        # https://github.com/cython/cython/issues/1106
+        idx = 0 if compute_ffill else N-1
+        for _ in range(N):
+            label = labels[idx]
+            if label == -1:  # na-group gets na-values
+                out[idx] = -1
+            elif mask[idx] == 1:  # is missing
+                # Stop filling once we've hit the limit
+                if limit != -1 and fill_count[label] >= limit:
+                    out[idx] = -1
+                else:
+                    out[idx] = last[label]
+                    fill_count[label] += 1
+            else:
+                fill_count[label] = 0  # reset items when not missing
+                last[label] = idx
+                out[idx] = idx
+
+            if compute_ffill:
+                idx += 1
+            else:
+                idx -= 1
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_any_all(
+    int8_t[:, ::1] out,
+    const int8_t[:, :] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, :] mask,
+    str val_test,
+    bint skipna,
+    uint8_t[:, ::1] result_mask,
+) -> None:
+    """
+    Aggregated boolean values to show truthfulness of group elements. If the
+    input is a nullable type (result_mask is not None), the result will be computed
+    using Kleene logic.
+
+    Parameters
+    ----------
+    out : np.ndarray[np.int8]
+        Values into which this method will write its results.
+    labels : np.ndarray[np.intp]
+        Array containing unique label for each group, with its
+        ordering matching up to the corresponding record in `values`
+    values : np.ndarray[np.int8]
+        Containing the truth value of each element.
+    mask : np.ndarray[np.uint8]
+        Indicating whether a value is na or not.
+    val_test : {'any', 'all'}
+        String object dictating whether to use any or all truth testing
+    skipna : bool
+        Flag to ignore nan values during truth testing
+    result_mask : ndarray[bool, ndim=2], optional
+        If not None, these specify locations in the output that are NA.
+        Modified in-place.
+
+    Notes
+    -----
+    This method modifies the `out` parameter rather than returning an object.
+    The returned values will either be 0, 1 (False or True, respectively), or
+    -1 to signify a masked position in the case of a nullable input.
+    """
+    cdef:
+        Py_ssize_t i, j, N = len(labels), K = out.shape[1]
+        intp_t lab
+        int8_t flag_val, val
+        bint uses_mask = result_mask is not None
+
+    if val_test == "all":
+        # Because the 'all' value of an empty iterable in Python is True we can
+        # start with an array full of ones and set to zero when a False value
+        # is encountered
+        flag_val = 0
+    elif val_test == "any":
+        # Because the 'any' value of an empty iterable in Python is False we
+        # can start with an array full of zeros and set to one only if any
+        # value encountered is True
+        flag_val = 1
+    else:
+        raise ValueError("'val_test' must be either 'any' or 'all'!")
+
+    out[:] = 1 - flag_val
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            for j in range(K):
+                if skipna and mask[i, j]:
+                    continue
+
+                if uses_mask and mask[i, j]:
+                    # Set the position as masked if `out[lab] != flag_val`, which
+                    # would indicate True/False has not yet been seen for any/all,
+                    # so by Kleene logic the result is currently unknown
+                    if out[lab, j] != flag_val:
+                        result_mask[lab, j] = 1
+                    continue
+
+                val = values[i, j]
+
+                # If True and 'any' or False and 'all', the result is
+                # already determined
+                if val == flag_val:
+                    out[lab, j] = flag_val
+                    if uses_mask:
+                        result_mask[lab, j] = 0
+
+
+# ----------------------------------------------------------------------
+# group_sum, group_prod, group_var, group_mean, group_ohlc
+# ----------------------------------------------------------------------
+
+ctypedef fused mean_t:
+    float64_t
+    float32_t
+    complex64_t
+    complex128_t
+
+ctypedef fused sum_t:
+    mean_t
+    int64_t
+    uint64_t
+    object
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_sum(
+    sum_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[sum_t, ndim=2] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, :] mask,
+    uint8_t[:, ::1] result_mask=None,
+    Py_ssize_t min_count=0,
+    bint is_datetimelike=False,
+    bint skipna=True,
+) -> None:
+    """
+    Only aggregates on axis=0 using Kahan summation
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        sum_t val, t, y, nan_val
+        sum_t[:, ::1] sumx, compensation
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint uses_mask = mask is not None
+        bint isna_entry, isna_result
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    # the below is equivalent to `np.zeros_like(out)` but faster
+    sumx = np.zeros((<object>out).shape, dtype=(<object>out).base.dtype)
+    compensation = np.zeros((<object>out).shape, dtype=(<object>out).base.dtype)
+
+    N, K = (<object>values).shape
+    if uses_mask:
+        nan_val = 0
+    elif is_datetimelike:
+        nan_val = NPY_NAT
+    elif sum_t is int64_t or sum_t is uint64_t:
+        # This has no effect as int64 can't be nan. Setting to 0 to avoid type error
+        nan_val = 0
+    else:
+        nan_val = NAN
+
+    with nogil(sum_t is not object):
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_result = result_mask[lab, j]
+                    else:
+                        isna_result = _treat_as_na(sumx[lab, j], is_datetimelike)
+
+                    if isna_result:
+                        # If sum is already NA, don't add to it. This is important for
+                        # datetimelikebecause adding a value to NPY_NAT may not result
+                        # in a NPY_NAT
+                        continue
+
+                if not isna_entry:
+                    nobs[lab, j] += 1
+
+                    if sum_t is object:
+                        # NB: this does not use 'compensation' like the non-object
+                        #  track does.
+                        if nobs[lab, j] == 1:
+                            # i.e. we haven't added anything yet; avoid TypeError
+                            #  if e.g. val is a str and sumx[lab, j] is 0
+                            t = val
+                        else:
+                            t = sumx[lab, j] + val
+                        sumx[lab, j] = t
+
+                    else:
+                        y = val - compensation[lab, j]
+                        t = sumx[lab, j] + y
+                        compensation[lab, j] = t - sumx[lab, j] - y
+                        if compensation[lab, j] != compensation[lab, j]:
+                            # GH#53606
+                            # If val is +/- infinity compensation is NaN
+                            # which would lead to results being NaN instead
+                            # of +/- infinity. We cannot use util.is_nan
+                            # because of no gil
+                            compensation[lab, j] = 0
+                        sumx[lab, j] = t
+                elif not skipna:
+                    if uses_mask:
+                        result_mask[lab, j] = True
+                    else:
+                        sumx[lab, j] = nan_val
+
+    _check_below_mincount(
+        out, uses_mask, result_mask, ncounts, K, nobs, min_count, sumx
+    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_prod(
+    int64float_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[int64float_t, ndim=2] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, ::1] mask,
+    uint8_t[:, ::1] result_mask=None,
+    Py_ssize_t min_count=0,
+    bint skipna=True,
+) -> None:
+    """
+    Only aggregates on axis=0
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        int64float_t val, nan_val
+        int64float_t[:, ::1] prodx
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint isna_entry, isna_result, uses_mask = mask is not None
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    prodx = np.ones((<object>out).shape, dtype=(<object>out).base.dtype)
+
+    N, K = (<object>values).shape
+    nan_val = _get_na_val(<int64float_t>0, False)
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, False)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_result = result_mask[lab, j]
+                    else:
+                        isna_result = _treat_as_na(prodx[lab, j], False)
+
+                    if isna_result:
+                        # If prod is already NA, no need to update it
+                        continue
+
+                if not isna_entry:
+                    nobs[lab, j] += 1
+                    prodx[lab, j] *= val
+                elif not skipna:
+                    if uses_mask:
+                        result_mask[lab, j] = True
+                    else:
+                        prodx[lab, j] = nan_val
+
+    _check_below_mincount(
+        out, uses_mask, result_mask, ncounts, K, nobs, min_count, prodx
+    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+@cython.cdivision(True)
+def group_var(
+    floating[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[floating, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    int64_t ddof=1,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint is_datetimelike=False,
+    str name="var",
+    bint skipna=True,
+) -> None:
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        floating val, ct, oldmean
+        floating[:, ::1] mean
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint isna_entry, isna_result, uses_mask = mask is not None
+        bint is_std = name == "std"
+        bint is_sem = name == "sem"
+
+    assert min_count == -1, "'min_count' only used in sum and prod"
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    mean = np.zeros((<object>out).shape, dtype=(<object>out).base.dtype)
+
+    N, K = (<object>values).shape
+
+    out[:, :] = 0.0
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                elif is_datetimelike:
+                    # With group_var, we cannot just use _treat_as_na bc
+                    #  datetimelike dtypes get cast to float64 instead of
+                    #  to int64.
+                    isna_entry = val == NPY_NAT
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_result = result_mask[lab, j]
+                    elif is_datetimelike:
+                        # With group_var, we cannot just use _treat_as_na bc
+                        #  datetimelike dtypes get cast to float64 instead of
+                        #  to int64.
+                        isna_result = out[lab, j] == NPY_NAT
+                    else:
+                        isna_result = _treat_as_na(out[lab, j], is_datetimelike)
+
+                    if isna_result:
+                        # If aggregate is already NA, don't add to it. This is
+                        # important for datetimelike because adding a value to NPY_NAT
+                        # may not result in a NPY_NAT
+                        continue
+
+                if not isna_entry:
+                    nobs[lab, j] += 1
+                    oldmean = mean[lab, j]
+                    mean[lab, j] += (val - oldmean) / nobs[lab, j]
+                    out[lab, j] += (val - mean[lab, j]) * (val - oldmean)
+                elif not skipna:
+                    nobs[lab, j] = 0
+                    if uses_mask:
+                        result_mask[lab, j] = True
+                    else:
+                        out[lab, j] = NAN
+
+        for i in range(ncounts):
+            for j in range(K):
+                ct = nobs[i, j]
+                if ct <= ddof:
+                    if uses_mask:
+                        result_mask[i, j] = True
+                    else:
+                        out[i, j] = NAN
+                else:
+                    if is_std:
+                        out[i, j] = sqrt(out[i, j] / (ct - ddof))
+                    elif is_sem:
+                        out[i, j] = sqrt(out[i, j] / (ct - ddof) / ct)
+                    else:
+                        # just "var"
+                        out[i, j] /= (ct - ddof)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+@cython.cdivision(True)
+@cython.cpow(True)
+def group_skew(
+    float64_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[float64_t, ndim=2] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ngroups = len(counts)
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint isna_entry, uses_mask = mask is not None
+        float64_t[:, ::1] M1, M2, M3
+        float64_t delta, delta_n, term1, val
+        int64_t n1, n
+        float64_t ct
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+
+    # M1, M2, and M3 correspond to 1st, 2nd, and third Moments
+    M1 = np.zeros((<object>out).shape, dtype=np.float64)
+    M2 = np.zeros((<object>out).shape, dtype=np.float64)
+    M3 = np.zeros((<object>out).shape, dtype=np.float64)
+
+    N, K = (<object>values).shape
+
+    out[:, :] = 0.0
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, False)
+
+                if not isna_entry:
+                    # Running stats update based on RunningStats::Push from
+                    #  https://www.johndcook.com/blog/skewness_kurtosis/
+                    n1 = nobs[lab, j]
+                    n = n1 + 1
+
+                    nobs[lab, j] = n
+                    delta = val - M1[lab, j]
+                    delta_n = delta / n
+                    term1 = delta * delta_n * n1
+
+                    M1[lab, j] += delta_n
+                    M3[lab, j] += term1 * delta_n * (n - 2) - 3 * delta_n * M2[lab, j]
+                    M2[lab, j] += term1
+                elif not skipna:
+                    M1[lab, j] = NaN
+                    M2[lab, j] = NaN
+                    M3[lab, j] = NaN
+
+        for i in range(ngroups):
+            for j in range(K):
+                ct = <float64_t>nobs[i, j]
+                if ct < 3:
+                    if result_mask is not None:
+                        result_mask[i, j] = 1
+                    out[i, j] = NaN
+                elif M2[i, j] == 0:
+                    out[i, j] = 0
+                else:
+                    out[i, j] = (
+                        (ct * (ct - 1) ** 0.5 / (ct - 2))
+                        * (M3[i, j] / M2[i, j] ** 1.5)
+                    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+@cython.cdivision(True)
+@cython.cpow(True)
+def group_kurt(
+    float64_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[float64_t, ndim=2] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ngroups = len(counts)
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint isna_entry, uses_mask = mask is not None
+        float64_t[:, ::1] M1, M2, M3, M4
+        float64_t delta, delta_n, delta_n2, term1, val
+        int64_t n1, n
+        float64_t ct, num, den, adj
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+
+    # M1, M2, M3 and M4 correspond to 1st, 2nd, 3rd and 4th Moments
+    M1 = np.zeros((<object>out).shape, dtype=np.float64)
+    M2 = np.zeros((<object>out).shape, dtype=np.float64)
+    M3 = np.zeros((<object>out).shape, dtype=np.float64)
+    M4 = np.zeros((<object>out).shape, dtype=np.float64)
+
+    N, K = (<object>values).shape
+
+    out[:, :] = 0.0
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, False)
+
+                if not isna_entry:
+                    # Running stats update based on RunningStats::Push from
+                    #  https://www.johndcook.com/blog/skewness_kurtosis/
+                    n1 = nobs[lab, j]
+                    n = n1 + 1
+
+                    nobs[lab, j] = n
+                    delta = val - M1[lab, j]
+                    delta_n = delta / n
+                    delta_n2 = delta_n * delta_n
+                    term1 = delta * delta_n * n1
+
+                    M1[lab, j] += delta_n
+                    M4[lab, j] += (term1 * delta_n2 * (n*n - 3*n + 3)
+                                   + 6 * delta_n2 * M2[lab, j]
+                                   - 4 * delta_n * M3[lab, j])
+                    M3[lab, j] += term1 * delta_n * (n - 2) - 3 * delta_n * M2[lab, j]
+                    M2[lab, j] += term1
+                elif not skipna:
+                    M1[lab, j] = NaN
+                    M2[lab, j] = NaN
+                    M3[lab, j] = NaN
+                    M4[lab, j] = NaN
+
+        for i in range(ngroups):
+            for j in range(K):
+                ct = <float64_t>nobs[i, j]
+                if ct < 4:
+                    if result_mask is not None:
+                        result_mask[i, j] = 1
+                    out[i, j] = NaN
+                elif M2[i, j] == 0:
+                    out[i, j] = 0
+                else:
+                    num = ct * (ct + 1) * (ct - 1) * M4[i, j]
+                    den = (ct - 2) * (ct - 3) * M2[i, j] ** 2
+                    adj = 3.0 * (ct - 1) ** 2 / ((ct - 2) * (ct - 3))
+                    out[i, j] = num / den - adj
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_mean(
+    mean_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[mean_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    """
+    Compute the mean per label given a label assignment for each value.
+    NaN values are ignored.
+
+    Parameters
+    ----------
+    out : np.ndarray[floating]
+        Values into which this method will write its results.
+    counts : np.ndarray[int64]
+        A zeroed array of the same shape as labels,
+        populated by group sizes during algorithm.
+    values : np.ndarray[floating]
+        2-d array of the values to find the mean of.
+    labels : np.ndarray[np.intp]
+        Array containing unique label for each group, with its
+        ordering matching up to the corresponding record in `values`.
+    min_count : Py_ssize_t
+        Only used in sum and prod. Always -1.
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    mask : ndarray[bool, ndim=2], optional
+        Mask of the input values.
+    result_mask : ndarray[bool, ndim=2], optional
+        Mask of the out array
+    skipna : bool, optional
+        If True, ignore nans in `values`.
+
+    Notes
+    -----
+    This method modifies the `out` parameter rather than returning an object.
+    `counts` is modified to hold group sizes
+    """
+
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        mean_t val, count, y, t, nan_val
+        mean_t[:, ::1] sumx, compensation
+        int64_t[:, ::1] nobs
+        Py_ssize_t len_values = len(values), len_labels = len(labels)
+        bint isna_entry, isna_result, uses_mask = mask is not None
+
+    assert min_count == -1, "'min_count' only used in sum and prod"
+
+    if len_values != len_labels:
+        raise ValueError("len(index) != len(labels)")
+
+    # the below is equivalent to `np.zeros_like(out)` but faster
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    sumx = np.zeros((<object>out).shape, dtype=(<object>out).base.dtype)
+    compensation = np.zeros((<object>out).shape, dtype=(<object>out).base.dtype)
+
+    N, K = (<object>values).shape
+    if uses_mask:
+        nan_val = 0
+    elif is_datetimelike:
+        nan_val = NPY_NAT
+    else:
+        nan_val = NAN
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                elif is_datetimelike:
+                    # With group_mean, we cannot just use _treat_as_na bc
+                    #  datetimelike dtypes get cast to float64 instead of
+                    #  to int64.
+                    isna_entry = val == NPY_NAT
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_result = result_mask[lab, j]
+                    elif is_datetimelike:
+                        # With group_mean, we cannot just use _treat_as_na bc
+                        #  datetimelike dtypes get cast to float64 instead of
+                        #  to int64.
+                        isna_result = sumx[lab, j] == NPY_NAT
+                    else:
+                        isna_result = _treat_as_na(sumx[lab, j], is_datetimelike)
+
+                    if isna_result:
+                        # If sum is already NA, don't add to it. This is important for
+                        # datetimelike because adding a value to NPY_NAT may not result
+                        # in NPY_NAT
+                        continue
+
+                if not isna_entry:
+                    nobs[lab, j] += 1
+                    y = val - compensation[lab, j]
+                    t = sumx[lab, j] + y
+                    compensation[lab, j] = t - sumx[lab, j] - y
+                    if compensation[lab, j] != compensation[lab, j]:
+                        # GH#50367
+                        # If val is +/- infinity, compensation is NaN
+                        # which would lead to results being NaN instead
+                        # of +/-infinity. We cannot use util.is_nan
+                        # because of no gil
+                        compensation[lab, j] = 0.
+                    sumx[lab, j] = t
+                elif not skipna:
+                    # Set the nobs to 0 so that in case of datetimelike,
+                    # dividing NPY_NAT by nobs may not result in a NPY_NAT
+                    nobs[lab, j] = 0
+                    if uses_mask:
+                        result_mask[lab, j] = True
+                    else:
+                        sumx[lab, j] = nan_val
+
+        for i in range(ncounts):
+            for j in range(K):
+                count = nobs[i, j]
+                if nobs[i, j] == 0:
+
+                    if uses_mask:
+                        result_mask[i, j] = True
+                    else:
+                        out[i, j] = nan_val
+
+                else:
+                    out[i, j] = sumx[i, j] / count
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_ohlc(
+    int64float_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[int64float_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+) -> None:
+    """
+    Only aggregates on axis=0
+    """
+    cdef:
+        Py_ssize_t i, N, K, lab
+        int64float_t val
+        uint8_t[::1] first_element_set
+        bint isna_entry, uses_mask = mask is not None
+
+    assert min_count == -1, "'min_count' only used in sum and prod"
+
+    if len(labels) == 0:
+        return
+
+    N, K = (<object>values).shape
+
+    if out.shape[1] != 4:
+        raise ValueError("Output array must have 4 columns")
+
+    if K > 1:
+        raise NotImplementedError("Argument 'values' must have only one dimension")
+
+    if int64float_t is float32_t or int64float_t is float64_t:
+        out[:] = NAN
+    else:
+        out[:] = 0
+
+    first_element_set = np.zeros((<object>counts).shape, dtype=np.uint8)
+    if uses_mask:
+        result_mask[:] = True
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab == -1:
+                continue
+
+            counts[lab] += 1
+            val = values[i, 0]
+
+            if uses_mask:
+                isna_entry = mask[i, 0]
+            else:
+                isna_entry = _treat_as_na(val, False)
+
+            if isna_entry:
+                continue
+
+            if not first_element_set[lab]:
+                out[lab, 0] = out[lab, 1] = out[lab, 2] = out[lab, 3] = val
+                first_element_set[lab] = True
+                if uses_mask:
+                    result_mask[lab] = False
+            else:
+                out[lab, 1] = max(out[lab, 1], val)
+                out[lab, 2] = min(out[lab, 2], val)
+                out[lab, 3] = val
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_quantile(
+    float64_t[:, ::1] out,
+    ndarray[numeric_t, ndim=1] values,
+    const intp_t[::1] labels,
+    const uint8_t[:] mask,
+    const float64_t[:] qs,
+    const int64_t[::1] starts,
+    const int64_t[::1] ends,
+    str interpolation,
+    uint8_t[:, ::1] result_mask,
+    bint is_datetimelike,
+) -> None:
+    """
+    Calculate the quantile per group.
+
+    Parameters
+    ----------
+    out : np.ndarray[np.float64, ndim=2]
+        Array of aggregated values that will be written to.
+    values : np.ndarray
+        Array containing the values to apply the function against.
+    labels : ndarray[np.intp]
+        Array containing the unique group labels.
+    qs : ndarray[float64_t]
+        The quantile values to search for.
+    starts : ndarray[int64]
+        Positions at which each group begins.
+    ends : ndarray[int64]
+        Positions at which each group ends.
+    interpolation : {'linear', 'lower', 'highest', 'nearest', 'midpoint'}
+    result_mask : ndarray[bool, ndim=2] or None
+    is_datetimelike : bool
+        Whether int64 values represent datetime64-like values.
+
+    Notes
+    -----
+    Rather than explicitly returning a value, this function modifies the
+    provided `out` parameter.
+    """
+    cdef:
+        Py_ssize_t i, N=len(labels), ngroups, non_na_sz, k, nqs
+        Py_ssize_t idx=0
+        Py_ssize_t grp_size
+        InterpolationEnumType interp
+        float64_t q_val, q_idx, frac, val, next_val
+        bint uses_result_mask = result_mask is not None
+        Py_ssize_t start, end
+        ndarray[numeric_t] grp
+        intp_t[::1] sort_indexer
+        const uint8_t[:] sub_mask
+
+    assert values.shape[0] == N
+    assert starts is not None
+    assert ends is not None
+    assert len(starts) == len(ends)
+
+    if any(not (0 <= q <= 1) for q in qs):
+        wrong = [x for x in qs if not (0 <= x <= 1)][0]
+        raise ValueError(
+            f"Each 'q' must be between 0 and 1. Got '{wrong}' instead"
+        )
+
+    inter_methods = {
+        "linear": INTERPOLATION_LINEAR,
+        "lower": INTERPOLATION_LOWER,
+        "higher": INTERPOLATION_HIGHER,
+        "nearest": INTERPOLATION_NEAREST,
+        "midpoint": INTERPOLATION_MIDPOINT,
+    }
+    interp = inter_methods[interpolation]
+
+    nqs = len(qs)
+    ngroups = len(out)
+
+    # TODO: get cnp.PyArray_ArgSort to work with nogil so we can restore the rest
+    #  of this function as being `with nogil:`
+    for i in range(ngroups):
+        start = starts[i]
+        end = ends[i]
+
+        grp = values[start:end]
+
+        # Figure out how many group elements there are
+        sub_mask = mask[start:end]
+        grp_size = sub_mask.size
+        non_na_sz = 0
+        for k in range(grp_size):
+            if sub_mask[k] == 0:
+                non_na_sz += 1
+
+        # equiv: sort_indexer = grp.argsort()
+        if is_datetimelike:
+            # We need the argsort to put NaTs at the end, not the beginning
+            sort_indexer = cnp.PyArray_ArgSort(grp.view("M8[ns]"), 0, cnp.NPY_QUICKSORT)
+        else:
+            sort_indexer = cnp.PyArray_ArgSort(grp, 0, cnp.NPY_QUICKSORT)
+
+        if non_na_sz == 0:
+            for k in range(nqs):
+                if uses_result_mask:
+                    result_mask[i, k] = 1
+                else:
+                    out[i, k] = NaN
+        else:
+            for k in range(nqs):
+                q_val = qs[k]
+
+                # Calculate where to retrieve the desired value
+                # Casting to int will intentionally truncate result
+                idx = <int64_t>(q_val * <float64_t>(non_na_sz - 1))
+
+                val = grp[sort_indexer[idx]]
+                # If requested quantile falls evenly on a particular index
+                # then write that index's value out. Otherwise interpolate
+                q_idx = q_val * (non_na_sz - 1)
+                frac = q_idx % 1
+
+                if frac == 0.0 or interp == INTERPOLATION_LOWER:
+                    out[i, k] = val
+                else:
+                    next_val = grp[sort_indexer[idx + 1]]
+                    if interp == INTERPOLATION_LINEAR:
+                        out[i, k] = val + (next_val - val) * frac
+                    elif interp == INTERPOLATION_HIGHER:
+                        out[i, k] = next_val
+                    elif interp == INTERPOLATION_MIDPOINT:
+                        out[i, k] = (val + next_val) / 2.0
+                    elif interp == INTERPOLATION_NEAREST:
+                        if frac > .5 or (frac == .5 and idx % 2 == 1):
+                            # If quantile lies in the middle of two indexes,
+                            # take the even index, as np.quantile.
+                            out[i, k] = next_val
+                        else:
+                            out[i, k] = val
+
+
+# ----------------------------------------------------------------------
+# group_nth, group_last, group_rank
+# ----------------------------------------------------------------------
+
+ctypedef fused numeric_object_complex_t:
+    numeric_object_t
+    complex64_t
+    complex128_t
+
+
+cdef bint _treat_as_na(numeric_object_complex_t val,
+                       bint is_datetimelike) noexcept nogil:
+    if numeric_object_complex_t is object:
+        with gil:
+            return checknull(val)
+
+    elif numeric_object_complex_t is int64_t:
+        return is_datetimelike and val == NPY_NAT
+    elif (
+        numeric_object_complex_t is float32_t
+        or numeric_object_complex_t is float64_t
+        or numeric_object_complex_t is complex64_t
+        or numeric_object_complex_t is complex128_t
+    ):
+        return val != val
+    else:
+        # non-datetimelike integer
+        return False
+
+
+cdef numeric_object_t _get_min_or_max(
+    numeric_object_t val,
+    bint compute_max,
+    bint is_datetimelike,
+):
+    """
+    Find either the min or the max supported by numeric_object_t; 'val' is a
+    placeholder to effectively make numeric_object_t an argument.
+    """
+    return get_rank_nan_fill_val(
+        not compute_max,
+        val=val,
+        is_datetimelike=is_datetimelike,
+    )
+
+
+cdef numeric_t _get_na_val(numeric_t val, bint is_datetimelike):
+    cdef:
+        numeric_t na_val
+
+    if numeric_t == float32_t or numeric_t == float64_t:
+        na_val = NaN
+    elif numeric_t is int64_t and is_datetimelike:
+        na_val = NPY_NAT
+    else:
+        # Used in case of masks
+        na_val = 0
+    return na_val
+
+
+ctypedef fused mincount_t:
+    numeric_object_t
+    complex64_t
+    complex128_t
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef inline void _check_below_mincount(
+    mincount_t[:, ::1] out,
+    bint uses_mask,
+    uint8_t[:, ::1] result_mask,
+    Py_ssize_t ncounts,
+    Py_ssize_t K,
+    const int64_t[:, ::1] nobs,
+    int64_t min_count,
+    mincount_t[:, ::1] resx,
+) noexcept:
+    """
+    Check if the number of observations for a group is below min_count,
+    and if so set the result for that group to the appropriate NA-like value.
+    """
+    cdef:
+        Py_ssize_t i, j
+
+    with nogil(mincount_t is not object):
+        for i in range(ncounts):
+            for j in range(K):
+                if nobs[i, j] >= min_count:
+                    out[i, j] = resx[i, j]
+                else:
+                    # if we are integer dtype, not is_datetimelike, and
+                    #  not uses_mask, then getting here implies that
+                    #  counts[i] < min_count, which means we will
+                    #  be cast to float64 and masked at the end
+                    #  of WrappedCythonOp._call_cython_op. So we can safely
+                    #  set a placeholder value in out[i, j].
+                    if uses_mask:
+                        result_mask[i, j] = True
+                        # set out[i, j] to 0 to be deterministic, as
+                        #  it was initialized with np.empty. Also ensures
+                        #  we can downcast out if appropriate.
+                        out[i, j] = 0
+                    elif (
+                        mincount_t is float32_t
+                        or mincount_t is float64_t
+                        or mincount_t is complex64_t
+                        or mincount_t is complex128_t
+                    ):
+                        out[i, j] = NAN
+                    elif mincount_t is int64_t:
+                        # Per above, this is a placeholder in
+                        #  non-is_datetimelike cases.
+                        out[i, j] = NPY_NAT
+                    elif mincount_t is object:
+                        out[i, j] = None
+                    else:
+                        # placeholder, see above
+                        out[i, j] = 0
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_last(
+    numeric_object_t[:, ::1] out,
+    int64_t[::1] counts,
+    const numeric_object_t[:, :] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, :] mask,
+    uint8_t[:, ::1] result_mask=None,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    bint skipna=True,
+) -> None:
+    """
+    Only aggregates on axis=0
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        numeric_object_t val
+        numeric_object_t[:, ::1] resx
+        int64_t[:, ::1] nobs
+        bint uses_mask = mask is not None
+        bint isna_entry
+
+    if not len(values) == len(labels):
+        raise AssertionError("len(index) != len(labels)")
+
+    min_count = max(min_count, 1)
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    if numeric_object_t is object:
+        resx = np.empty((<object>out).shape, dtype=object)
+    else:
+        resx = np.empty_like(out)
+
+    N, K = (<object>values).shape
+
+    with nogil(numeric_object_t is not object):
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+            for j in range(K):
+                val = values[i, j]
+
+                if skipna:
+                    if uses_mask:
+                        isna_entry = mask[i, j]
+                    else:
+                        isna_entry = _treat_as_na(val, is_datetimelike)
+                    if isna_entry:
+                        continue
+
+                nobs[lab, j] += 1
+                resx[lab, j] = val
+
+                if uses_mask and not skipna:
+                    result_mask[lab, j] = mask[i, j]
+
+    _check_below_mincount(
+        out, uses_mask, result_mask, ncounts, K, nobs, min_count, resx
+    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_nth(
+    numeric_object_t[:, ::1] out,
+    int64_t[::1] counts,
+    const numeric_object_t[:, :] values,
+    const intp_t[::1] labels,
+    const uint8_t[:, :] mask,
+    uint8_t[:, ::1] result_mask=None,
+    int64_t min_count=-1,
+    int64_t rank=1,
+    bint is_datetimelike=False,
+    bint skipna=True,
+) -> None:
+    """
+    Only aggregates on axis=0
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ncounts = len(counts)
+        numeric_object_t val
+        numeric_object_t[:, ::1] resx
+        int64_t[:, ::1] nobs
+        bint uses_mask = mask is not None
+        bint isna_entry
+
+    if not len(values) == len(labels):
+        raise AssertionError("len(index) != len(labels)")
+
+    min_count = max(min_count, 1)
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    if numeric_object_t is object:
+        resx = np.empty((<object>out).shape, dtype=object)
+    else:
+        resx = np.empty_like(out)
+
+    N, K = (<object>values).shape
+
+    with nogil(numeric_object_t is not object):
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+            for j in range(K):
+                val = values[i, j]
+
+                if skipna:
+                    if uses_mask:
+                        isna_entry = mask[i, j]
+                    else:
+                        isna_entry = _treat_as_na(val, is_datetimelike)
+                    if isna_entry:
+                        continue
+
+                nobs[lab, j] += 1
+                if nobs[lab, j] == rank:
+                    resx[lab, j] = val
+                    if uses_mask and not skipna:
+                        result_mask[lab, j] = mask[i, j]
+
+    _check_below_mincount(
+        out, uses_mask, result_mask, ncounts, K, nobs, min_count, resx
+    )
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_rank(
+    float64_t[:, ::1] out,
+    ndarray[numeric_object_t, ndim=2] values,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    str ties_method="average",
+    bint ascending=True,
+    bint pct=False,
+    str na_option="keep",
+    const uint8_t[:, :] mask=None,
+) -> None:
+    """
+    Provides the rank of values within each group.
+
+    Parameters
+    ----------
+    out : np.ndarray[np.float64, ndim=2]
+        Values to which this method will write its results.
+    values : np.ndarray of numeric_object_t values to be ranked
+    labels : np.ndarray[np.intp]
+        Array containing unique label for each group, with its ordering
+        matching up to the corresponding record in `values`
+    ngroups : int
+        This parameter is not used, is needed to match signatures of other
+        groupby functions.
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    ties_method : {'average', 'min', 'max', 'first', 'dense'}, default 'average'
+        * average: average rank of group
+        * min: lowest rank in group
+        * max: highest rank in group
+        * first: ranks assigned in order they appear in the array
+        * dense: like 'min', but rank always increases by 1 between groups
+    ascending : bool, default True
+        False for ranks by high (1) to low (N)
+        na_option : {'keep', 'top', 'bottom'}, default 'keep'
+    pct : bool, default False
+        Compute percentage rank of data within each group
+    na_option : {'keep', 'top', 'bottom'}, default 'keep'
+        * keep: leave NA values where they are
+        * top: smallest rank if ascending
+        * bottom: smallest rank if descending
+    mask : np.ndarray[bool] or None, default None
+
+    Notes
+    -----
+    This method modifies the `out` parameter rather than returning an object
+    """
+    cdef:
+        Py_ssize_t i, k, N
+        ndarray[float64_t, ndim=1] result
+        const uint8_t[:] sub_mask
+
+    N = values.shape[1]
+
+    for k in range(N):
+        if mask is None:
+            sub_mask = None
+        else:
+            sub_mask = mask[:, k]
+
+        result = rank_1d(
+            values=values[:, k],
+            labels=labels,
+            is_datetimelike=is_datetimelike,
+            ties_method=ties_method,
+            ascending=ascending,
+            pct=pct,
+            na_option=na_option,
+            mask=sub_mask,
+        )
+        for i in range(len(result)):
+            if labels[i] >= 0:
+                out[i, k] = result[i]
+
+
+# ----------------------------------------------------------------------
+# group_min, group_max
+# ----------------------------------------------------------------------
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef group_min_max(
+    numeric_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[numeric_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    bint compute_max=True,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+):
+    """
+    Compute minimum/maximum  of columns of `values`, in row groups `labels`.
+
+    Parameters
+    ----------
+    out : np.ndarray[numeric_t, ndim=2]
+        Array to store result in.
+    counts : np.ndarray[int64]
+        Input as a zeroed array, populated by group sizes during algorithm
+    values : array
+        Values to find column-wise min/max of.
+    labels : np.ndarray[np.intp]
+        Labels to group by.
+    min_count : Py_ssize_t, default -1
+        The minimum number of non-NA group elements, NA result if threshold
+        is not met
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    compute_max : bint, default True
+        True to compute group-wise max, False to compute min
+    mask : ndarray[bool, ndim=2], optional
+        If not None, indices represent missing values,
+        otherwise the mask will not be used
+    result_mask : ndarray[bool, ndim=2], optional
+        If not None, these specify locations in the output that are NA.
+        Modified in-place.
+    skipna : bool, default True
+        If True, ignore nans in `values`.
+
+    Notes
+    -----
+    This method modifies the `out` parameter, rather than returning an object.
+    `counts` is modified to hold group sizes
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab, ngroups = len(counts)
+        numeric_t val, nan_val
+        numeric_t[:, ::1] group_min_or_max
+        int64_t[:, ::1] nobs
+        bint uses_mask = mask is not None
+        bint isna_entry, isna_result
+
+    if not len(values) == len(labels):
+        raise AssertionError("len(index) != len(labels)")
+
+    min_count = max(min_count, 1)
+    nobs = np.zeros((<object>out).shape, dtype=np.int64)
+    nan_val = _get_na_val(<numeric_t>0, is_datetimelike)
+
+    group_min_or_max = np.empty_like(out)
+    group_min_or_max[:] = _get_min_or_max(<numeric_t>0, compute_max, is_datetimelike)
+
+    N, K = (<object>values).shape
+
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            counts[lab] += 1
+            for j in range(K):
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if not skipna:
+                    if uses_mask:
+                        isna_result = result_mask[lab, j]
+                    else:
+                        isna_result = _treat_as_na(
+                            group_min_or_max[lab, j], is_datetimelike
+                        )
+
+                    if isna_result:
+                        # If current min/max is already NA, it will always be NA
+                        continue
+
+                if not isna_entry:
+                    nobs[lab, j] += 1
+                    if compute_max:
+                        if val > group_min_or_max[lab, j]:
+                            group_min_or_max[lab, j] = val
+                    else:
+                        if val < group_min_or_max[lab, j]:
+                            group_min_or_max[lab, j] = val
+                elif not skipna:
+                    if uses_mask:
+                        result_mask[lab, j] = True
+                    else:
+                        group_min_or_max[lab, j] = nan_val
+
+    _check_below_mincount(
+        out, uses_mask, result_mask, ngroups, K, nobs, min_count, group_min_or_max
+    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_idxmin_idxmax(
+    intp_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[numeric_object_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    const uint8_t[:, ::1] mask=None,
+    str name="idxmin",
+    bint skipna=True,
+    uint8_t[:, ::1] result_mask=None,
+):
+    """
+    Compute index of minimum/maximum of columns of `values`, in row groups `labels`.
+
+    This function only computes the row number where the minimum/maximum occurs, we'll
+    take the corresponding index value after this function.
+
+    Parameters
+    ----------
+    out : np.ndarray[intp, ndim=2]
+        Array to store result in.
+    counts : np.ndarray[int64]
+        Input as a zeroed array, populated by group sizes during algorithm
+    values : np.ndarray[numeric_object_t, ndim=2]
+        Values to find column-wise min/max of.
+    labels : np.ndarray[np.intp]
+        Labels to group by.
+    min_count : Py_ssize_t, default -1
+        The minimum number of non-NA group elements, NA result if threshold
+        is not met.
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    name : {"idxmin", "idxmax"}, default "idxmin"
+        Whether to compute idxmin or idxmax.
+    mask : ndarray[bool, ndim=2], optional
+        If not None, indices represent missing values,
+        otherwise the mask will not be used
+    skipna : bool, default True
+        Flag to ignore nan values during truth testing
+    result_mask : ndarray[bool, ndim=2], optional
+        If not None, these specify locations in the output that are NA.
+        Modified in-place.
+
+    Notes
+    -----
+    This method modifies the `out` parameter, rather than returning an object.
+    `counts` is modified to hold group sizes
+    """
+    cdef:
+        Py_ssize_t i, j, N, K, lab
+        numeric_object_t val
+        numeric_object_t[:, ::1] group_min_or_max
+        uint8_t[:, ::1] seen
+        bint uses_mask = mask is not None
+        bint isna_entry
+        bint compute_max = name == "idxmax"
+
+    assert name == "idxmin" or name == "idxmax"
+
+    if not len(values) == len(labels):
+        raise AssertionError("len(index) != len(labels)")
+
+    N, K = (<object>values).shape
+
+    if numeric_object_t is object:
+        group_min_or_max = np.empty((<object>out).shape, dtype=object)
+        seen = np.zeros((<object>out).shape, dtype=np.uint8)
+    else:
+        group_min_or_max = np.empty_like(out, dtype=values.dtype)
+        seen = np.zeros_like(out, dtype=np.uint8)
+
+    # When using transform, we need a valid value for take in the case
+    # a category is not observed; these values will be dropped
+    out[:] = 0
+
+    with nogil(numeric_object_t is not object):
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+
+            for j in range(K):
+                if not skipna and out[lab, j] == -1:
+                    # Once we've hit NA there is no going back
+                    continue
+
+                val = values[i, j]
+
+                if uses_mask:
+                    isna_entry = mask[i, j]
+                else:
+                    isna_entry = _treat_as_na(val, is_datetimelike)
+
+                if isna_entry:
+                    if not skipna or not seen[lab, j]:
+                        out[lab, j] = -1
+                else:
+                    if not seen[lab, j]:
+                        seen[lab, j] = True
+                        group_min_or_max[lab, j] = val
+                        out[lab, j] = i
+                    elif compute_max:
+                        if val > group_min_or_max[lab, j]:
+                            group_min_or_max[lab, j] = val
+                            out[lab, j] = i
+                    else:
+                        if val < group_min_or_max[lab, j]:
+                            group_min_or_max[lab, j] = val
+                            out[lab, j] = i
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_max(
+    numeric_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[numeric_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    """See group_min_max.__doc__"""
+    group_min_max(
+        out,
+        counts,
+        values,
+        labels,
+        min_count=min_count,
+        is_datetimelike=is_datetimelike,
+        compute_max=True,
+        mask=mask,
+        result_mask=result_mask,
+        skipna=skipna,
+    )
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def group_min(
+    numeric_t[:, ::1] out,
+    int64_t[::1] counts,
+    ndarray[numeric_t, ndim=2] values,
+    const intp_t[::1] labels,
+    Py_ssize_t min_count=-1,
+    bint is_datetimelike=False,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    """See group_min_max.__doc__"""
+    group_min_max(
+        out,
+        counts,
+        values,
+        labels,
+        min_count=min_count,
+        is_datetimelike=is_datetimelike,
+        compute_max=False,
+        mask=mask,
+        result_mask=result_mask,
+        skipna=skipna,
+    )
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cdef group_cummin_max(
+    numeric_t[:, ::1] out,
+    ndarray[numeric_t, ndim=2] values,
+    const uint8_t[:, ::1] mask,
+    uint8_t[:, ::1] result_mask,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    bint skipna,
+    bint compute_max,
+):
+    """
+    Cumulative minimum/maximum of columns of `values`, in row groups `labels`.
+
+    Parameters
+    ----------
+    out : np.ndarray[numeric_t, ndim=2]
+        Array to store cummin/max in.
+    values : np.ndarray[numeric_t, ndim=2]
+        Values to take cummin/max of.
+    mask : np.ndarray[bool] or None
+        If not None, indices represent missing values,
+        otherwise the mask will not be used
+    result_mask : ndarray[bool, ndim=2], optional
+        If not None, these specify locations in the output that are NA.
+        Modified in-place.
+    labels : np.ndarray[np.intp]
+        Labels to group by.
+    ngroups : int
+        Number of groups, larger than all entries of `labels`.
+    is_datetimelike : bool
+        True if `values` contains datetime-like entries.
+    skipna : bool
+        If True, ignore nans in `values`.
+    compute_max : bool
+        True if cumulative maximum should be computed, False
+        if cumulative minimum should be computed
+
+    Notes
+    -----
+    This method modifies the `out` parameter, rather than returning an object.
+    """
+    cdef:
+        numeric_t[:, ::1] accum
+        Py_ssize_t i, j, N, K
+        numeric_t val, mval, na_val
+        uint8_t[:, ::1] seen_na
+        intp_t lab
+        bint na_possible
+        bint uses_mask = mask is not None
+        bint isna_entry
+
+    accum = np.empty((ngroups, (<object>values).shape[1]), dtype=values.dtype)
+    accum[:] = _get_min_or_max(<numeric_t>0, compute_max, is_datetimelike)
+
+    na_val = _get_na_val(<numeric_t>0, is_datetimelike)
+
+    if uses_mask:
+        na_possible = True
+        # Will never be used, just to avoid uninitialized warning
+        na_val = 0
+    elif numeric_t is float64_t or numeric_t is float32_t:
+        na_possible = True
+    elif is_datetimelike:
+        na_possible = True
+    else:
+        # Will never be used, just to avoid uninitialized warning
+        na_possible = False
+
+    if na_possible:
+        seen_na = np.zeros((<object>accum).shape, dtype=np.uint8)
+
+    N, K = (<object>values).shape
+    with nogil:
+        for i in range(N):
+            lab = labels[i]
+            if lab < 0:
+                continue
+            for j in range(K):
+
+                if not skipna and na_possible and seen_na[lab, j]:
+                    if uses_mask:
+                        result_mask[i, j] = 1
+                        # Set to 0 ensures that we are deterministic and can
+                        #  downcast if appropriate
+                        out[i, j] = 0
+
+                    else:
+                        out[i, j] = na_val
+                else:
+                    val = values[i, j]
+
+                    if uses_mask:
+                        isna_entry = mask[i, j]
+                    else:
+                        isna_entry = _treat_as_na(val, is_datetimelike)
+
+                    if not isna_entry:
+                        mval = accum[lab, j]
+                        if compute_max:
+                            if val > mval:
+                                accum[lab, j] = mval = val
+                        else:
+                            if val < mval:
+                                accum[lab, j] = mval = val
+                        out[i, j] = mval
+                    else:
+                        seen_na[lab, j] = 1
+                        out[i, j] = val
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_cummin(
+    numeric_t[:, ::1] out,
+    ndarray[numeric_t, ndim=2] values,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    """See group_cummin_max.__doc__"""
+    group_cummin_max(
+        out=out,
+        values=values,
+        mask=mask,
+        result_mask=result_mask,
+        labels=labels,
+        ngroups=ngroups,
+        is_datetimelike=is_datetimelike,
+        skipna=skipna,
+        compute_max=False,
+    )
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def group_cummax(
+    numeric_t[:, ::1] out,
+    ndarray[numeric_t, ndim=2] values,
+    const intp_t[::1] labels,
+    int ngroups,
+    bint is_datetimelike,
+    const uint8_t[:, ::1] mask=None,
+    uint8_t[:, ::1] result_mask=None,
+    bint skipna=True,
+) -> None:
+    """See group_cummin_max.__doc__"""
+    group_cummin_max(
+        out=out,
+        values=values,
+        mask=mask,
+        result_mask=result_mask,
+        labels=labels,
+        ngroups=ngroups,
+        is_datetimelike=is_datetimelike,
+        skipna=skipna,
+        compute_max=True,
+    )
diff --git a/pandas/pandas/_libs/hashing.pyi b/pandas/pandas/_libs/hashing.pyi
new file mode 100644
index 0000000000000..8361026e4a87d
--- /dev/null
+++ b/pandas/pandas/_libs/hashing.pyi
@@ -0,0 +1,9 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def hash_object_array(
+    arr: npt.NDArray[np.object_],
+    key: str,
+    encoding: str = ...,
+) -> npt.NDArray[np.uint64]: ...
diff --git a/pandas/pandas/_libs/hashing.pyx b/pandas/pandas/_libs/hashing.pyx
new file mode 100644
index 0000000000000..a1fd70529efa7
--- /dev/null
+++ b/pandas/pandas/_libs/hashing.pyx
@@ -0,0 +1,198 @@
+# Translated from the reference implementation
+# at https://github.com/veorq/SipHash
+
+cimport cython
+from libc.stdlib cimport (
+    free,
+    malloc,
+)
+
+import numpy as np
+
+from numpy cimport (
+    import_array,
+    ndarray,
+    uint8_t,
+    uint64_t,
+)
+
+import_array()
+
+from pandas._libs.util cimport is_nan
+
+
+@cython.boundscheck(False)
+def hash_object_array(
+    ndarray[object, ndim=1] arr, str key, str encoding="utf8"
+) -> np.ndarray[np.uint64]:
+    """
+    Parameters
+    ----------
+    arr : 1-d object ndarray of objects
+    key : hash key, must be 16 byte len encoded
+    encoding : encoding for key & arr, default to 'utf8'
+
+    Returns
+    -------
+    1-d uint64 ndarray of hashes.
+
+    Raises
+    ------
+    TypeError
+        If the array contains mixed types.
+
+    Notes
+    -----
+    Allowed values must be strings, or nulls
+    mixed array types will raise TypeError.
+    """
+    cdef:
+        Py_ssize_t i, n
+        uint64_t[::1] result
+        bytes data, k
+        uint8_t *kb
+        uint64_t *lens
+        char **vecs
+        char *cdata
+        object val
+        list data_list = []
+
+    k = <bytes>key.encode(encoding)
+    kb = <uint8_t *>k
+    if len(k) != 16:
+        raise ValueError(
+            f"key should be a 16-byte string encoded, got {k} (len {len(k)})"
+        )
+
+    n = len(arr)
+
+    # create an array of bytes
+    vecs = <char **>malloc(n * sizeof(char *))
+    if vecs is NULL:
+        raise MemoryError()
+    lens = <uint64_t*>malloc(n * sizeof(uint64_t))
+    if lens is NULL:
+        raise MemoryError()
+
+    for i in range(n):
+        val = arr[i]
+        if isinstance(val, bytes):
+            data = <bytes>val
+        elif isinstance(val, str):
+            data = <bytes>val.encode(encoding)
+        elif val is None or is_nan(val):
+            # null, stringify and encode
+            data = <bytes>str(val).encode(encoding)
+
+        elif isinstance(val, tuple):
+            # GH#28969 we could have a tuple, but need to ensure that
+            #  the tuple entries are themselves hashable before converting
+            #  to str
+            hash(val)
+            data = <bytes>str(val).encode(encoding)
+        else:
+            raise TypeError(
+                f"{val} of type {type(val)} is not a valid type for hashing, "
+                "must be string or null"
+            )
+
+        lens[i] = len(data)
+        cdata = data
+
+        # keep the references alive through the end of the
+        # function
+        data_list.append(data)
+        vecs[i] = cdata
+
+    result = np.empty(n, dtype=np.uint64)
+    with nogil:
+        for i in range(n):
+            result[i] = low_level_siphash(<uint8_t *>vecs[i], lens[i], kb)
+
+    free(vecs)
+    free(lens)
+    return result.base  # .base to retrieve underlying np.ndarray
+
+
+cdef uint64_t _rotl(uint64_t x, uint64_t b) noexcept nogil:
+    return (x << b) | (x >> (64 - b))
+
+
+cdef uint64_t u8to64_le(uint8_t* p) noexcept nogil:
+    return (<uint64_t>p[0] |
+            <uint64_t>p[1] << 8 |
+            <uint64_t>p[2] << 16 |
+            <uint64_t>p[3] << 24 |
+            <uint64_t>p[4] << 32 |
+            <uint64_t>p[5] << 40 |
+            <uint64_t>p[6] << 48 |
+            <uint64_t>p[7] << 56)
+
+
+cdef void _sipround(uint64_t* v0, uint64_t* v1,
+                    uint64_t* v2, uint64_t* v3) noexcept nogil:
+    v0[0] += v1[0]
+    v1[0] = _rotl(v1[0], 13)
+    v1[0] ^= v0[0]
+    v0[0] = _rotl(v0[0], 32)
+    v2[0] += v3[0]
+    v3[0] = _rotl(v3[0], 16)
+    v3[0] ^= v2[0]
+    v0[0] += v3[0]
+    v3[0] = _rotl(v3[0], 21)
+    v3[0] ^= v0[0]
+    v2[0] += v1[0]
+    v1[0] = _rotl(v1[0], 17)
+    v1[0] ^= v2[0]
+    v2[0] = _rotl(v2[0], 32)
+
+
+@cython.cdivision(True)
+cdef uint64_t low_level_siphash(uint8_t* data, size_t datalen,
+                                uint8_t* key) noexcept nogil:
+    cdef uint64_t v0 = 0x736f6d6570736575ULL
+    cdef uint64_t v1 = 0x646f72616e646f6dULL
+    cdef uint64_t v2 = 0x6c7967656e657261ULL
+    cdef uint64_t v3 = 0x7465646279746573ULL
+    cdef uint64_t b
+    cdef uint64_t k0 = u8to64_le(key)
+    cdef uint64_t k1 = u8to64_le(key + 8)
+    cdef uint64_t m
+    cdef int i
+    cdef uint8_t* end = data + datalen - (datalen % sizeof(uint64_t))
+    cdef int left = datalen & 7
+    cdef int cROUNDS = 2
+    cdef int dROUNDS = 4
+
+    b = (<uint64_t>datalen) << 56
+    v3 ^= k1
+    v2 ^= k0
+    v1 ^= k1
+    v0 ^= k0
+
+    while (data != end):
+        m = u8to64_le(data)
+        v3 ^= m
+        for i in range(cROUNDS):
+            _sipround(&v0, &v1, &v2, &v3)
+        v0 ^= m
+
+        data += sizeof(uint64_t)
+
+    for i in range(left-1, -1, -1):
+        b |= (<uint64_t>data[i]) << (i * 8)
+
+    v3 ^= b
+
+    for i in range(cROUNDS):
+        _sipround(&v0, &v1, &v2, &v3)
+
+    v0 ^= b
+    v2 ^= 0xff
+
+    for i in range(dROUNDS):
+        _sipround(&v0, &v1, &v2, &v3)
+
+    b = v0 ^ v1 ^ v2 ^ v3
+
+    return b
diff --git a/pandas/pandas/_libs/hashtable.pxd b/pandas/pandas/_libs/hashtable.pxd
new file mode 100644
index 0000000000000..a5a3edad63403
--- /dev/null
+++ b/pandas/pandas/_libs/hashtable.pxd
@@ -0,0 +1,189 @@
+from numpy cimport (
+    intp_t,
+    ndarray,
+)
+
+from pandas._libs.khash cimport (
+    complex64_t,
+    complex128_t,
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    kh_complex64_t,
+    kh_complex128_t,
+    kh_float32_t,
+    kh_float64_t,
+    kh_int8_t,
+    kh_int16_t,
+    kh_int32_t,
+    kh_int64_t,
+    kh_pymap_t,
+    kh_str_t,
+    kh_uint8_t,
+    kh_uint16_t,
+    kh_uint32_t,
+    kh_uint64_t,
+    khcomplex64_t,
+    khcomplex128_t,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+
+# prototypes for sharing
+
+cdef class HashTable:
+    pass
+
+cdef class UInt64HashTable(HashTable):
+    cdef kh_uint64_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, uint64_t val)
+    cpdef set_item(self, uint64_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Int64HashTable(HashTable):
+    cdef kh_int64_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, int64_t val)
+    cpdef set_item(self, int64_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class UInt32HashTable(HashTable):
+    cdef kh_uint32_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, uint32_t val)
+    cpdef set_item(self, uint32_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Int32HashTable(HashTable):
+    cdef kh_int32_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, int32_t val)
+    cpdef set_item(self, int32_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class UInt16HashTable(HashTable):
+    cdef kh_uint16_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, uint16_t val)
+    cpdef set_item(self, uint16_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Int16HashTable(HashTable):
+    cdef kh_int16_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, int16_t val)
+    cpdef set_item(self, int16_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class UInt8HashTable(HashTable):
+    cdef kh_uint8_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, uint8_t val)
+    cpdef set_item(self, uint8_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Int8HashTable(HashTable):
+    cdef kh_int8_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, int8_t val)
+    cpdef set_item(self, int8_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Float64HashTable(HashTable):
+    cdef kh_float64_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, float64_t val)
+    cpdef set_item(self, float64_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Float32HashTable(HashTable):
+    cdef kh_float32_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, float32_t val)
+    cpdef set_item(self, float32_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Complex64HashTable(HashTable):
+    cdef kh_complex64_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, complex64_t val)
+    cpdef set_item(self, complex64_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class Complex128HashTable(HashTable):
+    cdef kh_complex128_t *table
+    cdef int64_t na_position
+    cdef bint uses_mask
+
+    cpdef get_item(self, complex128_t val)
+    cpdef set_item(self, complex128_t key, Py_ssize_t val)
+    cpdef get_na(self)
+    cpdef set_na(self, Py_ssize_t val)
+
+cdef class PyObjectHashTable(HashTable):
+    cdef kh_pymap_t *table
+
+    cpdef get_item(self, object val)
+    cpdef set_item(self, object key, Py_ssize_t val)
+
+
+cdef class StringHashTable(HashTable):
+    cdef kh_str_t *table
+
+    cpdef get_item(self, str val)
+    cpdef set_item(self, str key, Py_ssize_t val)
+
+cdef struct Int64VectorData:
+    int64_t *data
+    Py_ssize_t size, capacity
+
+cdef class Vector:
+    cdef bint external_view_exists
+
+cdef class Int64Vector(Vector):
+    cdef Int64VectorData data
+    cdef ndarray ao
+
+    cdef resize(self, Py_ssize_t new_size)
+    cpdef ndarray to_array(self)
+    cdef void append(self, int64_t x) noexcept
+    cdef extend(self, int64_t[:] x)
diff --git a/pandas/pandas/_libs/hashtable.pyi b/pandas/pandas/_libs/hashtable.pyi
new file mode 100644
index 0000000000000..7a810a988e50e
--- /dev/null
+++ b/pandas/pandas/_libs/hashtable.pyi
@@ -0,0 +1,274 @@
+from typing import (
+    Any,
+    Hashable,
+    Literal,
+    overload,
+)
+
+import numpy as np
+
+from pandas._typing import npt
+
+def unique_label_indices(
+    labels: np.ndarray,  # const int64_t[:]
+) -> np.ndarray: ...
+
+class Factorizer:
+    count: int
+    uniques: Any
+    def __init__(self, size_hint: int, uses_mask: bool = False) -> None: ...
+    def get_count(self) -> int: ...
+    def factorize(
+        self,
+        values: np.ndarray,
+        na_sentinel=...,
+        na_value=...,
+        mask=...,
+    ) -> npt.NDArray[np.intp]: ...
+    def hash_inner_join(
+        self, values: np.ndarray, mask=...
+    ) -> tuple[np.ndarray, np.ndarray]: ...
+
+class ObjectFactorizer(Factorizer):
+    table: PyObjectHashTable
+    uniques: ObjectVector
+
+class Int64Factorizer(Factorizer):
+    table: Int64HashTable
+    uniques: Int64Vector
+
+class UInt64Factorizer(Factorizer):
+    table: UInt64HashTable
+    uniques: UInt64Vector
+
+class Int32Factorizer(Factorizer):
+    table: Int32HashTable
+    uniques: Int32Vector
+
+class UInt32Factorizer(Factorizer):
+    table: UInt32HashTable
+    uniques: UInt32Vector
+
+class Int16Factorizer(Factorizer):
+    table: Int16HashTable
+    uniques: Int16Vector
+
+class UInt16Factorizer(Factorizer):
+    table: UInt16HashTable
+    uniques: UInt16Vector
+
+class Int8Factorizer(Factorizer):
+    table: Int8HashTable
+    uniques: Int8Vector
+
+class UInt8Factorizer(Factorizer):
+    table: UInt8HashTable
+    uniques: UInt8Vector
+
+class Float64Factorizer(Factorizer):
+    table: Float64HashTable
+    uniques: Float64Vector
+
+class Float32Factorizer(Factorizer):
+    table: Float32HashTable
+    uniques: Float32Vector
+
+class Complex64Factorizer(Factorizer):
+    table: Complex64HashTable
+    uniques: Complex64Vector
+
+class Complex128Factorizer(Factorizer):
+    table: Complex128HashTable
+    uniques: Complex128Vector
+
+class Int64Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.int64]: ...
+
+class Int32Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.int32]: ...
+
+class Int16Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.int16]: ...
+
+class Int8Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.int8]: ...
+
+class UInt64Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.uint64]: ...
+
+class UInt32Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.uint32]: ...
+
+class UInt16Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.uint16]: ...
+
+class UInt8Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.uint8]: ...
+
+class Float64Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.float64]: ...
+
+class Float32Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.float32]: ...
+
+class Complex128Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.complex128]: ...
+
+class Complex64Vector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.complex64]: ...
+
+class StringVector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.object_]: ...
+
+class ObjectVector:
+    def __init__(self, *args) -> None: ...
+    def __len__(self) -> int: ...
+    def to_array(self) -> npt.NDArray[np.object_]: ...
+
+class HashTable:
+    # NB: The base HashTable class does _not_ actually have these methods;
+    #  we are putting them here for the sake of mypy to avoid
+    #  reproducing them in each subclass below.
+    def __init__(self, size_hint: int = ..., uses_mask: bool = ...) -> None: ...
+    def __len__(self) -> int: ...
+    def __contains__(self, key: Hashable) -> bool: ...
+    def sizeof(self, deep: bool = ...) -> int: ...
+    def get_state(self) -> dict[str, int]: ...
+    # TODO: `val/key` type is subclass-specific
+    def get_item(self, val): ...  # TODO: return type?
+    def set_item(self, key, val) -> None: ...
+    def get_na(self): ...  # TODO: return type?
+    def set_na(self, val) -> None: ...
+    def map_locations(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        mask: npt.NDArray[np.bool_] | None = ...,
+    ) -> None: ...
+    def lookup(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        mask: npt.NDArray[np.bool_] | None = ...,
+    ) -> npt.NDArray[np.intp]: ...
+    def get_labels(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        uniques,  # SubclassTypeVector
+        count_prior: int = ...,
+        na_sentinel: int = ...,
+        na_value: object = ...,
+        mask=...,
+    ) -> npt.NDArray[np.intp]: ...
+    @overload
+    def unique(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        *,
+        return_inverse: Literal[False] = ...,
+        mask: None = ...,
+    ) -> np.ndarray: ...  # np.ndarray[subclass-specific]
+    @overload
+    def unique(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        *,
+        return_inverse: Literal[True],
+        mask: None = ...,
+    ) -> tuple[np.ndarray, npt.NDArray[np.intp]]: ...  # np.ndarray[subclass-specific]
+    @overload
+    def unique(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        *,
+        return_inverse: Literal[False] = ...,
+        mask: npt.NDArray[np.bool_],
+    ) -> tuple[
+        np.ndarray,
+        npt.NDArray[np.bool_],
+    ]: ...  # np.ndarray[subclass-specific]
+    def factorize(
+        self,
+        values: np.ndarray,  # np.ndarray[subclass-specific]
+        na_sentinel: int = ...,
+        na_value: object = ...,
+        mask=...,
+        ignore_na: bool = True,
+    ) -> tuple[np.ndarray, npt.NDArray[np.intp]]: ...  # np.ndarray[subclass-specific]
+    def hash_inner_join(
+        self, values: np.ndarray, mask=...
+    ) -> tuple[np.ndarray, np.ndarray]: ...
+
+class Complex128HashTable(HashTable): ...
+class Complex64HashTable(HashTable): ...
+class Float64HashTable(HashTable): ...
+class Float32HashTable(HashTable): ...
+
+class Int64HashTable(HashTable):
+    # Only Int64HashTable has get_labels_groupby, map_keys_to_values
+    def get_labels_groupby(
+        self,
+        values: npt.NDArray[np.int64],  # const int64_t[:]
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.int64]]: ...
+    def map_keys_to_values(
+        self,
+        keys: npt.NDArray[np.int64],
+        values: npt.NDArray[np.int64],  # const int64_t[:]
+    ) -> None: ...
+
+class Int32HashTable(HashTable): ...
+class Int16HashTable(HashTable): ...
+class Int8HashTable(HashTable): ...
+class UInt64HashTable(HashTable): ...
+class UInt32HashTable(HashTable): ...
+class UInt16HashTable(HashTable): ...
+class UInt8HashTable(HashTable): ...
+class StringHashTable(HashTable): ...
+class PyObjectHashTable(HashTable): ...
+class IntpHashTable(HashTable): ...
+
+def duplicated(
+    values: np.ndarray,
+    keep: Literal["last", "first", False] = ...,
+    mask: npt.NDArray[np.bool_] | None = ...,
+) -> npt.NDArray[np.bool_]: ...
+def mode(
+    values: np.ndarray, dropna: bool, mask: npt.NDArray[np.bool_] | None = ...
+) -> np.ndarray: ...
+def value_count(
+    values: np.ndarray,
+    dropna: bool,
+    mask: npt.NDArray[np.bool_] | None = ...,
+) -> tuple[np.ndarray, npt.NDArray[np.int64], int]: ...  # np.ndarray[same-as-values]
+
+# arr and values should have same dtype
+def ismember(
+    arr: np.ndarray,
+    values: np.ndarray,
+) -> npt.NDArray[np.bool_]: ...
+def object_hash(obj) -> int: ...
+def objects_are_equal(a, b) -> bool: ...
diff --git a/pandas/pandas/_libs/hashtable.pyx b/pandas/pandas/_libs/hashtable.pyx
new file mode 100644
index 0000000000000..b5ae5a3440f39
--- /dev/null
+++ b/pandas/pandas/_libs/hashtable.pyx
@@ -0,0 +1,128 @@
+cimport cython
+from cpython.ref cimport (
+    Py_INCREF,
+    PyObject,
+)
+from libc.stdlib cimport (
+    free,
+    malloc,
+)
+from libc.string cimport memcpy
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport ndarray
+
+cnp.import_array()
+
+
+from pandas._libs cimport util
+from pandas._libs.dtypes cimport numeric_object_t
+from pandas._libs.khash cimport (
+    KHASH_TRACE_DOMAIN,
+    are_equivalent_float32_t,
+    are_equivalent_float64_t,
+    are_equivalent_khcomplex64_t,
+    are_equivalent_khcomplex128_t,
+    kh_needed_n_buckets,
+    kh_python_hash_equal,
+    kh_python_hash_func,
+    khiter_t,
+)
+from pandas._libs.missing cimport (
+    checknull,
+    is_matching_na,
+)
+
+
+def get_hashtable_trace_domain():
+    return KHASH_TRACE_DOMAIN
+
+
+def object_hash(obj):
+    return kh_python_hash_func(obj)
+
+
+def objects_are_equal(a, b):
+    return kh_python_hash_equal(a, b)
+
+
+cdef int64_t NPY_NAT = util.get_nat()
+SIZE_HINT_LIMIT = (1 << 20) + 7
+
+
+cdef Py_ssize_t _INIT_VEC_CAP = 128
+
+include "hashtable_class_helper.pxi"
+include "hashtable_func_helper.pxi"
+
+
+# map derived hash-map types onto basic hash-map types:
+if np.dtype(np.intp) == np.dtype(np.int64):
+    IntpHashTable = Int64HashTable
+    unique_label_indices = _unique_label_indices_int64
+elif np.dtype(np.intp) == np.dtype(np.int32):
+    IntpHashTable = Int32HashTable
+    unique_label_indices = _unique_label_indices_int32
+else:
+    raise ValueError(np.dtype(np.intp))
+
+
+cdef class Factorizer:
+    cdef readonly:
+        Py_ssize_t count
+
+    def __cinit__(self, size_hint: int, uses_mask: bool = False):
+        self.count = 0
+
+    def get_count(self) -> int:
+        return self.count
+
+    def factorize(self, values, na_sentinel=-1, na_value=None, mask=None) -> np.ndarray:
+        raise NotImplementedError
+
+    def hash_inner_join(self, values, mask=None):
+        raise NotImplementedError
+
+
+cdef class ObjectFactorizer(Factorizer):
+    cdef public:
+        PyObjectHashTable table
+        ObjectVector uniques
+
+    def __cinit__(self, size_hint: int, uses_mask: bool = False):
+        self.table = PyObjectHashTable(size_hint)
+        self.uniques = ObjectVector()
+
+    def factorize(
+        self, ndarray[object] values, na_sentinel=-1, na_value=None, mask=None
+    ) -> np.ndarray:
+        """
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+
+        Examples
+        --------
+        Factorize values with nans replaced by na_sentinel
+
+        >>> fac = ObjectFactorizer(3)
+        >>> fac.factorize(np.array([1,2,np.nan], dtype='O'), na_sentinel=20)
+        array([ 0,  1, 20])
+        """
+        cdef:
+            ndarray[intp_t] labels
+
+        if mask is not None:
+            raise NotImplementedError("mask not supported for ObjectFactorizer.")
+
+        if self.uniques.external_view_exists:
+            uniques = ObjectVector()
+            uniques.extend(self.uniques.to_array())
+            self.uniques = uniques
+        labels = self.table.get_labels(values, self.uniques,
+                                       self.count, na_sentinel, na_value)
+        self.count = len(self.uniques)
+        return labels
diff --git a/pandas/pandas/_libs/hashtable_class_helper.pxi.in b/pandas/pandas/_libs/hashtable_class_helper.pxi.in
new file mode 100644
index 0000000000000..210df09f07db6
--- /dev/null
+++ b/pandas/pandas/_libs/hashtable_class_helper.pxi.in
@@ -0,0 +1,1572 @@
+"""
+Template for each `dtype` helper function for hashtable
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+from cpython.unicode cimport PyUnicode_AsUTF8
+
+{{py:
+
+# name
+complex_types = ['complex64',
+                 'complex128']
+}}
+
+{{for name in complex_types}}
+cdef kh{{name}}_t to_kh{{name}}_t({{name}}_t val) noexcept nogil:
+    cdef kh{{name}}_t res
+    res.real = val.real
+    res.imag = val.imag
+    return res
+
+{{endfor}}
+
+
+{{py:
+
+
+# name
+c_types = ['khcomplex128_t',
+           'khcomplex64_t',
+           'float64_t',
+           'float32_t',
+           'int64_t',
+           'int32_t',
+           'int16_t',
+           'int8_t',
+           'uint64_t',
+           'uint32_t',
+           'uint16_t',
+           'uint8_t']
+}}
+
+{{for c_type in c_types}}
+
+cdef bint is_nan_{{c_type}}({{c_type}} val) noexcept nogil:
+    {{if c_type in {'khcomplex128_t', 'khcomplex64_t'} }}
+    return val.real != val.real or val.imag != val.imag
+    {{elif c_type in {'float64_t', 'float32_t'} }}
+    return val != val
+    {{else}}
+    return False
+    {{endif}}
+
+
+{{if c_type in {'khcomplex128_t', 'khcomplex64_t', 'float64_t', 'float32_t'} }}
+# are_equivalent_{{c_type}} is cimported via khash.pxd
+{{else}}
+cdef bint are_equivalent_{{c_type}}({{c_type}} val1, {{c_type}} val2) noexcept nogil:
+    return val1 == val2
+{{endif}}
+
+{{endfor}}
+
+
+{{py:
+
+# name
+cimported_types = ['complex64',
+                   'complex128',
+                   'float32',
+                   'float64',
+                   'int8',
+                   'int16',
+                   'int32',
+                   'int64',
+                   'pymap',
+                   'str',
+                   'strbox',
+                   'uint8',
+                   'uint16',
+                   'uint32',
+                   'uint64']
+}}
+
+{{for name in cimported_types}}
+from pandas._libs.khash cimport (
+    kh_destroy_{{name}},
+    kh_exist_{{name}},
+    kh_get_{{name}},
+    kh_init_{{name}},
+    kh_put_{{name}},
+    kh_resize_{{name}},
+)
+
+{{endfor}}
+
+# ----------------------------------------------------------------------
+# VectorData
+# ----------------------------------------------------------------------
+
+from pandas._libs.missing cimport C_NA
+
+
+{{py:
+
+# name, dtype, c_type
+# the generated StringVector is not actually used
+# but is included for completeness (rather ObjectVector is used
+# for uniques in hashtables)
+
+dtypes = [('Complex128', 'complex128', 'khcomplex128_t'),
+          ('Complex64', 'complex64', 'khcomplex64_t'),
+          ('Float64', 'float64', 'float64_t'),
+          ('Float32', 'float32', 'float32_t'),
+          ('Int64', 'int64', 'int64_t'),
+          ('Int32', 'int32', 'int32_t'),
+          ('Int16', 'int16', 'int16_t'),
+          ('Int8', 'int8', 'int8_t'),
+          ('String', 'string', 'char *'),
+          ('UInt64', 'uint64', 'uint64_t'),
+          ('UInt32', 'uint32', 'uint32_t'),
+          ('UInt16', 'uint16', 'uint16_t'),
+          ('UInt8', 'uint8', 'uint8_t')]
+}}
+
+{{for name, dtype, c_type in dtypes}}
+
+
+{{if dtype != 'int64'}}
+# Int64VectorData is defined in the .pxd file because it is needed (indirectly)
+#  by IntervalTree
+
+ctypedef struct {{name}}VectorData:
+    {{c_type}} *data
+    Py_ssize_t size, capacity
+
+{{endif}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef void append_data_{{dtype}}({{name}}VectorData *data,
+                                       {{c_type}} x) noexcept nogil:
+
+    data.data[data.size] = x
+    data.size += 1
+
+{{endfor}}
+
+ctypedef fused vector_data:
+    Int64VectorData
+    Int32VectorData
+    Int16VectorData
+    Int8VectorData
+    UInt64VectorData
+    UInt32VectorData
+    UInt16VectorData
+    UInt8VectorData
+    Float64VectorData
+    Float32VectorData
+    Complex128VectorData
+    Complex64VectorData
+    StringVectorData
+
+
+cdef bint needs_resize(Py_ssize_t nelems, Py_ssize_t capacity) noexcept nogil:
+    return nelems >= capacity
+
+# ----------------------------------------------------------------------
+# Vector
+# ----------------------------------------------------------------------
+
+cdef class Vector:
+    # cdef readonly:
+    #    bint external_view_exists
+
+    def __cinit__(self):
+        self.external_view_exists = False
+
+
+{{py:
+
+# name, dtype, c_type
+dtypes = [('Complex128', 'complex128', 'khcomplex128_t'),
+          ('Complex64', 'complex64', 'khcomplex64_t'),
+          ('Float64', 'float64', 'float64_t'),
+          ('UInt64', 'uint64', 'uint64_t'),
+          ('Int64', 'int64', 'int64_t'),
+          ('Float32', 'float32', 'float32_t'),
+          ('UInt32', 'uint32', 'uint32_t'),
+          ('Int32', 'int32', 'int32_t'),
+          ('UInt16', 'uint16', 'uint16_t'),
+          ('Int16', 'int16', 'int16_t'),
+          ('UInt8', 'uint8', 'uint8_t'),
+          ('Int8', 'int8', 'int8_t')]
+
+}}
+
+{{for name, dtype, c_type in dtypes}}
+
+cdef class {{name}}Vector(Vector):
+
+    # For int64 we have to put this declaration in the .pxd file;
+    # Int64Vector is the only one we need exposed for other cython files.
+    {{if dtype != 'int64'}}
+    cdef:
+        {{name}}VectorData data
+        ndarray ao
+    {{endif}}
+
+    def __cinit__(self):
+        self.data.size = 0
+        self.data.capacity = _INIT_VEC_CAP
+        self.ao = np.empty(self.data.capacity, dtype=np.{{dtype}})
+        self.data.data = <{{c_type}}*>self.ao.data
+
+    cdef resize(self, Py_ssize_t new_size):
+        self.data.capacity = max(new_size, _INIT_VEC_CAP)
+        self.ao.resize(self.data.capacity, refcheck=False)
+        self.data.data = <{{c_type}}*>self.ao.data
+
+    def __len__(self) -> int:
+        return self.data.size
+
+    cpdef ndarray to_array(self):
+        if self.data.capacity != self.data.size:
+            if self.external_view_exists:
+                # should never happen
+                raise ValueError("should have raised on append()")
+            self.ao.resize(self.data.size, refcheck=False)
+            self.data.capacity = self.data.size
+        self.external_view_exists = True
+        return self.ao
+
+    cdef void append(self, {{c_type}} x) noexcept:
+
+        if needs_resize(self.data.size, self.data.capacity):
+            if self.external_view_exists:
+                raise ValueError("external reference but "
+                                 "Vector.resize() needed")
+            self.resize(self.data.capacity * 4)
+
+        append_data_{{dtype}}(&self.data, x)
+
+    cdef extend(self, const {{c_type}}[:] x):
+        cdef Py_ssize_t x_size = len(x)
+        if x_size == 0:
+            return
+
+        cdef Py_ssize_t needed_size = self.data.size + x_size
+        if needs_resize(needed_size, self.data.capacity):
+            if self.external_view_exists:
+                raise ValueError("external reference but "
+                                 "Vector.resize() needed")
+            self.resize(needed_size)
+
+        memcpy(self.data.data + self.data.size, &x[0], x_size * sizeof({{c_type}}))
+        self.data.size = needed_size
+
+{{endfor}}
+
+cdef class StringVector(Vector):
+
+    cdef:
+        StringVectorData data
+
+    def __cinit__(self):
+        self.data.size = 0
+        self.data.capacity = _INIT_VEC_CAP
+        self.data.data = <char **>malloc(self.data.capacity * sizeof(char *))
+        if self.data.data is NULL:
+            raise MemoryError()
+
+    cdef resize(self, Py_ssize_t new_size):
+        cdef:
+            char **orig_data
+            Py_ssize_t i, orig_capacity
+
+        orig_capacity = self.data.capacity
+        self.data.capacity = max(self.data.capacity * 4, _INIT_VEC_CAP)
+
+        orig_data = self.data.data
+        self.data.data = <char **>malloc(self.data.capacity * sizeof(char *))
+        if self.data.data is NULL:
+            raise MemoryError()
+        for i in range(orig_capacity):
+            self.data.data[i] = orig_data[i]
+
+    def __dealloc__(self):
+        free(self.data.data)
+
+    def __len__(self) -> int:
+        return self.data.size
+
+    cpdef ndarray[object, ndim=1] to_array(self):
+        cdef:
+            ndarray ao
+            Py_ssize_t n
+            object val
+
+        ao = np.empty(self.data.size, dtype=object)
+        for i in range(self.data.size):
+            val = self.data.data[i]
+            ao[i] = val
+        self.external_view_exists = True
+        self.data.capacity = self.data.size
+        return ao
+
+    cdef void append(self, char *x) noexcept:
+
+        if needs_resize(self.data.size, self.data.capacity):
+            self.resize(self.data.capacity * 4)
+
+        append_data_string(&self.data, x)
+
+    cdef extend(self, ndarray[object] x):
+        for i in range(len(x)):
+            self.append(x[i])
+
+
+cdef class ObjectVector(Vector):
+
+    cdef:
+        PyObject **data
+        Py_ssize_t size, capacity
+        ndarray ao
+
+    def __cinit__(self):
+        self.size = 0
+        self.capacity = _INIT_VEC_CAP
+        self.ao = np.empty(_INIT_VEC_CAP, dtype=object)
+        self.data = <PyObject**>self.ao.data
+
+    def __len__(self) -> int:
+        return self.size
+
+    cdef append(self, object obj):
+        if self.size == self.capacity:
+            if self.external_view_exists:
+                raise ValueError("external reference but "
+                                 "Vector.resize() needed")
+            self.capacity = max(self.capacity * 2, _INIT_VEC_CAP)
+            self.ao.resize(self.capacity, refcheck=False)
+            self.data = <PyObject**>self.ao.data
+
+        Py_INCREF(obj)
+        self.data[self.size] = <PyObject*>obj
+        self.size += 1
+
+    cpdef ndarray[object, ndim=1] to_array(self):
+        if self.capacity != self.size:
+            if self.external_view_exists:
+                raise ValueError("should have raised on append()")
+            self.ao.resize(self.size, refcheck=False)
+            self.capacity = self.size
+        self.external_view_exists = True
+        return self.ao
+
+    cdef extend(self, ndarray[object] x):
+        for i in range(len(x)):
+            self.append(x[i])
+
+# ----------------------------------------------------------------------
+# HashTable
+# ----------------------------------------------------------------------
+
+
+cdef class HashTable:
+
+    pass
+
+{{py:
+
+# name, dtype, c_type, to_c_type
+dtypes = [('Complex128', 'complex128', 'khcomplex128_t', 'to_khcomplex128_t'),
+          ('Float64', 'float64', 'float64_t', ''),
+          ('UInt64', 'uint64', 'uint64_t', ''),
+          ('Int64', 'int64', 'int64_t', ''),
+          ('Complex64', 'complex64', 'khcomplex64_t', 'to_khcomplex64_t'),
+          ('Float32', 'float32', 'float32_t', ''),
+          ('UInt32', 'uint32', 'uint32_t', ''),
+          ('Int32', 'int32', 'int32_t', ''),
+          ('UInt16', 'uint16', 'uint16_t', ''),
+          ('Int16', 'int16', 'int16_t', ''),
+          ('UInt8', 'uint8', 'uint8_t', ''),
+          ('Int8', 'int8', 'int8_t', '')]
+
+}}
+
+
+{{for name, dtype, c_type, to_c_type in dtypes}}
+
+cdef class {{name}}HashTable(HashTable):
+
+    def __cinit__(self, int64_t size_hint=1, bint uses_mask=False):
+        self.table = kh_init_{{dtype}}()
+        size_hint = min(kh_needed_n_buckets(size_hint), SIZE_HINT_LIMIT)
+        kh_resize_{{dtype}}(self.table, size_hint)
+
+        self.uses_mask = uses_mask
+        self.na_position = -1
+
+    def __len__(self) -> int:
+        return self.table.size + (0 if self.na_position == -1 else 1)
+
+    def __dealloc__(self):
+        if self.table is not NULL:
+            kh_destroy_{{dtype}}(self.table)
+            self.table = NULL
+
+    def __contains__(self, object key) -> bool:
+        # The caller is responsible to check for compatible NA values in case
+        # of masked arrays.
+        cdef:
+            khiter_t k
+            {{c_type}} ckey
+
+        if self.uses_mask and checknull(key):
+            return -1 != self.na_position
+
+        ckey = {{to_c_type}}(key)
+        k = kh_get_{{dtype}}(self.table, ckey)
+        return k != self.table.n_buckets
+
+    def sizeof(self, deep: bool = False) -> int:
+        """ return the size of my table in bytes """
+        overhead = 4 * sizeof(uint32_t) + 3 * sizeof(uint32_t*)
+        for_flags = max(1, self.table.n_buckets >> 5) * sizeof(uint32_t)
+        for_pairs =  self.table.n_buckets * (sizeof({{dtype}}_t) + # keys
+                                             sizeof(Py_ssize_t))   # vals
+        return overhead + for_flags + for_pairs
+
+    def get_state(self) -> dict[str, int]:
+        """ returns infos about the state of the hashtable"""
+        return {
+            'n_buckets' : self.table.n_buckets,
+            'size' : self.table.size,
+            'n_occupied' : self.table.n_occupied,
+            'upper_bound' : self.table.upper_bound,
+        }
+
+    cpdef get_item(self, {{dtype}}_t val):
+        """Extracts the position of val from the hashtable.
+
+        Parameters
+        ----------
+        val : Scalar
+            The value that is looked up in the hashtable
+
+        Returns
+        -------
+        The position of the requested integer.
+        """
+
+        # Used in core.sorting, IndexEngine.get_loc
+        # Caller is responsible for checking for pd.NA
+        cdef:
+            khiter_t k
+            {{c_type}} cval
+
+        cval = {{to_c_type}}(val)
+        k = kh_get_{{dtype}}(self.table, cval)
+        if k != self.table.n_buckets:
+            return self.table.vals[k]
+        else:
+            raise KeyError(val)
+
+    cpdef get_na(self):
+        """Extracts the position of na_value from the hashtable.
+
+        Returns
+        -------
+        The position of the last na value.
+        """
+
+        if not self.uses_mask:
+            raise NotImplementedError
+
+        if self.na_position == -1:
+            raise KeyError("NA")
+        return self.na_position
+
+    cpdef set_item(self, {{dtype}}_t key, Py_ssize_t val):
+        # Used in libjoin
+        # Caller is responsible for checking for pd.NA
+        cdef:
+            khiter_t k
+            int ret = 0
+            {{c_type}} ckey
+
+        ckey = {{to_c_type}}(key)
+        k = kh_put_{{dtype}}(self.table, ckey, &ret)
+        if kh_exist_{{dtype}}(self.table, k):
+            self.table.vals[k] = val
+        else:
+            raise KeyError(key)
+
+    cpdef set_na(self, Py_ssize_t val):
+        # Caller is responsible for checking for pd.NA
+        cdef:
+            khiter_t k
+            int ret = 0
+            {{c_type}} ckey
+
+        if not self.uses_mask:
+            raise NotImplementedError
+
+        self.na_position = val
+
+    {{if dtype == "int64" }}
+    # We only use this for int64, can reduce build size and make .pyi
+    #  more accurate by only implementing it for int64
+    @cython.boundscheck(False)
+    def map_keys_to_values(
+        self, const {{dtype}}_t[:] keys, const int64_t[:] values
+    ) -> None:
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            {{c_type}} key
+            khiter_t k
+
+        with nogil:
+            for i in range(n):
+                key = {{to_c_type}}(keys[i])
+                k = kh_put_{{dtype}}(self.table, key, &ret)
+                self.table.vals[k] = <Py_ssize_t>values[i]
+    {{endif}}
+
+    @cython.boundscheck(False)
+    def map_locations(self, const {{dtype}}_t[:] values, const uint8_t[:] mask = None) -> None:
+        # Used in libindex, safe_sort
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            {{c_type}} val
+            khiter_t k
+            int8_t na_position = self.na_position
+
+        if self.uses_mask and mask is None:
+            raise NotImplementedError  # pragma: no cover
+
+        with nogil:
+            if self.uses_mask:
+                for i in range(n):
+                    if mask[i]:
+                        na_position = i
+                    else:
+                        val= {{to_c_type}}(values[i])
+                        k = kh_put_{{dtype}}(self.table, val, &ret)
+                        self.table.vals[k] = i
+            else:
+                for i in range(n):
+                    val= {{to_c_type}}(values[i])
+                    k = kh_put_{{dtype}}(self.table, val, &ret)
+                    self.table.vals[k] = i
+        self.na_position = na_position
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    def hash_inner_join(self, const {{dtype}}_t[:] values, const uint8_t[:] mask = None) -> tuple[ndarray, ndarray]:
+        cdef:
+            Py_ssize_t i, n = len(values)
+            {{c_type}} val
+            khiter_t k
+            Int64Vector locs = Int64Vector()
+            Int64Vector self_locs = Int64Vector()
+            Int64VectorData *l
+            Int64VectorData *sl
+            int8_t na_position = self.na_position
+
+        l = &locs.data
+        sl = &self_locs.data
+
+        if self.uses_mask and mask is None:
+            raise NotImplementedError  # pragma: no cover
+
+        with nogil:
+            for i in range(n):
+                if self.uses_mask and mask[i]:
+                    if self.na_position == -1:
+                        continue
+                    if needs_resize(l.size, l.capacity):
+                        with gil:
+                            locs.resize(locs.data.capacity * 4)
+                            self_locs.resize(locs.data.capacity * 4)
+                    append_data_int64(l, i)
+                    append_data_int64(sl, na_position)
+                else:
+                    val = {{to_c_type}}(values[i])
+                    k = kh_get_{{dtype}}(self.table, val)
+                    if k != self.table.n_buckets:
+                        if needs_resize(l.size, l.capacity):
+                            with gil:
+                                locs.resize(locs.data.capacity * 4)
+                                self_locs.resize(locs.data.capacity * 4)
+                        append_data_int64(l, i)
+                        append_data_int64(sl, self.table.vals[k])
+
+        return self_locs.to_array(), locs.to_array()
+
+    @cython.boundscheck(False)
+    def lookup(self, const {{dtype}}_t[:] values, const uint8_t[:] mask = None) -> ndarray:
+        # -> np.ndarray[np.intp]
+        # Used in safe_sort, IndexEngine.get_indexer
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            {{c_type}} val
+            khiter_t k
+            intp_t[::1] locs = np.empty(n, dtype=np.intp)
+            int8_t na_position = self.na_position
+
+        if self.uses_mask and mask is None:
+            raise NotImplementedError  # pragma: no cover
+
+        with nogil:
+            for i in range(n):
+                if self.uses_mask and mask[i]:
+                    locs[i] = na_position
+                else:
+                    val = {{to_c_type}}(values[i])
+                    k = kh_get_{{dtype}}(self.table, val)
+                    if k != self.table.n_buckets:
+                        locs[i] = self.table.vals[k]
+                    else:
+                        locs[i] = -1
+
+        return np.asarray(locs)
+
+    @cython.boundscheck(False)
+    @cython.wraparound(False)
+    def _unique(self, const {{dtype}}_t[:] values, {{name}}Vector uniques,
+                Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                object na_value=None, bint ignore_na=False,
+                object mask=None, bint return_inverse=False, bint use_result_mask=False):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[{{dtype}}]
+            Array of values of which unique will be calculated
+        uniques : {{name}}Vector
+            Vector into which uniques will be written
+        count_prior : Py_ssize_t, default 0
+            Number of existing entries in uniques
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then
+            any value "val" satisfying val != val is considered missing.
+            If na_value is not None, then _additionally_, any value "val"
+            satisfying val == na_value is considered missing.
+        ignore_na : bool, default False
+            Whether NA-values should be ignored for calculating the uniques. If
+            True, the labels corresponding to missing values will be set to
+            na_sentinel.
+        mask : ndarray[bool], optional
+            If not None, the mask is used as indicator for missing values
+            (True = missing, False = valid) instead of `na_value` or
+            condition "val != val".
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+        use_result_mask: bool, default False
+            Whether to create a result mask for the unique values. Not supported
+            with return_inverse=True.
+
+        Returns
+        -------
+        uniques : ndarray[{{dtype}}]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse=True)
+            The labels from values to uniques
+        result_mask: ndarray[bool], if use_result_mask is true
+            The mask for the result values.
+        """
+        cdef:
+            Py_ssize_t i, idx, count = count_prior, n = len(values)
+            intp_t[::1] labels
+            int ret = 0
+            {{c_type}} val, na_value2
+            khiter_t k
+            {{name}}VectorData *ud
+            UInt8Vector result_mask
+            UInt8VectorData *rmd
+            bint use_na_value, use_mask, seen_na = False
+            const uint8_t[:] mask_values
+
+        if return_inverse:
+            labels = np.empty(n, dtype=np.intp)
+        ud = &uniques.data
+        use_na_value = na_value is not None
+        use_mask = mask is not None
+        if not use_mask and use_result_mask:
+            raise NotImplementedError  # pragma: no cover
+
+        if use_result_mask and return_inverse:
+            raise NotImplementedError  # pragma: no cover
+
+        result_mask = UInt8Vector()
+        rmd = &result_mask.data
+
+        if use_mask:
+            mask_values = mask.view("uint8")
+
+        if use_na_value:
+            # We need this na_value2 because we want to allow users
+            # to *optionally* specify an NA sentinel *of the correct* type.
+            # We use None, to make it optional, which requires `object` type
+            # for the parameter. To please the compiler, we use na_value2,
+            # which is only used if it's *specified*.
+            na_value2 = {{to_c_type}}(na_value)
+        else:
+            na_value2 = {{to_c_type}}(0)
+
+        with nogil:
+            for i in range(n):
+                val = {{to_c_type}}(values[i])
+
+                if ignore_na and use_mask:
+                    if mask_values[i]:
+                        labels[i] = na_sentinel
+                        continue
+                elif ignore_na and (
+                   is_nan_{{c_type}}(val) or
+                   (use_na_value and are_equivalent_{{c_type}}(val, na_value2))
+                ):
+                    # if missing values do not count as unique values (i.e. if
+                    # ignore_na is True), skip the hashtable entry for them,
+                    # and replace the corresponding label with na_sentinel
+                    labels[i] = na_sentinel
+                    continue
+                elif not ignore_na and use_result_mask:
+                    if mask_values[i]:
+                        if seen_na:
+                            continue
+
+                        seen_na = True
+                        if needs_resize(ud.size, ud.capacity):
+                            with gil:
+                                if uniques.external_view_exists:
+                                    raise ValueError("external reference to "
+                                                     "uniques held, but "
+                                                     "Vector.resize() needed")
+                                uniques.resize(uniques.data.capacity * 4)
+                                if result_mask.external_view_exists:
+                                    raise ValueError("external reference to "
+                                                     "result_mask held, but "
+                                                     "Vector.resize() needed")
+                                result_mask.resize(result_mask.data.capacity * 4)
+                        append_data_{{dtype}}(ud, val)
+                        append_data_uint8(rmd, 1)
+                        continue
+
+                k = kh_get_{{dtype}}(self.table, val)
+
+                if k == self.table.n_buckets:
+                    # k hasn't been seen yet
+                    k = kh_put_{{dtype}}(self.table, val, &ret)
+
+                    if needs_resize(ud.size, ud.capacity):
+                        with gil:
+                            if uniques.external_view_exists:
+                                raise ValueError("external reference to "
+                                                 "uniques held, but "
+                                                 "Vector.resize() needed")
+                            uniques.resize(uniques.data.capacity * 4)
+                            if use_result_mask:
+                                if result_mask.external_view_exists:
+                                    raise ValueError("external reference to "
+                                                     "result_mask held, but "
+                                                     "Vector.resize() needed")
+                                result_mask.resize(result_mask.data.capacity * 4)
+                    append_data_{{dtype}}(ud, val)
+                    if use_result_mask:
+                        append_data_uint8(rmd, 0)
+
+                    if return_inverse:
+                        self.table.vals[k] = count
+                        labels[i] = count
+                        count += 1
+                elif return_inverse:
+                    # k falls into a previous bucket
+                    # only relevant in case we need to construct the inverse
+                    idx = self.table.vals[k]
+                    labels[i] = idx
+
+        if return_inverse:
+            return uniques.to_array(), labels.base  # .base -> underlying ndarray
+        if use_result_mask:
+            return uniques.to_array(), result_mask.to_array()
+        return uniques.to_array()
+
+    def unique(self, const {{dtype}}_t[:] values, *, bint return_inverse=False, object mask=None):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[{{dtype}}]
+            Array of values of which unique will be calculated
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+        mask : ndarray[bool], optional
+            If not None, the mask is used as indicator for missing values
+            (True = missing, False = valid) instead of `na_value` or
+
+        Returns
+        -------
+        uniques : ndarray[{{dtype}}]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse)
+            The labels from values to uniques
+        result_mask: ndarray[bool], if mask is given as input
+            The mask for the result values.
+        """
+        uniques = {{name}}Vector()
+        use_result_mask = True if mask is not None else False
+        return self._unique(values, uniques, ignore_na=False,
+                            return_inverse=return_inverse, mask=mask, use_result_mask=use_result_mask)
+
+    def factorize(self, const {{dtype}}_t[:] values, Py_ssize_t na_sentinel=-1,
+                  object na_value=None, object mask=None, ignore_na=True):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Missing values are not included in the "uniques" for this method.
+        The labels for any missing values will be set to "na_sentinel"
+
+        Parameters
+        ----------
+        values : ndarray[{{dtype}}]
+            Array of values of which unique will be calculated
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then
+            any value "val" satisfying val != val is considered missing.
+            If na_value is not None, then _additionally_, any value "val"
+            satisfying val == na_value is considered missing.
+        mask : ndarray[bool], optional
+            If not None, the mask is used as indicator for missing values
+            (True = missing, False = valid) instead of `na_value` or
+            condition "val != val".
+
+        Returns
+        -------
+        uniques : ndarray[{{dtype}}]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t]
+            The labels from values to uniques
+        """
+        uniques_vector = {{name}}Vector()
+        return self._unique(values, uniques_vector, na_sentinel=na_sentinel,
+                            na_value=na_value, ignore_na=ignore_na, mask=mask,
+                            return_inverse=True)
+
+    def get_labels(self, const {{dtype}}_t[:] values, {{name}}Vector uniques,
+                   Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                   object na_value=None, object mask=None):
+        # -> np.ndarray[np.intp]
+        _, labels = self._unique(values, uniques, count_prior=count_prior,
+                                 na_sentinel=na_sentinel, na_value=na_value,
+                                 ignore_na=True, return_inverse=True, mask=mask)
+        return labels
+
+    {{if dtype == 'int64'}}
+    @cython.boundscheck(False)
+    def get_labels_groupby(
+        self, const {{dtype}}_t[:] values
+    ) -> tuple[ndarray, ndarray]:
+        # tuple[np.ndarray[np.intp], np.ndarray[{{dtype}}]]
+        cdef:
+            Py_ssize_t i, n = len(values)
+            intp_t[::1] labels
+            Py_ssize_t idx, count = 0
+            int ret = 0
+            {{c_type}} val
+            khiter_t k
+            {{name}}Vector uniques = {{name}}Vector()
+            {{name}}VectorData *ud
+
+        labels = np.empty(n, dtype=np.intp)
+        ud = &uniques.data
+
+        with nogil:
+            for i in range(n):
+                val = {{to_c_type}}(values[i])
+
+                # specific for groupby
+                if val < 0:
+                    labels[i] = -1
+                    continue
+
+                k = kh_get_{{dtype}}(self.table, val)
+                if k != self.table.n_buckets:
+                    idx = self.table.vals[k]
+                    labels[i] = idx
+                else:
+                    k = kh_put_{{dtype}}(self.table, val, &ret)
+                    self.table.vals[k] = count
+
+                    if needs_resize(ud.size, ud.capacity):
+                        with gil:
+                            uniques.resize(uniques.data.capacity * 4)
+                    append_data_{{dtype}}(ud, val)
+                    labels[i] = count
+                    count += 1
+
+        arr_uniques = uniques.to_array()
+
+        return np.asarray(labels), arr_uniques
+    {{endif}}
+
+
+cdef class {{name}}Factorizer(Factorizer):
+    cdef public:
+        {{name}}HashTable table
+        {{name}}Vector uniques
+
+    def __cinit__(self, size_hint: int, uses_mask: bool = False):
+        self.table = {{name}}HashTable(size_hint, uses_mask=uses_mask)
+        self.uniques = {{name}}Vector()
+
+    def factorize(self, const {{c_type}}[:] values,
+                  na_sentinel=-1, na_value=None, object mask=None) -> np.ndarray:
+        """
+        Returns
+        -------
+        ndarray[intp_t]
+
+        Examples
+        --------
+        Factorize values with nans replaced by na_sentinel
+
+        >>> fac = {{name}}Factorizer(3)
+        >>> fac.factorize(np.array([1,2,3], dtype="{{dtype}}"), na_sentinel=20)
+        array([0, 1, 2])
+        """
+        cdef:
+            ndarray[intp_t] labels
+
+        if self.uniques.external_view_exists:
+            uniques = {{name}}Vector()
+            uniques.extend(self.uniques.to_array())
+            self.uniques = uniques
+        labels = self.table.get_labels(values, self.uniques,
+                                       self.count, na_sentinel,
+                                       na_value=na_value, mask=mask)
+        self.count = len(self.uniques)
+        return labels
+
+    def hash_inner_join(self, const {{c_type}}[:] values, const uint8_t[:] mask = None) -> tuple[np.ndarray, np.ndarray]:
+        return self.table.hash_inner_join(values, mask)
+
+{{endfor}}
+
+
+cdef class StringHashTable(HashTable):
+    # these by-definition *must* be strings
+    # or a sentinel np.nan / None missing value
+    na_string_sentinel = '__nan__'
+
+    def __init__(self, int64_t size_hint=1):
+        self.table = kh_init_str()
+        size_hint = min(kh_needed_n_buckets(size_hint), SIZE_HINT_LIMIT)
+        kh_resize_str(self.table, size_hint)
+
+    def __dealloc__(self):
+        if self.table is not NULL:
+            kh_destroy_str(self.table)
+            self.table = NULL
+
+    def __len__(self) -> int:
+        return self.table.size
+
+    def sizeof(self, deep: bool = False) -> int:
+        overhead = 4 * sizeof(uint32_t) + 3 * sizeof(uint32_t*)
+        for_flags = max(1, self.table.n_buckets >> 5) * sizeof(uint32_t)
+        for_pairs =  self.table.n_buckets * (sizeof(char *) +      # keys
+                                             sizeof(Py_ssize_t))   # vals
+        return overhead + for_flags + for_pairs
+
+    def get_state(self) -> dict[str, int]:
+        """ returns infos about the state of the hashtable"""
+        return {
+            'n_buckets' : self.table.n_buckets,
+            'size' : self.table.size,
+            'n_occupied' : self.table.n_occupied,
+            'upper_bound' : self.table.upper_bound,
+        }
+
+    cpdef get_item(self, str val):
+        cdef:
+            khiter_t k
+            const char *v
+        v = PyUnicode_AsUTF8(val)
+
+        k = kh_get_str(self.table, v)
+        if k != self.table.n_buckets:
+            return self.table.vals[k]
+        else:
+            raise KeyError(val)
+
+    cpdef set_item(self, str key, Py_ssize_t val):
+        cdef:
+            khiter_t k
+            int ret = 0
+            const char *v
+
+        v = PyUnicode_AsUTF8(key)
+
+        k = kh_put_str(self.table, v, &ret)
+        if kh_exist_str(self.table, k):
+            self.table.vals[k] = val
+        else:
+            raise KeyError(key)
+
+    @cython.boundscheck(False)
+    def get_indexer(self, ndarray[object] values) -> ndarray:
+        # -> np.ndarray[np.intp]
+        cdef:
+            Py_ssize_t i, n = len(values)
+            ndarray[intp_t] labels = np.empty(n, dtype=np.intp)
+            intp_t *resbuf = <intp_t*>labels.data
+            khiter_t k
+            kh_str_t *table = self.table
+            const char *v
+            const char **vecs
+
+        vecs = <const char **>malloc(n * sizeof(char *))
+        if vecs is NULL:
+            raise MemoryError()
+        for i in range(n):
+            val = values[i]
+            v = PyUnicode_AsUTF8(val)
+            vecs[i] = v
+
+        with nogil:
+            for i in range(n):
+                k = kh_get_str(table, vecs[i])
+                if k != table.n_buckets:
+                    resbuf[i] = table.vals[k]
+                else:
+                    resbuf[i] = -1
+
+        free(vecs)
+        return labels
+
+    @cython.boundscheck(False)
+    def lookup(self, ndarray[object] values, object mask = None) -> ndarray:
+        # -> np.ndarray[np.intp]
+        # mask not yet implemented
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            object val
+            const char *v
+            khiter_t k
+            intp_t[::1] locs = np.empty(n, dtype=np.intp)
+
+        # these by-definition *must* be strings
+        vecs = <const char **>malloc(n * sizeof(char *))
+        if vecs is NULL:
+            raise MemoryError()
+        for i in range(n):
+            val = values[i]
+
+            if isinstance(val, str):
+                # GH#31499 if we have a np.str_ PyUnicode_AsUTF8 won't recognize
+                #  it as a str, even though isinstance does.
+                v = PyUnicode_AsUTF8(<str>val)
+            else:
+                v = PyUnicode_AsUTF8(self.na_string_sentinel)
+            vecs[i] = v
+
+        with nogil:
+            for i in range(n):
+                v = vecs[i]
+                k = kh_get_str(self.table, v)
+                if k != self.table.n_buckets:
+                    locs[i] = self.table.vals[k]
+                else:
+                    locs[i] = -1
+
+        free(vecs)
+        return np.asarray(locs)
+
+    @cython.boundscheck(False)
+    def map_locations(self, ndarray[object] values, object mask = None) -> None:
+        # mask not yet implemented
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            object val
+            const char *v
+            const char **vecs
+            khiter_t k
+
+        # these by-definition *must* be strings
+        vecs = <const char **>malloc(n * sizeof(char *))
+        if vecs is NULL:
+            raise MemoryError()
+        for i in range(n):
+            val = values[i]
+
+            if isinstance(val, str):
+                # GH#31499 if we have a np.str_ PyUnicode_AsUTF8 won't recognize
+                #  it as a str, even though isinstance does.
+                v = PyUnicode_AsUTF8(<str>val)
+            else:
+                v = PyUnicode_AsUTF8(self.na_string_sentinel)
+            vecs[i] = v
+
+        with nogil:
+            for i in range(n):
+                v = vecs[i]
+                k = kh_put_str(self.table, v, &ret)
+                self.table.vals[k] = i
+        free(vecs)
+
+    @cython.boundscheck(False)
+    @cython.wraparound(False)
+    def _unique(self, ndarray[object] values, ObjectVector uniques,
+                Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                object na_value=None, bint ignore_na=False,
+                bint return_inverse=False):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        uniques : ObjectVector
+            Vector into which uniques will be written
+        count_prior : Py_ssize_t, default 0
+            Number of existing entries in uniques
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then any value
+            that is not a string is considered missing. If na_value is
+            not None, then _additionally_ any value "val" satisfying
+            val == na_value is considered missing.
+        ignore_na : bool, default False
+            Whether NA-values should be ignored for calculating the uniques. If
+            True, the labels corresponding to missing values will be set to
+            na_sentinel.
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse=True)
+            The labels from values to uniques
+        """
+        cdef:
+            Py_ssize_t i, idx, count = count_prior, n = len(values)
+            intp_t[::1] labels
+            int64_t[::1] uindexer
+            int ret = 0
+            object val
+            const char *v
+            const char **vecs
+            khiter_t k
+            bint use_na_value
+            bint non_null_na_value
+
+        if return_inverse:
+            labels = np.zeros(n, dtype=np.intp)
+        uindexer = np.empty(n, dtype=np.int64)
+        use_na_value = na_value is not None
+        non_null_na_value = not checknull(na_value)
+
+        # assign pointers and pre-filter out missing (if ignore_na)
+        vecs = <const char **>malloc(n * sizeof(char *))
+        if vecs is NULL:
+            raise MemoryError()
+        for i in range(n):
+            val = values[i]
+
+            if (ignore_na
+                and (not isinstance(val, str)
+                     or (use_na_value and (
+                        (non_null_na_value and val == na_value) or
+                        (not non_null_na_value and is_matching_na(val, na_value)))
+                        )
+                    )
+                ):
+                # if missing values do not count as unique values (i.e. if
+                # ignore_na is True), we can skip the actual value, and
+                # replace the label with na_sentinel directly
+                labels[i] = na_sentinel
+            else:
+                # if ignore_na is False, we also stringify NaN/None/etc.
+                try:
+                    v = PyUnicode_AsUTF8(<str>val)
+                except UnicodeEncodeError:
+                    v = PyUnicode_AsUTF8(<str>repr(val))
+                vecs[i] = v
+
+        # compute
+        with nogil:
+            for i in range(n):
+                if ignore_na and labels[i] == na_sentinel:
+                    # skip entries for ignored missing values (see above)
+                    continue
+
+                v = vecs[i]
+                k = kh_get_str(self.table, v)
+                if k == self.table.n_buckets:
+                    # k hasn't been seen yet
+                    k = kh_put_str(self.table, v, &ret)
+                    uindexer[count] = i
+                    if return_inverse:
+                        self.table.vals[k] = count
+                        labels[i] = count
+                    count += 1
+                elif return_inverse:
+                    # k falls into a previous bucket
+                    # only relevant in case we need to construct the inverse
+                    idx = self.table.vals[k]
+                    labels[i] = idx
+
+        free(vecs)
+
+        # uniques
+        for i in range(count):
+            uniques.append(values[uindexer[i]])
+
+        if return_inverse:
+            return uniques.to_array(), labels.base  # .base -> underlying ndarray
+        return uniques.to_array()
+
+    def unique(self, ndarray[object] values, *, bint return_inverse=False, object mask=None):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+        mask : ndarray[bool], optional
+            Not yet implemented for StringHashTable
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse)
+            The labels from values to uniques
+        """
+        uniques = ObjectVector()
+        return self._unique(values, uniques, ignore_na=False,
+                            return_inverse=return_inverse)
+
+    def factorize(self, ndarray[object] values, Py_ssize_t na_sentinel=-1,
+                  object na_value=None, object mask=None, ignore_na=True):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Missing values are not included in the "uniques" for this method.
+        The labels for any missing values will be set to "na_sentinel"
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then any value
+            that is not a string is considered missing. If na_value is
+            not None, then _additionally_ any value "val" satisfying
+            val == na_value is considered missing.
+        mask : ndarray[bool], optional
+            Not yet implemented for StringHashTable.
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp]
+            The labels from values to uniques
+        """
+        uniques_vector = ObjectVector()
+        return self._unique(values, uniques_vector, na_sentinel=na_sentinel,
+                            na_value=na_value, ignore_na=ignore_na,
+                            return_inverse=True)
+
+    # Add unused mask parameter for compat with other signatures
+    def get_labels(self, ndarray[object] values, ObjectVector uniques,
+                   Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                   object na_value=None, object mask=None):
+        # -> np.ndarray[np.intp]
+        _, labels = self._unique(values, uniques, count_prior=count_prior,
+                                 na_sentinel=na_sentinel, na_value=na_value,
+                                 ignore_na=True, return_inverse=True)
+        return labels
+
+
+cdef class PyObjectHashTable(HashTable):
+
+    def __init__(self, int64_t size_hint=1):
+        self.table = kh_init_pymap()
+        size_hint = min(kh_needed_n_buckets(size_hint), SIZE_HINT_LIMIT)
+        kh_resize_pymap(self.table, size_hint)
+
+    def __dealloc__(self):
+        if self.table is not NULL:
+            kh_destroy_pymap(self.table)
+            self.table = NULL
+
+    def __len__(self) -> int:
+        return self.table.size
+
+    def __contains__(self, object key) -> bool:
+        cdef:
+            khiter_t k
+        hash(key)
+
+        k = kh_get_pymap(self.table, <PyObject*>key)
+        return k != self.table.n_buckets
+
+    def sizeof(self, deep: bool = False) -> int:
+        """ return the size of my table in bytes """
+        overhead = 4 * sizeof(uint32_t) + 3 * sizeof(uint32_t*)
+        for_flags = max(1, self.table.n_buckets >> 5) * sizeof(uint32_t)
+        for_pairs =  self.table.n_buckets * (sizeof(PyObject *) +  # keys
+                                             sizeof(Py_ssize_t))   # vals
+        return overhead + for_flags + for_pairs
+
+    def get_state(self) -> dict[str, int]:
+        """
+        returns infos about the current state of the hashtable like size,
+        number of buckets and so on.
+        """
+        return {
+            'n_buckets' : self.table.n_buckets,
+            'size' : self.table.size,
+            'n_occupied' : self.table.n_occupied,
+            'upper_bound' : self.table.upper_bound,
+        }
+
+    cpdef get_item(self, object val):
+        cdef:
+            khiter_t k
+
+        k = kh_get_pymap(self.table, <PyObject*>val)
+        if k != self.table.n_buckets:
+            return self.table.vals[k]
+        else:
+            raise KeyError(val)
+
+    cpdef set_item(self, object key, Py_ssize_t val):
+        cdef:
+            khiter_t k
+            int ret = 0
+            char* buf
+
+        hash(key)
+
+        k = kh_put_pymap(self.table, <PyObject*>key, &ret)
+        if kh_exist_pymap(self.table, k):
+            self.table.vals[k] = val
+        else:
+            raise KeyError(key)
+
+    def map_locations(self, ndarray[object] values, object mask = None) -> None:
+        # mask not yet implemented
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            object val
+            khiter_t k
+
+        for i in range(n):
+            val = values[i]
+            hash(val)
+
+            k = kh_put_pymap(self.table, <PyObject*>val, &ret)
+            self.table.vals[k] = i
+
+    def lookup(self, ndarray[object] values, object mask = None) -> ndarray:
+        # -> np.ndarray[np.intp]
+        # mask not yet implemented
+        cdef:
+            Py_ssize_t i, n = len(values)
+            int ret = 0
+            object val
+            khiter_t k
+            intp_t[::1] locs = np.empty(n, dtype=np.intp)
+
+        for i in range(n):
+            val = values[i]
+            hash(val)
+
+            k = kh_get_pymap(self.table, <PyObject*>val)
+            if k != self.table.n_buckets:
+                locs[i] = self.table.vals[k]
+            else:
+                locs[i] = -1
+
+        return np.asarray(locs)
+
+    @cython.boundscheck(False)
+    @cython.wraparound(False)
+    def _unique(self, ndarray[object] values, ObjectVector uniques,
+                Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                object na_value=None, bint ignore_na=False,
+                bint return_inverse=False):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        uniques : ObjectVector
+            Vector into which uniques will be written
+        count_prior : Py_ssize_t, default 0
+            Number of existing entries in uniques
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then None _plus_
+            any value "val" satisfying val != val is considered missing.
+            If na_value is not None, then _additionally_, any value "val"
+            satisfying val == na_value is considered missing.
+        ignore_na : bool, default False
+            Whether NA-values should be ignored for calculating the uniques. If
+            True, the labels corresponding to missing values will be set to
+            na_sentinel.
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse=True)
+            The labels from values to uniques
+        """
+        cdef:
+            Py_ssize_t i, idx, count = count_prior, n = len(values)
+            intp_t[::1] labels
+            int ret = 0
+            object val
+            khiter_t k
+            bint use_na_value
+            bint non_null_na_value
+        if return_inverse:
+            labels = np.empty(n, dtype=np.intp)
+        use_na_value = na_value is not None
+        non_null_na_value = not checknull(na_value)
+
+        for i in range(n):
+            val = values[i]
+            hash(val)
+
+            if ignore_na and (
+                checknull(val)
+                or (use_na_value and (
+                    (non_null_na_value and val == na_value) or
+                    (not non_null_na_value and is_matching_na(val, na_value))
+                    )
+                )
+            ):
+                # if missing values do not count as unique values (i.e. if
+                # ignore_na is True), skip the hashtable entry for them, and
+                # replace the corresponding label with na_sentinel
+                labels[i] = na_sentinel
+                continue
+
+            k = kh_get_pymap(self.table, <PyObject*>val)
+            if k == self.table.n_buckets:
+                # k hasn't been seen yet
+                k = kh_put_pymap(self.table, <PyObject*>val, &ret)
+                uniques.append(val)
+                if return_inverse:
+                    self.table.vals[k] = count
+                    labels[i] = count
+                    count += 1
+            elif return_inverse:
+                # k falls into a previous bucket
+                # only relevant in case we need to construct the inverse
+                idx = self.table.vals[k]
+                labels[i] = idx
+
+        if return_inverse:
+            return uniques.to_array(), labels.base  # .base -> underlying ndarray
+        return uniques.to_array()
+
+    def unique(self, ndarray[object] values, *, bint return_inverse=False, object mask=None):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        return_inverse : bool, default False
+            Whether the mapping of the original array values to their location
+            in the vector of uniques should be returned.
+        mask : ndarray[bool], optional
+            Not yet implemented for PyObjectHashTable
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t] (if return_inverse)
+            The labels from values to uniques
+        """
+        uniques = ObjectVector()
+        return self._unique(values, uniques, ignore_na=False,
+                            return_inverse=return_inverse)
+
+    def factorize(self, ndarray[object] values, Py_ssize_t na_sentinel=-1,
+                  object na_value=None, object mask=None, ignore_na=True):
+        """
+        Calculate unique values and labels (no sorting!)
+
+        Missing values are not included in the "uniques" for this method.
+        The labels for any missing values will be set to "na_sentinel"
+
+        Parameters
+        ----------
+        values : ndarray[object]
+            Array of values of which unique will be calculated
+        na_sentinel : Py_ssize_t, default -1
+            Sentinel value used for all NA-values in inverse
+        na_value : object, default None
+            Value to identify as missing. If na_value is None, then None _plus_
+            any value "val" satisfying val != val is considered missing.
+            If na_value is not None, then _additionally_, any value "val"
+            satisfying val == na_value is considered missing.
+        mask : ndarray[bool], optional
+            Not yet implemented for PyObjectHashTable.
+
+        Returns
+        -------
+        uniques : ndarray[object]
+            Unique values of input, not sorted
+        labels : ndarray[intp_t]
+            The labels from values to uniques
+        """
+        uniques_vector = ObjectVector()
+        return self._unique(values, uniques_vector, na_sentinel=na_sentinel,
+                            na_value=na_value, ignore_na=ignore_na,
+                            return_inverse=True)
+
+    # Add unused mask parameter for compat with other signatures
+    def get_labels(self, ndarray[object] values, ObjectVector uniques,
+                   Py_ssize_t count_prior=0, Py_ssize_t na_sentinel=-1,
+                   object na_value=None, object mask=None):
+        # -> np.ndarray[np.intp]
+        _, labels = self._unique(values, uniques, count_prior=count_prior,
+                                 na_sentinel=na_sentinel, na_value=na_value,
+                                 ignore_na=True, return_inverse=True)
+        return labels
diff --git a/pandas/pandas/_libs/hashtable_func_helper.pxi.in b/pandas/pandas/_libs/hashtable_func_helper.pxi.in
new file mode 100644
index 0000000000000..f957ebdeaf67a
--- /dev/null
+++ b/pandas/pandas/_libs/hashtable_func_helper.pxi.in
@@ -0,0 +1,482 @@
+"""
+Template for each `dtype` helper function for hashtable
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+{{py:
+
+# name, dtype, ttype, c_type, to_c_type
+dtypes = [('Complex128', 'complex128', 'complex128',
+                         'khcomplex128_t', 'to_khcomplex128_t'),
+          ('Complex64', 'complex64', 'complex64',
+                        'khcomplex64_t', 'to_khcomplex64_t'),
+          ('Float64', 'float64', 'float64', 'float64_t', ''),
+          ('Float32', 'float32', 'float32', 'float32_t', ''),
+          ('UInt64', 'uint64', 'uint64', 'uint64_t', ''),
+          ('UInt32', 'uint32', 'uint32', 'uint32_t', ''),
+          ('UInt16', 'uint16', 'uint16', 'uint16_t', ''),
+          ('UInt8', 'uint8', 'uint8', 'uint8_t', ''),
+          ('Object', 'object', 'pymap', 'object', '<PyObject*>'),
+          ('Int64', 'int64', 'int64', 'int64_t', ''),
+          ('Int32', 'int32', 'int32', 'int32_t', ''),
+          ('Int16', 'int16', 'int16', 'int16_t', ''),
+          ('Int8', 'int8', 'int8', 'int8_t', '')]
+
+}}
+
+{{for name, dtype, ttype, c_type, to_c_type in dtypes}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if dtype == 'object'}}
+cdef value_count_{{dtype}}(ndarray[{{dtype}}] values, bint dropna, const uint8_t[:] mask=None):
+{{else}}
+cdef value_count_{{dtype}}(const {{dtype}}_t[:] values, bint dropna, const uint8_t[:] mask=None):
+{{endif}}
+    cdef:
+        Py_ssize_t i = 0, na_counter = 0, na_add = 0
+        Py_ssize_t n = len(values)
+        kh_{{ttype}}_t *table
+
+        # Don't use Py_ssize_t, since table.n_buckets is unsigned
+        khiter_t k
+
+        {{c_type}} val
+
+        int ret = 0
+        bint uses_mask = mask is not None
+        bint isna_entry = False
+
+    # we track the order in which keys are first seen (GH39009),
+    # khash-map isn't insertion-ordered, thus:
+    #    table maps keys to counts
+    #    result_keys remembers the original order of keys
+
+    result_keys = {{name}}Vector()
+    table = kh_init_{{ttype}}()
+
+    {{if dtype == 'object'}}
+    if uses_mask:
+        raise NotImplementedError("uses_mask not implemented with object dtype")
+
+    kh_resize_{{ttype}}(table, n // 10)
+
+    for i in range(n):
+        val = values[i]
+        if not dropna or not checknull(val):
+            k = kh_get_{{ttype}}(table, {{to_c_type}}val)
+            if k != table.n_buckets:
+                table.vals[k] += 1
+            else:
+                k = kh_put_{{ttype}}(table, {{to_c_type}}val, &ret)
+                table.vals[k] = 1
+                result_keys.append(val)
+    {{else}}
+    kh_resize_{{ttype}}(table, n)
+
+    for i in range(n):
+        val = {{to_c_type}}(values[i])
+
+        if uses_mask:
+            isna_entry = mask[i]
+
+        if dropna:
+            if not uses_mask:
+                isna_entry = is_nan_{{c_type}}(val)
+
+        if not dropna or not isna_entry:
+            if uses_mask and isna_entry:
+                na_counter += 1
+            else:
+                k = kh_get_{{ttype}}(table, val)
+                if k != table.n_buckets:
+                    table.vals[k] += 1
+                else:
+                    k = kh_put_{{ttype}}(table, val, &ret)
+                    table.vals[k] = 1
+                    result_keys.append(val)
+    {{endif}}
+
+    # collect counts in the order corresponding to result_keys:
+    if na_counter > 0:
+        na_add = 1
+    cdef:
+        int64_t[::1] result_counts = np.empty(table.size + na_add, dtype=np.int64)
+
+    for i in range(table.size):
+        {{if dtype == 'object'}}
+        k = kh_get_{{ttype}}(table, result_keys.data[i])
+        {{else}}
+        k = kh_get_{{ttype}}(table, result_keys.data.data[i])
+        {{endif}}
+        result_counts[i] = table.vals[k]
+
+    if na_counter > 0:
+        result_counts[table.size] = na_counter
+        result_keys.append(val)
+
+    kh_destroy_{{ttype}}(table)
+
+    return result_keys.to_array(), result_counts.base, na_counter
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if dtype == 'object'}}
+cdef duplicated_{{dtype}}(ndarray[{{dtype}}] values, object keep='first', const uint8_t[:] mask=None):
+{{else}}
+cdef duplicated_{{dtype}}(const {{dtype}}_t[:] values, object keep='first', const uint8_t[:] mask=None):
+{{endif}}
+    cdef:
+        int ret = 0
+        {{if dtype != 'object'}}
+        {{c_type}} value
+        {{else}}
+        PyObject* value
+        {{endif}}
+        Py_ssize_t i, n = len(values), first_na = -1
+        khiter_t k
+        kh_{{ttype}}_t *table = kh_init_{{ttype}}()
+        ndarray[uint8_t, ndim=1, cast=True] out = np.empty(n, dtype='bool')
+        bint seen_na = False, uses_mask = mask is not None
+        bint seen_multiple_na = False
+
+    kh_resize_{{ttype}}(table, min(kh_needed_n_buckets(n), SIZE_HINT_LIMIT))
+
+    if keep not in ('last', 'first', False):
+        raise ValueError('keep must be either "first", "last" or False')
+
+    {{for cond, keep in [('if', '"last"'), ('elif', '"first"')]}}
+    {{cond}} keep == {{keep}}:
+        {{if dtype == 'object'}}
+        if True:
+        {{else}}
+        with nogil:
+        {{endif}}
+            {{if keep == '"last"'}}
+            for i in range(n - 1, -1, -1):
+            {{else}}
+            for i in range(n):
+            {{endif}}
+                if uses_mask and mask[i]:
+                    if seen_na:
+                        out[i] = True
+                    else:
+                        out[i] = False
+                        seen_na = True
+                else:
+                    value = {{to_c_type}}(values[i])
+                    kh_put_{{ttype}}(table, value, &ret)
+                    out[i] = ret == 0
+    {{endfor}}
+
+    else:
+        {{if dtype == 'object'}}
+        if True:
+        {{else}}
+        with nogil:
+        {{endif}}
+            for i in range(n):
+                if uses_mask and mask[i]:
+                    if not seen_na:
+                        first_na = i
+                        seen_na = True
+                        out[i] = 0
+                    elif not seen_multiple_na:
+                        out[i] = 1
+                        out[first_na] = 1
+                        seen_multiple_na = True
+                    else:
+                        out[i] = 1
+
+                else:
+                    value = {{to_c_type}}(values[i])
+                    k = kh_get_{{ttype}}(table, value)
+                    if k != table.n_buckets:
+                        out[table.vals[k]] = 1
+                        out[i] = 1
+                    else:
+                        k = kh_put_{{ttype}}(table, value, &ret)
+                        table.vals[k] = i
+                        out[i] = 0
+
+    kh_destroy_{{ttype}}(table)
+    return out
+
+
+# ----------------------------------------------------------------------
+# Membership
+# ----------------------------------------------------------------------
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+{{if dtype == 'object'}}
+cdef ismember_{{dtype}}(ndarray[{{c_type}}] arr, ndarray[{{c_type}}] values):
+{{else}}
+cdef ismember_{{dtype}}(const {{dtype}}_t[:] arr, const {{dtype}}_t[:] values):
+{{endif}}
+    """
+    Return boolean of values in arr on an
+    element by-element basis
+
+    Parameters
+    ----------
+    arr : {{dtype}} ndarray
+    values : {{dtype}} ndarray
+
+    Returns
+    -------
+    boolean ndarray len of (arr)
+    """
+    cdef:
+        Py_ssize_t i, n
+        khiter_t k
+        int ret = 0
+        ndarray[uint8_t] result
+
+        {{if dtype == "object"}}
+        PyObject* val
+        {{else}}
+        {{c_type}} val
+        {{endif}}
+
+        kh_{{ttype}}_t *table = kh_init_{{ttype}}()
+
+    # construct the table
+    n = len(values)
+    kh_resize_{{ttype}}(table, n)
+
+    {{if dtype == 'object'}}
+    if True:
+    {{else}}
+    with nogil:
+    {{endif}}
+        for i in range(n):
+            val = {{to_c_type}}(values[i])
+            kh_put_{{ttype}}(table, val, &ret)
+
+    # test membership
+    n = len(arr)
+    result = np.empty(n, dtype=np.uint8)
+
+    {{if dtype == 'object'}}
+    if True:
+    {{else}}
+    with nogil:
+    {{endif}}
+        for i in range(n):
+            val = {{to_c_type}}(arr[i])
+            k = kh_get_{{ttype}}(table, val)
+            result[i] = (k != table.n_buckets)
+
+    kh_destroy_{{ttype}}(table)
+    return result.view(np.bool_)
+
+# ----------------------------------------------------------------------
+# Mode Computations
+# ----------------------------------------------------------------------
+
+{{endfor}}
+
+
+ctypedef fused htfunc_t:
+    numeric_object_t
+    complex128_t
+    complex64_t
+
+
+cpdef value_count(ndarray[htfunc_t] values, bint dropna, const uint8_t[:] mask=None):
+    if htfunc_t is object:
+        return value_count_object(values, dropna, mask=mask)
+
+    elif htfunc_t is int8_t:
+        return value_count_int8(values, dropna, mask=mask)
+    elif htfunc_t is int16_t:
+        return value_count_int16(values, dropna, mask=mask)
+    elif htfunc_t is int32_t:
+        return value_count_int32(values, dropna, mask=mask)
+    elif htfunc_t is int64_t:
+        return value_count_int64(values, dropna, mask=mask)
+
+    elif htfunc_t is uint8_t:
+        return value_count_uint8(values, dropna, mask=mask)
+    elif htfunc_t is uint16_t:
+        return value_count_uint16(values, dropna, mask=mask)
+    elif htfunc_t is uint32_t:
+        return value_count_uint32(values, dropna, mask=mask)
+    elif htfunc_t is uint64_t:
+        return value_count_uint64(values, dropna, mask=mask)
+
+    elif htfunc_t is float64_t:
+        return value_count_float64(values, dropna, mask=mask)
+    elif htfunc_t is float32_t:
+        return value_count_float32(values, dropna, mask=mask)
+
+    elif htfunc_t is complex128_t:
+        return value_count_complex128(values, dropna, mask=mask)
+    elif htfunc_t is complex64_t:
+        return value_count_complex64(values, dropna, mask=mask)
+
+    else:
+        raise TypeError(values.dtype)
+
+
+cpdef duplicated(ndarray[htfunc_t] values, object keep="first", const uint8_t[:] mask=None):
+    if htfunc_t is object:
+        return duplicated_object(values, keep, mask=mask)
+
+    elif htfunc_t is int8_t:
+        return duplicated_int8(values, keep, mask=mask)
+    elif htfunc_t is int16_t:
+        return duplicated_int16(values, keep, mask=mask)
+    elif htfunc_t is int32_t:
+        return duplicated_int32(values, keep, mask=mask)
+    elif htfunc_t is int64_t:
+        return duplicated_int64(values, keep, mask=mask)
+
+    elif htfunc_t is uint8_t:
+        return duplicated_uint8(values, keep, mask=mask)
+    elif htfunc_t is uint16_t:
+        return duplicated_uint16(values, keep, mask=mask)
+    elif htfunc_t is uint32_t:
+        return duplicated_uint32(values, keep, mask=mask)
+    elif htfunc_t is uint64_t:
+        return duplicated_uint64(values, keep, mask=mask)
+
+    elif htfunc_t is float64_t:
+        return duplicated_float64(values, keep, mask=mask)
+    elif htfunc_t is float32_t:
+        return duplicated_float32(values, keep, mask=mask)
+
+    elif htfunc_t is complex128_t:
+        return duplicated_complex128(values, keep, mask=mask)
+    elif htfunc_t is complex64_t:
+        return duplicated_complex64(values, keep, mask=mask)
+
+    else:
+        raise TypeError(values.dtype)
+
+
+cpdef ismember(ndarray[htfunc_t] arr, ndarray[htfunc_t] values):
+    if htfunc_t is object:
+        return ismember_object(arr, values)
+
+    elif htfunc_t is int8_t:
+        return ismember_int8(arr, values)
+    elif htfunc_t is int16_t:
+        return ismember_int16(arr, values)
+    elif htfunc_t is int32_t:
+        return ismember_int32(arr, values)
+    elif htfunc_t is int64_t:
+        return ismember_int64(arr, values)
+
+    elif htfunc_t is uint8_t:
+        return ismember_uint8(arr, values)
+    elif htfunc_t is uint16_t:
+        return ismember_uint16(arr, values)
+    elif htfunc_t is uint32_t:
+        return ismember_uint32(arr, values)
+    elif htfunc_t is uint64_t:
+        return ismember_uint64(arr, values)
+
+    elif htfunc_t is float64_t:
+        return ismember_float64(arr, values)
+    elif htfunc_t is float32_t:
+        return ismember_float32(arr, values)
+
+    elif htfunc_t is complex128_t:
+        return ismember_complex128(arr, values)
+    elif htfunc_t is complex64_t:
+        return ismember_complex64(arr, values)
+
+    else:
+        raise TypeError(values.dtype)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def mode(ndarray[htfunc_t] values, bint dropna, const uint8_t[:] mask=None):
+    # TODO(cython3): use const htfunct_t[:]
+
+    cdef:
+        ndarray[htfunc_t] keys
+        ndarray[htfunc_t] modes
+        ndarray[uint8_t] res_mask = None
+
+        int64_t[::1] counts
+        int64_t count, _, max_count = -1
+        Py_ssize_t nkeys, k, na_counter, j = 0
+
+    keys, counts, na_counter = value_count(values, dropna, mask=mask)
+    nkeys = len(keys)
+
+    modes = np.empty(nkeys, dtype=values.dtype)
+
+    with nogil(htfunc_t is not object):
+        for k in range(nkeys):
+            count = counts[k]
+            if count == max_count:
+                j += 1
+            elif count > max_count:
+                max_count = count
+                j = 0
+            else:
+                continue
+
+            modes[j] = keys[k]
+
+    if na_counter > 0:
+        res_mask = np.zeros(j+1, dtype=np.bool_)
+        res_mask[j] = True
+    return modes[:j + 1], res_mask
+
+
+{{py:
+
+# name, dtype, ttype, c_type
+dtypes = [('Int64', 'int64', 'int64', 'int64_t'),
+          ('Int32', 'int32', 'int32', 'int32_t'), ]
+
+}}
+
+{{for name, dtype, ttype, c_type in dtypes}}
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def _unique_label_indices_{{dtype}}(const {{c_type}}[:] labels) -> ndarray:
+    """
+    Indices of the first occurrences of the unique labels
+    *excluding* -1. equivalent to:
+        np.unique(labels, return_index=True)[1]
+    """
+    cdef:
+        int ret = 0
+        Py_ssize_t i, n = len(labels)
+        kh_{{ttype}}_t *table = kh_init_{{ttype}}()
+        {{name}}Vector idx = {{name}}Vector()
+        ndarray[{{c_type}}, ndim=1] arr
+        {{name}}VectorData *ud = &idx.data
+
+    kh_resize_{{ttype}}(table, min(kh_needed_n_buckets(n), SIZE_HINT_LIMIT))
+
+    with nogil:
+        for i in range(n):
+            kh_put_{{ttype}}(table, labels[i], &ret)
+            if ret != 0:
+                if needs_resize(ud.size, ud.capacity):
+                    with gil:
+                        idx.resize(idx.data.capacity * 4)
+                append_data_{{ttype}}(ud, i)
+
+    kh_destroy_{{ttype}}(table)
+
+    arr = idx.to_array()
+    arr = arr[np.asarray(labels)[arr].argsort()]
+
+    return arr[1:] if arr.size != 0 and labels[arr[0]] == -1 else arr
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/include/pandas/datetime/date_conversions.h b/pandas/pandas/_libs/include/pandas/datetime/date_conversions.h
new file mode 100644
index 0000000000000..043805a8b25f4
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/datetime/date_conversions.h
@@ -0,0 +1,25 @@
+/*
+Copyright (c) 2020, PyData Development Team
+All rights reserved.
+Distributed under the terms of the BSD Simplified License.
+The full license is in the LICENSE file, distributed with this software.
+*/
+
+#pragma once
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#include <numpy/ndarraytypes.h>
+
+// Scales value inplace from nanosecond resolution to unit resolution
+int scaleNanosecToUnit(int64_t *value, NPY_DATETIMEUNIT unit);
+
+// Converts an int64 object representing a date to ISO format
+// up to precision `base` e.g. base="s" yields 2020-01-03T00:00:00Z
+// while base="ns" yields "2020-01-01T00:00:00.000000000Z"
+// len is mutated to save the length of the returned string
+char *int64ToIso(int64_t value, NPY_DATETIMEUNIT valueUnit,
+                 NPY_DATETIMEUNIT base, size_t *len);
+
+char *int64ToIsoDuration(int64_t value, size_t *len);
diff --git a/pandas/pandas/_libs/include/pandas/datetime/pd_datetime.h b/pandas/pandas/_libs/include/pandas/datetime/pd_datetime.h
new file mode 100644
index 0000000000000..98e5521af2506
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/datetime/pd_datetime.h
@@ -0,0 +1,112 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+Written by Mark Wiebe (mwwiebe@gmail.com)
+
+Copyright (c) 2011 by Enthought, Inc.
+Copyright (c) 2005-2011, NumPy Developers
+
+All rights reserved.
+See NUMPY_LICENSE.txt for the license.
+*/
+
+#pragma once
+
+#ifndef NPY_NO_DEPRECATED_API
+#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
+#endif // NPY_NO_DEPRECATED_API
+
+#include "pandas/datetime/date_conversions.h"
+#include "pandas/vendored/numpy/datetime/np_datetime.h"
+#include "pandas/vendored/numpy/datetime/np_datetime_strings.h"
+#include <numpy/ndarraytypes.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef struct {
+  npy_datetime (*npy_datetimestruct_to_datetime)(NPY_DATETIMEUNIT,
+                                                 const npy_datetimestruct *);
+  int (*scaleNanosecToUnit)(int64_t *, NPY_DATETIMEUNIT);
+  char *(*int64ToIso)(int64_t, NPY_DATETIMEUNIT, NPY_DATETIMEUNIT, size_t *);
+  char *(*PyDateTimeToIso)(PyObject *, NPY_DATETIMEUNIT, size_t *);
+  npy_datetime (*PyDateTimeToEpoch)(PyObject *, NPY_DATETIMEUNIT);
+  char *(*int64ToIsoDuration)(int64_t, size_t *);
+  void (*pandas_datetime_to_datetimestruct)(npy_datetime, NPY_DATETIMEUNIT,
+                                            npy_datetimestruct *);
+  void (*pandas_timedelta_to_timedeltastruct)(npy_datetime, NPY_DATETIMEUNIT,
+                                              pandas_timedeltastruct *);
+  int (*convert_pydatetime_to_datetimestruct)(PyObject *, npy_datetimestruct *);
+  int (*cmp_npy_datetimestruct)(const npy_datetimestruct *,
+                                const npy_datetimestruct *);
+  PyArray_DatetimeMetaData (*get_datetime_metadata_from_dtype)(PyArray_Descr *);
+  int (*parse_iso_8601_datetime)(const char *, int, int, npy_datetimestruct *,
+                                 NPY_DATETIMEUNIT *, int *, int *, const char *,
+                                 int, FormatRequirement);
+  int (*get_datetime_iso_8601_strlen)(int, NPY_DATETIMEUNIT);
+  int (*make_iso_8601_datetime)(npy_datetimestruct *, char *, size_t, int,
+                                NPY_DATETIMEUNIT);
+  int (*make_iso_8601_timedelta)(pandas_timedeltastruct *, char *, size_t *);
+} PandasDateTime_CAPI;
+
+// The capsule name appears limited to module.attributename; see bpo-32414
+// cpython has an open PR gh-6898 to fix, but hasn't had traction for years
+#define PandasDateTime_CAPSULE_NAME "pandas._pandas_datetime_CAPI"
+
+/* block used as part of public API */
+#ifndef _PANDAS_DATETIME_IMPL
+static PandasDateTime_CAPI *PandasDateTimeAPI = NULL;
+
+#define PandasDateTime_IMPORT                                                  \
+  PandasDateTimeAPI =                                                          \
+      (PandasDateTime_CAPI *)PyCapsule_Import(PandasDateTime_CAPSULE_NAME, 0)
+
+#define npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT, npy_datetimestruct)   \
+  PandasDateTimeAPI->npy_datetimestruct_to_datetime((NPY_DATETIMEUNIT),        \
+                                                    (npy_datetimestruct))
+#define scaleNanosecToUnit(value, unit)                                        \
+  PandasDateTimeAPI->scaleNanosecToUnit((value), (unit))
+#define int64ToIso(value, valueUnit, base, len)                                \
+  PandasDateTimeAPI->int64ToIso((value), (valueUnit), (base), (len))
+#define NpyDateTimeToEpoch(dt, base)                                           \
+  PandasDateTimeAPI->NpyDateTimeToEpoch((dt), (base))
+#define PyDateTimeToIso(obj, base, len)                                        \
+  PandasDateTimeAPI->PyDateTimeToIso((obj), (base), (len))
+#define PyDateTimeToEpoch(dt, base)                                            \
+  PandasDateTimeAPI->PyDateTimeToEpoch((dt), (base))
+#define int64ToIsoDuration(value, len)                                         \
+  PandasDateTimeAPI->int64ToIsoDuration((value), (len))
+#define pandas_datetime_to_datetimestruct(dt, base, out)                       \
+  PandasDateTimeAPI->pandas_datetime_to_datetimestruct((dt), (base), (out))
+#define pandas_timedelta_to_timedeltastruct(td, base, out)                     \
+  PandasDateTimeAPI->pandas_timedelta_to_timedeltastruct((td), (base), (out))
+#define convert_pydatetime_to_datetimestruct(dtobj, out)                       \
+  PandasDateTimeAPI->convert_pydatetime_to_datetimestruct((dtobj), (out))
+#define cmp_npy_datetimestruct(a, b)                                           \
+  PandasDateTimeAPI->cmp_npy_datetimestruct((a), (b))
+#define get_datetime_metadata_from_dtype(dtype)                                \
+  PandasDateTimeAPI->get_datetime_metadata_from_dtype((dtype))
+#define parse_iso_8601_datetime(str, len, want_exc, out, out_bestunit,         \
+                                out_local, out_tzoffset, format, format_len,   \
+                                format_requirement)                            \
+  PandasDateTimeAPI->parse_iso_8601_datetime(                                  \
+      (str), (len), (want_exc), (out), (out_bestunit), (out_local),            \
+      (out_tzoffset), (format), (format_len), (format_requirement))
+#define get_datetime_iso_8601_strlen(local, base)                              \
+  PandasDateTimeAPI->get_datetime_iso_8601_strlen((local), (base))
+#define make_iso_8601_datetime(dts, outstr, outlen, utc, base)                 \
+  PandasDateTimeAPI->make_iso_8601_datetime((dts), (outstr), (outlen), (utc),  \
+                                            (base))
+#define make_iso_8601_timedelta(tds, outstr, outlen)                           \
+  PandasDateTimeAPI->make_iso_8601_timedelta((tds), (outstr), (outlen))
+#endif /* !defined(_PANDAS_DATETIME_IMPL) */
+
+#ifdef __cplusplus
+}
+#endif
diff --git a/pandas/pandas/_libs/include/pandas/parser/io.h b/pandas/pandas/_libs/include/pandas/parser/io.h
new file mode 100644
index 0000000000000..41f1bb9312724
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/parser/io.h
@@ -0,0 +1,32 @@
+/*
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+*/
+
+#pragma once
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#include "tokenizer.h"
+
+#define FS(source) ((file_source *)source)
+
+typedef struct _rd_source {
+  PyObject *obj;
+  PyObject *buffer;
+  size_t position;
+} rd_source;
+
+#define RDS(source) ((rd_source *)source)
+
+void *new_rd_source(PyObject *obj);
+
+void del_rd_source(void *src);
+
+char *buffer_rd_bytes(void *source, size_t nbytes, size_t *bytes_read,
+                      int *status, const char *encoding_errors);
diff --git a/pandas/pandas/_libs/include/pandas/parser/pd_parser.h b/pandas/pandas/_libs/include/pandas/parser/pd_parser.h
new file mode 100644
index 0000000000000..543839b5d75bf
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/parser/pd_parser.h
@@ -0,0 +1,111 @@
+/*
+
+Copyright (c) 2023, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+*/
+#pragma once
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#include "pandas/parser/tokenizer.h"
+
+typedef struct {
+  int (*to_double)(char *, double *, char, char, int *);
+  int (*floatify)(PyObject *, double *, int *);
+  void *(*new_rd_source)(PyObject *);
+  void (*del_rd_source)(void *);
+  char *(*buffer_rd_bytes)(void *, size_t, size_t *, int *, const char *);
+  void (*uint_state_init)(uint_state *);
+  int (*uint64_conflict)(uint_state *);
+  void (*coliter_setup)(coliter_t *, parser_t *, int64_t, int64_t);
+  parser_t *(*parser_new)(void);
+  int (*parser_init)(parser_t *);
+  void (*parser_free)(parser_t *);
+  void (*parser_del)(parser_t *);
+  int (*parser_add_skiprow)(parser_t *, int64_t);
+  void (*parser_set_skipfirstnrows)(parser_t *, int64_t);
+  void (*parser_set_default_options)(parser_t *);
+  int (*parser_consume_rows)(parser_t *, size_t);
+  int (*parser_trim_buffers)(parser_t *);
+  int (*tokenize_all_rows)(parser_t *, const char *);
+  int (*tokenize_nrows)(parser_t *, size_t, const char *);
+  int64_t (*str_to_int64)(const char *, int64_t, int64_t, int *, char);
+  uint64_t (*str_to_uint64)(uint_state *, const char *, int64_t, uint64_t,
+                            int *, char);
+  double (*xstrtod)(const char *, char **, char, char, char, int, int *, int *);
+  double (*precise_xstrtod)(const char *, char **, char, char, char, int, int *,
+                            int *);
+  double (*round_trip)(const char *, char **, char, char, char, int, int *,
+                       int *);
+  int (*to_boolean)(const char *, uint8_t *);
+} PandasParser_CAPI;
+
+#define PandasParser_CAPSULE_NAME "pandas._pandas_parser_CAPI"
+
+#ifndef _PANDAS_PARSER_IMPL
+static PandasParser_CAPI *PandasParserAPI = NULL;
+
+#define PandasParser_IMPORT                                                    \
+  PandasParserAPI =                                                            \
+      (PandasParser_CAPI *)PyCapsule_Import(PandasParser_CAPSULE_NAME, 0)
+
+#define to_double(item, p_value, sci, decimal, maybe_int)                      \
+  PandasParserAPI->to_double((item), (p_value), (sci), (decimal), (maybe_int))
+#define floatify(str, result, maybe_int)                                       \
+  PandasParserAPI->floatify((str), (result), (maybe_int))
+#define new_rd_source(obj) PandasParserAPI->new_rd_source((obj))
+#define del_rd_source(src) PandasParserAPI->del_rd_source((src))
+#define buffer_rd_bytes(source, nbytes, bytes_read, status, encoding_errors)   \
+  PandasParserAPI->buffer_rd_bytes((source), (nbytes), (bytes_read), (status), \
+                                   (encoding_errors))
+#define uint_state_init(self) PandasParserAPI->uint_state_init((self))
+#define uint64_conflict(self) PandasParserAPI->uint64_conflict((self))
+#define coliter_setup(self, parser, i, start)                                  \
+  PandasParserAPI->coliter_setup((self), (parser), (i), (start))
+#define parser_new PandasParserAPI->parser_new
+#define parser_init(self) PandasParserAPI->parser_init((self))
+#define parser_free(self) PandasParserAPI->parser_free((self))
+#define parser_del(self) PandasParserAPI->parser_del((self))
+#define parser_add_skiprow(self, row)                                          \
+  PandasParserAPI->parser_add_skiprow((self), (row))
+#define parser_set_skipfirstnrows(self, nrows)                                 \
+  PandasParserAPI->parser_set_skipfirstnrows((self), (nrows))
+#define parser_set_default_options(self)                                       \
+  PandasParserAPI->parser_set_default_options((self))
+#define parser_consume_rows(self, nrows)                                       \
+  PandasParserAPI->parser_consume_rows((self), (nrows))
+#define parser_trim_buffers(self) PandasParserAPI->parser_trim_buffers((self))
+#define tokenize_all_rows(self, encoding_errors)                               \
+  PandasParserAPI->tokenize_all_rows((self), (encoding_errors))
+#define tokenize_nrows(self, nrows, encoding_errors)                           \
+  PandasParserAPI->tokenize_nrows((self), (nrows), (encoding_errors))
+#define str_to_int64(p_item, int_min, int_max, error, t_sep)                   \
+  PandasParserAPI->str_to_int64((p_item), (int_min), (int_max), (error),       \
+                                (t_sep))
+#define str_to_uint64(state, p_item, int_max, uint_max, error, t_sep)          \
+  PandasParserAPI->str_to_uint64((state), (p_item), (int_max), (uint_max),     \
+                                 (error), (t_sep))
+#define xstrtod(p, q, decimal, sci, tsep, skip_trailing, error, maybe_int)     \
+  PandasParserAPI->xstrtod((p), (q), (decimal), (sci), (tsep),                 \
+                           (skip_trailing), (error), (maybe_int))
+#define precise_xstrtod(p, q, decimal, sci, tsep, skip_trailing, error,        \
+                        maybe_int)                                             \
+  PandasParserAPI->precise_xstrtod((p), (q), (decimal), (sci), (tsep),         \
+                                   (skip_trailing), (error), (maybe_int))
+#define round_trip(p, q, decimal, sci, tsep, skip_trailing, error, maybe_int)  \
+  PandasParserAPI->round_trip((p), (q), (decimal), (sci), (tsep),              \
+                              (skip_trailing), (error), (maybe_int))
+#define to_boolean(item, val) PandasParserAPI->to_boolean((item), (val))
+#endif /* !defined(_PANDAS_PARSER_IMPL) */
+
+#ifdef __cplusplus
+}
+#endif
diff --git a/pandas/pandas/_libs/include/pandas/parser/tokenizer.h b/pandas/pandas/_libs/include/pandas/parser/tokenizer.h
new file mode 100644
index 0000000000000..209f375a5bf6c
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/parser/tokenizer.h
@@ -0,0 +1,224 @@
+/*
+
+Copyright (c) 2012, Lambda Foundry, Inc., except where noted
+
+Incorporates components of WarrenWeckesser/textreader, licensed under 3-clause
+BSD
+
+See LICENSE for the license
+
+*/
+
+#pragma once
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#define ERROR_NO_DIGITS 1
+#define ERROR_OVERFLOW 2
+#define ERROR_INVALID_CHARS 3
+
+#include <stdint.h>
+
+#define STREAM_INIT_SIZE 32
+
+#define REACHED_EOF 1
+#define CALLING_READ_FAILED 2
+
+/*
+
+  C flat file parsing low level code for pandas / NumPy
+
+ */
+
+/*
+ *  Common set of error types for the read_rows() and tokenize()
+ *  functions.
+ */
+
+// #define VERBOSE
+#if defined(VERBOSE)
+#define TRACE(X) printf X;
+#else
+#define TRACE(X)
+#endif // VERBOSE
+
+#define PARSER_OUT_OF_MEMORY -1
+
+/*
+ *  TODO: Might want to couple count_rows() with read_rows() to avoid
+ *        duplication of some file I/O.
+ */
+
+typedef enum {
+  START_RECORD,
+  START_FIELD,
+  ESCAPED_CHAR,
+  IN_FIELD,
+  IN_QUOTED_FIELD,
+  ESCAPE_IN_QUOTED_FIELD,
+  QUOTE_IN_QUOTED_FIELD,
+  EAT_CRNL,
+  EAT_CRNL_NOP,
+  EAT_WHITESPACE,
+  EAT_COMMENT,
+  EAT_LINE_COMMENT,
+  WHITESPACE_LINE,
+  START_FIELD_IN_SKIP_LINE,
+  IN_FIELD_IN_SKIP_LINE,
+  IN_QUOTED_FIELD_IN_SKIP_LINE,
+  QUOTE_IN_QUOTED_FIELD_IN_SKIP_LINE,
+  FINISHED
+} ParserState;
+
+typedef enum {
+  QUOTE_MINIMAL,
+  QUOTE_ALL,
+  QUOTE_NONNUMERIC,
+  QUOTE_NONE
+} QuoteStyle;
+
+typedef enum { ERROR, WARN, SKIP } BadLineHandleMethod;
+
+typedef char *(*io_callback)(void *src, size_t nbytes, size_t *bytes_read,
+                             int *status, const char *encoding_errors);
+typedef void (*io_cleanup)(void *src);
+
+typedef struct parser_t {
+  void *source;
+  io_callback cb_io;
+  io_cleanup cb_cleanup;
+
+  int64_t chunksize; // Number of bytes to prepare for each chunk
+  char *data;        // pointer to data to be processed
+  int64_t datalen;   // amount of data available
+  int64_t datapos;
+
+  // where to write out tokenized data
+  char *stream;
+  uint64_t stream_len;
+  uint64_t stream_cap;
+
+  // Store words in (potentially ragged) matrix for now, hmm
+  char **words;
+  int64_t *word_starts; // where we are in the stream
+  uint64_t words_len;
+  uint64_t words_cap;
+  uint64_t max_words_cap; // maximum word cap encountered
+
+  char *pword_start;  // pointer to stream start of current field
+  int64_t word_start; // position start of current field
+
+  int64_t *line_start;  // position in words for start of line
+  int64_t *line_fields; // Number of fields in each line
+  uint64_t lines;       // Number of (good) lines observed
+  uint64_t file_lines;  // Number of lines (including bad or skipped)
+  uint64_t lines_cap;   // Vector capacity
+
+  // Tokenizing stuff
+  ParserState state;
+  int doublequote;      /* is " represented by ""? */
+  char delimiter;       /* field separator */
+  int delim_whitespace; /* delimit by consuming space/tabs instead */
+  char quotechar;       /* quote character */
+  char escapechar;      /* escape character */
+  char lineterminator;
+  int skipinitialspace; /* ignore spaces following delimiter? */
+  int quoting;          /* style of quoting to write */
+
+  char commentchar;
+  int allow_embedded_newline;
+
+  int usecols; // Boolean: 1: usecols provided, 0: none provided
+
+  Py_ssize_t expected_fields;
+  BadLineHandleMethod on_bad_lines;
+
+  // floating point options
+  char decimal;
+  char sci;
+
+  // thousands separator (comma, period)
+  char thousands;
+
+  int header;           // Boolean: 1: has header, 0: no header
+  int64_t header_start; // header row start
+  uint64_t header_end;  // header row end
+
+  void *skipset;
+  PyObject *skipfunc;
+  int64_t skip_first_N_rows;
+  int64_t skip_footer;
+  double (*double_converter)(const char *, char **, char, char, char, int,
+                             int *, int *);
+
+  // error handling
+  char *warn_msg;
+  char *error_msg;
+
+  int skip_empty_lines;
+} parser_t;
+
+typedef struct coliter_t {
+  char **words;
+  int64_t *line_start;
+  int64_t col;
+} coliter_t;
+
+void coliter_setup(coliter_t *self, parser_t *parser, int64_t i, int64_t start);
+
+#define COLITER_NEXT(iter, word)                                               \
+  do {                                                                         \
+    const int64_t i = *iter.line_start++ + iter.col;                           \
+    word = i >= *iter.line_start ? "" : iter.words[i];                         \
+  } while (0)
+
+parser_t *parser_new(void);
+
+int parser_init(parser_t *self);
+
+int parser_consume_rows(parser_t *self, size_t nrows);
+
+int parser_trim_buffers(parser_t *self);
+
+int parser_add_skiprow(parser_t *self, int64_t row);
+
+void parser_set_skipfirstnrows(parser_t *self, int64_t nrows);
+
+void parser_free(parser_t *self);
+
+void parser_del(parser_t *self);
+
+void parser_set_default_options(parser_t *self);
+
+int tokenize_nrows(parser_t *self, size_t nrows, const char *encoding_errors);
+
+int tokenize_all_rows(parser_t *self, const char *encoding_errors);
+
+// Have parsed / type-converted a chunk of data
+// and want to free memory from the token stream
+
+typedef struct uint_state {
+  int seen_sint;
+  int seen_uint;
+  int seen_null;
+} uint_state;
+
+void uint_state_init(uint_state *self);
+
+int uint64_conflict(uint_state *self);
+
+uint64_t str_to_uint64(uint_state *state, const char *p_item, int64_t int_max,
+                       uint64_t uint_max, int *error, char tsep);
+int64_t str_to_int64(const char *p_item, int64_t int_min, int64_t int_max,
+                     int *error, char tsep);
+double xstrtod(const char *p, char **q, char decimal, char sci, char tsep,
+               int skip_trailing, int *error, int *maybe_int);
+double precise_xstrtod(const char *p, char **q, char decimal, char sci,
+                       char tsep, int skip_trailing, int *error,
+                       int *maybe_int);
+
+// GH-15140 - round_trip requires and acquires the GIL on its own
+double round_trip(const char *p, char **q, char decimal, char sci, char tsep,
+                  int skip_trailing, int *error, int *maybe_int);
+int to_boolean(const char *item, uint8_t *val);
diff --git a/pandas/pandas/_libs/include/pandas/portable.h b/pandas/pandas/_libs/include/pandas/portable.h
new file mode 100644
index 0000000000000..1d0509d9e9724
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/portable.h
@@ -0,0 +1,37 @@
+/*
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+*/
+
+#pragma once
+
+#include <string.h>
+
+#if defined(_MSC_VER)
+#define strcasecmp(s1, s2) _stricmp(s1, s2)
+#endif
+
+// GH-23516 - works around locale perf issues
+// from MUSL libc, licence at LICENSES/MUSL_LICENSE
+#define isdigit_ascii(c) (((unsigned)(c) - '0') < 10u)
+#define getdigit_ascii(c, default)                                             \
+  (isdigit_ascii(c) ? ((int)((c) - '0')) : default)
+#define isspace_ascii(c) (((c) == ' ') || (((unsigned)(c) - '\t') < 5))
+#define toupper_ascii(c) ((((unsigned)(c) - 'a') < 26) ? ((c) & 0x5f) : (c))
+#define tolower_ascii(c) ((((unsigned)(c) - 'A') < 26) ? ((c) | 0x20) : (c))
+
+#if defined(_WIN32)
+#define PD_FALLTHROUGH                                                         \
+  do {                                                                         \
+  } while (0) /* fallthrough */
+#elif __has_attribute(__fallthrough__)
+#define PD_FALLTHROUGH __attribute__((__fallthrough__))
+#else
+#define PD_FALLTHROUGH                                                         \
+  do {                                                                         \
+  } while (0) /* fallthrough */
+#endif
diff --git a/pandas/pandas/_libs/include/pandas/skiplist.h b/pandas/pandas/_libs/include/pandas/skiplist.h
new file mode 100644
index 0000000000000..57e304b3a66c0
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/skiplist.h
@@ -0,0 +1,296 @@
+/*
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Flexibly-sized, index-able skiplist data structure for maintaining a sorted
+list of values
+
+Port of Wes McKinney's Cython version of Raymond Hettinger's original pure
+Python recipe (https://rhettinger.wordpress.com/2010/02/06/lost-knowledge/)
+*/
+
+#pragma once
+
+#include <math.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static inline float __skiplist_nanf(void) {
+  const union {
+    int __i;
+    float __f;
+  } __bint = {0x7fc00000UL};
+  return __bint.__f;
+}
+#define PANDAS_NAN ((double)__skiplist_nanf())
+
+static inline double Log2(double val) { return log(val) / log(2.); }
+
+typedef struct node_t node_t;
+
+struct node_t {
+  node_t **next;
+  int *width;
+  double value;
+  int is_nil;
+  int levels;
+  int ref_count;
+};
+
+typedef struct {
+  node_t *head;
+  node_t **tmp_chain;
+  int *tmp_steps;
+  int size;
+  int maxlevels;
+} skiplist_t;
+
+static inline double urand(void) {
+  return ((double)rand() + 1) / ((double)RAND_MAX + 2);
+}
+
+static inline int int_min(int a, int b) { return a < b ? a : b; }
+
+static inline node_t *node_init(double value, int levels) {
+  node_t *result;
+  result = (node_t *)malloc(sizeof(node_t));
+  if (result) {
+    result->value = value;
+    result->levels = levels;
+    result->is_nil = 0;
+    result->ref_count = 0;
+    result->next = (node_t **)malloc(levels * sizeof(node_t *));
+    result->width = (int *)malloc(levels * sizeof(int));
+    if (!(result->next && result->width) && (levels != 0)) {
+      free(result->next);
+      free(result->width);
+      free(result);
+      return NULL;
+    }
+  }
+  return result;
+}
+
+// do this ourselves
+static inline void node_incref(node_t *node) { ++(node->ref_count); }
+
+static inline void node_decref(node_t *node) { --(node->ref_count); }
+
+static void node_destroy(node_t *node) {
+  int i;
+  if (node) {
+    if (node->ref_count <= 1) {
+      for (i = 0; i < node->levels; ++i) {
+        node_destroy(node->next[i]);
+      }
+      free(node->next);
+      free(node->width);
+      // printf("Reference count was 1, freeing\n");
+      free(node);
+    } else {
+      node_decref(node);
+    }
+    // pretty sure that freeing the struct above will be enough
+  }
+}
+
+static inline void skiplist_destroy(skiplist_t *skp) {
+  if (skp) {
+    node_destroy(skp->head);
+    free(skp->tmp_steps);
+    free(skp->tmp_chain);
+    free(skp);
+  }
+}
+
+static inline skiplist_t *skiplist_init(int expected_size) {
+  skiplist_t *result;
+  node_t *NIL, *head;
+  int maxlevels, i;
+
+  maxlevels = 1 + Log2((double)expected_size);
+  result = (skiplist_t *)malloc(sizeof(skiplist_t));
+  if (!result) {
+    return NULL;
+  }
+  result->tmp_chain = (node_t **)malloc(maxlevels * sizeof(node_t *));
+  result->tmp_steps = (int *)malloc(maxlevels * sizeof(int));
+  result->maxlevels = maxlevels;
+  result->size = 0;
+
+  head = result->head = node_init(PANDAS_NAN, maxlevels);
+  NIL = node_init(0.0, 0);
+
+  if (!(result->tmp_chain && result->tmp_steps && result->head && NIL)) {
+    skiplist_destroy(result);
+    node_destroy(NIL);
+    return NULL;
+  }
+
+  node_incref(head);
+
+  NIL->is_nil = 1;
+
+  for (i = 0; i < maxlevels; ++i) {
+    head->next[i] = NIL;
+    head->width[i] = 1;
+    node_incref(NIL);
+  }
+
+  return result;
+}
+
+// 1 if left < right, 0 if left == right, -1 if left > right
+static inline int _node_cmp(node_t *node, double value) {
+  if (node->is_nil || node->value > value) {
+    return -1;
+  } else if (node->value < value) {
+    return 1;
+  } else {
+    return 0;
+  }
+}
+
+static inline double skiplist_get(skiplist_t *skp, int i, int *ret) {
+  node_t *node;
+  int level;
+
+  if (i < 0 || i >= skp->size) {
+    *ret = 0;
+    return 0;
+  }
+
+  node = skp->head;
+  ++i;
+  for (level = skp->maxlevels - 1; level >= 0; --level) {
+    while (node->width[level] <= i) {
+      i -= node->width[level];
+      node = node->next[level];
+    }
+  }
+
+  *ret = 1;
+  return node->value;
+}
+
+// Returns the lowest rank of all elements with value `value`, as opposed to the
+// highest rank returned by `skiplist_insert`.
+static inline int skiplist_min_rank(skiplist_t *skp, double value) {
+  node_t *node;
+  int level, rank = 0;
+
+  node = skp->head;
+  for (level = skp->maxlevels - 1; level >= 0; --level) {
+    while (_node_cmp(node->next[level], value) > 0) {
+      rank += node->width[level];
+      node = node->next[level];
+    }
+  }
+
+  return rank + 1;
+}
+
+// Returns the rank of the inserted element. When there are duplicates,
+// `rank` is the highest of the group, i.e. the 'max' method of
+// https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.rank.html
+static inline int skiplist_insert(skiplist_t *skp, double value) {
+  node_t *node, *prevnode, *newnode, *next_at_level;
+  int *steps_at_level;
+  int size, steps, level, rank = 0;
+  node_t **chain;
+
+  chain = skp->tmp_chain;
+
+  steps_at_level = skp->tmp_steps;
+  memset(steps_at_level, 0, skp->maxlevels * sizeof(int));
+
+  node = skp->head;
+
+  for (level = skp->maxlevels - 1; level >= 0; --level) {
+    next_at_level = node->next[level];
+    while (_node_cmp(next_at_level, value) >= 0) {
+      steps_at_level[level] += node->width[level];
+      rank += node->width[level];
+      node = next_at_level;
+      next_at_level = node->next[level];
+    }
+    chain[level] = node;
+  }
+
+  size = int_min(skp->maxlevels, 1 - ((int)Log2(urand())));
+
+  newnode = node_init(value, size);
+  if (!newnode) {
+    return -1;
+  }
+  steps = 0;
+
+  for (level = 0; level < size; ++level) {
+    prevnode = chain[level];
+    newnode->next[level] = prevnode->next[level];
+
+    prevnode->next[level] = newnode;
+    node_incref(newnode); // increment the reference count
+
+    newnode->width[level] = prevnode->width[level] - steps;
+    prevnode->width[level] = steps + 1;
+
+    steps += steps_at_level[level];
+  }
+
+  for (level = size; level < skp->maxlevels; ++level) {
+    chain[level]->width[level] += 1;
+  }
+
+  ++(skp->size);
+
+  return rank + 1;
+}
+
+static inline int skiplist_remove(skiplist_t *skp, double value) {
+  int level, size;
+  node_t *node, *prevnode, *tmpnode, *next_at_level;
+  node_t **chain;
+
+  chain = skp->tmp_chain;
+  node = skp->head;
+
+  for (level = skp->maxlevels - 1; level >= 0; --level) {
+    next_at_level = node->next[level];
+    while (_node_cmp(next_at_level, value) > 0) {
+      node = next_at_level;
+      next_at_level = node->next[level];
+    }
+    chain[level] = node;
+  }
+
+  if (value != chain[0]->next[0]->value) {
+    return 0;
+  }
+
+  size = chain[0]->next[0]->levels;
+
+  for (level = 0; level < size; ++level) {
+    prevnode = chain[level];
+
+    tmpnode = prevnode->next[level];
+
+    prevnode->width[level] += tmpnode->width[level] - 1;
+    prevnode->next[level] = tmpnode->next[level];
+
+    tmpnode->next[level] = NULL;
+    node_destroy(tmpnode); // decrement refcount or free
+  }
+
+  for (level = size; level < skp->maxlevels; ++level) {
+    --(chain[level]->width[level]);
+  }
+
+  --(skp->size);
+  return 1;
+}
diff --git a/pandas/pandas/_libs/include/pandas/vendored/klib/khash.h b/pandas/pandas/_libs/include/pandas/vendored/klib/khash.h
new file mode 100644
index 0000000000000..f072106e09596
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/vendored/klib/khash.h
@@ -0,0 +1,739 @@
+// Licence at LICENSES/KLIB_LICENSE
+
+/*
+  An example:
+
+#include "khash.h"
+KHASH_MAP_INIT_INT(32, char)
+int main() {
+        int ret, is_missing;
+        khiter_t k;
+        khash_t(32) *h = kh_init(32);
+        k = kh_put(32, h, 5, &ret);
+        if (!ret) kh_del(32, h, k);
+        kh_value(h, k) = 10;
+        k = kh_get(32, h, 10);
+        is_missing = (k == kh_end(h));
+        k = kh_get(32, h, 5);
+        kh_del(32, h, k);
+        for (k = kh_begin(h); k != kh_end(h); ++k)
+                if (kh_exist(h, k)) kh_value(h, k) = 1;
+        kh_destroy(32, h);
+        return 0;
+}
+*/
+
+/*
+  2011-09-16 (0.2.6):
+
+        * The capacity is a power of 2. This seems to dramatically improve the
+          speed for simple keys. Thank Zilong Tan for the suggestion. Reference:
+
+           - https://github.com/stefanocasazza/ULib
+           - https://nothings.org/computer/judy/
+
+        * Allow to optionally use linear probing which usually has better
+          performance for random input. Double hashing is still the default as
+  it is more robust to certain non-random input.
+
+        * Added Wang's integer hash function (not used by default). This hash
+          function is more robust to certain non-random input.
+
+  2011-02-14 (0.2.5):
+
+    * Allow to declare global functions.
+
+  2009-09-26 (0.2.4):
+
+    * Improve portability
+
+  2008-09-19 (0.2.3):
+
+        * Corrected the example
+        * Improved interfaces
+
+  2008-09-11 (0.2.2):
+
+        * Improved speed a little in kh_put()
+
+  2008-09-10 (0.2.1):
+
+        * Added kh_clear()
+        * Fixed a compiling error
+
+  2008-09-02 (0.2.0):
+
+        * Changed to token concatenation which increases flexibility.
+
+  2008-08-31 (0.1.2):
+
+        * Fixed a bug in kh_get(), which has not been tested previously.
+
+  2008-08-31 (0.1.1):
+
+        * Added destructor
+*/
+
+#ifndef __AC_KHASH_H
+#define __AC_KHASH_H
+
+/*!
+  @header
+
+  Generic hash table library.
+ */
+
+#define AC_VERSION_KHASH_H "0.2.6"
+
+#include <limits.h>
+#include <stdlib.h>
+#include <string.h>
+
+// hooks for memory allocator, C-runtime allocator used per default
+#ifndef KHASH_MALLOC
+#define KHASH_MALLOC malloc
+#endif
+
+#ifndef KHASH_REALLOC
+#define KHASH_REALLOC realloc
+#endif
+
+#ifndef KHASH_CALLOC
+#define KHASH_CALLOC calloc
+#endif
+
+#ifndef KHASH_FREE
+#define KHASH_FREE free
+#endif
+
+#if UINT_MAX == 0xffffffffu
+typedef unsigned int khuint32_t;
+typedef signed int khint32_t;
+#elif ULONG_MAX == 0xffffffffu
+typedef unsigned long khuint32_t;
+typedef signed long khint32_t;
+#endif
+
+#if ULONG_MAX == ULLONG_MAX
+typedef unsigned long khuint64_t;
+typedef signed long khint64_t;
+#else
+typedef unsigned long long khuint64_t;
+typedef signed long long khint64_t;
+#endif
+
+#if UINT_MAX == 0xffffu
+typedef unsigned int khuint16_t;
+typedef signed int khint16_t;
+#elif USHRT_MAX == 0xffffu
+typedef unsigned short khuint16_t;
+typedef signed short khint16_t;
+#endif
+
+#if UCHAR_MAX == 0xffu
+typedef unsigned char khuint8_t;
+typedef signed char khint8_t;
+#endif
+
+typedef double khfloat64_t;
+typedef float khfloat32_t;
+
+typedef khuint32_t khuint_t;
+typedef khuint_t khiter_t;
+
+#define __ac_isempty(flag, i) ((flag[i >> 5] >> (i & 0x1fU)) & 1)
+#define __ac_isdel(flag, i) (0)
+#define __ac_iseither(flag, i) __ac_isempty(flag, i)
+#define __ac_set_isdel_false(flag, i) (0)
+#define __ac_set_isempty_false(flag, i) (flag[i >> 5] &= ~(1ul << (i & 0x1fU)))
+#define __ac_set_isempty_true(flag, i) (flag[i >> 5] |= (1ul << (i & 0x1fU)))
+#define __ac_set_isboth_false(flag, i) __ac_set_isempty_false(flag, i)
+#define __ac_set_isdel_true(flag, i) ((void)0)
+
+// specializations of
+// https://github.com/aappleby/smhasher/blob/master/src/MurmurHash2.cpp
+static inline khuint32_t murmur2_32to32(khuint32_t k) {
+  const khuint32_t SEED = 0xc70f6907UL;
+  // 'm' and 'r' are mixing constants generated offline.
+  // They're not really 'magic', they just happen to work well.
+  const khuint32_t M_32 = 0x5bd1e995;
+  const int R_32 = 24;
+
+  // Initialize the hash to a 'random' value
+  khuint32_t h = SEED ^ 4;
+
+  // handle 4 bytes:
+  k *= M_32;
+  k ^= k >> R_32;
+  k *= M_32;
+
+  h *= M_32;
+  h ^= k;
+
+  // Do a few final mixes of the hash to ensure the "last few
+  // bytes" are well-incorporated. (Really needed here?)
+  h ^= h >> 13;
+  h *= M_32;
+  h ^= h >> 15;
+  return h;
+}
+
+// it is possible to have a special x64-version, which would need less
+// operations, but using 32bit version always has also some benefits:
+//    - one code for 32bit and 64bit builds
+//    - the same case for 32bit and 64bit builds
+//    - no performance difference could be measured compared to a possible
+//    x64-version
+
+static inline khuint32_t murmur2_32_32to32(khuint32_t k1, khuint32_t k2) {
+  const khuint32_t SEED = 0xc70f6907UL;
+  // 'm' and 'r' are mixing constants generated offline.
+  // They're not really 'magic', they just happen to work well.
+  const khuint32_t M_32 = 0x5bd1e995;
+  const int R_32 = 24;
+
+  // Initialize the hash to a 'random' value
+  khuint32_t h = SEED ^ 4;
+
+  // handle first 4 bytes:
+  k1 *= M_32;
+  k1 ^= k1 >> R_32;
+  k1 *= M_32;
+
+  h *= M_32;
+  h ^= k1;
+
+  // handle second 4 bytes:
+  k2 *= M_32;
+  k2 ^= k2 >> R_32;
+  k2 *= M_32;
+
+  h *= M_32;
+  h ^= k2;
+
+  // Do a few final mixes of the hash to ensure the "last few
+  // bytes" are well-incorporated.
+  h ^= h >> 13;
+  h *= M_32;
+  h ^= h >> 15;
+  return h;
+}
+
+static inline khuint32_t murmur2_64to32(khuint64_t k) {
+  khuint32_t k1 = (khuint32_t)k;
+  khuint32_t k2 = (khuint32_t)(k >> 32);
+
+  return murmur2_32_32to32(k1, k2);
+}
+
+#ifdef KHASH_LINEAR
+#define __ac_inc(k, m) 1
+#else
+#define __ac_inc(k, m) (murmur2_32to32(k) | 1) & (m)
+#endif
+
+#define __ac_fsize(m) ((m) < 32 ? 1 : (m) >> 5)
+
+#ifndef kroundup32
+#define kroundup32(x)                                                          \
+  (--(x), (x) |= (x) >> 1, (x) |= (x) >> 2, (x) |= (x) >> 4, (x) |= (x) >> 8,  \
+   (x) |= (x) >> 16, ++(x))
+#endif
+
+static const double __ac_HASH_UPPER = 0.77;
+
+#define KHASH_DECLARE(name, khkey_t, khval_t)                                  \
+  typedef struct {                                                             \
+    khuint_t n_buckets, size, n_occupied, upper_bound;                         \
+    khuint32_t *flags;                                                         \
+    khkey_t *keys;                                                             \
+    khval_t *vals;                                                             \
+  } kh_##name##_t;                                                             \
+  extern kh_##name##_t *kh_init_##name();                                      \
+  extern void kh_destroy_##name(kh_##name##_t *h);                             \
+  extern void kh_clear_##name(kh_##name##_t *h);                               \
+  extern khuint_t kh_get_##name(const kh_##name##_t *h, khkey_t key);          \
+  extern void kh_resize_##name(kh_##name##_t *h, khuint_t new_n_buckets);      \
+  extern khuint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret);      \
+  extern void kh_del_##name(kh_##name##_t *h, khuint_t x);
+
+#define KHASH_INIT2(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func,     \
+                    __hash_equal)                                              \
+  typedef struct {                                                             \
+    khuint_t n_buckets, size, n_occupied, upper_bound;                         \
+    khuint32_t *flags;                                                         \
+    khkey_t *keys;                                                             \
+    khval_t *vals;                                                             \
+  } kh_##name##_t;                                                             \
+  SCOPE kh_##name##_t *kh_init_##name(void) {                                  \
+    return (kh_##name##_t *)KHASH_CALLOC(1, sizeof(kh_##name##_t));            \
+  }                                                                            \
+  SCOPE void kh_destroy_##name(kh_##name##_t *h) {                             \
+    if (h) {                                                                   \
+      KHASH_FREE(h->keys);                                                     \
+      KHASH_FREE(h->flags);                                                    \
+      KHASH_FREE(h->vals);                                                     \
+      KHASH_FREE(h);                                                           \
+    }                                                                          \
+  }                                                                            \
+  SCOPE void kh_clear_##name(kh_##name##_t *h) {                               \
+    if (h && h->flags) {                                                       \
+      memset(h->flags, 0xaa, __ac_fsize(h->n_buckets) * sizeof(khuint32_t));   \
+      h->size = h->n_occupied = 0;                                             \
+    }                                                                          \
+  }                                                                            \
+  SCOPE khuint_t kh_get_##name(const kh_##name##_t *h, khkey_t key) {          \
+    if (h->n_buckets) {                                                        \
+      khuint_t inc, k, i, last, mask;                                          \
+      mask = h->n_buckets - 1;                                                 \
+      k = __hash_func(key);                                                    \
+      i = k & mask;                                                            \
+      inc = __ac_inc(k, mask);                                                 \
+      last = i; /* inc==1 for linear probing */                                \
+      while (!__ac_isempty(h->flags, i) &&                                     \
+             (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) {    \
+        i = (i + inc) & mask;                                                  \
+        if (i == last)                                                         \
+          return h->n_buckets;                                                 \
+      }                                                                        \
+      return __ac_iseither(h->flags, i) ? h->n_buckets : i;                    \
+    } else                                                                     \
+      return 0;                                                                \
+  }                                                                            \
+  SCOPE void kh_resize_##name(                                                 \
+      kh_##name##_t *h,                                                        \
+      khuint_t new_n_buckets) { /* This function uses 0.25*n_bucktes bytes of  \
+                                   working space instead of                    \
+                                   [sizeof(key_t+val_t)+.25]*n_buckets. */     \
+    khuint32_t *new_flags = 0;                                                 \
+    khuint_t j = 1;                                                            \
+    {                                                                          \
+      kroundup32(new_n_buckets);                                               \
+      if (new_n_buckets < 4)                                                   \
+        new_n_buckets = 4;                                                     \
+      if (h->size >= (khuint_t)(new_n_buckets * __ac_HASH_UPPER + 0.5))        \
+        j = 0; /* requested size is too small */                               \
+      else {   /* hash table size to be changed (shrink or expand); rehash */  \
+        new_flags = (khuint32_t *)KHASH_MALLOC(__ac_fsize(new_n_buckets) *     \
+                                               sizeof(khuint32_t));            \
+        memset(new_flags, 0xff,                                                \
+               __ac_fsize(new_n_buckets) * sizeof(khuint32_t));                \
+        if (h->n_buckets < new_n_buckets) { /* expand */                       \
+          h->keys = (khkey_t *)KHASH_REALLOC(h->keys,                          \
+                                             new_n_buckets * sizeof(khkey_t)); \
+          if (kh_is_map)                                                       \
+            h->vals = (khval_t *)KHASH_REALLOC(h->vals, new_n_buckets *        \
+                                                            sizeof(khval_t));  \
+        } /* otherwise shrink */                                               \
+      }                                                                        \
+    }                                                                          \
+    if (j) { /* rehashing is needed */                                         \
+      for (j = 0; j != h->n_buckets; ++j) {                                    \
+        if (__ac_iseither(h->flags, j) == 0) {                                 \
+          khkey_t key = h->keys[j];                                            \
+          khval_t val;                                                         \
+          khuint_t new_mask;                                                   \
+          new_mask = new_n_buckets - 1;                                        \
+          if (kh_is_map)                                                       \
+            val = h->vals[j];                                                  \
+          __ac_set_isempty_true(h->flags, j);                                  \
+          while (1) { /* kick-out process; sort of like in Cuckoo hashing */   \
+            khuint_t inc, k, i;                                                \
+            k = __hash_func(key);                                              \
+            i = k & new_mask;                                                  \
+            inc = __ac_inc(k, new_mask);                                       \
+            while (!__ac_isempty(new_flags, i))                                \
+              i = (i + inc) & new_mask;                                        \
+            __ac_set_isempty_false(new_flags, i);                              \
+            if (i < h->n_buckets &&                                            \
+                __ac_iseither(h->flags, i) ==                                  \
+                    0) { /* kick out the existing element */                   \
+              {                                                                \
+                khkey_t tmp = h->keys[i];                                      \
+                h->keys[i] = key;                                              \
+                key = tmp;                                                     \
+              }                                                                \
+              if (kh_is_map) {                                                 \
+                khval_t tmp = h->vals[i];                                      \
+                h->vals[i] = val;                                              \
+                val = tmp;                                                     \
+              }                                                                \
+              __ac_set_isempty_true(                                           \
+                  h->flags, i); /* mark it as deleted in the old hash table */ \
+            } else { /* write the element and jump out of the loop */          \
+              h->keys[i] = key;                                                \
+              if (kh_is_map)                                                   \
+                h->vals[i] = val;                                              \
+              break;                                                           \
+            }                                                                  \
+          }                                                                    \
+        }                                                                      \
+      }                                                                        \
+      if (h->n_buckets > new_n_buckets) { /* shrink the hash table */          \
+        h->keys = (khkey_t *)KHASH_REALLOC(h->keys,                            \
+                                           new_n_buckets * sizeof(khkey_t));   \
+        if (kh_is_map)                                                         \
+          h->vals = (khval_t *)KHASH_REALLOC(h->vals,                          \
+                                             new_n_buckets * sizeof(khval_t)); \
+      }                                                                        \
+      KHASH_FREE(h->flags); /* free the working space */                       \
+      h->flags = new_flags;                                                    \
+      h->n_buckets = new_n_buckets;                                            \
+      h->n_occupied = h->size;                                                 \
+      h->upper_bound = (khuint_t)(h->n_buckets * __ac_HASH_UPPER + 0.5);       \
+    }                                                                          \
+  }                                                                            \
+  SCOPE khuint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret) {      \
+    khuint_t x;                                                                \
+    if (h->n_occupied >= h->upper_bound) { /* update the hash table */         \
+      if (h->n_buckets > (h->size << 1))                                       \
+        kh_resize_##name(h, h->n_buckets - 1); /* clear "deleted" elements */  \
+      else                                                                     \
+        kh_resize_##name(h, h->n_buckets + 1); /* expand the hash table */     \
+    } /* TODO: to implement automatically shrinking; resize() already support  \
+         shrinking */                                                          \
+    {                                                                          \
+      khuint_t inc, k, i, site, last, mask = h->n_buckets - 1;                 \
+      x = site = h->n_buckets;                                                 \
+      k = __hash_func(key);                                                    \
+      i = k & mask;                                                            \
+      if (__ac_isempty(h->flags, i))                                           \
+        x = i; /* for speed up */                                              \
+      else {                                                                   \
+        inc = __ac_inc(k, mask);                                               \
+        last = i;                                                              \
+        while (!__ac_isempty(h->flags, i) &&                                   \
+               (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) {  \
+          if (__ac_isdel(h->flags, i))                                         \
+            site = i;                                                          \
+          i = (i + inc) & mask;                                                \
+          if (i == last) {                                                     \
+            x = site;                                                          \
+            break;                                                             \
+          }                                                                    \
+        }                                                                      \
+        if (x == h->n_buckets) {                                               \
+          if (__ac_isempty(h->flags, i) && site != h->n_buckets)               \
+            x = site;                                                          \
+          else                                                                 \
+            x = i;                                                             \
+        }                                                                      \
+      }                                                                        \
+    }                                                                          \
+    if (__ac_isempty(h->flags, x)) { /* not present at all */                  \
+      h->keys[x] = key;                                                        \
+      __ac_set_isboth_false(h->flags, x);                                      \
+      ++h->size;                                                               \
+      ++h->n_occupied;                                                         \
+      *ret = 1;                                                                \
+    } else if (__ac_isdel(h->flags, x)) { /* deleted */                        \
+      h->keys[x] = key;                                                        \
+      __ac_set_isboth_false(h->flags, x);                                      \
+      ++h->size;                                                               \
+      *ret = 2;                                                                \
+    } else                                                                     \
+      *ret = 0; /* Don't touch h->keys[x] if present and not deleted */        \
+    return x;                                                                  \
+  }                                                                            \
+  SCOPE void kh_del_##name(kh_##name##_t *h, khuint_t x) {                     \
+    if (x != h->n_buckets && !__ac_iseither(h->flags, x)) {                    \
+      __ac_set_isdel_true(h->flags, x);                                        \
+      --h->size;                                                               \
+    }                                                                          \
+  }
+
+#define KHASH_INIT(name, khkey_t, khval_t, kh_is_map, __hash_func,             \
+                   __hash_equal)                                               \
+  KHASH_INIT2(name, static inline, khkey_t, khval_t, kh_is_map, __hash_func,   \
+              __hash_equal)
+
+/* --- BEGIN OF HASH FUNCTIONS --- */
+
+/*! @function
+  @abstract     Integer hash function
+  @param  key   The integer [khuint32_t]
+  @return       The hash value [khuint_t]
+ */
+#define kh_int_hash_func(key) (khuint32_t)(key)
+/*! @function
+  @abstract     Integer comparison function
+ */
+#define kh_int_hash_equal(a, b) ((a) == (b))
+/*! @function
+  @abstract     64-bit integer hash function
+  @param  key   The integer [khuint64_t]
+  @return       The hash value [khuint_t]
+ */
+static inline khuint_t kh_int64_hash_func(khuint64_t key) {
+  return (khuint_t)((key) >> 33 ^ (key) ^ (key) << 11);
+}
+/*! @function
+  @abstract     64-bit integer comparison function
+ */
+#define kh_int64_hash_equal(a, b) ((a) == (b))
+
+/*! @function
+  @abstract     const char* hash function
+  @param  s     Pointer to a null terminated string
+  @return       The hash value
+ */
+static inline khuint_t __ac_X31_hash_string(const char *s) {
+  khuint_t h = *s;
+  if (h)
+    for (++s; *s; ++s)
+      h = (h << 5) - h + *s;
+  return h;
+}
+/*! @function
+  @abstract     Another interface to const char* hash function
+  @param  key   Pointer to a null terminated string [const char*]
+  @return       The hash value [khuint_t]
+ */
+#define kh_str_hash_func(key) __ac_X31_hash_string(key)
+/*! @function
+  @abstract     Const char* comparison function
+ */
+#define kh_str_hash_equal(a, b) (strcmp(a, b) == 0)
+
+static inline khuint_t __ac_Wang_hash(khuint_t key) {
+  key += ~(key << 15);
+  key ^= (key >> 10);
+  key += (key << 3);
+  key ^= (key >> 6);
+  key += ~(key << 11);
+  key ^= (key >> 16);
+  return key;
+}
+#define kh_int_hash_func2(k) __ac_Wang_hash((khuint_t)key)
+
+/* --- END OF HASH FUNCTIONS --- */
+
+/* Other convenient macros... */
+
+/*!
+  @abstract Type of the hash table.
+  @param  name  Name of the hash table [symbol]
+ */
+#define khash_t(name) kh_##name##_t
+
+/*! @function
+  @abstract     Initiate a hash table.
+  @param  name  Name of the hash table [symbol]
+  @return       Pointer to the hash table [khash_t(name)*]
+ */
+#define kh_init(name) kh_init_##name(void)
+
+/*! @function
+  @abstract     Destroy a hash table.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+ */
+#define kh_destroy(name, h) kh_destroy_##name(h)
+
+/*! @function
+  @abstract     Reset a hash table without deallocating memory.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+ */
+#define kh_clear(name, h) kh_clear_##name(h)
+
+/*! @function
+  @abstract     Resize a hash table.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  s     New size [khuint_t]
+ */
+#define kh_resize(name, h, s) kh_resize_##name(h, s)
+
+/*! @function
+  @abstract     Insert a key to the hash table.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  k     Key [type of keys]
+  @param  r     Extra return code: 0 if the key is present in the hash table;
+                1 if the bucket is empty (never used); 2 if the element in
+                                the bucket has been deleted [int*]
+  @return       Iterator to the inserted element [khuint_t]
+ */
+#define kh_put(name, h, k, r) kh_put_##name(h, k, r)
+
+/*! @function
+  @abstract     Retrieve a key from the hash table.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  k     Key [type of keys]
+  @return       Iterator to the found element, or kh_end(h) is the element is
+  absent [khuint_t]
+ */
+#define kh_get(name, h, k) kh_get_##name(h, k)
+
+/*! @function
+  @abstract     Remove a key from the hash table.
+  @param  name  Name of the hash table [symbol]
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  k     Iterator to the element to be deleted [khuint_t]
+ */
+#define kh_del(name, h, k) kh_del_##name(h, k)
+
+/*! @function
+  @abstract     Test whether a bucket contains data.
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  x     Iterator to the bucket [khuint_t]
+  @return       1 if containing data; 0 otherwise [int]
+ */
+#define kh_exist(h, x) (!__ac_iseither((h)->flags, (x)))
+
+/*! @function
+  @abstract     Get key given an iterator
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  x     Iterator to the bucket [khuint_t]
+  @return       Key [type of keys]
+ */
+#define kh_key(h, x) ((h)->keys[x])
+
+/*! @function
+  @abstract     Get value given an iterator
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @param  x     Iterator to the bucket [khuint_t]
+  @return       Value [type of values]
+  @discussion   For hash sets, calling this results in segfault.
+ */
+#define kh_val(h, x) ((h)->vals[x])
+
+/*! @function
+  @abstract     Alias of kh_val()
+ */
+#define kh_value(h, x) ((h)->vals[x])
+
+/*! @function
+  @abstract     Get the start iterator
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @return       The start iterator [khuint_t]
+ */
+#define kh_begin(h) (khuint_t)(0)
+
+/*! @function
+  @abstract     Get the end iterator
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @return       The end iterator [khuint_t]
+ */
+#define kh_end(h) ((h)->n_buckets)
+
+/*! @function
+  @abstract     Get the number of elements in the hash table
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @return       Number of elements in the hash table [khuint_t]
+ */
+#define kh_size(h) ((h)->size)
+
+/*! @function
+  @abstract     Get the number of buckets in the hash table
+  @param  h     Pointer to the hash table [khash_t(name)*]
+  @return       Number of buckets in the hash table [khuint_t]
+ */
+#define kh_n_buckets(h) ((h)->n_buckets)
+
+/* More convenient interfaces */
+
+/*! @function
+  @abstract     Instantiate a hash set containing integer keys
+  @param  name  Name of the hash table [symbol]
+ */
+#define KHASH_SET_INIT_INT(name)                                               \
+  KHASH_INIT(name, khint32_t, char, 0, kh_int_hash_func, kh_int_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing integer keys
+  @param  name  Name of the hash table [symbol]
+  @param  khval_t  Type of values [type]
+ */
+#define KHASH_MAP_INIT_INT(name, khval_t)                                      \
+  KHASH_INIT(name, khint32_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+#define KHASH_MAP_INIT_UINT(name, khval_t)                                     \
+  KHASH_INIT(name, khuint32_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing 64-bit integer keys
+  @param  name  Name of the hash table [symbol]
+ */
+#define KHASH_SET_INIT_UINT64(name)                                            \
+  KHASH_INIT(name, khuint64_t, char, 0, kh_int64_hash_func, kh_int64_hash_equal)
+
+#define KHASH_SET_INIT_INT64(name)                                             \
+  KHASH_INIT(name, khint64_t, char, 0, kh_int64_hash_func, kh_int64_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing 64-bit integer keys
+  @param  name  Name of the hash table [symbol]
+  @param  khval_t  Type of values [type]
+ */
+#define KHASH_MAP_INIT_UINT64(name, khval_t)                                   \
+  KHASH_INIT(name, khuint64_t, khval_t, 1, kh_int64_hash_func,                 \
+             kh_int64_hash_equal)
+
+#define KHASH_MAP_INIT_INT64(name, khval_t)                                    \
+  KHASH_INIT(name, khint64_t, khval_t, 1, kh_int64_hash_func,                  \
+             kh_int64_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing 16bit-integer keys
+  @param  name  Name of the hash table [symbol]
+  @param  khval_t  Type of values [type]
+ */
+#define KHASH_MAP_INIT_INT16(name, khval_t)                                    \
+  KHASH_INIT(name, khint16_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+#define KHASH_MAP_INIT_UINT16(name, khval_t)                                   \
+  KHASH_INIT(name, khuint16_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing 8bit-integer keys
+  @param  name  Name of the hash table [symbol]
+  @param  khval_t  Type of values [type]
+ */
+#define KHASH_MAP_INIT_INT8(name, khval_t)                                     \
+  KHASH_INIT(name, khint8_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+#define KHASH_MAP_INIT_UINT8(name, khval_t)                                    \
+  KHASH_INIT(name, khuint8_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
+
+typedef const char *kh_cstr_t;
+/*! @function
+  @abstract     Instantiate a hash map containing const char* keys
+  @param  name  Name of the hash table [symbol]
+ */
+#define KHASH_SET_INIT_STR(name)                                               \
+  KHASH_INIT(name, kh_cstr_t, char, 0, kh_str_hash_func, kh_str_hash_equal)
+
+/*! @function
+  @abstract     Instantiate a hash map containing const char* keys
+  @param  name  Name of the hash table [symbol]
+  @param  khval_t  Type of values [type]
+ */
+#define KHASH_MAP_INIT_STR(name, khval_t)                                      \
+  KHASH_INIT(name, kh_cstr_t, khval_t, 1, kh_str_hash_func, kh_str_hash_equal)
+
+#define kh_exist_str(h, k) (kh_exist(h, k))
+#define kh_exist_float64(h, k) (kh_exist(h, k))
+#define kh_exist_uint64(h, k) (kh_exist(h, k))
+#define kh_exist_int64(h, k) (kh_exist(h, k))
+#define kh_exist_float32(h, k) (kh_exist(h, k))
+#define kh_exist_int32(h, k) (kh_exist(h, k))
+#define kh_exist_uint32(h, k) (kh_exist(h, k))
+#define kh_exist_int16(h, k) (kh_exist(h, k))
+#define kh_exist_uint16(h, k) (kh_exist(h, k))
+#define kh_exist_int8(h, k) (kh_exist(h, k))
+#define kh_exist_uint8(h, k) (kh_exist(h, k))
+
+KHASH_MAP_INIT_STR(str, size_t)
+KHASH_MAP_INIT_INT(int32, size_t)
+KHASH_MAP_INIT_UINT(uint32, size_t)
+KHASH_MAP_INIT_INT64(int64, size_t)
+KHASH_MAP_INIT_UINT64(uint64, size_t)
+KHASH_MAP_INIT_INT16(int16, size_t)
+KHASH_MAP_INIT_UINT16(uint16, size_t)
+KHASH_MAP_INIT_INT8(int8, size_t)
+KHASH_MAP_INIT_UINT8(uint8, size_t)
+
+#endif /* __AC_KHASH_H */
diff --git a/pandas/pandas/_libs/include/pandas/vendored/klib/khash_python.h b/pandas/pandas/_libs/include/pandas/vendored/klib/khash_python.h
new file mode 100644
index 0000000000000..04e2d1e39ce2a
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/vendored/klib/khash_python.h
@@ -0,0 +1,410 @@
+// Licence at LICENSES/KLIB_LICENSE
+
+#pragma once
+
+#include <Python.h>
+
+#include <pymem.h>
+#include <string.h>
+
+typedef struct {
+  float real;
+  float imag;
+} khcomplex64_t;
+typedef struct {
+  double real;
+  double imag;
+} khcomplex128_t;
+
+static const int KHASH_TRACE_DOMAIN = 424242;
+static void *traced_malloc(size_t size) {
+  void *ptr = malloc(size);
+  if (ptr != NULL) {
+    PyTraceMalloc_Track(KHASH_TRACE_DOMAIN, (uintptr_t)ptr, size);
+  }
+  return ptr;
+}
+
+static void *traced_calloc(size_t num, size_t size) {
+  void *ptr = calloc(num, size);
+  if (ptr != NULL) {
+    PyTraceMalloc_Track(KHASH_TRACE_DOMAIN, (uintptr_t)ptr, num * size);
+  }
+  return ptr;
+}
+
+static void *traced_realloc(void *old_ptr, size_t size) {
+  PyTraceMalloc_Untrack(KHASH_TRACE_DOMAIN, (uintptr_t)old_ptr);
+  void *ptr = realloc(old_ptr, size);
+  if (ptr != NULL) {
+    PyTraceMalloc_Track(KHASH_TRACE_DOMAIN, (uintptr_t)ptr, size);
+  }
+  return ptr;
+}
+
+static void traced_free(void *ptr) {
+  if (ptr != NULL) {
+    PyTraceMalloc_Untrack(KHASH_TRACE_DOMAIN, (uintptr_t)ptr);
+  }
+  free(ptr);
+}
+
+#define KHASH_MALLOC traced_malloc
+#define KHASH_REALLOC traced_realloc
+#define KHASH_CALLOC traced_calloc
+#define KHASH_FREE traced_free
+#include "khash.h"
+
+// Previously we were using the built in cpython hash function for doubles
+// python 2.7 https://github.com/python/cpython/blob/2.7/Objects/object.c#L1021
+// python 3.5 https://github.com/python/cpython/blob/3.5/Python/pyhash.c#L85
+
+// The python 3 hash function has the invariant hash(x) == hash(int(x)) ==
+// hash(decimal(x)) and the size of hash may be different by platform / version
+// (long in py2, Py_ssize_t in py3). We don't need those invariants because
+// types will be cast before hashing, and if Py_ssize_t is 64 bits the
+// truncation causes collision issues.  Given all that, we use our own simple
+// hash, viewing the double bytes as an int64 and using khash's default hash for
+// 64 bit integers. GH 13436 showed that _Py_HashDouble doesn't work well with
+// khash GH 28303 showed, that the simple xoring-version isn't good enough See
+// GH 36729 for evaluation of the currently used murmur2-hash version An
+// interesting alternative to expensive murmur2-hash would be to change the
+// probing strategy and use e.g. the probing strategy from CPython's
+// implementation of dicts, which shines for smaller sizes but is more
+// predisposed to superlinear running times (see GH 36729 for comparison)
+
+static inline khuint64_t asuint64(double key) {
+  khuint64_t val;
+  memcpy(&val, &key, sizeof(double));
+  return val;
+}
+
+static inline khuint32_t asuint32(float key) {
+  khuint32_t val;
+  memcpy(&val, &key, sizeof(float));
+  return val;
+}
+
+#define ZERO_HASH 0
+#define NAN_HASH 0
+
+static inline khuint32_t kh_float64_hash_func(double val) {
+  // 0.0 and -0.0 should have the same hash:
+  if (val == 0.0) {
+    return ZERO_HASH;
+  }
+  // all nans should have the same hash:
+  if (val != val) {
+    return NAN_HASH;
+  }
+  khuint64_t as_int = asuint64(val);
+  return murmur2_64to32(as_int);
+}
+
+static inline khuint32_t kh_float32_hash_func(float val) {
+  // 0.0 and -0.0 should have the same hash:
+  if (val == 0.0f) {
+    return ZERO_HASH;
+  }
+  // all nans should have the same hash:
+  if (val != val) {
+    return NAN_HASH;
+  }
+  khuint32_t as_int = asuint32(val);
+  return murmur2_32to32(as_int);
+}
+
+#define kh_floats_hash_equal(a, b) ((a) == (b) || ((b) != (b) && (a) != (a)))
+
+#define KHASH_MAP_INIT_FLOAT64(name, khval_t)                                  \
+  KHASH_INIT(name, khfloat64_t, khval_t, 1, kh_float64_hash_func,              \
+             kh_floats_hash_equal)
+
+KHASH_MAP_INIT_FLOAT64(float64, size_t)
+
+#define KHASH_MAP_INIT_FLOAT32(name, khval_t)                                  \
+  KHASH_INIT(name, khfloat32_t, khval_t, 1, kh_float32_hash_func,              \
+             kh_floats_hash_equal)
+
+KHASH_MAP_INIT_FLOAT32(float32, size_t)
+
+static inline khint32_t kh_complex128_hash_func(khcomplex128_t val) {
+  return kh_float64_hash_func(val.real) ^ kh_float64_hash_func(val.imag);
+}
+static inline khint32_t kh_complex64_hash_func(khcomplex64_t val) {
+  return kh_float32_hash_func(val.real) ^ kh_float32_hash_func(val.imag);
+}
+
+#define kh_complex_hash_equal(a, b)                                            \
+  (kh_floats_hash_equal(a.real, b.real) && kh_floats_hash_equal(a.imag, b.imag))
+
+#define KHASH_MAP_INIT_COMPLEX64(name, khval_t)                                \
+  KHASH_INIT(name, khcomplex64_t, khval_t, 1, kh_complex64_hash_func,          \
+             kh_complex_hash_equal)
+
+KHASH_MAP_INIT_COMPLEX64(complex64, size_t)
+
+#define KHASH_MAP_INIT_COMPLEX128(name, khval_t)                               \
+  KHASH_INIT(name, khcomplex128_t, khval_t, 1, kh_complex128_hash_func,        \
+             kh_complex_hash_equal)
+
+KHASH_MAP_INIT_COMPLEX128(complex128, size_t)
+
+#define kh_exist_complex64(h, k) (kh_exist(h, k))
+#define kh_exist_complex128(h, k) (kh_exist(h, k))
+
+// NaN-floats should be in the same equivalency class, see GH 22119
+static inline int floatobject_cmp(PyFloatObject *a, PyFloatObject *b) {
+  return (isnan(PyFloat_AS_DOUBLE(a)) && isnan(PyFloat_AS_DOUBLE(b))) ||
+         (PyFloat_AS_DOUBLE(a) == PyFloat_AS_DOUBLE(b));
+}
+
+// NaNs should be in the same equivalency class, see GH 41836
+// PyObject_RichCompareBool for complexobjects has a different behavior
+// needs to be replaced
+static inline int complexobject_cmp(PyComplexObject *a, PyComplexObject *b) {
+  return (isnan(a->cval.real) && isnan(b->cval.real) && isnan(a->cval.imag) &&
+          isnan(b->cval.imag)) ||
+         (isnan(a->cval.real) && isnan(b->cval.real) &&
+          a->cval.imag == b->cval.imag) ||
+         (a->cval.real == b->cval.real && isnan(a->cval.imag) &&
+          isnan(b->cval.imag)) ||
+         (a->cval.real == b->cval.real && a->cval.imag == b->cval.imag);
+}
+
+static inline int pyobject_cmp(PyObject *a, PyObject *b);
+
+// replacing PyObject_RichCompareBool (NaN!=NaN) with pyobject_cmp (NaN==NaN),
+// which treats NaNs as equivalent
+// see GH 41836
+static inline int tupleobject_cmp(PyTupleObject *a, PyTupleObject *b) {
+  Py_ssize_t i;
+
+  if (Py_SIZE(a) != Py_SIZE(b)) {
+    return 0;
+  }
+
+  for (i = 0; i < Py_SIZE(a); ++i) {
+    if (!pyobject_cmp(PyTuple_GET_ITEM(a, i), PyTuple_GET_ITEM(b, i))) {
+      return 0;
+    }
+  }
+  return 1;
+}
+
+static inline int pyobject_cmp(PyObject *a, PyObject *b) {
+  if (a == b) {
+    return 1;
+  }
+  if (Py_TYPE(a) == Py_TYPE(b)) {
+    // special handling for some built-in types which could have NaNs
+    // as we would like to have them equivalent, but the usual
+    // PyObject_RichCompareBool would return False
+    if (PyFloat_CheckExact(a)) {
+      return floatobject_cmp((PyFloatObject *)a, (PyFloatObject *)b);
+    }
+    if (PyComplex_CheckExact(a)) {
+      return complexobject_cmp((PyComplexObject *)a, (PyComplexObject *)b);
+    }
+    if (PyTuple_Check(a)) {
+      // compare tuple subclasses as builtin tuples
+      return tupleobject_cmp((PyTupleObject *)a, (PyTupleObject *)b);
+    }
+    // frozenset isn't yet supported
+  }
+
+  int result = PyObject_RichCompareBool(a, b, Py_EQ);
+  if (result < 0) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+static inline Py_hash_t _Pandas_HashDouble(double val) {
+  // Since Python3.10, nan is no longer has hash 0
+  if (isnan(val)) {
+    return 0;
+  }
+#if PY_VERSION_HEX < 0x030A0000
+  return _Py_HashDouble(val);
+#else
+  return _Py_HashDouble(NULL, val);
+#endif
+}
+
+static inline Py_hash_t floatobject_hash(PyFloatObject *key) {
+  return _Pandas_HashDouble(PyFloat_AS_DOUBLE(key));
+}
+
+#define _PandasHASH_IMAG 1000003UL
+
+// replaces _Py_HashDouble with _Pandas_HashDouble
+static inline Py_hash_t complexobject_hash(PyComplexObject *key) {
+  Py_uhash_t realhash = (Py_uhash_t)_Pandas_HashDouble(key->cval.real);
+  Py_uhash_t imaghash = (Py_uhash_t)_Pandas_HashDouble(key->cval.imag);
+  if (realhash == (Py_uhash_t)-1 || imaghash == (Py_uhash_t)-1) {
+    return -1;
+  }
+  Py_uhash_t combined = realhash + _PandasHASH_IMAG * imaghash;
+  if (combined == (Py_uhash_t)-1) {
+    return -2;
+  }
+  return (Py_hash_t)combined;
+}
+
+static inline khuint32_t kh_python_hash_func(PyObject *key);
+
+// we could use any hashing algorithm, this is the original CPython's for tuples
+
+#if SIZEOF_PY_UHASH_T > 4
+#define _PandasHASH_XXPRIME_1 ((Py_uhash_t)11400714785074694791ULL)
+#define _PandasHASH_XXPRIME_2 ((Py_uhash_t)14029467366897019727ULL)
+#define _PandasHASH_XXPRIME_5 ((Py_uhash_t)2870177450012600261ULL)
+#define _PandasHASH_XXROTATE(x)                                                \
+  ((x << 31) | (x >> 33)) /* Rotate left 31 bits */
+#else
+#define _PandasHASH_XXPRIME_1 ((Py_uhash_t)2654435761UL)
+#define _PandasHASH_XXPRIME_2 ((Py_uhash_t)2246822519UL)
+#define _PandasHASH_XXPRIME_5 ((Py_uhash_t)374761393UL)
+#define _PandasHASH_XXROTATE(x)                                                \
+  ((x << 13) | (x >> 19)) /* Rotate left 13 bits */
+#endif
+
+static inline Py_hash_t tupleobject_hash(PyTupleObject *key) {
+  Py_ssize_t i, len = Py_SIZE(key);
+  PyObject **item = key->ob_item;
+
+  Py_uhash_t acc = _PandasHASH_XXPRIME_5;
+  for (i = 0; i < len; i++) {
+    Py_uhash_t lane = kh_python_hash_func(item[i]);
+    if (lane == (Py_uhash_t)-1) {
+      return -1;
+    }
+    acc += lane * _PandasHASH_XXPRIME_2;
+    acc = _PandasHASH_XXROTATE(acc);
+    acc *= _PandasHASH_XXPRIME_1;
+  }
+
+  /* Add input length, mangled to keep the historical value of hash(()). */
+  acc += len ^ (_PandasHASH_XXPRIME_5 ^ 3527539UL);
+
+  if (acc == (Py_uhash_t)-1) {
+    return 1546275796;
+  }
+  return acc;
+}
+
+static inline khuint32_t kh_python_hash_func(PyObject *key) {
+  Py_hash_t hash;
+  // For PyObject_Hash holds:
+  //    hash(0.0) == 0 == hash(-0.0)
+  //    yet for different nan-objects different hash-values
+  //    are possible
+  if (PyFloat_CheckExact(key)) {
+    // we cannot use kh_float64_hash_func
+    // because float(k) == k holds for any int-object k
+    // and kh_float64_hash_func doesn't respect it
+    hash = floatobject_hash((PyFloatObject *)key);
+  } else if (PyComplex_CheckExact(key)) {
+    // we cannot use kh_complex128_hash_func
+    // because complex(k,0) == k holds for any int-object k
+    // and kh_complex128_hash_func doesn't respect it
+    hash = complexobject_hash((PyComplexObject *)key);
+  } else if (PyTuple_Check(key)) {
+    // hash tuple subclasses as builtin tuples
+    hash = tupleobject_hash((PyTupleObject *)key);
+  } else {
+    hash = PyObject_Hash(key);
+  }
+
+  if (hash == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+#if SIZEOF_PY_HASH_T == 4
+  // it is already 32bit value
+  return hash;
+#else
+  // for 64bit builds,
+  // we need information of the upper 32bits as well
+  // see GH 37615
+  khuint64_t as_uint = (khuint64_t)hash;
+  // uints avoid undefined behavior of signed ints
+  return (as_uint >> 32) ^ as_uint;
+#endif
+}
+
+#define kh_python_hash_equal(a, b) (pyobject_cmp(a, b))
+
+// Python object
+
+typedef PyObject *kh_pyobject_t;
+
+#define KHASH_MAP_INIT_PYOBJECT(name, khval_t)                                 \
+  KHASH_INIT(name, kh_pyobject_t, khval_t, 1, kh_python_hash_func,             \
+             kh_python_hash_equal)
+
+KHASH_MAP_INIT_PYOBJECT(pymap, Py_ssize_t)
+
+#define KHASH_SET_INIT_PYOBJECT(name)                                          \
+  KHASH_INIT(name, kh_pyobject_t, char, 0, kh_python_hash_func,                \
+             kh_python_hash_equal)
+
+KHASH_SET_INIT_PYOBJECT(pyset)
+
+#define kh_exist_pymap(h, k) (kh_exist(h, k))
+#define kh_exist_pyset(h, k) (kh_exist(h, k))
+
+KHASH_MAP_INIT_STR(strbox, kh_pyobject_t)
+
+typedef struct {
+  kh_str_t *table;
+  int starts[256];
+} kh_str_starts_t;
+
+typedef kh_str_starts_t *p_kh_str_starts_t;
+
+static inline p_kh_str_starts_t kh_init_str_starts(void) {
+  kh_str_starts_t *result =
+      (kh_str_starts_t *)KHASH_CALLOC(1, sizeof(kh_str_starts_t));
+  result->table = kh_init_str();
+  return result;
+}
+
+static inline khuint_t kh_put_str_starts_item(kh_str_starts_t *table, char *key,
+                                              int *ret) {
+  khuint_t result = kh_put_str(table->table, key, ret);
+  if (*ret != 0) {
+    table->starts[(unsigned char)key[0]] = 1;
+  }
+  return result;
+}
+
+static inline khuint_t kh_get_str_starts_item(const kh_str_starts_t *table,
+                                              const char *key) {
+  unsigned char ch = *key;
+  if (table->starts[ch]) {
+    if (ch == '\0' || kh_get_str(table->table, key) != table->table->n_buckets)
+      return 1;
+  }
+  return 0;
+}
+
+static inline void kh_destroy_str_starts(kh_str_starts_t *table) {
+  kh_destroy_str(table->table);
+  KHASH_FREE(table);
+}
+
+static inline void kh_resize_str_starts(kh_str_starts_t *table, khuint_t val) {
+  kh_resize_str(table->table, val);
+}
+
+// utility function: given the number of elements
+// returns number of necessary buckets
+static inline khuint_t kh_needed_n_buckets(khuint_t n_elements) {
+  khuint_t candidate = n_elements;
+  kroundup32(candidate);
+  khuint_t upper_bound = (khuint_t)(candidate * __ac_HASH_UPPER + 0.5);
+  return (upper_bound < n_elements) ? 2 * candidate : candidate;
+}
diff --git a/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h b/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
new file mode 100644
index 0000000000000..e4e90a7ea24cf
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
@@ -0,0 +1,106 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Copyright (c) 2005-2011, NumPy Developers
+All rights reserved.
+
+This file is derived from NumPy 1.7. See NUMPY_LICENSE.txt
+
+*/
+
+#pragma once
+
+#ifndef NPY_NO_DEPRECATED_API
+#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
+#endif // NPY_NO_DEPRECATED_API
+
+#include <numpy/ndarraytypes.h>
+
+typedef struct {
+  npy_int64 days;
+  npy_int32 hrs, min, sec, ms, us, ns, seconds, microseconds, nanoseconds;
+} pandas_timedeltastruct;
+
+static const npy_datetimestruct _AS_MIN_DTS = {1969, 12,     31,     23,    59,
+                                               50,   776627, 963145, 224193};
+static const npy_datetimestruct _FS_MIN_DTS = {1969, 12,     31,     21,    26,
+                                               16,   627963, 145224, 193000};
+static const npy_datetimestruct _PS_MIN_DTS = {1969, 9,      16,     5, 57,
+                                               7,    963145, 224193, 0};
+static const npy_datetimestruct _NS_MIN_DTS = {1677, 9,      21,     0, 12,
+                                               43,   145224, 193000, 0};
+static const npy_datetimestruct _US_MIN_DTS = {-290308, 12,     21, 19, 59,
+                                               05,      224193, 0,  0};
+static const npy_datetimestruct _MS_MIN_DTS = {-292275055, 5,      16, 16, 47,
+                                               4,          193000, 0,  0};
+static const npy_datetimestruct _S_MIN_DTS = {
+    -292277022657, 1, 27, 8, 29, 53, 0, 0, 0};
+static const npy_datetimestruct _M_MIN_DTS = {
+    -17536621475646, 5, 4, 5, 53, 0, 0, 0, 0};
+
+static const npy_datetimestruct _AS_MAX_DTS = {1970, 1,      1,     0,     0,
+                                               9,    223372, 36854, 775807};
+static const npy_datetimestruct _FS_MAX_DTS = {1970, 1,      1,      2,     33,
+                                               43,   372036, 854775, 807000};
+static const npy_datetimestruct _PS_MAX_DTS = {1970, 4,     17,     18, 2,
+                                               52,   36854, 775807, 0};
+static const npy_datetimestruct _NS_MAX_DTS = {2262, 4,      11,     23, 47,
+                                               16,   854775, 807000, 0};
+static const npy_datetimestruct _US_MAX_DTS = {294247, 1,      10, 4, 0,
+                                               54,     775807, 0,  0};
+static const npy_datetimestruct _MS_MAX_DTS = {292278994, 8,      17, 7, 12,
+                                               55,        807000, 0,  0};
+static const npy_datetimestruct _S_MAX_DTS = {
+    292277026596, 12, 4, 15, 30, 7, 0, 0, 0};
+static const npy_datetimestruct _M_MAX_DTS = {
+    17536621479585, 8, 30, 18, 7, 0, 0, 0, 0};
+
+// stuff pandas needs
+// ----------------------------------------------------------------------------
+
+PyObject *extract_utc_offset(PyObject *obj);
+
+npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
+                                            const npy_datetimestruct *dts);
+
+void pandas_datetime_to_datetimestruct(npy_datetime val, NPY_DATETIMEUNIT fr,
+                                       npy_datetimestruct *result);
+
+void pandas_timedelta_to_timedeltastruct(npy_timedelta val, NPY_DATETIMEUNIT fr,
+                                         pandas_timedeltastruct *result);
+
+extern const int days_per_month_table[2][12];
+
+// stuff numpy-derived code needs in header
+// ----------------------------------------------------------------------------
+
+int is_leapyear(npy_int64 year);
+
+/*
+ * Calculates the days offset from the 1970 epoch.
+ */
+npy_int64 get_datetimestruct_days(const npy_datetimestruct *dts);
+
+/*
+ * Compares two npy_datetimestruct objects chronologically
+ */
+int cmp_npy_datetimestruct(const npy_datetimestruct *a,
+                           const npy_datetimestruct *b);
+
+/*
+ * Adjusts a datetimestruct based on a minutes offset. Assumes
+ * the current values are valid.
+ */
+void add_minutes_to_datetimestruct(npy_datetimestruct *dts, int minutes);
+
+/*
+ * This function returns the DateTimeMetaData
+ * contained within the provided datetime dtype.
+ */
+PyArray_DatetimeMetaData get_datetime_metadata_from_dtype(PyArray_Descr *dtype);
diff --git a/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime_strings.h b/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime_strings.h
new file mode 100644
index 0000000000000..75e69f30ada1e
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime_strings.h
@@ -0,0 +1,99 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Written by Mark Wiebe (mwwiebe@gmail.com)
+Copyright (c) 2011 by Enthought, Inc.
+
+Copyright (c) 2005-2011, NumPy Developers
+All rights reserved.
+
+See NUMPY_LICENSE.txt for the license.
+
+This file implements string parsing and creation for NumPy datetime.
+
+*/
+
+#pragma once
+
+#ifndef NPY_NO_DEPRECATED_API
+#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
+#endif // NPY_NO_DEPRECATED_API
+
+/* 'format_requirement' can be one of three values:
+ *      * PARTIAL_MATCH : Only require a partial match with 'format'.
+ *           For example, if the string is '2020-01-01 05:00:00' and
+ *           'format' is '%Y-%m-%d', then parse '2020-01-01';
+ *      * EXACT_MATCH : require an exact match with 'format'. If the
+ *           string is '2020-01-01', then the only format which will
+ *           be able to parse it without error is '%Y-%m-%d';
+ *      * INFER_FORMAT: parse without comparing 'format' (i.e. infer it).
+ */
+typedef enum { PARTIAL_MATCH, EXACT_MATCH, INFER_FORMAT } FormatRequirement;
+
+/*
+ * Parses (almost) standard ISO 8601 date strings. The differences are:
+ *
+ * + The date "20100312" is parsed as the year 20100312, not as
+ *   equivalent to "2010-03-12". The '-' in the dates are not optional.
+ * + Only seconds may have a decimal point, with up to 18 digits after it
+ *   (maximum attoseconds precision).
+ * + Either a 'T' as in ISO 8601 or a ' ' may be used to separate
+ *   the date and the time. Both are treated equivalently.
+ * + Doesn't (yet) handle the "YYYY-DDD" or "YYYY-Www" formats.
+ * + Doesn't handle leap seconds (seconds value has 60 in these cases).
+ * + Doesn't handle 24:00:00 as synonym for midnight (00:00:00) tomorrow
+ * + Accepts special values "NaT" (not a time), "Today", (current
+ *   day according to local time) and "Now" (current time in UTC).
+ *
+ * 'str' must be a NULL-terminated string, and 'len' must be its length.
+ *
+ * 'out' gets filled with the parsed date-time.
+ * 'out_local' gets whether returned value contains timezone. 0 for UTC, 1 for
+ * local time. 'out_tzoffset' gets set to timezone offset by minutes if the
+ * parsed time was in local time, to 0 otherwise. The values 'now' and 'today'
+ * don't get counted as local, and neither do UTC +/-#### timezone offsets,
+ * because they aren't using the computer's local timezone offset.
+ *
+ * Returns 0 on success, -1 on failure.
+ */
+int parse_iso_8601_datetime(const char *str, int len, int want_exc,
+                            npy_datetimestruct *out,
+                            NPY_DATETIMEUNIT *out_bestunit, int *out_local,
+                            int *out_tzoffset, const char *format,
+                            int format_len,
+                            FormatRequirement format_requirement);
+
+/*
+ * Provides a string length to use for converting datetime
+ * objects with the given local and unit settings.
+ */
+int get_datetime_iso_8601_strlen(int local, NPY_DATETIMEUNIT base);
+
+/*
+ * Converts an npy_datetimestruct to an (almost) ISO 8601
+ * NULL-terminated string using timezone Z (UTC).
+ *
+ * 'base' restricts the output to that unit. Set 'base' to
+ * -1 to auto-detect a base after which all the values are zero.
+ *
+ *  Returns 0 on success, -1 on failure (for example if the output
+ *  string was too short).
+ */
+int make_iso_8601_datetime(npy_datetimestruct *dts, char *outstr, size_t outlen,
+                           int utc, NPY_DATETIMEUNIT base);
+
+/*
+ * Converts an pandas_timedeltastruct to an ISO 8601 string.
+ *
+ * Mutates outlen to provide size of (non-NULL terminated) string.
+ *
+ * Currently has no error handling
+ */
+int make_iso_8601_timedelta(pandas_timedeltastruct *tds, char *outstr,
+                            size_t *outlen);
diff --git a/pandas/pandas/_libs/include/pandas/vendored/ujson/lib/ultrajson.h b/pandas/pandas/_libs/include/pandas/vendored/ujson/lib/ultrajson.h
new file mode 100644
index 0000000000000..51fdbc50bba57
--- /dev/null
+++ b/pandas/pandas/_libs/include/pandas/vendored/ujson/lib/ultrajson.h
@@ -0,0 +1,326 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+    * Neither the name of the ESN Social Software AB nor the
+      names of its contributors may be used to endorse or promote products
+      derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
+GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
+THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+ * Copyright (c) 1988-1993 The Regents of the University of California.
+ * Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+/*
+Ultra fast JSON encoder and decoder
+Developed by Jonas Tarnstrom (jonas@esn.me).
+
+Encoder notes:
+------------------
+
+:: Cyclic references ::
+Cyclic referenced objects are not detected.
+Set JSONObjectEncoder.recursionMax to suitable value or make sure input object
+tree doesn't have cyclic references.
+
+*/
+
+#pragma once
+
+#include "pandas/portable.h"
+#include <stdio.h>
+#include <wchar.h>
+
+// Don't output any extra whitespaces when encoding
+#define JSON_NO_EXTRA_WHITESPACE
+
+// Max decimals to encode double floating point numbers with
+#ifndef JSON_DOUBLE_MAX_DECIMALS
+#define JSON_DOUBLE_MAX_DECIMALS 15
+#endif
+
+// Max recursion depth, default for encoder
+#ifndef JSON_MAX_RECURSION_DEPTH
+#define JSON_MAX_RECURSION_DEPTH 1024
+#endif
+
+// Max recursion depth, default for decoder
+#ifndef JSON_MAX_OBJECT_DEPTH
+#define JSON_MAX_OBJECT_DEPTH 1024
+#endif
+
+/*
+Dictates and limits how much stack space for buffers UltraJSON will use before
+resorting to provided heap functions */
+#ifndef JSON_MAX_STACK_BUFFER_SIZE
+#define JSON_MAX_STACK_BUFFER_SIZE 131072
+#endif
+
+#ifdef _WIN32
+
+typedef __int64 JSINT64;
+typedef unsigned __int64 JSUINT64;
+
+typedef __int32 JSINT32;
+typedef unsigned __int32 JSUINT32;
+typedef unsigned __int8 JSUINT8;
+typedef unsigned __int16 JSUTF16;
+typedef unsigned __int32 JSUTF32;
+typedef __int64 JSLONG;
+
+#define EXPORTFUNCTION __declspec(dllexport)
+
+#define FASTCALL_MSVC __fastcall
+
+#define INLINE_PREFIX static __inline
+
+#else
+
+#include <stdint.h>
+typedef int64_t JSINT64;
+typedef uint64_t JSUINT64;
+
+typedef int32_t JSINT32;
+typedef uint32_t JSUINT32;
+
+#define FASTCALL_MSVC
+
+#define INLINE_PREFIX static inline
+
+typedef uint8_t JSUINT8;
+typedef uint16_t JSUTF16;
+typedef uint32_t JSUTF32;
+
+typedef int64_t JSLONG;
+
+#define EXPORTFUNCTION
+#endif
+
+#if !(defined(__LITTLE_ENDIAN__) || defined(__BIG_ENDIAN__))
+
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+#define __LITTLE_ENDIAN__
+#else
+
+#if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+#define __BIG_ENDIAN__
+#endif
+
+#endif
+
+#endif
+
+#if !defined(__LITTLE_ENDIAN__) && !defined(__BIG_ENDIAN__)
+#error "Endianness not supported"
+#endif
+
+enum JSTYPES {
+  JT_NULL,    // NULL
+  JT_TRUE,    // boolean true
+  JT_FALSE,   // boolean false
+  JT_INT,     // (JSINT32 (signed 32-bit))
+  JT_LONG,    // (JSINT64 (signed 64-bit))
+  JT_DOUBLE,  // (double)
+  JT_BIGNUM,  // integer larger than sys.maxsize
+  JT_UTF8,    // (char 8-bit)
+  JT_ARRAY,   // Array structure
+  JT_OBJECT,  // Key/Value structure
+  JT_INVALID, // Internal, do not return nor expect
+  JT_POS_INF, // Positive infinity
+  JT_NEG_INF, // Negative infinity
+};
+
+typedef void *JSOBJ;
+typedef void *JSITER;
+
+typedef struct __JSONTypeContext {
+  int type;
+  void *encoder;
+  void *prv;
+} JSONTypeContext;
+
+/*
+Function pointer declarations, suitable for implementing UltraJSON */
+typedef void (*JSPFN_ITERBEGIN)(JSOBJ obj, JSONTypeContext *tc);
+typedef int (*JSPFN_ITERNEXT)(JSOBJ obj, JSONTypeContext *tc);
+typedef void (*JSPFN_ITEREND)(JSOBJ obj, JSONTypeContext *tc);
+typedef JSOBJ (*JSPFN_ITERGETVALUE)(JSOBJ obj, JSONTypeContext *tc);
+typedef const char *(*JSPFN_ITERGETNAME)(JSOBJ obj, JSONTypeContext *tc,
+                                         size_t *outLen);
+typedef void *(*JSPFN_MALLOC)(size_t size);
+typedef void (*JSPFN_FREE)(void *pptr);
+typedef void *(*JSPFN_REALLOC)(void *base, size_t size);
+
+typedef struct __JSONObjectEncoder {
+  void (*beginTypeContext)(JSOBJ obj, JSONTypeContext *tc);
+  void (*endTypeContext)(JSOBJ obj, JSONTypeContext *tc);
+  const char *(*getStringValue)(JSOBJ obj, JSONTypeContext *tc,
+                                size_t *_outLen);
+  JSINT64 (*getLongValue)(JSOBJ obj, JSONTypeContext *tc);
+  JSINT32 (*getIntValue)(JSOBJ obj, JSONTypeContext *tc);
+  double (*getDoubleValue)(JSOBJ obj, JSONTypeContext *tc);
+  const char *(*getBigNumStringValue)(JSOBJ obj, JSONTypeContext *tc,
+                                      size_t *_outLen);
+
+  /*
+  Begin iteration of an iterable object (JS_ARRAY or JS_OBJECT)
+  Implementer should setup iteration state in ti->prv
+  */
+  JSPFN_ITERBEGIN iterBegin;
+
+  /*
+  Retrieve next object in an iteration. Should return 0 to indicate iteration
+  has reached end or 1 if there are more items. Implementer is responsible for
+  keeping state of the iteration. Use ti->prv fields for this
+  */
+  JSPFN_ITERNEXT iterNext;
+
+  /*
+  Ends the iteration of an iterable object.
+  Any iteration state stored in ti->prv can be freed here
+  */
+  JSPFN_ITEREND iterEnd;
+
+  /*
+  Returns a reference to the value object of an iterator
+  The is responsible for the life-cycle of the returned string. Use
+  iterNext/iterEnd and ti->prv to keep track of current object
+  */
+  JSPFN_ITERGETVALUE iterGetValue;
+
+  /*
+  Return name of iterator.
+  The is responsible for the life-cycle of the returned string. Use
+  iterNext/iterEnd and ti->prv to keep track of current object
+  */
+  JSPFN_ITERGETNAME iterGetName;
+
+  /*
+  Release a value as indicated by setting ti->release = 1 in the previous
+  getValue call. The ti->prv array should contain the necessary context to
+  release the value
+  */
+  void (*releaseObject)(JSOBJ obj);
+
+  /* Library functions
+  Set to NULL to use STDLIB malloc,realloc,free */
+  JSPFN_MALLOC malloc;
+  JSPFN_REALLOC realloc;
+  JSPFN_FREE free;
+
+  /*
+  Configuration for max recursion, set to 0 to use default (see
+  JSON_MAX_RECURSION_DEPTH)*/
+  int recursionMax;
+
+  /*
+  Configuration for max decimals of double floating point numbers to encode
+  (0-9) */
+  int doublePrecision;
+
+  /*
+  If true output will be ASCII with all characters above 127 encoded as \uXXXX.
+  If false output will be UTF-8 or what ever charset strings are brought as */
+  int forceASCII;
+
+  /*
+  If true, '<', '>', and '&' characters will be encoded as \u003c, \u003e, and
+  \u0026, respectively. If false, no special encoding will be used. */
+  int encodeHTMLChars;
+
+  /*
+  Configuration for spaces of indent */
+  int indent;
+
+  /*
+  Set to an error message if error occurred */
+  const char *errorMsg;
+  JSOBJ errorObj;
+
+  /* Buffer stuff */
+  char *start;
+  char *offset;
+  char *end;
+  int heap;
+  int level;
+} JSONObjectEncoder;
+
+/*
+Encode an object structure into JSON.
+
+Arguments:
+obj - An anonymous type representing the object
+enc - Function definitions for querying JSOBJ type
+buffer - Preallocated buffer to store result in. If NULL function allocates own
+buffer cbBuffer - Length of buffer (ignored if buffer is NULL)
+
+Returns:
+Encoded JSON object as a null terminated char string.
+
+NOTE:
+If the supplied buffer wasn't enough to hold the result the function will
+allocate a new buffer. Life cycle of the provided buffer must still be handled
+by caller.
+
+If the return value doesn't equal the specified buffer caller must release the
+memory using JSONObjectEncoder.free or free() as specified when calling this
+function.
+*/
+EXPORTFUNCTION char *JSON_EncodeObject(JSOBJ obj, JSONObjectEncoder *enc,
+                                       char *buffer, size_t cbBuffer);
+
+typedef struct __JSONObjectDecoder {
+  JSOBJ (*newString)(void *prv, wchar_t *start, wchar_t *end);
+  int (*objectAddKey)(void *prv, JSOBJ obj, JSOBJ name, JSOBJ value);
+  int (*arrayAddItem)(void *prv, JSOBJ obj, JSOBJ value);
+  JSOBJ (*newTrue)(void *prv);
+  JSOBJ (*newFalse)(void *prv);
+  JSOBJ (*newNull)(void *prv);
+  JSOBJ (*newPosInf)(void *prv);
+  JSOBJ (*newNegInf)(void *prv);
+  JSOBJ (*newObject)(void *prv, void *decoder);
+  JSOBJ (*endObject)(void *prv, JSOBJ obj);
+  JSOBJ (*newArray)(void *prv, void *decoder);
+  JSOBJ (*endArray)(void *prv, JSOBJ obj);
+  JSOBJ (*newInt)(void *prv, JSINT32 value);
+  JSOBJ (*newLong)(void *prv, JSINT64 value);
+  JSOBJ (*newUnsignedLong)(void *prv, JSUINT64 value);
+  JSOBJ (*newDouble)(void *prv, double value);
+  void (*releaseObject)(void *prv, JSOBJ obj, void *decoder);
+  JSPFN_MALLOC malloc;
+  JSPFN_FREE free;
+  JSPFN_REALLOC realloc;
+  char *errorStr;
+  char *errorOffset;
+  int preciseFloat;
+  void *prv;
+} JSONObjectDecoder;
+
+EXPORTFUNCTION JSOBJ JSON_DecodeObject(JSONObjectDecoder *dec,
+                                       const char *buffer, size_t cbBuffer);
+EXPORTFUNCTION void encode(JSOBJ, JSONObjectEncoder *, const char *, size_t);
diff --git a/pandas/pandas/_libs/index.pyi b/pandas/pandas/_libs/index.pyi
new file mode 100644
index 0000000000000..3af2856d2fbbf
--- /dev/null
+++ b/pandas/pandas/_libs/index.pyi
@@ -0,0 +1,107 @@
+import numpy as np
+
+from pandas._typing import npt
+
+from pandas import (
+    Index,
+    MultiIndex,
+)
+from pandas.core.arrays import ExtensionArray
+
+multiindex_nulls_shift: int
+
+class IndexEngine:
+    over_size_threshold: bool
+    def __init__(self, values: np.ndarray) -> None: ...
+    def __contains__(self, val: object) -> bool: ...
+
+    # -> int | slice | np.ndarray[bool]
+    def get_loc(self, val: object) -> int | slice | np.ndarray: ...
+    def sizeof(self, deep: bool = ...) -> int: ...
+    def __sizeof__(self) -> int: ...
+    @property
+    def is_unique(self) -> bool: ...
+    @property
+    def is_monotonic_increasing(self) -> bool: ...
+    @property
+    def is_monotonic_decreasing(self) -> bool: ...
+    @property
+    def is_mapping_populated(self) -> bool: ...
+    def clear_mapping(self): ...
+    def get_indexer(self, values: np.ndarray) -> npt.NDArray[np.intp]: ...
+    def get_indexer_non_unique(
+        self,
+        targets: np.ndarray,
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+
+class MaskedIndexEngine(IndexEngine):
+    def __init__(self, values: object) -> None: ...
+    def get_indexer_non_unique(
+        self, targets: object
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+
+class Float64Engine(IndexEngine): ...
+class Float32Engine(IndexEngine): ...
+class Complex128Engine(IndexEngine): ...
+class Complex64Engine(IndexEngine): ...
+class Int64Engine(IndexEngine): ...
+class Int32Engine(IndexEngine): ...
+class Int16Engine(IndexEngine): ...
+class Int8Engine(IndexEngine): ...
+class UInt64Engine(IndexEngine): ...
+class UInt32Engine(IndexEngine): ...
+class UInt16Engine(IndexEngine): ...
+class UInt8Engine(IndexEngine): ...
+class ObjectEngine(IndexEngine): ...
+class StringEngine(IndexEngine): ...
+class DatetimeEngine(Int64Engine): ...
+class TimedeltaEngine(DatetimeEngine): ...
+class PeriodEngine(Int64Engine): ...
+class BoolEngine(UInt8Engine): ...
+class MaskedFloat64Engine(MaskedIndexEngine): ...
+class MaskedFloat32Engine(MaskedIndexEngine): ...
+class MaskedComplex128Engine(MaskedIndexEngine): ...
+class MaskedComplex64Engine(MaskedIndexEngine): ...
+class MaskedInt64Engine(MaskedIndexEngine): ...
+class MaskedInt32Engine(MaskedIndexEngine): ...
+class MaskedInt16Engine(MaskedIndexEngine): ...
+class MaskedInt8Engine(MaskedIndexEngine): ...
+class MaskedUInt64Engine(MaskedIndexEngine): ...
+class MaskedUInt32Engine(MaskedIndexEngine): ...
+class MaskedUInt16Engine(MaskedIndexEngine): ...
+class MaskedUInt8Engine(MaskedIndexEngine): ...
+class MaskedBoolEngine(MaskedUInt8Engine): ...
+
+class StringObjectEngine(ObjectEngine):
+    def __init__(self, values: object, na_value) -> None: ...
+
+class BaseMultiIndexCodesEngine:
+    levels: list[np.ndarray]
+    offsets: np.ndarray  # np.ndarray[..., ndim=1]
+
+    def __init__(
+        self,
+        levels: list[Index],  # all entries hashable
+        labels: list[np.ndarray],  # all entries integer-dtyped
+        offsets: np.ndarray,  # np.ndarray[..., ndim=1]
+    ) -> None: ...
+    def get_indexer(self, target: npt.NDArray[np.object_]) -> npt.NDArray[np.intp]: ...
+    def _extract_level_codes(self, target: MultiIndex) -> np.ndarray: ...
+
+class ExtensionEngine:
+    def __init__(self, values: ExtensionArray) -> None: ...
+    def __contains__(self, val: object) -> bool: ...
+    def get_loc(self, val: object) -> int | slice | np.ndarray: ...
+    def get_indexer(self, values: np.ndarray) -> npt.NDArray[np.intp]: ...
+    def get_indexer_non_unique(
+        self,
+        targets: np.ndarray,
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+    @property
+    def is_unique(self) -> bool: ...
+    @property
+    def is_monotonic_increasing(self) -> bool: ...
+    @property
+    def is_monotonic_decreasing(self) -> bool: ...
+    def sizeof(self, deep: bool = ...) -> int: ...
+    def clear_mapping(self): ...
diff --git a/pandas/pandas/_libs/index.pyx b/pandas/pandas/_libs/index.pyx
new file mode 100644
index 0000000000000..c219d0b63870f
--- /dev/null
+++ b/pandas/pandas/_libs/index.pyx
@@ -0,0 +1,1317 @@
+cimport cython
+from cpython.sequence cimport PySequence_GetItem
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    intp_t,
+    ndarray,
+    uint8_t,
+)
+
+cnp.import_array()
+
+
+from pandas._libs cimport util
+from pandas._libs.hashtable cimport HashTable
+from pandas._libs.tslibs.nattype cimport c_NaT as NaT
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    get_unit_from_dtype,
+    import_pandas_datetime,
+)
+
+import_pandas_datetime()
+
+
+from pandas._libs.tslibs.period cimport is_period_object
+from pandas._libs.tslibs.timedeltas cimport _Timedelta
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+
+from pandas._libs import (
+    algos,
+    hashtable as _hash,
+)
+
+from pandas._libs.lib cimport eq_NA_compat
+from pandas._libs.missing cimport (
+    C_NA,
+    checknull,
+    is_matching_na,
+)
+
+from decimal import InvalidOperation
+
+# Defines shift of MultiIndex codes to avoid negative codes (missing values)
+multiindex_nulls_shift = 2
+
+
+cdef bint is_definitely_invalid_key(object val):
+    try:
+        hash(val)
+    except TypeError:
+        return True
+    return False
+
+
+cdef ndarray _get_bool_indexer(ndarray values, object val, ndarray mask = None):
+    """
+    Return a ndarray[bool] of locations where val matches self.values.
+
+    If val is not NA, this is equivalent to `self.values == val`
+    """
+    # Caller is responsible for ensuring _check_type has already been called
+    cdef:
+        ndarray[uint8_t, ndim=1, cast=True] indexer
+        Py_ssize_t i
+        object item
+
+    if values.descr.type_num == cnp.NPY_OBJECT:
+        assert mask is None  # no mask for object dtype
+        # i.e. values.dtype == object
+        if not checknull(val):
+            indexer = eq_NA_compat(values, val)
+
+        else:
+            # We need to check for _matching_ NA values
+            indexer = np.empty(len(values), dtype=np.uint8)
+
+            for i in range(len(values)):
+                item = PySequence_GetItem(values, i)
+                indexer[i] = is_matching_na(item, val)
+
+    else:
+        if mask is not None:
+            if val is C_NA:
+                indexer = mask == 1
+            else:
+                indexer = (values == val) & ~mask
+        else:
+            if util.is_nan(val):
+                indexer = np.isnan(values)
+            else:
+                indexer = values == val
+
+    return indexer.view(bool)
+
+
+cdef _maybe_resize_array(ndarray values, Py_ssize_t loc, Py_ssize_t max_length):
+    """
+    Resize array if loc is out of bounds.
+    """
+    cdef:
+        Py_ssize_t n = len(values)
+
+    if loc >= n:
+        while loc >= n:
+            n *= 2
+        values = np.resize(values, min(n, max_length))
+    return values
+
+
+# Don't populate hash tables in monotonic indexes larger than this
+_SIZE_CUTOFF = 1_000_000
+
+
+cdef _unpack_bool_indexer(ndarray[uint8_t, ndim=1, cast=True] indexer, object val):
+    """
+    Possibly unpack a boolean mask to a single indexer.
+    """
+    # Returns ndarray[bool] or int
+    cdef:
+        ndarray[intp_t, ndim=1] found
+        int count
+
+    found = np.where(indexer)[0]
+    count = len(found)
+
+    if count > 1:
+        return indexer
+    if count == 1:
+        return int(found[0])
+
+    raise KeyError(val)
+
+
+@cython.freelist(32)
+cdef class IndexEngine:
+
+    cdef readonly:
+        ndarray values
+        ndarray mask
+        HashTable mapping
+        bint over_size_threshold
+
+    cdef:
+        bint unique, monotonic_inc, monotonic_dec
+        bint need_monotonic_check, need_unique_check
+        object _np_type
+
+    def __init__(self, ndarray values):
+        self.values = values
+        self.mask = None
+
+        self.over_size_threshold = len(values) >= _SIZE_CUTOFF
+        self.clear_mapping()
+        self._np_type = values.dtype.type
+
+    def __contains__(self, val: object) -> bool:
+        hash(val)
+        try:
+            self.get_loc(val)
+        except KeyError:
+            return False
+        return True
+
+    cpdef get_loc(self, object val):
+        # -> Py_ssize_t | slice | ndarray[bool]
+        cdef:
+            Py_ssize_t loc
+
+        if is_definitely_invalid_key(val):
+            raise TypeError(f"'{val}' is an invalid key")
+
+        val = self._check_type(val)
+
+        if self.over_size_threshold and self.is_monotonic_increasing:
+            if not self.is_unique:
+                return self._get_loc_duplicates(val)
+            values = self.values
+
+            loc = self._searchsorted_left(val)
+            if loc >= len(values):
+                raise KeyError(val)
+            if values[loc] != val:
+                raise KeyError(val)
+            return loc
+
+        self._ensure_mapping_populated()
+        if not self.unique:
+            return self._get_loc_duplicates(val)
+        if self.mask is not None and val is C_NA:
+            return self.mapping.get_na()
+
+        try:
+            return self.mapping.get_item(val)
+        except OverflowError as err:
+            # GH#41775 OverflowError e.g. if we are uint64 and val is -1
+            #  or if we are int64 and value is np.iinfo(np.int64).max+1
+            #  (the uint64 with -1 case should actually be excluded by _check_type)
+            raise KeyError(val) from err
+
+    cdef Py_ssize_t _searchsorted_left(self, val) except? -1:
+        """
+        See ObjectEngine._searchsorted_left.__doc__.
+        """
+        # Caller is responsible for ensuring _check_type has already been called
+        loc = self.values.searchsorted(self._np_type(val), side="left")
+        return loc
+
+    cdef _get_loc_duplicates(self, object val):
+        # -> Py_ssize_t | slice | ndarray[bool]
+        cdef:
+            Py_ssize_t diff, left, right
+
+        if self.is_monotonic_increasing:
+            values = self.values
+            try:
+                left = values.searchsorted(val, side="left")
+                right = values.searchsorted(val, side="right")
+            except TypeError:
+                # e.g. GH#29189 get_loc(None) with a Float64Index
+                #  2021-09-29 Now only reached for object-dtype
+                raise KeyError(val)
+
+            diff = right - left
+            if diff == 0:
+                raise KeyError(val)
+            elif diff == 1:
+                return left
+            else:
+                return slice(left, right)
+
+        return self._maybe_get_bool_indexer(val)
+
+    cdef _maybe_get_bool_indexer(self, object val):
+        # Returns ndarray[bool] or int
+        cdef:
+            ndarray[uint8_t, ndim=1, cast=True] indexer
+
+        indexer = _get_bool_indexer(self.values, val, self.mask)
+        return _unpack_bool_indexer(indexer, val)
+
+    def sizeof(self, deep: bool = False) -> int:
+        """ return the sizeof our mapping """
+        if not self.is_mapping_populated:
+            return 0
+        return self.mapping.sizeof(deep=deep)
+
+    def __sizeof__(self) -> int:
+        return self.sizeof()
+
+    cpdef _update_from_sliced(self, IndexEngine other, reverse: bool):
+        if other.unique:
+            self.unique = other.unique
+            self.need_unique_check = other.need_unique_check
+
+        if not other.need_monotonic_check and (
+                other.is_monotonic_increasing or other.is_monotonic_decreasing):
+            self.need_monotonic_check = 0
+            if len(self.values) > 0 and self.values[0] != self.values[-1]:
+                # reverse=True means the index has been reversed
+                if reverse:
+                    self.monotonic_inc = other.monotonic_dec
+                    self.monotonic_dec = other.monotonic_inc
+                else:
+                    self.monotonic_inc = other.monotonic_inc
+                    self.monotonic_dec = other.monotonic_dec
+            else:
+                self.monotonic_inc = 1
+                self.monotonic_dec = 1
+
+    @property
+    def is_unique(self) -> bool:
+        # for why we check is_monotonic_increasing here, see
+        # https://github.com/pandas-dev/pandas/pull/55342#discussion_r1361405781
+        if self.need_monotonic_check:
+            self.is_monotonic_increasing
+        if self.need_unique_check:
+            self._do_unique_check()
+
+        return self.unique == 1
+
+    cdef _do_unique_check(self):
+        self._ensure_mapping_populated()
+
+    @property
+    def is_monotonic_increasing(self) -> bool:
+        if self.need_monotonic_check:
+            self._do_monotonic_check()
+
+        return self.monotonic_inc == 1
+
+    @property
+    def is_monotonic_decreasing(self) -> bool:
+        if self.need_monotonic_check:
+            self._do_monotonic_check()
+
+        return self.monotonic_dec == 1
+
+    cdef _do_monotonic_check(self):
+        cdef:
+            bint is_strict_monotonic
+        if self.mask is not None and np.any(self.mask):
+            self.monotonic_inc = 0
+            self.monotonic_dec = 0
+        else:
+            try:
+                values = self.values
+                self.monotonic_inc, self.monotonic_dec, is_strict_monotonic = \
+                    self._call_monotonic(values)
+            except (TypeError, InvalidOperation, ValueError):
+                self.monotonic_inc = 0
+                self.monotonic_dec = 0
+                is_strict_monotonic = 0
+
+            self.need_monotonic_check = 0
+
+            # we can only be sure of uniqueness if is_strict_monotonic=1
+            if is_strict_monotonic:
+                self.unique = 1
+                self.need_unique_check = 0
+
+    cdef _call_monotonic(self, values):
+        return algos.is_monotonic(values, timelike=False)
+
+    cdef _make_hash_table(self, Py_ssize_t n):
+        raise NotImplementedError  # pragma: no cover
+
+    cdef _check_type(self, object val):
+        hash(val)
+        return val
+
+    @property
+    def is_mapping_populated(self) -> bool:
+        return self.mapping is not None
+
+    cdef _ensure_mapping_populated(self):
+        # this populates the mapping
+        # if its not already populated
+        # also satisfies the need_unique_check
+
+        if not self.is_mapping_populated:
+
+            values = self.values
+            self.mapping = self._make_hash_table(len(values))
+            self.mapping.map_locations(values, self.mask)
+
+            if len(self.mapping) == len(values):
+                self.unique = 1
+
+        self.need_unique_check = 0
+
+    def clear_mapping(self):
+        self.mapping = None
+        self.need_monotonic_check = 1
+        self.need_unique_check = 1
+
+        self.unique = 0
+        self.monotonic_inc = 0
+        self.monotonic_dec = 0
+
+    def get_indexer(self, ndarray values) -> np.ndarray:
+        self._ensure_mapping_populated()
+        return self.mapping.lookup(values)
+
+    def get_indexer_non_unique(self, ndarray targets):
+        """
+        Return an indexer suitable for taking from a non unique index
+        return the labels in the same order as the target
+        and a missing indexer into the targets (which correspond
+        to the -1 indices in the results
+
+        Returns
+        -------
+        indexer : np.ndarray[np.intp]
+        missing : np.ndarray[np.intp]
+        """
+        cdef:
+            ndarray values
+            ndarray[intp_t] result, missing
+            set stargets, remaining_stargets, found_nas
+            dict d = {}
+            object val
+            Py_ssize_t count = 0, count_missing = 0
+            Py_ssize_t i, j, n, n_t, n_alloc, max_alloc, start, end
+            bint check_na_values = False
+
+        values = self.values
+        stargets = set(targets)
+
+        na_in_stargets = any(checknull(t) for t in stargets)
+
+        n = len(values)
+        n_t = len(targets)
+        max_alloc = n * n_t
+        if n > 10_000:
+            n_alloc = 10_000
+        else:
+            n_alloc = n
+
+        result = np.empty(n_alloc, dtype=np.intp)
+        missing = np.empty(n_t, dtype=np.intp)
+
+        # map each starget to its position in the index
+        if (
+                stargets and
+                len(stargets) < (n / (2 * n.bit_length())) and
+                not na_in_stargets and
+                self.is_monotonic_increasing
+        ):
+            # if there are few enough stargets and the index is monotonically
+            # increasing, then use binary search for each starget
+            remaining_stargets = set()
+            for starget in stargets:
+                try:
+                    start = values.searchsorted(starget, side="left")
+                    end = values.searchsorted(starget, side="right")
+                except TypeError:  # e.g. if we tried to search for string in int array
+                    remaining_stargets.add(starget)
+                else:
+                    if start != end:
+                        d[starget] = list(range(start, end))
+
+            stargets = remaining_stargets
+
+        if stargets:
+            # otherwise, map by iterating through all items in the index
+
+            # short-circuit na check
+            if na_in_stargets:
+                check_na_values = True
+                # keep track of nas in values
+                found_nas = set()
+
+            for i in range(n):
+                val = PySequence_GetItem(values, i)
+
+                # GH#43870
+                # handle lookup for nas
+                # (ie. np.nan, float("NaN"), Decimal("NaN"), dt64nat, td64nat)
+                if check_na_values and checknull(val):
+                    match = [na for na in found_nas if is_matching_na(val, na)]
+
+                    # matching na not found
+                    if not len(match):
+                        found_nas.add(val)
+
+                        # add na to stargets to utilize `in` for stargets/d lookup
+                        match_stargets = [
+                            x for x in stargets if is_matching_na(val, x)
+                        ]
+
+                        if len(match_stargets):
+                            # add our 'standardized' na
+                            stargets.add(val)
+
+                    # matching na found
+                    else:
+                        assert len(match) == 1
+                        val = match[0]
+
+                if val in stargets:
+                    if val not in d:
+                        d[val] = []
+                    d[val].append(i)
+
+        for i in range(n_t):
+            val = PySequence_GetItem(targets, i)
+
+            # ensure there are nas in values before looking for a matching na
+            if check_na_values and checknull(val):
+                match = [na for na in found_nas if is_matching_na(val, na)]
+                if len(match):
+                    assert len(match) == 1
+                    val = match[0]
+
+            # found
+            if val in d:
+                key = val
+                result = _maybe_resize_array(
+                    result,
+                    count + len(d[key]) - 1,
+                    max_alloc
+                )
+                for j in d[key]:
+                    result[count] = j
+                    count += 1
+
+            # value not found
+            else:
+                result = _maybe_resize_array(result, count, max_alloc)
+                result[count] = -1
+                count += 1
+                missing[count_missing] = i
+                count_missing += 1
+
+        return result[0:count], missing[0:count_missing]
+
+
+cdef Py_ssize_t _bin_search(ndarray values, object val) except -1:
+    # GH#1757 ndarray.searchsorted is not safe to use with array of tuples
+    #  (treats a tuple `val` as a sequence of keys instead of a single key),
+    #  so we implement something similar.
+    # This is equivalent to the stdlib's bisect.bisect_left
+
+    cdef:
+        Py_ssize_t mid = 0, lo = 0, hi = len(values) - 1
+        object pval
+
+    if hi == 0 or (hi > 0 and val > PySequence_GetItem(values, hi)):
+        return len(values)
+
+    while lo < hi:
+        mid = (lo + hi) // 2
+        pval = PySequence_GetItem(values, mid)
+        if val < pval:
+            hi = mid
+        elif val > pval:
+            lo = mid + 1
+        else:
+            while mid > 0 and val == PySequence_GetItem(values, mid - 1):
+                mid -= 1
+            return mid
+
+    if val <= PySequence_GetItem(values, mid):
+        return mid
+    else:
+        return mid + 1
+
+
+cdef class ObjectEngine(IndexEngine):
+    """
+    Index Engine for use with object-dtype Index, namely the base class Index.
+    """
+    cdef _make_hash_table(self, Py_ssize_t n):
+        return _hash.PyObjectHashTable(n)
+
+    cdef Py_ssize_t _searchsorted_left(self, val) except? -1:
+        # using values.searchsorted here would treat a tuple `val` as a sequence
+        #  instead of a single key, so we use a different implementation
+        try:
+            loc = _bin_search(self.values, val)
+        except TypeError as err:
+            raise KeyError(val) from err
+        return loc
+
+
+cdef class StringEngine(IndexEngine):
+
+    cdef _make_hash_table(self, Py_ssize_t n):
+        return _hash.StringHashTable(n)
+
+    cdef _check_type(self, object val):
+        if not isinstance(val, str):
+            raise KeyError(val)
+        return str(val)
+
+cdef class StringObjectEngine(ObjectEngine):
+
+    cdef:
+        object na_value
+
+    def __init__(self, ndarray values, na_value):
+        super().__init__(values)
+        self.na_value = na_value
+
+    cdef _check_type(self, object val):
+        if isinstance(val, str):
+            return val
+        elif checknull(val):
+            return self.na_value
+        else:
+            raise KeyError(val)
+
+
+cdef class DatetimeEngine(Int64Engine):
+
+    cdef:
+        NPY_DATETIMEUNIT _creso
+
+    def __init__(self, ndarray values):
+        super().__init__(values.view("i8"))
+        self._creso = get_unit_from_dtype(values.dtype)
+
+    cdef int64_t _unbox_scalar(self, scalar) except? -1:
+        # NB: caller is responsible for ensuring tzawareness compat
+        #  before we get here
+        if scalar is NaT:
+            return NaT._value
+        elif isinstance(scalar, _Timestamp):
+            if scalar._creso == self._creso:
+                return scalar._value
+            else:
+                # Note: caller is responsible for catching potential ValueError
+                #  from _as_creso
+                return (
+                    (<_Timestamp>scalar)._as_creso(self._creso, round_ok=False)._value
+                )
+        raise TypeError(scalar)
+
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
+        try:
+            self._unbox_scalar(val)
+        except ValueError:
+            return False
+
+        try:
+            self.get_loc(val)
+            return True
+        except KeyError:
+            return False
+
+    cdef _call_monotonic(self, values):
+        return algos.is_monotonic(values, timelike=True)
+
+    cpdef get_loc(self, object val):
+        # NB: the caller is responsible for ensuring that we are called
+        #  with either a Timestamp or NaT (Timedelta or NaT for TimedeltaEngine)
+
+        cdef:
+            Py_ssize_t loc
+
+        if is_definitely_invalid_key(val):
+            raise TypeError(f"'{val}' is an invalid key")
+
+        try:
+            conv = self._unbox_scalar(val)
+        except (TypeError, ValueError) as err:
+            raise KeyError(val) from err
+
+        # Welcome to the spaghetti factory
+        if self.over_size_threshold and self.is_monotonic_increasing:
+            if not self.is_unique:
+                return self._get_loc_duplicates(conv)
+            values = self.values
+
+            loc = values.searchsorted(conv, side="left")
+
+            if loc == len(values) or PySequence_GetItem(values, loc) != conv:
+                raise KeyError(val)
+            return loc
+
+        self._ensure_mapping_populated()
+        if not self.unique:
+            return self._get_loc_duplicates(conv)
+
+        try:
+            return self.mapping.get_item(conv)
+        except KeyError:
+            raise KeyError(val)
+
+
+cdef class TimedeltaEngine(DatetimeEngine):
+
+    cdef int64_t _unbox_scalar(self, scalar) except? -1:
+        if scalar is NaT:
+            return NaT._value
+        elif isinstance(scalar, _Timedelta):
+            if scalar._creso == self._creso:
+                return scalar._value
+            else:
+                # Note: caller is responsible for catching potential ValueError
+                #  from _as_creso
+                return (
+                    (<_Timedelta>scalar)._as_creso(self._creso, round_ok=False)._value
+                )
+        raise TypeError(scalar)
+
+
+cdef class PeriodEngine(Int64Engine):
+
+    cdef int64_t _unbox_scalar(self, scalar) except? -1:
+        if scalar is NaT:
+            return scalar._value
+        if is_period_object(scalar):
+            # NB: we assume that we have the correct freq here.
+            return scalar.ordinal
+        raise TypeError(scalar)
+
+    cpdef get_loc(self, object val):
+        # NB: the caller is responsible for ensuring that we are called
+        #  with either a Period or NaT
+        cdef:
+            int64_t conv
+
+        try:
+            conv = self._unbox_scalar(val)
+        except TypeError:
+            raise KeyError(val)
+
+        return Int64Engine.get_loc(self, conv)
+
+    cdef _call_monotonic(self, values):
+        return algos.is_monotonic(values, timelike=True)
+
+
+cdef class BaseMultiIndexCodesEngine:
+    """
+    Base class for MultiIndexUIntEngine and MultiIndexPyIntEngine, which
+    represent each label in a MultiIndex as an integer, by juxtaposing the bits
+    encoding each level, with appropriate offsets.
+
+    For instance: if 3 levels have respectively 3, 6 and 1 possible values,
+    then their labels can be represented using respectively 2, 3 and 1 bits,
+    as follows:
+     _ _ _ _____ _ __ __ __
+    |0|0|0| ... |0| 0|a1|a0| -> offset 0 (first level)
+     — — — ————— — —— —— ——
+    |0|0|0| ... |0|b2|b1|b0| -> offset 2 (bits required for first level)
+     — — — ————— — —— —— ——
+    |0|0|0| ... |0| 0| 0|c0| -> offset 5 (bits required for first two levels)
+     ‾ ‾ ‾ ‾‾‾‾‾ ‾ ‾‾ ‾‾ ‾‾
+    and the resulting unsigned integer representation will be:
+     _ _ _ _____ _ __ __ __ __ __ __
+    |0|0|0| ... |0|c0|b2|b1|b0|a1|a0|
+     ‾ ‾ ‾ ‾‾‾‾‾ ‾ ‾‾ ‾‾ ‾‾ ‾‾ ‾‾ ‾‾
+
+    Offsets are calculated at initialization, labels are transformed by method
+    _codes_to_ints.
+
+    Keys are located by first locating each component against the respective
+    level, then locating (the integer representation of) codes.
+    """
+    def __init__(self, object levels, object labels, ndarray offsets):
+        """
+        Parameters
+        ----------
+        levels : list-like of numpy arrays
+            Levels of the MultiIndex.
+        labels : list-like of numpy arrays of integer dtype
+            Labels of the MultiIndex.
+        offsets : numpy array of int dtype
+            Pre-calculated offsets, one for each level of the index.
+        """
+        self.levels = levels
+        self.offsets = offsets
+
+        # Transform labels in a single array, and add 2 so that we are working
+        # with positive integers (-1 for NaN becomes 1). This enables us to
+        # differentiate between values that are missing in other and matching
+        # NaNs. We will set values that are not found to 0 later:
+        codes = np.array(labels).T
+        codes += multiindex_nulls_shift  # inplace sum optimisation
+
+        self.level_has_nans = [-1 in lab for lab in labels]
+
+        # Map each codes combination in the index to an integer unambiguously
+        # (no collisions possible), based on the "offsets", which describe the
+        # number of bits to switch labels for each level:
+        lab_ints = self._codes_to_ints(codes)
+
+        # Initialize underlying index (e.g. libindex.UInt64Engine) with
+        # integers representing labels: we will use its get_loc and get_indexer
+        self._base.__init__(self, lab_ints)
+
+    def _codes_to_ints(self, ndarray codes) -> np.ndarray:
+        """
+        Transform combination(s) of uint in one uint or Python integer (each), in a
+        strictly monotonic way (i.e. respecting the lexicographic order of integer
+        combinations).
+
+        Parameters
+        ----------
+        codes : 1- or 2-dimensional array of dtype uint
+            Combinations of integers (one per row)
+
+        Returns
+        -------
+        scalar or 1-dimensional array, of dtype _codes_dtype
+            Integer(s) representing one combination (each).
+        """
+        # To avoid overflows, first make sure we are working with the right dtype:
+        codes = codes.astype(self._codes_dtype, copy=False)
+
+        # Shift the representation of each level by the pre-calculated number of bits:
+        codes <<= self.offsets  # inplace shift optimisation
+
+        # Now sum and OR are in fact interchangeable. This is a simple
+        # composition of the (disjunct) significant bits of each level (i.e.
+        # each column in "codes") in a single positive integer (per row):
+        if codes.ndim == 1:
+            # Single key
+            return np.bitwise_or.reduce(codes)
+
+        # Multiple keys
+        return np.bitwise_or.reduce(codes, axis=1)
+
+    def _extract_level_codes(self, target) -> np.ndarray:
+        """
+        Map the requested list of (tuple) keys to their integer representations
+        for searching in the underlying integer index.
+
+        Parameters
+        ----------
+        target : MultiIndex
+
+        Returns
+        ------
+        int_keys : 1-dimensional array of dtype uint64 or object
+            Integers representing one combination each
+        """
+        level_codes = list(target._recode_for_new_levels(self.levels))
+        for i, codes in enumerate(level_codes):
+            if self.levels[i].hasnans:
+                na_index = self.levels[i].isna().nonzero()[0][0]
+                codes[target.codes[i] == -1] = na_index
+            codes += 1
+            codes[codes > 0] += 1
+            if self.level_has_nans[i]:
+                codes[target.codes[i] == -1] += 1
+        return self._codes_to_ints(np.array(level_codes, dtype=self._codes_dtype).T)
+
+    def get_indexer(self, target: np.ndarray) -> np.ndarray:
+        """
+        Returns an array giving the positions of each value of `target` in
+        `self.values`, where -1 represents a value in `target` which does not
+        appear in `self.values`
+
+        Parameters
+        ----------
+        target : np.ndarray
+
+        Returns
+        -------
+        np.ndarray[intp_t, ndim=1] of the indexer of `target` into
+        `self.values`
+        """
+        return self._base.get_indexer(self, target)
+
+    def get_loc(self, object key):
+        if is_definitely_invalid_key(key):
+            raise TypeError(f"'{key}' is an invalid key")
+        if not isinstance(key, tuple):
+            raise KeyError(key)
+        try:
+            indices = [1 if checknull(v) else lev.get_loc(v) + multiindex_nulls_shift
+                       for lev, v in zip(self.levels, key)]
+        except KeyError:
+            raise KeyError(key)
+
+        # Transform indices into single integer:
+        lab_int = self._codes_to_ints(np.array(indices, dtype=self._codes_dtype))
+
+        return self._base.get_loc(self, lab_int)
+
+    def get_indexer_non_unique(self, target: np.ndarray) -> np.ndarray:
+        indexer = self._base.get_indexer_non_unique(self, target)
+
+        return indexer
+
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
+        # Default __contains__ looks in the underlying mapping, which in this
+        # case only contains integer representations.
+        try:
+            self.get_loc(val)
+            return True
+        except (KeyError, TypeError, ValueError):
+            return False
+
+
+# Generated from template.
+include "index_class_helper.pxi"
+
+
+cdef class BoolEngine(UInt8Engine):
+    cdef _check_type(self, object val):
+        if not util.is_bool_object(val):
+            raise KeyError(val)
+        return <uint8_t>val
+
+
+cdef class MaskedBoolEngine(MaskedUInt8Engine):
+    cdef _check_type(self, object val):
+        if val is C_NA:
+            return val
+        if not util.is_bool_object(val):
+            raise KeyError(val)
+        return <uint8_t>val
+
+
+@cython.internal
+@cython.freelist(32)
+cdef class SharedEngine:
+    cdef readonly:
+        object values  # ExtensionArray
+        bint over_size_threshold
+
+    cdef:
+        bint unique, monotonic_inc, monotonic_dec
+        bint need_monotonic_check, need_unique_check
+
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
+        try:
+            self.get_loc(val)
+            return True
+        except KeyError:
+            return False
+
+    def clear_mapping(self):
+        # for compat with IndexEngine
+        pass
+
+    cpdef _update_from_sliced(self, ExtensionEngine other, reverse: bool):
+        if other.unique:
+            self.unique = other.unique
+            self.need_unique_check = other.need_unique_check
+
+        if not other.need_monotonic_check and (
+                other.is_monotonic_increasing or other.is_monotonic_decreasing):
+            self.need_monotonic_check = 0
+            if len(self.values) > 0 and self.values[0] != self.values[-1]:
+                # reverse=True means the index has been reversed
+                if reverse:
+                    self.monotonic_inc = other.monotonic_dec
+                    self.monotonic_dec = other.monotonic_inc
+                else:
+                    self.monotonic_inc = other.monotonic_inc
+                    self.monotonic_dec = other.monotonic_dec
+            else:
+                self.monotonic_inc = 1
+                self.monotonic_dec = 1
+
+    @property
+    def is_unique(self) -> bool:
+        # for why we check is_monotonic_increasing here, see
+        # https://github.com/pandas-dev/pandas/pull/55342#discussion_r1361405781
+        if self.need_monotonic_check:
+            self.is_monotonic_increasing
+        if self.need_unique_check:
+            arr = self.values.unique()
+            self.unique = len(arr) == len(self.values)
+
+            self.need_unique_check = False
+        return self.unique
+
+    cdef _do_monotonic_check(self):
+        raise NotImplementedError
+
+    @property
+    def is_monotonic_increasing(self) -> bool:
+        if self.need_monotonic_check:
+            self._do_monotonic_check()
+
+        return self.monotonic_inc == 1
+
+    @property
+    def is_monotonic_decreasing(self) -> bool:
+        if self.need_monotonic_check:
+            self._do_monotonic_check()
+
+        return self.monotonic_dec == 1
+
+    cdef _call_monotonic(self, values):
+        return algos.is_monotonic(values, timelike=False)
+
+    def sizeof(self, deep: bool = False) -> int:
+        """ return the sizeof our mapping """
+        return 0
+
+    def __sizeof__(self) -> int:
+        return self.sizeof()
+
+    cdef _check_type(self, object obj):
+        raise NotImplementedError
+
+    cpdef get_loc(self, object val):
+        # -> Py_ssize_t | slice | ndarray[bool]
+        cdef:
+            Py_ssize_t loc
+
+        if is_definitely_invalid_key(val):
+            raise TypeError(f"'{val}' is an invalid key")
+
+        self._check_type(val)
+
+        if self.over_size_threshold and self.is_monotonic_increasing:
+            if not self.is_unique:
+                return self._get_loc_duplicates(val)
+
+            values = self.values
+
+            loc = self._searchsorted_left(val)
+            if loc >= len(values):
+                raise KeyError(val)
+            if values[loc] != val:
+                raise KeyError(val)
+            return loc
+
+        if not self.unique:
+            return self._get_loc_duplicates(val)
+
+        return self._get_loc_duplicates(val)
+
+    cdef _get_loc_duplicates(self, object val):
+        # -> Py_ssize_t | slice | ndarray[bool]
+        cdef:
+            Py_ssize_t diff
+
+        if self.is_monotonic_increasing:
+            values = self.values
+            try:
+                left = values.searchsorted(val, side="left")
+                right = values.searchsorted(val, side="right")
+            except TypeError:
+                # e.g. GH#29189 get_loc(None) with a Float64Index
+                raise KeyError(val)
+
+            diff = right - left
+            if diff == 0:
+                raise KeyError(val)
+            elif diff == 1:
+                return left
+            else:
+                return slice(left, right)
+
+        return self._maybe_get_bool_indexer(val)
+
+    cdef Py_ssize_t _searchsorted_left(self, val) except? -1:
+        """
+        See ObjectEngine._searchsorted_left.__doc__.
+        """
+        try:
+            loc = self.values.searchsorted(val, side="left")
+        except TypeError as err:
+            # GH#35788 e.g. val=None with float64 values
+            raise KeyError(val)
+        return loc
+
+    cdef ndarray _get_bool_indexer(self, val):
+        raise NotImplementedError
+
+    cdef _maybe_get_bool_indexer(self, object val):
+        # Returns ndarray[bool] or int
+        cdef:
+            ndarray[uint8_t, ndim=1, cast=True] indexer
+
+        indexer = self._get_bool_indexer(val)
+        return _unpack_bool_indexer(indexer, val)
+
+    def get_indexer(self, values) -> np.ndarray:
+        # values : type(self.values)
+        # Note: we only get here with self.is_unique
+        cdef:
+            Py_ssize_t i, N = len(values)
+
+        res = np.empty(N, dtype=np.intp)
+
+        for i in range(N):
+            val = PySequence_GetItem(values, i)
+            try:
+                loc = self.get_loc(val)
+                # Because we are unique, loc should always be an integer
+            except KeyError:
+                loc = -1
+            else:
+                assert util.is_integer_object(loc), (loc, val)
+            res[i] = loc
+
+        return res
+
+    def get_indexer_non_unique(self, targets):
+        """
+        Return an indexer suitable for taking from a non unique index
+        return the labels in the same order as the target
+        and a missing indexer into the targets (which correspond
+        to the -1 indices in the results
+        Parameters
+        ----------
+        targets : type(self.values)
+        Returns
+        -------
+        indexer : np.ndarray[np.intp]
+        missing : np.ndarray[np.intp]
+        """
+        cdef:
+            Py_ssize_t i, N = len(targets)
+
+        indexer = []
+        missing = []
+
+        # See also IntervalIndex.get_indexer_pointwise
+        for i in range(N):
+            val = PySequence_GetItem(targets, i)
+
+            try:
+                locs = self.get_loc(val)
+            except KeyError:
+                locs = np.array([-1], dtype=np.intp)
+                missing.append(i)
+            else:
+                if isinstance(locs, slice):
+                    # Only needed for get_indexer_non_unique
+                    locs = np.arange(locs.start, locs.stop, locs.step, dtype=np.intp)
+                elif util.is_integer_object(locs):
+                    locs = np.array([locs], dtype=np.intp)
+                else:
+                    assert locs.dtype.kind == "b"
+                    locs = locs.nonzero()[0]
+
+            indexer.append(locs)
+
+        try:
+            indexer = np.concatenate(indexer, dtype=np.intp)
+        except TypeError:
+            # numpy<1.20 doesn't accept dtype keyword
+            indexer = np.concatenate(indexer).astype(np.intp, copy=False)
+        missing = np.array(missing, dtype=np.intp)
+
+        return indexer, missing
+
+
+cdef class ExtensionEngine(SharedEngine):
+    def __init__(self, values: "ExtensionArray"):
+        self.values = values
+
+        self.over_size_threshold = len(values) >= _SIZE_CUTOFF
+        self.need_unique_check = True
+        self.need_monotonic_check = True
+        self.need_unique_check = True
+
+    cdef _do_monotonic_check(self):
+        cdef:
+            bint is_unique
+
+        values = self.values
+        if values._hasna:
+            self.monotonic_inc = 0
+            self.monotonic_dec = 0
+
+            nunique = len(values.unique())
+            self.unique = nunique == len(values)
+            self.need_unique_check = 0
+            return
+
+        try:
+            ranks = values._rank()
+
+        except TypeError:
+            self.monotonic_inc = 0
+            self.monotonic_dec = 0
+            is_unique = 0
+        else:
+            self.monotonic_inc, self.monotonic_dec, is_unique = \
+                self._call_monotonic(ranks)
+
+        self.need_monotonic_check = 0
+
+        # we can only be sure of uniqueness if is_unique=1
+        if is_unique:
+            self.unique = 1
+            self.need_unique_check = 0
+
+    cdef ndarray _get_bool_indexer(self, val):
+        if checknull(val):
+            return self.values.isna()
+
+        try:
+            return self.values == val
+        except TypeError:
+            # e.g. if __eq__ returns a BooleanArray instead of ndarray[bool]
+            try:
+                return (self.values == val).to_numpy(dtype=bool, na_value=False)
+            except (TypeError, AttributeError) as err:
+                # e.g. (self.values == val) returned a bool
+                #  see test_get_loc_generator[string[pyarrow]]
+                # e.g. self.value == val raises TypeError bc generator has no len
+                #  see test_get_loc_generator[string[python]]
+                raise KeyError from err
+
+    cdef _check_type(self, object val):
+        hash(val)
+
+
+cdef class MaskedIndexEngine(IndexEngine):
+    def __init__(self, object values):
+        super().__init__(self._get_data(values))
+        self.mask = self._get_mask(values)
+
+    def _get_data(self, object values) -> np.ndarray:
+        if hasattr(values, "_mask"):
+            return values._data
+        # We are an ArrowExtensionArray
+        # Set 1 as na_value to avoid ending up with NA and an object array
+        # TODO: Remove when arrow engine is implemented
+        return values.to_numpy(na_value=1, dtype=values.dtype.numpy_dtype)
+
+    def _get_mask(self, object values) -> np.ndarray:
+        if hasattr(values, "_mask"):
+            return values._mask
+        # We are an ArrowExtensionArray
+        return values.isna()
+
+    def get_indexer(self, object values) -> np.ndarray:
+        self._ensure_mapping_populated()
+        return self.mapping.lookup(self._get_data(values), self._get_mask(values))
+
+    def get_indexer_non_unique(self, object targets):
+        """
+        Return an indexer suitable for taking from a non unique index
+        return the labels in the same order as the target
+        and a missing indexer into the targets (which correspond
+        to the -1 indices in the results
+
+        Returns
+        -------
+        indexer : np.ndarray[np.intp]
+        missing : np.ndarray[np.intp]
+        """
+        # TODO: Unify with parent class
+        cdef:
+            ndarray values, mask, target_vals, target_mask
+            ndarray[intp_t] result, missing
+            set stargets
+            list na_pos
+            dict d = {}
+            object val
+            Py_ssize_t count = 0, count_missing = 0
+            Py_ssize_t i, j, n, n_t, n_alloc, max_alloc, start, end, na_idx
+
+        target_vals = self._get_data(targets)
+        target_mask = self._get_mask(targets)
+
+        values = self.values
+        assert not values.dtype == object  # go through object path instead
+
+        mask = self.mask
+        stargets = set(target_vals[~target_mask])
+
+        n = len(values)
+        n_t = len(target_vals)
+        max_alloc = n * n_t
+        if n > 10_000:
+            n_alloc = 10_000
+        else:
+            n_alloc = n
+
+        result = np.empty(n_alloc, dtype=np.intp)
+        missing = np.empty(n_t, dtype=np.intp)
+
+        # map each starget to its position in the index
+        if (
+                stargets and
+                len(stargets) < 5 and
+                not np.any(target_mask) and
+                self.is_monotonic_increasing
+        ):
+            # if there are few enough stargets and the index is monotonically
+            # increasing, then use binary search for each starget
+            for starget in stargets:
+                start = values.searchsorted(starget, side="left")
+                end = values.searchsorted(starget, side="right")
+                if start != end:
+                    d[starget] = list(range(start, end))
+
+            stargets = set()
+
+        if stargets:
+            # otherwise, map by iterating through all items in the index
+
+            na_pos = []
+
+            for i in range(n):
+                val = PySequence_GetItem(values, i)
+
+                if PySequence_GetItem(mask, i):
+                    na_pos.append(i)
+
+                else:
+                    if val in stargets:
+                        if val not in d:
+                            d[val] = []
+                        d[val].append(i)
+
+        for i in range(n_t):
+            val = PySequence_GetItem(target_vals, i)
+
+            if PySequence_GetItem(target_mask, i):
+                if na_pos:
+                    result = _maybe_resize_array(
+                        result,
+                        count + len(na_pos) - 1,
+                        max_alloc,
+                    )
+                    for na_idx in na_pos:
+                        result[count] = na_idx
+                        count += 1
+                    continue
+
+            elif val in d:
+                # found
+                key = val
+                result = _maybe_resize_array(
+                    result,
+                    count + len(d[key]) - 1,
+                    max_alloc,
+                )
+                for j in d[key]:
+                    result[count] = j
+                    count += 1
+                continue
+
+            # value not found
+            result = _maybe_resize_array(result, count, max_alloc)
+            result[count] = -1
+            count += 1
+            missing[count_missing] = i
+            count_missing += 1
+
+        return result[0:count], missing[0:count_missing]
diff --git a/pandas/pandas/_libs/index_class_helper.pxi.in b/pandas/pandas/_libs/index_class_helper.pxi.in
new file mode 100644
index 0000000000000..bf3d88edd9386
--- /dev/null
+++ b/pandas/pandas/_libs/index_class_helper.pxi.in
@@ -0,0 +1,78 @@
+"""
+Template for functions of IndexEngine subclasses.
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+# ----------------------------------------------------------------------
+# IndexEngine Subclass Methods
+# ----------------------------------------------------------------------
+
+{{py:
+
+# name, dtype
+dtypes = [('Float64', 'float64'),
+          ('Float32', 'float32'),
+          ('Int64', 'int64'),
+          ('Int32', 'int32'),
+          ('Int16', 'int16'),
+          ('Int8', 'int8'),
+          ('UInt64', 'uint64'),
+          ('UInt32', 'uint32'),
+          ('UInt16', 'uint16'),
+          ('UInt8', 'uint8'),
+          ('Complex64', 'complex64'),
+          ('Complex128', 'complex128'),
+          ]
+
+engines = [('', 'IndexEngine'), ('Masked', 'MaskedIndexEngine')]
+
+}}
+
+{{for name, dtype in dtypes}}
+
+{{for prefix, engine in engines}}
+
+cdef class {{prefix}}{{name}}Engine({{engine}}):
+
+    cdef _make_hash_table(self, Py_ssize_t n):
+    {{if engine == 'MaskedIndexEngine'}}
+        return _hash.{{name}}HashTable(n, uses_mask=True)
+    {{else}}
+        return _hash.{{name}}HashTable(n)
+    {{endif}}
+
+    cdef _check_type(self, object val):
+    {{if engine == 'MaskedIndexEngine'}}
+        if val is C_NA:
+            return val
+    {{endif}}
+    {{if name not in {'Float64', 'Float32', 'Complex64', 'Complex128'} }}
+        if not util.is_integer_object(val):
+            if util.is_float_object(val):
+                # Make sure Int64Index.get_loc(2.0) works
+                if val.is_integer():
+                    return int(val)
+            raise KeyError(val)
+        {{if name.startswith("U")}}
+        if val < 0:
+            # cannot have negative values with unsigned int dtype
+            raise KeyError(val)
+        {{endif}}
+    {{elif name not in {'Complex64', 'Complex128'} }}
+        if not util.is_integer_object(val) and not util.is_float_object(val):
+            # in particular catch bool and avoid casting True -> 1.0
+            raise KeyError(val)
+    {{else}}
+        if (not util.is_integer_object(val)
+            and not util.is_float_object(val)
+            and not util.is_complex_object(val)
+        ):
+            # in particular catch bool and avoid casting True -> 1.0
+            raise KeyError(val)
+    {{endif}}
+        return val
+
+{{endfor}}
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/indexing.pyi b/pandas/pandas/_libs/indexing.pyi
new file mode 100644
index 0000000000000..3ae5c5044a2f7
--- /dev/null
+++ b/pandas/pandas/_libs/indexing.pyi
@@ -0,0 +1,17 @@
+from typing import (
+    Generic,
+    TypeVar,
+)
+
+from pandas.core.indexing import IndexingMixin
+
+_IndexingMixinT = TypeVar("_IndexingMixinT", bound=IndexingMixin)
+
+class NDFrameIndexerBase(Generic[_IndexingMixinT]):
+    name: str
+    # in practice obj is either a DataFrame or a Series
+    obj: _IndexingMixinT
+
+    def __init__(self, name: str, obj: _IndexingMixinT) -> None: ...
+    @property
+    def ndim(self) -> int: ...
diff --git a/pandas/pandas/_libs/indexing.pyx b/pandas/pandas/_libs/indexing.pyx
new file mode 100644
index 0000000000000..c274b28b7559a
--- /dev/null
+++ b/pandas/pandas/_libs/indexing.pyx
@@ -0,0 +1,28 @@
+cdef class NDFrameIndexerBase:
+    """
+    A base class for _NDFrameIndexer for fast instantiation and attribute access.
+    """
+    cdef:
+        Py_ssize_t _ndim
+
+    cdef public:
+        str name
+        object obj
+
+    def __init__(self, name: str, obj):
+        self.obj = obj
+        self.name = name
+        self._ndim = -1
+
+    @property
+    def ndim(self) -> int:
+        # Delay `ndim` instantiation until required as reading it
+        # from `obj` isn't entirely cheap.
+        ndim = self._ndim
+        if ndim == -1:
+            ndim = self._ndim = self.obj.ndim
+            if ndim > 2:
+                raise ValueError(  # pragma: no cover
+                    "NDFrameIndexer does not support NDFrame objects with ndim > 2"
+                )
+        return ndim
diff --git a/pandas/pandas/_libs/internals.pyi b/pandas/pandas/_libs/internals.pyi
new file mode 100644
index 0000000000000..ffe6c7730bcdc
--- /dev/null
+++ b/pandas/pandas/_libs/internals.pyi
@@ -0,0 +1,94 @@
+from typing import (
+    Iterator,
+    Sequence,
+    final,
+    overload,
+)
+import weakref
+
+import numpy as np
+
+from pandas._typing import (
+    ArrayLike,
+    Self,
+    npt,
+)
+
+from pandas import Index
+from pandas.core.internals.blocks import Block as B
+
+def slice_len(slc: slice, objlen: int = ...) -> int: ...
+def get_concat_blkno_indexers(
+    blknos_list: list[npt.NDArray[np.intp]],
+) -> list[tuple[npt.NDArray[np.intp], BlockPlacement]]: ...
+def get_blkno_indexers(
+    blknos: np.ndarray,  # int64_t[:]
+    group: bool = ...,
+) -> list[tuple[int, slice | np.ndarray]]: ...
+def get_blkno_placements(
+    blknos: np.ndarray,
+    group: bool = ...,
+) -> Iterator[tuple[int, BlockPlacement]]: ...
+def update_blklocs_and_blknos(
+    blklocs: npt.NDArray[np.intp],
+    blknos: npt.NDArray[np.intp],
+    loc: int,
+    nblocks: int,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+@final
+class BlockPlacement:
+    def __init__(self, val: int | slice | np.ndarray) -> None: ...
+    @property
+    def indexer(self) -> np.ndarray | slice: ...
+    @property
+    def as_array(self) -> np.ndarray: ...
+    @property
+    def as_slice(self) -> slice: ...
+    @property
+    def is_slice_like(self) -> bool: ...
+    @overload
+    def __getitem__(
+        self, loc: slice | Sequence[int] | npt.NDArray[np.intp]
+    ) -> BlockPlacement: ...
+    @overload
+    def __getitem__(self, loc: int) -> int: ...
+    def __iter__(self) -> Iterator[int]: ...
+    def __len__(self) -> int: ...
+    def delete(self, loc) -> BlockPlacement: ...
+    def add(self, other) -> BlockPlacement: ...
+    def append(self, others: list[BlockPlacement]) -> BlockPlacement: ...
+    def tile_for_unstack(self, factor: int) -> npt.NDArray[np.intp]: ...
+
+class Block:
+    _mgr_locs: BlockPlacement
+    ndim: int
+    values: ArrayLike
+    refs: BlockValuesRefs
+    def __init__(
+        self,
+        values: ArrayLike,
+        placement: BlockPlacement,
+        ndim: int,
+        refs: BlockValuesRefs | None = ...,
+    ) -> None: ...
+    def slice_block_rows(self, slicer: slice) -> Self: ...
+
+class BlockManager:
+    blocks: tuple[B, ...]
+    axes: list[Index]
+    _known_consolidated: bool
+    _is_consolidated: bool
+    _blknos: np.ndarray
+    _blklocs: np.ndarray
+    def __init__(
+        self, blocks: tuple[B, ...], axes: list[Index], verify_integrity=...
+    ) -> None: ...
+    def get_slice(self, slobj: slice, axis: int = ...) -> Self: ...
+    def _rebuild_blknos_and_blklocs(self) -> None: ...
+
+class BlockValuesRefs:
+    referenced_blocks: list[weakref.ref]
+    def __init__(self, blk: Block | None = ...) -> None: ...
+    def add_reference(self, blk: Block) -> None: ...
+    def add_index_reference(self, index: Index) -> None: ...
+    def has_reference(self) -> bool: ...
diff --git a/pandas/pandas/_libs/internals.pyx b/pandas/pandas/_libs/internals.pyx
new file mode 100644
index 0000000000000..8caf7e5fb99b2
--- /dev/null
+++ b/pandas/pandas/_libs/internals.pyx
@@ -0,0 +1,998 @@
+from collections import defaultdict
+
+cimport cython
+from cpython.object cimport PyObject
+from cpython.pyport cimport PY_SSIZE_T_MAX
+from cpython.slice cimport PySlice_GetIndicesEx
+from cpython.weakref cimport PyWeakref_NewRef
+from cython cimport Py_ssize_t
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    NPY_INTP,
+    int64_t,
+    intp_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+from pandas._libs.algos import ensure_int64
+
+from pandas._libs.util cimport (
+    is_array,
+    is_integer_object,
+)
+
+include "free_threading_config.pxi"
+
+IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+    from cpython.ref cimport Py_DECREF
+    from cpython.weakref cimport PyWeakref_GetRef
+ELSE:
+    from cpython.weakref cimport PyWeakref_GetObject
+
+
+cdef extern from "Python.h":
+    PyObject* Py_None
+
+
+@cython.final
+@cython.freelist(32)
+cdef class BlockPlacement:
+    cdef:
+        slice _as_slice
+        ndarray _as_array  # Note: this still allows `None`; will be intp_t
+        bint _has_slice, _has_array, _is_known_slice_like
+
+    def __cinit__(self, val):
+        cdef:
+            slice slc
+
+        self._as_slice = None
+        self._as_array = None
+        self._has_slice = False
+        self._has_array = False
+
+        if is_integer_object(val):
+            slc = slice(val, val + 1, 1)
+            self._as_slice = slc
+            self._has_slice = True
+        elif isinstance(val, slice):
+            slc = slice_canonize(val)
+
+            if slc.start != slc.stop:
+                self._as_slice = slc
+                self._has_slice = True
+            else:
+                arr = np.empty(0, dtype=np.intp)
+                self._as_array = arr
+                self._has_array = True
+        else:
+            # Cython memoryview interface requires ndarray to be writeable.
+            if (
+                not is_array(val)
+                or not cnp.PyArray_ISWRITEABLE(val)
+                or (<ndarray>val).descr.type_num != cnp.NPY_INTP
+            ):
+                arr = np.require(val, dtype=np.intp, requirements="W")
+            else:
+                arr = val
+            # Caller is responsible for ensuring arr.ndim == 1
+            self._as_array = arr
+            self._has_array = True
+
+    def __str__(self) -> str:
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        if s is not None:
+            v = self._as_slice
+        else:
+            v = self._as_array
+
+        return f"{type(self).__name__}({v})"
+
+    def __repr__(self) -> str:
+        return str(self)
+
+    def __len__(self) -> int:
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        if s is not None:
+            return slice_len(s)
+        else:
+            return len(self._as_array)
+
+    def __iter__(self):
+        cdef:
+            slice s = self._ensure_has_slice()
+            Py_ssize_t start, stop, step, _
+
+        if s is not None:
+            start, stop, step, _ = slice_get_indices_ex(s)
+            return iter(range(start, stop, step))
+        else:
+            return iter(self._as_array)
+
+    @property
+    def as_slice(self) -> slice:
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        if s is not None:
+            return s
+        else:
+            raise TypeError("Not slice-like")
+
+    @property
+    def indexer(self):
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        if s is not None:
+            return s
+        else:
+            return self._as_array
+
+    @property
+    def as_array(self) -> np.ndarray:
+        cdef:
+            Py_ssize_t start, stop, _
+
+        if not self._has_array:
+            start, stop, step, _ = slice_get_indices_ex(self._as_slice)
+            # NOTE: this is the C-optimized equivalent of
+            #  `np.arange(start, stop, step, dtype=np.intp)`
+            self._as_array = cnp.PyArray_Arange(start, stop, step, NPY_INTP)
+            self._has_array = True
+
+        return self._as_array
+
+    @property
+    def is_slice_like(self) -> bool:
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        return s is not None
+
+    def __getitem__(self, loc):
+        cdef:
+            slice s = self._ensure_has_slice()
+
+        if s is not None:
+            val = slice_getitem(s, loc)
+        else:
+            val = self._as_array[loc]
+
+        if not isinstance(val, slice) and val.ndim == 0:
+            return val
+
+        return BlockPlacement(val)
+
+    def delete(self, loc) -> BlockPlacement:
+        return BlockPlacement(np.delete(self.as_array, loc, axis=0))
+
+    def append(self, others) -> BlockPlacement:
+        if not len(others):
+            return self
+
+        return BlockPlacement(
+            np.concatenate([self.as_array] + [o.as_array for o in others])
+        )
+
+    cdef BlockPlacement iadd(self, other):
+        cdef:
+            slice s = self._ensure_has_slice()
+            Py_ssize_t other_int, start, stop, step
+
+        if is_integer_object(other) and s is not None:
+            other_int = <Py_ssize_t>other
+
+            if other_int == 0:
+                # BlockPlacement is treated as immutable
+                return self
+
+            start, stop, step, _ = slice_get_indices_ex(s)
+            start += other_int
+            stop += other_int
+
+            if (step > 0 and start < 0) or (step < 0 and stop < step):
+                raise ValueError("iadd causes length change")
+
+            if stop < 0:
+                val = slice(start, None, step)
+            else:
+                val = slice(start, stop, step)
+
+            return BlockPlacement(val)
+        else:
+            newarr = self.as_array + other
+            if (newarr < 0).any():
+                raise ValueError("iadd causes length change")
+
+            val = newarr
+            return BlockPlacement(val)
+
+    def add(self, other) -> BlockPlacement:
+        # We can get here with int or ndarray
+        return self.iadd(other)
+
+    cdef slice _ensure_has_slice(self):
+        if not self._has_slice:
+            self._as_slice = indexer_as_slice(self._as_array)
+            self._has_slice = True
+
+        return self._as_slice
+
+    cpdef BlockPlacement increment_above(self, Py_ssize_t loc):
+        """
+        Increment any entries of 'loc' or above by one.
+        """
+        cdef:
+            slice nv, s = self._ensure_has_slice()
+            Py_ssize_t start, stop, step
+            ndarray[intp_t, ndim=1] newarr
+
+        if s is not None:
+            # see if we are either all-above or all-below, each of which
+            #  have fastpaths available.
+
+            start, stop, step, _ = slice_get_indices_ex(s)
+
+            if start < loc and stop <= loc:
+                # We are entirely below, nothing to increment
+                return self
+
+            if start >= loc and stop >= loc:
+                # We are entirely above, we can efficiently increment out slice
+                nv = slice(start + 1, stop + 1, step)
+                return BlockPlacement(nv)
+
+        if loc == 0:
+            # fastpath where we know everything is >= 0
+            newarr = self.as_array + 1
+            return BlockPlacement(newarr)
+
+        newarr = self.as_array.copy()
+        newarr[newarr >= loc] += 1
+        return BlockPlacement(newarr)
+
+    def tile_for_unstack(self, factor: int) -> np.ndarray:
+        """
+        Find the new mgr_locs for the un-stacked version of a Block.
+        """
+        cdef:
+            slice slc = self._ensure_has_slice()
+            ndarray[intp_t, ndim=1] new_placement
+
+        if slc is not None and slc.step == 1:
+            new_slc = slice(slc.start * factor, slc.stop * factor, 1)
+            # equiv: np.arange(new_slc.start, new_slc.stop, dtype=np.intp)
+            new_placement = cnp.PyArray_Arange(new_slc.start, new_slc.stop, 1, NPY_INTP)
+        else:
+            # Note: test_pivot_table_empty_aggfunc gets here with `slc is not None`
+            mapped = [
+                # equiv: np.arange(x * factor, (x + 1) * factor, dtype=np.intp)
+                cnp.PyArray_Arange(x * factor, (x + 1) * factor, 1, NPY_INTP)
+                for x in self
+            ]
+            new_placement = np.concatenate(mapped)
+        return new_placement
+
+
+cdef slice slice_canonize(slice s):
+    """
+    Convert slice to canonical bounded form.
+    """
+    cdef:
+        Py_ssize_t start = 0, stop = 0, step = 1
+
+    if s.step is None:
+        step = 1
+    else:
+        step = <Py_ssize_t>s.step
+        if step == 0:
+            raise ValueError("slice step cannot be zero")
+
+    if step > 0:
+        if s.stop is None:
+            raise ValueError("unbounded slice")
+
+        stop = <Py_ssize_t>s.stop
+        if s.start is None:
+            start = 0
+        else:
+            start = <Py_ssize_t>s.start
+            if start > stop:
+                start = stop
+    elif step < 0:
+        if s.start is None:
+            raise ValueError("unbounded slice")
+
+        start = <Py_ssize_t>s.start
+        if s.stop is None:
+            stop = -1
+        else:
+            stop = <Py_ssize_t>s.stop
+            if stop > start:
+                stop = start
+
+    if start < 0 or (stop < 0 and s.stop is not None and step > 0):
+        raise ValueError("unbounded slice")
+
+    if stop < 0:
+        return slice(start, None, step)
+    else:
+        return slice(start, stop, step)
+
+
+cpdef Py_ssize_t slice_len(slice slc, Py_ssize_t objlen=PY_SSIZE_T_MAX) except -1:
+    """
+    Get length of a bounded slice.
+
+    The slice must not have any "open" bounds that would create dependency on
+    container size, i.e.:
+    - if ``s.step is None or s.step > 0``, ``s.stop`` is not ``None``
+    - if ``s.step < 0``, ``s.start`` is not ``None``
+
+    Otherwise, the result is unreliable.
+    """
+    cdef:
+        Py_ssize_t start, stop, step, length
+
+    if slc is None:
+        raise TypeError("slc must be slice")  # pragma: no cover
+
+    PySlice_GetIndicesEx(slc, objlen, &start, &stop, &step, &length)
+
+    return length
+
+
+cdef (Py_ssize_t, Py_ssize_t, Py_ssize_t, Py_ssize_t) slice_get_indices_ex(
+    slice slc, Py_ssize_t objlen=PY_SSIZE_T_MAX
+):
+    """
+    Get (start, stop, step, length) tuple for a slice.
+
+    If `objlen` is not specified, slice must be bounded, otherwise the result
+    will be wrong.
+    """
+    cdef:
+        Py_ssize_t start, stop, step, length
+
+    if slc is None:
+        raise TypeError("slc should be a slice")  # pragma: no cover
+
+    PySlice_GetIndicesEx(slc, objlen, &start, &stop, &step, &length)
+
+    return start, stop, step, length
+
+
+cdef slice_getitem(slice slc, ind):
+    cdef:
+        Py_ssize_t s_start, s_stop, s_step, s_len
+        Py_ssize_t ind_start, ind_stop, ind_step, ind_len
+
+    s_start, s_stop, s_step, s_len = slice_get_indices_ex(slc)
+
+    if isinstance(ind, slice):
+        ind_start, ind_stop, ind_step, ind_len = slice_get_indices_ex(ind, s_len)
+
+        if ind_step > 0 and ind_len == s_len:
+            # short-cut for no-op slice
+            if ind_len == s_len:
+                return slc
+
+        if ind_step < 0:
+            s_start = s_stop - s_step
+            ind_step = -ind_step
+
+        s_step *= ind_step
+        s_stop = s_start + ind_stop * s_step
+        s_start = s_start + ind_start * s_step
+
+        if s_step < 0 and s_stop < 0:
+            return slice(s_start, None, s_step)
+        else:
+            return slice(s_start, s_stop, s_step)
+
+    else:
+        # NOTE:
+        # this is the C-optimized equivalent of
+        # `np.arange(s_start, s_stop, s_step, dtype=np.intp)[ind]`
+        return cnp.PyArray_Arange(s_start, s_stop, s_step, NPY_INTP)[ind]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cdef slice indexer_as_slice(intp_t[:] vals):
+    cdef:
+        Py_ssize_t i, n, start, stop
+        int64_t d
+
+    if vals is None:
+        raise TypeError("vals must be ndarray")  # pragma: no cover
+
+    n = vals.shape[0]
+
+    if n == 0 or vals[0] < 0:
+        return None
+
+    if n == 1:
+        return slice(vals[0], vals[0] + 1, 1)
+
+    if vals[1] < 0:
+        return None
+
+    # n > 2
+    d = vals[1] - vals[0]
+
+    if d == 0:
+        return None
+
+    for i in range(2, n):
+        if vals[i] < 0 or vals[i] - vals[i - 1] != d:
+            return None
+
+    start = vals[0]
+    stop = start + n * d
+    if stop < 0 and d < 0:
+        return slice(start, None, d)
+    else:
+        return slice(start, stop, d)
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def get_concat_blkno_indexers(list blknos_list not None):
+    """
+    Given the mgr.blknos for a list of mgrs, break range(len(mgrs[0])) into
+    slices such that within each slice blknos_list[i] is constant for each i.
+
+    This is a multi-Manager analogue to get_blkno_indexers with group=False.
+    """
+    # we have the blknos for each of several BlockManagers
+    # list[np.ndarray[int64_t]]
+    cdef:
+        Py_ssize_t i, j, k, start, ncols
+        cnp.npy_intp n_mgrs
+        ndarray[intp_t] blknos, cur_blknos, run_blknos
+        BlockPlacement bp
+        list result = []
+
+    n_mgrs = len(blknos_list)
+    cur_blknos = cnp.PyArray_EMPTY(1, &n_mgrs, cnp.NPY_INTP, 0)
+
+    blknos = blknos_list[0]
+    ncols = len(blknos)
+    if ncols == 0:
+        return []
+
+    start = 0
+    for i in range(n_mgrs):
+        blknos = blknos_list[i]
+        cur_blknos[i] = blknos[0]
+        assert len(blknos) == ncols
+
+    for i in range(1, ncols):
+        # For each column, we check if the Block it is part of (i.e. blknos[i])
+        #  is the same the previous column (i.e. blknos[i-1]) *and* if this is
+        #  the case for each blknos in blknos_list.  If not, we start a new "run".
+        for k in range(n_mgrs):
+            blknos = blknos_list[k]
+            # assert cur_blknos[k] == blknos[i - 1]
+
+            if blknos[i] != blknos[i - 1]:
+                bp = BlockPlacement(slice(start, i))
+                run_blknos = cnp.PyArray_Copy(cur_blknos)
+                result.append((run_blknos, bp))
+
+                start = i
+                for j in range(n_mgrs):
+                    blknos = blknos_list[j]
+                    cur_blknos[j] = blknos[i]
+                break  # break out of `for k in range(n_mgrs)` loop
+
+    if start != ncols:
+        bp = BlockPlacement(slice(start, ncols))
+        run_blknos = cnp.PyArray_Copy(cur_blknos)
+        result.append((run_blknos, bp))
+    return result
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def get_blkno_indexers(
+    const int64_t[:] blknos, bint group=True
+) -> list[tuple[int, slice | np.ndarray]]:
+    """
+    Enumerate contiguous runs of integers in ndarray.
+
+    Iterate over elements of `blknos` yielding ``(blkno, slice(start, stop))``
+    pairs for each contiguous run found.
+
+    If `group` is True and there is more than one run for a certain blkno,
+    ``(blkno, array)`` with an array containing positions of all elements equal
+    to blkno.
+
+    Returns
+    -------
+    list[tuple[int, slice | np.ndarray]]
+    """
+    # There's blkno in this function's name because it's used in block &
+    # blockno handling.
+    cdef:
+        int64_t cur_blkno
+        Py_ssize_t i, start, stop, n, diff
+        cnp.npy_intp tot_len
+        int64_t blkno
+        object group_dict = defaultdict(list)
+        ndarray[int64_t, ndim=1] arr
+
+    n = blknos.shape[0]
+    result = list()
+
+    if n == 0:
+        return result
+
+    start = 0
+    cur_blkno = blknos[start]
+
+    if group is False:
+        for i in range(1, n):
+            if blknos[i] != cur_blkno:
+                result.append((cur_blkno, slice(start, i)))
+
+                start = i
+                cur_blkno = blknos[i]
+
+        result.append((cur_blkno, slice(start, n)))
+    else:
+        for i in range(1, n):
+            if blknos[i] != cur_blkno:
+                group_dict[cur_blkno].append((start, i))
+
+                start = i
+                cur_blkno = blknos[i]
+
+        group_dict[cur_blkno].append((start, n))
+
+        for blkno, slices in group_dict.items():
+            if len(slices) == 1:
+                result.append((blkno, slice(slices[0][0], slices[0][1])))
+            else:
+                tot_len = sum(stop - start for start, stop in slices)
+                # equiv np.empty(tot_len, dtype=np.int64)
+                arr = cnp.PyArray_EMPTY(1, &tot_len, cnp.NPY_INT64, 0)
+
+                i = 0
+                for start, stop in slices:
+                    for diff in range(start, stop):
+                        arr[i] = diff
+                        i += 1
+
+                result.append((blkno, arr))
+
+    return result
+
+
+def get_blkno_placements(blknos, group: bool = True):
+    """
+    Parameters
+    ----------
+    blknos : np.ndarray[int64]
+    group : bool, default True
+
+    Returns
+    -------
+    iterator
+        yield (blkno, BlockPlacement)
+    """
+    blknos = ensure_int64(blknos)
+
+    for blkno, indexer in get_blkno_indexers(blknos, group):
+        yield blkno, BlockPlacement(indexer)
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cpdef update_blklocs_and_blknos(
+    const intp_t[:] blklocs,
+    const intp_t[:] blknos,
+    Py_ssize_t loc,
+    intp_t nblocks,
+):
+    """
+    Update blklocs and blknos when a new column is inserted at 'loc'.
+    """
+    cdef:
+        Py_ssize_t i
+        cnp.npy_intp length = blklocs.shape[0] + 1
+        ndarray[intp_t, ndim=1] new_blklocs, new_blknos
+
+    # equiv: new_blklocs = np.empty(length, dtype=np.intp)
+    new_blklocs = cnp.PyArray_EMPTY(1, &length, cnp.NPY_INTP, 0)
+    new_blknos = cnp.PyArray_EMPTY(1, &length, cnp.NPY_INTP, 0)
+
+    for i in range(loc):
+        new_blklocs[i] = blklocs[i]
+        new_blknos[i] = blknos[i]
+
+    new_blklocs[loc] = 0
+    new_blknos[loc] = nblocks
+
+    for i in range(loc, length - 1):
+        new_blklocs[i + 1] = blklocs[i]
+        new_blknos[i + 1] = blknos[i]
+
+    return new_blklocs, new_blknos
+
+
+def _unpickle_block(values, placement, ndim):
+    # We have to do some gymnastics b/c "ndim" is keyword-only
+
+    from pandas.core.internals.blocks import (
+        maybe_coerce_values,
+        new_block,
+    )
+    values = maybe_coerce_values(values)
+
+    if not isinstance(placement, BlockPlacement):
+        placement = BlockPlacement(placement)
+    return new_block(values, placement, ndim=ndim)
+
+
+@cython.freelist(64)
+cdef class Block:
+    """
+    Defining __init__ in a cython class significantly improves performance.
+    """
+    cdef:
+        public BlockPlacement _mgr_locs
+        public BlockValuesRefs refs
+        readonly int ndim
+        # 2023-08-15 no apparent performance improvement from declaring values
+        #  as ndarray in a type-special subclass (similar for NDArrayBacked).
+        #  This might change if slice_block_rows can be optimized with something
+        #  like https://github.com/numpy/numpy/issues/23934
+        public object values
+
+    def __cinit__(
+        self,
+        values,
+        placement: BlockPlacement,
+        ndim: int,
+        refs: BlockValuesRefs | None = None,
+    ):
+        """
+        Parameters
+        ----------
+        values : np.ndarray or ExtensionArray
+            We assume maybe_coerce_values has already been called.
+        placement : BlockPlacement
+        ndim : int
+            1 for SingleBlockManager/Series, 2 for BlockManager/DataFrame
+        refs: BlockValuesRefs, optional
+            Ref tracking object or None if block does not have any refs.
+        """
+        self.values = values
+
+        self._mgr_locs = placement
+        self.ndim = ndim
+        if refs is None:
+            # if no refs are passed, that means we are creating a Block from
+            # new values that it uniquely owns -> start a new BlockValuesRefs
+            # object that only references this block
+            self.refs = BlockValuesRefs(self)
+        else:
+            # if refs are passed, this is the BlockValuesRefs object that is shared
+            # with the parent blocks which share the values, and a reference to this
+            # new block is added
+            refs.add_reference(self)
+            self.refs = refs
+
+    cpdef __reduce__(self):
+        args = (self.values, self.mgr_locs.indexer, self.ndim)
+        return _unpickle_block, args
+
+    cpdef __setstate__(self, state):
+        from pandas.core.construction import extract_array
+
+        self.mgr_locs = BlockPlacement(state[0])
+        self.values = extract_array(state[1], extract_numpy=True)
+        if len(state) > 2:
+            # we stored ndim
+            self.ndim = state[2]
+        else:
+            # older pickle
+            from pandas.core.internals.api import maybe_infer_ndim
+
+            ndim = maybe_infer_ndim(self.values, self.mgr_locs)
+            self.ndim = ndim
+
+    cpdef Block slice_block_rows(self, slice slicer):
+        """
+        Perform __getitem__-like specialized to slicing along index.
+
+        Assumes self.ndim == 2
+        """
+        new_values = self.values[..., slicer]
+        return type(self)(new_values, self._mgr_locs, ndim=self.ndim, refs=self.refs)
+
+
+@cython.freelist(64)
+cdef class BlockManager:
+    cdef:
+        public tuple blocks
+        public list axes
+        public bint _known_consolidated, _is_consolidated
+        public ndarray _blknos, _blklocs
+
+    def __cinit__(
+        self,
+        blocks=None,
+        axes=None,
+        verify_integrity=True,
+    ):
+        # None as defaults for unpickling GH#42345
+        if blocks is None:
+            # This adds 1-2 microseconds to DataFrame(np.array([]))
+            return
+
+        if isinstance(blocks, list):
+            # Backward compat for e.g. pyarrow
+            blocks = tuple(blocks)
+
+        self.blocks = blocks
+        self.axes = axes.copy()  # copy to make sure we are not remotely-mutable
+
+        # Populate known_consolidate, blknos, and blklocs lazily
+        self._known_consolidated = False
+        self._is_consolidated = False
+        self._blknos = None
+        self._blklocs = None
+
+    # -------------------------------------------------------------------
+    # Block Placement
+
+    cpdef _rebuild_blknos_and_blklocs(self):
+        """
+        Update mgr._blknos / mgr._blklocs.
+        """
+        cdef:
+            intp_t blkno, i, j
+            cnp.npy_intp length = self.shape[0]
+            Block blk
+            BlockPlacement bp
+            ndarray[intp_t, ndim=1] new_blknos, new_blklocs
+
+        # equiv: np.empty(length, dtype=np.intp)
+        new_blknos = cnp.PyArray_EMPTY(1, &length, cnp.NPY_INTP, 0)
+        new_blklocs = cnp.PyArray_EMPTY(1, &length, cnp.NPY_INTP, 0)
+        # equiv: new_blknos.fill(-1)
+        cnp.PyArray_FILLWBYTE(new_blknos, -1)
+        cnp.PyArray_FILLWBYTE(new_blklocs, -1)
+
+        for blkno, blk in enumerate(self.blocks):
+            bp = blk._mgr_locs
+            # Iterating over `bp` is a faster equivalent to
+            #  new_blknos[bp.indexer] = blkno
+            #  new_blklocs[bp.indexer] = np.arange(len(bp))
+            for i, j in enumerate(bp):
+                new_blknos[j] = blkno
+                new_blklocs[j] = i
+
+        for i in range(length):
+            # faster than `for blkno in new_blknos`
+            #  https://github.com/cython/cython/issues/4393
+            blkno = new_blknos[i]
+
+            # If there are any -1s remaining, this indicates that our mgr_locs
+            #  are invalid.
+            if blkno == -1:
+                raise AssertionError("Gaps in blk ref_locs")
+
+        self._blknos = new_blknos
+        self._blklocs = new_blklocs
+
+    # -------------------------------------------------------------------
+    # Pickle
+
+    cpdef __reduce__(self):
+        if len(self.axes) == 1:
+            # SingleBlockManager, __init__ expects Block, axis
+            args = (self.blocks[0], self.axes[0])
+        else:
+            args = (self.blocks, self.axes)
+        return type(self), args
+
+    cpdef __setstate__(self, state):
+        from pandas.core.construction import extract_array
+        from pandas.core.internals.blocks import (
+            ensure_block_shape,
+            maybe_coerce_values,
+            new_block,
+        )
+        from pandas.core.internals.managers import ensure_index
+
+        if isinstance(state, tuple) and len(state) >= 4 and "0.14.1" in state[3]:
+            state = state[3]["0.14.1"]
+            axes = [ensure_index(ax) for ax in state["axes"]]
+            ndim = len(axes)
+
+            for blk in state["blocks"]:
+                vals = blk["values"]
+                # older versions may hold e.g. DatetimeIndex instead of DTA
+                vals = extract_array(vals, extract_numpy=True)
+                blk["values"] = maybe_coerce_values(ensure_block_shape(vals, ndim=ndim))
+
+                if not isinstance(blk["mgr_locs"], BlockPlacement):
+                    blk["mgr_locs"] = BlockPlacement(blk["mgr_locs"])
+
+            nbs = [
+                new_block(blk["values"], blk["mgr_locs"], ndim=ndim)
+                for blk in state["blocks"]
+            ]
+            blocks = tuple(nbs)
+            self.blocks = blocks
+            self.axes = axes
+
+        else:  # pragma: no cover
+            raise NotImplementedError("pre-0.14.1 pickles are no longer supported")
+
+        self._post_setstate()
+
+    def _post_setstate(self) -> None:
+        self._is_consolidated = False
+        self._known_consolidated = False
+        self._rebuild_blknos_and_blklocs()
+
+    # -------------------------------------------------------------------
+    # Indexing
+
+    cdef BlockManager _slice_mgr_rows(self, slice slobj):
+        cdef:
+            Block blk, nb
+            BlockManager mgr
+            ndarray blknos, blklocs
+
+        nbs = []
+        for blk in self.blocks:
+            nb = blk.slice_block_rows(slobj)
+            nbs.append(nb)
+
+        new_axes = [self.axes[0], self.axes[1]._getitem_slice(slobj)]
+        mgr = type(self)(tuple(nbs), new_axes, verify_integrity=False)
+
+        # We can avoid having to rebuild blklocs/blknos
+        blklocs = self._blklocs
+        blknos = self._blknos
+        if blknos is not None:
+            mgr._blknos = blknos.copy()
+            mgr._blklocs = blklocs.copy()
+        return mgr
+
+    def get_slice(self, slobj: slice, axis: int = 0) -> BlockManager:
+
+        if axis == 0:
+            new_blocks = self._slice_take_blocks_ax0(slobj)
+        elif axis == 1:
+            return self._slice_mgr_rows(slobj)
+        else:
+            raise IndexError("Requested axis not found in manager")
+
+        new_axes = list(self.axes)
+        new_axes[axis] = new_axes[axis]._getitem_slice(slobj)
+
+        return type(self)(tuple(new_blocks), new_axes, verify_integrity=False)
+
+
+cdef class BlockValuesRefs:
+    """Tracks all references to a given array.
+
+    Keeps track of all blocks (through weak references) that reference the same
+    data.
+    """
+    cdef:
+        public list referenced_blocks
+        public int clear_counter
+
+    def __cinit__(self, blk: Block | None = None) -> None:
+        if blk is not None:
+            self.referenced_blocks = [PyWeakref_NewRef(blk, None)]
+        else:
+            self.referenced_blocks = []
+        self.clear_counter = 500  # set reasonably high
+
+    cdef _clear_dead_references(self, bint force=False):
+        # Use exponential backoff to decide when we want to clear references
+        # if force=False. Clearing for every insertion causes slowdowns if
+        # all these objects stay alive, e.g. df.items() for wide DataFrames
+        # see GH#55245 and GH#55008
+        IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+            cdef PyObject* pobj
+            cdef bint status
+
+        if force or len(self.referenced_blocks) > self.clear_counter:
+            IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+                new_referenced_blocks = []
+                for ref in self.referenced_blocks:
+                    status = PyWeakref_GetRef(ref, &pobj)
+                    if status == -1:
+                        return
+                    elif status == 1:
+                        new_referenced_blocks.append(ref)
+                        Py_DECREF(<object>pobj)
+                self.referenced_blocks = new_referenced_blocks
+            ELSE:
+                self.referenced_blocks = [
+                    ref for ref in self.referenced_blocks
+                    if PyWeakref_GetObject(ref) != Py_None
+                ]
+
+            nr_of_refs = len(self.referenced_blocks)
+            if nr_of_refs < self.clear_counter // 2:
+                self.clear_counter = max(self.clear_counter // 2, 500)
+            elif nr_of_refs > self.clear_counter:
+                self.clear_counter = max(self.clear_counter * 2, nr_of_refs)
+
+    cpdef _add_reference_maybe_locked(self, Block blk):
+        self._clear_dead_references()
+        self.referenced_blocks.append(PyWeakref_NewRef(blk, None))
+
+    cpdef add_reference(self, Block blk):
+        """Adds a new reference to our reference collection.
+
+        Parameters
+        ----------
+        blk : Block
+            The block that the new references should point to.
+        """
+        IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+            with cython.critical_section(self):
+                self._add_reference_maybe_locked(blk)
+        ELSE:
+            self._add_reference_maybe_locked(blk)
+
+    def _add_index_reference_maybe_locked(self, index: object) -> None:
+        self._clear_dead_references()
+        self.referenced_blocks.append(PyWeakref_NewRef(index, None))
+
+    def add_index_reference(self, index: object) -> None:
+        """Adds a new reference to our reference collection when creating an index.
+
+        Parameters
+        ----------
+        index : Index
+            The index that the new reference should point to.
+        """
+        IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+            with cython.critical_section(self):
+                self._add_index_reference_maybe_locked(index)
+        ELSE:
+            self._add_index_reference_maybe_locked(index)
+
+    def _has_reference_maybe_locked(self) -> bool:
+        self._clear_dead_references(force=True)
+        # Checking for more references than block pointing to itself
+        return len(self.referenced_blocks) > 1
+
+    def has_reference(self) -> bool:
+        """Checks if block has foreign references.
+
+        A reference is only relevant if it is still alive. The reference to
+        ourselves does not count.
+
+        Returns
+        -------
+        bool
+        """
+        IF CYTHON_COMPATIBLE_WITH_FREE_THREADING:
+            with cython.critical_section(self):
+                return self._has_reference_maybe_locked()
+        ELSE:
+            return self._has_reference_maybe_locked()
diff --git a/pandas/pandas/_libs/interval.pyi b/pandas/pandas/_libs/interval.pyi
new file mode 100644
index 0000000000000..587fdf84f2f85
--- /dev/null
+++ b/pandas/pandas/_libs/interval.pyi
@@ -0,0 +1,174 @@
+from typing import (
+    Any,
+    Generic,
+    TypeVar,
+    overload,
+)
+
+import numpy as np
+import numpy.typing as npt
+
+from pandas._typing import (
+    IntervalClosedType,
+    Timedelta,
+    Timestamp,
+)
+
+VALID_CLOSED: frozenset[str]
+
+_OrderableScalarT = TypeVar("_OrderableScalarT", int, float)
+_OrderableTimesT = TypeVar("_OrderableTimesT", Timestamp, Timedelta)
+_OrderableT = TypeVar("_OrderableT", int, float, Timestamp, Timedelta)
+
+class _LengthDescriptor:
+    @overload
+    def __get__(
+        self, instance: Interval[_OrderableScalarT], owner: Any
+    ) -> _OrderableScalarT: ...
+    @overload
+    def __get__(
+        self, instance: Interval[_OrderableTimesT], owner: Any
+    ) -> Timedelta: ...
+
+class _MidDescriptor:
+    @overload
+    def __get__(self, instance: Interval[_OrderableScalarT], owner: Any) -> float: ...
+    @overload
+    def __get__(
+        self, instance: Interval[_OrderableTimesT], owner: Any
+    ) -> _OrderableTimesT: ...
+
+class IntervalMixin:
+    @property
+    def closed_left(self) -> bool: ...
+    @property
+    def closed_right(self) -> bool: ...
+    @property
+    def open_left(self) -> bool: ...
+    @property
+    def open_right(self) -> bool: ...
+    @property
+    def is_empty(self) -> bool: ...
+    def _check_closed_matches(self, other: IntervalMixin, name: str = ...) -> None: ...
+
+class Interval(IntervalMixin, Generic[_OrderableT]):
+    @property
+    def left(self: Interval[_OrderableT]) -> _OrderableT: ...
+    @property
+    def right(self: Interval[_OrderableT]) -> _OrderableT: ...
+    @property
+    def closed(self) -> IntervalClosedType: ...
+    mid: _MidDescriptor
+    length: _LengthDescriptor
+    def __init__(
+        self,
+        left: _OrderableT,
+        right: _OrderableT,
+        closed: IntervalClosedType = ...,
+    ) -> None: ...
+    def __hash__(self) -> int: ...
+    @overload
+    def __contains__(
+        self: Interval[Timedelta], key: Timedelta | Interval[Timedelta]
+    ) -> bool: ...
+    @overload
+    def __contains__(
+        self: Interval[Timestamp], key: Timestamp | Interval[Timestamp]
+    ) -> bool: ...
+    @overload
+    def __contains__(
+        self: Interval[_OrderableScalarT],
+        key: _OrderableScalarT | Interval[_OrderableScalarT],
+    ) -> bool: ...
+    @overload
+    def __add__(
+        self: Interval[_OrderableTimesT], y: Timedelta
+    ) -> Interval[_OrderableTimesT]: ...
+    @overload
+    def __add__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __add__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __radd__(
+        self: Interval[_OrderableTimesT], y: Timedelta
+    ) -> Interval[_OrderableTimesT]: ...
+    @overload
+    def __radd__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __radd__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __sub__(
+        self: Interval[_OrderableTimesT], y: Timedelta
+    ) -> Interval[_OrderableTimesT]: ...
+    @overload
+    def __sub__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __sub__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __rsub__(
+        self: Interval[_OrderableTimesT], y: Timedelta
+    ) -> Interval[_OrderableTimesT]: ...
+    @overload
+    def __rsub__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __rsub__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __mul__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __mul__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __rmul__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __rmul__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __truediv__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __truediv__(self: Interval[float], y: float) -> Interval[float]: ...
+    @overload
+    def __floordiv__(
+        self: Interval[int], y: _OrderableScalarT
+    ) -> Interval[_OrderableScalarT]: ...
+    @overload
+    def __floordiv__(self: Interval[float], y: float) -> Interval[float]: ...
+    def overlaps(self: Interval[_OrderableT], other: Interval[_OrderableT]) -> bool: ...
+
+def intervals_to_interval_bounds(
+    intervals: np.ndarray, validate_closed: bool = ...
+) -> tuple[np.ndarray, np.ndarray, IntervalClosedType]: ...
+
+class IntervalTree(IntervalMixin):
+    def __init__(
+        self,
+        left: np.ndarray,
+        right: np.ndarray,
+        closed: IntervalClosedType = ...,
+        leaf_size: int = ...,
+    ) -> None: ...
+    @property
+    def mid(self) -> np.ndarray: ...
+    @property
+    def length(self) -> np.ndarray: ...
+    def get_indexer(self, target) -> npt.NDArray[np.intp]: ...
+    def get_indexer_non_unique(
+        self, target
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+    _na_count: int
+    @property
+    def is_overlapping(self) -> bool: ...
+    @property
+    def is_monotonic_increasing(self) -> bool: ...
+    def clear_mapping(self) -> None: ...
diff --git a/pandas/pandas/_libs/interval.pyx b/pandas/pandas/_libs/interval.pyx
new file mode 100644
index 0000000000000..5d0876591a151
--- /dev/null
+++ b/pandas/pandas/_libs/interval.pyx
@@ -0,0 +1,683 @@
+import numbers
+from operator import (
+    le,
+    lt,
+)
+
+from cpython.datetime cimport (
+    PyDelta_Check,
+    import_datetime,
+)
+
+import_datetime()
+
+cimport cython
+from cpython.object cimport PyObject_RichCompare
+from cython cimport Py_ssize_t
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    NPY_QUICKSORT,
+    PyArray_ArgSort,
+    PyArray_Take,
+    float64_t,
+    int64_t,
+    ndarray,
+    uint64_t,
+)
+
+cnp.import_array()
+
+
+from pandas._libs cimport util
+from pandas._libs.hashtable cimport Int64Vector
+from pandas._libs.tslibs.timedeltas cimport _Timedelta
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+from pandas._libs.tslibs.timezones cimport tz_compare
+from pandas._libs.tslibs.util cimport (
+    is_float_object,
+    is_integer_object,
+)
+
+VALID_CLOSED = frozenset(["left", "right", "both", "neither"])
+
+
+cdef class IntervalMixin:
+
+    @property
+    def closed_left(self):
+        """
+        Check if the interval is closed on the left side.
+
+        For the meaning of `closed` and `open` see :class:`~pandas.Interval`.
+
+        Returns
+        -------
+        bool
+            True if the Interval is closed on the left-side.
+
+        See Also
+        --------
+        Interval.closed_right : Check if the interval is closed on the right side.
+        Interval.open_left : Boolean inverse of closed_left.
+
+        Examples
+        --------
+        >>> iv = pd.Interval(0, 5, closed='left')
+        >>> iv.closed_left
+        True
+
+        >>> iv = pd.Interval(0, 5, closed='right')
+        >>> iv.closed_left
+        False
+        """
+        return self.closed in ("left", "both")
+
+    @property
+    def closed_right(self):
+        """
+        Check if the interval is closed on the right side.
+
+        For the meaning of `closed` and `open` see :class:`~pandas.Interval`.
+
+        Returns
+        -------
+        bool
+            True if the Interval is closed on the left-side.
+
+        See Also
+        --------
+        Interval.closed_left : Check if the interval is closed on the left side.
+        Interval.open_right : Boolean inverse of closed_right.
+
+        Examples
+        --------
+        >>> iv = pd.Interval(0, 5, closed='both')
+        >>> iv.closed_right
+        True
+
+        >>> iv = pd.Interval(0, 5, closed='left')
+        >>> iv.closed_right
+        False
+        """
+        return self.closed in ("right", "both")
+
+    @property
+    def open_left(self):
+        """
+        Check if the interval is open on the left side.
+
+        For the meaning of `closed` and `open` see :class:`~pandas.Interval`.
+
+        Returns
+        -------
+        bool
+            True if the Interval is not closed on the left-side.
+
+        See Also
+        --------
+        Interval.open_right : Check if the interval is open on the right side.
+        Interval.closed_left : Boolean inverse of open_left.
+
+        Examples
+        --------
+        >>> iv = pd.Interval(0, 5, closed='neither')
+        >>> iv.open_left
+        True
+
+        >>> iv = pd.Interval(0, 5, closed='both')
+        >>> iv.open_left
+        False
+        """
+        return not self.closed_left
+
+    @property
+    def open_right(self):
+        """
+        Check if the interval is open on the right side.
+
+        For the meaning of `closed` and `open` see :class:`~pandas.Interval`.
+
+        Returns
+        -------
+        bool
+            True if the Interval is not closed on the left-side.
+
+        See Also
+        --------
+        Interval.open_left : Check if the interval is open on the left side.
+        Interval.closed_right : Boolean inverse of open_right.
+
+        Examples
+        --------
+        >>> iv = pd.Interval(0, 5, closed='left')
+        >>> iv.open_right
+        True
+
+        >>> iv = pd.Interval(0, 5)
+        >>> iv.open_right
+        False
+        """
+        return not self.closed_right
+
+    @property
+    def mid(self):
+        """
+        Return the midpoint of the Interval.
+
+        See Also
+        --------
+        Interval.left : Return the left bound for the interval.
+        Interval.right : Return the right bound for the interval.
+        Interval.length : Return the length of the interval.
+
+        Examples
+        --------
+        >>> iv = pd.Interval(0, 5)
+        >>> iv.mid
+        2.5
+        """
+        try:
+            return 0.5 * (self.left + self.right)
+        except TypeError:
+            # datetime safe version
+            return self.left + 0.5 * self.length
+
+    @property
+    def length(self):
+        """
+        Return the length of the Interval.
+
+        See Also
+        --------
+        Interval.is_empty : Indicates if an interval contains no points.
+
+        Examples
+        --------
+        >>> interval = pd.Interval(left=1, right=2, closed='left')
+        >>> interval
+        Interval(1, 2, closed='left')
+        >>> interval.length
+        1
+        """
+        return self.right - self.left
+
+    @property
+    def is_empty(self):
+        """
+        Indicates if an interval is empty, meaning it contains no points.
+
+        An interval is considered empty if its `left` and `right` endpoints
+        are equal, and it is not closed on both sides. This means that the
+        interval does not include any real points. In the case of an
+        :class:`pandas.arrays.IntervalArray` or :class:`IntervalIndex`, the
+        property returns a boolean array indicating the emptiness of each interval.
+
+        Returns
+        -------
+        bool or ndarray
+            A boolean indicating if a scalar :class:`Interval` is empty, or a
+            boolean ``ndarray`` positionally indicating if an ``Interval`` in
+            an :class:`~arrays.IntervalArray` or :class:`IntervalIndex` is
+            empty.
+
+        See Also
+        --------
+        Interval.length : Return the length of the Interval.
+
+        Examples
+        --------
+        An :class:`Interval` that contains points is not empty:
+
+        >>> pd.Interval(0, 1, closed='right').is_empty
+        False
+
+        An ``Interval`` that does not contain any points is empty:
+
+        >>> pd.Interval(0, 0, closed='right').is_empty
+        True
+        >>> pd.Interval(0, 0, closed='left').is_empty
+        True
+        >>> pd.Interval(0, 0, closed='neither').is_empty
+        True
+
+        An ``Interval`` that contains a single point is not empty:
+
+        >>> pd.Interval(0, 0, closed='both').is_empty
+        False
+
+        An :class:`~arrays.IntervalArray` or :class:`IntervalIndex` returns a
+        boolean ``ndarray`` positionally indicating if an ``Interval`` is
+        empty:
+
+        >>> ivs = [pd.Interval(0, 0, closed='neither'),
+        ...        pd.Interval(1, 2, closed='neither')]
+        >>> pd.arrays.IntervalArray(ivs).is_empty
+        array([ True, False])
+
+        Missing values are not considered empty:
+
+        >>> ivs = [pd.Interval(0, 0, closed='neither'), np.nan]
+        >>> pd.IntervalIndex(ivs).is_empty
+        array([ True, False])
+        """
+        return (self.right == self.left) & (self.closed != "both")
+
+    def _check_closed_matches(self, other, name="other"):
+        """
+        Check if the closed attribute of `other` matches.
+
+        Note that 'left' and 'right' are considered different from 'both'.
+
+        Parameters
+        ----------
+        other : Interval, IntervalIndex, IntervalArray
+        name : str
+            Name to use for 'other' in the error message.
+
+        Raises
+        ------
+        ValueError
+            When `other` is not closed exactly the same as self.
+        """
+        if self.closed != other.closed:
+            raise ValueError(f"'{name}.closed' is {repr(other.closed)}, "
+                             f"expected {repr(self.closed)}.")
+
+
+cdef bint _interval_like(other):
+    return (hasattr(other, "left")
+            and hasattr(other, "right")
+            and hasattr(other, "closed"))
+
+
+cdef class Interval(IntervalMixin):
+    """
+    Immutable object implementing an Interval, a bounded slice-like interval.
+
+    Attributes
+    ----------
+    left : orderable scalar
+        Left bound for the interval.
+    right : orderable scalar
+        Right bound for the interval.
+    closed : {'right', 'left', 'both', 'neither'}, default 'right'
+        Whether the interval is closed on the left-side, right-side, both or
+        neither. See the Notes for more detailed explanation.
+
+    See Also
+    --------
+    IntervalIndex : An Index of Interval objects that are all closed on the
+        same side.
+    cut : Convert continuous data into discrete bins (Categorical
+        of Interval objects).
+    qcut : Convert continuous data into bins (Categorical of Interval objects)
+        based on quantiles.
+    Period : Represents a period of time.
+
+    Notes
+    -----
+    The parameters `left` and `right` must be from the same type, you must be
+    able to compare them and they must satisfy ``left <= right``.
+
+    A closed interval (in mathematics denoted by square brackets) contains
+    its endpoints, i.e. the closed interval ``[0, 5]`` is characterized by the
+    conditions ``0 <= x <= 5``. This is what ``closed='both'`` stands for.
+    An open interval (in mathematics denoted by parentheses) does not contain
+    its endpoints, i.e. the open interval ``(0, 5)`` is characterized by the
+    conditions ``0 < x < 5``. This is what ``closed='neither'`` stands for.
+    Intervals can also be half-open or half-closed, i.e. ``[0, 5)`` is
+    described by ``0 <= x < 5`` (``closed='left'``) and ``(0, 5]`` is
+    described by ``0 < x <= 5`` (``closed='right'``).
+
+    Examples
+    --------
+    It is possible to build Intervals of different types, like numeric ones:
+
+    >>> iv = pd.Interval(left=0, right=5)
+    >>> iv
+    Interval(0, 5, closed='right')
+
+    You can check if an element belongs to it, or if it contains another interval:
+
+    >>> 2.5 in iv
+    True
+    >>> pd.Interval(left=2, right=5, closed='both') in iv
+    True
+
+    You can test the bounds (``closed='right'``, so ``0 < x <= 5``):
+
+    >>> 0 in iv
+    False
+    >>> 5 in iv
+    True
+    >>> 0.0001 in iv
+    True
+
+    Calculate its length
+
+    >>> iv.length
+    5
+
+    You can operate with `+` and `*` over an Interval and the operation
+    is applied to each of its bounds, so the result depends on the type
+    of the bound elements
+
+    >>> shifted_iv = iv + 3
+    >>> shifted_iv
+    Interval(3, 8, closed='right')
+    >>> extended_iv = iv * 10.0
+    >>> extended_iv
+    Interval(0.0, 50.0, closed='right')
+
+    To create a time interval you can use Timestamps as the bounds
+
+    >>> year_2017 = pd.Interval(pd.Timestamp('2017-01-01 00:00:00'),
+    ...                         pd.Timestamp('2018-01-01 00:00:00'),
+    ...                         closed='left')
+    >>> pd.Timestamp('2017-01-01 00:00') in year_2017
+    True
+    >>> year_2017.length
+    Timedelta('365 days 00:00:00')
+    """
+    _typ = "interval"
+    __array_priority__ = 1000
+
+    cdef readonly object left
+    """
+    Left bound for the interval.
+
+    See Also
+    --------
+    Interval.right : Return the right bound for the interval.
+    numpy.ndarray.left : A similar method in numpy for obtaining
+        the left endpoint(s) of intervals.
+
+    Examples
+    --------
+    >>> interval = pd.Interval(left=1, right=2, closed='left')
+    >>> interval
+    Interval(1, 2, closed='left')
+    >>> interval.left
+    1
+    """
+
+    cdef readonly object right
+    """
+    Right bound for the interval.
+
+    See Also
+    --------
+    Interval.left : Return the left bound for the interval.
+    numpy.ndarray.right : A similar method in numpy for obtaining
+        the right endpoint(s) of intervals.
+
+    Examples
+    --------
+    >>> interval = pd.Interval(left=1, right=2, closed='left')
+    >>> interval
+    Interval(1, 2, closed='left')
+    >>> interval.right
+    2
+    """
+
+    cdef readonly str closed
+    """
+    String describing the inclusive side the intervals.
+
+    Either ``left``, ``right``, ``both`` or ``neither``.
+
+    See Also
+    --------
+    Interval.closed_left : Check if the interval is closed on the left side.
+    Interval.closed_right : Check if the interval is closed on the right side.
+    Interval.open_left : Check if the interval is open on the left side.
+    Interval.open_right : Check if the interval is open on the right side.
+
+    Examples
+    --------
+    >>> interval = pd.Interval(left=1, right=2, closed='left')
+    >>> interval
+    Interval(1, 2, closed='left')
+    >>> interval.closed
+    'left'
+    """
+
+    def __init__(self, left, right, str closed="right"):
+        # note: it is faster to just do these checks than to use a special
+        # constructor (__cinit__/__new__) to avoid them
+
+        self._validate_endpoint(left)
+        self._validate_endpoint(right)
+
+        if closed not in VALID_CLOSED:
+            raise ValueError(f"invalid option for 'closed': {closed}")
+        if not left <= right:
+            raise ValueError("left side of interval must be <= right side")
+        if (isinstance(left, _Timestamp) and
+                not tz_compare(left.tzinfo, right.tzinfo)):
+            # GH 18538
+            raise ValueError("left and right must have the same time zone, got "
+                             f"{repr(left.tzinfo)}' and {repr(right.tzinfo)}")
+        self.left = left
+        self.right = right
+        self.closed = closed
+
+    def _validate_endpoint(self, endpoint):
+        # GH 23013
+        if not (is_integer_object(endpoint) or is_float_object(endpoint) or
+                isinstance(endpoint, (_Timestamp, _Timedelta))):
+            raise ValueError("Only numeric, Timestamp and Timedelta endpoints "
+                             "are allowed when constructing an Interval.")
+
+    def __hash__(self):
+        return hash((self.left, self.right, self.closed))
+
+    def __contains__(self, key) -> bool:
+        if _interval_like(key):
+            key_closed_left = key.closed in ("left", "both")
+            key_closed_right = key.closed in ("right", "both")
+            if self.open_left and key_closed_left:
+                left_contained = self.left < key.left
+            else:
+                left_contained = self.left <= key.left
+            if self.open_right and key_closed_right:
+                right_contained = key.right < self.right
+            else:
+                right_contained = key.right <= self.right
+            return left_contained and right_contained
+        return ((self.left < key if self.open_left else self.left <= key) and
+                (key < self.right if self.open_right else key <= self.right))
+
+    def __richcmp__(self, other, op: int):
+        if isinstance(other, Interval):
+            self_tuple = (self.left, self.right, self.closed)
+            other_tuple = (other.left, other.right, other.closed)
+            return PyObject_RichCompare(self_tuple, other_tuple, op)
+        elif util.is_array(other):
+            return np.array(
+                [PyObject_RichCompare(self, x, op) for x in other],
+                dtype=bool,
+            )
+
+        return NotImplemented
+
+    def __reduce__(self):
+        args = (self.left, self.right, self.closed)
+        return (type(self), args)
+
+    def __repr__(self) -> str:
+        disp = str if isinstance(self.left, (np.generic, _Timestamp)) else repr
+        name = type(self).__name__
+        repr_str = f"{name}({disp(self.left)}, {disp(self.right)}, closed={repr(self.closed)})"  # noqa: E501
+        return repr_str
+
+    def __str__(self) -> str:
+        start_symbol = "[" if self.closed_left else "("
+        end_symbol = "]" if self.closed_right else ")"
+        return f"{start_symbol}{self.left}, {self.right}{end_symbol}"
+
+    def __add__(self, y):
+        if (
+            isinstance(y, numbers.Number)
+            or PyDelta_Check(y)
+            or cnp.is_timedelta64_object(y)
+        ):
+            return Interval(self.left + y, self.right + y, closed=self.closed)
+        return NotImplemented
+
+    def __radd__(self, other):
+        if (
+                isinstance(other, numbers.Number)
+                or PyDelta_Check(other)
+                or cnp.is_timedelta64_object(other)
+        ):
+            return Interval(self.left + other, self.right + other, closed=self.closed)
+        return NotImplemented
+
+    def __sub__(self, y):
+        if (
+            isinstance(y, numbers.Number)
+            or PyDelta_Check(y)
+            or cnp.is_timedelta64_object(y)
+        ):
+            return Interval(self.left - y, self.right - y, closed=self.closed)
+        return NotImplemented
+
+    def __mul__(self, y):
+        if isinstance(y, numbers.Number):
+            return Interval(self.left * y, self.right * y, closed=self.closed)
+        return NotImplemented
+
+    def __rmul__(self, other):
+        if isinstance(other, numbers.Number):
+            return Interval(self.left * other, self.right * other, closed=self.closed)
+        return NotImplemented
+
+    def __truediv__(self, y):
+        if isinstance(y, numbers.Number):
+            return Interval(self.left / y, self.right / y, closed=self.closed)
+        return NotImplemented
+
+    def __floordiv__(self, y):
+        if isinstance(y, numbers.Number):
+            return Interval(
+                self.left // y, self.right // y, closed=self.closed)
+        return NotImplemented
+
+    def overlaps(self, other):
+        """
+        Check whether two Interval objects overlap.
+
+        Two intervals overlap if they share a common point, including closed
+        endpoints. Intervals that only have an open endpoint in common do not
+        overlap.
+
+        Parameters
+        ----------
+        other : Interval
+            Interval to check against for an overlap.
+
+        Returns
+        -------
+        bool
+            True if the two intervals overlap.
+
+        See Also
+        --------
+        IntervalArray.overlaps : The corresponding method for IntervalArray.
+        IntervalIndex.overlaps : The corresponding method for IntervalIndex.
+
+        Examples
+        --------
+        >>> i1 = pd.Interval(0, 2)
+        >>> i2 = pd.Interval(1, 3)
+        >>> i1.overlaps(i2)
+        True
+        >>> i3 = pd.Interval(4, 5)
+        >>> i1.overlaps(i3)
+        False
+
+        Intervals that share closed endpoints overlap:
+
+        >>> i4 = pd.Interval(0, 1, closed='both')
+        >>> i5 = pd.Interval(1, 2, closed='both')
+        >>> i4.overlaps(i5)
+        True
+
+        Intervals that only have an open endpoint in common do not overlap:
+
+        >>> i6 = pd.Interval(1, 2, closed='neither')
+        >>> i4.overlaps(i6)
+        False
+        """
+        if not isinstance(other, Interval):
+            raise TypeError("`other` must be an Interval, "
+                            f"got {type(other).__name__}")
+
+        # equality is okay if both endpoints are closed (overlap at a point)
+        op1 = le if (self.closed_left and other.closed_right) else lt
+        op2 = le if (other.closed_left and self.closed_right) else lt
+
+        # overlaps is equivalent negation of two interval being disjoint:
+        # disjoint = (A.left > B.right) or (B.left > A.right)
+        # (simplifying the negation allows this to be done in less operations)
+        return op1(self.left, other.right) and op2(other.left, self.right)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def intervals_to_interval_bounds(ndarray intervals, bint validate_closed=True):
+    """
+    Parameters
+    ----------
+    intervals : ndarray
+        Object array of Intervals / nulls.
+
+    validate_closed: bool, default True
+        Boolean indicating if all intervals must be closed on the same side.
+        Mismatching closed will raise if True, else return None for closed.
+
+    Returns
+    -------
+    tuple of
+        left : ndarray
+        right : ndarray
+        closed: IntervalClosedType
+    """
+    cdef:
+        object closed = None, interval
+        Py_ssize_t i, n = len(intervals)
+        ndarray left, right
+        bint seen_closed = False
+
+    left = np.empty(n, dtype=intervals.dtype)
+    right = np.empty(n, dtype=intervals.dtype)
+
+    for i in range(n):
+        interval = intervals[i]
+        if interval is None or util.is_nan(interval):
+            left[i] = np.nan
+            right[i] = np.nan
+            continue
+
+        if not isinstance(interval, Interval):
+            raise TypeError(f"type {type(interval)} with value "
+                            f"{interval} is not an interval")
+
+        left[i] = interval.left
+        right[i] = interval.right
+        if not seen_closed:
+            seen_closed = True
+            closed = interval.closed
+        elif closed != interval.closed:
+            closed = None
+            if validate_closed:
+                raise ValueError("intervals must all be closed on the same side")
+
+    return left, right, closed
+
+
+include "intervaltree.pxi"
diff --git a/pandas/pandas/_libs/intervaltree.pxi.in b/pandas/pandas/_libs/intervaltree.pxi.in
new file mode 100644
index 0000000000000..b94f60c272e5d
--- /dev/null
+++ b/pandas/pandas/_libs/intervaltree.pxi.in
@@ -0,0 +1,439 @@
+"""
+Template for intervaltree
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+from pandas._libs.algos import is_monotonic
+
+ctypedef fused int_scalar_t:
+    int64_t
+    float64_t
+
+ctypedef fused uint_scalar_t:
+    uint64_t
+    float64_t
+
+ctypedef fused scalar_t:
+    int_scalar_t
+    uint_scalar_t
+
+# ----------------------------------------------------------------------
+# IntervalTree
+# ----------------------------------------------------------------------
+
+cdef class IntervalTree(IntervalMixin):
+    """A centered interval tree
+
+    Based off the algorithm described on Wikipedia:
+    https://en.wikipedia.org/wiki/Interval_tree
+
+    we are emulating the IndexEngine interface
+    """
+    cdef readonly:
+        ndarray left, right
+        IntervalNode root
+        object dtype
+        str closed
+        object _is_overlapping, _left_sorter, _right_sorter
+        Py_ssize_t _na_count
+
+    def __init__(self, left, right, closed='right', leaf_size=100):
+        """
+        Parameters
+        ----------
+        left, right : np.ndarray[ndim=1]
+            Left and right bounds for each interval. Assumed to contain no
+            NaNs.
+        closed : {'left', 'right', 'both', 'neither'}, optional
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither. Defaults to 'right'.
+        leaf_size : int, optional
+            Parameter that controls when the tree switches from creating nodes
+            to brute-force search. Tune this parameter to optimize query
+            performance.
+        """
+        if closed not in ['left', 'right', 'both', 'neither']:
+            raise ValueError("invalid option for 'closed': %s" % closed)
+
+        left = np.asarray(left)
+        right = np.asarray(right)
+        self.dtype = np.result_type(left, right)
+        self.left = np.asarray(left, dtype=self.dtype)
+        self.right = np.asarray(right, dtype=self.dtype)
+
+        indices = np.arange(len(left), dtype='int64')
+
+        self.closed = closed
+
+        # GH 23352: ensure no nan in nodes
+        mask = ~np.isnan(self.left)
+        self._na_count = len(mask) - mask.sum()
+        self.left = self.left[mask]
+        self.right = self.right[mask]
+        indices = indices[mask]
+
+        node_cls = NODE_CLASSES[str(self.dtype), closed]
+        self.root = node_cls(self.left, self.right, indices, leaf_size)
+
+    @property
+    def left_sorter(self) -> np.ndarray:
+        """How to sort the left labels; this is used for binary search
+        """
+        if self._left_sorter is None:
+            values = [self.right, self.left]
+            self._left_sorter = np.lexsort(values)
+        return self._left_sorter
+
+    @property
+    def right_sorter(self) -> np.ndarray:
+        """How to sort the right labels
+        """
+        if self._right_sorter is None:
+            self._right_sorter = np.argsort(self.right)
+        return self._right_sorter
+
+    @property
+    def is_overlapping(self) -> bool:
+        """
+        Determine if the IntervalTree contains overlapping intervals.
+        Cached as self._is_overlapping.
+        """
+        if self._is_overlapping is not None:
+            return self._is_overlapping
+
+        # <= when both sides closed since endpoints can overlap
+        op = le if self.closed == 'both' else lt
+
+        # overlap if start of current interval < end of previous interval
+        # (current and previous in terms of sorted order by left/start side)
+        current = self.left[self.left_sorter[1:]]
+        previous = self.right[self.left_sorter[:-1]]
+        self._is_overlapping = bool(op(current, previous).any())
+
+        return self._is_overlapping
+
+    @property
+    def is_monotonic_increasing(self) -> bool:
+        """
+        Return True if the IntervalTree is monotonic increasing (only equal or
+        increasing values), else False
+        """
+        if self._na_count > 0:
+            return False
+
+        sort_order = self.left_sorter
+        return is_monotonic(sort_order, False)[0]
+
+    def get_indexer(self, ndarray[scalar_t, ndim=1] target) -> np.ndarray:
+        """Return the positions corresponding to unique intervals that overlap
+        with the given array of scalar targets.
+        """
+
+        # TODO: write get_indexer_intervals
+        cdef:
+            Py_ssize_t old_len
+            Py_ssize_t i
+            Int64Vector result
+
+        result = Int64Vector()
+        old_len = 0
+        for i in range(len(target)):
+            try:
+                self.root.query(result, target[i])
+            except OverflowError:
+                # overflow -> no match, which is already handled below
+                pass
+
+            if result.data.size == old_len:
+                result.append(-1)
+            elif result.data.size > old_len + 1:
+                raise KeyError(
+                    'indexer does not intersect a unique set of intervals')
+            old_len = result.data.size
+        return result.to_array().astype('intp')
+
+    def get_indexer_non_unique(self, ndarray[scalar_t, ndim=1] target):
+        """Return the positions corresponding to intervals that overlap with
+        the given array of scalar targets. Non-unique positions are repeated.
+        """
+        cdef:
+            Py_ssize_t old_len
+            Py_ssize_t i
+            Int64Vector result, missing
+
+        result = Int64Vector()
+        missing = Int64Vector()
+        old_len = 0
+        for i in range(len(target)):
+            try:
+                self.root.query(result, target[i])
+            except OverflowError:
+                # overflow -> no match, which is already handled below
+                pass
+
+            if result.data.size == old_len:
+                result.append(-1)
+                missing.append(i)
+            old_len = result.data.size
+        return (result.to_array().astype('intp'),
+                missing.to_array().astype('intp'))
+
+    def __repr__(self) -> str:
+        return ('<IntervalTree[{dtype},{closed}]: '
+                '{n_elements} elements>'.format(
+                    dtype=self.dtype, closed=self.closed,
+                    n_elements=self.root.n_elements))
+
+    # compat with IndexEngine interface
+    def clear_mapping(self) -> None:
+        pass
+
+
+cdef take(ndarray source, ndarray indices):
+    """Take the given positions from a 1D ndarray
+    """
+    return PyArray_Take(source, indices, 0)
+
+
+cdef sort_values_and_indices(all_values, all_indices, subset):
+    indices = take(all_indices, subset)
+    values = take(all_values, subset)
+    sorter = PyArray_ArgSort(values, 0, NPY_QUICKSORT)
+    sorted_values = take(values, sorter)
+    sorted_indices = take(indices, sorter)
+    return sorted_values, sorted_indices
+
+
+# ----------------------------------------------------------------------
+# Nodes
+# ----------------------------------------------------------------------
+
+@cython.internal
+cdef class IntervalNode:
+    cdef readonly:
+        int64_t n_elements, n_center, leaf_size
+        bint is_leaf_node
+
+    def __repr__(self) -> str:
+        if self.is_leaf_node:
+            return (
+                f"<{type(self).__name__}: {self.n_elements} elements (terminal)>"
+            )
+        else:
+            n_left = self.left_node.n_elements
+            n_right = self.right_node.n_elements
+            n_center = self.n_elements - n_left - n_right
+            return (
+                f"<{type(self).__name__}: "
+                f"pivot {self.pivot}, {self.n_elements} elements "
+                f"({n_left} left, {n_right} right, {n_center} overlapping)>"
+            )
+
+    def counts(self):
+        """
+        Inspect counts on this node
+        useful for debugging purposes
+        """
+        if self.is_leaf_node:
+            return self.n_elements
+        else:
+            m = len(self.center_left_values)
+            l = self.left_node.counts()
+            r = self.right_node.counts()
+            return (m, (l, r))
+
+
+# we need specialized nodes and leaves to optimize for different dtype and
+# closed values
+
+{{py:
+
+nodes = []
+for dtype in ['float64', 'int64', 'uint64']:
+    for closed, cmp_left, cmp_right in [
+        ('left', '<=', '<'),
+        ('right', '<', '<='),
+        ('both', '<=', '<='),
+        ('neither', '<', '<')]:
+        cmp_left_converse = '<' if cmp_left == '<=' else '<='
+        cmp_right_converse = '<' if cmp_right == '<=' else '<='
+        if dtype.startswith('int'):
+            fused_prefix = 'int_'
+        elif dtype.startswith('uint'):
+            fused_prefix = 'uint_'
+        elif dtype.startswith('float'):
+            fused_prefix = ''
+        nodes.append((dtype, dtype.title(),
+                      closed, closed.title(),
+                      cmp_left,
+                      cmp_right,
+                      cmp_left_converse,
+                      cmp_right_converse,
+                      fused_prefix))
+
+}}
+
+NODE_CLASSES = {}
+
+{{for dtype, dtype_title, closed, closed_title, cmp_left, cmp_right,
+      cmp_left_converse, cmp_right_converse, fused_prefix in nodes}}
+
+
+@cython.internal
+cdef class {{dtype_title}}Closed{{closed_title}}IntervalNode(IntervalNode):
+    """Non-terminal node for an IntervalTree
+
+    Categorizes intervals by those that fall to the left, those that fall to
+    the right, and those that overlap with the pivot.
+    """
+    cdef readonly:
+        {{dtype_title}}Closed{{closed_title}}IntervalNode left_node, right_node
+        {{dtype}}_t[:] center_left_values, center_right_values, left, right
+        int64_t[:] center_left_indices, center_right_indices, indices
+        {{dtype}}_t min_left, max_right
+        {{dtype}}_t pivot
+
+    def __init__(self,
+                 ndarray[{{dtype}}_t, ndim=1] left,
+                 ndarray[{{dtype}}_t, ndim=1] right,
+                 ndarray[int64_t, ndim=1] indices,
+                 int64_t leaf_size):
+
+        self.n_elements = len(left)
+        self.leaf_size = leaf_size
+
+        # min_left and min_right are used to speed-up query by skipping
+        # query on sub-nodes. If this node has size 0, query is cheap,
+        # so these values don't matter.
+        if left.size > 0:
+            self.min_left = left.min()
+            self.max_right = right.max()
+        else:
+            self.min_left = 0
+            self.max_right = 0
+
+        if self.n_elements <= leaf_size:
+            # make this a terminal (leaf) node
+            self.is_leaf_node = True
+            self.left = left
+            self.right = right
+            self.indices = indices
+            self.n_center = 0
+        else:
+            # calculate a pivot so we can create child nodes
+            self.is_leaf_node = False
+            self.pivot = np.median(left / 2 + right / 2)
+            if np.isinf(self.pivot):
+                self.pivot = cython.cast({{dtype}}_t, 0)
+                if self.pivot > np.max(right):
+                    self.pivot = np.max(left)
+                if self.pivot < np.min(left):
+                    self.pivot = np.min(right)
+
+            left_set, right_set, center_set = self.classify_intervals(
+                left, right)
+
+            self.left_node = self.new_child_node(left, right,
+                                                 indices, left_set)
+            self.right_node = self.new_child_node(left, right,
+                                                  indices, right_set)
+
+            self.center_left_values, self.center_left_indices = \
+                sort_values_and_indices(left, indices, center_set)
+            self.center_right_values, self.center_right_indices = \
+                sort_values_and_indices(right, indices, center_set)
+            self.n_center = len(self.center_left_indices)
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    cdef classify_intervals(self, {{dtype}}_t[:] left, {{dtype}}_t[:] right):
+        """Classify the given intervals based upon whether they fall to the
+        left, right, or overlap with this node's pivot.
+        """
+        cdef:
+            Int64Vector left_ind, right_ind, overlapping_ind
+            Py_ssize_t i
+
+        left_ind = Int64Vector()
+        right_ind = Int64Vector()
+        overlapping_ind = Int64Vector()
+
+        for i in range(self.n_elements):
+            if right[i] {{cmp_right_converse}} self.pivot:
+                left_ind.append(i)
+            elif self.pivot {{cmp_left_converse}} left[i]:
+                right_ind.append(i)
+            else:
+                overlapping_ind.append(i)
+
+        return (left_ind.to_array(),
+                right_ind.to_array(),
+                overlapping_ind.to_array())
+
+    cdef new_child_node(self,
+                        ndarray[{{dtype}}_t, ndim=1] left,
+                        ndarray[{{dtype}}_t, ndim=1] right,
+                        ndarray[int64_t, ndim=1] indices,
+                        ndarray[int64_t, ndim=1] subset):
+        """Create a new child node.
+        """
+        left = take(left, subset)
+        right = take(right, subset)
+        indices = take(indices, subset)
+        return {{dtype_title}}Closed{{closed_title}}IntervalNode(
+            left, right, indices, self.leaf_size)
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    @cython.initializedcheck(False)
+    cpdef query(self, Int64Vector result, {{fused_prefix}}scalar_t point):
+        """Recursively query this node and its sub-nodes for intervals that
+        overlap with the query point.
+        """
+
+        # GH 51826: ensures nan is handled properly during reindexing
+        if np.isnan(point):
+            return
+
+        cdef:
+            int64_t[:] indices
+            {{dtype}}_t[:] values
+            Py_ssize_t i
+
+        if self.is_leaf_node:
+            # Once we get down to a certain size, it doesn't make sense to
+            # continue the binary tree structure. Instead, we use linear
+            # search.
+            for i in range(self.n_elements):
+                if self.left[i] {{cmp_left}} point {{cmp_right}} self.right[i]:
+                    result.append(self.indices[i])
+        else:
+            # There are child nodes. Based on comparing our query to the pivot,
+            # look at the center values, then go to the relevant child.
+            if point < self.pivot:
+                values = self.center_left_values
+                indices = self.center_left_indices
+                for i in range(self.n_center):
+                    if not values[i] {{cmp_left}} point:
+                        break
+                    result.append(indices[i])
+                if point {{cmp_right}} self.left_node.max_right:
+                    self.left_node.query(result, point)
+            elif point > self.pivot:
+                values = self.center_right_values
+                indices = self.center_right_indices
+                for i in range(self.n_center - 1, -1, -1):
+                    if not point {{cmp_right}} values[i]:
+                        break
+                    result.append(indices[i])
+                if self.right_node.min_left {{cmp_left}} point:
+                    self.right_node.query(result, point)
+            else:
+                result.extend(self.center_left_indices)
+
+
+NODE_CLASSES['{{dtype}}',
+             '{{closed}}'] = {{dtype_title}}Closed{{closed_title}}IntervalNode
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/join.pyi b/pandas/pandas/_libs/join.pyi
new file mode 100644
index 0000000000000..1d4e8c90bc559
--- /dev/null
+++ b/pandas/pandas/_libs/join.pyi
@@ -0,0 +1,79 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def inner_join(
+    left: np.ndarray,  # const intp_t[:]
+    right: np.ndarray,  # const intp_t[:]
+    max_groups: int,
+    sort: bool = ...,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+def left_outer_join(
+    left: np.ndarray,  # const intp_t[:]
+    right: np.ndarray,  # const intp_t[:]
+    max_groups: int,
+    sort: bool = ...,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+def full_outer_join(
+    left: np.ndarray,  # const intp_t[:]
+    right: np.ndarray,  # const intp_t[:]
+    max_groups: int,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+def ffill_indexer(
+    indexer: np.ndarray,  # const intp_t[:]
+) -> npt.NDArray[np.intp]: ...
+def left_join_indexer_unique(
+    left: np.ndarray,  # ndarray[join_t]
+    right: np.ndarray,  # ndarray[join_t]
+) -> npt.NDArray[np.intp]: ...
+def left_join_indexer(
+    left: np.ndarray,  # ndarray[join_t]
+    right: np.ndarray,  # ndarray[join_t]
+) -> tuple[
+    np.ndarray,  # np.ndarray[join_t]
+    npt.NDArray[np.intp],
+    npt.NDArray[np.intp],
+]: ...
+def inner_join_indexer(
+    left: np.ndarray,  # ndarray[join_t]
+    right: np.ndarray,  # ndarray[join_t]
+) -> tuple[
+    np.ndarray,  # np.ndarray[join_t]
+    npt.NDArray[np.intp],
+    npt.NDArray[np.intp],
+]: ...
+def outer_join_indexer(
+    left: np.ndarray,  # ndarray[join_t]
+    right: np.ndarray,  # ndarray[join_t]
+) -> tuple[
+    np.ndarray,  # np.ndarray[join_t]
+    npt.NDArray[np.intp],
+    npt.NDArray[np.intp],
+]: ...
+def asof_join_backward_on_X_by_Y(
+    left_values: np.ndarray,  # ndarray[numeric_t]
+    right_values: np.ndarray,  # ndarray[numeric_t]
+    left_by_values: np.ndarray,  # const int64_t[:]
+    right_by_values: np.ndarray,  # const int64_t[:]
+    allow_exact_matches: bool = ...,
+    tolerance: np.number | float | None = ...,
+    use_hashtable: bool = ...,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+def asof_join_forward_on_X_by_Y(
+    left_values: np.ndarray,  # ndarray[numeric_t]
+    right_values: np.ndarray,  # ndarray[numeric_t]
+    left_by_values: np.ndarray,  # const int64_t[:]
+    right_by_values: np.ndarray,  # const int64_t[:]
+    allow_exact_matches: bool = ...,
+    tolerance: np.number | float | None = ...,
+    use_hashtable: bool = ...,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+def asof_join_nearest_on_X_by_Y(
+    left_values: np.ndarray,  # ndarray[numeric_t]
+    right_values: np.ndarray,  # ndarray[numeric_t]
+    left_by_values: np.ndarray,  # const int64_t[:]
+    right_by_values: np.ndarray,  # const int64_t[:]
+    allow_exact_matches: bool = ...,
+    tolerance: np.number | float | None = ...,
+    use_hashtable: bool = ...,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
diff --git a/pandas/pandas/_libs/join.pyx b/pandas/pandas/_libs/join.pyx
new file mode 100644
index 0000000000000..e6bd5a8de39dc
--- /dev/null
+++ b/pandas/pandas/_libs/join.pyx
@@ -0,0 +1,880 @@
+cimport cython
+from cython cimport Py_ssize_t
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    intp_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+from pandas._libs.algos import groupsort_indexer
+
+from pandas._libs.dtypes cimport (
+    numeric_object_t,
+    numeric_t,
+)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def inner_join(const intp_t[:] left, const intp_t[:] right,
+               Py_ssize_t max_groups, bint sort=True):
+    cdef:
+        Py_ssize_t i, j, k, count = 0
+        intp_t[::1] left_sorter, right_sorter
+        intp_t[::1] left_count, right_count
+        intp_t[::1] left_indexer, right_indexer
+        intp_t lc, rc
+        Py_ssize_t left_pos = 0, right_pos = 0, position = 0
+        Py_ssize_t offset
+
+    left_sorter, left_count = groupsort_indexer(left, max_groups)
+    right_sorter, right_count = groupsort_indexer(right, max_groups)
+
+    with nogil:
+        # First pass, determine size of result set, do not use the NA group
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc > 0 and lc > 0:
+                count += lc * rc
+
+    left_indexer = np.empty(count, dtype=np.intp)
+    right_indexer = np.empty(count, dtype=np.intp)
+
+    with nogil:
+        # exclude the NA group
+        left_pos = left_count[0]
+        right_pos = right_count[0]
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc > 0 and lc > 0:
+                for j in range(lc):
+                    offset = position + j * rc
+                    for k in range(rc):
+                        left_indexer[offset + k] = left_pos + j
+                        right_indexer[offset + k] = right_pos + k
+                position += lc * rc
+            left_pos += lc
+            right_pos += rc
+
+        # Will overwrite left/right indexer with the result
+        _get_result_indexer(left_sorter, left_indexer)
+        _get_result_indexer(right_sorter, right_indexer)
+
+    if not sort:
+        # if not asked to sort, revert to original order
+        if len(left) == len(left_indexer):
+            # no multiple matches for any row on the left
+            # this is a short-cut to avoid groupsort_indexer
+            # otherwise, the `else` path also works in this case
+            rev = np.empty(len(left), dtype=np.intp)
+            rev.put(np.asarray(left_sorter), np.arange(len(left)))
+        else:
+            rev, _ = groupsort_indexer(left_indexer, len(left))
+
+        return np.asarray(left_indexer).take(rev), np.asarray(right_indexer).take(rev)
+    else:
+        return np.asarray(left_indexer), np.asarray(right_indexer)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def left_outer_join(const intp_t[:] left, const intp_t[:] right,
+                    Py_ssize_t max_groups, bint sort=True):
+    cdef:
+        Py_ssize_t i, j, k, count = 0
+        ndarray[intp_t] rev
+        intp_t[::1] left_count, right_count
+        intp_t[::1] left_sorter, right_sorter
+        intp_t[::1] left_indexer, right_indexer
+        intp_t lc, rc
+        Py_ssize_t left_pos = 0, right_pos = 0, position = 0
+        Py_ssize_t offset
+
+    left_sorter, left_count = groupsort_indexer(left, max_groups)
+    right_sorter, right_count = groupsort_indexer(right, max_groups)
+
+    with nogil:
+        # First pass, determine size of result set, do not use the NA group
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc > 0:
+                count += lc * rc
+            else:
+                count += lc
+
+    left_indexer = np.empty(count, dtype=np.intp)
+    right_indexer = np.empty(count, dtype=np.intp)
+
+    with nogil:
+        # exclude the NA group
+        left_pos = left_count[0]
+        right_pos = right_count[0]
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc == 0:
+                for j in range(lc):
+                    left_indexer[position + j] = left_pos + j
+                    right_indexer[position + j] = -1
+                position += lc
+            else:
+                for j in range(lc):
+                    offset = position + j * rc
+                    for k in range(rc):
+                        left_indexer[offset + k] = left_pos + j
+                        right_indexer[offset + k] = right_pos + k
+                position += lc * rc
+            left_pos += lc
+            right_pos += rc
+
+        # Will overwrite left/right indexer with the result
+        _get_result_indexer(left_sorter, left_indexer)
+        _get_result_indexer(right_sorter, right_indexer)
+
+    if not sort:  # if not asked to sort, revert to original order
+        if len(left) == len(left_indexer):
+            # no multiple matches for any row on the left
+            # this is a short-cut to avoid groupsort_indexer
+            # otherwise, the `else` path also works in this case
+            rev = np.empty(len(left), dtype=np.intp)
+            rev.put(np.asarray(left_sorter), np.arange(len(left)))
+        else:
+            rev, _ = groupsort_indexer(left_indexer, len(left))
+
+        return np.asarray(left_indexer).take(rev), np.asarray(right_indexer).take(rev)
+    else:
+        return np.asarray(left_indexer), np.asarray(right_indexer)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def full_outer_join(const intp_t[:] left, const intp_t[:] right,
+                    Py_ssize_t max_groups):
+    cdef:
+        Py_ssize_t i, j, k, count = 0
+        intp_t[::1] left_sorter, right_sorter
+        intp_t[::1] left_count, right_count
+        intp_t[::1] left_indexer, right_indexer
+        intp_t lc, rc
+        intp_t left_pos = 0, right_pos = 0
+        Py_ssize_t offset, position = 0
+
+    left_sorter, left_count = groupsort_indexer(left, max_groups)
+    right_sorter, right_count = groupsort_indexer(right, max_groups)
+
+    with nogil:
+        # First pass, determine size of result set, do not use the NA group
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc > 0 and lc > 0:
+                count += lc * rc
+            else:
+                count += lc + rc
+
+    left_indexer = np.empty(count, dtype=np.intp)
+    right_indexer = np.empty(count, dtype=np.intp)
+
+    with nogil:
+        # exclude the NA group
+        left_pos = left_count[0]
+        right_pos = right_count[0]
+        for i in range(1, max_groups + 1):
+            lc = left_count[i]
+            rc = right_count[i]
+
+            if rc == 0:
+                for j in range(lc):
+                    left_indexer[position + j] = left_pos + j
+                    right_indexer[position + j] = -1
+                position += lc
+            elif lc == 0:
+                for j in range(rc):
+                    left_indexer[position + j] = -1
+                    right_indexer[position + j] = right_pos + j
+                position += rc
+            else:
+                for j in range(lc):
+                    offset = position + j * rc
+                    for k in range(rc):
+                        left_indexer[offset + k] = left_pos + j
+                        right_indexer[offset + k] = right_pos + k
+                position += lc * rc
+            left_pos += lc
+            right_pos += rc
+
+        # Will overwrite left/right indexer with the result
+        _get_result_indexer(left_sorter, left_indexer)
+        _get_result_indexer(right_sorter, right_indexer)
+
+    return np.asarray(left_indexer), np.asarray(right_indexer)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef void _get_result_indexer(
+    const intp_t[::1] sorter,
+    intp_t[::1] indexer,
+) noexcept nogil:
+    """NOTE: overwrites indexer with the result to avoid allocating another array"""
+    cdef:
+        Py_ssize_t i, n, idx
+
+    if len(sorter) > 0:
+        # cython-only equivalent to
+        #  `res = algos.take_nd(sorter, indexer, fill_value=-1)`
+        n = indexer.shape[0]
+        for i in range(n):
+            idx = indexer[i]
+            if idx == -1:
+                indexer[i] = -1
+            else:
+                indexer[i] = sorter[idx]
+    else:
+        # length-0 case
+        indexer[:] = -1
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def ffill_indexer(const intp_t[:] indexer) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, n = len(indexer)
+        ndarray[intp_t] result
+        intp_t val, last_obs
+
+    result = np.empty(n, dtype=np.intp)
+    last_obs = -1
+
+    for i in range(n):
+        val = indexer[i]
+        if val == -1:
+            result[i] = last_obs
+        else:
+            result[i] = val
+            last_obs = val
+
+    return result
+
+
+# ----------------------------------------------------------------------
+# left_join_indexer, inner_join_indexer, outer_join_indexer
+# ----------------------------------------------------------------------
+
+# Joins on ordered, unique indices
+
+# right might contain non-unique values
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def left_join_indexer_unique(
+    ndarray[numeric_object_t] left,
+    ndarray[numeric_object_t] right
+):
+    """
+    Both left and right are strictly monotonic increasing.
+    """
+    cdef:
+        Py_ssize_t i, j, nleft, nright
+        ndarray[intp_t] indexer
+        numeric_object_t rval
+
+    i = 0
+    j = 0
+    nleft = len(left)
+    nright = len(right)
+
+    indexer = np.empty(nleft, dtype=np.intp)
+    while True:
+        if i == nleft:
+            break
+
+        if j == nright:
+            indexer[i] = -1
+            i += 1
+            continue
+
+        rval = right[j]
+
+        while i < nleft - 1 and left[i] == rval:
+            indexer[i] = j
+            i += 1
+
+        if left[i] == rval:
+            indexer[i] = j
+            i += 1
+            while i < nleft - 1 and left[i] == rval:
+                indexer[i] = j
+                i += 1
+            j += 1
+        elif left[i] > rval:
+            indexer[i] = -1
+            j += 1
+        else:
+            indexer[i] = -1
+            i += 1
+    return indexer
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def left_join_indexer(ndarray[numeric_object_t] left, ndarray[numeric_object_t] right):
+    """
+    Two-pass algorithm for monotonic indexes. Handles many-to-one merges.
+
+    Both left and right are monotonic increasing, but at least one of them
+    is non-unique (if both were unique we'd use left_join_indexer_unique).
+    """
+    cdef:
+        Py_ssize_t i, j, nright, nleft, count
+        numeric_object_t lval, rval
+        ndarray[intp_t] lindexer, rindexer
+        ndarray[numeric_object_t] result
+
+    nleft = len(left)
+    nright = len(right)
+
+    # First pass is to find the size 'count' of our output indexers.
+    i = 0
+    j = 0
+    count = 0
+    if nleft > 0:
+        while i < nleft:
+            if j == nright:
+                count += nleft - i
+                break
+
+            lval = left[i]
+            rval = right[j]
+
+            if lval == rval:
+                # This block is identical across
+                #  left_join_indexer, inner_join_indexer, outer_join_indexer
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                count += 1
+                i += 1
+            else:
+                j += 1
+
+    # do it again now that result size is known
+
+    lindexer = np.empty(count, dtype=np.intp)
+    rindexer = np.empty(count, dtype=np.intp)
+    result = np.empty(count, dtype=left.dtype)
+
+    i = 0
+    j = 0
+    count = 0
+    if nleft > 0:
+        while i < nleft:
+            if j == nright:
+                while i < nleft:
+                    lindexer[count] = i
+                    rindexer[count] = -1
+                    result[count] = left[i]
+                    i += 1
+                    count += 1
+                break
+
+            lval = left[i]
+            rval = right[j]
+
+            if lval == rval:
+                lindexer[count] = i
+                rindexer[count] = j
+                result[count] = lval
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                # i.e. lval not in right; we keep for left_join_indexer
+                lindexer[count] = i
+                rindexer[count] = -1
+                result[count] = lval
+                count += 1
+                i += 1
+            else:
+                # i.e. rval not in left; we discard for left_join_indexer
+                j += 1
+
+    return result, lindexer, rindexer
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def inner_join_indexer(ndarray[numeric_object_t] left, ndarray[numeric_object_t] right):
+    """
+    Two-pass algorithm for monotonic indexes. Handles many-to-one merges.
+
+    Both left and right are monotonic increasing but not necessarily unique.
+    """
+    cdef:
+        Py_ssize_t i, j, nright, nleft, count
+        numeric_object_t lval, rval
+        ndarray[intp_t] lindexer, rindexer
+        ndarray[numeric_object_t] result
+
+    nleft = len(left)
+    nright = len(right)
+
+    # First pass is to find the size 'count' of our output indexers.
+    i = 0
+    j = 0
+    count = 0
+    if nleft > 0 and nright > 0:
+        while True:
+            if i == nleft:
+                break
+            if j == nright:
+                break
+
+            lval = left[i]
+            rval = right[j]
+            if lval == rval:
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                # i.e. lval not in right; we discard for inner_indexer
+                i += 1
+            else:
+                # i.e. rval not in left; we discard for inner_indexer
+                j += 1
+
+    # do it again now that result size is known
+
+    lindexer = np.empty(count, dtype=np.intp)
+    rindexer = np.empty(count, dtype=np.intp)
+    result = np.empty(count, dtype=left.dtype)
+
+    i = 0
+    j = 0
+    count = 0
+    if nleft > 0 and nright > 0:
+        while True:
+            if i == nleft:
+                break
+            if j == nright:
+                break
+
+            lval = left[i]
+            rval = right[j]
+            if lval == rval:
+                lindexer[count] = i
+                rindexer[count] = j
+                result[count] = lval
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                # i.e. lval not in right; we discard for inner_indexer
+                i += 1
+            else:
+                # i.e. rval not in left; we discard for inner_indexer
+                j += 1
+
+    return result, lindexer, rindexer
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def outer_join_indexer(ndarray[numeric_object_t] left, ndarray[numeric_object_t] right):
+    """
+    Both left and right are monotonic increasing but not necessarily unique.
+    """
+    cdef:
+        Py_ssize_t i, j, nright, nleft, count
+        numeric_object_t lval, rval
+        ndarray[intp_t] lindexer, rindexer
+        ndarray[numeric_object_t] result
+
+    nleft = len(left)
+    nright = len(right)
+
+    # First pass is to find the size 'count' of our output indexers.
+    # count will be length of left plus the number of elements of right not in
+    # left (counting duplicates)
+    i = 0
+    j = 0
+    count = 0
+    if nleft == 0:
+        count = nright
+    elif nright == 0:
+        count = nleft
+    else:
+        while True:
+            if i == nleft:
+                count += nright - j
+                break
+            if j == nright:
+                count += nleft - i
+                break
+
+            lval = left[i]
+            rval = right[j]
+            if lval == rval:
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                count += 1
+                i += 1
+            else:
+                count += 1
+                j += 1
+
+    lindexer = np.empty(count, dtype=np.intp)
+    rindexer = np.empty(count, dtype=np.intp)
+    result = np.empty(count, dtype=left.dtype)
+
+    # do it again, but populate the indexers / result
+
+    i = 0
+    j = 0
+    count = 0
+    if nleft == 0:
+        for j in range(nright):
+            lindexer[j] = -1
+            rindexer[j] = j
+            result[j] = right[j]
+    elif nright == 0:
+        for i in range(nleft):
+            lindexer[i] = i
+            rindexer[i] = -1
+            result[i] = left[i]
+    else:
+        while True:
+            if i == nleft:
+                while j < nright:
+                    lindexer[count] = -1
+                    rindexer[count] = j
+                    result[count] = right[j]
+                    count += 1
+                    j += 1
+                break
+            if j == nright:
+                while i < nleft:
+                    lindexer[count] = i
+                    rindexer[count] = -1
+                    result[count] = left[i]
+                    count += 1
+                    i += 1
+                break
+
+            lval = left[i]
+            rval = right[j]
+
+            if lval == rval:
+                lindexer[count] = i
+                rindexer[count] = j
+                result[count] = lval
+                count += 1
+                if i < nleft - 1:
+                    if j < nright - 1 and right[j + 1] == rval:
+                        j += 1
+                    else:
+                        i += 1
+                        if left[i] != rval:
+                            j += 1
+                elif j < nright - 1:
+                    j += 1
+                    if lval != right[j]:
+                        i += 1
+                else:
+                    # end of the road
+                    break
+            elif lval < rval:
+                # i.e. lval not in right; we keep for outer_join_indexer
+                lindexer[count] = i
+                rindexer[count] = -1
+                result[count] = lval
+                count += 1
+                i += 1
+            else:
+                # i.e. rval not in left; we keep for outer_join_indexer
+                lindexer[count] = -1
+                rindexer[count] = j
+                result[count] = rval
+                count += 1
+                j += 1
+
+    return result, lindexer, rindexer
+
+
+# ----------------------------------------------------------------------
+# asof_join_by
+# ----------------------------------------------------------------------
+
+from pandas._libs.hashtable cimport Int64HashTable
+
+
+def asof_join_backward_on_X_by_Y(const numeric_t[:] left_values,
+                                 const numeric_t[:] right_values,
+                                 const int64_t[:] left_by_values,
+                                 const int64_t[:] right_by_values,
+                                 bint allow_exact_matches=True,
+                                 tolerance=None,
+                                 bint use_hashtable=True):
+
+    cdef:
+        Py_ssize_t left_pos, right_pos, left_size, right_size, found_right_pos
+        ndarray[intp_t] left_indexer, right_indexer
+        bint has_tolerance = False
+        numeric_t tolerance_ = 0
+        numeric_t diff = 0
+        Int64HashTable hash_table
+
+    # if we are using tolerance, set our objects
+    if tolerance is not None:
+        has_tolerance = True
+        tolerance_ = tolerance
+
+    left_size = len(left_values)
+    right_size = len(right_values)
+
+    left_indexer = np.empty(left_size, dtype=np.intp)
+    right_indexer = np.empty(left_size, dtype=np.intp)
+
+    if use_hashtable:
+        hash_table = Int64HashTable(right_size)
+
+    right_pos = 0
+    for left_pos in range(left_size):
+        # restart right_pos if it went negative in a previous iteration
+        if right_pos < 0:
+            right_pos = 0
+
+        # find last position in right whose value is less than left's
+        if allow_exact_matches:
+            while (right_pos < right_size and
+                   right_values[right_pos] <= left_values[left_pos]):
+                if use_hashtable:
+                    hash_table.set_item(right_by_values[right_pos], right_pos)
+                right_pos += 1
+        else:
+            while (right_pos < right_size and
+                   right_values[right_pos] < left_values[left_pos]):
+                if use_hashtable:
+                    hash_table.set_item(right_by_values[right_pos], right_pos)
+                right_pos += 1
+        right_pos -= 1
+
+        # save positions as the desired index
+        if use_hashtable:
+            by_value = left_by_values[left_pos]
+            found_right_pos = (hash_table.get_item(by_value)
+                               if by_value in hash_table else -1)
+        else:
+            found_right_pos = right_pos
+
+        left_indexer[left_pos] = left_pos
+        right_indexer[left_pos] = found_right_pos
+
+        # if needed, verify that tolerance is met
+        if has_tolerance and found_right_pos != -1:
+            diff = left_values[left_pos] - right_values[found_right_pos]
+            if diff > tolerance_:
+                right_indexer[left_pos] = -1
+
+    return left_indexer, right_indexer
+
+
+def asof_join_forward_on_X_by_Y(const numeric_t[:] left_values,
+                                const numeric_t[:] right_values,
+                                const int64_t[:] left_by_values,
+                                const int64_t[:] right_by_values,
+                                bint allow_exact_matches=1,
+                                tolerance=None,
+                                bint use_hashtable=True):
+
+    cdef:
+        Py_ssize_t left_pos, right_pos, left_size, right_size, found_right_pos
+        ndarray[intp_t] left_indexer, right_indexer
+        bint has_tolerance = False
+        numeric_t tolerance_ = 0
+        numeric_t diff = 0
+        Int64HashTable hash_table
+
+    # if we are using tolerance, set our objects
+    if tolerance is not None:
+        has_tolerance = True
+        tolerance_ = tolerance
+
+    left_size = len(left_values)
+    right_size = len(right_values)
+
+    left_indexer = np.empty(left_size, dtype=np.intp)
+    right_indexer = np.empty(left_size, dtype=np.intp)
+
+    if use_hashtable:
+        hash_table = Int64HashTable(right_size)
+
+    right_pos = right_size - 1
+    for left_pos in range(left_size - 1, -1, -1):
+        # restart right_pos if it went over in a previous iteration
+        if right_pos == right_size:
+            right_pos = right_size - 1
+
+        # find first position in right whose value is greater than left's
+        if allow_exact_matches:
+            while (right_pos >= 0 and
+                   right_values[right_pos] >= left_values[left_pos]):
+                if use_hashtable:
+                    hash_table.set_item(right_by_values[right_pos], right_pos)
+                right_pos -= 1
+        else:
+            while (right_pos >= 0 and
+                   right_values[right_pos] > left_values[left_pos]):
+                if use_hashtable:
+                    hash_table.set_item(right_by_values[right_pos], right_pos)
+                right_pos -= 1
+        right_pos += 1
+
+        # save positions as the desired index
+        if use_hashtable:
+            by_value = left_by_values[left_pos]
+            found_right_pos = (hash_table.get_item(by_value)
+                               if by_value in hash_table else -1)
+        else:
+            found_right_pos = (right_pos
+                               if right_pos != right_size else -1)
+
+        left_indexer[left_pos] = left_pos
+        right_indexer[left_pos] = found_right_pos
+
+        # if needed, verify that tolerance is met
+        if has_tolerance and found_right_pos != -1:
+            diff = right_values[found_right_pos] - left_values[left_pos]
+            if diff > tolerance_:
+                right_indexer[left_pos] = -1
+
+    return left_indexer, right_indexer
+
+
+def asof_join_nearest_on_X_by_Y(const numeric_t[:] left_values,
+                                const numeric_t[:] right_values,
+                                const int64_t[:] left_by_values,
+                                const int64_t[:] right_by_values,
+                                bint allow_exact_matches=True,
+                                tolerance=None,
+                                bint use_hashtable=True):
+
+    cdef:
+        ndarray[intp_t] bli, bri, fli, fri
+
+        ndarray[intp_t] left_indexer, right_indexer
+        Py_ssize_t left_size, i
+        numeric_t bdiff, fdiff
+
+    # search both forward and backward
+    bli, bri = asof_join_backward_on_X_by_Y(
+        left_values,
+        right_values,
+        left_by_values,
+        right_by_values,
+        allow_exact_matches,
+        tolerance,
+        use_hashtable
+    )
+    fli, fri = asof_join_forward_on_X_by_Y(
+        left_values,
+        right_values,
+        left_by_values,
+        right_by_values,
+        allow_exact_matches,
+        tolerance,
+        use_hashtable
+    )
+
+    # choose the smaller timestamp
+    left_size = len(left_values)
+    left_indexer = np.empty(left_size, dtype=np.intp)
+    right_indexer = np.empty(left_size, dtype=np.intp)
+
+    for i in range(len(bri)):
+        # choose timestamp from right with smaller difference
+        if bri[i] != -1 and fri[i] != -1:
+            bdiff = left_values[bli[i]] - right_values[bri[i]]
+            fdiff = right_values[fri[i]] - left_values[fli[i]]
+            right_indexer[i] = bri[i] if bdiff <= fdiff else fri[i]
+        else:
+            right_indexer[i] = bri[i] if bri[i] != -1 else fri[i]
+        left_indexer[i] = bli[i]
+
+    return left_indexer, right_indexer
diff --git a/pandas/pandas/_libs/json.pyi b/pandas/pandas/_libs/json.pyi
new file mode 100644
index 0000000000000..bc4fe68573b94
--- /dev/null
+++ b/pandas/pandas/_libs/json.pyi
@@ -0,0 +1,23 @@
+from typing import (
+    Any,
+    Callable,
+)
+
+def ujson_dumps(
+    obj: Any,
+    ensure_ascii: bool = ...,
+    double_precision: int = ...,
+    indent: int = ...,
+    orient: str = ...,
+    date_unit: str = ...,
+    iso_dates: bool = ...,
+    default_handler: None
+    | Callable[[Any], str | float | bool | list | dict | None] = ...,
+) -> str: ...
+def ujson_loads(
+    s: str,
+    precise_float: bool = ...,
+    numpy: bool = ...,
+    dtype: None = ...,
+    labelled: bool = ...,
+) -> Any: ...
diff --git a/pandas/pandas/_libs/khash.pxd b/pandas/pandas/_libs/khash.pxd
new file mode 100644
index 0000000000000..c439e1cca772b
--- /dev/null
+++ b/pandas/pandas/_libs/khash.pxd
@@ -0,0 +1,129 @@
+from cpython.object cimport PyObject
+from numpy cimport (
+    complex64_t,
+    complex128_t,
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+
+
+cdef extern from "pandas/vendored/klib/khash_python.h":
+    const int KHASH_TRACE_DOMAIN
+
+    ctypedef uint32_t khuint_t
+    ctypedef khuint_t khiter_t
+
+    ctypedef struct khcomplex128_t:
+        double real
+        double imag
+
+    bint are_equivalent_khcomplex128_t \
+        "kh_complex_hash_equal" (khcomplex128_t a, khcomplex128_t b) nogil
+
+    ctypedef struct khcomplex64_t:
+        float real
+        float imag
+
+    bint are_equivalent_khcomplex64_t \
+        "kh_complex_hash_equal" (khcomplex64_t a, khcomplex64_t b) nogil
+
+    bint are_equivalent_float64_t \
+        "kh_floats_hash_equal" (float64_t a, float64_t b) nogil
+
+    bint are_equivalent_float32_t \
+        "kh_floats_hash_equal" (float32_t a, float32_t b) nogil
+
+    uint32_t kh_python_hash_func(object key)
+    bint kh_python_hash_equal(object a, object b)
+
+    ctypedef struct kh_pymap_t:
+        khuint_t n_buckets, size, n_occupied, upper_bound
+        uint32_t *flags
+        PyObject **keys
+        size_t *vals
+
+    kh_pymap_t* kh_init_pymap()
+    void kh_destroy_pymap(kh_pymap_t*)
+    void kh_clear_pymap(kh_pymap_t*)
+    khuint_t kh_get_pymap(kh_pymap_t*, PyObject*)
+    void kh_resize_pymap(kh_pymap_t*, khuint_t)
+    khuint_t kh_put_pymap(kh_pymap_t*, PyObject*, int*)
+    void kh_del_pymap(kh_pymap_t*, khuint_t)
+
+    bint kh_exist_pymap(kh_pymap_t*, khiter_t)
+
+    ctypedef struct kh_pyset_t:
+        khuint_t n_buckets, size, n_occupied, upper_bound
+        uint32_t *flags
+        PyObject **keys
+        size_t *vals
+
+    kh_pyset_t* kh_init_pyset()
+    void kh_destroy_pyset(kh_pyset_t*)
+    void kh_clear_pyset(kh_pyset_t*)
+    khuint_t kh_get_pyset(kh_pyset_t*, PyObject*)
+    void kh_resize_pyset(kh_pyset_t*, khuint_t)
+    khuint_t kh_put_pyset(kh_pyset_t*, PyObject*, int*)
+    void kh_del_pyset(kh_pyset_t*, khuint_t)
+
+    bint kh_exist_pyset(kh_pyset_t*, khiter_t)
+
+    ctypedef char* kh_cstr_t
+
+    ctypedef struct kh_str_t:
+        khuint_t n_buckets, size, n_occupied, upper_bound
+        uint32_t *flags
+        kh_cstr_t *keys
+        size_t *vals
+
+    kh_str_t* kh_init_str() nogil
+    void kh_destroy_str(kh_str_t*) nogil
+    void kh_clear_str(kh_str_t*) nogil
+    khuint_t kh_get_str(kh_str_t*, kh_cstr_t) nogil
+    void kh_resize_str(kh_str_t*, khuint_t) nogil
+    khuint_t kh_put_str(kh_str_t*, kh_cstr_t, int*) nogil
+    void kh_del_str(kh_str_t*, khuint_t) nogil
+
+    bint kh_exist_str(kh_str_t*, khiter_t) nogil
+
+    ctypedef struct kh_str_starts_t:
+        kh_str_t *table
+        int starts[256]
+
+    kh_str_starts_t* kh_init_str_starts() nogil
+    khuint_t kh_put_str_starts_item(kh_str_starts_t* table, char* key,
+                                    int* ret) nogil
+    khuint_t kh_get_str_starts_item(kh_str_starts_t* table, char* key) nogil
+    void kh_destroy_str_starts(kh_str_starts_t*) nogil
+    void kh_resize_str_starts(kh_str_starts_t*, khuint_t) nogil
+
+    # sweep factorize
+
+    ctypedef struct kh_strbox_t:
+        khuint_t n_buckets, size, n_occupied, upper_bound
+        uint32_t *flags
+        kh_cstr_t *keys
+        PyObject **vals
+
+    kh_strbox_t* kh_init_strbox() nogil
+    void kh_destroy_strbox(kh_strbox_t*) nogil
+    void kh_clear_strbox(kh_strbox_t*) nogil
+    khuint_t kh_get_strbox(kh_strbox_t*, kh_cstr_t) nogil
+    void kh_resize_strbox(kh_strbox_t*, khuint_t) nogil
+    khuint_t kh_put_strbox(kh_strbox_t*, kh_cstr_t, int*) nogil
+    void kh_del_strbox(kh_strbox_t*, khuint_t) nogil
+
+    bint kh_exist_strbox(kh_strbox_t*, khiter_t) nogil
+
+    khuint_t kh_needed_n_buckets(khuint_t element_n) nogil
+
+
+include "khash_for_primitive_helper.pxi"
diff --git a/pandas/pandas/_libs/khash_for_primitive_helper.pxi.in b/pandas/pandas/_libs/khash_for_primitive_helper.pxi.in
new file mode 100644
index 0000000000000..d3391d4028938
--- /dev/null
+++ b/pandas/pandas/_libs/khash_for_primitive_helper.pxi.in
@@ -0,0 +1,44 @@
+"""
+Template for wrapping khash-tables for each primitive `dtype`
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+{{py:
+
+# name, c_type
+primitive_types = [('int64', 'int64_t'),
+                   ('uint64', 'uint64_t'),
+                   ('float64', 'float64_t'),
+                   ('int32', 'int32_t'),
+                   ('uint32', 'uint32_t'),
+                   ('float32', 'float32_t'),
+                   ('int16', 'int16_t'),
+                   ('uint16', 'uint16_t'),
+                   ('int8', 'int8_t'),
+                   ('uint8', 'uint8_t'),
+                   ('complex64', 'khcomplex64_t'),
+                   ('complex128', 'khcomplex128_t'),
+                  ]
+}}
+
+{{for name, c_type in primitive_types}}
+
+cdef extern from "pandas/vendored/klib/khash_python.h":
+    ctypedef struct kh_{{name}}_t:
+        khuint_t n_buckets, size, n_occupied, upper_bound
+        uint32_t *flags
+        {{c_type}} *keys
+        size_t *vals
+
+    kh_{{name}}_t* kh_init_{{name}}() nogil
+    void kh_destroy_{{name}}(kh_{{name}}_t*) nogil
+    void kh_clear_{{name}}(kh_{{name}}_t*) nogil
+    khuint_t kh_get_{{name}}(kh_{{name}}_t*, {{c_type}}) nogil
+    void kh_resize_{{name}}(kh_{{name}}_t*, khuint_t) nogil
+    khuint_t kh_put_{{name}}(kh_{{name}}_t*, {{c_type}}, int*) nogil
+    void kh_del_{{name}}(kh_{{name}}_t*, khuint_t) nogil
+
+    bint kh_exist_{{name}}(kh_{{name}}_t*, khiter_t) nogil
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/lib.pxd b/pandas/pandas/_libs/lib.pxd
new file mode 100644
index 0000000000000..46a339f2e7cbb
--- /dev/null
+++ b/pandas/pandas/_libs/lib.pxd
@@ -0,0 +1,6 @@
+from numpy cimport ndarray
+
+
+cdef bint c_is_list_like(object, bint) except -1
+
+cpdef ndarray eq_NA_compat(ndarray[object] arr, object key)
diff --git a/pandas/pandas/_libs/lib.pyi b/pandas/pandas/_libs/lib.pyi
new file mode 100644
index 0000000000000..daaaacee3487d
--- /dev/null
+++ b/pandas/pandas/_libs/lib.pyi
@@ -0,0 +1,236 @@
+# TODO(npdtypes): Many types specified here can be made more specific/accurate;
+#  the more specific versions are specified in comments
+from decimal import Decimal
+from typing import (
+    Any,
+    Callable,
+    Final,
+    Generator,
+    Hashable,
+    Literal,
+    TypeAlias,
+    overload,
+)
+
+import numpy as np
+
+from pandas._typing import (
+    ArrayLike,
+    DtypeObj,
+    TypeGuard,
+    npt,
+)
+
+# placeholder until we can specify np.ndarray[object, ndim=2]
+ndarray_obj_2d = np.ndarray
+
+from enum import Enum
+
+class _NoDefault(Enum):
+    no_default = ...
+
+no_default: Final = _NoDefault.no_default
+NoDefault: TypeAlias = Literal[_NoDefault.no_default]
+
+i8max: int
+u8max: int
+
+def is_np_dtype(dtype: object, kinds: str | None = ...) -> TypeGuard[np.dtype]: ...
+def item_from_zerodim(val: object) -> object: ...
+def infer_dtype(value: object, skipna: bool = ...) -> str: ...
+def is_iterator(obj: object) -> bool: ...
+def is_scalar(val: object) -> bool: ...
+def is_list_like(obj: object, allow_sets: bool = ...) -> bool: ...
+def is_pyarrow_array(obj: object) -> bool: ...
+def is_decimal(obj: object) -> TypeGuard[Decimal]: ...
+def is_complex(obj: object) -> TypeGuard[complex]: ...
+def is_bool(obj: object) -> TypeGuard[bool | np.bool_]: ...
+def is_integer(obj: object) -> TypeGuard[int | np.integer]: ...
+def is_int_or_none(obj) -> bool: ...
+def is_float(obj: object) -> TypeGuard[float]: ...
+def is_interval_array(values: np.ndarray) -> bool: ...
+def is_datetime64_array(values: np.ndarray, skipna: bool = True) -> bool: ...
+def is_timedelta_or_timedelta64_array(
+    values: np.ndarray, skipna: bool = True
+) -> bool: ...
+def is_datetime_with_singletz_array(values: np.ndarray) -> bool: ...
+def is_time_array(values: np.ndarray, skipna: bool = ...): ...
+def is_date_array(values: np.ndarray, skipna: bool = ...): ...
+def is_datetime_array(values: np.ndarray, skipna: bool = ...): ...
+def is_string_array(values: np.ndarray, skipna: bool = ...): ...
+def is_float_array(values: np.ndarray): ...
+def is_integer_array(values: np.ndarray, skipna: bool = ...): ...
+def is_bool_array(values: np.ndarray, skipna: bool = ...): ...
+def fast_multiget(
+    mapping: dict,
+    keys: np.ndarray,  # object[:]
+    default=...,
+) -> ArrayLike: ...
+def fast_unique_multiple_list_gen(gen: Generator, sort: bool = ...) -> list: ...
+@overload
+def map_infer(
+    arr: np.ndarray,
+    f: Callable[[Any], Any],
+    *,
+    convert: Literal[False],
+    ignore_na: bool = ...,
+) -> np.ndarray: ...
+@overload
+def map_infer(
+    arr: np.ndarray,
+    f: Callable[[Any], Any],
+    *,
+    convert: bool = ...,
+    ignore_na: bool = ...,
+) -> ArrayLike: ...
+@overload
+def maybe_convert_objects(
+    objects: npt.NDArray[np.object_],
+    *,
+    try_float: bool = ...,
+    safe: bool = ...,
+    convert_numeric: bool = ...,
+    convert_non_numeric: Literal[False] = ...,
+    convert_to_nullable_dtype: Literal[False] = ...,
+    dtype_if_all_nat: DtypeObj | None = ...,
+) -> npt.NDArray[np.object_ | np.number]: ...
+@overload
+def maybe_convert_objects(
+    objects: npt.NDArray[np.object_],
+    *,
+    try_float: bool = ...,
+    safe: bool = ...,
+    convert_numeric: bool = ...,
+    convert_non_numeric: bool = ...,
+    convert_to_nullable_dtype: Literal[True] = ...,
+    dtype_if_all_nat: DtypeObj | None = ...,
+) -> ArrayLike: ...
+@overload
+def maybe_convert_objects(
+    objects: npt.NDArray[np.object_],
+    *,
+    try_float: bool = ...,
+    safe: bool = ...,
+    convert_numeric: bool = ...,
+    convert_non_numeric: bool = ...,
+    convert_to_nullable_dtype: bool = ...,
+    dtype_if_all_nat: DtypeObj | None = ...,
+) -> ArrayLike: ...
+@overload
+def maybe_convert_numeric(
+    values: npt.NDArray[np.object_],
+    na_values: set,
+    convert_empty: bool = ...,
+    coerce_numeric: bool = ...,
+    convert_to_masked_nullable: Literal[False] = ...,
+) -> tuple[np.ndarray, None]: ...
+@overload
+def maybe_convert_numeric(
+    values: npt.NDArray[np.object_],
+    na_values: set,
+    convert_empty: bool = ...,
+    coerce_numeric: bool = ...,
+    *,
+    convert_to_masked_nullable: Literal[True],
+) -> tuple[np.ndarray, np.ndarray]: ...
+
+# TODO: restrict `arr`?
+def ensure_string_array(
+    arr,
+    na_value: object = ...,
+    convert_na_value: bool = ...,
+    copy: bool = ...,
+    skipna: bool = ...,
+) -> npt.NDArray[np.object_]: ...
+def convert_nans_to_NA(
+    arr: npt.NDArray[np.object_],
+) -> npt.NDArray[np.object_]: ...
+def fast_zip(ndarrays: list) -> npt.NDArray[np.object_]: ...
+
+# TODO: can we be more specific about rows?
+def to_object_array_tuples(rows: object) -> ndarray_obj_2d: ...
+def tuples_to_object_array(
+    tuples: npt.NDArray[np.object_],
+) -> ndarray_obj_2d: ...
+
+# TODO: can we be more specific about rows?
+def to_object_array(rows: object, min_width: int = ...) -> ndarray_obj_2d: ...
+def dicts_to_array(dicts: list, columns: list) -> ndarray_obj_2d: ...
+def maybe_booleans_to_slice(
+    mask: npt.NDArray[np.uint8],
+) -> slice | npt.NDArray[np.uint8]: ...
+def maybe_indices_to_slice(
+    indices: npt.NDArray[np.intp],
+    max_len: int,
+) -> slice | npt.NDArray[np.intp]: ...
+def is_all_arraylike(obj: list) -> bool: ...
+
+# -----------------------------------------------------------------
+# Functions which in reality take memoryviews
+
+def memory_usage_of_objects(arr: np.ndarray) -> int: ...  # object[:]  # np.int64
+@overload
+def map_infer_mask(
+    arr: np.ndarray,
+    f: Callable[[Any], Any],
+    mask: np.ndarray,  # const uint8_t[:]
+    *,
+    convert: Literal[False],
+    na_value: Any = ...,
+    dtype: np.dtype = ...,
+) -> np.ndarray: ...
+@overload
+def map_infer_mask(
+    arr: np.ndarray,
+    f: Callable[[Any], Any],
+    mask: np.ndarray,  # const uint8_t[:]
+    *,
+    convert: bool = ...,
+    na_value: Any = ...,
+    dtype: np.dtype = ...,
+) -> ArrayLike: ...
+def indices_fast(
+    index: npt.NDArray[np.intp],
+    labels: np.ndarray,  # const int64_t[:]
+    keys: list,
+    sorted_labels: list[npt.NDArray[np.int64]],
+) -> dict[Hashable, npt.NDArray[np.intp]]: ...
+def generate_slices(
+    labels: np.ndarray,
+    ngroups: int,  # const intp_t[:]
+) -> tuple[npt.NDArray[np.int64], npt.NDArray[np.int64]]: ...
+def count_level_2d(
+    mask: np.ndarray,  # ndarray[uint8_t, ndim=2, cast=True],
+    labels: np.ndarray,  # const intp_t[:]
+    max_bin: int,
+) -> np.ndarray: ...  # np.ndarray[np.int64, ndim=2]
+def get_level_sorter(
+    codes: np.ndarray,  # const int64_t[:]
+    starts: np.ndarray,  # const intp_t[:]
+) -> np.ndarray: ...  # np.ndarray[np.intp, ndim=1]
+def generate_bins_dt64(
+    values: npt.NDArray[np.int64],
+    binner: np.ndarray,  # const int64_t[:]
+    closed: object = ...,
+    hasnans: bool = ...,
+) -> np.ndarray: ...  # np.ndarray[np.int64, ndim=1]
+def array_equivalent_object(
+    left: npt.NDArray[np.object_],
+    right: npt.NDArray[np.object_],
+) -> bool: ...
+def has_infs(arr: np.ndarray) -> bool: ...  # const floating[:]
+def has_only_ints_or_nan(arr: np.ndarray) -> bool: ...  # const floating[:]
+def get_reverse_indexer(
+    indexer: np.ndarray,  # const intp_t[:]
+    length: int,
+) -> npt.NDArray[np.intp]: ...
+def is_bool_list(obj: list) -> bool: ...
+def dtypes_all_equal(types: list[DtypeObj]) -> bool: ...
+def is_range_indexer(
+    left: np.ndarray,
+    n: int,  # np.ndarray[np.int64, ndim=1]
+) -> bool: ...
+def is_sequence_range(
+    sequence: np.ndarray,
+    step: int,  # np.ndarray[np.int64, ndim=1]
+) -> bool: ...
diff --git a/pandas/pandas/_libs/lib.pyx b/pandas/pandas/_libs/lib.pyx
new file mode 100644
index 0000000000000..3c509a3eae11a
--- /dev/null
+++ b/pandas/pandas/_libs/lib.pyx
@@ -0,0 +1,3291 @@
+from collections import abc
+from decimal import Decimal
+from enum import Enum
+from sys import getsizeof
+from typing import (
+    Literal,
+    _GenericAlias,
+)
+
+cimport cython
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    PyDelta_Check,
+    PyTime_Check,
+    date,
+    datetime,
+    import_datetime,
+    time,
+    timedelta,
+)
+from cpython.iterator cimport PyIter_Check
+from cpython.number cimport PyNumber_Check
+from cpython.object cimport (
+    Py_EQ,
+    PyObject,
+    PyObject_RichCompareBool,
+)
+from cpython.ref cimport Py_INCREF
+from cpython.sequence cimport PySequence_Check
+from cpython.tuple cimport (
+    PyTuple_New,
+    PyTuple_SET_ITEM,
+)
+from cython cimport (
+    Py_ssize_t,
+    floating,
+)
+
+from pandas._config import using_string_dtype
+
+from pandas._libs.missing import check_na_tuples_nonequal
+
+import_datetime()
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    NPY_OBJECT,
+    PyArray_Check,
+    PyArray_GETITEM,
+    PyArray_ITER_DATA,
+    PyArray_ITER_NEXT,
+    PyArray_IterNew,
+    PyArray_SETITEM,
+    complex128_t,
+    flatiter,
+    float64_t,
+    int32_t,
+    int64_t,
+    intp_t,
+    ndarray,
+    uint8_t,
+    uint64_t,
+)
+
+cnp.import_array()
+from pandas._libs.interval import Interval
+
+
+cdef extern from "pandas/parser/pd_parser.h":
+    int floatify(object, float64_t *result, int *maybe_int) except -1
+    void PandasParser_IMPORT()
+
+PandasParser_IMPORT
+
+from pandas._libs cimport util
+from pandas._libs.util cimport (
+    INT64_MAX,
+    INT64_MIN,
+    UINT64_MAX,
+    is_nan,
+)
+
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+)
+from pandas._libs.tslibs.period import Period
+
+from pandas._libs.missing cimport (
+    C_NA,
+    checknull,
+    is_matching_na,
+    is_null_datetime64,
+    is_null_timedelta64,
+)
+from pandas._libs.tslibs.conversion cimport convert_to_tsobject
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+    checknull_with_nat,
+)
+from pandas._libs.tslibs.offsets cimport is_offset_object
+from pandas._libs.tslibs.period cimport is_period_object
+from pandas._libs.tslibs.timedeltas cimport convert_to_timedelta64
+from pandas._libs.tslibs.timezones cimport tz_compare
+
+# constants that will be compared to potentially arbitrarily large
+# python int
+cdef:
+    object oINT64_MAX = <int64_t>INT64_MAX
+    object oINT64_MIN = <int64_t>INT64_MIN
+    object oUINT64_MAX = <uint64_t>UINT64_MAX
+
+    float64_t NaN = <float64_t>np.nan
+
+# python-visible
+i8max = <int64_t>INT64_MAX
+u8max = <uint64_t>UINT64_MAX
+
+
+cdef bint PYARROW_INSTALLED = False
+
+try:
+    import pyarrow as pa
+
+    PYARROW_INSTALLED = True
+except ImportError:
+    pa = None
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def memory_usage_of_objects(arr: object[:]) -> int64_t:
+    """
+    Return the memory usage of an object array in bytes.
+
+    Does not include the actual bytes of the pointers
+    """
+    cdef:
+        Py_ssize_t i
+        Py_ssize_t n
+        int64_t size = 0
+
+    n = len(arr)
+    for i in range(n):
+        size += getsizeof(arr[i])
+    return size
+
+
+# ----------------------------------------------------------------------
+
+
+def is_scalar(val: object) -> bool:
+    """
+    Return True if given object is scalar.
+
+    Parameters
+    ----------
+    val : object
+        This includes:
+
+        - numpy array scalar (e.g. np.int64)
+        - Python builtin numerics
+        - Python builtin byte arrays and strings
+        - None
+        - datetime.datetime
+        - datetime.timedelta
+        - Period
+        - decimal.Decimal
+        - Interval
+        - DateOffset
+        - Fraction
+        - Number.
+
+    Returns
+    -------
+    bool
+        Return True if given object is scalar.
+
+    See Also
+    --------
+    api.types.is_list_like : Check if the input is list-like.
+    api.types.is_integer : Check if the input is an integer.
+    api.types.is_float : Check if the input is a float.
+    api.types.is_bool : Check if the input is a boolean.
+
+    Examples
+    --------
+    >>> import datetime
+    >>> dt = datetime.datetime(2018, 10, 3)
+    >>> pd.api.types.is_scalar(dt)
+    True
+
+    >>> pd.api.types.is_scalar([2, 3])
+    False
+
+    >>> pd.api.types.is_scalar({0: 1, 2: 3})
+    False
+
+    >>> pd.api.types.is_scalar((0, 2))
+    False
+
+    pandas supports PEP 3141 numbers:
+
+    >>> from fractions import Fraction
+    >>> pd.api.types.is_scalar(Fraction(3, 5))
+    True
+    """
+
+    # Start with C-optimized checks
+    if (cnp.PyArray_IsAnyScalar(val)
+            # PyArray_IsAnyScalar is always False for bytearrays on Py3
+            or PyDate_Check(val)
+            or PyDelta_Check(val)
+            or PyTime_Check(val)
+            # We differ from numpy, which claims that None is not scalar;
+            # see np.isscalar
+            or val is C_NA
+            or val is None):
+        return True
+
+    # Next use C-optimized checks to exclude common non-scalars before falling
+    #  back to non-optimized checks.
+    if PySequence_Check(val):
+        # e.g. list, tuple
+        # includes np.ndarray, Series which PyNumber_Check can return True for
+        return False
+
+    # Note: PyNumber_Check check includes Decimal, Fraction, numbers.Number
+    return (PyNumber_Check(val)
+            or is_period_object(val)
+            or isinstance(val, Interval)
+            or is_offset_object(val))
+
+
+cdef int64_t get_itemsize(object val):
+    """
+    Get the itemsize of a NumPy scalar, -1 if not a NumPy scalar.
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_ndarray : bool
+    """
+    if cnp.PyArray_CheckScalar(val):
+        return cnp.PyArray_DescrFromScalar(val).itemsize
+    else:
+        return -1
+
+
+def is_iterator(obj: object) -> bool:
+    """
+    Check if the object is an iterator.
+
+    This is intended for generators, not list-like objects.
+    This method checks whether the passed object is an iterator. It
+    returns `True` if the object is an iterator, and `False` otherwise.
+
+    Parameters
+    ----------
+    obj : The object to check
+        The object to check for iterator type.
+
+    Returns
+    -------
+    is_iter : bool
+        Whether `obj` is an iterator.
+        `True` if the object is of iterator type, otherwise `False`.
+
+    See Also
+    --------
+    api.types.is_list_like : Check if the input is list-like.
+
+    Examples
+    --------
+    >>> import datetime
+    >>> from pandas.api.types import is_iterator
+    >>> is_iterator((x for x in []))
+    True
+    >>> is_iterator([1, 2, 3])
+    False
+    >>> is_iterator(datetime.datetime(2017, 1, 1))
+    False
+    >>> is_iterator("foo")
+    False
+    >>> is_iterator(1)
+    False
+    """
+    return PyIter_Check(obj)
+
+
+def item_from_zerodim(val: object) -> object:
+    """
+    If the value is a zerodim array, return the item it contains.
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    object
+
+    Examples
+    --------
+    >>> item_from_zerodim(1)
+    1
+    >>> item_from_zerodim('foobar')
+    'foobar'
+    >>> item_from_zerodim(np.array(1))
+    1
+    >>> item_from_zerodim(np.array([1]))
+    array([1])
+    """
+    if cnp.PyArray_IsZeroDim(val):
+        return cnp.PyArray_ToScalar(cnp.PyArray_DATA(val), val)
+    return val
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def fast_unique_multiple_list_gen(object gen, bint sort=True) -> list:
+    """
+    Generate a list of unique values from a generator of lists.
+
+    Parameters
+    ----------
+    gen : generator object
+        Generator of lists from which the unique list is created.
+    sort : bool
+        Whether or not to sort the resulting unique list.
+
+    Returns
+    -------
+    list of unique values
+    """
+    cdef:
+        list buf
+        Py_ssize_t j, n
+        list uniques = []
+        set table = set()
+        object val
+
+    for buf in gen:
+        n = len(buf)
+        for j in range(n):
+            val = buf[j]
+            if val not in table:
+                table.add(val)
+                uniques.append(val)
+    if sort:
+        try:
+            uniques.sort()
+        except TypeError:
+            pass
+
+    return uniques
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def dicts_to_array(dicts: list, columns: list):
+    cdef:
+        Py_ssize_t i, j, k, n
+        ndarray[object, ndim=2] result
+        dict row
+        object col, onan = np.nan
+
+    k = len(columns)
+    n = len(dicts)
+
+    result = np.empty((n, k), dtype="O")
+
+    for i in range(n):
+        row = dicts[i]
+        for j in range(k):
+            col = columns[j]
+            if col in row:
+                result[i, j] = row[col]
+            else:
+                result[i, j] = onan
+
+    return result
+
+
+def fast_zip(list ndarrays) -> ndarray[object]:
+    """
+    For zipping multiple ndarrays into an ndarray of tuples.
+    """
+    cdef:
+        Py_ssize_t i, j, k, n
+        ndarray[object, ndim=1] result
+        flatiter it
+        object val, tup
+
+    k = len(ndarrays)
+    n = len(ndarrays[0])
+
+    result = np.empty(n, dtype=object)
+
+    # initialize tuples on first pass
+    arr = ndarrays[0]
+    it = <flatiter>PyArray_IterNew(arr)
+    for i in range(n):
+        val = PyArray_GETITEM(arr, PyArray_ITER_DATA(it))
+        tup = PyTuple_New(k)
+
+        PyTuple_SET_ITEM(tup, 0, val)
+        Py_INCREF(val)
+        result[i] = tup
+        PyArray_ITER_NEXT(it)
+
+    for j in range(1, k):
+        arr = ndarrays[j]
+        it = <flatiter>PyArray_IterNew(arr)
+        if len(arr) != n:
+            raise ValueError("all arrays must be same length")
+
+        for i in range(n):
+            val = PyArray_GETITEM(arr, PyArray_ITER_DATA(it))
+            PyTuple_SET_ITEM(result[i], j, val)
+            Py_INCREF(val)
+            PyArray_ITER_NEXT(it)
+
+    return result
+
+
+def get_reverse_indexer(const intp_t[:] indexer, Py_ssize_t length) -> ndarray:
+    """
+    Reverse indexing operation.
+
+    Given `indexer`, make `indexer_inv` of it, such that::
+
+        indexer_inv[indexer[x]] = x
+
+    Parameters
+    ----------
+    indexer : np.ndarray[np.intp]
+    length : int
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+
+    Notes
+    -----
+    If indexer is not unique, only first occurrence is accounted.
+    """
+    cdef:
+        Py_ssize_t i, n = len(indexer)
+        ndarray[intp_t, ndim=1] rev_indexer
+        intp_t idx
+
+    rev_indexer = np.empty(length, dtype=np.intp)
+    rev_indexer[:] = -1
+    for i in range(n):
+        idx = indexer[i]
+        if idx != -1:
+            rev_indexer[idx] = i
+
+    return rev_indexer
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def has_infs(const floating[:] arr) -> bool:
+    cdef:
+        Py_ssize_t i, n = len(arr)
+        floating inf, neginf, val
+        bint ret = False
+
+    inf = np.inf
+    neginf = -inf
+    with nogil:
+        for i in range(n):
+            val = arr[i]
+            if val == inf or val == neginf:
+                ret = True
+                break
+    return ret
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def has_only_ints_or_nan(const floating[:] arr) -> bool:
+    cdef:
+        floating val
+        intp_t i
+
+    for i in range(len(arr)):
+        val = arr[i]
+        if (val != val) or (val == <int64_t>val):
+            continue
+        else:
+            return False
+    return True
+
+
+def maybe_indices_to_slice(ndarray[intp_t, ndim=1] indices, int max_len):
+    cdef:
+        Py_ssize_t i, n = len(indices)
+        intp_t k, vstart, vlast, v
+
+    if n == 0:
+        return slice(0, 0)
+
+    vstart = indices[0]
+    if vstart < 0 or max_len <= vstart:
+        return indices
+
+    if n == 1:
+        return slice(vstart, <intp_t>(vstart + 1))
+
+    vlast = indices[n - 1]
+    if vlast < 0 or max_len <= vlast:
+        return indices
+
+    k = indices[1] - indices[0]
+    if k == 0:
+        return indices
+    else:
+        for i in range(2, n):
+            v = indices[i]
+            if v - indices[i - 1] != k:
+                return indices
+
+        if k > 0:
+            return slice(vstart, <intp_t>(vlast + 1), k)
+        else:
+            if vlast == 0:
+                return slice(vstart, None, k)
+            else:
+                return slice(vstart, <intp_t>(vlast - 1), k)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def maybe_booleans_to_slice(ndarray[uint8_t, ndim=1] mask):
+    cdef:
+        Py_ssize_t i, n = len(mask)
+        Py_ssize_t start = 0, end = 0
+        bint started = False, finished = False
+
+    for i in range(n):
+        if mask[i]:
+            if finished:
+                return mask.view(np.bool_)
+            if not started:
+                started = True
+                start = i
+        else:
+            if finished:
+                continue
+
+            if started:
+                end = i
+                finished = True
+
+    if not started:
+        return slice(0, 0)
+    if not finished:
+        return slice(start, None)
+    else:
+        return slice(start, end)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def array_equivalent_object(ndarray left, ndarray right) -> bool:
+    """
+    Perform an element by element comparison on N-d object arrays
+    taking into account nan positions.
+    """
+    # left and right both have object dtype, but we cannot annotate that
+    #  without limiting ndim.
+    cdef:
+        Py_ssize_t i, n = left.size
+        object x, y
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(left, right)
+
+    # Caller is responsible for checking left.shape == right.shape
+
+    for i in range(n):
+        # Analogous to: x = left[i]
+        x = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 0))[0]
+        y = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        # we are either not equal or both nan
+        # I think None == None will be true here
+        try:
+            if PyArray_Check(x) and PyArray_Check(y):
+                if x.shape != y.shape:
+                    return False
+                if x.dtype == y.dtype == object:
+                    if not array_equivalent_object(x, y):
+                        return False
+                else:
+                    # Circular import isn't great, but so it goes.
+                    # TODO: could use np.array_equal?
+                    from pandas.core.dtypes.missing import array_equivalent
+
+                    if not array_equivalent(x, y):
+                        return False
+
+            elif PyArray_Check(x) or PyArray_Check(y):
+                return False
+            elif (x is C_NA) ^ (y is C_NA):
+                return False
+            elif not (
+                PyObject_RichCompareBool(x, y, Py_EQ)
+                or is_matching_na(x, y, nan_matches_none=True)
+            ):
+                return False
+        except (ValueError, TypeError):
+            # Avoid raising ValueError when comparing Numpy arrays to other types
+            if cnp.PyArray_IsAnyScalar(x) != cnp.PyArray_IsAnyScalar(y):
+                # Only compare scalars to scalars and non-scalars to non-scalars
+                return False
+            elif (not (cnp.PyArray_IsPythonScalar(x) or cnp.PyArray_IsPythonScalar(y))
+                  and not (isinstance(x, type(y)) or isinstance(y, type(x)))):
+                # Check if non-scalars have the same type
+                return False
+            elif check_na_tuples_nonequal(x, y):
+                # We have tuples where one Side has a NA and the other side does not
+                # Only condition we may end up with a TypeError
+                return False
+            raise
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return True
+
+
+ctypedef fused int6432_t:
+    int64_t
+    int32_t
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_range_indexer(const int6432_t[:] left, Py_ssize_t n) -> bool:
+    """
+    Perform an element by element comparison on 1-d integer arrays, meant for indexer
+    comparisons
+    """
+    cdef:
+        Py_ssize_t i
+
+    if left.size != n:
+        return False
+
+    for i in range(n):
+
+        if left[i] != i:
+            return False
+
+    return True
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_sequence_range(const int6432_t[:] sequence, int64_t step) -> bool:
+    """
+    Check if sequence is equivalent to a range with the specified step.
+    """
+    cdef:
+        Py_ssize_t i, n = len(sequence)
+        int6432_t first_element
+
+    if step == 0:
+        return False
+    if n == 0:
+        return True
+
+    first_element = sequence[0]
+    for i in range(1, n):
+        if sequence[i] != first_element + i * step:
+            return False
+    return True
+
+
+ctypedef fused ndarr_object:
+    ndarray[object, ndim=1]
+    ndarray[object, ndim=2]
+
+# TODO: get rid of this in StringArray and modify
+#  and go through ensure_string_array instead
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def convert_nans_to_NA(ndarr_object arr) -> ndarray:
+    """
+    Helper for StringArray that converts null values that
+    are not pd.NA(e.g. np.nan, None) to pd.NA. Assumes elements
+    have already been validated as null.
+    """
+    cdef:
+        Py_ssize_t i, m, n
+        object val
+        ndarr_object result
+    result = np.asarray(arr, dtype="object")
+    if arr.ndim == 2:
+        m, n = arr.shape[0], arr.shape[1]
+        for i in range(m):
+            for j in range(n):
+                val = arr[i, j]
+                if not isinstance(val, str):
+                    result[i, j] = <object>C_NA
+    else:
+        n = len(arr)
+        for i in range(n):
+            val = arr[i]
+            if not isinstance(val, str):
+                result[i] = <object>C_NA
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef ndarray[object] ensure_string_array(
+        arr,
+        object na_value=np.nan,
+        bint convert_na_value=True,
+        bint copy=True,
+        bint skipna=True,
+):
+    """
+    Returns a new numpy array with object dtype and only strings and na values.
+
+    Parameters
+    ----------
+    arr : array-like
+        The values to be converted to str, if needed.
+    na_value : Any, default np.nan
+        The value to use for na. For example, np.nan or pd.NA.
+    convert_na_value : bool, default True
+        If False, existing na values will be used unchanged in the new array.
+    copy : bool, default True
+        Whether to ensure that a new array is returned. When True, a new array
+        is always returned. When False, a new array is only returned when needed
+        to avoid mutating the input array.
+    skipna : bool, default True
+        Whether or not to coerce nulls to their stringified form
+        (e.g. if False, NaN becomes 'nan').
+
+    Returns
+    -------
+    np.ndarray[object]
+        An array with the input array's elements casted to str or nan-like.
+    """
+    cdef:
+        Py_ssize_t i = 0, n = len(arr)
+        bint already_copied = True
+        ndarray[object] newarr
+
+    if hasattr(arr, "to_numpy"):
+
+        if (
+            hasattr(arr, "dtype")
+            and arr.dtype.kind in "mM"
+            # TODO: we should add a custom ArrowExtensionArray.astype implementation
+            # that handles astype(str) specifically, avoiding ending up here and
+            # then we can remove the below check for `_pa_array` (for ArrowEA)
+            and not hasattr(arr, "_pa_array")
+        ):
+            # dtype check to exclude DataFrame
+            # GH#41409 TODO: not a great place for this
+            out = arr.astype(str).astype(object)
+            out[arr.isna()] = na_value
+            return out
+        arr = arr.to_numpy(dtype=object)
+    elif not util.is_array(arr):
+        arr = np.array(arr, dtype="object")
+
+    result = np.asarray(arr, dtype="object")
+
+    if result is arr or np.may_share_memory(arr, result):
+        # if np.asarray(..) did not make a copy of the input arr, we still need
+        #  to do that to avoid mutating the input array
+        # GH#54654: share_memory check is needed for rare cases where np.asarray
+        #  returns a new object without making a copy of the actual data
+        if copy:
+            result = result.copy()
+        else:
+            already_copied = False
+    elif not copy and not result.flags.writeable:
+        # Weird edge case where result is a view
+        already_copied = False
+
+    if issubclass(arr.dtype.type, np.str_):
+        # short-circuit, all elements are str
+        return result
+
+    if arr.dtype.kind == "f":  # non-optimized path
+        for i in range(n):
+            val = arr[i]
+
+            if not already_copied:
+                result = result.copy()
+                already_copied = True
+
+            if not checknull(val):
+                # f"{val}" is not always equivalent to str(val) for floats
+                result[i] = str(val)
+            else:
+                if convert_na_value:
+                    val = na_value
+                if skipna:
+                    result[i] = val
+                else:
+                    result[i] = f"{val}"
+
+        return result
+
+    newarr = np.asarray(arr, dtype=object)
+    for i in range(n):
+        val = newarr[i]
+
+        if isinstance(val, str):
+            continue
+
+        elif not already_copied:
+            result = result.copy()
+            already_copied = True
+
+        if not checknull(val):
+            if isinstance(val, bytes):
+                # GH#49658 discussion of desired behavior here
+                result[i] = val.decode()
+            elif not util.is_float_object(val):
+                # f"{val}" is faster than str(val)
+                result[i] = f"{val}"
+            else:
+                # f"{val}" is not always equivalent to str(val) for floats
+                result[i] = str(val)
+        else:
+            if convert_na_value:
+                val = na_value
+            if skipna:
+                result[i] = val
+            else:
+                result[i] = f"{val}"
+
+    return result
+
+
+def is_all_arraylike(obj: list) -> bool:
+    """
+    Should we treat these as levels of a MultiIndex, as opposed to Index items?
+    """
+    cdef:
+        Py_ssize_t i, n = len(obj)
+        object val
+        bint all_arrays = True
+
+    from pandas.core.dtypes.generic import (
+        ABCIndex,
+        ABCMultiIndex,
+        ABCSeries,
+    )
+
+    for i in range(n):
+        val = obj[i]
+        if (not (isinstance(val, (list, ABCSeries, ABCIndex)) or util.is_array(val))
+                or isinstance(val, ABCMultiIndex)):
+            # TODO: EA?
+            # exclude tuples, frozensets as they may be contained in an Index
+            all_arrays = False
+            break
+
+    return all_arrays
+
+
+# ------------------------------------------------------------------------------
+# Groupby-related functions
+
+# TODO: could do even better if we know something about the data. eg, index has
+# 1-min data, binner has 5-min data, then bins are just strides in index. This
+# is a general, O(max(len(values), len(binner))) method.
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def generate_bins_dt64(ndarray[int64_t, ndim=1] values, const int64_t[:] binner,
+                       object closed="left", bint hasnans=False):
+    """
+    Int64 (datetime64) version of generic python version in ``groupby.py``.
+    """
+    cdef:
+        Py_ssize_t lenidx, lenbin, i, j, bc
+        ndarray[int64_t, ndim=1] bins
+        int64_t r_bin, nat_count
+        bint right_closed = closed == "right"
+
+    nat_count = 0
+    if hasnans:
+        mask = values == NPY_NAT
+        nat_count = np.sum(mask)
+        values = values[~mask]
+
+    lenidx = len(values)
+    lenbin = len(binner)
+
+    if lenidx <= 0 or lenbin <= 0:
+        raise ValueError("Invalid length for values or for binner")
+
+    # check binner fits data
+    if values[0] < binner[0]:
+        raise ValueError("Values falls before first bin")
+
+    if values[lenidx - 1] > binner[lenbin - 1]:
+        raise ValueError("Values falls after last bin")
+
+    bins = np.empty(lenbin - 1, dtype=np.int64)
+
+    j = 0  # index into values
+    bc = 0  # bin count
+
+    # linear scan
+    if right_closed:
+        for i in range(0, lenbin - 1):
+            r_bin = binner[i + 1]
+            # count values in current bin, advance to next bin
+            while j < lenidx and values[j] <= r_bin:
+                j += 1
+            bins[bc] = j
+            bc += 1
+    else:
+        for i in range(0, lenbin - 1):
+            r_bin = binner[i + 1]
+            # count values in current bin, advance to next bin
+            while j < lenidx and values[j] < r_bin:
+                j += 1
+            bins[bc] = j
+            bc += 1
+
+    if nat_count > 0:
+        # shift bins by the number of NaT
+        bins = bins + nat_count
+        bins = np.insert(bins, 0, nat_count)
+
+    return bins
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def get_level_sorter(
+    ndarray[int64_t, ndim=1] codes, const intp_t[:] starts
+) -> ndarray:
+    """
+    Argsort for a single level of a multi-index, keeping the order of higher
+    levels unchanged. `starts` points to starts of same-key indices w.r.t
+    to leading levels; equivalent to:
+        np.hstack([codes[starts[i]:starts[i+1]].argsort(kind='mergesort')
+            + starts[i] for i in range(len(starts) - 1)])
+
+    Parameters
+    ----------
+    codes : np.ndarray[int64_t, ndim=1]
+    starts : np.ndarray[intp, ndim=1]
+
+    Returns
+    -------
+    np.ndarray[np.int, ndim=1]
+    """
+    cdef:
+        Py_ssize_t i, l, r
+        ndarray[intp_t, ndim=1] out = cnp.PyArray_EMPTY(1, codes.shape, cnp.NPY_INTP, 0)
+
+    for i in range(len(starts) - 1):
+        l, r = starts[i], starts[i + 1]
+        out[l:r] = l + codes[l:r].argsort(kind="mergesort")
+
+    return out
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def count_level_2d(const uint8_t[:, :] mask,
+                   const intp_t[:] labels,
+                   Py_ssize_t max_bin,
+                   ):
+    cdef:
+        Py_ssize_t i, j, k = mask.shape[1], n = mask.shape[0]
+        ndarray[int64_t, ndim=2] counts
+
+    counts = np.zeros((n, max_bin), dtype="i8")
+    with nogil:
+        for i in range(n):
+            for j in range(k):
+                if mask[i, j]:
+                    counts[i, labels[j]] += 1
+
+    return counts
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def generate_slices(const intp_t[:] labels, Py_ssize_t ngroups):
+    cdef:
+        Py_ssize_t i, group_size, n, start
+        intp_t lab
+        int64_t[::1] starts, ends
+
+    n = len(labels)
+
+    starts = np.zeros(ngroups, dtype=np.int64)
+    ends = np.zeros(ngroups, dtype=np.int64)
+
+    start = 0
+    group_size = 0
+    with nogil:
+        for i in range(n):
+            lab = labels[i]
+            if lab < 0:
+                start += 1
+            else:
+                group_size += 1
+                if i == n - 1 or lab != labels[i + 1]:
+                    starts[lab] = start
+                    ends[lab] = start + group_size
+                    start += group_size
+                    group_size = 0
+
+    return np.asarray(starts), np.asarray(ends)
+
+
+def indices_fast(ndarray[intp_t, ndim=1] index, const int64_t[:] labels, list keys,
+                 list sorted_labels) -> dict:
+    """
+    Parameters
+    ----------
+    index : ndarray[intp]
+    labels : ndarray[int64]
+    keys : list
+    sorted_labels : list[ndarray[int64]]
+    """
+    cdef:
+        Py_ssize_t i, j, k, lab, cur, start, n = len(labels)
+        dict result = {}
+        object tup
+
+    k = len(keys)
+
+    # Start at the first non-null entry
+    j = 0
+    for j in range(0, n):
+        if labels[j] != -1:
+            break
+    else:
+        return result
+    cur = labels[j]
+    start = j
+
+    for i in range(j+1, n):
+        lab = labels[i]
+
+        if lab != cur:
+            if lab != -1:
+                if k == 1:
+                    # When k = 1 we do not want to return a tuple as key
+                    tup = keys[0][sorted_labels[0][i - 1]]
+                else:
+                    tup = PyTuple_New(k)
+                    for j in range(k):
+                        val = keys[j][sorted_labels[j][i - 1]]
+                        PyTuple_SET_ITEM(tup, j, val)
+                        Py_INCREF(val)
+                result[tup] = index[start:i]
+            start = i
+        cur = lab
+
+    if k == 1:
+        # When k = 1 we do not want to return a tuple as key
+        tup = keys[0][sorted_labels[0][n - 1]]
+    else:
+        tup = PyTuple_New(k)
+        for j in range(k):
+            val = keys[j][sorted_labels[j][n - 1]]
+            PyTuple_SET_ITEM(tup, j, val)
+            Py_INCREF(val)
+    result[tup] = index[start:]
+
+    return result
+
+
+# core.common import for fast inference checks
+
+def is_float(obj: object) -> bool:
+    """
+    Return True if given object is float.
+
+    This method checks whether the passed object is a float type. It
+    returns `True` if the object is a float, and `False` otherwise.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check for float type.
+
+    Returns
+    -------
+    bool
+        `True` if the object is of float type, otherwise `False`.
+
+    See Also
+    --------
+    api.types.is_integer : Check if an object is of integer type.
+    api.types.is_numeric_dtype : Check if an object is of numeric type.
+
+    Examples
+    --------
+    >>> pd.api.types.is_float(1.0)
+    True
+
+    >>> pd.api.types.is_float(1)
+    False
+    """
+    return util.is_float_object(obj)
+
+
+def is_integer(obj: object) -> bool:
+    """
+    Return True if given object is integer.
+
+    This method checks whether the passed object is an integer type. It
+    returns `True` if the object is an integer, and `False` otherwise.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check for integer type.
+
+    Returns
+    -------
+    bool
+        `True` if the object is of integer type, otherwise `False`.
+
+    See Also
+    --------
+    api.types.is_float : Check if an object is of float type.
+    api.types.is_numeric_dtype : Check if an object is of numeric type.
+
+    Examples
+    --------
+    >>> pd.api.types.is_integer(1)
+    True
+
+    >>> pd.api.types.is_integer(1.0)
+    False
+    """
+    return util.is_integer_object(obj)
+
+
+def is_int_or_none(obj) -> bool:
+    """
+    Return True if given object is integer or None.
+
+    Returns
+    -------
+    bool
+    """
+    return obj is None or util.is_integer_object(obj)
+
+
+def is_bool(obj: object) -> bool:
+    """
+    Return True if given object is boolean.
+
+    Parameters
+    ----------
+    obj : object
+        Object to check.
+
+    Returns
+    -------
+    bool
+
+    See Also
+    --------
+    api.types.is_scalar : Check if the input is a scalar.
+    api.types.is_integer : Check if the input is an integer.
+    api.types.is_float : Check if the input is a float.
+
+    Examples
+    --------
+    >>> pd.api.types.is_bool(True)
+    True
+
+    >>> pd.api.types.is_bool(1)
+    False
+    """
+    return util.is_bool_object(obj)
+
+
+def is_complex(obj: object) -> bool:
+    """
+    Return True if given object is complex.
+
+    Parameters
+    ----------
+    obj : object
+        Object to check.
+
+    Returns
+    -------
+    bool
+
+    See Also
+    --------
+    api.types.is_complex_dtype: Check whether the provided array or
+                                dtype is of a complex dtype.
+    api.types.is_number: Check if the object is a number.
+    api.types.is_integer: Return True if given object is integer.
+
+    Examples
+    --------
+    >>> pd.api.types.is_complex(1 + 1j)
+    True
+
+    >>> pd.api.types.is_complex(1)
+    False
+    """
+    return util.is_complex_object(obj)
+
+
+cpdef bint is_decimal(object obj):
+    return isinstance(obj, Decimal)
+
+
+def is_list_like(obj: object, allow_sets: bool = True) -> bool:
+    """
+    Check if the object is list-like.
+
+    Objects that are considered list-like are for example Python
+    lists, tuples, sets, NumPy arrays, and Pandas Series.
+
+    Strings and datetime objects, however, are not considered list-like.
+
+    Parameters
+    ----------
+    obj : object
+        Object to check.
+    allow_sets : bool, default True
+        If this parameter is False, sets will not be considered list-like.
+
+    Returns
+    -------
+    bool
+        Whether `obj` has list-like properties.
+
+    See Also
+    --------
+    Series : One-dimensional ndarray with axis labels (including time series).
+    Index : Immutable sequence used for indexing and alignment.
+    numpy.ndarray : Array object from NumPy, which is considered list-like.
+
+    Examples
+    --------
+    >>> import datetime
+    >>> from pandas.api.types import is_list_like
+    >>> is_list_like([1, 2, 3])
+    True
+    >>> is_list_like({1, 2, 3})
+    True
+    >>> is_list_like(datetime.datetime(2017, 1, 1))
+    False
+    >>> is_list_like("foo")
+    False
+    >>> is_list_like(1)
+    False
+    >>> is_list_like(np.array([2]))
+    True
+    >>> is_list_like(np.array(2))
+    False
+    """
+    return c_is_list_like(obj, allow_sets)
+
+
+cdef bint c_is_list_like(object obj, bint allow_sets) except -1:
+    # first, performance short-cuts for the most common cases
+    if util.is_array(obj):
+        # exclude zero-dimensional numpy arrays, effectively scalars
+        return not cnp.PyArray_IsZeroDim(obj)
+    elif isinstance(obj, list):
+        return True
+    # then the generic implementation
+    return (
+        # equiv: `isinstance(obj, abc.Iterable)`
+        getattr(obj, "__iter__", None) is not None and not isinstance(obj, type)
+        # we do not count strings/unicode/bytes as list-like
+        # exclude Generic types that have __iter__
+        and not isinstance(obj, (str, bytes, _GenericAlias))
+        # exclude zero-dimensional duck-arrays, effectively scalars
+        and not (hasattr(obj, "ndim") and obj.ndim == 0)
+        # exclude sets if allow_sets is False
+        and not (allow_sets is False and isinstance(obj, abc.Set))
+    )
+
+
+def is_pyarrow_array(obj):
+    """
+    Return True if given object is a pyarrow Array or ChunkedArray.
+
+    Returns
+    -------
+    bool
+    """
+    if PYARROW_INSTALLED:
+        return isinstance(obj, (pa.Array, pa.ChunkedArray))
+    return False
+
+
+_TYPE_MAP = {
+    "categorical": "categorical",
+    "category": "categorical",
+    "int8": "integer",
+    "int16": "integer",
+    "int32": "integer",
+    "int64": "integer",
+    "i": "integer",
+    "uint8": "integer",
+    "uint16": "integer",
+    "uint32": "integer",
+    "uint64": "integer",
+    "u": "integer",
+    "float32": "floating",
+    "float64": "floating",
+    "float128": "floating",
+    "float256": "floating",
+    "f": "floating",
+    "complex64": "complex",
+    "complex128": "complex",
+    "complex256": "complex",
+    "c": "complex",
+    "string": "string",
+    str: "string",
+    "S": "bytes",
+    "U": "string",
+    "bool": "boolean",
+    "b": "boolean",
+    "datetime64[ns]": "datetime64",
+    "M": "datetime64",
+    "timedelta64[ns]": "timedelta64",
+    "m": "timedelta64",
+    "interval": "interval",
+    Period: "period",
+    datetime: "datetime64",
+    date: "date",
+    time: "time",
+    timedelta: "timedelta64",
+    Decimal: "decimal",
+    bytes: "bytes",
+}
+
+
+@cython.internal
+cdef class Seen:
+    """
+    Class for keeping track of the types of elements
+    encountered when trying to perform type conversions.
+    """
+
+    cdef:
+        bint int_             # seen_int
+        bint nat_             # seen nat
+        bint bool_            # seen_bool
+        bint null_            # seen_null
+        bint nan_             # seen_np.nan
+        bint uint_            # seen_uint (unsigned integer)
+        bint sint_            # seen_sint (signed integer)
+        bint float_           # seen_float
+        bint object_          # seen_object
+        bint complex_         # seen_complex
+        bint datetime_        # seen_datetime
+        bint coerce_numeric   # coerce data to numeric
+        bint timedelta_       # seen_timedelta
+        bint datetimetz_      # seen_datetimetz
+        bint period_          # seen_period
+        bint interval_        # seen_interval
+        bint str_             # seen_str
+
+    def __cinit__(self, bint coerce_numeric=False):
+        """
+        Initialize a Seen instance.
+
+        Parameters
+        ----------
+        coerce_numeric : bool, default False
+            Whether or not to force conversion to a numeric data type if
+            initial methods to convert to numeric fail.
+        """
+        self.int_ = False
+        self.nat_ = False
+        self.bool_ = False
+        self.null_ = False
+        self.nan_ = False
+        self.uint_ = False
+        self.sint_ = False
+        self.float_ = False
+        self.object_ = False
+        self.complex_ = False
+        self.datetime_ = False
+        self.timedelta_ = False
+        self.datetimetz_ = False
+        self.period_ = False
+        self.interval_ = False
+        self.str_ = False
+        self.coerce_numeric = coerce_numeric
+
+    cdef bint check_uint64_conflict(self) except -1:
+        """
+        Check whether we can safely convert a uint64 array to a numeric dtype.
+
+        There are two cases when conversion to numeric dtype with a uint64
+        array is not safe (and will therefore not be performed)
+
+        1) A NaN element is encountered.
+
+           uint64 cannot be safely cast to float64 due to truncation issues
+           at the extreme ends of the range.
+
+        2) A negative number is encountered.
+
+           There is no numerical dtype that can hold both negative numbers
+           and numbers greater than INT64_MAX. Hence, at least one number
+           will be improperly cast if we convert to a numeric dtype.
+
+        Returns
+        -------
+        bool
+            Whether or not we should return the original input array to avoid
+            data truncation.
+
+        Raises
+        ------
+        ValueError
+            uint64 elements were detected, and at least one of the
+            two conflict cases was also detected. However, we are
+            trying to force conversion to a numeric dtype.
+        """
+        return (self.uint_ and (self.null_ or self.sint_)
+                and not self.coerce_numeric)
+
+    cdef saw_null(self):
+        """
+        Set flags indicating that a null value was encountered.
+        """
+        self.null_ = True
+        self.float_ = True
+
+    cdef saw_int(self, object val):
+        """
+        Set flags indicating that an integer value was encountered.
+
+        In addition to setting a flag that an integer was seen, we
+        also set two flags depending on the type of integer seen:
+
+        1) sint_ : a signed numpy integer type or a negative (signed) number in the
+                   range of [-2**63, 0) was encountered
+        2) uint_ : an unsigned numpy integer type or a positive number in the range of
+                   [2**63, 2**64) was encountered
+
+        Parameters
+        ----------
+        val : Python int
+            Value with which to set the flags.
+        """
+        self.int_ = True
+        self.sint_ = (
+            self.sint_
+            or (oINT64_MIN <= val < 0)
+            or isinstance(val, cnp.signedinteger)
+        )
+        self.uint_ = (
+            self.uint_
+            or (oINT64_MAX < val <= oUINT64_MAX)
+            or isinstance(val, cnp.unsignedinteger)
+        )
+
+    @property
+    def numeric_(self):
+        return self.complex_ or self.float_ or self.int_
+
+    @property
+    def is_bool(self):
+        # i.e. not (anything but bool)
+        return self.is_bool_or_na and not (self.nan_ or self.null_)
+
+    @property
+    def is_bool_or_na(self):
+        # i.e. not (anything but bool or missing values)
+        return self.bool_ and not (
+            self.datetime_ or self.datetimetz_ or self.nat_ or self.timedelta_
+            or self.period_ or self.interval_ or self.numeric_ or self.object_
+        )
+
+
+cdef object _try_infer_map(object dtype):
+    """
+    If its in our map, just return the dtype.
+    """
+    cdef:
+        object val
+        str attr
+    for attr in ["type", "kind", "name", "base"]:
+        # Checking type before kind matters for ArrowDtype cases
+        val = getattr(dtype, attr, None)
+        if val in _TYPE_MAP:
+            return _TYPE_MAP[val]
+    return None
+
+
+def infer_dtype(value: object, skipna: bool = True) -> str:
+    """
+    Return a string label of the type of a scalar or list-like of values.
+
+    This method inspects the elements of the provided input and determines
+    classification of its data type. It is particularly useful for
+    handling heterogeneous data inputs where explicit dtype conversion may not
+    be possible or necessary.
+
+    Parameters
+    ----------
+    value : scalar, list, ndarray, or pandas type
+        The input data to infer the dtype.
+    skipna : bool, default True
+        Ignore NaN values when inferring the type.
+
+    Returns
+    -------
+    str
+        Describing the common type of the input data.
+    Results can include:
+
+    - string
+    - bytes
+    - floating
+    - integer
+    - mixed-integer
+    - mixed-integer-float
+    - decimal
+    - complex
+    - categorical
+    - boolean
+    - datetime64
+    - datetime
+    - date
+    - timedelta64
+    - timedelta
+    - time
+    - period
+    - mixed
+    - unknown-array
+
+    Raises
+    ------
+    TypeError
+        If ndarray-like but cannot infer the dtype
+
+    See Also
+    --------
+    api.types.is_scalar : Check if the input is a scalar.
+    api.types.is_list_like : Check if the input is list-like.
+    api.types.is_integer : Check if the input is an integer.
+    api.types.is_float : Check if the input is a float.
+    api.types.is_bool : Check if the input is a boolean.
+
+    Notes
+    -----
+    - 'mixed' is the catchall for anything that is not otherwise
+      specialized
+    - 'mixed-integer-float' are floats and integers
+    - 'mixed-integer' are integers mixed with non-integers
+    - 'unknown-array' is the catchall for something that *is* an array (has
+      a dtype attribute), but has a dtype unknown to pandas (e.g. external
+      extension array)
+
+    Examples
+    --------
+    >>> from pandas.api.types import infer_dtype
+    >>> infer_dtype(['foo', 'bar'])
+    'string'
+
+    >>> infer_dtype(['a', np.nan, 'b'], skipna=True)
+    'string'
+
+    >>> infer_dtype(['a', np.nan, 'b'], skipna=False)
+    'mixed'
+
+    >>> infer_dtype([b'foo', b'bar'])
+    'bytes'
+
+    >>> infer_dtype([1, 2, 3])
+    'integer'
+
+    >>> infer_dtype([1, 2, 3.5])
+    'mixed-integer-float'
+
+    >>> infer_dtype([1.0, 2.0, 3.5])
+    'floating'
+
+    >>> infer_dtype(['a', 1])
+    'mixed-integer'
+
+    >>> from decimal import Decimal
+    >>> infer_dtype([Decimal(1), Decimal(2.0)])
+    'decimal'
+
+    >>> infer_dtype([True, False])
+    'boolean'
+
+    >>> infer_dtype([True, False, np.nan])
+    'boolean'
+
+    >>> infer_dtype([pd.Timestamp('20130101')])
+    'datetime'
+
+    >>> import datetime
+    >>> infer_dtype([datetime.date(2013, 1, 1)])
+    'date'
+
+    >>> infer_dtype([np.datetime64('2013-01-01')])
+    'datetime64'
+
+    >>> infer_dtype([datetime.timedelta(0, 1, 1)])
+    'timedelta'
+
+    >>> infer_dtype(pd.Series(list('aabc')).astype('category'))
+    'categorical'
+    """
+    cdef:
+        Py_ssize_t i, n
+        object val
+        ndarray values
+        bint seen_pdnat = False
+        bint seen_val = False
+        flatiter it
+
+    if util.is_array(value):
+        values = value
+    elif hasattr(type(value), "inferred_type") and skipna is False:
+        # Index, use the cached attribute if possible, populate the cache otherwise
+        return value.inferred_type
+    elif hasattr(value, "dtype"):
+        inferred = _try_infer_map(value.dtype)
+        if inferred is not None:
+            return inferred
+        elif not cnp.PyArray_DescrCheck(value.dtype):
+            return "unknown-array"
+        # Unwrap Series/Index
+        values = np.asarray(value)
+    else:
+        if not isinstance(value, list):
+            value = list(value)
+        if not value:
+            return "empty"
+
+        from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+        values = construct_1d_object_array_from_listlike(value)
+
+    inferred = _try_infer_map(values.dtype)
+    if inferred is not None:
+        # Anything other than object-dtype should return here.
+        return inferred
+
+    if values.descr.type_num != NPY_OBJECT:
+        # i.e. values.dtype != np.object_
+        # This should not be reached
+        values = values.astype(object)
+
+    n = cnp.PyArray_SIZE(values)
+    if n == 0:
+        return "empty"
+
+    # Iterate until we find our first valid value. We will use this
+    #  value to decide which of the is_foo_array functions to call.
+    it = PyArray_IterNew(values)
+    for i in range(n):
+        # The PyArray_GETITEM and PyArray_ITER_NEXT are faster
+        #  equivalents to `val = values[i]`
+        val = PyArray_GETITEM(values, PyArray_ITER_DATA(it))
+        PyArray_ITER_NEXT(it)
+
+        # do not use checknull to keep
+        # np.datetime64('nat') and np.timedelta64('nat')
+        if val is None or util.is_nan(val) or val is C_NA:
+            pass
+        elif val is NaT:
+            seen_pdnat = True
+        else:
+            seen_val = True
+            break
+
+    # if all values are nan/NaT
+    if seen_val is False and seen_pdnat is True:
+        return "datetime"
+        # float/object nan is handled in latter logic
+    if seen_val is False and skipna:
+        return "empty"
+
+    if cnp.is_datetime64_object(val):
+        if is_datetime64_array(values, skipna=skipna):
+            return "datetime64"
+
+    elif is_timedelta(val):
+        if is_timedelta_or_timedelta64_array(values, skipna=skipna):
+            return "timedelta"
+
+    elif util.is_integer_object(val):
+        # ordering matters here; this check must come after the is_timedelta
+        #  check otherwise numpy timedelta64 objects would come through here
+
+        if is_integer_array(values, skipna=skipna):
+            return "integer"
+        elif is_integer_float_array(values, skipna=skipna):
+            if is_integer_na_array(values, skipna=skipna):
+                return "integer-na"
+            else:
+                return "mixed-integer-float"
+        return "mixed-integer"
+
+    elif PyDateTime_Check(val):
+        if is_datetime_array(values, skipna=skipna):
+            return "datetime"
+        elif is_date_array(values, skipna=skipna):
+            return "date"
+
+    elif PyDate_Check(val):
+        if is_date_array(values, skipna=skipna):
+            return "date"
+
+    elif PyTime_Check(val):
+        if is_time_array(values, skipna=skipna):
+            return "time"
+
+    elif is_decimal(val):
+        if is_decimal_array(values, skipna=skipna):
+            return "decimal"
+
+    elif util.is_complex_object(val):
+        if is_complex_array(values):
+            return "complex"
+
+    elif util.is_float_object(val):
+        if is_float_array(values):
+            return "floating"
+        elif is_integer_float_array(values, skipna=skipna):
+            if is_integer_na_array(values, skipna=skipna):
+                return "integer-na"
+            else:
+                return "mixed-integer-float"
+
+    elif util.is_bool_object(val):
+        if is_bool_array(values, skipna=skipna):
+            return "boolean"
+
+    elif isinstance(val, str):
+        if is_string_array(values, skipna=skipna):
+            return "string"
+
+    elif isinstance(val, bytes):
+        if is_bytes_array(values, skipna=skipna):
+            return "bytes"
+
+    elif is_period_object(val):
+        if is_period_array(values, skipna=skipna):
+            return "period"
+
+    elif isinstance(val, Interval):
+        if is_interval_array(values):
+            return "interval"
+
+    cnp.PyArray_ITER_RESET(it)
+    for i in range(n):
+        val = PyArray_GETITEM(values, PyArray_ITER_DATA(it))
+        PyArray_ITER_NEXT(it)
+
+        if util.is_integer_object(val):
+            return "mixed-integer"
+
+    return "mixed"
+
+
+cdef bint is_timedelta(object o):
+    return PyDelta_Check(o) or cnp.is_timedelta64_object(o)
+
+
+@cython.internal
+cdef class Validator:
+
+    cdef:
+        Py_ssize_t n
+        cnp.dtype dtype
+        bint skipna
+
+    def __cinit__(self, Py_ssize_t n, cnp.dtype dtype=np.dtype(np.object_),
+                  bint skipna=False):
+        self.n = n
+        self.dtype = dtype
+        self.skipna = skipna
+
+    cdef bint validate(self, ndarray values) except -1:
+        if not self.n:
+            return False
+
+        if self.is_array_typed():
+            # i.e. this ndarray is already of the desired dtype
+            return True
+        elif self.dtype.type_num == NPY_OBJECT:
+            if self.skipna:
+                return self._validate_skipna(values)
+            else:
+                return self._validate(values)
+        else:
+            return False
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    cdef bint _validate(self, ndarray values) except -1:
+        cdef:
+            Py_ssize_t i
+            Py_ssize_t n = values.size
+            flatiter it = PyArray_IterNew(values)
+
+        for i in range(n):
+            # The PyArray_GETITEM and PyArray_ITER_NEXT are faster
+            #  equivalents to `val = values[i]`
+            val = PyArray_GETITEM(values, PyArray_ITER_DATA(it))
+            PyArray_ITER_NEXT(it)
+            if not self.is_valid(val):
+                return False
+
+        return True
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    cdef bint _validate_skipna(self, ndarray values) except -1:
+        cdef:
+            Py_ssize_t i
+            Py_ssize_t n = values.size
+            flatiter it = PyArray_IterNew(values)
+
+        for i in range(n):
+            # The PyArray_GETITEM and PyArray_ITER_NEXT are faster
+            #  equivalents to `val = values[i]`
+            val = PyArray_GETITEM(values, PyArray_ITER_DATA(it))
+            PyArray_ITER_NEXT(it)
+            if not self.is_valid_skipna(val):
+                return False
+
+        return True
+
+    cdef bint is_valid(self, object value) except -1:
+        return self.is_value_typed(value)
+
+    cdef bint is_valid_skipna(self, object value) except -1:
+        return self.is_valid(value) or self.is_valid_null(value)
+
+    cdef bint is_value_typed(self, object value) except -1:
+        raise NotImplementedError(f"{type(self).__name__} child class "
+                                  "must define is_value_typed")
+
+    cdef bint is_valid_null(self, object value) except -1:
+        return value is None or value is C_NA or util.is_nan(value)
+        # TODO: include decimal NA?
+
+    cdef bint is_array_typed(self) except -1:
+        return False
+
+
+@cython.internal
+cdef class BoolValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return util.is_bool_object(value)
+
+    cdef bint is_array_typed(self) except -1:
+        return cnp.PyDataType_ISBOOL(self.dtype)
+
+
+cpdef bint is_bool_array(ndarray values, bint skipna=False):
+    cdef:
+        BoolValidator validator = BoolValidator(values.size,
+                                                values.dtype,
+                                                skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class IntegerValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return util.is_integer_object(value)
+
+    cdef bint is_array_typed(self) except -1:
+        return cnp.PyDataType_ISINTEGER(self.dtype)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_integer_array(ndarray values, bint skipna=True):
+    cdef:
+        IntegerValidator validator = IntegerValidator(values.size,
+                                                      values.dtype,
+                                                      skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class IntegerNaValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return (util.is_integer_object(value)
+                or (util.is_nan(value) and util.is_float_object(value)))
+
+
+cdef bint is_integer_na_array(ndarray values, bint skipna=True):
+    cdef:
+        IntegerNaValidator validator = IntegerNaValidator(values.size,
+                                                          values.dtype, skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class IntegerFloatValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return util.is_integer_object(value) or util.is_float_object(value)
+
+    cdef bint is_array_typed(self) except -1:
+        return cnp.PyDataType_ISINTEGER(self.dtype)
+
+
+cdef bint is_integer_float_array(ndarray values, bint skipna=True):
+    cdef:
+        IntegerFloatValidator validator = IntegerFloatValidator(values.size,
+                                                                values.dtype,
+                                                                skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class FloatValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return util.is_float_object(value)
+
+    cdef bint is_array_typed(self) except -1:
+        return cnp.PyDataType_ISFLOAT(self.dtype)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_float_array(ndarray values):
+    cdef:
+        FloatValidator validator = FloatValidator(values.size, values.dtype)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class ComplexValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return (
+            util.is_complex_object(value)
+            or (util.is_float_object(value) and is_nan(value))
+        )
+
+    cdef bint is_array_typed(self) except -1:
+        return cnp.PyDataType_ISCOMPLEX(self.dtype)
+
+
+cdef bint is_complex_array(ndarray values):
+    cdef:
+        ComplexValidator validator = ComplexValidator(values.size, values.dtype)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class DecimalValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return is_decimal(value)
+
+
+cdef bint is_decimal_array(ndarray values, bint skipna=False):
+    cdef:
+        DecimalValidator validator = DecimalValidator(
+            values.size, values.dtype, skipna=skipna
+        )
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class StringValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return isinstance(value, str)
+
+    cdef bint is_array_typed(self) except -1:
+        return self.dtype.type_num == cnp.NPY_UNICODE
+
+
+cpdef bint is_string_array(ndarray values, bint skipna=False):
+    cdef:
+        StringValidator validator = StringValidator(values.size,
+                                                    values.dtype,
+                                                    skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class BytesValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return isinstance(value, bytes)
+
+    cdef bint is_array_typed(self) except -1:
+        return self.dtype.type_num == cnp.NPY_STRING
+
+
+cdef bint is_bytes_array(ndarray values, bint skipna=False):
+    cdef:
+        BytesValidator validator = BytesValidator(values.size, values.dtype,
+                                                  skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class TemporalValidator(Validator):
+    cdef:
+        bint all_generic_na
+
+    def __cinit__(self, Py_ssize_t n, cnp.dtype dtype=np.dtype(np.object_),
+                  bint skipna=False):
+        self.n = n
+        self.dtype = dtype
+        self.skipna = skipna
+        self.all_generic_na = True
+
+    cdef bint is_valid(self, object value) except -1:
+        return self.is_value_typed(value) or self.is_valid_null(value)
+
+    cdef bint is_valid_null(self, object value) except -1:
+        raise NotImplementedError(f"{type(self).__name__} child class "
+                                  "must define is_valid_null")
+
+    cdef bint is_valid_skipna(self, object value) except -1:
+        cdef:
+            bint is_typed_null = self.is_valid_null(value)
+            bint is_generic_null = value is None or util.is_nan(value)
+        if not is_generic_null:
+            self.all_generic_na = False
+        return self.is_value_typed(value) or is_typed_null or is_generic_null
+
+    cdef bint _validate_skipna(self, ndarray values) except -1:
+        """
+        If we _only_ saw non-dtype-specific NA values, even if they are valid
+        for this dtype, we do not infer this dtype.
+        """
+        return Validator._validate_skipna(self, values) and not self.all_generic_na
+
+
+@cython.internal
+cdef class DatetimeValidator(TemporalValidator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return PyDateTime_Check(value)
+
+    cdef bint is_valid_null(self, object value) except -1:
+        return is_null_datetime64(value)
+
+
+cpdef bint is_datetime_array(ndarray values, bint skipna=True):
+    cdef:
+        DatetimeValidator validator = DatetimeValidator(values.size,
+                                                        skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class Datetime64Validator(DatetimeValidator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return cnp.is_datetime64_object(value)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_datetime64_array(ndarray values, bint skipna=True):
+    cdef:
+        Datetime64Validator validator = Datetime64Validator(values.size,
+                                                            skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class AnyDatetimeValidator(DatetimeValidator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return cnp.is_datetime64_object(value) or (
+            PyDateTime_Check(value) and value.tzinfo is None
+        )
+
+
+cdef bint is_datetime_or_datetime64_array(ndarray values, bint skipna=True):
+    cdef:
+        AnyDatetimeValidator validator = AnyDatetimeValidator(values.size,
+                                                              skipna=skipna)
+    return validator.validate(values)
+
+
+# Note: only python-exposed for tests
+def is_datetime_with_singletz_array(values: ndarray) -> bool:
+    """
+    Check values have the same tzinfo attribute.
+    Doesn't check values are datetime-like types.
+    """
+    cdef:
+        Py_ssize_t i = 0, j, n = values.size
+        object base_val, base_tz, val, tz
+
+    if n == 0:
+        return False
+
+    # Get a reference timezone to compare with the rest of the tzs in the array
+    for i in range(n):
+        base_val = values[i]
+        if base_val is not NaT and base_val is not None and not util.is_nan(base_val):
+            base_tz = getattr(base_val, "tzinfo", None)
+            break
+
+    for j in range(i, n):
+        # Compare val's timezone with the reference timezone
+        # NaT can coexist with tz-aware datetimes, so skip if encountered
+        val = values[j]
+        if val is not NaT and val is not None and not util.is_nan(val):
+            tz = getattr(val, "tzinfo", None)
+            if not tz_compare(base_tz, tz):
+                return False
+
+    # Note: we should only be called if a tzaware datetime has been seen,
+    #  so base_tz should always be set at this point.
+    return True
+
+
+@cython.internal
+cdef class TimedeltaValidator(TemporalValidator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return PyDelta_Check(value)
+
+    cdef bint is_valid_null(self, object value) except -1:
+        return is_null_timedelta64(value)
+
+
+@cython.internal
+cdef class AnyTimedeltaValidator(TimedeltaValidator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return is_timedelta(value)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_timedelta_or_timedelta64_array(ndarray values, bint skipna=True):
+    """
+    Infer with timedeltas and/or nat/none.
+    """
+    cdef:
+        AnyTimedeltaValidator validator = AnyTimedeltaValidator(values.size,
+                                                                skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class DateValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return PyDate_Check(value)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_date_array(ndarray values, bint skipna=False):
+    cdef:
+        DateValidator validator = DateValidator(values.size, skipna=skipna)
+    return validator.validate(values)
+
+
+@cython.internal
+cdef class TimeValidator(Validator):
+    cdef bint is_value_typed(self, object value) except -1:
+        return PyTime_Check(value)
+
+
+# Note: only python-exposed for tests
+cpdef bint is_time_array(ndarray values, bint skipna=False):
+    cdef:
+        TimeValidator validator = TimeValidator(values.size, skipna=skipna)
+    return validator.validate(values)
+
+
+# FIXME: actually use skipna
+cdef bint is_period_array(ndarray values, bint skipna=True):
+    """
+    Is this an ndarray of Period objects (or NaT) with a single `freq`?
+    """
+    # values should be object-dtype, but ndarray[object] assumes 1D, while
+    #  this _may_ be 2D.
+    cdef:
+        Py_ssize_t i, N = values.size
+        int dtype_code = -10000  # i.e. c_FreqGroup.FR_UND
+        object val
+        flatiter it
+
+    if N == 0:
+        return False
+
+    it = PyArray_IterNew(values)
+    for i in range(N):
+        # The PyArray_GETITEM and PyArray_ITER_NEXT are faster
+        #  equivalents to `val = values[i]`
+        val = PyArray_GETITEM(values, PyArray_ITER_DATA(it))
+        PyArray_ITER_NEXT(it)
+
+        if is_period_object(val):
+            if dtype_code == -10000:
+                dtype_code = val._dtype._dtype_code
+            elif dtype_code != val._dtype._dtype_code:
+                # mismatched freqs
+                return False
+        elif checknull_with_nat(val):
+            pass
+        else:
+            # Not a Period or NaT-like
+            return False
+
+    if dtype_code == -10000:
+        # we saw all-NaTs, no actual Periods
+        return False
+    return True
+
+
+# Note: only python-exposed for tests
+cpdef bint is_interval_array(ndarray values):
+    """
+    Is this an ndarray of Interval (or np.nan) with a single dtype?
+    """
+    cdef:
+        Py_ssize_t i, n = values.size
+        str closed = None
+        bint numeric = False
+        bint dt64 = False
+        bint td64 = False
+        object val
+
+    if n == 0:
+        return False
+
+    for i in range(n):
+        val = values[i]
+
+        if isinstance(val, Interval):
+            if closed is None:
+                closed = val.closed
+                numeric = (
+                    util.is_float_object(val.left)
+                    or util.is_integer_object(val.left)
+                )
+                td64 = is_timedelta(val.left)
+                dt64 = PyDateTime_Check(val.left)
+            elif val.closed != closed:
+                # mismatched closedness
+                return False
+            elif numeric:
+                if not (
+                    util.is_float_object(val.left)
+                    or util.is_integer_object(val.left)
+                ):
+                    # i.e. datetime64 or timedelta64
+                    return False
+            elif td64:
+                if not is_timedelta(val.left):
+                    return False
+            elif dt64:
+                if not PyDateTime_Check(val.left):
+                    return False
+            else:
+                raise ValueError(val)
+        elif util.is_nan(val) or val is None:
+            pass
+        else:
+            return False
+
+    if closed is None:
+        # we saw all-NAs, no actual Intervals
+        return False
+    return True
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def maybe_convert_numeric(
+    ndarray[object, ndim=1] values,
+    set na_values,
+    bint convert_empty=True,
+    bint coerce_numeric=False,
+    bint convert_to_masked_nullable=False,
+) -> tuple[np.ndarray, np.ndarray | None]:
+    """
+    Convert object array to a numeric array if possible.
+
+    Parameters
+    ----------
+    values : ndarray[object]
+        Array of object elements to convert.
+    na_values : set
+        Set of values that should be interpreted as NaN.
+    convert_empty : bool, default True
+        If an empty array-like object is encountered, whether to interpret
+        that element as NaN or not. If set to False, a ValueError will be
+        raised if such an element is encountered and 'coerce_numeric' is False.
+    coerce_numeric : bool, default False
+        If initial attempts to convert to numeric have failed, whether to
+        force conversion to numeric via alternative methods or by setting the
+        element to NaN. Otherwise, an Exception will be raised when such an
+        element is encountered.
+
+        This boolean also has an impact on how conversion behaves when a
+        numeric array has no suitable numerical dtype to return (i.e. uint64,
+        int32, uint8). If set to False, the original object array will be
+        returned. Otherwise, a ValueError will be raised.
+    convert_to_masked_nullable : bool, default False
+        Whether to return a mask for the converted values. This also disables
+        upcasting for ints with nulls to float64.
+    Returns
+    -------
+    np.ndarray
+        Array of converted object values to numerical ones.
+
+    Optional[np.ndarray]
+        If convert_to_masked_nullable is True,
+        returns a boolean mask for the converted values, otherwise returns None.
+    """
+    if len(values) == 0:
+        return (np.array([], dtype="i8"), None)
+
+    # fastpath for ints - try to convert all based on first value
+    cdef:
+        object val = values[0]
+
+    if util.is_integer_object(val):
+        try:
+            maybe_ints = values.astype("i8")
+            if (maybe_ints == values).all():
+                return (maybe_ints, None)
+        except (ValueError, OverflowError, TypeError):
+            pass
+
+    # Otherwise, iterate and do full inference.
+    cdef:
+        int maybe_int
+        Py_ssize_t i, n = values.size
+        Seen seen = Seen(coerce_numeric)
+        ndarray[float64_t, ndim=1] floats = cnp.PyArray_EMPTY(
+            1, values.shape, cnp.NPY_FLOAT64, 0
+        )
+        ndarray[complex128_t, ndim=1] complexes = cnp.PyArray_EMPTY(
+            1, values.shape, cnp.NPY_COMPLEX128, 0
+        )
+        ndarray[int64_t, ndim=1] ints = cnp.PyArray_EMPTY(
+            1, values.shape, cnp.NPY_INT64, 0
+        )
+        ndarray[uint64_t, ndim=1] uints = cnp.PyArray_EMPTY(
+            1, values.shape, cnp.NPY_UINT64, 0
+        )
+        ndarray[uint8_t, ndim=1] bools = cnp.PyArray_EMPTY(
+            1, values.shape, cnp.NPY_UINT8, 0
+        )
+        ndarray[uint8_t, ndim=1] mask = np.zeros(n, dtype="u1")
+        float64_t fval
+        bint allow_null_in_int = convert_to_masked_nullable
+
+    for i in range(n):
+        val = values[i]
+        # We only want to disable NaNs showing as float if
+        # a) convert_to_masked_nullable = True
+        # b) no floats have been seen ( assuming an int shows up later )
+        # However, if no ints present (all null array), we need to return floats
+        allow_null_in_int = convert_to_masked_nullable and not seen.float_
+
+        if val.__hash__ is not None and val in na_values:
+            if allow_null_in_int:
+                seen.null_ = True
+                mask[i] = 1
+            else:
+                if convert_to_masked_nullable:
+                    mask[i] = 1
+                seen.saw_null()
+            floats[i] = complexes[i] = NaN
+        elif util.is_float_object(val):
+            fval = val
+            if fval != fval:
+                seen.null_ = True
+                if allow_null_in_int:
+                    mask[i] = 1
+                else:
+                    if convert_to_masked_nullable:
+                        mask[i] = 1
+                    seen.float_ = True
+            else:
+                seen.float_ = True
+            floats[i] = complexes[i] = fval
+        elif util.is_integer_object(val):
+            floats[i] = complexes[i] = val
+
+            val = int(val)
+            seen.saw_int(val)
+
+            if val >= 0:
+                if val <= oUINT64_MAX:
+                    uints[i] = val
+                else:
+                    seen.float_ = True
+
+            if oINT64_MIN <= val <= oINT64_MAX:
+                ints[i] = val
+
+            if val < oINT64_MIN or (seen.sint_ and seen.uint_):
+                seen.float_ = True
+
+        elif util.is_bool_object(val):
+            floats[i] = uints[i] = ints[i] = bools[i] = val
+            seen.bool_ = True
+        elif val is None or val is C_NA:
+            if allow_null_in_int:
+                seen.null_ = True
+                mask[i] = 1
+            else:
+                if convert_to_masked_nullable:
+                    mask[i] = 1
+                seen.saw_null()
+            floats[i] = complexes[i] = NaN
+        elif hasattr(val, "__len__") and len(val) == 0:
+            if convert_empty or seen.coerce_numeric:
+                seen.saw_null()
+                floats[i] = complexes[i] = NaN
+                mask[i] = 1
+            else:
+                raise ValueError("Empty string encountered")
+        elif util.is_complex_object(val):
+            complexes[i] = val
+            seen.complex_ = True
+        elif is_decimal(val):
+            floats[i] = complexes[i] = val
+            seen.float_ = True
+        else:
+            try:
+                floatify(val, &fval, &maybe_int)
+
+                if fval in na_values:
+                    seen.saw_null()
+                    floats[i] = complexes[i] = NaN
+                    mask[i] = 1
+                else:
+                    if fval != fval:
+                        seen.null_ = True
+                        mask[i] = 1
+
+                    floats[i] = fval
+
+                if maybe_int:
+                    as_int = int(val)
+
+                    if as_int in na_values:
+                        mask[i] = 1
+                        seen.null_ = True
+                        if not allow_null_in_int:
+                            seen.float_ = True
+                    else:
+                        seen.saw_int(as_int)
+
+                    if as_int not in na_values:
+                        if as_int < oINT64_MIN or as_int > oUINT64_MAX:
+                            if seen.coerce_numeric:
+                                seen.float_ = True
+                            else:
+                                raise ValueError("Integer out of range.")
+                        else:
+                            if as_int >= 0:
+                                uints[i] = as_int
+
+                            if as_int <= oINT64_MAX:
+                                ints[i] = as_int
+
+                    seen.float_ = seen.float_ or (seen.uint_ and seen.sint_)
+                else:
+                    seen.float_ = True
+            except (TypeError, ValueError) as err:
+                if not seen.coerce_numeric:
+                    raise type(err)(f"{err} at position {i}")
+
+                mask[i] = 1
+
+                if allow_null_in_int:
+                    seen.null_ = True
+                else:
+                    seen.saw_null()
+                    floats[i] = NaN
+
+    if seen.check_uint64_conflict():
+        return (values, None)
+
+    # This occurs since we disabled float nulls showing as null in anticipation
+    # of seeing ints that were never seen. So then, we return float
+    if allow_null_in_int and seen.null_ and not seen.int_ and not seen.bool_:
+        seen.float_ = True
+
+    if seen.complex_:
+        return (complexes, None)
+    elif seen.float_:
+        if seen.null_ and convert_to_masked_nullable:
+            return (floats, mask.view(np.bool_))
+        return (floats, None)
+    elif seen.int_:
+        if seen.null_ and convert_to_masked_nullable:
+            if seen.uint_:
+                return (uints, mask.view(np.bool_))
+            else:
+                return (ints, mask.view(np.bool_))
+        if seen.uint_:
+            return (uints, None)
+        else:
+            return (ints, None)
+    elif seen.bool_:
+        if allow_null_in_int:
+            return (bools.view(np.bool_), mask.view(np.bool_))
+        return (bools.view(np.bool_), None)
+    elif seen.uint_:
+        return (uints, None)
+    return (ints, None)
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def maybe_convert_objects(ndarray[object] objects,
+                          *,
+                          bint try_float=False,
+                          bint safe=False,
+                          bint convert_numeric=True,  # NB: different default!
+                          bint convert_to_nullable_dtype=False,
+                          bint convert_non_numeric=False,
+                          object dtype_if_all_nat=None) -> "ArrayLike":
+    """
+    Type inference function-- convert object array to proper dtype
+
+    Parameters
+    ----------
+    objects : ndarray[object]
+        Array of object elements to convert.
+    try_float : bool, default False
+        If an array-like object contains only float or NaN values is
+        encountered, whether to convert and return an array of float dtype.
+    safe : bool, default False
+        Whether to upcast numeric type (e.g. int cast to float). If set to
+        True, no upcasting will be performed.
+    convert_numeric : bool, default True
+        Whether to convert numeric entries.
+    convert_to_nullable_dtype : bool, default False
+        If an array-like object contains only integer or boolean values (and NaN) is
+        encountered, whether to convert and return an Boolean/IntegerArray.
+    convert_non_numeric : bool, default False
+        Whether to convert datetime, timedelta, period, interval types.
+    dtype_if_all_nat : np.dtype, ExtensionDtype, or None, default None
+        Dtype to cast to if we have all-NaT.
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+        Array of converted object values to more specific dtypes if applicable.
+    """
+    cdef:
+        Py_ssize_t i, n, itemsize_max = 0
+        ndarray[float64_t] floats
+        ndarray[complex128_t] complexes
+        ndarray[int64_t] ints
+        ndarray[uint64_t] uints
+        ndarray[uint8_t] bools
+        ndarray[uint8_t] mask
+        Seen seen = Seen()
+        object val
+        float64_t fnan = NaN
+
+    if dtype_if_all_nat is not None:
+        # in practice we don't expect to ever pass dtype_if_all_nat
+        #  without both convert_non_numeric, so disallow
+        #  it to avoid needing to handle it below.
+        if not convert_non_numeric:
+            raise ValueError(
+                "Cannot specify 'dtype_if_all_nat' without convert_non_numeric=True"
+            )
+
+    n = len(objects)
+
+    floats = cnp.PyArray_EMPTY(1, objects.shape, cnp.NPY_FLOAT64, 0)
+    complexes = cnp.PyArray_EMPTY(1, objects.shape, cnp.NPY_COMPLEX128, 0)
+    ints = cnp.PyArray_EMPTY(1, objects.shape, cnp.NPY_INT64, 0)
+    uints = cnp.PyArray_EMPTY(1, objects.shape, cnp.NPY_UINT64, 0)
+    bools = cnp.PyArray_EMPTY(1, objects.shape, cnp.NPY_UINT8, 0)
+    mask = np.full(n, False)
+
+    for i in range(n):
+        val = objects[i]
+        if itemsize_max != -1:
+            itemsize = get_itemsize(val)
+            if itemsize > itemsize_max or itemsize == -1:
+                itemsize_max = itemsize
+
+        if val is None:
+            seen.null_ = True
+            floats[i] = complexes[i] = fnan
+            mask[i] = True
+        elif val is NaT:
+            seen.nat_ = True
+            if not convert_non_numeric:
+                seen.object_ = True
+                break
+        elif util.is_nan(val):
+            seen.nan_ = True
+            mask[i] = True
+            if util.is_complex_object(val):
+                floats[i] = fnan
+                complexes[i] = val
+                seen.complex_ = True
+                if not convert_numeric:
+                    break
+            else:
+                floats[i] = complexes[i] = val
+        elif util.is_bool_object(val):
+            seen.bool_ = True
+            bools[i] = val
+            if not convert_numeric:
+                break
+        elif util.is_float_object(val):
+            floats[i] = complexes[i] = val
+            seen.float_ = True
+            if not convert_numeric:
+                break
+        elif is_timedelta(val):
+            if convert_non_numeric:
+                seen.timedelta_ = True
+                try:
+                    convert_to_timedelta64(val, "ns")
+                except OutOfBoundsTimedelta:
+                    seen.object_ = True
+                    break
+                break
+            else:
+                seen.object_ = True
+                break
+        elif util.is_integer_object(val):
+            seen.int_ = True
+            floats[i] = <float64_t>val
+            complexes[i] = <double complex>val
+            if not seen.null_ or convert_to_nullable_dtype:
+                seen.saw_int(val)
+
+                if ((seen.uint_ and seen.sint_) or
+                        val > oUINT64_MAX or val < oINT64_MIN):
+                    seen.object_ = True
+                    break
+
+                if seen.uint_:
+                    uints[i] = val
+                elif seen.sint_:
+                    ints[i] = val
+                else:
+                    uints[i] = val
+                    ints[i] = val
+            if not convert_numeric:
+                break
+
+        elif util.is_complex_object(val):
+            complexes[i] = val
+            seen.complex_ = True
+            if not convert_numeric:
+                break
+        elif PyDateTime_Check(val) or cnp.is_datetime64_object(val):
+
+            # if we have an tz's attached then return the objects
+            if convert_non_numeric:
+                if getattr(val, "tzinfo", None) is not None:
+                    seen.datetimetz_ = True
+                    break
+                else:
+                    seen.datetime_ = True
+                    try:
+                        convert_to_tsobject(val, None, None, 0, 0)
+                    except OutOfBoundsDatetime:
+                        # e.g. test_out_of_s_bounds_datetime64
+                        seen.object_ = True
+                        break
+            else:
+                seen.object_ = True
+                break
+        elif is_period_object(val):
+            if convert_non_numeric:
+                seen.period_ = True
+                break
+            else:
+                seen.object_ = True
+                break
+        elif try_float and not isinstance(val, str):
+            # this will convert Decimal objects
+            try:
+                floats[i] = float(val)
+                complexes[i] = complex(val)
+                seen.float_ = True
+            except (ValueError, TypeError):
+                seen.object_ = True
+                break
+        elif isinstance(val, Interval):
+            if convert_non_numeric:
+                seen.interval_ = True
+                break
+            else:
+                seen.object_ = True
+                break
+        elif isinstance(val, str):
+            if convert_non_numeric:
+                seen.str_ = True
+                break
+            else:
+                seen.object_ = True
+                break
+        elif val is C_NA:
+            if convert_to_nullable_dtype:
+                seen.null_ = True
+                mask[i] = True
+            else:
+                seen.object_ = True
+            continue
+        else:
+            seen.object_ = True
+            break
+
+    # we try to coerce datetime w/tz but must all have the same tz
+    if seen.datetimetz_:
+        if is_datetime_with_singletz_array(objects):
+            from pandas import DatetimeIndex
+
+            try:
+                dti = DatetimeIndex(objects)
+            except OutOfBoundsDatetime:
+                # e.g. test_to_datetime_cache_coerce_50_lines_outofbounds
+                pass
+            else:
+                # unbox to DatetimeArray
+                return dti._data
+        seen.object_ = True
+
+    elif seen.datetime_:
+        if is_datetime_or_datetime64_array(objects):
+            from pandas import DatetimeIndex
+
+            try:
+                dti = DatetimeIndex(objects)
+            except OutOfBoundsDatetime:
+                pass
+            else:
+                # unbox to ndarray[datetime64[ns]]
+                return dti._data._ndarray
+        seen.object_ = True
+
+    elif seen.timedelta_:
+        if is_timedelta_or_timedelta64_array(objects):
+            from pandas import TimedeltaIndex
+
+            try:
+                tdi = TimedeltaIndex(objects)
+            except OutOfBoundsTimedelta:
+                pass
+            else:
+                # unbox to ndarray[timedelta64[ns]]
+                return tdi._data._ndarray
+        seen.object_ = True
+
+    elif seen.period_:
+        if is_period_array(objects):
+            from pandas import PeriodIndex
+            pi = PeriodIndex(objects)
+
+            # unbox to PeriodArray
+            return pi._data
+        seen.object_ = True
+
+    elif seen.str_:
+        if convert_to_nullable_dtype and is_string_array(objects, skipna=True):
+            from pandas.core.arrays.string_ import StringDtype
+
+            dtype = StringDtype()
+            return dtype.construct_array_type()._from_sequence(objects, dtype=dtype)
+
+        elif using_string_dtype() and is_string_array(objects, skipna=True):
+            from pandas.core.arrays.string_ import StringDtype
+
+            dtype = StringDtype(na_value=np.nan)
+            return dtype.construct_array_type()._from_sequence(objects, dtype=dtype)
+
+        seen.object_ = True
+    elif seen.interval_:
+        if is_interval_array(objects):
+            from pandas import IntervalIndex
+            ii = IntervalIndex(objects)
+
+            # unbox to IntervalArray
+            return ii._data
+
+        seen.object_ = True
+
+    elif seen.nat_:
+        if not seen.object_ and not seen.numeric_ and not seen.bool_:
+            # all NaT, None, or nan (at least one NaT)
+            # see GH#49340 for discussion of desired behavior
+            dtype = dtype_if_all_nat
+            if cnp.PyArray_DescrCheck(dtype):
+                # i.e. isinstance(dtype, np.dtype)
+                if dtype.kind not in "mM":
+                    raise ValueError(dtype)
+                else:
+                    res = np.empty((<object>objects).shape, dtype=dtype)
+                    res[:] = NPY_NAT
+                    return res
+            elif dtype is not None:
+                # i.e. PeriodDtype, DatetimeTZDtype
+                cls = dtype.construct_array_type()
+                obj = cls._from_sequence([], dtype=dtype)
+                taker = -np.ones((<object>objects).shape, dtype=np.intp)
+                return obj.take(taker, allow_fill=True)
+            else:
+                # we don't guess
+                seen.object_ = True
+        else:
+            seen.object_ = True
+
+    if not convert_numeric:
+        # Note: we count "bool" as numeric here. This is because
+        #  np.array(list_of_items) will convert bools just like it will numeric
+        #  entries.
+        return objects
+
+    if seen.bool_:
+        if convert_to_nullable_dtype and seen.is_bool_or_na:
+            from pandas.core.arrays import BooleanArray
+            return BooleanArray(bools.view(np.bool_), mask)
+        elif seen.is_bool:
+            # is_bool property rules out everything else
+            return bools.view(np.bool_)
+        seen.object_ = True
+
+    if not seen.object_:
+        result = None
+        if not safe:
+            if seen.null_ or seen.nan_:
+                if seen.complex_:
+                    result = complexes
+                elif seen.float_:
+                    result = floats
+                elif seen.int_ or seen.uint_:
+                    if convert_to_nullable_dtype:
+                        # Below we will wrap in IntegerArray
+                        if seen.uint_:
+                            result = uints
+                        else:
+                            result = ints
+                    else:
+                        result = floats
+                elif seen.nan_:
+                    result = floats
+            else:
+                if seen.complex_:
+                    result = complexes
+                elif seen.float_:
+                    result = floats
+                elif seen.int_:
+                    if seen.uint_:
+                        result = uints
+                    else:
+                        result = ints
+        else:
+            # don't cast int to float, etc.
+            if seen.null_:
+                if seen.complex_:
+                    if not seen.int_:
+                        result = complexes
+                elif seen.float_ or seen.nan_:
+                    if not seen.int_:
+                        result = floats
+            else:
+                if seen.complex_:
+                    if not seen.int_:
+                        result = complexes
+                elif seen.float_ or seen.nan_:
+                    if not seen.int_:
+                        result = floats
+                elif seen.int_:
+                    if seen.uint_:
+                        result = uints
+                    else:
+                        result = ints
+
+        # TODO: do these after the itemsize check?
+        if (result is ints or result is uints) and convert_to_nullable_dtype:
+            from pandas.core.arrays import IntegerArray
+
+            # Set these values to 1 to be deterministic, match
+            #  IntegerDtype._internal_fill_value
+            result[mask] = 1
+            result = IntegerArray(result, mask)
+        elif result is floats and convert_to_nullable_dtype:
+            from pandas.core.arrays import FloatingArray
+
+            # Set these values to 1.0 to be deterministic, match
+            #  FloatingDtype._internal_fill_value
+            result[mask] = 1.0
+            result = FloatingArray(result, mask)
+
+        if result is uints or result is ints or result is floats or result is complexes:
+            # cast to the largest itemsize when all values are NumPy scalars
+            if itemsize_max > 0 and itemsize_max != result.dtype.itemsize:
+                result = result.astype(result.dtype.kind + str(itemsize_max))
+            return result
+        elif result is not None:
+            return result
+
+    return objects
+
+
+class _NoDefault(Enum):
+    # We make this an Enum
+    # 1) because it round-trips through pickle correctly (see GH#40397)
+    # 2) because mypy does not understand singletons
+    no_default = "NO_DEFAULT"
+
+    def __repr__(self) -> str:
+        return "<no_default>"
+
+
+# Note: no_default is exported to the public API in pandas.api.extensions
+no_default = _NoDefault.no_default  # Sentinel indicating the default value.
+NoDefault = Literal[_NoDefault.no_default]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def map_infer_mask(
+    ndarray arr,
+    object f,
+    const uint8_t[:] mask,
+    *,
+    bint convert=True,
+    object na_value=no_default,
+    cnp.dtype dtype=np.dtype(object)
+) -> "ArrayLike":
+    """
+    Substitute for np.vectorize with pandas-friendly dtype inference.
+
+    Parameters
+    ----------
+    arr : ndarray
+    f : function
+    mask : ndarray
+        uint8 dtype ndarray indicating values not to apply `f` to.
+    convert : bool, default True
+        Whether to call `maybe_convert_objects` on the resulting ndarray.
+    na_value : Any, optional
+        The result value to use for masked values. By default, the
+        input value is used.
+    dtype : numpy.dtype
+        The numpy dtype to use for the result ndarray.
+
+    Returns
+    -------
+    np.ndarray or an ExtensionArray
+    """
+    cdef:
+        Py_ssize_t i
+        Py_ssize_t n = len(arr)
+        object val
+
+        ndarray result = np.empty(n, dtype=dtype)
+
+        flatiter arr_it = PyArray_IterNew(arr)
+        flatiter result_it = PyArray_IterNew(result)
+
+    for i in range(n):
+        if mask[i]:
+            if na_value is no_default:
+                val = PyArray_GETITEM(arr, PyArray_ITER_DATA(arr_it))
+            else:
+                val = na_value
+        else:
+            val = PyArray_GETITEM(arr, PyArray_ITER_DATA(arr_it))
+            val = f(val)
+
+            if cnp.PyArray_IsZeroDim(val):
+                # unbox 0-dim arrays, GH#690
+                val = val.item()
+
+        PyArray_SETITEM(result, PyArray_ITER_DATA(result_it), val)
+
+        PyArray_ITER_NEXT(arr_it)
+        PyArray_ITER_NEXT(result_it)
+
+    if convert:
+        return maybe_convert_objects(result)
+    else:
+        return result
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def map_infer(
+    ndarray arr, object f, *, bint convert=True, bint ignore_na=False
+) -> "ArrayLike":
+    """
+    Substitute for np.vectorize with pandas-friendly dtype inference.
+
+    Parameters
+    ----------
+    arr : ndarray
+    f : function
+    convert : bint
+    ignore_na : bint
+        If True, NA values will not have f applied
+
+    Returns
+    -------
+    np.ndarray or an ExtensionArray
+    """
+    cdef:
+        Py_ssize_t i, n
+        ndarray[object] result
+        object val
+
+    n = len(arr)
+    result = cnp.PyArray_EMPTY(1, arr.shape, cnp.NPY_OBJECT, 0)
+    for i in range(n):
+        if ignore_na and checknull(arr[i]):
+            result[i] = arr[i]
+            continue
+        val = f(arr[i])
+
+        if cnp.PyArray_IsZeroDim(val):
+            # unbox 0-dim arrays, GH#690
+            val = val.item()
+
+        result[i] = val
+
+    if convert:
+        return maybe_convert_objects(result)
+    else:
+        return result
+
+
+def to_object_array(rows: object, min_width: int = 0) -> ndarray:
+    """
+    Convert a list of lists into an object array.
+
+    Parameters
+    ----------
+    rows : 2-d array (N, K)
+        List of lists to be converted into an array.
+    min_width : int
+        Minimum width of the object array. If a list
+        in `rows` contains fewer than `width` elements,
+        the remaining elements in the corresponding row
+        will all be `NaN`.
+
+    Returns
+    -------
+    np.ndarray[object, ndim=2]
+    """
+    cdef:
+        Py_ssize_t i, j, n, k, tmp
+        ndarray[object, ndim=2] result
+        list row
+
+    rows = list(rows)
+    n = len(rows)
+
+    k = min_width
+    for i in range(n):
+        tmp = len(rows[i])
+        if tmp > k:
+            k = tmp
+
+    result = np.empty((n, k), dtype=object)
+
+    for i in range(n):
+        row = list(rows[i])
+
+        for j in range(len(row)):
+            result[i, j] = row[j]
+
+    return result
+
+
+def tuples_to_object_array(ndarray[object] tuples):
+    cdef:
+        Py_ssize_t i, j, n, k
+        ndarray[object, ndim=2] result
+        tuple tup
+
+    n = len(tuples)
+    k = len(tuples[0])
+    result = np.empty((n, k), dtype=object)
+    for i in range(n):
+        tup = tuples[i]
+        for j in range(k):
+            result[i, j] = tup[j]
+
+    return result
+
+
+def to_object_array_tuples(rows: object) -> np.ndarray:
+    """
+    Convert a list of tuples into an object array. Any subclass of
+    tuple in `rows` will be casted to tuple.
+
+    Parameters
+    ----------
+    rows : 2-d array (N, K)
+        List of tuples to be converted into an array.
+
+    Returns
+    -------
+    np.ndarray[object, ndim=2]
+    """
+    cdef:
+        Py_ssize_t i, j, n, k, tmp
+        ndarray[object, ndim=2] result
+        tuple row
+
+    rows = list(rows)
+    n = len(rows)
+
+    k = 0
+    for i in range(n):
+        tmp = 1 if checknull(rows[i]) else len(rows[i])
+        if tmp > k:
+            k = tmp
+
+    result = np.empty((n, k), dtype=object)
+
+    try:
+        for i in range(n):
+            row = rows[i]
+            for j in range(len(row)):
+                result[i, j] = row[j]
+    except TypeError:
+        # e.g. "Expected tuple, got list"
+        # upcast any subclasses to tuple
+        for i in range(n):
+            row = (rows[i],) if checknull(rows[i]) else tuple(rows[i])
+            for j in range(len(row)):
+                result[i, j] = row[j]
+
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def fast_multiget(dict mapping, object[:] keys, default=np.nan) -> "ArrayLike":
+    cdef:
+        Py_ssize_t i, n = len(keys)
+        object val
+        ndarray[object] output = np.empty(n, dtype="O")
+
+    if n == 0:
+        # kludge, for Series
+        return np.empty(0, dtype="f8")
+
+    for i in range(n):
+        val = keys[i]
+        if val in mapping:
+            output[i] = mapping[val]
+        else:
+            output[i] = default
+
+    return maybe_convert_objects(output)
+
+
+def is_bool_list(obj: list) -> bool:
+    """
+    Check if this list contains only bool or np.bool_ objects.
+
+    This is appreciably faster than checking `np.array(obj).dtype == bool`
+
+    obj1 = [True, False] * 100
+    obj2 = obj1 * 100
+    obj3 = obj2 * 100
+    obj4 = [True, None] + obj1
+
+    for obj in [obj1, obj2, obj3, obj4]:
+        %timeit is_bool_list(obj)
+        %timeit np.array(obj).dtype.kind == "b"
+
+    340 ns ± 8.22 ns
+    8.78 µs ± 253 ns
+
+    28.8 µs ± 704 ns
+    813 µs ± 17.8 µs
+
+    3.4 ms ± 168 µs
+    78.4 ms ± 1.05 ms
+
+    48.1 ns ± 1.26 ns
+    8.1 µs ± 198 ns
+    """
+    cdef:
+        object item
+
+    for item in obj:
+        if not util.is_bool_object(item):
+            return False
+
+    # Note: we return True for empty list
+    return True
+
+
+cpdef ndarray eq_NA_compat(ndarray[object] arr, object key):
+    """
+    Check for `arr == key`, treating all values as not-equal to pd.NA.
+
+    key is assumed to have `not isna(key)`
+    """
+    cdef:
+        ndarray[uint8_t, cast=True] result = cnp.PyArray_EMPTY(
+            arr.ndim, arr.shape, cnp.NPY_BOOL, 0
+        )
+        Py_ssize_t i
+        object item
+
+    for i in range(len(arr)):
+        item = arr[i]
+        if item is C_NA:
+            result[i] = False
+        else:
+            result[i] = item == key
+
+    return result
+
+
+def dtypes_all_equal(list types not None) -> bool:
+    """
+    Faster version for:
+
+    first = types[0]
+    all(is_dtype_equal(first, t) for t in types[1:])
+
+    And assuming all elements in the list are np.dtype/ExtensionDtype objects
+
+    See timings at https://github.com/pandas-dev/pandas/pull/44594
+    """
+    first = types[0]
+    for t in types[1:]:
+        if t is first:
+            # Fastpath can provide a nice boost for EADtypes
+            continue
+        try:
+            if not t == first:
+                return False
+        except (TypeError, AttributeError):
+            return False
+    else:
+        return True
+
+
+def is_np_dtype(object dtype, str kinds=None) -> bool:
+    """
+    Optimized check for `isinstance(dtype, np.dtype)` with
+    optional `and dtype.kind in kinds`.
+
+    dtype = np.dtype("m8[ns]")
+
+    In [7]: %timeit isinstance(dtype, np.dtype)
+    117 ns ± 1.91 ns per loop (mean ± std. dev. of 7 runs, 10,000,000 loops each)
+
+    In [8]: %timeit is_np_dtype(dtype)
+    64 ns ± 1.51 ns per loop (mean ± std. dev. of 7 runs, 10,000,000 loops each)
+
+    In [9]: %timeit is_timedelta64_dtype(dtype)
+    209 ns ± 6.96 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each)
+
+    In [10]: %timeit is_np_dtype(dtype, "m")
+    93.4 ns ± 1.11 ns per loop (mean ± std. dev. of 7 runs, 10,000,000 loops each)
+    """
+    if not cnp.PyArray_DescrCheck(dtype):
+        # i.e. not isinstance(dtype, np.dtype)
+        return False
+    if kinds is None:
+        return True
+    return dtype.kind in kinds
diff --git a/pandas/pandas/_libs/meson.build b/pandas/pandas/_libs/meson.build
new file mode 100644
index 0000000000000..a50976767928a
--- /dev/null
+++ b/pandas/pandas/_libs/meson.build
@@ -0,0 +1,196 @@
+_algos_take_helper = custom_target(
+    'algos_take_helper_pxi',
+    output: 'algos_take_helper.pxi',
+    input: 'algos_take_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_algos_common_helper = custom_target(
+    'algos_common_helper_pxi',
+    output: 'algos_common_helper.pxi',
+    input: 'algos_common_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_khash_primitive_helper = custom_target(
+    'khash_primitive_helper_pxi',
+    output: 'khash_for_primitive_helper.pxi',
+    input: 'khash_for_primitive_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_hashtable_class_helper = custom_target(
+    'hashtable_class_helper_pxi',
+    output: 'hashtable_class_helper.pxi',
+    input: 'hashtable_class_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_hashtable_func_helper = custom_target(
+    'hashtable_func_helper_pxi',
+    output: 'hashtable_func_helper.pxi',
+    input: 'hashtable_func_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_index_class_helper = custom_target(
+    'index_class_helper_pxi',
+    output: 'index_class_helper.pxi',
+    input: 'index_class_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_sparse_op_helper = custom_target(
+    'sparse_op_helper_pxi',
+    output: 'sparse_op_helper.pxi',
+    input: 'sparse_op_helper.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_intervaltree_helper = custom_target(
+    'intervaltree_helper_pxi',
+    output: 'intervaltree.pxi',
+    input: 'intervaltree.pxi.in',
+    command: [py, tempita, '@INPUT@', '-o', '@OUTDIR@'],
+)
+_khash_primitive_helper_dep = declare_dependency(
+    sources: _khash_primitive_helper,
+)
+
+cdata = configuration_data()
+if cy.version().version_compare('>=3.1.0')
+    cdata.set('freethreading_compatible', '1')
+else
+    cdata.set('freethreading_compatible', '0')
+endif
+_free_threading_config = configure_file(
+    input: 'free_threading_config.pxi.in',
+    output: 'free_threading_config.pxi',
+    configuration: cdata,
+    install: false,
+)
+
+subdir('tslibs')
+
+libs_sources = {
+    # Dict of extension name -> dict of {sources, include_dirs, and deps}
+    # numpy include dir is implicitly included
+    'algos': {
+        'sources': ['algos.pyx', _algos_common_helper, _algos_take_helper],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'arrays': {'sources': ['arrays.pyx']},
+    'groupby': {'sources': ['groupby.pyx']},
+    'hashing': {'sources': ['hashing.pyx']},
+    'hashtable': {
+        'sources': [
+            'hashtable.pyx',
+            _hashtable_class_helper,
+            _hashtable_func_helper,
+        ],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'index': {
+        'sources': ['index.pyx', _index_class_helper],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'indexing': {'sources': ['indexing.pyx']},
+    'internals': {'sources': ['internals.pyx']},
+    'interval': {
+        'sources': ['interval.pyx', _intervaltree_helper],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'join': {
+        'sources': ['join.pyx', _khash_primitive_helper],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'lib': {'sources': ['lib.pyx', 'src/parser/tokenizer.c']},
+    'missing': {'sources': ['missing.pyx']},
+    'pandas_datetime': {
+        'sources': [
+            'src/vendored/numpy/datetime/np_datetime.c',
+            'src/vendored/numpy/datetime/np_datetime_strings.c',
+            'src/datetime/date_conversions.c',
+            'src/datetime/pd_datetime.c',
+        ],
+    },
+    'pandas_parser': {
+        'sources': [
+            'src/parser/tokenizer.c',
+            'src/parser/io.c',
+            'src/parser/pd_parser.c',
+        ],
+    },
+    'parsers': {
+        'sources': ['parsers.pyx', 'src/parser/tokenizer.c', 'src/parser/io.c'],
+        'deps': _khash_primitive_helper_dep,
+    },
+    'json': {
+        'sources': [
+            'src/vendored/ujson/python/ujson.c',
+            'src/vendored/ujson/python/objToJSON.c',
+            'src/vendored/ujson/python/JSONtoObj.c',
+            'src/vendored/ujson/lib/ultrajsonenc.c',
+            'src/vendored/ujson/lib/ultrajsondec.c',
+        ],
+    },
+    'ops': {'sources': ['ops.pyx']},
+    'ops_dispatch': {'sources': ['ops_dispatch.pyx']},
+    'properties': {'sources': ['properties.pyx']},
+    'reshape': {'sources': ['reshape.pyx']},
+    'sas': {'sources': ['sas.pyx']},
+    'byteswap': {'sources': ['byteswap.pyx']},
+    'sparse': {'sources': ['sparse.pyx', _sparse_op_helper]},
+    'tslib': {'sources': ['tslib.pyx']},
+    'testing': {'sources': ['testing.pyx']},
+    'writers': {'sources': ['writers.pyx']},
+}
+
+cython_args = [
+    '--include-dir',
+    meson.current_build_dir(),
+    '-X always_allow_keywords=true',
+]
+if get_option('buildtype') == 'debug'
+    cython_args += ['--gdb']
+endif
+
+foreach ext_name, ext_dict : libs_sources
+    py.extension_module(
+        ext_name,
+        ext_dict.get('sources'),
+        cython_args: cython_args,
+        include_directories: [inc_np, inc_pd],
+        dependencies: ext_dict.get('deps', ''),
+        subdir: 'pandas/_libs',
+        install: true,
+    )
+endforeach
+
+# Basically just __init__.py and the .pyi files
+sources_to_install = [
+    '__init__.py',
+    'algos.pyi',
+    'arrays.pyi',
+    'byteswap.pyi',
+    'groupby.pyi',
+    'hashing.pyi',
+    'hashtable.pyi',
+    'index.pyi',
+    'indexing.pyi',
+    'internals.pyi',
+    'interval.pyi',
+    'join.pyi',
+    'json.pyi',
+    'lib.pyi',
+    'missing.pyi',
+    'ops.pyi',
+    'ops_dispatch.pyi',
+    'parsers.pyi',
+    'properties.pyi',
+    'reshape.pyi',
+    'sas.pyi',
+    'sparse.pyi',
+    'testing.pyi',
+    'tslib.pyi',
+    'writers.pyi',
+]
+
+foreach source : sources_to_install
+    py.install_sources(source, subdir: 'pandas/_libs')
+endforeach
+
+subdir('window')
diff --git a/pandas/pandas/_libs/missing.pxd b/pandas/pandas/_libs/missing.pxd
new file mode 100644
index 0000000000000..899d729690451
--- /dev/null
+++ b/pandas/pandas/_libs/missing.pxd
@@ -0,0 +1,20 @@
+from numpy cimport (
+    ndarray,
+    uint8_t,
+)
+
+
+cpdef bint is_matching_na(object left, object right, bint nan_matches_none=*)
+cpdef bint check_na_tuples_nonequal(object left, object right)
+
+cpdef bint checknull(object val)
+cpdef ndarray[uint8_t] isnaobj(ndarray arr)
+
+cdef bint is_null_datetime64(v)
+cdef bint is_null_timedelta64(v)
+cdef bint checknull_with_nat_and_na(object obj)
+
+cdef class C_NAType:
+    pass
+
+cdef C_NAType C_NA
diff --git a/pandas/pandas/_libs/missing.pyi b/pandas/pandas/_libs/missing.pyi
new file mode 100644
index 0000000000000..6bf30a03cef32
--- /dev/null
+++ b/pandas/pandas/_libs/missing.pyi
@@ -0,0 +1,16 @@
+import numpy as np
+from numpy import typing as npt
+
+class NAType:
+    def __new__(cls, *args, **kwargs): ...
+
+NA: NAType
+
+def is_matching_na(
+    left: object, right: object, nan_matches_none: bool = ...
+) -> bool: ...
+def isposinf_scalar(val: object) -> bool: ...
+def isneginf_scalar(val: object) -> bool: ...
+def checknull(val: object) -> bool: ...
+def isnaobj(arr: np.ndarray) -> npt.NDArray[np.bool_]: ...
+def is_numeric_na(values: np.ndarray) -> npt.NDArray[np.bool_]: ...
diff --git a/pandas/pandas/_libs/missing.pyx b/pandas/pandas/_libs/missing.pyx
new file mode 100644
index 0000000000000..390a527c22bbb
--- /dev/null
+++ b/pandas/pandas/_libs/missing.pyx
@@ -0,0 +1,521 @@
+from decimal import Decimal
+import numbers
+from sys import maxsize
+
+cimport cython
+from cpython.datetime cimport (
+    date,
+    time,
+    timedelta,
+)
+from cython cimport Py_ssize_t
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    flatiter,
+    float64_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+)
+
+cnp.import_array()
+
+from pandas._libs cimport util
+from pandas._libs.tslibs.nattype cimport (
+    c_NaT as NaT,
+    checknull_with_nat,
+    is_dt64nat,
+    is_td64nat,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    get_datetime64_unit,
+    import_pandas_datetime,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.ops_dispatch import maybe_dispatch_ufunc_to_dunder_op
+
+cdef:
+    float64_t INF = <float64_t>np.inf
+    float64_t NEGINF = -INF
+
+    int64_t NPY_NAT = util.get_nat()
+
+    bint is_32bit = maxsize <= 2 ** 32
+
+    type cDecimal = Decimal  # for faster isinstance checks
+
+
+cpdef bint check_na_tuples_nonequal(object left, object right):
+    """
+    When we have NA in one of the tuples but not the other we have to check here,
+    because our regular checks fail before with ambiguous boolean value.
+
+    Parameters
+    ----------
+    left: Any
+    right: Any
+
+    Returns
+    -------
+    True if we are dealing with tuples that have NA on one side and non NA on
+    the other side.
+
+    """
+    if not isinstance(left, tuple) or not isinstance(right, tuple):
+        return False
+
+    if len(left) != len(right):
+        return False
+
+    for left_element, right_element in zip(left, right):
+        if left_element is C_NA and right_element is not C_NA:
+            return True
+        elif right_element is C_NA and left_element is not C_NA:
+            return True
+
+    return False
+
+
+cpdef bint is_matching_na(object left, object right, bint nan_matches_none=False):
+    """
+    Check if two scalars are both NA of matching types.
+
+    Parameters
+    ----------
+    left : Any
+    right : Any
+    nan_matches_none : bool, default False
+        For backwards compatibility, consider NaN as matching None.
+
+    Returns
+    -------
+    bool
+    """
+    if left is None:
+        if nan_matches_none and util.is_nan(right):
+            return True
+        return right is None
+    elif left is C_NA:
+        return right is C_NA
+    elif left is NaT:
+        return right is NaT
+    elif util.is_float_object(left):
+        if nan_matches_none and right is None and util.is_nan(left):
+            return True
+        return (
+            util.is_nan(left)
+            and util.is_float_object(right)
+            and util.is_nan(right)
+        )
+    elif util.is_complex_object(left):
+        return (
+            util.is_nan(left)
+            and util.is_complex_object(right)
+            and util.is_nan(right)
+        )
+    elif cnp.is_datetime64_object(left):
+        return (
+            cnp.get_datetime64_value(left) == NPY_NAT
+            and cnp.is_datetime64_object(right)
+            and cnp.get_datetime64_value(right) == NPY_NAT
+            and get_datetime64_unit(left) == get_datetime64_unit(right)
+        )
+    elif cnp.is_timedelta64_object(left):
+        return (
+            cnp.get_timedelta64_value(left) == NPY_NAT
+            and cnp.is_timedelta64_object(right)
+            and cnp.get_timedelta64_value(right) == NPY_NAT
+            and get_datetime64_unit(left) == get_datetime64_unit(right)
+        )
+    elif is_decimal_na(left):
+        return is_decimal_na(right)
+    return False
+
+
+cpdef bint checknull(object val):
+    """
+    Return boolean describing of the input is NA-like, defined here as any
+    of:
+     - None
+     - nan
+     - NaT
+     - np.datetime64 representation of NaT
+     - np.timedelta64 representation of NaT
+     - NA
+     - Decimal("NaN")
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    bool
+    """
+    if val is None or val is NaT or val is C_NA:
+        return True
+    elif util.is_float_object(val) or util.is_complex_object(val):
+        if val != val:
+            return True
+        return False
+    elif cnp.is_timedelta64_object(val):
+        return cnp.get_timedelta64_value(val) == NPY_NAT
+    elif cnp.is_datetime64_object(val):
+        return cnp.get_datetime64_value(val) == NPY_NAT
+    else:
+        return is_decimal_na(val)
+
+
+cdef bint is_decimal_na(object val):
+    """
+    Is this a decimal.Decimal object Decimal("NAN").
+    """
+    return isinstance(val, cDecimal) and val != val
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef ndarray[uint8_t] isnaobj(ndarray arr):
+    """
+    Return boolean mask denoting which elements of a 1-D array are na-like,
+    according to the criteria defined in `checknull`:
+     - None
+     - nan
+     - NaT
+     - np.datetime64 representation of NaT
+     - np.timedelta64 representation of NaT
+     - NA
+     - Decimal("NaN")
+
+    Parameters
+    ----------
+    arr : ndarray
+
+    Returns
+    -------
+    result : ndarray (dtype=np.bool_)
+    """
+    cdef:
+        Py_ssize_t i, n = arr.size
+        object val
+        bint is_null
+        ndarray result = np.empty((<object>arr).shape, dtype=np.uint8)
+        flatiter it = cnp.PyArray_IterNew(arr)
+        flatiter it2 = cnp.PyArray_IterNew(result)
+
+    for i in range(n):
+        # The PyArray_GETITEM and PyArray_ITER_NEXT are faster
+        #  equivalents to `val = values[i]`
+        val = cnp.PyArray_GETITEM(arr, cnp.PyArray_ITER_DATA(it))
+        cnp.PyArray_ITER_NEXT(it)
+        is_null = checknull(val)
+        # Dereference pointer (set value)
+        (<uint8_t *>(cnp.PyArray_ITER_DATA(it2)))[0] = <uint8_t>is_null
+        cnp.PyArray_ITER_NEXT(it2)
+    return result.view(np.bool_)
+
+
+def isposinf_scalar(val: object) -> bool:
+    return util.is_float_object(val) and val == INF
+
+
+def isneginf_scalar(val: object) -> bool:
+    return util.is_float_object(val) and val == NEGINF
+
+
+cdef bint is_null_datetime64(v):
+    # determine if we have a null for a datetime (or integer versions),
+    # excluding np.timedelta64('nat')
+    if checknull_with_nat(v) or is_dt64nat(v):
+        return True
+    return False
+
+
+cdef bint is_null_timedelta64(v):
+    # determine if we have a null for a timedelta (or integer versions),
+    # excluding np.datetime64('nat')
+    if checknull_with_nat(v) or is_td64nat(v):
+        return True
+    return False
+
+
+cdef bint checknull_with_nat_and_na(object obj):
+    # See GH#32214
+    return checknull_with_nat(obj) or obj is C_NA
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_numeric_na(values: ndarray) -> ndarray:
+    """
+    Check for NA values consistent with IntegerArray/FloatingArray.
+
+    Similar to a vectorized is_valid_na_for_dtype restricted to numeric dtypes.
+
+    Returns
+    -------
+    ndarray[bool]
+    """
+    cdef:
+        ndarray[uint8_t] result
+        Py_ssize_t i, N
+        object val
+
+    N = len(values)
+    result = np.zeros(N, dtype=np.uint8)
+
+    for i in range(N):
+        val = values[i]
+        if checknull(val):
+            if val is None or val is C_NA or util.is_nan(val) or is_decimal_na(val):
+                result[i] = True
+            else:
+                raise TypeError(f"'values' contains non-numeric NA {val}")
+    return result.view(bool)
+
+
+# -----------------------------------------------------------------------------
+# Implementation of NA singleton
+
+
+def _create_binary_propagating_op(name, is_divmod=False):
+    is_cmp = name.strip("_") in ["eq", "ne", "le", "lt", "ge", "gt"]
+
+    def method(self, other):
+        if (other is C_NA or isinstance(other, (str, bytes))
+                or isinstance(other, (numbers.Number, np.bool_))
+                or util.is_array(other) and not other.shape):
+            # Need the other.shape clause to handle NumPy scalars,
+            # since we do a setitem on `out` below, which
+            # won't work for NumPy scalars.
+            if is_divmod:
+                return NA, NA
+            else:
+                return NA
+
+        elif util.is_array(other):
+            out = np.empty(other.shape, dtype=object)
+            out[:] = NA
+
+            if is_divmod:
+                return out, out.copy()
+            else:
+                return out
+
+        elif is_cmp and isinstance(other, (date, time, timedelta)):
+            return NA
+
+        elif isinstance(other, date):
+            if name in ["__sub__", "__rsub__"]:
+                return NA
+
+        elif isinstance(other, timedelta):
+            if name in ["__sub__", "__rsub__", "__add__", "__radd__"]:
+                return NA
+
+        return NotImplemented
+
+    method.__name__ = name
+    return method
+
+
+def _create_unary_propagating_op(name: str):
+    def method(self):
+        return NA
+
+    method.__name__ = name
+    return method
+
+
+cdef class C_NAType:
+    pass
+
+
+class NAType(C_NAType):
+    """
+    NA ("not available") missing value indicator.
+
+    .. warning::
+
+       Experimental: the behaviour of NA can still change without warning.
+
+    The NA singleton is a missing value indicator defined by pandas. It is
+    used in certain new extension dtypes (currently the "string" dtype).
+
+    See Also
+    --------
+    numpy.nan : Floating point representation of Not a Number (NaN) for numerical data.
+    isna : Detect missing values for an array-like object.
+    notna : Detect non-missing values for an array-like object.
+    DataFrame.fillna : Fill missing values in a DataFrame.
+    Series.fillna : Fill missing values in a Series.
+
+    Examples
+    --------
+    >>> pd.NA
+    <NA>
+
+    >>> True | pd.NA
+    True
+
+    >>> True & pd.NA
+    <NA>
+
+    >>> pd.NA != pd.NA
+    <NA>
+
+    >>> pd.NA == pd.NA
+    <NA>
+
+    >>> True | pd.NA
+    True
+    """
+
+    _instance = None
+
+    def __new__(cls, *args, **kwargs):
+        if NAType._instance is None:
+            NAType._instance = C_NAType.__new__(cls, *args, **kwargs)
+        return NAType._instance
+
+    def __repr__(self) -> str:
+        return "<NA>"
+
+    def __format__(self, format_spec) -> str:
+        try:
+            return self.__repr__().__format__(format_spec)
+        except ValueError:
+            return self.__repr__()
+
+    def __bool__(self):
+        raise TypeError("boolean value of NA is ambiguous")
+
+    def __hash__(self):
+        # GH 30013: Ensure hash is large enough to avoid hash collisions with integers
+        exponent = 31 if is_32bit else 61
+        return 2 ** exponent - 1
+
+    def __reduce__(self):
+        return "NA"
+
+    # Binary arithmetic and comparison ops -> propagate
+
+    __add__ = _create_binary_propagating_op("__add__")
+    __radd__ = _create_binary_propagating_op("__radd__")
+    __sub__ = _create_binary_propagating_op("__sub__")
+    __rsub__ = _create_binary_propagating_op("__rsub__")
+    __mul__ = _create_binary_propagating_op("__mul__")
+    __rmul__ = _create_binary_propagating_op("__rmul__")
+    __matmul__ = _create_binary_propagating_op("__matmul__")
+    __rmatmul__ = _create_binary_propagating_op("__rmatmul__")
+    __truediv__ = _create_binary_propagating_op("__truediv__")
+    __rtruediv__ = _create_binary_propagating_op("__rtruediv__")
+    __floordiv__ = _create_binary_propagating_op("__floordiv__")
+    __rfloordiv__ = _create_binary_propagating_op("__rfloordiv__")
+    __mod__ = _create_binary_propagating_op("__mod__")
+    __rmod__ = _create_binary_propagating_op("__rmod__")
+    __divmod__ = _create_binary_propagating_op("__divmod__", is_divmod=True)
+    __rdivmod__ = _create_binary_propagating_op("__rdivmod__", is_divmod=True)
+    # __lshift__ and __rshift__ are not implemented
+
+    __eq__ = _create_binary_propagating_op("__eq__")
+    __ne__ = _create_binary_propagating_op("__ne__")
+    __le__ = _create_binary_propagating_op("__le__")
+    __lt__ = _create_binary_propagating_op("__lt__")
+    __gt__ = _create_binary_propagating_op("__gt__")
+    __ge__ = _create_binary_propagating_op("__ge__")
+
+    # Unary ops
+
+    __neg__ = _create_unary_propagating_op("__neg__")
+    __pos__ = _create_unary_propagating_op("__pos__")
+    __abs__ = _create_unary_propagating_op("__abs__")
+    __invert__ = _create_unary_propagating_op("__invert__")
+
+    # pow has special
+    def __pow__(self, other):
+        if other is C_NA:
+            return NA
+        elif isinstance(other, (numbers.Number, np.bool_)):
+            if other == 0:
+                # returning positive is correct for +/- 0.
+                return type(other)(1)
+            else:
+                return NA
+        elif util.is_array(other):
+            return np.where(other == 0, other.dtype.type(1), NA)
+
+        return NotImplemented
+
+    def __rpow__(self, other):
+        if other is C_NA:
+            return NA
+        elif isinstance(other, (numbers.Number, np.bool_)):
+            if other == 1:
+                return other
+            else:
+                return NA
+        elif util.is_array(other):
+            return np.where(other == 1, other, NA)
+        return NotImplemented
+
+    # Logical ops using Kleene logic
+
+    def __and__(self, other):
+        if other is False:
+            return False
+        elif other is True or other is C_NA:
+            return NA
+        return NotImplemented
+
+    __rand__ = __and__
+
+    def __or__(self, other):
+        if other is True:
+            return True
+        elif other is False or other is C_NA:
+            return NA
+        return NotImplemented
+
+    __ror__ = __or__
+
+    def __xor__(self, other):
+        if other is False or other is True or other is C_NA:
+            return NA
+        return NotImplemented
+
+    __rxor__ = __xor__
+
+    __array_priority__ = 1000
+    _HANDLED_TYPES = (np.ndarray, numbers.Number, str, np.bool_)
+
+    def __array_ufunc__(self, ufunc, method, *inputs, **kwargs):
+        types = self._HANDLED_TYPES + (NAType,)
+        for x in inputs:
+            if not isinstance(x, types):
+                return NotImplemented
+
+        if method != "__call__":
+            raise ValueError(f"ufunc method '{method}' not supported for NA")
+        result = maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is NotImplemented:
+            # For a NumPy ufunc that's not a binop, like np.logaddexp
+            index = [i for i, x in enumerate(inputs) if x is NA][0]
+            result = np.broadcast_arrays(*inputs)[index]
+            if result.ndim == 0:
+                result = result.item()
+            if ufunc.nout > 1:
+                result = (NA,) * ufunc.nout
+
+        return result
+
+
+C_NA = NAType()   # C-visible
+NA = C_NA         # Python-visible
diff --git a/pandas/pandas/_libs/ops.pyi b/pandas/pandas/_libs/ops.pyi
new file mode 100644
index 0000000000000..6738a1dff4a9e
--- /dev/null
+++ b/pandas/pandas/_libs/ops.pyi
@@ -0,0 +1,51 @@
+from typing import (
+    Any,
+    Callable,
+    Iterable,
+    Literal,
+    TypeAlias,
+    overload,
+)
+
+import numpy as np
+
+from pandas._typing import npt
+
+_BinOp: TypeAlias = Callable[[Any, Any], Any]
+_BoolOp: TypeAlias = Callable[[Any, Any], bool]
+
+def scalar_compare(
+    values: np.ndarray,  # object[:]
+    val: object,
+    op: _BoolOp,  # {operator.eq, operator.ne, ...}
+) -> npt.NDArray[np.bool_]: ...
+def vec_compare(
+    left: npt.NDArray[np.object_],
+    right: npt.NDArray[np.object_],
+    op: _BoolOp,  # {operator.eq, operator.ne, ...}
+) -> npt.NDArray[np.bool_]: ...
+def scalar_binop(
+    values: np.ndarray,  # object[:]
+    val: object,
+    op: _BinOp,  # binary operator
+) -> np.ndarray: ...
+def vec_binop(
+    left: np.ndarray,  # object[:]
+    right: np.ndarray,  # object[:]
+    op: _BinOp,  # binary operator
+) -> np.ndarray: ...
+@overload
+def maybe_convert_bool(
+    arr: npt.NDArray[np.object_],
+    true_values: Iterable | None = None,
+    false_values: Iterable | None = None,
+    convert_to_masked_nullable: Literal[False] = ...,
+) -> tuple[np.ndarray, None]: ...
+@overload
+def maybe_convert_bool(
+    arr: npt.NDArray[np.object_],
+    true_values: Iterable = ...,
+    false_values: Iterable = ...,
+    *,
+    convert_to_masked_nullable: Literal[True],
+) -> tuple[np.ndarray, np.ndarray]: ...
diff --git a/pandas/pandas/_libs/ops.pyx b/pandas/pandas/_libs/ops.pyx
new file mode 100644
index 0000000000000..567bfc02a2950
--- /dev/null
+++ b/pandas/pandas/_libs/ops.pyx
@@ -0,0 +1,310 @@
+import operator
+
+cimport cython
+from cpython.object cimport (
+    Py_EQ,
+    Py_GE,
+    Py_GT,
+    Py_LE,
+    Py_LT,
+    Py_NE,
+    PyObject_RichCompareBool,
+)
+from cython cimport Py_ssize_t
+
+import numpy as np
+
+from numpy cimport (
+    import_array,
+    ndarray,
+    uint8_t,
+)
+
+import_array()
+
+
+from pandas._libs.missing cimport checknull
+from pandas._libs.util cimport is_nan
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def scalar_compare(ndarray[object] values, object val, object op) -> ndarray:
+    """
+    Compare each element of `values` array with the scalar `val`, with
+    the comparison operation described by `op`.
+
+    Parameters
+    ----------
+    values : ndarray[object]
+    val : object
+    op : {operator.eq, operator.ne,
+          operator.le, operator.lt,
+          operator.ge, operator.gt}
+
+    Returns
+    -------
+    result : ndarray[bool]
+    """
+    cdef:
+        Py_ssize_t i, n = len(values)
+        ndarray[uint8_t, cast=True] result
+        bint isnull_val
+        int flag
+        object x
+
+    if op is operator.lt:
+        flag = Py_LT
+    elif op is operator.le:
+        flag = Py_LE
+    elif op is operator.gt:
+        flag = Py_GT
+    elif op is operator.ge:
+        flag = Py_GE
+    elif op is operator.eq:
+        flag = Py_EQ
+    elif op is operator.ne:
+        flag = Py_NE
+    else:
+        raise ValueError("Unrecognized operator")
+
+    result = np.empty(n, dtype=bool).view(np.uint8)
+    isnull_val = checknull(val)
+
+    if flag == Py_NE:
+        for i in range(n):
+            x = values[i]
+            if checknull(x):
+                result[i] = True
+            elif isnull_val:
+                result[i] = True
+            else:
+                try:
+                    result[i] = PyObject_RichCompareBool(x, val, flag)
+                except TypeError:
+                    result[i] = True
+    elif flag == Py_EQ:
+        for i in range(n):
+            x = values[i]
+            if checknull(x):
+                result[i] = False
+            elif isnull_val:
+                result[i] = False
+            else:
+                try:
+                    result[i] = PyObject_RichCompareBool(x, val, flag)
+                except TypeError:
+                    result[i] = False
+
+    else:
+        for i in range(n):
+            x = values[i]
+            if checknull(x):
+                result[i] = False
+            elif isnull_val:
+                result[i] = False
+            else:
+                result[i] = PyObject_RichCompareBool(x, val, flag)
+
+    return result.view(bool)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def vec_compare(ndarray[object] left, ndarray[object] right, object op) -> ndarray:
+    """
+    Compare the elements of `left` with the elements of `right` pointwise,
+    with the comparison operation described by `op`.
+
+    Parameters
+    ----------
+    left : ndarray[object]
+    right : ndarray[object]
+    op : {operator.eq, operator.ne,
+          operator.le, operator.lt,
+          operator.ge, operator.gt}
+
+    Returns
+    -------
+    result : ndarray[bool]
+    """
+    cdef:
+        Py_ssize_t i, n = len(left)
+        ndarray[uint8_t, cast=True] result
+        int flag
+
+    if n != <Py_ssize_t>len(right):
+        raise ValueError(f"Arrays were different lengths: {n} vs {len(right)}")
+
+    if op is operator.lt:
+        flag = Py_LT
+    elif op is operator.le:
+        flag = Py_LE
+    elif op is operator.gt:
+        flag = Py_GT
+    elif op is operator.ge:
+        flag = Py_GE
+    elif op is operator.eq:
+        flag = Py_EQ
+    elif op is operator.ne:
+        flag = Py_NE
+    else:
+        raise ValueError("Unrecognized operator")
+
+    result = np.empty(n, dtype=bool).view(np.uint8)
+
+    if flag == Py_NE:
+        for i in range(n):
+            x = left[i]
+            y = right[i]
+
+            if checknull(x) or checknull(y):
+                result[i] = True
+            else:
+                result[i] = PyObject_RichCompareBool(x, y, flag)
+    else:
+        for i in range(n):
+            x = left[i]
+            y = right[i]
+
+            if checknull(x) or checknull(y):
+                result[i] = False
+            else:
+                result[i] = PyObject_RichCompareBool(x, y, flag)
+
+    return result.view(bool)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def scalar_binop(object[:] values, object val, object op) -> ndarray:
+    """
+    Apply the given binary operator `op` between each element of the array
+    `values` and the scalar `val`.
+
+    Parameters
+    ----------
+    values : ndarray[object]
+    val : object
+    op : binary operator
+
+    Returns
+    -------
+    result : ndarray[object]
+    """
+    cdef:
+        Py_ssize_t i, n = len(values)
+        object[::1] result
+        object x
+
+    result = np.empty(n, dtype=object)
+    if val is None or is_nan(val):
+        result[:] = val
+        return result.base  # `.base` to access underlying np.ndarray
+
+    for i in range(n):
+        x = values[i]
+        if x is None or is_nan(x):
+            result[i] = x
+        else:
+            result[i] = op(x, val)
+
+    return maybe_convert_bool(result.base)[0]
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def vec_binop(object[:] left, object[:] right, object op) -> ndarray:
+    """
+    Apply the given binary operator `op` pointwise to the elements of
+    arrays `left` and `right`.
+
+    Parameters
+    ----------
+    left : ndarray[object]
+    right : ndarray[object]
+    op : binary operator
+
+    Returns
+    -------
+    result : ndarray[object]
+    """
+    cdef:
+        Py_ssize_t i, n = len(left)
+        object[::1] result
+
+    if n != <Py_ssize_t>len(right):
+        raise ValueError(f"Arrays were different lengths: {n} vs {len(right)}")
+
+    result = np.empty(n, dtype=object)
+
+    for i in range(n):
+        x = left[i]
+        y = right[i]
+        try:
+            result[i] = op(x, y)
+        except TypeError:
+            if x is None or is_nan(x):
+                result[i] = x
+            elif y is None or is_nan(y):
+                result[i] = y
+            else:
+                raise
+
+    return maybe_convert_bool(result.base)[0]  # `.base` to access np.ndarray
+
+
+def maybe_convert_bool(ndarray[object] arr,
+                       true_values=None,
+                       false_values=None,
+                       convert_to_masked_nullable=False
+                       ) -> tuple[np.ndarray, np.ndarray | None]:
+    cdef:
+        Py_ssize_t i, n
+        ndarray[uint8_t] result
+        ndarray[uint8_t] mask
+        object val
+        set true_vals, false_vals
+        bint has_na = False
+
+    n = len(arr)
+    result = np.empty(n, dtype=np.uint8)
+    mask = np.zeros(n, dtype=np.uint8)
+    # the defaults
+    true_vals = {"True", "TRUE", "true"}
+    false_vals = {"False", "FALSE", "false"}
+
+    if true_values is not None:
+        true_vals = true_vals | set(true_values)
+
+    if false_values is not None:
+        false_vals = false_vals | set(false_values)
+
+    for i in range(n):
+        val = arr[i]
+
+        if isinstance(val, bool):
+            if val is True:
+                result[i] = 1
+            else:
+                result[i] = 0
+        elif val in true_vals:
+            result[i] = 1
+        elif val in false_vals:
+            result[i] = 0
+        elif is_nan(val) or val is None:
+            mask[i] = 1
+            result[i] = 0  # Value here doesn't matter, will be replaced w/ nan
+            has_na = True
+        else:
+            return (arr, None)
+
+    if has_na:
+        if convert_to_masked_nullable:
+            return (result.view(np.bool_), mask.view(np.bool_))
+        else:
+            arr = result.view(np.bool_).astype(object)
+            np.putmask(arr, mask, np.nan)
+            return (arr, None)
+    else:
+        return (result.view(np.bool_), None)
diff --git a/pandas/pandas/_libs/ops_dispatch.pyi b/pandas/pandas/_libs/ops_dispatch.pyi
new file mode 100644
index 0000000000000..91b5a4dbaaebc
--- /dev/null
+++ b/pandas/pandas/_libs/ops_dispatch.pyi
@@ -0,0 +1,5 @@
+import numpy as np
+
+def maybe_dispatch_ufunc_to_dunder_op(
+    self, ufunc: np.ufunc, method: str, *inputs, **kwargs
+): ...
diff --git a/pandas/pandas/_libs/ops_dispatch.pyx b/pandas/pandas/_libs/ops_dispatch.pyx
new file mode 100644
index 0000000000000..2b2a411e6635f
--- /dev/null
+++ b/pandas/pandas/_libs/ops_dispatch.pyx
@@ -0,0 +1,121 @@
+DISPATCHED_UFUNCS = {
+    "add",
+    "sub",
+    "mul",
+    "pow",
+    "mod",
+    "floordiv",
+    "truediv",
+    "divmod",
+    "eq",
+    "ne",
+    "lt",
+    "gt",
+    "le",
+    "ge",
+    "remainder",
+    "matmul",
+    "or",
+    "xor",
+    "and",
+    "neg",
+    "pos",
+    "abs",
+}
+UNARY_UFUNCS = {
+    "neg",
+    "pos",
+    "abs",
+}
+UFUNC_ALIASES = {
+    "subtract": "sub",
+    "multiply": "mul",
+    "floor_divide": "floordiv",
+    "true_divide": "truediv",
+    "power": "pow",
+    "remainder": "mod",
+    "divide": "truediv",
+    "equal": "eq",
+    "not_equal": "ne",
+    "less": "lt",
+    "less_equal": "le",
+    "greater": "gt",
+    "greater_equal": "ge",
+    "bitwise_or": "or",
+    "bitwise_and": "and",
+    "bitwise_xor": "xor",
+    "negative": "neg",
+    "absolute": "abs",
+    "positive": "pos",
+}
+
+# For op(., Array) -> Array.__r{op}__
+REVERSED_NAMES = {
+    "lt": "__gt__",
+    "le": "__ge__",
+    "gt": "__lt__",
+    "ge": "__le__",
+    "eq": "__eq__",
+    "ne": "__ne__",
+}
+
+
+def maybe_dispatch_ufunc_to_dunder_op(
+    object self, object ufunc, str method, *inputs, **kwargs
+):
+    """
+    Dispatch a ufunc to the equivalent dunder method.
+
+    Parameters
+    ----------
+    self : ArrayLike
+        The array whose dunder method we dispatch to
+    ufunc : Callable
+        A NumPy ufunc
+    method : {'reduce', 'accumulate', 'reduceat', 'outer', 'at', '__call__'}
+    inputs : ArrayLike
+        The input arrays.
+    kwargs : Any
+        The additional keyword arguments, e.g. ``out``.
+
+    Returns
+    -------
+    result : Any
+        The result of applying the ufunc
+    """
+    # special has the ufuncs we dispatch to the dunder op on
+
+    op_name = ufunc.__name__
+    op_name = UFUNC_ALIASES.get(op_name, op_name)
+
+    def not_implemented(*args, **kwargs):
+        return NotImplemented
+
+    if kwargs or ufunc.nin > 2:
+        return NotImplemented
+
+    if method == "__call__" and op_name in DISPATCHED_UFUNCS:
+
+        if inputs[0] is self:
+            name = f"__{op_name}__"
+            meth = getattr(self, name, not_implemented)
+
+            if op_name in UNARY_UFUNCS:
+                assert len(inputs) == 1
+                return meth()
+
+            return meth(inputs[1])
+
+        elif inputs[1] is self:
+            name = REVERSED_NAMES.get(op_name, f"__r{op_name}__")
+
+            meth = getattr(self, name, not_implemented)
+            result = meth(inputs[0])
+            return result
+
+        else:
+            # should not be reached, but covering our bases
+            return NotImplemented
+
+    else:
+        return NotImplemented
diff --git a/pandas/pandas/_libs/parsers.pyi b/pandas/pandas/_libs/parsers.pyi
new file mode 100644
index 0000000000000..253bb7303cefb
--- /dev/null
+++ b/pandas/pandas/_libs/parsers.pyi
@@ -0,0 +1,77 @@
+from typing import (
+    Hashable,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._typing import (
+    ArrayLike,
+    Dtype,
+    npt,
+)
+
+STR_NA_VALUES: set[str]
+DEFAULT_BUFFER_HEURISTIC: int
+
+def sanitize_objects(
+    values: npt.NDArray[np.object_],
+    na_values: set,
+) -> int: ...
+
+class TextReader:
+    unnamed_cols: set[str]
+    table_width: int  # int64_t
+    leading_cols: int  # int64_t
+    header: list[list[int]]  # non-negative integers
+    def __init__(
+        self,
+        source,
+        delimiter: bytes | str = ...,  # single-character only
+        header=...,
+        header_start: int = ...,  # int64_t
+        header_end: int = ...,  # uint64_t
+        index_col=...,
+        names=...,
+        tokenize_chunksize: int = ...,  # int64_t
+        delim_whitespace: bool = ...,
+        converters=...,
+        skipinitialspace: bool = ...,
+        escapechar: bytes | str | None = ...,  # single-character only
+        doublequote: bool = ...,
+        quotechar: str | bytes | None = ...,  # at most 1 character
+        quoting: int = ...,
+        lineterminator: bytes | str | None = ...,  # at most 1 character
+        comment=...,
+        decimal: bytes | str = ...,  # single-character only
+        thousands: bytes | str | None = ...,  # single-character only
+        dtype: Dtype | dict[Hashable, Dtype] = ...,
+        usecols=...,
+        error_bad_lines: bool = ...,
+        warn_bad_lines: bool = ...,
+        na_filter: bool = ...,
+        na_values=...,
+        na_fvalues=...,
+        keep_default_na: bool = ...,
+        true_values=...,
+        false_values=...,
+        allow_leading_cols: bool = ...,
+        skiprows=...,
+        skipfooter: int = ...,  # int64_t
+        verbose: bool = ...,
+        float_precision: Literal["round_trip", "legacy", "high"] | None = ...,
+        skip_blank_lines: bool = ...,
+        encoding_errors: bytes | str = ...,
+    ) -> None: ...
+    def set_noconvert(self, i: int) -> None: ...
+    def remove_noconvert(self, i: int) -> None: ...
+    def close(self) -> None: ...
+    def read(self, rows: int | None = ...) -> dict[int, ArrayLike]: ...
+    def read_low_memory(self, rows: int | None) -> list[dict[int, ArrayLike]]: ...
+
+# _maybe_upcast, na_values are only exposed for testing
+na_values: dict
+
+def _maybe_upcast(
+    arr, use_dtype_backend: bool = ..., dtype_backend: str = ...
+) -> np.ndarray: ...
diff --git a/pandas/pandas/_libs/parsers.pyx b/pandas/pandas/_libs/parsers.pyx
new file mode 100644
index 0000000000000..c29cdbcf5975e
--- /dev/null
+++ b/pandas/pandas/_libs/parsers.pyx
@@ -0,0 +1,2135 @@
+# Copyright (c) 2012, Lambda Foundry, Inc.
+# See LICENSE for the license
+from collections import defaultdict
+from csv import (
+    QUOTE_MINIMAL,
+    QUOTE_NONE,
+    QUOTE_NONNUMERIC,
+)
+import warnings
+
+from pandas.util._exceptions import find_stack_level
+
+from pandas import StringDtype
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BooleanArray,
+    FloatingArray,
+    IntegerArray,
+)
+
+cimport cython
+from cpython.bytes cimport PyBytes_AsString
+from cpython.exc cimport (
+    PyErr_Fetch,
+    PyErr_Occurred,
+)
+from cpython.object cimport PyObject
+from cpython.ref cimport (
+    Py_INCREF,
+    Py_XDECREF,
+)
+from cpython.unicode cimport (
+    PyUnicode_AsUTF8String,
+    PyUnicode_Decode,
+    PyUnicode_DecodeUTF8,
+    PyUnicode_FromString,
+)
+from cython cimport Py_ssize_t
+from libc.stdlib cimport free
+from libc.string cimport (
+    strcasecmp,
+    strlen,
+    strncpy,
+)
+
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    float64_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+    uint64_t,
+)
+
+cnp.import_array()
+
+from pandas._libs cimport util
+from pandas._libs.util cimport (
+    INT64_MAX,
+    INT64_MIN,
+    UINT64_MAX,
+)
+
+from pandas._libs import lib
+
+from pandas._libs.khash cimport (
+    kh_destroy_float64,
+    kh_destroy_str,
+    kh_destroy_str_starts,
+    kh_destroy_strbox,
+    kh_exist_str,
+    kh_float64_t,
+    kh_get_float64,
+    kh_get_str,
+    kh_get_str_starts_item,
+    kh_get_strbox,
+    kh_init_float64,
+    kh_init_str,
+    kh_init_str_starts,
+    kh_init_strbox,
+    kh_put_float64,
+    kh_put_str,
+    kh_put_str_starts_item,
+    kh_put_strbox,
+    kh_resize_float64,
+    kh_resize_str_starts,
+    kh_str_starts_t,
+    kh_str_t,
+    kh_strbox_t,
+    khiter_t,
+)
+
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+    ParserWarning,
+)
+
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.inference import is_dict_like
+
+from pandas.core.arrays.boolean import BooleanDtype
+
+cdef:
+    float64_t INF = <float64_t>np.inf
+    float64_t NEGINF = -INF
+    int64_t DEFAULT_CHUNKSIZE = 256 * 1024
+
+DEFAULT_BUFFER_HEURISTIC = 2 ** 20
+
+
+cdef extern from "pandas/portable.h":
+    # I *think* this is here so that strcasecmp is defined on Windows
+    # so we don't get
+    # `parsers.obj : error LNK2001: unresolved external symbol strcasecmp`
+    # in Appveyor.
+    # In a sane world, the `from libc.string cimport` above would fail
+    # loudly.
+    pass
+
+
+cdef extern from "pandas/parser/tokenizer.h":
+
+    ctypedef enum ParserState:
+        START_RECORD
+        START_FIELD
+        ESCAPED_CHAR
+        IN_FIELD
+        IN_QUOTED_FIELD
+        ESCAPE_IN_QUOTED_FIELD
+        QUOTE_IN_QUOTED_FIELD
+        EAT_CRNL
+        EAT_CRNL_NOP
+        EAT_WHITESPACE
+        EAT_COMMENT
+        EAT_LINE_COMMENT
+        WHITESPACE_LINE
+        SKIP_LINE
+        FINISHED
+
+    enum: ERROR_OVERFLOW
+
+    ctypedef enum BadLineHandleMethod:
+        ERROR,
+        WARN,
+        SKIP
+
+    ctypedef char* (*io_callback)(void *src, size_t nbytes, size_t *bytes_read,
+                                  int *status, const char *encoding_errors)
+    ctypedef void (*io_cleanup)(void *src)
+
+    ctypedef struct parser_t:
+        void *source
+        io_callback cb_io
+        io_cleanup cb_cleanup
+
+        int64_t chunksize  # Number of bytes to prepare for each chunk
+        char *data         # pointer to data to be processed
+        int64_t datalen    # amount of data available
+        int64_t datapos
+
+        # where to write out tokenized data
+        char *stream
+        uint64_t stream_len
+        uint64_t stream_cap
+
+        # Store words in (potentially ragged) matrix for now, hmm
+        char **words
+        int64_t *word_starts  # where we are in the stream
+        uint64_t words_len
+        uint64_t words_cap
+        uint64_t max_words_cap   # maximum word cap encountered
+
+        char *pword_start        # pointer to stream start of current field
+        int64_t word_start       # position start of current field
+
+        int64_t *line_start      # position in words for start of line
+        int64_t *line_fields     # Number of fields in each line
+        uint64_t lines           # Number of lines observed
+        uint64_t file_lines      # Number of lines observed (with bad/skipped)
+        uint64_t lines_cap       # Vector capacity
+
+        # Tokenizing stuff
+        ParserState state
+        int doublequote            # is " represented by ""? */
+        char delimiter             # field separator */
+        int delim_whitespace       # consume tabs / spaces instead
+        char quotechar             # quote character */
+        char escapechar            # escape character */
+        char lineterminator
+        int skipinitialspace       # ignore spaces following delimiter? */
+        int quoting                # style of quoting to write */
+
+        char commentchar
+        int allow_embedded_newline
+
+        int usecols
+
+        Py_ssize_t expected_fields
+        BadLineHandleMethod on_bad_lines
+
+        # floating point options
+        char decimal
+        char sci
+
+        # thousands separator (comma, period)
+        char thousands
+
+        int header                  # Boolean: 1: has header, 0: no header
+        int64_t header_start        # header row start
+        uint64_t header_end         # header row end
+
+        void *skipset
+        PyObject *skipfunc
+        int64_t skip_first_N_rows
+        int64_t skipfooter
+        # pick one, depending on whether the converter requires GIL
+        double (*double_converter)(const char *, char **,
+                                   char, char, char,
+                                   int, int *, int *) noexcept nogil
+
+        #  error handling
+        char *warn_msg
+        char *error_msg
+
+        int64_t skip_empty_lines
+
+    ctypedef struct coliter_t:
+        char **words
+        int64_t *line_start
+        int64_t col
+
+    ctypedef struct uint_state:
+        int seen_sint
+        int seen_uint
+        int seen_null
+
+    void COLITER_NEXT(coliter_t, const char *) nogil
+
+cdef extern from "pandas/parser/pd_parser.h":
+    void *new_rd_source(object obj) except NULL
+
+    void del_rd_source(void *src)
+
+    char* buffer_rd_bytes(void *source, size_t nbytes,
+                          size_t *bytes_read, int *status, const char *encoding_errors)
+
+    void uint_state_init(uint_state *self)
+    int uint64_conflict(uint_state *self)
+
+    void coliter_setup(coliter_t *it, parser_t *parser,
+                       int64_t i, int64_t start) nogil
+    void COLITER_NEXT(coliter_t, const char *) nogil
+
+    parser_t* parser_new()
+
+    int parser_init(parser_t *self) nogil
+    void parser_free(parser_t *self) nogil
+    void parser_del(parser_t *self) nogil
+    int parser_add_skiprow(parser_t *self, int64_t row)
+
+    void parser_set_skipfirstnrows(parser_t *self, int64_t nrows)
+
+    void parser_set_default_options(parser_t *self)
+
+    int parser_consume_rows(parser_t *self, size_t nrows)
+
+    int parser_trim_buffers(parser_t *self)
+
+    int tokenize_all_rows(parser_t *self, const char *encoding_errors) nogil
+    int tokenize_nrows(parser_t *self, size_t nrows, const char *encoding_errors) nogil
+
+    int64_t str_to_int64(char *p_item, int64_t int_min,
+                         int64_t int_max, int *error, char tsep) nogil
+    uint64_t str_to_uint64(uint_state *state, char *p_item, int64_t int_max,
+                           uint64_t uint_max, int *error, char tsep) nogil
+
+    double xstrtod(const char *p, char **q, char decimal,
+                   char sci, char tsep, int skip_trailing,
+                   int *error, int *maybe_int) nogil
+    double precise_xstrtod(const char *p, char **q, char decimal,
+                           char sci, char tsep, int skip_trailing,
+                           int *error, int *maybe_int) nogil
+    double round_trip(const char *p, char **q, char decimal,
+                      char sci, char tsep, int skip_trailing,
+                      int *error, int *maybe_int) nogil
+
+    int to_boolean(const char *item, uint8_t *val) nogil
+
+    void PandasParser_IMPORT()
+
+PandasParser_IMPORT
+
+# When not invoked directly but rather assigned as a function,
+# cdef extern'ed declarations seem to leave behind an undefined symbol
+cdef double xstrtod_wrapper(const char *p, char **q, char decimal,
+                            char sci, char tsep, int skip_trailing,
+                            int *error, int *maybe_int) noexcept nogil:
+    return xstrtod(p, q, decimal, sci, tsep, skip_trailing, error, maybe_int)
+
+
+cdef double precise_xstrtod_wrapper(const char *p, char **q, char decimal,
+                                    char sci, char tsep, int skip_trailing,
+                                    int *error, int *maybe_int) noexcept nogil:
+    return precise_xstrtod(p, q, decimal, sci, tsep, skip_trailing, error, maybe_int)
+
+
+cdef double round_trip_wrapper(const char *p, char **q, char decimal,
+                               char sci, char tsep, int skip_trailing,
+                               int *error, int *maybe_int) noexcept nogil:
+    return round_trip(p, q, decimal, sci, tsep, skip_trailing, error, maybe_int)
+
+
+cdef char* buffer_rd_bytes_wrapper(void *source, size_t nbytes,
+                                   size_t *bytes_read, int *status,
+                                   const char *encoding_errors) noexcept:
+    return buffer_rd_bytes(source, nbytes, bytes_read, status, encoding_errors)
+
+cdef void del_rd_source_wrapper(void *src) noexcept:
+    del_rd_source(src)
+
+
+cdef class TextReader:
+    """
+
+    # source: StringIO or file object
+
+    ..versionchange:: 1.2.0
+        removed 'compression', 'memory_map', and 'encoding' argument.
+        These arguments are outsourced to CParserWrapper.
+        'source' has to be a file handle.
+    """
+
+    cdef:
+        parser_t *parser
+        object na_fvalues
+        object true_values, false_values
+        object handle
+        object orig_header
+        bint na_filter, keep_default_na, has_usecols, has_mi_columns
+        bint allow_leading_cols
+        uint64_t parser_start  # this is modified after __init__
+        const char *encoding_errors
+        kh_str_starts_t *false_set
+        kh_str_starts_t *true_set
+        int64_t buffer_lines, skipfooter
+        list dtype_cast_order  # list[np.dtype]
+        list names   # can be None
+        set noconvert  # set[int]
+
+    cdef public:
+        int64_t leading_cols, table_width
+        object delimiter  # bytes or str
+        object converters
+        object na_values
+        list header  # list[list[non-negative integers]]
+        object index_col
+        object skiprows
+        object dtype
+        object usecols
+        set unnamed_cols  # set[str]
+        str dtype_backend
+
+    def __cinit__(self, source,
+                  delimiter=b",",  # bytes | str
+                  header=0,
+                  int64_t header_start=0,
+                  uint64_t header_end=0,
+                  index_col=None,
+                  names=None,
+                  tokenize_chunksize=DEFAULT_CHUNKSIZE,
+                  bint delim_whitespace=False,
+                  converters=None,
+                  bint skipinitialspace=False,
+                  escapechar=None,      # bytes | str
+                  bint doublequote=True,
+                  quotechar=b'"',
+                  quoting=0,            # int
+                  lineterminator=None,  # bytes | str
+                  comment=None,
+                  decimal=b".",         # bytes | str
+                  thousands=None,       # bytes | str
+                  dtype=None,
+                  usecols=None,
+                  on_bad_lines=ERROR,
+                  bint na_filter=True,
+                  na_values=None,
+                  na_fvalues=None,
+                  bint keep_default_na=True,
+                  true_values=None,
+                  false_values=None,
+                  bint allow_leading_cols=True,
+                  skiprows=None,
+                  skipfooter=0,         # int64_t
+                  float_precision=None,
+                  bint skip_blank_lines=True,
+                  encoding_errors=b"strict",
+                  dtype_backend="numpy"):
+
+        # set encoding for native Python and C library
+        if isinstance(encoding_errors, str):
+            encoding_errors = encoding_errors.encode("utf-8")
+        elif encoding_errors is None:
+            encoding_errors = b"strict"
+        Py_INCREF(encoding_errors)
+        self.encoding_errors = PyBytes_AsString(encoding_errors)
+
+        self.parser = parser_new()
+        self.parser.chunksize = tokenize_chunksize
+
+        self.parser.usecols = (usecols is not None)
+
+        self._setup_parser_source(source)
+        parser_set_default_options(self.parser)
+
+        parser_init(self.parser)
+
+        if delim_whitespace:
+            self.parser.delim_whitespace = delim_whitespace
+        else:
+            if len(delimiter) > 1:
+                raise ValueError("only length-1 separators excluded right now")
+            self.parser.delimiter = <char>ord(delimiter)
+
+        # ----------------------------------------
+        # parser options
+
+        self.parser.doublequote = doublequote
+        self.parser.skipinitialspace = skipinitialspace
+        self.parser.skip_empty_lines = skip_blank_lines
+
+        if lineterminator is not None:
+            if len(lineterminator) != 1:
+                raise ValueError("Only length-1 line terminators supported")
+            self.parser.lineterminator = <char>ord(lineterminator)
+
+        if len(decimal) != 1:
+            raise ValueError("Only length-1 decimal markers supported")
+        self.parser.decimal = <char>ord(decimal)
+
+        if thousands is not None:
+            if len(thousands) != 1:
+                raise ValueError("Only length-1 thousands markers supported")
+            self.parser.thousands = <char>ord(thousands)
+
+        if escapechar is not None:
+            if len(escapechar) != 1:
+                raise ValueError("Only length-1 escapes supported")
+            self.parser.escapechar = <char>ord(escapechar)
+
+        self._set_quoting(quotechar, quoting)
+
+        dtype_order = ["int64", "float64", "bool", "object"]
+        if quoting == QUOTE_NONNUMERIC:
+            # consistent with csv module semantics, cast all to float
+            dtype_order = dtype_order[1:]
+        self.dtype_cast_order = [np.dtype(x) for x in dtype_order]
+
+        if comment is not None:
+            if len(comment) > 1:
+                raise ValueError("Only length-1 comment characters supported")
+            self.parser.commentchar = <char>ord(comment)
+
+        self.parser.on_bad_lines = on_bad_lines
+
+        self.skiprows = skiprows
+        if skiprows is not None:
+            self._make_skiprow_set()
+
+        self.skipfooter = skipfooter
+
+        if usecols is not None:
+            self.has_usecols = 1
+            # GH-20558, validate usecols at higher level and only pass clean
+            # usecols into TextReader.
+            self.usecols = usecols
+
+        if skipfooter > 0:
+            self.parser.on_bad_lines = SKIP
+
+        self.delimiter = delimiter
+
+        self.na_values = na_values
+        if na_fvalues is None:
+            na_fvalues = set()
+        self.na_fvalues = na_fvalues
+
+        self.true_values = _maybe_encode(true_values) + _true_values
+        self.false_values = _maybe_encode(false_values) + _false_values
+
+        self.true_set = kset_from_list(self.true_values)
+        self.false_set = kset_from_list(self.false_values)
+
+        self.keep_default_na = keep_default_na
+        self.converters = converters
+        self.na_filter = na_filter
+
+        if float_precision == "round_trip":
+            # see gh-15140
+            self.parser.double_converter = round_trip_wrapper
+        elif float_precision == "legacy":
+            self.parser.double_converter = xstrtod_wrapper
+        elif float_precision == "high" or float_precision is None:
+            self.parser.double_converter = precise_xstrtod_wrapper
+        else:
+            raise ValueError(f"Unrecognized float_precision option: "
+                             f"{float_precision}")
+
+        # Caller is responsible for ensuring we have one of
+        # - None
+        # - DtypeObj
+        # - dict[Any, DtypeObj]
+        self.dtype = dtype
+        self.dtype_backend = dtype_backend
+
+        self.noconvert = set()
+
+        self.index_col = index_col
+
+        # ----------------------------------------
+        # header stuff
+
+        self.allow_leading_cols = allow_leading_cols
+        self.leading_cols = 0  # updated in _get_header
+
+        # TODO: no header vs. header is not the first row
+        self.has_mi_columns = 0
+        self.orig_header = header
+        if header is None:
+            # sentinel value
+            self.parser.header_start = -1
+            self.parser.header_end = -1
+            self.parser.header = -1
+            self.parser_start = 0
+            prelim_header = []
+        else:
+            if isinstance(header, list):
+                if len(header) > 1:
+                    # need to artificially skip the final line
+                    # which is still a header line
+                    header = list(header)
+                    header.append(header[-1] + 1)
+                    self.parser.header_end = header[-1]
+                    self.has_mi_columns = 1
+                else:
+                    self.parser.header_end = header[0]
+
+                self.parser_start = header[-1] + 1
+                self.parser.header_start = header[0]
+                self.parser.header = header[0]
+                prelim_header = header
+            else:
+                self.parser.header_start = header
+                self.parser.header_end = header
+                self.parser_start = header + 1
+                self.parser.header = header
+                prelim_header = [header]
+
+        self.names = names
+        header, table_width, unnamed_cols = self._get_header(prelim_header)
+        # header, table_width, and unnamed_cols are set here, never changed
+        self.header = header
+        self.table_width = table_width
+        self.unnamed_cols = unnamed_cols
+
+        if not self.table_width:
+            raise EmptyDataError("No columns to parse from file")
+
+        # Compute buffer_lines as function of table width.
+        heuristic = DEFAULT_BUFFER_HEURISTIC // self.table_width
+        self.buffer_lines = 1
+        while self.buffer_lines * 2 < heuristic:
+            self.buffer_lines *= 2
+
+    def __init__(self, *args, **kwargs):
+        pass
+
+    def __dealloc__(self):
+        _close(self)
+        parser_del(self.parser)
+
+    def close(self):
+        _close(self)
+
+    def _set_quoting(self, quote_char: str | bytes | None, quoting: int):
+        if not isinstance(quoting, int):
+            raise TypeError('"quoting" must be an integer')
+
+        if not QUOTE_MINIMAL <= quoting <= QUOTE_NONE:
+            raise TypeError('bad "quoting" value')
+
+        if not isinstance(quote_char, (str, bytes)) and quote_char is not None:
+            dtype = type(quote_char).__name__
+            raise TypeError(f'"quotechar" must be string, not {dtype}')
+
+        if quote_char is None or quote_char == "":
+            if quoting != QUOTE_NONE:
+                raise TypeError("quotechar must be set if quoting enabled")
+            self.parser.quoting = quoting
+            self.parser.quotechar = -1
+        elif len(quote_char) > 1:  # 0-len case handled earlier
+            raise TypeError('"quotechar" must be a 1-character string')
+        else:
+            self.parser.quoting = quoting
+            self.parser.quotechar = <char>ord(quote_char)
+
+    cdef _make_skiprow_set(self):
+        if util.is_integer_object(self.skiprows):
+            parser_set_skipfirstnrows(self.parser, self.skiprows)
+        elif not callable(self.skiprows):
+            for i in self.skiprows:
+                parser_add_skiprow(self.parser, i)
+        else:
+            self.parser.skipfunc = <PyObject *>self.skiprows
+
+    cdef _setup_parser_source(self, source):
+        cdef:
+            void *ptr
+
+        ptr = new_rd_source(source)
+        self.parser.source = ptr
+        self.parser.cb_io = buffer_rd_bytes_wrapper
+        self.parser.cb_cleanup = del_rd_source_wrapper
+
+    cdef _get_header(self, list prelim_header):
+        # header is now a list of lists, so field_count should use header[0]
+        #
+        # modifies:
+        #   self.parser attributes
+        #   self.parser_start
+        #   self.leading_cols
+
+        cdef:
+            Py_ssize_t i, start, field_count, passed_count, unnamed_count, level
+            char *word
+            str name
+            uint64_t hr, data_line = 0
+            list header = []
+            set unnamed_cols = set()
+
+        if self.parser.header_start >= 0:
+
+            # Header is in the file
+            for level, hr in enumerate(prelim_header):
+
+                this_header = []
+
+                if self.parser.lines < hr + 1:
+                    self._tokenize_rows(hr + 2)
+
+                if self.parser.lines == 0:
+                    field_count = 0
+                    start = self.parser.line_start[0]
+
+                # e.g., if header=3 and file only has 2 lines
+                elif (self.parser.lines < hr + 1
+                      and not isinstance(self.orig_header, list)) or (
+                          self.parser.lines < hr):
+                    msg = self.orig_header
+                    if isinstance(msg, list):
+                        joined = ",".join(str(m) for m in msg)
+                        msg = f"[{joined}], len of {len(msg)},"
+                    raise ParserError(
+                        f"Passed header={msg} but only "
+                        f"{self.parser.lines} lines in file")
+
+                else:
+                    field_count = self.parser.line_fields[hr]
+                    start = self.parser.line_start[hr]
+
+                unnamed_count = 0
+                unnamed_col_indices = []
+
+                for i in range(field_count):
+                    word = self.parser.words[start + i]
+
+                    name = PyUnicode_DecodeUTF8(word, strlen(word),
+                                                self.encoding_errors)
+
+                    if name == "":
+                        if self.has_mi_columns:
+                            name = f"Unnamed: {i}_level_{level}"
+                        else:
+                            name = f"Unnamed: {i}"
+
+                        unnamed_count += 1
+                        unnamed_col_indices.append(i)
+
+                    this_header.append(name)
+
+                if not self.has_mi_columns:
+                    # Ensure that regular columns are used before unnamed ones
+                    # to keep given names and mangle unnamed columns
+                    col_loop_order = [i for i in range(len(this_header))
+                                      if i not in unnamed_col_indices
+                                      ] + unnamed_col_indices
+                    counts = {}
+
+                    for i in col_loop_order:
+                        col = this_header[i]
+                        old_col = col
+                        cur_count = counts.get(col, 0)
+
+                        if cur_count > 0:
+                            while cur_count > 0:
+                                counts[old_col] = cur_count + 1
+                                col = f"{old_col}.{cur_count}"
+                                if col in this_header:
+                                    cur_count += 1
+                                else:
+                                    cur_count = counts.get(col, 0)
+
+                            if (
+                                self.dtype is not None
+                                and is_dict_like(self.dtype)
+                                and self.dtype.get(old_col) is not None
+                                and self.dtype.get(col) is None
+                            ):
+                                self.dtype.update({col: self.dtype.get(old_col)})
+
+                        this_header[i] = col
+                        counts[col] = cur_count + 1
+
+                if self.has_mi_columns:
+
+                    # If we have grabbed an extra line, but it's not in our
+                    # format, save in the buffer, and create an blank extra
+                    # line for the rest of the parsing code.
+                    if hr == prelim_header[-1]:
+                        lc = len(this_header)
+                        ic = (len(self.index_col) if self.index_col
+                              is not None else 0)
+
+                        # if wrong number of blanks or no index, not our format
+                        if (lc != unnamed_count and lc - ic > unnamed_count) or ic == 0:
+                            hr -= 1
+                            self.parser_start -= 1
+                            this_header = [None] * lc
+
+                data_line = hr + 1
+                header.append(this_header)
+                unnamed_cols.update({this_header[i] for i in unnamed_col_indices})
+
+            if self.names is not None:
+                header = [self.names]
+
+        elif self.names is not None:
+            # Names passed
+            if self.parser.lines < 1:
+                if not self.has_usecols:
+                    self.parser.expected_fields = len(self.names)
+                self._tokenize_rows(1)
+
+            header = [self.names]
+
+            if self.parser.lines < 1:
+                field_count = len(header[0])
+            else:
+                field_count = self.parser.line_fields[data_line]
+
+            # Enforce this unless usecols
+            if not self.has_usecols:
+                self.parser.expected_fields = max(field_count, len(self.names))
+
+        else:
+            # No header passed nor to be found in the file
+            if self.parser.lines < 1:
+                self._tokenize_rows(1)
+
+            return None, self.parser.line_fields[0], unnamed_cols
+
+        # Corner case, not enough lines in the file
+        if self.parser.lines < data_line + 1:
+            field_count = len(header[0])
+        else:
+
+            field_count = self.parser.line_fields[data_line]
+
+            # #2981
+            if self.names is not None:
+                field_count = max(field_count, len(self.names))
+
+            passed_count = len(header[0])
+
+            if (self.has_usecols and self.allow_leading_cols and
+                    not callable(self.usecols)):
+                nuse = len(self.usecols)
+                if nuse == passed_count:
+                    self.leading_cols = 0
+                elif self.names is None and nuse < passed_count:
+                    self.leading_cols = field_count - passed_count
+                elif passed_count != field_count:
+                    raise ValueError("Number of passed names did not match number of "
+                                     "header fields in the file")
+            # oh boy, #2442, #2981
+            elif self.allow_leading_cols and passed_count < field_count:
+                self.leading_cols = field_count - passed_count
+
+        return header, field_count, unnamed_cols
+
+    def read(self, rows: int | None = None) -> dict[int, "ArrayLike"]:
+        """
+        rows=None --> read all rows
+        """
+        # Don't care about memory usage
+        columns = self._read_rows(rows, 1)
+
+        return columns
+
+    def read_low_memory(self, rows: int | None)-> list[dict[int, "ArrayLike"]]:
+        """
+        rows=None --> read all rows
+        """
+        # Conserve intermediate space
+        # Caller is responsible for concatenating chunks,
+        #  see c_parser_wrapper._concatenate_chunks
+        cdef:
+            size_t rows_read = 0
+            list chunks = []
+
+        if rows is None:
+            while True:
+                try:
+                    chunk = self._read_rows(self.buffer_lines, 0)
+                    if len(chunk) == 0:
+                        break
+                except StopIteration:
+                    break
+                else:
+                    chunks.append(chunk)
+        else:
+            while rows_read < rows:
+                try:
+                    crows = min(self.buffer_lines, rows - rows_read)
+
+                    chunk = self._read_rows(crows, 0)
+                    if len(chunk) == 0:
+                        break
+
+                    rows_read += len(list(chunk.values())[0])
+                except StopIteration:
+                    break
+                else:
+                    chunks.append(chunk)
+
+        parser_trim_buffers(self.parser)
+
+        if len(chunks) == 0:
+            raise StopIteration
+
+        return chunks
+
+    cdef _tokenize_rows(self, size_t nrows):
+        cdef:
+            int status
+
+        with nogil:
+            status = tokenize_nrows(self.parser, nrows, self.encoding_errors)
+
+        self._check_tokenize_status(status)
+
+    cdef _check_tokenize_status(self, int status):
+        if self.parser.warn_msg != NULL:
+            warnings.warn(
+                PyUnicode_DecodeUTF8(
+                    self.parser.warn_msg,
+                    strlen(self.parser.warn_msg),
+                    self.encoding_errors
+                ),
+                ParserWarning,
+                stacklevel=find_stack_level()
+            )
+            free(self.parser.warn_msg)
+            self.parser.warn_msg = NULL
+
+        if status < 0:
+            raise_parser_error("Error tokenizing data", self.parser)
+
+    #  -> dict[int, "ArrayLike"]
+    cdef _read_rows(self, rows, bint trim):
+        cdef:
+            int64_t buffered_lines
+            int64_t irows
+
+        if rows is not None:
+            irows = rows
+            buffered_lines = self.parser.lines - self.parser_start
+            if buffered_lines < irows:
+                self._tokenize_rows(irows - buffered_lines)
+
+            if self.skipfooter > 0:
+                raise ValueError("skipfooter can only be used to read "
+                                 "the whole file")
+        else:
+            with nogil:
+                status = tokenize_all_rows(self.parser, self.encoding_errors)
+
+            self._check_tokenize_status(status)
+
+        if self.parser_start >= self.parser.lines:
+            raise StopIteration
+
+        columns = self._convert_column_data(rows)
+        if len(columns) > 0:
+            rows_read = len(list(columns.values())[0])
+            # trim
+            parser_consume_rows(self.parser, rows_read)
+            if trim:
+                parser_trim_buffers(self.parser)
+            self.parser_start -= rows_read
+
+        return columns
+
+    def set_noconvert(self, i: int) -> None:
+        self.noconvert.add(i)
+
+    def remove_noconvert(self, i: int) -> None:
+        self.noconvert.remove(i)
+
+    def _convert_column_data(self, rows: int | None) -> dict[int, "ArrayLike"]:
+        cdef:
+            int64_t i
+            int nused
+            kh_str_starts_t *na_hashset = NULL
+            int64_t start, end
+            object name, na_flist, col_dtype = None
+            bint na_filter = 0
+            int64_t num_cols
+            dict results
+
+        start = self.parser_start
+
+        if rows is None:
+            end = self.parser.lines
+        else:
+            end = min(start + rows, self.parser.lines)
+
+        num_cols = -1
+        # Py_ssize_t cast prevents build warning
+        for i in range(<Py_ssize_t>self.parser.lines):
+            num_cols = (num_cols < self.parser.line_fields[i]) * \
+                self.parser.line_fields[i] + \
+                (num_cols >= self.parser.line_fields[i]) * num_cols
+
+        usecols_not_callable_and_exists = not callable(self.usecols) and self.usecols
+        names_larger_num_cols = (self.names and
+                                 len(self.names) - self.leading_cols > num_cols)
+
+        if self.table_width - self.leading_cols > num_cols:
+            if (usecols_not_callable_and_exists
+                    and self.table_width - self.leading_cols < len(self.usecols)
+                    or names_larger_num_cols):
+                raise ParserError(f"Too many columns specified: expected "
+                                  f"{self.table_width - self.leading_cols} "
+                                  f"and found {num_cols}")
+
+        if (usecols_not_callable_and_exists and
+                all(isinstance(u, int) for u in self.usecols)):
+            missing_usecols = [col for col in self.usecols if col >= num_cols]
+            if missing_usecols:
+                raise ParserError(
+                    "Defining usecols with out-of-bounds indices is not allowed. "
+                    f"{missing_usecols} are out of bounds.",
+                )
+
+        results = {}
+        nused = 0
+        is_default_dict_dtype = isinstance(self.dtype, defaultdict)
+
+        for i in range(self.table_width):
+            if i < self.leading_cols:
+                # Pass through leading columns always
+                name = i
+            elif (self.usecols and not callable(self.usecols) and
+                    nused == len(self.usecols)):
+                # Once we've gathered all requested columns, stop. GH5766
+                break
+            else:
+                name = self._get_column_name(i, nused)
+                usecols = set()
+                if callable(self.usecols):
+                    if self.usecols(name):
+                        usecols = {i}
+                else:
+                    usecols = self.usecols
+                if self.has_usecols and not (i in usecols or
+                                             name in usecols):
+                    continue
+                nused += 1
+
+            conv = self._get_converter(i, name)
+
+            col_dtype = None
+            if self.dtype is not None:
+                if isinstance(self.dtype, dict):
+                    if name in self.dtype:
+                        col_dtype = self.dtype[name]
+                    elif i in self.dtype:
+                        col_dtype = self.dtype[i]
+                    elif is_default_dict_dtype:
+                        col_dtype = self.dtype[name]
+                else:
+                    if self.dtype.names:
+                        # structured array
+                        col_dtype = np.dtype(self.dtype.descr[i][1])
+                    else:
+                        col_dtype = self.dtype
+
+            if conv:
+                if col_dtype is not None:
+                    warnings.warn((f"Both a converter and dtype were specified "
+                                   f"for column {name} - only the converter will "
+                                   f"be used."), ParserWarning,
+                                  stacklevel=find_stack_level())
+                results[i] = _apply_converter(conv, self.parser, i, start, end)
+                continue
+
+            # Collect the list of NaN values associated with the column.
+            # If we aren't supposed to do that, or none are collected,
+            # we set `na_filter` to `0` (`1` otherwise).
+            na_flist = set()
+
+            if self.na_filter:
+                na_list, na_flist = self._get_na_list(i, name)
+                if na_list is None:
+                    na_filter = 0
+                else:
+                    na_filter = 1
+                    na_hashset = kset_from_list(na_list)
+            else:
+                na_filter = 0
+
+            # Attempt to parse tokens and infer dtype of the column.
+            # Should return as the desired dtype (inferred or specified).
+            try:
+                col_res, na_count = self._convert_tokens(
+                    i, start, end, name, na_filter, na_hashset,
+                    na_flist, col_dtype)
+            finally:
+                # gh-21353
+                #
+                # Cleanup the NaN hash that we generated
+                # to avoid memory leaks.
+                if na_filter:
+                    self._free_na_set(na_hashset)
+
+            # don't try to upcast EAs
+            if (
+                na_count > 0 and not isinstance(col_dtype, ExtensionDtype)
+                or self.dtype_backend != "numpy"
+            ):
+                use_dtype_backend = self.dtype_backend != "numpy" and col_dtype is None
+                col_res = _maybe_upcast(
+                    col_res,
+                    use_dtype_backend=use_dtype_backend,
+                    dtype_backend=self.dtype_backend,
+                )
+
+            if col_res is None:
+                raise ParserError(f"Unable to parse column {i}")
+
+            results[i] = col_res
+
+        self.parser_start += end - start
+
+        return results
+
+    # -> tuple["ArrayLike", int]:
+    cdef _convert_tokens(self, Py_ssize_t i, int64_t start,
+                         int64_t end, object name, bint na_filter,
+                         kh_str_starts_t *na_hashset,
+                         object na_flist, object col_dtype):
+
+        if col_dtype is not None:
+            col_res, na_count = self._convert_with_dtype(
+                col_dtype, i, start, end, na_filter,
+                1, na_hashset, na_flist)
+
+            # Fallback on the parse (e.g. we requested int dtype,
+            # but its actually a float).
+            if col_res is not None:
+                return col_res, na_count
+
+        if i in self.noconvert:
+            return self._string_convert(i, start, end, na_filter, na_hashset)
+        else:
+            col_res = None
+            for dt in self.dtype_cast_order:
+                try:
+                    col_res, na_count = self._convert_with_dtype(
+                        dt, i, start, end, na_filter, 0, na_hashset, na_flist)
+                except ValueError:
+                    # This error is raised from trying to convert to uint64,
+                    # and we discover that we cannot convert to any numerical
+                    # dtype successfully. As a result, we leave the data
+                    # column AS IS with object dtype.
+                    col_res, na_count = self._convert_with_dtype(
+                        np.dtype("object"), i, start, end, 0,
+                        0, na_hashset, na_flist)
+                except OverflowError:
+                    col_res, na_count = self._convert_with_dtype(
+                        np.dtype("object"), i, start, end, na_filter,
+                        0, na_hashset, na_flist)
+
+                if col_res is not None:
+                    break
+
+        # we had a fallback parse on the dtype, so now try to cast
+        if col_res is not None and col_dtype is not None:
+            # If col_res is bool, it might actually be a bool array mixed with NaNs
+            # (see _try_bool_flex()). Usually this would be taken care of using
+            # _maybe_upcast(), but if col_dtype is a floating type we should just
+            # take care of that cast here.
+            if col_res.dtype == np.bool_ and col_dtype.kind == "f":
+                mask = col_res.view(np.uint8) == na_values[np.uint8]
+                col_res = col_res.astype(col_dtype)
+                np.putmask(col_res, mask, np.nan)
+                return col_res, na_count
+
+            # NaNs are already cast to True here, so can not use astype
+            if col_res.dtype == np.bool_ and col_dtype.kind in "iu":
+                if na_count > 0:
+                    raise ValueError(
+                        f"cannot safely convert passed user dtype of "
+                        f"{col_dtype} for {np.bool_} dtyped data in "
+                        f"column {i} due to NA values"
+                    )
+
+            # only allow safe casts, eg. with a nan you cannot safely cast to int
+            try:
+                col_res = col_res.astype(col_dtype, casting="safe")
+            except TypeError:
+
+                # float -> int conversions can fail the above
+                # even with no nans
+                col_res_orig = col_res
+                col_res = col_res.astype(col_dtype)
+                if (col_res != col_res_orig).any():
+                    raise ValueError(
+                        f"cannot safely convert passed user dtype of "
+                        f"{col_dtype} for {col_res_orig.dtype.name} dtyped data in "
+                        f"column {i}")
+
+        return col_res, na_count
+
+    cdef _convert_with_dtype(self, object dtype, Py_ssize_t i,
+                             int64_t start, int64_t end,
+                             bint na_filter,
+                             bint user_dtype,
+                             kh_str_starts_t *na_hashset,
+                             object na_flist):
+        if isinstance(dtype, CategoricalDtype):
+            # TODO: I suspect that _categorical_convert could be
+            # optimized when dtype is an instance of CategoricalDtype
+            codes, cats, na_count = _categorical_convert(
+                self.parser, i, start, end, na_filter, na_hashset)
+
+            # Method accepts list of strings, not encoded ones.
+            true_values = [x.decode() for x in self.true_values]
+            array_type = dtype.construct_array_type()
+            cat = array_type._from_inferred_categories(
+                cats, codes, dtype, true_values=true_values)
+            return cat, na_count
+
+        elif isinstance(dtype, ExtensionDtype):
+            result, na_count = self._string_convert(i, start, end, na_filter,
+                                                    na_hashset)
+
+            array_type = dtype.construct_array_type()
+            try:
+                # use _from_sequence_of_strings if the class defines it
+                if isinstance(dtype, BooleanDtype):
+                    # xref GH 47534: BooleanArray._from_sequence_of_strings has extra
+                    # kwargs
+                    true_values = [x.decode() for x in self.true_values]
+                    false_values = [x.decode() for x in self.false_values]
+                    result = array_type._from_sequence_of_strings(
+                        result, dtype=dtype, true_values=true_values,
+                        false_values=false_values)
+                else:
+                    result = array_type._from_sequence_of_strings(result, dtype=dtype)
+            except NotImplementedError:
+                raise NotImplementedError(
+                    f"Extension Array: {array_type} must implement "
+                    f"_from_sequence_of_strings in order "
+                    f"to be used in parser methods")
+
+            return result, na_count
+
+        elif dtype.kind in "iu":
+            try:
+                result, na_count = _try_int64(self.parser, i, start,
+                                              end, na_filter, na_hashset)
+                if user_dtype and na_count is not None:
+                    if na_count > 0:
+                        raise ValueError(f"Integer column has NA values in column {i}")
+            except OverflowError:
+                result = _try_uint64(self.parser, i, start, end,
+                                     na_filter, na_hashset)
+                na_count = 0
+
+            if result is not None and dtype != "int64":
+                result = result.astype(dtype)
+
+            return result, na_count
+
+        elif dtype.kind == "f":
+            result, na_count = _try_double(self.parser, i, start, end,
+                                           na_filter, na_hashset, na_flist)
+
+            if result is not None and dtype != "float64":
+                result = result.astype(dtype)
+            return result, na_count
+        elif dtype.kind == "b":
+            result, na_count = _try_bool_flex(self.parser, i, start, end,
+                                              na_filter, na_hashset,
+                                              self.true_set, self.false_set)
+            if user_dtype and na_count is not None:
+                if na_count > 0:
+                    raise ValueError(f"Bool column has NA values in column {i}")
+            return result, na_count
+
+        elif dtype.kind == "S":
+            # TODO: na handling
+            width = dtype.itemsize
+            if width > 0:
+                result = _to_fw_string(self.parser, i, start, end, width)
+                return result, 0
+
+            # treat as a regular string parsing
+            return self._string_convert(i, start, end, na_filter,
+                                        na_hashset)
+        elif dtype.kind == "U":
+            width = dtype.itemsize
+            if width > 0:
+                raise TypeError(f"the dtype {dtype} is not supported for parsing")
+
+            # unicode variable width
+            return self._string_convert(i, start, end, na_filter,
+                                        na_hashset)
+        elif dtype == object:
+            return self._string_convert(i, start, end, na_filter,
+                                        na_hashset)
+        elif dtype.kind == "M":
+            raise TypeError(f"the dtype {dtype} is not supported "
+                            f"for parsing, pass this column "
+                            f"using parse_dates instead")
+        else:
+            raise TypeError(f"the dtype {dtype} is not supported for parsing")
+
+    # -> tuple[ndarray[object], int]
+    cdef _string_convert(self, Py_ssize_t i, int64_t start, int64_t end,
+                         bint na_filter, kh_str_starts_t *na_hashset):
+
+        return _string_box_utf8(self.parser, i, start, end, na_filter,
+                                na_hashset, self.encoding_errors)
+
+    def _get_converter(self, i: int, name):
+        if self.converters is None:
+            return None
+
+        if name is not None and name in self.converters:
+            return self.converters[name]
+
+        # Converter for position, if any
+        return self.converters.get(i)
+
+    cdef _get_na_list(self, Py_ssize_t i, name):
+        # Note: updates self.na_values, self.na_fvalues
+        if self.na_values is None:
+            return None, set()
+
+        if isinstance(self.na_values, dict):
+            key = None
+            values = None
+
+            if name is not None and name in self.na_values:
+                key = name
+            elif i in self.na_values:
+                key = i
+            else:  # No na_values provided for this column.
+                if self.keep_default_na:
+                    return _NA_VALUES, set()
+
+                return list(), set()
+
+            values = self.na_values[key]
+            if values is not None and not isinstance(values, list):
+                values = list(values)
+
+            fvalues = self.na_fvalues[key]
+            if fvalues is not None and not isinstance(fvalues, set):
+                fvalues = set(fvalues)
+
+            return _ensure_encoded(values), fvalues
+        else:
+            if not isinstance(self.na_values, list):
+                self.na_values = list(self.na_values)
+            if not isinstance(self.na_fvalues, set):
+                self.na_fvalues = set(self.na_fvalues)
+
+            return _ensure_encoded(self.na_values), self.na_fvalues
+
+    cdef _free_na_set(self, kh_str_starts_t *table):
+        kh_destroy_str_starts(table)
+
+    cdef _get_column_name(self, Py_ssize_t i, Py_ssize_t nused):
+        cdef int64_t j
+        if self.has_usecols and self.names is not None:
+            if (not callable(self.usecols) and
+                    len(self.names) == len(self.usecols)):
+                return self.names[nused]
+            else:
+                return self.names[i - self.leading_cols]
+        else:
+            if self.header is not None:
+                j = i - self.leading_cols
+                # generate extra (bogus) headers if there are more columns than headers
+                # These should be strings, not integers, because otherwise we might get
+                # issues with callables as usecols GH#46997
+                if j >= len(self.header[0]):
+                    return str(j)
+                elif self.has_mi_columns:
+                    return tuple(header_row[j] for header_row in self.header)
+                else:
+                    return self.header[0][j]
+            else:
+                return None
+
+
+# Factor out code common to TextReader.__dealloc__ and TextReader.close
+# It cannot be a class method, since calling self.close() in __dealloc__
+# which causes a class attribute lookup and violates best practices
+# https://cython.readthedocs.io/en/latest/src/userguide/special_methods.html#finalization-method-dealloc
+cdef _close(TextReader reader):
+    # also preemptively free all allocated memory
+    parser_free(reader.parser)
+    if reader.true_set:
+        kh_destroy_str_starts(reader.true_set)
+        reader.true_set = NULL
+    if reader.false_set:
+        kh_destroy_str_starts(reader.false_set)
+        reader.false_set = NULL
+
+
+cdef:
+    object _true_values = [b"True", b"TRUE", b"true"]
+    object _false_values = [b"False", b"FALSE", b"false"]
+
+
+def _ensure_encoded(list lst):
+    cdef:
+        list result = []
+    for x in lst:
+        if isinstance(x, str):
+            x = PyUnicode_AsUTF8String(x)
+        elif not isinstance(x, bytes):
+            x = str(x).encode("utf-8")
+
+        result.append(x)
+    return result
+
+
+# common NA values
+# no longer excluding inf representations
+# '1.#INF','-1.#INF', '1.#INF000000',
+STR_NA_VALUES = {
+    "-1.#IND",
+    "1.#QNAN",
+    "1.#IND",
+    "-1.#QNAN",
+    "#N/A N/A",
+    "#N/A",
+    "N/A",
+    "n/a",
+    "NA",
+    "<NA>",
+    "#NA",
+    "NULL",
+    "null",
+    "NaN",
+    "-NaN",
+    "nan",
+    "-nan",
+    "",
+    "None",
+}
+_NA_VALUES = _ensure_encoded(list(STR_NA_VALUES))
+
+
+def _maybe_upcast(
+    arr, use_dtype_backend: bool = False, dtype_backend: str = "numpy"
+):
+    """Sets nullable dtypes or upcasts if nans are present.
+
+    Upcast, if use_dtype_backend is false and nans are present so that the
+    current dtype can not hold the na value. We use nullable dtypes if the
+    flag is true for every array.
+
+    Parameters
+    ----------
+    arr: ndarray
+        Numpy array that is potentially being upcast.
+
+    use_dtype_backend: bool, default False
+        If true, we cast to the associated nullable dtypes.
+
+    Returns
+    -------
+    The casted array.
+    """
+    if isinstance(arr.dtype, ExtensionDtype):
+        # TODO: the docstring says arr is an ndarray, in which case this cannot
+        #  be reached. Is that incorrect?
+        return arr
+
+    na_value = na_values[arr.dtype]
+
+    if issubclass(arr.dtype.type, np.integer):
+        mask = arr == na_value
+
+        if use_dtype_backend:
+            arr = IntegerArray(arr, mask)
+        else:
+            arr = arr.astype(float)
+            np.putmask(arr, mask, np.nan)
+
+    elif arr.dtype == np.bool_:
+        mask = arr.view(np.uint8) == na_value
+
+        if use_dtype_backend:
+            arr = BooleanArray(arr, mask)
+        else:
+            arr = arr.astype(object)
+            np.putmask(arr, mask, np.nan)
+
+    elif issubclass(arr.dtype.type, float) or arr.dtype.type == np.float32:
+        if use_dtype_backend:
+            mask = np.isnan(arr)
+            arr = FloatingArray(arr, mask)
+
+    elif arr.dtype == np.object_:
+        if use_dtype_backend:
+            dtype = StringDtype()
+            cls = dtype.construct_array_type()
+            arr = cls._from_sequence(arr, dtype=dtype)
+
+    if use_dtype_backend and dtype_backend == "pyarrow":
+        import pyarrow as pa
+        if isinstance(arr, IntegerArray) and arr.isna().all():
+            # use null instead of int64 in pyarrow
+            arr = arr.to_numpy(na_value=None)
+        arr = ArrowExtensionArray(pa.array(arr, from_pandas=True))
+
+    return arr
+
+
+# ----------------------------------------------------------------------
+# Type conversions / inference support code
+
+
+# -> tuple[ndarray[object], int]
+cdef _string_box_utf8(parser_t *parser, int64_t col,
+                      int64_t line_start, int64_t line_end,
+                      bint na_filter, kh_str_starts_t *na_hashset,
+                      const char *encoding_errors):
+    cdef:
+        int na_count = 0
+        Py_ssize_t i, lines
+        coliter_t it
+        const char *word = NULL
+        ndarray[object] result
+
+        int ret = 0
+        kh_strbox_t *table
+
+        object pyval
+
+        object NA = na_values[np.object_]
+        khiter_t k
+
+    table = kh_init_strbox()
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.object_)
+    coliter_setup(&it, parser, col, line_start)
+
+    for i in range(lines):
+        COLITER_NEXT(it, word)
+
+        if na_filter:
+            if kh_get_str_starts_item(na_hashset, word):
+                # in the hash table
+                na_count += 1
+                result[i] = NA
+                continue
+
+        k = kh_get_strbox(table, word)
+
+        # in the hash table
+        if k != table.n_buckets:
+            # this increments the refcount, but need to test
+            pyval = <object>table.vals[k]
+        else:
+            # box it. new ref?
+            pyval = PyUnicode_Decode(word, strlen(word), "utf-8", encoding_errors)
+
+            k = kh_put_strbox(table, word, &ret)
+            table.vals[k] = <PyObject *>pyval
+
+        result[i] = pyval
+
+    kh_destroy_strbox(table)
+
+    return result, na_count
+
+
+@cython.boundscheck(False)
+cdef _categorical_convert(parser_t *parser, int64_t col,
+                          int64_t line_start, int64_t line_end,
+                          bint na_filter, kh_str_starts_t *na_hashset):
+    "Convert column data into codes, categories"
+    cdef:
+        int na_count = 0
+        Py_ssize_t i, lines
+        coliter_t it
+        const char *word = NULL
+
+        int64_t NA = -1
+        int64_t[::1] codes
+        int64_t current_category = 0
+
+        int ret = 0
+        kh_str_t *table
+        khiter_t k
+
+    lines = line_end - line_start
+    codes = np.empty(lines, dtype=np.int64)
+
+    # factorize parsed values, creating a hash table
+    # bytes -> category code
+    with nogil:
+        table = kh_init_str()
+        coliter_setup(&it, parser, col, line_start)
+
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+
+            if na_filter:
+                if kh_get_str_starts_item(na_hashset, word):
+                    # is in NA values
+                    na_count += 1
+                    codes[i] = NA
+                    continue
+
+            k = kh_get_str(table, word)
+            # not in the hash table
+            if k == table.n_buckets:
+                k = kh_put_str(table, word, &ret)
+                table.vals[k] = current_category
+                current_category += 1
+
+            codes[i] = table.vals[k]
+
+    # parse and box categories to python strings
+    result = np.empty(table.n_occupied, dtype=np.object_)
+    for k in range(table.n_buckets):
+        if kh_exist_str(table, k):
+            result[table.vals[k]] = PyUnicode_FromString(table.keys[k])
+
+    kh_destroy_str(table)
+    return np.asarray(codes), result, na_count
+
+
+# -> ndarray[f'|S{width}']
+cdef _to_fw_string(parser_t *parser, int64_t col, int64_t line_start,
+                   int64_t line_end, int64_t width):
+    cdef:
+        char *data
+        ndarray result
+
+    result = np.empty(line_end - line_start, dtype=f"|S{width}")
+    data = <char*>result.data
+
+    with nogil:
+        _to_fw_string_nogil(parser, col, line_start, line_end, width, data)
+
+    return result
+
+
+cdef void _to_fw_string_nogil(parser_t *parser, int64_t col,
+                              int64_t line_start, int64_t line_end,
+                              size_t width, char *data) noexcept nogil:
+    cdef:
+        int64_t i
+        coliter_t it
+        const char *word = NULL
+
+    coliter_setup(&it, parser, col, line_start)
+
+    for i in range(line_end - line_start):
+        COLITER_NEXT(it, word)
+        strncpy(data, word, width)
+        data += width
+
+
+cdef:
+    char* cinf = b"inf"
+    char* cposinf = b"+inf"
+    char* cneginf = b"-inf"
+
+    char* cinfty = b"Infinity"
+    char* cposinfty = b"+Infinity"
+    char* cneginfty = b"-Infinity"
+
+
+# -> tuple[ndarray[float64_t], int]  | tuple[None, None]
+cdef _try_double(parser_t *parser, int64_t col,
+                 int64_t line_start, int64_t line_end,
+                 bint na_filter, kh_str_starts_t *na_hashset, object na_flist):
+    cdef:
+        int error, na_count = 0
+        Py_ssize_t lines
+        float64_t *data
+        float64_t NA = na_values[np.float64]
+        kh_float64_t *na_fset
+        ndarray[float64_t] result
+        bint use_na_flist = len(na_flist) > 0
+
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.float64)
+    data = <float64_t *>result.data
+    na_fset = kset_float64_from_list(na_flist)
+    with nogil:
+        error = _try_double_nogil(parser, parser.double_converter,
+                                  col, line_start, line_end,
+                                  na_filter, na_hashset, use_na_flist,
+                                  na_fset, NA, data, &na_count)
+
+    kh_destroy_float64(na_fset)
+    if error != 0:
+        return None, None
+    return result, na_count
+
+
+cdef int _try_double_nogil(parser_t *parser,
+                           float64_t (*double_converter)(
+                               const char *, char **, char,
+                               char, char, int, int *, int *) noexcept nogil,
+                           int64_t col, int64_t line_start, int64_t line_end,
+                           bint na_filter, kh_str_starts_t *na_hashset,
+                           bint use_na_flist,
+                           const kh_float64_t *na_flist,
+                           float64_t NA, float64_t *data,
+                           int *na_count) nogil:
+    cdef:
+        int error = 0,
+        Py_ssize_t i, lines = line_end - line_start
+        coliter_t it
+        const char *word = NULL
+        char *p_end
+        khiter_t k64
+
+    na_count[0] = 0
+    coliter_setup(&it, parser, col, line_start)
+
+    if na_filter:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+
+            if kh_get_str_starts_item(na_hashset, word):
+                # in the hash table
+                na_count[0] += 1
+                data[0] = NA
+            else:
+                data[0] = double_converter(word, &p_end, parser.decimal,
+                                           parser.sci, parser.thousands,
+                                           1, &error, NULL)
+                if error != 0 or p_end == word or p_end[0]:
+                    error = 0
+                    if (strcasecmp(word, cinf) == 0 or
+                            strcasecmp(word, cposinf) == 0 or
+                            strcasecmp(word, cinfty) == 0 or
+                            strcasecmp(word, cposinfty) == 0):
+                        data[0] = INF
+                    elif (strcasecmp(word, cneginf) == 0 or
+                            strcasecmp(word, cneginfty) == 0):
+                        data[0] = NEGINF
+                    else:
+                        return 1
+                if use_na_flist:
+                    k64 = kh_get_float64(na_flist, data[0])
+                    if k64 != na_flist.n_buckets:
+                        na_count[0] += 1
+                        data[0] = NA
+            data += 1
+    else:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+            data[0] = double_converter(word, &p_end, parser.decimal,
+                                       parser.sci, parser.thousands,
+                                       1, &error, NULL)
+            if error != 0 or p_end == word or p_end[0]:
+                error = 0
+                if (strcasecmp(word, cinf) == 0 or
+                        strcasecmp(word, cposinf) == 0 or
+                        strcasecmp(word, cinfty) == 0 or
+                        strcasecmp(word, cposinfty) == 0):
+                    data[0] = INF
+                elif (strcasecmp(word, cneginf) == 0 or
+                        strcasecmp(word, cneginfty) == 0):
+                    data[0] = NEGINF
+                else:
+                    return 1
+            data += 1
+
+    return 0
+
+
+cdef _try_uint64(parser_t *parser, int64_t col,
+                 int64_t line_start, int64_t line_end,
+                 bint na_filter, kh_str_starts_t *na_hashset):
+    cdef:
+        int error
+        Py_ssize_t lines
+        coliter_t it
+        uint64_t *data
+        ndarray result
+        uint_state state
+
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.uint64)
+    data = <uint64_t *>result.data
+
+    uint_state_init(&state)
+    coliter_setup(&it, parser, col, line_start)
+    with nogil:
+        error = _try_uint64_nogil(parser, col, line_start, line_end,
+                                  na_filter, na_hashset, data, &state)
+    if error != 0:
+        if error == ERROR_OVERFLOW:
+            # Can't get the word variable
+            raise OverflowError("Overflow")
+        return None
+
+    if uint64_conflict(&state):
+        raise ValueError("Cannot convert to numerical dtype")
+
+    if state.seen_sint:
+        raise OverflowError("Overflow")
+
+    return result
+
+
+cdef int _try_uint64_nogil(parser_t *parser, int64_t col,
+                           int64_t line_start,
+                           int64_t line_end, bint na_filter,
+                           const kh_str_starts_t *na_hashset,
+                           uint64_t *data, uint_state *state) nogil:
+    cdef:
+        int error
+        Py_ssize_t i, lines = line_end - line_start
+        coliter_t it
+        const char *word = NULL
+
+    coliter_setup(&it, parser, col, line_start)
+
+    if na_filter:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+            if kh_get_str_starts_item(na_hashset, word):
+                # in the hash table
+                state.seen_null = 1
+                data[i] = 0
+                continue
+
+            data[i] = str_to_uint64(state, word, INT64_MAX, UINT64_MAX,
+                                    &error, parser.thousands)
+            if error != 0:
+                return error
+    else:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+            data[i] = str_to_uint64(state, word, INT64_MAX, UINT64_MAX,
+                                    &error, parser.thousands)
+            if error != 0:
+                return error
+
+    return 0
+
+
+cdef _try_int64(parser_t *parser, int64_t col,
+                int64_t line_start, int64_t line_end,
+                bint na_filter, kh_str_starts_t *na_hashset):
+    cdef:
+        int error, na_count = 0
+        Py_ssize_t lines
+        coliter_t it
+        int64_t *data
+        ndarray result
+        int64_t NA = na_values[np.int64]
+
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.int64)
+    data = <int64_t *>result.data
+    coliter_setup(&it, parser, col, line_start)
+    with nogil:
+        error = _try_int64_nogil(parser, col, line_start, line_end,
+                                 na_filter, na_hashset, NA, data, &na_count)
+    if error != 0:
+        if error == ERROR_OVERFLOW:
+            # Can't get the word variable
+            raise OverflowError("Overflow")
+        return None, None
+
+    return result, na_count
+
+
+cdef int _try_int64_nogil(parser_t *parser, int64_t col,
+                          int64_t line_start,
+                          int64_t line_end, bint na_filter,
+                          const kh_str_starts_t *na_hashset, int64_t NA,
+                          int64_t *data, int *na_count) nogil:
+    cdef:
+        int error
+        Py_ssize_t i, lines = line_end - line_start
+        coliter_t it
+        const char *word = NULL
+
+    na_count[0] = 0
+    coliter_setup(&it, parser, col, line_start)
+
+    if na_filter:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+            if kh_get_str_starts_item(na_hashset, word):
+                # in the hash table
+                na_count[0] += 1
+                data[i] = NA
+                continue
+
+            data[i] = str_to_int64(word, INT64_MIN, INT64_MAX,
+                                   &error, parser.thousands)
+            if error != 0:
+                return error
+    else:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+            data[i] = str_to_int64(word, INT64_MIN, INT64_MAX,
+                                   &error, parser.thousands)
+            if error != 0:
+                return error
+
+    return 0
+
+
+# -> tuple[ndarray[bool], int]
+cdef _try_bool_flex(parser_t *parser, int64_t col,
+                    int64_t line_start, int64_t line_end,
+                    bint na_filter, const kh_str_starts_t *na_hashset,
+                    const kh_str_starts_t *true_hashset,
+                    const kh_str_starts_t *false_hashset):
+    cdef:
+        int error, na_count = 0
+        Py_ssize_t lines
+        uint8_t *data
+        ndarray result
+        uint8_t NA = na_values[np.bool_]
+
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.uint8)
+    data = <uint8_t *>result.data
+    with nogil:
+        error = _try_bool_flex_nogil(parser, col, line_start, line_end,
+                                     na_filter, na_hashset, true_hashset,
+                                     false_hashset, NA, data, &na_count)
+    if error != 0:
+        return None, None
+    return result.view(np.bool_), na_count
+
+
+cdef int _try_bool_flex_nogil(parser_t *parser, int64_t col,
+                              int64_t line_start,
+                              int64_t line_end, bint na_filter,
+                              const kh_str_starts_t *na_hashset,
+                              const kh_str_starts_t *true_hashset,
+                              const kh_str_starts_t *false_hashset,
+                              uint8_t NA, uint8_t *data,
+                              int *na_count) nogil:
+    cdef:
+        int error = 0
+        Py_ssize_t i, lines = line_end - line_start
+        coliter_t it
+        const char *word = NULL
+
+    na_count[0] = 0
+    coliter_setup(&it, parser, col, line_start)
+
+    if na_filter:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+
+            if kh_get_str_starts_item(na_hashset, word):
+                # in the hash table
+                na_count[0] += 1
+                data[0] = NA
+                data += 1
+                continue
+
+            if kh_get_str_starts_item(true_hashset, word):
+                data[0] = 1
+                data += 1
+                continue
+            if kh_get_str_starts_item(false_hashset, word):
+                data[0] = 0
+                data += 1
+                continue
+
+            error = to_boolean(word, data)
+            if error != 0:
+                return error
+            data += 1
+    else:
+        for i in range(lines):
+            COLITER_NEXT(it, word)
+
+            if kh_get_str_starts_item(true_hashset, word):
+                data[0] = 1
+                data += 1
+                continue
+
+            if kh_get_str_starts_item(false_hashset, word):
+                data[0] = 0
+                data += 1
+                continue
+
+            error = to_boolean(word, data)
+            if error != 0:
+                return error
+            data += 1
+
+    return 0
+
+
+cdef kh_str_starts_t* kset_from_list(list values) except NULL:
+    # caller takes responsibility for freeing the hash table
+    cdef:
+        Py_ssize_t i
+        kh_str_starts_t *table
+        int ret = 0
+        object val
+
+    table = kh_init_str_starts()
+
+    for i in range(len(values)):
+        val = values[i]
+
+        # None creeps in sometimes, which isn't possible here
+        if not isinstance(val, bytes):
+            kh_destroy_str_starts(table)
+            raise ValueError("Must be all encoded bytes")
+
+        kh_put_str_starts_item(table, PyBytes_AsString(val), &ret)
+
+    if table.table.n_buckets <= 128:
+        # Resize the hash table to make it almost empty, this
+        # reduces amount of hash collisions on lookup thus
+        # "key not in table" case is faster.
+        # Note that this trades table memory footprint for lookup speed.
+        kh_resize_str_starts(table, table.table.n_buckets * 8)
+
+    return table
+
+
+cdef kh_float64_t* kset_float64_from_list(values) except NULL:
+    # caller takes responsibility for freeing the hash table
+    cdef:
+        kh_float64_t *table
+        int ret = 0
+        float64_t val
+        object value
+
+    table = kh_init_float64()
+
+    for value in values:
+        val = float(value)
+
+        kh_put_float64(table, val, &ret)
+
+    if table.n_buckets <= 128:
+        # See reasoning in kset_from_list
+        kh_resize_float64(table, table.n_buckets * 8)
+    return table
+
+
+cdef raise_parser_error(object base, parser_t *parser):
+    cdef:
+        object old_exc
+        object exc_type
+        PyObject *type
+        PyObject *value
+        PyObject *traceback
+
+    if PyErr_Occurred():
+        PyErr_Fetch(&type, &value, &traceback)
+        Py_XDECREF(traceback)
+
+        if value != NULL:
+            old_exc = <object>value
+            Py_XDECREF(value)
+
+            # PyErr_Fetch only returned the error message in *value,
+            # so the Exception class must be extracted from *type.
+            if isinstance(old_exc, str):
+                if type != NULL:
+                    exc_type = <object>type
+                else:
+                    exc_type = ParserError
+
+                Py_XDECREF(type)
+                raise exc_type(old_exc)
+            else:
+                Py_XDECREF(type)
+                raise old_exc
+
+    message = f"{base}. C error: "
+    if parser.error_msg != NULL:
+        message += parser.error_msg.decode("utf-8")
+    else:
+        message += "no error message set"
+
+    raise ParserError(message)
+
+
+# ----------------------------------------------------------------------
+# NA values
+def _compute_na_values():
+    int64info = np.iinfo(np.int64)
+    int32info = np.iinfo(np.int32)
+    int16info = np.iinfo(np.int16)
+    int8info = np.iinfo(np.int8)
+    uint64info = np.iinfo(np.uint64)
+    uint32info = np.iinfo(np.uint32)
+    uint16info = np.iinfo(np.uint16)
+    uint8info = np.iinfo(np.uint8)
+    na_values = {
+        np.float32: np.nan,
+        np.float64: np.nan,
+        np.int64: int64info.min,
+        np.int32: int32info.min,
+        np.int16: int16info.min,
+        np.int8: int8info.min,
+        np.uint64: uint64info.max,
+        np.uint32: uint32info.max,
+        np.uint16: uint16info.max,
+        np.uint8: uint8info.max,
+        np.bool_: uint8info.max,
+        np.object_: np.nan,
+    }
+    return na_values
+
+
+na_values = _compute_na_values()
+
+for k in list(na_values):
+    na_values[np.dtype(k)] = na_values[k]
+
+
+# -> ArrayLike
+cdef _apply_converter(object f, parser_t *parser, int64_t col,
+                      int64_t line_start, int64_t line_end):
+    cdef:
+        Py_ssize_t i, lines
+        coliter_t it
+        const char *word = NULL
+        ndarray[object] result
+        object val
+
+    lines = line_end - line_start
+    result = np.empty(lines, dtype=np.object_)
+
+    coliter_setup(&it, parser, col, line_start)
+
+    for i in range(lines):
+        COLITER_NEXT(it, word)
+        val = PyUnicode_FromString(word)
+        result[i] = f(val)
+
+    return lib.maybe_convert_objects(result)
+
+
+cdef list _maybe_encode(list values):
+    if values is None:
+        return []
+    return [x.encode("utf-8") if isinstance(x, str) else x for x in values]
+
+
+def sanitize_objects(ndarray[object] values, set na_values) -> int:
+    """
+    Convert specified values, including the given set na_values to np.nan.
+
+    Parameters
+    ----------
+    values : ndarray[object]
+    na_values : set
+
+    Returns
+    -------
+    na_count : int
+    """
+    cdef:
+        Py_ssize_t i, n
+        object val, onan
+        Py_ssize_t na_count = 0
+        dict memo = {}
+
+    n = len(values)
+    onan = np.nan
+
+    for i in range(n):
+        val = values[i]
+        if val in na_values:
+            values[i] = onan
+            na_count += 1
+        elif val in memo:
+            values[i] = memo[val]
+        else:
+            memo[val] = val
+
+    return na_count
diff --git a/pandas/pandas/_libs/properties.pyi b/pandas/pandas/_libs/properties.pyi
new file mode 100644
index 0000000000000..aaa44a0cf47bf
--- /dev/null
+++ b/pandas/pandas/_libs/properties.pyi
@@ -0,0 +1,27 @@
+from typing import (
+    Sequence,
+    overload,
+)
+
+from pandas._typing import (
+    AnyArrayLike,
+    DataFrame,
+    Index,
+    Series,
+)
+
+# note: this is a lie to make type checkers happy (they special
+# case property). cache_readonly uses attribute names similar to
+# property (fget) but it does not provide fset and fdel.
+cache_readonly = property
+
+class AxisProperty:
+    axis: int
+    def __init__(self, axis: int = ..., doc: str = ...) -> None: ...
+    @overload
+    def __get__(self, obj: DataFrame | Series, type) -> Index: ...
+    @overload
+    def __get__(self, obj: None, type) -> AxisProperty: ...
+    def __set__(
+        self, obj: DataFrame | Series, value: AnyArrayLike | Sequence
+    ) -> None: ...
diff --git a/pandas/pandas/_libs/properties.pyx b/pandas/pandas/_libs/properties.pyx
new file mode 100644
index 0000000000000..33cd2ef27a995
--- /dev/null
+++ b/pandas/pandas/_libs/properties.pyx
@@ -0,0 +1,69 @@
+from cpython.dict cimport (
+    PyDict_Contains,
+    PyDict_GetItem,
+    PyDict_SetItem,
+)
+from cython cimport Py_ssize_t
+
+
+cdef class CachedProperty:
+
+    cdef readonly:
+        object fget, name, __doc__
+
+    def __init__(self, fget):
+        self.fget = fget
+        self.name = fget.__name__
+        self.__doc__ = getattr(fget, "__doc__", None)
+
+    def __get__(self, obj, typ):
+        if obj is None:
+            # accessed on the class, not the instance
+            return self
+
+        # Get the cache or set a default one if needed
+        cache = getattr(obj, "_cache", None)
+        if cache is None:
+            try:
+                cache = obj._cache = {}
+            except (AttributeError):
+                return self
+
+        if PyDict_Contains(cache, self.name):
+            # not necessary to Py_INCREF
+            val = <object>PyDict_GetItem(cache, self.name)
+        else:
+            val = self.fget(obj)
+            PyDict_SetItem(cache, self.name, val)
+        return val
+
+    def __set__(self, obj, value):
+        raise AttributeError("Can't set attribute")
+
+
+cache_readonly = CachedProperty
+
+
+cdef class AxisProperty:
+
+    cdef readonly:
+        Py_ssize_t axis
+        object __doc__
+
+    def __init__(self, axis=0, doc=""):
+        self.axis = axis
+        self.__doc__ = doc
+
+    def __get__(self, obj, type):
+        cdef:
+            list axes
+
+        if obj is None:
+            # Only instances have _mgr, not classes
+            return self
+        else:
+            axes = obj._mgr.axes
+        return axes[self.axis]
+
+    def __set__(self, obj, value):
+        obj._set_axis(self.axis, value)
diff --git a/pandas/pandas/_libs/reshape.pyi b/pandas/pandas/_libs/reshape.pyi
new file mode 100644
index 0000000000000..110687fcd0c31
--- /dev/null
+++ b/pandas/pandas/_libs/reshape.pyi
@@ -0,0 +1,16 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def unstack(
+    values: np.ndarray,  # reshape_t[:, :]
+    mask: np.ndarray,  # const uint8_t[:]
+    stride: int,
+    length: int,
+    width: int,
+    new_values: np.ndarray,  # reshape_t[:, :]
+    new_mask: np.ndarray,  # uint8_t[:, :]
+) -> None: ...
+def explode(
+    values: npt.NDArray[np.object_],
+) -> tuple[npt.NDArray[np.object_], npt.NDArray[np.int64]]: ...
diff --git a/pandas/pandas/_libs/reshape.pyx b/pandas/pandas/_libs/reshape.pyx
new file mode 100644
index 0000000000000..51c1c75ba631b
--- /dev/null
+++ b/pandas/pandas/_libs/reshape.pyx
@@ -0,0 +1,119 @@
+cimport cython
+from cython cimport Py_ssize_t
+from numpy cimport (
+    int64_t,
+    ndarray,
+    uint8_t,
+)
+
+import numpy as np
+
+cimport numpy as cnp
+from libc.math cimport NAN
+
+cnp.import_array()
+
+from pandas._libs.dtypes cimport numeric_object_t
+from pandas._libs.lib cimport c_is_list_like
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def unstack(const numeric_object_t[:, :] values, const uint8_t[:] mask,
+            Py_ssize_t stride, Py_ssize_t length, Py_ssize_t width,
+            numeric_object_t[:, :] new_values, uint8_t[:, :] new_mask) -> None:
+    """
+    Transform long values to wide new_values.
+
+    Parameters
+    ----------
+    values : typed ndarray
+    mask : np.ndarray[bool]
+    stride : int
+    length : int
+    width : int
+    new_values : np.ndarray[bool]
+        result array
+    new_mask : np.ndarray[bool]
+        result mask
+    """
+    cdef:
+        Py_ssize_t i, j, w, nulls, s, offset
+
+    with nogil(numeric_object_t is not object):
+        for i in range(stride):
+
+            nulls = 0
+            for j in range(length):
+
+                for w in range(width):
+
+                    offset = j * width + w
+
+                    if mask[offset]:
+                        s = i * width + w
+                        new_values[j, s] = values[offset - nulls, i]
+                        new_mask[j, s] = 1
+                    else:
+                        nulls += 1
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def explode(object[:] values):
+    """
+    transform array list-likes to long form
+    preserve non-list entries
+
+    Parameters
+    ----------
+    values : ndarray[object]
+
+    Returns
+    -------
+    ndarray[object]
+        result
+    ndarray[int64_t]
+        counts
+    """
+    cdef:
+        Py_ssize_t i, j, count, n
+        object v
+        ndarray[object] result
+        ndarray[int64_t] counts
+
+    # find the resulting len
+    n = len(values)
+    counts = np.zeros(n, dtype="int64")
+    for i in range(n):
+        v = values[i]
+
+        if c_is_list_like(v, True):
+            if len(v):
+                counts[i] += len(v)
+            else:
+                # empty list-like, use a nan marker
+                counts[i] += 1
+        else:
+            counts[i] += 1
+
+    result = np.empty(counts.sum(), dtype="object")
+    count = 0
+    for i in range(n):
+        v = values[i]
+
+        if c_is_list_like(v, True):
+            if len(v):
+                v = list(v)
+                for j in range(len(v)):
+                    result[count] = v[j]
+                    count += 1
+            else:
+                # empty list-like, use a nan marker
+                result[count] = NAN
+                count += 1
+        else:
+            # replace with the existing scalar
+            result[count] = v
+            count += 1
+    return result, counts
diff --git a/pandas/pandas/_libs/sas.pyi b/pandas/pandas/_libs/sas.pyi
new file mode 100644
index 0000000000000..5d65e2b56b591
--- /dev/null
+++ b/pandas/pandas/_libs/sas.pyi
@@ -0,0 +1,7 @@
+from pandas.io.sas.sas7bdat import SAS7BDATReader
+
+class Parser:
+    def __init__(self, parser: SAS7BDATReader) -> None: ...
+    def read(self, nrows: int) -> None: ...
+
+def get_subheader_index(signature: bytes) -> int: ...
diff --git a/pandas/pandas/_libs/sas.pyx b/pandas/pandas/_libs/sas.pyx
new file mode 100644
index 0000000000000..209e82c6284f5
--- /dev/null
+++ b/pandas/pandas/_libs/sas.pyx
@@ -0,0 +1,549 @@
+# cython: language_level=3, initializedcheck=False
+# cython: warn.maybe_uninitialized=True, warn.unused=True
+from cython cimport Py_ssize_t
+from libc.stddef cimport size_t
+from libc.stdint cimport (
+    int64_t,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+from libc.stdlib cimport (
+    calloc,
+    free,
+)
+
+import numpy as np
+
+import pandas.io.sas.sas_constants as const
+
+
+cdef object np_nan = np.nan
+
+
+cdef struct Buffer:
+    # Convenience wrapper for uint8_t data to allow fast and safe reads and writes.
+    # We use this as a replacement for np.array(..., dtype=np.uint8) because it's
+    # much slower to create NumPy arrays and we create Buffer instances many times
+    # when reading a SAS7BDAT file (roughly once per row that is being read).
+    uint8_t *data
+    size_t length
+
+
+cdef uint8_t buf_get(Buffer buf, size_t offset) except? 255:
+    assert offset < buf.length, "Out of bounds read"
+    return buf.data[offset]
+
+
+cdef bint buf_set(Buffer buf, size_t offset, uint8_t value) except 0:
+    assert offset < buf.length, "Out of bounds write"
+    buf.data[offset] = value
+    return True
+
+
+cdef bytes buf_as_bytes(Buffer buf, size_t offset, size_t length):
+    assert offset + length <= buf.length, "Out of bounds read"
+    return buf.data[offset:offset+length]
+
+
+cdef Buffer buf_new(size_t length) except *:
+    cdef uint8_t *data = <uint8_t *>calloc(length, sizeof(uint8_t))
+    if data is NULL:
+        raise MemoryError(f"Failed to allocate {length} bytes")
+    return Buffer(data, length)
+
+
+cdef buf_free(Buffer buf):
+    if buf.data != NULL:
+        free(buf.data)
+
+# rle_decompress decompresses data using a Run Length Encoding
+# algorithm.  It is partially documented here:
+#
+# https://cran.r-project.org/package=sas7bdat/vignettes/sas7bdat.pdf
+# Licence at LICENSES/SAS7BDAT_LICENSE
+cdef int rle_decompress(Buffer inbuff, Buffer outbuff) except? 0:
+
+    cdef:
+        uint8_t control_byte, x
+        int rpos = 0
+        int i, nbytes, end_of_first_byte
+        size_t ipos = 0
+        Py_ssize_t _
+
+    while ipos < inbuff.length:
+        control_byte = buf_get(inbuff, ipos) & 0xF0
+        end_of_first_byte = <int>(buf_get(inbuff, ipos) & 0x0F)
+        ipos += 1
+
+        if control_byte == 0x00:
+            nbytes = <int>(buf_get(inbuff, ipos)) + 64 + end_of_first_byte * 256
+            ipos += 1
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos))
+                rpos += 1
+                ipos += 1
+        elif control_byte == 0x40:
+            # not documented
+            nbytes = <int>(buf_get(inbuff, ipos)) + 18 + end_of_first_byte * 256
+            ipos += 1
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos))
+                rpos += 1
+            ipos += 1
+        elif control_byte == 0x60:
+            nbytes = end_of_first_byte * 256 + <int>(buf_get(inbuff, ipos)) + 17
+            ipos += 1
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, 0x20)
+                rpos += 1
+        elif control_byte == 0x70:
+            nbytes = end_of_first_byte * 256 + <int>(buf_get(inbuff, ipos)) + 17
+            ipos += 1
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, 0x00)
+                rpos += 1
+        elif control_byte == 0x80:
+            nbytes = end_of_first_byte + 1
+            for i in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos + i))
+                rpos += 1
+            ipos += nbytes
+        elif control_byte == 0x90:
+            nbytes = end_of_first_byte + 17
+            for i in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos + i))
+                rpos += 1
+            ipos += nbytes
+        elif control_byte == 0xA0:
+            nbytes = end_of_first_byte + 33
+            for i in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos + i))
+                rpos += 1
+            ipos += nbytes
+        elif control_byte == 0xB0:
+            nbytes = end_of_first_byte + 49
+            for i in range(nbytes):
+                buf_set(outbuff, rpos, buf_get(inbuff, ipos + i))
+                rpos += 1
+            ipos += nbytes
+        elif control_byte == 0xC0:
+            nbytes = end_of_first_byte + 3
+            x = buf_get(inbuff, ipos)
+            ipos += 1
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, x)
+                rpos += 1
+        elif control_byte == 0xD0:
+            nbytes = end_of_first_byte + 2
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, 0x40)
+                rpos += 1
+        elif control_byte == 0xE0:
+            nbytes = end_of_first_byte + 2
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, 0x20)
+                rpos += 1
+        elif control_byte == 0xF0:
+            nbytes = end_of_first_byte + 2
+            for _ in range(nbytes):
+                buf_set(outbuff, rpos, 0x00)
+                rpos += 1
+        else:
+            raise ValueError(f"unknown control byte: {control_byte}")
+
+    return rpos
+
+
+# rdc_decompress decompresses data using the Ross Data Compression algorithm:
+#
+# http://collaboration.cmc.ec.gc.ca/science/rpn/biblio/ddj/Website/articles/CUJ/1992/9210/ross/ross.htm
+cdef int rdc_decompress(Buffer inbuff, Buffer outbuff) except? 0:
+
+    cdef:
+        uint8_t cmd
+        uint16_t ctrl_bits = 0, ctrl_mask = 0, ofs, cnt
+        int rpos = 0, k, ii
+        size_t ipos = 0
+
+    ii = -1
+
+    while ipos < inbuff.length:
+        ii += 1
+        ctrl_mask = ctrl_mask >> 1
+        if ctrl_mask == 0:
+            ctrl_bits = ((<uint16_t>buf_get(inbuff, ipos) << 8) +
+                         <uint16_t>buf_get(inbuff, ipos + 1))
+            ipos += 2
+            ctrl_mask = 0x8000
+
+        if ctrl_bits & ctrl_mask == 0:
+            buf_set(outbuff, rpos, buf_get(inbuff, ipos))
+            ipos += 1
+            rpos += 1
+            continue
+
+        cmd = (buf_get(inbuff, ipos) >> 4) & 0x0F
+        cnt = <uint16_t>(buf_get(inbuff, ipos) & 0x0F)
+        ipos += 1
+
+        # short RLE
+        if cmd == 0:
+            cnt += 3
+            for k in range(cnt):
+                buf_set(outbuff, rpos + k, buf_get(inbuff, ipos))
+            rpos += cnt
+            ipos += 1
+
+        # long RLE
+        elif cmd == 1:
+            cnt += <uint16_t>buf_get(inbuff, ipos) << 4
+            cnt += 19
+            ipos += 1
+            for k in range(cnt):
+                buf_set(outbuff, rpos + k, buf_get(inbuff, ipos))
+            rpos += cnt
+            ipos += 1
+
+        # long pattern
+        elif cmd == 2:
+            ofs = cnt + 3
+            ofs += <uint16_t>buf_get(inbuff, ipos) << 4
+            ipos += 1
+            cnt = <uint16_t>buf_get(inbuff, ipos)
+            ipos += 1
+            cnt += 16
+            for k in range(cnt):
+                buf_set(outbuff, rpos + k, buf_get(outbuff, rpos - <int>ofs + k))
+            rpos += cnt
+
+        # short pattern
+        else:
+            ofs = cnt + 3
+            ofs += <uint16_t>buf_get(inbuff, ipos) << 4
+            ipos += 1
+            for k in range(cmd):
+                buf_set(outbuff, rpos + k, buf_get(outbuff, rpos - <int>ofs + k))
+            rpos += cmd
+
+    return rpos
+
+
+cdef enum ColumnTypes:
+    column_type_decimal = 1
+    column_type_string = 2
+
+
+# Const aliases
+assert len(const.page_meta_types) == 2
+cdef:
+    int page_meta_types_0 = const.page_meta_types[0]
+    int page_meta_types_1 = const.page_meta_types[1]
+    int page_mix_type = const.page_mix_type
+    int page_data_type = const.page_data_type
+    int subheader_pointers_offset = const.subheader_pointers_offset
+
+    # Copy of subheader_signature_to_index that allows for much faster lookups.
+    # Lookups are done in get_subheader_index. The C structures are initialized
+    # in _init_subheader_signatures().
+    uint32_t subheader_signatures_32bit[13]
+    int subheader_indices_32bit[13]
+    uint64_t subheader_signatures_64bit[17]
+    int subheader_indices_64bit[17]
+    int data_subheader_index = const.SASIndex.data_subheader_index
+
+
+def _init_subheader_signatures():
+    subheaders_32bit = [
+        (sig, idx)
+        for sig, idx in const.subheader_signature_to_index.items()
+        if len(sig) == 4
+    ]
+    subheaders_64bit = [
+        (sig, idx)
+        for sig, idx in const.subheader_signature_to_index.items()
+        if len(sig) == 8
+    ]
+    assert len(subheaders_32bit) == 13
+    assert len(subheaders_64bit) == 17
+    assert len(const.subheader_signature_to_index) == 13 + 17
+    for i, (signature, idx) in enumerate(subheaders_32bit):
+        subheader_signatures_32bit[i] = (<uint32_t *><char *>signature)[0]
+        subheader_indices_32bit[i] = idx
+    for i, (signature, idx) in enumerate(subheaders_64bit):
+        subheader_signatures_64bit[i] = (<uint64_t *><char *>signature)[0]
+        subheader_indices_64bit[i] = idx
+
+
+_init_subheader_signatures()
+
+
+def get_subheader_index(bytes signature):
+    """Fast version of 'subheader_signature_to_index.get(signature)'."""
+    cdef:
+        uint32_t sig32
+        uint64_t sig64
+        Py_ssize_t i
+    assert len(signature) in (4, 8)
+    if len(signature) == 4:
+        sig32 = (<uint32_t *><char *>signature)[0]
+        for i in range(len(subheader_signatures_32bit)):
+            if subheader_signatures_32bit[i] == sig32:
+                return subheader_indices_32bit[i]
+    else:
+        sig64 = (<uint64_t *><char *>signature)[0]
+        for i in range(len(subheader_signatures_64bit)):
+            if subheader_signatures_64bit[i] == sig64:
+                return subheader_indices_64bit[i]
+
+    return data_subheader_index
+
+
+cdef class Parser:
+
+    cdef:
+        int column_count
+        int64_t[:] lengths
+        int64_t[:] offsets
+        int64_t[:] column_types
+        uint8_t[:, :] byte_chunk
+        object[:, :] string_chunk
+        uint8_t *cached_page
+        int cached_page_len
+        int current_row_on_page_index
+        int current_page_block_count
+        int current_page_data_subheader_pointers_len
+        int current_page_subheaders_count
+        int current_row_in_chunk_index
+        int current_row_in_file_index
+        bint blank_missing
+        int header_length
+        int row_length
+        int bit_offset
+        int subheader_pointer_length
+        int current_page_type
+        bint is_little_endian
+        int (*decompress)(Buffer, Buffer) except? 0
+        object parser
+
+    def __init__(self, object parser):
+        cdef:
+            int j
+            char[:] column_types
+
+        self.parser = parser
+        self.blank_missing = parser.blank_missing
+        self.header_length = self.parser.header_length
+        self.column_count = parser.column_count
+        self.lengths = parser.column_data_lengths()
+        self.offsets = parser.column_data_offsets()
+        self.byte_chunk = parser._byte_chunk
+        self.string_chunk = parser._string_chunk
+        self.row_length = parser.row_length
+        self.bit_offset = self.parser._page_bit_offset
+        self.subheader_pointer_length = self.parser._subheader_pointer_length
+        self.is_little_endian = parser.byte_order == "<"
+        self.column_types = np.empty(self.column_count, dtype="int64")
+
+        # page indicators
+        self.update_next_page()
+
+        column_types = parser.column_types()
+
+        # map column types
+        for j in range(self.column_count):
+            if column_types[j] == b"d":
+                self.column_types[j] = column_type_decimal
+            elif column_types[j] == b"s":
+                self.column_types[j] = column_type_string
+            else:
+                raise ValueError(f"unknown column type: {self.parser.columns[j].ctype}")
+
+        # compression
+        if parser.compression == const.rle_compression:
+            self.decompress = rle_decompress
+        elif parser.compression == const.rdc_compression:
+            self.decompress = rdc_decompress
+        else:
+            self.decompress = NULL
+
+        # update to current state of the parser
+        self.current_row_in_chunk_index = parser._current_row_in_chunk_index
+        self.current_row_in_file_index = parser._current_row_in_file_index
+        self.current_row_on_page_index = parser._current_row_on_page_index
+
+    def read(self, int nrows):
+        cdef:
+            bint done
+            Py_ssize_t _
+
+        for _ in range(nrows):
+            done = self.readline()
+            if done:
+                break
+
+        # update the parser
+        self.parser._current_row_on_page_index = self.current_row_on_page_index
+        self.parser._current_row_in_chunk_index = self.current_row_in_chunk_index
+        self.parser._current_row_in_file_index = self.current_row_in_file_index
+
+    cdef bint read_next_page(self) except? True:
+        cdef bint done
+
+        done = self.parser._read_next_page()
+        if done:
+            self.cached_page = NULL
+        else:
+            self.update_next_page()
+        return done
+
+    cdef update_next_page(self):
+        # update data for the current page
+
+        self.cached_page = <uint8_t *>self.parser._cached_page
+        self.cached_page_len = len(self.parser._cached_page)
+        self.current_row_on_page_index = 0
+        self.current_page_type = self.parser._current_page_type
+        self.current_page_block_count = self.parser._current_page_block_count
+        self.current_page_data_subheader_pointers_len = len(
+            self.parser._current_page_data_subheader_pointers
+        )
+        self.current_page_subheaders_count = self.parser._current_page_subheaders_count
+
+    cdef bint readline(self) except? True:
+
+        cdef:
+            int offset, length, bit_offset, align_correction
+            int subheader_pointer_length, mn
+            bint done, flag
+
+        bit_offset = self.bit_offset
+        subheader_pointer_length = self.subheader_pointer_length
+
+        # If there is no page, go to the end of the header and read a page.
+        if self.cached_page == NULL:
+            self.parser._path_or_buf.seek(self.header_length)
+            done = self.read_next_page()
+            if done:
+                return True
+
+        # Loop until a data row is read
+        while True:
+            if self.current_page_type in (page_meta_types_0, page_meta_types_1):
+                flag = self.current_row_on_page_index >=\
+                    self.current_page_data_subheader_pointers_len
+                if flag:
+                    done = self.read_next_page()
+                    if done:
+                        return True
+                    continue
+                offset, length = self.parser._current_page_data_subheader_pointers[
+                    self.current_row_on_page_index
+                ]
+                self.process_byte_array_with_data(offset, length)
+                return False
+            elif self.current_page_type == page_mix_type:
+                align_correction = (
+                    bit_offset
+                    + subheader_pointers_offset
+                    + self.current_page_subheaders_count * subheader_pointer_length
+                )
+                align_correction = align_correction % 8
+                offset = bit_offset + align_correction
+                offset += subheader_pointers_offset
+                offset += self.current_page_subheaders_count * subheader_pointer_length
+                offset += self.current_row_on_page_index * self.row_length
+                self.process_byte_array_with_data(offset, self.row_length)
+                mn = min(self.parser.row_count, self.parser._mix_page_row_count)
+                if self.current_row_on_page_index == mn:
+                    done = self.read_next_page()
+                    if done:
+                        return True
+                return False
+            elif self.current_page_type == page_data_type:
+                self.process_byte_array_with_data(
+                    bit_offset
+                    + subheader_pointers_offset
+                    + self.current_row_on_page_index * self.row_length,
+                    self.row_length,
+                )
+                flag = self.current_row_on_page_index == self.current_page_block_count
+                if flag:
+                    done = self.read_next_page()
+                    if done:
+                        return True
+                return False
+            else:
+                raise ValueError(f"unknown page type: {self.current_page_type}")
+
+    cdef void process_byte_array_with_data(self, int offset, int length) except *:
+
+        cdef:
+            Py_ssize_t j
+            int s, k, m, jb, js, current_row, rpos
+            int64_t lngt, start, ct
+            Buffer source, decompressed_source
+            int64_t[:] column_types
+            int64_t[:] lengths
+            int64_t[:] offsets
+            uint8_t[:, :] byte_chunk
+            object[:, :] string_chunk
+            bint compressed
+
+        assert offset + length <= self.cached_page_len, "Out of bounds read"
+        source = Buffer(&self.cached_page[offset], length)
+
+        compressed = self.decompress != NULL and length < self.row_length
+        if compressed:
+            decompressed_source = buf_new(self.row_length)
+            rpos = self.decompress(source, decompressed_source)
+            if rpos != self.row_length:
+                raise ValueError(
+                    f"Expected decompressed line of length {self.row_length} bytes "
+                    f"but decompressed {rpos} bytes"
+                )
+            source = decompressed_source
+
+        current_row = self.current_row_in_chunk_index
+        column_types = self.column_types
+        lengths = self.lengths
+        offsets = self.offsets
+        byte_chunk = self.byte_chunk
+        string_chunk = self.string_chunk
+        s = 8 * self.current_row_in_chunk_index
+        js = 0
+        jb = 0
+        for j in range(self.column_count):
+            lngt = lengths[j]
+            if lngt == 0:
+                break
+            start = offsets[j]
+            ct = column_types[j]
+            if ct == column_type_decimal:
+                # decimal
+                if self.is_little_endian:
+                    m = s + 8 - lngt
+                else:
+                    m = s
+                for k in range(lngt):
+                    byte_chunk[jb, m + k] = buf_get(source, start + k)
+                jb += 1
+            elif column_types[j] == column_type_string:
+                # string
+                # Skip trailing whitespace. This is equivalent to calling
+                # .rstrip(b"\x00 ") but without Python call overhead.
+                while lngt > 0 and buf_get(source, start + lngt - 1) in b"\x00 ":
+                    lngt -= 1
+                if lngt == 0 and self.blank_missing:
+                    string_chunk[js, current_row] = np_nan
+                else:
+                    string_chunk[js, current_row] = buf_as_bytes(source, start, lngt)
+                js += 1
+
+        self.current_row_on_page_index += 1
+        self.current_row_in_chunk_index += 1
+        self.current_row_in_file_index += 1
+
+        if compressed:
+            buf_free(decompressed_source)
diff --git a/pandas/pandas/_libs/sparse.pyi b/pandas/pandas/_libs/sparse.pyi
new file mode 100644
index 0000000000000..536265b25425e
--- /dev/null
+++ b/pandas/pandas/_libs/sparse.pyi
@@ -0,0 +1,51 @@
+from typing import Sequence
+
+import numpy as np
+
+from pandas._typing import (
+    Self,
+    npt,
+)
+
+class SparseIndex:
+    length: int
+    npoints: int
+    def __init__(self) -> None: ...
+    @property
+    def ngaps(self) -> int: ...
+    @property
+    def nbytes(self) -> int: ...
+    @property
+    def indices(self) -> npt.NDArray[np.int32]: ...
+    def equals(self, other) -> bool: ...
+    def lookup(self, index: int) -> np.int32: ...
+    def lookup_array(self, indexer: npt.NDArray[np.int32]) -> npt.NDArray[np.int32]: ...
+    def to_int_index(self) -> IntIndex: ...
+    def to_block_index(self) -> BlockIndex: ...
+    def intersect(self, y_: SparseIndex) -> Self: ...
+    def make_union(self, y_: SparseIndex) -> Self: ...
+
+class IntIndex(SparseIndex):
+    indices: npt.NDArray[np.int32]
+    def __init__(
+        self, length: int, indices: Sequence[int], check_integrity: bool = ...
+    ) -> None: ...
+
+class BlockIndex(SparseIndex):
+    nblocks: int
+    blocs: np.ndarray
+    blengths: np.ndarray
+    def __init__(
+        self, length: int, blocs: np.ndarray, blengths: np.ndarray
+    ) -> None: ...
+
+    # Override to have correct parameters
+    def intersect(self, other: SparseIndex) -> Self: ...
+    def make_union(self, y: SparseIndex) -> Self: ...
+
+def make_mask_object_ndarray(
+    arr: npt.NDArray[np.object_], fill_value
+) -> npt.NDArray[np.bool_]: ...
+def get_blocks(
+    indices: npt.NDArray[np.int32],
+) -> tuple[npt.NDArray[np.int32], npt.NDArray[np.int32]]: ...
diff --git a/pandas/pandas/_libs/sparse.pyx b/pandas/pandas/_libs/sparse.pyx
new file mode 100644
index 0000000000000..b42a9415fbefb
--- /dev/null
+++ b/pandas/pandas/_libs/sparse.pyx
@@ -0,0 +1,732 @@
+cimport cython
+
+import numpy as np
+
+cimport numpy as cnp
+from libc.math cimport (
+    INFINITY as INF,
+    NAN as NaN,
+)
+from numpy cimport (
+    float64_t,
+    int8_t,
+    int32_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+)
+
+cnp.import_array()
+
+
+# -----------------------------------------------------------------------------
+
+
+cdef class SparseIndex:
+    """
+    Abstract superclass for sparse index types.
+    """
+
+    def __init__(self):
+        raise NotImplementedError
+
+
+cdef class IntIndex(SparseIndex):
+    """
+    Object for holding exact integer sparse indexing information
+
+    Parameters
+    ----------
+    length : integer
+    indices : array-like
+        Contains integers corresponding to the indices.
+    check_integrity : bool, default=True
+        Check integrity of the input.
+    """
+
+    cdef readonly:
+        Py_ssize_t length, npoints
+        ndarray indices
+
+    def __init__(self, Py_ssize_t length, indices, bint check_integrity=True):
+        self.length = length
+        self.indices = np.ascontiguousarray(indices, dtype=np.int32)
+        self.npoints = len(self.indices)
+
+        if check_integrity:
+            self.check_integrity()
+
+    def __reduce__(self):
+        args = (self.length, self.indices)
+        return IntIndex, args
+
+    def __repr__(self) -> str:
+        output = "IntIndex\n"
+        output += f"Indices: {repr(self.indices)}\n"
+        return output
+
+    @property
+    def nbytes(self) -> int:
+        return self.indices.nbytes
+
+    cdef check_integrity(self):
+        """
+        Checks the following:
+
+        - Indices are strictly ascending
+        - Number of indices is at most self.length
+        - Indices are at least 0 and at most the total length less one
+
+        A ValueError is raised if any of these conditions is violated.
+        """
+
+        if self.npoints > self.length:
+            raise ValueError(
+                f"Too many indices. Expected {self.length} but found {self.npoints}"
+            )
+
+        # Indices are vacuously ordered and non-negative
+        # if the sequence of indices is empty.
+        if self.npoints == 0:
+            return
+
+        if self.indices.min() < 0:
+            raise ValueError("No index can be less than zero")
+
+        if self.indices.max() >= self.length:
+            raise ValueError("All indices must be less than the length")
+
+        monotonic = np.all(self.indices[:-1] < self.indices[1:])
+        if not monotonic:
+            raise ValueError("Indices must be strictly increasing")
+
+    def equals(self, other: object) -> bool:
+        if not isinstance(other, IntIndex):
+            return False
+
+        if self is other:
+            return True
+
+        same_length = self.length == other.length
+        same_indices = np.array_equal(self.indices, other.indices)
+        return same_length and same_indices
+
+    @property
+    def ngaps(self) -> int:
+        return self.length - self.npoints
+
+    cpdef to_int_index(self):
+        return self
+
+    def to_block_index(self):
+        locs, lens = get_blocks(self.indices)
+        return BlockIndex(self.length, locs, lens)
+
+    cpdef IntIndex intersect(self, SparseIndex y_):
+        cdef:
+            Py_ssize_t xi, yi = 0, result_indexer = 0
+            int32_t xind
+            ndarray[int32_t, ndim=1] xindices, yindices, new_indices
+            IntIndex y
+
+        # if is one already, returns self
+        y = y_.to_int_index()
+
+        if self.length != y.length:
+            raise Exception("Indices must reference same underlying length")
+
+        xindices = self.indices
+        yindices = y.indices
+        new_indices = np.empty(min(
+            len(xindices), len(yindices)), dtype=np.int32)
+
+        for xi in range(self.npoints):
+            xind = xindices[xi]
+
+            while yi < y.npoints and yindices[yi] < xind:
+                yi += 1
+
+            if yi >= y.npoints:
+                break
+
+            # TODO: would a two-pass algorithm be faster?
+            if yindices[yi] == xind:
+                new_indices[result_indexer] = xind
+                result_indexer += 1
+
+        new_indices = new_indices[:result_indexer]
+        return IntIndex(self.length, new_indices)
+
+    cpdef IntIndex make_union(self, SparseIndex y_):
+
+        cdef:
+            ndarray[int32_t, ndim=1] new_indices
+            IntIndex y
+
+        # if is one already, returns self
+        y = y_.to_int_index()
+
+        if self.length != y.length:
+            raise ValueError("Indices must reference same underlying length")
+
+        new_indices = np.union1d(self.indices, y.indices)
+        return IntIndex(self.length, new_indices)
+
+    @cython.wraparound(False)
+    cpdef int32_t lookup(self, Py_ssize_t index):
+        """
+        Return the internal location if value exists on given index.
+        Return -1 otherwise.
+        """
+        cdef:
+            int32_t res
+            ndarray[int32_t, ndim=1] inds
+
+        inds = self.indices
+        if self.npoints == 0:
+            return -1
+        elif index < 0 or self.length <= index:
+            return -1
+
+        res = inds.searchsorted(index)
+        if res == self.npoints:
+            return -1
+        elif inds[res] == index:
+            return res
+        else:
+            return -1
+
+    @cython.wraparound(False)
+    cpdef ndarray[int32_t] lookup_array(self, ndarray[int32_t, ndim=1] indexer):
+        """
+        Vectorized lookup, returns ndarray[int32_t]
+        """
+        cdef:
+            Py_ssize_t n
+            ndarray[int32_t, ndim=1] inds
+            ndarray[uint8_t, ndim=1, cast=True] mask
+            ndarray[int32_t, ndim=1] masked
+            ndarray[int32_t, ndim=1] res
+            ndarray[int32_t, ndim=1] results
+
+        n = len(indexer)
+        results = np.empty(n, dtype=np.int32)
+        results[:] = -1
+
+        if self.npoints == 0:
+            return results
+
+        inds = self.indices
+        mask = (inds[0] <= indexer) & (indexer <= inds[len(inds) - 1])
+
+        masked = indexer[mask]
+        res = inds.searchsorted(masked).astype(np.int32)
+
+        res[inds[res] != masked] = -1
+        results[mask] = res
+        return results
+
+
+cpdef get_blocks(ndarray[int32_t, ndim=1] indices):
+    cdef:
+        Py_ssize_t i, npoints, result_indexer = 0
+        int32_t block, length = 1, cur, prev
+        ndarray[int32_t, ndim=1] locs, lens
+
+    npoints = len(indices)
+
+    # just handle the special empty case separately
+    if npoints == 0:
+        return np.array([], dtype=np.int32), np.array([], dtype=np.int32)
+
+    # block size can't be longer than npoints
+    locs = np.empty(npoints, dtype=np.int32)
+    lens = np.empty(npoints, dtype=np.int32)
+
+    # TODO: two-pass algorithm faster?
+    prev = block = indices[0]
+    for i in range(1, npoints):
+        cur = indices[i]
+        if cur - prev > 1:
+            # new block
+            locs[result_indexer] = block
+            lens[result_indexer] = length
+            block = cur
+            length = 1
+            result_indexer += 1
+        else:
+            # same block, increment length
+            length += 1
+
+        prev = cur
+
+    locs[result_indexer] = block
+    lens[result_indexer] = length
+    result_indexer += 1
+    locs = locs[:result_indexer]
+    lens = lens[:result_indexer]
+    return locs, lens
+
+
+# -----------------------------------------------------------------------------
+# BlockIndex
+
+cdef class BlockIndex(SparseIndex):
+    """
+    Object for holding block-based sparse indexing information
+
+    Parameters
+    ----------
+    """
+    cdef readonly:
+        int32_t nblocks, npoints, length
+        ndarray blocs, blengths
+
+    cdef:
+        object __weakref__  # need to be picklable
+        int32_t *locbuf
+        int32_t *lenbuf
+
+    def __init__(self, length, blocs, blengths):
+
+        self.blocs = np.ascontiguousarray(blocs, dtype=np.int32)
+        self.blengths = np.ascontiguousarray(blengths, dtype=np.int32)
+
+        # in case we need
+        self.locbuf = <int32_t*>self.blocs.data
+        self.lenbuf = <int32_t*>self.blengths.data
+
+        self.length = length
+        self.nblocks = np.int32(len(self.blocs))
+        self.npoints = self.blengths.sum()
+
+        self.check_integrity()
+
+    def __reduce__(self):
+        args = (self.length, self.blocs, self.blengths)
+        return BlockIndex, args
+
+    def __repr__(self) -> str:
+        output = "BlockIndex\n"
+        output += f"Block locations: {repr(self.blocs)}\n"
+        output += f"Block lengths: {repr(self.blengths)}"
+
+        return output
+
+    @property
+    def nbytes(self) -> int:
+        return self.blocs.nbytes + self.blengths.nbytes
+
+    @property
+    def ngaps(self) -> int:
+        return self.length - self.npoints
+
+    cdef check_integrity(self):
+        """
+        Check:
+        - Locations are in ascending order
+        - No overlapping blocks
+        - Blocks to not start after end of index, nor extend beyond end
+        """
+        cdef:
+            Py_ssize_t i
+            ndarray[int32_t, ndim=1] blocs, blengths
+
+        blocs = self.blocs
+        blengths = self.blengths
+
+        if len(blocs) != len(blengths):
+            raise ValueError("block bound arrays must be same length")
+
+        for i in range(self.nblocks):
+            if i > 0:
+                if blocs[i] <= blocs[i - 1]:
+                    raise ValueError("Locations not in ascending order")
+
+            if i < self.nblocks - 1:
+                if blocs[i] + blengths[i] > blocs[i + 1]:
+                    raise ValueError(f"Block {i} overlaps")
+            else:
+                if blocs[i] + blengths[i] > self.length:
+                    raise ValueError(f"Block {i} extends beyond end")
+
+            # no zero-length blocks
+            if blengths[i] == 0:
+                raise ValueError(f"Zero-length block {i}")
+
+    def equals(self, other: object) -> bool:
+        if not isinstance(other, BlockIndex):
+            return False
+
+        if self is other:
+            return True
+
+        same_length = self.length == other.length
+        same_blocks = (np.array_equal(self.blocs, other.blocs) and
+                       np.array_equal(self.blengths, other.blengths))
+        return same_length and same_blocks
+
+    def to_block_index(self):
+        return self
+
+    cpdef to_int_index(self):
+        cdef:
+            int32_t i = 0, j, b
+            int32_t offset
+            ndarray[int32_t, ndim=1] indices
+
+        indices = np.empty(self.npoints, dtype=np.int32)
+
+        for b in range(self.nblocks):
+            offset = self.locbuf[b]
+
+            for j in range(self.lenbuf[b]):
+                indices[i] = offset + j
+                i += 1
+
+        return IntIndex(self.length, indices)
+
+    @property
+    def indices(self):
+        return self.to_int_index().indices
+
+    cpdef BlockIndex intersect(self, SparseIndex other):
+        """
+        Intersect two BlockIndex objects
+
+        Returns
+        -------
+        BlockIndex
+        """
+        cdef:
+            BlockIndex y
+            ndarray[int32_t, ndim=1] xloc, xlen, yloc, ylen, out_bloc, out_blen
+            Py_ssize_t xi = 0, yi = 0, max_len, result_indexer = 0
+            int32_t cur_loc, cur_length, diff
+
+        y = other.to_block_index()
+
+        if self.length != y.length:
+            raise Exception("Indices must reference same underlying length")
+
+        xloc = self.blocs
+        xlen = self.blengths
+        yloc = y.blocs
+        ylen = y.blengths
+
+        # block may be split, but can't exceed original len / 2 + 1
+        max_len = min(self.length, y.length) // 2 + 1
+        out_bloc = np.empty(max_len, dtype=np.int32)
+        out_blen = np.empty(max_len, dtype=np.int32)
+
+        while True:
+            # we are done (or possibly never began)
+            if xi >= self.nblocks or yi >= y.nblocks:
+                break
+
+            # completely symmetric...would like to avoid code dup but oh well
+            if xloc[xi] >= yloc[yi]:
+                cur_loc = xloc[xi]
+                diff = xloc[xi] - yloc[yi]
+
+                if ylen[yi] <= diff:
+                    # have to skip this block
+                    yi += 1
+                    continue
+
+                if ylen[yi] - diff < xlen[xi]:
+                    # take end of y block, move onward
+                    cur_length = ylen[yi] - diff
+                    yi += 1
+                else:
+                    # take end of x block
+                    cur_length = xlen[xi]
+                    xi += 1
+
+            else:  # xloc[xi] < yloc[yi]
+                cur_loc = yloc[yi]
+                diff = yloc[yi] - xloc[xi]
+
+                if xlen[xi] <= diff:
+                    # have to skip this block
+                    xi += 1
+                    continue
+
+                if xlen[xi] - diff < ylen[yi]:
+                    # take end of x block, move onward
+                    cur_length = xlen[xi] - diff
+                    xi += 1
+                else:
+                    # take end of y block
+                    cur_length = ylen[yi]
+                    yi += 1
+
+            out_bloc[result_indexer] = cur_loc
+            out_blen[result_indexer] = cur_length
+            result_indexer += 1
+
+        out_bloc = out_bloc[:result_indexer]
+        out_blen = out_blen[:result_indexer]
+
+        return BlockIndex(self.length, out_bloc, out_blen)
+
+    cpdef BlockIndex make_union(self, SparseIndex y):
+        """
+        Combine together two BlockIndex objects, accepting indices if contained
+        in one or the other
+
+        Parameters
+        ----------
+        other : SparseIndex
+
+        Notes
+        -----
+        union is a protected keyword in Cython, hence make_union
+
+        Returns
+        -------
+        BlockIndex
+        """
+        return BlockUnion(self, y.to_block_index()).result
+
+    cpdef Py_ssize_t lookup(self, Py_ssize_t index):
+        """
+        Return the internal location if value exists on given index.
+        Return -1 otherwise.
+        """
+        cdef:
+            Py_ssize_t i, cum_len
+            ndarray[int32_t, ndim=1] locs, lens
+
+        locs = self.blocs
+        lens = self.blengths
+
+        if self.nblocks == 0:
+            return -1
+        elif index < locs[0]:
+            return -1
+
+        cum_len = 0
+        for i in range(self.nblocks):
+            if index >= locs[i] and index < locs[i] + lens[i]:
+                return cum_len + index - locs[i]
+            cum_len += lens[i]
+
+        return -1
+
+    @cython.wraparound(False)
+    cpdef ndarray[int32_t] lookup_array(self, ndarray[int32_t, ndim=1] indexer):
+        """
+        Vectorized lookup, returns ndarray[int32_t]
+        """
+        cdef:
+            Py_ssize_t n, i, j, ind_val
+            ndarray[int32_t, ndim=1] locs, lens
+            ndarray[int32_t, ndim=1] results
+
+        locs = self.blocs
+        lens = self.blengths
+
+        n = len(indexer)
+        results = np.empty(n, dtype=np.int32)
+        results[:] = -1
+
+        if self.npoints == 0:
+            return results
+
+        for i in range(n):
+            ind_val = indexer[i]
+            if not (ind_val < 0 or self.length <= ind_val):
+                cum_len = 0
+                for j in range(self.nblocks):
+                    if ind_val >= locs[j] and ind_val < locs[j] + lens[j]:
+                        results[i] = cum_len + ind_val - locs[j]
+                    cum_len += lens[j]
+        return results
+
+
+@cython.internal
+cdef class BlockMerge:
+    """
+    Object-oriented approach makes sharing state between recursive functions a
+    lot easier and reduces code duplication
+    """
+    cdef:
+        BlockIndex x, y, result
+        ndarray xstart, xlen, xend, ystart, ylen, yend
+        int32_t xi, yi  # block indices
+
+    def __init__(self, BlockIndex x, BlockIndex y):
+        self.x = x
+        self.y = y
+
+        if x.length != y.length:
+            raise Exception("Indices must reference same underlying length")
+
+        self.xstart = self.x.blocs
+        self.ystart = self.y.blocs
+
+        self.xend = self.x.blocs + self.x.blengths
+        self.yend = self.y.blocs + self.y.blengths
+
+        # self.xlen = self.x.blengths
+        # self.ylen = self.y.blengths
+
+        self.xi = 0
+        self.yi = 0
+
+        self.result = self._make_merged_blocks()
+
+    cdef _make_merged_blocks(self):
+        raise NotImplementedError
+
+    cdef _set_current_indices(self, int32_t xi, int32_t yi, bint mode):
+        if mode == 0:
+            self.xi = xi
+            self.yi = yi
+        else:
+            self.xi = yi
+            self.yi = xi
+
+
+@cython.internal
+cdef class BlockUnion(BlockMerge):
+    """
+    Object-oriented approach makes sharing state between recursive functions a
+    lot easier and reduces code duplication
+    """
+
+    cdef _make_merged_blocks(self):
+        cdef:
+            ndarray[int32_t, ndim=1] xstart, xend, ystart
+            ndarray[int32_t, ndim=1] yend, out_bloc, out_blen
+            int32_t nstart, nend
+            Py_ssize_t max_len, result_indexer = 0
+
+        xstart = self.xstart
+        xend = self.xend
+        ystart = self.ystart
+        yend = self.yend
+
+        max_len = min(self.x.length, self.y.length) // 2 + 1
+        out_bloc = np.empty(max_len, dtype=np.int32)
+        out_blen = np.empty(max_len, dtype=np.int32)
+
+        while True:
+            # we are done (or possibly never began)
+            if self.xi >= self.x.nblocks and self.yi >= self.y.nblocks:
+                break
+            elif self.yi >= self.y.nblocks:
+                # through with y, just pass through x blocks
+                nstart = xstart[self.xi]
+                nend = xend[self.xi]
+                self.xi += 1
+            elif self.xi >= self.x.nblocks:
+                # through with x, just pass through y blocks
+                nstart = ystart[self.yi]
+                nend = yend[self.yi]
+                self.yi += 1
+            else:
+                # find end of new block
+                if xstart[self.xi] < ystart[self.yi]:
+                    nstart = xstart[self.xi]
+                    nend = self._find_next_block_end(0)
+                else:
+                    nstart = ystart[self.yi]
+                    nend = self._find_next_block_end(1)
+
+            out_bloc[result_indexer] = nstart
+            out_blen[result_indexer] = nend - nstart
+            result_indexer += 1
+
+        out_bloc = out_bloc[:result_indexer]
+        out_blen = out_blen[:result_indexer]
+
+        return BlockIndex(self.x.length, out_bloc, out_blen)
+
+    cdef int32_t _find_next_block_end(self, bint mode) except -1:
+        """
+        Wow, this got complicated in a hurry
+
+        mode 0: block started in index x
+        mode 1: block started in index y
+        """
+        cdef:
+            ndarray[int32_t, ndim=1] xstart, xend, ystart, yend
+            int32_t xi, yi, ynblocks, nend
+
+        if mode != 0 and mode != 1:
+            raise Exception("Mode must be 0 or 1")
+
+        # so symmetric code will work
+        if mode == 0:
+            xstart = self.xstart
+            xend = self.xend
+            xi = self.xi
+
+            ystart = self.ystart
+            yend = self.yend
+            yi = self.yi
+            ynblocks = self.y.nblocks
+        else:
+            xstart = self.ystart
+            xend = self.yend
+            xi = self.yi
+
+            ystart = self.xstart
+            yend = self.xend
+            yi = self.xi
+            ynblocks = self.x.nblocks
+
+        nend = xend[xi]
+
+        # done with y?
+        if yi == ynblocks:
+            self._set_current_indices(xi + 1, yi, mode)
+            return nend
+        elif nend < ystart[yi]:
+            # block ends before y block
+            self._set_current_indices(xi + 1, yi, mode)
+            return nend
+        else:
+            while yi < ynblocks and nend > yend[yi]:
+                yi += 1
+
+            self._set_current_indices(xi + 1, yi, mode)
+
+            if yi == ynblocks:
+                return nend
+
+            if nend < ystart[yi]:
+                # we're done, return the block end
+                return nend
+            else:
+                # merge blocks, continue searching
+                # this also catches the case where blocks
+                return self._find_next_block_end(1 - mode)
+
+
+# -----------------------------------------------------------------------------
+# Sparse arithmetic
+
+include "sparse_op_helper.pxi"
+
+
+# -----------------------------------------------------------------------------
+# SparseArray mask create operations
+
+def make_mask_object_ndarray(ndarray[object, ndim=1] arr, object fill_value):
+    cdef:
+        object value
+        Py_ssize_t i
+        Py_ssize_t new_length = len(arr)
+        ndarray[int8_t, ndim=1] mask
+
+    mask = np.ones(new_length, dtype=np.int8)
+
+    for i in range(new_length):
+        value = arr[i]
+        if value == fill_value and type(value) is type(fill_value):
+            mask[i] = 0
+
+    return mask.view(dtype=bool)
diff --git a/pandas/pandas/_libs/sparse_op_helper.pxi.in b/pandas/pandas/_libs/sparse_op_helper.pxi.in
new file mode 100644
index 0000000000000..774a8c579f6ce
--- /dev/null
+++ b/pandas/pandas/_libs/sparse_op_helper.pxi.in
@@ -0,0 +1,313 @@
+"""
+Template for each `dtype` helper function for sparse ops
+
+WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
+"""
+
+# ----------------------------------------------------------------------
+# Sparse op
+# ----------------------------------------------------------------------
+
+ctypedef fused sparse_t:
+    float64_t
+    int64_t
+
+
+cdef float64_t __div__(sparse_t a, sparse_t b):
+    if b == 0:
+        if a > 0:
+            return INF
+        elif a < 0:
+            return -INF
+        else:
+            return NaN
+    else:
+        return float(a) / b
+
+
+cdef float64_t __truediv__(sparse_t a, sparse_t b):
+    return __div__(a, b)
+
+
+cdef sparse_t __mod__(sparse_t a, sparse_t b):
+    if b == 0:
+        if sparse_t is float64_t:
+            return NaN
+        else:
+            return 0
+    else:
+        return a % b
+
+
+cdef sparse_t __floordiv__(sparse_t a, sparse_t b):
+    if b == 0:
+        if sparse_t is float64_t:
+            # Match non-sparse Series behavior implemented in mask_zero_div_zero
+            if a > 0:
+                return INF
+            elif a < 0:
+                return -INF
+            return NaN
+        else:
+            return 0
+    else:
+        return a // b
+
+
+# ----------------------------------------------------------------------
+# sparse array op
+# ----------------------------------------------------------------------
+
+{{py:
+
+# dtype, arith_comp_group, logical_group
+dtypes = [('float64', True, False),
+          ('int64', True, True),
+          ('uint8', False, True)]
+# do not generate arithmetic / comparison template for uint8,
+# it should be done in fused types
+
+def get_op(tup):
+    assert isinstance(tup, tuple)
+    assert len(tup) == 4
+
+    opname, lval, rval, dtype = tup
+
+    ops_dict = {'add': '{0} + {1}',
+                'sub': '{0} - {1}',
+                'mul': '{0} * {1}',
+                'div': '__div__({0}, {1})',
+                'mod': '__mod__({0}, {1})',
+                'truediv': '__truediv__({0}, {1})',
+                'floordiv': '__floordiv__({0}, {1})',
+                'pow': '{0} ** {1}',
+                'eq': '{0} == {1}',
+                'ne': '{0} != {1}',
+                'lt': '{0} < {1}',
+                'gt': '{0} > {1}',
+                'le': '{0} <= {1}',
+                'ge': '{0} >= {1}',
+
+                'and': '{0} & {1}',     # logical op
+                'or': '{0} | {1}',
+                'xor': '{0} ^ {1}'}
+
+    return ops_dict[opname].format(lval, rval)
+
+
+def get_dispatch(dtypes):
+
+    ops_list = ['add', 'sub', 'mul', 'div', 'mod', 'truediv',
+                'floordiv', 'pow',
+                'eq', 'ne', 'lt', 'gt', 'le', 'ge',
+                'and', 'or', 'xor']
+
+    for opname in ops_list:
+        for dtype, arith_comp_group, logical_group in dtypes:
+
+            if opname in ('div', 'truediv'):
+                rdtype = 'float64'
+            elif opname in ('eq', 'ne', 'lt', 'gt', 'le', 'ge'):
+                # comparison op
+                rdtype = 'uint8'
+            elif opname in ('and', 'or', 'xor'):
+                # logical op
+                rdtype = 'uint8'
+            else:
+                rdtype = dtype
+
+            if opname in ('and', 'or', 'xor'):
+                if logical_group:
+                    yield opname, dtype, rdtype
+            else:
+                if arith_comp_group:
+                    yield opname, dtype, rdtype
+
+}}
+
+
+{{for opname, dtype, rdtype in get_dispatch(dtypes)}}
+
+{{if opname == "pow"}}
+@cython.cpow(True) # Cython 3 matches Python pow, which isn't what we want here
+{{endif}}
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef tuple block_op_{{opname}}_{{dtype}}({{dtype}}_t[:] x_,
+                                                BlockIndex xindex,
+                                                {{dtype}}_t xfill,
+                                                {{dtype}}_t[:] y_,
+                                                BlockIndex yindex,
+                                                {{dtype}}_t yfill):
+    """
+    Binary operator on BlockIndex objects with fill values
+    """
+
+    cdef:
+        BlockIndex out_index
+        Py_ssize_t xi = 0, yi = 0, out_i = 0  # fp buf indices
+        int32_t xbp = 0, ybp = 0  # block positions
+        int32_t xloc, yloc
+        Py_ssize_t xblock = 0, yblock = 0  # block numbers
+
+        {{dtype}}_t[:] x, y
+        ndarray[{{rdtype}}_t, ndim=1] out
+
+    # to suppress Cython warning
+    x = x_
+    y = y_
+
+    out_index = xindex.make_union(yindex)
+    out = np.empty(out_index.npoints, dtype=np.{{rdtype}})
+
+    # Wow, what a hack job. Need to do something about this
+
+    # walk the two SparseVectors, adding matched locations...
+    for out_i in range(out_index.npoints):
+        if yblock == yindex.nblocks:
+            # use y fill value
+            out[out_i] = {{(opname, 'x[xi]', 'yfill', dtype) | get_op}}
+            xi += 1
+
+            # advance x location
+            xbp += 1
+            if xbp == xindex.lenbuf[xblock]:
+                xblock += 1
+                xbp = 0
+            continue
+
+        if xblock == xindex.nblocks:
+            # use x fill value
+            out[out_i] = {{(opname, 'xfill', 'y[yi]', dtype) | get_op}}
+            yi += 1
+
+            # advance y location
+            ybp += 1
+            if ybp == yindex.lenbuf[yblock]:
+                yblock += 1
+                ybp = 0
+            continue
+
+        yloc = yindex.locbuf[yblock] + ybp
+        xloc = xindex.locbuf[xblock] + xbp
+
+        # each index in the out_index had to come from either x, y, or both
+        if xloc == yloc:
+            out[out_i] = {{(opname, 'x[xi]', 'y[yi]', dtype) | get_op}}
+            xi += 1
+            yi += 1
+
+            # advance both locations
+            xbp += 1
+            if xbp == xindex.lenbuf[xblock]:
+                xblock += 1
+                xbp = 0
+
+            ybp += 1
+            if ybp == yindex.lenbuf[yblock]:
+                yblock += 1
+                ybp = 0
+
+        elif xloc < yloc:
+            # use y fill value
+            out[out_i] = {{(opname, 'x[xi]', 'yfill', dtype) | get_op}}
+            xi += 1
+
+            # advance x location
+            xbp += 1
+            if xbp == xindex.lenbuf[xblock]:
+                xblock += 1
+                xbp = 0
+        else:
+            # use x fill value
+            out[out_i] = {{(opname, 'xfill', 'y[yi]', dtype) | get_op}}
+            yi += 1
+
+            # advance y location
+            ybp += 1
+            if ybp == yindex.lenbuf[yblock]:
+                yblock += 1
+                ybp = 0
+
+    return out, out_index, {{(opname, 'xfill', 'yfill', dtype) | get_op}}
+
+{{if opname == "pow"}}
+@cython.cpow(True) # Cython 3 matches Python pow, which isn't what we want here
+{{endif}}
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef tuple int_op_{{opname}}_{{dtype}}({{dtype}}_t[:] x_,
+                                              IntIndex xindex,
+                                              {{dtype}}_t xfill,
+                                              {{dtype}}_t[:] y_,
+                                              IntIndex yindex,
+                                              {{dtype}}_t yfill):
+    cdef:
+        IntIndex out_index
+        Py_ssize_t xi = 0, yi = 0, out_i = 0  # fp buf indices
+        int32_t xloc, yloc
+        int32_t[:] xindices, yindices, out_indices
+        {{dtype}}_t[:] x, y
+        ndarray[{{rdtype}}_t, ndim=1] out
+
+    # suppress Cython compiler warnings due to inlining
+    x = x_
+    y = y_
+
+    # need to do this first to know size of result array
+    out_index = xindex.make_union(yindex)
+    out = np.empty(out_index.npoints, dtype=np.{{rdtype}})
+
+    xindices = xindex.indices
+    yindices = yindex.indices
+    out_indices = out_index.indices
+
+    # walk the two SparseVectors, adding matched locations...
+    for out_i in range(out_index.npoints):
+        if xi == xindex.npoints:
+            # use x fill value
+            out[out_i] = {{(opname, 'xfill', 'y[yi]', dtype) | get_op}}
+            yi += 1
+            continue
+
+        if yi == yindex.npoints:
+            # use y fill value
+            out[out_i] = {{(opname, 'x[xi]', 'yfill', dtype) | get_op}}
+            xi += 1
+            continue
+
+        xloc = xindices[xi]
+        yloc = yindices[yi]
+
+        # each index in the out_index had to come from either x, y, or both
+        if xloc == yloc:
+            out[out_i] = {{(opname, 'x[xi]', 'y[yi]', dtype) | get_op}}
+            xi += 1
+            yi += 1
+        elif xloc < yloc:
+            # use y fill value
+            out[out_i] = {{(opname, 'x[xi]', 'yfill', dtype) | get_op}}
+            xi += 1
+        else:
+            # use x fill value
+            out[out_i] = {{(opname, 'xfill', 'y[yi]', dtype) | get_op}}
+            yi += 1
+
+    return out, out_index, {{(opname, 'xfill', 'yfill', dtype) | get_op}}
+
+
+cpdef sparse_{{opname}}_{{dtype}}({{dtype}}_t[:] x,
+                                  SparseIndex xindex, {{dtype}}_t xfill,
+                                  {{dtype}}_t[:] y,
+                                  SparseIndex yindex, {{dtype}}_t yfill):
+
+    if isinstance(xindex, BlockIndex):
+        return block_op_{{opname}}_{{dtype}}(x, xindex.to_block_index(), xfill,
+                                             y, yindex.to_block_index(), yfill)
+    elif isinstance(xindex, IntIndex):
+        return int_op_{{opname}}_{{dtype}}(x, xindex.to_int_index(), xfill,
+                                           y, yindex.to_int_index(), yfill)
+    else:
+        raise NotImplementedError
+
+{{endfor}}
diff --git a/pandas/pandas/_libs/src/datetime/date_conversions.c b/pandas/pandas/_libs/src/datetime/date_conversions.c
new file mode 100644
index 0000000000000..99081746b2c97
--- /dev/null
+++ b/pandas/pandas/_libs/src/datetime/date_conversions.c
@@ -0,0 +1,96 @@
+/*
+Copyright (c) 2020, PyData Development Team
+All rights reserved.
+Distributed under the terms of the BSD Simplified License.
+The full license is in the LICENSE file, distributed with this software.
+*/
+
+// Conversion routines that are useful for serialization,
+// but which don't interact with JSON objects directly
+
+#include "pandas/datetime/date_conversions.h"
+#include "pandas/vendored/numpy/datetime/np_datetime.h"
+#include "pandas/vendored/numpy/datetime/np_datetime_strings.h"
+
+/*
+ * Function: scaleNanosecToUnit
+ * -----------------------------
+ *
+ * Scales an integer value representing time in nanoseconds to provided unit.
+ *
+ * Mutates the provided value directly. Returns 0 on success, non-zero on error.
+ */
+int scaleNanosecToUnit(int64_t *value, NPY_DATETIMEUNIT unit) {
+  switch (unit) {
+  case NPY_FR_ns:
+    break;
+  case NPY_FR_us:
+    *value /= 1000LL;
+    break;
+  case NPY_FR_ms:
+    *value /= 1000000LL;
+    break;
+  case NPY_FR_s:
+    *value /= 1000000000LL;
+    break;
+  default:
+    return -1;
+  }
+
+  return 0;
+}
+
+/* Converts the int64_t representation of a datetime to ISO; mutates len */
+char *int64ToIso(int64_t value, NPY_DATETIMEUNIT valueUnit,
+                 NPY_DATETIMEUNIT base, size_t *len) {
+  npy_datetimestruct dts;
+  int ret_code;
+
+  pandas_datetime_to_datetimestruct(value, valueUnit, &dts);
+
+  *len = (size_t)get_datetime_iso_8601_strlen(0, base);
+  char *result = PyObject_Malloc(*len);
+
+  if (result == NULL) {
+    PyErr_NoMemory();
+    return NULL;
+  }
+  // datetime64 is always naive
+  ret_code = make_iso_8601_datetime(&dts, result, *len, 0, base);
+  if (ret_code != 0) {
+    PyErr_SetString(PyExc_ValueError,
+                    "Could not convert datetime value to string");
+    PyObject_Free(result);
+  }
+
+  // Note that get_datetime_iso_8601_strlen just gives a generic size
+  // for ISO string conversion, not the actual size used
+  *len = strlen(result);
+  return result;
+}
+
+/* Converts the int64_t representation of a duration to ISO; mutates len */
+char *int64ToIsoDuration(int64_t value, size_t *len) {
+  pandas_timedeltastruct tds;
+  int ret_code;
+
+  pandas_timedelta_to_timedeltastruct(value, NPY_FR_ns, &tds);
+
+  // Max theoretical length of ISO Duration with 64 bit day
+  // as the largest unit is 70 characters + 1 for a null terminator
+  char *result = PyObject_Malloc(71);
+  if (result == NULL) {
+    PyErr_NoMemory();
+    return NULL;
+  }
+
+  ret_code = make_iso_8601_timedelta(&tds, result, len);
+  if (ret_code == -1) {
+    PyErr_SetString(PyExc_ValueError,
+                    "Could not convert timedelta value to string");
+    PyObject_Free(result);
+    return NULL;
+  }
+
+  return result;
+}
diff --git a/pandas/pandas/_libs/src/datetime/pd_datetime.c b/pandas/pandas/_libs/src/datetime/pd_datetime.c
new file mode 100644
index 0000000000000..2c32fb0481486
--- /dev/null
+++ b/pandas/pandas/_libs/src/datetime/pd_datetime.c
@@ -0,0 +1,268 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Copyright (c) 2005-2011, NumPy Developers
+All rights reserved.
+
+This file is derived from NumPy 1.7. See NUMPY_LICENSE.txt
+
+*/
+
+#define _PANDAS_DATETIME_IMPL
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#include "datetime.h"
+/* Need to import_array for np_datetime.c (for NumPy 1.x support only) */
+#define PY_ARRAY_UNIQUE_SYMBOL PANDAS_DATETIME_NUMPY
+#include "numpy/ndarrayobject.h"
+#include "pandas/datetime/pd_datetime.h"
+#include "pandas/portable.h"
+
+static void pandas_datetime_destructor(PyObject *op) {
+  void *ptr = PyCapsule_GetPointer(op, PandasDateTime_CAPSULE_NAME);
+  PyMem_Free(ptr);
+}
+
+/*
+ *
+ * Converts a Python datetime.datetime or datetime.date
+ * object into a NumPy npy_datetimestruct.  Uses tzinfo (if present)
+ * to convert to UTC time.
+ *
+ * The following implementation just asks for attributes, and thus
+ * supports datetime duck typing. The tzinfo time zone conversion
+ * requires this style of access as well.
+ *
+ * Returns -1 on error, 0 on success, and 1 (with no error set)
+ * if obj doesn't have the needed date or datetime attributes.
+ */
+static int convert_pydatetime_to_datetimestruct(PyObject *dtobj,
+                                                npy_datetimestruct *out) {
+  // Assumes that obj is a valid datetime object
+  PyObject *tmp;
+  PyObject *obj = (PyObject *)dtobj;
+
+  /* Initialize the output to all zeros */
+  memset(out, 0, sizeof(npy_datetimestruct));
+  out->month = 1;
+  out->day = 1;
+
+  out->year = PyLong_AsLong(PyObject_GetAttrString(obj, "year"));
+  out->month = PyLong_AsLong(PyObject_GetAttrString(obj, "month"));
+  out->day = PyLong_AsLong(PyObject_GetAttrString(obj, "day"));
+
+  // TODO(anyone): If we can get PyDateTime_IMPORT to work, we could use
+  // PyDateTime_Check here, and less verbose attribute lookups.
+
+  /* Check for time attributes (if not there, return success as a date) */
+  if (!PyObject_HasAttrString(obj, "hour") ||
+      !PyObject_HasAttrString(obj, "minute") ||
+      !PyObject_HasAttrString(obj, "second") ||
+      !PyObject_HasAttrString(obj, "microsecond")) {
+    return 0;
+  }
+
+  out->hour = PyLong_AsLong(PyObject_GetAttrString(obj, "hour"));
+  out->min = PyLong_AsLong(PyObject_GetAttrString(obj, "minute"));
+  out->sec = PyLong_AsLong(PyObject_GetAttrString(obj, "second"));
+  out->us = PyLong_AsLong(PyObject_GetAttrString(obj, "microsecond"));
+
+  if (PyObject_HasAttrString(obj, "tzinfo")) {
+    PyObject *offset = extract_utc_offset(obj);
+    /* Apply the time zone offset if datetime obj is tz-aware */
+    if (offset != NULL) {
+      if (offset == Py_None) {
+        Py_DECREF(offset);
+        return 0;
+      }
+      PyObject *tmp_int;
+      int seconds_offset, minutes_offset;
+      /*
+       * The timedelta should have a function "total_seconds"
+       * which contains the value we want.
+       */
+      tmp = PyObject_CallMethod(offset, "total_seconds", "");
+      Py_DECREF(offset);
+      if (tmp == NULL) {
+        return -1;
+      }
+      tmp_int = PyNumber_Long(tmp);
+      if (tmp_int == NULL) {
+        Py_DECREF(tmp);
+        return -1;
+      }
+      seconds_offset = PyLong_AsLong(tmp_int);
+      if (seconds_offset == -1 && PyErr_Occurred()) {
+        Py_DECREF(tmp_int);
+        Py_DECREF(tmp);
+        return -1;
+      }
+      Py_DECREF(tmp_int);
+      Py_DECREF(tmp);
+
+      /* Convert to a minutes offset and apply it */
+      minutes_offset = seconds_offset / 60;
+
+      add_minutes_to_datetimestruct(out, -minutes_offset);
+    }
+  }
+
+  return 0;
+}
+
+// Converts a Python object representing a Date / Datetime to ISO format
+// up to precision `base` e.g. base="s" yields 2020-01-03T00:00:00Z
+// while base="ns" yields "2020-01-01T00:00:00.000000000Z"
+// len is mutated to save the length of the returned string
+static char *PyDateTimeToIso(PyObject *obj, NPY_DATETIMEUNIT base,
+                             size_t *len) {
+  npy_datetimestruct dts;
+  int ret;
+
+  ret = convert_pydatetime_to_datetimestruct(obj, &dts);
+  if (ret != 0) {
+    if (!PyErr_Occurred()) {
+      PyErr_SetString(PyExc_ValueError,
+                      "Could not convert PyDateTime to numpy datetime");
+    }
+    return NULL;
+  }
+
+  *len = (size_t)get_datetime_iso_8601_strlen(0, base);
+  char *result = PyObject_Malloc(*len);
+  // Check to see if PyDateTime has a timezone.
+  // Don't convert to UTC if it doesn't.
+  int is_tz_aware = 0;
+  if (PyObject_HasAttrString(obj, "tzinfo")) {
+    PyObject *offset = extract_utc_offset(obj);
+    if (offset == NULL) {
+      PyObject_Free(result);
+      return NULL;
+    }
+    is_tz_aware = offset != Py_None;
+    Py_DECREF(offset);
+  }
+  ret = make_iso_8601_datetime(&dts, result, *len, is_tz_aware, base);
+
+  if (ret != 0) {
+    PyErr_SetString(PyExc_ValueError,
+                    "Could not convert datetime value to string");
+    PyObject_Free(result);
+    return NULL;
+  }
+
+  // Note that get_datetime_iso_8601_strlen just gives a generic size
+  // for ISO string conversion, not the actual size used
+  *len = strlen(result);
+  return result;
+}
+
+// Convert a Python Date/Datetime to Unix epoch with resolution base
+static npy_datetime PyDateTimeToEpoch(PyObject *dt, NPY_DATETIMEUNIT base) {
+  npy_datetimestruct dts;
+  int ret;
+
+  ret = convert_pydatetime_to_datetimestruct(dt, &dts);
+  if (ret != 0) {
+    if (!PyErr_Occurred()) {
+      PyErr_SetString(PyExc_ValueError,
+                      "Could not convert PyDateTime to numpy datetime");
+
+      return -1;
+    }
+  }
+
+  int64_t npy_dt = npy_datetimestruct_to_datetime(NPY_FR_ns, &dts);
+  if (scaleNanosecToUnit(&npy_dt, base) == -1) {
+    PyErr_Format(PyExc_ValueError,
+                 "Call to scaleNanosecToUnit with value %" NPY_DATETIME_FMT
+                 " and base %d failed",
+                 npy_dt, base);
+
+    return -1;
+  }
+  return npy_dt;
+}
+
+static int pandas_datetime_exec(PyObject *Py_UNUSED(module)) {
+  PyDateTime_IMPORT;
+  PandasDateTime_CAPI *capi = PyMem_Malloc(sizeof(PandasDateTime_CAPI));
+  if (capi == NULL) {
+    PyErr_NoMemory();
+    return -1;
+  }
+  capi->npy_datetimestruct_to_datetime = npy_datetimestruct_to_datetime;
+  capi->scaleNanosecToUnit = scaleNanosecToUnit;
+  capi->int64ToIso = int64ToIso;
+  capi->PyDateTimeToIso = PyDateTimeToIso;
+  capi->PyDateTimeToEpoch = PyDateTimeToEpoch;
+  capi->int64ToIsoDuration = int64ToIsoDuration;
+  capi->pandas_datetime_to_datetimestruct = pandas_datetime_to_datetimestruct;
+  capi->pandas_timedelta_to_timedeltastruct =
+      pandas_timedelta_to_timedeltastruct;
+  capi->convert_pydatetime_to_datetimestruct =
+      convert_pydatetime_to_datetimestruct;
+  capi->cmp_npy_datetimestruct = cmp_npy_datetimestruct;
+  capi->get_datetime_metadata_from_dtype = get_datetime_metadata_from_dtype;
+  capi->parse_iso_8601_datetime = parse_iso_8601_datetime;
+  capi->get_datetime_iso_8601_strlen = get_datetime_iso_8601_strlen;
+  capi->make_iso_8601_datetime = make_iso_8601_datetime;
+  capi->make_iso_8601_timedelta = make_iso_8601_timedelta;
+
+  PyObject *capsule = PyCapsule_New(capi, PandasDateTime_CAPSULE_NAME,
+                                    pandas_datetime_destructor);
+  if (capsule == NULL) {
+    PyMem_Free(capi);
+    return -1;
+  }
+
+  // Monkeypatch the top level pandas module to have an attribute for the
+  // C-API. This is required because Python capsules do not support setting
+  // this attribute on anything but the top level package. Ideally not
+  // done when cpython gh-6898 gets implemented
+  PyObject *pandas = PyImport_ImportModule("pandas");
+  if (!pandas) {
+    PyErr_SetString(PyExc_ImportError,
+                    "pd_datetime.c could not import module pandas");
+    Py_DECREF(capsule);
+    return -1;
+  }
+
+  if (PyModule_AddObject(pandas, "_pandas_datetime_CAPI", capsule) < 0) {
+    Py_DECREF(capsule);
+    return -1;
+  }
+
+  return 0;
+}
+
+static PyModuleDef_Slot pandas_datetime_slots[] = {
+    {Py_mod_exec, pandas_datetime_exec},
+#if PY_VERSION_HEX >= 0x030D0000
+    {Py_mod_gil, Py_MOD_GIL_NOT_USED},
+#endif
+    {0, NULL},
+};
+
+static struct PyModuleDef pandas_datetimemodule = {
+    PyModuleDef_HEAD_INIT,
+    .m_name = "pandas._libs.pandas_datetime",
+
+    .m_doc = "Internal module with datetime support for other extensions",
+    .m_size = 0,
+    .m_methods = NULL,
+    .m_slots = pandas_datetime_slots};
+
+PyMODINIT_FUNC PyInit_pandas_datetime(void) {
+  PyDateTime_IMPORT;
+  import_array();
+  return PyModuleDef_Init(&pandas_datetimemodule);
+}
diff --git a/pandas/pandas/_libs/src/parser/io.c b/pandas/pandas/_libs/src/parser/io.c
new file mode 100644
index 0000000000000..851901481d222
--- /dev/null
+++ b/pandas/pandas/_libs/src/parser/io.c
@@ -0,0 +1,99 @@
+/*
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+*/
+
+#include "pandas/parser/io.h"
+
+/*
+  On-disk FILE, uncompressed
+*/
+
+void *new_rd_source(PyObject *obj) {
+  rd_source *rds = (rd_source *)malloc(sizeof(rd_source));
+
+  if (rds == NULL) {
+    PyErr_NoMemory();
+    return NULL;
+  }
+  /* hold on to this object */
+  Py_INCREF(obj);
+  rds->obj = obj;
+  rds->buffer = NULL;
+  rds->position = 0;
+
+  return (void *)rds;
+}
+
+/*
+
+  Cleanup callbacks
+
+ */
+
+void del_rd_source(void *rds) {
+  Py_XDECREF(RDS(rds)->obj);
+  Py_XDECREF(RDS(rds)->buffer);
+  free(rds);
+}
+
+/*
+
+  IO callbacks
+
+ */
+
+char *buffer_rd_bytes(void *source, size_t nbytes, size_t *bytes_read,
+                      int *status, const char *encoding_errors) {
+  rd_source *src = RDS(source);
+  PyGILState_STATE state = PyGILState_Ensure();
+
+  /* delete old object */
+  Py_XDECREF(src->buffer);
+  src->buffer = NULL;
+  PyObject *args = Py_BuildValue("(i)", nbytes);
+
+  PyObject *func = PyObject_GetAttrString(src->obj, "read");
+
+  /* Note: PyObject_CallObject requires the GIL */
+  PyObject *result = PyObject_CallObject(func, args);
+  Py_XDECREF(args);
+  Py_XDECREF(func);
+
+  if (result == NULL) {
+    PyGILState_Release(state);
+    *bytes_read = 0;
+    *status = CALLING_READ_FAILED;
+    return NULL;
+  } else if (!PyBytes_Check(result)) {
+    PyObject *tmp = PyUnicode_AsEncodedString(result, "utf-8", encoding_errors);
+    Py_DECREF(result);
+    if (tmp == NULL) {
+      PyGILState_Release(state);
+      return NULL;
+    }
+    result = tmp;
+  }
+
+  const size_t length = PySequence_Length(result);
+
+  if (length == 0)
+    *status = REACHED_EOF;
+  else
+    *status = 0;
+
+  /* hang on to the Python object */
+  src->buffer = result;
+  char *retval = PyBytes_AsString(result);
+
+  PyGILState_Release(state);
+
+  /* TODO: more error handling */
+  *bytes_read = length;
+
+  return retval;
+}
diff --git a/pandas/pandas/_libs/src/parser/pd_parser.c b/pandas/pandas/_libs/src/parser/pd_parser.c
new file mode 100644
index 0000000000000..51cdf071a15cf
--- /dev/null
+++ b/pandas/pandas/_libs/src/parser/pd_parser.c
@@ -0,0 +1,182 @@
+/*
+
+Copyright (c) 2023, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+*/
+#define _PANDAS_PARSER_IMPL
+
+#include "pandas/parser/pd_parser.h"
+#include "pandas/parser/io.h"
+#include "pandas/portable.h"
+
+static int to_double(char *item, double *p_value, char sci, char decimal,
+                     int *maybe_int) {
+  char *p_end = NULL;
+  int error = 0;
+
+  /* Switch to precise xstrtod GH 31364 */
+  *p_value =
+      precise_xstrtod(item, &p_end, decimal, sci, '\0', 1, &error, maybe_int);
+
+  return (error == 0) && (!*p_end);
+}
+
+static int floatify(PyObject *str, double *result, int *maybe_int) {
+  char *data;
+  PyObject *tmp = NULL;
+  const char sci = 'E';
+  const char dec = '.';
+
+  if (PyBytes_Check(str)) {
+    data = PyBytes_AS_STRING(str);
+  } else if (PyUnicode_Check(str)) {
+    tmp = PyUnicode_AsUTF8String(str);
+    if (tmp == NULL) {
+      return -1;
+    }
+    data = PyBytes_AS_STRING(tmp);
+  } else {
+    PyErr_SetString(PyExc_TypeError, "Invalid object type");
+    return -1;
+  }
+
+  const int status = to_double(data, result, sci, dec, maybe_int);
+
+  if (!status) {
+    /* handle inf/-inf infinity/-infinity */
+    if (strlen(data) == 3) {
+      if (0 == strcasecmp(data, "inf")) {
+        *result = HUGE_VAL;
+        *maybe_int = 0;
+      } else {
+        goto parsingerror;
+      }
+    } else if (strlen(data) == 4) {
+      if (0 == strcasecmp(data, "-inf")) {
+        *result = -HUGE_VAL;
+        *maybe_int = 0;
+      } else if (0 == strcasecmp(data, "+inf")) {
+        *result = HUGE_VAL;
+        *maybe_int = 0;
+      } else {
+        goto parsingerror;
+      }
+    } else if (strlen(data) == 8) {
+      if (0 == strcasecmp(data, "infinity")) {
+        *result = HUGE_VAL;
+        *maybe_int = 0;
+      } else {
+        goto parsingerror;
+      }
+    } else if (strlen(data) == 9) {
+      if (0 == strcasecmp(data, "-infinity")) {
+        *result = -HUGE_VAL;
+        *maybe_int = 0;
+      } else if (0 == strcasecmp(data, "+infinity")) {
+        *result = HUGE_VAL;
+        *maybe_int = 0;
+      } else {
+        goto parsingerror;
+      }
+    } else {
+      goto parsingerror;
+    }
+  }
+
+  Py_XDECREF(tmp);
+  return 0;
+
+parsingerror:
+  PyErr_Format(PyExc_ValueError, "Unable to parse string \"%s\"", data);
+  Py_XDECREF(tmp);
+  return -1;
+}
+
+static void pandas_parser_destructor(PyObject *op) {
+  void *ptr = PyCapsule_GetPointer(op, PandasParser_CAPSULE_NAME);
+  PyMem_Free(ptr);
+}
+
+static int pandas_parser_exec(PyObject *Py_UNUSED(module)) {
+  PandasParser_CAPI *capi = PyMem_Malloc(sizeof(PandasParser_CAPI));
+  if (capi == NULL) {
+    PyErr_NoMemory();
+    return -1;
+  }
+
+  capi->to_double = to_double;
+  capi->floatify = floatify;
+  capi->new_rd_source = new_rd_source;
+  capi->del_rd_source = del_rd_source;
+  capi->buffer_rd_bytes = buffer_rd_bytes;
+  capi->uint_state_init = uint_state_init;
+  capi->uint64_conflict = uint64_conflict;
+  capi->coliter_setup = coliter_setup;
+  capi->parser_new = parser_new;
+  capi->parser_init = parser_init;
+  capi->parser_free = parser_free;
+  capi->parser_del = parser_del;
+  capi->parser_add_skiprow = parser_add_skiprow;
+  capi->parser_set_skipfirstnrows = parser_set_skipfirstnrows;
+  capi->parser_set_default_options = parser_set_default_options;
+  capi->parser_consume_rows = parser_consume_rows;
+  capi->parser_trim_buffers = parser_trim_buffers;
+  capi->tokenize_all_rows = tokenize_all_rows;
+  capi->tokenize_nrows = tokenize_nrows;
+  capi->str_to_int64 = str_to_int64;
+  capi->str_to_uint64 = str_to_uint64;
+  capi->xstrtod = xstrtod;
+  capi->precise_xstrtod = precise_xstrtod;
+  capi->round_trip = round_trip;
+  capi->to_boolean = to_boolean;
+
+  PyObject *capsule =
+      PyCapsule_New(capi, PandasParser_CAPSULE_NAME, pandas_parser_destructor);
+  if (capsule == NULL) {
+    PyMem_Free(capi);
+    return -1;
+  }
+
+  // Monkeypatch the top level pandas module to have an attribute for the
+  // C-API. This is required because Python capsules do not support setting
+  // this attribute on anything but the top level package. Ideally not
+  // done when cpython gh-6898 gets implemented
+  PyObject *pandas = PyImport_ImportModule("pandas");
+  if (!pandas) {
+    PyErr_SetString(PyExc_ImportError,
+                    "pd_parser.c could not import module pandas");
+    Py_DECREF(capsule);
+    return -1;
+  }
+
+  if (PyModule_AddObject(pandas, "_pandas_parser_CAPI", capsule) < 0) {
+    Py_DECREF(capsule);
+    return -1;
+  }
+
+  return 0;
+}
+
+static PyModuleDef_Slot pandas_parser_slots[] = {
+    {Py_mod_exec, pandas_parser_exec},
+#if PY_VERSION_HEX >= 0x030D0000
+    {Py_mod_gil, Py_MOD_GIL_NOT_USED},
+#endif
+    {0, NULL},
+};
+
+static struct PyModuleDef pandas_parsermodule = {
+    PyModuleDef_HEAD_INIT,
+    .m_name = "pandas._libs.pandas_parser",
+
+    .m_doc = "Internal module with parser support for other extensions",
+    .m_size = 0,
+    .m_methods = NULL,
+    .m_slots = pandas_parser_slots};
+
+PyMODINIT_FUNC PyInit_pandas_parser(void) {
+  return PyModuleDef_Init(&pandas_parsermodule);
+}
diff --git a/pandas/pandas/_libs/src/parser/tokenizer.c b/pandas/pandas/_libs/src/parser/tokenizer.c
new file mode 100644
index 0000000000000..61e96fc835e4d
--- /dev/null
+++ b/pandas/pandas/_libs/src/parser/tokenizer.c
@@ -0,0 +1,2034 @@
+/*
+
+Copyright (c) 2012, Lambda Foundry, Inc., except where noted
+
+Incorporates components of WarrenWeckesser/textreader, licensed under 3-clause
+BSD
+
+See LICENSE for the license
+
+*/
+
+/*
+
+Low-level ascii-file processing for pandas. Combines some elements from
+Python's built-in csv module and Warren Weckesser's textreader project on
+GitHub. See Python Software Foundation License and BSD licenses for these.
+
+*/
+#include "pandas/parser/tokenizer.h"
+#include "pandas/portable.h"
+
+#include <ctype.h>
+#include <float.h>
+#include <math.h>
+#include <stdbool.h>
+
+#include "pandas/portable.h"
+#include "pandas/vendored/klib/khash.h" // for kh_int64_t, kh_destroy_int64
+
+void coliter_setup(coliter_t *self, parser_t *parser, int64_t i,
+                   int64_t start) {
+  // column i, starting at 0
+  self->words = parser->words;
+  self->col = i;
+  self->line_start = parser->line_start + start;
+}
+
+static void free_if_not_null(void **ptr) {
+  TRACE(("free_if_not_null %p\n", *ptr))
+  if (*ptr != NULL) {
+    free(*ptr);
+    *ptr = NULL;
+  }
+}
+
+/*
+
+  Parser / tokenizer
+
+*/
+
+static void *grow_buffer(void *buffer, uint64_t length, uint64_t *capacity,
+                         int64_t space, int64_t elsize, int *error) {
+  uint64_t cap = *capacity;
+  void *newbuffer = buffer;
+
+  // Can we fit potentially nbytes tokens (+ null terminators) in the stream?
+  while ((length + space >= cap) && (newbuffer != NULL)) {
+    cap = cap ? cap << 1 : 2;
+    buffer = newbuffer;
+    newbuffer = realloc(newbuffer, elsize * cap);
+  }
+
+  if (newbuffer == NULL) {
+    // realloc failed so don't change *capacity, set *error to errno
+    // and return the last good realloc'd buffer so it can be freed
+    *error = errno;
+    newbuffer = buffer;
+  } else {
+    // realloc worked, update *capacity and set *error to 0
+    // sigh, multiple return values
+    *capacity = cap;
+    *error = 0;
+  }
+  return newbuffer;
+}
+
+void parser_set_default_options(parser_t *self) {
+  self->decimal = '.';
+  self->sci = 'E';
+
+  // For tokenization
+  self->state = START_RECORD;
+
+  self->delimiter = ','; // XXX
+  self->delim_whitespace = 0;
+
+  self->doublequote = 0;
+  self->quotechar = '"';
+  self->escapechar = 0;
+
+  self->lineterminator = '\0'; /* NUL->standard logic */
+
+  self->skipinitialspace = 0;
+  self->quoting = QUOTE_MINIMAL;
+  self->allow_embedded_newline = 1;
+
+  self->expected_fields = -1;
+  self->on_bad_lines = ERROR;
+
+  self->commentchar = '#';
+  self->thousands = '\0';
+
+  self->skipset = NULL;
+  self->skipfunc = NULL;
+  self->skip_first_N_rows = -1;
+  self->skip_footer = 0;
+}
+
+parser_t *parser_new(void) { return (parser_t *)calloc(1, sizeof(parser_t)); }
+
+static void parser_clear_data_buffers(parser_t *self) {
+  free_if_not_null((void *)&self->stream);
+  free_if_not_null((void *)&self->words);
+  free_if_not_null((void *)&self->word_starts);
+  free_if_not_null((void *)&self->line_start);
+  free_if_not_null((void *)&self->line_fields);
+}
+
+static void parser_cleanup(parser_t *self) {
+  // XXX where to put this
+  free_if_not_null((void *)&self->error_msg);
+  free_if_not_null((void *)&self->warn_msg);
+
+  if (self->skipset != NULL) {
+    kh_destroy_int64((kh_int64_t *)self->skipset);
+    self->skipset = NULL;
+  }
+
+  parser_clear_data_buffers(self);
+  if (self->cb_cleanup != NULL) {
+    self->cb_cleanup(self->source);
+    self->cb_cleanup = NULL;
+  }
+}
+
+int parser_init(parser_t *self) {
+  /*
+    Initialize data buffers
+  */
+
+  self->stream = NULL;
+  self->words = NULL;
+  self->word_starts = NULL;
+  self->line_start = NULL;
+  self->line_fields = NULL;
+  self->error_msg = NULL;
+  self->warn_msg = NULL;
+
+  // token stream
+  self->stream = malloc(STREAM_INIT_SIZE);
+  if (self->stream == NULL) {
+    parser_cleanup(self);
+    return PARSER_OUT_OF_MEMORY;
+  }
+  self->stream_cap = STREAM_INIT_SIZE;
+  self->stream_len = 0;
+
+  // word pointers and metadata
+  _Static_assert(STREAM_INIT_SIZE / 10 > 0,
+                 "STREAM_INIT_SIZE must be defined and >= 10");
+  const int64_t sz = STREAM_INIT_SIZE / 10;
+  self->words = malloc(sz * sizeof(char *));
+  self->word_starts = malloc(sz * sizeof(int64_t));
+  self->max_words_cap = sz;
+  self->words_cap = sz;
+  self->words_len = 0;
+
+  // line pointers and metadata
+  self->line_start = malloc(sz * sizeof(int64_t));
+
+  self->line_fields = malloc(sz * sizeof(int64_t));
+
+  self->lines_cap = sz;
+  self->lines = 0;
+  self->file_lines = 0;
+
+  if (self->stream == NULL || self->words == NULL ||
+      self->word_starts == NULL || self->line_start == NULL ||
+      self->line_fields == NULL) {
+    parser_cleanup(self);
+
+    return PARSER_OUT_OF_MEMORY;
+  }
+
+  /* amount of bytes buffered */
+  self->datalen = 0;
+  self->datapos = 0;
+
+  self->line_start[0] = 0;
+  self->line_fields[0] = 0;
+
+  self->pword_start = self->stream;
+  self->word_start = 0;
+
+  self->state = START_RECORD;
+
+  self->error_msg = NULL;
+  self->warn_msg = NULL;
+
+  self->commentchar = '\0';
+
+  return 0;
+}
+
+void parser_free(parser_t *self) {
+  // opposite of parser_init
+  parser_cleanup(self);
+}
+
+void parser_del(parser_t *self) { free(self); }
+
+static int make_stream_space(parser_t *self, size_t nbytes) {
+  // Can we fit potentially nbytes tokens (+ null terminators) in the stream?
+
+  /*
+    TOKEN STREAM
+  */
+
+  int status;
+  char *orig_ptr = (void *)self->stream;
+  TRACE(("\n\nmake_stream_space: nbytes = %zu.  grow_buffer(self->stream...)\n",
+         nbytes))
+  self->stream = (char *)grow_buffer((void *)self->stream, self->stream_len,
+                                     &self->stream_cap, nbytes * 2, 1, &status);
+  TRACE(("make_stream_space: self->stream=%p, self->stream_len = %zu, "
+         "self->stream_cap=%zu, status=%zu\n",
+         self->stream, self->stream_len, self->stream_cap, status))
+
+  if (status != 0) {
+    return PARSER_OUT_OF_MEMORY;
+  }
+
+  // realloc sets errno when moving buffer?
+  if (self->stream != orig_ptr) {
+    self->pword_start = self->stream + self->word_start;
+
+    for (uint64_t i = 0; i < self->words_len; ++i) {
+      self->words[i] = self->stream + self->word_starts[i];
+    }
+  }
+
+  /*
+    WORD VECTORS
+  */
+
+  const uint64_t words_cap = self->words_cap;
+
+  /**
+   * If we are reading in chunks, we need to be aware of the maximum number
+   * of words we have seen in previous chunks (self->max_words_cap), so
+   * that way, we can properly allocate when reading subsequent ones.
+   *
+   * Otherwise, we risk a buffer overflow if we mistakenly under-allocate
+   * just because a recent chunk did not have as many words.
+   */
+  const uint64_t length = self->words_len + nbytes < self->max_words_cap
+                              ? self->max_words_cap - nbytes - 1
+                              : self->words_len;
+
+  self->words =
+      (char **)grow_buffer((void *)self->words, length, &self->words_cap,
+                           nbytes, sizeof(char *), &status);
+  TRACE(("make_stream_space: grow_buffer(self->self->words, %zu, %zu, %zu, "
+         "%d)\n",
+         self->words_len, self->words_cap, nbytes, status))
+  if (status != 0) {
+    return PARSER_OUT_OF_MEMORY;
+  }
+
+  // realloc took place
+  if (words_cap != self->words_cap) {
+    TRACE(("make_stream_space: cap != self->words_cap, nbytes = %d, "
+           "self->words_cap=%d\n",
+           nbytes, self->words_cap))
+    int64_t *newptr = (int64_t *)realloc(self->word_starts,
+                                         sizeof(int64_t) * self->words_cap);
+    if (newptr == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    } else {
+      self->word_starts = newptr;
+    }
+  }
+
+  /*
+    LINE VECTORS
+  */
+  const uint64_t lines_cap = self->lines_cap;
+  self->line_start = (int64_t *)grow_buffer((void *)self->line_start,
+                                            self->lines + 1, &self->lines_cap,
+                                            nbytes, sizeof(int64_t), &status);
+  TRACE(
+      ("make_stream_space: grow_buffer(self->line_start, %zu, %zu, %zu, %d)\n",
+       self->lines + 1, self->lines_cap, nbytes, status))
+  if (status != 0) {
+    return PARSER_OUT_OF_MEMORY;
+  }
+
+  // realloc took place
+  if (lines_cap != self->lines_cap) {
+    TRACE(("make_stream_space: cap != self->lines_cap, nbytes = %d\n", nbytes))
+    int64_t *newptr = (int64_t *)realloc(self->line_fields,
+                                         sizeof(int64_t) * self->lines_cap);
+    if (newptr == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    } else {
+      self->line_fields = newptr;
+    }
+  }
+
+  return 0;
+}
+
+static int push_char(parser_t *self, char c) {
+  TRACE(("push_char: self->stream[%zu] = %x, stream_cap=%zu\n",
+         self->stream_len + 1, c, self->stream_cap))
+  if (self->stream_len >= self->stream_cap) {
+    TRACE(("push_char: ERROR!!! self->stream_len(%d) >= "
+           "self->stream_cap(%d)\n",
+           self->stream_len, self->stream_cap))
+    const size_t bufsize = 100;
+    self->error_msg = malloc(bufsize);
+    snprintf(self->error_msg, bufsize,
+             "Buffer overflow caught - possible malformed input file.\n");
+    return PARSER_OUT_OF_MEMORY;
+  }
+  self->stream[self->stream_len++] = c;
+  return 0;
+}
+
+static inline int end_field(parser_t *self) {
+  // XXX cruft
+  if (self->words_len >= self->words_cap) {
+    TRACE(("end_field: ERROR!!! self->words_len(%zu) >= "
+           "self->words_cap(%zu)\n",
+           self->words_len, self->words_cap))
+    const size_t bufsize = 100;
+    self->error_msg = malloc(bufsize);
+    snprintf(self->error_msg, bufsize,
+             "Buffer overflow caught - possible malformed input file.\n");
+    return PARSER_OUT_OF_MEMORY;
+  }
+
+  // null terminate token
+  push_char(self, '\0');
+
+  // set pointer and metadata
+  self->words[self->words_len] = self->pword_start;
+
+  TRACE(("end_field: Char diff: %d\n", self->pword_start - self->words[0]));
+
+  TRACE(("end_field: Saw word %s at: %d. Total: %d\n", self->pword_start,
+         self->word_start, self->words_len + 1))
+
+  self->word_starts[self->words_len] = self->word_start;
+  self->words_len++;
+
+  // increment line field count
+  self->line_fields[self->lines]++;
+
+  // New field begin in stream
+  self->pword_start = self->stream + self->stream_len;
+  self->word_start = self->stream_len;
+
+  return 0;
+}
+
+static void append_warning(parser_t *self, const char *msg) {
+  const int64_t length = strlen(msg);
+
+  if (self->warn_msg == NULL) {
+    self->warn_msg = malloc(length + 1);
+    snprintf(self->warn_msg, length + 1, "%s", msg);
+  } else {
+    const int64_t ex_length = strlen(self->warn_msg);
+    char *newptr = (char *)realloc(self->warn_msg, ex_length + length + 1);
+    if (newptr != NULL) {
+      self->warn_msg = newptr;
+      snprintf(self->warn_msg + ex_length, length + 1, "%s", msg);
+    }
+  }
+}
+
+static int end_line(parser_t *self) {
+  int64_t ex_fields = self->expected_fields;
+  int64_t fields = self->line_fields[self->lines];
+
+  TRACE(("end_line: Line end, nfields: %d\n", fields));
+
+  TRACE(("end_line: lines: %d\n", self->lines));
+  if (self->lines > 0) {
+    if (self->expected_fields >= 0) {
+      ex_fields = self->expected_fields;
+    } else {
+      ex_fields = self->line_fields[self->lines - 1];
+    }
+  }
+  TRACE(("end_line: ex_fields: %d\n", ex_fields));
+
+  if (self->state == START_FIELD_IN_SKIP_LINE ||
+      self->state == IN_FIELD_IN_SKIP_LINE ||
+      self->state == IN_QUOTED_FIELD_IN_SKIP_LINE ||
+      self->state == QUOTE_IN_QUOTED_FIELD_IN_SKIP_LINE) {
+    TRACE(("end_line: Skipping row %d\n", self->file_lines));
+    // increment file line count
+    self->file_lines++;
+
+    // skip the tokens from this bad line
+    self->line_start[self->lines] += fields;
+
+    // reset field count
+    self->line_fields[self->lines] = 0;
+    return 0;
+  }
+
+  if (!(self->lines <= self->header_end + 1) && (fields > ex_fields) &&
+      !(self->usecols)) {
+    // increment file line count
+    self->file_lines++;
+
+    // skip the tokens from this bad line
+    self->line_start[self->lines] += fields;
+
+    // reset field count
+    self->line_fields[self->lines] = 0;
+
+    // file_lines is now the actual file line number (starting at 1)
+    if (self->on_bad_lines == ERROR) {
+      const size_t bufsize = 100;
+      self->error_msg = malloc(bufsize);
+      snprintf(self->error_msg, bufsize,
+               "Expected %" PRId64 " fields in line %" PRIu64 ", saw %" PRId64
+               "\n",
+               ex_fields, self->file_lines, fields);
+
+      TRACE(("Error at line %d, %d fields\n", self->file_lines, fields));
+
+      return -1;
+    } else {
+      // simply skip bad lines
+      if (self->on_bad_lines == WARN) {
+        // pass up error message
+        const size_t bufsize = 100;
+        char *msg = (char *)malloc(bufsize);
+        snprintf(msg, bufsize,
+                 "Skipping line %" PRIu64 ": expected %" PRId64
+                 " fields, saw %" PRId64 "\n",
+                 self->file_lines, ex_fields, fields);
+        append_warning(self, msg);
+        free(msg);
+      }
+    }
+  } else {
+    // missing trailing delimiters
+    if ((self->lines >= self->header_end + 1) && fields < ex_fields) {
+      // might overrun the buffer when closing fields
+      if (make_stream_space(self, ex_fields - fields) < 0) {
+        const size_t bufsize = 100;
+        self->error_msg = malloc(bufsize);
+        snprintf(self->error_msg, bufsize, "out of memory");
+        return -1;
+      }
+
+      while (fields < ex_fields) {
+        end_field(self);
+        fields++;
+      }
+    }
+
+    // increment both line counts
+    self->file_lines++;
+    self->lines++;
+
+    // good line, set new start point
+    if (self->lines >= self->lines_cap) {
+      TRACE(("end_line: ERROR!!! self->lines(%zu) >= self->lines_cap(%zu)\n",
+             self->lines, self->lines_cap))
+      const size_t bufsize = 100;
+      self->error_msg = malloc(bufsize);
+      snprintf(self->error_msg, bufsize,
+               "Buffer overflow caught - "
+               "possible malformed input file.\n");
+      return PARSER_OUT_OF_MEMORY;
+    }
+    self->line_start[self->lines] =
+        (self->line_start[self->lines - 1] + fields);
+
+    TRACE(("end_line: new line start: %d\n", self->line_start[self->lines]));
+
+    // new line start with 0 fields
+    self->line_fields[self->lines] = 0;
+  }
+
+  TRACE(("end_line: Finished line, at %d\n", self->lines));
+
+  return 0;
+}
+
+int parser_add_skiprow(parser_t *self, int64_t row) {
+  khiter_t k;
+  kh_int64_t *set;
+  int ret = 0;
+
+  if (self->skipset == NULL) {
+    self->skipset = (void *)kh_init_int64();
+  }
+
+  set = (kh_int64_t *)self->skipset;
+
+  k = kh_put_int64(set, row, &ret);
+  set->keys[k] = row;
+
+  return 0;
+}
+
+void parser_set_skipfirstnrows(parser_t *self, int64_t nrows) {
+  // self->file_lines is zero based so subtract 1 from nrows
+  if (nrows > 0) {
+    self->skip_first_N_rows = nrows - 1;
+  }
+}
+
+static int parser_buffer_bytes(parser_t *self, size_t nbytes,
+                               const char *encoding_errors) {
+  int status;
+  size_t bytes_read;
+
+  status = 0;
+  self->datapos = 0;
+  self->data =
+      self->cb_io(self->source, nbytes, &bytes_read, &status, encoding_errors);
+  TRACE(
+      ("parser_buffer_bytes self->cb_io: nbytes=%zu, datalen: %d, status=%d\n",
+       nbytes, bytes_read, status));
+  self->datalen = bytes_read;
+
+  if (status != REACHED_EOF && self->data == NULL) {
+    const size_t bufsize = 200;
+    self->error_msg = malloc(bufsize);
+
+    if (status == CALLING_READ_FAILED) {
+      snprintf(self->error_msg, bufsize,
+               "Calling read(nbytes) on source failed. "
+               "Try engine='python'.");
+    } else {
+      snprintf(self->error_msg, bufsize, "Unknown error in IO callback");
+    }
+    return -1;
+  }
+
+  TRACE(("datalen: %d\n", self->datalen));
+
+  return status;
+}
+
+/*
+
+  Tokenization macros and state machine code
+
+*/
+
+#define PUSH_CHAR(c)                                                           \
+  TRACE(("PUSH_CHAR: Pushing %c, slen= %d, stream_cap=%zu, stream_len=%zu\n",  \
+         c, slen, self->stream_cap, self->stream_len))                         \
+  if (slen >= self->stream_cap) {                                              \
+    TRACE(("PUSH_CHAR: ERROR!!! slen(%d) >= stream_cap(%d)\n", slen,           \
+           self->stream_cap))                                                  \
+    const size_t bufsize = 100;                                                \
+    self->error_msg = malloc(bufsize);                                         \
+    snprintf(self->error_msg, bufsize,                                         \
+             "Buffer overflow caught - possible malformed input file.\n");     \
+    return PARSER_OUT_OF_MEMORY;                                               \
+  }                                                                            \
+  *stream++ = c;                                                               \
+  slen++;
+
+// This is a little bit of a hack but works for now
+
+#define END_FIELD()                                                            \
+  self->stream_len = slen;                                                     \
+  if (end_field(self) < 0) {                                                   \
+    goto parsingerror;                                                         \
+  }                                                                            \
+  stream = self->stream + self->stream_len;                                    \
+  slen = self->stream_len;
+
+#define END_LINE_STATE(STATE)                                                  \
+  self->stream_len = slen;                                                     \
+  if (end_line(self) < 0) {                                                    \
+    goto parsingerror;                                                         \
+  }                                                                            \
+  stream = self->stream + self->stream_len;                                    \
+  slen = self->stream_len;                                                     \
+  self->state = STATE;                                                         \
+  if (line_limit > 0 && self->lines == start_lines + line_limit) {             \
+    goto linelimit;                                                            \
+  }
+
+#define END_LINE_AND_FIELD_STATE(STATE)                                        \
+  self->stream_len = slen;                                                     \
+  if (end_line(self) < 0) {                                                    \
+    goto parsingerror;                                                         \
+  }                                                                            \
+  if (end_field(self) < 0) {                                                   \
+    goto parsingerror;                                                         \
+  }                                                                            \
+  stream = self->stream + self->stream_len;                                    \
+  slen = self->stream_len;                                                     \
+  self->state = STATE;                                                         \
+  if (line_limit > 0 && self->lines == start_lines + line_limit) {             \
+    goto linelimit;                                                            \
+  }
+
+#define END_LINE() END_LINE_STATE(START_RECORD)
+
+#define IS_TERMINATOR(c) (c == lineterminator)
+
+#define IS_QUOTE(c) ((c == self->quotechar && self->quoting != QUOTE_NONE))
+
+// don't parse '\r' with a custom line terminator
+#define IS_CARRIAGE(c) (c == carriage_symbol)
+
+#define IS_COMMENT_CHAR(c) (c == comment_symbol)
+
+#define IS_ESCAPE_CHAR(c) (c == escape_symbol)
+
+#define IS_SKIPPABLE_SPACE(c)                                                  \
+  ((!self->delim_whitespace && c == ' ' && self->skipinitialspace))
+
+// applied when in a field
+#define IS_DELIMITER(c)                                                        \
+  ((!delim_whitespace && c == delimiter) || (delim_whitespace && isblank(c)))
+
+#define _TOKEN_CLEANUP()                                                       \
+  self->stream_len = slen;                                                     \
+  self->datapos = i;                                                           \
+  TRACE(("_TOKEN_CLEANUP: datapos: %d, datalen: %d\n", self->datapos,          \
+         self->datalen));
+
+#define CHECK_FOR_BOM()                                                        \
+  if (*buf == '\xef' && *(buf + 1) == '\xbb' && *(buf + 2) == '\xbf') {        \
+    buf += 3;                                                                  \
+    self->datapos += 3;                                                        \
+  }
+
+static int skip_this_line(parser_t *self, int64_t rownum) {
+  if (self->skipfunc != NULL) {
+    PyGILState_STATE state = PyGILState_Ensure();
+    PyObject *result = PyObject_CallFunction(self->skipfunc, "i", rownum);
+
+    // Error occurred. It will be processed
+    // and caught at the Cython level.
+    const int should_skip = result == NULL ? -1 : PyObject_IsTrue(result);
+
+    Py_XDECREF(result);
+    PyGILState_Release(state);
+
+    return should_skip;
+  } else if (self->skipset != NULL) {
+    return (kh_get_int64((kh_int64_t *)self->skipset, self->file_lines) !=
+            ((kh_int64_t *)self->skipset)->n_buckets);
+  } else {
+    return (rownum <= self->skip_first_N_rows);
+  }
+}
+
+static int tokenize_bytes(parser_t *self, size_t line_limit,
+                          uint64_t start_lines) {
+  char *buf = self->data + self->datapos;
+
+  const char lineterminator =
+      (self->lineterminator == '\0') ? '\n' : self->lineterminator;
+
+  const int delim_whitespace = self->delim_whitespace;
+  const char delimiter = self->delimiter;
+
+  // 1000 is something that couldn't fit in "char"
+  // thus comparing a char to it would always be "false"
+  const int carriage_symbol = (self->lineterminator == '\0') ? '\r' : 1000;
+  const int comment_symbol =
+      (self->commentchar != '\0') ? self->commentchar : 1000;
+  const int escape_symbol =
+      (self->escapechar != '\0') ? self->escapechar : 1000;
+
+  if (make_stream_space(self, self->datalen - self->datapos) < 0) {
+    const size_t bufsize = 100;
+    self->error_msg = malloc(bufsize);
+    snprintf(self->error_msg, bufsize, "out of memory");
+    return -1;
+  }
+
+  char *stream = self->stream + self->stream_len;
+  uint64_t slen = self->stream_len;
+
+  TRACE(("%s\n", buf));
+
+  if (self->file_lines == 0) {
+    CHECK_FOR_BOM();
+  }
+
+  char c;
+  int64_t i;
+  for (i = self->datapos; i < self->datalen; ++i) {
+    // next character in file
+    c = *buf++;
+
+    TRACE(("tokenize_bytes - Iter: %d Char: 0x%x Line %d field_count %d, "
+           "state %d\n",
+           i, c, self->file_lines + 1, self->line_fields[self->lines],
+           self->state));
+
+    switch (self->state) {
+    case START_FIELD_IN_SKIP_LINE:
+      if (IS_TERMINATOR(c)) {
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        self->file_lines++;
+        self->state = EAT_CRNL_NOP;
+      } else if (IS_QUOTE(c)) {
+        self->state = IN_QUOTED_FIELD_IN_SKIP_LINE;
+      } else if (IS_DELIMITER(c)) {
+        // Do nothing, we're starting a new field again.
+      } else {
+        self->state = IN_FIELD_IN_SKIP_LINE;
+      }
+      break;
+
+    case IN_FIELD_IN_SKIP_LINE:
+      if (IS_TERMINATOR(c)) {
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        self->file_lines++;
+        self->state = EAT_CRNL_NOP;
+      } else if (IS_DELIMITER(c)) {
+        self->state = START_FIELD_IN_SKIP_LINE;
+      }
+      break;
+
+    case IN_QUOTED_FIELD_IN_SKIP_LINE:
+      if (IS_QUOTE(c)) {
+        if (self->doublequote) {
+          self->state = QUOTE_IN_QUOTED_FIELD_IN_SKIP_LINE;
+        } else {
+          self->state = IN_FIELD_IN_SKIP_LINE;
+        }
+      }
+      break;
+
+    case QUOTE_IN_QUOTED_FIELD_IN_SKIP_LINE:
+      if (IS_QUOTE(c)) {
+        self->state = IN_QUOTED_FIELD_IN_SKIP_LINE;
+      } else if (IS_TERMINATOR(c)) {
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        self->file_lines++;
+        self->state = EAT_CRNL_NOP;
+      } else if (IS_DELIMITER(c)) {
+        self->state = START_FIELD_IN_SKIP_LINE;
+      } else {
+        self->state = IN_FIELD_IN_SKIP_LINE;
+      }
+      break;
+
+    case WHITESPACE_LINE:
+      if (IS_TERMINATOR(c)) {
+        self->file_lines++;
+        self->state = START_RECORD;
+        break;
+      } else if (IS_CARRIAGE(c)) {
+        self->file_lines++;
+        self->state = EAT_CRNL_NOP;
+        break;
+      } else if (!self->delim_whitespace) {
+        if (isblank(c) && c != self->delimiter) {
+        } else { // backtrack
+          // use i + 1 because buf has been incremented but not i
+          do {
+            --buf;
+            --i;
+          } while (i + 1 > self->datapos && !IS_TERMINATOR(*buf));
+
+          // reached a newline rather than the beginning
+          if (IS_TERMINATOR(*buf)) {
+            ++buf; // move pointer to first char after newline
+            ++i;
+          }
+          self->state = START_FIELD;
+        }
+        break;
+      }
+      // fall through
+
+    case EAT_WHITESPACE:
+      if (IS_TERMINATOR(c)) {
+        END_LINE();
+        self->state = START_RECORD;
+        break;
+      } else if (IS_CARRIAGE(c)) {
+        self->state = EAT_CRNL;
+        break;
+      } else if (IS_COMMENT_CHAR(c)) {
+        self->state = EAT_COMMENT;
+        break;
+      } else if (!isblank(c)) {
+        self->state = START_FIELD;
+        PD_FALLTHROUGH; // fall through to subsequent state
+      } else {
+        // if whitespace char, keep slurping
+        break;
+      }
+
+    case START_RECORD: {
+      // start of record
+      const int should_skip = skip_this_line(self, self->file_lines);
+
+      if (should_skip == -1) {
+        goto parsingerror;
+      } else if (should_skip) {
+        if (IS_QUOTE(c)) {
+          self->state = IN_QUOTED_FIELD_IN_SKIP_LINE;
+        } else {
+          self->state = IN_FIELD_IN_SKIP_LINE;
+
+          if (IS_TERMINATOR(c)) {
+            END_LINE();
+          }
+        }
+        break;
+      } else if (IS_TERMINATOR(c)) {
+        // \n\r possible?
+        if (self->skip_empty_lines) {
+          self->file_lines++;
+        } else {
+          END_LINE();
+        }
+        break;
+      } else if (IS_CARRIAGE(c)) {
+        if (self->skip_empty_lines) {
+          self->file_lines++;
+          self->state = EAT_CRNL_NOP;
+        } else {
+          self->state = EAT_CRNL;
+        }
+        break;
+      } else if (IS_COMMENT_CHAR(c)) {
+        self->state = EAT_LINE_COMMENT;
+        break;
+      } else if (isblank(c)) {
+        if (self->delim_whitespace) {
+          if (self->skip_empty_lines) {
+            self->state = WHITESPACE_LINE;
+          } else {
+            self->state = EAT_WHITESPACE;
+          }
+          break;
+        } else if (c != self->delimiter && self->skip_empty_lines) {
+          self->state = WHITESPACE_LINE;
+          break;
+        }
+      }
+
+      // normal character - fall through
+      // to handle as START_FIELD
+      self->state = START_FIELD;
+      PD_FALLTHROUGH;
+    }
+    case START_FIELD:
+      // expecting field
+      if (IS_TERMINATOR(c)) {
+        END_FIELD();
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        END_FIELD();
+        self->state = EAT_CRNL;
+      } else if (IS_QUOTE(c)) {
+        // start quoted field
+        self->state = IN_QUOTED_FIELD;
+      } else if (IS_ESCAPE_CHAR(c)) {
+        // possible escaped character
+        self->state = ESCAPED_CHAR;
+      } else if (IS_SKIPPABLE_SPACE(c)) {
+        // ignore space at start of field
+      } else if (IS_DELIMITER(c)) {
+        if (self->delim_whitespace) {
+          self->state = EAT_WHITESPACE;
+        } else {
+          // save empty field
+          END_FIELD();
+        }
+      } else if (IS_COMMENT_CHAR(c)) {
+        END_FIELD();
+        self->state = EAT_COMMENT;
+      } else {
+        // begin new unquoted field
+        PUSH_CHAR(c);
+        self->state = IN_FIELD;
+      }
+      break;
+
+    case ESCAPED_CHAR:
+      PUSH_CHAR(c);
+      self->state = IN_FIELD;
+      break;
+
+    case EAT_LINE_COMMENT:
+      if (IS_TERMINATOR(c)) {
+        self->file_lines++;
+        self->state = START_RECORD;
+      } else if (IS_CARRIAGE(c)) {
+        self->file_lines++;
+        self->state = EAT_CRNL_NOP;
+      }
+      break;
+
+    case IN_FIELD:
+      // in unquoted field
+      if (IS_TERMINATOR(c)) {
+        END_FIELD();
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        END_FIELD();
+        self->state = EAT_CRNL;
+      } else if (IS_ESCAPE_CHAR(c)) {
+        // possible escaped character
+        self->state = ESCAPED_CHAR;
+      } else if (IS_DELIMITER(c)) {
+        // end of field - end of line not reached yet
+        END_FIELD();
+
+        if (self->delim_whitespace) {
+          self->state = EAT_WHITESPACE;
+        } else {
+          self->state = START_FIELD;
+        }
+      } else if (IS_COMMENT_CHAR(c)) {
+        END_FIELD();
+        self->state = EAT_COMMENT;
+      } else {
+        // normal character - save in field
+        PUSH_CHAR(c);
+      }
+      break;
+
+    case IN_QUOTED_FIELD:
+      // in quoted field
+      if (IS_ESCAPE_CHAR(c)) {
+        // possible escape character
+        self->state = ESCAPE_IN_QUOTED_FIELD;
+      } else if (IS_QUOTE(c)) {
+        if (self->doublequote) {
+          // double quote - " represented by ""
+          self->state = QUOTE_IN_QUOTED_FIELD;
+        } else {
+          // end of quote part of field
+          self->state = IN_FIELD;
+        }
+      } else {
+        // normal character - save in field
+        PUSH_CHAR(c);
+      }
+      break;
+
+    case ESCAPE_IN_QUOTED_FIELD:
+      PUSH_CHAR(c);
+      self->state = IN_QUOTED_FIELD;
+      break;
+
+    case QUOTE_IN_QUOTED_FIELD:
+      // double quote - seen a quote in an quoted field
+      if (IS_QUOTE(c)) {
+        // save "" as "
+
+        PUSH_CHAR(c);
+        self->state = IN_QUOTED_FIELD;
+      } else if (IS_DELIMITER(c)) {
+        // end of field - end of line not reached yet
+        END_FIELD();
+
+        if (self->delim_whitespace) {
+          self->state = EAT_WHITESPACE;
+        } else {
+          self->state = START_FIELD;
+        }
+      } else if (IS_TERMINATOR(c)) {
+        END_FIELD();
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        END_FIELD();
+        self->state = EAT_CRNL;
+      } else {
+        PUSH_CHAR(c);
+        self->state = IN_FIELD;
+      }
+      break;
+
+    case EAT_COMMENT:
+      if (IS_TERMINATOR(c)) {
+        END_LINE();
+      } else if (IS_CARRIAGE(c)) {
+        self->state = EAT_CRNL;
+      }
+      break;
+
+    // only occurs with non-custom line terminator,
+    // which is why we directly check for '\n'
+    case EAT_CRNL:
+      if (c == '\n') {
+        END_LINE();
+      } else if (IS_DELIMITER(c)) {
+        if (self->delim_whitespace) {
+          END_LINE_STATE(EAT_WHITESPACE);
+        } else {
+          // Handle \r-delimited files
+          END_LINE_AND_FIELD_STATE(START_FIELD);
+        }
+      } else {
+        if (self->delim_whitespace) {
+          /* XXX
+           * first character of a new record--need to back up and
+           * reread
+           * to handle properly...
+           */
+          i--;
+          buf--; // back up one character (HACK!)
+          END_LINE_STATE(START_RECORD);
+        } else {
+          // \r line terminator
+          // UGH. we don't actually want
+          // to consume the token. fix this later
+          self->stream_len = slen;
+          if (end_line(self) < 0) {
+            goto parsingerror;
+          }
+
+          stream = self->stream + self->stream_len;
+          slen = self->stream_len;
+          self->state = START_RECORD;
+
+          --i;
+          buf--; // let's try this character again (HACK!)
+          if (line_limit > 0 && self->lines == start_lines + line_limit) {
+            goto linelimit;
+          }
+        }
+      }
+      break;
+
+    // only occurs with non-custom line terminator,
+    // which is why we directly check for '\n'
+    case EAT_CRNL_NOP: // inside an ignored comment line
+      self->state = START_RECORD;
+      // \r line terminator -- parse this character again
+      if (c != '\n' && !IS_DELIMITER(c)) {
+        --i;
+        --buf;
+      }
+      break;
+    default:
+      break;
+    }
+  }
+
+  _TOKEN_CLEANUP();
+
+  TRACE(("Finished tokenizing input\n"))
+
+  return 0;
+
+parsingerror:
+  i++;
+  _TOKEN_CLEANUP();
+
+  return -1;
+
+linelimit:
+  i++;
+  _TOKEN_CLEANUP();
+
+  return 0;
+}
+
+static int parser_handle_eof(parser_t *self) {
+  const size_t bufsize = 100;
+
+  TRACE(("handling eof, datalen: %d, pstate: %d\n", self->datalen, self->state))
+
+  if (self->datalen != 0)
+    return -1;
+
+  switch (self->state) {
+  case START_RECORD:
+  case WHITESPACE_LINE:
+  case EAT_CRNL_NOP:
+  case EAT_LINE_COMMENT:
+    return 0;
+
+  case ESCAPE_IN_QUOTED_FIELD:
+  case IN_QUOTED_FIELD:
+    self->error_msg = (char *)malloc(bufsize);
+    snprintf(self->error_msg, bufsize,
+             "EOF inside string starting at row %" PRIu64, self->file_lines);
+    return -1;
+
+  case ESCAPED_CHAR:
+    self->error_msg = (char *)malloc(bufsize);
+    snprintf(self->error_msg, bufsize, "EOF following escape character");
+    return -1;
+
+  case IN_FIELD:
+  case START_FIELD:
+  case QUOTE_IN_QUOTED_FIELD:
+    if (end_field(self) < 0)
+      return -1;
+    break;
+
+  default:
+    break;
+  }
+
+  if (end_line(self) < 0)
+    return -1;
+  else
+    return 0;
+}
+
+int parser_consume_rows(parser_t *self, size_t nrows) {
+  if (nrows > self->lines) {
+    nrows = self->lines;
+  }
+
+  /* do nothing */
+  if (nrows == 0)
+    return 0;
+
+  /* cannot guarantee that nrows + 1 has been observed */
+  const int64_t word_deletions =
+      self->line_start[nrows - 1] + self->line_fields[nrows - 1];
+
+  /* if word_deletions == 0 (i.e. this case) then char_count must
+   * be 0 too, as no data needs to be skipped */
+  const uint64_t char_count =
+      word_deletions >= 1 ? (self->word_starts[word_deletions - 1] +
+                             strlen(self->words[word_deletions - 1]) + 1)
+                          : 0;
+
+  TRACE(("parser_consume_rows: Deleting %d words, %d chars\n", word_deletions,
+         char_count));
+
+  /* move stream, only if something to move */
+  if (char_count < self->stream_len) {
+    memmove(self->stream, (self->stream + char_count),
+            self->stream_len - char_count);
+  }
+  /* buffer counts */
+  self->stream_len -= char_count;
+
+  /* move token metadata */
+  // Note: We should always have words_len < word_deletions, so this
+  //  subtraction will remain appropriately-typed.
+  int64_t offset;
+  for (uint64_t i = 0; i < self->words_len - word_deletions; ++i) {
+    offset = i + word_deletions;
+
+    self->words[i] = self->words[offset] - char_count;
+    self->word_starts[i] = self->word_starts[offset] - char_count;
+  }
+  self->words_len -= word_deletions;
+
+  /* move current word pointer to stream */
+  self->pword_start -= char_count;
+  self->word_start -= char_count;
+
+  /* move line metadata */
+  // Note: We should always have self->lines - nrows + 1 >= 0, so this
+  //  subtraction will remain appropriately-typed.
+  for (uint64_t i = 0; i < self->lines - nrows + 1; ++i) {
+    offset = i + nrows;
+    self->line_start[i] = self->line_start[offset] - word_deletions;
+    self->line_fields[i] = self->line_fields[offset];
+  }
+  self->lines -= nrows;
+
+  return 0;
+}
+
+static size_t _next_pow2(size_t sz) {
+  size_t result = 1;
+  while (result < sz)
+    result *= 2;
+  return result;
+}
+
+int parser_trim_buffers(parser_t *self) {
+  /*
+    Free memory
+   */
+
+  /**
+   * Before we free up space and trim, we should
+   * save how many words we saw when parsing, if
+   * it exceeds the maximum number we saw before.
+   *
+   * This is important for when we read in chunks,
+   * so that we can inform subsequent chunk parsing
+   * as to how many words we could possibly see.
+   */
+  if (self->words_cap > self->max_words_cap) {
+    self->max_words_cap = self->words_cap;
+  }
+
+  /* trim words, word_starts */
+  size_t new_cap = _next_pow2(self->words_len) + 1;
+  if (new_cap < self->words_cap) {
+    TRACE(("parser_trim_buffers: new_cap < self->words_cap\n"));
+    self->words = realloc(self->words, new_cap * sizeof(char *));
+    if (self->words == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    }
+    self->word_starts = realloc(self->word_starts, new_cap * sizeof(int64_t));
+    if (self->word_starts == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    }
+    self->words_cap = new_cap;
+  }
+
+  /* trim stream */
+  new_cap = _next_pow2(self->stream_len) + 1;
+  TRACE(("parser_trim_buffers: new_cap = %zu, stream_cap = %zu, lines_cap = "
+         "%zu\n",
+         new_cap, self->stream_cap, self->lines_cap));
+  if (new_cap < self->stream_cap) {
+    TRACE(("parser_trim_buffers: new_cap < self->stream_cap, calling "
+           "realloc\n"));
+    void *newptr = realloc(self->stream, new_cap);
+    if (newptr == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    } else {
+      // Update the pointers in the self->words array (char **) if
+      // `realloc`
+      //  moved the `self->stream` buffer. This block mirrors a similar
+      //  block in
+      //  `make_stream_space`.
+      if (self->stream != newptr) {
+        self->pword_start = (char *)newptr + self->word_start;
+
+        for (uint64_t i = 0; i < self->words_len; ++i) {
+          self->words[i] = (char *)newptr + self->word_starts[i];
+        }
+      }
+
+      self->stream = newptr;
+      self->stream_cap = new_cap;
+    }
+  }
+
+  /* trim line_start, line_fields */
+  new_cap = _next_pow2(self->lines) + 1;
+  if (new_cap < self->lines_cap) {
+    TRACE(("parser_trim_buffers: new_cap < self->lines_cap\n"));
+    void *newptr = realloc(self->line_start, new_cap * sizeof(int64_t));
+    if (newptr == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    } else {
+      self->line_start = newptr;
+    }
+    newptr = realloc(self->line_fields, new_cap * sizeof(int64_t));
+    if (newptr == NULL) {
+      return PARSER_OUT_OF_MEMORY;
+    } else {
+      self->line_fields = newptr;
+      self->lines_cap = new_cap;
+    }
+  }
+
+  return 0;
+}
+
+/*
+  nrows : number of rows to tokenize (or until reach EOF)
+  all : tokenize all the data vs. certain number of rows
+ */
+
+static int _tokenize_helper(parser_t *self, size_t nrows, int all,
+                            const char *encoding_errors) {
+  int status = 0;
+  const uint64_t start_lines = self->lines;
+
+  if (self->state == FINISHED) {
+    return 0;
+  }
+
+  TRACE(
+      ("_tokenize_helper: Asked to tokenize %d rows, datapos=%d, datalen=%d\n",
+       nrows, self->datapos, self->datalen));
+
+  while (1) {
+    if (!all && self->lines - start_lines >= nrows)
+      break;
+
+    if (self->datapos == self->datalen) {
+      status = parser_buffer_bytes(self, self->chunksize, encoding_errors);
+
+      if (status == REACHED_EOF) {
+        // close out last line
+        status = parser_handle_eof(self);
+        self->state = FINISHED;
+        break;
+      } else if (status != 0) {
+        return status;
+      }
+    }
+
+    TRACE(("_tokenize_helper: Trying to process %d bytes, datalen=%d, "
+           "datapos= %d\n",
+           self->datalen - self->datapos, self->datalen, self->datapos));
+
+    status = tokenize_bytes(self, nrows, start_lines);
+
+    if (status < 0) {
+      // XXX
+      TRACE(("_tokenize_helper: Status %d returned from tokenize_bytes, "
+             "breaking\n",
+             status));
+      status = -1;
+      break;
+    }
+  }
+  TRACE(("leaving tokenize_helper\n"));
+  return status;
+}
+
+int tokenize_nrows(parser_t *self, size_t nrows, const char *encoding_errors) {
+  return _tokenize_helper(self, nrows, 0, encoding_errors);
+}
+
+int tokenize_all_rows(parser_t *self, const char *encoding_errors) {
+  return _tokenize_helper(self, -1, 1, encoding_errors);
+}
+
+/*
+ * Function: to_boolean
+ * --------------------
+ *
+ * Validate if item should be recognized as a boolean field.
+ *
+ * item: const char* representing parsed text
+ * val : pointer to a uint8_t of boolean representation
+ *
+ * If item is determined to be boolean, this method will set
+ * the appropriate value of val and return 0. A non-zero exit
+ * status means that item was not inferred to be boolean, and
+ * leaves the value of *val unmodified.
+ */
+int to_boolean(const char *item, uint8_t *val) {
+  if (strcasecmp(item, "TRUE") == 0) {
+    *val = 1;
+    return 0;
+  } else if (strcasecmp(item, "FALSE") == 0) {
+    *val = 0;
+    return 0;
+  }
+
+  return -1;
+}
+
+// ---------------------------------------------------------------------------
+// Implementation of xstrtod
+
+//
+// strtod.c
+//
+// Convert string to double
+//
+// Copyright (C) 2002 Michael Ringgaard. All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions
+// are met:
+//
+// 1. Redistributions of source code must retain the above copyright
+//    notice, this list of conditions and the following disclaimer.
+// 2. Redistributions in binary form must reproduce the above copyright
+//    notice, this list of conditions and the following disclaimer in the
+//    documentation and/or other materials provided with the distribution.
+// 3. Neither the name of the project nor the names of its contributors
+//    may be used to endorse or promote products derived from this software
+//    without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+// AND
+// ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+// ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+// LIABLE
+// FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+// DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+// OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+// HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+// LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+// OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+// SUCH DAMAGE.
+//
+// -----------------------------------------------------------------------
+// Modifications by Warren Weckesser, March 2011:
+// * Rename strtod() to xstrtod().
+// * Added decimal and sci arguments.
+// * Skip trailing spaces.
+// * Commented out the other functions.
+// Modifications by Richard T Guy, August 2013:
+// * Add tsep argument for thousands separator
+//
+
+double xstrtod(const char *str, char **endptr, char decimal, char sci,
+               char tsep, int skip_trailing, int *error, int *maybe_int) {
+  const char *p = str;
+  if (maybe_int != NULL)
+    *maybe_int = 1;
+  // Skip leading whitespace.
+  while (isspace_ascii(*p))
+    p++;
+
+  // Handle optional sign.
+  int negative = 0;
+  switch (*p) {
+  case '-':
+    negative = 1;
+    PD_FALLTHROUGH; // Fall through to increment position.
+  case '+':
+    p++;
+    break;
+  }
+
+  int exponent = 0;
+  int num_digits = 0;
+  int num_decimals = 0;
+
+  // pessimistic but quick assessment,
+  // assuming that each decimal digit requires 4 bits to store
+  // TODO: C23 has UINT64_WIDTH macro that can be used at compile time
+  const int max_int_decimal_digits = (sizeof(unsigned int) * 8) / 4;
+
+  // Process string of digits.
+  unsigned int i_number = 0;
+  while (isdigit_ascii(*p) && num_digits <= max_int_decimal_digits) {
+    i_number = i_number * 10 + (*p - '0');
+    p++;
+    num_digits++;
+
+    p += (tsep != '\0' && *p == tsep);
+  }
+  double number = i_number;
+
+  if (num_digits > max_int_decimal_digits) {
+    // process what's left as double
+    while (isdigit_ascii(*p)) {
+      number = number * 10. + (*p - '0');
+      p++;
+      num_digits++;
+
+      p += (tsep != '\0' && *p == tsep);
+    }
+  }
+
+  // Process decimal part.
+  if (*p == decimal) {
+    if (maybe_int != NULL)
+      *maybe_int = 0;
+    p++;
+
+    while (isdigit_ascii(*p)) {
+      number = number * 10. + (*p - '0');
+      p++;
+      num_digits++;
+      num_decimals++;
+    }
+
+    exponent -= num_decimals;
+  }
+
+  if (num_digits == 0) {
+    *error = ERANGE;
+    return 0.0;
+  }
+
+  // Correct for sign.
+  if (negative)
+    number = -number;
+
+  // Process an exponent string.
+  if (toupper_ascii(*p) == toupper_ascii(sci)) {
+    if (maybe_int != NULL)
+      *maybe_int = 0;
+
+    // Handle optional sign.
+    negative = 0;
+    switch (*++p) {
+    case '-':
+      negative = 1;
+      PD_FALLTHROUGH; // Fall through to increment position.
+    case '+':
+      p++;
+      break;
+    }
+
+    // Process string of digits.
+    num_digits = 0;
+    int n = 0;
+    while (isdigit_ascii(*p)) {
+      n = n * 10 + (*p - '0');
+      num_digits++;
+      p++;
+    }
+
+    if (negative)
+      exponent -= n;
+    else
+      exponent += n;
+
+    // If no digits, after the 'e'/'E', un-consume it
+    if (num_digits == 0)
+      p--;
+  }
+
+  if (exponent < DBL_MIN_EXP || exponent > DBL_MAX_EXP) {
+    *error = ERANGE;
+    return HUGE_VAL;
+  }
+
+  // Scale the result.
+  double p10 = 10.;
+  int n = exponent;
+  if (n < 0)
+    n = -n;
+  while (n) {
+    if (n & 1) {
+      if (exponent < 0)
+        number /= p10;
+      else
+        number *= p10;
+    }
+    n >>= 1;
+    p10 *= p10;
+  }
+
+  if (number == HUGE_VAL) {
+    *error = ERANGE;
+  }
+
+  if (skip_trailing) {
+    // Skip trailing whitespace.
+    while (isspace_ascii(*p))
+      p++;
+  }
+
+  if (endptr)
+    *endptr = (char *)p;
+  return number;
+}
+
+double precise_xstrtod(const char *str, char **endptr, char decimal, char sci,
+                       char tsep, int skip_trailing, int *error,
+                       int *maybe_int) {
+  const char *p = str;
+  const int max_digits = 17;
+
+  if (maybe_int != NULL)
+    *maybe_int = 1;
+  // Cache powers of 10 in memory.
+  static double e[] = {
+      1.,    1e1,   1e2,   1e3,   1e4,   1e5,   1e6,   1e7,   1e8,   1e9,
+      1e10,  1e11,  1e12,  1e13,  1e14,  1e15,  1e16,  1e17,  1e18,  1e19,
+      1e20,  1e21,  1e22,  1e23,  1e24,  1e25,  1e26,  1e27,  1e28,  1e29,
+      1e30,  1e31,  1e32,  1e33,  1e34,  1e35,  1e36,  1e37,  1e38,  1e39,
+      1e40,  1e41,  1e42,  1e43,  1e44,  1e45,  1e46,  1e47,  1e48,  1e49,
+      1e50,  1e51,  1e52,  1e53,  1e54,  1e55,  1e56,  1e57,  1e58,  1e59,
+      1e60,  1e61,  1e62,  1e63,  1e64,  1e65,  1e66,  1e67,  1e68,  1e69,
+      1e70,  1e71,  1e72,  1e73,  1e74,  1e75,  1e76,  1e77,  1e78,  1e79,
+      1e80,  1e81,  1e82,  1e83,  1e84,  1e85,  1e86,  1e87,  1e88,  1e89,
+      1e90,  1e91,  1e92,  1e93,  1e94,  1e95,  1e96,  1e97,  1e98,  1e99,
+      1e100, 1e101, 1e102, 1e103, 1e104, 1e105, 1e106, 1e107, 1e108, 1e109,
+      1e110, 1e111, 1e112, 1e113, 1e114, 1e115, 1e116, 1e117, 1e118, 1e119,
+      1e120, 1e121, 1e122, 1e123, 1e124, 1e125, 1e126, 1e127, 1e128, 1e129,
+      1e130, 1e131, 1e132, 1e133, 1e134, 1e135, 1e136, 1e137, 1e138, 1e139,
+      1e140, 1e141, 1e142, 1e143, 1e144, 1e145, 1e146, 1e147, 1e148, 1e149,
+      1e150, 1e151, 1e152, 1e153, 1e154, 1e155, 1e156, 1e157, 1e158, 1e159,
+      1e160, 1e161, 1e162, 1e163, 1e164, 1e165, 1e166, 1e167, 1e168, 1e169,
+      1e170, 1e171, 1e172, 1e173, 1e174, 1e175, 1e176, 1e177, 1e178, 1e179,
+      1e180, 1e181, 1e182, 1e183, 1e184, 1e185, 1e186, 1e187, 1e188, 1e189,
+      1e190, 1e191, 1e192, 1e193, 1e194, 1e195, 1e196, 1e197, 1e198, 1e199,
+      1e200, 1e201, 1e202, 1e203, 1e204, 1e205, 1e206, 1e207, 1e208, 1e209,
+      1e210, 1e211, 1e212, 1e213, 1e214, 1e215, 1e216, 1e217, 1e218, 1e219,
+      1e220, 1e221, 1e222, 1e223, 1e224, 1e225, 1e226, 1e227, 1e228, 1e229,
+      1e230, 1e231, 1e232, 1e233, 1e234, 1e235, 1e236, 1e237, 1e238, 1e239,
+      1e240, 1e241, 1e242, 1e243, 1e244, 1e245, 1e246, 1e247, 1e248, 1e249,
+      1e250, 1e251, 1e252, 1e253, 1e254, 1e255, 1e256, 1e257, 1e258, 1e259,
+      1e260, 1e261, 1e262, 1e263, 1e264, 1e265, 1e266, 1e267, 1e268, 1e269,
+      1e270, 1e271, 1e272, 1e273, 1e274, 1e275, 1e276, 1e277, 1e278, 1e279,
+      1e280, 1e281, 1e282, 1e283, 1e284, 1e285, 1e286, 1e287, 1e288, 1e289,
+      1e290, 1e291, 1e292, 1e293, 1e294, 1e295, 1e296, 1e297, 1e298, 1e299,
+      1e300, 1e301, 1e302, 1e303, 1e304, 1e305, 1e306, 1e307, 1e308};
+
+  // Skip leading whitespace.
+  while (isspace_ascii(*p))
+    p++;
+
+  // Handle optional sign.
+  int negative = 0;
+  switch (*p) {
+  case '-':
+    negative = 1;
+    PD_FALLTHROUGH; // Fall through to increment position.
+  case '+':
+    p++;
+    break;
+  }
+
+  double number = 0.;
+  int exponent = 0;
+  int num_digits = 0;
+  int num_decimals = 0;
+
+  // Process string of digits.
+  while (isdigit_ascii(*p)) {
+    if (num_digits < max_digits) {
+      number = number * 10. + (*p - '0');
+      num_digits++;
+    } else {
+      ++exponent;
+    }
+
+    p++;
+    p += (tsep != '\0' && *p == tsep);
+  }
+
+  // Process decimal part
+  if (*p == decimal) {
+    if (maybe_int != NULL)
+      *maybe_int = 0;
+    p++;
+
+    while (num_digits < max_digits && isdigit_ascii(*p)) {
+      number = number * 10. + (*p - '0');
+      p++;
+      num_digits++;
+      num_decimals++;
+    }
+
+    if (num_digits >= max_digits) // Consume extra decimal digits.
+      while (isdigit_ascii(*p))
+        ++p;
+
+    exponent -= num_decimals;
+  }
+
+  if (num_digits == 0) {
+    *error = ERANGE;
+    return 0.0;
+  }
+
+  // Correct for sign.
+  if (negative)
+    number = -number;
+
+  // Process an exponent string.
+  if (toupper_ascii(*p) == toupper_ascii(sci)) {
+    if (maybe_int != NULL)
+      *maybe_int = 0;
+
+    // Handle optional sign
+    negative = 0;
+    switch (*++p) {
+    case '-':
+      negative = 1;
+      PD_FALLTHROUGH; // Fall through to increment position.
+    case '+':
+      p++;
+      break;
+    }
+
+    // Process string of digits.
+    num_digits = 0;
+    int n = 0;
+    while (num_digits < max_digits && isdigit_ascii(*p)) {
+      n = n * 10 + (*p - '0');
+      num_digits++;
+      p++;
+    }
+
+    if (negative)
+      exponent -= n;
+    else
+      exponent += n;
+
+    // If no digits after the 'e'/'E', un-consume it.
+    if (num_digits == 0)
+      p--;
+  }
+
+  if (exponent > 308) {
+    *error = ERANGE;
+    return HUGE_VAL;
+  } else if (exponent > 0) {
+    number *= e[exponent];
+  } else if (exponent < -308) { // Subnormal
+    if (exponent < -616) {      // Prevent invalid array access.
+      number = 0.;
+    } else {
+      number /= e[-308 - exponent];
+      number /= e[308];
+    }
+
+  } else {
+    number /= e[-exponent];
+  }
+
+  if (number == HUGE_VAL || number == -HUGE_VAL)
+    *error = ERANGE;
+
+  if (skip_trailing) {
+    // Skip trailing whitespace.
+    while (isspace_ascii(*p))
+      p++;
+  }
+
+  if (endptr)
+    *endptr = (char *)p;
+  return number;
+}
+
+/* copy a decimal number string with `decimal`, `tsep` as decimal point
+   and thousands separator to an equivalent c-locale decimal string (striping
+   `tsep`, replacing `decimal` with '.'). The returned memory should be free-d
+   with a call to `free`.
+*/
+
+static char *_str_copy_decimal_str_c(const char *s, char **endpos, char decimal,
+                                     char tsep) {
+  const char *p = s;
+  const size_t length = strlen(s);
+  char *s_copy = malloc(length + 1);
+  char *dst = s_copy;
+  // Skip leading whitespace.
+  while (isspace_ascii(*p))
+    p++;
+  // Copy Leading sign
+  if (*p == '+' || *p == '-') {
+    *dst++ = *p++;
+  }
+  // Copy integer part dropping `tsep`
+  while (isdigit_ascii(*p)) {
+    *dst++ = *p++;
+    p += (tsep != '\0' && *p == tsep);
+  }
+  // Replace `decimal` with '.'
+  if (*p == decimal) {
+    *dst++ = '.';
+    p++;
+  }
+  // Copy fractional part after decimal (if any)
+  while (isdigit_ascii(*p)) {
+    *dst++ = *p++;
+  }
+  // Copy exponent if any
+  if (toupper_ascii(*p) == toupper_ascii('E')) {
+    *dst++ = *p++;
+    // Copy leading exponent sign (if any)
+    if (*p == '+' || *p == '-') {
+      *dst++ = *p++;
+    }
+    // Copy exponent digits
+    while (isdigit_ascii(*p)) {
+      *dst++ = *p++;
+    }
+  }
+  *dst++ = '\0'; // terminate
+  if (endpos != NULL)
+    *endpos = (char *)p;
+  return s_copy;
+}
+
+double round_trip(const char *p, char **q, char decimal, char Py_UNUSED(sci),
+                  char tsep, int skip_trailing, int *error, int *maybe_int) {
+  // 'normalize' representation to C-locale; replace decimal with '.' and
+  // remove thousands separator.
+  char *endptr;
+  char *pc = _str_copy_decimal_str_c(p, &endptr, decimal, tsep);
+  // This is called from a nogil block in parsers.pyx
+  // so need to explicitly get GIL before Python calls
+  PyGILState_STATE gstate = PyGILState_Ensure();
+  char *endpc;
+  const double r = PyOS_string_to_double(pc, &endpc, 0);
+  // PyOS_string_to_double needs to consume the whole string
+  if (endpc == pc + strlen(pc)) {
+    if (q != NULL) {
+      // report endptr from source string (p)
+      *q = endptr;
+    }
+  } else {
+    *error = -1;
+    if (q != NULL) {
+      // p and pc are different len due to tsep removal. Can't report
+      // how much it has consumed of p. Just rewind to beginning.
+      *q = (char *)p; // TODO(willayd): this could be undefined behavior
+    }
+  }
+  if (maybe_int != NULL)
+    *maybe_int = 0;
+  if (PyErr_Occurred() != NULL)
+    *error = -1;
+  else if (r == Py_HUGE_VAL)
+    *error = (int)Py_HUGE_VAL;
+  PyErr_Clear();
+
+  PyGILState_Release(gstate);
+  free(pc);
+  if (skip_trailing && q != NULL && *q != p) {
+    while (isspace_ascii(**q)) {
+      (*q)++;
+    }
+  }
+  return r;
+}
+
+// End of xstrtod code
+// ---------------------------------------------------------------------------
+
+void uint_state_init(uint_state *self) {
+  self->seen_sint = 0;
+  self->seen_uint = 0;
+  self->seen_null = 0;
+}
+
+int uint64_conflict(uint_state *self) {
+  return self->seen_uint && (self->seen_sint || self->seen_null);
+}
+
+int64_t str_to_int64(const char *p_item, int64_t int_min, int64_t int_max,
+                     int *error, char tsep) {
+  const char *p = p_item;
+  // Skip leading spaces.
+  while (isspace_ascii(*p)) {
+    ++p;
+  }
+
+  // Handle sign.
+  const bool isneg = *p == '-' ? true : false;
+  // Handle sign.
+  if (isneg || (*p == '+')) {
+    p++;
+  }
+
+  // Check that there is a first digit.
+  if (!isdigit_ascii(*p)) {
+    // Error...
+    *error = ERROR_NO_DIGITS;
+    return 0;
+  }
+
+  int64_t number = 0;
+  if (isneg) {
+    // If number is greater than pre_min, at least one more digit
+    // can be processed without overflowing.
+    int dig_pre_min = -(int_min % 10);
+    int64_t pre_min = int_min / 10;
+
+    // Process the digits.
+    char d = *p;
+    if (tsep != '\0') {
+      while (1) {
+        if (d == tsep) {
+          d = *++p;
+          continue;
+        } else if (!isdigit_ascii(d)) {
+          break;
+        }
+        if ((number > pre_min) ||
+            ((number == pre_min) && (d - '0' <= dig_pre_min))) {
+          number = number * 10 - (d - '0');
+          d = *++p;
+        } else {
+          *error = ERROR_OVERFLOW;
+          return 0;
+        }
+      }
+    } else {
+      while (isdigit_ascii(d)) {
+        if ((number > pre_min) ||
+            ((number == pre_min) && (d - '0' <= dig_pre_min))) {
+          number = number * 10 - (d - '0');
+          d = *++p;
+        } else {
+          *error = ERROR_OVERFLOW;
+          return 0;
+        }
+      }
+    }
+  } else {
+    // If number is less than pre_max, at least one more digit
+    // can be processed without overflowing.
+    int64_t pre_max = int_max / 10;
+    int dig_pre_max = int_max % 10;
+
+    // Process the digits.
+    char d = *p;
+    if (tsep != '\0') {
+      while (1) {
+        if (d == tsep) {
+          d = *++p;
+          continue;
+        } else if (!isdigit_ascii(d)) {
+          break;
+        }
+        if ((number < pre_max) ||
+            ((number == pre_max) && (d - '0' <= dig_pre_max))) {
+          number = number * 10 + (d - '0');
+          d = *++p;
+
+        } else {
+          *error = ERROR_OVERFLOW;
+          return 0;
+        }
+      }
+    } else {
+      while (isdigit_ascii(d)) {
+        if ((number < pre_max) ||
+            ((number == pre_max) && (d - '0' <= dig_pre_max))) {
+          number = number * 10 + (d - '0');
+          d = *++p;
+
+        } else {
+          *error = ERROR_OVERFLOW;
+          return 0;
+        }
+      }
+    }
+  }
+
+  // Skip trailing spaces.
+  while (isspace_ascii(*p)) {
+    ++p;
+  }
+
+  // Did we use up all the characters?
+  if (*p) {
+    *error = ERROR_INVALID_CHARS;
+    return 0;
+  }
+
+  *error = 0;
+  return number;
+}
+
+uint64_t str_to_uint64(uint_state *state, const char *p_item, int64_t int_max,
+                       uint64_t uint_max, int *error, char tsep) {
+  const char *p = p_item;
+  // Skip leading spaces.
+  while (isspace_ascii(*p)) {
+    ++p;
+  }
+
+  // Handle sign.
+  if (*p == '-') {
+    state->seen_sint = 1;
+    *error = 0;
+    return 0;
+  } else if (*p == '+') {
+    p++;
+  }
+
+  // Check that there is a first digit.
+  if (!isdigit_ascii(*p)) {
+    // Error...
+    *error = ERROR_NO_DIGITS;
+    return 0;
+  }
+
+  // If number is less than pre_max, at least one more digit
+  // can be processed without overflowing.
+  //
+  // Process the digits.
+  uint64_t number = 0;
+  const uint64_t pre_max = uint_max / 10;
+  const uint64_t dig_pre_max = uint_max % 10;
+  char d = *p;
+  if (tsep != '\0') {
+    while (1) {
+      if (d == tsep) {
+        d = *++p;
+        continue;
+      } else if (!isdigit_ascii(d)) {
+        break;
+      }
+      if ((number < pre_max) ||
+          ((number == pre_max) && ((uint64_t)(d - '0') <= dig_pre_max))) {
+        number = number * 10 + (d - '0');
+        d = *++p;
+
+      } else {
+        *error = ERROR_OVERFLOW;
+        return 0;
+      }
+    }
+  } else {
+    while (isdigit_ascii(d)) {
+      if ((number < pre_max) ||
+          ((number == pre_max) && ((uint64_t)(d - '0') <= dig_pre_max))) {
+        number = number * 10 + (d - '0');
+        d = *++p;
+
+      } else {
+        *error = ERROR_OVERFLOW;
+        return 0;
+      }
+    }
+  }
+
+  // Skip trailing spaces.
+  while (isspace_ascii(*p)) {
+    ++p;
+  }
+
+  // Did we use up all the characters?
+  if (*p) {
+    *error = ERROR_INVALID_CHARS;
+    return 0;
+  }
+
+  if (number > (uint64_t)int_max) {
+    state->seen_uint = 1;
+  }
+
+  *error = 0;
+  return number;
+}
diff --git a/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c b/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
new file mode 100644
index 0000000000000..9a022095feee9
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
@@ -0,0 +1,855 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Copyright (c) 2005-2011, NumPy Developers
+All rights reserved.
+
+This file is derived from NumPy 1.7. See NUMPY_LICENSE.txt
+
+*/
+
+// Licence at LICENSES/NUMPY_LICENSE
+
+#ifndef NPY_NO_DEPRECATED_API
+#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
+#endif // NPY_NO_DEPRECATED_API
+
+#include "pandas/vendored/numpy/datetime/np_datetime.h"
+#define NO_IMPORT_ARRAY
+#define PY_ARRAY_UNIQUE_SYMBOL PANDAS_DATETIME_NUMPY
+#include <numpy/ndarrayobject.h>
+#include <numpy/npy_common.h>
+#include <stdbool.h>
+
+#if defined(_WIN32)
+#ifndef ENABLE_INTSAFE_SIGNED_FUNCTIONS
+#define ENABLE_INTSAFE_SIGNED_FUNCTIONS
+#endif
+#include <intsafe.h>
+#define checked_int64_add(a, b, res) LongLongAdd(a, b, res)
+#define checked_int64_sub(a, b, res) LongLongSub(a, b, res)
+#define checked_int64_mul(a, b, res) LongLongMult(a, b, res)
+#else
+#if defined __has_builtin
+#if __has_builtin(__builtin_add_overflow)
+#define checked_int64_add(a, b, res) __builtin_add_overflow(a, b, res)
+#define checked_int64_sub(a, b, res) __builtin_sub_overflow(a, b, res)
+#define checked_int64_mul(a, b, res) __builtin_mul_overflow(a, b, res)
+#else
+_Static_assert(0,
+               "Overflow checking not detected; please try a newer compiler");
+#endif
+// __has_builtin was added in gcc 10, but our muslinux_1_1 build environment
+// only has gcc-9.3, so fall back to __GNUC__ macro as long as we have that
+#elif __GNUC__ > 7
+#define checked_int64_add(a, b, res) __builtin_add_overflow(a, b, res)
+#define checked_int64_sub(a, b, res) __builtin_sub_overflow(a, b, res)
+#define checked_int64_mul(a, b, res) __builtin_mul_overflow(a, b, res)
+#else
+_Static_assert(0, "__has_builtin not detected; please try a newer compiler");
+#endif
+#endif
+
+#define XSTR(a) STR(a)
+#define STR(a) #a
+
+#define PD_CHECK_OVERFLOW(FUNC)                                                \
+  do {                                                                         \
+    if ((FUNC) != 0) {                                                         \
+      PyGILState_STATE gstate = PyGILState_Ensure();                           \
+      PyErr_SetString(PyExc_OverflowError,                                     \
+                      "Overflow occurred at " __FILE__ ":" XSTR(__LINE__));    \
+      PyGILState_Release(gstate);                                              \
+      return -1;                                                               \
+    }                                                                          \
+  } while (0)
+
+const int days_per_month_table[2][12] = {
+    {31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31},
+    {31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31}};
+
+/*
+ * Returns 1 if the given year is a leap year, 0 otherwise.
+ */
+int is_leapyear(npy_int64 year) {
+  return (year & 0x3) == 0 && /* year % 4 == 0 */
+         ((year % 100) != 0 || (year % 400) == 0);
+}
+
+/*
+ * Adjusts a datetimestruct based on a minutes offset. Assumes
+ * the current values are valid.g
+ */
+void add_minutes_to_datetimestruct(npy_datetimestruct *dts, int minutes) {
+  int isleap;
+
+  /* MINUTES */
+  dts->min += minutes;
+  while (dts->min < 0) {
+    dts->min += 60;
+    dts->hour--;
+  }
+  while (dts->min >= 60) {
+    dts->min -= 60;
+    dts->hour++;
+  }
+
+  /* HOURS */
+  while (dts->hour < 0) {
+    dts->hour += 24;
+    dts->day--;
+  }
+  while (dts->hour >= 24) {
+    dts->hour -= 24;
+    dts->day++;
+  }
+
+  /* DAYS */
+  if (dts->day < 1) {
+    dts->month--;
+    if (dts->month < 1) {
+      dts->year--;
+      dts->month = 12;
+    }
+    isleap = is_leapyear(dts->year);
+    dts->day += days_per_month_table[isleap][dts->month - 1];
+  } else if (dts->day > 28) {
+    isleap = is_leapyear(dts->year);
+    if (dts->day > days_per_month_table[isleap][dts->month - 1]) {
+      dts->day -= days_per_month_table[isleap][dts->month - 1];
+      dts->month++;
+      if (dts->month > 12) {
+        dts->year++;
+        dts->month = 1;
+      }
+    }
+  }
+}
+
+/*
+ * Calculates the days offset from the 1970 epoch.
+ */
+npy_int64 get_datetimestruct_days(const npy_datetimestruct *dts) {
+  int i, month;
+  npy_int64 year, days = 0;
+  const int *month_lengths;
+
+  PD_CHECK_OVERFLOW(checked_int64_sub(dts->year, 1970, &year));
+  PD_CHECK_OVERFLOW(checked_int64_mul(year, 365, &days));
+
+  /* Adjust for leap years */
+  if (days >= 0) {
+    /*
+     * 1968 is the closest leap year before 1970.
+     * Exclude the current year, so add 1.
+     */
+    PD_CHECK_OVERFLOW(checked_int64_add(year, 1, &year));
+    /* Add one day for each 4 years */
+    PD_CHECK_OVERFLOW(checked_int64_add(days, year / 4, &days));
+    /* 1900 is the closest previous year divisible by 100 */
+    PD_CHECK_OVERFLOW(checked_int64_add(year, 68, &year));
+    /* Subtract one day for each 100 years */
+    PD_CHECK_OVERFLOW(checked_int64_sub(days, year / 100, &days));
+    /* 1600 is the closest previous year divisible by 400 */
+    PD_CHECK_OVERFLOW(checked_int64_add(year, 300, &year));
+    /* Add one day for each 400 years */
+    PD_CHECK_OVERFLOW(checked_int64_add(days, year / 400, &days));
+  } else {
+    /*
+     * 1972 is the closest later year after 1970.
+     * Include the current year, so subtract 2.
+     */
+    PD_CHECK_OVERFLOW(checked_int64_sub(year, 2, &year));
+    /* Subtract one day for each 4 years */
+    PD_CHECK_OVERFLOW(checked_int64_add(days, year / 4, &days));
+    /* 2000 is the closest later year divisible by 100 */
+    PD_CHECK_OVERFLOW(checked_int64_sub(year, 28, &year));
+    /* Add one day for each 100 years */
+    PD_CHECK_OVERFLOW(checked_int64_sub(days, year / 100, &days));
+    /* 2000 is also the closest later year divisible by 400 */
+    /* Subtract one day for each 400 years */
+    PD_CHECK_OVERFLOW(checked_int64_add(days, year / 400, &days));
+  }
+
+  month_lengths = days_per_month_table[is_leapyear(dts->year)];
+  month = dts->month - 1;
+
+  /* Add the months */
+  for (i = 0; i < month; ++i) {
+    PD_CHECK_OVERFLOW(checked_int64_add(days, month_lengths[i], &days));
+  }
+
+  /* Add the days */
+  PD_CHECK_OVERFLOW(checked_int64_add(days, dts->day - 1, &days));
+
+  return days;
+}
+
+/*
+ * Modifies '*days_' to be the day offset within the year,
+ * and returns the year.
+ */
+static npy_int64 days_to_yearsdays(npy_int64 *days_) {
+  const npy_int64 days_per_400years = (400 * 365 + 100 - 4 + 1);
+  /* Adjust so it's relative to the year 2000 (divisible by 400) */
+  npy_int64 days = (*days_) - (365 * 30 + 7);
+  npy_int64 year;
+
+  /* Break down the 400 year cycle to get the year and day within the year */
+  if (days >= 0) {
+    year = 400 * (days / days_per_400years);
+    days = days % days_per_400years;
+  } else {
+    year = 400 * ((days - (days_per_400years - 1)) / days_per_400years);
+    days = days % days_per_400years;
+    if (days < 0) {
+      days += days_per_400years;
+    }
+  }
+
+  /* Work out the year/day within the 400 year cycle */
+  if (days >= 366) {
+    year += 100 * ((days - 1) / (100 * 365 + 25 - 1));
+    days = (days - 1) % (100 * 365 + 25 - 1);
+    if (days >= 365) {
+      year += 4 * ((days + 1) / (4 * 365 + 1));
+      days = (days + 1) % (4 * 365 + 1);
+      if (days >= 366) {
+        year += (days - 1) / 365;
+        days = (days - 1) % 365;
+      }
+    }
+  }
+
+  *days_ = days;
+  return year + 2000;
+}
+
+/*
+ * Fills in the year, month, day in 'dts' based on the days
+ * offset from 1970.
+ */
+static void set_datetimestruct_days(npy_int64 days, npy_datetimestruct *dts) {
+  const int *month_lengths;
+  int i;
+
+  dts->year = days_to_yearsdays(&days);
+  month_lengths = days_per_month_table[is_leapyear(dts->year)];
+
+  for (i = 0; i < 12; ++i) {
+    if (days < month_lengths[i]) {
+      dts->month = i + 1;
+      dts->day = (npy_int32)days + 1;
+      return;
+    } else {
+      days -= month_lengths[i];
+    }
+  }
+}
+
+/*
+ * Compares two npy_datetimestruct objects chronologically
+ */
+int cmp_npy_datetimestruct(const npy_datetimestruct *a,
+                           const npy_datetimestruct *b) {
+  if (a->year > b->year) {
+    return 1;
+  } else if (a->year < b->year) {
+    return -1;
+  }
+
+  if (a->month > b->month) {
+    return 1;
+  } else if (a->month < b->month) {
+    return -1;
+  }
+
+  if (a->day > b->day) {
+    return 1;
+  } else if (a->day < b->day) {
+    return -1;
+  }
+
+  if (a->hour > b->hour) {
+    return 1;
+  } else if (a->hour < b->hour) {
+    return -1;
+  }
+
+  if (a->min > b->min) {
+    return 1;
+  } else if (a->min < b->min) {
+    return -1;
+  }
+
+  if (a->sec > b->sec) {
+    return 1;
+  } else if (a->sec < b->sec) {
+    return -1;
+  }
+
+  if (a->us > b->us) {
+    return 1;
+  } else if (a->us < b->us) {
+    return -1;
+  }
+
+  if (a->ps > b->ps) {
+    return 1;
+  } else if (a->ps < b->ps) {
+    return -1;
+  }
+
+  if (a->as > b->as) {
+    return 1;
+  } else if (a->as < b->as) {
+    return -1;
+  }
+
+  return 0;
+}
+/*
+ * Returns the offset from utc of the timezone as a timedelta.
+ * The caller is responsible for ensuring that the tzinfo
+ * attribute exists on the datetime object.
+ *
+ * If the passed object is timezone naive, Py_None is returned.
+ * If extraction of the offset fails, NULL is returned.
+ *
+ * NOTE: This function is not vendored from numpy.
+ */
+PyObject *extract_utc_offset(PyObject *obj) {
+  PyObject *tmp = PyObject_GetAttrString(obj, "tzinfo");
+  if (tmp == NULL) {
+    return NULL;
+  }
+  if (tmp != Py_None) {
+    PyObject *offset = PyObject_CallMethod(tmp, "utcoffset", "O", obj);
+    if (offset == NULL) {
+      Py_DECREF(tmp);
+      return NULL;
+    }
+    return offset;
+  }
+  return tmp;
+}
+
+static inline int scaleYearToEpoch(int64_t year, int64_t *result) {
+  return checked_int64_sub(year, 1970, result);
+}
+
+static inline int scaleYearsToMonths(int64_t years, int64_t *result) {
+  return checked_int64_mul(years, 12, result);
+}
+
+static inline int scaleDaysToWeeks(int64_t days, int64_t *result) {
+  if (days >= 0) {
+    *result = days / 7;
+    return 0;
+  } else {
+    int res;
+    int64_t checked_days;
+    if ((res = checked_int64_sub(days, 6, &checked_days))) {
+      return res;
+    }
+
+    *result = checked_days / 7;
+    return 0;
+  }
+}
+
+static inline int scaleDaysToHours(int64_t days, int64_t *result) {
+  return checked_int64_mul(days, 24, result);
+}
+
+static inline int scaleHoursToMinutes(int64_t hours, int64_t *result) {
+  return checked_int64_mul(hours, 60, result);
+}
+
+static inline int scaleMinutesToSeconds(int64_t minutes, int64_t *result) {
+  return checked_int64_mul(minutes, 60, result);
+}
+
+static inline int scaleSecondsToMilliseconds(int64_t seconds, int64_t *result) {
+  return checked_int64_mul(seconds, 1000, result);
+}
+
+static inline int scaleSecondsToMicroseconds(int64_t seconds, int64_t *result) {
+  return checked_int64_mul(seconds, 1000000, result);
+}
+
+static inline int scaleMicrosecondsToNanoseconds(int64_t microseconds,
+                                                 int64_t *result) {
+  return checked_int64_mul(microseconds, 1000, result);
+}
+
+static inline int scaleMicrosecondsToPicoseconds(int64_t microseconds,
+                                                 int64_t *result) {
+  return checked_int64_mul(microseconds, 1000000, result);
+}
+
+static inline int64_t scalePicosecondsToFemtoseconds(int64_t picoseconds,
+                                                     int64_t *result) {
+  return checked_int64_mul(picoseconds, 1000, result);
+}
+
+static inline int64_t scalePicosecondsToAttoseconds(int64_t picoseconds,
+                                                    int64_t *result) {
+  return checked_int64_mul(picoseconds, 1000000, result);
+}
+
+/*
+ * Converts a datetime from a datetimestruct to a datetime based
+ * on a metadata unit. Returns -1 on and sets PyErr on error.
+ */
+npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
+                                            const npy_datetimestruct *dts) {
+  if ((base == NPY_FR_Y) || (base == NPY_FR_M)) {
+    int64_t years;
+    PD_CHECK_OVERFLOW(scaleYearToEpoch(dts->year, &years));
+
+    if (base == NPY_FR_Y) {
+      return years;
+    }
+
+    int64_t months;
+    PD_CHECK_OVERFLOW(scaleYearsToMonths(years, &months));
+
+    int64_t months_adder;
+    PD_CHECK_OVERFLOW(checked_int64_sub(dts->month, 1, &months_adder));
+    PD_CHECK_OVERFLOW(checked_int64_add(months, months_adder, &months));
+
+    if (base == NPY_FR_M) {
+      return months;
+    }
+  }
+
+  const int64_t days = get_datetimestruct_days(dts);
+  if (days == -1) {
+    PyGILState_STATE gstate = PyGILState_Ensure();
+    bool did_error = PyErr_Occurred() == NULL ? false : true;
+    PyGILState_Release(gstate);
+    if (did_error) {
+      return -1;
+    }
+  }
+
+  if (base == NPY_FR_D) {
+    return days;
+  }
+
+  if (base == NPY_FR_W) {
+    int64_t weeks;
+    PD_CHECK_OVERFLOW(scaleDaysToWeeks(days, &weeks));
+    return weeks;
+  }
+
+  int64_t hours;
+  PD_CHECK_OVERFLOW(scaleDaysToHours(days, &hours));
+  PD_CHECK_OVERFLOW(checked_int64_add(hours, dts->hour, &hours));
+
+  if (base == NPY_FR_h) {
+    return hours;
+  }
+
+  int64_t minutes;
+  PD_CHECK_OVERFLOW(scaleHoursToMinutes(hours, &minutes));
+  PD_CHECK_OVERFLOW(checked_int64_add(minutes, dts->min, &minutes));
+
+  if (base == NPY_FR_m) {
+    return minutes;
+  }
+
+  int64_t seconds;
+  PD_CHECK_OVERFLOW(scaleMinutesToSeconds(minutes, &seconds));
+  PD_CHECK_OVERFLOW(checked_int64_add(seconds, dts->sec, &seconds));
+
+  if (base == NPY_FR_s) {
+    return seconds;
+  }
+
+  if (base == NPY_FR_ms) {
+    int64_t milliseconds;
+    PD_CHECK_OVERFLOW(scaleSecondsToMilliseconds(seconds, &milliseconds));
+    PD_CHECK_OVERFLOW(
+        checked_int64_add(milliseconds, dts->us / 1000, &milliseconds));
+
+    return milliseconds;
+  }
+
+  int64_t microseconds;
+  PD_CHECK_OVERFLOW(scaleSecondsToMicroseconds(seconds, &microseconds));
+  PD_CHECK_OVERFLOW(checked_int64_add(microseconds, dts->us, &microseconds));
+
+  if (base == NPY_FR_us) {
+    return microseconds;
+  }
+
+  if (base == NPY_FR_ns) {
+    int64_t nanoseconds;
+
+    // Minimum valid timestamp in nanoseconds (1677-09-21 00:12:43.145224193).
+    const int64_t min_nanoseconds = NPY_MIN_INT64 + 1;
+    if (microseconds == min_nanoseconds / 1000 - 1) {
+      // For values within one microsecond of min_nanoseconds, use it as base
+      // and offset it with nanosecond delta to avoid overflow during scaling.
+      PD_CHECK_OVERFLOW(checked_int64_add(
+          min_nanoseconds, (dts->ps - _NS_MIN_DTS.ps) / 1000, &nanoseconds));
+    } else {
+      PD_CHECK_OVERFLOW(
+          scaleMicrosecondsToNanoseconds(microseconds, &nanoseconds));
+      PD_CHECK_OVERFLOW(
+          checked_int64_add(nanoseconds, dts->ps / 1000, &nanoseconds));
+    }
+
+    return nanoseconds;
+  }
+
+  int64_t picoseconds;
+  PD_CHECK_OVERFLOW(scaleMicrosecondsToPicoseconds(microseconds, &picoseconds));
+  PD_CHECK_OVERFLOW(checked_int64_add(picoseconds, dts->ps, &picoseconds));
+
+  if (base == NPY_FR_ps) {
+    return picoseconds;
+  }
+
+  if (base == NPY_FR_fs) {
+    int64_t femtoseconds;
+    PD_CHECK_OVERFLOW(
+        scalePicosecondsToFemtoseconds(picoseconds, &femtoseconds));
+    PD_CHECK_OVERFLOW(
+        checked_int64_add(femtoseconds, dts->as / 1000, &femtoseconds));
+    return femtoseconds;
+  }
+
+  if (base == NPY_FR_as) {
+    int64_t attoseconds;
+    PD_CHECK_OVERFLOW(scalePicosecondsToAttoseconds(picoseconds, &attoseconds));
+    PD_CHECK_OVERFLOW(checked_int64_add(attoseconds, dts->as, &attoseconds));
+    return attoseconds;
+  }
+
+  /* Something got corrupted */
+  PyGILState_STATE gstate = PyGILState_Ensure();
+  PyErr_SetString(PyExc_ValueError,
+                  "NumPy datetime metadata with corrupt unit value");
+  PyGILState_Release(gstate);
+
+  return -1;
+}
+
+/*
+ * Port numpy#13188 https://github.com/numpy/numpy/pull/13188/
+ *
+ * Computes the python `ret, d = divmod(d, unit)`.
+ *
+ * Note that GCC is smart enough at -O2 to eliminate the `if(*d < 0)` branch
+ * for subsequent calls to this command - it is able to deduce that `*d >= 0`.
+ */
+npy_int64 extract_unit(npy_datetime *d, npy_datetime unit) {
+  assert(unit > 0);
+  npy_int64 div = *d / unit;
+  npy_int64 mod = *d % unit;
+  if (mod < 0) {
+    mod += unit;
+    div -= 1;
+  }
+  assert(mod >= 0);
+  *d = mod;
+  return div;
+}
+
+/*
+ * Converts a datetime based on the given metadata into a datetimestruct
+ */
+void pandas_datetime_to_datetimestruct(npy_datetime dt, NPY_DATETIMEUNIT base,
+                                       npy_datetimestruct *out) {
+  npy_int64 perday;
+
+  /* Initialize the output to all zeros */
+  memset(out, 0, sizeof(npy_datetimestruct));
+  out->year = 1970;
+  out->month = 1;
+  out->day = 1;
+
+  /*
+   * Note that care must be taken with the / and % operators
+   * for negative values.
+   */
+  switch (base) {
+  case NPY_FR_Y:
+    out->year = 1970 + dt;
+    break;
+
+  case NPY_FR_M:
+    out->year = 1970 + extract_unit(&dt, 12);
+    out->month = (npy_int32)dt + 1;
+    break;
+
+  case NPY_FR_W:
+    /* A week is 7 days */
+    set_datetimestruct_days(dt * 7, out);
+    break;
+
+  case NPY_FR_D:
+    set_datetimestruct_days(dt, out);
+    break;
+
+  case NPY_FR_h:
+    perday = 24LL;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)dt;
+    break;
+
+  case NPY_FR_m:
+    perday = 24LL * 60;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)extract_unit(&dt, 60);
+    out->min = (npy_int32)dt;
+    break;
+
+  case NPY_FR_s:
+    perday = 24LL * 60 * 60;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)extract_unit(&dt, 60 * 60);
+    out->min = (npy_int32)extract_unit(&dt, 60);
+    out->sec = (npy_int32)dt;
+    break;
+
+  case NPY_FR_ms:
+    perday = 24LL * 60 * 60 * 1000;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)extract_unit(&dt, 1000LL * 60 * 60);
+    out->min = (npy_int32)extract_unit(&dt, 1000LL * 60);
+    out->sec = (npy_int32)extract_unit(&dt, 1000LL);
+    out->us = (npy_int32)(dt * 1000);
+    break;
+
+  case NPY_FR_us:
+    perday = 24LL * 60LL * 60LL * 1000LL * 1000LL;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 60 * 60);
+    out->min = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 60);
+    out->sec = (npy_int32)extract_unit(&dt, 1000LL * 1000);
+    out->us = (npy_int32)dt;
+    break;
+
+  case NPY_FR_ns:
+    perday = 24LL * 60LL * 60LL * 1000LL * 1000LL * 1000LL;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 60 * 60);
+    out->min = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 60);
+    out->sec = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000);
+    out->us = (npy_int32)extract_unit(&dt, 1000LL);
+    out->ps = (npy_int32)(dt * 1000);
+    break;
+
+  case NPY_FR_ps:
+    perday = 24LL * 60 * 60 * 1000 * 1000 * 1000 * 1000;
+
+    set_datetimestruct_days(extract_unit(&dt, perday), out);
+    out->hour =
+        (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 * 60 * 60);
+    out->min = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 * 60);
+    out->sec = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000);
+    out->us = (npy_int32)extract_unit(&dt, 1000LL * 1000);
+    out->ps = (npy_int32)(dt);
+    break;
+
+  case NPY_FR_fs:
+    /* entire range is only +- 2.6 hours */
+    out->hour = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 *
+                                                 1000 * 60 * 60);
+    if (out->hour < 0) {
+      out->year = 1969;
+      out->month = 12;
+      out->day = 31;
+      out->hour += 24;
+      assert(out->hour >= 0);
+    }
+    out->min =
+        (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 * 1000 * 60);
+    out->sec = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 * 1000);
+    out->us = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000);
+    out->ps = (npy_int32)extract_unit(&dt, 1000LL);
+    out->as = (npy_int32)(dt * 1000);
+    break;
+
+  case NPY_FR_as:
+    /* entire range is only +- 9.2 seconds */
+    out->sec =
+        (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000 * 1000 * 1000);
+    if (out->sec < 0) {
+      out->year = 1969;
+      out->month = 12;
+      out->day = 31;
+      out->hour = 23;
+      out->min = 59;
+      out->sec += 60;
+      assert(out->sec >= 0);
+    }
+    out->us = (npy_int32)extract_unit(&dt, 1000LL * 1000 * 1000 * 1000);
+    out->ps = (npy_int32)extract_unit(&dt, 1000LL * 1000);
+    out->as = (npy_int32)dt;
+    break;
+
+  default:
+    PyErr_SetString(PyExc_RuntimeError,
+                    "NumPy datetime metadata is corrupted with invalid "
+                    "base unit");
+  }
+}
+
+/*
+ * Converts a timedelta from a timedeltastruct to a timedelta based
+ * on a metadata unit. The timedelta is assumed to be valid.
+ *
+ * Returns 0 on success, -1 on failure.
+ */
+void pandas_timedelta_to_timedeltastruct(npy_timedelta td,
+                                         NPY_DATETIMEUNIT base,
+                                         pandas_timedeltastruct *out) {
+  /* Initialize the output to all zeros */
+  memset(out, 0, sizeof(pandas_timedeltastruct));
+
+  const npy_int64 sec_per_hour = 3600;
+  const npy_int64 sec_per_min = 60;
+
+  switch (base) {
+  case NPY_FR_W:
+    out->days = 7 * td;
+    break;
+  case NPY_FR_D:
+    out->days = td;
+    break;
+  case NPY_FR_h:
+    out->days = td / 24LL;
+    td -= out->days * 24LL;
+    out->hrs = (npy_int32)td;
+    break;
+  case NPY_FR_m:
+    out->days = td / 1440LL;
+    td -= out->days * 1440LL;
+    out->hrs = (npy_int32)(td / 60LL);
+    td -= out->hrs * 60LL;
+    out->min = (npy_int32)td;
+    break;
+  case NPY_FR_s:
+  case NPY_FR_ms:
+  case NPY_FR_us:
+  case NPY_FR_ns: {
+    const npy_int64 sec_per_day = 86400;
+    npy_int64 per_sec;
+    if (base == NPY_FR_s) {
+      per_sec = 1;
+    } else if (base == NPY_FR_ms) {
+      per_sec = 1000;
+    } else if (base == NPY_FR_us) {
+      per_sec = 1000000;
+    } else {
+      per_sec = 1000000000;
+    }
+
+    const npy_int64 per_day = sec_per_day * per_sec;
+    npy_int64 frac;
+    // put frac in seconds
+    if (td < 0 && td % per_sec != 0)
+      frac = td / per_sec - 1;
+    else
+      frac = td / per_sec;
+
+    const int sign = frac < 0 ? -1 : 1;
+    if (frac < 0) {
+      // even fraction
+      if ((-frac % sec_per_day) != 0) {
+        out->days = -frac / sec_per_day + 1;
+        frac += sec_per_day * out->days;
+      } else {
+        frac = -frac;
+      }
+    }
+
+    if (frac >= sec_per_day) {
+      out->days += frac / sec_per_day;
+      frac -= out->days * sec_per_day;
+    }
+
+    if (frac >= sec_per_hour) {
+      out->hrs = (npy_int32)(frac / sec_per_hour);
+      frac -= out->hrs * sec_per_hour;
+    }
+
+    if (frac >= sec_per_min) {
+      out->min = (npy_int32)(frac / sec_per_min);
+      frac -= out->min * sec_per_min;
+    }
+
+    if (frac >= 0) {
+      out->sec = (npy_int32)frac;
+      frac -= out->sec;
+    }
+
+    if (sign < 0)
+      out->days = -out->days;
+
+    if (base > NPY_FR_s) {
+      const npy_int64 sfrac =
+          (out->hrs * sec_per_hour + out->min * sec_per_min + out->sec) *
+          per_sec;
+
+      npy_int64 ifrac = td - (out->days * per_day + sfrac);
+
+      if (base == NPY_FR_ms) {
+        out->ms = (npy_int32)ifrac;
+      } else if (base == NPY_FR_us) {
+        out->ms = (npy_int32)(ifrac / 1000LL);
+        ifrac = ifrac % 1000LL;
+        out->us = (npy_int32)ifrac;
+      } else if (base == NPY_FR_ns) {
+        out->ms = (npy_int32)(ifrac / (1000LL * 1000LL));
+        ifrac = ifrac % (1000LL * 1000LL);
+        out->us = (npy_int32)(ifrac / 1000LL);
+        ifrac = ifrac % 1000LL;
+        out->ns = (npy_int32)ifrac;
+      }
+    }
+
+  } break;
+  default:
+    PyErr_SetString(PyExc_RuntimeError,
+                    "NumPy timedelta metadata is corrupted with "
+                    "invalid base unit");
+    break;
+  }
+
+  out->seconds =
+      (npy_int32)(out->hrs * sec_per_hour + out->min * sec_per_min + out->sec);
+  out->microseconds = out->ms * 1000 + out->us;
+  out->nanoseconds = out->ns;
+}
+
+/*
+ * This function returns a pointer to the DateTimeMetaData
+ * contained within the provided datetime dtype.
+ *
+ * Copied near-verbatim from numpy/core/src/multiarray/datetime.c
+ */
+PyArray_DatetimeMetaData
+get_datetime_metadata_from_dtype(PyArray_Descr *dtype) {
+#if NPY_ABI_VERSION < 0x02000000
+#define PyDataType_C_METADATA(dtype) ((dtype)->c_metadata)
+#endif
+  return ((PyArray_DatetimeDTypeMetaData *)PyDataType_C_METADATA(dtype))->meta;
+}
diff --git a/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime_strings.c b/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime_strings.c
new file mode 100644
index 0000000000000..a46f5bc467c5d
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/numpy/datetime/np_datetime_strings.c
@@ -0,0 +1,1168 @@
+/*
+
+Copyright (c) 2016, PyData Development Team
+All rights reserved.
+
+Distributed under the terms of the BSD Simplified License.
+
+The full license is in the LICENSE file, distributed with this software.
+
+Written by Mark Wiebe (mwwiebe@gmail.com)
+Copyright (c) 2011 by Enthought, Inc.
+
+Copyright (c) 2005-2011, NumPy Developers
+All rights reserved.
+
+See NUMPY_LICENSE.txt for the license.
+
+This file implements string parsing and creation for NumPy datetime.
+
+*/
+
+// LICENSES/NUMPY_LICENSE
+
+#define PY_SSIZE_T_CLEAN
+#define NO_IMPORT
+
+#ifndef NPY_NO_DEPRECATED_API
+#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
+#endif // NPY_NO_DEPRECATED_API
+
+#include <Python.h>
+
+#include <time.h>
+
+#include <numpy/ndarraytypes.h>
+#include <numpy/npy_common.h>
+
+#include "pandas/portable.h"
+#include "pandas/vendored/numpy/datetime/np_datetime.h"
+#include "pandas/vendored/numpy/datetime/np_datetime_strings.h"
+
+/*
+ * Parses (almost) standard ISO 8601 date strings. The differences are:
+ *
+ * + Only seconds may have a decimal point, with up to 18 digits after it
+ *   (maximum attoseconds precision).
+ * + Either a 'T' as in ISO 8601 or a ' ' may be used to separate
+ *   the date and the time. Both are treated equivalently.
+ * + Doesn't (yet) handle the "YYYY-DDD" or "YYYY-Www" formats.
+ * + Doesn't handle leap seconds (seconds value has 60 in these cases).
+ * + Doesn't handle 24:00:00 as synonym for midnight (00:00:00) tomorrow
+ * + Accepts special values "NaT" (not a time), "Today", (current
+ *   day according to local time) and "Now" (current time in UTC).
+ * + ':' separator between hours, minutes, and seconds is optional. When
+ *   omitted, each component must be 2 digits if it appears. (GH-10041)
+ *
+ * 'str' must be a NULL-terminated string, and 'len' must be its length.
+ *
+ * 'out' gets filled with the parsed date-time.
+ * 'out_local' gets set to 1 if the parsed time contains timezone,
+ *      to 0 otherwise.
+ * 'out_tzoffset' gets set to timezone offset by minutes
+ *      if the parsed time was in local time,
+ *      to 0 otherwise. The values 'now' and 'today' don't get counted
+ *      as local, and neither do UTC +/-#### timezone offsets, because
+ *      they aren't using the computer's local timezone offset.
+ *
+ * Returns 0 on success, -1 on failure.
+ */
+
+typedef enum {
+  COMPARISON_SUCCESS,
+  COMPLETED_PARTIAL_MATCH,
+  COMPARISON_ERROR
+} DatetimePartParseResult;
+// This function will advance the pointer on format
+// and decrement characters_remaining by n on success
+// On failure will return COMPARISON_ERROR without incrementing
+// If `format_requirement` is PARTIAL_MATCH, and the `format` string has
+// been exhausted, then return COMPLETED_PARTIAL_MATCH.
+static DatetimePartParseResult
+compare_format(const char **format, int *characters_remaining,
+               const char *compare_to, int n,
+               const FormatRequirement format_requirement) {
+  if (format_requirement == INFER_FORMAT) {
+    return COMPARISON_SUCCESS;
+  }
+  if (*characters_remaining < 0) {
+    return COMPARISON_ERROR;
+  }
+  if (format_requirement == PARTIAL_MATCH && *characters_remaining == 0) {
+    return COMPLETED_PARTIAL_MATCH;
+  }
+  if (*characters_remaining < n) {
+    // TODO(pandas-dev): PyErr to differentiate what went wrong
+    return COMPARISON_ERROR;
+  } else {
+    if (strncmp(*format, compare_to, n)) {
+      // TODO(pandas-dev): PyErr to differentiate what went wrong
+      return COMPARISON_ERROR;
+    } else {
+      *format += n;
+      *characters_remaining -= n;
+      return COMPARISON_SUCCESS;
+    }
+  }
+  return COMPARISON_SUCCESS;
+}
+
+int parse_iso_8601_datetime(const char *str, int len, int want_exc,
+                            npy_datetimestruct *out,
+                            NPY_DATETIMEUNIT *out_bestunit, int *out_local,
+                            int *out_tzoffset, const char *format,
+                            int format_len,
+                            FormatRequirement format_requirement) {
+  if (len < 0 || format_len < 0)
+    goto parse_error;
+  int year_leap = 0;
+  int i, numdigits;
+  const char *substr;
+  int sublen;
+  NPY_DATETIMEUNIT bestunit = NPY_FR_GENERIC;
+  DatetimePartParseResult comparison;
+
+  /* If year-month-day are separated by a valid separator,
+   * months/days without leading zeroes will be parsed
+   * (though not iso8601). If the components aren't separated,
+   * 4 (YYYY) or 8 (YYYYMMDD) digits are expected. 6 digits are
+   * forbidden here (but parsed as YYMMDD elsewhere).
+   */
+  int has_ymd_sep = 0;
+  char ymd_sep = '\0';
+  char valid_ymd_sep[] = {'-', '.', '/', '\\', ' '};
+  int valid_ymd_sep_len = sizeof(valid_ymd_sep);
+
+  /* hour-minute-second may or may not separated by ':'. If not, then
+   * each component must be 2 digits. */
+  int has_hms_sep = 0;
+  int hour_was_2_digits = 0;
+
+  /* Initialize the output to all zeros */
+  memset(out, 0, sizeof(npy_datetimestruct));
+  out->month = 1;
+  out->day = 1;
+
+  substr = str;
+  sublen = len;
+
+  /* Skip leading whitespace */
+  while (sublen > 0 && isspace(*substr)) {
+    ++substr;
+    --sublen;
+    comparison =
+        compare_format(&format, &format_len, " ", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  }
+
+  /* Leading '-' sign for negative year */
+  if (*substr == '-') {
+    ++substr;
+    --sublen;
+  }
+
+  if (sublen == 0) {
+    goto parse_error;
+  }
+
+  /* PARSE THE YEAR (4 digits) */
+  comparison =
+      compare_format(&format, &format_len, "%Y", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+
+  out->year = 0;
+  if (sublen >= 4 && isdigit(substr[0]) && isdigit(substr[1]) &&
+      isdigit(substr[2]) && isdigit(substr[3])) {
+    out->year = 1000 * (substr[0] - '0') + 100 * (substr[1] - '0') +
+                10 * (substr[2] - '0') + (substr[3] - '0');
+
+    substr += 4;
+    sublen -= 4;
+  }
+
+  /* Negate the year if necessary */
+  if (str[0] == '-') {
+    out->year = -out->year;
+  }
+  /* Check whether it's a leap-year */
+  year_leap = is_leapyear(out->year);
+
+  /* Next character must be a separator, start of month, or end of string */
+  if (sublen == 0) {
+    if (out_local != NULL) {
+      *out_local = 0;
+    }
+    if (format_len) {
+      goto parse_error;
+    }
+    bestunit = NPY_FR_Y;
+    goto finish;
+  }
+
+  if (!isdigit(*substr)) {
+    for (i = 0; i < valid_ymd_sep_len; ++i) {
+      if (*substr == valid_ymd_sep[i]) {
+        break;
+      }
+    }
+    if (i == valid_ymd_sep_len) {
+      goto parse_error;
+    }
+    has_ymd_sep = 1;
+    ymd_sep = valid_ymd_sep[i];
+    ++substr;
+    --sublen;
+
+    comparison =
+        compare_format(&format, &format_len, &ymd_sep, 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+    /* Cannot have trailing separator */
+    if (sublen == 0 || !isdigit(*substr)) {
+      goto parse_error;
+    }
+  }
+
+  /* PARSE THE MONTH */
+  comparison =
+      compare_format(&format, &format_len, "%m", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  /* First digit required */
+  out->month = (*substr - '0');
+  ++substr;
+  --sublen;
+  /* Second digit optional if there was a separator */
+  if (isdigit(*substr)) {
+    out->month = 10 * out->month + (*substr - '0');
+    ++substr;
+    --sublen;
+  } else if (!has_ymd_sep) {
+    goto parse_error;
+  }
+  if (out->month < 1 || out->month > 12) {
+    if (want_exc) {
+      PyErr_Format(PyExc_ValueError,
+                   "Month out of range in datetime string \"%s\"", str);
+    }
+    goto error;
+  }
+
+  /* Next character must be the separator, start of day, or end of string */
+  if (sublen == 0) {
+    bestunit = NPY_FR_M;
+    /* Forbid YYYYMM. Parsed instead as YYMMDD by someone else. */
+    if (!has_ymd_sep) {
+      goto parse_error;
+    }
+    if (format_len) {
+      goto parse_error;
+    }
+    if (out_local != NULL) {
+      *out_local = 0;
+    }
+    goto finish;
+  }
+
+  if (has_ymd_sep) {
+    /* Must have separator, but cannot be trailing */
+    if (*substr != ymd_sep || sublen == 1) {
+      goto parse_error;
+    }
+    ++substr;
+    --sublen;
+    comparison =
+        compare_format(&format, &format_len, &ymd_sep, 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  }
+
+  /* PARSE THE DAY */
+  comparison =
+      compare_format(&format, &format_len, "%d", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  /* First digit required */
+  if (!isdigit(*substr)) {
+    goto parse_error;
+  }
+  out->day = (*substr - '0');
+  ++substr;
+  --sublen;
+  /* Second digit optional if there was a separator */
+  if (isdigit(*substr)) {
+    out->day = 10 * out->day + (*substr - '0');
+    ++substr;
+    --sublen;
+  } else if (!has_ymd_sep) {
+    goto parse_error;
+  }
+  if (out->day < 1 ||
+      out->day > days_per_month_table[year_leap][out->month - 1]) {
+    if (want_exc) {
+      PyErr_Format(PyExc_ValueError,
+                   "Day out of range in datetime string \"%s\"", str);
+    }
+    goto error;
+  }
+
+  /* Next character must be a 'T', ' ', or end of string */
+  if (sublen == 0) {
+    if (out_local != NULL) {
+      *out_local = 0;
+    }
+    if (format_len) {
+      goto parse_error;
+    }
+    bestunit = NPY_FR_D;
+    goto finish;
+  }
+
+  if ((*substr != 'T' && *substr != ' ') || sublen == 1) {
+    goto parse_error;
+  }
+  comparison =
+      compare_format(&format, &format_len, substr, 1, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  ++substr;
+  --sublen;
+
+  /* PARSE THE HOURS */
+  comparison =
+      compare_format(&format, &format_len, "%H", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  /* First digit required */
+  if (!isdigit(*substr)) {
+    goto parse_error;
+  }
+  out->hour = (*substr - '0');
+  bestunit = NPY_FR_h;
+  ++substr;
+  --sublen;
+  /* Second digit optional */
+  if (isdigit(*substr)) {
+    hour_was_2_digits = 1;
+    out->hour = 10 * out->hour + (*substr - '0');
+    ++substr;
+    --sublen;
+    if (out->hour >= 24) {
+      if (want_exc) {
+        PyErr_Format(PyExc_ValueError,
+                     "Hours out of range in datetime string \"%s\"", str);
+      }
+      goto error;
+    }
+  }
+
+  /* Next character must be a ':' or the end of the string */
+  if (sublen == 0) {
+    if (!hour_was_2_digits) {
+      goto parse_error;
+    }
+    if (format_len) {
+      goto parse_error;
+    }
+    bestunit = NPY_FR_h;
+    goto finish;
+  }
+
+  if (*substr == ':') {
+    has_hms_sep = 1;
+    ++substr;
+    --sublen;
+    /* Cannot have a trailing separator */
+    if (sublen == 0 || !isdigit(*substr)) {
+      goto parse_error;
+    }
+    comparison =
+        compare_format(&format, &format_len, ":", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  } else if (!isdigit(*substr)) {
+    if (!hour_was_2_digits) {
+      goto parse_error;
+    }
+    goto parse_timezone;
+  }
+
+  /* PARSE THE MINUTES */
+  comparison =
+      compare_format(&format, &format_len, "%M", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  /* First digit required */
+  out->min = (*substr - '0');
+  bestunit = NPY_FR_m;
+  ++substr;
+  --sublen;
+  /* Second digit optional if there was a separator */
+  if (isdigit(*substr)) {
+    out->min = 10 * out->min + (*substr - '0');
+    ++substr;
+    --sublen;
+    if (out->min >= 60) {
+      if (want_exc) {
+        PyErr_Format(PyExc_ValueError,
+                     "Minutes out of range in datetime string \"%s\"", str);
+      }
+      goto error;
+    }
+  } else if (!has_hms_sep) {
+    goto parse_error;
+  }
+
+  if (sublen == 0) {
+    bestunit = NPY_FR_m;
+    if (format_len) {
+      goto parse_error;
+    }
+    goto finish;
+  }
+
+  /* If we make it through this condition block, then the next
+   * character is a digit. */
+  if (has_hms_sep && *substr == ':') {
+    comparison =
+        compare_format(&format, &format_len, ":", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+    ++substr;
+    --sublen;
+    /* Cannot have a trailing ':' */
+    if (sublen == 0 || !isdigit(*substr)) {
+      goto parse_error;
+    }
+  } else if (!has_hms_sep && isdigit(*substr)) {
+  } else {
+    goto parse_timezone;
+  }
+
+  /* PARSE THE SECONDS */
+  comparison =
+      compare_format(&format, &format_len, "%S", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  /* First digit required */
+  out->sec = (*substr - '0');
+  ++substr;
+  --sublen;
+  /* Second digit optional if there was a separator */
+  if (isdigit(*substr)) {
+    out->sec = 10 * out->sec + (*substr - '0');
+    ++substr;
+    --sublen;
+    if (out->sec >= 60) {
+      if (want_exc) {
+        PyErr_Format(PyExc_ValueError,
+                     "Seconds out of range in datetime string \"%s\"", str);
+      }
+      goto error;
+    }
+  } else if (!has_hms_sep) {
+    goto parse_error;
+  }
+
+  /* Next character may be a '.' indicating fractional seconds */
+  if (sublen > 0 && *substr == '.') {
+    ++substr;
+    --sublen;
+    comparison =
+        compare_format(&format, &format_len, ".", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  } else {
+    bestunit = NPY_FR_s;
+    goto parse_timezone;
+  }
+
+  /* PARSE THE MICROSECONDS (0 to 6 digits) */
+  comparison =
+      compare_format(&format, &format_len, "%f", 2, format_requirement);
+  if (comparison == COMPARISON_ERROR) {
+    goto parse_error;
+  } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+    goto finish;
+  }
+  numdigits = 0;
+  for (i = 0; i < 6; ++i) {
+    out->us *= 10;
+    if (sublen > 0 && isdigit(*substr)) {
+      out->us += (*substr - '0');
+      ++substr;
+      --sublen;
+      ++numdigits;
+    }
+  }
+
+  if (sublen == 0 || !isdigit(*substr)) {
+    if (numdigits > 3) {
+      bestunit = NPY_FR_us;
+    } else {
+      bestunit = NPY_FR_ms;
+    }
+    goto parse_timezone;
+  }
+
+  /* PARSE THE PICOSECONDS (0 to 6 digits) */
+  numdigits = 0;
+  for (i = 0; i < 6; ++i) {
+    out->ps *= 10;
+    if (sublen > 0 && isdigit(*substr)) {
+      out->ps += (*substr - '0');
+      ++substr;
+      --sublen;
+      ++numdigits;
+    }
+  }
+
+  if (sublen == 0 || !isdigit(*substr)) {
+    if (numdigits > 3) {
+      bestunit = NPY_FR_ps;
+    } else {
+      bestunit = NPY_FR_ns;
+    }
+    goto parse_timezone;
+  }
+
+  /* PARSE THE ATTOSECONDS (0 to 6 digits) */
+  numdigits = 0;
+  for (i = 0; i < 6; ++i) {
+    out->as *= 10;
+    if (sublen > 0 && isdigit(*substr)) {
+      out->as += (*substr - '0');
+      ++substr;
+      --sublen;
+      ++numdigits;
+    }
+  }
+
+  if (numdigits > 3) {
+    bestunit = NPY_FR_as;
+  } else {
+    bestunit = NPY_FR_fs;
+  }
+
+parse_timezone:
+  /* trim any whitespace between time/timezone */
+  while (sublen > 0 && isspace(*substr)) {
+    ++substr;
+    --sublen;
+    comparison =
+        compare_format(&format, &format_len, " ", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  }
+
+  if (sublen == 0) {
+    // Unlike NumPy, treating no time zone as naive
+    if (format_len > 0) {
+      goto parse_error;
+    }
+    goto finish;
+  }
+
+  /* UTC specifier */
+  if (*substr == 'Z') {
+    comparison =
+        compare_format(&format, &format_len, "%z", 2, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+    /* "Z" should be equivalent to tz offset "+00:00" */
+    if (out_local != NULL) {
+      *out_local = 1;
+    }
+
+    if (out_tzoffset != NULL) {
+      *out_tzoffset = 0;
+    }
+
+    if (sublen == 1) {
+      if (format_len > 0) {
+        goto parse_error;
+      }
+      goto finish;
+    } else {
+      ++substr;
+      --sublen;
+    }
+  } else if (*substr == '-' || *substr == '+') {
+    comparison =
+        compare_format(&format, &format_len, "%z", 2, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+    /* Time zone offset */
+    int offset_neg = 0, offset_hour = 0, offset_minute = 0;
+
+    /*
+     * Since "local" means local with respect to the current
+     * machine, we say this is non-local.
+     */
+
+    if (*substr == '-') {
+      offset_neg = 1;
+    }
+    ++substr;
+    --sublen;
+
+    /* The hours offset */
+    if (sublen >= 2 && isdigit(substr[0]) && isdigit(substr[1])) {
+      offset_hour = 10 * (substr[0] - '0') + (substr[1] - '0');
+      substr += 2;
+      sublen -= 2;
+      if (offset_hour >= 24) {
+        if (want_exc) {
+          PyErr_Format(PyExc_ValueError,
+                       "Timezone hours offset out of range "
+                       "in datetime string \"%s\"",
+                       str);
+        }
+        goto error;
+      }
+    } else if (sublen >= 1 && isdigit(substr[0])) {
+      offset_hour = substr[0] - '0';
+      ++substr;
+      --sublen;
+    } else {
+      goto parse_error;
+    }
+
+    /* The minutes offset is optional */
+    if (sublen > 0) {
+      /* Optional ':' */
+      if (*substr == ':') {
+        ++substr;
+        --sublen;
+      }
+
+      /* The minutes offset (at the end of the string) */
+      if (sublen >= 2 && isdigit(substr[0]) && isdigit(substr[1])) {
+        offset_minute = 10 * (substr[0] - '0') + (substr[1] - '0');
+        substr += 2;
+        sublen -= 2;
+        if (offset_minute >= 60) {
+          if (want_exc) {
+            PyErr_Format(PyExc_ValueError,
+                         "Timezone minutes offset out of range "
+                         "in datetime string \"%s\"",
+                         str);
+          }
+          goto error;
+        }
+      } else if (sublen >= 1 && isdigit(substr[0])) {
+        offset_minute = substr[0] - '0';
+        ++substr;
+        --sublen;
+      } else {
+        goto parse_error;
+      }
+    }
+
+    /* Apply the time zone offset */
+    if (offset_neg) {
+      offset_hour = -offset_hour;
+      offset_minute = -offset_minute;
+    }
+    if (out_local != NULL) {
+      *out_local = 1;
+      // Unlike NumPy, do not change internal value to local time
+      *out_tzoffset = 60 * offset_hour + offset_minute;
+    }
+  }
+
+  /* Skip trailing whitespace */
+  while (sublen > 0 && isspace(*substr)) {
+    ++substr;
+    --sublen;
+    comparison =
+        compare_format(&format, &format_len, " ", 1, format_requirement);
+    if (comparison == COMPARISON_ERROR) {
+      goto parse_error;
+    } else if (comparison == COMPLETED_PARTIAL_MATCH) {
+      goto finish;
+    }
+  }
+
+  if ((sublen != 0) || (format_len != 0)) {
+    goto parse_error;
+  }
+
+finish:
+  if (out_bestunit != NULL) {
+    *out_bestunit = bestunit;
+  }
+  return 0;
+
+parse_error:
+  if (want_exc) {
+    PyErr_Format(PyExc_ValueError,
+                 "Error parsing datetime string \"%s\" at position %d", str,
+                 (int)(substr - str));
+  }
+  return -1;
+
+error:
+  return -1;
+}
+
+/*
+ * Provides a string length to use for converting datetime
+ * objects with the given local and unit settings.
+ */
+int get_datetime_iso_8601_strlen(int local, NPY_DATETIMEUNIT base) {
+  int len = 0;
+
+  switch (base) {
+  /* Generic units can only be used to represent NaT */
+  /*    return 4;*/
+  case NPY_FR_as:
+    len += 3; /* "###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_fs:
+    len += 3; /* "###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_ps:
+    len += 3; /* "###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_ns:
+    len += 3; /* "###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_us:
+    len += 3; /* "###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_ms:
+    len += 4; /* ".###" */
+    PD_FALLTHROUGH;
+  case NPY_FR_s:
+    len += 3; /* ":##" */
+    PD_FALLTHROUGH;
+  case NPY_FR_m:
+    len += 3; /* ":##" */
+    PD_FALLTHROUGH;
+  case NPY_FR_h:
+    len += 3; /* "T##" */
+    PD_FALLTHROUGH;
+  case NPY_FR_D:
+  case NPY_FR_W:
+    len += 3; /* "-##" */
+    PD_FALLTHROUGH;
+  case NPY_FR_M:
+    len += 3; /* "-##" */
+    PD_FALLTHROUGH;
+  case NPY_FR_Y:
+    len += 21; /* 64-bit year */
+    break;
+  default:
+    len += 3; /* handle the now defunct NPY_FR_B */
+    break;
+  }
+
+  if (base >= NPY_FR_h) {
+    if (local) {
+      len += 5; /* "+####" or "-####" */
+    } else {
+      len += 1; /* "Z" */
+    }
+  }
+
+  len += 1; /* NULL terminator */
+
+  return len;
+}
+
+/*
+ * Converts an npy_datetimestruct to an (almost) ISO 8601
+ * NULL-terminated string using timezone Z (UTC). If the string fits in
+ * the space exactly, it leaves out the NULL terminator and returns success.
+ *
+ * The differences from ISO 8601 are the 'NaT' string, and
+ * the number of year digits is >= 4 instead of strictly 4.
+ *
+ * 'base' restricts the output to that unit. Set 'base' to
+ * -1 to auto-detect a base after which all the values are zero.
+ *
+ *  Returns 0 on success, -1 on failure (for example if the output
+ *  string was too short).
+ */
+int make_iso_8601_datetime(npy_datetimestruct *dts, char *outstr, size_t outlen,
+                           int utc, NPY_DATETIMEUNIT base) {
+  char *substr = outstr;
+  size_t sublen = outlen;
+  int tmplen;
+
+  /*
+   * Print weeks with the same precision as days.
+   *
+   * TODO: Could print weeks with YYYY-Www format if the week
+   *       epoch is a Monday.
+   */
+  if (base == NPY_FR_W) {
+    base = NPY_FR_D;
+  }
+
+/* YEAR */
+/*
+ * Can't use PyOS_snprintf, because it always produces a '\0'
+ * character at the end, and NumPy string types are permitted
+ * to have data all the way to the end of the buffer.
+ */
+#ifdef _WIN32
+  tmplen = _snprintf(substr, sublen, "%04" NPY_INT64_FMT, dts->year);
+#else
+  tmplen = snprintf(substr, sublen, "%04" NPY_INT64_FMT, dts->year);
+#endif // _WIN32
+  /* If it ran out of space or there isn't space for the NULL terminator */
+  if (tmplen < 0 || (size_t)tmplen > sublen) {
+    goto string_too_short;
+  }
+  substr += tmplen;
+  sublen -= tmplen;
+
+  /* Stop if the unit is years */
+  if (base == NPY_FR_Y) {
+    if (sublen > 0) {
+      *substr = '\0';
+    }
+    return 0;
+  }
+
+  /* MONTH */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = '-';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->month / 10) + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->month % 10) + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is months */
+  if (base == NPY_FR_M) {
+    if (sublen > 0) {
+      *substr = '\0';
+    }
+    return 0;
+  }
+
+  /* DAY */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = '-';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->day / 10) + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->day % 10) + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is days */
+  if (base == NPY_FR_D) {
+    if (sublen > 0) {
+      *substr = '\0';
+    }
+    return 0;
+  }
+
+  /* HOUR */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = 'T';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->hour / 10) + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->hour % 10) + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is hours */
+  if (base == NPY_FR_h) {
+    goto add_time_zone;
+  }
+
+  /* MINUTE */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = ':';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->min / 10) + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->min % 10) + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is minutes */
+  if (base == NPY_FR_m) {
+    goto add_time_zone;
+  }
+
+  /* SECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = ':';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->sec / 10) + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->sec % 10) + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is seconds */
+  if (base == NPY_FR_s) {
+    goto add_time_zone;
+  }
+
+  /* MILLISECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = '.';
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->us / 100000) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->us / 10000) % 10 + '0');
+  if (sublen < 4) {
+    goto string_too_short;
+  }
+  substr[3] = (char)((dts->us / 1000) % 10 + '0');
+  substr += 4;
+  sublen -= 4;
+
+  /* Stop if the unit is milliseconds */
+  if (base == NPY_FR_ms) {
+    goto add_time_zone;
+  }
+
+  /* MICROSECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = (char)((dts->us / 100) % 10 + '0');
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->us / 10) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)(dts->us % 10 + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is microseconds */
+  if (base == NPY_FR_us) {
+    goto add_time_zone;
+  }
+
+  /* NANOSECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = (char)((dts->ps / 100000) % 10 + '0');
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->ps / 10000) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->ps / 1000) % 10 + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is nanoseconds */
+  if (base == NPY_FR_ns) {
+    goto add_time_zone;
+  }
+
+  /* PICOSECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = (char)((dts->ps / 100) % 10 + '0');
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->ps / 10) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)(dts->ps % 10 + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is picoseconds */
+  if (base == NPY_FR_ps) {
+    goto add_time_zone;
+  }
+
+  /* FEMTOSECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = (char)((dts->as / 100000) % 10 + '0');
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->as / 10000) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)((dts->as / 1000) % 10 + '0');
+  substr += 3;
+  sublen -= 3;
+
+  /* Stop if the unit is femtoseconds */
+  if (base == NPY_FR_fs) {
+    goto add_time_zone;
+  }
+
+  /* ATTOSECOND */
+  if (sublen < 1) {
+    goto string_too_short;
+  }
+  substr[0] = (char)((dts->as / 100) % 10 + '0');
+  if (sublen < 2) {
+    goto string_too_short;
+  }
+  substr[1] = (char)((dts->as / 10) % 10 + '0');
+  if (sublen < 3) {
+    goto string_too_short;
+  }
+  substr[2] = (char)(dts->as % 10 + '0');
+  substr += 3;
+  sublen -= 3;
+
+add_time_zone:
+  /* UTC "Zulu" time */
+  if (utc) {
+    if (sublen < 1) {
+      goto string_too_short;
+    }
+    substr[0] = 'Z';
+    substr += 1;
+    sublen -= 1;
+  }
+  /* Add a NULL terminator, and return */
+  if (sublen > 0) {
+    substr[0] = '\0';
+  }
+
+  return 0;
+
+string_too_short:
+  PyErr_Format(PyExc_RuntimeError,
+               "The string provided for NumPy ISO datetime formatting "
+               "was too short, with length %d",
+               outlen);
+  return -1;
+}
+
+int make_iso_8601_timedelta(pandas_timedeltastruct *tds, char *outstr,
+                            size_t *outlen) {
+  *outlen = 0;
+  *outlen += snprintf(outstr, 60, // NOLINT
+                      "P%" NPY_INT64_FMT "DT%" NPY_INT32_FMT "H%" NPY_INT32_FMT
+                      "M%" NPY_INT32_FMT,
+                      tds->days, tds->hrs, tds->min, tds->sec);
+  outstr += *outlen;
+
+  if (tds->ns != 0) {
+    *outlen += snprintf(outstr, 12, // NOLINT
+                        ".%03" NPY_INT32_FMT "%03" NPY_INT32_FMT
+                        "%03" NPY_INT32_FMT "S",
+                        tds->ms, tds->us, tds->ns);
+  } else if (tds->us != 0) {
+    *outlen += snprintf(outstr, 9, // NOLINT
+                        ".%03" NPY_INT32_FMT "%03" NPY_INT32_FMT "S", tds->ms,
+                        tds->us);
+  } else if (tds->ms != 0) {
+    *outlen += snprintf(outstr, 6, // NOLINT
+                        ".%03" NPY_INT32_FMT "S", tds->ms);
+  } else {
+    *outlen += snprintf(outstr, 2, // NOLINT
+                        "%s", "S");
+  }
+
+  return 0;
+}
diff --git a/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsondec.c b/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsondec.c
new file mode 100644
index 0000000000000..bf389b4dce1d0
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsondec.c
@@ -0,0 +1,1221 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+* Neither the name of the ESN Social Software AB nor the
+names of its contributors may be used to endorse or promote products
+derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+* Copyright (c) 1988-1993 The Regents of the University of California.
+* Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+// Licence at LICENSES/ULTRAJSON_LICENSE
+
+#include "pandas/vendored/ujson/lib/ultrajson.h"
+#include <errno.h>
+#include <limits.h>
+#include <locale.h>
+#include <math.h>
+#include <stdlib.h>
+#include <string.h>
+#include <wchar.h>
+
+#ifndef TRUE
+#define TRUE 1
+#define FALSE 0
+#endif
+#ifndef NULL
+#define NULL 0
+#endif
+
+struct DecoderState {
+  char *start;
+  char *end;
+  wchar_t *escStart;
+  wchar_t *escEnd;
+  int escHeap;
+  int lastType;
+  JSUINT32 objDepth;
+  void *prv;
+  JSONObjectDecoder *dec;
+};
+
+JSOBJ FASTCALL_MSVC decode_any(struct DecoderState *ds);
+typedef JSOBJ (*PFN_DECODER)(struct DecoderState *ds);
+
+static JSOBJ SetError(struct DecoderState *ds, int offset,
+                      const char *message) {
+  ds->dec->errorOffset = ds->start + offset;
+  ds->dec->errorStr = (char *)message;
+  return NULL;
+}
+
+double createDouble(double intNeg, double intValue, double frcValue,
+                    int frcDecimalCount) {
+  static const double g_pow10[] = {1.0,
+                                   0.1,
+                                   0.01,
+                                   0.001,
+                                   0.0001,
+                                   0.00001,
+                                   0.000001,
+                                   0.0000001,
+                                   0.00000001,
+                                   0.000000001,
+                                   0.0000000001,
+                                   0.00000000001,
+                                   0.000000000001,
+                                   0.0000000000001,
+                                   0.00000000000001,
+                                   0.000000000000001};
+  return (intValue + (frcValue * g_pow10[frcDecimalCount])) * intNeg;
+}
+
+JSOBJ FASTCALL_MSVC decodePreciseFloat(struct DecoderState *ds) {
+  char *end;
+  double value;
+  errno = 0;
+
+  value = strtod(ds->start, &end);
+
+  if (errno == ERANGE) {
+    return SetError(ds, -1, "Range error when decoding numeric as double");
+  }
+
+  ds->start = end;
+  return ds->dec->newDouble(ds->prv, value);
+}
+
+JSOBJ FASTCALL_MSVC decode_numeric(struct DecoderState *ds) {
+  int intNeg = 1;
+  JSUINT64 intValue;
+  JSUINT64 prevIntValue;
+  int chr;
+  int decimalCount = 0;
+  double frcValue = 0.0;
+  double expNeg;
+  double expValue;
+  char *offset = ds->start;
+
+  JSUINT64 overflowLimit = LLONG_MAX;
+
+  if (*(offset) == 'I') {
+    goto DECODE_INF;
+  } else if (*(offset) == 'N') {
+    goto DECODE_NAN;
+  } else if (*(offset) == '-') {
+    offset++;
+    intNeg = -1;
+    overflowLimit = LLONG_MIN;
+    if (*(offset) == 'I') {
+      goto DECODE_INF;
+    }
+  }
+
+  // Scan integer part
+  intValue = 0;
+
+  while (1) {
+    chr = (int)(unsigned char)*(offset);
+
+    switch (chr) {
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+    case '8':
+    case '9': {
+      // PERF: Don't do 64-bit arithmetic here unless we have to
+      prevIntValue = intValue;
+      intValue = intValue * 10ULL + (JSLONG)(chr - 48);
+
+      if (intNeg == 1 && prevIntValue > intValue) {
+        return SetError(ds, -1, "Value is too big!");
+      } else if (intNeg == -1 && intValue > overflowLimit) {
+        return SetError(ds, -1,
+                        overflowLimit == LLONG_MAX ? "Value is too big!"
+                                                   : "Value is too small");
+      }
+
+      offset++;
+      break;
+    }
+    case '.': {
+      offset++;
+      goto DECODE_FRACTION;
+      break;
+    }
+    case 'e':
+    case 'E': {
+      offset++;
+      goto DECODE_EXPONENT;
+      break;
+    }
+
+    default: {
+      goto BREAK_INT_LOOP;
+      break;
+    }
+    }
+  }
+
+BREAK_INT_LOOP:
+
+  ds->lastType = JT_INT;
+  ds->start = offset;
+
+  if (intNeg == 1 && (intValue & 0x8000000000000000ULL) != 0)
+    return ds->dec->newUnsignedLong(ds->prv, intValue);
+  else if ((intValue >> 31))
+    return ds->dec->newLong(ds->prv, (JSINT64)(intValue * (JSINT64)intNeg));
+  else
+    return ds->dec->newInt(ds->prv, (JSINT32)(intValue * intNeg));
+
+DECODE_FRACTION:
+
+  if (ds->dec->preciseFloat) {
+    return decodePreciseFloat(ds);
+  }
+
+  // Scan fraction part
+  frcValue = 0.0;
+  for (;;) {
+    chr = (int)(unsigned char)*(offset);
+
+    switch (chr) {
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+    case '8':
+    case '9': {
+      if (decimalCount < JSON_DOUBLE_MAX_DECIMALS) {
+        frcValue = frcValue * 10.0 + (double)(chr - 48);
+        decimalCount++;
+      }
+      offset++;
+      break;
+    }
+    case 'e':
+    case 'E': {
+      offset++;
+      goto DECODE_EXPONENT;
+      break;
+    }
+    default: {
+      goto BREAK_FRC_LOOP;
+    }
+    }
+  }
+
+BREAK_FRC_LOOP:
+  // FIXME: Check for arithmetic overflow here
+  ds->lastType = JT_DOUBLE;
+  ds->start = offset;
+  return ds->dec->newDouble(
+      ds->prv,
+      createDouble((double)intNeg, (double)intValue, frcValue, decimalCount));
+
+DECODE_EXPONENT:
+  if (ds->dec->preciseFloat) {
+    return decodePreciseFloat(ds);
+  }
+
+  expNeg = 1.0;
+
+  if (*(offset) == '-') {
+    expNeg = -1.0;
+    offset++;
+  } else if (*(offset) == '+') {
+    expNeg = +1.0;
+    offset++;
+  }
+
+  expValue = 0.0;
+
+  for (;;) {
+    chr = (int)(unsigned char)*(offset);
+
+    switch (chr) {
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+    case '8':
+    case '9': {
+      expValue = expValue * 10.0 + (double)(chr - 48);
+      offset++;
+      break;
+    }
+    default: {
+      goto BREAK_EXP_LOOP;
+    }
+    }
+  }
+
+DECODE_NAN:
+  offset++;
+  if (*(offset++) != 'a')
+    goto SET_NAN_ERROR;
+  if (*(offset++) != 'N')
+    goto SET_NAN_ERROR;
+
+  ds->lastType = JT_NULL;
+  ds->start = offset;
+  return ds->dec->newNull(ds->prv);
+
+SET_NAN_ERROR:
+  return SetError(ds, -1, "Unexpected character found when decoding 'NaN'");
+
+DECODE_INF:
+  offset++;
+  if (*(offset++) != 'n')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 'f')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 'i')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 'n')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 'i')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 't')
+    goto SET_INF_ERROR;
+  if (*(offset++) != 'y')
+    goto SET_INF_ERROR;
+
+  ds->start = offset;
+
+  if (intNeg == 1) {
+    ds->lastType = JT_POS_INF;
+    return ds->dec->newPosInf(ds->prv);
+  } else {
+    ds->lastType = JT_NEG_INF;
+    return ds->dec->newNegInf(ds->prv);
+  }
+
+SET_INF_ERROR:
+  if (intNeg == 1) {
+    const char *msg = "Unexpected character found when decoding 'Infinity'";
+    return SetError(ds, -1, msg);
+  } else {
+    const char *msg = "Unexpected character found when decoding '-Infinity'";
+    return SetError(ds, -1, msg);
+  }
+
+BREAK_EXP_LOOP:
+  // FIXME: Check for arithmetic overflow here
+  ds->lastType = JT_DOUBLE;
+  ds->start = offset;
+  return ds->dec->newDouble(
+      ds->prv,
+      createDouble((double)intNeg, (double)intValue, frcValue, decimalCount) *
+          pow(10.0, expValue * expNeg));
+}
+
+JSOBJ FASTCALL_MSVC decode_true(struct DecoderState *ds) {
+  char *offset = ds->start;
+  offset++;
+
+  if (*(offset++) != 'r')
+    goto SETERROR;
+  if (*(offset++) != 'u')
+    goto SETERROR;
+  if (*(offset++) != 'e')
+    goto SETERROR;
+
+  ds->lastType = JT_TRUE;
+  ds->start = offset;
+  return ds->dec->newTrue(ds->prv);
+
+SETERROR:
+  return SetError(ds, -1, "Unexpected character found when decoding 'true'");
+}
+
+JSOBJ FASTCALL_MSVC decode_false(struct DecoderState *ds) {
+  char *offset = ds->start;
+  offset++;
+
+  if (*(offset++) != 'a')
+    goto SETERROR;
+  if (*(offset++) != 'l')
+    goto SETERROR;
+  if (*(offset++) != 's')
+    goto SETERROR;
+  if (*(offset++) != 'e')
+    goto SETERROR;
+
+  ds->lastType = JT_FALSE;
+  ds->start = offset;
+  return ds->dec->newFalse(ds->prv);
+
+SETERROR:
+  return SetError(ds, -1, "Unexpected character found when decoding 'false'");
+}
+
+JSOBJ FASTCALL_MSVC decode_null(struct DecoderState *ds) {
+  char *offset = ds->start;
+  offset++;
+
+  if (*(offset++) != 'u')
+    goto SETERROR;
+  if (*(offset++) != 'l')
+    goto SETERROR;
+  if (*(offset++) != 'l')
+    goto SETERROR;
+
+  ds->lastType = JT_NULL;
+  ds->start = offset;
+  return ds->dec->newNull(ds->prv);
+
+SETERROR:
+  return SetError(ds, -1, "Unexpected character found when decoding 'null'");
+}
+
+void FASTCALL_MSVC SkipWhitespace(struct DecoderState *ds) {
+  char *offset;
+
+  for (offset = ds->start; (ds->end - offset) > 0; offset++) {
+    switch (*offset) {
+    case ' ':
+    case '\t':
+    case '\r':
+    case '\n':
+      break;
+
+    default:
+      ds->start = offset;
+      return;
+    }
+  }
+
+  if (offset == ds->end) {
+    ds->start = ds->end;
+  }
+}
+
+enum DECODESTRINGSTATE {
+  DS_ISNULL = 0x32,
+  DS_ISQUOTE,
+  DS_ISESCAPE,
+  DS_UTFLENERROR,
+};
+
+static const JSUINT8 g_decoderLookup[256] = {
+    /* 0x00 */ DS_ISNULL,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x10 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x20 */ 1,
+    1,
+    DS_ISQUOTE,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x30 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x40 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x50 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    DS_ISESCAPE,
+    1,
+    1,
+    1,
+    /* 0x60 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x70 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x80 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x90 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xa0 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xb0 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xc0 */ 2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    /* 0xd0 */ 2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    /* 0xe0 */ 3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    /* 0xf0 */ 4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+    DS_UTFLENERROR,
+};
+
+JSOBJ FASTCALL_MSVC decode_string(struct DecoderState *ds) {
+  JSUTF16 sur[2] = {0};
+  int iSur = 0;
+  int index;
+  wchar_t *escOffset;
+  wchar_t *escStart;
+  size_t escLen = (ds->escEnd - ds->escStart);
+  JSUINT8 *inputOffset;
+  JSUINT8 oct;
+  JSUTF32 ucs;
+  ds->lastType = JT_INVALID;
+  ds->start++;
+
+  if ((size_t)(ds->end - ds->start) > escLen) {
+    size_t newSize = (ds->end - ds->start);
+
+    if (ds->escHeap) {
+      if (newSize > (SIZE_MAX / sizeof(wchar_t))) {
+        return SetError(ds, -1, "Could not reserve memory block");
+      }
+      escStart =
+          (wchar_t *)ds->dec->realloc(ds->escStart, newSize * sizeof(wchar_t));
+      if (!escStart) {
+        ds->dec->free(ds->escStart);
+        return SetError(ds, -1, "Could not reserve memory block");
+      }
+      ds->escStart = escStart;
+    } else {
+      wchar_t *oldStart = ds->escStart;
+      if (newSize > (SIZE_MAX / sizeof(wchar_t))) {
+        return SetError(ds, -1, "Could not reserve memory block");
+      }
+      ds->escStart = (wchar_t *)ds->dec->malloc(newSize * sizeof(wchar_t));
+      if (!ds->escStart) {
+        return SetError(ds, -1, "Could not reserve memory block");
+      }
+      ds->escHeap = 1;
+      memcpy(ds->escStart, oldStart, escLen * sizeof(wchar_t));
+    }
+
+    ds->escEnd = ds->escStart + newSize;
+  }
+
+  escOffset = ds->escStart;
+  inputOffset = (JSUINT8 *)ds->start;
+
+  for (;;) {
+    switch (g_decoderLookup[(JSUINT8)(*inputOffset)]) {
+    case DS_ISNULL: {
+      return SetError(ds, -1, "Unmatched ''\"' when when decoding 'string'");
+    }
+    case DS_ISQUOTE: {
+      ds->lastType = JT_UTF8;
+      inputOffset++;
+      ds->start += ((char *)inputOffset - (ds->start));
+      return ds->dec->newString(ds->prv, ds->escStart, escOffset);
+    }
+    case DS_UTFLENERROR: {
+      return SetError(ds, -1,
+                      "Invalid UTF-8 sequence length when decoding 'string'");
+    }
+    case DS_ISESCAPE:
+      inputOffset++;
+      switch (*inputOffset) {
+      case '\\':
+        *(escOffset++) = L'\\';
+        inputOffset++;
+        continue;
+      case '\"':
+        *(escOffset++) = L'\"';
+        inputOffset++;
+        continue;
+      case '/':
+        *(escOffset++) = L'/';
+        inputOffset++;
+        continue;
+      case 'b':
+        *(escOffset++) = L'\b';
+        inputOffset++;
+        continue;
+      case 'f':
+        *(escOffset++) = L'\f';
+        inputOffset++;
+        continue;
+      case 'n':
+        *(escOffset++) = L'\n';
+        inputOffset++;
+        continue;
+      case 'r':
+        *(escOffset++) = L'\r';
+        inputOffset++;
+        continue;
+      case 't':
+        *(escOffset++) = L'\t';
+        inputOffset++;
+        continue;
+
+      case 'u': {
+        int index;
+        inputOffset++;
+
+        for (index = 0; index < 4; index++) {
+          switch (*inputOffset) {
+          case '\0':
+            return SetError(ds, -1,
+                            "Unterminated unicode "
+                            "escape sequence when "
+                            "decoding 'string'");
+          default:
+            return SetError(ds, -1,
+                            "Unexpected character in "
+                            "unicode escape sequence "
+                            "when decoding 'string'");
+
+          case '0':
+          case '1':
+          case '2':
+          case '3':
+          case '4':
+          case '5':
+          case '6':
+          case '7':
+          case '8':
+          case '9':
+            sur[iSur] = (sur[iSur] << 4) + (JSUTF16)(*inputOffset - '0');
+            break;
+
+          case 'a':
+          case 'b':
+          case 'c':
+          case 'd':
+          case 'e':
+          case 'f':
+            sur[iSur] = (sur[iSur] << 4) + 10 + (JSUTF16)(*inputOffset - 'a');
+            break;
+
+          case 'A':
+          case 'B':
+          case 'C':
+          case 'D':
+          case 'E':
+          case 'F':
+            sur[iSur] = (sur[iSur] << 4) + 10 + (JSUTF16)(*inputOffset - 'A');
+            break;
+          }
+
+          inputOffset++;
+        }
+
+        if (iSur == 0) {
+          if ((sur[iSur] & 0xfc00) == 0xd800) {
+            // First of a surrogate pair, continue parsing
+            iSur++;
+            break;
+          }
+          (*escOffset++) = (wchar_t)sur[iSur];
+          iSur = 0;
+        } else {
+          // Decode pair
+          if ((sur[1] & 0xfc00) != 0xdc00) {
+            return SetError(ds, -1,
+                            "Unpaired high surrogate when "
+                            "decoding 'string'");
+          }
+#if WCHAR_MAX == 0xffff
+          (*escOffset++) = (wchar_t)sur[0];
+          (*escOffset++) = (wchar_t)sur[1];
+#else
+          (*escOffset++) = (wchar_t)0x10000 +
+                           (((sur[0] - 0xd800) << 10) | (sur[1] - 0xdc00));
+#endif
+          iSur = 0;
+        }
+        break;
+      }
+
+      case '\0':
+        return SetError(ds, -1,
+                        "Unterminated escape sequence when "
+                        "decoding 'string'");
+      default:
+        return SetError(ds, -1,
+                        "Unrecognized escape sequence when "
+                        "decoding 'string'");
+      }
+      break;
+
+    case 1: {
+      *(escOffset++) = (wchar_t)(*inputOffset++);
+      break;
+    }
+
+    case 2: {
+      ucs = (*inputOffset++) & 0x1f;
+      ucs <<= 6;
+      if (((*inputOffset) & 0x80) != 0x80) {
+        return SetError(ds, -1,
+                        "Invalid octet in UTF-8 sequence when "
+                        "decoding 'string'");
+      }
+      ucs |= (*inputOffset++) & 0x3f;
+      if (ucs < 0x80)
+        return SetError(ds, -1,
+                        "Overlong 2 byte UTF-8 sequence detected "
+                        "when decoding 'string'");
+      *(escOffset++) = (wchar_t)ucs;
+      break;
+    }
+
+    case 3: {
+      JSUTF32 ucs = 0;
+      ucs |= (*inputOffset++) & 0x0f;
+
+      for (index = 0; index < 2; index++) {
+        ucs <<= 6;
+        oct = (*inputOffset++);
+
+        if ((oct & 0x80) != 0x80) {
+          return SetError(ds, -1,
+                          "Invalid octet in UTF-8 sequence when "
+                          "decoding 'string'");
+        }
+
+        ucs |= oct & 0x3f;
+      }
+
+      if (ucs < 0x800)
+        return SetError(ds, -1,
+                        "Overlong 3 byte UTF-8 sequence detected "
+                        "when encoding string");
+      *(escOffset++) = (wchar_t)ucs;
+      break;
+    }
+
+    case 4: {
+      JSUTF32 ucs = 0;
+      ucs |= (*inputOffset++) & 0x07;
+
+      for (index = 0; index < 3; index++) {
+        ucs <<= 6;
+        oct = (*inputOffset++);
+
+        if ((oct & 0x80) != 0x80) {
+          return SetError(ds, -1,
+                          "Invalid octet in UTF-8 sequence when "
+                          "decoding 'string'");
+        }
+
+        ucs |= oct & 0x3f;
+      }
+
+      if (ucs < 0x10000)
+        return SetError(ds, -1,
+                        "Overlong 4 byte UTF-8 sequence detected "
+                        "when decoding 'string'");
+
+#if WCHAR_MAX == 0xffff
+      if (ucs >= 0x10000) {
+        ucs -= 0x10000;
+        *(escOffset++) = (wchar_t)(ucs >> 10) + 0xd800;
+        *(escOffset++) = (wchar_t)(ucs & 0x3ff) + 0xdc00;
+      } else {
+        *(escOffset++) = (wchar_t)ucs;
+      }
+#else
+      *(escOffset++) = (wchar_t)ucs;
+#endif
+      break;
+    }
+    }
+  }
+}
+
+JSOBJ FASTCALL_MSVC decode_array(struct DecoderState *ds) {
+  JSOBJ itemValue;
+  JSOBJ newObj;
+  int len;
+  ds->objDepth++;
+  if (ds->objDepth > JSON_MAX_OBJECT_DEPTH) {
+    return SetError(ds, -1, "Reached object decoding depth limit");
+  }
+
+  newObj = ds->dec->newArray(ds->prv, ds->dec);
+  len = 0;
+
+  ds->lastType = JT_INVALID;
+  ds->start++;
+
+  for (;;) {
+    SkipWhitespace(ds);
+
+    if ((*ds->start) == ']') {
+      ds->objDepth--;
+      if (len == 0) {
+        ds->start++;
+        return ds->dec->endArray(ds->prv, newObj);
+      }
+
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return SetError(
+          ds, -1, "Unexpected character found when decoding array value (1)");
+    }
+
+    itemValue = decode_any(ds);
+
+    if (itemValue == NULL) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return NULL;
+    }
+
+    if (!ds->dec->arrayAddItem(ds->prv, newObj, itemValue)) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return NULL;
+    }
+
+    SkipWhitespace(ds);
+
+    switch (*(ds->start++)) {
+    case ']': {
+      ds->objDepth--;
+      return ds->dec->endArray(ds->prv, newObj);
+    }
+    case ',':
+      break;
+
+    default:
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return SetError(
+          ds, -1, "Unexpected character found when decoding array value (2)");
+    }
+
+    len++;
+  }
+}
+
+JSOBJ FASTCALL_MSVC decode_object(struct DecoderState *ds) {
+  JSOBJ itemName;
+  JSOBJ itemValue;
+  JSOBJ newObj;
+
+  ds->objDepth++;
+  if (ds->objDepth > JSON_MAX_OBJECT_DEPTH) {
+    return SetError(ds, -1, "Reached object decoding depth limit");
+  }
+
+  newObj = ds->dec->newObject(ds->prv, ds->dec);
+
+  ds->start++;
+
+  for (;;) {
+    SkipWhitespace(ds);
+
+    if ((*ds->start) == '}') {
+      ds->objDepth--;
+      ds->start++;
+      return ds->dec->endObject(ds->prv, newObj);
+    }
+
+    ds->lastType = JT_INVALID;
+    itemName = decode_any(ds);
+
+    if (itemName == NULL) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return NULL;
+    }
+
+    if (ds->lastType != JT_UTF8) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      ds->dec->releaseObject(ds->prv, itemName, ds->dec);
+      return SetError(
+          ds, -1, "Key name of object must be 'string' when decoding 'object'");
+    }
+
+    SkipWhitespace(ds);
+
+    if (*(ds->start++) != ':') {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      ds->dec->releaseObject(ds->prv, itemName, ds->dec);
+      return SetError(ds, -1, "No ':' found when decoding object value");
+    }
+
+    SkipWhitespace(ds);
+
+    itemValue = decode_any(ds);
+
+    if (itemValue == NULL) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      ds->dec->releaseObject(ds->prv, itemName, ds->dec);
+      return NULL;
+    }
+
+    if (!ds->dec->objectAddKey(ds->prv, newObj, itemName, itemValue)) {
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      ds->dec->releaseObject(ds->prv, itemName, ds->dec);
+      ds->dec->releaseObject(ds->prv, itemValue, ds->dec);
+      return NULL;
+    }
+
+    SkipWhitespace(ds);
+
+    switch (*(ds->start++)) {
+    case '}': {
+      ds->objDepth--;
+      return ds->dec->endObject(ds->prv, newObj);
+    }
+    case ',':
+      break;
+
+    default:
+      ds->dec->releaseObject(ds->prv, newObj, ds->dec);
+      return SetError(ds, -1,
+                      "Unexpected character found when decoding object value");
+    }
+  }
+}
+
+JSOBJ FASTCALL_MSVC decode_any(struct DecoderState *ds) {
+  for (;;) {
+    switch (*ds->start) {
+    case '\"':
+      return decode_string(ds);
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+    case '8':
+    case '9':
+    case 'I':
+    case 'N':
+    case '-':
+      return decode_numeric(ds);
+
+    case '[':
+      return decode_array(ds);
+    case '{':
+      return decode_object(ds);
+    case 't':
+      return decode_true(ds);
+    case 'f':
+      return decode_false(ds);
+    case 'n':
+      return decode_null(ds);
+
+    case ' ':
+    case '\t':
+    case '\r':
+    case '\n':
+      // White space
+      ds->start++;
+      break;
+
+    default:
+      return SetError(ds, -1, "Expected object or value");
+    }
+  }
+}
+
+JSOBJ JSON_DecodeObject(JSONObjectDecoder *dec, const char *buffer,
+                        size_t cbBuffer) {
+  /*
+  FIXME: Base the size of escBuffer of that of cbBuffer so that the unicode
+  escaping doesn't run into the wall each time */
+  char *locale;
+  struct DecoderState ds;
+  wchar_t escBuffer[(JSON_MAX_STACK_BUFFER_SIZE / sizeof(wchar_t))];
+  JSOBJ ret;
+
+  ds.start = (char *)buffer;
+  ds.end = ds.start + cbBuffer;
+
+  ds.escStart = escBuffer;
+  ds.escEnd = ds.escStart + (JSON_MAX_STACK_BUFFER_SIZE / sizeof(wchar_t));
+  ds.escHeap = 0;
+  ds.prv = dec->prv;
+  ds.dec = dec;
+  ds.dec->errorStr = NULL;
+  ds.dec->errorOffset = NULL;
+  ds.objDepth = 0;
+
+  ds.dec = dec;
+
+  locale = setlocale(LC_NUMERIC, NULL);
+  if (!locale) {
+    return SetError(&ds, -1, "setlocale call failed");
+  }
+
+  if (strcmp(locale, "C")) {
+    size_t len = strlen(locale) + 1;
+    char *saved_locale = malloc(len);
+    if (saved_locale == NULL) {
+      return SetError(&ds, -1, "Could not reserve memory block");
+    }
+    memcpy(saved_locale, locale, len);
+    setlocale(LC_NUMERIC, "C");
+    ret = decode_any(&ds);
+    setlocale(LC_NUMERIC, saved_locale);
+    free(saved_locale);
+  } else {
+    ret = decode_any(&ds);
+  }
+
+  if (ds.escHeap) {
+    dec->free(ds.escStart);
+  }
+
+  SkipWhitespace(&ds);
+
+  if (ds.start != ds.end && ret) {
+    dec->releaseObject(ds.prv, ret, ds.dec);
+    return SetError(&ds, -1, "Trailing data");
+  }
+
+  return ret;
+}
diff --git a/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsonenc.c b/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsonenc.c
new file mode 100644
index 0000000000000..1564ecb64b01d
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/ujson/lib/ultrajsonenc.c
@@ -0,0 +1,1205 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+    * Neither the name of the ESN Social Software AB nor the
+      names of its contributors may be used to endorse or promote products
+      derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+ * Copyright (c) 1988-1993 The Regents of the University of California.
+ * Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+// Licence at LICENSES/ULTRAJSON_LICENSE
+
+#include "pandas/portable.h"
+#include "pandas/vendored/ujson/lib/ultrajson.h"
+#include <locale.h>
+#include <math.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#ifndef TRUE
+#define TRUE 1
+#endif
+#ifndef FALSE
+#define FALSE 0
+#endif
+
+/*
+Worst cases being:
+
+Control characters (ASCII < 32)
+0x00 (1 byte) input => \u0000 output (6 bytes)
+1 * 6 => 6 (6 bytes required)
+
+or UTF-16 surrogate pairs
+4 bytes input in UTF-8 => \uXXXX\uYYYY (12 bytes).
+
+4 * 6 => 24 bytes (12 bytes required)
+
+The extra 2 bytes are for the quotes around the string
+
+*/
+#define RESERVE_STRING(_len) (2 + ((_len) * 6))
+
+static const double g_pow10[] = {1,
+                                 10,
+                                 100,
+                                 1000,
+                                 10000,
+                                 100000,
+                                 1000000,
+                                 10000000,
+                                 100000000,
+                                 1000000000,
+                                 10000000000,
+                                 100000000000,
+                                 1000000000000,
+                                 10000000000000,
+                                 100000000000000,
+                                 1000000000000000};
+static const char g_hexChars[] = "0123456789abcdef";
+static const char g_escapeChars[] = "0123456789\\b\\t\\n\\f\\r\\\"\\\\\\/";
+
+/*
+FIXME: While this is fine dandy and working it's a magic value mess which
+probably only the author understands.
+Needs a cleanup and more documentation */
+
+/*
+Table for pure ascii output escaping all characters above 127 to \uXXXX */
+static const JSUINT8 g_asciiOutputTable[256] = {
+    /* 0x00 */ 0,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    10,
+    12,
+    14,
+    30,
+    16,
+    18,
+    30,
+    30,
+    /* 0x10 */ 30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    30,
+    /* 0x20 */ 1,
+    1,
+    20,
+    1,
+    1,
+    1,
+    29,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    24,
+    /* 0x30 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    29,
+    1,
+    29,
+    1,
+    /* 0x40 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x50 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    22,
+    1,
+    1,
+    1,
+    /* 0x60 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x70 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x80 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0x90 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xa0 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xb0 */ 1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    /* 0xc0 */ 2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    /* 0xd0 */ 2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    /* 0xe0 */ 3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    /* 0xf0 */ 4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    5,
+    5,
+    5,
+    5,
+    6,
+    6,
+    1,
+    1};
+
+static void SetError(JSOBJ obj, JSONObjectEncoder *enc, const char *message) {
+  enc->errorMsg = message;
+  enc->errorObj = obj;
+}
+
+/*
+FIXME: Keep track of how big these get across several encoder calls and try to
+make an estimate
+That way we won't run our head into the wall each call */
+void Buffer_Realloc(JSONObjectEncoder *enc, size_t cbNeeded) {
+  size_t curSize = enc->end - enc->start;
+  size_t newSize = curSize * 2;
+  size_t offset = enc->offset - enc->start;
+
+  while (newSize < curSize + cbNeeded) {
+    newSize *= 2;
+  }
+
+  if (enc->heap) {
+    enc->start = (char *)enc->realloc(enc->start, newSize);
+    if (!enc->start) {
+      SetError(NULL, enc, "Could not reserve memory block");
+      return;
+    }
+  } else {
+    char *oldStart = enc->start;
+    enc->heap = 1;
+    enc->start = (char *)enc->malloc(newSize);
+    if (!enc->start) {
+      SetError(NULL, enc, "Could not reserve memory block");
+      return;
+    }
+    memcpy(enc->start, oldStart, offset);
+  }
+  enc->offset = enc->start + offset;
+  enc->end = enc->start + newSize;
+}
+
+INLINE_PREFIX void FASTCALL_MSVC
+Buffer_AppendShortHexUnchecked(char *outputOffset, unsigned short value) {
+  *(outputOffset++) = g_hexChars[(value & 0xf000) >> 12];
+  *(outputOffset++) = g_hexChars[(value & 0x0f00) >> 8];
+  *(outputOffset++) = g_hexChars[(value & 0x00f0) >> 4];
+  *(outputOffset++) = g_hexChars[(value & 0x000f) >> 0];
+}
+
+int Buffer_EscapeStringUnvalidated(JSONObjectEncoder *enc, const char *io,
+                                   const char *end) {
+  char *of = (char *)enc->offset;
+
+  for (;;) {
+    switch (*io) {
+    case 0x00: {
+      if (io < end) {
+        *(of++) = '\\';
+        *(of++) = 'u';
+        *(of++) = '0';
+        *(of++) = '0';
+        *(of++) = '0';
+        *(of++) = '0';
+        break;
+      } else {
+        enc->offset += (of - enc->offset);
+        return TRUE;
+      }
+    }
+    case '\"':
+      (*of++) = '\\';
+      (*of++) = '\"';
+      break;
+    case '\\':
+      (*of++) = '\\';
+      (*of++) = '\\';
+      break;
+    case '/':
+      (*of++) = '\\';
+      (*of++) = '/';
+      break;
+    case '\b':
+      (*of++) = '\\';
+      (*of++) = 'b';
+      break;
+    case '\f':
+      (*of++) = '\\';
+      (*of++) = 'f';
+      break;
+    case '\n':
+      (*of++) = '\\';
+      (*of++) = 'n';
+      break;
+    case '\r':
+      (*of++) = '\\';
+      (*of++) = 'r';
+      break;
+    case '\t':
+      (*of++) = '\\';
+      (*of++) = 't';
+      break;
+
+    case 0x26: // '/'
+    case 0x3c: // '<'
+    case 0x3e: // '>'
+    {
+      if (enc->encodeHTMLChars) {
+        // Fall through to \u00XX case below.
+        PD_FALLTHROUGH;
+      } else {
+        // Same as default case below.
+        (*of++) = (*io);
+        break;
+      }
+    }
+    case 0x01:
+    case 0x02:
+    case 0x03:
+    case 0x04:
+    case 0x05:
+    case 0x06:
+    case 0x07:
+    case 0x0b:
+    case 0x0e:
+    case 0x0f:
+    case 0x10:
+    case 0x11:
+    case 0x12:
+    case 0x13:
+    case 0x14:
+    case 0x15:
+    case 0x16:
+    case 0x17:
+    case 0x18:
+    case 0x19:
+    case 0x1a:
+    case 0x1b:
+    case 0x1c:
+    case 0x1d:
+    case 0x1e:
+    case 0x1f: {
+      *(of++) = '\\';
+      *(of++) = 'u';
+      *(of++) = '0';
+      *(of++) = '0';
+      *(of++) = g_hexChars[(unsigned char)(((*io) & 0xf0) >> 4)];
+      *(of++) = g_hexChars[(unsigned char)((*io) & 0x0f)];
+      break;
+    }
+    default:
+      (*of++) = (*io);
+      break;
+    }
+    io++;
+  }
+}
+
+int Buffer_EscapeStringValidated(JSOBJ obj, JSONObjectEncoder *enc,
+                                 const char *io, const char *end) {
+  JSUTF32 ucs;
+  char *of = (char *)enc->offset;
+
+  for (;;) {
+    JSUINT8 utflen = g_asciiOutputTable[(unsigned char)*io];
+
+    switch (utflen) {
+    case 0: {
+      if (io < end) {
+        *(of++) = '\\';
+        *(of++) = 'u';
+        *(of++) = '0';
+        *(of++) = '0';
+        *(of++) = '0';
+        *(of++) = '0';
+        io++;
+        continue;
+      } else {
+        enc->offset += (of - enc->offset);
+        return TRUE;
+      }
+    }
+
+    case 1: {
+      *(of++) = (*io++);
+      continue;
+    }
+
+    case 2: {
+      JSUTF32 in;
+      JSUTF16 in16;
+
+      if (end - io < 1) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc, "Unterminated UTF-8 sequence when encoding string");
+        return FALSE;
+      }
+
+      memcpy(&in16, io, sizeof(JSUTF16));
+      in = (JSUTF32)in16;
+
+#ifdef __LITTLE_ENDIAN__
+      ucs = ((in & 0x1f) << 6) | ((in >> 8) & 0x3f);
+#else
+      ucs = ((in & 0x1f00) >> 2) | (in & 0x3f);
+#endif
+
+      if (ucs < 0x80) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc,
+                 "Overlong 2 byte UTF-8 sequence detected when "
+                 "encoding string");
+        return FALSE;
+      }
+
+      io += 2;
+      break;
+    }
+
+    case 3: {
+      JSUTF32 in;
+      JSUTF16 in16;
+      JSUINT8 in8;
+
+      if (end - io < 2) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc, "Unterminated UTF-8 sequence when encoding string");
+        return FALSE;
+      }
+
+      memcpy(&in16, io, sizeof(JSUTF16));
+      memcpy(&in8, io + 2, sizeof(JSUINT8));
+#ifdef __LITTLE_ENDIAN__
+      in = (JSUTF32)in16;
+      in |= in8 << 16;
+      ucs =
+          ((in & 0x0f) << 12) | ((in & 0x3f00) >> 2) | ((in & 0x3f0000) >> 16);
+#else
+      in = in16 << 8;
+      in |= in8;
+      ucs = ((in & 0x0f0000) >> 4) | ((in & 0x3f00) >> 2) | (in & 0x3f);
+#endif
+
+      if (ucs < 0x800) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc,
+                 "Overlong 3 byte UTF-8 sequence detected when "
+                 "encoding string");
+        return FALSE;
+      }
+
+      io += 3;
+      break;
+    }
+    case 4: {
+      JSUTF32 in;
+
+      if (end - io < 3) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc, "Unterminated UTF-8 sequence when encoding string");
+        return FALSE;
+      }
+
+      memcpy(&in, io, sizeof(JSUTF32));
+#ifdef __LITTLE_ENDIAN__
+      ucs = ((in & 0x07) << 18) | ((in & 0x3f00) << 4) |
+            ((in & 0x3f0000) >> 10) | ((in & 0x3f000000) >> 24);
+#else
+      ucs = ((in & 0x07000000) >> 6) | ((in & 0x3f0000) >> 4) |
+            ((in & 0x3f00) >> 2) | (in & 0x3f);
+#endif
+      if (ucs < 0x10000) {
+        enc->offset += (of - enc->offset);
+        SetError(obj, enc,
+                 "Overlong 4 byte UTF-8 sequence detected when "
+                 "encoding string");
+        return FALSE;
+      }
+
+      io += 4;
+      break;
+    }
+
+    case 5:
+    case 6: {
+      enc->offset += (of - enc->offset);
+      SetError(obj, enc,
+               "Unsupported UTF-8 sequence length when encoding string");
+      return FALSE;
+    }
+
+    case 29: {
+      if (enc->encodeHTMLChars) {
+        // Fall through to \u00XX case 30 below.
+        PD_FALLTHROUGH;
+      } else {
+        // Same as case 1 above.
+        *(of++) = (*io++);
+        continue;
+      }
+    }
+
+    case 30: {
+      // \uXXXX encode
+      *(of++) = '\\';
+      *(of++) = 'u';
+      *(of++) = '0';
+      *(of++) = '0';
+      *(of++) = g_hexChars[(unsigned char)(((*io) & 0xf0) >> 4)];
+      *(of++) = g_hexChars[(unsigned char)((*io) & 0x0f)];
+      io++;
+      continue;
+    }
+    case 10:
+    case 12:
+    case 14:
+    case 16:
+    case 18:
+    case 20:
+    case 22:
+    case 24: {
+      *(of++) = *((char *)(g_escapeChars + utflen + 0));
+      *(of++) = *((char *)(g_escapeChars + utflen + 1));
+      io++;
+      continue;
+    }
+    // This can never happen, it's here to make L4 VC++ happy
+    default: {
+      ucs = 0;
+      break;
+    }
+    }
+
+    /*
+    If the character is a UTF8 sequence of length > 1 we end up here */
+    if (ucs >= 0x10000) {
+      ucs -= 0x10000;
+      *(of++) = '\\';
+      *(of++) = 'u';
+      Buffer_AppendShortHexUnchecked(of, (unsigned short)(ucs >> 10) + 0xd800);
+      of += 4;
+
+      *(of++) = '\\';
+      *(of++) = 'u';
+      Buffer_AppendShortHexUnchecked(of,
+                                     (unsigned short)(ucs & 0x3ff) + 0xdc00);
+      of += 4;
+    } else {
+      *(of++) = '\\';
+      *(of++) = 'u';
+      Buffer_AppendShortHexUnchecked(of, (unsigned short)ucs);
+      of += 4;
+    }
+  }
+}
+
+#define Buffer_Reserve(__enc, __len)                                           \
+  if ((size_t)((__enc)->end - (__enc)->offset) < (size_t)(__len)) {            \
+    Buffer_Realloc((__enc), (__len));                                          \
+  }
+
+#define Buffer_AppendCharUnchecked(__enc, __chr) *((__enc)->offset++) = __chr;
+
+INLINE_PREFIX void FASTCALL_MSVC strreverse(char *begin, char *end) {
+  char aux;
+  while (end > begin)
+    aux = *end, *end-- = *begin, *begin++ = aux;
+}
+
+void Buffer_AppendIndentNewlineUnchecked(JSONObjectEncoder *enc) {
+  if (enc->indent > 0)
+    Buffer_AppendCharUnchecked(enc, '\n');
+}
+
+// This function could be refactored to only accept enc as an argument,
+// but this is a straight vendor from ujson source
+void Buffer_AppendIndentUnchecked(JSONObjectEncoder *enc, JSINT32 value) {
+  int i;
+  if (enc->indent > 0) {
+    while (value-- > 0)
+      for (i = 0; i < enc->indent; i++)
+        Buffer_AppendCharUnchecked(enc, ' ');
+  }
+}
+
+void Buffer_AppendIntUnchecked(JSONObjectEncoder *enc, JSINT32 value) {
+  char *wstr;
+  JSUINT32 uvalue = (value < 0) ? -value : value;
+  wstr = enc->offset;
+
+  // Conversion. Number is reversed.
+  do {
+    *wstr++ = (char)(48 + (uvalue % 10));
+  } while (uvalue /= 10);
+  if (value < 0)
+    *wstr++ = '-';
+
+  // Reverse string
+  strreverse(enc->offset, wstr - 1);
+  enc->offset += (wstr - (enc->offset));
+}
+
+void Buffer_AppendLongUnchecked(JSONObjectEncoder *enc, JSINT64 value) {
+  char *wstr;
+  JSUINT64 uvalue;
+  if (value == INT64_MIN) {
+    uvalue = INT64_MAX + UINT64_C(1);
+  } else {
+    uvalue = (value < 0) ? -value : value;
+  }
+
+  wstr = enc->offset;
+  // Conversion. Number is reversed.
+
+  do {
+    *wstr++ = (char)(48 + (uvalue % 10ULL));
+  } while (uvalue /= 10ULL);
+  if (value < 0)
+    *wstr++ = '-';
+
+  // Reverse string
+  strreverse(enc->offset, wstr - 1);
+  enc->offset += (wstr - (enc->offset));
+}
+
+int Buffer_AppendDoubleUnchecked(JSOBJ obj, JSONObjectEncoder *enc,
+                                 double value) {
+  /* if input is beyond the thresholds, revert to exponential */
+  const double thres_max = (double)1e16 - 1;
+  const double thres_min = (double)1e-15;
+  char precision_str[20];
+  int count;
+  double diff = 0.0;
+  char *str = enc->offset;
+  char *wstr = str;
+  unsigned long long whole;
+  double tmp;
+  unsigned long long frac;
+  int neg;
+  double pow10;
+
+  if (value == HUGE_VAL || value == -HUGE_VAL) {
+    SetError(obj, enc, "Invalid Inf value when encoding double");
+    return FALSE;
+  }
+
+  if (!(value == value)) {
+    SetError(obj, enc, "Invalid Nan value when encoding double");
+    return FALSE;
+  }
+
+  /* we'll work in positive values and deal with the
+  negative sign issue later */
+  neg = 0;
+  if (value < 0) {
+    neg = 1;
+    value = -value;
+  }
+
+  /*
+  for very large or small numbers switch back to native sprintf for
+  exponentials.  anyone want to write code to replace this? */
+  if (value > thres_max || (value != 0.0 && fabs(value) < thres_min)) {
+    precision_str[0] = '%';
+    precision_str[1] = '.';
+#if defined(_WIN32) && defined(_MSC_VER)
+    sprintf_s(precision_str + 2, sizeof(precision_str) - 2, "%ug",
+              enc->doublePrecision);
+    enc->offset += sprintf_s(str, enc->end - enc->offset, precision_str,
+                             neg ? -value : value);
+#else
+    snprintf(precision_str + 2, sizeof(precision_str) - 2, "%ug",
+             enc->doublePrecision);
+    enc->offset += snprintf(str, enc->end - enc->offset, precision_str,
+                            neg ? -value : value);
+#endif
+    return TRUE;
+  }
+
+  pow10 = g_pow10[enc->doublePrecision];
+
+  whole = (unsigned long long)value;
+  tmp = (value - whole) * pow10;
+  frac = (unsigned long long)(tmp);
+  diff = tmp - frac;
+
+  if (diff > 0.5) {
+    ++frac;
+  } else if (diff == 0.5 && ((frac == 0) || (frac & 1))) {
+    /* if halfway, round up if odd, OR
+    if last digit is 0.  That last part is strange */
+    ++frac;
+  }
+
+  // handle rollover, e.g.
+  // case 0.99 with prec 1 is 1.0 and case 0.95 with prec is 1.0 as well
+  if (frac >= pow10) {
+    frac = 0;
+    ++whole;
+  }
+
+  if (enc->doublePrecision == 0) {
+    diff = value - whole;
+
+    if (diff > 0.5) {
+      /* greater than 0.5, round up, e.g. 1.6 -> 2 */
+      ++whole;
+    } else if (diff == 0.5 && (whole & 1)) {
+      /* exactly 0.5 and ODD, then round up */
+      /* 1.5 -> 2, but 2.5 -> 2 */
+      ++whole;
+    }
+
+    // vvvvvvvvvvvvvvvvvvv  Diff from modp_dto2
+  } else if (frac) {
+    count = enc->doublePrecision;
+    // now do fractional part, as an unsigned number
+    // we know it is not 0 but we can have leading zeros, these
+    // should be removed
+    while (!(frac % 10)) {
+      --count;
+      frac /= 10;
+    }
+    //^^^^^^^^^^^^^^^^^^^  Diff from modp_dto2
+
+    // now do fractional part, as an unsigned number
+    do {
+      --count;
+      *wstr++ = (char)(48 + (frac % 10));
+    } while (frac /= 10);
+    // add extra 0s
+    while (count-- > 0) {
+      *wstr++ = '0';
+    }
+    // add decimal
+    *wstr++ = '.';
+  } else {
+    *wstr++ = '0';
+    *wstr++ = '.';
+  }
+
+  // Do whole part. Take care of sign
+  // conversion. Number is reversed.
+  do {
+    *wstr++ = (char)(48 + (whole % 10));
+  } while (whole /= 10);
+
+  if (neg) {
+    *wstr++ = '-';
+  }
+  strreverse(str, wstr - 1);
+  enc->offset += (wstr - (enc->offset));
+
+  return TRUE;
+}
+
+/*
+FIXME:
+Handle integration functions returning NULL here */
+
+/*
+FIXME:
+Perhaps implement recursion detection */
+
+void encode(JSOBJ obj, JSONObjectEncoder *enc, const char *name,
+            size_t cbName) {
+  const char *value;
+  const char *objName;
+  int count;
+  JSOBJ iterObj;
+  size_t szlen;
+  JSONTypeContext tc;
+  tc.encoder = enc;
+
+  if (enc->level > enc->recursionMax) {
+    SetError(obj, enc, "Maximum recursion level reached");
+    return;
+  }
+
+  /*
+  This reservation must hold
+
+  length of _name as encoded worst case +
+  maxLength of double to string OR maxLength of JSLONG to string
+  */
+
+  Buffer_Reserve(enc, 256 + RESERVE_STRING(cbName));
+  if (enc->errorMsg) {
+    return;
+  }
+
+  if (name) {
+    Buffer_AppendCharUnchecked(enc, '\"');
+
+    if (enc->forceASCII) {
+      if (!Buffer_EscapeStringValidated(obj, enc, name, name + cbName)) {
+        return;
+      }
+    } else {
+      if (!Buffer_EscapeStringUnvalidated(enc, name, name + cbName)) {
+        return;
+      }
+    }
+
+    Buffer_AppendCharUnchecked(enc, '\"');
+
+    Buffer_AppendCharUnchecked(enc, ':');
+#ifndef JSON_NO_EXTRA_WHITESPACE
+    Buffer_AppendCharUnchecked(enc, ' ');
+#endif
+  }
+
+  enc->beginTypeContext(obj, &tc);
+
+  switch (tc.type) {
+  case JT_INVALID: {
+    return;
+  }
+
+  case JT_ARRAY: {
+    count = 0;
+    enc->iterBegin(obj, &tc);
+
+    Buffer_AppendCharUnchecked(enc, '[');
+    Buffer_AppendIndentNewlineUnchecked(enc);
+
+    while (enc->iterNext(obj, &tc)) {
+      if (count > 0) {
+        Buffer_AppendCharUnchecked(enc, ',');
+#ifndef JSON_NO_EXTRA_WHITESPACE
+        Buffer_AppendCharUnchecked(buffer, ' ');
+#endif
+        Buffer_AppendIndentNewlineUnchecked(enc);
+      }
+
+      iterObj = enc->iterGetValue(obj, &tc);
+
+      enc->level++;
+      Buffer_AppendIndentUnchecked(enc, enc->level);
+      encode(iterObj, enc, NULL, 0);
+      count++;
+    }
+
+    enc->iterEnd(obj, &tc);
+    Buffer_AppendIndentNewlineUnchecked(enc);
+    Buffer_AppendIndentUnchecked(enc, enc->level);
+    Buffer_AppendCharUnchecked(enc, ']');
+    break;
+  }
+
+  case JT_OBJECT: {
+    count = 0;
+    enc->iterBegin(obj, &tc);
+
+    Buffer_AppendCharUnchecked(enc, '{');
+    Buffer_AppendIndentNewlineUnchecked(enc);
+
+    while (enc->iterNext(obj, &tc)) {
+      if (count > 0) {
+        Buffer_AppendCharUnchecked(enc, ',');
+#ifndef JSON_NO_EXTRA_WHITESPACE
+        Buffer_AppendCharUnchecked(enc, ' ');
+#endif
+        Buffer_AppendIndentNewlineUnchecked(enc);
+      }
+
+      iterObj = enc->iterGetValue(obj, &tc);
+      objName = enc->iterGetName(obj, &tc, &szlen);
+
+      enc->level++;
+      Buffer_AppendIndentUnchecked(enc, enc->level);
+      encode(iterObj, enc, objName, szlen);
+      count++;
+    }
+
+    enc->iterEnd(obj, &tc);
+    Buffer_AppendIndentNewlineUnchecked(enc);
+    Buffer_AppendIndentUnchecked(enc, enc->level);
+    Buffer_AppendCharUnchecked(enc, '}');
+    break;
+  }
+
+  case JT_LONG: {
+    Buffer_AppendLongUnchecked(enc, enc->getLongValue(obj, &tc));
+    break;
+  }
+
+  case JT_INT: {
+    Buffer_AppendIntUnchecked(enc, enc->getIntValue(obj, &tc));
+    break;
+  }
+
+  case JT_TRUE: {
+    Buffer_AppendCharUnchecked(enc, 't');
+    Buffer_AppendCharUnchecked(enc, 'r');
+    Buffer_AppendCharUnchecked(enc, 'u');
+    Buffer_AppendCharUnchecked(enc, 'e');
+    break;
+  }
+
+  case JT_FALSE: {
+    Buffer_AppendCharUnchecked(enc, 'f');
+    Buffer_AppendCharUnchecked(enc, 'a');
+    Buffer_AppendCharUnchecked(enc, 'l');
+    Buffer_AppendCharUnchecked(enc, 's');
+    Buffer_AppendCharUnchecked(enc, 'e');
+    break;
+  }
+
+  case JT_NULL: {
+    Buffer_AppendCharUnchecked(enc, 'n');
+    Buffer_AppendCharUnchecked(enc, 'u');
+    Buffer_AppendCharUnchecked(enc, 'l');
+    Buffer_AppendCharUnchecked(enc, 'l');
+    break;
+  }
+
+  case JT_DOUBLE: {
+    if (!Buffer_AppendDoubleUnchecked(obj, enc,
+                                      enc->getDoubleValue(obj, &tc))) {
+      enc->endTypeContext(obj, &tc);
+      enc->level--;
+      return;
+    }
+    break;
+  }
+
+  case JT_UTF8: {
+    value = enc->getStringValue(obj, &tc, &szlen);
+    if (enc->errorMsg) {
+      enc->endTypeContext(obj, &tc);
+      return;
+    }
+    Buffer_Reserve(enc, RESERVE_STRING(szlen));
+    Buffer_AppendCharUnchecked(enc, '\"');
+
+    if (enc->forceASCII) {
+      if (!Buffer_EscapeStringValidated(obj, enc, value, value + szlen)) {
+        enc->endTypeContext(obj, &tc);
+        enc->level--;
+        return;
+      }
+    } else {
+      if (!Buffer_EscapeStringUnvalidated(enc, value, value + szlen)) {
+        enc->endTypeContext(obj, &tc);
+        enc->level--;
+        return;
+      }
+    }
+
+    Buffer_AppendCharUnchecked(enc, '\"');
+    break;
+  }
+
+  case JT_BIGNUM: {
+    value = enc->getBigNumStringValue(obj, &tc, &szlen);
+
+    Buffer_Reserve(enc, RESERVE_STRING(szlen));
+    if (enc->errorMsg) {
+      enc->endTypeContext(obj, &tc);
+      return;
+    }
+
+    if (enc->forceASCII) {
+      if (!Buffer_EscapeStringValidated(obj, enc, value, value + szlen)) {
+        enc->endTypeContext(obj, &tc);
+        enc->level--;
+        return;
+      }
+    } else {
+      if (!Buffer_EscapeStringUnvalidated(enc, value, value + szlen)) {
+        enc->endTypeContext(obj, &tc);
+        enc->level--;
+        return;
+      }
+    }
+
+    break;
+  }
+  }
+
+  enc->endTypeContext(obj, &tc);
+  enc->level--;
+}
+
+char *JSON_EncodeObject(JSOBJ obj, JSONObjectEncoder *enc, char *_buffer,
+                        size_t _cbBuffer) {
+  char *locale;
+  enc->malloc = enc->malloc ? enc->malloc : malloc;
+  enc->free = enc->free ? enc->free : free;
+  enc->realloc = enc->realloc ? enc->realloc : realloc;
+  enc->errorMsg = NULL;
+  enc->errorObj = NULL;
+  enc->level = 0;
+
+  if (enc->recursionMax < 1) {
+    enc->recursionMax = JSON_MAX_RECURSION_DEPTH;
+  }
+
+  if (enc->doublePrecision < 0 ||
+      enc->doublePrecision > JSON_DOUBLE_MAX_DECIMALS) {
+    enc->doublePrecision = JSON_DOUBLE_MAX_DECIMALS;
+  }
+
+  if (_buffer == NULL) {
+    _cbBuffer = 32768;
+    enc->start = (char *)enc->malloc(_cbBuffer);
+    if (!enc->start) {
+      SetError(obj, enc, "Could not reserve memory block");
+      return NULL;
+    }
+    enc->heap = 1;
+  } else {
+    enc->start = _buffer;
+    enc->heap = 0;
+  }
+
+  enc->end = enc->start + _cbBuffer;
+  enc->offset = enc->start;
+
+  locale = setlocale(LC_NUMERIC, NULL);
+  if (!locale) {
+    SetError(NULL, enc, "setlocale call failed");
+    return NULL;
+  }
+
+  if (strcmp(locale, "C")) {
+    size_t len = strlen(locale) + 1;
+    char *saved_locale = malloc(len);
+    if (saved_locale == NULL) {
+      SetError(NULL, enc, "Could not reserve memory block");
+      return NULL;
+    }
+    memcpy(saved_locale, locale, len);
+    setlocale(LC_NUMERIC, "C");
+    encode(obj, enc, NULL, 0);
+    setlocale(LC_NUMERIC, saved_locale);
+    free(saved_locale);
+  } else {
+    encode(obj, enc, NULL, 0);
+  }
+
+  Buffer_Reserve(enc, 1);
+  if (enc->errorMsg) {
+    return NULL;
+  }
+  Buffer_AppendCharUnchecked(enc, '\0');
+
+  return enc->start;
+}
diff --git a/pandas/pandas/_libs/src/vendored/ujson/python/JSONtoObj.c b/pandas/pandas/_libs/src/vendored/ujson/python/JSONtoObj.c
new file mode 100644
index 0000000000000..ef6f1104a1fb9
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/ujson/python/JSONtoObj.c
@@ -0,0 +1,171 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+    * Neither the name of the ESN Social Software AB nor the
+      names of its contributors may be used to endorse or promote products
+      derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
+GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
+THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+ * Copyright (c) 1988-1993 The Regents of the University of California.
+ * Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+// Licence at LICENSES/ULTRAJSON_LICENSE
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#include "pandas/vendored/ujson/lib/ultrajson.h"
+
+static int Object_objectAddKey(void *Py_UNUSED(prv), JSOBJ obj, JSOBJ name,
+                               JSOBJ value) {
+  int ret = PyDict_SetItem(obj, name, value);
+  Py_DECREF((PyObject *)name);
+  Py_DECREF((PyObject *)value);
+  return ret == 0 ? 1 : 0;
+}
+
+static int Object_arrayAddItem(void *Py_UNUSED(prv), JSOBJ obj, JSOBJ value) {
+  int ret = PyList_Append(obj, value);
+  Py_DECREF((PyObject *)value);
+  return ret == 0 ? 1 : 0;
+}
+
+static JSOBJ Object_newString(void *Py_UNUSED(prv), wchar_t *start,
+                              wchar_t *end) {
+  return PyUnicode_FromWideChar(start, (end - start));
+}
+
+static JSOBJ Object_newTrue(void *Py_UNUSED(prv)) { Py_RETURN_TRUE; }
+
+static JSOBJ Object_newFalse(void *Py_UNUSED(prv)) { Py_RETURN_FALSE; }
+
+static JSOBJ Object_newNull(void *Py_UNUSED(prv)) { Py_RETURN_NONE; }
+
+static JSOBJ Object_newPosInf(void *Py_UNUSED(prv)) {
+  return PyFloat_FromDouble(Py_HUGE_VAL);
+}
+
+static JSOBJ Object_newNegInf(void *Py_UNUSED(prv)) {
+  return PyFloat_FromDouble(-Py_HUGE_VAL);
+}
+
+static JSOBJ Object_newObject(void *Py_UNUSED(prv), void *Py_UNUSED(decoder)) {
+  return PyDict_New();
+}
+
+static JSOBJ Object_endObject(void *Py_UNUSED(prv), JSOBJ obj) { return obj; }
+
+static JSOBJ Object_newArray(void *Py_UNUSED(prv), void *Py_UNUSED(decoder)) {
+  return PyList_New(0);
+}
+
+static JSOBJ Object_endArray(void *Py_UNUSED(prv), JSOBJ obj) { return obj; }
+
+static JSOBJ Object_newInteger(void *Py_UNUSED(prv), JSINT32 value) {
+  return PyLong_FromLong(value);
+}
+
+static JSOBJ Object_newLong(void *Py_UNUSED(prv), JSINT64 value) {
+  return PyLong_FromLongLong(value);
+}
+
+static JSOBJ Object_newUnsignedLong(void *Py_UNUSED(prv), JSUINT64 value) {
+  return PyLong_FromUnsignedLongLong(value);
+}
+
+static JSOBJ Object_newDouble(void *Py_UNUSED(prv), double value) {
+  return PyFloat_FromDouble(value);
+}
+
+static void Object_releaseObject(void *Py_UNUSED(prv), JSOBJ obj,
+                                 void *Py_UNUSED(decoder)) {
+  Py_XDECREF(((PyObject *)obj));
+}
+
+PyObject *JSONToObj(PyObject *Py_UNUSED(self), PyObject *args,
+                    PyObject *kwargs) {
+  JSONObjectDecoder dec = {.newString = Object_newString,
+                           .objectAddKey = Object_objectAddKey,
+                           .arrayAddItem = Object_arrayAddItem,
+                           .newTrue = Object_newTrue,
+                           .newFalse = Object_newFalse,
+                           .newNull = Object_newNull,
+                           .newPosInf = Object_newPosInf,
+                           .newNegInf = Object_newNegInf,
+                           .newObject = Object_newObject,
+                           .endObject = Object_endObject,
+                           .newArray = Object_newArray,
+                           .endArray = Object_endArray,
+                           .newInt = Object_newInteger,
+                           .newLong = Object_newLong,
+                           .newUnsignedLong = Object_newUnsignedLong,
+                           .newDouble = Object_newDouble,
+                           .releaseObject = Object_releaseObject,
+                           .malloc = PyObject_Malloc,
+                           .free = PyObject_Free,
+                           .realloc = PyObject_Realloc,
+                           .errorStr = NULL,
+                           .errorOffset = NULL,
+                           .preciseFloat = 0,
+                           .prv = NULL};
+
+  char *kwlist[] = {"obj", "precise_float", NULL};
+  char *buf;
+  Py_ssize_t len;
+  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "s#|b", kwlist, &buf, &len,
+                                   &dec.preciseFloat)) {
+    return NULL;
+  }
+
+  PyObject *ret = JSON_DecodeObject(&dec, buf, len);
+
+  if (PyErr_Occurred()) {
+    if (ret) {
+      Py_DECREF((PyObject *)ret);
+    }
+    return NULL;
+  }
+
+  if (dec.errorStr) {
+    /*
+    FIXME: It's possible to give a much nicer error message here with actual
+    failing element in input etc*/
+
+    PyErr_Format(PyExc_ValueError, "%s", dec.errorStr);
+
+    if (ret) {
+      Py_DECREF((PyObject *)ret);
+    }
+
+    return NULL;
+  }
+
+  return ret;
+}
diff --git a/pandas/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/pandas/_libs/src/vendored/ujson/python/objToJSON.c
new file mode 100644
index 0000000000000..8342dbcd1763d
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -0,0 +1,2100 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+* Neither the name of the ESN Social Software AB nor the
+names of its contributors may be used to endorse or promote products
+derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
+GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
+THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+* Copyright (c) 1988-1993 The Regents of the University of California.
+* Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+// Licence at LICENSES/ULTRAJSON_LICENSE
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#define NO_IMPORT_ARRAY
+#define PY_ARRAY_UNIQUE_SYMBOL UJSON_NUMPY
+#include "datetime.h"
+#include "pandas/datetime/pd_datetime.h"
+#include "pandas/vendored/ujson/lib/ultrajson.h"
+#include <numpy/arrayobject.h>
+#include <numpy/arrayscalars.h>
+#include <numpy/ndarraytypes.h>
+#include <numpy/npy_math.h>
+
+npy_int64 get_nat(void) { return NPY_MIN_INT64; }
+
+typedef const char *(*PFN_PyTypeToUTF8)(JSOBJ obj, JSONTypeContext *ti,
+                                        size_t *_outLen);
+
+int object_is_decimal_type(PyObject *obj);
+int object_is_dataframe_type(PyObject *obj);
+int object_is_series_type(PyObject *obj);
+int object_is_index_type(PyObject *obj);
+int object_is_nat_type(PyObject *obj);
+int object_is_na_type(PyObject *obj);
+
+typedef struct __NpyArrContext {
+  PyObject *array;
+  char *dataptr;
+  npy_intp curdim;    // current dimension in array's order
+  npy_intp stridedim; // dimension we are striding over
+  int inc;            // stride dimension increment (+/- 1)
+  npy_intp dim;
+  npy_intp stride;
+  npy_intp ndim;
+  npy_intp index[NPY_MAXDIMS];
+  int type_num;
+
+  char **rowLabels;
+  char **columnLabels;
+} NpyArrContext;
+
+typedef struct __PdBlockContext {
+  Py_ssize_t colIdx;
+  Py_ssize_t ncols;
+  int transpose;
+
+  NpyArrContext **npyCtxts; // NpyArrContext for each column
+} PdBlockContext;
+
+typedef struct __TypeContext {
+  JSPFN_ITERBEGIN iterBegin;
+  JSPFN_ITEREND iterEnd;
+  JSPFN_ITERNEXT iterNext;
+  JSPFN_ITERGETNAME iterGetName;
+  JSPFN_ITERGETVALUE iterGetValue;
+  PFN_PyTypeToUTF8 PyTypeToUTF8;
+  PyObject *newObj;
+  PyObject *dictObj;
+  Py_ssize_t index;
+  Py_ssize_t size;
+  PyObject *itemValue;
+  PyObject *itemName;
+  PyObject *attrList;
+  PyObject *iterator;
+
+  double doubleValue;
+  JSINT64 longValue;
+
+  const char *cStr;
+  NpyArrContext *npyarr;
+  PdBlockContext *pdblock;
+  int transpose;
+  char **rowLabels;
+  char **columnLabels;
+  npy_intp rowLabelsLen;
+  npy_intp columnLabelsLen;
+} TypeContext;
+
+typedef struct __PyObjectEncoder {
+  JSONObjectEncoder enc;
+
+  // pass through the NpyArrContext when encoding multi-dimensional arrays
+  NpyArrContext *npyCtxtPassthru;
+
+  // pass through the PdBlockContext when encoding blocks
+  PdBlockContext *blkCtxtPassthru;
+
+  // pass-through to encode numpy data directly
+  int npyType;
+  void *npyValue;
+
+  int datetimeIso;
+  NPY_DATETIMEUNIT datetimeUnit;
+  NPY_DATETIMEUNIT valueUnit;
+
+  // output format style for pandas data types
+  int outputFormat;
+  int originalOutputFormat;
+
+  PyObject *defaultHandler;
+} PyObjectEncoder;
+
+#define GET_TC(__ptrtc) ((TypeContext *)((__ptrtc)->prv))
+
+enum PANDAS_FORMAT { SPLIT, RECORDS, INDEX, COLUMNS, VALUES };
+
+static int PdBlock_iterNext(JSOBJ, JSONTypeContext *);
+
+static TypeContext *createTypeContext(void) {
+  TypeContext *pc = PyObject_Malloc(sizeof(TypeContext));
+  if (!pc) {
+    PyErr_NoMemory();
+    return NULL;
+  }
+  pc->newObj = NULL;
+  pc->dictObj = NULL;
+  pc->itemValue = NULL;
+  pc->itemName = NULL;
+  pc->attrList = NULL;
+  pc->index = 0;
+  pc->size = 0;
+  pc->longValue = 0;
+  pc->doubleValue = 0.0;
+  pc->cStr = NULL;
+  pc->npyarr = NULL;
+  pc->pdblock = NULL;
+  pc->rowLabels = NULL;
+  pc->columnLabels = NULL;
+  pc->transpose = 0;
+  pc->rowLabelsLen = 0;
+  pc->columnLabelsLen = 0;
+
+  return pc;
+}
+
+static PyObject *get_values(PyObject *obj) {
+  PyObject *values = NULL;
+
+  if (object_is_index_type(obj) || object_is_series_type(obj)) {
+    // The special cases to worry about are dt64tz and category[dt64tz].
+    //  In both cases we want the UTC-localized datetime64 ndarray,
+    //  without going through and object array of Timestamps.
+    if (PyObject_HasAttrString(obj, "tz")) {
+      PyObject *tz = PyObject_GetAttrString(obj, "tz");
+      if (tz != Py_None) {
+        // Go through object array if we have dt64tz, since tz info will
+        // be lost if values is used directly.
+        Py_DECREF(tz);
+        values = PyObject_CallMethod(obj, "__array__", NULL);
+        return values;
+      }
+      Py_DECREF(tz);
+    }
+    values = PyObject_GetAttrString(obj, "values");
+    if (values == NULL) {
+      // Clear so we can subsequently try another method
+      PyErr_Clear();
+    } else if (PyObject_HasAttrString(values, "__array__")) {
+      // We may have gotten a Categorical or Sparse array so call np.array
+      PyObject *array_values = PyObject_CallMethod(values, "__array__", NULL);
+      Py_DECREF(values);
+      values = array_values;
+    } else if (!PyArray_CheckExact(values)) {
+      // Didn't get a numpy array, so keep trying
+      Py_DECREF(values);
+      values = NULL;
+    }
+  }
+
+  if (values == NULL) {
+    PyObject *typeRepr = PyObject_Repr((PyObject *)Py_TYPE(obj));
+    PyObject *repr;
+    if (PyObject_HasAttrString(obj, "dtype")) {
+      PyObject *dtype = PyObject_GetAttrString(obj, "dtype");
+      repr = PyObject_Repr(dtype);
+      Py_DECREF(dtype);
+    } else {
+      repr = PyUnicode_FromString("<unknown dtype>");
+    }
+
+    PyErr_Format(PyExc_ValueError, "%R or %R are not JSON serializable yet",
+                 repr, typeRepr);
+    Py_DECREF(repr);
+    Py_DECREF(typeRepr);
+
+    return NULL;
+  }
+
+  return values;
+}
+
+static PyObject *get_sub_attr(PyObject *obj, char *attr, char *subAttr) {
+  PyObject *tmp = PyObject_GetAttrString(obj, attr);
+  if (tmp == 0) {
+    return 0;
+  }
+  PyObject *ret = PyObject_GetAttrString(tmp, subAttr);
+  Py_DECREF(tmp);
+
+  return ret;
+}
+
+static Py_ssize_t get_attr_length(PyObject *obj, char *attr) {
+  PyObject *tmp = PyObject_GetAttrString(obj, attr);
+  if (tmp == 0) {
+    return 0;
+  }
+  Py_ssize_t ret = PyObject_Length(tmp);
+  Py_DECREF(tmp);
+
+  if (ret == -1) {
+    return 0;
+  }
+
+  return ret;
+}
+
+static npy_int64 get_long_attr(PyObject *o, const char *attr) {
+  // NB we are implicitly assuming that o is a Timedelta or Timestamp, or NaT
+
+  PyObject *value = PyObject_GetAttrString(o, attr);
+  const npy_int64 long_val =
+      (PyLong_Check(value) ? PyLong_AsLongLong(value) : PyLong_AsLong(value));
+
+  Py_DECREF(value);
+
+  if (object_is_nat_type(o)) {
+    // i.e. o is NaT, long_val will be NPY_MIN_INT64
+    return long_val;
+  }
+
+  // ensure we are in nanoseconds, similar to Timestamp._as_creso or _as_unit
+  PyObject *reso = PyObject_GetAttrString(o, "_creso");
+  if (!PyLong_Check(reso)) {
+    // https://github.com/pandas-dev/pandas/pull/49034#discussion_r1023165139
+    Py_DECREF(reso);
+    return -1;
+  }
+
+  long cReso = PyLong_AsLong(reso);
+  Py_DECREF(reso);
+  if (cReso == -1 && PyErr_Occurred()) {
+    return -1;
+  }
+
+  if (cReso == NPY_FR_us) {
+    return long_val * 1000L;
+  } else if (cReso == NPY_FR_ms) {
+    return long_val * 1000000L;
+  } else if (cReso == NPY_FR_s) {
+    return long_val * 1000000000L;
+  }
+
+  return long_val;
+}
+
+static npy_float64 total_seconds(PyObject *td) {
+  PyObject *value = PyObject_CallMethod(td, "total_seconds", NULL);
+  const npy_float64 double_val = PyFloat_AS_DOUBLE(value);
+  Py_DECREF(value);
+  return double_val;
+}
+
+static const char *PyBytesToUTF8(JSOBJ _obj, JSONTypeContext *Py_UNUSED(tc),
+                                 size_t *_outLen) {
+  PyObject *obj = (PyObject *)_obj;
+  *_outLen = PyBytes_GET_SIZE(obj);
+  return PyBytes_AS_STRING(obj);
+}
+
+static const char *PyUnicodeToUTF8(JSOBJ _obj, JSONTypeContext *tc,
+                                   size_t *_outLen) {
+  char *encoded = (char *)PyUnicode_AsUTF8AndSize(_obj, (Py_ssize_t *)_outLen);
+  if (encoded == NULL) {
+    /* Something went wrong.
+      Set errorMsg(to tell encoder to stop),
+      and let Python exception propagate. */
+    JSONObjectEncoder *enc = (JSONObjectEncoder *)tc->encoder;
+    enc->errorMsg = "Encoding failed.";
+  }
+  return encoded;
+}
+
+/* JSON callback. returns a char* and mutates the pointer to *len */
+static const char *NpyDateTimeToIsoCallback(JSOBJ Py_UNUSED(unused),
+                                            JSONTypeContext *tc, size_t *len) {
+  NPY_DATETIMEUNIT base = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+  NPY_DATETIMEUNIT valueUnit = ((PyObjectEncoder *)tc->encoder)->valueUnit;
+  GET_TC(tc)->cStr = int64ToIso(GET_TC(tc)->longValue, valueUnit, base, len);
+  return GET_TC(tc)->cStr;
+}
+
+/* JSON callback. returns a char* and mutates the pointer to *len */
+static const char *NpyTimeDeltaToIsoCallback(JSOBJ Py_UNUSED(unused),
+                                             JSONTypeContext *tc, size_t *len) {
+  GET_TC(tc)->cStr = int64ToIsoDuration(GET_TC(tc)->longValue, len);
+  return GET_TC(tc)->cStr;
+}
+
+/* JSON callback */
+static const char *PyDateTimeToIsoCallback(JSOBJ obj, JSONTypeContext *tc,
+                                           size_t *len) {
+  if (!PyDate_Check(obj) && !PyDateTime_Check(obj)) {
+    PyErr_SetString(PyExc_TypeError, "Expected date or datetime object");
+    ((JSONObjectEncoder *)tc->encoder)->errorMsg = "";
+    return NULL;
+  }
+
+  NPY_DATETIMEUNIT base = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+  return PyDateTimeToIso(obj, base, len);
+}
+
+static const char *PyTimeToJSON(JSOBJ _obj, JSONTypeContext *tc,
+                                size_t *outLen) {
+  PyObject *obj = (PyObject *)_obj;
+  PyObject *str = PyObject_CallMethod(obj, "isoformat", NULL);
+  if (str == NULL) {
+    *outLen = 0;
+    if (!PyErr_Occurred()) {
+      PyErr_SetString(PyExc_ValueError, "Failed to convert time");
+    }
+    ((JSONObjectEncoder *)tc->encoder)->errorMsg = "";
+    return NULL;
+  }
+  if (PyUnicode_Check(str)) {
+    PyObject *tmp = str;
+    str = PyUnicode_AsUTF8String(str);
+    Py_DECREF(tmp);
+  }
+
+  GET_TC(tc)->newObj = str;
+
+  *outLen = PyBytes_GET_SIZE(str);
+  char *outValue = PyBytes_AS_STRING(str);
+  return outValue;
+}
+
+static const char *PyDecimalToUTF8Callback(JSOBJ _obj, JSONTypeContext *tc,
+                                           size_t *len) {
+  PyObject *obj = (PyObject *)_obj;
+  PyObject *format_spec = PyUnicode_FromStringAndSize("f", 1);
+  PyObject *str = PyObject_Format(obj, format_spec);
+  Py_DECREF(format_spec);
+
+  if (str == NULL) {
+    ((JSONObjectEncoder *)tc->encoder)->errorMsg = "";
+    return NULL;
+  }
+
+  GET_TC(tc)->newObj = str;
+
+  Py_ssize_t s_len;
+  char *outValue = (char *)PyUnicode_AsUTF8AndSize(str, &s_len);
+  *len = s_len;
+
+  return outValue;
+}
+
+//=============================================================================
+// Numpy array iteration functions
+//=============================================================================
+
+static void NpyArr_freeItemValue(JSOBJ Py_UNUSED(_obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->npyarr &&
+      GET_TC(tc)->itemValue != GET_TC(tc)->npyarr->array) {
+    Py_XDECREF(GET_TC(tc)->itemValue);
+    GET_TC(tc)->itemValue = NULL;
+  }
+}
+
+static int NpyArr_iterNextNone(JSOBJ Py_UNUSED(_obj),
+                               JSONTypeContext *Py_UNUSED(tc)) {
+  return 0;
+}
+
+static void NpyArr_iterBegin(JSOBJ _obj, JSONTypeContext *tc) {
+  PyArrayObject *obj =
+      (PyArrayObject *)(GET_TC(tc)->newObj ? GET_TC(tc)->newObj : _obj);
+
+  NpyArrContext *npyarr = PyObject_Malloc(sizeof(NpyArrContext));
+  GET_TC(tc)->npyarr = npyarr;
+
+  if (!npyarr) {
+    PyErr_NoMemory();
+    GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+    return;
+  }
+
+  npyarr->array = (PyObject *)obj;
+  npyarr->dataptr = PyArray_DATA(obj);
+  npyarr->ndim = PyArray_NDIM(obj) - 1;
+  npyarr->curdim = 0;
+  npyarr->type_num = PyArray_DESCR(obj)->type_num;
+
+  if (GET_TC(tc)->transpose) {
+    npyarr->dim = PyArray_DIM(obj, (int)npyarr->ndim);
+    npyarr->stride = PyArray_STRIDE(obj, (int)npyarr->ndim);
+    npyarr->stridedim = npyarr->ndim;
+    npyarr->index[npyarr->ndim] = 0;
+    npyarr->inc = -1;
+  } else {
+    npyarr->dim = PyArray_DIM(obj, 0);
+    npyarr->stride = PyArray_STRIDE(obj, 0);
+    npyarr->stridedim = 0;
+    npyarr->index[0] = 0;
+    npyarr->inc = 1;
+  }
+
+  npyarr->columnLabels = GET_TC(tc)->columnLabels;
+  npyarr->rowLabels = GET_TC(tc)->rowLabels;
+}
+
+static void NpyArr_iterEnd(JSOBJ obj, JSONTypeContext *tc) {
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+
+  if (npyarr) {
+    NpyArr_freeItemValue(obj, tc);
+    PyObject_Free(npyarr);
+  }
+}
+
+static void NpyArrPassThru_iterBegin(JSOBJ Py_UNUSED(obj),
+                                     JSONTypeContext *Py_UNUSED(tc)) {}
+
+static void NpyArrPassThru_iterEnd(JSOBJ obj, JSONTypeContext *tc) {
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+  // finished this dimension, reset the data pointer
+  npyarr->curdim--;
+  npyarr->dataptr -= npyarr->stride * npyarr->index[npyarr->stridedim];
+  npyarr->stridedim -= npyarr->inc;
+
+  if (!PyArray_Check(npyarr->array)) {
+    PyErr_SetString(PyExc_TypeError,
+                    "NpyArrayPassThru_iterEnd received a non-array object");
+    return;
+  }
+  const PyArrayObject *arrayobj = (const PyArrayObject *)npyarr->array;
+  npyarr->dim = PyArray_DIM(arrayobj, (int)npyarr->stridedim);
+  npyarr->stride = PyArray_STRIDE(arrayobj, (int)npyarr->stridedim);
+  npyarr->dataptr += npyarr->stride;
+
+  NpyArr_freeItemValue(obj, tc);
+}
+
+static int NpyArr_iterNextItem(JSOBJ obj, JSONTypeContext *tc) {
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+
+  if (PyErr_Occurred()) {
+    return 0;
+  }
+
+  if (npyarr->index[npyarr->stridedim] >= npyarr->dim) {
+    return 0;
+  }
+
+  NpyArr_freeItemValue(obj, tc);
+
+  if (!PyArray_Check(npyarr->array)) {
+    PyErr_SetString(PyExc_TypeError,
+                    "NpyArr_iterNextItem received a non-array object");
+    return 0;
+  }
+  PyArrayObject *arrayobj = (PyArrayObject *)npyarr->array;
+
+  if (PyArray_ISDATETIME(arrayobj)) {
+    GET_TC(tc)->itemValue = obj;
+    Py_INCREF(obj);
+    ((PyObjectEncoder *)tc->encoder)->npyType = PyArray_TYPE(arrayobj);
+    // Also write the resolution (unit) of the ndarray
+    PyArray_Descr *dtype = PyArray_DESCR(arrayobj);
+    ((PyObjectEncoder *)tc->encoder)->valueUnit =
+        get_datetime_metadata_from_dtype(dtype).base;
+    ((PyObjectEncoder *)tc->encoder)->npyValue = npyarr->dataptr;
+    ((PyObjectEncoder *)tc->encoder)->npyCtxtPassthru = npyarr;
+  } else {
+    GET_TC(tc)->itemValue = PyArray_GETITEM(arrayobj, npyarr->dataptr);
+  }
+
+  npyarr->dataptr += npyarr->stride;
+  npyarr->index[npyarr->stridedim]++;
+  return 1;
+}
+
+static int NpyArr_iterNext(JSOBJ _obj, JSONTypeContext *tc) {
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+
+  if (PyErr_Occurred()) {
+    return 0;
+  }
+
+  if (npyarr->curdim >= npyarr->ndim ||
+      npyarr->index[npyarr->stridedim] >= npyarr->dim) {
+    // innermost dimension, start retrieving item values
+    GET_TC(tc)->iterNext = NpyArr_iterNextItem;
+    return NpyArr_iterNextItem(_obj, tc);
+  }
+
+  // dig a dimension deeper
+  npyarr->index[npyarr->stridedim]++;
+
+  npyarr->curdim++;
+  npyarr->stridedim += npyarr->inc;
+  if (!PyArray_Check(npyarr->array)) {
+    PyErr_SetString(PyExc_TypeError,
+                    "NpyArr_iterNext received a non-array object");
+    return 0;
+  }
+  const PyArrayObject *arrayobj = (const PyArrayObject *)npyarr->array;
+
+  npyarr->dim = PyArray_DIM(arrayobj, (int)npyarr->stridedim);
+  npyarr->stride = PyArray_STRIDE(arrayobj, (int)npyarr->stridedim);
+  npyarr->index[npyarr->stridedim] = 0;
+
+  ((PyObjectEncoder *)tc->encoder)->npyCtxtPassthru = npyarr;
+  GET_TC(tc)->itemValue = npyarr->array;
+  return 1;
+}
+
+static JSOBJ NpyArr_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *NpyArr_iterGetName(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc,
+                                      size_t *outLen) {
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+  const char *cStr;
+
+  if (GET_TC(tc)->iterNext == NpyArr_iterNextItem) {
+    const npy_intp idx = npyarr->index[npyarr->stridedim] - 1;
+    cStr = npyarr->columnLabels[idx];
+  } else {
+    const npy_intp idx = npyarr->index[npyarr->stridedim - npyarr->inc] - 1;
+    cStr = npyarr->rowLabels[idx];
+  }
+
+  *outLen = strlen(cStr);
+
+  return cStr;
+}
+
+//=============================================================================
+// Pandas block iteration functions
+//
+// Serialises a DataFrame column by column to avoid unnecessary data copies and
+// more representative serialisation when dealing with mixed dtypes.
+//
+// Uses a dedicated NpyArrContext for each column.
+//=============================================================================
+
+static void PdBlockPassThru_iterEnd(JSOBJ obj, JSONTypeContext *tc) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+
+  if (blkCtxt->transpose) {
+    blkCtxt->colIdx++;
+  } else {
+    blkCtxt->colIdx = 0;
+  }
+
+  NpyArr_freeItemValue(obj, tc);
+}
+
+static int PdBlock_iterNextItem(JSOBJ obj, JSONTypeContext *tc) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+
+  if (blkCtxt->colIdx >= blkCtxt->ncols) {
+    return 0;
+  }
+
+  GET_TC(tc)->npyarr = blkCtxt->npyCtxts[blkCtxt->colIdx];
+  blkCtxt->colIdx++;
+  return NpyArr_iterNextItem(obj, tc);
+}
+
+static const char *PdBlock_iterGetName(JSOBJ Py_UNUSED(obj),
+                                       JSONTypeContext *tc, size_t *outLen) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+  NpyArrContext *npyarr = blkCtxt->npyCtxts[0];
+  const char *cStr;
+
+  if (GET_TC(tc)->iterNext == PdBlock_iterNextItem) {
+    const npy_intp idx = blkCtxt->colIdx - 1;
+    cStr = npyarr->columnLabels[idx];
+  } else {
+    const npy_intp idx =
+        GET_TC(tc)->iterNext != PdBlock_iterNext
+            ? npyarr->index[npyarr->stridedim - npyarr->inc] - 1
+            : npyarr->index[npyarr->stridedim];
+
+    cStr = npyarr->rowLabels[idx];
+  }
+
+  *outLen = strlen(cStr);
+  return cStr;
+}
+
+static const char *PdBlock_iterGetName_Transpose(JSOBJ Py_UNUSED(obj),
+                                                 JSONTypeContext *tc,
+                                                 size_t *outLen) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+  NpyArrContext *npyarr = blkCtxt->npyCtxts[blkCtxt->colIdx];
+  const char *cStr;
+
+  if (GET_TC(tc)->iterNext == NpyArr_iterNextItem) {
+    const npy_intp idx = npyarr->index[npyarr->stridedim] - 1;
+    cStr = npyarr->columnLabels[idx];
+  } else {
+    const npy_intp idx = blkCtxt->colIdx;
+    cStr = npyarr->rowLabels[idx];
+  }
+
+  *outLen = strlen(cStr);
+  return cStr;
+}
+
+static int PdBlock_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+
+  if (PyErr_Occurred() || ((JSONObjectEncoder *)tc->encoder)->errorMsg) {
+    return 0;
+  }
+
+  if (blkCtxt->transpose) {
+    if (blkCtxt->colIdx >= blkCtxt->ncols) {
+      return 0;
+    }
+  } else {
+    const NpyArrContext *npyarr = blkCtxt->npyCtxts[0];
+    if (npyarr->index[npyarr->stridedim] >= npyarr->dim) {
+      return 0;
+    }
+  }
+
+  ((PyObjectEncoder *)tc->encoder)->blkCtxtPassthru = blkCtxt;
+  GET_TC(tc)->itemValue = obj;
+
+  return 1;
+}
+
+static void PdBlockPassThru_iterBegin(JSOBJ Py_UNUSED(obj),
+                                      JSONTypeContext *tc) {
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+
+  if (blkCtxt->transpose) {
+    // if transposed we exhaust each column before moving to the next
+    GET_TC(tc)->iterNext = NpyArr_iterNextItem;
+    GET_TC(tc)->iterGetName = PdBlock_iterGetName_Transpose;
+    GET_TC(tc)->npyarr = blkCtxt->npyCtxts[blkCtxt->colIdx];
+  }
+}
+
+static void PdBlock_iterBegin(JSOBJ _obj, JSONTypeContext *tc) {
+  PyObject *obj = (PyObject *)_obj;
+
+  GET_TC(tc)->iterGetName = GET_TC(tc)->transpose
+                                ? PdBlock_iterGetName_Transpose
+                                : PdBlock_iterGetName;
+
+  PdBlockContext *blkCtxt = PyObject_Malloc(sizeof(PdBlockContext));
+  if (!blkCtxt) {
+    PyErr_NoMemory();
+    GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+    return;
+  }
+  GET_TC(tc)->pdblock = blkCtxt;
+
+  blkCtxt->colIdx = 0;
+  blkCtxt->transpose = GET_TC(tc)->transpose;
+  blkCtxt->ncols = get_attr_length(obj, "columns");
+
+  if (blkCtxt->ncols == 0) {
+    blkCtxt->npyCtxts = NULL;
+
+    GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+    return;
+  }
+
+  blkCtxt->npyCtxts = PyObject_Malloc(sizeof(NpyArrContext *) * blkCtxt->ncols);
+  if (!blkCtxt->npyCtxts) {
+    PyErr_NoMemory();
+    GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+    return;
+  }
+
+  PyObject *arrays = get_sub_attr(obj, "_mgr", "column_arrays");
+  if (!arrays) {
+    GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+    return;
+  }
+
+  for (Py_ssize_t i = 0; i < PyObject_Length(arrays); i++) {
+    PyObject *array = PyList_GET_ITEM(arrays, i);
+    if (!array) {
+      GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+      goto ARR_RET;
+    }
+
+    // ensure we have a numpy array (i.e. np.asarray)
+    PyObject *values = PyObject_CallMethod(array, "__array__", NULL);
+    if ((!values) || (!PyArray_CheckExact(values))) {
+      // Didn't get a numpy array
+      ((JSONObjectEncoder *)tc->encoder)->errorMsg = "";
+      GET_TC(tc)->iterNext = NpyArr_iterNextNone;
+      goto ARR_RET;
+    }
+
+    GET_TC(tc)->newObj = values;
+
+    // init a dedicated context for this column
+    NpyArr_iterBegin(obj, tc);
+
+    GET_TC(tc)->itemValue = NULL;
+    ((PyObjectEncoder *)tc->encoder)->npyCtxtPassthru = NULL;
+
+    blkCtxt->npyCtxts[i] = GET_TC(tc)->npyarr;
+    GET_TC(tc)->newObj = NULL;
+  }
+  GET_TC(tc)->npyarr = blkCtxt->npyCtxts[0];
+  goto ARR_RET;
+
+ARR_RET:
+  Py_DECREF(arrays);
+}
+
+static void PdBlock_iterEnd(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->itemValue = NULL;
+  NpyArrContext *npyarr = GET_TC(tc)->npyarr;
+  PdBlockContext *blkCtxt = GET_TC(tc)->pdblock;
+
+  if (blkCtxt) {
+    for (int i = 0; i < blkCtxt->ncols; i++) {
+      npyarr = blkCtxt->npyCtxts[i];
+      if (npyarr) {
+        if (npyarr->array) {
+          Py_DECREF(npyarr->array);
+          npyarr->array = NULL;
+        }
+
+        GET_TC(tc)->npyarr = npyarr;
+        NpyArr_iterEnd(obj, tc);
+
+        blkCtxt->npyCtxts[i] = NULL;
+      }
+    }
+
+    if (blkCtxt->npyCtxts) {
+      PyObject_Free(blkCtxt->npyCtxts);
+    }
+    PyObject_Free(blkCtxt);
+  }
+}
+
+//=============================================================================
+// Tuple iteration functions
+// itemValue is borrowed reference, no ref counting
+//=============================================================================
+static void Tuple_iterBegin(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->index = 0;
+  GET_TC(tc)->size = PyTuple_GET_SIZE((PyObject *)obj);
+  GET_TC(tc)->itemValue = NULL;
+}
+
+static int Tuple_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+
+  if (GET_TC(tc)->index >= GET_TC(tc)->size) {
+    return 0;
+  }
+
+  PyObject *item = PyTuple_GET_ITEM(obj, GET_TC(tc)->index);
+
+  GET_TC(tc)->itemValue = item;
+  GET_TC(tc)->index++;
+  return 1;
+}
+
+static void Tuple_iterEnd(JSOBJ Py_UNUSED(obj),
+                          JSONTypeContext *Py_UNUSED(tc)) {}
+
+static JSOBJ Tuple_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Tuple_iterGetName(JSOBJ Py_UNUSED(obj),
+                                     JSONTypeContext *Py_UNUSED(tc),
+                                     size_t *Py_UNUSED(outLen)) {
+  return NULL;
+}
+
+//=============================================================================
+// Set iteration functions
+// itemValue is borrowed reference, no ref counting
+//=============================================================================
+static void Set_iterBegin(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->itemValue = NULL;
+  GET_TC(tc)->iterator = PyObject_GetIter(obj);
+}
+
+static int Set_iterNext(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->itemValue) {
+    Py_DECREF(GET_TC(tc)->itemValue);
+    GET_TC(tc)->itemValue = NULL;
+  }
+
+  PyObject *item = PyIter_Next(GET_TC(tc)->iterator);
+
+  if (item == NULL) {
+    return 0;
+  }
+
+  GET_TC(tc)->itemValue = item;
+  return 1;
+}
+
+static void Set_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->itemValue) {
+    Py_DECREF(GET_TC(tc)->itemValue);
+    GET_TC(tc)->itemValue = NULL;
+  }
+
+  if (GET_TC(tc)->iterator) {
+    Py_DECREF(GET_TC(tc)->iterator);
+    GET_TC(tc)->iterator = NULL;
+  }
+}
+
+static JSOBJ Set_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Set_iterGetName(JSOBJ Py_UNUSED(obj),
+                                   JSONTypeContext *Py_UNUSED(tc),
+                                   size_t *Py_UNUSED(outLen)) {
+  return NULL;
+}
+
+//=============================================================================
+// Dir iteration functions
+// itemName ref is borrowed from PyObject_Dir (attrList). No refcount
+// itemValue ref is from PyObject_GetAttr. Ref counted
+//=============================================================================
+static void Dir_iterBegin(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->attrList = PyObject_Dir(obj);
+  GET_TC(tc)->index = 0;
+  GET_TC(tc)->size = PyList_GET_SIZE(GET_TC(tc)->attrList);
+}
+
+static void Dir_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->itemValue) {
+    Py_DECREF(GET_TC(tc)->itemValue);
+    GET_TC(tc)->itemValue = NULL;
+  }
+
+  if (GET_TC(tc)->itemName) {
+    Py_DECREF(GET_TC(tc)->itemName);
+    GET_TC(tc)->itemName = NULL;
+  }
+
+  Py_DECREF((PyObject *)GET_TC(tc)->attrList);
+}
+
+static int Dir_iterNext(JSOBJ _obj, JSONTypeContext *tc) {
+  PyObject *obj = (PyObject *)_obj;
+  PyObject *itemValue = GET_TC(tc)->itemValue;
+  PyObject *itemName = GET_TC(tc)->itemName;
+
+  if (PyErr_Occurred() || ((JSONObjectEncoder *)tc->encoder)->errorMsg) {
+    return 0;
+  }
+
+  if (itemValue) {
+    Py_DECREF(GET_TC(tc)->itemValue);
+    GET_TC(tc)->itemValue = itemValue = NULL;
+  }
+
+  if (itemName) {
+    Py_DECREF(GET_TC(tc)->itemName);
+    GET_TC(tc)->itemName = itemName = NULL;
+  }
+
+  for (; GET_TC(tc)->index < GET_TC(tc)->size; GET_TC(tc)->index++) {
+    PyObject *attrName =
+        PyList_GET_ITEM(GET_TC(tc)->attrList, GET_TC(tc)->index);
+    PyObject *attr = PyUnicode_AsUTF8String(attrName);
+    const char *attrStr = PyBytes_AS_STRING(attr);
+
+    if (attrStr[0] == '_') {
+      Py_DECREF(attr);
+      continue;
+    }
+
+    itemValue = PyObject_GetAttr(obj, attrName);
+    if (itemValue == NULL) {
+      PyErr_Clear();
+      Py_DECREF(attr);
+      continue;
+    }
+
+    if (PyCallable_Check(itemValue)) {
+      Py_DECREF(itemValue);
+      Py_DECREF(attr);
+      continue;
+    }
+
+    GET_TC(tc)->itemName = itemName;
+    GET_TC(tc)->itemValue = itemValue;
+
+    itemName = attr;
+    break;
+  }
+
+  if (itemName == NULL) {
+    GET_TC(tc)->index = GET_TC(tc)->size;
+    GET_TC(tc)->itemValue = NULL;
+    return 0;
+  }
+
+  GET_TC(tc)->itemName = itemName;
+  GET_TC(tc)->itemValue = itemValue;
+  GET_TC(tc)->index++;
+
+  return 1;
+}
+
+static JSOBJ Dir_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Dir_iterGetName(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc,
+                                   size_t *outLen) {
+  *outLen = PyBytes_GET_SIZE(GET_TC(tc)->itemName);
+  return PyBytes_AS_STRING(GET_TC(tc)->itemName);
+}
+
+//=============================================================================
+// List iteration functions
+// itemValue is borrowed from object (which is list). No refcounting
+//=============================================================================
+static void List_iterBegin(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->index = 0;
+  GET_TC(tc)->size = PyList_GET_SIZE((PyObject *)obj);
+}
+
+static int List_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  if (GET_TC(tc)->index >= GET_TC(tc)->size) {
+    return 0;
+  }
+
+  GET_TC(tc)->itemValue = PyList_GET_ITEM(obj, GET_TC(tc)->index);
+  GET_TC(tc)->index++;
+  return 1;
+}
+
+static void List_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *Py_UNUSED(tc)) {
+}
+
+static JSOBJ List_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *List_iterGetName(JSOBJ Py_UNUSED(obj),
+                                    JSONTypeContext *Py_UNUSED(tc),
+                                    size_t *Py_UNUSED(outLen)) {
+  return NULL;
+}
+
+//=============================================================================
+// pandas Index iteration functions
+//=============================================================================
+static void Index_iterBegin(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  GET_TC(tc)->index = 0;
+}
+
+static int Index_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  const Py_ssize_t index = GET_TC(tc)->index;
+  Py_XDECREF(GET_TC(tc)->itemValue);
+  if (index == 0) {
+    GET_TC(tc)->cStr = "name";
+    GET_TC(tc)->itemValue = PyObject_GetAttrString(obj, "name");
+  } else if (index == 1) {
+    GET_TC(tc)->cStr = "data";
+    GET_TC(tc)->itemValue = get_values(obj);
+    if (!GET_TC(tc)->itemValue) {
+      return 0;
+    }
+  } else {
+    return 0;
+  }
+
+  GET_TC(tc)->index++;
+  return 1;
+}
+
+static void Index_iterEnd(JSOBJ Py_UNUSED(obj),
+                          JSONTypeContext *Py_UNUSED(tc)) {}
+
+static JSOBJ Index_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Index_iterGetName(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc,
+                                     size_t *outLen) {
+  *outLen = strlen(GET_TC(tc)->cStr);
+  return GET_TC(tc)->cStr;
+}
+
+//=============================================================================
+// pandas Series iteration functions
+//=============================================================================
+static void Series_iterBegin(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  PyObjectEncoder *enc = (PyObjectEncoder *)tc->encoder;
+  GET_TC(tc)->index = 0;
+  enc->outputFormat = VALUES; // for contained series
+}
+
+static int Series_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  const Py_ssize_t index = GET_TC(tc)->index;
+  Py_XDECREF(GET_TC(tc)->itemValue);
+  if (index == 0) {
+    GET_TC(tc)->cStr = "name";
+    GET_TC(tc)->itemValue = PyObject_GetAttrString(obj, "name");
+  } else if (index == 1) {
+    GET_TC(tc)->cStr = "index";
+    GET_TC(tc)->itemValue = PyObject_GetAttrString(obj, "index");
+  } else if (index == 2) {
+    GET_TC(tc)->cStr = "data";
+    GET_TC(tc)->itemValue = get_values(obj);
+    if (!GET_TC(tc)->itemValue) {
+      return 0;
+    }
+  } else {
+    return 0;
+  }
+
+  GET_TC(tc)->index++;
+  return 1;
+}
+
+static void Series_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  PyObjectEncoder *enc = (PyObjectEncoder *)tc->encoder;
+  enc->outputFormat = enc->originalOutputFormat;
+}
+
+static JSOBJ Series_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Series_iterGetName(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc,
+                                      size_t *outLen) {
+  *outLen = strlen(GET_TC(tc)->cStr);
+  return GET_TC(tc)->cStr;
+}
+
+//=============================================================================
+// pandas DataFrame iteration functions
+//=============================================================================
+static void DataFrame_iterBegin(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  PyObjectEncoder *enc = (PyObjectEncoder *)tc->encoder;
+  GET_TC(tc)->index = 0;
+  enc->outputFormat = VALUES; // for contained series & index
+}
+
+static int DataFrame_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  const Py_ssize_t index = GET_TC(tc)->index;
+  Py_XDECREF(GET_TC(tc)->itemValue);
+  if (index == 0) {
+    GET_TC(tc)->cStr = "columns";
+    GET_TC(tc)->itemValue = PyObject_GetAttrString(obj, "columns");
+  } else if (index == 1) {
+    GET_TC(tc)->cStr = "index";
+    GET_TC(tc)->itemValue = PyObject_GetAttrString(obj, "index");
+  } else if (index == 2) {
+    GET_TC(tc)->cStr = "data";
+    Py_INCREF(obj);
+    GET_TC(tc)->itemValue = obj;
+  } else {
+    return 0;
+  }
+
+  GET_TC(tc)->index++;
+  return 1;
+}
+
+static void DataFrame_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  PyObjectEncoder *enc = (PyObjectEncoder *)tc->encoder;
+  enc->outputFormat = enc->originalOutputFormat;
+}
+
+static JSOBJ DataFrame_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *DataFrame_iterGetName(JSOBJ Py_UNUSED(obj),
+                                         JSONTypeContext *tc, size_t *outLen) {
+  *outLen = strlen(GET_TC(tc)->cStr);
+  return GET_TC(tc)->cStr;
+}
+
+//=============================================================================
+// Dict iteration functions
+// itemName might converted to string (Python_Str). Do refCounting
+// itemValue is borrowed from object (which is dict). No refCounting
+//=============================================================================
+static void Dict_iterBegin(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  GET_TC(tc)->index = 0;
+}
+
+static int Dict_iterNext(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->itemName) {
+    Py_DECREF(GET_TC(tc)->itemName);
+    GET_TC(tc)->itemName = NULL;
+  }
+
+  if (!PyDict_Next((PyObject *)GET_TC(tc)->dictObj, &GET_TC(tc)->index,
+                   &GET_TC(tc)->itemName, &GET_TC(tc)->itemValue)) {
+    return 0;
+  }
+
+  if (PyUnicode_Check(GET_TC(tc)->itemName)) {
+    GET_TC(tc)->itemName = PyUnicode_AsUTF8String(GET_TC(tc)->itemName);
+  } else if (!PyBytes_Check(GET_TC(tc)->itemName)) {
+    GET_TC(tc)->itemName = PyObject_Str(GET_TC(tc)->itemName);
+    PyObject *itemNameTmp = GET_TC(tc)->itemName;
+    GET_TC(tc)->itemName = PyUnicode_AsUTF8String(GET_TC(tc)->itemName);
+    Py_DECREF(itemNameTmp);
+  } else {
+    Py_INCREF(GET_TC(tc)->itemName);
+  }
+  return 1;
+}
+
+static void Dict_iterEnd(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (GET_TC(tc)->itemName) {
+    Py_DECREF(GET_TC(tc)->itemName);
+    GET_TC(tc)->itemName = NULL;
+  }
+  Py_DECREF(GET_TC(tc)->dictObj);
+}
+
+static JSOBJ Dict_iterGetValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->itemValue;
+}
+
+static const char *Dict_iterGetName(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc,
+                                    size_t *outLen) {
+  *outLen = PyBytes_GET_SIZE(GET_TC(tc)->itemName);
+  return PyBytes_AS_STRING(GET_TC(tc)->itemName);
+}
+
+static void NpyArr_freeLabels(char **labels, npy_intp len) {
+  if (labels) {
+    for (npy_intp i = 0; i < len; i++) {
+      PyObject_Free(labels[i]);
+    }
+    PyObject_Free(labels);
+  }
+}
+
+/*
+ * Function: NpyArr_encodeLabels
+ * -----------------------------
+ *
+ * Builds an array of "encoded" labels.
+ *
+ * labels: PyArrayObject pointer for labels to be "encoded"
+ * num : number of labels
+ *
+ * "encode" is quoted above because we aren't really doing encoding
+ * For historical reasons this function would actually encode the entire
+ * array into a separate buffer with a separate call to JSON_Encode
+ * and would leave it to complex pointer manipulation from there to
+ * unpack values as needed. To make things simpler and more idiomatic
+ * this has instead just stringified any input save for datetime values,
+ * which may need to be represented in various formats.
+ */
+static char **NpyArr_encodeLabels(PyArrayObject *labels, PyObjectEncoder *enc,
+                                  npy_intp num) {
+  // NOTE this function steals a reference to labels.
+  PyObject *item = NULL;
+  const NPY_DATETIMEUNIT base = enc->datetimeUnit;
+
+  if (!labels) {
+    return 0;
+  }
+
+  if (PyArray_SIZE(labels) < num) {
+    PyErr_SetString(PyExc_ValueError,
+                    "Label array sizes do not match corresponding data shape");
+    Py_DECREF(labels);
+    return 0;
+  }
+
+  char **ret = PyObject_Malloc(sizeof(char *) * num);
+  if (!ret) {
+    PyErr_NoMemory();
+    Py_DECREF(labels);
+    return 0;
+  }
+
+  for (npy_intp i = 0; i < num; i++) {
+    ret[i] = NULL;
+  }
+
+  const npy_intp stride = PyArray_STRIDE(labels, 0);
+  char *dataptr = PyArray_DATA(labels);
+  const int type_num = PyArray_TYPE(labels);
+  PyArray_Descr *dtype = PyArray_DESCR(labels);
+
+  for (npy_intp i = 0; i < num; i++) {
+    item = PyArray_GETITEM(labels, dataptr);
+    if (!item) {
+      NpyArr_freeLabels(ret, num);
+      ret = 0;
+      break;
+    }
+
+    int is_datetimelike = 0;
+    int64_t i8date;
+    NPY_DATETIMEUNIT dateUnit = NPY_FR_ns;
+    if (PyTypeNum_ISDATETIME(type_num)) {
+      is_datetimelike = 1;
+      i8date = *(int64_t *)dataptr;
+      dateUnit = get_datetime_metadata_from_dtype(dtype).base;
+    } else if (PyDate_Check(item) || PyDelta_Check(item)) {
+      is_datetimelike = 1;
+      if (PyObject_HasAttrString(item, "_value")) {
+        // pd.Timestamp object or pd.NaT
+        // see test_date_index_and_values for case with non-nano
+        i8date = get_long_attr(item, "_value");
+      } else {
+        if (PyDelta_Check(item)) {
+          // TODO(anyone): cast below loses precision if total_seconds return
+          // value exceeds number of bits that significand can hold
+          // also liable to overflow
+          i8date = (int64_t)(total_seconds(item) *
+                             1000000000LL); // nanoseconds per second
+        } else {
+          // datetime.* objects don't follow above rules
+          i8date = PyDateTimeToEpoch(item, NPY_FR_ns);
+        }
+      }
+    }
+
+    size_t len;
+    char *cLabel;
+    if (is_datetimelike) {
+      if (i8date == get_nat()) {
+        len = 4;
+        cLabel = PyObject_Malloc(len + 1);
+        strncpy(cLabel, "null", len + 1);
+      } else {
+        if (enc->datetimeIso) {
+          if ((type_num == NPY_TIMEDELTA) || (PyDelta_Check(item))) {
+            // TODO(username): non-nano timedelta support?
+            cLabel = int64ToIsoDuration(i8date, &len);
+          } else {
+            if (type_num == NPY_DATETIME) {
+              cLabel = int64ToIso(i8date, dateUnit, base, &len);
+            } else {
+              cLabel = PyDateTimeToIso(item, base, &len);
+            }
+          }
+          if (cLabel == NULL) {
+            Py_DECREF(item);
+            NpyArr_freeLabels(ret, num);
+            ret = 0;
+            break;
+          }
+        } else {
+          int size_of_cLabel = 21; // 21 chars for int 64
+          cLabel = PyObject_Malloc(size_of_cLabel);
+          if (scaleNanosecToUnit(&i8date, base) == -1) {
+            NpyArr_freeLabels(ret, num);
+            ret = 0;
+            break;
+          }
+          snprintf(cLabel, size_of_cLabel, "%" PRId64, i8date);
+          len = strlen(cLabel);
+        }
+      }
+    } else { // Fallback to string representation
+      // Replace item with the string to keep it alive.
+      Py_SETREF(item, PyObject_Str(item));
+      if (item == NULL) {
+        NpyArr_freeLabels(ret, num);
+        ret = 0;
+        break;
+      }
+
+      cLabel = (char *)PyUnicode_AsUTF8(item);
+      len = strlen(cLabel);
+    }
+
+    // Add 1 to include NULL terminator
+    ret[i] = PyObject_Malloc(len + 1);
+    memcpy(ret[i], cLabel, len + 1);
+    Py_DECREF(item);
+
+    if (is_datetimelike) {
+      PyObject_Free(cLabel);
+    }
+
+    if (PyErr_Occurred()) {
+      NpyArr_freeLabels(ret, num);
+      ret = 0;
+      break;
+    }
+
+    if (!ret[i]) {
+      PyErr_NoMemory();
+      ret = 0;
+      break;
+    }
+
+    dataptr += stride;
+  }
+
+  Py_DECREF(labels);
+  return ret;
+}
+
+static void Object_invokeDefaultHandler(PyObject *obj, PyObjectEncoder *enc) {
+  PyObject *tmpObj = NULL;
+  tmpObj = PyObject_CallFunctionObjArgs(enc->defaultHandler, obj, NULL);
+  if (!PyErr_Occurred()) {
+    if (tmpObj == NULL) {
+      PyErr_SetString(PyExc_TypeError, "Failed to execute default handler");
+    } else {
+      encode(tmpObj, (JSONObjectEncoder *)enc, NULL, 0);
+    }
+  }
+  Py_XDECREF(tmpObj);
+  return;
+}
+
+static void Object_beginTypeContext(JSOBJ _obj, JSONTypeContext *tc) {
+  tc->prv = NULL;
+
+  if (!_obj) {
+    tc->type = JT_INVALID;
+    return;
+  }
+
+  PyObject *obj = (PyObject *)_obj;
+  PyObjectEncoder *enc = (PyObjectEncoder *)tc->encoder;
+
+  if (PyBool_Check(obj)) {
+    tc->type = (obj == Py_True) ? JT_TRUE : JT_FALSE;
+    return;
+  } else if (obj == Py_None) {
+    tc->type = JT_NULL;
+    return;
+  }
+
+  TypeContext *pc = createTypeContext();
+  if (!pc) {
+    tc->type = JT_INVALID;
+    return;
+  }
+  tc->prv = pc;
+
+  if (PyTypeNum_ISDATETIME(enc->npyType)) {
+    int64_t longVal = *(npy_int64 *)enc->npyValue;
+    if (longVal == get_nat()) {
+      tc->type = JT_NULL;
+    } else {
+      if (enc->datetimeIso) {
+        if (enc->npyType == NPY_TIMEDELTA) {
+          pc->PyTypeToUTF8 = NpyTimeDeltaToIsoCallback;
+        } else {
+          pc->PyTypeToUTF8 = NpyDateTimeToIsoCallback;
+        }
+        // Currently no way to pass longVal to iso function, so use
+        // state management
+        pc->longValue = longVal;
+        tc->type = JT_UTF8;
+      } else {
+        NPY_DATETIMEUNIT base = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+        if (scaleNanosecToUnit(&longVal, base) == -1) {
+          goto INVALID;
+        }
+        pc->longValue = longVal;
+        tc->type = JT_LONG;
+      }
+    }
+
+    // TODO(username): this prevents infinite loop with
+    // mixed-type DataFrames;
+    // refactor
+    enc->npyCtxtPassthru = NULL;
+    enc->npyType = -1;
+    return;
+  }
+
+  if (PyIter_Check(obj) || (PyArray_Check(obj) && !PyArray_CheckScalar(obj))) {
+    goto ISITERABLE;
+  }
+
+  if (PyLong_Check(obj)) {
+    tc->type = JT_LONG;
+    int overflow = 0;
+    pc->longValue = PyLong_AsLongLongAndOverflow(obj, &overflow);
+    int err;
+    err = (pc->longValue == -1) && PyErr_Occurred();
+
+    if (overflow) {
+      tc->type = JT_BIGNUM;
+    } else if (err) {
+      goto INVALID;
+    }
+
+    return;
+  } else if (PyFloat_Check(obj)) {
+    const double val = PyFloat_AS_DOUBLE(obj);
+    if (npy_isnan(val) || npy_isinf(val)) {
+      tc->type = JT_NULL;
+    } else {
+      pc->doubleValue = val;
+      tc->type = JT_DOUBLE;
+    }
+    return;
+  } else if (PyBytes_Check(obj)) {
+    pc->PyTypeToUTF8 = PyBytesToUTF8;
+    tc->type = JT_UTF8;
+    return;
+  } else if (PyUnicode_Check(obj)) {
+    pc->PyTypeToUTF8 = PyUnicodeToUTF8;
+    tc->type = JT_UTF8;
+    return;
+  } else if (object_is_decimal_type(obj)) {
+    PyObject *is_nan_py = PyObject_RichCompare(obj, obj, Py_NE);
+    if (is_nan_py == NULL) {
+      goto INVALID;
+    }
+    int is_nan = (is_nan_py == Py_True);
+    Py_DECREF(is_nan_py);
+    if (is_nan) {
+      tc->type = JT_NULL;
+      return;
+    }
+    pc->PyTypeToUTF8 = PyDecimalToUTF8Callback;
+    tc->type = JT_UTF8;
+    return;
+  } else if (PyDateTime_Check(obj) || PyDate_Check(obj)) {
+    if (object_is_nat_type(obj)) {
+      tc->type = JT_NULL;
+      return;
+    }
+
+    if (enc->datetimeIso) {
+      pc->PyTypeToUTF8 = PyDateTimeToIsoCallback;
+      tc->type = JT_UTF8;
+    } else {
+      NPY_DATETIMEUNIT base = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+      pc->longValue = PyDateTimeToEpoch(obj, base);
+      tc->type = JT_LONG;
+    }
+    return;
+  } else if (PyTime_Check(obj)) {
+    pc->PyTypeToUTF8 = PyTimeToJSON;
+    tc->type = JT_UTF8;
+    return;
+  } else if (PyArray_IsScalar(obj, Datetime)) {
+    npy_int64 longVal;
+    if (((PyDatetimeScalarObject *)obj)->obval == get_nat()) {
+      tc->type = JT_NULL;
+      return;
+    }
+    PyArray_Descr *dtype = PyArray_DescrFromScalar(obj);
+    if (!PyTypeNum_ISDATETIME(dtype->type_num)) {
+      PyErr_Format(PyExc_ValueError, "Could not get resolution of datetime");
+      return;
+    }
+
+    PyArray_Descr *outcode = PyArray_DescrFromType(NPY_INT64);
+    PyArray_CastScalarToCtype(obj, &longVal, outcode);
+    Py_DECREF(outcode);
+
+    if (enc->datetimeIso) {
+      GET_TC(tc)->longValue = longVal;
+      pc->PyTypeToUTF8 = NpyDateTimeToIsoCallback;
+      enc->valueUnit = get_datetime_metadata_from_dtype(dtype).base;
+      tc->type = JT_UTF8;
+    } else {
+      NPY_DATETIMEUNIT base = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+      pc->longValue = PyDateTimeToEpoch(obj, base);
+      tc->type = JT_LONG;
+    }
+    return;
+  } else if (PyDelta_Check(obj)) {
+    // pd.Timedelta object or pd.NaT should evaluate true here
+    // fallback to nanoseconds per sec for other objects
+    // TODO(anyone): cast below loses precision if total_seconds return
+    // value exceeds number of bits that significand can hold
+    // also liable to overflow
+    int64_t value = PyObject_HasAttrString(obj, "_value")
+                        ? get_long_attr(obj, "_value")
+                        : (int64_t)(total_seconds(obj) * 1000000000LL);
+
+    if (value == get_nat()) {
+      tc->type = JT_NULL;
+      return;
+    } else if (enc->datetimeIso) {
+      pc->PyTypeToUTF8 = NpyTimeDeltaToIsoCallback;
+      tc->type = JT_UTF8;
+    } else {
+      const int unit = ((PyObjectEncoder *)tc->encoder)->datetimeUnit;
+      if (scaleNanosecToUnit(&value, unit) != 0) {
+        // TODO(username): Add some kind of error handling here
+      }
+
+      if (PyErr_Occurred() && PyErr_ExceptionMatches(PyExc_OverflowError)) {
+        goto INVALID;
+      }
+
+      tc->type = JT_LONG;
+    }
+    pc->longValue = value;
+    return;
+  } else if (PyArray_IsScalar(obj, Integer)) {
+    tc->type = JT_LONG;
+    PyArray_CastScalarToCtype(obj, &(pc->longValue),
+                              PyArray_DescrFromType(NPY_INT64));
+
+    if (PyErr_Occurred() && PyErr_ExceptionMatches(PyExc_OverflowError)) {
+      goto INVALID;
+    }
+
+    return;
+  } else if (PyArray_IsScalar(obj, Bool)) {
+    PyArray_CastScalarToCtype(obj, &(pc->longValue),
+                              PyArray_DescrFromType(NPY_BOOL));
+    tc->type = (pc->longValue) ? JT_TRUE : JT_FALSE;
+    return;
+  } else if (PyArray_IsScalar(obj, Float) || PyArray_IsScalar(obj, Double)) {
+    PyArray_CastScalarToCtype(obj, &(pc->doubleValue),
+                              PyArray_DescrFromType(NPY_DOUBLE));
+    tc->type = JT_DOUBLE;
+    return;
+  } else if (PyArray_CheckScalar(obj)) {
+    PyErr_Format(PyExc_TypeError,
+                 "%R (numpy-scalar) is not JSON serializable at the moment",
+                 obj);
+    goto INVALID;
+  } else if (object_is_na_type(obj)) {
+    tc->type = JT_NULL;
+    return;
+  }
+
+ISITERABLE:
+
+  if (object_is_index_type(obj)) {
+    if (enc->outputFormat == SPLIT) {
+      tc->type = JT_OBJECT;
+      pc->iterBegin = Index_iterBegin;
+      pc->iterEnd = Index_iterEnd;
+      pc->iterNext = Index_iterNext;
+      pc->iterGetValue = Index_iterGetValue;
+      pc->iterGetName = Index_iterGetName;
+      return;
+    }
+
+    pc->newObj = get_values(obj);
+    if (pc->newObj) {
+      tc->type = JT_ARRAY;
+      pc->iterBegin = NpyArr_iterBegin;
+      pc->iterEnd = NpyArr_iterEnd;
+      pc->iterNext = NpyArr_iterNext;
+      pc->iterGetValue = NpyArr_iterGetValue;
+      pc->iterGetName = NpyArr_iterGetName;
+    } else {
+      goto INVALID;
+    }
+
+    return;
+  } else if (object_is_series_type(obj)) {
+    if (enc->outputFormat == SPLIT) {
+      tc->type = JT_OBJECT;
+      pc->iterBegin = Series_iterBegin;
+      pc->iterEnd = Series_iterEnd;
+      pc->iterNext = Series_iterNext;
+      pc->iterGetValue = Series_iterGetValue;
+      pc->iterGetName = Series_iterGetName;
+      return;
+    }
+
+    pc->newObj = get_values(obj);
+    if (!pc->newObj) {
+      goto INVALID;
+    }
+
+    if (enc->outputFormat == INDEX || enc->outputFormat == COLUMNS) {
+      tc->type = JT_OBJECT;
+      PyObject *tmpObj = PyObject_GetAttrString(obj, "index");
+      if (!tmpObj) {
+        goto INVALID;
+      }
+      PyObject *values = get_values(tmpObj);
+      Py_DECREF(tmpObj);
+      if (!values) {
+        goto INVALID;
+      }
+
+      if (!PyArray_Check(pc->newObj)) {
+        PyErr_SetString(PyExc_TypeError,
+                        "Object_beginTypeContext received a non-array object");
+        goto INVALID;
+      }
+      const PyArrayObject *arrayobj = (const PyArrayObject *)pc->newObj;
+      pc->columnLabelsLen = PyArray_DIM(arrayobj, 0);
+      pc->columnLabels = NpyArr_encodeLabels((PyArrayObject *)values, enc,
+                                             pc->columnLabelsLen);
+      if (!pc->columnLabels) {
+        goto INVALID;
+      }
+    } else {
+      tc->type = JT_ARRAY;
+    }
+    pc->iterBegin = NpyArr_iterBegin;
+    pc->iterEnd = NpyArr_iterEnd;
+    pc->iterNext = NpyArr_iterNext;
+    pc->iterGetValue = NpyArr_iterGetValue;
+    pc->iterGetName = NpyArr_iterGetName;
+    return;
+  } else if (PyArray_Check(obj)) {
+    if (enc->npyCtxtPassthru) {
+      pc->npyarr = enc->npyCtxtPassthru;
+      tc->type = (pc->npyarr->columnLabels ? JT_OBJECT : JT_ARRAY);
+
+      pc->iterBegin = NpyArrPassThru_iterBegin;
+      pc->iterNext = NpyArr_iterNext;
+      pc->iterEnd = NpyArrPassThru_iterEnd;
+      pc->iterGetValue = NpyArr_iterGetValue;
+      pc->iterGetName = NpyArr_iterGetName;
+
+      enc->npyCtxtPassthru = NULL;
+      return;
+    }
+
+    tc->type = JT_ARRAY;
+    pc->iterBegin = NpyArr_iterBegin;
+    pc->iterEnd = NpyArr_iterEnd;
+    pc->iterNext = NpyArr_iterNext;
+    pc->iterGetValue = NpyArr_iterGetValue;
+    pc->iterGetName = NpyArr_iterGetName;
+    return;
+  } else if (object_is_dataframe_type(obj)) {
+    if (enc->blkCtxtPassthru) {
+      pc->pdblock = enc->blkCtxtPassthru;
+      tc->type =
+          (pc->pdblock->npyCtxts[0]->columnLabels ? JT_OBJECT : JT_ARRAY);
+
+      pc->iterBegin = PdBlockPassThru_iterBegin;
+      pc->iterEnd = PdBlockPassThru_iterEnd;
+      pc->iterNext = PdBlock_iterNextItem;
+      pc->iterGetName = PdBlock_iterGetName;
+      pc->iterGetValue = NpyArr_iterGetValue;
+
+      enc->blkCtxtPassthru = NULL;
+      return;
+    }
+
+    if (enc->outputFormat == SPLIT) {
+      tc->type = JT_OBJECT;
+      pc->iterBegin = DataFrame_iterBegin;
+      pc->iterEnd = DataFrame_iterEnd;
+      pc->iterNext = DataFrame_iterNext;
+      pc->iterGetValue = DataFrame_iterGetValue;
+      pc->iterGetName = DataFrame_iterGetName;
+      return;
+    }
+
+    pc->iterBegin = PdBlock_iterBegin;
+    pc->iterEnd = PdBlock_iterEnd;
+    pc->iterNext = PdBlock_iterNext;
+    pc->iterGetName = PdBlock_iterGetName;
+    pc->iterGetValue = NpyArr_iterGetValue;
+
+    if (enc->outputFormat == VALUES) {
+      tc->type = JT_ARRAY;
+    } else if (enc->outputFormat == RECORDS) {
+      tc->type = JT_ARRAY;
+      PyObject *tmpObj = PyObject_GetAttrString(obj, "columns");
+      if (!tmpObj) {
+        goto INVALID;
+      }
+      PyObject *values = get_values(tmpObj);
+      if (!values) {
+        Py_DECREF(tmpObj);
+        goto INVALID;
+      }
+      pc->columnLabelsLen = PyObject_Size(tmpObj);
+      pc->columnLabels = NpyArr_encodeLabels((PyArrayObject *)values, enc,
+                                             pc->columnLabelsLen);
+      Py_DECREF(tmpObj);
+      if (!pc->columnLabels) {
+        goto INVALID;
+      }
+    } else if (enc->outputFormat == INDEX || enc->outputFormat == COLUMNS) {
+      tc->type = JT_OBJECT;
+      PyObject *tmpObj =
+          (enc->outputFormat == INDEX ? PyObject_GetAttrString(obj, "index")
+                                      : PyObject_GetAttrString(obj, "columns"));
+      if (!tmpObj) {
+        goto INVALID;
+      }
+      PyObject *values = get_values(tmpObj);
+      if (!values) {
+        Py_DECREF(tmpObj);
+        goto INVALID;
+      }
+      pc->rowLabelsLen = PyObject_Size(tmpObj);
+      pc->rowLabels =
+          NpyArr_encodeLabels((PyArrayObject *)values, enc, pc->rowLabelsLen);
+      Py_DECREF(tmpObj);
+      tmpObj =
+          (enc->outputFormat == INDEX ? PyObject_GetAttrString(obj, "columns")
+                                      : PyObject_GetAttrString(obj, "index"));
+      if (!tmpObj) {
+        NpyArr_freeLabels(pc->rowLabels, pc->rowLabelsLen);
+        pc->rowLabels = NULL;
+        goto INVALID;
+      }
+      values = get_values(tmpObj);
+      if (!values) {
+        Py_DECREF(tmpObj);
+        NpyArr_freeLabels(pc->rowLabels, pc->rowLabelsLen);
+        pc->rowLabels = NULL;
+        goto INVALID;
+      }
+      pc->columnLabelsLen = PyObject_Size(tmpObj);
+      pc->columnLabels = NpyArr_encodeLabels((PyArrayObject *)values, enc,
+                                             pc->columnLabelsLen);
+      Py_DECREF(tmpObj);
+      if (!pc->columnLabels) {
+        NpyArr_freeLabels(pc->rowLabels, pc->rowLabelsLen);
+        pc->rowLabels = NULL;
+        goto INVALID;
+      }
+
+      if (enc->outputFormat == COLUMNS) {
+        pc->transpose = 1;
+      }
+    } else {
+      goto INVALID;
+    }
+    return;
+  } else if (PyDict_Check(obj)) {
+    tc->type = JT_OBJECT;
+    pc->iterBegin = Dict_iterBegin;
+    pc->iterEnd = Dict_iterEnd;
+    pc->iterNext = Dict_iterNext;
+    pc->iterGetValue = Dict_iterGetValue;
+    pc->iterGetName = Dict_iterGetName;
+    pc->dictObj = obj;
+    Py_INCREF(obj);
+
+    return;
+  } else if (PyList_Check(obj)) {
+    tc->type = JT_ARRAY;
+    pc->iterBegin = List_iterBegin;
+    pc->iterEnd = List_iterEnd;
+    pc->iterNext = List_iterNext;
+    pc->iterGetValue = List_iterGetValue;
+    pc->iterGetName = List_iterGetName;
+    return;
+  } else if (PyTuple_Check(obj)) {
+    tc->type = JT_ARRAY;
+    pc->iterBegin = Tuple_iterBegin;
+    pc->iterEnd = Tuple_iterEnd;
+    pc->iterNext = Tuple_iterNext;
+    pc->iterGetValue = Tuple_iterGetValue;
+    pc->iterGetName = Tuple_iterGetName;
+    return;
+  } else if (PyAnySet_Check(obj)) {
+    tc->type = JT_ARRAY;
+    pc->iterBegin = Set_iterBegin;
+    pc->iterEnd = Set_iterEnd;
+    pc->iterNext = Set_iterNext;
+    pc->iterGetValue = Set_iterGetValue;
+    pc->iterGetName = Set_iterGetName;
+    return;
+  }
+
+  PyObject *toDictFunc = PyObject_GetAttrString(obj, "toDict");
+
+  if (toDictFunc) {
+    PyObject *tuple = PyTuple_New(0);
+    PyObject *toDictResult = PyObject_Call(toDictFunc, tuple, NULL);
+    Py_DECREF(tuple);
+    Py_DECREF(toDictFunc);
+
+    if (toDictResult == NULL) {
+      PyErr_Clear();
+      tc->type = JT_NULL;
+      return;
+    }
+
+    if (!PyDict_Check(toDictResult)) {
+      Py_DECREF(toDictResult);
+      tc->type = JT_NULL;
+      return;
+    }
+
+    tc->type = JT_OBJECT;
+    pc->iterBegin = Dict_iterBegin;
+    pc->iterEnd = Dict_iterEnd;
+    pc->iterNext = Dict_iterNext;
+    pc->iterGetValue = Dict_iterGetValue;
+    pc->iterGetName = Dict_iterGetName;
+    pc->dictObj = toDictResult;
+    return;
+  }
+
+  PyErr_Clear();
+
+  if (enc->defaultHandler) {
+    Object_invokeDefaultHandler(obj, enc);
+    goto INVALID;
+  }
+
+  tc->type = JT_OBJECT;
+  pc->iterBegin = Dir_iterBegin;
+  pc->iterEnd = Dir_iterEnd;
+  pc->iterNext = Dir_iterNext;
+  pc->iterGetValue = Dir_iterGetValue;
+  pc->iterGetName = Dir_iterGetName;
+  return;
+
+INVALID:
+  tc->type = JT_INVALID;
+  PyObject_Free(tc->prv);
+  tc->prv = NULL;
+  return;
+}
+
+static void Object_endTypeContext(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  if (tc->prv) {
+    Py_XDECREF(GET_TC(tc)->newObj);
+    GET_TC(tc)->newObj = NULL;
+    NpyArr_freeLabels(GET_TC(tc)->rowLabels, GET_TC(tc)->rowLabelsLen);
+    GET_TC(tc)->rowLabels = NULL;
+    NpyArr_freeLabels(GET_TC(tc)->columnLabels, GET_TC(tc)->columnLabelsLen);
+    GET_TC(tc)->columnLabels = NULL;
+    GET_TC(tc)->cStr = NULL;
+    PyObject_Free(tc->prv);
+    tc->prv = NULL;
+  }
+}
+
+static const char *Object_getStringValue(JSOBJ obj, JSONTypeContext *tc,
+                                         size_t *_outLen) {
+  return GET_TC(tc)->PyTypeToUTF8(obj, tc, _outLen);
+}
+
+static JSINT64 Object_getLongValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->longValue;
+}
+
+static double Object_getDoubleValue(JSOBJ Py_UNUSED(obj), JSONTypeContext *tc) {
+  return GET_TC(tc)->doubleValue;
+}
+
+static const char *Object_getBigNumStringValue(JSOBJ obj, JSONTypeContext *tc,
+                                               size_t *_outLen) {
+  PyObject *repr = PyObject_Str(obj);
+  const char *str = PyUnicode_AsUTF8AndSize(repr, (Py_ssize_t *)_outLen);
+  char *bytes = PyObject_Malloc(*_outLen + 1);
+  memcpy(bytes, str, *_outLen + 1);
+  GET_TC(tc)->cStr = bytes;
+
+  Py_DECREF(repr);
+
+  return GET_TC(tc)->cStr;
+}
+
+static void Object_releaseObject(JSOBJ _obj) { Py_DECREF((PyObject *)_obj); }
+
+static void Object_iterBegin(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->iterBegin(obj, tc);
+}
+
+static int Object_iterNext(JSOBJ obj, JSONTypeContext *tc) {
+  return GET_TC(tc)->iterNext(obj, tc);
+}
+
+static void Object_iterEnd(JSOBJ obj, JSONTypeContext *tc) {
+  GET_TC(tc)->iterEnd(obj, tc);
+}
+
+static JSOBJ Object_iterGetValue(JSOBJ obj, JSONTypeContext *tc) {
+  return GET_TC(tc)->iterGetValue(obj, tc);
+}
+
+static const char *Object_iterGetName(JSOBJ obj, JSONTypeContext *tc,
+                                      size_t *outLen) {
+  return GET_TC(tc)->iterGetName(obj, tc, outLen);
+}
+
+PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
+                    PyObject *kwargs) {
+  PyDateTime_IMPORT;
+  if (PyDateTimeAPI == NULL) {
+    return NULL;
+  }
+
+  PandasDateTime_IMPORT;
+  if (PandasDateTimeAPI == NULL) {
+    return NULL;
+  }
+
+  static char *kwlist[] = {"obj",
+                           "ensure_ascii",
+                           "double_precision",
+                           "encode_html_chars",
+                           "orient",
+                           "date_unit",
+                           "iso_dates",
+                           "default_handler",
+                           "indent",
+                           NULL};
+
+  PyObject *oinput = NULL;
+  PyObject *oensureAscii = NULL;
+  int idoublePrecision = 10; // default double precision setting
+  PyObject *oencodeHTMLChars = NULL;
+  char *sOrient = NULL;
+  char *sdateFormat = NULL;
+  PyObject *oisoDates = 0;
+  PyObject *odefHandler = 0;
+  int indent = 0;
+
+  PyObjectEncoder pyEncoder = {
+      {
+          .beginTypeContext = Object_beginTypeContext,
+          .endTypeContext = Object_endTypeContext,
+          .getStringValue = Object_getStringValue,
+          .getLongValue = Object_getLongValue,
+          .getIntValue = NULL,
+          .getDoubleValue = Object_getDoubleValue,
+          .getBigNumStringValue = Object_getBigNumStringValue,
+          .iterBegin = Object_iterBegin,
+          .iterNext = Object_iterNext,
+          .iterEnd = Object_iterEnd,
+          .iterGetValue = Object_iterGetValue,
+          .iterGetName = Object_iterGetName,
+          .releaseObject = Object_releaseObject,
+          .malloc = PyObject_Malloc,
+          .realloc = PyObject_Realloc,
+          .free = PyObject_Free,
+          .recursionMax = -1,
+          .doublePrecision = idoublePrecision,
+          .forceASCII = 1,
+          .encodeHTMLChars = 0,
+          .indent = indent,
+          .errorMsg = NULL,
+      },
+      .npyCtxtPassthru = NULL,
+      .blkCtxtPassthru = NULL,
+      .npyType = -1,
+      .npyValue = NULL,
+      .datetimeIso = 0,
+      .datetimeUnit = NPY_FR_ms,
+      .outputFormat = COLUMNS,
+      .defaultHandler = NULL,
+  };
+  JSONObjectEncoder *encoder = (JSONObjectEncoder *)&pyEncoder;
+
+  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O|OiOssOOi", kwlist, &oinput,
+                                   &oensureAscii, &idoublePrecision,
+                                   &oencodeHTMLChars, &sOrient, &sdateFormat,
+                                   &oisoDates, &odefHandler, &indent)) {
+    return NULL;
+  }
+
+  if (oensureAscii != NULL && !PyObject_IsTrue(oensureAscii)) {
+    encoder->forceASCII = 0;
+  }
+
+  if (oencodeHTMLChars != NULL && PyObject_IsTrue(oencodeHTMLChars)) {
+    encoder->encodeHTMLChars = 1;
+  }
+
+  if (idoublePrecision > JSON_DOUBLE_MAX_DECIMALS || idoublePrecision < 0) {
+    PyErr_Format(
+        PyExc_ValueError,
+        "Invalid value '%d' for option 'double_precision', max is '%u'",
+        idoublePrecision, JSON_DOUBLE_MAX_DECIMALS);
+    return NULL;
+  }
+  encoder->doublePrecision = idoublePrecision;
+
+  if (sOrient != NULL) {
+    if (strcmp(sOrient, "records") == 0) {
+      pyEncoder.outputFormat = RECORDS;
+    } else if (strcmp(sOrient, "index") == 0) {
+      pyEncoder.outputFormat = INDEX;
+    } else if (strcmp(sOrient, "split") == 0) {
+      pyEncoder.outputFormat = SPLIT;
+    } else if (strcmp(sOrient, "values") == 0) {
+      pyEncoder.outputFormat = VALUES;
+    } else if (strcmp(sOrient, "columns") != 0) {
+      PyErr_Format(PyExc_ValueError, "Invalid value '%s' for option 'orient'",
+                   sOrient);
+      return NULL;
+    }
+  }
+
+  if (sdateFormat != NULL) {
+    if (strcmp(sdateFormat, "s") == 0) {
+      pyEncoder.datetimeUnit = NPY_FR_s;
+    } else if (strcmp(sdateFormat, "ms") == 0) {
+      pyEncoder.datetimeUnit = NPY_FR_ms;
+    } else if (strcmp(sdateFormat, "us") == 0) {
+      pyEncoder.datetimeUnit = NPY_FR_us;
+    } else if (strcmp(sdateFormat, "ns") == 0) {
+      pyEncoder.datetimeUnit = NPY_FR_ns;
+    } else {
+      PyErr_Format(PyExc_ValueError,
+                   "Invalid value '%s' for option 'date_unit'", sdateFormat);
+      return NULL;
+    }
+  }
+
+  if (oisoDates != NULL && PyObject_IsTrue(oisoDates)) {
+    pyEncoder.datetimeIso = 1;
+  }
+
+  if (odefHandler != NULL && odefHandler != Py_None) {
+    if (!PyCallable_Check(odefHandler)) {
+      PyErr_SetString(PyExc_TypeError, "Default handler is not callable");
+      return NULL;
+    }
+    pyEncoder.defaultHandler = odefHandler;
+  }
+
+  encoder->indent = indent;
+
+  pyEncoder.originalOutputFormat = pyEncoder.outputFormat;
+
+  char buffer[65536];
+  char *ret = JSON_EncodeObject(oinput, encoder, buffer, sizeof(buffer));
+  if (PyErr_Occurred()) {
+    return NULL;
+  }
+
+  if (encoder->errorMsg) {
+    if (ret != buffer) {
+      encoder->free(ret);
+    }
+    PyErr_Format(PyExc_OverflowError, "%s", encoder->errorMsg);
+    return NULL;
+  }
+
+  PyObject *newobj = PyUnicode_FromString(ret);
+
+  if (ret != buffer) {
+    encoder->free(ret);
+  }
+
+  return newobj;
+}
diff --git a/pandas/pandas/_libs/src/vendored/ujson/python/ujson.c b/pandas/pandas/_libs/src/vendored/ujson/python/ujson.c
new file mode 100644
index 0000000000000..2ee084b9304f4
--- /dev/null
+++ b/pandas/pandas/_libs/src/vendored/ujson/python/ujson.c
@@ -0,0 +1,452 @@
+/*
+Copyright (c) 2011-2013, ESN Social Software AB and Jonas Tarnstrom
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+* Neither the name of the ESN Social Software AB nor the
+names of its contributors may be used to endorse or promote products
+derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL ESN SOCIAL SOFTWARE AB OR JONAS TARNSTROM BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
+GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
+THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Portions of code from MODP_ASCII - Ascii transformations (upper/lower, etc)
+https://github.com/client9/stringencoders
+Copyright (c) 2007  Nick Galbreath -- nickg [at] modp [dot] com. All rights
+reserved.
+
+Numeric decoder derived from TCL library
+https://www.opensource.apple.com/source/tcl/tcl-14/tcl/license.terms
+* Copyright (c) 1988-1993 The Regents of the University of California.
+* Copyright (c) 1994 Sun Microsystems, Inc.
+*/
+
+// Licence at LICENSES/ULTRAJSON_LICENSE
+
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
+#define PY_ARRAY_UNIQUE_SYMBOL UJSON_NUMPY
+#include "numpy/arrayobject.h"
+
+/* objToJSON */
+PyObject *objToJSON(PyObject *self, PyObject *args, PyObject *kwargs);
+void *initObjToJSON(void);
+
+/* JSONToObj */
+PyObject *JSONToObj(PyObject *self, PyObject *args, PyObject *kwargs);
+
+#define ENCODER_HELP_TEXT                                                      \
+  "Use ensure_ascii=false to output UTF-8. Pass in double_precision to "       \
+  "alter the maximum digit precision of doubles. Set "                         \
+  "encode_html_chars=True to encode < > & as unicode escape sequences."
+
+static PyMethodDef ujsonMethods[] = {
+    {"ujson_dumps", (PyCFunction)(void (*)(void))objToJSON,
+     METH_VARARGS | METH_KEYWORDS,
+     "Converts arbitrary object recursively into JSON. " ENCODER_HELP_TEXT},
+    {"ujson_loads", (PyCFunction)(void (*)(void))JSONToObj,
+     METH_VARARGS | METH_KEYWORDS,
+     "Converts JSON as string to dict object structure. Use precise_float=True "
+     "to use high precision float decoder."},
+    {NULL, NULL, 0, NULL} /* Sentinel */
+};
+
+typedef struct {
+  PyObject *type_decimal;
+  PyObject *type_dataframe;
+  PyObject *type_series;
+  PyObject *type_index;
+  PyObject *type_nat;
+  PyObject *type_na;
+} modulestate;
+
+#define modulestate(o) ((modulestate *)PyModule_GetState(o))
+
+static int module_traverse(PyObject *m, visitproc visit, void *arg);
+static int module_clear(PyObject *m);
+static void module_free(void *module);
+
+static struct PyModuleDef moduledef = {.m_base = PyModuleDef_HEAD_INIT,
+                                       .m_name = "pandas._libs.json",
+                                       .m_methods = ujsonMethods,
+                                       .m_size = sizeof(modulestate),
+                                       .m_traverse = module_traverse,
+                                       .m_clear = module_clear,
+                                       .m_free = module_free};
+
+#ifndef PYPY_VERSION
+/* Used in objToJSON.c */
+int object_is_decimal_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_decimal = state->type_decimal;
+  if (type_decimal == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_decimal);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_dataframe_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_dataframe = state->type_dataframe;
+  if (type_dataframe == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_dataframe);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_series_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_series = state->type_series;
+  if (type_series == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_series);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_index_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_index = state->type_index;
+  if (type_index == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_index);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_nat_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_nat = state->type_nat;
+  if (type_nat == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_nat);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_na_type(PyObject *obj) {
+  PyObject *module = PyState_FindModule(&moduledef);
+  if (module == NULL)
+    return 0;
+  modulestate *state = modulestate(module);
+  if (state == NULL)
+    return 0;
+  PyObject *type_na = state->type_na;
+  if (type_na == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_na);
+  if (result == -1) {
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+#else
+/* Used in objToJSON.c */
+int object_is_decimal_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("decimal");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_decimal = PyObject_GetAttrString(module, "Decimal");
+  if (type_decimal == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_decimal);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_decimal);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_dataframe_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("pandas");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_dataframe = PyObject_GetAttrString(module, "DataFrame");
+  if (type_dataframe == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_dataframe);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_dataframe);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_series_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("pandas");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_series = PyObject_GetAttrString(module, "Series");
+  if (type_series == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_series);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_series);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_index_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("pandas");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_index = PyObject_GetAttrString(module, "Index");
+  if (type_index == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_index);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_index);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_nat_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("pandas._libs.tslibs.nattype");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_nat = PyObject_GetAttrString(module, "NaTType");
+  if (type_nat == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_nat);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_nat);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+int object_is_na_type(PyObject *obj) {
+  PyObject *module = PyImport_ImportModule("pandas._libs.missing");
+  if (module == NULL) {
+    PyErr_Clear();
+    return 0;
+  }
+  PyObject *type_na = PyObject_GetAttrString(module, "NAType");
+  if (type_na == NULL) {
+    Py_DECREF(module);
+    PyErr_Clear();
+    return 0;
+  }
+  int result = PyObject_IsInstance(obj, type_na);
+  if (result == -1) {
+    Py_DECREF(module);
+    Py_DECREF(type_na);
+    PyErr_Clear();
+    return 0;
+  }
+  return result;
+}
+
+#endif
+
+static int module_traverse(PyObject *m, visitproc visit, void *arg) {
+  Py_VISIT(modulestate(m)->type_decimal);
+  Py_VISIT(modulestate(m)->type_dataframe);
+  Py_VISIT(modulestate(m)->type_series);
+  Py_VISIT(modulestate(m)->type_index);
+  Py_VISIT(modulestate(m)->type_nat);
+  Py_VISIT(modulestate(m)->type_na);
+  return 0;
+}
+
+static int module_clear(PyObject *m) {
+  Py_CLEAR(modulestate(m)->type_decimal);
+  Py_CLEAR(modulestate(m)->type_dataframe);
+  Py_CLEAR(modulestate(m)->type_series);
+  Py_CLEAR(modulestate(m)->type_index);
+  Py_CLEAR(modulestate(m)->type_nat);
+  Py_CLEAR(modulestate(m)->type_na);
+  return 0;
+}
+
+static void module_free(void *module) { module_clear((PyObject *)module); }
+
+PyMODINIT_FUNC PyInit_json(void) {
+  import_array() PyObject *module;
+
+#ifndef PYPY_VERSION
+  // This function is not supported in PyPy.
+  if ((module = PyState_FindModule(&moduledef)) != NULL) {
+    Py_INCREF(module);
+    return module;
+  }
+#endif
+
+  module = PyModule_Create(&moduledef);
+  if (module == NULL) {
+    return NULL;
+  }
+
+#ifdef Py_GIL_DISABLED
+  PyUnstable_Module_SetGIL(module, Py_MOD_GIL_NOT_USED);
+#endif
+
+#ifndef PYPY_VERSION
+  PyObject *mod_decimal = PyImport_ImportModule("decimal");
+  if (mod_decimal) {
+    PyObject *type_decimal = PyObject_GetAttrString(mod_decimal, "Decimal");
+    assert(type_decimal != NULL);
+    modulestate(module)->type_decimal = type_decimal;
+    Py_DECREF(mod_decimal);
+  }
+
+  PyObject *mod_pandas = PyImport_ImportModule("pandas");
+  if (mod_pandas) {
+    PyObject *type_dataframe = PyObject_GetAttrString(mod_pandas, "DataFrame");
+    assert(type_dataframe != NULL);
+    modulestate(module)->type_dataframe = type_dataframe;
+
+    PyObject *type_series = PyObject_GetAttrString(mod_pandas, "Series");
+    assert(type_series != NULL);
+    modulestate(module)->type_series = type_series;
+
+    PyObject *type_index = PyObject_GetAttrString(mod_pandas, "Index");
+    assert(type_index != NULL);
+    modulestate(module)->type_index = type_index;
+
+    Py_DECREF(mod_pandas);
+  }
+
+  PyObject *mod_nattype = PyImport_ImportModule("pandas._libs.tslibs.nattype");
+  if (mod_nattype) {
+    PyObject *type_nat = PyObject_GetAttrString(mod_nattype, "NaTType");
+    assert(type_nat != NULL);
+    modulestate(module)->type_nat = type_nat;
+
+    Py_DECREF(mod_nattype);
+  }
+
+  PyObject *mod_natype = PyImport_ImportModule("pandas._libs.missing");
+  if (mod_natype) {
+    PyObject *type_na = PyObject_GetAttrString(mod_natype, "NAType");
+    assert(type_na != NULL);
+    modulestate(module)->type_na = type_na;
+
+    Py_DECREF(mod_natype);
+  } else {
+    PyErr_Clear();
+  }
+#endif
+
+  /* Not vendored for now
+  JSONDecodeError = PyErr_NewException("ujson.JSONDecodeError",
+  PyExc_ValueError, NULL); Py_XINCREF(JSONDecodeError); if
+  (PyModule_AddObject(module, "JSONDecodeError", JSONDecodeError) < 0)
+  {
+    Py_XDECREF(JSONDecodeError);
+    Py_CLEAR(JSONDecodeError);
+    Py_DECREF(module);
+    return NULL;
+  }
+  */
+
+  return module;
+}
diff --git a/pandas/pandas/_libs/testing.pyi b/pandas/pandas/_libs/testing.pyi
new file mode 100644
index 0000000000000..ab87e58eba9b9
--- /dev/null
+++ b/pandas/pandas/_libs/testing.pyi
@@ -0,0 +1,14 @@
+from typing import Mapping
+
+def assert_dict_equal(a: Mapping, b: Mapping, compare_keys: bool = ...) -> bool: ...
+def assert_almost_equal(
+    a,
+    b,
+    rtol: float = ...,
+    atol: float = ...,
+    check_dtype: bool = ...,
+    obj=...,
+    lobj=...,
+    robj=...,
+    index_values=...,
+) -> bool: ...
diff --git a/pandas/pandas/_libs/testing.pyx b/pandas/pandas/_libs/testing.pyx
new file mode 100644
index 0000000000000..cfd31fa610e69
--- /dev/null
+++ b/pandas/pandas/_libs/testing.pyx
@@ -0,0 +1,212 @@
+import cmath
+import math
+
+import numpy as np
+
+from numpy cimport import_array
+
+import_array()
+
+from pandas._libs.missing cimport (
+    checknull,
+    is_matching_na,
+)
+from pandas._libs.util cimport (
+    is_array,
+    is_complex_object,
+    is_real_number_object,
+)
+
+
+from pandas.core.dtypes.missing import array_equivalent
+
+
+cdef bint isiterable(obj):
+    return hasattr(obj, "__iter__")
+
+
+cdef bint has_length(obj):
+    return hasattr(obj, "__len__")
+
+
+cdef bint is_dictlike(obj):
+    return hasattr(obj, "keys") and hasattr(obj, "__getitem__")
+
+
+cpdef assert_dict_equal(a, b, bint compare_keys=True):
+    assert is_dictlike(a) and is_dictlike(b), (
+        "Cannot compare dict objects, one or both is not dict-like"
+    )
+
+    a_keys = frozenset(a.keys())
+    b_keys = frozenset(b.keys())
+
+    if compare_keys:
+        assert a_keys == b_keys
+
+    for k in a_keys:
+        assert_almost_equal(a[k], b[k])
+
+    return True
+
+
+cpdef assert_almost_equal(a, b,
+                          rtol=1.e-5, atol=1.e-8,
+                          bint check_dtype=True,
+                          obj=None, lobj=None, robj=None, index_values=None):
+    """
+    Check that left and right objects are almost equal.
+
+    Parameters
+    ----------
+    a : object
+    b : object
+    rtol : float, default 1e-5
+        Relative tolerance.
+    atol : float, default 1e-8
+        Absolute tolerance.
+    check_dtype: bool, default True
+        check dtype if both a and b are np.ndarray.
+    obj : str, default None
+        Specify object name being compared, internally used to show
+        appropriate assertion message.
+    lobj : str, default None
+        Specify left object name being compared, internally used to show
+        appropriate assertion message.
+    robj : str, default None
+        Specify right object name being compared, internally used to show
+        appropriate assertion message.
+    index_values : Index | ndarray, default None
+        Specify shared index values of objects being compared, internally used
+        to show appropriate assertion message.
+
+    """
+    cdef:
+        double diff = 0.0
+        Py_ssize_t i, na, nb
+        double fa, fb
+        bint is_unequal = False, a_is_ndarray, b_is_ndarray
+        str first_diff = ""
+
+    if lobj is None:
+        lobj = a
+    if robj is None:
+        robj = b
+
+    if isinstance(a, set) or isinstance(b, set):
+        assert a == b, f"{a} != {b}"
+        return True
+
+    if isinstance(a, dict) or isinstance(b, dict):
+        return assert_dict_equal(a, b)
+
+    if isinstance(a, str) or isinstance(b, str):
+        assert a == b, f"{a} != {b}"
+        return True
+
+    a_is_ndarray = is_array(a)
+    b_is_ndarray = is_array(b)
+
+    if obj is None:
+        if a_is_ndarray or b_is_ndarray:
+            obj = "numpy array"
+        else:
+            obj = "Iterable"
+
+    if isiterable(a):
+
+        if not isiterable(b):
+            from pandas._testing import assert_class_equal
+
+            # classes can't be the same, to raise error
+            assert_class_equal(a, b, obj=obj)
+
+        assert has_length(a) and has_length(b), (
+            f"Can't compare objects without length, one or both is invalid: ({a}, {b})"
+        )
+
+        if a_is_ndarray and b_is_ndarray:
+            na, nb = a.size, b.size
+            if a.shape != b.shape:
+                from pandas._testing import raise_assert_detail
+                raise_assert_detail(
+                    obj, f"{obj} shapes are different", a.shape, b.shape)
+
+            if check_dtype and a.dtype != b.dtype:
+                from pandas._testing import assert_attr_equal
+                assert_attr_equal("dtype", a, b, obj=obj)
+
+            if array_equivalent(a, b, strict_nan=True):
+                return True
+
+        else:
+            na, nb = len(a), len(b)
+
+        if na != nb:
+            from pandas._testing import raise_assert_detail
+
+            # if we have a small diff set, print it
+            if abs(na - nb) < 10:
+                r = list(set(a) ^ set(b))
+            else:
+                r = None
+
+            raise_assert_detail(obj, f"{obj} length are different", na, nb, r)
+
+        for i in range(len(a)):
+            try:
+                assert_almost_equal(a[i], b[i], rtol=rtol, atol=atol)
+            except AssertionError:
+                is_unequal = True
+                diff += 1
+                if not first_diff:
+                    first_diff = (
+                        f"At positional index {i}, first diff: {a[i]} != {b[i]}"
+                    )
+
+        if is_unequal:
+            from pandas._testing import raise_assert_detail
+            msg = (f"{obj} values are different "
+                   f"({np.round(diff * 100.0 / na, 5)} %)")
+            raise_assert_detail(
+                obj, msg, lobj, robj, first_diff=first_diff, index_values=index_values
+            )
+
+        return True
+
+    elif isiterable(b):
+        from pandas._testing import assert_class_equal
+
+        # classes can't be the same, to raise error
+        assert_class_equal(a, b, obj=obj)
+
+    if checknull(a):
+        # nan / None comparison
+        if is_matching_na(a, b, nan_matches_none=False):
+            return True
+        elif checknull(b):
+            # GH#18463
+            raise AssertionError(f"Mismatched null-like values {a} != {b}")
+        raise AssertionError(f"{a} != {b}")
+    elif checknull(b):
+        raise AssertionError(f"{a} != {b}")
+
+    if a == b:
+        # object comparison
+        return True
+
+    if is_real_number_object(a) and is_real_number_object(b):
+        fa, fb = a, b
+
+        if not math.isclose(fa, fb, rel_tol=rtol, abs_tol=atol):
+            assert False, (f"expected {fb:.5f} but got {fa:.5f}, "
+                           f"with rtol={rtol}, atol={atol}")
+        return True
+
+    if is_complex_object(a) and is_complex_object(b):
+        if not cmath.isclose(a, b, rel_tol=rtol, abs_tol=atol):
+            assert False, (f"expected {b:.5f} but got {a:.5f}, "
+                           f"with rtol={rtol}, atol={atol}")
+        return True
+
+    raise AssertionError(f"{a} != {b}")
diff --git a/pandas/pandas/_libs/tslib.pyi b/pandas/pandas/_libs/tslib.pyi
new file mode 100644
index 0000000000000..7e3372a80db9d
--- /dev/null
+++ b/pandas/pandas/_libs/tslib.pyi
@@ -0,0 +1,33 @@
+from datetime import tzinfo
+
+import numpy as np
+
+from pandas._typing import npt
+
+def format_array_from_datetime(
+    values: npt.NDArray[np.int64],
+    tz: tzinfo | None = ...,
+    format: str | None = ...,
+    na_rep: str | float = ...,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.object_]: ...
+def first_non_null(values: np.ndarray) -> int: ...
+def array_to_datetime(
+    values: npt.NDArray[np.object_],
+    errors: str = ...,
+    dayfirst: bool = ...,
+    yearfirst: bool = ...,
+    utc: bool = ...,
+    creso: int = ...,
+    unit_for_numerics: str | None = ...,
+) -> tuple[np.ndarray, tzinfo | None]: ...
+
+# returned ndarray may be object dtype or datetime64[ns]
+
+def array_to_datetime_with_tz(
+    values: npt.NDArray[np.object_],
+    tz: tzinfo,
+    dayfirst: bool,
+    yearfirst: bool,
+    creso: int,
+) -> npt.NDArray[np.int64]: ...
diff --git a/pandas/pandas/_libs/tslib.pyx b/pandas/pandas/_libs/tslib.pyx
new file mode 100644
index 0000000000000..3c5854602df53
--- /dev/null
+++ b/pandas/pandas/_libs/tslib.pyx
@@ -0,0 +1,555 @@
+cimport cython
+
+from datetime import timezone
+
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    import_datetime,
+    timedelta,
+    tzinfo,
+)
+from cpython.object cimport PyObject
+
+# import datetime C API
+import_datetime()
+
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+import numpy as np
+
+cnp.import_array()
+
+from pandas._libs.tslibs.dtypes cimport (
+    get_supported_reso,
+    npy_unit_to_abbrev,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    get_datetime64_unit,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pandas_datetime_to_datetimestruct,
+    pydate_to_dt64,
+    string_to_dts,
+)
+
+import_pandas_datetime()
+
+
+from pandas._libs.tslibs.strptime cimport (
+    DatetimeParseState,
+    parse_today_now,
+)
+from pandas._libs.util cimport (
+    is_float_object,
+    is_integer_object,
+)
+
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+
+from pandas._libs.tslibs.conversion cimport (
+    _TSObject,
+    cast_from_unit,
+    convert_str_to_tsobject,
+    convert_to_tsobject,
+    get_datetime64_nanos,
+    parse_pydatetime,
+)
+from pandas._libs.tslibs.dtypes cimport (
+    get_supported_reso,
+    npy_unit_to_abbrev,
+)
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_nat_strings as nat_strings,
+)
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+
+from pandas._libs.tslibs import (
+    Resolution,
+    get_resolution,
+)
+from pandas._libs.tslibs.timestamps import Timestamp
+
+# Note: this is the only non-tslibs intra-pandas dependency here
+
+from pandas._libs.missing cimport checknull_with_nat_and_na
+from pandas._libs.tslibs.tzconversion cimport tz_localize_to_utc_single
+
+
+def _test_parse_iso8601(ts: str):
+    """
+    TESTING ONLY: Parse string into Timestamp using iso8601 parser. Used
+    only for testing, actual construction uses `convert_str_to_tsobject`
+    """
+    cdef:
+        _TSObject obj
+        int out_local = 0, out_tzoffset = 0
+        NPY_DATETIMEUNIT out_bestunit
+
+    obj = _TSObject()
+
+    string_to_dts(ts, &obj.dts, &out_bestunit, &out_local, &out_tzoffset, True)
+    try:
+        obj.value = npy_datetimestruct_to_datetime(NPY_FR_ns, &obj.dts)
+    except OverflowError as err:
+        raise OutOfBoundsDatetime(f"Out of bounds nanosecond timestamp: {ts}") from err
+    if out_local == 1:
+        obj.tzinfo = timezone(timedelta(minutes=out_tzoffset))
+        obj.value = tz_localize_to_utc_single(obj.value, obj.tzinfo)
+        return Timestamp(obj.value, tz=obj.tzinfo)
+    else:
+        return Timestamp(obj.value)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def format_array_from_datetime(
+    ndarray values,
+    tzinfo tz=None,
+    str format=None,
+    na_rep: str | float = "NaT",
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+) -> np.ndarray:
+    """
+    return a np object array of the string formatted values
+
+    Parameters
+    ----------
+    values : ndarray[int64_t], arbitrary ndim
+    tz : tzinfo or None, default None
+    format : str or None, default None
+          a strftime capable string
+    na_rep : optional, default is None
+          a nat format
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    np.ndarray[object]
+    """
+    cdef:
+        int64_t val, ns, N = values.size
+        bint show_ms = False, show_us = False, show_ns = False
+        bint basic_format = False, basic_format_day = False
+        _Timestamp ts
+        object res
+        npy_datetimestruct dts
+
+        # Note that `result` (and thus `result_flat`) is C-order and
+        #  `it` iterates C-order as well, so the iteration matches
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        ndarray result = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
+        object[::1] res_flat = result.ravel()     # should NOT be a copy
+        cnp.flatiter it = cnp.PyArray_IterNew(values)
+
+    if tz is None:
+        # if we don't have a format nor tz, then choose
+        # a format based on precision
+        basic_format = format is None
+        if basic_format:
+            reso_obj = get_resolution(values, tz=tz, reso=reso)
+            show_ns = reso_obj == Resolution.RESO_NS
+            show_us = reso_obj == Resolution.RESO_US
+            show_ms = reso_obj == Resolution.RESO_MS
+
+        elif format == "%Y-%m-%d %H:%M:%S":
+            # Same format as default, but with hardcoded precision (s)
+            basic_format = True
+            show_ns = show_us = show_ms = False
+
+        elif format == "%Y-%m-%d %H:%M:%S.%f":
+            # Same format as default, but with hardcoded precision (us)
+            basic_format = show_us = True
+            show_ns = show_ms = False
+
+        elif format == "%Y-%m-%d":
+            # Default format for dates
+            basic_format_day = True
+
+    assert not (basic_format_day and basic_format)
+
+    for i in range(N):
+        # Analogous to: utc_val = values[i]
+        val = (<int64_t*>cnp.PyArray_ITER_DATA(it))[0]
+
+        if val == NPY_NAT:
+            res = na_rep
+        elif basic_format_day:
+
+            pandas_datetime_to_datetimestruct(val, reso, &dts)
+            res = f"{dts.year}-{dts.month:02d}-{dts.day:02d}"
+
+        elif basic_format:
+
+            pandas_datetime_to_datetimestruct(val, reso, &dts)
+            res = (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                   f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}")
+
+            if show_ns:
+                ns = dts.ps // 1000
+                res += f".{ns + dts.us * 1000:09d}"
+            elif show_us:
+                res += f".{dts.us:06d}"
+            elif show_ms:
+                res += f".{dts.us // 1000:03d}"
+
+        else:
+
+            ts = Timestamp._from_value_and_reso(val, reso=reso, tz=tz)
+            if format is None:
+                # Use datetime.str, that returns ts.isoformat(sep=' ')
+                res = str(ts)
+            else:
+
+                # invalid format string
+                # requires dates > 1900
+                try:
+                    # Note: dispatches to pydatetime
+                    res = ts.strftime(format)
+                except ValueError:
+                    # Use datetime.str, that returns ts.isoformat(sep=' ')
+                    res = str(ts)
+
+        # Note: we can index result directly instead of using PyArray_MultiIter_DATA
+        #  like we do for the other functions because result is known C-contiguous
+        #  and is the first argument to PyArray_MultiIterNew2.  The usual pattern
+        #  does not seem to work with object dtype.
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        res_flat[i] = res
+
+        cnp.PyArray_ITER_NEXT(it)
+
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def first_non_null(values: ndarray) -> int:
+    """Find position of first non-null value, return -1 if there isn't one."""
+    cdef:
+        Py_ssize_t n = len(values)
+        Py_ssize_t i
+    for i in range(n):
+        val = values[i]
+        if checknull_with_nat_and_na(val):
+            continue
+        if (
+            isinstance(val, str)
+            and
+            (len(val) == 0 or val in nat_strings or val in ("now", "today"))
+        ):
+            continue
+        return i
+    else:
+        return -1
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef array_to_datetime(
+    ndarray values,  # object dtype, arbitrary ndim
+    str errors="raise",
+    bint dayfirst=False,
+    bint yearfirst=False,
+    bint utc=False,
+    NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_GENERIC,
+    str unit_for_numerics=None,
+):
+    """
+    Converts a 1D array of date-like values to a numpy array of either:
+        1) datetime64[ns] data
+        2) datetime.datetime objects, if OutOfBoundsDatetime or TypeError
+           is encountered
+
+    Also returns a fixed-offset tzinfo object if an array of strings with the same
+    timezone offset is passed and utc=True is not passed. Otherwise, None
+    is returned
+
+    Handles datetime.date, datetime.datetime, np.datetime64 objects, numeric,
+    strings
+
+    Parameters
+    ----------
+    values : ndarray of object
+        date-like objects to convert
+    errors : str, default 'raise'
+        error behavior when parsing
+    dayfirst : bool, default False
+        dayfirst parsing behavior when encountering datetime strings
+    yearfirst : bool, default False
+        yearfirst parsing behavior when encountering datetime strings
+    utc : bool, default False
+        indicator whether the dates should be UTC
+    creso : NPY_DATETIMEUNIT, default NPY_FR_GENERIC
+        If NPY_FR_GENERIC, conduct inference.
+    unit_for_numerics : str, default "ns"
+
+    Returns
+    -------
+    np.ndarray
+        May be datetime64[creso_unit] or object dtype
+    tzinfo or None
+    """
+    cdef:
+        Py_ssize_t i, n = values.size
+        object val
+        ndarray[int64_t] iresult
+        npy_datetimestruct dts
+        bint utc_convert = bool(utc)
+        bint is_raise = errors == "raise"
+        bint is_coerce = errors == "coerce"
+        _TSObject tsobj
+        tzinfo tz, tz_out = None
+        cnp.flatiter it = cnp.PyArray_IterNew(values)
+        NPY_DATETIMEUNIT item_reso
+        bint infer_reso = creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+        DatetimeParseState state = DatetimeParseState(creso)
+        str abbrev
+
+    # specify error conditions
+    assert is_raise or is_coerce
+
+    if infer_reso:
+        abbrev = "ns"
+    else:
+        abbrev = npy_unit_to_abbrev(creso)
+
+    if unit_for_numerics is not None:
+        # either creso or unit_for_numerics should be passed, not both
+        assert creso == NPY_FR_ns
+    else:
+        unit_for_numerics = abbrev
+
+    result = np.empty((<object>values).shape, dtype=f"M8[{abbrev}]")
+    iresult = result.view("i8").ravel()
+
+    for i in range(n):
+        # Analogous to `val = values[i]`
+        val = cnp.PyArray_GETITEM(values, cnp.PyArray_ITER_DATA(it))
+        cnp.PyArray_ITER_NEXT(it)
+
+        try:
+            if checknull_with_nat_and_na(val):
+                iresult[i] = NPY_NAT
+
+            elif PyDateTime_Check(val):
+                if isinstance(val, _Timestamp):
+                    item_reso = val._creso
+                else:
+                    item_reso = NPY_DATETIMEUNIT.NPY_FR_us
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                tz_out = state.process_datetime(val, tz_out, utc_convert)
+                iresult[i] = parse_pydatetime(val, &dts, creso=creso)
+
+            elif PyDate_Check(val):
+                item_reso = NPY_DATETIMEUNIT.NPY_FR_s
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                iresult[i] = pydate_to_dt64(val, &dts, reso=creso)
+                state.found_other = True
+
+            elif cnp.is_datetime64_object(val):
+                item_reso = get_supported_reso(get_datetime64_unit(val))
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                iresult[i] = get_datetime64_nanos(val, creso)
+                state.found_other = True
+
+            elif is_integer_object(val) or is_float_object(val):
+                # these must be ns unit by-definition
+
+                if val != val or val == NPY_NAT:
+                    iresult[i] = NPY_NAT
+                else:
+                    item_reso = NPY_FR_ns
+                    state.update_creso(item_reso)
+                    if infer_reso:
+                        creso = state.creso
+
+                    # we now need to parse this as if unit=abbrev
+                    iresult[i] = cast_from_unit(val, unit_for_numerics, out_reso=creso)
+
+                    state.found_other = True
+
+            elif isinstance(val, str):
+                # string
+                if type(val) is not str:
+                    # GH#32264 np.str_ object
+                    val = str(val)
+
+                if parse_today_now(val, &iresult[i], utc, creso, infer_reso=infer_reso):
+                    # We can't _quite_ dispatch this to convert_str_to_tsobject
+                    #  bc there isn't a nice way to pass "utc"
+                    item_reso = NPY_DATETIMEUNIT.NPY_FR_us
+                    state.update_creso(item_reso)
+                    if infer_reso:
+                        creso = state.creso
+                    continue
+
+                tsobj = convert_str_to_tsobject(
+                    val, None, dayfirst=dayfirst, yearfirst=yearfirst
+                )
+
+                if tsobj.value == NPY_NAT:
+                    # e.g. "NaT" string or empty string, we do not consider
+                    #  this as either tzaware or tznaive. See
+                    #  test_to_datetime_with_empty_str_utc_false_format_mixed
+                    # We also do not update resolution inference based on this,
+                    #  see test_infer_with_nat_int_float_str
+                    iresult[i] = tsobj.value
+                    continue
+
+                item_reso = tsobj.creso
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+
+                tsobj.ensure_reso(creso, val)
+                iresult[i] = tsobj.value
+
+                tz = tsobj.tzinfo
+                if tz is not None:
+                    # dateutil timezone objects cannot be hashed, so
+                    # store the UTC offsets in seconds instead
+                    nsecs = tz.utcoffset(None).total_seconds()
+                    state.out_tzoffset_vals.add(nsecs)
+                    state.found_aware_str = True
+                else:
+                    # Add a marker for naive string, to track if we are
+                    # parsing mixed naive and aware strings
+                    state.out_tzoffset_vals.add("naive")
+                    state.found_naive_str = True
+
+            else:
+                raise TypeError(f"{type(val)} is not convertible to datetime")
+
+        except (TypeError, OverflowError, ValueError) as ex:
+            ex.args = (f"{ex}",)
+            if is_coerce:
+                iresult[i] = NPY_NAT
+                continue
+            elif is_raise:
+                raise
+            return values, None
+
+    tz_out = state.check_for_mixed_inputs(tz_out, utc)
+
+    if infer_reso:
+        if state.creso_ever_changed:
+            # We encountered mismatched resolutions, need to re-parse with
+            #  the correct one.
+            return array_to_datetime(
+                values,
+                errors=errors,
+                yearfirst=yearfirst,
+                dayfirst=dayfirst,
+                utc=utc,
+                creso=state.creso,
+            )
+        elif state.creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+            # i.e. we never encountered anything non-NaT, default to "s". This
+            # ensures that insert and concat-like operations with NaT
+            # do not upcast units
+            result = iresult.view("M8[s]").reshape(result.shape)
+        else:
+            # Otherwise we can use the single reso that we encountered and avoid
+            #  a second pass.
+            abbrev = npy_unit_to_abbrev(state.creso)
+            result = iresult.view(f"M8[{abbrev}]").reshape(result.shape)
+    return result, tz_out
+
+
+def array_to_datetime_with_tz(
+    ndarray values, tzinfo tz, bint dayfirst, bint yearfirst, NPY_DATETIMEUNIT creso
+):
+    """
+    Vectorized analogue to pd.Timestamp(value, tz=tz)
+
+    values has object-dtype, unrestricted ndim.
+
+    Major differences between this and array_to_datetime with utc=True
+        - np.datetime64 objects are treated as _wall_ times.
+        - tznaive datetimes are treated as _wall_ times.
+    """
+    cdef:
+        ndarray result = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_INT64, 0)
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(result, values)
+        Py_ssize_t i, n = values.size
+        object item
+        int64_t ival
+        _TSObject tsobj
+        bint infer_reso = creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+        DatetimeParseState state = DatetimeParseState(creso)
+        str abbrev
+
+    if infer_reso:
+        # We treat ints/floats as nanoseconds
+        abbrev = "ns"
+    else:
+        abbrev = npy_unit_to_abbrev(creso)
+
+    for i in range(n):
+        # Analogous to `item = values[i]`
+        item = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if checknull_with_nat_and_na(item):
+            # this catches pd.NA which would raise in the Timestamp constructor
+            ival = NPY_NAT
+
+        else:
+            tsobj = convert_to_tsobject(
+                item,
+                tz=tz,
+                unit=abbrev,
+                dayfirst=dayfirst,
+                yearfirst=yearfirst,
+                nanos=0,
+            )
+            if tsobj.value != NPY_NAT:
+                state.update_creso(tsobj.creso)
+                if infer_reso:
+                    creso = state.creso
+                tsobj.ensure_reso(creso, item, round_ok=True)
+            ival = tsobj.value
+
+        # Analogous to: result[i] = ival
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = ival
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    if infer_reso:
+        if state.creso_ever_changed:
+            # We encountered mismatched resolutions, need to re-parse with
+            #  the correct one.
+            return array_to_datetime_with_tz(
+                values, tz=tz, dayfirst=dayfirst, yearfirst=yearfirst, creso=creso
+            )
+        elif creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+            # i.e. we never encountered anything non-NaT, default to "s". This
+            # ensures that insert and concat-like operations with NaT
+            # do not upcast units
+            result = result.view("M8[s]")
+        else:
+            # Otherwise we can use the single reso that we encountered and avoid
+            #  a second pass.
+            abbrev = npy_unit_to_abbrev(creso)
+            result = result.view(f"M8[{abbrev}]")
+    else:
+        abbrev = npy_unit_to_abbrev(creso)
+        result = result.view(f"M8[{abbrev}]")
+    return result
diff --git a/pandas/pandas/_libs/tslibs/__init__.py b/pandas/pandas/_libs/tslibs/__init__.py
new file mode 100644
index 0000000000000..f433a3acf356f
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/__init__.py
@@ -0,0 +1,87 @@
+__all__ = [
+    "BaseOffset",
+    "IncompatibleFrequency",
+    "NaT",
+    "NaTType",
+    "OutOfBoundsDatetime",
+    "OutOfBoundsTimedelta",
+    "Period",
+    "Resolution",
+    "Tick",
+    "Timedelta",
+    "Timestamp",
+    "add_overflowsafe",
+    "astype_overflowsafe",
+    "delta_to_nanoseconds",
+    "dt64arr_to_periodarr",
+    "dtypes",
+    "get_resolution",
+    "get_supported_dtype",
+    "get_unit_from_dtype",
+    "guess_datetime_format",
+    "iNaT",
+    "ints_to_pydatetime",
+    "ints_to_pytimedelta",
+    "is_date_array_normalized",
+    "is_supported_dtype",
+    "is_unitless",
+    "localize_pydatetime",
+    "nat_strings",
+    "normalize_i8_timestamps",
+    "periods_per_day",
+    "periods_per_second",
+    "to_offset",
+    "tz_compare",
+    "tz_convert_from_utc",
+    "tz_convert_from_utc_single",
+]
+
+from pandas._libs.tslibs import dtypes
+from pandas._libs.tslibs.conversion import localize_pydatetime
+from pandas._libs.tslibs.dtypes import (
+    Resolution,
+    periods_per_day,
+    periods_per_second,
+)
+from pandas._libs.tslibs.nattype import (
+    NaT,
+    NaTType,
+    iNaT,
+    nat_strings,
+)
+from pandas._libs.tslibs.np_datetime import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+    add_overflowsafe,
+    astype_overflowsafe,
+    get_supported_dtype,
+    is_supported_dtype,
+    is_unitless,
+    py_get_unit_from_dtype as get_unit_from_dtype,
+)
+from pandas._libs.tslibs.offsets import (
+    BaseOffset,
+    Tick,
+    to_offset,
+)
+from pandas._libs.tslibs.parsing import guess_datetime_format
+from pandas._libs.tslibs.period import (
+    IncompatibleFrequency,
+    Period,
+)
+from pandas._libs.tslibs.timedeltas import (
+    Timedelta,
+    delta_to_nanoseconds,
+    ints_to_pytimedelta,
+)
+from pandas._libs.tslibs.timestamps import Timestamp
+from pandas._libs.tslibs.timezones import tz_compare
+from pandas._libs.tslibs.tzconversion import tz_convert_from_utc_single
+from pandas._libs.tslibs.vectorized import (
+    dt64arr_to_periodarr,
+    get_resolution,
+    ints_to_pydatetime,
+    is_date_array_normalized,
+    normalize_i8_timestamps,
+    tz_convert_from_utc,
+)
diff --git a/pandas/pandas/_libs/tslibs/base.pxd b/pandas/pandas/_libs/tslibs/base.pxd
new file mode 100644
index 0000000000000..3bffff7aca43e
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/base.pxd
@@ -0,0 +1,5 @@
+from cpython.datetime cimport datetime
+
+
+cdef class ABCTimestamp(datetime):
+    pass
diff --git a/pandas/pandas/_libs/tslibs/base.pyx b/pandas/pandas/_libs/tslibs/base.pyx
new file mode 100644
index 0000000000000..1677a8b0be1ec
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/base.pyx
@@ -0,0 +1,12 @@
+"""
+We define base classes that will be inherited by Timestamp, Timedelta, etc
+in order to allow for fast isinstance checks without circular dependency issues.
+
+This is analogous to core.dtypes.generic.
+"""
+
+from cpython.datetime cimport datetime
+
+
+cdef class ABCTimestamp(datetime):
+    pass
diff --git a/pandas/pandas/_libs/tslibs/ccalendar.pxd b/pandas/pandas/_libs/tslibs/ccalendar.pxd
new file mode 100644
index 0000000000000..3c0f7ea824396
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/ccalendar.pxd
@@ -0,0 +1,20 @@
+from cython cimport Py_ssize_t
+from numpy cimport (
+    int32_t,
+    int64_t,
+)
+
+ctypedef (int32_t, int32_t, int32_t) iso_calendar_t
+
+cdef int dayofweek(int y, int m, int d) noexcept nogil
+cdef bint is_leapyear(int64_t year) noexcept nogil
+cpdef int32_t get_days_in_month(int year, Py_ssize_t month) noexcept nogil
+cpdef int32_t get_week_of_year(int year, int month, int day) noexcept nogil
+cpdef iso_calendar_t get_iso_calendar(int year, int month, int day) noexcept nogil
+cpdef int32_t get_day_of_year(int year, int month, int day) noexcept nogil
+cpdef int get_lastbday(int year, int month) noexcept nogil
+cpdef int get_firstbday(int year, int month) noexcept nogil
+
+cdef dict c_MONTH_NUMBERS, MONTH_TO_CAL_NUM
+
+cdef int32_t* month_offset
diff --git a/pandas/pandas/_libs/tslibs/ccalendar.pyi b/pandas/pandas/_libs/tslibs/ccalendar.pyi
new file mode 100644
index 0000000000000..993f18a61d74a
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/ccalendar.pyi
@@ -0,0 +1,12 @@
+DAYS: list[str]
+MONTH_ALIASES: dict[int, str]
+MONTH_NUMBERS: dict[str, int]
+MONTHS: list[str]
+int_to_weekday: dict[int, str]
+
+def get_firstbday(year: int, month: int) -> int: ...
+def get_lastbday(year: int, month: int) -> int: ...
+def get_day_of_year(year: int, month: int, day: int) -> int: ...
+def get_iso_calendar(year: int, month: int, day: int) -> tuple[int, int, int]: ...
+def get_week_of_year(year: int, month: int, day: int) -> int: ...
+def get_days_in_month(year: int, month: int) -> int: ...
diff --git a/pandas/pandas/_libs/tslibs/ccalendar.pyx b/pandas/pandas/_libs/tslibs/ccalendar.pyx
new file mode 100644
index 0000000000000..536fa19f4c5d7
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/ccalendar.pyx
@@ -0,0 +1,310 @@
+# cython: boundscheck=False
+"""
+Cython implementations of functions resembling the stdlib calendar module
+"""
+cimport cython
+from numpy cimport (
+    int32_t,
+    int64_t,
+)
+
+# ----------------------------------------------------------------------
+# Constants
+
+# Slightly more performant cython lookups than a 2D table
+# The first 12 entries correspond to month lengths for non-leap years.
+# The remaining 12 entries give month lengths for leap years
+cdef int32_t* days_per_month_array = [
+    31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31,
+    31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
+
+cdef int* em = [0, 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334]
+
+# The first 13 entries give the month days elapsed as of the first of month N
+# (or the total number of days in the year for N=13) in non-leap years.
+# The remaining 13 entries give the days elapsed in leap years.
+cdef int32_t* month_offset = [
+    0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365,
+    0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366]
+
+# Canonical location for other modules to find name constants
+MONTHS = ["JAN", "FEB", "MAR", "APR", "MAY", "JUN", "JUL",
+          "AUG", "SEP", "OCT", "NOV", "DEC"]
+# The first blank line is consistent with calendar.month_name in the calendar
+# standard library
+MONTHS_FULL = ["", "January", "February", "March", "April", "May", "June",
+               "July", "August", "September", "October", "November",
+               "December"]
+MONTH_NUMBERS = {name: num for num, name in enumerate(MONTHS)}
+cdef dict c_MONTH_NUMBERS = MONTH_NUMBERS
+MONTH_ALIASES = {(num + 1): name for num, name in enumerate(MONTHS)}
+cdef dict MONTH_TO_CAL_NUM = {name: num + 1 for num, name in enumerate(MONTHS)}
+
+DAYS = ["MON", "TUE", "WED", "THU", "FRI", "SAT", "SUN"]
+DAYS_FULL = ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday",
+             "Saturday", "Sunday"]
+int_to_weekday = {num: name for num, name in enumerate(DAYS)}
+weekday_to_int = {int_to_weekday[key]: key for key in int_to_weekday}
+
+
+# ----------------------------------------------------------------------
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef int32_t get_days_in_month(int year, Py_ssize_t month) noexcept nogil:
+    """
+    Return the number of days in the given month of the given year.
+
+    Parameters
+    ----------
+    year : int
+    month : int
+
+    Returns
+    -------
+    days_in_month : int
+
+    Notes
+    -----
+    Assumes that the arguments are valid.  Passing a month not between 1 and 12
+    risks a segfault.
+    """
+    return days_per_month_array[12 * is_leapyear(year) + month - 1]
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+@cython.cdivision(True)
+cdef long quot(long a , long b) noexcept nogil:
+    cdef long x
+    x = a/b
+    if (a < 0):
+        x -= (a % b != 0)
+    return x
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+@cython.cdivision(True)
+cdef int dayofweek(int y, int m, int d) noexcept nogil:
+    """
+    Find the day of week for the date described by the Y/M/D triple y, m, d
+    using Gauss' method, from wikipedia.
+
+    0 represents Monday.  See [1]_.
+
+    Parameters
+    ----------
+    y : int
+    m : int
+    d : int
+
+    Returns
+    -------
+    weekday : int
+
+    Notes
+    -----
+    Assumes that y, m, d, represents a valid date.
+
+    See Also
+    --------
+    [1] https://docs.python.org/3/library/calendar.html#calendar.weekday
+
+    [2] https://en.wikipedia.org/wiki/\
+    Determination_of_the_day_of_the_week#Gauss's_algorithm
+    """
+    # Note: this particular implementation comes from
+    # http://berndt-schwerdtfeger.de/wp-content/uploads/pdf/cal.pdf
+    cdef:
+        long c
+        int g
+        int f
+        int e
+
+    if (m < 3):
+        y -= 1
+
+    c = quot(y, 100)
+    g = y - c * 100
+    f = 5 * (c - quot(c, 4) * 4)
+    e = em[m]
+
+    if (m > 2):
+        e -= 1
+    return (-1 + d + e + f + g + g/4) % 7
+
+cdef bint is_leapyear(int64_t year) noexcept nogil:
+    """
+    Returns 1 if the given year is a leap year, 0 otherwise.
+
+    Parameters
+    ----------
+    year : int
+
+    Returns
+    -------
+    is_leap : bool
+    """
+    return ((year & 0x3) == 0 and  # year % 4 == 0
+            ((year % 100) != 0 or (year % 400) == 0))
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef int32_t get_week_of_year(int year, int month, int day) noexcept nogil:
+    """
+    Return the ordinal week-of-year for the given day.
+
+    Parameters
+    ----------
+    year : int
+    month : int
+    day : int
+
+    Returns
+    -------
+    week_of_year : int32_t
+
+    Notes
+    -----
+    Assumes the inputs describe a valid date.
+    """
+    return get_iso_calendar(year, month, day)[1]
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef iso_calendar_t get_iso_calendar(int year, int month, int day) noexcept nogil:
+    """
+    Return the year, week, and day of year corresponding to ISO 8601
+
+    Parameters
+    ----------
+    year : int
+    month : int
+    day : int
+
+    Returns
+    -------
+    year : int32_t
+    week : int32_t
+    day : int32_t
+
+    Notes
+    -----
+    Assumes the inputs describe a valid date.
+    """
+    cdef:
+        int32_t doy, dow
+        int32_t iso_year, iso_week
+
+    doy = get_day_of_year(year, month, day)
+    dow = dayofweek(year, month, day)
+
+    # estimate
+    iso_week = (doy - 1) - dow + 3
+    if iso_week >= 0:
+        iso_week = iso_week // 7 + 1
+
+    # verify
+    if iso_week < 0:
+        if (iso_week > -2) or (iso_week == -2 and is_leapyear(year - 1)):
+            iso_week = 53
+        else:
+            iso_week = 52
+    elif iso_week == 53:
+        if 31 - day + dow < 3:
+            iso_week = 1
+
+    iso_year = year
+    if iso_week == 1 and month == 12:
+        iso_year += 1
+
+    elif iso_week >= 52 and month == 1:
+        iso_year -= 1
+
+    return iso_year, iso_week, dow + 1
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef int32_t get_day_of_year(int year, int month, int day) noexcept nogil:
+    """
+    Return the ordinal day-of-year for the given day.
+
+    Parameters
+    ----------
+    year : int
+    month : int
+    day : int
+
+    Returns
+    -------
+    day_of_year : int32_t
+
+    Notes
+    -----
+    Assumes the inputs describe a valid date.
+    """
+    cdef:
+        bint isleap
+        int32_t mo_off
+        int day_of_year
+
+    isleap = is_leapyear(year)
+
+    mo_off = month_offset[isleap * 13 + month - 1]
+
+    day_of_year = mo_off + day
+    return day_of_year
+
+
+# ---------------------------------------------------------------------
+# Business Helpers
+
+cpdef int get_lastbday(int year, int month) noexcept nogil:
+    """
+    Find the last day of the month that is a business day.
+
+    Parameters
+    ----------
+    year : int
+    month : int
+
+    Returns
+    -------
+    last_bday : int
+    """
+    cdef:
+        int wkday, days_in_month
+
+    wkday = dayofweek(year, month, 1)
+    days_in_month = get_days_in_month(year, month)
+    return days_in_month - max(((wkday + days_in_month - 1) % 7) - 4, 0)
+
+
+cpdef int get_firstbday(int year, int month) noexcept nogil:
+    """
+    Find the first day of the month that is a business day.
+
+    Parameters
+    ----------
+    year : int
+    month : int
+
+    Returns
+    -------
+    first_bday : int
+    """
+    cdef:
+        int first, wkday
+
+    wkday = dayofweek(year, month, 1)
+    first = 1
+    if wkday == 5:  # on Saturday
+        first = 3
+    elif wkday == 6:  # on Sunday
+        first = 2
+    return first
diff --git a/pandas/pandas/_libs/tslibs/conversion.pxd b/pandas/pandas/_libs/tslibs/conversion.pxd
new file mode 100644
index 0000000000000..6b9f41b1bb06f
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/conversion.pxd
@@ -0,0 +1,59 @@
+from cpython.datetime cimport (
+    datetime,
+    tzinfo,
+)
+from numpy cimport (
+    int32_t,
+    int64_t,
+    ndarray,
+)
+
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    npy_datetimestruct,
+)
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+from pandas._libs.tslibs.timezones cimport tz_compare
+
+
+cdef class _TSObject:
+    cdef readonly:
+        npy_datetimestruct dts      # npy_datetimestruct
+        int64_t value               # numpy dt64
+        tzinfo tzinfo
+        bint fold
+        NPY_DATETIMEUNIT creso
+
+    cdef int64_t ensure_reso(
+        self, NPY_DATETIMEUNIT creso, val=*, bint round_ok=*
+    ) except? -1
+
+
+cdef _TSObject convert_to_tsobject(object ts, tzinfo tz, str unit,
+                                   bint dayfirst, bint yearfirst,
+                                   int32_t nanos=*)
+
+cdef _TSObject convert_datetime_to_tsobject(datetime ts, tzinfo tz,
+                                            int32_t nanos=*,
+                                            NPY_DATETIMEUNIT reso=*)
+
+cdef _TSObject convert_str_to_tsobject(str ts, tzinfo tz,
+                                       bint dayfirst=*,
+                                       bint yearfirst=*)
+
+cdef int64_t get_datetime64_nanos(object val, NPY_DATETIMEUNIT reso) except? -1
+
+cpdef datetime localize_pydatetime(datetime dt, tzinfo tz)
+cdef int64_t cast_from_unit(object ts, str unit, NPY_DATETIMEUNIT out_reso=*) except? -1
+cdef (int64_t, int) precision_from_unit(
+    NPY_DATETIMEUNIT in_reso, NPY_DATETIMEUNIT out_reso=*
+)
+
+cdef maybe_localize_tso(_TSObject obj, tzinfo tz, NPY_DATETIMEUNIT reso)
+
+
+cdef int64_t parse_pydatetime(
+    datetime val,
+    npy_datetimestruct *dts,
+    NPY_DATETIMEUNIT creso,
+) except? -1
diff --git a/pandas/pandas/_libs/tslibs/conversion.pyi b/pandas/pandas/_libs/tslibs/conversion.pyi
new file mode 100644
index 0000000000000..26affae577f4d
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/conversion.pyi
@@ -0,0 +1,14 @@
+from datetime import (
+    datetime,
+    tzinfo,
+)
+
+import numpy as np
+
+DT64NS_DTYPE: np.dtype
+TD64NS_DTYPE: np.dtype
+
+def localize_pydatetime(dt: datetime, tz: tzinfo | None) -> datetime: ...
+def cast_from_unit_vectorized(
+    values: np.ndarray, unit: str, out_unit: str = ...
+) -> np.ndarray: ...
diff --git a/pandas/pandas/_libs/tslibs/conversion.pyx b/pandas/pandas/_libs/tslibs/conversion.pyx
new file mode 100644
index 0000000000000..7a8b4df447aee
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/conversion.pyx
@@ -0,0 +1,818 @@
+cimport cython
+
+import numpy as np
+
+cimport numpy as cnp
+from libc.math cimport log10
+from numpy cimport (
+    float64_t,
+    int32_t,
+    int64_t,
+)
+
+cnp.import_array()
+
+# stdlib datetime imports
+
+from datetime import timezone
+
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    datetime,
+    import_datetime,
+    time,
+    timedelta,
+    tzinfo,
+)
+
+import_datetime()
+
+from pandas._libs.missing cimport checknull_with_nat_and_na
+from pandas._libs.tslibs.dtypes cimport (
+    abbrev_to_npy_unit,
+    get_supported_reso,
+    npy_unit_to_attrname,
+    periods_per_second,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    NPY_FR_us,
+    astype_overflowsafe,
+    check_dts_bounds,
+    convert_reso,
+    dts_to_iso_string,
+    get_conversion_factor,
+    get_datetime64_unit,
+    get_implementation_bounds,
+    import_pandas_datetime,
+    npy_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pandas_datetime_to_datetimestruct,
+    pydatetime_to_dt64,
+    pydatetime_to_dtstruct,
+    string_to_dts,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_nat_strings as nat_strings,
+)
+from pandas._libs.tslibs.parsing cimport parse_datetime_string
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+from pandas._libs.tslibs.timezones cimport (
+    get_utcoffset,
+    is_utc,
+    treat_tz_as_pytz,
+)
+from pandas._libs.tslibs.tzconversion cimport (
+    Localizer,
+    tz_localize_to_utc_single,
+)
+from pandas._libs.tslibs.util cimport (
+    is_float_object,
+    is_integer_object,
+    is_nan,
+)
+
+# ----------------------------------------------------------------------
+# Constants
+
+DT64NS_DTYPE = np.dtype("M8[ns]")
+TD64NS_DTYPE = np.dtype("m8[ns]")
+
+
+# ----------------------------------------------------------------------
+# Unit Conversion Helpers
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+@cython.overflowcheck(True)
+def cast_from_unit_vectorized(
+    ndarray values,
+    str unit,
+    str out_unit="ns",
+):
+    """
+    Vectorized analogue to cast_from_unit.
+    """
+    cdef:
+        int64_t m
+        int p
+        NPY_DATETIMEUNIT in_reso, out_reso
+        Py_ssize_t i
+
+    assert values.dtype.kind == "f"
+
+    if unit in "YM":
+        if not (((values % 1) == 0) | np.isnan(values)).all():
+            # GH#47267 it is clear that 2 "M" corresponds to 1970-02-01,
+            #  but not clear what 2.5 "M" corresponds to, so we will
+            #  disallow that case.
+            raise ValueError(
+                f"Conversion of non-round float with unit={unit} "
+                "is ambiguous"
+            )
+
+        # GH#47266 go through np.datetime64 to avoid weird results e.g. with "Y"
+        #  and 150 we'd get 2120-01-01 09:00:00
+        values = values.astype(f"M8[{unit}]")
+        dtype = np.dtype(f"M8[{out_unit}]")
+        return astype_overflowsafe(values, dtype=dtype, copy=False).view("i8")
+
+    in_reso = abbrev_to_npy_unit(unit)
+    out_reso = abbrev_to_npy_unit(out_unit)
+    m, p = precision_from_unit(in_reso, out_reso)
+
+    cdef:
+        ndarray[int64_t] base, out
+        ndarray[float64_t] frac
+        tuple shape = (<object>values).shape
+
+    out = np.empty(shape, dtype="i8")
+    base = np.empty(shape, dtype="i8")
+    frac = np.empty(shape, dtype="f8")
+
+    for i in range(len(values)):
+        if is_nan(values[i]):
+            base[i] = NPY_NAT
+        else:
+            base[i] = <int64_t>values[i]
+            frac[i] = values[i] - base[i]
+
+    if p:
+        frac = np.round(frac, p)
+
+    for i in range(len(values)):
+        try:
+            if base[i] == NPY_NAT:
+                out[i] = NPY_NAT
+            else:
+                out[i] = <int64_t>(base[i] * m) + <int64_t>(frac[i] * m)
+        except (OverflowError, FloatingPointError) as err:
+            # FloatingPointError can be issued if we have float dtype and have
+            #  set np.errstate(over="raise")
+            raise OutOfBoundsDatetime(
+                f"cannot convert input {values[i]} with the unit '{unit}'"
+            ) from err
+    return out
+
+
+cdef int64_t cast_from_unit(
+    object ts,
+    str unit,
+    NPY_DATETIMEUNIT out_reso=NPY_FR_ns
+) except? -1:
+    """
+    Return a casting of the unit represented to nanoseconds
+    round the fractional part of a float to our precision, p.
+
+    Parameters
+    ----------
+    ts : int, float, or None
+    unit : str
+
+    Returns
+    -------
+    int64_t
+    """
+    cdef:
+        int64_t m
+        int p
+        NPY_DATETIMEUNIT in_reso
+
+    if unit in ["Y", "M"]:
+        if is_float_object(ts) and not ts.is_integer():
+            # GH#47267 it is clear that 2 "M" corresponds to 1970-02-01,
+            #  but not clear what 2.5 "M" corresponds to, so we will
+            #  disallow that case.
+            raise ValueError(
+                f"Conversion of non-round float with unit={unit} "
+                "is ambiguous"
+            )
+        # GH#47266 go through np.datetime64 to avoid weird results e.g. with "Y"
+        #  and 150 we'd get 2120-01-01 09:00:00
+        if is_float_object(ts):
+            ts = int(ts)
+        dt64obj = np.datetime64(ts, unit)
+        return get_datetime64_nanos(dt64obj, out_reso)
+
+    in_reso = abbrev_to_npy_unit(unit)
+    if out_reso < in_reso and in_reso != NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        # We will end up rounding (always *down*), so don't need the fractional
+        #  part of `ts`.
+        m, _ = precision_from_unit(out_reso, in_reso)
+        return (<int64_t>ts) // m
+
+    m, p = precision_from_unit(in_reso, out_reso)
+
+    # cast the unit, multiply base/frac separately
+    # to avoid precision issues from float -> int
+    try:
+        base = <int64_t>ts
+    except OverflowError as err:
+        raise OutOfBoundsDatetime(
+            f"cannot convert input {ts} with the unit '{unit}'"
+        ) from err
+
+    frac = ts - base
+    if p:
+        frac = round(frac, p)
+
+    try:
+        return <int64_t>(base * m) + <int64_t>(frac * m)
+    except OverflowError as err:
+        raise OutOfBoundsDatetime(
+            f"cannot convert input {ts} with the unit '{unit}'"
+        ) from err
+
+
+cdef (int64_t, int) precision_from_unit(
+    NPY_DATETIMEUNIT in_reso,
+    NPY_DATETIMEUNIT out_reso=NPY_DATETIMEUNIT.NPY_FR_ns,
+):
+    """
+    Return a casting of the unit represented to nanoseconds + the precision
+    to round the fractional part.
+
+    Notes
+    -----
+    The caller is responsible for ensuring that the default value of "ns"
+    takes the place of None.
+    """
+    cdef:
+        int64_t m
+        int64_t multiplier
+        int p
+
+    if in_reso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        in_reso = NPY_DATETIMEUNIT.NPY_FR_ns
+    if in_reso == NPY_DATETIMEUNIT.NPY_FR_Y:
+        # each 400 years we have 97 leap years, for an average of 97/400=.2425
+        #  extra days each year. We get 31556952 by writing
+        #  3600*24*365.2425=31556952
+        multiplier = periods_per_second(out_reso)
+        m = multiplier * 31556952
+    elif in_reso == NPY_DATETIMEUNIT.NPY_FR_M:
+        # 2629746 comes from dividing the "Y" case by 12.
+        multiplier = periods_per_second(out_reso)
+        m = multiplier * 2629746
+    else:
+        # Careful: if get_conversion_factor raises, the exception does
+        #  not propagate, instead we get a warning about an ignored exception.
+        #  https://github.com/pandas-dev/pandas/pull/51483#discussion_r1115198951
+        m = get_conversion_factor(in_reso, out_reso)
+
+    p = <int>log10(m)  # number of digits in 'm' minus 1
+    return m, p
+
+
+cdef int64_t get_datetime64_nanos(object val, NPY_DATETIMEUNIT reso) except? -1:
+    """
+    Extract the value and unit from a np.datetime64 object, then convert the
+    value to nanoseconds if necessary.
+    """
+    cdef:
+        npy_datetimestruct dts
+        NPY_DATETIMEUNIT unit
+        npy_datetime ival
+
+    ival = cnp.get_datetime64_value(val)
+    if ival == NPY_NAT:
+        return NPY_NAT
+
+    unit = get_datetime64_unit(val)
+
+    if unit != reso:
+        pandas_datetime_to_datetimestruct(ival, unit, &dts)
+        try:
+            ival = npy_datetimestruct_to_datetime(reso, &dts)
+        except OverflowError as err:
+            attrname = npy_unit_to_attrname[reso]
+            raise OutOfBoundsDatetime(
+                f"Out of bounds {attrname} timestamp: {val}"
+            ) from err
+
+    return ival
+
+
+# ----------------------------------------------------------------------
+# _TSObject Conversion
+
+# lightweight C object to hold datetime & int64 pair
+cdef class _TSObject:
+    # cdef:
+    #    npy_datetimestruct dts      # npy_datetimestruct
+    #    int64_t value               # numpy dt64
+    #    tzinfo tzinfo
+    #    bint fold
+    #    NPY_DATETIMEUNIT creso
+
+    def __cinit__(self):
+        # GH 25057. As per PEP 495, set fold to 0 by default
+        self.fold = 0
+        self.creso = NPY_FR_ns  # default value
+
+    cdef int64_t ensure_reso(
+        self, NPY_DATETIMEUNIT creso, val=None, bint round_ok=False
+    ) except? -1:
+        if self.creso != creso:
+            try:
+                self.value = convert_reso(
+                    self.value, self.creso, creso, round_ok=round_ok
+                )
+            except OverflowError as err:
+                if val is not None:
+                    attrname = npy_unit_to_attrname[creso]
+                    raise OutOfBoundsDatetime(
+                        f"Out of bounds {attrname} timestamp: {val}"
+                    ) from err
+                raise OutOfBoundsDatetime from err
+
+            self.creso = creso
+        return self.value
+
+
+cdef _TSObject convert_to_tsobject(object ts, tzinfo tz, str unit,
+                                   bint dayfirst, bint yearfirst, int32_t nanos=0):
+    """
+    Extract datetime and int64 from any of:
+        - np.int64 (with unit providing a possible modifier)
+        - np.datetime64
+        - a float (with unit providing a possible modifier)
+        - python int or long object (with unit providing a possible modifier)
+        - iso8601 string object
+        - python datetime object
+        - another timestamp object
+
+    Raises
+    ------
+    OutOfBoundsDatetime : ts cannot be converted within implementation bounds
+    """
+    cdef:
+        _TSObject obj
+        NPY_DATETIMEUNIT reso
+
+    obj = _TSObject()
+
+    if isinstance(ts, str):
+        return convert_str_to_tsobject(ts, tz, dayfirst, yearfirst)
+
+    if checknull_with_nat_and_na(ts):
+        obj.value = NPY_NAT
+    elif cnp.is_datetime64_object(ts):
+        reso = get_supported_reso(get_datetime64_unit(ts))
+        obj.creso = reso
+        obj.value = get_datetime64_nanos(ts, reso)
+        if obj.value != NPY_NAT:
+            pandas_datetime_to_datetimestruct(obj.value, reso, &obj.dts)
+            if tz is not None:
+                # GH#24559, GH#42288 We treat np.datetime64 objects as *wall* times
+                obj.value = tz_localize_to_utc_single(
+                    obj.value, tz, ambiguous="raise", nonexistent=None, creso=reso
+                )
+    elif is_integer_object(ts):
+        try:
+            ts = <int64_t>ts
+        except OverflowError:
+            # GH#26651 re-raise as OutOfBoundsDatetime
+            raise OutOfBoundsDatetime(f"Out of bounds nanosecond timestamp {ts}")
+        if ts == NPY_NAT:
+            obj.value = NPY_NAT
+        else:
+            if unit is None:
+                unit = "ns"
+            in_reso = abbrev_to_npy_unit(unit)
+            reso = get_supported_reso(in_reso)
+            ts = cast_from_unit(ts, unit, reso)
+            obj.value = ts
+            obj.creso = reso
+            pandas_datetime_to_datetimestruct(ts, reso, &obj.dts)
+    elif is_float_object(ts):
+        if ts != ts or ts == NPY_NAT:
+            obj.value = NPY_NAT
+        else:
+            ts = cast_from_unit(ts, unit)
+            obj.value = ts
+            pandas_datetime_to_datetimestruct(ts, NPY_FR_ns, &obj.dts)
+    elif PyDateTime_Check(ts):
+        if nanos == 0:
+            if isinstance(ts, _Timestamp):
+                reso = (<_Timestamp>ts)._creso
+            else:
+                # TODO: what if user explicitly passes nanos=0?
+                reso = NPY_FR_us
+        else:
+            reso = NPY_FR_ns
+        return convert_datetime_to_tsobject(ts, tz, nanos, reso=reso)
+    elif PyDate_Check(ts):
+        # Keep the converter same as PyDateTime's
+        # For date object we give the lowest supported resolution, i.e. "s"
+        ts = datetime.combine(ts, time())
+        return convert_datetime_to_tsobject(
+            ts, tz, nanos=0, reso=NPY_DATETIMEUNIT.NPY_FR_s
+        )
+    else:
+        from .period import Period
+        if isinstance(ts, Period):
+            raise ValueError("Cannot convert Period to Timestamp "
+                             "unambiguously. Use to_timestamp")
+        raise TypeError(f"Cannot convert input [{ts}] of type {type(ts)} to "
+                        f"Timestamp")
+
+    maybe_localize_tso(obj, tz, obj.creso)
+    return obj
+
+
+cdef maybe_localize_tso(_TSObject obj, tzinfo tz, NPY_DATETIMEUNIT reso):
+    if tz is not None:
+        _localize_tso(obj, tz, reso)
+
+    if obj.value != NPY_NAT:
+        # check_overflows needs to run after _localize_tso
+        check_dts_bounds(&obj.dts, reso)
+        check_overflows(obj, reso)
+
+
+cdef _TSObject convert_datetime_to_tsobject(
+    datetime ts,
+    tzinfo tz,
+    int32_t nanos=0,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Convert a datetime (or Timestamp) input `ts`, along with optional timezone
+    object `tz` to a _TSObject.
+
+    The optional argument `nanos` allows for cases where datetime input
+    needs to be supplemented with higher-precision information.
+
+    Parameters
+    ----------
+    ts : datetime or Timestamp
+        Value to be converted to _TSObject
+    tz : tzinfo or None
+        timezone for the timezone-aware output
+    nanos : int32_t, default is 0
+        nanoseconds supplement the precision of the datetime input ts
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    obj : _TSObject
+    """
+    cdef:
+        _TSObject obj = _TSObject()
+        int64_t pps
+
+    obj.creso = reso
+    obj.fold = ts.fold
+    if tz is not None:
+
+        if ts.tzinfo is not None:
+            # Convert the current timezone to the passed timezone
+            ts = ts.astimezone(tz)
+            pydatetime_to_dtstruct(ts, &obj.dts)
+            obj.tzinfo = ts.tzinfo
+        elif not is_utc(tz):
+            ts = _localize_pydatetime(ts, tz)
+            pydatetime_to_dtstruct(ts, &obj.dts)
+            obj.tzinfo = ts.tzinfo
+        else:
+            # UTC
+            pydatetime_to_dtstruct(ts, &obj.dts)
+            obj.tzinfo = tz
+    else:
+        pydatetime_to_dtstruct(ts, &obj.dts)
+        obj.tzinfo = ts.tzinfo
+
+    if isinstance(ts, _Timestamp):
+        obj.dts.ps = ts.nanosecond * 1000
+
+    if nanos:
+        obj.dts.ps = nanos * 1000
+
+    try:
+        obj.value = npy_datetimestruct_to_datetime(reso, &obj.dts)
+    except OverflowError as err:
+        attrname = npy_unit_to_attrname[reso]
+        raise OutOfBoundsDatetime(f"Out of bounds {attrname} timestamp") from err
+
+    if obj.tzinfo is not None and not is_utc(obj.tzinfo):
+        offset = get_utcoffset(obj.tzinfo, ts)
+        pps = periods_per_second(reso)
+        obj.value -= int(offset.total_seconds() * pps)
+
+    check_overflows(obj, reso)
+    return obj
+
+
+cdef _adjust_tsobject_tz_using_offset(_TSObject obj, tzinfo tz):
+    """
+    Convert a datetimestruct `obj.dts`, with an attached tzinfo to a new
+    user-provided tz.
+
+    Parameters
+    ----------
+    obj : _TSObject
+    tz : tzinfo
+        timezone for the timezone-aware output.
+    """
+    cdef:
+        datetime dt
+        Py_ssize_t pos
+        int64_t ps = obj.dts.ps
+        Localizer info = Localizer(tz, obj.creso)
+
+    # Infer fold from offset-adjusted obj.value
+    # see PEP 495 https://www.python.org/dev/peps/pep-0495/#the-fold-attribute
+    if info.use_utc:
+        pass
+    elif info.use_tzlocal:
+        info.utc_val_to_local_val(obj.value, &pos, &obj.fold)
+    elif info.use_dst and not info.use_pytz:
+        # i.e. dateutil
+        info.utc_val_to_local_val(obj.value, &pos, &obj.fold)
+
+    # Keep the converter same as PyDateTime's
+    dt = datetime(obj.dts.year, obj.dts.month, obj.dts.day,
+                  obj.dts.hour, obj.dts.min, obj.dts.sec,
+                  obj.dts.us, obj.tzinfo, fold=obj.fold)
+
+    # The rest here is similar to the 2-tz path in convert_datetime_to_tsobject
+    #  but avoids re-calculating obj.value
+    dt = dt.astimezone(tz)
+    pydatetime_to_dtstruct(dt, &obj.dts)
+    obj.tzinfo = dt.tzinfo
+    obj.dts.ps = ps
+    check_dts_bounds(&obj.dts, obj.creso)
+    check_overflows(obj, obj.creso)
+
+
+cdef _TSObject convert_str_to_tsobject(str ts, tzinfo tz,
+                                       bint dayfirst=False,
+                                       bint yearfirst=False):
+    """
+    Convert a string input `ts`, along with optional timezone object`tz`
+    to a _TSObject.
+
+    The optional arguments `dayfirst` and `yearfirst` are passed to the
+    dateutil parser.
+
+    Parameters
+    ----------
+    ts : str
+        Value to be converted to _TSObject
+    tz : tzinfo or None
+        timezone for the timezone-aware output
+    dayfirst : bool, default False
+        When parsing an ambiguous date string, interpret e.g. "3/4/1975" as
+        April 3, as opposed to the standard US interpretation March 4.
+    yearfirst : bool, default False
+        When parsing an ambiguous date string, interpret e.g. "01/05/09"
+        as "May 9, 2001", as opposed to the default "Jan 5, 2009"
+
+    Returns
+    -------
+    obj : _TSObject
+    """
+    cdef:
+        npy_datetimestruct dts
+        int out_local = 0, out_tzoffset = 0, string_to_dts_failed
+        datetime dt
+        int64_t ival, nanos = 0
+        NPY_DATETIMEUNIT out_bestunit, reso
+        _TSObject obj
+
+    if len(ts) == 0 or ts in nat_strings:
+        obj = _TSObject()
+        obj.value = NPY_NAT
+        obj.tzinfo = tz
+        return obj
+    elif ts == "now":
+        # Issue 9000, we short-circuit rather than going
+        # into np_datetime_strings which returns utc
+        dt = datetime.now(tz)
+        return convert_datetime_to_tsobject(dt, tz, nanos=0, reso=NPY_FR_us)
+    elif ts == "today":
+        # Issue 9000, we short-circuit rather than going
+        # into np_datetime_strings which returns a normalized datetime
+        dt = datetime.now(tz)
+        # equiv: datetime.today().replace(tzinfo=tz)
+        return convert_datetime_to_tsobject(dt, tz, nanos=0, reso=NPY_FR_us)
+    else:
+        if not dayfirst:  # GH 58859
+            string_to_dts_failed = string_to_dts(
+                ts, &dts, &out_bestunit, &out_local,
+                &out_tzoffset, False
+            )
+            if not string_to_dts_failed:
+                reso = get_supported_reso(out_bestunit)
+                check_dts_bounds(&dts, reso)
+                obj = _TSObject()
+                obj.dts = dts
+                obj.creso = reso
+                ival = npy_datetimestruct_to_datetime(reso, &dts)
+
+                if out_local == 1:
+                    obj.tzinfo = timezone(timedelta(minutes=out_tzoffset))
+                    obj.value = tz_localize_to_utc_single(
+                        ival,
+                        obj.tzinfo,
+                        ambiguous="raise",
+                        nonexistent=None,
+                        creso=reso,
+                    )
+                    if tz is None:
+                        check_overflows(obj, reso)
+                        return obj
+                    _adjust_tsobject_tz_using_offset(obj, tz)
+                    return  obj
+                else:
+                    if tz is not None:
+                        # shift for _localize_tso
+                        ival = tz_localize_to_utc_single(
+                            ival, tz, ambiguous="raise", nonexistent=None, creso=reso
+                        )
+                    obj.value = ival
+                    maybe_localize_tso(obj, tz, obj.creso)
+                    return obj
+
+        dt = parse_datetime_string(
+            ts,
+            dayfirst=dayfirst,
+            yearfirst=yearfirst,
+            out_bestunit=&out_bestunit,
+            nanos=&nanos,
+        )
+        reso = get_supported_reso(out_bestunit)
+        return convert_datetime_to_tsobject(dt, tz, nanos=nanos, reso=reso)
+
+
+cdef check_overflows(_TSObject obj, NPY_DATETIMEUNIT reso=NPY_FR_ns):
+    """
+    Check that we haven't silently overflowed in timezone conversion
+
+    Parameters
+    ----------
+    obj : _TSObject
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    None
+
+    Raises
+    ------
+    OutOfBoundsDatetime
+    """
+    # GH#12677
+    cdef:
+        npy_datetimestruct lb, ub
+
+    get_implementation_bounds(reso, &lb, &ub)
+
+    if obj.dts.year == lb.year:
+        if not (obj.value < 0):
+            from pandas._libs.tslibs.timestamps import Timestamp
+            fmt = dts_to_iso_string(&obj.dts)
+            min_ts = (<_Timestamp>Timestamp(0))._as_creso(reso).min
+            raise OutOfBoundsDatetime(
+                f"Converting {fmt} underflows past {min_ts}"
+            )
+    elif obj.dts.year == ub.year:
+        if not (obj.value > 0):
+            from pandas._libs.tslibs.timestamps import Timestamp
+            fmt = dts_to_iso_string(&obj.dts)
+            max_ts = (<_Timestamp>Timestamp(0))._as_creso(reso).max
+            raise OutOfBoundsDatetime(
+                f"Converting {fmt} overflows past {max_ts}"
+            )
+
+# ----------------------------------------------------------------------
+# Localization
+
+cdef void _localize_tso(_TSObject obj, tzinfo tz, NPY_DATETIMEUNIT reso) noexcept:
+    """
+    Given the UTC nanosecond timestamp in obj.value, find the wall-clock
+    representation of that timestamp in the given timezone.
+
+    Parameters
+    ----------
+    obj : _TSObject
+    tz : tzinfo
+    reso : NPY_DATETIMEUNIT
+
+    Returns
+    -------
+    None
+
+    Notes
+    -----
+    Sets obj.tzinfo inplace, alters obj.dts inplace.
+    """
+    cdef:
+        int64_t local_val
+        Py_ssize_t outpos = -1
+        Localizer info = Localizer(tz, reso)
+
+    assert obj.tzinfo is None
+
+    if info.use_utc:
+        pass
+    elif obj.value == NPY_NAT:
+        pass
+    else:
+        local_val = info.utc_val_to_local_val(obj.value, &outpos, &obj.fold)
+
+        if info.use_pytz:
+            # infer we went through a pytz path, will have outpos!=-1
+            tz = tz._tzinfos[tz._transition_info[outpos]]
+
+        pandas_datetime_to_datetimestruct(local_val, reso, &obj.dts)
+
+    obj.tzinfo = tz
+
+
+cdef datetime _localize_pydatetime(datetime dt, tzinfo tz):
+    """
+    Take a datetime/Timestamp in UTC and localizes to timezone tz.
+
+    NB: Unlike the public version, this treats datetime and Timestamp objects
+        identically, i.e. discards nanos from Timestamps.
+        It also assumes that the `tz` input is not None.
+    """
+    if treat_tz_as_pytz(tz):
+        import pytz
+
+        # datetime.replace with pytz may be incorrect result
+        # TODO: try to respect `fold` attribute
+        try:
+            return tz.localize(dt, is_dst=None)
+        except (pytz.AmbiguousTimeError, pytz.NonExistentTimeError) as err:
+            # As of pandas 3.0, we raise ValueErrors instead of pytz exceptions
+            raise ValueError(str(err)) from err
+    else:
+        return dt.replace(tzinfo=tz)
+
+
+cpdef inline datetime localize_pydatetime(datetime dt, tzinfo tz):
+    """
+    Take a datetime/Timestamp in UTC and localizes to timezone tz.
+
+    Parameters
+    ----------
+    dt : datetime or Timestamp
+    tz : tzinfo or None
+
+    Returns
+    -------
+    localized : datetime or Timestamp
+    """
+    if tz is None:
+        return dt
+    elif isinstance(dt, _Timestamp):
+        return dt.tz_localize(tz)
+    return _localize_pydatetime(dt, tz)
+
+
+cdef int64_t parse_pydatetime(
+    datetime val,
+    npy_datetimestruct *dts,
+    NPY_DATETIMEUNIT creso,
+) except? -1:
+    """
+    Convert pydatetime to datetime64.
+
+    Parameters
+    ----------
+    val : datetime
+        Element being processed.
+    dts : *npy_datetimestruct
+        Needed to use in pydatetime_to_dt64, which writes to it.
+    creso : NPY_DATETIMEUNIT
+        Resolution to store the the result.
+
+    Raises
+    ------
+    OutOfBoundsDatetime
+    """
+    cdef:
+        _TSObject _ts
+        int64_t result
+
+    if val.tzinfo is not None:
+        _ts = convert_datetime_to_tsobject(val, None, nanos=0, reso=creso)
+        result = _ts.value
+    else:
+        if isinstance(val, _Timestamp):
+            result = (<_Timestamp>val)._as_creso(creso, round_ok=True)._value
+        else:
+            result = pydatetime_to_dt64(val, dts, reso=creso)
+    return result
diff --git a/pandas/pandas/_libs/tslibs/dtypes.pxd b/pandas/pandas/_libs/tslibs/dtypes.pxd
new file mode 100644
index 0000000000000..d8c536a34bc04
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/dtypes.pxd
@@ -0,0 +1,115 @@
+from numpy cimport int64_t
+
+from pandas._libs.tslibs.np_datetime cimport NPY_DATETIMEUNIT
+
+
+cdef str npy_unit_to_abbrev(NPY_DATETIMEUNIT unit)
+cpdef NPY_DATETIMEUNIT abbrev_to_npy_unit(str abbrev)
+cdef NPY_DATETIMEUNIT freq_group_code_to_npy_unit(int freq) noexcept nogil
+cpdef int64_t periods_per_day(NPY_DATETIMEUNIT reso=*) except? -1
+cpdef int64_t periods_per_second(NPY_DATETIMEUNIT reso) except? -1
+cdef NPY_DATETIMEUNIT get_supported_reso(NPY_DATETIMEUNIT reso)
+cdef bint is_supported_unit(NPY_DATETIMEUNIT reso)
+
+cdef dict c_OFFSET_TO_PERIOD_FREQSTR
+cdef dict c_PERIOD_TO_OFFSET_FREQSTR
+cdef dict c_OFFSET_RENAMED_FREQSTR
+cdef dict c_DEPR_UNITS
+cdef dict c_PERIOD_AND_OFFSET_DEPR_FREQSTR
+cdef dict attrname_to_abbrevs
+cdef dict npy_unit_to_attrname
+cdef dict attrname_to_npy_unit
+cdef str INVALID_FREQ_ERR_MSG
+
+cdef enum c_FreqGroup:
+    # Mirrors FreqGroup in the .pyx file
+    FR_ANN = 1000
+    FR_QTR = 2000
+    FR_MTH = 3000
+    FR_WK = 4000
+    FR_BUS = 5000
+    FR_DAY = 6000
+    FR_HR = 7000
+    FR_MIN = 8000
+    FR_SEC = 9000
+    FR_MS = 10000
+    FR_US = 11000
+    FR_NS = 12000
+    FR_UND = -10000  # undefined
+
+
+cdef enum c_Resolution:
+    # Mirrors Resolution in the .pyx file
+    RESO_NS = 0
+    RESO_US = 1
+    RESO_MS = 2
+    RESO_SEC = 3
+    RESO_MIN = 4
+    RESO_HR = 5
+    RESO_DAY = 6
+    RESO_MTH = 7
+    RESO_QTR = 8
+    RESO_YR = 9
+
+
+cdef enum PeriodDtypeCode:
+    # Annual freqs with various fiscal year ends.
+    # eg, 2005 for A_FEB runs Mar 1, 2004 to Feb 28, 2005
+    A = 1000      # Default alias
+    A_DEC = 1000  # Annual - December year end
+    A_JAN = 1001  # Annual - January year end
+    A_FEB = 1002  # Annual - February year end
+    A_MAR = 1003  # Annual - March year end
+    A_APR = 1004  # Annual - April year end
+    A_MAY = 1005  # Annual - May year end
+    A_JUN = 1006  # Annual - June year end
+    A_JUL = 1007  # Annual - July year end
+    A_AUG = 1008  # Annual - August year end
+    A_SEP = 1009  # Annual - September year end
+    A_OCT = 1010  # Annual - October year end
+    A_NOV = 1011  # Annual - November year end
+
+    # Quarterly frequencies with various fiscal year ends.
+    # eg, Q42005 for Q_OCT runs Aug 1, 2005 to Oct 31, 2005
+    Q_DEC = 2000    # Quarterly - December year end
+    Q_JAN = 2001    # Quarterly - January year end
+    Q_FEB = 2002    # Quarterly - February year end
+    Q_MAR = 2003    # Quarterly - March year end
+    Q_APR = 2004    # Quarterly - April year end
+    Q_MAY = 2005    # Quarterly - May year end
+    Q_JUN = 2006    # Quarterly - June year end
+    Q_JUL = 2007    # Quarterly - July year end
+    Q_AUG = 2008    # Quarterly - August year end
+    Q_SEP = 2009    # Quarterly - September year end
+    Q_OCT = 2010    # Quarterly - October year end
+    Q_NOV = 2011    # Quarterly - November year end
+
+    M = 3000        # Monthly
+
+    W_SUN = 4000    # Weekly - Sunday end of week
+    W_MON = 4001    # Weekly - Monday end of week
+    W_TUE = 4002    # Weekly - Tuesday end of week
+    W_WED = 4003    # Weekly - Wednesday end of week
+    W_THU = 4004    # Weekly - Thursday end of week
+    W_FRI = 4005    # Weekly - Friday end of week
+    W_SAT = 4006    # Weekly - Saturday end of week
+
+    B = 5000        # Business days
+    D = 6000        # Daily
+    H = 7000        # Hourly
+    T = 8000        # Minutely
+    S = 9000        # Secondly
+    L = 10000       # Millisecondly
+    U = 11000       # Microsecondly
+    N = 12000       # Nanosecondly
+
+    UNDEFINED = -10_000
+
+
+cdef class PeriodDtypeBase:
+    cdef readonly:
+        PeriodDtypeCode _dtype_code
+        int64_t _n
+
+    cpdef int _get_to_timestamp_base(self)
+    cpdef bint _is_tick_like(self)
diff --git a/pandas/pandas/_libs/tslibs/dtypes.pyi b/pandas/pandas/_libs/tslibs/dtypes.pyi
new file mode 100644
index 0000000000000..b435b9e2d8b31
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/dtypes.pyi
@@ -0,0 +1,84 @@
+from enum import Enum
+
+from pandas._typing import Self
+
+OFFSET_TO_PERIOD_FREQSTR: dict[str, str]
+
+def periods_per_day(reso: int = ...) -> int: ...
+def periods_per_second(reso: int) -> int: ...
+def abbrev_to_npy_unit(abbrev: str | None) -> int: ...
+
+class PeriodDtypeBase:
+    _dtype_code: int  # PeriodDtypeCode
+    _n: int
+
+    # actually __cinit__
+    def __new__(cls, code: int, n: int) -> Self: ...
+    @property
+    def _freq_group_code(self) -> int: ...
+    @property
+    def _resolution_obj(self) -> Resolution: ...
+    def _get_to_timestamp_base(self) -> int: ...
+    @property
+    def _freqstr(self) -> str: ...
+    def __hash__(self) -> int: ...
+    def _is_tick_like(self) -> bool: ...
+    @property
+    def _creso(self) -> int: ...
+    @property
+    def _td64_unit(self) -> str: ...
+
+class FreqGroup(Enum):
+    FR_ANN: int
+    FR_QTR: int
+    FR_MTH: int
+    FR_WK: int
+    FR_BUS: int
+    FR_DAY: int
+    FR_HR: int
+    FR_MIN: int
+    FR_SEC: int
+    FR_MS: int
+    FR_US: int
+    FR_NS: int
+    FR_UND: int
+    @staticmethod
+    def from_period_dtype_code(code: int) -> FreqGroup: ...
+
+class Resolution(Enum):
+    RESO_NS: int
+    RESO_US: int
+    RESO_MS: int
+    RESO_SEC: int
+    RESO_MIN: int
+    RESO_HR: int
+    RESO_DAY: int
+    RESO_MTH: int
+    RESO_QTR: int
+    RESO_YR: int
+    def __lt__(self, other: Resolution) -> bool: ...
+    def __ge__(self, other: Resolution) -> bool: ...
+    @property
+    def attrname(self) -> str: ...
+    @classmethod
+    def from_attrname(cls, attrname: str) -> Resolution: ...
+    @classmethod
+    def get_reso_from_freqstr(cls, freq: str) -> Resolution: ...
+    @property
+    def attr_abbrev(self) -> str: ...
+
+class NpyDatetimeUnit(Enum):
+    NPY_FR_Y: int
+    NPY_FR_M: int
+    NPY_FR_W: int
+    NPY_FR_D: int
+    NPY_FR_h: int
+    NPY_FR_m: int
+    NPY_FR_s: int
+    NPY_FR_ms: int
+    NPY_FR_us: int
+    NPY_FR_ns: int
+    NPY_FR_ps: int
+    NPY_FR_fs: int
+    NPY_FR_as: int
+    NPY_FR_GENERIC: int
diff --git a/pandas/pandas/_libs/tslibs/dtypes.pyx b/pandas/pandas/_libs/tslibs/dtypes.pyx
new file mode 100644
index 0000000000000..4100f3d90e817
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/dtypes.pyx
@@ -0,0 +1,650 @@
+# period frequency constants corresponding to scikits timeseries
+# originals
+from enum import Enum
+
+from pandas._libs.tslibs.ccalendar cimport c_MONTH_NUMBERS
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    get_conversion_factor,
+    import_pandas_datetime,
+)
+
+import_pandas_datetime()
+
+cdef class PeriodDtypeBase:
+    """
+    Similar to an actual dtype, this contains all of the information
+    describing a PeriodDtype in an integer code.
+    """
+    # cdef readonly:
+    #    PeriodDtypeCode _dtype_code
+    #    int64_t _n
+
+    def __cinit__(self, PeriodDtypeCode code, int64_t n):
+        self._dtype_code = code
+        self._n = n
+
+    def __eq__(self, other):
+        if not isinstance(other, PeriodDtypeBase):
+            return False
+        if not isinstance(self, PeriodDtypeBase):
+            # cython semantics, this is a reversed op
+            return False
+        return self._dtype_code == other._dtype_code and self._n == other._n
+
+    def __hash__(self) -> int:
+        return hash((self._n, self._dtype_code))
+
+    @property
+    def _freq_group_code(self) -> int:
+        # See also: libperiod.get_freq_group
+        return (self._dtype_code // 1000) * 1000
+
+    @property
+    def _resolution_obj(self) -> "Resolution":
+        fgc = self._freq_group_code
+        freq_group = FreqGroup(fgc)
+        abbrev = _period_code_to_abbrev[freq_group.value].split("-")[0]
+        if abbrev == "B":
+            return Resolution.RESO_DAY
+        attrname = _abbrev_to_attrnames[abbrev]
+        return Resolution.from_attrname(attrname)
+
+    @property
+    def _freqstr(self) -> str:
+        # Will be passed to to_offset in Period._maybe_convert_freq
+        out = _period_code_to_abbrev.get(self._dtype_code)
+        if self._n == 1:
+            return out
+        return str(self._n) + out
+
+    cpdef int _get_to_timestamp_base(self):
+        """
+        Return frequency code group used for base of to_timestamp against
+        frequency code.
+
+        Return day freq code against longer freq than day.
+        Return second freq code against hour between second.
+
+        Returns
+        -------
+        int
+        """
+        base = <c_FreqGroup>self._dtype_code
+        if base < FR_BUS:
+            return FR_DAY
+        elif FR_HR <= base <= FR_SEC:
+            return FR_SEC
+        return base
+
+    cpdef bint _is_tick_like(self):
+        return self._dtype_code >= PeriodDtypeCode.D
+
+    @property
+    def _creso(self) -> int:
+        return {
+            PeriodDtypeCode.D: NPY_DATETIMEUNIT.NPY_FR_D,
+            PeriodDtypeCode.H: NPY_DATETIMEUNIT.NPY_FR_h,
+            PeriodDtypeCode.T: NPY_DATETIMEUNIT.NPY_FR_m,
+            PeriodDtypeCode.S: NPY_DATETIMEUNIT.NPY_FR_s,
+            PeriodDtypeCode.L: NPY_DATETIMEUNIT.NPY_FR_ms,
+            PeriodDtypeCode.U: NPY_DATETIMEUNIT.NPY_FR_us,
+            PeriodDtypeCode.N: NPY_DATETIMEUNIT.NPY_FR_ns,
+        }[self._dtype_code]
+
+    @property
+    def _td64_unit(self) -> str:
+        return npy_unit_to_abbrev(self._creso)
+
+
+_period_code_map = {
+    # Annual freqs with various fiscal year ends.
+    # eg, 2005 for Y-FEB runs Mar 1, 2004 to Feb 28, 2005
+    "Y-DEC": PeriodDtypeCode.A_DEC,  # Annual - December year end
+    "Y-JAN": PeriodDtypeCode.A_JAN,  # Annual - January year end
+    "Y-FEB": PeriodDtypeCode.A_FEB,  # Annual - February year end
+    "Y-MAR": PeriodDtypeCode.A_MAR,  # Annual - March year end
+    "Y-APR": PeriodDtypeCode.A_APR,  # Annual - April year end
+    "Y-MAY": PeriodDtypeCode.A_MAY,  # Annual - May year end
+    "Y-JUN": PeriodDtypeCode.A_JUN,  # Annual - June year end
+    "Y-JUL": PeriodDtypeCode.A_JUL,  # Annual - July year end
+    "Y-AUG": PeriodDtypeCode.A_AUG,  # Annual - August year end
+    "Y-SEP": PeriodDtypeCode.A_SEP,  # Annual - September year end
+    "Y-OCT": PeriodDtypeCode.A_OCT,  # Annual - October year end
+    "Y-NOV": PeriodDtypeCode.A_NOV,  # Annual - November year end
+
+    # Quarterly frequencies with various fiscal year ends.
+    # eg, Q42005 for Q-OCT runs Aug 1, 2005 to Oct 31, 2005
+    "Q-DEC": PeriodDtypeCode.Q_DEC,    # Quarterly - December year end
+    "Q-JAN": PeriodDtypeCode.Q_JAN,    # Quarterly - January year end
+    "Q-FEB": PeriodDtypeCode.Q_FEB,    # Quarterly - February year end
+    "Q-MAR": PeriodDtypeCode.Q_MAR,    # Quarterly - March year end
+    "Q-APR": PeriodDtypeCode.Q_APR,    # Quarterly - April year end
+    "Q-MAY": PeriodDtypeCode.Q_MAY,    # Quarterly - May year end
+    "Q-JUN": PeriodDtypeCode.Q_JUN,    # Quarterly - June year end
+    "Q-JUL": PeriodDtypeCode.Q_JUL,    # Quarterly - July year end
+    "Q-AUG": PeriodDtypeCode.Q_AUG,    # Quarterly - August year end
+    "Q-SEP": PeriodDtypeCode.Q_SEP,    # Quarterly - September year end
+    "Q-OCT": PeriodDtypeCode.Q_OCT,    # Quarterly - October year end
+    "Q-NOV": PeriodDtypeCode.Q_NOV,    # Quarterly - November year end
+
+    "M": PeriodDtypeCode.M,        # Monthly
+
+    "W-SUN": PeriodDtypeCode.W_SUN,    # Weekly - Sunday end of week
+    "W-MON": PeriodDtypeCode.W_MON,    # Weekly - Monday end of week
+    "W-TUE": PeriodDtypeCode.W_TUE,    # Weekly - Tuesday end of week
+    "W-WED": PeriodDtypeCode.W_WED,    # Weekly - Wednesday end of week
+    "W-THU": PeriodDtypeCode.W_THU,    # Weekly - Thursday end of week
+    "W-FRI": PeriodDtypeCode.W_FRI,    # Weekly - Friday end of week
+    "W-SAT": PeriodDtypeCode.W_SAT,    # Weekly - Saturday end of week
+
+    "B": PeriodDtypeCode.B,        # Business days
+    "D": PeriodDtypeCode.D,        # Daily
+    "h": PeriodDtypeCode.H,        # Hourly
+    "min": PeriodDtypeCode.T,      # Minutely
+    "s": PeriodDtypeCode.S,        # Secondly
+    "ms": PeriodDtypeCode.L,       # Millisecondly
+    "us": PeriodDtypeCode.U,       # Microsecondly
+    "ns": PeriodDtypeCode.N,       # Nanosecondly
+}
+
+cdef dict _period_code_to_abbrev = {
+    _period_code_map[key]: key for key in _period_code_map}
+
+cdef set _month_names = set(c_MONTH_NUMBERS.keys())
+
+# Map attribute-name resolutions to resolution abbreviations
+cdef dict attrname_to_abbrevs = {
+    "year": "Y",
+    "quarter": "Q",
+    "month": "M",
+    "day": "D",
+    "hour": "h",
+    "minute": "min",
+    "second": "s",
+    "millisecond": "ms",
+    "microsecond": "us",
+    "nanosecond": "ns",
+}
+cdef dict _abbrev_to_attrnames = {v: k for k, v in attrname_to_abbrevs.items()}
+
+OFFSET_TO_PERIOD_FREQSTR: dict = {
+    "WEEKDAY": "D",
+    "EOM": "M",
+    "BME": "M",
+    "SME": "M",
+    "BMS": "M",
+    "CBME": "M",
+    "CBMS": "M",
+    "SMS": "M",
+    "BQS": "Q",
+    "QS": "Q",
+    "BQE": "Q",
+    "BQE-DEC": "Q-DEC",
+    "BQE-JAN": "Q-JAN",
+    "BQE-FEB": "Q-FEB",
+    "BQE-MAR": "Q-MAR",
+    "BQE-APR": "Q-APR",
+    "BQE-MAY": "Q-MAY",
+    "BQE-JUN": "Q-JUN",
+    "BQE-JUL": "Q-JUL",
+    "BQE-AUG": "Q-AUG",
+    "BQE-SEP": "Q-SEP",
+    "BQE-OCT": "Q-OCT",
+    "BQE-NOV": "Q-NOV",
+    "MS": "M",
+    "D": "D",
+    "B": "B",
+    "min": "min",
+    "s": "s",
+    "ms": "ms",
+    "us": "us",
+    "ns": "ns",
+    "h": "h",
+    "QE": "Q",
+    "QE-DEC": "Q-DEC",
+    "QE-JAN": "Q-JAN",
+    "QE-FEB": "Q-FEB",
+    "QE-MAR": "Q-MAR",
+    "QE-APR": "Q-APR",
+    "QE-MAY": "Q-MAY",
+    "QE-JUN": "Q-JUN",
+    "QE-JUL": "Q-JUL",
+    "QE-AUG": "Q-AUG",
+    "QE-SEP": "Q-SEP",
+    "QE-OCT": "Q-OCT",
+    "QE-NOV": "Q-NOV",
+    "YE": "Y",
+    "YE-DEC": "Y-DEC",
+    "YE-JAN": "Y-JAN",
+    "YE-FEB": "Y-FEB",
+    "YE-MAR": "Y-MAR",
+    "YE-APR": "Y-APR",
+    "YE-MAY": "Y-MAY",
+    "YE-JUN": "Y-JUN",
+    "YE-JUL": "Y-JUL",
+    "YE-AUG": "Y-AUG",
+    "YE-SEP": "Y-SEP",
+    "YE-OCT": "Y-OCT",
+    "YE-NOV": "Y-NOV",
+    "W": "W",
+    "ME": "M",
+    "BYE": "Y",
+    "BYE-DEC": "Y-DEC",
+    "BYE-JAN": "Y-JAN",
+    "BYE-FEB": "Y-FEB",
+    "BYE-MAR": "Y-MAR",
+    "BYE-APR": "Y-APR",
+    "BYE-MAY": "Y-MAY",
+    "BYE-JUN": "Y-JUN",
+    "BYE-JUL": "Y-JUL",
+    "BYE-AUG": "Y-AUG",
+    "BYE-SEP": "Y-SEP",
+    "BYE-OCT": "Y-OCT",
+    "BYE-NOV": "Y-NOV",
+    "YS": "Y",
+    "BYS": "Y",
+}
+cdef dict c_OFFSET_RENAMED_FREQSTR = {
+    "M": "ME",
+    "Q": "QE",
+    "Q-DEC": "QE-DEC",
+    "Q-JAN": "QE-JAN",
+    "Q-FEB": "QE-FEB",
+    "Q-MAR": "QE-MAR",
+    "Q-APR": "QE-APR",
+    "Q-MAY": "QE-MAY",
+    "Q-JUN": "QE-JUN",
+    "Q-JUL": "QE-JUL",
+    "Q-AUG": "QE-AUG",
+    "Q-SEP": "QE-SEP",
+    "Q-OCT": "QE-OCT",
+    "Q-NOV": "QE-NOV",
+    "Y": "YE",
+    "Y-DEC": "YE-DEC",
+    "Y-JAN": "YE-JAN",
+    "Y-FEB": "YE-FEB",
+    "Y-MAR": "YE-MAR",
+    "Y-APR": "YE-APR",
+    "Y-MAY": "YE-MAY",
+    "Y-JUN": "YE-JUN",
+    "Y-JUL": "YE-JUL",
+    "Y-AUG": "YE-AUG",
+    "Y-SEP": "YE-SEP",
+    "Y-OCT": "YE-OCT",
+    "Y-NOV": "YE-NOV",
+    "BY": "BYE",
+    "BY-DEC": "BYE-DEC",
+    "BY-JAN": "BYE-JAN",
+    "BY-FEB": "BYE-FEB",
+    "BY-MAR": "BYE-MAR",
+    "BY-APR": "BYE-APR",
+    "BY-MAY": "BYE-MAY",
+    "BY-JUN": "BYE-JUN",
+    "BY-JUL": "BYE-JUL",
+    "BY-AUG": "BYE-AUG",
+    "BY-SEP": "BYE-SEP",
+    "BY-OCT": "BYE-OCT",
+    "BY-NOV": "BYE-NOV",
+    "BM": "BME",
+    "CBM": "CBME",
+    "SM": "SME",
+    "BQ": "BQE",
+    "BQ-DEC": "BQE-DEC",
+    "BQ-JAN": "BQE-JAN",
+    "BQ-FEB": "BQE-FEB",
+    "BQ-MAR": "BQE-MAR",
+    "BQ-APR": "BQE-APR",
+    "BQ-MAY": "BQE-MAY",
+    "BQ-JUN": "BQE-JUN",
+    "BQ-JUL": "BQE-JUL",
+    "BQ-AUG": "BQE-AUG",
+    "BQ-SEP": "BQE-SEP",
+    "BQ-OCT": "BQE-OCT",
+    "BQ-NOV": "BQE-NOV",
+}
+PERIOD_TO_OFFSET_FREQSTR = {
+    "M": "ME",
+    "Q": "QE",
+    "Q-DEC": "QE-DEC",
+    "Q-JAN": "QE-JAN",
+    "Q-FEB": "QE-FEB",
+    "Q-MAR": "QE-MAR",
+    "Q-APR": "QE-APR",
+    "Q-MAY": "QE-MAY",
+    "Q-JUN": "QE-JUN",
+    "Q-JUL": "QE-JUL",
+    "Q-AUG": "QE-AUG",
+    "Q-SEP": "QE-SEP",
+    "Q-OCT": "QE-OCT",
+    "Q-NOV": "QE-NOV",
+    "Y": "YE",
+    "Y-DEC": "YE-DEC",
+    "Y-JAN": "YE-JAN",
+    "Y-FEB": "YE-FEB",
+    "Y-MAR": "YE-MAR",
+    "Y-APR": "YE-APR",
+    "Y-MAY": "YE-MAY",
+    "Y-JUN": "YE-JUN",
+    "Y-JUL": "YE-JUL",
+    "Y-AUG": "YE-AUG",
+    "Y-SEP": "YE-SEP",
+    "Y-OCT": "YE-OCT",
+    "Y-NOV": "YE-NOV",
+}
+cdef dict c_OFFSET_TO_PERIOD_FREQSTR = OFFSET_TO_PERIOD_FREQSTR
+cdef dict c_PERIOD_TO_OFFSET_FREQSTR = PERIOD_TO_OFFSET_FREQSTR
+
+cdef dict c_DEPR_UNITS = {
+    "w": "W",
+    "d": "D",
+    "H": "h",
+    "MIN": "min",
+    "S": "s",
+    "MS": "ms",
+    "US": "us",
+    "NS": "ns",
+}
+
+cdef dict c_PERIOD_AND_OFFSET_DEPR_FREQSTR = {
+    "w": "W",
+    "w-mon": "W-MON",
+    "w-tue": "W-TUE",
+    "w-wed": "W-WED",
+    "w-thu": "W-THU",
+    "w-fri": "W-FRI",
+    "w-sat": "W-SAT",
+    "w-sun": "W-SUN",
+    "d": "D",
+    "b": "B",
+    "c": "C",
+    "MIN": "min",
+    "US": "us",
+    "NS": "ns",
+}
+
+cdef str INVALID_FREQ_ERR_MSG = "Invalid frequency: {0}"
+
+
+class FreqGroup(Enum):
+    # Mirrors c_FreqGroup in the .pxd file
+    FR_ANN = c_FreqGroup.FR_ANN
+    FR_QTR = c_FreqGroup.FR_QTR
+    FR_MTH = c_FreqGroup.FR_MTH
+    FR_WK = c_FreqGroup.FR_WK
+    FR_BUS = c_FreqGroup.FR_BUS
+    FR_DAY = c_FreqGroup.FR_DAY
+    FR_HR = c_FreqGroup.FR_HR
+    FR_MIN = c_FreqGroup.FR_MIN
+    FR_SEC = c_FreqGroup.FR_SEC
+    FR_MS = c_FreqGroup.FR_MS
+    FR_US = c_FreqGroup.FR_US
+    FR_NS = c_FreqGroup.FR_NS
+    FR_UND = c_FreqGroup.FR_UND  # undefined
+
+    @staticmethod
+    def from_period_dtype_code(code: int) -> "FreqGroup":
+        # See also: PeriodDtypeBase._freq_group_code
+        code = (code // 1000) * 1000
+        return FreqGroup(code)
+
+
+class Resolution(Enum):
+    RESO_NS = c_Resolution.RESO_NS
+    RESO_US = c_Resolution.RESO_US
+    RESO_MS = c_Resolution.RESO_MS
+    RESO_SEC = c_Resolution.RESO_SEC
+    RESO_MIN = c_Resolution.RESO_MIN
+    RESO_HR = c_Resolution.RESO_HR
+    RESO_DAY = c_Resolution.RESO_DAY
+    RESO_MTH = c_Resolution.RESO_MTH
+    RESO_QTR = c_Resolution.RESO_QTR
+    RESO_YR = c_Resolution.RESO_YR
+
+    def __lt__(self, other):
+        return self.value < other.value
+
+    def __ge__(self, other):
+        return self.value >= other.value
+
+    @property
+    def attr_abbrev(self) -> str:
+        # string that we can pass to to_offset
+        return attrname_to_abbrevs[self.attrname]
+
+    @property
+    def attrname(self) -> str:
+        """
+        Return datetime attribute name corresponding to this Resolution.
+
+        Examples
+        --------
+        >>> Resolution.RESO_SEC.attrname
+        'second'
+        """
+        return _reso_str_map[self.value]
+
+    @classmethod
+    def from_attrname(cls, attrname: str) -> "Resolution":
+        """
+        Return resolution str against resolution code.
+
+        Examples
+        --------
+        >>> Resolution.from_attrname('second')
+        <Resolution.RESO_SEC: 3>
+
+        >>> Resolution.from_attrname('second') == Resolution.RESO_SEC
+        True
+        """
+        return cls(_str_reso_map[attrname])
+
+    @classmethod
+    def get_reso_from_freqstr(cls, freq: str) -> "Resolution":
+        """
+        Return resolution code against frequency str.
+
+        `freq` is given by the `offset.freqstr` for some DateOffset object.
+
+        Examples
+        --------
+        >>> Resolution.get_reso_from_freqstr('h')
+        <Resolution.RESO_HR: 5>
+
+        >>> Resolution.get_reso_from_freqstr('h') == Resolution.RESO_HR
+        True
+        """
+        try:
+            attr_name = _abbrev_to_attrnames[freq]
+        except KeyError as exc:
+            msg = INVALID_FREQ_ERR_MSG.format(freq)
+            # For quarterly and yearly resolutions, we need to chop off
+            #  a month string.
+            split_freq = freq.split("-")
+            if len(split_freq) != 2:
+                raise ValueError(msg) from exc
+            if split_freq[1] not in _month_names:
+                # i.e. we want e.g. "Q-DEC", not "Q-INVALID"
+                raise ValueError(msg) from exc
+            attr_name = _abbrev_to_attrnames[split_freq[0]]
+
+        return cls.from_attrname(attr_name)
+
+
+class NpyDatetimeUnit(Enum):
+    """
+    Python-space analogue to NPY_DATETIMEUNIT.
+    """
+    NPY_FR_Y = NPY_DATETIMEUNIT.NPY_FR_Y
+    NPY_FR_M = NPY_DATETIMEUNIT.NPY_FR_M
+    NPY_FR_W = NPY_DATETIMEUNIT.NPY_FR_W
+    NPY_FR_D = NPY_DATETIMEUNIT.NPY_FR_D
+    NPY_FR_h = NPY_DATETIMEUNIT.NPY_FR_h
+    NPY_FR_m = NPY_DATETIMEUNIT.NPY_FR_m
+    NPY_FR_s = NPY_DATETIMEUNIT.NPY_FR_s
+    NPY_FR_ms = NPY_DATETIMEUNIT.NPY_FR_ms
+    NPY_FR_us = NPY_DATETIMEUNIT.NPY_FR_us
+    NPY_FR_ns = NPY_DATETIMEUNIT.NPY_FR_ns
+    NPY_FR_ps = NPY_DATETIMEUNIT.NPY_FR_ps
+    NPY_FR_fs = NPY_DATETIMEUNIT.NPY_FR_fs
+    NPY_FR_as = NPY_DATETIMEUNIT.NPY_FR_as
+    NPY_FR_GENERIC = NPY_DATETIMEUNIT.NPY_FR_GENERIC
+
+
+cdef NPY_DATETIMEUNIT get_supported_reso(NPY_DATETIMEUNIT reso):
+    # If we have an unsupported reso, return the nearest supported reso.
+    if reso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        # TODO: or raise ValueError? trying this gives unraisable errors, but
+        #  "except? -1" breaks at compile-time for unknown reasons
+        return NPY_DATETIMEUNIT.NPY_FR_ns
+    if reso < NPY_DATETIMEUNIT.NPY_FR_s:
+        return NPY_DATETIMEUNIT.NPY_FR_s
+    elif reso > NPY_DATETIMEUNIT.NPY_FR_ns:
+        return NPY_DATETIMEUNIT.NPY_FR_ns
+    return reso
+
+
+cdef bint is_supported_unit(NPY_DATETIMEUNIT reso):
+    return (
+        reso == NPY_DATETIMEUNIT.NPY_FR_ns
+        or reso == NPY_DATETIMEUNIT.NPY_FR_us
+        or reso == NPY_DATETIMEUNIT.NPY_FR_ms
+        or reso == NPY_DATETIMEUNIT.NPY_FR_s
+    )
+
+
+cdef str npy_unit_to_abbrev(NPY_DATETIMEUNIT unit):
+    if unit == NPY_DATETIMEUNIT.NPY_FR_ns or unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        # generic -> default to nanoseconds
+        return "ns"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_us:
+        return "us"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_ms:
+        return "ms"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_s:
+        return "s"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_m:
+        return "m"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_h:
+        return "h"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_D:
+        return "D"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_W:
+        return "W"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_M:
+        return "M"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_Y:
+        return "Y"
+
+    # Checks for not-really-supported units go at the end, as we don't expect
+    #  to see these often
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_ps:
+        return "ps"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_fs:
+        return "fs"
+    elif unit == NPY_DATETIMEUNIT.NPY_FR_as:
+        return "as"
+
+    else:
+        raise NotImplementedError(unit)
+
+
+cpdef NPY_DATETIMEUNIT abbrev_to_npy_unit(str abbrev):
+    if abbrev == "Y":
+        return NPY_DATETIMEUNIT.NPY_FR_Y
+    elif abbrev == "M":
+        return NPY_DATETIMEUNIT.NPY_FR_M
+    elif abbrev == "W":
+        return NPY_DATETIMEUNIT.NPY_FR_W
+    elif abbrev == "D" or abbrev == "d":
+        return NPY_DATETIMEUNIT.NPY_FR_D
+    elif abbrev == "h":
+        return NPY_DATETIMEUNIT.NPY_FR_h
+    elif abbrev == "m":
+        return NPY_DATETIMEUNIT.NPY_FR_m
+    elif abbrev == "s":
+        return NPY_DATETIMEUNIT.NPY_FR_s
+    elif abbrev == "ms":
+        return NPY_DATETIMEUNIT.NPY_FR_ms
+    elif abbrev == "us":
+        return NPY_DATETIMEUNIT.NPY_FR_us
+    elif abbrev == "ns":
+        return NPY_DATETIMEUNIT.NPY_FR_ns
+    elif abbrev == "ps":
+        return NPY_DATETIMEUNIT.NPY_FR_ps
+    elif abbrev == "fs":
+        return NPY_DATETIMEUNIT.NPY_FR_fs
+    elif abbrev == "as":
+        return NPY_DATETIMEUNIT.NPY_FR_as
+    elif abbrev is None:
+        return NPY_DATETIMEUNIT.NPY_FR_GENERIC
+    else:
+        raise ValueError(f"Unrecognized unit {abbrev}")
+
+
+cdef NPY_DATETIMEUNIT freq_group_code_to_npy_unit(int freq) noexcept nogil:
+    """
+    Convert the freq to the corresponding NPY_DATETIMEUNIT to pass
+    to npy_datetimestruct_to_datetime.
+    """
+    if freq == FR_MTH:
+        return NPY_DATETIMEUNIT.NPY_FR_M
+    elif freq == FR_DAY:
+        return NPY_DATETIMEUNIT.NPY_FR_D
+    elif freq == FR_HR:
+        return NPY_DATETIMEUNIT.NPY_FR_h
+    elif freq == FR_MIN:
+        return NPY_DATETIMEUNIT.NPY_FR_m
+    elif freq == FR_SEC:
+        return NPY_DATETIMEUNIT.NPY_FR_s
+    elif freq == FR_MS:
+        return NPY_DATETIMEUNIT.NPY_FR_ms
+    elif freq == FR_US:
+        return NPY_DATETIMEUNIT.NPY_FR_us
+    elif freq == FR_NS:
+        return NPY_DATETIMEUNIT.NPY_FR_ns
+    elif freq == FR_UND:
+        # Default to Day
+        return NPY_DATETIMEUNIT.NPY_FR_D
+
+
+cpdef int64_t periods_per_day(
+    NPY_DATETIMEUNIT reso=NPY_DATETIMEUNIT.NPY_FR_ns
+) except? -1:
+    """
+    How many of the given time units fit into a single day?
+    """
+    return get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_D, reso)
+
+
+cpdef int64_t periods_per_second(NPY_DATETIMEUNIT reso) except? -1:
+    return get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_s, reso)
+
+
+cdef dict _reso_str_map = {
+    Resolution.RESO_NS.value: "nanosecond",
+    Resolution.RESO_US.value: "microsecond",
+    Resolution.RESO_MS.value: "millisecond",
+    Resolution.RESO_SEC.value: "second",
+    Resolution.RESO_MIN.value: "minute",
+    Resolution.RESO_HR.value: "hour",
+    Resolution.RESO_DAY.value: "day",
+    Resolution.RESO_MTH.value: "month",
+    Resolution.RESO_QTR.value: "quarter",
+    Resolution.RESO_YR.value: "year",
+}
+
+cdef dict _str_reso_map = {v: k for k, v in _reso_str_map.items()}
+
+cdef dict npy_unit_to_attrname = {
+    NPY_DATETIMEUNIT.NPY_FR_Y: "year",
+    NPY_DATETIMEUNIT.NPY_FR_M: "month",
+    NPY_DATETIMEUNIT.NPY_FR_D: "day",
+    NPY_DATETIMEUNIT.NPY_FR_h: "hour",
+    NPY_DATETIMEUNIT.NPY_FR_m: "minute",
+    NPY_DATETIMEUNIT.NPY_FR_s: "second",
+    NPY_DATETIMEUNIT.NPY_FR_ms: "millisecond",
+    NPY_DATETIMEUNIT.NPY_FR_us: "microsecond",
+    NPY_DATETIMEUNIT.NPY_FR_ns: "nanosecond",
+}
+cdef dict attrname_to_npy_unit = {v: k for k, v in npy_unit_to_attrname.items()}
diff --git a/pandas/pandas/_libs/tslibs/fields.pyi b/pandas/pandas/_libs/tslibs/fields.pyi
new file mode 100644
index 0000000000000..bc55e34f3d208
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/fields.pyi
@@ -0,0 +1,62 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def build_field_sarray(
+    dtindex: npt.NDArray[np.int64],  # const int64_t[:]
+    reso: int,  # NPY_DATETIMEUNIT
+) -> np.ndarray: ...
+def month_position_check(fields, weekdays) -> str | None: ...
+def get_date_name_field(
+    dtindex: npt.NDArray[np.int64],  # const int64_t[:]
+    field: str,
+    locale: str | None = ...,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.object_]: ...
+def get_start_end_field(
+    dtindex: npt.NDArray[np.int64],
+    field: str,
+    freq_name: str | None = ...,
+    month_kw: int = ...,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.bool_]: ...
+def get_date_field(
+    dtindex: npt.NDArray[np.int64],  # const int64_t[:]
+    field: str,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int32]: ...
+def get_timedelta_field(
+    tdindex: npt.NDArray[np.int64],  # const int64_t[:]
+    field: str,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int32]: ...
+def get_timedelta_days(
+    tdindex: npt.NDArray[np.int64],  # const int64_t[:]
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int64]: ...
+def isleapyear_arr(
+    years: np.ndarray,
+) -> npt.NDArray[np.bool_]: ...
+def build_isocalendar_sarray(
+    dtindex: npt.NDArray[np.int64],  # const int64_t[:]
+    reso: int,  # NPY_DATETIMEUNIT
+) -> np.ndarray: ...
+def _get_locale_names(name_type: str, locale: str | None = ...): ...
+
+class RoundTo:
+    @property
+    def MINUS_INFTY(self) -> int: ...
+    @property
+    def PLUS_INFTY(self) -> int: ...
+    @property
+    def NEAREST_HALF_EVEN(self) -> int: ...
+    @property
+    def NEAREST_HALF_PLUS_INFTY(self) -> int: ...
+    @property
+    def NEAREST_HALF_MINUS_INFTY(self) -> int: ...
+
+def round_nsint64(
+    values: npt.NDArray[np.int64],
+    mode: RoundTo,
+    nanos: int,
+) -> npt.NDArray[np.int64]: ...
diff --git a/pandas/pandas/_libs/tslibs/fields.pyx b/pandas/pandas/_libs/tslibs/fields.pyx
new file mode 100644
index 0000000000000..e523ac2e7b5c6
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/fields.pyx
@@ -0,0 +1,842 @@
+"""
+Functions for accessing attributes of Timestamp/datetime64/datetime-like
+objects and arrays
+"""
+from locale import LC_TIME
+
+from _strptime import LocaleTime
+
+cimport cython
+from cython cimport Py_ssize_t
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int8_t,
+    int32_t,
+    int64_t,
+    ndarray,
+    uint32_t,
+)
+
+cnp.import_array()
+
+from pandas._config.localization import set_locale
+
+from pandas._libs.tslibs.ccalendar import (
+    DAYS_FULL,
+    MONTHS_FULL,
+)
+
+from pandas._libs.tslibs.ccalendar cimport (
+    dayofweek,
+    get_day_of_year,
+    get_days_in_month,
+    get_firstbday,
+    get_iso_calendar,
+    get_lastbday,
+    get_week_of_year,
+    iso_calendar_t,
+)
+from pandas._libs.tslibs.nattype cimport NPY_NAT
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    pandas_datetime_to_datetimestruct,
+    pandas_timedelta_to_timedeltastruct,
+    pandas_timedeltastruct,
+)
+
+import_pandas_datetime()
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def build_field_sarray(const int64_t[:] dtindex, NPY_DATETIMEUNIT reso):
+    """
+    Datetime as int64 representation to a structured array of fields
+    """
+    cdef:
+        Py_ssize_t i, count = len(dtindex)
+        npy_datetimestruct dts
+        ndarray[int32_t] years, months, days, hours, minutes, seconds, mus
+
+    sa_dtype = [
+        ("Y", "i4"),  # year
+        ("M", "i4"),  # month
+        ("D", "i4"),  # day
+        ("h", "i4"),  # hour
+        ("m", "i4"),  # min
+        ("s", "i4"),  # second
+        ("u", "i4"),  # microsecond
+    ]
+
+    out = np.empty(count, dtype=sa_dtype)
+
+    years = out["Y"]
+    months = out["M"]
+    days = out["D"]
+    hours = out["h"]
+    minutes = out["m"]
+    seconds = out["s"]
+    mus = out["u"]
+
+    for i in range(count):
+        pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+        years[i] = dts.year
+        months[i] = dts.month
+        days[i] = dts.day
+        hours[i] = dts.hour
+        minutes[i] = dts.min
+        seconds[i] = dts.sec
+        mus[i] = dts.us
+
+    return out
+
+
+def month_position_check(fields, weekdays) -> str | None:
+    cdef:
+        int32_t daysinmonth, y, m, d
+        bint calendar_end = True
+        bint business_end = True
+        bint calendar_start = True
+        bint business_start = True
+        bint cal
+        int32_t[:] years = fields["Y"]
+        int32_t[:] months = fields["M"]
+        int32_t[:] days = fields["D"]
+
+    for y, m, d, wd in zip(years, months, days, weekdays):
+        if calendar_start:
+            calendar_start &= d == 1
+        if business_start:
+            business_start &= d == 1 or (d <= 3 and wd == 0)
+
+        if calendar_end or business_end:
+            daysinmonth = get_days_in_month(y, m)
+            cal = d == daysinmonth
+            if calendar_end:
+                calendar_end &= cal
+            if business_end:
+                business_end &= cal or (daysinmonth - d < 3 and wd == 4)
+        elif not calendar_start and not business_start:
+            break
+
+    if calendar_end:
+        return "ce"
+    elif business_end:
+        return "be"
+    elif calendar_start:
+        return "cs"
+    elif business_start:
+        return "bs"
+    else:
+        return None
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_date_name_field(
+    const int64_t[:] dtindex,
+    str field,
+    object locale=None,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Given a int64-based datetime index, return array of strings of date
+    name based on requested field (e.g. day_name)
+    """
+    cdef:
+        Py_ssize_t i
+        cnp.npy_intp count = dtindex.shape[0]
+        ndarray[object] out, names
+        npy_datetimestruct dts
+        int dow
+
+    out = cnp.PyArray_EMPTY(1, &count, cnp.NPY_OBJECT, 0)
+
+    if field == "day_name":
+        if locale is None:
+            names = np.array(DAYS_FULL, dtype=np.object_)
+        else:
+            names = np.array(_get_locale_names("f_weekday", locale),
+                             dtype=np.object_)
+        for i in range(count):
+            if dtindex[i] == NPY_NAT:
+                out[i] = np.nan
+                continue
+
+            pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+            dow = dayofweek(dts.year, dts.month, dts.day)
+            out[i] = names[dow].capitalize()
+
+    elif field == "month_name":
+        if locale is None:
+            names = np.array(MONTHS_FULL, dtype=np.object_)
+        else:
+            names = np.array(_get_locale_names("f_month", locale),
+                             dtype=np.object_)
+        for i in range(count):
+            if dtindex[i] == NPY_NAT:
+                out[i] = np.nan
+                continue
+
+            pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+            out[i] = names[dts.month].capitalize()
+
+    else:
+        raise ValueError(f"Field {field} not supported")
+
+    return out
+
+
+cdef bint _is_on_month(int month, int compare_month, int modby) noexcept nogil:
+    """
+    Analogous to DateOffset.is_on_offset checking for the month part of a date.
+    """
+    if modby == 1:
+        return True
+    elif modby == 3:
+        return (month - compare_month) % 3 == 0
+    else:
+        return month == compare_month
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_start_end_field(
+    const int64_t[:] dtindex,
+    str field,
+    str freq_name=None,
+    int month_kw=12,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Given an int64-based datetime index return array of indicators
+    of whether timestamps are at the start/end of the month/quarter/year
+    (defined by frequency).
+
+    Parameters
+    ----------
+    dtindex : ndarray[int64]
+    field : str
+    freq_name : str or None, default None
+    month_kw : int, default 12
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    ndarray[bool]
+    """
+    cdef:
+        Py_ssize_t i
+        int count = dtindex.shape[0]
+        bint is_business = 0
+        int end_month = 12
+        int start_month = 1
+        ndarray[int8_t] out
+        npy_datetimestruct dts
+        int compare_month, modby
+
+    out = np.zeros(count, dtype="int8")
+
+    if freq_name:
+        if freq_name == "C":
+            raise ValueError(f"Custom business days is not supported by {field}")
+        is_business = freq_name[0] == "B"
+
+        # YearBegin(), BYearBegin() use month = starting month of year.
+        # QuarterBegin(), BQuarterBegin() use startingMonth = starting
+        # month of year. Other offsets use month, startingMonth as ending
+        # month of year.
+
+        if freq_name.lstrip("B")[0:2] in ["QS", "YS"]:
+            end_month = 12 if month_kw == 1 else month_kw - 1
+            start_month = month_kw
+
+        else:
+            end_month = month_kw
+            start_month = (end_month % 12) + 1
+    else:
+        end_month = 12
+        start_month = 1
+
+    compare_month = start_month if "start" in field else end_month
+    if "month" in field:
+        modby = 1
+    elif "quarter" in field:
+        modby = 3
+    else:
+        modby = 12
+
+    if field in ["is_month_start", "is_quarter_start", "is_year_start"]:
+        if is_business:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = 0
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+
+                if _is_on_month(dts.month, compare_month, modby) and (
+                        dts.day == get_firstbday(dts.year, dts.month)):
+                    out[i] = 1
+
+        else:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = 0
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+
+                if _is_on_month(dts.month, compare_month, modby) and dts.day == 1:
+                    out[i] = 1
+
+    elif field in ["is_month_end", "is_quarter_end", "is_year_end"]:
+        if is_business:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = 0
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+
+                if _is_on_month(dts.month, compare_month, modby) and (
+                        dts.day == get_lastbday(dts.year, dts.month)):
+                    out[i] = 1
+
+        else:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = 0
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+
+                if _is_on_month(dts.month, compare_month, modby) and (
+                        dts.day == get_days_in_month(dts.year, dts.month)):
+                    out[i] = 1
+
+    else:
+        raise ValueError(f"Field {field} not supported")
+
+    return out.view(bool)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_date_field(
+    const int64_t[:] dtindex,
+    str field,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Given a int64-based datetime index, extract the year, month, etc.,
+    field and return an array of these values.
+    """
+    cdef:
+        Py_ssize_t i, count = len(dtindex)
+        ndarray[int32_t] out
+        npy_datetimestruct dts
+
+    out = np.empty(count, dtype="i4")
+
+    if field == "Y":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.year
+        return out
+
+    elif field == "M":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.month
+        return out
+
+    elif field == "D":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.day
+        return out
+
+    elif field == "h":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.hour
+                # TODO: can we de-dup with period.pyx <accessor>s?
+        return out
+
+    elif field == "m":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.min
+        return out
+
+    elif field == "s":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.sec
+        return out
+
+    elif field == "us":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.us
+        return out
+
+    elif field == "ns":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.ps // 1000
+        return out
+    elif field == "doy":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = get_day_of_year(dts.year, dts.month, dts.day)
+        return out
+
+    elif field == "dow":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dayofweek(dts.year, dts.month, dts.day)
+        return out
+
+    elif field == "woy":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = get_week_of_year(dts.year, dts.month, dts.day)
+        return out
+
+    elif field == "q":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = dts.month
+                out[i] = ((out[i] - 1) // 3) + 1
+        return out
+
+    elif field == "dim":
+        with nogil:
+            for i in range(count):
+                if dtindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                out[i] = get_days_in_month(dts.year, dts.month)
+        return out
+    elif field == "is_leap_year":
+        return isleapyear_arr(get_date_field(dtindex, "Y", reso=reso))
+
+    raise ValueError(f"Field {field} not supported")
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_timedelta_field(
+    const int64_t[:] tdindex,
+    str field,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Given a int64-based timedelta index, extract the days, hrs, sec.,
+    field and return an array of these values.
+    """
+    cdef:
+        Py_ssize_t i, count = len(tdindex)
+        ndarray[int32_t] out
+        pandas_timedeltastruct tds
+
+    out = np.empty(count, dtype="i4")
+
+    if field == "seconds":
+        with nogil:
+            for i in range(count):
+                if tdindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_timedelta_to_timedeltastruct(tdindex[i], reso, &tds)
+                out[i] = tds.seconds
+        return out
+
+    elif field == "microseconds":
+        with nogil:
+            for i in range(count):
+                if tdindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_timedelta_to_timedeltastruct(tdindex[i], reso, &tds)
+                out[i] = tds.microseconds
+        return out
+
+    elif field == "nanoseconds":
+        with nogil:
+            for i in range(count):
+                if tdindex[i] == NPY_NAT:
+                    out[i] = -1
+                    continue
+
+                pandas_timedelta_to_timedeltastruct(tdindex[i], reso, &tds)
+                out[i] = tds.nanoseconds
+        return out
+
+    raise ValueError(f"Field {field} not supported")
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_timedelta_days(
+    const int64_t[:] tdindex,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """
+    Given a int64-based timedelta index, extract the days,
+    field and return an array of these values.
+    """
+    cdef:
+        Py_ssize_t i, count = len(tdindex)
+        ndarray[int64_t] out
+        pandas_timedeltastruct tds
+
+    out = np.empty(count, dtype="i8")
+
+    with nogil:
+        for i in range(count):
+            if tdindex[i] == NPY_NAT:
+                out[i] = -1
+                continue
+
+            pandas_timedelta_to_timedeltastruct(tdindex[i], reso, &tds)
+            out[i] = tds.days
+    return out
+
+
+cpdef isleapyear_arr(ndarray years):
+    """vectorized version of isleapyear; NaT evaluates as False"""
+    cdef:
+        ndarray[int8_t] out
+
+    out = np.zeros(len(years), dtype="int8")
+    out[np.logical_or(years % 400 == 0,
+                      np.logical_and(years % 4 == 0,
+                                     years % 100 > 0))] = 1
+    return out.view(bool)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def build_isocalendar_sarray(const int64_t[:] dtindex, NPY_DATETIMEUNIT reso):
+    """
+    Given a int64-based datetime array, return the ISO 8601 year, week, and day
+    as a structured array.
+    """
+    cdef:
+        Py_ssize_t i, count = len(dtindex)
+        npy_datetimestruct dts
+        ndarray[uint32_t] iso_years, iso_weeks, days
+        iso_calendar_t ret_val
+
+    sa_dtype = [
+        ("year", "u4"),
+        ("week", "u4"),
+        ("day", "u4"),
+    ]
+
+    out = np.empty(count, dtype=sa_dtype)
+
+    iso_years = out["year"]
+    iso_weeks = out["week"]
+    days = out["day"]
+
+    with nogil:
+        for i in range(count):
+            if dtindex[i] == NPY_NAT:
+                ret_val = 0, 0, 0
+            else:
+                pandas_datetime_to_datetimestruct(dtindex[i], reso, &dts)
+                ret_val = get_iso_calendar(dts.year, dts.month, dts.day)
+
+            iso_years[i] = ret_val[0]
+            iso_weeks[i] = ret_val[1]
+            days[i] = ret_val[2]
+    return out
+
+
+def _get_locale_names(name_type: str, locale: object = None):
+    """
+    Returns an array of localized day or month names.
+
+    Parameters
+    ----------
+    name_type : str
+        Attribute of LocaleTime() in which to return localized names.
+    locale : str
+
+    Returns
+    -------
+    list of locale names
+    """
+    with set_locale(locale, LC_TIME):
+        return getattr(LocaleTime(), name_type)
+
+
+# ---------------------------------------------------------------------
+# Rounding
+
+
+class RoundTo:
+    """
+    enumeration defining the available rounding modes
+
+    Attributes
+    ----------
+    MINUS_INFTY
+        round towards -∞, or floor [2]_
+    PLUS_INFTY
+        round towards +∞, or ceil [3]_
+    NEAREST_HALF_EVEN
+        round to nearest, tie-break half to even [6]_
+    NEAREST_HALF_MINUS_INFTY
+        round to nearest, tie-break half to -∞ [5]_
+    NEAREST_HALF_PLUS_INFTY
+        round to nearest, tie-break half to +∞ [4]_
+
+
+    References
+    ----------
+    .. [1] "Rounding - Wikipedia"
+           https://en.wikipedia.org/wiki/Rounding
+    .. [2] "Rounding down"
+           https://en.wikipedia.org/wiki/Rounding#Rounding_down
+    .. [3] "Rounding up"
+           https://en.wikipedia.org/wiki/Rounding#Rounding_up
+    .. [4] "Round half up"
+           https://en.wikipedia.org/wiki/Rounding#Round_half_up
+    .. [5] "Round half down"
+           https://en.wikipedia.org/wiki/Rounding#Round_half_down
+    .. [6] "Round half to even"
+           https://en.wikipedia.org/wiki/Rounding#Round_half_to_even
+    """
+    @property
+    def MINUS_INFTY(self) -> int:
+        return 0
+
+    @property
+    def PLUS_INFTY(self) -> int:
+        return 1
+
+    @property
+    def NEAREST_HALF_EVEN(self) -> int:
+        return 2
+
+    @property
+    def NEAREST_HALF_PLUS_INFTY(self) -> int:
+        return 3
+
+    @property
+    def NEAREST_HALF_MINUS_INFTY(self) -> int:
+        return 4
+
+
+cdef ndarray[int64_t] _floor_int64(const int64_t[:] values, int64_t unit):
+    cdef:
+        Py_ssize_t i, n = len(values)
+        ndarray[int64_t] result = np.empty(n, dtype="i8")
+        int64_t res, value
+
+    with cython.overflowcheck(True):
+        for i in range(n):
+            value = values[i]
+            if value == NPY_NAT:
+                res = NPY_NAT
+            else:
+                res = value - value % unit
+            result[i] = res
+
+    return result
+
+
+cdef ndarray[int64_t] _ceil_int64(const int64_t[:] values, int64_t unit):
+    cdef:
+        Py_ssize_t i, n = len(values)
+        ndarray[int64_t] result = np.empty(n, dtype="i8")
+        int64_t res, value, remainder
+
+    with cython.overflowcheck(True):
+        for i in range(n):
+            value = values[i]
+
+            if value == NPY_NAT:
+                res = NPY_NAT
+            else:
+                remainder = value % unit
+                if remainder == 0:
+                    res = value
+                else:
+                    res = value + (unit - remainder)
+
+            result[i] = res
+
+    return result
+
+
+cdef ndarray[int64_t] _rounddown_int64(values, int64_t unit):
+    cdef:
+        Py_ssize_t i, n = len(values)
+        ndarray[int64_t] result = np.empty(n, dtype="i8")
+        int64_t res, value, remainder, half
+
+    half = unit // 2
+
+    with cython.overflowcheck(True):
+        for i in range(n):
+            value = values[i]
+
+            if value == NPY_NAT:
+                res = NPY_NAT
+            else:
+                # This adjustment is the only difference between rounddown_int64
+                #  and _ceil_int64
+                value = value - half
+                remainder = value % unit
+                if remainder == 0:
+                    res = value
+                else:
+                    res = value + (unit - remainder)
+
+            result[i] = res
+    return result
+
+
+cdef ndarray[int64_t] _roundup_int64(values, int64_t unit):
+    return _floor_int64(values + unit // 2, unit)
+
+
+cdef ndarray[int64_t] _round_nearest_int64(const int64_t[:] values, int64_t unit):
+    cdef:
+        Py_ssize_t i, n = len(values)
+        ndarray[int64_t] result = np.empty(n, dtype="i8")
+        int64_t res, value, half, remainder, quotient
+
+    half = unit // 2
+
+    with cython.overflowcheck(True):
+        for i in range(n):
+            value = values[i]
+
+            if value == NPY_NAT:
+                res = NPY_NAT
+            else:
+                quotient, remainder = divmod(value, unit)
+                if remainder > half:
+                    res = value + (unit - remainder)
+                elif remainder == half and quotient % 2:
+                    res = value + (unit - remainder)
+                else:
+                    res = value - remainder
+
+            result[i] = res
+
+    return result
+
+
+def round_nsint64(values: np.ndarray, mode: RoundTo, nanos: int) -> np.ndarray:
+    """
+    Applies rounding mode at given frequency
+
+    Parameters
+    ----------
+    values : np.ndarray[int64_t]`
+    mode : instance of `RoundTo` enumeration
+    nanos : np.int64
+        Freq to round to, expressed in nanoseconds
+
+    Returns
+    -------
+    np.ndarray[int64_t]
+    """
+    cdef:
+        int64_t unit = nanos
+
+    if mode == RoundTo.MINUS_INFTY:
+        return _floor_int64(values, unit)
+    elif mode == RoundTo.PLUS_INFTY:
+        return _ceil_int64(values, unit)
+    elif mode == RoundTo.NEAREST_HALF_MINUS_INFTY:
+        return _rounddown_int64(values, unit)
+    elif mode == RoundTo.NEAREST_HALF_PLUS_INFTY:
+        return _roundup_int64(values, unit)
+    elif mode == RoundTo.NEAREST_HALF_EVEN:
+        # for odd unit there is no need of a tie break
+        if unit % 2:
+            return _rounddown_int64(values, unit)
+        return _round_nearest_int64(values, unit)
+
+    # if/elif above should catch all rounding modes defined in enum 'RoundTo':
+    # if flow of control arrives here, it is a bug
+    raise ValueError("round_nsint64 called with an unrecognized rounding mode")
diff --git a/pandas/pandas/_libs/tslibs/meson.build b/pandas/pandas/_libs/tslibs/meson.build
new file mode 100644
index 0000000000000..052a8568b76af
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/meson.build
@@ -0,0 +1,64 @@
+tslibs_sources = {
+    # Dict of extension name -> dict of {sources, include_dirs, and deps}
+    # numpy include dir is implicitly included
+    'base': {'sources': ['base.pyx']},
+    'ccalendar': {'sources': ['ccalendar.pyx']},
+    'dtypes': {'sources': ['dtypes.pyx']},
+    'conversion': {'sources': ['conversion.pyx']},
+    'fields': {'sources': ['fields.pyx']},
+    'nattype': {'sources': ['nattype.pyx']},
+    'np_datetime': {'sources': ['np_datetime.pyx']},
+    'offsets': {'sources': ['offsets.pyx']},
+    'parsing': {'sources': ['parsing.pyx', '../src/parser/tokenizer.c']},
+    'period': {'sources': ['period.pyx']},
+    'strptime': {'sources': ['strptime.pyx']},
+    'timedeltas': {'sources': ['timedeltas.pyx']},
+    'timestamps': {'sources': ['timestamps.pyx']},
+    'timezones': {'sources': ['timezones.pyx']},
+    'tzconversion': {'sources': ['tzconversion.pyx']},
+    'vectorized': {'sources': ['vectorized.pyx']},
+}
+
+cython_args = [
+    '--include-dir',
+    meson.current_build_dir(),
+    '-X always_allow_keywords=true',
+]
+if get_option('buildtype') == 'debug'
+    cython_args += ['--gdb']
+endif
+
+foreach ext_name, ext_dict : tslibs_sources
+    py.extension_module(
+        ext_name,
+        ext_dict.get('sources'),
+        cython_args: cython_args,
+        include_directories: [inc_np, inc_pd],
+        dependencies: ext_dict.get('deps', ''),
+        subdir: 'pandas/_libs/tslibs',
+        install: true,
+    )
+endforeach
+
+sources_to_install = [
+    '__init__.py',
+    'ccalendar.pyi',
+    'conversion.pyi',
+    'dtypes.pyi',
+    'fields.pyi',
+    'nattype.pyi',
+    'np_datetime.pyi',
+    'offsets.pyi',
+    'parsing.pyi',
+    'period.pyi',
+    'strptime.pyi',
+    'timedeltas.pyi',
+    'timestamps.pyi',
+    'timezones.pyi',
+    'tzconversion.pyi',
+    'vectorized.pyi',
+]
+
+foreach source : sources_to_install
+    py.install_sources(source, subdir: 'pandas/_libs/tslibs')
+endforeach
diff --git a/pandas/pandas/_libs/tslibs/nattype.pxd b/pandas/pandas/_libs/tslibs/nattype.pxd
new file mode 100644
index 0000000000000..32705aa633135
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/nattype.pxd
@@ -0,0 +1,18 @@
+from cpython.datetime cimport datetime
+from numpy cimport int64_t
+
+
+cdef int64_t NPY_NAT
+
+cdef set c_nat_strings
+
+cdef class _NaT(datetime):
+    cdef readonly:
+        int64_t _value
+
+cdef _NaT c_NaT
+
+
+cdef bint checknull_with_nat(object val)
+cdef bint is_dt64nat(object val)
+cdef bint is_td64nat(object val)
diff --git a/pandas/pandas/_libs/tslibs/nattype.pyi b/pandas/pandas/_libs/tslibs/nattype.pyi
new file mode 100644
index 0000000000000..d3b10fbe79cb9
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/nattype.pyi
@@ -0,0 +1,186 @@
+from datetime import (
+    date as date_,
+    datetime,
+    time as time_,
+    timedelta,
+    tzinfo as _tzinfo,
+)
+from typing import (
+    Literal,
+    NoReturn,
+    TypeAlias,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs.period import Period
+from pandas._typing import (
+    Frequency,
+    Self,
+    TimestampNonexistent,
+)
+
+NaT: NaTType
+iNaT: int
+nat_strings: set[str]
+
+_TimeLike: TypeAlias = datetime | timedelta | Period | np.datetime64 | np.timedelta64
+_TimeDelta: TypeAlias = timedelta | np.timedelta64
+
+class NaTType:
+    _value: np.int64
+    @property
+    def value(self) -> int: ...
+    @property
+    def asm8(self) -> np.datetime64: ...
+    def to_datetime64(self) -> np.datetime64: ...
+    def to_numpy(
+        self, dtype: np.dtype | str | None = ..., copy: bool = ...
+    ) -> np.datetime64 | np.timedelta64: ...
+    @property
+    def is_leap_year(self) -> bool: ...
+    @property
+    def is_month_start(self) -> bool: ...
+    @property
+    def is_quarter_start(self) -> bool: ...
+    @property
+    def is_year_start(self) -> bool: ...
+    @property
+    def is_month_end(self) -> bool: ...
+    @property
+    def is_quarter_end(self) -> bool: ...
+    @property
+    def is_year_end(self) -> bool: ...
+    @property
+    def day_of_year(self) -> float: ...
+    @property
+    def dayofyear(self) -> float: ...
+    @property
+    def days_in_month(self) -> float: ...
+    @property
+    def daysinmonth(self) -> float: ...
+    @property
+    def day_of_week(self) -> float: ...
+    @property
+    def dayofweek(self) -> float: ...
+    @property
+    def week(self) -> float: ...
+    @property
+    def weekofyear(self) -> float: ...
+    @property
+    def fold(self) -> int: ...
+    def day_name(self) -> float: ...
+    def month_name(self) -> float: ...
+    def weekday(self) -> float: ...
+    def isoweekday(self) -> float: ...
+    def isoformat(self, sep: str = ..., timespec: str = ...) -> str: ...
+    def strftime(self, format: str) -> NoReturn: ...
+    def total_seconds(self) -> float: ...
+    def today(self, *args, **kwargs) -> NaTType: ...
+    def now(self, *args, **kwargs) -> NaTType: ...
+    def to_pydatetime(self) -> NaTType: ...
+    def date(self) -> NaTType: ...
+    def round(
+        self,
+        freq: Frequency,
+        ambiguous: bool | Literal["raise"] | NaTType = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> NaTType: ...
+    def floor(
+        self,
+        freq: Frequency,
+        ambiguous: bool | Literal["raise"] | NaTType = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> NaTType: ...
+    def ceil(
+        self,
+        freq: Frequency,
+        ambiguous: bool | Literal["raise"] | NaTType = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> NaTType: ...
+    def combine(cls, date: date_, time: time_) -> NoReturn: ...
+    @property
+    def tzinfo(self) -> None: ...
+    @property
+    def tz(self) -> None: ...
+    def tz_convert(self, tz: _tzinfo | str | None) -> NaTType: ...
+    def tz_localize(
+        self,
+        tz: _tzinfo | str | None,
+        ambiguous: bool | Literal["raise"] | NaTType = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> NaTType: ...
+    def replace(
+        self,
+        year: int | None = ...,
+        month: int | None = ...,
+        day: int | None = ...,
+        hour: int | None = ...,
+        minute: int | None = ...,
+        second: int | None = ...,
+        microsecond: int | None = ...,
+        nanosecond: int | None = ...,
+        tzinfo: _tzinfo | None = ...,
+        fold: int | None = ...,
+    ) -> NaTType: ...
+    @property
+    def year(self) -> float: ...
+    @property
+    def quarter(self) -> float: ...
+    @property
+    def month(self) -> float: ...
+    @property
+    def day(self) -> float: ...
+    @property
+    def hour(self) -> float: ...
+    @property
+    def minute(self) -> float: ...
+    @property
+    def second(self) -> float: ...
+    @property
+    def millisecond(self) -> float: ...
+    @property
+    def microsecond(self) -> float: ...
+    @property
+    def nanosecond(self) -> float: ...
+    # inject Timedelta properties
+    @property
+    def days(self) -> float: ...
+    @property
+    def seconds(self) -> float: ...
+    @property
+    def microseconds(self) -> float: ...
+    @property
+    def nanoseconds(self) -> float: ...
+    # inject Period properties
+    @property
+    def qyear(self) -> float: ...
+    # comparisons
+    def __eq__(self, other: object, /) -> Literal[False]: ...
+    def __ne__(self, other: object, /) -> Literal[True]: ...
+    def __lt__(self, other: Self | _TimeLike, /) -> Literal[False]: ...
+    def __le__(self, other: Self | _TimeLike, /) -> Literal[False]: ...
+    def __gt__(self, other: Self | _TimeLike, /) -> Literal[False]: ...
+    def __ge__(self, other: Self | _TimeLike, /) -> Literal[False]: ...
+    # unary operators
+    def __pos__(self) -> Self: ...
+    def __neg__(self) -> Self: ...
+    # binary operators
+    def __sub__(self, other: Self | _TimeLike, /) -> Self: ...
+    def __rsub__(self, other: Self | _TimeLike, /) -> Self: ...
+    def __add__(self, other: Self | _TimeLike, /) -> Self: ...
+    def __radd__(self, other: Self | _TimeLike, /) -> Self: ...
+    def __mul__(self, other: float, /) -> Self: ...  # analogous to timedelta
+    def __rmul__(self, other: float, /) -> Self: ...
+    @overload  # analogous to timedelta
+    def __truediv__(self, other: Self | _TimeDelta, /) -> float: ...  # Literal[NaN]
+    @overload
+    def __truediv__(self, other: float, /) -> Self: ...
+    @overload  # analogous to timedelta
+    def __floordiv__(self, other: Self | _TimeDelta, /) -> float: ...  # Literal[NaN]
+    @overload
+    def __floordiv__(self, other: float, /) -> Self: ...
+    # other
+    def __hash__(self) -> int: ...
+    def as_unit(self, unit: str, round_ok: bool = ...) -> NaTType: ...
diff --git a/pandas/pandas/_libs/tslibs/nattype.pyx b/pandas/pandas/_libs/tslibs/nattype.pyx
new file mode 100644
index 0000000000000..2657b1b9d197b
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/nattype.pyx
@@ -0,0 +1,1898 @@
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    PyDelta_Check,
+    datetime,
+    import_datetime,
+)
+
+import_datetime()
+from cpython.object cimport (
+    Py_EQ,
+    Py_NE,
+    PyObject_RichCompare,
+)
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport int64_t
+
+cnp.import_array()
+
+cimport pandas._libs.tslibs.util as util
+
+# ----------------------------------------------------------------------
+# Constants
+nat_strings = {"NaT", "nat", "NAT", "nan", "NaN", "NAN"}
+cdef set c_nat_strings = nat_strings
+
+cdef int64_t NPY_NAT = util.get_nat()
+iNaT = NPY_NAT  # python-visible constant
+
+# ----------------------------------------------------------------------
+
+
+def _make_nan_func(func_name: str, doc: str):
+    def f(*args, **kwargs):
+        return np.nan
+    f.__name__ = func_name
+    f.__doc__ = doc
+    return f
+
+
+def _make_nat_func(func_name: str, doc: str):
+    def f(*args, **kwargs):
+        return c_NaT
+    f.__name__ = func_name
+    f.__doc__ = doc
+    return f
+
+
+def _make_error_func(func_name: str, cls):
+    def f(*args, **kwargs):
+        raise ValueError(f"NaTType does not support {func_name}")
+
+    f.__name__ = func_name
+    if isinstance(cls, str):
+        # passed the literal docstring directly
+        f.__doc__ = cls
+    elif cls is not None:
+        f.__doc__ = getattr(cls, func_name).__doc__
+    return f
+
+
+cdef _nat_divide_op(self, other):
+    if PyDelta_Check(other) or cnp.is_timedelta64_object(other) or other is c_NaT:
+        return np.nan
+    if util.is_integer_object(other) or util.is_float_object(other):
+        return c_NaT
+    return NotImplemented
+
+
+cdef _nat_rdivide_op(self, other):
+    if PyDelta_Check(other):
+        return np.nan
+    return NotImplemented
+
+
+def _nat_unpickle(*args):
+    # return constant defined in the module
+    return c_NaT
+
+# ----------------------------------------------------------------------
+
+
+cdef class _NaT(datetime):
+    # cdef readonly:
+    #    int64_t value
+
+    # higher than np.ndarray and np.matrix
+    __array_priority__ = 100
+
+    def __richcmp__(_NaT self, object other, int op):
+        if cnp.is_datetime64_object(other) or PyDateTime_Check(other):
+            # We treat NaT as datetime-like for this comparison
+            return op == Py_NE
+
+        elif cnp.is_timedelta64_object(other) or PyDelta_Check(other):
+            # We treat NaT as timedelta-like for this comparison
+            return op == Py_NE
+
+        elif util.is_array(other):
+            if other.dtype.kind in "mM":
+                result = np.empty(other.shape, dtype=np.bool_)
+                result.fill(op == Py_NE)
+            elif other.dtype.kind == "O":
+                result = np.array([PyObject_RichCompare(self, x, op) for x in other])
+            elif op == Py_EQ:
+                result = np.zeros(other.shape, dtype=bool)
+            elif op == Py_NE:
+                result = np.ones(other.shape, dtype=bool)
+            else:
+                return NotImplemented
+            return result
+
+        elif PyDate_Check(other):
+            # GH#39151 don't defer to datetime.date object
+            if op == Py_EQ:
+                return False
+            if op == Py_NE:
+                return True
+            raise TypeError("Cannot compare NaT with datetime.date object")
+
+        return NotImplemented
+
+    def __add__(self, other):
+        if PyDateTime_Check(other):
+            return c_NaT
+        elif PyDelta_Check(other):
+            return c_NaT
+        elif cnp.is_datetime64_object(other) or cnp.is_timedelta64_object(other):
+            return c_NaT
+
+        elif util.is_integer_object(other):
+            # For Period compat
+            return c_NaT
+
+        elif util.is_array(other):
+            if other.dtype.kind in "mM":
+                # If we are adding to datetime64, we treat NaT as timedelta
+                #  Either way, result dtype is datetime64
+                result = np.empty(other.shape, dtype="datetime64[ns]")
+                result.fill("NaT")
+                return result
+            raise TypeError(f"Cannot add NaT to ndarray with dtype {other.dtype}")
+
+        # Includes Period, DateOffset going through here
+        return NotImplemented
+
+    def __radd__(self, other):
+        return self.__add__(other)
+
+    def __sub__(self, other):
+        # Duplicate some logic from _Timestamp.__sub__ to avoid needing
+        # to subclass; allows us to @final(_Timestamp.__sub__)
+
+        if PyDateTime_Check(other):
+            return c_NaT
+        elif PyDelta_Check(other):
+            return c_NaT
+        elif cnp.is_datetime64_object(other) or cnp.is_timedelta64_object(other):
+            return c_NaT
+
+        elif util.is_integer_object(other):
+            # For Period compat
+            return c_NaT
+
+        elif util.is_array(other):
+            if other.dtype.kind == "m":
+                # NaT - timedelta64 we treat NaT as datetime64, so result
+                #  is datetime64
+                result = np.empty(other.shape, dtype="datetime64[ns]")
+                result.fill("NaT")
+                return result
+
+            elif other.dtype.kind == "M":
+                # We treat NaT as a datetime, so regardless of whether this is
+                #  NaT - other or other - NaT, the result is timedelta64
+                result = np.empty(other.shape, dtype="timedelta64[ns]")
+                result.fill("NaT")
+                return result
+
+            raise TypeError(
+                f"Cannot subtract NaT from ndarray with dtype {other.dtype}"
+            )
+
+        # Includes Period, DateOffset going through here
+        return NotImplemented
+
+    def __rsub__(self, other):
+        if util.is_array(other):
+            if other.dtype.kind == "m":
+                # timedelta64 - NaT we have to treat NaT as timedelta64
+                #  for this to be meaningful, and the result is timedelta64
+                result = np.empty(other.shape, dtype="timedelta64[ns]")
+                result.fill("NaT")
+                return result
+
+            elif other.dtype.kind == "M":
+                # We treat NaT as a datetime, so regardless of whether this is
+                #  NaT - other or other - NaT, the result is timedelta64
+                result = np.empty(other.shape, dtype="timedelta64[ns]")
+                result.fill("NaT")
+                return result
+        # other cases are same, swap operands is allowed even though we subtract
+        # because this is NaT
+        return self.__sub__(other)
+
+    def __pos__(self):
+        return NaT
+
+    def __neg__(self):
+        return NaT
+
+    def __truediv__(self, other):
+        return _nat_divide_op(self, other)
+
+    def __floordiv__(self, other):
+        return _nat_divide_op(self, other)
+
+    def __mul__(self, other):
+        if util.is_integer_object(other) or util.is_float_object(other):
+            return NaT
+        return NotImplemented
+
+    @property
+    def asm8(self) -> np.datetime64:
+        return np.datetime64(NPY_NAT, "ns")
+
+    def to_datetime64(self) -> np.datetime64:
+        """
+        Return a NumPy datetime64 object with same precision.
+
+        This method returns a numpy.datetime64 object with the same
+        date and time information and precision as the pd.Timestamp object.
+
+        See Also
+        --------
+        numpy.datetime64 : Class to represent dates and times with high precision.
+        Timestamp.to_numpy : Alias for this method.
+        Timestamp.asm8 : Alias for this method.
+        pd.to_datetime : Convert argument to datetime.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(year=2023, month=1, day=1,
+        ...                   hour=10, second=15)
+        >>> ts
+        Timestamp('2023-01-01 10:00:15')
+        >>> ts.to_datetime64()
+        numpy.datetime64('2023-01-01T10:00:15.000000')
+        """
+        return np.datetime64("NaT", "ns")
+
+    def to_numpy(self, dtype=None, copy=False) -> np.datetime64 | np.timedelta64:
+        """
+        Convert the Timestamp to a NumPy datetime64.
+
+        This is an alias method for `Timestamp.to_datetime64()`. The dtype and
+        copy parameters are available here only for compatibility. Their values
+        will not affect the return value.
+
+        Parameters
+        ----------
+        dtype : dtype, optional
+            Data type of the output, ignored in this method as the return type
+            is always `numpy.datetime64`.
+        copy : bool, default False
+            Whether to ensure that the returned value is a new object. This
+            parameter is also ignored as the method does not support copying.
+
+        Returns
+        -------
+        numpy.datetime64
+
+        See Also
+        --------
+        DatetimeIndex.to_numpy : Similar method for DatetimeIndex.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.to_numpy()
+        numpy.datetime64('2020-03-14T15:32:52.192548651')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.to_numpy()
+        numpy.datetime64('NaT')
+        """
+        if dtype is not None:
+            # GH#44460
+            dtype = np.dtype(dtype)
+            if dtype.kind == "M":
+                return np.datetime64("NaT").astype(dtype)
+            elif dtype.kind == "m":
+                return np.timedelta64("NaT").astype(dtype)
+            else:
+                raise ValueError(
+                    "NaT.to_numpy dtype must be a datetime64 dtype, timedelta64 "
+                    "dtype, or None."
+                )
+        return self.to_datetime64()
+
+    def __repr__(self) -> str:
+        return "NaT"
+
+    def __str__(self) -> str:
+        return "NaT"
+
+    def isoformat(self, sep: str = "T", timespec: str = "auto") -> str:
+        # This allows Timestamp(ts.isoformat()) to always correctly roundtrip.
+        return "NaT"
+
+    def __hash__(self) -> int:
+        return NPY_NAT
+
+    @property
+    def is_leap_year(self) -> bool:
+        return False
+
+    @property
+    def is_month_start(self) -> bool:
+        return False
+
+    @property
+    def is_quarter_start(self) -> bool:
+        return False
+
+    @property
+    def is_year_start(self) -> bool:
+        return False
+
+    @property
+    def is_month_end(self) -> bool:
+        return False
+
+    @property
+    def is_quarter_end(self) -> bool:
+        return False
+
+    @property
+    def is_year_end(self) -> bool:
+        return False
+
+
+class NaTType(_NaT):
+    """
+    (N)ot-(A)-(T)ime, the time equivalent of NaN.
+
+    NaT is used to denote missing or null values in datetime and timedelta objects
+    in pandas. It functions similarly to how NaN is used for numerical data.
+    Operations with NaT will generally propagate the NaT value, similar to NaN.
+    NaT can be used in pandas data structures like Series and DataFrame
+    to represent missing datetime values. It is useful in data analysis
+    and time series analysis when working with incomplete or sparse
+    time-based data. Pandas provides robust handling of NaT to ensure
+    consistency and reliability in computations involving datetime objects.
+
+    See Also
+    --------
+    NA : NA ("not available") missing value indicator.
+    isna : Detect missing values (NaN or NaT) in an array-like object.
+    notna : Detect non-missing values.
+    numpy.nan : Floating point representation of Not a Number (NaN) for numerical data.
+
+    Examples
+    --------
+    >>> pd.DataFrame([pd.Timestamp("2023"), np.nan], columns=["col_1"])
+            col_1
+    0  2023-01-01
+    1         NaT
+    """
+
+    def __new__(cls):
+        cdef _NaT base
+
+        base = _NaT.__new__(cls, 1, 1, 1)
+        base._value= NPY_NAT
+
+        return base
+
+    @property
+    def value(self) -> int:
+        return self._value
+
+    def __reduce_ex__(self, protocol):
+        # python 3.6 compat
+        # https://bugs.python.org/issue28730
+        # now __reduce_ex__ is defined and higher priority than __reduce__
+        return self.__reduce__()
+
+    def __reduce__(self):
+        return (_nat_unpickle, (None, ))
+
+    def __rtruediv__(self, other):
+        return _nat_rdivide_op(self, other)
+
+    def __rfloordiv__(self, other):
+        return _nat_rdivide_op(self, other)
+
+    def __rmul__(self, other):
+        if util.is_integer_object(other) or util.is_float_object(other):
+            return c_NaT
+        return NotImplemented
+
+    # ----------------------------------------------------------------------
+    # inject the Timestamp field properties
+    # these by definition return np.nan
+
+    year = property(fget=lambda self: np.nan)
+    quarter = property(fget=lambda self: np.nan)
+    month = property(fget=lambda self: np.nan)
+    day = property(fget=lambda self: np.nan)
+    hour = property(fget=lambda self: np.nan)
+    minute = property(fget=lambda self: np.nan)
+    second = property(fget=lambda self: np.nan)
+    millisecond = property(fget=lambda self: np.nan)
+    microsecond = property(fget=lambda self: np.nan)
+    nanosecond = property(fget=lambda self: np.nan)
+
+    week = property(fget=lambda self: np.nan)
+    dayofyear = property(fget=lambda self: np.nan)
+    day_of_year = property(fget=lambda self: np.nan)
+    weekofyear = property(fget=lambda self: np.nan)
+    days_in_month = property(fget=lambda self: np.nan)
+    daysinmonth = property(fget=lambda self: np.nan)
+    dayofweek = property(fget=lambda self: np.nan)
+    day_of_week = property(fget=lambda self: np.nan)
+
+    # inject Timedelta properties
+    days = property(fget=lambda self: np.nan)
+    seconds = property(fget=lambda self: np.nan)
+    microseconds = property(fget=lambda self: np.nan)
+    nanoseconds = property(fget=lambda self: np.nan)
+
+    # inject pd.Period properties
+    qyear = property(fget=lambda self: np.nan)
+
+    # ----------------------------------------------------------------------
+    # GH9513 NaT methods (except to_datetime64) to raise, return np.nan, or
+    # return NaT create functions that raise, for binding to NaTType
+    # These are the ones that can get their docstrings from datetime.
+
+    # nan methods
+    weekday = _make_nan_func(
+        "weekday",
+        """
+        Return the day of the week represented by the date.
+
+        Monday == 0 ... Sunday == 6.
+
+        See Also
+        --------
+        Timestamp.dayofweek : Return the day of the week with Monday=0, Sunday=6.
+        Timestamp.isoweekday : Return the day of the week with Monday=1, Sunday=7.
+        datetime.date.weekday : Equivalent method in datetime module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01')
+        >>> ts
+        Timestamp('2023-01-01  00:00:00')
+        >>> ts.weekday()
+        6
+        """,
+    )
+    isoweekday = _make_nan_func(
+        "isoweekday",
+        """
+        Return the day of the week represented by the date.
+
+        Monday == 1 ... Sunday == 7.
+
+        See Also
+        --------
+        Timestamp.weekday : Return the day of the week with Monday=0, Sunday=6.
+        Timestamp.isocalendar : Return a tuple containing ISO year, week number
+            and weekday.
+        datetime.date.isoweekday : Equivalent method in datetime module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.isoweekday()
+        7
+        """,
+    )
+    total_seconds = _make_nan_func(
+        "total_seconds",
+        """
+        Total seconds in the duration.
+
+        This method calculates the total duration in seconds by combining
+        the days, seconds, and microseconds of the `Timedelta` object.
+
+        See Also
+        --------
+        to_timedelta : Convert argument to timedelta.
+        Timedelta : Represents a duration, the difference between two dates or times.
+        Timedelta.seconds : Returns the seconds component of the timedelta.
+        Timedelta.microseconds : Returns the microseconds component of the timedelta.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1min')
+        >>> td
+        Timedelta('0 days 00:01:00')
+        >>> td.total_seconds()
+        60.0
+        """,
+    )
+    month_name = _make_nan_func(
+        "month_name",
+        """
+        Return the month name of the Timestamp with specified locale.
+
+        This method returns the full name of the month corresponding to the
+        `Timestamp`, such as 'January', 'February', etc. The month name can
+        be returned in a specified locale if provided; otherwise, it defaults
+        to the English locale.
+
+        Parameters
+        ----------
+        locale : str, default None (English locale)
+            Locale determining the language in which to return the month name.
+
+        Returns
+        -------
+        str
+            The full month name as a string.
+
+        See Also
+        --------
+        Timestamp.day_name : Returns the name of the day of the week.
+        Timestamp.strftime : Returns a formatted string of the Timestamp.
+        datetime.datetime.strftime : Returns a string representing the date and time.
+
+        Examples
+        --------
+        Get the month name in English (default):
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.month_name()
+        'March'
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.month_name()
+        nan
+        """,
+    )
+    day_name = _make_nan_func(
+        "day_name",
+        """
+        Return the day name of the Timestamp with specified locale.
+
+        Parameters
+        ----------
+        locale : str, default None (English locale)
+            Locale determining the language in which to return the day name.
+
+        Returns
+        -------
+        str
+
+        See Also
+        --------
+        Timestamp.day_of_week : Return day of the week.
+        Timestamp.day_of_year : Return day of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.day_name()
+        'Saturday'
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.day_name()
+        nan
+        """,
+    )
+    # _nat_methods
+
+    # "fromisocalendar" was introduced in 3.8
+    fromisocalendar = _make_error_func("fromisocalendar", datetime)
+
+    # ----------------------------------------------------------------------
+    # The remaining methods have docstrings copy/pasted from the analogous
+    # Timestamp methods.
+    isocalendar = _make_error_func(
+        "isocalendar",
+        """
+        Return a named tuple containing ISO year, week number, and weekday.
+
+        See Also
+        --------
+        DatetimeIndex.isocalendar : Return a 3-tuple containing ISO year,
+            week number, and weekday for the given DatetimeIndex object.
+        datetime.date.isocalendar : The equivalent method for `datetime.date` objects.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.isocalendar()
+        datetime.IsoCalendarDate(year=2022, week=52, weekday=7)
+        """
+        )
+    dst = _make_error_func(
+        "dst",
+        """
+        Return the daylight saving time (DST) adjustment.
+
+        This method returns the DST adjustment as a `datetime.timedelta` object
+        if the Timestamp is timezone-aware and DST is applicable.
+
+        See Also
+        --------
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2000-06-01 00:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2000-06-01 00:00:00+0200', tz='Europe/Brussels')
+        >>> ts.dst()
+        datetime.timedelta(seconds=3600)
+        """
+        )
+    date = _make_nat_func(
+        "date",
+        """
+        Returns `datetime.date` with the same year, month, and day.
+
+        This method extracts the date component from the `Timestamp` and returns
+        it as a `datetime.date` object, discarding the time information.
+
+        Returns
+        -------
+        datetime.date
+            The date part of the `Timestamp`.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        datetime.datetime.date : Extract the date component from a `datetime` object.
+
+        Examples
+        --------
+        Extract the date from a Timestamp:
+
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00.00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.date()
+        datetime.date(2023, 1, 1)
+        """
+        )
+    utctimetuple = _make_error_func(
+        "utctimetuple",
+        """
+        Return UTC time tuple, compatible with `time.localtime()`.
+
+        This method converts the Timestamp to UTC and returns a time tuple
+        containing 9 components: year, month, day, hour, minute, second,
+        weekday, day of year, and DST flag. This is particularly useful for
+        converting a Timestamp to a format compatible with time module functions.
+
+        Returns
+        -------
+        time.struct_time
+            A time.struct_time object representing the UTC time.
+
+        See Also
+        --------
+        datetime.datetime.utctimetuple :
+            Return UTC time tuple, compatible with time.localtime().
+        Timestamp.timetuple : Return time tuple of local time.
+        time.struct_time : Time tuple structure used by time functions.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.utctimetuple()
+        time.struct_time(tm_year=2023, tm_mon=1, tm_mday=1, tm_hour=9,
+        tm_min=0, tm_sec=0, tm_wday=6, tm_yday=1, tm_isdst=0)
+        """
+        )
+    utcoffset = _make_error_func(
+        "utcoffset",
+        """
+        Return utc offset.
+
+        This method returns the difference between UTC and the local time
+        as a `timedelta` object. It is useful for understanding the time
+        difference between the current timezone and UTC.
+
+        Returns
+        -------
+        timedelta
+            The difference between UTC and the local time as a `timedelta` object.
+
+        See Also
+        --------
+        datetime.datetime.utcoffset :
+            Standard library method to get the UTC offset of a datetime object.
+        Timestamp.tzname : Return the name of the timezone.
+        Timestamp.dst : Return the daylight saving time (DST) adjustment.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.utcoffset()
+        datetime.timedelta(seconds=3600)
+        """
+        )
+    tzname = _make_error_func(
+        "tzname",
+        """
+        Return time zone name.
+
+        This method returns the name of the Timestamp's time zone as a string.
+
+        See Also
+        --------
+        Timestamp.tzinfo : Returns the timezone information of the Timestamp.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.tzname()
+        'CET'
+        """
+        )
+    time = _make_error_func(
+        "time",
+        """
+        Return time object with same time but with tzinfo=None.
+
+        This method extracts the time part of the `Timestamp` object, excluding any
+        timezone information. It returns a `datetime.time` object which only represents
+        the time (hours, minutes, seconds, and microseconds).
+
+        See Also
+        --------
+        Timestamp.date : Return date object with same year, month and day.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.time()
+        datetime.time(10, 0)
+        """,
+        )
+    timetuple = _make_error_func(
+        "timetuple",
+        """
+        Return time tuple, compatible with time.localtime().
+
+        This method converts the `Timestamp` into a time tuple, which is compatible
+        with functions like `time.localtime()`. The time tuple is a named tuple with
+        attributes such as year, month, day, hour, minute, second, weekday,
+        day of the year, and daylight savings indicator.
+
+        See Also
+        --------
+        time.localtime : Converts a POSIX timestamp into a time tuple.
+        Timestamp : The `Timestamp` that represents a specific point in time.
+        datetime.datetime.timetuple : Equivalent method in the `datetime` module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.timetuple()
+        time.struct_time(tm_year=2023, tm_mon=1, tm_mday=1,
+        tm_hour=10, tm_min=0, tm_sec=0, tm_wday=6, tm_yday=1, tm_isdst=-1)
+        """
+        )
+    timetz = _make_error_func(
+        "timetz",
+        """
+        Return time object with same time and tzinfo.
+
+        This method returns a datetime.time object with
+        the time and tzinfo corresponding to the pd.Timestamp
+        object, ignoring any information about the day/date.
+
+        See Also
+        --------
+        datetime.datetime.timetz : Return datetime.time object with the
+            same time attributes as the datetime object.
+        datetime.time : Class to represent the time of day, independent
+            of any particular day.
+        datetime.datetime.tzinfo : Attribute of datetime.datetime objects
+            representing the timezone of the datetime object.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.timetz()
+        datetime.time(10, 0, tzinfo=<DstTzInfo 'Europe/Brussels' CET+1:00:00 STD>)
+        """
+        )
+    toordinal = _make_error_func(
+        "toordinal",
+        """
+        Return proleptic Gregorian ordinal. January 1 of year 1 is day 1.
+
+        The proleptic Gregorian ordinal is a continuous count of days since
+        January 1 of year 1, which is considered day 1. This method converts
+        the `Timestamp` to its equivalent ordinal number, useful for date arithmetic
+        and comparison operations.
+
+        See Also
+        --------
+        datetime.datetime.toordinal : Equivalent method in the `datetime` module.
+        Timestamp : The `Timestamp` that represents a specific point in time.
+        Timestamp.fromordinal : Create a `Timestamp` from an ordinal.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:50')
+        >>> ts
+        Timestamp('2023-01-01 10:00:50')
+        >>> ts.toordinal()
+        738521
+        """
+        )
+    ctime = _make_error_func(
+        "ctime",
+        """
+        Return a ctime() style string representing the Timestamp.
+
+        This method returns a string representing the date and time
+        in the format returned by the standard library's `time.ctime()`
+        function, which is typically in the form 'Day Mon DD HH:MM:SS YYYY'.
+
+        If the `Timestamp` is outside the range supported by Python's
+        standard library, a `NotImplementedError` is raised.
+
+        Returns
+        -------
+        str
+            A string representing the Timestamp in ctime format.
+
+        See Also
+        --------
+        time.ctime : Return a string representing time in ctime format.
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        datetime.datetime.ctime : Return a ctime style string from a datetime object.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00.00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.ctime()
+        'Sun Jan  1 10:00:00 2023'
+        """,
+    )
+
+    strftime = _make_error_func(
+        "strftime",
+        """
+        Return a formatted string of the Timestamp.
+
+        Parameters
+        ----------
+        format : str
+            Format string to convert Timestamp to string.
+            See strftime documentation for more information on the format string:
+            https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior.
+
+        See Also
+        --------
+        Timestamp.isoformat : Return the time formatted according to ISO 8601.
+        pd.to_datetime : Convert argument to datetime.
+        Period.strftime : Format a single Period.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.strftime('%Y-%m-%d %X')
+        '2020-03-14 15:32:52'
+        """,
+    )
+
+    strptime = _make_error_func(
+        "strptime",
+        """
+        Convert string argument to datetime.
+
+        This method is not implemented; calling it will raise NotImplementedError.
+        Use pd.to_datetime() instead.
+
+        Parameters
+        ----------
+        date_string : str
+            String to convert to a datetime.
+        format : str, default None
+            The format string to parse time, e.g. "%d/%m/%Y".
+
+        See Also
+        --------
+        pd.to_datetime : Convert argument to datetime.
+        datetime.datetime.strptime : Return a datetime corresponding to a string
+            representing a date and time, parsed according to a separate
+            format string.
+        datetime.datetime.strftime : Return a string representing the date and
+            time, controlled by an explicit format string.
+        Timestamp.isoformat : Return the time formatted according to ISO 8601.
+
+        Examples
+        --------
+        >>> pd.Timestamp.strptime("2023-01-01", "%d/%m/%y")
+        Traceback (most recent call last):
+        NotImplementedError
+        """,
+    )
+
+    utcfromtimestamp = _make_error_func(
+        "utcfromtimestamp",
+        """
+        Timestamp.utcfromtimestamp(ts)
+
+        Construct a timezone-aware UTC datetime from a POSIX timestamp.
+
+        This method creates a datetime object from a POSIX timestamp, keeping the
+        Timestamp object's timezone.
+
+        Parameters
+        ----------
+        ts : float
+            POSIX timestamp.
+
+        See Also
+        --------
+        Timezone.tzname : Return time zone name.
+        Timestamp.utcnow : Return a new Timestamp representing UTC day and time.
+        Timestamp.fromtimestamp : Transform timestamp[, tz] to tz's local
+            time from POSIX timestamp.
+
+        Notes
+        -----
+        Timestamp.utcfromtimestamp behavior differs from datetime.utcfromtimestamp
+        in returning a timezone-aware object.
+
+        Examples
+        --------
+        >>> pd.Timestamp.utcfromtimestamp(1584199972)
+        Timestamp('2020-03-14 15:32:52+0000', tz='UTC')
+        """,
+    )
+    fromtimestamp = _make_error_func(
+        "fromtimestamp",
+        """
+        Create a `Timestamp` object from a POSIX timestamp.
+
+        This method converts a POSIX timestamp (the number of seconds since
+        January 1, 1970, 00:00:00 UTC) into a `Timestamp` object. The resulting
+        `Timestamp` can be localized to a specific time zone if provided.
+
+        Parameters
+        ----------
+        ts : float
+            The POSIX timestamp to convert, representing seconds since
+            the epoch (1970-01-01 00:00:00 UTC).
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile, optional
+            Time zone for the `Timestamp`. If not provided, the `Timestamp` will
+            be timezone-naive (i.e., without time zone information).
+
+        Returns
+        -------
+        Timestamp
+            A `Timestamp` object representing the given POSIX timestamp.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+        datetime.datetime.fromtimestamp : Returns a datetime from a POSIX timestamp.
+
+        Examples
+        --------
+        Convert a POSIX timestamp to a `Timestamp`:
+
+        >>> pd.Timestamp.fromtimestamp(1584199972)  # doctest: +SKIP
+        Timestamp('2020-03-14 15:32:52')
+
+        Note that the output may change depending on your local time and time zone:
+
+        >>> pd.Timestamp.fromtimestamp(1584199972, tz='UTC')  # doctest: +SKIP
+        Timestamp('2020-03-14 15:32:52+0000', tz='UTC')
+        """,
+    )
+    combine = _make_error_func(
+        "combine",
+        """
+        Timestamp.combine(date, time)
+
+        Combine a date and time into a single Timestamp object.
+
+        This method takes a `date` object and a `time` object
+        and combines them into a single `Timestamp`
+        that has the same date and time fields.
+
+        Parameters
+        ----------
+        date : datetime.date
+            The date part of the Timestamp.
+        time : datetime.time
+            The time part of the Timestamp.
+
+        Returns
+        -------
+        Timestamp
+            A new `Timestamp` object representing the combined date and time.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Examples
+        --------
+        >>> from datetime import date, time
+        >>> pd.Timestamp.combine(date(2020, 3, 14), time(15, 30, 15))
+        Timestamp('2020-03-14 15:30:15')
+        """,
+    )
+    utcnow = _make_error_func(
+        "utcnow",
+        """
+        Timestamp.utcnow()
+
+        Return a new Timestamp representing UTC day and time.
+
+        See Also
+        --------
+        Timestamp : Constructs an arbitrary datetime.
+        Timestamp.now : Return the current local date and time, which
+            can be timezone-aware.
+        Timestamp.today : Return the current local date and time with
+            timezone information set to None.
+        to_datetime : Convert argument to datetime.
+        date_range : Return a fixed frequency DatetimeIndex.
+        Timestamp.utctimetuple : Return UTC time tuple, compatible with
+            time.localtime().
+
+        Examples
+        --------
+        >>> pd.Timestamp.utcnow()   # doctest: +SKIP
+        Timestamp('2020-11-16 22:50:18.092888+0000', tz='UTC')
+        """,
+    )
+
+    timestamp = _make_error_func(
+        "timestamp",
+        """
+        Return POSIX timestamp as float.
+
+        This method converts the `Timestamp` object to a POSIX timestamp, which is
+        the number of seconds since the Unix epoch (January 1, 1970). The returned
+        value is a floating-point number, where the integer part represents the
+        seconds, and the fractional part represents the microseconds.
+
+        Returns
+        -------
+        float
+            The POSIX timestamp representation of the `Timestamp` object.
+
+        See Also
+        --------
+        Timestamp.fromtimestamp : Construct a `Timestamp` from a POSIX timestamp.
+        datetime.datetime.timestamp : Equivalent method from the `datetime` module.
+        Timestamp.to_pydatetime : Convert the `Timestamp` to a `datetime` object.
+        Timestamp.to_datetime64 : Converts `Timestamp` to `numpy.datetime64`.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548')
+        >>> ts.timestamp()
+        1584199972.192548
+        """
+    )
+
+    # GH9513 NaT methods (except to_datetime64) to raise, return np.nan, or
+    # return NaT create functions that raise, for binding to NaTType
+    astimezone = _make_error_func(
+        "astimezone",
+        """
+        Convert timezone-aware Timestamp to another time zone.
+
+        This method is used to convert a timezone-aware Timestamp object to a
+        different time zone. The original UTC time remains the same; only the
+        time zone information is changed. If the Timestamp is timezone-naive, a
+        TypeError is raised.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for time which Timestamp will be converted to.
+            None will remove timezone holding UTC time.
+
+        Returns
+        -------
+        converted : Timestamp
+
+        Raises
+        ------
+        TypeError
+            If Timestamp is tz-naive.
+
+        See Also
+        --------
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+        DatetimeIndex.tz_convert : Convert a DatetimeIndex to another time zone.
+        DatetimeIndex.tz_localize : Localize a DatetimeIndex to a specific time zone.
+        datetime.datetime.astimezone : Convert a datetime object to another time zone.
+
+        Examples
+        --------
+        Create a timestamp object with UTC timezone:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651', tz='UTC')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651+0000', tz='UTC')
+
+        Change to Tokyo timezone:
+
+        >>> ts.tz_convert(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Can also use ``astimezone``:
+
+        >>> ts.astimezone(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.tz_convert(tz='Asia/Tokyo')
+        NaT
+        """,
+    )
+    fromordinal = _make_error_func(
+        "fromordinal",
+        """
+        Construct a timestamp from a a proleptic Gregorian ordinal.
+
+        This method creates a `Timestamp` object corresponding to the given
+        proleptic Gregorian ordinal, which is a count of days from January 1,
+        0001 (using the proleptic Gregorian calendar). The time part of the
+        `Timestamp` is set to midnight (00:00:00) by default.
+
+        Parameters
+        ----------
+        ordinal : int
+            Date corresponding to a proleptic Gregorian ordinal.
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for the Timestamp.
+
+        Returns
+        -------
+        Timestamp
+            A `Timestamp` object representing the specified ordinal date.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Notes
+        -----
+        By definition there cannot be any tz info on the ordinal itself.
+
+        Examples
+        --------
+        Convert an ordinal to a `Timestamp`:
+
+        >>> pd.Timestamp.fromordinal(737425)
+        Timestamp('2020-01-01 00:00:00')
+
+        Create a `Timestamp` from an ordinal with timezone information:
+
+        >>> pd.Timestamp.fromordinal(737425, tz='UTC')
+        Timestamp('2020-01-01 00:00:00+0000', tz='UTC')
+        """,
+    )
+
+    # _nat_methods
+    to_pydatetime = _make_nat_func(
+        "to_pydatetime",
+        """
+        Convert a Timestamp object to a native Python datetime object.
+
+        This method is useful for when you need to utilize a pandas Timestamp
+        object in contexts where native Python datetime objects are expected
+        or required. The conversion discards the nanoseconds component, and a
+        warning can be issued in such cases if desired.
+
+        Parameters
+        ----------
+        warn : bool, default True
+            If True, issues a warning when the timestamp includes nonzero
+            nanoseconds, as these will be discarded during the conversion.
+
+        Returns
+        -------
+        datetime.datetime or NaT
+            Returns a datetime.datetime object representing the timestamp,
+            with year, month, day, hour, minute, second, and microsecond components.
+            If the timestamp is NaT (Not a Time), returns NaT.
+
+        See Also
+        --------
+        datetime.datetime : The standard Python datetime class that this method
+            returns.
+        Timestamp.timestamp : Convert a Timestamp object to POSIX timestamp.
+        Timestamp.to_datetime64 : Convert a Timestamp object to numpy.datetime64.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548')
+        >>> ts.to_pydatetime()
+        datetime.datetime(2020, 3, 14, 15, 32, 52, 192548)
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.to_pydatetime()
+        NaT
+        """,
+    )
+
+    now = _make_nat_func(
+        "now",
+        """
+        Return new Timestamp object representing current time local to tz.
+
+        This method returns a new `Timestamp` object that represents the current time.
+        If a timezone is provided, the current time will be localized to that timezone.
+        Otherwise, it returns the current local time.
+
+        Parameters
+        ----------
+        tz : str or timezone object, default None
+            Timezone to localize to.
+
+        See Also
+        --------
+        to_datetime : Convert argument to datetime.
+        Timestamp.utcnow : Return a new Timestamp representing UTC day and time.
+        Timestamp.today : Return the current time in the local timezone.
+
+        Examples
+        --------
+        >>> pd.Timestamp.now()  # doctest: +SKIP
+        Timestamp('2020-11-16 22:06:16.378782')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.now()
+        NaT
+        """,
+    )
+    today = _make_nat_func(
+        "today",
+        """
+        Return the current time in the local timezone.
+
+        This differs from datetime.today() in that it can be localized to a
+        passed timezone.
+
+        Parameters
+        ----------
+        tz : str or timezone object, default None
+            Timezone to localize to.
+
+        See Also
+        --------
+        datetime.datetime.today : Returns the current local date.
+        Timestamp.now : Returns current time with optional timezone.
+        Timestamp : A class representing a specific timestamp.
+
+        Examples
+        --------
+        >>> pd.Timestamp.today()    # doctest: +SKIP
+        Timestamp('2020-11-16 22:37:39.969883')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.today()
+        NaT
+        """,
+    )
+    round = _make_nat_func(
+        "round",
+        """
+        Round the Timestamp to the specified resolution.
+
+        This method rounds the given Timestamp down to a specified frequency
+        level. It is particularly useful in data analysis to normalize timestamps
+        to regular frequency intervals. For instance, rounding to the nearest
+        minute, hour, or day can help in time series comparisons or resampling
+        operations.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the rounding resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        a new Timestamp rounded to the given resolution of `freq`
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted
+
+        See Also
+        --------
+        datetime.round : Similar behavior in native Python datetime module.
+        Timestamp.floor : Round the Timestamp downward to the nearest multiple
+            of the specified frequency.
+        Timestamp.ceil : Round the Timestamp upward to the nearest multiple of
+            the specified frequency.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, rounding will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When rounding
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be rounded using multiple frequency units:
+
+        >>> ts.round(freq='h')  # hour
+        Timestamp('2020-03-14 16:00:00')
+
+        >>> ts.round(freq='min')  # minute
+        Timestamp('2020-03-14 15:33:00')
+
+        >>> ts.round(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:52')
+
+        >>> ts.round(freq='ms')  # milliseconds
+        Timestamp('2020-03-14 15:32:52.193000')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.round(freq='5min')
+        Timestamp('2020-03-14 15:35:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.round(freq='1h30min')
+        Timestamp('2020-03-14 15:00:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.round()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 01:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.round("h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.round("h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """,
+    )
+    floor = _make_nat_func(
+        "floor",
+        """
+        Return a new Timestamp floored to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the flooring resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted.
+
+        See Also
+        --------
+        Timestamp.ceil : Round up a Timestamp to the specified resolution.
+        Timestamp.round : Round a Timestamp to the specified resolution.
+        Series.dt.floor : Round down the datetime values in a Series.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, flooring will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When flooring
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be floored using multiple frequency units:
+
+        >>> ts.floor(freq='h')  # hour
+        Timestamp('2020-03-14 15:00:00')
+
+        >>> ts.floor(freq='min')  # minute
+        Timestamp('2020-03-14 15:32:00')
+
+        >>> ts.floor(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:52')
+
+        >>> ts.floor(freq='ns')  # nanoseconds
+        Timestamp('2020-03-14 15:32:52.192548651')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.floor(freq='5min')
+        Timestamp('2020-03-14 15:30:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.floor(freq='1h30min')
+        Timestamp('2020-03-14 15:00:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.floor()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 03:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.floor("2h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.floor("2h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """,
+    )
+    ceil = _make_nat_func(
+        "ceil",
+        """
+        Return a new Timestamp ceiled to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the ceiling resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted.
+
+        See Also
+        --------
+        Timestamp.floor : Round down a Timestamp to the specified resolution.
+        Timestamp.round : Round a Timestamp to the specified resolution.
+        Series.dt.ceil : Ceil the datetime values in a Series.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, ceiling will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When ceiling
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be ceiled using multiple frequency units:
+
+        >>> ts.ceil(freq='h')  # hour
+        Timestamp('2020-03-14 16:00:00')
+
+        >>> ts.ceil(freq='min')  # minute
+        Timestamp('2020-03-14 15:33:00')
+
+        >>> ts.ceil(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:53')
+
+        >>> ts.ceil(freq='us')  # microseconds
+        Timestamp('2020-03-14 15:32:52.192549')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.ceil(freq='5min')
+        Timestamp('2020-03-14 15:35:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.ceil(freq='1h30min')
+        Timestamp('2020-03-14 16:30:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.ceil()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 01:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.ceil("h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.ceil("h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """,
+    )
+
+    tz_convert = _make_nat_func(
+        "tz_convert",
+        """
+        Convert timezone-aware Timestamp to another time zone.
+
+        This method is used to convert a timezone-aware Timestamp object to a
+        different time zone. The original UTC time remains the same; only the
+        time zone information is changed. If the Timestamp is timezone-naive, a
+        TypeError is raised.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for time which Timestamp will be converted to.
+            None will remove timezone holding UTC time.
+
+        Returns
+        -------
+        converted : Timestamp
+
+        Raises
+        ------
+        TypeError
+            If Timestamp is tz-naive.
+
+        See Also
+        --------
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+        DatetimeIndex.tz_convert : Convert a DatetimeIndex to another time zone.
+        DatetimeIndex.tz_localize : Localize a DatetimeIndex to a specific time zone.
+        datetime.datetime.astimezone : Convert a datetime object to another time zone.
+
+        Examples
+        --------
+        Create a timestamp object with UTC timezone:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651', tz='UTC')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651+0000', tz='UTC')
+
+        Change to Tokyo timezone:
+
+        >>> ts.tz_convert(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Can also use ``astimezone``:
+
+        >>> ts.astimezone(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.tz_convert(tz='Asia/Tokyo')
+        NaT
+        """,
+    )
+    tz_localize = _make_nat_func(
+        "tz_localize",
+        """
+        Localize the Timestamp to a timezone.
+
+        Convert naive Timestamp to local time zone or remove
+        timezone from timezone-aware Timestamp.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for time which Timestamp will be converted to.
+            None will remove timezone holding local time.
+
+        ambiguous : bool, 'NaT', default 'raise'
+            When clocks moved backward due to DST, ambiguous times may arise.
+            For example in Central European Time (UTC+01), when going from
+            03:00 DST to 02:00 non-DST, 02:30:00 local time occurs both at
+            00:30:00 UTC and at 01:30:00 UTC. In such a situation, the
+            `ambiguous` parameter dictates how ambiguous times should be
+            handled.
+
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : 'shift_forward', 'shift_backward', 'NaT', timedelta, \
+default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            The behavior is as follows:
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        localized : Timestamp
+
+        Raises
+        ------
+        TypeError
+            If the Timestamp is tz-aware and tz is not None.
+
+        See Also
+        --------
+        Timestamp.tzinfo : Returns the timezone information of the Timestamp.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        DatetimeIndex.tz_localize : Localize a DatetimeIndex to a specific time zone.
+        datetime.datetime.astimezone : Convert a datetime object to another time zone.
+
+        Examples
+        --------
+        Create a naive timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651')
+
+        Add 'Europe/Stockholm' as timezone:
+
+        >>> ts.tz_localize(tz='Europe/Stockholm')
+        Timestamp('2020-03-14 15:32:52.192548651+0100', tz='Europe/Stockholm')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.tz_localize()
+        NaT
+        """,
+    )
+    replace = _make_nat_func(
+        "replace",
+        """
+        Implements datetime.replace, handles nanoseconds.
+
+        This method creates a new `Timestamp` object by replacing the specified
+        fields with new values. The new `Timestamp` retains the original fields
+        that are not explicitly replaced. This method handles nanoseconds, and
+        the `tzinfo` parameter allows for timezone replacement without conversion.
+
+        Parameters
+        ----------
+        year : int, optional
+            The year to replace. If `None`, the year is not changed.
+        month : int, optional
+            The month to replace. If `None`, the month is not changed.
+        day : int, optional
+            The day to replace. If `None`, the day is not changed.
+        hour : int, optional
+            The hour to replace. If `None`, the hour is not changed.
+        minute : int, optional
+            The minute to replace. If `None`, the minute is not changed.
+        second : int, optional
+            The second to replace. If `None`, the second is not changed.
+        microsecond : int, optional
+            The microsecond to replace. If `None`, the microsecond is not changed.
+        nanosecond : int, optional
+            The nanosecond to replace. If `None`, the nanosecond is not changed.
+        tzinfo : tz-convertible, optional
+            The timezone information to replace. If `None`, the timezone is not changed.
+        fold : int, optional
+            The fold information to replace. If `None`, the fold is not changed.
+
+        Returns
+        -------
+        Timestamp
+            A new `Timestamp` object with the specified fields replaced.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Notes
+        -----
+        The `replace` method does not perform timezone conversions. If you need
+        to convert the timezone, use the `tz_convert` method instead.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651', tz='UTC')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651+0000', tz='UTC')
+
+        Replace year and the hour:
+
+        >>> ts.replace(year=1999, hour=10)
+        Timestamp('1999-03-14 10:32:52.192548651+0000', tz='UTC')
+
+        Replace timezone (not a conversion):
+
+        >>> import zoneinfo
+        >>> ts.replace(tzinfo=zoneinfo.ZoneInfo('US/Pacific'))
+        Timestamp('2020-03-14 15:32:52.192548651-0700', tz='US/Pacific')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.replace(tzinfo=zoneinfo.ZoneInfo('US/Pacific'))
+        NaT
+        """,
+    )
+
+    @property
+    def tz(self) -> None:
+        return None
+
+    @property
+    def tzinfo(self) -> None:
+        return None
+
+    def as_unit(self, str unit, bint round_ok=True) -> "NaTType":
+        """
+        Convert the underlying int64 representation to the given unit.
+
+        Parameters
+        ----------
+        unit : {"ns", "us", "ms", "s"}
+        round_ok : bool, default True
+            If False and the conversion requires rounding, raise.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Timestamp.asm8 : Return numpy datetime64 format with same precision.
+        Timestamp.to_pydatetime : Convert Timestamp object to a native
+            Python datetime object.
+        to_timedelta : Convert argument into timedelta object,
+            which can represent differences in times.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 00:00:00.01')
+        >>> ts
+        Timestamp('2023-01-01 00:00:00.010000')
+        >>> ts.unit
+        'ms'
+        >>> ts = ts.as_unit('s')
+        >>> ts
+        Timestamp('2023-01-01 00:00:00')
+        >>> ts.unit
+        's'
+        """
+        return c_NaT
+
+
+c_NaT = NaTType()  # C-visible
+NaT = c_NaT        # Python-visible
+
+
+# ----------------------------------------------------------------------
+
+cdef bint checknull_with_nat(object val):
+    """
+    Utility to check if a value is a nat or not.
+    """
+    return val is None or util.is_nan(val) or val is c_NaT
+
+
+cdef bint is_dt64nat(object val):
+    """
+    Is this a np.datetime64 object np.datetime64("NaT").
+    """
+    if cnp.is_datetime64_object(val):
+        return cnp.get_datetime64_value(val) == NPY_NAT
+    return False
+
+
+cdef bint is_td64nat(object val):
+    """
+    Is this a np.timedelta64 object np.timedelta64("NaT").
+    """
+    if cnp.is_timedelta64_object(val):
+        return cnp.get_timedelta64_value(val) == NPY_NAT
+    return False
diff --git a/pandas/pandas/_libs/tslibs/np_datetime.pxd b/pandas/pandas/_libs/tslibs/np_datetime.pxd
new file mode 100644
index 0000000000000..3e5654b70cd92
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/np_datetime.pxd
@@ -0,0 +1,122 @@
+cimport numpy as cnp
+from cpython.datetime cimport (
+    date,
+    datetime,
+)
+from numpy cimport (
+    int32_t,
+    int64_t,
+    npy_datetime,
+    npy_timedelta,
+)
+
+
+# TODO(cython3): most of these can be cimported directly from numpy
+cdef extern from "numpy/ndarraytypes.h":
+    ctypedef struct npy_datetimestruct:
+        int64_t year
+        int32_t month, day, hour, min, sec, us, ps, as
+
+    ctypedef enum NPY_DATETIMEUNIT:
+        NPY_FR_Y
+        NPY_FR_M
+        NPY_FR_W
+        NPY_FR_D
+        NPY_FR_B
+        NPY_FR_h
+        NPY_FR_m
+        NPY_FR_s
+        NPY_FR_ms
+        NPY_FR_us
+        NPY_FR_ns
+        NPY_FR_ps
+        NPY_FR_fs
+        NPY_FR_as
+        NPY_FR_GENERIC
+
+    int64_t NPY_DATETIME_NAT  # elsewhere we call this NPY_NAT
+
+
+cdef extern from "pandas/datetime/pd_datetime.h":
+    ctypedef struct pandas_timedeltastruct:
+        int64_t days
+        int32_t hrs, min, sec, ms, us, ns, seconds, microseconds, nanoseconds
+
+    void pandas_datetime_to_datetimestruct(npy_datetime val,
+                                           NPY_DATETIMEUNIT fr,
+                                           npy_datetimestruct *result) nogil
+
+    npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT fr,
+                                                npy_datetimestruct *d) except? -1 nogil
+
+    void pandas_timedelta_to_timedeltastruct(npy_timedelta val,
+                                             NPY_DATETIMEUNIT fr,
+                                             pandas_timedeltastruct *result
+                                             ) nogil
+
+    void PandasDateTime_IMPORT()
+
+    ctypedef enum FormatRequirement:
+        PARTIAL_MATCH
+        EXACT_MATCH
+        INFER_FORMAT
+
+# You must call this before using the PandasDateTime CAPI functions
+cdef inline void import_pandas_datetime() noexcept:
+    PandasDateTime_IMPORT
+
+cdef bint cmp_scalar(int64_t lhs, int64_t rhs, int op) except -1
+
+cdef str dts_to_iso_string(npy_datetimestruct *dts)
+
+cdef check_dts_bounds(npy_datetimestruct *dts, NPY_DATETIMEUNIT unit=?)
+
+cdef int64_t pydatetime_to_dt64(
+    datetime val, npy_datetimestruct *dts, NPY_DATETIMEUNIT reso=?
+) except? -1
+cdef void pydatetime_to_dtstruct(datetime dt, npy_datetimestruct *dts) noexcept
+cdef int64_t pydate_to_dt64(
+    date val, npy_datetimestruct *dts, NPY_DATETIMEUNIT reso=?
+) except? -1
+cdef void pydate_to_dtstruct(date val, npy_datetimestruct *dts) noexcept
+
+cdef NPY_DATETIMEUNIT get_datetime64_unit(object obj) noexcept nogil
+
+cdef int string_to_dts(
+    str val,
+    npy_datetimestruct* dts,
+    NPY_DATETIMEUNIT* out_bestunit,
+    int* out_local,
+    int* out_tzoffset,
+    bint want_exc,
+    str format = *,
+    bint exact = *
+) except? -1
+
+cdef NPY_DATETIMEUNIT get_unit_from_dtype(cnp.dtype dtype)
+
+cpdef cnp.ndarray astype_overflowsafe(
+    cnp.ndarray values,  # ndarray[datetime64[anyunit]]
+    cnp.dtype dtype,  # ndarray[datetime64[anyunit]]
+    bint copy=*,
+    bint round_ok=*,
+    bint is_coerce=*,
+)
+cdef int64_t get_conversion_factor(
+    NPY_DATETIMEUNIT from_unit,
+    NPY_DATETIMEUNIT to_unit,
+) except? -1
+
+cdef bint cmp_dtstructs(npy_datetimestruct* left, npy_datetimestruct* right, int op)
+cdef get_implementation_bounds(
+    NPY_DATETIMEUNIT reso, npy_datetimestruct *lower, npy_datetimestruct *upper
+)
+
+cdef int64_t convert_reso(
+    int64_t value,
+    NPY_DATETIMEUNIT from_reso,
+    NPY_DATETIMEUNIT to_reso,
+    bint round_ok,
+) except? -1
+
+cpdef cnp.ndarray add_overflowsafe(cnp.ndarray left, cnp.ndarray right)
diff --git a/pandas/pandas/_libs/tslibs/np_datetime.pyi b/pandas/pandas/_libs/tslibs/np_datetime.pyi
new file mode 100644
index 0000000000000..00ef35c50e532
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/np_datetime.pyi
@@ -0,0 +1,27 @@
+import numpy as np
+
+from pandas._typing import npt
+
+class OutOfBoundsDatetime(ValueError): ...
+class OutOfBoundsTimedelta(ValueError): ...
+
+# only exposed for testing
+def py_get_unit_from_dtype(dtype: np.dtype): ...
+def py_td64_to_tdstruct(td64: int, unit: int) -> dict: ...
+def astype_overflowsafe(
+    values: np.ndarray,
+    dtype: np.dtype,
+    copy: bool = ...,
+    round_ok: bool = ...,
+    is_coerce: bool = ...,
+) -> np.ndarray: ...
+def is_unitless(dtype: np.dtype) -> bool: ...
+def compare_mismatched_resolutions(
+    left: np.ndarray, right: np.ndarray, op
+) -> npt.NDArray[np.bool_]: ...
+def add_overflowsafe(
+    left: npt.NDArray[np.int64],
+    right: npt.NDArray[np.int64],
+) -> npt.NDArray[np.int64]: ...
+def get_supported_dtype(dtype: np.dtype) -> np.dtype: ...
+def is_supported_dtype(dtype: np.dtype) -> bool: ...
diff --git a/pandas/pandas/_libs/tslibs/np_datetime.pyx b/pandas/pandas/_libs/tslibs/np_datetime.pyx
new file mode 100644
index 0000000000000..1b7f04fe17238
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/np_datetime.pyx
@@ -0,0 +1,758 @@
+cimport cython
+from cpython.datetime cimport (
+    PyDateTime_CheckExact,
+    PyDateTime_DATE_GET_HOUR,
+    PyDateTime_DATE_GET_MICROSECOND,
+    PyDateTime_DATE_GET_MINUTE,
+    PyDateTime_DATE_GET_SECOND,
+    PyDateTime_GET_DAY,
+    PyDateTime_GET_MONTH,
+    PyDateTime_GET_YEAR,
+    import_datetime,
+)
+from cpython.object cimport (
+    Py_EQ,
+    Py_GE,
+    Py_GT,
+    Py_LE,
+    Py_LT,
+    Py_NE,
+)
+from cpython.unicode cimport PyUnicode_AsUTF8AndSize
+from libc.stdint cimport INT64_MAX
+
+import_datetime()
+PandasDateTime_IMPORT
+
+import operator
+
+import numpy as np
+
+cimport numpy as cnp
+
+cnp.import_array()
+from numpy cimport (
+    PyArray_DatetimeMetaData,
+    PyDatetimeScalarObject,
+    int64_t,
+    ndarray,
+    uint8_t,
+)
+
+from pandas._libs.tslibs.dtypes cimport (
+    get_supported_reso,
+    is_supported_unit,
+    npy_unit_to_abbrev,
+    npy_unit_to_attrname,
+)
+
+
+cdef extern from "pandas/datetime/pd_datetime.h":
+    int cmp_npy_datetimestruct(npy_datetimestruct *a,
+                               npy_datetimestruct *b)
+
+    # AS, FS, PS versions exist but are not imported because they are not used.
+    npy_datetimestruct _NS_MIN_DTS, _NS_MAX_DTS
+    npy_datetimestruct _US_MIN_DTS, _US_MAX_DTS
+    npy_datetimestruct _MS_MIN_DTS, _MS_MAX_DTS
+    npy_datetimestruct _S_MIN_DTS, _S_MAX_DTS
+    npy_datetimestruct _M_MIN_DTS, _M_MAX_DTS
+
+    PyArray_DatetimeMetaData get_datetime_metadata_from_dtype(cnp.PyArray_Descr *dtype)
+
+    int parse_iso_8601_datetime(const char *str, int len, int want_exc,
+                                npy_datetimestruct *out,
+                                NPY_DATETIMEUNIT *out_bestunit,
+                                int *out_local, int *out_tzoffset,
+                                const char *format, int format_len,
+                                FormatRequirement exact)
+
+# ----------------------------------------------------------------------
+# numpy object inspection
+
+
+cdef NPY_DATETIMEUNIT get_datetime64_unit(object obj) noexcept nogil:
+    """
+    returns the unit part of the dtype for a numpy datetime64 object.
+    """
+    return <NPY_DATETIMEUNIT>(<PyDatetimeScalarObject*>obj).obmeta.base
+
+
+cdef NPY_DATETIMEUNIT get_unit_from_dtype(cnp.dtype dtype):
+    # NB: caller is responsible for ensuring this is *some* datetime64 or
+    #  timedelta64 dtype, otherwise we can segfault
+    cdef:
+        cnp.PyArray_Descr* descr = <cnp.PyArray_Descr*>dtype
+        PyArray_DatetimeMetaData meta
+    meta = get_datetime_metadata_from_dtype(descr)
+    return meta.base
+
+
+def py_get_unit_from_dtype(dtype):
+    # for testing get_unit_from_dtype; adds 896 bytes to the .so file.
+    return get_unit_from_dtype(dtype)
+
+
+def get_supported_dtype(dtype: cnp.dtype) -> cnp.dtype:
+    reso = get_unit_from_dtype(dtype)
+    new_reso = get_supported_reso(reso)
+    new_unit = npy_unit_to_abbrev(new_reso)
+
+    # Accessing dtype.kind here incorrectly(?) gives "" instead of "m"/"M",
+    #  so we check type_num instead
+    if dtype.type_num == cnp.NPY_DATETIME:
+        new_dtype = np.dtype(f"M8[{new_unit}]")
+    else:
+        new_dtype = np.dtype(f"m8[{new_unit}]")
+    return new_dtype
+
+
+def is_supported_dtype(dtype: cnp.dtype) -> bool:
+    if dtype.type_num not in [cnp.NPY_DATETIME, cnp.NPY_TIMEDELTA]:
+        raise ValueError("is_unitless dtype must be datetime64 or timedelta64")
+    cdef:
+        NPY_DATETIMEUNIT unit = get_unit_from_dtype(dtype)
+    return is_supported_unit(unit)
+
+
+def is_unitless(dtype: cnp.dtype) -> bool:
+    """
+    Check if a datetime64 or timedelta64 dtype has no attached unit.
+    """
+    if dtype.type_num not in [cnp.NPY_DATETIME, cnp.NPY_TIMEDELTA]:
+        raise ValueError("is_unitless dtype must be datetime64 or timedelta64")
+    cdef:
+        NPY_DATETIMEUNIT unit = get_unit_from_dtype(dtype)
+
+    return unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+
+
+# ----------------------------------------------------------------------
+# Comparison
+
+
+cdef bint cmp_dtstructs(
+    npy_datetimestruct* left, npy_datetimestruct* right, int op
+):
+    cdef:
+        int cmp_res
+
+    cmp_res = cmp_npy_datetimestruct(left, right)
+    if op == Py_EQ:
+        return cmp_res == 0
+    if op == Py_NE:
+        return cmp_res != 0
+    if op == Py_GT:
+        return cmp_res == 1
+    if op == Py_LT:
+        return cmp_res == -1
+    if op == Py_GE:
+        return cmp_res == 1 or cmp_res == 0
+    else:
+        # i.e. op == Py_LE
+        return cmp_res == -1 or cmp_res == 0
+
+
+cdef bint cmp_scalar(int64_t lhs, int64_t rhs, int op) except -1:
+    """
+    cmp_scalar is a more performant version of PyObject_RichCompare
+    typed for int64_t arguments.
+    """
+    if op == Py_EQ:
+        return lhs == rhs
+    elif op == Py_NE:
+        return lhs != rhs
+    elif op == Py_LT:
+        return lhs < rhs
+    elif op == Py_LE:
+        return lhs <= rhs
+    elif op == Py_GT:
+        return lhs > rhs
+    elif op == Py_GE:
+        return lhs >= rhs
+
+
+class OutOfBoundsDatetime(ValueError):
+    """
+    Raised when the datetime is outside the range that can be represented.
+
+    This error occurs when attempting to convert or parse a datetime value
+    that exceeds the bounds supported by pandas' internal datetime
+    representation.
+
+    See Also
+    --------
+    to_datetime : Convert argument to datetime.
+    Timestamp : Pandas replacement for python ``datetime.datetime`` object.
+
+    Examples
+    --------
+    >>> pd.to_datetime("08335394550")
+    Traceback (most recent call last):
+    OutOfBoundsDatetime: Parsing "08335394550" to datetime overflows,
+    at position 0
+    """
+    pass
+
+
+class OutOfBoundsTimedelta(ValueError):
+    """
+    Raised when encountering a timedelta value that cannot be represented.
+
+    Representation should be within a timedelta64[ns].
+
+    See Also
+    --------
+    date_range : Return a fixed frequency DatetimeIndex.
+
+    Examples
+    --------
+    >>> pd.date_range(start="1/1/1700", freq="B", periods=100000)
+    Traceback (most recent call last):
+    OutOfBoundsTimedelta: Cannot cast 139999 days 00:00:00
+    to unit='ns' without overflow.
+    """
+    # Timedelta analogue to OutOfBoundsDatetime
+    pass
+
+
+cdef get_implementation_bounds(
+    NPY_DATETIMEUNIT reso,
+    npy_datetimestruct *lower,
+    npy_datetimestruct *upper,
+):
+    if reso == NPY_FR_ns:
+        upper[0] = _NS_MAX_DTS
+        lower[0] = _NS_MIN_DTS
+    elif reso == NPY_FR_us:
+        upper[0] = _US_MAX_DTS
+        lower[0] = _US_MIN_DTS
+    elif reso == NPY_FR_ms:
+        upper[0] = _MS_MAX_DTS
+        lower[0] = _MS_MIN_DTS
+    elif reso == NPY_FR_s:
+        upper[0] = _S_MAX_DTS
+        lower[0] = _S_MIN_DTS
+    elif reso == NPY_FR_m:
+        upper[0] = _M_MAX_DTS
+        lower[0] = _M_MIN_DTS
+    else:
+        raise NotImplementedError(reso)
+
+
+cdef str dts_to_iso_string(npy_datetimestruct *dts):
+    return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+            f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}")
+
+
+cdef check_dts_bounds(npy_datetimestruct *dts, NPY_DATETIMEUNIT unit=NPY_FR_ns):
+    """Raises OutOfBoundsDatetime if the given date is outside the range that
+    can be represented by nanosecond-resolution 64-bit integers."""
+    cdef:
+        bint error = False
+        npy_datetimestruct cmp_upper, cmp_lower
+
+    get_implementation_bounds(unit, &cmp_lower, &cmp_upper)
+
+    if cmp_npy_datetimestruct(dts, &cmp_lower) == -1:
+        error = True
+    elif cmp_npy_datetimestruct(dts, &cmp_upper) == 1:
+        error = True
+
+    if error:
+        fmt = dts_to_iso_string(dts)
+        attrname = npy_unit_to_attrname[unit]
+        raise OutOfBoundsDatetime(f"Out of bounds {attrname} timestamp: {fmt}")
+
+
+# ----------------------------------------------------------------------
+# Conversion
+
+
+# just exposed for testing at the moment
+def py_td64_to_tdstruct(int64_t td64, NPY_DATETIMEUNIT unit):
+    cdef:
+        pandas_timedeltastruct tds
+    pandas_timedelta_to_timedeltastruct(td64, unit, &tds)
+    return tds  # <- returned as a dict to python
+
+
+cdef void pydatetime_to_dtstruct(datetime dt, npy_datetimestruct *dts) noexcept:
+    if PyDateTime_CheckExact(dt):
+        dts.year = PyDateTime_GET_YEAR(dt)
+    else:
+        # We use dt.year instead of PyDateTime_GET_YEAR because with Timestamp
+        #  we override year such that PyDateTime_GET_YEAR is incorrect.
+        dts.year = dt.year
+
+    dts.month = PyDateTime_GET_MONTH(dt)
+    dts.day = PyDateTime_GET_DAY(dt)
+    dts.hour = PyDateTime_DATE_GET_HOUR(dt)
+    dts.min = PyDateTime_DATE_GET_MINUTE(dt)
+    dts.sec = PyDateTime_DATE_GET_SECOND(dt)
+    dts.us = PyDateTime_DATE_GET_MICROSECOND(dt)
+    dts.ps = dts.as = 0
+
+
+cdef int64_t pydatetime_to_dt64(datetime val,
+                                npy_datetimestruct *dts,
+                                NPY_DATETIMEUNIT reso=NPY_FR_ns) except? -1:
+    """
+    Note we are assuming that the datetime object is timezone-naive.
+    """
+    cdef int64_t result
+    pydatetime_to_dtstruct(val, dts)
+    try:
+        result = npy_datetimestruct_to_datetime(reso, dts)
+    except OverflowError as err:
+        attrname = npy_unit_to_attrname[reso]
+        raise OutOfBoundsDatetime(
+            f"Out of bounds {attrname} timestamp: {val}"
+        ) from err
+
+    return result
+
+
+cdef void pydate_to_dtstruct(date val, npy_datetimestruct *dts) noexcept:
+    dts.year = PyDateTime_GET_YEAR(val)
+    dts.month = PyDateTime_GET_MONTH(val)
+    dts.day = PyDateTime_GET_DAY(val)
+    dts.hour = dts.min = dts.sec = dts.us = 0
+    dts.ps = dts.as = 0
+    return
+
+
+cdef int64_t pydate_to_dt64(
+    date val, npy_datetimestruct *dts, NPY_DATETIMEUNIT reso=NPY_FR_ns
+) except? -1:
+    cdef int64_t result
+    pydate_to_dtstruct(val, dts)
+
+    try:
+        result = npy_datetimestruct_to_datetime(reso, dts)
+    except OverflowError as err:
+        attrname = npy_unit_to_attrname[reso]
+        raise OutOfBoundsDatetime(f"Out of bounds {attrname} timestamp: {val}") from err
+
+    return result
+
+
+cdef int string_to_dts(
+    str val,
+    npy_datetimestruct* dts,
+    NPY_DATETIMEUNIT* out_bestunit,
+    int* out_local,
+    int* out_tzoffset,
+    bint want_exc,
+    str format=None,
+    bint exact=True,
+) except? -1:
+    cdef:
+        Py_ssize_t length
+        const char* buf
+        Py_ssize_t format_length
+        const char* format_buf
+        FormatRequirement format_requirement
+
+    buf = PyUnicode_AsUTF8AndSize(val, &length)
+    if format is None:
+        format_buf = b""
+        format_length = 0
+        format_requirement = INFER_FORMAT
+    else:
+        format_buf = PyUnicode_AsUTF8AndSize(format, &format_length)
+        format_requirement = <FormatRequirement>exact
+    return parse_iso_8601_datetime(buf, length, want_exc,
+                                   dts, out_bestunit, out_local, out_tzoffset,
+                                   format_buf, format_length,
+                                   format_requirement)
+
+
+cpdef ndarray astype_overflowsafe(
+    ndarray values,
+    cnp.dtype dtype,
+    bint copy=True,
+    bint round_ok=True,
+    bint is_coerce=False,
+):
+    """
+    Convert an ndarray with datetime64[X] to datetime64[Y]
+    or timedelta64[X] to timedelta64[Y],
+    raising on overflow.
+    """
+    if values.descr.type_num == dtype.type_num == cnp.NPY_DATETIME:
+        # i.e. dtype.kind == "M"
+        dtype_name = "datetime64"
+    elif values.descr.type_num == dtype.type_num == cnp.NPY_TIMEDELTA:
+        # i.e. dtype.kind == "m"
+        dtype_name = "timedelta64"
+    else:
+        raise TypeError(
+            "astype_overflowsafe values.dtype and dtype must be either "
+            "both-datetime64 or both-timedelta64."
+        )
+
+    cdef:
+        NPY_DATETIMEUNIT from_unit = get_unit_from_dtype(values.dtype)
+        NPY_DATETIMEUNIT to_unit = get_unit_from_dtype(dtype)
+
+    if from_unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        raise TypeError(f"{dtype_name} values must have a unit specified")
+
+    if to_unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+        # without raising explicitly here, we end up with a SystemError
+        # built-in function [...] returned a result with an error
+        raise ValueError(
+            f"{dtype_name} dtype must have a unit specified"
+        )
+
+    if from_unit == to_unit:
+        # Check this before allocating result for perf, might save some memory
+        if copy:
+            return values.copy()
+        return values
+
+    elif from_unit > to_unit:
+        iresult2 = _astype_overflowsafe_to_smaller_unit(
+            values.view("i8"), from_unit, to_unit, round_ok=round_ok
+        )
+        return iresult2.view(dtype)
+
+    if (<object>values).dtype.byteorder == ">":
+        # GH#29684 we incorrectly get OutOfBoundsDatetime if we dont swap
+        values = values.astype(values.dtype.newbyteorder("<"))
+
+    cdef:
+        ndarray i8values = values.view("i8")
+
+        # equiv: result = np.empty((<object>values).shape, dtype="i8")
+        ndarray iresult = cnp.PyArray_EMPTY(
+            values.ndim, values.shape, cnp.NPY_INT64, 0
+        )
+
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(iresult, i8values)
+        Py_ssize_t i, N = values.size
+        int64_t value, new_value
+        npy_datetimestruct dts
+        bint is_td = dtype.type_num == cnp.NPY_TIMEDELTA
+
+    for i in range(N):
+        # Analogous to: item = values[i]
+        value = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if value == NPY_DATETIME_NAT:
+            new_value = NPY_DATETIME_NAT
+        else:
+            pandas_datetime_to_datetimestruct(value, from_unit, &dts)
+
+            try:
+                check_dts_bounds(&dts, to_unit)
+            except OutOfBoundsDatetime as err:
+                if is_coerce:
+                    new_value = NPY_DATETIME_NAT
+                elif is_td:
+                    from_abbrev = np.datetime_data(values.dtype)[0]
+                    np_val = np.timedelta64(value, from_abbrev)
+                    msg = (
+                        "Cannot convert {np_val} to {dtype} without overflow"
+                        .format(np_val=str(np_val), dtype=str(dtype))
+                    )
+                    raise OutOfBoundsTimedelta(msg) from err
+                else:
+                    raise
+            else:
+                new_value = npy_datetimestruct_to_datetime(to_unit, &dts)
+
+        # Analogous to: iresult[i] = new_value
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = new_value
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return iresult.view(dtype)
+
+
+# TODO(numpy#16352): try to upstream this fix to numpy
+def compare_mismatched_resolutions(ndarray left, ndarray right, op):
+    """
+    Overflow-safe comparison of timedelta64/datetime64 with mismatched resolutions.
+
+    >>> left = np.array([500], dtype="M8[Y]")
+    >>> right = np.array([0], dtype="M8[ns]")
+    >>> left < right  # <- wrong!
+    array([ True])
+    """
+
+    if left.dtype.kind != right.dtype.kind or left.dtype.kind not in "mM":
+        raise ValueError("left and right must both be timedelta64 or both datetime64")
+
+    cdef:
+        int op_code = op_to_op_code(op)
+        NPY_DATETIMEUNIT left_unit = get_unit_from_dtype(left.dtype)
+        NPY_DATETIMEUNIT right_unit = get_unit_from_dtype(right.dtype)
+
+        # equiv: result = np.empty((<object>left).shape, dtype="bool")
+        ndarray result = cnp.PyArray_EMPTY(
+            left.ndim, left.shape, cnp.NPY_BOOL, 0
+        )
+
+        ndarray lvalues = left.view("i8")
+        ndarray rvalues = right.view("i8")
+
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew3(result, lvalues, rvalues)
+        int64_t lval, rval
+        bint res_value
+
+        Py_ssize_t i, N = left.size
+        npy_datetimestruct ldts, rdts
+
+    for i in range(N):
+        # Analogous to: lval = lvalues[i]
+        lval = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        # Analogous to: rval = rvalues[i]
+        rval = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 2))[0]
+
+        if lval == NPY_DATETIME_NAT or rval == NPY_DATETIME_NAT:
+            res_value = op_code == Py_NE
+
+        else:
+            pandas_datetime_to_datetimestruct(lval, left_unit, &ldts)
+            pandas_datetime_to_datetimestruct(rval, right_unit, &rdts)
+
+            res_value = cmp_dtstructs(&ldts, &rdts, op_code)
+
+        # Analogous to: result[i] = res_value
+        (<uint8_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_value
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return result
+
+
+cdef int op_to_op_code(op):
+    if op is operator.eq:
+        return Py_EQ
+    if op is operator.ne:
+        return Py_NE
+    if op is operator.le:
+        return Py_LE
+    if op is operator.lt:
+        return Py_LT
+    if op is operator.ge:
+        return Py_GE
+    if op is operator.gt:
+        return Py_GT
+
+
+cdef ndarray _astype_overflowsafe_to_smaller_unit(
+    ndarray i8values,
+    NPY_DATETIMEUNIT from_unit,
+    NPY_DATETIMEUNIT to_unit,
+    bint round_ok,
+):
+    """
+    Overflow-safe conversion for cases with from_unit > to_unit, e.g. ns->us.
+    In addition for checking for overflows (which can occur near the lower
+    implementation bound, see numpy#22346), this checks for truncation,
+    e.g. 1500ns->1us.
+    """
+    # e.g. test_astype_ns_to_ms_near_bounds is a case with round_ok=True where
+    #  just using numpy's astype silently fails
+    cdef:
+        Py_ssize_t i, N = i8values.size
+
+        # equiv: iresult = np.empty((<object>i8values).shape, dtype="i8")
+        ndarray iresult = cnp.PyArray_EMPTY(
+            i8values.ndim, i8values.shape, cnp.NPY_INT64, 0
+        )
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(iresult, i8values)
+
+        # Note the arguments to_unit, from unit are swapped vs how they
+        #  are passed when going to a higher-frequency reso.
+        int64_t mult = get_conversion_factor(to_unit, from_unit)
+        int64_t value, mod
+
+    for i in range(N):
+        # Analogous to: item = i8values[i]
+        value = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if value == NPY_DATETIME_NAT:
+            new_value = NPY_DATETIME_NAT
+        else:
+            new_value, mod = divmod(value, mult)
+            if not round_ok and mod != 0:
+                from_abbrev = npy_unit_to_abbrev(from_unit)
+                to_abbrev = npy_unit_to_abbrev(to_unit)
+                raise ValueError(
+                    f"Cannot losslessly cast '{value} {from_abbrev}' to {to_abbrev}"
+                )
+
+        # Analogous to: iresult[i] = new_value
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = new_value
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return iresult
+
+
+cdef int64_t get_conversion_factor(
+    NPY_DATETIMEUNIT from_unit,
+    NPY_DATETIMEUNIT to_unit
+) except? -1:
+    """
+    Find the factor by which we need to multiply to convert from from_unit to to_unit.
+    """
+    cdef int64_t value, overflow_limit, factor
+    if (
+        from_unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+        or to_unit == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+    ):
+        raise ValueError("unit-less resolutions are not supported")
+    if from_unit > to_unit:
+        raise ValueError("from_unit must be <= to_unit")
+
+    if from_unit == to_unit:
+        return 1
+
+    if from_unit == NPY_DATETIMEUNIT.NPY_FR_W:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_D, to_unit)
+        factor = 7
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_D:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_h, to_unit)
+        factor = 24
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_h:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_m, to_unit)
+        factor = 60
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_m:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_s, to_unit)
+        factor = 60
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_s:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_ms, to_unit)
+        factor = 1000
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_ms:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_us, to_unit)
+        factor = 1000
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_us:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_ns, to_unit)
+        factor = 1000
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_ns:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_ps, to_unit)
+        factor = 1000
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_ps:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_fs, to_unit)
+        factor = 1000
+    elif from_unit == NPY_DATETIMEUNIT.NPY_FR_fs:
+        value = get_conversion_factor(NPY_DATETIMEUNIT.NPY_FR_as, to_unit)
+        factor = 1000
+    else:
+        raise ValueError("Converting from M or Y units is not supported.")
+
+    overflow_limit = INT64_MAX // factor
+    if value > overflow_limit or value < -overflow_limit:
+        raise OverflowError("result would overflow")
+
+    return factor * value
+
+
+cdef int64_t convert_reso(
+    int64_t value,
+    NPY_DATETIMEUNIT from_reso,
+    NPY_DATETIMEUNIT to_reso,
+    bint round_ok,
+) except? -1:
+    cdef:
+        int64_t res_value, mult, div, mod, overflow_limit
+
+    if from_reso == to_reso:
+        return value
+
+    elif to_reso < from_reso:
+        # e.g. ns -> us, no risk of overflow, but can be lossy rounding
+        mult = get_conversion_factor(to_reso, from_reso)
+        div, mod = divmod(value, mult)
+        if mod > 0 and not round_ok:
+            raise ValueError("Cannot losslessly convert units")
+
+        # Note that when mod > 0, we follow np.timedelta64 in always
+        #  rounding down.
+        res_value = div
+
+    elif (
+        from_reso == NPY_FR_Y
+        or from_reso == NPY_FR_M
+        or to_reso == NPY_FR_Y
+        or to_reso == NPY_FR_M
+    ):
+        # Converting by multiplying isn't _quite_ right bc the number of
+        #  seconds in a month/year isn't fixed.
+        res_value = _convert_reso_with_dtstruct(value, from_reso, to_reso)
+
+    else:
+        # e.g. ns -> us, risk of overflow, but no risk of lossy rounding
+        mult = get_conversion_factor(from_reso, to_reso)
+        overflow_limit = INT64_MAX // mult
+        if value > overflow_limit or value < -overflow_limit:
+            # Note: caller is responsible for re-raising as OutOfBoundsTimedelta
+            raise OverflowError("result would overflow")
+
+        res_value = value * mult
+
+    return res_value
+
+
+cdef int64_t _convert_reso_with_dtstruct(
+    int64_t value,
+    NPY_DATETIMEUNIT from_unit,
+    NPY_DATETIMEUNIT to_unit,
+) except? -1:
+    cdef:
+        npy_datetimestruct dts
+        int64_t result
+
+    pandas_datetime_to_datetimestruct(value, from_unit, &dts)
+    try:
+        result = npy_datetimestruct_to_datetime(to_unit, &dts)
+    except OverflowError as err:
+        raise OutOfBoundsDatetime from err
+
+    return result
+
+
+@cython.overflowcheck(True)
+cpdef cnp.ndarray add_overflowsafe(cnp.ndarray left, cnp.ndarray right):
+    """
+    Overflow-safe addition for datetime64/timedelta64 dtypes.
+
+    `right` may either be zero-dim or of the same shape as `left`.
+    """
+    cdef:
+        Py_ssize_t N = left.size
+        int64_t lval, rval, res_value
+        ndarray iresult = cnp.PyArray_EMPTY(
+            left.ndim, left.shape, cnp.NPY_INT64, 0
+        )
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew3(iresult, left, right)
+
+    # Note: doing this try/except outside the loop improves performance over
+    #  doing it inside the loop.
+    try:
+        for i in range(N):
+            # Analogous to: lval = lvalues[i]
+            lval = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+            # Analogous to: rval = rvalues[i]
+            rval = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 2))[0]
+
+            if lval == NPY_DATETIME_NAT or rval == NPY_DATETIME_NAT:
+                res_value = NPY_DATETIME_NAT
+            else:
+                res_value = lval + rval
+
+            # Analogous to: result[i] = res_value
+            (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_value
+
+            cnp.PyArray_MultiIter_NEXT(mi)
+    except OverflowError as err:
+        raise OverflowError("Overflow in int64 addition") from err
+
+    return iresult
diff --git a/pandas/pandas/_libs/tslibs/offsets.pxd b/pandas/pandas/_libs/tslibs/offsets.pxd
new file mode 100644
index 0000000000000..cda6825de35be
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/offsets.pxd
@@ -0,0 +1,12 @@
+from numpy cimport int64_t
+
+
+cpdef to_offset(object obj, bint is_period=*)
+cdef bint is_offset_object(object obj)
+cdef bint is_tick_object(object obj)
+
+cdef class BaseOffset:
+    cdef readonly:
+        int64_t n
+        bint normalize
+        dict _cache
diff --git a/pandas/pandas/_libs/tslibs/offsets.pyi b/pandas/pandas/_libs/tslibs/offsets.pyi
new file mode 100644
index 0000000000000..3f942d6aa3622
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/offsets.pyi
@@ -0,0 +1,285 @@
+from datetime import (
+    datetime,
+    time,
+    timedelta,
+)
+from typing import (
+    Any,
+    Collection,
+    Literal,
+    TypeVar,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs.nattype import NaTType
+from pandas._typing import (
+    OffsetCalendar,
+    Self,
+    npt,
+)
+
+_BaseOffsetT = TypeVar("_BaseOffsetT", bound=BaseOffset)
+_DatetimeT = TypeVar("_DatetimeT", bound=datetime)
+_TimedeltaT = TypeVar("_TimedeltaT", bound=timedelta)
+
+_relativedelta_kwds: set[str]
+prefix_mapping: dict[str, type]
+
+class ApplyTypeError(TypeError): ...
+
+class BaseOffset:
+    n: int
+    normalize: bool
+    def __init__(self, n: int = ..., normalize: bool = ...) -> None: ...
+    def __eq__(self, other) -> bool: ...
+    def __ne__(self, other) -> bool: ...
+    def __hash__(self) -> int: ...
+    @property
+    def kwds(self) -> dict: ...
+    @property
+    def base(self) -> BaseOffset: ...
+    @overload
+    def __add__(self, other: npt.NDArray[np.object_]) -> npt.NDArray[np.object_]: ...
+    @overload
+    def __add__(self, other: BaseOffset) -> Self: ...
+    @overload
+    def __add__(self, other: _DatetimeT) -> _DatetimeT: ...
+    @overload
+    def __add__(self, other: _TimedeltaT) -> _TimedeltaT: ...
+    @overload
+    def __radd__(self, other: npt.NDArray[np.object_]) -> npt.NDArray[np.object_]: ...
+    @overload
+    def __radd__(self, other: BaseOffset) -> Self: ...
+    @overload
+    def __radd__(self, other: _DatetimeT) -> _DatetimeT: ...
+    @overload
+    def __radd__(self, other: _TimedeltaT) -> _TimedeltaT: ...
+    @overload
+    def __radd__(self, other: NaTType) -> NaTType: ...
+    def __sub__(self, other: BaseOffset) -> Self: ...
+    @overload
+    def __rsub__(self, other: npt.NDArray[np.object_]) -> npt.NDArray[np.object_]: ...
+    @overload
+    def __rsub__(self, other: BaseOffset) -> Self: ...
+    @overload
+    def __rsub__(self, other: _DatetimeT) -> _DatetimeT: ...
+    @overload
+    def __rsub__(self, other: _TimedeltaT) -> _TimedeltaT: ...
+    @overload
+    def __mul__(self, other: np.ndarray) -> np.ndarray: ...
+    @overload
+    def __mul__(self, other: int) -> Self: ...
+    @overload
+    def __rmul__(self, other: np.ndarray) -> np.ndarray: ...
+    @overload
+    def __rmul__(self, other: int) -> Self: ...
+    def __neg__(self) -> Self: ...
+    def copy(self) -> Self: ...
+    @property
+    def name(self) -> str: ...
+    @property
+    def rule_code(self) -> str: ...
+    @property
+    def freqstr(self) -> str: ...
+    def _apply(self, other): ...
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray: ...
+    def rollback(self, dt: datetime) -> datetime: ...
+    def rollforward(self, dt: datetime) -> datetime: ...
+    def is_on_offset(self, dt: datetime) -> bool: ...
+    def __setstate__(self, state) -> None: ...
+    def __getstate__(self): ...
+    @property
+    def nanos(self) -> int: ...
+
+def _get_offset(name: str) -> BaseOffset: ...
+
+class SingleConstructorOffset(BaseOffset):
+    @classmethod
+    def _from_name(cls, suffix: None = ...) -> Self: ...
+    def __reduce__(self): ...
+
+@overload
+def to_offset(freq: None, is_period: bool = ...) -> None: ...
+@overload
+def to_offset(freq: _BaseOffsetT, is_period: bool = ...) -> _BaseOffsetT: ...
+@overload
+def to_offset(freq: timedelta | str, is_period: bool = ...) -> BaseOffset: ...
+
+class Tick(SingleConstructorOffset):
+    _creso: int
+    _prefix: str
+    def __init__(self, n: int = ..., normalize: bool = ...) -> None: ...
+    @property
+    def nanos(self) -> int: ...
+
+def delta_to_tick(delta: timedelta) -> Tick: ...
+
+class Day(Tick): ...
+class Hour(Tick): ...
+class Minute(Tick): ...
+class Second(Tick): ...
+class Milli(Tick): ...
+class Micro(Tick): ...
+class Nano(Tick): ...
+
+class RelativeDeltaOffset(BaseOffset):
+    def __init__(self, n: int = ..., normalize: bool = ..., **kwds: Any) -> None: ...
+
+class BusinessMixin(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., offset: timedelta = ...
+    ) -> None: ...
+
+class BusinessDay(BusinessMixin): ...
+
+class BusinessHour(BusinessMixin):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        start: str | time | Collection[str | time] = ...,
+        end: str | time | Collection[str | time] = ...,
+        offset: timedelta = ...,
+    ) -> None: ...
+
+class WeekOfMonthMixin(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., weekday: int = ...
+    ) -> None: ...
+
+class YearOffset(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., month: int | None = ...
+    ) -> None: ...
+
+class BYearEnd(YearOffset): ...
+class BYearBegin(YearOffset): ...
+class YearEnd(YearOffset): ...
+class YearBegin(YearOffset): ...
+
+class QuarterOffset(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., startingMonth: int | None = ...
+    ) -> None: ...
+
+class BQuarterEnd(QuarterOffset): ...
+class BQuarterBegin(QuarterOffset): ...
+class QuarterEnd(QuarterOffset): ...
+class QuarterBegin(QuarterOffset): ...
+class MonthOffset(SingleConstructorOffset): ...
+class MonthEnd(MonthOffset): ...
+class MonthBegin(MonthOffset): ...
+class BusinessMonthEnd(MonthOffset): ...
+class BusinessMonthBegin(MonthOffset): ...
+
+class SemiMonthOffset(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., day_of_month: int | None = ...
+    ) -> None: ...
+
+class SemiMonthEnd(SemiMonthOffset): ...
+class SemiMonthBegin(SemiMonthOffset): ...
+
+class Week(SingleConstructorOffset):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., weekday: int | None = ...
+    ) -> None: ...
+
+class WeekOfMonth(WeekOfMonthMixin):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., week: int = ..., weekday: int = ...
+    ) -> None: ...
+
+class LastWeekOfMonth(WeekOfMonthMixin):
+    def __init__(
+        self, n: int = ..., normalize: bool = ..., weekday: int = ...
+    ) -> None: ...
+
+class FY5253Mixin(SingleConstructorOffset):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        weekday: int = ...,
+        startingMonth: int = ...,
+        variation: Literal["nearest", "last"] = ...,
+    ) -> None: ...
+
+class FY5253(FY5253Mixin): ...
+
+class FY5253Quarter(FY5253Mixin):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        weekday: int = ...,
+        startingMonth: int = ...,
+        qtr_with_extra_week: int = ...,
+        variation: Literal["nearest", "last"] = ...,
+    ) -> None: ...
+
+class Easter(SingleConstructorOffset): ...
+
+class _CustomBusinessMonth(BusinessMixin):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        weekmask: str = ...,
+        holidays: list | None = ...,
+        calendar: OffsetCalendar | None = ...,
+        offset: timedelta = ...,
+    ) -> None: ...
+
+class CustomBusinessDay(BusinessDay):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        weekmask: str = ...,
+        holidays: list | None = ...,
+        calendar: OffsetCalendar | None = ...,
+        offset: timedelta = ...,
+    ) -> None: ...
+
+class CustomBusinessHour(BusinessHour):
+    def __init__(
+        self,
+        n: int = ...,
+        normalize: bool = ...,
+        weekmask: str = ...,
+        holidays: list | None = ...,
+        calendar: OffsetCalendar | None = ...,
+        start: str | time | Collection[str | time] = ...,
+        end: str | time | Collection[str | time] = ...,
+        offset: timedelta = ...,
+    ) -> None: ...
+
+class CustomBusinessMonthEnd(_CustomBusinessMonth): ...
+class CustomBusinessMonthBegin(_CustomBusinessMonth): ...
+class OffsetMeta(type): ...
+class DateOffset(RelativeDeltaOffset, metaclass=OffsetMeta): ...
+
+BDay = BusinessDay
+BMonthEnd = BusinessMonthEnd
+BMonthBegin = BusinessMonthBegin
+CBMonthEnd = CustomBusinessMonthEnd
+CBMonthBegin = CustomBusinessMonthBegin
+CDay = CustomBusinessDay
+
+def roll_qtrday(
+    other: datetime, n: int, month: int, day_opt: str, modby: int
+) -> int: ...
+
+INVALID_FREQ_ERR_MSG: Literal["Invalid frequency: {0}"]
+
+def shift_months(
+    dtindex: npt.NDArray[np.int64],
+    months: int,
+    day_opt: str | None = ...,
+    reso: int = ...,
+) -> npt.NDArray[np.int64]: ...
+
+_offset_map: dict[str, BaseOffset]
diff --git a/pandas/pandas/_libs/tslibs/offsets.pyx b/pandas/pandas/_libs/tslibs/offsets.pyx
new file mode 100644
index 0000000000000..36b431974c121
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/offsets.pyx
@@ -0,0 +1,5494 @@
+import re
+import time
+import warnings
+
+from pandas.util._exceptions import find_stack_level
+
+cimport cython
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    PyDelta_Check,
+    date,
+    datetime,
+    import_datetime,
+    time as dt_time,
+    timedelta,
+)
+
+import warnings
+
+import_datetime()
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+# TODO: formalize having _libs.properties "above" tslibs in the dependency structure
+
+from pandas._libs.properties import cache_readonly
+
+from pandas._libs.tslibs cimport util
+from pandas._libs.tslibs.util cimport (
+    is_float_object,
+    is_integer_object,
+)
+
+from pandas._libs.tslibs.ccalendar import (
+    MONTH_ALIASES,
+    int_to_weekday,
+    weekday_to_int,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas._libs.tslibs.ccalendar cimport (
+    MONTH_TO_CAL_NUM,
+    dayofweek,
+    get_days_in_month,
+    get_firstbday,
+    get_lastbday,
+)
+from pandas._libs.tslibs.conversion cimport localize_pydatetime
+from pandas._libs.tslibs.dtypes cimport (
+    c_OFFSET_RENAMED_FREQSTR,
+    c_OFFSET_TO_PERIOD_FREQSTR,
+    c_PERIOD_AND_OFFSET_DEPR_FREQSTR,
+    c_PERIOD_TO_OFFSET_FREQSTR,
+    periods_per_day,
+)
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    get_unit_from_dtype,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pandas_datetime_to_datetimestruct,
+    pydate_to_dtstruct,
+)
+
+import_pandas_datetime()
+
+from .dtypes cimport PeriodDtypeCode
+from .timedeltas cimport (
+    _Timedelta,
+    delta_to_nanoseconds,
+    is_any_td_scalar,
+)
+
+from .timedeltas import Timedelta
+
+from .timestamps cimport _Timestamp
+
+from .timestamps import Timestamp
+
+# ---------------------------------------------------------------------
+# Misc Helpers
+
+cdef bint is_offset_object(object obj):
+    return isinstance(obj, BaseOffset)
+
+
+cdef bint is_tick_object(object obj):
+    return isinstance(obj, Tick)
+
+
+cdef bint _is_normalized(datetime dt):
+    if dt.hour != 0 or dt.minute != 0 or dt.second != 0 or dt.microsecond != 0:
+        # Regardless of whether dt is datetime vs Timestamp
+        return False
+    if isinstance(dt, _Timestamp):
+        return dt.nanosecond == 0
+    return True
+
+
+def apply_wraps(func):
+    # Note: normally we would use `@functools.wraps(func)`, but this does
+    # not play nicely with cython class methods
+
+    def wrapper(self, other):
+
+        if other is NaT:
+            return NaT
+        elif (
+            isinstance(other, BaseOffset)
+            or PyDelta_Check(other)
+            or cnp.is_timedelta64_object(other)
+        ):
+            # timedelta path
+            return func(self, other)
+        elif cnp.is_datetime64_object(other) or PyDate_Check(other):
+            # PyDate_Check includes date, datetime
+            other = Timestamp(other)
+        else:
+            # This will end up returning NotImplemented back in __add__
+            raise ApplyTypeError
+
+        tz = other.tzinfo
+        nano = other.nanosecond
+
+        if self._adjust_dst:
+            other = other.tz_localize(None)
+
+        result = func(self, other)
+
+        result2 = Timestamp(result).as_unit(other.unit)
+        if result == result2:
+            # i.e. the conversion is non-lossy, not the case for e.g.
+            #  test_milliseconds_combination
+            result = result2
+
+        if self._adjust_dst:
+            result = result.tz_localize(tz)
+
+        if self.normalize:
+            result = result.normalize()
+
+        # If the offset object does not have a nanoseconds component,
+        # the result's nanosecond component may be lost.
+        if not self.normalize and nano != 0 and not hasattr(self, "nanoseconds"):
+            if result.nanosecond != nano:
+                if result.tz is not None:
+                    # convert to UTC
+                    res = result.tz_localize(None)
+                else:
+                    res = result
+                value = res.as_unit("ns")._value
+                result = Timestamp(value + nano)
+
+        if tz is not None and result.tzinfo is None:
+            result = result.tz_localize(tz)
+
+        return result
+
+    # do @functools.wraps(func) manually since it doesn't work on cdef funcs
+    wrapper.__name__ = func.__name__
+    wrapper.__doc__ = func.__doc__
+    return wrapper
+
+
+cdef _wrap_timedelta_result(result):
+    """
+    Tick operations dispatch to their Timedelta counterparts.  Wrap the result
+    of these operations in a Tick if possible.
+
+    Parameters
+    ----------
+    result : object
+
+    Returns
+    -------
+    object
+    """
+    if PyDelta_Check(result):
+        # convert Timedelta back to a Tick
+        return delta_to_tick(result)
+
+    return result
+
+# ---------------------------------------------------------------------
+# Business Helpers
+
+
+cdef _get_calendar(weekmask, holidays, calendar):
+    """
+    Generate busdaycalendar
+    """
+    if isinstance(calendar, np.busdaycalendar):
+        if not holidays:
+            holidays = tuple(calendar.holidays)
+        elif not isinstance(holidays, tuple):
+            holidays = tuple(holidays)
+        else:
+            # trust that calendar.holidays and holidays are
+            # consistent
+            pass
+        return calendar, holidays
+
+    if holidays is None:
+        holidays = []
+    try:
+        holidays = holidays + calendar.holidays().tolist()
+    except AttributeError:
+        pass
+    holidays = tuple(sorted(_to_dt64D(dt) for dt in holidays))
+
+    kwargs = {"weekmask": weekmask}
+    if holidays:
+        kwargs["holidays"] = holidays
+
+    busdaycalendar = np.busdaycalendar(**kwargs)
+    return busdaycalendar, holidays
+
+
+cdef _to_dt64D(dt):
+    # Currently
+    # > np.datetime64(dt.datetime(2013,5,1),dtype='datetime64[D]')
+    # numpy.datetime64('2013-05-01T02:00:00.000000+0200')
+    # Thus astype is needed to cast datetime to datetime64[D]
+    if getattr(dt, "tzinfo", None) is not None:
+        # Get the nanosecond timestamp,
+        #  equiv `Timestamp(dt).value` or `dt.timestamp() * 10**9`
+        # The `naive` must be the `dt` naive wall time
+        #  instead of the naive absolute time (GH#49441)
+        naive = dt.replace(tzinfo=None)
+        dt = np.datetime64(naive, "D")
+    else:
+        dt = np.datetime64(dt)
+    if dt.dtype.name != "datetime64[D]":
+        dt = dt.astype("datetime64[D]")
+    return dt
+
+
+# ---------------------------------------------------------------------
+# Validation
+
+
+cdef _validate_business_time(t_input):
+    if isinstance(t_input, str):
+        try:
+            t = time.strptime(t_input, "%H:%M")
+            return dt_time(hour=t.tm_hour, minute=t.tm_min)
+        except ValueError:
+            raise ValueError("time data must match '%H:%M' format")
+    elif isinstance(t_input, dt_time):
+        if t_input.second != 0 or t_input.microsecond != 0:
+            raise ValueError(
+                "time data must be specified only with hour and minute")
+        return t_input
+    else:
+        raise ValueError("time data must be string or datetime.time")
+
+
+# ---------------------------------------------------------------------
+# Constructor Helpers
+
+_relativedelta_kwds = {"years", "months", "weeks", "days", "year", "month",
+                       "day", "weekday", "hour", "minute", "second",
+                       "microsecond", "millisecond", "nanosecond",
+                       "nanoseconds", "hours", "minutes", "seconds",
+                       "milliseconds", "microseconds"}
+
+
+cdef _determine_offset(kwds):
+    if not kwds:
+        # GH 45643/45890: (historically) defaults to 1 day
+        return timedelta(days=1), False
+
+    if "millisecond" in kwds:
+        raise NotImplementedError(
+            "Using DateOffset to replace `millisecond` component in "
+            "datetime object is not supported. Use "
+            "`microsecond=timestamp.microsecond % 1000 + ms * 1000` "
+            "instead."
+        )
+
+    nanos = {"nanosecond", "nanoseconds"}
+
+    # nanos are handled by apply_wraps
+    if all(k in nanos for k in kwds):
+        return timedelta(days=0), False
+
+    kwds_no_nanos = {k: v for k, v in kwds.items() if k not in nanos}
+
+    kwds_use_relativedelta = {
+        "year", "month", "day", "hour", "minute",
+        "second", "microsecond", "weekday", "years", "months", "weeks", "days",
+        "hours", "minutes", "seconds", "microseconds"
+    }
+
+    # "weeks" and "days" are left out despite being valid args for timedelta,
+    # because (historically) timedelta is used only for sub-daily.
+    kwds_use_timedelta = {
+        "seconds", "microseconds", "milliseconds", "minutes", "hours",
+    }
+
+    if all(k in kwds_use_timedelta for k in kwds_no_nanos):
+        # Sub-daily offset - use timedelta (tz-aware)
+        # This also handles "milliseconds" (plur): see GH 49897
+        return timedelta(**kwds_no_nanos), False
+
+    # convert milliseconds to microseconds, so relativedelta can parse it
+    if "milliseconds" in kwds_no_nanos:
+        micro = kwds_no_nanos.pop("milliseconds") * 1000
+        kwds_no_nanos["microseconds"] = kwds_no_nanos.get("microseconds", 0) + micro
+
+    if all(k in kwds_use_relativedelta for k in kwds_no_nanos):
+        from dateutil.relativedelta import relativedelta
+
+        return relativedelta(**kwds_no_nanos), True
+
+    raise ValueError(
+        f"Invalid argument/s or bad combination of arguments: {list(kwds.keys())}"
+    )
+
+# ---------------------------------------------------------------------
+# Mixins & Singletons
+
+
+class ApplyTypeError(TypeError):
+    # sentinel class for catching the apply error to return NotImplemented
+    pass
+
+
+# ---------------------------------------------------------------------
+# Base Classes
+
+cdef class BaseOffset:
+    """
+    Base class for DateOffset methods that are not overridden by subclasses.
+
+    Parameters
+    ----------
+    n : int
+        Number of multiples of the frequency.
+
+    normalize : bool
+        Whether the frequency can align with midnight.
+
+    Examples
+    --------
+    >>> pd.offsets.Hour(5).n
+    5
+    >>> pd.offsets.Hour(5).normalize
+    False
+    """
+    # ensure that reversed-ops with numpy scalars return NotImplemented
+    __array_priority__ = 1000
+
+    _day_opt = None
+    _attributes = tuple(["n", "normalize"])
+    _use_relativedelta = False
+    _adjust_dst = True
+
+    # cdef readonly:
+    #    int64_t n
+    #    bint normalize
+    #    dict _cache
+
+    def __init__(self, n=1, normalize=False):
+        n = self._validate_n(n)
+        self.n = n
+        self.normalize = normalize
+        self._cache = {}
+
+    def __eq__(self, other) -> bool:
+        if isinstance(other, str):
+            try:
+                # GH#23524 if to_offset fails, we are dealing with an
+                #  incomparable type so == is False and != is True
+                other = to_offset(other)
+            except ValueError:
+                # e.g. "infer"
+                return False
+        try:
+            return self._params == other._params
+        except AttributeError:
+            # other is not a DateOffset object
+            return False
+
+    def __ne__(self, other):
+        return not self == other
+
+    def __hash__(self) -> int:
+        return hash(self._params)
+
+    @cache_readonly
+    def _params(self):
+        """
+        Returns a tuple containing all of the attributes needed to evaluate
+        equality between two DateOffset objects.
+        """
+        d = getattr(self, "__dict__", {})
+        all_paras = d.copy()
+        all_paras["n"] = self.n
+        all_paras["normalize"] = self.normalize
+        for attr in self._attributes:
+            if hasattr(self, attr) and attr not in d:
+                # cython attributes are not in __dict__
+                all_paras[attr] = getattr(self, attr)
+
+        if "holidays" in all_paras and not all_paras["holidays"]:
+            all_paras.pop("holidays")
+        exclude = {"kwds", "name", "calendar"}
+        attrs = {(k, v) for k, v in all_paras.items()
+                 if (k not in exclude) and (k[0] != "_")}
+        params = tuple([str(type(self))] + sorted(attrs))
+        return params
+
+    @property
+    def kwds(self) -> dict:
+        """
+        Return a dict of extra parameters for the offset.
+
+        See Also
+        --------
+        tseries.offsets.DateOffset : The base class for all pandas date offsets.
+        tseries.offsets.WeekOfMonth : Represents the week of the month.
+        tseries.offsets.LastWeekOfMonth : Represents the last week of the month.
+
+        Examples
+        --------
+        >>> pd.DateOffset(5).kwds
+        {}
+
+        >>> pd.offsets.FY5253Quarter().kwds
+        {'weekday': 0,
+         'startingMonth': 1,
+         'qtr_with_extra_week': 1,
+         'variation': 'nearest'}
+        """
+        # for backwards-compatibility
+        kwds = {name: getattr(self, name, None) for name in self._attributes
+                if name not in ["n", "normalize"]}
+        return {name: kwds[name] for name in kwds if kwds[name] is not None}
+
+    @property
+    def base(self):
+        """
+        Returns a copy of the calling offset object with n=1 and all other
+        attributes equal.
+        """
+        return type(self)(n=1, normalize=self.normalize, **self.kwds)
+
+    def __add__(self, other):
+        if util.is_array(other) and other.dtype == object:
+            return np.array([self + x for x in other])
+
+        try:
+            return self._apply(other)
+        except ApplyTypeError:
+            return NotImplemented
+
+    def __radd__(self, other):
+        return self.__add__(other)
+
+    def __sub__(self, other):
+        if PyDateTime_Check(other):
+            raise TypeError("Cannot subtract datetime from offset.")
+        elif type(other) is type(self):
+            return type(self)(self.n - other.n, normalize=self.normalize,
+                              **self.kwds)
+        else:
+            # e.g. PeriodIndex
+            return NotImplemented
+
+    def __rsub__(self, other):
+        return (-self).__add__(other)
+
+    def __mul__(self, other):
+        if util.is_array(other):
+            return np.array([self * x for x in other])
+        elif is_integer_object(other):
+            return type(self)(n=other * self.n, normalize=self.normalize,
+                              **self.kwds)
+        elif isinstance(other, BaseOffset):
+            # Otherwise raises RecurrsionError due to __rmul__
+            raise TypeError(
+                f"Cannot multiply {type(self).__name__} with "
+                f"{type(other).__name__}."
+            )
+        return NotImplemented
+
+    def __rmul__(self, other):
+        return self.__mul__(other)
+
+    def __neg__(self):
+        # Note: we are deferring directly to __mul__ instead of __rmul__, as
+        # that allows us to use methods that can go in a `cdef class`
+        return self * -1
+
+    def copy(self):
+        # Note: we are deferring directly to __mul__ instead of __rmul__, as
+        # that allows us to use methods that can go in a `cdef class`
+        """
+        Return a copy of the frequency.
+
+        See Also
+        --------
+        tseries.offsets.Week.copy : Return a copy of Week offset.
+        tseries.offsets.DateOffset.copy : Return a copy of date offset.
+        tseries.offsets.MonthEnd.copy : Return a copy of MonthEnd offset.
+        tseries.offsets.YearBegin.copy : Return a copy of YearBegin offset.
+
+        Examples
+        --------
+        >>> freq = pd.DateOffset(1)
+        >>> freq_copy = freq.copy()
+        >>> freq is freq_copy
+        False
+        """
+        return self * 1
+
+    # ------------------------------------------------------------------
+    # Name and Rendering Methods
+
+    def __repr__(self) -> str:
+        # _output_name used by B(Year|Quarter)(End|Begin) to
+        #  expand "B" -> "Business"
+        class_name = getattr(self, "_output_name", type(self).__name__)
+
+        if abs(self.n) != 1:
+            plural = "s"
+        else:
+            plural = ""
+
+        n_str = ""
+        if self.n != 1:
+            n_str = f"{self.n} * "
+
+        out = f"<{n_str}{class_name}{plural}{self._repr_attrs()}>"
+        return out
+
+    def _repr_attrs(self) -> str:
+        exclude = {"n", "inc", "normalize"}
+        attrs = []
+        for attr in sorted(self._attributes):
+            # _attributes instead of __dict__ because cython attrs are not in __dict__
+            if attr.startswith("_") or attr == "kwds" or not hasattr(self, attr):
+                # DateOffset may not have some of these attributes
+                continue
+            elif attr not in exclude:
+                value = getattr(self, attr)
+                attrs.append(f"{attr}={value}")
+
+        out = ""
+        if attrs:
+            out += ": " + ", ".join(attrs)
+        return out
+
+    @property
+    def name(self) -> str:
+        """
+        Return a string representing the base frequency.
+
+        See Also
+        --------
+        tseries.offsets.Week : Represents a weekly offset.
+        DateOffset : Base class for all other offset classes.
+        tseries.offsets.Day : Represents a single day offset.
+        tseries.offsets.MonthEnd : Represents a monthly offset that
+            snaps to the end of the month.
+
+        Examples
+        --------
+        >>> pd.offsets.Hour().name
+        'h'
+
+        >>> pd.offsets.Hour(5).name
+        'h'
+        """
+        return self.rule_code
+
+    @property
+    def _prefix(self) -> str:
+        raise NotImplementedError("Prefix not defined")
+
+    @property
+    def rule_code(self) -> str:
+        """
+        Return a string representing the base frequency.
+
+        See Also
+        --------
+        tseries.offsets.Hour.rule_code :
+            Returns a string representing the base frequency of 'h'.
+        tseries.offsets.Day.rule_code :
+            Returns a string representing the base frequency of 'D'.
+
+        Examples
+        --------
+        >>> pd.offsets.Hour().rule_code
+        'h'
+
+        >>> pd.offsets.Week(5).rule_code
+        'W'
+        """
+        return self._prefix
+
+    @cache_readonly
+    def freqstr(self) -> str:
+        """
+        Return a string representing the frequency.
+
+        See Also
+        --------
+        tseries.offsets.BusinessDay.freqstr :
+            Return a string representing an offset frequency in Business Days.
+        tseries.offsets.BusinessHour.freqstr :
+            Return a string representing an offset frequency in Business Hours.
+        tseries.offsets.Week.freqstr :
+            Return a string representing an offset frequency in Weeks.
+        tseries.offsets.Hour.freqstr :
+            Return a string representing an offset frequency in Hours.
+
+        Examples
+        --------
+        >>> pd.DateOffset(5).freqstr
+        '<5 * DateOffsets>'
+
+        >>> pd.offsets.BusinessHour(2).freqstr
+        '2bh'
+
+        >>> pd.offsets.Nano().freqstr
+        'ns'
+
+        >>> pd.offsets.Nano(-3).freqstr
+        '-3ns'
+        """
+        try:
+            code = self.rule_code
+        except NotImplementedError:
+            return str(repr(self))
+
+        if self.n != 1:
+            fstr = f"{self.n}{code}"
+        else:
+            fstr = code
+
+        try:
+            if self._offset:
+                fstr += self._offset_str()
+        except AttributeError:
+            # TODO: standardize `_offset` vs `offset` naming convention
+            pass
+
+        return fstr
+
+    def _offset_str(self) -> str:
+        return ""
+
+    # ------------------------------------------------------------------
+
+    def _apply(self, other):
+        raise NotImplementedError("implemented by subclasses")
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        # NB: _apply_array does not handle respecting `self.normalize`, the
+        #  caller (DatetimeArray) handles that in post-processing.
+        raise NotImplementedError(
+            f"DateOffset subclass {type(self).__name__} "
+            "does not have a vectorized implementation"
+        )
+
+    def rollback(self, dt) -> datetime:
+        """
+        Roll provided date backward to next offset only if not on offset.
+
+        Returns
+        -------
+        TimeStamp
+            Rolled timestamp if not on offset, otherwise unchanged timestamp.
+        """
+        dt = Timestamp(dt)
+        if not self.is_on_offset(dt):
+            dt = dt - type(self)(1, normalize=self.normalize, **self.kwds)
+        return dt
+
+    def rollforward(self, dt) -> datetime:
+        """
+        Roll provided date forward to next offset only if not on offset.
+
+        Returns
+        -------
+        TimeStamp
+            Rolled timestamp if not on offset, otherwise unchanged timestamp.
+        """
+        dt = Timestamp(dt)
+        if not self.is_on_offset(dt):
+            dt = dt + type(self)(1, normalize=self.normalize, **self.kwds)
+        return dt
+
+    def _get_offset_day(self, other: datetime) -> int:
+        # subclass must implement `_day_opt`; calling from the base class
+        # will implicitly assume day_opt = "business_end", see get_day_of_month.
+        cdef:
+            npy_datetimestruct dts
+        pydate_to_dtstruct(other, &dts)
+        return get_day_of_month(&dts, self._day_opt)
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        """
+        Return boolean whether a timestamp intersects with this frequency.
+
+        This method determines if a given timestamp aligns with the start
+        of a custom business month, as defined by this offset. It accounts
+        for custom rules, such as skipping weekends or other non-business days,
+        and checks whether the provided datetime falls on a valid business day
+        that marks the beginning of the custom business month.
+
+        Parameters
+        ----------
+        dt : datetime.datetime
+            Timestamp to check intersections with frequency.
+
+        See Also
+        --------
+        tseries.offsets.CustomBusinessMonthBegin : Represents the start of a custom
+            business month.
+        tseries.offsets.CustomBusinessMonthEnd : Represents the end of a custom
+            business month.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Day(1)
+        >>> freq.is_on_offset(ts)
+        True
+
+        >>> ts = pd.Timestamp(2022, 8, 6)
+        >>> ts.day_name()
+        'Saturday'
+        >>> freq = pd.offsets.BusinessDay(1)
+        >>> freq.is_on_offset(ts)
+        False
+        """
+        if self.normalize and not _is_normalized(dt):
+            return False
+
+        # Default (slow) method for determining if some date is a member of the
+        # date range generated by this offset. Subclasses may have this
+        # re-implemented in a nicer way.
+        a = dt
+        b = (dt + self) - self
+        return a == b
+
+    # ------------------------------------------------------------------
+
+    # Staticmethod so we can call from Tick.__init__, will be unnecessary
+    #  once BaseOffset is a cdef class and is inherited by Tick
+    @staticmethod
+    def _validate_n(n) -> int:
+        """
+        Require that `n` be an integer.
+
+        Parameters
+        ----------
+        n : int
+
+        Returns
+        -------
+        nint : int
+
+        Raises
+        ------
+        TypeError if `int(n)` raises
+        ValueError if n != int(n)
+        """
+        if cnp.is_timedelta64_object(n):
+            raise TypeError(f"`n` argument must be an integer, got {type(n)}")
+        try:
+            nint = int(n)
+        except (ValueError, TypeError):
+            raise TypeError(f"`n` argument must be an integer, got {type(n)}")
+        if n != nint:
+            raise ValueError(f"`n` argument must be an integer, got {n}")
+        return nint
+
+    def __setstate__(self, state):
+        """
+        Reconstruct an instance from a pickled state
+        """
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self._cache = state.pop("_cache", {})
+        # At this point we expect state to be empty
+
+    def __getstate__(self):
+        """
+        Return a picklable state
+        """
+        state = {}
+        state["n"] = self.n
+        state["normalize"] = self.normalize
+
+        # we don't want to actually pickle the calendar object
+        # as its a np.busyday; we recreate on deserialization
+        state.pop("calendar", None)
+        if "kwds" in state:
+            state["kwds"].pop("calendar", None)
+
+        return state
+
+    @property
+    def nanos(self):
+        """
+        Returns a integer of the total number of nanoseconds for fixed frequencies.
+
+        Raises
+        ------
+        ValueError
+            If the frequency is non-fixed.
+
+        See Also
+        --------
+        tseries.offsets.Hour.nanos :
+            Returns an integer of the total number of nanoseconds.
+        tseries.offsets.Day.nanos :
+            Returns an integer of the total number of nanoseconds.
+
+        Examples
+        --------
+        >>> pd.offsets.Week(n=1).nanos
+        ValueError: Week: weekday=None is a non-fixed frequency
+        """
+        raise ValueError(f"{self} is a non-fixed frequency")
+
+    # ------------------------------------------------------------------
+
+    def is_month_start(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the month start.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_month_end : Return boolean whether a timestamp occurs on the month end.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_month_start(ts)
+        True
+        """
+        return ts._get_start_end_field("is_month_start", self)
+
+    def is_month_end(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the month end.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_month_start : Return boolean whether a timestamp occurs on the month start.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_month_end(ts)
+        False
+        """
+        return ts._get_start_end_field("is_month_end", self)
+
+    def is_quarter_start(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the quarter start.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_quarter_end : Return boolean whether a timestamp occurs on the quarter end.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_quarter_start(ts)
+        True
+        """
+        return ts._get_start_end_field("is_quarter_start", self)
+
+    def is_quarter_end(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the quarter end.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_quarter_start : Return boolean whether a timestamp
+            occurs on the quarter start.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_quarter_end(ts)
+        False
+        """
+        return ts._get_start_end_field("is_quarter_end", self)
+
+    def is_year_start(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the year start.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_year_end : Return boolean whether a timestamp occurs on the year end.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_year_start(ts)
+        True
+        """
+        return ts._get_start_end_field("is_year_start", self)
+
+    def is_year_end(self, _Timestamp ts):
+        """
+        Return boolean whether a timestamp occurs on the year end.
+
+        Parameters
+        ----------
+        ts : Timestamp
+            The timestamp to check.
+
+        See Also
+        --------
+        is_year_start : Return boolean whether a timestamp occurs on the year start.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2022, 1, 1)
+        >>> freq = pd.offsets.Hour(5)
+        >>> freq.is_year_end(ts)
+        False
+        """
+        return ts._get_start_end_field("is_year_end", self)
+
+
+cdef class SingleConstructorOffset(BaseOffset):
+    @classmethod
+    def _from_name(cls, suffix=None):
+        # default _from_name calls cls with no args
+        if suffix:
+            raise ValueError(f"Bad freq suffix {suffix}")
+        return cls()
+
+    def __reduce__(self):
+        # This __reduce__ implementation is for all BaseOffset subclasses
+        #  except for RelativeDeltaOffset
+        # np.busdaycalendar objects do not pickle nicely, but we can reconstruct
+        #  from attributes that do get pickled.
+        tup = tuple(
+            getattr(self, attr) if attr != "calendar" else None
+            for attr in self._attributes
+        )
+        return type(self), tup
+
+
+# ---------------------------------------------------------------------
+# Tick Offsets
+
+cdef class Tick(SingleConstructorOffset):
+    _adjust_dst = False
+    _prefix = "undefined"
+    _attributes = tuple(["n", "normalize"])
+
+    def __init__(self, n=1, normalize=False):
+        n = self._validate_n(n)
+        self.n = n
+        self.normalize = False
+        self._cache = {}
+        if normalize:
+            # GH#21427
+            raise ValueError(
+                "Tick offset with `normalize=True` are not allowed."
+            )
+
+    # Note: Without making this cpdef, we get AttributeError when calling
+    #  from __mul__
+    cpdef Tick _next_higher_resolution(Tick self):
+        if type(self) is Day:
+            return Hour(self.n * 24)
+        if type(self) is Hour:
+            return Minute(self.n * 60)
+        if type(self) is Minute:
+            return Second(self.n * 60)
+        if type(self) is Second:
+            return Milli(self.n * 1000)
+        if type(self) is Milli:
+            return Micro(self.n * 1000)
+        if type(self) is Micro:
+            return Nano(self.n * 1000)
+        raise ValueError("Could not convert to integer offset at any resolution")
+
+    # --------------------------------------------------------------------
+
+    def _repr_attrs(self) -> str:
+        # Since cdef classes have no __dict__, we need to override
+        return ""
+
+    @cache_readonly
+    def _as_pd_timedelta(self):
+        return Timedelta(self)
+
+    @property
+    def nanos(self) -> int64_t:
+        """
+        Returns an integer of the total number of nanoseconds.
+
+        See Also
+        --------
+        tseries.offsets.Hour.nanos :
+            Returns an integer of the total number of nanoseconds.
+        tseries.offsets.Day.nanos :
+            Returns an integer of the total number of nanoseconds.
+
+        Examples
+        --------
+        >>> pd.offsets.Hour(5).nanos
+        18000000000000
+        """
+        return self.n * self._nanos_inc
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        return True
+
+    # This is identical to BaseOffset.__hash__, but has to be redefined here
+    # for Python 3, because we've redefined __eq__.
+    def __hash__(self) -> int:
+        return hash(self._params)
+
+    # --------------------------------------------------------------------
+    # Comparison and Arithmetic Methods
+
+    def __eq__(self, other):
+        if isinstance(other, str):
+            try:
+                # GH#23524 if to_offset fails, we are dealing with an
+                #  incomparable type so == is False and != is True
+                other = to_offset(other)
+            except ValueError:
+                # e.g. "infer"
+                return False
+        return self._as_pd_timedelta == other
+
+    def __ne__(self, other):
+        return not (self == other)
+
+    def __le__(self, other):
+        return self._as_pd_timedelta.__le__(other)
+
+    def __lt__(self, other):
+        return self._as_pd_timedelta.__lt__(other)
+
+    def __ge__(self, other):
+        return self._as_pd_timedelta.__ge__(other)
+
+    def __gt__(self, other):
+        return self._as_pd_timedelta.__gt__(other)
+
+    def __mul__(self, other):
+        if is_float_object(other):
+            n = other * self.n
+            # If the new `n` is an integer, we can represent it using the
+            #  same Tick subclass as self, otherwise we need to move up
+            #  to a higher-resolution subclass
+            if np.isclose(n % 1, 0):
+                return type(self)(int(n))
+            new_self = self._next_higher_resolution()
+            return new_self * other
+        return BaseOffset.__mul__(self, other)
+
+    def __rmul__(self, other):
+        return self.__mul__(other)
+
+    def __truediv__(self, other):
+        if not isinstance(self, Tick):
+            # cython semantics mean the args are sometimes swapped
+            result = other._as_pd_timedelta.__rtruediv__(self)
+        else:
+            result = self._as_pd_timedelta.__truediv__(other)
+        return _wrap_timedelta_result(result)
+
+    def __rtruediv__(self, other):
+        result = self._as_pd_timedelta.__rtruediv__(other)
+        return _wrap_timedelta_result(result)
+
+    def __add__(self, other):
+        if isinstance(other, Tick):
+            if type(self) is type(other):
+                return type(self)(self.n + other.n)
+            else:
+                return delta_to_tick(self._as_pd_timedelta + other._as_pd_timedelta)
+        try:
+            return self._apply(other)
+        except ApplyTypeError:
+            # Includes pd.Period
+            return NotImplemented
+        except OverflowError as err:
+            raise OverflowError(
+                f"the add operation between {self} and {other} will overflow"
+            ) from err
+
+    def __radd__(self, other):
+        return self.__add__(other)
+
+    def _apply(self, other):
+        # Timestamp can handle tz and nano sec, thus no need to use apply_wraps
+        if isinstance(other, _Timestamp):
+            # GH#15126
+            return other + self._as_pd_timedelta
+        elif other is NaT:
+            return NaT
+        elif cnp.is_datetime64_object(other) or PyDate_Check(other):
+            # PyDate_Check includes date, datetime
+            return Timestamp(other) + self
+
+        if cnp.is_timedelta64_object(other) or PyDelta_Check(other):
+            return other + self._as_pd_timedelta
+
+        raise ApplyTypeError(f"Unhandled type: {type(other).__name__}")
+
+    # --------------------------------------------------------------------
+    # Pickle Methods
+
+    def __setstate__(self, state):
+        self.n = state["n"]
+        self.normalize = False
+
+
+cdef class Day(Tick):
+    """
+    Offset ``n`` days.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of days represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n days.
+
+    >>> from pandas.tseries.offsets import Day
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Day()
+    Timestamp('2022-12-10 15:00:00')
+    >>> ts - Day(4)
+    Timestamp('2022-12-05 15:00:00')
+
+    >>> ts + Day(-4)
+    Timestamp('2022-12-05 15:00:00')
+    """
+    _nanos_inc = 24 * 3600 * 1_000_000_000
+    _prefix = "D"
+    _period_dtype_code = PeriodDtypeCode.D
+    _creso = NPY_DATETIMEUNIT.NPY_FR_D
+
+
+cdef class Hour(Tick):
+    """
+    Offset ``n`` hours.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of hours represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n hours.
+
+    >>> from pandas.tseries.offsets import Hour
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Hour()
+    Timestamp('2022-12-09 16:00:00')
+    >>> ts - Hour(4)
+    Timestamp('2022-12-09 11:00:00')
+
+    >>> ts + Hour(-4)
+    Timestamp('2022-12-09 11:00:00')
+    """
+    _nanos_inc = 3600 * 1_000_000_000
+    _prefix = "h"
+    _period_dtype_code = PeriodDtypeCode.H
+    _creso = NPY_DATETIMEUNIT.NPY_FR_h
+
+
+cdef class Minute(Tick):
+    """
+    Offset ``n`` minutes.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of minutes represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n minutes.
+
+    >>> from pandas.tseries.offsets import Minute
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Minute(n=10)
+    Timestamp('2022-12-09 15:10:00')
+    >>> ts - Minute(n=10)
+    Timestamp('2022-12-09 14:50:00')
+
+    >>> ts + Minute(n=-10)
+    Timestamp('2022-12-09 14:50:00')
+    """
+    _nanos_inc = 60 * 1_000_000_000
+    _prefix = "min"
+    _period_dtype_code = PeriodDtypeCode.T
+    _creso = NPY_DATETIMEUNIT.NPY_FR_m
+
+
+cdef class Second(Tick):
+    """
+    Offset ``n`` seconds.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of seconds represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n seconds.
+
+    >>> from pandas.tseries.offsets import Second
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Second(n=10)
+    Timestamp('2022-12-09 15:00:10')
+    >>> ts - Second(n=10)
+    Timestamp('2022-12-09 14:59:50')
+
+    >>> ts + Second(n=-10)
+    Timestamp('2022-12-09 14:59:50')
+    """
+    _nanos_inc = 1_000_000_000
+    _prefix = "s"
+    _period_dtype_code = PeriodDtypeCode.S
+    _creso = NPY_DATETIMEUNIT.NPY_FR_s
+
+
+cdef class Milli(Tick):
+    """
+    Offset ``n`` milliseconds.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of milliseconds represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n milliseconds.
+
+    >>> from pandas.tseries.offsets import Milli
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Milli(n=10)
+    Timestamp('2022-12-09 15:00:00.010000')
+
+    >>> ts - Milli(n=10)
+    Timestamp('2022-12-09 14:59:59.990000')
+
+    >>> ts + Milli(n=-10)
+    Timestamp('2022-12-09 14:59:59.990000')
+    """
+    _nanos_inc = 1_000_000
+    _prefix = "ms"
+    _period_dtype_code = PeriodDtypeCode.L
+    _creso = NPY_DATETIMEUNIT.NPY_FR_ms
+
+
+cdef class Micro(Tick):
+    """
+    Offset ``n`` microseconds.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of microseconds represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n microseconds.
+
+    >>> from pandas.tseries.offsets import Micro
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Micro(n=1000)
+    Timestamp('2022-12-09 15:00:00.001000')
+
+    >>> ts - Micro(n=1000)
+    Timestamp('2022-12-09 14:59:59.999000')
+
+    >>> ts + Micro(n=-1000)
+    Timestamp('2022-12-09 14:59:59.999000')
+    """
+    _nanos_inc = 1000
+    _prefix = "us"
+    _period_dtype_code = PeriodDtypeCode.U
+    _creso = NPY_DATETIMEUNIT.NPY_FR_us
+
+
+cdef class Nano(Tick):
+    """
+    Offset ``n`` nanoseconds.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of nanoseconds represented.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n nanoseconds.
+
+    >>> from pandas.tseries.offsets import Nano
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts
+    Timestamp('2022-12-09 15:00:00')
+
+    >>> ts + Nano(n=1000)
+    Timestamp('2022-12-09 15:00:00.000001')
+
+    >>> ts - Nano(n=1000)
+    Timestamp('2022-12-09 14:59:59.999999')
+
+    >>> ts + Nano(n=-1000)
+    Timestamp('2022-12-09 14:59:59.999999')
+    """
+    _nanos_inc = 1
+    _prefix = "ns"
+    _period_dtype_code = PeriodDtypeCode.N
+    _creso = NPY_DATETIMEUNIT.NPY_FR_ns
+
+
+def delta_to_tick(delta: timedelta) -> Tick:
+    if delta.microseconds == 0 and getattr(delta, "nanoseconds", 0) == 0:
+        # nanoseconds only for pd.Timedelta
+        if delta.seconds == 0:
+            return Day(delta.days)
+        else:
+            seconds = delta.days * 86400 + delta.seconds
+            if seconds % 3600 == 0:
+                return Hour(seconds / 3600)
+            elif seconds % 60 == 0:
+                return Minute(seconds / 60)
+            else:
+                return Second(seconds)
+    else:
+        nanos = delta_to_nanoseconds(delta)
+        if nanos % 1_000_000 == 0:
+            return Milli(nanos // 1_000_000)
+        elif nanos % 1000 == 0:
+            return Micro(nanos // 1000)
+        else:  # pragma: no cover
+            return Nano(nanos)
+
+
+# --------------------------------------------------------------------
+
+cdef class RelativeDeltaOffset(BaseOffset):
+    """
+    DateOffset subclass backed by a dateutil relativedelta object.
+    """
+    _attributes = tuple(["n", "normalize"] + list(_relativedelta_kwds))
+    _adjust_dst = False
+
+    def __init__(self, n=1, normalize=False, **kwds):
+        BaseOffset.__init__(self, n, normalize)
+        off, use_rd = _determine_offset(kwds)
+        object.__setattr__(self, "_offset", off)
+        object.__setattr__(self, "_use_relativedelta", use_rd)
+        for key in kwds:
+            val = kwds[key]
+            object.__setattr__(self, key, val)
+
+    def __getstate__(self):
+        """
+        Return a picklable state
+        """
+        # RelativeDeltaOffset (technically DateOffset) is the only non-cdef
+        #  class, so the only one with __dict__
+        state = self.__dict__.copy()
+        state["n"] = self.n
+        state["normalize"] = self.normalize
+        return state
+
+    def __setstate__(self, state):
+        """
+        Reconstruct an instance from a pickled state
+        """
+
+        if "offset" in state:
+            # Older (<0.22.0) versions have offset attribute instead of _offset
+            if "_offset" in state:  # pragma: no cover
+                raise AssertionError("Unexpected key `_offset`")
+            state["_offset"] = state.pop("offset")
+            state["kwds"]["offset"] = state["_offset"]
+
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self._cache = state.pop("_cache", {})
+
+        self.__dict__.update(state)
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        other_nanos = 0
+        if self._use_relativedelta:
+            if isinstance(other, _Timestamp):
+                other_nanos = other.nanosecond
+                other = other.to_pydatetime(warn=False)
+
+        if len(self.kwds) > 0:
+            tzinfo = getattr(other, "tzinfo", None)
+            if tzinfo is not None and self._use_relativedelta:
+                # perform calculation in UTC
+                other = other.replace(tzinfo=None)
+
+            other = other + (self._offset * self.n)
+
+            if hasattr(self, "nanoseconds"):
+                other = self.n * Timedelta(nanoseconds=self.nanoseconds) + other
+            if other_nanos != 0:
+                other = Timedelta(nanoseconds=other_nanos) + other
+
+            if tzinfo is not None and self._use_relativedelta:
+                # bring tz back from UTC calculation
+                other = localize_pydatetime(other, tzinfo)
+
+            return Timestamp(other)
+        else:
+            return other + timedelta(self.n)
+
+    @cache_readonly
+    def _pd_timedelta(self) -> Timedelta:
+        # components of _offset that can be cast to pd.Timedelta
+
+        kwds = self.kwds
+        relativedelta_fast = {
+            "years",
+            "months",
+            "weeks",
+            "days",
+            "hours",
+            "minutes",
+            "seconds",
+            "microseconds",
+            "milliseconds",
+        }
+        # relativedelta/_offset path only valid for base DateOffset
+        if self._use_relativedelta and set(kwds).issubset(relativedelta_fast):
+            td_args = {
+                "days",
+                "hours",
+                "minutes",
+                "seconds",
+                "microseconds",
+                "milliseconds"
+            }
+            td_kwds = {
+                key: val
+                for key, val in kwds.items()
+                if key in td_args
+            }
+            if "weeks" in kwds:
+                days = td_kwds.get("days", 0)
+                td_kwds["days"] = days + 7 * kwds["weeks"]
+
+            if td_kwds:
+                delta = Timedelta(**td_kwds)
+                if "microseconds" in kwds:
+                    delta = delta.as_unit("us")
+                elif "milliseconds" in kwds:
+                    delta = delta.as_unit("ms")
+                else:
+                    delta = delta.as_unit("s")
+            else:
+                delta = Timedelta(0).as_unit("s")
+
+            return delta * self.n
+
+        elif not self._use_relativedelta and hasattr(self, "_offset"):
+            # timedelta
+            num_nano = getattr(self, "nanoseconds", 0)
+            if num_nano != 0:
+                rem_nano = Timedelta(nanoseconds=num_nano)
+                delta = Timedelta((self._offset + rem_nano) * self.n)
+            else:
+                delta = Timedelta(self._offset * self.n)
+                if "microseconds" in kwds:
+                    delta = delta.as_unit("us")
+                elif "milliseconds" in kwds:
+                    delta = delta.as_unit("ms")
+                else:
+                    delta = delta.as_unit("s")
+            return delta
+
+        else:
+            # relativedelta with other keywords
+            kwd = set(kwds) - relativedelta_fast
+            raise NotImplementedError(
+                "DateOffset with relativedelta "
+                f"keyword(s) {kwd} not able to be "
+                "applied vectorized"
+            )
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        reso = get_unit_from_dtype(dtarr.dtype)
+        dt64other = np.asarray(dtarr)
+
+        delta = self._pd_timedelta  # may raise NotImplementedError
+
+        kwds = self.kwds
+        months = (kwds.get("years", 0) * 12 + kwds.get("months", 0)) * self.n
+        if months:
+            shifted = shift_months(dt64other.view("i8"), months, reso=reso)
+            dt64other = shifted.view(dtarr.dtype)
+        return dt64other + delta
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return True
+
+
+class OffsetMeta(type):
+    """
+    Metaclass that allows us to pretend that all BaseOffset subclasses
+    inherit from DateOffset (which is needed for backward-compatibility).
+    """
+
+    @classmethod
+    def __instancecheck__(cls, obj) -> bool:
+        return isinstance(obj, BaseOffset)
+
+    @classmethod
+    def __subclasscheck__(cls, obj) -> bool:
+        return issubclass(obj, BaseOffset)
+
+
+# TODO: figure out a way to use a metaclass with a cdef class
+class DateOffset(RelativeDeltaOffset, metaclass=OffsetMeta):
+    """
+    Standard kind of date increment used for a date range.
+
+    Works exactly like the keyword argument form of relativedelta.
+    Note that the positional argument form of relativedelta is not
+    supported. Use of the keyword n is discouraged-- you would be better
+    off specifying n in the keywords you use, but regardless it is
+    there for you. n is needed for DateOffset subclasses.
+
+    DateOffset works as follows.  Each offset specify a set of dates
+    that conform to the DateOffset.  For example, Bday defines this
+    set to be the set of dates that are weekdays (M-F).  To test if a
+    date is in the set of a DateOffset dateOffset we can use the
+    is_on_offset method: dateOffset.is_on_offset(date).
+
+    If a date is not on a valid date, the rollback and rollforward
+    methods can be used to roll the date to the nearest valid date
+    before/after the date.
+
+    DateOffsets can be created to move dates forward a given number of
+    valid dates.  For example, Bday(2) can be added to a date to move
+    it two business days forward.  If the date does not start on a
+    valid date, first it is moved to a valid date.  Thus pseudo code
+    is::
+
+        def __add__(date):
+          date = rollback(date) # does nothing if date is valid
+          return date + <n number of periods>
+
+    When a date offset is created for a negative number of periods,
+    the date is first rolled forward.  The pseudo code is::
+
+        def __add__(date):
+          date = rollforward(date) # does nothing if date is valid
+          return date + <n number of periods>
+
+    Zero presents a problem.  Should it roll forward or back?  We
+    arbitrarily have it rollforward:
+
+    date + BDay(0) == BDay.rollforward(date)
+
+    Since 0 is a bit weird, we suggest avoiding its use.
+
+    Besides, adding a DateOffsets specified by the singular form of the date
+    component can be used to replace certain component of the timestamp.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of time periods the offset represents.
+        If specified without a temporal pattern, defaults to n days.
+    normalize : bool, default False
+        Whether to round the result of a DateOffset addition down to the
+        previous midnight.
+    weekday : int {0, 1, ..., 6}, default 0
+
+        A specific integer for the day of the week.
+
+        - 0 is Monday
+        - 1 is Tuesday
+        - 2 is Wednesday
+        - 3 is Thursday
+        - 4 is Friday
+        - 5 is Saturday
+        - 6 is Sunday
+
+        Instead Weekday type from dateutil.relativedelta can be used.
+
+        - MO is Monday
+        - TU is Tuesday
+        - WE is Wednesday
+        - TH is Thursday
+        - FR is Friday
+        - SA is Saturday
+        - SU is Sunday.
+
+    **kwds
+        Temporal parameter that add to or replace the offset value.
+
+        Parameters that **add** to the offset (like Timedelta):
+
+        - years
+        - months
+        - weeks
+        - days
+        - hours
+        - minutes
+        - seconds
+        - milliseconds
+        - microseconds
+        - nanoseconds
+
+        Parameters that **replace** the offset value:
+
+        - year
+        - month
+        - day
+        - weekday
+        - hour
+        - minute
+        - second
+        - microsecond
+        - nanosecond.
+
+    See Also
+    --------
+    dateutil.relativedelta.relativedelta : The relativedelta type is designed
+        to be applied to an existing datetime an can replace specific components of
+        that datetime, or represents an interval of time.
+
+    Examples
+    --------
+    >>> from pandas.tseries.offsets import DateOffset
+    >>> ts = pd.Timestamp('2017-01-01 09:10:11')
+    >>> ts + DateOffset(months=3)
+    Timestamp('2017-04-01 09:10:11')
+
+    >>> ts = pd.Timestamp('2017-01-01 09:10:11')
+    >>> ts + DateOffset(months=2)
+    Timestamp('2017-03-01 09:10:11')
+    >>> ts + DateOffset(day=31)
+    Timestamp('2017-01-31 09:10:11')
+
+    >>> ts + pd.DateOffset(hour=8)
+    Timestamp('2017-01-01 08:10:11')
+    """
+    def __setattr__(self, name, value):
+        raise AttributeError("DateOffset objects are immutable.")
+
+# --------------------------------------------------------------------
+
+
+cdef class BusinessMixin(SingleConstructorOffset):
+    """
+    Mixin to business types to provide related functions.
+    """
+
+    cdef readonly:
+        timedelta _offset
+        # Only Custom subclasses use weekmask, holiday, calendar
+        object weekmask, holidays, calendar
+
+    def __init__(self, n=1, normalize=False, offset=timedelta(0)):
+        BaseOffset.__init__(self, n, normalize)
+        self._offset = offset
+
+    cpdef _init_custom(self, weekmask, holidays, calendar):
+        """
+        Additional __init__ for Custom subclasses.
+        """
+        calendar, holidays = _get_calendar(
+            weekmask=weekmask, holidays=holidays, calendar=calendar
+        )
+        # Custom offset instances are identified by the
+        # following two attributes. See DateOffset._params()
+        # holidays, weekmask
+        self.weekmask = weekmask
+        self.holidays = holidays
+        self.calendar = calendar
+
+    @property
+    def offset(self):
+        """
+        Alias for self._offset.
+        """
+        # Alias for backward compat
+        return self._offset
+
+    def _repr_attrs(self) -> str:
+        if self.offset:
+            attrs = [f"offset={repr(self.offset)}"]
+        else:
+            attrs = []
+        out = ""
+        if attrs:
+            out += ": " + ", ".join(attrs)
+        return out
+
+    cpdef __setstate__(self, state):
+        # We need to use a cdef/cpdef method to set the readonly _offset attribute
+        if "_offset" in state:
+            self._offset = state.pop("_offset")
+        elif "offset" in state:
+            # Older (<0.22.0) versions have offset attribute instead of _offset
+            self._offset = state.pop("offset")
+
+        if self._prefix.startswith(("C", "c")):
+            # i.e. this is a Custom class
+            weekmask = state.pop("weekmask")
+            holidays = state.pop("holidays")
+            calendar, holidays = _get_calendar(weekmask=weekmask,
+                                               holidays=holidays,
+                                               calendar=None)
+            self.weekmask = weekmask
+            self.calendar = calendar
+            self.holidays = holidays
+
+        BaseOffset.__setstate__(self, state)
+
+
+cdef class BusinessDay(BusinessMixin):
+    """
+    DateOffset subclass representing possibly n business days.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of days represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n business days.
+
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts.strftime('%a %d %b %Y %H:%M')
+    'Fri 09 Dec 2022 15:00'
+    >>> (ts + pd.offsets.BusinessDay(n=5)).strftime('%a %d %b %Y %H:%M')
+    'Fri 16 Dec 2022 15:00'
+
+    Passing the parameter ``normalize`` equal to True, you shift the start
+    of the next business day to midnight.
+
+    >>> ts = pd.Timestamp(2022, 12, 9, 15)
+    >>> ts + pd.offsets.BusinessDay(normalize=True)
+    Timestamp('2022-12-12 00:00:00')
+    """
+    _period_dtype_code = PeriodDtypeCode.B
+    _prefix = "B"
+    _attributes = tuple(["n", "normalize", "offset"])
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        if "_offset" in state:
+            self._offset = state.pop("_offset")
+        elif "offset" in state:
+            self._offset = state.pop("offset")
+        self._cache = state.pop("_cache", {})
+
+    def _offset_str(self) -> str:
+        def get_str(td):
+            off_str = ""
+            if td.days > 0:
+                off_str += str(td.days) + "D"
+            if td.seconds > 0:
+                s = td.seconds
+                hrs = int(s / 3600)
+                if hrs != 0:
+                    off_str += str(hrs) + "h"
+                    s -= hrs * 3600
+                mts = int(s / 60)
+                if mts != 0:
+                    off_str += str(mts) + "Min"
+                    s -= mts * 60
+                if s != 0:
+                    off_str += str(s) + "s"
+            if td.microseconds > 0:
+                off_str += str(td.microseconds) + "us"
+            return off_str
+
+        if PyDelta_Check(self.offset):
+            zero = timedelta(0, 0, 0)
+            if self.offset >= zero:
+                off_str = "+" + get_str(self.offset)
+            else:
+                off_str = "-" + get_str(-self.offset)
+            return off_str
+        else:
+            return "+" + repr(self.offset)
+
+    @apply_wraps
+    def _apply(self, other):
+        if PyDateTime_Check(other):
+            n = self.n
+            wday = other.weekday()
+
+            # avoid slowness below by operating on weeks first
+            weeks = n // 5
+            days = self._adjust_ndays(wday, weeks)
+
+            result = other + timedelta(days=7 * weeks + days)
+            if self.offset:
+                result = result + self.offset
+            return result
+
+        elif is_any_td_scalar(other):
+            td = Timedelta(self.offset) + other
+            return BusinessDay(
+                self.n, offset=td.to_pytimedelta(), normalize=self.normalize
+            )
+        else:
+            raise ApplyTypeError(
+                "Only know how to combine business day with datetime or timedelta."
+            )
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    cdef ndarray _shift_bdays(
+        self,
+        ndarray i8other,
+        NPY_DATETIMEUNIT reso=NPY_DATETIMEUNIT.NPY_FR_ns,
+    ):
+        """
+        Implementation of BusinessDay.apply_offset.
+
+        Parameters
+        ----------
+        i8other : const int64_t[:]
+        reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+        Returns
+        -------
+        ndarray[int64_t]
+        """
+        cdef:
+            int periods = self.n
+            Py_ssize_t i, n = i8other.size
+            ndarray result = cnp.PyArray_EMPTY(
+                i8other.ndim, i8other.shape, cnp.NPY_INT64, 0
+            )
+            int64_t val, res_val
+            int wday, days
+            npy_datetimestruct dts
+            int64_t DAY_PERIODS = periods_per_day(reso)
+            cnp.broadcast mi = cnp.PyArray_MultiIterNew2(result, i8other)
+
+        for i in range(n):
+            # Analogous to: val = i8other[i]
+            val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+            if val == NPY_NAT:
+                res_val = NPY_NAT
+            else:
+                # The rest of this is effectively a copy of BusinessDay.apply
+                weeks = periods // 5
+                pandas_datetime_to_datetimestruct(val, reso, &dts)
+                wday = dayofweek(dts.year, dts.month, dts.day)
+
+                days = self._adjust_ndays(wday, weeks)
+                res_val = val + (7 * weeks + days) * DAY_PERIODS
+
+            # Analogous to: out[i] = res_val
+            (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+            cnp.PyArray_MultiIter_NEXT(mi)
+
+        return result
+
+    cdef int _adjust_ndays(self, int wday, int weeks):
+        cdef:
+            int n = self.n
+            int days
+
+        if n <= 0 and wday > 4:
+            # roll forward
+            n += 1
+
+        n -= 5 * weeks
+
+        # n is always >= 0 at this point
+        if n == 0 and wday > 4:
+            # roll back
+            days = 4 - wday
+        elif wday > 4:
+            # roll forward
+            days = (7 - wday) + (n - 1)
+        elif wday + n <= 4:
+            # shift by n days without leaving the current week
+            days = n
+        else:
+            # shift by n days plus 2 to get past the weekend
+            days = n + 2
+        return days
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        i8other = dtarr.view("i8")
+        reso = get_unit_from_dtype(dtarr.dtype)
+        res = self._shift_bdays(i8other, reso=reso)
+        if self.offset:
+            res = res.view(dtarr.dtype) + Timedelta(self.offset)
+        return res
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return dt.weekday() < 5
+
+
+cdef class BusinessHour(BusinessMixin):
+    """
+    DateOffset subclass representing possibly n business hours.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of hours represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    start : str, time, or list of str/time, default "09:00"
+        Start time of your custom business hour in 24h format.
+    end : str, time, or list of str/time, default: "17:00"
+        End time of your custom business hour in 24h format.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    Examples
+    --------
+    You can use the parameter ``n`` to represent a shift of n hours.
+
+    >>> ts = pd.Timestamp(2022, 12, 9, 8)
+    >>> ts + pd.offsets.BusinessHour(n=5)
+    Timestamp('2022-12-09 14:00:00')
+
+    You can also change the start and the end of business hours.
+
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.BusinessHour(start="11:00")
+    Timestamp('2022-08-08 11:00:00')
+
+    >>> from datetime import time as dt_time
+    >>> ts = pd.Timestamp(2022, 8, 5, 22)
+    >>> ts + pd.offsets.BusinessHour(end=dt_time(19, 0))
+    Timestamp('2022-08-08 10:00:00')
+
+    Passing the parameter ``normalize`` equal to True, you shift the start
+    of the next business hour to midnight.
+
+    >>> ts = pd.Timestamp(2022, 12, 9, 8)
+    >>> ts + pd.offsets.BusinessHour(normalize=True)
+    Timestamp('2022-12-09 00:00:00')
+
+    You can divide your business day hours into several parts.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.BusinessHour(start=["06:00", "10:00", "15:00"],
+    ...                                end=["08:00", "12:00", "17:00"])
+    >>> pd.date_range(dt.datetime(2022, 12, 9), dt.datetime(2022, 12, 13), freq=freq)
+    DatetimeIndex(['2022-12-09 06:00:00', '2022-12-09 07:00:00',
+                   '2022-12-09 10:00:00', '2022-12-09 11:00:00',
+                   '2022-12-09 15:00:00', '2022-12-09 16:00:00',
+                   '2022-12-12 06:00:00', '2022-12-12 07:00:00',
+                   '2022-12-12 10:00:00', '2022-12-12 11:00:00',
+                   '2022-12-12 15:00:00', '2022-12-12 16:00:00'],
+                   dtype='datetime64[ns]', freq='bh')
+    """
+
+    _prefix = "bh"
+    _anchor = 0
+    _attributes = tuple(["n", "normalize", "start", "end", "offset"])
+    _adjust_dst = False
+
+    cdef readonly:
+        tuple start, end
+
+    def __init__(
+            self, n=1, normalize=False, start="09:00", end="17:00", offset=timedelta(0)
+    ):
+        BusinessMixin.__init__(self, n, normalize, offset)
+
+        # must be validated here to equality check
+        if np.ndim(start) == 0:
+            # i.e. not is_list_like
+            start = [start]
+        if not len(start):
+            raise ValueError("Must include at least 1 start time")
+
+        if np.ndim(end) == 0:
+            # i.e. not is_list_like
+            end = [end]
+        if not len(end):
+            raise ValueError("Must include at least 1 end time")
+
+        start = np.array([_validate_business_time(x) for x in start])
+        end = np.array([_validate_business_time(x) for x in end])
+
+        # Validation of input
+        if len(start) != len(end):
+            raise ValueError("number of starting time and ending time must be the same")
+        num_openings = len(start)
+
+        # sort starting and ending time by starting time
+        index = np.argsort(start)
+
+        # convert to tuple so that start and end are hashable
+        start = tuple(start[index])
+        end = tuple(end[index])
+
+        total_secs = 0
+        for i in range(num_openings):
+            total_secs += self._get_business_hours_by_sec(start[i], end[i])
+            total_secs += self._get_business_hours_by_sec(
+                end[i], start[(i + 1) % num_openings]
+            )
+        if total_secs != 24 * 60 * 60:
+            raise ValueError(
+                "invalid starting and ending time(s): "
+                "opening hours should not touch or overlap with "
+                "one another"
+            )
+
+        self.start = start
+        self.end = end
+
+    cpdef __setstate__(self, state):
+        start = state.pop("start")
+        start = (start,) if np.ndim(start) == 0 else tuple(start)
+        end = state.pop("end")
+        end = (end,) if np.ndim(end) == 0 else tuple(end)
+        self.start = start
+        self.end = end
+
+        state.pop("kwds", {})
+        state.pop("next_bday", None)
+        BusinessMixin.__setstate__(self, state)
+
+    def _repr_attrs(self) -> str:
+        out = super()._repr_attrs()
+        # Use python string formatting to be faster than strftime
+        hours = ",".join(
+            f"{st.hour:02d}:{st.minute:02d}-{en.hour:02d}:{en.minute:02d}"
+            for st, en in zip(self.start, self.end)
+        )
+        attrs = [f"{self._prefix}={hours}"]
+        out += ": " + ", ".join(attrs)
+        return out
+
+    def _get_business_hours_by_sec(self, start, end):
+        """
+        Return business hours in a day by seconds.
+        """
+        # create dummy datetime to calculate business hours in a day
+        dtstart = datetime(2014, 4, 1, start.hour, start.minute)
+        day = 1 if start < end else 2
+        until = datetime(2014, 4, day, end.hour, end.minute)
+        return int((until - dtstart).total_seconds())
+
+    def _get_closing_time(self, dt: datetime) -> datetime:
+        """
+        Get the closing time of a business hour interval by its opening time.
+
+        Parameters
+        ----------
+        dt : datetime
+            Opening time of a business hour interval.
+
+        Returns
+        -------
+        result : datetime
+            Corresponding closing time.
+        """
+        for i, st in enumerate(self.start):
+            if st.hour == dt.hour and st.minute == dt.minute:
+                return dt + timedelta(
+                    seconds=self._get_business_hours_by_sec(st, self.end[i])
+                )
+        assert False
+
+    @cache_readonly
+    def next_bday(self):
+        """
+        Used for moving to next business day.
+        """
+        if self.n >= 0:
+            nb_offset = 1
+        else:
+            nb_offset = -1
+        if self._prefix.startswith(("c")):
+            # CustomBusinessHour
+            return CustomBusinessDay(
+                n=nb_offset,
+                weekmask=self.weekmask,
+                holidays=self.holidays,
+                calendar=self.calendar,
+            )
+        else:
+            return BusinessDay(n=nb_offset)
+
+    def _next_opening_time(self, other, sign=1):
+        """
+        If self.n and sign have the same sign, return the earliest opening time
+        later than or equal to current time.
+        Otherwise the latest opening time earlier than or equal to current
+        time.
+
+        Opening time always locates on BusinessDay.
+        However, closing time may not if business hour extends over midnight.
+
+        Parameters
+        ----------
+        other : datetime
+            Current time.
+        sign : int, default 1.
+            Either 1 or -1. Going forward in time if it has the same sign as
+            self.n. Going backward in time otherwise.
+
+        Returns
+        -------
+        result : datetime
+            Next opening time.
+        """
+        earliest_start = self.start[0]
+        latest_start = self.start[-1]
+
+        if self.n == 0:
+            is_same_sign = sign > 0
+        else:
+            is_same_sign = self.n * sign >= 0
+
+        if not self.next_bday.is_on_offset(other):
+            # today is not business day
+            other = other + sign * self.next_bday
+            if is_same_sign:
+                hour, minute = earliest_start.hour, earliest_start.minute
+            else:
+                hour, minute = latest_start.hour, latest_start.minute
+        else:
+            if is_same_sign:
+                if latest_start < other.time():
+                    # current time is after latest starting time in today
+                    other = other + sign * self.next_bday
+                    hour, minute = earliest_start.hour, earliest_start.minute
+                else:
+                    # find earliest starting time no earlier than current time
+                    for st in self.start:
+                        if other.time() <= st:
+                            hour, minute = st.hour, st.minute
+                            break
+            else:
+                if other.time() < earliest_start:
+                    # current time is before earliest starting time in today
+                    other = other + sign * self.next_bday
+                    hour, minute = latest_start.hour, latest_start.minute
+                else:
+                    # find latest starting time no later than current time
+                    for st in reversed(self.start):
+                        if other.time() >= st:
+                            hour, minute = st.hour, st.minute
+                            break
+
+        return datetime(other.year, other.month, other.day, hour, minute)
+
+    def _prev_opening_time(self, other: datetime) -> datetime:
+        """
+        If n is positive, return the latest opening time earlier than or equal
+        to current time.
+        Otherwise the earliest opening time later than or equal to current
+        time.
+
+        Parameters
+        ----------
+        other : datetime
+            Current time.
+
+        Returns
+        -------
+        result : datetime
+            Previous opening time.
+        """
+        return self._next_opening_time(other, sign=-1)
+
+    @apply_wraps
+    def rollback(self, dt: datetime) -> datetime:
+        """
+        Roll provided date backward to next offset only if not on offset.
+        """
+        if not self.is_on_offset(dt):
+            if self.n >= 0:
+                dt = self._prev_opening_time(dt)
+            else:
+                dt = self._next_opening_time(dt)
+            return self._get_closing_time(dt)
+        return dt
+
+    @apply_wraps
+    def rollforward(self, dt: datetime) -> datetime:
+        """
+        Roll provided date forward to next offset only if not on offset.
+        """
+        if not self.is_on_offset(dt):
+            if self.n >= 0:
+                return self._next_opening_time(dt)
+            else:
+                return self._prev_opening_time(dt)
+        return dt
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        # used for detecting edge condition
+        nanosecond = getattr(other, "nanosecond", 0)
+        # reset timezone and nanosecond
+        # other may be a Timestamp, thus not use replace
+        other = datetime(
+            other.year,
+            other.month,
+            other.day,
+            other.hour,
+            other.minute,
+            other.second,
+            other.microsecond,
+        )
+        n = self.n
+
+        # adjust other to reduce number of cases to handle
+        if n >= 0:
+            if other.time() in self.end or not self._is_on_offset(other):
+                other = self._next_opening_time(other)
+        else:
+            if other.time() in self.start:
+                # adjustment to move to previous business day
+                other = other - timedelta(seconds=1)
+            if not self._is_on_offset(other):
+                other = self._next_opening_time(other)
+                other = self._get_closing_time(other)
+
+        # get total business hours by sec in one business day
+        businesshours = sum(
+            self._get_business_hours_by_sec(st, en)
+            for st, en in zip(self.start, self.end)
+        )
+
+        bd, r = divmod(abs(n * 60), businesshours // 60)
+        if n < 0:
+            bd, r = -bd, -r
+
+        # adjust by business days first
+        if bd != 0:
+            if self._prefix.startswith("c"):
+                # GH#30593 this is a Custom offset
+                skip_bd = CustomBusinessDay(
+                    n=bd,
+                    weekmask=self.weekmask,
+                    holidays=self.holidays,
+                    calendar=self.calendar,
+                )
+            else:
+                skip_bd = BusinessDay(n=bd)
+            # midnight business hour may not on BusinessDay
+            if not self.next_bday.is_on_offset(other):
+                prev_open = self._prev_opening_time(other)
+                remain = other - prev_open
+                other = prev_open + skip_bd + remain
+            else:
+                other = other + skip_bd
+
+        # remaining business hours to adjust
+        bhour_remain = timedelta(minutes=r)
+
+        if n >= 0:
+            while bhour_remain != timedelta(0):
+                # business hour left in this business time interval
+                bhour = (
+                    self._get_closing_time(self._prev_opening_time(other)) - other
+                )
+                if bhour_remain < bhour:
+                    # finish adjusting if possible
+                    other += bhour_remain
+                    bhour_remain = timedelta(0)
+                else:
+                    # go to next business time interval
+                    bhour_remain -= bhour
+                    other = self._next_opening_time(other + bhour)
+        else:
+            while bhour_remain != timedelta(0):
+                # business hour left in this business time interval
+                bhour = self._next_opening_time(other) - other
+                if (
+                    bhour_remain > bhour
+                    or bhour_remain == bhour
+                    and nanosecond != 0
+                ):
+                    # finish adjusting if possible
+                    other += bhour_remain
+                    bhour_remain = timedelta(0)
+                else:
+                    # go to next business time interval
+                    bhour_remain -= bhour
+                    other = self._get_closing_time(
+                        self._next_opening_time(
+                            other + bhour - timedelta(seconds=1)
+                        )
+                    )
+
+        return other
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+
+        if dt.tzinfo is not None:
+            dt = datetime(
+                dt.year, dt.month, dt.day, dt.hour, dt.minute, dt.second, dt.microsecond
+            )
+        # Valid bh can be on the different BusinessDay during midnight
+        # Distinguish by the time spent from previous opening time
+        return self._is_on_offset(dt)
+
+    def _is_on_offset(self, dt: datetime) -> bool:
+        """
+        Slight speedups using calculated values.
+        """
+        # if self.normalize and not _is_normalized(dt):
+        #     return False
+        # Valid bh can be on the different BusinessDay during midnight
+        # Distinguish by the time spent from previous opening time
+        if self.n >= 0:
+            op = self._prev_opening_time(dt)
+        else:
+            op = self._next_opening_time(dt)
+        span = (dt - op).total_seconds()
+        businesshours = 0
+        for i, st in enumerate(self.start):
+            if op.hour == st.hour and op.minute == st.minute:
+                businesshours = self._get_business_hours_by_sec(st, self.end[i])
+        if span <= businesshours:
+            return True
+        else:
+            return False
+
+
+cdef class WeekOfMonthMixin(SingleConstructorOffset):
+    """
+    Mixin for methods common to WeekOfMonth and LastWeekOfMonth.
+    """
+
+    cdef readonly:
+        int weekday, week
+
+    def __init__(self, n=1, normalize=False, weekday=0):
+        BaseOffset.__init__(self, n, normalize)
+        self.weekday = weekday
+
+        if weekday < 0 or weekday > 6:
+            raise ValueError(f"Day must be 0<=day<=6, got {weekday}")
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        compare_day = self._get_offset_day(other)
+
+        months = self.n
+        months = roll_convention(other.day, months, compare_day)
+
+        shifted = shift_month(other, months, "start")
+        to_day = self._get_offset_day(shifted)
+        return _shift_day(shifted, to_day - shifted.day)
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return dt.day == self._get_offset_day(dt)
+
+    @property
+    def rule_code(self) -> str:
+        """
+        Return a string representing the base frequency.
+
+        See Also
+        --------
+        tseries.offsets.Hour.rule_code :
+            Returns a string representing the base frequency of 'h'.
+        tseries.offsets.Day.rule_code :
+            Returns a string representing the base frequency of 'D'.
+
+        Examples
+        --------
+        >>> pd.offsets.Week(5).rule_code
+        'W'
+
+        >>> pd.offsets.WeekOfMonth(n=1, week=0, weekday=0).rule_code
+        'WOM-1MON'
+        """
+        weekday = int_to_weekday.get(self.weekday, "")
+        if self.week == -1:
+            # LastWeekOfMonth
+            return f"{self._prefix}-{weekday}"
+        return f"{self._prefix}-{self.week + 1}{weekday}"
+
+
+# ----------------------------------------------------------------------
+# Year-Based Offset Classes
+
+cdef class YearOffset(SingleConstructorOffset):
+    """
+    DateOffset that just needs a month.
+    """
+    _attributes = tuple(["n", "normalize", "month"])
+
+    # FIXME(cython#4446): python annotation here gives compile-time errors
+    # _default_month: int
+
+    cdef readonly:
+        int month
+
+    def __init__(self, n=1, normalize=False, month=None):
+        BaseOffset.__init__(self, n, normalize)
+
+        month = month if month is not None else self._default_month
+        self.month = month
+
+        if month < 1 or month > 12:
+            raise ValueError("Month must go from 1 to 12")
+
+    cpdef __setstate__(self, state):
+        self.month = state.pop("month")
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self._cache = {}
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        kwargs = {}
+        if suffix:
+            kwargs["month"] = MONTH_TO_CAL_NUM[suffix]
+        return cls(**kwargs)
+
+    @property
+    def rule_code(self) -> str:
+        """
+        Return a string representing the base frequency.
+
+        See Also
+        --------
+        tseries.offsets.Hour.rule_code :
+            Returns a string representing the base frequency of 'h'.
+        tseries.offsets.Day.rule_code :
+            Returns a string representing the base frequency of 'D'.
+
+        Examples
+        --------
+        >>> pd.tseries.offsets.YearBegin(n=1, month=2).rule_code
+        'YS-FEB'
+
+        >>> pd.tseries.offsets.YearEnd(n=1, month=6).rule_code
+        'YE-JUN'
+        """
+        month = MONTH_ALIASES[self.month]
+        return f"{self._prefix}-{month}"
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return dt.month == self.month and dt.day == self._get_offset_day(dt)
+
+    def _get_offset_day(self, other: datetime) -> int:
+        # override BaseOffset method to use self.month instead of other.month
+        cdef:
+            npy_datetimestruct dts
+        pydate_to_dtstruct(other, &dts)
+        dts.month = self.month
+        return get_day_of_month(&dts, self._day_opt)
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        years = roll_qtrday(other, self.n, self.month, self._day_opt, modby=12)
+        months = years * 12 + (self.month - other.month)
+        return shift_month(other, months, self._day_opt)
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        reso = get_unit_from_dtype(dtarr.dtype)
+        shifted = shift_quarters(
+            dtarr.view("i8"), self.n, self.month, self._day_opt, modby=12, reso=reso
+        )
+        return shifted
+
+
+cdef class BYearEnd(YearOffset):
+    """
+    DateOffset increments between the last business day of the year.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    month : int, default 12
+        A specific integer for the month of the year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> from pandas.tseries.offsets import BYearEnd
+    >>> ts = pd.Timestamp('2020-05-24 05:01:15')
+    >>> ts - BYearEnd()
+    Timestamp('2019-12-31 05:01:15')
+    >>> ts + BYearEnd()
+    Timestamp('2020-12-31 05:01:15')
+    >>> ts + BYearEnd(3)
+    Timestamp('2022-12-30 05:01:15')
+    >>> ts + BYearEnd(-3)
+    Timestamp('2017-12-29 05:01:15')
+    >>> ts + BYearEnd(month=11)
+    Timestamp('2020-11-30 05:01:15')
+    """
+
+    _outputName = "BusinessYearEnd"
+    _default_month = 12
+    _prefix = "BYE"
+    _day_opt = "business_end"
+
+
+cdef class BYearBegin(YearOffset):
+    """
+    DateOffset increments between the first business day of the year.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    month : int, default 1
+        A specific integer for the month of the year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> from pandas.tseries.offsets import BYearBegin
+    >>> ts = pd.Timestamp('2020-05-24 05:01:15')
+    >>> ts + BYearBegin()
+    Timestamp('2021-01-01 05:01:15')
+    >>> ts - BYearBegin()
+    Timestamp('2020-01-01 05:01:15')
+    >>> ts + BYearBegin(-1)
+    Timestamp('2020-01-01 05:01:15')
+    >>> ts + BYearBegin(2)
+    Timestamp('2022-01-03 05:01:15')
+    >>> ts + BYearBegin(month=11)
+    Timestamp('2020-11-02 05:01:15')
+    """
+
+    _outputName = "BusinessYearBegin"
+    _default_month = 1
+    _prefix = "BYS"
+    _day_opt = "business_start"
+
+
+cdef class YearEnd(YearOffset):
+    """
+    DateOffset increments between calendar year end dates.
+
+    YearEnd goes to the next date which is the end of the year.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    month : int, default 12
+        A specific integer for the month of the year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.YearEnd()
+    Timestamp('2022-12-31 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 12, 31)
+    >>> ts + pd.offsets.YearEnd()
+    Timestamp('2023-12-31 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.YearEnd(month=2)
+    Timestamp('2022-02-28 00:00:00')
+
+    If you want to get the end of the current year:
+
+    >>> ts = pd.Timestamp(2022, 12, 31)
+    >>> pd.offsets.YearEnd().rollforward(ts)
+    Timestamp('2022-12-31 00:00:00')
+    """
+
+    _default_month = 12
+    _prefix = "YE"
+    _day_opt = "end"
+
+    cdef readonly:
+        int _period_dtype_code
+
+    def __init__(self, n=1, normalize=False, month=None):
+        # Because YearEnd can be the freq for a Period, define its
+        #  _period_dtype_code at construction for performance
+        YearOffset.__init__(self, n, normalize, month)
+        self._period_dtype_code = PeriodDtypeCode.A + self.month % 12
+
+
+cdef class YearBegin(YearOffset):
+    """
+    DateOffset increments between calendar year begin dates.
+
+    YearBegin goes to the next date which is the start of the year.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    month : int, default 1
+        A specific integer for the month of the year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 12, 1)
+    >>> ts + pd.offsets.YearBegin()
+    Timestamp('2023-01-01 00:00:00')
+
+    >>> ts = pd.Timestamp(2023, 1, 1)
+    >>> ts + pd.offsets.YearBegin()
+    Timestamp('2024-01-01 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.YearBegin(month=2)
+    Timestamp('2022-02-01 00:00:00')
+
+    If you want to get the start of the current year:
+
+    >>> ts = pd.Timestamp(2023, 1, 1)
+    >>> pd.offsets.YearBegin().rollback(ts)
+    Timestamp('2023-01-01 00:00:00')
+    """
+
+    _default_month = 1
+    _prefix = "YS"
+    _day_opt = "start"
+
+
+# ----------------------------------------------------------------------
+# Quarter-Based Offset Classes
+
+cdef class QuarterOffset(SingleConstructorOffset):
+    _attributes = tuple(["n", "normalize", "startingMonth"])
+    # TODO: Consider combining QuarterOffset and YearOffset __init__ at some
+    #       point.  Also apply_index, is_on_offset, rule_code if
+    #       startingMonth vs month attr names are resolved
+
+    # FIXME(cython#4446): python annotation here gives compile-time errors
+    # _default_starting_month: int
+    # _from_name_starting_month: int
+
+    cdef readonly:
+        int startingMonth
+
+    def __init__(self, n=1, normalize=False, startingMonth=None):
+        BaseOffset.__init__(self, n, normalize)
+
+        if startingMonth is None:
+            startingMonth = self._default_starting_month
+        self.startingMonth = startingMonth
+
+    cpdef __setstate__(self, state):
+        self.startingMonth = state.pop("startingMonth")
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        kwargs = {}
+        if suffix:
+            kwargs["startingMonth"] = MONTH_TO_CAL_NUM[suffix]
+        else:
+            if cls._from_name_starting_month is not None:
+                kwargs["startingMonth"] = cls._from_name_starting_month
+        return cls(**kwargs)
+
+    @property
+    def rule_code(self) -> str:
+        month = MONTH_ALIASES[self.startingMonth]
+        return f"{self._prefix}-{month}"
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        mod_month = (dt.month - self.startingMonth) % 3
+        return mod_month == 0 and dt.day == self._get_offset_day(dt)
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        # months_since: find the calendar quarter containing other.month,
+        # e.g. if other.month == 8, the calendar quarter is [Jul, Aug, Sep].
+        # Then find the month in that quarter containing an is_on_offset date for
+        # self.  `months_since` is the number of months to shift other.month
+        # to get to this on-offset month.
+        months_since = other.month % 3 - self.startingMonth % 3
+        qtrs = roll_qtrday(
+            other, self.n, self.startingMonth, day_opt=self._day_opt, modby=3
+        )
+        months = qtrs * 3 - months_since
+        return shift_month(other, months, self._day_opt)
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        reso = get_unit_from_dtype(dtarr.dtype)
+        shifted = shift_quarters(
+            dtarr.view("i8"),
+            self.n,
+            self.startingMonth,
+            self._day_opt,
+            modby=3,
+            reso=reso,
+        )
+        return shifted
+
+
+cdef class BQuarterEnd(QuarterOffset):
+    """
+    DateOffset increments between the last business day of each Quarter.
+
+    startingMonth = 1 corresponds to dates like 1/31/2007, 4/30/2007, ...
+    startingMonth = 2 corresponds to dates like 2/28/2007, 5/31/2007, ...
+    startingMonth = 3 corresponds to dates like 3/30/2007, 6/29/2007, ...
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of quarters represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    startingMonth : int, default 3
+        A specific integer for the month of the year from which we start quarters.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> from pandas.tseries.offsets import BQuarterEnd
+    >>> ts = pd.Timestamp('2020-05-24 05:01:15')
+    >>> ts + BQuarterEnd()
+    Timestamp('2020-06-30 05:01:15')
+    >>> ts + BQuarterEnd(2)
+    Timestamp('2020-09-30 05:01:15')
+    >>> ts + BQuarterEnd(1, startingMonth=2)
+    Timestamp('2020-05-29 05:01:15')
+    >>> ts + BQuarterEnd(startingMonth=2)
+    Timestamp('2020-05-29 05:01:15')
+    """
+    _output_name = "BusinessQuarterEnd"
+    _default_starting_month = 3
+    _from_name_starting_month = 12
+    _prefix = "BQE"
+    _day_opt = "business_end"
+
+
+cdef class BQuarterBegin(QuarterOffset):
+    """
+    DateOffset increments between the first business day of each Quarter.
+
+    startingMonth = 1 corresponds to dates like 1/01/2007, 4/01/2007, ...
+    startingMonth = 2 corresponds to dates like 2/01/2007, 5/01/2007, ...
+    startingMonth = 3 corresponds to dates like 3/01/2007, 6/01/2007, ...
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of quarters represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    startingMonth : int, default 3
+        A specific integer for the month of the year from which we start quarters.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> from pandas.tseries.offsets import BQuarterBegin
+    >>> ts = pd.Timestamp('2020-05-24 05:01:15')
+    >>> ts + BQuarterBegin()
+    Timestamp('2020-06-01 05:01:15')
+    >>> ts + BQuarterBegin(2)
+    Timestamp('2020-09-01 05:01:15')
+    >>> ts + BQuarterBegin(startingMonth=2)
+    Timestamp('2020-08-03 05:01:15')
+    >>> ts + BQuarterBegin(-1)
+    Timestamp('2020-03-02 05:01:15')
+    """
+    _output_name = "BusinessQuarterBegin"
+    _default_starting_month = 3
+    _from_name_starting_month = 1
+    _prefix = "BQS"
+    _day_opt = "business_start"
+
+
+cdef class QuarterEnd(QuarterOffset):
+    """
+    DateOffset increments between Quarter end dates.
+
+    startingMonth = 1 corresponds to dates like 1/31/2007, 4/30/2007, ...
+    startingMonth = 2 corresponds to dates like 2/28/2007, 5/31/2007, ...
+    startingMonth = 3 corresponds to dates like 3/31/2007, 6/30/2007, ...
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of quarters represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    startingMonth : int, default 3
+        A specific integer for the month of the year from which we start quarters.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.QuarterEnd()
+    Timestamp('2022-03-31 00:00:00')
+    """
+    _default_starting_month = 3
+    _prefix = "QE"
+    _day_opt = "end"
+
+    cdef readonly:
+        int _period_dtype_code
+
+    def __init__(self, n=1, normalize=False, startingMonth=None):
+        # Because QuarterEnd can be the freq for a Period, define its
+        #  _period_dtype_code at construction for performance
+        QuarterOffset.__init__(self, n, normalize, startingMonth)
+        self._period_dtype_code = PeriodDtypeCode.Q_DEC + self.startingMonth % 12
+
+
+cdef class QuarterBegin(QuarterOffset):
+    """
+    DateOffset increments between Quarter start dates.
+
+    startingMonth = 1 corresponds to dates like 1/01/2007, 4/01/2007, ...
+    startingMonth = 2 corresponds to dates like 2/01/2007, 5/01/2007, ...
+    startingMonth = 3 corresponds to dates like 3/01/2007, 6/01/2007, ...
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of quarters represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    startingMonth : int, default 3
+        A specific integer for the month of the year from which we start quarters.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.QuarterBegin()
+    Timestamp('2022-03-01 00:00:00')
+    """
+    _default_starting_month = 3
+    _from_name_starting_month = 1
+    _prefix = "QS"
+    _day_opt = "start"
+
+
+# ----------------------------------------------------------------------
+# Month-Based Offset Classes
+
+cdef class MonthOffset(SingleConstructorOffset):
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return dt.day == self._get_offset_day(dt)
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        compare_day = self._get_offset_day(other)
+        n = roll_convention(other.day, self.n, compare_day)
+        return shift_month(other, n, self._day_opt)
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        reso = get_unit_from_dtype(dtarr.dtype)
+        shifted = shift_months(dtarr.view("i8"), self.n, self._day_opt, reso=reso)
+        return shifted
+
+    cpdef __setstate__(self, state):
+        state.pop("_use_relativedelta", False)
+        state.pop("offset", None)
+        state.pop("_offset", None)
+        state.pop("kwds", {})
+
+        BaseOffset.__setstate__(self, state)
+
+
+cdef class MonthEnd(MonthOffset):
+    """
+    DateOffset of one month end.
+
+    MonthEnd goes to the next date which is an end of the month.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 30)
+    >>> ts + pd.offsets.MonthEnd()
+    Timestamp('2022-01-31 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 1, 31)
+    >>> ts + pd.offsets.MonthEnd()
+    Timestamp('2022-02-28 00:00:00')
+
+    If you want to get the end of the current month:
+
+    >>> ts = pd.Timestamp(2022, 1, 31)
+    >>> pd.offsets.MonthEnd().rollforward(ts)
+    Timestamp('2022-01-31 00:00:00')
+    """
+    _period_dtype_code = PeriodDtypeCode.M
+    _prefix = "ME"
+    _day_opt = "end"
+
+
+cdef class MonthBegin(MonthOffset):
+    """
+    DateOffset of one month at beginning.
+
+    MonthBegin goes to the next date which is a start of the month.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 11, 30)
+    >>> ts + pd.offsets.MonthBegin()
+    Timestamp('2022-12-01 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 12, 1)
+    >>> ts + pd.offsets.MonthBegin()
+    Timestamp('2023-01-01 00:00:00')
+
+    If you want to get the start of the current month:
+
+    >>> ts = pd.Timestamp(2022, 12, 1)
+    >>> pd.offsets.MonthBegin().rollback(ts)
+    Timestamp('2022-12-01 00:00:00')
+    """
+    _prefix = "MS"
+    _day_opt = "start"
+
+
+cdef class BusinessMonthEnd(MonthOffset):
+    """
+    DateOffset increments between the last business day of the month.
+
+    BusinessMonthEnd goes to the next date which is the last business day of the month.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 11, 29)
+    >>> ts + pd.offsets.BMonthEnd()
+    Timestamp('2022-11-30 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 11, 30)
+    >>> ts + pd.offsets.BMonthEnd()
+    Timestamp('2022-12-30 00:00:00')
+
+    If you want to get the end of the current business month:
+
+    >>> ts = pd.Timestamp(2022, 11, 30)
+    >>> pd.offsets.BMonthEnd().rollforward(ts)
+    Timestamp('2022-11-30 00:00:00')
+    """
+    _prefix = "BME"
+    _day_opt = "business_end"
+
+
+cdef class BusinessMonthBegin(MonthOffset):
+    """
+    DateOffset of one month at the first business day.
+
+    BusinessMonthBegin goes to the next date which is the first business day
+    of the month.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 11, 30)
+    >>> ts + pd.offsets.BMonthBegin()
+    Timestamp('2022-12-01 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 12, 1)
+    >>> ts + pd.offsets.BMonthBegin()
+    Timestamp('2023-01-02 00:00:00')
+
+    If you want to get the start of the current business month:
+
+    >>> ts = pd.Timestamp(2022, 12, 1)
+    >>> pd.offsets.BMonthBegin().rollback(ts)
+    Timestamp('2022-12-01 00:00:00')
+    """
+    _prefix = "BMS"
+    _day_opt = "business_start"
+
+
+# ---------------------------------------------------------------------
+# Semi-Month Based Offsets
+
+cdef class SemiMonthOffset(SingleConstructorOffset):
+    _default_day_of_month = 15
+    _min_day_of_month = 2
+    _attributes = tuple(["n", "normalize", "day_of_month"])
+
+    cdef readonly:
+        int day_of_month
+
+    def __init__(self, n=1, normalize=False, day_of_month=None):
+        BaseOffset.__init__(self, n, normalize)
+
+        if day_of_month is None:
+            day_of_month = self._default_day_of_month
+
+        self.day_of_month = int(day_of_month)
+        if not self._min_day_of_month <= self.day_of_month <= 27:
+            raise ValueError(
+                "day_of_month must be "
+                f"{self._min_day_of_month}<=day_of_month<=27, "
+                f"got {self.day_of_month}"
+            )
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self.day_of_month = state.pop("day_of_month")
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        return cls(day_of_month=suffix)
+
+    @property
+    def rule_code(self) -> str:
+        suffix = f"-{self.day_of_month}"
+        return self._prefix + suffix
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        is_start = isinstance(self, SemiMonthBegin)
+
+        # shift `other` to self.day_of_month, incrementing `n` if necessary
+        n = roll_convention(other.day, self.n, self.day_of_month)
+
+        days_in_month = get_days_in_month(other.year, other.month)
+        # For SemiMonthBegin on other.day == 1 and
+        # SemiMonthEnd on other.day == days_in_month,
+        # shifting `other` to `self.day_of_month` _always_ requires
+        # incrementing/decrementing `n`, regardless of whether it is
+        # initially positive.
+        if is_start and (self.n <= 0 and other.day == 1):
+            n -= 1
+        elif (not is_start) and (self.n > 0 and other.day == days_in_month):
+            n += 1
+
+        if is_start:
+            months = n // 2 + n % 2
+            to_day = 1 if n % 2 else self.day_of_month
+        else:
+            months = n // 2
+            to_day = 31 if n % 2 else self.day_of_month
+
+        return shift_month(other, months, to_day)
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        cdef:
+            ndarray i8other = dtarr.view("i8")
+            Py_ssize_t i, count = dtarr.size
+            int64_t val, res_val
+            ndarray out = cnp.PyArray_EMPTY(
+                i8other.ndim, i8other.shape, cnp.NPY_INT64, 0
+            )
+            npy_datetimestruct dts
+            int months, to_day, nadj, n = self.n
+            int days_in_month, day, anchor_dom = self.day_of_month
+            bint is_start = isinstance(self, SemiMonthBegin)
+            NPY_DATETIMEUNIT reso = get_unit_from_dtype(dtarr.dtype)
+            cnp.broadcast mi = cnp.PyArray_MultiIterNew2(out, i8other)
+
+        with nogil:
+            for i in range(count):
+                # Analogous to: val = i8other[i]
+                val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+                if val == NPY_NAT:
+                    res_val = NPY_NAT
+
+                else:
+                    pandas_datetime_to_datetimestruct(val, reso, &dts)
+                    day = dts.day
+
+                    # Adjust so that we are always looking at self.day_of_month,
+                    #  incrementing/decrementing n if necessary.
+                    nadj = roll_convention(day, n, anchor_dom)
+
+                    days_in_month = get_days_in_month(dts.year, dts.month)
+                    # For SemiMonthBegin on other.day == 1 and
+                    #  SemiMonthEnd on other.day == days_in_month,
+                    #  shifting `other` to `self.day_of_month` _always_ requires
+                    #  incrementing/decrementing `n`, regardless of whether it is
+                    #  initially positive.
+                    if is_start and (n <= 0 and day == 1):
+                        nadj -= 1
+                    elif (not is_start) and (n > 0 and day == days_in_month):
+                        nadj += 1
+
+                    if is_start:
+                        # See also: SemiMonthBegin._apply
+                        months = nadj // 2 + nadj % 2
+                        to_day = 1 if nadj % 2 else anchor_dom
+
+                    else:
+                        # See also: SemiMonthEnd._apply
+                        months = nadj // 2
+                        to_day = 31 if nadj % 2 else anchor_dom
+
+                    dts.year = year_add_months(dts, months)
+                    dts.month = month_add_months(dts, months)
+                    days_in_month = get_days_in_month(dts.year, dts.month)
+                    dts.day = min(to_day, days_in_month)
+
+                    res_val = npy_datetimestruct_to_datetime(reso, &dts)
+
+                # Analogous to: out[i] = res_val
+                (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+                cnp.PyArray_MultiIter_NEXT(mi)
+
+        return out
+
+
+cdef class SemiMonthEnd(SemiMonthOffset):
+    """
+    Two DateOffset's per month repeating on the last day of the month & day_of_month.
+
+    This offset allows for flexibility in generating date ranges or adjusting dates
+    to the end of a month or a specific day in the month, such as the 15th or the last
+    day of the month. It is useful for financial or scheduling applications where
+    events occur bi-monthly.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    day_of_month : int, {1, 3,...,27}, default 15
+        A specific integer for the day of the month.
+
+    See Also
+    --------
+    tseries.offsets.SemiMonthBegin : Offset for semi-monthly frequencies, starting at
+        the beginning of the month.
+    tseries.offsets.MonthEnd : Offset to the last calendar day of the month.
+    tseries.offsets.MonthBegin : Offset to the first calendar day of the month.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 14)
+    >>> ts + pd.offsets.SemiMonthEnd()
+    Timestamp('2022-01-15 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 1, 15)
+    >>> ts + pd.offsets.SemiMonthEnd()
+    Timestamp('2022-01-31 00:00:00')
+
+    >>> ts = pd.Timestamp(2022, 1, 31)
+    >>> ts + pd.offsets.SemiMonthEnd()
+    Timestamp('2022-02-15 00:00:00')
+
+    If you want to get the result for the current month:
+
+    >>> ts = pd.Timestamp(2022, 1, 15)
+    >>> pd.offsets.SemiMonthEnd().rollforward(ts)
+    Timestamp('2022-01-15 00:00:00')
+    """
+    _prefix = "SME"
+    _min_day_of_month = 1
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        days_in_month = get_days_in_month(dt.year, dt.month)
+        return dt.day in (self.day_of_month, days_in_month)
+
+
+cdef class SemiMonthBegin(SemiMonthOffset):
+    """
+    Two DateOffset's per month repeating on the first day of the month & day_of_month.
+
+    This offset moves dates to the first day of the month and an additional specified
+    day (typically the 15th by default), useful in scenarios where bi-monthly processing
+    occurs on set days.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    day_of_month : int, {1, 3,...,27}, default 15
+        A specific integer for the day of the month.
+
+    See Also
+    --------
+    tseries.offsets.SemiMonthEnd : Two DateOffset's per month repeating on the last day
+        of the month & day_of_month.
+    tseries.offsets.MonthEnd : Offset to the last calendar day of the month.
+    tseries.offsets.MonthBegin : Offset to the first calendar day of the month.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.SemiMonthBegin()
+    Timestamp('2022-01-15 00:00:00')
+    """
+
+    _prefix = "SMS"
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        return dt.day in (1, self.day_of_month)
+
+
+# ---------------------------------------------------------------------
+# Week-Based Offset Classes
+
+
+cdef class Week(SingleConstructorOffset):
+    """
+    Weekly offset.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of weeks represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekday : int or None, default None
+        Always generate specific day of week.
+        0 for Monday and 6 for Sunday.
+
+    See Also
+    --------
+    pd.tseries.offsets.WeekOfMonth :
+     Describes monthly dates like, the Tuesday of the
+     2nd week of each month.
+
+    Examples
+    --------
+
+    >>> date_object = pd.Timestamp("2023-01-13")
+    >>> date_object
+    Timestamp('2023-01-13 00:00:00')
+
+    >>> date_plus_one_week = date_object + pd.tseries.offsets.Week(n=1)
+    >>> date_plus_one_week
+    Timestamp('2023-01-20 00:00:00')
+
+    >>> date_next_monday = date_object + pd.tseries.offsets.Week(weekday=0)
+    >>> date_next_monday
+    Timestamp('2023-01-16 00:00:00')
+
+    >>> date_next_sunday = date_object + pd.tseries.offsets.Week(weekday=6)
+    >>> date_next_sunday
+    Timestamp('2023-01-15 00:00:00')
+    """
+
+    _inc = timedelta(weeks=1)
+    _prefix = "W"
+    _attributes = tuple(["n", "normalize", "weekday"])
+
+    cdef readonly:
+        object weekday  # int or None
+        int _period_dtype_code
+
+    def __init__(self, n=1, normalize=False, weekday=None):
+        BaseOffset.__init__(self, n, normalize)
+        self.weekday = weekday
+
+        if self.weekday is not None:
+            if self.weekday < 0 or self.weekday > 6:
+                raise ValueError(f"Day must be 0<=day<=6, got {self.weekday}")
+
+            self._period_dtype_code = PeriodDtypeCode.W_SUN + (weekday + 1) % 7
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self.weekday = state.pop("weekday")
+        self._cache = state.pop("_cache", {})
+
+    @apply_wraps
+    def _apply(self, other):
+        if self.weekday is None:
+            return other + self.n * self._inc
+
+        if not PyDateTime_Check(other):
+            raise TypeError(
+                f"Cannot add {type(other).__name__} to {type(self).__name__}"
+            )
+
+        k = self.n
+        otherDay = other.weekday()
+        if otherDay != self.weekday:
+            other = other + timedelta((self.weekday - otherDay) % 7)
+            if k > 0:
+                k -= 1
+
+        return other + timedelta(weeks=k)
+
+    def _apply_array(self, dtarr: np.ndarray) -> np.ndarray:
+        if self.weekday is None:
+            td = timedelta(days=7 * self.n)
+            unit = np.datetime_data(dtarr.dtype)[0]
+            td64 = np.timedelta64(td, unit)
+            return dtarr + td64
+        else:
+            reso = get_unit_from_dtype(dtarr.dtype)
+            i8other = dtarr.view("i8")
+            return self._end_apply_index(i8other, reso=reso)
+
+    @cython.wraparound(False)
+    @cython.boundscheck(False)
+    cdef ndarray _end_apply_index(self, ndarray i8other, NPY_DATETIMEUNIT reso):
+        """
+        Add self to the given DatetimeIndex, specialized for case where
+        self.weekday is non-null.
+
+        Parameters
+        ----------
+        i8other : const int64_t[:]
+        reso : NPY_DATETIMEUNIT
+
+        Returns
+        -------
+        ndarray[int64_t]
+        """
+        cdef:
+            Py_ssize_t i, count = i8other.size
+            int64_t val, res_val
+            ndarray out = cnp.PyArray_EMPTY(
+                i8other.ndim, i8other.shape, cnp.NPY_INT64, 0
+            )
+            npy_datetimestruct dts
+            int wday, days, weeks, n = self.n
+            int anchor_weekday = self.weekday
+            int64_t DAY_PERIODS = periods_per_day(reso)
+            cnp.broadcast mi = cnp.PyArray_MultiIterNew2(out, i8other)
+
+        with nogil:
+            for i in range(count):
+                # Analogous to: val = i8other[i]
+                val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+                if val == NPY_NAT:
+                    res_val = NPY_NAT
+                else:
+                    pandas_datetime_to_datetimestruct(val, reso, &dts)
+                    wday = dayofweek(dts.year, dts.month, dts.day)
+
+                    days = 0
+                    weeks = n
+                    if wday != anchor_weekday:
+                        days = (anchor_weekday - wday) % 7
+                        if weeks > 0:
+                            weeks -= 1
+
+                    res_val = val + (7 * weeks + days) * DAY_PERIODS
+
+                # Analogous to: out[i] = res_val
+                (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+                cnp.PyArray_MultiIter_NEXT(mi)
+
+        return out
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        elif self.weekday is None:
+            return True
+        return dt.weekday() == self.weekday
+
+    @property
+    def rule_code(self) -> str:
+        """
+        Return a string representing the base frequency.
+
+        See Also
+        --------
+        tseries.offsets.Hour.name :
+            Returns a string representing the base frequency of 'h'.
+        tseries.offsets.Day.name :
+            Returns a string representing the base frequency of 'D'.
+
+        Examples
+        --------
+        >>> pd.offsets.Hour().rule_code
+        'h'
+
+        >>> pd.offsets.Week(5).rule_code
+        'W'
+        """
+        suffix = ""
+        if self.weekday is not None:
+            weekday = int_to_weekday[self.weekday]
+            suffix = f"-{weekday}"
+        return self._prefix + suffix
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        if not suffix:
+            weekday = None
+        else:
+            weekday = weekday_to_int[suffix]
+        return cls(weekday=weekday)
+
+
+cdef class WeekOfMonth(WeekOfMonthMixin):
+    """
+    Describes monthly dates like "the Tuesday of the 2nd week of each month".
+
+    This offset allows for generating or adjusting dates by specifying
+    a particular week and weekday within a month. The week is zero-indexed,
+    where 0 corresponds to the first week of the month, and weekday follows
+    a Monday=0 convention.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    week : int {0, 1, 2, 3, ...}, default 0
+        A specific integer for the week of the month.
+        e.g. 0 is 1st week of month, 1 is the 2nd week, etc.
+    weekday : int {0, 1, ..., 6}, default 0
+        A specific integer for the day of the week.
+
+        - 0 is Monday
+        - 1 is Tuesday
+        - 2 is Wednesday
+        - 3 is Thursday
+        - 4 is Friday
+        - 5 is Saturday
+        - 6 is Sunday.
+
+    See Also
+    --------
+    offsets.Week : Describes weekly frequency adjustments.
+    offsets.MonthEnd : Describes month-end frequency adjustments.
+    date_range : Generates a range of dates based on a specific frequency.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.WeekOfMonth()
+    Timestamp('2022-01-03 00:00:00')
+    """
+
+    _prefix = "WOM"
+    _attributes = tuple(["n", "normalize", "week", "weekday"])
+
+    def __init__(self, n=1, normalize=False, week=0, weekday=0):
+        WeekOfMonthMixin.__init__(self, n, normalize, weekday)
+        self.week = week
+
+        if self.week < 0 or self.week > 3:
+            raise ValueError(f"Week must be 0<=week<=3, got {self.week}")
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self.weekday = state.pop("weekday")
+        self.week = state.pop("week")
+
+    def _get_offset_day(self, other: datetime) -> int:
+        """
+        Find the day in the same month as other that has the same
+        weekday as self.weekday and is the self.week'th such day in the month.
+
+        Parameters
+        ----------
+        other : datetime
+
+        Returns
+        -------
+        day : int
+        """
+        mstart = datetime(other.year, other.month, 1)
+        wday = mstart.weekday()
+        shift_days = (self.weekday - wday) % 7
+        return 1 + shift_days + self.week * 7
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        if not suffix:
+            raise ValueError(f"Prefix {repr(cls._prefix)} requires a suffix.")
+        # only one digit weeks (1 --> week 0, 2 --> week 1, etc.)
+        week = int(suffix[0]) - 1
+        weekday = weekday_to_int[suffix[1:]]
+        return cls(week=week, weekday=weekday)
+
+
+cdef class LastWeekOfMonth(WeekOfMonthMixin):
+    """
+    Describes monthly dates in last week of month.
+
+    For example "the last Tuesday of each month".
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekday : int {0, 1, ..., 6}, default 0
+        A specific integer for the day of the week.
+
+        - 0 is Monday
+        - 1 is Tuesday
+        - 2 is Wednesday
+        - 3 is Thursday
+        - 4 is Friday
+        - 5 is Saturday
+        - 6 is Sunday.
+
+    See Also
+    --------
+    tseries.offsets.WeekOfMonth :
+        Date offset for a specific weekday in a month.
+    tseries.offsets.MonthEnd :
+        Date offset for the end of the month.
+    tseries.offsets.BMonthEnd :
+        Date offset for the last business day of the month.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.LastWeekOfMonth()
+    Timestamp('2022-01-31 00:00:00')
+    """
+
+    _prefix = "LWOM"
+    _attributes = tuple(["n", "normalize", "weekday"])
+
+    def __init__(self, n=1, normalize=False, weekday=0):
+        WeekOfMonthMixin.__init__(self, n, normalize, weekday)
+        self.week = -1
+
+        if self.n == 0:
+            raise ValueError("N cannot be 0")
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self.weekday = state.pop("weekday")
+        self.week = -1
+
+    def _get_offset_day(self, other: datetime) -> int:
+        """
+        Find the day in the same month as other that has the same
+        weekday as self.weekday and is the last such day in the month.
+
+        Parameters
+        ----------
+        other: datetime
+
+        Returns
+        -------
+        day: int
+        """
+        dim = get_days_in_month(other.year, other.month)
+        mend = datetime(other.year, other.month, dim)
+        wday = mend.weekday()
+        shift_days = (wday - self.weekday) % 7
+        return dim - shift_days
+
+    @classmethod
+    def _from_name(cls, suffix=None):
+        if not suffix:
+            raise ValueError(f"Prefix {repr(cls._prefix)} requires a suffix.")
+        weekday = weekday_to_int[suffix]
+        return cls(weekday=weekday)
+
+
+# ---------------------------------------------------------------------
+# Special Offset Classes
+
+cdef class FY5253Mixin(SingleConstructorOffset):
+    cdef readonly:
+        int startingMonth
+        int weekday
+        str variation
+
+    def __init__(
+        self, n=1, normalize=False, weekday=0, startingMonth=1, variation="nearest"
+    ):
+        BaseOffset.__init__(self, n, normalize)
+        self.startingMonth = startingMonth
+        self.weekday = weekday
+        self.variation = variation
+
+        if self.n == 0:
+            raise ValueError("N cannot be 0")
+
+        if self.variation not in ["nearest", "last"]:
+            raise ValueError(f"{self.variation} is not a valid variation")
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+        self.weekday = state.pop("weekday")
+        self.variation = state.pop("variation")
+
+    # --------------------------------------------------------------------
+    # Name-related methods
+
+    @property
+    def rule_code(self) -> str:
+        prefix = self._prefix
+        suffix = self.get_rule_code_suffix()
+        return f"{prefix}-{suffix}"
+
+    def _get_suffix_prefix(self) -> str:
+        if self.variation == "nearest":
+            return "N"
+        else:
+            return "L"
+
+    def get_rule_code_suffix(self) -> str:
+        prefix = self._get_suffix_prefix()
+        month = MONTH_ALIASES[self.startingMonth]
+        weekday = int_to_weekday[self.weekday]
+        return f"{prefix}-{month}-{weekday}"
+
+
+cdef class FY5253(FY5253Mixin):
+    """
+    Describes 52-53 week fiscal year. This is also known as a 4-4-5 calendar.
+
+    It is used by companies that desire that their
+    fiscal year always end on the same day of the week.
+
+    It is a method of managing accounting periods.
+    It is a common calendar structure for some industries,
+    such as retail, manufacturing and parking industry.
+
+    For more information see:
+    https://en.wikipedia.org/wiki/4-4-5_calendar
+
+    The year may either:
+
+    - end on the last X day of the Y month.
+    - end on the last X day closest to the last day of the Y month.
+
+    X is a specific day of the week.
+    Y is a certain month of the year
+
+    Attributes
+    ----------
+    n : int
+        The number of fiscal years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekday : int {0, 1, ..., 6}, default 0
+        A specific integer for the day of the week.
+
+        - 0 is Monday
+        - 1 is Tuesday
+        - 2 is Wednesday
+        - 3 is Thursday
+        - 4 is Friday
+        - 5 is Saturday
+        - 6 is Sunday.
+
+    startingMonth : int {1, 2, ... 12}, default 1
+        The month in which the fiscal year ends.
+
+    variation : str, default "nearest"
+        Method of employing 4-4-5 calendar.
+
+        There are two options:
+
+        - "nearest" means year end is **weekday** closest to last day of month in year.
+        - "last" means year end is final **weekday** of the final month in fiscal year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    In the example below the default parameters give the next 52-53 week fiscal year.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253()
+    Timestamp('2022-01-31 00:00:00')
+
+    By the parameter ``startingMonth`` we can specify
+    the month in which fiscal years end.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253(startingMonth=3)
+    Timestamp('2022-03-28 00:00:00')
+
+    52-53 week fiscal year can be specified by
+    ``weekday`` and ``variation`` parameters.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253(weekday=5, startingMonth=12, variation="last")
+    Timestamp('2022-12-31 00:00:00')
+    """
+
+    _prefix = "RE"
+    _attributes = tuple(["n", "normalize", "weekday", "startingMonth", "variation"])
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        dt = datetime(dt.year, dt.month, dt.day)
+        year_end = self.get_year_end(dt)
+
+        if self.variation == "nearest":
+            # We have to check the year end of "this" cal year AND the previous
+            return year_end == dt or self.get_year_end(shift_month(dt, -1, None)) == dt
+        else:
+            return year_end == dt
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        norm = Timestamp(other).normalize()
+
+        n = self.n
+        prev_year = self.get_year_end(datetime(other.year - 1, self.startingMonth, 1))
+        cur_year = self.get_year_end(datetime(other.year, self.startingMonth, 1))
+        next_year = self.get_year_end(datetime(other.year + 1, self.startingMonth, 1))
+
+        prev_year = localize_pydatetime(prev_year, other.tzinfo)
+        cur_year = localize_pydatetime(cur_year, other.tzinfo)
+        next_year = localize_pydatetime(next_year, other.tzinfo)
+
+        # Note: next_year.year == other.year + 1, so we will always
+        # have other < next_year
+        if norm == prev_year:
+            n -= 1
+        elif norm == cur_year:
+            pass
+        elif n > 0:
+            if norm < prev_year:
+                n -= 2
+            elif prev_year < norm < cur_year:
+                n -= 1
+            elif cur_year < norm < next_year:
+                pass
+        else:
+            if cur_year < norm < next_year:
+                n += 1
+            elif prev_year < norm < cur_year:
+                pass
+            elif (
+                norm.year == prev_year.year
+                and norm < prev_year
+                and prev_year - norm <= timedelta(6)
+            ):
+                # GH#14774, error when next_year.year == cur_year.year
+                # e.g. prev_year == datetime(2004, 1, 3),
+                # other == datetime(2004, 1, 1)
+                n -= 1
+            else:
+                assert False
+
+        shifted = datetime(other.year + n, self.startingMonth, 1)
+        result = self.get_year_end(shifted)
+        result = datetime(
+            result.year,
+            result.month,
+            result.day,
+            other.hour,
+            other.minute,
+            other.second,
+            other.microsecond,
+        )
+        return result
+
+    def get_year_end(self, dt: datetime) -> datetime:
+        assert dt.tzinfo is None
+
+        dim = get_days_in_month(dt.year, self.startingMonth)
+        target_date = datetime(dt.year, self.startingMonth, dim)
+        wkday_diff = self.weekday - target_date.weekday()
+        if wkday_diff == 0:
+            # year_end is the same for "last" and "nearest" cases
+            return target_date
+
+        if self.variation == "last":
+            days_forward = (wkday_diff % 7) - 7
+
+            # days_forward is always negative, so we always end up
+            # in the same year as dt
+            return target_date + timedelta(days=days_forward)
+        else:
+            # variation == "nearest":
+            days_forward = wkday_diff % 7
+            if days_forward <= 3:
+                # The upcoming self.weekday is closer than the previous one
+                return target_date + timedelta(days_forward)
+            else:
+                # The previous self.weekday is closer than the upcoming one
+                return target_date + timedelta(days_forward - 7)
+
+    @classmethod
+    def _parse_suffix(cls, varion_code, startingMonth_code, weekday_code):
+        if varion_code == "N":
+            variation = "nearest"
+        elif varion_code == "L":
+            variation = "last"
+        else:
+            raise ValueError(f"Unable to parse varion_code: {varion_code}")
+
+        startingMonth = MONTH_TO_CAL_NUM[startingMonth_code]
+        weekday = weekday_to_int[weekday_code]
+
+        return {
+            "weekday": weekday,
+            "startingMonth": startingMonth,
+            "variation": variation,
+        }
+
+    @classmethod
+    def _from_name(cls, *args):
+        return cls(**cls._parse_suffix(*args))
+
+
+cdef class FY5253Quarter(FY5253Mixin):
+    """
+    DateOffset increments between business quarter dates for 52-53 week fiscal year.
+
+    Also known as a 4-4-5 calendar.
+
+    It is used by companies that desire that their
+    fiscal year always end on the same day of the week.
+
+    It is a method of managing accounting periods.
+    It is a common calendar structure for some industries,
+    such as retail, manufacturing and parking industry.
+
+    For more information see:
+    https://en.wikipedia.org/wiki/4-4-5_calendar
+
+    The year may either:
+
+    - end on the last X day of the Y month.
+    - end on the last X day closest to the last day of the Y month.
+
+    X is a specific day of the week.
+    Y is a certain month of the year
+
+    startingMonth = 1 corresponds to dates like 1/31/2007, 4/30/2007, ...
+    startingMonth = 2 corresponds to dates like 2/28/2007, 5/31/2007, ...
+    startingMonth = 3 corresponds to dates like 3/30/2007, 6/29/2007, ...
+
+    Attributes
+    ----------
+    n : int
+        The number of business quarters represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekday : int {0, 1, ..., 6}, default 0
+        A specific integer for the day of the week.
+
+        - 0 is Monday
+        - 1 is Tuesday
+        - 2 is Wednesday
+        - 3 is Thursday
+        - 4 is Friday
+        - 5 is Saturday
+        - 6 is Sunday.
+
+    startingMonth : int {1, 2, ..., 12}, default 1
+        The month in which fiscal years end.
+
+    qtr_with_extra_week : int {1, 2, 3, 4}, default 1
+        The quarter number that has the leap or 14 week when needed.
+
+    variation : str, default "nearest"
+        Method of employing 4-4-5 calendar.
+
+        There are two options:
+
+        - "nearest" means year end is **weekday** closest to last day of month in year.
+        - "last" means year end is final **weekday** of the final month in fiscal year.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    In the example below the default parameters give
+    the next business quarter for 52-53 week fiscal year.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253Quarter()
+    Timestamp('2022-01-31 00:00:00')
+
+    By the parameter ``startingMonth`` we can specify
+    the month in which fiscal years end.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253Quarter(startingMonth=3)
+    Timestamp('2022-03-28 00:00:00')
+
+    Business quarters for 52-53 week fiscal year can be specified by
+    ``weekday`` and ``variation`` parameters.
+
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.FY5253Quarter(weekday=5, startingMonth=12, variation="last")
+    Timestamp('2022-04-02 00:00:00')
+    """
+
+    _prefix = "REQ"
+    _attributes = tuple(
+        [
+            "n",
+            "normalize",
+            "weekday",
+            "startingMonth",
+            "qtr_with_extra_week",
+            "variation",
+        ]
+    )
+
+    cdef readonly:
+        int qtr_with_extra_week
+
+    def __init__(
+        self,
+        n=1,
+        normalize=False,
+        weekday=0,
+        startingMonth=1,
+        qtr_with_extra_week=1,
+        variation="nearest",
+    ):
+        FY5253Mixin.__init__(
+            self, n, normalize, weekday, startingMonth, variation
+        )
+        self.qtr_with_extra_week = qtr_with_extra_week
+
+    cpdef __setstate__(self, state):
+        FY5253Mixin.__setstate__(self, state)
+        self.qtr_with_extra_week = state.pop("qtr_with_extra_week")
+
+    @cache_readonly
+    def _offset(self):
+        return FY5253(
+            startingMonth=self.startingMonth,
+            weekday=self.weekday,
+            variation=self.variation,
+        )
+
+    def _rollback_to_year(self, other: datetime):
+        """
+        Roll `other` back to the most recent date that was on a fiscal year
+        end.
+
+        Return the date of that year-end, the number of full quarters
+        elapsed between that year-end and other, and the remaining Timedelta
+        since the most recent quarter-end.
+
+        Parameters
+        ----------
+        other : datetime or Timestamp
+
+        Returns
+        -------
+        tuple of
+        prev_year_end : Timestamp giving most recent fiscal year end
+        num_qtrs : int
+        tdelta : Timedelta
+        """
+        num_qtrs = 0
+
+        norm = Timestamp(other).tz_localize(None)
+        start = self._offset.rollback(norm)
+        # Note: start <= norm and self._offset.is_on_offset(start)
+
+        if start < norm:
+            # roll adjustment
+            qtr_lens = self.get_weeks(norm)
+
+            # check that qtr_lens is consistent with self._offset addition
+            end = _shift_day(start, days=7 * sum(qtr_lens))
+            assert self._offset.is_on_offset(end), (start, end, qtr_lens)
+
+            tdelta = norm - start
+            for qlen in qtr_lens:
+                if qlen * 7 <= tdelta.days:
+                    num_qtrs += 1
+                    tdelta -= (
+                        <_Timedelta>Timedelta(days=qlen * 7)
+                    )._as_creso(norm._creso)
+                else:
+                    break
+        else:
+            tdelta = Timedelta(0)
+
+        # Note: we always have tdelta._value>= 0
+        return start, num_qtrs, tdelta
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        # Note: self.n == 0 is not allowed.
+
+        n = self.n
+
+        prev_year_end, num_qtrs, tdelta = self._rollback_to_year(other)
+        res = prev_year_end
+        n += num_qtrs
+        if self.n <= 0 and tdelta._value > 0:
+            n += 1
+
+        # Possible speedup by handling years first.
+        years = n // 4
+        if years:
+            res += self._offset * years
+            n -= years * 4
+
+        # Add an extra day to make *sure* we are getting the quarter lengths
+        # for the upcoming year, not the previous year
+        qtr_lens = self.get_weeks(res + Timedelta(days=1))
+
+        # Note: we always have 0 <= n < 4
+        weeks = sum(qtr_lens[:n])
+        if weeks:
+            res = _shift_day(res, days=weeks * 7)
+
+        return res
+
+    def get_weeks(self, dt: datetime):
+        ret = [13] * 4
+
+        year_has_extra_week = self.year_has_extra_week(dt)
+
+        if year_has_extra_week:
+            ret[self.qtr_with_extra_week - 1] = 14
+
+        return ret
+
+    def year_has_extra_week(self, dt: datetime) -> bool:
+        # Avoid round-down errors --> normalize to get
+        # e.g. '370D' instead of '360D23H'
+        norm = Timestamp(dt).normalize().tz_localize(None)
+
+        next_year_end = self._offset.rollforward(norm)
+        prev_year_end = norm - self._offset
+        weeks_in_year = (next_year_end - prev_year_end).days / 7
+        assert weeks_in_year in [52, 53], weeks_in_year
+        return weeks_in_year == 53
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        if self._offset.is_on_offset(dt):
+            return True
+
+        next_year_end = dt - self._offset
+
+        qtr_lens = self.get_weeks(dt)
+
+        current = next_year_end
+        for qtr_len in qtr_lens:
+            current = _shift_day(current, days=qtr_len * 7)
+            if dt == current:
+                return True
+        return False
+
+    @property
+    def rule_code(self) -> str:
+        suffix = FY5253Mixin.rule_code.__get__(self)
+        qtr = self.qtr_with_extra_week
+        return f"{suffix}-{qtr}"
+
+    @classmethod
+    def _from_name(cls, *args):
+        return cls(
+            **dict(FY5253._parse_suffix(*args[:-1]), qtr_with_extra_week=int(args[-1]))
+        )
+
+
+cdef class Easter(SingleConstructorOffset):
+    """
+    DateOffset for the Easter holiday using logic defined in dateutil.
+
+    Right now uses the revised method which is valid in years 1583-4099.
+
+    Attributes
+    ----------
+    n : int, default 1
+        The number of years represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    >>> ts = pd.Timestamp(2022, 1, 1)
+    >>> ts + pd.offsets.Easter()
+    Timestamp('2022-04-17 00:00:00')
+    """
+
+    cpdef __setstate__(self, state):
+        self.n = state.pop("n")
+        self.normalize = state.pop("normalize")
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        from dateutil.easter import easter
+
+        current_easter = easter(other.year)
+        current_easter = datetime(
+            current_easter.year, current_easter.month, current_easter.day
+        )
+        current_easter = localize_pydatetime(current_easter, other.tzinfo)
+
+        n = self.n
+        if n >= 0 and other < current_easter:
+            n -= 1
+        elif n < 0 and other > current_easter:
+            n += 1
+        # TODO: Why does this handle the 0 case the opposite of others?
+
+        # NOTE: easter returns a datetime.date so we have to convert to type of
+        # other
+        new = easter(other.year + n)
+        new = datetime(
+            new.year,
+            new.month,
+            new.day,
+            other.hour,
+            other.minute,
+            other.second,
+            other.microsecond,
+        )
+        return new
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+
+        from dateutil.easter import easter
+
+        return date(dt.year, dt.month, dt.day) == easter(dt.year)
+
+
+# ----------------------------------------------------------------------
+# Custom Offset classes
+
+
+cdef class CustomBusinessDay(BusinessDay):
+    """
+    DateOffset subclass representing possibly n custom business days.
+
+    In CustomBusinessDay we can use custom weekmask, holidays, and calendar.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of days represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekmask : str, Default 'Mon Tue Wed Thu Fri'
+        Weekmask of valid business days, passed to ``numpy.busdaycalendar``.
+    holidays : list
+        List/array of dates to exclude from the set of valid business days,
+        passed to ``numpy.busdaycalendar``.
+    calendar : np.busdaycalendar
+        Calendar to integrate.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    Examples
+    --------
+    In the example below the default parameters give the next business day.
+
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.CustomBusinessDay()
+    Timestamp('2022-08-08 16:00:00')
+
+    Business days can be specified by ``weekmask`` parameter. To convert
+    the returned datetime object to its string representation
+    the function strftime() is used in the next example.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.CustomBusinessDay(weekmask="Mon Wed Fri")
+    >>> pd.date_range(dt.datetime(2022, 12, 10), dt.datetime(2022, 12, 21),
+    ...               freq=freq).strftime('%a %d %b %Y %H:%M')
+    Index(['Mon 12 Dec 2022 00:00', 'Wed 14 Dec 2022 00:00',
+           'Fri 16 Dec 2022 00:00', 'Mon 19 Dec 2022 00:00',
+           'Wed 21 Dec 2022 00:00'],
+           dtype='object')
+
+    Using NumPy business day calendar you can define custom holidays.
+
+    >>> import datetime as dt
+    >>> bdc = np.busdaycalendar(holidays=['2022-12-12', '2022-12-14'])
+    >>> freq = pd.offsets.CustomBusinessDay(calendar=bdc)
+    >>> pd.date_range(dt.datetime(2022, 12, 10), dt.datetime(2022, 12, 25), freq=freq)
+    DatetimeIndex(['2022-12-13', '2022-12-15', '2022-12-16', '2022-12-19',
+                   '2022-12-20', '2022-12-21', '2022-12-22', '2022-12-23'],
+                   dtype='datetime64[ns]', freq='C')
+
+    If you want to shift the result on n day you can use the parameter ``offset``.
+
+    >>> pd.Timestamp(2022, 8, 5, 16) + pd.offsets.CustomBusinessDay(1)
+    Timestamp('2022-08-08 16:00:00')
+
+    >>> import datetime as dt
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.CustomBusinessDay(1, offset=dt.timedelta(days=1))
+    Timestamp('2022-08-09 16:00:00')
+    """
+
+    _prefix = "C"
+    _attributes = tuple(
+        ["n", "normalize", "weekmask", "holidays", "calendar", "offset"]
+    )
+
+    @property
+    def _period_dtype_code(self):
+        # GH#52534
+        raise ValueError(f"{self.base} is not supported as period frequency")
+
+    _apply_array = BaseOffset._apply_array
+
+    def __init__(
+        self,
+        n=1,
+        normalize=False,
+        weekmask="Mon Tue Wed Thu Fri",
+        holidays=None,
+        calendar=None,
+        offset=timedelta(0),
+    ):
+        BusinessDay.__init__(self, n, normalize, offset)
+        self._init_custom(weekmask, holidays, calendar)
+
+    cpdef __setstate__(self, state):
+        self.holidays = state.pop("holidays")
+        self.weekmask = state.pop("weekmask")
+        BusinessDay.__setstate__(self, state)
+
+    @apply_wraps
+    def _apply(self, other):
+        if self.n <= 0:
+            roll = "forward"
+        else:
+            roll = "backward"
+
+        if PyDateTime_Check(other):
+            date_in = other
+            np_dt = np.datetime64(date_in.date())
+
+            np_incr_dt = np.busday_offset(
+                np_dt, self.n, roll=roll, busdaycal=self.calendar
+            )
+
+            dt_date = np_incr_dt.astype(datetime)
+            result = datetime.combine(dt_date, date_in.time())
+
+            if self.offset:
+                result = result + self.offset
+            return result
+
+        elif is_any_td_scalar(other):
+            td = Timedelta(self.offset) + other
+            return BDay(self.n, offset=td.to_pytimedelta(), normalize=self.normalize)
+        else:
+            raise ApplyTypeError(
+                "Only know how to combine trading day with "
+                "datetime, datetime64 or timedelta."
+            )
+
+    def is_on_offset(self, dt: datetime) -> bool:
+        if self.normalize and not _is_normalized(dt):
+            return False
+        day64 = _to_dt64D(dt)
+        return np.is_busday(day64, busdaycal=self.calendar)
+
+
+cdef class CustomBusinessHour(BusinessHour):
+    """
+    DateOffset subclass representing possibly n custom business days.
+
+    In CustomBusinessHour we can use custom weekmask, holidays, and calendar.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of hours represented.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    weekmask : str, Default 'Mon Tue Wed Thu Fri'
+        Weekmask of valid business days, passed to ``numpy.busdaycalendar``.
+    holidays : list
+        List/array of dates to exclude from the set of valid business days,
+        passed to ``numpy.busdaycalendar``.
+    calendar : np.busdaycalendar
+        Calendar to integrate.
+    start : str, time, or list of str/time, default "09:00"
+        Start time of your custom business hour in 24h format.
+    end : str, time, or list of str/time, default: "17:00"
+        End time of your custom business hour in 24h format.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    Examples
+    --------
+    In the example below the default parameters give the next business hour.
+
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.CustomBusinessHour()
+    Timestamp('2022-08-08 09:00:00')
+
+    We can also change the start and the end of business hours.
+
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.CustomBusinessHour(start="11:00")
+    Timestamp('2022-08-08 11:00:00')
+
+    >>> from datetime import time as dt_time
+    >>> ts = pd.Timestamp(2022, 8, 5, 16)
+    >>> ts + pd.offsets.CustomBusinessHour(end=dt_time(19, 0))
+    Timestamp('2022-08-05 17:00:00')
+
+    >>> ts = pd.Timestamp(2022, 8, 5, 22)
+    >>> ts + pd.offsets.CustomBusinessHour(end=dt_time(19, 0))
+    Timestamp('2022-08-08 10:00:00')
+
+    You can divide your business day hours into several parts.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.CustomBusinessHour(start=["06:00", "10:00", "15:00"],
+    ...                                      end=["08:00", "12:00", "17:00"])
+    >>> pd.date_range(dt.datetime(2022, 12, 9), dt.datetime(2022, 12, 13), freq=freq)
+    DatetimeIndex(['2022-12-09 06:00:00', '2022-12-09 07:00:00',
+                   '2022-12-09 10:00:00', '2022-12-09 11:00:00',
+                   '2022-12-09 15:00:00', '2022-12-09 16:00:00',
+                   '2022-12-12 06:00:00', '2022-12-12 07:00:00',
+                   '2022-12-12 10:00:00', '2022-12-12 11:00:00',
+                   '2022-12-12 15:00:00', '2022-12-12 16:00:00'],
+                   dtype='datetime64[ns]', freq='cbh')
+
+    Business days can be specified by ``weekmask`` parameter. To convert
+    the returned datetime object to its string representation
+    the function strftime() is used in the next example.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.CustomBusinessHour(weekmask="Mon Wed Fri",
+    ...                                      start="10:00", end="13:00")
+    >>> pd.date_range(dt.datetime(2022, 12, 10), dt.datetime(2022, 12, 18),
+    ...               freq=freq).strftime('%a %d %b %Y %H:%M')
+    Index(['Mon 12 Dec 2022 10:00', 'Mon 12 Dec 2022 11:00',
+           'Mon 12 Dec 2022 12:00', 'Wed 14 Dec 2022 10:00',
+           'Wed 14 Dec 2022 11:00', 'Wed 14 Dec 2022 12:00',
+           'Fri 16 Dec 2022 10:00', 'Fri 16 Dec 2022 11:00',
+           'Fri 16 Dec 2022 12:00'],
+           dtype='object')
+
+    Using NumPy business day calendar you can define custom holidays.
+
+    >>> import datetime as dt
+    >>> bdc = np.busdaycalendar(holidays=['2022-12-12', '2022-12-14'])
+    >>> freq = pd.offsets.CustomBusinessHour(calendar=bdc, start="10:00", end="13:00")
+    >>> pd.date_range(dt.datetime(2022, 12, 10), dt.datetime(2022, 12, 18), freq=freq)
+    DatetimeIndex(['2022-12-13 10:00:00', '2022-12-13 11:00:00',
+                   '2022-12-13 12:00:00', '2022-12-15 10:00:00',
+                   '2022-12-15 11:00:00', '2022-12-15 12:00:00',
+                   '2022-12-16 10:00:00', '2022-12-16 11:00:00',
+                   '2022-12-16 12:00:00'],
+                   dtype='datetime64[ns]', freq='cbh')
+    """
+
+    _prefix = "cbh"
+    _anchor = 0
+    _attributes = tuple(
+        ["n", "normalize", "weekmask", "holidays", "calendar", "start", "end", "offset"]
+    )
+
+    def __init__(
+        self,
+        n=1,
+        normalize=False,
+        weekmask="Mon Tue Wed Thu Fri",
+        holidays=None,
+        calendar=None,
+        start="09:00",
+        end="17:00",
+        offset=timedelta(0),
+    ):
+        BusinessHour.__init__(self, n, normalize, start=start, end=end, offset=offset)
+        self._init_custom(weekmask, holidays, calendar)
+
+
+cdef class _CustomBusinessMonth(BusinessMixin):
+    _attributes = tuple(
+        ["n", "normalize", "weekmask", "holidays", "calendar", "offset"]
+    )
+
+    def __init__(
+        self,
+        n=1,
+        normalize=False,
+        weekmask="Mon Tue Wed Thu Fri",
+        holidays=None,
+        calendar=None,
+        offset=timedelta(0),
+    ):
+        BusinessMixin.__init__(self, n, normalize, offset)
+        self._init_custom(weekmask, holidays, calendar)
+
+    @cache_readonly
+    def cbday_roll(self):
+        """
+        Define default roll function to be called in apply method.
+        """
+        cbday_kwds = self.kwds.copy()
+        cbday_kwds["offset"] = timedelta(0)
+
+        cbday = CustomBusinessDay(n=1, normalize=False, **cbday_kwds)
+
+        if self._prefix.endswith("S"):
+            # MonthBegin
+            roll_func = cbday.rollforward
+        else:
+            # MonthEnd
+            roll_func = cbday.rollback
+        return roll_func
+
+    @cache_readonly
+    def m_offset(self):
+        if self._prefix.endswith("S"):
+            # MonthBegin
+            moff = MonthBegin(n=1, normalize=False)
+        else:
+            # MonthEnd
+            moff = MonthEnd(n=1, normalize=False)
+        return moff
+
+    @cache_readonly
+    def month_roll(self):
+        """
+        Define default roll function to be called in apply method.
+        """
+        if self._prefix.endswith("S"):
+            # MonthBegin
+            roll_func = self.m_offset.rollback
+        else:
+            # MonthEnd
+            roll_func = self.m_offset.rollforward
+        return roll_func
+
+    @apply_wraps
+    def _apply(self, other: datetime) -> datetime:
+        # First move to month offset
+        cur_month_offset_date = self.month_roll(other)
+
+        # Find this custom month offset
+        compare_date = self.cbday_roll(cur_month_offset_date)
+        n = roll_convention(other.day, self.n, compare_date.day)
+
+        new = cur_month_offset_date + n * self.m_offset
+        result = self.cbday_roll(new)
+
+        if self.offset:
+            result = result + self.offset
+        return result
+
+
+cdef class CustomBusinessMonthEnd(_CustomBusinessMonth):
+    """
+    DateOffset subclass representing custom business month(s).
+
+    Increments between end of month dates.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize end dates to midnight before generating date range.
+    weekmask : str, Default 'Mon Tue Wed Thu Fri'
+        Weekmask of valid business days, passed to ``numpy.busdaycalendar``.
+    holidays : list
+        List/array of dates to exclude from the set of valid business days,
+        passed to ``numpy.busdaycalendar``.
+    calendar : np.busdaycalendar
+        Calendar to integrate.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    In the example below we use the default parameters.
+
+    >>> ts = pd.Timestamp(2022, 8, 5)
+    >>> ts + pd.offsets.CustomBusinessMonthEnd()
+    Timestamp('2022-08-31 00:00:00')
+
+    Custom business month end can be specified by ``weekmask`` parameter.
+    To convert the returned datetime object to its string representation
+    the function strftime() is used in the next example.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.CustomBusinessMonthEnd(weekmask="Wed Thu")
+    >>> pd.date_range(dt.datetime(2022, 7, 10), dt.datetime(2022, 12, 18),
+    ...               freq=freq).strftime('%a %d %b %Y %H:%M')
+    Index(['Thu 28 Jul 2022 00:00', 'Wed 31 Aug 2022 00:00',
+           'Thu 29 Sep 2022 00:00', 'Thu 27 Oct 2022 00:00',
+           'Wed 30 Nov 2022 00:00'],
+           dtype='object')
+
+    Using NumPy business day calendar you can define custom holidays.
+
+    >>> import datetime as dt
+    >>> bdc = np.busdaycalendar(holidays=['2022-08-01', '2022-09-30',
+    ...                                   '2022-10-31', '2022-11-01'])
+    >>> freq = pd.offsets.CustomBusinessMonthEnd(calendar=bdc)
+    >>> pd.date_range(dt.datetime(2022, 7, 10), dt.datetime(2022, 11, 10), freq=freq)
+    DatetimeIndex(['2022-07-29', '2022-08-31', '2022-09-29', '2022-10-28'],
+                   dtype='datetime64[ns]', freq='CBME')
+    """
+
+    _prefix = "CBME"
+
+
+cdef class CustomBusinessMonthBegin(_CustomBusinessMonth):
+    """
+    DateOffset subclass representing custom business month(s).
+
+    Increments between beginning of month dates.
+
+    Parameters
+    ----------
+    n : int, default 1
+        The number of months represented.
+    normalize : bool, default False
+        Normalize start dates to midnight before generating date range.
+    weekmask : str, Default 'Mon Tue Wed Thu Fri'
+        Weekmask of valid business days, passed to ``numpy.busdaycalendar``.
+    holidays : list
+        List/array of dates to exclude from the set of valid business days,
+        passed to ``numpy.busdaycalendar``.
+    calendar : np.busdaycalendar
+        Calendar to integrate.
+    offset : timedelta, default timedelta(0)
+        Time offset to apply.
+
+    See Also
+    --------
+    :class:`~pandas.tseries.offsets.DateOffset` : Standard kind of date increment.
+
+    Examples
+    --------
+    In the example below we use the default parameters.
+
+    >>> ts = pd.Timestamp(2022, 8, 5)
+    >>> ts + pd.offsets.CustomBusinessMonthBegin()
+    Timestamp('2022-09-01 00:00:00')
+
+    Custom business month start can be specified by ``weekmask`` parameter.
+    To convert the returned datetime object to its string representation
+    the function strftime() is used in the next example.
+
+    >>> import datetime as dt
+    >>> freq = pd.offsets.CustomBusinessMonthBegin(weekmask="Wed Thu")
+    >>> pd.date_range(dt.datetime(2022, 7, 10), dt.datetime(2022, 12, 18),
+    ...               freq=freq).strftime('%a %d %b %Y %H:%M')
+    Index(['Wed 03 Aug 2022 00:00', 'Thu 01 Sep 2022 00:00',
+           'Wed 05 Oct 2022 00:00', 'Wed 02 Nov 2022 00:00',
+           'Thu 01 Dec 2022 00:00'],
+           dtype='object')
+
+    Using NumPy business day calendar you can define custom holidays.
+
+    >>> import datetime as dt
+    >>> bdc = np.busdaycalendar(holidays=['2022-08-01', '2022-09-30',
+    ...                                   '2022-10-31', '2022-11-01'])
+    >>> freq = pd.offsets.CustomBusinessMonthBegin(calendar=bdc)
+    >>> pd.date_range(dt.datetime(2022, 7, 10), dt.datetime(2022, 11, 10), freq=freq)
+    DatetimeIndex(['2022-08-02', '2022-09-01', '2022-10-03', '2022-11-02'],
+                   dtype='datetime64[ns]', freq='CBMS')
+    """
+
+    _prefix = "CBMS"
+
+
+BDay = BusinessDay
+BMonthEnd = BusinessMonthEnd
+BMonthBegin = BusinessMonthBegin
+CBMonthEnd = CustomBusinessMonthEnd
+CBMonthBegin = CustomBusinessMonthBegin
+CDay = CustomBusinessDay
+
+# ----------------------------------------------------------------------
+# to_offset helpers
+
+prefix_mapping = {
+    offset._prefix: offset
+    for offset in [
+        YearBegin,  # 'YS'
+        YearEnd,  # 'YE'
+        BYearBegin,  # 'BYS'
+        BYearEnd,  # 'BYE'
+        BusinessDay,  # 'B'
+        BusinessMonthBegin,  # 'BMS'
+        BusinessMonthEnd,  # 'BME'
+        BQuarterEnd,  # 'BQE'
+        BQuarterBegin,  # 'BQS'
+        BusinessHour,  # 'bh'
+        CustomBusinessDay,  # 'C'
+        CustomBusinessMonthEnd,  # 'CBME'
+        CustomBusinessMonthBegin,  # 'CBMS'
+        CustomBusinessHour,  # 'cbh'
+        MonthEnd,  # 'ME'
+        MonthBegin,  # 'MS'
+        Nano,  # 'ns'
+        SemiMonthEnd,  # 'SME'
+        SemiMonthBegin,  # 'SMS'
+        Week,  # 'W'
+        Second,  # 's'
+        Minute,  # 'min'
+        Micro,  # 'us'
+        QuarterEnd,  # 'QE'
+        QuarterBegin,  # 'QS'
+        Milli,  # 'ms'
+        Hour,  # 'h'
+        Day,  # 'D'
+        WeekOfMonth,  # 'WOM'
+        LastWeekOfMonth,  # 'LWOM'
+        FY5253,
+        FY5253Quarter,
+    ]
+}
+
+# hack to handle WOM-1MON
+opattern = re.compile(
+    r"([+\-]?\d*|[+\-]?\d*\.\d*)\s*([A-Za-z]+([\-][\dA-Za-z\-]+)?)"
+)
+
+_lite_rule_alias = {
+    "W": "W-SUN",
+    "QE": "QE-DEC",
+
+    "YE": "YE-DEC",      # YearEnd(month=12),
+    "YS": "YS-JAN",    # YearBegin(month=1),
+    "BYE": "BYE-DEC",    # BYearEnd(month=12),
+    "BYS": "BYS-JAN",  # BYearBegin(month=1),
+
+    "Min": "min",
+}
+
+_dont_uppercase = {"min", "h", "bh", "cbh", "s", "ms", "us", "ns"}
+
+
+INVALID_FREQ_ERR_MSG = "Invalid frequency: {0}"
+
+# TODO: still needed?
+# cache of previously seen offsets
+_offset_map = {}
+
+
+def _warn_about_deprecated_aliases(name: str, is_period: bool) -> str:
+    if name in _lite_rule_alias:
+        return name
+    if name in c_PERIOD_AND_OFFSET_DEPR_FREQSTR:
+        warnings.warn(
+            f"\'{name}\' is deprecated and will be removed "
+            f"in a future version, please use "
+            f"\'{c_PERIOD_AND_OFFSET_DEPR_FREQSTR.get(name)}\' "
+            f" instead.",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+            )
+        return c_PERIOD_AND_OFFSET_DEPR_FREQSTR[name]
+
+    for _name in (name.lower(), name.upper()):
+        if name == _name:
+            continue
+        if _name in c_PERIOD_AND_OFFSET_DEPR_FREQSTR.values():
+            warnings.warn(
+                f"\'{name}\' is deprecated and will be removed "
+                f"in a future version, please use "
+                f"\'{_name}\' "
+                f" instead.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+                )
+            return _name
+
+    return name
+
+
+def _validate_to_offset_alias(alias: str, is_period: bool) -> None:
+    if not is_period:
+        if alias.upper() in c_OFFSET_RENAMED_FREQSTR:
+            raise ValueError(
+                f"\'{alias}\' is no longer supported for offsets. Please "
+                f"use \'{c_OFFSET_RENAMED_FREQSTR.get(alias.upper())}\' "
+                f"instead."
+            )
+        if (alias.upper() != alias and
+                alias.lower() not in {"s", "ms", "us", "ns"} and
+                alias.upper().split("-")[0].endswith(("S", "E"))):
+            raise ValueError(INVALID_FREQ_ERR_MSG.format(alias))
+    if (
+        is_period and
+        alias in c_OFFSET_TO_PERIOD_FREQSTR and
+        alias != c_OFFSET_TO_PERIOD_FREQSTR[alias]
+    ):
+        alias_msg = c_OFFSET_TO_PERIOD_FREQSTR.get(alias)
+        raise ValueError(
+            f"for Period, please use \'{alias_msg}\' "
+            f"instead of \'{alias}\'"
+        )
+
+
+# TODO: better name?
+def _get_offset(name: str) -> BaseOffset:
+    """
+    Return DateOffset object associated with rule name.
+
+    Examples
+    --------
+    _get_offset('EOM') --> BMonthEnd(1)
+    """
+    if name not in _offset_map:
+        try:
+            split = name.split("-")
+            klass = prefix_mapping[split[0]]
+            # handles case where there's no suffix (and will TypeError if too
+            # many '-')
+            offset = klass._from_name(*split[1:])
+        except (ValueError, TypeError, KeyError) as err:
+            # bad prefix or suffix
+            raise ValueError(INVALID_FREQ_ERR_MSG.format(
+                f"{name}, failed to parse with error message: {repr(err)}")
+            )
+        # cache
+        _offset_map[name] = offset
+
+    return _offset_map[name]
+
+
+cpdef to_offset(freq, bint is_period=False):
+    """
+    Return DateOffset object from string or datetime.timedelta object.
+
+    Parameters
+    ----------
+    freq : str, datetime.timedelta, BaseOffset or None
+        The frequency represented.
+    is_period : bool, default False
+        Convert string denoting period frequency to corresponding offsets
+        frequency if is_period=True.
+
+    Returns
+    -------
+    BaseOffset subclass or None
+
+    Raises
+    ------
+    ValueError
+        If freq is an invalid frequency
+
+    See Also
+    --------
+    BaseOffset : Standard kind of date increment used for a date range.
+
+    Examples
+    --------
+    >>> from pandas.tseries.frequencies import to_offset
+    >>> to_offset("5min")
+    <5 * Minutes>
+
+    >>> to_offset("1D1h")
+    <25 * Hours>
+
+    >>> to_offset("2W")
+    <2 * Weeks: weekday=6>
+
+    >>> to_offset("2B")
+    <2 * BusinessDays>
+
+    >>> to_offset(pd.Timedelta(days=1))
+    <Day>
+
+    >>> to_offset(pd.offsets.Hour())
+    <Hour>
+
+    Passing the parameter ``is_period`` equal to True, you can use a string
+    denoting period frequency:
+
+    >>> freq = to_offset(freq="ME", is_period=False)
+    >>> freq.rule_code
+    'ME'
+
+    >>> freq = to_offset(freq="M", is_period=True)
+    >>> freq.rule_code
+    'ME'
+    """
+    if freq is None:
+        return None
+
+    if isinstance(freq, tuple):
+        raise TypeError(
+            f"to_offset does not support tuples {freq}, pass as a string instead"
+        )
+
+    if isinstance(freq, BaseOffset):
+        result = freq
+
+    elif PyDelta_Check(freq):
+        result = delta_to_tick(freq)
+
+    elif isinstance(freq, str):
+        result = None
+        stride_sign = None
+
+        try:
+            split = opattern.split(freq)
+            if split[-1] != "" and not split[-1].isspace():
+                # the last element must be blank
+                raise ValueError("last element must be blank")
+
+            tups = zip(split[0::4], split[1::4], split[2::4])
+            for n, (sep, stride, name) in enumerate(tups):
+                name = _warn_about_deprecated_aliases(name, is_period)
+                _validate_to_offset_alias(name, is_period)
+                if is_period:
+                    if name.upper() in c_PERIOD_TO_OFFSET_FREQSTR:
+                        if name.upper() != name:
+                            raise ValueError(
+                                f"\'{name}\' is no longer supported, "
+                                f"please use \'{name.upper()}\' instead.",
+                            )
+                        name = c_PERIOD_TO_OFFSET_FREQSTR[name.upper()]
+                name = _lite_rule_alias.get(name, name)
+
+                if sep != "" and not sep.isspace():
+                    raise ValueError("separator must be spaces")
+                if stride_sign is None:
+                    stride_sign = -1 if stride.startswith("-") else 1
+                if not stride:
+                    stride = 1
+
+                if name in {"D", "h", "min", "s", "ms", "us", "ns"}:
+                    # For these prefixes, we have something like "3h" or
+                    #  "2.5min", so we can construct a Timedelta with the
+                    #  matching unit and get our offset from delta_to_tick
+                    td = Timedelta(1, unit=name)
+                    off = delta_to_tick(td)
+                    offset = off * float(stride)
+                    if n != 0:
+                        # If n==0, then stride_sign is already incorporated
+                        #  into the offset
+                        offset *= stride_sign
+                else:
+                    stride = int(stride)
+                    offset = _get_offset(name)
+                    offset = offset * int(np.fabs(stride) * stride_sign)
+
+                if result is None:
+                    result = offset
+                else:
+                    result = result + offset
+        except (ValueError, TypeError) as err:
+            raise ValueError(INVALID_FREQ_ERR_MSG.format(
+                f"{freq}, failed to parse with error message: {repr(err)}")
+            ) from err
+    else:
+        result = None
+
+    if result is None:
+        raise ValueError(INVALID_FREQ_ERR_MSG.format(freq))
+
+    try:
+        has_period_dtype_code = hasattr(result, "_period_dtype_code")
+    except ValueError:
+        has_period_dtype_code = False
+
+    if is_period and not has_period_dtype_code:
+        if isinstance(freq, str):
+            raise ValueError(f"{result.name} is not supported as period frequency")
+        else:
+            raise ValueError(f"{freq} is not supported as period frequency")
+
+    return result
+
+
+# ----------------------------------------------------------------------
+# RelativeDelta Arithmetic
+
+cdef datetime _shift_day(datetime other, int days):
+    """
+    Increment the datetime `other` by the given number of days, retaining
+    the time-portion of the datetime.  For tz-naive datetimes this is
+    equivalent to adding a timedelta.  For tz-aware datetimes it is similar to
+    dateutil's relativedelta.__add__, but handles pytz tzinfo objects.
+
+    Parameters
+    ----------
+    other : datetime or Timestamp
+    days : int
+
+    Returns
+    -------
+    shifted: datetime or Timestamp
+    """
+    if other.tzinfo is None:
+        return other + timedelta(days=days)
+
+    tz = other.tzinfo
+    naive = other.replace(tzinfo=None)
+    shifted = naive + timedelta(days=days)
+    return localize_pydatetime(shifted, tz)
+
+
+cdef int year_add_months(npy_datetimestruct dts, int months) noexcept nogil:
+    """
+    New year number after shifting npy_datetimestruct number of months.
+    """
+    return dts.year + (dts.month + months - 1) // 12
+
+
+cdef int month_add_months(npy_datetimestruct dts, int months) noexcept nogil:
+    """
+    New month number after shifting npy_datetimestruct
+    number of months.
+    """
+    cdef:
+        int new_month = (dts.month + months) % 12
+    return 12 if new_month == 0 else new_month
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef ndarray shift_quarters(
+    ndarray dtindex,
+    int quarters,
+    int q1start_month,
+    str day_opt,
+    int modby=3,
+    NPY_DATETIMEUNIT reso=NPY_DATETIMEUNIT.NPY_FR_ns,
+):
+    """
+    Given an int64 array representing nanosecond timestamps, shift all elements
+    by the specified number of quarters using DateOffset semantics.
+
+    Parameters
+    ----------
+    dtindex : int64_t[:] timestamps for input dates
+    quarters : int number of quarters to shift
+    q1start_month : int month in which Q1 begins by convention
+    day_opt : {'start', 'end', 'business_start', 'business_end'}
+    modby : int (3 for quarters, 12 for years)
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    out : ndarray[int64_t]
+    """
+    if day_opt not in ["start", "end", "business_start", "business_end"]:
+        raise ValueError("day must be None, 'start', 'end', "
+                         "'business_start', or 'business_end'")
+
+    cdef:
+        Py_ssize_t count = dtindex.size
+        ndarray out = cnp.PyArray_EMPTY(dtindex.ndim, dtindex.shape, cnp.NPY_INT64, 0)
+        Py_ssize_t i
+        int64_t val, res_val
+        int months_since, n
+        npy_datetimestruct dts
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(out, dtindex)
+
+    with nogil:
+        for i in range(count):
+            # Analogous to: val = dtindex[i]
+            val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+            if val == NPY_NAT:
+                res_val = NPY_NAT
+            else:
+                pandas_datetime_to_datetimestruct(val, reso, &dts)
+                n = quarters
+
+                months_since = (dts.month - q1start_month) % modby
+                n = _roll_qtrday(&dts, n, months_since, day_opt)
+
+                dts.year = year_add_months(dts, modby * n - months_since)
+                dts.month = month_add_months(dts, modby * n - months_since)
+                dts.day = get_day_of_month(&dts, day_opt)
+
+                res_val = npy_datetimestruct_to_datetime(reso, &dts)
+
+            # Analogous to: out[i] = res_val
+            (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+            cnp.PyArray_MultiIter_NEXT(mi)
+
+    return out
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def shift_months(
+    ndarray dtindex,  # int64_t, arbitrary ndim
+    int months,
+    str day_opt=None,
+    NPY_DATETIMEUNIT reso=NPY_DATETIMEUNIT.NPY_FR_ns,
+):
+    """
+    Given an int64-based datetime index, shift all elements
+    specified number of months using DateOffset semantics
+
+    day_opt: {None, 'start', 'end', 'business_start', 'business_end'}
+       * None: day of month
+       * 'start' 1st day of month
+       * 'end' last day of month
+    """
+    cdef:
+        Py_ssize_t i
+        npy_datetimestruct dts
+        int count = dtindex.size
+        ndarray out = cnp.PyArray_EMPTY(dtindex.ndim, dtindex.shape, cnp.NPY_INT64, 0)
+        int months_to_roll
+        int64_t val, res_val
+
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(out, dtindex)
+
+    if day_opt is not None and day_opt not in {
+            "start", "end", "business_start", "business_end"
+    }:
+        raise ValueError("day must be None, 'start', 'end', "
+                         "'business_start', or 'business_end'")
+
+    if day_opt is None:
+        # TODO: can we combine this with the non-None case?
+        with nogil:
+            for i in range(count):
+                # Analogous to: val = i8other[i]
+                val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+                if val == NPY_NAT:
+                    res_val = NPY_NAT
+                else:
+                    pandas_datetime_to_datetimestruct(val, reso, &dts)
+                    dts.year = year_add_months(dts, months)
+                    dts.month = month_add_months(dts, months)
+
+                    dts.day = min(dts.day, get_days_in_month(dts.year, dts.month))
+                    res_val = npy_datetimestruct_to_datetime(reso, &dts)
+
+                # Analogous to: out[i] = res_val
+                (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+                cnp.PyArray_MultiIter_NEXT(mi)
+
+    else:
+        with nogil:
+            for i in range(count):
+
+                # Analogous to: val = i8other[i]
+                val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+                if val == NPY_NAT:
+                    res_val = NPY_NAT
+                else:
+                    pandas_datetime_to_datetimestruct(val, reso, &dts)
+                    months_to_roll = months
+
+                    months_to_roll = _roll_qtrday(&dts, months_to_roll, 0, day_opt)
+
+                    dts.year = year_add_months(dts, months_to_roll)
+                    dts.month = month_add_months(dts, months_to_roll)
+                    dts.day = get_day_of_month(&dts, day_opt)
+
+                    res_val = npy_datetimestruct_to_datetime(reso, &dts)
+
+                # Analogous to: out[i] = res_val
+                (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+                cnp.PyArray_MultiIter_NEXT(mi)
+
+    return out
+
+
+def shift_month(stamp: datetime, months: int, day_opt: object = None) -> datetime:
+    """
+    Given a datetime (or Timestamp) `stamp`, an integer `months` and an
+    option `day_opt`, return a new datetimelike that many months later,
+    with day determined by `day_opt` using relativedelta semantics.
+
+    Scalar analogue of shift_months.
+
+    Parameters
+    ----------
+    stamp : datetime or Timestamp
+    months : int
+    day_opt : None, 'start', 'end', 'business_start', 'business_end', or int
+        None: returned datetimelike has the same day as the input, or the
+              last day of the month if the new month is too short
+        'start': returned datetimelike has day=1
+        'end': returned datetimelike has day on the last day of the month
+        'business_start': returned datetimelike has day on the first
+            business day of the month
+        'business_end': returned datetimelike has day on the last
+            business day of the month
+        int: returned datetimelike has day equal to day_opt
+
+    Returns
+    -------
+    shifted : datetime or Timestamp (same as input `stamp`)
+    """
+    cdef:
+        int year, month, day
+        int days_in_month, dy
+
+    dy = (stamp.month + months) // 12
+    month = (stamp.month + months) % 12
+
+    if month == 0:
+        month = 12
+        dy -= 1
+    year = stamp.year + dy
+
+    if day_opt is None:
+        days_in_month = get_days_in_month(year, month)
+        day = min(stamp.day, days_in_month)
+    elif day_opt == "start":
+        day = 1
+    elif day_opt == "end":
+        day = get_days_in_month(year, month)
+    elif day_opt == "business_start":
+        # first business day of month
+        day = get_firstbday(year, month)
+    elif day_opt == "business_end":
+        # last business day of month
+        day = get_lastbday(year, month)
+    elif is_integer_object(day_opt):
+        days_in_month = get_days_in_month(year, month)
+        day = min(day_opt, days_in_month)
+    else:
+        raise ValueError(day_opt)
+    return stamp.replace(year=year, month=month, day=day)
+
+
+cdef int get_day_of_month(npy_datetimestruct* dts, str day_opt) noexcept nogil:
+    """
+    Find the day in `other`'s month that satisfies a DateOffset's is_on_offset
+    policy, as described by the `day_opt` argument.
+
+    Parameters
+    ----------
+    dts : npy_datetimestruct*
+    day_opt : {'start', 'end', 'business_start', 'business_end'}
+        'start': returns 1
+        'end': returns last day of the month
+        'business_start': returns the first business day of the month
+        'business_end': returns the last business day of the month
+
+    Returns
+    -------
+    day_of_month : int
+
+    Examples
+    -------
+    >>> other = datetime(2017, 11, 14)
+    >>> get_day_of_month(other, 'start')
+    1
+    >>> get_day_of_month(other, 'end')
+    30
+
+    Notes
+    -----
+    Caller is responsible for ensuring one of the four accepted day_opt values
+    is passed.
+    """
+
+    if day_opt == "start":
+        return 1
+    elif day_opt == "end":
+        return get_days_in_month(dts.year, dts.month)
+    elif day_opt == "business_start":
+        # first business day of month
+        return get_firstbday(dts.year, dts.month)
+    else:
+        # i.e. day_opt == "business_end":
+        # last business day of month
+        return get_lastbday(dts.year, dts.month)
+
+
+cpdef int roll_convention(int other, int n, int compare) noexcept nogil:
+    """
+    Possibly increment or decrement the number of periods to shift
+    based on rollforward/rollbackward conventions.
+
+    Parameters
+    ----------
+    other : int, generally the day component of a datetime
+    n : number of periods to increment, before adjusting for rolling
+    compare : int, generally the day component of a datetime, in the same
+              month as the datetime form which `other` was taken.
+
+    Returns
+    -------
+    n : int number of periods to increment
+    """
+    if n > 0 and other < compare:
+        n -= 1
+    elif n <= 0 and other > compare:
+        # as if rolled forward already
+        n += 1
+    return n
+
+
+def roll_qtrday(other: datetime, n: int, month: int,
+                day_opt: str, modby: int) -> int:
+    """
+    Possibly increment or decrement the number of periods to shift
+    based on rollforward/rollbackward conventions.
+
+    Parameters
+    ----------
+    other : datetime or Timestamp
+    n : number of periods to increment, before adjusting for rolling
+    month : int reference month giving the first month of the year
+    day_opt : {'start', 'end', 'business_start', 'business_end'}
+        The convention to use in finding the day in a given month against
+        which to compare for rollforward/rollbackward decisions.
+    modby : int 3 for quarters, 12 for years
+
+    Returns
+    -------
+    n : int number of periods to increment
+
+    See Also
+    --------
+    get_day_of_month : Find the day in a month provided an offset.
+    """
+    cdef:
+        int months_since
+        npy_datetimestruct dts
+
+    if day_opt not in ["start", "end", "business_start", "business_end"]:
+        raise ValueError(day_opt)
+
+    pydate_to_dtstruct(other, &dts)
+
+    if modby == 12:
+        # We care about the month-of-year, not month-of-quarter, so skip mod
+        months_since = other.month - month
+    else:
+        months_since = other.month % modby - month % modby
+
+    return _roll_qtrday(&dts, n, months_since, day_opt)
+
+
+cdef int _roll_qtrday(npy_datetimestruct* dts,
+                      int n,
+                      int months_since,
+                      str day_opt) except? -1 nogil:
+    """
+    See roll_qtrday.__doc__
+    """
+
+    if n > 0:
+        if months_since < 0 or (months_since == 0 and
+                                dts.day < get_day_of_month(dts, day_opt)):
+            # pretend to roll back if on same month but
+            # before compare_day
+            n -= 1
+    else:
+        if months_since > 0 or (months_since == 0 and
+                                dts.day > get_day_of_month(dts, day_opt)):
+            # make sure to roll forward, so negate
+            n += 1
+    return n
diff --git a/pandas/pandas/_libs/tslibs/parsing.pxd b/pandas/pandas/_libs/tslibs/parsing.pxd
new file mode 100644
index 0000000000000..fbe07e68f5adf
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/parsing.pxd
@@ -0,0 +1,16 @@
+from cpython.datetime cimport datetime
+from numpy cimport int64_t
+
+from pandas._libs.tslibs.np_datetime cimport NPY_DATETIMEUNIT
+
+
+cpdef str get_rule_month(str source)
+cpdef quarter_to_myear(int year, int quarter, str freq)
+
+cdef datetime parse_datetime_string(
+    str date_string,
+    bint dayfirst,
+    bint yearfirst,
+    NPY_DATETIMEUNIT* out_bestunit,
+    int64_t* nanos,
+)
diff --git a/pandas/pandas/_libs/tslibs/parsing.pyi b/pandas/pandas/_libs/tslibs/parsing.pyi
new file mode 100644
index 0000000000000..845bd9a5a5635
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/parsing.pyi
@@ -0,0 +1,30 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas._typing import npt
+
+class DateParseError(ValueError): ...
+
+def py_parse_datetime_string(
+    date_string: str,
+    dayfirst: bool = ...,
+    yearfirst: bool = ...,
+) -> datetime: ...
+def parse_datetime_string_with_reso(
+    date_string: str,
+    freq: str | None = ...,
+    dayfirst: bool | None = ...,
+    yearfirst: bool | None = ...,
+) -> tuple[datetime, str]: ...
+def _does_string_look_like_datetime(py_string: str) -> bool: ...
+def quarter_to_myear(year: int, quarter: int, freq: str) -> tuple[int, int]: ...
+def try_parse_dates(
+    values: npt.NDArray[np.object_],  # object[:]
+    parser,
+) -> npt.NDArray[np.object_]: ...
+def guess_datetime_format(
+    dt_str: str,
+    dayfirst: bool | None = ...,
+) -> str | None: ...
+def get_rule_month(source: str) -> str: ...
diff --git a/pandas/pandas/_libs/tslibs/parsing.pyx b/pandas/pandas/_libs/tslibs/parsing.pyx
new file mode 100644
index 0000000000000..308183402198d
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/parsing.pyx
@@ -0,0 +1,1124 @@
+"""
+Parsing functions for datetime and datetime-like strings.
+"""
+import re
+import time
+import warnings
+
+from pandas.util._exceptions import find_stack_level
+
+from cpython.datetime cimport (
+    datetime,
+    datetime_new,
+    import_datetime,
+    timedelta,
+    tzinfo,
+)
+
+from datetime import timezone
+
+from cpython.unicode cimport PyUnicode_AsUTF8AndSize
+from cython cimport Py_ssize_t
+from libc.string cimport strchr
+
+import_datetime()
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport int64_t
+
+cnp.import_array()
+
+# dateutil compat
+
+from decimal import InvalidOperation
+
+from dateutil.parser import DEFAULTPARSER
+from dateutil.tz import (
+    tzoffset,
+    tzutc as _dateutil_tzutc,
+)
+
+from pandas._config import get_option
+
+from pandas._libs.tslibs.ccalendar cimport MONTH_TO_CAL_NUM
+from pandas._libs.tslibs.dtypes cimport (
+    attrname_to_npy_unit,
+    npy_unit_to_attrname,
+)
+from pandas._libs.tslibs.nattype cimport (
+    c_NaT as NaT,
+    c_nat_strings as nat_strings,
+)
+
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    string_to_dts,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.strptime import array_strptime
+
+
+cdef extern from "pandas/portable.h":
+    int getdigit_ascii(char c, int default) nogil
+
+cdef extern from "pandas/parser/tokenizer.h":
+    double xstrtod(const char *p, char **q, char decimal, char sci, char tsep,
+                   int skip_trailing, int *error, int *maybe_int)
+
+
+# ----------------------------------------------------------------------
+# Constants
+
+
+class DateParseError(ValueError):
+    pass
+
+
+_DEFAULT_DATETIME = datetime(1, 1, 1).replace(hour=0, minute=0,
+                                              second=0, microsecond=0)
+
+cdef:
+    set _not_datelike_strings = {"a", "A", "m", "M", "p", "P", "t", "T"}
+
+    # _timestamp_units -> units that we round to nanos
+    set _timestamp_units = {
+        NPY_DATETIMEUNIT.NPY_FR_ns,
+        NPY_DATETIMEUNIT.NPY_FR_ps,
+        NPY_DATETIMEUNIT.NPY_FR_fs,
+        NPY_DATETIMEUNIT.NPY_FR_as,
+    }
+
+# ----------------------------------------------------------------------
+cdef:
+    const char* delimiters = " /-."
+    int MAX_DAYS_IN_MONTH = 31, MAX_MONTH = 12
+
+
+cdef bint _is_delimiter(const char ch):
+    return strchr(delimiters, ch) != NULL
+
+
+cdef int _parse_1digit(const char* s):
+    cdef int result = 0
+    result += getdigit_ascii(s[0], -10) * 1
+    return result
+
+
+cdef int _parse_2digit(const char* s):
+    cdef int result = 0
+    result += getdigit_ascii(s[0], -10) * 10
+    result += getdigit_ascii(s[1], -100) * 1
+    return result
+
+
+cdef int _parse_4digit(const char* s):
+    cdef int result = 0
+    result += getdigit_ascii(s[0], -10) * 1000
+    result += getdigit_ascii(s[1], -100) * 100
+    result += getdigit_ascii(s[2], -1000) * 10
+    result += getdigit_ascii(s[3], -10000) * 1
+    return result
+
+
+cdef datetime _parse_delimited_date(
+    str date_string, bint dayfirst, NPY_DATETIMEUNIT* out_bestunit
+):
+    """
+    Parse special cases of dates: MM/DD/YYYY, DD/MM/YYYY, MM/YYYY.
+
+    At the beginning function tries to parse date in MM/DD/YYYY format, but
+    if month > 12 - in DD/MM/YYYY (`dayfirst == False`).
+    With `dayfirst == True` function makes an attempt to parse date in
+    DD/MM/YYYY, if an attempt is wrong - in DD/MM/YYYY
+
+    For MM/DD/YYYY, DD/MM/YYYY: delimiter can be a space or one of /-.
+    For MM/YYYY: delimiter can be a space or one of /-
+    If `date_string` can't be converted to date, then function returns
+    None, None
+
+    Parameters
+    ----------
+    date_string : str
+    dayfirst : bool
+    out_bestunit : NPY_DATETIMEUNIT*
+        For specifying identified resolution.
+
+    Returns:
+    --------
+    datetime or None
+    """
+    cdef:
+        const char* buf
+        Py_ssize_t length
+        int day = 1, month = 1, year
+        bint can_swap = 0
+
+    buf = PyUnicode_AsUTF8AndSize(date_string, &length)
+    if length == 10 and _is_delimiter(buf[2]) and _is_delimiter(buf[5]):
+        # parsing MM?DD?YYYY and DD?MM?YYYY dates
+        month = _parse_2digit(buf)
+        day = _parse_2digit(buf + 3)
+        year = _parse_4digit(buf + 6)
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_D
+        can_swap = 1
+    elif length == 9 and _is_delimiter(buf[1]) and _is_delimiter(buf[4]):
+        # parsing M?DD?YYYY and D?MM?YYYY dates
+        month = _parse_1digit(buf)
+        day = _parse_2digit(buf + 2)
+        year = _parse_4digit(buf + 5)
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_D
+        can_swap = 1
+    elif length == 9 and _is_delimiter(buf[2]) and _is_delimiter(buf[4]):
+        # parsing MM?D?YYYY and DD?M?YYYY dates
+        month = _parse_2digit(buf)
+        day = _parse_1digit(buf + 3)
+        year = _parse_4digit(buf + 5)
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_D
+        can_swap = 1
+    elif length == 8 and _is_delimiter(buf[1]) and _is_delimiter(buf[3]):
+        # parsing M?D?YYYY and D?M?YYYY dates
+        month = _parse_1digit(buf)
+        day = _parse_1digit(buf + 2)
+        year = _parse_4digit(buf + 4)
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_D
+        can_swap = 1
+    elif length == 7 and _is_delimiter(buf[2]):
+        # parsing MM?YYYY dates
+        if buf[2] == b".":
+            # we cannot reliably tell whether e.g. 10.2010 is a float
+            # or a date, thus we refuse to parse it here
+            return None
+        month = _parse_2digit(buf)
+        year = _parse_4digit(buf + 3)
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_M
+    else:
+        return None
+
+    if month < 0 or day < 0 or year < 1000:
+        # some part is not an integer, so
+        # date_string can't be converted to date, above format
+        return None
+
+    if 1 <= month <= MAX_DAYS_IN_MONTH and 1 <= day <= MAX_DAYS_IN_MONTH \
+            and (month <= MAX_MONTH or day <= MAX_MONTH):
+        if (month > MAX_MONTH or (day <= MAX_MONTH and dayfirst)) and can_swap:
+            day, month = month, day
+        # In Python <= 3.6.0 there is no range checking for invalid dates
+        # in C api, thus we call faster C version for 3.6.1 or newer
+        return datetime_new(year, month, day, 0, 0, 0, 0, None)
+
+    raise DateParseError(f"Invalid date specified ({month}/{day})")
+
+
+cdef bint _does_string_look_like_time(str parse_string):
+    """
+    Checks whether given string is a time: it has to start either from
+    H:MM or from HH:MM, and hour and minute values must be valid.
+
+    Parameters
+    ----------
+    parse_string : str
+
+    Returns:
+    --------
+    bool
+        Whether given string is potentially a time.
+    """
+    cdef:
+        const char* buf
+        Py_ssize_t length
+        int hour = -1, minute = -1
+
+    buf = PyUnicode_AsUTF8AndSize(parse_string, &length)
+    if length >= 4:
+        if buf[1] == b":":
+            # h:MM format
+            hour = getdigit_ascii(buf[0], -1)
+            minute = _parse_2digit(buf + 2)
+        elif buf[2] == b":":
+            # HH:MM format
+            hour = _parse_2digit(buf)
+            minute = _parse_2digit(buf + 3)
+
+    return 0 <= hour <= 23 and 0 <= minute <= 59
+
+
+def py_parse_datetime_string(
+    str date_string, bint dayfirst=False, bint yearfirst=False
+):
+    # Python-accessible version for testing (we can't just make
+    #  parse_datetime_string cpdef bc it has a pointer argument)
+    cdef:
+        NPY_DATETIMEUNIT out_bestunit
+        int64_t nanos
+
+    return parse_datetime_string(
+        date_string, dayfirst, yearfirst, &out_bestunit, &nanos
+    )
+
+
+cdef datetime parse_datetime_string(
+    # NB: This will break with np.str_ (GH#32264) even though
+    #  isinstance(npstrobj, str) evaluates to True, so caller must ensure
+    #  the argument is *exactly* 'str'
+    str date_string,
+    bint dayfirst,
+    bint yearfirst,
+    NPY_DATETIMEUNIT* out_bestunit,
+    int64_t* nanos,
+):
+    """
+    Parse datetime string, only returns datetime.
+    Also cares special handling matching time patterns.
+
+    Returns
+    -------
+    datetime
+
+    Notes
+    -----
+    Does not handle "today" or "now", which caller is responsible for handling.
+    """
+
+    cdef:
+        datetime dt
+        bint is_quarter = 0
+
+    if not _does_string_look_like_datetime(date_string):
+        raise ValueError(f'Given date string "{date_string}" not likely a datetime')
+
+    if _does_string_look_like_time(date_string):
+        # time without date e.g. "01:01:01.111"
+        # use current datetime as default, not pass _DEFAULT_DATETIME
+        default = datetime.now().replace(hour=0, minute=0, second=0, microsecond=0)
+        dt = dateutil_parse(date_string, default=default,
+                            dayfirst=dayfirst, yearfirst=yearfirst,
+                            ignoretz=False, out_bestunit=out_bestunit, nanos=nanos)
+        return dt
+
+    dt = _parse_delimited_date(date_string, dayfirst, out_bestunit)
+    if dt is not None:
+        return dt
+
+    try:
+        dt = _parse_dateabbr_string(
+            date_string, _DEFAULT_DATETIME, None, out_bestunit, &is_quarter
+        )
+        return dt
+    except DateParseError:
+        raise
+    except ValueError:
+        pass
+
+    dt = dateutil_parse(date_string, default=_DEFAULT_DATETIME,
+                        dayfirst=dayfirst, yearfirst=yearfirst,
+                        ignoretz=False, out_bestunit=out_bestunit, nanos=nanos)
+    return dt
+
+
+def parse_datetime_string_with_reso(
+    str date_string, str freq=None, dayfirst=None, yearfirst=None
+):
+    # NB: This will break with np.str_ (GH#45580) even though
+    #  isinstance(npstrobj, str) evaluates to True, so caller must ensure
+    #  the argument is *exactly* 'str'
+    """
+    Try hard to parse datetime string, leveraging dateutil plus some extra
+    goodies like quarter recognition.
+
+    Parameters
+    ----------
+    date_string : str
+    freq : str or None, default None
+        Helps with interpreting time string if supplied
+        Corresponds to `offset.rule_code`
+    dayfirst : bool, default None
+        If None uses default from print_config
+    yearfirst : bool, default None
+        If None uses default from print_config
+
+    Returns
+    -------
+    datetime
+    str
+        Describing resolution of parsed string.
+
+    Raises
+    ------
+    ValueError : preliminary check suggests string is not datetime
+    DateParseError : error within dateutil
+    """
+
+    if dayfirst is None:
+        dayfirst = get_option("display.date_dayfirst")
+    if yearfirst is None:
+        yearfirst = get_option("display.date_yearfirst")
+
+    cdef:
+        datetime parsed
+        str reso
+        bint string_to_dts_failed
+        npy_datetimestruct dts
+        NPY_DATETIMEUNIT out_bestunit
+        int out_local = 0
+        int out_tzoffset
+        tzinfo tz
+        bint is_quarter = 0
+
+    if not _does_string_look_like_datetime(date_string):
+        raise ValueError(f'Given date string "{date_string}" not likely a datetime')
+
+    # Try iso8601 first, as it handles nanoseconds
+    if not dayfirst:  # GH 58859
+        string_to_dts_failed = string_to_dts(
+            date_string, &dts, &out_bestunit, &out_local,
+            &out_tzoffset, False
+        )
+        if not string_to_dts_failed:
+            # Match Timestamp and drop picoseconds, femtoseconds, attoseconds
+            # The new resolution will just be nano
+            # GH#50417
+            if out_bestunit in _timestamp_units:
+                out_bestunit = NPY_DATETIMEUNIT.NPY_FR_ns
+
+            if out_bestunit == NPY_DATETIMEUNIT.NPY_FR_ns:
+                # TODO: avoid circular import
+                from pandas import Timestamp
+                parsed = Timestamp(date_string)
+            else:
+                if out_local:
+                    tz = timezone(timedelta(minutes=out_tzoffset))
+                else:
+                    tz = None
+                parsed = datetime_new(
+                    dts.year, dts.month, dts.day, dts.hour, dts.min, dts.sec, dts.us, tz
+                )
+
+            reso = npy_unit_to_attrname[out_bestunit]
+            return parsed, reso
+
+    parsed = _parse_delimited_date(date_string, dayfirst, &out_bestunit)
+    if parsed is not None:
+        reso = npy_unit_to_attrname[out_bestunit]
+        return parsed, reso
+
+    try:
+        parsed = _parse_dateabbr_string(
+            date_string, _DEFAULT_DATETIME, freq, &out_bestunit, &is_quarter
+        )
+    except DateParseError:
+        raise
+    except ValueError:
+        pass
+    else:
+        if is_quarter:
+            reso = "quarter"
+        else:
+            reso = npy_unit_to_attrname[out_bestunit]
+        return parsed, reso
+
+    parsed = dateutil_parse(date_string, _DEFAULT_DATETIME,
+                            dayfirst=dayfirst, yearfirst=yearfirst,
+                            ignoretz=False, out_bestunit=&out_bestunit, nanos=NULL)
+    reso = npy_unit_to_attrname[out_bestunit]
+    return parsed, reso
+
+
+cpdef bint _does_string_look_like_datetime(str py_string):
+    """
+    Checks whether given string is a datetime: it has to start with '0' or
+    be greater than 1000.
+
+    Parameters
+    ----------
+    py_string: str
+
+    Returns
+    -------
+    bool
+        Whether given string is potentially a datetime.
+    """
+    cdef:
+        const char *buf
+        char *endptr = NULL
+        Py_ssize_t length = -1
+        double converted_date
+        char first
+        int error = 0
+
+    buf = PyUnicode_AsUTF8AndSize(py_string, &length)
+    if length >= 1:
+        first = buf[0]
+        if first == b"0":
+            # Strings starting with 0 are more consistent with a
+            # date-like string than a number
+            return True
+        elif py_string in _not_datelike_strings:
+            return False
+        else:
+            # xstrtod with such parameters copies behavior of python `float`
+            # cast; for example, " 35.e-1 " is valid string for this cast so,
+            # for correctly xstrtod call necessary to pass these params:
+            # b'.' - a dot is used as separator, b'e' - an exponential form of
+            # a float number can be used, b'\0' - not to use a thousand
+            # separator, 1 - skip extra spaces before and after,
+            converted_date = xstrtod(buf, &endptr,
+                                     b".", b"e", b"\0", 1, &error, NULL)
+            # if there were no errors and the whole line was parsed, then ...
+            if error == 0 and endptr == buf + length:
+                return converted_date >= 1000
+
+    return True
+
+
+cdef datetime _parse_dateabbr_string(str date_string, datetime default,
+                                     str freq, NPY_DATETIMEUNIT* out_bestunit,
+                                     bint* is_quarter):
+    # special handling for possibilities eg, 2Q2005, 2Q05, 2005Q1, 05Q1
+    cdef:
+        datetime ret
+        # year initialized to prevent compiler warnings
+        int year = -1, quarter = -1, month
+        Py_ssize_t date_len
+        const char* buf
+
+    if date_string in nat_strings:
+        # default to nanos, could also reasonably do NPY_FR_GENERIC
+        out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_ns
+        return NaT
+
+    date_string = date_string.upper()
+    date_len = len(date_string)
+
+    if date_len == 4:
+        # parse year only like 2000
+        try:
+            ret = default.replace(year=int(date_string))
+            out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_Y
+            return ret
+        except ValueError:
+            pass
+
+    if 4 <= date_len <= 7:
+        buf = PyUnicode_AsUTF8AndSize(date_string, &date_len)
+        try:
+            i = date_string.index("Q", 1, 6)
+            if i == 1:
+                quarter = _parse_1digit(buf)  # i.e. int(date_string[0])
+                if date_len == 4 or (date_len == 5
+                                     and date_string[i + 1] == "-"):
+                    # r'(\d)Q-?(\d\d)')
+                    year = 2000 + int(date_string[-2:])
+                elif date_len == 6 or (date_len == 7
+                                       and date_string[i + 1] == "-"):
+                    # r'(\d)Q-?(\d\d\d\d)')
+                    year = int(date_string[-4:])
+                else:
+                    raise ValueError
+            elif i == 2 or i == 3:
+                # r'(\d\d)-?Q(\d)'
+                if date_len == 4 or (date_len == 5
+                                     and date_string[i - 1] == "-"):
+                    # i.e. quarter = int(date_string[-1])
+                    quarter = _parse_1digit(buf + date_len - 1)
+                    year = 2000 + int(date_string[:2])
+                else:
+                    raise ValueError
+            elif i == 4 or i == 5:
+                if date_len == 6 or (date_len == 7
+                                     and date_string[i - 1] == "-"):
+                    # r'(\d\d\d\d)-?Q(\d)'
+                    # i.e. quarter = int(date_string[-1])
+                    quarter = _parse_1digit(buf + date_len - 1)
+                    year = int(date_string[:4])
+                else:
+                    raise ValueError
+
+            if not (1 <= quarter <= 4):
+                raise DateParseError(f"Incorrect quarterly string is given, "
+                                     f"quarter must be "
+                                     f"between 1 and 4: {date_string}")
+
+            try:
+                # GH#1228
+                year, month = quarter_to_myear(year, quarter, freq)
+            except KeyError:
+                raise DateParseError("Unable to retrieve month "
+                                     "information from given "
+                                     f"freq: {freq}")
+
+            ret = default.replace(year=year, month=month)
+            # Monthly is as close as we can get to a non-existent NPY_FR_Q
+            out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_M
+            is_quarter[0] = 1
+            return ret
+
+        except DateParseError:
+            raise
+        except ValueError:
+            # e.g. if "Q" is not in date_string and .index raised
+            pass
+
+    if date_len == 6 and freq == "ME":
+        year = int(date_string[:4])
+        month = int(date_string[4:6])
+        try:
+            ret = default.replace(year=year, month=month)
+            out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_M
+            return ret
+        except ValueError as err:
+            # We can infer that none of the patterns below will match
+            raise ValueError(f"Unable to parse {date_string}") from err
+
+    for pat in ["%Y-%m", "%b %Y", "%b-%Y"]:
+        try:
+            ret = datetime.strptime(date_string, pat)
+            out_bestunit[0] = NPY_DATETIMEUNIT.NPY_FR_M
+            return ret
+        except ValueError:
+            pass
+
+    raise ValueError(f"Unable to parse {date_string}")
+
+
+cpdef quarter_to_myear(int year, int quarter, str freq):
+    """
+    A quarterly frequency defines a "year" which may not coincide with
+    the calendar-year.  Find the calendar-year and calendar-month associated
+    with the given year and quarter under the `freq`-derived calendar.
+
+    Parameters
+    ----------
+    year : int
+    quarter : int
+    freq : str or None
+
+    Returns
+    -------
+    year : int
+    month : int
+
+    See Also
+    --------
+    Period.qyear
+    """
+    if quarter <= 0 or quarter > 4:
+        raise ValueError("Quarter must be 1 <= q <= 4")
+
+    if freq is not None:
+        mnum = MONTH_TO_CAL_NUM[get_rule_month(freq)]
+        month = (mnum + (quarter - 1) * 3) % 12 + 1
+        if month > mnum:
+            year -= 1
+    else:
+        month = (quarter - 1) * 3 + 1
+
+    return year, month
+
+
+cdef datetime dateutil_parse(
+    str timestr,
+    datetime default,
+    bint ignoretz,
+    bint dayfirst,
+    bint yearfirst,
+    NPY_DATETIMEUNIT* out_bestunit,
+    int64_t* nanos,
+):
+    """ lifted from dateutil to get resolution"""
+
+    cdef:
+        str attr
+        datetime ret
+        object res
+        str reso = None
+        dict repl = {}
+
+    try:
+        res, _ = DEFAULTPARSER._parse(timestr, dayfirst=dayfirst, yearfirst=yearfirst)
+    except InvalidOperation:
+        # GH#51157 dateutil can raise decimal.InvalidOperation
+        res = None
+
+    if res is None:
+        raise DateParseError(
+            f"Unknown datetime string format, unable to parse: {timestr}"
+        )
+
+    for attr in ["year", "month", "day", "hour",
+                 "minute", "second", "microsecond"]:
+        value = getattr(res, attr)
+        if value is not None:
+            repl[attr] = value
+            reso = attr
+
+    if reso is None:
+        raise DateParseError(f"Unable to parse datetime string: {timestr}")
+
+    if reso == "microsecond" and repl["microsecond"] % 1000 == 0:
+        reso = _find_subsecond_reso(timestr, nanos=nanos)
+
+    try:
+        ret = default.replace(**repl)
+    except ValueError as err:
+        # e.g. "day is out of range for month"
+        # we re-raise to match dateutil's exception message
+        raise DateParseError(str(err) + ": " + timestr) from err
+    except OverflowError as err:
+        # with e.g. "08335394550" dateutil raises when trying to pass
+        #  year=8335394550 to datetime.replace
+        raise OutOfBoundsDatetime(
+            f'Parsing "{timestr}" to datetime overflows'
+        ) from err
+
+    if res.weekday is not None and not res.day:
+        # GH#52659
+        raise ValueError(
+            "Parsing datetimes with weekday but no day information is "
+            "not supported"
+        )
+    if not ignoretz:
+        if res.tzname and res.tzname in time.tzname:
+            # GH#50791
+            if res.tzname != "UTC":
+                raise ValueError(
+                    f"Parsing '{res.tzname}' as tzlocal (dependent on system timezone) "
+                    "is no longer supported. Pass the 'tz' "
+                    "keyword or call tz_localize after construction instead",
+                )
+            ret = ret.replace(tzinfo=timezone.utc)
+        elif res.tzoffset == 0:
+            ret = ret.replace(tzinfo=_dateutil_tzutc())
+        elif res.tzoffset:
+            ret = ret.replace(tzinfo=tzoffset(res.tzname, res.tzoffset))
+
+            # dateutil can return a datetime with a tzoffset outside of (-24h, 24h)
+            #  bounds, which is invalid (can be constructed, but raises if we call
+            #  str(ret)).  Check that and raise here if necessary.
+            try:
+                ret.utcoffset()
+            except ValueError as err:
+                # offset must be a timedelta strictly between -timedelta(hours=24)
+                #  and timedelta(hours=24)
+                raise ValueError(
+                    f'Parsed string "{timestr}" gives an invalid tzoffset, '
+                    "which must be between -timedelta(hours=24) and timedelta(hours=24)"
+                )
+        elif res.tzname is not None:
+            # e.g. "1994 Jan 15 05:16 FOO" where FOO is not recognized
+            # GH#18702, # GH 50235 enforced in 3.0
+            raise ValueError(
+                f'Parsed string "{timestr}" included an un-recognized timezone '
+                f'"{res.tzname}".'
+            )
+
+    out_bestunit[0] = attrname_to_npy_unit[reso]
+    return ret
+
+
+cdef object _reso_pattern = re.compile(r"\d:\d{2}:\d{2}\.(?P<frac>\d+)")
+
+cdef _find_subsecond_reso(str timestr, int64_t* nanos):
+    # GH#55737
+    # Check for trailing zeros in a H:M:S.f pattern
+    match = _reso_pattern.search(timestr)
+    if not match:
+        reso = "second"
+    else:
+        frac = match.groupdict()["frac"]
+        if len(frac) <= 3:
+            reso = "millisecond"
+        elif len(frac) > 6:
+            if frac[6:] == "0" * len(frac[6:]):
+                # corner case where we haven't lost any data
+                reso = "nanosecond"
+            elif len(frac) <= 9:
+                reso = "nanosecond"
+                if nanos is not NULL:
+                    if len(frac) < 9:
+                        frac = frac + "0" * (9 - len(frac))
+                    nanos[0] = int(frac[6:])
+            else:
+                # TODO: should we warn/raise in higher-than-nano cases?
+                reso = "nanosecond"
+                if nanos is not NULL:
+                    nanos[0] = int(frac[6:9])
+        else:
+            reso = "microsecond"
+    return reso
+
+
+# ----------------------------------------------------------------------
+# Parsing for type-inference
+
+
+def try_parse_dates(object[:] values, parser) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, n
+        object[::1] result
+
+    n = len(values)
+    result = np.empty(n, dtype="O")
+
+    for i in range(n):
+        if values[i] == "":
+            result[i] = np.nan
+        else:
+            result[i] = parser(values[i])
+
+    return result.base  # .base to access underlying ndarray
+
+
+# ----------------------------------------------------------------------
+# Miscellaneous
+
+
+# Class copied verbatim from https://github.com/dateutil/dateutil/pull/732
+#
+# We use this class to parse and tokenize date strings. However, as it is
+# a private class in the dateutil library, relying on backwards compatibility
+# is not practical. In fact, using this class issues warnings (xref gh-21322).
+# Thus, we port the class over so that both issues are resolved.
+#
+# Licence at LICENSES/DATEUTIL_LICENSE
+class _timelex:
+    def __init__(self, instream):
+        if getattr(instream, "decode", None) is not None:
+            instream = instream.decode()
+
+        if isinstance(instream, str):
+            self.stream = instream
+        elif getattr(instream, "read", None) is None:
+            raise TypeError(
+                "Parser must be a string or character stream, not "
+                f"{type(instream).__name__}")
+        else:
+            self.stream = instream.read()
+
+    def get_tokens(self):
+        """
+        This function breaks the time string into lexical units (tokens), which
+        can be parsed by the parser. Lexical units are demarcated by changes in
+        the character set, so any continuous string of letters is considered
+        one unit, any continuous string of numbers is considered one unit.
+        The main complication arises from the fact that dots ('.') can be used
+        both as separators (e.g. "Sep.20.2009") or decimal points (e.g.
+        "4:30:21.447"). As such, it is necessary to read the full context of
+        any dot-separated strings before breaking it into tokens; as such, this
+        function maintains a "token stack", for when the ambiguous context
+        demands that multiple tokens be parsed at once.
+        """
+        cdef:
+            Py_ssize_t n
+
+        stream = self.stream.replace("\x00", "")
+
+        # TODO: Change \s --> \s+ (this doesn't match existing behavior)
+        # TODO: change the punctuation block to punc+ (does not match existing)
+        # TODO: can we merge the two digit patterns?
+        tokens = re.findall(r"\s|"
+                            r"(?<![\.\d])\d+\.\d+(?![\.\d])"
+                            r"|\d+"
+                            r"|[a-zA-Z]+"
+                            r"|[\./:]+"
+                            r"|[^\da-zA-Z\./:\s]+", stream)
+
+        # Re-combine token tuples of the form ["59", ",", "456"] because
+        # in this context the "," is treated as a decimal
+        # (e.g. in python's default logging format)
+        for n, token in enumerate(tokens[:-2]):
+            # Kludge to match ,-decimal behavior; it'd be better to do this
+            # later in the process and have a simpler tokenization
+            if (token is not None and token.isdigit() and
+                    tokens[n + 1] == "," and tokens[n + 2].isdigit()):
+                # Have to check None b/c it might be replaced during the loop
+                # TODO: I _really_ don't faking the value here
+                tokens[n] = token + "." + tokens[n + 2]
+                tokens[n + 1] = None
+                tokens[n + 2] = None
+
+        tokens = [x for x in tokens if x is not None]
+        return tokens
+
+    @classmethod
+    def split(cls, s):
+        return cls(s).get_tokens()
+
+
+_DATEUTIL_LEXER_SPLIT = _timelex.split
+
+
+def guess_datetime_format(dt_str: str, bint dayfirst=False) -> str | None:
+    """
+    Guess the datetime format of a given datetime string.
+
+    This function attempts to deduce the format of a given datetime string. It is
+    useful for situations where the datetime format is unknown and needs to be
+    determined for proper parsing. The function is not guaranteed to return a format.
+
+    Parameters
+    ----------
+    dt_str : str
+        Datetime string to guess the format of.
+    dayfirst : bool, default False
+        If True parses dates with the day first, eg 20/01/2005
+
+        .. warning::
+            dayfirst=True is not strict, but will prefer to parse
+            with day first (this is a known bug).
+
+    Returns
+    -------
+    str or None : ret
+        datetime format string (for `strftime` or `strptime`),
+        or None if it can't be guessed.
+
+    See Also
+    --------
+    to_datetime : Convert argument to datetime.
+    Timestamp : Pandas replacement for python datetime.datetime object.
+    DatetimeIndex : Immutable ndarray-like of datetime64 data.
+
+    Examples
+    --------
+    >>> from pandas.tseries.api import guess_datetime_format
+    >>> guess_datetime_format('09/13/2023')
+    '%m/%d/%Y'
+
+    >>> guess_datetime_format('2023|September|13')
+    """
+    cdef:
+        NPY_DATETIMEUNIT out_bestunit
+    day_attribute_and_format = (("day",), "%d", 2)
+
+    # attr name, format, padding (if any)
+    datetime_attrs_to_format = [
+        (("year", "month", "day", "hour", "minute", "second"), "%Y%m%d%H%M%S", 0),
+        (("year", "month", "day", "hour", "minute"), "%Y%m%d%H%M", 0),
+        (("year", "month", "day", "hour"), "%Y%m%d%H", 0),
+        (("year", "month", "day"), "%Y%m%d", 0),
+        (("hour", "minute", "second"), "%H%M%S", 0),
+        (("year",), "%Y", 0),
+        (("hour", "minute"), "%H%M", 0),
+        (("month",), "%B", 0),
+        (("month",), "%b", 0),
+        (("month",), "%m", 2),
+        day_attribute_and_format,
+        (("hour",), "%H", 2),
+        (("minute",), "%M", 2),
+        (("second",), "%S", 2),
+        (("second", "microsecond"), "%S.%f", 0),
+        (("tzinfo",), "%z", 0),
+        (("tzinfo",), "%Z", 0),
+        (("day_of_week",), "%a", 0),
+        (("day_of_week",), "%A", 0),
+        (("meridiem",), "%p", 0),
+    ]
+
+    if dayfirst:
+        datetime_attrs_to_format.remove(day_attribute_and_format)
+        datetime_attrs_to_format.insert(0, day_attribute_and_format)
+
+    # Use this instead of the dateutil default of
+    # `datetime.now().replace(hour=0, minute=0, second=0, microsecond=0)`
+    # as that causes issues on the 29th of February.
+    default = datetime(1970, 1, 1)
+    try:
+        parsed_datetime = dateutil_parse(
+            dt_str,
+            default=default,
+            dayfirst=dayfirst,
+            yearfirst=False,
+            ignoretz=False,
+            out_bestunit=&out_bestunit,
+            nanos=NULL,
+        )
+    except (ValueError, OverflowError, InvalidOperation):
+        # In case the datetime can't be parsed, its format cannot be guessed
+        return None
+
+    if parsed_datetime is None:
+        return None
+
+    # _DATEUTIL_LEXER_SPLIT from dateutil will never raise here
+    tokens = _DATEUTIL_LEXER_SPLIT(dt_str)
+
+    # Normalize offset part of tokens.
+    # There are multiple formats for the timezone offset.
+    # To pass the comparison condition between the output of `strftime` and
+    # joined tokens, which is carried out at the final step of the function,
+    # the offset part of the tokens must match the '%z' format like '+0900'
+    # instead of ‘+09:00’.
+    if parsed_datetime.tzinfo is not None:
+        offset_index = None
+        if len(tokens) > 0 and tokens[-1] == "Z":
+            # the last 'Z' means zero offset
+            offset_index = -1
+        elif len(tokens) > 1 and tokens[-2] in ("+", "-"):
+            # ex. [..., '+', '0900']
+            offset_index = -2
+        elif len(tokens) > 3 and tokens[-4] in ("+", "-"):
+            # ex. [..., '+', '09', ':', '00']
+            offset_index = -4
+
+        if offset_index is not None:
+            # If the input string has a timezone offset like '+0900',
+            # the offset is separated into two tokens, ex. ['+', '0900’].
+            # This separation will prevent subsequent processing
+            # from correctly parsing the time zone format.
+            # So in addition to the format normalization, we rejoin them here.
+            try:
+                tokens[offset_index] = parsed_datetime.strftime("%z")
+            except ValueError:
+                # Invalid offset might not have raised in du_parse
+                # https://github.com/dateutil/dateutil/issues/188
+                return None
+            tokens = tokens[:offset_index + 1 or None]
+
+    format_guess = [None] * len(tokens)
+    found_attrs = set()
+
+    for attrs, attr_format, padding in datetime_attrs_to_format:
+        # If a given attribute has been placed in the format string, skip
+        # over other formats for that same underlying attribute (IE, month
+        # can be represented in multiple different ways)
+        if set(attrs) & found_attrs:
+            continue
+
+        if parsed_datetime.tzinfo is None and attr_format in ("%Z", "%z"):
+            continue
+
+        parsed_formatted = parsed_datetime.strftime(attr_format)
+        for i, token_format in enumerate(format_guess):
+            token_filled = _fill_token(tokens[i], padding)
+            if token_format is None and token_filled == parsed_formatted:
+                format_guess[i] = attr_format
+                tokens[i] = token_filled
+                found_attrs.update(attrs)
+                break
+
+    # Only consider it a valid guess if we have a year, month and day.
+    # We make exceptions for %Y and %Y-%m (only with the `-` separator)
+    # as they conform with ISO8601.
+    if (
+        len({"year", "month", "day"} & found_attrs) != 3
+        and format_guess != ["%Y"]
+        and not (
+            format_guess == ["%Y", None, "%m"] and tokens[1] == "-"
+        )
+    ):
+        return None
+
+    output_format = []
+    for i, guess in enumerate(format_guess):
+        if guess is not None:
+            # Either fill in the format placeholder (like %Y)
+            output_format.append(guess)
+        else:
+            # Or just the token separate (IE, the dashes in "01-01-2013")
+            try:
+                # If the token is numeric, then we likely didn't parse it
+                # properly, so our guess is wrong
+                float(tokens[i])
+                return None
+            except ValueError:
+                pass
+
+            output_format.append(tokens[i])
+
+    # if am/pm token present, replace 24-hour %H, with 12-hour %I
+    if "%p" in output_format and "%H" in output_format:
+        i = output_format.index("%H")
+        output_format[i] = "%I"
+
+    guessed_format = "".join(output_format)
+
+    try:
+        array_strptime(np.asarray([dt_str], dtype=object), guessed_format)
+    except ValueError:
+        # Doesn't parse, so this can't be the correct format.
+        return None
+    # rebuild string, capturing any inferred padding
+    dt_str = "".join(tokens)
+    if parsed_datetime.strftime(guessed_format) == dt_str:
+        _maybe_warn_about_dayfirst(guessed_format, dayfirst)
+        return guessed_format
+    else:
+        return None
+
+
+cdef str _fill_token(token: str, padding: int):
+    cdef str token_filled
+    if re.search(r"\d+\.\d+", token) is None:
+        # For example: 98
+        token_filled = token.zfill(padding)
+    else:
+        # For example: 00.123
+        seconds, nanoseconds = token.split(".")
+        seconds = f"{int(seconds):02d}"
+        # right-pad so we get nanoseconds, then only take
+        # first 6 digits (microseconds) as stdlib datetime
+        # doesn't support nanoseconds
+        nanoseconds = nanoseconds.ljust(9, "0")[:6]
+        token_filled = f"{seconds}.{nanoseconds}"
+    return token_filled
+
+
+cdef void _maybe_warn_about_dayfirst(format: str, bint dayfirst) noexcept:
+    """Warn if guessed datetime format doesn't respect dayfirst argument."""
+    cdef:
+        int day_index = format.find("%d")
+        int month_index = format.find("%m")
+
+    if (day_index != -1) and (month_index != -1):
+        if (day_index > month_index) and dayfirst:
+            warnings.warn(
+                f"Parsing dates in {format} format when dayfirst=True was specified. "
+                "Pass `dayfirst=False` or specify a format to silence this warning.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+        if (day_index < month_index) and not dayfirst:
+            warnings.warn(
+                f"Parsing dates in {format} format when dayfirst=False (the default) "
+                "was specified. "
+                "Pass `dayfirst=True` or specify a format to silence this warning.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+
+cpdef str get_rule_month(str source):
+    """
+    Return starting month of given freq, default is December.
+
+    Parameters
+    ----------
+    source : str
+        Derived from `freq.rule_code` or `freq.freqstr`.
+
+    Returns
+    -------
+    rule_month: str
+
+    Examples
+    --------
+    >>> get_rule_month('D')
+    'DEC'
+
+    >>> get_rule_month('A-JAN')
+    'JAN'
+    """
+    source = source.upper()
+    if "-" not in source:
+        return "DEC"
+    else:
+        return source.split("-")[1]
diff --git a/pandas/pandas/_libs/tslibs/period.pxd b/pandas/pandas/_libs/tslibs/period.pxd
new file mode 100644
index 0000000000000..8afe7c9b3f693
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/period.pxd
@@ -0,0 +1,7 @@
+from numpy cimport int64_t
+
+from .np_datetime cimport npy_datetimestruct
+
+
+cdef bint is_period_object(object obj)
+cdef int64_t get_period_ordinal(npy_datetimestruct *dts, int freq) noexcept nogil
diff --git a/pandas/pandas/_libs/tslibs/period.pyi b/pandas/pandas/_libs/tslibs/period.pyi
new file mode 100644
index 0000000000000..22f3bdbe668de
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/period.pyi
@@ -0,0 +1,135 @@
+from datetime import timedelta
+from typing import Literal
+
+import numpy as np
+
+from pandas._libs.tslibs.dtypes import PeriodDtypeBase
+from pandas._libs.tslibs.nattype import NaTType
+from pandas._libs.tslibs.offsets import BaseOffset
+from pandas._libs.tslibs.timestamps import Timestamp
+from pandas._typing import (
+    Frequency,
+    npt,
+)
+
+INVALID_FREQ_ERR_MSG: str
+DIFFERENT_FREQ: str
+
+class IncompatibleFrequency(ValueError): ...
+
+def periodarr_to_dt64arr(
+    periodarr: npt.NDArray[np.int64],  # const int64_t[:]
+    freq: int,
+) -> npt.NDArray[np.int64]: ...
+def period_asfreq_arr(
+    arr: npt.NDArray[np.int64],
+    freq1: int,
+    freq2: int,
+    end: bool,
+) -> npt.NDArray[np.int64]: ...
+def get_period_field_arr(
+    field: str,
+    arr: npt.NDArray[np.int64],  # const int64_t[:]
+    freq: int,
+) -> npt.NDArray[np.int64]: ...
+def from_ordinals(
+    values: npt.NDArray[np.int64],  # const int64_t[:]
+    freq: timedelta | BaseOffset | str,
+) -> npt.NDArray[np.int64]: ...
+def extract_ordinals(
+    values: npt.NDArray[np.object_],
+    freq: Frequency | int,
+) -> npt.NDArray[np.int64]: ...
+def extract_freq(
+    values: npt.NDArray[np.object_],
+) -> BaseOffset: ...
+def period_array_strftime(
+    values: npt.NDArray[np.int64],
+    dtype_code: int,
+    na_rep,
+    date_format: str | None,
+) -> npt.NDArray[np.object_]: ...
+
+# exposed for tests
+def period_asfreq(ordinal: int, freq1: int, freq2: int, end: bool) -> int: ...
+def period_ordinal(
+    y: int, m: int, d: int, h: int, min: int, s: int, us: int, ps: int, freq: int
+) -> int: ...
+def freq_to_dtype_code(freq: BaseOffset) -> int: ...
+def validate_end_alias(how: str) -> Literal["E", "S"]: ...
+
+class PeriodMixin:
+    @property
+    def end_time(self) -> Timestamp: ...
+    @property
+    def start_time(self) -> Timestamp: ...
+    def _require_matching_freq(self, other: BaseOffset, base: bool = ...) -> None: ...
+
+class Period(PeriodMixin):
+    ordinal: int  # int64_t
+    freq: BaseOffset
+    _dtype: PeriodDtypeBase
+
+    # error: "__new__" must return a class instance (got "Union[Period, NaTType]")
+    def __new__(  # type: ignore[misc]
+        cls,
+        value=...,
+        freq: int | str | BaseOffset | None = ...,
+        ordinal: int | None = ...,
+        year: int | None = ...,
+        month: int | None = ...,
+        quarter: int | None = ...,
+        day: int | None = ...,
+        hour: int | None = ...,
+        minute: int | None = ...,
+        second: int | None = ...,
+    ) -> Period | NaTType: ...
+    @classmethod
+    def _maybe_convert_freq(cls, freq) -> BaseOffset: ...
+    @classmethod
+    def _from_ordinal(cls, ordinal: int, freq: BaseOffset) -> Period: ...
+    @classmethod
+    def now(cls, freq: Frequency) -> Period: ...
+    def strftime(self, fmt: str | None) -> str: ...
+    def to_timestamp(
+        self,
+        freq: str | BaseOffset | None = ...,
+        how: str = ...,
+    ) -> Timestamp: ...
+    def asfreq(self, freq: str | BaseOffset, how: str = ...) -> Period: ...
+    @property
+    def freqstr(self) -> str: ...
+    @property
+    def is_leap_year(self) -> bool: ...
+    @property
+    def daysinmonth(self) -> int: ...
+    @property
+    def days_in_month(self) -> int: ...
+    @property
+    def qyear(self) -> int: ...
+    @property
+    def quarter(self) -> int: ...
+    @property
+    def day_of_year(self) -> int: ...
+    @property
+    def weekday(self) -> int: ...
+    @property
+    def day_of_week(self) -> int: ...
+    @property
+    def week(self) -> int: ...
+    @property
+    def weekofyear(self) -> int: ...
+    @property
+    def second(self) -> int: ...
+    @property
+    def minute(self) -> int: ...
+    @property
+    def hour(self) -> int: ...
+    @property
+    def day(self) -> int: ...
+    @property
+    def month(self) -> int: ...
+    @property
+    def year(self) -> int: ...
+    def __sub__(self, other) -> Period | BaseOffset: ...
+    def __add__(self, other) -> Period: ...
diff --git a/pandas/pandas/_libs/tslibs/period.pyx b/pandas/pandas/_libs/tslibs/period.pyx
new file mode 100644
index 0000000000000..bef1956996b4f
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/period.pyx
@@ -0,0 +1,3081 @@
+import re
+
+cimport numpy as cnp
+from cpython.object cimport (
+    Py_EQ,
+    Py_NE,
+    PyObject,
+    PyObject_RichCompare,
+    PyObject_RichCompareBool,
+)
+from numpy cimport (
+    int32_t,
+    int64_t,
+    ndarray,
+)
+
+import numpy as np
+
+cnp.import_array()
+
+cimport cython
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    datetime,
+    import_datetime,
+)
+from libc.stdlib cimport (
+    free,
+    malloc,
+)
+from libc.string cimport (
+    memset,
+    strlen,
+)
+from libc.time cimport (
+    strftime,
+    tm,
+)
+
+from pandas._libs.tslibs.dtypes cimport c_OFFSET_TO_PERIOD_FREQSTR
+
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+
+# import datetime C API
+import_datetime()
+
+cimport pandas._libs.tslibs.util as util
+from pandas._libs.missing cimport C_NA
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_D,
+    astype_overflowsafe,
+    dts_to_iso_string,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pandas_datetime_to_datetimestruct,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.timestamps import Timestamp
+
+from pandas._libs.tslibs.ccalendar cimport (
+    dayofweek,
+    get_day_of_year,
+    get_days_in_month,
+    get_week_of_year,
+    is_leapyear,
+)
+from pandas._libs.tslibs.timedeltas cimport (
+    delta_to_nanoseconds,
+    is_any_td_scalar,
+)
+
+from pandas._libs.tslibs.conversion import DT64NS_DTYPE
+
+from pandas._libs.tslibs.dtypes cimport (
+    FR_ANN,
+    FR_BUS,
+    FR_DAY,
+    FR_HR,
+    FR_MIN,
+    FR_MS,
+    FR_MTH,
+    FR_NS,
+    FR_QTR,
+    FR_SEC,
+    FR_UND,
+    FR_US,
+    FR_WK,
+    PeriodDtypeBase,
+    attrname_to_abbrevs,
+    freq_group_code_to_npy_unit,
+)
+from pandas._libs.tslibs.parsing cimport quarter_to_myear
+
+from pandas._libs.tslibs.parsing import parse_datetime_string_with_reso
+
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+    c_nat_strings as nat_strings,
+    checknull_with_nat,
+)
+from pandas._libs.tslibs.offsets cimport (
+    BaseOffset,
+    is_offset_object,
+    to_offset,
+)
+
+from pandas._libs.tslibs.offsets import (
+    INVALID_FREQ_ERR_MSG,
+    BDay,
+)
+from pandas.util._decorators import set_module
+
+cdef:
+    enum:
+        INT32_MIN = -2_147_483_648LL
+
+
+ctypedef struct asfreq_info:
+    int64_t intraday_conversion_factor
+    int is_end
+    int to_end
+    int from_end
+
+ctypedef int64_t (*freq_conv_func)(int64_t, asfreq_info*) noexcept nogil
+
+
+cdef extern from *:
+    """
+    // must use npy typedef b/c int64_t is aliased in cython-generated c
+    // unclear why we need LL for that row.
+    // see https://github.com/pandas-dev/pandas/pull/34416/
+    static npy_int64 daytime_conversion_factor_matrix[7][7] = {
+        {1, 24, 1440, 86400, 86400000, 86400000000, 86400000000000},
+        {0LL,  1LL,   60LL,  3600LL,  3600000LL,  3600000000LL,  3600000000000LL},
+        {0,  0,   1,     60,    60000,    60000000,    60000000000},
+        {0,  0,   0,      1,     1000,     1000000,     1000000000},
+        {0,  0,   0,      0,        1,        1000,        1000000},
+        {0,  0,   0,      0,        0,           1,           1000},
+        {0,  0,   0,      0,        0,           0,              1}};
+    """
+    int64_t daytime_conversion_factor_matrix[7][7]
+
+
+cdef int max_value(int left, int right) noexcept nogil:
+    if left > right:
+        return left
+    return right
+
+
+cdef int min_value(int left, int right) noexcept nogil:
+    if left < right:
+        return left
+    return right
+
+
+cdef int64_t get_daytime_conversion_factor(int from_index, int to_index) noexcept nogil:
+    cdef:
+        int row = min_value(from_index, to_index)
+        int col = max_value(from_index, to_index)
+    # row or col < 6 means frequency strictly lower than Daily, which
+    # do not use daytime_conversion_factors
+    if row < 6:
+        return 0
+    elif col < 6:
+        return 0
+    return daytime_conversion_factor_matrix[row - 6][col - 6]
+
+
+cdef int64_t nofunc(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return INT32_MIN
+
+
+cdef int64_t no_op(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return ordinal
+
+
+cdef freq_conv_func get_asfreq_func(int from_freq, int to_freq) noexcept nogil:
+    cdef:
+        int from_group = get_freq_group(from_freq)
+        int to_group = get_freq_group(to_freq)
+
+    if from_group == FR_UND:
+        from_group = FR_DAY
+
+    if from_group == FR_BUS:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_BtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_BtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>asfreq_BtoM
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_BtoW
+        elif to_group == FR_BUS:
+            return <freq_conv_func>no_op
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_BtoDT
+        else:
+            return <freq_conv_func>nofunc
+
+    elif to_group == FR_BUS:
+        if from_group == FR_ANN:
+            return <freq_conv_func>asfreq_AtoB
+        elif from_group == FR_QTR:
+            return <freq_conv_func>asfreq_QtoB
+        elif from_group == FR_MTH:
+            return <freq_conv_func>asfreq_MtoB
+        elif from_group == FR_WK:
+            return <freq_conv_func>asfreq_WtoB
+        elif from_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_DTtoB
+        else:
+            return <freq_conv_func>nofunc
+
+    elif from_group == FR_ANN:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_AtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_AtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>asfreq_AtoM
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_AtoW
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_AtoDT
+        else:
+            return <freq_conv_func>nofunc
+
+    elif from_group == FR_QTR:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_QtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_QtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>asfreq_QtoM
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_QtoW
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_QtoDT
+        else:
+            return <freq_conv_func>nofunc
+
+    elif from_group == FR_MTH:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_MtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_MtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>no_op
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_MtoW
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_MtoDT
+        else:
+            return <freq_conv_func>nofunc
+
+    elif from_group == FR_WK:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_WtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_WtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>asfreq_WtoM
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_WtoW
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            return <freq_conv_func>asfreq_WtoDT
+        else:
+            return <freq_conv_func>nofunc
+
+    elif from_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+        if to_group == FR_ANN:
+            return <freq_conv_func>asfreq_DTtoA
+        elif to_group == FR_QTR:
+            return <freq_conv_func>asfreq_DTtoQ
+        elif to_group == FR_MTH:
+            return <freq_conv_func>asfreq_DTtoM
+        elif to_group == FR_WK:
+            return <freq_conv_func>asfreq_DTtoW
+        elif to_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
+            if from_group > to_group:
+                return <freq_conv_func>downsample_daytime
+            else:
+                return <freq_conv_func>upsample_daytime
+
+        else:
+            return <freq_conv_func>nofunc
+
+    else:
+        return <freq_conv_func>nofunc
+
+
+# --------------------------------------------------------------------
+# Frequency Conversion Helpers
+
+cdef int64_t DtoB_weekday(int64_t unix_date) noexcept nogil:
+    return ((unix_date + 4) // 7) * 5 + ((unix_date + 4) % 7) - 4
+
+
+cdef int64_t DtoB(npy_datetimestruct *dts, int roll_back,
+                  int64_t unix_date) noexcept nogil:
+    # calculate the current week (counting from 1970-01-01) treating
+    # sunday as last day of a week
+    cdef:
+        int day_of_week = dayofweek(dts.year, dts.month, dts.day)
+
+    if roll_back == 1:
+        if day_of_week > 4:
+            # change to friday before weekend
+            unix_date -= (day_of_week - 4)
+    else:
+        if day_of_week > 4:
+            # change to Monday after weekend
+            unix_date += (7 - day_of_week)
+
+    return DtoB_weekday(unix_date)
+
+
+cdef int64_t upsample_daytime(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    if af_info.is_end:
+        return (ordinal + 1) * af_info.intraday_conversion_factor - 1
+    else:
+        return ordinal * af_info.intraday_conversion_factor
+
+
+cdef int64_t downsample_daytime(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return ordinal // af_info.intraday_conversion_factor
+
+
+cdef int64_t transform_via_day(int64_t ordinal,
+                               asfreq_info *af_info,
+                               freq_conv_func first_func,
+                               freq_conv_func second_func) noexcept nogil:
+    cdef:
+        int64_t result
+
+    result = first_func(ordinal, af_info)
+    result = second_func(result, af_info)
+    return result
+
+
+# --------------------------------------------------------------------
+# Conversion _to_ Daily Freq
+
+cdef int64_t asfreq_AtoDT(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int64_t unix_date
+        npy_datetimestruct dts
+
+    ordinal += af_info.is_end
+
+    dts.year = ordinal + 1970
+    dts.month = 1
+    adjust_dts_for_month(&dts, af_info.from_end)
+
+    unix_date = unix_date_from_ymd(dts.year, dts.month, 1)
+    unix_date -= af_info.is_end
+    return upsample_daytime(unix_date, af_info)
+
+
+cdef int64_t asfreq_QtoDT(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int64_t unix_date
+        npy_datetimestruct dts
+
+    ordinal += af_info.is_end
+
+    dts.year = ordinal // 4 + 1970
+    dts.month = (ordinal % 4) * 3 + 1
+    adjust_dts_for_month(&dts, af_info.from_end)
+
+    unix_date = unix_date_from_ymd(dts.year, dts.month, 1)
+    unix_date -= af_info.is_end
+    return upsample_daytime(unix_date, af_info)
+
+
+cdef int64_t asfreq_MtoDT(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int64_t unix_date
+        int year, month
+
+    ordinal += af_info.is_end
+
+    year = ordinal // 12 + 1970
+    month = ordinal % 12 + 1
+
+    unix_date = unix_date_from_ymd(year, month, 1)
+    unix_date -= af_info.is_end
+    return upsample_daytime(unix_date, af_info)
+
+
+cdef int64_t asfreq_WtoDT(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    ordinal = (ordinal * 7 + af_info.from_end - 4 +
+               (7 - 1) * (af_info.is_end - 1))
+    return upsample_daytime(ordinal, af_info)
+
+
+# --------------------------------------------------------------------
+# Conversion _to_ BusinessDay Freq
+
+cdef int64_t asfreq_AtoB(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int roll_back
+        npy_datetimestruct dts
+        int64_t unix_date = asfreq_AtoDT(ordinal, af_info)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, &dts)
+    roll_back = af_info.is_end
+    return DtoB(&dts, roll_back, unix_date)
+
+
+cdef int64_t asfreq_QtoB(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int roll_back
+        npy_datetimestruct dts
+        int64_t unix_date = asfreq_QtoDT(ordinal, af_info)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, &dts)
+    roll_back = af_info.is_end
+    return DtoB(&dts, roll_back, unix_date)
+
+
+cdef int64_t asfreq_MtoB(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int roll_back
+        npy_datetimestruct dts
+        int64_t unix_date = asfreq_MtoDT(ordinal, af_info)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, &dts)
+    roll_back = af_info.is_end
+    return DtoB(&dts, roll_back, unix_date)
+
+
+cdef int64_t asfreq_WtoB(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int roll_back
+        npy_datetimestruct dts
+        int64_t unix_date = asfreq_WtoDT(ordinal, af_info)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, &dts)
+    roll_back = af_info.is_end
+    return DtoB(&dts, roll_back, unix_date)
+
+
+cdef int64_t asfreq_DTtoB(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int roll_back
+        npy_datetimestruct dts
+        int64_t unix_date = downsample_daytime(ordinal, af_info)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, &dts)
+    # This usage defines roll_back the opposite way from the others
+    roll_back = 1 - af_info.is_end
+    return DtoB(&dts, roll_back, unix_date)
+
+
+# ----------------------------------------------------------------------
+# Conversion _from_ Daily Freq
+
+cdef int64_t asfreq_DTtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        npy_datetimestruct dts
+
+    ordinal = downsample_daytime(ordinal, af_info)
+    pandas_datetime_to_datetimestruct(ordinal, NPY_FR_D, &dts)
+    return dts_to_year_ordinal(&dts, af_info.to_end)
+
+
+cdef int DtoQ_yq(int64_t ordinal, asfreq_info *af_info,
+                 npy_datetimestruct* dts) noexcept nogil:
+    cdef:
+        int quarter
+
+    pandas_datetime_to_datetimestruct(ordinal, NPY_FR_D, dts)
+    adjust_dts_for_qtr(dts, af_info.to_end)
+
+    quarter = month_to_quarter(dts.month)
+    return quarter
+
+
+cdef int64_t asfreq_DTtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        int quarter
+        npy_datetimestruct dts
+
+    ordinal = downsample_daytime(ordinal, af_info)
+
+    quarter = DtoQ_yq(ordinal, af_info, &dts)
+    return <int64_t>((dts.year - 1970) * 4 + quarter - 1)
+
+
+cdef int64_t asfreq_DTtoM(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    cdef:
+        npy_datetimestruct dts
+
+    ordinal = downsample_daytime(ordinal, af_info)
+    pandas_datetime_to_datetimestruct(ordinal, NPY_FR_D, &dts)
+    return dts_to_month_ordinal(&dts)
+
+
+cdef int64_t asfreq_DTtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    ordinal = downsample_daytime(ordinal, af_info)
+    return unix_date_to_week(ordinal, af_info.to_end)
+
+
+cdef int64_t unix_date_to_week(int64_t unix_date, int to_end) noexcept nogil:
+    return (unix_date + 3 - to_end) // 7 + 1
+
+
+# --------------------------------------------------------------------
+# Conversion _from_ BusinessDay Freq
+
+cdef int64_t asfreq_BtoDT(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    ordinal = ((ordinal + 3) // 5) * 7 + (ordinal + 3) % 5 - 3
+    return upsample_daytime(ordinal, af_info)
+
+
+cdef int64_t asfreq_BtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_BtoDT,
+                             <freq_conv_func>asfreq_DTtoA)
+
+
+cdef int64_t asfreq_BtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_BtoDT,
+                             <freq_conv_func>asfreq_DTtoQ)
+
+
+cdef int64_t asfreq_BtoM(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_BtoDT,
+                             <freq_conv_func>asfreq_DTtoM)
+
+
+cdef int64_t asfreq_BtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_BtoDT,
+                             <freq_conv_func>asfreq_DTtoW)
+
+
+# ----------------------------------------------------------------------
+# Conversion _from_ Annual Freq
+
+cdef int64_t asfreq_AtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_AtoDT,
+                             <freq_conv_func>asfreq_DTtoA)
+
+
+cdef int64_t asfreq_AtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_AtoDT,
+                             <freq_conv_func>asfreq_DTtoQ)
+
+
+cdef int64_t asfreq_AtoM(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_AtoDT,
+                             <freq_conv_func>asfreq_DTtoM)
+
+
+cdef int64_t asfreq_AtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_AtoDT,
+                             <freq_conv_func>asfreq_DTtoW)
+
+
+# ----------------------------------------------------------------------
+# Conversion _from_ Quarterly Freq
+
+cdef int64_t asfreq_QtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_QtoDT,
+                             <freq_conv_func>asfreq_DTtoQ)
+
+
+cdef int64_t asfreq_QtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_QtoDT,
+                             <freq_conv_func>asfreq_DTtoA)
+
+
+cdef int64_t asfreq_QtoM(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_QtoDT,
+                             <freq_conv_func>asfreq_DTtoM)
+
+
+cdef int64_t asfreq_QtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_QtoDT,
+                             <freq_conv_func>asfreq_DTtoW)
+
+
+# ----------------------------------------------------------------------
+# Conversion _from_ Monthly Freq
+
+cdef int64_t asfreq_MtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_MtoDT,
+                             <freq_conv_func>asfreq_DTtoA)
+
+
+cdef int64_t asfreq_MtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_MtoDT,
+                             <freq_conv_func>asfreq_DTtoQ)
+
+
+cdef int64_t asfreq_MtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_MtoDT,
+                             <freq_conv_func>asfreq_DTtoW)
+
+
+# ----------------------------------------------------------------------
+# Conversion _from_ Weekly Freq
+
+cdef int64_t asfreq_WtoA(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_WtoDT,
+                             <freq_conv_func>asfreq_DTtoA)
+
+
+cdef int64_t asfreq_WtoQ(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_WtoDT,
+                             <freq_conv_func>asfreq_DTtoQ)
+
+
+cdef int64_t asfreq_WtoM(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_WtoDT,
+                             <freq_conv_func>asfreq_DTtoM)
+
+
+cdef int64_t asfreq_WtoW(int64_t ordinal, asfreq_info *af_info) noexcept nogil:
+    return transform_via_day(ordinal, af_info,
+                             <freq_conv_func>asfreq_WtoDT,
+                             <freq_conv_func>asfreq_DTtoW)
+
+
+# ----------------------------------------------------------------------
+
+@cython.cdivision
+cdef char* c_strftime(npy_datetimestruct *dts, char *fmt):
+    """
+    Generate a nice string representation of the period
+    object, originally from DateObject_strftime
+
+    Parameters
+    ----------
+    dts : npy_datetimestruct*
+    fmt : char*
+
+    Returns
+    -------
+    result : char*
+    """
+    cdef:
+        tm c_date
+        char *result
+        int result_len = strlen(fmt) + 50
+
+    c_date.tm_sec = dts.sec
+    c_date.tm_min = dts.min
+    c_date.tm_hour = dts.hour
+    c_date.tm_mday = dts.day
+    c_date.tm_mon = dts.month - 1
+    c_date.tm_year = dts.year - 1900
+    c_date.tm_wday = (dayofweek(dts.year, dts.month, dts.day) + 1) % 7
+    c_date.tm_yday = get_day_of_year(dts.year, dts.month, dts.day) - 1
+    c_date.tm_isdst = -1
+
+    result = <char*>malloc(result_len)
+    if result is NULL:
+        raise MemoryError()
+
+    strftime(result, result_len, fmt, &c_date)
+
+    return result
+
+
+# ----------------------------------------------------------------------
+# Conversion between date_info and npy_datetimestruct
+
+cdef int get_freq_group(int freq) noexcept nogil:
+    # See also FreqGroup.get_freq_group
+    return (freq // 1000) * 1000
+
+
+cdef int get_freq_group_index(int freq) noexcept nogil:
+    return freq // 1000
+
+
+cdef void adjust_dts_for_month(npy_datetimestruct* dts, int from_end) noexcept nogil:
+    if from_end != 12:
+        dts.month += from_end
+        if dts.month > 12:
+            dts.month -= 12
+        else:
+            dts.year -= 1
+
+
+cdef void adjust_dts_for_qtr(npy_datetimestruct* dts, int to_end) noexcept nogil:
+    if to_end != 12:
+        dts.month -= to_end
+        if dts.month <= 0:
+            dts.month += 12
+        else:
+            dts.year += 1
+
+
+# Find the unix_date (days elapsed since datetime(1970, 1, 1)
+# for the given year/month/day.
+# Assumes GREGORIAN_CALENDAR */
+cdef int64_t unix_date_from_ymd(int year, int month, int day) noexcept nogil:
+    # Calculate the absolute date
+    cdef:
+        npy_datetimestruct dts
+        int64_t unix_date
+
+    memset(&dts, 0, sizeof(npy_datetimestruct))
+    dts.year = year
+    dts.month = month
+    dts.day = day
+    unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, &dts)
+    return unix_date
+
+
+cdef int64_t dts_to_month_ordinal(npy_datetimestruct* dts) noexcept nogil:
+    # AKA: use npy_datetimestruct_to_datetime(NPY_FR_M, &dts)
+    return <int64_t>((dts.year - 1970) * 12 + dts.month - 1)
+
+
+cdef int64_t dts_to_year_ordinal(npy_datetimestruct *dts, int to_end) noexcept nogil:
+    cdef:
+        int64_t result
+
+    result = npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT.NPY_FR_Y, dts)
+    if dts.month > to_end:
+        return result + 1
+    else:
+        return result
+
+
+cdef int64_t dts_to_qtr_ordinal(npy_datetimestruct* dts, int to_end) noexcept nogil:
+    cdef:
+        int quarter
+
+    adjust_dts_for_qtr(dts, to_end)
+    quarter = month_to_quarter(dts.month)
+    return <int64_t>((dts.year - 1970) * 4 + quarter - 1)
+
+
+cdef int get_anchor_month(int freq, int freq_group) noexcept nogil:
+    cdef:
+        int fmonth
+    fmonth = freq - freq_group
+    if fmonth == 0:
+        fmonth = 12
+    return fmonth
+
+
+# specifically _dont_ use cdvision or else ordinals near -1 are assigned to
+# incorrect dates GH#19643
+@cython.cdivision(False)
+cdef int64_t get_period_ordinal(npy_datetimestruct *dts, int freq) noexcept nogil:
+    """
+    Generate an ordinal in period space
+
+    Parameters
+    ----------
+    dts : npy_datetimestruct*
+    freq : int
+
+    Returns
+    -------
+    period_ordinal : int64_t
+    """
+    cdef:
+        int64_t unix_date
+        int freq_group, fmonth
+        NPY_DATETIMEUNIT unit
+
+    freq_group = get_freq_group(freq)
+
+    if freq_group == FR_ANN:
+        fmonth = get_anchor_month(freq, freq_group)
+        return dts_to_year_ordinal(dts, fmonth)
+
+    elif freq_group == FR_QTR:
+        fmonth = get_anchor_month(freq, freq_group)
+        return dts_to_qtr_ordinal(dts, fmonth)
+
+    elif freq_group == FR_WK:
+        unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
+        return unix_date_to_week(unix_date, freq - FR_WK)
+
+    elif freq == FR_BUS:
+        unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
+        return DtoB(dts, 0, unix_date)
+
+    unit = freq_group_code_to_npy_unit(freq)
+    return npy_datetimestruct_to_datetime(unit, dts)
+
+
+cdef void get_date_info(int64_t ordinal,
+                        int freq, npy_datetimestruct *dts) noexcept nogil:
+    cdef:
+        int64_t unix_date, nanos
+        npy_datetimestruct dts2
+
+    unix_date = get_unix_date(ordinal, freq)
+    nanos = get_time_nanos(freq, unix_date, ordinal)
+
+    pandas_datetime_to_datetimestruct(unix_date, NPY_FR_D, dts)
+
+    pandas_datetime_to_datetimestruct(nanos, NPY_DATETIMEUNIT.NPY_FR_ns, &dts2)
+    dts.hour = dts2.hour
+    dts.min = dts2.min
+    dts.sec = dts2.sec
+    dts.us = dts2.us
+    dts.ps = dts2.ps
+
+
+cdef int64_t get_unix_date(int64_t period_ordinal, int freq) noexcept nogil:
+    """
+    Returns the proleptic Gregorian ordinal of the date, as an integer.
+    This corresponds to the number of days since Jan., 1st, 1970 AD.
+    When the instance has a frequency less than daily, the proleptic date
+    is calculated for the last day of the period.
+
+    Parameters
+    ----------
+    period_ordinal : int64_t
+    freq : int
+
+    Returns
+    -------
+    unix_date : int64_t number of days since datetime(1970, 1, 1)
+    """
+    cdef:
+        asfreq_info af_info
+        freq_conv_func toDaily = NULL
+
+    if freq == FR_DAY:
+        return period_ordinal
+
+    toDaily = get_asfreq_func(freq, FR_DAY)
+    get_asfreq_info(freq, FR_DAY, True, &af_info)
+    return toDaily(period_ordinal, &af_info)
+
+
+@cython.cdivision
+cdef int64_t get_time_nanos(int freq, int64_t unix_date,
+                            int64_t ordinal) noexcept nogil:
+    """
+    Find the number of nanoseconds after midnight on the given unix_date
+    that the ordinal represents in the given frequency.
+
+    Parameters
+    ----------
+    freq : int
+    unix_date : int64_t
+    ordinal : int64_t
+
+    Returns
+    -------
+    int64_t
+    """
+    cdef:
+        int64_t sub, factor
+        int64_t nanos_in_day = 24 * 3600 * 10**9
+
+    freq = get_freq_group(freq)
+
+    if freq <= FR_DAY:
+        return 0
+
+    elif freq == FR_NS:
+        factor = 1
+
+    elif freq == FR_US:
+        factor = 10**3
+
+    elif freq == FR_MS:
+        factor = 10**6
+
+    elif freq == FR_SEC:
+        factor = 10 **9
+
+    elif freq == FR_MIN:
+        factor = 10**9 * 60
+
+    else:
+        # We must have freq == FR_HR
+        factor = 10**9 * 3600
+
+    sub = ordinal - unix_date * (nanos_in_day / factor)
+    return sub * factor
+
+
+cdef int get_yq(int64_t ordinal, int freq, npy_datetimestruct* dts):
+    """
+    Find the year and quarter of a Period with the given ordinal and frequency
+
+    Parameters
+    ----------
+    ordinal : int64_t
+    freq : int
+    dts : *npy_datetimestruct
+
+    Returns
+    -------
+    quarter : int
+        describes the implied quarterly frequency associated with `freq`
+
+    Notes
+    -----
+    Sets dts.year in-place.
+    """
+    cdef:
+        asfreq_info af_info
+        int qtr_freq
+        int64_t unix_date
+        int quarter
+
+    unix_date = get_unix_date(ordinal, freq)
+
+    if get_freq_group(freq) == FR_QTR:
+        qtr_freq = freq
+    else:
+        qtr_freq = FR_QTR
+
+    get_asfreq_info(FR_DAY, qtr_freq, True, &af_info)
+
+    quarter = DtoQ_yq(unix_date, &af_info, dts)
+    return quarter
+
+
+cdef int month_to_quarter(int month) noexcept nogil:
+    return (month - 1) // 3 + 1
+
+
+# ----------------------------------------------------------------------
+# Period logic
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def periodarr_to_dt64arr(const int64_t[:] periodarr, int freq):
+    """
+    Convert array to datetime64 values from a set of ordinals corresponding to
+    periods per period convention.
+    """
+    cdef:
+        int64_t[::1] out
+        Py_ssize_t i, N
+
+    if freq < 6000:  # i.e. FR_DAY, hard-code to avoid need to cast
+        N = len(periodarr)
+        out = np.empty(N, dtype="i8")
+
+        # We get here with freqs that do not correspond to a datetime64 unit
+        for i in range(N):
+            out[i] = period_ordinal_to_dt64(periodarr[i], freq)
+
+        return out.base  # .base to access underlying np.ndarray
+
+    else:
+        # Short-circuit for performance
+        if freq == FR_NS:
+            # TODO: copy?
+            return periodarr.base
+
+        if freq == FR_US:
+            dta = periodarr.base.view("M8[us]")
+        elif freq == FR_MS:
+            dta = periodarr.base.view("M8[ms]")
+        elif freq == FR_SEC:
+            dta = periodarr.base.view("M8[s]")
+        elif freq == FR_MIN:
+            dta = periodarr.base.view("M8[m]")
+        elif freq == FR_HR:
+            dta = periodarr.base.view("M8[h]")
+        elif freq == FR_DAY:
+            dta = periodarr.base.view("M8[D]")
+        return astype_overflowsafe(dta, dtype=DT64NS_DTYPE)
+
+
+cdef void get_asfreq_info(int from_freq, int to_freq,
+                          bint is_end, asfreq_info *af_info) noexcept nogil:
+    """
+    Construct the `asfreq_info` object used to convert an ordinal from
+    `from_freq` to `to_freq`.
+
+    Parameters
+    ----------
+    from_freq : int
+    to_freq int
+    is_end : bool
+    af_info : *asfreq_info
+    """
+    cdef:
+        int from_group = get_freq_group(from_freq)
+        int to_group = get_freq_group(to_freq)
+
+    af_info.is_end = is_end
+
+    af_info.intraday_conversion_factor = get_daytime_conversion_factor(
+        get_freq_group_index(max_value(from_group, FR_DAY)),
+        get_freq_group_index(max_value(to_group, FR_DAY)))
+
+    if from_group == FR_WK:
+        af_info.from_end = calc_week_end(from_freq, from_group)
+    elif from_group == FR_ANN:
+        af_info.from_end = calc_a_year_end(from_freq, from_group)
+    elif from_group == FR_QTR:
+        af_info.from_end = calc_a_year_end(from_freq, from_group)
+
+    if to_group == FR_WK:
+        af_info.to_end = calc_week_end(to_freq, to_group)
+    elif to_group == FR_ANN:
+        af_info.to_end = calc_a_year_end(to_freq, to_group)
+    elif to_group == FR_QTR:
+        af_info.to_end = calc_a_year_end(to_freq, to_group)
+
+
+@cython.cdivision
+cdef int calc_a_year_end(int freq, int group) noexcept nogil:
+    cdef:
+        int result = (freq - group) % 12
+    if result == 0:
+        return 12
+    else:
+        return result
+
+
+cdef int calc_week_end(int freq, int group) noexcept nogil:
+    return freq - group
+
+
+cpdef int64_t period_asfreq(int64_t ordinal, int freq1, int freq2, bint end):
+    """
+    Convert period ordinal from one frequency to another, and if upsampling,
+    choose to use start ('S') or end ('E') of period.
+    """
+    cdef:
+        int64_t retval
+
+    _period_asfreq(&ordinal, &retval, 1, freq1, freq2, end)
+    return retval
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def period_asfreq_arr(ndarray[int64_t] arr, int freq1, int freq2, bint end):
+    """
+    Convert int64-array of period ordinals from one frequency to another, and
+    if upsampling, choose to use start ('S') or end ('E') of period.
+    """
+    cdef:
+        Py_ssize_t n = len(arr)
+        Py_ssize_t increment = arr.strides[0] // 8
+        ndarray[int64_t] result = cnp.PyArray_EMPTY(
+            arr.ndim, arr.shape, cnp.NPY_INT64, 0
+        )
+
+    _period_asfreq(
+        <int64_t*>cnp.PyArray_DATA(arr),
+        <int64_t*>cnp.PyArray_DATA(result),
+        n,
+        freq1,
+        freq2,
+        end,
+        increment,
+    )
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cdef void _period_asfreq(
+    int64_t* ordinals,
+    int64_t* out,
+    Py_ssize_t length,
+    int freq1,
+    int freq2,
+    bint end,
+    Py_ssize_t increment=1,
+) noexcept:
+    """See period_asfreq.__doc__"""
+    cdef:
+        Py_ssize_t i
+        freq_conv_func func
+        asfreq_info af_info
+        int64_t val
+
+    if length == 1 and ordinals[0] == NPY_NAT:
+        # fastpath avoid calling get_asfreq_func
+        out[0] = NPY_NAT
+        return
+
+    func = get_asfreq_func(freq1, freq2)
+    get_asfreq_info(freq1, freq2, end, &af_info)
+
+    for i in range(length):
+        val = ordinals[i * increment]
+        if val != NPY_NAT:
+            val = func(val, &af_info)
+        out[i] = val
+
+
+cpdef int64_t period_ordinal(int y, int m, int d, int h, int min,
+                             int s, int us, int ps, int freq):
+    """
+    Find the ordinal representation of the given datetime components at the
+    frequency `freq`.
+
+    Parameters
+    ----------
+    y : int
+    m : int
+    d : int
+    h : int
+    min : int
+    s : int
+    us : int
+    ps : int
+
+    Returns
+    -------
+    ordinal : int64_t
+    """
+    cdef:
+        npy_datetimestruct dts
+    dts.year = y
+    dts.month = m
+    dts.day = d
+    dts.hour = h
+    dts.min = min
+    dts.sec = s
+    dts.us = us
+    dts.ps = ps
+    return get_period_ordinal(&dts, freq)
+
+
+cdef int64_t period_ordinal_to_dt64(int64_t ordinal, int freq) except? -1:
+    cdef:
+        npy_datetimestruct dts
+        int64_t result
+
+    if ordinal == NPY_NAT:
+        return NPY_NAT
+
+    get_date_info(ordinal, freq, &dts)
+
+    try:
+        result = npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT.NPY_FR_ns, &dts)
+    except OverflowError as err:
+        fmt = dts_to_iso_string(&dts)
+        raise OutOfBoundsDatetime(f"Out of bounds nanosecond timestamp: {fmt}") from err
+
+    return result
+
+
+cdef str period_format(int64_t value, int freq, object fmt=None):
+
+    cdef:
+        int freq_group, quarter
+        npy_datetimestruct dts
+        bint is_fmt_none
+
+    if value == NPY_NAT:
+        return "NaT"
+
+    # fill dts and freq group
+    get_date_info(value, freq, &dts)
+    freq_group = get_freq_group(freq)
+
+    # use the appropriate default format depending on frequency group
+    is_fmt_none = fmt is None
+    if freq_group == FR_ANN and (is_fmt_none or fmt == "%Y"):
+        return f"{dts.year}"
+
+    elif freq_group == FR_QTR and (is_fmt_none or fmt == "%FQ%q"):
+        # get quarter and modify dts.year to be the 'Fiscal' year
+        quarter = get_yq(value, freq, &dts)
+        return f"{dts.year}Q{quarter}"
+
+    elif freq_group == FR_MTH and (is_fmt_none or fmt == "%Y-%m"):
+        return f"{dts.year}-{dts.month:02d}"
+
+    elif freq_group == FR_WK and is_fmt_none:
+        # special: start_date/end_date. Recurse
+        left = period_asfreq(value, freq, FR_DAY, 0)
+        right = period_asfreq(value, freq, FR_DAY, 1)
+        return f"{period_format(left, FR_DAY)}/{period_format(right, FR_DAY)}"
+
+    elif (
+        (freq_group == FR_BUS or freq_group == FR_DAY)
+        and (is_fmt_none or fmt == "%Y-%m-%d")
+    ):
+        return f"{dts.year}-{dts.month:02d}-{dts.day:02d}"
+
+    elif freq_group == FR_HR and (is_fmt_none or fmt == "%Y-%m-%d %H:00"):
+        return f"{dts.year}-{dts.month:02d}-{dts.day:02d} {dts.hour:02d}:00"
+
+    elif freq_group == FR_MIN and (is_fmt_none or fmt == "%Y-%m-%d %H:%M"):
+        return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                f"{dts.hour:02d}:{dts.min:02d}")
+
+    elif freq_group == FR_SEC and (is_fmt_none or fmt == "%Y-%m-%d %H:%M:%S"):
+        return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}")
+
+    elif freq_group == FR_MS and (is_fmt_none or fmt == "%Y-%m-%d %H:%M:%S.%l"):
+        return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}"
+                f".{(dts.us // 1_000):03d}")
+
+    elif freq_group == FR_US and (is_fmt_none or fmt == "%Y-%m-%d %H:%M:%S.%u"):
+        return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}"
+                f".{(dts.us):06d}")
+
+    elif freq_group == FR_NS and (is_fmt_none or fmt == "%Y-%m-%d %H:%M:%S.%n"):
+        return (f"{dts.year}-{dts.month:02d}-{dts.day:02d} "
+                f"{dts.hour:02d}:{dts.min:02d}:{dts.sec:02d}"
+                f".{((dts.us * 1000) + (dts.ps // 1000)):09d}")
+
+    elif is_fmt_none:
+        # `freq_group` is invalid, raise
+        raise ValueError(f"Unknown freq: {freq}")
+
+    else:
+        # A custom format is requested
+        if isinstance(fmt, str):
+            # Encode using current locale, in case fmt contains non-utf8 chars
+            fmt = <bytes>util.string_encode_locale(fmt)
+
+        return _period_strftime(value, freq, fmt, dts)
+
+
+cdef list extra_fmts = [(b"%q", b"^`AB`^"),
+                        (b"%f", b"^`CD`^"),
+                        (b"%F", b"^`EF`^"),
+                        (b"%l", b"^`GH`^"),
+                        (b"%u", b"^`IJ`^"),
+                        (b"%n", b"^`KL`^")]
+
+cdef list str_extra_fmts = ["^`AB`^", "^`CD`^", "^`EF`^",
+                            "^`GH`^", "^`IJ`^", "^`KL`^"]
+
+cdef str _period_strftime(int64_t value, int freq, bytes fmt, npy_datetimestruct dts):
+    cdef:
+        Py_ssize_t i
+        char *formatted
+        bytes pat, brepl
+        list found_pat = [False] * len(extra_fmts)
+        int quarter
+        int32_t us, ps
+        str result, repl
+
+    # Find our additional directives in the pattern and replace them with
+    # placeholders that are not processed by c_strftime
+    for i in range(len(extra_fmts)):
+        pat = extra_fmts[i][0]
+        brepl = extra_fmts[i][1]
+        if pat in fmt:
+            fmt = fmt.replace(pat, brepl)
+            found_pat[i] = True
+
+    # Execute c_strftime to process the usual datetime directives
+    formatted = c_strftime(&dts, <char*>fmt)
+
+    # Decode result according to current locale
+    result = util.char_to_string_locale(formatted)
+    free(formatted)
+
+    # Now we will fill the placeholders corresponding to our additional directives
+
+    # First prepare the contents
+    # Save these to local vars as dts can be modified by get_yq below
+    us = dts.us
+    ps = dts.ps
+    if any(found_pat[0:3]):
+        # Note: this modifies `dts` in-place so that year becomes fiscal year
+        # However it looses the us and ps
+        quarter = get_yq(value, freq, &dts)
+    else:
+        quarter = 0
+
+    # Now do the filling per se
+    for i in range(len(extra_fmts)):
+        if found_pat[i]:
+
+            if i == 0:  # %q, 1-digit quarter.
+                repl = f"{quarter}"
+            elif i == 1:  # %f, 2-digit 'Fiscal' year
+                repl = f"{(dts.year % 100):02d}"
+            elif i == 2:  # %F, 'Fiscal' year with a century
+                repl = str(dts.year)
+            elif i == 3:  # %l, milliseconds
+                repl = f"{(us // 1_000):03d}"
+            elif i == 4:  # %u, microseconds
+                repl = f"{(us):06d}"
+            elif i == 5:  # %n, nanoseconds
+                repl = f"{((us * 1000) + (ps // 1000)):09d}"
+
+            result = result.replace(str_extra_fmts[i], repl)
+
+    return result
+
+
+def period_array_strftime(
+    ndarray values, int dtype_code, object na_rep, str date_format
+):
+    """
+    Vectorized Period.strftime used for PeriodArray._format_native_types.
+
+    Parameters
+    ----------
+    values : ndarray[int64_t], ndim unrestricted
+    dtype_code : int
+        Corresponds to PeriodDtype._dtype_code
+    na_rep : any
+    date_format : str or None
+    """
+    cdef:
+        Py_ssize_t i, n = values.size
+        int64_t ordinal
+        object item_repr
+        ndarray out = cnp.PyArray_EMPTY(
+            values.ndim, values.shape, cnp.NPY_OBJECT, 0
+        )
+        object[::1] out_flat = out.ravel()
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(out, values)
+
+    for i in range(n):
+        # Analogous to: ordinal = values[i]
+        ordinal = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if ordinal == NPY_NAT:
+            item_repr = na_rep
+        else:
+            # This is equivalent to
+            # freq = frequency_corresponding_to_dtype_code(dtype_code)
+            # per = Period(ordinal, freq=freq)
+            # if date_format:
+            #     item_repr = per.strftime(date_format)
+            # else:
+            #     item_repr = str(per)
+            item_repr = period_format(ordinal, dtype_code, date_format)
+
+        # Analogous to: ordinals[i] = ordinal
+        out_flat[i] = item_repr
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return out
+
+
+# ----------------------------------------------------------------------
+# period accessors
+
+ctypedef int (*accessor)(int64_t ordinal, int freq) except INT32_MIN
+
+
+cdef int pyear(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dts.year
+
+
+cdef int pqyear(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+
+    get_yq(ordinal, freq, &dts)
+    return dts.year
+
+
+cdef int pquarter(int64_t ordinal, int freq):
+    cdef:
+        int quarter
+        npy_datetimestruct dts
+    quarter = get_yq(ordinal, freq, &dts)
+    return quarter
+
+
+cdef int pmonth(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dts.month
+
+
+cdef int pday(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dts.day
+
+
+cdef int pweekday(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dayofweek(dts.year, dts.month, dts.day)
+
+
+cdef int pday_of_year(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return get_day_of_year(dts.year, dts.month, dts.day)
+
+
+cdef int pweek(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return get_week_of_year(dts.year, dts.month, dts.day)
+
+
+cdef int phour(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dts.hour
+
+
+cdef int pminute(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return dts.min
+
+
+cdef int psecond(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return <int>dts.sec
+
+
+cdef int pdays_in_month(int64_t ordinal, int freq):
+    cdef:
+        npy_datetimestruct dts
+    get_date_info(ordinal, freq, &dts)
+    return get_days_in_month(dts.year, dts.month)
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_period_field_arr(str field, const int64_t[:] arr, int freq):
+    cdef:
+        Py_ssize_t i, sz
+        int64_t[::1] out
+
+    func = _get_accessor_func(field)
+    if func is NULL:
+        raise ValueError(f"Unrecognized field name: {field}")
+
+    sz = len(arr)
+    out = np.empty(sz, dtype=np.int64)
+
+    for i in range(sz):
+        if arr[i] == NPY_NAT:
+            out[i] = -1
+            continue
+        out[i] = func(arr[i], freq)
+
+    return out.base  # .base to access underlying np.ndarray
+
+
+cdef accessor _get_accessor_func(str field):
+    if field == "year":
+        return <accessor>pyear
+    elif field == "qyear":
+        return <accessor>pqyear
+    elif field == "quarter":
+        return <accessor>pquarter
+    elif field == "month":
+        return <accessor>pmonth
+    elif field == "day":
+        return <accessor>pday
+    elif field == "hour":
+        return <accessor>phour
+    elif field == "minute":
+        return <accessor>pminute
+    elif field == "second":
+        return <accessor>psecond
+    elif field == "week":
+        return <accessor>pweek
+    elif field == "day_of_year":
+        return <accessor>pday_of_year
+    elif field == "weekday" or field == "day_of_week":
+        return <accessor>pweekday
+    elif field == "days_in_month":
+        return <accessor>pdays_in_month
+    return NULL
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def from_ordinals(const int64_t[:] values, freq):
+    cdef:
+        Py_ssize_t i, n = len(values)
+        int64_t[::1] result = np.empty(len(values), dtype="i8")
+        int64_t val
+
+    freq = to_offset(freq, is_period=True)
+    if not isinstance(freq, BaseOffset):
+        raise ValueError("freq not specified and cannot be inferred")
+
+    for i in range(n):
+        val = values[i]
+        if val == NPY_NAT:
+            result[i] = NPY_NAT
+        else:
+            result[i] = Period(val, freq=freq).ordinal
+
+    return result.base
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def extract_ordinals(ndarray values, freq) -> np.ndarray:
+    # values is object-dtype, may be 2D
+
+    cdef:
+        Py_ssize_t i, n = values.size
+        int64_t ordinal
+        ndarray ordinals = cnp.PyArray_EMPTY(
+            values.ndim, values.shape, cnp.NPY_INT64, 0
+        )
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(ordinals, values)
+        object p
+
+    if values.descr.type_num != cnp.NPY_OBJECT:
+        # if we don't raise here, we'll segfault later!
+        raise TypeError("extract_ordinals values must be object-dtype")
+
+    freqstr = PeriodDtypeBase(freq._period_dtype_code, freq.n)._freqstr
+
+    for i in range(n):
+        # Analogous to: p = values[i]
+        p = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        ordinal = _extract_ordinal(p, freqstr, freq)
+
+        # Analogous to: ordinals[i] = ordinal
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = ordinal
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return ordinals
+
+
+cdef int64_t _extract_ordinal(object item, str freqstr, freq) except? -1:
+    """
+    See extract_ordinals.
+    """
+    cdef:
+        int64_t ordinal
+
+    if checknull_with_nat(item) or item is C_NA:
+        ordinal = NPY_NAT
+    elif util.is_integer_object(item):
+        if item == NPY_NAT:
+            ordinal = NPY_NAT
+        else:
+            raise TypeError(item)
+    else:
+        try:
+            ordinal = item.ordinal
+
+            if item.freqstr != freqstr:
+                msg = DIFFERENT_FREQ.format(cls="PeriodIndex",
+                                            own_freq=freqstr,
+                                            other_freq=item.freqstr)
+                raise IncompatibleFrequency(msg)
+
+        except AttributeError:
+            item = Period(item, freq=freq)
+            if item is NaT:
+                # input may contain NaT-like string
+                ordinal = NPY_NAT
+            else:
+                ordinal = item.ordinal
+
+    return ordinal
+
+
+def extract_freq(ndarray[object] values) -> BaseOffset:
+    # TODO: Change type to const object[:] when Cython supports that.
+
+    cdef:
+        Py_ssize_t i, n = len(values)
+        object value
+
+    for i in range(n):
+        value = values[i]
+
+        if is_period_object(value):
+            return value.freq
+
+    raise ValueError("freq not specified and cannot be inferred")
+
+# -----------------------------------------------------------------------
+# period helpers
+
+
+DIFFERENT_FREQ = ("Input has different freq={other_freq} "
+                  "from {cls}(freq={own_freq})")
+
+
+class IncompatibleFrequency(ValueError):
+    pass
+
+
+cdef class PeriodMixin:
+    # Methods shared between Period and PeriodArray
+
+    @property
+    def start_time(self) -> Timestamp:
+        """
+        Get the Timestamp for the start of the period.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Period.end_time : Return the end Timestamp.
+        Period.dayofyear : Return the day of year.
+        Period.daysinmonth : Return the days in that month.
+        Period.dayofweek : Return the day of the week.
+
+        Examples
+        --------
+        >>> period = pd.Period('2012-1-1', freq='D')
+        >>> period
+        Period('2012-01-01', 'D')
+
+        >>> period.start_time
+        Timestamp('2012-01-01 00:00:00')
+
+        >>> period.end_time
+        Timestamp('2012-01-01 23:59:59.999999999')
+        """
+        return self.to_timestamp(how="start")
+
+    @property
+    def end_time(self) -> Timestamp:
+        """
+        Get the Timestamp for the end of the period.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Period.start_time : Return the start Timestamp.
+        Period.dayofyear : Return the day of year.
+        Period.daysinmonth : Return the days in that month.
+        Period.dayofweek : Return the day of the week.
+
+        Examples
+        --------
+        For Period:
+
+        >>> pd.Period('2020-01', 'D').end_time
+        Timestamp('2020-01-01 23:59:59.999999999')
+
+        For Series:
+
+        >>> period_index = pd.period_range('2020-1-1 00:00', '2020-3-1 00:00', freq='M')
+        >>> s = pd.Series(period_index)
+        >>> s
+        0   2020-01
+        1   2020-02
+        2   2020-03
+        dtype: period[M]
+        >>> s.dt.end_time
+        0   2020-01-31 23:59:59.999999999
+        1   2020-02-29 23:59:59.999999999
+        2   2020-03-31 23:59:59.999999999
+        dtype: datetime64[ns]
+
+        For PeriodIndex:
+
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.end_time
+        DatetimeIndex(['2023-01-31 23:59:59.999999999',
+                       '2023-02-28 23:59:59.999999999',
+                       '2023-03-31 23:59:59.999999999'],
+                       dtype='datetime64[ns]', freq=None)
+        """
+        return self.to_timestamp(how="end")
+
+    def _require_matching_freq(self, other: BaseOffset, bint base=False):
+        # See also arrays.period.raise_on_incompatible
+        if base:
+            condition = self.freq.base != other.base
+        else:
+            condition = self.freq != other
+
+        if condition:
+            freqstr = PeriodDtypeBase(
+                self.freq._period_dtype_code, self.freq.n
+            )._freqstr
+            if hasattr(other, "_period_dtype_code"):
+                other_freqstr = PeriodDtypeBase(
+                    other._period_dtype_code, other.n
+                )._freqstr
+            else:
+                other_freqstr = other.freqstr
+            msg = DIFFERENT_FREQ.format(
+                cls=type(self).__name__,
+                own_freq=freqstr,
+                other_freq=other_freqstr,
+            )
+            raise IncompatibleFrequency(msg)
+
+
+cdef class _Period(PeriodMixin):
+
+    cdef readonly:
+        int64_t ordinal
+        PeriodDtypeBase _dtype
+        BaseOffset freq
+
+    # higher than np.ndarray, np.matrix, np.timedelta64
+    __array_priority__ = 100
+
+    dayofweek = _Period.day_of_week
+    dayofyear = _Period.day_of_year
+
+    def __cinit__(self, int64_t ordinal, BaseOffset freq):
+        self.ordinal = ordinal
+        self.freq = freq
+        # Note: this is more performant than PeriodDtype.from_date_offset(freq)
+        #  because from_date_offset cannot be made a cdef method (until cython
+        #  supported cdef classmethods)
+        self._dtype = PeriodDtypeBase(freq._period_dtype_code, freq.n)
+
+    @classmethod
+    def _maybe_convert_freq(cls, object freq) -> BaseOffset:
+        """
+        A Period's freq attribute must have `freq.n > 0`, which we check for here.
+
+        Returns
+        -------
+        DateOffset
+        """
+        freq = to_offset(freq, is_period=True)
+
+        if freq.n <= 0:
+            raise ValueError("Frequency must be positive, because it "
+                             f"represents span: {freq.freqstr}")
+
+        return freq
+
+    @classmethod
+    def _from_ordinal(cls, ordinal: int64_t, freq: BaseOffset) -> "Period":
+        """
+        Fast creation from an ordinal and freq that are already validated!
+        """
+        if ordinal == NPY_NAT:
+            return NaT
+        else:
+            freq = cls._maybe_convert_freq(freq)
+            self = _Period.__new__(cls, ordinal, freq)
+            return self
+
+    def __richcmp__(self, other, op):
+        if is_period_object(other):
+            if other._dtype != self._dtype:
+                if op == Py_EQ:
+                    return False
+                elif op == Py_NE:
+                    return True
+                self._require_matching_freq(other.freq)
+            return PyObject_RichCompareBool(self.ordinal, other.ordinal, op)
+        elif other is NaT:
+            return op == Py_NE
+        elif util.is_array(other):
+            # GH#44285
+            if cnp.PyArray_IsZeroDim(other):
+                return PyObject_RichCompare(self, other.item(), op)
+            else:
+                # in particular ndarray[object]; see test_pi_cmp_period
+                return np.array([PyObject_RichCompare(self, x, op) for x in other])
+        return NotImplemented
+
+    def __hash__(self):
+        return hash((self.ordinal, self.freqstr))
+
+    def _add_timedeltalike_scalar(self, other) -> "Period":
+        cdef:
+            int64_t inc, ordinal
+
+        if not self._dtype._is_tick_like():
+            raise IncompatibleFrequency("Input cannot be converted to "
+                                        f"Period(freq={self.freqstr})")
+
+        if (
+            cnp.is_timedelta64_object(other) and
+            cnp.get_timedelta64_value(other) == NPY_NAT
+        ):
+            # i.e. np.timedelta64("nat")
+            return NaT
+
+        try:
+            inc = delta_to_nanoseconds(other, reso=self._dtype._creso, round_ok=False)
+        except ValueError as err:
+            raise IncompatibleFrequency("Input cannot be converted to "
+                                        f"Period(freq={self.freqstr})") from err
+        with cython.overflowcheck(True):
+            ordinal = self.ordinal + inc
+        return Period(ordinal=ordinal, freq=self.freq)
+
+    def _add_offset(self, other) -> "Period":
+        # Non-Tick DateOffset other
+        cdef:
+            int64_t ordinal
+
+        self._require_matching_freq(other, base=True)
+
+        ordinal = self.ordinal + other.n
+        return Period(ordinal=ordinal, freq=self.freq)
+
+    @cython.overflowcheck(True)
+    def __add__(self, other):
+        if is_any_td_scalar(other):
+            return self._add_timedeltalike_scalar(other)
+        elif is_offset_object(other):
+            return self._add_offset(other)
+        elif other is NaT:
+            return NaT
+        elif util.is_integer_object(other):
+            ordinal = self.ordinal + other * self._dtype._n
+            return Period(ordinal=ordinal, freq=self.freq)
+
+        elif is_period_object(other):
+            # can't add datetime-like
+            # GH#17983; can't just return NotImplemented bc we get a RecursionError
+            #  when called via np.add.reduce see TestNumpyReductions.test_add
+            #  in npdev build
+            sname = type(self).__name__
+            oname = type(other).__name__
+            raise TypeError(f"unsupported operand type(s) for +: '{sname}' "
+                            f"and '{oname}'")
+
+        elif util.is_array(other):
+            if other.dtype == object:
+                # GH#50162
+                return np.array([self + x for x in other], dtype=object)
+
+        return NotImplemented
+
+    def __radd__(self, other):
+        return self.__add__(other)
+
+    def __sub__(self, other):
+        if (
+            is_any_td_scalar(other)
+            or is_offset_object(other)
+            or util.is_integer_object(other)
+        ):
+            return self + (-other)
+        elif is_period_object(other):
+            self._require_matching_freq(other.freq)
+            # GH 23915 - mul by base freq since __add__ is agnostic of n
+            return (self.ordinal - other.ordinal) * self.freq.base
+        elif other is NaT:
+            return NaT
+
+        elif util.is_array(other):
+            if other.dtype == object:
+                # GH#50162
+                return np.array([self - x for x in other], dtype=object)
+
+        return NotImplemented
+
+    def __rsub__(self, other):
+        if other is NaT:
+            return NaT
+
+        elif util.is_array(other):
+            if other.dtype == object:
+                # GH#50162
+                return np.array([x - self for x in other], dtype=object)
+
+        return NotImplemented
+
+    def asfreq(self, freq, how="E") -> "Period":
+        """
+        Convert Period to desired frequency, at the start or end of the interval.
+
+        Parameters
+        ----------
+        freq : str, BaseOffset
+            The target frequency to convert the Period object to.
+            If a string is provided,
+            it must be a valid :ref:`period alias <timeseries.period_aliases>`.
+
+        how : {'E', 'S', 'end', 'start'}, default 'end'
+            Specifies whether to align the period to the start or end of the interval:
+            - 'E' or 'end': Align to the end of the interval.
+            - 'S' or 'start': Align to the start of the interval.
+
+        Returns
+        -------
+        Period : Period object with the specified frequency, aligned to the parameter.
+
+        See Also
+        --------
+        Period.end_time : Return the end Timestamp.
+        Period.start_time : Return the start Timestamp.
+        Period.dayofyear : Return the day of the year.
+        Period.dayofweek : Return the day of the week.
+
+        Examples
+        --------
+        Convert a daily period to an hourly period, aligning to the end of the day:
+
+        >>> period = pd.Period('2023-01-01', freq='D')
+        >>> period.asfreq('h')
+        Period('2023-01-01 23:00', 'h')
+
+        Convert a monthly period to a daily period, aligning to the start of the month:
+
+        >>> period = pd.Period('2023-01', freq='M')
+        >>> period.asfreq('D', how='start')
+        Period('2023-01-01', 'D')
+
+        Convert a yearly period to a monthly period, aligning to the last month:
+
+        >>> period = pd.Period('2023', freq='Y')
+        >>> period.asfreq('M', how='end')
+        Period('2023-12', 'M')
+
+        Convert a monthly period to an hourly period,
+        aligning to the first day of the month:
+
+        >>> period = pd.Period('2023-01', freq='M')
+        >>> period.asfreq('h', how='start')
+        Period('2023-01-01 00:00', 'H')
+
+        Convert a weekly period to a daily period, aligning to the last day of the week:
+
+        >>> period = pd.Period('2023-08-01', freq='W')
+        >>> period.asfreq('D', how='end')
+        Period('2023-08-04', 'D')
+        """
+        freq = self._maybe_convert_freq(freq)
+        how = validate_end_alias(how)
+        base1 = self._dtype._dtype_code
+        base2 = freq_to_dtype_code(freq)
+
+        # self.n can't be negative or 0
+        end = how == "E"
+        if end:
+            ordinal = self.ordinal + self._dtype._n - 1
+        else:
+            ordinal = self.ordinal
+        ordinal = period_asfreq(ordinal, base1, base2, end)
+
+        return Period(ordinal=ordinal, freq=freq)
+
+    def to_timestamp(self, freq=None, how="start") -> Timestamp:
+        """
+        Return the Timestamp representation of the Period.
+
+        Uses the target frequency specified at the part of the period specified
+        by `how`, which is either `Start` or `Finish`.
+
+        Parameters
+        ----------
+        freq : str or DateOffset
+            Target frequency. Default is 'D' if self.freq is week or
+            longer and 'S' otherwise.
+        how : str, default 'S' (start)
+            One of 'S', 'E'. Can be aliased as case insensitive
+            'Start', 'Finish', 'Begin', 'End'.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Timestamp : A class representing a single point in time.
+        Period : Represents a span of time with a fixed frequency.
+        PeriodIndex.to_timestamp : Convert a `PeriodIndex` to a `DatetimeIndex`.
+
+        Examples
+        --------
+        >>> period = pd.Period('2023-1-1', freq='D')
+        >>> timestamp = period.to_timestamp()
+        >>> timestamp
+        Timestamp('2023-01-01 00:00:00')
+        """
+        how = validate_end_alias(how)
+
+        end = how == "E"
+        if end:
+            if freq == "B" or self.freq == "B":
+                # roll forward to ensure we land on B date
+                adjust = np.timedelta64(1, "D") - np.timedelta64(1, "ns")
+                return self.to_timestamp(how="start") + adjust
+            endpoint = (self + self.freq).to_timestamp(how="start")
+            return endpoint - np.timedelta64(1, "ns")
+
+        if freq is None:
+            freq_code = self._dtype._get_to_timestamp_base()
+            dtype = PeriodDtypeBase(freq_code, 1)
+            freq = dtype._freqstr
+            base = freq_code
+        else:
+            freq = self._maybe_convert_freq(freq)
+            base = freq._period_dtype_code
+
+        val = self.asfreq(freq, how)
+
+        dt64 = period_ordinal_to_dt64(val.ordinal, base)
+        return Timestamp(dt64)
+
+    @property
+    def year(self) -> int:
+        """
+        Return the year this Period falls on.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        period.month : Get the month of the year for the given Period.
+        period.day : Return the day of the month the Period falls on.
+
+        Notes
+        -----
+        The year is based on the `ordinal` and `base` attributes of the Period.
+
+        Examples
+        --------
+        Create a Period object for January 2023 and get the year:
+
+        >>> period = pd.Period('2023-01', 'M')
+        >>> period.year
+        2023
+
+        Create a Period object for 01 January 2023 and get the year:
+
+        >>> period = pd.Period('2023', 'D')
+        >>> period.year
+        2023
+
+        Get the year for a period representing a quarter:
+
+        >>> period = pd.Period('2023Q2', 'Q')
+        >>> period.year
+        2023
+
+        Handle a case where the Period object is empty, which results in `NaN`:
+
+        >>> period = pd.Period('nan', 'M')
+        >>> period.year
+        nan
+        """
+        base = self._dtype._dtype_code
+        return pyear(self.ordinal, base)
+
+    @property
+    def month(self) -> int:
+        """
+        Return the month this Period falls on.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        period.week : Get the week of the year on the given Period.
+        Period.year : Return the year this Period falls on.
+        Period.day : Return the day of the month this Period falls on.
+
+        Notes
+        -----
+        The month is based on the `ordinal` and `base` attributes of the Period.
+
+        Examples
+        --------
+        Create a Period object for January 2022 and get the month:
+
+        >>> period = pd.Period('2022-01', 'M')
+        >>> period.month
+        1
+
+        Period object with no specified frequency, resulting in a default frequency:
+
+        >>> period = pd.Period('2022', 'Y')
+        >>> period.month
+        12
+
+        Create a Period object with a specified frequency but an incomplete date string:
+
+        >>> period = pd.Period('2022', 'M')
+        >>> period.month
+        1
+
+        Handle a case where the Period object is empty, which results in `NaN`:
+
+        >>> period = pd.Period('nan', 'M')
+        >>> period.month
+        nan
+        """
+        base = self._dtype._dtype_code
+        return pmonth(self.ordinal, base)
+
+    @property
+    def day(self) -> int:
+        """
+        Get day of the month that a Period falls on.
+
+        The `day` property provides a simple way to access the day component
+        of a `Period` object, which represents time spans in various frequencies
+        (e.g., daily, hourly, monthly). If the period's frequency does not include
+        a day component (e.g., yearly or quarterly periods), the returned day
+        corresponds to the first day of that period.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Period.dayofweek : Get the day of the week.
+        Period.dayofyear : Get the day of the year.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11", freq='h')
+        >>> p.day
+        11
+        """
+        base = self._dtype._dtype_code
+        return pday(self.ordinal, base)
+
+    @property
+    def hour(self) -> int:
+        """
+        Get the hour of the day component of the Period.
+
+        Returns
+        -------
+        int
+            The hour as an integer, between 0 and 23.
+
+        See Also
+        --------
+        Period.second : Get the second component of the Period.
+        Period.minute : Get the minute component of the Period.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11 13:03:12.050000")
+        >>> p.hour
+        13
+
+        Period longer than a day
+
+        >>> p = pd.Period("2018-03-11", freq="M")
+        >>> p.hour
+        0
+        """
+        base = self._dtype._dtype_code
+        return phour(self.ordinal, base)
+
+    @property
+    def minute(self) -> int:
+        """
+        Get minute of the hour component of the Period.
+
+        Returns
+        -------
+        int
+            The minute as an integer, between 0 and 59.
+
+        See Also
+        --------
+        Period.hour : Get the hour component of the Period.
+        Period.second : Get the second component of the Period.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11 13:03:12.050000")
+        >>> p.minute
+        3
+        """
+        base = self._dtype._dtype_code
+        return pminute(self.ordinal, base)
+
+    @property
+    def second(self) -> int:
+        """
+        Get the second component of the Period.
+
+        Returns
+        -------
+        int
+            The second of the Period (ranges from 0 to 59).
+
+        See Also
+        --------
+        Period.hour : Get the hour component of the Period.
+        Period.minute : Get the minute component of the Period.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11 13:03:12.050000")
+        >>> p.second
+        12
+        """
+        base = self._dtype._dtype_code
+        return psecond(self.ordinal, base)
+
+    @property
+    def weekofyear(self) -> int:
+        """
+        Get the week of the year on the given Period.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Period.dayofweek : Get the day component of the Period.
+        Period.weekday : Get the day component of the Period.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11", "h")
+        >>> p.weekofyear
+        10
+
+        >>> p = pd.Period("2018-02-01", "D")
+        >>> p.weekofyear
+        5
+
+        >>> p = pd.Period("2018-01-06", "D")
+        >>> p.weekofyear
+        1
+        """
+        base = self._dtype._dtype_code
+        return pweek(self.ordinal, base)
+
+    @property
+    def week(self) -> int:
+        """
+        Get the week of the year on the given Period.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Period.dayofweek : Get the day component of the Period.
+        Period.weekday : Get the day component of the Period.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11", "h")
+        >>> p.week
+        10
+
+        >>> p = pd.Period("2018-02-01", "D")
+        >>> p.week
+        5
+
+        >>> p = pd.Period("2018-01-06", "D")
+        >>> p.week
+        1
+        """
+        return self.weekofyear
+
+    @property
+    def day_of_week(self) -> int:
+        """
+        Day of the week the period lies in, with Monday=0 and Sunday=6.
+
+        If the period frequency is lower than daily (e.g. hourly), and the
+        period spans over multiple days, the day at the start of the period is
+        used.
+
+        If the frequency is higher than daily (e.g. monthly), the last day
+        of the period is used.
+
+        Returns
+        -------
+        int
+            Day of the week.
+
+        See Also
+        --------
+        Period.day_of_week : Day of the week the period lies in.
+        Period.weekday : Alias of Period.day_of_week.
+        Period.day : Day of the month.
+        Period.dayofyear : Day of the year.
+
+        Examples
+        --------
+        >>> per = pd.Period('2017-12-31 22:00', 'h')
+        >>> per.day_of_week
+        6
+
+        For periods that span over multiple days, the day at the beginning of
+        the period is returned.
+
+        >>> per = pd.Period('2017-12-31 22:00', '4h')
+        >>> per.day_of_week
+        6
+        >>> per.start_time.day_of_week
+        6
+
+        For periods with a frequency higher than days, the last day of the
+        period is returned.
+
+        >>> per = pd.Period('2018-01', 'M')
+        >>> per.day_of_week
+        2
+        >>> per.end_time.day_of_week
+        2
+        """
+        base = self._dtype._dtype_code
+        return pweekday(self.ordinal, base)
+
+    @property
+    def weekday(self) -> int:
+        """
+        Day of the week the period lies in, with Monday=0 and Sunday=6.
+
+        If the period frequency is lower than daily (e.g. hourly), and the
+        period spans over multiple days, the day at the start of the period is
+        used.
+
+        If the frequency is higher than daily (e.g. monthly), the last day
+        of the period is used.
+
+        Returns
+        -------
+        int
+            Day of the week.
+
+        See Also
+        --------
+        Period.dayofweek : Day of the week the period lies in.
+        Period.weekday : Alias of Period.dayofweek.
+        Period.day : Day of the month.
+        Period.dayofyear : Day of the year.
+
+        Examples
+        --------
+        >>> per = pd.Period('2017-12-31 22:00', 'h')
+        >>> per.dayofweek
+        6
+
+        For periods that span over multiple days, the day at the beginning of
+        the period is returned.
+
+        >>> per = pd.Period('2017-12-31 22:00', '4h')
+        >>> per.dayofweek
+        6
+        >>> per.start_time.dayofweek
+        6
+
+        For periods with a frequency higher than days, the last day of the
+        period is returned.
+
+        >>> per = pd.Period('2018-01', 'M')
+        >>> per.dayofweek
+        2
+        >>> per.end_time.dayofweek
+        2
+        """
+        # Docstring is a duplicate from dayofweek. Reusing docstrings with
+        # Appender doesn't work for properties in Cython files, and setting
+        # the __doc__ attribute is also not possible.
+        return self.dayofweek
+
+    @property
+    def day_of_year(self) -> int:
+        """
+        Return the day of the year.
+
+        This attribute returns the day of the year on which the particular
+        date occurs. The return value ranges between 1 to 365 for regular
+        years and 1 to 366 for leap years.
+
+        Returns
+        -------
+        int
+            The day of year.
+
+        See Also
+        --------
+        Period.day : Return the day of the month.
+        Period.day_of_week : Return the day of week.
+        PeriodIndex.day_of_year : Return the day of year of all indexes.
+
+        Examples
+        --------
+        >>> period = pd.Period("2015-10-23", freq='h')
+        >>> period.day_of_year
+        296
+        >>> period = pd.Period("2012-12-31", freq='D')
+        >>> period.day_of_year
+        366
+        >>> period = pd.Period("2013-01-01", freq='D')
+        >>> period.day_of_year
+        1
+        """
+        base = self._dtype._dtype_code
+        return pday_of_year(self.ordinal, base)
+
+    @property
+    def quarter(self) -> int:
+        """
+        Return the quarter this Period falls on.
+
+        See Also
+        --------
+        Timestamp.quarter : Return the quarter of the Timestamp.
+        Period.year : Return the year of the period.
+        Period.month : Return the month of the period.
+
+        Examples
+        --------
+        >>> period = pd.Period('2022-04', 'M')
+        >>> period.quarter
+        2
+        """
+        base = self._dtype._dtype_code
+        return pquarter(self.ordinal, base)
+
+    @property
+    def qyear(self) -> int:
+        """
+        Fiscal year the Period lies in according to its starting-quarter.
+
+        The `year` and the `qyear` of the period will be the same if the fiscal
+        and calendar years are the same. When they are not, the fiscal year
+        can be different from the calendar year of the period.
+
+        Returns
+        -------
+        int
+            The fiscal year of the period.
+
+        See Also
+        --------
+        Period.year : Return the calendar year of the period.
+
+        Examples
+        --------
+        If the natural and fiscal year are the same, `qyear` and `year` will
+        be the same.
+
+        >>> per = pd.Period('2018Q1', freq='Q')
+        >>> per.qyear
+        2018
+        >>> per.year
+        2018
+
+        If the fiscal year starts in April (`Q-MAR`), the first quarter of
+        2018 will start in April 2017. `year` will then be 2017, but `qyear`
+        will be the fiscal year, 2018.
+
+        >>> per = pd.Period('2018Q1', freq='Q-MAR')
+        >>> per.start_time
+        Timestamp('2017-04-01 00:00:00')
+        >>> per.qyear
+        2018
+        >>> per.year
+        2017
+        """
+        base = self._dtype._dtype_code
+        return pqyear(self.ordinal, base)
+
+    @property
+    def days_in_month(self) -> int:
+        """
+        Get the total number of days in the month that this period falls on.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Period.daysinmonth : Gets the number of days in the month.
+        DatetimeIndex.daysinmonth : Gets the number of days in the month.
+        calendar.monthrange : Returns a tuple containing weekday
+            (0-6 ~ Mon-Sun) and number of days (28-31).
+
+        Examples
+        --------
+        >>> p = pd.Period('2018-2-17')
+        >>> p.days_in_month
+        28
+
+        >>> pd.Period('2018-03-01').days_in_month
+        31
+
+        Handles the leap year case as well:
+
+        >>> p = pd.Period('2016-2-17')
+        >>> p.days_in_month
+        29
+        """
+        base = self._dtype._dtype_code
+        return pdays_in_month(self.ordinal, base)
+
+    @property
+    def daysinmonth(self) -> int:
+        """
+        Get the total number of days of the month that this period falls on.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Period.days_in_month : Return the days of the month.
+        Period.dayofyear : Return the day of the year.
+
+        Examples
+        --------
+        >>> p = pd.Period("2018-03-11", freq='h')
+        >>> p.daysinmonth
+        31
+        """
+        return self.days_in_month
+
+    @property
+    def is_leap_year(self) -> bool:
+        """
+        Return True if the period's year is in a leap year.
+
+        See Also
+        --------
+        Timestamp.is_leap_year : Check if the year in a Timestamp is a leap year.
+        DatetimeIndex.is_leap_year : Boolean indicator if the date belongs to a
+            leap year.
+
+        Examples
+        --------
+        >>> period = pd.Period('2022-01', 'M')
+        >>> period.is_leap_year
+        False
+
+        >>> period = pd.Period('2020-01', 'M')
+        >>> period.is_leap_year
+        True
+        """
+        return bool(is_leapyear(self.year))
+
+    @classmethod
+    def now(cls, freq):
+        """
+        Return the period of now's date.
+
+        The `now` method provides a convenient way to generate a period
+        object for the current date and time. This can be particularly
+        useful in financial and economic analysis, where data is often
+        collected and analyzed in regular intervals (e.g., hourly, daily,
+        monthly). By specifying the frequency, users can create periods
+        that match the granularity of their data.
+
+        Parameters
+        ----------
+        freq : str, BaseOffset
+            Frequency to use for the returned period.
+
+        See Also
+        --------
+        to_datetime : Convert argument to datetime.
+        Period : Represents a period of time.
+        Period.to_timestamp : Return the Timestamp representation of the Period.
+
+        Examples
+        --------
+        >>> pd.Period.now('h')  # doctest: +SKIP
+        Period('2023-06-12 11:00', 'h')
+        """
+        return Period(datetime.now(), freq=freq)
+
+    @property
+    def freqstr(self) -> str:
+        """
+        Return a string representation of the frequency.
+
+        This property provides the frequency string associated with the `Period`
+        object. The frequency string describes the granularity of the time span
+        represented by the `Period`. Common frequency strings include 'D' for
+        daily, 'M' for monthly, 'Y' for yearly, etc.
+
+        See Also
+        --------
+        Period.asfreq : Convert Period to desired frequency, at the start or end
+            of the interval.
+        period_range : Return a fixed frequency PeriodIndex.
+
+        Examples
+        --------
+        >>> pd.Period('2020-01', 'D').freqstr
+        'D'
+        """
+        freqstr = PeriodDtypeBase(self.freq._period_dtype_code, self.freq.n)._freqstr
+        return freqstr
+
+    def __repr__(self) -> str:
+        base = self._dtype._dtype_code
+        formatted = period_format(self.ordinal, base)
+        return f"Period('{formatted}', '{self.freqstr}')"
+
+    def __str__(self) -> str:
+        """
+        Return a string representation for a particular DataFrame
+        """
+        base = self._dtype._dtype_code
+        formatted = period_format(self.ordinal, base)
+        value = str(formatted)
+        return value
+
+    def __setstate__(self, state):
+        self.freq = state[1]
+        self.ordinal = state[2]
+
+    def __reduce__(self):
+        object_state = None, self.freq, self.ordinal
+        return (Period, object_state)
+
+    def strftime(self, fmt: str | None) -> str:
+        r"""
+        Returns a formatted string representation of the :class:`Period`.
+
+        ``fmt`` must be ``None`` or a string containing one or several directives.
+        When ``None``, the format will be determined from the frequency of the Period.
+        The method recognizes the same directives as the :func:`time.strftime`
+        function of the standard Python distribution, as well as the specific
+        additional directives ``%f``, ``%F``, ``%q``, ``%l``, ``%u``, ``%n``.
+        (formatting & docs originally from scikits.timeries).
+
+        +-----------+--------------------------------+-------+
+        | Directive | Meaning                        | Notes |
+        +===========+================================+=======+
+        | ``%a``    | Locale's abbreviated weekday   |       |
+        |           | name.                          |       |
+        +-----------+--------------------------------+-------+
+        | ``%A``    | Locale's full weekday name.    |       |
+        +-----------+--------------------------------+-------+
+        | ``%b``    | Locale's abbreviated month     |       |
+        |           | name.                          |       |
+        +-----------+--------------------------------+-------+
+        | ``%B``    | Locale's full month name.      |       |
+        +-----------+--------------------------------+-------+
+        | ``%c``    | Locale's appropriate date and  |       |
+        |           | time representation.           |       |
+        +-----------+--------------------------------+-------+
+        | ``%d``    | Day of the month as a decimal  |       |
+        |           | number [01,31].                |       |
+        +-----------+--------------------------------+-------+
+        | ``%f``    | 'Fiscal' year without a        | \(1)  |
+        |           | century  as a decimal number   |       |
+        |           | [00,99]                        |       |
+        +-----------+--------------------------------+-------+
+        | ``%F``    | 'Fiscal' year with a century   | \(2)  |
+        |           | as a decimal number            |       |
+        +-----------+--------------------------------+-------+
+        | ``%H``    | Hour (24-hour clock) as a      |       |
+        |           | decimal number [00,23].        |       |
+        +-----------+--------------------------------+-------+
+        | ``%I``    | Hour (12-hour clock) as a      |       |
+        |           | decimal number [01,12].        |       |
+        +-----------+--------------------------------+-------+
+        | ``%j``    | Day of the year as a decimal   |       |
+        |           | number [001,366].              |       |
+        +-----------+--------------------------------+-------+
+        | ``%m``    | Month as a decimal number      |       |
+        |           | [01,12].                       |       |
+        +-----------+--------------------------------+-------+
+        | ``%M``    | Minute as a decimal number     |       |
+        |           | [00,59].                       |       |
+        +-----------+--------------------------------+-------+
+        | ``%p``    | Locale's equivalent of either  | \(3)  |
+        |           | AM or PM.                      |       |
+        +-----------+--------------------------------+-------+
+        | ``%q``    | Quarter as a decimal number    |       |
+        |           | [1,4]                          |       |
+        +-----------+--------------------------------+-------+
+        | ``%S``    | Second as a decimal number     | \(4)  |
+        |           | [00,61].                       |       |
+        +-----------+--------------------------------+-------+
+        | ``%l``    | Millisecond as a decimal number|       |
+        |           | [000,999].                     |       |
+        +-----------+--------------------------------+-------+
+        | ``%u``    | Microsecond as a decimal number|       |
+        |           | [000000,999999].               |       |
+        +-----------+--------------------------------+-------+
+        | ``%n``    | Nanosecond as a decimal number |       |
+        |           | [000000000,999999999].         |       |
+        +-----------+--------------------------------+-------+
+        | ``%U``    | Week number of the year        | \(5)  |
+        |           | (Sunday as the first day of    |       |
+        |           | the week) as a decimal number  |       |
+        |           | [00,53].  All days in a new    |       |
+        |           | year preceding the first       |       |
+        |           | Sunday are considered to be in |       |
+        |           | week 0.                        |       |
+        +-----------+--------------------------------+-------+
+        | ``%w``    | Weekday as a decimal number    |       |
+        |           | [0(Sunday),6].                 |       |
+        +-----------+--------------------------------+-------+
+        | ``%W``    | Week number of the year        | \(5)  |
+        |           | (Monday as the first day of    |       |
+        |           | the week) as a decimal number  |       |
+        |           | [00,53].  All days in a new    |       |
+        |           | year preceding the first       |       |
+        |           | Monday are considered to be in |       |
+        |           | week 0.                        |       |
+        +-----------+--------------------------------+-------+
+        | ``%x``    | Locale's appropriate date      |       |
+        |           | representation.                |       |
+        +-----------+--------------------------------+-------+
+        | ``%X``    | Locale's appropriate time      |       |
+        |           | representation.                |       |
+        +-----------+--------------------------------+-------+
+        | ``%y``    | Year without century as a      |       |
+        |           | decimal number [00,99].        |       |
+        +-----------+--------------------------------+-------+
+        | ``%Y``    | Year with century as a decimal |       |
+        |           | number.                        |       |
+        +-----------+--------------------------------+-------+
+        | ``%Z``    | Time zone name (no characters  |       |
+        |           | if no time zone exists).       |       |
+        +-----------+--------------------------------+-------+
+        | ``%%``    | A literal ``'%'`` character.   |       |
+        +-----------+--------------------------------+-------+
+
+        The `strftime` method provides a way to represent a :class:`Period`
+        object as a string in a specified format. This is particularly useful
+        when displaying date and time data in different locales or customized
+        formats, suitable for reports or user interfaces. It extends the standard
+        Python string formatting capabilities with additional directives specific
+        to `pandas`, accommodating features like fiscal years and precise
+        sub-second components.
+
+        Parameters
+        ----------
+        fmt : str or None
+            String containing the desired format directives. If ``None``, the
+            format is determined based on the Period's frequency.
+
+        See Also
+        --------
+        Timestamp.strftime : Return a formatted string of the Timestamp.
+        to_datetime : Convert argument to datetime.
+        time.strftime : Format a time object as a string according to a
+            specified format string in the standard Python library.
+
+        Notes
+        -----
+
+        (1)
+            The ``%f`` directive is the same as ``%y`` if the frequency is
+            not quarterly.
+            Otherwise, it corresponds to the 'fiscal' year, as defined by
+            the :attr:`qyear` attribute.
+
+        (2)
+            The ``%F`` directive is the same as ``%Y`` if the frequency is
+            not quarterly.
+            Otherwise, it corresponds to the 'fiscal' year, as defined by
+            the :attr:`qyear` attribute.
+
+        (3)
+            The ``%p`` directive only affects the output hour field
+            if the ``%I`` directive is used to parse the hour.
+
+        (4)
+            The range really is ``0`` to ``61``; this accounts for leap
+            seconds and the (very rare) double leap seconds.
+
+        (5)
+            The ``%U`` and ``%W`` directives are only used in calculations
+            when the day of the week and the year are specified.
+
+        Examples
+        --------
+
+        >>> from pandas import Period
+        >>> a = Period(freq='Q-JUL', year=2006, quarter=1)
+        >>> a.strftime('%F-Q%q')
+        '2006-Q1'
+        >>> # Output the last month in the quarter of this date
+        >>> a.strftime('%b-%Y')
+        'Oct-2005'
+        >>>
+        >>> a = Period(freq='D', year=2001, month=1, day=1)
+        >>> a.strftime('%d-%b-%Y')
+        '01-Jan-2001'
+        >>> a.strftime('%b. %d, %Y was a %A')
+        'Jan. 01, 2001 was a Monday'
+        """
+        base = self._dtype._dtype_code
+        return period_format(self.ordinal, base, fmt)
+
+
+@set_module("pandas")
+class Period(_Period):
+    """
+    Represents a period of time.
+
+    A `Period` represents a specific time span rather than a point in time.
+    Unlike `Timestamp`, which represents a single instant, a `Period` defines a
+    duration, such as a month, quarter, or year. The exact representation is
+    determined by the `freq` parameter.
+
+    Parameters
+    ----------
+    value : Period, str, datetime, date or pandas.Timestamp, default None
+        The time period represented (e.g., '4Q2005'). This represents neither
+        the start or the end of the period, but rather the entire period itself.
+    freq : str, default None
+        One of pandas period strings or corresponding objects. Accepted
+        strings are listed in the
+        :ref:`period alias section <timeseries.period_aliases>` in the user docs.
+        If value is datetime, freq is required.
+    ordinal : int, default None
+        The period offset from the proleptic Gregorian epoch.
+    year : int, default None
+        Year value of the period.
+    month : int, default 1
+        Month value of the period.
+    quarter : int, default None
+        Quarter value of the period.
+    day : int, default 1
+        Day value of the period.
+    hour : int, default 0
+        Hour value of the period.
+    minute : int, default 0
+        Minute value of the period.
+    second : int, default 0
+        Second value of the period.
+
+    See Also
+    --------
+    Timestamp : Pandas replacement for python datetime.datetime object.
+    date_range : Return a fixed frequency DatetimeIndex.
+    timedelta_range : Generates a fixed frequency range of timedeltas.
+
+    Examples
+    --------
+    >>> period = pd.Period('2012-1-1', freq='D')
+    >>> period
+    Period('2012-01-01', 'D')
+    """
+
+    def __new__(cls, value=None, freq=None, ordinal=None,
+                year=None, month=None, quarter=None, day=None,
+                hour=None, minute=None, second=None):
+        # freq points to a tuple (base, mult);  base is one of the defined
+        # periods such as Y, Q, etc. Every five minutes would be, e.g.,
+        # ('min', 5) but may be passed in as a string like '5min'
+
+        # ordinal is the period offset from the gregorian proleptic epoch
+
+        if freq is not None:
+            freq = cls._maybe_convert_freq(freq)
+            try:
+                period_dtype_code = freq._period_dtype_code
+            except (AttributeError, TypeError):
+                # AttributeError: _period_dtype_code might not exist
+                # TypeError: _period_dtype_code might intentionally raise
+                raise TypeError(
+                    f"{(type(freq).__name__)} is not supported as period frequency"
+                )
+        nanosecond = 0
+
+        if ordinal is not None and value is not None:
+            raise ValueError("Only value or ordinal but not both should be "
+                             "given but not both")
+        elif ordinal is not None:
+            if not util.is_integer_object(ordinal):
+                raise ValueError("Ordinal must be an integer")
+            if freq is None:
+                raise ValueError("Must supply freq for ordinal value")
+
+        elif value is None:
+            if (year is None and month is None and
+                    quarter is None and day is None and
+                    hour is None and minute is None and second is None):
+                ordinal = NPY_NAT
+            else:
+                if freq is None:
+                    raise ValueError("If value is None, freq cannot be None")
+
+                # set defaults
+                month = 1 if month is None else month
+                day = 1 if day is None else day
+                hour = 0 if hour is None else hour
+                minute = 0 if minute is None else minute
+                second = 0 if second is None else second
+
+                ordinal = _ordinal_from_fields(year, month, quarter, day,
+                                               hour, minute, second, freq)
+
+        elif is_period_object(value):
+            other = value
+            if freq is None or period_dtype_code == other._dtype._dtype_code:
+                ordinal = other.ordinal
+                freq = other.freq
+            else:
+                converted = other.asfreq(freq)
+                ordinal = converted.ordinal
+
+        elif checknull_with_nat(value) or (isinstance(value, str) and
+                                           (value in nat_strings or len(value) == 0)):
+            # explicit str check is necessary to avoid raising incorrectly
+            #  if we have a non-hashable value.
+            ordinal = NPY_NAT
+
+        elif isinstance(value, str) or util.is_integer_object(value):
+            if util.is_integer_object(value):
+                if value == NPY_NAT:
+                    value = "NaT"
+
+                value = str(value)
+            value = value.upper()
+
+            freqstr = freq.rule_code if freq is not None else None
+            try:
+                dt, reso = parse_datetime_string_with_reso(value, freqstr)
+            except ValueError as err:
+                match = re.search(r"^\d{4}-\d{2}-\d{2}/\d{4}-\d{2}-\d{2}", value)
+                if match:
+                    # Case that cannot be parsed (correctly) by our datetime
+                    #  parsing logic
+                    dt, freq = _parse_weekly_str(value, freq)
+                else:
+                    raise err
+
+            else:
+                if reso == "nanosecond":
+                    nanosecond = dt.nanosecond
+                if dt is NaT:
+                    ordinal = NPY_NAT
+
+                if freq is None and ordinal != NPY_NAT:
+                    # Skip NaT, since it doesn't have a resolution
+                    freq = attrname_to_abbrevs[reso]
+                    freq = c_OFFSET_TO_PERIOD_FREQSTR.get(freq, freq)
+                    freq = to_offset(freq, is_period=True)
+
+        elif PyDateTime_Check(value):
+            dt = value
+            if freq is None:
+                raise ValueError("Must supply freq for datetime value")
+            if isinstance(dt, Timestamp):
+                nanosecond = dt.nanosecond
+        elif cnp.is_datetime64_object(value):
+            dt = Timestamp(value)
+            if freq is None:
+                raise ValueError("Must supply freq for datetime value")
+            nanosecond = dt.nanosecond
+        elif PyDate_Check(value):
+            dt = datetime(year=value.year, month=value.month, day=value.day)
+            if freq is None:
+                raise ValueError("Must supply freq for datetime value")
+        else:
+            msg = "Value must be Period, string, integer, or datetime"
+            raise ValueError(msg)
+
+        if ordinal is None:
+            base = freq_to_dtype_code(freq)
+            ordinal = period_ordinal(dt.year, dt.month, dt.day,
+                                     dt.hour, dt.minute, dt.second,
+                                     dt.microsecond, 1000*nanosecond, base)
+
+        if isinstance(freq, BDay):
+            # GH#53446
+            import warnings
+
+            from pandas.util._exceptions import find_stack_level
+            warnings.warn(
+                "Period with BDay freq is deprecated and will be removed "
+                "in a future version. Use a DatetimeIndex with BDay freq instead.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        if ordinal == NPY_NAT:
+            return NaT
+        return cls._from_ordinal(ordinal, freq)
+
+
+cdef bint is_period_object(object obj):
+    return isinstance(obj, _Period)
+
+
+cpdef int freq_to_dtype_code(BaseOffset freq) except? -1:
+    try:
+        return freq._period_dtype_code
+    except AttributeError as err:
+        raise ValueError(INVALID_FREQ_ERR_MSG.format(freq)) from err
+
+
+cdef int64_t _ordinal_from_fields(int year, int month, quarter, int day,
+                                  int hour, int minute, int second,
+                                  BaseOffset freq):
+    base = freq_to_dtype_code(freq)
+    if quarter is not None:
+        year, month = quarter_to_myear(year, quarter, freq.freqstr)
+
+    return period_ordinal(year, month, day, hour,
+                          minute, second, 0, 0, base)
+
+
+def validate_end_alias(how: str) -> str:  # Literal["E", "S"]
+    how_dict = {"S": "S", "E": "E",
+                "START": "S", "FINISH": "E",
+                "BEGIN": "S", "END": "E"}
+    how = how_dict.get(str(how).upper())
+    if how not in {"S", "E"}:
+        raise ValueError("How must be one of S or E")
+    return how
+
+
+cdef _parse_weekly_str(value, BaseOffset freq):
+    """
+    Parse e.g. "2017-01-23/2017-01-29", which cannot be parsed by the general
+    datetime-parsing logic.  This ensures that we can round-trip with
+    Period.__str__ with weekly freq.
+    """
+    # GH#50803
+    start, end = value.split("/")
+    start = Timestamp(start)
+    end = Timestamp(end)
+
+    if (end - start).days != 6:
+        # We are interested in cases where this is str(period)
+        #  of a Week-freq period
+        raise ValueError("Could not parse as weekly-freq Period")
+
+    if freq is None:
+        day_name = end.day_name()[:3].upper()
+        freqstr = f"W-{day_name}"
+        freq = to_offset(freqstr, is_period=True)
+        # We _should_ have freq.is_on_offset(end)
+
+    return end, freq
diff --git a/pandas/pandas/_libs/tslibs/strptime.pxd b/pandas/pandas/_libs/tslibs/strptime.pxd
new file mode 100644
index 0000000000000..d2eae910a87b5
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/strptime.pxd
@@ -0,0 +1,29 @@
+from cpython.datetime cimport (
+    datetime,
+    tzinfo,
+)
+from numpy cimport int64_t
+
+from pandas._libs.tslibs.np_datetime cimport NPY_DATETIMEUNIT
+
+
+cdef bint parse_today_now(
+    str val, int64_t* iresult, bint utc, NPY_DATETIMEUNIT creso, bint infer_reso=*
+)
+
+
+cdef class DatetimeParseState:
+    cdef:
+        # See comments describing these attributes in the __cinit__ method
+        bint found_tz
+        bint found_naive
+        bint found_naive_str
+        bint found_aware_str
+        bint found_other
+        bint creso_ever_changed
+        NPY_DATETIMEUNIT creso
+        set out_tzoffset_vals
+
+    cdef tzinfo process_datetime(self, datetime dt, tzinfo tz, bint utc_convert)
+    cdef bint update_creso(self, NPY_DATETIMEUNIT item_reso) noexcept
+    cdef tzinfo check_for_mixed_inputs(self, tzinfo tz_out, bint utc)
diff --git a/pandas/pandas/_libs/tslibs/strptime.pyi b/pandas/pandas/_libs/tslibs/strptime.pyi
new file mode 100644
index 0000000000000..0ec1a1e25a2b3
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/strptime.pyi
@@ -0,0 +1,14 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def array_strptime(
+    values: npt.NDArray[np.object_],
+    fmt: str | None,
+    exact: bool = ...,
+    errors: str = ...,
+    utc: bool = ...,
+    creso: int = ...,  # NPY_DATETIMEUNIT
+) -> tuple[np.ndarray, np.ndarray]: ...
+
+# first ndarray is M8[ns], second is object ndarray of tzinfo | None
diff --git a/pandas/pandas/_libs/tslibs/strptime.pyx b/pandas/pandas/_libs/tslibs/strptime.pyx
new file mode 100644
index 0000000000000..fb89f1328529d
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/strptime.pyx
@@ -0,0 +1,988 @@
+"""Strptime-related classes and functions.
+
+TimeRE, _calc_julian_from_U_or_W are vendored
+from the standard library, see
+https://github.com/python/cpython/blob/main/Lib/_strptime.py
+Licence at LICENSES/PSF_LICENSE
+The original module-level docstring follows.
+
+Strptime-related classes and functions.
+CLASSES:
+    LocaleTime -- Discovers and stores locale-specific time information
+    TimeRE -- Creates regexes for pattern matching a string of text containing
+                time information
+FUNCTIONS:
+    _getlang -- Figure out what language is being used for the locale
+    strptime -- Calculates the time struct represented by the passed-in string
+"""
+from datetime import timezone
+import zoneinfo
+
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    date,
+    import_datetime,
+    timedelta,
+    tzinfo,
+)
+
+from _strptime import (
+    TimeRE as _TimeRE,
+    _getlang,
+)
+from _strptime import LocaleTime  # no-cython-lint
+
+import_datetime()
+
+from _thread import allocate_lock as _thread_allocate_lock
+import re
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+from pandas._libs.missing cimport checknull_with_nat_and_na
+from pandas._libs.tslibs.conversion cimport (
+    get_datetime64_nanos,
+    parse_pydatetime,
+)
+from pandas._libs.tslibs.dtypes cimport (
+    get_supported_reso,
+    npy_unit_to_abbrev,
+    npy_unit_to_attrname,
+)
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_nat_strings as nat_strings,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    get_datetime64_unit,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pydate_to_dt64,
+    string_to_dts,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+
+from pandas._libs.tslibs.timestamps cimport _Timestamp
+from pandas._libs.tslibs.timezones cimport tz_compare
+from pandas._libs.util cimport (
+    is_float_object,
+    is_integer_object,
+)
+
+from pandas._libs.tslibs.timestamps import Timestamp
+
+from pandas._libs.tslibs.tzconversion cimport tz_localize_to_utc_single
+
+cnp.import_array()
+
+
+cdef bint format_is_iso(f: str):
+    """
+    Does format match the iso8601 set that can be handled by the C parser?
+    Generally of form YYYY-MM-DDTHH:MM:SS - date separator can be different
+    but must be consistent.  Leading 0s in dates and times are optional.
+    """
+    iso_regex = re.compile(
+        r"""
+        ^                     # start of string
+        %Y                    # Year
+        (?:([-/ \\.]?)%m      # month with or without separators
+        (?: \1%d              # day with same separator as for year-month
+        (?:[ T]%H             # hour with separator
+        (?:\:%M               # minute with separator
+        (?:\:%S               # second with separator
+        (?:%z|\.%f(?:%z)?     # timezone or fractional second
+        )?)?)?)?)?)?          # optional
+        $                     # end of string
+        """,
+        re.VERBOSE,
+    )
+    excluded_formats = ["%Y%m"]
+    return re.match(iso_regex, f) is not None and f not in excluded_formats
+
+
+def _test_format_is_iso(f: str) -> bool:
+    """Only used in testing."""
+    return format_is_iso(f)
+
+
+cdef bint parse_today_now(
+    str val, int64_t* iresult, bint utc, NPY_DATETIMEUNIT creso, bint infer_reso = False
+):
+    # We delay this check for as long as possible
+    # because it catches relatively rare cases
+    cdef:
+        _Timestamp ts
+
+    if val == "now":
+        if infer_reso:
+            creso = NPY_DATETIMEUNIT.NPY_FR_us
+        if utc:
+            ts = <_Timestamp>Timestamp.now(timezone.utc)
+            iresult[0] = ts._as_creso(creso)._value
+        else:
+            # GH#18705 make sure to_datetime("now") matches Timestamp("now")
+            # Note using Timestamp.now() is faster than Timestamp("now")
+            ts = <_Timestamp>Timestamp.now()
+            iresult[0] = ts._as_creso(creso)._value
+        return True
+    elif val == "today":
+        if infer_reso:
+            creso = NPY_DATETIMEUNIT.NPY_FR_us
+        ts = <_Timestamp>Timestamp.today()
+        iresult[0] = ts._as_creso(creso)._value
+        return True
+    return False
+
+cdef dict _parse_code_table = {"y": 0,
+                               "Y": 1,
+                               "m": 2,
+                               "B": 3,
+                               "b": 4,
+                               "d": 5,
+                               "H": 6,
+                               "I": 7,
+                               "M": 8,
+                               "S": 9,
+                               "f": 10,
+                               "A": 11,
+                               "a": 12,
+                               "w": 13,
+                               "j": 14,
+                               "U": 15,
+                               "W": 16,
+                               "Z": 17,
+                               "p": 18,  # an additional key, only with I
+                               "z": 19,
+                               "G": 20,
+                               "V": 21,
+                               "u": 22}
+
+
+cdef _validate_fmt(str fmt):
+    if "%W" in fmt or "%U" in fmt:
+        if "%Y" not in fmt and "%y" not in fmt:
+            raise ValueError("Cannot use '%W' or '%U' without day and year")
+        if "%A" not in fmt and "%a" not in fmt and "%w" not in fmt:
+            raise ValueError("Cannot use '%W' or '%U' without day and year")
+    elif "%Z" in fmt and "%z" in fmt:
+        raise ValueError("Cannot parse both %Z and %z")
+    elif "%j" in fmt and "%G" in fmt:
+        raise ValueError("Day of the year directive '%j' is not "
+                         "compatible with ISO year directive '%G'. "
+                         "Use '%Y' instead.")
+    elif "%G" in fmt and (
+        "%V" not in fmt
+        or not (
+            "%A" in fmt
+            or "%a" in fmt
+            or "%w" in fmt
+            or "%u" in fmt
+        )
+    ):
+        raise ValueError("ISO year directive '%G' must be used with "
+                         "the ISO week directive '%V' and a weekday "
+                         "directive '%A', '%a', '%w', or '%u'.")
+    elif "%V" in fmt and "%Y" in fmt:
+        raise ValueError("ISO week directive '%V' is incompatible with "
+                         "the year directive '%Y'. Use the ISO year "
+                         "'%G' instead.")
+    elif "%V" in fmt and (
+        "%G" not in fmt
+        or not (
+            "%A" in fmt
+            or "%a" in fmt
+            or "%w" in fmt
+            or "%u" in fmt
+        )
+    ):
+        raise ValueError("ISO week directive '%V' must be used with "
+                         "the ISO year directive '%G' and a weekday "
+                         "directive '%A', '%a', '%w', or '%u'.")
+
+
+cdef _get_format_regex(str fmt):
+    global _TimeRE_cache, _regex_cache
+    with _cache_lock:
+        if _getlang() != _TimeRE_cache.locale_time.lang:
+            _TimeRE_cache = TimeRE()
+            _regex_cache.clear()
+        if len(_regex_cache) > _CACHE_MAX_SIZE:
+            _regex_cache.clear()
+        locale_time = _TimeRE_cache.locale_time
+        format_regex = _regex_cache.get(fmt)
+        if not format_regex:
+            try:
+                format_regex = _TimeRE_cache.compile(fmt)
+            except KeyError, err:
+                # KeyError raised when a bad format is found; can be specified as
+                # \\, in which case it was a stray % but with a space after it
+                bad_directive = err.args[0]
+                if bad_directive == "\\":
+                    bad_directive = "%"
+                del err
+                raise ValueError(f"'{bad_directive}' is a bad directive "
+                                 f"in format '{fmt}'")
+            except IndexError:
+                # IndexError only occurs when the format string is "%"
+                raise ValueError(f"stray % in format '{fmt}'")
+            _regex_cache[fmt] = format_regex
+    return format_regex, locale_time
+
+
+cdef class DatetimeParseState:
+    def __cinit__(self, NPY_DATETIMEUNIT creso):
+        # found_tz and found_naive are specifically about datetime/Timestamp
+        #  objects with and without tzinfos attached.
+        self.found_tz = False
+        self.found_naive = False
+        # found_naive_str refers to a string that was parsed to a timezone-naive
+        #  datetime.
+        self.found_naive_str = False
+        self.found_aware_str = False
+        self.found_other = False
+
+        self.out_tzoffset_vals = set()
+
+        self.creso = creso
+        self.creso_ever_changed = False
+
+    cdef bint update_creso(self, NPY_DATETIMEUNIT item_reso) noexcept:
+        # Return a bool indicating whether we bumped to a higher resolution
+        if self.creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+            self.creso = item_reso
+        elif item_reso > self.creso:
+            self.creso = item_reso
+            self.creso_ever_changed = True
+            return True
+        return False
+
+    cdef tzinfo process_datetime(self, datetime dt, tzinfo tz, bint utc_convert):
+        if dt.tzinfo is not None:
+            self.found_tz = True
+        else:
+            self.found_naive = True
+
+        if dt.tzinfo is not None:
+            if utc_convert:
+                pass
+            elif self.found_naive:
+                raise ValueError("Tz-aware datetime.datetime "
+                                 "cannot be converted to "
+                                 "datetime64 unless utc=True")
+            elif tz is not None and not tz_compare(tz, dt.tzinfo):
+                raise ValueError("Tz-aware datetime.datetime "
+                                 "cannot be converted to "
+                                 "datetime64 unless utc=True")
+            else:
+                tz = dt.tzinfo
+        else:
+            if self.found_tz and not utc_convert:
+                raise ValueError("Cannot mix tz-aware with "
+                                 "tz-naive values")
+        return tz
+
+    cdef tzinfo check_for_mixed_inputs(
+        self,
+        tzinfo tz_out,
+        bint utc,
+    ):
+        cdef:
+            bint is_same_offsets
+            float tz_offset
+
+        if self.found_aware_str and not utc:
+            # GH#17697, GH#57275
+            # 1) If all the offsets are equal, return one offset for
+            #    the parsed dates to (maybe) pass to DatetimeIndex
+            # 2) If the offsets are different, then do not force the parsing
+            #    and raise a ValueError: "cannot parse datetimes with
+            #    mixed time zones unless `utc=True`" instead
+            is_same_offsets = len(self.out_tzoffset_vals) == 1
+            if not is_same_offsets or (self.found_naive or self.found_other):
+                # e.g. test_to_datetime_mixed_awareness_mixed_types (array_to_datetime)
+                raise ValueError(
+                    "Mixed timezones detected. Pass utc=True in to_datetime "
+                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                )
+            elif tz_out is not None:
+                # GH#55693
+                tz_offset = self.out_tzoffset_vals.pop()
+                tz_out2 = timezone(timedelta(seconds=tz_offset))
+                if not tz_compare(tz_out, tz_out2):
+                    # e.g. (array_strptime)
+                    #  test_to_datetime_mixed_offsets_with_utc_false_removed
+                    # e.g. test_to_datetime_mixed_tzs_mixed_types (array_to_datetime)
+                    raise ValueError(
+                        "Mixed timezones detected. Pass utc=True in to_datetime "
+                        "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                    )
+                # e.g. (array_strptime)
+                #  test_guess_datetime_format_with_parseable_formats
+                # e.g. test_to_datetime_mixed_types_matching_tzs (array_to_datetime)
+            else:
+                # e.g. test_to_datetime_iso8601_with_timezone_valid (array_strptime)
+                tz_offset = self.out_tzoffset_vals.pop()
+                tz_out = timezone(timedelta(seconds=tz_offset))
+        elif not utc:
+            if tz_out and (self.found_other or self.found_naive_str):
+                # found_other indicates a tz-naive int, float, dt64, or date
+                # e.g. test_to_datetime_mixed_awareness_mixed_types (array_to_datetime)
+                raise ValueError(
+                    "Mixed timezones detected. Pass utc=True in to_datetime "
+                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                )
+        return tz_out
+
+
+def array_strptime(
+    ndarray[object] values,
+    str fmt,
+    bint exact=True,
+    errors="raise",
+    bint utc=False,
+    NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_GENERIC,
+):
+    """
+    Calculates the datetime structs represented by the passed array of strings
+
+    Parameters
+    ----------
+    values : ndarray of string-like objects
+    fmt : string-like regex
+    exact : matches must be exact if True, search if False
+    errors : string specifying error handling, {'raise', 'coerce'}
+    creso : NPY_DATETIMEUNIT, default NPY_FR_GENERIC
+        Set to NPY_FR_GENERIC to infer a resolution.
+    """
+
+    cdef:
+        Py_ssize_t i, n = len(values)
+        npy_datetimestruct dts
+        int64_t[::1] iresult
+        object val
+        bint is_raise = errors=="raise"
+        bint is_coerce = errors=="coerce"
+        tzinfo tz, tz_out = None
+        bint iso_format = format_is_iso(fmt)
+        NPY_DATETIMEUNIT out_bestunit, item_reso
+        int out_local = 0, out_tzoffset = 0
+        bint string_to_dts_succeeded = 0
+        bint infer_reso = creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC
+        DatetimeParseState state = DatetimeParseState(creso)
+
+    assert is_raise or is_coerce
+
+    _validate_fmt(fmt)
+    format_regex, locale_time = _get_format_regex(fmt)
+
+    if infer_reso:
+        abbrev = "ns"
+    else:
+        abbrev = npy_unit_to_abbrev(creso)
+    result = np.empty(n, dtype=f"M8[{abbrev}]")
+    iresult = result.view("i8")
+
+    dts.us = dts.ps = dts.as = 0
+
+    for i in range(n):
+        val = values[i]
+        try:
+            if isinstance(val, str):
+                if len(val) == 0 or val in nat_strings:
+                    iresult[i] = NPY_NAT
+                    continue
+            elif checknull_with_nat_and_na(val):
+                iresult[i] = NPY_NAT
+                continue
+            elif PyDateTime_Check(val):
+                if isinstance(val, _Timestamp):
+                    item_reso = val._creso
+                else:
+                    item_reso = NPY_DATETIMEUNIT.NPY_FR_us
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                tz_out = state.process_datetime(val, tz_out, utc)
+                iresult[i] = parse_pydatetime(val, &dts, state.creso)
+                continue
+            elif PyDate_Check(val):
+                state.found_other = True
+                item_reso = NPY_DATETIMEUNIT.NPY_FR_s
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                iresult[i] = pydate_to_dt64(val, &dts, reso=creso)
+                continue
+            elif cnp.is_datetime64_object(val):
+                state.found_other = True
+                item_reso = get_supported_reso(get_datetime64_unit(val))
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                iresult[i] = get_datetime64_nanos(val, creso)
+                continue
+            elif (
+                    (is_integer_object(val) or is_float_object(val))
+                    and (val != val or val == NPY_NAT)
+            ):
+                iresult[i] = NPY_NAT
+                continue
+            else:
+                val = str(val)
+
+            if fmt == "ISO8601":
+                string_to_dts_succeeded = not string_to_dts(
+                    val, &dts, &out_bestunit, &out_local,
+                    &out_tzoffset, False, None, False
+                )
+            elif iso_format:
+                string_to_dts_succeeded = not string_to_dts(
+                    val, &dts, &out_bestunit, &out_local,
+                    &out_tzoffset, False, fmt, exact
+                )
+            if string_to_dts_succeeded:
+                # No error reported by string_to_dts, pick back up
+                # where we left off
+                item_reso = get_supported_reso(out_bestunit)
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                try:
+                    value = npy_datetimestruct_to_datetime(creso, &dts)
+                except OverflowError as err:
+                    attrname = npy_unit_to_attrname[creso]
+                    raise OutOfBoundsDatetime(
+                        f"Out of bounds {attrname} timestamp: {val}"
+                    ) from err
+                if out_local == 1:
+                    nsecs = out_tzoffset * 60
+                    state.out_tzoffset_vals.add(nsecs)
+                    state.found_aware_str = True
+                    tz = timezone(timedelta(minutes=out_tzoffset))
+                    value = tz_localize_to_utc_single(
+                        value, tz, ambiguous="raise", nonexistent=None, creso=creso
+                    )
+                else:
+                    tz = None
+                    state.out_tzoffset_vals.add("naive")
+                    state.found_naive_str = True
+                iresult[i] = value
+                continue
+
+            if parse_today_now(val, &iresult[i], utc, creso, infer_reso=infer_reso):
+                item_reso = NPY_DATETIMEUNIT.NPY_FR_us
+                state.update_creso(item_reso)
+                if infer_reso:
+                    creso = state.creso
+                continue
+
+            # Some ISO formats can't be parsed by string_to_dts
+            # For example, 6-digit YYYYMD. So, if there's an error, and a format
+            # was specified, then try the string-matching code below. If the format
+            # specified was 'ISO8601', then we need to error, because
+            # only string_to_dts handles mixed ISO8601 formats.
+            if not string_to_dts_succeeded and fmt == "ISO8601":
+                raise ValueError(f"Time data {val} is not ISO8601 format")
+
+            tz = _parse_with_format(
+                val, fmt, exact, format_regex, locale_time, &dts, &item_reso
+            )
+
+            state.update_creso(item_reso)
+            if infer_reso:
+                creso = state.creso
+
+            try:
+                iresult[i] = npy_datetimestruct_to_datetime(creso, &dts)
+            except OverflowError as err:
+                attrname = npy_unit_to_attrname[creso]
+                raise OutOfBoundsDatetime(
+                    f"Out of bounds {attrname} timestamp: {val}"
+                ) from err
+
+            if tz is not None:
+                ival = iresult[i]
+                iresult[i] = tz_localize_to_utc_single(
+                    ival, tz, ambiguous="raise", nonexistent=None, creso=creso
+                )
+                nsecs = (ival - iresult[i])
+                if creso == NPY_FR_ns:
+                    nsecs = nsecs // 10**9
+                elif creso == NPY_DATETIMEUNIT.NPY_FR_us:
+                    nsecs = nsecs // 10**6
+                elif creso == NPY_DATETIMEUNIT.NPY_FR_ms:
+                    nsecs = nsecs // 10**3
+
+                state.out_tzoffset_vals.add(nsecs)
+                state.found_aware_str = True
+            else:
+                state.found_naive_str = True
+                tz = None
+                state.out_tzoffset_vals.add("naive")
+
+        except ValueError as ex:
+            ex.args = (
+                f"{str(ex)}. You might want to try:\n"
+                "    - passing `format` if your strings have a consistent format;\n"
+                "    - passing `format='ISO8601'` if your strings are "
+                "all ISO8601 but not necessarily in exactly the same format;\n"
+                "    - passing `format='mixed'`, and the format will be "
+                "inferred for each element individually. "
+                "You might want to use `dayfirst` alongside this.",
+            )
+            if is_coerce:
+                iresult[i] = NPY_NAT
+                continue
+            elif is_raise:
+                raise
+            return values, None
+
+    tz_out = state.check_for_mixed_inputs(tz_out, utc)
+
+    if infer_reso:
+        if state.creso_ever_changed:
+            # We encountered mismatched resolutions, need to re-parse with
+            #  the correct one.
+            return array_strptime(
+                values,
+                fmt=fmt,
+                exact=exact,
+                errors=errors,
+                utc=utc,
+                creso=state.creso,
+            )
+        elif state.creso == NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+            # i.e. we never encountered anything non-NaT, default to "s". This
+            # ensures that insert and concat-like operations with NaT
+            # do not upcast units
+            result = iresult.base.view("M8[s]")
+        else:
+            # Otherwise we can use the single reso that we encountered and avoid
+            #  a second pass.
+            abbrev = npy_unit_to_abbrev(state.creso)
+            result = iresult.base.view(f"M8[{abbrev}]")
+    return result, tz_out
+
+
+cdef tzinfo _parse_with_format(
+    str val,
+    str fmt,
+    bint exact,
+    format_regex,
+    locale_time,
+    npy_datetimestruct* dts,
+    NPY_DATETIMEUNIT* item_reso,
+):
+    # Based on https://github.com/python/cpython/blob/main/Lib/_strptime.py#L293
+    cdef:
+        int year, month, day, minute, hour, second, weekday, julian
+        int week_of_year, week_of_year_start, parse_code, ordinal
+        int iso_week, iso_year
+        int64_t us, ns
+        object found
+        tzinfo tz
+        dict found_dict
+        str group_key, ampm
+
+    if exact:
+        # exact matching
+        found = format_regex.match(val)
+        if not found:
+            raise ValueError(
+                f"time data \"{val}\" doesn't match format \"{fmt}\""
+            )
+        if len(val) != found.end():
+            raise ValueError(
+                "unconverted data remains when parsing with "
+                f"format \"{fmt}\": \"{val[found.end():]}\""
+            )
+
+    else:
+        # search
+        found = format_regex.search(val)
+        if not found:
+            raise ValueError(
+                f"time data \"{val}\" doesn't match format \"{fmt}\""
+            )
+
+    item_reso[0] = NPY_DATETIMEUNIT.NPY_FR_s
+
+    iso_year = -1
+    year = 1900
+    month = day = 1
+    hour = minute = second = ns = us = 0
+    tz = None
+    # Default to -1 to signify that values not known; not critical to have,
+    # though
+    iso_week = week_of_year = -1
+    week_of_year_start = -1
+    # weekday and julian defaulted to -1 so as to signal need to calculate
+    # values
+    weekday = julian = -1
+    found_dict = found.groupdict()
+    for group_key in found_dict.iterkeys():
+        # Directives not explicitly handled below:
+        #   c, x, X
+        #      handled by making out of other directives
+        #   U, W
+        #      worthless without day of the week
+        parse_code = _parse_code_table[group_key]
+
+        if parse_code == 0:
+            year = int(found_dict["y"])
+            # Open Group specification for strptime() states that a %y
+            # value in the range of [00, 68] is in the century 2000, while
+            # [69,99] is in the century 1900
+            if year <= 68:
+                # e.g. val='May 04'; fmt='%b %y'
+                year += 2000
+            else:
+                year += 1900
+                # TODO: not reached in tests 2023-10-28
+        elif parse_code == 1:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            year = int(found_dict["Y"])
+        elif parse_code == 2:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            month = int(found_dict["m"])
+        # elif group_key == 'B':
+        elif parse_code == 3:
+            # e.g. val='30/December/2011'; fmt='%d/%B/%Y'
+            month = locale_time.f_month.index(found_dict["B"].lower())
+        # elif group_key == 'b':
+        elif parse_code == 4:
+            # e.g. val='30/Dec/2011 00:00:00'; fmt='%d/%b/%Y %H:%M:%S'
+            month = locale_time.a_month.index(found_dict["b"].lower())
+        # elif group_key == 'd':
+        elif parse_code == 5:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            day = int(found_dict["d"])
+        # elif group_key == 'H':
+        elif parse_code == 6:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            hour = int(found_dict["H"])
+        elif parse_code == 7:
+            hour = int(found_dict["I"])
+            ampm = found_dict.get("p", "").lower()
+            # If there was no AM/PM indicator, we'll treat this like AM
+            if ampm in ("", locale_time.am_pm[0]):
+                # We're in AM so the hour is correct unless we're
+                # looking at 12 midnight.
+                # 12 midnight == 12 AM == hour 0
+                if hour == 12:
+                    hour = 0
+                    # TODO: not reached in tests 2023-10-28; the implicit `else`
+                    #  branch is tested with e.g.
+                    #  val='Tuesday 24 Aug 2021 01:30:48 AM'
+                    #  fmt='%A %d %b %Y %I:%M:%S %p'
+            elif ampm == locale_time.am_pm[1]:
+                # We're in PM so we need to add 12 to the hour unless
+                # we're looking at 12 noon.
+                # 12 noon == 12 PM == hour 12
+                if hour != 12:
+                    # e.g. val='01/10/2010 08:14 PM'; fmt='%m/%d/%Y %I:%M %p'
+                    hour += 12
+                    # TODO: the implicit `else` branch is not tested 2023-10-28
+            # TODO: the implicit `else` branch is not reached 2023-10-28; possible?
+        elif parse_code == 8:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            minute = int(found_dict["M"])
+        elif parse_code == 9:
+            # e.g. val='17-10-2010 07:15:30'; fmt='%d-%m-%Y %H:%M:%S'
+            second = int(found_dict["S"])
+        elif parse_code == 10:
+            # e.g. val='10:10:10.100'; fmt='%H:%M:%S.%f'
+            s = found_dict["f"]
+            if len(s) <= 3:
+                item_reso[0] = NPY_DATETIMEUNIT.NPY_FR_ms
+            elif len(s) <= 6:
+                item_reso[0] = NPY_DATETIMEUNIT.NPY_FR_us
+            else:
+                item_reso[0] = NPY_FR_ns
+            # Pad to always return nanoseconds
+            s += "0" * (9 - len(s))
+            us = int(s)
+            ns = us % 1000
+            us = us // 1000
+        elif parse_code == 11:
+            # e.g val='Tuesday 24 Aug 2021 01:30:48 AM'; fmt='%A %d %b %Y %I:%M:%S %p'
+            weekday = locale_time.f_weekday.index(found_dict["A"].lower())
+        elif parse_code == 12:
+            # e.g. val='Tue 24 Aug 2021 01:30:48 AM'; fmt='%a %d %b %Y %I:%M:%S %p'
+            weekday = locale_time.a_weekday.index(found_dict["a"].lower())
+        elif parse_code == 13:
+            weekday = int(found_dict["w"])
+            if weekday == 0:
+                # e.g. val='2013020'; fmt='%Y%U%w'
+                weekday = 6
+            else:
+                # e.g. val='2009324'; fmt='%Y%W%w'
+                weekday -= 1
+        elif parse_code == 14:
+            # e.g. val='2009164202000'; fmt='%Y%j%H%M%S'
+            julian = int(found_dict["j"])
+        elif parse_code == 15 or parse_code == 16:
+            week_of_year = int(found_dict[group_key])
+            if group_key == "U":
+                # e.g. val='2013020'; fmt='%Y%U%w'
+                # U starts week on Sunday.
+                week_of_year_start = 6
+            else:
+                # e.g. val='2009324'; fmt='%Y%W%w'
+                # W starts week on Monday.
+                week_of_year_start = 0
+        elif parse_code == 17:
+            # e.g. val='2011-12-30T00:00:00.000000UTC'; fmt='%Y-%m-%dT%H:%M:%S.%f%Z'
+            tz = zoneinfo.ZoneInfo(found_dict["Z"])
+        elif parse_code == 19:
+            # e.g. val='March 1, 2018 12:00:00+0400'; fmt='%B %d, %Y %H:%M:%S%z'
+            tz = parse_timezone_directive(found_dict["z"])
+        elif parse_code == 20:
+            # e.g. val='2015-1-7'; fmt='%G-%V-%u'
+            iso_year = int(found_dict["G"])
+        elif parse_code == 21:
+            # e.g. val='2015-1-7'; fmt='%G-%V-%u'
+            iso_week = int(found_dict["V"])
+        elif parse_code == 22:
+            # e.g. val='2015-1-7'; fmt='%G-%V-%u'
+            weekday = int(found_dict["u"])
+            weekday -= 1
+
+    # If we know the wk of the year and what day of that wk, we can figure
+    # out the Julian day of the year.
+    if julian == -1 and weekday != -1:
+        if week_of_year != -1:
+            # e.g. val='2013020'; fmt='%Y%U%w'
+            week_starts_Mon = week_of_year_start == 0
+            julian = _calc_julian_from_U_or_W(year, week_of_year, weekday,
+                                              week_starts_Mon)
+        elif iso_year != -1 and iso_week != -1:
+            # e.g. val='2015-1-7'; fmt='%G-%V-%u'
+            year, julian = _calc_julian_from_V(iso_year, iso_week,
+                                               weekday + 1)
+        # else:
+        #    # e.g. val='Thu Sep 25 2003'; fmt='%a %b %d %Y'
+        #    pass
+
+    # Cannot pre-calculate date() since can change in Julian
+    # calculation and thus could have different value for the day of the wk
+    # calculation.
+    if julian == -1:
+        # Need to add 1 to result since first day of the year is 1, not
+        # 0.
+        # We don't actually need ordinal/julian here, but need to raise
+        #  on e.g. val='2015-04-31'; fmt='%Y-%m-%d'
+        ordinal = date(year, month, day).toordinal()
+        julian = ordinal - date(year, 1, 1).toordinal() + 1
+    else:
+        # Assume that if they bothered to include Julian day it will
+        # be accurate.
+        datetime_result = date.fromordinal(
+            (julian - 1) + date(year, 1, 1).toordinal())
+        year = datetime_result.year
+        month = datetime_result.month
+        day = datetime_result.day
+    if weekday == -1:
+        # We don't actually use weekday here, but need to do this in order to
+        #  raise on y/m/d combinations
+        # TODO: not reached in tests 2023-10-28; necessary?
+        weekday = date(year, month, day).weekday()
+
+    dts.year = year
+    dts.month = month
+    dts.day = day
+    dts.hour = hour
+    dts.min = minute
+    dts.sec = second
+    dts.us = us
+    dts.ps = ns * 1000
+    return tz
+
+
+class TimeRE(_TimeRE):
+    """
+    Handle conversion from format directives to regexes.
+
+    Creates regexes for pattern matching a string of text containing
+    time information
+    """
+
+    def __init__(self, locale_time=None):
+        """
+        Create keys/values.
+
+        Order of execution is important for dependency reasons.
+        """
+        self._Z = None
+        super().__init__(locale_time=locale_time)
+        # GH 48767: Overrides for cpython's TimeRE
+        #  1) Parse up to nanos instead of micros
+        self.update({"f": r"(?P<f>[0-9]{1,9})"}),
+
+    def __getitem__(self, key):
+        if key == "Z":
+            # lazy computation
+            if self._Z is None:
+                self._Z = self.__seqToRE(zoneinfo.available_timezones(), "Z")
+            # Note: handling Z is the key difference vs using the stdlib
+            # _strptime.TimeRE. test_to_datetime_parse_tzname_or_tzoffset with
+            # fmt='%Y-%m-%d %H:%M:%S %Z' fails with the stdlib version.
+            return self._Z
+        return super().__getitem__(key)
+
+
+_cache_lock = _thread_allocate_lock()
+# DO NOT modify _TimeRE_cache or _regex_cache without acquiring the cache lock
+# first!
+_TimeRE_cache = TimeRE()
+_CACHE_MAX_SIZE = 5  # Max number of regexes stored in _regex_cache
+_regex_cache = {}
+
+
+cdef int _calc_julian_from_U_or_W(int year, int week_of_year,
+                                  int day_of_week, int week_starts_Mon):
+    """
+    Calculate the Julian day based on the year, week of the year, and day of
+    the week, with week_start_day representing whether the week of the year
+    assumes the week starts on Sunday or Monday (6 or 0).
+
+    Parameters
+    ----------
+    year : int
+        the year
+    week_of_year : int
+        week taken from format U or W
+    week_starts_Mon : int
+        represents whether the week of the year
+        assumes the week starts on Sunday or Monday (6 or 0)
+
+    Returns
+    -------
+    int
+        converted julian day
+    """
+
+    cdef:
+        int first_weekday, week_0_length, days_to_week
+
+    first_weekday = date(year, 1, 1).weekday()
+    # If we are dealing with the %U directive (week starts on Sunday), it's
+    # easier to just shift the view to Sunday being the first day of the
+    # week.
+    if not week_starts_Mon:
+        first_weekday = (first_weekday + 1) % 7
+        day_of_week = (day_of_week + 1) % 7
+
+    # Need to watch out for a week 0 (when the first day of the year is not
+    # the same as that specified by %U or %W).
+    week_0_length = (7 - first_weekday) % 7
+    if week_of_year == 0:
+        return 1 + day_of_week - first_weekday
+    else:
+        days_to_week = week_0_length + (7 * (week_of_year - 1))
+        return 1 + days_to_week + day_of_week
+
+
+cdef (int, int) _calc_julian_from_V(int iso_year, int iso_week, int iso_weekday):
+    """
+    Calculate the Julian day based on the ISO 8601 year, week, and weekday.
+
+    ISO weeks start on Mondays, with week 01 being the week containing 4 Jan.
+    ISO week days range from 1 (Monday) to 7 (Sunday).
+
+    Parameters
+    ----------
+    iso_year : int
+        the year taken from format %G
+    iso_week : int
+        the week taken from format %V
+    iso_weekday : int
+        weekday taken from format %u
+
+    Returns
+    -------
+    (int, int)
+        the iso year and the Gregorian ordinal date / julian date
+    """
+
+    cdef:
+        int correction, ordinal
+
+    correction = date(iso_year, 1, 4).isoweekday() + 3
+    ordinal = (iso_week * 7) + iso_weekday - correction
+
+    if iso_week == 53:
+        now = date.fromordinal(date(iso_year, 1, 1).toordinal() + ordinal - iso_weekday)
+        jan_4th = date(iso_year+1, 1, 4)
+        if (jan_4th - now).days < 7:
+            raise ValueError(f"Week 53 does not exist in ISO year {iso_year}.")
+
+    # ordinal may be negative or 0 now, which means the date is in the previous
+    # calendar year
+    if ordinal < 1:
+        ordinal += date(iso_year, 1, 1).toordinal()
+        iso_year -= 1
+        ordinal -= date(iso_year, 1, 1).toordinal()
+    return iso_year, ordinal
+
+
+cdef tzinfo parse_timezone_directive(str z):
+    """
+    Parse the '%z' directive and return a datetime.timezone object.
+
+    Parameters
+    ----------
+    z : string of the UTC offset
+
+    Returns
+    -------
+    datetime.timezone
+
+    Notes
+    -----
+    This is essentially similar to the cpython implementation
+    https://github.com/python/cpython/blob/546cab84448b892c92e68d9c1a3d3b58c13b3463/Lib/_strptime.py#L437-L454
+    Licence at LICENSES/PSF_LICENSE
+    """
+
+    cdef:
+        int hours, minutes, seconds, pad_number, microseconds
+        int total_minutes
+        str gmtoff_remainder, gmtoff_remainder_padding
+
+    if z == "Z":
+        return timezone(timedelta(0))
+    if z[3] == ":":
+        z = z[:3] + z[4:]
+        if len(z) > 5:
+            if z[5] != ":":
+                raise ValueError(f"Inconsistent use of : in {z}")
+            z = z[:5] + z[6:]
+    hours = int(z[1:3])
+    minutes = int(z[3:5])
+    seconds = int(z[5:7] or 0)
+
+    # Pad to always return microseconds.
+    gmtoff_remainder = z[8:]
+    pad_number = 6 - len(gmtoff_remainder)
+    gmtoff_remainder_padding = "0" * pad_number
+    microseconds = int(gmtoff_remainder + gmtoff_remainder_padding)
+
+    total_minutes = ((hours * 60) + minutes + (seconds // 60) +
+                     (microseconds // 60_000_000))
+    total_minutes = -total_minutes if z.startswith("-") else total_minutes
+    return timezone(timedelta(minutes=total_minutes))
diff --git a/pandas/pandas/_libs/tslibs/timedeltas.pxd b/pandas/pandas/_libs/tslibs/timedeltas.pxd
new file mode 100644
index 0000000000000..f3473e46b6699
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timedeltas.pxd
@@ -0,0 +1,29 @@
+from cpython.datetime cimport timedelta
+from numpy cimport int64_t
+
+from .np_datetime cimport NPY_DATETIMEUNIT
+
+
+cpdef int64_t get_unit_for_round(freq, NPY_DATETIMEUNIT creso) except? -1
+# Exposed for tslib, not intended for outside use.
+cpdef int64_t delta_to_nanoseconds(
+    delta, NPY_DATETIMEUNIT reso=*, bint round_ok=*
+) except? -1
+cdef convert_to_timedelta64(object ts, str unit)
+cdef bint is_any_td_scalar(object obj)
+
+
+cdef class _Timedelta(timedelta):
+    cdef readonly:
+        int64_t _value      # nanoseconds
+        bint _is_populated  # are my components populated
+        int64_t _d, _h, _m, _s, _ms, _us, _ns
+        NPY_DATETIMEUNIT _creso
+
+    cpdef timedelta to_pytimedelta(_Timedelta self)
+    cdef bint _has_ns(self)
+    cdef bint _is_in_pytimedelta_bounds(self)
+    cdef _ensure_components(_Timedelta self)
+    cdef bint _compare_mismatched_resos(self, _Timedelta other, op)
+    cdef _Timedelta _as_creso(self, NPY_DATETIMEUNIT reso, bint round_ok=*)
+    cpdef _maybe_cast_to_matching_resos(self, _Timedelta other)
diff --git a/pandas/pandas/_libs/tslibs/timedeltas.pyi b/pandas/pandas/_libs/tslibs/timedeltas.pyi
new file mode 100644
index 0000000000000..979a5666661b2
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timedeltas.pyi
@@ -0,0 +1,168 @@
+from datetime import timedelta
+from typing import (
+    ClassVar,
+    Literal,
+    TypeAlias,
+    TypeVar,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import (
+    NaTType,
+    Tick,
+)
+from pandas._typing import (
+    Frequency,
+    Self,
+    npt,
+)
+
+# This should be kept consistent with the keys in the dict timedelta_abbrevs
+# in pandas/_libs/tslibs/timedeltas.pyx
+UnitChoices: TypeAlias = Literal[
+    "Y",
+    "y",
+    "M",
+    "W",
+    "w",
+    "D",
+    "d",
+    "days",
+    "day",
+    "hours",
+    "hour",
+    "hr",
+    "h",
+    "m",
+    "minute",
+    "min",
+    "minutes",
+    "s",
+    "seconds",
+    "sec",
+    "second",
+    "ms",
+    "milliseconds",
+    "millisecond",
+    "milli",
+    "millis",
+    "us",
+    "microseconds",
+    "microsecond",
+    "µs",
+    "micro",
+    "micros",
+    "ns",
+    "nanoseconds",
+    "nano",
+    "nanos",
+    "nanosecond",
+]
+_S = TypeVar("_S", bound=timedelta)
+
+def get_unit_for_round(freq, creso: int) -> int: ...
+def disallow_ambiguous_unit(unit: str | None) -> None: ...
+def ints_to_pytimedelta(
+    m8values: npt.NDArray[np.timedelta64],
+    box: bool = ...,
+) -> npt.NDArray[np.object_]: ...
+def array_to_timedelta64(
+    values: npt.NDArray[np.object_],
+    unit: str | None = ...,
+    errors: str = ...,
+) -> np.ndarray: ...  # np.ndarray[m8ns]
+def parse_timedelta_unit(unit: str | None) -> UnitChoices: ...
+def delta_to_nanoseconds(
+    delta: np.timedelta64 | timedelta | Tick,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+    round_ok: bool = ...,
+) -> int: ...
+def floordiv_object_array(
+    left: np.ndarray, right: npt.NDArray[np.object_]
+) -> np.ndarray: ...
+def truediv_object_array(
+    left: np.ndarray, right: npt.NDArray[np.object_]
+) -> np.ndarray: ...
+
+class Timedelta(timedelta):
+    _creso: int
+    min: ClassVar[Timedelta]
+    max: ClassVar[Timedelta]
+    resolution: ClassVar[Timedelta]
+    value: int  # np.int64
+    _value: int  # np.int64
+    # error: "__new__" must return a class instance (got "Union[Timestamp, NaTType]")
+    def __new__(  # type: ignore[misc]
+        cls: type[_S],
+        value=...,
+        unit: str | None = ...,
+        **kwargs: float | np.integer | np.floating,
+    ) -> _S | NaTType: ...
+    @classmethod
+    def _from_value_and_reso(cls, value: np.int64, reso: int) -> Timedelta: ...
+    @property
+    def days(self) -> int: ...
+    @property
+    def seconds(self) -> int: ...
+    @property
+    def microseconds(self) -> int: ...
+    def total_seconds(self) -> float: ...
+    def to_pytimedelta(self) -> timedelta: ...
+    def to_timedelta64(self) -> np.timedelta64: ...
+    @property
+    def asm8(self) -> np.timedelta64: ...
+    # TODO: round/floor/ceil could return NaT?
+    def round(self, freq: Frequency) -> Self: ...
+    def floor(self, freq: Frequency) -> Self: ...
+    def ceil(self, freq: Frequency) -> Self: ...
+    @property
+    def resolution_string(self) -> str: ...
+    def __add__(self, other: timedelta) -> Timedelta: ...
+    def __radd__(self, other: timedelta) -> Timedelta: ...
+    def __sub__(self, other: timedelta) -> Timedelta: ...
+    def __rsub__(self, other: timedelta) -> Timedelta: ...
+    def __neg__(self) -> Timedelta: ...
+    def __pos__(self) -> Timedelta: ...
+    def __abs__(self) -> Timedelta: ...
+    def __mul__(self, other: float) -> Timedelta: ...
+    def __rmul__(self, other: float) -> Timedelta: ...
+    # error: Signature of "__floordiv__" incompatible with supertype "timedelta"
+    @overload  # type: ignore[override]
+    def __floordiv__(self, other: timedelta) -> int: ...
+    @overload
+    def __floordiv__(self, other: float) -> Timedelta: ...
+    @overload
+    def __floordiv__(
+        self, other: npt.NDArray[np.timedelta64]
+    ) -> npt.NDArray[np.intp]: ...
+    @overload
+    def __floordiv__(
+        self, other: npt.NDArray[np.number]
+    ) -> npt.NDArray[np.timedelta64] | Timedelta: ...
+    @overload
+    def __rfloordiv__(self, other: timedelta | str) -> int: ...
+    @overload
+    def __rfloordiv__(self, other: None | NaTType) -> NaTType: ...
+    @overload
+    def __rfloordiv__(self, other: np.ndarray) -> npt.NDArray[np.timedelta64]: ...
+    @overload
+    def __truediv__(self, other: timedelta) -> float: ...
+    @overload
+    def __truediv__(self, other: float) -> Timedelta: ...
+    def __mod__(self, other: timedelta) -> Timedelta: ...
+    def __divmod__(self, other: timedelta) -> tuple[int, Timedelta]: ...
+    def __le__(self, other: timedelta) -> bool: ...
+    def __lt__(self, other: timedelta) -> bool: ...
+    def __ge__(self, other: timedelta) -> bool: ...
+    def __gt__(self, other: timedelta) -> bool: ...
+    def __hash__(self) -> int: ...
+    def isoformat(self) -> str: ...
+    def to_numpy(
+        self, dtype: npt.DTypeLike = ..., copy: bool = False
+    ) -> np.timedelta64: ...
+    def view(self, dtype: npt.DTypeLike) -> object: ...
+    @property
+    def unit(self) -> str: ...
+    def as_unit(self, unit: str, round_ok: bool = ...) -> Timedelta: ...
diff --git a/pandas/pandas/_libs/tslibs/timedeltas.pyx b/pandas/pandas/_libs/tslibs/timedeltas.pyx
new file mode 100644
index 0000000000000..e320aca04683c
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timedeltas.pyx
@@ -0,0 +1,2504 @@
+import collections
+import warnings
+
+from pandas.util._decorators import set_module
+from pandas.util._exceptions import find_stack_level
+
+cimport cython
+from cpython.object cimport (
+    Py_EQ,
+    Py_GE,
+    Py_GT,
+    Py_LE,
+    Py_LT,
+    Py_NE,
+    PyObject,
+    PyObject_RichCompare,
+)
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+from cpython.datetime cimport (
+    PyDateTime_Check,
+    PyDelta_Check,
+    import_datetime,
+    timedelta,
+)
+
+import_datetime()
+
+
+cimport pandas._libs.tslibs.util as util
+from pandas._libs.missing cimport checknull_with_nat_and_na
+from pandas._libs.tslibs.base cimport ABCTimestamp
+from pandas._libs.tslibs.conversion cimport (
+    cast_from_unit,
+    precision_from_unit,
+)
+from pandas._libs.tslibs.dtypes cimport (
+    c_DEPR_UNITS,
+    get_supported_reso,
+    is_supported_unit,
+    npy_unit_to_abbrev,
+)
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+    c_nat_strings as nat_strings,
+    checknull_with_nat,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    cmp_dtstructs,
+    cmp_scalar,
+    convert_reso,
+    get_datetime64_unit,
+    get_unit_from_dtype,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    pandas_datetime_to_datetimestruct,
+    pandas_timedelta_to_timedeltastruct,
+    pandas_timedeltastruct,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.np_datetime import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+)
+
+from pandas._libs.tslibs.offsets cimport is_tick_object
+from pandas._libs.tslibs.util cimport (
+    is_array,
+    is_float_object,
+    is_integer_object,
+)
+
+from pandas._libs.tslibs.fields import (
+    RoundTo,
+    round_nsint64,
+)
+
+# ----------------------------------------------------------------------
+# Constants
+
+# components named tuple
+Components = collections.namedtuple(
+    "Components",
+    [
+        "days",
+        "hours",
+        "minutes",
+        "seconds",
+        "milliseconds",
+        "microseconds",
+        "nanoseconds",
+    ],
+)
+
+# This should be kept consistent with UnitChoices in pandas/_libs/tslibs/timedeltas.pyi
+cdef dict timedelta_abbrevs = {
+    "Y": "Y",
+    "y": "Y",
+    "M": "M",
+    "W": "W",
+    "w": "W",
+    "D": "D",
+    "d": "D",
+    "days": "D",
+    "day": "D",
+    "hours": "h",
+    "hour": "h",
+    "hr": "h",
+    "h": "h",
+    "m": "m",
+    "minute": "m",
+    "min": "m",
+    "minutes": "m",
+    "s": "s",
+    "seconds": "s",
+    "sec": "s",
+    "second": "s",
+    "ms": "ms",
+    "milliseconds": "ms",
+    "millisecond": "ms",
+    "milli": "ms",
+    "millis": "ms",
+    "us": "us",
+    "microseconds": "us",
+    "microsecond": "us",
+    "µs": "us",
+    "micro": "us",
+    "micros": "us",
+    "ns": "ns",
+    "nanoseconds": "ns",
+    "nano": "ns",
+    "nanos": "ns",
+    "nanosecond": "ns",
+}
+
+_no_input = object()
+
+# ----------------------------------------------------------------------
+# API
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def ints_to_pytimedelta(ndarray m8values, box=False):
+    """
+    convert an i8 repr to an ndarray of timedelta or Timedelta (if box ==
+    True)
+
+    Parameters
+    ----------
+    arr : ndarray[timedelta64]
+    box : bool, default False
+
+    Returns
+    -------
+    result : ndarray[object]
+        array of Timedelta or timedeltas objects
+    """
+    cdef:
+        NPY_DATETIMEUNIT reso = get_unit_from_dtype(m8values.dtype)
+        Py_ssize_t i, n = m8values.size
+        int64_t value
+        object res_val
+
+        # Note that `result` (and thus `result_flat`) is C-order and
+        #  `it` iterates C-order as well, so the iteration matches
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        ndarray result = cnp.PyArray_EMPTY(
+            m8values.ndim, m8values.shape, cnp.NPY_OBJECT, 0
+        )
+        object[::1] res_flat = result.ravel()     # should NOT be a copy
+
+        ndarray arr = m8values.view("i8")
+        cnp.flatiter it = cnp.PyArray_IterNew(arr)
+
+    for i in range(n):
+        # Analogous to: value = arr[i]
+        value = (<int64_t*>cnp.PyArray_ITER_DATA(it))[0]
+
+        if value == NPY_NAT:
+            res_val = <object>NaT
+        else:
+            if box:
+                res_val = _timedelta_from_value_and_reso(Timedelta, value, reso=reso)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_ns:
+                res_val = timedelta(microseconds=int(value) / 1000)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_us:
+                res_val = timedelta(microseconds=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_ms:
+                res_val = timedelta(milliseconds=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_s:
+                res_val = timedelta(seconds=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_m:
+                res_val = timedelta(minutes=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_h:
+                res_val = timedelta(hours=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_D:
+                res_val = timedelta(days=value)
+            elif reso == NPY_DATETIMEUNIT.NPY_FR_W:
+                res_val = timedelta(weeks=value)
+            else:
+                # Month, Year, NPY_FR_GENERIC, pico, femto, atto
+                raise NotImplementedError(reso)
+
+        # Note: we can index result directly instead of using PyArray_MultiIter_DATA
+        #  like we do for the other functions because result is known C-contiguous
+        #  and is the first argument to PyArray_MultiIterNew2.  The usual pattern
+        #  does not seem to work with object dtype.
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        res_flat[i] = res_val
+
+        cnp.PyArray_ITER_NEXT(it)
+
+    return result
+
+
+# ----------------------------------------------------------------------
+
+
+cpdef int64_t delta_to_nanoseconds(
+    delta,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+    bint round_ok=True,
+) except? -1:
+    # Note: this will raise on timedelta64 with Y or M unit
+
+    cdef:
+        NPY_DATETIMEUNIT in_reso
+        int64_t n
+
+    if is_tick_object(delta):
+        n = delta.n
+        in_reso = delta._creso
+
+    elif isinstance(delta, _Timedelta):
+        n = delta._value
+        in_reso = delta._creso
+
+    elif cnp.is_timedelta64_object(delta):
+        in_reso = get_datetime64_unit(delta)
+        if in_reso == NPY_DATETIMEUNIT.NPY_FR_Y or in_reso == NPY_DATETIMEUNIT.NPY_FR_M:
+            raise ValueError(
+                "delta_to_nanoseconds does not support Y or M units, "
+                "as their duration in nanoseconds is ambiguous."
+            )
+        n = cnp.get_timedelta64_value(delta)
+
+    elif PyDelta_Check(delta):
+        in_reso = NPY_DATETIMEUNIT.NPY_FR_us
+        try:
+            n = (
+                delta.days * 24 * 3600 * 1_000_000
+                + delta.seconds * 1_000_000
+                + delta.microseconds
+                )
+        except OverflowError as err:
+            raise OutOfBoundsTimedelta(*err.args) from err
+
+    else:
+        raise TypeError(type(delta))
+
+    try:
+        return convert_reso(n, in_reso, reso, round_ok=round_ok)
+    except (OutOfBoundsDatetime, OverflowError) as err:
+        # Catch OutOfBoundsDatetime bc convert_reso can call check_dts_bounds
+        #  for Y/M-resolution cases
+        unit_str = npy_unit_to_abbrev(reso)
+        raise OutOfBoundsTimedelta(
+            f"Cannot cast {str(delta)} to unit={unit_str} without overflow."
+        ) from err
+
+
+@cython.overflowcheck(True)
+cdef object ensure_td64ns(object ts):
+    """
+    Overflow-safe implementation of td64.astype("m8[ns]")
+
+    Parameters
+    ----------
+    ts : np.timedelta64
+
+    Returns
+    -------
+    np.timedelta64[ns]
+    """
+    cdef:
+        NPY_DATETIMEUNIT td64_unit
+        int64_t td64_value, mult
+
+    td64_unit = get_datetime64_unit(ts)
+    if (
+        td64_unit != NPY_DATETIMEUNIT.NPY_FR_ns
+        and td64_unit != NPY_DATETIMEUNIT.NPY_FR_GENERIC
+    ):
+
+        td64_value = cnp.get_timedelta64_value(ts)
+
+        mult = precision_from_unit(td64_unit)[0]
+        try:
+            # NB: cython#1381 this cannot be *=
+            td64_value = td64_value * mult
+        except OverflowError as err:
+            raise OutOfBoundsTimedelta(ts) from err
+
+        return np.timedelta64(td64_value, "ns")
+
+    return ts
+
+
+cdef convert_to_timedelta64(object ts, str unit):
+    """
+    Convert an incoming object to a timedelta64 if possible.
+    Before calling, unit must be standardized to avoid repeated unit conversion
+
+    Handle these types of objects:
+        - timedelta/Timedelta
+        - timedelta64
+        - an offset
+        - np.int64 (with unit providing a possible modifier)
+        - None/NaT
+
+    Return an ns based int64
+    """
+    # Caller is responsible for checking unit not in ["Y", "y", "M"]
+    if checknull_with_nat_and_na(ts):
+        return np.timedelta64(NPY_NAT, "ns")
+    elif isinstance(ts, _Timedelta):
+        # already in the proper format
+        if ts._creso != NPY_FR_ns:
+            ts = ts.as_unit("ns").asm8
+        else:
+            ts = np.timedelta64(ts._value, "ns")
+    elif cnp.is_timedelta64_object(ts):
+        ts = ensure_td64ns(ts)
+    elif is_integer_object(ts):
+        if ts == NPY_NAT:
+            return np.timedelta64(NPY_NAT, "ns")
+        else:
+            ts = _maybe_cast_from_unit(ts, unit)
+    elif is_float_object(ts):
+        ts = _maybe_cast_from_unit(ts, unit)
+    elif isinstance(ts, str):
+        if (len(ts) > 0 and ts[0] == "P") or (len(ts) > 1 and ts[:2] == "-P"):
+            ts = parse_iso_format_string(ts)
+        else:
+            ts = parse_timedelta_string(ts)
+        ts = np.timedelta64(ts, "ns")
+    elif is_tick_object(ts):
+        ts = np.timedelta64(ts.nanos, "ns")
+
+    if PyDelta_Check(ts):
+        ts = np.timedelta64(delta_to_nanoseconds(ts), "ns")
+    elif not cnp.is_timedelta64_object(ts):
+        raise TypeError(f"Invalid type for timedelta scalar: {type(ts)}")
+    return ts.astype("timedelta64[ns]")
+
+
+cdef _maybe_cast_from_unit(ts, str unit):
+    # caller is responsible for checking
+    #  assert unit not in ["Y", "y", "M"]
+    try:
+        ts = cast_from_unit(ts, unit)
+    except OutOfBoundsDatetime as err:
+        raise OutOfBoundsTimedelta(
+            f"Cannot cast {ts} from {unit} to 'ns' without overflow."
+        ) from err
+
+    ts = np.timedelta64(ts, "ns")
+    return ts
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def array_to_timedelta64(
+    ndarray values, str unit=None, str errors="raise"
+) -> ndarray:
+    # values is object-dtype, may be 2D
+    """
+    Convert an ndarray to an array of timedeltas. If errors == 'coerce',
+    coerce non-convertible objects to NaT. Otherwise, raise.
+
+    Returns
+    -------
+    np.ndarray[timedelta64ns]
+    """
+    # Caller is responsible for checking
+    assert unit not in ["Y", "y", "M"]
+
+    cdef:
+        Py_ssize_t i, n = values.size
+        ndarray result = np.empty((<object>values).shape, dtype="m8[ns]")
+        object item
+        int64_t ival
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(result, values)
+        cnp.flatiter it
+
+    if values.descr.type_num != cnp.NPY_OBJECT:
+        # raise here otherwise we segfault below
+        raise TypeError("array_to_timedelta64 'values' must have object dtype")
+
+    if errors not in {"ignore", "raise", "coerce"}:
+        raise ValueError("errors must be one of {'ignore', 'raise', or 'coerce'}")
+
+    if unit is not None and errors != "coerce":
+        it = cnp.PyArray_IterNew(values)
+        for i in range(n):
+            # Analogous to: item = values[i]
+            item = cnp.PyArray_GETITEM(values, cnp.PyArray_ITER_DATA(it))
+            if isinstance(item, str):
+                raise ValueError(
+                    "unit must not be specified if the input contains a str"
+                )
+            cnp.PyArray_ITER_NEXT(it)
+
+    # Usually, we have all strings. If so, we hit the fast path.
+    # If this path fails, we try conversion a different way, and
+    # this is where all of the error handling will take place.
+    try:
+        for i in range(n):
+            # Analogous to: item = values[i]
+            item = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+            ival = _item_to_timedelta64_fastpath(item)
+
+            # Analogous to: iresult[i] = ival
+            (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = ival
+
+            cnp.PyArray_MultiIter_NEXT(mi)
+
+    except (TypeError, ValueError):
+        cnp.PyArray_MultiIter_RESET(mi)
+
+        parsed_unit = parse_timedelta_unit(unit or "ns")
+        for i in range(n):
+            item = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+            ival = _item_to_timedelta64(item, parsed_unit, errors)
+
+            (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = ival
+
+            cnp.PyArray_MultiIter_NEXT(mi)
+
+    return result
+
+
+cdef int64_t _item_to_timedelta64_fastpath(object item) except? -1:
+    """
+    See array_to_timedelta64.
+    """
+    if item is NaT:
+        # we allow this check in the fast-path because NaT is a C-object
+        #  so this is an inexpensive check
+        return NPY_NAT
+    else:
+        return parse_timedelta_string(item)
+
+
+cdef int64_t _item_to_timedelta64(
+    object item,
+    str parsed_unit,
+    str errors
+) except? -1:
+    """
+    See array_to_timedelta64.
+    """
+    try:
+        return cnp.get_timedelta64_value(convert_to_timedelta64(item, parsed_unit))
+    except ValueError as err:
+        if errors == "coerce":
+            return NPY_NAT
+        elif "unit abbreviation w/o a number" in str(err):
+            # re-raise with more pertinent message
+            msg = f"Could not convert '{item}' to NumPy timedelta"
+            raise ValueError(msg) from err
+        else:
+            raise
+
+
+@cython.cpow(True)
+cdef int64_t parse_timedelta_string(str ts) except? -1:
+    """
+    Parse a regular format timedelta string. Return an int64_t (in ns)
+    or raise a ValueError on an invalid parse.
+    """
+
+    cdef:
+        str c
+        bint neg = 0, have_dot = 0, have_value = 0, have_hhmmss = 0
+        str current_unit = None
+        int64_t result = 0, m = 0, r
+        list number = [], frac = [], unit = []
+
+    # neg : tracks if we have a leading negative for the value
+    # have_dot : tracks if we are processing a dot (either post hhmmss or
+    #            inside an expression)
+    # have_value : track if we have at least 1 leading unit
+    # have_hhmmss : tracks if we have a regular format hh:mm:ss
+
+    if len(ts) == 0 or ts in nat_strings:
+        return NPY_NAT
+
+    for c in ts:
+
+        # skip whitespace / commas
+        if c == " " or c == ",":
+            pass
+
+        # positive signs are ignored
+        elif c == "+":
+            pass
+
+        # neg
+        elif c == "-":
+
+            if neg or have_value or have_hhmmss:
+                raise ValueError("only leading negative signs are allowed")
+
+            neg = 1
+
+        # number (ascii codes)
+        elif ord(c) >= 48 and ord(c) <= 57:
+
+            if have_dot:
+
+                # we found a dot, but now its just a fraction
+                if len(unit):
+                    number.append(c)
+                    have_dot = 0
+                else:
+                    frac.append(c)
+
+            elif not len(unit):
+                number.append(c)
+
+            else:
+                r = timedelta_from_spec(number, frac, unit)
+                unit, number, frac = [], [c], []
+
+                result += timedelta_as_neg(r, neg)
+
+        # hh:mm:ss.
+        elif c == ":":
+
+            # we flip this off if we have a leading value
+            if have_value:
+                neg = 0
+
+            # we are in the pattern hh:mm:ss pattern
+            if len(number):
+                if current_unit is None:
+                    current_unit = "h"
+                    m = 1000000000 * 3600
+                elif current_unit == "h":
+                    current_unit = "m"
+                    m = 1000000000 * 60
+                elif current_unit == "m":
+                    current_unit = "s"
+                    m = 1000000000
+                r = <int64_t>int("".join(number)) * m
+                result += timedelta_as_neg(r, neg)
+                have_hhmmss = 1
+            else:
+                raise ValueError(f"expecting hh:mm:ss format, received: {ts}")
+
+            unit, number = [], []
+
+        # after the decimal point
+        elif c == ".":
+
+            if len(number) and current_unit is not None:
+
+                # by definition we had something like
+                # so we need to evaluate the final field from a
+                # hh:mm:ss (so current_unit is 'm')
+                if current_unit != "m":
+                    raise ValueError("expected hh:mm:ss format before .")
+                m = 1000000000
+                r = <int64_t>int("".join(number)) * m
+                result += timedelta_as_neg(r, neg)
+                have_value = 1
+                unit, number, frac = [], [], []
+
+            have_dot = 1
+
+        # unit
+        else:
+            unit.append(c)
+            have_value = 1
+            have_dot = 0
+
+    # we had a dot, but we have a fractional
+    # value since we have an unit
+    if have_dot and len(unit):
+        r = timedelta_from_spec(number, frac, unit)
+        result += timedelta_as_neg(r, neg)
+
+    # we have a dot as part of a regular format
+    # e.g. hh:mm:ss.fffffff
+    elif have_dot:
+
+        if ((len(number) or len(frac)) and not len(unit)
+                and current_unit is None):
+            raise ValueError("no units specified")
+
+        if len(frac) > 0 and len(frac) <= 3:
+            m = 10**(3 -len(frac)) * 1000 * 1000
+        elif len(frac) > 3 and len(frac) <= 6:
+            m = 10**(6 -len(frac)) * 1000
+        elif len(frac) > 6 and len(frac) <= 9:
+            m = 10**(9 -len(frac))
+        else:
+            m = 1
+            frac = frac[:9]
+        r = <int64_t>int("".join(frac)) * m
+        result += timedelta_as_neg(r, neg)
+
+    # we have a regular format
+    # we must have seconds at this point (hence the unit is still 'm')
+    elif current_unit is not None:
+        if current_unit != "m":
+            raise ValueError("expected hh:mm:ss format")
+        m = 1000000000
+        r = <int64_t>int("".join(number)) * m
+        result += timedelta_as_neg(r, neg)
+
+    # we have a last abbreviation
+    elif len(unit):
+        if len(number):
+            r = timedelta_from_spec(number, frac, unit)
+            result += timedelta_as_neg(r, neg)
+        else:
+            raise ValueError("unit abbreviation w/o a number")
+
+    # we only have symbols and no numbers
+    elif len(number) == 0:
+        raise ValueError("symbols w/o a number")
+
+    # treat as nanoseconds
+    # but only if we don't have anything else
+    else:
+        if have_value:
+            raise ValueError("have leftover units")
+        if len(number):
+            r = timedelta_from_spec(number, frac, "ns")
+            result += timedelta_as_neg(r, neg)
+
+    return result
+
+
+cdef int64_t timedelta_as_neg(int64_t value, bint neg):
+    """
+
+    Parameters
+    ----------
+    value : int64_t of the timedelta value
+    neg : bool if the a negative value
+    """
+    if neg:
+        return -value
+    return value
+
+
+cdef timedelta_from_spec(object number, object frac, object unit):
+    """
+
+    Parameters
+    ----------
+    number : a list of number digits
+    frac : a list of frac digits
+    unit : a list of unit characters
+    """
+    cdef:
+        str n
+
+    unit = "".join(unit)
+    if unit in ["M", "Y", "y"]:
+        raise ValueError(
+            "Units 'M', 'Y' and 'y' do not represent unambiguous timedelta "
+            "values and are not supported."
+        )
+
+    unit = parse_timedelta_unit(unit)
+
+    n = "".join(number) + "." + "".join(frac)
+    return cast_from_unit(float(n), unit)
+
+
+cpdef inline str parse_timedelta_unit(str unit):
+    """
+    Parameters
+    ----------
+    unit : str or None
+
+    Returns
+    -------
+    str
+        Canonical unit string.
+
+    Raises
+    ------
+    ValueError : on non-parseable input
+    """
+    if unit is None:
+        return "ns"
+    elif unit == "M":
+        return unit
+    elif unit in c_DEPR_UNITS:
+        warnings.warn(
+            f"\'{unit}\' is deprecated and will be removed in a "
+            f"future version. Please use \'{c_DEPR_UNITS.get(unit)}\' "
+            f"instead of \'{unit}\'.",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        unit = c_DEPR_UNITS[unit]
+    try:
+        return timedelta_abbrevs[unit.lower()]
+    except KeyError:
+        raise ValueError(f"invalid unit abbreviation: {unit}")
+
+# ----------------------------------------------------------------------
+# Timedelta ops utilities
+
+cdef bint _validate_ops_compat(other):
+    # return True if we are compat with operating
+    if checknull_with_nat(other):
+        return True
+    elif is_any_td_scalar(other):
+        return True
+    elif isinstance(other, str):
+        return True
+    return False
+
+
+def _op_unary_method(func, name):
+    def f(self):
+        new_value = func(self._value)
+        return _timedelta_from_value_and_reso(Timedelta, new_value, self._creso)
+    f.__name__ = name
+    return f
+
+
+def _binary_op_method_timedeltalike(op, name):
+    # define a binary operation that only works if the other argument is
+    # timedelta like or an array of timedeltalike
+    def f(self, other):
+        if other is NaT:
+            return NaT
+
+        elif cnp.is_datetime64_object(other) or (
+            PyDateTime_Check(other) and not isinstance(other, ABCTimestamp)
+        ):
+            # this case is for a datetime object that is specifically
+            # *not* a Timestamp, as the Timestamp case will be
+            # handled after `_validate_ops_compat` returns False below
+            from pandas._libs.tslibs.timestamps import Timestamp
+            return op(self, Timestamp(other))
+            # We are implicitly requiring the canonical behavior to be
+            # defined by Timestamp methods.
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return f(self, item)
+
+            elif other.dtype.kind in "mM":
+                return op(self.to_timedelta64(), other)
+            elif other.dtype.kind == "O":
+                return np.array([op(self, x) for x in other])
+            else:
+                return NotImplemented
+
+        elif not _validate_ops_compat(other):
+            # Includes any of our non-cython classes
+            return NotImplemented
+
+        try:
+            other = Timedelta(other)
+        except ValueError:
+            # failed to parse as timedelta
+            return NotImplemented
+
+        if other is NaT:
+            # e.g. if original other was timedelta64('NaT')
+            return NaT
+
+        # Matching numpy, we cast to the higher resolution. Unlike numpy,
+        #  we raise instead of silently overflowing during this casting.
+        if self._creso < other._creso:
+            self = (<_Timedelta>self)._as_creso(other._creso, round_ok=True)
+        elif self._creso > other._creso:
+            other = (<_Timedelta>other)._as_creso(self._creso, round_ok=True)
+
+        res = op(self._value, other._value)
+        if res == NPY_NAT:
+            # e.g. test_implementation_limits
+            # TODO: more generally could do an overflowcheck in op?
+            return NaT
+
+        return _timedelta_from_value_and_reso(Timedelta, res, reso=self._creso)
+
+    f.__name__ = name
+    return f
+
+
+# ----------------------------------------------------------------------
+# Timedelta Construction
+
+cpdef disallow_ambiguous_unit(unit):
+    if unit in {"Y", "y", "M"}:
+        raise ValueError(
+            "Units 'M', 'Y', and 'y' are no longer supported, as they do not "
+            "represent unambiguous timedelta values durations."
+        )
+
+
+cdef int64_t parse_iso_format_string(str ts) except? -1:
+    """
+    Extracts and cleanses the appropriate values from a match object with
+    groups for each component of an ISO 8601 duration
+
+    Parameters
+    ----------
+    ts: str
+        ISO 8601 Duration formatted string
+
+    Returns
+    -------
+    ns: int64_t
+        Precision in nanoseconds of matched ISO 8601 duration
+
+    Raises
+    ------
+    ValueError
+        If ``ts`` cannot be parsed
+    """
+
+    cdef:
+        unicode c
+        int64_t result = 0, r
+        int p = 0, sign = 1
+        object dec_unit = "ms", err_msg
+        bint have_dot = 0, have_value = 0, neg = 0
+        list number = [], unit = []
+
+    err_msg = f"Invalid ISO 8601 Duration format - {ts}"
+
+    if ts[0] == "-":
+        sign = -1
+        ts = ts[1:]
+
+    for c in ts:
+        # number (ascii codes)
+        if 48 <= ord(c) <= 57:
+
+            have_value = 1
+            if have_dot:
+                if p == 3 and dec_unit != "ns":
+                    unit.append(dec_unit)
+                    if dec_unit == "ms":
+                        dec_unit = "us"
+                    elif dec_unit == "us":
+                        dec_unit = "ns"
+                    p = 0
+                p += 1
+
+            if not len(unit):
+                number.append(c)
+            else:
+                r = timedelta_from_spec(number, "0", unit)
+                result += timedelta_as_neg(r, neg)
+
+                neg = 0
+                unit, number = [], [c]
+        else:
+            if c == "P" or c == "T":
+                pass  # ignore marking characters P and T
+            elif c == "-":
+                if neg or have_value:
+                    raise ValueError(err_msg)
+                else:
+                    neg = 1
+            elif c == "+":
+                pass
+            elif c in ["W", "D", "H", "M"]:
+                if c in ["H", "M"] and len(number) > 2:
+                    raise ValueError(err_msg)
+                if c in ["M", "H"]:
+                    c = c.replace("M", "min").replace("H", "h")
+                unit.append(c)
+                r = timedelta_from_spec(number, "0", unit)
+                result += timedelta_as_neg(r, neg)
+
+                neg = 0
+                unit, number = [], []
+            elif c == ".":
+                # append any seconds
+                if len(number):
+                    r = timedelta_from_spec(number, "0", "s")
+                    result += timedelta_as_neg(r, neg)
+                    unit, number = [], []
+                have_dot = 1
+            elif c == "S":
+                if have_dot:  # ms, us, or ns
+                    if not len(number) or p > 3:
+                        raise ValueError(err_msg)
+                    # pad to 3 digits as required
+                    pad = 3 - p
+                    while pad > 0:
+                        number.append("0")
+                        pad -= 1
+
+                    r = timedelta_from_spec(number, "0", dec_unit)
+                    result += timedelta_as_neg(r, neg)
+                else:  # seconds
+                    r = timedelta_from_spec(number, "0", "s")
+                    result += timedelta_as_neg(r, neg)
+            else:
+                raise ValueError(err_msg)
+
+    if not have_value:
+        # Received string only - never parsed any values
+        raise ValueError(err_msg)
+
+    return sign*result
+
+
+cdef _to_py_int_float(v):
+    # Note: This used to be defined inside Timedelta.__new__
+    # but cython will not allow `cdef` functions to be defined dynamically.
+    if is_integer_object(v):
+        return int(v)
+    elif is_float_object(v):
+        return float(v)
+    raise TypeError(f"Invalid type {type(v)}. Must be int or float.")
+
+
+def _timedelta_unpickle(value, reso):
+    return _timedelta_from_value_and_reso(Timedelta, value, reso)
+
+
+cdef _timedelta_from_value_and_reso(cls, int64_t value, NPY_DATETIMEUNIT reso):
+    # Could make this a classmethod if/when cython supports cdef classmethods
+    cdef:
+        _Timedelta td_base
+
+    assert value != NPY_NAT
+    # For millisecond and second resos, we cannot actually pass int(value) because
+    #  many cases would fall outside of the pytimedelta implementation bounds.
+    #  We pass 0 instead, and override seconds, microseconds, days.
+    #  In principle we could pass 0 for ns and us too.
+    if reso == NPY_FR_ns:
+        td_base = _Timedelta.__new__(cls, microseconds=int(value) // 1000)
+    elif reso == NPY_DATETIMEUNIT.NPY_FR_us:
+        td_base = _Timedelta.__new__(cls, microseconds=int(value))
+    elif reso == NPY_DATETIMEUNIT.NPY_FR_ms:
+        td_base = _Timedelta.__new__(cls, milliseconds=0)
+    elif reso == NPY_DATETIMEUNIT.NPY_FR_s:
+        td_base = _Timedelta.__new__(cls, seconds=0)
+    # Other resolutions are disabled but could potentially be implemented here:
+    # elif reso == NPY_DATETIMEUNIT.NPY_FR_m:
+    #    td_base = _Timedelta.__new__(Timedelta, minutes=int(value))
+    # elif reso == NPY_DATETIMEUNIT.NPY_FR_h:
+    #    td_base = _Timedelta.__new__(Timedelta, hours=int(value))
+    # elif reso == NPY_DATETIMEUNIT.NPY_FR_D:
+    #    td_base = _Timedelta.__new__(Timedelta, days=int(value))
+    else:
+        raise NotImplementedError(
+            "Only resolutions 's', 'ms', 'us', 'ns' are supported."
+        )
+
+    td_base._value = value
+    td_base._is_populated = 0
+    td_base._creso = reso
+    return td_base
+
+
+class MinMaxReso:
+    """
+    We need to define min/max/resolution on both the Timedelta _instance_
+    and Timedelta class.  On an instance, these depend on the object's _reso.
+    On the class, we default to the values we would get with nanosecond _reso.
+    """
+    def __init__(self, name):
+        self._name = name
+
+    def __get__(self, obj, type=None):
+        if self._name == "min":
+            val = np.iinfo(np.int64).min + 1
+        elif self._name == "max":
+            val = np.iinfo(np.int64).max
+        else:
+            assert self._name == "resolution"
+            val = 1
+
+        if obj is None:
+            # i.e. this is on the class, default to nanos
+            return Timedelta(val)
+        else:
+            return Timedelta._from_value_and_reso(val, obj._creso)
+
+    def __set__(self, obj, value):
+        raise AttributeError(f"{self._name} is not settable.")
+
+
+# Similar to Timestamp/datetime, this is a construction requirement for
+# timedeltas that we need to do object instantiation in python. This will
+# serve as a C extension type that shadows the Python class, where we do any
+# heavy lifting.
+cdef class _Timedelta(timedelta):
+    # cdef readonly:
+    #    int64_t value      # nanoseconds
+    #    bint _is_populated  # are my components populated
+    #    int64_t _d, _h, _m, _s, _ms, _us, _ns
+    #    NPY_DATETIMEUNIT _reso
+
+    # higher than np.ndarray and np.matrix
+    __array_priority__ = 100
+    min = MinMaxReso("min")
+    max = MinMaxReso("max")
+    resolution = MinMaxReso("resolution")
+
+    @property
+    def value(self):
+        """
+        Return the value of Timedelta object in nanoseconds.
+
+        Return the total seconds, milliseconds and microseconds
+        of the timedelta as nanoseconds.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timedelta.unit : Return the unit of Timedelta object.
+
+        Examples
+        --------
+        >>> pd.Timedelta(1, "us").value
+        1000
+        """
+        try:
+            return convert_reso(self._value, self._creso, NPY_FR_ns, False)
+        except OverflowError:
+            raise OverflowError(
+                "Cannot convert Timedelta to nanoseconds without overflow. "
+                "Use `.asm8.view('i8')` to cast represent Timedelta in its own "
+                f"unit (here, {self.unit})."
+            )
+
+    @property
+    def _unit(self) -> str:
+        """
+        The abbreviation associated with self._creso.
+        """
+        return npy_unit_to_abbrev(self._creso)
+
+    @property
+    def days(self) -> int:  # TODO(cython3): make cdef property
+        """
+        Returns the days of the timedelta.
+
+        The `days` attribute of a `pandas.Timedelta` object provides the number
+        of days represented by the `Timedelta`. This is useful for extracting
+        the day component from a `Timedelta` that may also include hours, minutes,
+        seconds, and smaller time units. This attribute simplifies the process
+        of working with durations where only the day component is of interest.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timedelta.seconds : Returns the seconds component of the timedelta.
+        Timedelta.microseconds : Returns the microseconds component of the timedelta.
+        Timedelta.total_seconds : Returns the total duration in seconds.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta(1, "d")
+        >>> td.days
+        1
+
+        >>> td = pd.Timedelta('4 min 3 us 42 ns')
+        >>> td.days
+        0
+        """
+        # NB: using the python C-API PyDateTime_DELTA_GET_DAYS will fail
+        #  (or be incorrect)
+        self._ensure_components()
+        return self._d
+
+    @property
+    def seconds(self) -> int:  # TODO(cython3): make cdef property
+        """
+        Return the total hours, minutes, and seconds of the timedelta as seconds.
+
+        Timedelta.seconds = hours * 3600 + minutes * 60 + seconds.
+
+        Returns
+        -------
+        int
+            Number of seconds.
+
+        See Also
+        --------
+        Timedelta.components : Return all attributes with assigned values
+            (i.e. days, hours, minutes, seconds, milliseconds, microseconds,
+            nanoseconds).
+        Timedelta.total_seconds : Express the Timedelta as total number of seconds.
+
+        Examples
+        --------
+        **Using string input**
+
+        >>> td = pd.Timedelta('1 days 2 min 3 us 42 ns')
+        >>> td.seconds
+        120
+
+        **Using integer input**
+
+        >>> td = pd.Timedelta(42, unit='s')
+        >>> td.seconds
+        42
+        """
+        # NB: using the python C-API PyDateTime_DELTA_GET_SECONDS will fail
+        #  (or be incorrect)
+        self._ensure_components()
+        return self._h * 3600 + self._m * 60 + self._s
+
+    @property
+    def microseconds(self) -> int:  # TODO(cython3): make cdef property
+        # NB: using the python C-API PyDateTime_DELTA_GET_MICROSECONDS will fail
+        #  (or be incorrect)
+        """
+        Return the number of microseconds (n), where 0 <= n < 1 millisecond.
+
+        Timedelta.microseconds = milliseconds * 1000 + microseconds.
+
+        Returns
+        -------
+        int
+            Number of microseconds.
+
+        See Also
+        --------
+        Timedelta.components : Return all attributes with assigned values
+            (i.e. days, hours, minutes, seconds, milliseconds, microseconds,
+            nanoseconds).
+
+        Examples
+        --------
+        **Using string input**
+
+        >>> td = pd.Timedelta('1 days 2 min 3 us')
+
+        >>> td.microseconds
+        3
+
+        **Using integer input**
+
+        >>> td = pd.Timedelta(42, unit='us')
+        >>> td.microseconds
+        42
+        """
+        self._ensure_components()
+        return self._ms * 1000 + self._us
+
+    def total_seconds(self) -> float:
+        """
+        Total seconds in the duration.
+
+        This method calculates the total duration in seconds by combining
+        the days, seconds, and microseconds of the `Timedelta` object.
+
+        See Also
+        --------
+        to_timedelta : Convert argument to timedelta.
+        Timedelta : Represents a duration, the difference between two dates or times.
+        Timedelta.seconds : Returns the seconds component of the timedelta.
+        Timedelta.microseconds : Returns the microseconds component of the timedelta.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1min')
+        >>> td
+        Timedelta('0 days 00:01:00')
+        >>> td.total_seconds()
+        60.0
+        """
+        # We need to override bc we overrode days/seconds/microseconds
+        # TODO: add nanos/1e9?
+        return self.days * 24 * 3600 + self.seconds + self.microseconds / 1_000_000
+
+    @property
+    def unit(self) -> str:
+        """
+        Return the unit of Timedelta object.
+
+        The unit of Timedelta object is nanosecond, i.e., 'ns' by default.
+
+        Returns
+        -------
+        str
+
+        See Also
+        --------
+        Timedelta.value : Return the value of Timedelta object in nanoseconds.
+        Timedelta.as_unit : Convert the underlying int64 representation to
+            the given unit.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta(42, unit='us')
+        'ns'
+        """
+        return npy_unit_to_abbrev(self._creso)
+
+    def __hash__(_Timedelta self):
+        if self._has_ns():
+            # Note: this does *not* satisfy the invariance
+            #  td1 == td2 \\Rightarrow hash(td1) == hash(td2)
+            #  if td1 and td2 have different _resos. timedelta64 also has this
+            #  non-invariant behavior.
+            #  see GH#44504
+            return hash(self._value)
+        elif self._is_in_pytimedelta_bounds() and (
+            self._creso == NPY_FR_ns or self._creso == NPY_DATETIMEUNIT.NPY_FR_us
+        ):
+            # If we can defer to timedelta.__hash__, do so, as that
+            #  ensures the hash is invariant to our _reso.
+            # We can only defer for ns and us, as for these two resos we
+            #  call _Timedelta.__new__ with the correct input in
+            #  _timedelta_from_value_and_reso; so timedelta.__hash__
+            #  will be correct
+            return timedelta.__hash__(self)
+        else:
+            # We want to ensure that two equivalent Timedelta objects
+            #  have the same hash.  So we try downcasting to the next-lowest
+            #  resolution.
+            try:
+                obj = (<_Timedelta>self)._as_creso(<NPY_DATETIMEUNIT>(self._creso + 1))
+            except OutOfBoundsTimedelta:
+                # Doesn't fit, so we're off the hook
+                return hash(self._value)
+            else:
+                return hash(obj)
+
+    def __richcmp__(_Timedelta self, object other, int op):
+        cdef:
+            _Timedelta ots
+
+        if isinstance(other, _Timedelta):
+            ots = other
+        elif is_any_td_scalar(other):
+            try:
+                ots = Timedelta(other)
+            except OutOfBoundsTimedelta as err:
+                # GH#49021 pytimedelta.max overflows
+                if not PyDelta_Check(other):
+                    # TODO: handle this case
+                    raise
+                ltup = (self.days, self.seconds, self.microseconds, self.nanoseconds)
+                rtup = (other.days, other.seconds, other.microseconds, 0)
+                if op == Py_EQ:
+                    return ltup == rtup
+                elif op == Py_NE:
+                    return ltup != rtup
+                elif op == Py_LT:
+                    return ltup < rtup
+                elif op == Py_LE:
+                    return ltup <= rtup
+                elif op == Py_GT:
+                    return ltup > rtup
+                elif op == Py_GE:
+                    return ltup >= rtup
+
+        elif other is NaT:
+            return op == Py_NE
+
+        elif util.is_array(other):
+            if other.dtype.kind == "m":
+                return PyObject_RichCompare(self.asm8, other, op)
+            elif other.dtype.kind == "O":
+                # operate element-wise
+                return np.array(
+                    [PyObject_RichCompare(self, x, op) for x in other],
+                    dtype=bool,
+                )
+            if op == Py_EQ:
+                return np.zeros(other.shape, dtype=bool)
+            elif op == Py_NE:
+                return np.ones(other.shape, dtype=bool)
+            return NotImplemented  # let other raise TypeError
+
+        else:
+            return NotImplemented
+
+        if self._creso == ots._creso:
+            return cmp_scalar(self._value, ots._value, op)
+        return self._compare_mismatched_resos(ots, op)
+
+    # TODO: reuse/share with Timestamp
+    cdef bint _compare_mismatched_resos(self, _Timedelta other, op):
+        # Can't just dispatch to numpy as they silently overflow and get it wrong
+        cdef:
+            npy_datetimestruct dts_self
+            npy_datetimestruct dts_other
+
+        # dispatch to the datetimestruct utils instead of writing new ones!
+        pandas_datetime_to_datetimestruct(self._value, self._creso, &dts_self)
+        pandas_datetime_to_datetimestruct(other._value, other._creso, &dts_other)
+        return cmp_dtstructs(&dts_self,  &dts_other, op)
+
+    cdef bint _has_ns(self):
+        if self._creso == NPY_FR_ns:
+            return self._value % 1000 != 0
+        elif self._creso < NPY_FR_ns:
+            # i.e. seconds, millisecond, microsecond
+            return False
+        else:
+            raise NotImplementedError(self._creso)
+
+    cdef bint _is_in_pytimedelta_bounds(self):
+        """
+        Check if we are within the bounds of datetime.timedelta.
+        """
+        self._ensure_components()
+        return -999999999 <= self._d and self._d <= 999999999
+
+    cdef _ensure_components(_Timedelta self):
+        """
+        compute the components
+        """
+        if self._is_populated:
+            return
+
+        cdef:
+            pandas_timedeltastruct tds
+
+        pandas_timedelta_to_timedeltastruct(self._value, self._creso, &tds)
+        self._d = tds.days
+        self._h = tds.hrs
+        self._m = tds.min
+        self._s = tds.sec
+        self._ms = tds.ms
+        self._us = tds.us
+        self._ns = tds.ns
+        self._seconds = tds.seconds
+        self._microseconds = tds.microseconds
+
+        self._is_populated = 1
+
+    cpdef timedelta to_pytimedelta(_Timedelta self):
+        """
+        Convert a pandas Timedelta object into a python ``datetime.timedelta`` object.
+
+        Timedelta objects are internally saved as numpy datetime64[ns] dtype.
+        Use to_pytimedelta() to convert to object dtype.
+
+        Returns
+        -------
+        datetime.timedelta or numpy.array of datetime.timedelta
+
+        See Also
+        --------
+        to_timedelta : Convert argument to Timedelta type.
+
+        Notes
+        -----
+        Any nanosecond resolution will be lost.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('3D')
+        >>> td
+        Timedelta('3 days 00:00:00')
+        >>> td.to_pytimedelta()
+        datetime.timedelta(days=3)
+        """
+        if self._creso == NPY_FR_ns:
+            us, remainder = divmod(self._value, 1000)
+            if remainder >= 500:
+                us += 1
+            return timedelta(microseconds=us)
+
+        # TODO(@WillAyd): is this the right way to use components?
+        self._ensure_components()
+        return timedelta(
+            days=self._d, seconds=self._seconds, microseconds=self._microseconds
+        )
+
+    def to_timedelta64(self) -> np.timedelta64:
+        """
+        Return a numpy.timedelta64 object with 'ns' precision.
+
+        Since NumPy uses ``timedelta64`` objects for its time operations, converting
+        a pandas ``Timedelta`` into a NumPy ``timedelta64`` provides seamless
+        integration between the two libraries, especially when working in environments
+        that heavily rely on NumPy for array-based calculations.
+
+        See Also
+        --------
+        to_timedelta : Convert argument to timedelta.
+        numpy.timedelta64 : A NumPy object for time duration.
+        Timedelta : Represents a duration, the difference between two dates
+            or times.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('3D')
+        >>> td
+        Timedelta('3 days 00:00:00')
+        >>> td.to_timedelta64()
+        numpy.timedelta64(259200000000000,'ns')
+        """
+        cdef:
+            str abbrev = npy_unit_to_abbrev(self._creso)
+        # TODO: way to create a np.timedelta64 obj with the reso directly
+        #  instead of having to get the abbrev?
+        return np.timedelta64(self._value, abbrev)
+
+    def to_numpy(self, dtype=None, copy=False) -> np.timedelta64:
+        """
+        Convert the Timedelta to a NumPy timedelta64.
+
+        This is an alias method for `Timedelta.to_timedelta64()`.
+
+        Parameters
+        ----------
+        dtype : NoneType
+            It is available here only for compatibility. Its value will not
+            affect the return value.
+        copy : bool, default False
+            It is available here only for compatibility. Its value will not
+            affect the return value.
+
+        Returns
+        -------
+        numpy.timedelta64
+
+        See Also
+        --------
+        Series.to_numpy : Similar method for Series.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('3D')
+        >>> td
+        Timedelta('3 days 00:00:00')
+        >>> td.to_numpy()
+        numpy.timedelta64(259200000000000,'ns')
+        """
+        if dtype is not None or copy is not False:
+            raise ValueError(
+                "Timedelta.to_numpy dtype and copy arguments are ignored"
+            )
+        return self.to_timedelta64()
+
+    def view(self, dtype):
+        """
+        Array view compatibility.
+
+        This method allows you to reinterpret the underlying data of a Timedelta
+        object as a different dtype. The `view` method provides a way to reinterpret
+        the internal representation of the `Timedelta` object without modifying its
+        data. This is particularly useful when you need to work with the underlying
+        data directly, such as for performance optimizations or interfacing with
+        low-level APIs. The returned value is typically the number of nanoseconds
+        since the epoch, represented as an integer or another specified dtype.
+
+        Parameters
+        ----------
+        dtype : str or dtype
+            The dtype to view the underlying data as.
+
+        See Also
+        --------
+        Timedelta.asm8 : Return a numpy timedelta64 array scalar view.
+        numpy.ndarray.view : Returns a view of an array with the same data.
+        Timedelta.to_numpy : Converts the Timedelta to a NumPy timedelta64.
+        Timedelta.total_seconds : Returns the total duration of the Timedelta
+            object in seconds.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('3D')
+        >>> td
+        Timedelta('3 days 00:00:00')
+        >>> td.view(int)
+        259200000000000
+        """
+        return np.timedelta64(self._value).view(dtype)
+
+    @property
+    def components(self):
+        """
+        Return a components namedtuple-like.
+
+        Each component represents a different time unit, allowing you to access the
+        breakdown of the total duration in terms of days, hours, minutes, seconds,
+        milliseconds, microseconds, and nanoseconds.
+
+        See Also
+        --------
+        Timedelta.total_seconds : Returns the total duration of the Timedelta in
+            seconds.
+        to_timedelta : Convert argument to Timedelta.
+        Timedelta : Represents a duration, the difference between two dates or times.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('2 day 4 min 3 us 42 ns')
+        >>> td.components
+        Components(days=2, hours=0, minutes=4, seconds=0, milliseconds=0,
+            microseconds=3, nanoseconds=42)
+        """
+        self._ensure_components()
+        # return the named tuple
+        return Components(self._d, self._h, self._m, self._s,
+                          self._ms, self._us, self._ns)
+
+    @property
+    def asm8(self) -> np.timedelta64:
+        """
+        Return a numpy timedelta64 array scalar view.
+
+        Provides access to the array scalar view (i.e. a combination of the
+        value and the units) associated with the numpy.timedelta64().view(),
+        including a 64-bit integer representation of the timedelta in
+        nanoseconds (Python int compatible).
+
+        Returns
+        -------
+        numpy timedelta64 array scalar view
+            Array scalar view of the timedelta in nanoseconds.
+
+        See Also
+        --------
+            Timedelta.total_seconds : Return the total seconds in the duration.
+            Timedelta.components : Return a namedtuple of the Timedelta's components.
+            Timedelta.to_timedelta64 : Convert the Timedelta to a numpy.timedelta64.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1 days 2 min 3 us 42 ns')
+        >>> td.asm8
+        numpy.timedelta64(86520000003042,'ns')
+
+        >>> td = pd.Timedelta('2 min 3 s')
+        >>> td.asm8
+        numpy.timedelta64(123000000000,'ns')
+
+        >>> td = pd.Timedelta('3 ms 5 us')
+        >>> td.asm8
+        numpy.timedelta64(3005000,'ns')
+
+        >>> td = pd.Timedelta(42, unit='ns')
+        >>> td.asm8
+        numpy.timedelta64(42,'ns')
+        """
+        return self.to_timedelta64()
+
+    @property
+    def resolution_string(self) -> str:
+        """
+        Return a string representing the lowest timedelta resolution.
+
+        Each timedelta has a defined resolution that represents the lowest OR
+        most granular level of precision. Each level of resolution is
+        represented by a short string as defined below:
+
+        Resolution:     Return value
+
+        * Days:         'D'
+        * Hours:        'h'
+        * Minutes:      'min'
+        * Seconds:      's'
+        * Milliseconds: 'ms'
+        * Microseconds: 'us'
+        * Nanoseconds:  'ns'
+
+        Returns
+        -------
+        str
+            Timedelta resolution.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1 days 2 min 3 us 42 ns')
+        >>> td.resolution_string
+        'ns'
+
+        >>> td = pd.Timedelta('1 days 2 min 3 us')
+        >>> td.resolution_string
+        'us'
+
+        >>> td = pd.Timedelta('2 min 3 s')
+        >>> td.resolution_string
+        's'
+
+        >>> td = pd.Timedelta(36, unit='us')
+        >>> td.resolution_string
+        'us'
+        """
+        self._ensure_components()
+        if self._ns:
+            return "ns"
+        elif self._us:
+            return "us"
+        elif self._ms:
+            return "ms"
+        elif self._s:
+            return "s"
+        elif self._m:
+            return "min"
+        elif self._h:
+            return "h"
+        else:
+            return "D"
+
+    @property
+    def nanoseconds(self):
+        """
+        Return the number of nanoseconds (n), where 0 <= n < 1 microsecond.
+
+        Returns
+        -------
+        int
+            Number of nanoseconds.
+
+        See Also
+        --------
+        Timedelta.components : Return all attributes with assigned values
+            (i.e. days, hours, minutes, seconds, milliseconds, microseconds,
+            nanoseconds).
+
+        Examples
+        --------
+        **Using string input**
+
+        >>> td = pd.Timedelta('1 days 2 min 3 us 42 ns')
+
+        >>> td.nanoseconds
+        42
+
+        **Using integer input**
+
+        >>> td = pd.Timedelta(42, unit='ns')
+        >>> td.nanoseconds
+        42
+        """
+        self._ensure_components()
+        return self._ns
+
+    def _repr_base(self, format=None) -> str:
+        """
+
+        Parameters
+        ----------
+        format : None|all|sub_day|long
+
+        Returns
+        -------
+        converted : string of a Timedelta
+
+        """
+        cdef:
+            str sign, fmt
+            dict comp_dict
+            object subs
+
+        self._ensure_components()
+
+        if self._d < 0:
+            sign = " +"
+        else:
+            sign = " "
+
+        if format == "all":
+            fmt = ("{days} days{sign}{hours:02}:{minutes:02}:{seconds:02}."
+                   "{milliseconds:03}{microseconds:03}{nanoseconds:03}")
+        else:
+            # if we have a partial day
+            subs = (self._h or self._m or self._s or
+                    self._ms or self._us or self._ns)
+
+            if self._ms or self._us or self._ns:
+                seconds_fmt = "{seconds:02}.{milliseconds:03}{microseconds:03}"
+                if self._ns:
+                    # GH#9309
+                    seconds_fmt += "{nanoseconds:03}"
+            else:
+                seconds_fmt = "{seconds:02}"
+
+            if format == "sub_day" and not self._d:
+                fmt = "{hours:02}:{minutes:02}:" + seconds_fmt
+            elif subs or format == "long":
+                fmt = "{days} days{sign}{hours:02}:{minutes:02}:" + seconds_fmt
+            else:
+                fmt = "{days} days"
+
+        comp_dict = self.components._asdict()
+        comp_dict["sign"] = sign
+
+        return fmt.format(**comp_dict)
+
+    def __repr__(self) -> str:
+        repr_based = self._repr_base(format="long")
+        return f"Timedelta('{repr_based}')"
+
+    def __str__(self) -> str:
+        return self._repr_base(format="long")
+
+    def __bool__(self) -> bool:
+        return self._value!= 0
+
+    def isoformat(self) -> str:
+        """
+        Format the Timedelta as ISO 8601 Duration.
+
+        ``P[n]Y[n]M[n]DT[n]H[n]M[n]S``, where the ``[n]`` s are replaced by the
+        values. See https://en.wikipedia.org/wiki/ISO_8601#Durations.
+
+        Returns
+        -------
+        str
+
+        See Also
+        --------
+        Timestamp.isoformat : Function is used to convert the given
+            Timestamp object into the ISO format.
+
+        Notes
+        -----
+        The longest component is days, whose value may be larger than
+        365.
+        Every component is always included, even if its value is 0.
+        Pandas uses nanosecond precision, so up to 9 decimal places may
+        be included in the seconds component.
+        Trailing 0's are removed from the seconds component after the decimal.
+        We do not 0 pad components, so it's `...T5H...`, not `...T05H...`
+
+        Examples
+        --------
+        >>> td = pd.Timedelta(days=6, minutes=50, seconds=3,
+        ...                   milliseconds=10, microseconds=10, nanoseconds=12)
+
+        >>> td.isoformat()
+        'P6DT0H50M3.010010012S'
+        >>> pd.Timedelta(hours=1, seconds=10).isoformat()
+        'P0DT1H0M10S'
+        >>> pd.Timedelta(days=500.5).isoformat()
+        'P500DT12H0M0S'
+        """
+        components = self.components
+        seconds = (f"{components.seconds}."
+                   f"{components.milliseconds:0>3}"
+                   f"{components.microseconds:0>3}"
+                   f"{components.nanoseconds:0>3}")
+        # Trim unnecessary 0s, 1.000000000 -> 1
+        seconds = seconds.rstrip("0").rstrip(".")
+        tpl = (f"P{components.days}DT{components.hours}"
+               f"H{components.minutes}M{seconds}S")
+        return tpl
+
+    # ----------------------------------------------------------------
+    # Constructors
+
+    @classmethod
+    def _from_value_and_reso(cls, int64_t value, NPY_DATETIMEUNIT reso):
+        # exposing as classmethod for testing
+        return _timedelta_from_value_and_reso(cls, value, reso)
+
+    def as_unit(self, str unit, bint round_ok=True):
+        """
+        Convert the underlying int64 representation to the given unit.
+
+        Parameters
+        ----------
+        unit : {"ns", "us", "ms", "s"}
+        round_ok : bool, default True
+            If False and the conversion requires rounding, raise.
+
+        Returns
+        -------
+        Timedelta
+
+        See Also
+        --------
+        Timedelta : Represents a duration, the difference between two dates or times.
+        to_timedelta : Convert argument to timedelta.
+        Timedelta.asm8 : Return a numpy timedelta64 array scalar view.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1001ms')
+        >>> td
+        Timedelta('0 days 00:00:01.001000')
+        >>> td.as_unit('s')
+        Timedelta('0 days 00:00:01')
+        """
+        dtype = np.dtype(f"m8[{unit}]")
+        reso = get_unit_from_dtype(dtype)
+        return self._as_creso(reso, round_ok=round_ok)
+
+    @cython.cdivision(False)
+    cdef _Timedelta _as_creso(self, NPY_DATETIMEUNIT reso, bint round_ok=True):
+        cdef:
+            int64_t value
+
+        if reso == self._creso:
+            return self
+
+        try:
+            value = convert_reso(self._value, self._creso, reso, round_ok=round_ok)
+        except OverflowError as err:
+            unit = npy_unit_to_abbrev(reso)
+            raise OutOfBoundsTimedelta(
+                f"Cannot cast {self} to unit='{unit}' without overflow."
+            ) from err
+
+        return type(self)._from_value_and_reso(value, reso=reso)
+
+    cpdef _maybe_cast_to_matching_resos(self, _Timedelta other):
+        """
+        If _resos do not match, cast to the higher resolution, raising on overflow.
+        """
+        if self._creso > other._creso:
+            other = other._as_creso(self._creso)
+        elif self._creso < other._creso:
+            self = self._as_creso(other._creso)
+        return self, other
+
+
+# Python front end to C extension type _Timedelta
+# This serves as the box for timedelta64
+@set_module("pandas")
+class Timedelta(_Timedelta):
+    """
+    Represents a duration, the difference between two dates or times.
+
+    Timedelta is the pandas equivalent of python's ``datetime.timedelta``
+    and is interchangeable with it in most cases.
+
+    Parameters
+    ----------
+    value : Timedelta, timedelta, np.timedelta64, str, int or float
+        Input value.
+    unit : str, default 'ns'
+        If input is an integer, denote the unit of the input.
+        If input is a float, denote the unit of the integer parts.
+        The decimal parts with resolution lower than 1 nanosecond are ignored.
+
+        Possible values:
+
+        * 'W', or 'D'
+        * 'days', or 'day'
+        * 'hours', 'hour', 'hr', or 'h'
+        * 'minutes', 'minute', 'min', or 'm'
+        * 'seconds', 'second', 'sec', or 's'
+        * 'milliseconds', 'millisecond', 'millis', 'milli', or 'ms'
+        * 'microseconds', 'microsecond', 'micros', 'micro', or 'us'
+        * 'nanoseconds', 'nanosecond', 'nanos', 'nano', or 'ns'.
+
+        .. deprecated:: 3.0.0
+
+            Allowing the values `w`, `d`, `MIN`, `MS`, `US` and `NS` to denote units
+            are deprecated in favour of the values `W`, `D`, `min`, `ms`, `us` and `ns`.
+
+    **kwargs
+        Available kwargs: {days, seconds, microseconds,
+        milliseconds, minutes, hours, weeks}.
+        Values for construction in compat with datetime.timedelta.
+        Numpy ints and floats will be coerced to python ints and floats.
+
+    See Also
+    --------
+    Timestamp : Represents a single timestamp in time.
+    TimedeltaIndex : Immutable Index of timedelta64 data.
+    DateOffset : Standard kind of date increment used for a date range.
+    to_timedelta : Convert argument to timedelta.
+    datetime.timedelta : Represents a duration in the datetime module.
+    numpy.timedelta64 : Represents a duration compatible with NumPy.
+
+    Notes
+    -----
+    The constructor may take in either both values of value and unit or
+    kwargs as above. Either one of them must be used during initialization
+
+    The ``.value`` attribute is always in ns.
+
+    If the precision is higher than nanoseconds, the precision of the duration is
+    truncated to nanoseconds.
+
+    Examples
+    --------
+    Here we initialize Timedelta object with both value and unit
+
+    >>> td = pd.Timedelta(1, "D")
+    >>> td
+    Timedelta('1 days 00:00:00')
+
+    Here we initialize the Timedelta object with kwargs
+
+    >>> td2 = pd.Timedelta(days=1)
+    >>> td2
+    Timedelta('1 days 00:00:00')
+
+    We see that either way we get the same result
+    """
+
+    _req_any_kwargs_new = {"weeks", "days", "hours", "minutes", "seconds",
+                           "milliseconds", "microseconds", "nanoseconds"}
+
+    def __new__(cls, object value=_no_input, unit=None, **kwargs):
+        if value is _no_input:
+            if not len(kwargs):
+                raise ValueError("cannot construct a Timedelta without a "
+                                 "value/unit or descriptive keywords "
+                                 "(days,seconds....)")
+
+            kwargs = {key: _to_py_int_float(kwargs[key]) for key in kwargs}
+
+            unsupported_kwargs = set(kwargs)
+            unsupported_kwargs.difference_update(cls._req_any_kwargs_new)
+            if unsupported_kwargs or not cls._req_any_kwargs_new.intersection(kwargs):
+                raise ValueError(
+                    "cannot construct a Timedelta from the passed arguments, "
+                    "allowed keywords are "
+                    "[weeks, days, hours, minutes, seconds, "
+                    "milliseconds, microseconds, nanoseconds]"
+                )
+
+            # GH43764, convert any input to nanoseconds first and then
+            # create the timedelta. This ensures that any potential
+            # nanosecond contributions from kwargs parsed as floats
+            # are taken into consideration.
+            seconds = int((
+                (
+                    (kwargs.get("days", 0) + kwargs.get("weeks", 0) * 7) * 24
+                    + kwargs.get("hours", 0)
+                ) * 3600
+                + kwargs.get("minutes", 0) * 60
+                + kwargs.get("seconds", 0)
+                ) * 1_000_000_000
+            )
+
+            ns = kwargs.get("nanoseconds", 0)
+            us = kwargs.get("microseconds", 0)
+            ms = kwargs.get("milliseconds", 0)
+            try:
+                value = np.timedelta64(
+                    int(ns)
+                    + int(us * 1_000)
+                    + int(ms * 1_000_000)
+                    + seconds
+                )
+            except OverflowError as err:
+                # GH#55503
+                msg = (
+                    f"seconds={seconds}, milliseconds={ms}, "
+                    f"microseconds={us}, nanoseconds={ns}"
+                )
+                raise OutOfBoundsTimedelta(msg) from err
+
+        disallow_ambiguous_unit(unit)
+
+        # GH 30543 if pd.Timedelta already passed, return it
+        # check that only value is passed
+        if isinstance(value, _Timedelta):
+            # 'unit' is benign in this case, but e.g. days or seconds
+            #  doesn't make sense here.
+            if len(kwargs):
+                # GH#48898
+                raise ValueError(
+                    "Cannot pass both a Timedelta input and timedelta keyword "
+                    "arguments, got "
+                    f"{list(kwargs.keys())}"
+                )
+            return value
+        elif isinstance(value, str):
+            if unit is not None:
+                raise ValueError("unit must not be specified if the value is a str")
+            if (len(value) > 0 and value[0] == "P") or (
+                len(value) > 1 and value[:2] == "-P"
+            ):
+                value = parse_iso_format_string(value)
+            else:
+                value = parse_timedelta_string(value)
+            value = np.timedelta64(value)
+        elif PyDelta_Check(value):
+            # pytimedelta object -> microsecond resolution
+            new_value = delta_to_nanoseconds(
+                value, reso=NPY_DATETIMEUNIT.NPY_FR_us
+            )
+            return cls._from_value_and_reso(
+                new_value, reso=NPY_DATETIMEUNIT.NPY_FR_us
+            )
+        elif cnp.is_timedelta64_object(value):
+            # Retain the resolution if possible, otherwise cast to the nearest
+            #  supported resolution.
+            new_value = cnp.get_timedelta64_value(value)
+            if new_value == NPY_NAT:
+                # i.e. np.timedelta64("NaT")
+                return NaT
+
+            reso = get_datetime64_unit(value)
+            if not (is_supported_unit(reso) or
+                    reso in [NPY_DATETIMEUNIT.NPY_FR_m,
+                             NPY_DATETIMEUNIT.NPY_FR_h,
+                             NPY_DATETIMEUNIT.NPY_FR_D,
+                             NPY_DATETIMEUNIT.NPY_FR_W,
+                             NPY_DATETIMEUNIT.NPY_FR_GENERIC]):
+                err = npy_unit_to_abbrev(reso)
+                raise ValueError(
+                    f"Unit {err} is not supported. "
+                    "Only unambiguous timedelta values durations are supported. "
+                    "Allowed units are 'W', 'D', 'h', 'm', 's', 'ms', 'us', 'ns'")
+
+            new_reso = get_supported_reso(reso)
+            if reso != NPY_DATETIMEUNIT.NPY_FR_GENERIC:
+                try:
+                    new_value = convert_reso(
+                        new_value,
+                        reso,
+                        new_reso,
+                        round_ok=True,
+                    )
+                except (OverflowError, OutOfBoundsDatetime) as err:
+                    raise OutOfBoundsTimedelta(value) from err
+            return cls._from_value_and_reso(new_value, reso=new_reso)
+
+        elif is_tick_object(value):
+            new_reso = get_supported_reso(value._creso)
+            new_value = delta_to_nanoseconds(value, reso=new_reso)
+            return cls._from_value_and_reso(new_value, reso=new_reso)
+
+        elif is_integer_object(value) or is_float_object(value):
+            # unit=None is de-facto 'ns'
+            unit = parse_timedelta_unit(unit)
+            value = convert_to_timedelta64(value, unit)
+        elif checknull_with_nat_and_na(value):
+            return NaT
+        else:
+            raise ValueError(
+                "Value must be Timedelta, string, integer, "
+                f"float, timedelta or convertible, not {type(value).__name__}"
+            )
+
+        if cnp.is_timedelta64_object(value):
+            value = value.view("i8")
+
+        # nat
+        if value == NPY_NAT:
+            return NaT
+
+        return _timedelta_from_value_and_reso(cls, value, NPY_FR_ns)
+
+    def __setstate__(self, state):
+        if len(state) == 1:
+            # older pickle, only supported nanosecond
+            value = state[0]
+            reso = NPY_FR_ns
+        else:
+            value, reso = state
+        self._value= value
+        self._creso = reso
+
+    def __reduce__(self):
+        object_state = self._value, self._creso
+        return (_timedelta_unpickle, object_state)
+
+    @cython.cdivision(True)
+    def _round(self, freq, mode):
+        cdef:
+            int64_t result, unit
+            ndarray[int64_t] arr
+
+        unit = get_unit_for_round(freq, self._creso)
+
+        arr = np.array([self._value], dtype="i8")
+        try:
+            result = round_nsint64(arr, mode, unit)[0]
+        except OverflowError as err:
+            raise OutOfBoundsTimedelta(
+                f"Cannot round {self} to freq={freq} without overflow"
+            ) from err
+        return Timedelta._from_value_and_reso(result, self._creso)
+
+    def round(self, freq):
+        """
+        Round the Timedelta to the specified resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the rounding resolution.
+            It uses the same units as class constructor :class:`~pandas.Timedelta`.
+
+        Returns
+        -------
+        a new Timedelta rounded to the given resolution of `freq`
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted
+
+        See Also
+        --------
+            Timedelta.floor : Floor the Timedelta to the specified resolution.
+            Timedelta.round : Round the Timedelta to the nearest specified resolution.
+            Timestamp.ceil : Similar method for Timestamp objects.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1001ms')
+        >>> td
+        Timedelta('0 days 00:00:01.001000')
+        >>> td.round('s')
+        Timedelta('0 days 00:00:01')
+        """
+        return self._round(freq, RoundTo.NEAREST_HALF_EVEN)
+
+    def floor(self, freq):
+        """
+        Return a new Timedelta floored to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the flooring resolution.
+            It uses the same units as class constructor :class:`~pandas.Timedelta`.
+
+        Returns
+        -------
+        Timedelta
+            A new Timedelta object floored to the specified resolution.
+
+        See Also
+        --------
+            Timestamp.ceil : Round the Timestamp up to the nearest specified resolution.
+            Timestamp.round : Round the Timestamp to the nearest specified resolution.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1001ms')
+        >>> td
+        Timedelta('0 days 00:00:01.001000')
+        >>> td.floor('s')
+        Timedelta('0 days 00:00:01')
+        """
+        return self._round(freq, RoundTo.MINUS_INFTY)
+
+    def ceil(self, freq):
+        """
+        Return a new Timedelta ceiled to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the ceiling resolution. Must be a fixed
+            frequency like 's' (second) not 'ME' (month end). See
+            :ref:`frequency aliases <timeseries.offset_aliases>` for
+            a list of possible `freq` values.
+
+        Returns
+        -------
+        Timedelta
+            A new Timedelta object ceiled to the specified resolution.
+
+        See Also
+        --------
+            Timedelta.floor : Floor the Timedelta to the specified resolution.
+            Timedelta.round : Round the Timedelta to the nearest specified resolution.
+
+        Examples
+        --------
+        >>> td = pd.Timedelta('1001ms')
+        >>> td
+        Timedelta('0 days 00:00:01.001000')
+        >>> td.ceil('s')
+        Timedelta('0 days 00:00:02')
+        """
+        return self._round(freq, RoundTo.PLUS_INFTY)
+
+    # ----------------------------------------------------------------
+    # Arithmetic Methods
+    # TODO: Can some of these be defined in the cython class?
+
+    __neg__ = _op_unary_method(lambda x: -x, "__neg__")
+    __pos__ = _op_unary_method(lambda x: x, "__pos__")
+    __abs__ = _op_unary_method(lambda x: abs(x), "__abs__")
+
+    __add__ = _binary_op_method_timedeltalike(lambda x, y: x + y, "__add__")
+    __radd__ = _binary_op_method_timedeltalike(lambda x, y: x + y, "__radd__")
+    __sub__ = _binary_op_method_timedeltalike(lambda x, y: x - y, "__sub__")
+    __rsub__ = _binary_op_method_timedeltalike(lambda x, y: y - x, "__rsub__")
+
+    def __mul__(self, other):
+        if is_integer_object(other) or is_float_object(other):
+            if util.is_nan(other):
+                # np.nan * timedelta -> np.timedelta64("NaT"), in this case NaT
+                return NaT
+
+            return _timedelta_from_value_and_reso(
+                Timedelta,
+                <int64_t>(other * self._value),
+                reso=self._creso,
+            )
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return self.__mul__(item)
+            return other * self.to_timedelta64()
+
+        return NotImplemented
+
+    __rmul__ = __mul__
+
+    def __truediv__(self, other):
+        if _should_cast_to_timedelta(other):
+            # We interpret NaT as timedelta64("NaT")
+            other = Timedelta(other)
+            if other is NaT:
+                return np.nan
+            if other._creso != self._creso:
+                self, other = self._maybe_cast_to_matching_resos(other)
+            return self._value/ float(other._value)
+
+        elif is_integer_object(other) or is_float_object(other):
+            # integers or floats
+            if util.is_nan(other):
+                return NaT
+            # We want NumPy numeric scalars to behave like Python scalars
+            # post NEP 50
+            if isinstance(other, cnp.integer):
+                other = int(other)
+            if isinstance(other, cnp.floating):
+                other = float(other)
+            return Timedelta._from_value_and_reso(
+                <int64_t>(self._value/ other), self._creso
+            )
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return self.__truediv__(item)
+            return self.to_timedelta64() / other
+
+        return NotImplemented
+
+    def __rtruediv__(self, other):
+        if _should_cast_to_timedelta(other):
+            # We interpret NaT as timedelta64("NaT")
+            other = Timedelta(other)
+            if other is NaT:
+                return np.nan
+            if self._creso != other._creso:
+                self, other = self._maybe_cast_to_matching_resos(other)
+            return float(other._value) / self._value
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return self.__rtruediv__(item)
+            elif other.dtype.kind == "O":
+                # GH#31869
+                return np.array([x / self for x in other])
+
+            # TODO: if other.dtype.kind == "m" and other.dtype != self.asm8.dtype
+            #  then should disallow for consistency with scalar behavior; requires
+            #  deprecation cycle. (or changing scalar behavior)
+            return other / self.to_timedelta64()
+
+        return NotImplemented
+
+    def __floordiv__(self, other):
+        # numpy does not implement floordiv for timedelta64 dtype, so we cannot
+        # just defer
+        if _should_cast_to_timedelta(other):
+            # We interpret NaT as timedelta64("NaT")
+            other = Timedelta(other)
+            if other is NaT:
+                return np.nan
+            if self._creso != other._creso:
+                self, other = self._maybe_cast_to_matching_resos(other)
+            return self._value// other._value
+
+        elif is_integer_object(other) or is_float_object(other):
+            if util.is_nan(other):
+                return NaT
+            # We want NumPy numeric scalars to behave like Python scalars
+            # post NEP 50
+            if isinstance(other, cnp.integer):
+                other = int(other)
+            if isinstance(other, cnp.floating):
+                other = float(other)
+            return type(self)._from_value_and_reso(self._value// other, self._creso)
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return self.__floordiv__(item)
+
+            if other.dtype.kind == "m":
+                # also timedelta-like
+                with warnings.catch_warnings():
+                    warnings.filterwarnings(
+                        "ignore",
+                        "invalid value encountered in floor_divide",
+                        RuntimeWarning
+                    )
+                    result = self.asm8 // other
+                mask = other.view("i8") == NPY_NAT
+                if mask.any():
+                    # We differ from numpy here
+                    result = result.astype("f8")
+                    result[mask] = np.nan
+                return result
+
+            elif other.dtype.kind in "iuf":
+                if other.ndim == 0:
+                    return self // other.item()
+                else:
+                    return self.to_timedelta64() // other
+
+            raise TypeError(f"Invalid dtype {other.dtype} for __floordiv__")
+
+        return NotImplemented
+
+    def __rfloordiv__(self, other):
+        # numpy does not implement floordiv for timedelta64 dtype, so we cannot
+        # just defer
+        if _should_cast_to_timedelta(other):
+            # We interpret NaT as timedelta64("NaT")
+            other = Timedelta(other)
+            if other is NaT:
+                return np.nan
+            if self._creso != other._creso:
+                self, other = self._maybe_cast_to_matching_resos(other)
+            return other._value// self._value
+
+        elif is_array(other):
+            if other.ndim == 0:
+                # see also: item_from_zerodim
+                item = cnp.PyArray_ToScalar(cnp.PyArray_DATA(other), other)
+                return self.__rfloordiv__(item)
+
+            if other.dtype.kind == "m":
+                # also timedelta-like
+                with warnings.catch_warnings():
+                    warnings.filterwarnings(
+                        "ignore",
+                        "invalid value encountered in floor_divide",
+                        RuntimeWarning
+                    )
+                    result = other // self.asm8
+                mask = other.view("i8") == NPY_NAT
+                if mask.any():
+                    # We differ from numpy here
+                    result = result.astype("f8")
+                    result[mask] = np.nan
+                return result
+
+            # Includes integer array // Timedelta, disallowed in GH#19761
+            raise TypeError(f"Invalid dtype {other.dtype} for __floordiv__")
+
+        return NotImplemented
+
+    def __mod__(self, other):
+        # Naive implementation, room for optimization
+        return self.__divmod__(other)[1]
+
+    def __rmod__(self, other):
+        # Naive implementation, room for optimization
+        return self.__rdivmod__(other)[1]
+
+    def __divmod__(self, other):
+        # Naive implementation, room for optimization
+        div = self // other
+        return div, self - div * other
+
+    def __rdivmod__(self, other):
+        # Naive implementation, room for optimization
+        div = other // self
+        return div, other - div * self
+
+
+def truediv_object_array(ndarray left, ndarray right):
+    cdef:
+        ndarray[object] result = np.empty((<object>left).shape, dtype=object)
+        object td64  # really timedelta64 if we find a way to declare that
+        object obj, res_value
+        _Timedelta td
+        Py_ssize_t i
+
+    for i in range(len(left)):
+        td64 = left[i]
+        obj = right[i]
+
+        if cnp.get_timedelta64_value(td64) == NPY_NAT:
+            # td here should be interpreted as a td64 NaT
+            if _should_cast_to_timedelta(obj):
+                res_value = np.nan
+            else:
+                # if its a number then let numpy handle division, otherwise
+                #  numpy will raise
+                res_value = td64 / obj
+        else:
+            td = Timedelta(td64)
+            res_value = td / obj
+
+        result[i] = res_value
+
+    return result
+
+
+def floordiv_object_array(ndarray left, ndarray right):
+    cdef:
+        ndarray[object] result = np.empty((<object>left).shape, dtype=object)
+        object td64  # really timedelta64 if we find a way to declare that
+        object obj, res_value
+        _Timedelta td
+        Py_ssize_t i
+
+    for i in range(len(left)):
+        td64 = left[i]
+        obj = right[i]
+
+        if cnp.get_timedelta64_value(td64) == NPY_NAT:
+            # td here should be interpreted as a td64 NaT
+            if _should_cast_to_timedelta(obj):
+                res_value = np.nan
+            else:
+                # if its a number then let numpy handle division, otherwise
+                #  numpy will raise
+                res_value = td64 // obj
+        else:
+            td = Timedelta(td64)
+            res_value = td // obj
+
+        result[i] = res_value
+
+    return result
+
+
+cdef bint is_any_td_scalar(object obj):
+    """
+    Cython equivalent for `isinstance(obj, (timedelta, np.timedelta64, Tick))`
+
+    Parameters
+    ----------
+    obj : object
+
+    Returns
+    -------
+    bool
+    """
+    return (
+        PyDelta_Check(obj) or cnp.is_timedelta64_object(obj) or is_tick_object(obj)
+    )
+
+
+cdef bint _should_cast_to_timedelta(object obj):
+    """
+    Should we treat this object as a Timedelta for the purpose of a binary op
+    """
+    return (
+        is_any_td_scalar(obj) or obj is None or obj is NaT or isinstance(obj, str)
+    )
+
+
+cpdef int64_t get_unit_for_round(freq, NPY_DATETIMEUNIT creso) except? -1:
+    from pandas._libs.tslibs.offsets import to_offset
+
+    freq = to_offset(freq)
+    freq.nanos  # raises on non-fixed freq
+    return delta_to_nanoseconds(freq, creso)
diff --git a/pandas/pandas/_libs/tslibs/timestamps.pxd b/pandas/pandas/_libs/tslibs/timestamps.pxd
new file mode 100644
index 0000000000000..c5ec92fabc7f8
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timestamps.pxd
@@ -0,0 +1,36 @@
+from cpython.datetime cimport (
+    datetime,
+    tzinfo,
+)
+from numpy cimport int64_t
+
+from pandas._libs.tslibs.base cimport ABCTimestamp
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    npy_datetimestruct,
+)
+from pandas._libs.tslibs.offsets cimport BaseOffset
+
+
+cdef _Timestamp create_timestamp_from_ts(int64_t value,
+                                         npy_datetimestruct dts,
+                                         tzinfo tz,
+                                         bint fold,
+                                         NPY_DATETIMEUNIT reso=*)
+
+
+cdef class _Timestamp(ABCTimestamp):
+    cdef readonly:
+        int64_t _value, _nanosecond, _year
+        NPY_DATETIMEUNIT _creso
+
+    cdef bint _get_start_end_field(self, str field, freq)
+    cdef _get_date_name_field(self, str field, object locale)
+    cdef int64_t _maybe_convert_value_to_local(self) except? -1
+    cdef bint _can_compare(self, datetime other)
+    cpdef to_datetime64(self)
+    cpdef datetime to_pydatetime(_Timestamp self, bint warn=*)
+    cdef bint _compare_outside_nanorange(_Timestamp self, datetime other,
+                                         int op) except -1
+    cdef bint _compare_mismatched_resos(_Timestamp self, _Timestamp other, int op)
+    cdef _Timestamp _as_creso(_Timestamp self, NPY_DATETIMEUNIT creso, bint round_ok=*)
diff --git a/pandas/pandas/_libs/tslibs/timestamps.pyi b/pandas/pandas/_libs/tslibs/timestamps.pyi
new file mode 100644
index 0000000000000..c769b09d1b7a1
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timestamps.pyi
@@ -0,0 +1,241 @@
+from datetime import (
+    date as _date,
+    datetime,
+    time as _time,
+    timedelta,
+    tzinfo as _tzinfo,
+)
+from time import struct_time
+from typing import (
+    ClassVar,
+    Literal,
+    TypeAlias,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import (
+    BaseOffset,
+    NaTType,
+    Period,
+    Tick,
+    Timedelta,
+)
+from pandas._typing import (
+    Self,
+    TimestampNonexistent,
+)
+
+_TimeZones: TypeAlias = str | _tzinfo | None | int
+
+def integer_op_not_supported(obj: object) -> TypeError: ...
+
+class Timestamp(datetime):
+    _creso: int
+    min: ClassVar[Timestamp]
+    max: ClassVar[Timestamp]
+
+    resolution: ClassVar[Timedelta]
+    _value: int  # np.int64
+    # error: "__new__" must return a class instance (got "Union[Timestamp, NaTType]")
+    def __new__(  # type: ignore[misc]
+        cls: type[Self],
+        ts_input: np.integer | float | str | _date | datetime | np.datetime64 = ...,
+        year: int | None = ...,
+        month: int | None = ...,
+        day: int | None = ...,
+        hour: int | None = ...,
+        minute: int | None = ...,
+        second: int | None = ...,
+        microsecond: int | None = ...,
+        tzinfo: _tzinfo | None = ...,
+        *,
+        nanosecond: int | None = ...,
+        tz: _TimeZones = ...,
+        unit: str | int | None = ...,
+        fold: int | None = ...,
+    ) -> Self | NaTType: ...
+    @classmethod
+    def _from_value_and_reso(
+        cls, value: int, reso: int, tz: _TimeZones
+    ) -> Timestamp: ...
+    @property
+    def value(self) -> int: ...  # np.int64
+    @property
+    def year(self) -> int: ...
+    @property
+    def month(self) -> int: ...
+    @property
+    def day(self) -> int: ...
+    @property
+    def hour(self) -> int: ...
+    @property
+    def minute(self) -> int: ...
+    @property
+    def second(self) -> int: ...
+    @property
+    def microsecond(self) -> int: ...
+    @property
+    def nanosecond(self) -> int: ...
+    @property
+    def tzinfo(self) -> _tzinfo | None: ...
+    @property
+    def tz(self) -> _tzinfo | None: ...
+    @property
+    def fold(self) -> int: ...
+    @classmethod
+    def fromtimestamp(cls, ts: float, tz: _TimeZones = ...) -> Self: ...
+    @classmethod
+    def utcfromtimestamp(cls, ts: float) -> Self: ...
+    @classmethod
+    def today(cls, tz: _TimeZones = ...) -> Self: ...
+    @classmethod
+    def fromordinal(
+        cls,
+        ordinal: int,
+        tz: _TimeZones = ...,
+    ) -> Self: ...
+    @classmethod
+    def now(cls, tz: _TimeZones = ...) -> Self: ...
+    @classmethod
+    def utcnow(cls) -> Self: ...
+    # error: Signature of "combine" incompatible with supertype "datetime"
+    @classmethod
+    def combine(  # type: ignore[override]
+        cls, date: _date, time: _time
+    ) -> datetime: ...
+    @classmethod
+    def fromisoformat(cls, date_string: str) -> Self: ...
+    def strftime(self, format: str) -> str: ...
+    def __format__(self, fmt: str) -> str: ...
+    def toordinal(self) -> int: ...
+    def timetuple(self) -> struct_time: ...
+    def timestamp(self) -> float: ...
+    def utctimetuple(self) -> struct_time: ...
+    def date(self) -> _date: ...
+    def time(self) -> _time: ...
+    def timetz(self) -> _time: ...
+    # LSP violation: nanosecond is not present in datetime.datetime.replace
+    # and has positional args following it
+    def replace(  # type: ignore[override]
+        self,
+        year: int | None = ...,
+        month: int | None = ...,
+        day: int | None = ...,
+        hour: int | None = ...,
+        minute: int | None = ...,
+        second: int | None = ...,
+        microsecond: int | None = ...,
+        nanosecond: int | None = ...,
+        tzinfo: _tzinfo | type[object] | None = ...,
+        fold: int | None = ...,
+    ) -> Self: ...
+    # LSP violation: datetime.datetime.astimezone has a default value for tz
+    def astimezone(self, tz: _TimeZones) -> Self: ...  # type: ignore[override]
+    def ctime(self) -> str: ...
+    def isoformat(self, sep: str = ..., timespec: str = ...) -> str: ...
+    @classmethod
+    def strptime(
+        # Note: strptime is actually disabled and raises NotImplementedError
+        cls,
+        date_string: str,
+        format: str,
+    ) -> Self: ...
+    def utcoffset(self) -> timedelta | None: ...
+    def tzname(self) -> str | None: ...
+    def dst(self) -> timedelta | None: ...
+    def __le__(self, other: datetime) -> bool: ...  # type: ignore[override]
+    def __lt__(self, other: datetime) -> bool: ...  # type: ignore[override]
+    def __ge__(self, other: datetime) -> bool: ...  # type: ignore[override]
+    def __gt__(self, other: datetime) -> bool: ...  # type: ignore[override]
+    # error: Signature of "__add__" incompatible with supertype "date"/"datetime"
+    @overload  # type: ignore[override]
+    def __add__(self, other: np.ndarray) -> np.ndarray: ...
+    @overload
+    def __add__(self, other: timedelta | np.timedelta64 | Tick) -> Self: ...
+    def __radd__(self, other: timedelta) -> Self: ...
+    @overload  # type: ignore[override]
+    def __sub__(self, other: datetime) -> Timedelta: ...
+    @overload
+    def __sub__(self, other: timedelta | np.timedelta64 | Tick) -> Self: ...
+    def __hash__(self) -> int: ...
+    def weekday(self) -> int: ...
+    def isoweekday(self) -> int: ...
+    # Return type "Tuple[int, int, int]" of "isocalendar" incompatible with return
+    # type "_IsoCalendarDate" in supertype "date"
+    def isocalendar(self) -> tuple[int, int, int]: ...  # type: ignore[override]
+    @property
+    def is_leap_year(self) -> bool: ...
+    @property
+    def is_month_start(self) -> bool: ...
+    @property
+    def is_quarter_start(self) -> bool: ...
+    @property
+    def is_year_start(self) -> bool: ...
+    @property
+    def is_month_end(self) -> bool: ...
+    @property
+    def is_quarter_end(self) -> bool: ...
+    @property
+    def is_year_end(self) -> bool: ...
+    def to_pydatetime(self, warn: bool = ...) -> datetime: ...
+    def to_datetime64(self) -> np.datetime64: ...
+    def to_period(self, freq: BaseOffset | str | None = None) -> Period: ...
+    def to_julian_date(self) -> np.float64: ...
+    @property
+    def asm8(self) -> np.datetime64: ...
+    def tz_convert(self, tz: _TimeZones) -> Self: ...
+    # TODO: could return NaT?
+    def tz_localize(
+        self,
+        tz: _TimeZones,
+        ambiguous: bool | Literal["raise", "NaT"] = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> Self: ...
+    def normalize(self) -> Self: ...
+    # TODO: round/floor/ceil could return NaT?
+    def round(
+        self,
+        freq: str,
+        ambiguous: bool | Literal["raise", "NaT"] = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> Self: ...
+    def floor(
+        self,
+        freq: str,
+        ambiguous: bool | Literal["raise", "NaT"] = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> Self: ...
+    def ceil(
+        self,
+        freq: str,
+        ambiguous: bool | Literal["raise", "NaT"] = ...,
+        nonexistent: TimestampNonexistent = ...,
+    ) -> Self: ...
+    def day_name(self, locale: str | None = ...) -> str: ...
+    def month_name(self, locale: str | None = ...) -> str: ...
+    @property
+    def day_of_week(self) -> int: ...
+    @property
+    def dayofweek(self) -> int: ...
+    @property
+    def day_of_year(self) -> int: ...
+    @property
+    def dayofyear(self) -> int: ...
+    @property
+    def quarter(self) -> int: ...
+    @property
+    def week(self) -> int: ...
+    def to_numpy(
+        self, dtype: np.dtype | None = ..., copy: bool = ...
+    ) -> np.datetime64: ...
+    @property
+    def _date_repr(self) -> str: ...
+    @property
+    def days_in_month(self) -> int: ...
+    @property
+    def daysinmonth(self) -> int: ...
+    @property
+    def unit(self) -> str: ...
+    def as_unit(self, unit: str, round_ok: bool = ...) -> Timestamp: ...
diff --git a/pandas/pandas/_libs/tslibs/timestamps.pyx b/pandas/pandas/_libs/tslibs/timestamps.pyx
new file mode 100644
index 0000000000000..6b4b90167e625
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timestamps.pyx
@@ -0,0 +1,3457 @@
+"""
+_Timestamp is a c-defined subclass of datetime.datetime
+
+_Timestamp is PITA. Because we inherit from datetime, which has very specific
+construction requirements, we need to do object instantiation in python
+(see Timestamp class below). This will serve as a C extension type that
+shadows the python class, where we do any heavy lifting.
+"""
+
+import warnings
+
+cimport cython
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    ndarray,
+    uint8_t,
+)
+
+cnp.import_array()
+
+from cpython.datetime cimport (  # alias tzinfo_type bc `tzinfo` is a kwarg below
+    PyDate_Check,
+    PyDateTime_Check,
+    PyDelta_Check,
+    PyTZInfo_Check,
+    datetime,
+    import_datetime,
+    time as dt_time,
+    tzinfo as tzinfo_type,
+)
+from cpython.object cimport (
+    Py_EQ,
+    Py_GE,
+    Py_GT,
+    Py_LE,
+    Py_LT,
+    Py_NE,
+    PyObject_RichCompare,
+    PyObject_RichCompareBool,
+)
+
+import_datetime()
+
+import datetime as dt
+
+from pandas._libs.tslibs cimport ccalendar
+from pandas._libs.tslibs.base cimport ABCTimestamp
+
+from pandas.util._decorators import set_module
+from pandas.util._exceptions import find_stack_level
+
+from pandas._libs.tslibs.conversion cimport (
+    _TSObject,
+    convert_datetime_to_tsobject,
+    convert_to_tsobject,
+    maybe_localize_tso,
+)
+from pandas._libs.tslibs.dtypes cimport (
+    npy_unit_to_abbrev,
+    npy_unit_to_attrname,
+    periods_per_day,
+    periods_per_second,
+)
+from pandas._libs.tslibs.util cimport (
+    is_array,
+    is_integer_object,
+)
+
+from pandas._libs.tslibs.fields import (
+    RoundTo,
+    get_date_name_field,
+    get_start_end_field,
+    round_nsint64,
+)
+
+from pandas._libs.tslibs.nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+)
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    cmp_dtstructs,
+    cmp_scalar,
+    convert_reso,
+    dts_to_iso_string,
+    get_datetime64_unit,
+    get_unit_from_dtype,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    npy_datetimestruct_to_datetime,
+    pandas_datetime_to_datetimestruct,
+    pydatetime_to_dtstruct,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.np_datetime import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+)
+
+from pandas._libs.tslibs.offsets cimport to_offset
+from pandas._libs.tslibs.timedeltas cimport (
+    _Timedelta,
+    get_unit_for_round,
+    is_any_td_scalar,
+)
+
+from pandas._libs.tslibs.timedeltas import Timedelta
+
+from pandas._libs.tslibs.timezones cimport (
+    get_timezone,
+    is_utc,
+    maybe_get_tz,
+    treat_tz_as_pytz,
+    utc_stdlib as UTC,
+)
+from pandas._libs.tslibs.tzconversion cimport (
+    tz_convert_from_utc_single,
+    tz_localize_to_utc_single,
+)
+
+# ----------------------------------------------------------------------
+# Constants
+_zero_time = dt_time(0, 0)
+_no_input = object()
+
+# ----------------------------------------------------------------------
+
+
+cdef _Timestamp create_timestamp_from_ts(
+    int64_t value,
+    npy_datetimestruct dts,
+    tzinfo tz,
+    bint fold,
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+):
+    """ convenience routine to construct a Timestamp from its parts """
+    cdef:
+        _Timestamp ts_base
+        int64_t pass_year = dts.year
+
+    # We pass year=1970/1972 here and set year below because with non-nanosecond
+    #  resolution we may have datetimes outside of the stdlib pydatetime
+    #  implementation bounds, which would raise.
+    # NB: this means the C-API macro PyDateTime_GET_YEAR is unreliable.
+    if 1 <= pass_year <= 9999:
+        # we are in-bounds for pydatetime
+        pass
+    elif ccalendar.is_leapyear(dts.year):
+        pass_year = 1972
+    else:
+        pass_year = 1970
+
+    ts_base = _Timestamp.__new__(Timestamp, pass_year, dts.month,
+                                 dts.day, dts.hour, dts.min,
+                                 dts.sec, dts.us, tz, fold=fold)
+
+    ts_base._value = value
+    ts_base._year = dts.year
+    ts_base._nanosecond = dts.ps // 1000
+    ts_base._creso = reso
+
+    return ts_base
+
+
+def _unpickle_timestamp(value, freq, tz, reso=NPY_FR_ns):
+    # GH#41949 dont warn on unpickle if we have a freq
+    ts = Timestamp._from_value_and_reso(value, reso, tz)
+    return ts
+
+
+# ----------------------------------------------------------------------
+
+def integer_op_not_supported(obj):
+    # GH#22535 add/sub of integers and int-arrays is no longer allowed
+    # Note we return rather than raise the exception so we can raise in
+    #  the caller; mypy finds this more palatable.
+    cls = type(obj).__name__
+
+    # GH#30886 using an fstring raises SystemError
+    int_addsub_msg = (
+        f"Addition/subtraction of integers and integer-arrays with {cls} is "
+        "no longer supported.  Instead of adding/subtracting `n`, "
+        "use `n * obj.freq`"
+    )
+    return TypeError(int_addsub_msg)
+
+
+class MinMaxReso:
+    """
+    We need to define min/max/resolution on both the Timestamp _instance_
+    and Timestamp class.  On an instance, these depend on the object's _reso.
+    On the class, we default to the values we would get with nanosecond _reso.
+
+    See also: timedeltas.MinMaxReso
+    """
+    def __init__(self, name):
+        self._name = name
+
+    def __get__(self, obj, type=None):
+        cls = Timestamp
+        if self._name == "min":
+            val = np.iinfo(np.int64).min + 1
+        elif self._name == "max":
+            val = np.iinfo(np.int64).max
+        else:
+            assert self._name == "resolution"
+            val = 1
+            cls = Timedelta
+
+        if obj is None:
+            # i.e. this is on the class, default to nanos
+            return cls(val)
+        elif self._name == "resolution":
+            return Timedelta._from_value_and_reso(val, obj._creso)
+        else:
+            return Timestamp._from_value_and_reso(val, obj._creso, tz=None)
+
+    def __set__(self, obj, value):
+        raise AttributeError(f"{self._name} is not settable.")
+
+
+# ----------------------------------------------------------------------
+
+cdef class _Timestamp(ABCTimestamp):
+
+    # higher than np.ndarray and np.matrix
+    __array_priority__ = 100
+    dayofweek = _Timestamp.day_of_week
+    dayofyear = _Timestamp.day_of_year
+
+    min = MinMaxReso("min")
+    max = MinMaxReso("max")
+    resolution = MinMaxReso("resolution")  # GH#21336, GH#21365
+
+    @property
+    def value(self) -> int:
+        """
+        Return the value of the Timestamp.
+
+        Returns
+        -------
+        int
+            The integer representation of the Timestamp object in nanoseconds
+            since the Unix epoch (1970-01-01 00:00:00 UTC).
+
+        See Also
+        --------
+        Timestamp.second : Return the second of the Timestamp.
+        Timestamp.minute : Return the minute of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.value
+        1725120990000000000
+        """
+
+        try:
+            return convert_reso(self._value, self._creso, NPY_FR_ns, False)
+        except OverflowError:
+            raise OverflowError(
+                "Cannot convert Timestamp to nanoseconds without overflow. "
+                "Use `.asm8.view('i8')` to cast represent Timestamp in its own "
+                f"unit (here, {self.unit})."
+            )
+
+    @property
+    def unit(self) -> str:
+        """
+        The abbreviation associated with self._creso.
+
+        This property returns a string representing the time unit of the Timestamp's
+        resolution. It corresponds to the smallest time unit that can be represented
+        by this Timestamp object. The possible values are:
+        - 's' (second)
+        - 'ms' (millisecond)
+        - 'us' (microsecond)
+        - 'ns' (nanosecond)
+
+        Returns
+        -------
+        str
+            A string abbreviation of the Timestamp's resolution unit:
+            - 's' for second
+            - 'ms' for millisecond
+            - 'us' for microsecond
+            - 'ns' for nanosecond
+
+        See Also
+        --------
+        Timestamp.resolution : Return resolution of the Timestamp.
+        Timedelta : A duration expressing the difference between two dates or times.
+
+        Examples
+        --------
+        >>> pd.Timestamp("2020-01-01 12:34:56").unit
+        's'
+
+        >>> pd.Timestamp("2020-01-01 12:34:56.123").unit
+        'ms'
+
+        >>> pd.Timestamp("2020-01-01 12:34:56.123456").unit
+        'us'
+
+        >>> pd.Timestamp("2020-01-01 12:34:56.123456789").unit
+        'ns'
+        """
+        return npy_unit_to_abbrev(self._creso)
+
+    # -----------------------------------------------------------------
+    # Constructors
+
+    @classmethod
+    def _from_value_and_reso(cls, int64_t value, NPY_DATETIMEUNIT reso, tzinfo tz):
+        cdef:
+            _TSObject obj = _TSObject()
+
+        if value == NPY_NAT:
+            return NaT
+
+        if reso < NPY_DATETIMEUNIT.NPY_FR_s or reso > NPY_DATETIMEUNIT.NPY_FR_ns:
+            raise NotImplementedError(
+                "Only resolutions 's', 'ms', 'us', 'ns' are supported."
+            )
+
+        obj.value = value
+        obj.creso = reso
+        pandas_datetime_to_datetimestruct(value, reso, &obj.dts)
+        maybe_localize_tso(obj, tz, reso)
+
+        return create_timestamp_from_ts(
+            value, obj.dts, tz=obj.tzinfo, fold=obj.fold, reso=reso
+        )
+
+    @classmethod
+    def _from_dt64(cls, dt64: np.datetime64):
+        # construct a Timestamp from a np.datetime64 object, keeping the
+        #  resolution of the input.
+        # This is here mainly so we can incrementally implement non-nano
+        #  (e.g. only tznaive at first)
+        cdef:
+            int64_t value
+            NPY_DATETIMEUNIT reso
+
+        reso = get_datetime64_unit(dt64)
+        value = cnp.get_datetime64_value(dt64)
+        return cls._from_value_and_reso(value, reso, None)
+
+    # -----------------------------------------------------------------
+
+    def __hash__(_Timestamp self):
+        if self._nanosecond:
+            return hash(self._value)
+        if not (1 <= self._year <= 9999):
+            # out of bounds for pydatetime
+            return hash(self._value)
+        if self.fold:
+            return datetime.__hash__(self.replace(fold=0))
+        return datetime.__hash__(self)
+
+    def __richcmp__(_Timestamp self, object other, int op):
+        cdef:
+            _Timestamp ots
+
+        if isinstance(other, _Timestamp):
+            ots = other
+        elif other is NaT:
+            return op == Py_NE
+        elif cnp.is_datetime64_object(other):
+            ots = Timestamp(other)
+        elif PyDateTime_Check(other):
+            if self._nanosecond == 0:
+                val = self.to_pydatetime()
+                return PyObject_RichCompareBool(val, other, op)
+
+            try:
+                ots = type(self)(other)
+            except ValueError:
+                return self._compare_outside_nanorange(other, op)
+
+        elif is_array(other):
+            # avoid recursion error GH#15183
+            if other.dtype.kind == "M":
+                if self.tz is None:
+                    return PyObject_RichCompare(self.asm8, other, op)
+                elif op == Py_NE:
+                    return np.ones(other.shape, dtype=np.bool_)
+                elif op == Py_EQ:
+                    return np.zeros(other.shape, dtype=np.bool_)
+                raise TypeError(
+                    "Cannot compare tz-naive and tz-aware timestamps"
+                )
+            elif other.dtype.kind == "O":
+                # Operate element-wise
+                return np.array(
+                    [PyObject_RichCompare(self, x, op) for x in other],
+                    dtype=bool,
+                )
+            elif op == Py_NE:
+                return np.ones(other.shape, dtype=np.bool_)
+            elif op == Py_EQ:
+                return np.zeros(other.shape, dtype=np.bool_)
+            return NotImplemented
+
+        elif PyDate_Check(other):
+            # returning NotImplemented defers to the `date` implementation
+            #  which incorrectly drops tz and normalizes to midnight
+            #  before comparing
+            # We follow the stdlib datetime behavior of never being equal
+            if op == Py_EQ:
+                return False
+            elif op == Py_NE:
+                return True
+            raise TypeError(
+                "Cannot compare Timestamp with datetime.date. "
+                "Use ts == pd.Timestamp(date) or ts.date() == date instead."
+            )
+        else:
+            return NotImplemented
+
+        if not self._can_compare(ots):
+            if op == Py_NE or op == Py_EQ:
+                return NotImplemented
+            raise TypeError(
+                "Cannot compare tz-naive and tz-aware timestamps"
+            )
+        if self._creso == ots._creso:
+            return cmp_scalar(self._value, ots._value, op)
+        return self._compare_mismatched_resos(ots, op)
+
+    # TODO: copied from Timedelta; try to de-duplicate
+    cdef bint _compare_mismatched_resos(self, _Timestamp other, int op):
+        # Can't just dispatch to numpy as they silently overflow and get it wrong
+        cdef:
+            npy_datetimestruct dts_self
+            npy_datetimestruct dts_other
+
+        # dispatch to the datetimestruct utils instead of writing new ones!
+        pandas_datetime_to_datetimestruct(self._value, self._creso, &dts_self)
+        pandas_datetime_to_datetimestruct(other._value, other._creso, &dts_other)
+        return cmp_dtstructs(&dts_self,  &dts_other, op)
+
+    cdef bint _compare_outside_nanorange(_Timestamp self, datetime other,
+                                         int op) except -1:
+        cdef:
+            datetime dtval = self.to_pydatetime(warn=False)
+
+        if not self._can_compare(other):
+            return NotImplemented
+
+        if self._nanosecond == 0:
+            return PyObject_RichCompareBool(dtval, other, op)
+
+        # otherwise we have dtval < self
+        if op == Py_NE:
+            return True
+        if op == Py_EQ:
+            return False
+        if op == Py_LE or op == Py_LT:
+            return self._year <= other.year
+        if op == Py_GE or op == Py_GT:
+            return self._year >= other.year
+
+    cdef bint _can_compare(self, datetime other):
+        if self.tzinfo is not None:
+            return other.tzinfo is not None
+        return other.tzinfo is None
+
+    @cython.overflowcheck(True)
+    def __add__(self, other):
+        cdef:
+            int64_t nanos = 0
+
+        if is_any_td_scalar(other):
+            other = Timedelta(other)
+
+            # TODO: share this with __sub__, Timedelta.__add__
+            # Matching numpy, we cast to the higher resolution. Unlike numpy,
+            #  we raise instead of silently overflowing during this casting.
+            if self._creso < other._creso:
+                self = (<_Timestamp>self)._as_creso(other._creso, round_ok=True)
+            elif self._creso > other._creso:
+                other = (<_Timedelta>other)._as_creso(self._creso, round_ok=True)
+
+            nanos = other._value
+
+            try:
+                new_value = self._value + nanos
+                result = type(self)._from_value_and_reso(
+                    new_value, reso=self._creso, tz=self.tzinfo
+                )
+            except OverflowError as err:
+                new_value = int(self._value) + int(nanos)
+                attrname = npy_unit_to_attrname[self._creso]
+                raise OutOfBoundsDatetime(
+                    f"Out of bounds {attrname} timestamp: {new_value}"
+                ) from err
+
+            return result
+
+        elif is_integer_object(other):
+            raise integer_op_not_supported(self)
+
+        elif is_array(other):
+            if other.dtype.kind in "iu":
+                raise integer_op_not_supported(self)
+            if other.dtype.kind == "m":
+                if self.tz is None:
+                    return self.asm8 + other
+                return np.asarray(
+                    [self + other[n] for n in range(len(other))],
+                    dtype=object,
+                )
+
+        return NotImplemented
+
+    def __radd__(self, other):
+        # Have to duplicate checks to avoid infinite recursion due to NotImplemented
+        if is_any_td_scalar(other) or is_integer_object(other) or is_array(other):
+            return self.__add__(other)
+        return NotImplemented
+
+    def __sub__(self, other):
+        if other is NaT:
+            return NaT
+
+        elif is_any_td_scalar(other) or is_integer_object(other):
+            neg_other = -other
+            return self + neg_other
+
+        elif is_array(other):
+            if other.dtype.kind in "iu":
+                raise integer_op_not_supported(self)
+            if other.dtype.kind == "m":
+                if self.tz is None:
+                    return self.asm8 - other
+                return np.asarray(
+                    [self - other[n] for n in range(len(other))],
+                    dtype=object,
+                )
+            return NotImplemented
+
+        # coerce if necessary if we are a Timestamp-like
+        if PyDateTime_Check(other) or cnp.is_datetime64_object(other):
+            # both_timestamps is to determine whether Timedelta(self - other)
+            # should raise the OOB error, or fall back returning a timedelta.
+            both_timestamps = isinstance(other, _Timestamp)
+            other = type(self)(other)
+
+            if (self.tzinfo is None) ^ (other.tzinfo is None):
+                raise TypeError(
+                    "Cannot subtract tz-naive and tz-aware datetime-like objects."
+                )
+
+            # Matching numpy, we cast to the higher resolution. Unlike numpy,
+            #  we raise instead of silently overflowing during this casting.
+            if self._creso < other._creso:
+                self = (<_Timestamp>self)._as_creso(other._creso, round_ok=True)
+            elif self._creso > other._creso:
+                other = (<_Timestamp>other)._as_creso(self._creso, round_ok=True)
+
+            # scalar Timestamp/datetime - Timestamp/datetime -> yields a
+            # Timedelta
+            try:
+                res_value = self._value - other._value
+                return Timedelta._from_value_and_reso(res_value, self._creso)
+            except (OverflowError, OutOfBoundsDatetime, OutOfBoundsTimedelta) as err:
+                if both_timestamps:
+                    raise OutOfBoundsDatetime(
+                        "Result is too large for pandas.Timedelta. Convert inputs "
+                        "to datetime.datetime with 'Timestamp.to_pydatetime()' "
+                        "before subtracting."
+                    ) from err
+                # We get here in stata tests, fall back to stdlib datetime
+                #  method and return stdlib timedelta object
+                pass
+
+        return NotImplemented
+
+    def __rsub__(self, other):
+        if PyDateTime_Check(other):
+            try:
+                return type(self)(other) - self
+            except (OverflowError, OutOfBoundsDatetime) as err:
+                # We get here in stata tests, fall back to stdlib datetime
+                #  method and return stdlib timedelta object
+                pass
+        elif cnp.is_datetime64_object(other):
+            return type(self)(other) - self
+        return NotImplemented
+
+    # -----------------------------------------------------------------
+
+    cdef int64_t _maybe_convert_value_to_local(self) except? -1:
+        """Convert UTC i8 value to local i8 value if tz exists"""
+        cdef:
+            int64_t val
+            tzinfo own_tz = self.tzinfo
+            npy_datetimestruct dts
+
+        if own_tz is not None and not is_utc(own_tz):
+            pydatetime_to_dtstruct(self, &dts)
+            val = npy_datetimestruct_to_datetime(self._creso, &dts) + self._nanosecond
+        else:
+            val = self._value
+        return val
+
+    @cython.boundscheck(False)
+    cdef bint _get_start_end_field(self, str field, freq):
+        cdef:
+            int64_t val
+            dict kwds
+            ndarray[uint8_t, cast=True] out
+            int month_kw
+
+        if freq:
+            kwds = freq.kwds
+            month_kw = kwds.get("startingMonth", kwds.get("month", 12))
+            freq_name = freq.name
+        else:
+            month_kw = 12
+            freq_name = None
+
+        val = self._maybe_convert_value_to_local()
+
+        out = get_start_end_field(np.array([val], dtype=np.int64),
+                                  field, freq_name, month_kw, self._creso)
+        return out[0]
+
+    @property
+    def is_month_start(self) -> bool:
+        """
+        Check if the date is the first day of the month.
+
+        Returns
+        -------
+        bool
+            True if the date is the first day of the month.
+
+        See Also
+        --------
+        Timestamp.is_month_end : Similar property indicating the last day of the month.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_month_start
+        False
+
+        >>> ts = pd.Timestamp(2020, 1, 1)
+        >>> ts.is_month_start
+        True
+        """
+        return self.day == 1
+
+    @property
+    def is_month_end(self) -> bool:
+        """
+        Check if the date is the last day of the month.
+
+        Returns
+        -------
+        bool
+            True if the date is the last day of the month.
+
+        See Also
+        --------
+        Timestamp.is_month_start : Similar property indicating month start.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_month_end
+        False
+
+        >>> ts = pd.Timestamp(2020, 12, 31)
+        >>> ts.is_month_end
+        True
+        """
+        return self.day == self.days_in_month
+
+    @property
+    def is_quarter_start(self) -> bool:
+        """
+        Check if the date is the first day of the quarter.
+
+        Returns
+        -------
+        bool
+            True if date is first day of the quarter.
+
+        See Also
+        --------
+        Timestamp.is_quarter_end : Similar property indicating the quarter end.
+        Timestamp.quarter : Return the quarter of the date.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_quarter_start
+        False
+
+        >>> ts = pd.Timestamp(2020, 4, 1)
+        >>> ts.is_quarter_start
+        True
+        """
+        return self.day == 1 and self.month % 3 == 1
+
+    @property
+    def is_quarter_end(self) -> bool:
+        """
+        Check if date is last day of the quarter.
+
+        Returns
+        -------
+        bool
+            True if date is last day of the quarter.
+
+        See Also
+        --------
+        Timestamp.is_quarter_start : Similar property indicating the quarter start.
+        Timestamp.quarter : Return the quarter of the date.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_quarter_end
+        False
+
+        >>> ts = pd.Timestamp(2020, 3, 31)
+        >>> ts.is_quarter_end
+        True
+        """
+        return (self.month % 3) == 0 and self.day == self.days_in_month
+
+    @property
+    def is_year_start(self) -> bool:
+        """
+        Return True if date is first day of the year.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Timestamp.is_year_end : Similar property indicating the end of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_year_start
+        False
+
+        >>> ts = pd.Timestamp(2020, 1, 1)
+        >>> ts.is_year_start
+        True
+        """
+        return self.day == self.month == 1
+
+    @property
+    def is_year_end(self) -> bool:
+        """
+        Return True if date is last day of the year.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Timestamp.is_year_start : Similar property indicating the start of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_year_end
+        False
+
+        >>> ts = pd.Timestamp(2020, 12, 31)
+        >>> ts.is_year_end
+        True
+        """
+        return self.month == 12 and self.day == 31
+
+    @cython.boundscheck(False)
+    cdef _get_date_name_field(self, str field, object locale):
+        cdef:
+            int64_t val
+            object[::1] out
+
+        val = self._maybe_convert_value_to_local()
+
+        out = get_date_name_field(np.array([val], dtype=np.int64),
+                                  field, locale=locale, reso=self._creso)
+        return out[0]
+
+    def day_name(self, locale=None) -> str:
+        """
+        Return the day name of the Timestamp with specified locale.
+
+        Parameters
+        ----------
+        locale : str, default None (English locale)
+            Locale determining the language in which to return the day name.
+
+        Returns
+        -------
+        str
+
+        See Also
+        --------
+        Timestamp.day_of_week : Return day of the week.
+        Timestamp.day_of_year : Return day of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.day_name()
+        'Saturday'
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.day_name()
+        nan
+        """
+        return self._get_date_name_field("day_name", locale)
+
+    def month_name(self, locale=None) -> str:
+        """
+        Return the month name of the Timestamp with specified locale.
+
+        This method returns the full name of the month corresponding to the
+        `Timestamp`, such as 'January', 'February', etc. The month name can
+        be returned in a specified locale if provided; otherwise, it defaults
+        to the English locale.
+
+        Parameters
+        ----------
+        locale : str, default None (English locale)
+            Locale determining the language in which to return the month name.
+
+        Returns
+        -------
+        str
+            The full month name as a string.
+
+        See Also
+        --------
+        Timestamp.day_name : Returns the name of the day of the week.
+        Timestamp.strftime : Returns a formatted string of the Timestamp.
+        datetime.datetime.strftime : Returns a string representing the date and time.
+
+        Examples
+        --------
+        Get the month name in English (default):
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.month_name()
+        'March'
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.month_name()
+        nan
+        """
+        return self._get_date_name_field("month_name", locale)
+
+    @property
+    def is_leap_year(self) -> bool:
+        """
+        Return True if year is a leap year.
+
+        A leap year is a year, which has 366 days (instead of 365) including 29th of
+        February as an intercalary day. Leap years are years which are multiples of
+        four with the exception of years divisible by 100 but not by 400.
+
+        Returns
+        -------
+        bool
+            True if year is a leap year, else False
+
+        See Also
+        --------
+        Period.is_leap_year : Return True if the period’s year is in a leap year.
+        DatetimeIndex.is_leap_year : Boolean indicator if the date belongs to a
+            leap year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.is_leap_year
+        True
+        """
+        return bool(ccalendar.is_leapyear(self._year))
+
+    @property
+    def day_of_week(self) -> int:
+        """
+        Return day of the week.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timestamp.isoweekday : Return the ISO day of the week represented by the date.
+        Timestamp.weekday : Return the day of the week represented by the date.
+        Timestamp.day_of_year : Return day of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.day_of_week
+        5
+        """
+        return self.weekday()
+
+    @property
+    def day_of_year(self) -> int:
+        """
+        Return the day of the year.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timestamp.day_of_week : Return day of the week.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.day_of_year
+        74
+        """
+        return ccalendar.get_day_of_year(self._year, self.month, self.day)
+
+    @property
+    def quarter(self) -> int:
+        """
+        Return the quarter of the year for the `Timestamp`.
+
+        This property returns an integer representing the quarter of the year in
+        which the `Timestamp` falls. The quarters are defined as follows:
+        - Q1: January 1 to March 31
+        - Q2: April 1 to June 30
+        - Q3: July 1 to September 30
+        - Q4: October 1 to December 31
+
+        Returns
+        -------
+        int
+            The quarter of the year (1 through 4).
+
+        See Also
+        --------
+        Timestamp.month : Returns the month of the `Timestamp`.
+        Timestamp.year : Returns the year of the `Timestamp`.
+
+        Examples
+        --------
+        Get the quarter for a `Timestamp`:
+
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.quarter
+        1
+
+        For a `Timestamp` in the fourth quarter:
+
+        >>> ts = pd.Timestamp(2020, 10, 14)
+        >>> ts.quarter
+        4
+        """
+        return ((self.month - 1) // 3) + 1
+
+    @property
+    def day(self) -> int:
+        """
+        Return the day of the Timestamp.
+
+        Returns
+        -------
+        int
+            The day of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.week : Return the week number of the year.
+        Timestamp.weekday : Return the day of the week.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.day
+        31
+        """
+        return super().day
+
+    @property
+    def fold(self) -> int:
+        """
+        Return the fold value of the Timestamp.
+
+        Returns
+        -------
+        int
+            The fold value of the Timestamp, where 0 indicates the first occurrence
+            of the ambiguous time, and 1 indicates the second.
+
+        See Also
+        --------
+        Timestamp.dst : Return the daylight saving time (DST) adjustment.
+        Timestamp.tzinfo : Return the timezone information associated.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-11-03 01:30:00")
+        >>> ts.fold
+        0
+        """
+        return super().fold
+
+    @property
+    def year(self) -> int:
+        """
+        Return the year of the Timestamp.
+
+        Returns
+        -------
+        int
+            The year of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.month : Return the month of the Timestamp.
+        Timestamp.day : Return the day of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.year
+        2024
+        """
+        return self._year
+
+    @property
+    def month(self) -> int:
+        """
+        Return the month of the Timestamp.
+
+        Returns
+        -------
+        int
+            The month of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.day : Return the day of the Timestamp.
+        Timestamp.year : Return the year of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.month
+        8
+        """
+        return super().month
+
+    @property
+    def hour(self) -> int:
+        """
+        Return the hour of the Timestamp.
+
+        Returns
+        -------
+        int
+            The hour of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.minute : Return the minute of the Timestamp.
+        Timestamp.second : Return the second of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.hour
+        16
+        """
+        return super().hour
+
+    @property
+    def minute(self) -> int:
+        """
+        Return the minute of the Timestamp.
+
+        Returns
+        -------
+        int
+            The minute of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.hour : Return the hour of the Timestamp.
+        Timestamp.second : Return the second of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.minute
+        16
+        """
+        return super().minute
+
+    @property
+    def second(self) -> int:
+        """
+        Return the second of the Timestamp.
+
+        Returns
+        -------
+        int
+            The second of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.microsecond : Return the microsecond of the Timestamp.
+        Timestamp.minute : Return the minute of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30")
+        >>> ts.second
+        30
+        """
+        return super().second
+
+    @property
+    def microsecond(self) -> int:
+        """
+        Return the microsecond of the Timestamp.
+
+        Returns
+        -------
+        int
+            The microsecond of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.second : Return the second of the Timestamp.
+        Timestamp.minute : Return the minute of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30.2304")
+        >>> ts.microsecond
+        230400
+        """
+        return super().microsecond
+
+    @property
+    def nanosecond(self) -> int:
+        """
+        Return the nanosecond of the Timestamp.
+
+        Returns
+        -------
+        int
+            The nanosecond of the Timestamp.
+
+        See Also
+        --------
+        Timestamp.second : Return the second of the Timestamp.
+        Timestamp.microsecond : Return the microsecond of the Timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2024-08-31 16:16:30.230400015")
+        >>> ts.nanosecond
+        15
+        """
+        return self._nanosecond
+
+    @property
+    def week(self) -> int:
+        """
+        Return the week number of the year.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timestamp.weekday : Return the day of the week.
+        Timestamp.quarter : Return the quarter of the year.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.week
+        11
+        """
+        return ccalendar.get_week_of_year(self._year, self.month, self.day)
+
+    @property
+    def days_in_month(self) -> int:
+        """
+        Return the number of days in the month.
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        Timestamp.month_name : Return the month name of the Timestamp with
+            specified locale.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14)
+        >>> ts.days_in_month
+        31
+        """
+        return ccalendar.get_days_in_month(self._year, self.month)
+
+    # -----------------------------------------------------------------
+    # Transformation Methods
+
+    def normalize(self) -> "Timestamp":
+        """
+        Normalize Timestamp to midnight, preserving tz information.
+
+        This method sets the time component of the `Timestamp` to midnight (00:00:00),
+        while preserving the date and time zone information. It is useful when you
+        need to standardize the time across different `Timestamp` objects without
+        altering the time zone or the date.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Timestamp.floor : Rounds `Timestamp` down to the nearest frequency.
+        Timestamp.ceil : Rounds `Timestamp` up to the nearest frequency.
+        Timestamp.round : Rounds `Timestamp` to the nearest frequency.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14, 15, 30)
+        >>> ts.normalize()
+        Timestamp('2020-03-14 00:00:00')
+        """
+        cdef:
+            local_val = self._maybe_convert_value_to_local()
+            int64_t normalized
+            int64_t ppd = periods_per_day(self._creso)
+            _Timestamp ts
+
+        normalized = normalize_i8_stamp(local_val, ppd)
+        ts = type(self)._from_value_and_reso(normalized, reso=self._creso, tz=None)
+        return ts.tz_localize(self.tzinfo)
+
+    # -----------------------------------------------------------------
+    # Pickle Methods
+
+    def __reduce_ex__(self, protocol):
+        # python 3.6 compat
+        # https://bugs.python.org/issue28730
+        # now __reduce_ex__ is defined and higher priority than __reduce__
+        return self.__reduce__()
+
+    def __setstate__(self, state):
+        self._value= state[0]
+        self.tzinfo = state[2]
+
+        if len(state) == 3:
+            # pre-non-nano pickle
+            # TODO: no tests get here 2022-05-10
+            reso = NPY_FR_ns
+        else:
+            reso = state[4]
+        self._creso = reso
+
+    def __reduce__(self):
+        object_state = self._value, None, self.tzinfo, self._creso
+        return (_unpickle_timestamp, object_state)
+
+    # -----------------------------------------------------------------
+    # Rendering Methods
+
+    def isoformat(self, sep: str = "T", timespec: str = "auto") -> str:
+        """
+        Return the time formatted according to ISO 8601.
+
+        The full format looks like 'YYYY-MM-DD HH:MM:SS.mmmmmmnnn'.
+        By default, the fractional part is omitted if self.microsecond == 0
+        and self._nanosecond == 0.
+
+        If self.tzinfo is not None, the UTC offset is also attached, giving
+        giving a full format of 'YYYY-MM-DD HH:MM:SS.mmmmmmnnn+HH:MM'.
+
+        Parameters
+        ----------
+        sep : str, default 'T'
+            String used as the separator between the date and time.
+
+        timespec : str, default 'auto'
+            Specifies the number of additional terms of the time to include.
+            The valid values are 'auto', 'hours', 'minutes', 'seconds',
+            'milliseconds', 'microseconds', and 'nanoseconds'.
+
+        Returns
+        -------
+        str
+
+        See Also
+        --------
+        Timestamp.strftime : Return a formatted string.
+        Timestamp.isocalendar : Return a tuple containing ISO year, week number and
+            weekday.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.isoformat()
+        '2020-03-14T15:32:52.192548651'
+        >>> ts.isoformat(timespec='microseconds')
+        '2020-03-14T15:32:52.192548'
+        """
+        base_ts = "microseconds" if timespec == "nanoseconds" else timespec
+        base = super(_Timestamp, self).isoformat(sep=sep, timespec=base_ts)
+        # We need to replace the fake year 1970 with our real year
+        base = f"{self._year:04d}-" + base.split("-", 1)[1]
+
+        if self._nanosecond == 0 and timespec != "nanoseconds":
+            return base
+
+        if self.tzinfo is not None:
+            base1, base2 = base[:-6], base[-6:]
+        else:
+            base1, base2 = base, ""
+
+        if timespec == "nanoseconds" or (timespec == "auto" and self._nanosecond):
+            if self.microsecond or timespec == "nanoseconds":
+                base1 += f"{self._nanosecond:03d}"
+            else:
+                base1 += f".{self._nanosecond:09d}"
+
+        return base1 + base2
+
+    def __repr__(self) -> str:
+        stamp = self._repr_base
+        zone = None
+
+        if self.tzinfo is not None:
+            try:
+                stamp += self.strftime("%z")
+            except ValueError:
+                year2000 = self.replace(year=2000)
+                stamp += year2000.strftime("%z")
+
+            zone = get_timezone(self.tzinfo)
+            try:
+                stamp += zone.strftime(" %%Z")
+            except AttributeError:
+                # e.g. tzlocal has no `strftime`
+                pass
+
+        tz = f", tz='{zone}'" if zone is not None else ""
+
+        return f"Timestamp('{stamp}'{tz})"
+
+    @property
+    def _repr_base(self) -> str:
+        return f"{self._date_repr} {self._time_repr}"
+
+    @property
+    def _date_repr(self) -> str:
+        # Ideal here would be self.strftime("%Y-%m-%d"), but
+        # the datetime strftime() methods require year >= 1900 and is slower
+        return f"{self._year}-{self.month:02d}-{self.day:02d}"
+
+    @property
+    def _time_repr(self) -> str:
+        result = f"{self.hour:02d}:{self.minute:02d}:{self.second:02d}"
+
+        if self._nanosecond != 0:
+            result += f".{self._nanosecond + 1000 * self.microsecond:09d}"
+        elif self.microsecond != 0:
+            result += f".{self.microsecond:06d}"
+
+        return result
+
+    # -----------------------------------------------------------------
+    # Conversion Methods
+
+    @cython.cdivision(False)
+    cdef _Timestamp _as_creso(self, NPY_DATETIMEUNIT creso, bint round_ok=True):
+        cdef:
+            int64_t value
+
+        if creso == self._creso:
+            return self
+
+        try:
+            value = convert_reso(self._value, self._creso, creso, round_ok=round_ok)
+        except OverflowError as err:
+            unit = npy_unit_to_abbrev(creso)
+            raise OutOfBoundsDatetime(
+                f"Cannot cast {self} to unit='{unit}' without overflow."
+            ) from err
+
+        return type(self)._from_value_and_reso(value, reso=creso, tz=self.tzinfo)
+
+    def as_unit(self, str unit, bint round_ok=True):
+        """
+        Convert the underlying int64 representation to the given unit.
+
+        Parameters
+        ----------
+        unit : {"ns", "us", "ms", "s"}
+        round_ok : bool, default True
+            If False and the conversion requires rounding, raise.
+
+        Returns
+        -------
+        Timestamp
+
+        See Also
+        --------
+        Timestamp.asm8 : Return numpy datetime64 format with same precision.
+        Timestamp.to_pydatetime : Convert Timestamp object to a native
+            Python datetime object.
+        to_timedelta : Convert argument into timedelta object,
+            which can represent differences in times.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 00:00:00.01')
+        >>> ts
+        Timestamp('2023-01-01 00:00:00.010000')
+        >>> ts.unit
+        'ms'
+        >>> ts = ts.as_unit('s')
+        >>> ts
+        Timestamp('2023-01-01 00:00:00')
+        >>> ts.unit
+        's'
+        """
+        dtype = np.dtype(f"M8[{unit}]")
+        reso = get_unit_from_dtype(dtype)
+        try:
+            return self._as_creso(reso, round_ok=round_ok)
+        except OverflowError as err:
+            raise OutOfBoundsDatetime(
+                f"Cannot cast {self} to unit='{unit}' without overflow."
+            ) from err
+
+    @property
+    def asm8(self) -> np.datetime64:
+        """
+        Return numpy datetime64 format with same precision.
+
+        See Also
+        --------
+        numpy.datetime64 : Numpy datatype for dates and times with high precision.
+        Timestamp.to_numpy : Convert the Timestamp to a NumPy datetime64.
+        to_datetime : Convert argument to datetime.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(2020, 3, 14, 15)
+        >>> ts.asm8
+        numpy.datetime64('2020-03-14T15:00:00.000000')
+        """
+        return self.to_datetime64()
+
+    def timestamp(self):
+        """
+        Return POSIX timestamp as float.
+
+        This method converts the `Timestamp` object to a POSIX timestamp, which is
+        the number of seconds since the Unix epoch (January 1, 1970). The returned
+        value is a floating-point number, where the integer part represents the
+        seconds, and the fractional part represents the microseconds.
+
+        Returns
+        -------
+        float
+            The POSIX timestamp representation of the `Timestamp` object.
+
+        See Also
+        --------
+        Timestamp.fromtimestamp : Construct a `Timestamp` from a POSIX timestamp.
+        datetime.datetime.timestamp : Equivalent method from the `datetime` module.
+        Timestamp.to_pydatetime : Convert the `Timestamp` to a `datetime` object.
+        Timestamp.to_datetime64 : Converts `Timestamp` to `numpy.datetime64`.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548')
+        >>> ts.timestamp()
+        1584199972.192548
+        """
+        # GH 17329
+        # Note: Naive timestamps will not match datetime.stdlib
+
+        denom = periods_per_second(self._creso)
+
+        return round(self._value/ denom, 6)
+
+    cpdef datetime to_pydatetime(_Timestamp self, bint warn=True):
+        """
+        Convert a Timestamp object to a native Python datetime object.
+
+        This method is useful for when you need to utilize a pandas Timestamp
+        object in contexts where native Python datetime objects are expected
+        or required. The conversion discards the nanoseconds component, and a
+        warning can be issued in such cases if desired.
+
+        Parameters
+        ----------
+        warn : bool, default True
+            If True, issues a warning when the timestamp includes nonzero
+            nanoseconds, as these will be discarded during the conversion.
+
+        Returns
+        -------
+        datetime.datetime or NaT
+            Returns a datetime.datetime object representing the timestamp,
+            with year, month, day, hour, minute, second, and microsecond components.
+            If the timestamp is NaT (Not a Time), returns NaT.
+
+        See Also
+        --------
+        datetime.datetime : The standard Python datetime class that this method
+            returns.
+        Timestamp.timestamp : Convert a Timestamp object to POSIX timestamp.
+        Timestamp.to_datetime64 : Convert a Timestamp object to numpy.datetime64.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548')
+        >>> ts.to_pydatetime()
+        datetime.datetime(2020, 3, 14, 15, 32, 52, 192548)
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.to_pydatetime()
+        NaT
+        """
+        if self._nanosecond != 0 and warn:
+            warnings.warn("Discarding nonzero nanoseconds in conversion.",
+                          UserWarning, stacklevel=find_stack_level())
+
+        return datetime(self._year, self.month, self.day,
+                        self.hour, self.minute, self.second,
+                        self.microsecond, self.tzinfo, fold=self.fold)
+
+    cpdef to_datetime64(self):
+        """
+        Return a NumPy datetime64 object with same precision.
+
+        This method returns a numpy.datetime64 object with the same
+        date and time information and precision as the pd.Timestamp object.
+
+        See Also
+        --------
+        numpy.datetime64 : Class to represent dates and times with high precision.
+        Timestamp.to_numpy : Alias for this method.
+        Timestamp.asm8 : Alias for this method.
+        pd.to_datetime : Convert argument to datetime.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(year=2023, month=1, day=1,
+        ...                   hour=10, second=15)
+        >>> ts
+        Timestamp('2023-01-01 10:00:15')
+        >>> ts.to_datetime64()
+        numpy.datetime64('2023-01-01T10:00:15.000000')
+        """
+        # TODO: find a way to construct dt64 directly from _reso
+        abbrev = npy_unit_to_abbrev(self._creso)
+        return np.datetime64(self._value, abbrev)
+
+    def to_numpy(self, dtype=None, copy=False) -> np.datetime64:
+        """
+        Convert the Timestamp to a NumPy datetime64.
+
+        This is an alias method for `Timestamp.to_datetime64()`. The dtype and
+        copy parameters are available here only for compatibility. Their values
+        will not affect the return value.
+
+        Parameters
+        ----------
+        dtype : dtype, optional
+            Data type of the output, ignored in this method as the return type
+            is always `numpy.datetime64`.
+        copy : bool, default False
+            Whether to ensure that the returned value is a new object. This
+            parameter is also ignored as the method does not support copying.
+
+        Returns
+        -------
+        numpy.datetime64
+
+        See Also
+        --------
+        DatetimeIndex.to_numpy : Similar method for DatetimeIndex.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.to_numpy()
+        numpy.datetime64('2020-03-14T15:32:52.192548651')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.to_numpy()
+        numpy.datetime64('NaT')
+        """
+        if dtype is not None or copy is not False:
+            raise ValueError(
+                "Timestamp.to_numpy dtype and copy arguments are ignored."
+            )
+        return self.to_datetime64()
+
+    def to_period(self, freq=None):
+        """
+        Return an period of which this timestamp is an observation.
+
+        This method converts the given Timestamp to a Period object,
+        which represents a span of time,such as a year, month, etc.,
+        based on the specified frequency.
+
+        Parameters
+        ----------
+        freq : str, optional
+            Frequency string for the period (e.g., 'Y', 'M', 'W'). Defaults to `None`.
+
+        See Also
+        --------
+        Timestamp : Represents a specific timestamp.
+        Period : Represents a span of time.
+        to_period : Converts an object to a Period.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> # Year end frequency
+        >>> ts.to_period(freq='Y')
+        Period('2020', 'Y-DEC')
+
+        >>> # Month end frequency
+        >>> ts.to_period(freq='M')
+        Period('2020-03', 'M')
+
+        >>> # Weekly frequency
+        >>> ts.to_period(freq='W')
+        Period('2020-03-09/2020-03-15', 'W-SUN')
+
+        >>> # Quarter end frequency
+        >>> ts.to_period(freq='Q')
+        Period('2020Q1', 'Q-DEC')
+        """
+        from pandas import Period
+
+        if self.tz is not None:
+            # GH#21333
+            warnings.warn(
+                "Converting to Period representation will drop timezone information.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        return Period(self, freq=freq)
+
+
+# ----------------------------------------------------------------------
+
+# Python front end to C extension type _Timestamp
+# This serves as the box for datetime64
+
+@set_module("pandas")
+class Timestamp(_Timestamp):
+    """
+    Pandas replacement for python datetime.datetime object.
+
+    Timestamp is the pandas equivalent of python's Datetime
+    and is interchangeable with it in most cases. It's the type used
+    for the entries that make up a DatetimeIndex, and other timeseries
+    oriented data structures in pandas.
+
+    Parameters
+    ----------
+    ts_input : datetime-like, str, int, float
+        Value to be converted to Timestamp.
+    year : int
+        Value of year.
+    month : int
+        Value of month.
+    day : int
+        Value of day.
+    hour : int, optional, default 0
+        Value of hour.
+    minute : int, optional, default 0
+        Value of minute.
+    second : int, optional, default 0
+        Value of second.
+    microsecond : int, optional, default 0
+        Value of microsecond.
+    tzinfo : datetime.tzinfo, optional, default None
+        Timezone info.
+    nanosecond : int, optional, default 0
+        Value of nanosecond.
+    tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+        Time zone for time which Timestamp will have.
+    unit : str
+        Unit used for conversion if ts_input is of type int or float. The
+        valid values are 'W', 'D', 'h', 'm', 's', 'ms', 'us', and 'ns'. For
+        example, 's' means seconds and 'ms' means milliseconds.
+
+        For float inputs, the result will be stored in nanoseconds, and
+        the unit attribute will be set as ``'ns'``.
+    fold : {0, 1}, default None, keyword-only
+        Due to daylight saving time, one wall clock time can occur twice
+        when shifting from summer to winter time; fold describes whether the
+        datetime-like corresponds  to the first (0) or the second time (1)
+        the wall clock hits the ambiguous time.
+
+    See Also
+    --------
+    Timedelta : Represents a duration, the difference between two dates or times.
+    datetime.datetime : Python datetime.datetime object.
+
+    Notes
+    -----
+    There are essentially three calling conventions for the constructor. The
+    primary form accepts four parameters. They can be passed by position or
+    keyword.
+
+    The other two forms mimic the parameters from ``datetime.datetime``. They
+    can be passed by either position or keyword, but not both mixed together.
+
+    Examples
+    --------
+    Using the primary calling convention:
+
+    This converts a datetime-like string
+
+    >>> pd.Timestamp('2017-01-01T12')
+    Timestamp('2017-01-01 12:00:00')
+
+    This converts a float representing a Unix epoch in units of seconds
+
+    >>> pd.Timestamp(1513393355.5, unit='s')
+    Timestamp('2017-12-16 03:02:35.500000')
+
+    This converts an int representing a Unix-epoch in units of weeks
+
+    >>> pd.Timestamp(1535, unit='W')
+    Timestamp('1999-06-03 00:00:00')
+
+    This converts an int representing a Unix-epoch in units of seconds
+    and for a particular timezone
+
+    >>> pd.Timestamp(1513393355, unit='s', tz='US/Pacific')
+    Timestamp('2017-12-15 19:02:35-0800', tz='US/Pacific')
+
+    Using the other two forms that mimic the API for ``datetime.datetime``:
+
+    >>> pd.Timestamp(2017, 1, 1, 12)
+    Timestamp('2017-01-01 12:00:00')
+
+    >>> pd.Timestamp(year=2017, month=1, day=1, hour=12)
+    Timestamp('2017-01-01 12:00:00')
+    """
+
+    @classmethod
+    def fromordinal(cls, ordinal, tz=None):
+        """
+        Construct a timestamp from a a proleptic Gregorian ordinal.
+
+        This method creates a `Timestamp` object corresponding to the given
+        proleptic Gregorian ordinal, which is a count of days from January 1,
+        0001 (using the proleptic Gregorian calendar). The time part of the
+        `Timestamp` is set to midnight (00:00:00) by default.
+
+        Parameters
+        ----------
+        ordinal : int
+            Date corresponding to a proleptic Gregorian ordinal.
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for the Timestamp.
+
+        Returns
+        -------
+        Timestamp
+            A `Timestamp` object representing the specified ordinal date.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Notes
+        -----
+        By definition there cannot be any tz info on the ordinal itself.
+
+        Examples
+        --------
+        Convert an ordinal to a `Timestamp`:
+
+        >>> pd.Timestamp.fromordinal(737425)
+        Timestamp('2020-01-01 00:00:00')
+
+        Create a `Timestamp` from an ordinal with timezone information:
+
+        >>> pd.Timestamp.fromordinal(737425, tz='UTC')
+        Timestamp('2020-01-01 00:00:00+0000', tz='UTC')
+        """
+        return cls(datetime.fromordinal(ordinal), tz=tz)
+
+    @classmethod
+    def now(cls, tz=None):
+        """
+        Return new Timestamp object representing current time local to tz.
+
+        This method returns a new `Timestamp` object that represents the current time.
+        If a timezone is provided, the current time will be localized to that timezone.
+        Otherwise, it returns the current local time.
+
+        Parameters
+        ----------
+        tz : str or timezone object, default None
+            Timezone to localize to.
+
+        See Also
+        --------
+        to_datetime : Convert argument to datetime.
+        Timestamp.utcnow : Return a new Timestamp representing UTC day and time.
+        Timestamp.today : Return the current time in the local timezone.
+
+        Examples
+        --------
+        >>> pd.Timestamp.now()  # doctest: +SKIP
+        Timestamp('2020-11-16 22:06:16.378782')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.now()
+        NaT
+        """
+        if isinstance(tz, str):
+            tz = maybe_get_tz(tz)
+        return cls(datetime.now(tz))
+
+    @classmethod
+    def today(cls, tz=None):
+        """
+        Return the current time in the local timezone.
+
+        This differs from datetime.today() in that it can be localized to a
+        passed timezone.
+
+        Parameters
+        ----------
+        tz : str or timezone object, default None
+            Timezone to localize to.
+
+        See Also
+        --------
+        datetime.datetime.today : Returns the current local date.
+        Timestamp.now : Returns current time with optional timezone.
+        Timestamp : A class representing a specific timestamp.
+
+        Examples
+        --------
+        >>> pd.Timestamp.today()    # doctest: +SKIP
+        Timestamp('2020-11-16 22:37:39.969883')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.today()
+        NaT
+        """
+        return cls.now(tz)
+
+    @classmethod
+    def utcnow(cls):
+        """
+        Timestamp.utcnow()
+
+        Return a new Timestamp representing UTC day and time.
+
+        See Also
+        --------
+        Timestamp : Constructs an arbitrary datetime.
+        Timestamp.now : Return the current local date and time, which
+            can be timezone-aware.
+        Timestamp.today : Return the current local date and time with
+            timezone information set to None.
+        to_datetime : Convert argument to datetime.
+        date_range : Return a fixed frequency DatetimeIndex.
+        Timestamp.utctimetuple : Return UTC time tuple, compatible with
+            time.localtime().
+
+        Examples
+        --------
+        >>> pd.Timestamp.utcnow()   # doctest: +SKIP
+        Timestamp('2020-11-16 22:50:18.092888+0000', tz='UTC')
+        """
+        warnings.warn(
+            # The stdlib datetime.utcnow is deprecated, so we deprecate to match.
+            #  GH#56680
+            "Timestamp.utcnow is deprecated and will be removed in a future "
+            "version. Use Timestamp.now('UTC') instead.",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        return cls.now(UTC)
+
+    @classmethod
+    def utcfromtimestamp(cls, ts):
+        """
+        Timestamp.utcfromtimestamp(ts)
+
+        Construct a timezone-aware UTC datetime from a POSIX timestamp.
+
+        This method creates a datetime object from a POSIX timestamp, keeping the
+        Timestamp object's timezone.
+
+        Parameters
+        ----------
+        ts : float
+            POSIX timestamp.
+
+        See Also
+        --------
+        Timezone.tzname : Return time zone name.
+        Timestamp.utcnow : Return a new Timestamp representing UTC day and time.
+        Timestamp.fromtimestamp : Transform timestamp[, tz] to tz's local
+            time from POSIX timestamp.
+
+        Notes
+        -----
+        Timestamp.utcfromtimestamp behavior differs from datetime.utcfromtimestamp
+        in returning a timezone-aware object.
+
+        Examples
+        --------
+        >>> pd.Timestamp.utcfromtimestamp(1584199972)
+        Timestamp('2020-03-14 15:32:52+0000', tz='UTC')
+        """
+        # GH#22451
+        warnings.warn(
+            # The stdlib datetime.utcfromtimestamp is deprecated, so we deprecate
+            #  to match. GH#56680
+            "Timestamp.utcfromtimestamp is deprecated and will be removed in a "
+            "future version. Use Timestamp.fromtimestamp(ts, 'UTC') instead.",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        return cls.fromtimestamp(ts, tz="UTC")
+
+    @classmethod
+    def fromtimestamp(cls, ts, tz=None):
+        """
+        Create a `Timestamp` object from a POSIX timestamp.
+
+        This method converts a POSIX timestamp (the number of seconds since
+        January 1, 1970, 00:00:00 UTC) into a `Timestamp` object. The resulting
+        `Timestamp` can be localized to a specific time zone if provided.
+
+        Parameters
+        ----------
+        ts : float
+            The POSIX timestamp to convert, representing seconds since
+            the epoch (1970-01-01 00:00:00 UTC).
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile, optional
+            Time zone for the `Timestamp`. If not provided, the `Timestamp` will
+            be timezone-naive (i.e., without time zone information).
+
+        Returns
+        -------
+        Timestamp
+            A `Timestamp` object representing the given POSIX timestamp.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+        datetime.datetime.fromtimestamp : Returns a datetime from a POSIX timestamp.
+
+        Examples
+        --------
+        Convert a POSIX timestamp to a `Timestamp`:
+
+        >>> pd.Timestamp.fromtimestamp(1584199972)  # doctest: +SKIP
+        Timestamp('2020-03-14 15:32:52')
+
+        Note that the output may change depending on your local time and time zone:
+
+        >>> pd.Timestamp.fromtimestamp(1584199972, tz='UTC')  # doctest: +SKIP
+        Timestamp('2020-03-14 15:32:52+0000', tz='UTC')
+        """
+        tz = maybe_get_tz(tz)
+        return cls(datetime.fromtimestamp(ts, tz))
+
+    def strftime(self, format):
+        """
+        Return a formatted string of the Timestamp.
+
+        Parameters
+        ----------
+        format : str
+            Format string to convert Timestamp to string.
+            See strftime documentation for more information on the format string:
+            https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior.
+
+        See Also
+        --------
+        Timestamp.isoformat : Return the time formatted according to ISO 8601.
+        pd.to_datetime : Convert argument to datetime.
+        Period.strftime : Format a single Period.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts.strftime('%Y-%m-%d %X')
+        '2020-03-14 15:32:52'
+        """
+        try:
+            _dt = datetime(self._year, self.month, self.day,
+                           self.hour, self.minute, self.second,
+                           self.microsecond, self.tzinfo, fold=self.fold)
+        except ValueError as err:
+            raise NotImplementedError(
+                "strftime not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+                "For now, please call the components you need (such as `.year` "
+                "and `.month`) and construct your string from there."
+            ) from err
+        return _dt.strftime(format)
+
+    def ctime(self):
+        """
+        Return a ctime() style string representing the Timestamp.
+
+        This method returns a string representing the date and time
+        in the format returned by the standard library's `time.ctime()`
+        function, which is typically in the form 'Day Mon DD HH:MM:SS YYYY'.
+
+        If the `Timestamp` is outside the range supported by Python's
+        standard library, a `NotImplementedError` is raised.
+
+        Returns
+        -------
+        str
+            A string representing the Timestamp in ctime format.
+
+        See Also
+        --------
+        time.ctime : Return a string representing time in ctime format.
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        datetime.datetime.ctime : Return a ctime style string from a datetime object.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00.00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.ctime()
+        'Sun Jan  1 10:00:00 2023'
+        """
+        try:
+            _dt = datetime(self._year, self.month, self.day,
+                           self.hour, self.minute, self.second,
+                           self.microsecond, self.tzinfo, fold=self.fold)
+        except ValueError as err:
+            raise NotImplementedError(
+                "ctime not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+                "For now, please call the components you need (such as `.year` "
+                "and `.month`) and construct your string from there."
+            ) from err
+        return _dt.ctime()
+
+    def date(self):
+        """
+        Returns `datetime.date` with the same year, month, and day.
+
+        This method extracts the date component from the `Timestamp` and returns
+        it as a `datetime.date` object, discarding the time information.
+
+        Returns
+        -------
+        datetime.date
+            The date part of the `Timestamp`.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        datetime.datetime.date : Extract the date component from a `datetime` object.
+
+        Examples
+        --------
+        Extract the date from a Timestamp:
+
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00.00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.date()
+        datetime.date(2023, 1, 1)
+        """
+        try:
+            _dt = dt.date(self._year, self.month, self.day)
+        except ValueError as err:
+            raise NotImplementedError(
+                "date not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+            ) from err
+        return _dt
+
+    def dst(self):
+        """
+        Return the daylight saving time (DST) adjustment.
+
+        This method returns the DST adjustment as a `datetime.timedelta` object
+        if the Timestamp is timezone-aware and DST is applicable.
+
+        See Also
+        --------
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2000-06-01 00:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2000-06-01 00:00:00+0200', tz='Europe/Brussels')
+        >>> ts.dst()
+        datetime.timedelta(seconds=3600)
+        """
+        return super().dst()
+
+    def isocalendar(self):
+        """
+        Return a named tuple containing ISO year, week number, and weekday.
+
+        See Also
+        --------
+        DatetimeIndex.isocalendar : Return a 3-tuple containing ISO year,
+            week number, and weekday for the given DatetimeIndex object.
+        datetime.date.isocalendar : The equivalent method for `datetime.date` objects.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.isocalendar()
+        datetime.IsoCalendarDate(year=2022, week=52, weekday=7)
+        """
+        try:
+            _dt = datetime(self._year, self.month, self.day,
+                           self.hour, self.minute, self.second,
+                           self.microsecond, self.tzinfo, fold=self.fold)
+        except ValueError as err:
+            raise NotImplementedError(
+                "isocalendar not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+            ) from err
+        return _dt.isocalendar()
+
+    def tzname(self):
+        """
+        Return time zone name.
+
+        This method returns the name of the Timestamp's time zone as a string.
+
+        See Also
+        --------
+        Timestamp.tzinfo : Returns the timezone information of the Timestamp.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.tzname()
+        'CET'
+        """
+        return super().tzname()
+
+    def utcoffset(self):
+        """
+        Return utc offset.
+
+        This method returns the difference between UTC and the local time
+        as a `timedelta` object. It is useful for understanding the time
+        difference between the current timezone and UTC.
+
+        Returns
+        -------
+        timedelta
+            The difference between UTC and the local time as a `timedelta` object.
+
+        See Also
+        --------
+        datetime.datetime.utcoffset :
+            Standard library method to get the UTC offset of a datetime object.
+        Timestamp.tzname : Return the name of the timezone.
+        Timestamp.dst : Return the daylight saving time (DST) adjustment.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.utcoffset()
+        datetime.timedelta(seconds=3600)
+        """
+        return super().utcoffset()
+
+    def utctimetuple(self):
+        """
+        Return UTC time tuple, compatible with `time.localtime()`.
+
+        This method converts the Timestamp to UTC and returns a time tuple
+        containing 9 components: year, month, day, hour, minute, second,
+        weekday, day of year, and DST flag. This is particularly useful for
+        converting a Timestamp to a format compatible with time module functions.
+
+        Returns
+        -------
+        time.struct_time
+            A time.struct_time object representing the UTC time.
+
+        See Also
+        --------
+        datetime.datetime.utctimetuple :
+            Return UTC time tuple, compatible with time.localtime().
+        Timestamp.timetuple : Return time tuple of local time.
+        time.struct_time : Time tuple structure used by time functions.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.utctimetuple()
+        time.struct_time(tm_year=2023, tm_mon=1, tm_mday=1, tm_hour=9,
+        tm_min=0, tm_sec=0, tm_wday=6, tm_yday=1, tm_isdst=0)
+        """
+        return super().utctimetuple()
+
+    def time(self):
+        """
+        Return time object with same time but with tzinfo=None.
+
+        This method extracts the time part of the `Timestamp` object, excluding any
+        timezone information. It returns a `datetime.time` object which only represents
+        the time (hours, minutes, seconds, and microseconds).
+
+        See Also
+        --------
+        Timestamp.date : Return date object with same year, month and day.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.time()
+        datetime.time(10, 0)
+        """
+        return super().time()
+
+    def timetuple(self):
+        """
+        Return time tuple, compatible with time.localtime().
+
+        This method converts the `Timestamp` into a time tuple, which is compatible
+        with functions like `time.localtime()`. The time tuple is a named tuple with
+        attributes such as year, month, day, hour, minute, second, weekday,
+        day of the year, and daylight savings indicator.
+
+        See Also
+        --------
+        time.localtime : Converts a POSIX timestamp into a time tuple.
+        Timestamp : The `Timestamp` that represents a specific point in time.
+        datetime.datetime.timetuple : Equivalent method in the `datetime` module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.timetuple()
+        time.struct_time(tm_year=2023, tm_mon=1, tm_mday=1,
+        tm_hour=10, tm_min=0, tm_sec=0, tm_wday=6, tm_yday=1, tm_isdst=-1)
+        """
+        try:
+            _dt = datetime(self._year, self.month, self.day,
+                           self.hour, self.minute, self.second,
+                           self.microsecond, self.tzinfo, fold=self.fold)
+        except ValueError as err:
+            raise NotImplementedError(
+                "timetuple not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+            ) from err
+        return _dt.timetuple()
+
+    def timetz(self):
+        """
+        Return time object with same time and tzinfo.
+
+        This method returns a datetime.time object with
+        the time and tzinfo corresponding to the pd.Timestamp
+        object, ignoring any information about the day/date.
+
+        See Also
+        --------
+        datetime.datetime.timetz : Return datetime.time object with the
+            same time attributes as the datetime object.
+        datetime.time : Class to represent the time of day, independent
+            of any particular day.
+        datetime.datetime.tzinfo : Attribute of datetime.datetime objects
+            representing the timezone of the datetime object.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00', tz='Europe/Brussels')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00+0100', tz='Europe/Brussels')
+        >>> ts.timetz()
+        datetime.time(10, 0, tzinfo=<DstTzInfo 'Europe/Brussels' CET+1:00:00 STD>)
+        """
+        return super().timetz()
+
+    def toordinal(self):
+        """
+        Return proleptic Gregorian ordinal. January 1 of year 1 is day 1.
+
+        The proleptic Gregorian ordinal is a continuous count of days since
+        January 1 of year 1, which is considered day 1. This method converts
+        the `Timestamp` to its equivalent ordinal number, useful for date arithmetic
+        and comparison operations.
+
+        See Also
+        --------
+        datetime.datetime.toordinal : Equivalent method in the `datetime` module.
+        Timestamp : The `Timestamp` that represents a specific point in time.
+        Timestamp.fromordinal : Create a `Timestamp` from an ordinal.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:50')
+        >>> ts
+        Timestamp('2023-01-01 10:00:50')
+        >>> ts.toordinal()
+        738521
+        """
+        try:
+            _dt = datetime(self._year, self.month, self.day,
+                           self.hour, self.minute, self.second,
+                           self.microsecond, self.tzinfo, fold=self.fold)
+        except ValueError as err:
+            raise NotImplementedError(
+                "toordinal not yet supported on Timestamps which "
+                "are outside the range of Python's standard library. "
+            ) from err
+        return _dt.toordinal()
+
+    # Issue 25016.
+    @classmethod
+    def strptime(cls, date_string, format):
+        """
+        Convert string argument to datetime.
+
+        This method is not implemented; calling it will raise NotImplementedError.
+        Use pd.to_datetime() instead.
+
+        Parameters
+        ----------
+        date_string : str
+            String to convert to a datetime.
+        format : str, default None
+            The format string to parse time, e.g. "%d/%m/%Y".
+
+        See Also
+        --------
+        pd.to_datetime : Convert argument to datetime.
+        datetime.datetime.strptime : Return a datetime corresponding to a string
+            representing a date and time, parsed according to a separate
+            format string.
+        datetime.datetime.strftime : Return a string representing the date and
+            time, controlled by an explicit format string.
+        Timestamp.isoformat : Return the time formatted according to ISO 8601.
+
+        Examples
+        --------
+        >>> pd.Timestamp.strptime("2023-01-01", "%d/%m/%y")
+        Traceback (most recent call last):
+        NotImplementedError
+        """
+        raise NotImplementedError(
+            "Timestamp.strptime() is not implemented. "
+            "Use to_datetime() to parse date strings."
+        )
+
+    @classmethod
+    def combine(cls, date, time):
+        """
+        Timestamp.combine(date, time)
+
+        Combine a date and time into a single Timestamp object.
+
+        This method takes a `date` object and a `time` object
+        and combines them into a single `Timestamp`
+        that has the same date and time fields.
+
+        Parameters
+        ----------
+        date : datetime.date
+            The date part of the Timestamp.
+        time : datetime.time
+            The time part of the Timestamp.
+
+        Returns
+        -------
+        Timestamp
+            A new `Timestamp` object representing the combined date and time.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Examples
+        --------
+        >>> from datetime import date, time
+        >>> pd.Timestamp.combine(date(2020, 3, 14), time(15, 30, 15))
+        Timestamp('2020-03-14 15:30:15')
+        """
+        return cls(datetime.combine(date, time))
+
+    def __new__(
+        cls,
+        object ts_input=_no_input,
+        year=None,
+        month=None,
+        day=None,
+        hour=None,
+        minute=None,
+        second=None,
+        microsecond=None,
+        tzinfo_type tzinfo=None,
+        *,
+        nanosecond=None,
+        tz=_no_input,
+        unit=None,
+        fold=None,
+    ):
+        # The parameter list folds together legacy parameter names (the first
+        # four) and positional and keyword parameter names from pydatetime.
+        #
+        # There are three calling forms:
+        #
+        # - In the legacy form, the first parameter, ts_input, is required
+        #   and may be datetime-like, str, int, or float. The second
+        #   parameter, offset, is optional and may be str or DateOffset.
+        #
+        # - ints in the first, second, and third arguments indicate
+        #   pydatetime positional arguments. Only the first 8 arguments
+        #   (standing in for year, month, day, hour, minute, second,
+        #   microsecond, tzinfo) may be non-None. As a shortcut, we just
+        #   check that the second argument is an int.
+        #
+        # - Nones for the first four (legacy) arguments indicate pydatetime
+        #   keyword arguments. year, month, and day are required. As a
+        #   shortcut, we just check that the first argument was not passed.
+        #
+        # Mixing pydatetime positional and keyword arguments is forbidden!
+
+        cdef:
+            _TSObject ts
+            tzinfo_type tzobj
+
+        _date_attributes = [year, month, day, hour, minute, second,
+                            microsecond, nanosecond]
+
+        explicit_tz_none = tz is None
+        if tz is _no_input:
+            tz = None
+
+        if tzinfo is not None:
+            # GH#17690 tzinfo must be a datetime.tzinfo object, ensured
+            #  by the cython annotation.
+            if tz is not None:
+                raise ValueError("Can provide at most one of tz, tzinfo")
+
+            # User passed tzinfo instead of tz; avoid silently ignoring
+            tz, tzinfo = tzinfo, None
+
+        # Allow fold only for unambiguous input
+        if fold is not None:
+            if fold not in [0, 1]:
+                raise ValueError(
+                    "Valid values for the fold argument are None, 0, or 1."
+                )
+
+            if (ts_input is not _no_input and not (
+                    PyDateTime_Check(ts_input) and
+                    getattr(ts_input, "tzinfo", None) is None)):
+                raise ValueError(
+                    "Cannot pass fold with possibly unambiguous input: int, "
+                    "float, numpy.datetime64, str, or timezone-aware "
+                    "datetime-like. Pass naive datetime-like or build "
+                    "Timestamp from components."
+                )
+
+            if tz is not None and PyTZInfo_Check(tz) and treat_tz_as_pytz(tz):
+                raise ValueError(
+                    "pytz timezones do not support fold. Please use dateutil "
+                    "timezones."
+                )
+
+            if hasattr(ts_input, "fold"):
+                ts_input = ts_input.replace(fold=fold)
+
+        # GH 30543 if pd.Timestamp already passed, return it
+        # check that only ts_input is passed
+        # checking verbosely, because cython doesn't optimize
+        # list comprehensions (as of cython 0.29.x)
+        if (isinstance(ts_input, _Timestamp) and
+                tz is None and unit is None and year is None and
+                month is None and day is None and hour is None and
+                minute is None and second is None and
+                microsecond is None and nanosecond is None and
+                tzinfo is None):
+            return ts_input
+        elif isinstance(ts_input, str):
+            # User passed a date string to parse.
+            # Check that the user didn't also pass a date attribute kwarg.
+            if any(arg is not None for arg in _date_attributes):
+                raise ValueError(
+                    "Cannot pass a date attribute keyword "
+                    "argument when passing a date string; 'tz' is keyword-only"
+                )
+
+        elif ts_input is _no_input:
+            # GH 31200
+            # When year, month or day is not given, we call the datetime
+            # constructor to make sure we get the same error message
+            # since Timestamp inherits datetime
+            datetime_kwargs = {
+                "hour": hour or 0,
+                "minute": minute or 0,
+                "second": second or 0,
+                "microsecond": microsecond or 0,
+                "fold": fold or 0
+            }
+            if year is not None:
+                datetime_kwargs["year"] = year
+            if month is not None:
+                datetime_kwargs["month"] = month
+            if day is not None:
+                datetime_kwargs["day"] = day
+
+            ts_input = datetime(**datetime_kwargs)
+
+        elif is_integer_object(year):
+            # User passed positional arguments:
+            # Timestamp(year, month, day[, hour[, minute[, second[,
+            # microsecond[, tzinfo]]]]])
+            ts_input = datetime(ts_input, year, month, day or 0,
+                                hour or 0, minute or 0, second or 0, fold=fold or 0)
+            unit = None
+
+        if getattr(ts_input, "tzinfo", None) is not None and tz is not None:
+            raise ValueError("Cannot pass a datetime or Timestamp with tzinfo with "
+                             "the tz parameter. Use tz_convert instead.")
+
+        tzobj = maybe_get_tz(tz)
+
+        if nanosecond is None:
+            nanosecond = 0
+        elif not (999 >= nanosecond >= 0):
+            raise ValueError("nanosecond must be in 0..999")
+
+        ts = convert_to_tsobject(ts_input, tzobj, unit, 0, 0, nanosecond)
+
+        if ts.value == NPY_NAT:
+            return NaT
+
+        if ts.tzinfo is not None and explicit_tz_none:
+            raise ValueError(
+                "Passed data is timezone-aware, incompatible with 'tz=None'."
+            )
+
+        return create_timestamp_from_ts(ts.value, ts.dts, ts.tzinfo, ts.fold, ts.creso)
+
+    def _round(self, freq, mode, ambiguous="raise", nonexistent="raise"):
+        cdef:
+            int64_t nanos
+
+        freq = to_offset(freq, is_period=False)
+        nanos = get_unit_for_round(freq, self._creso)
+        if nanos == 0:
+            if freq.nanos == 0:
+                raise ValueError("Division by zero in rounding")
+
+            # e.g. self.unit == "s" and sub-second freq
+            return self
+
+        if self.tz is not None:
+            value = self.tz_localize(None)._value
+        else:
+            value = self._value
+
+        value = np.array([value], dtype=np.int64)
+
+        # Will only ever contain 1 element for timestamp
+        try:
+            r = round_nsint64(value, mode, nanos)[0]
+        except OverflowError as err:
+            raise OutOfBoundsDatetime(
+                f"Cannot round {self} to freq={freq} without overflow"
+            ) from err
+
+        result = Timestamp._from_value_and_reso(r, self._creso, None)
+        if self.tz is not None:
+            result = result.tz_localize(
+                self.tz, ambiguous=ambiguous, nonexistent=nonexistent
+            )
+        return result
+
+    def round(self, freq, ambiguous="raise", nonexistent="raise"):
+        """
+        Round the Timestamp to the specified resolution.
+
+        This method rounds the given Timestamp down to a specified frequency
+        level. It is particularly useful in data analysis to normalize timestamps
+        to regular frequency intervals. For instance, rounding to the nearest
+        minute, hour, or day can help in time series comparisons or resampling
+        operations.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the rounding resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        a new Timestamp rounded to the given resolution of `freq`
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted
+
+        See Also
+        --------
+        datetime.round : Similar behavior in native Python datetime module.
+        Timestamp.floor : Round the Timestamp downward to the nearest multiple
+            of the specified frequency.
+        Timestamp.ceil : Round the Timestamp upward to the nearest multiple of
+            the specified frequency.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, rounding will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When rounding
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be rounded using multiple frequency units:
+
+        >>> ts.round(freq='h')  # hour
+        Timestamp('2020-03-14 16:00:00')
+
+        >>> ts.round(freq='min')  # minute
+        Timestamp('2020-03-14 15:33:00')
+
+        >>> ts.round(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:52')
+
+        >>> ts.round(freq='ms')  # milliseconds
+        Timestamp('2020-03-14 15:32:52.193000')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.round(freq='5min')
+        Timestamp('2020-03-14 15:35:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.round(freq='1h30min')
+        Timestamp('2020-03-14 15:00:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.round()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 01:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.round("h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.round("h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """
+        return self._round(
+            freq, RoundTo.NEAREST_HALF_EVEN, ambiguous, nonexistent
+        )
+
+    def floor(self, freq, ambiguous="raise", nonexistent="raise"):
+        """
+        Return a new Timestamp floored to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the flooring resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted.
+
+        See Also
+        --------
+        Timestamp.ceil : Round up a Timestamp to the specified resolution.
+        Timestamp.round : Round a Timestamp to the specified resolution.
+        Series.dt.floor : Round down the datetime values in a Series.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, flooring will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When flooring
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be floored using multiple frequency units:
+
+        >>> ts.floor(freq='h')  # hour
+        Timestamp('2020-03-14 15:00:00')
+
+        >>> ts.floor(freq='min')  # minute
+        Timestamp('2020-03-14 15:32:00')
+
+        >>> ts.floor(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:52')
+
+        >>> ts.floor(freq='ns')  # nanoseconds
+        Timestamp('2020-03-14 15:32:52.192548651')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.floor(freq='5min')
+        Timestamp('2020-03-14 15:30:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.floor(freq='1h30min')
+        Timestamp('2020-03-14 15:00:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.floor()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 03:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.floor("2h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.floor("2h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """
+        return self._round(freq, RoundTo.MINUS_INFTY, ambiguous, nonexistent)
+
+    def ceil(self, freq, ambiguous="raise", nonexistent="raise"):
+        """
+        Return a new Timestamp ceiled to this resolution.
+
+        Parameters
+        ----------
+        freq : str
+            Frequency string indicating the ceiling resolution.
+        ambiguous : bool or {'raise', 'NaT'}, default 'raise'
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : {'raise', 'shift_forward', 'shift_backward', 'NaT', \
+timedelta}, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Raises
+        ------
+        ValueError if the freq cannot be converted.
+
+        See Also
+        --------
+        Timestamp.floor : Round down a Timestamp to the specified resolution.
+        Timestamp.round : Round a Timestamp to the specified resolution.
+        Series.dt.ceil : Ceil the datetime values in a Series.
+
+        Notes
+        -----
+        If the Timestamp has a timezone, ceiling will take place relative to the
+        local ("wall") time and re-localized to the same timezone. When ceiling
+        near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+        control the re-localization behavior.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+
+        A timestamp can be ceiled using multiple frequency units:
+
+        >>> ts.ceil(freq='h')  # hour
+        Timestamp('2020-03-14 16:00:00')
+
+        >>> ts.ceil(freq='min')  # minute
+        Timestamp('2020-03-14 15:33:00')
+
+        >>> ts.ceil(freq='s')  # seconds
+        Timestamp('2020-03-14 15:32:53')
+
+        >>> ts.ceil(freq='us')  # microseconds
+        Timestamp('2020-03-14 15:32:52.192549')
+
+        ``freq`` can also be a multiple of a single unit, like '5min' (i.e.  5 minutes):
+
+        >>> ts.ceil(freq='5min')
+        Timestamp('2020-03-14 15:35:00')
+
+        or a combination of multiple units, like '1h30min' (i.e. 1 hour and 30 minutes):
+
+        >>> ts.ceil(freq='1h30min')
+        Timestamp('2020-03-14 16:30:00')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.ceil()
+        NaT
+
+        When rounding near a daylight savings time transition, use ``ambiguous`` or
+        ``nonexistent`` to control how the timestamp should be re-localized.
+
+        >>> ts_tz = pd.Timestamp("2021-10-31 01:30:00").tz_localize("Europe/Amsterdam")
+
+        >>> ts_tz.ceil("h", ambiguous=False)
+        Timestamp('2021-10-31 02:00:00+0100', tz='Europe/Amsterdam')
+
+        >>> ts_tz.ceil("h", ambiguous=True)
+        Timestamp('2021-10-31 02:00:00+0200', tz='Europe/Amsterdam')
+        """
+        return self._round(freq, RoundTo.PLUS_INFTY, ambiguous, nonexistent)
+
+    @property
+    def tz(self):
+        """
+        Alias for tzinfo.
+
+        The `tz` property provides a simple and direct way to retrieve the timezone
+        information of a `Timestamp` object. It is particularly useful when working
+        with time series data that includes timezone information, allowing for easy
+        access and manipulation of the timezone context.
+
+        See Also
+        --------
+        Timestamp.tzinfo : Returns the timezone information of the Timestamp.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp(1584226800, unit='s', tz='Europe/Stockholm')
+        >>> ts.tz
+        zoneinfo.ZoneInfo(key='Europe/Stockholm')
+        """
+        return self.tzinfo
+
+    @tz.setter
+    def tz(self, value):
+        # GH 3746: Prevent localizing or converting the index by setting tz
+        raise AttributeError(
+            "Cannot directly set timezone. "
+            "Use tz_localize() or tz_convert() as appropriate"
+        )
+
+    def tz_localize(self, tz, ambiguous="raise", nonexistent="raise"):
+        """
+        Localize the Timestamp to a timezone.
+
+        Convert naive Timestamp to local time zone or remove
+        timezone from timezone-aware Timestamp.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for time which Timestamp will be converted to.
+            None will remove timezone holding local time.
+
+        ambiguous : bool, 'NaT', default 'raise'
+            When clocks moved backward due to DST, ambiguous times may arise.
+            For example in Central European Time (UTC+01), when going from
+            03:00 DST to 02:00 non-DST, 02:30:00 local time occurs both at
+            00:30:00 UTC and at 01:30:00 UTC. In such a situation, the
+            `ambiguous` parameter dictates how ambiguous times should be
+            handled.
+
+            The behavior is as follows:
+
+            * bool contains flags to determine if time is dst or not (note
+              that this flag is only applicable for ambiguous fall dst dates).
+            * 'NaT' will return NaT for an ambiguous time.
+            * 'raise' will raise a ValueError for an ambiguous time.
+
+        nonexistent : 'shift_forward', 'shift_backward', 'NaT', timedelta, \
+default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            The behavior is as follows:
+
+            * 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time.
+            * 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time.
+            * 'NaT' will return NaT where there are nonexistent times.
+            * timedelta objects will shift nonexistent times by the timedelta.
+            * 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        localized : Timestamp
+
+        Raises
+        ------
+        TypeError
+            If the Timestamp is tz-aware and tz is not None.
+
+        See Also
+        --------
+        Timestamp.tzinfo : Returns the timezone information of the Timestamp.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        DatetimeIndex.tz_localize : Localize a DatetimeIndex to a specific time zone.
+        datetime.datetime.astimezone : Convert a datetime object to another time zone.
+
+        Examples
+        --------
+        Create a naive timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651')
+
+        Add 'Europe/Stockholm' as timezone:
+
+        >>> ts.tz_localize(tz='Europe/Stockholm')
+        Timestamp('2020-03-14 15:32:52.192548651+0100', tz='Europe/Stockholm')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.tz_localize()
+        NaT
+        """
+        if not isinstance(ambiguous, bool) and ambiguous not in {"NaT", "raise"}:
+            raise ValueError(
+                        "'ambiguous' parameter must be one of: "
+                        "True, False, 'NaT', 'raise' (default)"
+                    )
+
+        nonexistent_options = ("raise", "NaT", "shift_forward", "shift_backward")
+        if nonexistent not in nonexistent_options and not PyDelta_Check(nonexistent):
+            raise ValueError(
+                "The nonexistent argument must be one of 'raise', "
+                "'NaT', 'shift_forward', 'shift_backward' or a timedelta object"
+            )
+
+        if self.tzinfo is None:
+            # tz naive, localize
+            tz = maybe_get_tz(tz)
+            if not isinstance(ambiguous, str):
+                ambiguous = [ambiguous]
+            value = tz_localize_to_utc_single(self._value, tz,
+                                              ambiguous=ambiguous,
+                                              nonexistent=nonexistent,
+                                              creso=self._creso)
+        elif tz is None:
+            # reset tz
+            value = tz_convert_from_utc_single(self._value, self.tz, creso=self._creso)
+
+        else:
+            raise TypeError(
+                "Cannot localize tz-aware Timestamp, use tz_convert for conversions"
+            )
+
+        out = type(self)._from_value_and_reso(value, self._creso, tz=tz)
+        return out
+
+    def tz_convert(self, tz):
+        """
+        Convert timezone-aware Timestamp to another time zone.
+
+        This method is used to convert a timezone-aware Timestamp object to a
+        different time zone. The original UTC time remains the same; only the
+        time zone information is changed. If the Timestamp is timezone-naive, a
+        TypeError is raised.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile or None
+            Time zone for time which Timestamp will be converted to.
+            None will remove timezone holding UTC time.
+
+        Returns
+        -------
+        converted : Timestamp
+
+        Raises
+        ------
+        TypeError
+            If Timestamp is tz-naive.
+
+        See Also
+        --------
+        Timestamp.tz_localize : Localize the Timestamp to a timezone.
+        DatetimeIndex.tz_convert : Convert a DatetimeIndex to another time zone.
+        DatetimeIndex.tz_localize : Localize a DatetimeIndex to a specific time zone.
+        datetime.datetime.astimezone : Convert a datetime object to another time zone.
+
+        Examples
+        --------
+        Create a timestamp object with UTC timezone:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651', tz='UTC')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651+0000', tz='UTC')
+
+        Change to Tokyo timezone:
+
+        >>> ts.tz_convert(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Can also use ``astimezone``:
+
+        >>> ts.astimezone(tz='Asia/Tokyo')
+        Timestamp('2020-03-15 00:32:52.192548651+0900', tz='Asia/Tokyo')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.tz_convert(tz='Asia/Tokyo')
+        NaT
+        """
+        if self.tzinfo is None:
+            # tz naive, use tz_localize
+            raise TypeError(
+                "Cannot convert tz-naive Timestamp, use tz_localize to localize"
+            )
+        else:
+            # Same UTC timestamp, different time zone
+            tz = maybe_get_tz(tz)
+            out = type(self)._from_value_and_reso(self._value, reso=self._creso, tz=tz)
+            return out
+
+    astimezone = tz_convert
+
+    def replace(
+        self,
+        year=None,
+        month=None,
+        day=None,
+        hour=None,
+        minute=None,
+        second=None,
+        microsecond=None,
+        nanosecond=None,
+        tzinfo=object,
+        fold=None,
+    ):
+        """
+        Implements datetime.replace, handles nanoseconds.
+
+        This method creates a new `Timestamp` object by replacing the specified
+        fields with new values. The new `Timestamp` retains the original fields
+        that are not explicitly replaced. This method handles nanoseconds, and
+        the `tzinfo` parameter allows for timezone replacement without conversion.
+
+        Parameters
+        ----------
+        year : int, optional
+            The year to replace. If `None`, the year is not changed.
+        month : int, optional
+            The month to replace. If `None`, the month is not changed.
+        day : int, optional
+            The day to replace. If `None`, the day is not changed.
+        hour : int, optional
+            The hour to replace. If `None`, the hour is not changed.
+        minute : int, optional
+            The minute to replace. If `None`, the minute is not changed.
+        second : int, optional
+            The second to replace. If `None`, the second is not changed.
+        microsecond : int, optional
+            The microsecond to replace. If `None`, the microsecond is not changed.
+        nanosecond : int, optional
+            The nanosecond to replace. If `None`, the nanosecond is not changed.
+        tzinfo : tz-convertible, optional
+            The timezone information to replace. If `None`, the timezone is not changed.
+        fold : int, optional
+            The fold information to replace. If `None`, the fold is not changed.
+
+        Returns
+        -------
+        Timestamp
+            A new `Timestamp` object with the specified fields replaced.
+
+        See Also
+        --------
+        Timestamp : Represents a single timestamp, similar to `datetime`.
+        to_datetime : Converts various types of data to datetime.
+
+        Notes
+        -----
+        The `replace` method does not perform timezone conversions. If you need
+        to convert the timezone, use the `tz_convert` method instead.
+
+        Examples
+        --------
+        Create a timestamp object:
+
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52.192548651', tz='UTC')
+        >>> ts
+        Timestamp('2020-03-14 15:32:52.192548651+0000', tz='UTC')
+
+        Replace year and the hour:
+
+        >>> ts.replace(year=1999, hour=10)
+        Timestamp('1999-03-14 10:32:52.192548651+0000', tz='UTC')
+
+        Replace timezone (not a conversion):
+
+        >>> import zoneinfo
+        >>> ts.replace(tzinfo=zoneinfo.ZoneInfo('US/Pacific'))
+        Timestamp('2020-03-14 15:32:52.192548651-0700', tz='US/Pacific')
+
+        Analogous for ``pd.NaT``:
+
+        >>> pd.NaT.replace(tzinfo=zoneinfo.ZoneInfo('US/Pacific'))
+        NaT
+        """
+
+        cdef:
+            npy_datetimestruct dts
+            int64_t value
+            object k, v
+            datetime ts_input
+            tzinfo_type tzobj
+            _TSObject ts
+
+        # set to naive if needed
+        tzobj = self.tzinfo
+        value = self._value
+
+        # GH 37610. Preserve fold when replacing.
+        if fold is None:
+            fold = self.fold
+
+        if tzobj is not None:
+            value = tz_convert_from_utc_single(value, tzobj, creso=self._creso)
+
+        # setup components
+        pandas_datetime_to_datetimestruct(value, self._creso, &dts)
+        dts.ps = self._nanosecond * 1000
+
+        # replace
+        def validate(k, v):
+            """ validate integers """
+            if not is_integer_object(v):
+                raise ValueError(
+                    f"value must be an integer, received {type(v)} for {k}"
+                )
+            return v
+
+        if year is not None:
+            dts.year = validate("year", year)
+        if month is not None:
+            dts.month = validate("month", month)
+        if day is not None:
+            dts.day = validate("day", day)
+        if hour is not None:
+            dts.hour = validate("hour", hour)
+        if minute is not None:
+            dts.min = validate("minute", minute)
+        if second is not None:
+            dts.sec = validate("second", second)
+        if microsecond is not None:
+            dts.us = validate("microsecond", microsecond)
+        if nanosecond is not None:
+            dts.ps = validate("nanosecond", nanosecond) * 1000
+        if tzinfo is not object:
+            tzobj = tzinfo
+
+        # reconstruct & check bounds
+        if tzobj is None:
+            # We can avoid going through pydatetime paths, which is robust
+            #  to datetimes outside of pydatetime range.
+            ts = _TSObject()
+            try:
+                ts.value = npy_datetimestruct_to_datetime(self._creso, &dts)
+            except OverflowError as err:
+                fmt = dts_to_iso_string(&dts)
+                raise OutOfBoundsDatetime(
+                    f"Out of bounds timestamp: {fmt} with frequency '{self.unit}'"
+                ) from err
+            ts.dts = dts
+            ts.creso = self._creso
+            ts.fold = fold
+            return create_timestamp_from_ts(
+                ts.value, dts, tzobj, fold, reso=self._creso
+            )
+
+        elif tzobj is not None and treat_tz_as_pytz(tzobj):
+            # replacing across a DST boundary may induce a new tzinfo object
+            # see GH#18319
+            ts_input = tzobj.localize(datetime(dts.year, dts.month, dts.day,
+                                               dts.hour, dts.min, dts.sec,
+                                               dts.us),
+                                      is_dst=not bool(fold))
+            tzobj = ts_input.tzinfo
+        else:
+            kwargs = {"year": dts.year, "month": dts.month, "day": dts.day,
+                      "hour": dts.hour, "minute": dts.min, "second": dts.sec,
+                      "microsecond": dts.us, "tzinfo": tzobj,
+                      "fold": fold}
+            ts_input = datetime(**kwargs)
+
+        ts = convert_datetime_to_tsobject(
+            ts_input, tzobj, nanos=dts.ps // 1000, reso=self._creso
+        )
+        return create_timestamp_from_ts(
+            ts.value, dts, tzobj, fold, reso=self._creso
+        )
+
+    def to_julian_date(self) -> np.float64:
+        """
+        Convert TimeStamp to a Julian Date.
+
+        This method returns the number of days as a float since
+        0 Julian date, which is noon January 1, 4713 BC.
+
+        See Also
+        --------
+        Timestamp.toordinal : Return proleptic Gregorian ordinal.
+        Timestamp.timestamp : Return POSIX timestamp as float.
+        Timestamp : Represents a single timestamp.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2020-03-14T15:32:52')
+        >>> ts.to_julian_date()
+        2458923.147824074
+        """
+        year = self._year
+        month = self.month
+        day = self.day
+        if month <= 2:
+            year -= 1
+            month += 12
+        return (day +
+                np.fix((153 * month - 457) / 5) +
+                365 * year +
+                np.floor(year / 4) -
+                np.floor(year / 100) +
+                np.floor(year / 400) +
+                1721118.5 +
+                (self.hour +
+                 self.minute / 60.0 +
+                 self.second / 3600.0 +
+                 self.microsecond / 3600.0 / 1e+6 +
+                 self._nanosecond / 3600.0 / 1e+9
+                 ) / 24.0)
+
+    def isoweekday(self):
+        """
+        Return the day of the week represented by the date.
+
+        Monday == 1 ... Sunday == 7.
+
+        See Also
+        --------
+        Timestamp.weekday : Return the day of the week with Monday=0, Sunday=6.
+        Timestamp.isocalendar : Return a tuple containing ISO year, week number
+            and weekday.
+        datetime.date.isoweekday : Equivalent method in datetime module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01 10:00:00')
+        >>> ts
+        Timestamp('2023-01-01 10:00:00')
+        >>> ts.isoweekday()
+        7
+        """
+        # same as super().isoweekday(), but that breaks because of how
+        #  we have overridden year, see note in create_timestamp_from_ts
+
+        return self.weekday() + 1
+
+    def weekday(self):
+        """
+        Return the day of the week represented by the date.
+
+        Monday == 0 ... Sunday == 6.
+
+        See Also
+        --------
+        Timestamp.dayofweek : Return the day of the week with Monday=0, Sunday=6.
+        Timestamp.isoweekday : Return the day of the week with Monday=1, Sunday=7.
+        datetime.date.weekday : Equivalent method in datetime module.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp('2023-01-01')
+        >>> ts
+        Timestamp('2023-01-01  00:00:00')
+        >>> ts.weekday()
+        6
+        """
+        # same as super().weekday(), but that breaks because of how
+        #  we have overridden year, see note in create_timestamp_from_ts
+        return ccalendar.dayofweek(self._year, self.month, self.day)
+
+
+# Aliases
+Timestamp.weekofyear = Timestamp.week
+Timestamp.daysinmonth = Timestamp.days_in_month
+
+
+# ----------------------------------------------------------------------
+# Scalar analogues to functions in vectorized.pyx
+
+
+@cython.cdivision(False)
+cdef int64_t normalize_i8_stamp(int64_t local_val, int64_t ppd) noexcept nogil:
+    """
+    Round the localized nanosecond timestamp down to the previous midnight.
+
+    Parameters
+    ----------
+    local_val : int64_t
+    ppd : int64_t
+        Periods per day in the Timestamp's resolution.
+
+    Returns
+    -------
+    int64_t
+    """
+    return local_val - (local_val % ppd)
diff --git a/pandas/pandas/_libs/tslibs/timezones.pxd b/pandas/pandas/_libs/tslibs/timezones.pxd
new file mode 100644
index 0000000000000..5629deb965a2d
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timezones.pxd
@@ -0,0 +1,23 @@
+from cpython.datetime cimport (
+    datetime,
+    timedelta,
+    tzinfo,
+)
+
+
+cdef tzinfo utc_stdlib
+
+cpdef bint is_utc(tzinfo tz)
+cdef bint is_tzlocal(tzinfo tz)
+cdef bint is_zoneinfo(tzinfo tz)
+
+cdef bint treat_tz_as_pytz(tzinfo tz)
+
+cpdef bint tz_compare(tzinfo start, tzinfo end)
+cpdef object get_timezone(tzinfo tz)
+cpdef tzinfo maybe_get_tz(object tz)
+
+cdef timedelta get_utcoffset(tzinfo tz, datetime obj)
+cpdef bint is_fixed_offset(tzinfo tz)
+
+cdef object get_dst_info(tzinfo tz)
diff --git a/pandas/pandas/_libs/tslibs/timezones.pyi b/pandas/pandas/_libs/tslibs/timezones.pyi
new file mode 100644
index 0000000000000..4e9f0c6ae6c33
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timezones.pyi
@@ -0,0 +1,21 @@
+from datetime import (
+    datetime,
+    tzinfo,
+)
+from typing import Callable
+
+import numpy as np
+
+# imported from dateutil.tz
+dateutil_gettz: Callable[[str], tzinfo]
+
+def tz_standardize(tz: tzinfo) -> tzinfo: ...
+def tz_compare(start: tzinfo | None, end: tzinfo | None) -> bool: ...
+def infer_tzinfo(
+    start: datetime | None,
+    end: datetime | None,
+) -> tzinfo | None: ...
+def maybe_get_tz(tz: str | int | np.int64 | tzinfo | None) -> tzinfo | None: ...
+def get_timezone(tz: tzinfo) -> tzinfo | str: ...
+def is_utc(tz: tzinfo | None) -> bool: ...
+def is_fixed_offset(tz: tzinfo) -> bool: ...
diff --git a/pandas/pandas/_libs/tslibs/timezones.pyx b/pandas/pandas/_libs/tslibs/timezones.pyx
new file mode 100644
index 0000000000000..36b644ffc826d
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/timezones.pyx
@@ -0,0 +1,440 @@
+from datetime import (
+    timedelta,
+    timezone,
+)
+import zoneinfo
+
+from pandas.compat._optional import import_optional_dependency
+
+from cpython.datetime cimport (
+    datetime,
+    timedelta,
+    tzinfo,
+)
+
+# dateutil compat
+
+from dateutil.tz import (
+    gettz as dateutil_gettz,
+    tzfile as _dateutil_tzfile,
+    tzlocal as _dateutil_tzlocal,
+    tzutc as _dateutil_tzutc,
+)
+import numpy as np
+
+pytz = import_optional_dependency("pytz", errors="ignore")
+
+cimport numpy as cnp
+from numpy cimport int64_t
+
+cnp.import_array()
+
+# ----------------------------------------------------------------------
+from pandas._libs.tslibs.util cimport (
+    get_nat,
+    is_integer_object,
+)
+
+
+cdef int64_t NPY_NAT = get_nat()
+cdef tzinfo utc_stdlib = timezone.utc
+cdef tzinfo utc_pytz = pytz.UTC if pytz else None
+cdef tzinfo utc_dateutil_str = dateutil_gettz("UTC")  # NB: *not* the same as tzutc()
+
+cdef tzinfo utc_zoneinfo = None
+cdef type ZoneInfo = zoneinfo.ZoneInfo
+
+
+# ----------------------------------------------------------------------
+
+cdef bint is_utc_zoneinfo(tzinfo tz):
+    # Workaround for cases with missing tzdata
+    #  https://github.com/pandas-dev/pandas/pull/46425#discussion_r830633025
+    if tz is None:
+        return False
+
+    global utc_zoneinfo
+    if utc_zoneinfo is None:
+        try:
+            utc_zoneinfo = zoneinfo.ZoneInfo("UTC")
+        except zoneinfo.ZoneInfoNotFoundError:
+            return False
+        # Warn if tzdata is too old, even if there is a system tzdata to alert
+        # users about the mismatch between local/system tzdata
+        import_optional_dependency("tzdata", errors="warn", min_version="2022.7")
+
+    return tz is utc_zoneinfo
+
+
+cpdef inline bint is_utc(tzinfo tz):
+    return (
+        tz is utc_stdlib
+        or isinstance(tz, _dateutil_tzutc)
+        or tz is utc_dateutil_str
+        or is_utc_zoneinfo(tz)
+        or (utc_pytz is not None and tz is utc_pytz)
+    )
+
+
+cdef bint is_zoneinfo(tzinfo tz):
+    return isinstance(tz, ZoneInfo)
+
+
+cdef bint is_tzlocal(tzinfo tz):
+    return isinstance(tz, _dateutil_tzlocal)
+
+
+cdef bint treat_tz_as_pytz(tzinfo tz):
+    return (hasattr(tz, "_utc_transition_times") and
+            hasattr(tz, "_transition_info"))
+
+
+cdef bint treat_tz_as_dateutil(tzinfo tz):
+    return hasattr(tz, "_trans_list") and hasattr(tz, "_trans_idx")
+
+
+# Returns str or tzinfo object
+cpdef inline object get_timezone(tzinfo tz):
+    """
+    We need to do several things here:
+    1) Distinguish between pytz and dateutil timezones
+    2) Not be over-specific (e.g. US/Eastern with/without DST is same *zone*
+       but a different tz object)
+    3) Provide something to serialize when we're storing a datetime object
+       in pytables.
+
+    We return a string prefaced with dateutil if it's a dateutil tz, else just
+    the tz name. It needs to be a string so that we can serialize it with
+    UJSON/pytables. maybe_get_tz (below) is the inverse of this process.
+    """
+    if tz is None:
+        raise TypeError("tz argument cannot be None")
+    if is_utc(tz):
+        return tz
+    elif is_zoneinfo(tz):
+        return tz.key
+    elif treat_tz_as_pytz(tz):
+        zone = tz.zone
+        if zone is None:
+            return tz
+        return zone
+    elif treat_tz_as_dateutil(tz):
+        if ".tar.gz" in tz._filename:
+            raise ValueError(
+                "Bad tz filename. Dateutil on python 3 on windows has a "
+                "bug which causes tzfile._filename to be the same for all "
+                "timezone files. Please construct dateutil timezones "
+                'implicitly by passing a string like "dateutil/Europe'
+                '/London" when you construct your pandas objects instead '
+                "of passing a timezone object. See "
+                "https://github.com/pandas-dev/pandas/pull/7362")
+        return "dateutil/" + tz._filename
+    else:
+        return tz
+
+
+cpdef inline tzinfo maybe_get_tz(object tz):
+    """
+    (Maybe) Construct a timezone object from a string. If tz is a string, use
+    it to construct a timezone object. Otherwise, just return tz.
+    """
+    if isinstance(tz, str):
+        if tz == "tzlocal()":
+            tz = _dateutil_tzlocal()
+        elif tz.startswith("dateutil/"):
+            zone = tz[9:]
+            tz = dateutil_gettz(zone)
+            # On Python 3 on Windows, the filename is not always set correctly.
+            if isinstance(tz, _dateutil_tzfile) and ".tar.gz" in tz._filename:
+                tz._filename = zone
+        elif tz[0] in {"-", "+"}:
+            hours = int(tz[0:3])
+            minutes = int(tz[0] + tz[4:6])
+            tz = timezone(timedelta(hours=hours, minutes=minutes))
+        elif tz[0:4] in {"UTC-", "UTC+"}:
+            hours = int(tz[3:6])
+            minutes = int(tz[3] + tz[7:9])
+            tz = timezone(timedelta(hours=hours, minutes=minutes))
+        elif tz == "UTC" or tz == "utc":
+            tz = utc_stdlib
+        else:
+            tz = zoneinfo.ZoneInfo(tz)
+    elif is_integer_object(tz):
+        tz = timezone(timedelta(seconds=tz))
+    elif isinstance(tz, tzinfo):
+        pass
+    elif tz is None:
+        pass
+    else:
+        raise TypeError(type(tz))
+    return tz
+
+
+def _p_tz_cache_key(tz: tzinfo):
+    """
+    Python interface for cache function to facilitate testing.
+    """
+    return tz_cache_key(tz)
+
+
+# Timezone data caches, key is the pytz string or dateutil file name.
+dst_cache = {}
+
+
+cdef object tz_cache_key(tzinfo tz):
+    """
+    Return the key in the cache for the timezone info object or None
+    if unknown.
+
+    The key is currently the tz string for pytz timezones, the filename for
+    dateutil timezones.
+
+    Notes
+    -----
+    This cannot just be the hash of a timezone object. Unfortunately, the
+    hashes of two dateutil tz objects which represent the same timezone are
+    not equal (even though the tz objects will compare equal and represent
+    the same tz file). Also, pytz objects are not always hashable so we use
+    str(tz) instead.
+    """
+    if pytz is not None and isinstance(tz, pytz.tzinfo.BaseTzInfo):
+        return tz.zone
+    elif isinstance(tz, _dateutil_tzfile):
+        if ".tar.gz" in tz._filename:
+            raise ValueError("Bad tz filename. Dateutil on python 3 on "
+                             "windows has a bug which causes tzfile._filename "
+                             "to be the same for all timezone files. Please "
+                             "construct dateutil timezones implicitly by "
+                             'passing a string like "dateutil/Europe/London" '
+                             "when you construct your pandas objects instead "
+                             "of passing a timezone object. See "
+                             "https://github.com/pandas-dev/pandas/pull/7362")
+        return "dateutil" + tz._filename
+    else:
+        return None
+
+
+# ----------------------------------------------------------------------
+# UTC Offsets
+
+
+cdef timedelta get_utcoffset(tzinfo tz, datetime obj):
+    try:
+        return tz._utcoffset
+    except AttributeError:
+        return tz.utcoffset(obj)
+
+
+cpdef inline bint is_fixed_offset(tzinfo tz):
+    if treat_tz_as_dateutil(tz):
+        if len(tz._trans_idx) == 0 and len(tz._trans_list) == 0:
+            return 1
+        else:
+            return 0
+    elif treat_tz_as_pytz(tz) and pytz is not None:
+        if (len(tz._transition_info) == 0
+                and len(tz._utc_transition_times) == 0):
+            return 1
+        else:
+            return 0
+    elif is_zoneinfo(tz):
+        return 0
+    # This also implicitly accepts datetime.timezone objects which are
+    # considered fixed
+    return 1
+
+
+cdef object _get_utc_trans_times_from_dateutil_tz(tzinfo tz):
+    """
+    Transition times in dateutil timezones are stored in local non-dst
+    time.  This code converts them to UTC. It's the reverse of the code
+    in dateutil.tz.tzfile.__init__.
+    """
+    new_trans = list(tz._trans_list)
+    last_std_offset = 0
+    for i, (trans, tti) in enumerate(zip(tz._trans_list, tz._trans_idx)):
+        if not tti.isdst:
+            last_std_offset = tti.offset
+        new_trans[i] = trans - last_std_offset
+    return new_trans
+
+
+cdef int64_t[::1] unbox_utcoffsets(object transinfo):
+    cdef:
+        Py_ssize_t i
+        cnp.npy_intp sz
+        int64_t[::1] arr
+
+    sz = len(transinfo)
+    arr = cnp.PyArray_EMPTY(1, &sz, cnp.NPY_INT64, 0)
+
+    for i in range(sz):
+        arr[i] = int(transinfo[i][0].total_seconds()) * 1_000_000_000
+
+    return arr
+
+
+# ----------------------------------------------------------------------
+# Daylight Savings
+
+
+cdef object get_dst_info(tzinfo tz):
+    """
+    Returns
+    -------
+    ndarray[int64_t]
+        Nanosecond UTC times of DST transitions.
+    ndarray[int64_t]
+        Nanosecond UTC offsets corresponding to DST transitions.
+    str
+        Describing the type of tzinfo object.
+    """
+    cache_key = tz_cache_key(tz)
+    if cache_key is None:
+        # e.g. pytz.FixedOffset, matplotlib.dates._UTC,
+        # psycopg2.tz.FixedOffsetTimezone
+        num = int(get_utcoffset(tz, None).total_seconds()) * 1_000_000_000
+        # If we have e.g. ZoneInfo here, the get_utcoffset call will return None,
+        #  so the total_seconds() call will raise AttributeError.
+        return (np.array([NPY_NAT + 1], dtype=np.int64),
+                np.array([num], dtype=np.int64),
+                "unknown")
+
+    if cache_key not in dst_cache:
+        if treat_tz_as_pytz(tz):
+            trans = np.array(tz._utc_transition_times, dtype="M8[ns]")
+            trans = trans.view("i8")
+            if tz._utc_transition_times[0].year == 1:
+                trans[0] = NPY_NAT + 1
+            deltas = unbox_utcoffsets(tz._transition_info)
+            typ = "pytz"
+
+        elif treat_tz_as_dateutil(tz):
+            if len(tz._trans_list):
+                # get utc trans times
+                trans_list = _get_utc_trans_times_from_dateutil_tz(tz)
+                trans = np.hstack([
+                    np.array([0], dtype="M8[s]"),  # place holder for 1st item
+                    np.array(trans_list, dtype="M8[s]")]).astype(
+                    "M8[ns]")  # all trans listed
+                trans = trans.view("i8")
+                trans[0] = NPY_NAT + 1
+
+                # deltas
+                deltas = np.array([v.offset for v in (
+                    tz._ttinfo_before,) + tz._trans_idx], dtype="i8")
+                deltas *= 1_000_000_000
+                typ = "dateutil"
+
+            elif is_fixed_offset(tz):
+                trans = np.array([NPY_NAT + 1], dtype=np.int64)
+                deltas = np.array([tz._ttinfo_std.offset],
+                                  dtype="i8") * 1_000_000_000
+                typ = "fixed"
+            else:
+                # 2018-07-12 this is not reached in the tests, and this case
+                # is not handled in any of the functions that call
+                # get_dst_info.  If this case _were_ hit the calling
+                # functions would then hit an IndexError because they assume
+                # `deltas` is non-empty.
+                # (under the just-deleted code that returned empty arrays)
+                raise AssertionError("dateutil tzinfo is not a FixedOffset "
+                                     "and has an empty `_trans_list`.", tz)
+        else:
+            # static tzinfo, we can get here with pytz.StaticTZInfo
+            #  which are not caught by treat_tz_as_pytz
+            trans = np.array([NPY_NAT + 1], dtype=np.int64)
+            num = int(get_utcoffset(tz, None).total_seconds()) * 1_000_000_000
+            deltas = np.array([num], dtype=np.int64)
+            typ = "static"
+
+        dst_cache[cache_key] = (trans, deltas, typ)
+
+    return dst_cache[cache_key]
+
+
+def infer_tzinfo(datetime start, datetime end):
+    if start is not None and end is not None:
+        tz = start.tzinfo
+        if not tz_compare(tz, end.tzinfo):
+            raise AssertionError(f"Inputs must both have the same timezone, "
+                                 f"{tz} != {end.tzinfo}")
+    elif start is not None:
+        tz = start.tzinfo
+    elif end is not None:
+        tz = end.tzinfo
+    else:
+        tz = None
+    return tz
+
+
+cpdef bint tz_compare(tzinfo start, tzinfo end):
+    """
+    Compare string representations of timezones
+
+    The same timezone can be represented as different instances of
+    timezones. For example
+    `<DstTzInfo 'Europe/Paris' LMT+0:09:00 STD>` and
+    `<DstTzInfo 'Europe/Paris' CET+1:00:00 STD>` are essentially same
+    timezones but aren't evaluated such, but the string representation
+    for both of these is `'Europe/Paris'`.
+
+    This exists only to add a notion of equality to pytz-style zones
+    that is compatible with the notion of equality expected of tzinfo
+    subclasses.
+
+    Parameters
+    ----------
+    start : tzinfo
+    end : tzinfo
+
+    Returns:
+    -------
+    bool
+    """
+    # GH 18523
+    if is_utc(start):
+        # GH#38851 consider pytz/dateutil/stdlib UTCs as equivalent
+        return is_utc(end)
+    elif is_utc(end):
+        # Ensure we don't treat tzlocal as equal to UTC when running in UTC
+        return False
+    elif start is None or end is None:
+        return start is None and end is None
+    return get_timezone(start) == get_timezone(end)
+
+
+def tz_standardize(tz: tzinfo) -> tzinfo:
+    """
+    If the passed tz is a pytz timezone object, "normalize" it to the a
+    consistent version
+
+    Parameters
+    ----------
+    tz : tzinfo
+
+    Returns
+    -------
+    tzinfo
+
+    Examples
+    --------
+    >>> from datetime import datetime
+    >>> from pytz import timezone
+    >>> tz = timezone('US/Pacific').normalize(
+    ...     datetime(2014, 1, 1, tzinfo=pytz.utc)
+    ... ).tzinfo
+    >>> tz
+    <DstTzInfo 'US/Pacific' PST-1 day, 16:00:00 STD>
+    >>> tz_standardize(tz)
+    <DstTzInfo 'US/Pacific' LMT-1 day, 16:07:00 STD>
+
+    >>> tz = timezone('US/Pacific')
+    >>> tz
+    <DstTzInfo 'US/Pacific' LMT-1 day, 16:07:00 STD>
+    >>> tz_standardize(tz)
+    <DstTzInfo 'US/Pacific' LMT-1 day, 16:07:00 STD>
+    """
+    if treat_tz_as_pytz(tz):
+        return pytz.timezone(str(tz))
+    return tz
diff --git a/pandas/pandas/_libs/tslibs/tzconversion.pxd b/pandas/pandas/_libs/tslibs/tzconversion.pxd
new file mode 100644
index 0000000000000..7c1dd04e2b2cc
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/tzconversion.pxd
@@ -0,0 +1,39 @@
+from cpython.datetime cimport tzinfo
+from numpy cimport (
+    int64_t,
+    intp_t,
+    ndarray,
+)
+
+from pandas._libs.tslibs.np_datetime cimport NPY_DATETIMEUNIT
+
+
+cpdef int64_t tz_convert_from_utc_single(
+    int64_t utc_val, tzinfo tz, NPY_DATETIMEUNIT creso=*
+) except? -1
+cdef int64_t tz_localize_to_utc_single(
+    int64_t val,
+    tzinfo tz,
+    object ambiguous=*,
+    object nonexistent=*,
+    NPY_DATETIMEUNIT creso=*,
+) except? -1
+
+
+cdef class Localizer:
+    cdef:
+        tzinfo tz
+        NPY_DATETIMEUNIT _creso
+        bint use_utc, use_fixed, use_tzlocal, use_dst, use_pytz
+        ndarray trans
+        Py_ssize_t ntrans
+        const int64_t[::1] deltas
+        int64_t delta
+        int64_t* tdata
+
+    cdef int64_t utc_val_to_local_val(
+        self,
+        int64_t utc_val,
+        Py_ssize_t* pos,
+        bint* fold=?,
+    ) except? -1
diff --git a/pandas/pandas/_libs/tslibs/tzconversion.pyi b/pandas/pandas/_libs/tslibs/tzconversion.pyi
new file mode 100644
index 0000000000000..2108fa0f35547
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/tzconversion.pyi
@@ -0,0 +1,21 @@
+from datetime import (
+    timedelta,
+    tzinfo,
+)
+from typing import Iterable
+
+import numpy as np
+
+from pandas._typing import npt
+
+# tz_convert_from_utc_single exposed for testing
+def tz_convert_from_utc_single(
+    utc_val: np.int64, tz: tzinfo, creso: int = ...
+) -> np.int64: ...
+def tz_localize_to_utc(
+    vals: npt.NDArray[np.int64],
+    tz: tzinfo | None,
+    ambiguous: str | bool | Iterable[bool] | None = ...,
+    nonexistent: str | timedelta | np.timedelta64 | None = ...,
+    creso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int64]: ...
diff --git a/pandas/pandas/_libs/tslibs/tzconversion.pyx b/pandas/pandas/_libs/tslibs/tzconversion.pyx
new file mode 100644
index 0000000000000..c100f315e9a19
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/tzconversion.pyx
@@ -0,0 +1,839 @@
+"""
+timezone conversion
+"""
+cimport cython
+from cpython.datetime cimport (
+    PyDelta_Check,
+    datetime,
+    datetime_new,
+    import_datetime,
+    timedelta,
+    tzinfo,
+)
+from cython cimport Py_ssize_t
+
+import_datetime()
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    int64_t,
+    intp_t,
+    ndarray,
+    uint8_t,
+)
+
+cnp.import_array()
+
+from pandas._libs.tslibs.dtypes cimport (
+    periods_per_day,
+    periods_per_second,
+)
+from pandas._libs.tslibs.nattype cimport NPY_NAT
+from pandas._libs.tslibs.np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    pandas_datetime_to_datetimestruct,
+    pydatetime_to_dt64,
+)
+
+import_pandas_datetime()
+
+from pandas._libs.tslibs.timezones cimport (
+    get_dst_info,
+    is_fixed_offset,
+    is_tzlocal,
+    is_utc,
+    is_zoneinfo,
+    utc_stdlib,
+)
+
+
+cdef const int64_t[::1] _deltas_placeholder = np.array([], dtype=np.int64)
+
+
+@cython.freelist(16)
+@cython.final
+cdef class Localizer:
+    # cdef:
+    #    tzinfo tz
+    #    NPY_DATETIMEUNIT _creso
+    #    bint use_utc, use_fixed, use_tzlocal, use_dst, use_pytz
+    #    ndarray trans
+    #    Py_ssize_t ntrans
+    #    const int64_t[::1] deltas
+    #    int64_t delta
+    #    int64_t* tdata
+
+    @cython.initializedcheck(False)
+    @cython.boundscheck(False)
+    def __cinit__(self, tzinfo tz, NPY_DATETIMEUNIT creso):
+        self.tz = tz
+        self._creso = creso
+        self.use_utc = self.use_tzlocal = self.use_fixed = False
+        self.use_dst = self.use_pytz = False
+        self.ntrans = -1  # placeholder
+        self.delta = -1  # placeholder
+        self.deltas = _deltas_placeholder
+        self.tdata = NULL
+
+        if is_utc(tz) or tz is None:
+            self.use_utc = True
+
+        elif is_tzlocal(tz) or is_zoneinfo(tz):
+            self.use_tzlocal = True
+
+        else:
+            trans, deltas, typ = get_dst_info(tz)
+            if creso != NPY_DATETIMEUNIT.NPY_FR_ns:
+                # NB: using floordiv here is implicitly assuming we will
+                #  never see trans or deltas that are not an integer number
+                #  of seconds.
+                # TODO: avoid these np.array calls
+                if creso == NPY_DATETIMEUNIT.NPY_FR_us:
+                    trans = np.array(trans) // 1_000
+                    deltas = np.array(deltas) // 1_000
+                elif creso == NPY_DATETIMEUNIT.NPY_FR_ms:
+                    trans = np.array(trans) // 1_000_000
+                    deltas = np.array(deltas) // 1_000_000
+                elif creso == NPY_DATETIMEUNIT.NPY_FR_s:
+                    trans = np.array(trans) // 1_000_000_000
+                    deltas = np.array(deltas) // 1_000_000_000
+                else:
+                    raise NotImplementedError(creso)
+
+            self.trans = trans
+            self.ntrans = self.trans.shape[0]
+            self.deltas = deltas
+
+            if typ != "pytz" and typ != "dateutil":
+                # static/fixed; in this case we know that len(delta) == 1
+                self.use_fixed = True
+                self.delta = deltas[0]
+            else:
+                self.use_dst = True
+                if typ == "pytz":
+                    self.use_pytz = True
+                self.tdata = <int64_t*>cnp.PyArray_DATA(trans)
+
+    @cython.boundscheck(False)
+    cdef int64_t utc_val_to_local_val(
+        self, int64_t utc_val, Py_ssize_t* pos, bint* fold=NULL
+    ) except? -1:
+        if self.use_utc:
+            return utc_val
+        elif self.use_tzlocal:
+            return utc_val + _tz_localize_using_tzinfo_api(
+                utc_val, self.tz, to_utc=False, creso=self._creso, fold=fold
+            )
+        elif self.use_fixed:
+            return utc_val + self.delta
+        else:
+            pos[0] = bisect_right_i8(self.tdata, utc_val, self.ntrans) - 1
+            if fold is not NULL:
+                fold[0] = _infer_dateutil_fold(
+                    utc_val, self.trans, self.deltas, pos[0]
+                )
+
+            return utc_val + self.deltas[pos[0]]
+
+
+cdef int64_t tz_localize_to_utc_single(
+    int64_t val,
+    tzinfo tz,
+    object ambiguous=None,
+    object nonexistent=None,
+    NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_ns,
+) except? -1:
+    """See tz_localize_to_utc.__doc__"""
+    cdef:
+        int64_t delta
+        int64_t[::1] deltas
+
+    if val == NPY_NAT:
+        return val
+
+    elif is_utc(tz) or tz is None:
+        return val
+
+    elif is_tzlocal(tz):
+        return val - _tz_localize_using_tzinfo_api(val, tz, to_utc=True, creso=creso)
+
+    elif is_fixed_offset(tz):
+        _, deltas, _ = get_dst_info(tz)
+        delta = deltas[0]
+        # TODO: de-duplicate with Localizer.__init__
+        if creso != NPY_DATETIMEUNIT.NPY_FR_ns:
+            if creso == NPY_DATETIMEUNIT.NPY_FR_us:
+                delta = delta // 1000
+            elif creso == NPY_DATETIMEUNIT.NPY_FR_ms:
+                delta = delta // 1_000_000
+            elif creso == NPY_DATETIMEUNIT.NPY_FR_s:
+                delta = delta // 1_000_000_000
+
+        return val - delta
+
+    else:
+        return tz_localize_to_utc(
+            np.array([val], dtype="i8"),
+            tz,
+            ambiguous=ambiguous,
+            nonexistent=nonexistent,
+            creso=creso,
+        )[0]
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def tz_localize_to_utc(
+    ndarray[int64_t] vals,
+    tzinfo tz,
+    object ambiguous=None,
+    object nonexistent=None,
+    NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_ns,
+):
+    """
+    Localize tzinfo-naive i8 to given time zone. If
+    there are ambiguities in the values, raise ValueError.
+
+    Parameters
+    ----------
+    vals : ndarray[int64_t]
+    tz : tzinfo or None
+    ambiguous : str, bool, or arraylike
+        When clocks moved backward due to DST, ambiguous times may arise.
+        For example in Central European Time (UTC+01), when going from 03:00
+        DST to 02:00 non-DST, 02:30:00 local time occurs both at 00:30:00 UTC
+        and at 01:30:00 UTC. In such a situation, the `ambiguous` parameter
+        dictates how ambiguous times should be handled.
+
+        - 'infer' will attempt to infer fall dst-transition hours based on
+          order
+        - bool-ndarray where True signifies a DST time, False signifies a
+          non-DST time (note that this flag is only applicable for ambiguous
+          times, but the array must have the same length as vals)
+        - bool if True, treat all vals as DST. If False, treat them as non-DST
+        - 'NaT' will return NaT where there are ambiguous times
+
+    nonexistent : {None, "NaT", "shift_forward", "shift_backward", "raise", \
+timedelta-like}
+        How to handle non-existent times when converting wall times to UTC
+    creso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    localized : ndarray[int64_t]
+    """
+
+    if tz is None or is_utc(tz) or vals.size == 0:
+        # Fastpath, avoid overhead of creating Localizer
+        return vals.copy()
+
+    cdef:
+        ndarray[uint8_t, cast=True] ambiguous_array
+        Py_ssize_t i, n = vals.shape[0]
+        Py_ssize_t delta_idx_offset, delta_idx
+        int64_t v, left, right, val, new_local, remaining_mins
+        int64_t first_delta, delta
+        int64_t shift_delta = 0
+        ndarray[int64_t] result_a, result_b, dst_hours
+        int64_t[::1] result
+        bint is_zi = False
+        bint infer_dst = False, is_dst = False, fill = False
+        bint shift_forward = False, shift_backward = False
+        bint fill_nonexist = False
+        str stamp
+        Localizer info = Localizer(tz, creso=creso)
+        int64_t pph = periods_per_day(creso) // 24
+        int64_t pps = periods_per_second(creso)
+        npy_datetimestruct dts
+
+    # Vectorized version of DstTzInfo.localize
+
+    # silence false-positive compiler warning
+    ambiguous_array = np.empty(0, dtype=bool)
+    if isinstance(ambiguous, str):
+        if ambiguous == "infer":
+            infer_dst = True
+        elif ambiguous == "NaT":
+            fill = True
+    elif isinstance(ambiguous, bool):
+        is_dst = True
+        if ambiguous:
+            ambiguous_array = np.ones(len(vals), dtype=bool)
+        else:
+            ambiguous_array = np.zeros(len(vals), dtype=bool)
+    elif hasattr(ambiguous, "__iter__"):
+        is_dst = True
+        if len(ambiguous) != len(vals):
+            raise ValueError("Length of ambiguous bool-array must be "
+                             "the same size as vals")
+        ambiguous_array = np.asarray(ambiguous, dtype=bool)
+
+    if nonexistent == "NaT":
+        fill_nonexist = True
+    elif nonexistent == "shift_forward":
+        shift_forward = True
+    elif nonexistent == "shift_backward":
+        shift_backward = True
+    elif PyDelta_Check(nonexistent):
+        from .timedeltas import delta_to_nanoseconds
+        shift_delta = delta_to_nanoseconds(nonexistent, reso=creso)
+    elif nonexistent not in ("raise", None):
+        msg = ("nonexistent must be one of {'NaT', 'raise', 'shift_forward', "
+               "shift_backwards} or a timedelta object")
+        raise ValueError(msg)
+
+    result = cnp.PyArray_EMPTY(vals.ndim, vals.shape, cnp.NPY_INT64, 0)
+
+    if info.use_tzlocal and not is_zoneinfo(tz):
+        for i in range(n):
+            v = vals[i]
+            if v == NPY_NAT:
+                result[i] = NPY_NAT
+            else:
+                result[i] = v - _tz_localize_using_tzinfo_api(
+                    v, tz, to_utc=True, creso=creso
+                )
+        return result.base  # to return underlying ndarray
+
+    elif info.use_fixed:
+        delta = info.delta
+        for i in range(n):
+            v = vals[i]
+            if v == NPY_NAT:
+                result[i] = NPY_NAT
+            else:
+                result[i] = v - delta
+        return result.base  # to return underlying ndarray
+
+    # Determine whether each date lies left of the DST transition (store in
+    # result_a) or right of the DST transition (store in result_b)
+    if is_zoneinfo(tz):
+        is_zi = True
+        result_a, result_b =_get_utc_bounds_zoneinfo(
+            vals, tz, creso=creso
+        )
+    else:
+        result_a, result_b =_get_utc_bounds(
+            vals, info.tdata, info.ntrans, info.deltas, creso=creso
+        )
+
+    # silence false-positive compiler warning
+    dst_hours = np.empty(0, dtype=np.int64)
+    if infer_dst:
+        dst_hours = _get_dst_hours(vals, result_a, result_b, creso=creso)
+
+    # Pre-compute delta_idx_offset that will be used if we go down non-existent
+    #  paths.
+    # Shift the delta_idx by if the UTC offset of
+    # the target tz is greater than 0 and we're moving forward
+    # or vice versa
+    # TODO: delta_idx_offset and info.deltas are needed for zoneinfo timezones,
+    # but are not applicable for all timezones. Setting the former to 0 and
+    # length checking the latter avoids UB, but this could use a larger refactor
+    delta_idx_offset = 0
+    if len(info.deltas):
+        first_delta = info.deltas[0]
+        if (shift_forward or shift_delta > 0) and first_delta > 0:
+            delta_idx_offset = 1
+        elif (shift_backward or shift_delta < 0) and first_delta < 0:
+            delta_idx_offset = 1
+
+    for i in range(n):
+        val = vals[i]
+        left = result_a[i]
+        right = result_b[i]
+        if val == NPY_NAT:
+            # TODO: test with non-nano
+            result[i] = val
+        elif left != NPY_NAT and right != NPY_NAT:
+            if left == right:
+                # TODO: test with non-nano
+                result[i] = left
+            else:
+                if infer_dst and dst_hours[i] != NPY_NAT:
+                    # TODO: test with non-nano
+                    result[i] = dst_hours[i]
+                elif is_dst:
+                    if ambiguous_array[i]:
+                        result[i] = left
+                    else:
+                        result[i] = right
+                elif fill:
+                    # TODO: test with non-nano; parametrize test_dt_round_tz_ambiguous
+                    result[i] = NPY_NAT
+                else:
+                    stamp = _render_tstamp(val, creso=creso)
+                    raise ValueError(
+                        f"Cannot infer dst time from {stamp}, try using the "
+                        "'ambiguous' argument"
+                    )
+        elif left != NPY_NAT:
+            result[i] = left
+        elif right != NPY_NAT:
+            # TODO: test with non-nano
+            result[i] = right
+        else:
+            # Handle nonexistent times
+            if shift_forward or shift_backward or shift_delta != 0:
+                # Shift the nonexistent time to the closest existing time
+                remaining_mins = val % pph
+                if shift_delta != 0:
+                    # Validate that we don't relocalize on another nonexistent
+                    # time
+                    if -1 < shift_delta + remaining_mins < pph:
+                        raise ValueError(
+                            "The provided timedelta will relocalize on a "
+                            f"nonexistent time: {nonexistent}"
+                        )
+                    new_local = val + shift_delta
+                elif shift_forward:
+                    new_local = val + (pph - remaining_mins)
+                else:
+                    # Subtract 1 since the beginning hour is _inclusive_ of
+                    # nonexistent times
+                    new_local = val - remaining_mins - 1
+
+                if is_zi:
+                    # use the same construction as in _get_utc_bounds_zoneinfo
+                    pandas_datetime_to_datetimestruct(new_local, creso, &dts)
+                    extra = (dts.ps // 1000) * (pps // 1_000_000_000)
+
+                    dt = datetime_new(dts.year, dts.month, dts.day, dts.hour,
+                                      dts.min, dts.sec, dts.us, None)
+
+                    if shift_forward or shift_delta > 0:
+                        dt = dt.replace(tzinfo=tz, fold=1)
+                    else:
+                        dt = dt.replace(tzinfo=tz, fold=0)
+                    dt = dt.astimezone(utc_stdlib)
+                    dt = dt.replace(tzinfo=None)
+                    result[i] = pydatetime_to_dt64(dt, &dts, creso) + extra
+
+                else:
+                    delta_idx = bisect_right_i8(info.tdata, new_local, info.ntrans)
+                    # Logic similar to the precompute section. But check the current
+                    # delta in case we are moving between UTC+0 and non-zero timezone
+                    if (shift_forward or shift_delta > 0) and \
+                       info.deltas[delta_idx - 1] >= 0:
+                        delta_idx = delta_idx - 1
+                    else:
+                        delta_idx = delta_idx - delta_idx_offset
+                    result[i] = new_local - info.deltas[delta_idx]
+            elif fill_nonexist:
+                result[i] = NPY_NAT
+            else:
+                stamp = _render_tstamp(val, creso=creso)
+                raise ValueError(
+                    f"{stamp} is a nonexistent time due to daylight savings time. "
+                    "Try using the 'nonexistent' argument."
+                )
+
+    return result.base  # .base to get underlying ndarray
+
+
+cdef Py_ssize_t bisect_right_i8(
+    const int64_t *data,
+    int64_t val,
+    Py_ssize_t n
+) noexcept:
+    # Caller is responsible for checking n > 0
+    # This looks very similar to local_search_right in the ndarray.searchsorted
+    #  implementation.
+    cdef:
+        Py_ssize_t pivot, left = 0, right = n
+
+    # edge cases
+    if val > data[n - 1]:
+        return n
+
+    # Caller is responsible for ensuring 'val >= data[0]'. This is
+    #  ensured by the fact that 'data' comes from get_dst_info where data[0]
+    #  is *always* NPY_NAT+1. If that ever changes, we will need to restore
+    #  the following disabled check.
+    # if val < data[0]:
+    #    return 0
+
+    while left < right:
+        pivot = left + (right - left) // 2
+
+        if data[pivot] <= val:
+            left = pivot + 1
+        else:
+            right = pivot
+
+    return left
+
+
+cdef str _render_tstamp(int64_t val, NPY_DATETIMEUNIT creso):
+    """ Helper function to render exception messages"""
+    from pandas._libs.tslibs.timestamps import Timestamp
+    ts = Timestamp._from_value_and_reso(val, creso, None)
+    return str(ts)
+
+
+cdef _get_utc_bounds(
+    ndarray[int64_t] vals,
+    const int64_t* tdata,
+    Py_ssize_t ntrans,
+    const int64_t[::1] deltas,
+    NPY_DATETIMEUNIT creso,
+):
+    # Determine whether each date lies left of the DST transition (store in
+    # result_a) or right of the DST transition (store in result_b)
+
+    cdef:
+        ndarray[int64_t] result_a, result_b
+        Py_ssize_t i, n = vals.size
+        int64_t val, v_left, v_right
+        Py_ssize_t isl, isr, pos_left, pos_right
+        int64_t ppd = periods_per_day(creso)
+
+    result_a = cnp.PyArray_EMPTY(vals.ndim, vals.shape, cnp.NPY_INT64, 0)
+    result_b = cnp.PyArray_EMPTY(vals.ndim, vals.shape, cnp.NPY_INT64, 0)
+
+    for i in range(n):
+        # This loops resembles the "Find the two best possibilities" block
+        #  in pytz's DstTZInfo.localize method.
+        result_a[i] = NPY_NAT
+        result_b[i] = NPY_NAT
+
+        val = vals[i]
+        if val == NPY_NAT:
+            continue
+
+        # TODO: be careful of overflow in val-ppd
+        isl = bisect_right_i8(tdata, val - ppd, ntrans) - 1
+        if isl < 0:
+            isl = 0
+
+        v_left = val - deltas[isl]
+        pos_left = bisect_right_i8(tdata, v_left, ntrans) - 1
+        # timestamp falls to the left side of the DST transition
+        if v_left + deltas[pos_left] == val:
+            result_a[i] = v_left
+
+        # TODO: be careful of overflow in val+ppd
+        isr = bisect_right_i8(tdata, val + ppd, ntrans) - 1
+        if isr < 0:
+            isr = 0
+
+        v_right = val - deltas[isr]
+        pos_right = bisect_right_i8(tdata, v_right, ntrans) - 1
+        # timestamp falls to the right side of the DST transition
+        if v_right + deltas[pos_right] == val:
+            result_b[i] = v_right
+
+    return result_a, result_b
+
+
+cdef _get_utc_bounds_zoneinfo(ndarray vals, tz, NPY_DATETIMEUNIT creso):
+    """
+    For each point in 'vals', find the UTC time that it corresponds to if
+    with fold=0 and fold=1. In non-ambiguous cases, these will match.
+
+    Parameters
+    ----------
+    vals : ndarray[int64_t]
+    tz : ZoneInfo
+    creso : NPY_DATETIMEUNIT
+
+    Returns
+    -------
+    ndarray[int64_t]
+    ndarray[int64_t]
+    """
+    cdef:
+        Py_ssize_t i, n = vals.size
+        npy_datetimestruct dts
+        datetime dt, rt, left, right, aware, as_utc
+        int64_t val, pps = periods_per_second(creso)
+        ndarray result_a, result_b
+
+    result_a = cnp.PyArray_EMPTY(vals.ndim, vals.shape, cnp.NPY_INT64, 0)
+    result_b = cnp.PyArray_EMPTY(vals.ndim, vals.shape, cnp.NPY_INT64, 0)
+
+    for i in range(n):
+        val = vals[i]
+        if val == NPY_NAT:
+            result_a[i] = NPY_NAT
+            result_b[i] = NPY_NAT
+            continue
+
+        pandas_datetime_to_datetimestruct(val, creso, &dts)
+        # casting to pydatetime drops nanoseconds etc, which we will
+        #  need to re-add later as 'extra'
+        extra = (dts.ps // 1000) * (pps // 1_000_000_000)
+
+        dt = datetime_new(dts.year, dts.month, dts.day, dts.hour,
+                          dts.min, dts.sec, dts.us, None)
+
+        aware = dt.replace(tzinfo=tz)
+        as_utc = aware.astimezone(utc_stdlib)
+        rt = as_utc.astimezone(tz)
+        if aware != rt:
+            # AFAICT this means that 'aware' is non-existent
+            # TODO: better way to check this?
+            #  mail.python.org/archives/list/datetime-sig@python.org/
+            #  thread/57Y3IQAASJOKHX4D27W463XTZIS2NR3M/
+            result_a[i] = NPY_NAT
+        else:
+            left = as_utc.replace(tzinfo=None)
+            result_a[i] = pydatetime_to_dt64(left, &dts, creso) + extra
+
+        aware = dt.replace(fold=1, tzinfo=tz)
+        as_utc = aware.astimezone(utc_stdlib)
+        rt = as_utc.astimezone(tz)
+        if aware != rt:
+            result_b[i] = NPY_NAT
+        else:
+            right = as_utc.replace(tzinfo=None)
+            result_b[i] = pydatetime_to_dt64(right, &dts, creso) + extra
+
+    return result_a, result_b
+
+
+@cython.boundscheck(False)
+cdef ndarray[int64_t] _get_dst_hours(
+    # vals, creso only needed here to potential render an exception message
+    const int64_t[:] vals,
+    ndarray[int64_t] result_a,
+    ndarray[int64_t] result_b,
+    NPY_DATETIMEUNIT creso,
+):
+    cdef:
+        Py_ssize_t i, n = vals.shape[0]
+        ndarray[uint8_t, cast=True] mismatch
+        ndarray[int64_t] delta, dst_hours
+        ndarray[intp_t] switch_idxs, trans_idx, grp, a_idx, b_idx, one_diff
+        # TODO: Can uncomment when numpy >=2 is the minimum
+        # tuple trans_grp
+        intp_t switch_idx
+        int64_t left, right
+
+    dst_hours = cnp.PyArray_EMPTY(result_a.ndim, result_a.shape, cnp.NPY_INT64, 0)
+    dst_hours[:] = NPY_NAT
+
+    mismatch = cnp.PyArray_ZEROS(result_a.ndim, result_a.shape, cnp.NPY_BOOL, 0)
+
+    for i in range(n):
+        left = result_a[i]
+        right = result_b[i]
+
+        # Get the ambiguous hours (given the above, these are the hours
+        # where result_a != result_b and neither of them are NAT)
+        if left != right and left != NPY_NAT and right != NPY_NAT:
+            mismatch[i] = 1
+
+    trans_idx = mismatch.nonzero()[0]
+
+    if trans_idx.size == 1:
+        # see test_tz_localize_to_utc_ambiguous_infer
+        stamp = _render_tstamp(vals[trans_idx[0]], creso=creso)
+        raise ValueError(
+            f"Cannot infer dst time from {stamp} as there "
+            "are no repeated times"
+        )
+
+    # Split the array into contiguous chunks (where the difference between
+    # indices is 1).  These are effectively dst transitions in different
+    # years which is useful for checking that there is not an ambiguous
+    # transition in an individual year.
+    if trans_idx.size > 0:
+        one_diff = np.where(np.diff(trans_idx) != 1)[0] + 1
+        trans_grp = np.array_split(trans_idx, one_diff)
+
+        # Iterate through each day, if there are no hours where the
+        # delta is negative (indicates a repeat of hour) the switch
+        # cannot be inferred
+        for grp in trans_grp:
+
+            delta = np.diff(result_a[grp])
+            if grp.size == 1 or np.all(delta > 0):
+                # see test_tz_localize_to_utc_ambiguous_infer
+                stamp = _render_tstamp(vals[grp[0]], creso=creso)
+                raise ValueError(
+                    f"{stamp} is an ambiguous time and cannot be inferred."
+                )
+
+            # Find the index for the switch and pull from a for dst and b
+            # for standard
+            switch_idxs = (delta <= 0).nonzero()[0]
+            if switch_idxs.size > 1:
+                # see test_tz_localize_to_utc_ambiguous_infer
+                raise ValueError(
+                    f"There are {switch_idxs.size} dst switches when "
+                    "there should only be 1."
+                )
+
+            switch_idx = switch_idxs[0] + 1
+            # Pull the only index and adjust
+            a_idx = grp[:switch_idx]
+            b_idx = grp[switch_idx:]
+            dst_hours[grp] = np.hstack((result_a[a_idx], result_b[b_idx]))
+
+    return dst_hours
+
+
+# ----------------------------------------------------------------------
+# Timezone Conversion
+
+cpdef int64_t tz_convert_from_utc_single(
+    int64_t utc_val, tzinfo tz, NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_ns
+) except? -1:
+    """
+    Convert the val (in i8) from UTC to tz
+
+    This is a single value version of tz_convert_from_utc.
+
+    Parameters
+    ----------
+    utc_val : int64
+    tz : tzinfo
+    creso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    converted: int64
+    """
+    cdef:
+        Localizer info = Localizer(tz, creso=creso)
+        Py_ssize_t pos
+
+    # Note: caller is responsible for ensuring utc_val != NPY_NAT
+    return info.utc_val_to_local_val(utc_val, &pos)
+
+
+# OSError may be thrown by tzlocal on windows at or close to 1970-01-01
+#  see https://github.com/pandas-dev/pandas/pull/37591#issuecomment-720628241
+cdef int64_t _tz_localize_using_tzinfo_api(
+    int64_t val,
+    tzinfo tz,
+    bint to_utc=True,
+    NPY_DATETIMEUNIT creso=NPY_DATETIMEUNIT.NPY_FR_ns,
+    bint* fold=NULL,
+) except? -1:
+    """
+    Convert the i8 representation of a datetime from a general-case timezone to
+    UTC, or vice-versa using the datetime/tzinfo API.
+
+    Private, not intended for use outside of tslibs.tzconversion.
+
+    Parameters
+    ----------
+    val : int64_t
+    tz : tzinfo
+    to_utc : bint
+        True if converting _to_ UTC, False if going the other direction.
+    creso : NPY_DATETIMEUNIT
+    fold : bint*, default NULL
+        pointer to fold: whether datetime ends up in a fold or not
+        after adjustment.
+        Only passed with to_utc=False.
+
+    Returns
+    -------
+    delta : int64_t
+        Value to add when converting from utc, subtract when converting to utc.
+
+    Notes
+    -----
+    Sets fold by pointer
+    """
+    cdef:
+        npy_datetimestruct dts
+        datetime dt
+        int64_t delta
+        timedelta td
+        int64_t pps = periods_per_second(creso)
+
+    pandas_datetime_to_datetimestruct(val, creso, &dts)
+
+    # datetime_new is cython-optimized constructor
+    if not to_utc:
+        # tz.utcoffset only makes sense if datetime
+        # is _wall time_, so if val is a UTC timestamp convert to wall time
+        dt = _astimezone(dts, tz)
+
+        if fold is not NULL:
+            # NB: fold is only passed with to_utc=False
+            fold[0] = dt.fold
+    else:
+        dt = datetime_new(dts.year, dts.month, dts.day, dts.hour,
+                          dts.min, dts.sec, dts.us, None)
+
+    td = tz.utcoffset(dt)
+    delta = int(td.total_seconds() * pps)
+    return delta
+
+
+cdef datetime _astimezone(npy_datetimestruct dts, tzinfo tz):
+    """
+    Optimized equivalent to:
+
+    dt = datetime(dts.year, dts.month, dts.day, dts.hour,
+                  dts.min, dts.sec, dts.us, utc_stdlib)
+    dt = dt.astimezone(tz)
+
+    Derived from the datetime.astimezone implementation at
+    https://github.com/python/cpython/blob/main/Modules/_datetimemodule.c#L6187
+
+    NB: we are assuming tz is not None.
+    """
+    cdef:
+        datetime result
+
+    result = datetime_new(dts.year, dts.month, dts.day, dts.hour,
+                          dts.min, dts.sec, dts.us, tz)
+    return tz.fromutc(result)
+
+
+# NB: relies on dateutil internals, subject to change.
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cdef bint _infer_dateutil_fold(
+    int64_t value,
+    const int64_t[::1] trans,
+    const int64_t[::1] deltas,
+    Py_ssize_t pos,
+):
+    """
+    Infer _TSObject fold property from value by assuming 0 and then setting
+    to 1 if necessary.
+
+    Parameters
+    ----------
+    value : int64_t
+    trans : ndarray[int64_t]
+        ndarray of offset transition points in nanoseconds since epoch.
+    deltas : int64_t[:]
+        array of offsets corresponding to transition points in trans.
+    pos : Py_ssize_t
+        Position of the last transition point before taking fold into account.
+
+    Returns
+    -------
+    bint
+        Due to daylight saving time, one wall clock time can occur twice
+        when shifting from summer to winter time; fold describes whether the
+        datetime-like corresponds  to the first (0) or the second time (1)
+        the wall clock hits the ambiguous time
+
+    References
+    ----------
+    .. [1] "PEP 495 - Local Time Disambiguation"
+           https://www.python.org/dev/peps/pep-0495/#the-fold-attribute
+    """
+    cdef:
+        bint fold = 0
+        int64_t fold_delta
+
+    if pos > 0:
+        fold_delta = deltas[pos - 1] - deltas[pos]
+        if value - fold_delta < trans[pos]:
+            fold = 1
+
+    return fold
diff --git a/pandas/pandas/_libs/tslibs/util.pxd b/pandas/pandas/_libs/tslibs/util.pxd
new file mode 100644
index 0000000000000..f144275e0ee6a
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/util.pxd
@@ -0,0 +1,164 @@
+
+from cpython.object cimport PyTypeObject
+
+
+cdef extern from "Python.h":
+    # Note: importing extern-style allows us to declare these as nogil
+    # functions, whereas `from cpython cimport` does not.
+    bint PyBool_Check(object obj) nogil
+    bint PyFloat_Check(object obj) nogil
+    bint PyComplex_Check(object obj) nogil
+    bint PyObject_TypeCheck(object obj, PyTypeObject* type) nogil
+
+    # Note that following functions can potentially raise an exception,
+    # thus they cannot be declared 'nogil'.
+    object PyUnicode_EncodeLocale(object obj, const char *errors) nogil
+    object PyUnicode_DecodeLocale(const char *str, const char *errors) nogil
+
+
+cimport numpy as cnp
+from numpy cimport (
+    PyArray_Check,
+    float64_t,
+    int64_t,
+    is_timedelta64_object,
+)
+
+
+cdef extern from "numpy/arrayobject.h":
+    PyTypeObject PyFloatingArrType_Type
+
+cdef extern from "numpy/ndarrayobject.h":
+    PyTypeObject PyComplexFloatingArrType_Type
+    PyTypeObject PyBoolArrType_Type
+
+    bint PyArray_IsIntegerScalar(obj) nogil
+
+cdef extern from "numpy/npy_common.h":
+    int64_t NPY_MIN_INT64
+
+
+cdef inline int64_t get_nat() noexcept:
+    return NPY_MIN_INT64
+
+
+# --------------------------------------------------------------------
+# Type Checking
+
+cdef inline bint is_integer_object(object obj) noexcept:
+    """
+    Cython equivalent of
+
+    `isinstance(val, (int, np.integer)) and not isinstance(val, (bool, np.timedelta64))`
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_integer : bool
+
+    Notes
+    -----
+    This counts np.timedelta64 objects as integers.
+    """
+    return (not PyBool_Check(obj) and isinstance(obj, (int, cnp.integer))
+            and not is_timedelta64_object(obj))
+
+
+cdef inline bint is_float_object(object obj) noexcept nogil:
+    """
+    Cython equivalent of `isinstance(val, (float, np.floating))`
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_float : bool
+    """
+    return (PyFloat_Check(obj) or
+            (PyObject_TypeCheck(obj, &PyFloatingArrType_Type)))
+
+
+cdef inline bint is_complex_object(object obj) noexcept nogil:
+    """
+    Cython equivalent of `isinstance(val, (complex, np.complexfloating))`
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_complex : bool
+    """
+    return (PyComplex_Check(obj) or
+            PyObject_TypeCheck(obj, &PyComplexFloatingArrType_Type))
+
+
+cdef inline bint is_bool_object(object obj) noexcept nogil:
+    """
+    Cython equivalent of `isinstance(val, (bool, np.bool_))`
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_bool : bool
+    """
+    return (PyBool_Check(obj) or
+            PyObject_TypeCheck(obj, &PyBoolArrType_Type))
+
+
+cdef inline bint is_real_number_object(object obj) noexcept:
+    return is_bool_object(obj) or is_integer_object(obj) or is_float_object(obj)
+
+
+cdef inline bint is_array(object val) noexcept:
+    """
+    Cython equivalent of `isinstance(val, np.ndarray)`
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_ndarray : bool
+    """
+    return PyArray_Check(val)
+
+
+cdef inline bint is_nan(object val):
+    """
+    Check if val is a Not-A-Number float or complex, including
+    float('NaN') and np.nan.
+
+    Parameters
+    ----------
+    val : object
+
+    Returns
+    -------
+    is_nan : bool
+    """
+    cdef float64_t fval
+    if is_float_object(val):
+        fval = val
+        return fval != fval
+    return is_complex_object(val) and val != val
+
+
+cdef inline bytes string_encode_locale(str py_string):
+    """As opposed to PyUnicode_Encode, use current system locale to encode."""
+    return PyUnicode_EncodeLocale(py_string, NULL)
+
+
+cdef inline object char_to_string_locale(const char* data):
+    """As opposed to PyUnicode_FromString, use current system locale to decode."""
+    return PyUnicode_DecodeLocale(data, NULL)
diff --git a/pandas/pandas/_libs/tslibs/vectorized.pyi b/pandas/pandas/_libs/tslibs/vectorized.pyi
new file mode 100644
index 0000000000000..f377c2e26ab81
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/vectorized.pyi
@@ -0,0 +1,41 @@
+# For cython types that cannot be represented precisely, closest-available
+# python equivalents are used, and the precise types kept as adjacent comments.
+from datetime import tzinfo
+
+import numpy as np
+
+from pandas._libs.tslibs.dtypes import Resolution
+from pandas._typing import npt
+
+def dt64arr_to_periodarr(
+    stamps: npt.NDArray[np.int64],
+    freq: int,
+    tz: tzinfo | None,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int64]: ...
+def is_date_array_normalized(
+    stamps: npt.NDArray[np.int64],
+    tz: tzinfo | None,
+    reso: int,  # NPY_DATETIMEUNIT
+) -> bool: ...
+def normalize_i8_timestamps(
+    stamps: npt.NDArray[np.int64],
+    tz: tzinfo | None,
+    reso: int,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int64]: ...
+def get_resolution(
+    stamps: npt.NDArray[np.int64],
+    tz: tzinfo | None = ...,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> Resolution: ...
+def ints_to_pydatetime(
+    stamps: npt.NDArray[np.int64],
+    tz: tzinfo | None = ...,
+    box: str = ...,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.object_]: ...
+def tz_convert_from_utc(
+    stamps: npt.NDArray[np.int64],
+    tz: tzinfo | None,
+    reso: int = ...,  # NPY_DATETIMEUNIT
+) -> npt.NDArray[np.int64]: ...
diff --git a/pandas/pandas/_libs/tslibs/vectorized.pyx b/pandas/pandas/_libs/tslibs/vectorized.pyx
new file mode 100644
index 0000000000000..1e09874639d4f
--- /dev/null
+++ b/pandas/pandas/_libs/tslibs/vectorized.pyx
@@ -0,0 +1,381 @@
+cimport cython
+cimport numpy as cnp
+from cpython.datetime cimport (
+    date,
+    datetime,
+    time,
+    tzinfo,
+)
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+from .dtypes import Resolution
+
+from .dtypes cimport (
+    c_Resolution,
+    periods_per_day,
+)
+from .nattype cimport (
+    NPY_NAT,
+    c_NaT as NaT,
+)
+from .np_datetime cimport (
+    NPY_DATETIMEUNIT,
+    NPY_FR_ns,
+    import_pandas_datetime,
+    npy_datetimestruct,
+    pandas_datetime_to_datetimestruct,
+)
+
+import_pandas_datetime()
+
+from .period cimport get_period_ordinal
+from .timestamps cimport create_timestamp_from_ts
+from .timezones cimport is_utc
+from .tzconversion cimport Localizer
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def tz_convert_from_utc(ndarray stamps, tzinfo tz, NPY_DATETIMEUNIT reso=NPY_FR_ns):
+    # stamps is int64_t, arbitrary ndim
+    """
+    Convert the values (in i8) from UTC to tz
+
+    Parameters
+    ----------
+    stamps : ndarray[int64]
+    tz : tzinfo
+
+    Returns
+    -------
+    ndarray[int64]
+    """
+    if tz is None or is_utc(tz) or stamps.size == 0:
+        # Much faster than going through the "standard" pattern below;
+        #  do this before initializing Localizer.
+        return stamps.copy()
+
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        int64_t utc_val, local_val
+        Py_ssize_t pos, i, n = stamps.size
+
+        ndarray result
+        cnp.broadcast mi
+
+    result = cnp.PyArray_EMPTY(stamps.ndim, stamps.shape, cnp.NPY_INT64, 0)
+    mi = cnp.PyArray_MultiIterNew2(result, stamps)
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if utc_val == NPY_NAT:
+            local_val = NPY_NAT
+        else:
+            local_val = info.utc_val_to_local_val(utc_val, &pos)
+
+        # Analogous to: result[i] = local_val
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = local_val
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return result
+
+
+# -------------------------------------------------------------------------
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def ints_to_pydatetime(
+    ndarray stamps,
+    tzinfo tz=None,
+    str box="datetime",
+    NPY_DATETIMEUNIT reso=NPY_FR_ns,
+) -> ndarray:
+    # stamps is int64, arbitrary ndim
+    """
+    Convert an i8 repr to an ndarray of datetimes, date, time or Timestamp.
+
+    Parameters
+    ----------
+    stamps : array of i8
+    tz : str, optional
+         convert to this timezone
+    box : {'datetime', 'timestamp', 'date', 'time'}, default 'datetime'
+        * If datetime, convert to datetime.datetime
+        * If date, convert to datetime.date
+        * If time, convert to datetime.time
+        * If Timestamp, convert to pandas.Timestamp
+
+    reso : NPY_DATETIMEUNIT, default NPY_FR_ns
+
+    Returns
+    -------
+    ndarray[object] of type specified by box
+    """
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        int64_t utc_val, local_val
+        Py_ssize_t i, n = stamps.size
+        Py_ssize_t pos = -1  # unused, avoid not-initialized warning
+
+        npy_datetimestruct dts
+        tzinfo new_tz
+        bint use_date = False, use_ts = False, use_pydt = False
+        object res_val
+        bint fold = 0
+
+        # Note that `result` (and thus `result_flat`) is C-order and
+        #  `it` iterates C-order as well, so the iteration matches
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        ndarray result = cnp.PyArray_EMPTY(stamps.ndim, stamps.shape, cnp.NPY_OBJECT, 0)
+        object[::1] res_flat = result.ravel()     # should NOT be a copy
+        cnp.flatiter it = cnp.PyArray_IterNew(stamps)
+
+    if box == "date":
+        assert (tz is None), "tz should be None when converting to date"
+        use_date = True
+    elif box == "timestamp":
+        use_ts = True
+    elif box == "datetime":
+        use_pydt = True
+    elif box != "time":
+        raise ValueError(
+            "box must be one of 'datetime', 'date', 'time' or 'timestamp'"
+        )
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_ITER_DATA(it))[0]
+
+        new_tz = tz
+
+        if utc_val == NPY_NAT:
+            res_val = <object>NaT
+
+        else:
+
+            local_val = info.utc_val_to_local_val(utc_val, &pos, &fold)
+            if info.use_pytz:
+                # find right representation of dst etc in pytz timezone
+                new_tz = tz._tzinfos[tz._transition_info[pos]]
+
+            pandas_datetime_to_datetimestruct(local_val, reso, &dts)
+
+            if use_ts:
+                res_val = create_timestamp_from_ts(
+                    utc_val, dts, new_tz, fold, reso=reso
+                )
+            elif use_pydt:
+                res_val = datetime(
+                    dts.year, dts.month, dts.day, dts.hour, dts.min, dts.sec, dts.us,
+                    new_tz, fold=fold,
+                )
+            elif use_date:
+                res_val = date(dts.year, dts.month, dts.day)
+            else:
+                res_val = time(dts.hour, dts.min, dts.sec, dts.us, new_tz, fold=fold)
+
+        # Note: we can index result directly instead of using PyArray_MultiIter_DATA
+        #  like we do for the other functions because result is known C-contiguous
+        #  and is the first argument to PyArray_MultiIterNew2.  The usual pattern
+        #  does not seem to work with object dtype.
+        #  See discussion at
+        #  github.com/pandas-dev/pandas/pull/46886#discussion_r860261305
+        res_flat[i] = res_val
+
+        cnp.PyArray_ITER_NEXT(it)
+
+    return result
+
+
+# -------------------------------------------------------------------------
+
+
+cdef c_Resolution _reso_stamp(npy_datetimestruct *dts):
+    if dts.ps != 0:
+        return c_Resolution.RESO_NS
+    elif dts.us != 0:
+        if dts.us % 1000 == 0:
+            return c_Resolution.RESO_MS
+        return c_Resolution.RESO_US
+    elif dts.sec != 0:
+        return c_Resolution.RESO_SEC
+    elif dts.min != 0:
+        return c_Resolution.RESO_MIN
+    elif dts.hour != 0:
+        return c_Resolution.RESO_HR
+    return c_Resolution.RESO_DAY
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def get_resolution(
+    ndarray stamps, tzinfo tz=None, NPY_DATETIMEUNIT reso=NPY_FR_ns
+) -> Resolution:
+    # stamps is int64_t, any ndim
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        int64_t utc_val, local_val
+        Py_ssize_t i, n = stamps.size
+        Py_ssize_t pos = -1  # unused, avoid not-initialized warning
+        cnp.flatiter it = cnp.PyArray_IterNew(stamps)
+
+        npy_datetimestruct dts
+        c_Resolution pd_reso = c_Resolution.RESO_DAY, curr_reso
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_ITER_DATA(it))[0]
+
+        if utc_val == NPY_NAT:
+            pass
+        else:
+            local_val = info.utc_val_to_local_val(utc_val, &pos)
+
+            pandas_datetime_to_datetimestruct(local_val, reso, &dts)
+            curr_reso = _reso_stamp(&dts)
+            if curr_reso < pd_reso:
+                pd_reso = curr_reso
+
+        cnp.PyArray_ITER_NEXT(it)
+
+    return Resolution(pd_reso)
+
+
+# -------------------------------------------------------------------------
+
+
+@cython.cdivision(False)
+@cython.wraparound(False)
+@cython.boundscheck(False)
+cpdef ndarray normalize_i8_timestamps(ndarray stamps, tzinfo tz, NPY_DATETIMEUNIT reso):
+    # stamps is int64_t, arbitrary ndim
+    """
+    Normalize each of the (nanosecond) timezone aware timestamps in the given
+    array by rounding down to the beginning of the day (i.e. midnight).
+    This is midnight for timezone, `tz`.
+
+    Parameters
+    ----------
+    stamps : int64 ndarray
+    tz : tzinfo or None
+    reso : NPY_DATETIMEUNIT
+
+    Returns
+    -------
+    result : int64 ndarray of converted of normalized nanosecond timestamps
+    """
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        int64_t utc_val, local_val, res_val
+        Py_ssize_t i, n = stamps.size
+        Py_ssize_t pos = -1  # unused, avoid not-initialized warning
+
+        ndarray result = cnp.PyArray_EMPTY(stamps.ndim, stamps.shape, cnp.NPY_INT64, 0)
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(result, stamps)
+        int64_t ppd = periods_per_day(reso)
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if utc_val == NPY_NAT:
+            res_val = NPY_NAT
+        else:
+            local_val = info.utc_val_to_local_val(utc_val, &pos)
+            res_val = local_val - (local_val % ppd)
+
+        # Analogous to: result[i] = res_val
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return result
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_date_array_normalized(ndarray stamps, tzinfo tz, NPY_DATETIMEUNIT reso) -> bool:
+    # stamps is int64_t, arbitrary ndim
+    """
+    Check if all of the given (nanosecond) timestamps are normalized to
+    midnight, i.e. hour == minute == second == 0.  If the optional timezone
+    `tz` is not None, then this is midnight for this timezone.
+
+    Parameters
+    ----------
+    stamps : int64 ndarray
+    tz : tzinfo or None
+    reso : NPY_DATETIMEUNIT
+
+    Returns
+    -------
+    is_normalized : bool True if all stamps are normalized
+    """
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        int64_t utc_val, local_val
+        Py_ssize_t i, n = stamps.size
+        Py_ssize_t pos = -1  # unused, avoid not-initialized warning
+        cnp.flatiter it = cnp.PyArray_IterNew(stamps)
+        int64_t ppd = periods_per_day(reso)
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_ITER_DATA(it))[0]
+
+        local_val = info.utc_val_to_local_val(utc_val, &pos)
+
+        if local_val % ppd != 0:
+            return False
+
+        cnp.PyArray_ITER_NEXT(it)
+
+    return True
+
+
+# -------------------------------------------------------------------------
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def dt64arr_to_periodarr(
+    ndarray stamps, int freq, tzinfo tz, NPY_DATETIMEUNIT reso=NPY_FR_ns
+):
+    # stamps is int64_t, arbitrary ndim
+    cdef:
+        Localizer info = Localizer(tz, creso=reso)
+        Py_ssize_t i, n = stamps.size
+        Py_ssize_t pos = -1  # unused, avoid not-initialized warning
+        int64_t utc_val, local_val, res_val
+
+        npy_datetimestruct dts
+        ndarray result = cnp.PyArray_EMPTY(stamps.ndim, stamps.shape, cnp.NPY_INT64, 0)
+        cnp.broadcast mi = cnp.PyArray_MultiIterNew2(result, stamps)
+
+    for i in range(n):
+        # Analogous to: utc_val = stamps[i]
+        utc_val = (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
+        if utc_val == NPY_NAT:
+            res_val = NPY_NAT
+        else:
+            local_val = info.utc_val_to_local_val(utc_val, &pos)
+            pandas_datetime_to_datetimestruct(local_val, reso, &dts)
+            res_val = get_period_ordinal(&dts, freq)
+
+        # Analogous to: result[i] = res_val
+        (<int64_t*>cnp.PyArray_MultiIter_DATA(mi, 0))[0] = res_val
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return result
diff --git a/pandas/pandas/_libs/util.pxd b/pandas/pandas/_libs/util.pxd
new file mode 100644
index 0000000000000..18009a1a25de4
--- /dev/null
+++ b/pandas/pandas/_libs/util.pxd
@@ -0,0 +1,17 @@
+cimport numpy as cnp
+from libc.stdint cimport (
+    INT8_MAX,
+    INT8_MIN,
+    INT16_MAX,
+    INT16_MIN,
+    INT32_MAX,
+    INT32_MIN,
+    INT64_MAX,
+    INT64_MIN,
+    UINT8_MAX,
+    UINT16_MAX,
+    UINT32_MAX,
+    UINT64_MAX,
+)
+
+from pandas._libs.tslibs.util cimport *
diff --git a/pandas/pandas/_libs/window/__init__.py b/pandas/pandas/_libs/window/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/_libs/window/aggregations.pyi b/pandas/pandas/_libs/window/aggregations.pyi
new file mode 100644
index 0000000000000..ee735761e3dc6
--- /dev/null
+++ b/pandas/pandas/_libs/window/aggregations.pyi
@@ -0,0 +1,139 @@
+from typing import (
+    Any,
+    Callable,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._typing import (
+    WindowingRankType,
+    npt,
+)
+
+def roll_sum(
+    values: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_mean(
+    values: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_var(
+    values: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+    ddof: int = ...,
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_skew(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_kurt(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_median_c(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_max(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_min(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_first(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_last(
+    values: np.ndarray,  # np.ndarray[np.float64]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_quantile(
+    values: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+    quantile: float,  # float64_t
+    interpolation: Literal["linear", "lower", "higher", "nearest", "midpoint"],
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_rank(
+    values: np.ndarray,
+    start: np.ndarray,
+    end: np.ndarray,
+    minp: int,
+    percentile: bool,
+    method: WindowingRankType,
+    ascending: bool,
+) -> np.ndarray: ...  # np.ndarray[float]
+def roll_apply(
+    obj: object,
+    start: np.ndarray,  # np.ndarray[np.int64]
+    end: np.ndarray,  # np.ndarray[np.int64]
+    minp: int,  # int64_t
+    function: Callable[..., Any],
+    raw: bool,
+    args: tuple[Any, ...],
+    kwargs: dict[str, Any],
+) -> npt.NDArray[np.float64]: ...
+def roll_weighted_sum(
+    values: np.ndarray,  # const float64_t[:]
+    weights: np.ndarray,  # const float64_t[:]
+    minp: int,
+) -> np.ndarray: ...  # np.ndarray[np.float64]
+def roll_weighted_mean(
+    values: np.ndarray,  # const float64_t[:]
+    weights: np.ndarray,  # const float64_t[:]
+    minp: int,
+) -> np.ndarray: ...  # np.ndarray[np.float64]
+def roll_weighted_var(
+    values: np.ndarray,  # const float64_t[:]
+    weights: np.ndarray,  # const float64_t[:]
+    minp: int,  # int64_t
+    ddof: int,  # unsigned int
+) -> np.ndarray: ...  # np.ndarray[np.float64]
+def ewm(
+    vals: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # const int64_t[:]
+    end: np.ndarray,  # const int64_t[:]
+    minp: int,
+    com: float,  # float64_t
+    adjust: bool,
+    ignore_na: bool,
+    deltas: np.ndarray | None = None,  # const float64_t[:]
+    normalize: bool = True,
+) -> np.ndarray: ...  # np.ndarray[np.float64]
+def ewmcov(
+    input_x: np.ndarray,  # const float64_t[:]
+    start: np.ndarray,  # const int64_t[:]
+    end: np.ndarray,  # const int64_t[:]
+    minp: int,
+    input_y: np.ndarray,  # const float64_t[:]
+    com: float,  # float64_t
+    adjust: bool,
+    ignore_na: bool,
+    bias: bool,
+) -> np.ndarray: ...  # np.ndarray[np.float64]
diff --git a/pandas/pandas/_libs/window/aggregations.pyx b/pandas/pandas/_libs/window/aggregations.pyx
new file mode 100644
index 0000000000000..d33c840371d2a
--- /dev/null
+++ b/pandas/pandas/_libs/window/aggregations.pyx
@@ -0,0 +1,2038 @@
+# cython: boundscheck=False, wraparound=False, cdivision=True
+
+from libc.math cimport (
+    round,
+    signbit,
+    sqrt,
+)
+from libcpp.deque cimport deque
+
+from pandas._libs.algos cimport TiebreakEnumType
+
+import numpy as np
+
+cimport numpy as cnp
+from numpy cimport (
+    float32_t,
+    float64_t,
+    int64_t,
+    ndarray,
+)
+
+cnp.import_array()
+
+import cython
+
+from pandas._libs.algos import is_monotonic
+
+
+cdef extern from "pandas/skiplist.h":
+    ctypedef struct node_t:
+        node_t **next
+        int *width
+        double value
+        int is_nil
+        int levels
+        int ref_count
+
+    ctypedef struct skiplist_t:
+        node_t *head
+        node_t **tmp_chain
+        int *tmp_steps
+        int size
+        int maxlevels
+
+    skiplist_t* skiplist_init(int) nogil
+    void skiplist_destroy(skiplist_t*) nogil
+    double skiplist_get(skiplist_t*, int, int*) nogil
+    int skiplist_insert(skiplist_t*, double) nogil
+    int skiplist_remove(skiplist_t*, double) nogil
+    int skiplist_rank(skiplist_t*, double) nogil
+    int skiplist_min_rank(skiplist_t*, double) nogil
+
+cdef:
+    float32_t MINfloat32 = -np.inf
+    float64_t MINfloat64 = -np.inf
+
+    float32_t MAXfloat32 = np.inf
+    float64_t MAXfloat64 = np.inf
+
+    float64_t NaN = <float64_t>np.nan
+
+cdef bint is_monotonic_increasing_start_end_bounds(
+    ndarray[int64_t, ndim=1] start, ndarray[int64_t, ndim=1] end
+):
+    return is_monotonic(start, False)[0] and is_monotonic(end, False)[0]
+
+# ----------------------------------------------------------------------
+# Rolling sum
+
+
+cdef float64_t calc_sum(int64_t minp, int64_t nobs, float64_t sum_x,
+                        int64_t num_consecutive_same_value, float64_t prev_value
+                        ) noexcept nogil:
+    cdef:
+        float64_t result
+
+    if nobs == 0 == minp:
+        result = 0
+    elif nobs >= minp:
+        if num_consecutive_same_value >= nobs:
+            result = prev_value * nobs
+        else:
+            result = sum_x
+    else:
+        result = NaN
+
+    return result
+
+
+cdef void add_sum(float64_t val, int64_t *nobs, float64_t *sum_x,
+                  float64_t *compensation, int64_t *num_consecutive_same_value,
+                  float64_t *prev_value) noexcept nogil:
+    """ add a value from the sum calc using Kahan summation """
+
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] + 1
+        y = val - compensation[0]
+        t = sum_x[0] + y
+        compensation[0] = t - sum_x[0] - y
+        sum_x[0] = t
+
+        # GH#42064, record num of same values to remove floating point artifacts
+        if val == prev_value[0]:
+            num_consecutive_same_value[0] += 1
+        else:
+            # reset to 1 (include current value itself)
+            num_consecutive_same_value[0] = 1
+        prev_value[0] = val
+
+
+cdef void remove_sum(float64_t val, int64_t *nobs, float64_t *sum_x,
+                     float64_t *compensation) noexcept nogil:
+    """ remove a value from the sum calc using Kahan summation """
+
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] - 1
+        y = - val - compensation[0]
+        t = sum_x[0] + y
+        compensation[0] = t - sum_x[0] - y
+        sum_x[0] = t
+
+
+def roll_sum(const float64_t[:] values, ndarray[int64_t] start,
+             ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, j
+        float64_t sum_x, compensation_add, compensation_remove, prev_value
+        int64_t s, e, num_consecutive_same_value
+        int64_t nobs = 0, N = len(start)
+        ndarray[float64_t] output
+        bint is_monotonic_increasing_bounds
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    output = np.empty(N, dtype=np.float64)
+
+    with nogil:
+
+        for i in range(0, N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                # setup
+                prev_value = values[s]
+                num_consecutive_same_value = 0
+                sum_x = compensation_add = compensation_remove = 0
+                nobs = 0
+                for j in range(s, e):
+                    add_sum(values[j], &nobs, &sum_x, &compensation_add,
+                            &num_consecutive_same_value, &prev_value)
+
+            else:
+
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    remove_sum(values[j], &nobs, &sum_x, &compensation_remove)
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    add_sum(values[j], &nobs, &sum_x, &compensation_add,
+                            &num_consecutive_same_value, &prev_value)
+
+            output[i] = calc_sum(
+                minp, nobs, sum_x, num_consecutive_same_value, prev_value
+            )
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+                sum_x = 0.0
+                compensation_remove = 0.0
+
+    return output
+
+
+# ----------------------------------------------------------------------
+# Rolling mean
+
+
+cdef float64_t calc_mean(int64_t minp, Py_ssize_t nobs, Py_ssize_t neg_ct,
+                         float64_t sum_x, int64_t num_consecutive_same_value,
+                         float64_t prev_value) noexcept nogil:
+    cdef:
+        float64_t result
+
+    if nobs >= minp and nobs > 0:
+        result = sum_x / <float64_t>nobs
+        if num_consecutive_same_value >= nobs:
+            result = prev_value
+        elif neg_ct == 0 and result < 0:
+            # all positive
+            result = 0
+        elif neg_ct == nobs and result > 0:
+            # all negative
+            result = 0
+        else:
+            pass
+    else:
+        result = NaN
+    return result
+
+
+cdef void add_mean(
+    float64_t val,
+    Py_ssize_t *nobs,
+    float64_t *sum_x,
+    Py_ssize_t *neg_ct,
+    float64_t *compensation,
+    int64_t *num_consecutive_same_value,
+    float64_t *prev_value
+) noexcept nogil:
+    """ add a value from the mean calc using Kahan summation """
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] + 1
+        y = val - compensation[0]
+        t = sum_x[0] + y
+        compensation[0] = t - sum_x[0] - y
+        sum_x[0] = t
+        if signbit(val):
+            neg_ct[0] = neg_ct[0] + 1
+
+        # GH#42064, record num of same values to remove floating point artifacts
+        if val == prev_value[0]:
+            num_consecutive_same_value[0] += 1
+        else:
+            # reset to 1 (include current value itself)
+            num_consecutive_same_value[0] = 1
+        prev_value[0] = val
+
+
+cdef void remove_mean(float64_t val, Py_ssize_t *nobs, float64_t *sum_x,
+                      Py_ssize_t *neg_ct, float64_t *compensation) noexcept nogil:
+    """ remove a value from the mean calc using Kahan summation """
+    cdef:
+        float64_t y, t
+
+    if val == val:
+        nobs[0] = nobs[0] - 1
+        y = - val - compensation[0]
+        t = sum_x[0] + y
+        compensation[0] = t - sum_x[0] - y
+        sum_x[0] = t
+        if signbit(val):
+            neg_ct[0] = neg_ct[0] - 1
+
+
+def roll_mean(const float64_t[:] values, ndarray[int64_t] start,
+              ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    cdef:
+        float64_t val, compensation_add, compensation_remove, sum_x, prev_value
+        int64_t s, e, num_consecutive_same_value
+        Py_ssize_t nobs, i, j, neg_ct, N = len(start)
+        ndarray[float64_t] output
+        bint is_monotonic_increasing_bounds
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    output = np.empty(N, dtype=np.float64)
+
+    with nogil:
+
+        for i in range(0, N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                # setup
+                compensation_add = compensation_remove = sum_x = 0
+                nobs = neg_ct = 0
+                prev_value = values[s]
+                num_consecutive_same_value = 0
+                for j in range(s, e):
+                    val = values[j]
+                    add_mean(val, &nobs, &sum_x, &neg_ct, &compensation_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            else:
+
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    val = values[j]
+                    remove_mean(val, &nobs, &sum_x, &neg_ct, &compensation_remove)
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    val = values[j]
+                    add_mean(val, &nobs, &sum_x, &neg_ct, &compensation_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            output[i] = calc_mean(
+                minp, nobs, neg_ct, sum_x, num_consecutive_same_value, prev_value
+            )
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+                neg_ct = 0
+                sum_x = 0.0
+                compensation_remove = 0.0
+    return output
+
+# ----------------------------------------------------------------------
+# Rolling variance
+
+
+cdef float64_t calc_var(
+    int64_t minp,
+    int ddof,
+    float64_t nobs,
+    float64_t ssqdm_x,
+    int64_t num_consecutive_same_value
+) noexcept nogil:
+    cdef:
+        float64_t result
+
+    # Variance is unchanged if no observation is added or removed
+    if (nobs >= minp) and (nobs > ddof):
+
+        # pathological case & repeatedly same values case
+        if nobs == 1 or num_consecutive_same_value >= nobs:
+            result = 0
+        else:
+            result = ssqdm_x / (nobs - <float64_t>ddof)
+    else:
+        result = NaN
+
+    return result
+
+
+cdef void add_var(
+    float64_t val,
+    float64_t *nobs,
+    float64_t *mean_x,
+    float64_t *ssqdm_x,
+    float64_t *compensation,
+    int64_t *num_consecutive_same_value,
+    float64_t *prev_value,
+) noexcept nogil:
+    """ add a value from the var calc """
+    cdef:
+        float64_t delta, prev_mean, y, t
+
+    # GH#21813, if msvc 2017 bug is resolved, we should be OK with != instead of `isnan`
+    if val != val:
+        return
+
+    nobs[0] = nobs[0] + 1
+
+    # GH#42064, record num of same values to remove floating point artifacts
+    if val == prev_value[0]:
+        num_consecutive_same_value[0] += 1
+    else:
+        # reset to 1 (include current value itself)
+        num_consecutive_same_value[0] = 1
+    prev_value[0] = val
+
+    # Welford's method for the online variance-calculation
+    # using Kahan summation
+    # https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+    prev_mean = mean_x[0] - compensation[0]
+    y = val - compensation[0]
+    t = y - mean_x[0]
+    compensation[0] = t + mean_x[0] - y
+    delta = t
+    if nobs[0]:
+        mean_x[0] = mean_x[0] + delta / nobs[0]
+    else:
+        mean_x[0] = 0
+    ssqdm_x[0] = ssqdm_x[0] + (val - prev_mean) * (val - mean_x[0])
+
+
+cdef void remove_var(
+    float64_t val,
+    float64_t *nobs,
+    float64_t *mean_x,
+    float64_t *ssqdm_x,
+    float64_t *compensation
+) noexcept nogil:
+    """ remove a value from the var calc """
+    cdef:
+        float64_t delta, prev_mean, y, t
+    if val == val:
+        nobs[0] = nobs[0] - 1
+        if nobs[0]:
+            # Welford's method for the online variance-calculation
+            # using Kahan summation
+            # https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+            prev_mean = mean_x[0] - compensation[0]
+            y = val - compensation[0]
+            t = y - mean_x[0]
+            compensation[0] = t + mean_x[0] - y
+            delta = t
+            mean_x[0] = mean_x[0] - delta / nobs[0]
+            ssqdm_x[0] = ssqdm_x[0] - (val - prev_mean) * (val - mean_x[0])
+        else:
+            mean_x[0] = 0
+            ssqdm_x[0] = 0
+
+
+def roll_var(const float64_t[:] values, ndarray[int64_t] start,
+             ndarray[int64_t] end, int64_t minp, int ddof=1) -> np.ndarray:
+    """
+    Numerically stable implementation using Welford's method.
+    """
+    cdef:
+        float64_t mean_x, ssqdm_x, nobs, compensation_add,
+        float64_t compensation_remove, prev_value
+        int64_t s, e, num_consecutive_same_value
+        Py_ssize_t i, j, N = len(start)
+        ndarray[float64_t] output
+        bint is_monotonic_increasing_bounds
+
+    minp = max(minp, 1)
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    output = np.empty(N, dtype=np.float64)
+
+    with nogil:
+
+        for i in range(0, N):
+
+            s = start[i]
+            e = end[i]
+
+            # Over the first window, observations can only be added
+            # never removed
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                prev_value = values[s]
+                num_consecutive_same_value = 0
+
+                mean_x = ssqdm_x = nobs = compensation_add = compensation_remove = 0
+                for j in range(s, e):
+                    add_var(values[j], &nobs, &mean_x, &ssqdm_x, &compensation_add,
+                            &num_consecutive_same_value, &prev_value)
+
+            else:
+
+                # After the first window, observations can both be added
+                # and removed
+
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    remove_var(values[j], &nobs, &mean_x, &ssqdm_x,
+                               &compensation_remove)
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    add_var(values[j], &nobs, &mean_x, &ssqdm_x, &compensation_add,
+                            &num_consecutive_same_value, &prev_value)
+
+            output[i] = calc_var(minp, ddof, nobs, ssqdm_x, num_consecutive_same_value)
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0.0
+                mean_x = 0.0
+                ssqdm_x = 0.0
+                compensation_remove = 0.0
+
+    return output
+
+# ----------------------------------------------------------------------
+# Rolling skewness
+
+
+cdef float64_t calc_skew(int64_t minp, int64_t nobs,
+                         float64_t x, float64_t xx, float64_t xxx,
+                         int64_t num_consecutive_same_value
+                         ) noexcept nogil:
+    cdef:
+        float64_t result, dnobs
+        float64_t A, B, C, R
+
+    if nobs >= minp:
+        dnobs = <float64_t>nobs
+        A = x / dnobs
+        B = xx / dnobs - A * A
+        C = xxx / dnobs - A * A * A - 3 * A * B
+
+        if nobs < 3:
+            result = NaN
+        # GH 42064 46431
+        # uniform case, force result to be 0
+        elif num_consecutive_same_value >= nobs:
+            result = 0.0
+        # #18044: with uniform distribution, floating issue will
+        #         cause B != 0. and cause the result is a very
+        #         large number.
+        #
+        #         in core/nanops.py nanskew/nankurt call the function
+        #         _zero_out_fperr(m2) to fix floating error.
+        #         if the variance is less than 1e-14, it could be
+        #         treat as zero, here we follow the original
+        #         skew/kurt behaviour to check B <= 1e-14
+        elif B <= 1e-14:
+            result = NaN
+        else:
+            R = sqrt(B)
+            result = ((sqrt(dnobs * (dnobs - 1.)) * C) /
+                      ((dnobs - 2) * R * R * R))
+    else:
+        result = NaN
+
+    return result
+
+
+cdef void add_skew(float64_t val, int64_t *nobs,
+                   float64_t *x, float64_t *xx,
+                   float64_t *xxx,
+                   float64_t *compensation_x,
+                   float64_t *compensation_xx,
+                   float64_t *compensation_xxx,
+                   int64_t *num_consecutive_same_value,
+                   float64_t *prev_value,
+                   ) noexcept nogil:
+    """ add a value from the skew calc """
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] + 1
+
+        y = val - compensation_x[0]
+        t = x[0] + y
+        compensation_x[0] = t - x[0] - y
+        x[0] = t
+        y = val * val - compensation_xx[0]
+        t = xx[0] + y
+        compensation_xx[0] = t - xx[0] - y
+        xx[0] = t
+        y = val * val * val - compensation_xxx[0]
+        t = xxx[0] + y
+        compensation_xxx[0] = t - xxx[0] - y
+        xxx[0] = t
+
+        # GH#42064, record num of same values to remove floating point artifacts
+        if val == prev_value[0]:
+            num_consecutive_same_value[0] += 1
+        else:
+            # reset to 1 (include current value itself)
+            num_consecutive_same_value[0] = 1
+        prev_value[0] = val
+
+
+cdef void remove_skew(float64_t val, int64_t *nobs,
+                      float64_t *x, float64_t *xx,
+                      float64_t *xxx,
+                      float64_t *compensation_x,
+                      float64_t *compensation_xx,
+                      float64_t *compensation_xxx) noexcept nogil:
+    """ remove a value from the skew calc """
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] - 1
+
+        y = - val - compensation_x[0]
+        t = x[0] + y
+        compensation_x[0] = t - x[0] - y
+        x[0] = t
+        y = - val * val - compensation_xx[0]
+        t = xx[0] + y
+        compensation_xx[0] = t - xx[0] - y
+        xx[0] = t
+        y = - val * val * val - compensation_xxx[0]
+        t = xxx[0] + y
+        compensation_xxx[0] = t - xxx[0] - y
+        xxx[0] = t
+
+
+def roll_skew(ndarray[float64_t] values, ndarray[int64_t] start,
+              ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, j
+        float64_t val, min_val, mean_val, sum_val = 0
+        float64_t compensation_xxx_add, compensation_xxx_remove
+        float64_t compensation_xx_add, compensation_xx_remove
+        float64_t compensation_x_add, compensation_x_remove
+        float64_t x, xx, xxx
+        float64_t prev_value
+        int64_t nobs = 0, N = len(start), V = len(values), nobs_mean = 0
+        int64_t s, e, num_consecutive_same_value
+        ndarray[float64_t] output, values_copy
+        bint is_monotonic_increasing_bounds
+
+    minp = max(minp, 3)
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    output = np.empty(N, dtype=np.float64)
+    min_val = np.nanmin(values)
+    values_copy = np.copy(values)
+
+    with nogil:
+        for i in range(0, V):
+            val = values_copy[i]
+            if val == val:
+                nobs_mean += 1
+                sum_val += val
+        mean_val = sum_val / nobs_mean
+        # Other cases would lead to imprecision for smallest values
+        if min_val - mean_val > -1e5:
+            mean_val = round(mean_val)
+            for i in range(0, V):
+                values_copy[i] = values_copy[i] - mean_val
+
+        for i in range(0, N):
+
+            s = start[i]
+            e = end[i]
+
+            # Over the first window, observations can only be added
+            # never removed
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                prev_value = values[s]
+                num_consecutive_same_value = 0
+
+                compensation_xxx_add = compensation_xxx_remove = 0
+                compensation_xx_add = compensation_xx_remove = 0
+                compensation_x_add = compensation_x_remove = 0
+                x = xx = xxx = 0
+                nobs = 0
+                for j in range(s, e):
+                    val = values_copy[j]
+                    add_skew(val, &nobs, &x, &xx, &xxx, &compensation_x_add,
+                             &compensation_xx_add, &compensation_xxx_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            else:
+
+                # After the first window, observations can both be added
+                # and removed
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    val = values_copy[j]
+                    remove_skew(val, &nobs, &x, &xx, &xxx, &compensation_x_remove,
+                                &compensation_xx_remove, &compensation_xxx_remove)
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    val = values_copy[j]
+                    add_skew(val, &nobs, &x, &xx, &xxx, &compensation_x_add,
+                             &compensation_xx_add, &compensation_xxx_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            output[i] = calc_skew(minp, nobs, x, xx, xxx, num_consecutive_same_value)
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+                x = 0.0
+                xx = 0.0
+                xxx = 0.0
+
+    return output
+
+# ----------------------------------------------------------------------
+# Rolling kurtosis
+
+
+cdef float64_t calc_kurt(int64_t minp, int64_t nobs,
+                         float64_t x, float64_t xx,
+                         float64_t xxx, float64_t xxxx,
+                         int64_t num_consecutive_same_value,
+                         ) noexcept nogil:
+    cdef:
+        float64_t result, dnobs
+        float64_t A, B, C, D, R, K
+
+    if nobs >= minp:
+        if nobs < 4:
+            result = NaN
+        # GH 42064 46431
+        # uniform case, force result to be -3.
+        elif num_consecutive_same_value >= nobs:
+            result = -3.
+        else:
+            dnobs = <float64_t>nobs
+            A = x / dnobs
+            R = A * A
+            B = xx / dnobs - R
+            R = R * A
+            C = xxx / dnobs - R - 3 * A * B
+            R = R * A
+            D = xxxx / dnobs - R - 6 * B * A * A - 4 * C * A
+
+            # #18044: with uniform distribution, floating issue will
+            #         cause B != 0. and cause the result is a very
+            #         large number.
+            #
+            #         in core/nanops.py nanskew/nankurt call the function
+            #         _zero_out_fperr(m2) to fix floating error.
+            #         if the variance is less than 1e-14, it could be
+            #         treat as zero, here we follow the original
+            #         skew/kurt behaviour to check B <= 1e-14
+            if B <= 1e-14:
+                result = NaN
+            else:
+                K = (dnobs * dnobs - 1.) * D / (B * B) - 3 * ((dnobs - 1.) ** 2)
+                result = K / ((dnobs - 2.) * (dnobs - 3.))
+    else:
+        result = NaN
+
+    return result
+
+
+cdef void add_kurt(float64_t val, int64_t *nobs,
+                   float64_t *x, float64_t *xx,
+                   float64_t *xxx, float64_t *xxxx,
+                   float64_t *compensation_x,
+                   float64_t *compensation_xx,
+                   float64_t *compensation_xxx,
+                   float64_t *compensation_xxxx,
+                   int64_t *num_consecutive_same_value,
+                   float64_t *prev_value
+                   ) noexcept nogil:
+    """ add a value from the kurotic calc """
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] + 1
+
+        y = val - compensation_x[0]
+        t = x[0] + y
+        compensation_x[0] = t - x[0] - y
+        x[0] = t
+        y = val * val - compensation_xx[0]
+        t = xx[0] + y
+        compensation_xx[0] = t - xx[0] - y
+        xx[0] = t
+        y = val * val * val - compensation_xxx[0]
+        t = xxx[0] + y
+        compensation_xxx[0] = t - xxx[0] - y
+        xxx[0] = t
+        y = val * val * val * val - compensation_xxxx[0]
+        t = xxxx[0] + y
+        compensation_xxxx[0] = t - xxxx[0] - y
+        xxxx[0] = t
+
+        # GH#42064, record num of same values to remove floating point artifacts
+        if val == prev_value[0]:
+            num_consecutive_same_value[0] += 1
+        else:
+            # reset to 1 (include current value itself)
+            num_consecutive_same_value[0] = 1
+        prev_value[0] = val
+
+
+cdef void remove_kurt(float64_t val, int64_t *nobs,
+                      float64_t *x, float64_t *xx,
+                      float64_t *xxx, float64_t *xxxx,
+                      float64_t *compensation_x,
+                      float64_t *compensation_xx,
+                      float64_t *compensation_xxx,
+                      float64_t *compensation_xxxx) noexcept nogil:
+    """ remove a value from the kurotic calc """
+    cdef:
+        float64_t y, t
+
+    # Not NaN
+    if val == val:
+        nobs[0] = nobs[0] - 1
+
+        y = - val - compensation_x[0]
+        t = x[0] + y
+        compensation_x[0] = t - x[0] - y
+        x[0] = t
+        y = - val * val - compensation_xx[0]
+        t = xx[0] + y
+        compensation_xx[0] = t - xx[0] - y
+        xx[0] = t
+        y = - val * val * val - compensation_xxx[0]
+        t = xxx[0] + y
+        compensation_xxx[0] = t - xxx[0] - y
+        xxx[0] = t
+        y = - val * val * val * val - compensation_xxxx[0]
+        t = xxxx[0] + y
+        compensation_xxxx[0] = t - xxxx[0] - y
+        xxxx[0] = t
+
+
+def roll_kurt(ndarray[float64_t] values, ndarray[int64_t] start,
+              ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, j
+        float64_t val, mean_val, min_val, sum_val = 0
+        float64_t compensation_xxxx_add, compensation_xxxx_remove
+        float64_t compensation_xxx_remove, compensation_xxx_add
+        float64_t compensation_xx_remove, compensation_xx_add
+        float64_t compensation_x_remove, compensation_x_add
+        float64_t x, xx, xxx, xxxx
+        float64_t prev_value
+        int64_t nobs, s, e, num_consecutive_same_value
+        int64_t N = len(start), V = len(values), nobs_mean = 0
+        ndarray[float64_t] output, values_copy
+        bint is_monotonic_increasing_bounds
+
+    minp = max(minp, 4)
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    output = np.empty(N, dtype=np.float64)
+    values_copy = np.copy(values)
+    min_val = np.nanmin(values)
+
+    with nogil:
+        for i in range(0, V):
+            val = values_copy[i]
+            if val == val:
+                nobs_mean += 1
+                sum_val += val
+        mean_val = sum_val / nobs_mean
+        # Other cases would lead to imprecision for smallest values
+        if min_val - mean_val > -1e4:
+            mean_val = round(mean_val)
+            for i in range(0, V):
+                values_copy[i] = values_copy[i] - mean_val
+
+        for i in range(0, N):
+
+            s = start[i]
+            e = end[i]
+
+            # Over the first window, observations can only be added
+            # never removed
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                prev_value = values[s]
+                num_consecutive_same_value = 0
+
+                compensation_xxxx_add = compensation_xxxx_remove = 0
+                compensation_xxx_remove = compensation_xxx_add = 0
+                compensation_xx_remove = compensation_xx_add = 0
+                compensation_x_remove = compensation_x_add = 0
+                x = xx = xxx = xxxx = 0
+                nobs = 0
+                for j in range(s, e):
+                    add_kurt(values_copy[j], &nobs, &x, &xx, &xxx, &xxxx,
+                             &compensation_x_add, &compensation_xx_add,
+                             &compensation_xxx_add, &compensation_xxxx_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            else:
+
+                # After the first window, observations can both be added
+                # and removed
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    remove_kurt(values_copy[j], &nobs, &x, &xx, &xxx, &xxxx,
+                                &compensation_x_remove, &compensation_xx_remove,
+                                &compensation_xxx_remove, &compensation_xxxx_remove)
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    add_kurt(values_copy[j], &nobs, &x, &xx, &xxx, &xxxx,
+                             &compensation_x_add, &compensation_xx_add,
+                             &compensation_xxx_add, &compensation_xxxx_add,
+                             &num_consecutive_same_value, &prev_value)
+
+            output[i] = calc_kurt(minp, nobs, x, xx, xxx, xxxx,
+                                  num_consecutive_same_value)
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+                x = 0.0
+                xx = 0.0
+                xxx = 0.0
+                xxxx = 0.0
+
+    return output
+
+
+# ----------------------------------------------------------------------
+# Rolling median, min, max
+
+
+def roll_median_c(const float64_t[:] values, ndarray[int64_t] start,
+                  ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    cdef:
+        Py_ssize_t i, j
+        bint err = False, is_monotonic_increasing_bounds
+        int midpoint, ret = 0
+        int64_t nobs = 0, N = len(start), s, e, win
+        float64_t val, res
+        skiplist_t *sl
+        ndarray[float64_t] output
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+
+    # we use the Fixed/Variable Indexer here as the
+    # actual skiplist ops outweigh any window computation costs
+    output = np.empty(N, dtype=np.float64)
+
+    if (end - start).max() == 0:
+        output[:] = NaN
+        return output
+    win = (end - start).max()
+    sl = skiplist_init(<int>win)
+    if sl == NULL:
+        raise MemoryError("skiplist_init failed")
+
+    with nogil:
+
+        for i in range(0, N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+
+                if i != 0:
+                    skiplist_destroy(sl)
+                    sl = skiplist_init(<int>win)
+                    nobs = 0
+                # setup
+                for j in range(s, e):
+                    val = values[j]
+                    if val == val:
+                        nobs += 1
+                        err = skiplist_insert(sl, val) == -1
+                        if err:
+                            break
+
+            else:
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    val = values[j]
+                    if val == val:
+                        nobs += 1
+                        err = skiplist_insert(sl, val) == -1
+                        if err:
+                            break
+
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    val = values[j]
+                    if val == val:
+                        skiplist_remove(sl, val)
+                        nobs -= 1
+            if nobs >= minp:
+                midpoint = <int>(nobs / 2)
+                if nobs % 2:
+                    res = skiplist_get(sl, midpoint, &ret)
+                else:
+                    res = (skiplist_get(sl, midpoint, &ret) +
+                           skiplist_get(sl, (midpoint - 1), &ret)) / 2
+                if ret == 0:
+                    res = NaN
+            else:
+                res = NaN
+
+            output[i] = res
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+                skiplist_destroy(sl)
+                sl = skiplist_init(<int>win)
+
+    skiplist_destroy(sl)
+    if err:
+        raise MemoryError("skiplist_insert failed")
+    return output
+
+
+# ----------------------------------------------------------------------
+
+# Moving maximum / minimum code taken from Bottleneck
+# Licence at LICENSES/BOTTLENECK_LICENCE
+
+
+cdef float64_t init_mm(float64_t ai, Py_ssize_t *nobs, bint is_max) noexcept nogil:
+
+    if ai == ai:
+        nobs[0] = nobs[0] + 1
+    elif is_max:
+        ai = MINfloat64
+    else:
+        ai = MAXfloat64
+
+    return ai
+
+
+cdef void remove_mm(float64_t aold, Py_ssize_t *nobs) noexcept nogil:
+    """ remove a value from the mm calc """
+    if aold == aold:
+        nobs[0] = nobs[0] - 1
+
+
+cdef float64_t calc_mm(int64_t minp, Py_ssize_t nobs,
+                       float64_t value) noexcept nogil:
+    cdef:
+        float64_t result
+
+    if nobs >= minp:
+        result = value
+    else:
+        result = NaN
+
+    return result
+
+
+def roll_max(ndarray[float64_t] values, ndarray[int64_t] start,
+             ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    """
+    Moving max of 1d array of any numeric type along axis=0 ignoring NaNs.
+
+    Parameters
+    ----------
+    values : np.ndarray[np.float64]
+    window : int, size of rolling window
+    minp : if number of observations in window
+          is below this, output a NaN
+    index : ndarray, optional
+       index for window computation
+    closed : 'right', 'left', 'both', 'neither'
+            make the interval closed on the right, left,
+            both or neither endpoints
+
+    Returns
+    -------
+    np.ndarray[float]
+    """
+    return _roll_min_max(values, start, end, minp, is_max=1)
+
+
+def roll_min(ndarray[float64_t] values, ndarray[int64_t] start,
+             ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    """
+    Moving min of 1d array of any numeric type along axis=0 ignoring NaNs.
+
+    Parameters
+    ----------
+    values : np.ndarray[np.float64]
+    window : int, size of rolling window
+    minp : if number of observations in window
+          is below this, output a NaN
+    index : ndarray, optional
+       index for window computation
+
+    Returns
+    -------
+    np.ndarray[float]
+    """
+    return _roll_min_max(values, start, end, minp, is_max=0)
+
+
+cdef _roll_min_max(ndarray[float64_t] values,
+                   ndarray[int64_t] starti,
+                   ndarray[int64_t] endi,
+                   int64_t minp,
+                   bint is_max):
+    cdef:
+        float64_t ai
+        int64_t curr_win_size, start
+        Py_ssize_t i, k, nobs = 0, N = len(starti)
+        deque Q[int64_t]  # min/max always the front
+        deque W[int64_t]  # track the whole window for nobs compute
+        ndarray[float64_t, ndim=1] output
+
+    output = np.empty(N, dtype=np.float64)
+    Q = deque[int64_t]()
+    W = deque[int64_t]()
+
+    with nogil:
+
+        # This is using a modified version of the C++ code in this
+        # SO post: https://stackoverflow.com/a/12239580
+        # The original impl didn't deal with variable window sizes
+        # So the code was optimized for that
+
+        # first window's size
+        curr_win_size = endi[0] - starti[0]
+        # GH 32865
+        # Anchor output index to values index to provide custom
+        # BaseIndexer support
+        for i in range(N):
+
+            curr_win_size = endi[i] - starti[i]
+            if i == 0:
+                start = starti[i]
+            else:
+                start = endi[i - 1]
+
+            for k in range(start, endi[i]):
+                ai = init_mm(values[k], &nobs, is_max)
+                # Discard previous entries if we find new min or max
+                if is_max:
+                    while not Q.empty() and ((ai >= values[Q.back()]) or
+                                             values[Q.back()] != values[Q.back()]):
+                        Q.pop_back()
+                else:
+                    while not Q.empty() and ((ai <= values[Q.back()]) or
+                                             values[Q.back()] != values[Q.back()]):
+                        Q.pop_back()
+                Q.push_back(k)
+                W.push_back(k)
+
+            # Discard entries outside and left of current window
+            while not Q.empty() and Q.front() <= starti[i] - 1:
+                Q.pop_front()
+            while not W.empty() and W.front() <= starti[i] - 1:
+                remove_mm(values[W.front()], &nobs)
+                W.pop_front()
+
+            # Save output based on index in input value array
+            if not Q.empty() and curr_win_size > 0:
+                output[i] = calc_mm(minp, nobs, values[Q.front()])
+            else:
+                output[i] = NaN
+
+    return output
+
+# ----------------------------------------------------------------------
+# Rolling first, last
+
+
+def roll_first(const float64_t[:] values, ndarray[int64_t] start,
+               ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    return _roll_first_last(values, start, end, minp, is_first=1)
+
+
+def roll_last(const float64_t[:] values, ndarray[int64_t] start,
+              ndarray[int64_t] end, int64_t minp) -> np.ndarray:
+    return _roll_first_last(values, start, end, minp, is_first=0)
+
+
+cdef _roll_first_last(const float64_t[:] values, ndarray[int64_t] start,
+                      ndarray[int64_t] end, int64_t minp, bint is_first):
+    cdef:
+        Py_ssize_t i, j, fl_idx
+        bint is_monotonic_increasing_bounds
+        int64_t nobs = 0, N = len(start), s, e
+        float64_t val, res
+        ndarray[float64_t] output
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+
+    output = np.empty(N, dtype=np.float64)
+
+    if (end - start).max() == 0:
+        output[:] = NaN
+        return output
+
+    with nogil:
+        for i in range(0, N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+                fl_idx = -1
+                nobs = 0
+                for j in range(s, e):
+                    val = values[j]
+                    if val == val:
+                        if not is_first or fl_idx < s:
+                            fl_idx = j
+                        nobs += 1
+            else:
+                # handle deletes
+                for j in range(start[i - 1], s):
+                    val = values[j]
+                    if val == val:
+                        nobs -= 1
+
+                # update fl_idx if out of range, if first
+                if is_first and fl_idx < s:
+                    fl_idx = -1
+                    for j in range(s, end[i - 1]):
+                        val = values[j]
+                        if val == val:
+                            fl_idx = j
+                            break
+
+                # handle adds
+                for j in range(end[i - 1], e):
+                    val = values[j]
+                    if val == val:
+                        if not is_first or fl_idx < s:
+                            fl_idx = j
+                        nobs += 1
+
+            if nobs >= minp and fl_idx >= s:
+                res = values[fl_idx]
+            else:
+                res = NaN
+
+            output[i] = res
+
+            if not is_monotonic_increasing_bounds:
+                nobs = 0
+
+    return output
+
+
+cdef enum InterpolationType:
+    LINEAR,
+    LOWER,
+    HIGHER,
+    NEAREST,
+    MIDPOINT
+
+
+interpolation_types = {
+    "linear": LINEAR,
+    "lower": LOWER,
+    "higher": HIGHER,
+    "nearest": NEAREST,
+    "midpoint": MIDPOINT,
+}
+
+
+def roll_quantile(const float64_t[:] values, ndarray[int64_t] start,
+                  ndarray[int64_t] end, int64_t minp,
+                  float64_t quantile, str interpolation) -> np.ndarray:
+    """
+    O(N log(window)) implementation using skip list
+    """
+    cdef:
+        Py_ssize_t i, j, s, e, N = len(start), idx
+        int ret = 0
+        int64_t nobs = 0, win
+        float64_t val, idx_with_fraction
+        float64_t vlow, vhigh
+        skiplist_t *skiplist
+        InterpolationType interpolation_type
+        ndarray[float64_t] output
+
+    if quantile <= 0.0 or quantile >= 1.0:
+        raise ValueError(f"quantile value {quantile} not in [0, 1]")
+
+    try:
+        interpolation_type = interpolation_types[interpolation]
+    except KeyError:
+        raise ValueError(f"Interpolation '{interpolation}' is not supported")
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    # we use the Fixed/Variable Indexer here as the
+    # actual skiplist ops outweigh any window computation costs
+    output = np.empty(N, dtype=np.float64)
+
+    win = (end - start).max()
+    if win == 0:
+        output[:] = NaN
+        return output
+    skiplist = skiplist_init(<int>win)
+    if skiplist == NULL:
+        raise MemoryError("skiplist_init failed")
+
+    with nogil:
+        for i in range(0, N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+                if i != 0:
+                    nobs = 0
+                    skiplist_destroy(skiplist)
+                    skiplist = skiplist_init(<int>win)
+
+                # setup
+                for j in range(s, e):
+                    val = values[j]
+                    if val == val:
+                        nobs += 1
+                        skiplist_insert(skiplist, val)
+
+            else:
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    val = values[j]
+                    if val == val:
+                        nobs += 1
+                        skiplist_insert(skiplist, val)
+
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    val = values[j]
+                    if val == val:
+                        skiplist_remove(skiplist, val)
+                        nobs -= 1
+            if nobs >= minp:
+                if nobs == 1:
+                    # Single value in skip list
+                    output[i] = skiplist_get(skiplist, 0, &ret)
+                else:
+                    idx_with_fraction = quantile * (nobs - 1)
+                    idx = <int>idx_with_fraction
+
+                    if idx_with_fraction == idx:
+                        # no need to interpolate
+                        output[i] = skiplist_get(skiplist, idx, &ret)
+                        continue
+
+                    if interpolation_type == LINEAR:
+                        vlow = skiplist_get(skiplist, idx, &ret)
+                        vhigh = skiplist_get(skiplist, idx + 1, &ret)
+                        output[i] = ((vlow + (vhigh - vlow) *
+                                      (idx_with_fraction - idx)))
+                    elif interpolation_type == LOWER:
+                        output[i] = skiplist_get(skiplist, idx, &ret)
+                    elif interpolation_type == HIGHER:
+                        output[i] = skiplist_get(skiplist, idx + 1, &ret)
+                    elif interpolation_type == NEAREST:
+                        # the same behaviour as round()
+                        if idx_with_fraction - idx == 0.5:
+                            if idx % 2 == 0:
+                                output[i] = skiplist_get(skiplist, idx, &ret)
+                            else:
+                                output[i] = skiplist_get(
+                                    skiplist, idx + 1, &ret)
+                        elif idx_with_fraction - idx < 0.5:
+                            output[i] = skiplist_get(skiplist, idx, &ret)
+                        else:
+                            output[i] = skiplist_get(skiplist, idx + 1, &ret)
+                    elif interpolation_type == MIDPOINT:
+                        vlow = skiplist_get(skiplist, idx, &ret)
+                        vhigh = skiplist_get(skiplist, idx + 1, &ret)
+                        output[i] = <float64_t>(vlow + vhigh) / 2
+
+                    if ret == 0:
+                        output[i] = NaN
+            else:
+                output[i] = NaN
+
+    skiplist_destroy(skiplist)
+
+    return output
+
+
+rolling_rank_tiebreakers = {
+    "average": TiebreakEnumType.TIEBREAK_AVERAGE,
+    "min": TiebreakEnumType.TIEBREAK_MIN,
+    "max": TiebreakEnumType.TIEBREAK_MAX,
+}
+
+
+def roll_rank(const float64_t[:] values, ndarray[int64_t] start,
+              ndarray[int64_t] end, int64_t minp, bint percentile,
+              str method, bint ascending) -> np.ndarray:
+    """
+    O(N log(window)) implementation using skip list
+
+    derived from roll_quantile
+    """
+    cdef:
+        Py_ssize_t i, j, s, e, N = len(start)
+        float64_t rank_min = 0, rank = 0
+        int64_t nobs = 0, win
+        float64_t val
+        skiplist_t *skiplist
+        float64_t[::1] output
+        TiebreakEnumType rank_type
+
+    try:
+        rank_type = rolling_rank_tiebreakers[method]
+    except KeyError:
+        raise ValueError(f"Method '{method}' is not supported")
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing_start_end_bounds(
+        start, end
+    )
+    # we use the Fixed/Variable Indexer here as the
+    # actual skiplist ops outweigh any window computation costs
+    output = np.empty(N, dtype=np.float64)
+
+    win = (end - start).max()
+    if win == 0:
+        output[:] = NaN
+        return np.asarray(output)
+    skiplist = skiplist_init(<int>win)
+    if skiplist == NULL:
+        raise MemoryError("skiplist_init failed")
+
+    with nogil:
+        for i in range(N):
+            s = start[i]
+            e = end[i]
+
+            if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
+                if i != 0:
+                    nobs = 0
+                    skiplist_destroy(skiplist)
+                    skiplist = skiplist_init(<int>win)
+
+                # setup
+                for j in range(s, e):
+                    val = values[j] if ascending else -values[j]
+                    if val == val:
+                        nobs += 1
+                        rank = skiplist_insert(skiplist, val)
+                        if rank == -1:
+                            raise MemoryError("skiplist_insert failed")
+                        if rank_type == TiebreakEnumType.TIEBREAK_AVERAGE:
+                            # The average rank of `val` is the sum of the ranks of all
+                            # instances of `val` in the skip list divided by the number
+                            # of instances. The sum of consecutive integers from 1 to N
+                            # is N * (N + 1) / 2.
+                            # The sum of the ranks is the sum of integers from the
+                            # lowest rank to the highest rank, which is the sum of
+                            # integers from 1 to the highest rank minus the sum of
+                            # integers from 1 to one less than the lowest rank.
+                            rank_min = skiplist_min_rank(skiplist, val)
+                            rank = (((rank * (rank + 1) / 2)
+                                    - ((rank_min - 1) * rank_min / 2))
+                                    / (rank - rank_min + 1))
+                        elif rank_type == TiebreakEnumType.TIEBREAK_MIN:
+                            rank = skiplist_min_rank(skiplist, val)
+                    else:
+                        rank = NaN
+
+            else:
+                # calculate deletes
+                for j in range(start[i - 1], s):
+                    val = values[j] if ascending else -values[j]
+                    if val == val:
+                        skiplist_remove(skiplist, val)
+                        nobs -= 1
+
+                # calculate adds
+                for j in range(end[i - 1], e):
+                    val = values[j] if ascending else -values[j]
+                    if val == val:
+                        nobs += 1
+                        rank = skiplist_insert(skiplist, val)
+                        if rank == -1:
+                            raise MemoryError("skiplist_insert failed")
+                        if rank_type == TiebreakEnumType.TIEBREAK_AVERAGE:
+                            rank_min = skiplist_min_rank(skiplist, val)
+                            rank = (((rank * (rank + 1) / 2)
+                                    - ((rank_min - 1) * rank_min / 2))
+                                    / (rank - rank_min + 1))
+                        elif rank_type == TiebreakEnumType.TIEBREAK_MIN:
+                            rank = skiplist_min_rank(skiplist, val)
+                    else:
+                        rank = NaN
+            if nobs >= minp:
+                output[i] = rank / nobs if percentile else rank
+            else:
+                output[i] = NaN
+
+    skiplist_destroy(skiplist)
+
+    return np.asarray(output)
+
+
+def roll_apply(object obj,
+               ndarray[int64_t] start, ndarray[int64_t] end,
+               int64_t minp,
+               object function, bint raw,
+               tuple args, dict kwargs) -> np.ndarray:
+    cdef:
+        ndarray[float64_t] output, counts
+        ndarray[float64_t, cast=True] arr
+        Py_ssize_t i, s, e, N = len(start), n = len(obj)
+
+    if n == 0:
+        return np.array([], dtype=np.float64)
+
+    arr = np.asarray(obj)
+
+    # ndarray input
+    if raw and not arr.flags.c_contiguous:
+        arr = arr.copy("C")
+
+    counts = roll_sum(np.isfinite(arr).astype(float), start, end, minp)
+
+    output = np.empty(N, dtype=np.float64)
+
+    for i in range(N):
+
+        s = start[i]
+        e = end[i]
+
+        if counts[i] >= minp:
+            if raw:
+                output[i] = function(arr[s:e], *args, **kwargs)
+            else:
+                output[i] = function(obj.iloc[s:e], *args, **kwargs)
+        else:
+            output[i] = NaN
+
+    return output
+
+
+# ----------------------------------------------------------------------
+# Rolling sum and mean for weighted window
+
+
+def roll_weighted_sum(
+    const float64_t[:] values, const float64_t[:] weights, int minp
+) -> np.ndarray:
+    return _roll_weighted_sum_mean(values, weights, minp, avg=0)
+
+
+def roll_weighted_mean(
+    const float64_t[:] values, const float64_t[:] weights, int minp
+) -> np.ndarray:
+    return _roll_weighted_sum_mean(values, weights, minp, avg=1)
+
+
+cdef float64_t[:] _roll_weighted_sum_mean(const float64_t[:] values,
+                                          const float64_t[:] weights,
+                                          int minp, bint avg):
+    """
+    Assume len(weights) << len(values)
+    """
+    cdef:
+        float64_t[:] output, tot_wgt, counts
+        Py_ssize_t in_i, win_i, win_n, in_n
+        float64_t val_in, val_win, c, w
+
+    in_n = len(values)
+    win_n = len(weights)
+
+    output = np.zeros(in_n, dtype=np.float64)
+    counts = np.zeros(in_n, dtype=np.float64)
+    if avg:
+        tot_wgt = np.zeros(in_n, dtype=np.float64)
+
+    elif minp > in_n:
+        minp = in_n + 1
+
+    minp = max(minp, 1)
+
+    with nogil:
+        if avg:
+            for win_i in range(win_n):
+                val_win = weights[win_i]
+                if val_win != val_win:
+                    continue
+
+                for in_i in range(in_n - (win_n - win_i) + 1):
+                    val_in = values[in_i]
+                    if val_in == val_in:
+                        output[in_i + (win_n - win_i) - 1] += val_in * val_win
+                        counts[in_i + (win_n - win_i) - 1] += 1
+                        tot_wgt[in_i + (win_n - win_i) - 1] += val_win
+
+            for in_i in range(in_n):
+                c = counts[in_i]
+                if c < minp:
+                    output[in_i] = NaN
+                else:
+                    w = tot_wgt[in_i]
+                    if w == 0:
+                        output[in_i] = NaN
+                    else:
+                        output[in_i] /= tot_wgt[in_i]
+
+        else:
+            for win_i in range(win_n):
+                val_win = weights[win_i]
+                if val_win != val_win:
+                    continue
+
+                for in_i in range(in_n - (win_n - win_i) + 1):
+                    val_in = values[in_i]
+
+                    if val_in == val_in:
+                        output[in_i + (win_n - win_i) - 1] += val_in * val_win
+                        counts[in_i + (win_n - win_i) - 1] += 1
+
+            for in_i in range(in_n):
+                c = counts[in_i]
+                if c < minp:
+                    output[in_i] = NaN
+
+    return output
+
+
+# ----------------------------------------------------------------------
+# Rolling var for weighted window
+
+
+cdef float64_t calc_weighted_var(float64_t t,
+                                 float64_t sum_w,
+                                 Py_ssize_t win_n,
+                                 unsigned int ddof,
+                                 float64_t nobs,
+                                 int64_t minp) noexcept nogil:
+    """
+    Calculate weighted variance for a window using West's method.
+
+    Paper: https://dl.acm.org/citation.cfm?id=359153
+
+    Parameters
+    ----------
+    t: float64_t
+        sum of weighted squared differences
+    sum_w: float64_t
+        sum of weights
+    win_n: Py_ssize_t
+        window size
+    ddof: unsigned int
+        delta degrees of freedom
+    nobs: float64_t
+        number of observations
+    minp: int64_t
+        minimum number of observations
+
+    Returns
+    -------
+    result : float64_t
+        weighted variance of the window
+    """
+
+    cdef:
+        float64_t result
+
+    # Variance is unchanged if no observation is added or removed
+    if (nobs >= minp) and (nobs > ddof):
+
+        # pathological case
+        if nobs == 1:
+            result = 0
+        else:
+            result = t * win_n / ((win_n - ddof) * sum_w)
+            if result < 0:
+                result = 0
+    else:
+        result = NaN
+
+    return result
+
+
+cdef void add_weighted_var(float64_t val,
+                           float64_t w,
+                           float64_t *t,
+                           float64_t *sum_w,
+                           float64_t *mean,
+                           float64_t *nobs) noexcept nogil:
+    """
+    Update weighted mean, sum of weights and sum of weighted squared
+    differences to include value and weight pair in weighted variance
+    calculation using West's method.
+
+    Paper: https://dl.acm.org/citation.cfm?id=359153
+
+    Parameters
+    ----------
+    val: float64_t
+        window values
+    w: float64_t
+        window weights
+    t: float64_t
+        sum of weighted squared differences
+    sum_w: float64_t
+        sum of weights
+    mean: float64_t
+        weighted mean
+    nobs: float64_t
+        number of observations
+    """
+
+    cdef:
+        float64_t temp, q, r
+
+    if val != val:
+        return
+
+    nobs[0] = nobs[0] + 1
+
+    q = val - mean[0]
+    temp = sum_w[0] + w
+    r = q * w / temp
+
+    mean[0] = mean[0] + r
+    t[0] = t[0] + r * sum_w[0] * q
+    sum_w[0] = temp
+
+
+cdef void remove_weighted_var(float64_t val,
+                              float64_t w,
+                              float64_t *t,
+                              float64_t *sum_w,
+                              float64_t *mean,
+                              float64_t *nobs) noexcept nogil:
+    """
+    Update weighted mean, sum of weights and sum of weighted squared
+    differences to remove value and weight pair from weighted variance
+    calculation using West's method.
+
+    Paper: https://dl.acm.org/citation.cfm?id=359153
+
+    Parameters
+    ----------
+    val: float64_t
+        window values
+    w: float64_t
+        window weights
+    t: float64_t
+        sum of weighted squared differences
+    sum_w: float64_t
+        sum of weights
+    mean: float64_t
+        weighted mean
+    nobs: float64_t
+        number of observations
+    """
+
+    cdef:
+        float64_t temp, q, r
+
+    if val == val:
+        nobs[0] = nobs[0] - 1
+
+        if nobs[0]:
+            q = val - mean[0]
+            temp = sum_w[0] - w
+            r = q * w / temp
+
+            mean[0] = mean[0] - r
+            t[0] = t[0] - r * sum_w[0] * q
+            sum_w[0] = temp
+
+        else:
+            t[0] = 0
+            sum_w[0] = 0
+            mean[0] = 0
+
+
+def roll_weighted_var(const float64_t[:] values, const float64_t[:] weights,
+                      int64_t minp, unsigned int ddof):
+    """
+    Calculates weighted rolling variance using West's online algorithm.
+
+    Paper: https://dl.acm.org/citation.cfm?id=359153
+
+    Parameters
+    ----------
+    values: float64_t[:]
+        values to roll window over
+    weights: float64_t[:]
+        array of weights whose length is window size
+    minp: int64_t
+        minimum number of observations to calculate
+        variance of a window
+    ddof: unsigned int
+         the divisor used in variance calculations
+         is the window size - ddof
+
+    Returns
+    -------
+    output: float64_t[:]
+        weighted variances of windows
+    """
+
+    cdef:
+        float64_t t = 0, sum_w = 0, mean = 0, nobs = 0
+        float64_t val, pre_val, w, pre_w
+        Py_ssize_t i, n, win_n
+        float64_t[:] output
+
+    n = len(values)
+    win_n = len(weights)
+    output = np.empty(n, dtype=np.float64)
+
+    with nogil:
+
+        for i in range(min(win_n, n)):
+            add_weighted_var(values[i], weights[i], &t,
+                             &sum_w, &mean, &nobs)
+
+            output[i] = calc_weighted_var(t, sum_w, win_n,
+                                          ddof, nobs, minp)
+
+        for i in range(win_n, n):
+            val = values[i]
+            pre_val = values[i - win_n]
+
+            w = weights[i % win_n]
+            pre_w = weights[(i - win_n) % win_n]
+
+            if val == val:
+                if pre_val == pre_val:
+                    remove_weighted_var(pre_val, pre_w, &t,
+                                        &sum_w, &mean, &nobs)
+
+                add_weighted_var(val, w, &t, &sum_w, &mean, &nobs)
+
+            elif pre_val == pre_val:
+                remove_weighted_var(pre_val, pre_w, &t,
+                                    &sum_w, &mean, &nobs)
+
+            output[i] = calc_weighted_var(t, sum_w, win_n,
+                                          ddof, nobs, minp)
+
+    return output
+
+
+# ----------------------------------------------------------------------
+# Exponentially weighted moving
+@cython.cpow(True)
+def ewm(const float64_t[:] vals, const int64_t[:] start, const int64_t[:] end,
+        int minp, float64_t com, bint adjust, bint ignore_na,
+        const float64_t[:] deltas=None, bint normalize=True) -> np.ndarray:
+    """
+    Compute exponentially-weighted moving average or sum using center-of-mass.
+
+    Parameters
+    ----------
+    vals : ndarray (float64 type)
+    start: ndarray (int64 type)
+    end: ndarray (int64 type)
+    minp : int
+    com : float64
+    adjust : bool
+    ignore_na : bool
+    deltas : ndarray (float64 type), optional. If None, implicitly assumes equally
+             spaced points (used when `times` is not passed)
+    normalize : bool, optional.
+                If True, calculate the mean. If False, calculate the sum.
+
+    Returns
+    -------
+    np.ndarray[float64_t]
+    """
+
+    cdef:
+        Py_ssize_t i, j, s, e, nobs, win_size, N = len(vals), M = len(start)
+        const float64_t[:] sub_vals
+        const float64_t[:] sub_deltas=None
+        ndarray[float64_t] sub_output, output = np.empty(N, dtype=np.float64)
+        float64_t alpha, old_wt_factor, new_wt, weighted, old_wt, cur
+        bint is_observation, use_deltas
+
+    if N == 0:
+        return output
+
+    use_deltas = deltas is not None
+
+    alpha = 1. / (1. + com)
+    old_wt_factor = 1. - alpha
+    new_wt = 1. if adjust else alpha
+
+    for j in range(M):
+        s = start[j]
+        e = end[j]
+        sub_vals = vals[s:e]
+        # note that len(deltas) = len(vals) - 1 and deltas[i] is to be used in
+        # conjunction with vals[i+1]
+        if use_deltas:
+            sub_deltas = deltas[s:e - 1]
+        win_size = len(sub_vals)
+        sub_output = np.empty(win_size, dtype=np.float64)
+
+        weighted = sub_vals[0]
+        is_observation = weighted == weighted
+        nobs = int(is_observation)
+        sub_output[0] = weighted if nobs >= minp else NaN
+        old_wt = 1.
+
+        with nogil:
+            for i in range(1, win_size):
+                cur = sub_vals[i]
+                is_observation = cur == cur
+                nobs += is_observation
+                if weighted == weighted:
+
+                    if is_observation or not ignore_na:
+                        if normalize:
+                            if use_deltas:
+                                old_wt *= old_wt_factor ** sub_deltas[i - 1]
+                            else:
+                                old_wt *= old_wt_factor
+                        else:
+                            weighted = old_wt_factor * weighted
+                        if is_observation:
+                            if normalize:
+                                # avoid numerical errors on constant series
+                                if weighted != cur:
+                                    if not adjust and com == 1:
+                                        # update in case of irregular-interval series
+                                        new_wt = 1. - old_wt
+                                    weighted = old_wt * weighted + new_wt * cur
+                                    weighted /= (old_wt + new_wt)
+                                if adjust:
+                                    old_wt += new_wt
+                                else:
+                                    old_wt = 1.
+                            else:
+                                weighted += cur
+                elif is_observation:
+                    weighted = cur
+
+                sub_output[i] = weighted if nobs >= minp else NaN
+
+        output[s:e] = sub_output
+
+    return output
+
+
+def ewmcov(const float64_t[:] input_x, const int64_t[:] start, const int64_t[:] end,
+           int minp, const float64_t[:] input_y, float64_t com, bint adjust,
+           bint ignore_na, bint bias) -> np.ndarray:
+    """
+    Compute exponentially-weighted moving variance using center-of-mass.
+
+    Parameters
+    ----------
+    input_x : ndarray (float64 type)
+    start: ndarray (int64 type)
+    end: ndarray (int64 type)
+    minp : int
+    input_y : ndarray (float64 type)
+    com : float64
+    adjust : bool
+    ignore_na : bool
+    bias : bool
+
+    Returns
+    -------
+    np.ndarray[float64_t]
+    """
+
+    cdef:
+        Py_ssize_t i, j, s, e, win_size, nobs
+        Py_ssize_t N = len(input_x), M = len(input_y), L = len(start)
+        float64_t alpha, old_wt_factor, new_wt, mean_x, mean_y, cov
+        float64_t sum_wt, sum_wt2, old_wt, cur_x, cur_y, old_mean_x, old_mean_y
+        float64_t numerator, denominator
+        const float64_t[:] sub_x_vals, sub_y_vals
+        ndarray[float64_t] sub_out, output = np.empty(N, dtype=np.float64)
+        bint is_observation
+
+    if M != N:
+        raise ValueError(f"arrays are of different lengths ({N} and {M})")
+
+    if N == 0:
+        return output
+
+    alpha = 1. / (1. + com)
+    old_wt_factor = 1. - alpha
+    new_wt = 1. if adjust else alpha
+
+    for j in range(L):
+        s = start[j]
+        e = end[j]
+        sub_x_vals = input_x[s:e]
+        sub_y_vals = input_y[s:e]
+        win_size = len(sub_x_vals)
+        sub_out = np.empty(win_size, dtype=np.float64)
+
+        mean_x = sub_x_vals[0]
+        mean_y = sub_y_vals[0]
+        is_observation = (mean_x == mean_x) and (mean_y == mean_y)
+        nobs = int(is_observation)
+        if not is_observation:
+            mean_x = NaN
+            mean_y = NaN
+        sub_out[0] = (0. if bias else NaN) if nobs >= minp else NaN
+        cov = 0.
+        sum_wt = 1.
+        sum_wt2 = 1.
+        old_wt = 1.
+
+        with nogil:
+            for i in range(1, win_size):
+                cur_x = sub_x_vals[i]
+                cur_y = sub_y_vals[i]
+                is_observation = (cur_x == cur_x) and (cur_y == cur_y)
+                nobs += is_observation
+                if mean_x == mean_x:
+                    if is_observation or not ignore_na:
+                        sum_wt *= old_wt_factor
+                        sum_wt2 *= (old_wt_factor * old_wt_factor)
+                        old_wt *= old_wt_factor
+                        if is_observation:
+                            old_mean_x = mean_x
+                            old_mean_y = mean_y
+
+                            # avoid numerical errors on constant series
+                            if mean_x != cur_x:
+                                mean_x = ((old_wt * old_mean_x) +
+                                          (new_wt * cur_x)) / (old_wt + new_wt)
+
+                            # avoid numerical errors on constant series
+                            if mean_y != cur_y:
+                                mean_y = ((old_wt * old_mean_y) +
+                                          (new_wt * cur_y)) / (old_wt + new_wt)
+                            cov = ((old_wt * (cov + ((old_mean_x - mean_x) *
+                                                     (old_mean_y - mean_y)))) +
+                                   (new_wt * ((cur_x - mean_x) *
+                                              (cur_y - mean_y)))) / (old_wt + new_wt)
+                            sum_wt += new_wt
+                            sum_wt2 += (new_wt * new_wt)
+                            old_wt += new_wt
+                            if not adjust:
+                                sum_wt /= old_wt
+                                sum_wt2 /= (old_wt * old_wt)
+                                old_wt = 1.
+                elif is_observation:
+                    mean_x = cur_x
+                    mean_y = cur_y
+
+                if nobs >= minp:
+                    if not bias:
+                        numerator = sum_wt * sum_wt
+                        denominator = numerator - sum_wt2
+                        if denominator > 0:
+                            sub_out[i] = (numerator / denominator) * cov
+                        else:
+                            sub_out[i] = NaN
+                    else:
+                        sub_out[i] = cov
+                else:
+                    sub_out[i] = NaN
+
+        output[s:e] = sub_out
+
+    return output
diff --git a/pandas/pandas/_libs/window/indexers.pyi b/pandas/pandas/_libs/window/indexers.pyi
new file mode 100644
index 0000000000000..c9bc64be34ac9
--- /dev/null
+++ b/pandas/pandas/_libs/window/indexers.pyi
@@ -0,0 +1,12 @@
+import numpy as np
+
+from pandas._typing import npt
+
+def calculate_variable_window_bounds(
+    num_values: int,  # int64_t
+    window_size: int,  # int64_t
+    min_periods,
+    center: bool,
+    closed: str | None,
+    index: np.ndarray,  # const int64_t[:]
+) -> tuple[npt.NDArray[np.int64], npt.NDArray[np.int64]]: ...
diff --git a/pandas/pandas/_libs/window/indexers.pyx b/pandas/pandas/_libs/window/indexers.pyx
new file mode 100644
index 0000000000000..7b306c5e681e0
--- /dev/null
+++ b/pandas/pandas/_libs/window/indexers.pyx
@@ -0,0 +1,151 @@
+# cython: boundscheck=False, wraparound=False, cdivision=True
+
+import numpy as np
+
+from numpy cimport (
+    int64_t,
+    ndarray,
+)
+
+# Cython routines for window indexers
+
+
+def calculate_variable_window_bounds(
+    int64_t num_values,
+    int64_t window_size,
+    object min_periods,  # unused but here to match get_window_bounds signature
+    bint center,
+    str closed,
+    const int64_t[:] index
+):
+    """
+    Calculate window boundaries for rolling windows from a time offset.
+
+    Parameters
+    ----------
+    num_values : int64
+        total number of values
+
+    window_size : int64
+        window size calculated from the offset
+
+    min_periods : object
+        ignored, exists for compatibility
+
+    center : bint
+        center the rolling window on the current observation
+
+    closed : str
+        string of side of the window that should be closed
+
+    index : ndarray[int64]
+        time series index to roll over
+
+    Returns
+    -------
+    (ndarray[int64], ndarray[int64])
+    """
+    cdef:
+        bint left_closed = False
+        bint right_closed = False
+        ndarray[int64_t, ndim=1] start, end
+        int64_t start_bound, end_bound, index_growth_sign = 1
+        Py_ssize_t i, j
+
+    if num_values <= 0:
+        return np.empty(0, dtype="int64"), np.empty(0, dtype="int64")
+
+    # default is 'right'
+    if closed is None:
+        closed = "right"
+
+    if closed in ["right", "both"]:
+        right_closed = True
+
+    if closed in ["left", "both"]:
+        left_closed = True
+
+    # GH 43997:
+    # If the forward and the backward facing windows
+    # would result in a fraction of 1/2 a nanosecond
+    # we need to make both interval ends inclusive.
+    if center and window_size % 2 == 1:
+        right_closed = True
+        left_closed = True
+
+    if index[num_values - 1] < index[0]:
+        index_growth_sign = -1
+
+    start = np.empty(num_values, dtype="int64")
+    start.fill(-1)
+    end = np.empty(num_values, dtype="int64")
+    end.fill(-1)
+
+    start[0] = 0
+
+    # right endpoint is closed
+    if right_closed:
+        end[0] = 1
+    # right endpoint is open
+    else:
+        end[0] = 0
+    if center:
+        end_bound = index[0] + index_growth_sign * window_size / 2
+        for j in range(0, num_values):
+            if (index[j] - end_bound) * index_growth_sign < 0:
+                end[0] = j + 1
+            elif (index[j] - end_bound) * index_growth_sign == 0 and right_closed:
+                end[0] = j + 1
+            elif (index[j] - end_bound) * index_growth_sign >= 0:
+                end[0] = j
+                break
+
+    with nogil:
+
+        # start is start of slice interval (including)
+        # end is end of slice interval (not including)
+        for i in range(1, num_values):
+            if center:
+                end_bound = index[i] + index_growth_sign * window_size / 2
+                start_bound = index[i] - index_growth_sign * window_size / 2
+            else:
+                end_bound = index[i]
+                start_bound = index[i] - index_growth_sign * window_size
+
+            # left endpoint is closed
+            if left_closed:
+                start_bound -= 1 * index_growth_sign
+
+            # advance the start bound until we are
+            # within the constraint
+            start[i] = i
+            for j in range(start[i - 1], i):
+                if (index[j] - start_bound) * index_growth_sign > 0:
+                    start[i] = j
+                    break
+
+            # for centered window advance the end bound until we are
+            # outside the constraint
+            if center:
+                for j in range(end[i - 1], num_values + 1):
+                    if j == num_values:
+                        end[i] = j
+                    elif ((index[j] - end_bound) * index_growth_sign == 0 and
+                          right_closed):
+                        end[i] = j + 1
+                    elif (index[j] - end_bound) * index_growth_sign >= 0:
+                        end[i] = j
+                        break
+            # end bound is previous end
+            # or current index
+            elif index[end[i - 1]] == end_bound and not right_closed:
+                end[i] = end[i - 1] + 1
+            elif (index[end[i - 1]] - end_bound) * index_growth_sign <= 0:
+                end[i] = i + 1
+            else:
+                end[i] = end[i - 1]
+
+            # right endpoint is open
+            if not right_closed and not center:
+                end[i] -= 1
+    return start, end
diff --git a/pandas/pandas/_libs/window/meson.build b/pandas/pandas/_libs/window/meson.build
new file mode 100644
index 0000000000000..1d49bba47e139
--- /dev/null
+++ b/pandas/pandas/_libs/window/meson.build
@@ -0,0 +1,24 @@
+py.extension_module(
+    'aggregations',
+    ['aggregations.pyx'],
+    cython_args: ['-X always_allow_keywords=true'],
+    include_directories: [inc_np, inc_pd],
+    subdir: 'pandas/_libs/window',
+    override_options: ['cython_language=cpp'],
+    install: true,
+)
+
+py.extension_module(
+    'indexers',
+    ['indexers.pyx'],
+    cython_args: ['-X always_allow_keywords=true'],
+    include_directories: [inc_np, inc_pd],
+    subdir: 'pandas/_libs/window',
+    install: true,
+)
+
+sources_to_install = ['__init__.py', 'aggregations.pyi', 'indexers.pyi']
+
+foreach source : sources_to_install
+    py.install_sources(source, subdir: 'pandas/_libs/window')
+endforeach
diff --git a/pandas/pandas/_libs/writers.pyi b/pandas/pandas/_libs/writers.pyi
new file mode 100644
index 0000000000000..7b41856525dad
--- /dev/null
+++ b/pandas/pandas/_libs/writers.pyi
@@ -0,0 +1,20 @@
+import numpy as np
+
+from pandas._typing import ArrayLike
+
+def write_csv_rows(
+    data: list[ArrayLike],
+    data_index: np.ndarray,
+    nlevels: int,
+    cols: np.ndarray,
+    writer: object,  # _csv.writer
+) -> None: ...
+def convert_json_to_lines(arr: str) -> str: ...
+def max_len_string_array(
+    arr: np.ndarray,  # pandas_string[:]
+) -> int: ...
+def word_len(val: object) -> int: ...
+def string_array_replace_from_nan_rep(
+    arr: np.ndarray,  # np.ndarray[object, ndim=1]
+    nan_rep: object,
+) -> None: ...
diff --git a/pandas/pandas/_libs/writers.pyx b/pandas/pandas/_libs/writers.pyx
new file mode 100644
index 0000000000000..48eedd1fd1af6
--- /dev/null
+++ b/pandas/pandas/_libs/writers.pyx
@@ -0,0 +1,174 @@
+cimport cython
+from cython cimport Py_ssize_t
+import numpy as np
+
+from cpython cimport (
+    PyBytes_GET_SIZE,
+    PyUnicode_GET_LENGTH,
+)
+from numpy cimport (
+    ndarray,
+    uint8_t,
+)
+
+ctypedef fused pandas_string:
+    str
+    bytes
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def write_csv_rows(
+    list data,
+    ndarray data_index,
+    Py_ssize_t nlevels,
+    ndarray cols,
+    object writer
+) -> None:
+    """
+    Write the given data to the writer object, pre-allocating where possible
+    for performance improvements.
+
+    Parameters
+    ----------
+    data : list[ArrayLike]
+    data_index : ndarray
+    nlevels : int
+    cols : ndarray
+    writer : _csv.writer
+    """
+    # In crude testing, N>100 yields little marginal improvement
+    cdef:
+        Py_ssize_t i, j = 0, k = len(data_index), N = 100, ncols = len(cols)
+        list rows
+
+    # pre-allocate rows
+    rows = [[None] * (nlevels + ncols) for _ in range(N)]
+
+    if nlevels == 1:
+        for j in range(k):
+            row = rows[j % N]
+            row[0] = data_index[j]
+            for i in range(ncols):
+                row[1 + i] = data[i][j]
+
+            if j >= N - 1 and j % N == N - 1:
+                writer.writerows(rows)
+    elif nlevels > 1:
+        for j in range(k):
+            row = rows[j % N]
+            row[:nlevels] = list(data_index[j])
+            for i in range(ncols):
+                row[nlevels + i] = data[i][j]
+
+            if j >= N - 1 and j % N == N - 1:
+                writer.writerows(rows)
+    else:
+        for j in range(k):
+            row = rows[j % N]
+            for i in range(ncols):
+                row[i] = data[i][j]
+
+            if j >= N - 1 and j % N == N - 1:
+                writer.writerows(rows)
+
+    if j >= 0 and (j < N - 1 or (j % N) != N - 1):
+        writer.writerows(rows[:((j + 1) % N)])
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def convert_json_to_lines(arr: str) -> str:
+    """
+    replace comma separated json with line feeds, paying special attention
+    to quotes & brackets
+    """
+    cdef:
+        Py_ssize_t i = 0, num_open_brackets_seen = 0, length
+        bint in_quotes = False, is_escaping = False
+        ndarray[uint8_t, ndim=1] narr
+        unsigned char val, newline, comma, left_bracket, right_bracket, quote
+        unsigned char backslash
+
+    newline = ord("\n")
+    comma = ord(",")
+    left_bracket = ord("{")
+    right_bracket = ord("}")
+    quote = ord('"')
+    backslash = ord("\\")
+
+    narr = np.frombuffer(arr.encode("utf-8"), dtype="u1").copy()
+    length = narr.shape[0]
+    for i in range(length):
+        val = narr[i]
+        if val == quote and i > 0 and not is_escaping:
+            in_quotes = ~in_quotes
+        if val == backslash or is_escaping:
+            is_escaping = ~is_escaping
+        if val == comma:  # commas that should be \n
+            if num_open_brackets_seen == 0 and not in_quotes:
+                narr[i] = newline
+        elif val == left_bracket:
+            if not in_quotes:
+                num_open_brackets_seen += 1
+        elif val == right_bracket:
+            if not in_quotes:
+                num_open_brackets_seen -= 1
+
+    return narr.tobytes().decode("utf-8") + "\n"  # GH:36888
+
+
+# stata, pytables
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def max_len_string_array(pandas_string[:] arr) -> Py_ssize_t:
+    """
+    Return the maximum size of elements in a 1-dim string array.
+    """
+    cdef:
+        Py_ssize_t i, m = 0, wlen = 0, length = arr.shape[0]
+        pandas_string val
+
+    for i in range(length):
+        val = arr[i]
+        wlen = word_len(val)
+
+        if wlen > m:
+            m = wlen
+
+    return m
+
+
+cpdef inline Py_ssize_t word_len(object val):
+    """
+    Return the maximum length of a string or bytes value.
+    """
+    cdef:
+        Py_ssize_t wlen = 0
+
+    if isinstance(val, str):
+        wlen = PyUnicode_GET_LENGTH(val)
+    elif isinstance(val, bytes):
+        wlen = PyBytes_GET_SIZE(val)
+
+    return wlen
+
+# ------------------------------------------------------------------
+# PyTables Helpers
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+def string_array_replace_from_nan_rep(
+    ndarray[object, ndim=1] arr,
+    object nan_rep,
+) -> None:
+    """
+    Replace the values in the array with np.nan if they are nan_rep.
+    """
+    cdef:
+        Py_ssize_t length = len(arr), i = 0
+
+    for i in range(length):
+        if arr[i] == nan_rep:
+            arr[i] = np.nan
diff --git a/pandas/pandas/_testing/__init__.py b/pandas/pandas/_testing/__init__.py
new file mode 100644
index 0000000000000..ec9b5098c97c9
--- /dev/null
+++ b/pandas/pandas/_testing/__init__.py
@@ -0,0 +1,613 @@
+from __future__ import annotations
+
+from decimal import Decimal
+import operator
+import os
+from sys import byteorder
+from typing import (
+    TYPE_CHECKING,
+    ContextManager,
+)
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+from pandas._config.localization import (
+    can_set_locale,
+    get_locales,
+    set_locale,
+)
+
+from pandas.compat import pa_version_under10p1
+
+import pandas as pd
+from pandas import (
+    ArrowDtype,
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+)
+from pandas._testing._io import (
+    round_trip_pathlib,
+    round_trip_pickle,
+    write_to_compressed,
+)
+from pandas._testing._warnings import (
+    assert_produces_warning,
+    maybe_produces_warning,
+)
+from pandas._testing.asserters import (
+    assert_almost_equal,
+    assert_attr_equal,
+    assert_categorical_equal,
+    assert_class_equal,
+    assert_contains_all,
+    assert_copy,
+    assert_datetime_array_equal,
+    assert_dict_equal,
+    assert_equal,
+    assert_extension_array_equal,
+    assert_frame_equal,
+    assert_index_equal,
+    assert_indexing_slices_equivalent,
+    assert_interval_array_equal,
+    assert_is_sorted,
+    assert_metadata_equivalent,
+    assert_numpy_array_equal,
+    assert_period_array_equal,
+    assert_series_equal,
+    assert_sp_array_equal,
+    assert_timedelta_array_equal,
+    raise_assert_detail,
+)
+from pandas._testing.compat import (
+    get_dtype,
+    get_obj,
+)
+from pandas._testing.contexts import (
+    decompress_file,
+    ensure_clean,
+    raises_chained_assignment_error,
+    set_timezone,
+    with_csv_dialect,
+)
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BaseMaskedArray,
+    NumpyExtensionArray,
+)
+from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+from pandas.core.construction import extract_array
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        Dtype,
+        NpDtype,
+    )
+
+
+UNSIGNED_INT_NUMPY_DTYPES: list[NpDtype] = ["uint8", "uint16", "uint32", "uint64"]
+UNSIGNED_INT_EA_DTYPES: list[Dtype] = ["UInt8", "UInt16", "UInt32", "UInt64"]
+SIGNED_INT_NUMPY_DTYPES: list[NpDtype] = [int, "int8", "int16", "int32", "int64"]
+SIGNED_INT_EA_DTYPES: list[Dtype] = ["Int8", "Int16", "Int32", "Int64"]
+ALL_INT_NUMPY_DTYPES = UNSIGNED_INT_NUMPY_DTYPES + SIGNED_INT_NUMPY_DTYPES
+ALL_INT_EA_DTYPES = UNSIGNED_INT_EA_DTYPES + SIGNED_INT_EA_DTYPES
+ALL_INT_DTYPES: list[Dtype] = [*ALL_INT_NUMPY_DTYPES, *ALL_INT_EA_DTYPES]
+
+FLOAT_NUMPY_DTYPES: list[NpDtype] = [float, "float32", "float64"]
+FLOAT_EA_DTYPES: list[Dtype] = ["Float32", "Float64"]
+ALL_FLOAT_DTYPES: list[Dtype] = [*FLOAT_NUMPY_DTYPES, *FLOAT_EA_DTYPES]
+
+COMPLEX_DTYPES: list[Dtype] = [complex, "complex64", "complex128"]
+if using_string_dtype():
+    STRING_DTYPES: list[Dtype] = ["U"]
+else:
+    STRING_DTYPES: list[Dtype] = [str, "str", "U"]  # type: ignore[no-redef]
+COMPLEX_FLOAT_DTYPES: list[Dtype] = [*COMPLEX_DTYPES, *FLOAT_NUMPY_DTYPES]
+
+DATETIME64_DTYPES: list[Dtype] = ["datetime64[ns]", "M8[ns]"]
+TIMEDELTA64_DTYPES: list[Dtype] = ["timedelta64[ns]", "m8[ns]"]
+
+BOOL_DTYPES: list[Dtype] = [bool, "bool"]
+BYTES_DTYPES: list[Dtype] = [bytes, "bytes"]
+OBJECT_DTYPES: list[Dtype] = [object, "object"]
+
+ALL_REAL_NUMPY_DTYPES = FLOAT_NUMPY_DTYPES + ALL_INT_NUMPY_DTYPES
+ALL_REAL_EXTENSION_DTYPES = FLOAT_EA_DTYPES + ALL_INT_EA_DTYPES
+ALL_REAL_DTYPES: list[Dtype] = [*ALL_REAL_NUMPY_DTYPES, *ALL_REAL_EXTENSION_DTYPES]
+ALL_NUMERIC_DTYPES: list[Dtype] = [*ALL_REAL_DTYPES, *COMPLEX_DTYPES]
+
+ALL_NUMPY_DTYPES = (
+    ALL_REAL_NUMPY_DTYPES
+    + COMPLEX_DTYPES
+    + STRING_DTYPES
+    + DATETIME64_DTYPES
+    + TIMEDELTA64_DTYPES
+    + BOOL_DTYPES
+    + OBJECT_DTYPES
+    + BYTES_DTYPES
+)
+
+NARROW_NP_DTYPES = [
+    np.float16,
+    np.float32,
+    np.int8,
+    np.int16,
+    np.int32,
+    np.uint8,
+    np.uint16,
+    np.uint32,
+]
+
+PYTHON_DATA_TYPES = [
+    str,
+    int,
+    float,
+    complex,
+    list,
+    tuple,
+    range,
+    dict,
+    set,
+    frozenset,
+    bool,
+    bytes,
+    bytearray,
+    memoryview,
+]
+
+ENDIAN = {"little": "<", "big": ">"}[byteorder]
+
+NULL_OBJECTS = [None, np.nan, pd.NaT, float("nan"), pd.NA, Decimal("NaN")]
+NP_NAT_OBJECTS = [
+    cls("NaT", unit)
+    for cls in [np.datetime64, np.timedelta64]
+    for unit in [
+        "Y",
+        "M",
+        "W",
+        "D",
+        "h",
+        "m",
+        "s",
+        "ms",
+        "us",
+        "ns",
+        "ps",
+        "fs",
+        "as",
+    ]
+]
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+
+    UNSIGNED_INT_PYARROW_DTYPES = [pa.uint8(), pa.uint16(), pa.uint32(), pa.uint64()]
+    SIGNED_INT_PYARROW_DTYPES = [pa.int8(), pa.int16(), pa.int32(), pa.int64()]
+    ALL_INT_PYARROW_DTYPES = UNSIGNED_INT_PYARROW_DTYPES + SIGNED_INT_PYARROW_DTYPES
+    ALL_INT_PYARROW_DTYPES_STR_REPR = [
+        str(ArrowDtype(typ)) for typ in ALL_INT_PYARROW_DTYPES
+    ]
+
+    # pa.float16 doesn't seem supported
+    # https://github.com/apache/arrow/blob/master/python/pyarrow/src/arrow/python/helpers.cc#L86
+    FLOAT_PYARROW_DTYPES = [pa.float32(), pa.float64()]
+    FLOAT_PYARROW_DTYPES_STR_REPR = [
+        str(ArrowDtype(typ)) for typ in FLOAT_PYARROW_DTYPES
+    ]
+    DECIMAL_PYARROW_DTYPES = [pa.decimal128(7, 3)]
+    STRING_PYARROW_DTYPES = [pa.string()]
+    BINARY_PYARROW_DTYPES = [pa.binary()]
+
+    TIME_PYARROW_DTYPES = [
+        pa.time32("s"),
+        pa.time32("ms"),
+        pa.time64("us"),
+        pa.time64("ns"),
+    ]
+    DATE_PYARROW_DTYPES = [pa.date32(), pa.date64()]
+    DATETIME_PYARROW_DTYPES = [
+        pa.timestamp(unit=unit, tz=tz)
+        for unit in ["s", "ms", "us", "ns"]
+        for tz in [None, "UTC", "US/Pacific", "US/Eastern"]
+    ]
+    TIMEDELTA_PYARROW_DTYPES = [pa.duration(unit) for unit in ["s", "ms", "us", "ns"]]
+
+    BOOL_PYARROW_DTYPES = [pa.bool_()]
+
+    # TODO: Add container like pyarrow types:
+    #  https://arrow.apache.org/docs/python/api/datatypes.html#factory-functions
+    ALL_PYARROW_DTYPES = (
+        ALL_INT_PYARROW_DTYPES
+        + FLOAT_PYARROW_DTYPES
+        + DECIMAL_PYARROW_DTYPES
+        + STRING_PYARROW_DTYPES
+        + BINARY_PYARROW_DTYPES
+        + TIME_PYARROW_DTYPES
+        + DATE_PYARROW_DTYPES
+        + DATETIME_PYARROW_DTYPES
+        + TIMEDELTA_PYARROW_DTYPES
+        + BOOL_PYARROW_DTYPES
+    )
+    ALL_REAL_PYARROW_DTYPES_STR_REPR = (
+        ALL_INT_PYARROW_DTYPES_STR_REPR + FLOAT_PYARROW_DTYPES_STR_REPR
+    )
+else:
+    FLOAT_PYARROW_DTYPES_STR_REPR = []
+    ALL_INT_PYARROW_DTYPES_STR_REPR = []
+    ALL_PYARROW_DTYPES = []
+    ALL_REAL_PYARROW_DTYPES_STR_REPR = []
+
+ALL_REAL_NULLABLE_DTYPES = (
+    FLOAT_NUMPY_DTYPES + ALL_REAL_EXTENSION_DTYPES + ALL_REAL_PYARROW_DTYPES_STR_REPR
+)
+
+arithmetic_dunder_methods = [
+    "__add__",
+    "__radd__",
+    "__sub__",
+    "__rsub__",
+    "__mul__",
+    "__rmul__",
+    "__floordiv__",
+    "__rfloordiv__",
+    "__truediv__",
+    "__rtruediv__",
+    "__pow__",
+    "__rpow__",
+    "__mod__",
+    "__rmod__",
+]
+
+comparison_dunder_methods = ["__eq__", "__ne__", "__le__", "__lt__", "__ge__", "__gt__"]
+
+
+# -----------------------------------------------------------------------------
+# Comparators
+
+
+def box_expected(expected, box_cls, transpose: bool = True):
+    """
+    Helper function to wrap the expected output of a test in a given box_class.
+
+    Parameters
+    ----------
+    expected : np.ndarray, Index, Series
+    box_cls : {Index, Series, DataFrame}
+
+    Returns
+    -------
+    subclass of box_cls
+    """
+    if box_cls is pd.array:
+        if isinstance(expected, RangeIndex):
+            # pd.array would return an IntegerArray
+            expected = NumpyExtensionArray(np.asarray(expected._values))
+        else:
+            expected = pd.array(expected, copy=False)
+    elif box_cls is Index:
+        expected = Index(expected)
+    elif box_cls is Series:
+        expected = Series(expected)
+    elif box_cls is DataFrame:
+        expected = Series(expected).to_frame()
+        if transpose:
+            # for vector operations, we need a DataFrame to be a single-row,
+            #  not a single-column, in order to operate against non-DataFrame
+            #  vectors of the same length. But convert to two rows to avoid
+            #  single-row special cases in datetime arithmetic
+            expected = expected.T
+            expected = pd.concat([expected] * 2, ignore_index=True)
+    elif box_cls is np.ndarray or box_cls is np.array:
+        expected = np.array(expected)
+    elif box_cls is to_array:
+        expected = to_array(expected)
+    else:
+        raise NotImplementedError(box_cls)
+    return expected
+
+
+def to_array(obj):
+    """
+    Similar to pd.array, but does not cast numpy dtypes to nullable dtypes.
+    """
+    # temporary implementation until we get pd.array in place
+    dtype = getattr(obj, "dtype", None)
+
+    if dtype is None:
+        return np.asarray(obj)
+
+    return extract_array(obj, extract_numpy=True)
+
+
+class SubclassedSeries(Series):
+    _metadata = ["testattr", "name"]
+
+    @property
+    def _constructor(self):
+        # For testing, those properties return a generic callable, and not
+        # the actual class. In this case that is equivalent, but it is to
+        # ensure we don't rely on the property returning a class
+        # See https://github.com/pandas-dev/pandas/pull/46018 and
+        # https://github.com/pandas-dev/pandas/issues/32638 and linked issues
+        return lambda *args, **kwargs: SubclassedSeries(*args, **kwargs)
+
+    @property
+    def _constructor_expanddim(self):
+        return lambda *args, **kwargs: SubclassedDataFrame(*args, **kwargs)
+
+
+class SubclassedDataFrame(DataFrame):
+    _metadata = ["testattr"]
+
+    @property
+    def _constructor(self):
+        return lambda *args, **kwargs: SubclassedDataFrame(*args, **kwargs)
+
+    @property
+    def _constructor_sliced(self):
+        return lambda *args, **kwargs: SubclassedSeries(*args, **kwargs)
+
+
+def convert_rows_list_to_csv_str(rows_list: list[str]) -> str:
+    """
+    Convert list of CSV rows to single CSV-formatted string for current OS.
+
+    This method is used for creating expected value of to_csv() method.
+
+    Parameters
+    ----------
+    rows_list : List[str]
+        Each element represents the row of csv.
+
+    Returns
+    -------
+    str
+        Expected output of to_csv() in current OS.
+    """
+    sep = os.linesep
+    return sep.join(rows_list) + sep
+
+
+def external_error_raised(expected_exception: type[Exception]) -> ContextManager:
+    """
+    Helper function to mark pytest.raises that have an external error message.
+
+    Parameters
+    ----------
+    expected_exception : Exception
+        Expected error to raise.
+
+    Returns
+    -------
+    Callable
+        Regular `pytest.raises` function with `match` equal to `None`.
+    """
+    import pytest
+
+    return pytest.raises(expected_exception, match=None)
+
+
+def get_cython_table_params(ndframe, func_names_and_expected):
+    """
+    Combine frame, functions from com._cython_table
+    keys and expected result.
+
+    Parameters
+    ----------
+    ndframe : DataFrame or Series
+    func_names_and_expected : Sequence of two items
+        The first item is a name of a NDFrame method ('sum', 'prod') etc.
+        The second item is the expected return value.
+
+    Returns
+    -------
+    list
+        List of three items (DataFrame, function, expected result)
+    """
+    results = []
+    for func_name, expected in func_names_and_expected:
+        results.append((ndframe, func_name, expected))
+    return results
+
+
+def get_op_from_name(op_name: str) -> Callable:
+    """
+    The operator function for a given op name.
+
+    Parameters
+    ----------
+    op_name : str
+        The op name, in form of "add" or "__add__".
+
+    Returns
+    -------
+    function
+        A function performing the operation.
+    """
+    short_opname = op_name.strip("_")
+    try:
+        op = getattr(operator, short_opname)
+    except AttributeError:
+        # Assume it is the reverse operator
+        rop = getattr(operator, short_opname[1:])
+        op = lambda x, y: rop(y, x)
+
+    return op
+
+
+# -----------------------------------------------------------------------------
+# Indexing test helpers
+
+
+def getitem(x):
+    return x
+
+
+def setitem(x):
+    return x
+
+
+def loc(x):
+    return x.loc
+
+
+def iloc(x):
+    return x.iloc
+
+
+def at(x):
+    return x.at
+
+
+def iat(x):
+    return x.iat
+
+
+# -----------------------------------------------------------------------------
+
+_UNITS = ["s", "ms", "us", "ns"]
+
+
+def get_finest_unit(left: str, right: str) -> str:
+    """
+    Find the higher of two datetime64 units.
+    """
+    if _UNITS.index(left) >= _UNITS.index(right):
+        return left
+    return right
+
+
+def shares_memory(left, right) -> bool:
+    """
+    Pandas-compat for np.shares_memory.
+    """
+    if isinstance(left, np.ndarray) and isinstance(right, np.ndarray):
+        return np.shares_memory(left, right)
+    elif isinstance(left, np.ndarray):
+        # Call with reversed args to get to unpacking logic below.
+        return shares_memory(right, left)
+
+    if isinstance(left, RangeIndex):
+        return False
+    if isinstance(left, MultiIndex):
+        return shares_memory(left._codes, right)
+    if isinstance(left, (Index, Series)):
+        if isinstance(right, (Index, Series)):
+            return shares_memory(left._values, right._values)
+        return shares_memory(left._values, right)
+
+    if isinstance(left, NDArrayBackedExtensionArray):
+        return shares_memory(left._ndarray, right)
+    if isinstance(left, pd.core.arrays.SparseArray):
+        return shares_memory(left.sp_values, right)
+    if isinstance(left, pd.core.arrays.IntervalArray):
+        return shares_memory(left._left, right) or shares_memory(left._right, right)
+
+    if isinstance(left, ArrowExtensionArray):
+        if isinstance(right, ArrowExtensionArray):
+            # https://github.com/pandas-dev/pandas/pull/43930#discussion_r736862669
+            left_pa_data = left._pa_array
+            right_pa_data = right._pa_array
+            left_buf1 = left_pa_data.chunk(0).buffers()[1]
+            right_buf1 = right_pa_data.chunk(0).buffers()[1]
+            return left_buf1.address == right_buf1.address
+        else:
+            # if we have one one ArrowExtensionArray and one other array, assume
+            # they can only share memory if they share the same numpy buffer
+            return np.shares_memory(left, right)
+
+    if isinstance(left, BaseMaskedArray) and isinstance(right, BaseMaskedArray):
+        # By convention, we'll say these share memory if they share *either*
+        #  the _data or the _mask
+        return np.shares_memory(left._data, right._data) or np.shares_memory(
+            left._mask, right._mask
+        )
+
+    if isinstance(left, DataFrame) and len(left._mgr.blocks) == 1:
+        arr = left._mgr.blocks[0].values
+        return shares_memory(arr, right)
+
+    raise NotImplementedError(type(left), type(right))
+
+
+__all__ = [
+    "ALL_INT_EA_DTYPES",
+    "ALL_INT_NUMPY_DTYPES",
+    "ALL_NUMPY_DTYPES",
+    "ALL_REAL_NUMPY_DTYPES",
+    "BOOL_DTYPES",
+    "BYTES_DTYPES",
+    "COMPLEX_DTYPES",
+    "DATETIME64_DTYPES",
+    "ENDIAN",
+    "FLOAT_EA_DTYPES",
+    "FLOAT_NUMPY_DTYPES",
+    "NARROW_NP_DTYPES",
+    "NP_NAT_OBJECTS",
+    "NULL_OBJECTS",
+    "OBJECT_DTYPES",
+    "SIGNED_INT_EA_DTYPES",
+    "SIGNED_INT_NUMPY_DTYPES",
+    "STRING_DTYPES",
+    "TIMEDELTA64_DTYPES",
+    "UNSIGNED_INT_EA_DTYPES",
+    "UNSIGNED_INT_NUMPY_DTYPES",
+    "SubclassedDataFrame",
+    "SubclassedSeries",
+    "assert_almost_equal",
+    "assert_attr_equal",
+    "assert_categorical_equal",
+    "assert_class_equal",
+    "assert_contains_all",
+    "assert_copy",
+    "assert_datetime_array_equal",
+    "assert_dict_equal",
+    "assert_equal",
+    "assert_extension_array_equal",
+    "assert_frame_equal",
+    "assert_index_equal",
+    "assert_indexing_slices_equivalent",
+    "assert_interval_array_equal",
+    "assert_is_sorted",
+    "assert_metadata_equivalent",
+    "assert_numpy_array_equal",
+    "assert_period_array_equal",
+    "assert_produces_warning",
+    "assert_series_equal",
+    "assert_sp_array_equal",
+    "assert_timedelta_array_equal",
+    "at",
+    "box_expected",
+    "can_set_locale",
+    "convert_rows_list_to_csv_str",
+    "decompress_file",
+    "ensure_clean",
+    "external_error_raised",
+    "get_cython_table_params",
+    "get_dtype",
+    "get_finest_unit",
+    "get_locales",
+    "get_obj",
+    "get_op_from_name",
+    "getitem",
+    "iat",
+    "iloc",
+    "loc",
+    "maybe_produces_warning",
+    "raise_assert_detail",
+    "raises_chained_assignment_error",
+    "round_trip_pathlib",
+    "round_trip_pickle",
+    "set_locale",
+    "set_timezone",
+    "setitem",
+    "shares_memory",
+    "to_array",
+    "with_csv_dialect",
+    "write_to_compressed",
+]
diff --git a/pandas/pandas/_testing/_hypothesis.py b/pandas/pandas/_testing/_hypothesis.py
new file mode 100644
index 0000000000000..bbad21d8ab8d1
--- /dev/null
+++ b/pandas/pandas/_testing/_hypothesis.py
@@ -0,0 +1,89 @@
+"""
+Hypothesis data generator helpers.
+"""
+
+from datetime import datetime
+
+from hypothesis import strategies as st
+from hypothesis.extra.dateutil import timezones as dateutil_timezones
+
+from pandas.compat import is_platform_windows
+
+import pandas as pd
+
+from pandas.tseries.offsets import (
+    BMonthBegin,
+    BMonthEnd,
+    BQuarterBegin,
+    BQuarterEnd,
+    BYearBegin,
+    BYearEnd,
+    MonthBegin,
+    MonthEnd,
+    QuarterBegin,
+    QuarterEnd,
+    YearBegin,
+    YearEnd,
+)
+
+OPTIONAL_INTS = st.lists(st.one_of(st.integers(), st.none()), max_size=10, min_size=3)
+
+OPTIONAL_FLOATS = st.lists(st.one_of(st.floats(), st.none()), max_size=10, min_size=3)
+
+OPTIONAL_TEXT = st.lists(st.one_of(st.none(), st.text()), max_size=10, min_size=3)
+
+OPTIONAL_DICTS = st.lists(
+    st.one_of(st.none(), st.dictionaries(st.text(), st.integers())),
+    max_size=10,
+    min_size=3,
+)
+
+OPTIONAL_LISTS = st.lists(
+    st.one_of(st.none(), st.lists(st.text(), max_size=10, min_size=3)),
+    max_size=10,
+    min_size=3,
+)
+
+OPTIONAL_ONE_OF_ALL = st.one_of(
+    OPTIONAL_DICTS, OPTIONAL_FLOATS, OPTIONAL_INTS, OPTIONAL_LISTS, OPTIONAL_TEXT
+)
+
+if is_platform_windows():
+    DATETIME_NO_TZ = st.datetimes(min_value=datetime(1900, 1, 1))
+else:
+    DATETIME_NO_TZ = st.datetimes()
+
+DATETIME_JAN_1_1900_OPTIONAL_TZ = st.datetimes(
+    min_value=pd.Timestamp(1900, 1, 1).to_pydatetime(),  # pyright: ignore[reportArgumentType]
+    max_value=pd.Timestamp(1900, 1, 1).to_pydatetime(),  # pyright: ignore[reportArgumentType]
+    timezones=st.one_of(st.none(), dateutil_timezones(), st.timezones()),
+)
+
+DATETIME_IN_PD_TIMESTAMP_RANGE_NO_TZ = st.datetimes(
+    min_value=pd.Timestamp.min.to_pydatetime(warn=False),
+    max_value=pd.Timestamp.max.to_pydatetime(warn=False),
+)
+
+INT_NEG_999_TO_POS_999 = st.integers(-999, 999)
+
+# The strategy for each type is registered in conftest.py, as they don't carry
+# enough runtime information (e.g. type hints) to infer how to build them.
+YQM_OFFSET = st.one_of(
+    *map(
+        st.from_type,
+        [
+            MonthBegin,
+            MonthEnd,
+            BMonthBegin,
+            BMonthEnd,
+            QuarterBegin,
+            QuarterEnd,
+            BQuarterBegin,
+            BQuarterEnd,
+            YearBegin,
+            YearEnd,
+            BYearBegin,
+            BYearEnd,
+        ],
+    )
+)
diff --git a/pandas/pandas/_testing/_io.py b/pandas/pandas/_testing/_io.py
new file mode 100644
index 0000000000000..e1841c95dcdfe
--- /dev/null
+++ b/pandas/pandas/_testing/_io.py
@@ -0,0 +1,142 @@
+from __future__ import annotations
+
+import gzip
+import io
+import pathlib
+import tarfile
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import uuid
+import zipfile
+
+from pandas.compat._optional import import_optional_dependency
+
+import pandas as pd
+from pandas._testing.contexts import ensure_clean
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        FilePath,
+        ReadPickleBuffer,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+# ------------------------------------------------------------------
+# File-IO
+
+
+def round_trip_pickle(
+    obj: Any, path: FilePath | ReadPickleBuffer | None = None
+) -> DataFrame | Series:
+    """
+    Pickle an object and then read it again.
+
+    Parameters
+    ----------
+    obj : any object
+        The object to pickle and then re-read.
+    path : str, path object or file-like object, default None
+        The path where the pickled object is written and then read.
+
+    Returns
+    -------
+    pandas object
+        The original object that was pickled and then re-read.
+    """
+    _path = path
+    if _path is None:
+        _path = f"__{uuid.uuid4()}__.pickle"
+    with ensure_clean(_path) as temp_path:
+        pd.to_pickle(obj, temp_path)
+        return pd.read_pickle(temp_path)
+
+
+def round_trip_pathlib(writer, reader, path: str | None = None):
+    """
+    Write an object to file specified by a pathlib.Path and read it back
+
+    Parameters
+    ----------
+    writer : callable bound to pandas object
+        IO writing function (e.g. DataFrame.to_csv )
+    reader : callable
+        IO reading function (e.g. pd.read_csv )
+    path : str, default None
+        The path where the object is written and then read.
+
+    Returns
+    -------
+    pandas object
+        The original object that was serialized and then re-read.
+    """
+    Path = pathlib.Path
+    if path is None:
+        path = "___pathlib___"
+    with ensure_clean(path) as path:
+        writer(Path(path))  # type: ignore[arg-type]
+        obj = reader(Path(path))  # type: ignore[arg-type]
+    return obj
+
+
+def write_to_compressed(compression, path, data, dest: str = "test") -> None:
+    """
+    Write data to a compressed file.
+
+    Parameters
+    ----------
+    compression : {'gzip', 'bz2', 'zip', 'xz', 'zstd'}
+        The compression type to use.
+    path : str
+        The file path to write the data.
+    data : str
+        The data to write.
+    dest : str, default "test"
+        The destination file (for ZIP only)
+
+    Raises
+    ------
+    ValueError : An invalid compression value was passed in.
+    """
+    args: tuple[Any, ...] = (data,)
+    mode = "wb"
+    method = "write"
+    compress_method: Callable
+
+    if compression == "zip":
+        compress_method = zipfile.ZipFile
+        mode = "w"
+        args = (dest, data)
+        method = "writestr"
+    elif compression == "tar":
+        compress_method = tarfile.TarFile
+        mode = "w"
+        file = tarfile.TarInfo(name=dest)
+        bytes = io.BytesIO(data)
+        file.size = len(data)
+        args = (file, bytes)
+        method = "addfile"
+    elif compression == "gzip":
+        compress_method = gzip.GzipFile
+    elif compression == "bz2":
+        import bz2
+
+        compress_method = bz2.BZ2File
+    elif compression == "zstd":
+        compress_method = import_optional_dependency("zstandard").open
+    elif compression == "xz":
+        import lzma
+
+        compress_method = lzma.LZMAFile
+    else:
+        raise ValueError(f"Unrecognized compression type: {compression}")
+
+    with compress_method(path, mode=mode) as f:
+        getattr(f, method)(*args)
diff --git a/pandas/pandas/_testing/_warnings.py b/pandas/pandas/_testing/_warnings.py
new file mode 100644
index 0000000000000..a752c8db90f38
--- /dev/null
+++ b/pandas/pandas/_testing/_warnings.py
@@ -0,0 +1,266 @@
+from __future__ import annotations
+
+from contextlib import (
+    AbstractContextManager,
+    contextmanager,
+    nullcontext,
+)
+import inspect
+import re
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    Union,
+    cast,
+)
+import warnings
+
+from pandas.compat import PY311
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Sequence,
+    )
+
+
+@contextmanager
+def assert_produces_warning(
+    expected_warning: type[Warning] | bool | tuple[type[Warning], ...] | None = Warning,
+    filter_level: Literal[
+        "error", "ignore", "always", "default", "module", "once"
+    ] = "always",
+    check_stacklevel: bool = True,
+    raise_on_extra_warnings: bool = True,
+    match: str | tuple[str | None, ...] | None = None,
+    must_find_all_warnings: bool = True,
+) -> Generator[list[warnings.WarningMessage]]:
+    """
+    Context manager for running code expected to either raise a specific warning,
+    multiple specific warnings, or not raise any warnings. Verifies that the code
+    raises the expected warning(s), and that it does not raise any other unexpected
+    warnings. It is basically a wrapper around ``warnings.catch_warnings``.
+
+    Parameters
+    ----------
+    expected_warning : {Warning, False, tuple[Warning, ...], None}, default Warning
+        The type of Exception raised. ``exception.Warning`` is the base
+        class for all warnings. To raise multiple types of exceptions,
+        pass them as a tuple. To check that no warning is returned,
+        specify ``False`` or ``None``.
+    filter_level : str or None, default "always"
+        Specifies whether warnings are ignored, displayed, or turned
+        into errors.
+        Valid values are:
+
+        * "error" - turns matching warnings into exceptions
+        * "ignore" - discard the warning
+        * "always" - always emit a warning
+        * "default" - print the warning the first time it is generated
+          from each location
+        * "module" - print the warning the first time it is generated
+          from each module
+        * "once" - print the warning the first time it is generated
+
+    check_stacklevel : bool, default True
+        If True, displays the line that called the function containing
+        the warning to show were the function is called. Otherwise, the
+        line that implements the function is displayed.
+    raise_on_extra_warnings : bool, default True
+        Whether extra warnings not of the type `expected_warning` should
+        cause the test to fail.
+    match : {str, tuple[str, ...]}, optional
+        Match warning message. If it's a tuple, it has to be the size of
+        `expected_warning`. If additionally `must_find_all_warnings` is
+        True, each expected warning's message gets matched with a respective
+        match. Otherwise, multiple values get treated as an alternative.
+    must_find_all_warnings : bool, default True
+        If True and `expected_warning` is a tuple, each expected warning
+        type must get encountered. Otherwise, even one expected warning
+        results in success.
+
+    Examples
+    --------
+    >>> import warnings
+    >>> with assert_produces_warning():
+    ...     warnings.warn(UserWarning())
+    >>> with assert_produces_warning(False):
+    ...     warnings.warn(RuntimeWarning())
+    Traceback (most recent call last):
+        ...
+    AssertionError: Caused unexpected warning(s): ['RuntimeWarning'].
+    >>> with assert_produces_warning(UserWarning):
+    ...     warnings.warn(RuntimeWarning())
+    Traceback (most recent call last):
+        ...
+    AssertionError: Did not see expected warning of class 'UserWarning'.
+
+    ..warn:: This is *not* thread-safe.
+    """
+    __tracebackhide__ = True
+
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter(filter_level)
+        try:
+            yield w
+        finally:
+            if expected_warning:
+                if isinstance(expected_warning, tuple) and must_find_all_warnings:
+                    match = (
+                        match
+                        if isinstance(match, tuple)
+                        else (match,) * len(expected_warning)
+                    )
+                    for warning_type, warning_match in zip(expected_warning, match):
+                        _assert_caught_expected_warnings(
+                            caught_warnings=w,
+                            expected_warning=warning_type,
+                            match=warning_match,
+                            check_stacklevel=check_stacklevel,
+                        )
+                else:
+                    expected_warning = cast(
+                        Union[type[Warning], tuple[type[Warning], ...]],
+                        expected_warning,
+                    )
+                    match = (
+                        "|".join(m for m in match if m)
+                        if isinstance(match, tuple)
+                        else match
+                    )
+                    _assert_caught_expected_warnings(
+                        caught_warnings=w,
+                        expected_warning=expected_warning,
+                        match=match,
+                        check_stacklevel=check_stacklevel,
+                    )
+            if raise_on_extra_warnings:
+                _assert_caught_no_extra_warnings(
+                    caught_warnings=w,
+                    expected_warning=expected_warning,
+                )
+
+
+def maybe_produces_warning(
+    warning: type[Warning], condition: bool, **kwargs
+) -> AbstractContextManager:
+    """
+    Return a context manager that possibly checks a warning based on the condition
+    """
+    if condition:
+        return assert_produces_warning(warning, **kwargs)
+    else:
+        return nullcontext()
+
+
+def _assert_caught_expected_warnings(
+    *,
+    caught_warnings: Sequence[warnings.WarningMessage],
+    expected_warning: type[Warning] | tuple[type[Warning], ...],
+    match: str | None,
+    check_stacklevel: bool,
+) -> None:
+    """Assert that there was the expected warning among the caught warnings."""
+    saw_warning = False
+    matched_message = False
+    unmatched_messages = []
+    warning_name = (
+        tuple(x.__name__ for x in expected_warning)
+        if isinstance(expected_warning, tuple)
+        else expected_warning.__name__
+    )
+
+    for actual_warning in caught_warnings:
+        if issubclass(actual_warning.category, expected_warning):
+            saw_warning = True
+
+            if check_stacklevel:
+                _assert_raised_with_correct_stacklevel(actual_warning)
+
+            if match is not None:
+                if re.search(match, str(actual_warning.message)):
+                    matched_message = True
+                else:
+                    unmatched_messages.append(actual_warning.message)
+
+    if not saw_warning:
+        raise AssertionError(f"Did not see expected warning of class {warning_name!r}")
+
+    if match and not matched_message:
+        raise AssertionError(
+            f"Did not see warning {warning_name!r} "
+            f"matching '{match}'. The emitted warning messages are "
+            f"{unmatched_messages}"
+        )
+
+
+def _assert_caught_no_extra_warnings(
+    *,
+    caught_warnings: Sequence[warnings.WarningMessage],
+    expected_warning: type[Warning] | bool | tuple[type[Warning], ...] | None,
+) -> None:
+    """Assert that no extra warnings apart from the expected ones are caught."""
+    extra_warnings = []
+
+    for actual_warning in caught_warnings:
+        if _is_unexpected_warning(actual_warning, expected_warning):
+            # GH#38630 pytest.filterwarnings does not suppress these.
+            if actual_warning.category == ResourceWarning:
+                # GH 44732: Don't make the CI flaky by filtering SSL-related
+                # ResourceWarning from dependencies
+                if "unclosed <ssl.SSLSocket" in str(actual_warning.message):
+                    continue
+                # GH 44844: Matplotlib leaves font files open during the entire process
+                # upon import. Don't make CI flaky if ResourceWarning raised
+                # due to these open files.
+                if any("matplotlib" in mod for mod in sys.modules):
+                    continue
+            if PY311 and actual_warning.category == EncodingWarning:
+                # EncodingWarnings are checked in the CI
+                # pyproject.toml errors on EncodingWarnings in pandas
+                # Ignore EncodingWarnings from other libraries
+                continue
+            extra_warnings.append(
+                (
+                    actual_warning.category.__name__,
+                    actual_warning.message,
+                    actual_warning.filename,
+                    actual_warning.lineno,
+                )
+            )
+
+    if extra_warnings:
+        raise AssertionError(f"Caused unexpected warning(s): {extra_warnings!r}")
+
+
+def _is_unexpected_warning(
+    actual_warning: warnings.WarningMessage,
+    expected_warning: type[Warning] | bool | tuple[type[Warning], ...] | None,
+) -> bool:
+    """Check if the actual warning issued is unexpected."""
+    if actual_warning and not expected_warning:
+        return True
+    expected_warning = cast(type[Warning], expected_warning)
+    return bool(not issubclass(actual_warning.category, expected_warning))
+
+
+def _assert_raised_with_correct_stacklevel(
+    actual_warning: warnings.WarningMessage,
+) -> None:
+    # https://stackoverflow.com/questions/17407119/python-inspect-stack-is-slow
+    frame = inspect.currentframe()
+    for _ in range(4):
+        frame = frame.f_back  # type: ignore[union-attr]
+    try:
+        caller_filename = inspect.getfile(frame)  # type: ignore[arg-type]
+    finally:
+        # See note in
+        # https://docs.python.org/3/library/inspect.html#inspect.Traceback
+        del frame
+    msg = (
+        "Warning not set with correct stacklevel. "
+        f"File where warning is raised: {actual_warning.filename} != "
+        f"{caller_filename}. Warning message: {actual_warning.message}"
+    )
+    assert actual_warning.filename == caller_filename, msg
diff --git a/pandas/pandas/_testing/asserters.py b/pandas/pandas/_testing/asserters.py
new file mode 100644
index 0000000000000..daa5187cdb636
--- /dev/null
+++ b/pandas/pandas/_testing/asserters.py
@@ -0,0 +1,1479 @@
+from __future__ import annotations
+
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    NoReturn,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.missing import is_matching_na
+from pandas._libs.sparse import SparseIndex
+import pandas._libs.testing as _testing
+from pandas._libs.tslibs.np_datetime import compare_mismatched_resolutions
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_float_dtype,
+    is_integer_dtype,
+    is_number,
+    is_numeric_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    NumpyEADtype,
+)
+from pandas.core.dtypes.missing import array_equivalent
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IntervalDtype,
+    IntervalIndex,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    TimedeltaIndex,
+)
+from pandas.core.arrays import (
+    DatetimeArray,
+    ExtensionArray,
+    IntervalArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.datetimelike import DatetimeLikeArrayMixin
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.indexes.api import safe_sort_index
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from pandas._typing import DtypeObj
+
+
+def assert_almost_equal(
+    left,
+    right,
+    check_dtype: bool | Literal["equiv"] = "equiv",
+    rtol: float = 1.0e-5,
+    atol: float = 1.0e-8,
+    **kwargs,
+) -> None:
+    """
+    Check that the left and right objects are approximately equal.
+
+    By approximately equal, we refer to objects that are numbers or that
+    contain numbers which may be equivalent to specific levels of precision.
+
+    Parameters
+    ----------
+    left : object
+    right : object
+    check_dtype : bool or {'equiv'}, default 'equiv'
+        Check dtype if both a and b are the same type. If 'equiv' is passed in,
+        then `RangeIndex` and `Index` with int64 dtype are also considered
+        equivalent when doing type checking.
+    rtol : float, default 1e-5
+        Relative tolerance.
+    atol : float, default 1e-8
+        Absolute tolerance.
+    """
+    if isinstance(left, Index):
+        assert_index_equal(
+            left,
+            right,
+            check_exact=False,
+            exact=check_dtype,
+            rtol=rtol,
+            atol=atol,
+            **kwargs,
+        )
+
+    elif isinstance(left, Series):
+        assert_series_equal(
+            left,
+            right,
+            check_exact=False,
+            check_dtype=check_dtype,
+            rtol=rtol,
+            atol=atol,
+            **kwargs,
+        )
+
+    elif isinstance(left, DataFrame):
+        assert_frame_equal(
+            left,
+            right,
+            check_exact=False,
+            check_dtype=check_dtype,
+            rtol=rtol,
+            atol=atol,
+            **kwargs,
+        )
+
+    else:
+        # Other sequences.
+        if check_dtype:
+            if is_number(left) and is_number(right):
+                # Do not compare numeric classes, like np.float64 and float.
+                pass
+            elif is_bool(left) and is_bool(right):
+                # Do not compare bool classes, like np.bool_ and bool.
+                pass
+            else:
+                if isinstance(left, np.ndarray) or isinstance(right, np.ndarray):
+                    obj = "numpy array"
+                else:
+                    obj = "Input"
+                assert_class_equal(left, right, obj=obj)
+
+        # if we have "equiv", this becomes True
+        _testing.assert_almost_equal(
+            left, right, check_dtype=bool(check_dtype), rtol=rtol, atol=atol, **kwargs
+        )
+
+
+def _check_isinstance(left, right, cls) -> None:
+    """
+    Helper method for our assert_* methods that ensures that
+    the two objects being compared have the right type before
+    proceeding with the comparison.
+
+    Parameters
+    ----------
+    left : The first object being compared.
+    right : The second object being compared.
+    cls : The class type to check against.
+
+    Raises
+    ------
+    AssertionError : Either `left` or `right` is not an instance of `cls`.
+    """
+    cls_name = cls.__name__
+
+    if not isinstance(left, cls):
+        raise AssertionError(
+            f"{cls_name} Expected type {cls}, found {type(left)} instead"
+        )
+    if not isinstance(right, cls):
+        raise AssertionError(
+            f"{cls_name} Expected type {cls}, found {type(right)} instead"
+        )
+
+
+def assert_dict_equal(left, right, compare_keys: bool = True) -> None:
+    _check_isinstance(left, right, dict)
+    _testing.assert_dict_equal(left, right, compare_keys=compare_keys)
+
+
+def assert_index_equal(
+    left: Index,
+    right: Index,
+    exact: bool | str = "equiv",
+    check_names: bool = True,
+    check_exact: bool = True,
+    check_categorical: bool = True,
+    check_order: bool = True,
+    rtol: float = 1.0e-5,
+    atol: float = 1.0e-8,
+    obj: str | None = None,
+) -> None:
+    """
+    Check that left and right Index are equal.
+
+    Parameters
+    ----------
+    left : Index
+        The first index to compare.
+    right : Index
+        The second index to compare.
+    exact : bool or {'equiv'}, default 'equiv'
+        Whether to check the Index class, dtype and inferred_type
+        are identical. If 'equiv', then RangeIndex can be substituted for
+        Index with an int64 dtype as well.
+    check_names : bool, default True
+        Whether to check the names attribute.
+    check_exact : bool, default True
+        Whether to compare number exactly.
+    check_categorical : bool, default True
+        Whether to compare internal Categorical exactly.
+    check_order : bool, default True
+        Whether to compare the order of index entries as well as their values.
+        If True, both indexes must contain the same elements, in the same order.
+        If False, both indexes must contain the same elements, but in any order.
+    rtol : float, default 1e-5
+        Relative tolerance. Only used when check_exact is False.
+    atol : float, default 1e-8
+        Absolute tolerance. Only used when check_exact is False.
+    obj : str, default 'Index' or 'MultiIndex'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+
+    See Also
+    --------
+    testing.assert_series_equal : Check that two Series are equal.
+    testing.assert_frame_equal : Check that two DataFrames are equal.
+
+    Examples
+    --------
+    >>> from pandas import testing as tm
+    >>> a = pd.Index([1, 2, 3])
+    >>> b = pd.Index([1, 2, 3])
+    >>> tm.assert_index_equal(a, b)
+    """
+    __tracebackhide__ = True
+
+    if obj is None:
+        obj = "MultiIndex" if isinstance(left, MultiIndex) else "Index"
+
+    def _check_types(left, right, obj: str = "Index") -> None:
+        if not exact:
+            return
+
+        assert_class_equal(left, right, exact=exact, obj=obj)
+        assert_attr_equal("inferred_type", left, right, obj=obj)
+
+        # Skip exact dtype checking when `check_categorical` is False
+        if isinstance(left.dtype, CategoricalDtype) and isinstance(
+            right.dtype, CategoricalDtype
+        ):
+            if check_categorical:
+                assert_attr_equal("dtype", left, right, obj=obj)
+                assert_index_equal(left.categories, right.categories, exact=exact)
+            return
+
+        assert_attr_equal("dtype", left, right, obj=obj)
+
+    # instance validation
+    _check_isinstance(left, right, Index)
+
+    # class / dtype comparison
+    _check_types(left, right, obj=obj)
+
+    # level comparison
+    if left.nlevels != right.nlevels:
+        msg1 = f"{obj} levels are different"
+        msg2 = f"{left.nlevels}, {left}"
+        msg3 = f"{right.nlevels}, {right}"
+        raise_assert_detail(obj, msg1, msg2, msg3)
+
+    # length comparison
+    if len(left) != len(right):
+        msg1 = f"{obj} length are different"
+        msg2 = f"{len(left)}, {left}"
+        msg3 = f"{len(right)}, {right}"
+        raise_assert_detail(obj, msg1, msg2, msg3)
+
+    # If order doesn't matter then sort the index entries
+    if not check_order:
+        left = safe_sort_index(left)
+        right = safe_sort_index(right)
+
+    # MultiIndex special comparison for little-friendly error messages
+    if isinstance(left, MultiIndex):
+        right = cast(MultiIndex, right)
+
+        for level in range(left.nlevels):
+            lobj = f"{obj} level [{level}]"
+            try:
+                # try comparison on levels/codes to avoid densifying MultiIndex
+                assert_index_equal(
+                    left.levels[level],
+                    right.levels[level],
+                    exact=exact,
+                    check_names=check_names,
+                    check_exact=check_exact,
+                    check_categorical=check_categorical,
+                    rtol=rtol,
+                    atol=atol,
+                    obj=lobj,
+                )
+                assert_numpy_array_equal(left.codes[level], right.codes[level])
+            except AssertionError:
+                llevel = left.get_level_values(level)
+                rlevel = right.get_level_values(level)
+
+                assert_index_equal(
+                    llevel,
+                    rlevel,
+                    exact=exact,
+                    check_names=check_names,
+                    check_exact=check_exact,
+                    check_categorical=check_categorical,
+                    rtol=rtol,
+                    atol=atol,
+                    obj=lobj,
+                )
+            # get_level_values may change dtype
+            _check_types(left.levels[level], right.levels[level], obj=lobj)
+
+    # skip exact index checking when `check_categorical` is False
+    elif check_exact and check_categorical:
+        if not left.equals(right):
+            mismatch = left._values != right._values
+
+            if not isinstance(mismatch, np.ndarray):
+                mismatch = cast("ExtensionArray", mismatch).fillna(True)
+
+            diff = np.sum(mismatch.astype(int)) * 100.0 / len(left)
+            msg = f"{obj} values are different ({np.round(diff, 5)} %)"
+            raise_assert_detail(obj, msg, left, right)
+    else:
+        # if we have "equiv", this becomes True
+        exact_bool = bool(exact)
+        _testing.assert_almost_equal(
+            left.values,
+            right.values,
+            rtol=rtol,
+            atol=atol,
+            check_dtype=exact_bool,
+            obj=obj,
+            lobj=left,
+            robj=right,
+        )
+
+    # metadata comparison
+    if check_names:
+        assert_attr_equal("names", left, right, obj=obj)
+    if isinstance(left, PeriodIndex) or isinstance(right, PeriodIndex):
+        assert_attr_equal("dtype", left, right, obj=obj)
+    if isinstance(left, IntervalIndex) or isinstance(right, IntervalIndex):
+        assert_interval_array_equal(left._values, right._values)
+
+    if check_categorical:
+        if isinstance(left.dtype, CategoricalDtype) or isinstance(
+            right.dtype, CategoricalDtype
+        ):
+            assert_categorical_equal(left._values, right._values, obj=f"{obj} category")
+
+
+def assert_class_equal(
+    left, right, exact: bool | str = True, obj: str = "Input"
+) -> None:
+    """
+    Checks classes are equal.
+    """
+    __tracebackhide__ = True
+
+    def repr_class(x):
+        if isinstance(x, Index):
+            # return Index as it is to include values in the error message
+            return x
+
+        return type(x).__name__
+
+    def is_class_equiv(idx: Index) -> bool:
+        """Classes that are a RangeIndex (sub-)instance or exactly an `Index` .
+
+        This only checks class equivalence. There is a separate check that the
+        dtype is int64.
+        """
+        return type(idx) is Index or isinstance(idx, RangeIndex)
+
+    if type(left) == type(right):
+        return
+
+    if exact == "equiv":
+        if is_class_equiv(left) and is_class_equiv(right):
+            return
+
+    msg = f"{obj} classes are different"
+    raise_assert_detail(obj, msg, repr_class(left), repr_class(right))
+
+
+def assert_attr_equal(attr: str, left, right, obj: str = "Attributes") -> None:
+    """
+    Check attributes are equal. Both objects must have attribute.
+
+    Parameters
+    ----------
+    attr : str
+        Attribute name being compared.
+    left : object
+    right : object
+    obj : str, default 'Attributes'
+        Specify object name being compared, internally used to show appropriate
+        assertion message
+    """
+    __tracebackhide__ = True
+
+    left_attr = getattr(left, attr)
+    right_attr = getattr(right, attr)
+
+    if left_attr is right_attr or is_matching_na(left_attr, right_attr):
+        # e.g. both np.nan, both NaT, both pd.NA, ...
+        return None
+
+    try:
+        result = left_attr == right_attr
+    except TypeError:
+        # datetimetz on rhs may raise TypeError
+        result = False
+    if (left_attr is pd.NA) ^ (right_attr is pd.NA):
+        result = False
+    elif not isinstance(result, bool):
+        result = result.all()
+
+    if not result:
+        msg = f'Attribute "{attr}" are different'
+        raise_assert_detail(obj, msg, left_attr, right_attr)
+    return None
+
+
+def assert_is_sorted(seq) -> None:
+    """Assert that the sequence is sorted."""
+    if isinstance(seq, (Index, Series)):
+        seq = seq.values
+    # sorting does not change precisions
+    if isinstance(seq, np.ndarray):
+        assert_numpy_array_equal(seq, np.sort(np.array(seq)))
+    else:
+        assert_extension_array_equal(seq, seq[seq.argsort()])
+
+
+def assert_categorical_equal(
+    left,
+    right,
+    check_dtype: bool = True,
+    check_category_order: bool = True,
+    obj: str = "Categorical",
+) -> None:
+    """
+    Test that Categoricals are equivalent.
+
+    Parameters
+    ----------
+    left : Categorical
+    right : Categorical
+    check_dtype : bool, default True
+        Check that integer dtype of the codes are the same.
+    check_category_order : bool, default True
+        Whether the order of the categories should be compared, which
+        implies identical integer codes.  If False, only the resulting
+        values are compared.  The ordered attribute is
+        checked regardless.
+    obj : str, default 'Categorical'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+    """
+    _check_isinstance(left, right, Categorical)
+
+    exact: bool | str
+    if isinstance(left.categories, RangeIndex) or isinstance(
+        right.categories, RangeIndex
+    ):
+        exact = "equiv"
+    else:
+        # We still want to require exact matches for Index
+        exact = True
+
+    if check_category_order:
+        assert_index_equal(
+            left.categories, right.categories, obj=f"{obj}.categories", exact=exact
+        )
+        assert_numpy_array_equal(
+            left.codes, right.codes, check_dtype=check_dtype, obj=f"{obj}.codes"
+        )
+    else:
+        try:
+            lc = left.categories.sort_values()
+            rc = right.categories.sort_values()
+        except TypeError:
+            # e.g. '<' not supported between instances of 'int' and 'str'
+            lc, rc = left.categories, right.categories
+        assert_index_equal(lc, rc, obj=f"{obj}.categories", exact=exact)
+        assert_index_equal(
+            left.categories.take(left.codes),
+            right.categories.take(right.codes),
+            obj=f"{obj}.values",
+            exact=exact,
+        )
+
+    assert_attr_equal("ordered", left, right, obj=obj)
+
+
+def assert_interval_array_equal(
+    left, right, exact: bool | Literal["equiv"] = "equiv", obj: str = "IntervalArray"
+) -> None:
+    """
+    Test that two IntervalArrays are equivalent.
+
+    Parameters
+    ----------
+    left, right : IntervalArray
+        The IntervalArrays to compare.
+    exact : bool or {'equiv'}, default 'equiv'
+        Whether to check the Index class, dtype and inferred_type
+        are identical. If 'equiv', then RangeIndex can be substituted for
+        Index with an int64 dtype as well.
+    obj : str, default 'IntervalArray'
+        Specify object name being compared, internally used to show appropriate
+        assertion message
+    """
+    _check_isinstance(left, right, IntervalArray)
+
+    kwargs = {}
+    if left._left.dtype.kind in "mM":
+        # We have a DatetimeArray or TimedeltaArray
+        kwargs["check_freq"] = False
+
+    assert_equal(left._left, right._left, obj=f"{obj}.left", **kwargs)
+    assert_equal(left._right, right._right, obj=f"{obj}.right", **kwargs)
+
+    assert_attr_equal("closed", left, right, obj=obj)
+
+
+def assert_period_array_equal(left, right, obj: str = "PeriodArray") -> None:
+    _check_isinstance(left, right, PeriodArray)
+
+    assert_numpy_array_equal(left._ndarray, right._ndarray, obj=f"{obj}._ndarray")
+    assert_attr_equal("dtype", left, right, obj=obj)
+
+
+def assert_datetime_array_equal(
+    left, right, obj: str = "DatetimeArray", check_freq: bool = True
+) -> None:
+    __tracebackhide__ = True
+    _check_isinstance(left, right, DatetimeArray)
+
+    assert_numpy_array_equal(left._ndarray, right._ndarray, obj=f"{obj}._ndarray")
+    if check_freq:
+        assert_attr_equal("freq", left, right, obj=obj)
+    assert_attr_equal("tz", left, right, obj=obj)
+
+
+def assert_timedelta_array_equal(
+    left, right, obj: str = "TimedeltaArray", check_freq: bool = True
+) -> None:
+    __tracebackhide__ = True
+    _check_isinstance(left, right, TimedeltaArray)
+    assert_numpy_array_equal(left._ndarray, right._ndarray, obj=f"{obj}._ndarray")
+    if check_freq:
+        assert_attr_equal("freq", left, right, obj=obj)
+
+
+def raise_assert_detail(
+    obj, message, left, right, diff=None, first_diff=None, index_values=None
+) -> NoReturn:
+    __tracebackhide__ = True
+
+    msg = f"""{obj} are different
+
+{message}"""
+
+    if isinstance(index_values, Index):
+        index_values = np.asarray(index_values)
+
+    if isinstance(index_values, np.ndarray):
+        msg += f"\n[index]: {pprint_thing(index_values)}"
+
+    if isinstance(left, np.ndarray):
+        left = pprint_thing(left)
+    elif isinstance(left, (CategoricalDtype, NumpyEADtype)):
+        left = repr(left)
+    elif isinstance(left, StringDtype):
+        # TODO(infer_string) this special case could be avoided if we have
+        # a more informative repr https://github.com/pandas-dev/pandas/issues/59342
+        left = f"StringDtype(storage={left.storage}, na_value={left.na_value})"
+
+    if isinstance(right, np.ndarray):
+        right = pprint_thing(right)
+    elif isinstance(right, (CategoricalDtype, NumpyEADtype)):
+        right = repr(right)
+    elif isinstance(right, StringDtype):
+        right = f"StringDtype(storage={right.storage}, na_value={right.na_value})"
+
+    msg += f"""
+[left]:  {left}
+[right]: {right}"""
+
+    if diff is not None:
+        msg += f"\n[diff]: {diff}"
+
+    if first_diff is not None:
+        msg += f"\n{first_diff}"
+
+    raise AssertionError(msg)
+
+
+def assert_numpy_array_equal(
+    left,
+    right,
+    strict_nan: bool = False,
+    check_dtype: bool | Literal["equiv"] = True,
+    err_msg=None,
+    check_same=None,
+    obj: str = "numpy array",
+    index_values=None,
+) -> None:
+    """
+    Check that 'np.ndarray' is equivalent.
+
+    Parameters
+    ----------
+    left, right : numpy.ndarray or iterable
+        The two arrays to be compared.
+    strict_nan : bool, default False
+        If True, consider NaN and None to be different.
+    check_dtype : bool, default True
+        Check dtype if both a and b are np.ndarray.
+    err_msg : str, default None
+        If provided, used as assertion message.
+    check_same : None|'copy'|'same', default None
+        Ensure left and right refer/do not refer to the same memory area.
+    obj : str, default 'numpy array'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+    index_values : Index | numpy.ndarray, default None
+        optional index (shared by both left and right), used in output.
+    """
+    __tracebackhide__ = True
+
+    # instance validation
+    # Show a detailed error message when classes are different
+    assert_class_equal(left, right, obj=obj)
+    # both classes must be an np.ndarray
+    _check_isinstance(left, right, np.ndarray)
+
+    def _get_base(obj):
+        return obj.base if getattr(obj, "base", None) is not None else obj
+
+    left_base = _get_base(left)
+    right_base = _get_base(right)
+
+    if check_same == "same":
+        if left_base is not right_base:
+            raise AssertionError(f"{left_base!r} is not {right_base!r}")
+    elif check_same == "copy":
+        if left_base is right_base:
+            raise AssertionError(f"{left_base!r} is {right_base!r}")
+
+    def _raise(left, right, err_msg) -> NoReturn:
+        if err_msg is None:
+            if left.shape != right.shape:
+                raise_assert_detail(
+                    obj, f"{obj} shapes are different", left.shape, right.shape
+                )
+
+            diff = 0
+            for left_arr, right_arr in zip(left, right):
+                # count up differences
+                if not array_equivalent(left_arr, right_arr, strict_nan=strict_nan):
+                    diff += 1
+
+            diff = diff * 100.0 / left.size
+            msg = f"{obj} values are different ({np.round(diff, 5)} %)"
+            raise_assert_detail(obj, msg, left, right, index_values=index_values)
+
+        raise AssertionError(err_msg)
+
+    # compare shape and values
+    if not array_equivalent(left, right, strict_nan=strict_nan):
+        _raise(left, right, err_msg)
+
+    if check_dtype:
+        if isinstance(left, np.ndarray) and isinstance(right, np.ndarray):
+            assert_attr_equal("dtype", left, right, obj=obj)
+
+
+def assert_extension_array_equal(
+    left,
+    right,
+    check_dtype: bool | Literal["equiv"] = True,
+    index_values=None,
+    check_exact: bool | lib.NoDefault = lib.no_default,
+    rtol: float | lib.NoDefault = lib.no_default,
+    atol: float | lib.NoDefault = lib.no_default,
+    obj: str = "ExtensionArray",
+) -> None:
+    """
+    Check that left and right ExtensionArrays are equal.
+
+    This method compares two ``ExtensionArray`` instances for equality,
+    including checks for missing values, the dtype of the arrays, and
+    the exactness of the comparison (or tolerance when comparing floats).
+
+    Parameters
+    ----------
+    left, right : ExtensionArray
+        The two arrays to compare.
+    check_dtype : bool, default True
+        Whether to check if the ExtensionArray dtypes are identical.
+    index_values : Index | numpy.ndarray, default None
+        Optional index (shared by both left and right), used in output.
+    check_exact : bool, default False
+        Whether to compare number exactly.
+
+        .. versionchanged:: 2.2.0
+
+            Defaults to True for integer dtypes if none of
+            ``check_exact``, ``rtol`` and ``atol`` are specified.
+    rtol : float, default 1e-5
+        Relative tolerance. Only used when check_exact is False.
+    atol : float, default 1e-8
+        Absolute tolerance. Only used when check_exact is False.
+    obj : str, default 'ExtensionArray'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+
+        .. versionadded:: 2.0.0
+
+    See Also
+    --------
+    testing.assert_series_equal : Check that left and right ``Series`` are equal.
+    testing.assert_frame_equal : Check that left and right ``DataFrame`` are equal.
+    testing.assert_index_equal : Check that left and right ``Index`` are equal.
+
+    Notes
+    -----
+    Missing values are checked separately from valid values.
+    A mask of missing values is computed for each and checked to match.
+    The remaining all-valid values are cast to object dtype and checked.
+
+    Examples
+    --------
+    >>> from pandas import testing as tm
+    >>> a = pd.Series([1, 2, 3, 4])
+    >>> b, c = a.array, a.array
+    >>> tm.assert_extension_array_equal(b, c)
+    """
+    if (
+        check_exact is lib.no_default
+        and rtol is lib.no_default
+        and atol is lib.no_default
+    ):
+        check_exact = (
+            is_numeric_dtype(left.dtype) and not is_float_dtype(left.dtype)
+        ) or (is_numeric_dtype(right.dtype) and not is_float_dtype(right.dtype))
+    elif check_exact is lib.no_default:
+        check_exact = False
+
+    rtol = rtol if rtol is not lib.no_default else 1.0e-5
+    atol = atol if atol is not lib.no_default else 1.0e-8
+
+    assert isinstance(left, ExtensionArray), "left is not an ExtensionArray"
+    assert isinstance(right, ExtensionArray), "right is not an ExtensionArray"
+    if check_dtype:
+        assert_attr_equal("dtype", left, right, obj=f"Attributes of {obj}")
+
+    if (
+        isinstance(left, DatetimeLikeArrayMixin)
+        and isinstance(right, DatetimeLikeArrayMixin)
+        and type(right) == type(left)
+    ):
+        # GH 52449
+        if not check_dtype and left.dtype.kind in "mM":
+            if not isinstance(left.dtype, np.dtype):
+                l_unit = cast(DatetimeTZDtype, left.dtype).unit
+            else:
+                l_unit = np.datetime_data(left.dtype)[0]
+            if not isinstance(right.dtype, np.dtype):
+                r_unit = cast(DatetimeTZDtype, right.dtype).unit
+            else:
+                r_unit = np.datetime_data(right.dtype)[0]
+            if (
+                l_unit != r_unit
+                and compare_mismatched_resolutions(
+                    left._ndarray, right._ndarray, operator.eq
+                ).all()
+            ):
+                return
+        # Avoid slow object-dtype comparisons
+        # np.asarray for case where we have a np.MaskedArray
+        assert_numpy_array_equal(
+            np.asarray(left.asi8),
+            np.asarray(right.asi8),
+            index_values=index_values,
+            obj=obj,
+        )
+        return
+
+    left_na = np.asarray(left.isna())
+    right_na = np.asarray(right.isna())
+    assert_numpy_array_equal(
+        left_na, right_na, obj=f"{obj} NA mask", index_values=index_values
+    )
+
+    # Specifically for StringArrayNumpySemantics, validate here we have a valid array
+    if (
+        isinstance(left.dtype, StringDtype)
+        and left.dtype.storage == "python"
+        and left.dtype.na_value is np.nan
+    ):
+        assert np.all(
+            [np.isnan(val) for val in left._ndarray[left_na]]  # type: ignore[attr-defined]
+        ), "wrong missing value sentinels"
+    if (
+        isinstance(right.dtype, StringDtype)
+        and right.dtype.storage == "python"
+        and right.dtype.na_value is np.nan
+    ):
+        assert np.all(
+            [np.isnan(val) for val in right._ndarray[right_na]]  # type: ignore[attr-defined]
+        ), "wrong missing value sentinels"
+
+    left_valid = left[~left_na].to_numpy(dtype=object)
+    right_valid = right[~right_na].to_numpy(dtype=object)
+    if check_exact:
+        assert_numpy_array_equal(
+            left_valid, right_valid, obj=obj, index_values=index_values
+        )
+    else:
+        _testing.assert_almost_equal(
+            left_valid,
+            right_valid,
+            check_dtype=bool(check_dtype),
+            rtol=rtol,
+            atol=atol,
+            obj=obj,
+            index_values=index_values,
+        )
+
+
+# This could be refactored to use the NDFrame.equals method
+def assert_series_equal(
+    left,
+    right,
+    check_dtype: bool | Literal["equiv"] = True,
+    check_index_type: bool | Literal["equiv"] = "equiv",
+    check_series_type: bool = True,
+    check_names: bool = True,
+    check_exact: bool | lib.NoDefault = lib.no_default,
+    check_datetimelike_compat: bool = False,
+    check_categorical: bool = True,
+    check_category_order: bool = True,
+    check_freq: bool = True,
+    check_flags: bool = True,
+    rtol: float | lib.NoDefault = lib.no_default,
+    atol: float | lib.NoDefault = lib.no_default,
+    obj: str = "Series",
+    *,
+    check_index: bool = True,
+    check_like: bool = False,
+) -> None:
+    """
+    Check that left and right Series are equal.
+
+    Parameters
+    ----------
+    left : Series
+        First Series to compare.
+    right : Series
+        Second Series to compare.
+    check_dtype : bool, default True
+        Whether to check the Series dtype is identical.
+    check_index_type : bool or {'equiv'}, default 'equiv'
+        Whether to check the Index class, dtype and inferred_type
+        are identical.
+    check_series_type : bool, default True
+         Whether to check the Series class is identical.
+    check_names : bool, default True
+        Whether to check the Series and Index names attribute.
+    check_exact : bool, default False
+        Whether to compare number exactly. This also applies when checking
+        Index equivalence.
+
+        .. versionchanged:: 2.2.0
+
+            Defaults to True for integer dtypes if none of
+            ``check_exact``, ``rtol`` and ``atol`` are specified.
+
+        .. versionchanged:: 3.0.0
+
+            check_exact for comparing the Indexes defaults to True by
+            checking if an Index is of integer dtypes.
+
+    check_datetimelike_compat : bool, default False
+        Compare datetime-like which is comparable ignoring dtype.
+    check_categorical : bool, default True
+        Whether to compare internal Categorical exactly.
+    check_category_order : bool, default True
+        Whether to compare category order of internal Categoricals.
+    check_freq : bool, default True
+        Whether to check the `freq` attribute on a DatetimeIndex or TimedeltaIndex.
+    check_flags : bool, default True
+        Whether to check the `flags` attribute.
+    rtol : float, default 1e-5
+        Relative tolerance. Only used when check_exact is False.
+    atol : float, default 1e-8
+        Absolute tolerance. Only used when check_exact is False.
+    obj : str, default 'Series'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+    check_index : bool, default True
+        Whether to check index equivalence. If False, then compare only values.
+
+        .. versionadded:: 1.3.0
+    check_like : bool, default False
+        If True, ignore the order of the index. Must be False if check_index is False.
+        Note: same labels must be with the same data.
+
+        .. versionadded:: 1.5.0
+
+    See Also
+    --------
+    testing.assert_index_equal : Check that two Indexes are equal.
+    testing.assert_frame_equal : Check that two DataFrames are equal.
+
+    Examples
+    --------
+    >>> from pandas import testing as tm
+    >>> a = pd.Series([1, 2, 3, 4])
+    >>> b = pd.Series([1, 2, 3, 4])
+    >>> tm.assert_series_equal(a, b)
+    """
+    __tracebackhide__ = True
+    if (
+        check_exact is lib.no_default
+        and rtol is lib.no_default
+        and atol is lib.no_default
+    ):
+        check_exact = (
+            is_numeric_dtype(left.dtype) and not is_float_dtype(left.dtype)
+        ) or (is_numeric_dtype(right.dtype) and not is_float_dtype(right.dtype))
+        left_index_dtypes = (
+            [left.index.dtype] if left.index.nlevels == 1 else left.index.dtypes
+        )
+        right_index_dtypes = (
+            [right.index.dtype] if right.index.nlevels == 1 else right.index.dtypes
+        )
+        check_exact_index = all(
+            dtype.kind in "iu" for dtype in left_index_dtypes
+        ) or all(dtype.kind in "iu" for dtype in right_index_dtypes)
+    elif check_exact is lib.no_default:
+        check_exact = False
+        check_exact_index = False
+    else:
+        check_exact_index = check_exact
+
+    rtol = rtol if rtol is not lib.no_default else 1.0e-5
+    atol = atol if atol is not lib.no_default else 1.0e-8
+
+    if not check_index and check_like:
+        raise ValueError("check_like must be False if check_index is False")
+
+    # instance validation
+    _check_isinstance(left, right, Series)
+
+    if check_series_type:
+        assert_class_equal(left, right, obj=obj)
+
+    # length comparison
+    if len(left) != len(right):
+        msg1 = f"{len(left)}, {left.index}"
+        msg2 = f"{len(right)}, {right.index}"
+        raise_assert_detail(obj, "Series length are different", msg1, msg2)
+
+    if check_flags:
+        assert left.flags == right.flags, f"{left.flags!r} != {right.flags!r}"
+
+    if check_index:
+        # GH #38183
+        assert_index_equal(
+            left.index,
+            right.index,
+            exact=check_index_type,
+            check_names=check_names,
+            check_exact=check_exact_index,
+            check_categorical=check_categorical,
+            check_order=not check_like,
+            rtol=rtol,
+            atol=atol,
+            obj=f"{obj}.index",
+        )
+
+    if check_like:
+        left = left.reindex_like(right)
+
+    if check_freq and isinstance(left.index, (DatetimeIndex, TimedeltaIndex)):
+        lidx = left.index
+        ridx = right.index
+        assert lidx.freq == ridx.freq, (lidx.freq, ridx.freq)
+
+    if check_dtype:
+        # We want to skip exact dtype checking when `check_categorical`
+        # is False. We'll still raise if only one is a `Categorical`,
+        # regardless of `check_categorical`
+        if (
+            isinstance(left.dtype, CategoricalDtype)
+            and isinstance(right.dtype, CategoricalDtype)
+            and not check_categorical
+        ):
+            pass
+        else:
+            assert_attr_equal("dtype", left, right, obj=f"Attributes of {obj}")
+    if check_exact:
+        left_values = left._values
+        right_values = right._values
+        # Only check exact if dtype is numeric
+        if isinstance(left_values, ExtensionArray) and isinstance(
+            right_values, ExtensionArray
+        ):
+            assert_extension_array_equal(
+                left_values,
+                right_values,
+                check_dtype=check_dtype,
+                index_values=left.index,
+                obj=str(obj),
+            )
+        else:
+            # convert both to NumPy if not, check_dtype would raise earlier
+            lv, rv = left_values, right_values
+            if isinstance(left_values, ExtensionArray):
+                lv = left_values.to_numpy()
+            if isinstance(right_values, ExtensionArray):
+                rv = right_values.to_numpy()
+            assert_numpy_array_equal(
+                lv,
+                rv,
+                check_dtype=check_dtype,
+                obj=str(obj),
+                index_values=left.index,
+            )
+    elif check_datetimelike_compat and (
+        needs_i8_conversion(left.dtype) or needs_i8_conversion(right.dtype)
+    ):
+        # we want to check only if we have compat dtypes
+        # e.g. integer and M|m are NOT compat, but we can simply check
+        # the values in that case
+
+        # datetimelike may have different objects (e.g. datetime.datetime
+        # vs Timestamp) but will compare equal
+        if not Index(left._values).equals(Index(right._values)):
+            msg = (
+                f"[datetimelike_compat=True] {left._values} "
+                f"is not equal to {right._values}."
+            )
+            raise AssertionError(msg)
+    elif isinstance(left.dtype, IntervalDtype) and isinstance(
+        right.dtype, IntervalDtype
+    ):
+        assert_interval_array_equal(left.array, right.array)
+    elif isinstance(left.dtype, CategoricalDtype) or isinstance(
+        right.dtype, CategoricalDtype
+    ):
+        _testing.assert_almost_equal(
+            left._values,
+            right._values,
+            rtol=rtol,
+            atol=atol,
+            check_dtype=bool(check_dtype),
+            obj=str(obj),
+            index_values=left.index,
+        )
+    elif isinstance(left.dtype, ExtensionDtype) and isinstance(
+        right.dtype, ExtensionDtype
+    ):
+        assert_extension_array_equal(
+            left._values,
+            right._values,
+            rtol=rtol,
+            atol=atol,
+            check_dtype=check_dtype,
+            index_values=left.index,
+            obj=str(obj),
+        )
+    elif is_extension_array_dtype_and_needs_i8_conversion(
+        left.dtype, right.dtype
+    ) or is_extension_array_dtype_and_needs_i8_conversion(right.dtype, left.dtype):
+        assert_extension_array_equal(
+            left._values,
+            right._values,
+            check_dtype=check_dtype,
+            index_values=left.index,
+            obj=str(obj),
+        )
+    elif needs_i8_conversion(left.dtype) and needs_i8_conversion(right.dtype):
+        # DatetimeArray or TimedeltaArray
+        assert_extension_array_equal(
+            left._values,
+            right._values,
+            check_dtype=check_dtype,
+            index_values=left.index,
+            obj=str(obj),
+        )
+    else:
+        _testing.assert_almost_equal(
+            left._values,
+            right._values,
+            rtol=rtol,
+            atol=atol,
+            check_dtype=bool(check_dtype),
+            obj=str(obj),
+            index_values=left.index,
+        )
+
+    # metadata comparison
+    if check_names:
+        assert_attr_equal("name", left, right, obj=obj)
+
+    if check_categorical:
+        if isinstance(left.dtype, CategoricalDtype) or isinstance(
+            right.dtype, CategoricalDtype
+        ):
+            assert_categorical_equal(
+                left._values,
+                right._values,
+                obj=f"{obj} category",
+                check_category_order=check_category_order,
+            )
+
+
+# This could be refactored to use the NDFrame.equals method
+def assert_frame_equal(
+    left,
+    right,
+    check_dtype: bool | Literal["equiv"] = True,
+    check_index_type: bool | Literal["equiv"] = "equiv",
+    check_column_type: bool | Literal["equiv"] = "equiv",
+    check_frame_type: bool = True,
+    check_names: bool = True,
+    by_blocks: bool = False,
+    check_exact: bool | lib.NoDefault = lib.no_default,
+    check_datetimelike_compat: bool = False,
+    check_categorical: bool = True,
+    check_like: bool = False,
+    check_freq: bool = True,
+    check_flags: bool = True,
+    rtol: float | lib.NoDefault = lib.no_default,
+    atol: float | lib.NoDefault = lib.no_default,
+    obj: str = "DataFrame",
+) -> None:
+    """
+    Check that left and right DataFrame are equal.
+
+    This function is intended to compare two DataFrames and output any
+    differences. It is mostly intended for use in unit tests.
+    Additional parameters allow varying the strictness of the
+    equality checks performed.
+
+    Parameters
+    ----------
+    left : DataFrame
+        First DataFrame to compare.
+    right : DataFrame
+        Second DataFrame to compare.
+    check_dtype : bool, default True
+        Whether to check the DataFrame dtype is identical.
+    check_index_type : bool or {'equiv'}, default 'equiv'
+        Whether to check the Index class, dtype and inferred_type
+        are identical.
+    check_column_type : bool or {'equiv'}, default 'equiv'
+        Whether to check the columns class, dtype and inferred_type
+        are identical. Is passed as the ``exact`` argument of
+        :func:`assert_index_equal`.
+    check_frame_type : bool, default True
+        Whether to check the DataFrame class is identical.
+    check_names : bool, default True
+        Whether to check that the `names` attribute for both the `index`
+        and `column` attributes of the DataFrame is identical.
+    by_blocks : bool, default False
+        Specify how to compare internal data. If False, compare by columns.
+        If True, compare by blocks.
+    check_exact : bool, default False
+        Whether to compare number exactly. If False, the comparison uses the
+        relative tolerance (``rtol``) and absolute tolerance (``atol``)
+        parameters to determine if two values are considered close,
+        according to the formula: ``|a - b| <= (atol + rtol * |b|)``.
+
+        .. versionchanged:: 2.2.0
+
+            Defaults to True for integer dtypes if none of
+            ``check_exact``, ``rtol`` and ``atol`` are specified.
+    check_datetimelike_compat : bool, default False
+        Compare datetime-like which is comparable ignoring dtype.
+    check_categorical : bool, default True
+        Whether to compare internal Categorical exactly.
+    check_like : bool, default False
+        If True, ignore the order of index & columns.
+        Note: index labels must match their respective rows
+        (same as in columns) - same labels must be with the same data.
+    check_freq : bool, default True
+        Whether to check the `freq` attribute on a DatetimeIndex or TimedeltaIndex.
+    check_flags : bool, default True
+        Whether to check the `flags` attribute.
+    rtol : float, default 1e-5
+        Relative tolerance. Only used when check_exact is False.
+    atol : float, default 1e-8
+        Absolute tolerance. Only used when check_exact is False.
+    obj : str, default 'DataFrame'
+        Specify object name being compared, internally used to show appropriate
+        assertion message.
+
+    See Also
+    --------
+    assert_series_equal : Equivalent method for asserting Series equality.
+    DataFrame.equals : Check DataFrame equality.
+
+    Examples
+    --------
+    This example shows comparing two DataFrames that are equal
+    but with columns of differing dtypes.
+
+    >>> from pandas.testing import assert_frame_equal
+    >>> df1 = pd.DataFrame({"a": [1, 2], "b": [3, 4]})
+    >>> df2 = pd.DataFrame({"a": [1, 2], "b": [3.0, 4.0]})
+
+    df1 equals itself.
+
+    >>> assert_frame_equal(df1, df1)
+
+    df1 differs from df2 as column 'b' is of a different type.
+
+    >>> assert_frame_equal(df1, df2)
+    Traceback (most recent call last):
+    ...
+    AssertionError: Attributes of DataFrame.iloc[:, 1] (column name="b") are different
+
+    Attribute "dtype" are different
+    [left]:  int64
+    [right]: float64
+
+    Ignore differing dtypes in columns with check_dtype.
+
+    >>> assert_frame_equal(df1, df2, check_dtype=False)
+    """
+    __tracebackhide__ = True
+    _rtol = rtol if rtol is not lib.no_default else 1.0e-5
+    _atol = atol if atol is not lib.no_default else 1.0e-8
+    _check_exact = check_exact if check_exact is not lib.no_default else False
+
+    # instance validation
+    _check_isinstance(left, right, DataFrame)
+
+    if check_frame_type:
+        assert isinstance(left, type(right))
+        # assert_class_equal(left, right, obj=obj)
+
+    # shape comparison
+    if left.shape != right.shape:
+        raise_assert_detail(
+            obj, f"{obj} shape mismatch", f"{left.shape!r}", f"{right.shape!r}"
+        )
+
+    if check_flags:
+        assert left.flags == right.flags, f"{left.flags!r} != {right.flags!r}"
+
+    # index comparison
+    assert_index_equal(
+        left.index,
+        right.index,
+        exact=check_index_type,
+        check_names=check_names,
+        check_exact=_check_exact,
+        check_categorical=check_categorical,
+        check_order=not check_like,
+        rtol=_rtol,
+        atol=_atol,
+        obj=f"{obj}.index",
+    )
+
+    # column comparison
+    assert_index_equal(
+        left.columns,
+        right.columns,
+        exact=check_column_type,
+        check_names=check_names,
+        check_exact=_check_exact,
+        check_categorical=check_categorical,
+        check_order=not check_like,
+        rtol=_rtol,
+        atol=_atol,
+        obj=f"{obj}.columns",
+    )
+
+    if check_like:
+        left = left.reindex_like(right)
+
+    # compare by blocks
+    if by_blocks:
+        rblocks = right._to_dict_of_blocks()
+        lblocks = left._to_dict_of_blocks()
+        for dtype in list(set(list(lblocks.keys()) + list(rblocks.keys()))):
+            assert dtype in lblocks
+            assert dtype in rblocks
+            assert_frame_equal(
+                lblocks[dtype], rblocks[dtype], check_dtype=check_dtype, obj=obj
+            )
+
+    # compare by columns
+    else:
+        for i, col in enumerate(left.columns):
+            # We have already checked that columns match, so we can do
+            #  fast location-based lookups
+            lcol = left._ixs(i, axis=1)
+            rcol = right._ixs(i, axis=1)
+
+            # GH #38183
+            # use check_index=False, because we do not want to run
+            # assert_index_equal for each column,
+            # as we already checked it for the whole dataframe before.
+            assert_series_equal(
+                lcol,
+                rcol,
+                check_dtype=check_dtype,
+                check_index_type=check_index_type,
+                check_exact=check_exact,
+                check_names=check_names,
+                check_datetimelike_compat=check_datetimelike_compat,
+                check_categorical=check_categorical,
+                check_freq=check_freq,
+                obj=f'{obj}.iloc[:, {i}] (column name="{col}")',
+                rtol=rtol,
+                atol=atol,
+                check_index=False,
+                check_flags=False,
+            )
+
+
+def assert_equal(left, right, **kwargs) -> None:
+    """
+    Wrapper for tm.assert_*_equal to dispatch to the appropriate test function.
+
+    Parameters
+    ----------
+    left, right : Index, Series, DataFrame, ExtensionArray, or np.ndarray
+        The two items to be compared.
+    **kwargs
+        All keyword arguments are passed through to the underlying assert method.
+    """
+    __tracebackhide__ = True
+
+    if isinstance(left, Index):
+        assert_index_equal(left, right, **kwargs)
+        if isinstance(left, (DatetimeIndex, TimedeltaIndex)):
+            assert left.freq == right.freq, (left.freq, right.freq)
+    elif isinstance(left, Series):
+        assert_series_equal(left, right, **kwargs)
+    elif isinstance(left, DataFrame):
+        assert_frame_equal(left, right, **kwargs)
+    elif isinstance(left, IntervalArray):
+        assert_interval_array_equal(left, right, **kwargs)
+    elif isinstance(left, PeriodArray):
+        assert_period_array_equal(left, right, **kwargs)
+    elif isinstance(left, DatetimeArray):
+        assert_datetime_array_equal(left, right, **kwargs)
+    elif isinstance(left, TimedeltaArray):
+        assert_timedelta_array_equal(left, right, **kwargs)
+    elif isinstance(left, ExtensionArray):
+        assert_extension_array_equal(left, right, **kwargs)
+    elif isinstance(left, np.ndarray):
+        assert_numpy_array_equal(left, right, **kwargs)
+    elif isinstance(left, str):
+        assert kwargs == {}
+        assert left == right
+    else:
+        assert kwargs == {}
+        assert_almost_equal(left, right)
+
+
+def assert_sp_array_equal(left, right) -> None:
+    """
+    Check that the left and right SparseArray are equal.
+
+    Parameters
+    ----------
+    left : SparseArray
+    right : SparseArray
+    """
+    _check_isinstance(left, right, pd.arrays.SparseArray)
+
+    assert_numpy_array_equal(left.sp_values, right.sp_values)
+
+    # SparseIndex comparison
+    assert isinstance(left.sp_index, SparseIndex)
+    assert isinstance(right.sp_index, SparseIndex)
+
+    left_index = left.sp_index
+    right_index = right.sp_index
+
+    if not left_index.equals(right_index):
+        raise_assert_detail(
+            "SparseArray.index", "index are not equal", left_index, right_index
+        )
+    else:
+        # Just ensure a
+        pass
+
+    assert_attr_equal("fill_value", left, right)
+    assert_attr_equal("dtype", left, right)
+    assert_numpy_array_equal(left.to_dense(), right.to_dense())
+
+
+def assert_contains_all(iterable, dic) -> None:
+    for k in iterable:
+        assert k in dic, f"Did not contain item: {k!r}"
+
+
+def assert_copy(iter1, iter2, **eql_kwargs) -> None:
+    """
+    iter1, iter2: iterables that produce elements
+    comparable with assert_almost_equal
+
+    Checks that the elements are equal, but not
+    the same object. (Does not check that items
+    in sequences are also not the same object)
+    """
+    for elem1, elem2 in zip(iter1, iter2):
+        assert_almost_equal(elem1, elem2, **eql_kwargs)
+        msg = (
+            f"Expected object {type(elem1)!r} and object {type(elem2)!r} to be "
+            "different objects, but they were the same object."
+        )
+        assert elem1 is not elem2, msg
+
+
+def is_extension_array_dtype_and_needs_i8_conversion(
+    left_dtype: DtypeObj, right_dtype: DtypeObj
+) -> bool:
+    """
+    Checks that we have the combination of an ExtensionArraydtype and
+    a dtype that should be converted to int64
+
+    Returns
+    -------
+    bool
+
+    Related to issue #37609
+    """
+    return isinstance(left_dtype, ExtensionDtype) and needs_i8_conversion(right_dtype)
+
+
+def assert_indexing_slices_equivalent(ser: Series, l_slc: slice, i_slc: slice) -> None:
+    """
+    Check that ser.iloc[i_slc] matches ser.loc[l_slc] and, if applicable,
+    ser[l_slc].
+    """
+    expected = ser.iloc[i_slc]
+
+    assert_series_equal(ser.loc[l_slc], expected)
+
+    if not is_integer_dtype(ser.index):
+        # For integer indices, .loc and plain getitem are position-based.
+        assert_series_equal(ser[l_slc], expected)
+
+
+def assert_metadata_equivalent(
+    left: DataFrame | Series, right: DataFrame | Series | None = None
+) -> None:
+    """
+    Check that ._metadata attributes are equivalent.
+    """
+    for attr in left._metadata:
+        val = getattr(left, attr, None)
+        if right is None:
+            assert val is None
+        else:
+            assert val == getattr(right, attr, None)
diff --git a/pandas/pandas/_testing/compat.py b/pandas/pandas/_testing/compat.py
new file mode 100644
index 0000000000000..722ba61a3227f
--- /dev/null
+++ b/pandas/pandas/_testing/compat.py
@@ -0,0 +1,30 @@
+"""
+Helpers for sharing tests between DataFrame/Series
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas import DataFrame
+
+if TYPE_CHECKING:
+    from pandas._typing import DtypeObj
+
+
+def get_dtype(obj) -> DtypeObj:
+    if isinstance(obj, DataFrame):
+        # Note: we are assuming only one column
+        return obj.dtypes.iat[0]
+    else:
+        return obj.dtype
+
+
+def get_obj(df: DataFrame, klass):
+    """
+    For sharing tests using frame_or_series, either return the DataFrame
+    unchanged or return it's first column as a Series.
+    """
+    if klass is DataFrame:
+        return df
+    return df._ixs(0, axis=1)
diff --git a/pandas/pandas/_testing/contexts.py b/pandas/pandas/_testing/contexts.py
new file mode 100644
index 0000000000000..99826de51e1bf
--- /dev/null
+++ b/pandas/pandas/_testing/contexts.py
@@ -0,0 +1,184 @@
+from __future__ import annotations
+
+from contextlib import contextmanager
+import os
+from pathlib import Path
+import tempfile
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+)
+import uuid
+
+from pandas.compat import PYPY
+from pandas.errors import ChainedAssignmentError
+
+from pandas.io.common import get_handle
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+    from pandas._typing import (
+        BaseBuffer,
+        CompressionOptions,
+        FilePath,
+    )
+
+
+@contextmanager
+def decompress_file(
+    path: FilePath | BaseBuffer, compression: CompressionOptions
+) -> Generator[IO[bytes]]:
+    """
+    Open a compressed file and return a file object.
+
+    Parameters
+    ----------
+    path : str
+        The path where the file is read from.
+
+    compression : {'gzip', 'bz2', 'zip', 'xz', 'zstd', None}
+        Name of the decompression to use
+
+    Returns
+    -------
+    file object
+    """
+    with get_handle(path, "rb", compression=compression, is_text=False) as handle:
+        yield handle.handle
+
+
+@contextmanager
+def set_timezone(tz: str) -> Generator[None]:
+    """
+    Context manager for temporarily setting a timezone.
+
+    Parameters
+    ----------
+    tz : str
+        A string representing a valid timezone.
+
+    Examples
+    --------
+    >>> from datetime import datetime
+    >>> from dateutil.tz import tzlocal
+    >>> tzlocal().tzname(datetime(2021, 1, 1))  # doctest: +SKIP
+    'IST'
+
+    >>> with set_timezone("US/Eastern"):
+    ...     tzlocal().tzname(datetime(2021, 1, 1))
+    'EST'
+    """
+    import time
+
+    def setTZ(tz) -> None:
+        if hasattr(time, "tzset"):
+            if tz is None:
+                try:
+                    del os.environ["TZ"]
+                except KeyError:
+                    pass
+            else:
+                os.environ["TZ"] = tz
+                time.tzset()
+
+    orig_tz = os.environ.get("TZ")
+    setTZ(tz)
+    try:
+        yield
+    finally:
+        setTZ(orig_tz)
+
+
+@contextmanager
+def ensure_clean(filename=None) -> Generator[Any]:
+    """
+    Gets a temporary path and agrees to remove on close.
+
+    This implementation does not use tempfile.mkstemp to avoid having a file handle.
+    If the code using the returned path wants to delete the file itself, windows
+    requires that no program has a file handle to it.
+
+    Parameters
+    ----------
+    filename : str (optional)
+        suffix of the created file.
+    """
+    folder = Path(tempfile.gettempdir())
+
+    if filename is None:
+        filename = ""
+    filename = str(uuid.uuid4()) + filename
+    path = folder / filename
+
+    path.touch()
+
+    handle_or_str = str(path)
+
+    try:
+        yield handle_or_str
+    finally:
+        if path.is_file():
+            path.unlink()
+
+
+@contextmanager
+def with_csv_dialect(name: str, **kwargs) -> Generator[None]:
+    """
+    Context manager to temporarily register a CSV dialect for parsing CSV.
+
+    Parameters
+    ----------
+    name : str
+        The name of the dialect.
+    kwargs : mapping
+        The parameters for the dialect.
+
+    Raises
+    ------
+    ValueError : the name of the dialect conflicts with a builtin one.
+
+    See Also
+    --------
+    csv : Python's CSV library.
+    """
+    import csv
+
+    _BUILTIN_DIALECTS = {"excel", "excel-tab", "unix"}
+
+    if name in _BUILTIN_DIALECTS:
+        raise ValueError("Cannot override builtin dialect.")
+
+    csv.register_dialect(name, **kwargs)
+    try:
+        yield
+    finally:
+        csv.unregister_dialect(name)
+
+
+def raises_chained_assignment_error(extra_warnings=(), extra_match=()):
+    from pandas._testing import assert_produces_warning
+
+    if PYPY:
+        if not extra_warnings:
+            from contextlib import nullcontext
+
+            return nullcontext()
+        else:
+            return assert_produces_warning(
+                extra_warnings,
+                match=extra_match,
+            )
+    else:
+        warning = ChainedAssignmentError
+        match = (
+            "A value is trying to be set on a copy of a DataFrame or Series "
+            "through chained assignment"
+        )
+        if extra_warnings:
+            warning = (warning, *extra_warnings)  # type: ignore[assignment]
+        return assert_produces_warning(
+            warning,
+            match=(match, *extra_match),
+        )
diff --git a/pandas/pandas/_typing.py b/pandas/pandas/_typing.py
new file mode 100644
index 0000000000000..4365ee85f72e3
--- /dev/null
+++ b/pandas/pandas/_typing.py
@@ -0,0 +1,532 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterator,
+    Mapping,
+    MutableMapping,
+    Sequence,
+)
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+    tzinfo,
+)
+from os import PathLike
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    Optional,
+    Protocol,
+    Type as type_t,
+    TypeVar,
+    Union,
+    overload,
+)
+
+import numpy as np
+
+# To prevent import cycles place any internal imports in the branch below
+# and use a string literal forward reference to it in subsequent types
+# https://mypy.readthedocs.io/en/latest/common_issues.html#import-cycles
+if TYPE_CHECKING:
+    import numpy.typing as npt
+
+    from pandas._libs import (
+        NaTType,
+        Period,
+        Timedelta,
+        Timestamp,
+    )
+    from pandas._libs.tslibs import BaseOffset
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+    from pandas import (
+        DatetimeIndex,
+        Interval,
+        PeriodIndex,
+        TimedeltaIndex,
+    )
+    from pandas.arrays import (
+        DatetimeArray,
+        TimedeltaArray,
+    )
+    from pandas.core.arrays.base import ExtensionArray
+    from pandas.core.frame import DataFrame
+    from pandas.core.generic import NDFrame
+    from pandas.core.groupby.generic import (
+        DataFrameGroupBy,
+        GroupBy,
+        SeriesGroupBy,
+    )
+    from pandas.core.indexes.base import Index
+    from pandas.core.internals import (
+        BlockManager,
+        SingleBlockManager,
+    )
+    from pandas.core.resample import Resampler
+    from pandas.core.series import Series
+    from pandas.core.window.rolling import BaseWindow
+
+    from pandas.io.formats.format import EngFormatter
+    from pandas.tseries.holiday import AbstractHolidayCalendar
+
+    ScalarLike_co = Union[
+        int,
+        float,
+        complex,
+        str,
+        bytes,
+        np.generic,
+    ]
+
+    # numpy compatible types
+    NumpyValueArrayLike = Union[ScalarLike_co, npt.ArrayLike]
+    # Name "npt._ArrayLikeInt_co" is not defined  [name-defined]
+    NumpySorter = Optional[npt._ArrayLikeInt_co]  # type: ignore[name-defined]
+
+    from typing import (
+        ParamSpec,
+        SupportsIndex,
+    )
+    from typing import Concatenate  # pyright: ignore[reportUnusedImport]
+    from typing import TypeGuard  # pyright: ignore[reportUnusedImport]
+
+    P = ParamSpec("P")
+
+    if sys.version_info >= (3, 11):
+        from typing import Self  # pyright: ignore[reportUnusedImport]
+        from typing import Unpack  # pyright: ignore[reportUnusedImport]
+    else:
+        from typing_extensions import Self  # pyright: ignore[reportUnusedImport]
+        from typing_extensions import Unpack  # pyright: ignore[reportUnusedImport]
+
+else:
+    npt: Any = None
+    ParamSpec: Any = None
+    Self: Any = None
+    TypeGuard: Any = None
+    Concatenate: Any = None
+    Unpack: Any = None
+
+HashableT = TypeVar("HashableT", bound=Hashable)
+HashableT2 = TypeVar("HashableT2", bound=Hashable)
+MutableMappingT = TypeVar("MutableMappingT", bound=MutableMapping)
+
+# array-like
+
+ArrayLike = Union["ExtensionArray", np.ndarray]
+ArrayLikeT = TypeVar("ArrayLikeT", "ExtensionArray", np.ndarray)
+AnyArrayLike = Union[ArrayLike, "Index", "Series"]
+TimeArrayLike = Union["DatetimeArray", "TimedeltaArray"]
+
+# list-like
+
+# from https://github.com/hauntsaninja/useful_types
+# includes Sequence-like objects but excludes str and bytes
+_T_co = TypeVar("_T_co", covariant=True)
+
+
+class SequenceNotStr(Protocol[_T_co]):
+    @overload
+    def __getitem__(self, index: SupportsIndex, /) -> _T_co: ...
+
+    @overload
+    def __getitem__(self, index: slice, /) -> Sequence[_T_co]: ...
+
+    def __contains__(self, value: object, /) -> bool: ...
+
+    def __len__(self) -> int: ...
+
+    def __iter__(self) -> Iterator[_T_co]: ...
+
+    def index(self, value: Any, start: int = ..., stop: int = ..., /) -> int: ...
+
+    def count(self, value: Any, /) -> int: ...
+
+    def __reversed__(self) -> Iterator[_T_co]: ...
+
+
+ListLike = Union[AnyArrayLike, SequenceNotStr, range]
+
+# scalars
+
+PythonScalar = Union[str, float, bool]
+DatetimeLikeScalar = Union["Period", "Timestamp", "Timedelta"]
+PandasScalar = Union["Period", "Timestamp", "Timedelta", "Interval"]
+Scalar = Union[PythonScalar, PandasScalar, np.datetime64, np.timedelta64, date]
+IntStrT = TypeVar("IntStrT", bound=Union[int, str])
+
+
+# timestamp and timedelta convertible types
+
+TimestampConvertibleTypes = Union[
+    "Timestamp", date, np.datetime64, np.int64, float, str
+]
+TimestampNonexistent = Union[
+    Literal["shift_forward", "shift_backward", "NaT", "raise"], timedelta
+]
+TimedeltaConvertibleTypes = Union[
+    "Timedelta", timedelta, np.timedelta64, np.int64, float, str
+]
+Timezone = Union[str, tzinfo]
+
+ToTimestampHow = Literal["s", "e", "start", "end"]
+
+# NDFrameT is stricter and ensures that the same subclass of NDFrame always is
+# used. E.g. `def func(a: NDFrameT) -> NDFrameT: ...` means that if a
+# Series is passed into a function, a Series is always returned and if a DataFrame is
+# passed in, a DataFrame is always returned.
+NDFrameT = TypeVar("NDFrameT", bound="NDFrame")
+
+IndexT = TypeVar("IndexT", bound="Index")
+FreqIndexT = TypeVar("FreqIndexT", "DatetimeIndex", "PeriodIndex", "TimedeltaIndex")
+NumpyIndexT = TypeVar("NumpyIndexT", np.ndarray, "Index")
+
+AxisInt = int
+Axis = Union[AxisInt, Literal["index", "columns", "rows"]]
+IndexLabel = Union[Hashable, Sequence[Hashable]]
+Level = Hashable
+Shape = tuple[int, ...]
+Suffixes = Sequence[Optional[str]]
+Ordered = Optional[bool]
+JSONSerializable = Optional[Union[PythonScalar, list, dict]]
+Frequency = Union[str, "BaseOffset"]
+Axes = ListLike
+
+RandomState = Union[
+    int,
+    np.ndarray,
+    np.random.Generator,
+    np.random.BitGenerator,
+    np.random.RandomState,
+]
+
+# dtypes
+NpDtype = Union[str, np.dtype, type_t[Union[str, complex, bool, object]]]
+Dtype = Union["ExtensionDtype", NpDtype]
+AstypeArg = Union["ExtensionDtype", "npt.DTypeLike"]
+# DtypeArg specifies all allowable dtypes in a functions its dtype argument
+DtypeArg = Union[Dtype, Mapping[Hashable, Dtype]]
+DtypeObj = Union[np.dtype, "ExtensionDtype"]
+
+# converters
+ConvertersArg = dict[Hashable, Callable[[Dtype], Dtype]]
+
+# parse_dates
+ParseDatesArg = Union[
+    bool, list[Hashable], list[list[Hashable]], dict[Hashable, list[Hashable]]
+]
+
+# For functions like rename that convert one label to another
+Renamer = Union[Mapping[Any, Hashable], Callable[[Any], Hashable]]
+
+# to maintain type information across generic functions and parametrization
+T = TypeVar("T")
+
+# used in decorators to preserve the signature of the function it decorates
+# see https://mypy.readthedocs.io/en/stable/generics.html#declaring-decorators
+FuncType = Callable[..., Any]
+F = TypeVar("F", bound=FuncType)
+TypeT = TypeVar("TypeT", bound=type)
+
+# types of vectorized key functions for DataFrame::sort_values and
+# DataFrame::sort_index, among others
+ValueKeyFunc = Optional[Callable[["Series"], Union["Series", AnyArrayLike]]]
+IndexKeyFunc = Optional[Callable[["Index"], Union["Index", AnyArrayLike]]]
+
+# types of `func` kwarg for DataFrame.aggregate and Series.aggregate
+AggFuncTypeBase = Union[Callable, str]
+AggFuncTypeDict = MutableMapping[
+    Hashable, Union[AggFuncTypeBase, list[AggFuncTypeBase]]
+]
+AggFuncType = Union[
+    AggFuncTypeBase,
+    list[AggFuncTypeBase],
+    AggFuncTypeDict,
+]
+AggObjType = Union[
+    "Series",
+    "DataFrame",
+    "GroupBy",
+    "SeriesGroupBy",
+    "DataFrameGroupBy",
+    "BaseWindow",
+    "Resampler",
+]
+
+PythonFuncType = Callable[[Any], Any]
+
+# filenames and file-like-objects
+AnyStr_co = TypeVar("AnyStr_co", str, bytes, covariant=True)
+AnyStr_contra = TypeVar("AnyStr_contra", str, bytes, contravariant=True)
+
+
+class BaseBuffer(Protocol):
+    @property
+    def mode(self) -> str:
+        # for _get_filepath_or_buffer
+        ...
+
+    def seek(self, offset: int, whence: int = ..., /) -> int:
+        # with one argument: gzip.GzipFile, bz2.BZ2File
+        # with two arguments: zip.ZipFile, read_sas
+        ...
+
+    def seekable(self) -> bool:
+        # for bz2.BZ2File
+        ...
+
+    def tell(self) -> int:
+        # for zip.ZipFile, read_stata, to_stata
+        ...
+
+
+class ReadBuffer(BaseBuffer, Protocol[AnyStr_co]):
+    def read(self, n: int = ..., /) -> AnyStr_co:
+        # for BytesIOWrapper, gzip.GzipFile, bz2.BZ2File
+        ...
+
+
+class WriteBuffer(BaseBuffer, Protocol[AnyStr_contra]):
+    def write(self, b: AnyStr_contra, /) -> Any:
+        # for gzip.GzipFile, bz2.BZ2File
+        ...
+
+    def flush(self) -> Any:
+        # for gzip.GzipFile, bz2.BZ2File
+        ...
+
+
+class ReadPickleBuffer(ReadBuffer[bytes], Protocol):
+    def readline(self) -> bytes: ...
+
+
+class WriteExcelBuffer(WriteBuffer[bytes], Protocol):
+    def truncate(self, size: int | None = ..., /) -> int: ...
+
+
+class ReadCsvBuffer(ReadBuffer[AnyStr_co], Protocol):
+    def __iter__(self) -> Iterator[AnyStr_co]:
+        # for engine=python
+        ...
+
+    def fileno(self) -> int:
+        # for _MMapWrapper
+        ...
+
+    def readline(self) -> AnyStr_co:
+        # for engine=python
+        ...
+
+    @property
+    def closed(self) -> bool:
+        # for engine=pyarrow
+        ...
+
+
+FilePath = Union[str, "PathLike[str]"]
+
+# for arbitrary kwargs passed during reading/writing files
+StorageOptions = Optional[dict[str, Any]]
+
+
+# compression keywords and compression
+CompressionDict = dict[str, Any]
+CompressionOptions = Optional[
+    Union[Literal["infer", "gzip", "bz2", "zip", "xz", "zstd", "tar"], CompressionDict]
+]
+
+# types in DataFrameFormatter
+FormattersType = Union[
+    list[Callable], tuple[Callable, ...], Mapping[Union[str, int], Callable]
+]
+ColspaceType = Mapping[Hashable, Union[str, int]]
+FloatFormatType = Union[str, Callable, "EngFormatter"]
+ColspaceArgType = Union[
+    str, int, Sequence[Union[str, int]], Mapping[Hashable, Union[str, int]]
+]
+
+# Arguments for fillna()
+FillnaOptions = Literal["backfill", "bfill", "ffill", "pad"]
+InterpolateOptions = Literal[
+    "linear",
+    "time",
+    "index",
+    "values",
+    "nearest",
+    "zero",
+    "slinear",
+    "quadratic",
+    "cubic",
+    "barycentric",
+    "polynomial",
+    "krogh",
+    "piecewise_polynomial",
+    "spline",
+    "pchip",
+    "akima",
+    "cubicspline",
+    "from_derivatives",
+]
+
+# internals
+Manager = Union["BlockManager", "SingleBlockManager"]
+
+# indexing
+# PositionalIndexer -> valid 1D positional indexer, e.g. can pass
+# to ndarray.__getitem__
+# ScalarIndexer is for a single value as the index
+# SequenceIndexer is for list like or slices (but not tuples)
+# PositionalIndexerTuple is extends the PositionalIndexer for 2D arrays
+# These are used in various __getitem__ overloads
+# TODO(typing#684): add Ellipsis, see
+# https://github.com/python/typing/issues/684#issuecomment-548203158
+# https://bugs.python.org/issue41810
+# Using List[int] here rather than Sequence[int] to disallow tuples.
+ScalarIndexer = Union[int, np.integer]
+SequenceIndexer = Union[slice, list[int], np.ndarray]
+PositionalIndexer = Union[ScalarIndexer, SequenceIndexer]
+PositionalIndexerTuple = tuple[PositionalIndexer, PositionalIndexer]
+PositionalIndexer2D = Union[PositionalIndexer, PositionalIndexerTuple]
+if TYPE_CHECKING:
+    TakeIndexer = Union[Sequence[int], Sequence[np.integer], npt.NDArray[np.integer]]
+else:
+    TakeIndexer = Any
+
+# Shared by functions such as drop and astype
+IgnoreRaise = Literal["ignore", "raise"]
+
+# Windowing rank methods
+WindowingRankType = Literal["average", "min", "max"]
+
+# read_csv engines
+CSVEngine = Literal["c", "python", "pyarrow", "python-fwf"]
+
+# read_json engines
+JSONEngine = Literal["ujson", "pyarrow"]
+
+# read_xml parsers
+XMLParsers = Literal["lxml", "etree"]
+
+# read_html flavors
+HTMLFlavors = Literal["lxml", "html5lib", "bs4"]
+
+# Interval closed type
+IntervalLeftRight = Literal["left", "right"]
+IntervalClosedType = Union[IntervalLeftRight, Literal["both", "neither"]]
+
+# datetime and NaTType
+DatetimeNaTType = Union[datetime, "NaTType"]
+DateTimeErrorChoices = Literal["raise", "coerce"]
+
+# sort_index
+SortKind = Literal["quicksort", "mergesort", "heapsort", "stable"]
+NaPosition = Literal["first", "last"]
+
+# Arguments for nsmallest and nlargest
+NsmallestNlargestKeep = Literal["first", "last", "all"]
+
+# quantile interpolation
+QuantileInterpolation = Literal["linear", "lower", "higher", "midpoint", "nearest"]
+
+# plotting
+PlottingOrientation = Literal["horizontal", "vertical"]
+
+# dropna
+AnyAll = Literal["any", "all"]
+
+# merge
+MergeHow = Literal[
+    "left", "right", "inner", "outer", "cross", "left_anti", "right_anti"
+]
+MergeValidate = Literal[
+    "one_to_one",
+    "1:1",
+    "one_to_many",
+    "1:m",
+    "many_to_one",
+    "m:1",
+    "many_to_many",
+    "m:m",
+]
+
+# join
+JoinHow = Literal["left", "right", "inner", "outer"]
+JoinValidate = Literal[
+    "one_to_one",
+    "1:1",
+    "one_to_many",
+    "1:m",
+    "many_to_one",
+    "m:1",
+    "many_to_many",
+    "m:m",
+]
+
+# reindex
+ReindexMethod = Union[FillnaOptions, Literal["nearest"]]
+
+MatplotlibColor = Union[str, Sequence[float]]
+TimeGrouperOrigin = Union[
+    "Timestamp", Literal["epoch", "start", "start_day", "end", "end_day"]
+]
+TimeAmbiguous = Union[Literal["infer", "NaT", "raise"], "npt.NDArray[np.bool_]"]
+TimeNonexistent = Union[
+    Literal["shift_forward", "shift_backward", "NaT", "raise"], timedelta
+]
+DropKeep = Literal["first", "last", False]
+CorrelationMethod = Union[
+    Literal["pearson", "kendall", "spearman"], Callable[[np.ndarray, np.ndarray], float]
+]
+AlignJoin = Literal["outer", "inner", "left", "right"]
+DtypeBackend = Literal["pyarrow", "numpy_nullable"]
+
+TimeUnit = Literal["s", "ms", "us", "ns"]
+OpenFileErrors = Literal[
+    "strict",
+    "ignore",
+    "replace",
+    "surrogateescape",
+    "xmlcharrefreplace",
+    "backslashreplace",
+    "namereplace",
+]
+
+# update
+UpdateJoin = Literal["left"]
+
+# applymap
+NaAction = Literal["ignore"]
+
+# from_dict
+FromDictOrient = Literal["columns", "index", "tight"]
+
+# to_stata
+ToStataByteorder = Literal[">", "<", "little", "big"]
+
+# ExcelWriter
+ExcelWriterIfSheetExists = Literal["error", "new", "replace", "overlay"]
+ExcelWriterMergeCells = Union[bool, Literal["columns"]]
+
+# Offsets
+OffsetCalendar = Union[np.busdaycalendar, "AbstractHolidayCalendar"]
+
+# read_csv: usecols
+UsecolsArgType = Union[
+    SequenceNotStr[Hashable],
+    range,
+    AnyArrayLike,
+    Callable[[HashableT], bool],
+    None,
+]
+
+# maintain the sub-type of any hashable sequence
+SequenceT = TypeVar("SequenceT", bound=Sequence[Hashable])
+
+SliceType = Optional[Hashable]
diff --git a/pandas/pandas/_version.py b/pandas/pandas/_version.py
new file mode 100644
index 0000000000000..c5e3c16d3f773
--- /dev/null
+++ b/pandas/pandas/_version.py
@@ -0,0 +1,692 @@
+# This file helps to compute a version number in source trees obtained from
+# git-archive tarball (such as those provided by github's download-from-tag
+# feature). Distribution tarballs (built by setup.py sdist) and build
+# directories (produced by setup.py build) will contain a much shorter file
+# that just contains the computed version number.
+
+# This file is released into the public domain.
+# Generated by versioneer-0.28
+# https://github.com/python-versioneer/python-versioneer
+
+"""Git implementation of _version.py."""
+
+from collections.abc import Callable
+import errno
+import functools
+import os
+import re
+import subprocess
+import sys
+
+
+def get_keywords():
+    """Get the keywords needed to look up the version information."""
+    # these strings will be replaced by git during git-archive.
+    # setup.py/versioneer.py will grep for the variable names, so they must
+    # each be defined on a line of their own. _version.py will just call
+    # get_keywords().
+    git_refnames = "$Format:%d$"
+    git_full = "$Format:%H$"
+    git_date = "$Format:%ci$"
+    keywords = {"refnames": git_refnames, "full": git_full, "date": git_date}
+    return keywords
+
+
+class VersioneerConfig:
+    """Container for Versioneer configuration parameters."""
+
+
+def get_config():
+    """Create, populate and return the VersioneerConfig() object."""
+    # these strings are filled in when 'setup.py versioneer' creates
+    # _version.py
+    cfg = VersioneerConfig()
+    cfg.VCS = "git"
+    cfg.style = "pep440"
+    cfg.tag_prefix = "v"
+    cfg.parentdir_prefix = "pandas-"
+    cfg.versionfile_source = "pandas/_version.py"
+    cfg.verbose = False
+    return cfg
+
+
+class NotThisMethod(Exception):
+    """Exception raised if a method is not valid for the current scenario."""
+
+
+LONG_VERSION_PY: dict[str, str] = {}
+HANDLERS: dict[str, dict[str, Callable]] = {}
+
+
+def register_vcs_handler(vcs, method):  # decorator
+    """Create decorator to mark a method as the handler of a VCS."""
+
+    def decorate(f):
+        """Store f in HANDLERS[vcs][method]."""
+        if vcs not in HANDLERS:
+            HANDLERS[vcs] = {}
+        HANDLERS[vcs][method] = f
+        return f
+
+    return decorate
+
+
+def run_command(commands, args, cwd=None, verbose=False, hide_stderr=False, env=None):
+    """Call the given command(s)."""
+    assert isinstance(commands, list)
+    process = None
+
+    popen_kwargs = {}
+    if sys.platform == "win32":
+        # This hides the console window if pythonw.exe is used
+        startupinfo = subprocess.STARTUPINFO()
+        startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW
+        popen_kwargs["startupinfo"] = startupinfo
+
+    for command in commands:
+        dispcmd = str([command] + args)
+        try:
+            # remember shell=False, so use git.cmd on windows, not just git
+            process = subprocess.Popen(
+                [command] + args,
+                cwd=cwd,
+                env=env,
+                stdout=subprocess.PIPE,
+                stderr=(subprocess.PIPE if hide_stderr else None),
+                **popen_kwargs,
+            )
+            break
+        except OSError:
+            e = sys.exc_info()[1]
+            if e.errno == errno.ENOENT:
+                continue
+            if verbose:
+                print(f"unable to run {dispcmd}")
+                print(e)
+            return None, None
+    else:
+        if verbose:
+            print(f"unable to find command, tried {commands}")
+        return None, None
+    stdout = process.communicate()[0].strip().decode()
+    if process.returncode != 0:
+        if verbose:
+            print(f"unable to run {dispcmd} (error)")
+            print(f"stdout was {stdout}")
+        return None, process.returncode
+    return stdout, process.returncode
+
+
+def versions_from_parentdir(parentdir_prefix, root, verbose):
+    """Try to determine the version from the parent directory name.
+
+    Source tarballs conventionally unpack into a directory that includes both
+    the project name and a version string. We will also support searching up
+    two directory levels for an appropriately named parent directory
+    """
+    rootdirs = []
+
+    for _ in range(3):
+        dirname = os.path.basename(root)
+        if dirname.startswith(parentdir_prefix):
+            return {
+                "version": dirname[len(parentdir_prefix) :],
+                "full-revisionid": None,
+                "dirty": False,
+                "error": None,
+                "date": None,
+            }
+        rootdirs.append(root)
+        root = os.path.dirname(root)  # up a level
+
+    if verbose:
+        print(
+            f"Tried directories {rootdirs!s} \
+            but none started with prefix {parentdir_prefix}"
+        )
+    raise NotThisMethod("rootdir doesn't start with parentdir_prefix")
+
+
+@register_vcs_handler("git", "get_keywords")
+def git_get_keywords(versionfile_abs):
+    """Extract version information from the given file."""
+    # the code embedded in _version.py can just fetch the value of these
+    # keywords. When used from setup.py, we don't want to import _version.py,
+    # so we do it with a regexp instead. This function is not used from
+    # _version.py.
+    keywords = {}
+    try:
+        with open(versionfile_abs, encoding="utf-8") as fobj:
+            for line in fobj:
+                if line.strip().startswith("git_refnames ="):
+                    mo = re.search(r'=\s*"(.*)"', line)
+                    if mo:
+                        keywords["refnames"] = mo.group(1)
+                if line.strip().startswith("git_full ="):
+                    mo = re.search(r'=\s*"(.*)"', line)
+                    if mo:
+                        keywords["full"] = mo.group(1)
+                if line.strip().startswith("git_date ="):
+                    mo = re.search(r'=\s*"(.*)"', line)
+                    if mo:
+                        keywords["date"] = mo.group(1)
+    except OSError:
+        pass
+    return keywords
+
+
+@register_vcs_handler("git", "keywords")
+def git_versions_from_keywords(keywords, tag_prefix, verbose):
+    """Get version information from git keywords."""
+    if "refnames" not in keywords:
+        raise NotThisMethod("Short version file found")
+    date = keywords.get("date")
+    if date is not None:
+        # Use only the last line.  Previous lines may contain GPG signature
+        # information.
+        date = date.splitlines()[-1]
+
+        # git-2.2.0 added "%cI", which expands to an ISO-8601 -compliant
+        # datestamp. However we prefer "%ci" (which expands to an "ISO-8601
+        # -like" string, which we must then edit to make compliant), because
+        # it's been around since git-1.5.3, and it's too difficult to
+        # discover which version we're using, or to work around using an
+        # older one.
+        date = date.strip().replace(" ", "T", 1).replace(" ", "", 1)
+    refnames = keywords["refnames"].strip()
+    if refnames.startswith("$Format"):
+        if verbose:
+            print("keywords are unexpanded, not using")
+        raise NotThisMethod("unexpanded keywords, not a git-archive tarball")
+    refs = {r.strip() for r in refnames.strip("()").split(",")}
+    # starting in git-1.8.3, tags are listed as "tag: foo-1.0" instead of
+    # just "foo-1.0". If we see a "tag: " prefix, prefer those.
+    TAG = "tag: "
+    tags = {r[len(TAG) :] for r in refs if r.startswith(TAG)}
+    if not tags:
+        # Either we're using git < 1.8.3, or there really are no tags. We use
+        # a heuristic: assume all version tags have a digit. The old git %d
+        # expansion behaves like git log --decorate=short and strips out the
+        # refs/heads/ and refs/tags/ prefixes that would let us distinguish
+        # between branches and tags. By ignoring refnames without digits, we
+        # filter out many common branch names like "release" and
+        # "stabilization", as well as "HEAD" and "master".
+        tags = {r for r in refs if re.search(r"\d", r)}
+        if verbose:
+            print(f"discarding '{','.join(refs - tags)}', no digits")
+    if verbose:
+        print(f"likely tags: {','.join(sorted(tags))}")
+    for ref in sorted(tags):
+        # sorting will prefer e.g. "2.0" over "2.0rc1"
+        if ref.startswith(tag_prefix):
+            r = ref[len(tag_prefix) :]
+            # Filter out refs that exactly match prefix or that don't start
+            # with a number once the prefix is stripped (mostly a concern
+            # when prefix is '')
+            if not re.match(r"\d", r):
+                continue
+            if verbose:
+                print(f"picking {r}")
+            return {
+                "version": r,
+                "full-revisionid": keywords["full"].strip(),
+                "dirty": False,
+                "error": None,
+                "date": date,
+            }
+    # no suitable tags, so version is "0+unknown", but full hex is still there
+    if verbose:
+        print("no suitable tags, using unknown + full revision id")
+    return {
+        "version": "0+unknown",
+        "full-revisionid": keywords["full"].strip(),
+        "dirty": False,
+        "error": "no suitable tags",
+        "date": None,
+    }
+
+
+@register_vcs_handler("git", "pieces_from_vcs")
+def git_pieces_from_vcs(tag_prefix, root, verbose, runner=run_command):
+    """Get version from 'git describe' in the root of the source tree.
+
+    This only gets called if the git-archive 'subst' keywords were *not*
+    expanded, and _version.py hasn't already been rewritten with a short
+    version string, meaning we're inside a checked out source tree.
+    """
+    GITS = ["git"]
+    if sys.platform == "win32":
+        GITS = ["git.cmd", "git.exe"]
+
+    # GIT_DIR can interfere with correct operation of Versioneer.
+    # It may be intended to be passed to the Versioneer-versioned project,
+    # but that should not change where we get our version from.
+    env = os.environ.copy()
+    env.pop("GIT_DIR", None)
+    runner = functools.partial(runner, env=env)
+
+    _, rc = runner(GITS, ["rev-parse", "--git-dir"], cwd=root, hide_stderr=not verbose)
+    if rc != 0:
+        if verbose:
+            print(f"Directory {root} not under git control")
+        raise NotThisMethod("'git rev-parse --git-dir' returned error")
+
+    # if there is a tag matching tag_prefix, this yields TAG-NUM-gHEX[-dirty]
+    # if there isn't one, this yields HEX[-dirty] (no NUM)
+    describe_out, rc = runner(
+        GITS,
+        [
+            "describe",
+            "--tags",
+            "--dirty",
+            "--always",
+            "--long",
+            "--match",
+            f"{tag_prefix}[[:digit:]]*",
+        ],
+        cwd=root,
+    )
+    # --long was added in git-1.5.5
+    if describe_out is None:
+        raise NotThisMethod("'git describe' failed")
+    describe_out = describe_out.strip()
+    full_out, rc = runner(GITS, ["rev-parse", "HEAD"], cwd=root)
+    if full_out is None:
+        raise NotThisMethod("'git rev-parse' failed")
+    full_out = full_out.strip()
+
+    pieces = {}
+    pieces["long"] = full_out
+    pieces["short"] = full_out[:7]  # maybe improved later
+    pieces["error"] = None
+
+    branch_name, rc = runner(GITS, ["rev-parse", "--abbrev-ref", "HEAD"], cwd=root)
+    # --abbrev-ref was added in git-1.6.3
+    if rc != 0 or branch_name is None:
+        raise NotThisMethod("'git rev-parse --abbrev-ref' returned error")
+    branch_name = branch_name.strip()
+
+    if branch_name == "HEAD":
+        # If we aren't exactly on a branch, pick a branch which represents
+        # the current commit. If all else fails, we are on a branchless
+        # commit.
+        branches, rc = runner(GITS, ["branch", "--contains"], cwd=root)
+        # --contains was added in git-1.5.4
+        if rc != 0 or branches is None:
+            raise NotThisMethod("'git branch --contains' returned error")
+        branches = branches.split("\n")
+
+        # Remove the first line if we're running detached
+        if "(" in branches[0]:
+            branches.pop(0)
+
+        # Strip off the leading "* " from the list of branches.
+        branches = [branch[2:] for branch in branches]
+        if "master" in branches:
+            branch_name = "master"
+        elif not branches:
+            branch_name = None
+        else:
+            # Pick the first branch that is returned. Good or bad.
+            branch_name = branches[0]
+
+    pieces["branch"] = branch_name
+
+    # parse describe_out. It will be like TAG-NUM-gHEX[-dirty] or HEX[-dirty]
+    # TAG might have hyphens.
+    git_describe = describe_out
+
+    # look for -dirty suffix
+    dirty = git_describe.endswith("-dirty")
+    pieces["dirty"] = dirty
+    if dirty:
+        git_describe = git_describe[: git_describe.rindex("-dirty")]
+
+    # now we have TAG-NUM-gHEX or HEX
+
+    if "-" in git_describe:
+        # TAG-NUM-gHEX
+        mo = re.search(r"^(.+)-(\d+)-g([0-9a-f]+)$", git_describe)
+        if not mo:
+            # unparsable. Maybe git-describe is misbehaving?
+            pieces["error"] = f"unable to parse git-describe output: '{describe_out}'"
+            return pieces
+
+        # tag
+        full_tag = mo.group(1)
+        if not full_tag.startswith(tag_prefix):
+            if verbose:
+                fmt = "tag '%s' doesn't start with prefix '%s'"
+                print(fmt % (full_tag, tag_prefix))
+            pieces["error"] = (
+                f"tag '{full_tag}' doesn't start with prefix '{tag_prefix}'"
+            )
+            return pieces
+        pieces["closest-tag"] = full_tag[len(tag_prefix) :]
+
+        # distance: number of commits since tag
+        pieces["distance"] = int(mo.group(2))
+
+        # commit: short hex revision ID
+        pieces["short"] = mo.group(3)
+
+    else:
+        # HEX: no tags
+        pieces["closest-tag"] = None
+        out, rc = runner(GITS, ["rev-list", "HEAD", "--left-right"], cwd=root)
+        pieces["distance"] = len(out.split())  # total number of commits
+
+    # commit date: see ISO-8601 comment in git_versions_from_keywords()
+    date = runner(GITS, ["show", "-s", "--format=%ci", "HEAD"], cwd=root)[0].strip()
+    # Use only the last line.  Previous lines may contain GPG signature
+    # information.
+    date = date.splitlines()[-1]
+    pieces["date"] = date.strip().replace(" ", "T", 1).replace(" ", "", 1)
+
+    return pieces
+
+
+def plus_or_dot(pieces) -> str:
+    """Return a + if we don't already have one, else return a ."""
+    if "+" in pieces.get("closest-tag", ""):
+        return "."
+    return "+"
+
+
+def render_pep440(pieces):
+    """Build up version string, with post-release "local version identifier".
+
+    Our goal: TAG[+DISTANCE.gHEX[.dirty]] . Note that if you
+    get a tagged build and then dirty it, you'll get TAG+0.gHEX.dirty
+
+    Exceptions:
+    1: no tags. git_describe was just HEX. 0+untagged.DISTANCE.gHEX[.dirty]
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"] or pieces["dirty"]:
+            rendered += plus_or_dot(pieces)
+            rendered += f"{pieces['distance']}.g{pieces['short']}"
+            if pieces["dirty"]:
+                rendered += ".dirty"
+    else:
+        # exception #1
+        rendered = f"0+untagged.{pieces['distance']}.g{pieces['short']}"
+        if pieces["dirty"]:
+            rendered += ".dirty"
+    return rendered
+
+
+def render_pep440_branch(pieces):
+    """TAG[[.dev0]+DISTANCE.gHEX[.dirty]] .
+
+    The ".dev0" means not master branch. Note that .dev0 sorts backwards
+    (a feature branch will appear "older" than the master branch).
+
+    Exceptions:
+    1: no tags. 0[.dev0]+untagged.DISTANCE.gHEX[.dirty]
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"] or pieces["dirty"]:
+            if pieces["branch"] != "master":
+                rendered += ".dev0"
+            rendered += plus_or_dot(pieces)
+            rendered += f"{pieces['distance']}.g{pieces['short']}"
+            if pieces["dirty"]:
+                rendered += ".dirty"
+    else:
+        # exception #1
+        rendered = "0"
+        if pieces["branch"] != "master":
+            rendered += ".dev0"
+        rendered += f"+untagged.{pieces['distance']}.g{pieces['short']}"
+        if pieces["dirty"]:
+            rendered += ".dirty"
+    return rendered
+
+
+def pep440_split_post(ver):
+    """Split pep440 version string at the post-release segment.
+
+    Returns the release segments before the post-release and the
+    post-release version number (or -1 if no post-release segment is present).
+    """
+    vc = str.split(ver, ".post")
+    return vc[0], int(vc[1] or 0) if len(vc) == 2 else None
+
+
+def render_pep440_pre(pieces):
+    """TAG[.postN.devDISTANCE] -- No -dirty.
+
+    Exceptions:
+    1: no tags. 0.post0.devDISTANCE
+    """
+    if pieces["closest-tag"]:
+        if pieces["distance"]:
+            # update the post release segment
+            tag_version, post_version = pep440_split_post(pieces["closest-tag"])
+            rendered = tag_version
+            if post_version is not None:
+                rendered += f".post{post_version + 1}.dev{pieces['distance']}"
+            else:
+                rendered += f".post0.dev{pieces['distance']}"
+        else:
+            # no commits, use the tag as the version
+            rendered = pieces["closest-tag"]
+    else:
+        # exception #1
+        rendered = f"0.post0.dev{pieces['distance']}"
+    return rendered
+
+
+def render_pep440_post(pieces):
+    """TAG[.postDISTANCE[.dev0]+gHEX] .
+
+    The ".dev0" means dirty. Note that .dev0 sorts backwards
+    (a dirty tree will appear "older" than the corresponding clean one),
+    but you shouldn't be releasing software with -dirty anyways.
+
+    Exceptions:
+    1: no tags. 0.postDISTANCE[.dev0]
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"] or pieces["dirty"]:
+            rendered += f".post{pieces['distance']}"
+            if pieces["dirty"]:
+                rendered += ".dev0"
+            rendered += plus_or_dot(pieces)
+            rendered += f"g{pieces['short']}"
+    else:
+        # exception #1
+        rendered = f"0.post{pieces['distance']}"
+        if pieces["dirty"]:
+            rendered += ".dev0"
+        rendered += f"+g{pieces['short']}"
+    return rendered
+
+
+def render_pep440_post_branch(pieces):
+    """TAG[.postDISTANCE[.dev0]+gHEX[.dirty]] .
+
+    The ".dev0" means not master branch.
+
+    Exceptions:
+    1: no tags. 0.postDISTANCE[.dev0]+gHEX[.dirty]
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"] or pieces["dirty"]:
+            rendered += f".post{pieces['distance']}"
+            if pieces["branch"] != "master":
+                rendered += ".dev0"
+            rendered += plus_or_dot(pieces)
+            rendered += f"g{pieces['short']}"
+            if pieces["dirty"]:
+                rendered += ".dirty"
+    else:
+        # exception #1
+        rendered = f"0.post{pieces['distance']}"
+        if pieces["branch"] != "master":
+            rendered += ".dev0"
+        rendered += f"+g{pieces['short']}"
+        if pieces["dirty"]:
+            rendered += ".dirty"
+    return rendered
+
+
+def render_pep440_old(pieces):
+    """TAG[.postDISTANCE[.dev0]] .
+
+    The ".dev0" means dirty.
+
+    Exceptions:
+    1: no tags. 0.postDISTANCE[.dev0]
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"] or pieces["dirty"]:
+            rendered += f"0.post{pieces['distance']}"
+            if pieces["dirty"]:
+                rendered += ".dev0"
+    else:
+        # exception #1
+        rendered = f"0.post{pieces['distance']}"
+        if pieces["dirty"]:
+            rendered += ".dev0"
+    return rendered
+
+
+def render_git_describe(pieces):
+    """TAG[-DISTANCE-gHEX][-dirty].
+
+    Like 'git describe --tags --dirty --always'.
+
+    Exceptions:
+    1: no tags. HEX[-dirty]  (note: no 'g' prefix)
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        if pieces["distance"]:
+            rendered += f"-{pieces['distance']}-g{pieces['short']}"
+    else:
+        # exception #1
+        rendered = pieces["short"]
+    if pieces["dirty"]:
+        rendered += "-dirty"
+    return rendered
+
+
+def render_git_describe_long(pieces):
+    """TAG-DISTANCE-gHEX[-dirty].
+
+    Like 'git describe --tags --dirty --always -long'.
+    The distance/hash is unconditional.
+
+    Exceptions:
+    1: no tags. HEX[-dirty]  (note: no 'g' prefix)
+    """
+    if pieces["closest-tag"]:
+        rendered = pieces["closest-tag"]
+        rendered += f"-{pieces['distance']}-g{pieces['short']}"
+    else:
+        # exception #1
+        rendered = pieces["short"]
+    if pieces["dirty"]:
+        rendered += "-dirty"
+    return rendered
+
+
+def render(pieces, style):
+    """Render the given version pieces into the requested style."""
+    if pieces["error"]:
+        return {
+            "version": "unknown",
+            "full-revisionid": pieces.get("long"),
+            "dirty": None,
+            "error": pieces["error"],
+            "date": None,
+        }
+
+    if not style or style == "default":
+        style = "pep440"  # the default
+
+    if style == "pep440":
+        rendered = render_pep440(pieces)
+    elif style == "pep440-branch":
+        rendered = render_pep440_branch(pieces)
+    elif style == "pep440-pre":
+        rendered = render_pep440_pre(pieces)
+    elif style == "pep440-post":
+        rendered = render_pep440_post(pieces)
+    elif style == "pep440-post-branch":
+        rendered = render_pep440_post_branch(pieces)
+    elif style == "pep440-old":
+        rendered = render_pep440_old(pieces)
+    elif style == "git-describe":
+        rendered = render_git_describe(pieces)
+    elif style == "git-describe-long":
+        rendered = render_git_describe_long(pieces)
+    else:
+        raise ValueError(f"unknown style '{style}'")
+
+    return {
+        "version": rendered,
+        "full-revisionid": pieces["long"],
+        "dirty": pieces["dirty"],
+        "error": None,
+        "date": pieces.get("date"),
+    }
+
+
+def get_versions():
+    """Get version information or return default if unable to do so."""
+    # I am in _version.py, which lives at ROOT/VERSIONFILE_SOURCE. If we have
+    # __file__, we can work backwards from there to the root. Some
+    # py2exe/bbfreeze/non-CPython implementations don't do __file__, in which
+    # case we can only use expanded keywords.
+
+    cfg = get_config()
+    verbose = cfg.verbose
+
+    try:
+        return git_versions_from_keywords(get_keywords(), cfg.tag_prefix, verbose)
+    except NotThisMethod:
+        pass
+
+    try:
+        root = os.path.realpath(__file__)
+        # versionfile_source is the relative path from the top of the source
+        # tree (where the .git directory might live) to this file. Invert
+        # this to find the root from __file__.
+        for _ in cfg.versionfile_source.split("/"):
+            root = os.path.dirname(root)
+    except NameError:
+        return {
+            "version": "0+unknown",
+            "full-revisionid": None,
+            "dirty": None,
+            "error": "unable to find root of source tree",
+            "date": None,
+        }
+
+    try:
+        pieces = git_pieces_from_vcs(cfg.tag_prefix, root, verbose)
+        return render(pieces, cfg.style)
+    except NotThisMethod:
+        pass
+
+    try:
+        if cfg.parentdir_prefix:
+            return versions_from_parentdir(cfg.parentdir_prefix, root, verbose)
+    except NotThisMethod:
+        pass
+
+    return {
+        "version": "0+unknown",
+        "full-revisionid": None,
+        "dirty": None,
+        "error": "unable to compute version",
+        "date": None,
+    }
diff --git a/pandas/pandas/api/__init__.py b/pandas/pandas/api/__init__.py
new file mode 100644
index 0000000000000..8f659e3cd14c8
--- /dev/null
+++ b/pandas/pandas/api/__init__.py
@@ -0,0 +1,17 @@
+"""public toolkit API"""
+
+from pandas.api import (
+    extensions,
+    indexers,
+    interchange,
+    types,
+    typing,
+)
+
+__all__ = [
+    "extensions",
+    "indexers",
+    "interchange",
+    "types",
+    "typing",
+]
diff --git a/pandas/pandas/api/extensions/__init__.py b/pandas/pandas/api/extensions/__init__.py
new file mode 100644
index 0000000000000..1c88c0d35b4d7
--- /dev/null
+++ b/pandas/pandas/api/extensions/__init__.py
@@ -0,0 +1,33 @@
+"""
+Public API for extending pandas objects.
+"""
+
+from pandas._libs.lib import no_default
+
+from pandas.core.dtypes.base import (
+    ExtensionDtype,
+    register_extension_dtype,
+)
+
+from pandas.core.accessor import (
+    register_dataframe_accessor,
+    register_index_accessor,
+    register_series_accessor,
+)
+from pandas.core.algorithms import take
+from pandas.core.arrays import (
+    ExtensionArray,
+    ExtensionScalarOpsMixin,
+)
+
+__all__ = [
+    "ExtensionArray",
+    "ExtensionDtype",
+    "ExtensionScalarOpsMixin",
+    "no_default",
+    "register_dataframe_accessor",
+    "register_extension_dtype",
+    "register_index_accessor",
+    "register_series_accessor",
+    "take",
+]
diff --git a/pandas/pandas/api/indexers/__init__.py b/pandas/pandas/api/indexers/__init__.py
new file mode 100644
index 0000000000000..f3c6546218de4
--- /dev/null
+++ b/pandas/pandas/api/indexers/__init__.py
@@ -0,0 +1,17 @@
+"""
+Public API for Rolling Window Indexers.
+"""
+
+from pandas.core.indexers import check_array_indexer
+from pandas.core.indexers.objects import (
+    BaseIndexer,
+    FixedForwardWindowIndexer,
+    VariableOffsetWindowIndexer,
+)
+
+__all__ = [
+    "BaseIndexer",
+    "FixedForwardWindowIndexer",
+    "VariableOffsetWindowIndexer",
+    "check_array_indexer",
+]
diff --git a/pandas/pandas/api/interchange/__init__.py b/pandas/pandas/api/interchange/__init__.py
new file mode 100644
index 0000000000000..aded37abc7224
--- /dev/null
+++ b/pandas/pandas/api/interchange/__init__.py
@@ -0,0 +1,8 @@
+"""
+Public API for DataFrame interchange protocol.
+"""
+
+from pandas.core.interchange.dataframe_protocol import DataFrame
+from pandas.core.interchange.from_dataframe import from_dataframe
+
+__all__ = ["DataFrame", "from_dataframe"]
diff --git a/pandas/pandas/api/internals.py b/pandas/pandas/api/internals.py
new file mode 100644
index 0000000000000..03d8992a87575
--- /dev/null
+++ b/pandas/pandas/api/internals.py
@@ -0,0 +1,62 @@
+import numpy as np
+
+from pandas._typing import ArrayLike
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+from pandas.core.internals.api import _make_block
+from pandas.core.internals.managers import BlockManager as _BlockManager
+
+
+def create_dataframe_from_blocks(
+    blocks: list[tuple[ArrayLike, np.ndarray]], index: Index, columns: Index
+) -> DataFrame:
+    """
+    Low-level function to create a DataFrame from arrays as they are
+    representing the block structure of the resulting DataFrame.
+
+    Attention: this is an advanced, low-level function that should only be
+    used if you know that the below-mentioned assumptions are guaranteed.
+    If passing data that do not follow those assumptions, subsequent
+    subsequent operations on the resulting DataFrame might lead to strange
+    errors.
+    For almost all use cases, you should use the standard pd.DataFrame(..)
+    constructor instead. If you are planning to use this function, let us
+    know by opening an issue at https://github.com/pandas-dev/pandas/issues.
+
+    Assumptions:
+
+    - The block arrays are either a 2D numpy array or a pandas ExtensionArray
+    - In case of a numpy array, it is assumed to already be in the expected
+      shape for Blocks (2D, (cols, rows), i.e. transposed compared to the
+      DataFrame columns).
+    - All arrays are taken as is (no type inference) and expected to have the
+      correct size.
+    - The placement arrays have the correct length (equalling the number of
+      columns that its equivalent block array represents), and all placement
+      arrays together form a complete set of 0 to n_columns - 1.
+
+    Parameters
+    ----------
+    blocks : list of tuples of (block_array, block_placement)
+        This should be a list of tuples existing of (block_array, block_placement),
+        where:
+
+        - block_array is a 2D numpy array or a 1D ExtensionArray, following the
+          requirements listed above.
+        - block_placement is a 1D integer numpy array
+    index : Index
+        The Index object for the `index` of the resulting DataFrame.
+    columns : Index
+        The Index object for the `columns` of the resulting DataFrame.
+
+    Returns
+    -------
+    DataFrame
+    """
+    block_objs = [_make_block(*block) for block in blocks]
+    axes = [columns, index]
+    mgr = _BlockManager(block_objs, axes)
+    return DataFrame._from_mgr(mgr, mgr.axes)
diff --git a/pandas/pandas/api/types/__init__.py b/pandas/pandas/api/types/__init__.py
new file mode 100644
index 0000000000000..4a5c742b1628b
--- /dev/null
+++ b/pandas/pandas/api/types/__init__.py
@@ -0,0 +1,23 @@
+"""
+Public toolkit API.
+"""
+
+from pandas._libs.lib import infer_dtype
+
+from pandas.core.dtypes.api import *  # noqa: F403
+from pandas.core.dtypes.concat import union_categoricals
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+
+__all__ = [
+    "CategoricalDtype",
+    "DatetimeTZDtype",
+    "IntervalDtype",
+    "PeriodDtype",
+    "infer_dtype",
+    "union_categoricals",
+]
diff --git a/pandas/pandas/api/typing/__init__.py b/pandas/pandas/api/typing/__init__.py
new file mode 100644
index 0000000000000..c1178c72f3edc
--- /dev/null
+++ b/pandas/pandas/api/typing/__init__.py
@@ -0,0 +1,59 @@
+"""
+Public API classes that store intermediate results useful for type-hinting.
+"""
+
+from pandas._libs import NaTType
+from pandas._libs.lib import NoDefault
+from pandas._libs.missing import NAType
+
+from pandas.core.groupby import (
+    DataFrameGroupBy,
+    SeriesGroupBy,
+)
+from pandas.core.indexes.frozen import FrozenList
+from pandas.core.resample import (
+    DatetimeIndexResamplerGroupby,
+    PeriodIndexResamplerGroupby,
+    Resampler,
+    TimedeltaIndexResamplerGroupby,
+    TimeGrouper,
+)
+from pandas.core.window import (
+    Expanding,
+    ExpandingGroupby,
+    ExponentialMovingWindow,
+    ExponentialMovingWindowGroupby,
+    Rolling,
+    RollingGroupby,
+    Window,
+)
+
+# TODO: Can't import Styler without importing jinja2
+# from pandas.io.formats.style import Styler
+from pandas.io.json._json import JsonReader
+from pandas.io.sas.sasreader import SASReader
+from pandas.io.stata import StataReader
+
+__all__ = [
+    "DataFrameGroupBy",
+    "DatetimeIndexResamplerGroupby",
+    "Expanding",
+    "ExpandingGroupby",
+    "ExponentialMovingWindow",
+    "ExponentialMovingWindowGroupby",
+    "FrozenList",
+    "JsonReader",
+    "NAType",
+    "NaTType",
+    "NoDefault",
+    "PeriodIndexResamplerGroupby",
+    "Resampler",
+    "Rolling",
+    "RollingGroupby",
+    "SASReader",
+    "SeriesGroupBy",
+    "StataReader",
+    "TimeGrouper",
+    "TimedeltaIndexResamplerGroupby",
+    "Window",
+]
diff --git a/pandas/pandas/arrays/__init__.py b/pandas/pandas/arrays/__init__.py
new file mode 100644
index 0000000000000..b5c1c98da1c78
--- /dev/null
+++ b/pandas/pandas/arrays/__init__.py
@@ -0,0 +1,37 @@
+"""
+All of pandas' ExtensionArrays.
+
+See :ref:`extending.extension-types` for more.
+"""
+
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    ArrowStringArray,
+    BooleanArray,
+    Categorical,
+    DatetimeArray,
+    FloatingArray,
+    IntegerArray,
+    IntervalArray,
+    NumpyExtensionArray,
+    PeriodArray,
+    SparseArray,
+    StringArray,
+    TimedeltaArray,
+)
+
+__all__ = [
+    "ArrowExtensionArray",
+    "ArrowStringArray",
+    "BooleanArray",
+    "Categorical",
+    "DatetimeArray",
+    "FloatingArray",
+    "IntegerArray",
+    "IntervalArray",
+    "NumpyExtensionArray",
+    "PeriodArray",
+    "SparseArray",
+    "StringArray",
+    "TimedeltaArray",
+]
diff --git a/pandas/pandas/compat/__init__.py b/pandas/pandas/compat/__init__.py
new file mode 100644
index 0000000000000..138456f877c5f
--- /dev/null
+++ b/pandas/pandas/compat/__init__.py
@@ -0,0 +1,171 @@
+"""
+compat
+======
+
+Cross-compatible functions for different versions of Python.
+
+Other items:
+* platform checker
+"""
+
+from __future__ import annotations
+
+import os
+import platform
+import sys
+from typing import TYPE_CHECKING
+
+from pandas.compat._constants import (
+    IS64,
+    ISMUSL,
+    PY311,
+    PY312,
+    PYPY,
+    WASM,
+)
+from pandas.compat.numpy import is_numpy_dev
+from pandas.compat.pyarrow import (
+    HAS_PYARROW,
+    pa_version_under10p1,
+    pa_version_under11p0,
+    pa_version_under13p0,
+    pa_version_under14p0,
+    pa_version_under14p1,
+    pa_version_under16p0,
+    pa_version_under17p0,
+    pa_version_under18p0,
+    pa_version_under19p0,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import F
+
+
+def set_function_name(f: F, name: str, cls: type) -> F:
+    """
+    Bind the name/qualname attributes of the function.
+    """
+    f.__name__ = name
+    f.__qualname__ = f"{cls.__name__}.{name}"
+    f.__module__ = cls.__module__
+    return f
+
+
+def is_platform_little_endian() -> bool:
+    """
+    Checking if the running platform is little endian.
+
+    Returns
+    -------
+    bool
+        True if the running platform is little endian.
+    """
+    return sys.byteorder == "little"
+
+
+def is_platform_windows() -> bool:
+    """
+    Checking if the running platform is windows.
+
+    Returns
+    -------
+    bool
+        True if the running platform is windows.
+    """
+    return sys.platform in ["win32", "cygwin"]
+
+
+def is_platform_linux() -> bool:
+    """
+    Checking if the running platform is linux.
+
+    Returns
+    -------
+    bool
+        True if the running platform is linux.
+    """
+    return sys.platform == "linux"
+
+
+def is_platform_mac() -> bool:
+    """
+    Checking if the running platform is mac.
+
+    Returns
+    -------
+    bool
+        True if the running platform is mac.
+    """
+    return sys.platform == "darwin"
+
+
+def is_platform_arm() -> bool:
+    """
+    Checking if the running platform use ARM architecture.
+
+    Returns
+    -------
+    bool
+        True if the running platform uses ARM architecture.
+    """
+    return platform.machine() in ("arm64", "aarch64") or platform.machine().startswith(
+        "armv"
+    )
+
+
+def is_platform_power() -> bool:
+    """
+    Checking if the running platform use Power architecture.
+
+    Returns
+    -------
+    bool
+        True if the running platform uses ARM architecture.
+    """
+    return platform.machine() in ("ppc64", "ppc64le")
+
+
+def is_platform_riscv64() -> bool:
+    """
+    Checking if the running platform use riscv64 architecture.
+
+    Returns
+    -------
+    bool
+        True if the running platform uses riscv64 architecture.
+    """
+    return platform.machine() == "riscv64"
+
+
+def is_ci_environment() -> bool:
+    """
+    Checking if running in a continuous integration environment by checking
+    the PANDAS_CI environment variable.
+
+    Returns
+    -------
+    bool
+        True if the running in a continuous integration environment.
+    """
+    return os.environ.get("PANDAS_CI", "0") == "1"
+
+
+__all__ = [
+    "HAS_PYARROW",
+    "IS64",
+    "ISMUSL",
+    "PY311",
+    "PY312",
+    "PYPY",
+    "WASM",
+    "is_numpy_dev",
+    "pa_version_under10p1",
+    "pa_version_under11p0",
+    "pa_version_under13p0",
+    "pa_version_under14p0",
+    "pa_version_under14p1",
+    "pa_version_under16p0",
+    "pa_version_under17p0",
+    "pa_version_under18p0",
+    "pa_version_under19p0",
+]
diff --git a/pandas/pandas/compat/_constants.py b/pandas/pandas/compat/_constants.py
new file mode 100644
index 0000000000000..c7b7341013251
--- /dev/null
+++ b/pandas/pandas/compat/_constants.py
@@ -0,0 +1,30 @@
+"""
+_constants
+======
+
+Constants relevant for the Python implementation.
+"""
+
+from __future__ import annotations
+
+import platform
+import sys
+import sysconfig
+
+IS64 = sys.maxsize > 2**32
+
+PY311 = sys.version_info >= (3, 11)
+PY312 = sys.version_info >= (3, 12)
+PYPY = platform.python_implementation() == "PyPy"
+WASM = (sys.platform == "emscripten") or (platform.machine() in ["wasm32", "wasm64"])
+ISMUSL = "musl" in (sysconfig.get_config_var("HOST_GNU_TYPE") or "")
+REF_COUNT = 2 if PY311 else 3
+
+__all__ = [
+    "IS64",
+    "ISMUSL",
+    "PY311",
+    "PY312",
+    "PYPY",
+    "WASM",
+]
diff --git a/pandas/pandas/compat/_optional.py b/pandas/pandas/compat/_optional.py
new file mode 100644
index 0000000000000..6b90389a62056
--- /dev/null
+++ b/pandas/pandas/compat/_optional.py
@@ -0,0 +1,192 @@
+from __future__ import annotations
+
+import importlib
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    overload,
+)
+import warnings
+
+from pandas.util._exceptions import find_stack_level
+
+from pandas.util.version import Version
+
+if TYPE_CHECKING:
+    import types
+
+# Update install.rst, actions-310-minimum_versions.yaml,
+# deps_minimum.toml & pyproject.toml when updating versions!
+
+VERSIONS = {
+    "adbc-driver-postgresql": "0.10.0",
+    "adbc-driver-sqlite": "0.8.0",
+    "bs4": "4.11.2",
+    "blosc": "1.21.3",
+    "bottleneck": "1.3.6",
+    "fastparquet": "2023.10.0",
+    "fsspec": "2022.11.0",
+    "html5lib": "1.1",
+    "hypothesis": "6.84.0",
+    "gcsfs": "2022.11.0",
+    "jinja2": "3.1.2",
+    "lxml.etree": "4.9.2",
+    "matplotlib": "3.6.3",
+    "numba": "0.56.4",
+    "numexpr": "2.8.4",
+    "odfpy": "1.4.1",
+    "openpyxl": "3.1.0",
+    "psycopg2": "2.9.6",  # (dt dec pq3 ext lo64)
+    "pymysql": "1.0.2",
+    "pyarrow": "10.0.1",
+    "pyreadstat": "1.2.0",
+    "pytest": "7.3.2",
+    "python-calamine": "0.1.7",
+    "pytz": "2023.4",
+    "pyxlsb": "1.0.10",
+    "s3fs": "2022.11.0",
+    "scipy": "1.10.0",
+    "sqlalchemy": "2.0.0",
+    "tables": "3.8.0",
+    "tabulate": "0.9.0",
+    "xarray": "2022.12.0",
+    "xlrd": "2.0.1",
+    "xlsxwriter": "3.0.5",
+    "zstandard": "0.19.0",
+    "tzdata": "2022.7",
+    "qtpy": "2.3.0",
+    "pyqt5": "5.15.9",
+}
+
+# A mapping from import name to package name (on PyPI) for packages where
+# these two names are different.
+
+INSTALL_MAPPING = {
+    "bs4": "beautifulsoup4",
+    "bottleneck": "Bottleneck",
+    "jinja2": "Jinja2",
+    "lxml.etree": "lxml",
+    "odf": "odfpy",
+    "python_calamine": "python-calamine",
+    "sqlalchemy": "SQLAlchemy",
+    "tables": "pytables",
+}
+
+
+def get_version(module: types.ModuleType) -> str:
+    version = getattr(module, "__version__", None)
+
+    if version is None:
+        raise ImportError(f"Can't determine version for {module.__name__}")
+    if module.__name__ == "psycopg2":
+        # psycopg2 appends " (dt dec pq3 ext lo64)" to it's version
+        version = version.split()[0]
+    return version
+
+
+@overload
+def import_optional_dependency(
+    name: str,
+    extra: str = ...,
+    min_version: str | None = ...,
+    *,
+    errors: Literal["raise"] = ...,
+) -> types.ModuleType: ...
+
+
+@overload
+def import_optional_dependency(
+    name: str,
+    extra: str = ...,
+    min_version: str | None = ...,
+    *,
+    errors: Literal["warn", "ignore"],
+) -> types.ModuleType | None: ...
+
+
+def import_optional_dependency(
+    name: str,
+    extra: str = "",
+    min_version: str | None = None,
+    *,
+    errors: Literal["raise", "warn", "ignore"] = "raise",
+) -> types.ModuleType | None:
+    """
+    Import an optional dependency.
+
+    By default, if a dependency is missing an ImportError with a nice
+    message will be raised. If a dependency is present, but too old,
+    we raise.
+
+    Parameters
+    ----------
+    name : str
+        The module name.
+    extra : str
+        Additional text to include in the ImportError message.
+    errors : str {'raise', 'warn', 'ignore'}
+        What to do when a dependency is not found or its version is too old.
+
+        * raise : Raise an ImportError
+        * warn : Only applicable when a module's version is to old.
+          Warns that the version is too old and returns None
+        * ignore: If the module is not installed, return None, otherwise,
+          return the module, even if the version is too old.
+          It's expected that users validate the version locally when
+          using ``errors="ignore"`` (see. ``io/html.py``)
+    min_version : str, default None
+        Specify a minimum version that is different from the global pandas
+        minimum version required.
+    Returns
+    -------
+    maybe_module : Optional[ModuleType]
+        The imported module, when found and the version is correct.
+        None is returned when the package is not found and `errors`
+        is False, or when the package's version is too old and `errors`
+        is ``'warn'`` or ``'ignore'``.
+    """
+    assert errors in {"warn", "raise", "ignore"}
+
+    package_name = INSTALL_MAPPING.get(name)
+    install_name = package_name if package_name is not None else name
+
+    msg = (
+        f"Missing optional dependency '{install_name}'. {extra} "
+        f"Use pip or conda to install {install_name}."
+    )
+    try:
+        module = importlib.import_module(name)
+    except ImportError as err:
+        if errors == "raise":
+            raise ImportError(msg) from err
+        return None
+
+    # Handle submodules: if we have submodule, grab parent module from sys.modules
+    parent = name.split(".")[0]
+    if parent != name:
+        install_name = parent
+        module_to_get = sys.modules[install_name]
+    else:
+        module_to_get = module
+    minimum_version = min_version if min_version is not None else VERSIONS.get(parent)
+    if minimum_version:
+        version = get_version(module_to_get)
+        if version and Version(version) < Version(minimum_version):
+            msg = (
+                f"Pandas requires version '{minimum_version}' or newer of '{parent}' "
+                f"(version '{version}' currently installed)."
+            )
+            if errors == "warn":
+                warnings.warn(
+                    msg,
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+                return None
+            elif errors == "raise":
+                raise ImportError(msg)
+            else:
+                return None
+
+    return module
diff --git a/pandas/pandas/compat/numpy/__init__.py b/pandas/pandas/compat/numpy/__init__.py
new file mode 100644
index 0000000000000..3306b36d71806
--- /dev/null
+++ b/pandas/pandas/compat/numpy/__init__.py
@@ -0,0 +1,53 @@
+"""support numpy compatibility across versions"""
+
+import warnings
+
+import numpy as np
+
+from pandas.util.version import Version
+
+# numpy versioning
+_np_version = np.__version__
+_nlv = Version(_np_version)
+np_version_gte1p24 = _nlv >= Version("1.24")
+np_version_gte1p24p3 = _nlv >= Version("1.24.3")
+np_version_gte1p25 = _nlv >= Version("1.25")
+np_version_gt2 = _nlv >= Version("2.0.0")
+is_numpy_dev = _nlv.dev is not None
+_min_numpy_ver = "1.23.5"
+
+
+if _nlv < Version(_min_numpy_ver):
+    raise ImportError(
+        f"this version of pandas is incompatible with numpy < {_min_numpy_ver}\n"
+        f"your numpy version is {_np_version}.\n"
+        f"Please upgrade numpy to >= {_min_numpy_ver} to use this pandas version"
+    )
+
+
+np_long: type
+np_ulong: type
+
+if np_version_gt2:
+    try:
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore",
+                r".*In the future `np\.long` will be defined as.*",
+                FutureWarning,
+            )
+            np_long = np.long  # type: ignore[attr-defined]
+            np_ulong = np.ulong  # type: ignore[attr-defined]
+    except AttributeError:
+        np_long = np.int_
+        np_ulong = np.uint
+else:
+    np_long = np.int_
+    np_ulong = np.uint
+
+
+__all__ = [
+    "_np_version",
+    "is_numpy_dev",
+    "np",
+]
diff --git a/pandas/pandas/compat/numpy/function.py b/pandas/pandas/compat/numpy/function.py
new file mode 100644
index 0000000000000..abf86fc415641
--- /dev/null
+++ b/pandas/pandas/compat/numpy/function.py
@@ -0,0 +1,376 @@
+"""
+For compatibility with numpy libraries, pandas functions or methods have to
+accept '*args' and '**kwargs' parameters to accommodate numpy arguments that
+are not actually used or respected in the pandas implementation.
+
+To ensure that users do not abuse these parameters, validation is performed in
+'validators.py' to make sure that any extra parameters passed correspond ONLY
+to those in the numpy signature. Part of that validation includes whether or
+not the user attempted to pass in non-default values for these extraneous
+parameters. As we want to discourage users from relying on these parameters
+when calling the pandas implementation, we want them only to pass in the
+default values for these parameters.
+
+This module provides a set of commonly used default arguments for functions and
+methods that are spread throughout the codebase. This module will make it
+easier to adjust to future upstream changes in the analogous numpy signatures.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypeVar,
+    cast,
+    overload,
+)
+
+import numpy as np
+from numpy import ndarray
+
+from pandas._libs.lib import (
+    is_bool,
+    is_integer,
+)
+from pandas.errors import UnsupportedFunctionCall
+from pandas.util._validators import (
+    validate_args,
+    validate_args_and_kwargs,
+    validate_kwargs,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        Axis,
+        AxisInt,
+    )
+
+    AxisNoneT = TypeVar("AxisNoneT", Axis, None)
+
+
+class CompatValidator:
+    def __init__(
+        self,
+        defaults,
+        fname=None,
+        method: str | None = None,
+        max_fname_arg_count=None,
+    ) -> None:
+        self.fname = fname
+        self.method = method
+        self.defaults = defaults
+        self.max_fname_arg_count = max_fname_arg_count
+
+    def __call__(
+        self,
+        args,
+        kwargs,
+        fname=None,
+        max_fname_arg_count=None,
+        method: str | None = None,
+    ) -> None:
+        if not args and not kwargs:
+            return None
+
+        fname = self.fname if fname is None else fname
+        max_fname_arg_count = (
+            self.max_fname_arg_count
+            if max_fname_arg_count is None
+            else max_fname_arg_count
+        )
+        method = self.method if method is None else method
+
+        if method == "args":
+            validate_args(fname, args, max_fname_arg_count, self.defaults)
+        elif method == "kwargs":
+            validate_kwargs(fname, kwargs, self.defaults)
+        elif method == "both":
+            validate_args_and_kwargs(
+                fname, args, kwargs, max_fname_arg_count, self.defaults
+            )
+        else:
+            raise ValueError(f"invalid validation method '{method}'")
+
+
+ARGMINMAX_DEFAULTS = {"out": None}
+validate_argmin = CompatValidator(
+    ARGMINMAX_DEFAULTS, fname="argmin", method="both", max_fname_arg_count=1
+)
+validate_argmax = CompatValidator(
+    ARGMINMAX_DEFAULTS, fname="argmax", method="both", max_fname_arg_count=1
+)
+
+
+def process_skipna(skipna: bool | ndarray | None, args) -> tuple[bool, Any]:
+    if isinstance(skipna, ndarray) or skipna is None:
+        args = (skipna,) + args
+        skipna = True
+
+    return skipna, args
+
+
+def validate_argmin_with_skipna(skipna: bool | ndarray | None, args, kwargs) -> bool:
+    """
+    If 'Series.argmin' is called via the 'numpy' library, the third parameter
+    in its signature is 'out', which takes either an ndarray or 'None', so
+    check if the 'skipna' parameter is either an instance of ndarray or is
+    None, since 'skipna' itself should be a boolean
+    """
+    skipna, args = process_skipna(skipna, args)
+    validate_argmin(args, kwargs)
+    return skipna
+
+
+def validate_argmax_with_skipna(skipna: bool | ndarray | None, args, kwargs) -> bool:
+    """
+    If 'Series.argmax' is called via the 'numpy' library, the third parameter
+    in its signature is 'out', which takes either an ndarray or 'None', so
+    check if the 'skipna' parameter is either an instance of ndarray or is
+    None, since 'skipna' itself should be a boolean
+    """
+    skipna, args = process_skipna(skipna, args)
+    validate_argmax(args, kwargs)
+    return skipna
+
+
+ARGSORT_DEFAULTS: dict[str, int | str | None] = {}
+ARGSORT_DEFAULTS["axis"] = -1
+ARGSORT_DEFAULTS["kind"] = "quicksort"
+ARGSORT_DEFAULTS["order"] = None
+ARGSORT_DEFAULTS["kind"] = None
+ARGSORT_DEFAULTS["stable"] = None
+
+
+validate_argsort = CompatValidator(
+    ARGSORT_DEFAULTS, fname="argsort", max_fname_arg_count=0, method="both"
+)
+
+# two different signatures of argsort, this second validation for when the
+# `kind` param is supported
+ARGSORT_DEFAULTS_KIND: dict[str, int | None] = {}
+ARGSORT_DEFAULTS_KIND["axis"] = -1
+ARGSORT_DEFAULTS_KIND["order"] = None
+ARGSORT_DEFAULTS_KIND["stable"] = None
+validate_argsort_kind = CompatValidator(
+    ARGSORT_DEFAULTS_KIND, fname="argsort", max_fname_arg_count=0, method="both"
+)
+
+
+def validate_argsort_with_ascending(ascending: bool | int | None, args, kwargs) -> bool:
+    """
+    If 'Categorical.argsort' is called via the 'numpy' library, the first
+    parameter in its signature is 'axis', which takes either an integer or
+    'None', so check if the 'ascending' parameter has either integer type or is
+    None, since 'ascending' itself should be a boolean
+    """
+    if is_integer(ascending) or ascending is None:
+        args = (ascending,) + args
+        ascending = True
+
+    validate_argsort_kind(args, kwargs, max_fname_arg_count=3)
+    ascending = cast(bool, ascending)
+    return ascending
+
+
+CLIP_DEFAULTS: dict[str, Any] = {"out": None}
+validate_clip = CompatValidator(
+    CLIP_DEFAULTS, fname="clip", method="both", max_fname_arg_count=3
+)
+
+
+@overload
+def validate_clip_with_axis(axis: ndarray, args, kwargs) -> None: ...
+
+
+@overload
+def validate_clip_with_axis(axis: AxisNoneT, args, kwargs) -> AxisNoneT: ...
+
+
+def validate_clip_with_axis(
+    axis: ndarray | AxisNoneT, args, kwargs
+) -> AxisNoneT | None:
+    """
+    If 'NDFrame.clip' is called via the numpy library, the third parameter in
+    its signature is 'out', which can takes an ndarray, so check if the 'axis'
+    parameter is an instance of ndarray, since 'axis' itself should either be
+    an integer or None
+    """
+    if isinstance(axis, ndarray):
+        args = (axis,) + args
+        # error: Incompatible types in assignment (expression has type "None",
+        # variable has type "Union[ndarray[Any, Any], str, int]")
+        axis = None  # type: ignore[assignment]
+
+    validate_clip(args, kwargs)
+    # error: Incompatible return value type (got "Union[ndarray[Any, Any],
+    # str, int]", expected "Union[str, int, None]")
+    return axis  # type: ignore[return-value]
+
+
+CUM_FUNC_DEFAULTS: dict[str, Any] = {}
+CUM_FUNC_DEFAULTS["dtype"] = None
+CUM_FUNC_DEFAULTS["out"] = None
+validate_cum_func = CompatValidator(
+    CUM_FUNC_DEFAULTS, method="both", max_fname_arg_count=1
+)
+validate_cumsum = CompatValidator(
+    CUM_FUNC_DEFAULTS, fname="cumsum", method="both", max_fname_arg_count=1
+)
+
+
+def validate_cum_func_with_skipna(skipna: bool, args, kwargs, name) -> bool:
+    """
+    If this function is called via the 'numpy' library, the third parameter in
+    its signature is 'dtype', which takes either a 'numpy' dtype or 'None', so
+    check if the 'skipna' parameter is a boolean or not
+    """
+    if not is_bool(skipna):
+        args = (skipna,) + args
+        skipna = True
+    elif isinstance(skipna, np.bool_):
+        skipna = bool(skipna)
+
+    validate_cum_func(args, kwargs, fname=name)
+    return skipna
+
+
+ALLANY_DEFAULTS: dict[str, bool | None] = {}
+ALLANY_DEFAULTS["dtype"] = None
+ALLANY_DEFAULTS["out"] = None
+ALLANY_DEFAULTS["keepdims"] = False
+ALLANY_DEFAULTS["axis"] = None
+validate_all = CompatValidator(
+    ALLANY_DEFAULTS, fname="all", method="both", max_fname_arg_count=1
+)
+validate_any = CompatValidator(
+    ALLANY_DEFAULTS, fname="any", method="both", max_fname_arg_count=1
+)
+
+LOGICAL_FUNC_DEFAULTS = {"out": None, "keepdims": False}
+validate_logical_func = CompatValidator(LOGICAL_FUNC_DEFAULTS, method="kwargs")
+
+MINMAX_DEFAULTS = {"axis": None, "dtype": None, "out": None, "keepdims": False}
+validate_min = CompatValidator(
+    MINMAX_DEFAULTS, fname="min", method="both", max_fname_arg_count=1
+)
+validate_max = CompatValidator(
+    MINMAX_DEFAULTS, fname="max", method="both", max_fname_arg_count=1
+)
+
+
+REPEAT_DEFAULTS: dict[str, Any] = {"axis": None}
+validate_repeat = CompatValidator(
+    REPEAT_DEFAULTS, fname="repeat", method="both", max_fname_arg_count=1
+)
+
+ROUND_DEFAULTS: dict[str, Any] = {"out": None}
+validate_round = CompatValidator(
+    ROUND_DEFAULTS, fname="round", method="both", max_fname_arg_count=1
+)
+
+STAT_FUNC_DEFAULTS: dict[str, Any | None] = {}
+STAT_FUNC_DEFAULTS["dtype"] = None
+STAT_FUNC_DEFAULTS["out"] = None
+
+SUM_DEFAULTS = STAT_FUNC_DEFAULTS.copy()
+SUM_DEFAULTS["axis"] = None
+SUM_DEFAULTS["keepdims"] = False
+SUM_DEFAULTS["initial"] = None
+
+PROD_DEFAULTS = SUM_DEFAULTS.copy()
+
+MEAN_DEFAULTS = SUM_DEFAULTS.copy()
+
+MEDIAN_DEFAULTS = STAT_FUNC_DEFAULTS.copy()
+MEDIAN_DEFAULTS["overwrite_input"] = False
+MEDIAN_DEFAULTS["keepdims"] = False
+
+STAT_FUNC_DEFAULTS["keepdims"] = False
+
+validate_stat_func = CompatValidator(STAT_FUNC_DEFAULTS, method="kwargs")
+validate_sum = CompatValidator(
+    SUM_DEFAULTS, fname="sum", method="both", max_fname_arg_count=1
+)
+validate_prod = CompatValidator(
+    PROD_DEFAULTS, fname="prod", method="both", max_fname_arg_count=1
+)
+validate_mean = CompatValidator(
+    MEAN_DEFAULTS, fname="mean", method="both", max_fname_arg_count=1
+)
+validate_median = CompatValidator(
+    MEDIAN_DEFAULTS, fname="median", method="both", max_fname_arg_count=1
+)
+
+STAT_DDOF_FUNC_DEFAULTS: dict[str, bool | None] = {}
+STAT_DDOF_FUNC_DEFAULTS["dtype"] = None
+STAT_DDOF_FUNC_DEFAULTS["out"] = None
+STAT_DDOF_FUNC_DEFAULTS["keepdims"] = False
+validate_stat_ddof_func = CompatValidator(STAT_DDOF_FUNC_DEFAULTS, method="kwargs")
+
+TAKE_DEFAULTS: dict[str, str | None] = {}
+TAKE_DEFAULTS["out"] = None
+TAKE_DEFAULTS["mode"] = "raise"
+validate_take = CompatValidator(TAKE_DEFAULTS, fname="take", method="kwargs")
+
+
+TRANSPOSE_DEFAULTS = {"axes": None}
+validate_transpose = CompatValidator(
+    TRANSPOSE_DEFAULTS, fname="transpose", method="both", max_fname_arg_count=0
+)
+
+
+def validate_groupby_func(name: str, args, kwargs, allowed=None) -> None:
+    """
+    'args' and 'kwargs' should be empty, except for allowed kwargs because all
+    of their necessary parameters are explicitly listed in the function
+    signature
+    """
+    if allowed is None:
+        allowed = []
+
+    kwargs = set(kwargs) - set(allowed)
+
+    if len(args) + len(kwargs) > 0:
+        raise UnsupportedFunctionCall(
+            "numpy operations are not valid with groupby. "
+            f"Use .groupby(...).{name}() instead"
+        )
+
+
+def validate_minmax_axis(axis: AxisInt | None, ndim: int = 1) -> None:
+    """
+    Ensure that the axis argument passed to min, max, argmin, or argmax is zero
+    or None, as otherwise it will be incorrectly ignored.
+
+    Parameters
+    ----------
+    axis : int or None
+    ndim : int, default 1
+
+    Raises
+    ------
+    ValueError
+    """
+    if axis is None:
+        return
+    if axis >= ndim or (axis < 0 and ndim + axis < 0):
+        raise ValueError(f"`axis` must be fewer than the number of dimensions ({ndim})")
+
+
+_validation_funcs = {
+    "median": validate_median,
+    "mean": validate_mean,
+    "min": validate_min,
+    "max": validate_max,
+    "sum": validate_sum,
+    "prod": validate_prod,
+}
+
+
+def validate_func(fname, args, kwargs) -> None:
+    if fname not in _validation_funcs:
+        return validate_stat_func(args, kwargs, fname=fname)
+
+    validation_func = _validation_funcs[fname]
+    return validation_func(args, kwargs)
diff --git a/pandas/pandas/compat/pickle_compat.py b/pandas/pandas/compat/pickle_compat.py
new file mode 100644
index 0000000000000..beaaa3f8ed3cc
--- /dev/null
+++ b/pandas/pandas/compat/pickle_compat.py
@@ -0,0 +1,143 @@
+"""
+Pickle compatibility to pandas version 1.0
+"""
+
+from __future__ import annotations
+
+import contextlib
+import io
+import pickle
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs.arrays import NDArrayBacked
+from pandas._libs.tslibs import BaseOffset
+
+from pandas.core.arrays import (
+    DatetimeArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.internals import BlockManager
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+
+# If classes are moved, provide compat here.
+_class_locations_map = {
+    # Re-routing unpickle block logic to go through _unpickle_block instead
+    # for pandas <= 1.3.5
+    ("pandas.core.internals.blocks", "new_block"): (
+        "pandas._libs.internals",
+        "_unpickle_block",
+    ),
+    # Avoid Cython's warning "contradiction to to Python 'class private name' rules"
+    ("pandas._libs.tslibs.nattype", "__nat_unpickle"): (
+        "pandas._libs.tslibs.nattype",
+        "_nat_unpickle",
+    ),
+    # 50775, remove Int64Index, UInt64Index & Float64Index from codebase
+    ("pandas.core.indexes.numeric", "Int64Index"): (
+        "pandas.core.indexes.base",
+        "Index",
+    ),
+    ("pandas.core.indexes.numeric", "UInt64Index"): (
+        "pandas.core.indexes.base",
+        "Index",
+    ),
+    ("pandas.core.indexes.numeric", "Float64Index"): (
+        "pandas.core.indexes.base",
+        "Index",
+    ),
+    ("pandas.core.arrays.sparse.dtype", "SparseDtype"): (
+        "pandas.core.dtypes.dtypes",
+        "SparseDtype",
+    ),
+}
+
+
+# our Unpickler sub-class to override methods and some dispatcher
+# functions for compat and uses a non-public class of the pickle module.
+class Unpickler(pickle._Unpickler):
+    def find_class(self, module: str, name: str) -> Any:
+        key = (module, name)
+        module, name = _class_locations_map.get(key, key)
+        return super().find_class(module, name)
+
+    dispatch = pickle._Unpickler.dispatch.copy()
+
+    def load_reduce(self) -> None:
+        stack = self.stack  # type: ignore[attr-defined]
+        args = stack.pop()
+        func = stack[-1]
+
+        try:
+            stack[-1] = func(*args)
+        except TypeError:
+            # If we have a deprecated function,
+            # try to replace and try again.
+            if args and isinstance(args[0], type) and issubclass(args[0], BaseOffset):
+                # TypeError: object.__new__(Day) is not safe, use Day.__new__()
+                cls = args[0]
+                stack[-1] = cls.__new__(*args)
+                return
+            elif args and issubclass(args[0], PeriodArray):
+                cls = args[0]
+                stack[-1] = NDArrayBacked.__new__(*args)
+                return
+            raise
+
+    dispatch[pickle.REDUCE[0]] = load_reduce  # type: ignore[assignment]
+
+    def load_newobj(self) -> None:
+        args = self.stack.pop()  # type: ignore[attr-defined]
+        cls = self.stack.pop()  # type: ignore[attr-defined]
+
+        # compat
+        if issubclass(cls, DatetimeArray) and not args:
+            arr = np.array([], dtype="M8[ns]")
+            obj = cls.__new__(cls, arr, arr.dtype)
+        elif issubclass(cls, TimedeltaArray) and not args:
+            arr = np.array([], dtype="m8[ns]")
+            obj = cls.__new__(cls, arr, arr.dtype)
+        elif cls is BlockManager and not args:
+            obj = cls.__new__(cls, (), [], False)
+        else:
+            obj = cls.__new__(cls, *args)
+        self.append(obj)  # type: ignore[attr-defined]
+
+    dispatch[pickle.NEWOBJ[0]] = load_newobj  # type: ignore[assignment]
+
+
+def loads(
+    bytes_object: bytes,
+    *,
+    fix_imports: bool = True,
+    encoding: str = "ASCII",
+    errors: str = "strict",
+) -> Any:
+    """
+    Analogous to pickle._loads.
+    """
+    fd = io.BytesIO(bytes_object)
+    return Unpickler(
+        fd, fix_imports=fix_imports, encoding=encoding, errors=errors
+    ).load()
+
+
+@contextlib.contextmanager
+def patch_pickle() -> Generator[None]:
+    """
+    Temporarily patch pickle to use our unpickler.
+    """
+    orig_loads = pickle.loads
+    try:
+        setattr(pickle, "loads", loads)
+        yield
+    finally:
+        setattr(pickle, "loads", orig_loads)
diff --git a/pandas/pandas/compat/pyarrow.py b/pandas/pandas/compat/pyarrow.py
new file mode 100644
index 0000000000000..c501c06b93813
--- /dev/null
+++ b/pandas/pandas/compat/pyarrow.py
@@ -0,0 +1,35 @@
+"""support pyarrow compatibility across versions"""
+
+from __future__ import annotations
+
+from pandas.util.version import Version
+
+try:
+    import pyarrow as pa
+
+    _palv = Version(Version(pa.__version__).base_version)
+    pa_version_under10p1 = _palv < Version("10.0.1")
+    pa_version_under11p0 = _palv < Version("11.0.0")
+    pa_version_under12p0 = _palv < Version("12.0.0")
+    pa_version_under13p0 = _palv < Version("13.0.0")
+    pa_version_under14p0 = _palv < Version("14.0.0")
+    pa_version_under14p1 = _palv < Version("14.0.1")
+    pa_version_under15p0 = _palv < Version("15.0.0")
+    pa_version_under16p0 = _palv < Version("16.0.0")
+    pa_version_under17p0 = _palv < Version("17.0.0")
+    pa_version_under18p0 = _palv < Version("18.0.0")
+    pa_version_under19p0 = _palv < Version("19.0.0")
+    HAS_PYARROW = True
+except ImportError:
+    pa_version_under10p1 = True
+    pa_version_under11p0 = True
+    pa_version_under12p0 = True
+    pa_version_under13p0 = True
+    pa_version_under14p0 = True
+    pa_version_under14p1 = True
+    pa_version_under15p0 = True
+    pa_version_under16p0 = True
+    pa_version_under17p0 = True
+    pa_version_under18p0 = True
+    pa_version_under19p0 = True
+    HAS_PYARROW = False
diff --git a/pandas/pandas/conftest.py b/pandas/pandas/conftest.py
new file mode 100644
index 0000000000000..f9c10a7758bd2
--- /dev/null
+++ b/pandas/pandas/conftest.py
@@ -0,0 +1,2118 @@
+"""
+This file is very long and growing, but it was decided to not split it yet, as
+it's still manageable (2020-03-17, ~1.1k LoC). See gh-31989
+
+Instead of splitting it was decided to define sections here:
+- Configuration / Settings
+- Autouse fixtures
+- Common arguments
+- Missing values & co.
+- Classes
+- Indices
+- Series'
+- DataFrames
+- Operators & Operations
+- Data sets/files
+- Time zones
+- Dtypes
+- Misc
+"""
+
+from __future__ import annotations
+
+from collections import abc
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+    timezone,
+)
+from decimal import Decimal
+import gc
+import operator
+import os
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import uuid
+
+from dateutil.tz import (
+    tzlocal,
+    tzutc,
+)
+import hypothesis
+from hypothesis import strategies as st
+import numpy as np
+import pytest
+
+from pandas.compat._optional import import_optional_dependency
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    IntervalDtype,
+)
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Interval,
+    IntervalIndex,
+    Period,
+    RangeIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core import ops
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterator,
+    )
+
+try:
+    import pyarrow as pa
+except ImportError:
+    has_pyarrow = False
+else:
+    del pa
+    has_pyarrow = True
+
+pytz = import_optional_dependency("pytz", errors="ignore")
+
+
+# ----------------------------------------------------------------
+# Configuration / Settings
+# ----------------------------------------------------------------
+# pytest
+
+
+def pytest_addoption(parser) -> None:
+    parser.addoption(
+        "--no-strict-data-files",
+        action="store_false",
+        help="Don't fail if a test is skipped for missing data file.",
+    )
+
+
+def ignore_doctest_warning(item: pytest.Item, path: str, message: str) -> None:
+    """Ignore doctest warning.
+
+    Parameters
+    ----------
+    item : pytest.Item
+        pytest test item.
+    path : str
+        Module path to Python object, e.g. "pandas.DataFrame.append". A
+        warning will be filtered when item.name ends with in given path. So it is
+        sufficient to specify e.g. "DataFrame.append".
+    message : str
+        Message to be filtered.
+    """
+    if item.name.endswith(path):
+        item.add_marker(pytest.mark.filterwarnings(f"ignore:{message}"))
+
+
+def pytest_collection_modifyitems(items, config) -> None:
+    is_doctest = config.getoption("--doctest-modules") or config.getoption(
+        "--doctest-cython", default=False
+    )
+
+    # Warnings from doctests that can be ignored; place reason in comment above.
+    # Each entry specifies (path, message) - see the ignore_doctest_warning function
+    ignored_doctest_warnings = [
+        ("is_int64_dtype", "is_int64_dtype is deprecated"),
+        ("is_interval_dtype", "is_interval_dtype is deprecated"),
+        ("is_period_dtype", "is_period_dtype is deprecated"),
+        ("is_datetime64tz_dtype", "is_datetime64tz_dtype is deprecated"),
+        ("is_categorical_dtype", "is_categorical_dtype is deprecated"),
+        ("is_sparse", "is_sparse is deprecated"),
+        ("DataFrameGroupBy.fillna", "DataFrameGroupBy.fillna is deprecated"),
+        ("DataFrameGroupBy.corrwith", "DataFrameGroupBy.corrwith is deprecated"),
+        ("NDFrame.replace", "Series.replace without 'value'"),
+        ("NDFrame.clip", "Downcasting behavior in Series and DataFrame methods"),
+        ("Series.idxmin", "The behavior of Series.idxmin"),
+        ("Series.idxmax", "The behavior of Series.idxmax"),
+        ("SeriesGroupBy.fillna", "SeriesGroupBy.fillna is deprecated"),
+        ("SeriesGroupBy.idxmin", "The behavior of Series.idxmin"),
+        ("SeriesGroupBy.idxmax", "The behavior of Series.idxmax"),
+        ("to_pytimedelta", "The behavior of TimedeltaProperties.to_pytimedelta"),
+        ("NDFrame.reindex_like", "keyword argument 'method' is deprecated"),
+        # Docstring divides by zero to show behavior difference
+        ("missing.mask_zero_div_zero", "divide by zero encountered"),
+        (
+            "pandas.core.generic.NDFrame.first",
+            "first is deprecated and will be removed in a future version. "
+            "Please create a mask and filter using `.loc` instead",
+        ),
+        (
+            "Resampler.fillna",
+            "DatetimeIndexResampler.fillna is deprecated",
+        ),
+        (
+            "DataFrameGroupBy.fillna",
+            "DataFrameGroupBy.fillna with 'method' is deprecated",
+        ),
+        ("read_parquet", "Passing a BlockManager to DataFrame is deprecated"),
+    ]
+
+    if is_doctest:
+        for item in items:
+            for path, message in ignored_doctest_warnings:
+                ignore_doctest_warning(item, path, message)
+
+
+hypothesis_health_checks = [
+    hypothesis.HealthCheck.too_slow,
+    hypothesis.HealthCheck.differing_executors,
+]
+
+# Hypothesis
+hypothesis.settings.register_profile(
+    "ci",
+    # Hypothesis timing checks are tuned for scalars by default, so we bump
+    # them from 200ms to 500ms per test case as the global default.  If this
+    # is too short for a specific test, (a) try to make it faster, and (b)
+    # if it really is slow add `@settings(deadline=...)` with a working value,
+    # or `deadline=None` to entirely disable timeouts for that test.
+    # 2022-02-09: Changed deadline from 500 -> None. Deadline leads to
+    # non-actionable, flaky CI failures (# GH 24641, 44969, 45118, 44969)
+    deadline=None,
+    suppress_health_check=tuple(hypothesis_health_checks),
+)
+hypothesis.settings.load_profile("ci")
+
+# Registering these strategies makes them globally available via st.from_type,
+# which is use for offsets in tests/tseries/offsets/test_offsets_properties.py
+for name in "MonthBegin MonthEnd BMonthBegin BMonthEnd".split():
+    cls = getattr(pd.tseries.offsets, name)
+    st.register_type_strategy(
+        cls, st.builds(cls, n=st.integers(-99, 99), normalize=st.booleans())
+    )
+
+for name in "YearBegin YearEnd BYearBegin BYearEnd".split():
+    cls = getattr(pd.tseries.offsets, name)
+    st.register_type_strategy(
+        cls,
+        st.builds(
+            cls,
+            n=st.integers(-5, 5),
+            normalize=st.booleans(),
+            month=st.integers(min_value=1, max_value=12),
+        ),
+    )
+
+for name in "QuarterBegin QuarterEnd BQuarterBegin BQuarterEnd".split():
+    cls = getattr(pd.tseries.offsets, name)
+    st.register_type_strategy(
+        cls,
+        st.builds(
+            cls,
+            n=st.integers(-24, 24),
+            normalize=st.booleans(),
+            startingMonth=st.integers(min_value=1, max_value=12),
+        ),
+    )
+
+
+# ----------------------------------------------------------------
+# Autouse fixtures
+# ----------------------------------------------------------------
+
+
+# https://github.com/pytest-dev/pytest/issues/11873
+# Would like to avoid autouse=True, but cannot as of pytest 8.0.0
+@pytest.fixture(autouse=True)
+def add_doctest_imports(doctest_namespace) -> None:
+    """
+    Make `np` and `pd` names available for doctests.
+    """
+    doctest_namespace["np"] = np
+    doctest_namespace["pd"] = pd
+
+
+@pytest.fixture(autouse=True)
+def configure_tests() -> None:
+    """
+    Configure settings for all tests and test modules.
+    """
+    pd.set_option("chained_assignment", "raise")
+
+
+# ----------------------------------------------------------------
+# Common arguments
+# ----------------------------------------------------------------
+@pytest.fixture(params=[0, 1, "index", "columns"], ids=lambda x: f"axis={x!r}")
+def axis(request):
+    """
+    Fixture for returning the axis numbers of a DataFrame.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def observed(request):
+    """
+    Pass in the observed keyword to groupby for [True, False]
+    This indicates whether categoricals should return values for
+    values which are not in the grouper [False / None], or only values which
+    appear in the grouper [True]. [None] is supported for future compatibility
+    if we decide to change the default (and would need to warn if this
+    parameter is not passed).
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False, None])
+def ordered(request):
+    """
+    Boolean 'ordered' parameter for Categorical.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def dropna(request):
+    """
+    Boolean 'dropna' parameter.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def sort(request):
+    """
+    Boolean 'sort' parameter.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def skipna(request):
+    """
+    Boolean 'skipna' parameter.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["first", "last", False])
+def keep(request):
+    """
+    Valid values for the 'keep' parameter used in
+    .duplicated or .drop_duplicates
+    """
+    return request.param
+
+
+@pytest.fixture(params=["both", "neither", "left", "right"])
+def inclusive_endpoints_fixture(request):
+    """
+    Fixture for trying all interval 'inclusive' parameters.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["left", "right", "both", "neither"])
+def closed(request):
+    """
+    Fixture for trying all interval closed parameters.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["left", "right", "both", "neither"])
+def other_closed(request):
+    """
+    Secondary closed fixture to allow parametrizing over all pairs of closed.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        None,
+        "gzip",
+        "bz2",
+        "zip",
+        "xz",
+        "tar",
+        pytest.param("zstd", marks=td.skip_if_no("zstandard")),
+    ]
+)
+def compression(request):
+    """
+    Fixture for trying common compression types in compression tests.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        "gzip",
+        "bz2",
+        "zip",
+        "xz",
+        "tar",
+        pytest.param("zstd", marks=td.skip_if_no("zstandard")),
+    ]
+)
+def compression_only(request):
+    """
+    Fixture for trying common compression types in compression tests excluding
+    uncompressed case.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def writable(request):
+    """
+    Fixture that an array is writable.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["inner", "outer", "left", "right"])
+def join_type(request):
+    """
+    Fixture for trying all types of join operations.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["nlargest", "nsmallest"])
+def nselect_method(request):
+    """
+    Fixture for trying all nselect methods.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[None, "ignore"])
+def na_action(request):
+    """
+    Fixture for 'na_action' argument in map.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def ascending(request):
+    """
+    Fixture for 'na_action' argument in sort_values/sort_index/rank.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["average", "min", "max", "first", "dense"])
+def rank_method(request):
+    """
+    Fixture for 'rank' argument in rank.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def as_index(request):
+    """
+    Fixture for 'as_index' argument in groupby.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def cache(request):
+    """
+    Fixture for 'cache' argument in to_datetime.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def parallel(request):
+    """
+    Fixture for parallel keyword argument for numba.jit.
+    """
+    return request.param
+
+
+# Can parameterize nogil & nopython over True | False, but limiting per
+# https://github.com/pandas-dev/pandas/pull/41971#issuecomment-860607472
+
+
+@pytest.fixture(params=[False])
+def nogil(request):
+    """
+    Fixture for nogil keyword argument for numba.jit.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True])
+def nopython(request):
+    """
+    Fixture for nopython keyword argument for numba.jit.
+    """
+    return request.param
+
+
+# ----------------------------------------------------------------
+# Missing values & co.
+# ----------------------------------------------------------------
+@pytest.fixture(params=tm.NULL_OBJECTS, ids=lambda x: type(x).__name__)
+def nulls_fixture(request):
+    """
+    Fixture for each null type in pandas.
+    """
+    return request.param
+
+
+nulls_fixture2 = nulls_fixture  # Generate cartesian product of nulls_fixture
+
+
+@pytest.fixture(params=[None, np.nan, pd.NaT])
+def unique_nulls_fixture(request):
+    """
+    Fixture for each null type in pandas, each null type exactly once.
+    """
+    return request.param
+
+
+# Generate cartesian product of unique_nulls_fixture:
+unique_nulls_fixture2 = unique_nulls_fixture
+
+
+@pytest.fixture(params=tm.NP_NAT_OBJECTS, ids=lambda x: type(x).__name__)
+def np_nat_fixture(request):
+    """
+    Fixture for each NaT type in numpy.
+    """
+    return request.param
+
+
+# Generate cartesian product of np_nat_fixture:
+np_nat_fixture2 = np_nat_fixture
+
+
+# ----------------------------------------------------------------
+# Classes
+# ----------------------------------------------------------------
+
+
+@pytest.fixture(params=[DataFrame, Series])
+def frame_or_series(request):
+    """
+    Fixture to parametrize over DataFrame and Series.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[Index, Series], ids=["index", "series"])
+def index_or_series(request):
+    """
+    Fixture to parametrize over Index and Series, made necessary by a mypy
+    bug, giving an error:
+
+    List item 0 has incompatible type "Type[Series]"; expected "Type[PandasObject]"
+
+    See GH#29725
+    """
+    return request.param
+
+
+@pytest.fixture(params=[Index, Series, pd.array], ids=["index", "series", "array"])
+def index_or_series_or_array(request):
+    """
+    Fixture to parametrize over Index, Series, and ExtensionArray
+    """
+    return request.param
+
+
+@pytest.fixture(params=[Index, Series, DataFrame, pd.array], ids=lambda x: x.__name__)
+def box_with_array(request):
+    """
+    Fixture to test behavior for Index, Series, DataFrame, and pandas Array
+    classes
+    """
+    return request.param
+
+
+box_with_array2 = box_with_array
+
+
+@pytest.fixture
+def dict_subclass() -> type[dict]:
+    """
+    Fixture for a dictionary subclass.
+    """
+
+    class TestSubDict(dict):
+        def __init__(self, *args, **kwargs) -> None:
+            dict.__init__(self, *args, **kwargs)
+
+    return TestSubDict
+
+
+@pytest.fixture
+def non_dict_mapping_subclass() -> type[abc.Mapping]:
+    """
+    Fixture for a non-mapping dictionary subclass.
+    """
+
+    class TestNonDictMapping(abc.Mapping):
+        def __init__(self, underlying_dict) -> None:
+            self._data = underlying_dict
+
+        def __getitem__(self, key):
+            return self._data.__getitem__(key)
+
+        def __iter__(self) -> Iterator:
+            return self._data.__iter__()
+
+        def __len__(self) -> int:
+            return self._data.__len__()
+
+    return TestNonDictMapping
+
+
+# ----------------------------------------------------------------
+# Indices
+# ----------------------------------------------------------------
+@pytest.fixture
+def multiindex_year_month_day_dataframe_random_data():
+    """
+    DataFrame with 3 level MultiIndex (year, month, day) covering
+    first 100 business days from 2000-01-01 with random data
+    """
+    tdf = DataFrame(
+        np.random.default_rng(2).standard_normal((100, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=100, freq="B"),
+    )
+    ymd = tdf.groupby([lambda x: x.year, lambda x: x.month, lambda x: x.day]).sum()
+    # use int64 Index, to make sure things work
+    ymd.index = ymd.index.set_levels([lev.astype("i8") for lev in ymd.index.levels])
+    ymd.index.set_names(["year", "month", "day"], inplace=True)
+    return ymd
+
+
+@pytest.fixture
+def lexsorted_two_level_string_multiindex() -> MultiIndex:
+    """
+    2-level MultiIndex, lexsorted, with string names.
+    """
+    return MultiIndex(
+        levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],
+        codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+        names=["first", "second"],
+    )
+
+
+@pytest.fixture
+def multiindex_dataframe_random_data(
+    lexsorted_two_level_string_multiindex,
+) -> DataFrame:
+    """DataFrame with 2 level MultiIndex with random data"""
+    index = lexsorted_two_level_string_multiindex
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((10, 3)),
+        index=index,
+        columns=Index(["A", "B", "C"], name="exp"),
+    )
+
+
+def _create_multiindex():
+    """
+    MultiIndex used to test the general functionality of this object
+    """
+
+    # See Also: tests.multi.conftest.idx
+    major_axis = Index(["foo", "bar", "baz", "qux"])
+    minor_axis = Index(["one", "two"])
+
+    major_codes = np.array([0, 0, 1, 2, 3, 3])
+    minor_codes = np.array([0, 1, 0, 1, 0, 1])
+    index_names = ["first", "second"]
+    return MultiIndex(
+        levels=[major_axis, minor_axis],
+        codes=[major_codes, minor_codes],
+        names=index_names,
+        verify_integrity=False,
+    )
+
+
+def _create_mi_with_dt64tz_level():
+    """
+    MultiIndex with a level that is a tzaware DatetimeIndex.
+    """
+    # GH#8367 round trip with pickle
+    return MultiIndex.from_product(
+        [[1, 2], ["a", "b"], date_range("20130101", periods=3, tz="US/Eastern")],
+        names=["one", "two", "three"],
+    )
+
+
+indices_dict = {
+    "object": Index([f"pandas_{i}" for i in range(10)], dtype=object),
+    "string": Index([f"pandas_{i}" for i in range(10)], dtype="str"),
+    "datetime": date_range("2020-01-01", periods=10),
+    "datetime-tz": date_range("2020-01-01", periods=10, tz="US/Pacific"),
+    "period": period_range("2020-01-01", periods=10, freq="D"),
+    "timedelta": timedelta_range(start="1 day", periods=10, freq="D"),
+    "range": RangeIndex(10),
+    "int8": Index(np.arange(10), dtype="int8"),
+    "int16": Index(np.arange(10), dtype="int16"),
+    "int32": Index(np.arange(10), dtype="int32"),
+    "int64": Index(np.arange(10), dtype="int64"),
+    "uint8": Index(np.arange(10), dtype="uint8"),
+    "uint16": Index(np.arange(10), dtype="uint16"),
+    "uint32": Index(np.arange(10), dtype="uint32"),
+    "uint64": Index(np.arange(10), dtype="uint64"),
+    "float32": Index(np.arange(10), dtype="float32"),
+    "float64": Index(np.arange(10), dtype="float64"),
+    "bool-object": Index([True, False] * 5, dtype=object),
+    "bool-dtype": Index([True, False] * 5, dtype=bool),
+    "complex64": Index(
+        np.arange(10, dtype="complex64") + 1.0j * np.arange(10, dtype="complex64")
+    ),
+    "complex128": Index(
+        np.arange(10, dtype="complex128") + 1.0j * np.arange(10, dtype="complex128")
+    ),
+    "categorical": CategoricalIndex(list("abcd") * 2),
+    "interval": IntervalIndex.from_breaks(np.linspace(0, 100, num=11)),
+    "empty": Index([]),
+    "tuples": MultiIndex.from_tuples(zip(["foo", "bar", "baz"], [1, 2, 3])),
+    "mi-with-dt64tz-level": _create_mi_with_dt64tz_level(),
+    "multi": _create_multiindex(),
+    "repeats": Index([0, 0, 1, 1, 2, 2]),
+    "nullable_int": Index(np.arange(10), dtype="Int64"),
+    "nullable_uint": Index(np.arange(10), dtype="UInt16"),
+    "nullable_float": Index(np.arange(10), dtype="Float32"),
+    "nullable_bool": Index(np.arange(10).astype(bool), dtype="boolean"),
+    "string-python": Index(
+        pd.array([f"pandas_{i}" for i in range(10)], dtype="string[python]")
+    ),
+}
+if has_pyarrow:
+    idx = Index(pd.array([f"pandas_{i}" for i in range(10)], dtype="string[pyarrow]"))
+    indices_dict["string-pyarrow"] = idx
+
+
+@pytest.fixture(params=indices_dict.keys())
+def index(request):
+    """
+    Fixture for many "simple" kinds of indices.
+
+    These indices are unlikely to cover corner cases, e.g.
+        - no names
+        - no NaTs/NaNs
+        - no values near implementation bounds
+        - ...
+    """
+    # copy to avoid mutation, e.g. setting .name
+    return indices_dict[request.param].copy()
+
+
+@pytest.fixture(
+    params=[
+        key for key, value in indices_dict.items() if not isinstance(value, MultiIndex)
+    ]
+)
+def index_flat(request):
+    """
+    index fixture, but excluding MultiIndex cases.
+    """
+    key = request.param
+    return indices_dict[key].copy()
+
+
+@pytest.fixture(
+    params=[
+        key
+        for key, value in indices_dict.items()
+        if not (
+            key.startswith(("int", "uint", "float"))
+            or key in ["range", "empty", "repeats", "bool-dtype"]
+        )
+        and not isinstance(value, MultiIndex)
+    ]
+)
+def index_with_missing(request):
+    """
+    Fixture for indices with missing values.
+
+    Integer-dtype and empty cases are excluded because they cannot hold missing
+    values.
+
+    MultiIndex is excluded because isna() is not defined for MultiIndex.
+    """
+
+    # GH 35538. Use deep copy to avoid illusive bug on np-dev
+    # GHA pipeline that writes into indices_dict despite copy
+    ind = indices_dict[request.param].copy(deep=True)
+    vals = ind.values.copy()
+    if request.param in ["tuples", "mi-with-dt64tz-level", "multi"]:
+        # For setting missing values in the top level of MultiIndex
+        vals = ind.tolist()
+        vals[0] = (None,) + vals[0][1:]
+        vals[-1] = (None,) + vals[-1][1:]
+        return MultiIndex.from_tuples(vals)
+    else:
+        vals[0] = None
+        vals[-1] = None
+        return type(ind)(vals)
+
+
+# ----------------------------------------------------------------
+# Series'
+# ----------------------------------------------------------------
+@pytest.fixture
+def string_series() -> Series:
+    """
+    Fixture for Series of floats with Index of unique strings
+    """
+    return Series(
+        np.arange(30, dtype=np.float64) * 1.1,
+        index=Index([f"i_{i}" for i in range(30)]),
+        name="series",
+    )
+
+
+@pytest.fixture
+def object_series() -> Series:
+    """
+    Fixture for Series of dtype object with Index of unique strings
+    """
+    data = [f"foo_{i}" for i in range(30)]
+    index = Index([f"bar_{i}" for i in range(30)])
+    return Series(data, index=index, name="objects", dtype=object)
+
+
+@pytest.fixture
+def datetime_series() -> Series:
+    """
+    Fixture for Series of floats with DatetimeIndex
+    """
+    return Series(
+        np.random.default_rng(2).standard_normal(30),
+        index=date_range("2000-01-01", periods=30, freq="B"),
+        name="ts",
+    )
+
+
+def _create_series(index):
+    """Helper for the _series dict"""
+    size = len(index)
+    data = np.random.default_rng(2).standard_normal(size)
+    return Series(data, index=index, name="a", copy=False)
+
+
+_series = {
+    f"series-with-{index_id}-index": _create_series(index)
+    for index_id, index in indices_dict.items()
+}
+
+
+@pytest.fixture
+def series_with_simple_index(index) -> Series:
+    """
+    Fixture for tests on series with changing types of indices.
+    """
+    return _create_series(index)
+
+
+_narrow_series = {
+    f"{dtype.__name__}-series": Series(
+        range(30), index=[f"i-{i}" for i in range(30)], name="a", dtype=dtype
+    )
+    for dtype in tm.NARROW_NP_DTYPES
+}
+
+
+_index_or_series_objs = {**indices_dict, **_series, **_narrow_series}
+
+
+@pytest.fixture(params=_index_or_series_objs.keys())
+def index_or_series_obj(request):
+    """
+    Fixture for tests on indexes, series and series with a narrow dtype
+    copy to avoid mutation, e.g. setting .name
+    """
+    return _index_or_series_objs[request.param].copy(deep=True)
+
+
+_typ_objects_series = {
+    f"{dtype.__name__}-series": Series(dtype) for dtype in tm.PYTHON_DATA_TYPES
+}
+
+
+_index_or_series_memory_objs = {
+    **indices_dict,
+    **_series,
+    **_narrow_series,
+    **_typ_objects_series,
+}
+
+
+@pytest.fixture(params=_index_or_series_memory_objs.keys())
+def index_or_series_memory_obj(request):
+    """
+    Fixture for tests on indexes, series, series with a narrow dtype and
+    series with empty objects type
+    copy to avoid mutation, e.g. setting .name
+    """
+    return _index_or_series_memory_objs[request.param].copy(deep=True)
+
+
+# ----------------------------------------------------------------
+# DataFrames
+# ----------------------------------------------------------------
+@pytest.fixture
+def int_frame() -> DataFrame:
+    """
+    Fixture for DataFrame of ints with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D']
+    """
+    return DataFrame(
+        np.ones((30, 4), dtype=np.int64),
+        index=Index([f"foo_{i}" for i in range(30)]),
+        columns=Index(list("ABCD")),
+    )
+
+
+@pytest.fixture
+def float_frame() -> DataFrame:
+    """
+    Fixture for DataFrame of floats with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D'].
+    """
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        index=Index([f"foo_{i}" for i in range(30)]),
+        columns=Index(list("ABCD")),
+    )
+
+
+@pytest.fixture
+def rand_series_with_duplicate_datetimeindex() -> Series:
+    """
+    Fixture for Series with a DatetimeIndex that has duplicates.
+    """
+    dates = [
+        datetime(2000, 1, 2),
+        datetime(2000, 1, 2),
+        datetime(2000, 1, 2),
+        datetime(2000, 1, 3),
+        datetime(2000, 1, 3),
+        datetime(2000, 1, 3),
+        datetime(2000, 1, 4),
+        datetime(2000, 1, 4),
+        datetime(2000, 1, 4),
+        datetime(2000, 1, 5),
+    ]
+
+    return Series(np.random.default_rng(2).standard_normal(len(dates)), index=dates)
+
+
+# ----------------------------------------------------------------
+# Scalars
+# ----------------------------------------------------------------
+@pytest.fixture(
+    params=[
+        (Interval(left=0, right=5), IntervalDtype("int64", "right")),
+        (Interval(left=0.1, right=0.5), IntervalDtype("float64", "right")),
+        (Period("2012-01", freq="M"), "period[M]"),
+        (Period("2012-02-01", freq="D"), "period[D]"),
+        (
+            Timestamp("2011-01-01", tz="US/Eastern"),
+            DatetimeTZDtype(unit="s", tz="US/Eastern"),
+        ),
+        (Timedelta(seconds=500), "timedelta64[ns]"),
+    ]
+)
+def ea_scalar_and_dtype(request):
+    """
+    Fixture that tests each scalar and datetime type.
+    """
+    return request.param
+
+
+# ----------------------------------------------------------------
+# Operators & Operations
+# ----------------------------------------------------------------
+
+
+@pytest.fixture(params=tm.arithmetic_dunder_methods)
+def all_arithmetic_operators(request):
+    """
+    Fixture for dunder names for common arithmetic operations.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        operator.add,
+        ops.radd,
+        operator.sub,
+        ops.rsub,
+        operator.mul,
+        ops.rmul,
+        operator.truediv,
+        ops.rtruediv,
+        operator.floordiv,
+        ops.rfloordiv,
+        operator.mod,
+        ops.rmod,
+        operator.pow,
+        ops.rpow,
+        operator.eq,
+        operator.ne,
+        operator.lt,
+        operator.le,
+        operator.gt,
+        operator.ge,
+        operator.and_,
+        ops.rand_,
+        operator.xor,
+        ops.rxor,
+        operator.or_,
+        ops.ror_,
+    ]
+)
+def all_binary_operators(request):
+    """
+    Fixture for operator and roperator arithmetic, comparison, and logical ops.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        operator.add,
+        ops.radd,
+        operator.sub,
+        ops.rsub,
+        operator.mul,
+        ops.rmul,
+        operator.truediv,
+        ops.rtruediv,
+        operator.floordiv,
+        ops.rfloordiv,
+        operator.mod,
+        ops.rmod,
+        operator.pow,
+        ops.rpow,
+    ]
+)
+def all_arithmetic_functions(request):
+    """
+    Fixture for operator and roperator arithmetic functions.
+
+    Notes
+    -----
+    This includes divmod and rdivmod, whereas all_arithmetic_operators
+    does not.
+    """
+    return request.param
+
+
+_all_numeric_reductions = [
+    "count",
+    "sum",
+    "max",
+    "min",
+    "mean",
+    "prod",
+    "std",
+    "var",
+    "median",
+    "kurt",
+    "skew",
+    "sem",
+]
+
+
+@pytest.fixture(params=_all_numeric_reductions)
+def all_numeric_reductions(request):
+    """
+    Fixture for numeric reduction names.
+    """
+    return request.param
+
+
+_all_boolean_reductions = ["all", "any"]
+
+
+@pytest.fixture(params=_all_boolean_reductions)
+def all_boolean_reductions(request):
+    """
+    Fixture for boolean reduction names.
+    """
+    return request.param
+
+
+_all_reductions = _all_numeric_reductions + _all_boolean_reductions
+
+
+@pytest.fixture(params=_all_reductions)
+def all_reductions(request):
+    """
+    Fixture for all (boolean + numeric) reduction names.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        operator.eq,
+        operator.ne,
+        operator.gt,
+        operator.ge,
+        operator.lt,
+        operator.le,
+    ]
+)
+def comparison_op(request):
+    """
+    Fixture for operator module comparison functions.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["__le__", "__lt__", "__ge__", "__gt__"])
+def compare_operators_no_eq_ne(request):
+    """
+    Fixture for dunder names for compare operations except == and !=
+
+    * >=
+    * >
+    * <
+    * <=
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=["__and__", "__rand__", "__or__", "__ror__", "__xor__", "__rxor__"]
+)
+def all_logical_operators(request):
+    """
+    Fixture for dunder names for common logical operations
+
+    * |
+    * &
+    * ^
+    """
+    return request.param
+
+
+_all_numeric_accumulations = ["cumsum", "cumprod", "cummin", "cummax"]
+
+
+@pytest.fixture(params=_all_numeric_accumulations)
+def all_numeric_accumulations(request):
+    """
+    Fixture for numeric accumulation names
+    """
+    return request.param
+
+
+# ----------------------------------------------------------------
+# Data sets/files
+# ----------------------------------------------------------------
+@pytest.fixture
+def strict_data_files(pytestconfig):
+    """
+    Returns the configuration for the test setting `--no-strict-data-files`.
+    """
+    return pytestconfig.getoption("--no-strict-data-files")
+
+
+@pytest.fixture
+def datapath(strict_data_files: str) -> Callable[..., str]:
+    """
+    Get the path to a data file.
+
+    Parameters
+    ----------
+    path : str
+        Path to the file, relative to ``pandas/tests/``
+
+    Returns
+    -------
+    path including ``pandas/tests``.
+
+    Raises
+    ------
+    ValueError
+        If the path doesn't exist and the --no-strict-data-files option is not set.
+    """
+    BASE_PATH = os.path.join(os.path.dirname(__file__), "tests")
+
+    def deco(*args):
+        path = os.path.join(BASE_PATH, *args)
+        if not os.path.exists(path):
+            if strict_data_files:
+                raise ValueError(
+                    f"Could not find file {path} and --no-strict-data-files is not set."
+                )
+            pytest.skip(f"Could not find {path}.")
+        return path
+
+    return deco
+
+
+# ----------------------------------------------------------------
+# Time zones
+# ----------------------------------------------------------------
+TIMEZONES = [
+    None,
+    "UTC",
+    "US/Eastern",
+    "Asia/Tokyo",
+    "dateutil/US/Pacific",
+    "dateutil/Asia/Singapore",
+    "+01:15",
+    "-02:15",
+    "UTC+01:15",
+    "UTC-02:15",
+    tzutc(),
+    tzlocal(),
+    timezone.utc,
+    timezone(timedelta(hours=1)),
+    timezone(timedelta(hours=-1), name="foo"),
+]
+if pytz is not None:
+    TIMEZONES.extend(
+        (
+            pytz.FixedOffset(300),
+            pytz.FixedOffset(0),
+            pytz.FixedOffset(-300),
+            pytz.timezone("US/Pacific"),
+            pytz.timezone("UTC"),
+        )
+    )
+TIMEZONE_IDS = [repr(i) for i in TIMEZONES]
+
+
+@td.parametrize_fixture_doc(str(TIMEZONE_IDS))
+@pytest.fixture(params=TIMEZONES, ids=TIMEZONE_IDS)
+def tz_naive_fixture(request):
+    """
+    Fixture for trying timezones including default (None): {0}
+    """
+    return request.param
+
+
+@td.parametrize_fixture_doc(str(TIMEZONE_IDS[1:]))
+@pytest.fixture(params=TIMEZONES[1:], ids=TIMEZONE_IDS[1:])
+def tz_aware_fixture(request):
+    """
+    Fixture for trying explicit timezones: {0}
+    """
+    return request.param
+
+
+_UTCS = ["utc", "dateutil/UTC", tzutc(), timezone.utc]
+
+if pytz is not None:
+    _UTCS.append(pytz.utc)
+
+
+@pytest.fixture(params=_UTCS)
+def utc_fixture(request):
+    """
+    Fixture to provide variants of UTC timezone strings and tzinfo objects.
+    """
+    return request.param
+
+
+utc_fixture2 = utc_fixture
+
+
+@pytest.fixture(params=["s", "ms", "us", "ns"])
+def unit(request):
+    """
+    datetime64 units we support.
+    """
+    return request.param
+
+
+unit2 = unit
+
+
+# ----------------------------------------------------------------
+# Dtypes
+# ----------------------------------------------------------------
+@pytest.fixture(params=tm.STRING_DTYPES)
+def string_dtype(request):
+    """
+    Parametrized fixture for string dtypes.
+
+    * str
+    * 'str'
+    * 'U'
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ("python", pd.NA),
+        pytest.param(("pyarrow", pd.NA), marks=td.skip_if_no("pyarrow")),
+        pytest.param(("pyarrow", np.nan), marks=td.skip_if_no("pyarrow")),
+        ("python", np.nan),
+    ],
+    ids=[
+        "string=string[python]",
+        "string=string[pyarrow]",
+        "string=str[pyarrow]",
+        "string=str[python]",
+    ],
+)
+def string_dtype_no_object(request):
+    """
+    Parametrized fixture for string dtypes.
+    * 'string[python]' (NA variant)
+    * 'string[pyarrow]' (NA variant)
+    * 'str' (NaN variant, with pyarrow)
+    * 'str' (NaN variant, without pyarrow)
+    """
+    # need to instantiate the StringDtype here instead of in the params
+    # to avoid importing pyarrow during test collection
+    storage, na_value = request.param
+    return pd.StringDtype(storage, na_value)
+
+
+@pytest.fixture(
+    params=[
+        "string[python]",
+        pytest.param("string[pyarrow]", marks=td.skip_if_no("pyarrow")),
+    ]
+)
+def nullable_string_dtype(request):
+    """
+    Parametrized fixture for string dtypes.
+
+    * 'string[python]'
+    * 'string[pyarrow]'
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        pytest.param(("pyarrow", np.nan), marks=td.skip_if_no("pyarrow")),
+        pytest.param(("pyarrow", pd.NA), marks=td.skip_if_no("pyarrow")),
+    ]
+)
+def pyarrow_string_dtype(request):
+    """
+    Parametrized fixture for string dtypes backed by Pyarrow.
+
+    * 'str[pyarrow]'
+    * 'string[pyarrow]'
+    """
+    return pd.StringDtype(*request.param)
+
+
+@pytest.fixture(
+    params=[
+        "python",
+        pytest.param("pyarrow", marks=td.skip_if_no("pyarrow")),
+    ]
+)
+def string_storage(request):
+    """
+    Parametrized fixture for pd.options.mode.string_storage.
+
+    * 'python'
+    * 'pyarrow'
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ("python", pd.NA),
+        pytest.param(("pyarrow", pd.NA), marks=td.skip_if_no("pyarrow")),
+        pytest.param(("pyarrow", np.nan), marks=td.skip_if_no("pyarrow")),
+        ("python", np.nan),
+    ],
+    ids=[
+        "string=string[python]",
+        "string=string[pyarrow]",
+        "string=str[pyarrow]",
+        "string=str[python]",
+    ],
+)
+def string_dtype_arguments(request):
+    """
+    Parametrized fixture for StringDtype storage and na_value.
+
+    * 'python' + pd.NA
+    * 'pyarrow' + pd.NA
+    * 'pyarrow' + np.nan
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        "numpy_nullable",
+        pytest.param("pyarrow", marks=td.skip_if_no("pyarrow")),
+    ]
+)
+def dtype_backend(request):
+    """
+    Parametrized fixture for pd.options.mode.string_storage.
+
+    * 'python'
+    * 'pyarrow'
+    """
+    return request.param
+
+
+# Alias so we can test with cartesian product of string_storage
+string_storage2 = string_storage
+string_dtype_arguments2 = string_dtype_arguments
+
+
+@pytest.fixture(params=tm.BYTES_DTYPES)
+def bytes_dtype(request):
+    """
+    Parametrized fixture for bytes dtypes.
+
+    * bytes
+    * 'bytes'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.OBJECT_DTYPES)
+def object_dtype(request):
+    """
+    Parametrized fixture for object dtypes.
+
+    * object
+    * 'object'
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        np.dtype("object"),
+        ("python", pd.NA),
+        pytest.param(("pyarrow", pd.NA), marks=td.skip_if_no("pyarrow")),
+        pytest.param(("pyarrow", np.nan), marks=td.skip_if_no("pyarrow")),
+        ("python", np.nan),
+    ],
+    ids=[
+        "string=object",
+        "string=string[python]",
+        "string=string[pyarrow]",
+        "string=str[pyarrow]",
+        "string=str[python]",
+    ],
+)
+def any_string_dtype(request):
+    """
+    Parametrized fixture for string dtypes.
+    * 'object'
+    * 'string[python]' (NA variant)
+    * 'string[pyarrow]' (NA variant)
+    * 'str' (NaN variant, with pyarrow)
+    * 'str' (NaN variant, without pyarrow)
+    """
+    if isinstance(request.param, np.dtype):
+        return request.param
+    else:
+        # need to instantiate the StringDtype here instead of in the params
+        # to avoid importing pyarrow during test collection
+        storage, na_value = request.param
+        return pd.StringDtype(storage, na_value)
+
+
+@pytest.fixture(params=tm.DATETIME64_DTYPES)
+def datetime64_dtype(request):
+    """
+    Parametrized fixture for datetime64 dtypes.
+
+    * 'datetime64[ns]'
+    * 'M8[ns]'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.TIMEDELTA64_DTYPES)
+def timedelta64_dtype(request):
+    """
+    Parametrized fixture for timedelta64 dtypes.
+
+    * 'timedelta64[ns]'
+    * 'm8[ns]'
+    """
+    return request.param
+
+
+@pytest.fixture
+def fixed_now_ts() -> Timestamp:
+    """
+    Fixture emits fixed Timestamp.now()
+    """
+    return Timestamp(  # pyright: ignore[reportReturnType]
+        year=2021, month=1, day=1, hour=12, minute=4, second=13, microsecond=22
+    )
+
+
+@pytest.fixture(params=tm.FLOAT_NUMPY_DTYPES)
+def float_numpy_dtype(request):
+    """
+    Parameterized fixture for float dtypes.
+
+    * float
+    * 'float32'
+    * 'float64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.FLOAT_EA_DTYPES)
+def float_ea_dtype(request):
+    """
+    Parameterized fixture for float dtypes.
+
+    * 'Float32'
+    * 'Float64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_FLOAT_DTYPES)
+def any_float_dtype(request):
+    """
+    Parameterized fixture for float dtypes.
+
+    * float
+    * 'float32'
+    * 'float64'
+    * 'Float32'
+    * 'Float64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.COMPLEX_DTYPES)
+def complex_dtype(request):
+    """
+    Parameterized fixture for complex dtypes.
+
+    * complex
+    * 'complex64'
+    * 'complex128'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.COMPLEX_FLOAT_DTYPES)
+def complex_or_float_dtype(request):
+    """
+    Parameterized fixture for complex and numpy float dtypes.
+
+    * complex
+    * 'complex64'
+    * 'complex128'
+    * float
+    * 'float32'
+    * 'float64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.SIGNED_INT_NUMPY_DTYPES)
+def any_signed_int_numpy_dtype(request):
+    """
+    Parameterized fixture for signed integer dtypes.
+
+    * int
+    * 'int8'
+    * 'int16'
+    * 'int32'
+    * 'int64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.UNSIGNED_INT_NUMPY_DTYPES)
+def any_unsigned_int_numpy_dtype(request):
+    """
+    Parameterized fixture for unsigned integer dtypes.
+
+    * 'uint8'
+    * 'uint16'
+    * 'uint32'
+    * 'uint64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_INT_NUMPY_DTYPES)
+def any_int_numpy_dtype(request):
+    """
+    Parameterized fixture for any integer dtype.
+
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_INT_EA_DTYPES)
+def any_int_ea_dtype(request):
+    """
+    Parameterized fixture for any nullable integer dtype.
+
+    * 'UInt8'
+    * 'Int8'
+    * 'UInt16'
+    * 'Int16'
+    * 'UInt32'
+    * 'Int32'
+    * 'UInt64'
+    * 'Int64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_INT_DTYPES)
+def any_int_dtype(request):
+    """
+    Parameterized fixture for any nullable integer dtype.
+
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    * 'UInt8'
+    * 'Int8'
+    * 'UInt16'
+    * 'Int16'
+    * 'UInt32'
+    * 'Int32'
+    * 'UInt64'
+    * 'Int64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_INT_EA_DTYPES + tm.FLOAT_EA_DTYPES)
+def any_numeric_ea_dtype(request):
+    """
+    Parameterized fixture for any nullable integer dtype and
+    any float ea dtypes.
+
+    * 'UInt8'
+    * 'Int8'
+    * 'UInt16'
+    * 'Int16'
+    * 'UInt32'
+    * 'Int32'
+    * 'UInt64'
+    * 'Int64'
+    * 'Float32'
+    * 'Float64'
+    """
+    return request.param
+
+
+#  Unsupported operand types for + ("List[Union[str, ExtensionDtype, dtype[Any],
+#  Type[object]]]" and "List[str]")
+@pytest.fixture(
+    params=tm.ALL_INT_EA_DTYPES
+    + tm.FLOAT_EA_DTYPES
+    + tm.ALL_INT_PYARROW_DTYPES_STR_REPR
+    + tm.FLOAT_PYARROW_DTYPES_STR_REPR  # type: ignore[operator]
+)
+def any_numeric_ea_and_arrow_dtype(request):
+    """
+    Parameterized fixture for any nullable integer dtype and
+    any float ea dtypes.
+
+    * 'UInt8'
+    * 'Int8'
+    * 'UInt16'
+    * 'Int16'
+    * 'UInt32'
+    * 'Int32'
+    * 'UInt64'
+    * 'Int64'
+    * 'Float32'
+    * 'Float64'
+    * 'uint8[pyarrow]'
+    * 'int8[pyarrow]'
+    * 'uint16[pyarrow]'
+    * 'int16[pyarrow]'
+    * 'uint32[pyarrow]'
+    * 'int32[pyarrow]'
+    * 'uint64[pyarrow]'
+    * 'int64[pyarrow]'
+    * 'float32[pyarrow]'
+    * 'float64[pyarrow]'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.SIGNED_INT_EA_DTYPES)
+def any_signed_int_ea_dtype(request):
+    """
+    Parameterized fixture for any signed nullable integer dtype.
+
+    * 'Int8'
+    * 'Int16'
+    * 'Int32'
+    * 'Int64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_REAL_NUMPY_DTYPES)
+def any_real_numpy_dtype(request):
+    """
+    Parameterized fixture for any (purely) real numeric dtype.
+
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    * float
+    * 'float32'
+    * 'float64'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_REAL_DTYPES)
+def any_real_numeric_dtype(request):
+    """
+    Parameterized fixture for any (purely) real numeric dtype.
+
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    * float
+    * 'float32'
+    * 'float64'
+
+    and associated ea dtypes.
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_NUMPY_DTYPES)
+def any_numpy_dtype(request):
+    """
+    Parameterized fixture for all numpy dtypes.
+
+    * bool
+    * 'bool'
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    * float
+    * 'float32'
+    * 'float64'
+    * complex
+    * 'complex64'
+    * 'complex128'
+    * str
+    * 'str'
+    * 'U'
+    * bytes
+    * 'bytes'
+    * 'datetime64[ns]'
+    * 'M8[ns]'
+    * 'timedelta64[ns]'
+    * 'm8[ns]'
+    * object
+    * 'object'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_REAL_NULLABLE_DTYPES)
+def any_real_nullable_dtype(request):
+    """
+    Parameterized fixture for all real dtypes that can hold NA.
+
+    * float
+    * 'float32'
+    * 'float64'
+    * 'Float32'
+    * 'Float64'
+    * 'UInt8'
+    * 'UInt16'
+    * 'UInt32'
+    * 'UInt64'
+    * 'Int8'
+    * 'Int16'
+    * 'Int32'
+    * 'Int64'
+    * 'uint8[pyarrow]'
+    * 'uint16[pyarrow]'
+    * 'uint32[pyarrow]'
+    * 'uint64[pyarrow]'
+    * 'int8[pyarrow]'
+    * 'int16[pyarrow]'
+    * 'int32[pyarrow]'
+    * 'int64[pyarrow]'
+    * 'float[pyarrow]'
+    * 'double[pyarrow]'
+    """
+    return request.param
+
+
+@pytest.fixture(params=tm.ALL_NUMERIC_DTYPES)
+def any_numeric_dtype(request):
+    """
+    Parameterized fixture for all numeric dtypes.
+
+    * int
+    * 'int8'
+    * 'uint8'
+    * 'int16'
+    * 'uint16'
+    * 'int32'
+    * 'uint32'
+    * 'int64'
+    * 'uint64'
+    * float
+    * 'float32'
+    * 'float64'
+    * complex
+    * 'complex64'
+    * 'complex128'
+    * 'UInt8'
+    * 'Int8'
+    * 'UInt16'
+    * 'Int16'
+    * 'UInt32'
+    * 'Int32'
+    * 'UInt64'
+    * 'Int64'
+    * 'Float32'
+    * 'Float64'
+    """
+    return request.param
+
+
+# categoricals are handled separately
+_any_skipna_inferred_dtype = [
+    ("string", ["a", np.nan, "c"]),
+    ("string", ["a", pd.NA, "c"]),
+    ("mixed", ["a", pd.NaT, "c"]),  # pd.NaT not considered valid by is_string_array
+    ("bytes", [b"a", np.nan, b"c"]),
+    ("empty", [np.nan, np.nan, np.nan]),
+    ("empty", []),
+    ("mixed-integer", ["a", np.nan, 2]),
+    ("mixed", ["a", np.nan, 2.0]),
+    ("floating", [1.0, np.nan, 2.0]),
+    ("integer", [1, np.nan, 2]),
+    ("mixed-integer-float", [1, np.nan, 2.0]),
+    ("decimal", [Decimal(1), np.nan, Decimal(2)]),
+    ("boolean", [True, np.nan, False]),
+    ("boolean", [True, pd.NA, False]),
+    ("datetime64", [np.datetime64("2013-01-01"), np.nan, np.datetime64("2018-01-01")]),
+    ("datetime", [Timestamp("20130101"), np.nan, Timestamp("20180101")]),
+    ("date", [date(2013, 1, 1), np.nan, date(2018, 1, 1)]),
+    ("complex", [1 + 1j, np.nan, 2 + 2j]),
+    # The following dtype is commented out due to GH 23554
+    # ('timedelta64', [np.timedelta64(1, 'D'),
+    #                  np.nan, np.timedelta64(2, 'D')]),
+    ("timedelta", [timedelta(1), np.nan, timedelta(2)]),
+    ("time", [time(1), np.nan, time(2)]),
+    ("period", [Period(2013), pd.NaT, Period(2018)]),
+    ("interval", [Interval(0, 1), np.nan, Interval(0, 2)]),
+]
+ids, _ = zip(*_any_skipna_inferred_dtype)  # use inferred type as fixture-id
+
+
+@pytest.fixture(params=_any_skipna_inferred_dtype, ids=ids)
+def any_skipna_inferred_dtype(request):
+    """
+    Fixture for all inferred dtypes from _libs.lib.infer_dtype
+
+    The covered (inferred) types are:
+    * 'string'
+    * 'empty'
+    * 'bytes'
+    * 'mixed'
+    * 'mixed-integer'
+    * 'mixed-integer-float'
+    * 'floating'
+    * 'integer'
+    * 'decimal'
+    * 'boolean'
+    * 'datetime64'
+    * 'datetime'
+    * 'date'
+    * 'timedelta'
+    * 'time'
+    * 'period'
+    * 'interval'
+
+    Returns
+    -------
+    inferred_dtype : str
+        The string for the inferred dtype from _libs.lib.infer_dtype
+    values : np.ndarray
+        An array of object dtype that will be inferred to have
+        `inferred_dtype`
+
+    Examples
+    --------
+    >>> from pandas._libs import lib
+    >>>
+    >>> def test_something(any_skipna_inferred_dtype):
+    ...     inferred_dtype, values = any_skipna_inferred_dtype
+    ...     # will pass
+    ...     assert lib.infer_dtype(values, skipna=True) == inferred_dtype
+    """
+    inferred_dtype, values = request.param
+    values = np.array(values, dtype=object)  # object dtype to avoid casting
+
+    # correctness of inference tested in tests/dtypes/test_inference.py
+    return inferred_dtype, values
+
+
+# ----------------------------------------------------------------
+# Misc
+# ----------------------------------------------------------------
+@pytest.fixture
+def ip():
+    """
+    Get an instance of IPython.InteractiveShell.
+
+    Will raise a skip if IPython is not installed.
+    """
+    pytest.importorskip("IPython", minversion="6.0.0")
+    from IPython.core.interactiveshell import InteractiveShell
+
+    # GH#35711 make sure sqlite history file handle is not leaked
+    from traitlets.config import Config  # isort:skip
+
+    c = Config()
+    c.HistoryManager.hist_file = ":memory:"
+
+    return InteractiveShell(config=c)
+
+
+@pytest.fixture
+def mpl_cleanup():
+    """
+    Ensure Matplotlib is cleaned up around a test.
+
+    Before a test is run:
+
+    1) Set the backend to "template" to avoid requiring a GUI.
+
+    After a test is run:
+
+    1) Reset units registry
+    2) Reset rc_context
+    3) Close all figures
+
+    See matplotlib/testing/decorators.py#L24.
+    """
+    mpl = pytest.importorskip("matplotlib")
+    mpl_units = pytest.importorskip("matplotlib.units")
+    plt = pytest.importorskip("matplotlib.pyplot")
+    orig_units_registry = mpl_units.registry.copy()
+    try:
+        with mpl.rc_context():
+            mpl.use("template")
+            yield
+    finally:
+        mpl_units.registry.clear()
+        mpl_units.registry.update(orig_units_registry)
+        plt.close("all")
+        # https://matplotlib.org/stable/users/prev_whats_new/whats_new_3.6.0.html#garbage-collection-is-no-longer-run-on-figure-close  # noqa: E501
+        gc.collect(1)
+
+
+@pytest.fixture(
+    params=[
+        getattr(pd.offsets, o)
+        for o in pd.offsets.__all__
+        if issubclass(getattr(pd.offsets, o), pd.offsets.Tick) and o != "Tick"
+    ]
+)
+def tick_classes(request):
+    """
+    Fixture for Tick based datetime offsets available for a time series.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[None, lambda x: x])
+def sort_by_key(request):
+    """
+    Simple fixture for testing keys in sorting methods.
+    Tests None (no key) and the identity key.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ("foo", None, None),
+        ("Egon", "Venkman", None),
+        ("NCC1701D", "NCC1701D", "NCC1701D"),
+        # possibly-matching NAs
+        (np.nan, np.nan, np.nan),
+        (np.nan, pd.NaT, None),
+        (np.nan, pd.NA, None),
+        (pd.NA, pd.NA, pd.NA),
+    ]
+)
+def names(request) -> tuple[Hashable, Hashable, Hashable]:
+    """
+    A 3-tuple of names, the first two for operands, the last for a result.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.setitem, tm.loc, tm.iloc])
+def indexer_sli(request):
+    """
+    Parametrize over __setitem__, loc.__setitem__, iloc.__setitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.loc, tm.iloc])
+def indexer_li(request):
+    """
+    Parametrize over loc.__getitem__, iloc.__getitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.setitem, tm.iloc])
+def indexer_si(request):
+    """
+    Parametrize over __setitem__, iloc.__setitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.setitem, tm.loc])
+def indexer_sl(request):
+    """
+    Parametrize over __setitem__, loc.__setitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.at, tm.loc])
+def indexer_al(request):
+    """
+    Parametrize over at.__setitem__, loc.__setitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[tm.iat, tm.iloc])
+def indexer_ial(request):
+    """
+    Parametrize over iat.__setitem__, iloc.__setitem__
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def performance_warning(request) -> Iterator[bool | type[Warning]]:
+    """
+    Fixture to check if performance warnings are enabled. Either produces
+    ``PerformanceWarning`` if they are enabled, otherwise ``False``.
+    """
+    with pd.option_context("mode.performance_warnings", request.param):
+        yield pd.errors.PerformanceWarning if request.param else False
+
+
+@pytest.fixture
+def using_infer_string() -> bool:
+    """
+    Fixture to check if infer string option is enabled.
+    """
+    return pd.options.future.infer_string is True
+
+
+_warsaws: list[Any] = ["Europe/Warsaw", "dateutil/Europe/Warsaw"]
+if pytz is not None:
+    _warsaws.append(pytz.timezone("Europe/Warsaw"))
+
+
+@pytest.fixture(params=_warsaws)
+def warsaw(request) -> str:
+    """
+    tzinfo for Europe/Warsaw using pytz, dateutil, or zoneinfo.
+    """
+    return request.param
+
+
+@pytest.fixture
+def temp_file(tmp_path):
+    """
+    Generate a unique file for testing use. See link for removal policy.
+    https://docs.pytest.org/en/7.1.x/how-to/tmp_path.html#the-default-base-temporary-directory
+    """
+    file_path = tmp_path / str(uuid.uuid4())
+    file_path.touch()
+    return file_path
diff --git a/pandas/pandas/core/__init__.py b/pandas/pandas/core/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/_numba/__init__.py b/pandas/pandas/core/_numba/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/_numba/executor.py b/pandas/pandas/core/_numba/executor.py
new file mode 100644
index 0000000000000..3f3ebe8dbe023
--- /dev/null
+++ b/pandas/pandas/core/_numba/executor.py
@@ -0,0 +1,241 @@
+from __future__ import annotations
+
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+    from pandas._typing import Scalar
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.util.numba_ import jit_user_function
+
+
+@functools.cache
+def generate_apply_looper(func, nopython=True, nogil=True, parallel=False):
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+    nb_compat_func = jit_user_function(func)
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def nb_looper(values, axis, *args):
+        # Operate on the first row/col in order to get
+        # the output shape
+        if axis == 0:
+            first_elem = values[:, 0]
+            dim0 = values.shape[1]
+        else:
+            first_elem = values[0]
+            dim0 = values.shape[0]
+        res0 = nb_compat_func(first_elem, *args)
+        # Use np.asarray to get shape for
+        # https://github.com/numba/numba/issues/4202#issuecomment-1185981507
+        buf_shape = (dim0,) + np.atleast_1d(np.asarray(res0)).shape
+        if axis == 0:
+            buf_shape = buf_shape[::-1]
+        buff = np.empty(buf_shape)
+
+        if axis == 1:
+            buff[0] = res0
+            for i in numba.prange(1, values.shape[0]):
+                buff[i] = nb_compat_func(values[i], *args)
+        else:
+            buff[:, 0] = res0
+            for j in numba.prange(1, values.shape[1]):
+                buff[:, j] = nb_compat_func(values[:, j], *args)
+        return buff
+
+    return nb_looper
+
+
+@functools.cache
+def make_looper(func, result_dtype, is_grouped_kernel, nopython, nogil, parallel):
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    if is_grouped_kernel:
+
+        @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+        def column_looper(
+            values: np.ndarray,
+            labels: np.ndarray,
+            ngroups: int,
+            min_periods: int,
+            *args,
+        ):
+            result = np.empty((values.shape[0], ngroups), dtype=result_dtype)
+            na_positions = {}
+            for i in numba.prange(values.shape[0]):
+                output, na_pos = func(
+                    values[i], result_dtype, labels, ngroups, min_periods, *args
+                )
+                result[i] = output
+                if len(na_pos) > 0:
+                    na_positions[i] = np.array(na_pos)
+            return result, na_positions
+
+    else:
+
+        @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+        def column_looper(
+            values: np.ndarray,
+            start: np.ndarray,
+            end: np.ndarray,
+            min_periods: int,
+            *args,
+        ):
+            result = np.empty((values.shape[0], len(start)), dtype=result_dtype)
+            na_positions = {}
+            for i in numba.prange(values.shape[0]):
+                output, na_pos = func(
+                    values[i], result_dtype, start, end, min_periods, *args
+                )
+                result[i] = output
+                if len(na_pos) > 0:
+                    na_positions[i] = np.array(na_pos)
+            return result, na_positions
+
+    return column_looper
+
+
+default_dtype_mapping: dict[np.dtype, Any] = {
+    np.dtype("int8"): np.int64,
+    np.dtype("int16"): np.int64,
+    np.dtype("int32"): np.int64,
+    np.dtype("int64"): np.int64,
+    np.dtype("uint8"): np.uint64,
+    np.dtype("uint16"): np.uint64,
+    np.dtype("uint32"): np.uint64,
+    np.dtype("uint64"): np.uint64,
+    np.dtype("float32"): np.float64,
+    np.dtype("float64"): np.float64,
+    np.dtype("complex64"): np.complex128,
+    np.dtype("complex128"): np.complex128,
+}
+
+
+# TODO: Preserve complex dtypes
+
+float_dtype_mapping: dict[np.dtype, Any] = {
+    np.dtype("int8"): np.float64,
+    np.dtype("int16"): np.float64,
+    np.dtype("int32"): np.float64,
+    np.dtype("int64"): np.float64,
+    np.dtype("uint8"): np.float64,
+    np.dtype("uint16"): np.float64,
+    np.dtype("uint32"): np.float64,
+    np.dtype("uint64"): np.float64,
+    np.dtype("float32"): np.float64,
+    np.dtype("float64"): np.float64,
+    np.dtype("complex64"): np.float64,
+    np.dtype("complex128"): np.float64,
+}
+
+identity_dtype_mapping: dict[np.dtype, Any] = {
+    np.dtype("int8"): np.int8,
+    np.dtype("int16"): np.int16,
+    np.dtype("int32"): np.int32,
+    np.dtype("int64"): np.int64,
+    np.dtype("uint8"): np.uint8,
+    np.dtype("uint16"): np.uint16,
+    np.dtype("uint32"): np.uint32,
+    np.dtype("uint64"): np.uint64,
+    np.dtype("float32"): np.float32,
+    np.dtype("float64"): np.float64,
+    np.dtype("complex64"): np.complex64,
+    np.dtype("complex128"): np.complex128,
+}
+
+
+def generate_shared_aggregator(
+    func: Callable[..., Scalar],
+    dtype_mapping: dict[np.dtype, np.dtype],
+    is_grouped_kernel: bool,
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+):
+    """
+    Generate a Numba function that loops over the columns 2D object and applies
+    a 1D numba kernel over each column.
+
+    Parameters
+    ----------
+    func : function
+        aggregation function to be applied to each column
+    dtype_mapping: dict or None
+        If not None, maps a dtype to a result dtype.
+        Otherwise, will fall back to default mapping.
+    is_grouped_kernel: bool, default False
+        Whether func operates using the group labels (True)
+        or using starts/ends arrays
+
+        If true, you also need to pass the number of groups to this function
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+
+    # A wrapper around the looper function,
+    # to dispatch based on dtype since numba is unable to do that in nopython mode
+
+    # It also post-processes the values by inserting nans where number of observations
+    # is less than min_periods
+    # Cannot do this in numba nopython mode
+    # (you'll run into type-unification error when you cast int -> float)
+    def looper_wrapper(
+        values,
+        start=None,
+        end=None,
+        labels=None,
+        ngroups=None,
+        min_periods: int = 0,
+        **kwargs,
+    ):
+        result_dtype = dtype_mapping[values.dtype]
+        column_looper = make_looper(
+            func, result_dtype, is_grouped_kernel, nopython, nogil, parallel
+        )
+        # Need to unpack kwargs since numba only supports *args
+        if is_grouped_kernel:
+            result, na_positions = column_looper(
+                values, labels, ngroups, min_periods, *kwargs.values()
+            )
+        else:
+            result, na_positions = column_looper(
+                values, start, end, min_periods, *kwargs.values()
+            )
+        if result.dtype.kind == "i":
+            # Look if na_positions is not empty
+            # If so, convert the whole block
+            # This is OK since int dtype cannot hold nan,
+            # so if min_periods not satisfied for 1 col, it is not satisfied for
+            # all columns at that index
+            for na_pos in na_positions.values():
+                if len(na_pos) > 0:
+                    result = result.astype("float64")
+                    break
+        # TODO: Optimize this
+        for i, na_pos in na_positions.items():
+            if len(na_pos) > 0:
+                result[i, na_pos] = np.nan
+        return result
+
+    return looper_wrapper
diff --git a/pandas/pandas/core/_numba/extensions.py b/pandas/pandas/core/_numba/extensions.py
new file mode 100644
index 0000000000000..413fdafc7fd04
--- /dev/null
+++ b/pandas/pandas/core/_numba/extensions.py
@@ -0,0 +1,589 @@
+# Disable type checking for this module since numba's internals
+# are not typed, and we use numba's internals via its extension API
+# mypy: ignore-errors
+"""
+Utility classes/functions to let numba recognize
+pandas Index/Series/DataFrame
+
+Mostly vendored from https://github.com/numba/numba/blob/main/numba/tests/pdlike_usecase.py
+"""
+
+from __future__ import annotations
+
+from contextlib import contextmanager
+import operator
+from typing import TYPE_CHECKING
+
+import numba
+from numba import types
+from numba.core import cgutils
+from numba.core.datamodel import models
+from numba.core.extending import (
+    NativeValue,
+    box,
+    lower_builtin,
+    make_attribute_wrapper,
+    overload,
+    overload_attribute,
+    overload_method,
+    register_model,
+    type_callable,
+    typeof_impl,
+    unbox,
+)
+from numba.core.imputils import impl_ret_borrowed
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.indexes.base import Index
+from pandas.core.indexing import _iLocIndexer
+from pandas.core.internals import SingleBlockManager
+from pandas.core.series import Series
+
+if TYPE_CHECKING:
+    from pandas._typing import Self
+
+
+# Helper function to hack around fact that Index casts numpy string dtype to object
+#
+# Idea is to set an attribute on a Index called _numba_data
+# that is the original data, or the object data casted to numpy string dtype,
+# with a context manager that is unset afterwards
+@contextmanager
+def set_numba_data(index: Index):
+    numba_data = index._data
+    if numba_data.dtype in (object, "string"):
+        numba_data = np.asarray(numba_data)
+        if not lib.is_string_array(numba_data):
+            raise ValueError(
+                "The numba engine only supports using string or numeric column names"
+            )
+        numba_data = numba_data.astype("U")
+    try:
+        index._numba_data = numba_data
+        yield index
+    finally:
+        del index._numba_data
+
+
+# TODO: Range index support
+# (this currently lowers OK, but does not round-trip)
+class IndexType(types.Type):
+    """
+    The type class for Index objects.
+    """
+
+    def __init__(self, dtype, layout, pyclass: any) -> None:
+        self.pyclass = pyclass
+        name = f"index({dtype}, {layout})"
+        self.dtype = dtype
+        self.layout = layout
+        super().__init__(name)
+
+    @property
+    def key(self):
+        return self.pyclass, self.dtype, self.layout
+
+    @property
+    def as_array(self):
+        return types.Array(self.dtype, 1, self.layout)
+
+    def copy(self, dtype=None, ndim: int = 1, layout=None) -> Self:
+        assert ndim == 1
+        if dtype is None:
+            dtype = self.dtype
+        layout = layout or self.layout
+        return type(self)(dtype, layout, self.pyclass)
+
+
+class SeriesType(types.Type):
+    """
+    The type class for Series objects.
+    """
+
+    def __init__(self, dtype, index, namety) -> None:
+        assert isinstance(index, IndexType)
+        self.dtype = dtype
+        self.index = index
+        self.values = types.Array(self.dtype, 1, "C")
+        self.namety = namety
+        name = f"series({dtype}, {index}, {namety})"
+        super().__init__(name)
+
+    @property
+    def key(self):
+        return self.dtype, self.index, self.namety
+
+    @property
+    def as_array(self):
+        return self.values
+
+    def copy(self, dtype=None, ndim: int = 1, layout: str = "C") -> Self:
+        assert ndim == 1
+        assert layout == "C"
+        if dtype is None:
+            dtype = self.dtype
+        return type(self)(dtype, self.index, self.namety)
+
+
+@typeof_impl.register(Index)
+def typeof_index(val, c) -> IndexType:
+    """
+    This will assume that only strings are in object dtype
+    index.
+    (you should check this before this gets lowered down to numba)
+    """
+    # arrty = typeof_impl(val._data, c)
+    arrty = typeof_impl(val._numba_data, c)
+    assert arrty.ndim == 1
+    return IndexType(arrty.dtype, arrty.layout, type(val))
+
+
+@typeof_impl.register(Series)
+def typeof_series(val, c) -> SeriesType:
+    index = typeof_impl(val.index, c)
+    arrty = typeof_impl(val.values, c)
+    namety = typeof_impl(val.name, c)
+    assert arrty.ndim == 1
+    assert arrty.layout == "C"
+    return SeriesType(arrty.dtype, index, namety)
+
+
+@type_callable(Series)
+def type_series_constructor(context):
+    def typer(data, index, name=None):
+        if isinstance(index, IndexType) and isinstance(data, types.Array):
+            assert data.ndim == 1
+            if name is None:
+                name = types.intp
+            return SeriesType(data.dtype, index, name)
+
+    return typer
+
+
+@type_callable(Index)
+def type_index_constructor(context):
+    def typer(data, hashmap=None):
+        if isinstance(data, types.Array):
+            assert data.layout == "C"
+            assert data.ndim == 1
+            assert hashmap is None or isinstance(hashmap, types.DictType)
+            return IndexType(data.dtype, layout=data.layout, pyclass=Index)
+
+    return typer
+
+
+# Backend extensions for Index and Series and Frame
+@register_model(IndexType)
+class IndexModel(models.StructModel):
+    def __init__(self, dmm, fe_type) -> None:
+        # We don't want the numpy string scalar type in our hashmap
+        members = [
+            ("data", fe_type.as_array),
+            # This is an attempt to emulate our hashtable code with a numba
+            # typed dict
+            # It maps from values in the index to their integer positions in the array
+            ("hashmap", types.DictType(fe_type.dtype, types.intp)),
+            # Pointer to the Index object this was created from, or that it
+            # boxes to
+            # https://numba.discourse.group/t/qst-how-to-cache-the-boxing-of-an-object/2128/2?u=lithomas1
+            ("parent", types.pyobject),
+        ]
+        models.StructModel.__init__(self, dmm, fe_type, members)
+
+
+@register_model(SeriesType)
+class SeriesModel(models.StructModel):
+    def __init__(self, dmm, fe_type) -> None:
+        members = [
+            ("index", fe_type.index),
+            ("values", fe_type.as_array),
+            ("name", fe_type.namety),
+        ]
+        models.StructModel.__init__(self, dmm, fe_type, members)
+
+
+make_attribute_wrapper(IndexType, "data", "_data")
+make_attribute_wrapper(IndexType, "hashmap", "hashmap")
+
+make_attribute_wrapper(SeriesType, "index", "index")
+make_attribute_wrapper(SeriesType, "values", "values")
+make_attribute_wrapper(SeriesType, "name", "name")
+
+
+@lower_builtin(Series, types.Array, IndexType)
+def pdseries_constructor(context, builder, sig, args):
+    data, index = args
+    series = cgutils.create_struct_proxy(sig.return_type)(context, builder)
+    series.index = index
+    series.values = data
+    series.name = context.get_constant(types.intp, 0)
+    return impl_ret_borrowed(context, builder, sig.return_type, series._getvalue())
+
+
+@lower_builtin(Series, types.Array, IndexType, types.intp)
+@lower_builtin(Series, types.Array, IndexType, types.float64)
+@lower_builtin(Series, types.Array, IndexType, types.unicode_type)
+def pdseries_constructor_with_name(context, builder, sig, args):
+    data, index, name = args
+    series = cgutils.create_struct_proxy(sig.return_type)(context, builder)
+    series.index = index
+    series.values = data
+    series.name = name
+    return impl_ret_borrowed(context, builder, sig.return_type, series._getvalue())
+
+
+@lower_builtin(Index, types.Array, types.DictType, types.pyobject)
+def index_constructor_2arg(context, builder, sig, args):
+    (data, hashmap, parent) = args
+    index = cgutils.create_struct_proxy(sig.return_type)(context, builder)
+
+    index.data = data
+    index.hashmap = hashmap
+    index.parent = parent
+    return impl_ret_borrowed(context, builder, sig.return_type, index._getvalue())
+
+
+@lower_builtin(Index, types.Array, types.DictType)
+def index_constructor_2arg_parent(context, builder, sig, args):
+    # Basically same as index_constructor_1arg, but also lets you specify the
+    # parent object
+    (data, hashmap) = args
+    index = cgutils.create_struct_proxy(sig.return_type)(context, builder)
+
+    index.data = data
+    index.hashmap = hashmap
+    return impl_ret_borrowed(context, builder, sig.return_type, index._getvalue())
+
+
+@lower_builtin(Index, types.Array)
+def index_constructor_1arg(context, builder, sig, args):
+    from numba.typed import Dict
+
+    key_type = sig.return_type.dtype
+    value_type = types.intp
+
+    def index_impl(data):
+        return Index(data, Dict.empty(key_type, value_type))
+
+    return context.compile_internal(builder, index_impl, sig, args)
+
+
+# Helper to convert the unicodecharseq (numpy string scalar) into a unicode_type
+# (regular string)
+def maybe_cast_str(x):
+    # Dummy function that numba can overload
+    pass
+
+
+@overload(maybe_cast_str)
+def maybe_cast_str_impl(x):
+    """Converts numba UnicodeCharSeq (numpy string scalar) -> unicode type (string).
+    Is a no-op for other types."""
+    if isinstance(x, types.UnicodeCharSeq):
+        return lambda x: str(x)
+    else:
+        return lambda x: x
+
+
+@unbox(IndexType)
+def unbox_index(typ, obj, c):
+    """
+    Convert a Index object to a native structure.
+
+    Note: Object dtype is not allowed here
+    """
+    data_obj = c.pyapi.object_getattr_string(obj, "_numba_data")
+    index = cgutils.create_struct_proxy(typ)(c.context, c.builder)
+    # If we see an object array, assume its been validated as only containing strings
+    # We still need to do the conversion though
+    index.data = c.unbox(typ.as_array, data_obj).value
+    typed_dict_obj = c.pyapi.unserialize(c.pyapi.serialize_object(numba.typed.Dict))
+    # Create an empty typed dict in numba for the hashmap for indexing
+    # equiv of numba.typed.Dict.empty(typ.dtype, types.intp)
+    arr_type_obj = c.pyapi.unserialize(c.pyapi.serialize_object(typ.dtype))
+    intp_type_obj = c.pyapi.unserialize(c.pyapi.serialize_object(types.intp))
+    hashmap_obj = c.pyapi.call_method(
+        typed_dict_obj, "empty", (arr_type_obj, intp_type_obj)
+    )
+    index.hashmap = c.unbox(types.DictType(typ.dtype, types.intp), hashmap_obj).value
+    # Set the parent for speedy boxing.
+    index.parent = obj
+
+    # Decrefs
+    c.pyapi.decref(data_obj)
+    c.pyapi.decref(arr_type_obj)
+    c.pyapi.decref(intp_type_obj)
+    c.pyapi.decref(typed_dict_obj)
+
+    return NativeValue(index._getvalue())
+
+
+@unbox(SeriesType)
+def unbox_series(typ, obj, c):
+    """
+    Convert a Series object to a native structure.
+    """
+    index_obj = c.pyapi.object_getattr_string(obj, "index")
+    values_obj = c.pyapi.object_getattr_string(obj, "values")
+    name_obj = c.pyapi.object_getattr_string(obj, "name")
+
+    series = cgutils.create_struct_proxy(typ)(c.context, c.builder)
+    series.index = c.unbox(typ.index, index_obj).value
+    series.values = c.unbox(typ.values, values_obj).value
+    series.name = c.unbox(typ.namety, name_obj).value
+
+    # Decrefs
+    c.pyapi.decref(index_obj)
+    c.pyapi.decref(values_obj)
+    c.pyapi.decref(name_obj)
+
+    return NativeValue(series._getvalue())
+
+
+@box(IndexType)
+def box_index(typ, val, c):
+    """
+    Convert a native index structure to a Index object.
+
+    If our native index is of a numpy string dtype, we'll cast it to
+    object.
+    """
+    # First build a Numpy array object, then wrap it in a Index
+    index = cgutils.create_struct_proxy(typ)(c.context, c.builder, value=val)
+
+    res = cgutils.alloca_once_value(c.builder, index.parent)
+
+    # Does parent exist?
+    # (it means already boxed once, or Index same as original df.index or df.columns)
+    # xref https://github.com/numba/numba/blob/596e8a55334cc46854e3192766e643767bd7c934/numba/core/boxing.py#L593C17-L593C17
+    with c.builder.if_else(cgutils.is_not_null(c.builder, index.parent)) as (
+        has_parent,
+        otherwise,
+    ):
+        with has_parent:
+            c.pyapi.incref(index.parent)
+        with otherwise:
+            # TODO: preserve the original class for the index
+            # Also need preserve the name of the Index
+            # class_obj = c.pyapi.unserialize(c.pyapi.serialize_object(typ.pyclass))
+            class_obj = c.pyapi.unserialize(c.pyapi.serialize_object(Index))
+            array_obj = c.box(typ.as_array, index.data)
+            if isinstance(typ.dtype, types.UnicodeCharSeq):
+                # We converted to numpy string dtype, convert back
+                # to object since _simple_new won't do that for uss
+                object_str_obj = c.pyapi.unserialize(c.pyapi.serialize_object("object"))
+                array_obj = c.pyapi.call_method(array_obj, "astype", (object_str_obj,))
+                c.pyapi.decref(object_str_obj)
+            # this is basically Index._simple_new(array_obj, name_obj) in python
+            index_obj = c.pyapi.call_method(class_obj, "_simple_new", (array_obj,))
+            index.parent = index_obj
+            c.builder.store(index_obj, res)
+
+            # Decrefs
+            c.pyapi.decref(class_obj)
+            c.pyapi.decref(array_obj)
+    return c.builder.load(res)
+
+
+@box(SeriesType)
+def box_series(typ, val, c):
+    """
+    Convert a native series structure to a Series object.
+    """
+    series = cgutils.create_struct_proxy(typ)(c.context, c.builder, value=val)
+    series_const_obj = c.pyapi.unserialize(c.pyapi.serialize_object(Series._from_mgr))
+    mgr_const_obj = c.pyapi.unserialize(
+        c.pyapi.serialize_object(SingleBlockManager.from_array)
+    )
+    index_obj = c.box(typ.index, series.index)
+    array_obj = c.box(typ.as_array, series.values)
+    name_obj = c.box(typ.namety, series.name)
+    # This is basically equivalent of
+    # pd.Series(data=array_obj, index=index_obj)
+    # To improve perf, we will construct the Series from a manager
+    # object to avoid checks.
+    # We'll also set the name attribute manually to avoid validation
+    mgr_obj = c.pyapi.call_function_objargs(
+        mgr_const_obj,
+        (
+            array_obj,
+            index_obj,
+        ),
+    )
+    mgr_axes_obj = c.pyapi.object_getattr_string(mgr_obj, "axes")
+    # Series._constructor_from_mgr(mgr, axes)
+    series_obj = c.pyapi.call_function_objargs(
+        series_const_obj, (mgr_obj, mgr_axes_obj)
+    )
+    c.pyapi.object_setattr_string(series_obj, "_name", name_obj)
+
+    # Decrefs
+    c.pyapi.decref(series_const_obj)
+    c.pyapi.decref(mgr_axes_obj)
+    c.pyapi.decref(mgr_obj)
+    c.pyapi.decref(mgr_const_obj)
+    c.pyapi.decref(index_obj)
+    c.pyapi.decref(array_obj)
+    c.pyapi.decref(name_obj)
+
+    return series_obj
+
+
+# Add common series reductions (e.g. mean, sum),
+# and also add common binops (e.g. add, sub, mul, div)
+def generate_series_reduction(ser_reduction, ser_method):
+    @overload_method(SeriesType, ser_reduction)
+    def series_reduction(series):
+        def series_reduction_impl(series):
+            return ser_method(series.values)
+
+        return series_reduction_impl
+
+    return series_reduction
+
+
+def generate_series_binop(binop):
+    @overload(binop)
+    def series_binop(series1, value):
+        if isinstance(series1, SeriesType):
+            if isinstance(value, SeriesType):
+
+                def series_binop_impl(series1, series2):
+                    # TODO: Check index matching?
+                    return Series(
+                        binop(series1.values, series2.values),
+                        series1.index,
+                        series1.name,
+                    )
+
+                return series_binop_impl
+            else:
+
+                def series_binop_impl(series1, value):
+                    return Series(
+                        binop(series1.values, value), series1.index, series1.name
+                    )
+
+                return series_binop_impl
+
+    return series_binop
+
+
+series_reductions = [
+    ("sum", np.sum),
+    ("mean", np.mean),
+    # Disabled due to discrepancies between numba std. dev
+    # and pandas std. dev (no way to specify dof)
+    # ("std", np.std),
+    # ("var", np.var),
+    ("min", np.min),
+    ("max", np.max),
+]
+for reduction, reduction_method in series_reductions:
+    generate_series_reduction(reduction, reduction_method)
+
+series_binops = [operator.add, operator.sub, operator.mul, operator.truediv]
+
+for ser_binop in series_binops:
+    generate_series_binop(ser_binop)
+
+
+# get_loc on Index
+@overload_method(IndexType, "get_loc")
+def index_get_loc(index, item):
+    def index_get_loc_impl(index, item):
+        # Initialize the hash table if not initialized
+        if len(index.hashmap) == 0:
+            for i, val in enumerate(index._data):
+                index.hashmap[val] = i
+        return index.hashmap[item]
+
+    return index_get_loc_impl
+
+
+# Indexing for Series/Index
+@overload(operator.getitem)
+def series_indexing(series, item):
+    if isinstance(series, SeriesType):
+
+        def series_getitem(series, item):
+            loc = series.index.get_loc(item)
+            return series.iloc[loc]
+
+        return series_getitem
+
+
+@overload(operator.getitem)
+def index_indexing(index, idx):
+    if isinstance(index, IndexType):
+
+        def index_getitem(index, idx):
+            return index._data[idx]
+
+        return index_getitem
+
+
+class IlocType(types.Type):
+    def __init__(self, obj_type) -> None:
+        self.obj_type = obj_type
+        name = f"iLocIndexer({obj_type})"
+        super().__init__(name=name)
+
+    @property
+    def key(self):
+        return self.obj_type
+
+
+@typeof_impl.register(_iLocIndexer)
+def typeof_iloc(val, c) -> IlocType:
+    objtype = typeof_impl(val.obj, c)
+    return IlocType(objtype)
+
+
+@type_callable(_iLocIndexer)
+def type_iloc_constructor(context):
+    def typer(obj):
+        if isinstance(obj, SeriesType):
+            return IlocType(obj)
+
+    return typer
+
+
+@lower_builtin(_iLocIndexer, SeriesType)
+def iloc_constructor(context, builder, sig, args):
+    (obj,) = args
+    iloc_indexer = cgutils.create_struct_proxy(sig.return_type)(context, builder)
+    iloc_indexer.obj = obj
+    return impl_ret_borrowed(
+        context, builder, sig.return_type, iloc_indexer._getvalue()
+    )
+
+
+@register_model(IlocType)
+class ILocModel(models.StructModel):
+    def __init__(self, dmm, fe_type) -> None:
+        members = [("obj", fe_type.obj_type)]
+        models.StructModel.__init__(self, dmm, fe_type, members)
+
+
+make_attribute_wrapper(IlocType, "obj", "obj")
+
+
+@overload_attribute(SeriesType, "iloc")
+def series_iloc(series):
+    def get(series):
+        return _iLocIndexer(series)
+
+    return get
+
+
+@overload(operator.getitem)
+def iloc_getitem(iloc_indexer, i):
+    if isinstance(iloc_indexer, IlocType):
+
+        def getitem_impl(iloc_indexer, i):
+            return iloc_indexer.obj.values[i]
+
+        return getitem_impl
diff --git a/pandas/pandas/core/_numba/kernels/__init__.py b/pandas/pandas/core/_numba/kernels/__init__.py
new file mode 100644
index 0000000000000..6983711480455
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/__init__.py
@@ -0,0 +1,27 @@
+from pandas.core._numba.kernels.mean_ import (
+    grouped_mean,
+    sliding_mean,
+)
+from pandas.core._numba.kernels.min_max_ import (
+    grouped_min_max,
+    sliding_min_max,
+)
+from pandas.core._numba.kernels.sum_ import (
+    grouped_sum,
+    sliding_sum,
+)
+from pandas.core._numba.kernels.var_ import (
+    grouped_var,
+    sliding_var,
+)
+
+__all__ = [
+    "grouped_mean",
+    "grouped_min_max",
+    "grouped_sum",
+    "grouped_var",
+    "sliding_mean",
+    "sliding_min_max",
+    "sliding_sum",
+    "sliding_var",
+]
diff --git a/pandas/pandas/core/_numba/kernels/mean_.py b/pandas/pandas/core/_numba/kernels/mean_.py
new file mode 100644
index 0000000000000..2b59ea2fe12a5
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/mean_.py
@@ -0,0 +1,198 @@
+"""
+Numba 1D mean kernels that can be shared by
+* Dataframe / Series
+* groupby
+* rolling / expanding
+
+Mirrors pandas/_libs/window/aggregation.pyx
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numba
+import numpy as np
+
+from pandas.core._numba.kernels.shared import is_monotonic_increasing
+from pandas.core._numba.kernels.sum_ import grouped_kahan_sum
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def add_mean(
+    val: float,
+    nobs: int,
+    sum_x: float,
+    neg_ct: int,
+    compensation: float,
+    num_consecutive_same_value: int,
+    prev_value: float,
+) -> tuple[int, float, int, float, int, float]:
+    if not np.isnan(val):
+        nobs += 1
+        y = val - compensation
+        t = sum_x + y
+        compensation = t - sum_x - y
+        sum_x = t
+        if val < 0:
+            neg_ct += 1
+
+        if val == prev_value:
+            num_consecutive_same_value += 1
+        else:
+            num_consecutive_same_value = 1
+        prev_value = val
+
+    return nobs, sum_x, neg_ct, compensation, num_consecutive_same_value, prev_value
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def remove_mean(
+    val: float, nobs: int, sum_x: float, neg_ct: int, compensation: float
+) -> tuple[int, float, int, float]:
+    if not np.isnan(val):
+        nobs -= 1
+        y = -val - compensation
+        t = sum_x + y
+        compensation = t - sum_x - y
+        sum_x = t
+        if val < 0:
+            neg_ct -= 1
+    return nobs, sum_x, neg_ct, compensation
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def sliding_mean(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    start: np.ndarray,
+    end: np.ndarray,
+    min_periods: int,
+) -> tuple[np.ndarray, list[int]]:
+    N = len(start)
+    nobs = 0
+    sum_x = 0.0
+    neg_ct = 0
+    compensation_add = 0.0
+    compensation_remove = 0.0
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing(
+        start
+    ) and is_monotonic_increasing(end)
+
+    output = np.empty(N, dtype=result_dtype)
+
+    for i in range(N):
+        s = start[i]
+        e = end[i]
+        if i == 0 or not is_monotonic_increasing_bounds:
+            prev_value = values[s]
+            num_consecutive_same_value = 0
+
+            for j in range(s, e):
+                val = values[j]
+                (
+                    nobs,
+                    sum_x,
+                    neg_ct,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_mean(
+                    val,
+                    nobs,
+                    sum_x,
+                    neg_ct,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,  # pyright: ignore[reportArgumentType]
+                )
+        else:
+            for j in range(start[i - 1], s):
+                val = values[j]
+                nobs, sum_x, neg_ct, compensation_remove = remove_mean(
+                    val, nobs, sum_x, neg_ct, compensation_remove
+                )
+
+            for j in range(end[i - 1], e):
+                val = values[j]
+                (
+                    nobs,
+                    sum_x,
+                    neg_ct,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_mean(
+                    val,
+                    nobs,
+                    sum_x,
+                    neg_ct,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,  # pyright: ignore[reportArgumentType]
+                )
+
+        if nobs >= min_periods and nobs > 0:
+            result = sum_x / nobs
+            if num_consecutive_same_value >= nobs:
+                result = prev_value
+            elif neg_ct == 0 and result < 0:
+                result = 0
+            elif neg_ct == nobs and result > 0:
+                result = 0
+        else:
+            result = np.nan
+
+        output[i] = result
+
+        if not is_monotonic_increasing_bounds:
+            nobs = 0
+            sum_x = 0.0
+            neg_ct = 0
+            compensation_remove = 0.0
+
+    # na_position is empty list since float64 can already hold nans
+    # Do list comprehension, since numba cannot figure out that na_pos is
+    # empty list of ints on its own
+    na_pos = [0 for i in range(0)]
+    return output, na_pos
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def grouped_mean(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    labels: npt.NDArray[np.intp],
+    ngroups: int,
+    min_periods: int,
+    skipna: bool,
+) -> tuple[np.ndarray, list[int]]:
+    output, nobs_arr, comp_arr, consecutive_counts, prev_vals = grouped_kahan_sum(
+        values, result_dtype, labels, ngroups, skipna
+    )
+
+    # Post-processing, replace sums that don't satisfy min_periods
+    for lab in range(ngroups):
+        nobs = nobs_arr[lab]
+        num_consecutive_same_value = consecutive_counts[lab]
+        prev_value = prev_vals[lab]
+        sum_x = output[lab]
+        if nobs >= min_periods:
+            if num_consecutive_same_value >= nobs:
+                result = prev_value * nobs
+            else:
+                result = sum_x
+        else:
+            result = np.nan
+        result /= nobs
+        output[lab] = result
+
+    # na_position is empty list since float64 can already hold nans
+    # Do list comprehension, since numba cannot figure out that na_pos is
+    # empty list of ints on its own
+    na_pos = [0 for i in range(0)]
+    return output, na_pos
diff --git a/pandas/pandas/core/_numba/kernels/min_max_.py b/pandas/pandas/core/_numba/kernels/min_max_.py
new file mode 100644
index 0000000000000..68aa1446bbe3c
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/min_max_.py
@@ -0,0 +1,130 @@
+"""
+Numba 1D min/max kernels that can be shared by
+* Dataframe / Series
+* groupby
+* rolling / expanding
+
+Mirrors pandas/_libs/window/aggregation.pyx
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numba
+import numpy as np
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def sliding_min_max(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    start: np.ndarray,
+    end: np.ndarray,
+    min_periods: int,
+    is_max: bool,
+) -> tuple[np.ndarray, list[int]]:
+    N = len(start)
+    nobs = 0
+    output = np.empty(N, dtype=result_dtype)
+    na_pos = []
+    # Use deque once numba supports it
+    # https://github.com/numba/numba/issues/7417
+    Q: list = []
+    W: list = []
+    for i in range(N):
+        curr_win_size = end[i] - start[i]
+        if i == 0:
+            st = start[i]
+        else:
+            st = end[i - 1]
+
+        for k in range(st, end[i]):
+            ai = values[k]
+            if not np.isnan(ai):
+                nobs += 1
+            elif is_max:
+                ai = -np.inf
+            else:
+                ai = np.inf
+            # Discard previous entries if we find new min or max
+            if is_max:
+                while Q and ((ai >= values[Q[-1]]) or values[Q[-1]] != values[Q[-1]]):
+                    Q.pop()
+            else:
+                while Q and ((ai <= values[Q[-1]]) or values[Q[-1]] != values[Q[-1]]):
+                    Q.pop()
+            Q.append(k)
+            W.append(k)
+
+        # Discard entries outside and left of current window
+        while Q and Q[0] <= start[i] - 1:
+            Q.pop(0)
+        while W and W[0] <= start[i] - 1:
+            if not np.isnan(values[W[0]]):
+                nobs -= 1
+            W.pop(0)
+
+        # Save output based on index in input value array
+        if Q and curr_win_size > 0 and nobs >= min_periods:
+            output[i] = values[Q[0]]
+        else:
+            if values.dtype.kind != "i":
+                output[i] = np.nan
+            else:
+                na_pos.append(i)
+
+    return output, na_pos
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def grouped_min_max(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    labels: npt.NDArray[np.intp],
+    ngroups: int,
+    min_periods: int,
+    is_max: bool,
+    skipna: bool = True,
+) -> tuple[np.ndarray, list[int]]:
+    N = len(labels)
+    nobs = np.zeros(ngroups, dtype=np.int64)
+    na_pos = []
+    output = np.empty(ngroups, dtype=result_dtype)
+
+    for i in range(N):
+        lab = labels[i]
+        val = values[i]
+        if lab < 0 or (not skipna and nobs[lab] >= 1 and np.isnan(output[lab])):
+            continue
+
+        if values.dtype.kind == "i" or not np.isnan(val):
+            nobs[lab] += 1
+        else:
+            if not skipna:
+                # If skipna is False and we encounter a NaN,
+                # both min and max of the group will be NaN
+                output[lab] = np.nan
+            continue
+
+        if nobs[lab] == 1:
+            # First element in group, set output equal to this
+            output[lab] = val
+            continue
+
+        if is_max:
+            if val > output[lab]:
+                output[lab] = val
+        else:
+            if val < output[lab]:
+                output[lab] = val
+
+    # Set labels that don't satisfy min_periods as np.nan
+    for lab, count in enumerate(nobs):
+        if count < min_periods:
+            na_pos.append(lab)
+
+    return output, na_pos
diff --git a/pandas/pandas/core/_numba/kernels/shared.py b/pandas/pandas/core/_numba/kernels/shared.py
new file mode 100644
index 0000000000000..c52372fe6b08f
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/shared.py
@@ -0,0 +1,29 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numba
+
+if TYPE_CHECKING:
+    import numpy as np
+
+
+@numba.jit(
+    # error: Any? not callable
+    numba.boolean(numba.int64[:]),  # type: ignore[misc]
+    nopython=True,
+    nogil=True,
+    parallel=False,
+)
+def is_monotonic_increasing(bounds: np.ndarray) -> bool:
+    """Check if int64 values are monotonically increasing."""
+    n = len(bounds)
+    if n < 2:
+        return True
+    prev = bounds[0]
+    for i in range(1, n):
+        cur = bounds[i]
+        if cur < prev:
+            return False
+        prev = cur
+    return True
diff --git a/pandas/pandas/core/_numba/kernels/sum_.py b/pandas/pandas/core/_numba/kernels/sum_.py
new file mode 100644
index 0000000000000..9f2e9541b31d0
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/sum_.py
@@ -0,0 +1,255 @@
+"""
+Numba 1D sum kernels that can be shared by
+* Dataframe / Series
+* groupby
+* rolling / expanding
+
+Mirrors pandas/_libs/window/aggregation.pyx
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numba
+from numba.extending import register_jitable
+import numpy as np
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+from pandas.core._numba.kernels.shared import is_monotonic_increasing
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def add_sum(
+    val: Any,
+    nobs: int,
+    sum_x: Any,
+    compensation: Any,
+    num_consecutive_same_value: int,
+    prev_value: Any,
+) -> tuple[int, Any, Any, int, Any]:
+    if not np.isnan(val):
+        nobs += 1
+        y = val - compensation
+        t = sum_x + y
+        compensation = t - sum_x - y
+        sum_x = t
+
+        if val == prev_value:
+            num_consecutive_same_value += 1
+        else:
+            num_consecutive_same_value = 1
+        prev_value = val
+
+    return nobs, sum_x, compensation, num_consecutive_same_value, prev_value
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def remove_sum(
+    val: Any, nobs: int, sum_x: Any, compensation: Any
+) -> tuple[int, Any, Any]:
+    if not np.isnan(val):
+        nobs -= 1
+        y = -val - compensation
+        t = sum_x + y
+        compensation = t - sum_x - y
+        sum_x = t
+    return nobs, sum_x, compensation
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def sliding_sum(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    start: np.ndarray,
+    end: np.ndarray,
+    min_periods: int,
+) -> tuple[np.ndarray, list[int]]:
+    dtype = values.dtype
+
+    na_val: object = np.nan
+    if dtype.kind == "i":
+        na_val = 0
+
+    N = len(start)
+    nobs = 0
+    sum_x = 0
+    compensation_add = 0
+    compensation_remove = 0
+    na_pos = []
+
+    is_monotonic_increasing_bounds = is_monotonic_increasing(
+        start
+    ) and is_monotonic_increasing(end)
+
+    output = np.empty(N, dtype=result_dtype)
+
+    for i in range(N):
+        s = start[i]
+        e = end[i]
+        if i == 0 or not is_monotonic_increasing_bounds:
+            prev_value = values[s]
+            num_consecutive_same_value = 0
+
+            for j in range(s, e):
+                val = values[j]
+                (
+                    nobs,
+                    sum_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_sum(
+                    val,
+                    nobs,
+                    sum_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                )
+        else:
+            for j in range(start[i - 1], s):
+                val = values[j]
+                nobs, sum_x, compensation_remove = remove_sum(
+                    val, nobs, sum_x, compensation_remove
+                )
+
+            for j in range(end[i - 1], e):
+                val = values[j]
+                (
+                    nobs,
+                    sum_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_sum(
+                    val,
+                    nobs,
+                    sum_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                )
+
+        if nobs == 0 == min_periods:
+            result: object = 0
+        elif nobs >= min_periods:
+            if num_consecutive_same_value >= nobs:
+                result = prev_value * nobs
+            else:
+                result = sum_x
+        else:
+            result = na_val
+            if dtype.kind == "i":
+                na_pos.append(i)
+
+        output[i] = result
+
+        if not is_monotonic_increasing_bounds:
+            nobs = 0
+            sum_x = 0
+            compensation_remove = 0
+
+    return output, na_pos
+
+
+# Mypy/pyright don't like the fact that the decorator is untyped
+@register_jitable  # type: ignore[misc]
+def grouped_kahan_sum(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    labels: npt.NDArray[np.intp],
+    ngroups: int,
+    skipna: bool,
+) -> tuple[
+    np.ndarray, npt.NDArray[np.int64], np.ndarray, npt.NDArray[np.int64], np.ndarray
+]:
+    N = len(labels)
+
+    nobs_arr = np.zeros(ngroups, dtype=np.int64)
+    comp_arr = np.zeros(ngroups, dtype=values.dtype)
+    consecutive_counts = np.zeros(ngroups, dtype=np.int64)
+    prev_vals = np.zeros(ngroups, dtype=values.dtype)
+    output = np.zeros(ngroups, dtype=result_dtype)
+
+    for i in range(N):
+        lab = labels[i]
+        val = values[i]
+
+        if lab < 0 or np.isnan(output[lab]):
+            continue
+
+        if not skipna and np.isnan(val):
+            output[lab] = np.nan
+            nobs_arr[lab] += 1
+            comp_arr[lab] = np.nan
+            consecutive_counts[lab] = 1
+            prev_vals[lab] = np.nan
+            continue
+
+        sum_x = output[lab]
+        nobs = nobs_arr[lab]
+        compensation_add = comp_arr[lab]
+        num_consecutive_same_value = consecutive_counts[lab]
+        prev_value = prev_vals[lab]
+
+        (
+            nobs,
+            sum_x,
+            compensation_add,
+            num_consecutive_same_value,
+            prev_value,
+        ) = add_sum(
+            val,
+            nobs,
+            sum_x,
+            compensation_add,
+            num_consecutive_same_value,
+            prev_value,
+        )
+
+        output[lab] = sum_x
+        consecutive_counts[lab] = num_consecutive_same_value
+        prev_vals[lab] = prev_value
+        comp_arr[lab] = compensation_add
+        nobs_arr[lab] = nobs
+    return output, nobs_arr, comp_arr, consecutive_counts, prev_vals
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def grouped_sum(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    labels: npt.NDArray[np.intp],
+    ngroups: int,
+    min_periods: int,
+    skipna: bool,
+) -> tuple[np.ndarray, list[int]]:
+    na_pos = []
+
+    output, nobs_arr, comp_arr, consecutive_counts, prev_vals = grouped_kahan_sum(
+        values, result_dtype, labels, ngroups, skipna
+    )
+
+    # Post-processing, replace sums that don't satisfy min_periods
+    for lab in range(ngroups):
+        nobs = nobs_arr[lab]
+        num_consecutive_same_value = consecutive_counts[lab]
+        prev_value = prev_vals[lab]
+        sum_x = output[lab]
+        if nobs >= min_periods:
+            if num_consecutive_same_value >= nobs:
+                result = prev_value * nobs
+            else:
+                result = sum_x
+        else:
+            result = sum_x  # Don't change val, will be replaced by nan later
+            na_pos.append(lab)
+        output[lab] = result
+
+    return output, na_pos
diff --git a/pandas/pandas/core/_numba/kernels/var_.py b/pandas/pandas/core/_numba/kernels/var_.py
new file mode 100644
index 0000000000000..5d720c877815d
--- /dev/null
+++ b/pandas/pandas/core/_numba/kernels/var_.py
@@ -0,0 +1,251 @@
+"""
+Numba 1D var kernels that can be shared by
+* Dataframe / Series
+* groupby
+* rolling / expanding
+
+Mirrors pandas/_libs/window/aggregation.pyx
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numba
+import numpy as np
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+from pandas.core._numba.kernels.shared import is_monotonic_increasing
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def add_var(
+    val: float,
+    nobs: int,
+    mean_x: float,
+    ssqdm_x: float,
+    compensation: float,
+    num_consecutive_same_value: int,
+    prev_value: float,
+) -> tuple[int, float, float, float, int, float]:
+    if not np.isnan(val):
+        if val == prev_value:
+            num_consecutive_same_value += 1
+        else:
+            num_consecutive_same_value = 1
+        prev_value = val
+
+        nobs += 1
+        prev_mean = mean_x - compensation
+        y = val - compensation
+        t = y - mean_x
+        compensation = t + mean_x - y
+        delta = t
+        if nobs:
+            mean_x += delta / nobs
+        else:
+            mean_x = 0
+        ssqdm_x += (val - prev_mean) * (val - mean_x)
+    return nobs, mean_x, ssqdm_x, compensation, num_consecutive_same_value, prev_value
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def remove_var(
+    val: float, nobs: int, mean_x: float, ssqdm_x: float, compensation: float
+) -> tuple[int, float, float, float]:
+    if not np.isnan(val):
+        nobs -= 1
+        if nobs:
+            prev_mean = mean_x - compensation
+            y = val - compensation
+            t = y - mean_x
+            compensation = t + mean_x - y
+            delta = t
+            mean_x -= delta / nobs
+            ssqdm_x -= (val - prev_mean) * (val - mean_x)
+        else:
+            mean_x = 0
+            ssqdm_x = 0
+    return nobs, mean_x, ssqdm_x, compensation
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def sliding_var(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    start: np.ndarray,
+    end: np.ndarray,
+    min_periods: int,
+    ddof: int = 1,
+) -> tuple[np.ndarray, list[int]]:
+    N = len(start)
+    nobs = 0
+    mean_x = 0.0
+    ssqdm_x = 0.0
+    compensation_add = 0.0
+    compensation_remove = 0.0
+
+    min_periods = max(min_periods, 1)
+    is_monotonic_increasing_bounds = is_monotonic_increasing(
+        start
+    ) and is_monotonic_increasing(end)
+
+    output = np.empty(N, dtype=result_dtype)
+
+    for i in range(N):
+        s = start[i]
+        e = end[i]
+        if i == 0 or not is_monotonic_increasing_bounds:
+            prev_value = values[s]
+            num_consecutive_same_value = 0
+
+            for j in range(s, e):
+                val = values[j]
+                (
+                    nobs,
+                    mean_x,
+                    ssqdm_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_var(
+                    val,
+                    nobs,
+                    mean_x,
+                    ssqdm_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                )
+        else:
+            for j in range(start[i - 1], s):
+                val = values[j]
+                nobs, mean_x, ssqdm_x, compensation_remove = remove_var(
+                    val, nobs, mean_x, ssqdm_x, compensation_remove
+                )
+
+            for j in range(end[i - 1], e):
+                val = values[j]
+                (
+                    nobs,
+                    mean_x,
+                    ssqdm_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                ) = add_var(
+                    val,
+                    nobs,
+                    mean_x,
+                    ssqdm_x,
+                    compensation_add,
+                    num_consecutive_same_value,
+                    prev_value,
+                )
+
+        if nobs >= min_periods and nobs > ddof:
+            if nobs == 1 or num_consecutive_same_value >= nobs:
+                result = 0.0
+            else:
+                result = ssqdm_x / (nobs - ddof)
+        else:
+            result = np.nan
+
+        output[i] = result
+
+        if not is_monotonic_increasing_bounds:
+            nobs = 0
+            mean_x = 0.0
+            ssqdm_x = 0.0
+            compensation_remove = 0.0
+
+    # na_position is empty list since float64 can already hold nans
+    # Do list comprehension, since numba cannot figure out that na_pos is
+    # empty list of ints on its own
+    na_pos = [0 for i in range(0)]
+    return output, na_pos
+
+
+@numba.jit(nopython=True, nogil=True, parallel=False)
+def grouped_var(
+    values: np.ndarray,
+    result_dtype: np.dtype,
+    labels: npt.NDArray[np.intp],
+    ngroups: int,
+    min_periods: int,
+    ddof: int = 1,
+    skipna: bool = True,
+) -> tuple[np.ndarray, list[int]]:
+    N = len(labels)
+
+    nobs_arr = np.zeros(ngroups, dtype=np.int64)
+    comp_arr = np.zeros(ngroups, dtype=values.dtype)
+    consecutive_counts = np.zeros(ngroups, dtype=np.int64)
+    prev_vals = np.zeros(ngroups, dtype=values.dtype)
+    output = np.zeros(ngroups, dtype=result_dtype)
+    means = np.zeros(ngroups, dtype=result_dtype)
+
+    for i in range(N):
+        lab = labels[i]
+        val = values[i]
+
+        if lab < 0 or np.isnan(output[lab]):
+            continue
+
+        if not skipna and np.isnan(val):
+            output[lab] = np.nan
+            continue
+
+        mean_x = means[lab]
+        ssqdm_x = output[lab]
+        nobs = nobs_arr[lab]
+        compensation_add = comp_arr[lab]
+        num_consecutive_same_value = consecutive_counts[lab]
+        prev_value = prev_vals[lab]
+
+        (
+            nobs,
+            mean_x,
+            ssqdm_x,
+            compensation_add,
+            num_consecutive_same_value,
+            prev_value,
+        ) = add_var(
+            val,
+            nobs,
+            mean_x,
+            ssqdm_x,
+            compensation_add,
+            num_consecutive_same_value,
+            prev_value,
+        )
+
+        output[lab] = ssqdm_x
+        means[lab] = mean_x
+        consecutive_counts[lab] = num_consecutive_same_value
+        prev_vals[lab] = prev_value
+        comp_arr[lab] = compensation_add
+        nobs_arr[lab] = nobs
+
+    # Post-processing, replace vars that don't satisfy min_periods
+    for lab in range(ngroups):
+        nobs = nobs_arr[lab]
+        num_consecutive_same_value = consecutive_counts[lab]
+        ssqdm_x = output[lab]
+        if nobs >= min_periods and nobs > ddof:
+            if nobs == 1 or num_consecutive_same_value >= nobs:
+                result = 0.0
+            else:
+                result = ssqdm_x / (nobs - ddof)
+        else:
+            result = np.nan
+        output[lab] = result
+
+    # Second pass to get the std.dev
+    # na_position is empty list since float64 can already hold nans
+    # Do list comprehension, since numba cannot figure out that na_pos is
+    # empty list of ints on its own
+    na_pos = [0 for i in range(0)]
+    return output, na_pos
diff --git a/pandas/pandas/core/accessor.py b/pandas/pandas/core/accessor.py
new file mode 100644
index 0000000000000..78684eacf2d66
--- /dev/null
+++ b/pandas/pandas/core/accessor.py
@@ -0,0 +1,395 @@
+"""
+
+accessor.py contains base classes for implementing accessor properties
+that can be mixed into or pinned onto other pandas classes.
+
+"""
+
+from __future__ import annotations
+
+import functools
+from typing import (
+    TYPE_CHECKING,
+    final,
+)
+import warnings
+
+from pandas.util._decorators import doc
+from pandas.util._exceptions import find_stack_level
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import TypeT
+
+    from pandas import Index
+    from pandas.core.generic import NDFrame
+
+
+class DirNamesMixin:
+    _accessors: set[str] = set()
+    _hidden_attrs: frozenset[str] = frozenset()
+
+    @final
+    def _dir_deletions(self) -> set[str]:
+        """
+        Delete unwanted __dir__ for this object.
+        """
+        return self._accessors | self._hidden_attrs
+
+    def _dir_additions(self) -> set[str]:
+        """
+        Add additional __dir__ for this object.
+        """
+        return {accessor for accessor in self._accessors if hasattr(self, accessor)}
+
+    def __dir__(self) -> list[str]:
+        """
+        Provide method name lookup and completion.
+
+        Notes
+        -----
+        Only provide 'public' methods.
+        """
+        rv = set(super().__dir__())
+        rv = (rv - self._dir_deletions()) | self._dir_additions()
+        return sorted(rv)
+
+
+class PandasDelegate:
+    """
+    Abstract base class for delegating methods/properties.
+    """
+
+    def _delegate_property_get(self, name: str, *args, **kwargs):
+        raise TypeError(f"You cannot access the property {name}")
+
+    def _delegate_property_set(self, name: str, value, *args, **kwargs) -> None:
+        raise TypeError(f"The property {name} cannot be set")
+
+    def _delegate_method(self, name: str, *args, **kwargs):
+        raise TypeError(f"You cannot call method {name}")
+
+    @classmethod
+    def _add_delegate_accessors(
+        cls,
+        delegate,
+        accessors: list[str],
+        typ: str,
+        overwrite: bool = False,
+        accessor_mapping: Callable[[str], str] = lambda x: x,
+        raise_on_missing: bool = True,
+    ) -> None:
+        """
+        Add accessors to cls from the delegate class.
+
+        Parameters
+        ----------
+        cls
+            Class to add the methods/properties to.
+        delegate
+            Class to get methods/properties and doc-strings.
+        accessors : list of str
+            List of accessors to add.
+        typ : {'property', 'method'}
+        overwrite : bool, default False
+            Overwrite the method/property in the target class if it exists.
+        accessor_mapping: Callable, default lambda x: x
+            Callable to map the delegate's function to the cls' function.
+        raise_on_missing: bool, default True
+            Raise if an accessor does not exist on delegate.
+            False skips the missing accessor.
+        """
+
+        def _create_delegator_property(name: str):
+            def _getter(self):
+                return self._delegate_property_get(name)
+
+            def _setter(self, new_values):
+                return self._delegate_property_set(name, new_values)
+
+            _getter.__name__ = name
+            _setter.__name__ = name
+
+            return property(
+                fget=_getter,
+                fset=_setter,
+                doc=getattr(delegate, accessor_mapping(name)).__doc__,
+            )
+
+        def _create_delegator_method(name: str):
+            method = getattr(delegate, accessor_mapping(name))
+
+            @functools.wraps(method)
+            def f(self, *args, **kwargs):
+                return self._delegate_method(name, *args, **kwargs)
+
+            return f
+
+        for name in accessors:
+            if (
+                not raise_on_missing
+                and getattr(delegate, accessor_mapping(name), None) is None
+            ):
+                continue
+
+            if typ == "property":
+                f = _create_delegator_property(name)
+            else:
+                f = _create_delegator_method(name)
+
+            # don't overwrite existing methods/properties
+            if overwrite or not hasattr(cls, name):
+                setattr(cls, name, f)
+
+
+def delegate_names(
+    delegate,
+    accessors: list[str],
+    typ: str,
+    overwrite: bool = False,
+    accessor_mapping: Callable[[str], str] = lambda x: x,
+    raise_on_missing: bool = True,
+):
+    """
+    Add delegated names to a class using a class decorator.  This provides
+    an alternative usage to directly calling `_add_delegate_accessors`
+    below a class definition.
+
+    Parameters
+    ----------
+    delegate : object
+        The class to get methods/properties & doc-strings.
+    accessors : Sequence[str]
+        List of accessor to add.
+    typ : {'property', 'method'}
+    overwrite : bool, default False
+       Overwrite the method/property in the target class if it exists.
+    accessor_mapping: Callable, default lambda x: x
+        Callable to map the delegate's function to the cls' function.
+    raise_on_missing: bool, default True
+        Raise if an accessor does not exist on delegate.
+        False skips the missing accessor.
+
+    Returns
+    -------
+    callable
+        A class decorator.
+
+    Examples
+    --------
+    @delegate_names(Categorical, ["categories", "ordered"], "property")
+    class CategoricalAccessor(PandasDelegate):
+        [...]
+    """
+
+    def add_delegate_accessors(cls):
+        cls._add_delegate_accessors(
+            delegate,
+            accessors,
+            typ,
+            overwrite=overwrite,
+            accessor_mapping=accessor_mapping,
+            raise_on_missing=raise_on_missing,
+        )
+        return cls
+
+    return add_delegate_accessors
+
+
+class Accessor:
+    """
+    Custom property-like object.
+
+    A descriptor for accessors.
+
+    Parameters
+    ----------
+    name : str
+        Namespace that will be accessed under, e.g. ``df.foo``.
+    accessor : cls
+        Class with the extension methods.
+
+    Notes
+    -----
+    For accessor, The class's __init__ method assumes that one of
+    ``Series``, ``DataFrame`` or ``Index`` as the
+    single argument ``data``.
+    """
+
+    def __init__(self, name: str, accessor) -> None:
+        self._name = name
+        self._accessor = accessor
+
+    def __get__(self, obj, cls):
+        if obj is None:
+            # we're accessing the attribute of the class, i.e., Dataset.geo
+            return self._accessor
+        return self._accessor(obj)
+
+
+# Alias kept for downstream libraries
+# TODO: Deprecate as name is now misleading
+CachedAccessor = Accessor
+
+
+@doc(klass="", examples="", others="")
+def _register_accessor(
+    name: str, cls: type[NDFrame | Index]
+) -> Callable[[TypeT], TypeT]:
+    """
+    Register a custom accessor on {klass} objects.
+
+    Parameters
+    ----------
+    name : str
+        Name under which the accessor should be registered. A warning is issued
+        if this name conflicts with a preexisting attribute.
+
+    Returns
+    -------
+    callable
+        A class decorator.
+
+    See Also
+    --------
+    register_dataframe_accessor : Register a custom accessor on DataFrame objects.
+    register_series_accessor : Register a custom accessor on Series objects.
+    register_index_accessor : Register a custom accessor on Index objects.
+
+    Notes
+    -----
+    This function allows you to register a custom-defined accessor class for {klass}.
+    The requirements for the accessor class are as follows:
+
+    * Must contain an init method that:
+
+      * accepts a single {klass} object
+
+      * raises an AttributeError if the {klass} object does not have correctly
+        matching inputs for the accessor
+
+    * Must contain a method for each access pattern.
+
+      * The methods should be able to take any argument signature.
+
+      * Accessible using the @property decorator if no additional arguments are
+        needed.
+
+    Examples
+    --------
+    {examples}
+    """
+
+    def decorator(accessor: TypeT) -> TypeT:
+        if hasattr(cls, name):
+            warnings.warn(
+                f"registration of accessor {accessor!r} under name "
+                f"{name!r} for type {cls!r} is overriding a preexisting "
+                f"attribute with the same name.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+        setattr(cls, name, Accessor(name, accessor))
+        cls._accessors.add(name)
+        return accessor
+
+    return decorator
+
+
+_register_df_examples = """
+An accessor that only accepts integers could
+have a class defined like this:
+
+>>> @pd.api.extensions.register_dataframe_accessor("int_accessor")
+... class IntAccessor:
+...     def __init__(self, pandas_obj):
+...         if not all(pandas_obj[col].dtype == 'int64' for col in pandas_obj.columns):
+...             raise AttributeError("All columns must contain integer values only")
+...         self._obj = pandas_obj
+...
+...     def sum(self):
+...         return self._obj.sum()
+...
+>>> df = pd.DataFrame([[1, 2], ['x', 'y']])
+>>> df.int_accessor
+Traceback (most recent call last):
+...
+AttributeError: All columns must contain integer values only.
+>>> df = pd.DataFrame([[1, 2], [3, 4]])
+>>> df.int_accessor.sum()
+0    4
+1    6
+dtype: int64"""
+
+
+@doc(_register_accessor, klass="DataFrame", examples=_register_df_examples)
+def register_dataframe_accessor(name: str) -> Callable[[TypeT], TypeT]:
+    from pandas import DataFrame
+
+    return _register_accessor(name, DataFrame)
+
+
+_register_series_examples = """
+An accessor that only accepts integers could
+have a class defined like this:
+
+>>> @pd.api.extensions.register_series_accessor("int_accessor")
+... class IntAccessor:
+...     def __init__(self, pandas_obj):
+...         if not pandas_obj.dtype == 'int64':
+...             raise AttributeError("The series must contain integer data only")
+...         self._obj = pandas_obj
+...
+...     def sum(self):
+...         return self._obj.sum()
+...
+>>> df = pd.Series([1, 2, 'x'])
+>>> df.int_accessor
+Traceback (most recent call last):
+...
+AttributeError: The series must contain integer data only.
+>>> df = pd.Series([1, 2, 3])
+>>> df.int_accessor.sum()
+6"""
+
+
+@doc(_register_accessor, klass="Series", examples=_register_series_examples)
+def register_series_accessor(name: str) -> Callable[[TypeT], TypeT]:
+    from pandas import Series
+
+    return _register_accessor(name, Series)
+
+
+_register_index_examples = """
+An accessor that only accepts integers could
+have a class defined like this:
+
+>>> @pd.api.extensions.register_index_accessor("int_accessor")
+... class IntAccessor:
+...     def __init__(self, pandas_obj):
+...         if not all(isinstance(x, int) for x in pandas_obj):
+...             raise AttributeError("The index must only be an integer value")
+...         self._obj = pandas_obj
+...
+...     def even(self):
+...         return [x for x in self._obj if x % 2 == 0]
+>>> df = pd.DataFrame.from_dict(
+...     {"row1": {"1": 1, "2": "a"}, "row2": {"1": 2, "2": "b"}}, orient="index"
+... )
+>>> df.index.int_accessor
+Traceback (most recent call last):
+...
+AttributeError: The index must only be an integer value.
+>>> df = pd.DataFrame(
+...     {"col1": [1, 2, 3, 4], "col2": ["a", "b", "c", "d"]}, index=[1, 2, 5, 8]
+... )
+>>> df.index.int_accessor.even()
+[2, 8]"""
+
+
+@doc(_register_accessor, klass="Index", examples=_register_index_examples)
+def register_index_accessor(name: str) -> Callable[[TypeT], TypeT]:
+    from pandas import Index
+
+    return _register_accessor(name, Index)
diff --git a/pandas/pandas/core/algorithms.py b/pandas/pandas/core/algorithms.py
new file mode 100644
index 0000000000000..aafd802b827a5
--- /dev/null
+++ b/pandas/pandas/core/algorithms.py
@@ -0,0 +1,1701 @@
+"""
+Generic data algorithms. This module is experimental at the moment and not
+intended for public consumption
+"""
+
+from __future__ import annotations
+
+import decimal
+import operator
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    algos,
+    hashtable as htable,
+    iNaT,
+    lib,
+)
+from pandas._libs.missing import NA
+from pandas._typing import (
+    AnyArrayLike,
+    ArrayLike,
+    ArrayLikeT,
+    AxisInt,
+    DtypeObj,
+    TakeIndexer,
+    npt,
+)
+from pandas.util._decorators import doc
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import (
+    construct_1d_object_array_from_listlike,
+    np_find_common_type,
+)
+from pandas.core.dtypes.common import (
+    ensure_float64,
+    ensure_object,
+    ensure_platform_int,
+    is_bool_dtype,
+    is_complex_dtype,
+    is_dict_like,
+    is_extension_array_dtype,
+    is_float,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_object_dtype,
+    is_signed_integer_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import (
+    BaseMaskedDtype,
+    CategoricalDtype,
+    ExtensionDtype,
+    NumpyEADtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDatetimeArray,
+    ABCExtensionArray,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCNumpyExtensionArray,
+    ABCSeries,
+    ABCTimedeltaArray,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+)
+
+from pandas.core.array_algos.take import take_nd
+from pandas.core.construction import (
+    array as pd_array,
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import validate_indices
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ListLike,
+        NumpySorter,
+        NumpyValueArrayLike,
+    )
+
+    from pandas import (
+        Categorical,
+        Index,
+        Series,
+    )
+    from pandas.core.arrays import (
+        BaseMaskedArray,
+        ExtensionArray,
+    )
+
+
+# --------------- #
+# dtype access    #
+# --------------- #
+def _ensure_data(values: ArrayLike) -> np.ndarray:
+    """
+    routine to ensure that our data is of the correct
+    input dtype for lower-level routines
+
+    This will coerce:
+    - ints -> int64
+    - uint -> uint64
+    - bool -> uint8
+    - datetimelike -> i8
+    - datetime64tz -> i8 (in local tz)
+    - categorical -> codes
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+
+    Returns
+    -------
+    np.ndarray
+    """
+
+    if not isinstance(values, ABCMultiIndex):
+        # extract_array would raise
+        values = extract_array(values, extract_numpy=True)
+
+    if is_object_dtype(values.dtype):
+        return ensure_object(np.asarray(values))
+
+    elif isinstance(values.dtype, BaseMaskedDtype):
+        # i.e. BooleanArray, FloatingArray, IntegerArray
+        values = cast("BaseMaskedArray", values)
+        if not values._hasna:
+            # No pd.NAs -> We can avoid an object-dtype cast (and copy) GH#41816
+            #  recurse to avoid re-implementing logic for eg bool->uint8
+            return _ensure_data(values._data)
+        return np.asarray(values)
+
+    elif isinstance(values.dtype, CategoricalDtype):
+        # NB: cases that go through here should NOT be using _reconstruct_data
+        #  on the back-end.
+        values = cast("Categorical", values)
+        return values.codes
+
+    elif is_bool_dtype(values.dtype):
+        if isinstance(values, np.ndarray):
+            # i.e. actually dtype == np.dtype("bool")
+            return np.asarray(values).view("uint8")
+        else:
+            # e.g. Sparse[bool, False]  # TODO: no test cases get here
+            return np.asarray(values).astype("uint8", copy=False)
+
+    elif is_integer_dtype(values.dtype):
+        return np.asarray(values)
+
+    elif is_float_dtype(values.dtype):
+        # Note: checking `values.dtype == "float128"` raises on Windows and 32bit
+        # error: Item "ExtensionDtype" of "Union[Any, ExtensionDtype, dtype[Any]]"
+        # has no attribute "itemsize"
+        if values.dtype.itemsize in [2, 12, 16]:  # type: ignore[union-attr]
+            # we dont (yet) have float128 hashtable support
+            return ensure_float64(values)
+        return np.asarray(values)
+
+    elif is_complex_dtype(values.dtype):
+        return cast(np.ndarray, values)
+
+    # datetimelike
+    elif needs_i8_conversion(values.dtype):
+        npvalues = values.view("i8")
+        npvalues = cast(np.ndarray, npvalues)
+        return npvalues
+
+    # we have failed, return object
+    values = np.asarray(values, dtype=object)
+    return ensure_object(values)
+
+
+def _reconstruct_data(
+    values: ArrayLikeT, dtype: DtypeObj, original: AnyArrayLike
+) -> ArrayLikeT:
+    """
+    reverse of _ensure_data
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+    dtype : np.dtype or ExtensionDtype
+    original : AnyArrayLike
+
+    Returns
+    -------
+    ExtensionArray or np.ndarray
+    """
+    if isinstance(values, ABCExtensionArray) and values.dtype == dtype:
+        # Catch DatetimeArray/TimedeltaArray
+        return values
+
+    if not isinstance(dtype, np.dtype):
+        # i.e. ExtensionDtype; note we have ruled out above the possibility
+        #  that values.dtype == dtype
+        cls = dtype.construct_array_type()
+
+        # error: Incompatible types in assignment (expression has type
+        # "ExtensionArray", variable has type "ndarray[Any, Any]")
+        values = cls._from_sequence(values, dtype=dtype)  # type: ignore[assignment]
+
+    else:
+        values = values.astype(dtype, copy=False)
+
+    return values
+
+
+def _ensure_arraylike(values, func_name: str) -> ArrayLike:
+    """
+    ensure that we are arraylike if not already
+    """
+    if not isinstance(
+        values,
+        (ABCIndex, ABCSeries, ABCExtensionArray, np.ndarray, ABCNumpyExtensionArray),
+    ):
+        # GH#52986
+        if func_name != "isin-targets":
+            # Make an exception for the comps argument in isin.
+            raise TypeError(
+                f"{func_name} requires a Series, Index, "
+                f"ExtensionArray, np.ndarray or NumpyExtensionArray "
+                f"got {type(values).__name__}."
+            )
+
+        inferred = lib.infer_dtype(values, skipna=False)
+        if inferred in ["mixed", "string", "mixed-integer"]:
+            # "mixed-integer" to ensure we do not cast ["ss", 42] to str GH#22160
+            if isinstance(values, tuple):
+                values = list(values)
+            values = construct_1d_object_array_from_listlike(values)
+        else:
+            values = np.asarray(values)
+    return values
+
+
+_hashtables = {
+    "complex128": htable.Complex128HashTable,
+    "complex64": htable.Complex64HashTable,
+    "float64": htable.Float64HashTable,
+    "float32": htable.Float32HashTable,
+    "uint64": htable.UInt64HashTable,
+    "uint32": htable.UInt32HashTable,
+    "uint16": htable.UInt16HashTable,
+    "uint8": htable.UInt8HashTable,
+    "int64": htable.Int64HashTable,
+    "int32": htable.Int32HashTable,
+    "int16": htable.Int16HashTable,
+    "int8": htable.Int8HashTable,
+    "string": htable.StringHashTable,
+    "object": htable.PyObjectHashTable,
+}
+
+
+def _get_hashtable_algo(
+    values: np.ndarray,
+) -> tuple[type[htable.HashTable], np.ndarray]:
+    """
+    Parameters
+    ----------
+    values : np.ndarray
+
+    Returns
+    -------
+    htable : HashTable subclass
+    values : ndarray
+    """
+    values = _ensure_data(values)
+
+    ndtype = _check_object_for_strings(values)
+    hashtable = _hashtables[ndtype]
+    return hashtable, values
+
+
+def _check_object_for_strings(values: np.ndarray) -> str:
+    """
+    Check if we can use string hashtable instead of object hashtable.
+
+    Parameters
+    ----------
+    values : ndarray
+
+    Returns
+    -------
+    str
+    """
+    ndtype = values.dtype.name
+    if ndtype == "object":
+        # it's cheaper to use a String Hash Table than Object; we infer
+        # including nulls because that is the only difference between
+        # StringHashTable and ObjectHashtable
+        if lib.is_string_array(values, skipna=False):
+            ndtype = "string"
+    return ndtype
+
+
+# --------------- #
+# top-level algos #
+# --------------- #
+
+
+def unique(values):
+    """
+    Return unique values based on a hash table.
+
+    Uniques are returned in order of appearance. This does NOT sort.
+
+    Significantly faster than numpy.unique for long enough sequences.
+    Includes NA values.
+
+    Parameters
+    ----------
+    values : 1d array-like
+        The input array-like object containing values from which to extract
+        unique values.
+
+    Returns
+    -------
+    numpy.ndarray, ExtensionArray or NumpyExtensionArray
+
+        The return can be:
+
+        * Index : when the input is an Index
+        * Categorical : when the input is a Categorical dtype
+        * ndarray : when the input is a Series/ndarray
+
+        Return numpy.ndarray, ExtensionArray or NumpyExtensionArray.
+
+    See Also
+    --------
+    Index.unique : Return unique values from an Index.
+    Series.unique : Return unique values of Series object.
+
+    Examples
+    --------
+    >>> pd.unique(pd.Series([2, 1, 3, 3]))
+    array([2, 1, 3])
+
+    >>> pd.unique(pd.Series([2] + [1] * 5))
+    array([2, 1])
+
+    >>> pd.unique(pd.Series([pd.Timestamp("20160101"), pd.Timestamp("20160101")]))
+    array(['2016-01-01T00:00:00'], dtype='datetime64[s]')
+
+    >>> pd.unique(
+    ...     pd.Series(
+    ...         [
+    ...             pd.Timestamp("20160101", tz="US/Eastern"),
+    ...             pd.Timestamp("20160101", tz="US/Eastern"),
+    ...         ],
+    ...         dtype="M8[ns, US/Eastern]",
+    ...     )
+    ... )
+    <DatetimeArray>
+    ['2016-01-01 00:00:00-05:00']
+    Length: 1, dtype: datetime64[ns, US/Eastern]
+
+    >>> pd.unique(
+    ...     pd.Index(
+    ...         [
+    ...             pd.Timestamp("20160101", tz="US/Eastern"),
+    ...             pd.Timestamp("20160101", tz="US/Eastern"),
+    ...         ],
+    ...         dtype="M8[ns, US/Eastern]",
+    ...     )
+    ... )
+    DatetimeIndex(['2016-01-01 00:00:00-05:00'],
+            dtype='datetime64[ns, US/Eastern]',
+            freq=None)
+
+    >>> pd.unique(np.array(list("baabc"), dtype="O"))
+    array(['b', 'a', 'c'], dtype=object)
+
+    An unordered Categorical will return categories in the
+    order of appearance.
+
+    >>> pd.unique(pd.Series(pd.Categorical(list("baabc"))))
+    ['b', 'a', 'c']
+    Categories (3, object): ['a', 'b', 'c']
+
+    >>> pd.unique(pd.Series(pd.Categorical(list("baabc"), categories=list("abc"))))
+    ['b', 'a', 'c']
+    Categories (3, object): ['a', 'b', 'c']
+
+    An ordered Categorical preserves the category ordering.
+
+    >>> pd.unique(
+    ...     pd.Series(
+    ...         pd.Categorical(list("baabc"), categories=list("abc"), ordered=True)
+    ...     )
+    ... )
+    ['b', 'a', 'c']
+    Categories (3, object): ['a' < 'b' < 'c']
+
+    An array of tuples
+
+    >>> pd.unique(pd.Series([("a", "b"), ("b", "a"), ("a", "c"), ("b", "a")]).values)
+    array([('a', 'b'), ('b', 'a'), ('a', 'c')], dtype=object)
+
+    An NumpyExtensionArray of complex
+
+    >>> pd.unique(pd.array([1 + 1j, 2, 3]))
+    <NumpyExtensionArray>
+    [(1+1j), (2+0j), (3+0j)]
+    Length: 3, dtype: complex128
+    """
+    return unique_with_mask(values)
+
+
+def nunique_ints(values: ArrayLike) -> int:
+    """
+    Return the number of unique values for integer array-likes.
+
+    Significantly faster than pandas.unique for long enough sequences.
+    No checks are done to ensure input is integral.
+
+    Parameters
+    ----------
+    values : 1d array-like
+
+    Returns
+    -------
+    int : The number of unique values in ``values``
+    """
+    if len(values) == 0:
+        return 0
+    values = _ensure_data(values)
+    # bincount requires intp
+    result = (np.bincount(values.ravel().astype("intp")) != 0).sum()
+    return result
+
+
+def unique_with_mask(values, mask: npt.NDArray[np.bool_] | None = None):
+    """See algorithms.unique for docs. Takes a mask for masked arrays."""
+    values = _ensure_arraylike(values, func_name="unique")
+
+    if isinstance(values.dtype, ExtensionDtype):
+        # Dispatch to extension dtype's unique.
+        return values.unique()
+
+    if isinstance(values, ABCIndex):
+        # Dispatch to Index's unique.
+        return values.unique()
+
+    original = values
+    hashtable, values = _get_hashtable_algo(values)
+
+    table = hashtable(len(values))
+    if mask is None:
+        uniques = table.unique(values)
+        uniques = _reconstruct_data(uniques, original.dtype, original)
+        return uniques
+
+    else:
+        uniques, mask = table.unique(values, mask=mask)
+        uniques = _reconstruct_data(uniques, original.dtype, original)
+        assert mask is not None  # for mypy
+        return uniques, mask.astype("bool")
+
+
+unique1d = unique
+
+
+_MINIMUM_COMP_ARR_LEN = 1_000_000
+
+
+def isin(comps: ListLike, values: ListLike) -> npt.NDArray[np.bool_]:
+    """
+    Compute the isin boolean array.
+
+    Parameters
+    ----------
+    comps : list-like
+    values : list-like
+
+    Returns
+    -------
+    ndarray[bool]
+        Same length as `comps`.
+    """
+    if not is_list_like(comps):
+        raise TypeError(
+            "only list-like objects are allowed to be passed "
+            f"to isin(), you passed a `{type(comps).__name__}`"
+        )
+    if not is_list_like(values):
+        raise TypeError(
+            "only list-like objects are allowed to be passed "
+            f"to isin(), you passed a `{type(values).__name__}`"
+        )
+
+    if not isinstance(values, (ABCIndex, ABCSeries, ABCExtensionArray, np.ndarray)):
+        orig_values = list(values)
+        values = _ensure_arraylike(orig_values, func_name="isin-targets")
+
+        if (
+            len(values) > 0
+            and values.dtype.kind in "iufcb"
+            and not is_signed_integer_dtype(comps)
+        ):
+            # GH#46485 Use object to avoid upcast to float64 later
+            # TODO: Share with _find_common_type_compat
+            values = construct_1d_object_array_from_listlike(orig_values)
+
+    elif isinstance(values, ABCMultiIndex):
+        # Avoid raising in extract_array
+        values = np.array(values)
+    else:
+        values = extract_array(values, extract_numpy=True, extract_range=True)
+
+    comps_array = _ensure_arraylike(comps, func_name="isin")
+    comps_array = extract_array(comps_array, extract_numpy=True)
+    if not isinstance(comps_array, np.ndarray):
+        # i.e. Extension Array
+        return comps_array.isin(values)
+
+    elif needs_i8_conversion(comps_array.dtype):
+        # Dispatch to DatetimeLikeArrayMixin.isin
+        return pd_array(comps_array).isin(values)
+    elif needs_i8_conversion(values.dtype) and not is_object_dtype(comps_array.dtype):
+        # e.g. comps_array are integers and values are datetime64s
+        return np.zeros(comps_array.shape, dtype=bool)
+        # TODO: not quite right ... Sparse/Categorical
+    elif needs_i8_conversion(values.dtype):
+        return isin(comps_array, values.astype(object))
+
+    elif isinstance(values.dtype, ExtensionDtype):
+        return isin(np.asarray(comps_array), np.asarray(values))
+
+    # GH16012
+    # Ensure np.isin doesn't get object types or it *may* throw an exception
+    # Albeit hashmap has O(1) look-up (vs. O(logn) in sorted array),
+    # isin is faster for small sizes
+
+    # GH60678
+    # Ensure values don't contain <NA>, otherwise it throws exception with np.in1d
+
+    if (
+        len(comps_array) > _MINIMUM_COMP_ARR_LEN
+        and len(values) <= 26
+        and comps_array.dtype != object
+        and not any(v is NA for v in values)
+    ):
+        # If the values include nan we need to check for nan explicitly
+        # since np.nan it not equal to np.nan
+        if isna(values).any():
+
+            def f(c, v):
+                return np.logical_or(np.isin(c, v).ravel(), np.isnan(c))
+
+        else:
+            f = lambda a, b: np.isin(a, b).ravel()
+
+    else:
+        common = np_find_common_type(values.dtype, comps_array.dtype)
+        values = values.astype(common, copy=False)
+        comps_array = comps_array.astype(common, copy=False)
+        f = htable.ismember
+
+    return f(comps_array, values)
+
+
+def factorize_array(
+    values: np.ndarray,
+    use_na_sentinel: bool = True,
+    size_hint: int | None = None,
+    na_value: object = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> tuple[npt.NDArray[np.intp], np.ndarray]:
+    """
+    Factorize a numpy array to codes and uniques.
+
+    This doesn't do any coercion of types or unboxing before factorization.
+
+    Parameters
+    ----------
+    values : ndarray
+    use_na_sentinel : bool, default True
+        If True, the sentinel -1 will be used for NaN values. If False,
+        NaN values will be encoded as non-negative integers and will not drop the
+        NaN from the uniques of the values.
+    size_hint : int, optional
+        Passed through to the hashtable's 'get_labels' method
+    na_value : object, optional
+        A value in `values` to consider missing. Note: only use this
+        parameter when you know that you don't have any values pandas would
+        consider missing in the array (NaN for float data, iNaT for
+        datetimes, etc.).
+    mask : ndarray[bool], optional
+        If not None, the mask is used as indicator for missing values
+        (True = missing, False = valid) instead of `na_value` or
+        condition "val != val".
+
+    Returns
+    -------
+    codes : ndarray[np.intp]
+    uniques : ndarray
+    """
+    original = values
+    if values.dtype.kind in "mM":
+        # _get_hashtable_algo will cast dt64/td64 to i8 via _ensure_data, so we
+        #  need to do the same to na_value. We are assuming here that the passed
+        #  na_value is an appropriately-typed NaT.
+        # e.g. test_where_datetimelike_categorical
+        na_value = iNaT
+
+    hash_klass, values = _get_hashtable_algo(values)
+
+    table = hash_klass(size_hint or len(values))
+    uniques, codes = table.factorize(
+        values,
+        na_sentinel=-1,
+        na_value=na_value,
+        mask=mask,
+        ignore_na=use_na_sentinel,
+    )
+
+    # re-cast e.g. i8->dt64/td64, uint8->bool
+    uniques = _reconstruct_data(uniques, original.dtype, original)
+
+    codes = ensure_platform_int(codes)
+    return codes, uniques
+
+
+@doc(
+    values=dedent(
+        """\
+    values : sequence
+        A 1-D sequence. Sequences that aren't pandas objects are
+        coerced to ndarrays before factorization.
+    """
+    ),
+    sort=dedent(
+        """\
+    sort : bool, default False
+        Sort `uniques` and shuffle `codes` to maintain the
+        relationship.
+    """
+    ),
+    size_hint=dedent(
+        """\
+    size_hint : int, optional
+        Hint to the hashtable sizer.
+    """
+    ),
+)
+def factorize(
+    values,
+    sort: bool = False,
+    use_na_sentinel: bool = True,
+    size_hint: int | None = None,
+) -> tuple[np.ndarray, np.ndarray | Index]:
+    """
+    Encode the object as an enumerated type or categorical variable.
+
+    This method is useful for obtaining a numeric representation of an
+    array when all that matters is identifying distinct values. `factorize`
+    is available as both a top-level function :func:`pandas.factorize`,
+    and as a method :meth:`Series.factorize` and :meth:`Index.factorize`.
+
+    Parameters
+    ----------
+    {values}{sort}
+    use_na_sentinel : bool, default True
+        If True, the sentinel -1 will be used for NaN values. If False,
+        NaN values will be encoded as non-negative integers and will not drop the
+        NaN from the uniques of the values.
+
+        .. versionadded:: 1.5.0
+    {size_hint}\
+
+    Returns
+    -------
+    codes : ndarray
+        An integer ndarray that's an indexer into `uniques`.
+        ``uniques.take(codes)`` will have the same values as `values`.
+    uniques : ndarray, Index, or Categorical
+        The unique valid values. When `values` is Categorical, `uniques`
+        is a Categorical. When `values` is some other pandas object, an
+        `Index` is returned. Otherwise, a 1-D ndarray is returned.
+
+        .. note::
+
+           Even if there's a missing value in `values`, `uniques` will
+           *not* contain an entry for it.
+
+    See Also
+    --------
+    cut : Discretize continuous-valued array.
+    unique : Find the unique value in an array.
+
+    Notes
+    -----
+    Reference :ref:`the user guide <reshaping.factorize>` for more examples.
+
+    Examples
+    --------
+    These examples all show factorize as a top-level method like
+    ``pd.factorize(values)``. The results are identical for methods like
+    :meth:`Series.factorize`.
+
+    >>> codes, uniques = pd.factorize(np.array(['b', 'b', 'a', 'c', 'b'], dtype="O"))
+    >>> codes
+    array([0, 0, 1, 2, 0])
+    >>> uniques
+    array(['b', 'a', 'c'], dtype=object)
+
+    With ``sort=True``, the `uniques` will be sorted, and `codes` will be
+    shuffled so that the relationship is the maintained.
+
+    >>> codes, uniques = pd.factorize(np.array(['b', 'b', 'a', 'c', 'b'], dtype="O"),
+    ...                               sort=True)
+    >>> codes
+    array([1, 1, 0, 2, 1])
+    >>> uniques
+    array(['a', 'b', 'c'], dtype=object)
+
+    When ``use_na_sentinel=True`` (the default), missing values are indicated in
+    the `codes` with the sentinel value ``-1`` and missing values are not
+    included in `uniques`.
+
+    >>> codes, uniques = pd.factorize(np.array(['b', None, 'a', 'c', 'b'], dtype="O"))
+    >>> codes
+    array([ 0, -1,  1,  2,  0])
+    >>> uniques
+    array(['b', 'a', 'c'], dtype=object)
+
+    Thus far, we've only factorized lists (which are internally coerced to
+    NumPy arrays). When factorizing pandas objects, the type of `uniques`
+    will differ. For Categoricals, a `Categorical` is returned.
+
+    >>> cat = pd.Categorical(['a', 'a', 'c'], categories=['a', 'b', 'c'])
+    >>> codes, uniques = pd.factorize(cat)
+    >>> codes
+    array([0, 0, 1])
+    >>> uniques
+    ['a', 'c']
+    Categories (3, object): ['a', 'b', 'c']
+
+    Notice that ``'b'`` is in ``uniques.categories``, despite not being
+    present in ``cat.values``.
+
+    For all other pandas objects, an Index of the appropriate type is
+    returned.
+
+    >>> cat = pd.Series(['a', 'a', 'c'])
+    >>> codes, uniques = pd.factorize(cat)
+    >>> codes
+    array([0, 0, 1])
+    >>> uniques
+    Index(['a', 'c'], dtype='object')
+
+    If NaN is in the values, and we want to include NaN in the uniques of the
+    values, it can be achieved by setting ``use_na_sentinel=False``.
+
+    >>> values = np.array([1, 2, 1, np.nan])
+    >>> codes, uniques = pd.factorize(values)  # default: use_na_sentinel=True
+    >>> codes
+    array([ 0,  1,  0, -1])
+    >>> uniques
+    array([1., 2.])
+
+    >>> codes, uniques = pd.factorize(values, use_na_sentinel=False)
+    >>> codes
+    array([0, 1, 0, 2])
+    >>> uniques
+    array([ 1.,  2., nan])
+    """
+    # Implementation notes: This method is responsible for 3 things
+    # 1.) coercing data to array-like (ndarray, Index, extension array)
+    # 2.) factorizing codes and uniques
+    # 3.) Maybe boxing the uniques in an Index
+    #
+    # Step 2 is dispatched to extension types (like Categorical). They are
+    # responsible only for factorization. All data coercion, sorting and boxing
+    # should happen here.
+    if isinstance(values, (ABCIndex, ABCSeries)):
+        return values.factorize(sort=sort, use_na_sentinel=use_na_sentinel)
+
+    values = _ensure_arraylike(values, func_name="factorize")
+    original = values
+
+    if (
+        isinstance(values, (ABCDatetimeArray, ABCTimedeltaArray))
+        and values.freq is not None
+    ):
+        # The presence of 'freq' means we can fast-path sorting and know there
+        #  aren't NAs
+        codes, uniques = values.factorize(sort=sort)
+        return codes, uniques
+
+    elif not isinstance(values, np.ndarray):
+        # i.e. ExtensionArray
+        codes, uniques = values.factorize(use_na_sentinel=use_na_sentinel)
+
+    else:
+        values = np.asarray(values)  # convert DTA/TDA/MultiIndex
+
+        if not use_na_sentinel and values.dtype == object:
+            # factorize can now handle differentiating various types of null values.
+            # These can only occur when the array has object dtype.
+            # However, for backwards compatibility we only use the null for the
+            # provided dtype. This may be revisited in the future, see GH#48476.
+            null_mask = isna(values)
+            if null_mask.any():
+                na_value = na_value_for_dtype(values.dtype, compat=False)
+                # Don't modify (potentially user-provided) array
+                values = np.where(null_mask, na_value, values)
+
+        codes, uniques = factorize_array(
+            values,
+            use_na_sentinel=use_na_sentinel,
+            size_hint=size_hint,
+        )
+
+    if sort and len(uniques) > 0:
+        uniques, codes = safe_sort(
+            uniques,
+            codes,
+            use_na_sentinel=use_na_sentinel,
+            assume_unique=True,
+            verify=False,
+        )
+
+    uniques = _reconstruct_data(uniques, original.dtype, original)
+
+    return codes, uniques
+
+
+def value_counts_internal(
+    values,
+    sort: bool = True,
+    ascending: bool = False,
+    normalize: bool = False,
+    bins=None,
+    dropna: bool = True,
+) -> Series:
+    from pandas import (
+        Index,
+        Series,
+    )
+
+    index_name = getattr(values, "name", None)
+    name = "proportion" if normalize else "count"
+
+    if bins is not None:
+        from pandas.core.reshape.tile import cut
+
+        if isinstance(values, Series):
+            values = values._values
+
+        try:
+            ii = cut(values, bins, include_lowest=True)
+        except TypeError as err:
+            raise TypeError("bins argument only works with numeric data.") from err
+
+        # count, remove nulls (from the index), and but the bins
+        result = ii.value_counts(dropna=dropna)
+        result.name = name
+        result = result[result.index.notna()]
+        result.index = result.index.astype("interval")
+        result = result.sort_index()
+
+        # if we are dropna and we have NO values
+        if dropna and (result._values == 0).all():
+            result = result.iloc[0:0]
+
+        # normalizing is by len of all (regardless of dropna)
+        counts = np.array([len(ii)])
+
+    else:
+        if is_extension_array_dtype(values):
+            # handle Categorical and sparse,
+            result = Series(values, copy=False)._values.value_counts(dropna=dropna)
+            result.name = name
+            result.index.name = index_name
+            counts = result._values
+            if not isinstance(counts, np.ndarray):
+                # e.g. ArrowExtensionArray
+                counts = np.asarray(counts)
+
+        elif isinstance(values, ABCMultiIndex):
+            # GH49558
+            levels = list(range(values.nlevels))
+            result = (
+                Series(index=values, name=name)
+                .groupby(level=levels, dropna=dropna)
+                .size()
+            )
+            result.index.names = values.names
+            counts = result._values
+
+        else:
+            values = _ensure_arraylike(values, func_name="value_counts")
+            keys, counts, _ = value_counts_arraylike(values, dropna)
+            if keys.dtype == np.float16:
+                keys = keys.astype(np.float32)
+
+            # Starting in 3.0, we no longer perform dtype inference on the
+            #  Index object we construct here, xref GH#56161
+            idx = Index(keys, dtype=keys.dtype, name=index_name)
+            result = Series(counts, index=idx, name=name, copy=False)
+
+    if sort:
+        result = result.sort_values(ascending=ascending)
+
+    if normalize:
+        result = result / counts.sum()
+
+    return result
+
+
+# Called once from SparseArray, otherwise could be private
+def value_counts_arraylike(
+    values: np.ndarray, dropna: bool, mask: npt.NDArray[np.bool_] | None = None
+) -> tuple[ArrayLike, npt.NDArray[np.int64], int]:
+    """
+    Parameters
+    ----------
+    values : np.ndarray
+    dropna : bool
+    mask : np.ndarray[bool] or None, default None
+
+    Returns
+    -------
+    uniques : np.ndarray
+    counts : np.ndarray[np.int64]
+    """
+    original = values
+    values = _ensure_data(values)
+
+    keys, counts, na_counter = htable.value_count(values, dropna, mask=mask)
+
+    if needs_i8_conversion(original.dtype):
+        # datetime, timedelta, or period
+
+        if dropna:
+            mask = keys != iNaT
+            keys, counts = keys[mask], counts[mask]
+
+    res_keys = _reconstruct_data(keys, original.dtype, original)
+    return res_keys, counts, na_counter
+
+
+def duplicated(
+    values: ArrayLike,
+    keep: Literal["first", "last", False] = "first",
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> npt.NDArray[np.bool_]:
+    """
+    Return boolean ndarray denoting duplicate values.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+        Array over which to check for duplicate values.
+    keep : {'first', 'last', False}, default 'first'
+        - ``first`` : Mark duplicates as ``True`` except for the first
+          occurrence.
+        - ``last`` : Mark duplicates as ``True`` except for the last
+          occurrence.
+        - False : Mark all duplicates as ``True``.
+    mask : ndarray[bool], optional
+        array indicating which elements to exclude from checking
+
+    Returns
+    -------
+    duplicated : ndarray[bool]
+    """
+    values = _ensure_data(values)
+    return htable.duplicated(values, keep=keep, mask=mask)
+
+
+def mode(
+    values: ArrayLike, dropna: bool = True, mask: npt.NDArray[np.bool_] | None = None
+) -> ArrayLike:
+    """
+    Returns the mode(s) of an array.
+
+    Parameters
+    ----------
+    values : array-like
+        Array over which to check for duplicate values.
+    dropna : bool, default True
+        Don't consider counts of NaN/NaT.
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+    """
+    values = _ensure_arraylike(values, func_name="mode")
+    original = values
+
+    if needs_i8_conversion(values.dtype):
+        # Got here with ndarray; dispatch to DatetimeArray/TimedeltaArray.
+        values = ensure_wrapped_if_datetimelike(values)
+        values = cast("ExtensionArray", values)
+        return values._mode(dropna=dropna)
+
+    values = _ensure_data(values)
+
+    npresult, res_mask = htable.mode(values, dropna=dropna, mask=mask)
+    if res_mask is not None:
+        return npresult, res_mask  # type: ignore[return-value]
+
+    try:
+        npresult = safe_sort(npresult)
+    except TypeError as err:
+        warnings.warn(
+            f"Unable to sort modes: {err}",
+            stacklevel=find_stack_level(),
+        )
+
+    result = _reconstruct_data(npresult, original.dtype, original)
+    return result
+
+
+def rank(
+    values: ArrayLike,
+    axis: AxisInt = 0,
+    method: str = "average",
+    na_option: str = "keep",
+    ascending: bool = True,
+    pct: bool = False,
+) -> npt.NDArray[np.float64]:
+    """
+    Rank the values along a given axis.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+        Array whose values will be ranked. The number of dimensions in this
+        array must not exceed 2.
+    axis : int, default 0
+        Axis over which to perform rankings.
+    method : {'average', 'min', 'max', 'first', 'dense'}, default 'average'
+        The method by which tiebreaks are broken during the ranking.
+    na_option : {'keep', 'top'}, default 'keep'
+        The method by which NaNs are placed in the ranking.
+        - ``keep``: rank each NaN value with a NaN ranking
+        - ``top``: replace each NaN with either +/- inf so that they
+                   there are ranked at the top
+    ascending : bool, default True
+        Whether or not the elements should be ranked in ascending order.
+    pct : bool, default False
+        Whether or not to the display the returned rankings in integer form
+        (e.g. 1, 2, 3) or in percentile form (e.g. 0.333..., 0.666..., 1).
+    """
+    is_datetimelike = needs_i8_conversion(values.dtype)
+    values = _ensure_data(values)
+
+    if values.ndim == 1:
+        ranks = algos.rank_1d(
+            values,
+            is_datetimelike=is_datetimelike,
+            ties_method=method,
+            ascending=ascending,
+            na_option=na_option,
+            pct=pct,
+        )
+    elif values.ndim == 2:
+        ranks = algos.rank_2d(
+            values,
+            axis=axis,
+            is_datetimelike=is_datetimelike,
+            ties_method=method,
+            ascending=ascending,
+            na_option=na_option,
+            pct=pct,
+        )
+    else:
+        raise TypeError("Array with ndim > 2 are not supported.")
+
+    return ranks
+
+
+# ---- #
+# take #
+# ---- #
+
+
+def take(
+    arr,
+    indices: TakeIndexer,
+    axis: AxisInt = 0,
+    allow_fill: bool = False,
+    fill_value=None,
+):
+    """
+    Take elements from an array.
+
+    Parameters
+    ----------
+    arr : array-like or scalar value
+        Non array-likes (sequences/scalars without a dtype) are coerced
+        to an ndarray.
+
+        .. deprecated:: 2.1.0
+            Passing an argument other than a numpy.ndarray, ExtensionArray,
+            Index, or Series is deprecated.
+
+    indices : sequence of int or one-dimensional np.ndarray of int
+        Indices to be taken.
+    axis : int, default 0
+        The axis over which to select values.
+    allow_fill : bool, default False
+        How to handle negative values in `indices`.
+
+        * False: negative values in `indices` indicate positional indices
+          from the right (the default). This is similar to :func:`numpy.take`.
+
+        * True: negative values in `indices` indicate
+          missing values. These values are set to `fill_value`. Any other
+          negative values raise a ``ValueError``.
+
+    fill_value : any, optional
+        Fill value to use for NA-indices when `allow_fill` is True.
+        This may be ``None``, in which case the default NA value for
+        the type (``self.dtype.na_value``) is used.
+
+        For multi-dimensional `arr`, each *element* is filled with
+        `fill_value`.
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+        Same type as the input.
+
+    Raises
+    ------
+    IndexError
+        When `indices` is out of bounds for the array.
+    ValueError
+        When the indexer contains negative values other than ``-1``
+        and `allow_fill` is True.
+
+    Notes
+    -----
+    When `allow_fill` is False, `indices` may be whatever dimensionality
+    is accepted by NumPy for `arr`.
+
+    When `allow_fill` is True, `indices` should be 1-D.
+
+    See Also
+    --------
+    numpy.take : Take elements from an array along an axis.
+
+    Examples
+    --------
+    >>> import pandas as pd
+
+    With the default ``allow_fill=False``, negative numbers indicate
+    positional indices from the right.
+
+    >>> pd.api.extensions.take(np.array([10, 20, 30]), [0, 0, -1])
+    array([10, 10, 30])
+
+    Setting ``allow_fill=True`` will place `fill_value` in those positions.
+
+    >>> pd.api.extensions.take(np.array([10, 20, 30]), [0, 0, -1], allow_fill=True)
+    array([10., 10., nan])
+
+    >>> pd.api.extensions.take(
+    ...     np.array([10, 20, 30]), [0, 0, -1], allow_fill=True, fill_value=-10
+    ... )
+    array([ 10,  10, -10])
+    """
+    if not isinstance(
+        arr,
+        (np.ndarray, ABCExtensionArray, ABCIndex, ABCSeries, ABCNumpyExtensionArray),
+    ):
+        # GH#52981
+        raise TypeError(
+            "pd.api.extensions.take requires a numpy.ndarray, ExtensionArray, "
+            f"Index, Series, or NumpyExtensionArray got {type(arr).__name__}."
+        )
+
+    indices = ensure_platform_int(indices)
+
+    if allow_fill:
+        # Pandas style, -1 means NA
+        validate_indices(indices, arr.shape[axis])
+        # error: Argument 1 to "take_nd" has incompatible type
+        # "ndarray[Any, Any] | ExtensionArray | Index | Series"; expected
+        # "ndarray[Any, Any]"
+        result = take_nd(
+            arr,  # type: ignore[arg-type]
+            indices,
+            axis=axis,
+            allow_fill=True,
+            fill_value=fill_value,
+        )
+    else:
+        # NumPy style
+        # error: Unexpected keyword argument "axis" for "take" of "ExtensionArray"
+        result = arr.take(indices, axis=axis)  # type: ignore[call-arg,assignment]
+    return result
+
+
+# ------------ #
+# searchsorted #
+# ------------ #
+
+
+def searchsorted(
+    arr: ArrayLike,
+    value: NumpyValueArrayLike | ExtensionArray,
+    side: Literal["left", "right"] = "left",
+    sorter: NumpySorter | None = None,
+) -> npt.NDArray[np.intp] | np.intp:
+    """
+    Find indices where elements should be inserted to maintain order.
+
+    Find the indices into a sorted array `arr` (a) such that, if the
+    corresponding elements in `value` were inserted before the indices,
+    the order of `arr` would be preserved.
+
+    Assuming that `arr` is sorted:
+
+    ======  ================================
+    `side`  returned index `i` satisfies
+    ======  ================================
+    left    ``arr[i-1] < value <= self[i]``
+    right   ``arr[i-1] <= value < self[i]``
+    ======  ================================
+
+    Parameters
+    ----------
+    arr: np.ndarray, ExtensionArray, Series
+        Input array. If `sorter` is None, then it must be sorted in
+        ascending order, otherwise `sorter` must be an array of indices
+        that sort it.
+    value : array-like or scalar
+        Values to insert into `arr`.
+    side : {'left', 'right'}, optional
+        If 'left', the index of the first suitable location found is given.
+        If 'right', return the last such index.  If there is no suitable
+        index, return either 0 or N (where N is the length of `self`).
+    sorter : 1-D array-like, optional
+        Optional array of integer indices that sort array a into ascending
+        order. They are typically the result of argsort.
+
+    Returns
+    -------
+    array of ints or int
+        If value is array-like, array of insertion points.
+        If value is scalar, a single integer.
+
+    See Also
+    --------
+    numpy.searchsorted : Similar method from NumPy.
+    """
+    if sorter is not None:
+        sorter = ensure_platform_int(sorter)
+
+    if (
+        isinstance(arr, np.ndarray)
+        and arr.dtype.kind in "iu"
+        and (is_integer(value) or is_integer_dtype(value))
+    ):
+        # if `arr` and `value` have different dtypes, `arr` would be
+        # recast by numpy, causing a slow search.
+        # Before searching below, we therefore try to give `value` the
+        # same dtype as `arr`, while guarding against integer overflows.
+        iinfo = np.iinfo(arr.dtype.type)
+        value_arr = np.array([value]) if is_integer(value) else np.array(value)
+        if (value_arr >= iinfo.min).all() and (value_arr <= iinfo.max).all():
+            # value within bounds, so no overflow, so can convert value dtype
+            # to dtype of arr
+            dtype = arr.dtype
+        else:
+            dtype = value_arr.dtype
+
+        if is_integer(value):
+            # We know that value is int
+            value = cast(int, dtype.type(value))
+        else:
+            value = pd_array(cast(ArrayLike, value), dtype=dtype)
+    else:
+        # E.g. if `arr` is an array with dtype='datetime64[ns]'
+        # and `value` is a pd.Timestamp, we may need to convert value
+        arr = ensure_wrapped_if_datetimelike(arr)
+
+    # Argument 1 to "searchsorted" of "ndarray" has incompatible type
+    # "Union[NumpyValueArrayLike, ExtensionArray]"; expected "NumpyValueArrayLike"
+    return arr.searchsorted(value, side=side, sorter=sorter)  # type: ignore[arg-type]
+
+
+# ---- #
+# diff #
+# ---- #
+
+_diff_special = {"float64", "float32", "int64", "int32", "int16", "int8"}
+
+
+def diff(arr, n: int, axis: AxisInt = 0):
+    """
+    difference of n between self,
+    analogous to s-s.shift(n)
+
+    Parameters
+    ----------
+    arr : ndarray or ExtensionArray
+    n : int
+        number of periods
+    axis : {0, 1}
+        axis to shift on
+    stacklevel : int, default 3
+        The stacklevel for the lost dtype warning.
+
+    Returns
+    -------
+    shifted
+    """
+
+    # added a check on the integer value of period
+    # see https://github.com/pandas-dev/pandas/issues/56607
+    if not lib.is_integer(n):
+        if not (is_float(n) and n.is_integer()):
+            raise ValueError("periods must be an integer")
+        n = int(n)
+    na = np.nan
+    dtype = arr.dtype
+
+    is_bool = is_bool_dtype(dtype)
+    if is_bool:
+        op = operator.xor
+    else:
+        op = operator.sub
+
+    if isinstance(dtype, NumpyEADtype):
+        # NumpyExtensionArray cannot necessarily hold shifted versions of itself.
+        arr = arr.to_numpy()
+        dtype = arr.dtype
+
+    if not isinstance(arr, np.ndarray):
+        # i.e ExtensionArray
+        if hasattr(arr, f"__{op.__name__}__"):
+            if axis != 0:
+                raise ValueError(f"cannot diff {type(arr).__name__} on axis={axis}")
+            return op(arr, arr.shift(n))
+        else:
+            raise TypeError(
+                f"{type(arr).__name__} has no 'diff' method. "
+                "Convert to a suitable dtype prior to calling 'diff'."
+            )
+
+    is_timedelta = False
+    if arr.dtype.kind in "mM":
+        dtype = np.int64
+        arr = arr.view("i8")
+        na = iNaT
+        is_timedelta = True
+
+    elif is_bool:
+        # We have to cast in order to be able to hold np.nan
+        dtype = np.object_
+
+    elif dtype.kind in "iu":
+        # We have to cast in order to be able to hold np.nan
+
+        # int8, int16 are incompatible with float64,
+        # see https://github.com/cython/cython/issues/2646
+        if arr.dtype.name in ["int8", "int16"]:
+            dtype = np.float32
+        else:
+            dtype = np.float64
+
+    orig_ndim = arr.ndim
+    if orig_ndim == 1:
+        # reshape so we can always use algos.diff_2d
+        arr = arr.reshape(-1, 1)
+        # TODO: require axis == 0
+
+    dtype = np.dtype(dtype)
+    out_arr = np.empty(arr.shape, dtype=dtype)
+
+    na_indexer = [slice(None)] * 2
+    na_indexer[axis] = slice(None, n) if n >= 0 else slice(n, None)
+    out_arr[tuple(na_indexer)] = na
+
+    if arr.dtype.name in _diff_special:
+        # TODO: can diff_2d dtype specialization troubles be fixed by defining
+        #  out_arr inside diff_2d?
+        algos.diff_2d(arr, out_arr, n, axis, datetimelike=is_timedelta)
+    else:
+        # To keep mypy happy, _res_indexer is a list while res_indexer is
+        #  a tuple, ditto for lag_indexer.
+        _res_indexer = [slice(None)] * 2
+        _res_indexer[axis] = slice(n, None) if n >= 0 else slice(None, n)
+        res_indexer = tuple(_res_indexer)
+
+        _lag_indexer = [slice(None)] * 2
+        _lag_indexer[axis] = slice(None, -n) if n > 0 else slice(-n, None)
+        lag_indexer = tuple(_lag_indexer)
+
+        out_arr[res_indexer] = op(arr[res_indexer], arr[lag_indexer])
+
+    if is_timedelta:
+        out_arr = out_arr.view("timedelta64[ns]")
+
+    if orig_ndim == 1:
+        out_arr = out_arr[:, 0]
+    return out_arr
+
+
+# --------------------------------------------------------------------
+# Helper functions
+
+
+# Note: safe_sort is in algorithms.py instead of sorting.py because it is
+#  low-dependency, is used in this module, and used private methods from
+#  this module.
+def safe_sort(
+    values: Index | ArrayLike,
+    codes: npt.NDArray[np.intp] | None = None,
+    use_na_sentinel: bool = True,
+    assume_unique: bool = False,
+    verify: bool = True,
+) -> AnyArrayLike | tuple[AnyArrayLike, np.ndarray]:
+    """
+    Sort ``values`` and reorder corresponding ``codes``.
+
+    ``values`` should be unique if ``codes`` is not None.
+    Safe for use with mixed types (int, str), orders ints before strs.
+
+    Parameters
+    ----------
+    values : list-like
+        Sequence; must be unique if ``codes`` is not None.
+    codes : np.ndarray[intp] or None, default None
+        Indices to ``values``. All out of bound indices are treated as
+        "not found" and will be masked with ``-1``.
+    use_na_sentinel : bool, default True
+        If True, the sentinel -1 will be used for NaN values. If False,
+        NaN values will be encoded as non-negative integers and will not drop the
+        NaN from the uniques of the values.
+    assume_unique : bool, default False
+        When True, ``values`` are assumed to be unique, which can speed up
+        the calculation. Ignored when ``codes`` is None.
+    verify : bool, default True
+        Check if codes are out of bound for the values and put out of bound
+        codes equal to ``-1``. If ``verify=False``, it is assumed there
+        are no out of bound codes. Ignored when ``codes`` is None.
+
+    Returns
+    -------
+    ordered : AnyArrayLike
+        Sorted ``values``
+    new_codes : ndarray
+        Reordered ``codes``; returned when ``codes`` is not None.
+
+    Raises
+    ------
+    TypeError
+        * If ``values`` is not list-like or if ``codes`` is neither None
+        nor list-like
+        * If ``values`` cannot be sorted
+    ValueError
+        * If ``codes`` is not None and ``values`` contain duplicates.
+    """
+    if not isinstance(values, (np.ndarray, ABCExtensionArray, ABCIndex)):
+        raise TypeError(
+            "Only np.ndarray, ExtensionArray, and Index objects are allowed to "
+            "be passed to safe_sort as values"
+        )
+
+    sorter = None
+    ordered: AnyArrayLike
+
+    if (
+        not isinstance(values.dtype, ExtensionDtype)
+        and lib.infer_dtype(values, skipna=False) == "mixed-integer"
+    ):
+        ordered = _sort_mixed(values)
+    else:
+        try:
+            sorter = values.argsort()
+            ordered = values.take(sorter)
+        except (TypeError, decimal.InvalidOperation):
+            # Previous sorters failed or were not applicable, try `_sort_mixed`
+            # which would work, but which fails for special case of 1d arrays
+            # with tuples.
+            if values.size and isinstance(values[0], tuple):
+                # error: Argument 1 to "_sort_tuples" has incompatible type
+                # "Union[Index, ExtensionArray, ndarray[Any, Any]]"; expected
+                # "ndarray[Any, Any]"
+                ordered = _sort_tuples(values)  # type: ignore[arg-type]
+            else:
+                ordered = _sort_mixed(values)
+
+    # codes:
+
+    if codes is None:
+        return ordered
+
+    if not is_list_like(codes):
+        raise TypeError(
+            "Only list-like objects or None are allowed to "
+            "be passed to safe_sort as codes"
+        )
+    codes = ensure_platform_int(np.asarray(codes))
+
+    if not assume_unique and not len(unique(values)) == len(values):
+        raise ValueError("values should be unique if codes is not None")
+
+    if sorter is None:
+        # mixed types
+        # error: Argument 1 to "_get_hashtable_algo" has incompatible type
+        # "Union[Index, ExtensionArray, ndarray[Any, Any]]"; expected
+        # "ndarray[Any, Any]"
+        hash_klass, values = _get_hashtable_algo(values)  # type: ignore[arg-type]
+        t = hash_klass(len(values))
+        t.map_locations(values)
+        # error: Argument 1 to "lookup" of "HashTable" has incompatible type
+        # "ExtensionArray | ndarray[Any, Any] | Index | Series"; expected "ndarray"
+        sorter = ensure_platform_int(t.lookup(ordered))  # type: ignore[arg-type]
+
+    if use_na_sentinel:
+        # take_nd is faster, but only works for na_sentinels of -1
+        order2 = sorter.argsort()
+        if verify:
+            mask = (codes < -len(values)) | (codes >= len(values))
+            codes[mask] = -1
+        new_codes = take_nd(order2, codes, fill_value=-1)
+    else:
+        reverse_indexer = np.empty(len(sorter), dtype=int)
+        reverse_indexer.put(sorter, np.arange(len(sorter)))
+        # Out of bound indices will be masked with `-1` next, so we
+        # may deal with them here without performance loss using `mode='wrap'`
+        new_codes = reverse_indexer.take(codes, mode="wrap")
+
+    return ordered, ensure_platform_int(new_codes)
+
+
+def _sort_mixed(values) -> AnyArrayLike:
+    """order ints before strings before nulls in 1d arrays"""
+    str_pos = np.array([isinstance(x, str) for x in values], dtype=bool)
+    null_pos = np.array([isna(x) for x in values], dtype=bool)
+    num_pos = ~str_pos & ~null_pos
+    str_argsort = np.argsort(values[str_pos])
+    num_argsort = np.argsort(values[num_pos])
+    # convert boolean arrays to positional indices, then order by underlying values
+    str_locs = str_pos.nonzero()[0].take(str_argsort)
+    num_locs = num_pos.nonzero()[0].take(num_argsort)
+    null_locs = null_pos.nonzero()[0]
+    locs = np.concatenate([num_locs, str_locs, null_locs])
+    return values.take(locs)
+
+
+def _sort_tuples(values: np.ndarray) -> np.ndarray:
+    """
+    Convert array of tuples (1d) to array of arrays (2d).
+    We need to keep the columns separately as they contain different types and
+    nans (can't use `np.sort` as it may fail when str and nan are mixed in a
+    column as types cannot be compared).
+    """
+    from pandas.core.internals.construction import to_arrays
+    from pandas.core.sorting import lexsort_indexer
+
+    arrays, _ = to_arrays(values, None)
+    indexer = lexsort_indexer(arrays, orders=True)
+    return values[indexer]
+
+
+def union_with_duplicates(
+    lvals: ArrayLike | Index, rvals: ArrayLike | Index
+) -> ArrayLike | Index:
+    """
+    Extracts the union from lvals and rvals with respect to duplicates and nans in
+    both arrays.
+
+    Parameters
+    ----------
+    lvals: np.ndarray or ExtensionArray
+        left values which is ordered in front.
+    rvals: np.ndarray or ExtensionArray
+        right values ordered after lvals.
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+        Containing the unsorted union of both arrays.
+
+    Notes
+    -----
+    Caller is responsible for ensuring lvals.dtype == rvals.dtype.
+    """
+    from pandas import Series
+
+    l_count = value_counts_internal(lvals, dropna=False)
+    r_count = value_counts_internal(rvals, dropna=False)
+    l_count, r_count = l_count.align(r_count, fill_value=0)
+    final_count = np.maximum(l_count.values, r_count.values)
+    final_count = Series(final_count, index=l_count.index, dtype="int", copy=False)
+    if isinstance(lvals, ABCMultiIndex) and isinstance(rvals, ABCMultiIndex):
+        unique_vals = lvals.append(rvals).unique()
+    else:
+        if isinstance(lvals, ABCIndex):
+            lvals = lvals._values
+        if isinstance(rvals, ABCIndex):
+            rvals = rvals._values
+        # error: List item 0 has incompatible type "Union[ExtensionArray,
+        # ndarray[Any, Any], Index]"; expected "Union[ExtensionArray,
+        # ndarray[Any, Any]]"
+        combined = concat_compat([lvals, rvals])  # type: ignore[list-item]
+        unique_vals = unique(combined)
+        unique_vals = ensure_wrapped_if_datetimelike(unique_vals)
+    repeats = final_count.reindex(unique_vals).values
+    return np.repeat(unique_vals, repeats)
+
+
+def map_array(
+    arr: ArrayLike,
+    mapper,
+    na_action: Literal["ignore"] | None = None,
+) -> np.ndarray | ExtensionArray | Index:
+    """
+    Map values using an input mapping or function.
+
+    Parameters
+    ----------
+    mapper : function, dict, or Series
+        Mapping correspondence.
+    na_action : {None, 'ignore'}, default None
+        If 'ignore', propagate NA values, without passing them to the
+        mapping correspondence.
+
+    Returns
+    -------
+    Union[ndarray, Index, ExtensionArray]
+        The output of the mapping function applied to the array.
+        If the function returns a tuple with more than one element
+        a MultiIndex will be returned.
+    """
+    if na_action not in (None, "ignore"):
+        msg = f"na_action must either be 'ignore' or None, {na_action} was passed"
+        raise ValueError(msg)
+
+    # we can fastpath dict/Series to an efficient map
+    # as we know that we are not going to have to yield
+    # python types
+    if is_dict_like(mapper):
+        if isinstance(mapper, dict) and hasattr(mapper, "__missing__"):
+            # If a dictionary subclass defines a default value method,
+            # convert mapper to a lookup function (GH #15999).
+            dict_with_default = mapper
+            mapper = lambda x: dict_with_default[
+                np.nan if isinstance(x, float) and np.isnan(x) else x
+            ]
+        else:
+            # Dictionary does not have a default. Thus it's safe to
+            # convert to an Series for efficiency.
+            # we specify the keys here to handle the
+            # possibility that they are tuples
+
+            # The return value of mapping with an empty mapper is
+            # expected to be pd.Series(np.nan, ...). As np.nan is
+            # of dtype float64 the return value of this method should
+            # be float64 as well
+            from pandas import Series
+
+            if len(mapper) == 0:
+                mapper = Series(mapper, dtype=np.float64)
+            else:
+                mapper = Series(mapper)
+
+    if isinstance(mapper, ABCSeries):
+        if na_action == "ignore":
+            mapper = mapper[mapper.index.notna()]
+
+        # Since values were input this means we came from either
+        # a dict or a series and mapper should be an index
+        indexer = mapper.index.get_indexer(arr)
+        new_values = take_nd(mapper._values, indexer)
+
+        return new_values
+
+    if not len(arr):
+        return arr.copy()
+
+    # we must convert to python types
+    values = arr.astype(object, copy=False)
+    if na_action is None:
+        return lib.map_infer(values, mapper)
+    else:
+        return lib.map_infer_mask(values, mapper, mask=isna(values).view(np.uint8))
diff --git a/pandas/pandas/core/api.py b/pandas/pandas/core/api.py
new file mode 100644
index 0000000000000..ec12d543d8389
--- /dev/null
+++ b/pandas/pandas/core/api.py
@@ -0,0 +1,138 @@
+from pandas._libs import (
+    NaT,
+    Period,
+    Timedelta,
+    Timestamp,
+)
+from pandas._libs.missing import NA
+
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    isnull,
+    notna,
+    notnull,
+)
+
+from pandas.core.algorithms import (
+    factorize,
+    unique,
+)
+from pandas.core.arrays import Categorical
+from pandas.core.arrays.boolean import BooleanDtype
+from pandas.core.arrays.floating import (
+    Float32Dtype,
+    Float64Dtype,
+)
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    Int16Dtype,
+    Int32Dtype,
+    Int64Dtype,
+    UInt8Dtype,
+    UInt16Dtype,
+    UInt32Dtype,
+    UInt64Dtype,
+)
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.construction import array  # noqa: ICN001
+from pandas.core.flags import Flags
+from pandas.core.groupby import (
+    Grouper,
+    NamedAgg,
+)
+from pandas.core.indexes.api import (
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    TimedeltaIndex,
+)
+from pandas.core.indexes.datetimes import (
+    bdate_range,
+    date_range,
+)
+from pandas.core.indexes.interval import (
+    Interval,
+    interval_range,
+)
+from pandas.core.indexes.period import period_range
+from pandas.core.indexes.timedeltas import timedelta_range
+from pandas.core.indexing import IndexSlice
+from pandas.core.series import Series
+from pandas.core.tools.datetimes import to_datetime
+from pandas.core.tools.numeric import to_numeric
+from pandas.core.tools.timedeltas import to_timedelta
+
+from pandas.io.formats.format import set_eng_float_format
+from pandas.tseries.offsets import DateOffset
+
+# DataFrame needs to be imported after NamedAgg to avoid a circular import
+from pandas.core.frame import DataFrame  # isort:skip
+
+__all__ = [
+    "NA",
+    "ArrowDtype",
+    "BooleanDtype",
+    "Categorical",
+    "CategoricalDtype",
+    "CategoricalIndex",
+    "DataFrame",
+    "DateOffset",
+    "DatetimeIndex",
+    "DatetimeTZDtype",
+    "Flags",
+    "Float32Dtype",
+    "Float64Dtype",
+    "Grouper",
+    "Index",
+    "IndexSlice",
+    "Int8Dtype",
+    "Int16Dtype",
+    "Int32Dtype",
+    "Int64Dtype",
+    "Interval",
+    "IntervalDtype",
+    "IntervalIndex",
+    "MultiIndex",
+    "NaT",
+    "NamedAgg",
+    "Period",
+    "PeriodDtype",
+    "PeriodIndex",
+    "RangeIndex",
+    "Series",
+    "StringDtype",
+    "Timedelta",
+    "TimedeltaIndex",
+    "Timestamp",
+    "UInt8Dtype",
+    "UInt16Dtype",
+    "UInt32Dtype",
+    "UInt64Dtype",
+    "array",
+    "bdate_range",
+    "date_range",
+    "factorize",
+    "interval_range",
+    "isna",
+    "isnull",
+    "notna",
+    "notnull",
+    "period_range",
+    "set_eng_float_format",
+    "timedelta_range",
+    "to_datetime",
+    "to_numeric",
+    "to_timedelta",
+    "unique",
+]
diff --git a/pandas/pandas/core/apply.py b/pandas/pandas/core/apply.py
new file mode 100644
index 0000000000000..f36fc82fb1a11
--- /dev/null
+++ b/pandas/pandas/core/apply.py
@@ -0,0 +1,2000 @@
+from __future__ import annotations
+
+import abc
+from collections import defaultdict
+from collections.abc import Callable
+import functools
+from functools import partial
+import inspect
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs.internals import BlockValuesRefs
+from pandas._typing import (
+    AggFuncType,
+    AggFuncTypeBase,
+    AggFuncTypeDict,
+    AggObjType,
+    Axis,
+    AxisInt,
+    NDFrameT,
+    npt,
+)
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import SpecificationError
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.cast import is_nested_object
+from pandas.core.dtypes.common import (
+    is_dict_like,
+    is_extension_array_dtype,
+    is_list_like,
+    is_numeric_dtype,
+    is_sequence,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCNDFrame,
+    ABCSeries,
+)
+
+from pandas.core._numba.executor import generate_apply_looper
+import pandas.core.common as com
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+from pandas.core.util.numba_ import (
+    get_jit_arguments,
+    prepare_function_arguments,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Hashable,
+        Iterable,
+        MutableMapping,
+        Sequence,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+    from pandas.core.groupby import GroupBy
+    from pandas.core.resample import Resampler
+    from pandas.core.window.rolling import BaseWindow
+
+ResType = dict[int, Any]
+
+
+def frame_apply(
+    obj: DataFrame,
+    func: AggFuncType,
+    axis: Axis = 0,
+    raw: bool = False,
+    result_type: str | None = None,
+    by_row: Literal[False, "compat"] = "compat",
+    engine: str = "python",
+    engine_kwargs: dict[str, bool] | None = None,
+    args=None,
+    kwargs=None,
+) -> FrameApply:
+    """construct and return a row or column based frame apply object"""
+    _, func, columns, _ = reconstruct_func(func, **kwargs)
+
+    axis = obj._get_axis_number(axis)
+    klass: type[FrameApply]
+    if axis == 0:
+        klass = FrameRowApply
+    elif axis == 1:
+        if columns:
+            raise NotImplementedError(
+                f"Named aggregation is not supported when {axis=}."
+            )
+        klass = FrameColumnApply
+
+    return klass(
+        obj,
+        func,
+        raw=raw,
+        result_type=result_type,
+        by_row=by_row,
+        engine=engine,
+        engine_kwargs=engine_kwargs,
+        args=args,
+        kwargs=kwargs,
+    )
+
+
+class Apply(metaclass=abc.ABCMeta):
+    axis: AxisInt
+
+    def __init__(
+        self,
+        obj: AggObjType,
+        func: AggFuncType,
+        raw: bool,
+        result_type: str | None,
+        *,
+        by_row: Literal[False, "compat", "_compat"] = "compat",
+        engine: str = "python",
+        engine_kwargs: dict[str, bool] | None = None,
+        args,
+        kwargs,
+    ) -> None:
+        self.obj = obj
+        self.raw = raw
+
+        assert by_row is False or by_row in ["compat", "_compat"]
+        self.by_row = by_row
+
+        self.args = args or ()
+        self.kwargs = kwargs or {}
+
+        self.engine = engine
+        self.engine_kwargs = {} if engine_kwargs is None else engine_kwargs
+
+        if result_type not in [None, "reduce", "broadcast", "expand"]:
+            raise ValueError(
+                "invalid value for result_type, must be one "
+                "of {None, 'reduce', 'broadcast', 'expand'}"
+            )
+
+        self.result_type = result_type
+
+        self.func = func
+
+    @abc.abstractmethod
+    def apply(self) -> DataFrame | Series:
+        pass
+
+    @abc.abstractmethod
+    def agg_or_apply_list_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        pass
+
+    @abc.abstractmethod
+    def agg_or_apply_dict_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        pass
+
+    def agg(self) -> DataFrame | Series | None:
+        """
+        Provide an implementation for the aggregators.
+
+        Returns
+        -------
+        Result of aggregation, or None if agg cannot be performed by
+        this method.
+        """
+        func = self.func
+
+        if isinstance(func, str):
+            return self.apply_str()
+
+        if is_dict_like(func):
+            return self.agg_dict_like()
+        elif is_list_like(func):
+            # we require a list, but not a 'str'
+            return self.agg_list_like()
+
+        # caller can react
+        return None
+
+    def transform(self) -> DataFrame | Series:
+        """
+        Transform a DataFrame or Series.
+
+        Returns
+        -------
+        DataFrame or Series
+            Result of applying ``func`` along the given axis of the
+            Series or DataFrame.
+
+        Raises
+        ------
+        ValueError
+            If the transform function fails or does not transform.
+        """
+        obj = self.obj
+        func = self.func
+        axis = self.axis
+        args = self.args
+        kwargs = self.kwargs
+
+        is_series = obj.ndim == 1
+
+        if obj._get_axis_number(axis) == 1:
+            assert not is_series
+            return obj.T.transform(func, 0, *args, **kwargs).T
+
+        if is_list_like(func) and not is_dict_like(func):
+            func = cast(list[AggFuncTypeBase], func)
+            # Convert func equivalent dict
+            if is_series:
+                func = {com.get_callable_name(v) or v: v for v in func}
+            else:
+                func = {col: func for col in obj}
+
+        if is_dict_like(func):
+            func = cast(AggFuncTypeDict, func)
+            return self.transform_dict_like(func)
+
+        # func is either str or callable
+        func = cast(AggFuncTypeBase, func)
+        try:
+            result = self.transform_str_or_callable(func)
+        except TypeError:
+            raise
+        except Exception as err:
+            raise ValueError("Transform function failed") from err
+
+        # Functions that transform may return empty Series/DataFrame
+        # when the dtype is not appropriate
+        if (
+            isinstance(result, (ABCSeries, ABCDataFrame))
+            and result.empty
+            and not obj.empty
+        ):
+            raise ValueError("Transform function failed")
+        if not isinstance(result, (ABCSeries, ABCDataFrame)) or not result.index.equals(
+            obj.index
+        ):
+            raise ValueError("Function did not transform")
+
+        return result
+
+    def transform_dict_like(self, func) -> DataFrame:
+        """
+        Compute transform in the case of a dict-like func
+        """
+        from pandas.core.reshape.concat import concat
+
+        obj = self.obj
+        args = self.args
+        kwargs = self.kwargs
+
+        # transform is currently only for Series/DataFrame
+        assert isinstance(obj, ABCNDFrame)
+
+        if len(func) == 0:
+            raise ValueError("No transform functions were provided")
+
+        func = self.normalize_dictlike_arg("transform", obj, func)
+
+        results: dict[Hashable, DataFrame | Series] = {}
+        for name, how in func.items():
+            colg = obj._gotitem(name, ndim=1)
+            results[name] = colg.transform(how, 0, *args, **kwargs)
+        return concat(results, axis=1)
+
+    def transform_str_or_callable(self, func) -> DataFrame | Series:
+        """
+        Compute transform in the case of a string or callable func
+        """
+        obj = self.obj
+        args = self.args
+        kwargs = self.kwargs
+
+        if isinstance(func, str):
+            return self._apply_str(obj, func, *args, **kwargs)
+
+        # Two possible ways to use a UDF - apply or call directly
+        try:
+            return obj.apply(func, args=args, **kwargs)
+        except Exception:
+            return func(obj, *args, **kwargs)
+
+    def agg_list_like(self) -> DataFrame | Series:
+        """
+        Compute aggregation in the case of a list-like argument.
+
+        Returns
+        -------
+        Result of aggregation.
+        """
+        return self.agg_or_apply_list_like(op_name="agg")
+
+    def compute_list_like(
+        self,
+        op_name: Literal["agg", "apply"],
+        selected_obj: Series | DataFrame,
+        kwargs: dict[str, Any],
+    ) -> tuple[list[Hashable] | Index, list[Any]]:
+        """
+        Compute agg/apply results for like-like input.
+
+        Parameters
+        ----------
+        op_name : {"agg", "apply"}
+            Operation being performed.
+        selected_obj : Series or DataFrame
+            Data to perform operation on.
+        kwargs : dict
+            Keyword arguments to pass to the functions.
+
+        Returns
+        -------
+        keys : list[Hashable] or Index
+            Index labels for result.
+        results : list
+            Data for result. When aggregating with a Series, this can contain any
+            Python objects.
+        """
+        func = cast(list[AggFuncTypeBase], self.func)
+        obj = self.obj
+
+        results = []
+        keys = []
+
+        # degenerate case
+        if selected_obj.ndim == 1:
+            for a in func:
+                colg = obj._gotitem(selected_obj.name, ndim=1, subset=selected_obj)
+                args = (
+                    [self.axis, *self.args]
+                    if include_axis(op_name, colg)
+                    else self.args
+                )
+                new_res = getattr(colg, op_name)(a, *args, **kwargs)
+                results.append(new_res)
+
+                # make sure we find a good name
+                name = com.get_callable_name(a) or a
+                keys.append(name)
+
+        else:
+            indices = []
+            for index, col in enumerate(selected_obj):
+                colg = obj._gotitem(col, ndim=1, subset=selected_obj.iloc[:, index])
+                args = (
+                    [self.axis, *self.args]
+                    if include_axis(op_name, colg)
+                    else self.args
+                )
+                new_res = getattr(colg, op_name)(func, *args, **kwargs)
+                results.append(new_res)
+                indices.append(index)
+            # error: Incompatible types in assignment (expression has type "Any |
+            # Index", variable has type "list[Any | Callable[..., Any] | str]")
+            keys = selected_obj.columns.take(indices)  # type: ignore[assignment]
+
+        return keys, results
+
+    def wrap_results_list_like(
+        self, keys: Iterable[Hashable], results: list[Series | DataFrame]
+    ):
+        from pandas.core.reshape.concat import concat
+
+        obj = self.obj
+
+        try:
+            return concat(results, keys=keys, axis=1, sort=False)
+        except TypeError as err:
+            # we are concatting non-NDFrame objects,
+            # e.g. a list of scalars
+            from pandas import Series
+
+            result = Series(results, index=keys, name=obj.name)
+            if is_nested_object(result):
+                raise ValueError(
+                    "cannot combine transform and aggregation operations"
+                ) from err
+            return result
+
+    def agg_dict_like(self) -> DataFrame | Series:
+        """
+        Compute aggregation in the case of a dict-like argument.
+
+        Returns
+        -------
+        Result of aggregation.
+        """
+        return self.agg_or_apply_dict_like(op_name="agg")
+
+    def compute_dict_like(
+        self,
+        op_name: Literal["agg", "apply"],
+        selected_obj: Series | DataFrame,
+        selection: Hashable | Sequence[Hashable],
+        kwargs: dict[str, Any],
+    ) -> tuple[list[Hashable], list[Any]]:
+        """
+        Compute agg/apply results for dict-like input.
+
+        Parameters
+        ----------
+        op_name : {"agg", "apply"}
+            Operation being performed.
+        selected_obj : Series or DataFrame
+            Data to perform operation on.
+        selection : hashable or sequence of hashables
+            Used by GroupBy, Window, and Resample if selection is applied to the object.
+        kwargs : dict
+            Keyword arguments to pass to the functions.
+
+        Returns
+        -------
+        keys : list[hashable]
+            Index labels for result.
+        results : list
+            Data for result. When aggregating with a Series, this can contain any
+            Python object.
+        """
+        from pandas.core.groupby.generic import (
+            DataFrameGroupBy,
+            SeriesGroupBy,
+        )
+
+        obj = self.obj
+        is_groupby = isinstance(obj, (DataFrameGroupBy, SeriesGroupBy))
+        func = cast(AggFuncTypeDict, self.func)
+        func = self.normalize_dictlike_arg(op_name, selected_obj, func)
+
+        is_non_unique_col = (
+            selected_obj.ndim == 2
+            and selected_obj.columns.nunique() < len(selected_obj.columns)
+        )
+
+        if selected_obj.ndim == 1:
+            # key only used for output
+            colg = obj._gotitem(selection, ndim=1)
+            results = [getattr(colg, op_name)(how, **kwargs) for _, how in func.items()]
+            keys = list(func.keys())
+        elif not is_groupby and is_non_unique_col:
+            # key used for column selection and output
+            # GH#51099
+            results = []
+            keys = []
+            for key, how in func.items():
+                indices = selected_obj.columns.get_indexer_for([key])
+                labels = selected_obj.columns.take(indices)
+                label_to_indices = defaultdict(list)
+                for index, label in zip(indices, labels):
+                    label_to_indices[label].append(index)
+
+                key_data = [
+                    getattr(selected_obj._ixs(indice, axis=1), op_name)(how, **kwargs)
+                    for label, indices in label_to_indices.items()
+                    for indice in indices
+                ]
+
+                keys += [key] * len(key_data)
+                results += key_data
+        elif is_groupby:
+            # key used for column selection and output
+
+            df = selected_obj
+            results, keys = [], []
+            for key, how in func.items():
+                cols = df[key]
+
+                if cols.ndim == 1:
+                    series = obj._gotitem(key, ndim=1, subset=cols)
+                    results.append(getattr(series, op_name)(how, **kwargs))
+                    keys.append(key)
+                else:
+                    for _, col in cols.items():
+                        series = obj._gotitem(key, ndim=1, subset=col)
+                        results.append(getattr(series, op_name)(how, **kwargs))
+                        keys.append(key)
+        else:
+            results = [
+                getattr(obj._gotitem(key, ndim=1), op_name)(how, **kwargs)
+                for key, how in func.items()
+            ]
+            keys = list(func.keys())
+
+        return keys, results
+
+    def wrap_results_dict_like(
+        self,
+        selected_obj: Series | DataFrame,
+        result_index: list[Hashable],
+        result_data: list,
+    ):
+        from pandas import Index
+        from pandas.core.reshape.concat import concat
+
+        obj = self.obj
+
+        # Avoid making two isinstance calls in all and any below
+        is_ndframe = [isinstance(r, ABCNDFrame) for r in result_data]
+
+        if all(is_ndframe):
+            results = [result for result in result_data if not result.empty]
+            keys_to_use: Iterable[Hashable]
+            keys_to_use = [k for k, v in zip(result_index, result_data) if not v.empty]
+            # Have to check, if at least one DataFrame is not empty.
+            if keys_to_use == []:
+                keys_to_use = result_index
+                results = result_data
+
+            if selected_obj.ndim == 2:
+                # keys are columns, so we can preserve names
+                ktu = Index(keys_to_use)
+                ktu._set_names(selected_obj.columns.names)
+                keys_to_use = ktu
+
+            axis: AxisInt = 0 if isinstance(obj, ABCSeries) else 1
+            result = concat(
+                results,
+                axis=axis,
+                keys=keys_to_use,
+            )
+        elif any(is_ndframe):
+            # There is a mix of NDFrames and scalars
+            raise ValueError(
+                "cannot perform both aggregation "
+                "and transformation operations "
+                "simultaneously"
+            )
+        else:
+            from pandas import Series
+
+            # we have a list of scalars
+            # GH 36212 use name only if obj is a series
+            if obj.ndim == 1:
+                obj = cast("Series", obj)
+                name = obj.name
+            else:
+                name = None
+
+            result = Series(result_data, index=result_index, name=name)
+
+        return result
+
+    def apply_str(self) -> DataFrame | Series:
+        """
+        Compute apply in case of a string.
+
+        Returns
+        -------
+        result: Series or DataFrame
+        """
+        # Caller is responsible for checking isinstance(self.f, str)
+        func = cast(str, self.func)
+
+        obj = self.obj
+
+        from pandas.core.groupby.generic import (
+            DataFrameGroupBy,
+            SeriesGroupBy,
+        )
+
+        # Support for `frame.transform('method')`
+        # Some methods (shift, etc.) require the axis argument, others
+        # don't, so inspect and insert if necessary.
+        method = getattr(obj, func, None)
+        if callable(method):
+            sig = inspect.getfullargspec(method)
+            arg_names = (*sig.args, *sig.kwonlyargs)
+            if self.axis != 0 and (
+                "axis" not in arg_names or func in ("corrwith", "skew")
+            ):
+                raise ValueError(f"Operation {func} does not support axis=1")
+            if "axis" in arg_names and not isinstance(
+                obj, (SeriesGroupBy, DataFrameGroupBy)
+            ):
+                self.kwargs["axis"] = self.axis
+        return self._apply_str(obj, func, *self.args, **self.kwargs)
+
+    def apply_list_or_dict_like(self) -> DataFrame | Series:
+        """
+        Compute apply in case of a list-like or dict-like.
+
+        Returns
+        -------
+        result: Series, DataFrame, or None
+            Result when self.func is a list-like or dict-like, None otherwise.
+        """
+
+        if self.engine == "numba":
+            raise NotImplementedError(
+                "The 'numba' engine doesn't support list-like/"
+                "dict likes of callables yet."
+            )
+
+        if self.axis == 1 and isinstance(self.obj, ABCDataFrame):
+            return self.obj.T.apply(self.func, 0, args=self.args, **self.kwargs).T
+
+        func = self.func
+        kwargs = self.kwargs
+
+        if is_dict_like(func):
+            result = self.agg_or_apply_dict_like(op_name="apply")
+        else:
+            result = self.agg_or_apply_list_like(op_name="apply")
+
+        result = reconstruct_and_relabel_result(result, func, **kwargs)
+
+        return result
+
+    def normalize_dictlike_arg(
+        self, how: str, obj: DataFrame | Series, func: AggFuncTypeDict
+    ) -> AggFuncTypeDict:
+        """
+        Handler for dict-like argument.
+
+        Ensures that necessary columns exist if obj is a DataFrame, and
+        that a nested renamer is not passed. Also normalizes to all lists
+        when values consists of a mix of list and non-lists.
+        """
+        assert how in ("apply", "agg", "transform")
+
+        # Can't use func.values(); wouldn't work for a Series
+        if (
+            how == "agg"
+            and isinstance(obj, ABCSeries)
+            and any(is_list_like(v) for _, v in func.items())
+        ) or (any(is_dict_like(v) for _, v in func.items())):
+            # GH 15931 - deprecation of renaming keys
+            raise SpecificationError("nested renamer is not supported")
+
+        if obj.ndim != 1:
+            # Check for missing columns on a frame
+            from pandas import Index
+
+            cols = Index(list(func.keys())).difference(obj.columns, sort=True)
+            if len(cols) > 0:
+                # GH 58474
+                raise KeyError(f"Label(s) {list(cols)} do not exist")
+
+        aggregator_types = (list, tuple, dict)
+
+        # if we have a dict of any non-scalars
+        # eg. {'A' : ['mean']}, normalize all to
+        # be list-likes
+        # Cannot use func.values() because arg may be a Series
+        if any(isinstance(x, aggregator_types) for _, x in func.items()):
+            new_func: AggFuncTypeDict = {}
+            for k, v in func.items():
+                if not isinstance(v, aggregator_types):
+                    new_func[k] = [v]
+                else:
+                    new_func[k] = v
+            func = new_func
+        return func
+
+    def _apply_str(self, obj, func: str, *args, **kwargs):
+        """
+        if arg is a string, then try to operate on it:
+        - try to find a function (or attribute) on obj
+        - try to find a numpy function
+        - raise
+        """
+        assert isinstance(func, str)
+
+        if hasattr(obj, func):
+            f = getattr(obj, func)
+            if callable(f):
+                return f(*args, **kwargs)
+
+            # people may aggregate on a non-callable attribute
+            # but don't let them think they can pass args to it
+            assert len(args) == 0
+            assert not any(kwarg == "axis" for kwarg in kwargs)
+            return f
+        elif hasattr(np, func) and hasattr(obj, "__array__"):
+            # in particular exclude Window
+            f = getattr(np, func)
+            return f(obj, *args, **kwargs)
+        else:
+            msg = f"'{func}' is not a valid function for '{type(obj).__name__}' object"
+            raise AttributeError(msg)
+
+
+class NDFrameApply(Apply):
+    """
+    Methods shared by FrameApply and SeriesApply but
+    not GroupByApply or ResamplerWindowApply
+    """
+
+    obj: DataFrame | Series
+
+    @property
+    def index(self) -> Index:
+        return self.obj.index
+
+    @property
+    def agg_axis(self) -> Index:
+        return self.obj._get_agg_axis(self.axis)
+
+    def agg_or_apply_list_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        obj = self.obj
+        kwargs = self.kwargs
+
+        if op_name == "apply":
+            if isinstance(self, FrameApply):
+                by_row = self.by_row
+
+            elif isinstance(self, SeriesApply):
+                by_row = "_compat" if self.by_row else False
+            else:
+                by_row = False
+            kwargs = {**kwargs, "by_row": by_row}
+
+        if getattr(obj, "axis", 0) == 1:
+            raise NotImplementedError("axis other than 0 is not supported")
+
+        keys, results = self.compute_list_like(op_name, obj, kwargs)
+        result = self.wrap_results_list_like(keys, results)
+        return result
+
+    def agg_or_apply_dict_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        assert op_name in ["agg", "apply"]
+        obj = self.obj
+
+        kwargs = {}
+        if op_name == "apply":
+            by_row = "_compat" if self.by_row else False
+            kwargs.update({"by_row": by_row})
+
+        if getattr(obj, "axis", 0) == 1:
+            raise NotImplementedError("axis other than 0 is not supported")
+
+        selection = None
+        result_index, result_data = self.compute_dict_like(
+            op_name, obj, selection, kwargs
+        )
+        result = self.wrap_results_dict_like(obj, result_index, result_data)
+        return result
+
+
+class FrameApply(NDFrameApply):
+    obj: DataFrame
+
+    def __init__(
+        self,
+        obj: AggObjType,
+        func: AggFuncType,
+        raw: bool,
+        result_type: str | None,
+        *,
+        by_row: Literal[False, "compat"] = False,
+        engine: str = "python",
+        engine_kwargs: dict[str, bool] | None = None,
+        args,
+        kwargs,
+    ) -> None:
+        if by_row is not False and by_row != "compat":
+            raise ValueError(f"by_row={by_row} not allowed")
+        super().__init__(
+            obj,
+            func,
+            raw,
+            result_type,
+            by_row=by_row,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    # ---------------------------------------------------------------
+    # Abstract Methods
+
+    @property
+    @abc.abstractmethod
+    def result_index(self) -> Index:
+        pass
+
+    @property
+    @abc.abstractmethod
+    def result_columns(self) -> Index:
+        pass
+
+    @property
+    @abc.abstractmethod
+    def series_generator(self) -> Generator[Series]:
+        pass
+
+    @staticmethod
+    @functools.cache
+    @abc.abstractmethod
+    def generate_numba_apply_func(
+        func, nogil=True, nopython=True, parallel=False
+    ) -> Callable[[npt.NDArray, Index, Index], dict[int, Any]]:
+        pass
+
+    @abc.abstractmethod
+    def apply_with_numba(self):
+        pass
+
+    def validate_values_for_numba(self) -> None:
+        # Validate column dtyps all OK
+        for colname, dtype in self.obj.dtypes.items():
+            if not is_numeric_dtype(dtype):
+                raise ValueError(
+                    f"Column {colname} must have a numeric dtype. "
+                    f"Found '{dtype}' instead"
+                )
+            if is_extension_array_dtype(dtype):
+                raise ValueError(
+                    f"Column {colname} is backed by an extension array, "
+                    f"which is not supported by the numba engine."
+                )
+
+    @abc.abstractmethod
+    def wrap_results_for_axis(
+        self, results: ResType, res_index: Index
+    ) -> DataFrame | Series:
+        pass
+
+    # ---------------------------------------------------------------
+
+    @property
+    def res_columns(self) -> Index:
+        return self.result_columns
+
+    @property
+    def columns(self) -> Index:
+        return self.obj.columns
+
+    @cache_readonly
+    def values(self):
+        return self.obj.values
+
+    def apply(self) -> DataFrame | Series:
+        """compute the results"""
+
+        # dispatch to handle list-like or dict-like
+        if is_list_like(self.func):
+            if self.engine == "numba":
+                raise NotImplementedError(
+                    "the 'numba' engine doesn't support lists of callables yet"
+                )
+            return self.apply_list_or_dict_like()
+
+        # all empty
+        if len(self.columns) == 0 and len(self.index) == 0:
+            return self.apply_empty_result()
+
+        # string dispatch
+        if isinstance(self.func, str):
+            if self.engine == "numba":
+                raise NotImplementedError(
+                    "the 'numba' engine doesn't support using "
+                    "a string as the callable function"
+                )
+            return self.apply_str()
+
+        # ufunc
+        elif isinstance(self.func, np.ufunc):
+            if self.engine == "numba":
+                raise NotImplementedError(
+                    "the 'numba' engine doesn't support "
+                    "using a numpy ufunc as the callable function"
+                )
+            with np.errstate(all="ignore"):
+                results = self.obj._mgr.apply("apply", func=self.func)
+            # _constructor will retain self.index and self.columns
+            return self.obj._constructor_from_mgr(results, axes=results.axes)
+
+        # broadcasting
+        if self.result_type == "broadcast":
+            if self.engine == "numba":
+                raise NotImplementedError(
+                    "the 'numba' engine doesn't support result_type='broadcast'"
+                )
+            return self.apply_broadcast(self.obj)
+
+        # one axis empty
+        elif not all(self.obj.shape):
+            return self.apply_empty_result()
+
+        # raw
+        elif self.raw:
+            return self.apply_raw(engine=self.engine, engine_kwargs=self.engine_kwargs)
+
+        return self.apply_standard()
+
+    def agg(self):
+        obj = self.obj
+        axis = self.axis
+
+        # TODO: Avoid having to change state
+        self.obj = self.obj if self.axis == 0 else self.obj.T
+        self.axis = 0
+
+        result = None
+        try:
+            result = super().agg()
+        finally:
+            self.obj = obj
+            self.axis = axis
+
+        if axis == 1:
+            result = result.T if result is not None else result
+
+        if result is None:
+            result = self.obj.apply(self.func, axis, args=self.args, **self.kwargs)
+
+        return result
+
+    def apply_empty_result(self):
+        """
+        we have an empty result; at least 1 axis is 0
+
+        we will try to apply the function to an empty
+        series in order to see if this is a reduction function
+        """
+        assert callable(self.func)
+
+        # we are not asked to reduce or infer reduction
+        # so just return a copy of the existing object
+        if self.result_type not in ["reduce", None]:
+            return self.obj.copy()
+
+        # we may need to infer
+        should_reduce = self.result_type == "reduce"
+
+        from pandas import Series
+
+        if not should_reduce:
+            try:
+                if self.axis == 0:
+                    r = self.func(
+                        Series([], dtype=np.float64), *self.args, **self.kwargs
+                    )
+                else:
+                    r = self.func(
+                        Series(index=self.columns, dtype=np.float64),
+                        *self.args,
+                        **self.kwargs,
+                    )
+            except Exception:
+                pass
+            else:
+                should_reduce = not isinstance(r, Series)
+
+        if should_reduce:
+            if len(self.agg_axis):
+                r = self.func(Series([], dtype=np.float64), *self.args, **self.kwargs)
+            else:
+                r = np.nan
+
+            return self.obj._constructor_sliced(r, index=self.agg_axis)
+        else:
+            return self.obj.copy()
+
+    def apply_raw(self, engine="python", engine_kwargs=None):
+        """apply to the values as a numpy array"""
+
+        def wrap_function(func):
+            """
+            Wrap user supplied function to work around numpy issue.
+
+            see https://github.com/numpy/numpy/issues/8352
+            """
+
+            def wrapper(*args, **kwargs):
+                result = func(*args, **kwargs)
+                if isinstance(result, str):
+                    result = np.array(result, dtype=object)
+                return result
+
+            return wrapper
+
+        if engine == "numba":
+            args, kwargs = prepare_function_arguments(
+                self.func,  # type: ignore[arg-type]
+                self.args,
+                self.kwargs,
+                num_required_args=1,
+            )
+            # error: Argument 1 to "__call__" of "_lru_cache_wrapper" has
+            # incompatible type "Callable[..., Any] | str | list[Callable
+            # [..., Any] | str] | dict[Hashable,Callable[..., Any] | str |
+            # list[Callable[..., Any] | str]]"; expected "Hashable"
+            nb_looper = generate_apply_looper(
+                self.func,  # type: ignore[arg-type]
+                **get_jit_arguments(engine_kwargs),
+            )
+            result = nb_looper(self.values, self.axis, *args)
+            # If we made the result 2-D, squeeze it back to 1-D
+            result = np.squeeze(result)
+        else:
+            result = np.apply_along_axis(
+                wrap_function(self.func),
+                self.axis,
+                self.values,
+                *self.args,
+                **self.kwargs,
+            )
+
+        # TODO: mixed type case
+        if result.ndim == 2:
+            return self.obj._constructor(result, index=self.index, columns=self.columns)
+        else:
+            return self.obj._constructor_sliced(result, index=self.agg_axis)
+
+    def apply_broadcast(self, target: DataFrame) -> DataFrame:
+        assert callable(self.func)
+
+        result_values = np.empty_like(target.values)
+
+        # axis which we want to compare compliance
+        result_compare = target.shape[0]
+
+        for i, col in enumerate(target.columns):
+            res = self.func(target[col], *self.args, **self.kwargs)
+            ares = np.asarray(res).ndim
+
+            # must be a scalar or 1d
+            if ares > 1:
+                raise ValueError("too many dims to broadcast")
+            if ares == 1:
+                # must match return dim
+                if result_compare != len(res):
+                    raise ValueError("cannot broadcast result")
+
+            result_values[:, i] = res
+
+        # we *always* preserve the original index / columns
+        result = self.obj._constructor(
+            result_values, index=target.index, columns=target.columns
+        )
+        return result
+
+    def apply_standard(self):
+        if self.engine == "python":
+            results, res_index = self.apply_series_generator()
+        else:
+            results, res_index = self.apply_series_numba()
+
+        # wrap results
+        return self.wrap_results(results, res_index)
+
+    def apply_series_generator(self) -> tuple[ResType, Index]:
+        assert callable(self.func)
+
+        series_gen = self.series_generator
+        res_index = self.result_index
+
+        results = {}
+
+        for i, v in enumerate(series_gen):
+            results[i] = self.func(v, *self.args, **self.kwargs)
+            if isinstance(results[i], ABCSeries):
+                # If we have a view on v, we need to make a copy because
+                #  series_generator will swap out the underlying data
+                results[i] = results[i].copy(deep=False)
+
+        return results, res_index
+
+    def apply_series_numba(self):
+        if self.engine_kwargs.get("parallel", False):
+            raise NotImplementedError(
+                "Parallel apply is not supported when raw=False and engine='numba'"
+            )
+        if not self.obj.index.is_unique or not self.columns.is_unique:
+            raise NotImplementedError(
+                "The index/columns must be unique when raw=False and engine='numba'"
+            )
+        self.validate_values_for_numba()
+        results = self.apply_with_numba()
+        return results, self.result_index
+
+    def wrap_results(self, results: ResType, res_index: Index) -> DataFrame | Series:
+        from pandas import Series
+
+        # see if we can infer the results
+        if len(results) > 0 and 0 in results and is_sequence(results[0]):
+            return self.wrap_results_for_axis(results, res_index)
+
+        # dict of scalars
+
+        # the default dtype of an empty Series is `object`, but this
+        # code can be hit by df.mean() where the result should have dtype
+        # float64 even if it's an empty Series.
+        constructor_sliced = self.obj._constructor_sliced
+        if len(results) == 0 and constructor_sliced is Series:
+            result = constructor_sliced(results, dtype=np.float64)
+        else:
+            result = constructor_sliced(results)
+        result.index = res_index
+
+        return result
+
+    def apply_str(self) -> DataFrame | Series:
+        # Caller is responsible for checking isinstance(self.func, str)
+        # TODO: GH#39993 - Avoid special-casing by replacing with lambda
+        if self.func == "size":
+            # Special-cased because DataFrame.size returns a single scalar
+            obj = self.obj
+            value = obj.shape[self.axis]
+            return obj._constructor_sliced(value, index=self.agg_axis)
+        return super().apply_str()
+
+
+class FrameRowApply(FrameApply):
+    axis: AxisInt = 0
+
+    @property
+    def series_generator(self) -> Generator[Series]:
+        return (self.obj._ixs(i, axis=1) for i in range(len(self.columns)))
+
+    @staticmethod
+    @functools.cache
+    def generate_numba_apply_func(
+        func, nogil=True, nopython=True, parallel=False
+    ) -> Callable[[npt.NDArray, Index, Index], dict[int, Any]]:
+        numba = import_optional_dependency("numba")
+        from pandas import Series
+
+        # Import helper from extensions to cast string object -> np strings
+        # Note: This also has the side effect of loading our numba extensions
+        from pandas.core._numba.extensions import maybe_cast_str
+
+        jitted_udf = numba.extending.register_jitable(func)
+
+        # Currently the parallel argument doesn't get passed through here
+        # (it's disabled) since the dicts in numba aren't thread-safe.
+        @numba.jit(nogil=nogil, nopython=nopython, parallel=parallel)
+        def numba_func(values, col_names, df_index, *args):
+            results = {}
+            for j in range(values.shape[1]):
+                # Create the series
+                ser = Series(
+                    values[:, j], index=df_index, name=maybe_cast_str(col_names[j])
+                )
+                results[j] = jitted_udf(ser, *args)
+            return results
+
+        return numba_func
+
+    def apply_with_numba(self) -> dict[int, Any]:
+        func = cast(Callable, self.func)
+        args, kwargs = prepare_function_arguments(
+            func, self.args, self.kwargs, num_required_args=1
+        )
+        nb_func = self.generate_numba_apply_func(
+            func, **get_jit_arguments(self.engine_kwargs)
+        )
+        from pandas.core._numba.extensions import set_numba_data
+
+        index = self.obj.index
+        columns = self.obj.columns
+
+        # Convert from numba dict to regular dict
+        # Our isinstance checks in the df constructor don't pass for numbas typed dict
+        with set_numba_data(index) as index, set_numba_data(columns) as columns:
+            res = dict(nb_func(self.values, columns, index, *args))
+        return res
+
+    @property
+    def result_index(self) -> Index:
+        return self.columns
+
+    @property
+    def result_columns(self) -> Index:
+        return self.index
+
+    def wrap_results_for_axis(
+        self, results: ResType, res_index: Index
+    ) -> DataFrame | Series:
+        """return the results for the rows"""
+
+        if self.result_type == "reduce":
+            # e.g. test_apply_dict GH#8735
+            res = self.obj._constructor_sliced(results)
+            res.index = res_index
+            return res
+
+        elif self.result_type is None and all(
+            isinstance(x, dict) for x in results.values()
+        ):
+            # Our operation was a to_dict op e.g.
+            #  test_apply_dict GH#8735, test_apply_reduce_to_dict GH#25196 #37544
+            res = self.obj._constructor_sliced(results)
+            res.index = res_index
+            return res
+
+        try:
+            result = self.obj._constructor(data=results)
+        except ValueError as err:
+            if "All arrays must be of the same length" in str(err):
+                # e.g. result = [[2, 3], [1.5], ['foo', 'bar']]
+                #  see test_agg_listlike_result GH#29587
+                res = self.obj._constructor_sliced(results)
+                res.index = res_index
+                return res
+            else:
+                raise
+
+        if not isinstance(results[0], ABCSeries):
+            if len(result.index) == len(self.res_columns):
+                result.index = self.res_columns
+
+        if len(result.columns) == len(res_index):
+            result.columns = res_index
+
+        return result
+
+
+class FrameColumnApply(FrameApply):
+    axis: AxisInt = 1
+
+    def apply_broadcast(self, target: DataFrame) -> DataFrame:
+        result = super().apply_broadcast(target.T)
+        return result.T
+
+    @property
+    def series_generator(self) -> Generator[Series]:
+        values = self.values
+        values = ensure_wrapped_if_datetimelike(values)
+        assert len(values) > 0
+
+        # We create one Series object, and will swap out the data inside
+        #  of it.  Kids: don't do this at home.
+        ser = self.obj._ixs(0, axis=0)
+        mgr = ser._mgr
+
+        is_view = mgr.blocks[0].refs.has_reference()
+
+        if isinstance(ser.dtype, ExtensionDtype):
+            # values will be incorrect for this block
+            # TODO(EA2D): special case would be unnecessary with 2D EAs
+            obj = self.obj
+            for i in range(len(obj)):
+                yield obj._ixs(i, axis=0)
+
+        else:
+            for arr, name in zip(values, self.index):
+                # GH#35462 re-pin mgr in case setitem changed it
+                ser._mgr = mgr
+                mgr.set_values(arr)
+                object.__setattr__(ser, "_name", name)
+                if not is_view:
+                    # In apply_series_generator we store the a shallow copy of the
+                    # result, which potentially increases the ref count of this reused
+                    # `ser` object (depending on the result of the applied function)
+                    # -> if that happened and `ser` is already a copy, then we reset
+                    # the refs here to avoid triggering a unnecessary CoW inside the
+                    # applied function (https://github.com/pandas-dev/pandas/pull/56212)
+                    mgr.blocks[0].refs = BlockValuesRefs(mgr.blocks[0])
+                yield ser
+
+    @staticmethod
+    @functools.cache
+    def generate_numba_apply_func(
+        func, nogil=True, nopython=True, parallel=False
+    ) -> Callable[[npt.NDArray, Index, Index], dict[int, Any]]:
+        numba = import_optional_dependency("numba")
+        from pandas import Series
+        from pandas.core._numba.extensions import maybe_cast_str
+
+        jitted_udf = numba.extending.register_jitable(func)
+
+        @numba.jit(nogil=nogil, nopython=nopython, parallel=parallel)
+        def numba_func(values, col_names_index, index, *args):
+            results = {}
+            # Currently the parallel argument doesn't get passed through here
+            # (it's disabled) since the dicts in numba aren't thread-safe.
+            for i in range(values.shape[0]):
+                # Create the series
+                # TODO: values corrupted without the copy
+                ser = Series(
+                    values[i].copy(),
+                    index=col_names_index,
+                    name=maybe_cast_str(index[i]),
+                )
+                results[i] = jitted_udf(ser, *args)
+
+            return results
+
+        return numba_func
+
+    def apply_with_numba(self) -> dict[int, Any]:
+        func = cast(Callable, self.func)
+        args, kwargs = prepare_function_arguments(
+            func, self.args, self.kwargs, num_required_args=1
+        )
+        nb_func = self.generate_numba_apply_func(
+            func, **get_jit_arguments(self.engine_kwargs)
+        )
+
+        from pandas.core._numba.extensions import set_numba_data
+
+        # Convert from numba dict to regular dict
+        # Our isinstance checks in the df constructor don't pass for numbas typed dict
+        with (
+            set_numba_data(self.obj.index) as index,
+            set_numba_data(self.columns) as columns,
+        ):
+            res = dict(nb_func(self.values, columns, index, *args))
+
+        return res
+
+    @property
+    def result_index(self) -> Index:
+        return self.index
+
+    @property
+    def result_columns(self) -> Index:
+        return self.columns
+
+    def wrap_results_for_axis(
+        self, results: ResType, res_index: Index
+    ) -> DataFrame | Series:
+        """return the results for the columns"""
+        result: DataFrame | Series
+
+        # we have requested to expand
+        if self.result_type == "expand":
+            result = self.infer_to_same_shape(results, res_index)
+
+        # we have a non-series and don't want inference
+        elif not isinstance(results[0], ABCSeries):
+            result = self.obj._constructor_sliced(results)
+            result.index = res_index
+
+        # we may want to infer results
+        else:
+            result = self.infer_to_same_shape(results, res_index)
+
+        return result
+
+    def infer_to_same_shape(self, results: ResType, res_index: Index) -> DataFrame:
+        """infer the results to the same shape as the input object"""
+        result = self.obj._constructor(data=results)
+        result = result.T
+
+        # set the index
+        result.index = res_index
+
+        # infer dtypes
+        result = result.infer_objects()
+
+        return result
+
+
+class SeriesApply(NDFrameApply):
+    obj: Series
+    axis: AxisInt = 0
+    by_row: Literal[False, "compat", "_compat"]  # only relevant for apply()
+
+    def __init__(
+        self,
+        obj: Series,
+        func: AggFuncType,
+        *,
+        by_row: Literal[False, "compat", "_compat"] = "compat",
+        args,
+        kwargs,
+    ) -> None:
+        super().__init__(
+            obj,
+            func,
+            raw=False,
+            result_type=None,
+            by_row=by_row,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    def apply(self) -> DataFrame | Series:
+        obj = self.obj
+
+        if len(obj) == 0:
+            return self.apply_empty_result()
+
+        # dispatch to handle list-like or dict-like
+        if is_list_like(self.func):
+            return self.apply_list_or_dict_like()
+
+        if isinstance(self.func, str):
+            # if we are a string, try to dispatch
+            return self.apply_str()
+
+        if self.by_row == "_compat":
+            return self.apply_compat()
+
+        # self.func is Callable
+        return self.apply_standard()
+
+    def agg(self):
+        result = super().agg()
+        if result is None:
+            obj = self.obj
+            func = self.func
+            # string, list-like, and dict-like are entirely handled in super
+            assert callable(func)
+            result = func(obj, *self.args, **self.kwargs)
+        return result
+
+    def apply_empty_result(self) -> Series:
+        obj = self.obj
+        return obj._constructor(dtype=obj.dtype, index=obj.index).__finalize__(
+            obj, method="apply"
+        )
+
+    def apply_compat(self):
+        """compat apply method for funcs in listlikes and dictlikes.
+
+         Used for each callable when giving listlikes and dictlikes of callables to
+         apply. Needed for compatibility with Pandas < v2.1.
+
+        .. versionadded:: 2.1.0
+        """
+        obj = self.obj
+        func = self.func
+
+        if callable(func):
+            f = com.get_cython_func(func)
+            if f and not self.args and not self.kwargs:
+                return obj.apply(func, by_row=False)
+
+        try:
+            result = obj.apply(func, by_row="compat")
+        except (ValueError, AttributeError, TypeError):
+            result = obj.apply(func, by_row=False)
+        return result
+
+    def apply_standard(self) -> DataFrame | Series:
+        # caller is responsible for ensuring that f is Callable
+        func = cast(Callable, self.func)
+        obj = self.obj
+
+        if isinstance(func, np.ufunc):
+            with np.errstate(all="ignore"):
+                return func(obj, *self.args, **self.kwargs)
+        elif not self.by_row:
+            return func(obj, *self.args, **self.kwargs)
+
+        if self.args or self.kwargs:
+            # _map_values does not support args/kwargs
+            def curried(x):
+                return func(x, *self.args, **self.kwargs)
+
+        else:
+            curried = func
+        mapped = obj._map_values(mapper=curried)
+
+        if len(mapped) and isinstance(mapped[0], ABCSeries):
+            # GH#43986 Need to do list(mapped) in order to get treated as nested
+            #  See also GH#25959 regarding EA support
+            return obj._constructor_expanddim(list(mapped), index=obj.index)
+        else:
+            return obj._constructor(mapped, index=obj.index).__finalize__(
+                obj, method="apply"
+            )
+
+
+class GroupByApply(Apply):
+    obj: GroupBy | Resampler | BaseWindow
+
+    def __init__(
+        self,
+        obj: GroupBy[NDFrameT],
+        func: AggFuncType,
+        *,
+        args,
+        kwargs,
+    ) -> None:
+        kwargs = kwargs.copy()
+        self.axis = obj.obj._get_axis_number(kwargs.get("axis", 0))
+        super().__init__(
+            obj,
+            func,
+            raw=False,
+            result_type=None,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    def apply(self):
+        raise NotImplementedError
+
+    def transform(self):
+        raise NotImplementedError
+
+    def agg_or_apply_list_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        obj = self.obj
+        kwargs = self.kwargs
+        if op_name == "apply":
+            kwargs = {**kwargs, "by_row": False}
+
+        if getattr(obj, "axis", 0) == 1:
+            raise NotImplementedError("axis other than 0 is not supported")
+
+        if obj._selected_obj.ndim == 1:
+            # For SeriesGroupBy this matches _obj_with_exclusions
+            selected_obj = obj._selected_obj
+        else:
+            selected_obj = obj._obj_with_exclusions
+
+        # Only set as_index=True on groupby objects, not Window or Resample
+        # that inherit from this class.
+        with com.temp_setattr(
+            obj, "as_index", True, condition=hasattr(obj, "as_index")
+        ):
+            keys, results = self.compute_list_like(op_name, selected_obj, kwargs)
+        result = self.wrap_results_list_like(keys, results)
+        return result
+
+    def agg_or_apply_dict_like(
+        self, op_name: Literal["agg", "apply"]
+    ) -> DataFrame | Series:
+        from pandas.core.groupby.generic import (
+            DataFrameGroupBy,
+            SeriesGroupBy,
+        )
+
+        assert op_name in ["agg", "apply"]
+
+        obj = self.obj
+        kwargs = {}
+        if op_name == "apply":
+            by_row = "_compat" if self.by_row else False
+            kwargs.update({"by_row": by_row})
+
+        if getattr(obj, "axis", 0) == 1:
+            raise NotImplementedError("axis other than 0 is not supported")
+
+        selected_obj = obj._selected_obj
+        selection = obj._selection
+
+        is_groupby = isinstance(obj, (DataFrameGroupBy, SeriesGroupBy))
+
+        # Numba Groupby engine/engine-kwargs passthrough
+        if is_groupby:
+            engine = self.kwargs.get("engine", None)
+            engine_kwargs = self.kwargs.get("engine_kwargs", None)
+            kwargs.update({"engine": engine, "engine_kwargs": engine_kwargs})
+
+        with com.temp_setattr(
+            obj, "as_index", True, condition=hasattr(obj, "as_index")
+        ):
+            result_index, result_data = self.compute_dict_like(
+                op_name, selected_obj, selection, kwargs
+            )
+        result = self.wrap_results_dict_like(selected_obj, result_index, result_data)
+        return result
+
+
+class ResamplerWindowApply(GroupByApply):
+    axis: AxisInt = 0
+    obj: Resampler | BaseWindow
+
+    def __init__(
+        self,
+        obj: Resampler | BaseWindow,
+        func: AggFuncType,
+        *,
+        args,
+        kwargs,
+    ) -> None:
+        super(GroupByApply, self).__init__(
+            obj,
+            func,
+            raw=False,
+            result_type=None,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    def apply(self):
+        raise NotImplementedError
+
+    def transform(self):
+        raise NotImplementedError
+
+
+def reconstruct_func(
+    func: AggFuncType | None, **kwargs
+) -> tuple[bool, AggFuncType, tuple[str, ...] | None, npt.NDArray[np.intp] | None]:
+    """
+    This is the internal function to reconstruct func given if there is relabeling
+    or not and also normalize the keyword to get new order of columns.
+
+    If named aggregation is applied, `func` will be None, and kwargs contains the
+    column and aggregation function information to be parsed;
+    If named aggregation is not applied, `func` is either string (e.g. 'min') or
+    Callable, or list of them (e.g. ['min', np.max]), or the dictionary of column name
+    and str/Callable/list of them (e.g. {'A': 'min'}, or {'A': [np.min, lambda x: x]})
+
+    If relabeling is True, will return relabeling, reconstructed func, column
+    names, and the reconstructed order of columns.
+    If relabeling is False, the columns and order will be None.
+
+    Parameters
+    ----------
+    func: agg function (e.g. 'min' or Callable) or list of agg functions
+        (e.g. ['min', np.max]) or dictionary (e.g. {'A': ['min', np.max]}).
+    **kwargs: dict, kwargs used in is_multi_agg_with_relabel and
+        normalize_keyword_aggregation function for relabelling
+
+    Returns
+    -------
+    relabelling: bool, if there is relabelling or not
+    func: normalized and mangled func
+    columns: tuple of column names
+    order: array of columns indices
+
+    Examples
+    --------
+    >>> reconstruct_func(None, **{"foo": ("col", "min")})
+    (True, defaultdict(<class 'list'>, {'col': ['min']}), ('foo',), array([0]))
+
+    >>> reconstruct_func("min")
+    (False, 'min', None, None)
+    """
+    relabeling = func is None and is_multi_agg_with_relabel(**kwargs)
+    columns: tuple[str, ...] | None = None
+    order: npt.NDArray[np.intp] | None = None
+
+    if not relabeling:
+        if isinstance(func, list) and len(func) > len(set(func)):
+            # GH 28426 will raise error if duplicated function names are used and
+            # there is no reassigned name
+            raise SpecificationError(
+                "Function names must be unique if there is no new column names assigned"
+            )
+        if func is None:
+            # nicer error message
+            raise TypeError("Must provide 'func' or tuples of '(column, aggfunc).")
+
+    if relabeling:
+        # error: Incompatible types in assignment (expression has type
+        # "MutableMapping[Hashable, list[Callable[..., Any] | str]]", variable has type
+        # "Callable[..., Any] | str | list[Callable[..., Any] | str] |
+        # MutableMapping[Hashable, Callable[..., Any] | str | list[Callable[..., Any] |
+        # str]] | None")
+        func, columns, order = normalize_keyword_aggregation(  # type: ignore[assignment]
+            kwargs
+        )
+    assert func is not None
+
+    return relabeling, func, columns, order
+
+
+def is_multi_agg_with_relabel(**kwargs) -> bool:
+    """
+    Check whether kwargs passed to .agg look like multi-agg with relabeling.
+
+    Parameters
+    ----------
+    **kwargs : dict
+
+    Returns
+    -------
+    bool
+
+    Examples
+    --------
+    >>> is_multi_agg_with_relabel(a="max")
+    False
+    >>> is_multi_agg_with_relabel(a_max=("a", "max"), a_min=("a", "min"))
+    True
+    >>> is_multi_agg_with_relabel()
+    False
+    """
+    return all(isinstance(v, tuple) and len(v) == 2 for v in kwargs.values()) and (
+        len(kwargs) > 0
+    )
+
+
+def normalize_keyword_aggregation(
+    kwargs: dict,
+) -> tuple[
+    MutableMapping[Hashable, list[AggFuncTypeBase]],
+    tuple[str, ...],
+    npt.NDArray[np.intp],
+]:
+    """
+    Normalize user-provided "named aggregation" kwargs.
+    Transforms from the new ``Mapping[str, NamedAgg]`` style kwargs
+    to the old Dict[str, List[scalar]]].
+
+    Parameters
+    ----------
+    kwargs : dict
+
+    Returns
+    -------
+    aggspec : dict
+        The transformed kwargs.
+    columns : tuple[str, ...]
+        The user-provided keys.
+    col_idx_order : List[int]
+        List of columns indices.
+
+    Examples
+    --------
+    >>> normalize_keyword_aggregation({"output": ("input", "sum")})
+    (defaultdict(<class 'list'>, {'input': ['sum']}), ('output',), array([0]))
+    """
+    from pandas.core.indexes.base import Index
+
+    # Normalize the aggregation functions as Mapping[column, List[func]],
+    # process normally, then fixup the names.
+    # TODO: aggspec type: typing.Dict[str, List[AggScalar]]
+    aggspec = defaultdict(list)
+    order = []
+    columns = tuple(kwargs.keys())
+
+    for column, aggfunc in kwargs.values():
+        aggspec[column].append(aggfunc)
+        order.append((column, com.get_callable_name(aggfunc) or aggfunc))
+
+    # uniquify aggfunc name if duplicated in order list
+    uniquified_order = _make_unique_kwarg_list(order)
+
+    # GH 25719, due to aggspec will change the order of assigned columns in aggregation
+    # uniquified_aggspec will store uniquified order list and will compare it with order
+    # based on index
+    aggspec_order = [
+        (column, com.get_callable_name(aggfunc) or aggfunc)
+        for column, aggfuncs in aggspec.items()
+        for aggfunc in aggfuncs
+    ]
+    uniquified_aggspec = _make_unique_kwarg_list(aggspec_order)
+
+    # get the new index of columns by comparison
+    col_idx_order = Index(uniquified_aggspec).get_indexer(uniquified_order)
+    return aggspec, columns, col_idx_order
+
+
+def _make_unique_kwarg_list(
+    seq: Sequence[tuple[Any, Any]],
+) -> Sequence[tuple[Any, Any]]:
+    """
+    Uniquify aggfunc name of the pairs in the order list
+
+    Examples:
+    --------
+    >>> kwarg_list = [("a", "<lambda>"), ("a", "<lambda>"), ("b", "<lambda>")]
+    >>> _make_unique_kwarg_list(kwarg_list)
+    [('a', '<lambda>_0'), ('a', '<lambda>_1'), ('b', '<lambda>')]
+    """
+    return [
+        (pair[0], f"{pair[1]}_{seq[:i].count(pair)}") if seq.count(pair) > 1 else pair
+        for i, pair in enumerate(seq)
+    ]
+
+
+def relabel_result(
+    result: DataFrame | Series,
+    func: dict[str, list[Callable | str]],
+    columns: Iterable[Hashable],
+    order: Iterable[int],
+) -> dict[Hashable, Series]:
+    """
+    Internal function to reorder result if relabelling is True for
+    dataframe.agg, and return the reordered result in dict.
+
+    Parameters:
+    ----------
+    result: Result from aggregation
+    func: Dict of (column name, funcs)
+    columns: New columns name for relabelling
+    order: New order for relabelling
+
+    Examples
+    --------
+    >>> from pandas.core.apply import relabel_result
+    >>> result = pd.DataFrame(
+    ...     {"A": [np.nan, 2, np.nan], "C": [6, np.nan, np.nan], "B": [np.nan, 4, 2.5]},
+    ...     index=["max", "mean", "min"],
+    ... )
+    >>> funcs = {"A": ["max"], "C": ["max"], "B": ["mean", "min"]}
+    >>> columns = ("foo", "aab", "bar", "dat")
+    >>> order = [0, 1, 2, 3]
+    >>> result_in_dict = relabel_result(result, funcs, columns, order)
+    >>> pd.DataFrame(result_in_dict, index=columns)
+           A    C    B
+    foo  2.0  NaN  NaN
+    aab  NaN  6.0  NaN
+    bar  NaN  NaN  4.0
+    dat  NaN  NaN  2.5
+    """
+    from pandas.core.indexes.base import Index
+
+    reordered_indexes = [
+        pair[0] for pair in sorted(zip(columns, order), key=lambda t: t[1])
+    ]
+    reordered_result_in_dict: dict[Hashable, Series] = {}
+    idx = 0
+
+    reorder_mask = not isinstance(result, ABCSeries) and len(result.columns) > 1
+    for col, fun in func.items():
+        s = result[col].dropna()
+
+        # In the `_aggregate`, the callable names are obtained and used in `result`, and
+        # these names are ordered alphabetically. e.g.
+        #           C2   C1
+        # <lambda>   1  NaN
+        # amax     NaN  4.0
+        # max      NaN  4.0
+        # sum     18.0  6.0
+        # Therefore, the order of functions for each column could be shuffled
+        # accordingly so need to get the callable name if it is not parsed names, and
+        # reorder the aggregated result for each column.
+        # e.g. if df.agg(c1=("C2", sum), c2=("C2", lambda x: min(x))), correct order is
+        # [sum, <lambda>], but in `result`, it will be [<lambda>, sum], and we need to
+        # reorder so that aggregated values map to their functions regarding the order.
+
+        # However there is only one column being used for aggregation, not need to
+        # reorder since the index is not sorted, and keep as is in `funcs`, e.g.
+        #         A
+        # min   1.0
+        # mean  1.5
+        # mean  1.5
+        if reorder_mask:
+            fun = [
+                com.get_callable_name(f) if not isinstance(f, str) else f for f in fun
+            ]
+            col_idx_order = Index(s.index).get_indexer(fun)
+            valid_idx = col_idx_order != -1
+            if valid_idx.any():
+                s = s.iloc[col_idx_order[valid_idx]]
+        # assign the new user-provided "named aggregation" as index names, and reindex
+        # it based on the whole user-provided names.
+        if not s.empty:
+            s.index = reordered_indexes[idx : idx + len(fun)]
+        reordered_result_in_dict[col] = s.reindex(columns)
+        idx = idx + len(fun)
+    return reordered_result_in_dict
+
+
+def reconstruct_and_relabel_result(result, func, **kwargs) -> DataFrame | Series:
+    from pandas import DataFrame
+
+    relabeling, func, columns, order = reconstruct_func(func, **kwargs)
+
+    if relabeling:
+        # This is to keep the order to columns occurrence unchanged, and also
+        # keep the order of new columns occurrence unchanged
+
+        # For the return values of reconstruct_func, if relabeling is
+        # False, columns and order will be None.
+        assert columns is not None
+        assert order is not None
+
+        result_in_dict = relabel_result(result, func, columns, order)
+        result = DataFrame(result_in_dict, index=columns)
+
+    return result
+
+
+# TODO: Can't use, because mypy doesn't like us setting __name__
+#   error: "partial[Any]" has no attribute "__name__"
+# the type is:
+#   typing.Sequence[Callable[..., ScalarResult]]
+#     -> typing.Sequence[Callable[..., ScalarResult]]:
+
+
+def _managle_lambda_list(aggfuncs: Sequence[Any]) -> Sequence[Any]:
+    """
+    Possibly mangle a list of aggfuncs.
+
+    Parameters
+    ----------
+    aggfuncs : Sequence
+
+    Returns
+    -------
+    mangled: list-like
+        A new AggSpec sequence, where lambdas have been converted
+        to have unique names.
+
+    Notes
+    -----
+    If just one aggfunc is passed, the name will not be mangled.
+    """
+    if len(aggfuncs) <= 1:
+        # don't mangle for .agg([lambda x: .])
+        return aggfuncs
+    i = 0
+    mangled_aggfuncs = []
+    for aggfunc in aggfuncs:
+        if com.get_callable_name(aggfunc) == "<lambda>":
+            aggfunc = partial(aggfunc)
+            aggfunc.__name__ = f"<lambda_{i}>"
+            i += 1
+        mangled_aggfuncs.append(aggfunc)
+
+    return mangled_aggfuncs
+
+
+def maybe_mangle_lambdas(agg_spec: Any) -> Any:
+    """
+    Make new lambdas with unique names.
+
+    Parameters
+    ----------
+    agg_spec : Any
+        An argument to GroupBy.agg.
+        Non-dict-like `agg_spec` are pass through as is.
+        For dict-like `agg_spec` a new spec is returned
+        with name-mangled lambdas.
+
+    Returns
+    -------
+    mangled : Any
+        Same type as the input.
+
+    Examples
+    --------
+    >>> maybe_mangle_lambdas("sum")
+    'sum'
+    >>> maybe_mangle_lambdas([lambda: 1, lambda: 2])  # doctest: +SKIP
+    [<function __main__.<lambda_0>,
+     <function pandas...._make_lambda.<locals>.f(*args, **kwargs)>]
+    """
+    is_dict = is_dict_like(agg_spec)
+    if not (is_dict or is_list_like(agg_spec)):
+        return agg_spec
+    mangled_aggspec = type(agg_spec)()  # dict or OrderedDict
+
+    if is_dict:
+        for key, aggfuncs in agg_spec.items():
+            if is_list_like(aggfuncs) and not is_dict_like(aggfuncs):
+                mangled_aggfuncs = _managle_lambda_list(aggfuncs)
+            else:
+                mangled_aggfuncs = aggfuncs
+
+            mangled_aggspec[key] = mangled_aggfuncs
+    else:
+        mangled_aggspec = _managle_lambda_list(agg_spec)
+
+    return mangled_aggspec
+
+
+def validate_func_kwargs(
+    kwargs: dict,
+) -> tuple[list[str], list[str | Callable[..., Any]]]:
+    """
+    Validates types of user-provided "named aggregation" kwargs.
+    `TypeError` is raised if aggfunc is not `str` or callable.
+
+    Parameters
+    ----------
+    kwargs : dict
+
+    Returns
+    -------
+    columns : List[str]
+        List of user-provided keys.
+    func : List[Union[str, callable[...,Any]]]
+        List of user-provided aggfuncs
+
+    Examples
+    --------
+    >>> validate_func_kwargs({"one": "min", "two": "max"})
+    (['one', 'two'], ['min', 'max'])
+    """
+    tuple_given_message = "func is expected but received {} in **kwargs."
+    columns = list(kwargs)
+    func = []
+    for col_func in kwargs.values():
+        if not (isinstance(col_func, str) or callable(col_func)):
+            raise TypeError(tuple_given_message.format(type(col_func).__name__))
+        func.append(col_func)
+    if not columns:
+        no_arg_message = "Must provide 'func' or named aggregation **kwargs."
+        raise TypeError(no_arg_message)
+    return columns, func
+
+
+def include_axis(op_name: Literal["agg", "apply"], colg: Series | DataFrame) -> bool:
+    return isinstance(colg, ABCDataFrame) or (
+        isinstance(colg, ABCSeries) and op_name == "agg"
+    )
diff --git a/pandas/pandas/core/array_algos/__init__.py b/pandas/pandas/core/array_algos/__init__.py
new file mode 100644
index 0000000000000..a7655a013c6cf
--- /dev/null
+++ b/pandas/pandas/core/array_algos/__init__.py
@@ -0,0 +1,9 @@
+"""
+core.array_algos is for algorithms that operate on ndarray and ExtensionArray.
+These should:
+
+- Assume that any Index, Series, or DataFrame objects have already been unwrapped.
+- Assume that any list arguments have already been cast to ndarray/EA.
+- Not depend on Index, Series, or DataFrame, nor import any of these.
+- May dispatch to ExtensionArray methods, but should not import from core.arrays.
+"""
diff --git a/pandas/pandas/core/array_algos/datetimelike_accumulations.py b/pandas/pandas/core/array_algos/datetimelike_accumulations.py
new file mode 100644
index 0000000000000..bc10dbfbec90d
--- /dev/null
+++ b/pandas/pandas/core/array_algos/datetimelike_accumulations.py
@@ -0,0 +1,73 @@
+"""
+datetimelke_accumulations.py is for accumulations of datetimelike extension arrays
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import iNaT
+
+from pandas.core.dtypes.missing import isna
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+
+def _cum_func(
+    func: Callable,
+    values: np.ndarray,
+    *,
+    skipna: bool = True,
+) -> np.ndarray:
+    """
+    Accumulations for 1D datetimelike arrays.
+
+    Parameters
+    ----------
+    func : np.cumsum, np.maximum.accumulate, np.minimum.accumulate
+    values : np.ndarray
+        Numpy array with the values (can be of any dtype that support the
+        operation). Values is changed is modified inplace.
+    skipna : bool, default True
+        Whether to skip NA.
+    """
+    try:
+        fill_value = {
+            np.maximum.accumulate: np.iinfo(np.int64).min,
+            np.cumsum: 0,
+            np.minimum.accumulate: np.iinfo(np.int64).max,
+        }[func]
+    except KeyError as err:
+        raise ValueError(
+            f"No accumulation for {func} implemented on BaseMaskedArray"
+        ) from err
+
+    mask = isna(values)
+    y = values.view("i8")
+    y[mask] = fill_value
+
+    if not skipna:
+        mask = np.maximum.accumulate(mask)
+
+    # GH 57956
+    result = func(y, axis=0)
+    result[mask] = iNaT
+
+    if values.dtype.kind in "mM":
+        return result.view(values.dtype.base)
+    return result
+
+
+def cumsum(values: np.ndarray, *, skipna: bool = True) -> np.ndarray:
+    return _cum_func(np.cumsum, values, skipna=skipna)
+
+
+def cummin(values: np.ndarray, *, skipna: bool = True) -> np.ndarray:
+    return _cum_func(np.minimum.accumulate, values, skipna=skipna)
+
+
+def cummax(values: np.ndarray, *, skipna: bool = True) -> np.ndarray:
+    return _cum_func(np.maximum.accumulate, values, skipna=skipna)
diff --git a/pandas/pandas/core/array_algos/masked_accumulations.py b/pandas/pandas/core/array_algos/masked_accumulations.py
new file mode 100644
index 0000000000000..b4e116388b85e
--- /dev/null
+++ b/pandas/pandas/core/array_algos/masked_accumulations.py
@@ -0,0 +1,97 @@
+"""
+masked_accumulations.py is for accumulation algorithms using a mask-based approach
+for missing values.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import npt
+
+
+def _cum_func(
+    func: Callable,
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    """
+    Accumulations for 1D masked array.
+
+    We will modify values in place to replace NAs with the appropriate fill value.
+
+    Parameters
+    ----------
+    func : np.cumsum, np.cumprod, np.maximum.accumulate, np.minimum.accumulate
+    values : np.ndarray
+        Numpy array with the values (can be of any dtype that support the
+        operation).
+    mask : np.ndarray
+        Boolean numpy array (True values indicate missing values).
+    skipna : bool, default True
+        Whether to skip NA.
+    """
+    dtype_info: np.iinfo | np.finfo
+    if values.dtype.kind == "f":
+        dtype_info = np.finfo(values.dtype.type)
+    elif values.dtype.kind in "iu":
+        dtype_info = np.iinfo(values.dtype.type)
+    elif values.dtype.kind == "b":
+        # Max value of bool is 1, but since we are setting into a boolean
+        # array, 255 is fine as well. Min value has to be 0 when setting
+        # into the boolean array.
+        dtype_info = np.iinfo(np.uint8)
+    else:
+        raise NotImplementedError(
+            f"No masked accumulation defined for dtype {values.dtype.type}"
+        )
+    try:
+        fill_value = {
+            np.cumprod: 1,
+            np.maximum.accumulate: dtype_info.min,
+            np.cumsum: 0,
+            np.minimum.accumulate: dtype_info.max,
+        }[func]
+    except KeyError as err:
+        raise NotImplementedError(
+            f"No accumulation for {func} implemented on BaseMaskedArray"
+        ) from err
+
+    values[mask] = fill_value
+
+    if not skipna:
+        mask = np.maximum.accumulate(mask)
+
+    values = func(values)
+    return values, mask
+
+
+def cumsum(
+    values: np.ndarray, mask: npt.NDArray[np.bool_], *, skipna: bool = True
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    return _cum_func(np.cumsum, values, mask, skipna=skipna)
+
+
+def cumprod(
+    values: np.ndarray, mask: npt.NDArray[np.bool_], *, skipna: bool = True
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    return _cum_func(np.cumprod, values, mask, skipna=skipna)
+
+
+def cummin(
+    values: np.ndarray, mask: npt.NDArray[np.bool_], *, skipna: bool = True
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    return _cum_func(np.minimum.accumulate, values, mask, skipna=skipna)
+
+
+def cummax(
+    values: np.ndarray, mask: npt.NDArray[np.bool_], *, skipna: bool = True
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    return _cum_func(np.maximum.accumulate, values, mask, skipna=skipna)
diff --git a/pandas/pandas/core/array_algos/masked_reductions.py b/pandas/pandas/core/array_algos/masked_reductions.py
new file mode 100644
index 0000000000000..bdf88f2e9fa07
--- /dev/null
+++ b/pandas/pandas/core/array_algos/masked_reductions.py
@@ -0,0 +1,201 @@
+"""
+masked_reductions.py is for reduction algorithms using a mask-based approach
+for missing values.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._libs import missing as libmissing
+
+from pandas.core.nanops import check_below_min_count
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        AxisInt,
+        npt,
+    )
+
+
+def _reductions(
+    func: Callable,
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    min_count: int = 0,
+    axis: AxisInt | None = None,
+    **kwargs,
+):
+    """
+    Sum, mean or product for 1D masked array.
+
+    Parameters
+    ----------
+    func : np.sum or np.prod
+    values : np.ndarray
+        Numpy array with the values (can be of any dtype that support the
+        operation).
+    mask : np.ndarray[bool]
+        Boolean numpy array (True values indicate missing values).
+    skipna : bool, default True
+        Whether to skip NA.
+    min_count : int, default 0
+        The required number of valid values to perform the operation. If fewer than
+        ``min_count`` non-NA values are present the result will be NA.
+    axis : int, optional, default None
+    """
+    if not skipna:
+        if mask.any() or check_below_min_count(values.shape, None, min_count):
+            return libmissing.NA
+        else:
+            return func(values, axis=axis, **kwargs)
+    else:
+        if check_below_min_count(values.shape, mask, min_count) and (
+            axis is None or values.ndim == 1
+        ):
+            return libmissing.NA
+
+        if values.dtype == np.dtype(object):
+            # object dtype does not support `where` without passing an initial
+            values = values[~mask]
+            return func(values, axis=axis, **kwargs)
+        return func(values, where=~mask, axis=axis, **kwargs)
+
+
+def sum(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    min_count: int = 0,
+    axis: AxisInt | None = None,
+):
+    return _reductions(
+        np.sum, values=values, mask=mask, skipna=skipna, min_count=min_count, axis=axis
+    )
+
+
+def prod(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    min_count: int = 0,
+    axis: AxisInt | None = None,
+):
+    return _reductions(
+        np.prod, values=values, mask=mask, skipna=skipna, min_count=min_count, axis=axis
+    )
+
+
+def _minmax(
+    func: Callable,
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+):
+    """
+    Reduction for 1D masked array.
+
+    Parameters
+    ----------
+    func : np.min or np.max
+    values : np.ndarray
+        Numpy array with the values (can be of any dtype that support the
+        operation).
+    mask : np.ndarray[bool]
+        Boolean numpy array (True values indicate missing values).
+    skipna : bool, default True
+        Whether to skip NA.
+    axis : int, optional, default None
+    """
+    if not skipna:
+        if mask.any() or not values.size:
+            # min/max with empty array raise in numpy, pandas returns NA
+            return libmissing.NA
+        else:
+            return func(values, axis=axis)
+    else:
+        subset = values[~mask]
+        if subset.size:
+            return func(subset, axis=axis)
+        else:
+            # min/max with empty array raise in numpy, pandas returns NA
+            return libmissing.NA
+
+
+def min(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+):
+    return _minmax(np.min, values=values, mask=mask, skipna=skipna, axis=axis)
+
+
+def max(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+):
+    return _minmax(np.max, values=values, mask=mask, skipna=skipna, axis=axis)
+
+
+def mean(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+):
+    if not values.size or mask.all():
+        return libmissing.NA
+    return _reductions(np.mean, values=values, mask=mask, skipna=skipna, axis=axis)
+
+
+def var(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+    ddof: int = 1,
+):
+    if not values.size or mask.all():
+        return libmissing.NA
+
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore", RuntimeWarning)
+        return _reductions(
+            np.var, values=values, mask=mask, skipna=skipna, axis=axis, ddof=ddof
+        )
+
+
+def std(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    *,
+    skipna: bool = True,
+    axis: AxisInt | None = None,
+    ddof: int = 1,
+):
+    if not values.size or mask.all():
+        return libmissing.NA
+
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore", RuntimeWarning)
+        return _reductions(
+            np.std, values=values, mask=mask, skipna=skipna, axis=axis, ddof=ddof
+        )
diff --git a/pandas/pandas/core/array_algos/putmask.py b/pandas/pandas/core/array_algos/putmask.py
new file mode 100644
index 0000000000000..464a4d552af68
--- /dev/null
+++ b/pandas/pandas/core/array_algos/putmask.py
@@ -0,0 +1,150 @@
+"""
+EA-compatible analogue to np.putmask
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.cast import infer_dtype_from
+from pandas.core.dtypes.common import is_list_like
+
+from pandas.core.arrays import ExtensionArray
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        npt,
+    )
+
+    from pandas import MultiIndex
+
+
+def putmask_inplace(values: ArrayLike, mask: npt.NDArray[np.bool_], value: Any) -> None:
+    """
+    ExtensionArray-compatible implementation of np.putmask.  The main
+    difference is we do not handle repeating or truncating like numpy.
+
+    Parameters
+    ----------
+    values: np.ndarray or ExtensionArray
+    mask : np.ndarray[bool]
+        We assume extract_bool_array has already been called.
+    value : Any
+    """
+
+    if (
+        not isinstance(values, np.ndarray)
+        or (values.dtype == object and not lib.is_scalar(value))
+        # GH#43424: np.putmask raises TypeError if we cannot cast between types with
+        # rule = "safe", a stricter guarantee we may not have here
+        or (
+            isinstance(value, np.ndarray) and not np.can_cast(value.dtype, values.dtype)
+        )
+    ):
+        # GH#19266 using np.putmask gives unexpected results with listlike value
+        #  along with object dtype
+        if is_list_like(value) and len(value) == len(values):
+            values[mask] = value[mask]
+        else:
+            values[mask] = value
+    else:
+        # GH#37833 np.putmask is more performant than __setitem__
+        np.putmask(values, mask, value)
+
+
+def putmask_without_repeat(
+    values: np.ndarray, mask: npt.NDArray[np.bool_], new: Any
+) -> None:
+    """
+    np.putmask will truncate or repeat if `new` is a listlike with
+    len(new) != len(values).  We require an exact match.
+
+    Parameters
+    ----------
+    values : np.ndarray
+    mask : np.ndarray[bool]
+    new : Any
+    """
+    if getattr(new, "ndim", 0) >= 1:
+        new = new.astype(values.dtype, copy=False)
+
+    # TODO: this prob needs some better checking for 2D cases
+    nlocs = mask.sum()
+    if nlocs > 0 and is_list_like(new) and getattr(new, "ndim", 1) == 1:
+        shape = np.shape(new)
+        # np.shape compat for if setitem_datetimelike_compat
+        #  changed arraylike to list e.g. test_where_dt64_2d
+        if nlocs == shape[-1]:
+            # GH#30567
+            # If length of ``new`` is less than the length of ``values``,
+            # `np.putmask` would first repeat the ``new`` array and then
+            # assign the masked values hence produces incorrect result.
+            # `np.place` on the other hand uses the ``new`` values at it is
+            # to place in the masked locations of ``values``
+            np.place(values, mask, new)
+            # i.e. values[mask] = new
+        elif mask.shape[-1] == shape[-1] or shape[-1] == 1:
+            np.putmask(values, mask, new)
+        else:
+            raise ValueError("cannot assign mismatch length to masked array")
+    else:
+        np.putmask(values, mask, new)
+
+
+def validate_putmask(
+    values: ArrayLike | MultiIndex, mask: np.ndarray
+) -> tuple[npt.NDArray[np.bool_], bool]:
+    """
+    Validate mask and check if this putmask operation is a no-op.
+    """
+    mask = extract_bool_array(mask)
+    if mask.shape != values.shape:
+        raise ValueError("putmask: mask and data must be the same size")
+
+    noop = not mask.any()
+    return mask, noop
+
+
+def extract_bool_array(mask: ArrayLike) -> npt.NDArray[np.bool_]:
+    """
+    If we have a SparseArray or BooleanArray, convert it to ndarray[bool].
+    """
+    if isinstance(mask, ExtensionArray):
+        # We could have BooleanArray, Sparse[bool], ...
+        #  Except for BooleanArray, this is equivalent to just
+        #  np.asarray(mask, dtype=bool)
+        mask = mask.to_numpy(dtype=bool, na_value=False)
+
+    mask = np.asarray(mask, dtype=bool)
+    return mask
+
+
+def setitem_datetimelike_compat(values: np.ndarray, num_set: int, other):
+    """
+    Parameters
+    ----------
+    values : np.ndarray
+    num_set : int
+        For putmask, this is mask.sum()
+    other : Any
+    """
+    if values.dtype == object:
+        dtype, _ = infer_dtype_from(other)
+
+        if lib.is_np_dtype(dtype, "mM"):
+            # https://github.com/numpy/numpy/issues/12550
+            #  timedelta64 will incorrectly cast to int
+            if not is_list_like(other):
+                other = [other] * num_set
+            else:
+                other = list(other)
+
+    return other
diff --git a/pandas/pandas/core/array_algos/quantile.py b/pandas/pandas/core/array_algos/quantile.py
new file mode 100644
index 0000000000000..8a920d1849bb3
--- /dev/null
+++ b/pandas/pandas/core/array_algos/quantile.py
@@ -0,0 +1,226 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Scalar,
+        npt,
+    )
+
+
+def quantile_compat(
+    values: ArrayLike, qs: npt.NDArray[np.float64], interpolation: str
+) -> ArrayLike:
+    """
+    Compute the quantiles of the given values for each quantile in `qs`.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+    qs : np.ndarray[float64]
+    interpolation : str
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+    """
+    if isinstance(values, np.ndarray):
+        fill_value = na_value_for_dtype(values.dtype, compat=False)
+        mask = isna(values)
+        return quantile_with_mask(values, mask, fill_value, qs, interpolation)
+    else:
+        return values._quantile(qs, interpolation)
+
+
+def quantile_with_mask(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    fill_value,
+    qs: npt.NDArray[np.float64],
+    interpolation: str,
+) -> np.ndarray:
+    """
+    Compute the quantiles of the given values for each quantile in `qs`.
+
+    Parameters
+    ----------
+    values : np.ndarray
+        For ExtensionArray, this is _values_for_factorize()[0]
+    mask : np.ndarray[bool]
+        mask = isna(values)
+        For ExtensionArray, this is computed before calling _value_for_factorize
+    fill_value : Scalar
+        The value to interpret fill NA entries with
+        For ExtensionArray, this is _values_for_factorize()[1]
+    qs : np.ndarray[float64]
+    interpolation : str
+        Type of interpolation
+
+    Returns
+    -------
+    np.ndarray
+
+    Notes
+    -----
+    Assumes values is already 2D.  For ExtensionArray this means np.atleast_2d
+    has been called on _values_for_factorize()[0]
+
+    Quantile is computed along axis=1.
+    """
+    assert values.shape == mask.shape
+    if values.ndim == 1:
+        # unsqueeze, operate, re-squeeze
+        values = np.atleast_2d(values)
+        mask = np.atleast_2d(mask)
+        res_values = quantile_with_mask(values, mask, fill_value, qs, interpolation)
+        return res_values[0]
+
+    assert values.ndim == 2
+
+    is_empty = values.shape[1] == 0
+
+    if is_empty:
+        # create the array of na_values
+        # 2d len(values) * len(qs)
+        flat = np.full(len(qs), fill_value)
+        result = np.repeat(flat, len(values)).reshape(len(values), len(qs))
+    else:
+        result = _nanquantile(
+            values,
+            qs,
+            na_value=fill_value,
+            mask=mask,
+            interpolation=interpolation,
+        )
+
+        result = np.asarray(result)
+        result = result.T
+
+    return result
+
+
+def _nanquantile_1d(
+    values: np.ndarray,
+    mask: npt.NDArray[np.bool_],
+    qs: npt.NDArray[np.float64],
+    na_value: Scalar,
+    interpolation: str,
+) -> Scalar | np.ndarray:
+    """
+    Wrapper for np.quantile that skips missing values, specialized to
+    1-dimensional case.
+
+    Parameters
+    ----------
+    values : array over which to find quantiles
+    mask : ndarray[bool]
+        locations in values that should be considered missing
+    qs : np.ndarray[float64] of quantile indices to find
+    na_value : scalar
+        value to return for empty or all-null values
+    interpolation : str
+
+    Returns
+    -------
+    quantiles : scalar or array
+    """
+    # mask is Union[ExtensionArray, ndarray]
+    values = values[~mask]
+
+    if len(values) == 0:
+        # Can't pass dtype=values.dtype here bc we might have na_value=np.nan
+        #  with values.dtype=int64 see test_quantile_empty
+        # equiv: 'np.array([na_value] * len(qs))' but much faster
+        return np.full(len(qs), na_value)
+
+    return np.quantile(
+        values,
+        qs,
+        # error: No overload variant of "percentile" matches argument
+        # types "ndarray[Any, Any]", "ndarray[Any, dtype[floating[_64Bit]]]"
+        # , "Dict[str, str]"  [call-overload]
+        method=interpolation,  # type: ignore[call-overload]
+    )
+
+
+def _nanquantile(
+    values: np.ndarray,
+    qs: npt.NDArray[np.float64],
+    *,
+    na_value,
+    mask: npt.NDArray[np.bool_],
+    interpolation: str,
+):
+    """
+    Wrapper for np.quantile that skips missing values.
+
+    Parameters
+    ----------
+    values : np.ndarray[ndim=2]  over which to find quantiles
+    qs : np.ndarray[float64] of quantile indices to find
+    na_value : scalar
+        value to return for empty or all-null values
+    mask : np.ndarray[bool]
+        locations in values that should be considered missing
+    interpolation : str
+
+    Returns
+    -------
+    quantiles : scalar or array
+    """
+
+    if values.dtype.kind in "mM":
+        # need to cast to integer to avoid rounding errors in numpy
+        result = _nanquantile(
+            values.view("i8"),
+            qs=qs,
+            na_value=na_value.view("i8"),
+            mask=mask,
+            interpolation=interpolation,
+        )
+
+        # Note: we have to do `astype` and not view because in general we
+        #  have float result at this point, not i8
+        return result.astype(values.dtype)
+
+    if mask.any():
+        # Caller is responsible for ensuring mask shape match
+        assert mask.shape == values.shape
+        result = [
+            _nanquantile_1d(val, m, qs, na_value, interpolation=interpolation)
+            for (val, m) in zip(list(values), list(mask))
+        ]
+        if values.dtype.kind == "f":
+            # preserve itemsize
+            result = np.asarray(result, dtype=values.dtype).T
+        else:
+            result = np.asarray(result).T
+            if (
+                result.dtype != values.dtype
+                and not mask.all()
+                and (result == result.astype(values.dtype, copy=False)).all()
+            ):
+                # mask.all() will never get cast back to int
+                # e.g. values id integer dtype and result is floating dtype,
+                #  only cast back to integer dtype if result values are all-integer.
+                result = result.astype(values.dtype, copy=False)
+        return result
+    else:
+        return np.quantile(
+            values,
+            qs,
+            axis=1,
+            # error: No overload variant of "percentile" matches argument types
+            # "ndarray[Any, Any]", "ndarray[Any, dtype[floating[_64Bit]]]",
+            # "int", "Dict[str, str]"  [call-overload]
+            method=interpolation,  # type: ignore[call-overload]
+        )
diff --git a/pandas/pandas/core/array_algos/replace.py b/pandas/pandas/core/array_algos/replace.py
new file mode 100644
index 0000000000000..debd6368e98a4
--- /dev/null
+++ b/pandas/pandas/core/array_algos/replace.py
@@ -0,0 +1,157 @@
+"""
+Methods used by Block.replace and related methods.
+"""
+
+from __future__ import annotations
+
+import operator
+import re
+from re import Pattern
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_re,
+    is_re_compilable,
+)
+from pandas.core.dtypes.missing import isna
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Scalar,
+        npt,
+    )
+
+
+def should_use_regex(regex: bool, to_replace: Any) -> bool:
+    """
+    Decide whether to treat `to_replace` as a regular expression.
+    """
+    if is_re(to_replace):
+        regex = True
+
+    regex = regex and is_re_compilable(to_replace)
+
+    # Don't use regex if the pattern is empty.
+    regex = regex and re.compile(to_replace).pattern != ""
+    return regex
+
+
+def compare_or_regex_search(
+    a: ArrayLike, b: Scalar | Pattern, regex: bool, mask: npt.NDArray[np.bool_]
+) -> ArrayLike:
+    """
+    Compare two array-like inputs of the same shape or two scalar values
+
+    Calls operator.eq or re.search, depending on regex argument. If regex is
+    True, perform an element-wise regex matching.
+
+    Parameters
+    ----------
+    a : array-like
+    b : scalar or regex pattern
+    regex : bool
+    mask : np.ndarray[bool]
+
+    Returns
+    -------
+    mask : array-like of bool
+    """
+    if isna(b):
+        return ~mask
+
+    def _check_comparison_types(
+        result: ArrayLike | bool, a: ArrayLike, b: Scalar | Pattern
+    ) -> None:
+        """
+        Raises an error if the two arrays (a,b) cannot be compared.
+        Otherwise, returns the comparison result as expected.
+        """
+        if is_bool(result) and isinstance(a, np.ndarray):
+            type_names = [type(a).__name__, type(b).__name__]
+
+            type_names[0] = f"ndarray(dtype={a.dtype})"
+
+            raise TypeError(
+                f"Cannot compare types {type_names[0]!r} and {type_names[1]!r}"
+            )
+
+    if not regex or not should_use_regex(regex, b):
+        # TODO: should use missing.mask_missing?
+        op = lambda x: operator.eq(x, b)
+    else:
+        op = np.vectorize(
+            lambda x: bool(re.search(b, x))
+            if isinstance(x, str) and isinstance(b, (str, Pattern))
+            else False,
+            otypes=[bool],
+        )
+
+    # GH#32621 use mask to avoid comparing to NAs
+    if isinstance(a, np.ndarray) and mask is not None:
+        a = a[mask]
+        result = op(a)
+
+        if isinstance(result, np.ndarray):
+            # The shape of the mask can differ to that of the result
+            # since we may compare only a subset of a's or b's elements
+            tmp = np.zeros(mask.shape, dtype=np.bool_)
+            np.place(tmp, mask, result)
+            result = tmp
+    else:
+        result = op(a)
+
+    _check_comparison_types(result, a, b)
+    return result
+
+
+def replace_regex(
+    values: ArrayLike, rx: re.Pattern, value, mask: npt.NDArray[np.bool_] | None
+) -> None:
+    """
+    Parameters
+    ----------
+    values : ArrayLike
+        Object dtype.
+    rx : re.Pattern
+    value : Any
+    mask : np.ndarray[bool], optional
+
+    Notes
+    -----
+    Alters values in-place.
+    """
+
+    # deal with replacing values with objects (strings) that match but
+    # whose replacement is not a string (numeric, nan, object)
+    if isna(value) or not isinstance(value, str):
+
+        def re_replacer(s):
+            if is_re(rx) and isinstance(s, str):
+                return value if rx.search(s) is not None else s
+            else:
+                return s
+
+    else:
+        # value is guaranteed to be a string here, s can be either a string
+        # or null if it's null it gets returned
+        def re_replacer(s):
+            if is_re(rx) and isinstance(s, str):
+                return rx.sub(value, s)
+            else:
+                return s
+
+    f = np.vectorize(re_replacer, otypes=[np.object_])
+
+    if mask is None:
+        values[:] = f(values)
+    else:
+        if values.ndim != mask.ndim:
+            mask = np.broadcast_to(mask, values.shape)
+        values[mask] = f(values[mask])
diff --git a/pandas/pandas/core/array_algos/take.py b/pandas/pandas/core/array_algos/take.py
new file mode 100644
index 0000000000000..4d33b01f616cc
--- /dev/null
+++ b/pandas/pandas/core/array_algos/take.py
@@ -0,0 +1,546 @@
+from __future__ import annotations
+
+import functools
+from typing import (
+    TYPE_CHECKING,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    algos as libalgos,
+    lib,
+)
+
+from pandas.core.dtypes.cast import maybe_promote
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    is_1d_only_ea_dtype,
+)
+from pandas.core.dtypes.missing import na_value_for_dtype
+
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        npt,
+    )
+
+    from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+    from pandas.core.arrays.base import ExtensionArray
+
+
+@overload
+def take_nd(
+    arr: np.ndarray,
+    indexer,
+    axis: AxisInt = ...,
+    fill_value=...,
+    allow_fill: bool = ...,
+) -> np.ndarray: ...
+
+
+@overload
+def take_nd(
+    arr: ExtensionArray,
+    indexer,
+    axis: AxisInt = ...,
+    fill_value=...,
+    allow_fill: bool = ...,
+) -> ArrayLike: ...
+
+
+def take_nd(
+    arr: ArrayLike,
+    indexer,
+    axis: AxisInt = 0,
+    fill_value=lib.no_default,
+    allow_fill: bool = True,
+) -> ArrayLike:
+    """
+    Specialized Cython take which sets NaN values in one pass
+
+    This dispatches to ``take`` defined on ExtensionArrays.
+
+    Note: this function assumes that the indexer is a valid(ated) indexer with
+    no out of bound indices.
+
+    Parameters
+    ----------
+    arr : np.ndarray or ExtensionArray
+        Input array.
+    indexer : ndarray
+        1-D array of indices to take, subarrays corresponding to -1 value
+        indices are filed with fill_value
+    axis : int, default 0
+        Axis to take from
+    fill_value : any, default np.nan
+        Fill value to replace -1 values with
+    allow_fill : bool, default True
+        If False, indexer is assumed to contain no -1 values so no filling
+        will be done.  This short-circuits computation of a mask.  Result is
+        undefined if allow_fill == False and -1 is present in indexer.
+
+    Returns
+    -------
+    subarray : np.ndarray or ExtensionArray
+        May be the same type as the input, or cast to an ndarray.
+    """
+    if fill_value is lib.no_default:
+        fill_value = na_value_for_dtype(arr.dtype, compat=False)
+    elif lib.is_np_dtype(arr.dtype, "mM"):
+        dtype, fill_value = maybe_promote(arr.dtype, fill_value)
+        if arr.dtype != dtype:
+            # EA.take is strict about returning a new object of the same type
+            # so for that case cast upfront
+            arr = arr.astype(dtype)
+
+    if not isinstance(arr, np.ndarray):
+        # i.e. ExtensionArray,
+        # includes for EA to catch DatetimeArray, TimedeltaArray
+        if not is_1d_only_ea_dtype(arr.dtype):
+            # i.e. DatetimeArray, TimedeltaArray
+            arr = cast("NDArrayBackedExtensionArray", arr)
+            return arr.take(
+                indexer, fill_value=fill_value, allow_fill=allow_fill, axis=axis
+            )
+
+        return arr.take(indexer, fill_value=fill_value, allow_fill=allow_fill)
+
+    arr = np.asarray(arr)
+    return _take_nd_ndarray(arr, indexer, axis, fill_value, allow_fill)
+
+
+def _take_nd_ndarray(
+    arr: np.ndarray,
+    indexer: npt.NDArray[np.intp] | None,
+    axis: AxisInt,
+    fill_value,
+    allow_fill: bool,
+) -> np.ndarray:
+    if indexer is None:
+        indexer = np.arange(arr.shape[axis], dtype=np.intp)
+        dtype, fill_value = arr.dtype, arr.dtype.type()
+    else:
+        indexer = ensure_platform_int(indexer)
+
+    dtype, fill_value, mask_info = _take_preprocess_indexer_and_fill_value(
+        arr, indexer, fill_value, allow_fill
+    )
+
+    flip_order = False
+    if arr.ndim == 2 and arr.flags.f_contiguous:
+        flip_order = True
+
+    if flip_order:
+        arr = arr.T
+        axis = arr.ndim - axis - 1
+
+    # at this point, it's guaranteed that dtype can hold both the arr values
+    # and the fill_value
+    out_shape_ = list(arr.shape)
+    out_shape_[axis] = len(indexer)
+    out_shape = tuple(out_shape_)
+    if arr.flags.f_contiguous and axis == arr.ndim - 1:
+        # minor tweak that can make an order-of-magnitude difference
+        # for dataframes initialized directly from 2-d ndarrays
+        # (s.t. df.values is c-contiguous and df._mgr.blocks[0] is its
+        # f-contiguous transpose)
+        out = np.empty(out_shape, dtype=dtype, order="F")
+    else:
+        out = np.empty(out_shape, dtype=dtype)
+
+    func = _get_take_nd_function(
+        arr.ndim, arr.dtype, out.dtype, axis=axis, mask_info=mask_info
+    )
+    func(arr, indexer, out, fill_value)
+
+    if flip_order:
+        out = out.T
+    return out
+
+
+def take_2d_multi(
+    arr: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    fill_value=np.nan,
+) -> np.ndarray:
+    """
+    Specialized Cython take which sets NaN values in one pass.
+    """
+    # This is only called from one place in DataFrame._reindex_multi,
+    #  so we know indexer is well-behaved.
+    assert indexer is not None
+    assert indexer[0] is not None
+    assert indexer[1] is not None
+
+    row_idx, col_idx = indexer
+
+    row_idx = ensure_platform_int(row_idx)
+    col_idx = ensure_platform_int(col_idx)
+    indexer = row_idx, col_idx
+    mask_info = None
+
+    # check for promotion based on types only (do this first because
+    # it's faster than computing a mask)
+    dtype, fill_value = maybe_promote(arr.dtype, fill_value)
+    if dtype != arr.dtype:
+        # check if promotion is actually required based on indexer
+        row_mask = row_idx == -1
+        col_mask = col_idx == -1
+        row_needs = row_mask.any()
+        col_needs = col_mask.any()
+        mask_info = (row_mask, col_mask), (row_needs, col_needs)
+
+        if not (row_needs or col_needs):
+            # if not, then depromote, set fill_value to dummy
+            # (it won't be used but we don't want the cython code
+            # to crash when trying to cast it to dtype)
+            dtype, fill_value = arr.dtype, arr.dtype.type()
+
+    # at this point, it's guaranteed that dtype can hold both the arr values
+    # and the fill_value
+    out_shape = len(row_idx), len(col_idx)
+    out = np.empty(out_shape, dtype=dtype)
+
+    func = _take_2d_multi_dict.get((arr.dtype.name, out.dtype.name), None)
+    if func is None and arr.dtype != out.dtype:
+        func = _take_2d_multi_dict.get((out.dtype.name, out.dtype.name), None)
+        if func is not None:
+            func = _convert_wrapper(func, out.dtype)
+
+    if func is not None:
+        func(arr, indexer, out=out, fill_value=fill_value)
+    else:
+        # test_reindex_multi
+        _take_2d_multi_object(
+            arr, indexer, out, fill_value=fill_value, mask_info=mask_info
+        )
+
+    return out
+
+
+@functools.lru_cache
+def _get_take_nd_function_cached(
+    ndim: int, arr_dtype: np.dtype, out_dtype: np.dtype, axis: AxisInt
+):
+    """
+    Part of _get_take_nd_function below that doesn't need `mask_info` and thus
+    can be cached (mask_info potentially contains a numpy ndarray which is not
+    hashable and thus cannot be used as argument for cached function).
+    """
+    tup = (arr_dtype.name, out_dtype.name)
+    if ndim == 1:
+        func = _take_1d_dict.get(tup, None)
+    elif ndim == 2:
+        if axis == 0:
+            func = _take_2d_axis0_dict.get(tup, None)
+        else:
+            func = _take_2d_axis1_dict.get(tup, None)
+    if func is not None:
+        return func
+
+    # We get here with string, uint, float16, and complex dtypes that could
+    #  potentially be handled in algos_take_helper.
+    #  Also a couple with (M8[ns], object) and (m8[ns], object)
+    tup = (out_dtype.name, out_dtype.name)
+    if ndim == 1:
+        func = _take_1d_dict.get(tup, None)
+    elif ndim == 2:
+        if axis == 0:
+            func = _take_2d_axis0_dict.get(tup, None)
+        else:
+            func = _take_2d_axis1_dict.get(tup, None)
+    if func is not None:
+        func = _convert_wrapper(func, out_dtype)
+        return func
+
+    return None
+
+
+def _get_take_nd_function(
+    ndim: int,
+    arr_dtype: np.dtype,
+    out_dtype: np.dtype,
+    axis: AxisInt = 0,
+    mask_info=None,
+):
+    """
+    Get the appropriate "take" implementation for the given dimension, axis
+    and dtypes.
+    """
+    func = None
+    if ndim <= 2:
+        # for this part we don't need `mask_info` -> use the cached algo lookup
+        func = _get_take_nd_function_cached(ndim, arr_dtype, out_dtype, axis)
+
+    if func is None:
+
+        def func(arr, indexer, out, fill_value=np.nan) -> None:
+            indexer = ensure_platform_int(indexer)
+            _take_nd_object(
+                arr, indexer, out, axis=axis, fill_value=fill_value, mask_info=mask_info
+            )
+
+    return func
+
+
+def _view_wrapper(f, arr_dtype=None, out_dtype=None, fill_wrap=None):
+    def wrapper(
+        arr: np.ndarray, indexer: np.ndarray, out: np.ndarray, fill_value=np.nan
+    ) -> None:
+        if arr_dtype is not None:
+            arr = arr.view(arr_dtype)
+        if out_dtype is not None:
+            out = out.view(out_dtype)
+        if fill_wrap is not None:
+            # FIXME: if we get here with dt64/td64 we need to be sure we have
+            #  matching resos
+            if fill_value.dtype.kind == "m":
+                fill_value = fill_value.astype("m8[ns]")
+            else:
+                fill_value = fill_value.astype("M8[ns]")
+            fill_value = fill_wrap(fill_value)
+
+        f(arr, indexer, out, fill_value=fill_value)
+
+    return wrapper
+
+
+def _convert_wrapper(f, conv_dtype):
+    def wrapper(
+        arr: np.ndarray, indexer: np.ndarray, out: np.ndarray, fill_value=np.nan
+    ) -> None:
+        if conv_dtype == object:
+            # GH#39755 avoid casting dt64/td64 to integers
+            arr = ensure_wrapped_if_datetimelike(arr)
+        arr = arr.astype(conv_dtype)
+        f(arr, indexer, out, fill_value=fill_value)
+
+    return wrapper
+
+
+_take_1d_dict = {
+    ("int8", "int8"): libalgos.take_1d_int8_int8,
+    ("int8", "int32"): libalgos.take_1d_int8_int32,
+    ("int8", "int64"): libalgos.take_1d_int8_int64,
+    ("int8", "float64"): libalgos.take_1d_int8_float64,
+    ("int16", "int16"): libalgos.take_1d_int16_int16,
+    ("int16", "int32"): libalgos.take_1d_int16_int32,
+    ("int16", "int64"): libalgos.take_1d_int16_int64,
+    ("int16", "float64"): libalgos.take_1d_int16_float64,
+    ("int32", "int32"): libalgos.take_1d_int32_int32,
+    ("int32", "int64"): libalgos.take_1d_int32_int64,
+    ("int32", "float64"): libalgos.take_1d_int32_float64,
+    ("int64", "int64"): libalgos.take_1d_int64_int64,
+    ("uint8", "uint8"): libalgos.take_1d_bool_bool,
+    ("uint16", "int64"): libalgos.take_1d_uint16_uint16,
+    ("uint32", "int64"): libalgos.take_1d_uint32_uint32,
+    ("uint64", "int64"): libalgos.take_1d_uint64_uint64,
+    ("int64", "float64"): libalgos.take_1d_int64_float64,
+    ("float32", "float32"): libalgos.take_1d_float32_float32,
+    ("float32", "float64"): libalgos.take_1d_float32_float64,
+    ("float64", "float64"): libalgos.take_1d_float64_float64,
+    ("object", "object"): libalgos.take_1d_object_object,
+    ("bool", "bool"): _view_wrapper(libalgos.take_1d_bool_bool, np.uint8, np.uint8),
+    ("bool", "object"): _view_wrapper(libalgos.take_1d_bool_object, np.uint8, None),
+    ("datetime64[ns]", "datetime64[ns]"): _view_wrapper(
+        libalgos.take_1d_int64_int64, np.int64, np.int64, np.int64
+    ),
+    ("timedelta64[ns]", "timedelta64[ns]"): _view_wrapper(
+        libalgos.take_1d_int64_int64, np.int64, np.int64, np.int64
+    ),
+}
+
+_take_2d_axis0_dict = {
+    ("int8", "int8"): libalgos.take_2d_axis0_int8_int8,
+    ("int8", "int32"): libalgos.take_2d_axis0_int8_int32,
+    ("int8", "int64"): libalgos.take_2d_axis0_int8_int64,
+    ("int8", "float64"): libalgos.take_2d_axis0_int8_float64,
+    ("int16", "int16"): libalgos.take_2d_axis0_int16_int16,
+    ("int16", "int32"): libalgos.take_2d_axis0_int16_int32,
+    ("int16", "int64"): libalgos.take_2d_axis0_int16_int64,
+    ("int16", "float64"): libalgos.take_2d_axis0_int16_float64,
+    ("int32", "int32"): libalgos.take_2d_axis0_int32_int32,
+    ("int32", "int64"): libalgos.take_2d_axis0_int32_int64,
+    ("int32", "float64"): libalgos.take_2d_axis0_int32_float64,
+    ("int64", "int64"): libalgos.take_2d_axis0_int64_int64,
+    ("int64", "float64"): libalgos.take_2d_axis0_int64_float64,
+    ("uint8", "uint8"): libalgos.take_2d_axis0_bool_bool,
+    ("uint16", "uint16"): libalgos.take_2d_axis0_uint16_uint16,
+    ("uint32", "uint32"): libalgos.take_2d_axis0_uint32_uint32,
+    ("uint64", "uint64"): libalgos.take_2d_axis0_uint64_uint64,
+    ("float32", "float32"): libalgos.take_2d_axis0_float32_float32,
+    ("float32", "float64"): libalgos.take_2d_axis0_float32_float64,
+    ("float64", "float64"): libalgos.take_2d_axis0_float64_float64,
+    ("object", "object"): libalgos.take_2d_axis0_object_object,
+    ("bool", "bool"): _view_wrapper(
+        libalgos.take_2d_axis0_bool_bool, np.uint8, np.uint8
+    ),
+    ("bool", "object"): _view_wrapper(
+        libalgos.take_2d_axis0_bool_object, np.uint8, None
+    ),
+    ("datetime64[ns]", "datetime64[ns]"): _view_wrapper(
+        libalgos.take_2d_axis0_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+    ("timedelta64[ns]", "timedelta64[ns]"): _view_wrapper(
+        libalgos.take_2d_axis0_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+}
+
+_take_2d_axis1_dict = {
+    ("int8", "int8"): libalgos.take_2d_axis1_int8_int8,
+    ("int8", "int32"): libalgos.take_2d_axis1_int8_int32,
+    ("int8", "int64"): libalgos.take_2d_axis1_int8_int64,
+    ("int8", "float64"): libalgos.take_2d_axis1_int8_float64,
+    ("int16", "int16"): libalgos.take_2d_axis1_int16_int16,
+    ("int16", "int32"): libalgos.take_2d_axis1_int16_int32,
+    ("int16", "int64"): libalgos.take_2d_axis1_int16_int64,
+    ("int16", "float64"): libalgos.take_2d_axis1_int16_float64,
+    ("int32", "int32"): libalgos.take_2d_axis1_int32_int32,
+    ("int32", "int64"): libalgos.take_2d_axis1_int32_int64,
+    ("int32", "float64"): libalgos.take_2d_axis1_int32_float64,
+    ("int64", "int64"): libalgos.take_2d_axis1_int64_int64,
+    ("int64", "float64"): libalgos.take_2d_axis1_int64_float64,
+    ("uint8", "uint8"): libalgos.take_2d_axis1_bool_bool,
+    ("uint16", "uint16"): libalgos.take_2d_axis1_uint16_uint16,
+    ("uint32", "uint32"): libalgos.take_2d_axis1_uint32_uint32,
+    ("uint64", "uint64"): libalgos.take_2d_axis1_uint64_uint64,
+    ("float32", "float32"): libalgos.take_2d_axis1_float32_float32,
+    ("float32", "float64"): libalgos.take_2d_axis1_float32_float64,
+    ("float64", "float64"): libalgos.take_2d_axis1_float64_float64,
+    ("object", "object"): libalgos.take_2d_axis1_object_object,
+    ("bool", "bool"): _view_wrapper(
+        libalgos.take_2d_axis1_bool_bool, np.uint8, np.uint8
+    ),
+    ("bool", "object"): _view_wrapper(
+        libalgos.take_2d_axis1_bool_object, np.uint8, None
+    ),
+    ("datetime64[ns]", "datetime64[ns]"): _view_wrapper(
+        libalgos.take_2d_axis1_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+    ("timedelta64[ns]", "timedelta64[ns]"): _view_wrapper(
+        libalgos.take_2d_axis1_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+}
+
+_take_2d_multi_dict = {
+    ("int8", "int8"): libalgos.take_2d_multi_int8_int8,
+    ("int8", "int32"): libalgos.take_2d_multi_int8_int32,
+    ("int8", "int64"): libalgos.take_2d_multi_int8_int64,
+    ("int8", "float64"): libalgos.take_2d_multi_int8_float64,
+    ("int16", "int16"): libalgos.take_2d_multi_int16_int16,
+    ("int16", "int32"): libalgos.take_2d_multi_int16_int32,
+    ("int16", "int64"): libalgos.take_2d_multi_int16_int64,
+    ("int16", "float64"): libalgos.take_2d_multi_int16_float64,
+    ("int32", "int32"): libalgos.take_2d_multi_int32_int32,
+    ("int32", "int64"): libalgos.take_2d_multi_int32_int64,
+    ("int32", "float64"): libalgos.take_2d_multi_int32_float64,
+    ("int64", "int64"): libalgos.take_2d_multi_int64_int64,
+    ("int64", "float64"): libalgos.take_2d_multi_int64_float64,
+    ("float32", "float32"): libalgos.take_2d_multi_float32_float32,
+    ("float32", "float64"): libalgos.take_2d_multi_float32_float64,
+    ("float64", "float64"): libalgos.take_2d_multi_float64_float64,
+    ("object", "object"): libalgos.take_2d_multi_object_object,
+    ("bool", "bool"): _view_wrapper(
+        libalgos.take_2d_multi_bool_bool, np.uint8, np.uint8
+    ),
+    ("bool", "object"): _view_wrapper(
+        libalgos.take_2d_multi_bool_object, np.uint8, None
+    ),
+    ("datetime64[ns]", "datetime64[ns]"): _view_wrapper(
+        libalgos.take_2d_multi_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+    ("timedelta64[ns]", "timedelta64[ns]"): _view_wrapper(
+        libalgos.take_2d_multi_int64_int64, np.int64, np.int64, fill_wrap=np.int64
+    ),
+}
+
+
+def _take_nd_object(
+    arr: np.ndarray,
+    indexer: npt.NDArray[np.intp],
+    out: np.ndarray,
+    axis: AxisInt,
+    fill_value,
+    mask_info,
+) -> None:
+    if mask_info is not None:
+        mask, needs_masking = mask_info
+    else:
+        mask = indexer == -1
+        needs_masking = mask.any()
+    if arr.dtype != out.dtype:
+        arr = arr.astype(out.dtype)
+    if arr.shape[axis] > 0:
+        arr.take(indexer, axis=axis, out=out)
+    if needs_masking:
+        outindexer = [slice(None)] * arr.ndim
+        outindexer[axis] = mask
+        out[tuple(outindexer)] = fill_value
+
+
+def _take_2d_multi_object(
+    arr: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    out: np.ndarray,
+    fill_value,
+    mask_info,
+) -> None:
+    # this is not ideal, performance-wise, but it's better than raising
+    # an exception (best to optimize in Cython to avoid getting here)
+    row_idx, col_idx = indexer  # both np.intp
+    if mask_info is not None:
+        (row_mask, col_mask), (row_needs, col_needs) = mask_info
+    else:
+        row_mask = row_idx == -1
+        col_mask = col_idx == -1
+        row_needs = row_mask.any()
+        col_needs = col_mask.any()
+    if fill_value is not None:
+        if row_needs:
+            out[row_mask, :] = fill_value
+        if col_needs:
+            out[:, col_mask] = fill_value
+    for i, u_ in enumerate(row_idx):
+        if u_ != -1:
+            for j, v in enumerate(col_idx):
+                if v != -1:
+                    out[i, j] = arr[u_, v]
+
+
+def _take_preprocess_indexer_and_fill_value(
+    arr: np.ndarray,
+    indexer: npt.NDArray[np.intp],
+    fill_value,
+    allow_fill: bool,
+    mask: npt.NDArray[np.bool_] | None = None,
+):
+    mask_info: tuple[np.ndarray | None, bool] | None = None
+
+    if not allow_fill:
+        dtype, fill_value = arr.dtype, arr.dtype.type()
+        mask_info = None, False
+    else:
+        # check for promotion based on types only (do this first because
+        # it's faster than computing a mask)
+        dtype, fill_value = maybe_promote(arr.dtype, fill_value)
+        if dtype != arr.dtype:
+            # check if promotion is actually required based on indexer
+            if mask is not None:
+                needs_masking = True
+            else:
+                mask = indexer == -1
+                needs_masking = bool(mask.any())
+            mask_info = mask, needs_masking
+            if not needs_masking:
+                # if not, then depromote, set fill_value to dummy
+                # (it won't be used but we don't want the cython code
+                # to crash when trying to cast it to dtype)
+                dtype, fill_value = arr.dtype, arr.dtype.type()
+
+    return dtype, fill_value, mask_info
diff --git a/pandas/pandas/core/array_algos/transforms.py b/pandas/pandas/core/array_algos/transforms.py
new file mode 100644
index 0000000000000..ec67244949e3d
--- /dev/null
+++ b/pandas/pandas/core/array_algos/transforms.py
@@ -0,0 +1,50 @@
+"""
+transforms.py is for shape-preserving functions.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        AxisInt,
+        Scalar,
+    )
+
+
+def shift(
+    values: np.ndarray, periods: int, axis: AxisInt, fill_value: Scalar
+) -> np.ndarray:
+    new_values = values
+
+    if periods == 0 or values.size == 0:
+        return new_values.copy()
+
+    # make sure array sent to np.roll is c_contiguous
+    f_ordered = values.flags.f_contiguous
+    if f_ordered:
+        new_values = new_values.T
+        axis = new_values.ndim - axis - 1
+
+    if new_values.size:
+        new_values = np.roll(
+            new_values,
+            np.intp(periods),
+            axis=axis,
+        )
+
+    axis_indexer = [slice(None)] * values.ndim
+    if periods > 0:
+        axis_indexer[axis] = slice(None, periods)
+    else:
+        axis_indexer[axis] = slice(periods, None)
+    new_values[tuple(axis_indexer)] = fill_value
+
+    # restore original order
+    if f_ordered:
+        new_values = new_values.T
+
+    return new_values
diff --git a/pandas/pandas/core/arraylike.py b/pandas/pandas/core/arraylike.py
new file mode 100644
index 0000000000000..51ddd9e91b227
--- /dev/null
+++ b/pandas/pandas/core/arraylike.py
@@ -0,0 +1,530 @@
+"""
+Methods that can be shared by many array-like classes or subclasses:
+    Series
+    Index
+    ExtensionArray
+"""
+
+from __future__ import annotations
+
+import operator
+from typing import Any
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.ops_dispatch import maybe_dispatch_ufunc_to_dunder_op
+
+from pandas.core.dtypes.generic import ABCNDFrame
+
+from pandas.core import roperator
+from pandas.core.construction import extract_array
+from pandas.core.ops.common import unpack_zerodim_and_defer
+
+REDUCTION_ALIASES = {
+    "maximum": "max",
+    "minimum": "min",
+    "add": "sum",
+    "multiply": "prod",
+}
+
+
+class OpsMixin:
+    # -------------------------------------------------------------
+    # Comparisons
+
+    def _cmp_method(self, other, op):
+        return NotImplemented
+
+    @unpack_zerodim_and_defer("__eq__")
+    def __eq__(self, other):
+        return self._cmp_method(other, operator.eq)
+
+    @unpack_zerodim_and_defer("__ne__")
+    def __ne__(self, other):
+        return self._cmp_method(other, operator.ne)
+
+    @unpack_zerodim_and_defer("__lt__")
+    def __lt__(self, other):
+        return self._cmp_method(other, operator.lt)
+
+    @unpack_zerodim_and_defer("__le__")
+    def __le__(self, other):
+        return self._cmp_method(other, operator.le)
+
+    @unpack_zerodim_and_defer("__gt__")
+    def __gt__(self, other):
+        return self._cmp_method(other, operator.gt)
+
+    @unpack_zerodim_and_defer("__ge__")
+    def __ge__(self, other):
+        return self._cmp_method(other, operator.ge)
+
+    # -------------------------------------------------------------
+    # Logical Methods
+
+    def _logical_method(self, other, op):
+        return NotImplemented
+
+    @unpack_zerodim_and_defer("__and__")
+    def __and__(self, other):
+        return self._logical_method(other, operator.and_)
+
+    @unpack_zerodim_and_defer("__rand__")
+    def __rand__(self, other):
+        return self._logical_method(other, roperator.rand_)
+
+    @unpack_zerodim_and_defer("__or__")
+    def __or__(self, other):
+        return self._logical_method(other, operator.or_)
+
+    @unpack_zerodim_and_defer("__ror__")
+    def __ror__(self, other):
+        return self._logical_method(other, roperator.ror_)
+
+    @unpack_zerodim_and_defer("__xor__")
+    def __xor__(self, other):
+        return self._logical_method(other, operator.xor)
+
+    @unpack_zerodim_and_defer("__rxor__")
+    def __rxor__(self, other):
+        return self._logical_method(other, roperator.rxor)
+
+    # -------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _arith_method(self, other, op):
+        return NotImplemented
+
+    @unpack_zerodim_and_defer("__add__")
+    def __add__(self, other):
+        """
+        Get Addition of DataFrame and other, column-wise.
+
+        Equivalent to ``DataFrame.add(other)``.
+
+        Parameters
+        ----------
+        other : scalar, sequence, Series, dict or DataFrame
+            Object to be added to the DataFrame.
+
+        Returns
+        -------
+        DataFrame
+            The result of adding ``other`` to DataFrame.
+
+        See Also
+        --------
+        DataFrame.add : Add a DataFrame and another object, with option for index-
+            or column-oriented addition.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"height": [1.5, 2.6], "weight": [500, 800]}, index=["elk", "moose"]
+        ... )
+        >>> df
+               height  weight
+        elk       1.5     500
+        moose     2.6     800
+
+        Adding a scalar affects all rows and columns.
+
+        >>> df[["height", "weight"]] + 1.5
+               height  weight
+        elk       3.0   501.5
+        moose     4.1   801.5
+
+        Each element of a list is added to a column of the DataFrame, in order.
+
+        >>> df[["height", "weight"]] + [0.5, 1.5]
+               height  weight
+        elk       2.0   501.5
+        moose     3.1   801.5
+
+        Keys of a dictionary are aligned to the DataFrame, based on column names;
+        each value in the dictionary is added to the corresponding column.
+
+        >>> df[["height", "weight"]] + {"height": 0.5, "weight": 1.5}
+               height  weight
+        elk       2.0   501.5
+        moose     3.1   801.5
+
+        When `other` is a :class:`Series`, the index of `other` is aligned with the
+        columns of the DataFrame.
+
+        >>> s1 = pd.Series([0.5, 1.5], index=["weight", "height"])
+        >>> df[["height", "weight"]] + s1
+               height  weight
+        elk       3.0   500.5
+        moose     4.1   800.5
+
+        Even when the index of `other` is the same as the index of the DataFrame,
+        the :class:`Series` will not be reoriented. If index-wise alignment is desired,
+        :meth:`DataFrame.add` should be used with `axis='index'`.
+
+        >>> s2 = pd.Series([0.5, 1.5], index=["elk", "moose"])
+        >>> df[["height", "weight"]] + s2
+               elk  height  moose  weight
+        elk    NaN     NaN    NaN     NaN
+        moose  NaN     NaN    NaN     NaN
+
+        >>> df[["height", "weight"]].add(s2, axis="index")
+               height  weight
+        elk       2.0   500.5
+        moose     4.1   801.5
+
+        When `other` is a :class:`DataFrame`, both columns names and the
+        index are aligned.
+
+        >>> other = pd.DataFrame(
+        ...     {"height": [0.2, 0.4, 0.6]}, index=["elk", "moose", "deer"]
+        ... )
+        >>> df[["height", "weight"]] + other
+               height  weight
+        deer      NaN     NaN
+        elk       1.7     NaN
+        moose     3.0     NaN
+        """
+        return self._arith_method(other, operator.add)
+
+    @unpack_zerodim_and_defer("__radd__")
+    def __radd__(self, other):
+        return self._arith_method(other, roperator.radd)
+
+    @unpack_zerodim_and_defer("__sub__")
+    def __sub__(self, other):
+        return self._arith_method(other, operator.sub)
+
+    @unpack_zerodim_and_defer("__rsub__")
+    def __rsub__(self, other):
+        return self._arith_method(other, roperator.rsub)
+
+    @unpack_zerodim_and_defer("__mul__")
+    def __mul__(self, other):
+        return self._arith_method(other, operator.mul)
+
+    @unpack_zerodim_and_defer("__rmul__")
+    def __rmul__(self, other):
+        return self._arith_method(other, roperator.rmul)
+
+    @unpack_zerodim_and_defer("__truediv__")
+    def __truediv__(self, other):
+        return self._arith_method(other, operator.truediv)
+
+    @unpack_zerodim_and_defer("__rtruediv__")
+    def __rtruediv__(self, other):
+        return self._arith_method(other, roperator.rtruediv)
+
+    @unpack_zerodim_and_defer("__floordiv__")
+    def __floordiv__(self, other):
+        return self._arith_method(other, operator.floordiv)
+
+    @unpack_zerodim_and_defer("__rfloordiv")
+    def __rfloordiv__(self, other):
+        return self._arith_method(other, roperator.rfloordiv)
+
+    @unpack_zerodim_and_defer("__mod__")
+    def __mod__(self, other):
+        return self._arith_method(other, operator.mod)
+
+    @unpack_zerodim_and_defer("__rmod__")
+    def __rmod__(self, other):
+        return self._arith_method(other, roperator.rmod)
+
+    @unpack_zerodim_and_defer("__divmod__")
+    def __divmod__(self, other):
+        return self._arith_method(other, divmod)
+
+    @unpack_zerodim_and_defer("__rdivmod__")
+    def __rdivmod__(self, other):
+        return self._arith_method(other, roperator.rdivmod)
+
+    @unpack_zerodim_and_defer("__pow__")
+    def __pow__(self, other):
+        return self._arith_method(other, operator.pow)
+
+    @unpack_zerodim_and_defer("__rpow__")
+    def __rpow__(self, other):
+        return self._arith_method(other, roperator.rpow)
+
+
+# -----------------------------------------------------------------------------
+# Helpers to implement __array_ufunc__
+
+
+def array_ufunc(self, ufunc: np.ufunc, method: str, *inputs: Any, **kwargs: Any):
+    """
+    Compatibility with numpy ufuncs.
+
+    See also
+    --------
+    numpy.org/doc/stable/reference/arrays.classes.html#numpy.class.__array_ufunc__
+    """
+    from pandas.core.frame import (
+        DataFrame,
+        Series,
+    )
+    from pandas.core.generic import NDFrame
+    from pandas.core.internals import BlockManager
+
+    cls = type(self)
+
+    kwargs = _standardize_out_kwarg(**kwargs)
+
+    # for binary ops, use our custom dunder methods
+    result = maybe_dispatch_ufunc_to_dunder_op(self, ufunc, method, *inputs, **kwargs)
+    if result is not NotImplemented:
+        return result
+
+    # Determine if we should defer.
+    no_defer = (
+        np.ndarray.__array_ufunc__,
+        cls.__array_ufunc__,
+    )
+
+    for item in inputs:
+        higher_priority = (
+            hasattr(item, "__array_priority__")
+            and item.__array_priority__ > self.__array_priority__
+        )
+        has_array_ufunc = (
+            hasattr(item, "__array_ufunc__")
+            and type(item).__array_ufunc__ not in no_defer
+            and not isinstance(item, self._HANDLED_TYPES)
+        )
+        if higher_priority or has_array_ufunc:
+            return NotImplemented
+
+    # align all the inputs.
+    types = tuple(type(x) for x in inputs)
+    alignable = [x for x, t in zip(inputs, types) if issubclass(t, NDFrame)]
+
+    if len(alignable) > 1:
+        # This triggers alignment.
+        # At the moment, there aren't any ufuncs with more than two inputs
+        # so this ends up just being x1.index | x2.index, but we write
+        # it to handle *args.
+        set_types = set(types)
+        if len(set_types) > 1 and {DataFrame, Series}.issubset(set_types):
+            # We currently don't handle ufunc(DataFrame, Series)
+            # well. Previously this raised an internal ValueError. We might
+            # support it someday, so raise a NotImplementedError.
+            raise NotImplementedError(
+                f"Cannot apply ufunc {ufunc} to mixed DataFrame and Series inputs."
+            )
+        axes = self.axes
+        for obj in alignable[1:]:
+            # this relies on the fact that we aren't handling mixed
+            # series / frame ufuncs.
+            for i, (ax1, ax2) in enumerate(zip(axes, obj.axes)):
+                axes[i] = ax1.union(ax2)
+
+        reconstruct_axes = dict(zip(self._AXIS_ORDERS, axes))
+        inputs = tuple(
+            x.reindex(**reconstruct_axes) if issubclass(t, NDFrame) else x
+            for x, t in zip(inputs, types)
+        )
+    else:
+        reconstruct_axes = dict(zip(self._AXIS_ORDERS, self.axes))
+
+    if self.ndim == 1:
+        names = {x.name for x in inputs if hasattr(x, "name")}
+        name = names.pop() if len(names) == 1 else None
+        reconstruct_kwargs = {"name": name}
+    else:
+        reconstruct_kwargs = {}
+
+    def reconstruct(result):
+        if ufunc.nout > 1:
+            # np.modf, np.frexp, np.divmod
+            return tuple(_reconstruct(x) for x in result)
+
+        return _reconstruct(result)
+
+    def _reconstruct(result):
+        if lib.is_scalar(result):
+            return result
+
+        if result.ndim != self.ndim:
+            if method == "outer":
+                raise NotImplementedError
+            return result
+        if isinstance(result, BlockManager):
+            # we went through BlockManager.apply e.g. np.sqrt
+            result = self._constructor_from_mgr(result, axes=result.axes)
+        else:
+            # we converted an array, lost our axes
+            result = self._constructor(
+                result, **reconstruct_axes, **reconstruct_kwargs, copy=False
+            )
+        # TODO: When we support multiple values in __finalize__, this
+        # should pass alignable to `__finalize__` instead of self.
+        # Then `np.add(a, b)` would consider attrs from both a and b
+        # when a and b are NDFrames.
+        if len(alignable) == 1:
+            result = result.__finalize__(self)
+        return result
+
+    if "out" in kwargs:
+        # e.g. test_multiindex_get_loc
+        result = dispatch_ufunc_with_out(self, ufunc, method, *inputs, **kwargs)
+        return reconstruct(result)
+
+    if method == "reduce":
+        # e.g. test.series.test_ufunc.test_reduce
+        result = dispatch_reduction_ufunc(self, ufunc, method, *inputs, **kwargs)
+        if result is not NotImplemented:
+            return result
+
+    # We still get here with kwargs `axis` for e.g. np.maximum.accumulate
+    #  and `dtype` and `keepdims` for np.ptp
+
+    if self.ndim > 1 and (len(inputs) > 1 or ufunc.nout > 1):
+        # Just give up on preserving types in the complex case.
+        # In theory we could preserve them for them.
+        # * nout>1 is doable if BlockManager.apply took nout and
+        #   returned a Tuple[BlockManager].
+        # * len(inputs) > 1 is doable when we know that we have
+        #   aligned blocks / dtypes.
+
+        # e.g. my_ufunc, modf, logaddexp, heaviside, subtract, add
+        inputs = tuple(np.asarray(x) for x in inputs)
+        # Note: we can't use default_array_ufunc here bc reindexing means
+        #  that `self` may not be among `inputs`
+        result = getattr(ufunc, method)(*inputs, **kwargs)
+    elif self.ndim == 1:
+        # ufunc(series, ...)
+        inputs = tuple(extract_array(x, extract_numpy=True) for x in inputs)
+        result = getattr(ufunc, method)(*inputs, **kwargs)
+    else:
+        # ufunc(dataframe)
+        if method == "__call__" and not kwargs:
+            # for np.<ufunc>(..) calls
+            # kwargs cannot necessarily be handled block-by-block, so only
+            # take this path if there are no kwargs
+            mgr = inputs[0]._mgr  # pyright: ignore[reportGeneralTypeIssues]
+            result = mgr.apply(getattr(ufunc, method))
+        else:
+            # otherwise specific ufunc methods (eg np.<ufunc>.accumulate(..))
+            # Those can have an axis keyword and thus can't be called block-by-block
+            result = default_array_ufunc(inputs[0], ufunc, method, *inputs, **kwargs)  # pyright: ignore[reportGeneralTypeIssues]
+            # e.g. np.negative (only one reached), with "where" and "out" in kwargs
+
+    result = reconstruct(result)
+    return result
+
+
+def _standardize_out_kwarg(**kwargs) -> dict:
+    """
+    If kwargs contain "out1" and "out2", replace that with a tuple "out"
+
+    np.divmod, np.modf, np.frexp can have either `out=(out1, out2)` or
+    `out1=out1, out2=out2)`
+    """
+    if "out" not in kwargs and "out1" in kwargs and "out2" in kwargs:
+        out1 = kwargs.pop("out1")
+        out2 = kwargs.pop("out2")
+        out = (out1, out2)
+        kwargs["out"] = out
+    return kwargs
+
+
+def dispatch_ufunc_with_out(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+    """
+    If we have an `out` keyword, then call the ufunc without `out` and then
+    set the result into the given `out`.
+    """
+
+    # Note: we assume _standardize_out_kwarg has already been called.
+    out = kwargs.pop("out")
+    where = kwargs.pop("where", None)
+
+    result = getattr(ufunc, method)(*inputs, **kwargs)
+
+    if result is NotImplemented:
+        return NotImplemented
+
+    if isinstance(result, tuple):
+        # i.e. np.divmod, np.modf, np.frexp
+        if not isinstance(out, tuple) or len(out) != len(result):
+            raise NotImplementedError
+
+        for arr, res in zip(out, result):
+            _assign_where(arr, res, where)
+
+        return out
+
+    if isinstance(out, tuple):
+        if len(out) == 1:
+            out = out[0]
+        else:
+            raise NotImplementedError
+
+    _assign_where(out, result, where)
+    return out
+
+
+def _assign_where(out, result, where) -> None:
+    """
+    Set a ufunc result into 'out', masking with a 'where' argument if necessary.
+    """
+    if where is None:
+        # no 'where' arg passed to ufunc
+        out[:] = result
+    else:
+        np.putmask(out, where, result)
+
+
+def default_array_ufunc(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+    """
+    Fallback to the behavior we would get if we did not define __array_ufunc__.
+
+    Notes
+    -----
+    We are assuming that `self` is among `inputs`.
+    """
+    if not any(x is self for x in inputs):
+        raise NotImplementedError
+
+    new_inputs = [x if x is not self else np.asarray(x) for x in inputs]
+
+    return getattr(ufunc, method)(*new_inputs, **kwargs)
+
+
+def dispatch_reduction_ufunc(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+    """
+    Dispatch ufunc reductions to self's reduction methods.
+    """
+    assert method == "reduce"
+
+    if len(inputs) != 1 or inputs[0] is not self:
+        return NotImplemented
+
+    if ufunc.__name__ not in REDUCTION_ALIASES:
+        return NotImplemented
+
+    method_name = REDUCTION_ALIASES[ufunc.__name__]
+
+    # NB: we are assuming that min/max represent minimum/maximum methods,
+    #  which would not be accurate for e.g. Timestamp.min
+    if not hasattr(self, method_name):
+        return NotImplemented
+
+    if self.ndim > 1:
+        if isinstance(self, ABCNDFrame):
+            # TODO: test cases where this doesn't hold, i.e. 2D DTA/TDA
+            kwargs["numeric_only"] = False
+
+        if "axis" not in kwargs:
+            # For DataFrame reductions we don't want the default axis=0
+            # Note: np.min is not a ufunc, but uses array_function_dispatch,
+            #  so calls DataFrame.min (without ever getting here) with the np.min
+            #  default of axis=None, which DataFrame.min catches and changes to axis=0.
+            # np.minimum.reduce(df) gets here bc axis is not in kwargs,
+            #  so we set axis=0 to match the behavior of np.minimum.reduce(df.values)
+            kwargs["axis"] = 0
+
+    # By default, numpy's reductions do not skip NaNs, so we have to
+    #  pass skipna=False
+    return getattr(self, method_name)(skipna=False, **kwargs)
diff --git a/pandas/pandas/core/arrays/__init__.py b/pandas/pandas/core/arrays/__init__.py
new file mode 100644
index 0000000000000..f183e9236471e
--- /dev/null
+++ b/pandas/pandas/core/arrays/__init__.py
@@ -0,0 +1,43 @@
+from pandas.core.arrays.arrow import ArrowExtensionArray
+from pandas.core.arrays.base import (
+    ExtensionArray,
+    ExtensionOpsMixin,
+    ExtensionScalarOpsMixin,
+)
+from pandas.core.arrays.boolean import BooleanArray
+from pandas.core.arrays.categorical import Categorical
+from pandas.core.arrays.datetimes import DatetimeArray
+from pandas.core.arrays.floating import FloatingArray
+from pandas.core.arrays.integer import IntegerArray
+from pandas.core.arrays.interval import IntervalArray
+from pandas.core.arrays.masked import BaseMaskedArray
+from pandas.core.arrays.numpy_ import NumpyExtensionArray
+from pandas.core.arrays.period import (
+    PeriodArray,
+    period_array,
+)
+from pandas.core.arrays.sparse import SparseArray
+from pandas.core.arrays.string_ import StringArray
+from pandas.core.arrays.string_arrow import ArrowStringArray
+from pandas.core.arrays.timedeltas import TimedeltaArray
+
+__all__ = [
+    "ArrowExtensionArray",
+    "ArrowStringArray",
+    "BaseMaskedArray",
+    "BooleanArray",
+    "Categorical",
+    "DatetimeArray",
+    "ExtensionArray",
+    "ExtensionOpsMixin",
+    "ExtensionScalarOpsMixin",
+    "FloatingArray",
+    "IntegerArray",
+    "IntervalArray",
+    "NumpyExtensionArray",
+    "PeriodArray",
+    "SparseArray",
+    "StringArray",
+    "TimedeltaArray",
+    "period_array",
+]
diff --git a/pandas/pandas/core/arrays/_arrow_string_mixins.py b/pandas/pandas/core/arrays/_arrow_string_mixins.py
new file mode 100644
index 0000000000000..1ca52ce64bd77
--- /dev/null
+++ b/pandas/pandas/core/arrays/_arrow_string_mixins.py
@@ -0,0 +1,360 @@
+from __future__ import annotations
+
+from functools import partial
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.compat import (
+    pa_version_under10p1,
+    pa_version_under11p0,
+    pa_version_under13p0,
+    pa_version_under17p0,
+)
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+    import pyarrow.compute as pc
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        Scalar,
+        Self,
+    )
+
+
+class ArrowStringArrayMixin:
+    _pa_array: pa.ChunkedArray
+
+    def __init__(self, *args, **kwargs) -> None:
+        raise NotImplementedError
+
+    def _convert_bool_result(self, result, na=lib.no_default, method_name=None):
+        # Convert a bool-dtype result to the appropriate result type
+        raise NotImplementedError
+
+    def _convert_int_result(self, result):
+        # Convert an integer-dtype result to the appropriate result type
+        raise NotImplementedError
+
+    def _apply_elementwise(self, func: Callable) -> list[list[Any]]:
+        raise NotImplementedError
+
+    def _str_len(self):
+        result = pc.utf8_length(self._pa_array)
+        return self._convert_int_result(result)
+
+    def _str_lower(self) -> Self:
+        return type(self)(pc.utf8_lower(self._pa_array))
+
+    def _str_upper(self) -> Self:
+        return type(self)(pc.utf8_upper(self._pa_array))
+
+    def _str_strip(self, to_strip=None) -> Self:
+        if to_strip is None:
+            result = pc.utf8_trim_whitespace(self._pa_array)
+        else:
+            result = pc.utf8_trim(self._pa_array, characters=to_strip)
+        return type(self)(result)
+
+    def _str_lstrip(self, to_strip=None) -> Self:
+        if to_strip is None:
+            result = pc.utf8_ltrim_whitespace(self._pa_array)
+        else:
+            result = pc.utf8_ltrim(self._pa_array, characters=to_strip)
+        return type(self)(result)
+
+    def _str_rstrip(self, to_strip=None) -> Self:
+        if to_strip is None:
+            result = pc.utf8_rtrim_whitespace(self._pa_array)
+        else:
+            result = pc.utf8_rtrim(self._pa_array, characters=to_strip)
+        return type(self)(result)
+
+    def _str_pad(
+        self,
+        width: int,
+        side: Literal["left", "right", "both"] = "left",
+        fillchar: str = " ",
+    ) -> Self:
+        if side == "left":
+            pa_pad = pc.utf8_lpad
+        elif side == "right":
+            pa_pad = pc.utf8_rpad
+        elif side == "both":
+            if pa_version_under17p0:
+                # GH#59624 fall back to object dtype
+                from pandas import array
+
+                obj_arr = self.astype(object, copy=False)  # type: ignore[attr-defined]
+                obj = array(obj_arr, dtype=object)
+                result = obj._str_pad(width, side, fillchar)  # type: ignore[attr-defined]
+                return type(self)._from_sequence(result, dtype=self.dtype)  # type: ignore[attr-defined]
+            else:
+                # GH#54792
+                # https://github.com/apache/arrow/issues/15053#issuecomment-2317032347
+                lean_left = (width % 2) == 0
+                pa_pad = partial(pc.utf8_center, lean_left_on_odd_padding=lean_left)
+        else:
+            raise ValueError(
+                f"Invalid side: {side}. Side must be one of 'left', 'right', 'both'"
+            )
+        return type(self)(pa_pad(self._pa_array, width=width, padding=fillchar))
+
+    def _str_get(self, i: int) -> Self:
+        lengths = pc.utf8_length(self._pa_array)
+        if i >= 0:
+            out_of_bounds = pc.greater_equal(i, lengths)
+            start = i
+            stop = i + 1
+            step = 1
+        else:
+            out_of_bounds = pc.greater(-i, lengths)
+            start = i
+            stop = i - 1
+            step = -1
+        not_out_of_bounds = pc.invert(out_of_bounds.fill_null(True))
+        selected = pc.utf8_slice_codeunits(
+            self._pa_array, start=start, stop=stop, step=step
+        )
+        null_value = pa.scalar(None, type=self._pa_array.type)
+        result = pc.if_else(not_out_of_bounds, selected, null_value)
+        return type(self)(result)
+
+    def _str_slice(
+        self, start: int | None = None, stop: int | None = None, step: int | None = None
+    ) -> Self:
+        if pa_version_under11p0:
+            # GH#59724
+            result = self._apply_elementwise(lambda val: val[start:stop:step])
+            return type(self)(pa.chunked_array(result, type=self._pa_array.type))
+        if start is None:
+            if step is not None and step < 0:
+                # GH#59710
+                start = -1
+            else:
+                start = 0
+        if step is None:
+            step = 1
+        return type(self)(
+            pc.utf8_slice_codeunits(self._pa_array, start=start, stop=stop, step=step)
+        )
+
+    def _str_slice_replace(
+        self, start: int | None = None, stop: int | None = None, repl: str | None = None
+    ) -> Self:
+        if repl is None:
+            repl = ""
+        if start is None:
+            start = 0
+        if stop is None:
+            stop = np.iinfo(np.int64).max
+        return type(self)(pc.utf8_replace_slice(self._pa_array, start, stop, repl))
+
+    def _str_replace(
+        self,
+        pat: str | re.Pattern,
+        repl: str | Callable,
+        n: int = -1,
+        case: bool = True,
+        flags: int = 0,
+        regex: bool = True,
+    ) -> Self:
+        if isinstance(pat, re.Pattern) or callable(repl) or not case or flags:
+            raise NotImplementedError(
+                "replace is not supported with a re.Pattern, callable repl, "
+                "case=False, or flags!=0"
+            )
+
+        func = pc.replace_substring_regex if regex else pc.replace_substring
+        # https://github.com/apache/arrow/issues/39149
+        # GH 56404, unexpected behavior with negative max_replacements with pyarrow.
+        pa_max_replacements = None if n < 0 else n
+        result = func(
+            self._pa_array,
+            pattern=pat,
+            replacement=repl,
+            max_replacements=pa_max_replacements,
+        )
+        return type(self)(result)
+
+    def _str_capitalize(self) -> Self:
+        return type(self)(pc.utf8_capitalize(self._pa_array))
+
+    def _str_title(self) -> Self:
+        return type(self)(pc.utf8_title(self._pa_array))
+
+    def _str_swapcase(self) -> Self:
+        return type(self)(pc.utf8_swapcase(self._pa_array))
+
+    def _str_removeprefix(self, prefix: str):
+        if not pa_version_under13p0:
+            starts_with = pc.starts_with(self._pa_array, pattern=prefix)
+            removed = pc.utf8_slice_codeunits(self._pa_array, len(prefix))
+            result = pc.if_else(starts_with, removed, self._pa_array)
+            return type(self)(result)
+        predicate = lambda val: val.removeprefix(prefix)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_removesuffix(self, suffix: str):
+        ends_with = pc.ends_with(self._pa_array, pattern=suffix)
+        removed = pc.utf8_slice_codeunits(self._pa_array, 0, stop=-len(suffix))
+        result = pc.if_else(ends_with, removed, self._pa_array)
+        return type(self)(result)
+
+    def _str_startswith(
+        self, pat: str | tuple[str, ...], na: Scalar | lib.NoDefault = lib.no_default
+    ):
+        if isinstance(pat, str):
+            result = pc.starts_with(self._pa_array, pattern=pat)
+        else:
+            if len(pat) == 0:
+                # For empty tuple we return null for missing values and False
+                #  for valid values.
+                result = pc.if_else(pc.is_null(self._pa_array), None, False)
+            else:
+                result = pc.starts_with(self._pa_array, pattern=pat[0])
+
+                for p in pat[1:]:
+                    result = pc.or_(result, pc.starts_with(self._pa_array, pattern=p))
+        return self._convert_bool_result(result, na=na, method_name="startswith")
+
+    def _str_endswith(
+        self, pat: str | tuple[str, ...], na: Scalar | lib.NoDefault = lib.no_default
+    ):
+        if isinstance(pat, str):
+            result = pc.ends_with(self._pa_array, pattern=pat)
+        else:
+            if len(pat) == 0:
+                # For empty tuple we return null for missing values and False
+                #  for valid values.
+                result = pc.if_else(pc.is_null(self._pa_array), None, False)
+            else:
+                result = pc.ends_with(self._pa_array, pattern=pat[0])
+
+                for p in pat[1:]:
+                    result = pc.or_(result, pc.ends_with(self._pa_array, pattern=p))
+        return self._convert_bool_result(result, na=na, method_name="endswith")
+
+    def _str_isalnum(self):
+        result = pc.utf8_is_alnum(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isalpha(self):
+        result = pc.utf8_is_alpha(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isascii(self):
+        result = pc.string_is_ascii(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isdecimal(self):
+        result = pc.utf8_is_decimal(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isdigit(self):
+        result = pc.utf8_is_digit(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_islower(self):
+        result = pc.utf8_is_lower(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isnumeric(self):
+        result = pc.utf8_is_numeric(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isspace(self):
+        result = pc.utf8_is_space(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_istitle(self):
+        result = pc.utf8_is_title(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_isupper(self):
+        result = pc.utf8_is_upper(self._pa_array)
+        return self._convert_bool_result(result)
+
+    def _str_contains(
+        self,
+        pat,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+        regex: bool = True,
+    ):
+        if flags:
+            raise NotImplementedError(f"contains not implemented with {flags=}")
+
+        if regex:
+            pa_contains = pc.match_substring_regex
+        else:
+            pa_contains = pc.match_substring
+        result = pa_contains(self._pa_array, pat, ignore_case=not case)
+        return self._convert_bool_result(result, na=na, method_name="contains")
+
+    def _str_match(
+        self,
+        pat: str,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        if not pat.startswith("^"):
+            pat = f"^{pat}"
+        return self._str_contains(pat, case, flags, na, regex=True)
+
+    def _str_fullmatch(
+        self,
+        pat,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        if not pat.endswith("$") or pat.endswith("\\$"):
+            pat = f"{pat}$"
+        return self._str_match(pat, case, flags, na)
+
+    def _str_find(self, sub: str, start: int = 0, end: int | None = None):
+        if (
+            pa_version_under13p0
+            and not (start != 0 and end is not None)
+            and not (start == 0 and end is None)
+        ):
+            # GH#59562
+            res_list = self._apply_elementwise(lambda val: val.find(sub, start, end))
+            return self._convert_int_result(pa.chunked_array(res_list))
+
+        if (start == 0 or start is None) and end is None:
+            result = pc.find_substring(self._pa_array, sub)
+        else:
+            if sub == "":
+                # GH#56792
+                res_list = self._apply_elementwise(
+                    lambda val: val.find(sub, start, end)
+                )
+                return self._convert_int_result(pa.chunked_array(res_list))
+            if start is None:
+                start_offset = 0
+                start = 0
+            elif start < 0:
+                start_offset = pc.add(start, pc.utf8_length(self._pa_array))
+                start_offset = pc.if_else(pc.less(start_offset, 0), 0, start_offset)
+            else:
+                start_offset = start
+            slices = pc.utf8_slice_codeunits(self._pa_array, start, stop=end)
+            result = pc.find_substring(slices, sub)
+            found = pc.not_equal(result, pa.scalar(-1, type=result.type))
+            offset_result = pc.add(result, start_offset)
+            result = pc.if_else(found, offset_result, -1)
+        return self._convert_int_result(result)
diff --git a/pandas/pandas/core/arrays/_mixins.py b/pandas/pandas/core/arrays/_mixins.py
new file mode 100644
index 0000000000000..4e6f20e6ad3dd
--- /dev/null
+++ b/pandas/pandas/core/arrays/_mixins.py
@@ -0,0 +1,525 @@
+from __future__ import annotations
+
+from functools import wraps
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.arrays import NDArrayBacked
+from pandas._libs.tslibs import is_supported_dtype
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    Dtype,
+    F,
+    FillnaOptions,
+    PositionalIndexer2D,
+    PositionalIndexerTuple,
+    ScalarIndexer,
+    Self,
+    SequenceIndexer,
+    Shape,
+    TakeIndexer,
+    npt,
+)
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import doc
+from pandas.util._validators import (
+    validate_bool_kwarg,
+    validate_insert_loc,
+)
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import array_equivalent
+
+from pandas.core import missing
+from pandas.core.algorithms import (
+    take,
+    unique,
+    value_counts_internal as value_counts,
+)
+from pandas.core.array_algos.quantile import quantile_with_mask
+from pandas.core.array_algos.transforms import shift
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.construction import extract_array
+from pandas.core.indexers import check_array_indexer
+from pandas.core.sorting import nargminmax
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from pandas._typing import (
+        NumpySorter,
+        NumpyValueArrayLike,
+    )
+
+    from pandas import Series
+
+
+def ravel_compat(meth: F) -> F:
+    """
+    Decorator to ravel a 2D array before passing it to a cython operation,
+    then reshape the result to our own shape.
+    """
+
+    @wraps(meth)
+    def method(self, *args, **kwargs):
+        if self.ndim == 1:
+            return meth(self, *args, **kwargs)
+
+        flags = self._ndarray.flags
+        flat = self.ravel("K")
+        result = meth(flat, *args, **kwargs)
+        order = "F" if flags.f_contiguous else "C"
+        return result.reshape(self.shape, order=order)
+
+    return cast(F, method)
+
+
+# error: Definition of "delete/ravel/T/repeat/copy" in base class "NDArrayBacked"
+# is incompatible with definition in base class "ExtensionArray"
+class NDArrayBackedExtensionArray(NDArrayBacked, ExtensionArray):  # type: ignore[misc]
+    """
+    ExtensionArray that is backed by a single NumPy ndarray.
+    """
+
+    _ndarray: np.ndarray
+
+    # scalar used to denote NA value inside our self._ndarray, e.g. -1
+    #  for Categorical, iNaT for Period. Outside of object dtype,
+    #  self.isna() should be exactly locations in self._ndarray with
+    #  _internal_fill_value.
+    _internal_fill_value: Any
+
+    def _box_func(self, x):
+        """
+        Wrap numpy type in our dtype.type if necessary.
+        """
+        return x
+
+    def _validate_scalar(self, value):
+        # used by NDArrayBackedExtensionIndex.insert
+        raise AbstractMethodError(self)
+
+    # ------------------------------------------------------------------------
+
+    def view(self, dtype: Dtype | None = None) -> ArrayLike:
+        # We handle datetime64, datetime64tz, timedelta64, and period
+        #  dtypes here. Everything else we pass through to the underlying
+        #  ndarray.
+        if dtype is None or dtype is self.dtype:
+            return self._from_backing_data(self._ndarray)
+
+        if isinstance(dtype, type):
+            # we sometimes pass non-dtype objects, e.g np.ndarray;
+            #  pass those through to the underlying ndarray
+            return self._ndarray.view(dtype)
+
+        dtype = pandas_dtype(dtype)
+        arr = self._ndarray
+
+        if isinstance(dtype, PeriodDtype):
+            cls = dtype.construct_array_type()
+            return cls(arr.view("i8"), dtype=dtype)
+        elif isinstance(dtype, DatetimeTZDtype):
+            dt_cls = dtype.construct_array_type()
+            dt64_values = arr.view(f"M8[{dtype.unit}]")
+            return dt_cls._simple_new(dt64_values, dtype=dtype)
+        elif lib.is_np_dtype(dtype, "M") and is_supported_dtype(dtype):
+            from pandas.core.arrays import DatetimeArray
+
+            dt64_values = arr.view(dtype)
+            return DatetimeArray._simple_new(dt64_values, dtype=dtype)
+
+        elif lib.is_np_dtype(dtype, "m") and is_supported_dtype(dtype):
+            from pandas.core.arrays import TimedeltaArray
+
+            td64_values = arr.view(dtype)
+            return TimedeltaArray._simple_new(td64_values, dtype=dtype)
+
+        # error: Argument "dtype" to "view" of "_ArrayOrScalarCommon" has incompatible
+        # type "Union[ExtensionDtype, dtype[Any]]"; expected "Union[dtype[Any], None,
+        # type, _SupportsDType, str, Union[Tuple[Any, int], Tuple[Any, Union[int,
+        # Sequence[int]]], List[Any], _DTypeDict, Tuple[Any, Any]]]"
+        return arr.view(dtype=dtype)  # type: ignore[arg-type]
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        *,
+        allow_fill: bool = False,
+        fill_value: Any = None,
+        axis: AxisInt = 0,
+    ) -> Self:
+        if allow_fill:
+            fill_value = self._validate_scalar(fill_value)
+
+        new_data = take(
+            self._ndarray,
+            indices,
+            allow_fill=allow_fill,
+            fill_value=fill_value,
+            axis=axis,
+        )
+        return self._from_backing_data(new_data)
+
+    # ------------------------------------------------------------------------
+
+    def equals(self, other) -> bool:
+        if type(self) is not type(other):
+            return False
+        if self.dtype != other.dtype:
+            return False
+        return bool(array_equivalent(self._ndarray, other._ndarray, dtype_equal=True))
+
+    @classmethod
+    def _from_factorized(cls, values, original):
+        assert values.dtype == original._ndarray.dtype
+        return original._from_backing_data(values)
+
+    def _values_for_argsort(self) -> np.ndarray:
+        return self._ndarray
+
+    def _values_for_factorize(self):
+        return self._ndarray, self._internal_fill_value
+
+    def _hash_pandas_object(
+        self, *, encoding: str, hash_key: str, categorize: bool
+    ) -> npt.NDArray[np.uint64]:
+        from pandas.core.util.hashing import hash_array
+
+        values = self._ndarray
+        return hash_array(
+            values, encoding=encoding, hash_key=hash_key, categorize=categorize
+        )
+
+    # Signature of "argmin" incompatible with supertype "ExtensionArray"
+    def argmin(self, axis: AxisInt = 0, skipna: bool = True):  # type: ignore[override]
+        # override base class by adding axis keyword
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return nargminmax(self, "argmin", axis=axis)
+
+    # Signature of "argmax" incompatible with supertype "ExtensionArray"
+    def argmax(self, axis: AxisInt = 0, skipna: bool = True):  # type: ignore[override]
+        # override base class by adding axis keyword
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return nargminmax(self, "argmax", axis=axis)
+
+    def unique(self) -> Self:
+        new_data = unique(self._ndarray)
+        return self._from_backing_data(new_data)
+
+    @classmethod
+    @doc(ExtensionArray._concat_same_type)
+    def _concat_same_type(
+        cls,
+        to_concat: Sequence[Self],
+        axis: AxisInt = 0,
+    ) -> Self:
+        if not lib.dtypes_all_equal([x.dtype for x in to_concat]):
+            dtypes = {str(x.dtype) for x in to_concat}
+            raise ValueError("to_concat must have the same dtype", dtypes)
+
+        return super()._concat_same_type(to_concat, axis=axis)
+
+    @doc(ExtensionArray.searchsorted)
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        npvalue = self._validate_setitem_value(value)
+        return self._ndarray.searchsorted(npvalue, side=side, sorter=sorter)
+
+    @doc(ExtensionArray.shift)
+    def shift(self, periods: int = 1, fill_value=None) -> Self:
+        # NB: shift is always along axis=0
+        axis = 0
+        fill_value = self._validate_scalar(fill_value)
+        new_values = shift(self._ndarray, periods, axis, fill_value)
+
+        return self._from_backing_data(new_values)
+
+    def __setitem__(self, key, value) -> None:
+        key = check_array_indexer(self, key)
+        value = self._validate_setitem_value(value)
+        self._ndarray[key] = value
+
+    def _validate_setitem_value(self, value):
+        return value
+
+    @overload
+    def __getitem__(self, key: ScalarIndexer) -> Any: ...
+
+    @overload
+    def __getitem__(
+        self,
+        key: SequenceIndexer | PositionalIndexerTuple,
+    ) -> Self: ...
+
+    def __getitem__(
+        self,
+        key: PositionalIndexer2D,
+    ) -> Self | Any:
+        if lib.is_integer(key):
+            # fast-path
+            result = self._ndarray[key]
+            if self.ndim == 1:
+                return self._box_func(result)
+            return self._from_backing_data(result)
+
+        # error: Incompatible types in assignment (expression has type "ExtensionArray",
+        # variable has type "Union[int, slice, ndarray]")
+        key = extract_array(key, extract_numpy=True)  # type: ignore[assignment]
+        key = check_array_indexer(self, key)
+        result = self._ndarray[key]
+        if lib.is_scalar(result):
+            return self._box_func(result)
+
+        result = self._from_backing_data(result)
+        return result
+
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        mask = self.isna()
+        if mask.any():
+            # (for now) when self.ndim == 2, we assume axis=0
+            func = missing.get_fill_func(method, ndim=self.ndim)
+
+            npvalues = self._ndarray.T
+            if copy:
+                npvalues = npvalues.copy()
+            func(npvalues, limit=limit, limit_area=limit_area, mask=mask.T)
+            npvalues = npvalues.T
+
+            if copy:
+                new_values = self._from_backing_data(npvalues)
+            else:
+                new_values = self
+
+        else:
+            if copy:
+                new_values = self.copy()
+            else:
+                new_values = self
+        return new_values
+
+    @doc(ExtensionArray.fillna)
+    def fillna(self, value, limit: int | None = None, copy: bool = True) -> Self:
+        mask = self.isna()
+        if limit is not None and limit < len(self):
+            # mypy doesn't like that mask can be an EA which need not have `cumsum`
+            modify = mask.cumsum() > limit  # type: ignore[union-attr]
+            if modify.any():
+                # Only copy mask if necessary
+                mask = mask.copy()
+                mask[modify] = False
+        # error: Argument 2 to "check_value_size" has incompatible type
+        # "ExtensionArray"; expected "ndarray"
+        value = missing.check_value_size(
+            value,
+            mask,  # type: ignore[arg-type]
+            len(self),
+        )
+
+        if mask.any():
+            # fill with value
+            if copy:
+                new_values = self.copy()
+            else:
+                new_values = self[:]
+            new_values[mask] = value
+        else:
+            # We validate the fill_value even if there is nothing to fill
+            self._validate_setitem_value(value)
+
+            if not copy:
+                new_values = self[:]
+            else:
+                new_values = self.copy()
+        return new_values
+
+    # ------------------------------------------------------------------------
+    # Reductions
+
+    def _wrap_reduction_result(self, axis: AxisInt | None, result) -> Any:
+        if axis is None or self.ndim == 1:
+            return self._box_func(result)
+        return self._from_backing_data(result)
+
+    # ------------------------------------------------------------------------
+    # __array_function__ methods
+
+    def _putmask(self, mask: npt.NDArray[np.bool_], value) -> None:
+        """
+        Analogue to np.putmask(self, mask, value)
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+        value : scalar or listlike
+
+        Raises
+        ------
+        TypeError
+            If value cannot be cast to self.dtype.
+        """
+        value = self._validate_setitem_value(value)
+
+        np.putmask(self._ndarray, mask, value)
+
+    def _where(self: Self, mask: npt.NDArray[np.bool_], value) -> Self:
+        """
+        Analogue to np.where(mask, self, value)
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+        value : scalar or listlike
+
+        Raises
+        ------
+        TypeError
+            If value cannot be cast to self.dtype.
+        """
+        value = self._validate_setitem_value(value)
+
+        res_values = np.where(mask, self._ndarray, value)
+        if res_values.dtype != self._ndarray.dtype:
+            raise AssertionError(
+                # GH#56410
+                "Something has gone wrong, please report a bug at "
+                "github.com/pandas-dev/pandas/"
+            )
+        return self._from_backing_data(res_values)
+
+    # ------------------------------------------------------------------------
+    # Index compat methods
+
+    def insert(self, loc: int, item) -> Self:
+        """
+        Make new ExtensionArray inserting new item at location. Follows
+        Python list.append semantics for negative values.
+
+        Parameters
+        ----------
+        loc : int
+        item : object
+
+        Returns
+        -------
+        type(self)
+        """
+        loc = validate_insert_loc(loc, len(self))
+
+        code = self._validate_scalar(item)
+
+        new_vals = np.concatenate(
+            (
+                self._ndarray[:loc],
+                np.asarray([code], dtype=self._ndarray.dtype),
+                self._ndarray[loc:],
+            )
+        )
+        return self._from_backing_data(new_vals)
+
+    # ------------------------------------------------------------------------
+    # Additional array methods
+    #  These are not part of the EA API, but we implement them because
+    #  pandas assumes they're there.
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Return a Series containing counts of unique values.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of NA values.
+
+        Returns
+        -------
+        Series
+        """
+        if self.ndim != 1:
+            raise NotImplementedError
+
+        from pandas import (
+            Index,
+            Series,
+        )
+
+        if dropna:
+            # error: Unsupported operand type for ~ ("ExtensionArray")
+            values = self[~self.isna()]._ndarray  # type: ignore[operator]
+        else:
+            values = self._ndarray
+
+        result = value_counts(values, sort=False, dropna=dropna)
+
+        index_arr = self._from_backing_data(np.asarray(result.index._data))
+        index = Index(index_arr, name=result.index.name)
+        return Series(result._values, index=index, name=result.name, copy=False)
+
+    def _quantile(
+        self,
+        qs: npt.NDArray[np.float64],
+        interpolation: str,
+    ) -> Self:
+        # TODO: disable for Categorical if not ordered?
+
+        mask = np.asarray(self.isna())
+        arr = self._ndarray
+        fill_value = self._internal_fill_value
+
+        res_values = quantile_with_mask(arr, mask, fill_value, qs, interpolation)
+        if res_values.dtype == self._ndarray.dtype:
+            return self._from_backing_data(res_values)
+        else:
+            # e.g. test_quantile_empty we are empty integer dtype and res_values
+            #  has floating dtype
+            # TODO: technically __init__ isn't defined here.
+            #  Should we raise NotImplementedError and handle this on NumpyEA?
+            return type(self)(res_values)  # type: ignore[call-arg]
+
+    # ------------------------------------------------------------------------
+    # numpy-like methods
+
+    @classmethod
+    def _empty(cls, shape: Shape, dtype: ExtensionDtype) -> Self:
+        """
+        Analogous to np.empty(shape, dtype=dtype)
+
+        Parameters
+        ----------
+        shape : tuple[int]
+        dtype : ExtensionDtype
+        """
+        # The base implementation uses a naive approach to find the dtype
+        #  for the backing ndarray
+        arr = cls._from_sequence([], dtype=dtype)
+        backing = np.empty(shape, dtype=arr._ndarray.dtype)
+        return arr._from_backing_data(backing)
diff --git a/pandas/pandas/core/arrays/_ranges.py b/pandas/pandas/core/arrays/_ranges.py
new file mode 100644
index 0000000000000..88f5ac4ebdea4
--- /dev/null
+++ b/pandas/pandas/core/arrays/_ranges.py
@@ -0,0 +1,200 @@
+"""
+Helper functions to generate range-like data for DatetimeArray
+(and possibly TimedeltaArray/PeriodArray)
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs.lib import i8max
+from pandas._libs.tslibs import (
+    BaseOffset,
+    OutOfBoundsDatetime,
+    Timedelta,
+    Timestamp,
+    iNaT,
+)
+
+from pandas.core.construction import range_to_ndarray
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+
+def generate_regular_range(
+    start: Timestamp | Timedelta | None,
+    end: Timestamp | Timedelta | None,
+    periods: int | None,
+    freq: BaseOffset,
+    unit: str = "ns",
+) -> npt.NDArray[np.intp]:
+    """
+    Generate a range of dates or timestamps with the spans between dates
+    described by the given `freq` DateOffset.
+
+    Parameters
+    ----------
+    start : Timedelta, Timestamp or None
+        First point of produced date range.
+    end : Timedelta, Timestamp or None
+        Last point of produced date range.
+    periods : int or None
+        Number of periods in produced date range.
+    freq : Tick
+        Describes space between dates in produced date range.
+    unit : str, default "ns"
+        The resolution the output is meant to represent.
+
+    Returns
+    -------
+    ndarray[np.int64]
+        Representing the given resolution.
+    """
+    istart = start._value if start is not None else None
+    iend = end._value if end is not None else None
+    freq.nanos  # raises if non-fixed frequency
+    td = Timedelta(freq)
+    b: int
+    e: int
+    try:
+        td = td.as_unit(unit, round_ok=False)
+    except ValueError as err:
+        raise ValueError(
+            f"freq={freq} is incompatible with unit={unit}. "
+            "Use a lower freq or a higher unit instead."
+        ) from err
+    stride = int(td._value)
+
+    if periods is None and istart is not None and iend is not None:
+        b = istart
+        # cannot just use e = Timestamp(end) + 1 because arange breaks when
+        # stride is too large, see GH10887
+        e = b + (iend - b) // stride * stride + stride // 2 + 1
+    elif istart is not None and periods is not None:
+        b = istart
+        e = _generate_range_overflow_safe(b, periods, stride, side="start")
+    elif iend is not None and periods is not None:
+        e = iend + stride
+        b = _generate_range_overflow_safe(e, periods, stride, side="end")
+    else:
+        raise ValueError(
+            "at least 'start' or 'end' should be specified if a 'period' is given."
+        )
+
+    return range_to_ndarray(range(b, e, stride))
+
+
+def _generate_range_overflow_safe(
+    endpoint: int, periods: int, stride: int, side: str = "start"
+) -> int:
+    """
+    Calculate the second endpoint for passing to np.arange, checking
+    to avoid an integer overflow.  Catch OverflowError and re-raise
+    as OutOfBoundsDatetime.
+
+    Parameters
+    ----------
+    endpoint : int
+        nanosecond timestamp of the known endpoint of the desired range
+    periods : int
+        number of periods in the desired range
+    stride : int
+        nanoseconds between periods in the desired range
+    side : {'start', 'end'}
+        which end of the range `endpoint` refers to
+
+    Returns
+    -------
+    other_end : int
+
+    Raises
+    ------
+    OutOfBoundsDatetime
+    """
+    # GH#14187 raise instead of incorrectly wrapping around
+    assert side in ["start", "end"]
+
+    i64max = np.uint64(i8max)
+    msg = f"Cannot generate range with {side}={endpoint} and periods={periods}"
+
+    with np.errstate(over="raise"):
+        # if periods * strides cannot be multiplied within the *uint64* bounds,
+        #  we cannot salvage the operation by recursing, so raise
+        try:
+            addend = np.uint64(periods) * np.uint64(np.abs(stride))
+        except FloatingPointError as err:
+            raise OutOfBoundsDatetime(msg) from err
+
+    if np.abs(addend) <= i64max:
+        # relatively easy case without casting concerns
+        return _generate_range_overflow_safe_signed(endpoint, periods, stride, side)
+
+    elif (endpoint > 0 and side == "start" and stride > 0) or (
+        endpoint < 0 < stride and side == "end"
+    ):
+        # no chance of not-overflowing
+        raise OutOfBoundsDatetime(msg)
+
+    elif side == "end" and endpoint - stride <= i64max < endpoint:
+        # in _generate_regular_range we added `stride` thereby overflowing
+        #  the bounds.  Adjust to fix this.
+        return _generate_range_overflow_safe(
+            endpoint - stride, periods - 1, stride, side
+        )
+
+    # split into smaller pieces
+    mid_periods = periods // 2
+    remaining = periods - mid_periods
+    assert 0 < remaining < periods, (remaining, periods, endpoint, stride)
+
+    midpoint = int(_generate_range_overflow_safe(endpoint, mid_periods, stride, side))
+    return _generate_range_overflow_safe(midpoint, remaining, stride, side)
+
+
+def _generate_range_overflow_safe_signed(
+    endpoint: int, periods: int, stride: int, side: str
+) -> int:
+    """
+    A special case for _generate_range_overflow_safe where `periods * stride`
+    can be calculated without overflowing int64 bounds.
+    """
+    assert side in ["start", "end"]
+    if side == "end":
+        stride *= -1
+
+    with np.errstate(over="raise"):
+        addend = np.int64(periods) * np.int64(stride)
+        try:
+            # easy case with no overflows
+            result = np.int64(endpoint) + addend
+            if result == iNaT:
+                # Putting this into a DatetimeArray/TimedeltaArray
+                #  would incorrectly be interpreted as NaT
+                raise OverflowError
+            return int(result)
+        except (FloatingPointError, OverflowError):
+            # with endpoint negative and addend positive we risk
+            #  FloatingPointError; with reversed signed we risk OverflowError
+            pass
+
+        # if stride and endpoint had opposite signs, then endpoint + addend
+        #  should never overflow.  so they must have the same signs
+        assert (stride > 0 and endpoint >= 0) or (stride < 0 and endpoint <= 0)
+
+        if stride > 0:
+            # watch out for very special case in which we just slightly
+            #  exceed implementation bounds, but when passing the result to
+            #  np.arange will get a result slightly within the bounds
+
+            uresult = np.uint64(endpoint) + np.uint64(addend)
+            i64max = np.uint64(i8max)
+            assert uresult > i64max
+            if uresult <= i64max + np.uint64(stride):
+                return int(uresult)
+
+    raise OutOfBoundsDatetime(
+        f"Cannot generate range with {side}={endpoint} and periods={periods}"
+    )
diff --git a/pandas/pandas/core/arrays/_utils.py b/pandas/pandas/core/arrays/_utils.py
new file mode 100644
index 0000000000000..6b46396d5efdf
--- /dev/null
+++ b/pandas/pandas/core/arrays/_utils.py
@@ -0,0 +1,63 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.errors import LossySetitemError
+
+from pandas.core.dtypes.cast import np_can_hold_element
+from pandas.core.dtypes.common import is_numeric_dtype
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        npt,
+    )
+
+
+def to_numpy_dtype_inference(
+    arr: ArrayLike, dtype: npt.DTypeLike | None, na_value, hasna: bool
+) -> tuple[npt.DTypeLike, Any]:
+    if dtype is None and is_numeric_dtype(arr.dtype):
+        dtype_given = False
+        if hasna:
+            if arr.dtype.kind == "b":
+                dtype = np.dtype(np.object_)
+            else:
+                if arr.dtype.kind in "iu":
+                    dtype = np.dtype(np.float64)
+                else:
+                    dtype = arr.dtype.numpy_dtype  # type: ignore[union-attr]
+                if na_value is lib.no_default:
+                    na_value = np.nan
+        else:
+            dtype = arr.dtype.numpy_dtype  # type: ignore[union-attr]
+    elif dtype is not None:
+        dtype = np.dtype(dtype)
+        dtype_given = True
+    else:
+        dtype_given = True
+
+    if na_value is lib.no_default:
+        if dtype is None or not hasna:
+            na_value = arr.dtype.na_value
+        elif dtype.kind == "f":  # type: ignore[union-attr]
+            na_value = np.nan
+        elif dtype.kind == "M":  # type: ignore[union-attr]
+            na_value = np.datetime64("nat")
+        elif dtype.kind == "m":  # type: ignore[union-attr]
+            na_value = np.timedelta64("nat")
+        else:
+            na_value = arr.dtype.na_value
+
+    if not dtype_given and hasna:
+        try:
+            np_can_hold_element(dtype, na_value)  # type: ignore[arg-type]
+        except LossySetitemError:
+            dtype = np.dtype(np.object_)
+    return dtype, na_value
diff --git a/pandas/pandas/core/arrays/arrow/__init__.py b/pandas/pandas/core/arrays/arrow/__init__.py
new file mode 100644
index 0000000000000..50274a2de2cc1
--- /dev/null
+++ b/pandas/pandas/core/arrays/arrow/__init__.py
@@ -0,0 +1,7 @@
+from pandas.core.arrays.arrow.accessors import (
+    ListAccessor,
+    StructAccessor,
+)
+from pandas.core.arrays.arrow.array import ArrowExtensionArray
+
+__all__ = ["ArrowExtensionArray", "ListAccessor", "StructAccessor"]
diff --git a/pandas/pandas/core/arrays/arrow/_arrow_utils.py b/pandas/pandas/core/arrays/arrow/_arrow_utils.py
new file mode 100644
index 0000000000000..285c3fd465ffc
--- /dev/null
+++ b/pandas/pandas/core/arrays/arrow/_arrow_utils.py
@@ -0,0 +1,50 @@
+from __future__ import annotations
+
+import numpy as np
+import pyarrow
+
+
+def pyarrow_array_to_numpy_and_mask(
+    arr, dtype: np.dtype
+) -> tuple[np.ndarray, np.ndarray]:
+    """
+    Convert a primitive pyarrow.Array to a numpy array and boolean mask based
+    on the buffers of the Array.
+
+    At the moment pyarrow.BooleanArray is not supported.
+
+    Parameters
+    ----------
+    arr : pyarrow.Array
+    dtype : numpy.dtype
+
+    Returns
+    -------
+    (data, mask)
+        Tuple of two numpy arrays with the raw data (with specified dtype) and
+        a boolean mask (validity mask, so False means missing)
+    """
+    dtype = np.dtype(dtype)
+
+    if pyarrow.types.is_null(arr.type):
+        # No initialization of data is needed since everything is null
+        data = np.empty(len(arr), dtype=dtype)
+        mask = np.zeros(len(arr), dtype=bool)
+        return data, mask
+    buflist = arr.buffers()
+    # Since Arrow buffers might contain padding and the data might be offset,
+    # the buffer gets sliced here before handing it to numpy.
+    # See also https://github.com/pandas-dev/pandas/issues/40896
+    offset = arr.offset * dtype.itemsize
+    length = len(arr) * dtype.itemsize
+    data_buf = buflist[1][offset : offset + length]
+    data = np.frombuffer(data_buf, dtype=dtype)
+    bitmask = buflist[0]
+    if bitmask is not None:
+        mask = pyarrow.BooleanArray.from_buffers(
+            pyarrow.bool_(), len(arr), [None, bitmask], offset=arr.offset
+        )
+        mask = np.asarray(mask)
+    else:
+        mask = np.ones(len(arr), dtype=bool)
+    return data, mask
diff --git a/pandas/pandas/core/arrays/arrow/accessors.py b/pandas/pandas/core/arrays/arrow/accessors.py
new file mode 100644
index 0000000000000..b220a94d032b5
--- /dev/null
+++ b/pandas/pandas/core/arrays/arrow/accessors.py
@@ -0,0 +1,509 @@
+"""Accessors for arrow-backed data."""
+
+from __future__ import annotations
+
+from abc import (
+    ABCMeta,
+    abstractmethod,
+)
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+from pandas.compat import (
+    pa_version_under10p1,
+    pa_version_under11p0,
+)
+
+from pandas.core.dtypes.common import is_list_like
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+    import pyarrow.compute as pc
+
+    from pandas.core.dtypes.dtypes import ArrowDtype
+
+if TYPE_CHECKING:
+    from collections.abc import Iterator
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+class ArrowAccessor(metaclass=ABCMeta):
+    @abstractmethod
+    def __init__(self, data, validation_msg: str) -> None:
+        self._data = data
+        self._validation_msg = validation_msg
+        self._validate(data)
+
+    @abstractmethod
+    def _is_valid_pyarrow_dtype(self, pyarrow_dtype) -> bool:
+        pass
+
+    def _validate(self, data) -> None:
+        dtype = data.dtype
+        if pa_version_under10p1 or not isinstance(dtype, ArrowDtype):
+            # Raise AttributeError so that inspect can handle non-struct Series.
+            raise AttributeError(self._validation_msg.format(dtype=dtype))
+
+        if not self._is_valid_pyarrow_dtype(dtype.pyarrow_dtype):
+            # Raise AttributeError so that inspect can handle invalid Series.
+            raise AttributeError(self._validation_msg.format(dtype=dtype))
+
+    @property
+    def _pa_array(self):
+        return self._data.array._pa_array
+
+
+class ListAccessor(ArrowAccessor):
+    """
+    Accessor object for list data properties of the Series values.
+
+    Parameters
+    ----------
+    data : Series
+        Series containing Arrow list data.
+    """
+
+    def __init__(self, data=None) -> None:
+        super().__init__(
+            data,
+            validation_msg="Can only use the '.list' accessor with "
+            "'list[pyarrow]' dtype, not {dtype}.",
+        )
+
+    def _is_valid_pyarrow_dtype(self, pyarrow_dtype) -> bool:
+        return (
+            pa.types.is_list(pyarrow_dtype)
+            or pa.types.is_fixed_size_list(pyarrow_dtype)
+            or pa.types.is_large_list(pyarrow_dtype)
+        )
+
+    def len(self) -> Series:
+        """
+        Return the length of each list in the Series.
+
+        Returns
+        -------
+        pandas.Series
+            The length of each list.
+
+        See Also
+        --------
+        str.len : Python built-in function returning the length of an object.
+        Series.size : Returns the length of the Series.
+        StringMethods.len : Compute the length of each element in the Series/Index.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         [1, 2, 3],
+        ...         [3],
+        ...     ],
+        ...     dtype=pd.ArrowDtype(pa.list_(pa.int64())),
+        ... )
+        >>> s.list.len()
+        0    3
+        1    1
+        dtype: int32[pyarrow]
+        """
+        from pandas import Series
+
+        value_lengths = pc.list_value_length(self._pa_array)
+        return Series(
+            value_lengths,
+            dtype=ArrowDtype(value_lengths.type),
+            index=self._data.index,
+            name=self._data.name,
+        )
+
+    def __getitem__(self, key: int | slice) -> Series:
+        """
+        Index or slice lists in the Series.
+
+        Parameters
+        ----------
+        key : int | slice
+            Index or slice of indices to access from each list.
+
+        Returns
+        -------
+        pandas.Series
+            The list at requested index.
+
+        See Also
+        --------
+        ListAccessor.flatten : Flatten list values.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         [1, 2, 3],
+        ...         [3],
+        ...     ],
+        ...     dtype=pd.ArrowDtype(pa.list_(pa.int64())),
+        ... )
+        >>> s.list[0]
+        0    1
+        1    3
+        dtype: int64[pyarrow]
+        """
+        from pandas import Series
+
+        if isinstance(key, int):
+            # TODO: Support negative key but pyarrow does not allow
+            # element index to be an array.
+            # if key < 0:
+            #     key = pc.add(key, pc.list_value_length(self._pa_array))
+            element = pc.list_element(self._pa_array, key)
+            return Series(
+                element,
+                dtype=ArrowDtype(element.type),
+                index=self._data.index,
+                name=self._data.name,
+            )
+        elif isinstance(key, slice):
+            if pa_version_under11p0:
+                raise NotImplementedError(
+                    f"List slice not supported by pyarrow {pa.__version__}."
+                )
+
+            # TODO: Support negative start/stop/step, ideally this would be added
+            # upstream in pyarrow.
+            start, stop, step = key.start, key.stop, key.step
+            if start is None:
+                # TODO: When adding negative step support
+                #  this should be setto last element of array
+                # when step is negative.
+                start = 0
+            if step is None:
+                step = 1
+            sliced = pc.list_slice(self._pa_array, start, stop, step)
+            return Series(
+                sliced,
+                dtype=ArrowDtype(sliced.type),
+                index=self._data.index,
+                name=self._data.name,
+            )
+        else:
+            raise ValueError(f"key must be an int or slice, got {type(key).__name__}")
+
+    def __iter__(self) -> Iterator:
+        raise TypeError(f"'{type(self).__name__}' object is not iterable")
+
+    def flatten(self) -> Series:
+        """
+        Flatten list values.
+
+        Returns
+        -------
+        pandas.Series
+            The data from all lists in the series flattened.
+
+        See Also
+        --------
+        ListAccessor.__getitem__ : Index or slice values in the Series.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         [1, 2, 3],
+        ...         [3],
+        ...     ],
+        ...     dtype=pd.ArrowDtype(pa.list_(pa.int64())),
+        ... )
+        >>> s.list.flatten()
+        0    1
+        0    2
+        0    3
+        1    3
+        dtype: int64[pyarrow]
+        """
+        from pandas import Series
+
+        counts = pa.compute.list_value_length(self._pa_array)
+        flattened = pa.compute.list_flatten(self._pa_array)
+        index = self._data.index.repeat(counts.fill_null(pa.scalar(0, counts.type)))
+        return Series(
+            flattened,
+            dtype=ArrowDtype(flattened.type),
+            index=index,
+            name=self._data.name,
+        )
+
+
+class StructAccessor(ArrowAccessor):
+    """
+    Accessor object for structured data properties of the Series values.
+
+    Parameters
+    ----------
+    data : Series
+        Series containing Arrow struct data.
+    """
+
+    def __init__(self, data=None) -> None:
+        super().__init__(
+            data,
+            validation_msg=(
+                "Can only use the '.struct' accessor with 'struct[pyarrow]' "
+                "dtype, not {dtype}."
+            ),
+        )
+
+    def _is_valid_pyarrow_dtype(self, pyarrow_dtype) -> bool:
+        return pa.types.is_struct(pyarrow_dtype)
+
+    @property
+    def dtypes(self) -> Series:
+        """
+        Return the dtype object of each child field of the struct.
+
+        Returns
+        -------
+        pandas.Series
+            The data type of each child field.
+
+        See Also
+        --------
+        Series.dtype: Return the dtype object of the underlying data.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         {"version": 1, "project": "pandas"},
+        ...         {"version": 2, "project": "pandas"},
+        ...         {"version": 1, "project": "numpy"},
+        ...     ],
+        ...     dtype=pd.ArrowDtype(
+        ...         pa.struct([("version", pa.int64()), ("project", pa.string())])
+        ...     ),
+        ... )
+        >>> s.struct.dtypes
+        version     int64[pyarrow]
+        project    string[pyarrow]
+        dtype: object
+        """
+        from pandas import (
+            Index,
+            Series,
+        )
+
+        pa_type = self._data.dtype.pyarrow_dtype
+        types = [ArrowDtype(struct.type) for struct in pa_type]
+        names = [struct.name for struct in pa_type]
+        return Series(types, index=Index(names))
+
+    def field(
+        self,
+        name_or_index: list[str]
+        | list[bytes]
+        | list[int]
+        | pc.Expression
+        | bytes
+        | str
+        | int,
+    ) -> Series:
+        """
+        Extract a child field of a struct as a Series.
+
+        Parameters
+        ----------
+        name_or_index : str | bytes | int | expression | list
+            Name or index of the child field to extract.
+
+            For list-like inputs, this will index into a nested
+            struct.
+
+        Returns
+        -------
+        pandas.Series
+            The data corresponding to the selected child field.
+
+        See Also
+        --------
+        Series.struct.explode : Return all child fields as a DataFrame.
+
+        Notes
+        -----
+        The name of the resulting Series will be set using the following
+        rules:
+
+        - For string, bytes, or integer `name_or_index` (or a list of these, for
+          a nested selection), the Series name is set to the selected
+          field's name.
+        - For a :class:`pyarrow.compute.Expression`, this is set to
+          the string form of the expression.
+        - For list-like `name_or_index`, the name will be set to the
+          name of the final field selected.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         {"version": 1, "project": "pandas"},
+        ...         {"version": 2, "project": "pandas"},
+        ...         {"version": 1, "project": "numpy"},
+        ...     ],
+        ...     dtype=pd.ArrowDtype(
+        ...         pa.struct([("version", pa.int64()), ("project", pa.string())])
+        ...     ),
+        ... )
+
+        Extract by field name.
+
+        >>> s.struct.field("project")
+        0    pandas
+        1    pandas
+        2     numpy
+        Name: project, dtype: string[pyarrow]
+
+        Extract by field index.
+
+        >>> s.struct.field(0)
+        0    1
+        1    2
+        2    1
+        Name: version, dtype: int64[pyarrow]
+
+        Or an expression
+
+        >>> import pyarrow.compute as pc
+        >>> s.struct.field(pc.field("project"))
+        0    pandas
+        1    pandas
+        2     numpy
+        Name: project, dtype: string[pyarrow]
+
+        For nested struct types, you can pass a list of values to index
+        multiple levels:
+
+        >>> version_type = pa.struct(
+        ...     [
+        ...         ("major", pa.int64()),
+        ...         ("minor", pa.int64()),
+        ...     ]
+        ... )
+        >>> s = pd.Series(
+        ...     [
+        ...         {"version": {"major": 1, "minor": 5}, "project": "pandas"},
+        ...         {"version": {"major": 2, "minor": 1}, "project": "pandas"},
+        ...         {"version": {"major": 1, "minor": 26}, "project": "numpy"},
+        ...     ],
+        ...     dtype=pd.ArrowDtype(
+        ...         pa.struct([("version", version_type), ("project", pa.string())])
+        ...     ),
+        ... )
+        >>> s.struct.field(["version", "minor"])
+        0     5
+        1     1
+        2    26
+        Name: minor, dtype: int64[pyarrow]
+        >>> s.struct.field([0, 0])
+        0    1
+        1    2
+        2    1
+        Name: major, dtype: int64[pyarrow]
+        """
+        from pandas import Series
+
+        def get_name(
+            level_name_or_index: list[str]
+            | list[bytes]
+            | list[int]
+            | pc.Expression
+            | bytes
+            | str
+            | int,
+            data: pa.ChunkedArray,
+        ):
+            if isinstance(level_name_or_index, int):
+                name = data.type.field(level_name_or_index).name
+            elif isinstance(level_name_or_index, (str, bytes)):
+                name = level_name_or_index
+            elif isinstance(level_name_or_index, pc.Expression):
+                name = str(level_name_or_index)
+            elif is_list_like(level_name_or_index):
+                # For nested input like [2, 1, 2]
+                # iteratively get the struct and field name. The last
+                # one is used for the name of the index.
+                level_name_or_index = list(reversed(level_name_or_index))
+                selected = data
+                while level_name_or_index:
+                    # we need the cast, otherwise mypy complains about
+                    # getting ints, bytes, or str here, which isn't possible.
+                    level_name_or_index = cast(list, level_name_or_index)
+                    name_or_index = level_name_or_index.pop()
+                    name = get_name(name_or_index, selected)
+                    selected = selected.type.field(selected.type.get_field_index(name))
+                    name = selected.name
+            else:
+                raise ValueError(
+                    "name_or_index must be an int, str, bytes, "
+                    "pyarrow.compute.Expression, or list of those"
+                )
+            return name
+
+        pa_arr = self._data.array._pa_array
+        name = get_name(name_or_index, pa_arr)
+        field_arr = pc.struct_field(pa_arr, name_or_index)
+
+        return Series(
+            field_arr,
+            dtype=ArrowDtype(field_arr.type),
+            index=self._data.index,
+            name=name,
+        )
+
+    def explode(self) -> DataFrame:
+        """
+        Extract all child fields of a struct as a DataFrame.
+
+        Returns
+        -------
+        pandas.DataFrame
+            The data corresponding to all child fields.
+
+        See Also
+        --------
+        Series.struct.field : Return a single child field as a Series.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> s = pd.Series(
+        ...     [
+        ...         {"version": 1, "project": "pandas"},
+        ...         {"version": 2, "project": "pandas"},
+        ...         {"version": 1, "project": "numpy"},
+        ...     ],
+        ...     dtype=pd.ArrowDtype(
+        ...         pa.struct([("version", pa.int64()), ("project", pa.string())])
+        ...     ),
+        ... )
+
+        >>> s.struct.explode()
+           version project
+        0        1  pandas
+        1        2  pandas
+        2        1   numpy
+        """
+        from pandas import concat
+
+        pa_type = self._pa_array.type
+        return concat(
+            [self.field(i) for i in range(pa_type.num_fields)], axis="columns"
+        )
diff --git a/pandas/pandas/core/arrays/arrow/array.py b/pandas/pandas/core/arrays/arrow/array.py
new file mode 100644
index 0000000000000..e2feda495c103
--- /dev/null
+++ b/pandas/pandas/core/arrays/arrow/array.py
@@ -0,0 +1,2973 @@
+from __future__ import annotations
+
+import functools
+import operator
+import re
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import unicodedata
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    Timedelta,
+    Timestamp,
+    timezones,
+)
+from pandas.compat import (
+    pa_version_under10p1,
+    pa_version_under11p0,
+    pa_version_under13p0,
+)
+from pandas.util._decorators import doc
+
+from pandas.core.dtypes.cast import (
+    can_hold_element,
+    infer_dtype_from_scalar,
+)
+from pandas.core.dtypes.common import (
+    CategoricalDtype,
+    is_array_like,
+    is_bool_dtype,
+    is_float_dtype,
+    is_integer,
+    is_list_like,
+    is_numeric_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import (
+    algorithms as algos,
+    missing,
+    ops,
+    roperator,
+)
+from pandas.core.algorithms import map_array
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays._arrow_string_mixins import ArrowStringArrayMixin
+from pandas.core.arrays._utils import to_numpy_dtype_inference
+from pandas.core.arrays.base import (
+    ExtensionArray,
+    ExtensionArraySupportsAnyAll,
+)
+from pandas.core.arrays.masked import BaseMaskedArray
+from pandas.core.arrays.string_ import StringDtype
+import pandas.core.common as com
+from pandas.core.indexers import (
+    check_array_indexer,
+    unpack_tuple_and_ellipses,
+    validate_indices,
+)
+from pandas.core.nanops import check_below_min_count
+from pandas.core.strings.base import BaseStringArrayMethods
+
+from pandas.io._util import _arrow_dtype_mapping
+from pandas.tseries.frequencies import to_offset
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+    import pyarrow.compute as pc
+
+    from pandas.core.dtypes.dtypes import ArrowDtype
+
+    ARROW_CMP_FUNCS = {
+        "eq": pc.equal,
+        "ne": pc.not_equal,
+        "lt": pc.less,
+        "gt": pc.greater,
+        "le": pc.less_equal,
+        "ge": pc.greater_equal,
+    }
+
+    ARROW_LOGICAL_FUNCS = {
+        "and_": pc.and_kleene,
+        "rand_": lambda x, y: pc.and_kleene(y, x),
+        "or_": pc.or_kleene,
+        "ror_": lambda x, y: pc.or_kleene(y, x),
+        "xor": pc.xor,
+        "rxor": lambda x, y: pc.xor(y, x),
+    }
+
+    ARROW_BIT_WISE_FUNCS = {
+        "and_": pc.bit_wise_and,
+        "rand_": lambda x, y: pc.bit_wise_and(y, x),
+        "or_": pc.bit_wise_or,
+        "ror_": lambda x, y: pc.bit_wise_or(y, x),
+        "xor": pc.bit_wise_xor,
+        "rxor": lambda x, y: pc.bit_wise_xor(y, x),
+    }
+
+    def cast_for_truediv(
+        arrow_array: pa.ChunkedArray, pa_object: pa.Array | pa.Scalar
+    ) -> tuple[pa.ChunkedArray, pa.Array | pa.Scalar]:
+        # Ensure int / int -> float mirroring Python/Numpy behavior
+        # as pc.divide_checked(int, int) -> int
+        if pa.types.is_integer(arrow_array.type) and pa.types.is_integer(
+            pa_object.type
+        ):
+            # GH: 56645.
+            # https://github.com/apache/arrow/issues/35563
+            return pc.cast(arrow_array, pa.float64(), safe=False), pc.cast(
+                pa_object, pa.float64(), safe=False
+            )
+
+        return arrow_array, pa_object
+
+    def floordiv_compat(
+        left: pa.ChunkedArray | pa.Array | pa.Scalar,
+        right: pa.ChunkedArray | pa.Array | pa.Scalar,
+    ) -> pa.ChunkedArray:
+        # TODO: Replace with pyarrow floordiv kernel.
+        # https://github.com/apache/arrow/issues/39386
+        if pa.types.is_integer(left.type) and pa.types.is_integer(right.type):
+            divided = pc.divide_checked(left, right)
+            if pa.types.is_signed_integer(divided.type):
+                # GH 56676
+                has_remainder = pc.not_equal(pc.multiply(divided, right), left)
+                has_one_negative_operand = pc.less(
+                    pc.bit_wise_xor(left, right),
+                    pa.scalar(0, type=divided.type),
+                )
+                result = pc.if_else(
+                    pc.and_(
+                        has_remainder,
+                        has_one_negative_operand,
+                    ),
+                    # GH: 55561
+                    pc.subtract(divided, pa.scalar(1, type=divided.type)),
+                    divided,
+                )
+            else:
+                result = divided
+            result = result.cast(left.type)
+        else:
+            divided = pc.divide(left, right)
+            result = pc.floor(divided)
+        return result
+
+    ARROW_ARITHMETIC_FUNCS = {
+        "add": pc.add_checked,
+        "radd": lambda x, y: pc.add_checked(y, x),
+        "sub": pc.subtract_checked,
+        "rsub": lambda x, y: pc.subtract_checked(y, x),
+        "mul": pc.multiply_checked,
+        "rmul": lambda x, y: pc.multiply_checked(y, x),
+        "truediv": lambda x, y: pc.divide(*cast_for_truediv(x, y)),
+        "rtruediv": lambda x, y: pc.divide(*cast_for_truediv(y, x)),
+        "floordiv": lambda x, y: floordiv_compat(x, y),
+        "rfloordiv": lambda x, y: floordiv_compat(y, x),
+        "mod": NotImplemented,
+        "rmod": NotImplemented,
+        "divmod": NotImplemented,
+        "rdivmod": NotImplemented,
+        "pow": pc.power_checked,
+        "rpow": lambda x, y: pc.power_checked(y, x),
+    }
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+
+    from pandas._libs.missing import NAType
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        Dtype,
+        FillnaOptions,
+        InterpolateOptions,
+        Iterator,
+        NpDtype,
+        NumpySorter,
+        NumpyValueArrayLike,
+        PositionalIndexer,
+        Scalar,
+        Self,
+        SortKind,
+        TakeIndexer,
+        TimeAmbiguous,
+        TimeNonexistent,
+        npt,
+    )
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+    from pandas import Series
+    from pandas.core.arrays.datetimes import DatetimeArray
+    from pandas.core.arrays.timedeltas import TimedeltaArray
+
+
+def get_unit_from_pa_dtype(pa_dtype) -> str:
+    # https://github.com/pandas-dev/pandas/pull/50998#discussion_r1100344804
+    if pa_version_under11p0:
+        unit = str(pa_dtype).split("[", 1)[-1][:-1]
+        if unit not in ["s", "ms", "us", "ns"]:
+            raise ValueError(pa_dtype)
+        return unit
+    return pa_dtype.unit
+
+
+def to_pyarrow_type(
+    dtype: ArrowDtype | pa.DataType | Dtype | None,
+) -> pa.DataType | None:
+    """
+    Convert dtype to a pyarrow type instance.
+    """
+    if isinstance(dtype, ArrowDtype):
+        return dtype.pyarrow_dtype
+    elif isinstance(dtype, pa.DataType):
+        return dtype
+    elif isinstance(dtype, DatetimeTZDtype):
+        return pa.timestamp(dtype.unit, dtype.tz)
+    elif dtype:
+        try:
+            # Accepts python types too
+            # Doesn't handle all numpy types
+            return pa.from_numpy_dtype(dtype)
+        except pa.ArrowNotImplementedError:
+            pass
+    return None
+
+
+class ArrowExtensionArray(
+    OpsMixin,
+    ExtensionArraySupportsAnyAll,
+    ArrowStringArrayMixin,
+    BaseStringArrayMethods,
+):
+    """
+    Pandas ExtensionArray backed by a PyArrow ChunkedArray.
+
+    .. warning::
+
+       ArrowExtensionArray is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Parameters
+    ----------
+    values : pyarrow.Array or pyarrow.ChunkedArray
+        The input data to initialize the ArrowExtensionArray.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    Returns
+    -------
+    ArrowExtensionArray
+
+    See Also
+    --------
+    array : Create a Pandas array with a specified dtype.
+    DataFrame.to_feather : Write a DataFrame to the binary Feather format.
+    read_feather : Load a feather-format object from the file path.
+
+    Notes
+    -----
+    Most methods are implemented using `pyarrow compute functions. <https://arrow.apache.org/docs/python/api/compute.html>`__
+    Some methods may either raise an exception or raise a ``PerformanceWarning`` if an
+    associated compute function is not available based on the installed version of PyArrow.
+
+    Please install the latest version of PyArrow to enable the best functionality and avoid
+    potential bugs in prior versions of PyArrow.
+
+    Examples
+    --------
+    Create an ArrowExtensionArray with :func:`pandas.array`:
+
+    >>> pd.array([1, 1, None], dtype="int64[pyarrow]")
+    <ArrowExtensionArray>
+    [1, 1, <NA>]
+    Length: 3, dtype: int64[pyarrow]
+    """  # noqa: E501 (http link too long)
+
+    _pa_array: pa.ChunkedArray
+    _dtype: ArrowDtype
+
+    def __init__(self, values: pa.Array | pa.ChunkedArray) -> None:
+        if pa_version_under10p1:
+            msg = "pyarrow>=10.0.1 is required for PyArrow backed ArrowExtensionArray."
+            raise ImportError(msg)
+        if isinstance(values, pa.Array):
+            self._pa_array = pa.chunked_array([values])
+        elif isinstance(values, pa.ChunkedArray):
+            self._pa_array = values
+        else:
+            raise ValueError(
+                f"Unsupported type '{type(values)}' for ArrowExtensionArray"
+            )
+        self._dtype = ArrowDtype(self._pa_array.type)
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        """
+        Construct a new ExtensionArray from a sequence of scalars.
+        """
+        pa_type = to_pyarrow_type(dtype)
+        pa_array = cls._box_pa_array(scalars, pa_type=pa_type, copy=copy)
+        arr = cls(pa_array)
+        return arr
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        """
+        Construct a new ExtensionArray from a sequence of strings.
+        """
+        pa_type = to_pyarrow_type(dtype)
+        if (
+            pa_type is None
+            or pa.types.is_binary(pa_type)
+            or pa.types.is_string(pa_type)
+            or pa.types.is_large_string(pa_type)
+        ):
+            # pa_type is None: Let pa.array infer
+            # pa_type is string/binary: scalars already correct type
+            scalars = strings
+        elif pa.types.is_timestamp(pa_type):
+            from pandas.core.tools.datetimes import to_datetime
+
+            scalars = to_datetime(strings, errors="raise")
+        elif pa.types.is_date(pa_type):
+            from pandas.core.tools.datetimes import to_datetime
+
+            scalars = to_datetime(strings, errors="raise").date
+        elif pa.types.is_duration(pa_type):
+            from pandas.core.tools.timedeltas import to_timedelta
+
+            scalars = to_timedelta(strings, errors="raise")
+            if pa_type.unit != "ns":
+                # GH51175: test_from_sequence_of_strings_pa_array
+                # attempt to parse as int64 reflecting pyarrow's
+                # duration to string casting behavior
+                mask = isna(scalars)
+                if not isinstance(strings, (pa.Array, pa.ChunkedArray)):
+                    strings = pa.array(strings, type=pa.string(), from_pandas=True)
+                strings = pc.if_else(mask, None, strings)
+                try:
+                    scalars = strings.cast(pa.int64())
+                except pa.ArrowInvalid:
+                    pass
+        elif pa.types.is_time(pa_type):
+            from pandas.core.tools.times import to_time
+
+            # "coerce" to allow "null times" (None) to not raise
+            scalars = to_time(strings, errors="coerce")
+        elif pa.types.is_boolean(pa_type):
+            # pyarrow string->bool casting is case-insensitive:
+            #   "true" or "1" -> True
+            #   "false" or "0" -> False
+            # Note: BooleanArray was previously used to parse these strings
+            #   and allows "1.0" and "0.0". Pyarrow casting does not support
+            #   this, but we allow it here.
+            if isinstance(strings, (pa.Array, pa.ChunkedArray)):
+                scalars = strings
+            else:
+                scalars = pa.array(strings, type=pa.string(), from_pandas=True)
+            scalars = pc.if_else(pc.equal(scalars, "1.0"), "1", scalars)
+            scalars = pc.if_else(pc.equal(scalars, "0.0"), "0", scalars)
+            scalars = scalars.cast(pa.bool_())
+        elif (
+            pa.types.is_integer(pa_type)
+            or pa.types.is_floating(pa_type)
+            or pa.types.is_decimal(pa_type)
+        ):
+            from pandas.core.tools.numeric import to_numeric
+
+            scalars = to_numeric(strings, errors="raise")
+        else:
+            raise NotImplementedError(
+                f"Converting strings to {pa_type} is not implemented."
+            )
+        return cls._from_sequence(scalars, dtype=pa_type, copy=copy)
+
+    @classmethod
+    def _box_pa(
+        cls, value, pa_type: pa.DataType | None = None
+    ) -> pa.Array | pa.ChunkedArray | pa.Scalar:
+        """
+        Box value into a pyarrow Array, ChunkedArray or Scalar.
+
+        Parameters
+        ----------
+        value : any
+        pa_type : pa.DataType | None
+
+        Returns
+        -------
+        pa.Array or pa.ChunkedArray or pa.Scalar
+        """
+        if isinstance(value, pa.Scalar) or not is_list_like(value):
+            return cls._box_pa_scalar(value, pa_type)
+        return cls._box_pa_array(value, pa_type)
+
+    @classmethod
+    def _box_pa_scalar(cls, value, pa_type: pa.DataType | None = None) -> pa.Scalar:
+        """
+        Box value into a pyarrow Scalar.
+
+        Parameters
+        ----------
+        value : any
+        pa_type : pa.DataType | None
+
+        Returns
+        -------
+        pa.Scalar
+        """
+        if isinstance(value, pa.Scalar):
+            pa_scalar = value
+        elif isna(value):
+            pa_scalar = pa.scalar(None, type=pa_type)
+        else:
+            # Workaround https://github.com/apache/arrow/issues/37291
+            if isinstance(value, Timedelta):
+                if pa_type is None:
+                    pa_type = pa.duration(value.unit)
+                elif value.unit != pa_type.unit:
+                    value = value.as_unit(pa_type.unit)
+                value = value._value
+            elif isinstance(value, Timestamp):
+                if pa_type is None:
+                    pa_type = pa.timestamp(value.unit, tz=value.tz)
+                elif value.unit != pa_type.unit:
+                    value = value.as_unit(pa_type.unit)
+                value = value._value
+
+            pa_scalar = pa.scalar(value, type=pa_type, from_pandas=True)
+
+        if pa_type is not None and pa_scalar.type != pa_type:
+            pa_scalar = pa_scalar.cast(pa_type)
+
+        return pa_scalar
+
+    @classmethod
+    def _box_pa_array(
+        cls, value, pa_type: pa.DataType | None = None, copy: bool = False
+    ) -> pa.Array | pa.ChunkedArray:
+        """
+        Box value into a pyarrow Array or ChunkedArray.
+
+        Parameters
+        ----------
+        value : Sequence
+        pa_type : pa.DataType | None
+
+        Returns
+        -------
+        pa.Array or pa.ChunkedArray
+        """
+        if isinstance(value, cls):
+            pa_array = value._pa_array
+        elif isinstance(value, (pa.Array, pa.ChunkedArray)):
+            pa_array = value
+        elif isinstance(value, BaseMaskedArray):
+            # GH 52625
+            if copy:
+                value = value.copy()
+            pa_array = value.__arrow_array__()
+        else:
+            if (
+                isinstance(value, np.ndarray)
+                and pa_type is not None
+                and (
+                    pa.types.is_large_binary(pa_type)
+                    or pa.types.is_large_string(pa_type)
+                )
+            ):
+                # See https://github.com/apache/arrow/issues/35289
+                value = value.tolist()
+            elif copy and is_array_like(value):
+                # pa array should not get updated when numpy array is updated
+                value = value.copy()
+
+            if (
+                pa_type is not None
+                and pa.types.is_duration(pa_type)
+                and (not isinstance(value, np.ndarray) or value.dtype.kind not in "mi")
+            ):
+                # Workaround https://github.com/apache/arrow/issues/37291
+                from pandas.core.tools.timedeltas import to_timedelta
+
+                value = to_timedelta(value, unit=pa_type.unit).as_unit(pa_type.unit)
+                value = value.to_numpy()
+
+            try:
+                pa_array = pa.array(value, type=pa_type, from_pandas=True)
+            except (pa.ArrowInvalid, pa.ArrowTypeError):
+                # GH50430: let pyarrow infer type, then cast
+                pa_array = pa.array(value, from_pandas=True)
+
+            if pa_type is None and pa.types.is_duration(pa_array.type):
+                # Workaround https://github.com/apache/arrow/issues/37291
+                from pandas.core.tools.timedeltas import to_timedelta
+
+                value = to_timedelta(value)
+                value = value.to_numpy()
+                pa_array = pa.array(value, type=pa_type, from_pandas=True)
+
+            if pa.types.is_duration(pa_array.type) and pa_array.null_count > 0:
+                # GH52843: upstream bug for duration types when originally
+                # constructed with data containing numpy NaT.
+                # https://github.com/apache/arrow/issues/35088
+                arr = cls(pa_array)
+                arr = arr.fillna(arr.dtype.na_value)
+                pa_array = arr._pa_array
+
+        if pa_type is not None and pa_array.type != pa_type:
+            if pa.types.is_dictionary(pa_type):
+                pa_array = pa_array.dictionary_encode()
+                if pa_array.type != pa_type:
+                    pa_array = pa_array.cast(pa_type)
+            else:
+                try:
+                    pa_array = pa_array.cast(pa_type)
+                except (pa.ArrowNotImplementedError, pa.ArrowTypeError):
+                    if pa.types.is_string(pa_array.type) or pa.types.is_large_string(
+                        pa_array.type
+                    ):
+                        # TODO: Move logic in _from_sequence_of_strings into
+                        # _box_pa_array
+                        dtype = ArrowDtype(pa_type)
+                        return cls._from_sequence_of_strings(
+                            value, dtype=dtype
+                        )._pa_array
+                    else:
+                        raise
+
+        return pa_array
+
+    def __getitem__(self, item: PositionalIndexer):
+        """Select a subset of self.
+
+        Parameters
+        ----------
+        item : int, slice, or ndarray
+            * int: The position in 'self' to get.
+            * slice: A slice object, where 'start', 'stop', and 'step' are
+              integers or None
+            * ndarray: A 1-d boolean NumPy ndarray the same length as 'self'
+
+        Returns
+        -------
+        item : scalar or ExtensionArray
+
+        Notes
+        -----
+        For scalar ``item``, return a scalar value suitable for the array's
+        type. This should be an instance of ``self.dtype.type``.
+        For slice ``key``, return an instance of ``ExtensionArray``, even
+        if the slice is length 0 or 1.
+        For a boolean mask, return an instance of ``ExtensionArray``, filtered
+        to the values where ``item`` is True.
+        """
+        item = check_array_indexer(self, item)
+
+        if isinstance(item, np.ndarray):
+            if not len(item):
+                # Removable once we migrate StringDtype[pyarrow] to ArrowDtype[string]
+                if (
+                    isinstance(self._dtype, StringDtype)
+                    and self._dtype.storage == "pyarrow"
+                ):
+                    # TODO(infer_string) should this be large_string?
+                    pa_dtype = pa.string()
+                else:
+                    pa_dtype = self._dtype.pyarrow_dtype
+                return type(self)(pa.chunked_array([], type=pa_dtype))
+            elif item.dtype.kind in "iu":
+                return self.take(item)
+            elif item.dtype.kind == "b":
+                return type(self)(self._pa_array.filter(item))
+            else:
+                raise IndexError(
+                    "Only integers, slices and integer or "
+                    "boolean arrays are valid indices."
+                )
+        elif isinstance(item, tuple):
+            item = unpack_tuple_and_ellipses(item)
+
+        if item is Ellipsis:
+            # TODO: should be handled by pyarrow?
+            item = slice(None)
+
+        if is_scalar(item) and not is_integer(item):
+            # e.g. "foo" or 2.5
+            # exception message copied from numpy
+            raise IndexError(
+                r"only integers, slices (`:`), ellipsis (`...`), numpy.newaxis "
+                r"(`None`) and integer or boolean arrays are valid indices"
+            )
+        # We are not an array indexer, so maybe e.g. a slice or integer
+        # indexer. We dispatch to pyarrow.
+        if isinstance(item, slice):
+            # Arrow bug https://github.com/apache/arrow/issues/38768
+            if item.start == item.stop:
+                pass
+            elif (
+                item.stop is not None
+                and item.stop < -len(self)
+                and item.step is not None
+                and item.step < 0
+            ):
+                item = slice(item.start, None, item.step)
+
+        value = self._pa_array[item]
+        if isinstance(value, pa.ChunkedArray):
+            return type(self)(value)
+        else:
+            pa_type = self._pa_array.type
+            scalar = value.as_py()
+            if scalar is None:
+                return self._dtype.na_value
+            elif pa.types.is_timestamp(pa_type) and pa_type.unit != "ns":
+                # GH 53326
+                return Timestamp(scalar).as_unit(pa_type.unit)
+            elif pa.types.is_duration(pa_type) and pa_type.unit != "ns":
+                # GH 53326
+                return Timedelta(scalar).as_unit(pa_type.unit)
+            else:
+                return scalar
+
+    def __iter__(self) -> Iterator[Any]:
+        """
+        Iterate over elements of the array.
+        """
+        na_value = self._dtype.na_value
+        # GH 53326
+        pa_type = self._pa_array.type
+        box_timestamp = pa.types.is_timestamp(pa_type) and pa_type.unit != "ns"
+        box_timedelta = pa.types.is_duration(pa_type) and pa_type.unit != "ns"
+        for value in self._pa_array:
+            val = value.as_py()
+            if val is None:
+                yield na_value
+            elif box_timestamp:
+                yield Timestamp(val).as_unit(pa_type.unit)
+            elif box_timedelta:
+                yield Timedelta(val).as_unit(pa_type.unit)
+            else:
+                yield val
+
+    def __arrow_array__(self, type=None):
+        """Convert myself to a pyarrow ChunkedArray."""
+        return self._pa_array
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        """Correctly construct numpy arrays when passed to `np.asarray()`."""
+        if copy is False:
+            # TODO: By using `zero_copy_only` it may be possible to implement this
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+        elif copy is None:
+            # `to_numpy(copy=False)` has the meaning of NumPy `copy=None`.
+            copy = False
+
+        return self.to_numpy(dtype=dtype, copy=copy)
+
+    def __invert__(self) -> Self:
+        # This is a bit wise op for integer types
+        if pa.types.is_integer(self._pa_array.type):
+            return type(self)(pc.bit_wise_not(self._pa_array))
+        elif pa.types.is_string(self._pa_array.type) or pa.types.is_large_string(
+            self._pa_array.type
+        ):
+            # Raise TypeError instead of pa.ArrowNotImplementedError
+            raise TypeError("__invert__ is not supported for string dtypes")
+        else:
+            return type(self)(pc.invert(self._pa_array))
+
+    def __neg__(self) -> Self:
+        try:
+            return type(self)(pc.negate_checked(self._pa_array))
+        except pa.ArrowNotImplementedError as err:
+            raise TypeError(
+                f"unary '-' not supported for dtype '{self.dtype}'"
+            ) from err
+
+    def __pos__(self) -> Self:
+        return type(self)(self._pa_array)
+
+    def __abs__(self) -> Self:
+        return type(self)(pc.abs_checked(self._pa_array))
+
+    # GH 42600: __getstate__/__setstate__ not necessary once
+    # https://issues.apache.org/jira/browse/ARROW-10739 is addressed
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["_pa_array"] = self._pa_array.combine_chunks()
+        return state
+
+    def __setstate__(self, state) -> None:
+        if "_data" in state:
+            data = state.pop("_data")
+        else:
+            data = state["_pa_array"]
+        state["_pa_array"] = pa.chunked_array(data)
+        self.__dict__.update(state)
+
+    def _cmp_method(self, other, op) -> ArrowExtensionArray:
+        pc_func = ARROW_CMP_FUNCS[op.__name__]
+        if isinstance(
+            other, (ArrowExtensionArray, np.ndarray, list, BaseMaskedArray)
+        ) or isinstance(getattr(other, "dtype", None), CategoricalDtype):
+            try:
+                result = pc_func(self._pa_array, self._box_pa(other))
+            except pa.ArrowNotImplementedError:
+                # TODO: could this be wrong if other is object dtype?
+                #  in which case we need to operate pointwise?
+                result = ops.invalid_comparison(self, other, op)
+                result = pa.array(result, type=pa.bool_())
+        elif is_scalar(other):
+            try:
+                result = pc_func(self._pa_array, self._box_pa(other))
+            except (pa.lib.ArrowNotImplementedError, pa.lib.ArrowInvalid):
+                mask = isna(self) | isna(other)
+                valid = ~mask
+                result = np.zeros(len(self), dtype="bool")
+                np_array = np.array(self)
+                try:
+                    result[valid] = op(np_array[valid], other)
+                except TypeError:
+                    result = ops.invalid_comparison(self, other, op)
+                result = pa.array(result, type=pa.bool_())
+                result = pc.if_else(valid, result, None)
+        else:
+            raise NotImplementedError(
+                f"{op.__name__} not implemented for {type(other)}"
+            )
+        return ArrowExtensionArray(result)
+
+    def _op_method_error_message(self, other, op) -> str:
+        if hasattr(other, "dtype"):
+            other_type = f"dtype '{other.dtype}'"
+        else:
+            other_type = f"object of type {type(other)}"
+        return (
+            f"operation '{op.__name__}' not supported for "
+            f"dtype '{self.dtype}' with {other_type}"
+        )
+
+    def _evaluate_op_method(self, other, op, arrow_funcs) -> Self:
+        pa_type = self._pa_array.type
+        other_original = other
+        other = self._box_pa(other)
+
+        if (
+            pa.types.is_string(pa_type)
+            or pa.types.is_large_string(pa_type)
+            or pa.types.is_binary(pa_type)
+        ):
+            if op in [operator.add, roperator.radd]:
+                sep = pa.scalar("", type=pa_type)
+                try:
+                    if op is operator.add:
+                        result = pc.binary_join_element_wise(self._pa_array, other, sep)
+                    elif op is roperator.radd:
+                        result = pc.binary_join_element_wise(other, self._pa_array, sep)
+                except pa.ArrowNotImplementedError as err:
+                    raise TypeError(
+                        self._op_method_error_message(other_original, op)
+                    ) from err
+                return type(self)(result)
+            elif op in [operator.mul, roperator.rmul]:
+                binary = self._pa_array
+                integral = other
+                if not pa.types.is_integer(integral.type):
+                    raise TypeError("Can only string multiply by an integer.")
+                pa_integral = pc.if_else(pc.less(integral, 0), 0, integral)
+                result = pc.binary_repeat(binary, pa_integral)
+                return type(self)(result)
+        elif (
+            pa.types.is_string(other.type)
+            or pa.types.is_binary(other.type)
+            or pa.types.is_large_string(other.type)
+        ) and op in [operator.mul, roperator.rmul]:
+            binary = other
+            integral = self._pa_array
+            if not pa.types.is_integer(integral.type):
+                raise TypeError("Can only string multiply by an integer.")
+            pa_integral = pc.if_else(pc.less(integral, 0), 0, integral)
+            result = pc.binary_repeat(binary, pa_integral)
+            return type(self)(result)
+        if (
+            isinstance(other, pa.Scalar)
+            and pc.is_null(other).as_py()
+            and op.__name__ in ARROW_LOGICAL_FUNCS
+        ):
+            # pyarrow kleene ops require null to be typed
+            other = other.cast(pa_type)
+
+        pc_func = arrow_funcs[op.__name__]
+        if pc_func is NotImplemented:
+            if pa.types.is_string(pa_type) or pa.types.is_large_string(pa_type):
+                raise TypeError(self._op_method_error_message(other_original, op))
+            raise NotImplementedError(f"{op.__name__} not implemented.")
+
+        try:
+            result = pc_func(self._pa_array, other)
+        except pa.ArrowNotImplementedError as err:
+            raise TypeError(self._op_method_error_message(other_original, op)) from err
+        return type(self)(result)
+
+    def _logical_method(self, other, op) -> Self:
+        # For integer types `^`, `|`, `&` are bitwise operators and return
+        # integer types. Otherwise these are boolean ops.
+        if pa.types.is_integer(self._pa_array.type):
+            return self._evaluate_op_method(other, op, ARROW_BIT_WISE_FUNCS)
+        else:
+            return self._evaluate_op_method(other, op, ARROW_LOGICAL_FUNCS)
+
+    def _arith_method(self, other, op) -> Self:
+        return self._evaluate_op_method(other, op, ARROW_ARITHMETIC_FUNCS)
+
+    def equals(self, other) -> bool:
+        if not isinstance(other, ArrowExtensionArray):
+            return False
+        # I'm told that pyarrow makes __eq__ behave like pandas' equals;
+        #  TODO: is this documented somewhere?
+        return self._pa_array == other._pa_array
+
+    @property
+    def dtype(self) -> ArrowDtype:
+        """
+        An instance of 'ExtensionDtype'.
+        """
+        return self._dtype
+
+    @property
+    def nbytes(self) -> int:
+        """
+        The number of bytes needed to store this object in memory.
+        """
+        return self._pa_array.nbytes
+
+    def __len__(self) -> int:
+        """
+        Length of this array.
+
+        Returns
+        -------
+        length : int
+        """
+        return len(self._pa_array)
+
+    def __contains__(self, key) -> bool:
+        # https://github.com/pandas-dev/pandas/pull/51307#issuecomment-1426372604
+        if isna(key) and key is not self.dtype.na_value:
+            if self.dtype.kind == "f" and lib.is_float(key):
+                return pc.any(pc.is_nan(self._pa_array)).as_py()
+
+            # e.g. date or timestamp types we do not allow None here to match pd.NA
+            return False
+            # TODO: maybe complex? object?
+
+        return bool(super().__contains__(key))
+
+    @property
+    def _hasna(self) -> bool:
+        return self._pa_array.null_count > 0
+
+    def isna(self) -> npt.NDArray[np.bool_]:
+        """
+        Boolean NumPy array indicating if each value is missing.
+
+        This should return a 1-D array the same length as 'self'.
+        """
+        # GH51630: fast paths
+        null_count = self._pa_array.null_count
+        if null_count == 0:
+            return np.zeros(len(self), dtype=np.bool_)
+        elif null_count == len(self):
+            return np.ones(len(self), dtype=np.bool_)
+
+        return self._pa_array.is_null().to_numpy()
+
+    @overload
+    def any(self, *, skipna: Literal[True] = ..., **kwargs) -> bool: ...
+
+    @overload
+    def any(self, *, skipna: bool, **kwargs) -> bool | NAType: ...
+
+    def any(self, *, skipna: bool = True, **kwargs) -> bool | NAType:
+        """
+        Return whether any element is truthy.
+
+        Returns False unless there is at least one element that is truthy.
+        By default, NAs are skipped. If ``skipna=False`` is specified and
+        missing values are present, similar :ref:`Kleene logic <boolean.kleene>`
+        is used as for logical operations.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values. If the entire array is NA and `skipna` is
+            True, then the result will be False, as for an empty array.
+            If `skipna` is False, the result will still be True if there is
+            at least one element that is truthy, otherwise NA will be returned
+            if there are NA's present.
+
+        Returns
+        -------
+        bool or :attr:`pandas.NA`
+
+        See Also
+        --------
+        ArrowExtensionArray.all : Return whether all elements are truthy.
+
+        Examples
+        --------
+        The result indicates whether any element is truthy (and by default
+        skips NAs):
+
+        >>> pd.array([True, False, True], dtype="boolean[pyarrow]").any()
+        True
+        >>> pd.array([True, False, pd.NA], dtype="boolean[pyarrow]").any()
+        True
+        >>> pd.array([False, False, pd.NA], dtype="boolean[pyarrow]").any()
+        False
+        >>> pd.array([], dtype="boolean[pyarrow]").any()
+        False
+        >>> pd.array([pd.NA], dtype="boolean[pyarrow]").any()
+        False
+        >>> pd.array([pd.NA], dtype="float64[pyarrow]").any()
+        False
+
+        With ``skipna=False``, the result can be NA if this is logically
+        required (whether ``pd.NA`` is True or False influences the result):
+
+        >>> pd.array([True, False, pd.NA], dtype="boolean[pyarrow]").any(skipna=False)
+        True
+        >>> pd.array([1, 0, pd.NA], dtype="boolean[pyarrow]").any(skipna=False)
+        True
+        >>> pd.array([False, False, pd.NA], dtype="boolean[pyarrow]").any(skipna=False)
+        <NA>
+        >>> pd.array([0, 0, pd.NA], dtype="boolean[pyarrow]").any(skipna=False)
+        <NA>
+        """
+        return self._reduce("any", skipna=skipna, **kwargs)
+
+    @overload
+    def all(self, *, skipna: Literal[True] = ..., **kwargs) -> bool: ...
+
+    @overload
+    def all(self, *, skipna: bool, **kwargs) -> bool | NAType: ...
+
+    def all(self, *, skipna: bool = True, **kwargs) -> bool | NAType:
+        """
+        Return whether all elements are truthy.
+
+        Returns True unless there is at least one element that is falsey.
+        By default, NAs are skipped. If ``skipna=False`` is specified and
+        missing values are present, similar :ref:`Kleene logic <boolean.kleene>`
+        is used as for logical operations.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values. If the entire array is NA and `skipna` is
+            True, then the result will be True, as for an empty array.
+            If `skipna` is False, the result will still be False if there is
+            at least one element that is falsey, otherwise NA will be returned
+            if there are NA's present.
+
+        Returns
+        -------
+        bool or :attr:`pandas.NA`
+
+        See Also
+        --------
+        ArrowExtensionArray.any : Return whether any element is truthy.
+
+        Examples
+        --------
+        The result indicates whether all elements are truthy (and by default
+        skips NAs):
+
+        >>> pd.array([True, True, pd.NA], dtype="boolean[pyarrow]").all()
+        True
+        >>> pd.array([1, 1, pd.NA], dtype="boolean[pyarrow]").all()
+        True
+        >>> pd.array([True, False, pd.NA], dtype="boolean[pyarrow]").all()
+        False
+        >>> pd.array([], dtype="boolean[pyarrow]").all()
+        True
+        >>> pd.array([pd.NA], dtype="boolean[pyarrow]").all()
+        True
+        >>> pd.array([pd.NA], dtype="float64[pyarrow]").all()
+        True
+
+        With ``skipna=False``, the result can be NA if this is logically
+        required (whether ``pd.NA`` is True or False influences the result):
+
+        >>> pd.array([True, True, pd.NA], dtype="boolean[pyarrow]").all(skipna=False)
+        <NA>
+        >>> pd.array([1, 1, pd.NA], dtype="boolean[pyarrow]").all(skipna=False)
+        <NA>
+        >>> pd.array([True, False, pd.NA], dtype="boolean[pyarrow]").all(skipna=False)
+        False
+        >>> pd.array([1, 0, pd.NA], dtype="boolean[pyarrow]").all(skipna=False)
+        False
+        """
+        return self._reduce("all", skipna=skipna, **kwargs)
+
+    def argsort(
+        self,
+        *,
+        ascending: bool = True,
+        kind: SortKind = "quicksort",
+        na_position: str = "last",
+        **kwargs,
+    ) -> np.ndarray:
+        order = "ascending" if ascending else "descending"
+        null_placement = {"last": "at_end", "first": "at_start"}.get(na_position, None)
+        if null_placement is None:
+            raise ValueError(f"invalid na_position: {na_position}")
+
+        result = pc.array_sort_indices(
+            self._pa_array, order=order, null_placement=null_placement
+        )
+        np_result = result.to_numpy()
+        return np_result.astype(np.intp, copy=False)
+
+    def _argmin_max(self, skipna: bool, method: str) -> int:
+        if self._pa_array.length() in (0, self._pa_array.null_count) or (
+            self._hasna and not skipna
+        ):
+            # For empty or all null, pyarrow returns -1 but pandas expects TypeError
+            # For skipna=False and data w/ null, pandas expects NotImplementedError
+            # let ExtensionArray.arg{max|min} raise
+            return getattr(super(), f"arg{method}")(skipna=skipna)
+
+        data = self._pa_array
+        if pa.types.is_duration(data.type):
+            data = data.cast(pa.int64())
+
+        value = getattr(pc, method)(data, skip_nulls=skipna)
+        return pc.index(data, value).as_py()
+
+    def argmin(self, skipna: bool = True) -> int:
+        return self._argmin_max(skipna, "min")
+
+    def argmax(self, skipna: bool = True) -> int:
+        return self._argmin_max(skipna, "max")
+
+    def copy(self) -> Self:
+        """
+        Return a shallow copy of the array.
+
+        Underlying ChunkedArray is immutable, so a deep copy is unnecessary.
+
+        Returns
+        -------
+        type(self)
+        """
+        return type(self)(self._pa_array)
+
+    def dropna(self) -> Self:
+        """
+        Return ArrowExtensionArray without NA values.
+
+        Returns
+        -------
+        ArrowExtensionArray
+        """
+        return type(self)(pc.drop_null(self._pa_array))
+
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        if not self._hasna:
+            return self
+
+        if limit is None and limit_area is None:
+            method = missing.clean_fill_method(method)
+            try:
+                if method == "pad":
+                    return type(self)(pc.fill_null_forward(self._pa_array))
+                elif method == "backfill":
+                    return type(self)(pc.fill_null_backward(self._pa_array))
+            except pa.ArrowNotImplementedError:
+                # ArrowNotImplementedError: Function 'coalesce' has no kernel
+                #   matching input types (duration[ns], duration[ns])
+                # TODO: remove try/except wrapper if/when pyarrow implements
+                #   a kernel for duration types.
+                pass
+
+        # TODO: Why do we no longer need the above cases?
+        # TODO(3.0): after EA.fillna 'method' deprecation is enforced, we can remove
+        #  this method entirely.
+        return super()._pad_or_backfill(
+            method=method, limit=limit, limit_area=limit_area, copy=copy
+        )
+
+    @doc(ExtensionArray.fillna)
+    def fillna(
+        self,
+        value: object | ArrayLike,
+        limit: int | None = None,
+        copy: bool = True,
+    ) -> Self:
+        if not self._hasna:
+            return self.copy()
+
+        if limit is not None:
+            return super().fillna(value=value, limit=limit, copy=copy)
+
+        if isinstance(value, (np.ndarray, ExtensionArray)):
+            # Similar to check_value_size, but we do not mask here since we may
+            #  end up passing it to the super() method.
+            if len(value) != len(self):
+                raise ValueError(
+                    f"Length of 'value' does not match. Got ({len(value)}) "
+                    f" expected {len(self)}"
+                )
+
+        try:
+            fill_value = self._box_pa(value, pa_type=self._pa_array.type)
+        except pa.ArrowTypeError as err:
+            msg = f"Invalid value '{value!s}' for dtype '{self.dtype}'"
+            raise TypeError(msg) from err
+
+        try:
+            return type(self)(pc.fill_null(self._pa_array, fill_value=fill_value))
+        except pa.ArrowNotImplementedError:
+            # ArrowNotImplementedError: Function 'coalesce' has no kernel
+            #   matching input types (duration[ns], duration[ns])
+            # TODO: remove try/except wrapper if/when pyarrow implements
+            #   a kernel for duration types.
+            pass
+
+        return super().fillna(value=value, limit=limit, copy=copy)
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        # short-circuit to return all False array.
+        if not len(values):
+            return np.zeros(len(self), dtype=bool)
+
+        result = pc.is_in(self._pa_array, value_set=pa.array(values, from_pandas=True))
+        # pyarrow 2.0.0 returned nulls, so we explicitly specify dtype to convert nulls
+        # to False
+        return np.array(result, dtype=np.bool_)
+
+    def _values_for_factorize(self) -> tuple[np.ndarray, Any]:
+        """
+        Return an array and missing value suitable for factorization.
+
+        Returns
+        -------
+        values : ndarray
+        na_value : pd.NA
+
+        Notes
+        -----
+        The values returned by this method are also used in
+        :func:`pandas.util.hash_pandas_object`.
+        """
+        values = self._pa_array.to_numpy()
+        return values, self.dtype.na_value
+
+    @doc(ExtensionArray.factorize)
+    def factorize(
+        self,
+        use_na_sentinel: bool = True,
+    ) -> tuple[np.ndarray, ExtensionArray]:
+        null_encoding = "mask" if use_na_sentinel else "encode"
+
+        data = self._pa_array
+        pa_type = data.type
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            # https://github.com/apache/arrow/issues/15226#issuecomment-1376578323
+            data = data.cast(pa.int64())
+
+        if pa.types.is_dictionary(data.type):
+            if null_encoding == "encode":
+                # dictionary encode does nothing if an already encoded array is given
+                data = data.cast(data.type.value_type)
+                encoded = data.dictionary_encode(null_encoding=null_encoding)
+            else:
+                encoded = data
+        else:
+            encoded = data.dictionary_encode(null_encoding=null_encoding)
+        if encoded.length() == 0:
+            indices = np.array([], dtype=np.intp)
+            uniques = type(self)(pa.chunked_array([], type=encoded.type.value_type))
+        else:
+            # GH 54844
+            combined = encoded.combine_chunks()
+            pa_indices = combined.indices
+            if pa_indices.null_count > 0:
+                pa_indices = pc.fill_null(pa_indices, -1)
+            indices = pa_indices.to_numpy(zero_copy_only=False, writable=True).astype(
+                np.intp, copy=False
+            )
+            uniques = type(self)(combined.dictionary)
+
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            uniques = cast(ArrowExtensionArray, uniques.astype(self.dtype))
+        return indices, uniques
+
+    def reshape(self, *args, **kwargs):
+        raise NotImplementedError(
+            f"{type(self)} does not support reshape "
+            f"as backed by a 1D pyarrow.ChunkedArray."
+        )
+
+    def round(self, decimals: int = 0, *args, **kwargs) -> Self:
+        """
+        Round each value in the array a to the given number of decimals.
+
+        Parameters
+        ----------
+        decimals : int, default 0
+            Number of decimal places to round to. If decimals is negative,
+            it specifies the number of positions to the left of the decimal point.
+        *args, **kwargs
+            Additional arguments and keywords have no effect.
+
+        Returns
+        -------
+        ArrowExtensionArray
+            Rounded values of the ArrowExtensionArray.
+
+        See Also
+        --------
+        DataFrame.round : Round values of a DataFrame.
+        Series.round : Round values of a Series.
+        """
+        return type(self)(pc.round(self._pa_array, ndigits=decimals))
+
+    @doc(ExtensionArray.searchsorted)
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if self._hasna:
+            raise ValueError(
+                "searchsorted requires array to be sorted, which is impossible "
+                "with NAs present."
+            )
+        if isinstance(value, ExtensionArray):
+            value = value.astype(object)
+        # Base class searchsorted would cast to object, which is *much* slower.
+        dtype = None
+        if isinstance(self.dtype, ArrowDtype):
+            pa_dtype = self.dtype.pyarrow_dtype
+            if (
+                pa.types.is_timestamp(pa_dtype) or pa.types.is_duration(pa_dtype)
+            ) and pa_dtype.unit == "ns":
+                # np.array[datetime/timedelta].searchsorted(datetime/timedelta)
+                # erroneously fails when numpy type resolution is nanoseconds
+                dtype = object
+        return self.to_numpy(dtype=dtype).searchsorted(value, side=side, sorter=sorter)
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        allow_fill: bool = False,
+        fill_value: Any = None,
+    ) -> ArrowExtensionArray:
+        """
+        Take elements from an array.
+
+        Parameters
+        ----------
+        indices : sequence of int or one-dimensional np.ndarray of int
+            Indices to be taken.
+        allow_fill : bool, default False
+            How to handle negative values in `indices`.
+
+            * False: negative values in `indices` indicate positional indices
+              from the right (the default). This is similar to
+              :func:`numpy.take`.
+
+            * True: negative values in `indices` indicate
+              missing values. These values are set to `fill_value`. Any other
+              other negative values raise a ``ValueError``.
+
+        fill_value : any, optional
+            Fill value to use for NA-indices when `allow_fill` is True.
+            This may be ``None``, in which case the default NA value for
+            the type, ``self.dtype.na_value``, is used.
+
+            For many ExtensionArrays, there will be two representations of
+            `fill_value`: a user-facing "boxed" scalar, and a low-level
+            physical NA value. `fill_value` should be the user-facing version,
+            and the implementation should handle translating that to the
+            physical version for processing the take if necessary.
+
+        Returns
+        -------
+        ExtensionArray
+
+        Raises
+        ------
+        IndexError
+            When the indices are out of bounds for the array.
+        ValueError
+            When `indices` contains negative values other than ``-1``
+            and `allow_fill` is True.
+
+        See Also
+        --------
+        numpy.take
+        api.extensions.take
+
+        Notes
+        -----
+        ExtensionArray.take is called by ``Series.__getitem__``, ``.loc``,
+        ``iloc``, when `indices` is a sequence of values. Additionally,
+        it's called by :meth:`Series.reindex`, or any other method
+        that causes realignment, with a `fill_value`.
+        """
+        indices_array = np.asanyarray(indices)
+
+        if len(self._pa_array) == 0 and (indices_array >= 0).any():
+            raise IndexError("cannot do a non-empty take")
+        if indices_array.size > 0 and indices_array.max() >= len(self._pa_array):
+            raise IndexError("out of bounds value in 'indices'.")
+
+        if allow_fill:
+            fill_mask = indices_array < 0
+            if fill_mask.any():
+                validate_indices(indices_array, len(self._pa_array))
+                # TODO(ARROW-9433): Treat negative indices as NULL
+                indices_array = pa.array(indices_array, mask=fill_mask)
+                result = self._pa_array.take(indices_array)
+                if isna(fill_value):
+                    return type(self)(result)
+                # TODO: ArrowNotImplementedError: Function fill_null has no
+                # kernel matching input types (array[string], scalar[string])
+                result = type(self)(result)
+                result[fill_mask] = fill_value
+                return result
+                # return type(self)(pc.fill_null(result, pa.scalar(fill_value)))
+            else:
+                # Nothing to fill
+                return type(self)(self._pa_array.take(indices))
+        else:  # allow_fill=False
+            # TODO(ARROW-9432): Treat negative indices as indices from the right.
+            if (indices_array < 0).any():
+                # Don't modify in-place
+                indices_array = np.copy(indices_array)
+                indices_array[indices_array < 0] += len(self._pa_array)
+            return type(self)(self._pa_array.take(indices_array))
+
+    def _maybe_convert_datelike_array(self):
+        """Maybe convert to a datelike array."""
+        pa_type = self._pa_array.type
+        if pa.types.is_timestamp(pa_type):
+            return self._to_datetimearray()
+        elif pa.types.is_duration(pa_type):
+            return self._to_timedeltaarray()
+        return self
+
+    def _to_datetimearray(self) -> DatetimeArray:
+        """Convert a pyarrow timestamp typed array to a DatetimeArray."""
+        from pandas.core.arrays.datetimes import (
+            DatetimeArray,
+            tz_to_dtype,
+        )
+
+        pa_type = self._pa_array.type
+        assert pa.types.is_timestamp(pa_type)
+        np_dtype = np.dtype(f"M8[{pa_type.unit}]")
+        dtype = tz_to_dtype(pa_type.tz, pa_type.unit)
+        np_array = self._pa_array.to_numpy()
+        np_array = np_array.astype(np_dtype, copy=False)
+        return DatetimeArray._simple_new(np_array, dtype=dtype)
+
+    def _to_timedeltaarray(self) -> TimedeltaArray:
+        """Convert a pyarrow duration typed array to a TimedeltaArray."""
+        from pandas.core.arrays.timedeltas import TimedeltaArray
+
+        pa_type = self._pa_array.type
+        assert pa.types.is_duration(pa_type)
+        np_dtype = np.dtype(f"m8[{pa_type.unit}]")
+        np_array = self._pa_array.to_numpy()
+        np_array = np_array.astype(np_dtype, copy=False)
+        return TimedeltaArray._simple_new(np_array, dtype=np_dtype)
+
+    def _values_for_json(self) -> np.ndarray:
+        if is_numeric_dtype(self.dtype):
+            return np.asarray(self, dtype=object)
+        return super()._values_for_json()
+
+    @doc(ExtensionArray.to_numpy)
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        original_na_value = na_value
+        dtype, na_value = to_numpy_dtype_inference(self, dtype, na_value, self._hasna)
+        pa_type = self._pa_array.type
+        if not self._hasna or isna(na_value) or pa.types.is_null(pa_type):
+            data = self
+        else:
+            data = self.fillna(na_value)
+            copy = False
+
+        if pa.types.is_timestamp(pa_type) or pa.types.is_duration(pa_type):
+            # GH 55997
+            if dtype != object and na_value is self.dtype.na_value:
+                na_value = lib.no_default
+            result = data._maybe_convert_datelike_array().to_numpy(
+                dtype=dtype, na_value=na_value
+            )
+        elif pa.types.is_time(pa_type) or pa.types.is_date(pa_type):
+            # convert to list of python datetime.time objects before
+            # wrapping in ndarray
+            result = np.array(list(data), dtype=dtype)
+            if data._hasna:
+                result[data.isna()] = na_value
+        elif pa.types.is_null(pa_type):
+            if dtype is not None and isna(na_value):
+                na_value = None
+            result = np.full(len(data), fill_value=na_value, dtype=dtype)
+        elif not data._hasna or (
+            pa.types.is_floating(pa_type)
+            and (
+                na_value is np.nan
+                or (original_na_value is lib.no_default and is_float_dtype(dtype))
+            )
+        ):
+            result = data._pa_array.to_numpy()
+            if dtype is not None:
+                result = result.astype(dtype, copy=False)
+            if copy:
+                result = result.copy()
+        else:
+            if dtype is None:
+                empty = pa.array([], type=pa_type).to_numpy(zero_copy_only=False)
+                if can_hold_element(empty, na_value):
+                    dtype = empty.dtype
+                else:
+                    dtype = np.object_
+            result = np.empty(len(data), dtype=dtype)
+            mask = data.isna()
+            result[mask] = na_value
+            result[~mask] = data[~mask]._pa_array.to_numpy()
+        return result
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        if is_numeric_dtype(self.dtype):
+            return map_array(self.to_numpy(), mapper, na_action=na_action)
+        else:
+            return super().map(mapper, na_action)
+
+    @doc(ExtensionArray.duplicated)
+    def duplicated(
+        self, keep: Literal["first", "last", False] = "first"
+    ) -> npt.NDArray[np.bool_]:
+        pa_type = self._pa_array.type
+        if pa.types.is_floating(pa_type) or pa.types.is_integer(pa_type):
+            values = self.to_numpy(na_value=0)
+        elif pa.types.is_boolean(pa_type):
+            values = self.to_numpy(na_value=False)
+        elif pa.types.is_temporal(pa_type):
+            if pa_type.bit_width == 32:
+                pa_type = pa.int32()
+            else:
+                pa_type = pa.int64()
+            arr = self.astype(ArrowDtype(pa_type))
+            values = arr.to_numpy(na_value=0)
+        else:
+            # factorize the values to avoid the performance penalty of
+            # converting to object dtype
+            values = self.factorize()[0]
+
+        mask = self.isna() if self._hasna else None
+        return algos.duplicated(values, keep=keep, mask=mask)
+
+    def unique(self) -> Self:
+        """
+        Compute the ArrowExtensionArray of unique values.
+
+        Returns
+        -------
+        ArrowExtensionArray
+        """
+        pa_type = self._pa_array.type
+
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            # https://github.com/apache/arrow/issues/15226#issuecomment-1376578323
+            data = self._pa_array.cast(pa.int64())
+        else:
+            data = self._pa_array
+
+        pa_result = pc.unique(data)
+
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            pa_result = pa_result.cast(pa_type)
+
+        return type(self)(pa_result)
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Return a Series containing counts of each unique value.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of missing values.
+
+        Returns
+        -------
+        counts : Series
+
+        See Also
+        --------
+        Series.value_counts
+        """
+        pa_type = self._pa_array.type
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            # https://github.com/apache/arrow/issues/15226#issuecomment-1376578323
+            data = self._pa_array.cast(pa.int64())
+        else:
+            data = self._pa_array
+
+        from pandas import (
+            Index,
+            Series,
+        )
+
+        vc = data.value_counts()
+
+        values = vc.field(0)
+        counts = vc.field(1)
+        if dropna and data.null_count > 0:
+            mask = values.is_valid()
+            values = values.filter(mask)
+            counts = counts.filter(mask)
+
+        if pa_version_under11p0 and pa.types.is_duration(pa_type):
+            values = values.cast(pa_type)
+
+        counts = ArrowExtensionArray(counts)
+
+        index = Index(type(self)(values))
+
+        return Series(counts, index=index, name="count", copy=False)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat) -> Self:
+        """
+        Concatenate multiple ArrowExtensionArrays.
+
+        Parameters
+        ----------
+        to_concat : sequence of ArrowExtensionArrays
+
+        Returns
+        -------
+        ArrowExtensionArray
+        """
+        chunks = [array for ea in to_concat for array in ea._pa_array.iterchunks()]
+        if to_concat[0].dtype == "string":
+            # StringDtype has no attribute pyarrow_dtype
+            pa_dtype = pa.large_string()
+        else:
+            pa_dtype = to_concat[0].dtype.pyarrow_dtype
+        arr = pa.chunked_array(chunks, type=pa_dtype)
+        return cls(arr)
+
+    def _accumulate(
+        self, name: str, *, skipna: bool = True, **kwargs
+    ) -> ArrowExtensionArray | ExtensionArray:
+        """
+        Return an ExtensionArray performing an accumulation operation.
+
+        The underlying data type might change.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            - cummin
+            - cummax
+            - cumsum
+            - cumprod
+        skipna : bool, default True
+            If True, skip NA values.
+        **kwargs
+            Additional keyword arguments passed to the accumulation function.
+            Currently, there is no supported kwarg.
+
+        Returns
+        -------
+        array
+
+        Raises
+        ------
+        NotImplementedError : subclass does not define accumulations
+        """
+        if is_string_dtype(self):
+            return self._str_accumulate(name=name, skipna=skipna, **kwargs)
+
+        pyarrow_name = {
+            "cummax": "cumulative_max",
+            "cummin": "cumulative_min",
+            "cumprod": "cumulative_prod_checked",
+            "cumsum": "cumulative_sum_checked",
+        }.get(name, name)
+        pyarrow_meth = getattr(pc, pyarrow_name, None)
+        if pyarrow_meth is None:
+            return super()._accumulate(name, skipna=skipna, **kwargs)
+
+        data_to_accum = self._pa_array
+
+        pa_dtype = data_to_accum.type
+
+        convert_to_int = (
+            pa.types.is_temporal(pa_dtype) and name in ["cummax", "cummin"]
+        ) or (pa.types.is_duration(pa_dtype) and name == "cumsum")
+
+        if convert_to_int:
+            if pa_dtype.bit_width == 32:
+                data_to_accum = data_to_accum.cast(pa.int32())
+            else:
+                data_to_accum = data_to_accum.cast(pa.int64())
+
+        try:
+            result = pyarrow_meth(data_to_accum, skip_nulls=skipna, **kwargs)
+        except pa.ArrowNotImplementedError as err:
+            msg = f"operation '{name}' not supported for dtype '{self.dtype}'"
+            raise TypeError(msg) from err
+
+        if convert_to_int:
+            result = result.cast(pa_dtype)
+
+        return type(self)(result)
+
+    def _str_accumulate(
+        self, name: str, *, skipna: bool = True, **kwargs
+    ) -> ArrowExtensionArray | ExtensionArray:
+        """
+        Accumulate implementation for strings, see `_accumulate` docstring for details.
+
+        pyarrow.compute does not implement these methods for strings.
+        """
+        if name == "cumprod":
+            msg = f"operation '{name}' not supported for dtype '{self.dtype}'"
+            raise TypeError(msg)
+
+        # We may need to strip out trailing NA values
+        tail: pa.array | None = None
+        na_mask: pa.array | None = None
+        pa_array = self._pa_array
+        np_func = {
+            "cumsum": np.cumsum,
+            "cummin": np.minimum.accumulate,
+            "cummax": np.maximum.accumulate,
+        }[name]
+
+        if self._hasna:
+            na_mask = pc.is_null(pa_array)
+            if pc.all(na_mask) == pa.scalar(True):
+                return type(self)(pa_array)
+            if skipna:
+                if name == "cumsum":
+                    pa_array = pc.fill_null(pa_array, "")
+                else:
+                    # We can retain the running min/max by forward/backward filling.
+                    pa_array = pc.fill_null_forward(pa_array)
+                    pa_array = pc.fill_null_backward(pa_array)
+            else:
+                # When not skipping NA values, the result should be null from
+                # the first NA value onward.
+                idx = pc.index(na_mask, True).as_py()
+                tail = pa.nulls(len(pa_array) - idx, type=pa_array.type)
+                pa_array = pa_array[:idx]
+
+        # error: Cannot call function of unknown type
+        pa_result = pa.array(np_func(pa_array), type=pa_array.type)  # type: ignore[operator]
+
+        if tail is not None:
+            pa_result = pa.concat_arrays([pa_result, tail])
+        elif na_mask is not None:
+            pa_result = pc.if_else(na_mask, None, pa_result)
+
+        result = type(self)(pa_result)
+        return result
+
+    def _reduce_pyarrow(self, name: str, *, skipna: bool = True, **kwargs) -> pa.Scalar:
+        """
+        Return a pyarrow scalar result of performing the reduction operation.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            { any, all, min, max, sum, mean, median, prod,
+            std, var, sem, kurt, skew }.
+        skipna : bool, default True
+            If True, skip NaN values.
+        **kwargs
+            Additional keyword arguments passed to the reduction function.
+            Currently, `ddof` is the only supported kwarg.
+
+        Returns
+        -------
+        pyarrow scalar
+
+        Raises
+        ------
+        TypeError : subclass does not define reductions
+        """
+        pa_type = self._pa_array.type
+
+        data_to_reduce = self._pa_array
+
+        cast_kwargs = {} if pa_version_under13p0 else {"safe": False}
+
+        if name in ["any", "all"] and (
+            pa.types.is_integer(pa_type)
+            or pa.types.is_floating(pa_type)
+            or pa.types.is_duration(pa_type)
+            or pa.types.is_decimal(pa_type)
+        ):
+            # pyarrow only supports any/all for boolean dtype, we allow
+            #  for other dtypes, matching our non-pyarrow behavior
+
+            if pa.types.is_duration(pa_type):
+                data_to_cmp = self._pa_array.cast(pa.int64())
+            else:
+                data_to_cmp = self._pa_array
+
+            not_eq = pc.not_equal(data_to_cmp, 0)
+            data_to_reduce = not_eq
+
+        elif name in ["min", "max", "sum"] and pa.types.is_duration(pa_type):
+            data_to_reduce = self._pa_array.cast(pa.int64())
+
+        elif name in ["median", "mean", "std", "sem"] and pa.types.is_temporal(pa_type):
+            nbits = pa_type.bit_width
+            if nbits == 32:
+                data_to_reduce = self._pa_array.cast(pa.int32())
+            else:
+                data_to_reduce = self._pa_array.cast(pa.int64())
+
+        if name == "sem":
+
+            def pyarrow_meth(data, skip_nulls, **kwargs):
+                numerator = pc.stddev(data, skip_nulls=skip_nulls, **kwargs)
+                denominator = pc.sqrt_checked(pc.count(self._pa_array))
+                return pc.divide_checked(numerator, denominator)
+
+        elif name == "sum" and (
+            pa.types.is_string(pa_type) or pa.types.is_large_string(pa_type)
+        ):
+
+            def pyarrow_meth(data, skip_nulls, min_count=0):  # type: ignore[misc]
+                mask = pc.is_null(data) if data.null_count > 0 else None
+                if skip_nulls:
+                    if min_count > 0 and check_below_min_count(
+                        (len(data),),
+                        None if mask is None else mask.to_numpy(),
+                        min_count,
+                    ):
+                        return pa.scalar(None, type=data.type)
+                    if data.null_count > 0:
+                        # binary_join returns null if there is any null ->
+                        # have to filter out any nulls
+                        data = data.filter(pc.invert(mask))
+                else:
+                    if mask is not None or check_below_min_count(
+                        (len(data),), None, min_count
+                    ):
+                        return pa.scalar(None, type=data.type)
+
+                if pa.types.is_large_string(data.type):
+                    # binary_join only supports string, not large_string
+                    data = data.cast(pa.string())
+                data_list = pa.ListArray.from_arrays(
+                    [0, len(data)], data.combine_chunks()
+                )[0]
+                return pc.binary_join(data_list, "")
+
+        else:
+            pyarrow_name = {
+                "median": "quantile",
+                "prod": "product",
+                "std": "stddev",
+                "var": "variance",
+            }.get(name, name)
+            # error: Incompatible types in assignment
+            # (expression has type "Optional[Any]", variable has type
+            # "Callable[[Any, Any, KwArg(Any)], Any]")
+            pyarrow_meth = getattr(pc, pyarrow_name, None)  # type: ignore[assignment]
+            if pyarrow_meth is None:
+                # Let ExtensionArray._reduce raise the TypeError
+                return super()._reduce(name, skipna=skipna, **kwargs)
+
+        # GH51624: pyarrow defaults to min_count=1, pandas behavior is min_count=0
+        if name in ["any", "all"] and "min_count" not in kwargs:
+            kwargs["min_count"] = 0
+        elif name == "median":
+            # GH 52679: Use quantile instead of approximate_median
+            kwargs["q"] = 0.5
+
+        try:
+            result = pyarrow_meth(data_to_reduce, skip_nulls=skipna, **kwargs)
+        except (AttributeError, NotImplementedError, TypeError) as err:
+            msg = (
+                f"'{type(self).__name__}' with dtype {self.dtype} "
+                f"does not support operation '{name}' with pyarrow "
+                f"version {pa.__version__}. '{name}' may be supported by "
+                f"upgrading pyarrow."
+            )
+            raise TypeError(msg) from err
+        if name == "median":
+            # GH 52679: Use quantile instead of approximate_median; returns array
+            result = result[0]
+
+        if name in ["min", "max", "sum"] and pa.types.is_duration(pa_type):
+            result = result.cast(pa_type)
+        if name in ["median", "mean"] and pa.types.is_temporal(pa_type):
+            if not pa_version_under13p0:
+                nbits = pa_type.bit_width
+                if nbits == 32:
+                    result = result.cast(pa.int32(), **cast_kwargs)
+                else:
+                    result = result.cast(pa.int64(), **cast_kwargs)
+            result = result.cast(pa_type)
+        if name in ["std", "sem"] and pa.types.is_temporal(pa_type):
+            result = result.cast(pa.int64(), **cast_kwargs)
+            if pa.types.is_duration(pa_type):
+                result = result.cast(pa_type)
+            elif pa.types.is_time(pa_type):
+                unit = get_unit_from_pa_dtype(pa_type)
+                result = result.cast(pa.duration(unit))
+            elif pa.types.is_date(pa_type):
+                # go with closest available unit, i.e. "s"
+                result = result.cast(pa.duration("s"))
+            else:
+                # i.e. timestamp
+                result = result.cast(pa.duration(pa_type.unit))
+
+        return result
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        """
+        Return a scalar result of performing the reduction operation.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            { any, all, min, max, sum, mean, median, prod,
+            std, var, sem, kurt, skew }.
+        skipna : bool, default True
+            If True, skip NaN values.
+        **kwargs
+            Additional keyword arguments passed to the reduction function.
+            Currently, `ddof` is the only supported kwarg.
+
+        Returns
+        -------
+        scalar
+
+        Raises
+        ------
+        TypeError : subclass does not define reductions
+        """
+        result = self._reduce_calc(name, skipna=skipna, keepdims=keepdims, **kwargs)
+        if isinstance(result, pa.Array):
+            return type(self)(result)
+        else:
+            return result
+
+    def _reduce_calc(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        pa_result = self._reduce_pyarrow(name, skipna=skipna, **kwargs)
+
+        if keepdims:
+            if isinstance(pa_result, pa.Scalar):
+                result = pa.array([pa_result.as_py()], type=pa_result.type)
+            else:
+                result = pa.array(
+                    [pa_result],
+                    type=to_pyarrow_type(infer_dtype_from_scalar(pa_result)[0]),
+                )
+            return result
+
+        if pc.is_null(pa_result).as_py():
+            return self.dtype.na_value
+        elif isinstance(pa_result, pa.Scalar):
+            return pa_result.as_py()
+        else:
+            return pa_result
+
+    def _explode(self):
+        """
+        See Series.explode.__doc__.
+        """
+        # child class explode method supports only list types; return
+        # default implementation for non list types.
+        if not pa.types.is_list(self.dtype.pyarrow_dtype):
+            return super()._explode()
+        values = self
+        counts = pa.compute.list_value_length(values._pa_array)
+        counts = counts.fill_null(1).to_numpy()
+        fill_value = pa.scalar([None], type=self._pa_array.type)
+        mask = counts == 0
+        if mask.any():
+            values = values.copy()
+            values[mask] = fill_value
+            counts = counts.copy()
+            counts[mask] = 1
+        values = values.fillna(fill_value)
+        values = type(self)(pa.compute.list_flatten(values._pa_array))
+        return values, counts
+
+    def __setitem__(self, key, value) -> None:
+        """Set one or more values inplace.
+
+        Parameters
+        ----------
+        key : int, ndarray, or slice
+            When called from, e.g. ``Series.__setitem__``, ``key`` will be
+            one of
+
+            * scalar int
+            * ndarray of integers.
+            * boolean ndarray
+            * slice object
+
+        value : ExtensionDtype.type, Sequence[ExtensionDtype.type], or object
+            value or values to be set of ``key``.
+
+        Returns
+        -------
+        None
+        """
+        # GH50085: unwrap 1D indexers
+        if isinstance(key, tuple) and len(key) == 1:
+            key = key[0]
+
+        key = check_array_indexer(self, key)
+        value = self._maybe_convert_setitem_value(value)
+
+        if com.is_null_slice(key):
+            # fast path (GH50248)
+            data = self._if_else(True, value, self._pa_array)
+
+        elif is_integer(key):
+            # fast path
+            key = cast(int, key)
+            n = len(self)
+            if key < 0:
+                key += n
+            if not 0 <= key < n:
+                raise IndexError(
+                    f"index {key} is out of bounds for axis 0 with size {n}"
+                )
+            if isinstance(value, pa.Scalar):
+                value = value.as_py()
+            elif is_list_like(value):
+                raise ValueError("Length of indexer and values mismatch")
+            chunks = [
+                *self._pa_array[:key].chunks,
+                pa.array([value], type=self._pa_array.type, from_pandas=True),
+                *self._pa_array[key + 1 :].chunks,
+            ]
+            data = pa.chunked_array(chunks).combine_chunks()
+
+        elif is_bool_dtype(key):
+            key = np.asarray(key, dtype=np.bool_)
+            data = self._replace_with_mask(self._pa_array, key, value)
+
+        elif is_scalar(value) or isinstance(value, pa.Scalar):
+            mask = np.zeros(len(self), dtype=np.bool_)
+            mask[key] = True
+            data = self._if_else(mask, value, self._pa_array)
+
+        else:
+            indices = np.arange(len(self))[key]
+            if len(indices) != len(value):
+                raise ValueError("Length of indexer and values mismatch")
+            if len(indices) == 0:
+                return
+            # GH#58530 wrong item assignment by repeated key
+            _, argsort = np.unique(indices, return_index=True)
+            indices = indices[argsort]
+            value = value.take(argsort)
+            mask = np.zeros(len(self), dtype=np.bool_)
+            mask[indices] = True
+            data = self._replace_with_mask(self._pa_array, mask, value)
+
+        if isinstance(data, pa.Array):
+            data = pa.chunked_array([data])
+        self._pa_array = data
+
+    def _rank_calc(
+        self,
+        *,
+        axis: AxisInt = 0,
+        method: str = "average",
+        na_option: str = "keep",
+        ascending: bool = True,
+        pct: bool = False,
+    ):
+        if axis != 0:
+            ranked = super()._rank(
+                axis=axis,
+                method=method,
+                na_option=na_option,
+                ascending=ascending,
+                pct=pct,
+            )
+            # keep dtypes consistent with the implementation below
+            if method == "average" or pct:
+                pa_type = pa.float64()
+            else:
+                pa_type = pa.uint64()
+            result = pa.array(ranked, type=pa_type, from_pandas=True)
+            return result
+
+        data = self._pa_array.combine_chunks()
+        sort_keys = "ascending" if ascending else "descending"
+        null_placement = "at_start" if na_option == "top" else "at_end"
+        tiebreaker = "min" if method == "average" else method
+
+        result = pc.rank(
+            data,
+            sort_keys=sort_keys,
+            null_placement=null_placement,
+            tiebreaker=tiebreaker,
+        )
+
+        if na_option == "keep":
+            mask = pc.is_null(self._pa_array)
+            null = pa.scalar(None, type=result.type)
+            result = pc.if_else(mask, null, result)
+
+        if method == "average":
+            result_max = pc.rank(
+                data,
+                sort_keys=sort_keys,
+                null_placement=null_placement,
+                tiebreaker="max",
+            )
+            result_max = result_max.cast(pa.float64())
+            result_min = result.cast(pa.float64())
+            result = pc.divide(pc.add(result_min, result_max), 2)
+
+        if pct:
+            if not pa.types.is_floating(result.type):
+                result = result.cast(pa.float64())
+            if method == "dense":
+                divisor = pc.max(result)
+            else:
+                divisor = pc.count(result)
+            result = pc.divide(result, divisor)
+
+        return result
+
+    def _rank(
+        self,
+        *,
+        axis: AxisInt = 0,
+        method: str = "average",
+        na_option: str = "keep",
+        ascending: bool = True,
+        pct: bool = False,
+    ) -> Self:
+        """
+        See Series.rank.__doc__.
+        """
+        return self._convert_rank_result(
+            self._rank_calc(
+                axis=axis,
+                method=method,
+                na_option=na_option,
+                ascending=ascending,
+                pct=pct,
+            )
+        )
+
+    def _quantile(self, qs: npt.NDArray[np.float64], interpolation: str) -> Self:
+        """
+        Compute the quantiles of self for each quantile in `qs`.
+
+        Parameters
+        ----------
+        qs : np.ndarray[float64]
+        interpolation: str
+
+        Returns
+        -------
+        same type as self
+        """
+        pa_dtype = self._pa_array.type
+
+        data = self._pa_array
+        if pa.types.is_temporal(pa_dtype):
+            # https://github.com/apache/arrow/issues/33769 in these cases
+            #  we can cast to ints and back
+            nbits = pa_dtype.bit_width
+            if nbits == 32:
+                data = data.cast(pa.int32())
+            else:
+                data = data.cast(pa.int64())
+
+        result = pc.quantile(data, q=qs, interpolation=interpolation)
+
+        if pa.types.is_temporal(pa_dtype):
+            if pa.types.is_floating(result.type):
+                result = pc.floor(result)
+            nbits = pa_dtype.bit_width
+            if nbits == 32:
+                result = result.cast(pa.int32())
+            else:
+                result = result.cast(pa.int64())
+            result = result.cast(pa_dtype)
+
+        return type(self)(result)
+
+    def _mode(self, dropna: bool = True) -> Self:
+        """
+        Returns the mode(s) of the ExtensionArray.
+
+        Always returns `ExtensionArray` even if only one value.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't consider counts of NA values.
+
+        Returns
+        -------
+        same type as self
+            Sorted, if possible.
+        """
+        pa_type = self._pa_array.type
+        if pa.types.is_temporal(pa_type):
+            nbits = pa_type.bit_width
+            if nbits == 32:
+                data = self._pa_array.cast(pa.int32())
+            elif nbits == 64:
+                data = self._pa_array.cast(pa.int64())
+            else:
+                raise NotImplementedError(pa_type)
+        else:
+            data = self._pa_array
+
+        if dropna:
+            data = data.drop_null()
+
+        res = pc.value_counts(data)
+        most_common = res.field("values").filter(
+            pc.equal(res.field("counts"), pc.max(res.field("counts")))
+        )
+
+        if pa.types.is_temporal(pa_type):
+            most_common = most_common.cast(pa_type)
+
+        most_common = most_common.take(pc.array_sort_indices(most_common))
+        return type(self)(most_common)
+
+    def _maybe_convert_setitem_value(self, value):
+        """Maybe convert value to be pyarrow compatible."""
+        try:
+            value = self._box_pa(value, self._pa_array.type)
+        except pa.ArrowTypeError as err:
+            msg = f"Invalid value '{value!s}' for dtype '{self.dtype}'"
+            raise TypeError(msg) from err
+        return value
+
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        axis: int,
+        index,
+        limit,
+        limit_direction,
+        limit_area,
+        copy: bool,
+        **kwargs,
+    ) -> Self:
+        """
+        See NDFrame.interpolate.__doc__.
+        """
+        # NB: we return type(self) even if copy=False
+        if not self.dtype._is_numeric:
+            raise TypeError(f"Cannot interpolate with {self.dtype} dtype")
+
+        if (
+            not pa_version_under13p0
+            and method == "linear"
+            and limit_area is None
+            and limit is None
+            and limit_direction == "forward"
+        ):
+            values = self._pa_array.combine_chunks()
+            na_value = pa.array([None], type=values.type)
+            y_diff_2 = pc.fill_null_backward(pc.pairwise_diff_checked(values, period=2))
+            prev_values = pa.concat_arrays([na_value, values[:-2], na_value])
+            interps = pc.add_checked(prev_values, pc.divide_checked(y_diff_2, 2))
+            return type(self)(pc.coalesce(self._pa_array, interps))
+
+        mask = self.isna()
+        if self.dtype.kind == "f":
+            data = self._pa_array.to_numpy()
+        elif self.dtype.kind in "iu":
+            data = self.to_numpy(dtype="f8", na_value=0.0)
+        else:
+            raise NotImplementedError(
+                f"interpolate is not implemented for dtype={self.dtype}"
+            )
+
+        missing.interpolate_2d_inplace(
+            data,
+            method=method,
+            axis=0,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            mask=mask,
+            **kwargs,
+        )
+        return type(self)(self._box_pa_array(pa.array(data, mask=mask)))
+
+    @classmethod
+    def _if_else(
+        cls,
+        cond: npt.NDArray[np.bool_] | bool,
+        left: ArrayLike | Scalar,
+        right: ArrayLike | Scalar,
+    ) -> pa.Array:
+        """
+        Choose values based on a condition.
+
+        Analogous to pyarrow.compute.if_else, with logic
+        to fallback to numpy for unsupported types.
+
+        Parameters
+        ----------
+        cond : npt.NDArray[np.bool_] or bool
+        left : ArrayLike | Scalar
+        right : ArrayLike | Scalar
+
+        Returns
+        -------
+        pa.Array
+        """
+        try:
+            return pc.if_else(cond, left, right)
+        except pa.ArrowNotImplementedError:
+            pass
+
+        def _to_numpy_and_type(value) -> tuple[np.ndarray, pa.DataType | None]:
+            if isinstance(value, (pa.Array, pa.ChunkedArray)):
+                pa_type = value.type
+            elif isinstance(value, pa.Scalar):
+                pa_type = value.type
+                value = value.as_py()
+            else:
+                pa_type = None
+            return np.array(value, dtype=object), pa_type
+
+        left, left_type = _to_numpy_and_type(left)
+        right, right_type = _to_numpy_and_type(right)
+        pa_type = left_type or right_type
+        result = np.where(cond, left, right)
+        return pa.array(result, type=pa_type, from_pandas=True)
+
+    @classmethod
+    def _replace_with_mask(
+        cls,
+        values: pa.Array | pa.ChunkedArray,
+        mask: npt.NDArray[np.bool_] | bool,
+        replacements: ArrayLike | Scalar,
+    ) -> pa.Array | pa.ChunkedArray:
+        """
+        Replace items selected with a mask.
+
+        Analogous to pyarrow.compute.replace_with_mask, with logic
+        to fallback to numpy for unsupported types.
+
+        Parameters
+        ----------
+        values : pa.Array or pa.ChunkedArray
+        mask : npt.NDArray[np.bool_] or bool
+        replacements : ArrayLike or Scalar
+            Replacement value(s)
+
+        Returns
+        -------
+        pa.Array or pa.ChunkedArray
+        """
+        if isinstance(replacements, pa.ChunkedArray):
+            # replacements must be array or scalar, not ChunkedArray
+            replacements = replacements.combine_chunks()
+        if isinstance(values, pa.ChunkedArray) and pa.types.is_boolean(values.type):
+            # GH#52059 replace_with_mask segfaults for chunked array
+            # https://github.com/apache/arrow/issues/34634
+            values = values.combine_chunks()
+        try:
+            return pc.replace_with_mask(values, mask, replacements)
+        except pa.ArrowNotImplementedError:
+            pass
+        if isinstance(replacements, pa.Array):
+            replacements = np.array(replacements, dtype=object)
+        elif isinstance(replacements, pa.Scalar):
+            replacements = replacements.as_py()
+        result = np.array(values, dtype=object)
+        result[mask] = replacements
+        return pa.array(result, type=values.type, from_pandas=True)
+
+    # ------------------------------------------------------------------
+    # GroupBy Methods
+
+    def _to_masked(self):
+        pa_dtype = self._pa_array.type
+
+        if pa.types.is_floating(pa_dtype) or pa.types.is_integer(pa_dtype):
+            na_value = 1
+        elif pa.types.is_boolean(pa_dtype):
+            na_value = True
+        else:
+            raise NotImplementedError
+
+        dtype = _arrow_dtype_mapping()[pa_dtype]
+        mask = self.isna()
+        arr = self.to_numpy(dtype=dtype.numpy_dtype, na_value=na_value)
+        return dtype.construct_array_type()(arr, mask)
+
+    def _groupby_op(
+        self,
+        *,
+        how: str,
+        has_dropped_na: bool,
+        min_count: int,
+        ngroups: int,
+        ids: npt.NDArray[np.intp],
+        **kwargs,
+    ):
+        if isinstance(self.dtype, StringDtype):
+            if how in [
+                "prod",
+                "mean",
+                "median",
+                "cumsum",
+                "cumprod",
+                "std",
+                "sem",
+                "var",
+                "skew",
+            ]:
+                raise TypeError(
+                    f"dtype '{self.dtype}' does not support operation '{how}'"
+                )
+            return super()._groupby_op(
+                how=how,
+                has_dropped_na=has_dropped_na,
+                min_count=min_count,
+                ngroups=ngroups,
+                ids=ids,
+                **kwargs,
+            )
+
+        # maybe convert to a compatible dtype optimized for groupby
+        values: ExtensionArray
+        pa_type = self._pa_array.type
+        if pa.types.is_timestamp(pa_type):
+            values = self._to_datetimearray()
+        elif pa.types.is_duration(pa_type):
+            values = self._to_timedeltaarray()
+        else:
+            values = self._to_masked()
+
+        result = values._groupby_op(
+            how=how,
+            has_dropped_na=has_dropped_na,
+            min_count=min_count,
+            ngroups=ngroups,
+            ids=ids,
+            **kwargs,
+        )
+        if isinstance(result, np.ndarray):
+            return result
+        elif isinstance(result, BaseMaskedArray):
+            pa_result = result.__arrow_array__()
+            return type(self)(pa_result)
+        else:
+            # DatetimeArray, TimedeltaArray
+            pa_result = pa.array(result, from_pandas=True)
+            return type(self)(pa_result)
+
+    def _apply_elementwise(self, func: Callable) -> list[list[Any]]:
+        """Apply a callable to each element while maintaining the chunking structure."""
+        return [
+            [
+                None if val is None else func(val)
+                for val in chunk.to_numpy(zero_copy_only=False)
+            ]
+            for chunk in self._pa_array.iterchunks()
+        ]
+
+    def _convert_bool_result(self, result, na=lib.no_default, method_name=None):
+        if na is not lib.no_default and not isna(na):  # pyright: ignore [reportGeneralTypeIssues]
+            result = result.fill_null(na)
+        return type(self)(result)
+
+    def _convert_int_result(self, result):
+        return type(self)(result)
+
+    def _convert_rank_result(self, result):
+        return type(self)(result)
+
+    def _str_count(self, pat: str, flags: int = 0) -> Self:
+        if flags:
+            raise NotImplementedError(f"count not implemented with {flags=}")
+        return type(self)(pc.count_substring_regex(self._pa_array, pat))
+
+    def _str_repeat(self, repeats: int | Sequence[int]) -> Self:
+        if not isinstance(repeats, int):
+            raise NotImplementedError(
+                f"repeat is not implemented when repeats is {type(repeats).__name__}"
+            )
+        return type(self)(pc.binary_repeat(self._pa_array, repeats))
+
+    def _str_join(self, sep: str) -> Self:
+        if pa.types.is_string(self._pa_array.type) or pa.types.is_large_string(
+            self._pa_array.type
+        ):
+            result = self._apply_elementwise(list)
+            result = pa.chunked_array(result, type=pa.list_(pa.string()))
+        else:
+            result = self._pa_array
+        return type(self)(pc.binary_join(result, sep))
+
+    def _str_partition(self, sep: str, expand: bool) -> Self:
+        predicate = lambda val: val.partition(sep)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_rpartition(self, sep: str, expand: bool) -> Self:
+        predicate = lambda val: val.rpartition(sep)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_casefold(self) -> Self:
+        predicate = lambda val: val.casefold()
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_encode(self, encoding: str, errors: str = "strict") -> Self:
+        predicate = lambda val: val.encode(encoding, errors)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_extract(self, pat: str, flags: int = 0, expand: bool = True):
+        if flags:
+            raise NotImplementedError("Only flags=0 is implemented.")
+        groups = re.compile(pat).groupindex.keys()
+        if len(groups) == 0:
+            raise ValueError(f"{pat=} must contain a symbolic group name.")
+        result = pc.extract_regex(self._pa_array, pat)
+        if expand:
+            return {
+                col: type(self)(pc.struct_field(result, [i]))
+                for col, i in zip(groups, range(result.type.num_fields))
+            }
+        else:
+            return type(self)(pc.struct_field(result, [0]))
+
+    def _str_findall(self, pat: str, flags: int = 0) -> Self:
+        regex = re.compile(pat, flags=flags)
+        predicate = lambda val: regex.findall(val)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
+        if dtype is None:
+            dtype = np.bool_
+        split = pc.split_pattern(self._pa_array, sep)
+        flattened_values = pc.list_flatten(split)
+        uniques = flattened_values.unique()
+        uniques_sorted = uniques.take(pa.compute.array_sort_indices(uniques))
+        lengths = pc.list_value_length(split).fill_null(0).to_numpy()
+        n_rows = len(self)
+        n_cols = len(uniques)
+        indices = pc.index_in(flattened_values, uniques_sorted).to_numpy()
+        indices = indices + np.arange(n_rows).repeat(lengths) * n_cols
+        _dtype = pandas_dtype(dtype)
+        dummies_dtype: NpDtype
+        if isinstance(_dtype, np.dtype):
+            dummies_dtype = _dtype
+        else:
+            dummies_dtype = np.bool_
+        dummies = np.zeros(n_rows * n_cols, dtype=dummies_dtype)
+        dummies[indices] = True
+        dummies = dummies.reshape((n_rows, n_cols))
+        result = type(self)(pa.array(list(dummies)))
+        return result, uniques_sorted.to_pylist()
+
+    def _str_index(self, sub: str, start: int = 0, end: int | None = None) -> Self:
+        predicate = lambda val: val.index(sub, start, end)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_rindex(self, sub: str, start: int = 0, end: int | None = None) -> Self:
+        predicate = lambda val: val.rindex(sub, start, end)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_normalize(self, form: Literal["NFC", "NFD", "NFKC", "NFKD"]) -> Self:
+        predicate = lambda val: unicodedata.normalize(form, val)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_rfind(self, sub: str, start: int = 0, end=None) -> Self:
+        predicate = lambda val: val.rfind(sub, start, end)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_split(
+        self,
+        pat: str | None = None,
+        n: int | None = -1,
+        expand: bool = False,
+        regex: bool | None = None,
+    ) -> Self:
+        if n in {-1, 0}:
+            n = None
+        if pat is None:
+            split_func = pc.utf8_split_whitespace
+        elif regex:
+            split_func = functools.partial(pc.split_pattern_regex, pattern=pat)
+        else:
+            split_func = functools.partial(pc.split_pattern, pattern=pat)
+        return type(self)(split_func(self._pa_array, max_splits=n))
+
+    def _str_rsplit(self, pat: str | None = None, n: int | None = -1) -> Self:
+        if n in {-1, 0}:
+            n = None
+        if pat is None:
+            return type(self)(
+                pc.utf8_split_whitespace(self._pa_array, max_splits=n, reverse=True)
+            )
+        return type(self)(
+            pc.split_pattern(self._pa_array, pat, max_splits=n, reverse=True)
+        )
+
+    def _str_translate(self, table: dict[int, str]) -> Self:
+        predicate = lambda val: val.translate(table)
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    def _str_wrap(self, width: int, **kwargs) -> Self:
+        kwargs["width"] = width
+        tw = textwrap.TextWrapper(**kwargs)
+        predicate = lambda val: "\n".join(tw.wrap(val))
+        result = self._apply_elementwise(predicate)
+        return type(self)(pa.chunked_array(result))
+
+    @property
+    def _dt_days(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.days,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_hours(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.hours,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_minutes(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.minutes,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_seconds(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.seconds,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_milliseconds(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.milliseconds,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_microseconds(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.microseconds,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    @property
+    def _dt_nanoseconds(self) -> Self:
+        return type(self)(
+            pa.array(
+                self._to_timedeltaarray().components.nanoseconds,
+                from_pandas=True,
+                type=pa.int32(),
+            )
+        )
+
+    def _dt_to_pytimedelta(self) -> np.ndarray:
+        data = self._pa_array.to_pylist()
+        if self._dtype.pyarrow_dtype.unit == "ns":
+            data = [None if ts is None else ts.to_pytimedelta() for ts in data]
+        return np.array(data, dtype=object)
+
+    def _dt_total_seconds(self) -> Self:
+        return type(self)(
+            pa.array(self._to_timedeltaarray().total_seconds(), from_pandas=True)
+        )
+
+    def _dt_as_unit(self, unit: str) -> Self:
+        if pa.types.is_date(self.dtype.pyarrow_dtype):
+            raise NotImplementedError("as_unit not implemented for date types")
+        pd_array = self._maybe_convert_datelike_array()
+        # Don't just cast _pa_array in order to follow pandas unit conversion rules
+        return type(self)(pa.array(pd_array.as_unit(unit), from_pandas=True))
+
+    @property
+    def _dt_year(self) -> Self:
+        return type(self)(pc.year(self._pa_array))
+
+    @property
+    def _dt_day(self) -> Self:
+        return type(self)(pc.day(self._pa_array))
+
+    @property
+    def _dt_day_of_week(self) -> Self:
+        return type(self)(pc.day_of_week(self._pa_array))
+
+    _dt_dayofweek = _dt_day_of_week
+    _dt_weekday = _dt_day_of_week
+
+    @property
+    def _dt_day_of_year(self) -> Self:
+        return type(self)(pc.day_of_year(self._pa_array))
+
+    _dt_dayofyear = _dt_day_of_year
+
+    @property
+    def _dt_hour(self) -> Self:
+        return type(self)(pc.hour(self._pa_array))
+
+    def _dt_isocalendar(self) -> Self:
+        return type(self)(pc.iso_calendar(self._pa_array))
+
+    @property
+    def _dt_is_leap_year(self) -> Self:
+        return type(self)(pc.is_leap_year(self._pa_array))
+
+    @property
+    def _dt_is_month_start(self) -> Self:
+        return type(self)(pc.equal(pc.day(self._pa_array), 1))
+
+    @property
+    def _dt_is_month_end(self) -> Self:
+        result = pc.equal(
+            pc.days_between(
+                pc.floor_temporal(self._pa_array, unit="day"),
+                pc.ceil_temporal(self._pa_array, unit="month"),
+            ),
+            1,
+        )
+        return type(self)(result)
+
+    @property
+    def _dt_is_year_start(self) -> Self:
+        return type(self)(
+            pc.and_(
+                pc.equal(pc.month(self._pa_array), 1),
+                pc.equal(pc.day(self._pa_array), 1),
+            )
+        )
+
+    @property
+    def _dt_is_year_end(self) -> Self:
+        return type(self)(
+            pc.and_(
+                pc.equal(pc.month(self._pa_array), 12),
+                pc.equal(pc.day(self._pa_array), 31),
+            )
+        )
+
+    @property
+    def _dt_is_quarter_start(self) -> Self:
+        result = pc.equal(
+            pc.floor_temporal(self._pa_array, unit="quarter"),
+            pc.floor_temporal(self._pa_array, unit="day"),
+        )
+        return type(self)(result)
+
+    @property
+    def _dt_is_quarter_end(self) -> Self:
+        result = pc.equal(
+            pc.days_between(
+                pc.floor_temporal(self._pa_array, unit="day"),
+                pc.ceil_temporal(self._pa_array, unit="quarter"),
+            ),
+            1,
+        )
+        return type(self)(result)
+
+    @property
+    def _dt_days_in_month(self) -> Self:
+        result = pc.days_between(
+            pc.floor_temporal(self._pa_array, unit="month"),
+            pc.ceil_temporal(self._pa_array, unit="month"),
+        )
+        return type(self)(result)
+
+    _dt_daysinmonth = _dt_days_in_month
+
+    @property
+    def _dt_microsecond(self) -> Self:
+        # GH 59154
+        us = pc.microsecond(self._pa_array)
+        ms_to_us = pc.multiply(pc.millisecond(self._pa_array), 1000)
+        return type(self)(pc.add(us, ms_to_us))
+
+    @property
+    def _dt_minute(self) -> Self:
+        return type(self)(pc.minute(self._pa_array))
+
+    @property
+    def _dt_month(self) -> Self:
+        return type(self)(pc.month(self._pa_array))
+
+    @property
+    def _dt_nanosecond(self) -> Self:
+        return type(self)(pc.nanosecond(self._pa_array))
+
+    @property
+    def _dt_quarter(self) -> Self:
+        return type(self)(pc.quarter(self._pa_array))
+
+    @property
+    def _dt_second(self) -> Self:
+        return type(self)(pc.second(self._pa_array))
+
+    @property
+    def _dt_date(self) -> Self:
+        return type(self)(self._pa_array.cast(pa.date32()))
+
+    @property
+    def _dt_time(self) -> Self:
+        unit = (
+            self.dtype.pyarrow_dtype.unit
+            if self.dtype.pyarrow_dtype.unit in {"us", "ns"}
+            else "ns"
+        )
+        return type(self)(self._pa_array.cast(pa.time64(unit)))
+
+    @property
+    def _dt_tz(self):
+        return timezones.maybe_get_tz(self.dtype.pyarrow_dtype.tz)
+
+    @property
+    def _dt_unit(self):
+        return self.dtype.pyarrow_dtype.unit
+
+    def _dt_normalize(self) -> Self:
+        return type(self)(pc.floor_temporal(self._pa_array, 1, "day"))
+
+    def _dt_strftime(self, format: str) -> Self:
+        return type(self)(pc.strftime(self._pa_array, format=format))
+
+    def _round_temporally(
+        self,
+        method: Literal["ceil", "floor", "round"],
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        if ambiguous != "raise":
+            raise NotImplementedError("ambiguous is not supported.")
+        if nonexistent != "raise":
+            raise NotImplementedError("nonexistent is not supported.")
+        offset = to_offset(freq)
+        if offset is None:
+            raise ValueError(f"Must specify a valid frequency: {freq}")
+        pa_supported_unit = {
+            "Y": "year",
+            "YS": "year",
+            "Q": "quarter",
+            "QS": "quarter",
+            "M": "month",
+            "MS": "month",
+            "W": "week",
+            "D": "day",
+            "h": "hour",
+            "min": "minute",
+            "s": "second",
+            "ms": "millisecond",
+            "us": "microsecond",
+            "ns": "nanosecond",
+        }
+        unit = pa_supported_unit.get(offset._prefix, None)
+        if unit is None:
+            raise ValueError(f"{freq=} is not supported")
+        multiple = offset.n
+        rounding_method = getattr(pc, f"{method}_temporal")
+        return type(self)(rounding_method(self._pa_array, multiple=multiple, unit=unit))
+
+    def _dt_ceil(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round_temporally("ceil", freq, ambiguous, nonexistent)
+
+    def _dt_floor(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round_temporally("floor", freq, ambiguous, nonexistent)
+
+    def _dt_round(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round_temporally("round", freq, ambiguous, nonexistent)
+
+    def _dt_day_name(self, locale: str | None = None) -> Self:
+        if locale is None:
+            locale = "C"
+        return type(self)(pc.strftime(self._pa_array, format="%A", locale=locale))
+
+    def _dt_month_name(self, locale: str | None = None) -> Self:
+        if locale is None:
+            locale = "C"
+        return type(self)(pc.strftime(self._pa_array, format="%B", locale=locale))
+
+    def _dt_to_pydatetime(self) -> Series:
+        from pandas import Series
+
+        if pa.types.is_date(self.dtype.pyarrow_dtype):
+            raise ValueError(
+                f"to_pydatetime cannot be called with {self.dtype.pyarrow_dtype} type. "
+                "Convert to pyarrow timestamp type."
+            )
+        data = self._pa_array.to_pylist()
+        if self._dtype.pyarrow_dtype.unit == "ns":
+            data = [None if ts is None else ts.to_pydatetime(warn=False) for ts in data]
+        return Series(data, dtype=object)
+
+    def _dt_tz_localize(
+        self,
+        tz,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        if ambiguous != "raise":
+            raise NotImplementedError(f"{ambiguous=} is not supported")
+        nonexistent_pa = {
+            "raise": "raise",
+            "shift_backward": "earliest",
+            "shift_forward": "latest",
+        }.get(
+            nonexistent,  # type: ignore[arg-type]
+            None,
+        )
+        if nonexistent_pa is None:
+            raise NotImplementedError(f"{nonexistent=} is not supported")
+        if tz is None:
+            result = self._pa_array.cast(pa.timestamp(self.dtype.pyarrow_dtype.unit))
+        else:
+            result = pc.assume_timezone(
+                self._pa_array, str(tz), ambiguous=ambiguous, nonexistent=nonexistent_pa
+            )
+        return type(self)(result)
+
+    def _dt_tz_convert(self, tz) -> Self:
+        if self.dtype.pyarrow_dtype.tz is None:
+            raise TypeError(
+                "Cannot convert tz-naive timestamps, use tz_localize to localize"
+            )
+        current_unit = self.dtype.pyarrow_dtype.unit
+        result = self._pa_array.cast(pa.timestamp(current_unit, tz))
+        return type(self)(result)
+
+
+def transpose_homogeneous_pyarrow(
+    arrays: Sequence[ArrowExtensionArray],
+) -> list[ArrowExtensionArray]:
+    """Transpose arrow extension arrays in a list, but faster.
+
+    Input should be a list of arrays of equal length and all have the same
+    dtype. The caller is responsible for ensuring validity of input data.
+    """
+    arrays = list(arrays)
+    nrows, ncols = len(arrays[0]), len(arrays)
+    indices = np.arange(nrows * ncols).reshape(ncols, nrows).T.reshape(-1)
+    arr = pa.chunked_array([chunk for arr in arrays for chunk in arr._pa_array.chunks])
+    arr = arr.take(indices)
+    return [ArrowExtensionArray(arr.slice(i * ncols, ncols)) for i in range(nrows)]
diff --git a/pandas/pandas/core/arrays/arrow/extension_types.py b/pandas/pandas/core/arrays/arrow/extension_types.py
new file mode 100644
index 0000000000000..2fa5f7a882cc7
--- /dev/null
+++ b/pandas/pandas/core/arrays/arrow/extension_types.py
@@ -0,0 +1,174 @@
+from __future__ import annotations
+
+import json
+from typing import TYPE_CHECKING
+
+import pyarrow
+
+from pandas.compat import pa_version_under14p1
+
+from pandas.core.dtypes.dtypes import (
+    IntervalDtype,
+    PeriodDtype,
+)
+
+from pandas.core.arrays.interval import VALID_CLOSED
+
+if TYPE_CHECKING:
+    from pandas._typing import IntervalClosedType
+
+
+class ArrowPeriodType(pyarrow.ExtensionType):
+    def __init__(self, freq) -> None:
+        # attributes need to be set first before calling
+        # super init (as that calls serialize)
+        self._freq = freq
+        pyarrow.ExtensionType.__init__(self, pyarrow.int64(), "pandas.period")
+
+    @property
+    def freq(self):
+        return self._freq
+
+    def __arrow_ext_serialize__(self) -> bytes:
+        metadata = {"freq": self.freq}
+        return json.dumps(metadata).encode()
+
+    @classmethod
+    def __arrow_ext_deserialize__(cls, storage_type, serialized) -> ArrowPeriodType:
+        metadata = json.loads(serialized.decode())
+        return ArrowPeriodType(metadata["freq"])
+
+    def __eq__(self, other):
+        if isinstance(other, pyarrow.BaseExtensionType):
+            return type(self) == type(other) and self.freq == other.freq
+        else:
+            return NotImplemented
+
+    def __ne__(self, other) -> bool:
+        return not self == other
+
+    def __hash__(self) -> int:
+        return hash((str(self), self.freq))
+
+    def to_pandas_dtype(self) -> PeriodDtype:
+        return PeriodDtype(freq=self.freq)
+
+
+# register the type with a dummy instance
+_period_type = ArrowPeriodType("D")
+pyarrow.register_extension_type(_period_type)
+
+
+class ArrowIntervalType(pyarrow.ExtensionType):
+    def __init__(self, subtype, closed: IntervalClosedType) -> None:
+        # attributes need to be set first before calling
+        # super init (as that calls serialize)
+        assert closed in VALID_CLOSED
+        self._closed: IntervalClosedType = closed
+        if not isinstance(subtype, pyarrow.DataType):
+            subtype = pyarrow.type_for_alias(str(subtype))
+        self._subtype = subtype
+
+        storage_type = pyarrow.struct([("left", subtype), ("right", subtype)])
+        pyarrow.ExtensionType.__init__(self, storage_type, "pandas.interval")
+
+    @property
+    def subtype(self):
+        return self._subtype
+
+    @property
+    def closed(self) -> IntervalClosedType:
+        return self._closed
+
+    def __arrow_ext_serialize__(self) -> bytes:
+        metadata = {"subtype": str(self.subtype), "closed": self.closed}
+        return json.dumps(metadata).encode()
+
+    @classmethod
+    def __arrow_ext_deserialize__(cls, storage_type, serialized) -> ArrowIntervalType:
+        metadata = json.loads(serialized.decode())
+        subtype = pyarrow.type_for_alias(metadata["subtype"])
+        closed = metadata["closed"]
+        return ArrowIntervalType(subtype, closed)
+
+    def __eq__(self, other):
+        if isinstance(other, pyarrow.BaseExtensionType):
+            return (
+                type(self) == type(other)
+                and self.subtype == other.subtype
+                and self.closed == other.closed
+            )
+        else:
+            return NotImplemented
+
+    def __ne__(self, other) -> bool:
+        return not self == other
+
+    def __hash__(self) -> int:
+        return hash((str(self), str(self.subtype), self.closed))
+
+    def to_pandas_dtype(self) -> IntervalDtype:
+        return IntervalDtype(self.subtype.to_pandas_dtype(), self.closed)
+
+
+# register the type with a dummy instance
+_interval_type = ArrowIntervalType(pyarrow.int64(), "left")
+pyarrow.register_extension_type(_interval_type)
+
+
+_ERROR_MSG = """\
+Disallowed deserialization of 'arrow.py_extension_type':
+storage_type = {storage_type}
+serialized = {serialized}
+pickle disassembly:\n{pickle_disassembly}
+
+Reading of untrusted Parquet or Feather files with a PyExtensionType column
+allows arbitrary code execution.
+If you trust this file, you can enable reading the extension type by one of:
+
+- upgrading to pyarrow >= 14.0.1, and call `pa.PyExtensionType.set_auto_load(True)`
+- install pyarrow-hotfix (`pip install pyarrow-hotfix`) and disable it by running
+  `import pyarrow_hotfix; pyarrow_hotfix.uninstall()`
+
+We strongly recommend updating your Parquet/Feather files to use extension types
+derived from `pyarrow.ExtensionType` instead, and register this type explicitly.
+"""
+
+
+def patch_pyarrow() -> None:
+    # starting from pyarrow 14.0.1, it has its own mechanism
+    if not pa_version_under14p1:
+        return
+
+    # if https://github.com/pitrou/pyarrow-hotfix was installed and enabled
+    if getattr(pyarrow, "_hotfix_installed", False):
+        return
+
+    class ForbiddenExtensionType(pyarrow.ExtensionType):
+        def __arrow_ext_serialize__(self) -> bytes:
+            return b""
+
+        @classmethod
+        def __arrow_ext_deserialize__(cls, storage_type, serialized):
+            import io
+            import pickletools
+
+            out = io.StringIO()
+            pickletools.dis(serialized, out)
+            raise RuntimeError(
+                _ERROR_MSG.format(
+                    storage_type=storage_type,
+                    serialized=serialized,
+                    pickle_disassembly=out.getvalue(),
+                )
+            )
+
+    pyarrow.unregister_extension_type("arrow.py_extension_type")
+    pyarrow.register_extension_type(
+        ForbiddenExtensionType(pyarrow.null(), "arrow.py_extension_type")
+    )
+
+    pyarrow._hotfix_installed = True
+
+
+patch_pyarrow()
diff --git a/pandas/pandas/core/arrays/base.py b/pandas/pandas/core/arrays/base.py
new file mode 100644
index 0000000000000..33745438e2aea
--- /dev/null
+++ b/pandas/pandas/core/arrays/base.py
@@ -0,0 +1,2856 @@
+"""
+An interface for extending pandas with custom arrays.
+
+.. warning::
+
+   This is an experimental API and subject to breaking changes
+   without warning.
+"""
+
+from __future__ import annotations
+
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    ClassVar,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    algos as libalgos,
+    lib,
+)
+from pandas.compat import set_function_name
+from pandas.compat.numpy import function as nv
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    cache_readonly,
+)
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import (
+    validate_bool_kwarg,
+    validate_insert_loc,
+)
+
+from pandas.core.dtypes.cast import maybe_cast_pointwise_result
+from pandas.core.dtypes.common import (
+    is_list_like,
+    is_scalar,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import (
+    arraylike,
+    missing,
+    roperator,
+)
+from pandas.core.algorithms import (
+    duplicated,
+    factorize_array,
+    isin,
+    map_array,
+    mode,
+    rank,
+    unique,
+)
+from pandas.core.array_algos.quantile import quantile_with_mask
+from pandas.core.missing import _fill_limit_area_1d
+from pandas.core.sorting import (
+    nargminmax,
+    nargsort,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Iterator,
+        Sequence,
+    )
+
+    from pandas._libs.missing import NAType
+    from pandas._typing import (
+        ArrayLike,
+        AstypeArg,
+        AxisInt,
+        Dtype,
+        DtypeObj,
+        FillnaOptions,
+        InterpolateOptions,
+        NumpySorter,
+        NumpyValueArrayLike,
+        PositionalIndexer,
+        ScalarIndexer,
+        Self,
+        SequenceIndexer,
+        Shape,
+        SortKind,
+        TakeIndexer,
+        npt,
+    )
+
+    from pandas import Index
+
+_extension_array_shared_docs: dict[str, str] = {}
+
+
+class ExtensionArray:
+    """
+    Abstract base class for custom 1-D array types.
+
+    pandas will recognize instances of this class as proper arrays
+    with a custom type and will not attempt to coerce them to objects. They
+    may be stored directly inside a :class:`DataFrame` or :class:`Series`.
+
+    Attributes
+    ----------
+    dtype
+    nbytes
+    ndim
+    shape
+
+    Methods
+    -------
+    argsort
+    astype
+    copy
+    dropna
+    duplicated
+    factorize
+    fillna
+    equals
+    insert
+    interpolate
+    isin
+    isna
+    ravel
+    repeat
+    searchsorted
+    shift
+    take
+    tolist
+    unique
+    view
+    _accumulate
+    _concat_same_type
+    _explode
+    _formatter
+    _from_factorized
+    _from_sequence
+    _from_sequence_of_strings
+    _hash_pandas_object
+    _pad_or_backfill
+    _reduce
+    _values_for_argsort
+    _values_for_factorize
+
+    See Also
+    --------
+    api.extensions.ExtensionDtype : A custom data type, to be paired with an
+        ExtensionArray.
+    api.extensions.ExtensionArray.dtype : An instance of ExtensionDtype.
+
+    Notes
+    -----
+    The interface includes the following abstract methods that must be
+    implemented by subclasses:
+
+    * _from_sequence
+    * _from_factorized
+    * __getitem__
+    * __len__
+    * __eq__
+    * dtype
+    * nbytes
+    * isna
+    * take
+    * copy
+    * _concat_same_type
+    * interpolate
+
+    A default repr displaying the type, (truncated) data, length,
+    and dtype is provided. It can be customized or replaced by
+    by overriding:
+
+    * __repr__ : A default repr for the ExtensionArray.
+    * _formatter : Print scalars inside a Series or DataFrame.
+
+    Some methods require casting the ExtensionArray to an ndarray of Python
+    objects with ``self.astype(object)``, which may be expensive. When
+    performance is a concern, we highly recommend overriding the following
+    methods:
+
+    * fillna
+    * _pad_or_backfill
+    * dropna
+    * unique
+    * factorize / _values_for_factorize
+    * argsort, argmax, argmin / _values_for_argsort
+    * searchsorted
+    * map
+
+    The remaining methods implemented on this class should be performant,
+    as they only compose abstract methods. Still, a more efficient
+    implementation may be available, and these methods can be overridden.
+
+    One can implement methods to handle array accumulations or reductions.
+
+    * _accumulate
+    * _reduce
+
+    One can implement methods to handle parsing from strings that will be used
+    in methods such as ``pandas.io.parsers.read_csv``.
+
+    * _from_sequence_of_strings
+
+    This class does not inherit from 'abc.ABCMeta' for performance reasons.
+    Methods and properties required by the interface raise
+    ``pandas.errors.AbstractMethodError`` and no ``register`` method is
+    provided for registering virtual subclasses.
+
+    ExtensionArrays are limited to 1 dimension.
+
+    They may be backed by none, one, or many NumPy arrays. For example,
+    ``pandas.Categorical`` is an extension array backed by two arrays,
+    one for codes and one for categories. An array of IPv6 address may
+    be backed by a NumPy structured array with two fields, one for the
+    lower 64 bits and one for the upper 64 bits. Or they may be backed
+    by some other storage type, like Python lists. Pandas makes no
+    assumptions on how the data are stored, just that it can be converted
+    to a NumPy array.
+    The ExtensionArray interface does not impose any rules on how this data
+    is stored. However, currently, the backing data cannot be stored in
+    attributes called ``.values`` or ``._values`` to ensure full compatibility
+    with pandas internals. But other names as ``.data``, ``._data``,
+    ``._items``, ... can be freely used.
+
+    If implementing NumPy's ``__array_ufunc__`` interface, pandas expects
+    that
+
+    1. You defer by returning ``NotImplemented`` when any Series are present
+       in `inputs`. Pandas will extract the arrays and call the ufunc again.
+    2. You define a ``_HANDLED_TYPES`` tuple as an attribute on the class.
+       Pandas inspect this to determine whether the ufunc is valid for the
+       types present.
+
+    See :ref:`extending.extension.ufunc` for more.
+
+    By default, ExtensionArrays are not hashable.  Immutable subclasses may
+    override this behavior.
+
+    Examples
+    --------
+    Please see the following:
+
+    https://github.com/pandas-dev/pandas/blob/main/pandas/tests/extension/list/array.py
+    """
+
+    # '_typ' is for pandas.core.dtypes.generic.ABCExtensionArray.
+    # Don't override this.
+    _typ = "extension"
+
+    # similar to __array_priority__, positions ExtensionArray after Index,
+    #  Series, and DataFrame.  EA subclasses may override to choose which EA
+    #  subclass takes priority. If overriding, the value should always be
+    #  strictly less than 2000 to be below Index.__pandas_priority__.
+    __pandas_priority__ = 1000
+
+    # ------------------------------------------------------------------------
+    # Constructors
+    # ------------------------------------------------------------------------
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        """
+        Construct a new ExtensionArray from a sequence of scalars.
+
+        Parameters
+        ----------
+        scalars : Sequence
+            Each element will be an instance of the scalar type for this
+            array, ``cls.dtype.type`` or be converted into this type in this method.
+        dtype : dtype, optional
+            Construct for this particular dtype. This should be a Dtype
+            compatible with the ExtensionArray.
+        copy : bool, default False
+            If True, copy the underlying data.
+
+        Returns
+        -------
+        ExtensionArray
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._from_sequence_of_strings : Construct a new
+            ExtensionArray from a sequence of strings.
+        api.extensions.ExtensionArray._hash_pandas_object : Hook for
+            hash_pandas_object.
+
+        Examples
+        --------
+        >>> pd.arrays.IntegerArray._from_sequence([4, 5])
+        <IntegerArray>
+        [4, 5]
+        Length: 2, dtype: Int64
+        """
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _from_scalars(cls, scalars, *, dtype: DtypeObj) -> Self:
+        """
+        Strict analogue to _from_sequence, allowing only sequences of scalars
+        that should be specifically inferred to the given dtype.
+
+        Parameters
+        ----------
+        scalars : sequence
+        dtype : ExtensionDtype
+
+        Raises
+        ------
+        TypeError or ValueError
+
+        Notes
+        -----
+        This is called in a try/except block when casting the result of a
+        pointwise operation.
+        """
+        try:
+            return cls._from_sequence(scalars, dtype=dtype, copy=False)
+        except (ValueError, TypeError):
+            raise
+        except Exception:
+            warnings.warn(
+                "_from_scalars should only raise ValueError or TypeError. "
+                "Consider overriding _from_scalars where appropriate.",
+                stacklevel=find_stack_level(),
+            )
+            raise
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        """
+        Construct a new ExtensionArray from a sequence of strings.
+
+        Parameters
+        ----------
+        strings : Sequence
+            Each element will be an instance of the scalar type for this
+            array, ``cls.dtype.type``.
+        dtype : ExtensionDtype
+            Construct for this particular dtype. This should be a Dtype
+            compatible with the ExtensionArray.
+        copy : bool, default False
+            If True, copy the underlying data.
+
+        Returns
+        -------
+        ExtensionArray
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._from_sequence : Construct a new ExtensionArray
+            from a sequence of scalars.
+        api.extensions.ExtensionArray._from_factorized : Reconstruct an ExtensionArray
+            after factorization.
+        api.extensions.ExtensionArray._from_scalars : Strict analogue to _from_sequence,
+            allowing only sequences of scalars that should be specifically inferred to
+            the given dtype.
+
+        Examples
+        --------
+        >>> pd.arrays.IntegerArray._from_sequence_of_strings(
+        ...     ["1", "2", "3"], dtype=pd.Int64Dtype()
+        ... )
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _from_factorized(cls, values, original):
+        """
+        Reconstruct an ExtensionArray after factorization.
+
+        Parameters
+        ----------
+        values : ndarray
+            An integer ndarray with the factorized values.
+        original : ExtensionArray
+            The original ExtensionArray that factorize was called on.
+
+        See Also
+        --------
+        factorize : Top-level factorize method that dispatches here.
+        ExtensionArray.factorize : Encode the extension array as an enumerated type.
+
+        Examples
+        --------
+        >>> interv_arr = pd.arrays.IntervalArray(
+        ...     [pd.Interval(0, 1), pd.Interval(1, 5), pd.Interval(1, 5)]
+        ... )
+        >>> codes, uniques = pd.factorize(interv_arr)
+        >>> pd.arrays.IntervalArray._from_factorized(uniques, interv_arr)
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        """
+        raise AbstractMethodError(cls)
+
+    # ------------------------------------------------------------------------
+    # Must be a Sequence
+    # ------------------------------------------------------------------------
+    @overload
+    def __getitem__(self, item: ScalarIndexer) -> Any: ...
+
+    @overload
+    def __getitem__(self, item: SequenceIndexer) -> Self: ...
+
+    def __getitem__(self, item: PositionalIndexer) -> Self | Any:
+        """
+        Select a subset of self.
+
+        Parameters
+        ----------
+        item : int, slice, or ndarray
+            * int: The position in 'self' to get.
+
+            * slice: A slice object, where 'start', 'stop', and 'step' are
+              integers or None
+
+            * ndarray: A 1-d boolean NumPy ndarray the same length as 'self'
+
+            * list[int]:  A list of int
+
+        Returns
+        -------
+        item : scalar or ExtensionArray
+
+        Notes
+        -----
+        For scalar ``item``, return a scalar value suitable for the array's
+        type. This should be an instance of ``self.dtype.type``.
+
+        For slice ``key``, return an instance of ``ExtensionArray``, even
+        if the slice is length 0 or 1.
+
+        For a boolean mask, return an instance of ``ExtensionArray``, filtered
+        to the values where ``item`` is True.
+        """
+        raise AbstractMethodError(self)
+
+    def __setitem__(self, key, value) -> None:
+        """
+        Set one or more values inplace.
+
+        This method is not required to satisfy the pandas extension array
+        interface.
+
+        Parameters
+        ----------
+        key : int, ndarray, or slice
+            When called from, e.g. ``Series.__setitem__``, ``key`` will be
+            one of
+
+            * scalar int
+            * ndarray of integers.
+            * boolean ndarray
+            * slice object
+
+        value : ExtensionDtype.type, Sequence[ExtensionDtype.type], or object
+            value or values to be set of ``key``.
+
+        Returns
+        -------
+        None
+        """
+        # Some notes to the ExtensionArray implementer who may have ended up
+        # here. While this method is not required for the interface, if you
+        # *do* choose to implement __setitem__, then some semantics should be
+        # observed:
+        #
+        # * Setting multiple values : ExtensionArrays should support setting
+        #   multiple values at once, 'key' will be a sequence of integers and
+        #  'value' will be a same-length sequence.
+        #
+        # * Broadcasting : For a sequence 'key' and a scalar 'value',
+        #   each position in 'key' should be set to 'value'.
+        #
+        # * Coercion : Most users will expect basic coercion to work. For
+        #   example, a string like '2018-01-01' is coerced to a datetime
+        #   when setting on a datetime64ns array. In general, if the
+        #   __init__ method coerces that value, then so should __setitem__
+        # Note, also, that Series/DataFrame.where internally use __setitem__
+        # on a copy of the data.
+        raise NotImplementedError(f"{type(self)} does not implement __setitem__.")
+
+    def __len__(self) -> int:
+        """
+        Length of this array
+
+        Returns
+        -------
+        length : int
+        """
+        raise AbstractMethodError(self)
+
+    def __iter__(self) -> Iterator[Any]:
+        """
+        Iterate over elements of the array.
+        """
+        # This needs to be implemented so that pandas recognizes extension
+        # arrays as list-like. The default implementation makes successive
+        # calls to ``__getitem__``, which may be slower than necessary.
+        for i in range(len(self)):
+            yield self[i]
+
+    def __contains__(self, item: object) -> bool | np.bool_:
+        """
+        Return for `item in self`.
+        """
+        # GH37867
+        # comparisons of any item to pd.NA always return pd.NA, so e.g. "a" in [pd.NA]
+        # would raise a TypeError. The implementation below works around that.
+        if is_scalar(item) and isna(item):
+            if not self._can_hold_na:
+                return False
+            elif item is self.dtype.na_value or isinstance(item, self.dtype.type):
+                return self._hasna
+            else:
+                return False
+        else:
+            # error: Item "ExtensionArray" of "Union[ExtensionArray, ndarray]" has no
+            # attribute "any"
+            return (item == self).any()  # type: ignore[union-attr]
+
+    # error: Signature of "__eq__" incompatible with supertype "object"
+    def __eq__(self, other: object) -> ArrayLike:  # type: ignore[override]
+        """
+        Return for `self == other` (element-wise equality).
+        """
+        # Implementer note: this should return a boolean numpy ndarray or
+        # a boolean ExtensionArray.
+        # When `other` is one of Series, Index, or DataFrame, this method should
+        # return NotImplemented (to ensure that those objects are responsible for
+        # first unpacking the arrays, and then dispatch the operation to the
+        # underlying arrays)
+        raise AbstractMethodError(self)
+
+    # error: Signature of "__ne__" incompatible with supertype "object"
+    def __ne__(self, other: object) -> ArrayLike:  # type: ignore[override]
+        """
+        Return for `self != other` (element-wise in-equality).
+        """
+        # error: Unsupported operand type for ~ ("ExtensionArray")
+        return ~(self == other)  # type: ignore[operator]
+
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        """
+        Convert to a NumPy ndarray.
+
+        This is similar to :meth:`numpy.asarray`, but may provide additional control
+        over how the conversion is done.
+
+        Parameters
+        ----------
+        dtype : str or numpy.dtype, optional
+            The dtype to pass to :meth:`numpy.asarray`.
+        copy : bool, default False
+            Whether to ensure that the returned value is a not a view on
+            another array. Note that ``copy=False`` does not *ensure* that
+            ``to_numpy()`` is no-copy. Rather, ``copy=True`` ensure that
+            a copy is made, even if not strictly necessary.
+        na_value : Any, optional
+            The value to use for missing values. The default value depends
+            on `dtype` and the type of the array.
+
+        Returns
+        -------
+        numpy.ndarray
+        """
+        result = np.asarray(self, dtype=dtype)
+        if copy or na_value is not lib.no_default:
+            result = result.copy()
+        if na_value is not lib.no_default:
+            result[self.isna()] = na_value
+        return result
+
+    # ------------------------------------------------------------------------
+    # Required attributes
+    # ------------------------------------------------------------------------
+
+    @property
+    def dtype(self) -> ExtensionDtype:
+        """
+        An instance of ExtensionDtype.
+
+        See Also
+        --------
+        api.extensions.ExtensionDtype : Base class for extension dtypes.
+        api.extensions.ExtensionArray : Base class for extension array types.
+        api.extensions.ExtensionArray.dtype : The dtype of an ExtensionArray.
+        Series.dtype : The dtype of a Series.
+        DataFrame.dtype : The dtype of a DataFrame.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3]).dtype
+        Int64Dtype()
+        """
+        raise AbstractMethodError(self)
+
+    @property
+    def shape(self) -> Shape:
+        """
+        Return a tuple of the array dimensions.
+
+        See Also
+        --------
+        numpy.ndarray.shape : Similar attribute which returns the shape of an array.
+        DataFrame.shape : Return a tuple representing the dimensionality of the
+            DataFrame.
+        Series.shape : Return a tuple representing the dimensionality of the Series.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.shape
+        (3,)
+        """
+        return (len(self),)
+
+    @property
+    def size(self) -> int:
+        """
+        The number of elements in the array.
+        """
+        # error: Incompatible return value type (got "signedinteger[_64Bit]",
+        # expected "int")  [return-value]
+        return np.prod(self.shape)  # type: ignore[return-value]
+
+    @property
+    def ndim(self) -> int:
+        """
+        Extension Arrays are only allowed to be 1-dimensional.
+
+        See Also
+        --------
+        ExtensionArray.shape: Return a tuple of the array dimensions.
+        ExtensionArray.size: The number of elements in the array.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.ndim
+        1
+        """
+        return 1
+
+    @property
+    def nbytes(self) -> int:
+        """
+        The number of bytes needed to store this object in memory.
+
+        See Also
+        --------
+        ExtensionArray.shape: Return a tuple of the array dimensions.
+        ExtensionArray.size: The number of elements in the array.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3]).nbytes
+        27
+        """
+        # If this is expensive to compute, return an approximate lower bound
+        # on the number of bytes needed.
+        raise AbstractMethodError(self)
+
+    # ------------------------------------------------------------------------
+    # Additional Methods
+    # ------------------------------------------------------------------------
+
+    @overload
+    def astype(self, dtype: npt.DTypeLike, copy: bool = ...) -> np.ndarray: ...
+
+    @overload
+    def astype(self, dtype: ExtensionDtype, copy: bool = ...) -> ExtensionArray: ...
+
+    @overload
+    def astype(self, dtype: AstypeArg, copy: bool = ...) -> ArrayLike: ...
+
+    def astype(self, dtype: AstypeArg, copy: bool = True) -> ArrayLike:
+        """
+        Cast to a NumPy array or ExtensionArray with 'dtype'.
+
+        Parameters
+        ----------
+        dtype : str or dtype
+            Typecode or data-type to which the array is cast.
+        copy : bool, default True
+            Whether to copy the data, even if not necessary. If False,
+            a copy is made only if the old dtype does not match the
+            new dtype.
+
+        Returns
+        -------
+        np.ndarray or pandas.api.extensions.ExtensionArray
+            An ``ExtensionArray`` if ``dtype`` is ``ExtensionDtype``,
+            otherwise a Numpy ndarray with ``dtype`` for its dtype.
+
+        See Also
+        --------
+        Series.astype : Cast a Series to a different dtype.
+        DataFrame.astype : Cast a DataFrame to a different dtype.
+        api.extensions.ExtensionArray : Base class for ExtensionArray objects.
+        core.arrays.DatetimeArray._from_sequence : Create a DatetimeArray from a
+            sequence.
+        core.arrays.TimedeltaArray._from_sequence : Create a TimedeltaArray from
+            a sequence.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+
+        Casting to another ``ExtensionDtype`` returns an ``ExtensionArray``:
+
+        >>> arr1 = arr.astype("Float64")
+        >>> arr1
+        <FloatingArray>
+        [1.0, 2.0, 3.0]
+        Length: 3, dtype: Float64
+        >>> arr1.dtype
+        Float64Dtype()
+
+        Otherwise, we will get a Numpy ndarray:
+
+        >>> arr2 = arr.astype("float64")
+        >>> arr2
+        array([1., 2., 3.])
+        >>> arr2.dtype
+        dtype('float64')
+        """
+        dtype = pandas_dtype(dtype)
+        if dtype == self.dtype:
+            if not copy:
+                return self
+            else:
+                return self.copy()
+
+        if isinstance(dtype, ExtensionDtype):
+            cls = dtype.construct_array_type()
+            return cls._from_sequence(self, dtype=dtype, copy=copy)
+
+        elif lib.is_np_dtype(dtype, "M"):
+            from pandas.core.arrays import DatetimeArray
+
+            return DatetimeArray._from_sequence(self, dtype=dtype, copy=copy)
+
+        elif lib.is_np_dtype(dtype, "m"):
+            from pandas.core.arrays import TimedeltaArray
+
+            return TimedeltaArray._from_sequence(self, dtype=dtype, copy=copy)
+
+        if not copy:
+            return np.asarray(self, dtype=dtype)
+        else:
+            return np.array(self, dtype=dtype, copy=copy)
+
+    def isna(self) -> np.ndarray | ExtensionArraySupportsAnyAll:
+        """
+        A 1-D array indicating if each value is missing.
+
+        Returns
+        -------
+        numpy.ndarray or pandas.api.extensions.ExtensionArray
+            In most cases, this should return a NumPy ndarray. For
+            exceptional cases like ``SparseArray``, where returning
+            an ndarray would be expensive, an ExtensionArray may be
+            returned.
+
+        See Also
+        --------
+        ExtensionArray.dropna: Return ExtensionArray without NA values.
+        ExtensionArray.fillna: Fill NA/NaN values using the specified method.
+
+        Notes
+        -----
+        If returning an ExtensionArray, then
+
+        * ``na_values._is_boolean`` should be True
+        * ``na_values`` should implement :func:`ExtensionArray._reduce`
+        * ``na_values`` should implement :func:`ExtensionArray._accumulate`
+        * ``na_values.any`` and ``na_values.all`` should be implemented
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, np.nan, np.nan])
+        >>> arr.isna()
+        array([False, False,  True,  True])
+        """
+        raise AbstractMethodError(self)
+
+    @property
+    def _hasna(self) -> bool:
+        # GH#22680
+        """
+        Equivalent to `self.isna().any()`.
+
+        Some ExtensionArray subclasses may be able to optimize this check.
+        """
+        return bool(self.isna().any())
+
+    def _values_for_argsort(self) -> np.ndarray:
+        """
+        Return values for sorting.
+
+        Returns
+        -------
+        ndarray
+            The transformed values should maintain the ordering between values
+            within the array.
+
+        See Also
+        --------
+        ExtensionArray.argsort : Return the indices that would sort this array.
+
+        Notes
+        -----
+        The caller is responsible for *not* modifying these values in-place, so
+        it is safe for implementers to give views on ``self``.
+
+        Functions that use this (e.g. ``ExtensionArray.argsort``) should ignore
+        entries with missing values in the original array (according to
+        ``self.isna()``). This means that the corresponding entries in the returned
+        array don't need to be modified to sort correctly.
+
+        Examples
+        --------
+        In most cases, this is the underlying Numpy array of the ``ExtensionArray``:
+
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr._values_for_argsort()
+        array([1, 2, 3])
+        """
+        # Note: this is used in `ExtensionArray.argsort/argmin/argmax`.
+        return np.array(self)
+
+    def argsort(
+        self,
+        *,
+        ascending: bool = True,
+        kind: SortKind = "quicksort",
+        na_position: str = "last",
+        **kwargs,
+    ) -> np.ndarray:
+        """
+        Return the indices that would sort this array.
+
+        Parameters
+        ----------
+        ascending : bool, default True
+            Whether the indices should result in an ascending
+            or descending sort.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, optional
+            Sorting algorithm.
+        na_position : {'first', 'last'}, default 'last'
+            If ``'first'``, put ``NaN`` values at the beginning.
+            If ``'last'``, put ``NaN`` values at the end.
+        **kwargs
+            Passed through to :func:`numpy.argsort`.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            Array of indices that sort ``self``. If NaN values are contained,
+            NaN values are placed at the end.
+
+        See Also
+        --------
+        numpy.argsort : Sorting implementation used internally.
+
+        Examples
+        --------
+        >>> arr = pd.array([3, 1, 2, 5, 4])
+        >>> arr.argsort()
+        array([1, 2, 0, 4, 3])
+        """
+        # Implementer note: You have two places to override the behavior of
+        # argsort.
+        # 1. _values_for_argsort : construct the values passed to np.argsort
+        # 2. argsort : total control over sorting. In case of overriding this,
+        #    it is recommended to also override argmax/argmin
+        ascending = nv.validate_argsort_with_ascending(ascending, (), kwargs)
+
+        values = self._values_for_argsort()
+        return nargsort(
+            values,
+            kind=kind,
+            ascending=ascending,
+            na_position=na_position,
+            mask=np.asarray(self.isna()),
+        )
+
+    def argmin(self, skipna: bool = True) -> int:
+        """
+        Return the index of minimum value.
+
+        In case of multiple occurrences of the minimum value, the index
+        corresponding to the first occurrence is returned.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        ExtensionArray.argmax : Return the index of the maximum value.
+
+        Examples
+        --------
+        >>> arr = pd.array([3, 1, 2, 5, 4])
+        >>> arr.argmin()
+        1
+        """
+        # Implementer note: You have two places to override the behavior of
+        # argmin.
+        # 1. _values_for_argsort : construct the values used in nargminmax
+        # 2. argmin itself : total control over sorting.
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return nargminmax(self, "argmin")
+
+    def argmax(self, skipna: bool = True) -> int:
+        """
+        Return the index of maximum value.
+
+        In case of multiple occurrences of the maximum value, the index
+        corresponding to the first occurrence is returned.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+
+        Returns
+        -------
+        int
+
+        See Also
+        --------
+        ExtensionArray.argmin : Return the index of the minimum value.
+
+        Examples
+        --------
+        >>> arr = pd.array([3, 1, 2, 5, 4])
+        >>> arr.argmax()
+        3
+        """
+        # Implementer note: You have two places to override the behavior of
+        # argmax.
+        # 1. _values_for_argsort : construct the values used in nargminmax
+        # 2. argmax itself : total control over sorting.
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return nargminmax(self, "argmax")
+
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        axis: int,
+        index: Index,
+        limit,
+        limit_direction,
+        limit_area,
+        copy: bool,
+        **kwargs,
+    ) -> Self:
+        """
+        Fill NaN values using an interpolation method.
+
+        Parameters
+        ----------
+        method : str, default 'linear'
+            Interpolation technique to use. One of:
+            * 'linear': Ignore the index and treat the values as equally spaced.
+            This is the only method supported on MultiIndexes.
+            * 'time': Works on daily and higher resolution data to interpolate
+            given length of interval.
+            * 'index', 'values': use the actual numerical values of the index.
+            * 'nearest', 'zero', 'slinear', 'quadratic', 'cubic', 'barycentric',
+            'polynomial': Passed to scipy.interpolate.interp1d, whereas 'spline'
+            is passed to scipy.interpolate.UnivariateSpline. These methods use
+            the numerical values of the index.
+            Both 'polynomial' and 'spline' require that you also specify an
+            order (int), e.g. arr.interpolate(method='polynomial', order=5).
+            * 'krogh', 'piecewise_polynomial', 'spline', 'pchip', 'akima',
+            'cubicspline': Wrappers around the SciPy interpolation methods
+            of similar names. See Notes.
+            * 'from_derivatives': Refers to scipy.interpolate.BPoly.from_derivatives.
+        axis : int
+            Axis to interpolate along. For 1-dimensional data, use 0.
+        index : Index
+            Index to use for interpolation.
+        limit : int or None
+            Maximum number of consecutive NaNs to fill. Must be greater than 0.
+        limit_direction : {'forward', 'backward', 'both'}
+            Consecutive NaNs will be filled in this direction.
+        limit_area : {'inside', 'outside'} or None
+            If limit is specified, consecutive NaNs will be filled with this
+            restriction.
+            * None: No fill restriction.
+            * 'inside': Only fill NaNs surrounded by valid values (interpolate).
+            * 'outside': Only fill NaNs outside valid values (extrapolate).
+        copy : bool
+            If True, a copy of the object is returned with interpolated values.
+        **kwargs : optional
+            Keyword arguments to pass on to the interpolating function.
+
+        Returns
+        -------
+        ExtensionArray
+            An ExtensionArray with interpolated values.
+
+        See Also
+        --------
+        Series.interpolate : Interpolate values in a Series.
+        DataFrame.interpolate : Interpolate values in a DataFrame.
+
+        Notes
+        -----
+        - All parameters must be specified as keyword arguments.
+        - The 'krogh', 'piecewise_polynomial', 'spline', 'pchip' and 'akima'
+          methods are wrappers around the respective SciPy implementations of
+          similar names. These use the actual numerical values of the index.
+
+        Examples
+        --------
+        Interpolating values in a NumPy array:
+
+        >>> arr = pd.arrays.NumpyExtensionArray(np.array([0, 1, np.nan, 3]))
+        >>> arr.interpolate(
+        ...     method="linear",
+        ...     limit=3,
+        ...     limit_direction="forward",
+        ...     index=pd.Index(range(len(arr))),
+        ...     fill_value=1,
+        ...     copy=False,
+        ...     axis=0,
+        ...     limit_area="inside",
+        ... )
+        <NumpyExtensionArray>
+        [0.0, 1.0, 2.0, 3.0]
+        Length: 4, dtype: float64
+
+        Interpolating values in a FloatingArray:
+
+        >>> arr = pd.array([1.0, pd.NA, 3.0, 4.0, pd.NA, 6.0], dtype="Float64")
+        >>> arr.interpolate(
+        ...     method="linear",
+        ...     axis=0,
+        ...     index=pd.Index(range(len(arr))),
+        ...     limit=None,
+        ...     limit_direction="both",
+        ...     limit_area=None,
+        ...     copy=True,
+        ... )
+        <FloatingArray>
+        [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]
+        Length: 6, dtype: Float64
+        """
+        # NB: we return type(self) even if copy=False
+        raise NotImplementedError(
+            f"{type(self).__name__} does not implement interpolate"
+        )
+
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        """
+        Pad or backfill values, used by Series/DataFrame ffill and bfill.
+
+        Parameters
+        ----------
+        method : {'backfill', 'bfill', 'pad', 'ffill'}
+            Method to use for filling holes in reindexed Series:
+
+            * pad / ffill: propagate last valid observation forward to next valid.
+            * backfill / bfill: use NEXT valid observation to fill gap.
+
+        limit : int, default None
+            This is the maximum number of consecutive
+            NaN values to forward/backward fill. In other words, if there is
+            a gap with more than this number of consecutive NaNs, it will only
+            be partially filled. If method is not specified, this is the
+            maximum number of entries along the entire axis where NaNs will be
+            filled.
+
+        limit_area : {'inside', 'outside'} or None, default None
+            Specifies which area to limit filling.
+            - 'inside': Limit the filling to the area within the gaps.
+            - 'outside': Limit the filling to the area outside the gaps.
+            If `None`, no limitation is applied.
+
+        copy : bool, default True
+            Whether to make a copy of the data before filling. If False, then
+            the original should be modified and no new memory should be allocated.
+            For ExtensionArray subclasses that cannot do this, it is at the
+            author's discretion whether to ignore "copy=False" or to raise.
+            The base class implementation ignores the keyword if any NAs are
+            present.
+
+        Returns
+        -------
+        Same type as self
+            The filled array with the same type as the original.
+
+        See Also
+        --------
+        Series.ffill : Forward fill missing values.
+        Series.bfill : Backward fill missing values.
+        DataFrame.ffill : Forward fill missing values in DataFrame.
+        DataFrame.bfill : Backward fill missing values in DataFrame.
+        api.types.isna : Check for missing values.
+        api.types.isnull : Check for missing values.
+
+        Examples
+        --------
+        >>> arr = pd.array([np.nan, np.nan, 2, 3, np.nan, np.nan])
+        >>> arr._pad_or_backfill(method="backfill", limit=1)
+        <IntegerArray>
+        [<NA>, 2, 2, 3, <NA>, <NA>]
+        Length: 6, dtype: Int64
+        """
+        mask = self.isna()
+
+        if mask.any():
+            # NB: the base class does not respect the "copy" keyword
+            meth = missing.clean_fill_method(method)
+
+            npmask = np.asarray(mask)
+            if limit_area is not None and not npmask.all():
+                _fill_limit_area_1d(npmask, limit_area)
+            if meth == "pad":
+                indexer = libalgos.get_fill_indexer(npmask, limit=limit)
+                return self.take(indexer, allow_fill=True)
+            else:
+                # i.e. meth == "backfill"
+                indexer = libalgos.get_fill_indexer(npmask[::-1], limit=limit)[::-1]
+                return self[::-1].take(indexer, allow_fill=True)
+
+        else:
+            if not copy:
+                return self
+            new_values = self.copy()
+        return new_values
+
+    def fillna(
+        self,
+        value: object | ArrayLike,
+        limit: int | None = None,
+        copy: bool = True,
+    ) -> Self:
+        """
+        Fill NA/NaN values using the specified method.
+
+        Parameters
+        ----------
+        value : scalar, array-like
+            If a scalar value is passed it is used to fill all missing values.
+            Alternatively, an array-like "value" can be given. It's expected
+            that the array-like have the same length as 'self'.
+        limit : int, default None
+            The maximum number of entries where NA values will be filled.
+        copy : bool, default True
+            Whether to make a copy of the data before filling. If False, then
+            the original should be modified and no new memory should be allocated.
+            For ExtensionArray subclasses that cannot do this, it is at the
+            author's discretion whether to ignore "copy=False" or to raise.
+
+        Returns
+        -------
+        ExtensionArray
+            With NA/NaN filled.
+
+        See Also
+        --------
+        api.extensions.ExtensionArray.dropna : Return ExtensionArray without
+            NA values.
+        api.extensions.ExtensionArray.isna : A 1-D array indicating if
+            each value is missing.
+
+        Examples
+        --------
+        >>> arr = pd.array([np.nan, np.nan, 2, 3, np.nan, np.nan])
+        >>> arr.fillna(0)
+        <IntegerArray>
+        [0, 0, 2, 3, 0, 0]
+        Length: 6, dtype: Int64
+        """
+        mask = self.isna()
+        if limit is not None and limit < len(self):
+            # isna can return an ExtensionArray, we're assuming that comparisons
+            # are implemented.
+            # mypy doesn't like that mask can be an EA which need not have `cumsum`
+            modify = mask.cumsum() > limit  # type: ignore[union-attr]
+            if modify.any():
+                # Only copy mask if necessary
+                mask = mask.copy()
+                mask[modify] = False
+        # error: Argument 2 to "check_value_size" has incompatible type
+        # "ExtensionArray"; expected "ndarray"
+        value = missing.check_value_size(
+            value,
+            mask,  # type: ignore[arg-type]
+            len(self),
+        )
+
+        if mask.any():
+            # fill with value
+            if not copy:
+                new_values = self[:]
+            else:
+                new_values = self.copy()
+            new_values[mask] = value
+        else:
+            if not copy:
+                new_values = self[:]
+            else:
+                new_values = self.copy()
+        return new_values
+
+    def dropna(self) -> Self:
+        """
+        Return ExtensionArray without NA values.
+
+        Returns
+        -------
+        Self
+            An ExtensionArray of the same type as the original but with all
+            NA values removed.
+
+        See Also
+        --------
+        Series.dropna : Remove missing values from a Series.
+        DataFrame.dropna : Remove missing values from a DataFrame.
+        api.extensions.ExtensionArray.isna : Check for missing values in
+            an ExtensionArray.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, np.nan]).dropna()
+        <IntegerArray>
+        [1, 2]
+        Length: 2, dtype: Int64
+        """
+        # error: Unsupported operand type for ~ ("ExtensionArray")
+        return self[~self.isna()]  # type: ignore[operator]
+
+    def duplicated(
+        self, keep: Literal["first", "last", False] = "first"
+    ) -> npt.NDArray[np.bool_]:
+        """
+        Return boolean ndarray denoting duplicate values.
+
+        Parameters
+        ----------
+        keep : {'first', 'last', False}, default 'first'
+            - ``first`` : Mark duplicates as ``True`` except for the first occurrence.
+            - ``last`` : Mark duplicates as ``True`` except for the last occurrence.
+            - False : Mark all duplicates as ``True``.
+
+        Returns
+        -------
+        ndarray[bool]
+            With true in indices where elements are duplicated and false otherwise.
+
+        See Also
+        --------
+        DataFrame.duplicated : Return boolean Series denoting
+            duplicate rows.
+        Series.duplicated : Indicate duplicate Series values.
+        api.extensions.ExtensionArray.unique : Compute the ExtensionArray
+            of unique values.
+
+        Examples
+        --------
+        >>> pd.array([1, 1, 2, 3, 3], dtype="Int64").duplicated()
+        array([False,  True, False, False,  True])
+        """
+        mask = self.isna().astype(np.bool_, copy=False)
+        return duplicated(values=self, keep=keep, mask=mask)
+
+    def shift(self, periods: int = 1, fill_value: object = None) -> ExtensionArray:
+        """
+        Shift values by desired number.
+
+        Newly introduced missing values are filled with
+        ``self.dtype.na_value``.
+
+        Parameters
+        ----------
+        periods : int, default 1
+            The number of periods to shift. Negative values are allowed
+            for shifting backwards.
+
+        fill_value : object, optional
+            The scalar value to use for newly introduced missing values.
+            The default is ``self.dtype.na_value``.
+
+        Returns
+        -------
+        ExtensionArray
+            Shifted.
+
+        See Also
+        --------
+        api.extensions.ExtensionArray.transpose : Return a transposed view on
+            this array.
+        api.extensions.ExtensionArray.factorize : Encode the extension array as an
+            enumerated type.
+
+        Notes
+        -----
+        If ``self`` is empty or ``periods`` is 0, a copy of ``self`` is
+        returned.
+
+        If ``periods > len(self)``, then an array of size
+        len(self) is returned, with all values filled with
+        ``self.dtype.na_value``.
+
+        For 2-dimensional ExtensionArrays, we are always shifting along axis=0.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.shift(2)
+        <IntegerArray>
+        [<NA>, <NA>, 1]
+        Length: 3, dtype: Int64
+        """
+        # Note: this implementation assumes that `self.dtype.na_value` can be
+        # stored in an instance of your ExtensionArray with `self.dtype`.
+        if not len(self) or periods == 0:
+            return self.copy()
+
+        if isna(fill_value):
+            fill_value = self.dtype.na_value
+
+        empty = self._from_sequence(
+            [fill_value] * min(abs(periods), len(self)), dtype=self.dtype
+        )
+        if periods > 0:
+            a = empty
+            b = self[:-periods]
+        else:
+            a = self[abs(periods) :]
+            b = empty
+        return self._concat_same_type([a, b])
+
+    def unique(self) -> Self:
+        """
+        Compute the ExtensionArray of unique values.
+
+        Returns
+        -------
+        pandas.api.extensions.ExtensionArray
+            With unique values from the input array.
+
+        See Also
+        --------
+        Index.unique: Return unique values in the index.
+        Series.unique: Return unique values of Series object.
+        unique: Return unique values based on a hash table.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3, 1, 2, 3])
+        >>> arr.unique()
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        uniques = unique(self.astype(object))
+        return self._from_sequence(uniques, dtype=self.dtype)
+
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        """
+        Find indices where elements should be inserted to maintain order.
+
+        Find the indices into a sorted array `self` (a) such that, if the
+        corresponding elements in `value` were inserted before the indices,
+        the order of `self` would be preserved.
+
+        Assuming that `self` is sorted:
+
+        ======  ================================
+        `side`  returned index `i` satisfies
+        ======  ================================
+        left    ``self[i-1] < value <= self[i]``
+        right   ``self[i-1] <= value < self[i]``
+        ======  ================================
+
+        Parameters
+        ----------
+        value : array-like, list or scalar
+            Value(s) to insert into `self`.
+        side : {'left', 'right'}, optional
+            If 'left', the index of the first suitable location found is given.
+            If 'right', return the last such index.  If there is no suitable
+            index, return either 0 or N (where N is the length of `self`).
+        sorter : 1-D array-like, optional
+            Optional array of integer indices that sort array a into ascending
+            order. They are typically the result of argsort.
+
+        Returns
+        -------
+        array of ints or int
+            If value is array-like, array of insertion points.
+            If value is scalar, a single integer.
+
+        See Also
+        --------
+        numpy.searchsorted : Similar method from NumPy.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3, 5])
+        >>> arr.searchsorted([4])
+        array([3])
+        """
+        # Note: the base tests provided by pandas only test the basics.
+        # We do not test
+        # 1. Values outside the range of the `data_for_sorting` fixture
+        # 2. Values between the values in the `data_for_sorting` fixture
+        # 3. Missing values.
+        arr = self.astype(object)
+        if isinstance(value, ExtensionArray):
+            value = value.astype(object)
+        return arr.searchsorted(value, side=side, sorter=sorter)
+
+    def equals(self, other: object) -> bool:
+        """
+        Return if another array is equivalent to this array.
+
+        Equivalent means that both arrays have the same shape and dtype, and
+        all values compare equal. Missing values in the same location are
+        considered equal (in contrast with normal equality).
+
+        Parameters
+        ----------
+        other : ExtensionArray
+            Array to compare to this Array.
+
+        Returns
+        -------
+        boolean
+            Whether the arrays are equivalent.
+
+        See Also
+        --------
+        numpy.array_equal : Equivalent method for numpy array.
+        Series.equals : Equivalent method for Series.
+        DataFrame.equals : Equivalent method for DataFrame.
+
+        Examples
+        --------
+        >>> arr1 = pd.array([1, 2, np.nan])
+        >>> arr2 = pd.array([1, 2, np.nan])
+        >>> arr1.equals(arr2)
+        True
+
+        >>> arr1 = pd.array([1, 3, np.nan])
+        >>> arr2 = pd.array([1, 2, np.nan])
+        >>> arr1.equals(arr2)
+        False
+        """
+        if type(self) != type(other):
+            return False
+        other = cast(ExtensionArray, other)
+        if self.dtype != other.dtype:
+            return False
+        elif len(self) != len(other):
+            return False
+        else:
+            equal_values = self == other
+            if isinstance(equal_values, ExtensionArray):
+                # boolean array with NA -> fill with False
+                equal_values = equal_values.fillna(False)
+            # error: Unsupported left operand type for & ("ExtensionArray")
+            equal_na = self.isna() & other.isna()  # type: ignore[operator]
+            return bool((equal_values | equal_na).all())
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        """
+        Pointwise comparison for set containment in the given values.
+
+        Roughly equivalent to `np.array([x in values for x in self])`
+
+        Parameters
+        ----------
+        values : np.ndarray or ExtensionArray
+            Values to compare every element in the array against.
+
+        Returns
+        -------
+        np.ndarray[bool]
+            With true at indices where value is in `values`.
+
+        See Also
+        --------
+        DataFrame.isin: Whether each element in the DataFrame is contained in values.
+        Index.isin: Return a boolean array where the index values are in values.
+        Series.isin: Whether elements in Series are contained in values.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.isin([1])
+        <BooleanArray>
+        [True, False, False]
+        Length: 3, dtype: boolean
+        """
+        return isin(np.asarray(self), values)
+
+    def _values_for_factorize(self) -> tuple[np.ndarray, Any]:
+        """
+        Return an array and missing value suitable for factorization.
+
+        Returns
+        -------
+        values : ndarray
+            An array suitable for factorization. This should maintain order
+            and be a supported dtype (Float64, Int64, UInt64, String, Object).
+            By default, the extension array is cast to object dtype.
+        na_value : object
+            The value in `values` to consider missing. This will be treated
+            as NA in the factorization routines, so it will be coded as
+            `-1` and not included in `uniques`. By default,
+            ``np.nan`` is used.
+
+        See Also
+        --------
+        util.hash_pandas_object : Hash the pandas object.
+
+        Notes
+        -----
+        The values returned by this method are also used in
+        :func:`pandas.util.hash_pandas_object`. If needed, this can be
+        overridden in the ``self._hash_pandas_object()`` method.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3])._values_for_factorize()
+        (array([1, 2, 3], dtype=object), nan)
+        """
+        return self.astype(object), np.nan
+
+    def factorize(
+        self,
+        use_na_sentinel: bool = True,
+    ) -> tuple[np.ndarray, ExtensionArray]:
+        """
+        Encode the extension array as an enumerated type.
+
+        Parameters
+        ----------
+        use_na_sentinel : bool, default True
+            If True, the sentinel -1 will be used for NaN values. If False,
+            NaN values will be encoded as non-negative integers and will not drop the
+            NaN from the uniques of the values.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        codes : ndarray
+            An integer NumPy array that's an indexer into the original
+            ExtensionArray.
+        uniques : ExtensionArray
+            An ExtensionArray containing the unique values of `self`.
+
+            .. note::
+
+               uniques will *not* contain an entry for the NA value of
+               the ExtensionArray if there are any missing values present
+               in `self`.
+
+        See Also
+        --------
+        factorize : Top-level factorize method that dispatches here.
+
+        Notes
+        -----
+        :meth:`pandas.factorize` offers a `sort` keyword as well.
+
+        Examples
+        --------
+        >>> idx1 = pd.PeriodIndex(
+        ...     ["2014-01", "2014-01", "2014-02", "2014-02", "2014-03", "2014-03"],
+        ...     freq="M",
+        ... )
+        >>> arr, idx = idx1.factorize()
+        >>> arr
+        array([0, 0, 1, 1, 2, 2])
+        >>> idx
+        PeriodIndex(['2014-01', '2014-02', '2014-03'], dtype='period[M]')
+        """
+        # Implementer note: There are two ways to override the behavior of
+        # pandas.factorize
+        # 1. _values_for_factorize and _from_factorize.
+        #    Specify the values passed to pandas' internal factorization
+        #    routines, and how to convert from those values back to the
+        #    original ExtensionArray.
+        # 2. ExtensionArray.factorize.
+        #    Complete control over factorization.
+        arr, na_value = self._values_for_factorize()
+
+        codes, uniques = factorize_array(
+            arr, use_na_sentinel=use_na_sentinel, na_value=na_value
+        )
+
+        uniques_ea = self._from_factorized(uniques, self)
+        return codes, uniques_ea
+
+    _extension_array_shared_docs["repeat"] = """
+        Repeat elements of a %(klass)s.
+
+        Returns a new %(klass)s where each element of the current %(klass)s
+        is repeated consecutively a given number of times.
+
+        Parameters
+        ----------
+        repeats : int or array of ints
+            The number of repetitions for each element. This should be a
+            non-negative integer. Repeating 0 times will return an empty
+            %(klass)s.
+        axis : None
+            Must be ``None``. Has no effect but is accepted for compatibility
+            with numpy.
+
+        Returns
+        -------
+        %(klass)s
+            Newly created %(klass)s with repeated elements.
+
+        See Also
+        --------
+        Series.repeat : Equivalent function for Series.
+        Index.repeat : Equivalent function for Index.
+        numpy.repeat : Similar method for :class:`numpy.ndarray`.
+        ExtensionArray.take : Take arbitrary positions.
+
+        Examples
+        --------
+        >>> cat = pd.Categorical(['a', 'b', 'c'])
+        >>> cat
+        ['a', 'b', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> cat.repeat(2)
+        ['a', 'a', 'b', 'b', 'c', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> cat.repeat([1, 2, 3])
+        ['a', 'b', 'b', 'c', 'c', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        """
+
+    @Substitution(klass="ExtensionArray")
+    @Appender(_extension_array_shared_docs["repeat"])
+    def repeat(self, repeats: int | Sequence[int], axis: AxisInt | None = None) -> Self:
+        nv.validate_repeat((), {"axis": axis})
+        ind = np.arange(len(self)).repeat(repeats)
+        return self.take(ind)
+
+    # ------------------------------------------------------------------------
+    # Indexing methods
+    # ------------------------------------------------------------------------
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        *,
+        allow_fill: bool = False,
+        fill_value: Any = None,
+    ) -> Self:
+        """
+        Take elements from an array.
+
+        Parameters
+        ----------
+        indices : sequence of int or one-dimensional np.ndarray of int
+            Indices to be taken.
+        allow_fill : bool, default False
+            How to handle negative values in `indices`.
+
+            * False: negative values in `indices` indicate positional indices
+              from the right (the default). This is similar to
+              :func:`numpy.take`.
+
+            * True: negative values in `indices` indicate
+              missing values. These values are set to `fill_value`. Any other
+              other negative values raise a ``ValueError``.
+
+        fill_value : any, optional
+            Fill value to use for NA-indices when `allow_fill` is True.
+            This may be ``None``, in which case the default NA value for
+            the type, ``self.dtype.na_value``, is used.
+
+            For many ExtensionArrays, there will be two representations of
+            `fill_value`: a user-facing "boxed" scalar, and a low-level
+            physical NA value. `fill_value` should be the user-facing version,
+            and the implementation should handle translating that to the
+            physical version for processing the take if necessary.
+
+        Returns
+        -------
+        ExtensionArray
+            An array formed with selected `indices`.
+
+        Raises
+        ------
+        IndexError
+            When the indices are out of bounds for the array.
+        ValueError
+            When `indices` contains negative values other than ``-1``
+            and `allow_fill` is True.
+
+        See Also
+        --------
+        numpy.take : Take elements from an array along an axis.
+        api.extensions.take : Take elements from an array.
+
+        Notes
+        -----
+        ExtensionArray.take is called by ``Series.__getitem__``, ``.loc``,
+        ``iloc``, when `indices` is a sequence of values. Additionally,
+        it's called by :meth:`Series.reindex`, or any other method
+        that causes realignment, with a `fill_value`.
+
+        Examples
+        --------
+        Here's an example implementation, which relies on casting the
+        extension array to object dtype. This uses the helper method
+        :func:`pandas.api.extensions.take`.
+
+        .. code-block:: python
+
+           def take(self, indices, allow_fill=False, fill_value=None):
+               from pandas.core.algorithms import take
+
+               # If the ExtensionArray is backed by an ndarray, then
+               # just pass that here instead of coercing to object.
+               data = self.astype(object)
+
+               if allow_fill and fill_value is None:
+                   fill_value = self.dtype.na_value
+
+               # fill value should always be translated from the scalar
+               # type for the array, to the physical storage type for
+               # the data, before passing to take.
+
+               result = take(
+                   data, indices, fill_value=fill_value, allow_fill=allow_fill
+               )
+               return self._from_sequence(result, dtype=self.dtype)
+        """
+        # Implementer note: The `fill_value` parameter should be a user-facing
+        # value, an instance of self.dtype.type. When passed `fill_value=None`,
+        # the default of `self.dtype.na_value` should be used.
+        # This may differ from the physical storage type your ExtensionArray
+        # uses. In this case, your implementation is responsible for casting
+        # the user-facing type to the storage type, before using
+        # pandas.api.extensions.take
+        raise AbstractMethodError(self)
+
+    def copy(self) -> Self:
+        """
+        Return a copy of the array.
+
+        This method creates a copy of the `ExtensionArray` where modifying the
+        data in the copy will not affect the original array. This is useful when
+        you want to manipulate data without altering the original dataset.
+
+        Returns
+        -------
+        ExtensionArray
+            A new `ExtensionArray` object that is a copy of the current instance.
+
+        See Also
+        --------
+        DataFrame.copy : Return a copy of the DataFrame.
+        Series.copy : Return a copy of the Series.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr2 = arr.copy()
+        >>> arr[0] = 2
+        >>> arr2
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        raise AbstractMethodError(self)
+
+    def view(self, dtype: Dtype | None = None) -> ArrayLike:
+        """
+        Return a view on the array.
+
+        Parameters
+        ----------
+        dtype : str, np.dtype, or ExtensionDtype, optional
+            Default None.
+
+        Returns
+        -------
+        ExtensionArray or np.ndarray
+            A view on the :class:`ExtensionArray`'s data.
+
+        See Also
+        --------
+        api.extensions.ExtensionArray.ravel: Return a flattened view on input array.
+        Index.view: Equivalent function for Index.
+        ndarray.view: New view of array with the same data.
+
+        Examples
+        --------
+        This gives view on the underlying data of an ``ExtensionArray`` and is not a
+        copy. Modifications on either the view or the original ``ExtensionArray``
+        will be reflected on the underlying data:
+
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr2 = arr.view()
+        >>> arr[0] = 2
+        >>> arr2
+        <IntegerArray>
+        [2, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        # NB:
+        # - This must return a *new* object referencing the same data, not self.
+        # - The only case that *must* be implemented is with dtype=None,
+        #   giving a view with the same dtype as self.
+        if dtype is not None:
+            raise NotImplementedError(dtype)
+        return self[:]
+
+    # ------------------------------------------------------------------------
+    # Printing
+    # ------------------------------------------------------------------------
+
+    def __repr__(self) -> str:
+        if self.ndim > 1:
+            return self._repr_2d()
+
+        from pandas.io.formats.printing import format_object_summary
+
+        # the short repr has no trailing newline, while the truncated
+        # repr does. So we include a newline in our template, and strip
+        # any trailing newlines from format_object_summary
+        data = format_object_summary(
+            self, self._formatter(), indent_for_name=False
+        ).rstrip(", \n")
+        class_name = f"<{type(self).__name__}>\n"
+        footer = self._get_repr_footer()
+        return f"{class_name}{data}\n{footer}"
+
+    def _get_repr_footer(self) -> str:
+        # GH#24278
+        if self.ndim > 1:
+            return f"Shape: {self.shape}, dtype: {self.dtype}"
+        return f"Length: {len(self)}, dtype: {self.dtype}"
+
+    def _repr_2d(self) -> str:
+        from pandas.io.formats.printing import format_object_summary
+
+        # the short repr has no trailing newline, while the truncated
+        # repr does. So we include a newline in our template, and strip
+        # any trailing newlines from format_object_summary
+        lines = [
+            format_object_summary(x, self._formatter(), indent_for_name=False).rstrip(
+                ", \n"
+            )
+            for x in self
+        ]
+        data = ",\n".join(lines)
+        class_name = f"<{type(self).__name__}>"
+        footer = self._get_repr_footer()
+        return f"{class_name}\n[\n{data}\n]\n{footer}"
+
+    def _formatter(self, boxed: bool = False) -> Callable[[Any], str | None]:
+        """
+        Formatting function for scalar values.
+
+        This is used in the default '__repr__'. The returned formatting
+        function receives instances of your scalar type.
+
+        Parameters
+        ----------
+        boxed : bool, default False
+            An indicated for whether or not your array is being printed
+            within a Series, DataFrame, or Index (True), or just by
+            itself (False). This may be useful if you want scalar values
+            to appear differently within a Series versus on its own (e.g.
+            quoted or not).
+
+        Returns
+        -------
+        Callable[[Any], str]
+            A callable that gets instances of the scalar type and
+            returns a string. By default, :func:`repr` is used
+            when ``boxed=False`` and :func:`str` is used when
+            ``boxed=True``.
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._concat_same_type : Concatenate multiple
+            array of this dtype.
+        api.extensions.ExtensionArray._explode : Transform each element of
+            list-like to a row.
+        api.extensions.ExtensionArray._from_factorized : Reconstruct an
+            ExtensionArray after factorization.
+        api.extensions.ExtensionArray._from_sequence : Construct a new
+            ExtensionArray from a sequence of scalars.
+
+        Examples
+        --------
+        >>> class MyExtensionArray(pd.arrays.NumpyExtensionArray):
+        ...     def _formatter(self, boxed=False):
+        ...         return lambda x: "*" + str(x) + "*" if boxed else repr(x) + "*"
+        >>> MyExtensionArray(np.array([1, 2, 3, 4]))
+        <MyExtensionArray>
+        [1*, 2*, 3*, 4*]
+        Length: 4, dtype: int64
+        """
+        if boxed:
+            return str
+        return repr
+
+    # ------------------------------------------------------------------------
+    # Reshaping
+    # ------------------------------------------------------------------------
+
+    def transpose(self, *axes: int) -> Self:
+        """
+        Return a transposed view on this array.
+
+        Because ExtensionArrays are always 1D, this is a no-op.  It is included
+        for compatibility with np.ndarray.
+
+        Returns
+        -------
+        ExtensionArray
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3]).transpose()
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        return self[:]
+
+    @property
+    def T(self) -> Self:
+        return self.transpose()
+
+    def ravel(self, order: Literal["C", "F", "A", "K"] | None = "C") -> Self:
+        """
+        Return a flattened view on this array.
+
+        Parameters
+        ----------
+        order : {None, 'C', 'F', 'A', 'K'}, default 'C'
+
+        Returns
+        -------
+        ExtensionArray
+            A flattened view on the array.
+
+        See Also
+        --------
+        ExtensionArray.tolist: Return a list of the values.
+
+        Notes
+        -----
+        - Because ExtensionArrays are 1D-only, this is a no-op.
+        - The "order" argument is ignored, is for compatibility with NumPy.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3]).ravel()
+        <IntegerArray>
+        [1, 2, 3]
+        Length: 3, dtype: Int64
+        """
+        return self
+
+    @classmethod
+    def _concat_same_type(cls, to_concat: Sequence[Self]) -> Self:
+        """
+        Concatenate multiple array of this dtype.
+
+        Parameters
+        ----------
+        to_concat : sequence of this type
+            An array of the same dtype to concatenate.
+
+        Returns
+        -------
+        ExtensionArray
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._explode : Transform each element of
+            list-like to a row.
+        api.extensions.ExtensionArray._formatter : Formatting function for
+            scalar values.
+        api.extensions.ExtensionArray._from_factorized : Reconstruct an
+            ExtensionArray after factorization.
+
+        Examples
+        --------
+        >>> arr1 = pd.array([1, 2, 3])
+        >>> arr2 = pd.array([4, 5, 6])
+        >>> pd.arrays.IntegerArray._concat_same_type([arr1, arr2])
+        <IntegerArray>
+        [1, 2, 3, 4, 5, 6]
+        Length: 6, dtype: Int64
+        """
+        # Implementer note: this method will only be called with a sequence of
+        # ExtensionArrays of this class and with the same dtype as self. This
+        # should allow "easy" concatenation (no upcasting needed), and result
+        # in a new ExtensionArray of the same dtype.
+        # Note: this strict behaviour is only guaranteed starting with pandas 1.1
+        raise AbstractMethodError(cls)
+
+    # The _can_hold_na attribute is set to True so that pandas internals
+    # will use the ExtensionDtype.na_value as the NA value in operations
+    # such as take(), reindex(), shift(), etc.  In addition, those results
+    # will then be of the ExtensionArray subclass rather than an array
+    # of objects
+    @cache_readonly
+    def _can_hold_na(self) -> bool:
+        return self.dtype._can_hold_na
+
+    def _accumulate(
+        self, name: str, *, skipna: bool = True, **kwargs
+    ) -> ExtensionArray:
+        """
+        Return an ExtensionArray performing an accumulation operation.
+
+        The underlying data type might change.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            - cummin
+            - cummax
+            - cumsum
+            - cumprod
+        skipna : bool, default True
+            If True, skip NA values.
+        **kwargs
+            Additional keyword arguments passed to the accumulation function.
+            Currently, there is no supported kwarg.
+
+        Returns
+        -------
+        array
+            An array performing the accumulation operation.
+
+        Raises
+        ------
+        NotImplementedError : subclass does not define accumulations
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._concat_same_type : Concatenate multiple
+            array of this dtype.
+        api.extensions.ExtensionArray.view : Return a view on the array.
+        api.extensions.ExtensionArray._explode : Transform each element of
+            list-like to a row.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr._accumulate(name="cumsum")
+        <IntegerArray>
+        [1, 3, 6]
+        Length: 3, dtype: Int64
+        """
+        raise NotImplementedError(f"cannot perform {name} with type {self.dtype}")
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        """
+        Return a scalar result of performing the reduction operation.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            { any, all, min, max, sum, mean, median, prod,
+            std, var, sem, kurt, skew }.
+        skipna : bool, default True
+            If True, skip NaN values.
+        keepdims : bool, default False
+            If False, a scalar is returned.
+            If True, the result has dimension with size one along the reduced axis.
+        **kwargs
+            Additional keyword arguments passed to the reduction function.
+            Currently, `ddof` is the only supported kwarg.
+
+        Returns
+        -------
+        scalar or ndarray:
+            The result of the reduction operation. The type of the result
+            depends on `keepdims`:
+            - If `keepdims` is `False`, a scalar value is returned.
+            - If `keepdims` is `True`, the result is wrapped in a numpy array with
+            a single element.
+
+        Raises
+        ------
+        TypeError : subclass does not define operations
+
+        See Also
+        --------
+        Series.min : Return the minimum value.
+        Series.max : Return the maximum value.
+        Series.sum : Return the sum of values.
+        Series.mean : Return the mean of values.
+        Series.median : Return the median of values.
+        Series.std : Return the standard deviation.
+        Series.var : Return the variance.
+        Series.prod : Return the product of values.
+        Series.sem : Return the standard error of the mean.
+        Series.kurt : Return the kurtosis.
+        Series.skew : Return the skewness.
+
+        Examples
+        --------
+        >>> pd.array([1, 2, 3])._reduce("min")
+        1
+        >>> pd.array([1, 2, 3])._reduce("max")
+        3
+        >>> pd.array([1, 2, 3])._reduce("sum")
+        6
+        >>> pd.array([1, 2, 3])._reduce("mean")
+        2.0
+        >>> pd.array([1, 2, 3])._reduce("median")
+        2.0
+        """
+        meth = getattr(self, name, None)
+        if meth is None:
+            raise TypeError(
+                f"'{type(self).__name__}' with dtype {self.dtype} "
+                f"does not support operation '{name}'"
+            )
+        result = meth(skipna=skipna, **kwargs)
+        if keepdims:
+            if name in ["min", "max"]:
+                result = self._from_sequence([result], dtype=self.dtype)
+            else:
+                result = np.array([result])
+
+        return result
+
+    # https://github.com/python/typeshed/issues/2148#issuecomment-520783318
+    # Incompatible types in assignment (expression has type "None", base class
+    # "object" defined the type as "Callable[[object], int]")
+    __hash__: ClassVar[None]  # type: ignore[assignment]
+
+    # ------------------------------------------------------------------------
+    # Non-Optimized Default Methods; in the case of the private methods here,
+    #  these are not guaranteed to be stable across pandas versions.
+
+    def _values_for_json(self) -> np.ndarray:
+        """
+        Specify how to render our entries in to_json.
+
+        Notes
+        -----
+        The dtype on the returned ndarray is not restricted, but for non-native
+        types that are not specifically handled in objToJSON.c, to_json is
+        liable to raise. In these cases, it may be safer to return an ndarray
+        of strings.
+        """
+        return np.asarray(self)
+
+    def _hash_pandas_object(
+        self, *, encoding: str, hash_key: str, categorize: bool
+    ) -> npt.NDArray[np.uint64]:
+        """
+        Hook for hash_pandas_object.
+
+        Default is to use the values returned by _values_for_factorize.
+
+        Parameters
+        ----------
+        encoding : str
+            Encoding for data & key when strings.
+        hash_key : str
+            Hash_key for string key to encode.
+        categorize : bool
+            Whether to first categorize object arrays before hashing. This is more
+            efficient when the array contains duplicate values.
+
+        Returns
+        -------
+        np.ndarray[uint64]
+            An array of hashed values.
+
+        See Also
+        --------
+        api.extensions.ExtensionArray._values_for_factorize : Return an array and
+            missing value suitable for factorization.
+        util.hash_array : Given a 1d array, return an array of hashed values.
+
+        Examples
+        --------
+        >>> pd.array([1, 2])._hash_pandas_object(
+        ...     encoding="utf-8", hash_key="1000000000000000", categorize=False
+        ... )
+        array([ 6238072747940578789, 15839785061582574730], dtype=uint64)
+        """
+        from pandas.core.util.hashing import hash_array
+
+        values, _ = self._values_for_factorize()
+        return hash_array(
+            values, encoding=encoding, hash_key=hash_key, categorize=categorize
+        )
+
+    def _explode(self) -> tuple[Self, npt.NDArray[np.uint64]]:
+        """
+        Transform each element of list-like to a row.
+
+        For arrays that do not contain list-like elements the default
+        implementation of this method just returns a copy and an array
+        of ones (unchanged index).
+
+        Returns
+        -------
+        ExtensionArray
+            Array with the exploded values.
+        np.ndarray[uint64]
+            The original lengths of each list-like for determining the
+            resulting index.
+
+        See Also
+        --------
+        Series.explode : The method on the ``Series`` object that this
+            extension array method is meant to support.
+
+        Examples
+        --------
+        >>> import pyarrow as pa
+        >>> a = pd.array(
+        ...     [[1, 2, 3], [4], [5, 6]], dtype=pd.ArrowDtype(pa.list_(pa.int64()))
+        ... )
+        >>> a._explode()
+        (<ArrowExtensionArray>
+        [1, 2, 3, 4, 5, 6]
+        Length: 6, dtype: int64[pyarrow], array([3, 1, 2], dtype=int32))
+        """
+        values = self.copy()
+        counts = np.ones(shape=(len(self),), dtype=np.uint64)
+        return values, counts
+
+    def tolist(self) -> list:
+        """
+        Return a list of the values.
+
+        These are each a scalar type, which is a Python scalar
+        (for str, int, float) or a pandas scalar
+        (for Timestamp/Timedelta/Interval/Period)
+
+        Returns
+        -------
+        list
+            Python list of values in array.
+
+        See Also
+        --------
+        Index.to_list: Return a list of the values in the Index.
+        Series.to_list: Return a list of the values in the Series.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.tolist()
+        [1, 2, 3]
+        """
+        if self.ndim > 1:
+            return [x.tolist() for x in self]
+        return list(self)
+
+    def delete(self, loc: PositionalIndexer) -> Self:
+        indexer = np.delete(np.arange(len(self)), loc)
+        return self.take(indexer)
+
+    def insert(self, loc: int, item) -> Self:
+        """
+        Insert an item at the given position.
+
+        Parameters
+        ----------
+        loc : int
+            Index where the `item` needs to be inserted.
+        item : scalar-like
+            Value to be inserted.
+
+        Returns
+        -------
+        ExtensionArray
+            With `item` inserted at `loc`.
+
+        See Also
+        --------
+        Index.insert: Make new Index inserting new item at location.
+
+        Notes
+        -----
+        This method should be both type and dtype-preserving.  If the item
+        cannot be held in an array of this type/dtype, either ValueError or
+        TypeError should be raised.
+
+        The default implementation relies on _from_sequence to raise on invalid
+        items.
+
+        Examples
+        --------
+        >>> arr = pd.array([1, 2, 3])
+        >>> arr.insert(2, -1)
+        <IntegerArray>
+        [1, 2, -1, 3]
+        Length: 4, dtype: Int64
+        """
+        loc = validate_insert_loc(loc, len(self))
+
+        item_arr = type(self)._from_sequence([item], dtype=self.dtype)
+
+        return type(self)._concat_same_type([self[:loc], item_arr, self[loc:]])
+
+    def _putmask(self, mask: npt.NDArray[np.bool_], value) -> None:
+        """
+        Analogue to np.putmask(self, mask, value)
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+        value : scalar or listlike
+            If listlike, must be arraylike with same length as self.
+
+        Returns
+        -------
+        None
+
+        Notes
+        -----
+        Unlike np.putmask, we do not repeat listlike values with mismatched length.
+        'value' should either be a scalar or an arraylike with the same length
+        as self.
+        """
+        if is_list_like(value):
+            val = value[mask]
+        else:
+            val = value
+
+        self[mask] = val
+
+    def _where(self, mask: npt.NDArray[np.bool_], value) -> Self:
+        """
+        Analogue to np.where(mask, self, value)
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+        value : scalar or listlike
+
+        Returns
+        -------
+        same type as self
+        """
+        result = self.copy()
+
+        if is_list_like(value):
+            val = value[~mask]
+        else:
+            val = value
+
+        result[~mask] = val
+        return result
+
+    def _rank(
+        self,
+        *,
+        axis: AxisInt = 0,
+        method: str = "average",
+        na_option: str = "keep",
+        ascending: bool = True,
+        pct: bool = False,
+    ):
+        """
+        See Series.rank.__doc__.
+        """
+        if axis != 0:
+            raise NotImplementedError
+
+        return rank(
+            self,
+            axis=axis,
+            method=method,
+            na_option=na_option,
+            ascending=ascending,
+            pct=pct,
+        )
+
+    @classmethod
+    def _empty(cls, shape: Shape, dtype: ExtensionDtype):
+        """
+        Create an ExtensionArray with the given shape and dtype.
+
+        See also
+        --------
+        ExtensionDtype.empty
+            ExtensionDtype.empty is the 'official' public version of this API.
+        """
+        # Implementer note: while ExtensionDtype.empty is the public way to
+        # call this method, it is still required to implement this `_empty`
+        # method as well (it is called internally in pandas)
+        obj = cls._from_sequence([], dtype=dtype)
+
+        taker = np.broadcast_to(np.intp(-1), shape)
+        result = obj.take(taker, allow_fill=True)
+        if not isinstance(result, cls) or dtype != result.dtype:
+            raise NotImplementedError(
+                f"Default 'empty' implementation is invalid for dtype='{dtype}'"
+            )
+        return result
+
+    def _quantile(self, qs: npt.NDArray[np.float64], interpolation: str) -> Self:
+        """
+        Compute the quantiles of self for each quantile in `qs`.
+
+        Parameters
+        ----------
+        qs : np.ndarray[float64]
+        interpolation: str
+
+        Returns
+        -------
+        same type as self
+        """
+        mask = np.asarray(self.isna())
+        arr = np.asarray(self)
+        fill_value = np.nan
+
+        res_values = quantile_with_mask(arr, mask, fill_value, qs, interpolation)
+        return type(self)._from_sequence(res_values)
+
+    def _mode(self, dropna: bool = True) -> Self:
+        """
+        Returns the mode(s) of the ExtensionArray.
+
+        Always returns `ExtensionArray` even if only one value.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't consider counts of NA values.
+
+        Returns
+        -------
+        same type as self
+            Sorted, if possible.
+        """
+        # error: Incompatible return value type (got "Union[ExtensionArray,
+        # ndarray[Any, Any]]", expected "Self")
+        return mode(self, dropna=dropna)  # type: ignore[return-value]
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        if any(
+            isinstance(other, (ABCSeries, ABCIndex, ABCDataFrame)) for other in inputs
+        ):
+            return NotImplemented
+
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                return result
+
+        return arraylike.default_array_ufunc(self, ufunc, method, *inputs, **kwargs)
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        """
+        Map values using an input mapping or function.
+
+        Parameters
+        ----------
+        mapper : function, dict, or Series
+            Mapping correspondence.
+        na_action : {None, 'ignore'}, default None
+            If 'ignore', propagate NA values, without passing them to the
+            mapping correspondence. If 'ignore' is not supported, a
+            ``NotImplementedError`` should be raised.
+
+        Returns
+        -------
+        Union[ndarray, Index, ExtensionArray]
+            The output of the mapping function applied to the array.
+            If the function returns a tuple with more than one element
+            a MultiIndex will be returned.
+        """
+        return map_array(self, mapper, na_action=na_action)
+
+    # ------------------------------------------------------------------------
+    # GroupBy Methods
+
+    def _groupby_op(
+        self,
+        *,
+        how: str,
+        has_dropped_na: bool,
+        min_count: int,
+        ngroups: int,
+        ids: npt.NDArray[np.intp],
+        **kwargs,
+    ) -> ArrayLike:
+        """
+        Dispatch GroupBy reduction or transformation operation.
+
+        This is an *experimental* API to allow ExtensionArray authors to implement
+        reductions and transformations. The API is subject to change.
+
+        Parameters
+        ----------
+        how : {'any', 'all', 'sum', 'prod', 'min', 'max', 'mean', 'median',
+               'median', 'var', 'std', 'sem', 'nth', 'last', 'ohlc',
+               'cumprod', 'cumsum', 'cummin', 'cummax', 'rank'}
+        has_dropped_na : bool
+        min_count : int
+        ngroups : int
+        ids : np.ndarray[np.intp]
+            ids[i] gives the integer label for the group that self[i] belongs to.
+        **kwargs : operation-specific
+            'any', 'all' -> ['skipna']
+            'var', 'std', 'sem' -> ['ddof']
+            'cumprod', 'cumsum', 'cummin', 'cummax' -> ['skipna']
+            'rank' -> ['ties_method', 'ascending', 'na_option', 'pct']
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray
+        """
+        from pandas.core.arrays.string_ import StringDtype
+        from pandas.core.groupby.ops import WrappedCythonOp
+
+        kind = WrappedCythonOp.get_kind_from_how(how)
+        op = WrappedCythonOp(how=how, kind=kind, has_dropped_na=has_dropped_na)
+
+        # GH#43682
+        if isinstance(self.dtype, StringDtype):
+            # StringArray
+            if op.how in [
+                "prod",
+                "mean",
+                "median",
+                "cumsum",
+                "cumprod",
+                "std",
+                "sem",
+                "var",
+                "skew",
+                "kurt",
+            ]:
+                raise TypeError(
+                    f"dtype '{self.dtype}' does not support operation '{how}'"
+                )
+            if op.how not in ["any", "all"]:
+                # Fail early to avoid conversion to object
+                op._get_cython_function(op.kind, op.how, np.dtype(object), False)
+            npvalues = self.to_numpy(object, na_value=np.nan)
+        else:
+            raise NotImplementedError(
+                f"function is not implemented for this dtype: {self.dtype}"
+            )
+
+        res_values = op._cython_op_ndim_compat(
+            npvalues,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=ids,
+            mask=None,
+            **kwargs,
+        )
+
+        if op.how in op.cast_blocklist:
+            # i.e. how in ["rank"], since other cast_blocklist methods don't go
+            #  through cython_operation
+            return res_values
+
+        if isinstance(self.dtype, StringDtype):
+            dtype = self.dtype
+            string_array_cls = dtype.construct_array_type()
+            return string_array_cls._from_sequence(res_values, dtype=dtype)
+
+        else:
+            raise NotImplementedError
+
+
+class ExtensionArraySupportsAnyAll(ExtensionArray):
+    @overload
+    def any(self, *, skipna: Literal[True] = ...) -> bool: ...
+
+    @overload
+    def any(self, *, skipna: bool) -> bool | NAType: ...
+
+    def any(self, *, skipna: bool = True) -> bool | NAType:
+        raise AbstractMethodError(self)
+
+    @overload
+    def all(self, *, skipna: Literal[True] = ...) -> bool: ...
+
+    @overload
+    def all(self, *, skipna: bool) -> bool | NAType: ...
+
+    def all(self, *, skipna: bool = True) -> bool | NAType:
+        raise AbstractMethodError(self)
+
+
+class ExtensionOpsMixin:
+    """
+    A base class for linking the operators to their dunder names.
+
+    .. note::
+
+       You may want to set ``__array_priority__`` if you want your
+       implementation to be called when involved in binary operations
+       with NumPy arrays.
+    """
+
+    @classmethod
+    def _create_arithmetic_method(cls, op):
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _add_arithmetic_ops(cls) -> None:
+        setattr(cls, "__add__", cls._create_arithmetic_method(operator.add))
+        setattr(cls, "__radd__", cls._create_arithmetic_method(roperator.radd))
+        setattr(cls, "__sub__", cls._create_arithmetic_method(operator.sub))
+        setattr(cls, "__rsub__", cls._create_arithmetic_method(roperator.rsub))
+        setattr(cls, "__mul__", cls._create_arithmetic_method(operator.mul))
+        setattr(cls, "__rmul__", cls._create_arithmetic_method(roperator.rmul))
+        setattr(cls, "__pow__", cls._create_arithmetic_method(operator.pow))
+        setattr(cls, "__rpow__", cls._create_arithmetic_method(roperator.rpow))
+        setattr(cls, "__mod__", cls._create_arithmetic_method(operator.mod))
+        setattr(cls, "__rmod__", cls._create_arithmetic_method(roperator.rmod))
+        setattr(cls, "__floordiv__", cls._create_arithmetic_method(operator.floordiv))
+        setattr(
+            cls, "__rfloordiv__", cls._create_arithmetic_method(roperator.rfloordiv)
+        )
+        setattr(cls, "__truediv__", cls._create_arithmetic_method(operator.truediv))
+        setattr(cls, "__rtruediv__", cls._create_arithmetic_method(roperator.rtruediv))
+        setattr(cls, "__divmod__", cls._create_arithmetic_method(divmod))
+        setattr(cls, "__rdivmod__", cls._create_arithmetic_method(roperator.rdivmod))
+
+    @classmethod
+    def _create_comparison_method(cls, op):
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _add_comparison_ops(cls) -> None:
+        setattr(cls, "__eq__", cls._create_comparison_method(operator.eq))
+        setattr(cls, "__ne__", cls._create_comparison_method(operator.ne))
+        setattr(cls, "__lt__", cls._create_comparison_method(operator.lt))
+        setattr(cls, "__gt__", cls._create_comparison_method(operator.gt))
+        setattr(cls, "__le__", cls._create_comparison_method(operator.le))
+        setattr(cls, "__ge__", cls._create_comparison_method(operator.ge))
+
+    @classmethod
+    def _create_logical_method(cls, op):
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _add_logical_ops(cls) -> None:
+        setattr(cls, "__and__", cls._create_logical_method(operator.and_))
+        setattr(cls, "__rand__", cls._create_logical_method(roperator.rand_))
+        setattr(cls, "__or__", cls._create_logical_method(operator.or_))
+        setattr(cls, "__ror__", cls._create_logical_method(roperator.ror_))
+        setattr(cls, "__xor__", cls._create_logical_method(operator.xor))
+        setattr(cls, "__rxor__", cls._create_logical_method(roperator.rxor))
+
+
+class ExtensionScalarOpsMixin(ExtensionOpsMixin):
+    """
+    A mixin for defining ops on an ExtensionArray.
+
+    It is assumed that the underlying scalar objects have the operators
+    already defined.
+
+    Notes
+    -----
+    If you have defined a subclass MyExtensionArray(ExtensionArray), then
+    use MyExtensionArray(ExtensionArray, ExtensionScalarOpsMixin) to
+    get the arithmetic operators.  After the definition of MyExtensionArray,
+    insert the lines
+
+    MyExtensionArray._add_arithmetic_ops()
+    MyExtensionArray._add_comparison_ops()
+
+    to link the operators to your class.
+
+    .. note::
+
+       You may want to set ``__array_priority__`` if you want your
+       implementation to be called when involved in binary operations
+       with NumPy arrays.
+    """
+
+    @classmethod
+    def _create_method(cls, op, coerce_to_dtype: bool = True, result_dtype=None):
+        """
+        A class method that returns a method that will correspond to an
+        operator for an ExtensionArray subclass, by dispatching to the
+        relevant operator defined on the individual elements of the
+        ExtensionArray.
+
+        Parameters
+        ----------
+        op : function
+            An operator that takes arguments op(a, b)
+        coerce_to_dtype : bool, default True
+            boolean indicating whether to attempt to convert
+            the result to the underlying ExtensionArray dtype.
+            If it's not possible to create a new ExtensionArray with the
+            values, an ndarray is returned instead.
+
+        Returns
+        -------
+        Callable[[Any, Any], Union[ndarray, ExtensionArray]]
+            A method that can be bound to a class. When used, the method
+            receives the two arguments, one of which is the instance of
+            this class, and should return an ExtensionArray or an ndarray.
+
+            Returning an ndarray may be necessary when the result of the
+            `op` cannot be stored in the ExtensionArray. The dtype of the
+            ndarray uses NumPy's normal inference rules.
+
+        Examples
+        --------
+        Given an ExtensionArray subclass called MyExtensionArray, use
+
+            __add__ = cls._create_method(operator.add)
+
+        in the class definition of MyExtensionArray to create the operator
+        for addition, that will be based on the operator implementation
+        of the underlying elements of the ExtensionArray
+        """
+
+        def _binop(self, other):
+            def convert_values(param):
+                if isinstance(param, ExtensionArray) or is_list_like(param):
+                    ovalues = param
+                else:  # Assume its an object
+                    ovalues = [param] * len(self)
+                return ovalues
+
+            if isinstance(other, (ABCSeries, ABCIndex, ABCDataFrame)):
+                # rely on pandas to unbox and dispatch to us
+                return NotImplemented
+
+            lvalues = self
+            rvalues = convert_values(other)
+
+            # If the operator is not defined for the underlying objects,
+            # a TypeError should be raised
+            res = [op(a, b) for (a, b) in zip(lvalues, rvalues)]
+
+            def _maybe_convert(arr):
+                if coerce_to_dtype:
+                    # https://github.com/pandas-dev/pandas/issues/22850
+                    # We catch all regular exceptions here, and fall back
+                    # to an ndarray.
+                    res = maybe_cast_pointwise_result(arr, self.dtype, same_dtype=False)
+                    if not isinstance(res, type(self)):
+                        # exception raised in _from_sequence; ensure we have ndarray
+                        res = np.asarray(arr)
+                else:
+                    res = np.asarray(arr, dtype=result_dtype)
+                return res
+
+            if op.__name__ in {"divmod", "rdivmod"}:
+                a, b = zip(*res)
+                return _maybe_convert(a), _maybe_convert(b)
+
+            return _maybe_convert(res)
+
+        op_name = f"__{op.__name__}__"
+        return set_function_name(_binop, op_name, cls)
+
+    @classmethod
+    def _create_arithmetic_method(cls, op):
+        return cls._create_method(op)
+
+    @classmethod
+    def _create_comparison_method(cls, op):
+        return cls._create_method(op, coerce_to_dtype=False, result_dtype=bool)
diff --git a/pandas/pandas/core/arrays/boolean.py b/pandas/pandas/core/arrays/boolean.py
new file mode 100644
index 0000000000000..87c18fe346c62
--- /dev/null
+++ b/pandas/pandas/core/arrays/boolean.py
@@ -0,0 +1,421 @@
+from __future__ import annotations
+
+import numbers
+from typing import (
+    TYPE_CHECKING,
+    ClassVar,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+
+from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.dtypes import register_extension_dtype
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import ops
+from pandas.core.array_algos import masked_accumulations
+from pandas.core.arrays.masked import (
+    BaseMaskedArray,
+    BaseMaskedDtype,
+)
+
+if TYPE_CHECKING:
+    import pyarrow
+
+    from pandas._typing import (
+        DtypeObj,
+        Self,
+        npt,
+        type_t,
+    )
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+
+@register_extension_dtype
+class BooleanDtype(BaseMaskedDtype):
+    """
+    Extension dtype for boolean data.
+
+    .. warning::
+
+       BooleanDtype is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    StringDtype : Extension dtype for string data.
+
+    Examples
+    --------
+    >>> pd.BooleanDtype()
+    BooleanDtype
+    """
+
+    name: ClassVar[str] = "boolean"
+
+    # The value used to fill '_data' to avoid upcasting
+    _internal_fill_value = False
+
+    # https://github.com/python/mypy/issues/4125
+    # error: Signature of "type" incompatible with supertype "BaseMaskedDtype"
+    @property
+    def type(self) -> type:  # type: ignore[override]
+        return np.bool_
+
+    @property
+    def kind(self) -> str:
+        return "b"
+
+    @property
+    def numpy_dtype(self) -> np.dtype:
+        return np.dtype("bool")
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[BooleanArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return BooleanArray
+
+    def __repr__(self) -> str:
+        return "BooleanDtype"
+
+    @property
+    def _is_boolean(self) -> bool:
+        return True
+
+    @property
+    def _is_numeric(self) -> bool:
+        return True
+
+    def __from_arrow__(
+        self, array: pyarrow.Array | pyarrow.ChunkedArray
+    ) -> BooleanArray:
+        """
+        Construct BooleanArray from pyarrow Array/ChunkedArray.
+        """
+        import pyarrow
+
+        if array.type != pyarrow.bool_() and not pyarrow.types.is_null(array.type):
+            raise TypeError(f"Expected array of boolean type, got {array.type} instead")
+
+        if isinstance(array, pyarrow.Array):
+            chunks = [array]
+            length = len(array)
+        else:
+            # pyarrow.ChunkedArray
+            chunks = array.chunks
+            length = array.length()
+
+        if pyarrow.types.is_null(array.type):
+            mask = np.ones(length, dtype=bool)
+            # No need to init data, since all null
+            data = np.empty(length, dtype=bool)
+            return BooleanArray(data, mask)
+
+        results = []
+        for arr in chunks:
+            buflist = arr.buffers()
+            data = pyarrow.BooleanArray.from_buffers(
+                arr.type, len(arr), [None, buflist[1]], offset=arr.offset
+            ).to_numpy(zero_copy_only=False)
+            if arr.null_count != 0:
+                mask = pyarrow.BooleanArray.from_buffers(
+                    arr.type, len(arr), [None, buflist[0]], offset=arr.offset
+                ).to_numpy(zero_copy_only=False)
+                mask = ~mask
+            else:
+                mask = np.zeros(len(arr), dtype=bool)
+
+            bool_arr = BooleanArray(data, mask)
+            results.append(bool_arr)
+
+        if not results:
+            return BooleanArray(
+                np.array([], dtype=np.bool_), np.array([], dtype=np.bool_)
+            )
+        else:
+            return BooleanArray._concat_same_type(results)
+
+
+def coerce_to_array(
+    values, mask=None, copy: bool = False
+) -> tuple[np.ndarray, np.ndarray]:
+    """
+    Coerce the input values array to numpy arrays with a mask.
+
+    Parameters
+    ----------
+    values : 1D list-like
+    mask : bool 1D array, optional
+    copy : bool, default False
+        if True, copy the input
+
+    Returns
+    -------
+    tuple of (values, mask)
+    """
+    if isinstance(values, BooleanArray):
+        if mask is not None:
+            raise ValueError("cannot pass mask for BooleanArray input")
+        values, mask = values._data, values._mask
+        if copy:
+            values = values.copy()
+            mask = mask.copy()
+        return values, mask
+
+    mask_values = None
+    if isinstance(values, np.ndarray) and values.dtype == np.bool_:
+        if copy:
+            values = values.copy()
+    elif isinstance(values, np.ndarray) and values.dtype.kind in "iufcb":
+        mask_values = isna(values)
+
+        values_bool = np.zeros(len(values), dtype=bool)
+        values_bool[~mask_values] = values[~mask_values].astype(bool)
+
+        if not np.all(
+            values_bool[~mask_values].astype(values.dtype) == values[~mask_values]
+        ):
+            raise TypeError("Need to pass bool-like values")
+
+        values = values_bool
+    else:
+        values_object = np.asarray(values, dtype=object)
+
+        inferred_dtype = lib.infer_dtype(values_object, skipna=True)
+        integer_like = ("floating", "integer", "mixed-integer-float")
+        if inferred_dtype not in ("boolean", "empty") + integer_like:
+            raise TypeError("Need to pass bool-like values")
+
+        # mypy does not narrow the type of mask_values to npt.NDArray[np.bool_]
+        # within this branch, it assumes it can also be None
+        mask_values = cast("npt.NDArray[np.bool_]", isna(values_object))
+        values = np.zeros(len(values), dtype=bool)
+        values[~mask_values] = values_object[~mask_values].astype(bool)
+
+        # if the values were integer-like, validate it were actually 0/1's
+        if (inferred_dtype in integer_like) and not (
+            np.all(
+                values[~mask_values].astype(float)
+                == values_object[~mask_values].astype(float)
+            )
+        ):
+            raise TypeError("Need to pass bool-like values")
+
+    if mask is None and mask_values is None:
+        mask = np.zeros(values.shape, dtype=bool)
+    elif mask is None:
+        mask = mask_values
+    else:
+        if isinstance(mask, np.ndarray) and mask.dtype == np.bool_:
+            if mask_values is not None:
+                mask = mask | mask_values
+            else:
+                if copy:
+                    mask = mask.copy()
+        else:
+            mask = np.array(mask, dtype=bool)
+            if mask_values is not None:
+                mask = mask | mask_values
+
+    if values.shape != mask.shape:
+        raise ValueError("values.shape and mask.shape must match")
+
+    return values, mask
+
+
+class BooleanArray(BaseMaskedArray):
+    """
+    Array of boolean (True/False) data with missing values.
+
+    This is a pandas Extension array for boolean data, under the hood
+    represented by 2 numpy arrays: a boolean array with the data and
+    a boolean array with the mask (True indicating missing).
+
+    BooleanArray implements Kleene logic (sometimes called three-value
+    logic) for logical operations. See :ref:`boolean.kleene` for more.
+
+    To construct an BooleanArray from generic array-like input, use
+    :func:`pandas.array` specifying ``dtype="boolean"`` (see examples
+    below).
+
+    .. warning::
+
+       BooleanArray is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Parameters
+    ----------
+    values : numpy.ndarray
+        A 1-d boolean-dtype array with the data.
+    mask : numpy.ndarray
+        A 1-d boolean-dtype array indicating missing values (True
+        indicates missing).
+    copy : bool, default False
+        Whether to copy the `values` and `mask` arrays.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    Returns
+    -------
+    BooleanArray
+
+    See Also
+    --------
+    array : Create an array from data with the appropriate dtype.
+    BooleanDtype : Extension dtype for boolean data.
+    Series : One-dimensional ndarray with axis labels (including time series).
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+
+    Examples
+    --------
+    Create an BooleanArray with :func:`pandas.array`:
+
+    >>> pd.array([True, False, None], dtype="boolean")
+    <BooleanArray>
+    [True, False, <NA>]
+    Length: 3, dtype: boolean
+    """
+
+    _TRUE_VALUES = {"True", "TRUE", "true", "1", "1.0"}
+    _FALSE_VALUES = {"False", "FALSE", "false", "0", "0.0"}
+
+    @classmethod
+    def _simple_new(cls, values: np.ndarray, mask: npt.NDArray[np.bool_]) -> Self:
+        result = super()._simple_new(values, mask)
+        result._dtype = BooleanDtype()
+        return result
+
+    def __init__(
+        self, values: np.ndarray, mask: np.ndarray, copy: bool = False
+    ) -> None:
+        if not (isinstance(values, np.ndarray) and values.dtype == np.bool_):
+            raise TypeError(
+                "values should be boolean numpy array. Use "
+                "the 'pd.array' function instead"
+            )
+        self._dtype = BooleanDtype()
+        super().__init__(values, mask, copy=copy)
+
+    @property
+    def dtype(self) -> BooleanDtype:
+        return self._dtype
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls,
+        strings: list[str],
+        *,
+        dtype: ExtensionDtype,
+        copy: bool = False,
+        true_values: list[str] | None = None,
+        false_values: list[str] | None = None,
+        none_values: list[str] | None = None,
+    ) -> BooleanArray:
+        true_values_union = cls._TRUE_VALUES.union(true_values or [])
+        false_values_union = cls._FALSE_VALUES.union(false_values or [])
+
+        if none_values is None:
+            none_values = []
+
+        def map_string(s) -> bool | None:
+            if s in true_values_union:
+                return True
+            elif s in false_values_union:
+                return False
+            elif s in none_values:
+                return None
+            else:
+                raise ValueError(f"{s} cannot be cast to bool")
+
+        scalars = np.array(strings, dtype=object)
+        mask = isna(scalars)
+        scalars[~mask] = list(map(map_string, scalars[~mask]))
+        return cls._from_sequence(scalars, dtype=dtype, copy=copy)
+
+    _HANDLED_TYPES = (np.ndarray, numbers.Number, bool, np.bool_)
+
+    @classmethod
+    def _coerce_to_array(
+        cls, value, *, dtype: DtypeObj, copy: bool = False
+    ) -> tuple[np.ndarray, np.ndarray]:
+        if dtype:
+            assert dtype == "boolean"
+        return coerce_to_array(value, copy=copy)
+
+    def _logical_method(self, other, op):  # type: ignore[override]
+        assert op.__name__ in {"or_", "ror_", "and_", "rand_", "xor", "rxor"}
+        other_is_scalar = lib.is_scalar(other)
+        mask = None
+
+        if isinstance(other, BooleanArray):
+            other, mask = other._data, other._mask
+        elif is_list_like(other):
+            other = np.asarray(other, dtype="bool")
+            if other.ndim > 1:
+                raise NotImplementedError("can only perform ops with 1-d structures")
+            other, mask = coerce_to_array(other, copy=False)
+        elif isinstance(other, np.bool_):
+            other = other.item()
+
+        if other_is_scalar and other is not libmissing.NA and not lib.is_bool(other):
+            raise TypeError(
+                "'other' should be pandas.NA or a bool. "
+                f"Got {type(other).__name__} instead."
+            )
+
+        if not other_is_scalar and len(self) != len(other):
+            raise ValueError("Lengths must match")
+
+        if op.__name__ in {"or_", "ror_"}:
+            result, mask = ops.kleene_or(self._data, other, self._mask, mask)
+        elif op.__name__ in {"and_", "rand_"}:
+            result, mask = ops.kleene_and(self._data, other, self._mask, mask)
+        else:
+            # i.e. xor, rxor
+            result, mask = ops.kleene_xor(self._data, other, self._mask, mask)
+
+        # i.e. BooleanArray
+        return self._maybe_mask_result(result, mask)
+
+    def _accumulate(
+        self, name: str, *, skipna: bool = True, **kwargs
+    ) -> BaseMaskedArray:
+        data = self._data
+        mask = self._mask
+        if name in ("cummin", "cummax"):
+            op = getattr(masked_accumulations, name)
+            data, mask = op(data, mask, skipna=skipna, **kwargs)
+            return self._simple_new(data, mask)
+        else:
+            from pandas.core.arrays import IntegerArray
+
+            return IntegerArray(data.astype(int), mask)._accumulate(
+                name, skipna=skipna, **kwargs
+            )
diff --git a/pandas/pandas/core/arrays/categorical.py b/pandas/pandas/core/arrays/categorical.py
new file mode 100644
index 0000000000000..0ce700772fdcc
--- /dev/null
+++ b/pandas/pandas/core/arrays/categorical.py
@@ -0,0 +1,3106 @@
+from __future__ import annotations
+
+from csv import QUOTE_NONNUMERIC
+from functools import partial
+import operator
+from shutil import get_terminal_size
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs import (
+    NaT,
+    algos as libalgos,
+    lib,
+)
+from pandas._libs.arrays import NDArrayBacked
+from pandas.compat.numpy import function as nv
+from pandas.util._validators import validate_bool_kwarg
+
+from pandas.core.dtypes.cast import (
+    coerce_indexer_dtype,
+    find_common_type,
+)
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    ensure_platform_int,
+    is_any_real_numeric_dtype,
+    is_bool_dtype,
+    is_dict_like,
+    is_hashable,
+    is_integer_dtype,
+    is_list_like,
+    is_scalar,
+    needs_i8_conversion,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    CategoricalDtypeType,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+)
+
+from pandas.core import (
+    algorithms,
+    arraylike,
+    ops,
+)
+from pandas.core.accessor import (
+    PandasDelegate,
+    delegate_names,
+)
+from pandas.core.algorithms import (
+    factorize,
+    take_nd,
+)
+from pandas.core.arrays._mixins import (
+    NDArrayBackedExtensionArray,
+    ravel_compat,
+)
+from pandas.core.base import (
+    ExtensionArray,
+    NoNewAttributesMixin,
+    PandasObject,
+)
+import pandas.core.common as com
+from pandas.core.construction import (
+    extract_array,
+    sanitize_array,
+)
+from pandas.core.ops.common import unpack_zerodim_and_defer
+from pandas.core.sorting import nargsort
+from pandas.core.strings.object_array import ObjectStringArrayMixin
+
+from pandas.io.formats import console
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterator,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        AstypeArg,
+        AxisInt,
+        Dtype,
+        DtypeObj,
+        NpDtype,
+        Ordered,
+        Self,
+        Shape,
+        SortKind,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+
+def _cat_compare_op(op):
+    opname = f"__{op.__name__}__"
+    fill_value = op is operator.ne
+
+    @unpack_zerodim_and_defer(opname)
+    def func(self, other):
+        hashable = is_hashable(other)
+        if is_list_like(other) and len(other) != len(self) and not hashable:
+            # in hashable case we may have a tuple that is itself a category
+            raise ValueError("Lengths must match.")
+
+        if not self.ordered:
+            if opname in ["__lt__", "__gt__", "__le__", "__ge__"]:
+                raise TypeError(
+                    "Unordered Categoricals can only compare equality or not"
+                )
+        if isinstance(other, Categorical):
+            # Two Categoricals can only be compared if the categories are
+            # the same (maybe up to ordering, depending on ordered)
+
+            msg = "Categoricals can only be compared if 'categories' are the same."
+            if not self._categories_match_up_to_permutation(other):
+                raise TypeError(msg)
+
+            if not self.ordered and not self.categories.equals(other.categories):
+                # both unordered and different order
+                other_codes = recode_for_categories(
+                    other.codes, other.categories, self.categories, copy=False
+                )
+            else:
+                other_codes = other._codes
+
+            ret = op(self._codes, other_codes)
+            mask = (self._codes == -1) | (other_codes == -1)
+            if mask.any():
+                ret[mask] = fill_value
+            return ret
+
+        if hashable:
+            if other in self.categories:
+                i = self._unbox_scalar(other)
+                ret = op(self._codes, i)
+
+                if opname not in {"__eq__", "__ge__", "__gt__"}:
+                    # GH#29820 performance trick; get_loc will always give i>=0,
+                    #  so in the cases (__ne__, __le__, __lt__) the setting
+                    #  here is a no-op, so can be skipped.
+                    mask = self._codes == -1
+                    ret[mask] = fill_value
+                return ret
+            else:
+                return ops.invalid_comparison(self, other, op)
+        else:
+            # allow categorical vs object dtype array comparisons for equality
+            # these are only positional comparisons
+            if opname not in ["__eq__", "__ne__"]:
+                raise TypeError(
+                    f"Cannot compare a Categorical for op {opname} with "
+                    f"type {type(other)}.\nIf you want to compare values, "
+                    "use 'np.asarray(cat) <op> other'."
+                )
+
+            if isinstance(other, ExtensionArray) and needs_i8_conversion(other.dtype):
+                # We would return NotImplemented here, but that messes up
+                #  ExtensionIndex's wrapped methods
+                return op(other, self)
+            return getattr(np.array(self), opname)(np.array(other))
+
+    func.__name__ = opname
+
+    return func
+
+
+def contains(cat, key, container) -> bool:
+    """
+    Helper for membership check for ``key`` in ``cat``.
+
+    This is a helper method for :method:`__contains__`
+    and :class:`CategoricalIndex.__contains__`.
+
+    Returns True if ``key`` is in ``cat.categories`` and the
+    location of ``key`` in ``categories`` is in ``container``.
+
+    Parameters
+    ----------
+    cat : :class:`Categorical`or :class:`categoricalIndex`
+    key : a hashable object
+        The key to check membership for.
+    container : Container (e.g. list-like or mapping)
+        The container to check for membership in.
+
+    Returns
+    -------
+    is_in : bool
+        True if ``key`` is in ``self.categories`` and location of
+        ``key`` in ``categories`` is in ``container``, else False.
+
+    Notes
+    -----
+    This method does not check for NaN values. Do that separately
+    before calling this method.
+    """
+    hash(key)
+
+    # get location of key in categories.
+    # If a KeyError, the key isn't in categories, so logically
+    #  can't be in container either.
+    try:
+        loc = cat.categories.get_loc(key)
+    except (KeyError, TypeError):
+        return False
+
+    # loc is the location of key in categories, but also the *value*
+    # for key in container. So, `key` may be in categories,
+    # but still not in `container`. Example ('b' in categories,
+    # but not in values):
+    # 'b' in Categorical(['a'], categories=['a', 'b'])  # False
+    if is_scalar(loc):
+        return loc in container
+    else:
+        # if categories is an IntervalIndex, loc is an array.
+        return any(loc_ in container for loc_ in loc)
+
+
+# error: Definition of "delete/ravel/T/repeat/copy" in base class "NDArrayBacked"
+# is incompatible with definition in base class "ExtensionArray"
+class Categorical(NDArrayBackedExtensionArray, PandasObject, ObjectStringArrayMixin):  # type: ignore[misc]
+    """
+    Represent a categorical variable in classic R / S-plus fashion.
+
+    `Categoricals` can only take on a limited, and usually fixed, number
+    of possible values (`categories`). In contrast to statistical categorical
+    variables, a `Categorical` might have an order, but numerical operations
+    (additions, divisions, ...) are not possible.
+
+    All values of the `Categorical` are either in `categories` or `np.nan`.
+    Assigning values outside of `categories` will raise a `ValueError`. Order
+    is defined by the order of the `categories`, not lexical order of the
+    values.
+
+    Parameters
+    ----------
+    values : list-like
+        The values of the categorical. If categories are given, values not in
+        categories will be replaced with NaN.
+    categories : Index-like (unique), optional
+        The unique categories for this categorical. If not given, the
+        categories are assumed to be the unique values of `values` (sorted, if
+        possible, otherwise in the order in which they appear).
+    ordered : bool, default False
+        Whether or not this categorical is treated as a ordered categorical.
+        If True, the resulting categorical will be ordered.
+        An ordered categorical respects, when sorted, the order of its
+        `categories` attribute (which in turn is the `categories` argument, if
+        provided).
+    dtype : CategoricalDtype
+        An instance of ``CategoricalDtype`` to use for this categorical.
+    copy : bool, default True
+        Whether to copy if the codes are unchanged.
+
+    Attributes
+    ----------
+    categories : Index
+        The categories of this categorical.
+    codes : ndarray
+        The codes (integer positions, which point to the categories) of this
+        categorical, read only.
+    ordered : bool
+        Whether or not this Categorical is ordered.
+    dtype : CategoricalDtype
+        The instance of ``CategoricalDtype`` storing the ``categories``
+        and ``ordered``.
+
+    Methods
+    -------
+    from_codes
+    as_ordered
+    as_unordered
+    set_categories
+    rename_categories
+    reorder_categories
+    add_categories
+    remove_categories
+    remove_unused_categories
+    map
+    __array__
+
+    Raises
+    ------
+    ValueError
+        If the categories do not validate.
+    TypeError
+        If an explicit ``ordered=True`` is given but no `categories` and the
+        `values` are not sortable.
+
+    See Also
+    --------
+    CategoricalDtype : Type for categorical data.
+    CategoricalIndex : An Index with an underlying ``Categorical``.
+
+    Notes
+    -----
+    See the `user guide
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html>`__
+    for more.
+
+    Examples
+    --------
+    >>> pd.Categorical([1, 2, 3, 1, 2, 3])
+    [1, 2, 3, 1, 2, 3]
+    Categories (3, int64): [1, 2, 3]
+
+    >>> pd.Categorical(["a", "b", "c", "a", "b", "c"])
+    ['a', 'b', 'c', 'a', 'b', 'c']
+    Categories (3, object): ['a', 'b', 'c']
+
+    Missing values are not included as a category.
+
+    >>> c = pd.Categorical([1, 2, 3, 1, 2, 3, np.nan])
+    >>> c
+    [1, 2, 3, 1, 2, 3, NaN]
+    Categories (3, int64): [1, 2, 3]
+
+    However, their presence is indicated in the `codes` attribute
+    by code `-1`.
+
+    >>> c.codes
+    array([ 0,  1,  2,  0,  1,  2, -1], dtype=int8)
+
+    Ordered `Categoricals` can be sorted according to the custom order
+    of the categories and can have a min and max value.
+
+    >>> c = pd.Categorical(
+    ...     ["a", "b", "c", "a", "b", "c"], ordered=True, categories=["c", "b", "a"]
+    ... )
+    >>> c
+    ['a', 'b', 'c', 'a', 'b', 'c']
+    Categories (3, object): ['c' < 'b' < 'a']
+    >>> c.min()
+    'c'
+    """
+
+    # For comparisons, so that numpy uses our implementation if the compare
+    # ops, which raise
+    __array_priority__ = 1000
+    # tolist is not actually deprecated, just suppressed in the __dir__
+    _hidden_attrs = PandasObject._hidden_attrs | frozenset(["tolist"])
+    _typ = "categorical"
+
+    _dtype: CategoricalDtype
+
+    @classmethod
+    # error: Argument 2 of "_simple_new" is incompatible with supertype
+    # "NDArrayBacked"; supertype defines the argument type as
+    # "Union[dtype[Any], ExtensionDtype]"
+    def _simple_new(  # type: ignore[override]
+        cls, codes: np.ndarray, dtype: CategoricalDtype
+    ) -> Self:
+        # NB: This is not _quite_ as simple as the "usual" _simple_new
+        codes = coerce_indexer_dtype(codes, dtype.categories)
+        dtype = CategoricalDtype(ordered=False).update_dtype(dtype)
+        return super()._simple_new(codes, dtype)
+
+    def __init__(
+        self,
+        values,
+        categories=None,
+        ordered=None,
+        dtype: Dtype | None = None,
+        copy: bool = True,
+    ) -> None:
+        dtype = CategoricalDtype._from_values_or_dtype(
+            values, categories, ordered, dtype
+        )
+        # At this point, dtype is always a CategoricalDtype, but
+        # we may have dtype.categories be None, and we need to
+        # infer categories in a factorization step further below
+
+        if not is_list_like(values):
+            # GH#38433
+            raise TypeError("Categorical input must be list-like")
+
+        # null_mask indicates missing values we want to exclude from inference.
+        # This means: only missing values in list-likes (not arrays/ndframes).
+        null_mask = np.array(False)
+
+        # sanitize input
+        vdtype = getattr(values, "dtype", None)
+        if isinstance(vdtype, CategoricalDtype):
+            if dtype.categories is None:
+                dtype = CategoricalDtype(values.categories, dtype.ordered)
+        elif isinstance(values, range):
+            from pandas.core.indexes.range import RangeIndex
+
+            values = RangeIndex(values)
+        elif not isinstance(values, (ABCIndex, ABCSeries, ExtensionArray)):
+            values = com.convert_to_list_like(values)
+            if isinstance(values, list) and len(values) == 0:
+                # By convention, empty lists result in object dtype:
+                values = np.array([], dtype=object)
+            elif isinstance(values, np.ndarray):
+                if values.ndim > 1:
+                    # preempt sanitize_array from raising ValueError
+                    raise NotImplementedError(
+                        "> 1 ndim Categorical are not supported at this time"
+                    )
+                values = sanitize_array(values, None)
+            else:
+                # i.e. must be a list
+                arr = sanitize_array(values, None)
+                null_mask = isna(arr)
+                if null_mask.any():
+                    # We remove null values here, then below will re-insert
+                    #  them, grep "full_codes"
+                    arr_list = [values[idx] for idx in np.where(~null_mask)[0]]
+
+                    # GH#44900 Do not cast to float if we have only missing values
+                    if arr_list or arr.dtype == "object":
+                        sanitize_dtype = None
+                    else:
+                        sanitize_dtype = arr.dtype
+
+                    arr = sanitize_array(arr_list, None, dtype=sanitize_dtype)
+                values = arr
+
+        if dtype.categories is None:
+            if isinstance(values.dtype, ArrowDtype) and issubclass(
+                values.dtype.type, CategoricalDtypeType
+            ):
+                from pandas import Index
+
+                if isinstance(values, Index):
+                    arr = values._data._pa_array.combine_chunks()
+                else:
+                    arr = values._pa_array.combine_chunks()
+                categories = arr.dictionary.to_pandas(types_mapper=ArrowDtype)
+                codes = arr.indices.to_numpy()
+                dtype = CategoricalDtype(categories, values.dtype.pyarrow_dtype.ordered)
+            else:
+                if not isinstance(values, ABCIndex):
+                    # in particular RangeIndex xref test_index_equal_range_categories
+                    values = sanitize_array(values, None)
+                try:
+                    codes, categories = factorize(values, sort=True)
+                except TypeError as err:
+                    codes, categories = factorize(values, sort=False)
+                    if dtype.ordered:
+                        # raise, as we don't have a sortable data structure and so
+                        # the user should give us one by specifying categories
+                        raise TypeError(
+                            "'values' is not ordered, please "
+                            "explicitly specify the categories order "
+                            "by passing in a categories argument."
+                        ) from err
+
+                # we're inferring from values
+                dtype = CategoricalDtype(categories, dtype.ordered)
+
+        elif isinstance(values.dtype, CategoricalDtype):
+            old_codes = extract_array(values)._codes
+            codes = recode_for_categories(
+                old_codes, values.dtype.categories, dtype.categories, copy=copy
+            )
+
+        else:
+            codes = _get_codes_for_values(values, dtype.categories)
+
+        if null_mask.any():
+            # Reinsert -1 placeholders for previously removed missing values
+            full_codes = -np.ones(null_mask.shape, dtype=codes.dtype)
+            full_codes[~null_mask] = codes
+            codes = full_codes
+
+        dtype = CategoricalDtype(ordered=False).update_dtype(dtype)
+        arr = coerce_indexer_dtype(codes, dtype.categories)
+        super().__init__(arr, dtype)
+
+    @property
+    def dtype(self) -> CategoricalDtype:
+        """
+        The :class:`~pandas.api.types.CategoricalDtype` for this instance.
+
+        See Also
+        --------
+        astype : Cast argument to a specified dtype.
+        CategoricalDtype : Type for categorical data.
+
+        Examples
+        --------
+        >>> cat = pd.Categorical(["a", "b"], ordered=True)
+        >>> cat
+        ['a', 'b']
+        Categories (2, object): ['a' < 'b']
+        >>> cat.dtype
+        CategoricalDtype(categories=['a', 'b'], ordered=True, categories_dtype=object)
+        """
+        return self._dtype
+
+    @property
+    def _internal_fill_value(self) -> int:
+        # using the specific numpy integer instead of python int to get
+        #  the correct dtype back from _quantile in the all-NA case
+        dtype = self._ndarray.dtype
+        return dtype.type(-1)
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        return cls(scalars, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _from_scalars(cls, scalars, *, dtype: DtypeObj) -> Self:
+        if dtype is None:
+            # The _from_scalars strictness doesn't make much sense in this case.
+            raise NotImplementedError
+
+        res = cls._from_sequence(scalars, dtype=dtype)
+
+        # if there are any non-category elements in scalars, these will be
+        #  converted to NAs in res.
+        mask = isna(scalars)
+        if not (mask == res.isna()).all():
+            # Some non-category element in scalars got converted to NA in res.
+            raise ValueError
+        return res
+
+    @overload
+    def astype(self, dtype: npt.DTypeLike, copy: bool = ...) -> np.ndarray: ...
+
+    @overload
+    def astype(self, dtype: ExtensionDtype, copy: bool = ...) -> ExtensionArray: ...
+
+    @overload
+    def astype(self, dtype: AstypeArg, copy: bool = ...) -> ArrayLike: ...
+
+    def astype(self, dtype: AstypeArg, copy: bool = True) -> ArrayLike:
+        """
+        Coerce this type to another dtype
+
+        Parameters
+        ----------
+        dtype : numpy dtype or pandas type
+        copy : bool, default True
+            By default, astype always returns a newly allocated object.
+            If copy is set to False and dtype is categorical, the original
+            object is returned.
+        """
+        dtype = pandas_dtype(dtype)
+        result: Categorical | np.ndarray
+        if self.dtype is dtype:
+            result = self.copy() if copy else self
+
+        elif isinstance(dtype, CategoricalDtype):
+            # GH 10696/18593/18630
+            dtype = self.dtype.update_dtype(dtype)
+            self = self.copy() if copy else self
+            result = self._set_dtype(dtype)
+
+        elif isinstance(dtype, ExtensionDtype):
+            return super().astype(dtype, copy=copy)
+
+        elif dtype.kind in "iu" and self.isna().any():
+            raise ValueError("Cannot convert float NaN to integer")
+
+        elif len(self.codes) == 0 or len(self.categories) == 0:
+            # For NumPy 1.x compatibility we cannot use copy=None.  And
+            # `copy=False` has the meaning of `copy=None` here:
+            if not copy:
+                result = np.asarray(self, dtype=dtype)
+            else:
+                result = np.array(self, dtype=dtype)
+
+        else:
+            # GH8628 (PERF): astype category codes instead of astyping array
+            new_cats = self.categories._values
+
+            try:
+                new_cats = new_cats.astype(dtype=dtype, copy=copy)
+                fill_value = self.categories._na_value
+                if not is_valid_na_for_dtype(fill_value, dtype):
+                    fill_value = lib.item_from_zerodim(
+                        np.array(self.categories._na_value).astype(dtype)
+                    )
+            except (
+                TypeError,  # downstream error msg for CategoricalIndex is misleading
+                ValueError,
+            ) as err:
+                msg = f"Cannot cast {self.categories.dtype} dtype to {dtype}"
+                raise ValueError(msg) from err
+
+            result = take_nd(
+                new_cats, ensure_platform_int(self._codes), fill_value=fill_value
+            )
+
+        return result
+
+    @classmethod
+    def _from_inferred_categories(
+        cls, inferred_categories, inferred_codes, dtype, true_values=None
+    ) -> Self:
+        """
+        Construct a Categorical from inferred values.
+
+        For inferred categories (`dtype` is None) the categories are sorted.
+        For explicit `dtype`, the `inferred_categories` are cast to the
+        appropriate type.
+
+        Parameters
+        ----------
+        inferred_categories : Index
+        inferred_codes : Index
+        dtype : CategoricalDtype or 'category'
+        true_values : list, optional
+            If none are provided, the default ones are
+            "True", "TRUE", and "true."
+
+        Returns
+        -------
+        Categorical
+        """
+        from pandas import (
+            Index,
+            to_datetime,
+            to_numeric,
+            to_timedelta,
+        )
+
+        cats = Index(inferred_categories)
+        known_categories = (
+            isinstance(dtype, CategoricalDtype) and dtype.categories is not None
+        )
+
+        if known_categories:
+            # Convert to a specialized type with `dtype` if specified.
+            if is_any_real_numeric_dtype(dtype.categories.dtype):
+                cats = to_numeric(inferred_categories, errors="coerce")
+            elif lib.is_np_dtype(dtype.categories.dtype, "M"):
+                cats = to_datetime(inferred_categories, errors="coerce")
+            elif lib.is_np_dtype(dtype.categories.dtype, "m"):
+                cats = to_timedelta(inferred_categories, errors="coerce")
+            elif is_bool_dtype(dtype.categories.dtype):
+                if true_values is None:
+                    true_values = ["True", "TRUE", "true"]
+
+                # error: Incompatible types in assignment (expression has type
+                # "ndarray", variable has type "Index")
+                cats = cats.isin(true_values)  # type: ignore[assignment]
+
+        if known_categories:
+            # Recode from observation order to dtype.categories order.
+            categories = dtype.categories
+            codes = recode_for_categories(inferred_codes, cats, categories)
+        elif not cats.is_monotonic_increasing:
+            # Sort categories and recode for unknown categories.
+            unsorted = cats.copy()
+            categories = cats.sort_values()
+
+            codes = recode_for_categories(inferred_codes, unsorted, categories)
+            dtype = CategoricalDtype(categories, ordered=False)
+        else:
+            dtype = CategoricalDtype(cats, ordered=False)
+            codes = inferred_codes
+
+        return cls._simple_new(codes, dtype=dtype)
+
+    @classmethod
+    def from_codes(
+        cls,
+        codes,
+        categories=None,
+        ordered=None,
+        dtype: Dtype | None = None,
+        validate: bool = True,
+    ) -> Self:
+        """
+        Make a Categorical type from codes and categories or dtype.
+
+        This constructor is useful if you already have codes and
+        categories/dtype and so do not need the (computation intensive)
+        factorization step, which is usually done on the constructor.
+
+        If your data does not follow this convention, please use the normal
+        constructor.
+
+        Parameters
+        ----------
+        codes : array-like of int
+            An integer array, where each integer points to a category in
+            categories or dtype.categories, or else is -1 for NaN.
+        categories : index-like, optional
+            The categories for the categorical. Items need to be unique.
+            If the categories are not given here, then they must be provided
+            in `dtype`.
+        ordered : bool, optional
+            Whether or not this categorical is treated as an ordered
+            categorical. If not given here or in `dtype`, the resulting
+            categorical will be unordered.
+        dtype : CategoricalDtype or "category", optional
+            If :class:`CategoricalDtype`, cannot be used together with
+            `categories` or `ordered`.
+        validate : bool, default True
+            If True, validate that the codes are valid for the dtype.
+            If False, don't validate that the codes are valid. Be careful about skipping
+            validation, as invalid codes can lead to severe problems, such as segfaults.
+
+            .. versionadded:: 2.1.0
+
+        Returns
+        -------
+        Categorical
+
+        See Also
+        --------
+        codes : The category codes of the categorical.
+        CategoricalIndex : An Index with an underlying ``Categorical``.
+
+        Examples
+        --------
+        >>> dtype = pd.CategoricalDtype(["a", "b"], ordered=True)
+        >>> pd.Categorical.from_codes(codes=[0, 1, 0, 1], dtype=dtype)
+        ['a', 'b', 'a', 'b']
+        Categories (2, object): ['a' < 'b']
+        """
+        dtype = CategoricalDtype._from_values_or_dtype(
+            categories=categories, ordered=ordered, dtype=dtype
+        )
+        if dtype.categories is None:
+            msg = (
+                "The categories must be provided in 'categories' or "
+                "'dtype'. Both were None."
+            )
+            raise ValueError(msg)
+
+        if validate:
+            # beware: non-valid codes may segfault
+            codes = cls._validate_codes_for_dtype(codes, dtype=dtype)
+
+        return cls._simple_new(codes, dtype=dtype)
+
+    # ------------------------------------------------------------------
+    # Categories/Codes/Ordered
+
+    @property
+    def categories(self) -> Index:
+        """
+        The categories of this categorical.
+
+        Setting assigns new values to each category (effectively a rename of
+        each individual category).
+
+        The assigned value has to be a list-like object. All items must be
+        unique and the number of items in the new categories must be the same
+        as the number of items in the old categories.
+
+        Raises
+        ------
+        ValueError
+            If the new categories do not validate as categories or if the
+            number of new categories is unequal the number of old categories
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        reorder_categories : Reorder categories.
+        add_categories : Add new categories.
+        remove_categories : Remove the specified categories.
+        remove_unused_categories : Remove categories which are not used.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> ser = pd.Series(["a", "b", "c", "a"], dtype="category")
+        >>> ser.cat.categories
+        Index(['a', 'b', 'c'], dtype='object')
+
+        >>> raw_cat = pd.Categorical(["a", "b", "c", "a"], categories=["b", "c", "d"])
+        >>> ser = pd.Series(raw_cat)
+        >>> ser.cat.categories
+        Index(['b', 'c', 'd'], dtype='object')
+
+        For :class:`pandas.Categorical`:
+
+        >>> cat = pd.Categorical(["a", "b"], ordered=True)
+        >>> cat.categories
+        Index(['a', 'b'], dtype='object')
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "c", "b", "a", "c", "b"])
+        >>> ci.categories
+        Index(['a', 'b', 'c'], dtype='object')
+
+        >>> ci = pd.CategoricalIndex(["a", "c"], categories=["c", "b", "a"])
+        >>> ci.categories
+        Index(['c', 'b', 'a'], dtype='object')
+        """
+        return self.dtype.categories
+
+    @property
+    def ordered(self) -> Ordered:
+        """
+        Whether the categories have an ordered relationship.
+
+        See Also
+        --------
+        set_ordered : Set the ordered attribute.
+        as_ordered : Set the Categorical to be ordered.
+        as_unordered : Set the Categorical to be unordered.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> ser = pd.Series(["a", "b", "c", "a"], dtype="category")
+        >>> ser.cat.ordered
+        False
+
+        >>> raw_cat = pd.Categorical(["a", "b", "c", "a"], ordered=True)
+        >>> ser = pd.Series(raw_cat)
+        >>> ser.cat.ordered
+        True
+
+        For :class:`pandas.Categorical`:
+
+        >>> cat = pd.Categorical(["a", "b"], ordered=True)
+        >>> cat.ordered
+        True
+
+        >>> cat = pd.Categorical(["a", "b"], ordered=False)
+        >>> cat.ordered
+        False
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "b"], ordered=True)
+        >>> ci.ordered
+        True
+
+        >>> ci = pd.CategoricalIndex(["a", "b"], ordered=False)
+        >>> ci.ordered
+        False
+        """
+        return self.dtype.ordered
+
+    @property
+    def codes(self) -> np.ndarray:
+        """
+        The category codes of this categorical index.
+
+        Codes are an array of integers which are the positions of the actual
+        values in the categories array.
+
+        There is no setter, use the other categorical methods and the normal item
+        setter to change values in the categorical.
+
+        Returns
+        -------
+        ndarray[int]
+            A non-writable view of the ``codes`` array.
+
+        See Also
+        --------
+        Categorical.from_codes : Make a Categorical from codes.
+        CategoricalIndex : An Index with an underlying ``Categorical``.
+
+        Examples
+        --------
+        For :class:`pandas.Categorical`:
+
+        >>> cat = pd.Categorical(["a", "b"], ordered=True)
+        >>> cat.codes
+        array([0, 1], dtype=int8)
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "b", "c", "a", "b", "c"])
+        >>> ci.codes
+        array([0, 1, 2, 0, 1, 2], dtype=int8)
+
+        >>> ci = pd.CategoricalIndex(["a", "c"], categories=["c", "b", "a"])
+        >>> ci.codes
+        array([2, 0], dtype=int8)
+        """
+        v = self._codes.view()
+        v.flags.writeable = False
+        return v
+
+    def _set_categories(self, categories, fastpath: bool = False) -> None:
+        """
+        Sets new categories inplace
+
+        Parameters
+        ----------
+        fastpath : bool, default False
+           Don't perform validation of the categories for uniqueness or nulls
+
+        Examples
+        --------
+        >>> c = pd.Categorical(["a", "b"])
+        >>> c
+        ['a', 'b']
+        Categories (2, object): ['a', 'b']
+
+        >>> c._set_categories(pd.Index(["a", "c"]))
+        >>> c
+        ['a', 'c']
+        Categories (2, object): ['a', 'c']
+        """
+        if fastpath:
+            new_dtype = CategoricalDtype._from_fastpath(categories, self.ordered)
+        else:
+            new_dtype = CategoricalDtype(categories, ordered=self.ordered)
+        if (
+            not fastpath
+            and self.dtype.categories is not None
+            and len(new_dtype.categories) != len(self.dtype.categories)
+        ):
+            raise ValueError(
+                "new categories need to have the same number of "
+                "items as the old categories!"
+            )
+
+        super().__init__(self._ndarray, new_dtype)
+
+    def _set_dtype(self, dtype: CategoricalDtype) -> Self:
+        """
+        Internal method for directly updating the CategoricalDtype
+
+        Parameters
+        ----------
+        dtype : CategoricalDtype
+
+        Notes
+        -----
+        We don't do any validation here. It's assumed that the dtype is
+        a (valid) instance of `CategoricalDtype`.
+        """
+        codes = recode_for_categories(self.codes, self.categories, dtype.categories)
+        return type(self)._simple_new(codes, dtype=dtype)
+
+    def set_ordered(self, value: bool) -> Self:
+        """
+        Set the ordered attribute to the boolean value.
+
+        Parameters
+        ----------
+        value : bool
+           Set whether this categorical is ordered (True) or not (False).
+        """
+        new_dtype = CategoricalDtype(self.categories, ordered=value)
+        cat = self.copy()
+        NDArrayBacked.__init__(cat, cat._ndarray, new_dtype)
+        return cat
+
+    def as_ordered(self) -> Self:
+        """
+        Set the Categorical to be ordered.
+
+        Returns
+        -------
+        Categorical
+            Ordered Categorical.
+
+        See Also
+        --------
+        as_unordered : Set the Categorical to be unordered.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> ser = pd.Series(["a", "b", "c", "a"], dtype="category")
+        >>> ser.cat.ordered
+        False
+        >>> ser = ser.cat.as_ordered()
+        >>> ser.cat.ordered
+        True
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "b", "c", "a"])
+        >>> ci.ordered
+        False
+        >>> ci = ci.as_ordered()
+        >>> ci.ordered
+        True
+        """
+        return self.set_ordered(True)
+
+    def as_unordered(self) -> Self:
+        """
+        Set the Categorical to be unordered.
+
+        Returns
+        -------
+        Categorical
+            Unordered Categorical.
+
+        See Also
+        --------
+        as_ordered : Set the Categorical to be ordered.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> raw_cat = pd.Categorical(["a", "b", "c", "a"], ordered=True)
+        >>> ser = pd.Series(raw_cat)
+        >>> ser.cat.ordered
+        True
+        >>> ser = ser.cat.as_unordered()
+        >>> ser.cat.ordered
+        False
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "b", "c", "a"], ordered=True)
+        >>> ci.ordered
+        True
+        >>> ci = ci.as_unordered()
+        >>> ci.ordered
+        False
+        """
+        return self.set_ordered(False)
+
+    def set_categories(
+        self, new_categories, ordered=None, rename: bool = False
+    ) -> Self:
+        """
+        Set the categories to the specified new categories.
+
+        ``new_categories`` can include new categories (which will result in
+        unused categories) or remove old categories (which results in values
+        set to ``NaN``). If ``rename=True``, the categories will simply be renamed
+        (less or more items than in old categories will result in values set to
+        ``NaN`` or in unused categories respectively).
+
+        This method can be used to perform more than one action of adding,
+        removing, and reordering simultaneously and is therefore faster than
+        performing the individual steps via the more specialised methods.
+
+        On the other hand this methods does not do checks (e.g., whether the
+        old categories are included in the new categories on a reorder), which
+        can result in surprising changes, for example when using special string
+        dtypes, which do not consider a S1 string equal to a single char
+        python string.
+
+        Parameters
+        ----------
+        new_categories : Index-like
+           The categories in new order.
+        ordered : bool, default None
+           Whether or not the categorical is treated as a ordered categorical.
+           If not given, do not change the ordered information.
+        rename : bool, default False
+           Whether or not the new_categories should be considered as a rename
+           of the old categories or as reordered categories.
+
+        Returns
+        -------
+        Categorical
+            New categories to be used, with optional ordering changes.
+
+        Raises
+        ------
+        ValueError
+            If new_categories does not validate as categories
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        reorder_categories : Reorder categories.
+        add_categories : Add new categories.
+        remove_categories : Remove the specified categories.
+        remove_unused_categories : Remove categories which are not used.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> raw_cat = pd.Categorical(
+        ...     ["a", "b", "c", "A"], categories=["a", "b", "c"], ordered=True
+        ... )
+        >>> ser = pd.Series(raw_cat)
+        >>> ser
+        0   a
+        1   b
+        2   c
+        3   NaN
+        dtype: category
+        Categories (3, object): ['a' < 'b' < 'c']
+
+        >>> ser.cat.set_categories(["A", "B", "C"], rename=True)
+        0   A
+        1   B
+        2   C
+        3   NaN
+        dtype: category
+        Categories (3, object): ['A' < 'B' < 'C']
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(
+        ...     ["a", "b", "c", "A"], categories=["a", "b", "c"], ordered=True
+        ... )
+        >>> ci
+        CategoricalIndex(['a', 'b', 'c', nan], categories=['a', 'b', 'c'],
+                         ordered=True, dtype='category')
+
+        >>> ci.set_categories(["A", "b", "c"])
+        CategoricalIndex([nan, 'b', 'c', nan], categories=['A', 'b', 'c'],
+                         ordered=True, dtype='category')
+        >>> ci.set_categories(["A", "b", "c"], rename=True)
+        CategoricalIndex(['A', 'b', 'c', nan], categories=['A', 'b', 'c'],
+                         ordered=True, dtype='category')
+        """
+
+        if ordered is None:
+            ordered = self.dtype.ordered
+        new_dtype = CategoricalDtype(new_categories, ordered=ordered)
+
+        cat = self.copy()
+        if rename:
+            if cat.dtype.categories is not None and len(new_dtype.categories) < len(
+                cat.dtype.categories
+            ):
+                # remove all _codes which are larger and set to -1/NaN
+                cat._codes[cat._codes >= len(new_dtype.categories)] = -1
+            codes = cat._codes
+        else:
+            codes = recode_for_categories(
+                cat.codes, cat.categories, new_dtype.categories
+            )
+        NDArrayBacked.__init__(cat, codes, new_dtype)
+        return cat
+
+    def rename_categories(self, new_categories) -> Self:
+        """
+        Rename categories.
+
+        This method is commonly used to re-label or adjust the
+        category names in categorical data without changing the
+        underlying data. It is useful in situations where you want
+        to modify the labels used for clarity, consistency,
+        or readability.
+
+        Parameters
+        ----------
+        new_categories : list-like, dict-like or callable
+
+            New categories which will replace old categories.
+
+            * list-like: all items must be unique and the number of items in
+              the new categories must match the existing number of categories.
+
+            * dict-like: specifies a mapping from
+              old categories to new. Categories not contained in the mapping
+              are passed through and extra categories in the mapping are
+              ignored.
+
+            * callable : a callable that is called on all items in the old
+              categories and whose return values comprise the new categories.
+
+        Returns
+        -------
+        Categorical
+            Categorical with renamed categories.
+
+        Raises
+        ------
+        ValueError
+            If new categories are list-like and do not have the same number of
+            items than the current categories or do not validate as categories
+
+        See Also
+        --------
+        reorder_categories : Reorder categories.
+        add_categories : Add new categories.
+        remove_categories : Remove the specified categories.
+        remove_unused_categories : Remove categories which are not used.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        >>> c = pd.Categorical(["a", "a", "b"])
+        >>> c.rename_categories([0, 1])
+        [0, 0, 1]
+        Categories (2, int64): [0, 1]
+
+        For dict-like ``new_categories``, extra keys are ignored and
+        categories not in the dictionary are passed through
+
+        >>> c.rename_categories({"a": "A", "c": "C"})
+        ['A', 'A', 'b']
+        Categories (2, object): ['A', 'b']
+
+        You may also provide a callable to create the new categories
+
+        >>> c.rename_categories(lambda x: x.upper())
+        ['A', 'A', 'B']
+        Categories (2, object): ['A', 'B']
+        """
+
+        if is_dict_like(new_categories):
+            new_categories = [
+                new_categories.get(item, item) for item in self.categories
+            ]
+        elif callable(new_categories):
+            new_categories = [new_categories(item) for item in self.categories]
+
+        cat = self.copy()
+        cat._set_categories(new_categories)
+        return cat
+
+    def reorder_categories(self, new_categories, ordered=None) -> Self:
+        """
+        Reorder categories as specified in new_categories.
+
+        ``new_categories`` need to include all old categories and no new category
+        items.
+
+        Parameters
+        ----------
+        new_categories : Index-like
+           The categories in new order.
+        ordered : bool, optional
+           Whether or not the categorical is treated as a ordered categorical.
+           If not given, do not change the ordered information.
+
+        Returns
+        -------
+        Categorical
+            Categorical with reordered categories.
+
+        Raises
+        ------
+        ValueError
+            If the new categories do not contain all old category items or any
+            new ones
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        add_categories : Add new categories.
+        remove_categories : Remove the specified categories.
+        remove_unused_categories : Remove categories which are not used.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        For :class:`pandas.Series`:
+
+        >>> ser = pd.Series(["a", "b", "c", "a"], dtype="category")
+        >>> ser = ser.cat.reorder_categories(["c", "b", "a"], ordered=True)
+        >>> ser
+        0   a
+        1   b
+        2   c
+        3   a
+        dtype: category
+        Categories (3, object): ['c' < 'b' < 'a']
+
+        >>> ser.sort_values()
+        2   c
+        1   b
+        0   a
+        3   a
+        dtype: category
+        Categories (3, object): ['c' < 'b' < 'a']
+
+        For :class:`pandas.CategoricalIndex`:
+
+        >>> ci = pd.CategoricalIndex(["a", "b", "c", "a"])
+        >>> ci
+        CategoricalIndex(['a', 'b', 'c', 'a'], categories=['a', 'b', 'c'],
+                         ordered=False, dtype='category')
+        >>> ci.reorder_categories(["c", "b", "a"], ordered=True)
+        CategoricalIndex(['a', 'b', 'c', 'a'], categories=['c', 'b', 'a'],
+                         ordered=True, dtype='category')
+        """
+        if (
+            len(self.categories) != len(new_categories)
+            or not self.categories.difference(new_categories).empty
+        ):
+            raise ValueError(
+                "items in new_categories are not the same as in old categories"
+            )
+        return self.set_categories(new_categories, ordered=ordered)
+
+    def add_categories(self, new_categories) -> Self:
+        """
+        Add new categories.
+
+        `new_categories` will be included at the last/highest place in the
+        categories and will be unused directly after this call.
+
+        Parameters
+        ----------
+        new_categories : category or list-like of category
+            The new categories to be included.
+
+        Returns
+        -------
+        Categorical
+            Categorical with new categories added.
+
+        Raises
+        ------
+        ValueError
+            If the new categories include old categories or do not validate as
+            categories
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        reorder_categories : Reorder categories.
+        remove_categories : Remove the specified categories.
+        remove_unused_categories : Remove categories which are not used.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        >>> c = pd.Categorical(["c", "b", "c"])
+        >>> c
+        ['c', 'b', 'c']
+        Categories (2, object): ['b', 'c']
+
+        >>> c.add_categories(["d", "a"])
+        ['c', 'b', 'c']
+        Categories (4, object): ['b', 'c', 'd', 'a']
+        """
+
+        if not is_list_like(new_categories):
+            new_categories = [new_categories]
+        already_included = set(new_categories) & set(self.dtype.categories)
+        if len(already_included) != 0:
+            raise ValueError(
+                f"new categories must not include old categories: {already_included}"
+            )
+
+        if hasattr(new_categories, "dtype"):
+            from pandas import Series
+
+            dtype = find_common_type(
+                [self.dtype.categories.dtype, new_categories.dtype]
+            )
+            new_categories = Series(
+                list(self.dtype.categories) + list(new_categories), dtype=dtype
+            )
+        else:
+            new_categories = list(self.dtype.categories) + list(new_categories)
+
+        new_dtype = CategoricalDtype(new_categories, self.ordered)
+        cat = self.copy()
+        codes = coerce_indexer_dtype(cat._ndarray, new_dtype.categories)
+        NDArrayBacked.__init__(cat, codes, new_dtype)
+        return cat
+
+    def remove_categories(self, removals) -> Self:
+        """
+        Remove the specified categories.
+
+        The ``removals`` argument must be a subset of the current categories.
+        Any values that were part of the removed categories will be set to NaN.
+
+        Parameters
+        ----------
+        removals : category or list of categories
+           The categories which should be removed.
+
+        Returns
+        -------
+        Categorical
+            Categorical with removed categories.
+
+        Raises
+        ------
+        ValueError
+            If the removals are not contained in the categories
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        reorder_categories : Reorder categories.
+        add_categories : Add new categories.
+        remove_unused_categories : Remove categories which are not used.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        >>> c = pd.Categorical(["a", "c", "b", "c", "d"])
+        >>> c
+        ['a', 'c', 'b', 'c', 'd']
+        Categories (4, object): ['a', 'b', 'c', 'd']
+
+        >>> c.remove_categories(["d", "a"])
+        [NaN, 'c', 'b', 'c', NaN]
+        Categories (2, object): ['b', 'c']
+        """
+        from pandas import Index
+
+        if not is_list_like(removals):
+            removals = [removals]
+
+        removals = Index(removals).unique().dropna()
+        new_categories = (
+            self.dtype.categories.difference(removals, sort=False)
+            if self.dtype.ordered is True
+            else self.dtype.categories.difference(removals)
+        )
+        not_included = removals.difference(self.dtype.categories)
+
+        if len(not_included) != 0:
+            not_included = set(not_included)
+            raise ValueError(f"removals must all be in old categories: {not_included}")
+
+        return self.set_categories(new_categories, ordered=self.ordered, rename=False)
+
+    def remove_unused_categories(self) -> Self:
+        """
+        Remove categories which are not used.
+
+        This method is useful when working with datasets
+        that undergo dynamic changes where categories may no longer be
+        relevant, allowing to maintain a clean, efficient data structure.
+
+        Returns
+        -------
+        Categorical
+            Categorical with unused categories dropped.
+
+        See Also
+        --------
+        rename_categories : Rename categories.
+        reorder_categories : Reorder categories.
+        add_categories : Add new categories.
+        remove_categories : Remove the specified categories.
+        set_categories : Set the categories to the specified ones.
+
+        Examples
+        --------
+        >>> c = pd.Categorical(["a", "c", "b", "c", "d"])
+        >>> c
+        ['a', 'c', 'b', 'c', 'd']
+        Categories (4, object): ['a', 'b', 'c', 'd']
+
+        >>> c[2] = "a"
+        >>> c[4] = "c"
+        >>> c
+        ['a', 'c', 'a', 'c', 'c']
+        Categories (4, object): ['a', 'b', 'c', 'd']
+
+        >>> c.remove_unused_categories()
+        ['a', 'c', 'a', 'c', 'c']
+        Categories (2, object): ['a', 'c']
+        """
+        idx, inv = np.unique(self._codes, return_inverse=True)
+
+        if idx.size != 0 and idx[0] == -1:  # na sentinel
+            idx, inv = idx[1:], inv - 1
+
+        new_categories = self.dtype.categories.take(idx)
+        new_dtype = CategoricalDtype._from_fastpath(
+            new_categories, ordered=self.ordered
+        )
+        new_codes = coerce_indexer_dtype(inv, new_dtype.categories)
+
+        cat = self.copy()
+        NDArrayBacked.__init__(cat, new_codes, new_dtype)
+        return cat
+
+    # ------------------------------------------------------------------
+
+    def map(
+        self,
+        mapper,
+        na_action: Literal["ignore"] | None = None,
+    ):
+        """
+        Map categories using an input mapping or function.
+
+        Maps the categories to new categories. If the mapping correspondence is
+        one-to-one the result is a :class:`~pandas.Categorical` which has the
+        same order property as the original, otherwise a :class:`~pandas.Index`
+        is returned. NaN values are unaffected.
+
+        If a `dict` or :class:`~pandas.Series` is used any unmapped category is
+        mapped to `NaN`. Note that if this happens an :class:`~pandas.Index`
+        will be returned.
+
+        Parameters
+        ----------
+        mapper : function, dict, or Series
+            Mapping correspondence.
+        na_action : {None, 'ignore'}, default None
+            If 'ignore', propagate NaN values, without passing them to the
+            mapping correspondence.
+
+        Returns
+        -------
+        pandas.Categorical or pandas.Index
+            Mapped categorical.
+
+        See Also
+        --------
+        CategoricalIndex.map : Apply a mapping correspondence on a
+            :class:`~pandas.CategoricalIndex`.
+        Index.map : Apply a mapping correspondence on an
+            :class:`~pandas.Index`.
+        Series.map : Apply a mapping correspondence on a
+            :class:`~pandas.Series`.
+        Series.apply : Apply more complex functions on a
+            :class:`~pandas.Series`.
+
+        Examples
+        --------
+        >>> cat = pd.Categorical(["a", "b", "c"])
+        >>> cat
+        ['a', 'b', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> cat.map(lambda x: x.upper(), na_action=None)
+        ['A', 'B', 'C']
+        Categories (3, object): ['A', 'B', 'C']
+        >>> cat.map({"a": "first", "b": "second", "c": "third"}, na_action=None)
+        ['first', 'second', 'third']
+        Categories (3, object): ['first', 'second', 'third']
+
+        If the mapping is one-to-one the ordering of the categories is
+        preserved:
+
+        >>> cat = pd.Categorical(["a", "b", "c"], ordered=True)
+        >>> cat
+        ['a', 'b', 'c']
+        Categories (3, object): ['a' < 'b' < 'c']
+        >>> cat.map({"a": 3, "b": 2, "c": 1}, na_action=None)
+        [3, 2, 1]
+        Categories (3, int64): [3 < 2 < 1]
+
+        If the mapping is not one-to-one an :class:`~pandas.Index` is returned:
+
+        >>> cat.map({"a": "first", "b": "second", "c": "first"}, na_action=None)
+        Index(['first', 'second', 'first'], dtype='object')
+
+        If a `dict` is used, all unmapped categories are mapped to `NaN` and
+        the result is an :class:`~pandas.Index`:
+
+        >>> cat.map({"a": "first", "b": "second"}, na_action=None)
+        Index(['first', 'second', nan], dtype='object')
+        """
+        assert callable(mapper) or is_dict_like(mapper)
+
+        new_categories = self.categories.map(mapper)
+
+        has_nans = np.any(self._codes == -1)
+
+        na_val = np.nan
+        if na_action is None and has_nans:
+            na_val = mapper(np.nan) if callable(mapper) else mapper.get(np.nan, np.nan)
+
+        if new_categories.is_unique and not new_categories.hasnans and na_val is np.nan:
+            new_dtype = CategoricalDtype(new_categories, ordered=self.ordered)
+            return self.from_codes(self._codes.copy(), dtype=new_dtype, validate=False)
+
+        if has_nans:
+            new_categories = new_categories.insert(len(new_categories), na_val)
+
+        return np.take(new_categories, self._codes)
+
+    __eq__ = _cat_compare_op(operator.eq)
+    __ne__ = _cat_compare_op(operator.ne)
+    __lt__ = _cat_compare_op(operator.lt)
+    __gt__ = _cat_compare_op(operator.gt)
+    __le__ = _cat_compare_op(operator.le)
+    __ge__ = _cat_compare_op(operator.ge)
+
+    # -------------------------------------------------------------
+    # Validators; ideally these can be de-duplicated
+
+    def _validate_setitem_value(self, value):
+        if not is_hashable(value):
+            # wrap scalars and hashable-listlikes in list
+            return self._validate_listlike(value)
+        else:
+            return self._validate_scalar(value)
+
+    def _validate_scalar(self, fill_value):
+        """
+        Convert a user-facing fill_value to a representation to use with our
+        underlying ndarray, raising TypeError if this is not possible.
+
+        Parameters
+        ----------
+        fill_value : object
+
+        Returns
+        -------
+        fill_value : int
+
+        Raises
+        ------
+        TypeError
+        """
+
+        if is_valid_na_for_dtype(fill_value, self.categories.dtype):
+            fill_value = -1
+        elif fill_value in self.categories:
+            fill_value = self._unbox_scalar(fill_value)
+        else:
+            raise TypeError(
+                "Cannot setitem on a Categorical with a new "
+                f"category ({fill_value}), set the categories first"
+            ) from None
+        return fill_value
+
+    @classmethod
+    def _validate_codes_for_dtype(cls, codes, *, dtype: CategoricalDtype) -> np.ndarray:
+        if isinstance(codes, ExtensionArray) and is_integer_dtype(codes.dtype):
+            # Avoid the implicit conversion of Int to object
+            if isna(codes).any():
+                raise ValueError("codes cannot contain NA values")
+            codes = codes.to_numpy(dtype=np.int64)
+        else:
+            codes = np.asarray(codes)
+        if len(codes) and codes.dtype.kind not in "iu":
+            raise ValueError("codes need to be array-like integers")
+
+        if len(codes) and (codes.max() >= len(dtype.categories) or codes.min() < -1):
+            raise ValueError("codes need to be between -1 and len(categories)-1")
+        return codes
+
+    # -------------------------------------------------------------
+
+    @ravel_compat
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        """
+        The numpy array interface.
+
+        Users should not call this directly. Rather, it is invoked by
+        :func:`numpy.array` and :func:`numpy.asarray`.
+
+        Parameters
+        ----------
+        dtype : np.dtype or None
+            Specifies the the dtype for the array.
+
+        copy : bool or None, optional
+            See :func:`numpy.asarray`.
+
+        Returns
+        -------
+        numpy.array
+            A numpy array of either the specified dtype or,
+            if dtype==None (default), the same dtype as
+            categorical.categories.dtype.
+
+        See Also
+        --------
+        numpy.asarray : Convert input to numpy.ndarray.
+
+        Examples
+        --------
+
+        >>> cat = pd.Categorical(["a", "b"], ordered=True)
+
+        The following calls ``cat.__array__``
+
+        >>> np.asarray(cat)
+        array(['a', 'b'], dtype=object)
+        """
+        if copy is False:
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        ret = take_nd(self.categories._values, self._codes)
+        # When we're a Categorical[ExtensionArray], like Interval,
+        # we need to ensure __array__ gets all the way to an
+        # ndarray.
+
+        # `take_nd` should already make a copy, so don't force again.
+        return np.asarray(ret, dtype=dtype)
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        # for binary ops, use our custom dunder methods
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            # e.g. test_numpy_ufuncs_out
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        if method == "reduce":
+            # e.g. TestCategoricalAnalytics::test_min_max_ordered
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                return result
+
+        # for all other cases, raise for now (similarly as what happens in
+        # Series.__array_prepare__)
+        raise TypeError(
+            f"Object with dtype {self.dtype} cannot perform "
+            f"the numpy op {ufunc.__name__}"
+        )
+
+    def __setstate__(self, state) -> None:
+        """Necessary for making this object picklable"""
+        if not isinstance(state, dict):
+            return super().__setstate__(state)
+
+        if "_dtype" not in state:
+            state["_dtype"] = CategoricalDtype(state["_categories"], state["_ordered"])
+
+        if "_codes" in state and "_ndarray" not in state:
+            # backward compat, changed what is property vs attribute
+            state["_ndarray"] = state.pop("_codes")
+
+        super().__setstate__(state)
+
+    @property
+    def nbytes(self) -> int:
+        return self._codes.nbytes + self.dtype.categories.values.nbytes
+
+    def memory_usage(self, deep: bool = False) -> int:
+        """
+        Memory usage of my values
+
+        Parameters
+        ----------
+        deep : bool
+            Introspect the data deeply, interrogate
+            `object` dtypes for system-level memory consumption
+
+        Returns
+        -------
+        bytes used
+
+        Notes
+        -----
+        Memory usage does not include memory consumed by elements that
+        are not components of the array if deep=False
+
+        See Also
+        --------
+        numpy.ndarray.nbytes
+        """
+        return self._codes.nbytes + self.dtype.categories.memory_usage(deep=deep)
+
+    def isna(self) -> npt.NDArray[np.bool_]:
+        """
+        Detect missing values
+
+        Missing values (-1 in .codes) are detected.
+
+        Returns
+        -------
+        np.ndarray[bool] of whether my values are null
+
+        See Also
+        --------
+        isna : Top-level isna.
+        isnull : Alias of isna.
+        Categorical.notna : Boolean inverse of Categorical.isna.
+
+        """
+        return self._codes == -1
+
+    isnull = isna
+
+    def notna(self) -> npt.NDArray[np.bool_]:
+        """
+        Inverse of isna
+
+        Both missing values (-1 in .codes) and NA as a category are detected as
+        null.
+
+        Returns
+        -------
+        np.ndarray[bool] of whether my values are not null
+
+        See Also
+        --------
+        notna : Top-level notna.
+        notnull : Alias of notna.
+        Categorical.isna : Boolean inverse of Categorical.notna.
+
+        """
+        return ~self.isna()
+
+    notnull = notna
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Return a Series containing counts of each category.
+
+        Every category will have an entry, even those with a count of 0.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of NaN.
+
+        Returns
+        -------
+        counts : Series
+
+        See Also
+        --------
+        Series.value_counts
+        """
+        from pandas import (
+            CategoricalIndex,
+            Series,
+        )
+
+        code, cat = self._codes, self.categories
+        ncat, mask = (len(cat), code >= 0)
+        ix, clean = np.arange(ncat), mask.all()
+
+        if dropna or clean:
+            obs = code if clean else code[mask]
+            count = np.bincount(obs, minlength=ncat or 0)
+        else:
+            count = np.bincount(np.where(mask, code, ncat))
+            ix = np.append(ix, -1)
+
+        ix = coerce_indexer_dtype(ix, self.dtype.categories)
+        ix_categorical = self._from_backing_data(ix)
+
+        return Series(
+            count,
+            index=CategoricalIndex(ix_categorical),
+            dtype="int64",
+            name="count",
+            copy=False,
+        )
+
+    # error: Argument 2 of "_empty" is incompatible with supertype
+    # "NDArrayBackedExtensionArray"; supertype defines the argument type as
+    # "ExtensionDtype"
+    @classmethod
+    def _empty(  # type: ignore[override]
+        cls, shape: Shape, dtype: CategoricalDtype
+    ) -> Self:
+        """
+        Analogous to np.empty(shape, dtype=dtype)
+
+        Parameters
+        ----------
+        shape : tuple[int]
+        dtype : CategoricalDtype
+        """
+        arr = cls._from_sequence([], dtype=dtype)
+
+        # We have to use np.zeros instead of np.empty otherwise the resulting
+        #  ndarray may contain codes not supported by this dtype, in which
+        #  case repr(result) could segfault.
+        backing = np.zeros(shape, dtype=arr._ndarray.dtype)
+
+        return arr._from_backing_data(backing)
+
+    def _internal_get_values(self) -> ArrayLike:
+        """
+        Return the values.
+
+        For internal compatibility with pandas formatting.
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray
+            A numpy array or ExtensionArray of the same dtype as
+            categorical.categories.dtype.
+        """
+        # if we are a datetime and period index, return Index to keep metadata
+        if needs_i8_conversion(self.categories.dtype):
+            return self.categories.take(self._codes, fill_value=NaT)._values
+        elif is_integer_dtype(self.categories.dtype) and -1 in self._codes:
+            return (
+                self.categories.astype("object")
+                .take(self._codes, fill_value=np.nan)
+                ._values
+            )
+        return np.array(self)
+
+    def check_for_ordered(self, op) -> None:
+        """assert that we are ordered"""
+        if not self.ordered:
+            raise TypeError(
+                f"Categorical is not ordered for operation {op}\n"
+                "you can use .as_ordered() to change the "
+                "Categorical to an ordered one\n"
+            )
+
+    def argsort(
+        self, *, ascending: bool = True, kind: SortKind = "quicksort", **kwargs
+    ) -> npt.NDArray[np.intp]:
+        """
+        Return the indices that would sort the Categorical.
+
+        Missing values are sorted at the end.
+
+        Parameters
+        ----------
+        ascending : bool, default True
+            Whether the indices should result in an ascending
+            or descending sort.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, optional
+            Sorting algorithm.
+        **kwargs:
+            passed through to :func:`numpy.argsort`.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+
+        See Also
+        --------
+        numpy.ndarray.argsort
+
+        Notes
+        -----
+        While an ordering is applied to the category values, arg-sorting
+        in this context refers more to organizing and grouping together
+        based on matching category values. Thus, this function can be
+        called on an unordered Categorical instance unlike the functions
+        'Categorical.min' and 'Categorical.max'.
+
+        Examples
+        --------
+        >>> pd.Categorical(["b", "b", "a", "c"]).argsort()
+        array([2, 0, 1, 3])
+
+        >>> cat = pd.Categorical(
+        ...     ["b", "b", "a", "c"], categories=["c", "b", "a"], ordered=True
+        ... )
+        >>> cat.argsort()
+        array([3, 0, 1, 2])
+
+        Missing values are placed at the end
+
+        >>> cat = pd.Categorical([2, None, 1])
+        >>> cat.argsort()
+        array([2, 0, 1])
+        """
+        return super().argsort(ascending=ascending, kind=kind, **kwargs)
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        inplace: Literal[False] = ...,
+        ascending: bool = ...,
+        na_position: str = ...,
+    ) -> Self: ...
+
+    @overload
+    def sort_values(
+        self, *, inplace: Literal[True], ascending: bool = ..., na_position: str = ...
+    ) -> None: ...
+
+    def sort_values(
+        self,
+        *,
+        inplace: bool = False,
+        ascending: bool = True,
+        na_position: str = "last",
+    ) -> Self | None:
+        """
+        Sort the Categorical by category value returning a new
+        Categorical by default.
+
+        While an ordering is applied to the category values, sorting in this
+        context refers more to organizing and grouping together based on
+        matching category values. Thus, this function can be called on an
+        unordered Categorical instance unlike the functions 'Categorical.min'
+        and 'Categorical.max'.
+
+        Parameters
+        ----------
+        inplace : bool, default False
+            Do operation in place.
+        ascending : bool, default True
+            Order ascending. Passing False orders descending. The
+            ordering parameter provides the method by which the
+            category values are organized.
+        na_position : {'first', 'last'} (optional, default='last')
+            'first' puts NaNs at the beginning
+            'last' puts NaNs at the end
+
+        Returns
+        -------
+        Categorical or None
+
+        See Also
+        --------
+        Categorical.sort
+        Series.sort_values
+
+        Examples
+        --------
+        >>> c = pd.Categorical([1, 2, 2, 1, 5])
+        >>> c
+        [1, 2, 2, 1, 5]
+        Categories (3, int64): [1, 2, 5]
+        >>> c.sort_values()
+        [1, 1, 2, 2, 5]
+        Categories (3, int64): [1, 2, 5]
+        >>> c.sort_values(ascending=False)
+        [5, 2, 2, 1, 1]
+        Categories (3, int64): [1, 2, 5]
+
+        >>> c = pd.Categorical([1, 2, 2, 1, 5])
+
+        'sort_values' behaviour with NaNs. Note that 'na_position'
+        is independent of the 'ascending' parameter:
+
+        >>> c = pd.Categorical([np.nan, 2, 2, np.nan, 5])
+        >>> c
+        [NaN, 2, 2, NaN, 5]
+        Categories (2, int64): [2, 5]
+        >>> c.sort_values()
+        [2, 2, 5, NaN, NaN]
+        Categories (2, int64): [2, 5]
+        >>> c.sort_values(ascending=False)
+        [5, 2, 2, NaN, NaN]
+        Categories (2, int64): [2, 5]
+        >>> c.sort_values(na_position="first")
+        [NaN, NaN, 2, 2, 5]
+        Categories (2, int64): [2, 5]
+        >>> c.sort_values(ascending=False, na_position="first")
+        [NaN, NaN, 5, 2, 2]
+        Categories (2, int64): [2, 5]
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if na_position not in ["last", "first"]:
+            raise ValueError(f"invalid na_position: {na_position!r}")
+
+        sorted_idx = nargsort(self, ascending=ascending, na_position=na_position)
+
+        if not inplace:
+            codes = self._codes[sorted_idx]
+            return self._from_backing_data(codes)
+        self._codes[:] = self._codes[sorted_idx]
+        return None
+
+    def _rank(
+        self,
+        *,
+        axis: AxisInt = 0,
+        method: str = "average",
+        na_option: str = "keep",
+        ascending: bool = True,
+        pct: bool = False,
+    ):
+        """
+        See Series.rank.__doc__.
+        """
+        if axis != 0:
+            raise NotImplementedError
+        vff = self._values_for_rank()
+        return algorithms.rank(
+            vff,
+            axis=axis,
+            method=method,
+            na_option=na_option,
+            ascending=ascending,
+            pct=pct,
+        )
+
+    def _values_for_rank(self) -> np.ndarray:
+        """
+        For correctly ranking ordered categorical data. See GH#15420
+
+        Ordered categorical data should be ranked on the basis of
+        codes with -1 translated to NaN.
+
+        Returns
+        -------
+        numpy.array
+
+        """
+        from pandas import Series
+
+        if self.ordered:
+            values = self.codes
+            mask = values == -1
+            if mask.any():
+                values = values.astype("float64")
+                values[mask] = np.nan
+        elif is_any_real_numeric_dtype(self.categories.dtype):
+            values = np.array(self)
+        else:
+            #  reorder the categories (so rank can use the float codes)
+            #  instead of passing an object array to rank
+            values = np.array(
+                self.rename_categories(
+                    Series(self.categories, copy=False).rank().values
+                )
+            )
+        return values
+
+    def _hash_pandas_object(
+        self, *, encoding: str, hash_key: str, categorize: bool
+    ) -> npt.NDArray[np.uint64]:
+        """
+        Hash a Categorical by hashing its categories, and then mapping the codes
+        to the hashes.
+
+        Parameters
+        ----------
+        encoding : str
+        hash_key : str
+        categorize : bool
+            Ignored for Categorical.
+
+        Returns
+        -------
+        np.ndarray[uint64]
+        """
+        # Note we ignore categorize, as we are already Categorical.
+        from pandas.core.util.hashing import hash_array
+
+        # Convert ExtensionArrays to ndarrays
+        values = np.asarray(self.categories._values)
+        hashed = hash_array(values, encoding, hash_key, categorize=False)
+
+        # we have uint64, as we don't directly support missing values
+        # we don't want to use take_nd which will coerce to float
+        # instead, directly construct the result with a
+        # max(np.uint64) as the missing value indicator
+        #
+        # TODO: GH#15362
+
+        mask = self.isna()
+        if len(hashed):
+            result = hashed.take(self._codes)
+        else:
+            result = np.zeros(len(mask), dtype="uint64")
+
+        if mask.any():
+            result[mask] = lib.u8max
+
+        return result
+
+    # ------------------------------------------------------------------
+    # NDArrayBackedExtensionArray compat
+
+    @property
+    def _codes(self) -> np.ndarray:
+        return self._ndarray
+
+    def _box_func(self, i: int):
+        if i == -1:
+            return np.nan
+        return self.categories[i]
+
+    def _unbox_scalar(self, key) -> int:
+        # searchsorted is very performance sensitive. By converting codes
+        # to same dtype as self.codes, we get much faster performance.
+        code = self.categories.get_loc(key)
+        code = self._ndarray.dtype.type(code)
+        return code
+
+    # ------------------------------------------------------------------
+
+    def __iter__(self) -> Iterator:
+        """
+        Returns an Iterator over the values of this Categorical.
+        """
+        if self.ndim == 1:
+            return iter(self._internal_get_values().tolist())
+        else:
+            return (self[n] for n in range(len(self)))
+
+    def __contains__(self, key) -> bool:
+        """
+        Returns True if `key` is in this Categorical.
+        """
+        # if key is a NaN, check if any NaN is in self.
+        if is_valid_na_for_dtype(key, self.categories.dtype):
+            return bool(self.isna().any())
+
+        return contains(self, key, container=self._codes)
+
+    # ------------------------------------------------------------------
+    # Rendering Methods
+
+    # error: Return type "None" of "_formatter" incompatible with return
+    # type "Callable[[Any], str | None]" in supertype "ExtensionArray"
+    def _formatter(self, boxed: bool = False) -> None:  # type: ignore[override]
+        # Returning None here will cause format_array to do inference.
+        return None
+
+    def _repr_categories(self) -> list[str]:
+        """
+        return the base repr for the categories
+        """
+        max_categories = (
+            10
+            if get_option("display.max_categories") == 0
+            else get_option("display.max_categories")
+        )
+        from pandas.io.formats import format as fmt
+
+        format_array = partial(
+            fmt.format_array, formatter=None, quoting=QUOTE_NONNUMERIC
+        )
+        if len(self.categories) > max_categories:
+            num = max_categories // 2
+            head = format_array(self.categories[:num]._values)
+            tail = format_array(self.categories[-num:]._values)
+            category_strs = head + ["..."] + tail
+        else:
+            category_strs = format_array(self.categories._values)
+
+        # Strip all leading spaces, which format_array adds for columns...
+        category_strs = [x.strip() for x in category_strs]
+        return category_strs
+
+    def _get_repr_footer(self) -> str:
+        """
+        Returns a string representation of the footer.
+        """
+        category_strs = self._repr_categories()
+        dtype = str(self.categories.dtype)
+        levheader = f"Categories ({len(self.categories)}, {dtype}): "
+        width, _ = get_terminal_size()
+        max_width = get_option("display.width") or width
+        if console.in_ipython_frontend():
+            # 0 = no breaks
+            max_width = 0
+        levstring = ""
+        start = True
+        cur_col_len = len(levheader)  # header
+        sep_len, sep = (3, " < ") if self.ordered else (2, ", ")
+        linesep = f"{sep.rstrip()}\n"  # remove whitespace
+        for val in category_strs:
+            if max_width != 0 and cur_col_len + sep_len + len(val) > max_width:
+                levstring += linesep + (" " * (len(levheader) + 1))
+                cur_col_len = len(levheader) + 1  # header + a whitespace
+            elif not start:
+                levstring += sep
+                cur_col_len += len(val)
+            levstring += val
+            start = False
+        # replace to simple save space by
+        return f"{levheader}[{levstring.replace(' < ... < ', ' ... ')}]"
+
+    def _get_values_repr(self) -> str:
+        from pandas.io.formats import format as fmt
+
+        assert len(self) > 0
+
+        vals = self._internal_get_values()
+        fmt_values = fmt.format_array(
+            vals,
+            None,
+            float_format=None,
+            na_rep="NaN",
+            quoting=QUOTE_NONNUMERIC,
+        )
+
+        fmt_values = [i.strip() for i in fmt_values]
+        joined = ", ".join(fmt_values)
+        result = "[" + joined + "]"
+        return result
+
+    def __repr__(self) -> str:
+        """
+        String representation.
+        """
+        footer = self._get_repr_footer()
+        length = len(self)
+        max_len = 10
+        if length > max_len:
+            # In long cases we do not display all entries, so we add Length
+            #  information to the __repr__.
+            num = max_len // 2
+            head = self[:num]._get_values_repr()
+            tail = self[-(max_len - num) :]._get_values_repr()
+            body = f"{head[:-1]}, ..., {tail[1:]}"
+            length_info = f"Length: {len(self)}"
+            result = f"{body}\n{length_info}\n{footer}"
+        elif length > 0:
+            body = self._get_values_repr()
+            result = f"{body}\n{footer}"
+        else:
+            # In the empty case we use a comma instead of newline to get
+            #  a more compact __repr__
+            body = "[]"
+            result = f"{body}, {footer}"
+
+        return result
+
+    # ------------------------------------------------------------------
+
+    def _validate_listlike(self, value):
+        # NB: here we assume scalar-like tuples have already been excluded
+        value = extract_array(value, extract_numpy=True)
+
+        # require identical categories set
+        if isinstance(value, Categorical):
+            if self.dtype != value.dtype:
+                raise TypeError(
+                    "Cannot set a Categorical with another, "
+                    "without identical categories"
+                )
+            # dtype equality implies categories_match_up_to_permutation
+            value = self._encode_with_my_categories(value)
+            return value._codes
+
+        from pandas import Index
+
+        # tupleize_cols=False for e.g. test_fillna_iterable_category GH#41914
+        to_add = Index._with_infer(value, tupleize_cols=False).difference(
+            self.categories
+        )
+
+        # no assignments of values not in categories, but it's always ok to set
+        # something to np.nan
+        if len(to_add) and not isna(to_add).all():
+            raise TypeError(
+                "Cannot setitem on a Categorical with a new "
+                "category, set the categories first"
+            )
+
+        codes = self.categories.get_indexer(value)
+        return codes.astype(self._ndarray.dtype, copy=False)
+
+    def _reverse_indexer(self) -> dict[Hashable, npt.NDArray[np.intp]]:
+        """
+        Compute the inverse of a categorical, returning
+        a dict of categories -> indexers.
+
+        *This is an internal function*
+
+        Returns
+        -------
+        Dict[Hashable, np.ndarray[np.intp]]
+            dict of categories -> indexers
+
+        Examples
+        --------
+        >>> c = pd.Categorical(list("aabca"))
+        >>> c
+        ['a', 'a', 'b', 'c', 'a']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> c.categories
+        Index(['a', 'b', 'c'], dtype='object')
+        >>> c.codes
+        array([0, 0, 1, 2, 0], dtype=int8)
+        >>> c._reverse_indexer()
+        {'a': array([0, 1, 4]), 'b': array([2]), 'c': array([3])}
+
+        """
+        categories = self.categories
+        r, counts = libalgos.groupsort_indexer(
+            ensure_platform_int(self.codes), categories.size
+        )
+        counts = ensure_int64(counts).cumsum()
+        _result = (r[start:end] for start, end in zip(counts, counts[1:]))
+        return dict(zip(categories, _result))
+
+    # ------------------------------------------------------------------
+    # Reductions
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        result = super()._reduce(name, skipna=skipna, keepdims=keepdims, **kwargs)
+        if name in ["argmax", "argmin"]:
+            # don't wrap in Categorical!
+            return result
+        if keepdims:
+            return type(self)(result, dtype=self.dtype)
+        else:
+            return result
+
+    def min(self, *, skipna: bool = True, **kwargs):
+        """
+        The minimum value of the object.
+
+        Only ordered `Categoricals` have a minimum!
+
+        Raises
+        ------
+        TypeError
+            If the `Categorical` is not `ordered`.
+
+        Returns
+        -------
+        min : the minimum of this `Categorical`, NA value if empty
+        """
+        nv.validate_minmax_axis(kwargs.get("axis", 0))
+        nv.validate_min((), kwargs)
+        self.check_for_ordered("min")
+
+        if not len(self._codes):
+            return self.dtype.na_value
+
+        good = self._codes != -1
+        if not good.all():
+            if skipna and good.any():
+                pointer = self._codes[good].min()
+            else:
+                return np.nan
+        else:
+            pointer = self._codes.min()
+        return self._wrap_reduction_result(None, pointer)
+
+    def max(self, *, skipna: bool = True, **kwargs):
+        """
+        The maximum value of the object.
+
+        Only ordered `Categoricals` have a maximum!
+
+        Raises
+        ------
+        TypeError
+            If the `Categorical` is not `ordered`.
+
+        Returns
+        -------
+        max : the maximum of this `Categorical`, NA if array is empty
+        """
+        nv.validate_minmax_axis(kwargs.get("axis", 0))
+        nv.validate_max((), kwargs)
+        self.check_for_ordered("max")
+
+        if not len(self._codes):
+            return self.dtype.na_value
+
+        good = self._codes != -1
+        if not good.all():
+            if skipna and good.any():
+                pointer = self._codes[good].max()
+            else:
+                return np.nan
+        else:
+            pointer = self._codes.max()
+        return self._wrap_reduction_result(None, pointer)
+
+    def _mode(self, dropna: bool = True) -> Categorical:
+        codes = self._codes
+        mask = None
+        if dropna:
+            mask = self.isna()
+
+        res_codes = algorithms.mode(codes, mask=mask)
+        res_codes = cast(np.ndarray, res_codes)
+        assert res_codes.dtype == codes.dtype
+        res = self._from_backing_data(res_codes)
+        return res
+
+    # ------------------------------------------------------------------
+    # ExtensionArray Interface
+
+    def unique(self) -> Self:
+        """
+        Return the ``Categorical`` which ``categories`` and ``codes`` are
+        unique.
+
+        .. versionchanged:: 1.3.0
+
+            Previously, unused categories were dropped from the new categories.
+
+        Returns
+        -------
+        Categorical
+
+        See Also
+        --------
+        pandas.unique
+        CategoricalIndex.unique
+        Series.unique : Return unique values of Series object.
+
+        Examples
+        --------
+        >>> pd.Categorical(list("baabc")).unique()
+        ['b', 'a', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> pd.Categorical(list("baab"), categories=list("abc"), ordered=True).unique()
+        ['b', 'a']
+        Categories (3, object): ['a' < 'b' < 'c']
+        """
+        return super().unique()
+
+    def equals(self, other: object) -> bool:
+        """
+        Returns True if categorical arrays are equal.
+
+        Parameters
+        ----------
+        other : `Categorical`
+
+        Returns
+        -------
+        bool
+        """
+        if not isinstance(other, Categorical):
+            return False
+        elif self._categories_match_up_to_permutation(other):
+            other = self._encode_with_my_categories(other)
+            return np.array_equal(self._codes, other._codes)
+        return False
+
+    def _accumulate(self, name: str, skipna: bool = True, **kwargs) -> Self:
+        func: Callable
+        if name == "cummin":
+            func = np.minimum.accumulate
+        elif name == "cummax":
+            func = np.maximum.accumulate
+        else:
+            raise TypeError(f"Accumulation {name} not supported for {type(self)}")
+        self.check_for_ordered(name)
+
+        codes = self.codes.copy()
+        mask = self.isna()
+        if func == np.minimum.accumulate:
+            codes[mask] = np.iinfo(codes.dtype.type).max
+        # no need to change codes for maximum because codes[mask] is already -1
+        if not skipna:
+            mask = np.maximum.accumulate(mask)
+
+        codes = func(codes)
+        codes[mask] = -1
+        return self._simple_new(codes, dtype=self._dtype)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat: Sequence[Self], axis: AxisInt = 0) -> Self:
+        from pandas.core.dtypes.concat import union_categoricals
+
+        first = to_concat[0]
+        if axis >= first.ndim:
+            raise ValueError(
+                f"axis {axis} is out of bounds for array of dimension {first.ndim}"
+            )
+
+        if axis == 1:
+            # Flatten, concatenate then reshape
+            if not all(x.ndim == 2 for x in to_concat):
+                raise ValueError
+
+            # pass correctly-shaped to union_categoricals
+            tc_flat = []
+            for obj in to_concat:
+                tc_flat.extend([obj[:, i] for i in range(obj.shape[1])])
+
+            res_flat = cls._concat_same_type(tc_flat, axis=0)
+
+            result = res_flat.reshape(len(first), -1, order="F")
+            return result
+
+        # error: Incompatible types in assignment (expression has type "Categorical",
+        # variable has type "Self")
+        result = union_categoricals(to_concat)  # type: ignore[assignment]
+        return result
+
+    # ------------------------------------------------------------------
+
+    def _encode_with_my_categories(self, other: Categorical) -> Categorical:
+        """
+        Re-encode another categorical using this Categorical's categories.
+
+        Notes
+        -----
+        This assumes we have already checked
+        self._categories_match_up_to_permutation(other).
+        """
+        # Indexing on codes is more efficient if categories are the same,
+        #  so we can apply some optimizations based on the degree of
+        #  dtype-matching.
+        codes = recode_for_categories(
+            other.codes, other.categories, self.categories, copy=False
+        )
+        return self._from_backing_data(codes)
+
+    def _categories_match_up_to_permutation(self, other: Categorical) -> bool:
+        """
+        Returns True if categoricals are the same dtype
+          same categories, and same ordered
+
+        Parameters
+        ----------
+        other : Categorical
+
+        Returns
+        -------
+        bool
+        """
+        return hash(self.dtype) == hash(other.dtype)
+
+    def describe(self) -> DataFrame:
+        """
+        Describes this Categorical
+
+        Returns
+        -------
+        description: `DataFrame`
+            A dataframe with frequency and counts by category.
+        """
+        counts = self.value_counts(dropna=False)
+        freqs = counts / counts.sum()
+
+        from pandas import Index
+        from pandas.core.reshape.concat import concat
+
+        result = concat([counts, freqs], ignore_index=True, axis=1)
+        result.columns = Index(["counts", "freqs"])
+        result.index.name = "categories"
+
+        return result
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        """
+        Check whether `values` are contained in Categorical.
+
+        Return a boolean NumPy Array showing whether each element in
+        the Categorical matches an element in the passed sequence of
+        `values` exactly.
+
+        Parameters
+        ----------
+        values : np.ndarray or ExtensionArray
+            The sequence of values to test. Passing in a single string will
+            raise a ``TypeError``. Instead, turn a single string into a
+            list of one element.
+
+        Returns
+        -------
+        np.ndarray[bool]
+
+        Raises
+        ------
+        TypeError
+          * If `values` is not a set or list-like
+
+        See Also
+        --------
+        pandas.Series.isin : Equivalent method on Series.
+
+        Examples
+        --------
+        >>> s = pd.Categorical(["llama", "cow", "llama", "beetle", "llama", "hippo"])
+        >>> s.isin(["cow", "llama"])
+        array([ True,  True,  True, False,  True, False])
+
+        Passing a single string as ``s.isin('llama')`` will raise an error. Use
+        a list of one element instead:
+
+        >>> s.isin(["llama"])
+        array([ True, False,  True, False,  True, False])
+        """
+        null_mask = np.asarray(isna(values))
+        code_values = self.categories.get_indexer_for(values)
+        code_values = code_values[null_mask | (code_values >= 0)]
+        return algorithms.isin(self.codes, code_values)
+
+    # ------------------------------------------------------------------------
+    # String methods interface
+    def _str_map(
+        self, f, na_value=lib.no_default, dtype=np.dtype("object"), convert: bool = True
+    ):
+        # Optimization to apply the callable `f` to the categories once
+        # and rebuild the result by `take`ing from the result with the codes.
+        # Returns the same type as the object-dtype implementation though.
+        categories = self.categories
+        codes = self.codes
+        if categories.dtype == "string":
+            result = categories.array._str_map(f, na_value, dtype)  # type: ignore[attr-defined]
+            if (
+                categories.dtype.na_value is np.nan  # type: ignore[union-attr]
+                and is_bool_dtype(dtype)
+                and (na_value is lib.no_default or isna(na_value))
+            ):
+                # NaN propagates as False for functions with boolean return type
+                na_value = False
+        else:
+            from pandas.core.arrays import NumpyExtensionArray
+
+            result = NumpyExtensionArray(categories.to_numpy())._str_map(
+                f, na_value, dtype
+            )
+        return take_nd(result, codes, fill_value=na_value)
+
+    def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
+        # sep may not be in categories. Just bail on this.
+        from pandas.core.arrays import NumpyExtensionArray
+
+        return NumpyExtensionArray(self.to_numpy(str, na_value="NaN"))._str_get_dummies(
+            sep, dtype
+        )
+
+    # ------------------------------------------------------------------------
+    # GroupBy Methods
+
+    def _groupby_op(
+        self,
+        *,
+        how: str,
+        has_dropped_na: bool,
+        min_count: int,
+        ngroups: int,
+        ids: npt.NDArray[np.intp],
+        **kwargs,
+    ):
+        from pandas.core.groupby.ops import WrappedCythonOp
+
+        kind = WrappedCythonOp.get_kind_from_how(how)
+        op = WrappedCythonOp(how=how, kind=kind, has_dropped_na=has_dropped_na)
+
+        dtype = self.dtype
+        if how in ["sum", "prod", "cumsum", "cumprod", "skew", "kurt"]:
+            raise TypeError(f"{dtype} type does not support {how} operations")
+        if how in ["min", "max", "rank", "idxmin", "idxmax"] and not dtype.ordered:
+            # raise TypeError instead of NotImplementedError to ensure we
+            #  don't go down a group-by-group path, since in the empty-groups
+            #  case that would fail to raise
+            raise TypeError(f"Cannot perform {how} with non-ordered Categorical")
+        if how not in [
+            "rank",
+            "any",
+            "all",
+            "first",
+            "last",
+            "min",
+            "max",
+            "idxmin",
+            "idxmax",
+        ]:
+            if kind == "transform":
+                raise TypeError(f"{dtype} type does not support {how} operations")
+            raise TypeError(f"{dtype} dtype does not support aggregation '{how}'")
+
+        result_mask = None
+        mask = self.isna()
+        if how == "rank":
+            assert self.ordered  # checked earlier
+            npvalues = self._ndarray
+        elif how in ["first", "last", "min", "max", "idxmin", "idxmax"]:
+            npvalues = self._ndarray
+            result_mask = np.zeros(ngroups, dtype=bool)
+        else:
+            # any/all
+            npvalues = self.astype(bool)
+
+        res_values = op._cython_op_ndim_compat(
+            npvalues,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=ids,
+            mask=mask,
+            result_mask=result_mask,
+            **kwargs,
+        )
+
+        if how in op.cast_blocklist:
+            return res_values
+        elif how in ["first", "last", "min", "max"]:
+            res_values[result_mask == 1] = -1
+        return self._from_backing_data(res_values)
+
+
+# The Series.cat accessor
+
+
+@delegate_names(
+    delegate=Categorical, accessors=["categories", "ordered"], typ="property"
+)
+@delegate_names(
+    delegate=Categorical,
+    accessors=[
+        "rename_categories",
+        "reorder_categories",
+        "add_categories",
+        "remove_categories",
+        "remove_unused_categories",
+        "set_categories",
+        "as_ordered",
+        "as_unordered",
+    ],
+    typ="method",
+)
+class CategoricalAccessor(PandasDelegate, PandasObject, NoNewAttributesMixin):
+    """
+    Accessor object for categorical properties of the Series values.
+
+    Parameters
+    ----------
+    data : Series or CategoricalIndex
+        The object to which the categorical accessor is attached.
+
+    See Also
+    --------
+    Series.dt : Accessor object for datetimelike properties of the Series values.
+    Series.sparse : Accessor for sparse matrix data types.
+
+    Examples
+    --------
+    >>> s = pd.Series(list("abbccc")).astype("category")
+    >>> s
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (3, object): ['a', 'b', 'c']
+
+    >>> s.cat.categories
+    Index(['a', 'b', 'c'], dtype='object')
+
+    >>> s.cat.rename_categories(list("cba"))
+    0    c
+    1    b
+    2    b
+    3    a
+    4    a
+    5    a
+    dtype: category
+    Categories (3, object): ['c', 'b', 'a']
+
+    >>> s.cat.reorder_categories(list("cba"))
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (3, object): ['c', 'b', 'a']
+
+    >>> s.cat.add_categories(["d", "e"])
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (5, object): ['a', 'b', 'c', 'd', 'e']
+
+    >>> s.cat.remove_categories(["a", "c"])
+    0    NaN
+    1      b
+    2      b
+    3    NaN
+    4    NaN
+    5    NaN
+    dtype: category
+    Categories (1, object): ['b']
+
+    >>> s1 = s.cat.add_categories(["d", "e"])
+    >>> s1.cat.remove_unused_categories()
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (3, object): ['a', 'b', 'c']
+
+    >>> s.cat.set_categories(list("abcde"))
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (5, object): ['a', 'b', 'c', 'd', 'e']
+
+    >>> s.cat.as_ordered()
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (3, object): ['a' < 'b' < 'c']
+
+    >>> s.cat.as_unordered()
+    0    a
+    1    b
+    2    b
+    3    c
+    4    c
+    5    c
+    dtype: category
+    Categories (3, object): ['a', 'b', 'c']
+    """
+
+    def __init__(self, data) -> None:
+        self._validate(data)
+        self._parent = data.values
+        self._index = data.index
+        self._name = data.name
+        self._freeze()
+
+    @staticmethod
+    def _validate(data) -> None:
+        if not isinstance(data.dtype, CategoricalDtype):
+            raise AttributeError("Can only use .cat accessor with a 'category' dtype")
+
+    def _delegate_property_get(self, name: str):
+        return getattr(self._parent, name)
+
+    # error: Signature of "_delegate_property_set" incompatible with supertype
+    # "PandasDelegate"
+    def _delegate_property_set(self, name: str, new_values) -> None:  # type: ignore[override]
+        setattr(self._parent, name, new_values)
+
+    @property
+    def codes(self) -> Series:
+        """
+        Return Series of codes as well as the index.
+
+        See Also
+        --------
+        Series.cat.categories : Return the categories of this categorical.
+        Series.cat.as_ordered : Set the Categorical to be ordered.
+        Series.cat.as_unordered : Set the Categorical to be unordered.
+
+        Examples
+        --------
+        >>> raw_cate = pd.Categorical(["a", "b", "c", "a"], categories=["a", "b"])
+        >>> ser = pd.Series(raw_cate)
+        >>> ser.cat.codes
+        0   0
+        1   1
+        2  -1
+        3   0
+        dtype: int8
+        """
+        from pandas import Series
+
+        return Series(self._parent.codes, index=self._index)
+
+    def _delegate_method(self, name: str, *args, **kwargs):
+        from pandas import Series
+
+        method = getattr(self._parent, name)
+        res = method(*args, **kwargs)
+        if res is not None:
+            return Series(res, index=self._index, name=self._name)
+
+
+# utility routines
+
+
+def _get_codes_for_values(
+    values: Index | Series | ExtensionArray | np.ndarray,
+    categories: Index,
+) -> np.ndarray:
+    """
+    utility routine to turn values into codes given the specified categories
+
+    If `values` is known to be a Categorical, use recode_for_categories instead.
+    """
+    codes = categories.get_indexer_for(values)
+    return coerce_indexer_dtype(codes, categories)
+
+
+def recode_for_categories(
+    codes: np.ndarray, old_categories, new_categories, copy: bool = True
+) -> np.ndarray:
+    """
+    Convert a set of codes for to a new set of categories
+
+    Parameters
+    ----------
+    codes : np.ndarray
+    old_categories, new_categories : Index
+    copy: bool, default True
+        Whether to copy if the codes are unchanged.
+
+    Returns
+    -------
+    new_codes : np.ndarray[np.int64]
+
+    Examples
+    --------
+    >>> old_cat = pd.Index(["b", "a", "c"])
+    >>> new_cat = pd.Index(["a", "b"])
+    >>> codes = np.array([0, 1, 1, 2])
+    >>> recode_for_categories(codes, old_cat, new_cat)
+    array([ 1,  0,  0, -1], dtype=int8)
+    """
+    if len(old_categories) == 0:
+        # All null anyway, so just retain the nulls
+        if copy:
+            return codes.copy()
+        return codes
+    elif new_categories.equals(old_categories):
+        # Same categories, so no need to actually recode
+        if copy:
+            return codes.copy()
+        return codes
+
+    indexer = coerce_indexer_dtype(
+        new_categories.get_indexer_for(old_categories), new_categories
+    )
+    new_codes = take_nd(indexer, codes, fill_value=-1)
+    return new_codes
+
+
+def factorize_from_iterable(values) -> tuple[np.ndarray, Index]:
+    """
+    Factorize an input `values` into `categories` and `codes`. Preserves
+    categorical dtype in `categories`.
+
+    Parameters
+    ----------
+    values : list-like
+
+    Returns
+    -------
+    codes : ndarray
+    categories : Index
+        If `values` has a categorical dtype, then `categories` is
+        a CategoricalIndex keeping the categories and order of `values`.
+    """
+    from pandas import CategoricalIndex
+
+    if not is_list_like(values):
+        raise TypeError("Input must be list-like")
+
+    categories: Index
+
+    vdtype = getattr(values, "dtype", None)
+    if isinstance(vdtype, CategoricalDtype):
+        values = extract_array(values)
+        # The Categorical we want to build has the same categories
+        # as values but its codes are by def [0, ..., len(n_categories) - 1]
+        cat_codes = np.arange(len(values.categories), dtype=values.codes.dtype)
+        cat = Categorical.from_codes(cat_codes, dtype=values.dtype, validate=False)
+
+        categories = CategoricalIndex(cat)
+        codes = values.codes
+    else:
+        # The value of ordered is irrelevant since we don't use cat as such,
+        # but only the resulting categories, the order of which is independent
+        # from ordered. Set ordered to False as default. See GH #15457
+        cat = Categorical(values, ordered=False)
+        categories = cat.categories
+        codes = cat.codes
+    return codes, categories
+
+
+def factorize_from_iterables(iterables) -> tuple[list[np.ndarray], list[Index]]:
+    """
+    A higher-level wrapper over `factorize_from_iterable`.
+
+    Parameters
+    ----------
+    iterables : list-like of list-likes
+
+    Returns
+    -------
+    codes : list of ndarrays
+    categories : list of Indexes
+
+    Notes
+    -----
+    See `factorize_from_iterable` for more info.
+    """
+    if len(iterables) == 0:
+        # For consistency, it should return two empty lists.
+        return [], []
+
+    codes, categories = zip(*(factorize_from_iterable(it) for it in iterables))
+    return list(codes), list(categories)
diff --git a/pandas/pandas/core/arrays/datetimelike.py b/pandas/pandas/core/arrays/datetimelike.py
new file mode 100644
index 0000000000000..8a79ab53442c3
--- /dev/null
+++ b/pandas/pandas/core/arrays/datetimelike.py
@@ -0,0 +1,2554 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+from functools import wraps
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    Union,
+    cast,
+    final,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+from pandas._config.config import get_option
+
+from pandas._libs import (
+    algos,
+    lib,
+)
+from pandas._libs.tslibs import (
+    BaseOffset,
+    IncompatibleFrequency,
+    NaT,
+    NaTType,
+    Period,
+    Resolution,
+    Tick,
+    Timedelta,
+    Timestamp,
+    add_overflowsafe,
+    astype_overflowsafe,
+    get_unit_from_dtype,
+    iNaT,
+    ints_to_pydatetime,
+    ints_to_pytimedelta,
+    periods_per_day,
+    to_offset,
+)
+from pandas._libs.tslibs.fields import (
+    RoundTo,
+    round_nsint64,
+)
+from pandas._libs.tslibs.np_datetime import compare_mismatched_resolutions
+from pandas._libs.tslibs.timedeltas import get_unit_for_round
+from pandas._libs.tslibs.timestamps import integer_op_not_supported
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    DatetimeLikeScalar,
+    Dtype,
+    DtypeObj,
+    F,
+    InterpolateOptions,
+    NpDtype,
+    PositionalIndexer2D,
+    PositionalIndexerTuple,
+    ScalarIndexer,
+    Self,
+    SequenceIndexer,
+    TakeIndexer,
+    TimeAmbiguous,
+    TimeNonexistent,
+    npt,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    AbstractMethodError,
+    InvalidComparison,
+    PerformanceWarning,
+)
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    cache_readonly,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+from pandas.core.dtypes.common import (
+    is_all_strings,
+    is_integer_dtype,
+    is_list_like,
+    is_object_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCCategorical,
+    ABCMultiIndex,
+)
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+)
+
+from pandas.core import (
+    algorithms,
+    missing,
+    nanops,
+    ops,
+)
+from pandas.core.algorithms import (
+    isin,
+    map_array,
+    unique1d,
+)
+from pandas.core.array_algos import datetimelike_accumulations
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays._mixins import (
+    NDArrayBackedExtensionArray,
+    ravel_compat,
+)
+from pandas.core.arrays.arrow.array import ArrowExtensionArray
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.arrays.integer import IntegerArray
+import pandas.core.common as com
+from pandas.core.construction import (
+    array as pd_array,
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import (
+    check_array_indexer,
+    check_setitem_lengths,
+)
+from pandas.core.ops.common import unpack_zerodim_and_defer
+from pandas.core.ops.invalid import (
+    invalid_comparison,
+    make_invalid_op,
+)
+
+from pandas.tseries import frequencies
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Iterator,
+        Sequence,
+    )
+
+    from pandas import Index
+    from pandas.core.arrays import (
+        DatetimeArray,
+        PeriodArray,
+        TimedeltaArray,
+    )
+
+DTScalarOrNaT = Union[DatetimeLikeScalar, NaTType]
+
+
+def _make_unpacked_invalid_op(op_name: str):
+    op = make_invalid_op(op_name)
+    return unpack_zerodim_and_defer(op_name)(op)
+
+
+def _period_dispatch(meth: F) -> F:
+    """
+    For PeriodArray methods, dispatch to DatetimeArray and re-wrap the results
+    in PeriodArray.  We cannot use ._ndarray directly for the affected
+    methods because the i8 data has different semantics on NaT values.
+    """
+
+    @wraps(meth)
+    def new_meth(self, *args, **kwargs):
+        if not isinstance(self.dtype, PeriodDtype):
+            return meth(self, *args, **kwargs)
+
+        arr = self.view("M8[ns]")
+        result = meth(arr, *args, **kwargs)
+        if result is NaT:
+            return NaT
+        elif isinstance(result, Timestamp):
+            return self._box_func(result._value)
+
+        res_i8 = result.view("i8")
+        return self._from_backing_data(res_i8)
+
+    return cast(F, new_meth)
+
+
+# error: Definition of "_concat_same_type" in base class "NDArrayBacked" is
+# incompatible with definition in base class "ExtensionArray"
+class DatetimeLikeArrayMixin(  # type: ignore[misc]
+    OpsMixin, NDArrayBackedExtensionArray
+):
+    """
+    Shared Base/Mixin class for DatetimeArray, TimedeltaArray, PeriodArray
+
+    Assumes that __new__/__init__ defines:
+        _ndarray
+
+    and that inheriting subclass implements:
+        freq
+    """
+
+    # _infer_matches -> which infer_dtype strings are close enough to our own
+    _infer_matches: tuple[str, ...]
+    _is_recognized_dtype: Callable[[DtypeObj], bool]
+    _recognized_scalars: tuple[type, ...]
+    _ndarray: np.ndarray
+    freq: BaseOffset | None
+
+    @cache_readonly
+    def _can_hold_na(self) -> bool:
+        return True
+
+    def __init__(
+        self, data, dtype: Dtype | None = None, freq=None, copy: bool = False
+    ) -> None:
+        raise AbstractMethodError(self)
+
+    @property
+    def _scalar_type(self) -> type[DatetimeLikeScalar]:
+        """
+        The scalar associated with this datelike
+
+        * PeriodArray : Period
+        * DatetimeArray : Timestamp
+        * TimedeltaArray : Timedelta
+        """
+        raise AbstractMethodError(self)
+
+    def _scalar_from_string(self, value: str) -> DTScalarOrNaT:
+        """
+        Construct a scalar type from a string.
+
+        Parameters
+        ----------
+        value : str
+
+        Returns
+        -------
+        Period, Timestamp, or Timedelta, or NaT
+            Whatever the type of ``self._scalar_type`` is.
+
+        Notes
+        -----
+        This should call ``self._check_compatible_with`` before
+        unboxing the result.
+        """
+        raise AbstractMethodError(self)
+
+    def _unbox_scalar(
+        self, value: DTScalarOrNaT
+    ) -> np.int64 | np.datetime64 | np.timedelta64:
+        """
+        Unbox the integer value of a scalar `value`.
+
+        Parameters
+        ----------
+        value : Period, Timestamp, Timedelta, or NaT
+            Depending on subclass.
+
+        Returns
+        -------
+        int
+
+        Examples
+        --------
+        >>> arr = pd.array(np.array(["1970-01-01"], "datetime64[ns]"))
+        >>> arr._unbox_scalar(arr[0])
+        numpy.datetime64('1970-01-01T00:00:00.000000000')
+        """
+        raise AbstractMethodError(self)
+
+    def _check_compatible_with(self, other: DTScalarOrNaT) -> None:
+        """
+        Verify that `self` and `other` are compatible.
+
+        * DatetimeArray verifies that the timezones (if any) match
+        * PeriodArray verifies that the freq matches
+        * Timedelta has no verification
+
+        In each case, NaT is considered compatible.
+
+        Parameters
+        ----------
+        other
+
+        Raises
+        ------
+        Exception
+        """
+        raise AbstractMethodError(self)
+
+    # ------------------------------------------------------------------
+
+    def _box_func(self, x):
+        """
+        box function to get object from internal representation
+        """
+        raise AbstractMethodError(self)
+
+    def _box_values(self, values) -> np.ndarray:
+        """
+        apply box func to passed values
+        """
+        return lib.map_infer(values, self._box_func, convert=False)
+
+    def __iter__(self) -> Iterator:
+        if self.ndim > 1:
+            return (self[n] for n in range(len(self)))
+        else:
+            return (self._box_func(v) for v in self.asi8)
+
+    @property
+    def asi8(self) -> npt.NDArray[np.int64]:
+        """
+        Integer representation of the values.
+
+        Returns
+        -------
+        ndarray
+            An ndarray with int64 dtype.
+        """
+        # do not cache or you'll create a memory leak
+        return self._ndarray.view("i8")
+
+    # ----------------------------------------------------------------
+    # Rendering Methods
+
+    def _format_native_types(
+        self, *, na_rep: str | float = "NaT", date_format=None
+    ) -> npt.NDArray[np.object_]:
+        """
+        Helper method for astype when converting to strings.
+
+        Returns
+        -------
+        ndarray[str]
+        """
+        raise AbstractMethodError(self)
+
+    def _formatter(self, boxed: bool = False) -> Callable[[object], str]:
+        # TODO: Remove Datetime & DatetimeTZ formatters.
+        return "'{}'".format
+
+    # ----------------------------------------------------------------
+    # Array-Like / EA-Interface Methods
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        # used for Timedelta/DatetimeArray, overwritten by PeriodArray
+        if is_object_dtype(dtype):
+            if copy is False:
+                raise ValueError(
+                    "Unable to avoid copy while creating an array as requested."
+                )
+            return np.array(list(self), dtype=object)
+
+        if copy is True:
+            return np.array(self._ndarray, dtype=dtype)
+        return self._ndarray
+
+    @overload
+    def __getitem__(self, key: ScalarIndexer) -> DTScalarOrNaT: ...
+
+    @overload
+    def __getitem__(
+        self,
+        key: SequenceIndexer | PositionalIndexerTuple,
+    ) -> Self: ...
+
+    def __getitem__(self, key: PositionalIndexer2D) -> Self | DTScalarOrNaT:
+        """
+        This getitem defers to the underlying array, which by-definition can
+        only handle list-likes, slices, and integer scalars
+        """
+        # Use cast as we know we will get back a DatetimeLikeArray or DTScalar,
+        # but skip evaluating the Union at runtime for performance
+        # (see https://github.com/pandas-dev/pandas/pull/44624)
+        result = cast("Union[Self, DTScalarOrNaT]", super().__getitem__(key))
+        if lib.is_scalar(result):
+            return result
+        else:
+            # At this point we know the result is an array.
+            result = cast(Self, result)
+        result._freq = self._get_getitem_freq(key)
+        return result
+
+    def _get_getitem_freq(self, key) -> BaseOffset | None:
+        """
+        Find the `freq` attribute to assign to the result of a __getitem__ lookup.
+        """
+        is_period = isinstance(self.dtype, PeriodDtype)
+        if is_period:
+            freq = self.freq
+        elif self.ndim != 1:
+            freq = None
+        else:
+            key = check_array_indexer(self, key)  # maybe ndarray[bool] -> slice
+            freq = None
+            if isinstance(key, slice):
+                if self.freq is not None and key.step is not None:
+                    freq = key.step * self.freq
+                else:
+                    freq = self.freq
+            elif key is Ellipsis:
+                # GH#21282 indexing with Ellipsis is similar to a full slice,
+                #  should preserve `freq` attribute
+                freq = self.freq
+            elif com.is_bool_indexer(key):
+                new_key = lib.maybe_booleans_to_slice(key.view(np.uint8))
+                if isinstance(new_key, slice):
+                    return self._get_getitem_freq(new_key)
+        return freq
+
+    # error: Argument 1 of "__setitem__" is incompatible with supertype
+    # "ExtensionArray"; supertype defines the argument type as "Union[int,
+    # ndarray]"
+    def __setitem__(
+        self,
+        key: int | Sequence[int] | Sequence[bool] | slice,
+        value: NaTType | Any | Sequence[Any],
+    ) -> None:
+        # I'm fudging the types a bit here. "Any" above really depends
+        # on type(self). For PeriodArray, it's Period (or stuff coercible
+        # to a period in from_sequence). For DatetimeArray, it's Timestamp...
+        # I don't know if mypy can do that, possibly with Generics.
+        # https://mypy.readthedocs.io/en/latest/generics.html
+
+        no_op = check_setitem_lengths(key, value, self)
+
+        # Calling super() before the no_op short-circuit means that we raise
+        #  on invalid 'value' even if this is a no-op, e.g. wrong-dtype empty array.
+        super().__setitem__(key, value)
+
+        if no_op:
+            return
+
+        self._maybe_clear_freq()
+
+    def _maybe_clear_freq(self) -> None:
+        # inplace operations like __setitem__ may invalidate the freq of
+        # DatetimeArray and TimedeltaArray
+        pass
+
+    def astype(self, dtype, copy: bool = True):
+        # Some notes on cases we don't have to handle here in the base class:
+        #   1. PeriodArray.astype handles period -> period
+        #   2. DatetimeArray.astype handles conversion between tz.
+        #   3. DatetimeArray.astype handles datetime -> period
+        dtype = pandas_dtype(dtype)
+
+        if dtype == object:
+            if self.dtype.kind == "M":
+                self = cast("DatetimeArray", self)
+                # *much* faster than self._box_values
+                #  for e.g. test_get_loc_tuple_monotonic_above_size_cutoff
+                i8data = self.asi8
+                converted = ints_to_pydatetime(
+                    i8data,
+                    tz=self.tz,
+                    box="timestamp",
+                    reso=self._creso,
+                )
+                return converted
+
+            elif self.dtype.kind == "m":
+                return ints_to_pytimedelta(self._ndarray, box=True)
+
+            return self._box_values(self.asi8.ravel()).reshape(self.shape)
+
+        elif is_string_dtype(dtype):
+            if isinstance(dtype, ExtensionDtype):
+                arr_object = self._format_native_types(na_rep=dtype.na_value)  # type: ignore[arg-type]
+                cls = dtype.construct_array_type()
+                return cls._from_sequence(arr_object, dtype=dtype, copy=False)
+            else:
+                return self._format_native_types()
+
+        elif isinstance(dtype, ExtensionDtype):
+            return super().astype(dtype, copy=copy)
+        elif dtype.kind in "iu":
+            # we deliberately ignore int32 vs. int64 here.
+            # See https://github.com/pandas-dev/pandas/issues/24381 for more.
+            values = self.asi8
+            if dtype != np.int64:
+                raise TypeError(
+                    f"Converting from {self.dtype} to {dtype} is not supported. "
+                    "Do obj.astype('int64').astype(dtype) instead"
+                )
+
+            if copy:
+                values = values.copy()
+            return values
+        elif (dtype.kind in "mM" and self.dtype != dtype) or dtype.kind == "f":
+            # disallow conversion between datetime/timedelta,
+            # and conversions for any datetimelike to float
+            msg = f"Cannot cast {type(self).__name__} to dtype {dtype}"
+            raise TypeError(msg)
+        else:
+            return np.asarray(self, dtype=dtype)
+
+    @overload
+    def view(self) -> Self: ...
+
+    @overload
+    def view(self, dtype: Literal["M8[ns]"]) -> DatetimeArray: ...
+
+    @overload
+    def view(self, dtype: Literal["m8[ns]"]) -> TimedeltaArray: ...
+
+    @overload
+    def view(self, dtype: Dtype | None = ...) -> ArrayLike: ...
+
+    def view(self, dtype: Dtype | None = None) -> ArrayLike:
+        # we need to explicitly call super() method as long as the `@overload`s
+        #  are present in this file.
+        return super().view(dtype)
+
+    # ------------------------------------------------------------------
+    # Validation Methods
+    # TODO: try to de-duplicate these, ensure identical behavior
+
+    def _validate_comparison_value(self, other):
+        if isinstance(other, str):
+            try:
+                # GH#18435 strings get a pass from tzawareness compat
+                other = self._scalar_from_string(other)
+            except (ValueError, IncompatibleFrequency) as err:
+                # failed to parse as Timestamp/Timedelta/Period
+                raise InvalidComparison(other) from err
+
+        if isinstance(other, self._recognized_scalars) or other is NaT:
+            other = self._scalar_type(other)
+            try:
+                self._check_compatible_with(other)
+            except (TypeError, IncompatibleFrequency) as err:
+                # e.g. tzawareness mismatch
+                raise InvalidComparison(other) from err
+
+        elif not is_list_like(other):
+            raise InvalidComparison(other)
+
+        elif len(other) != len(self):
+            raise ValueError("Lengths must match")
+
+        else:
+            try:
+                other = self._validate_listlike(other, allow_object=True)
+                self._check_compatible_with(other)
+            except (TypeError, IncompatibleFrequency) as err:
+                if is_object_dtype(getattr(other, "dtype", None)):
+                    # We will have to operate element-wise
+                    pass
+                else:
+                    raise InvalidComparison(other) from err
+
+        return other
+
+    def _validate_scalar(
+        self,
+        value,
+        *,
+        allow_listlike: bool = False,
+        unbox: bool = True,
+    ):
+        """
+        Validate that the input value can be cast to our scalar_type.
+
+        Parameters
+        ----------
+        value : object
+        allow_listlike: bool, default False
+            When raising an exception, whether the message should say
+            listlike inputs are allowed.
+        unbox : bool, default True
+            Whether to unbox the result before returning.  Note: unbox=False
+            skips the setitem compatibility check.
+
+        Returns
+        -------
+        self._scalar_type or NaT
+        """
+        if isinstance(value, self._scalar_type):
+            pass
+
+        elif isinstance(value, str):
+            # NB: Careful about tzawareness
+            try:
+                value = self._scalar_from_string(value)
+            except ValueError as err:
+                msg = self._validation_error_message(value, allow_listlike)
+                raise TypeError(msg) from err
+
+        elif is_valid_na_for_dtype(value, self.dtype):
+            # GH#18295
+            value = NaT
+
+        elif isna(value):
+            # if we are dt64tz and value is dt64("NaT"), dont cast to NaT,
+            #  or else we'll fail to raise in _unbox_scalar
+            msg = self._validation_error_message(value, allow_listlike)
+            raise TypeError(msg)
+
+        elif isinstance(value, self._recognized_scalars):
+            # error: Argument 1 to "Timestamp" has incompatible type "object"; expected
+            # "integer[Any] | float | str | date | datetime | datetime64"
+            value = self._scalar_type(value)  # type: ignore[arg-type]
+
+        else:
+            msg = self._validation_error_message(value, allow_listlike)
+            raise TypeError(msg)
+
+        if not unbox:
+            # NB: In general NDArrayBackedExtensionArray will unbox here;
+            #  this option exists to prevent a performance hit in
+            #  TimedeltaIndex.get_loc
+            return value
+        return self._unbox_scalar(value)
+
+    def _validation_error_message(self, value, allow_listlike: bool = False) -> str:
+        """
+        Construct an exception message on validation error.
+
+        Some methods allow only scalar inputs, while others allow either scalar
+        or listlike.
+
+        Parameters
+        ----------
+        allow_listlike: bool, default False
+
+        Returns
+        -------
+        str
+        """
+        if hasattr(value, "dtype") and getattr(value, "ndim", 0) > 0:
+            msg_got = f"{value.dtype} array"
+        else:
+            msg_got = f"'{type(value).__name__}'"
+        if allow_listlike:
+            msg = (
+                f"value should be a '{self._scalar_type.__name__}', 'NaT', "
+                f"or array of those. Got {msg_got} instead."
+            )
+        else:
+            msg = (
+                f"value should be a '{self._scalar_type.__name__}' or 'NaT'. "
+                f"Got {msg_got} instead."
+            )
+        return msg
+
+    def _validate_listlike(self, value, allow_object: bool = False):
+        if isinstance(value, type(self)):
+            if self.dtype.kind in "mM" and not allow_object and self.unit != value.unit:  # type: ignore[attr-defined]
+                # error: "DatetimeLikeArrayMixin" has no attribute "as_unit"
+                value = value.as_unit(self.unit, round_ok=False)  # type: ignore[attr-defined]
+            return value
+
+        if isinstance(value, list) and len(value) == 0:
+            # We treat empty list as our own dtype.
+            return type(self)._from_sequence([], dtype=self.dtype)
+
+        if hasattr(value, "dtype") and value.dtype == object:
+            # `array` below won't do inference if value is an Index or Series.
+            #  so do so here.  in the Index case, inferred_type may be cached.
+            if lib.infer_dtype(value) in self._infer_matches:
+                try:
+                    value = type(self)._from_sequence(value)
+                except (ValueError, TypeError) as err:
+                    if allow_object:
+                        return value
+                    msg = self._validation_error_message(value, True)
+                    raise TypeError(msg) from err
+
+        # Do type inference if necessary up front (after unpacking
+        # NumpyExtensionArray)
+        # e.g. we passed PeriodIndex.values and got an ndarray of Periods
+        value = extract_array(value, extract_numpy=True)
+        value = pd_array(value)
+        value = extract_array(value, extract_numpy=True)
+
+        if is_all_strings(value):
+            # We got a StringArray
+            try:
+                # TODO: Could use from_sequence_of_strings if implemented
+                # Note: passing dtype is necessary for PeriodArray tests
+                value = type(self)._from_sequence(value, dtype=self.dtype)
+            except ValueError:
+                pass
+
+        if isinstance(value.dtype, CategoricalDtype):
+            # e.g. we have a Categorical holding self.dtype
+            if value.categories.dtype == self.dtype:
+                # TODO: do we need equal dtype or just comparable?
+                value = value._internal_get_values()
+                value = extract_array(value, extract_numpy=True)
+
+        if allow_object and is_object_dtype(value.dtype):
+            pass
+
+        elif not type(self)._is_recognized_dtype(value.dtype):
+            msg = self._validation_error_message(value, True)
+            raise TypeError(msg)
+
+        if self.dtype.kind in "mM" and not allow_object:
+            # error: "DatetimeLikeArrayMixin" has no attribute "as_unit"
+            value = value.as_unit(self.unit, round_ok=False)  # type: ignore[attr-defined]
+        return value
+
+    def _validate_setitem_value(self, value):
+        if is_list_like(value):
+            value = self._validate_listlike(value)
+        else:
+            return self._validate_scalar(value, allow_listlike=True)
+
+        return self._unbox(value)
+
+    @final
+    def _unbox(self, other) -> np.int64 | np.datetime64 | np.timedelta64 | np.ndarray:
+        """
+        Unbox either a scalar with _unbox_scalar or an instance of our own type.
+        """
+        if lib.is_scalar(other):
+            other = self._unbox_scalar(other)
+        else:
+            # same type as self
+            self._check_compatible_with(other)
+            other = other._ndarray
+        return other
+
+    # ------------------------------------------------------------------
+    # Additional array methods
+    #  These are not part of the EA API, but we implement them because
+    #  pandas assumes they're there.
+
+    @ravel_compat
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        from pandas import Index
+
+        result = map_array(self, mapper, na_action=na_action)
+        result = Index(result)
+
+        if isinstance(result, ABCMultiIndex):
+            return result.to_numpy()
+        else:
+            return result.array
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        """
+        Compute boolean array of whether each value is found in the
+        passed set of values.
+
+        Parameters
+        ----------
+        values : np.ndarray or ExtensionArray
+
+        Returns
+        -------
+        ndarray[bool]
+        """
+        if values.dtype.kind in "fiuc":
+            # TODO: de-duplicate with equals, validate_comparison_value
+            return np.zeros(self.shape, dtype=bool)
+
+        values = ensure_wrapped_if_datetimelike(values)
+
+        if not isinstance(values, type(self)):
+            if values.dtype == object:
+                values = lib.maybe_convert_objects(
+                    values,  # type: ignore[arg-type]
+                    convert_non_numeric=True,
+                    dtype_if_all_nat=self.dtype,
+                )
+                if values.dtype != object:
+                    return self.isin(values)
+                else:
+                    # TODO: Deprecate this case
+                    # https://github.com/pandas-dev/pandas/pull/58645/files#r1604055791
+                    return isin(self.astype(object), values)
+            return np.zeros(self.shape, dtype=bool)
+
+        if self.dtype.kind in "mM":
+            self = cast("DatetimeArray | TimedeltaArray", self)
+            # error: "DatetimeLikeArrayMixin" has no attribute "as_unit"
+            values = values.as_unit(self.unit)  # type: ignore[attr-defined]
+
+        try:
+            # error: Argument 1 to "_check_compatible_with" of "DatetimeLikeArrayMixin"
+            # has incompatible type "ExtensionArray | ndarray[Any, Any]"; expected
+            # "Period | Timestamp | Timedelta | NaTType"
+            self._check_compatible_with(values)  # type: ignore[arg-type]
+        except (TypeError, ValueError):
+            # Includes tzawareness mismatch and IncompatibleFrequencyError
+            return np.zeros(self.shape, dtype=bool)
+
+        # error: Item "ExtensionArray" of "ExtensionArray | ndarray[Any, Any]"
+        # has no attribute "asi8"
+        return isin(self.asi8, values.asi8)  # type: ignore[union-attr]
+
+    # ------------------------------------------------------------------
+    # Null Handling
+
+    def isna(self) -> npt.NDArray[np.bool_]:
+        return self._isnan
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def _isnan(self) -> npt.NDArray[np.bool_]:
+        """
+        return if each value is nan
+        """
+        return self.asi8 == iNaT
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def _hasna(self) -> bool:
+        """
+        return if I have any nans; enables various perf speedups
+        """
+        return bool(self._isnan.any())
+
+    def _maybe_mask_results(
+        self, result: np.ndarray, fill_value=iNaT, convert=None
+    ) -> np.ndarray:
+        """
+        Parameters
+        ----------
+        result : np.ndarray
+        fill_value : object, default iNaT
+        convert : str, dtype or None
+
+        Returns
+        -------
+        result : ndarray with values replace by the fill_value
+
+        mask the result if needed, convert to the provided dtype if its not
+        None
+
+        This is an internal routine.
+        """
+        if self._hasna:
+            if convert:
+                result = result.astype(convert)
+            if fill_value is None:
+                fill_value = np.nan
+            np.putmask(result, self._isnan, fill_value)
+        return result
+
+    # ------------------------------------------------------------------
+    # Frequency Properties/Methods
+
+    @property
+    def freqstr(self) -> str | None:
+        """
+        Return the frequency object as a string if it's set, otherwise None.
+
+        See Also
+        --------
+        DatetimeIndex.inferred_freq : Returns a string representing a frequency
+            generated by infer_freq.
+
+        Examples
+        --------
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(["1/1/2020 10:00:00+00:00"], freq="D")
+        >>> idx.freqstr
+        'D'
+
+        The frequency can be inferred if there are more than 2 points:
+
+        >>> idx = pd.DatetimeIndex(
+        ...     ["2018-01-01", "2018-01-03", "2018-01-05"], freq="infer"
+        ... )
+        >>> idx.freqstr
+        '2D'
+
+        For PeriodIndex:
+
+        >>> idx = pd.PeriodIndex(["2023-1", "2023-2", "2023-3"], freq="M")
+        >>> idx.freqstr
+        'M'
+        """
+        if self.freq is None:
+            return None
+        return self.freq.freqstr
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def inferred_freq(self) -> str | None:
+        """
+        Tries to return a string representing a frequency generated by infer_freq.
+
+        Returns None if it can't autodetect the frequency.
+
+        See Also
+        --------
+        DatetimeIndex.freqstr : Return the frequency object as a string if it's set,
+            otherwise None.
+
+        Examples
+        --------
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(["2018-01-01", "2018-01-03", "2018-01-05"])
+        >>> idx.inferred_freq
+        '2D'
+
+        For TimedeltaIndex:
+
+        >>> tdelta_idx = pd.to_timedelta(["0 days", "10 days", "20 days"])
+        >>> tdelta_idx
+        TimedeltaIndex(['0 days', '10 days', '20 days'],
+                       dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.inferred_freq
+        '10D'
+        """
+        if self.ndim != 1:
+            return None
+        try:
+            return frequencies.infer_freq(self)
+        except ValueError:
+            return None
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def _resolution_obj(self) -> Resolution | None:
+        freqstr = self.freqstr
+        if freqstr is None:
+            return None
+        try:
+            return Resolution.get_reso_from_freqstr(freqstr)
+        except KeyError:
+            return None
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def resolution(self) -> str:
+        """
+        Returns day, hour, minute, second, millisecond or microsecond
+        """
+        # error: Item "None" of "Optional[Any]" has no attribute "attrname"
+        return self._resolution_obj.attrname  # type: ignore[union-attr]
+
+    # monotonicity/uniqueness properties are called via frequencies.infer_freq,
+    #  see GH#23789
+
+    @property
+    def _is_monotonic_increasing(self) -> bool:
+        return algos.is_monotonic(self.asi8, timelike=True)[0]
+
+    @property
+    def _is_monotonic_decreasing(self) -> bool:
+        return algos.is_monotonic(self.asi8, timelike=True)[1]
+
+    @property
+    def _is_unique(self) -> bool:
+        return len(unique1d(self.asi8.ravel("K"))) == self.size
+
+    # ------------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _cmp_method(self, other, op):
+        if self.ndim > 1 and getattr(other, "shape", None) == self.shape:
+            # TODO: handle 2D-like listlikes
+            return op(self.ravel(), other.ravel()).reshape(self.shape)
+
+        try:
+            other = self._validate_comparison_value(other)
+        except InvalidComparison:
+            return invalid_comparison(self, other, op)
+
+        dtype = getattr(other, "dtype", None)
+        if is_object_dtype(dtype):
+            # We have to use comp_method_OBJECT_ARRAY instead of numpy
+            #  comparison otherwise it would raise when comparing to None
+            result = ops.comp_method_OBJECT_ARRAY(
+                op, np.asarray(self.astype(object)), other
+            )
+            return result
+        if other is NaT:
+            if op is operator.ne:
+                result = np.ones(self.shape, dtype=bool)
+            else:
+                result = np.zeros(self.shape, dtype=bool)
+            return result
+
+        if not isinstance(self.dtype, PeriodDtype):
+            self = cast(TimelikeOps, self)
+            if self._creso != other._creso:
+                if not isinstance(other, type(self)):
+                    # i.e. Timedelta/Timestamp, cast to ndarray and let
+                    #  compare_mismatched_resolutions handle broadcasting
+                    try:
+                        # GH#52080 see if we can losslessly cast to shared unit
+                        other = other.as_unit(self.unit, round_ok=False)
+                    except ValueError:
+                        other_arr = np.array(other.asm8)
+                        return compare_mismatched_resolutions(
+                            self._ndarray, other_arr, op
+                        )
+                else:
+                    other_arr = other._ndarray
+                    return compare_mismatched_resolutions(self._ndarray, other_arr, op)
+
+        other_vals = self._unbox(other)
+        # GH#37462 comparison on i8 values is almost 2x faster than M8/m8
+        result = op(self._ndarray.view("i8"), other_vals.view("i8"))
+
+        o_mask = isna(other)
+        mask = self._isnan | o_mask
+        if mask.any():
+            nat_result = op is operator.ne
+            np.putmask(result, mask, nat_result)
+
+        return result
+
+    # pow is invalid for all three subclasses; TimedeltaArray will override
+    #  the multiplication and division ops
+    __pow__ = _make_unpacked_invalid_op("__pow__")
+    __rpow__ = _make_unpacked_invalid_op("__rpow__")
+    __mul__ = _make_unpacked_invalid_op("__mul__")
+    __rmul__ = _make_unpacked_invalid_op("__rmul__")
+    __truediv__ = _make_unpacked_invalid_op("__truediv__")
+    __rtruediv__ = _make_unpacked_invalid_op("__rtruediv__")
+    __floordiv__ = _make_unpacked_invalid_op("__floordiv__")
+    __rfloordiv__ = _make_unpacked_invalid_op("__rfloordiv__")
+    __mod__ = _make_unpacked_invalid_op("__mod__")
+    __rmod__ = _make_unpacked_invalid_op("__rmod__")
+    __divmod__ = _make_unpacked_invalid_op("__divmod__")
+    __rdivmod__ = _make_unpacked_invalid_op("__rdivmod__")
+
+    @final
+    def _get_i8_values_and_mask(
+        self, other
+    ) -> tuple[int | npt.NDArray[np.int64], None | npt.NDArray[np.bool_]]:
+        """
+        Get the int64 values and b_mask to pass to add_overflowsafe.
+        """
+        if isinstance(other, Period):
+            i8values = other.ordinal
+            mask = None
+        elif isinstance(other, (Timestamp, Timedelta)):
+            i8values = other._value
+            mask = None
+        else:
+            # PeriodArray, DatetimeArray, TimedeltaArray
+            mask = other._isnan
+            i8values = other.asi8
+        return i8values, mask
+
+    @final
+    def _get_arithmetic_result_freq(self, other) -> BaseOffset | None:
+        """
+        Check if we can preserve self.freq in addition or subtraction.
+        """
+        # Adding or subtracting a Timedelta/Timestamp scalar is freq-preserving
+        #  whenever self.freq is a Tick
+        if isinstance(self.dtype, PeriodDtype):
+            return self.freq
+        elif not lib.is_scalar(other):
+            return None
+        elif isinstance(self.freq, Tick):
+            # In these cases
+            return self.freq
+        return None
+
+    @final
+    def _add_datetimelike_scalar(self, other) -> DatetimeArray:
+        if not lib.is_np_dtype(self.dtype, "m"):
+            raise TypeError(
+                f"cannot add {type(self).__name__} and {type(other).__name__}"
+            )
+
+        self = cast("TimedeltaArray", self)
+
+        from pandas.core.arrays import DatetimeArray
+        from pandas.core.arrays.datetimes import tz_to_dtype
+
+        assert other is not NaT
+        if isna(other):
+            # i.e. np.datetime64("NaT")
+            # In this case we specifically interpret NaT as a datetime, not
+            # the timedelta interpretation we would get by returning self + NaT
+            result = self._ndarray + NaT.to_datetime64().astype(f"M8[{self.unit}]")
+            # Preserve our resolution
+            return DatetimeArray._simple_new(result, dtype=result.dtype)
+
+        other = Timestamp(other)
+        self, other = self._ensure_matching_resos(other)
+        self = cast("TimedeltaArray", self)
+
+        other_i8, o_mask = self._get_i8_values_and_mask(other)
+        result = add_overflowsafe(self.asi8, np.asarray(other_i8, dtype="i8"))
+        res_values = result.view(f"M8[{self.unit}]")
+
+        dtype = tz_to_dtype(tz=other.tz, unit=self.unit)
+        res_values = result.view(f"M8[{self.unit}]")
+        new_freq = self._get_arithmetic_result_freq(other)
+        return DatetimeArray._simple_new(res_values, dtype=dtype, freq=new_freq)
+
+    @final
+    def _add_datetime_arraylike(self, other: DatetimeArray) -> DatetimeArray:
+        if not lib.is_np_dtype(self.dtype, "m"):
+            raise TypeError(
+                f"cannot add {type(self).__name__} and {type(other).__name__}"
+            )
+
+        # defer to DatetimeArray.__add__
+        return other + self
+
+    @final
+    def _sub_datetimelike_scalar(
+        self, other: datetime | np.datetime64
+    ) -> TimedeltaArray:
+        if self.dtype.kind != "M":
+            raise TypeError(f"cannot subtract a datelike from a {type(self).__name__}")
+
+        self = cast("DatetimeArray", self)
+        # subtract a datetime from myself, yielding a ndarray[timedelta64[ns]]
+
+        if isna(other):
+            # i.e. np.datetime64("NaT")
+            return self - NaT
+
+        ts = Timestamp(other)
+
+        self, ts = self._ensure_matching_resos(ts)
+        return self._sub_datetimelike(ts)
+
+    @final
+    def _sub_datetime_arraylike(self, other: DatetimeArray) -> TimedeltaArray:
+        if self.dtype.kind != "M":
+            raise TypeError(f"cannot subtract a datelike from a {type(self).__name__}")
+
+        if len(self) != len(other):
+            raise ValueError("cannot add indices of unequal length")
+
+        self = cast("DatetimeArray", self)
+
+        self, other = self._ensure_matching_resos(other)
+        return self._sub_datetimelike(other)
+
+    @final
+    def _sub_datetimelike(self, other: Timestamp | DatetimeArray) -> TimedeltaArray:
+        self = cast("DatetimeArray", self)
+
+        from pandas.core.arrays import TimedeltaArray
+
+        try:
+            self._assert_tzawareness_compat(other)
+        except TypeError as err:
+            new_message = str(err).replace("compare", "subtract")
+            raise type(err)(new_message) from err
+
+        other_i8, o_mask = self._get_i8_values_and_mask(other)
+        res_values = add_overflowsafe(self.asi8, np.asarray(-other_i8, dtype="i8"))
+        res_m8 = res_values.view(f"timedelta64[{self.unit}]")
+
+        new_freq = self._get_arithmetic_result_freq(other)
+        new_freq = cast("Tick | None", new_freq)
+        return TimedeltaArray._simple_new(res_m8, dtype=res_m8.dtype, freq=new_freq)
+
+    @final
+    def _add_period(self, other: Period) -> PeriodArray:
+        if not lib.is_np_dtype(self.dtype, "m"):
+            raise TypeError(f"cannot add Period to a {type(self).__name__}")
+
+        # We will wrap in a PeriodArray and defer to the reversed operation
+        from pandas.core.arrays.period import PeriodArray
+
+        i8vals = np.broadcast_to(other.ordinal, self.shape)
+        dtype = PeriodDtype(other.freq)
+        parr = PeriodArray(i8vals, dtype=dtype)
+        return parr + self
+
+    def _add_offset(self, offset):
+        raise AbstractMethodError(self)
+
+    def _add_timedeltalike_scalar(self, other):
+        """
+        Add a delta of a timedeltalike
+
+        Returns
+        -------
+        Same type as self
+        """
+        if isna(other):
+            # i.e np.timedelta64("NaT")
+            new_values = np.empty(self.shape, dtype="i8").view(self._ndarray.dtype)
+            new_values.fill(iNaT)
+            return type(self)._simple_new(new_values, dtype=self.dtype)
+
+        # PeriodArray overrides, so we only get here with DTA/TDA
+        self = cast("DatetimeArray | TimedeltaArray", self)
+        other = Timedelta(other)
+        self, other = self._ensure_matching_resos(other)
+        return self._add_timedeltalike(other)
+
+    def _add_timedelta_arraylike(self, other: TimedeltaArray) -> Self:
+        """
+        Add a delta of a TimedeltaIndex
+
+        Returns
+        -------
+        Same type as self
+        """
+        # overridden by PeriodArray
+
+        if len(self) != len(other):
+            raise ValueError("cannot add indices of unequal length")
+
+        self, other = cast(
+            "DatetimeArray | TimedeltaArray", self
+        )._ensure_matching_resos(other)
+        return self._add_timedeltalike(other)
+
+    @final
+    def _add_timedeltalike(self, other: Timedelta | TimedeltaArray) -> Self:
+        other_i8, o_mask = self._get_i8_values_and_mask(other)
+        new_values = add_overflowsafe(self.asi8, np.asarray(other_i8, dtype="i8"))
+        res_values = new_values.view(self._ndarray.dtype)
+
+        new_freq = self._get_arithmetic_result_freq(other)
+
+        # error: Unexpected keyword argument "freq" for "_simple_new" of "NDArrayBacked"
+        return type(self)._simple_new(
+            res_values,
+            dtype=self.dtype,
+            freq=new_freq,  # type: ignore[call-arg]
+        )
+
+    @final
+    def _add_nat(self) -> Self:
+        """
+        Add pd.NaT to self
+        """
+        if isinstance(self.dtype, PeriodDtype):
+            raise TypeError(
+                f"Cannot add {type(self).__name__} and {type(NaT).__name__}"
+            )
+
+        # GH#19124 pd.NaT is treated like a timedelta for both timedelta
+        # and datetime dtypes
+        result = np.empty(self.shape, dtype=np.int64)
+        result.fill(iNaT)
+        result = result.view(self._ndarray.dtype)  # preserve reso
+        # error: Unexpected keyword argument "freq" for "_simple_new" of "NDArrayBacked"
+        return type(self)._simple_new(
+            result,
+            dtype=self.dtype,
+            freq=None,  # type: ignore[call-arg]
+        )
+
+    @final
+    def _sub_nat(self) -> np.ndarray:
+        """
+        Subtract pd.NaT from self
+        """
+        # GH#19124 Timedelta - datetime is not in general well-defined.
+        # We make an exception for pd.NaT, which in this case quacks
+        # like a timedelta.
+        # For datetime64 dtypes by convention we treat NaT as a datetime, so
+        # this subtraction returns a timedelta64 dtype.
+        # For period dtype, timedelta64 is a close-enough return dtype.
+        result = np.empty(self.shape, dtype=np.int64)
+        result.fill(iNaT)
+        if self.dtype.kind in "mM":
+            # We can retain unit in dtype
+            self = cast("DatetimeArray| TimedeltaArray", self)
+            return result.view(f"timedelta64[{self.unit}]")
+        else:
+            return result.view("timedelta64[ns]")
+
+    @final
+    def _sub_periodlike(self, other: Period | PeriodArray) -> npt.NDArray[np.object_]:
+        # If the operation is well-defined, we return an object-dtype ndarray
+        # of DateOffsets.  Null entries are filled with pd.NaT
+        if not isinstance(self.dtype, PeriodDtype):
+            raise TypeError(
+                f"cannot subtract {type(other).__name__} from {type(self).__name__}"
+            )
+
+        self = cast("PeriodArray", self)
+        self._check_compatible_with(other)
+
+        other_i8, o_mask = self._get_i8_values_and_mask(other)
+        new_i8_data = add_overflowsafe(self.asi8, np.asarray(-other_i8, dtype="i8"))
+        new_data = np.array([self.freq.base * x for x in new_i8_data])
+
+        if o_mask is None:
+            # i.e. Period scalar
+            mask = self._isnan
+        else:
+            # i.e. PeriodArray
+            mask = self._isnan | o_mask
+        new_data[mask] = NaT
+        return new_data
+
+    @final
+    def _addsub_object_array(self, other: npt.NDArray[np.object_], op) -> np.ndarray:
+        """
+        Add or subtract array-like of DateOffset objects
+
+        Parameters
+        ----------
+        other : np.ndarray[object]
+        op : {operator.add, operator.sub}
+
+        Returns
+        -------
+        np.ndarray[object]
+            Except in fastpath case with length 1 where we operate on the
+            contained scalar.
+        """
+        assert op in [operator.add, operator.sub]
+        if len(other) == 1 and self.ndim == 1:
+            # Note: without this special case, we could annotate return type
+            #  as ndarray[object]
+            # If both 1D then broadcasting is unambiguous
+            return op(self, other[0])
+
+        if get_option("performance_warnings"):
+            warnings.warn(
+                "Adding/subtracting object-dtype array to "
+                f"{type(self).__name__} not vectorized.",
+                PerformanceWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        # Caller is responsible for broadcasting if necessary
+        assert self.shape == other.shape, (self.shape, other.shape)
+
+        res_values = op(self.astype("O"), np.asarray(other))
+        return res_values
+
+    def _accumulate(self, name: str, *, skipna: bool = True, **kwargs) -> Self:
+        if name not in {"cummin", "cummax"}:
+            raise TypeError(f"Accumulation {name} not supported for {type(self)}")
+
+        op = getattr(datetimelike_accumulations, name)
+        result = op(self.copy(), skipna=skipna, **kwargs)
+
+        return type(self)._simple_new(result, dtype=self.dtype)
+
+    @unpack_zerodim_and_defer("__add__")
+    def __add__(self, other):
+        other_dtype = getattr(other, "dtype", None)
+        other = ensure_wrapped_if_datetimelike(other)
+
+        # scalar others
+        if other is NaT:
+            result: np.ndarray | DatetimeLikeArrayMixin = self._add_nat()
+        elif isinstance(other, (Tick, timedelta, np.timedelta64)):
+            result = self._add_timedeltalike_scalar(other)
+        elif isinstance(other, BaseOffset):
+            # specifically _not_ a Tick
+            result = self._add_offset(other)
+        elif isinstance(other, (datetime, np.datetime64)):
+            result = self._add_datetimelike_scalar(other)
+        elif isinstance(other, Period) and lib.is_np_dtype(self.dtype, "m"):
+            result = self._add_period(other)
+        elif lib.is_integer(other):
+            # This check must come after the check for np.timedelta64
+            # as is_integer returns True for these
+            if not isinstance(self.dtype, PeriodDtype):
+                raise integer_op_not_supported(self)
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.dtype._n, operator.add)
+
+        # array-like others
+        elif lib.is_np_dtype(other_dtype, "m"):
+            # TimedeltaIndex, ndarray[timedelta64]
+            result = self._add_timedelta_arraylike(other)
+        elif is_object_dtype(other_dtype):
+            # e.g. Array/Index of DateOffset objects
+            result = self._addsub_object_array(other, operator.add)
+        elif lib.is_np_dtype(other_dtype, "M") or isinstance(
+            other_dtype, DatetimeTZDtype
+        ):
+            # DatetimeIndex, ndarray[datetime64]
+            return self._add_datetime_arraylike(other)
+        elif is_integer_dtype(other_dtype):
+            if not isinstance(self.dtype, PeriodDtype):
+                raise integer_op_not_supported(self)
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.dtype._n, operator.add)
+        else:
+            # Includes Categorical, other ExtensionArrays
+            # For PeriodDtype, if self is a TimedeltaArray and other is a
+            #  PeriodArray with  a timedelta-like (i.e. Tick) freq, this
+            #  operation is valid.  Defer to the PeriodArray implementation.
+            #  In remaining cases, this will end up raising TypeError.
+            return NotImplemented
+
+        if isinstance(result, np.ndarray) and lib.is_np_dtype(result.dtype, "m"):
+            from pandas.core.arrays import TimedeltaArray
+
+            return TimedeltaArray._from_sequence(result, dtype=result.dtype)
+        return result
+
+    def __radd__(self, other):
+        # alias for __add__
+        return self.__add__(other)
+
+    @unpack_zerodim_and_defer("__sub__")
+    def __sub__(self, other):
+        other_dtype = getattr(other, "dtype", None)
+        other = ensure_wrapped_if_datetimelike(other)
+
+        # scalar others
+        if other is NaT:
+            result: np.ndarray | DatetimeLikeArrayMixin = self._sub_nat()
+        elif isinstance(other, (Tick, timedelta, np.timedelta64)):
+            result = self._add_timedeltalike_scalar(-other)
+        elif isinstance(other, BaseOffset):
+            # specifically _not_ a Tick
+            result = self._add_offset(-other)
+        elif isinstance(other, (datetime, np.datetime64)):
+            result = self._sub_datetimelike_scalar(other)
+        elif lib.is_integer(other):
+            # This check must come after the check for np.timedelta64
+            # as is_integer returns True for these
+            if not isinstance(self.dtype, PeriodDtype):
+                raise integer_op_not_supported(self)
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.dtype._n, operator.sub)
+
+        elif isinstance(other, Period):
+            result = self._sub_periodlike(other)
+
+        # array-like others
+        elif lib.is_np_dtype(other_dtype, "m"):
+            # TimedeltaIndex, ndarray[timedelta64]
+            result = self._add_timedelta_arraylike(-other)
+        elif is_object_dtype(other_dtype):
+            # e.g. Array/Index of DateOffset objects
+            result = self._addsub_object_array(other, operator.sub)
+        elif lib.is_np_dtype(other_dtype, "M") or isinstance(
+            other_dtype, DatetimeTZDtype
+        ):
+            # DatetimeIndex, ndarray[datetime64]
+            result = self._sub_datetime_arraylike(other)
+        elif isinstance(other_dtype, PeriodDtype):
+            # PeriodIndex
+            result = self._sub_periodlike(other)
+        elif is_integer_dtype(other_dtype):
+            if not isinstance(self.dtype, PeriodDtype):
+                raise integer_op_not_supported(self)
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.dtype._n, operator.sub)
+        else:
+            # Includes ExtensionArrays, float_dtype
+            return NotImplemented
+
+        if isinstance(result, np.ndarray) and lib.is_np_dtype(result.dtype, "m"):
+            from pandas.core.arrays import TimedeltaArray
+
+            return TimedeltaArray._from_sequence(result, dtype=result.dtype)
+        return result
+
+    def __rsub__(self, other):
+        other_dtype = getattr(other, "dtype", None)
+        other_is_dt64 = lib.is_np_dtype(other_dtype, "M") or isinstance(
+            other_dtype, DatetimeTZDtype
+        )
+
+        if other_is_dt64 and lib.is_np_dtype(self.dtype, "m"):
+            # ndarray[datetime64] cannot be subtracted from self, so
+            # we need to wrap in DatetimeArray/Index and flip the operation
+            if lib.is_scalar(other):
+                # i.e. np.datetime64 object
+                return Timestamp(other) - self
+            if not isinstance(other, DatetimeLikeArrayMixin):
+                # Avoid down-casting DatetimeIndex
+                from pandas.core.arrays import DatetimeArray
+
+                other = DatetimeArray._from_sequence(other, dtype=other.dtype)
+            return other - self
+        elif self.dtype.kind == "M" and hasattr(other, "dtype") and not other_is_dt64:
+            # GH#19959 datetime - datetime is well-defined as timedelta,
+            # but any other type - datetime is not well-defined.
+            raise TypeError(
+                f"cannot subtract {type(self).__name__} from {type(other).__name__}"
+            )
+        elif isinstance(self.dtype, PeriodDtype) and lib.is_np_dtype(other_dtype, "m"):
+            # TODO: Can we simplify/generalize these cases at all?
+            raise TypeError(f"cannot subtract {type(self).__name__} from {other.dtype}")
+        elif lib.is_np_dtype(self.dtype, "m"):
+            self = cast("TimedeltaArray", self)
+            return (-self) + other
+
+        # We get here with e.g. datetime objects
+        return -(self - other)
+
+    def __iadd__(self, other) -> Self:
+        result = self + other
+        self[:] = result[:]
+
+        if not isinstance(self.dtype, PeriodDtype):
+            # restore freq, which is invalidated by setitem
+            self._freq = result.freq
+        return self
+
+    def __isub__(self, other) -> Self:
+        result = self - other
+        self[:] = result[:]
+
+        if not isinstance(self.dtype, PeriodDtype):
+            # restore freq, which is invalidated by setitem
+            self._freq = result.freq
+        return self
+
+    # --------------------------------------------------------------
+    # Reductions
+
+    @_period_dispatch
+    def _quantile(
+        self,
+        qs: npt.NDArray[np.float64],
+        interpolation: str,
+    ) -> Self:
+        return super()._quantile(qs=qs, interpolation=interpolation)
+
+    @_period_dispatch
+    def min(self, *, axis: AxisInt | None = None, skipna: bool = True, **kwargs):
+        """
+        Return the minimum value of the Array or minimum along
+        an axis.
+
+        See Also
+        --------
+        numpy.ndarray.min
+        Index.min : Return the minimum value in an Index.
+        Series.min : Return the minimum value in a Series.
+        """
+        nv.validate_min((), kwargs)
+        nv.validate_minmax_axis(axis, self.ndim)
+
+        result = nanops.nanmin(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    @_period_dispatch
+    def max(self, *, axis: AxisInt | None = None, skipna: bool = True, **kwargs):
+        """
+        Return the maximum value of the Array or maximum along
+        an axis.
+
+        See Also
+        --------
+        numpy.ndarray.max
+        Index.max : Return the maximum value in an Index.
+        Series.max : Return the maximum value in a Series.
+        """
+        nv.validate_max((), kwargs)
+        nv.validate_minmax_axis(axis, self.ndim)
+
+        result = nanops.nanmax(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def mean(self, *, skipna: bool = True, axis: AxisInt | None = 0):
+        """
+        Return the mean value of the Array.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Whether to ignore any NaT elements.
+        axis : int, optional, default 0
+            Axis for the function to be applied on.
+
+        Returns
+        -------
+        scalar
+            Timestamp or Timedelta.
+
+        See Also
+        --------
+        numpy.ndarray.mean : Returns the average of array elements along a given axis.
+        Series.mean : Return the mean value in a Series.
+
+        Notes
+        -----
+        mean is only defined for Datetime and Timedelta dtypes, not for Period.
+
+        Examples
+        --------
+        For :class:`pandas.DatetimeIndex`:
+
+        >>> idx = pd.date_range("2001-01-01 00:00", periods=3)
+        >>> idx
+        DatetimeIndex(['2001-01-01', '2001-01-02', '2001-01-03'],
+                      dtype='datetime64[ns]', freq='D')
+        >>> idx.mean()
+        Timestamp('2001-01-02 00:00:00')
+
+        For :class:`pandas.TimedeltaIndex`:
+
+        >>> tdelta_idx = pd.to_timedelta([1, 2, 3], unit="D")
+        >>> tdelta_idx
+        TimedeltaIndex(['1 days', '2 days', '3 days'],
+                        dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.mean()
+        Timedelta('2 days 00:00:00')
+        """
+        if isinstance(self.dtype, PeriodDtype):
+            # See discussion in GH#24757
+            raise TypeError(
+                f"mean is not implemented for {type(self).__name__} since the "
+                "meaning is ambiguous.  An alternative is "
+                "obj.to_timestamp(how='start').mean()"
+            )
+
+        result = nanops.nanmean(
+            self._ndarray, axis=axis, skipna=skipna, mask=self.isna()
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    @_period_dispatch
+    def median(self, *, axis: AxisInt | None = None, skipna: bool = True, **kwargs):
+        nv.validate_median((), kwargs)
+
+        if axis is not None and abs(axis) >= self.ndim:
+            raise ValueError("abs(axis) must be less than ndim")
+
+        result = nanops.nanmedian(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def _mode(self, dropna: bool = True):
+        mask = None
+        if dropna:
+            mask = self.isna()
+
+        i8modes = algorithms.mode(self.view("i8"), mask=mask)
+        npmodes = i8modes.view(self._ndarray.dtype)
+        npmodes = cast(np.ndarray, npmodes)
+        return self._from_backing_data(npmodes)
+
+    # ------------------------------------------------------------------
+    # GroupBy Methods
+
+    def _groupby_op(
+        self,
+        *,
+        how: str,
+        has_dropped_na: bool,
+        min_count: int,
+        ngroups: int,
+        ids: npt.NDArray[np.intp],
+        **kwargs,
+    ):
+        dtype = self.dtype
+        if dtype.kind == "M":
+            # Adding/multiplying datetimes is not valid
+            if how in ["sum", "prod", "cumsum", "cumprod", "var", "skew", "kurt"]:
+                raise TypeError(f"datetime64 type does not support operation '{how}'")
+            if how in ["any", "all"]:
+                # GH#34479
+                raise TypeError(
+                    f"'{how}' with datetime64 dtypes is no longer supported. "
+                    f"Use (obj != pd.Timestamp(0)).{how}() instead."
+                )
+
+        elif isinstance(dtype, PeriodDtype):
+            # Adding/multiplying Periods is not valid
+            if how in ["sum", "prod", "cumsum", "cumprod", "var", "skew", "kurt"]:
+                raise TypeError(f"Period type does not support {how} operations")
+            if how in ["any", "all"]:
+                # GH#34479
+                raise TypeError(
+                    f"'{how}' with PeriodDtype is no longer supported. "
+                    f"Use (obj != pd.Period(0, freq)).{how}() instead."
+                )
+        else:
+            # timedeltas we can add but not multiply
+            if how in ["prod", "cumprod", "skew", "kurt", "var"]:
+                raise TypeError(f"timedelta64 type does not support {how} operations")
+
+        # All of the functions implemented here are ordinal, so we can
+        #  operate on the tz-naive equivalents
+        npvalues = self._ndarray.view("M8[ns]")
+
+        from pandas.core.groupby.ops import WrappedCythonOp
+
+        kind = WrappedCythonOp.get_kind_from_how(how)
+        op = WrappedCythonOp(how=how, kind=kind, has_dropped_na=has_dropped_na)
+
+        res_values = op._cython_op_ndim_compat(
+            npvalues,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=ids,
+            mask=None,
+            **kwargs,
+        )
+
+        if op.how in op.cast_blocklist:
+            # i.e. how in ["rank"], since other cast_blocklist methods don't go
+            #  through cython_operation
+            return res_values
+
+        # We did a view to M8[ns] above, now we go the other direction
+        assert res_values.dtype == "M8[ns]"
+        if how in ["std", "sem"]:
+            from pandas.core.arrays import TimedeltaArray
+
+            if isinstance(self.dtype, PeriodDtype):
+                raise TypeError("'std' and 'sem' are not valid for PeriodDtype")
+            self = cast("DatetimeArray | TimedeltaArray", self)
+            new_dtype = f"m8[{self.unit}]"
+            res_values = res_values.view(new_dtype)
+            return TimedeltaArray._simple_new(res_values, dtype=res_values.dtype)
+
+        res_values = res_values.view(self._ndarray.dtype)
+        return self._from_backing_data(res_values)
+
+
+class DatelikeOps(DatetimeLikeArrayMixin):
+    """
+    Common ops for DatetimeIndex/PeriodIndex, but not TimedeltaIndex.
+    """
+
+    @Substitution(
+        URL="https://docs.python.org/3/library/datetime.html"
+        "#strftime-and-strptime-behavior"
+    )
+    def strftime(self, date_format: str) -> npt.NDArray[np.object_]:
+        """
+        Convert to Index using specified date_format.
+
+        Return an Index of formatted strings specified by date_format, which
+        supports the same string format as the python standard library. Details
+        of the string format can be found in `python string format
+        doc <%(URL)s>`__.
+
+        Formats supported by the C `strftime` API but not by the python string format
+        doc (such as `"%%R"`, `"%%r"`) are not officially supported and should be
+        preferably replaced with their supported equivalents (such as `"%%H:%%M"`,
+        `"%%I:%%M:%%S %%p"`).
+
+        Note that `PeriodIndex` support additional directives, detailed in
+        `Period.strftime`.
+
+        Parameters
+        ----------
+        date_format : str
+            Date format string (e.g. "%%Y-%%m-%%d").
+
+        Returns
+        -------
+        ndarray[object]
+            NumPy ndarray of formatted strings.
+
+        See Also
+        --------
+        to_datetime : Convert the given argument to datetime.
+        DatetimeIndex.normalize : Return DatetimeIndex with times to midnight.
+        DatetimeIndex.round : Round the DatetimeIndex to the specified freq.
+        DatetimeIndex.floor : Floor the DatetimeIndex to the specified freq.
+        Timestamp.strftime : Format a single Timestamp.
+        Period.strftime : Format a single Period.
+
+        Examples
+        --------
+        >>> rng = pd.date_range(pd.Timestamp("2018-03-10 09:00"), periods=3, freq="s")
+        >>> rng.strftime("%%B %%d, %%Y, %%r")
+        Index(['March 10, 2018, 09:00:00 AM', 'March 10, 2018, 09:00:01 AM',
+               'March 10, 2018, 09:00:02 AM'],
+              dtype='object')
+        """
+        result = self._format_native_types(date_format=date_format, na_rep=np.nan)
+        if using_string_dtype():
+            from pandas import StringDtype
+
+            return pd_array(result, dtype=StringDtype(na_value=np.nan))  # type: ignore[return-value]
+        return result.astype(object, copy=False)
+
+
+_round_doc = """
+    Perform {op} operation on the data to the specified `freq`.
+
+    Parameters
+    ----------
+    freq : str or Offset
+        The frequency level to {op} the index to. Must be a fixed
+        frequency like 's' (second) not 'ME' (month end). See
+        :ref:`frequency aliases <timeseries.offset_aliases>` for
+        a list of possible `freq` values.
+    ambiguous : 'infer', bool-ndarray, 'NaT', default 'raise'
+        Only relevant for DatetimeIndex:
+
+        - 'infer' will attempt to infer fall dst-transition hours based on
+          order
+        - bool-ndarray where True signifies a DST time, False designates
+          a non-DST time (note that this flag is only applicable for
+          ambiguous times)
+        - 'NaT' will return NaT where there are ambiguous times
+        - 'raise' will raise a ValueError if there are ambiguous
+          times.
+
+    nonexistent : 'shift_forward', 'shift_backward', 'NaT', timedelta, default 'raise'
+        A nonexistent time does not exist in a particular timezone
+        where clocks moved forward due to DST.
+
+        - 'shift_forward' will shift the nonexistent time forward to the
+          closest existing time
+        - 'shift_backward' will shift the nonexistent time backward to the
+          closest existing time
+        - 'NaT' will return NaT where there are nonexistent times
+        - timedelta objects will shift nonexistent times by the timedelta
+        - 'raise' will raise a ValueError if there are
+          nonexistent times.
+
+    Returns
+    -------
+    DatetimeIndex, TimedeltaIndex, or Series
+        Index of the same type for a DatetimeIndex or TimedeltaIndex,
+        or a Series with the same index for a Series.
+
+    Raises
+    ------
+    ValueError if the `freq` cannot be converted.
+
+    See Also
+    --------
+    DatetimeIndex.floor : Perform floor operation on the data to the specified `freq`.
+    DatetimeIndex.snap : Snap time stamps to nearest occurring frequency.
+
+    Notes
+    -----
+    If the timestamps have a timezone, {op}ing will take place relative to the
+    local ("wall") time and re-localized to the same timezone. When {op}ing
+    near daylight savings time, use ``nonexistent`` and ``ambiguous`` to
+    control the re-localization behavior.
+
+    Examples
+    --------
+    **DatetimeIndex**
+
+    >>> rng = pd.date_range('1/1/2018 11:59:00', periods=3, freq='min')
+    >>> rng
+    DatetimeIndex(['2018-01-01 11:59:00', '2018-01-01 12:00:00',
+                   '2018-01-01 12:01:00'],
+                  dtype='datetime64[ns]', freq='min')
+    """
+
+_round_example = """>>> rng.round('h')
+    DatetimeIndex(['2018-01-01 12:00:00', '2018-01-01 12:00:00',
+                   '2018-01-01 12:00:00'],
+                  dtype='datetime64[ns]', freq=None)
+
+    **Series**
+
+    >>> pd.Series(rng).dt.round("h")
+    0   2018-01-01 12:00:00
+    1   2018-01-01 12:00:00
+    2   2018-01-01 12:00:00
+    dtype: datetime64[ns]
+
+    When rounding near a daylight savings time transition, use ``ambiguous`` or
+    ``nonexistent`` to control how the timestamp should be re-localized.
+
+    >>> rng_tz = pd.DatetimeIndex(["2021-10-31 03:30:00"], tz="Europe/Amsterdam")
+
+    >>> rng_tz.floor("2h", ambiguous=False)
+    DatetimeIndex(['2021-10-31 02:00:00+01:00'],
+                  dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+
+    >>> rng_tz.floor("2h", ambiguous=True)
+    DatetimeIndex(['2021-10-31 02:00:00+02:00'],
+                  dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+    """
+
+_floor_example = """>>> rng.floor('h')
+    DatetimeIndex(['2018-01-01 11:00:00', '2018-01-01 12:00:00',
+                   '2018-01-01 12:00:00'],
+                  dtype='datetime64[ns]', freq=None)
+
+    **Series**
+
+    >>> pd.Series(rng).dt.floor("h")
+    0   2018-01-01 11:00:00
+    1   2018-01-01 12:00:00
+    2   2018-01-01 12:00:00
+    dtype: datetime64[ns]
+
+    When rounding near a daylight savings time transition, use ``ambiguous`` or
+    ``nonexistent`` to control how the timestamp should be re-localized.
+
+    >>> rng_tz = pd.DatetimeIndex(["2021-10-31 03:30:00"], tz="Europe/Amsterdam")
+
+    >>> rng_tz.floor("2h", ambiguous=False)
+    DatetimeIndex(['2021-10-31 02:00:00+01:00'],
+                 dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+
+    >>> rng_tz.floor("2h", ambiguous=True)
+    DatetimeIndex(['2021-10-31 02:00:00+02:00'],
+                  dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+    """
+
+_ceil_example = """>>> rng.ceil('h')
+    DatetimeIndex(['2018-01-01 12:00:00', '2018-01-01 12:00:00',
+                   '2018-01-01 13:00:00'],
+                  dtype='datetime64[ns]', freq=None)
+
+    **Series**
+
+    >>> pd.Series(rng).dt.ceil("h")
+    0   2018-01-01 12:00:00
+    1   2018-01-01 12:00:00
+    2   2018-01-01 13:00:00
+    dtype: datetime64[ns]
+
+    When rounding near a daylight savings time transition, use ``ambiguous`` or
+    ``nonexistent`` to control how the timestamp should be re-localized.
+
+    >>> rng_tz = pd.DatetimeIndex(["2021-10-31 01:30:00"], tz="Europe/Amsterdam")
+
+    >>> rng_tz.ceil("h", ambiguous=False)
+    DatetimeIndex(['2021-10-31 02:00:00+01:00'],
+                  dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+
+    >>> rng_tz.ceil("h", ambiguous=True)
+    DatetimeIndex(['2021-10-31 02:00:00+02:00'],
+                  dtype='datetime64[s, Europe/Amsterdam]', freq=None)
+    """
+
+
+class TimelikeOps(DatetimeLikeArrayMixin):
+    """
+    Common ops for TimedeltaIndex/DatetimeIndex, but not PeriodIndex.
+    """
+
+    @classmethod
+    def _validate_dtype(cls, values, dtype):
+        raise AbstractMethodError(cls)
+
+    @property
+    def freq(self):
+        """
+        Return the frequency object if it is set, otherwise None.
+
+        To learn more about the frequency strings, please see
+        :ref:`this link<timeseries.offset_aliases>`.
+
+        See Also
+        --------
+        DatetimeIndex.freq : Return the frequency object if it is set, otherwise None.
+        PeriodIndex.freq : Return the frequency object if it is set, otherwise None.
+
+        Examples
+        --------
+        >>> datetimeindex = pd.date_range(
+        ...     "2022-02-22 02:22:22", periods=10, tz="America/Chicago", freq="h"
+        ... )
+        >>> datetimeindex
+        DatetimeIndex(['2022-02-22 02:22:22-06:00', '2022-02-22 03:22:22-06:00',
+                       '2022-02-22 04:22:22-06:00', '2022-02-22 05:22:22-06:00',
+                       '2022-02-22 06:22:22-06:00', '2022-02-22 07:22:22-06:00',
+                       '2022-02-22 08:22:22-06:00', '2022-02-22 09:22:22-06:00',
+                       '2022-02-22 10:22:22-06:00', '2022-02-22 11:22:22-06:00'],
+                      dtype='datetime64[ns, America/Chicago]', freq='h')
+        >>> datetimeindex.freq
+        <Hour>
+        """
+        return self._freq
+
+    @freq.setter
+    def freq(self, value) -> None:
+        if value is not None:
+            value = to_offset(value)
+            self._validate_frequency(self, value)
+            if self.dtype.kind == "m" and not isinstance(value, Tick):
+                raise TypeError("TimedeltaArray/Index freq must be a Tick")
+
+            if self.ndim > 1:
+                raise ValueError("Cannot set freq with ndim > 1")
+
+        self._freq = value
+
+    @final
+    def _maybe_pin_freq(self, freq, validate_kwds: dict) -> None:
+        """
+        Constructor helper to pin the appropriate `freq` attribute.  Assumes
+        that self._freq is currently set to any freq inferred in
+        _from_sequence_not_strict.
+        """
+        if freq is None:
+            # user explicitly passed None -> override any inferred_freq
+            self._freq = None
+        elif freq == "infer":
+            # if self._freq is *not* None then we already inferred a freq
+            #  and there is nothing left to do
+            if self._freq is None:
+                # Set _freq directly to bypass duplicative _validate_frequency
+                # check.
+                self._freq = to_offset(self.inferred_freq)
+        elif freq is lib.no_default:
+            # user did not specify anything, keep inferred freq if the original
+            #  data had one, otherwise do nothing
+            pass
+        elif self._freq is None:
+            # We cannot inherit a freq from the data, so we need to validate
+            #  the user-passed freq
+            freq = to_offset(freq)
+            type(self)._validate_frequency(self, freq, **validate_kwds)
+            self._freq = freq
+        else:
+            # Otherwise we just need to check that the user-passed freq
+            #  doesn't conflict with the one we already have.
+            freq = to_offset(freq)
+            _validate_inferred_freq(freq, self._freq)
+
+    @final
+    @classmethod
+    def _validate_frequency(cls, index, freq: BaseOffset, **kwargs) -> None:
+        """
+        Validate that a frequency is compatible with the values of a given
+        Datetime Array/Index or Timedelta Array/Index
+
+        Parameters
+        ----------
+        index : DatetimeIndex or TimedeltaIndex
+            The index on which to determine if the given frequency is valid
+        freq : DateOffset
+            The frequency to validate
+        """
+        inferred = index.inferred_freq
+        if index.size == 0 or inferred == freq.freqstr:
+            return None
+
+        try:
+            on_freq = cls._generate_range(
+                start=index[0],
+                end=None,
+                periods=len(index),
+                freq=freq,
+                unit=index.unit,
+                **kwargs,
+            )
+            if not np.array_equal(index.asi8, on_freq.asi8):
+                raise ValueError
+        except ValueError as err:
+            if "non-fixed" in str(err):
+                # non-fixed frequencies are not meaningful for timedelta64;
+                #  we retain that error message
+                raise err
+            # GH#11587 the main way this is reached is if the `np.array_equal`
+            #  check above is False.  This can also be reached if index[0]
+            #  is `NaT`, in which case the call to `cls._generate_range` will
+            #  raise a ValueError, which we re-raise with a more targeted
+            #  message.
+            raise ValueError(
+                f"Inferred frequency {inferred} from passed values "
+                f"does not conform to passed frequency {freq.freqstr}"
+            ) from err
+
+    @classmethod
+    def _generate_range(
+        cls, start, end, periods: int | None, freq, *args, **kwargs
+    ) -> Self:
+        raise AbstractMethodError(cls)
+
+    # --------------------------------------------------------------
+
+    @cache_readonly
+    def _creso(self) -> int:
+        return get_unit_from_dtype(self._ndarray.dtype)
+
+    @cache_readonly
+    def unit(self) -> str:
+        """
+        The precision unit of the datetime data.
+
+        Returns the precision unit for the dtype.
+        It means the smallest time frame that can be stored within this dtype.
+
+        Returns
+        -------
+        str
+            Unit string representation (e.g. "ns").
+
+        See Also
+        --------
+        TimelikeOps.as_unit : Converts to a specific unit.
+
+        Examples
+        --------
+        >>> idx = pd.DatetimeIndex(["2020-01-02 01:02:03.004005006"])
+        >>> idx.unit
+        'ns'
+        >>> idx.as_unit("s").unit
+        's'
+        """
+        # error: Argument 1 to "dtype_to_unit" has incompatible type
+        # "ExtensionDtype"; expected "Union[DatetimeTZDtype, dtype[Any]]"
+        return dtype_to_unit(self.dtype)  # type: ignore[arg-type]
+
+    def as_unit(self, unit: str, round_ok: bool = True) -> Self:
+        """
+        Convert to a dtype with the given unit resolution.
+
+        The limits of timestamp representation depend on the chosen resolution.
+        Different resolutions can be converted to each other through as_unit.
+
+        Parameters
+        ----------
+        unit : {'s', 'ms', 'us', 'ns'}
+        round_ok : bool, default True
+            If False and the conversion requires rounding, raise ValueError.
+
+        Returns
+        -------
+        same type as self
+            Converted to the specified unit.
+
+        See Also
+        --------
+        Timestamp.as_unit : Convert to the given unit.
+
+        Examples
+        --------
+        For :class:`pandas.DatetimeIndex`:
+
+        >>> idx = pd.DatetimeIndex(["2020-01-02 01:02:03.004005006"])
+        >>> idx
+        DatetimeIndex(['2020-01-02 01:02:03.004005006'],
+                      dtype='datetime64[ns]', freq=None)
+        >>> idx.as_unit("s")
+        DatetimeIndex(['2020-01-02 01:02:03'], dtype='datetime64[s]', freq=None)
+
+        For :class:`pandas.TimedeltaIndex`:
+
+        >>> tdelta_idx = pd.to_timedelta(["1 day 3 min 2 us 42 ns"])
+        >>> tdelta_idx
+        TimedeltaIndex(['1 days 00:03:00.000002042'],
+                        dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.as_unit("s")
+        TimedeltaIndex(['1 days 00:03:00'], dtype='timedelta64[s]', freq=None)
+        """
+        if unit not in ["s", "ms", "us", "ns"]:
+            raise ValueError("Supported units are 's', 'ms', 'us', 'ns'")
+
+        dtype = np.dtype(f"{self.dtype.kind}8[{unit}]")
+        new_values = astype_overflowsafe(self._ndarray, dtype, round_ok=round_ok)
+
+        if isinstance(self.dtype, np.dtype):
+            new_dtype = new_values.dtype
+        else:
+            tz = cast("DatetimeArray", self).tz
+            new_dtype = DatetimeTZDtype(tz=tz, unit=unit)
+
+        # error: Unexpected keyword argument "freq" for "_simple_new" of
+        # "NDArrayBacked"  [call-arg]
+        return type(self)._simple_new(
+            new_values,
+            dtype=new_dtype,
+            freq=self.freq,  # type: ignore[call-arg]
+        )
+
+    # TODO: annotate other as DatetimeArray | TimedeltaArray | Timestamp | Timedelta
+    #  with the return type matching input type.  TypeVar?
+    def _ensure_matching_resos(self, other):
+        if self._creso != other._creso:
+            # Just as with Timestamp/Timedelta, we cast to the higher resolution
+            if self._creso < other._creso:
+                self = self.as_unit(other.unit)
+            else:
+                other = other.as_unit(self.unit)
+        return self, other
+
+    # --------------------------------------------------------------
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        if (
+            ufunc in [np.isnan, np.isinf, np.isfinite]
+            and len(inputs) == 1
+            and inputs[0] is self
+        ):
+            # numpy 1.18 changed isinf and isnan to not raise on dt64/td64
+            return getattr(ufunc, method)(self._ndarray, **kwargs)
+
+        return super().__array_ufunc__(ufunc, method, *inputs, **kwargs)
+
+    def _round(self, freq, mode, ambiguous, nonexistent):
+        # round the local times
+        if isinstance(self.dtype, DatetimeTZDtype):
+            # operate on naive timestamps, then convert back to aware
+            self = cast("DatetimeArray", self)
+            naive = self.tz_localize(None)
+            result = naive._round(freq, mode, ambiguous, nonexistent)
+            return result.tz_localize(
+                self.tz, ambiguous=ambiguous, nonexistent=nonexistent
+            )
+
+        values = self.view("i8")
+        values = cast(np.ndarray, values)
+        nanos = get_unit_for_round(freq, self._creso)
+        if nanos == 0:
+            # GH 52761
+            return self.copy()
+        result_i8 = round_nsint64(values, mode, nanos)
+        result = self._maybe_mask_results(result_i8, fill_value=iNaT)
+        result = result.view(self._ndarray.dtype)
+        return self._simple_new(result, dtype=self.dtype)
+
+    @Appender((_round_doc + _round_example).format(op="round"))
+    def round(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round(freq, RoundTo.NEAREST_HALF_EVEN, ambiguous, nonexistent)
+
+    @Appender((_round_doc + _floor_example).format(op="floor"))
+    def floor(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round(freq, RoundTo.MINUS_INFTY, ambiguous, nonexistent)
+
+    @Appender((_round_doc + _ceil_example).format(op="ceil"))
+    def ceil(
+        self,
+        freq,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        return self._round(freq, RoundTo.PLUS_INFTY, ambiguous, nonexistent)
+
+    # --------------------------------------------------------------
+    # Reductions
+
+    def any(self, *, axis: AxisInt | None = None, skipna: bool = True) -> bool:
+        # GH#34479 the nanops call will raise a TypeError for non-td64 dtype
+        return nanops.nanany(self._ndarray, axis=axis, skipna=skipna, mask=self.isna())
+
+    def all(self, *, axis: AxisInt | None = None, skipna: bool = True) -> bool:
+        # GH#34479 the nanops call will raise a TypeError for non-td64 dtype
+
+        return nanops.nanall(self._ndarray, axis=axis, skipna=skipna, mask=self.isna())
+
+    # --------------------------------------------------------------
+    # Frequency Methods
+
+    def _maybe_clear_freq(self) -> None:
+        self._freq = None
+
+    def _with_freq(self, freq) -> Self:
+        """
+        Helper to get a view on the same data, with a new freq.
+
+        Parameters
+        ----------
+        freq : DateOffset, None, or "infer"
+
+        Returns
+        -------
+        Same type as self
+        """
+        # GH#29843
+        if freq is None:
+            # Always valid
+            pass
+        elif len(self) == 0 and isinstance(freq, BaseOffset):
+            # Always valid.  In the TimedeltaArray case, we require a Tick offset
+            if self.dtype.kind == "m" and not isinstance(freq, Tick):
+                raise TypeError("TimedeltaArray/Index freq must be a Tick")
+        else:
+            # As an internal method, we can ensure this assertion always holds
+            assert freq == "infer"
+            freq = to_offset(self.inferred_freq)
+
+        arr = self.view()
+        arr._freq = freq
+        return arr
+
+    # --------------------------------------------------------------
+    # ExtensionArray Interface
+
+    def _values_for_json(self) -> np.ndarray:
+        # Small performance bump vs the base class which calls np.asarray(self)
+        if isinstance(self.dtype, np.dtype):
+            return self._ndarray
+        return super()._values_for_json()
+
+    def factorize(
+        self,
+        use_na_sentinel: bool = True,
+        sort: bool = False,
+    ):
+        if self.freq is not None:
+            # We must be unique, so can short-circuit (and retain freq)
+            if sort and self.freq.n < 0:
+                codes = np.arange(len(self) - 1, -1, -1, dtype=np.intp)
+                uniques = self[::-1]
+            else:
+                codes = np.arange(len(self), dtype=np.intp)
+                uniques = self.copy()  # TODO: copy or view?
+            return codes, uniques
+
+        if sort:
+            # algorithms.factorize only passes sort=True here when freq is
+            #  not None, so this should not be reached.
+            raise NotImplementedError(
+                f"The 'sort' keyword in {type(self).__name__}.factorize is "
+                "ignored unless arr.freq is not None. To factorize with sort, "
+                "call pd.factorize(obj, sort=True) instead."
+            )
+        return super().factorize(use_na_sentinel=use_na_sentinel)
+
+    @classmethod
+    def _concat_same_type(
+        cls,
+        to_concat: Sequence[Self],
+        axis: AxisInt = 0,
+    ) -> Self:
+        new_obj = super()._concat_same_type(to_concat, axis)
+
+        obj = to_concat[0]
+
+        if axis == 0:
+            # GH 3232: If the concat result is evenly spaced, we can retain the
+            # original frequency
+            to_concat = [x for x in to_concat if len(x)]
+
+            if obj.freq is not None and all(x.freq == obj.freq for x in to_concat):
+                pairs = zip(to_concat[:-1], to_concat[1:])
+                if all(pair[0][-1] + obj.freq == pair[1][0] for pair in pairs):
+                    new_freq = obj.freq
+                    new_obj._freq = new_freq
+        return new_obj
+
+    def copy(self, order: str = "C") -> Self:
+        new_obj = super().copy(order=order)
+        new_obj._freq = self.freq
+        return new_obj
+
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        axis: int,
+        index: Index,
+        limit,
+        limit_direction,
+        limit_area,
+        copy: bool,
+        **kwargs,
+    ) -> Self:
+        """
+        See NDFrame.interpolate.__doc__.
+        """
+        # NB: we return type(self) even if copy=False
+        if method != "linear":
+            raise NotImplementedError
+
+        if not copy:
+            out_data = self._ndarray
+        else:
+            out_data = self._ndarray.copy()
+
+        missing.interpolate_2d_inplace(
+            out_data,
+            method=method,
+            axis=axis,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            **kwargs,
+        )
+        if not copy:
+            return self
+        return type(self)._simple_new(out_data, dtype=self.dtype)
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        *,
+        allow_fill: bool = False,
+        fill_value: Any = None,
+        axis: AxisInt = 0,
+    ) -> Self:
+        result = super().take(
+            indices=indices, allow_fill=allow_fill, fill_value=fill_value, axis=axis
+        )
+
+        indices = np.asarray(indices, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(self))
+
+        if isinstance(maybe_slice, slice):
+            freq = self._get_getitem_freq(maybe_slice)
+            result._freq = freq
+
+        return result
+
+    # --------------------------------------------------------------
+    # Unsorted
+
+    @property
+    def _is_dates_only(self) -> bool:
+        """
+        Check if we are round times at midnight (and no timezone), which will
+        be given a more compact __repr__ than other cases. For TimedeltaArray
+        we are checking for multiples of 24H.
+        """
+        if not lib.is_np_dtype(self.dtype):
+            # i.e. we have a timezone
+            return False
+
+        values_int = self.asi8
+        consider_values = values_int != iNaT
+        reso = get_unit_from_dtype(self.dtype)
+        ppd = periods_per_day(reso)
+
+        # TODO: can we reuse is_date_array_normalized?  would need a skipna kwd
+        #  (first attempt at this was less performant than this implementation)
+        even_days = np.logical_and(consider_values, values_int % ppd != 0).sum() == 0
+        return even_days
+
+
+# -------------------------------------------------------------------
+# Shared Constructor Helpers
+
+
+def ensure_arraylike_for_datetimelike(
+    data, copy: bool, cls_name: str
+) -> tuple[ArrayLike, bool]:
+    if not hasattr(data, "dtype"):
+        # e.g. list, tuple
+        if not isinstance(data, (list, tuple)) and np.ndim(data) == 0:
+            # i.e. generator
+            data = list(data)
+
+        data = construct_1d_object_array_from_listlike(data)
+        copy = False
+    elif isinstance(data, ABCMultiIndex):
+        raise TypeError(f"Cannot create a {cls_name} from a MultiIndex.")
+    else:
+        data = extract_array(data, extract_numpy=True)
+
+    if isinstance(data, IntegerArray) or (
+        isinstance(data, ArrowExtensionArray) and data.dtype.kind in "iu"
+    ):
+        data = data.to_numpy("int64", na_value=iNaT)
+        copy = False
+    elif isinstance(data, ArrowExtensionArray):
+        data = data._maybe_convert_datelike_array()
+        data = data.to_numpy()
+        copy = False
+    elif not isinstance(data, (np.ndarray, ExtensionArray)):
+        # GH#24539 e.g. xarray, dask object
+        data = np.asarray(data)
+
+    elif isinstance(data, ABCCategorical):
+        # GH#18664 preserve tz in going DTI->Categorical->DTI
+        # TODO: cases where we need to do another pass through maybe_convert_dtype,
+        #  e.g. the categories are timedelta64s
+        data = data.categories.take(data.codes, fill_value=NaT)._values
+        copy = False
+
+    return data, copy
+
+
+@overload
+def validate_periods(periods: None) -> None: ...
+
+
+@overload
+def validate_periods(periods: int) -> int: ...
+
+
+def validate_periods(periods: int | None) -> int | None:
+    """
+    If a `periods` argument is passed to the Datetime/Timedelta Array/Index
+    constructor, cast it to an integer.
+
+    Parameters
+    ----------
+    periods : None, int
+
+    Returns
+    -------
+    periods : None or int
+
+    Raises
+    ------
+    TypeError
+        if periods is not None or int
+    """
+    if periods is not None and not lib.is_integer(periods):
+        raise TypeError(f"periods must be an integer, got {periods}")
+    # error: Incompatible return value type (got "int | integer[Any] | None",
+    # expected "int | None")
+    return periods  # type: ignore[return-value]
+
+
+def _validate_inferred_freq(
+    freq: BaseOffset | None, inferred_freq: BaseOffset | None
+) -> BaseOffset | None:
+    """
+    If the user passes a freq and another freq is inferred from passed data,
+    require that they match.
+
+    Parameters
+    ----------
+    freq : DateOffset or None
+    inferred_freq : DateOffset or None
+
+    Returns
+    -------
+    freq : DateOffset or None
+    """
+    if inferred_freq is not None:
+        if freq is not None and freq != inferred_freq:
+            raise ValueError(
+                f"Inferred frequency {inferred_freq} from passed "
+                "values does not conform to passed frequency "
+                f"{freq.freqstr}"
+            )
+        if freq is None:
+            freq = inferred_freq
+
+    return freq
+
+
+def dtype_to_unit(dtype: DatetimeTZDtype | np.dtype | ArrowDtype) -> str:
+    """
+    Return the unit str corresponding to the dtype's resolution.
+
+    Parameters
+    ----------
+    dtype : DatetimeTZDtype or np.dtype
+        If np.dtype, we assume it is a datetime64 dtype.
+
+    Returns
+    -------
+    str
+    """
+    if isinstance(dtype, DatetimeTZDtype):
+        return dtype.unit
+    elif isinstance(dtype, ArrowDtype):
+        if dtype.kind not in "mM":
+            raise ValueError(f"{dtype=} does not have a resolution.")
+        return dtype.pyarrow_dtype.unit
+    return np.datetime_data(dtype)[0]
diff --git a/pandas/pandas/core/arrays/datetimes.py b/pandas/pandas/core/arrays/datetimes.py
new file mode 100644
index 0000000000000..df40c9c11b117
--- /dev/null
+++ b/pandas/pandas/core/arrays/datetimes.py
@@ -0,0 +1,3021 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+    tzinfo,
+)
+from typing import (
+    TYPE_CHECKING,
+    TypeVar,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+from pandas._config.config import get_option
+
+from pandas._libs import (
+    lib,
+    tslib,
+)
+from pandas._libs.tslibs import (
+    BaseOffset,
+    NaT,
+    NaTType,
+    Resolution,
+    Timestamp,
+    astype_overflowsafe,
+    fields,
+    get_resolution,
+    get_supported_dtype,
+    get_unit_from_dtype,
+    ints_to_pydatetime,
+    is_date_array_normalized,
+    is_supported_dtype,
+    is_unitless,
+    normalize_i8_timestamps,
+    timezones,
+    to_offset,
+    tz_convert_from_utc,
+    tzconversion,
+)
+from pandas._libs.tslibs.dtypes import abbrev_to_npy_unit
+from pandas.errors import PerformanceWarning
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import validate_inclusive
+
+from pandas.core.dtypes.common import (
+    DT64NS_DTYPE,
+    INT64_DTYPE,
+    is_bool_dtype,
+    is_float_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.arrays import datetimelike as dtl
+from pandas.core.arrays._ranges import generate_regular_range
+import pandas.core.common as com
+
+from pandas.tseries.frequencies import get_period_alias
+from pandas.tseries.offsets import (
+    Day,
+    Tick,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Iterator,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        DateTimeErrorChoices,
+        DtypeObj,
+        IntervalClosedType,
+        Self,
+        TimeAmbiguous,
+        TimeNonexistent,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Timedelta,
+    )
+    from pandas.core.arrays import PeriodArray
+
+    _TimestampNoneT1 = TypeVar("_TimestampNoneT1", Timestamp, None)
+    _TimestampNoneT2 = TypeVar("_TimestampNoneT2", Timestamp, None)
+
+
+_ITER_CHUNKSIZE = 10_000
+
+
+@overload
+def tz_to_dtype(tz: tzinfo, unit: str = ...) -> DatetimeTZDtype: ...
+
+
+@overload
+def tz_to_dtype(tz: None, unit: str = ...) -> np.dtype[np.datetime64]: ...
+
+
+def tz_to_dtype(
+    tz: tzinfo | None, unit: str = "ns"
+) -> np.dtype[np.datetime64] | DatetimeTZDtype:
+    """
+    Return a datetime64[ns] dtype appropriate for the given timezone.
+
+    Parameters
+    ----------
+    tz : tzinfo or None
+    unit : str, default "ns"
+
+    Returns
+    -------
+    np.dtype or Datetime64TZDType
+    """
+    if tz is None:
+        return np.dtype(f"M8[{unit}]")
+    else:
+        return DatetimeTZDtype(tz=tz, unit=unit)
+
+
+def _field_accessor(name: str, field: str, docstring: str | None = None):
+    def f(self):
+        values = self._local_timestamps()
+
+        if field in self._bool_ops:
+            result: np.ndarray
+
+            if field.endswith(("start", "end")):
+                freq = self.freq
+                month_kw = 12
+                if freq:
+                    kwds = freq.kwds
+                    month_kw = kwds.get("startingMonth", kwds.get("month", month_kw))
+
+                if freq is not None:
+                    freq_name = freq.name
+                else:
+                    freq_name = None
+                result = fields.get_start_end_field(
+                    values, field, freq_name, month_kw, reso=self._creso
+                )
+            else:
+                result = fields.get_date_field(values, field, reso=self._creso)
+
+            # these return a boolean by-definition
+            return result
+
+        result = fields.get_date_field(values, field, reso=self._creso)
+        result = self._maybe_mask_results(result, fill_value=None, convert="float64")
+
+        return result
+
+    f.__name__ = name
+    f.__doc__ = docstring
+    return property(f)
+
+
+# error: Definition of "_concat_same_type" in base class "NDArrayBacked" is
+# incompatible with definition in base class "ExtensionArray"
+class DatetimeArray(dtl.TimelikeOps, dtl.DatelikeOps):  # type: ignore[misc]
+    """
+    Pandas ExtensionArray for tz-naive or tz-aware datetime data.
+
+    .. warning::
+
+       DatetimeArray is currently experimental, and its API may change
+       without warning. In particular, :attr:`DatetimeArray.dtype` is
+       expected to change to always be an instance of an ``ExtensionDtype``
+       subclass.
+
+    Parameters
+    ----------
+    data : Series, Index, DatetimeArray, ndarray
+        The datetime data.
+
+        For DatetimeArray `values` (or a Series or Index boxing one),
+        `dtype` and `freq` will be extracted from `values`.
+
+    dtype : numpy.dtype or DatetimeTZDtype
+        Note that the only NumPy dtype allowed is 'datetime64[ns]'.
+    freq : str or Offset, optional
+        The frequency.
+    copy : bool, default False
+        Whether to copy the underlying array of values.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    DatetimeIndex : Immutable Index for datetime-like data.
+    Series : One-dimensional labeled array capable of holding datetime-like data.
+    Timestamp : Pandas replacement for python datetime.datetime object.
+    to_datetime : Convert argument to datetime.
+    period_range : Return a fixed frequency PeriodIndex.
+
+    Examples
+    --------
+    >>> pd.arrays.DatetimeArray._from_sequence(
+    ...     pd.DatetimeIndex(["2023-01-01", "2023-01-02"], freq="D")
+    ... )
+    <DatetimeArray>
+    ['2023-01-01 00:00:00', '2023-01-02 00:00:00']
+    Length: 2, dtype: datetime64[s]
+    """
+
+    _typ = "datetimearray"
+    _internal_fill_value = np.datetime64("NaT", "ns")
+    _recognized_scalars = (datetime, np.datetime64)
+    _is_recognized_dtype = lambda x: lib.is_np_dtype(x, "M") or isinstance(
+        x, DatetimeTZDtype
+    )
+    _infer_matches = ("datetime", "datetime64", "date")
+
+    @property
+    def _scalar_type(self) -> type[Timestamp]:
+        return Timestamp
+
+    # define my properties & methods for delegation
+    _bool_ops: list[str] = [
+        "is_month_start",
+        "is_month_end",
+        "is_quarter_start",
+        "is_quarter_end",
+        "is_year_start",
+        "is_year_end",
+        "is_leap_year",
+    ]
+    _field_ops: list[str] = [
+        "year",
+        "month",
+        "day",
+        "hour",
+        "minute",
+        "second",
+        "weekday",
+        "dayofweek",
+        "day_of_week",
+        "dayofyear",
+        "day_of_year",
+        "quarter",
+        "days_in_month",
+        "daysinmonth",
+        "microsecond",
+        "nanosecond",
+    ]
+    _other_ops: list[str] = ["date", "time", "timetz"]
+    _datetimelike_ops: list[str] = (
+        _field_ops + _bool_ops + _other_ops + ["unit", "freq", "tz"]
+    )
+    _datetimelike_methods: list[str] = [
+        "to_period",
+        "tz_localize",
+        "tz_convert",
+        "normalize",
+        "strftime",
+        "round",
+        "floor",
+        "ceil",
+        "month_name",
+        "day_name",
+        "as_unit",
+    ]
+
+    # ndim is inherited from ExtensionArray, must exist to ensure
+    #  Timestamp.__richcmp__(DateTimeArray) operates pointwise
+
+    # ensure that operations with numpy arrays defer to our implementation
+    __array_priority__ = 1000
+
+    # -----------------------------------------------------------------
+    # Constructors
+
+    _dtype: np.dtype[np.datetime64] | DatetimeTZDtype
+    _freq: BaseOffset | None = None
+
+    @classmethod
+    def _from_scalars(cls, scalars, *, dtype: DtypeObj) -> Self:
+        if lib.infer_dtype(scalars, skipna=True) not in ["datetime", "datetime64"]:
+            # TODO: require any NAs be valid-for-DTA
+            # TODO: if dtype is passed, check for tzawareness compat?
+            raise ValueError
+        return cls._from_sequence(scalars, dtype=dtype)
+
+    @classmethod
+    def _validate_dtype(cls, values, dtype):
+        # used in TimeLikeOps.__init__
+        dtype = _validate_dt64_dtype(dtype)
+        _validate_dt64_dtype(values.dtype)
+        if isinstance(dtype, np.dtype):
+            if values.dtype != dtype:
+                raise ValueError("Values resolution does not match dtype.")
+        else:
+            vunit = np.datetime_data(values.dtype)[0]
+            if vunit != dtype.unit:
+                raise ValueError("Values resolution does not match dtype.")
+        return dtype
+
+    # error: Signature of "_simple_new" incompatible with supertype "NDArrayBacked"
+    @classmethod
+    def _simple_new(  # type: ignore[override]
+        cls,
+        values: npt.NDArray[np.datetime64],
+        freq: BaseOffset | None = None,
+        dtype: np.dtype[np.datetime64] | DatetimeTZDtype = DT64NS_DTYPE,
+    ) -> Self:
+        assert isinstance(values, np.ndarray)
+        assert dtype.kind == "M"
+        if isinstance(dtype, np.dtype):
+            assert dtype == values.dtype
+            assert not is_unitless(dtype)
+        else:
+            # DatetimeTZDtype. If we have e.g. DatetimeTZDtype[us, UTC],
+            #  then values.dtype should be M8[us].
+            assert dtype._creso == get_unit_from_dtype(values.dtype)
+
+        result = super()._simple_new(values, dtype)
+        result._freq = freq
+        return result
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy: bool = False) -> Self:
+        return cls._from_sequence_not_strict(scalars, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _from_sequence_not_strict(
+        cls,
+        data,
+        *,
+        dtype=None,
+        copy: bool = False,
+        tz=lib.no_default,
+        freq: str | BaseOffset | lib.NoDefault | None = lib.no_default,
+        dayfirst: bool = False,
+        yearfirst: bool = False,
+        ambiguous: TimeAmbiguous = "raise",
+    ) -> Self:
+        """
+        A non-strict version of _from_sequence, called from DatetimeIndex.__new__.
+        """
+
+        # if the user either explicitly passes tz=None or a tz-naive dtype, we
+        #  disallows inferring a tz.
+        explicit_tz_none = tz is None
+        if tz is lib.no_default:
+            tz = None
+        else:
+            tz = timezones.maybe_get_tz(tz)
+
+        dtype = _validate_dt64_dtype(dtype)
+        # if dtype has an embedded tz, capture it
+        tz = _validate_tz_from_dtype(dtype, tz, explicit_tz_none)
+
+        unit = None
+        if dtype is not None:
+            unit = dtl.dtype_to_unit(dtype)
+
+        data, copy = dtl.ensure_arraylike_for_datetimelike(
+            data, copy, cls_name="DatetimeArray"
+        )
+        inferred_freq = None
+        if isinstance(data, DatetimeArray):
+            inferred_freq = data.freq
+
+        subarr, tz = _sequence_to_dt64(
+            data,
+            copy=copy,
+            tz=tz,
+            dayfirst=dayfirst,
+            yearfirst=yearfirst,
+            ambiguous=ambiguous,
+            out_unit=unit,
+        )
+        # We have to call this again after possibly inferring a tz above
+        _validate_tz_from_dtype(dtype, tz, explicit_tz_none)
+        if tz is not None and explicit_tz_none:
+            raise ValueError(
+                "Passed data is timezone-aware, incompatible with 'tz=None'. "
+                "Use obj.tz_localize(None) instead."
+            )
+
+        data_unit = np.datetime_data(subarr.dtype)[0]
+        data_dtype = tz_to_dtype(tz, data_unit)
+        result = cls._simple_new(subarr, freq=inferred_freq, dtype=data_dtype)
+        if unit is not None and unit != result.unit:
+            # If unit was specified in user-passed dtype, cast to it here
+            result = result.as_unit(unit)
+
+        validate_kwds = {"ambiguous": ambiguous}
+        result._maybe_pin_freq(freq, validate_kwds)
+        return result
+
+    @classmethod
+    def _generate_range(
+        cls,
+        start,
+        end,
+        periods: int | None,
+        freq,
+        tz=None,
+        normalize: bool = False,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+        inclusive: IntervalClosedType = "both",
+        *,
+        unit: str | None = None,
+    ) -> Self:
+        periods = dtl.validate_periods(periods)
+        if freq is None and any(x is None for x in [periods, start, end]):
+            raise ValueError("Must provide freq argument if no data is supplied")
+
+        if com.count_not_none(start, end, periods, freq) != 3:
+            raise ValueError(
+                "Of the four parameters: start, end, periods, "
+                "and freq, exactly three must be specified"
+            )
+        freq = to_offset(freq)
+
+        if start is not None:
+            start = Timestamp(start)
+
+        if end is not None:
+            end = Timestamp(end)
+
+        if start is NaT or end is NaT:
+            raise ValueError("Neither `start` nor `end` can be NaT")
+
+        if unit is not None:
+            if unit not in ["s", "ms", "us", "ns"]:
+                raise ValueError("'unit' must be one of 's', 'ms', 'us', 'ns'")
+        else:
+            unit = "ns"
+
+        if start is not None:
+            start = start.as_unit(unit, round_ok=False)
+        if end is not None:
+            end = end.as_unit(unit, round_ok=False)
+
+        left_inclusive, right_inclusive = validate_inclusive(inclusive)
+        start, end = _maybe_normalize_endpoints(start, end, normalize)
+        tz = _infer_tz_from_endpoints(start, end, tz)
+
+        if tz is not None:
+            # Localize the start and end arguments
+            start = _maybe_localize_point(start, freq, tz, ambiguous, nonexistent)
+            end = _maybe_localize_point(end, freq, tz, ambiguous, nonexistent)
+
+        if freq is not None:
+            # We break Day arithmetic (fixed 24 hour) here and opt for
+            # Day to mean calendar day (23/24/25 hour). Therefore, strip
+            # tz info from start and day to avoid DST arithmetic
+            if isinstance(freq, Day):
+                if start is not None:
+                    start = start.tz_localize(None)
+                if end is not None:
+                    end = end.tz_localize(None)
+
+            if isinstance(freq, Tick):
+                i8values = generate_regular_range(start, end, periods, freq, unit=unit)
+            else:
+                xdr = _generate_range(
+                    start=start, end=end, periods=periods, offset=freq, unit=unit
+                )
+                i8values = np.array([x._value for x in xdr], dtype=np.int64)
+
+            endpoint_tz = start.tz if start is not None else end.tz
+
+            if tz is not None and endpoint_tz is None:
+                if not timezones.is_utc(tz):
+                    # short-circuit tz_localize_to_utc which would make
+                    #  an unnecessary copy with UTC but be a no-op.
+                    creso = abbrev_to_npy_unit(unit)
+                    i8values = tzconversion.tz_localize_to_utc(
+                        i8values,
+                        tz,
+                        ambiguous=ambiguous,
+                        nonexistent=nonexistent,
+                        creso=creso,
+                    )
+
+                # i8values is localized datetime64 array -> have to convert
+                # start/end as well to compare
+                if start is not None:
+                    start = start.tz_localize(tz, ambiguous, nonexistent)
+                if end is not None:
+                    end = end.tz_localize(tz, ambiguous, nonexistent)
+        else:
+            # Create a linearly spaced date_range in local time
+            # Nanosecond-granularity timestamps aren't always correctly
+            # representable with doubles, so we limit the range that we
+            # pass to np.linspace as much as possible
+            periods = cast(int, periods)
+            i8values = (
+                np.linspace(0, end._value - start._value, periods, dtype="int64")
+                + start._value
+            )
+            if i8values.dtype != "i8":
+                # 2022-01-09 I (brock) am not sure if it is possible for this
+                #  to overflow and cast to e.g. f8, but if it does we need to cast
+                i8values = i8values.astype("i8")
+
+        if start == end:
+            if not left_inclusive and not right_inclusive:
+                i8values = i8values[1:-1]
+        else:
+            start_i8 = Timestamp(start)._value
+            end_i8 = Timestamp(end)._value
+            if not left_inclusive or not right_inclusive:
+                if not left_inclusive and len(i8values) and i8values[0] == start_i8:
+                    i8values = i8values[1:]
+                if not right_inclusive and len(i8values) and i8values[-1] == end_i8:
+                    i8values = i8values[:-1]
+
+        dt64_values = i8values.view(f"datetime64[{unit}]")
+        dtype = tz_to_dtype(tz, unit=unit)
+        return cls._simple_new(dt64_values, freq=freq, dtype=dtype)
+
+    # -----------------------------------------------------------------
+    # DatetimeLike Interface
+
+    def _unbox_scalar(self, value) -> np.datetime64:
+        if not isinstance(value, self._scalar_type) and value is not NaT:
+            raise ValueError("'value' should be a Timestamp.")
+        self._check_compatible_with(value)
+        if value is NaT:
+            return np.datetime64(value._value, self.unit)
+        else:
+            return value.as_unit(self.unit, round_ok=False).asm8
+
+    def _scalar_from_string(self, value) -> Timestamp | NaTType:
+        return Timestamp(value, tz=self.tz)
+
+    def _check_compatible_with(self, other) -> None:
+        if other is NaT:
+            return
+        self._assert_tzawareness_compat(other)
+
+    # -----------------------------------------------------------------
+    # Descriptive Properties
+
+    def _box_func(self, x: np.datetime64) -> Timestamp | NaTType:
+        # GH#42228
+        value = x.view("i8")
+        ts = Timestamp._from_value_and_reso(value, reso=self._creso, tz=self.tz)
+        return ts
+
+    @property
+    # error: Return type "Union[dtype, DatetimeTZDtype]" of "dtype"
+    # incompatible with return type "ExtensionDtype" in supertype
+    # "ExtensionArray"
+    def dtype(self) -> np.dtype[np.datetime64] | DatetimeTZDtype:  # type: ignore[override]
+        """
+        The dtype for the DatetimeArray.
+
+        .. warning::
+
+           A future version of pandas will change dtype to never be a
+           ``numpy.dtype``. Instead, :attr:`DatetimeArray.dtype` will
+           always be an instance of an ``ExtensionDtype`` subclass.
+
+        Returns
+        -------
+        numpy.dtype or DatetimeTZDtype
+            If the values are tz-naive, then ``np.dtype('datetime64[ns]')``
+            is returned.
+
+            If the values are tz-aware, then the ``DatetimeTZDtype``
+            is returned.
+        """
+        return self._dtype
+
+    @property
+    def tz(self) -> tzinfo | None:
+        """
+        Return the timezone.
+
+        Returns
+        -------
+        zoneinfo.ZoneInfo,, datetime.tzinfo, pytz.tzinfo.BaseTZInfo, dateutil.tz.tz.tzfile, or None
+            Returns None when the array is tz-naive.
+
+        See Also
+        --------
+        DatetimeIndex.tz_localize : Localize tz-naive DatetimeIndex to a
+            given time zone, or remove timezone from a tz-aware DatetimeIndex.
+        DatetimeIndex.tz_convert : Convert tz-aware DatetimeIndex from
+            one time zone to another.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.tz
+        datetime.timezone.utc
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(
+        ...     ["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"]
+        ... )
+        >>> idx.tz
+        datetime.timezone.utc
+        """  # noqa: E501
+        # GH 18595
+        return getattr(self.dtype, "tz", None)
+
+    @tz.setter
+    def tz(self, value):
+        # GH 3746: Prevent localizing or converting the index by setting tz
+        raise AttributeError(
+            "Cannot directly set timezone. Use tz_localize() "
+            "or tz_convert() as appropriate"
+        )
+
+    @property
+    def tzinfo(self) -> tzinfo | None:
+        """
+        Alias for tz attribute
+        """
+        return self.tz
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def is_normalized(self) -> bool:
+        """
+        Returns True if all of the dates are at midnight ("no time")
+        """
+        return is_date_array_normalized(self.asi8, self.tz, reso=self._creso)
+
+    @property  # NB: override with cache_readonly in immutable subclasses
+    def _resolution_obj(self) -> Resolution:
+        return get_resolution(self.asi8, self.tz, reso=self._creso)
+
+    # ----------------------------------------------------------------
+    # Array-Like / EA-Interface Methods
+
+    def __array__(self, dtype=None, copy=None) -> np.ndarray:
+        if dtype is None and self.tz:
+            # The default for tz-aware is object, to preserve tz info
+            dtype = object
+
+        return super().__array__(dtype=dtype, copy=copy)
+
+    def __iter__(self) -> Iterator:
+        """
+        Return an iterator over the boxed values
+
+        Yields
+        ------
+        tstamp : Timestamp
+        """
+        if self.ndim > 1:
+            for i in range(len(self)):
+                yield self[i]
+        else:
+            # convert in chunks of 10k for efficiency
+            data = self.asi8
+            length = len(self)
+            chunksize = _ITER_CHUNKSIZE
+            chunks = (length // chunksize) + 1
+
+            for i in range(chunks):
+                start_i = i * chunksize
+                end_i = min((i + 1) * chunksize, length)
+                converted = ints_to_pydatetime(
+                    data[start_i:end_i],
+                    tz=self.tz,
+                    box="timestamp",
+                    reso=self._creso,
+                )
+                yield from converted
+
+    def astype(self, dtype, copy: bool = True):
+        # We handle
+        #   --> datetime
+        #   --> period
+        # DatetimeLikeArrayMixin Super handles the rest.
+        dtype = pandas_dtype(dtype)
+
+        if dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+
+        elif isinstance(dtype, ExtensionDtype):
+            if not isinstance(dtype, DatetimeTZDtype):
+                # e.g. Sparse[datetime64[ns]]
+                return super().astype(dtype, copy=copy)
+            elif self.tz is None:
+                # pre-2.0 this did self.tz_localize(dtype.tz), which did not match
+                #  the Series behavior which did
+                #  values.tz_localize("UTC").tz_convert(dtype.tz)
+                raise TypeError(
+                    "Cannot use .astype to convert from timezone-naive dtype to "
+                    "timezone-aware dtype. Use obj.tz_localize instead or "
+                    "series.dt.tz_localize instead"
+                )
+            else:
+                # tzaware unit conversion e.g. datetime64[s, UTC]
+                np_dtype = np.dtype(dtype.str)
+                res_values = astype_overflowsafe(self._ndarray, np_dtype, copy=copy)
+                return type(self)._simple_new(res_values, dtype=dtype, freq=self.freq)
+
+        elif (
+            self.tz is None
+            and lib.is_np_dtype(dtype, "M")
+            and not is_unitless(dtype)
+            and is_supported_dtype(dtype)
+        ):
+            # unit conversion e.g. datetime64[s]
+            res_values = astype_overflowsafe(self._ndarray, dtype, copy=True)
+            return type(self)._simple_new(res_values, dtype=res_values.dtype)
+            # TODO: preserve freq?
+
+        elif self.tz is not None and lib.is_np_dtype(dtype, "M"):
+            # pre-2.0 behavior for DTA/DTI was
+            #  values.tz_convert("UTC").tz_localize(None), which did not match
+            #  the Series behavior
+            raise TypeError(
+                "Cannot use .astype to convert from timezone-aware dtype to "
+                "timezone-naive dtype. Use obj.tz_localize(None) or "
+                "obj.tz_convert('UTC').tz_localize(None) instead."
+            )
+
+        elif (
+            self.tz is None
+            and lib.is_np_dtype(dtype, "M")
+            and dtype != self.dtype
+            and is_unitless(dtype)
+        ):
+            raise TypeError(
+                "Casting to unit-less dtype 'datetime64' is not supported. "
+                "Pass e.g. 'datetime64[ns]' instead."
+            )
+
+        elif isinstance(dtype, PeriodDtype):
+            return self.to_period(freq=dtype.freq)
+        return dtl.DatetimeLikeArrayMixin.astype(self, dtype, copy)
+
+    # -----------------------------------------------------------------
+    # Rendering Methods
+
+    def _format_native_types(
+        self, *, na_rep: str | float = "NaT", date_format=None, **kwargs
+    ) -> npt.NDArray[np.object_]:
+        if date_format is None and self._is_dates_only:
+            # Only dates and no timezone: provide a default format
+            date_format = "%Y-%m-%d"
+
+        return tslib.format_array_from_datetime(
+            self.asi8, tz=self.tz, format=date_format, na_rep=na_rep, reso=self._creso
+        )
+
+    # -----------------------------------------------------------------
+    # Comparison Methods
+
+    def _assert_tzawareness_compat(self, other) -> None:
+        # adapted from _Timestamp._assert_tzawareness_compat
+        other_tz = getattr(other, "tzinfo", None)
+        other_dtype = getattr(other, "dtype", None)
+
+        if isinstance(other_dtype, DatetimeTZDtype):
+            # Get tzinfo from Series dtype
+            other_tz = other.dtype.tz
+        if other is NaT:
+            # pd.NaT quacks both aware and naive
+            pass
+        elif self.tz is None:
+            if other_tz is not None:
+                raise TypeError(
+                    "Cannot compare tz-naive and tz-aware datetime-like objects."
+                )
+        elif other_tz is None:
+            raise TypeError(
+                "Cannot compare tz-naive and tz-aware datetime-like objects"
+            )
+
+    # -----------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _add_offset(self, offset: BaseOffset) -> Self:
+        assert not isinstance(offset, Tick)
+
+        if self.tz is not None:
+            values = self.tz_localize(None)
+        else:
+            values = self
+
+        try:
+            res_values = offset._apply_array(values._ndarray)
+            if res_values.dtype.kind == "i":
+                # error: Argument 1 to "view" of "ndarray" has incompatible type
+                # "dtype[datetime64] | DatetimeTZDtype"; expected
+                # "dtype[Any] | type[Any] | _SupportsDType[dtype[Any]]"
+                res_values = res_values.view(values.dtype)  # type: ignore[arg-type]
+        except NotImplementedError:
+            if get_option("performance_warnings"):
+                warnings.warn(
+                    "Non-vectorized DateOffset being applied to Series or "
+                    "DatetimeIndex.",
+                    PerformanceWarning,
+                    stacklevel=find_stack_level(),
+                )
+            res_values = self.astype("O") + offset
+            result = type(self)._from_sequence(res_values, dtype=self.dtype)
+
+        else:
+            result = type(self)._simple_new(res_values, dtype=res_values.dtype)
+            if offset.normalize:
+                result = result.normalize()
+                result._freq = None
+
+            if self.tz is not None:
+                result = result.tz_localize(self.tz)
+
+        return result
+
+    # -----------------------------------------------------------------
+    # Timezone Conversion and Localization Methods
+
+    def _local_timestamps(self) -> npt.NDArray[np.int64]:
+        """
+        Convert to an i8 (unix-like nanosecond timestamp) representation
+        while keeping the local timezone and not using UTC.
+        This is used to calculate time-of-day information as if the timestamps
+        were timezone-naive.
+        """
+        if self.tz is None or timezones.is_utc(self.tz):
+            # Avoid the copy that would be made in tzconversion
+            return self.asi8
+        return tz_convert_from_utc(self.asi8, self.tz, reso=self._creso)
+
+    def tz_convert(self, tz) -> Self:
+        """
+        Convert tz-aware Datetime Array/Index from one time zone to another.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile, datetime.tzinfo or None
+            Time zone for time. Corresponding timestamps would be converted
+            to this time zone of the Datetime Array/Index. A `tz` of None will
+            convert to UTC and remove the timezone information.
+
+        Returns
+        -------
+        Array or Index
+            Datetme Array/Index with target `tz`.
+
+        Raises
+        ------
+        TypeError
+            If Datetime Array/Index is tz-naive.
+
+        See Also
+        --------
+        DatetimeIndex.tz : A timezone that has a variable offset from UTC.
+        DatetimeIndex.tz_localize : Localize tz-naive DatetimeIndex to a
+            given time zone, or remove timezone from a tz-aware DatetimeIndex.
+
+        Examples
+        --------
+        With the `tz` parameter, we can change the DatetimeIndex
+        to other time zones:
+
+        >>> dti = pd.date_range(
+        ...     start="2014-08-01 09:00", freq="h", periods=3, tz="Europe/Berlin"
+        ... )
+
+        >>> dti
+        DatetimeIndex(['2014-08-01 09:00:00+02:00',
+                       '2014-08-01 10:00:00+02:00',
+                       '2014-08-01 11:00:00+02:00'],
+                      dtype='datetime64[ns, Europe/Berlin]', freq='h')
+
+        >>> dti.tz_convert("US/Central")
+        DatetimeIndex(['2014-08-01 02:00:00-05:00',
+                       '2014-08-01 03:00:00-05:00',
+                       '2014-08-01 04:00:00-05:00'],
+                      dtype='datetime64[ns, US/Central]', freq='h')
+
+        With the ``tz=None``, we can remove the timezone (after converting
+        to UTC if necessary):
+
+        >>> dti = pd.date_range(
+        ...     start="2014-08-01 09:00", freq="h", periods=3, tz="Europe/Berlin"
+        ... )
+
+        >>> dti
+        DatetimeIndex(['2014-08-01 09:00:00+02:00',
+                       '2014-08-01 10:00:00+02:00',
+                       '2014-08-01 11:00:00+02:00'],
+                        dtype='datetime64[ns, Europe/Berlin]', freq='h')
+
+        >>> dti.tz_convert(None)
+        DatetimeIndex(['2014-08-01 07:00:00',
+                       '2014-08-01 08:00:00',
+                       '2014-08-01 09:00:00'],
+                        dtype='datetime64[ns]', freq='h')
+        """  # noqa: E501
+        tz = timezones.maybe_get_tz(tz)
+
+        if self.tz is None:
+            # tz naive, use tz_localize
+            raise TypeError(
+                "Cannot convert tz-naive timestamps, use tz_localize to localize"
+            )
+
+        # No conversion since timestamps are all UTC to begin with
+        dtype = tz_to_dtype(tz, unit=self.unit)
+        return self._simple_new(self._ndarray, dtype=dtype, freq=self.freq)
+
+    @dtl.ravel_compat
+    def tz_localize(
+        self,
+        tz,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        """
+        Localize tz-naive Datetime Array/Index to tz-aware Datetime Array/Index.
+
+        This method takes a time zone (tz) naive Datetime Array/Index object
+        and makes this time zone aware. It does not move the time to another
+        time zone.
+
+        This method can also be used to do the inverse -- to create a time
+        zone unaware object from an aware object. To that end, pass `tz=None`.
+
+        Parameters
+        ----------
+        tz : str, zoneinfo.ZoneInfo,, pytz.timezone, dateutil.tz.tzfile, datetime.tzinfo or None
+            Time zone to convert timestamps to. Passing ``None`` will
+            remove the time zone information preserving local time.
+        ambiguous : 'infer', 'NaT', bool array, default 'raise'
+            When clocks moved backward due to DST, ambiguous times may arise.
+            For example in Central European Time (UTC+01), when going from
+            03:00 DST to 02:00 non-DST, 02:30:00 local time occurs both at
+            00:30:00 UTC and at 01:30:00 UTC. In such a situation, the
+            `ambiguous` parameter dictates how ambiguous times should be
+            handled.
+
+            - 'infer' will attempt to infer fall dst-transition hours based on
+              order
+            - bool-ndarray where True signifies a DST time, False signifies a
+              non-DST time (note that this flag is only applicable for
+              ambiguous times)
+            - 'NaT' will return NaT where there are ambiguous times
+            - 'raise' will raise a ValueError if there are ambiguous
+              times.
+
+        nonexistent : 'shift_forward', 'shift_backward, 'NaT', timedelta, \
+default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST.
+
+            - 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time
+            - 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time
+            - 'NaT' will return NaT where there are nonexistent times
+            - timedelta objects will shift nonexistent times by the timedelta
+            - 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        Same type as self
+            Array/Index converted to the specified time zone.
+
+        Raises
+        ------
+        TypeError
+            If the Datetime Array/Index is tz-aware and tz is not None.
+
+        See Also
+        --------
+        DatetimeIndex.tz_convert : Convert tz-aware DatetimeIndex from
+            one time zone to another.
+
+        Examples
+        --------
+        >>> tz_naive = pd.date_range('2018-03-01 09:00', periods=3)
+        >>> tz_naive
+        DatetimeIndex(['2018-03-01 09:00:00', '2018-03-02 09:00:00',
+                       '2018-03-03 09:00:00'],
+                      dtype='datetime64[ns]', freq='D')
+
+        Localize DatetimeIndex in US/Eastern time zone:
+
+        >>> tz_aware = tz_naive.tz_localize(tz='US/Eastern')
+        >>> tz_aware
+        DatetimeIndex(['2018-03-01 09:00:00-05:00',
+                       '2018-03-02 09:00:00-05:00',
+                       '2018-03-03 09:00:00-05:00'],
+                      dtype='datetime64[ns, US/Eastern]', freq=None)
+
+        With the ``tz=None``, we can remove the time zone information
+        while keeping the local time (not converted to UTC):
+
+        >>> tz_aware.tz_localize(None)
+        DatetimeIndex(['2018-03-01 09:00:00', '2018-03-02 09:00:00',
+                       '2018-03-03 09:00:00'],
+                      dtype='datetime64[ns]', freq=None)
+
+        Be careful with DST changes. When there is sequential data, pandas can
+        infer the DST time:
+
+        >>> s = pd.to_datetime(pd.Series(['2018-10-28 01:30:00',
+        ...                               '2018-10-28 02:00:00',
+        ...                               '2018-10-28 02:30:00',
+        ...                               '2018-10-28 02:00:00',
+        ...                               '2018-10-28 02:30:00',
+        ...                               '2018-10-28 03:00:00',
+        ...                               '2018-10-28 03:30:00']))
+        >>> s.dt.tz_localize('CET', ambiguous='infer')
+        0   2018-10-28 01:30:00+02:00
+        1   2018-10-28 02:00:00+02:00
+        2   2018-10-28 02:30:00+02:00
+        3   2018-10-28 02:00:00+01:00
+        4   2018-10-28 02:30:00+01:00
+        5   2018-10-28 03:00:00+01:00
+        6   2018-10-28 03:30:00+01:00
+        dtype: datetime64[s, CET]
+
+        In some cases, inferring the DST is impossible. In such cases, you can
+        pass an ndarray to the ambiguous parameter to set the DST explicitly
+
+        >>> s = pd.to_datetime(pd.Series(['2018-10-28 01:20:00',
+        ...                               '2018-10-28 02:36:00',
+        ...                               '2018-10-28 03:46:00']))
+        >>> s.dt.tz_localize('CET', ambiguous=np.array([True, True, False]))
+        0   2018-10-28 01:20:00+02:00
+        1   2018-10-28 02:36:00+02:00
+        2   2018-10-28 03:46:00+01:00
+        dtype: datetime64[s, CET]
+
+        If the DST transition causes nonexistent times, you can shift these
+        dates forward or backwards with a timedelta object or `'shift_forward'`
+        or `'shift_backwards'`.
+
+        >>> s = pd.to_datetime(pd.Series(['2015-03-29 02:30:00',
+        ...                               '2015-03-29 03:30:00'], dtype="M8[ns]"))
+        >>> s.dt.tz_localize('Europe/Warsaw', nonexistent='shift_forward')
+        0   2015-03-29 03:00:00+02:00
+        1   2015-03-29 03:30:00+02:00
+        dtype: datetime64[ns, Europe/Warsaw]
+
+        >>> s.dt.tz_localize('Europe/Warsaw', nonexistent='shift_backward')
+        0   2015-03-29 01:59:59.999999999+01:00
+        1   2015-03-29 03:30:00+02:00
+        dtype: datetime64[ns, Europe/Warsaw]
+
+        >>> s.dt.tz_localize('Europe/Warsaw', nonexistent=pd.Timedelta('1h'))
+        0   2015-03-29 03:30:00+02:00
+        1   2015-03-29 03:30:00+02:00
+        dtype: datetime64[ns, Europe/Warsaw]
+        """  # noqa: E501
+        nonexistent_options = ("raise", "NaT", "shift_forward", "shift_backward")
+        if nonexistent not in nonexistent_options and not isinstance(
+            nonexistent, timedelta
+        ):
+            raise ValueError(
+                "The nonexistent argument must be one of 'raise', "
+                "'NaT', 'shift_forward', 'shift_backward' or "
+                "a timedelta object"
+            )
+
+        if self.tz is not None:
+            if tz is None:
+                new_dates = tz_convert_from_utc(self.asi8, self.tz, reso=self._creso)
+            else:
+                raise TypeError("Already tz-aware, use tz_convert to convert.")
+        else:
+            tz = timezones.maybe_get_tz(tz)
+            # Convert to UTC
+
+            new_dates = tzconversion.tz_localize_to_utc(
+                self.asi8,
+                tz,
+                ambiguous=ambiguous,
+                nonexistent=nonexistent,
+                creso=self._creso,
+            )
+        new_dates_dt64 = new_dates.view(f"M8[{self.unit}]")
+        dtype = tz_to_dtype(tz, unit=self.unit)
+
+        freq = None
+        if timezones.is_utc(tz) or (len(self) == 1 and not isna(new_dates_dt64[0])):
+            # we can preserve freq
+            # TODO: Also for fixed-offsets
+            freq = self.freq
+        elif tz is None and self.tz is None:
+            # no-op
+            freq = self.freq
+        return self._simple_new(new_dates_dt64, dtype=dtype, freq=freq)
+
+    # ----------------------------------------------------------------
+    # Conversion Methods - Vectorized analogues of Timestamp methods
+
+    def to_pydatetime(self) -> npt.NDArray[np.object_]:
+        """
+        Return an ndarray of ``datetime.datetime`` objects.
+
+        Returns
+        -------
+        numpy.ndarray
+            An ndarray of ``datetime.datetime`` objects.
+
+        See Also
+        --------
+        DatetimeIndex.to_julian_date : Converts Datetime Array to float64 ndarray
+            of Julian Dates.
+
+        Examples
+        --------
+        >>> idx = pd.date_range("2018-02-27", periods=3)
+        >>> idx.to_pydatetime()
+        array([datetime.datetime(2018, 2, 27, 0, 0),
+               datetime.datetime(2018, 2, 28, 0, 0),
+               datetime.datetime(2018, 3, 1, 0, 0)], dtype=object)
+        """
+        return ints_to_pydatetime(self.asi8, tz=self.tz, reso=self._creso)
+
+    def normalize(self) -> Self:
+        """
+        Convert times to midnight.
+
+        The time component of the date-time is converted to midnight i.e.
+        00:00:00. This is useful in cases, when the time does not matter.
+        Length is unaltered. The timezones are unaffected.
+
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on Datetime Array/Index.
+
+        Returns
+        -------
+        DatetimeArray, DatetimeIndex or Series
+            The same type as the original data. Series will have the same
+            name and index. DatetimeIndex will have the same name.
+
+        See Also
+        --------
+        floor : Floor the datetimes to the specified freq.
+        ceil : Ceil the datetimes to the specified freq.
+        round : Round the datetimes to the specified freq.
+
+        Examples
+        --------
+        >>> idx = pd.date_range(
+        ...     start="2014-08-01 10:00", freq="h", periods=3, tz="Asia/Calcutta"
+        ... )
+        >>> idx
+        DatetimeIndex(['2014-08-01 10:00:00+05:30',
+                       '2014-08-01 11:00:00+05:30',
+                       '2014-08-01 12:00:00+05:30'],
+                        dtype='datetime64[ns, Asia/Calcutta]', freq='h')
+        >>> idx.normalize()
+        DatetimeIndex(['2014-08-01 00:00:00+05:30',
+                       '2014-08-01 00:00:00+05:30',
+                       '2014-08-01 00:00:00+05:30'],
+                       dtype='datetime64[ns, Asia/Calcutta]', freq=None)
+        """
+        new_values = normalize_i8_timestamps(self.asi8, self.tz, reso=self._creso)
+        dt64_values = new_values.view(self._ndarray.dtype)
+
+        dta = type(self)._simple_new(dt64_values, dtype=dt64_values.dtype)
+        dta = dta._with_freq("infer")
+        if self.tz is not None:
+            dta = dta.tz_localize(self.tz)
+        return dta
+
+    def to_period(self, freq=None) -> PeriodArray:
+        """
+        Cast to PeriodArray/PeriodIndex at a particular frequency.
+
+        Converts DatetimeArray/Index to PeriodArray/PeriodIndex.
+
+        Parameters
+        ----------
+        freq : str or Period, optional
+            One of pandas' :ref:`period aliases <timeseries.period_aliases>`
+            or an Period object. Will be inferred by default.
+
+        Returns
+        -------
+        PeriodArray/PeriodIndex
+            Immutable ndarray holding ordinal values at a particular frequency.
+
+        Raises
+        ------
+        ValueError
+            When converting a DatetimeArray/Index with non-regular values,
+            so that a frequency cannot be inferred.
+
+        See Also
+        --------
+        PeriodIndex: Immutable ndarray holding ordinal values.
+        DatetimeIndex.to_pydatetime: Return DatetimeIndex as object.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"y": [1, 2, 3]},
+        ...     index=pd.to_datetime(
+        ...         [
+        ...             "2000-03-31 00:00:00",
+        ...             "2000-05-31 00:00:00",
+        ...             "2000-08-31 00:00:00",
+        ...         ]
+        ...     ),
+        ... )
+        >>> df.index.to_period("M")
+        PeriodIndex(['2000-03', '2000-05', '2000-08'],
+                    dtype='period[M]')
+
+        Infer the daily frequency
+
+        >>> idx = pd.date_range("2017-01-01", periods=2)
+        >>> idx.to_period()
+        PeriodIndex(['2017-01-01', '2017-01-02'],
+                    dtype='period[D]')
+        """
+        from pandas.core.arrays import PeriodArray
+
+        if self.tz is not None:
+            warnings.warn(
+                "Converting to PeriodArray/Index representation "
+                "will drop timezone information.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        if freq is None:
+            freq = self.freqstr or self.inferred_freq
+            if isinstance(self.freq, BaseOffset) and hasattr(
+                self.freq, "_period_dtype_code"
+            ):
+                freq = PeriodDtype(self.freq)._freqstr
+
+            if freq is None:
+                raise ValueError(
+                    "You must pass a freq argument as current index has none."
+                )
+
+            res = get_period_alias(freq)
+
+            #  https://github.com/pandas-dev/pandas/issues/33358
+            if res is None:
+                res = freq
+
+            freq = res
+        return PeriodArray._from_datetime64(self._ndarray, freq, tz=self.tz)
+
+    # -----------------------------------------------------------------
+    # Properties - Vectorized Timestamp Properties/Methods
+
+    def month_name(self, locale=None) -> npt.NDArray[np.object_]:
+        """
+        Return the month names with specified locale.
+
+        Parameters
+        ----------
+        locale : str, optional
+            Locale determining the language in which to return the month name.
+            Default is English locale (``'en_US.utf8'``). Use the command
+            ``locale -a`` on your terminal on Unix systems to find your locale
+            language code.
+
+        Returns
+        -------
+        Series or Index
+            Series or Index of month names.
+
+        See Also
+        --------
+        DatetimeIndex.day_name : Return the day names with specified locale.
+
+        Examples
+        --------
+        >>> s = pd.Series(pd.date_range(start="2018-01", freq="ME", periods=3))
+        >>> s
+        0   2018-01-31
+        1   2018-02-28
+        2   2018-03-31
+        dtype: datetime64[ns]
+        >>> s.dt.month_name()
+        0     January
+        1    February
+        2       March
+        dtype: object
+
+        >>> idx = pd.date_range(start="2018-01", freq="ME", periods=3)
+        >>> idx
+        DatetimeIndex(['2018-01-31', '2018-02-28', '2018-03-31'],
+                      dtype='datetime64[ns]', freq='ME')
+        >>> idx.month_name()
+        Index(['January', 'February', 'March'], dtype='object')
+
+        Using the ``locale`` parameter you can set a different locale language,
+        for example: ``idx.month_name(locale='pt_BR.utf8')`` will return month
+        names in Brazilian Portuguese language.
+
+        >>> idx = pd.date_range(start="2018-01", freq="ME", periods=3)
+        >>> idx
+        DatetimeIndex(['2018-01-31', '2018-02-28', '2018-03-31'],
+                      dtype='datetime64[ns]', freq='ME')
+        >>> idx.month_name(locale="pt_BR.utf8")  # doctest: +SKIP
+        Index(['Janeiro', 'Fevereiro', 'Março'], dtype='object')
+        """
+        values = self._local_timestamps()
+
+        result = fields.get_date_name_field(
+            values, "month_name", locale=locale, reso=self._creso
+        )
+        result = self._maybe_mask_results(result, fill_value=None)
+        if using_string_dtype():
+            from pandas import (
+                StringDtype,
+                array as pd_array,
+            )
+
+            return pd_array(result, dtype=StringDtype(na_value=np.nan))  # type: ignore[return-value]
+        return result
+
+    def day_name(self, locale=None) -> npt.NDArray[np.object_]:
+        """
+        Return the day names with specified locale.
+
+        Parameters
+        ----------
+        locale : str, optional
+            Locale determining the language in which to return the day name.
+            Default is English locale (``'en_US.utf8'``). Use the command
+            ``locale -a`` on your terminal on Unix systems to find your locale
+            language code.
+
+        Returns
+        -------
+        Series or Index
+            Series or Index of day names.
+
+        See Also
+        --------
+        DatetimeIndex.month_name : Return the month names with specified locale.
+
+        Examples
+        --------
+        >>> s = pd.Series(pd.date_range(start="2018-01-01", freq="D", periods=3))
+        >>> s
+        0   2018-01-01
+        1   2018-01-02
+        2   2018-01-03
+        dtype: datetime64[ns]
+        >>> s.dt.day_name()
+        0       Monday
+        1      Tuesday
+        2    Wednesday
+        dtype: object
+
+        >>> idx = pd.date_range(start="2018-01-01", freq="D", periods=3)
+        >>> idx
+        DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03'],
+                      dtype='datetime64[ns]', freq='D')
+        >>> idx.day_name()
+        Index(['Monday', 'Tuesday', 'Wednesday'], dtype='object')
+
+        Using the ``locale`` parameter you can set a different locale language,
+        for example: ``idx.day_name(locale='pt_BR.utf8')`` will return day
+        names in Brazilian Portuguese language.
+
+        >>> idx = pd.date_range(start="2018-01-01", freq="D", periods=3)
+        >>> idx
+        DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03'],
+                      dtype='datetime64[ns]', freq='D')
+        >>> idx.day_name(locale="pt_BR.utf8")  # doctest: +SKIP
+        Index(['Segunda', 'Terça', 'Quarta'], dtype='object')
+        """
+        values = self._local_timestamps()
+
+        result = fields.get_date_name_field(
+            values, "day_name", locale=locale, reso=self._creso
+        )
+        result = self._maybe_mask_results(result, fill_value=None)
+        if using_string_dtype():
+            # TODO: no tests that check for dtype of result as of 2024-08-15
+            from pandas import (
+                StringDtype,
+                array as pd_array,
+            )
+
+            return pd_array(result, dtype=StringDtype(na_value=np.nan))  # type: ignore[return-value]
+        return result
+
+    @property
+    def time(self) -> npt.NDArray[np.object_]:
+        """
+        Returns numpy array of :class:`datetime.time` objects.
+
+        The time part of the Timestamps.
+
+        See Also
+        --------
+        DatetimeIndex.timetz : Returns numpy array of :class:`datetime.time`
+            objects with timezones. The time part of the Timestamps.
+        DatetimeIndex.date : Returns numpy array of python :class:`datetime.date`
+            objects. Namely, the date part of Timestamps without time and timezone
+            information.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.time
+        0    10:00:00
+        1    11:00:00
+        dtype: object
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(
+        ...     ["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"]
+        ... )
+        >>> idx.time
+        array([datetime.time(10, 0), datetime.time(11, 0)], dtype=object)
+        """
+        # If the Timestamps have a timezone that is not UTC,
+        # convert them into their i8 representation while
+        # keeping their timezone and not using UTC
+        timestamps = self._local_timestamps()
+
+        return ints_to_pydatetime(timestamps, box="time", reso=self._creso)
+
+    @property
+    def timetz(self) -> npt.NDArray[np.object_]:
+        """
+        Returns numpy array of :class:`datetime.time` objects with timezones.
+
+        The time part of the Timestamps.
+
+        See Also
+        --------
+        DatetimeIndex.time : Returns numpy array of :class:`datetime.time` objects.
+            The time part of the Timestamps.
+        DatetimeIndex.tz : Return the timezone.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.timetz
+        0    10:00:00+00:00
+        1    11:00:00+00:00
+        dtype: object
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(
+        ...     ["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"]
+        ... )
+        >>> idx.timetz
+        array([datetime.time(10, 0, tzinfo=datetime.timezone.utc),
+        datetime.time(11, 0, tzinfo=datetime.timezone.utc)], dtype=object)
+        """
+        return ints_to_pydatetime(self.asi8, self.tz, box="time", reso=self._creso)
+
+    @property
+    def date(self) -> npt.NDArray[np.object_]:
+        """
+        Returns numpy array of python :class:`datetime.date` objects.
+
+        Namely, the date part of Timestamps without time and
+        timezone information.
+
+        See Also
+        --------
+        DatetimeIndex.time : Returns numpy array of :class:`datetime.time` objects.
+            The time part of the Timestamps.
+        DatetimeIndex.year : The year of the datetime.
+        DatetimeIndex.month : The month as January=1, December=12.
+        DatetimeIndex.day : The day of the datetime.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.date
+        0    2020-01-01
+        1    2020-02-01
+        dtype: object
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(
+        ...     ["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"]
+        ... )
+        >>> idx.date
+        array([datetime.date(2020, 1, 1), datetime.date(2020, 2, 1)], dtype=object)
+        """
+        # If the Timestamps have a timezone that is not UTC,
+        # convert them into their i8 representation while
+        # keeping their timezone and not using UTC
+        timestamps = self._local_timestamps()
+
+        return ints_to_pydatetime(timestamps, box="date", reso=self._creso)
+
+    def isocalendar(self) -> DataFrame:
+        """
+        Calculate year, week, and day according to the ISO 8601 standard.
+
+        Returns
+        -------
+        DataFrame
+            With columns year, week and day.
+
+        See Also
+        --------
+        Timestamp.isocalendar : Function return a 3-tuple containing ISO year,
+            week number, and weekday for the given Timestamp object.
+        datetime.date.isocalendar : Return a named tuple object with
+            three components: year, week and weekday.
+
+        Examples
+        --------
+        >>> idx = pd.date_range(start="2019-12-29", freq="D", periods=4)
+        >>> idx.isocalendar()
+                    year  week  day
+        2019-12-29  2019    52    7
+        2019-12-30  2020     1    1
+        2019-12-31  2020     1    2
+        2020-01-01  2020     1    3
+        >>> idx.isocalendar().week
+        2019-12-29    52
+        2019-12-30     1
+        2019-12-31     1
+        2020-01-01     1
+        Freq: D, Name: week, dtype: UInt32
+        """
+        from pandas import DataFrame
+
+        values = self._local_timestamps()
+        sarray = fields.build_isocalendar_sarray(values, reso=self._creso)
+        iso_calendar_df = DataFrame(
+            sarray, columns=["year", "week", "day"], dtype="UInt32"
+        )
+        if self._hasna:
+            iso_calendar_df.iloc[self._isnan] = None
+        return iso_calendar_df
+
+    year = _field_accessor(
+        "year",
+        "Y",
+        """
+        The year of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.month: The month as January=1, December=12.
+        DatetimeIndex.day: The day of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="YE")
+        ... )
+        >>> datetime_series
+        0   2000-12-31
+        1   2001-12-31
+        2   2002-12-31
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.year
+        0    2000
+        1    2001
+        2    2002
+        dtype: int32
+        """,
+    )
+    month = _field_accessor(
+        "month",
+        "M",
+        """
+        The month as January=1, December=12.
+
+        See Also
+        --------
+        DatetimeIndex.year: The year of the datetime.
+        DatetimeIndex.day: The day of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="ME")
+        ... )
+        >>> datetime_series
+        0   2000-01-31
+        1   2000-02-29
+        2   2000-03-31
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.month
+        0    1
+        1    2
+        2    3
+        dtype: int32
+        """,
+    )
+    day = _field_accessor(
+        "day",
+        "D",
+        """
+        The day of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.year: The year of the datetime.
+        DatetimeIndex.month: The month as January=1, December=12.
+        DatetimeIndex.hour: The hours of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="D")
+        ... )
+        >>> datetime_series
+        0   2000-01-01
+        1   2000-01-02
+        2   2000-01-03
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.day
+        0    1
+        1    2
+        2    3
+        dtype: int32
+        """,
+    )
+    hour = _field_accessor(
+        "hour",
+        "h",
+        """
+        The hours of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.day: The day of the datetime.
+        DatetimeIndex.minute: The minutes of the datetime.
+        DatetimeIndex.second: The seconds of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="h")
+        ... )
+        >>> datetime_series
+        0   2000-01-01 00:00:00
+        1   2000-01-01 01:00:00
+        2   2000-01-01 02:00:00
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.hour
+        0    0
+        1    1
+        2    2
+        dtype: int32
+        """,
+    )
+    minute = _field_accessor(
+        "minute",
+        "m",
+        """
+        The minutes of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.hour: The hours of the datetime.
+        DatetimeIndex.second: The seconds of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="min")
+        ... )
+        >>> datetime_series
+        0   2000-01-01 00:00:00
+        1   2000-01-01 00:01:00
+        2   2000-01-01 00:02:00
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.minute
+        0    0
+        1    1
+        2    2
+        dtype: int32
+        """,
+    )
+    second = _field_accessor(
+        "second",
+        "s",
+        """
+        The seconds of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.minute: The minutes of the datetime.
+        DatetimeIndex.microsecond: The microseconds of the datetime.
+        DatetimeIndex.nanosecond: The nanoseconds of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="s")
+        ... )
+        >>> datetime_series
+        0   2000-01-01 00:00:00
+        1   2000-01-01 00:00:01
+        2   2000-01-01 00:00:02
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.second
+        0    0
+        1    1
+        2    2
+        dtype: int32
+        """,
+    )
+    microsecond = _field_accessor(
+        "microsecond",
+        "us",
+        """
+        The microseconds of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.second: The seconds of the datetime.
+        DatetimeIndex.nanosecond: The nanoseconds of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="us")
+        ... )
+        >>> datetime_series
+        0   2000-01-01 00:00:00.000000
+        1   2000-01-01 00:00:00.000001
+        2   2000-01-01 00:00:00.000002
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.microsecond
+        0       0
+        1       1
+        2       2
+        dtype: int32
+        """,
+    )
+    nanosecond = _field_accessor(
+        "nanosecond",
+        "ns",
+        """
+        The nanoseconds of the datetime.
+
+        See Also
+        --------
+        DatetimeIndex.second: The seconds of the datetime.
+        DatetimeIndex.microsecond: The microseconds of the datetime.
+
+        Examples
+        --------
+        >>> datetime_series = pd.Series(
+        ...     pd.date_range("2000-01-01", periods=3, freq="ns")
+        ... )
+        >>> datetime_series
+        0   2000-01-01 00:00:00.000000000
+        1   2000-01-01 00:00:00.000000001
+        2   2000-01-01 00:00:00.000000002
+        dtype: datetime64[ns]
+        >>> datetime_series.dt.nanosecond
+        0       0
+        1       1
+        2       2
+        dtype: int32
+        """,
+    )
+    _dayofweek_doc = """
+    The day of the week with Monday=0, Sunday=6.
+
+    Return the day of the week. It is assumed the week starts on
+    Monday, which is denoted by 0 and ends on Sunday which is denoted
+    by 6. This method is available on both Series with datetime
+    values (using the `dt` accessor) or DatetimeIndex.
+
+    Returns
+    -------
+    Series or Index
+        Containing integers indicating the day number.
+
+    See Also
+    --------
+    Series.dt.dayofweek : Alias.
+    Series.dt.weekday : Alias.
+    Series.dt.day_name : Returns the name of the day of the week.
+
+    Examples
+    --------
+    >>> s = pd.date_range('2016-12-31', '2017-01-08', freq='D').to_series()
+    >>> s.dt.dayofweek
+    2016-12-31    5
+    2017-01-01    6
+    2017-01-02    0
+    2017-01-03    1
+    2017-01-04    2
+    2017-01-05    3
+    2017-01-06    4
+    2017-01-07    5
+    2017-01-08    6
+    Freq: D, dtype: int32
+    """
+    day_of_week = _field_accessor("day_of_week", "dow", _dayofweek_doc)
+    dayofweek = day_of_week
+    weekday = day_of_week
+
+    day_of_year = _field_accessor(
+        "dayofyear",
+        "doy",
+        """
+        The ordinal day of the year.
+
+        See Also
+        --------
+        DatetimeIndex.dayofweek : The day of the week with Monday=0, Sunday=6.
+        DatetimeIndex.day : The day of the datetime.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.dayofyear
+        0    1
+        1   32
+        dtype: int32
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(["1/1/2020 10:00:00+00:00",
+        ...                         "2/1/2020 11:00:00+00:00"])
+        >>> idx.dayofyear
+        Index([1, 32], dtype='int32')
+        """,
+    )
+    dayofyear = day_of_year
+    quarter = _field_accessor(
+        "quarter",
+        "q",
+        """
+        The quarter of the date.
+
+        See Also
+        --------
+        DatetimeIndex.snap : Snap time stamps to nearest occurring frequency.
+        DatetimeIndex.time : Returns numpy array of datetime.time objects.
+            The time part of the Timestamps.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "4/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-04-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.quarter
+        0    1
+        1    2
+        dtype: int32
+
+        For DatetimeIndex:
+
+        >>> idx = pd.DatetimeIndex(["1/1/2020 10:00:00+00:00",
+        ...                         "2/1/2020 11:00:00+00:00"])
+        >>> idx.quarter
+        Index([1, 1], dtype='int32')
+        """,
+    )
+    days_in_month = _field_accessor(
+        "days_in_month",
+        "dim",
+        """
+        The number of days in the month.
+
+        See Also
+        --------
+        Series.dt.day : Return the day of the month.
+        Series.dt.is_month_end : Return a boolean indicating if the
+            date is the last day of the month.
+        Series.dt.is_month_start : Return a boolean indicating if the
+            date is the first day of the month.
+        Series.dt.month : Return the month as January=1 through December=12.
+
+        Examples
+        --------
+        >>> s = pd.Series(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+        >>> s = pd.to_datetime(s)
+        >>> s
+        0   2020-01-01 10:00:00+00:00
+        1   2020-02-01 11:00:00+00:00
+        dtype: datetime64[s, UTC]
+        >>> s.dt.daysinmonth
+        0    31
+        1    29
+        dtype: int32
+        """,
+    )
+    daysinmonth = days_in_month
+    _is_month_doc = """
+        Indicates whether the date is the {first_or_last} day of the month.
+
+        Returns
+        -------
+        Series or array
+            For Series, returns a Series with boolean values.
+            For DatetimeIndex, returns a boolean array.
+
+        See Also
+        --------
+        is_month_start : Return a boolean indicating whether the date
+            is the first day of the month.
+        is_month_end : Return a boolean indicating whether the date
+            is the last day of the month.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> s = pd.Series(pd.date_range("2018-02-27", periods=3))
+        >>> s
+        0   2018-02-27
+        1   2018-02-28
+        2   2018-03-01
+        dtype: datetime64[ns]
+        >>> s.dt.is_month_start
+        0    False
+        1    False
+        2    True
+        dtype: bool
+        >>> s.dt.is_month_end
+        0    False
+        1    True
+        2    False
+        dtype: bool
+
+        >>> idx = pd.date_range("2018-02-27", periods=3)
+        >>> idx.is_month_start
+        array([False, False, True])
+        >>> idx.is_month_end
+        array([False, True, False])
+    """
+    is_month_start = _field_accessor(
+        "is_month_start", "is_month_start", _is_month_doc.format(first_or_last="first")
+    )
+
+    is_month_end = _field_accessor(
+        "is_month_end", "is_month_end", _is_month_doc.format(first_or_last="last")
+    )
+
+    is_quarter_start = _field_accessor(
+        "is_quarter_start",
+        "is_quarter_start",
+        """
+        Indicator for whether the date is the first day of a quarter.
+
+        Returns
+        -------
+        is_quarter_start : Series or DatetimeIndex
+            The same type as the original data with boolean values. Series will
+            have the same name and index. DatetimeIndex will have the same
+            name.
+
+        See Also
+        --------
+        quarter : Return the quarter of the date.
+        is_quarter_end : Similar property for indicating the quarter end.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> df = pd.DataFrame({'dates': pd.date_range("2017-03-30",
+        ...                   periods=4)})
+        >>> df.assign(quarter=df.dates.dt.quarter,
+        ...           is_quarter_start=df.dates.dt.is_quarter_start)
+               dates  quarter  is_quarter_start
+        0 2017-03-30        1             False
+        1 2017-03-31        1             False
+        2 2017-04-01        2              True
+        3 2017-04-02        2             False
+
+        >>> idx = pd.date_range('2017-03-30', periods=4)
+        >>> idx
+        DatetimeIndex(['2017-03-30', '2017-03-31', '2017-04-01', '2017-04-02'],
+                      dtype='datetime64[ns]', freq='D')
+
+        >>> idx.is_quarter_start
+        array([False, False,  True, False])
+        """,
+    )
+    is_quarter_end = _field_accessor(
+        "is_quarter_end",
+        "is_quarter_end",
+        """
+        Indicator for whether the date is the last day of a quarter.
+
+        Returns
+        -------
+        is_quarter_end : Series or DatetimeIndex
+            The same type as the original data with boolean values. Series will
+            have the same name and index. DatetimeIndex will have the same
+            name.
+
+        See Also
+        --------
+        quarter : Return the quarter of the date.
+        is_quarter_start : Similar property indicating the quarter start.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> df = pd.DataFrame({'dates': pd.date_range("2017-03-30",
+        ...                    periods=4)})
+        >>> df.assign(quarter=df.dates.dt.quarter,
+        ...           is_quarter_end=df.dates.dt.is_quarter_end)
+               dates  quarter    is_quarter_end
+        0 2017-03-30        1             False
+        1 2017-03-31        1              True
+        2 2017-04-01        2             False
+        3 2017-04-02        2             False
+
+        >>> idx = pd.date_range('2017-03-30', periods=4)
+        >>> idx
+        DatetimeIndex(['2017-03-30', '2017-03-31', '2017-04-01', '2017-04-02'],
+                      dtype='datetime64[ns]', freq='D')
+
+        >>> idx.is_quarter_end
+        array([False,  True, False, False])
+        """,
+    )
+    is_year_start = _field_accessor(
+        "is_year_start",
+        "is_year_start",
+        """
+        Indicate whether the date is the first day of a year.
+
+        Returns
+        -------
+        Series or DatetimeIndex
+            The same type as the original data with boolean values. Series will
+            have the same name and index. DatetimeIndex will have the same
+            name.
+
+        See Also
+        --------
+        is_year_end : Similar property indicating the last day of the year.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> dates = pd.Series(pd.date_range("2017-12-30", periods=3))
+        >>> dates
+        0   2017-12-30
+        1   2017-12-31
+        2   2018-01-01
+        dtype: datetime64[ns]
+
+        >>> dates.dt.is_year_start
+        0    False
+        1    False
+        2    True
+        dtype: bool
+
+        >>> idx = pd.date_range("2017-12-30", periods=3)
+        >>> idx
+        DatetimeIndex(['2017-12-30', '2017-12-31', '2018-01-01'],
+                      dtype='datetime64[ns]', freq='D')
+
+        >>> idx.is_year_start
+        array([False, False,  True])
+
+        This method, when applied to Series with datetime values under
+        the ``.dt`` accessor, will lose information about Business offsets.
+
+        >>> dates = pd.Series(pd.date_range("2020-10-30", periods=4, freq="BYS"))
+        >>> dates
+        0   2021-01-01
+        1   2022-01-03
+        2   2023-01-02
+        3   2024-01-01
+        dtype: datetime64[ns]
+
+        >>> dates.dt.is_year_start
+        0    True
+        1    False
+        2    False
+        3    True
+        dtype: bool
+
+        >>> idx = pd.date_range("2020-10-30", periods=4, freq="BYS")
+        >>> idx
+        DatetimeIndex(['2021-01-01', '2022-01-03', '2023-01-02', '2024-01-01'],
+                      dtype='datetime64[ns]', freq='BYS-JAN')
+
+        >>> idx.is_year_start
+        array([ True,  True,  True,  True])
+        """,
+    )
+    is_year_end = _field_accessor(
+        "is_year_end",
+        "is_year_end",
+        """
+        Indicate whether the date is the last day of the year.
+
+        Returns
+        -------
+        Series or DatetimeIndex
+            The same type as the original data with boolean values. Series will
+            have the same name and index. DatetimeIndex will have the same
+            name.
+
+        See Also
+        --------
+        is_year_start : Similar property indicating the start of the year.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> dates = pd.Series(pd.date_range("2017-12-30", periods=3))
+        >>> dates
+        0   2017-12-30
+        1   2017-12-31
+        2   2018-01-01
+        dtype: datetime64[ns]
+
+        >>> dates.dt.is_year_end
+        0    False
+        1     True
+        2    False
+        dtype: bool
+
+        >>> idx = pd.date_range("2017-12-30", periods=3)
+        >>> idx
+        DatetimeIndex(['2017-12-30', '2017-12-31', '2018-01-01'],
+                      dtype='datetime64[ns]', freq='D')
+
+        >>> idx.is_year_end
+        array([False,  True, False])
+        """,
+    )
+    is_leap_year = _field_accessor(
+        "is_leap_year",
+        "is_leap_year",
+        """
+        Boolean indicator if the date belongs to a leap year.
+
+        A leap year is a year, which has 366 days (instead of 365) including
+        29th of February as an intercalary day.
+        Leap years are years which are multiples of four with the exception
+        of years divisible by 100 but not by 400.
+
+        Returns
+        -------
+        Series or ndarray
+             Booleans indicating if dates belong to a leap year.
+
+        See Also
+        --------
+        DatetimeIndex.is_year_end : Indicate whether the date is the
+            last day of the year.
+        DatetimeIndex.is_year_start : Indicate whether the date is the first
+            day of a year.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> idx = pd.date_range("2012-01-01", "2015-01-01", freq="YE")
+        >>> idx
+        DatetimeIndex(['2012-12-31', '2013-12-31', '2014-12-31'],
+                      dtype='datetime64[ns]', freq='YE-DEC')
+        >>> idx.is_leap_year
+        array([ True, False, False])
+
+        >>> dates_series = pd.Series(idx)
+        >>> dates_series
+        0   2012-12-31
+        1   2013-12-31
+        2   2014-12-31
+        dtype: datetime64[ns]
+        >>> dates_series.dt.is_leap_year
+        0     True
+        1    False
+        2    False
+        dtype: bool
+        """,
+    )
+
+    def to_julian_date(self) -> npt.NDArray[np.float64]:
+        """
+        Convert Datetime Array to float64 ndarray of Julian Dates.
+        0 Julian date is noon January 1, 4713 BC.
+        https://en.wikipedia.org/wiki/Julian_day
+        """
+
+        # http://mysite.verizon.net/aesir_research/date/jdalg2.htm
+        year = np.asarray(self.year)
+        month = np.asarray(self.month)
+        day = np.asarray(self.day)
+        testarr = month < 3
+        year[testarr] -= 1
+        month[testarr] += 12
+        return (
+            day
+            + np.fix((153 * month - 457) / 5)
+            + 365 * year
+            + np.floor(year / 4)
+            - np.floor(year / 100)
+            + np.floor(year / 400)
+            + 1_721_118.5
+            + (
+                self.hour
+                + self.minute / 60
+                + self.second / 3600
+                + self.microsecond / 3600 / 10**6
+                + self.nanosecond / 3600 / 10**9
+            )
+            / 24
+        )
+
+    # -----------------------------------------------------------------
+    # Reductions
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        result = super()._reduce(name, skipna=skipna, keepdims=keepdims, **kwargs)
+        if keepdims and isinstance(result, np.ndarray):
+            if name == "std":
+                from pandas.core.arrays import TimedeltaArray
+
+                return TimedeltaArray._from_sequence(result)
+            else:
+                return self._from_sequence(result, dtype=self.dtype)
+        return result
+
+    def std(
+        self,
+        axis=None,
+        dtype=None,
+        out=None,
+        ddof: int = 1,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ) -> Timedelta:
+        """
+        Return sample standard deviation over requested axis.
+
+        Normalized by `N-1` by default. This can be changed using ``ddof``.
+
+        Parameters
+        ----------
+        axis : int, optional
+            Axis for the function to be applied on. For :class:`pandas.Series`
+            this parameter is unused and defaults to ``None``.
+        dtype : dtype, optional, default None
+            Type to use in computing the standard deviation. For arrays of
+            integer type the default is float64, for arrays of float types
+            it is the same as the array type.
+        out : ndarray, optional, default None
+            Alternative output array in which to place the result. It must have
+            the same shape as the expected output but the type (of the
+            calculated values) will be cast if necessary.
+        ddof : int, default 1
+            Degrees of Freedom. The divisor used in calculations is `N - ddof`,
+            where `N` represents the number of elements.
+        keepdims : bool, optional
+            If this is set to True, the axes which are reduced are left in the
+            result as dimensions with size one. With this option, the result
+            will broadcast correctly against the input array. If the default
+            value is passed, then keepdims will not be passed through to the
+            std method of sub-classes of ndarray, however any non-default value
+            will be. If the sub-class method does not implement keepdims any
+            exceptions will be raised.
+        skipna : bool, default True
+            Exclude NA/null values. If an entire row/column is ``NA``, the result
+            will be ``NA``.
+
+        Returns
+        -------
+        Timedelta
+            Standard deviation over requested axis.
+
+        See Also
+        --------
+        numpy.ndarray.std : Returns the standard deviation of the array elements
+            along given axis.
+        Series.std : Return sample standard deviation over requested axis.
+
+        Examples
+        --------
+        For :class:`pandas.DatetimeIndex`:
+
+        >>> idx = pd.date_range("2001-01-01 00:00", periods=3)
+        >>> idx
+        DatetimeIndex(['2001-01-01', '2001-01-02', '2001-01-03'],
+                      dtype='datetime64[ns]', freq='D')
+        >>> idx.std()
+        Timedelta('1 days 00:00:00')
+        """
+        # Because std is translation-invariant, we can get self.std
+        #  by calculating (self - Timestamp(0)).std, and we can do it
+        #  without creating a copy by using a view on self._ndarray
+        from pandas.core.arrays import TimedeltaArray
+
+        # Find the td64 dtype with the same resolution as our dt64 dtype
+        dtype_str = self._ndarray.dtype.name.replace("datetime64", "timedelta64")
+        dtype = np.dtype(dtype_str)
+
+        tda = TimedeltaArray._simple_new(self._ndarray.view(dtype), dtype=dtype)
+
+        return tda.std(axis=axis, out=out, ddof=ddof, keepdims=keepdims, skipna=skipna)
+
+
+# -------------------------------------------------------------------
+# Constructor Helpers
+
+
+def _sequence_to_dt64(
+    data: ArrayLike,
+    *,
+    copy: bool = False,
+    tz: tzinfo | None = None,
+    dayfirst: bool = False,
+    yearfirst: bool = False,
+    ambiguous: TimeAmbiguous = "raise",
+    out_unit: str | None = None,
+) -> tuple[np.ndarray, tzinfo | None]:
+    """
+    Parameters
+    ----------
+    data : np.ndarray or ExtensionArray
+        dtl.ensure_arraylike_for_datetimelike has already been called.
+    copy : bool, default False
+    tz : tzinfo or None, default None
+    dayfirst : bool, default False
+    yearfirst : bool, default False
+    ambiguous : str, bool, or arraylike, default 'raise'
+        See pandas._libs.tslibs.tzconversion.tz_localize_to_utc.
+    out_unit : str or None, default None
+        Desired output resolution.
+
+    Returns
+    -------
+    result : numpy.ndarray
+        The sequence converted to a numpy array with dtype ``datetime64[unit]``.
+        Where `unit` is "ns" unless specified otherwise by `out_unit`.
+    tz : tzinfo or None
+        Either the user-provided tzinfo or one inferred from the data.
+
+    Raises
+    ------
+    TypeError : PeriodDType data is passed
+    """
+
+    # By this point we are assured to have either a numpy array or Index
+    data, copy = maybe_convert_dtype(data, copy, tz=tz)
+    data_dtype = getattr(data, "dtype", None)
+
+    out_dtype = DT64NS_DTYPE
+    if out_unit is not None:
+        out_dtype = np.dtype(f"M8[{out_unit}]")
+
+    if data_dtype == object or is_string_dtype(data_dtype):
+        # TODO: We do not have tests specific to string-dtypes,
+        #  also complex or categorical or other extension
+        data = cast(np.ndarray, data)
+        copy = False
+        if lib.infer_dtype(data, skipna=False) == "integer":
+            # Much more performant than going through array_to_datetime
+            data = data.astype(np.int64)
+        elif tz is not None and ambiguous == "raise":
+            obj_data = np.asarray(data, dtype=object)
+            result = tslib.array_to_datetime_with_tz(
+                obj_data,
+                tz=tz,
+                dayfirst=dayfirst,
+                yearfirst=yearfirst,
+                creso=abbrev_to_npy_unit(out_unit),
+            )
+            return result, tz
+        else:
+            converted, inferred_tz = objects_to_datetime64(
+                data,
+                dayfirst=dayfirst,
+                yearfirst=yearfirst,
+                allow_object=False,
+                out_unit=out_unit,
+            )
+            copy = False
+            if tz and inferred_tz:
+                #  two timezones: convert to intended from base UTC repr
+                # GH#42505 by convention, these are _already_ UTC
+                result = converted
+
+            elif inferred_tz:
+                tz = inferred_tz
+                result = converted
+
+            else:
+                result, _ = _construct_from_dt64_naive(
+                    converted, tz=tz, copy=copy, ambiguous=ambiguous
+                )
+            return result, tz
+
+        data_dtype = data.dtype
+
+    # `data` may have originally been a Categorical[datetime64[ns, tz]],
+    # so we need to handle these types.
+    if isinstance(data_dtype, DatetimeTZDtype):
+        # DatetimeArray -> ndarray
+        data = cast(DatetimeArray, data)
+        tz = _maybe_infer_tz(tz, data.tz)
+        result = data._ndarray
+
+    elif lib.is_np_dtype(data_dtype, "M"):
+        # tz-naive DatetimeArray or ndarray[datetime64]
+        if isinstance(data, DatetimeArray):
+            data = data._ndarray
+
+        data = cast(np.ndarray, data)
+        result, copy = _construct_from_dt64_naive(
+            data, tz=tz, copy=copy, ambiguous=ambiguous
+        )
+
+    else:
+        # must be integer dtype otherwise
+        # assume this data are epoch timestamps
+        if data.dtype != INT64_DTYPE:
+            data = data.astype(np.int64, copy=False)
+            copy = False
+        data = cast(np.ndarray, data)
+        result = data.view(out_dtype)
+
+    if copy:
+        result = result.copy()
+
+    assert isinstance(result, np.ndarray), type(result)
+    assert result.dtype.kind == "M"
+    assert result.dtype != "M8"
+    assert is_supported_dtype(result.dtype)
+    return result, tz
+
+
+def _construct_from_dt64_naive(
+    data: np.ndarray, *, tz: tzinfo | None, copy: bool, ambiguous: TimeAmbiguous
+) -> tuple[np.ndarray, bool]:
+    """
+    Convert datetime64 data to a supported dtype, localizing if necessary.
+    """
+    # Caller is responsible for ensuring
+    #  lib.is_np_dtype(data.dtype)
+
+    new_dtype = data.dtype
+    if not is_supported_dtype(new_dtype):
+        # Cast to the nearest supported unit, generally "s"
+        new_dtype = get_supported_dtype(new_dtype)
+        data = astype_overflowsafe(data, dtype=new_dtype, copy=False)
+        copy = False
+
+    if data.dtype.byteorder == ">":
+        # TODO: better way to handle this?  non-copying alternative?
+        #  without this, test_constructor_datetime64_bigendian fails
+        data = data.astype(data.dtype.newbyteorder("<"))
+        new_dtype = data.dtype
+        copy = False
+
+    if tz is not None:
+        # Convert tz-naive to UTC
+        # TODO: if tz is UTC, are there situations where we *don't* want a
+        #  copy?  tz_localize_to_utc always makes one.
+        shape = data.shape
+        if data.ndim > 1:
+            data = data.ravel()
+
+        data_unit = get_unit_from_dtype(new_dtype)
+        data = tzconversion.tz_localize_to_utc(
+            data.view("i8"), tz, ambiguous=ambiguous, creso=data_unit
+        )
+        data = data.view(new_dtype)
+        data = data.reshape(shape)
+
+    assert data.dtype == new_dtype, data.dtype
+    result = data
+
+    return result, copy
+
+
+def objects_to_datetime64(
+    data: np.ndarray,
+    dayfirst,
+    yearfirst,
+    utc: bool = False,
+    errors: DateTimeErrorChoices = "raise",
+    allow_object: bool = False,
+    out_unit: str | None = None,
+) -> tuple[np.ndarray, tzinfo | None]:
+    """
+    Convert data to array of timestamps.
+
+    Parameters
+    ----------
+    data : np.ndarray[object]
+    dayfirst : bool
+    yearfirst : bool
+    utc : bool, default False
+        Whether to convert/localize timestamps to UTC.
+    errors : {'raise', 'coerce'}
+    allow_object : bool
+        Whether to return an object-dtype ndarray instead of raising if the
+        data contains more than one timezone.
+    out_unit : str or None, default None
+        None indicates we should do resolution inference.
+
+    Returns
+    -------
+    result : ndarray
+        np.datetime64[out_unit] if returned values represent wall times or UTC
+        timestamps.
+        object if mixed timezones
+    inferred_tz : tzinfo or None
+        If not None, then the datetime64 values in `result` denote UTC timestamps.
+
+    Raises
+    ------
+    ValueError : if data cannot be converted to datetimes
+    TypeError  : When a type cannot be converted to datetime
+    """
+    assert errors in ["raise", "coerce"]
+
+    # if str-dtype, convert
+    data = np.asarray(data, dtype=np.object_)
+
+    result, tz_parsed = tslib.array_to_datetime(
+        data,
+        errors=errors,
+        utc=utc,
+        dayfirst=dayfirst,
+        yearfirst=yearfirst,
+        creso=abbrev_to_npy_unit(out_unit),
+    )
+
+    if tz_parsed is not None:
+        # We can take a shortcut since the datetime64 numpy array
+        #  is in UTC
+        return result, tz_parsed
+    elif result.dtype.kind == "M":
+        return result, tz_parsed
+    elif result.dtype == object:
+        # GH#23675 when called via `pd.to_datetime`, returning an object-dtype
+        #  array is allowed.  When called via `pd.DatetimeIndex`, we can
+        #  only accept datetime64 dtype, so raise TypeError if object-dtype
+        #  is returned, as that indicates the values can be recognized as
+        #  datetimes but they have conflicting timezones/awareness
+        if allow_object:
+            return result, tz_parsed
+        raise TypeError("DatetimeIndex has mixed timezones")
+    else:  # pragma: no cover
+        # GH#23675 this TypeError should never be hit, whereas the TypeError
+        #  in the object-dtype branch above is reachable.
+        raise TypeError(result)
+
+
+def maybe_convert_dtype(data, copy: bool, tz: tzinfo | None = None):
+    """
+    Convert data based on dtype conventions, issuing
+    errors where appropriate.
+
+    Parameters
+    ----------
+    data : np.ndarray or pd.Index
+    copy : bool
+    tz : tzinfo or None, default None
+
+    Returns
+    -------
+    data : np.ndarray or pd.Index
+    copy : bool
+
+    Raises
+    ------
+    TypeError : PeriodDType data is passed
+    """
+    if not hasattr(data, "dtype"):
+        # e.g. collections.deque
+        return data, copy
+
+    if is_float_dtype(data.dtype):
+        # pre-2.0 we treated these as wall-times, inconsistent with ints
+        # GH#23675, GH#45573 deprecated to treat symmetrically with integer dtypes.
+        # Note: data.astype(np.int64) fails ARM tests, see
+        # https://github.com/pandas-dev/pandas/issues/49468.
+        data = data.astype(DT64NS_DTYPE).view("i8")
+        copy = False
+
+    elif lib.is_np_dtype(data.dtype, "m") or is_bool_dtype(data.dtype):
+        # GH#29794 enforcing deprecation introduced in GH#23539
+        raise TypeError(f"dtype {data.dtype} cannot be converted to datetime64[ns]")
+    elif isinstance(data.dtype, PeriodDtype):
+        # Note: without explicitly raising here, PeriodIndex
+        #  test_setops.test_join_does_not_recur fails
+        raise TypeError(
+            "Passing PeriodDtype data is invalid. Use `data.to_timestamp()` instead"
+        )
+
+    elif isinstance(data.dtype, ExtensionDtype) and not isinstance(
+        data.dtype, DatetimeTZDtype
+    ):
+        # TODO: We have no tests for these
+        data = np.array(data, dtype=np.object_)
+        copy = False
+
+    return data, copy
+
+
+# -------------------------------------------------------------------
+# Validation and Inference
+
+
+def _maybe_infer_tz(tz: tzinfo | None, inferred_tz: tzinfo | None) -> tzinfo | None:
+    """
+    If a timezone is inferred from data, check that it is compatible with
+    the user-provided timezone, if any.
+
+    Parameters
+    ----------
+    tz : tzinfo or None
+    inferred_tz : tzinfo or None
+
+    Returns
+    -------
+    tz : tzinfo or None
+
+    Raises
+    ------
+    TypeError : if both timezones are present but do not match
+    """
+    if tz is None:
+        tz = inferred_tz
+    elif inferred_tz is None:
+        pass
+    elif not timezones.tz_compare(tz, inferred_tz):
+        raise TypeError(
+            f"data is already tz-aware {inferred_tz}, unable to set specified tz: {tz}"
+        )
+    return tz
+
+
+def _validate_dt64_dtype(dtype):
+    """
+    Check that a dtype, if passed, represents either a numpy datetime64[ns]
+    dtype or a pandas DatetimeTZDtype.
+
+    Parameters
+    ----------
+    dtype : object
+
+    Returns
+    -------
+    dtype : None, numpy.dtype, or DatetimeTZDtype
+
+    Raises
+    ------
+    ValueError : invalid dtype
+
+    Notes
+    -----
+    Unlike _validate_tz_from_dtype, this does _not_ allow non-existent
+    tz errors to go through
+    """
+    if dtype is not None:
+        dtype = pandas_dtype(dtype)
+        if dtype == np.dtype("M8"):
+            # no precision, disallowed GH#24806
+            msg = (
+                "Passing in 'datetime64' dtype with no precision is not allowed. "
+                "Please pass in 'datetime64[ns]' instead."
+            )
+            raise ValueError(msg)
+
+        if (
+            isinstance(dtype, np.dtype)
+            and (dtype.kind != "M" or not is_supported_dtype(dtype))
+        ) or not isinstance(dtype, (np.dtype, DatetimeTZDtype)):
+            raise ValueError(
+                f"Unexpected value for 'dtype': '{dtype}'. "
+                "Must be 'datetime64[s]', 'datetime64[ms]', 'datetime64[us]', "
+                "'datetime64[ns]' or DatetimeTZDtype'."
+            )
+
+        if getattr(dtype, "tz", None):
+            # https://github.com/pandas-dev/pandas/issues/18595
+            # Ensure that we have a standard timezone for pytz objects.
+            # Without this, things like adding an array of timedeltas and
+            # a  tz-aware Timestamp (with a tz specific to its datetime) will
+            # be incorrect(ish?) for the array as a whole
+            dtype = cast(DatetimeTZDtype, dtype)
+            dtype = DatetimeTZDtype(
+                unit=dtype.unit, tz=timezones.tz_standardize(dtype.tz)
+            )
+
+    return dtype
+
+
+def _validate_tz_from_dtype(
+    dtype, tz: tzinfo | None, explicit_tz_none: bool = False
+) -> tzinfo | None:
+    """
+    If the given dtype is a DatetimeTZDtype, extract the implied
+    tzinfo object from it and check that it does not conflict with the given
+    tz.
+
+    Parameters
+    ----------
+    dtype : dtype, str
+    tz : None, tzinfo
+    explicit_tz_none : bool, default False
+        Whether tz=None was passed explicitly, as opposed to lib.no_default.
+
+    Returns
+    -------
+    tz : consensus tzinfo
+
+    Raises
+    ------
+    ValueError : on tzinfo mismatch
+    """
+    if dtype is not None:
+        if isinstance(dtype, str):
+            try:
+                dtype = DatetimeTZDtype.construct_from_string(dtype)
+            except TypeError:
+                # Things like `datetime64[ns]`, which is OK for the
+                # constructors, but also nonsense, which should be validated
+                # but not by us. We *do* allow non-existent tz errors to
+                # go through
+                pass
+        dtz = getattr(dtype, "tz", None)
+        if dtz is not None:
+            if tz is not None and not timezones.tz_compare(tz, dtz):
+                raise ValueError("cannot supply both a tz and a dtype with a tz")
+            if explicit_tz_none:
+                raise ValueError("Cannot pass both a timezone-aware dtype and tz=None")
+            tz = dtz
+
+        if tz is not None and lib.is_np_dtype(dtype, "M"):
+            # We also need to check for the case where the user passed a
+            #  tz-naive dtype (i.e. datetime64[ns])
+            if tz is not None and not timezones.tz_compare(tz, dtz):
+                raise ValueError(
+                    "cannot supply both a tz and a "
+                    "timezone-naive dtype (i.e. datetime64[ns])"
+                )
+
+    return tz
+
+
+def _infer_tz_from_endpoints(
+    start: Timestamp, end: Timestamp, tz: tzinfo | None
+) -> tzinfo | None:
+    """
+    If a timezone is not explicitly given via `tz`, see if one can
+    be inferred from the `start` and `end` endpoints.  If more than one
+    of these inputs provides a timezone, require that they all agree.
+
+    Parameters
+    ----------
+    start : Timestamp
+    end : Timestamp
+    tz : tzinfo or None
+
+    Returns
+    -------
+    tz : tzinfo or None
+
+    Raises
+    ------
+    TypeError : if start and end timezones do not agree
+    """
+    try:
+        inferred_tz = timezones.infer_tzinfo(start, end)
+    except AssertionError as err:
+        # infer_tzinfo raises AssertionError if passed mismatched timezones
+        raise TypeError(
+            "Start and end cannot both be tz-aware with different timezones"
+        ) from err
+
+    inferred_tz = timezones.maybe_get_tz(inferred_tz)
+    tz = timezones.maybe_get_tz(tz)
+
+    if tz is not None and inferred_tz is not None:
+        if not timezones.tz_compare(inferred_tz, tz):
+            raise AssertionError("Inferred time zone not equal to passed time zone")
+
+    elif inferred_tz is not None:
+        tz = inferred_tz
+
+    return tz
+
+
+def _maybe_normalize_endpoints(
+    start: _TimestampNoneT1, end: _TimestampNoneT2, normalize: bool
+) -> tuple[_TimestampNoneT1, _TimestampNoneT2]:
+    if normalize:
+        if start is not None:
+            start = start.normalize()
+
+        if end is not None:
+            end = end.normalize()
+
+    return start, end
+
+
+def _maybe_localize_point(
+    ts: Timestamp | None, freq, tz, ambiguous, nonexistent
+) -> Timestamp | None:
+    """
+    Localize a start or end Timestamp to the timezone of the corresponding
+    start or end Timestamp
+
+    Parameters
+    ----------
+    ts : start or end Timestamp to potentially localize
+    freq : Tick, DateOffset, or None
+    tz : str, timezone object or None
+    ambiguous: str, localization behavior for ambiguous times
+    nonexistent: str, localization behavior for nonexistent times
+
+    Returns
+    -------
+    ts : Timestamp
+    """
+    # Make sure start and end are timezone localized if:
+    # 1) freq = a Timedelta-like frequency (Tick)
+    # 2) freq = None i.e. generating a linspaced range
+    if ts is not None and ts.tzinfo is None:
+        # Note: We can't ambiguous='infer' a singular ambiguous time; however,
+        # we have historically defaulted ambiguous=False
+        ambiguous = ambiguous if ambiguous != "infer" else False
+        localize_args = {"ambiguous": ambiguous, "nonexistent": nonexistent, "tz": None}
+        if isinstance(freq, Tick) or freq is None:
+            localize_args["tz"] = tz
+        ts = ts.tz_localize(**localize_args)
+    return ts
+
+
+def _generate_range(
+    start: Timestamp | None,
+    end: Timestamp | None,
+    periods: int | None,
+    offset: BaseOffset,
+    *,
+    unit: str,
+) -> Generator[Timestamp]:
+    """
+    Generates a sequence of dates corresponding to the specified time
+    offset. Similar to dateutil.rrule except uses pandas DateOffset
+    objects to represent time increments.
+
+    Parameters
+    ----------
+    start : Timestamp or None
+    end : Timestamp or None
+    periods : int or None
+    offset : DateOffset
+    unit : str
+
+    Notes
+    -----
+    * This method is faster for generating weekdays than dateutil.rrule
+    * At least two of (start, end, periods) must be specified.
+    * If both start and end are specified, the returned dates will
+    satisfy start <= date <= end.
+
+    Returns
+    -------
+    dates : generator object
+    """
+    offset = to_offset(offset)
+
+    # Argument 1 to "Timestamp" has incompatible type "Optional[Timestamp]";
+    # expected "Union[integer[Any], float, str, date, datetime64]"
+    start = Timestamp(start)  # type: ignore[arg-type]
+    if start is not NaT:
+        start = start.as_unit(unit)
+    else:
+        start = None
+
+    # Argument 1 to "Timestamp" has incompatible type "Optional[Timestamp]";
+    # expected "Union[integer[Any], float, str, date, datetime64]"
+    end = Timestamp(end)  # type: ignore[arg-type]
+    if end is not NaT:
+        end = end.as_unit(unit)
+    else:
+        end = None
+
+    if start and not offset.is_on_offset(start):
+        # Incompatible types in assignment (expression has type "datetime",
+        # variable has type "Optional[Timestamp]")
+
+        # GH #56147 account for negative direction and range bounds
+        if offset.n >= 0:
+            start = offset.rollforward(start)  # type: ignore[assignment]
+        else:
+            start = offset.rollback(start)  # type: ignore[assignment]
+
+    # Unsupported operand types for < ("Timestamp" and "None")
+    if periods is None and end < start and offset.n >= 0:  # type: ignore[operator]
+        end = None
+        periods = 0
+
+    if end is None:
+        # error: No overload variant of "__radd__" of "BaseOffset" matches
+        # argument type "None"
+        end = start + (periods - 1) * offset  # type: ignore[operator]
+
+    if start is None:
+        # error: No overload variant of "__radd__" of "BaseOffset" matches
+        # argument type "None"
+        start = end - (periods - 1) * offset  # type: ignore[operator]
+
+    start = cast(Timestamp, start)
+    end = cast(Timestamp, end)
+
+    cur = start
+    if offset.n >= 0:
+        while cur <= end:
+            yield cur
+
+            if cur == end:
+                # GH#24252 avoid overflows by not performing the addition
+                # in offset.apply unless we have to
+                break
+
+            # faster than cur + offset
+            next_date = offset._apply(cur)
+            next_date = next_date.as_unit(unit)
+            if next_date <= cur:
+                raise ValueError(f"Offset {offset} did not increment date")
+            cur = next_date
+    else:
+        while cur >= end:
+            yield cur
+
+            if cur == end:
+                # GH#24252 avoid overflows by not performing the addition
+                # in offset.apply unless we have to
+                break
+
+            # faster than cur + offset
+            next_date = offset._apply(cur)
+            next_date = next_date.as_unit(unit)
+            if next_date >= cur:
+                raise ValueError(f"Offset {offset} did not decrement date")
+            cur = next_date
diff --git a/pandas/pandas/core/arrays/floating.py b/pandas/pandas/core/arrays/floating.py
new file mode 100644
index 0000000000000..67c23f4825a7f
--- /dev/null
+++ b/pandas/pandas/core/arrays/floating.py
@@ -0,0 +1,181 @@
+from __future__ import annotations
+
+from typing import ClassVar
+
+import numpy as np
+
+from pandas.core.dtypes.base import register_extension_dtype
+from pandas.core.dtypes.common import is_float_dtype
+
+from pandas.core.arrays.numeric import (
+    NumericArray,
+    NumericDtype,
+)
+
+
+class FloatingDtype(NumericDtype):
+    """
+    An ExtensionDtype to hold a single size of floating dtype.
+
+    These specific implementations are subclasses of the non-public
+    FloatingDtype. For example we have Float32Dtype to represent float32.
+
+    The attributes name & type are set when these subclasses are created.
+    """
+
+    # The value used to fill '_data' to avoid upcasting
+    _internal_fill_value = np.nan
+    _default_np_dtype = np.dtype(np.float64)
+    _checker = is_float_dtype
+
+    @classmethod
+    def construct_array_type(cls) -> type[FloatingArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return FloatingArray
+
+    @classmethod
+    def _get_dtype_mapping(cls) -> dict[np.dtype, FloatingDtype]:
+        return NUMPY_FLOAT_TO_DTYPE
+
+    @classmethod
+    def _safe_cast(cls, values: np.ndarray, dtype: np.dtype, copy: bool) -> np.ndarray:
+        """
+        Safely cast the values to the given dtype.
+
+        "safe" in this context means the casting is lossless.
+        """
+        # This is really only here for compatibility with IntegerDtype
+        # Here for compat with IntegerDtype
+        return values.astype(dtype, copy=copy)
+
+
+class FloatingArray(NumericArray):
+    """
+    Array of floating (optional missing) values.
+
+    .. warning::
+
+       FloatingArray is currently experimental, and its API or internal
+       implementation may change without warning. Especially the behaviour
+       regarding NaN (distinct from NA missing values) is subject to change.
+
+    We represent a FloatingArray with 2 numpy arrays:
+
+    - data: contains a numpy float array of the appropriate dtype
+    - mask: a boolean array holding a mask on the data, True is missing
+
+    To construct an FloatingArray from generic array-like input, use
+    :func:`pandas.array` with one of the float dtypes (see examples).
+
+    See :ref:`integer_na` for more.
+
+    Parameters
+    ----------
+    values : numpy.ndarray
+        A 1-d float-dtype array.
+    mask : numpy.ndarray
+        A 1-d boolean-dtype array indicating missing values.
+    copy : bool, default False
+        Whether to copy the `values` and `mask`.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    Returns
+    -------
+    FloatingArray
+
+    See Also
+    --------
+    array : Create an array.
+    Float32Dtype : Float32 dtype for FloatingArray.
+    Float64Dtype : Float64 dtype for FloatingArray.
+    Series : One-dimensional labeled array capable of holding data.
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+
+    Examples
+    --------
+    Create an FloatingArray with :func:`pandas.array`:
+
+    >>> pd.array([0.1, None, 0.3], dtype=pd.Float32Dtype())
+    <FloatingArray>
+    [0.1, <NA>, 0.3]
+    Length: 3, dtype: Float32
+
+    String aliases for the dtypes are also available. They are capitalized.
+
+    >>> pd.array([0.1, None, 0.3], dtype="Float32")
+    <FloatingArray>
+    [0.1, <NA>, 0.3]
+    Length: 3, dtype: Float32
+    """
+
+    _dtype_cls = FloatingDtype
+
+
+_dtype_docstring = """
+An ExtensionDtype for {dtype} data.
+
+This dtype uses ``pd.NA`` as missing value indicator.
+
+Attributes
+----------
+None
+
+Methods
+-------
+None
+
+See Also
+--------
+CategoricalDtype : Type for categorical data with the categories and orderedness.
+IntegerDtype : An ExtensionDtype to hold a single size & kind of integer dtype.
+StringDtype : An ExtensionDtype for string data.
+
+Examples
+--------
+For Float32Dtype:
+
+>>> ser = pd.Series([2.25, pd.NA], dtype=pd.Float32Dtype())
+>>> ser.dtype
+Float32Dtype()
+
+For Float64Dtype:
+
+>>> ser = pd.Series([2.25, pd.NA], dtype=pd.Float64Dtype())
+>>> ser.dtype
+Float64Dtype()
+"""
+
+# create the Dtype
+
+
+@register_extension_dtype
+class Float32Dtype(FloatingDtype):
+    type = np.float32
+    name: ClassVar[str] = "Float32"
+    __doc__ = _dtype_docstring.format(dtype="float32")
+
+
+@register_extension_dtype
+class Float64Dtype(FloatingDtype):
+    type = np.float64
+    name: ClassVar[str] = "Float64"
+    __doc__ = _dtype_docstring.format(dtype="float64")
+
+
+NUMPY_FLOAT_TO_DTYPE: dict[np.dtype, FloatingDtype] = {
+    np.dtype(np.float32): Float32Dtype(),
+    np.dtype(np.float64): Float64Dtype(),
+}
diff --git a/pandas/pandas/core/arrays/integer.py b/pandas/pandas/core/arrays/integer.py
new file mode 100644
index 0000000000000..afbadd754cdbc
--- /dev/null
+++ b/pandas/pandas/core/arrays/integer.py
@@ -0,0 +1,279 @@
+from __future__ import annotations
+
+from typing import ClassVar
+
+import numpy as np
+
+from pandas.core.dtypes.base import register_extension_dtype
+from pandas.core.dtypes.common import is_integer_dtype
+
+from pandas.core.arrays.numeric import (
+    NumericArray,
+    NumericDtype,
+)
+
+
+class IntegerDtype(NumericDtype):
+    """
+    An ExtensionDtype to hold a single size & kind of integer dtype.
+
+    These specific implementations are subclasses of the non-public
+    IntegerDtype. For example, we have Int8Dtype to represent signed int 8s.
+
+    The attributes name & type are set when these subclasses are created.
+    """
+
+    # The value used to fill '_data' to avoid upcasting
+    _internal_fill_value = 1
+    _default_np_dtype = np.dtype(np.int64)
+    _checker = is_integer_dtype
+
+    @classmethod
+    def construct_array_type(cls) -> type[IntegerArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return IntegerArray
+
+    @classmethod
+    def _get_dtype_mapping(cls) -> dict[np.dtype, IntegerDtype]:
+        return NUMPY_INT_TO_DTYPE
+
+    @classmethod
+    def _safe_cast(cls, values: np.ndarray, dtype: np.dtype, copy: bool) -> np.ndarray:
+        """
+        Safely cast the values to the given dtype.
+
+        "safe" in this context means the casting is lossless. e.g. if 'values'
+        has a floating dtype, each value must be an integer.
+        """
+        try:
+            return values.astype(dtype, casting="safe", copy=copy)
+        except TypeError as err:
+            casted = values.astype(dtype, copy=copy)
+            if (casted == values).all():
+                return casted
+
+            raise TypeError(
+                f"cannot safely cast non-equivalent {values.dtype} to {np.dtype(dtype)}"
+            ) from err
+
+
+class IntegerArray(NumericArray):
+    """
+    Array of integer (optional missing) values.
+
+    Uses :attr:`pandas.NA` as the missing value.
+
+    .. warning::
+
+       IntegerArray is currently experimental, and its API or internal
+       implementation may change without warning.
+
+    We represent an IntegerArray with 2 numpy arrays:
+
+    - data: contains a numpy integer array of the appropriate dtype
+    - mask: a boolean array holding a mask on the data, True is missing
+
+    To construct an IntegerArray from generic array-like input, use
+    :func:`pandas.array` with one of the integer dtypes (see examples).
+
+    See :ref:`integer_na` for more.
+
+    Parameters
+    ----------
+    values : numpy.ndarray
+        A 1-d integer-dtype array.
+    mask : numpy.ndarray
+        A 1-d boolean-dtype array indicating missing values.
+    copy : bool, default False
+        Whether to copy the `values` and `mask`.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    Returns
+    -------
+    IntegerArray
+
+    See Also
+    --------
+    array : Create an array using the appropriate dtype, including ``IntegerArray``.
+    Int32Dtype : An ExtensionDtype for int32 integer data.
+    UInt16Dtype : An ExtensionDtype for uint16 integer data.
+
+    Examples
+    --------
+    Create an IntegerArray with :func:`pandas.array`.
+
+    >>> int_array = pd.array([1, None, 3], dtype=pd.Int32Dtype())
+    >>> int_array
+    <IntegerArray>
+    [1, <NA>, 3]
+    Length: 3, dtype: Int32
+
+    String aliases for the dtypes are also available. They are capitalized.
+
+    >>> pd.array([1, None, 3], dtype="Int32")
+    <IntegerArray>
+    [1, <NA>, 3]
+    Length: 3, dtype: Int32
+
+    >>> pd.array([1, None, 3], dtype="UInt16")
+    <IntegerArray>
+    [1, <NA>, 3]
+    Length: 3, dtype: UInt16
+    """
+
+    _dtype_cls = IntegerDtype
+
+
+_dtype_docstring = """
+An ExtensionDtype for {dtype} integer data.
+
+Uses :attr:`pandas.NA` as its missing value, rather than :attr:`numpy.nan`.
+
+Attributes
+----------
+None
+
+Methods
+-------
+None
+
+See Also
+--------
+Int8Dtype : 8-bit nullable integer type.
+Int16Dtype : 16-bit nullable integer type.
+Int32Dtype : 32-bit nullable integer type.
+Int64Dtype : 64-bit nullable integer type.
+
+Examples
+--------
+For Int8Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.Int8Dtype())
+>>> ser.dtype
+Int8Dtype()
+
+For Int16Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.Int16Dtype())
+>>> ser.dtype
+Int16Dtype()
+
+For Int32Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.Int32Dtype())
+>>> ser.dtype
+Int32Dtype()
+
+For Int64Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.Int64Dtype())
+>>> ser.dtype
+Int64Dtype()
+
+For UInt8Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.UInt8Dtype())
+>>> ser.dtype
+UInt8Dtype()
+
+For UInt16Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.UInt16Dtype())
+>>> ser.dtype
+UInt16Dtype()
+
+For UInt32Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.UInt32Dtype())
+>>> ser.dtype
+UInt32Dtype()
+
+For UInt64Dtype:
+
+>>> ser = pd.Series([2, pd.NA], dtype=pd.UInt64Dtype())
+>>> ser.dtype
+UInt64Dtype()
+"""
+
+# create the Dtype
+
+
+@register_extension_dtype
+class Int8Dtype(IntegerDtype):
+    type = np.int8
+    name: ClassVar[str] = "Int8"
+    __doc__ = _dtype_docstring.format(dtype="int8")
+
+
+@register_extension_dtype
+class Int16Dtype(IntegerDtype):
+    type = np.int16
+    name: ClassVar[str] = "Int16"
+    __doc__ = _dtype_docstring.format(dtype="int16")
+
+
+@register_extension_dtype
+class Int32Dtype(IntegerDtype):
+    type = np.int32
+    name: ClassVar[str] = "Int32"
+    __doc__ = _dtype_docstring.format(dtype="int32")
+
+
+@register_extension_dtype
+class Int64Dtype(IntegerDtype):
+    type = np.int64
+    name: ClassVar[str] = "Int64"
+    __doc__ = _dtype_docstring.format(dtype="int64")
+
+
+@register_extension_dtype
+class UInt8Dtype(IntegerDtype):
+    type = np.uint8
+    name: ClassVar[str] = "UInt8"
+    __doc__ = _dtype_docstring.format(dtype="uint8")
+
+
+@register_extension_dtype
+class UInt16Dtype(IntegerDtype):
+    type = np.uint16
+    name: ClassVar[str] = "UInt16"
+    __doc__ = _dtype_docstring.format(dtype="uint16")
+
+
+@register_extension_dtype
+class UInt32Dtype(IntegerDtype):
+    type = np.uint32
+    name: ClassVar[str] = "UInt32"
+    __doc__ = _dtype_docstring.format(dtype="uint32")
+
+
+@register_extension_dtype
+class UInt64Dtype(IntegerDtype):
+    type = np.uint64
+    name: ClassVar[str] = "UInt64"
+    __doc__ = _dtype_docstring.format(dtype="uint64")
+
+
+NUMPY_INT_TO_DTYPE: dict[np.dtype, IntegerDtype] = {
+    np.dtype(np.int8): Int8Dtype(),
+    np.dtype(np.int16): Int16Dtype(),
+    np.dtype(np.int32): Int32Dtype(),
+    np.dtype(np.int64): Int64Dtype(),
+    np.dtype(np.uint8): UInt8Dtype(),
+    np.dtype(np.uint16): UInt16Dtype(),
+    np.dtype(np.uint32): UInt32Dtype(),
+    np.dtype(np.uint64): UInt64Dtype(),
+}
diff --git a/pandas/pandas/core/arrays/interval.py b/pandas/pandas/core/arrays/interval.py
new file mode 100644
index 0000000000000..0bf2089df5f85
--- /dev/null
+++ b/pandas/pandas/core/arrays/interval.py
@@ -0,0 +1,2037 @@
+from __future__ import annotations
+
+import operator
+from operator import (
+    le,
+    lt,
+)
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    Union,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.interval import (
+    VALID_CLOSED,
+    Interval,
+    IntervalMixin,
+    intervals_to_interval_bounds,
+)
+from pandas._libs.missing import NA
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    Dtype,
+    IntervalClosedType,
+    NpDtype,
+    PositionalIndexer,
+    ScalarIndexer,
+    Self,
+    SequenceIndexer,
+    SortKind,
+    TimeArrayLike,
+    npt,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import IntCastingNaNError
+from pandas.util._decorators import Appender
+
+from pandas.core.dtypes.cast import (
+    LossySetitemError,
+    maybe_upcast_numeric_to_64bit,
+)
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_integer_dtype,
+    is_list_like,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    needs_i8_conversion,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    IntervalDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCDatetimeIndex,
+    ABCIntervalIndex,
+    ABCPeriodIndex,
+)
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+    notna,
+)
+
+from pandas.core.algorithms import (
+    isin,
+    take,
+    unique,
+    value_counts_internal as value_counts,
+)
+from pandas.core.arrays import ArrowExtensionArray
+from pandas.core.arrays.base import (
+    ExtensionArray,
+    _extension_array_shared_docs,
+)
+from pandas.core.arrays.datetimes import DatetimeArray
+from pandas.core.arrays.timedeltas import TimedeltaArray
+import pandas.core.common as com
+from pandas.core.construction import (
+    array as pd_array,
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import check_array_indexer
+from pandas.core.ops import (
+    invalid_comparison,
+    unpack_zerodim_and_defer,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Iterator,
+        Sequence,
+    )
+
+    from pandas import (
+        Index,
+        Series,
+    )
+
+
+IntervalSide = Union[TimeArrayLike, np.ndarray]
+IntervalOrNA = Union[Interval, float]
+
+_interval_shared_docs: dict[str, str] = {}
+
+_shared_docs_kwargs = {
+    "klass": "IntervalArray",
+    "qualname": "arrays.IntervalArray",
+    "name": "",
+}
+
+
+_interval_shared_docs["class"] = """
+%(summary)s
+
+Parameters
+----------
+data : array-like (1-dimensional)
+    Array-like (ndarray, :class:`DateTimeArray`, :class:`TimeDeltaArray`) containing
+    Interval objects from which to build the %(klass)s.
+closed : {'left', 'right', 'both', 'neither'}, default 'right'
+    Whether the intervals are closed on the left-side, right-side, both or
+    neither.
+dtype : dtype or None, default None
+    If None, dtype will be inferred.
+copy : bool, default False
+    Copy the input data.
+%(name)s\
+verify_integrity : bool, default True
+    Verify that the %(klass)s is valid.
+
+Attributes
+----------
+left
+right
+closed
+mid
+length
+is_empty
+is_non_overlapping_monotonic
+%(extra_attributes)s\
+
+Methods
+-------
+from_arrays
+from_tuples
+from_breaks
+contains
+overlaps
+set_closed
+to_tuples
+%(extra_methods)s\
+
+See Also
+--------
+Index : The base pandas Index type.
+Interval : A bounded slice-like interval; the elements of an %(klass)s.
+interval_range : Function to create a fixed frequency IntervalIndex.
+cut : Bin values into discrete Intervals.
+qcut : Bin values into equal-sized Intervals based on rank or sample quantiles.
+
+Notes
+-----
+See the `user guide
+<https://pandas.pydata.org/pandas-docs/stable/user_guide/advanced.html#intervalindex>`__
+for more.
+
+%(examples)s\
+"""
+
+
+@Appender(
+    _interval_shared_docs["class"]
+    % {
+        "klass": "IntervalArray",
+        "summary": "Pandas array for interval data that are closed on the same side.",
+        "name": "",
+        "extra_attributes": "",
+        "extra_methods": "",
+        "examples": textwrap.dedent(
+            """\
+    Examples
+    --------
+    A new ``IntervalArray`` can be constructed directly from an array-like of
+    ``Interval`` objects:
+
+    >>> pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+    <IntervalArray>
+    [(0, 1], (1, 5]]
+    Length: 2, dtype: interval[int64, right]
+
+    It may also be constructed using one of the constructor
+    methods: :meth:`IntervalArray.from_arrays`,
+    :meth:`IntervalArray.from_breaks`, and :meth:`IntervalArray.from_tuples`.
+    """
+        ),
+    }
+)
+class IntervalArray(IntervalMixin, ExtensionArray):
+    can_hold_na = True
+    _na_value = _fill_value = np.nan
+
+    @property
+    def ndim(self) -> Literal[1]:
+        return 1
+
+    # To make mypy recognize the fields
+    _left: IntervalSide
+    _right: IntervalSide
+    _dtype: IntervalDtype
+
+    # ---------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data,
+        closed: IntervalClosedType | None = None,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        verify_integrity: bool = True,
+    ) -> Self:
+        data = extract_array(data, extract_numpy=True)
+
+        if isinstance(data, cls):
+            left: IntervalSide = data._left
+            right: IntervalSide = data._right
+            closed = closed or data.closed
+            dtype = IntervalDtype(left.dtype, closed=closed)
+        else:
+            # don't allow scalars
+            if is_scalar(data):
+                msg = (
+                    f"{cls.__name__}(...) must be called with a collection "
+                    f"of some kind, {data} was passed"
+                )
+                raise TypeError(msg)
+
+            # might need to convert empty or purely na data
+            data = _maybe_convert_platform_interval(data)
+            left, right, infer_closed = intervals_to_interval_bounds(
+                data, validate_closed=closed is None
+            )
+            if left.dtype == object:
+                left = lib.maybe_convert_objects(left)
+                right = lib.maybe_convert_objects(right)
+            closed = closed or infer_closed
+
+            left, right, dtype = cls._ensure_simple_new_inputs(
+                left,
+                right,
+                closed=closed,
+                copy=copy,
+                dtype=dtype,
+            )
+
+        if verify_integrity:
+            cls._validate(left, right, dtype=dtype)
+
+        return cls._simple_new(
+            left,
+            right,
+            dtype=dtype,
+        )
+
+    @classmethod
+    def _simple_new(
+        cls,
+        left: IntervalSide,
+        right: IntervalSide,
+        dtype: IntervalDtype,
+    ) -> Self:
+        result = IntervalMixin.__new__(cls)
+        result._left = left
+        result._right = right
+        result._dtype = dtype
+
+        return result
+
+    @classmethod
+    def _ensure_simple_new_inputs(
+        cls,
+        left,
+        right,
+        closed: IntervalClosedType | None = None,
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> tuple[IntervalSide, IntervalSide, IntervalDtype]:
+        """Ensure correctness of input parameters for cls._simple_new."""
+        from pandas.core.indexes.base import ensure_index
+
+        left = ensure_index(left, copy=copy)
+        left = maybe_upcast_numeric_to_64bit(left)
+
+        right = ensure_index(right, copy=copy)
+        right = maybe_upcast_numeric_to_64bit(right)
+
+        if closed is None and isinstance(dtype, IntervalDtype):
+            closed = dtype.closed
+
+        closed = closed or "right"
+
+        if dtype is not None:
+            # GH 19262: dtype must be an IntervalDtype to override inferred
+            dtype = pandas_dtype(dtype)
+            if isinstance(dtype, IntervalDtype):
+                if dtype.subtype is not None:
+                    left = left.astype(dtype.subtype)
+                    right = right.astype(dtype.subtype)
+            else:
+                msg = f"dtype must be an IntervalDtype, got {dtype}"
+                raise TypeError(msg)
+
+            if dtype.closed is None:
+                # possibly loading an old pickle
+                dtype = IntervalDtype(dtype.subtype, closed)
+            elif closed != dtype.closed:
+                raise ValueError("closed keyword does not match dtype.closed")
+
+        # coerce dtypes to match if needed
+        if is_float_dtype(left.dtype) and is_integer_dtype(right.dtype):
+            right = right.astype(left.dtype)
+        elif is_float_dtype(right.dtype) and is_integer_dtype(left.dtype):
+            left = left.astype(right.dtype)
+
+        if type(left) != type(right):
+            msg = (
+                f"must not have differing left [{type(left).__name__}] and "
+                f"right [{type(right).__name__}] types"
+            )
+            raise ValueError(msg)
+        if isinstance(left.dtype, CategoricalDtype) or is_string_dtype(left.dtype):
+            # GH 19016
+            msg = (
+                "category, object, and string subtypes are not supported "
+                "for IntervalArray"
+            )
+            raise TypeError(msg)
+        if isinstance(left, ABCPeriodIndex):
+            msg = "Period dtypes are not supported, use a PeriodIndex instead"
+            raise ValueError(msg)
+        if isinstance(left, ABCDatetimeIndex) and str(left.tz) != str(right.tz):
+            msg = (
+                "left and right must have the same time zone, got "
+                f"'{left.tz}' and '{right.tz}'"
+            )
+            raise ValueError(msg)
+        elif needs_i8_conversion(left.dtype) and left.unit != right.unit:
+            # e.g. m8[s] vs m8[ms], try to cast to a common dtype GH#55714
+            left_arr, right_arr = left._data._ensure_matching_resos(right._data)
+            left = ensure_index(left_arr)
+            right = ensure_index(right_arr)
+
+        # For dt64/td64 we want DatetimeArray/TimedeltaArray instead of ndarray
+        left = ensure_wrapped_if_datetimelike(left)
+        left = extract_array(left, extract_numpy=True)
+        right = ensure_wrapped_if_datetimelike(right)
+        right = extract_array(right, extract_numpy=True)
+
+        if isinstance(left, ArrowExtensionArray) or isinstance(
+            right, ArrowExtensionArray
+        ):
+            pass
+        else:
+            lbase = getattr(left, "_ndarray", left)
+            lbase = getattr(lbase, "_data", lbase).base
+            rbase = getattr(right, "_ndarray", right)
+            rbase = getattr(rbase, "_data", rbase).base
+            if lbase is not None and lbase is rbase:
+                # If these share data, then setitem could corrupt our IA
+                right = right.copy()
+
+        dtype = IntervalDtype(left.dtype, closed=closed)
+
+        return left, right, dtype
+
+    @classmethod
+    def _from_sequence(
+        cls,
+        scalars,
+        *,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+    ) -> Self:
+        return cls(scalars, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _from_factorized(cls, values: np.ndarray, original: IntervalArray) -> Self:
+        return cls._from_sequence(values, dtype=original.dtype)
+
+    _interval_shared_docs["from_breaks"] = textwrap.dedent(
+        """
+        Construct an %(klass)s from an array of splits.
+
+        Parameters
+        ----------
+        breaks : array-like (1-dimensional)
+            Left and right bounds for each interval.
+        closed : {'left', 'right', 'both', 'neither'}, default 'right'
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither.\
+        %(name)s
+        copy : bool, default False
+            Copy the data.
+        dtype : dtype or None, default None
+            If None, dtype will be inferred.
+
+        Returns
+        -------
+        %(klass)s
+
+        See Also
+        --------
+        interval_range : Function to create a fixed frequency IntervalIndex.
+        %(klass)s.from_arrays : Construct from a left and right array.
+        %(klass)s.from_tuples : Construct from a sequence of tuples.
+
+        %(examples)s\
+        """
+    )
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_breaks"]
+        % {
+            "klass": "IntervalArray",
+            "name": "",
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.arrays.IntervalArray.from_breaks([0, 1, 2, 3])
+        <IntervalArray>
+        [(0, 1], (1, 2], (2, 3]]
+        Length: 3, dtype: interval[int64, right]
+        """
+            ),
+        }
+    )
+    def from_breaks(
+        cls,
+        breaks,
+        closed: IntervalClosedType | None = "right",
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> Self:
+        breaks = _maybe_convert_platform_interval(breaks)
+
+        return cls.from_arrays(breaks[:-1], breaks[1:], closed, copy=copy, dtype=dtype)
+
+    _interval_shared_docs["from_arrays"] = textwrap.dedent(
+        """
+        Construct from two arrays defining the left and right bounds.
+
+        Parameters
+        ----------
+        left : array-like (1-dimensional)
+            Left bounds for each interval.
+        right : array-like (1-dimensional)
+            Right bounds for each interval.
+        closed : {'left', 'right', 'both', 'neither'}, default 'right'
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither.\
+        %(name)s
+        copy : bool, default False
+            Copy the data.
+        dtype : dtype, optional
+            If None, dtype will be inferred.
+
+        Returns
+        -------
+        %(klass)s
+
+        Raises
+        ------
+        ValueError
+            When a value is missing in only one of `left` or `right`.
+            When a value in `left` is greater than the corresponding value
+            in `right`.
+
+        See Also
+        --------
+        interval_range : Function to create a fixed frequency IntervalIndex.
+        %(klass)s.from_breaks : Construct an %(klass)s from an array of
+            splits.
+        %(klass)s.from_tuples : Construct an %(klass)s from an
+            array-like of tuples.
+
+        Notes
+        -----
+        Each element of `left` must be less than or equal to the `right`
+        element at the same position. If an element is missing, it must be
+        missing in both `left` and `right`. A TypeError is raised when
+        using an unsupported type for `left` or `right`. At the moment,
+        'category', 'object', and 'string' subtypes are not supported.
+
+        %(examples)s\
+        """
+    )
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_arrays"]
+        % {
+            "klass": "IntervalArray",
+            "name": "",
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.arrays.IntervalArray.from_arrays([0, 1, 2], [1, 2, 3])
+        <IntervalArray>
+        [(0, 1], (1, 2], (2, 3]]
+        Length: 3, dtype: interval[int64, right]
+        """
+            ),
+        }
+    )
+    def from_arrays(
+        cls,
+        left,
+        right,
+        closed: IntervalClosedType | None = "right",
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> Self:
+        left = _maybe_convert_platform_interval(left)
+        right = _maybe_convert_platform_interval(right)
+
+        left, right, dtype = cls._ensure_simple_new_inputs(
+            left,
+            right,
+            closed=closed,
+            copy=copy,
+            dtype=dtype,
+        )
+        cls._validate(left, right, dtype=dtype)
+
+        return cls._simple_new(left, right, dtype=dtype)
+
+    _interval_shared_docs["from_tuples"] = textwrap.dedent(
+        """
+        Construct an %(klass)s from an array-like of tuples.
+
+        Parameters
+        ----------
+        data : array-like (1-dimensional)
+            Array of tuples.
+        closed : {'left', 'right', 'both', 'neither'}, default 'right'
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither.\
+        %(name)s
+        copy : bool, default False
+            By-default copy the data, this is compat only and ignored.
+        dtype : dtype or None, default None
+            If None, dtype will be inferred.
+
+        Returns
+        -------
+        %(klass)s
+
+        See Also
+        --------
+        interval_range : Function to create a fixed frequency IntervalIndex.
+        %(klass)s.from_arrays : Construct an %(klass)s from a left and
+                                    right array.
+        %(klass)s.from_breaks : Construct an %(klass)s from an array of
+                                    splits.
+
+        %(examples)s\
+        """
+    )
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_tuples"]
+        % {
+            "klass": "IntervalArray",
+            "name": "",
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.arrays.IntervalArray.from_tuples([(0, 1), (1, 2)])
+        <IntervalArray>
+        [(0, 1], (1, 2]]
+        Length: 2, dtype: interval[int64, right]
+        """
+            ),
+        }
+    )
+    def from_tuples(
+        cls,
+        data,
+        closed: IntervalClosedType | None = "right",
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> Self:
+        if len(data):
+            left, right = [], []
+        else:
+            # ensure that empty data keeps input dtype
+            left = right = data
+
+        for d in data:
+            if not isinstance(d, tuple) and isna(d):
+                lhs = rhs = np.nan
+            else:
+                name = cls.__name__
+                try:
+                    # need list of length 2 tuples, e.g. [(0, 1), (1, 2), ...]
+                    lhs, rhs = d
+                except ValueError as err:
+                    msg = f"{name}.from_tuples requires tuples of length 2, got {d}"
+                    raise ValueError(msg) from err
+                except TypeError as err:
+                    msg = f"{name}.from_tuples received an invalid item, {d}"
+                    raise TypeError(msg) from err
+            left.append(lhs)
+            right.append(rhs)
+
+        return cls.from_arrays(left, right, closed, copy=False, dtype=dtype)
+
+    @classmethod
+    def _validate(cls, left, right, dtype: IntervalDtype) -> None:
+        """
+        Verify that the IntervalArray is valid.
+
+        Checks that
+
+        * dtype is correct
+        * left and right match lengths
+        * left and right have the same missing values
+        * left is always below right
+        """
+        if not isinstance(dtype, IntervalDtype):
+            msg = f"invalid dtype: {dtype}"
+            raise ValueError(msg)
+        if len(left) != len(right):
+            msg = "left and right must have the same length"
+            raise ValueError(msg)
+        left_mask = notna(left)
+        right_mask = notna(right)
+        if not (left_mask == right_mask).all():
+            msg = (
+                "missing values must be missing in the same "
+                "location both left and right sides"
+            )
+            raise ValueError(msg)
+        if not (left[left_mask] <= right[left_mask]).all():
+            msg = "left side of interval must be <= right side"
+            raise ValueError(msg)
+
+    def _shallow_copy(self, left, right) -> Self:
+        """
+        Return a new IntervalArray with the replacement attributes
+
+        Parameters
+        ----------
+        left : Index
+            Values to be used for the left-side of the intervals.
+        right : Index
+            Values to be used for the right-side of the intervals.
+        """
+        dtype = IntervalDtype(left.dtype, closed=self.closed)
+        left, right, dtype = self._ensure_simple_new_inputs(left, right, dtype=dtype)
+
+        return self._simple_new(left, right, dtype=dtype)
+
+    # ---------------------------------------------------------------------
+    # Descriptive
+
+    @property
+    def dtype(self) -> IntervalDtype:
+        return self._dtype
+
+    @property
+    def nbytes(self) -> int:
+        return self.left.nbytes + self.right.nbytes
+
+    @property
+    def size(self) -> int:
+        # Avoid materializing self.values
+        return self.left.size
+
+    # ---------------------------------------------------------------------
+    # EA Interface
+
+    def __iter__(self) -> Iterator:
+        return iter(np.asarray(self))
+
+    def __len__(self) -> int:
+        return len(self._left)
+
+    @overload
+    def __getitem__(self, key: ScalarIndexer) -> IntervalOrNA: ...
+
+    @overload
+    def __getitem__(self, key: SequenceIndexer) -> Self: ...
+
+    def __getitem__(self, key: PositionalIndexer) -> Self | IntervalOrNA:
+        key = check_array_indexer(self, key)
+        left = self._left[key]
+        right = self._right[key]
+
+        if not isinstance(left, (np.ndarray, ExtensionArray)):
+            # scalar
+            if is_scalar(left) and isna(left):
+                return self._fill_value
+            return Interval(left, right, self.closed)
+        if np.ndim(left) > 1:
+            # GH#30588 multi-dimensional indexer disallowed
+            raise ValueError("multi-dimensional indexing not allowed")
+        # Argument 2 to "_simple_new" of "IntervalArray" has incompatible type
+        # "Union[Period, Timestamp, Timedelta, NaTType, DatetimeArray, TimedeltaArray,
+        # ndarray[Any, Any]]"; expected "Union[Union[DatetimeArray, TimedeltaArray],
+        # ndarray[Any, Any]]"
+        return self._simple_new(left, right, dtype=self.dtype)  # type: ignore[arg-type]
+
+    def __setitem__(self, key, value) -> None:
+        value_left, value_right = self._validate_setitem_value(value)
+        key = check_array_indexer(self, key)
+
+        self._left[key] = value_left
+        self._right[key] = value_right
+
+    def _cmp_method(self, other, op):
+        # ensure pandas array for list-like and eliminate non-interval scalars
+        if is_list_like(other):
+            if len(self) != len(other):
+                raise ValueError("Lengths must match to compare")
+            other = pd_array(other)
+        elif not isinstance(other, Interval):
+            # non-interval scalar -> no matches
+            if other is NA:
+                # GH#31882
+                from pandas.core.arrays import BooleanArray
+
+                arr = np.empty(self.shape, dtype=bool)
+                mask = np.ones(self.shape, dtype=bool)
+                return BooleanArray(arr, mask)
+            return invalid_comparison(self, other, op)
+
+        # determine the dtype of the elements we want to compare
+        if isinstance(other, Interval):
+            other_dtype = pandas_dtype("interval")
+        elif not isinstance(other.dtype, CategoricalDtype):
+            other_dtype = other.dtype
+        else:
+            # for categorical defer to categories for dtype
+            other_dtype = other.categories.dtype
+
+            # extract intervals if we have interval categories with matching closed
+            if isinstance(other_dtype, IntervalDtype):
+                if self.closed != other.categories.closed:
+                    return invalid_comparison(self, other, op)
+
+                other = other.categories._values.take(
+                    other.codes, allow_fill=True, fill_value=other.categories._na_value
+                )
+
+        # interval-like -> need same closed and matching endpoints
+        if isinstance(other_dtype, IntervalDtype):
+            if self.closed != other.closed:
+                return invalid_comparison(self, other, op)
+            elif not isinstance(other, Interval):
+                other = type(self)(other)
+
+            if op is operator.eq:
+                return (self._left == other.left) & (self._right == other.right)
+            elif op is operator.ne:
+                return (self._left != other.left) | (self._right != other.right)
+            elif op is operator.gt:
+                return (self._left > other.left) | (
+                    (self._left == other.left) & (self._right > other.right)
+                )
+            elif op is operator.ge:
+                return (self == other) | (self > other)
+            elif op is operator.lt:
+                return (self._left < other.left) | (
+                    (self._left == other.left) & (self._right < other.right)
+                )
+            else:
+                # operator.lt
+                return (self == other) | (self < other)
+
+        # non-interval/non-object dtype -> no matches
+        if not is_object_dtype(other_dtype):
+            return invalid_comparison(self, other, op)
+
+        # object dtype -> iteratively check for intervals
+        result = np.zeros(len(self), dtype=bool)
+        for i, obj in enumerate(other):
+            try:
+                result[i] = op(self[i], obj)
+            except TypeError:
+                if obj is NA:
+                    # comparison with np.nan returns NA
+                    # github.com/pandas-dev/pandas/pull/37124#discussion_r509095092
+                    result = result.astype(object)
+                    result[i] = NA
+                else:
+                    raise
+        return result
+
+    @unpack_zerodim_and_defer("__eq__")
+    def __eq__(self, other):
+        return self._cmp_method(other, operator.eq)
+
+    @unpack_zerodim_and_defer("__ne__")
+    def __ne__(self, other):
+        return self._cmp_method(other, operator.ne)
+
+    @unpack_zerodim_and_defer("__gt__")
+    def __gt__(self, other):
+        return self._cmp_method(other, operator.gt)
+
+    @unpack_zerodim_and_defer("__ge__")
+    def __ge__(self, other):
+        return self._cmp_method(other, operator.ge)
+
+    @unpack_zerodim_and_defer("__lt__")
+    def __lt__(self, other):
+        return self._cmp_method(other, operator.lt)
+
+    @unpack_zerodim_and_defer("__le__")
+    def __le__(self, other):
+        return self._cmp_method(other, operator.le)
+
+    def argsort(
+        self,
+        *,
+        ascending: bool = True,
+        kind: SortKind = "quicksort",
+        na_position: str = "last",
+        **kwargs,
+    ) -> np.ndarray:
+        ascending = nv.validate_argsort_with_ascending(ascending, (), kwargs)
+
+        if ascending and kind == "quicksort" and na_position == "last":
+            # TODO: in an IntervalIndex we can reuse the cached
+            #  IntervalTree.left_sorter
+            return np.lexsort((self.right, self.left))
+
+        # TODO: other cases we can use lexsort for?  much more performant.
+        return super().argsort(
+            ascending=ascending, kind=kind, na_position=na_position, **kwargs
+        )
+
+    def min(self, *, axis: AxisInt | None = None, skipna: bool = True) -> IntervalOrNA:
+        nv.validate_minmax_axis(axis, self.ndim)
+
+        if not len(self):
+            return self._na_value
+
+        mask = self.isna()
+        if mask.any():
+            if not skipna:
+                return self._na_value
+            obj = self[~mask]
+        else:
+            obj = self
+
+        indexer = obj.argsort()[0]
+        return obj[indexer]
+
+    def max(self, *, axis: AxisInt | None = None, skipna: bool = True) -> IntervalOrNA:
+        nv.validate_minmax_axis(axis, self.ndim)
+
+        if not len(self):
+            return self._na_value
+
+        mask = self.isna()
+        if mask.any():
+            if not skipna:
+                return self._na_value
+            obj = self[~mask]
+        else:
+            obj = self
+
+        indexer = obj.argsort()[-1]
+        return obj[indexer]
+
+    def fillna(self, value, limit: int | None = None, copy: bool = True) -> Self:
+        """
+        Fill NA/NaN values using the specified method.
+
+        Parameters
+        ----------
+        value : scalar, dict, Series
+            If a scalar value is passed it is used to fill all missing values.
+            Alternatively, a Series or dict can be used to fill in different
+            values for each index. The value should not be a list. The
+            value(s) passed should be either Interval objects or NA/NaN.
+        limit : int, default None
+            (Not implemented yet for IntervalArray)
+            The maximum number of entries where NA values will be filled.
+        copy : bool, default True
+            Whether to make a copy of the data before filling. If False, then
+            the original should be modified and no new memory should be allocated.
+            For ExtensionArray subclasses that cannot do this, it is at the
+            author's discretion whether to ignore "copy=False" or to raise.
+
+        Returns
+        -------
+        filled : IntervalArray with NA/NaN filled
+        """
+        if copy is False:
+            raise NotImplementedError
+        if limit is not None:
+            raise ValueError("limit must be None")
+
+        value_left, value_right = self._validate_scalar(value)
+
+        left = self.left.fillna(value=value_left)
+        right = self.right.fillna(value=value_right)
+        return self._shallow_copy(left, right)
+
+    def astype(self, dtype, copy: bool = True):
+        """
+        Cast to an ExtensionArray or NumPy array with dtype 'dtype'.
+
+        Parameters
+        ----------
+        dtype : str or dtype
+            Typecode or data-type to which the array is cast.
+
+        copy : bool, default True
+            Whether to copy the data, even if not necessary. If False,
+            a copy is made only if the old dtype does not match the
+            new dtype.
+
+        Returns
+        -------
+        array : ExtensionArray or ndarray
+            ExtensionArray or NumPy ndarray with 'dtype' for its dtype.
+        """
+        from pandas import Index
+
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        if isinstance(dtype, IntervalDtype):
+            if dtype == self.dtype:
+                return self.copy() if copy else self
+
+            if is_float_dtype(self.dtype.subtype) and needs_i8_conversion(
+                dtype.subtype
+            ):
+                # This is allowed on the Index.astype but we disallow it here
+                msg = (
+                    f"Cannot convert {self.dtype} to {dtype}; subtypes are incompatible"
+                )
+                raise TypeError(msg)
+
+            # need to cast to different subtype
+            try:
+                # We need to use Index rules for astype to prevent casting
+                #  np.nan entries to int subtypes
+                new_left = Index(self._left, copy=False).astype(dtype.subtype)
+                new_right = Index(self._right, copy=False).astype(dtype.subtype)
+            except IntCastingNaNError:
+                # e.g test_subtype_integer
+                raise
+            except (TypeError, ValueError) as err:
+                # e.g. test_subtype_integer_errors f8->u8 can be lossy
+                #  and raises ValueError
+                msg = (
+                    f"Cannot convert {self.dtype} to {dtype}; subtypes are incompatible"
+                )
+                raise TypeError(msg) from err
+            return self._shallow_copy(new_left, new_right)
+        else:
+            try:
+                return super().astype(dtype, copy=copy)
+            except (TypeError, ValueError) as err:
+                msg = f"Cannot cast {type(self).__name__} to dtype {dtype}"
+                raise TypeError(msg) from err
+
+    def equals(self, other) -> bool:
+        if type(self) != type(other):
+            return False
+
+        return bool(
+            self.closed == other.closed
+            and self.left.equals(other.left)
+            and self.right.equals(other.right)
+        )
+
+    @classmethod
+    def _concat_same_type(cls, to_concat: Sequence[IntervalArray]) -> Self:
+        """
+        Concatenate multiple IntervalArray
+
+        Parameters
+        ----------
+        to_concat : sequence of IntervalArray
+
+        Returns
+        -------
+        IntervalArray
+        """
+        closed_set = {interval.closed for interval in to_concat}
+        if len(closed_set) != 1:
+            raise ValueError("Intervals must all be closed on the same side.")
+        closed = closed_set.pop()
+
+        left: IntervalSide = np.concatenate([interval.left for interval in to_concat])
+        right: IntervalSide = np.concatenate([interval.right for interval in to_concat])
+
+        left, right, dtype = cls._ensure_simple_new_inputs(left, right, closed=closed)
+
+        return cls._simple_new(left, right, dtype=dtype)
+
+    def copy(self) -> Self:
+        """
+        Return a copy of the array.
+
+        Returns
+        -------
+        IntervalArray
+        """
+        left = self._left.copy()
+        right = self._right.copy()
+        dtype = self.dtype
+        return self._simple_new(left, right, dtype=dtype)
+
+    def isna(self) -> np.ndarray:
+        return isna(self._left)
+
+    def shift(self, periods: int = 1, fill_value: object = None) -> IntervalArray:
+        if not len(self) or periods == 0:
+            return self.copy()
+
+        self._validate_scalar(fill_value)
+
+        # ExtensionArray.shift doesn't work for two reasons
+        # 1. IntervalArray.dtype.na_value may not be correct for the dtype.
+        # 2. IntervalArray._from_sequence only accepts NaN for missing values,
+        #    not other values like NaT
+
+        empty_len = min(abs(periods), len(self))
+        if isna(fill_value):
+            from pandas import Index
+
+            fill_value = Index(self._left, copy=False)._na_value
+            empty = IntervalArray.from_breaks(
+                [fill_value] * (empty_len + 1), closed=self.closed
+            )
+        else:
+            empty = self._from_sequence([fill_value] * empty_len, dtype=self.dtype)
+
+        if periods > 0:
+            a = empty
+            b = self[:-periods]
+        else:
+            a = self[abs(periods) :]
+            b = empty
+        return self._concat_same_type([a, b])
+
+    def take(
+        self,
+        indices,
+        *,
+        allow_fill: bool = False,
+        fill_value=None,
+        axis=None,
+        **kwargs,
+    ) -> Self:
+        """
+        Take elements from the IntervalArray.
+
+        Parameters
+        ----------
+        indices : sequence of integers
+            Indices to be taken.
+
+        allow_fill : bool, default False
+            How to handle negative values in `indices`.
+
+            * False: negative values in `indices` indicate positional indices
+              from the right (the default). This is similar to
+              :func:`numpy.take`.
+
+            * True: negative values in `indices` indicate
+              missing values. These values are set to `fill_value`. Any other
+              other negative values raise a ``ValueError``.
+
+        fill_value : Interval or NA, optional
+            Fill value to use for NA-indices when `allow_fill` is True.
+            This may be ``None``, in which case the default NA value for
+            the type, ``self.dtype.na_value``, is used.
+
+            For many ExtensionArrays, there will be two representations of
+            `fill_value`: a user-facing "boxed" scalar, and a low-level
+            physical NA value. `fill_value` should be the user-facing version,
+            and the implementation should handle translating that to the
+            physical version for processing the take if necessary.
+
+        axis : any, default None
+            Present for compat with IntervalIndex; does nothing.
+
+        Returns
+        -------
+        IntervalArray
+
+        Raises
+        ------
+        IndexError
+            When the indices are out of bounds for the array.
+        ValueError
+            When `indices` contains negative values other than ``-1``
+            and `allow_fill` is True.
+        """
+        nv.validate_take((), kwargs)
+
+        fill_left = fill_right = fill_value
+        if allow_fill:
+            fill_left, fill_right = self._validate_scalar(fill_value)
+
+        left_take = take(
+            self._left, indices, allow_fill=allow_fill, fill_value=fill_left
+        )
+        right_take = take(
+            self._right, indices, allow_fill=allow_fill, fill_value=fill_right
+        )
+
+        return self._shallow_copy(left_take, right_take)
+
+    def _validate_listlike(self, value):
+        # list-like of intervals
+        try:
+            array = IntervalArray(value)
+            self._check_closed_matches(array, name="value")
+            value_left, value_right = array.left, array.right
+        except TypeError as err:
+            # wrong type: not interval or NA
+            msg = f"'value' should be an interval type, got {type(value)} instead."
+            raise TypeError(msg) from err
+
+        try:
+            self.left._validate_fill_value(value_left)
+        except (LossySetitemError, TypeError) as err:
+            msg = (
+                "'value' should be a compatible interval type, "
+                f"got {type(value)} instead."
+            )
+            raise TypeError(msg) from err
+
+        return value_left, value_right
+
+    def _validate_scalar(self, value):
+        if isinstance(value, Interval):
+            self._check_closed_matches(value, name="value")
+            left, right = value.left, value.right
+            # TODO: check subdtype match like _validate_setitem_value?
+        elif is_valid_na_for_dtype(value, self.left.dtype):
+            # GH#18295
+            left = right = self.left._na_value
+        else:
+            raise TypeError(
+                "can only insert Interval objects and NA into an IntervalArray"
+            )
+        return left, right
+
+    def _validate_setitem_value(self, value):
+        if is_valid_na_for_dtype(value, self.left.dtype):
+            # na value: need special casing to set directly on numpy arrays
+            value = self.left._na_value
+            if is_integer_dtype(self.dtype.subtype):
+                # can't set NaN on a numpy integer array
+                # GH#45484 TypeError, not ValueError, matches what we get with
+                #  non-NA un-holdable value.
+                raise TypeError("Cannot set float NaN to integer-backed IntervalArray")
+            value_left, value_right = value, value
+
+        elif isinstance(value, Interval):
+            # scalar interval
+            self._check_closed_matches(value, name="value")
+            value_left, value_right = value.left, value.right
+            self.left._validate_fill_value(value_left)
+            self.left._validate_fill_value(value_right)
+
+        else:
+            return self._validate_listlike(value)
+
+        return value_left, value_right
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Returns a Series containing counts of each interval.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of NaN.
+
+        Returns
+        -------
+        counts : Series
+
+        See Also
+        --------
+        Series.value_counts
+        """
+        # TODO: implement this is a non-naive way!
+        result = value_counts(np.asarray(self), dropna=dropna)
+        result.index = result.index.astype(self.dtype)
+        return result
+
+    # ---------------------------------------------------------------------
+    # Rendering Methods
+
+    def _formatter(self, boxed: bool = False) -> Callable[[object], str]:
+        # returning 'str' here causes us to render as e.g. "(0, 1]" instead of
+        #  "Interval(0, 1, closed='right')"
+        return str
+
+    # ---------------------------------------------------------------------
+    # Vectorized Interval Properties/Attributes
+
+    @property
+    def left(self) -> Index:
+        """
+        Return the left endpoints of each Interval in the IntervalArray as an Index.
+
+        This property provides access to the left endpoints of the intervals
+        contained within the IntervalArray. This can be useful for analyses where
+        the starting point of each interval is of interest, such as in histogram
+        creation, data aggregation, or any scenario requiring the identification
+        of the beginning of defined ranges. This property returns a ``pandas.Index``
+        object containing the midpoint for each interval.
+
+        See Also
+        --------
+        arrays.IntervalArray.right : Return the right endpoints of each Interval in
+            the IntervalArray as an Index.
+        arrays.IntervalArray.mid : Return the midpoint of each Interval in the
+            IntervalArray as an Index.
+        arrays.IntervalArray.contains : Check elementwise if the Intervals contain
+            the value.
+
+        Examples
+        --------
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(2, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (2, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.left
+        Index([0, 2], dtype='int64')
+        """
+        from pandas import Index
+
+        return Index(self._left, copy=False)
+
+    @property
+    def right(self) -> Index:
+        """
+        Return the right endpoints of each Interval in the IntervalArray as an Index.
+
+        This property extracts the right endpoints from each interval contained within
+        the IntervalArray. This can be helpful in use cases where you need to work
+        with or compare only the upper bounds of intervals, such as when performing
+        range-based filtering, determining interval overlaps, or visualizing the end
+        boundaries of data segments.
+
+        See Also
+        --------
+        arrays.IntervalArray.left : Return the left endpoints of each Interval in
+            the IntervalArray as an Index.
+        arrays.IntervalArray.mid : Return the midpoint of each Interval in the
+            IntervalArray as an Index.
+        arrays.IntervalArray.contains : Check elementwise if the Intervals contain
+            the value.
+
+        Examples
+        --------
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(2, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (2, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.right
+        Index([1, 5], dtype='int64')
+        """
+        from pandas import Index
+
+        return Index(self._right, copy=False)
+
+    @property
+    def length(self) -> Index:
+        """
+        Return an Index with entries denoting the length of each Interval.
+
+        The length of an interval is calculated as the difference between
+        its `right` and `left` bounds. This property is particularly useful
+        when working with intervals where the size of the interval is an important
+        attribute, such as in time-series analysis or spatial data analysis.
+
+        See Also
+        --------
+        arrays.IntervalArray.left : Return the left endpoints of each Interval in
+            the IntervalArray as an Index.
+        arrays.IntervalArray.right : Return the right endpoints of each Interval in
+            the IntervalArray as an Index.
+        arrays.IntervalArray.mid : Return the midpoint of each Interval in the
+            IntervalArray as an Index.
+
+        Examples
+        --------
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.length
+        Index([1, 4], dtype='int64')
+        """
+        return self.right - self.left
+
+    @property
+    def mid(self) -> Index:
+        """
+        Return the midpoint of each Interval in the IntervalArray as an Index.
+
+        The midpoint of an interval is calculated as the average of its
+        ``left`` and ``right`` bounds. This property returns a ``pandas.Index`` object
+        containing the midpoint for each interval.
+
+        See Also
+        --------
+        Interval.left : Return left bound for the interval.
+        Interval.right : Return right bound for the interval.
+        Interval.length : Return the length of each interval.
+
+        Examples
+        --------
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.mid
+        Index([0.5, 3.0], dtype='float64')
+        """
+        try:
+            return 0.5 * (self.left + self.right)
+        except TypeError:
+            # datetime safe version
+            return self.left + 0.5 * self.length
+
+    _interval_shared_docs["overlaps"] = textwrap.dedent(
+        """
+        Check elementwise if an Interval overlaps the values in the %(klass)s.
+
+        Two intervals overlap if they share a common point, including closed
+        endpoints. Intervals that only have an open endpoint in common do not
+        overlap.
+
+        Parameters
+        ----------
+        other : %(klass)s
+            Interval to check against for an overlap.
+
+        Returns
+        -------
+        ndarray
+            Boolean array positionally indicating where an overlap occurs.
+
+        See Also
+        --------
+        Interval.overlaps : Check whether two Interval objects overlap.
+
+        Examples
+        --------
+        %(examples)s
+        >>> intervals.overlaps(pd.Interval(0.5, 1.5))
+        array([ True,  True, False])
+
+        Intervals that share closed endpoints overlap:
+
+        >>> intervals.overlaps(pd.Interval(1, 3, closed='left'))
+        array([ True,  True, True])
+
+        Intervals that only have an open endpoint in common do not overlap:
+
+        >>> intervals.overlaps(pd.Interval(1, 2, closed='right'))
+        array([False,  True, False])
+        """
+    )
+
+    @Appender(
+        _interval_shared_docs["overlaps"]
+        % {
+            "klass": "IntervalArray",
+            "examples": textwrap.dedent(
+                """\
+        >>> data = [(0, 1), (1, 3), (2, 4)]
+        >>> intervals = pd.arrays.IntervalArray.from_tuples(data)
+        >>> intervals
+        <IntervalArray>
+        [(0, 1], (1, 3], (2, 4]]
+        Length: 3, dtype: interval[int64, right]
+        """
+            ),
+        }
+    )
+    def overlaps(self, other):
+        if isinstance(other, (IntervalArray, ABCIntervalIndex)):
+            raise NotImplementedError
+        if not isinstance(other, Interval):
+            msg = f"`other` must be Interval-like, got {type(other).__name__}"
+            raise TypeError(msg)
+
+        # equality is okay if both endpoints are closed (overlap at a point)
+        op1 = le if (self.closed_left and other.closed_right) else lt
+        op2 = le if (other.closed_left and self.closed_right) else lt
+
+        # overlaps is equivalent negation of two interval being disjoint:
+        # disjoint = (A.left > B.right) or (B.left > A.right)
+        # (simplifying the negation allows this to be done in less operations)
+        return op1(self.left, other.right) & op2(other.left, self.right)
+
+    # ---------------------------------------------------------------------
+
+    @property
+    def closed(self) -> IntervalClosedType:
+        """
+        String describing the inclusive side the intervals.
+
+        Either ``left``, ``right``, ``both`` or ``neither``.
+
+        See Also
+        --------
+        IntervalArray.closed : Returns inclusive side of the IntervalArray.
+        Interval.closed : Returns inclusive side of the Interval.
+        IntervalIndex.closed : Returns inclusive side of the IntervalIndex.
+
+        Examples
+        --------
+
+        For arrays:
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.closed
+        'right'
+
+        For Interval Index:
+
+        >>> interv_idx = pd.interval_range(start=0, end=2)
+        >>> interv_idx
+        IntervalIndex([(0, 1], (1, 2]], dtype='interval[int64, right]')
+        >>> interv_idx.closed
+        'right'
+        """
+        return self.dtype.closed
+
+    _interval_shared_docs["set_closed"] = textwrap.dedent(
+        """
+        Return an identical %(klass)s closed on the specified side.
+
+        Parameters
+        ----------
+        closed : {'left', 'right', 'both', 'neither'}
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither.
+
+        Returns
+        -------
+        %(klass)s
+
+        %(examples)s\
+        """
+    )
+
+    def set_closed(self, closed: IntervalClosedType) -> Self:
+        """
+        Return an identical IntervalArray closed on the specified side.
+
+        Parameters
+        ----------
+        closed : {'left', 'right', 'both', 'neither'}
+            Whether the intervals are closed on the left-side, right-side, both
+            or neither.
+
+        Returns
+        -------
+        IntervalArray
+            A new IntervalArray with the specified side closures.
+
+        See Also
+        --------
+        IntervalArray.closed : Returns inclusive side of the Interval.
+        arrays.IntervalArray.closed : Returns inclusive side of the IntervalArray.
+
+        Examples
+        --------
+        >>> index = pd.arrays.IntervalArray.from_breaks(range(4))
+        >>> index
+        <IntervalArray>
+        [(0, 1], (1, 2], (2, 3]]
+        Length: 3, dtype: interval[int64, right]
+        >>> index.set_closed("both")
+        <IntervalArray>
+        [[0, 1], [1, 2], [2, 3]]
+        Length: 3, dtype: interval[int64, both]
+        """
+        if closed not in VALID_CLOSED:
+            msg = f"invalid option for 'closed': {closed}"
+            raise ValueError(msg)
+
+        left, right = self._left, self._right
+        dtype = IntervalDtype(left.dtype, closed=closed)
+        return self._simple_new(left, right, dtype=dtype)
+
+    _interval_shared_docs["is_non_overlapping_monotonic"] = """
+        Return a boolean whether the %(klass)s is non-overlapping and monotonic.
+
+        Non-overlapping means (no Intervals share points), and monotonic means
+        either monotonic increasing or monotonic decreasing.
+
+        Examples
+        --------
+        For arrays:
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.is_non_overlapping_monotonic
+        True
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1),
+        ...                                       pd.Interval(-1, 0.1)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0.0, 1.0], (-1.0, 0.1]]
+        Length: 2, dtype: interval[float64, right]
+        >>> interv_arr.is_non_overlapping_monotonic
+        False
+
+        For Interval Index:
+
+        >>> interv_idx = pd.interval_range(start=0, end=2)
+        >>> interv_idx
+        IntervalIndex([(0, 1], (1, 2]], dtype='interval[int64, right]')
+        >>> interv_idx.is_non_overlapping_monotonic
+        True
+
+        >>> interv_idx = pd.interval_range(start=0, end=2, closed='both')
+        >>> interv_idx
+        IntervalIndex([[0, 1], [1, 2]], dtype='interval[int64, both]')
+        >>> interv_idx.is_non_overlapping_monotonic
+        False
+        """
+
+    @property
+    def is_non_overlapping_monotonic(self) -> bool:
+        """
+        Return a boolean whether the IntervalArray/IntervalIndex\
+        is non-overlapping and monotonic.
+
+        Non-overlapping means (no Intervals share points), and monotonic means
+        either monotonic increasing or monotonic decreasing.
+
+        See Also
+        --------
+        overlaps : Check if two IntervalIndex objects overlap.
+
+        Examples
+        --------
+        For arrays:
+
+        >>> interv_arr = pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)])
+        >>> interv_arr
+        <IntervalArray>
+        [(0, 1], (1, 5]]
+        Length: 2, dtype: interval[int64, right]
+        >>> interv_arr.is_non_overlapping_monotonic
+        True
+
+        >>> interv_arr = pd.arrays.IntervalArray(
+        ...     [pd.Interval(0, 1), pd.Interval(-1, 0.1)]
+        ... )
+        >>> interv_arr
+        <IntervalArray>
+        [(0.0, 1.0], (-1.0, 0.1]]
+        Length: 2, dtype: interval[float64, right]
+        >>> interv_arr.is_non_overlapping_monotonic
+        False
+
+        For Interval Index:
+
+        >>> interv_idx = pd.interval_range(start=0, end=2)
+        >>> interv_idx
+        IntervalIndex([(0, 1], (1, 2]], dtype='interval[int64, right]')
+        >>> interv_idx.is_non_overlapping_monotonic
+        True
+
+        >>> interv_idx = pd.interval_range(start=0, end=2, closed="both")
+        >>> interv_idx
+        IntervalIndex([[0, 1], [1, 2]], dtype='interval[int64, both]')
+        >>> interv_idx.is_non_overlapping_monotonic
+        False
+        """
+        # must be increasing  (e.g., [0, 1), [1, 2), [2, 3), ... )
+        # or decreasing (e.g., [-1, 0), [-2, -1), [-3, -2), ...)
+        # we already require left <= right
+
+        # strict inequality for closed == 'both'; equality implies overlapping
+        # at a point when both sides of intervals are included
+        if self.closed == "both":
+            return bool(
+                (self._right[:-1] < self._left[1:]).all()
+                or (self._left[:-1] > self._right[1:]).all()
+            )
+
+        # non-strict inequality when closed != 'both'; at least one side is
+        # not included in the intervals, so equality does not imply overlapping
+        return bool(
+            (self._right[:-1] <= self._left[1:]).all()
+            or (self._left[:-1] >= self._right[1:]).all()
+        )
+
+    # ---------------------------------------------------------------------
+    # Conversion
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        """
+        Return the IntervalArray's data as a numpy array of Interval
+        objects (with dtype='object')
+        """
+        if copy is False:
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        left = self._left
+        right = self._right
+        mask = self.isna()
+        closed = self.closed
+
+        result = np.empty(len(left), dtype=object)
+        for i, left_value in enumerate(left):
+            if mask[i]:
+                result[i] = np.nan
+            else:
+                result[i] = Interval(left_value, right[i], closed)
+        return result
+
+    def __arrow_array__(self, type=None):
+        """
+        Convert myself into a pyarrow Array.
+        """
+        import pyarrow
+
+        from pandas.core.arrays.arrow.extension_types import ArrowIntervalType
+
+        try:
+            subtype = pyarrow.from_numpy_dtype(self.dtype.subtype)
+        except TypeError as err:
+            raise TypeError(
+                f"Conversion to arrow with subtype '{self.dtype.subtype}' "
+                "is not supported"
+            ) from err
+        interval_type = ArrowIntervalType(subtype, self.closed)
+        storage_array = pyarrow.StructArray.from_arrays(
+            [
+                pyarrow.array(self._left, type=subtype, from_pandas=True),
+                pyarrow.array(self._right, type=subtype, from_pandas=True),
+            ],
+            names=["left", "right"],
+        )
+        mask = self.isna()
+        if mask.any():
+            # if there are missing values, set validity bitmap also on the array level
+            null_bitmap = pyarrow.array(~mask).buffers()[1]
+            storage_array = pyarrow.StructArray.from_buffers(
+                storage_array.type,
+                len(storage_array),
+                [null_bitmap],
+                children=[storage_array.field(0), storage_array.field(1)],
+            )
+
+        if type is not None:
+            if type.equals(interval_type.storage_type):
+                return storage_array
+            elif isinstance(type, ArrowIntervalType):
+                # ensure we have the same subtype and closed attributes
+                if not type.equals(interval_type):
+                    raise TypeError(
+                        "Not supported to convert IntervalArray to type with "
+                        f"different 'subtype' ({self.dtype.subtype} vs {type.subtype}) "
+                        f"and 'closed' ({self.closed} vs {type.closed}) attributes"
+                    )
+            else:
+                raise TypeError(
+                    f"Not supported to convert IntervalArray to '{type}' type"
+                )
+
+        return pyarrow.ExtensionArray.from_storage(interval_type, storage_array)
+
+    _interval_shared_docs["to_tuples"] = textwrap.dedent(
+        """
+        Return an %(return_type)s of tuples of the form (left, right).
+
+        Parameters
+        ----------
+        na_tuple : bool, default True
+            If ``True``, return ``NA`` as a tuple ``(nan, nan)``. If ``False``,
+            just return ``NA`` as ``nan``.
+
+        Returns
+        -------
+        tuples: %(return_type)s
+        %(examples)s\
+        """
+    )
+
+    def to_tuples(self, na_tuple: bool = True) -> np.ndarray:
+        """
+        Return an ndarray (if self is IntervalArray) or Index \
+        (if self is IntervalIndex) of tuples of the form (left, right).
+
+        Parameters
+        ----------
+        na_tuple : bool, default True
+            If ``True``, return ``NA`` as a tuple ``(nan, nan)``. If ``False``,
+            just return ``NA`` as ``nan``.
+
+        Returns
+        -------
+        ndarray or Index
+            An ndarray of tuples representing the intervals
+                if `self` is an IntervalArray.
+            An Index of tuples representing the intervals
+                if `self` is an IntervalIndex.
+
+        See Also
+        --------
+        IntervalArray.to_list : Convert IntervalArray to a list of tuples.
+        IntervalArray.to_numpy : Convert IntervalArray to a numpy array.
+        IntervalArray.unique : Find unique intervals in an IntervalArray.
+
+        Examples
+        --------
+        For :class:`pandas.IntervalArray`:
+
+        >>> idx = pd.arrays.IntervalArray.from_tuples([(0, 1), (1, 2)])
+        >>> idx
+        <IntervalArray>
+        [(0, 1], (1, 2]]
+        Length: 2, dtype: interval[int64, right]
+        >>> idx.to_tuples()
+        array([(0, 1), (1, 2)], dtype=object)
+
+        For :class:`pandas.IntervalIndex`:
+
+        >>> idx = pd.interval_range(start=0, end=2)
+        >>> idx
+        IntervalIndex([(0, 1], (1, 2]], dtype='interval[int64, right]')
+        >>> idx.to_tuples()
+        Index([(0, 1), (1, 2)], dtype='object')
+        """
+        tuples = com.asarray_tuplesafe(zip(self._left, self._right))
+        if not na_tuple:
+            # GH 18756
+            tuples = np.where(~self.isna(), tuples, np.nan)
+        return tuples
+
+    # ---------------------------------------------------------------------
+
+    def _putmask(self, mask: npt.NDArray[np.bool_], value) -> None:
+        value_left, value_right = self._validate_setitem_value(value)
+
+        if isinstance(self._left, np.ndarray):
+            np.putmask(self._left, mask, value_left)
+            assert isinstance(self._right, np.ndarray)
+            np.putmask(self._right, mask, value_right)
+        else:
+            self._left._putmask(mask, value_left)
+            assert not isinstance(self._right, np.ndarray)
+            self._right._putmask(mask, value_right)
+
+    def insert(self, loc: int, item: Interval) -> Self:
+        """
+        Return a new IntervalArray inserting new item at location. Follows
+        Python numpy.insert semantics for negative values.  Only Interval
+        objects and NA can be inserted into an IntervalIndex
+
+        Parameters
+        ----------
+        loc : int
+        item : Interval
+
+        Returns
+        -------
+        IntervalArray
+        """
+        left_insert, right_insert = self._validate_scalar(item)
+
+        new_left = self.left.insert(loc, left_insert)
+        new_right = self.right.insert(loc, right_insert)
+
+        return self._shallow_copy(new_left, new_right)
+
+    def delete(self, loc) -> Self:
+        new_left: np.ndarray | DatetimeArray | TimedeltaArray
+        new_right: np.ndarray | DatetimeArray | TimedeltaArray
+        if isinstance(self._left, np.ndarray):
+            new_left = np.delete(self._left, loc)
+            assert isinstance(self._right, np.ndarray)
+            new_right = np.delete(self._right, loc)
+        else:
+            new_left = self._left.delete(loc)
+            assert not isinstance(self._right, np.ndarray)
+            new_right = self._right.delete(loc)
+        return self._shallow_copy(left=new_left, right=new_right)
+
+    @Appender(_extension_array_shared_docs["repeat"] % _shared_docs_kwargs)
+    def repeat(
+        self,
+        repeats: int | Sequence[int],
+        axis: AxisInt | None = None,
+    ) -> Self:
+        nv.validate_repeat((), {"axis": axis})
+        left_repeat = self.left.repeat(repeats)
+        right_repeat = self.right.repeat(repeats)
+        return self._shallow_copy(left=left_repeat, right=right_repeat)
+
+    _interval_shared_docs["contains"] = textwrap.dedent(
+        """
+        Check elementwise if the Intervals contain the value.
+
+        Return a boolean mask whether the value is contained in the Intervals
+        of the %(klass)s.
+
+        Parameters
+        ----------
+        other : scalar
+            The value to check whether it is contained in the Intervals.
+
+        Returns
+        -------
+        boolean array
+
+        See Also
+        --------
+        Interval.contains : Check whether Interval object contains value.
+        %(klass)s.overlaps : Check if an Interval overlaps the values in the
+            %(klass)s.
+
+        Examples
+        --------
+        %(examples)s
+        >>> intervals.contains(0.5)
+        array([ True, False, False])
+    """
+    )
+
+    def contains(self, other):
+        """
+        Check elementwise if the Intervals contain the value.
+
+        Return a boolean mask whether the value is contained in the Intervals
+        of the IntervalArray.
+
+        Parameters
+        ----------
+        other : scalar
+            The value to check whether it is contained in the Intervals.
+
+        Returns
+        -------
+        boolean array
+            A boolean mask whether the value is contained in the Intervals.
+
+        See Also
+        --------
+        Interval.contains : Check whether Interval object contains value.
+        IntervalArray.overlaps : Check if an Interval overlaps the values in the
+            IntervalArray.
+
+        Examples
+        --------
+        >>> intervals = pd.arrays.IntervalArray.from_tuples([(0, 1), (1, 3), (2, 4)])
+        >>> intervals
+        <IntervalArray>
+        [(0, 1], (1, 3], (2, 4]]
+        Length: 3, dtype: interval[int64, right]
+
+        >>> intervals.contains(0.5)
+        array([ True, False, False])
+        """
+        if isinstance(other, Interval):
+            raise NotImplementedError("contains not implemented for two intervals")
+
+        return (self._left < other if self.open_left else self._left <= other) & (
+            other < self._right if self.open_right else other <= self._right
+        )
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        if isinstance(values, IntervalArray):
+            if self.closed != values.closed:
+                # not comparable -> no overlap
+                return np.zeros(self.shape, dtype=bool)
+
+            if self.dtype == values.dtype:
+                # GH#38353 instead of casting to object, operating on a
+                #  complex128 ndarray is much more performant.
+                left = self._combined.view("complex128")
+                right = values._combined.view("complex128")
+                # error: Argument 1 to "isin" has incompatible type
+                # "Union[ExtensionArray, ndarray[Any, Any],
+                # ndarray[Any, dtype[Any]]]"; expected
+                # "Union[_SupportsArray[dtype[Any]],
+                # _NestedSequence[_SupportsArray[dtype[Any]]], bool,
+                # int, float, complex, str, bytes, _NestedSequence[
+                # Union[bool, int, float, complex, str, bytes]]]"
+                return np.isin(left, right).ravel()  # type: ignore[arg-type]
+
+            elif needs_i8_conversion(self.left.dtype) ^ needs_i8_conversion(
+                values.left.dtype
+            ):
+                # not comparable -> no overlap
+                return np.zeros(self.shape, dtype=bool)
+
+        return isin(self.astype(object), values.astype(object))
+
+    @property
+    def _combined(self) -> IntervalSide:
+        # error: Item "ExtensionArray" of "ExtensionArray | ndarray[Any, Any]"
+        # has no attribute "reshape"  [union-attr]
+        left = self.left._values.reshape(-1, 1)  # type: ignore[union-attr]
+        right = self.right._values.reshape(-1, 1)  # type: ignore[union-attr]
+        if needs_i8_conversion(left.dtype):
+            # error: Item "ndarray[Any, Any]" of "Any | ndarray[Any, Any]" has
+            # no attribute "_concat_same_type"
+            comb = left._concat_same_type(  # type: ignore[union-attr]
+                [left, right], axis=1
+            )
+        else:
+            comb = np.concatenate([left, right], axis=1)
+        return comb
+
+    def _from_combined(self, combined: np.ndarray) -> IntervalArray:
+        """
+        Create a new IntervalArray with our dtype from a 1D complex128 ndarray.
+        """
+        nc = combined.view("i8").reshape(-1, 2)
+
+        dtype = self._left.dtype
+        if needs_i8_conversion(dtype):
+            assert isinstance(self._left, (DatetimeArray, TimedeltaArray))
+            new_left: DatetimeArray | TimedeltaArray | np.ndarray = type(
+                self._left
+            )._from_sequence(nc[:, 0], dtype=dtype)
+            assert isinstance(self._right, (DatetimeArray, TimedeltaArray))
+            new_right: DatetimeArray | TimedeltaArray | np.ndarray = type(
+                self._right
+            )._from_sequence(nc[:, 1], dtype=dtype)
+        else:
+            assert isinstance(dtype, np.dtype)
+            new_left = nc[:, 0].view(dtype)
+            new_right = nc[:, 1].view(dtype)
+        return self._shallow_copy(left=new_left, right=new_right)
+
+    def unique(self) -> IntervalArray:
+        # No overload variant of "__getitem__" of "ExtensionArray" matches argument
+        # type "Tuple[slice, int]"
+        nc = unique(
+            self._combined.view("complex128")[:, 0]  # type: ignore[call-overload]
+        )
+        nc = nc[:, None]
+        return self._from_combined(nc)
+
+
+def _maybe_convert_platform_interval(values) -> ArrayLike:
+    """
+    Try to do platform conversion, with special casing for IntervalArray.
+    Wrapper around maybe_convert_platform that alters the default return
+    dtype in certain cases to be compatible with IntervalArray.  For example,
+    empty lists return with integer dtype instead of object dtype, which is
+    prohibited for IntervalArray.
+
+    Parameters
+    ----------
+    values : array-like
+
+    Returns
+    -------
+    array
+    """
+    if isinstance(values, (list, tuple)) and len(values) == 0:
+        # GH 19016
+        # empty lists/tuples get object dtype by default, but this is
+        # prohibited for IntervalArray, so coerce to integer instead
+        return np.array([], dtype=np.int64)
+    elif not is_list_like(values) or isinstance(values, ABCDataFrame):
+        # This will raise later, but we avoid passing to maybe_convert_platform
+        return values
+    elif isinstance(getattr(values, "dtype", None), CategoricalDtype):
+        values = np.asarray(values)
+    elif not hasattr(values, "dtype") and not isinstance(values, (list, tuple, range)):
+        # TODO: should we just cast these to list?
+        return values
+    else:
+        values = extract_array(values, extract_numpy=True)
+
+    if not hasattr(values, "dtype"):
+        values = np.asarray(values)
+        if values.dtype.kind in "iu" and values.dtype != np.int64:
+            values = values.astype(np.int64)
+    return values
diff --git a/pandas/pandas/core/arrays/masked.py b/pandas/pandas/core/arrays/masked.py
new file mode 100644
index 0000000000000..f3a0cc0dccdb3
--- /dev/null
+++ b/pandas/pandas/core/arrays/masked.py
@@ -0,0 +1,1658 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas._libs.tslibs import is_supported_dtype
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import doc
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_integer_dtype,
+    is_list_like,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import BaseMaskedDtype
+from pandas.core.dtypes.missing import (
+    array_equivalent,
+    is_valid_na_for_dtype,
+    isna,
+    notna,
+)
+
+from pandas.core import (
+    algorithms as algos,
+    arraylike,
+    missing,
+    nanops,
+    ops,
+)
+from pandas.core.algorithms import (
+    factorize_array,
+    isin,
+    map_array,
+    mode,
+    take,
+)
+from pandas.core.array_algos import (
+    masked_accumulations,
+    masked_reductions,
+)
+from pandas.core.array_algos.quantile import quantile_with_mask
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays._utils import to_numpy_dtype_inference
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.construction import (
+    array as pd_array,
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import check_array_indexer
+from pandas.core.ops import invalid_comparison
+from pandas.core.util.hashing import hash_array
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+    from collections.abc import (
+        Iterator,
+        Sequence,
+    )
+    from pandas import Series
+    from pandas.core.arrays import BooleanArray
+    from pandas._typing import (
+        NumpySorter,
+        NumpyValueArrayLike,
+        ArrayLike,
+        AstypeArg,
+        AxisInt,
+        DtypeObj,
+        FillnaOptions,
+        InterpolateOptions,
+        NpDtype,
+        PositionalIndexer,
+        Scalar,
+        ScalarIndexer,
+        Self,
+        SequenceIndexer,
+        Shape,
+        npt,
+    )
+    from pandas._libs.missing import NAType
+    from pandas.core.arrays import FloatingArray
+
+from pandas.compat.numpy import function as nv
+
+
+class BaseMaskedArray(OpsMixin, ExtensionArray):
+    """
+    Base class for masked arrays (which use _data and _mask to store the data).
+
+    numpy based
+    """
+
+    # our underlying data and mask are each ndarrays
+    _data: np.ndarray
+    _mask: npt.NDArray[np.bool_]
+
+    @classmethod
+    def _simple_new(cls, values: np.ndarray, mask: npt.NDArray[np.bool_]) -> Self:
+        result = BaseMaskedArray.__new__(cls)
+        result._data = values
+        result._mask = mask
+        return result
+
+    def __init__(
+        self, values: np.ndarray, mask: npt.NDArray[np.bool_], copy: bool = False
+    ) -> None:
+        # values is supposed to already be validated in the subclass
+        if not (isinstance(mask, np.ndarray) and mask.dtype == np.bool_):
+            raise TypeError(
+                "mask should be boolean numpy array. Use "
+                "the 'pd.array' function instead"
+            )
+        if values.shape != mask.shape:
+            raise ValueError("values.shape must match mask.shape")
+
+        if copy:
+            values = values.copy()
+            mask = mask.copy()
+
+        self._data = values
+        self._mask = mask
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy: bool = False) -> Self:
+        values, mask = cls._coerce_to_array(scalars, dtype=dtype, copy=copy)
+        return cls(values, mask)
+
+    @classmethod
+    @doc(ExtensionArray._empty)
+    def _empty(cls, shape: Shape, dtype: ExtensionDtype) -> Self:
+        dtype = cast(BaseMaskedDtype, dtype)
+        values: np.ndarray = np.empty(shape, dtype=dtype.type)
+        values.fill(dtype._internal_fill_value)
+        mask = np.ones(shape, dtype=bool)
+        result = cls(values, mask)
+        if not isinstance(result, cls) or dtype != result.dtype:
+            raise NotImplementedError(
+                f"Default 'empty' implementation is invalid for dtype='{dtype}'"
+            )
+        return result
+
+    def _formatter(self, boxed: bool = False) -> Callable[[Any], str | None]:
+        # NEP 51: https://github.com/numpy/numpy/pull/22449
+        return str
+
+    @property
+    def dtype(self) -> BaseMaskedDtype:
+        raise AbstractMethodError(self)
+
+    @overload
+    def __getitem__(self, item: ScalarIndexer) -> Any: ...
+
+    @overload
+    def __getitem__(self, item: SequenceIndexer) -> Self: ...
+
+    def __getitem__(self, item: PositionalIndexer) -> Self | Any:
+        item = check_array_indexer(self, item)
+
+        newmask = self._mask[item]
+        if is_bool(newmask):
+            # This is a scalar indexing
+            if newmask:
+                return self.dtype.na_value
+            return self._data[item]
+
+        return self._simple_new(self._data[item], newmask)
+
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        mask = self._mask
+
+        if mask.any():
+            func = missing.get_fill_func(method, ndim=self.ndim)
+
+            npvalues = self._data.T
+            new_mask = mask.T
+            if copy:
+                npvalues = npvalues.copy()
+                new_mask = new_mask.copy()
+            elif limit_area is not None:
+                mask = mask.copy()
+            func(npvalues, limit=limit, mask=new_mask)
+
+            if limit_area is not None and not mask.all():
+                mask = mask.T
+                neg_mask = ~mask
+                first = neg_mask.argmax()
+                last = len(neg_mask) - neg_mask[::-1].argmax() - 1
+                if limit_area == "inside":
+                    new_mask[:first] |= mask[:first]
+                    new_mask[last + 1 :] |= mask[last + 1 :]
+                elif limit_area == "outside":
+                    new_mask[first + 1 : last] |= mask[first + 1 : last]
+
+            if copy:
+                return self._simple_new(npvalues.T, new_mask.T)
+            else:
+                return self
+        else:
+            if copy:
+                new_values = self.copy()
+            else:
+                new_values = self
+        return new_values
+
+    @doc(ExtensionArray.fillna)
+    def fillna(self, value, limit: int | None = None, copy: bool = True) -> Self:
+        mask = self._mask
+        if limit is not None and limit < len(self):
+            modify = mask.cumsum() > limit
+            if modify.any():
+                # Only copy mask if necessary
+                mask = mask.copy()
+                mask[modify] = False
+
+        value = missing.check_value_size(value, mask, len(self))
+
+        if mask.any():
+            # fill with value
+            if copy:
+                new_values = self.copy()
+            else:
+                new_values = self[:]
+            new_values[mask] = value
+        else:
+            if copy:
+                new_values = self.copy()
+            else:
+                new_values = self[:]
+        return new_values
+
+    @classmethod
+    def _coerce_to_array(
+        cls, values, *, dtype: DtypeObj, copy: bool = False
+    ) -> tuple[np.ndarray, np.ndarray]:
+        raise AbstractMethodError(cls)
+
+    def _validate_setitem_value(self, value):
+        """
+        Check if we have a scalar that we can cast losslessly.
+
+        Raises
+        ------
+        TypeError
+        """
+        kind = self.dtype.kind
+        # TODO: get this all from np_can_hold_element?
+        if kind == "b":
+            if lib.is_bool(value):
+                return value
+
+        elif kind == "f":
+            if lib.is_integer(value) or lib.is_float(value):
+                return value
+
+        else:
+            if lib.is_integer(value) or (lib.is_float(value) and value.is_integer()):
+                return value
+            # TODO: unsigned checks
+
+        # Note: without the "str" here, the f-string rendering raises in
+        #  py38 builds.
+        raise TypeError(f"Invalid value '{value!s}' for dtype '{self.dtype}'")
+
+    def __setitem__(self, key, value) -> None:
+        key = check_array_indexer(self, key)
+
+        if is_scalar(value):
+            if is_valid_na_for_dtype(value, self.dtype):
+                self._mask[key] = True
+            else:
+                value = self._validate_setitem_value(value)
+                self._data[key] = value
+                self._mask[key] = False
+            return
+
+        value, mask = self._coerce_to_array(value, dtype=self.dtype)
+
+        self._data[key] = value
+        self._mask[key] = mask
+
+    def __contains__(self, key) -> bool:
+        if isna(key) and key is not self.dtype.na_value:
+            # GH#52840
+            if self._data.dtype.kind == "f" and lib.is_float(key):
+                return bool((np.isnan(self._data) & ~self._mask).any())
+
+        return bool(super().__contains__(key))
+
+    def __iter__(self) -> Iterator:
+        if self.ndim == 1:
+            if not self._hasna:
+                for val in self._data:
+                    yield val
+            else:
+                na_value = self.dtype.na_value
+                for isna_, val in zip(self._mask, self._data):
+                    if isna_:
+                        yield na_value
+                    else:
+                        yield val
+        else:
+            for i in range(len(self)):
+                yield self[i]
+
+    def __len__(self) -> int:
+        return len(self._data)
+
+    @property
+    def shape(self) -> Shape:
+        return self._data.shape
+
+    @property
+    def ndim(self) -> int:
+        return self._data.ndim
+
+    def swapaxes(self, axis1, axis2) -> Self:
+        data = self._data.swapaxes(axis1, axis2)
+        mask = self._mask.swapaxes(axis1, axis2)
+        return self._simple_new(data, mask)
+
+    def delete(self, loc, axis: AxisInt = 0) -> Self:
+        data = np.delete(self._data, loc, axis=axis)
+        mask = np.delete(self._mask, loc, axis=axis)
+        return self._simple_new(data, mask)
+
+    def reshape(self, *args, **kwargs) -> Self:
+        data = self._data.reshape(*args, **kwargs)
+        mask = self._mask.reshape(*args, **kwargs)
+        return self._simple_new(data, mask)
+
+    def ravel(self, *args, **kwargs) -> Self:
+        # TODO: need to make sure we have the same order for data/mask
+        data = self._data.ravel(*args, **kwargs)
+        mask = self._mask.ravel(*args, **kwargs)
+        return type(self)(data, mask)
+
+    @property
+    def T(self) -> Self:
+        return self._simple_new(self._data.T, self._mask.T)
+
+    def round(self, decimals: int = 0, *args, **kwargs):
+        """
+        Round each value in the array a to the given number of decimals.
+
+        Parameters
+        ----------
+        decimals : int, default 0
+            Number of decimal places to round to. If decimals is negative,
+            it specifies the number of positions to the left of the decimal point.
+        *args, **kwargs
+            Additional arguments and keywords have no effect but might be
+            accepted for compatibility with NumPy.
+
+        Returns
+        -------
+        NumericArray
+            Rounded values of the NumericArray.
+
+        See Also
+        --------
+        numpy.around : Round values of an np.array.
+        DataFrame.round : Round values of a DataFrame.
+        Series.round : Round values of a Series.
+        """
+        if self.dtype.kind == "b":
+            return self
+        nv.validate_round(args, kwargs)
+        values = np.round(self._data, decimals=decimals, **kwargs)
+
+        # Usually we'll get same type as self, but ndarray[bool] casts to float
+        return self._maybe_mask_result(values, self._mask.copy())
+
+    # ------------------------------------------------------------------
+    # Unary Methods
+
+    def __invert__(self) -> Self:
+        return self._simple_new(~self._data, self._mask.copy())
+
+    def __neg__(self) -> Self:
+        return self._simple_new(-self._data, self._mask.copy())
+
+    def __pos__(self) -> Self:
+        return self.copy()
+
+    def __abs__(self) -> Self:
+        return self._simple_new(abs(self._data), self._mask.copy())
+
+    # ------------------------------------------------------------------
+
+    def _values_for_json(self) -> np.ndarray:
+        return np.asarray(self, dtype=object)
+
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        """
+        Convert to a NumPy Array.
+
+        By default converts to an object-dtype NumPy array. Specify the `dtype` and
+        `na_value` keywords to customize the conversion.
+
+        Parameters
+        ----------
+        dtype : dtype, default object
+            The numpy dtype to convert to.
+        copy : bool, default False
+            Whether to ensure that the returned value is a not a view on
+            the array. Note that ``copy=False`` does not *ensure* that
+            ``to_numpy()`` is no-copy. Rather, ``copy=True`` ensure that
+            a copy is made, even if not strictly necessary. This is typically
+            only possible when no missing values are present and `dtype`
+            is the equivalent numpy dtype.
+        na_value : scalar, optional
+             Scalar missing value indicator to use in numpy array. Defaults
+             to the native missing value indicator of this array (pd.NA).
+
+        Returns
+        -------
+        numpy.ndarray
+
+        Examples
+        --------
+        An object-dtype is the default result
+
+        >>> a = pd.array([True, False, pd.NA], dtype="boolean")
+        >>> a.to_numpy()
+        array([True, False, <NA>], dtype=object)
+
+        When no missing values are present, an equivalent dtype can be used.
+
+        >>> pd.array([True, False], dtype="boolean").to_numpy(dtype="bool")
+        array([ True, False])
+        >>> pd.array([1, 2], dtype="Int64").to_numpy("int64")
+        array([1, 2])
+
+        However, requesting such dtype will raise a ValueError if
+        missing values are present and the default missing value :attr:`NA`
+        is used.
+
+        >>> a = pd.array([True, False, pd.NA], dtype="boolean")
+        >>> a
+        <BooleanArray>
+        [True, False, <NA>]
+        Length: 3, dtype: boolean
+
+        >>> a.to_numpy(dtype="bool")
+        Traceback (most recent call last):
+        ...
+        ValueError: cannot convert to bool numpy array in presence of missing values
+
+        Specify a valid `na_value` instead
+
+        >>> a.to_numpy(dtype="bool", na_value=False)
+        array([ True, False, False])
+        """
+        hasna = self._hasna
+        dtype, na_value = to_numpy_dtype_inference(self, dtype, na_value, hasna)
+        if dtype is None:
+            dtype = object
+
+        if hasna:
+            if (
+                dtype != object
+                and not is_string_dtype(dtype)
+                and na_value is libmissing.NA
+            ):
+                raise ValueError(
+                    f"cannot convert to '{dtype}'-dtype NumPy array "
+                    "with missing values. Specify an appropriate 'na_value' "
+                    "for this dtype."
+                )
+            # don't pass copy to astype -> always need a copy since we are mutating
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore", category=RuntimeWarning)
+                data = self._data.astype(dtype)
+            data[self._mask] = na_value
+        else:
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore", category=RuntimeWarning)
+                data = self._data.astype(dtype, copy=copy)
+        return data
+
+    @doc(ExtensionArray.tolist)
+    def tolist(self) -> list:
+        if self.ndim > 1:
+            return [x.tolist() for x in self]
+        dtype = None if self._hasna else self._data.dtype
+        return self.to_numpy(dtype=dtype, na_value=libmissing.NA).tolist()
+
+    @overload
+    def astype(self, dtype: npt.DTypeLike, copy: bool = ...) -> np.ndarray: ...
+
+    @overload
+    def astype(self, dtype: ExtensionDtype, copy: bool = ...) -> ExtensionArray: ...
+
+    @overload
+    def astype(self, dtype: AstypeArg, copy: bool = ...) -> ArrayLike: ...
+
+    def astype(self, dtype: AstypeArg, copy: bool = True) -> ArrayLike:
+        dtype = pandas_dtype(dtype)
+
+        if dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+
+        # if we are astyping to another nullable masked dtype, we can fastpath
+        if isinstance(dtype, BaseMaskedDtype):
+            # TODO deal with NaNs for FloatingArray case
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore", category=RuntimeWarning)
+                # TODO: Is rounding what we want long term?
+                data = self._data.astype(dtype.numpy_dtype, copy=copy)
+            # mask is copied depending on whether the data was copied, and
+            # not directly depending on the `copy` keyword
+            mask = self._mask if data is self._data else self._mask.copy()
+            cls = dtype.construct_array_type()
+            return cls(data, mask, copy=False)
+
+        if isinstance(dtype, ExtensionDtype):
+            eacls = dtype.construct_array_type()
+            return eacls._from_sequence(self, dtype=dtype, copy=copy)
+
+        na_value: float | np.datetime64 | lib.NoDefault
+
+        # coerce
+        if dtype.kind == "f":
+            # In astype, we consider dtype=float to also mean na_value=np.nan
+            na_value = np.nan
+        elif dtype.kind == "M":
+            na_value = np.datetime64("NaT")
+        else:
+            na_value = lib.no_default
+
+        # to_numpy will also raise, but we get somewhat nicer exception messages here
+        if dtype.kind in "iu" and self._hasna:
+            raise ValueError("cannot convert NA to integer")
+        if dtype.kind == "b" and self._hasna:
+            # careful: astype_nansafe converts np.nan to True
+            raise ValueError("cannot convert float NaN to bool")
+
+        data = self.to_numpy(dtype=dtype, na_value=na_value, copy=copy)
+        return data
+
+    __array_priority__ = 1000  # higher than ndarray so ops dispatch to us
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        """
+        the array interface, return my values
+        We return an object array here to preserve our scalar values
+        """
+        if copy is False:
+            if not self._hasna:
+                # special case, here we can simply return the underlying data
+                return np.array(self._data, dtype=dtype, copy=copy)
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        if copy is None:
+            copy = False  # The NumPy copy=False meaning is different here.
+        return self.to_numpy(dtype=dtype, copy=copy)
+
+    _HANDLED_TYPES: tuple[type, ...]
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        # For MaskedArray inputs, we apply the ufunc to ._data
+        # and mask the result.
+
+        out = kwargs.get("out", ())
+
+        for x in inputs + out:
+            if not isinstance(x, self._HANDLED_TYPES + (BaseMaskedArray,)):
+                return NotImplemented
+
+        # for binary ops, use our custom dunder methods
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            # e.g. test_ufunc_with_out
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                return result
+
+        mask = np.zeros(len(self), dtype=bool)
+        inputs2 = []
+        for x in inputs:
+            if isinstance(x, BaseMaskedArray):
+                mask |= x._mask
+                inputs2.append(x._data)
+            else:
+                inputs2.append(x)
+
+        def reconstruct(x: np.ndarray):
+            # we don't worry about scalar `x` here, since we
+            # raise for reduce up above.
+            from pandas.core.arrays import (
+                BooleanArray,
+                FloatingArray,
+                IntegerArray,
+            )
+
+            if x.dtype.kind == "b":
+                m = mask.copy()
+                return BooleanArray(x, m)
+            elif x.dtype.kind in "iu":
+                m = mask.copy()
+                return IntegerArray(x, m)
+            elif x.dtype.kind == "f":
+                m = mask.copy()
+                if x.dtype == np.float16:
+                    # reached in e.g. np.sqrt on BooleanArray
+                    # we don't support float16
+                    x = x.astype(np.float32)
+                return FloatingArray(x, m)
+            else:
+                x[mask] = np.nan
+            return x
+
+        result = getattr(ufunc, method)(*inputs2, **kwargs)
+        if ufunc.nout > 1:
+            # e.g. np.divmod
+            return tuple(reconstruct(x) for x in result)
+        elif method == "reduce":
+            # e.g. np.add.reduce; test_ufunc_reduce_raises
+            if self._mask.any():
+                return self._na_value
+            return result
+        else:
+            return reconstruct(result)
+
+    def __arrow_array__(self, type=None):
+        """
+        Convert myself into a pyarrow Array.
+        """
+        import pyarrow as pa
+
+        return pa.array(self._data, mask=self._mask, type=type)
+
+    @property
+    def _hasna(self) -> bool:
+        # Note: this is expensive right now! The hope is that we can
+        # make this faster by having an optional mask, but not have to change
+        # source code using it..
+
+        # error: Incompatible return value type (got "bool_", expected "bool")
+        return self._mask.any()  # type: ignore[return-value]
+
+    def _propagate_mask(
+        self, mask: npt.NDArray[np.bool_] | None, other
+    ) -> npt.NDArray[np.bool_]:
+        if mask is None:
+            mask = self._mask.copy()  # TODO: need test for BooleanArray needing a copy
+            if other is libmissing.NA:
+                # GH#45421 don't alter inplace
+                mask = mask | True
+            elif is_list_like(other) and len(other) == len(mask):
+                mask = mask | isna(other)
+        else:
+            mask = self._mask | mask
+        # Incompatible return value type (got "Optional[ndarray[Any, dtype[bool_]]]",
+        # expected "ndarray[Any, dtype[bool_]]")
+        return mask  # type: ignore[return-value]
+
+    def _arith_method(self, other, op):
+        op_name = op.__name__
+        omask = None
+
+        if (
+            not hasattr(other, "dtype")
+            and is_list_like(other)
+            and len(other) == len(self)
+        ):
+            # Try inferring masked dtype instead of casting to object
+            other = pd_array(other)
+            other = extract_array(other, extract_numpy=True)
+
+        if isinstance(other, BaseMaskedArray):
+            other, omask = other._data, other._mask
+
+        elif is_list_like(other):
+            if not isinstance(other, ExtensionArray):
+                other = np.asarray(other)
+            if other.ndim > 1:
+                raise NotImplementedError("can only perform ops with 1-d structures")
+
+        # We wrap the non-masked arithmetic logic used for numpy dtypes
+        #  in Series/Index arithmetic ops.
+        other = ops.maybe_prepare_scalar_for_op(other, (len(self),))
+        pd_op = ops.get_array_op(op)
+        other = ensure_wrapped_if_datetimelike(other)
+
+        if op_name in {"pow", "rpow"} and isinstance(other, np.bool_):
+            # Avoid DeprecationWarning: In future, it will be an error
+            #  for 'np.bool_' scalars to be interpreted as an index
+            #  e.g. test_array_scalar_like_equivalence
+            other = bool(other)
+
+        mask = self._propagate_mask(omask, other)
+
+        if other is libmissing.NA:
+            result = np.ones_like(self._data)
+            if self.dtype.kind == "b":
+                if op_name in {
+                    "floordiv",
+                    "rfloordiv",
+                    "pow",
+                    "rpow",
+                    "truediv",
+                    "rtruediv",
+                }:
+                    # GH#41165 Try to match non-masked Series behavior
+                    #  This is still imperfect GH#46043
+                    raise NotImplementedError(
+                        f"operator '{op_name}' not implemented for bool dtypes"
+                    )
+                if op_name in {"mod", "rmod"}:
+                    dtype = "int8"
+                else:
+                    dtype = "bool"
+                result = result.astype(dtype)
+            elif "truediv" in op_name and self.dtype.kind != "f":
+                # The actual data here doesn't matter since the mask
+                #  will be all-True, but since this is division, we want
+                #  to end up with floating dtype.
+                result = result.astype(np.float64)
+        else:
+            # Make sure we do this before the "pow" mask checks
+            #  to get an expected exception message on shape mismatch.
+            if self.dtype.kind in "iu" and op_name in ["floordiv", "mod"]:
+                # TODO(GH#30188) ATM we don't match the behavior of non-masked
+                #  types with respect to floordiv-by-zero
+                pd_op = op
+
+            with np.errstate(all="ignore"):
+                result = pd_op(self._data, other)
+
+        if op_name == "pow":
+            # 1 ** x is 1.
+            mask = np.where((self._data == 1) & ~self._mask, False, mask)
+            # x ** 0 is 1.
+            if omask is not None:
+                mask = np.where((other == 0) & ~omask, False, mask)
+            elif other is not libmissing.NA:
+                mask = np.where(other == 0, False, mask)
+
+        elif op_name == "rpow":
+            # 1 ** x is 1.
+            if omask is not None:
+                mask = np.where((other == 1) & ~omask, False, mask)
+            elif other is not libmissing.NA:
+                mask = np.where(other == 1, False, mask)
+            # x ** 0 is 1.
+            mask = np.where((self._data == 0) & ~self._mask, False, mask)
+
+        return self._maybe_mask_result(result, mask)
+
+    _logical_method = _arith_method
+
+    def _cmp_method(self, other, op) -> BooleanArray:
+        from pandas.core.arrays import BooleanArray
+
+        mask = None
+
+        if isinstance(other, BaseMaskedArray):
+            other, mask = other._data, other._mask
+
+        elif is_list_like(other):
+            other = np.asarray(other)
+            if other.ndim > 1:
+                raise NotImplementedError("can only perform ops with 1-d structures")
+            if len(self) != len(other):
+                raise ValueError("Lengths must match to compare")
+
+        if other is libmissing.NA:
+            # numpy does not handle pd.NA well as "other" scalar (it returns
+            # a scalar False instead of an array)
+            # This may be fixed by NA.__array_ufunc__. Revisit this check
+            # once that's implemented.
+            result = np.zeros(self._data.shape, dtype="bool")
+            mask = np.ones(self._data.shape, dtype="bool")
+        else:
+            with warnings.catch_warnings():
+                # numpy may show a FutureWarning or DeprecationWarning:
+                #     elementwise comparison failed; returning scalar instead,
+                #     but in the future will perform elementwise comparison
+                # before returning NotImplemented. We fall back to the correct
+                # behavior today, so that should be fine to ignore.
+                warnings.filterwarnings("ignore", "elementwise", FutureWarning)
+                warnings.filterwarnings("ignore", "elementwise", DeprecationWarning)
+                method = getattr(self._data, f"__{op.__name__}__")
+                result = method(other)
+
+                if result is NotImplemented:
+                    result = invalid_comparison(self._data, other, op)
+
+        mask = self._propagate_mask(mask, other)
+        return BooleanArray(result, mask, copy=False)
+
+    def _maybe_mask_result(
+        self, result: np.ndarray | tuple[np.ndarray, np.ndarray], mask: np.ndarray
+    ):
+        """
+        Parameters
+        ----------
+        result : array-like or tuple[array-like]
+        mask : array-like bool
+        """
+        if isinstance(result, tuple):
+            # i.e. divmod
+            div, mod = result
+            return (
+                self._maybe_mask_result(div, mask),
+                self._maybe_mask_result(mod, mask),
+            )
+
+        if result.dtype.kind == "f":
+            from pandas.core.arrays import FloatingArray
+
+            return FloatingArray(result, mask, copy=False)
+
+        elif result.dtype.kind == "b":
+            from pandas.core.arrays import BooleanArray
+
+            return BooleanArray(result, mask, copy=False)
+
+        elif lib.is_np_dtype(result.dtype, "m") and is_supported_dtype(result.dtype):
+            # e.g. test_numeric_arr_mul_tdscalar_numexpr_path
+            from pandas.core.arrays import TimedeltaArray
+
+            result[mask] = result.dtype.type("NaT")
+
+            if not isinstance(result, TimedeltaArray):
+                return TimedeltaArray._simple_new(result, dtype=result.dtype)
+
+            return result
+
+        elif result.dtype.kind in "iu":
+            from pandas.core.arrays import IntegerArray
+
+            return IntegerArray(result, mask, copy=False)
+
+        else:
+            result[mask] = np.nan
+            return result
+
+    def isna(self) -> np.ndarray:
+        return self._mask.copy()
+
+    @property
+    def _na_value(self):
+        return self.dtype.na_value
+
+    @property
+    def nbytes(self) -> int:
+        return self._data.nbytes + self._mask.nbytes
+
+    @classmethod
+    def _concat_same_type(
+        cls,
+        to_concat: Sequence[Self],
+        axis: AxisInt = 0,
+    ) -> Self:
+        data = np.concatenate([x._data for x in to_concat], axis=axis)
+        mask = np.concatenate([x._mask for x in to_concat], axis=axis)
+        return cls(data, mask)
+
+    def _hash_pandas_object(
+        self, *, encoding: str, hash_key: str, categorize: bool
+    ) -> npt.NDArray[np.uint64]:
+        hashed_array = hash_array(
+            self._data, encoding=encoding, hash_key=hash_key, categorize=categorize
+        )
+        hashed_array[self.isna()] = hash(self.dtype.na_value)
+        return hashed_array
+
+    def take(
+        self,
+        indexer,
+        *,
+        allow_fill: bool = False,
+        fill_value: Scalar | None = None,
+        axis: AxisInt = 0,
+    ) -> Self:
+        # we always fill with 1 internally
+        # to avoid upcasting
+        data_fill_value = (
+            self.dtype._internal_fill_value if isna(fill_value) else fill_value
+        )
+        result = take(
+            self._data,
+            indexer,
+            fill_value=data_fill_value,
+            allow_fill=allow_fill,
+            axis=axis,
+        )
+
+        mask = take(
+            self._mask, indexer, fill_value=True, allow_fill=allow_fill, axis=axis
+        )
+
+        # if we are filling
+        # we only fill where the indexer is null
+        # not existing missing values
+        # TODO(jreback) what if we have a non-na float as a fill value?
+        if allow_fill and notna(fill_value):
+            fill_mask = np.asarray(indexer) == -1
+            result[fill_mask] = fill_value
+            mask = mask ^ fill_mask
+
+        return self._simple_new(result, mask)
+
+    # error: Return type "BooleanArray" of "isin" incompatible with return type
+    # "ndarray" in supertype "ExtensionArray"
+    def isin(self, values: ArrayLike) -> BooleanArray:  # type: ignore[override]
+        from pandas.core.arrays import BooleanArray
+
+        # algorithms.isin will eventually convert values to an ndarray, so no extra
+        # cost to doing it here first
+        values_arr = np.asarray(values)
+        result = isin(self._data, values_arr)
+
+        if self._hasna:
+            values_have_NA = values_arr.dtype == object and any(
+                val is self.dtype.na_value for val in values_arr
+            )
+
+            # For now, NA does not propagate so set result according to presence of NA,
+            # see https://github.com/pandas-dev/pandas/pull/38379 for some discussion
+            result[self._mask] = values_have_NA
+
+        mask = np.zeros(self._data.shape, dtype=bool)
+        return BooleanArray(result, mask, copy=False)
+
+    def copy(self) -> Self:
+        data = self._data.copy()
+        mask = self._mask.copy()
+        return self._simple_new(data, mask)
+
+    @doc(ExtensionArray.duplicated)
+    def duplicated(
+        self, keep: Literal["first", "last", False] = "first"
+    ) -> npt.NDArray[np.bool_]:
+        values = self._data
+        mask = self._mask
+        return algos.duplicated(values, keep=keep, mask=mask)
+
+    def unique(self) -> Self:
+        """
+        Compute the BaseMaskedArray of unique values.
+
+        Returns
+        -------
+        uniques : BaseMaskedArray
+        """
+        uniques, mask = algos.unique_with_mask(self._data, self._mask)
+        return self._simple_new(uniques, mask)
+
+    @doc(ExtensionArray.searchsorted)
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if self._hasna:
+            raise ValueError(
+                "searchsorted requires array to be sorted, which is impossible "
+                "with NAs present."
+            )
+        if isinstance(value, ExtensionArray):
+            value = value.astype(object)
+        # Base class searchsorted would cast to object, which is *much* slower.
+        return self._data.searchsorted(value, side=side, sorter=sorter)
+
+    @doc(ExtensionArray.factorize)
+    def factorize(
+        self,
+        use_na_sentinel: bool = True,
+    ) -> tuple[np.ndarray, ExtensionArray]:
+        arr = self._data
+        mask = self._mask
+
+        # Use a sentinel for na; recode and add NA to uniques if necessary below
+        codes, uniques = factorize_array(arr, use_na_sentinel=True, mask=mask)
+
+        # check that factorize_array correctly preserves dtype.
+        assert uniques.dtype == self.dtype.numpy_dtype, (uniques.dtype, self.dtype)
+
+        has_na = mask.any()
+        if use_na_sentinel or not has_na:
+            size = len(uniques)
+        else:
+            # Make room for an NA value
+            size = len(uniques) + 1
+        uniques_mask = np.zeros(size, dtype=bool)
+        if not use_na_sentinel and has_na:
+            na_index = mask.argmax()
+            # Insert na with the proper code
+            if na_index == 0:
+                na_code = np.intp(0)
+            else:
+                na_code = codes[:na_index].max() + 1
+            codes[codes >= na_code] += 1
+            codes[codes == -1] = na_code
+            # dummy value for uniques; not used since uniques_mask will be True
+            uniques = np.insert(uniques, na_code, 0)
+            uniques_mask[na_code] = True
+        uniques_ea = self._simple_new(uniques, uniques_mask)
+
+        return codes, uniques_ea
+
+    @doc(ExtensionArray._values_for_argsort)
+    def _values_for_argsort(self) -> np.ndarray:
+        return self._data
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Returns a Series containing counts of each unique value.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of missing values.
+
+        Returns
+        -------
+        counts : Series
+
+        See Also
+        --------
+        Series.value_counts
+        """
+        from pandas import (
+            Index,
+            Series,
+        )
+        from pandas.arrays import IntegerArray
+
+        keys, value_counts, na_counter = algos.value_counts_arraylike(
+            self._data, dropna=dropna, mask=self._mask
+        )
+        mask_index = np.zeros((len(value_counts),), dtype=np.bool_)
+        mask = mask_index.copy()
+
+        if na_counter > 0:
+            mask_index[-1] = True
+
+        arr = IntegerArray(value_counts, mask)
+        index = Index(
+            self.dtype.construct_array_type()(
+                keys,  # type: ignore[arg-type]
+                mask_index,
+            )
+        )
+        return Series(arr, index=index, name="count", copy=False)
+
+    def _mode(self, dropna: bool = True) -> Self:
+        if dropna:
+            result = mode(self._data, dropna=dropna, mask=self._mask)
+            res_mask = np.zeros(result.shape, dtype=np.bool_)
+        else:
+            result, res_mask = mode(self._data, dropna=dropna, mask=self._mask)
+        result = type(self)(result, res_mask)  # type: ignore[arg-type]
+        return result[result.argsort()]
+
+    @doc(ExtensionArray.equals)
+    def equals(self, other) -> bool:
+        if type(self) != type(other):
+            return False
+        if other.dtype != self.dtype:
+            return False
+
+        # GH#44382 if e.g. self[1] is np.nan and other[1] is pd.NA, we are NOT
+        #  equal.
+        if not np.array_equal(self._mask, other._mask):
+            return False
+
+        left = self._data[~self._mask]
+        right = other._data[~other._mask]
+        return array_equivalent(left, right, strict_nan=True, dtype_equal=True)
+
+    def _quantile(
+        self, qs: npt.NDArray[np.float64], interpolation: str
+    ) -> BaseMaskedArray:
+        """
+        Dispatch to quantile_with_mask, needed because we do not have
+        _from_factorized.
+
+        Notes
+        -----
+        We assume that all impacted cases are 1D-only.
+        """
+        res = quantile_with_mask(
+            self._data,
+            mask=self._mask,
+            # TODO(GH#40932): na_value_for_dtype(self.dtype.numpy_dtype)
+            #  instead of np.nan
+            fill_value=np.nan,
+            qs=qs,
+            interpolation=interpolation,
+        )
+
+        if self._hasna:
+            # Our result mask is all-False unless we are all-NA, in which
+            #  case it is all-True.
+            if self.ndim == 2:
+                # I think this should be out_mask=self.isna().all(axis=1)
+                #  but am holding off until we have tests
+                raise NotImplementedError
+            if self.isna().all():
+                out_mask = np.ones(res.shape, dtype=bool)
+
+                if is_integer_dtype(self.dtype):
+                    # We try to maintain int dtype if possible for not all-na case
+                    # as well
+                    res = np.zeros(res.shape, dtype=self.dtype.numpy_dtype)
+            else:
+                out_mask = np.zeros(res.shape, dtype=bool)
+        else:
+            out_mask = np.zeros(res.shape, dtype=bool)
+        return self._maybe_mask_result(res, mask=out_mask)
+
+    # ------------------------------------------------------------------
+    # Reductions
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        if name in {"any", "all", "min", "max", "sum", "prod", "mean", "var", "std"}:
+            result = getattr(self, name)(skipna=skipna, **kwargs)
+        else:
+            # median, skew, kurt, sem
+            data = self._data
+            mask = self._mask
+            op = getattr(nanops, f"nan{name}")
+            axis = kwargs.pop("axis", None)
+            result = op(data, axis=axis, skipna=skipna, mask=mask, **kwargs)
+
+        if keepdims:
+            if isna(result):
+                return self._wrap_na_result(name=name, axis=0, mask_size=(1,))
+            else:
+                result = result.reshape(1)
+                mask = np.zeros(1, dtype=bool)
+                return self._maybe_mask_result(result, mask)
+
+        if isna(result):
+            return libmissing.NA
+        else:
+            return result
+
+    def _wrap_reduction_result(self, name: str, result, *, skipna, axis):
+        if isinstance(result, np.ndarray):
+            if skipna:
+                # we only retain mask for all-NA rows/columns
+                mask = self._mask.all(axis=axis)
+            else:
+                mask = self._mask.any(axis=axis)
+
+            return self._maybe_mask_result(result, mask)
+        return result
+
+    def _wrap_na_result(self, *, name, axis, mask_size):
+        mask = np.ones(mask_size, dtype=bool)
+
+        float_dtyp = "float32" if self.dtype == "Float32" else "float64"
+        if name in ["mean", "median", "var", "std", "skew", "kurt", "sem"]:
+            np_dtype = float_dtyp
+        elif name in ["min", "max"] or self.dtype.itemsize == 8:
+            np_dtype = self.dtype.numpy_dtype.name
+        else:
+            is_windows_or_32bit = is_platform_windows() or not IS64
+            int_dtyp = "int32" if is_windows_or_32bit else "int64"
+            uint_dtyp = "uint32" if is_windows_or_32bit else "uint64"
+            np_dtype = {"b": int_dtyp, "i": int_dtyp, "u": uint_dtyp, "f": float_dtyp}[
+                self.dtype.kind
+            ]
+
+        value = np.array([1], dtype=np_dtype)
+        return self._maybe_mask_result(value, mask=mask)
+
+    def _wrap_min_count_reduction_result(
+        self, name: str, result, *, skipna, min_count, axis
+    ):
+        if min_count == 0 and isinstance(result, np.ndarray):
+            return self._maybe_mask_result(result, np.zeros(result.shape, dtype=bool))
+        return self._wrap_reduction_result(name, result, skipna=skipna, axis=axis)
+
+    def sum(
+        self,
+        *,
+        skipna: bool = True,
+        min_count: int = 0,
+        axis: AxisInt | None = 0,
+        **kwargs,
+    ):
+        nv.validate_sum((), kwargs)
+
+        result = masked_reductions.sum(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            min_count=min_count,
+            axis=axis,
+        )
+        return self._wrap_min_count_reduction_result(
+            "sum", result, skipna=skipna, min_count=min_count, axis=axis
+        )
+
+    def prod(
+        self,
+        *,
+        skipna: bool = True,
+        min_count: int = 0,
+        axis: AxisInt | None = 0,
+        **kwargs,
+    ):
+        nv.validate_prod((), kwargs)
+
+        result = masked_reductions.prod(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            min_count=min_count,
+            axis=axis,
+        )
+        return self._wrap_min_count_reduction_result(
+            "prod", result, skipna=skipna, min_count=min_count, axis=axis
+        )
+
+    def mean(self, *, skipna: bool = True, axis: AxisInt | None = 0, **kwargs):
+        nv.validate_mean((), kwargs)
+        result = masked_reductions.mean(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            axis=axis,
+        )
+        return self._wrap_reduction_result("mean", result, skipna=skipna, axis=axis)
+
+    def var(
+        self, *, skipna: bool = True, axis: AxisInt | None = 0, ddof: int = 1, **kwargs
+    ):
+        nv.validate_stat_ddof_func((), kwargs, fname="var")
+        result = masked_reductions.var(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            axis=axis,
+            ddof=ddof,
+        )
+        return self._wrap_reduction_result("var", result, skipna=skipna, axis=axis)
+
+    def std(
+        self, *, skipna: bool = True, axis: AxisInt | None = 0, ddof: int = 1, **kwargs
+    ):
+        nv.validate_stat_ddof_func((), kwargs, fname="std")
+        result = masked_reductions.std(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            axis=axis,
+            ddof=ddof,
+        )
+        return self._wrap_reduction_result("std", result, skipna=skipna, axis=axis)
+
+    def min(self, *, skipna: bool = True, axis: AxisInt | None = 0, **kwargs):
+        nv.validate_min((), kwargs)
+        result = masked_reductions.min(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            axis=axis,
+        )
+        return self._wrap_reduction_result("min", result, skipna=skipna, axis=axis)
+
+    def max(self, *, skipna: bool = True, axis: AxisInt | None = 0, **kwargs):
+        nv.validate_max((), kwargs)
+        result = masked_reductions.max(
+            self._data,
+            self._mask,
+            skipna=skipna,
+            axis=axis,
+        )
+        return self._wrap_reduction_result("max", result, skipna=skipna, axis=axis)
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        return map_array(self.to_numpy(), mapper, na_action=na_action)
+
+    @overload
+    def any(
+        self, *, skipna: Literal[True] = ..., axis: AxisInt | None = ..., **kwargs
+    ) -> np.bool_: ...
+
+    @overload
+    def any(
+        self, *, skipna: bool, axis: AxisInt | None = ..., **kwargs
+    ) -> np.bool_ | NAType: ...
+
+    def any(
+        self, *, skipna: bool = True, axis: AxisInt | None = 0, **kwargs
+    ) -> np.bool_ | NAType:
+        """
+        Return whether any element is truthy.
+
+        Returns False unless there is at least one element that is truthy.
+        By default, NAs are skipped. If ``skipna=False`` is specified and
+        missing values are present, similar :ref:`Kleene logic <boolean.kleene>`
+        is used as for logical operations.
+
+        .. versionchanged:: 1.4.0
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values. If the entire array is NA and `skipna` is
+            True, then the result will be False, as for an empty array.
+            If `skipna` is False, the result will still be True if there is
+            at least one element that is truthy, otherwise NA will be returned
+            if there are NA's present.
+        axis : int, optional, default 0
+        **kwargs : any, default None
+            Additional keywords have no effect but might be accepted for
+            compatibility with NumPy.
+
+        Returns
+        -------
+        bool or :attr:`pandas.NA`
+
+        See Also
+        --------
+        numpy.any : Numpy version of this method.
+        BaseMaskedArray.all : Return whether all elements are truthy.
+
+        Examples
+        --------
+        The result indicates whether any element is truthy (and by default
+        skips NAs):
+
+        >>> pd.array([True, False, True]).any()
+        True
+        >>> pd.array([True, False, pd.NA]).any()
+        True
+        >>> pd.array([False, False, pd.NA]).any()
+        False
+        >>> pd.array([], dtype="boolean").any()
+        False
+        >>> pd.array([pd.NA], dtype="boolean").any()
+        False
+        >>> pd.array([pd.NA], dtype="Float64").any()
+        False
+
+        With ``skipna=False``, the result can be NA if this is logically
+        required (whether ``pd.NA`` is True or False influences the result):
+
+        >>> pd.array([True, False, pd.NA]).any(skipna=False)
+        True
+        >>> pd.array([1, 0, pd.NA]).any(skipna=False)
+        True
+        >>> pd.array([False, False, pd.NA]).any(skipna=False)
+        <NA>
+        >>> pd.array([0, 0, pd.NA]).any(skipna=False)
+        <NA>
+        """
+        nv.validate_any((), kwargs)
+
+        values = self._data.copy()
+        np.putmask(values, self._mask, self.dtype._falsey_value)
+        result = values.any()
+        if skipna:
+            return result
+        else:
+            if result or len(self) == 0 or not self._mask.any():
+                return result
+            else:
+                return self.dtype.na_value
+
+    @overload
+    def all(
+        self, *, skipna: Literal[True] = ..., axis: AxisInt | None = ..., **kwargs
+    ) -> np.bool_: ...
+
+    @overload
+    def all(
+        self, *, skipna: bool, axis: AxisInt | None = ..., **kwargs
+    ) -> np.bool_ | NAType: ...
+
+    def all(
+        self, *, skipna: bool = True, axis: AxisInt | None = 0, **kwargs
+    ) -> np.bool_ | NAType:
+        """
+        Return whether all elements are truthy.
+
+        Returns True unless there is at least one element that is falsey.
+        By default, NAs are skipped. If ``skipna=False`` is specified and
+        missing values are present, similar :ref:`Kleene logic <boolean.kleene>`
+        is used as for logical operations.
+
+        .. versionchanged:: 1.4.0
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values. If the entire array is NA and `skipna` is
+            True, then the result will be True, as for an empty array.
+            If `skipna` is False, the result will still be False if there is
+            at least one element that is falsey, otherwise NA will be returned
+            if there are NA's present.
+        axis : int, optional, default 0
+        **kwargs : any, default None
+            Additional keywords have no effect but might be accepted for
+            compatibility with NumPy.
+
+        Returns
+        -------
+        bool or :attr:`pandas.NA`
+
+        See Also
+        --------
+        numpy.all : Numpy version of this method.
+        BooleanArray.any : Return whether any element is truthy.
+
+        Examples
+        --------
+        The result indicates whether all elements are truthy (and by default
+        skips NAs):
+
+        >>> pd.array([True, True, pd.NA]).all()
+        True
+        >>> pd.array([1, 1, pd.NA]).all()
+        True
+        >>> pd.array([True, False, pd.NA]).all()
+        False
+        >>> pd.array([], dtype="boolean").all()
+        True
+        >>> pd.array([pd.NA], dtype="boolean").all()
+        True
+        >>> pd.array([pd.NA], dtype="Float64").all()
+        True
+
+        With ``skipna=False``, the result can be NA if this is logically
+        required (whether ``pd.NA`` is True or False influences the result):
+
+        >>> pd.array([True, True, pd.NA]).all(skipna=False)
+        <NA>
+        >>> pd.array([1, 1, pd.NA]).all(skipna=False)
+        <NA>
+        >>> pd.array([True, False, pd.NA]).all(skipna=False)
+        False
+        >>> pd.array([1, 0, pd.NA]).all(skipna=False)
+        False
+        """
+        nv.validate_all((), kwargs)
+
+        values = self._data.copy()
+        np.putmask(values, self._mask, self.dtype._truthy_value)
+        result = values.all(axis=axis)
+
+        if skipna:
+            return result
+        else:
+            if not result or len(self) == 0 or not self._mask.any():
+                return result
+            else:
+                return self.dtype.na_value
+
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        axis: int,
+        index,
+        limit,
+        limit_direction,
+        limit_area,
+        copy: bool,
+        **kwargs,
+    ) -> FloatingArray:
+        """
+        See NDFrame.interpolate.__doc__.
+        """
+        # NB: we return type(self) even if copy=False
+        if self.dtype.kind == "f":
+            if copy:
+                data = self._data.copy()
+                mask = self._mask.copy()
+            else:
+                data = self._data
+                mask = self._mask
+        elif self.dtype.kind in "iu":
+            copy = True
+            data = self._data.astype("f8")
+            mask = self._mask.copy()
+        else:
+            raise NotImplementedError(
+                f"interpolate is not implemented for dtype={self.dtype}"
+            )
+
+        missing.interpolate_2d_inplace(
+            data,
+            method=method,
+            axis=0,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            mask=mask,
+            **kwargs,
+        )
+        if not copy:
+            return self  # type: ignore[return-value]
+        if self.dtype.kind == "f":
+            return type(self)._simple_new(data, mask)  # type: ignore[return-value]
+        else:
+            from pandas.core.arrays import FloatingArray
+
+            return FloatingArray._simple_new(data, mask)
+
+    def _accumulate(
+        self, name: str, *, skipna: bool = True, **kwargs
+    ) -> BaseMaskedArray:
+        data = self._data
+        mask = self._mask
+
+        op = getattr(masked_accumulations, name)
+        data, mask = op(data, mask, skipna=skipna, **kwargs)
+
+        return self._simple_new(data, mask)
+
+    # ------------------------------------------------------------------
+    # GroupBy Methods
+
+    def _groupby_op(
+        self,
+        *,
+        how: str,
+        has_dropped_na: bool,
+        min_count: int,
+        ngroups: int,
+        ids: npt.NDArray[np.intp],
+        **kwargs,
+    ):
+        from pandas.core.groupby.ops import WrappedCythonOp
+
+        kind = WrappedCythonOp.get_kind_from_how(how)
+        op = WrappedCythonOp(how=how, kind=kind, has_dropped_na=has_dropped_na)
+
+        # libgroupby functions are responsible for NOT altering mask
+        mask = self._mask
+        if op.kind != "aggregate":
+            result_mask = mask.copy()
+        else:
+            result_mask = np.zeros(ngroups, dtype=bool)
+
+        if how == "rank" and kwargs.get("na_option") in ["top", "bottom"]:
+            result_mask[:] = False
+
+        res_values = op._cython_op_ndim_compat(
+            self._data,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=ids,
+            mask=mask,
+            result_mask=result_mask,
+            **kwargs,
+        )
+
+        if op.how == "ohlc":
+            arity = op._cython_arity.get(op.how, 1)
+            result_mask = np.tile(result_mask, (arity, 1)).T
+
+        if op.how in ["idxmin", "idxmax"]:
+            # Result values are indexes to take, keep as ndarray
+            return res_values
+        else:
+            # res_values should already have the correct dtype, we just need to
+            #  wrap in a MaskedArray
+            return self._maybe_mask_result(res_values, result_mask)
+
+
+def transpose_homogeneous_masked_arrays(
+    masked_arrays: Sequence[BaseMaskedArray],
+) -> list[BaseMaskedArray]:
+    """Transpose masked arrays in a list, but faster.
+
+    Input should be a list of 1-dim masked arrays of equal length and all have the
+    same dtype. The caller is responsible for ensuring validity of input data.
+    """
+    masked_arrays = list(masked_arrays)
+    dtype = masked_arrays[0].dtype
+
+    values = [arr._data.reshape(1, -1) for arr in masked_arrays]
+    transposed_values = np.concatenate(
+        values,
+        axis=0,
+        out=np.empty(
+            (len(masked_arrays), len(masked_arrays[0])),
+            order="F",
+            dtype=dtype.numpy_dtype,
+        ),
+    )
+
+    masks = [arr._mask.reshape(1, -1) for arr in masked_arrays]
+    transposed_masks = np.concatenate(
+        masks, axis=0, out=np.empty_like(transposed_values, dtype=bool)
+    )
+
+    arr_type = dtype.construct_array_type()
+    transposed_arrays: list[BaseMaskedArray] = []
+    for i in range(transposed_values.shape[1]):
+        transposed_arr = arr_type(transposed_values[:, i], mask=transposed_masks[:, i])
+        transposed_arrays.append(transposed_arr)
+
+    return transposed_arrays
diff --git a/pandas/pandas/core/arrays/numeric.py b/pandas/pandas/core/arrays/numeric.py
new file mode 100644
index 0000000000000..f319a3cc05575
--- /dev/null
+++ b/pandas/pandas/core/arrays/numeric.py
@@ -0,0 +1,295 @@
+from __future__ import annotations
+
+import numbers
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+
+from pandas.core.arrays.masked import (
+    BaseMaskedArray,
+    BaseMaskedDtype,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Mapping,
+    )
+
+    import pyarrow
+
+    from pandas._typing import (
+        DtypeObj,
+        Self,
+        npt,
+    )
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+
+class NumericDtype(BaseMaskedDtype):
+    _default_np_dtype: np.dtype
+    _checker: Callable[[Any], bool]  # is_foo_dtype
+
+    def __repr__(self) -> str:
+        return f"{self.name}Dtype()"
+
+    @cache_readonly
+    def is_signed_integer(self) -> bool:
+        return self.kind == "i"
+
+    @cache_readonly
+    def is_unsigned_integer(self) -> bool:
+        return self.kind == "u"
+
+    @property
+    def _is_numeric(self) -> bool:
+        return True
+
+    def __from_arrow__(
+        self, array: pyarrow.Array | pyarrow.ChunkedArray
+    ) -> BaseMaskedArray:
+        """
+        Construct IntegerArray/FloatingArray from pyarrow Array/ChunkedArray.
+        """
+        import pyarrow
+
+        from pandas.core.arrays.arrow._arrow_utils import (
+            pyarrow_array_to_numpy_and_mask,
+        )
+
+        array_class = self.construct_array_type()
+
+        pyarrow_type = pyarrow.from_numpy_dtype(self.type)
+        if not array.type.equals(pyarrow_type) and not pyarrow.types.is_null(
+            array.type
+        ):
+            # test_from_arrow_type_error raise for string, but allow
+            #  through itemsize conversion GH#31896
+            rt_dtype = pandas_dtype(array.type.to_pandas_dtype())
+            if rt_dtype.kind not in "iuf":
+                # Could allow "c" or potentially disallow float<->int conversion,
+                #  but at the moment we specifically test that uint<->int works
+                raise TypeError(
+                    f"Expected array of {self} type, got {array.type} instead"
+                )
+
+            array = array.cast(pyarrow_type)
+
+        if isinstance(array, pyarrow.ChunkedArray):
+            # TODO this "if" can be removed when requiring pyarrow >= 10.0, which fixed
+            # combine_chunks for empty arrays https://github.com/apache/arrow/pull/13757
+            if array.num_chunks == 0:
+                array = pyarrow.array([], type=array.type)
+            else:
+                array = array.combine_chunks()
+
+        data, mask = pyarrow_array_to_numpy_and_mask(array, dtype=self.numpy_dtype)
+        return array_class(data.copy(), ~mask, copy=False)
+
+    @classmethod
+    def _get_dtype_mapping(cls) -> Mapping[np.dtype, NumericDtype]:
+        raise AbstractMethodError(cls)
+
+    @classmethod
+    def _standardize_dtype(cls, dtype: NumericDtype | str | np.dtype) -> NumericDtype:
+        """
+        Convert a string representation or a numpy dtype to NumericDtype.
+        """
+        if isinstance(dtype, str) and (dtype.startswith(("Int", "UInt", "Float"))):
+            # Avoid DeprecationWarning from NumPy about np.dtype("Int64")
+            # https://github.com/numpy/numpy/pull/7476
+            dtype = dtype.lower()
+
+        if not isinstance(dtype, NumericDtype):
+            mapping = cls._get_dtype_mapping()
+            try:
+                dtype = mapping[np.dtype(dtype)]
+            except KeyError as err:
+                raise ValueError(f"invalid dtype specified {dtype}") from err
+        return dtype
+
+    @classmethod
+    def _safe_cast(cls, values: np.ndarray, dtype: np.dtype, copy: bool) -> np.ndarray:
+        """
+        Safely cast the values to the given dtype.
+
+        "safe" in this context means the casting is lossless.
+        """
+        raise AbstractMethodError(cls)
+
+
+def _coerce_to_data_and_mask(
+    values, dtype, copy: bool, dtype_cls: type[NumericDtype], default_dtype: np.dtype
+):
+    checker = dtype_cls._checker
+
+    mask = None
+    inferred_type = None
+
+    if dtype is None and hasattr(values, "dtype"):
+        if checker(values.dtype):
+            dtype = values.dtype
+
+    if dtype is not None:
+        dtype = dtype_cls._standardize_dtype(dtype)
+
+    cls = dtype_cls.construct_array_type()
+    if isinstance(values, cls):
+        values, mask = values._data, values._mask
+        if dtype is not None:
+            values = values.astype(dtype.numpy_dtype, copy=False)
+
+        if copy:
+            values = values.copy()
+            mask = mask.copy()
+        return values, mask, dtype, inferred_type
+
+    original = values
+    if not copy:
+        values = np.asarray(values)
+    else:
+        values = np.array(values, copy=copy)
+    inferred_type = None
+    if values.dtype == object or is_string_dtype(values.dtype):
+        inferred_type = lib.infer_dtype(values, skipna=True)
+        if inferred_type == "boolean" and dtype is None:
+            name = dtype_cls.__name__.strip("_")
+            raise TypeError(f"{values.dtype} cannot be converted to {name}")
+
+    elif values.dtype.kind == "b" and checker(dtype):
+        # fastpath
+        mask = np.zeros(len(values), dtype=np.bool_)
+        if not copy:
+            values = np.asarray(values, dtype=default_dtype)
+        else:
+            values = np.array(values, dtype=default_dtype, copy=copy)
+
+    elif values.dtype.kind not in "iuf":
+        name = dtype_cls.__name__.strip("_")
+        raise TypeError(f"{values.dtype} cannot be converted to {name}")
+
+    if values.ndim != 1:
+        raise TypeError("values must be a 1D list-like")
+
+    if mask is None:
+        if values.dtype.kind in "iu":
+            # fastpath
+            mask = np.zeros(len(values), dtype=np.bool_)
+        elif values.dtype.kind == "f":
+            # np.isnan is faster than is_numeric_na() for floats
+            # github issue: #60066
+            mask = np.isnan(values)
+        else:
+            mask = libmissing.is_numeric_na(values)
+    else:
+        assert len(mask) == len(values)
+
+    if mask.ndim != 1:
+        raise TypeError("mask must be a 1D list-like")
+
+    # infer dtype if needed
+    if dtype is None:
+        dtype = default_dtype
+    else:
+        dtype = dtype.numpy_dtype
+
+    if is_integer_dtype(dtype) and values.dtype.kind == "f" and len(values) > 0:
+        if mask.all():
+            values = np.ones(values.shape, dtype=dtype)
+        else:
+            idx = np.nanargmax(values)
+            if int(values[idx]) != original[idx]:
+                # We have ints that lost precision during the cast.
+                inferred_type = lib.infer_dtype(original, skipna=True)
+                if (
+                    inferred_type not in ["floating", "mixed-integer-float"]
+                    and not mask.any()
+                ):
+                    values = np.asarray(original, dtype=dtype)
+                else:
+                    values = np.asarray(original, dtype="object")
+
+    # we copy as need to coerce here
+    if mask.any():
+        values = values.copy()
+        values[mask] = dtype_cls._internal_fill_value
+    if inferred_type in ("string", "unicode"):
+        # casts from str are always safe since they raise
+        # a ValueError if the str cannot be parsed into a float
+        values = values.astype(dtype, copy=copy)
+    else:
+        values = dtype_cls._safe_cast(values, dtype, copy=False)
+
+    return values, mask, dtype, inferred_type
+
+
+class NumericArray(BaseMaskedArray):
+    """
+    Base class for IntegerArray and FloatingArray.
+    """
+
+    _dtype_cls: type[NumericDtype]
+
+    def __init__(
+        self, values: np.ndarray, mask: npt.NDArray[np.bool_], copy: bool = False
+    ) -> None:
+        checker = self._dtype_cls._checker
+        if not (isinstance(values, np.ndarray) and checker(values.dtype)):
+            descr = (
+                "floating"
+                if self._dtype_cls.kind == "f"  # type: ignore[comparison-overlap]
+                else "integer"
+            )
+            raise TypeError(
+                f"values should be {descr} numpy array. Use "
+                "the 'pd.array' function instead"
+            )
+        if values.dtype == np.float16:
+            # If we don't raise here, then accessing self.dtype would raise
+            raise TypeError("FloatingArray does not support np.float16 dtype.")
+
+        super().__init__(values, mask, copy=copy)
+
+    @cache_readonly
+    def dtype(self) -> NumericDtype:
+        mapping = self._dtype_cls._get_dtype_mapping()
+        return mapping[self._data.dtype]
+
+    @classmethod
+    def _coerce_to_array(
+        cls, value, *, dtype: DtypeObj, copy: bool = False
+    ) -> tuple[np.ndarray, np.ndarray]:
+        dtype_cls = cls._dtype_cls
+        default_dtype = dtype_cls._default_np_dtype
+        values, mask, _, _ = _coerce_to_data_and_mask(
+            value, dtype, copy, dtype_cls, default_dtype
+        )
+        return values, mask
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        from pandas.core.tools.numeric import to_numeric
+
+        scalars = to_numeric(strings, errors="raise", dtype_backend="numpy_nullable")
+        return cls._from_sequence(scalars, dtype=dtype, copy=copy)
+
+    _HANDLED_TYPES = (np.ndarray, numbers.Number)
diff --git a/pandas/pandas/core/arrays/numpy_.py b/pandas/pandas/core/arrays/numpy_.py
new file mode 100644
index 0000000000000..ac0823ed903b3
--- /dev/null
+++ b/pandas/pandas/core/arrays/numpy_.py
@@ -0,0 +1,567 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs import is_supported_dtype
+from pandas.compat.numpy import function as nv
+
+from pandas.core.dtypes.astype import astype_array
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.dtypes import NumpyEADtype
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import (
+    arraylike,
+    missing,
+    nanops,
+    ops,
+)
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+from pandas.core.strings.object_array import ObjectStringArrayMixin
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        AxisInt,
+        Dtype,
+        FillnaOptions,
+        InterpolateOptions,
+        NpDtype,
+        Scalar,
+        Self,
+        npt,
+    )
+
+    from pandas import Index
+
+
+# error: Definition of "_concat_same_type" in base class "NDArrayBacked" is
+# incompatible with definition in base class "ExtensionArray"
+class NumpyExtensionArray(  # type: ignore[misc]
+    OpsMixin,
+    NDArrayBackedExtensionArray,
+    ObjectStringArrayMixin,
+):
+    """
+    A pandas ExtensionArray for NumPy data.
+
+    This is mostly for internal compatibility, and is not especially
+    useful on its own.
+
+    Parameters
+    ----------
+    values : ndarray
+        The NumPy ndarray to wrap. Must be 1-dimensional.
+    copy : bool, default False
+        Whether to copy `values`.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    array : Create an array.
+    Series.to_numpy : Convert a Series to a NumPy array.
+
+    Examples
+    --------
+    >>> pd.arrays.NumpyExtensionArray(np.array([0, 1, 2, 3]))
+    <NumpyExtensionArray>
+    [0, 1, 2, 3]
+    Length: 4, dtype: int64
+    """
+
+    # If you're wondering why pd.Series(cls) doesn't put the array in an
+    # ExtensionBlock, search for `ABCNumpyExtensionArray`. We check for
+    # that _typ to ensure that users don't unnecessarily use EAs inside
+    # pandas internals, which turns off things like block consolidation.
+    _typ = "npy_extension"
+    __array_priority__ = 1000
+    _ndarray: np.ndarray
+    _dtype: NumpyEADtype
+    _internal_fill_value = np.nan
+
+    # ------------------------------------------------------------------------
+    # Constructors
+
+    def __init__(
+        self, values: np.ndarray | NumpyExtensionArray, copy: bool = False
+    ) -> None:
+        if isinstance(values, type(self)):
+            values = values._ndarray
+        if not isinstance(values, np.ndarray):
+            raise ValueError(
+                f"'values' must be a NumPy array, not {type(values).__name__}"
+            )
+
+        if values.ndim == 0:
+            # Technically we support 2, but do not advertise that fact.
+            raise ValueError("NumpyExtensionArray must be 1-dimensional.")
+
+        if copy:
+            values = values.copy()
+
+        dtype = NumpyEADtype(values.dtype)
+        super().__init__(values, dtype)
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> NumpyExtensionArray:
+        if isinstance(dtype, NumpyEADtype):
+            dtype = dtype._dtype
+
+        # error: Argument "dtype" to "asarray" has incompatible type
+        # "Union[ExtensionDtype, str, dtype[Any], dtype[floating[_64Bit]], Type[object],
+        # None]"; expected "Union[dtype[Any], None, type, _SupportsDType, str,
+        # Union[Tuple[Any, int], Tuple[Any, Union[int, Sequence[int]]], List[Any],
+        # _DTypeDict, Tuple[Any, Any]]]"
+        result = np.asarray(scalars, dtype=dtype)  # type: ignore[arg-type]
+        if (
+            result.ndim > 1
+            and not hasattr(scalars, "dtype")
+            and (dtype is None or dtype == object)
+        ):
+            # e.g. list-of-tuples
+            result = construct_1d_object_array_from_listlike(scalars)
+
+        if copy and result is scalars:
+            result = result.copy()
+        return cls(result)
+
+    # ------------------------------------------------------------------------
+    # Data
+
+    @property
+    def dtype(self) -> NumpyEADtype:
+        return self._dtype
+
+    # ------------------------------------------------------------------------
+    # NumPy Array Interface
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        if copy is not None:
+            # Note: branch avoids `copy=None` for NumPy 1.x support
+            return np.array(self._ndarray, dtype=dtype, copy=copy)
+        return np.asarray(self._ndarray, dtype=dtype)
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        # Lightly modified version of
+        # https://numpy.org/doc/stable/reference/generated/numpy.lib.mixins.NDArrayOperatorsMixin.html
+        # The primary modification is not boxing scalar return values
+        # in NumpyExtensionArray, since pandas' ExtensionArrays are 1-d.
+        out = kwargs.get("out", ())
+
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            # e.g. test_ufunc_unary
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                # e.g. tests.series.test_ufunc.TestNumpyReductions
+                return result
+
+        # Defer to the implementation of the ufunc on unwrapped values.
+        inputs = tuple(
+            x._ndarray if isinstance(x, NumpyExtensionArray) else x for x in inputs
+        )
+        if out:
+            kwargs["out"] = tuple(
+                x._ndarray if isinstance(x, NumpyExtensionArray) else x for x in out
+            )
+        result = getattr(ufunc, method)(*inputs, **kwargs)
+
+        if ufunc.nout > 1:
+            # multiple return values; re-box array-like results
+            return tuple(type(self)(x) for x in result)
+        elif method == "at":
+            # no return value
+            return None
+        elif method == "reduce":
+            if isinstance(result, np.ndarray):
+                # e.g. test_np_reduce_2d
+                return type(self)(result)
+
+            # e.g. test_np_max_nested_tuples
+            return result
+        else:
+            # one return value; re-box array-like results
+            return type(self)(result)
+
+    # ------------------------------------------------------------------------
+    # Pandas ExtensionArray Interface
+
+    def astype(self, dtype, copy: bool = True):
+        dtype = pandas_dtype(dtype)
+
+        if dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+
+        result = astype_array(self._ndarray, dtype=dtype, copy=copy)
+        return result
+
+    def isna(self) -> np.ndarray:
+        return isna(self._ndarray)
+
+    def _validate_scalar(self, fill_value):
+        if fill_value is None:
+            # Primarily for subclasses
+            fill_value = self.dtype.na_value
+        return fill_value
+
+    def _values_for_factorize(self) -> tuple[np.ndarray, float | None]:
+        if self.dtype.kind in "iub":
+            fv = None
+        else:
+            fv = np.nan
+        return self._ndarray, fv
+
+    # Base EA class (and all other EA classes) don't have limit_area keyword
+    # This can be removed here as well when the interpolate ffill/bfill method
+    # deprecation is enforced
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        """
+        ffill or bfill along axis=0.
+        """
+        if copy:
+            out_data = self._ndarray.copy()
+        else:
+            out_data = self._ndarray
+
+        meth = missing.clean_fill_method(method)
+        missing.pad_or_backfill_inplace(
+            out_data.T,
+            method=meth,
+            axis=0,
+            limit=limit,
+            limit_area=limit_area,
+        )
+
+        if not copy:
+            return self
+        return type(self)._simple_new(out_data, dtype=self.dtype)
+
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        axis: int,
+        index: Index,
+        limit,
+        limit_direction,
+        limit_area,
+        copy: bool,
+        **kwargs,
+    ) -> Self:
+        """
+        See NDFrame.interpolate.__doc__.
+        """
+        # NB: we return type(self) even if copy=False
+        if not self.dtype._is_numeric:
+            raise TypeError(f"Cannot interpolate with {self.dtype} dtype")
+
+        if not copy:
+            out_data = self._ndarray
+        else:
+            out_data = self._ndarray.copy()
+
+        # TODO: assert we have floating dtype?
+        missing.interpolate_2d_inplace(
+            out_data,
+            method=method,
+            axis=axis,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            **kwargs,
+        )
+        if not copy:
+            return self
+        return type(self)._simple_new(out_data, dtype=self.dtype)
+
+    # ------------------------------------------------------------------------
+    # Reductions
+
+    def any(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        out=None,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_any((), {"out": out, "keepdims": keepdims})
+        result = nanops.nanany(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def all(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        out=None,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_all((), {"out": out, "keepdims": keepdims})
+        result = nanops.nanall(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def min(
+        self, *, axis: AxisInt | None = None, skipna: bool = True, **kwargs
+    ) -> Scalar:
+        nv.validate_min((), kwargs)
+        result = nanops.nanmin(
+            values=self._ndarray, axis=axis, mask=self.isna(), skipna=skipna
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def max(
+        self, *, axis: AxisInt | None = None, skipna: bool = True, **kwargs
+    ) -> Scalar:
+        nv.validate_max((), kwargs)
+        result = nanops.nanmax(
+            values=self._ndarray, axis=axis, mask=self.isna(), skipna=skipna
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def sum(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        skipna: bool = True,
+        min_count: int = 0,
+        **kwargs,
+    ) -> Scalar:
+        nv.validate_sum((), kwargs)
+        result = nanops.nansum(
+            self._ndarray, axis=axis, skipna=skipna, min_count=min_count
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def prod(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        skipna: bool = True,
+        min_count: int = 0,
+        **kwargs,
+    ) -> Scalar:
+        nv.validate_prod((), kwargs)
+        result = nanops.nanprod(
+            self._ndarray, axis=axis, skipna=skipna, min_count=min_count
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def mean(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_mean((), {"dtype": dtype, "out": out, "keepdims": keepdims})
+        result = nanops.nanmean(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def median(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        out=None,
+        overwrite_input: bool = False,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_median(
+            (), {"out": out, "overwrite_input": overwrite_input, "keepdims": keepdims}
+        )
+        result = nanops.nanmedian(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def std(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        ddof: int = 1,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="std"
+        )
+        result = nanops.nanstd(self._ndarray, axis=axis, skipna=skipna, ddof=ddof)
+        return self._wrap_reduction_result(axis, result)
+
+    def var(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        ddof: int = 1,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="var"
+        )
+        result = nanops.nanvar(self._ndarray, axis=axis, skipna=skipna, ddof=ddof)
+        return self._wrap_reduction_result(axis, result)
+
+    def sem(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        ddof: int = 1,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="sem"
+        )
+        result = nanops.nansem(self._ndarray, axis=axis, skipna=skipna, ddof=ddof)
+        return self._wrap_reduction_result(axis, result)
+
+    def kurt(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="kurt"
+        )
+        result = nanops.nankurt(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    def skew(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="skew"
+        )
+        result = nanops.nanskew(self._ndarray, axis=axis, skipna=skipna)
+        return self._wrap_reduction_result(axis, result)
+
+    # ------------------------------------------------------------------------
+    # Additional Methods
+
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        mask = self.isna()
+        if na_value is not lib.no_default and mask.any():
+            result = self._ndarray.copy()
+            result[mask] = na_value
+        else:
+            result = self._ndarray
+
+        result = np.asarray(result, dtype=dtype)
+
+        if copy and result is self._ndarray:
+            result = result.copy()
+
+        return result
+
+    # ------------------------------------------------------------------------
+    # Ops
+
+    def __invert__(self) -> NumpyExtensionArray:
+        return type(self)(~self._ndarray)
+
+    def __neg__(self) -> NumpyExtensionArray:
+        return type(self)(-self._ndarray)
+
+    def __pos__(self) -> NumpyExtensionArray:
+        return type(self)(+self._ndarray)
+
+    def __abs__(self) -> NumpyExtensionArray:
+        return type(self)(abs(self._ndarray))
+
+    def _cmp_method(self, other, op):
+        if isinstance(other, NumpyExtensionArray):
+            other = other._ndarray
+
+        other = ops.maybe_prepare_scalar_for_op(other, (len(self),))
+        pd_op = ops.get_array_op(op)
+        other = ensure_wrapped_if_datetimelike(other)
+        result = pd_op(self._ndarray, other)
+
+        if op is divmod or op is ops.rdivmod:
+            a, b = result
+            if isinstance(a, np.ndarray):
+                # for e.g. op vs TimedeltaArray, we may already
+                #  have an ExtensionArray, in which case we do not wrap
+                return self._wrap_ndarray_result(a), self._wrap_ndarray_result(b)
+            return a, b
+
+        if isinstance(result, np.ndarray):
+            # for e.g. multiplication vs TimedeltaArray, we may already
+            #  have an ExtensionArray, in which case we do not wrap
+            return self._wrap_ndarray_result(result)
+        return result
+
+    _arith_method = _cmp_method
+
+    def _wrap_ndarray_result(self, result: np.ndarray):
+        # If we have timedelta64[ns] result, return a TimedeltaArray instead
+        #  of a NumpyExtensionArray
+        if result.dtype.kind == "m" and is_supported_dtype(result.dtype):
+            from pandas.core.arrays import TimedeltaArray
+
+            return TimedeltaArray._simple_new(result, dtype=result.dtype)
+        return type(self)(result)
diff --git a/pandas/pandas/core/arrays/period.py b/pandas/pandas/core/arrays/period.py
new file mode 100644
index 0000000000000..ae92e17332c76
--- /dev/null
+++ b/pandas/pandas/core/arrays/period.py
@@ -0,0 +1,1464 @@
+from __future__ import annotations
+
+from datetime import timedelta
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    TypeVar,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    algos as libalgos,
+    lib,
+)
+from pandas._libs.arrays import NDArrayBacked
+from pandas._libs.tslibs import (
+    BaseOffset,
+    NaT,
+    NaTType,
+    Timedelta,
+    add_overflowsafe,
+    astype_overflowsafe,
+    dt64arr_to_periodarr as c_dt64arr_to_periodarr,
+    get_unit_from_dtype,
+    iNaT,
+    parsing,
+    period as libperiod,
+    to_offset,
+)
+from pandas._libs.tslibs.dtypes import (
+    FreqGroup,
+    PeriodDtypeBase,
+)
+from pandas._libs.tslibs.fields import isleapyear_arr
+from pandas._libs.tslibs.offsets import (
+    Tick,
+    delta_to_tick,
+)
+from pandas._libs.tslibs.period import (
+    DIFFERENT_FREQ,
+    IncompatibleFrequency,
+    Period,
+    get_period_field_arr,
+    period_asfreq_arr,
+)
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+)
+
+from pandas.core.dtypes.common import (
+    ensure_object,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCPeriodIndex,
+    ABCSeries,
+    ABCTimedeltaArray,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.arrays import datetimelike as dtl
+import pandas.core.common as com
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        AnyArrayLike,
+        Dtype,
+        FillnaOptions,
+        NpDtype,
+        NumpySorter,
+        NumpyValueArrayLike,
+        Self,
+        npt,
+    )
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+    from pandas.core.arrays import (
+        DatetimeArray,
+        TimedeltaArray,
+    )
+    from pandas.core.arrays.base import ExtensionArray
+
+
+BaseOffsetT = TypeVar("BaseOffsetT", bound=BaseOffset)
+
+
+_shared_doc_kwargs = {
+    "klass": "PeriodArray",
+}
+
+
+def _field_accessor(name: str, docstring: str | None = None):
+    def f(self):
+        base = self.dtype._dtype_code
+        result = get_period_field_arr(name, self.asi8, base)
+        return result
+
+    f.__name__ = name
+    f.__doc__ = docstring
+    return property(f)
+
+
+# error: Definition of "_concat_same_type" in base class "NDArrayBacked" is
+# incompatible with definition in base class "ExtensionArray"
+class PeriodArray(dtl.DatelikeOps, libperiod.PeriodMixin):  # type: ignore[misc]
+    """
+    Pandas ExtensionArray for storing Period data.
+
+    Users should use :func:`~pandas.array` to create new instances.
+
+    Parameters
+    ----------
+    values : Union[PeriodArray, Series[period], ndarray[int], PeriodIndex]
+        The data to store. These should be arrays that can be directly
+        converted to ordinals without inference or copy (PeriodArray,
+        ndarray[int64]), or a box around such an array (Series[period],
+        PeriodIndex).
+    dtype : PeriodDtype, optional
+        A PeriodDtype instance from which to extract a `freq`. If both
+        `freq` and `dtype` are specified, then the frequencies must match.
+    copy : bool, default False
+        Whether to copy the ordinals before storing.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    Period: Represents a period of time.
+    PeriodIndex : Immutable Index for period data.
+    period_range: Create a fixed-frequency PeriodArray.
+    array: Construct a pandas array.
+
+    Notes
+    -----
+    There are two components to a PeriodArray
+
+    - ordinals : integer ndarray
+    - freq : pd.tseries.offsets.Offset
+
+    The values are physically stored as a 1-D ndarray of integers. These are
+    called "ordinals" and represent some kind of offset from a base.
+
+    The `freq` indicates the span covered by each element of the array.
+    All elements in the PeriodArray have the same `freq`.
+
+    Examples
+    --------
+    >>> pd.arrays.PeriodArray(pd.PeriodIndex(["2023-01-01", "2023-01-02"], freq="D"))
+    <PeriodArray>
+    ['2023-01-01', '2023-01-02']
+    Length: 2, dtype: period[D]
+    """
+
+    # array priority higher than numpy scalars
+    __array_priority__ = 1000
+    _typ = "periodarray"  # ABCPeriodArray
+    _internal_fill_value = np.int64(iNaT)
+    _recognized_scalars = (Period,)
+    _is_recognized_dtype = lambda x: isinstance(
+        x, PeriodDtype
+    )  # check_compatible_with checks freq match
+    _infer_matches = ("period",)
+
+    @property
+    def _scalar_type(self) -> type[Period]:
+        return Period
+
+    # Names others delegate to us
+    _other_ops: list[str] = []
+    _bool_ops: list[str] = ["is_leap_year"]
+    _object_ops: list[str] = ["start_time", "end_time", "freq"]
+    _field_ops: list[str] = [
+        "year",
+        "month",
+        "day",
+        "hour",
+        "minute",
+        "second",
+        "weekofyear",
+        "weekday",
+        "week",
+        "dayofweek",
+        "day_of_week",
+        "dayofyear",
+        "day_of_year",
+        "quarter",
+        "qyear",
+        "days_in_month",
+        "daysinmonth",
+    ]
+    _datetimelike_ops: list[str] = _field_ops + _object_ops + _bool_ops
+    _datetimelike_methods: list[str] = ["strftime", "to_timestamp", "asfreq"]
+
+    _dtype: PeriodDtype
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __init__(self, values, dtype: Dtype | None = None, copy: bool = False) -> None:
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+            if not isinstance(dtype, PeriodDtype):
+                raise ValueError(f"Invalid dtype {dtype} for PeriodArray")
+
+        if isinstance(values, ABCSeries):
+            values = values._values
+            if not isinstance(values, type(self)):
+                raise TypeError("Incorrect dtype")
+
+        elif isinstance(values, ABCPeriodIndex):
+            values = values._values
+
+        if isinstance(values, type(self)):
+            if dtype is not None and dtype != values.dtype:
+                raise raise_on_incompatible(values, dtype.freq)
+            values, dtype = values._ndarray, values.dtype
+
+        if not copy:
+            values = np.asarray(values, dtype="int64")
+        else:
+            values = np.array(values, dtype="int64", copy=copy)
+        if dtype is None:
+            raise ValueError("dtype is not specified and cannot be inferred")
+        dtype = cast(PeriodDtype, dtype)
+        NDArrayBacked.__init__(self, values, dtype)
+
+    # error: Signature of "_simple_new" incompatible with supertype "NDArrayBacked"
+    @classmethod
+    def _simple_new(  # type: ignore[override]
+        cls,
+        values: npt.NDArray[np.int64],
+        dtype: PeriodDtype,
+    ) -> Self:
+        # alias for PeriodArray.__init__
+        assertion_msg = "Should be numpy array of type i8"
+        assert isinstance(values, np.ndarray) and values.dtype == "i8", assertion_msg
+        return cls(values, dtype=dtype)
+
+    @classmethod
+    def _from_sequence(
+        cls,
+        scalars,
+        *,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+    ) -> Self:
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+        if dtype and isinstance(dtype, PeriodDtype):
+            freq = dtype.freq
+        else:
+            freq = None
+
+        if isinstance(scalars, cls):
+            validate_dtype_freq(scalars.dtype, freq)
+            if copy:
+                scalars = scalars.copy()
+            return scalars
+
+        periods = np.asarray(scalars, dtype=object)
+
+        freq = freq or libperiod.extract_freq(periods)
+        ordinals = libperiod.extract_ordinals(periods, freq)
+        dtype = PeriodDtype(freq)
+        return cls(ordinals, dtype=dtype)
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        return cls._from_sequence(strings, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _from_datetime64(cls, data, freq, tz=None) -> Self:
+        """
+        Construct a PeriodArray from a datetime64 array
+
+        Parameters
+        ----------
+        data : ndarray[datetime64[ns], datetime64[ns, tz]]
+        freq : str or Tick
+        tz : tzinfo, optional
+
+        Returns
+        -------
+        PeriodArray[freq]
+        """
+        if isinstance(freq, BaseOffset):
+            freq = PeriodDtype(freq)._freqstr
+        data, freq = dt64arr_to_periodarr(data, freq, tz)
+        dtype = PeriodDtype(freq)
+        return cls(data, dtype=dtype)
+
+    @classmethod
+    def _generate_range(cls, start, end, periods, freq):
+        periods = dtl.validate_periods(periods)
+
+        if freq is not None:
+            freq = Period._maybe_convert_freq(freq)
+
+        if start is not None or end is not None:
+            subarr, freq = _get_ordinal_range(start, end, periods, freq)
+        else:
+            raise ValueError("Not enough parameters to construct Period range")
+
+        return subarr, freq
+
+    @classmethod
+    def _from_fields(cls, *, fields: dict, freq) -> Self:
+        subarr, freq = _range_from_fields(freq=freq, **fields)
+        dtype = PeriodDtype(freq)
+        return cls._simple_new(subarr, dtype=dtype)
+
+    # -----------------------------------------------------------------
+    # DatetimeLike Interface
+
+    # error: Argument 1 of "_unbox_scalar" is incompatible with supertype
+    # "DatetimeLikeArrayMixin"; supertype defines the argument type as
+    # "Union[Union[Period, Any, Timedelta], NaTType]"
+    def _unbox_scalar(  # type: ignore[override]
+        self,
+        value: Period | NaTType,
+    ) -> np.int64:
+        if value is NaT:
+            # error: Item "Period" of "Union[Period, NaTType]" has no attribute "value"
+            return np.int64(value._value)  # type: ignore[union-attr]
+        elif isinstance(value, self._scalar_type):
+            self._check_compatible_with(value)
+            return np.int64(value.ordinal)
+        else:
+            raise ValueError(f"'value' should be a Period. Got '{value}' instead.")
+
+    def _scalar_from_string(self, value: str) -> Period:
+        return Period(value, freq=self.freq)
+
+    # error: Argument 1 of "_check_compatible_with" is incompatible with
+    # supertype "DatetimeLikeArrayMixin"; supertype defines the argument type
+    # as "Period | Timestamp | Timedelta | NaTType"
+    def _check_compatible_with(self, other: Period | NaTType | PeriodArray) -> None:  # type: ignore[override]
+        if other is NaT:
+            return
+        # error: Item "NaTType" of "Period | NaTType | PeriodArray" has no
+        # attribute "freq"
+        self._require_matching_freq(other.freq)  # type: ignore[union-attr]
+
+    # --------------------------------------------------------------------
+    # Data / Attributes
+
+    @cache_readonly
+    def dtype(self) -> PeriodDtype:
+        return self._dtype
+
+    # error: Cannot override writeable attribute with read-only property
+    @property  # type: ignore[override]
+    def freq(self) -> BaseOffset:
+        """
+        Return the frequency object for this PeriodArray.
+        """
+        return self.dtype.freq
+
+    @property
+    def freqstr(self) -> str:
+        return PeriodDtype(self.freq)._freqstr
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        if dtype == "i8":
+            # For NumPy 1.x compatibility we cannot use copy=None.  And
+            # `copy=False` has the meaning of `copy=None` here:
+            if not copy:
+                return np.asarray(self.asi8, dtype=dtype)
+            else:
+                return np.array(self.asi8, dtype=dtype)
+
+        if copy is False:
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        if dtype == bool:
+            return ~self._isnan
+
+        # This will raise TypeError for non-object dtypes
+        return np.array(list(self), dtype=object)
+
+    def __arrow_array__(self, type=None):
+        """
+        Convert myself into a pyarrow Array.
+        """
+        import pyarrow
+
+        from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+        if type is not None:
+            if pyarrow.types.is_integer(type):
+                return pyarrow.array(self._ndarray, mask=self.isna(), type=type)
+            elif isinstance(type, ArrowPeriodType):
+                # ensure we have the same freq
+                if self.freqstr != type.freq:
+                    raise TypeError(
+                        "Not supported to convert PeriodArray to array with different "
+                        f"'freq' ({self.freqstr} vs {type.freq})"
+                    )
+            else:
+                raise TypeError(
+                    f"Not supported to convert PeriodArray to '{type}' type"
+                )
+
+        period_type = ArrowPeriodType(self.freqstr)
+        storage_array = pyarrow.array(self._ndarray, mask=self.isna(), type="int64")
+        return pyarrow.ExtensionArray.from_storage(period_type, storage_array)
+
+    # --------------------------------------------------------------------
+    # Vectorized analogues of Period properties
+
+    year = _field_accessor(
+        "year",
+        """
+        The year of the period.
+
+        See Also
+        --------
+        PeriodIndex.day_of_year : The ordinal day of the year.
+        PeriodIndex.dayofyear : The ordinal day of the year.
+        PeriodIndex.is_leap_year : Logical indicating if the date belongs to a
+            leap year.
+        PeriodIndex.weekofyear : The week ordinal of the year.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023", "2024", "2025"], freq="Y")
+        >>> idx.year
+        Index([2023, 2024, 2025], dtype='int64')
+        """,
+    )
+    month = _field_accessor(
+        "month",
+        """
+        The month as January=1, December=12.
+
+        See Also
+        --------
+        PeriodIndex.days_in_month : The number of days in the month.
+        PeriodIndex.daysinmonth : The number of days in the month.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.month
+        Index([1, 2, 3], dtype='int64')
+        """,
+    )
+    day = _field_accessor(
+        "day",
+        """
+        The days of the period.
+
+        See Also
+        --------
+        PeriodIndex.day_of_week : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.day_of_year : The ordinal day of the year.
+        PeriodIndex.dayofweek : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.dayofyear : The ordinal day of the year.
+        PeriodIndex.days_in_month : The number of days in the month.
+        PeriodIndex.daysinmonth : The number of days in the month.
+        PeriodIndex.weekday : The day of the week with Monday=0, Sunday=6.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(['2020-01-31', '2020-02-28'], freq='D')
+        >>> idx.day
+        Index([31, 28], dtype='int64')
+        """,
+    )
+    hour = _field_accessor(
+        "hour",
+        """
+        The hour of the period.
+
+        See Also
+        --------
+        PeriodIndex.minute : The minute of the period.
+        PeriodIndex.second : The second of the period.
+        PeriodIndex.to_timestamp : Cast to DatetimeArray/Index.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01-01 10:00", "2023-01-01 11:00"], freq='h')
+        >>> idx.hour
+        Index([10, 11], dtype='int64')
+        """,
+    )
+    minute = _field_accessor(
+        "minute",
+        """
+        The minute of the period.
+
+        See Also
+        --------
+        PeriodIndex.hour : The hour of the period.
+        PeriodIndex.second : The second of the period.
+        PeriodIndex.to_timestamp : Cast to DatetimeArray/Index.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01-01 10:30:00",
+        ...                       "2023-01-01 11:50:00"], freq='min')
+        >>> idx.minute
+        Index([30, 50], dtype='int64')
+        """,
+    )
+    second = _field_accessor(
+        "second",
+        """
+        The second of the period.
+
+        See Also
+        --------
+        PeriodIndex.hour : The hour of the period.
+        PeriodIndex.minute : The minute of the period.
+        PeriodIndex.to_timestamp : Cast to DatetimeArray/Index.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01-01 10:00:30",
+        ...                       "2023-01-01 10:00:31"], freq='s')
+        >>> idx.second
+        Index([30, 31], dtype='int64')
+        """,
+    )
+    weekofyear = _field_accessor(
+        "week",
+        """
+        The week ordinal of the year.
+
+        See Also
+        --------
+        PeriodIndex.day_of_week : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.dayofweek : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.week : The week ordinal of the year.
+        PeriodIndex.weekday : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.week  # It can be written `weekofyear`
+        Index([5, 9, 13], dtype='int64')
+        """,
+    )
+    week = weekofyear
+    day_of_week = _field_accessor(
+        "day_of_week",
+        """
+        The day of the week with Monday=0, Sunday=6.
+
+        See Also
+        --------
+        PeriodIndex.day : The days of the period.
+        PeriodIndex.day_of_week : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.day_of_year : The ordinal day of the year.
+        PeriodIndex.dayofweek : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.dayofyear : The ordinal day of the year.
+        PeriodIndex.week : The week ordinal of the year.
+        PeriodIndex.weekday : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.weekofyear : The week ordinal of the year.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01-01", "2023-01-02", "2023-01-03"], freq="D")
+        >>> idx.weekday
+        Index([6, 0, 1], dtype='int64')
+        """,
+    )
+    dayofweek = day_of_week
+    weekday = dayofweek
+    dayofyear = day_of_year = _field_accessor(
+        "day_of_year",
+        """
+        The ordinal day of the year.
+
+        See Also
+        --------
+        PeriodIndex.day : The days of the period.
+        PeriodIndex.day_of_week : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.day_of_year : The ordinal day of the year.
+        PeriodIndex.dayofweek : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.dayofyear : The ordinal day of the year.
+        PeriodIndex.weekday : The day of the week with Monday=0, Sunday=6.
+        PeriodIndex.weekofyear : The week ordinal of the year.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01-10", "2023-02-01", "2023-03-01"], freq="D")
+        >>> idx.dayofyear
+        Index([10, 32, 60], dtype='int64')
+
+        >>> idx = pd.PeriodIndex(["2023", "2024", "2025"], freq="Y")
+        >>> idx
+        PeriodIndex(['2023', '2024', '2025'], dtype='period[Y-DEC]')
+        >>> idx.dayofyear
+        Index([365, 366, 365], dtype='int64')
+        """,
+    )
+    quarter = _field_accessor(
+        "quarter",
+        """
+        The quarter of the date.
+
+        See Also
+        --------
+        PeriodIndex.qyear : Fiscal year the Period lies in according to its
+            starting-quarter.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.quarter
+        Index([1, 1, 1], dtype='int64')
+        """,
+    )
+    qyear = _field_accessor(
+        "qyear",
+        """
+        Fiscal year the Period lies in according to its starting-quarter.
+
+        The `year` and the `qyear` of the period will be the same if the fiscal
+        and calendar years are the same. When they are not, the fiscal year
+        can be different from the calendar year of the period.
+
+        Returns
+        -------
+        int
+            The fiscal year of the period.
+
+        See Also
+        --------
+        PeriodIndex.quarter : The quarter of the date.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        If the natural and fiscal year are the same, `qyear` and `year` will
+        be the same.
+
+        >>> per = pd.Period('2018Q1', freq='Q')
+        >>> per.qyear
+        2018
+        >>> per.year
+        2018
+
+        If the fiscal year starts in April (`Q-MAR`), the first quarter of
+        2018 will start in April 2017. `year` will then be 2017, but `qyear`
+        will be the fiscal year, 2018.
+
+        >>> per = pd.Period('2018Q1', freq='Q-MAR')
+        >>> per.start_time
+        Timestamp('2017-04-01 00:00:00')
+        >>> per.qyear
+        2018
+        >>> per.year
+        2017
+        """,
+    )
+
+    days_in_month = _field_accessor(
+        "days_in_month",
+        """
+        The number of days in the month.
+
+        See Also
+        --------
+        PeriodIndex.day : The days of the period.
+        PeriodIndex.days_in_month : The number of days in the month.
+        PeriodIndex.daysinmonth : The number of days in the month.
+        PeriodIndex.month : The month as January=1, December=12.
+
+        Examples
+        --------
+        For Series:
+
+        >>> period = pd.period_range('2020-1-1 00:00', '2020-3-1 00:00', freq='M')
+        >>> s = pd.Series(period)
+        >>> s
+        0   2020-01
+        1   2020-02
+        2   2020-03
+        dtype: period[M]
+        >>> s.dt.days_in_month
+        0    31
+        1    29
+        2    31
+        dtype: int64
+
+        For PeriodIndex:
+
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.days_in_month   # It can be also entered as `daysinmonth`
+        Index([31, 28, 31], dtype='int64')
+        """,
+    )
+    daysinmonth = days_in_month
+
+    @property
+    def is_leap_year(self) -> npt.NDArray[np.bool_]:
+        """
+        Logical indicating if the date belongs to a leap year.
+
+        See Also
+        --------
+        PeriodIndex.qyear : Fiscal year the Period lies in according to its
+            starting-quarter.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023", "2024", "2025"], freq="Y")
+        >>> idx.is_leap_year
+        array([False,  True, False])
+        """
+        return isleapyear_arr(np.asarray(self.year))
+
+    def to_timestamp(self, freq=None, how: str = "start") -> DatetimeArray:
+        """
+        Cast to DatetimeArray/Index.
+
+        Parameters
+        ----------
+        freq : str or DateOffset, optional
+            Target frequency. The default is 'D' for week or longer,
+            's' otherwise.
+        how : {'s', 'e', 'start', 'end'}
+            Whether to use the start or end of the time period being converted.
+
+        Returns
+        -------
+        DatetimeArray/Index
+            Timestamp representation of given Period-like object.
+
+        See Also
+        --------
+        PeriodIndex.day : The days of the period.
+        PeriodIndex.from_fields : Construct a PeriodIndex from fields
+            (year, month, day, etc.).
+        PeriodIndex.from_ordinals : Construct a PeriodIndex from ordinals.
+        PeriodIndex.hour : The hour of the period.
+        PeriodIndex.minute : The minute of the period.
+        PeriodIndex.month : The month as January=1, December=12.
+        PeriodIndex.second : The second of the period.
+        PeriodIndex.year : The year of the period.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02", "2023-03"], freq="M")
+        >>> idx.to_timestamp()
+        DatetimeIndex(['2023-01-01', '2023-02-01', '2023-03-01'],
+        dtype='datetime64[ns]', freq='MS')
+
+        The frequency will not be inferred if the index contains less than
+        three elements, or if the values of index are not strictly monotonic:
+
+        >>> idx = pd.PeriodIndex(["2023-01", "2023-02"], freq="M")
+        >>> idx.to_timestamp()
+        DatetimeIndex(['2023-01-01', '2023-02-01'], dtype='datetime64[ns]', freq=None)
+
+        >>> idx = pd.PeriodIndex(
+        ...     ["2023-01", "2023-02", "2023-02", "2023-03"], freq="2M"
+        ... )
+        >>> idx.to_timestamp()
+        DatetimeIndex(['2023-01-01', '2023-02-01', '2023-02-01', '2023-03-01'],
+        dtype='datetime64[ns]', freq=None)
+        """
+        from pandas.core.arrays import DatetimeArray
+
+        how = libperiod.validate_end_alias(how)
+
+        end = how == "E"
+        if end:
+            if freq == "B" or self.freq == "B":
+                # roll forward to ensure we land on B date
+                adjust = Timedelta(1, "D") - Timedelta(1, "ns")
+                return self.to_timestamp(how="start") + adjust
+            else:
+                adjust = Timedelta(1, "ns")
+                return (self + self.freq).to_timestamp(how="start") - adjust
+
+        if freq is None:
+            freq_code = self._dtype._get_to_timestamp_base()
+            dtype = PeriodDtypeBase(freq_code, 1)
+            freq = dtype._freqstr
+            base = freq_code
+        else:
+            freq = Period._maybe_convert_freq(freq)
+            base = freq._period_dtype_code
+
+        new_parr = self.asfreq(freq, how=how)
+
+        new_data = libperiod.periodarr_to_dt64arr(new_parr.asi8, base)
+        dta = DatetimeArray._from_sequence(new_data, dtype=np.dtype("M8[ns]"))
+
+        if self.freq.name == "B":
+            # See if we can retain BDay instead of Day in cases where
+            #  len(self) is too small for infer_freq to distinguish between them
+            diffs = libalgos.unique_deltas(self.asi8)
+            if len(diffs) == 1:
+                diff = diffs[0]
+                if diff == self.dtype._n:
+                    dta._freq = self.freq
+                elif diff == 1:
+                    dta._freq = self.freq.base
+                # TODO: other cases?
+            return dta
+        else:
+            return dta._with_freq("infer")
+
+    # --------------------------------------------------------------------
+
+    def _box_func(self, x) -> Period | NaTType:
+        return Period._from_ordinal(ordinal=x, freq=self.freq)
+
+    @doc(**_shared_doc_kwargs, other="PeriodIndex", other_name="PeriodIndex")
+    def asfreq(self, freq=None, how: str = "E") -> Self:
+        """
+        Convert the {klass} to the specified frequency `freq`.
+
+        Equivalent to applying :meth:`pandas.Period.asfreq` with the given arguments
+        to each :class:`~pandas.Period` in this {klass}.
+
+        Parameters
+        ----------
+        freq : str
+            A frequency.
+        how : str {{'E', 'S'}}, default 'E'
+            Whether the elements should be aligned to the end
+            or start within pa period.
+
+            * 'E', 'END', or 'FINISH' for end,
+            * 'S', 'START', or 'BEGIN' for start.
+
+            January 31st ('END') vs. January 1st ('START') for example.
+
+        Returns
+        -------
+        {klass}
+            The transformed {klass} with the new frequency.
+
+        See Also
+        --------
+        {other}.asfreq: Convert each Period in a {other_name} to the given frequency.
+        Period.asfreq : Convert a :class:`~pandas.Period` object to the given frequency.
+
+        Examples
+        --------
+        >>> pidx = pd.period_range("2010-01-01", "2015-01-01", freq="Y")
+        >>> pidx
+        PeriodIndex(['2010', '2011', '2012', '2013', '2014', '2015'],
+        dtype='period[Y-DEC]')
+
+        >>> pidx.asfreq("M")
+        PeriodIndex(['2010-12', '2011-12', '2012-12', '2013-12', '2014-12',
+        '2015-12'], dtype='period[M]')
+
+        >>> pidx.asfreq("M", how="S")
+        PeriodIndex(['2010-01', '2011-01', '2012-01', '2013-01', '2014-01',
+        '2015-01'], dtype='period[M]')
+        """
+        how = libperiod.validate_end_alias(how)
+        if isinstance(freq, BaseOffset) and hasattr(freq, "_period_dtype_code"):
+            freq = PeriodDtype(freq)._freqstr
+        freq = Period._maybe_convert_freq(freq)
+
+        base1 = self._dtype._dtype_code
+        base2 = freq._period_dtype_code
+
+        asi8 = self.asi8
+        # self.freq.n can't be negative or 0
+        end = how == "E"
+        if end:
+            ordinal = asi8 + self.dtype._n - 1
+        else:
+            ordinal = asi8
+
+        new_data = period_asfreq_arr(ordinal, base1, base2, end)
+
+        if self._hasna:
+            new_data[self._isnan] = iNaT
+
+        dtype = PeriodDtype(freq)
+        return type(self)(new_data, dtype=dtype)
+
+    # ------------------------------------------------------------------
+    # Rendering Methods
+
+    def _formatter(self, boxed: bool = False) -> Callable[[object], str]:
+        if boxed:
+            return str
+        return "'{}'".format
+
+    def _format_native_types(
+        self, *, na_rep: str | float = "NaT", date_format=None, **kwargs
+    ) -> npt.NDArray[np.object_]:
+        """
+        actually format my specific types
+        """
+        return libperiod.period_array_strftime(
+            self.asi8, self.dtype._dtype_code, na_rep, date_format
+        )
+
+    # ------------------------------------------------------------------
+
+    def astype(self, dtype, copy: bool = True):
+        # We handle Period[T] -> Period[U]
+        # Our parent handles everything else.
+        dtype = pandas_dtype(dtype)
+        if dtype == self._dtype:
+            if not copy:
+                return self
+            else:
+                return self.copy()
+        if isinstance(dtype, PeriodDtype):
+            return self.asfreq(dtype.freq)
+
+        if lib.is_np_dtype(dtype, "M") or isinstance(dtype, DatetimeTZDtype):
+            # GH#45038 match PeriodIndex behavior.
+            tz = getattr(dtype, "tz", None)
+            unit = dtl.dtype_to_unit(dtype)
+            return self.to_timestamp().tz_localize(tz).as_unit(unit)
+
+        return super().astype(dtype, copy=copy)
+
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        npvalue = self._validate_setitem_value(value).view("M8[ns]")
+
+        # Cast to M8 to get datetime-like NaT placement,
+        #  similar to dtl._period_dispatch
+        m8arr = self._ndarray.view("M8[ns]")
+        return m8arr.searchsorted(npvalue, side=side, sorter=sorter)
+
+    def _pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        copy: bool = True,
+    ) -> Self:
+        # view as dt64 so we get treated as timelike in core.missing,
+        #  similar to dtl._period_dispatch
+        dta = self.view("M8[ns]")
+        result = dta._pad_or_backfill(
+            method=method, limit=limit, limit_area=limit_area, copy=copy
+        )
+        if copy:
+            return cast("Self", result.view(self.dtype))
+        else:
+            return self
+
+    # ------------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _addsub_int_array_or_scalar(
+        self, other: np.ndarray | int, op: Callable[[Any, Any], Any]
+    ) -> Self:
+        """
+        Add or subtract array of integers.
+
+        Parameters
+        ----------
+        other : np.ndarray[int64] or int
+        op : {operator.add, operator.sub}
+
+        Returns
+        -------
+        result : PeriodArray
+        """
+        assert op in [operator.add, operator.sub]
+        if op is operator.sub:
+            other = -other
+        res_values = add_overflowsafe(self.asi8, np.asarray(other, dtype="i8"))
+        return type(self)(res_values, dtype=self.dtype)
+
+    def _add_offset(self, other: BaseOffset):
+        assert not isinstance(other, Tick)
+
+        self._require_matching_freq(other, base=True)
+        return self._addsub_int_array_or_scalar(other.n, operator.add)
+
+    # TODO: can we de-duplicate with Period._add_timedeltalike_scalar?
+    def _add_timedeltalike_scalar(self, other):
+        """
+        Parameters
+        ----------
+        other : timedelta, Tick, np.timedelta64
+
+        Returns
+        -------
+        PeriodArray
+        """
+        if not isinstance(self.freq, Tick):
+            # We cannot add timedelta-like to non-tick PeriodArray
+            raise raise_on_incompatible(self, other)
+
+        if isna(other):
+            # i.e. np.timedelta64("NaT")
+            return super()._add_timedeltalike_scalar(other)
+
+        td = np.asarray(Timedelta(other).asm8)
+        return self._add_timedelta_arraylike(td)
+
+    def _add_timedelta_arraylike(
+        self, other: TimedeltaArray | npt.NDArray[np.timedelta64]
+    ) -> Self:
+        """
+        Parameters
+        ----------
+        other : TimedeltaArray or ndarray[timedelta64]
+
+        Returns
+        -------
+        PeriodArray
+        """
+        if not self.dtype._is_tick_like():
+            # We cannot add timedelta-like to non-tick PeriodArray
+            raise TypeError(
+                f"Cannot add or subtract timedelta64[ns] dtype from {self.dtype}"
+            )
+
+        dtype = np.dtype(f"m8[{self.dtype._td64_unit}]")
+
+        # Similar to _check_timedeltalike_freq_compat, but we raise with a
+        #  more specific exception message if necessary.
+        try:
+            delta = astype_overflowsafe(
+                np.asarray(other), dtype=dtype, copy=False, round_ok=False
+            )
+        except ValueError as err:
+            # e.g. if we have minutes freq and try to add 30s
+            # "Cannot losslessly convert units"
+            raise IncompatibleFrequency(
+                "Cannot add/subtract timedelta-like from PeriodArray that is "
+                "not an integer multiple of the PeriodArray's freq."
+            ) from err
+
+        res_values = add_overflowsafe(self.asi8, np.asarray(delta.view("i8")))
+        return type(self)(res_values, dtype=self.dtype)
+
+    def _check_timedeltalike_freq_compat(self, other):
+        """
+        Arithmetic operations with timedelta-like scalars or array `other`
+        are only valid if `other` is an integer multiple of `self.freq`.
+        If the operation is valid, find that integer multiple.  Otherwise,
+        raise because the operation is invalid.
+
+        Parameters
+        ----------
+        other : timedelta, np.timedelta64, Tick,
+                ndarray[timedelta64], TimedeltaArray, TimedeltaIndex
+
+        Returns
+        -------
+        multiple : int or ndarray[int64]
+
+        Raises
+        ------
+        IncompatibleFrequency
+        """
+        assert self.dtype._is_tick_like()  # checked by calling function
+
+        dtype = np.dtype(f"m8[{self.dtype._td64_unit}]")
+
+        if isinstance(other, (timedelta, np.timedelta64, Tick)):
+            td = np.asarray(Timedelta(other).asm8)
+        else:
+            td = np.asarray(other)
+
+        try:
+            delta = astype_overflowsafe(td, dtype=dtype, copy=False, round_ok=False)
+        except ValueError as err:
+            raise raise_on_incompatible(self, other) from err
+
+        delta = delta.view("i8")
+        return lib.item_from_zerodim(delta)
+
+    # ------------------------------------------------------------------
+    # Reductions
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        result = super()._reduce(name, skipna=skipna, keepdims=keepdims, **kwargs)
+        if keepdims and isinstance(result, np.ndarray):
+            return self._from_sequence(result, dtype=self.dtype)
+        return result
+
+
+def raise_on_incompatible(left, right) -> IncompatibleFrequency:
+    """
+    Helper function to render a consistent error message when raising
+    IncompatibleFrequency.
+
+    Parameters
+    ----------
+    left : PeriodArray
+    right : None, DateOffset, Period, ndarray, or timedelta-like
+
+    Returns
+    -------
+    IncompatibleFrequency
+        Exception to be raised by the caller.
+    """
+    # GH#24283 error message format depends on whether right is scalar
+    if isinstance(right, (np.ndarray, ABCTimedeltaArray)) or right is None:
+        other_freq = None
+    elif isinstance(right, BaseOffset):
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore", r"PeriodDtype\[B\] is deprecated", category=FutureWarning
+            )
+            other_freq = PeriodDtype(right)._freqstr
+    elif isinstance(right, (ABCPeriodIndex, PeriodArray, Period)):
+        other_freq = right.freqstr
+    else:
+        other_freq = delta_to_tick(Timedelta(right)).freqstr
+
+    own_freq = PeriodDtype(left.freq)._freqstr
+    msg = DIFFERENT_FREQ.format(
+        cls=type(left).__name__, own_freq=own_freq, other_freq=other_freq
+    )
+    return IncompatibleFrequency(msg)
+
+
+# -------------------------------------------------------------------
+# Constructor Helpers
+
+
+def period_array(
+    data: Sequence[Period | str | None] | AnyArrayLike,
+    freq: str | Tick | BaseOffset | None = None,
+    copy: bool = False,
+) -> PeriodArray:
+    """
+    Construct a new PeriodArray from a sequence of Period scalars.
+
+    Parameters
+    ----------
+    data : Sequence of Period objects
+        A sequence of Period objects. These are required to all have
+        the same ``freq.`` Missing values can be indicated by ``None``
+        or ``pandas.NaT``.
+    freq : str, Tick, or Offset
+        The frequency of every element of the array. This can be specified
+        to avoid inferring the `freq` from `data`.
+    copy : bool, default False
+        Whether to ensure a copy of the data is made.
+
+    Returns
+    -------
+    PeriodArray
+
+    See Also
+    --------
+    PeriodArray
+    pandas.PeriodIndex
+
+    Examples
+    --------
+    >>> period_array([pd.Period("2017", freq="Y"), pd.Period("2018", freq="Y")])
+    <PeriodArray>
+    ['2017', '2018']
+    Length: 2, dtype: period[Y-DEC]
+
+    >>> period_array([pd.Period("2017", freq="Y"), pd.Period("2018", freq="Y"), pd.NaT])
+    <PeriodArray>
+    ['2017', '2018', 'NaT']
+    Length: 3, dtype: period[Y-DEC]
+
+    Integers that look like years are handled
+
+    >>> period_array([2000, 2001, 2002], freq="D")
+    <PeriodArray>
+    ['2000-01-01', '2001-01-01', '2002-01-01']
+    Length: 3, dtype: period[D]
+
+    Datetime-like strings may also be passed
+
+    >>> period_array(["2000-Q1", "2000-Q2", "2000-Q3", "2000-Q4"], freq="Q")
+    <PeriodArray>
+    ['2000Q1', '2000Q2', '2000Q3', '2000Q4']
+    Length: 4, dtype: period[Q-DEC]
+    """
+    data_dtype = getattr(data, "dtype", None)
+
+    if lib.is_np_dtype(data_dtype, "M"):
+        return PeriodArray._from_datetime64(data, freq)
+    if isinstance(data_dtype, PeriodDtype):
+        out = PeriodArray(data)
+        if freq is not None:
+            if freq == data_dtype.freq:
+                return out
+            return out.asfreq(freq)
+        return out
+
+    # other iterable of some kind
+    if not isinstance(data, (np.ndarray, list, tuple, ABCSeries)):
+        data = list(data)
+
+    arrdata = np.asarray(data)
+
+    dtype: PeriodDtype | None
+    if freq:
+        dtype = PeriodDtype(freq)
+    else:
+        dtype = None
+
+    if arrdata.dtype.kind == "f" and len(arrdata) > 0:
+        raise TypeError("PeriodIndex does not allow floating point in construction")
+
+    if arrdata.dtype.kind in "iu":
+        arr = arrdata.astype(np.int64, copy=False)
+        # error: Argument 2 to "from_ordinals" has incompatible type "Union[str,
+        # Tick, None]"; expected "Union[timedelta, BaseOffset, str]"
+        ordinals = libperiod.from_ordinals(arr, freq)  # type: ignore[arg-type]
+        return PeriodArray(ordinals, dtype=dtype)
+
+    data = ensure_object(arrdata)
+    if freq is None:
+        freq = libperiod.extract_freq(data)
+    dtype = PeriodDtype(freq)
+    return PeriodArray._from_sequence(data, dtype=dtype)
+
+
+@overload
+def validate_dtype_freq(dtype, freq: BaseOffsetT) -> BaseOffsetT: ...
+
+
+@overload
+def validate_dtype_freq(dtype, freq: timedelta | str | None) -> BaseOffset: ...
+
+
+def validate_dtype_freq(
+    dtype, freq: BaseOffsetT | BaseOffset | timedelta | str | None
+) -> BaseOffsetT:
+    """
+    If both a dtype and a freq are available, ensure they match.  If only
+    dtype is available, extract the implied freq.
+
+    Parameters
+    ----------
+    dtype : dtype
+    freq : DateOffset or None
+
+    Returns
+    -------
+    freq : DateOffset
+
+    Raises
+    ------
+    ValueError : non-period dtype
+    IncompatibleFrequency : mismatch between dtype and freq
+    """
+    if freq is not None:
+        freq = to_offset(freq, is_period=True)
+
+    if dtype is not None:
+        dtype = pandas_dtype(dtype)
+        if not isinstance(dtype, PeriodDtype):
+            raise ValueError("dtype must be PeriodDtype")
+        if freq is None:
+            freq = dtype.freq
+        elif freq != dtype.freq:
+            raise IncompatibleFrequency("specified freq and dtype are different")
+    # error: Incompatible return value type (got "Union[BaseOffset, Any, None]",
+    # expected "BaseOffset")
+    return freq  # type: ignore[return-value]
+
+
+def dt64arr_to_periodarr(
+    data, freq, tz=None
+) -> tuple[npt.NDArray[np.int64], BaseOffset]:
+    """
+    Convert an datetime-like array to values Period ordinals.
+
+    Parameters
+    ----------
+    data : Union[Series[datetime64[ns]], DatetimeIndex, ndarray[datetime64ns]]
+    freq : Optional[Union[str, Tick]]
+        Must match the `freq` on the `data` if `data` is a DatetimeIndex
+        or Series.
+    tz : Optional[tzinfo]
+
+    Returns
+    -------
+    ordinals : ndarray[int64]
+    freq : Tick
+        The frequency extracted from the Series or DatetimeIndex if that's
+        used.
+
+    """
+    if not isinstance(data.dtype, np.dtype) or data.dtype.kind != "M":
+        raise ValueError(f"Wrong dtype: {data.dtype}")
+
+    if freq is None:
+        if isinstance(data, ABCIndex):
+            data, freq = data._values, data.freq
+        elif isinstance(data, ABCSeries):
+            data, freq = data._values, data.dt.freq
+
+    elif isinstance(data, (ABCIndex, ABCSeries)):
+        data = data._values
+
+    reso = get_unit_from_dtype(data.dtype)
+    freq = Period._maybe_convert_freq(freq)
+    base = freq._period_dtype_code
+    return c_dt64arr_to_periodarr(data.view("i8"), base, tz, reso=reso), freq
+
+
+def _get_ordinal_range(start, end, periods, freq, mult: int = 1):
+    if com.count_not_none(start, end, periods) != 2:
+        raise ValueError(
+            "Of the three parameters: start, end, and periods, "
+            "exactly two must be specified"
+        )
+
+    if freq is not None:
+        freq = to_offset(freq, is_period=True)
+        mult = freq.n
+
+    if start is not None:
+        start = Period(start, freq)
+    if end is not None:
+        end = Period(end, freq)
+
+    is_start_per = isinstance(start, Period)
+    is_end_per = isinstance(end, Period)
+
+    if is_start_per and is_end_per and start.freq != end.freq:
+        raise ValueError("start and end must have same freq")
+    if start is NaT or end is NaT:
+        raise ValueError("start and end must not be NaT")
+
+    if freq is None:
+        if is_start_per:
+            freq = start.freq
+        elif is_end_per:
+            freq = end.freq
+        else:  # pragma: no cover
+            raise ValueError("Could not infer freq from start/end")
+        mult = freq.n
+
+    if periods is not None:
+        periods = periods * mult
+        if start is None:
+            data = np.arange(
+                end.ordinal - periods + mult, end.ordinal + 1, mult, dtype=np.int64
+            )
+        else:
+            data = np.arange(
+                start.ordinal, start.ordinal + periods, mult, dtype=np.int64
+            )
+    else:
+        data = np.arange(start.ordinal, end.ordinal + 1, mult, dtype=np.int64)
+
+    return data, freq
+
+
+def _range_from_fields(
+    year=None,
+    month=None,
+    quarter=None,
+    day=None,
+    hour=None,
+    minute=None,
+    second=None,
+    freq=None,
+) -> tuple[np.ndarray, BaseOffset]:
+    if hour is None:
+        hour = 0
+    if minute is None:
+        minute = 0
+    if second is None:
+        second = 0
+    if day is None:
+        day = 1
+
+    ordinals = []
+
+    if quarter is not None:
+        if freq is None:
+            freq = to_offset("Q", is_period=True)
+            base = FreqGroup.FR_QTR.value
+        else:
+            freq = to_offset(freq, is_period=True)
+            base = libperiod.freq_to_dtype_code(freq)
+            if base != FreqGroup.FR_QTR.value:
+                raise AssertionError("base must equal FR_QTR")
+
+        freqstr = freq.freqstr
+        year, quarter = _make_field_arrays(year, quarter)
+        for y, q in zip(year, quarter):
+            calendar_year, calendar_month = parsing.quarter_to_myear(y, q, freqstr)
+            val = libperiod.period_ordinal(
+                calendar_year, calendar_month, 1, 1, 1, 1, 0, 0, base
+            )
+            ordinals.append(val)
+    else:
+        freq = to_offset(freq, is_period=True)
+        base = libperiod.freq_to_dtype_code(freq)
+        arrays = _make_field_arrays(year, month, day, hour, minute, second)
+        for y, mth, d, h, mn, s in zip(*arrays):
+            ordinals.append(libperiod.period_ordinal(y, mth, d, h, mn, s, 0, 0, base))
+
+    return np.array(ordinals, dtype=np.int64), freq
+
+
+def _make_field_arrays(*fields) -> list[np.ndarray]:
+    length = None
+    for x in fields:
+        if isinstance(x, (list, np.ndarray, ABCSeries)):
+            if length is not None and len(x) != length:
+                raise ValueError("Mismatched Period array lengths")
+            if length is None:
+                length = len(x)
+
+    # error: Argument 2 to "repeat" has incompatible type "Optional[int]"; expected
+    # "Union[Union[int, integer[Any]], Union[bool, bool_], ndarray, Sequence[Union[int,
+    # integer[Any]]], Sequence[Union[bool, bool_]], Sequence[Sequence[Any]]]"
+    return [
+        np.asarray(x)
+        if isinstance(x, (np.ndarray, list, ABCSeries))
+        else np.repeat(x, length)  # type: ignore[arg-type]
+        for x in fields
+    ]
diff --git a/pandas/pandas/core/arrays/sparse/__init__.py b/pandas/pandas/core/arrays/sparse/__init__.py
new file mode 100644
index 0000000000000..93d5cb8cc335a
--- /dev/null
+++ b/pandas/pandas/core/arrays/sparse/__init__.py
@@ -0,0 +1,19 @@
+from pandas.core.arrays.sparse.accessor import (
+    SparseAccessor,
+    SparseFrameAccessor,
+)
+from pandas.core.arrays.sparse.array import (
+    BlockIndex,
+    IntIndex,
+    SparseArray,
+    make_sparse_index,
+)
+
+__all__ = [
+    "BlockIndex",
+    "IntIndex",
+    "SparseAccessor",
+    "SparseArray",
+    "SparseFrameAccessor",
+    "make_sparse_index",
+]
diff --git a/pandas/pandas/core/arrays/sparse/accessor.py b/pandas/pandas/core/arrays/sparse/accessor.py
new file mode 100644
index 0000000000000..eab8527eef526
--- /dev/null
+++ b/pandas/pandas/core/arrays/sparse/accessor.py
@@ -0,0 +1,488 @@
+"""Sparse accessor"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.dtypes.cast import find_common_type
+from pandas.core.dtypes.dtypes import SparseDtype
+
+from pandas.core.accessor import (
+    PandasDelegate,
+    delegate_names,
+)
+from pandas.core.arrays.sparse.array import SparseArray
+
+if TYPE_CHECKING:
+    from scipy.sparse import (
+        coo_matrix,
+        spmatrix,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+class BaseAccessor:
+    _validation_msg = "Can only use the '.sparse' accessor with Sparse data."
+
+    def __init__(self, data=None) -> None:
+        self._parent = data
+        self._validate(data)
+
+    def _validate(self, data) -> None:
+        raise NotImplementedError
+
+
+@delegate_names(
+    SparseArray, ["npoints", "density", "fill_value", "sp_values"], typ="property"
+)
+class SparseAccessor(BaseAccessor, PandasDelegate):
+    """
+    Accessor for SparseSparse from other sparse matrix data types.
+
+    Parameters
+    ----------
+    data : Series or DataFrame
+        The Series or DataFrame to which the SparseAccessor is attached.
+
+    See Also
+    --------
+    Series.sparse.to_coo : Create a scipy.sparse.coo_matrix from a Series with
+        MultiIndex.
+    Series.sparse.from_coo : Create a Series with sparse values from a
+        scipy.sparse.coo_matrix.
+
+    Examples
+    --------
+    >>> ser = pd.Series([0, 0, 2, 2, 2], dtype="Sparse[int]")
+    >>> ser.sparse.density
+    0.6
+    >>> ser.sparse.sp_values
+    array([2, 2, 2])
+    """
+
+    def _validate(self, data) -> None:
+        if not isinstance(data.dtype, SparseDtype):
+            raise AttributeError(self._validation_msg)
+
+    def _delegate_property_get(self, name: str, *args, **kwargs):
+        return getattr(self._parent.array, name)
+
+    def _delegate_method(self, name: str, *args, **kwargs):
+        if name == "from_coo":
+            return self.from_coo(*args, **kwargs)
+        elif name == "to_coo":
+            return self.to_coo(*args, **kwargs)
+        else:
+            raise ValueError
+
+    @classmethod
+    def from_coo(cls, A, dense_index: bool = False) -> Series:
+        """
+        Create a Series with sparse values from a scipy.sparse.coo_matrix.
+
+        This method takes a ``scipy.sparse.coo_matrix`` (coordinate format) as input and
+        returns a pandas ``Series`` where the non-zero elements are represented as
+        sparse values. The index of the Series can either include only the coordinates
+        of non-zero elements (default behavior) or the full sorted set of coordinates
+        from the matrix if ``dense_index`` is set to `True`.
+
+        Parameters
+        ----------
+        A : scipy.sparse.coo_matrix
+            The sparse matrix in coordinate format from which the sparse Series
+            will be created.
+        dense_index : bool, default False
+            If False (default), the index consists of only the
+            coords of the non-null entries of the original coo_matrix.
+            If True, the index consists of the full sorted
+            (row, col) coordinates of the coo_matrix.
+
+        Returns
+        -------
+        s : Series
+            A Series with sparse values.
+
+        See Also
+        --------
+        DataFrame.sparse.from_spmatrix : Create a new DataFrame from a scipy sparse
+            matrix.
+        scipy.sparse.coo_matrix : A sparse matrix in COOrdinate format.
+
+        Examples
+        --------
+        >>> from scipy import sparse
+
+        >>> A = sparse.coo_matrix(
+        ...     ([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])), shape=(3, 4)
+        ... )
+        >>> A
+        <COOrdinate sparse matrix of dtype 'float64'
+            with 3 stored elements and shape (3, 4)>
+
+        >>> A.todense()
+        matrix([[0., 0., 1., 2.],
+        [3., 0., 0., 0.],
+        [0., 0., 0., 0.]])
+
+        >>> ss = pd.Series.sparse.from_coo(A)
+        >>> ss
+        0  2    1.0
+           3    2.0
+        1  0    3.0
+        dtype: Sparse[float64, nan]
+        """
+        from pandas import Series
+        from pandas.core.arrays.sparse.scipy_sparse import coo_to_sparse_series
+
+        result = coo_to_sparse_series(A, dense_index=dense_index)
+        result = Series(result.array, index=result.index, copy=False)
+
+        return result
+
+    def to_coo(
+        self, row_levels=(0,), column_levels=(1,), sort_labels: bool = False
+    ) -> tuple[coo_matrix, list, list]:
+        """
+        Create a scipy.sparse.coo_matrix from a Series with MultiIndex.
+
+        Use row_levels and column_levels to determine the row and column
+        coordinates respectively. row_levels and column_levels are the names
+        (labels) or numbers of the levels. {row_levels, column_levels} must be
+        a partition of the MultiIndex level names (or numbers).
+
+        Parameters
+        ----------
+        row_levels : tuple/list
+            MultiIndex levels to use for row coordinates, specified by name or index.
+        column_levels : tuple/list
+            MultiIndex levels to use for column coordinates, specified by name or index.
+        sort_labels : bool, default False
+            Sort the row and column labels before forming the sparse matrix.
+            When `row_levels` and/or `column_levels` refer to a single level,
+            set to `True` for a faster execution.
+
+        Returns
+        -------
+        y : scipy.sparse.coo_matrix
+            The sparse matrix in coordinate format.
+        rows : list (row labels)
+            Labels corresponding to the row coordinates.
+        columns : list (column labels)
+            Labels corresponding to the column coordinates.
+
+        See Also
+        --------
+        Series.sparse.from_coo : Create a Series with sparse values from a
+            scipy.sparse.coo_matrix.
+
+        Examples
+        --------
+        >>> s = pd.Series([3.0, np.nan, 1.0, 3.0, np.nan, np.nan])
+        >>> s.index = pd.MultiIndex.from_tuples(
+        ...     [
+        ...         (1, 2, "a", 0),
+        ...         (1, 2, "a", 1),
+        ...         (1, 1, "b", 0),
+        ...         (1, 1, "b", 1),
+        ...         (2, 1, "b", 0),
+        ...         (2, 1, "b", 1),
+        ...     ],
+        ...     names=["A", "B", "C", "D"],
+        ... )
+        >>> s
+        A  B  C  D
+        1  2  a  0    3.0
+                 1    NaN
+           1  b  0    1.0
+                 1    3.0
+        2  1  b  0    NaN
+                 1    NaN
+        dtype: float64
+
+        >>> ss = s.astype("Sparse")
+        >>> ss
+        A  B  C  D
+        1  2  a  0    3.0
+                 1    NaN
+           1  b  0    1.0
+                 1    3.0
+        2  1  b  0    NaN
+                 1    NaN
+        dtype: Sparse[float64, nan]
+
+        >>> A, rows, columns = ss.sparse.to_coo(
+        ...     row_levels=["A", "B"], column_levels=["C", "D"], sort_labels=True
+        ... )
+        >>> A
+        <COOrdinate sparse matrix of dtype 'float64'
+            with 3 stored elements and shape (3, 4)>
+        >>> A.todense()
+        matrix([[0., 0., 1., 3.],
+        [3., 0., 0., 0.],
+        [0., 0., 0., 0.]])
+
+        >>> rows
+        [(1, 1), (1, 2), (2, 1)]
+        >>> columns
+        [('a', 0), ('a', 1), ('b', 0), ('b', 1)]
+        """
+        from pandas.core.arrays.sparse.scipy_sparse import sparse_series_to_coo
+
+        A, rows, columns = sparse_series_to_coo(
+            self._parent, row_levels, column_levels, sort_labels=sort_labels
+        )
+        return A, rows, columns
+
+    def to_dense(self) -> Series:
+        """
+        Convert a Series from sparse values to dense.
+
+        Returns
+        -------
+        Series:
+            A Series with the same values, stored as a dense array.
+
+        Examples
+        --------
+        >>> series = pd.Series(pd.arrays.SparseArray([0, 1, 0]))
+        >>> series
+        0    0
+        1    1
+        2    0
+        dtype: Sparse[int64, 0]
+
+        >>> series.sparse.to_dense()
+        0    0
+        1    1
+        2    0
+        dtype: int64
+        """
+        from pandas import Series
+
+        return Series(
+            self._parent.array.to_dense(),
+            index=self._parent.index,
+            name=self._parent.name,
+            copy=False,
+        )
+
+
+class SparseFrameAccessor(BaseAccessor, PandasDelegate):
+    """
+    DataFrame accessor for sparse data.
+
+    It allows users to interact with a `DataFrame` that contains sparse data types
+    (`SparseDtype`). It provides methods and attributes to efficiently work with sparse
+    storage, reducing memory usage while maintaining compatibility with standard pandas
+    operations.
+
+    Parameters
+    ----------
+    data : scipy.sparse.spmatrix
+        Must be convertible to csc format.
+
+    See Also
+    --------
+    DataFrame.sparse.density : Ratio of non-sparse points to total (dense) data points.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"a": [1, 2, 0, 0], "b": [3, 0, 0, 4]}, dtype="Sparse[int]")
+    >>> df.sparse.density
+    0.5
+    """
+
+    def _validate(self, data) -> None:
+        dtypes = data.dtypes
+        if not all(isinstance(t, SparseDtype) for t in dtypes):
+            raise AttributeError(self._validation_msg)
+
+    @classmethod
+    def from_spmatrix(cls, data, index=None, columns=None) -> DataFrame:
+        """
+        Create a new DataFrame from a scipy sparse matrix.
+
+        Parameters
+        ----------
+        data : scipy.sparse.spmatrix
+            Must be convertible to csc format.
+        index, columns : Index, optional
+            Row and column labels to use for the resulting DataFrame.
+            Defaults to a RangeIndex.
+
+        Returns
+        -------
+        DataFrame
+            Each column of the DataFrame is stored as a
+            :class:`arrays.SparseArray`.
+
+        See Also
+        --------
+        DataFrame.sparse.to_coo : Return the contents of the frame as a
+            sparse SciPy COO matrix.
+
+        Examples
+        --------
+        >>> import scipy.sparse
+        >>> mat = scipy.sparse.eye(3, dtype=int)
+        >>> pd.DataFrame.sparse.from_spmatrix(mat)
+             0    1    2
+        0    1    0    0
+        1    0    1    0
+        2    0    0    1
+        """
+        from pandas._libs.sparse import IntIndex
+
+        from pandas import DataFrame
+
+        data = data.tocsc()
+        index, columns = cls._prep_index(data, index, columns)
+        n_rows, n_columns = data.shape
+        # We need to make sure indices are sorted, as we create
+        # IntIndex with no input validation (i.e. check_integrity=False ).
+        # Indices may already be sorted in scipy in which case this adds
+        # a small overhead.
+        data.sort_indices()
+        indices = data.indices
+        indptr = data.indptr
+        array_data = data.data
+        dtype = SparseDtype(array_data.dtype)
+        arrays = []
+        for i in range(n_columns):
+            sl = slice(indptr[i], indptr[i + 1])
+            idx = IntIndex(n_rows, indices[sl], check_integrity=False)
+            arr = SparseArray._simple_new(array_data[sl], idx, dtype)
+            arrays.append(arr)
+        return DataFrame._from_arrays(
+            arrays, columns=columns, index=index, verify_integrity=False
+        )
+
+    def to_dense(self) -> DataFrame:
+        """
+        Convert a DataFrame with sparse values to dense.
+
+        Returns
+        -------
+        DataFrame
+            A DataFrame with the same values stored as dense arrays.
+
+        See Also
+        --------
+        DataFrame.sparse.density : Ratio of non-sparse points to total
+            (dense) data points.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 1, 0])})
+        >>> df.sparse.to_dense()
+           A
+        0  0
+        1  1
+        2  0
+        """
+        data = {k: v.array.to_dense() for k, v in self._parent.items()}
+        return self._parent._constructor(
+            data, index=self._parent.index, columns=self._parent.columns
+        )
+
+    def to_coo(self) -> spmatrix:
+        """
+        Return the contents of the frame as a sparse SciPy COO matrix.
+
+        Returns
+        -------
+        scipy.sparse.spmatrix
+            If the caller is heterogeneous and contains booleans or objects,
+            the result will be of dtype=object. See Notes.
+
+        See Also
+        --------
+        DataFrame.sparse.to_dense : Convert a DataFrame with sparse values to dense.
+
+        Notes
+        -----
+        The dtype will be the lowest-common-denominator type (implicit
+        upcasting); that is to say if the dtypes (even of numeric types)
+        are mixed, the one that accommodates all will be chosen.
+
+        e.g. If the dtypes are float16 and float32, dtype will be upcast to
+        float32. By numpy.find_common_type convention, mixing int64 and
+        and uint64 will result in a float64 dtype.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 1, 0, 1])})
+        >>> df.sparse.to_coo()
+        <COOrdinate sparse matrix of dtype 'int64'
+            with 2 stored elements and shape (4, 1)>
+        """
+        import_optional_dependency("scipy")
+        from scipy.sparse import coo_matrix
+
+        dtype = find_common_type(self._parent.dtypes.to_list())
+        if isinstance(dtype, SparseDtype):
+            dtype = dtype.subtype
+
+        cols, rows, data = [], [], []
+        for col, (_, ser) in enumerate(self._parent.items()):
+            sp_arr = ser.array
+
+            row = sp_arr.sp_index.indices
+            cols.append(np.repeat(col, len(row)))
+            rows.append(row)
+            data.append(sp_arr.sp_values.astype(dtype, copy=False))
+
+        cols = np.concatenate(cols)
+        rows = np.concatenate(rows)
+        data = np.concatenate(data)
+        return coo_matrix((data, (rows, cols)), shape=self._parent.shape)
+
+    @property
+    def density(self) -> float:
+        """
+        Ratio of non-sparse points to total (dense) data points.
+
+        See Also
+        --------
+        DataFrame.sparse.from_spmatrix : Create a new DataFrame from a
+            scipy sparse matrix.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 1, 0, 1])})
+        >>> df.sparse.density
+        0.5
+        """
+        tmp = np.mean([column.array.density for _, column in self._parent.items()])
+        return tmp
+
+    @staticmethod
+    def _prep_index(data, index, columns):
+        from pandas.core.indexes.api import (
+            default_index,
+            ensure_index,
+        )
+
+        N, K = data.shape
+        if index is None:
+            index = default_index(N)
+        else:
+            index = ensure_index(index)
+        if columns is None:
+            columns = default_index(K)
+        else:
+            columns = ensure_index(columns)
+
+        if len(columns) != K:
+            raise ValueError(f"Column length mismatch: {len(columns)} vs. {K}")
+        if len(index) != N:
+            raise ValueError(f"Index length mismatch: {len(index)} vs. {N}")
+        return index, columns
diff --git a/pandas/pandas/core/arrays/sparse/array.py b/pandas/pandas/core/arrays/sparse/array.py
new file mode 100644
index 0000000000000..137dbb6e4d139
--- /dev/null
+++ b/pandas/pandas/core/arrays/sparse/array.py
@@ -0,0 +1,1941 @@
+"""
+SparseArray data structure
+"""
+
+from __future__ import annotations
+
+from collections import abc
+import numbers
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config.config import get_option
+
+from pandas._libs import lib
+import pandas._libs.sparse as splib
+from pandas._libs.sparse import (
+    BlockIndex,
+    IntIndex,
+    SparseIndex,
+)
+from pandas._libs.tslibs import NaT
+from pandas.compat.numpy import function as nv
+from pandas.errors import PerformanceWarning
+from pandas.util._decorators import doc
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import (
+    validate_bool_kwarg,
+    validate_insert_loc,
+)
+
+from pandas.core.dtypes.astype import astype_array
+from pandas.core.dtypes.cast import (
+    find_common_type,
+    maybe_box_datetimelike,
+)
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_integer,
+    is_list_like,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    SparseDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+    notna,
+)
+
+from pandas.core import arraylike
+import pandas.core.algorithms as algos
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays import ExtensionArray
+from pandas.core.base import PandasObject
+import pandas.core.common as com
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+    sanitize_array,
+)
+from pandas.core.indexers import (
+    check_array_indexer,
+    unpack_tuple_and_ellipses,
+)
+from pandas.core.nanops import check_below_min_count
+
+from pandas.io.formats import printing
+
+# See https://github.com/python/typing/issues/684
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+    from enum import Enum
+
+    class ellipsis(Enum):
+        Ellipsis = "..."
+
+    Ellipsis = ellipsis.Ellipsis
+
+    from scipy.sparse import spmatrix
+
+    from pandas._typing import NumpySorter
+
+    SparseIndexKind = Literal["integer", "block"]
+
+    from pandas._typing import (
+        ArrayLike,
+        AstypeArg,
+        Axis,
+        AxisInt,
+        Dtype,
+        NpDtype,
+        PositionalIndexer,
+        Scalar,
+        ScalarIndexer,
+        Self,
+        SequenceIndexer,
+        npt,
+    )
+
+    from pandas import Series
+
+else:
+    ellipsis = type(Ellipsis)
+
+
+# ----------------------------------------------------------------------------
+# Array
+
+_sparray_doc_kwargs = {"klass": "SparseArray"}
+
+
+def _get_fill(arr: SparseArray) -> np.ndarray:
+    """
+    Create a 0-dim ndarray containing the fill value
+
+    Parameters
+    ----------
+    arr : SparseArray
+
+    Returns
+    -------
+    fill_value : ndarray
+        0-dim ndarray with just the fill value.
+
+    Notes
+    -----
+    coerce fill_value to arr dtype if possible
+    int64 SparseArray can have NaN as fill_value if there is no missing
+    """
+    try:
+        return np.asarray(arr.fill_value, dtype=arr.dtype.subtype)
+    except ValueError:
+        return np.asarray(arr.fill_value)
+
+
+def _sparse_array_op(
+    left: SparseArray, right: SparseArray, op: Callable, name: str
+) -> SparseArray:
+    """
+    Perform a binary operation between two arrays.
+
+    Parameters
+    ----------
+    left : Union[SparseArray, ndarray]
+    right : Union[SparseArray, ndarray]
+    op : Callable
+        The binary operation to perform
+    name str
+        Name of the callable.
+
+    Returns
+    -------
+    SparseArray
+    """
+    if name.startswith("__"):
+        # For lookups in _libs.sparse we need non-dunder op name
+        name = name[2:-2]
+
+    # dtype used to find corresponding sparse method
+    ltype = left.dtype.subtype
+    rtype = right.dtype.subtype
+
+    if ltype != rtype:
+        subtype = find_common_type([ltype, rtype])
+        ltype = SparseDtype(subtype, left.fill_value)
+        rtype = SparseDtype(subtype, right.fill_value)
+
+        left = left.astype(ltype, copy=False)
+        right = right.astype(rtype, copy=False)
+        dtype = ltype.subtype
+    else:
+        dtype = ltype
+
+    # dtype the result must have
+    result_dtype = None
+
+    if left.sp_index.ngaps == 0 or right.sp_index.ngaps == 0:
+        with np.errstate(all="ignore"):
+            result = op(left.to_dense(), right.to_dense())
+            fill = op(_get_fill(left), _get_fill(right))
+
+        if left.sp_index.ngaps == 0:
+            index = left.sp_index
+        else:
+            index = right.sp_index
+    elif left.sp_index.equals(right.sp_index):
+        with np.errstate(all="ignore"):
+            result = op(left.sp_values, right.sp_values)
+            fill = op(_get_fill(left), _get_fill(right))
+        index = left.sp_index
+    else:
+        if name[0] == "r":
+            left, right = right, left
+            name = name[1:]
+
+        if name in ("and", "or", "xor") and dtype == "bool":
+            opname = f"sparse_{name}_uint8"
+            # to make template simple, cast here
+            left_sp_values = left.sp_values.view(np.uint8)
+            right_sp_values = right.sp_values.view(np.uint8)
+            result_dtype = bool
+        else:
+            opname = f"sparse_{name}_{dtype}"
+            left_sp_values = left.sp_values
+            right_sp_values = right.sp_values
+
+        if (
+            name in ["floordiv", "mod"]
+            and (right == 0).any()
+            and left.dtype.kind in "iu"
+        ):
+            # Match the non-Sparse Series behavior
+            opname = f"sparse_{name}_float64"
+            left_sp_values = left_sp_values.astype("float64")
+            right_sp_values = right_sp_values.astype("float64")
+
+        sparse_op = getattr(splib, opname)
+
+        with np.errstate(all="ignore"):
+            result, index, fill = sparse_op(
+                left_sp_values,
+                left.sp_index,
+                left.fill_value,
+                right_sp_values,
+                right.sp_index,
+                right.fill_value,
+            )
+
+    if name == "divmod":
+        # result is a 2-tuple
+        # error: Incompatible return value type (got "Tuple[SparseArray,
+        # SparseArray]", expected "SparseArray")
+        return (  # type: ignore[return-value]
+            _wrap_result(name, result[0], index, fill[0], dtype=result_dtype),
+            _wrap_result(name, result[1], index, fill[1], dtype=result_dtype),
+        )
+
+    if result_dtype is None:
+        result_dtype = result.dtype
+
+    return _wrap_result(name, result, index, fill, dtype=result_dtype)
+
+
+def _wrap_result(
+    name: str, data, sparse_index, fill_value, dtype: Dtype | None = None
+) -> SparseArray:
+    """
+    wrap op result to have correct dtype
+    """
+    if name.startswith("__"):
+        # e.g. __eq__ --> eq
+        name = name[2:-2]
+
+    if name in ("eq", "ne", "lt", "gt", "le", "ge"):
+        dtype = bool
+
+    fill_value = lib.item_from_zerodim(fill_value)
+
+    if is_bool_dtype(dtype):
+        # fill_value may be np.bool_
+        fill_value = bool(fill_value)
+    return SparseArray(
+        data, sparse_index=sparse_index, fill_value=fill_value, dtype=dtype
+    )
+
+
+class SparseArray(OpsMixin, PandasObject, ExtensionArray):
+    """
+    An ExtensionArray for storing sparse data.
+
+    SparseArray efficiently stores data with a high frequency of a
+    specific fill value (e.g., zeros), saving memory by only retaining
+    non-fill elements and their indices. This class is particularly
+    useful for large datasets where most values are redundant.
+
+    Parameters
+    ----------
+    data : array-like or scalar
+        A dense array of values to store in the SparseArray. This may contain
+        `fill_value`.
+    sparse_index : SparseIndex, optional
+        Index indicating the locations of sparse elements.
+    fill_value : scalar, optional
+        Elements in data that are ``fill_value`` are not stored in the
+        SparseArray. For memory savings, this should be the most common value
+        in `data`. By default, `fill_value` depends on the dtype of `data`:
+
+        =========== ==========
+        data.dtype  na_value
+        =========== ==========
+        float       ``np.nan``
+        int         ``0``
+        bool        False
+        datetime64  ``pd.NaT``
+        timedelta64 ``pd.NaT``
+        =========== ==========
+
+        The fill value is potentially specified in three ways. In order of
+        precedence, these are
+
+        1. The `fill_value` argument
+        2. ``dtype.fill_value`` if `fill_value` is None and `dtype` is
+           a ``SparseDtype``
+        3. ``data.dtype.fill_value`` if `fill_value` is None and `dtype`
+           is not a ``SparseDtype`` and `data` is a ``SparseArray``.
+
+    kind : str
+        Can be 'integer' or 'block', default is 'integer'.
+        The type of storage for sparse locations.
+
+        * 'block': Stores a `block` and `block_length` for each
+          contiguous *span* of sparse values. This is best when
+          sparse data tends to be clumped together, with large
+          regions of ``fill-value`` values between sparse values.
+        * 'integer': uses an integer to store the location of
+          each sparse value.
+
+    dtype : np.dtype or SparseDtype, optional
+        The dtype to use for the SparseArray. For numpy dtypes, this
+        determines the dtype of ``self.sp_values``. For SparseDtype,
+        this determines ``self.sp_values`` and ``self.fill_value``.
+    copy : bool, default False
+        Whether to explicitly copy the incoming `data` array.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    SparseDtype : Dtype for sparse data.
+
+    Examples
+    --------
+    >>> from pandas.arrays import SparseArray
+    >>> arr = SparseArray([0, 0, 1, 2])
+    >>> arr
+    [0, 0, 1, 2]
+    Fill: 0
+    IntIndex
+    Indices: array([2, 3], dtype=int32)
+    """
+
+    _subtyp = "sparse_array"  # register ABCSparseArray
+    _hidden_attrs = PandasObject._hidden_attrs | frozenset([])
+    _sparse_index: SparseIndex
+    _sparse_values: np.ndarray
+    _dtype: SparseDtype
+
+    def __init__(
+        self,
+        data,
+        sparse_index=None,
+        fill_value=None,
+        kind: SparseIndexKind = "integer",
+        dtype: Dtype | None = None,
+        copy: bool = False,
+    ) -> None:
+        if fill_value is None and isinstance(dtype, SparseDtype):
+            fill_value = dtype.fill_value
+
+        if isinstance(data, type(self)):
+            # disable normal inference on dtype, sparse_index, & fill_value
+            if sparse_index is None:
+                sparse_index = data.sp_index
+            if fill_value is None:
+                fill_value = data.fill_value
+            if dtype is None:
+                dtype = data.dtype
+            # TODO: make kind=None, and use data.kind?
+            data = data.sp_values
+
+        # Handle use-provided dtype
+        if isinstance(dtype, str):
+            # Two options: dtype='int', regular numpy dtype
+            # or dtype='Sparse[int]', a sparse dtype
+            try:
+                dtype = SparseDtype.construct_from_string(dtype)
+            except TypeError:
+                dtype = pandas_dtype(dtype)
+
+        if isinstance(dtype, SparseDtype):
+            if fill_value is None:
+                fill_value = dtype.fill_value
+            dtype = dtype.subtype
+
+        if is_scalar(data):
+            raise TypeError(
+                f"Cannot construct {type(self).__name__} from scalar data. "
+                "Pass a sequence instead."
+            )
+
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        # TODO: disentangle the fill_value dtype inference from
+        # dtype inference
+        if data is None:
+            # TODO: What should the empty dtype be? Object or float?
+
+            # error: Argument "dtype" to "array" has incompatible type
+            # "Union[ExtensionDtype, dtype[Any], None]"; expected "Union[dtype[Any],
+            # None, type, _SupportsDType, str, Union[Tuple[Any, int], Tuple[Any,
+            # Union[int, Sequence[int]]], List[Any], _DTypeDict, Tuple[Any, Any]]]"
+            data = np.array([], dtype=dtype)  # type: ignore[arg-type]
+
+        try:
+            data = sanitize_array(data, index=None)
+        except ValueError:
+            # NumPy may raise a ValueError on data like [1, []]
+            # we retry with object dtype here.
+            if dtype is None:
+                dtype = np.dtype(object)
+                data = np.atleast_1d(np.asarray(data, dtype=dtype))
+            else:
+                raise
+
+        if copy:
+            # TODO: avoid double copy when dtype forces cast.
+            data = data.copy()
+
+        if fill_value is None:
+            fill_value_dtype = data.dtype if dtype is None else dtype
+            if fill_value_dtype is None:
+                fill_value = np.nan
+            else:
+                fill_value = na_value_for_dtype(fill_value_dtype)
+
+        if isinstance(data, type(self)) and sparse_index is None:
+            sparse_index = data._sparse_index
+            # error: Argument "dtype" to "asarray" has incompatible type
+            # "Union[ExtensionDtype, dtype[Any], None]"; expected "None"
+            sparse_values = np.asarray(
+                data.sp_values,
+                dtype=dtype,  # type: ignore[arg-type]
+            )
+        elif sparse_index is None:
+            data = extract_array(data, extract_numpy=True)
+            if not isinstance(data, np.ndarray):
+                # EA
+                if isinstance(data.dtype, DatetimeTZDtype):
+                    warnings.warn(
+                        f"Creating SparseArray from {data.dtype} data "
+                        "loses timezone information. Cast to object before "
+                        "sparse to retain timezone information.",
+                        UserWarning,
+                        stacklevel=find_stack_level(),
+                    )
+                    data = np.asarray(data, dtype="datetime64[ns]")
+                    if fill_value is NaT:
+                        fill_value = np.datetime64("NaT", "ns")
+                data = np.asarray(data)
+            sparse_values, sparse_index, fill_value = _make_sparse(
+                # error: Argument "dtype" to "_make_sparse" has incompatible type
+                # "Union[ExtensionDtype, dtype[Any], None]"; expected
+                # "Optional[dtype[Any]]"
+                data,
+                kind=kind,
+                fill_value=fill_value,
+                dtype=dtype,  # type: ignore[arg-type]
+            )
+        else:
+            # error: Argument "dtype" to "asarray" has incompatible type
+            # "Union[ExtensionDtype, dtype[Any], None]"; expected "None"
+            sparse_values = np.asarray(data, dtype=dtype)  # type: ignore[arg-type]
+            if len(sparse_values) != sparse_index.npoints:
+                raise AssertionError(
+                    f"Non array-like type {type(sparse_values)} must "
+                    "have the same length as the index"
+                )
+        self._sparse_index = sparse_index
+        self._sparse_values = sparse_values
+        self._dtype = SparseDtype(sparse_values.dtype, fill_value)
+
+    @classmethod
+    def _simple_new(
+        cls,
+        sparse_array: np.ndarray,
+        sparse_index: SparseIndex,
+        dtype: SparseDtype,
+    ) -> Self:
+        new = object.__new__(cls)
+        new._sparse_index = sparse_index
+        new._sparse_values = sparse_array
+        new._dtype = dtype
+        return new
+
+    @classmethod
+    def from_spmatrix(cls, data: spmatrix) -> Self:
+        """
+        Create a SparseArray from a scipy.sparse matrix.
+
+        Parameters
+        ----------
+        data : scipy.sparse.sp_matrix
+            This should be a SciPy sparse matrix where the size
+            of the second dimension is 1. In other words, a
+            sparse matrix with a single column.
+
+        Returns
+        -------
+        SparseArray
+
+        Examples
+        --------
+        >>> import scipy.sparse
+        >>> mat = scipy.sparse.coo_matrix((4, 1))
+        >>> pd.arrays.SparseArray.from_spmatrix(mat)
+        [0.0, 0.0, 0.0, 0.0]
+        Fill: 0.0
+        IntIndex
+        Indices: array([], dtype=int32)
+        """
+        length, ncol = data.shape
+
+        if ncol != 1:
+            raise ValueError(f"'data' must have a single column, not '{ncol}'")
+
+        # our sparse index classes require that the positions be strictly
+        # increasing. So we need to sort loc, and arr accordingly.
+        data = data.tocsc()
+        data.sort_indices()
+        arr = data.data
+        idx = data.indices
+
+        zero = np.array(0, dtype=arr.dtype).item()
+        dtype = SparseDtype(arr.dtype, zero)
+        index = IntIndex(length, idx)
+
+        return cls._simple_new(arr, index, dtype)
+
+    def __array__(
+        self, dtype: NpDtype | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        if self.sp_index.ngaps == 0:
+            # Compat for na dtype and int values.
+            if copy is True:
+                return np.array(self.sp_values)
+            else:
+                return self.sp_values
+
+        if copy is False:
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        fill_value = self.fill_value
+
+        if dtype is None:
+            # Can NumPy represent this type?
+            # If not, `np.result_type` will raise. We catch that
+            # and return object.
+            if self.sp_values.dtype.kind == "M":
+                # However, we *do* special-case the common case of
+                # a datetime64 with pandas NaT.
+                if fill_value is NaT:
+                    # Can't put pd.NaT in a datetime64[ns]
+                    fill_value = np.datetime64("NaT")
+            try:
+                dtype = np.result_type(self.sp_values.dtype, type(fill_value))
+            except TypeError:
+                dtype = object
+
+        out = np.full(self.shape, fill_value, dtype=dtype)
+        out[self.sp_index.indices] = self.sp_values
+        return out
+
+    def __setitem__(self, key, value) -> None:
+        # I suppose we could allow setting of non-fill_value elements.
+        # TODO(SparseArray.__setitem__): remove special cases in
+        # ExtensionBlock.where
+        msg = "SparseArray does not support item assignment via setitem"
+        raise TypeError(msg)
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        return cls(scalars, dtype=dtype)
+
+    @classmethod
+    def _from_factorized(cls, values, original) -> Self:
+        return cls(values, dtype=original.dtype)
+
+    # ------------------------------------------------------------------------
+    # Data
+    # ------------------------------------------------------------------------
+    @property
+    def sp_index(self) -> SparseIndex:
+        """
+        The SparseIndex containing the location of non- ``fill_value`` points.
+        """
+        return self._sparse_index
+
+    @property
+    def sp_values(self) -> np.ndarray:
+        """
+        An ndarray containing the non- ``fill_value`` values.
+
+        This property returns the actual data values stored in the sparse
+        representation, excluding the values that are equal to the ``fill_value``.
+        The result is an ndarray of the underlying values, preserving the sparse
+        structure by omitting the default ``fill_value`` entries.
+
+        See Also
+        --------
+        Series.sparse.to_dense : Convert a Series from sparse values to dense.
+        Series.sparse.fill_value : Elements in `data` that are `fill_value` are
+            not stored.
+        Series.sparse.density : The percent of non- ``fill_value`` points, as decimal.
+
+        Examples
+        --------
+        >>> from pandas.arrays import SparseArray
+        >>> s = SparseArray([0, 0, 1, 0, 2], fill_value=0)
+        >>> s.sp_values
+        array([1, 2])
+        """
+        return self._sparse_values
+
+    @property
+    def dtype(self) -> SparseDtype:
+        return self._dtype
+
+    @property
+    def fill_value(self):
+        """
+        Elements in `data` that are `fill_value` are not stored.
+
+        For memory savings, this should be the most common value in the array.
+
+        See Also
+        --------
+        SparseDtype : Dtype for data stored in :class:`SparseArray`.
+        Series.value_counts : Return a Series containing counts of unique values.
+        Series.fillna : Fill NA/NaN in a Series with a specified value.
+
+        Examples
+        --------
+        >>> ser = pd.Series([0, 0, 2, 2, 2], dtype="Sparse[int]")
+        >>> ser.sparse.fill_value
+        0
+        >>> spa_dtype = pd.SparseDtype(dtype=np.int32, fill_value=2)
+        >>> ser = pd.Series([0, 0, 2, 2, 2], dtype=spa_dtype)
+        >>> ser.sparse.fill_value
+        2
+        """
+        return self.dtype.fill_value
+
+    @fill_value.setter
+    def fill_value(self, value) -> None:
+        self._dtype = SparseDtype(self.dtype.subtype, value)
+
+    @property
+    def kind(self) -> SparseIndexKind:
+        """
+        The kind of sparse index for this array. One of {'integer', 'block'}.
+        """
+        if isinstance(self.sp_index, IntIndex):
+            return "integer"
+        else:
+            return "block"
+
+    @property
+    def _valid_sp_values(self) -> np.ndarray:
+        sp_vals = self.sp_values
+        mask = notna(sp_vals)
+        return sp_vals[mask]
+
+    def __len__(self) -> int:
+        return self.sp_index.length
+
+    @property
+    def _null_fill_value(self) -> bool:
+        return self._dtype._is_na_fill_value
+
+    @property
+    def nbytes(self) -> int:
+        return self.sp_values.nbytes + self.sp_index.nbytes
+
+    @property
+    def density(self) -> float:
+        """
+        The percent of non- ``fill_value`` points, as decimal.
+
+        See Also
+        --------
+        DataFrame.sparse.from_spmatrix : Create a new DataFrame from a
+            scipy sparse matrix.
+
+        Examples
+        --------
+        >>> from pandas.arrays import SparseArray
+        >>> s = SparseArray([0, 0, 1, 1, 1], fill_value=0)
+        >>> s.density
+        0.6
+        """
+        return self.sp_index.npoints / self.sp_index.length
+
+    @property
+    def npoints(self) -> int:
+        """
+        The number of non- ``fill_value`` points.
+
+        This property returns the number of elements in the sparse series that are
+        not equal to the ``fill_value``. Sparse data structures store only the
+        non-``fill_value`` elements, reducing memory usage when the majority of
+        values are the same.
+
+        See Also
+        --------
+        Series.sparse.to_dense : Convert a Series from sparse values to dense.
+        Series.sparse.fill_value : Elements in ``data`` that are ``fill_value`` are
+            not stored.
+        Series.sparse.density : The percent of non- ``fill_value`` points, as decimal.
+
+        Examples
+        --------
+        >>> from pandas.arrays import SparseArray
+        >>> s = SparseArray([0, 0, 1, 1, 1], fill_value=0)
+        >>> s.npoints
+        3
+        """
+        return self.sp_index.npoints
+
+    # error: Return type "SparseArray" of "isna" incompatible with return type
+    # "ndarray[Any, Any] | ExtensionArraySupportsAnyAll" in supertype "ExtensionArray"
+    def isna(self) -> Self:  # type: ignore[override]
+        # If null fill value, we want SparseDtype[bool, true]
+        # to preserve the same memory usage.
+        dtype = SparseDtype(bool, self._null_fill_value)
+        if self._null_fill_value:
+            return type(self)._simple_new(isna(self.sp_values), self.sp_index, dtype)
+        mask = np.full(len(self), False, dtype=np.bool_)
+        mask[self.sp_index.indices] = isna(self.sp_values)
+        return type(self)(mask, fill_value=False, dtype=dtype)
+
+    def fillna(
+        self,
+        value,
+        limit: int | None = None,
+        copy: bool = True,
+    ) -> Self:
+        """
+        Fill missing values with `value`.
+
+        Parameters
+        ----------
+        value : scalar
+        limit : int, optional
+            Not supported for SparseArray, must be None.
+        copy: bool, default True
+            Ignored for SparseArray.
+
+        Returns
+        -------
+        SparseArray
+
+        Notes
+        -----
+        When `value` is specified, the result's ``fill_value`` depends on
+        ``self.fill_value``. The goal is to maintain low-memory use.
+
+        If ``self.fill_value`` is NA, the result dtype will be
+        ``SparseDtype(self.dtype, fill_value=value)``. This will preserve
+        amount of memory used before and after filling.
+
+        When ``self.fill_value`` is not NA, the result dtype will be
+        ``self.dtype``. Again, this preserves the amount of memory used.
+        """
+        if limit is not None:
+            raise ValueError("limit must be None")
+        new_values = np.where(isna(self.sp_values), value, self.sp_values)
+
+        if self._null_fill_value:
+            # This is essentially just updating the dtype.
+            new_dtype = SparseDtype(self.dtype.subtype, fill_value=value)
+        else:
+            new_dtype = self.dtype
+
+        return self._simple_new(new_values, self._sparse_index, new_dtype)
+
+    def shift(self, periods: int = 1, fill_value=None) -> Self:
+        if not len(self) or periods == 0:
+            return self.copy()
+
+        if isna(fill_value):
+            fill_value = self.dtype.na_value
+
+        subtype = np.result_type(fill_value, self.dtype.subtype)
+
+        if subtype != self.dtype.subtype:
+            # just coerce up front
+            arr = self.astype(SparseDtype(subtype, self.fill_value))
+        else:
+            arr = self
+
+        empty = self._from_sequence(
+            [fill_value] * min(abs(periods), len(self)), dtype=arr.dtype
+        )
+
+        if periods > 0:
+            a = empty
+            b = arr[:-periods]
+        else:
+            a = arr[abs(periods) :]
+            b = empty
+        return arr._concat_same_type([a, b])
+
+    def _first_fill_value_loc(self):
+        """
+        Get the location of the first fill value.
+
+        Returns
+        -------
+        int
+        """
+        if len(self) == 0 or self.sp_index.npoints == len(self):
+            return -1
+
+        indices = self.sp_index.indices
+        if not len(indices) or indices[0] > 0:
+            return 0
+
+        # a number larger than 1 should be appended to
+        # the last in case of fill value only appears
+        # in the tail of array
+        diff = np.r_[np.diff(indices), 2]
+        return indices[(diff > 1).argmax()] + 1
+
+    @doc(ExtensionArray.duplicated)
+    def duplicated(
+        self, keep: Literal["first", "last", False] = "first"
+    ) -> npt.NDArray[np.bool_]:
+        values = np.asarray(self)
+        mask = np.asarray(self.isna())
+        return algos.duplicated(values, keep=keep, mask=mask)
+
+    def unique(self) -> Self:
+        uniques = algos.unique(self.sp_values)
+        if len(self.sp_values) != len(self):
+            fill_loc = self._first_fill_value_loc()
+            # Inorder to align the behavior of pd.unique or
+            # pd.Series.unique, we should keep the original
+            # order, here we use unique again to find the
+            # insertion place. Since the length of sp_values
+            # is not large, maybe minor performance hurt
+            # is worthwhile to the correctness.
+            insert_loc = len(algos.unique(self.sp_values[:fill_loc]))
+            uniques = np.insert(uniques, insert_loc, self.fill_value)
+        return type(self)._from_sequence(uniques, dtype=self.dtype)
+
+    def _values_for_factorize(self):
+        # Still override this for hash_pandas_object
+        return np.asarray(self), self.fill_value
+
+    def factorize(
+        self,
+        use_na_sentinel: bool = True,
+    ) -> tuple[np.ndarray, SparseArray]:
+        # Currently, ExtensionArray.factorize -> Tuple[ndarray, EA]
+        # The sparsity on this is backwards from what Sparse would want. Want
+        # ExtensionArray.factorize -> Tuple[EA, EA]
+        # Given that we have to return a dense array of codes, why bother
+        # implementing an efficient factorize?
+        codes, uniques = algos.factorize(
+            np.asarray(self), use_na_sentinel=use_na_sentinel
+        )
+        uniques_sp = SparseArray(uniques, dtype=self.dtype)
+        return codes, uniques_sp
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        """
+        Returns a Series containing counts of unique values.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include counts of NaN, even if NaN is in sp_values.
+
+        Returns
+        -------
+        counts : Series
+        """
+        from pandas import (
+            Index,
+            Series,
+        )
+
+        keys, counts, _ = algos.value_counts_arraylike(self.sp_values, dropna=dropna)
+        fcounts = self.sp_index.ngaps
+        if fcounts > 0 and (not self._null_fill_value or not dropna):
+            mask = isna(keys) if self._null_fill_value else keys == self.fill_value
+            if mask.any():
+                counts[mask] += fcounts
+            else:
+                # error: Argument 1 to "insert" has incompatible type "Union[
+                # ExtensionArray,ndarray[Any, Any]]"; expected "Union[
+                # _SupportsArray[dtype[Any]], Sequence[_SupportsArray[dtype
+                # [Any]]], Sequence[Sequence[_SupportsArray[dtype[Any]]]],
+                # Sequence[Sequence[Sequence[_SupportsArray[dtype[Any]]]]], Sequence
+                # [Sequence[Sequence[Sequence[_SupportsArray[dtype[Any]]]]]]]"
+                keys = np.insert(keys, 0, self.fill_value)  # type: ignore[arg-type]
+                counts = np.insert(counts, 0, fcounts)
+
+        if not isinstance(keys, ABCIndex):
+            index = Index(keys)
+        else:
+            index = keys
+        return Series(counts, index=index, copy=False)
+
+    # --------
+    # Indexing
+    # --------
+    @overload
+    def __getitem__(self, key: ScalarIndexer) -> Any: ...
+
+    @overload
+    def __getitem__(
+        self,
+        key: SequenceIndexer | tuple[int | ellipsis, ...],
+    ) -> Self: ...
+
+    def __getitem__(
+        self,
+        key: PositionalIndexer | tuple[int | ellipsis, ...],
+    ) -> Self | Any:
+        if isinstance(key, tuple):
+            key = unpack_tuple_and_ellipses(key)
+            if key is Ellipsis:
+                raise ValueError("Cannot slice with Ellipsis")
+
+        if is_integer(key):
+            return self._get_val_at(key)
+        elif isinstance(key, tuple):
+            # error: Invalid index type "Tuple[Union[int, ellipsis], ...]"
+            # for "ndarray[Any, Any]"; expected type
+            # "Union[SupportsIndex, _SupportsArray[dtype[Union[bool_,
+            # integer[Any]]]], _NestedSequence[_SupportsArray[dtype[
+            # Union[bool_, integer[Any]]]]], _NestedSequence[Union[
+            # bool, int]], Tuple[Union[SupportsIndex, _SupportsArray[
+            # dtype[Union[bool_, integer[Any]]]], _NestedSequence[
+            # _SupportsArray[dtype[Union[bool_, integer[Any]]]]],
+            # _NestedSequence[Union[bool, int]]], ...]]"
+            data_slice = self.to_dense()[key]  # type: ignore[index]
+        elif isinstance(key, slice):
+            # Avoid densifying when handling contiguous slices
+            if key.step is None or key.step == 1:
+                start = 0 if key.start is None else key.start
+                if start < 0:
+                    start += len(self)
+
+                end = len(self) if key.stop is None else key.stop
+                if end < 0:
+                    end += len(self)
+
+                indices = self.sp_index.indices
+                keep_inds = np.flatnonzero((indices >= start) & (indices < end))
+                sp_vals = self.sp_values[keep_inds]
+
+                sp_index = indices[keep_inds].copy()
+
+                # If we've sliced to not include the start of the array, all our indices
+                # should be shifted. NB: here we are careful to also not shift by a
+                # negative value for a case like [0, 1][-100:] where the start index
+                # should be treated like 0
+                if start > 0:
+                    sp_index -= start
+
+                # Length of our result should match applying this slice to a range
+                # of the length of our original array
+                new_len = len(range(len(self))[key])
+                new_sp_index = make_sparse_index(new_len, sp_index, self.kind)
+                return type(self)._simple_new(sp_vals, new_sp_index, self.dtype)
+            else:
+                indices = np.arange(len(self), dtype=np.int32)[key]
+                return self.take(indices)
+
+        elif not is_list_like(key):
+            # e.g. "foo" or 2.5
+            # exception message copied from numpy
+            raise IndexError(
+                r"only integers, slices (`:`), ellipsis (`...`), numpy.newaxis "
+                r"(`None`) and integer or boolean arrays are valid indices"
+            )
+
+        else:
+            if isinstance(key, SparseArray):
+                # NOTE: If we guarantee that SparseDType(bool)
+                # has only fill_value - true, false or nan
+                # (see GH PR 44955)
+                # we can apply mask very fast:
+                if is_bool_dtype(key):
+                    if isna(key.fill_value):
+                        return self.take(key.sp_index.indices[key.sp_values])
+                    if not key.fill_value:
+                        return self.take(key.sp_index.indices)
+                    n = len(self)
+                    mask = np.full(n, True, dtype=np.bool_)
+                    mask[key.sp_index.indices] = False
+                    return self.take(np.arange(n)[mask])
+                else:
+                    key = np.asarray(key)
+
+            key = check_array_indexer(self, key)
+
+            if com.is_bool_indexer(key):
+                # mypy doesn't know we have an array here
+                key = cast(np.ndarray, key)
+                return self.take(np.arange(len(key), dtype=np.int32)[key])
+            elif hasattr(key, "__len__"):
+                return self.take(key)
+            else:
+                raise ValueError(f"Cannot slice with '{key}'")
+
+        return type(self)(data_slice, kind=self.kind)
+
+    def _get_val_at(self, loc):
+        loc = validate_insert_loc(loc, len(self))
+
+        sp_loc = self.sp_index.lookup(loc)
+        if sp_loc == -1:
+            return self.fill_value
+        else:
+            val = self.sp_values[sp_loc]
+            val = maybe_box_datetimelike(val, self.sp_values.dtype)
+            return val
+
+    def take(self, indices, *, allow_fill: bool = False, fill_value=None) -> Self:
+        if is_scalar(indices):
+            raise ValueError(f"'indices' must be an array, not a scalar '{indices}'.")
+        indices = np.asarray(indices, dtype=np.int32)
+
+        dtype = None
+        if indices.size == 0:
+            result = np.array([], dtype="object")
+            dtype = self.dtype
+        elif allow_fill:
+            result = self._take_with_fill(indices, fill_value=fill_value)
+        else:
+            return self._take_without_fill(indices)
+
+        return type(self)(
+            result, fill_value=self.fill_value, kind=self.kind, dtype=dtype
+        )
+
+    def _take_with_fill(self, indices, fill_value=None) -> np.ndarray:
+        if fill_value is None:
+            fill_value = self.dtype.na_value
+
+        if indices.min() < -1:
+            raise ValueError(
+                "Invalid value in 'indices'. Must be between -1 "
+                "and the length of the array."
+            )
+
+        if indices.max() >= len(self):
+            raise IndexError("out of bounds value in 'indices'.")
+
+        if len(self) == 0:
+            # Empty... Allow taking only if all empty
+            if (indices == -1).all():
+                dtype = np.result_type(self.sp_values, type(fill_value))
+                taken = np.empty_like(indices, dtype=dtype)
+                taken.fill(fill_value)
+                return taken
+            else:
+                raise IndexError("cannot do a non-empty take from an empty axes.")
+
+        # sp_indexer may be -1 for two reasons
+        # 1.) we took for an index of -1 (new)
+        # 2.) we took a value that was self.fill_value (old)
+        sp_indexer = self.sp_index.lookup_array(indices)
+        new_fill_indices = indices == -1
+        old_fill_indices = (sp_indexer == -1) & ~new_fill_indices
+
+        if self.sp_index.npoints == 0 and old_fill_indices.all():
+            # We've looked up all valid points on an all-sparse array.
+            taken = np.full(
+                sp_indexer.shape, fill_value=self.fill_value, dtype=self.dtype.subtype
+            )
+
+        elif self.sp_index.npoints == 0:
+            # Use the old fill_value unless we took for an index of -1
+            _dtype = np.result_type(self.dtype.subtype, type(fill_value))
+            taken = np.full(sp_indexer.shape, fill_value=fill_value, dtype=_dtype)
+            taken[old_fill_indices] = self.fill_value
+        else:
+            taken = self.sp_values.take(sp_indexer)
+
+            # Fill in two steps.
+            # Old fill values
+            # New fill values
+            # potentially coercing to a new dtype at each stage.
+
+            m0 = sp_indexer[old_fill_indices] < 0
+            m1 = sp_indexer[new_fill_indices] < 0
+
+            result_type = taken.dtype
+
+            if m0.any():
+                result_type = np.result_type(result_type, type(self.fill_value))
+                taken = taken.astype(result_type)
+                taken[old_fill_indices] = self.fill_value
+
+            if m1.any():
+                result_type = np.result_type(result_type, type(fill_value))
+                taken = taken.astype(result_type)
+                taken[new_fill_indices] = fill_value
+
+        return taken
+
+    def _take_without_fill(self, indices) -> Self:
+        to_shift = indices < 0
+
+        n = len(self)
+
+        if (indices.max() >= n) or (indices.min() < -n):
+            if n == 0:
+                raise IndexError("cannot do a non-empty take from an empty axes.")
+            raise IndexError("out of bounds value in 'indices'.")
+
+        if to_shift.any():
+            indices = indices.copy()
+            indices[to_shift] += n
+
+        sp_indexer = self.sp_index.lookup_array(indices)
+        value_mask = sp_indexer != -1
+        new_sp_values = self.sp_values[sp_indexer[value_mask]]
+
+        value_indices = np.flatnonzero(value_mask).astype(np.int32, copy=False)
+
+        new_sp_index = make_sparse_index(len(indices), value_indices, kind=self.kind)
+        return type(self)._simple_new(new_sp_values, new_sp_index, dtype=self.dtype)
+
+    def searchsorted(
+        self,
+        v: ArrayLike | object,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if get_option("performance_warnings"):
+            msg = "searchsorted requires high memory usage."
+            warnings.warn(msg, PerformanceWarning, stacklevel=find_stack_level())
+        v = np.asarray(v)
+        return np.asarray(self, dtype=self.dtype.subtype).searchsorted(v, side, sorter)
+
+    def copy(self) -> Self:
+        values = self.sp_values.copy()
+        return self._simple_new(values, self.sp_index, self.dtype)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat: Sequence[Self]) -> Self:
+        fill_value = to_concat[0].fill_value
+
+        values = []
+        length = 0
+
+        if to_concat:
+            sp_kind = to_concat[0].kind
+        else:
+            sp_kind = "integer"
+
+        sp_index: SparseIndex
+        if sp_kind == "integer":
+            indices = []
+
+            for arr in to_concat:
+                int_idx = arr.sp_index.indices.copy()
+                int_idx += length  # TODO: wraparound
+                length += arr.sp_index.length
+
+                values.append(arr.sp_values)
+                indices.append(int_idx)
+
+            data = np.concatenate(values)
+            indices_arr = np.concatenate(indices)
+            # error: Argument 2 to "IntIndex" has incompatible type
+            # "ndarray[Any, dtype[signedinteger[_32Bit]]]";
+            # expected "Sequence[int]"
+            sp_index = IntIndex(length, indices_arr)  # type: ignore[arg-type]
+
+        else:
+            # when concatenating block indices, we don't claim that you'll
+            # get an identical index as concatenating the values and then
+            # creating a new index. We don't want to spend the time trying
+            # to merge blocks across arrays in `to_concat`, so the resulting
+            # BlockIndex may have more blocks.
+            blengths = []
+            blocs = []
+
+            for arr in to_concat:
+                block_idx = arr.sp_index.to_block_index()
+
+                values.append(arr.sp_values)
+                blocs.append(block_idx.blocs.copy() + length)
+                blengths.append(block_idx.blengths)
+                length += arr.sp_index.length
+
+            data = np.concatenate(values)
+            blocs_arr = np.concatenate(blocs)
+            blengths_arr = np.concatenate(blengths)
+
+            sp_index = BlockIndex(length, blocs_arr, blengths_arr)
+
+        return cls(data, sparse_index=sp_index, fill_value=fill_value)
+
+    def astype(self, dtype: AstypeArg | None = None, copy: bool = True):
+        """
+        Change the dtype of a SparseArray.
+
+        The output will always be a SparseArray. To convert to a dense
+        ndarray with a certain dtype, use :meth:`numpy.asarray`.
+
+        Parameters
+        ----------
+        dtype : np.dtype or ExtensionDtype
+            For SparseDtype, this changes the dtype of
+            ``self.sp_values`` and the ``self.fill_value``.
+
+            For other dtypes, this only changes the dtype of
+            ``self.sp_values``.
+
+        copy : bool, default True
+            Whether to ensure a copy is made, even if not necessary.
+
+        Returns
+        -------
+        SparseArray
+
+        Examples
+        --------
+        >>> arr = pd.arrays.SparseArray([0, 0, 1, 2])
+        >>> arr
+        [0, 0, 1, 2]
+        Fill: 0
+        IntIndex
+        Indices: array([2, 3], dtype=int32)
+
+        >>> arr.astype(SparseDtype(np.dtype("int32")))
+        [0, 0, 1, 2]
+        Fill: 0
+        IntIndex
+        Indices: array([2, 3], dtype=int32)
+
+        Using a NumPy dtype with a different kind (e.g. float) will coerce
+        just ``self.sp_values``.
+
+        >>> arr.astype(SparseDtype(np.dtype("float64")))
+        ... # doctest: +NORMALIZE_WHITESPACE
+        [nan, nan, 1.0, 2.0]
+        Fill: nan
+        IntIndex
+        Indices: array([2, 3], dtype=int32)
+
+        Using a SparseDtype, you can also change the fill value as well.
+
+        >>> arr.astype(SparseDtype("float64", fill_value=0.0))
+        ... # doctest: +NORMALIZE_WHITESPACE
+        [0.0, 0.0, 1.0, 2.0]
+        Fill: 0.0
+        IntIndex
+        Indices: array([2, 3], dtype=int32)
+        """
+        if dtype == self._dtype:
+            if not copy:
+                return self
+            else:
+                return self.copy()
+
+        future_dtype = pandas_dtype(dtype)
+        if not isinstance(future_dtype, SparseDtype):
+            # GH#34457
+            values = np.asarray(self)
+            values = ensure_wrapped_if_datetimelike(values)
+            return astype_array(values, dtype=future_dtype, copy=False)
+
+        dtype = self.dtype.update_dtype(dtype)
+        subtype = pandas_dtype(dtype._subtype_with_str)
+        subtype = cast(np.dtype, subtype)  # ensured by update_dtype
+        values = ensure_wrapped_if_datetimelike(self.sp_values)
+        sp_values = astype_array(values, subtype, copy=copy)
+        sp_values = np.asarray(sp_values)
+
+        return self._simple_new(sp_values, self.sp_index, dtype)
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None) -> Self:
+        """
+        Map categories using an input mapping or function.
+
+        Parameters
+        ----------
+        mapper : dict, Series, callable
+            The correspondence from old values to new.
+        na_action : {None, 'ignore'}, default None
+            If 'ignore', propagate NA values, without passing them to the
+            mapping correspondence.
+
+        Returns
+        -------
+        SparseArray
+            The output array will have the same density as the input.
+            The output fill value will be the result of applying the
+            mapping to ``self.fill_value``
+
+        Examples
+        --------
+        >>> arr = pd.arrays.SparseArray([0, 1, 2])
+        >>> arr.map(lambda x: x + 10)
+        [10, 11, 12]
+        Fill: 10
+        IntIndex
+        Indices: array([1, 2], dtype=int32)
+
+        >>> arr.map({0: 10, 1: 11, 2: 12})
+        [10, 11, 12]
+        Fill: 10
+        IntIndex
+        Indices: array([1, 2], dtype=int32)
+
+        >>> arr.map(pd.Series([10, 11, 12], index=[0, 1, 2]))
+        [10, 11, 12]
+        Fill: 10
+        IntIndex
+        Indices: array([1, 2], dtype=int32)
+        """
+        is_map = isinstance(mapper, (abc.Mapping, ABCSeries))
+
+        fill_val = self.fill_value
+
+        if na_action is None or notna(fill_val):
+            fill_val = mapper.get(fill_val, fill_val) if is_map else mapper(fill_val)
+
+        def func(sp_val):
+            new_sp_val = mapper.get(sp_val, None) if is_map else mapper(sp_val)
+            # check identity and equality because nans are not equal to each other
+            if new_sp_val is fill_val or new_sp_val == fill_val:
+                msg = "fill value in the sparse values not supported"
+                raise ValueError(msg)
+            return new_sp_val
+
+        sp_values = [func(x) for x in self.sp_values]
+
+        return type(self)(sp_values, sparse_index=self.sp_index, fill_value=fill_val)
+
+    def to_dense(self) -> np.ndarray:
+        """
+        Convert SparseArray to a NumPy array.
+
+        Returns
+        -------
+        arr : NumPy array
+        """
+        return np.asarray(self, dtype=self.sp_values.dtype)
+
+    def _where(self, mask, value):
+        # NB: may not preserve dtype, e.g. result may be Sparse[float64]
+        #  while self is Sparse[int64]
+        naive_implementation = np.where(mask, self, value)
+        dtype = SparseDtype(naive_implementation.dtype, fill_value=self.fill_value)
+        result = type(self)._from_sequence(naive_implementation, dtype=dtype)
+        return result
+
+    # ------------------------------------------------------------------------
+    # IO
+    # ------------------------------------------------------------------------
+    def __setstate__(self, state) -> None:
+        """Necessary for making this object picklable"""
+        if isinstance(state, tuple):
+            # Compat for pandas < 0.24.0
+            nd_state, (fill_value, sp_index) = state
+            sparse_values = np.array([])
+            sparse_values.__setstate__(nd_state)
+
+            self._sparse_values = sparse_values
+            self._sparse_index = sp_index
+            self._dtype = SparseDtype(sparse_values.dtype, fill_value)
+        else:
+            self.__dict__.update(state)
+
+    def nonzero(self) -> tuple[npt.NDArray[np.int32]]:
+        if self.fill_value == 0:
+            return (self.sp_index.indices,)
+        else:
+            return (self.sp_index.indices[self.sp_values != 0],)
+
+    # ------------------------------------------------------------------------
+    # Reductions
+    # ------------------------------------------------------------------------
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        method = getattr(self, name, None)
+
+        if method is None:
+            raise TypeError(f"cannot perform {name} with type {self.dtype}")
+
+        if skipna:
+            arr = self
+        else:
+            arr = self.dropna()
+
+        result = getattr(arr, name)(**kwargs)
+
+        if keepdims:
+            return type(self)([result], dtype=self.dtype)
+        else:
+            return result
+
+    def all(self, axis=None, *args, **kwargs):
+        """
+        Tests whether all elements evaluate True
+
+        Returns
+        -------
+        all : bool
+
+        See Also
+        --------
+        numpy.all
+        """
+        nv.validate_all(args, kwargs)
+
+        values = self.sp_values
+
+        if len(values) != len(self) and not np.all(self.fill_value):
+            return False
+
+        return values.all()
+
+    def any(self, axis: AxisInt = 0, *args, **kwargs) -> bool:
+        """
+        Tests whether at least one of elements evaluate True
+
+        Returns
+        -------
+        any : bool
+
+        See Also
+        --------
+        numpy.any
+        """
+        nv.validate_any(args, kwargs)
+
+        values = self.sp_values
+
+        if len(values) != len(self) and np.any(self.fill_value):
+            return True
+
+        return values.any().item()
+
+    def sum(
+        self,
+        axis: AxisInt = 0,
+        min_count: int = 0,
+        skipna: bool = True,
+        *args,
+        **kwargs,
+    ) -> Scalar:
+        """
+        Sum of non-NA/null values
+
+        Parameters
+        ----------
+        axis : int, default 0
+            Not Used. NumPy compatibility.
+        min_count : int, default 0
+            The required number of valid values to perform the summation. If fewer
+            than ``min_count`` valid values are present, the result will be the missing
+            value indicator for subarray type.
+        *args, **kwargs
+            Not Used. NumPy compatibility.
+
+        Returns
+        -------
+        scalar
+        """
+        nv.validate_sum(args, kwargs)
+        valid_vals = self._valid_sp_values
+        sp_sum = valid_vals.sum()
+        has_na = self.sp_index.ngaps > 0 and not self._null_fill_value
+
+        if has_na and not skipna:
+            return na_value_for_dtype(self.dtype.subtype, compat=False)
+
+        if self._null_fill_value:
+            if check_below_min_count(valid_vals.shape, None, min_count):
+                return na_value_for_dtype(self.dtype.subtype, compat=False)
+            return sp_sum
+        else:
+            nsparse = self.sp_index.ngaps
+            if check_below_min_count(valid_vals.shape, None, min_count - nsparse):
+                return na_value_for_dtype(self.dtype.subtype, compat=False)
+            return sp_sum + self.fill_value * nsparse
+
+    def cumsum(self, axis: AxisInt = 0, *args, **kwargs) -> SparseArray:
+        """
+        Cumulative sum of non-NA/null values.
+
+        When performing the cumulative summation, any non-NA/null values will
+        be skipped. The resulting SparseArray will preserve the locations of
+        NaN values, but the fill value will be `np.nan` regardless.
+
+        Parameters
+        ----------
+        axis : int or None
+            Axis over which to perform the cumulative summation. If None,
+            perform cumulative summation over flattened array.
+
+        Returns
+        -------
+        cumsum : SparseArray
+        """
+        nv.validate_cumsum(args, kwargs)
+
+        if axis is not None and axis >= self.ndim:  # Mimic ndarray behaviour.
+            raise ValueError(f"axis(={axis}) out of bounds")
+
+        if not self._null_fill_value:
+            return SparseArray(self.to_dense()).cumsum()
+
+        return SparseArray(
+            self.sp_values.cumsum(),
+            sparse_index=self.sp_index,
+            fill_value=self.fill_value,
+        )
+
+    def mean(self, axis: Axis = 0, *args, **kwargs):
+        """
+        Mean of non-NA/null values
+
+        Returns
+        -------
+        mean : float
+        """
+        nv.validate_mean(args, kwargs)
+        valid_vals = self._valid_sp_values
+        sp_sum = valid_vals.sum()
+        ct = len(valid_vals)
+
+        if self._null_fill_value:
+            return sp_sum / ct
+        else:
+            nsparse = self.sp_index.ngaps
+            return (sp_sum + self.fill_value * nsparse) / (ct + nsparse)
+
+    def max(self, *, axis: AxisInt | None = None, skipna: bool = True):
+        """
+        Max of array values, ignoring NA values if specified.
+
+        Parameters
+        ----------
+        axis : int, default 0
+            Not Used. NumPy compatibility.
+        skipna : bool, default True
+            Whether to ignore NA values.
+
+        Returns
+        -------
+        scalar
+        """
+        nv.validate_minmax_axis(axis, self.ndim)
+        return self._min_max("max", skipna=skipna)
+
+    def min(self, *, axis: AxisInt | None = None, skipna: bool = True):
+        """
+        Min of array values, ignoring NA values if specified.
+
+        Parameters
+        ----------
+        axis : int, default 0
+            Not Used. NumPy compatibility.
+        skipna : bool, default True
+            Whether to ignore NA values.
+
+        Returns
+        -------
+        scalar
+        """
+        nv.validate_minmax_axis(axis, self.ndim)
+        return self._min_max("min", skipna=skipna)
+
+    def _min_max(self, kind: Literal["min", "max"], skipna: bool) -> Scalar:
+        """
+        Min/max of non-NA/null values
+
+        Parameters
+        ----------
+        kind : {"min", "max"}
+        skipna : bool
+
+        Returns
+        -------
+        scalar
+        """
+        valid_vals = self._valid_sp_values
+        has_nonnull_fill_vals = not self._null_fill_value and self.sp_index.ngaps > 0
+
+        if len(valid_vals) > 0:
+            sp_min_max = getattr(valid_vals, kind)()
+
+            # If a non-null fill value is currently present, it might be the min/max
+            if has_nonnull_fill_vals:
+                func = max if kind == "max" else min
+                return func(sp_min_max, self.fill_value)
+            elif skipna:
+                return sp_min_max
+            elif self.sp_index.ngaps == 0:
+                # No NAs present
+                return sp_min_max
+            else:
+                return na_value_for_dtype(self.dtype.subtype, compat=False)
+        elif has_nonnull_fill_vals:
+            return self.fill_value
+        else:
+            return na_value_for_dtype(self.dtype.subtype, compat=False)
+
+    def _argmin_argmax(self, kind: Literal["argmin", "argmax"]) -> int:
+        values = self._sparse_values
+        index = self._sparse_index.indices
+        mask = np.asarray(isna(values))
+        func = np.argmax if kind == "argmax" else np.argmin
+
+        idx = np.arange(values.shape[0])
+        non_nans = values[~mask]
+        non_nan_idx = idx[~mask]
+
+        _candidate = non_nan_idx[func(non_nans)]
+        candidate = index[_candidate]
+
+        if isna(self.fill_value):
+            return candidate
+        if kind == "argmin" and self[candidate] < self.fill_value:
+            return candidate
+        if kind == "argmax" and self[candidate] > self.fill_value:
+            return candidate
+        _loc = self._first_fill_value_loc()
+        if _loc == -1:
+            # fill_value doesn't exist
+            return candidate
+        else:
+            return _loc
+
+    def argmax(self, skipna: bool = True) -> int:
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return self._argmin_argmax("argmax")
+
+    def argmin(self, skipna: bool = True) -> int:
+        validate_bool_kwarg(skipna, "skipna")
+        if not skipna and self._hasna:
+            raise ValueError("Encountered an NA value with skipna=False")
+        return self._argmin_argmax("argmin")
+
+    # ------------------------------------------------------------------------
+    # Ufuncs
+    # ------------------------------------------------------------------------
+
+    _HANDLED_TYPES = (np.ndarray, numbers.Number)
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        out = kwargs.get("out", ())
+
+        for x in inputs + out:
+            if not isinstance(x, self._HANDLED_TYPES + (SparseArray,)):
+                return NotImplemented
+
+        # for binary ops, use our custom dunder methods
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            # e.g. tests.arrays.sparse.test_arithmetics.test_ndarray_inplace
+            res = arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            return res
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                # e.g. tests.series.test_ufunc.TestNumpyReductions
+                return result
+
+        if len(inputs) == 1:
+            # No alignment necessary.
+            sp_values = getattr(ufunc, method)(self.sp_values, **kwargs)
+            fill_value = getattr(ufunc, method)(self.fill_value, **kwargs)
+
+            if ufunc.nout > 1:
+                # multiple outputs. e.g. modf
+                arrays = tuple(
+                    self._simple_new(
+                        sp_value, self.sp_index, SparseDtype(sp_value.dtype, fv)
+                    )
+                    for sp_value, fv in zip(sp_values, fill_value)
+                )
+                return arrays
+            elif method == "reduce":
+                # e.g. reductions
+                return sp_values
+
+            return self._simple_new(
+                sp_values, self.sp_index, SparseDtype(sp_values.dtype, fill_value)
+            )
+
+        new_inputs = tuple(np.asarray(x) for x in inputs)
+        result = getattr(ufunc, method)(*new_inputs, **kwargs)
+        if out:
+            if len(out) == 1:
+                out = out[0]
+            return out
+
+        if ufunc.nout > 1:
+            return tuple(type(self)(x) for x in result)
+        elif method == "at":
+            # no return value
+            return None
+        else:
+            return type(self)(result)
+
+    # ------------------------------------------------------------------------
+    # Ops
+    # ------------------------------------------------------------------------
+
+    def _arith_method(self, other, op):
+        op_name = op.__name__
+
+        if isinstance(other, SparseArray):
+            return _sparse_array_op(self, other, op, op_name)
+
+        elif is_scalar(other):
+            with np.errstate(all="ignore"):
+                fill = op(_get_fill(self), np.asarray(other))
+                result = op(self.sp_values, other)
+
+            if op_name == "divmod":
+                left, right = result
+                lfill, rfill = fill
+                return (
+                    _wrap_result(op_name, left, self.sp_index, lfill),
+                    _wrap_result(op_name, right, self.sp_index, rfill),
+                )
+
+            return _wrap_result(op_name, result, self.sp_index, fill)
+
+        else:
+            other = np.asarray(other)
+            with np.errstate(all="ignore"):
+                if len(self) != len(other):
+                    raise AssertionError(
+                        f"length mismatch: {len(self)} vs. {len(other)}"
+                    )
+                if not isinstance(other, SparseArray):
+                    dtype = getattr(other, "dtype", None)
+                    other = SparseArray(other, fill_value=self.fill_value, dtype=dtype)
+                return _sparse_array_op(self, other, op, op_name)
+
+    def _cmp_method(self, other, op) -> SparseArray:
+        if not is_scalar(other) and not isinstance(other, type(self)):
+            # convert list-like to ndarray
+            other = np.asarray(other)
+
+        if isinstance(other, np.ndarray):
+            # TODO: make this more flexible than just ndarray...
+            other = SparseArray(other, fill_value=self.fill_value)
+
+        if isinstance(other, SparseArray):
+            if len(self) != len(other):
+                raise ValueError(
+                    f"operands have mismatched length {len(self)} and {len(other)}"
+                )
+
+            op_name = op.__name__.strip("_")
+            return _sparse_array_op(self, other, op, op_name)
+        else:
+            # scalar
+            fill_value = op(self.fill_value, other)
+            result = np.full(len(self), fill_value, dtype=np.bool_)
+            result[self.sp_index.indices] = op(self.sp_values, other)
+
+            return type(self)(
+                result,
+                fill_value=fill_value,
+                dtype=np.bool_,
+            )
+
+    _logical_method = _cmp_method
+
+    def _unary_method(self, op) -> SparseArray:
+        fill_value = op(np.array(self.fill_value)).item()
+        dtype = SparseDtype(self.dtype.subtype, fill_value)
+        # NOTE: if fill_value doesn't change
+        # we just have to apply op to sp_values
+        if isna(self.fill_value) or fill_value == self.fill_value:
+            values = op(self.sp_values)
+            return type(self)._simple_new(values, self.sp_index, self.dtype)
+        # In the other case we have to recalc indexes
+        return type(self)(op(self.to_dense()), dtype=dtype)
+
+    def __pos__(self) -> SparseArray:
+        return self._unary_method(operator.pos)
+
+    def __neg__(self) -> SparseArray:
+        return self._unary_method(operator.neg)
+
+    def __invert__(self) -> SparseArray:
+        return self._unary_method(operator.invert)
+
+    def __abs__(self) -> SparseArray:
+        return self._unary_method(operator.abs)
+
+    # ----------
+    # Formatting
+    # -----------
+    def __repr__(self) -> str:
+        pp_str = printing.pprint_thing(self)
+        pp_fill = printing.pprint_thing(self.fill_value)
+        pp_index = printing.pprint_thing(self.sp_index)
+        return f"{pp_str}\nFill: {pp_fill}\n{pp_index}"
+
+    # error: Return type "None" of "_formatter" incompatible with return
+    # type "Callable[[Any], str | None]" in supertype "ExtensionArray"
+    def _formatter(self, boxed: bool = False) -> None:  # type: ignore[override]
+        # Defer to the formatter from the GenericArrayFormatter calling us.
+        # This will infer the correct formatter from the dtype of the values.
+        return None
+
+
+def _make_sparse(
+    arr: np.ndarray,
+    kind: SparseIndexKind = "block",
+    fill_value=None,
+    dtype: np.dtype | None = None,
+):
+    """
+    Convert ndarray to sparse format
+
+    Parameters
+    ----------
+    arr : ndarray
+    kind : {'block', 'integer'}
+    fill_value : NaN or another value
+    dtype : np.dtype, optional
+    copy : bool, default False
+
+    Returns
+    -------
+    (sparse_values, index, fill_value) : (ndarray, SparseIndex, Scalar)
+    """
+    assert isinstance(arr, np.ndarray)
+
+    if arr.ndim > 1:
+        raise TypeError("expected dimension <= 1 data")
+
+    if fill_value is None:
+        fill_value = na_value_for_dtype(arr.dtype)
+
+    if isna(fill_value):
+        mask = notna(arr)
+    else:
+        # cast to object comparison to be safe
+        if is_string_dtype(arr.dtype):
+            arr = arr.astype(object)
+
+        if is_object_dtype(arr.dtype):
+            # element-wise equality check method in numpy doesn't treat
+            # each element type, eg. 0, 0.0, and False are treated as
+            # same. So we have to check the both of its type and value.
+            mask = splib.make_mask_object_ndarray(arr, fill_value)
+        else:
+            mask = arr != fill_value
+
+    length = len(arr)
+    if length != len(mask):
+        # the arr is a SparseArray
+        indices = mask.sp_index.indices
+    else:
+        indices = mask.nonzero()[0].astype(np.int32)
+
+    index = make_sparse_index(length, indices, kind)
+    sparsified_values = arr[mask]
+    if dtype is not None:
+        sparsified_values = ensure_wrapped_if_datetimelike(sparsified_values)
+        sparsified_values = astype_array(sparsified_values, dtype=dtype)
+        sparsified_values = np.asarray(sparsified_values)
+
+    # TODO: copy
+    return sparsified_values, index, fill_value
+
+
+@overload
+def make_sparse_index(length: int, indices, kind: Literal["block"]) -> BlockIndex: ...
+
+
+@overload
+def make_sparse_index(length: int, indices, kind: Literal["integer"]) -> IntIndex: ...
+
+
+def make_sparse_index(length: int, indices, kind: SparseIndexKind) -> SparseIndex:
+    index: SparseIndex
+    if kind == "block":
+        locs, lens = splib.get_blocks(indices)
+        index = BlockIndex(length, locs, lens)
+    elif kind == "integer":
+        index = IntIndex(length, indices)
+    else:  # pragma: no cover
+        raise ValueError("must be block or integer type")
+    return index
diff --git a/pandas/pandas/core/arrays/sparse/scipy_sparse.py b/pandas/pandas/core/arrays/sparse/scipy_sparse.py
new file mode 100644
index 0000000000000..cc9fd2d5fb8b0
--- /dev/null
+++ b/pandas/pandas/core/arrays/sparse/scipy_sparse.py
@@ -0,0 +1,208 @@
+"""
+Interaction with scipy.sparse matrices.
+
+Currently only includes to_coo helpers.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.missing import notna
+
+from pandas.core.algorithms import factorize
+from pandas.core.indexes.api import MultiIndex
+from pandas.core.series import Series
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+    import numpy as np
+    import scipy.sparse
+
+    from pandas._typing import (
+        IndexLabel,
+        npt,
+    )
+
+
+def _check_is_partition(parts: Iterable, whole: Iterable) -> None:
+    whole = set(whole)
+    parts = [set(x) for x in parts]
+    if set.intersection(*parts) != set():
+        raise ValueError("Is not a partition because intersection is not null.")
+    if set.union(*parts) != whole:
+        raise ValueError("Is not a partition because union is not the whole.")
+
+
+def _levels_to_axis(
+    ss,
+    levels: tuple[int] | list[int],
+    valid_ilocs: npt.NDArray[np.intp],
+    sort_labels: bool = False,
+) -> tuple[npt.NDArray[np.intp], list[IndexLabel]]:
+    """
+    For a MultiIndexed sparse Series `ss`, return `ax_coords` and `ax_labels`,
+    where `ax_coords` are the coordinates along one of the two axes of the
+    destination sparse matrix, and `ax_labels` are the labels from `ss`' Index
+    which correspond to these coordinates.
+
+    Parameters
+    ----------
+    ss : Series
+    levels : tuple/list
+    valid_ilocs : numpy.ndarray
+        Array of integer positions of valid values for the sparse matrix in ss.
+    sort_labels : bool, default False
+        Sort the axis labels before forming the sparse matrix. When `levels`
+        refers to a single level, set to True for a faster execution.
+
+    Returns
+    -------
+    ax_coords : numpy.ndarray (axis coordinates)
+    ax_labels : list (axis labels)
+    """
+    # Since the labels are sorted in `Index.levels`, when we wish to sort and
+    # there is only one level of the MultiIndex for this axis, the desired
+    # output can be obtained in the following simpler, more efficient way.
+    if sort_labels and len(levels) == 1:
+        ax_coords = ss.index.codes[levels[0]][valid_ilocs]
+        ax_labels = ss.index.levels[levels[0]]
+
+    else:
+        levels_values = lib.fast_zip(
+            [ss.index.get_level_values(lvl).to_numpy() for lvl in levels]
+        )
+        codes, ax_labels = factorize(levels_values, sort=sort_labels)
+        ax_coords = codes[valid_ilocs]
+
+    ax_labels = ax_labels.tolist()
+    return ax_coords, ax_labels
+
+
+def _to_ijv(
+    ss,
+    row_levels: tuple[int] | list[int] = (0,),
+    column_levels: tuple[int] | list[int] = (1,),
+    sort_labels: bool = False,
+) -> tuple[
+    np.ndarray,
+    npt.NDArray[np.intp],
+    npt.NDArray[np.intp],
+    list[IndexLabel],
+    list[IndexLabel],
+]:
+    """
+    For an arbitrary MultiIndexed sparse Series return (v, i, j, ilabels,
+    jlabels) where (v, (i, j)) is suitable for passing to scipy.sparse.coo
+    constructor, and ilabels and jlabels are the row and column labels
+    respectively.
+
+    Parameters
+    ----------
+    ss : Series
+    row_levels : tuple/list
+    column_levels : tuple/list
+    sort_labels : bool, default False
+        Sort the row and column labels before forming the sparse matrix.
+        When `row_levels` and/or `column_levels` refer to a single level,
+        set to `True` for a faster execution.
+
+    Returns
+    -------
+    values : numpy.ndarray
+        Valid values to populate a sparse matrix, extracted from
+        ss.
+    i_coords : numpy.ndarray (row coordinates of the values)
+    j_coords : numpy.ndarray (column coordinates of the values)
+    i_labels : list (row labels)
+    j_labels : list (column labels)
+    """
+    # index and column levels must be a partition of the index
+    _check_is_partition([row_levels, column_levels], range(ss.index.nlevels))
+    # From the sparse Series, get the integer indices and data for valid sparse
+    # entries.
+    sp_vals = ss.array.sp_values
+    na_mask = notna(sp_vals)
+    values = sp_vals[na_mask]
+    valid_ilocs = ss.array.sp_index.indices[na_mask]
+
+    i_coords, i_labels = _levels_to_axis(
+        ss, row_levels, valid_ilocs, sort_labels=sort_labels
+    )
+
+    j_coords, j_labels = _levels_to_axis(
+        ss, column_levels, valid_ilocs, sort_labels=sort_labels
+    )
+
+    return values, i_coords, j_coords, i_labels, j_labels
+
+
+def sparse_series_to_coo(
+    ss: Series,
+    row_levels: Iterable[int] = (0,),
+    column_levels: Iterable[int] = (1,),
+    sort_labels: bool = False,
+) -> tuple[scipy.sparse.coo_matrix, list[IndexLabel], list[IndexLabel]]:
+    """
+    Convert a sparse Series to a scipy.sparse.coo_matrix using index
+    levels row_levels, column_levels as the row and column
+    labels respectively. Returns the sparse_matrix, row and column labels.
+    """
+    import scipy.sparse
+
+    if ss.index.nlevels < 2:
+        raise ValueError("to_coo requires MultiIndex with nlevels >= 2.")
+    if not ss.index.is_unique:
+        raise ValueError(
+            "Duplicate index entries are not allowed in to_coo transformation."
+        )
+
+    # to keep things simple, only rely on integer indexing (not labels)
+    row_levels = [ss.index._get_level_number(x) for x in row_levels]
+    column_levels = [ss.index._get_level_number(x) for x in column_levels]
+
+    v, i, j, rows, columns = _to_ijv(
+        ss, row_levels=row_levels, column_levels=column_levels, sort_labels=sort_labels
+    )
+    sparse_matrix = scipy.sparse.coo_matrix(
+        (v, (i, j)), shape=(len(rows), len(columns))
+    )
+    return sparse_matrix, rows, columns
+
+
+def coo_to_sparse_series(
+    A: scipy.sparse.coo_matrix, dense_index: bool = False
+) -> Series:
+    """
+    Convert a scipy.sparse.coo_matrix to a Series with type sparse.
+
+    Parameters
+    ----------
+    A : scipy.sparse.coo_matrix
+    dense_index : bool, default False
+
+    Returns
+    -------
+    Series
+
+    Raises
+    ------
+    TypeError if A is not a coo_matrix
+    """
+    from pandas import SparseDtype
+
+    try:
+        ser = Series(A.data, MultiIndex.from_arrays((A.row, A.col)), copy=False)
+    except AttributeError as err:
+        raise TypeError(
+            f"Expected coo_matrix. Got {type(A).__name__} instead."
+        ) from err
+    ser = ser.sort_index()
+    ser = ser.astype(SparseDtype(ser.dtype))
+    if dense_index:
+        ind = MultiIndex.from_product([A.row, A.col])
+        ser = ser.reindex(ind)
+    return ser
diff --git a/pandas/pandas/core/arrays/string_.py b/pandas/pandas/core/arrays/string_.py
new file mode 100644
index 0000000000000..7227ea77ca433
--- /dev/null
+++ b/pandas/pandas/core/arrays/string_.py
@@ -0,0 +1,1084 @@
+from __future__ import annotations
+
+from functools import partial
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import (
+    get_option,
+    using_string_dtype,
+)
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas._libs.arrays import NDArrayBacked
+from pandas._libs.lib import ensure_string_array
+from pandas.compat import (
+    HAS_PYARROW,
+    pa_version_under10p1,
+)
+from pandas.compat.numpy import function as nv
+from pandas.util._decorators import (
+    doc,
+    set_module,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import (
+    ExtensionDtype,
+    StorageExtensionDtype,
+    register_extension_dtype,
+)
+from pandas.core.dtypes.common import (
+    is_array_like,
+    is_bool_dtype,
+    is_integer_dtype,
+    is_object_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+
+from pandas.core import (
+    missing,
+    nanops,
+    ops,
+)
+from pandas.core.algorithms import isin
+from pandas.core.array_algos import masked_reductions
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.arrays.floating import (
+    FloatingArray,
+    FloatingDtype,
+)
+from pandas.core.arrays.integer import (
+    IntegerArray,
+    IntegerDtype,
+)
+from pandas.core.arrays.numpy_ import NumpyExtensionArray
+from pandas.core.construction import extract_array
+from pandas.core.indexers import check_array_indexer
+from pandas.core.missing import isna
+
+from pandas.io.formats import printing
+
+if TYPE_CHECKING:
+    import pyarrow
+
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        Dtype,
+        DtypeObj,
+        NumpySorter,
+        NumpyValueArrayLike,
+        Scalar,
+        Self,
+        npt,
+        type_t,
+    )
+
+    from pandas import Series
+
+
+@set_module("pandas")
+@register_extension_dtype
+class StringDtype(StorageExtensionDtype):
+    """
+    Extension dtype for string data.
+
+    .. warning::
+
+       StringDtype is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Parameters
+    ----------
+    storage : {"python", "pyarrow"}, optional
+        If not given, the value of ``pd.options.mode.string_storage``.
+    na_value : {np.nan, pd.NA}, default pd.NA
+        Whether the dtype follows NaN or NA missing value semantics.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    BooleanDtype : Extension dtype for boolean data.
+
+    Examples
+    --------
+    >>> pd.StringDtype()
+    string[python]
+
+    >>> pd.StringDtype(storage="pyarrow")
+    string[pyarrow]
+    """
+
+    @property
+    def name(self) -> str:  # type: ignore[override]
+        if self._na_value is libmissing.NA:
+            return "string"
+        else:
+            return "str"
+
+    #: StringDtype().na_value uses pandas.NA except the implementation that
+    # follows NumPy semantics, which uses nan.
+    @property
+    def na_value(self) -> libmissing.NAType | float:  # type: ignore[override]
+        return self._na_value
+
+    _metadata = ("storage", "_na_value")  # type: ignore[assignment]
+
+    def __init__(
+        self,
+        storage: str | None = None,
+        na_value: libmissing.NAType | float = libmissing.NA,
+    ) -> None:
+        # infer defaults
+        if storage is None:
+            if na_value is not libmissing.NA:
+                storage = get_option("mode.string_storage")
+                if storage == "auto":
+                    if HAS_PYARROW:
+                        storage = "pyarrow"
+                    else:
+                        storage = "python"
+            else:
+                storage = get_option("mode.string_storage")
+                if storage == "auto":
+                    storage = "python"
+
+        if storage == "pyarrow_numpy":
+            warnings.warn(
+                "The 'pyarrow_numpy' storage option name is deprecated and will be "
+                'removed in pandas 3.0. Use \'pd.StringDtype(storage="pyarrow", '
+                "na_value-np.nan)' to construct the same dtype.\nOr enable the "
+                "'pd.options.future.infer_string = True' option globally and use "
+                'the "str" alias as a shorthand notation to specify a dtype '
+                '(instead of "string[pyarrow_numpy]").',
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+            storage = "pyarrow"
+            na_value = np.nan
+
+        # validate options
+        if storage not in {"python", "pyarrow"}:
+            raise ValueError(
+                f"Storage must be 'python' or 'pyarrow'. Got {storage} instead."
+            )
+        if storage == "pyarrow" and pa_version_under10p1:
+            raise ImportError(
+                "pyarrow>=10.0.1 is required for PyArrow backed StringArray."
+            )
+
+        if isinstance(na_value, float) and np.isnan(na_value):
+            # when passed a NaN value, always set to np.nan to ensure we use
+            # a consistent NaN value (and we can use `dtype.na_value is np.nan`)
+            na_value = np.nan
+        elif na_value is not libmissing.NA:
+            raise ValueError(f"'na_value' must be np.nan or pd.NA, got {na_value}")
+
+        self.storage = cast(str, storage)
+        self._na_value = na_value
+
+    def __repr__(self) -> str:
+        if self._na_value is libmissing.NA:
+            return f"{self.name}[{self.storage}]"
+        else:
+            # TODO add more informative repr
+            return self.name
+
+    def __eq__(self, other: object) -> bool:
+        # we need to override the base class __eq__ because na_value (NA or NaN)
+        # cannot be checked with normal `==`
+        if isinstance(other, str):
+            # TODO should dtype == "string" work for the NaN variant?
+            if other == "string" or other == self.name:
+                return True
+            try:
+                other = self.construct_from_string(other)
+            except (TypeError, ImportError):
+                # TypeError if `other` is not a valid string for StringDtype
+                # ImportError if pyarrow is not installed for "string[pyarrow]"
+                return False
+        if isinstance(other, type(self)):
+            return self.storage == other.storage and self.na_value is other.na_value
+        return False
+
+    def __hash__(self) -> int:
+        # need to override __hash__ as well because of overriding __eq__
+        return super().__hash__()
+
+    def __reduce__(self):
+        return StringDtype, (self.storage, self.na_value)
+
+    @property
+    def type(self) -> type[str]:
+        return str
+
+    @classmethod
+    def construct_from_string(cls, string) -> Self:
+        """
+        Construct a StringDtype from a string.
+
+        Parameters
+        ----------
+        string : str
+            The type of the name. The storage type will be taking from `string`.
+            Valid options and their storage types are
+
+            ========================== ==============================================
+            string                     result storage
+            ========================== ==============================================
+            ``'string'``               pd.options.mode.string_storage, default python
+            ``'string[python]'``       python
+            ``'string[pyarrow]'``      pyarrow
+            ========================== ==============================================
+
+        Returns
+        -------
+        StringDtype
+
+        Raise
+        -----
+        TypeError
+            If the string is not a valid option.
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+        if string == "string":
+            return cls()
+        elif string == "str" and using_string_dtype():
+            return cls(na_value=np.nan)
+        elif string == "string[python]":
+            return cls(storage="python")
+        elif string == "string[pyarrow]":
+            return cls(storage="pyarrow")
+        elif string == "string[pyarrow_numpy]":
+            # this is deprecated in the dtype __init__, remove this in pandas 3.0
+            return cls(storage="pyarrow_numpy")
+        else:
+            raise TypeError(f"Cannot construct a '{cls.__name__}' from '{string}'")
+
+    # https://github.com/pandas-dev/pandas/issues/36126
+    # error: Signature of "construct_array_type" incompatible with supertype
+    # "ExtensionDtype"
+    def construct_array_type(  # type: ignore[override]
+        self,
+    ) -> type_t[BaseStringArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays.string_arrow import (
+            ArrowStringArray,
+            ArrowStringArrayNumpySemantics,
+        )
+
+        if self.storage == "python" and self._na_value is libmissing.NA:
+            return StringArray
+        elif self.storage == "pyarrow" and self._na_value is libmissing.NA:
+            return ArrowStringArray
+        elif self.storage == "python":
+            return StringArrayNumpySemantics
+        else:
+            return ArrowStringArrayNumpySemantics
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        storages = set()
+        na_values = set()
+
+        for dtype in dtypes:
+            if isinstance(dtype, StringDtype):
+                storages.add(dtype.storage)
+                na_values.add(dtype.na_value)
+            elif isinstance(dtype, np.dtype) and dtype.kind in ("U", "T"):
+                continue
+            else:
+                return None
+
+        if len(storages) == 2:
+            # if both python and pyarrow storage -> priority to pyarrow
+            storage = "pyarrow"
+        else:
+            storage = next(iter(storages))  # type: ignore[assignment]
+
+        na_value: libmissing.NAType | float
+        if len(na_values) == 2:
+            # if both NaN and NA -> priority to NA
+            na_value = libmissing.NA
+        else:
+            na_value = next(iter(na_values))
+
+        return StringDtype(storage=storage, na_value=na_value)
+
+    def __from_arrow__(
+        self, array: pyarrow.Array | pyarrow.ChunkedArray
+    ) -> BaseStringArray:
+        """
+        Construct StringArray from pyarrow Array/ChunkedArray.
+        """
+        if self.storage == "pyarrow":
+            if self._na_value is libmissing.NA:
+                from pandas.core.arrays.string_arrow import ArrowStringArray
+
+                return ArrowStringArray(array)
+            else:
+                from pandas.core.arrays.string_arrow import (
+                    ArrowStringArrayNumpySemantics,
+                )
+
+                return ArrowStringArrayNumpySemantics(array)
+
+        else:
+            import pyarrow
+
+            if isinstance(array, pyarrow.Array):
+                chunks = [array]
+            else:
+                # pyarrow.ChunkedArray
+                chunks = array.chunks
+
+            results = []
+            for arr in chunks:
+                # convert chunk by chunk to numpy and concatenate then, to avoid
+                # overflow for large string data when concatenating the pyarrow arrays
+                arr = arr.to_numpy(zero_copy_only=False)
+                arr = ensure_string_array(arr, na_value=self.na_value)
+                results.append(arr)
+
+        if len(chunks) == 0:
+            arr = np.array([], dtype=object)
+        else:
+            arr = np.concatenate(results)
+
+        # Bypass validation inside StringArray constructor, see GH#47781
+        new_string_array = StringArray.__new__(StringArray)
+        NDArrayBacked.__init__(new_string_array, arr, self)
+        return new_string_array
+
+
+class BaseStringArray(ExtensionArray):
+    """
+    Mixin class for StringArray, ArrowStringArray.
+    """
+
+    dtype: StringDtype
+
+    @doc(ExtensionArray.tolist)
+    def tolist(self) -> list:
+        if self.ndim > 1:
+            return [x.tolist() for x in self]
+        return list(self.to_numpy())
+
+    @classmethod
+    def _from_scalars(cls, scalars, dtype: DtypeObj) -> Self:
+        if lib.infer_dtype(scalars, skipna=True) not in ["string", "empty"]:
+            # TODO: require any NAs be valid-for-string
+            raise ValueError
+        return cls._from_sequence(scalars, dtype=dtype)
+
+    def _formatter(self, boxed: bool = False):
+        formatter = partial(
+            printing.pprint_thing,
+            escape_chars=("\t", "\r", "\n"),
+            quote_strings=not boxed,
+        )
+        return formatter
+
+    def _str_map(
+        self,
+        f,
+        na_value=lib.no_default,
+        dtype: Dtype | None = None,
+        convert: bool = True,
+    ):
+        if self.dtype.na_value is np.nan:
+            return self._str_map_nan_semantics(f, na_value=na_value, dtype=dtype)
+
+        from pandas.arrays import BooleanArray
+
+        if dtype is None:
+            dtype = self.dtype
+        if na_value is lib.no_default:
+            na_value = self.dtype.na_value
+
+        mask = isna(self)
+        arr = np.asarray(self)
+
+        if is_integer_dtype(dtype) or is_bool_dtype(dtype):
+            constructor: type[IntegerArray | BooleanArray]
+            if is_integer_dtype(dtype):
+                constructor = IntegerArray
+            else:
+                constructor = BooleanArray
+
+            na_value_is_na = isna(na_value)
+            if na_value_is_na:
+                na_value = 1
+            elif dtype == np.dtype("bool"):
+                # GH#55736
+                na_value = bool(na_value)
+            result = lib.map_infer_mask(
+                arr,
+                f,
+                mask.view("uint8"),
+                convert=False,
+                na_value=na_value,
+                # error: Argument 1 to "dtype" has incompatible type
+                # "Union[ExtensionDtype, str, dtype[Any], Type[object]]"; expected
+                # "Type[object]"
+                dtype=np.dtype(cast(type, dtype)),
+            )
+
+            if not na_value_is_na:
+                mask[:] = False
+
+            return constructor(result, mask)
+
+        else:
+            return self._str_map_str_or_object(dtype, na_value, arr, f, mask)
+
+    def _str_map_str_or_object(
+        self,
+        dtype,
+        na_value,
+        arr: np.ndarray,
+        f,
+        mask: npt.NDArray[np.bool_],
+    ):
+        # _str_map helper for case where dtype is either string dtype or object
+        if is_string_dtype(dtype) and not is_object_dtype(dtype):
+            # i.e. StringDtype
+            result = lib.map_infer_mask(
+                arr, f, mask.view("uint8"), convert=False, na_value=na_value
+            )
+            if self.dtype.storage == "pyarrow":
+                import pyarrow as pa
+
+                result = pa.array(
+                    result, mask=mask, type=pa.large_string(), from_pandas=True
+                )
+            # error: Too many arguments for "BaseStringArray"
+            return type(self)(result)  # type: ignore[call-arg]
+
+        else:
+            # This is when the result type is object. We reach this when
+            # -> We know the result type is truly object (e.g. .encode returns bytes
+            #    or .findall returns a list).
+            # -> We don't know the result type. E.g. `.get` can return anything.
+            return lib.map_infer_mask(arr, f, mask.view("uint8"))
+
+    def _str_map_nan_semantics(
+        self, f, na_value=lib.no_default, dtype: Dtype | None = None
+    ):
+        if dtype is None:
+            dtype = self.dtype
+        if na_value is lib.no_default:
+            if is_bool_dtype(dtype):
+                # NaN propagates as False
+                na_value = False
+            else:
+                na_value = self.dtype.na_value
+
+        mask = isna(self)
+        arr = np.asarray(self)
+
+        if is_integer_dtype(dtype) or is_bool_dtype(dtype):
+            na_value_is_na = isna(na_value)
+            if na_value_is_na:
+                if is_integer_dtype(dtype):
+                    na_value = 0
+                else:
+                    # NaN propagates as False
+                    na_value = False
+
+            result = lib.map_infer_mask(
+                arr,
+                f,
+                mask.view("uint8"),
+                convert=False,
+                na_value=na_value,
+                dtype=np.dtype(cast(type, dtype)),
+            )
+            if na_value_is_na and is_integer_dtype(dtype) and mask.any():
+                # TODO: we could alternatively do this check before map_infer_mask
+                #  and adjust the dtype/na_value we pass there. Which is more
+                #  performant?
+                result = result.astype("float64")
+                result[mask] = np.nan
+
+            return result
+
+        else:
+            return self._str_map_str_or_object(dtype, na_value, arr, f, mask)
+
+    def view(self, dtype: Dtype | None = None) -> ArrayLike:
+        if dtype is not None:
+            raise TypeError("Cannot change data-type for string array.")
+        return super().view(dtype=dtype)
+
+
+# error: Definition of "_concat_same_type" in base class "NDArrayBacked" is
+# incompatible with definition in base class "ExtensionArray"
+class StringArray(BaseStringArray, NumpyExtensionArray):  # type: ignore[misc]
+    """
+    Extension array for string data.
+
+    .. warning::
+
+       StringArray is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Parameters
+    ----------
+    values : array-like
+        The array of data.
+
+        .. warning::
+
+           Currently, this expects an object-dtype ndarray
+           where the elements are Python strings
+           or nan-likes (``None``, ``np.nan``, ``NA``).
+           This may change without warning in the future. Use
+           :meth:`pandas.array` with ``dtype="string"`` for a stable way of
+           creating a `StringArray` from any sequence.
+
+        .. versionchanged:: 1.5.0
+
+           StringArray now accepts array-likes containing
+           nan-likes(``None``, ``np.nan``) for the ``values`` parameter
+           in addition to strings and :attr:`pandas.NA`
+
+    copy : bool, default False
+        Whether to copy the array of data.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    :func:`array`
+        The recommended function for creating a StringArray.
+    Series.str
+        The string methods are available on Series backed by
+        a StringArray.
+
+    Notes
+    -----
+    StringArray returns a BooleanArray for comparison methods.
+
+    Examples
+    --------
+    >>> pd.array(["This is", "some text", None, "data."], dtype="string")
+    <StringArray>
+    ['This is', 'some text', <NA>, 'data.']
+    Length: 4, dtype: string
+
+    Unlike arrays instantiated with ``dtype="object"``, ``StringArray``
+    will convert the values to strings.
+
+    >>> pd.array(["1", 1], dtype="object")
+    <NumpyExtensionArray>
+    ['1', 1]
+    Length: 2, dtype: object
+    >>> pd.array(["1", 1], dtype="string")
+    <StringArray>
+    ['1', '1']
+    Length: 2, dtype: string
+
+    However, instantiating StringArrays directly with non-strings will raise an error.
+
+    For comparison methods, `StringArray` returns a :class:`pandas.BooleanArray`:
+
+    >>> pd.array(["a", None, "c"], dtype="string") == "a"
+    <BooleanArray>
+    [True, <NA>, False]
+    Length: 3, dtype: boolean
+    """
+
+    # undo the NumpyExtensionArray hack
+    _typ = "extension"
+    _storage = "python"
+    _na_value: libmissing.NAType | float = libmissing.NA
+
+    def __init__(self, values, copy: bool = False) -> None:
+        values = extract_array(values)
+
+        super().__init__(values, copy=copy)
+        if not isinstance(values, type(self)):
+            self._validate()
+        NDArrayBacked.__init__(
+            self,
+            self._ndarray,
+            StringDtype(storage=self._storage, na_value=self._na_value),
+        )
+
+    def _validate(self) -> None:
+        """Validate that we only store NA or strings."""
+        if len(self._ndarray) and not lib.is_string_array(self._ndarray, skipna=True):
+            raise ValueError("StringArray requires a sequence of strings or pandas.NA")
+        if self._ndarray.dtype != "object":
+            raise ValueError(
+                "StringArray requires a sequence of strings or pandas.NA. Got "
+                f"'{self._ndarray.dtype}' dtype instead."
+            )
+        # Check to see if need to convert Na values to pd.NA
+        if self._ndarray.ndim > 2:
+            # Ravel if ndims > 2 b/c no cythonized version available
+            lib.convert_nans_to_NA(self._ndarray.ravel("K"))
+        else:
+            lib.convert_nans_to_NA(self._ndarray)
+
+    def _validate_scalar(self, value):
+        # used by NDArrayBackedExtensionIndex.insert
+        if isna(value):
+            return self.dtype.na_value
+        elif not isinstance(value, str):
+            raise TypeError(
+                f"Invalid value '{value}' for dtype '{self.dtype}'. Value should be a "
+                f"string or missing value, got '{type(value).__name__}' instead."
+            )
+        return value
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        if dtype and not (isinstance(dtype, str) and dtype == "string"):
+            dtype = pandas_dtype(dtype)
+            assert isinstance(dtype, StringDtype) and dtype.storage == "python"
+        else:
+            if using_string_dtype():
+                dtype = StringDtype(storage="python", na_value=np.nan)
+            else:
+                dtype = StringDtype(storage="python")
+
+        from pandas.core.arrays.masked import BaseMaskedArray
+
+        na_value = dtype.na_value
+        if isinstance(scalars, BaseMaskedArray):
+            # avoid costly conversion to object dtype
+            na_values = scalars._mask
+            result = scalars._data
+            result = lib.ensure_string_array(result, copy=copy, convert_na_value=False)
+            result[na_values] = na_value
+
+        else:
+            if lib.is_pyarrow_array(scalars):
+                # pyarrow array; we cannot rely on the "to_numpy" check in
+                #  ensure_string_array because calling scalars.to_numpy would set
+                #  zero_copy_only to True which caused problems see GH#52076
+                scalars = np.array(scalars)
+            # convert non-na-likes to str, and nan-likes to StringDtype().na_value
+            result = lib.ensure_string_array(scalars, na_value=na_value, copy=copy)
+
+        # Manually creating new array avoids the validation step in the __init__, so is
+        # faster. Refactor need for validation?
+        new_string_array = cls.__new__(cls)
+        NDArrayBacked.__init__(new_string_array, result, dtype)
+
+        return new_string_array
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        return cls._from_sequence(strings, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _empty(cls, shape, dtype) -> StringArray:
+        values = np.empty(shape, dtype=object)
+        values[:] = libmissing.NA
+        return cls(values).astype(dtype, copy=False)
+
+    def __arrow_array__(self, type=None):
+        """
+        Convert myself into a pyarrow Array.
+        """
+        import pyarrow as pa
+
+        if type is None:
+            type = pa.string()
+
+        values = self._ndarray.copy()
+        values[self.isna()] = None
+        return pa.array(values, type=type, from_pandas=True)
+
+    def _values_for_factorize(self) -> tuple[np.ndarray, libmissing.NAType | float]:  # type: ignore[override]
+        arr = self._ndarray
+
+        return arr, self.dtype.na_value
+
+    def _maybe_convert_setitem_value(self, value):
+        """Maybe convert value to be pyarrow compatible."""
+        if lib.is_scalar(value):
+            if isna(value):
+                value = self.dtype.na_value
+            elif not isinstance(value, str):
+                raise TypeError(
+                    f"Invalid value '{value}' for dtype '{self.dtype}'. Value should "
+                    f"be a string or missing value, got '{type(value).__name__}' "
+                    "instead."
+                )
+        else:
+            value = extract_array(value, extract_numpy=True)
+            if not is_array_like(value):
+                value = np.asarray(value, dtype=object)
+            elif isinstance(value.dtype, type(self.dtype)):
+                return value
+            else:
+                # cast categories and friends to arrays to see if values are
+                # compatible, compatibility with arrow backed strings
+                value = np.asarray(value)
+            if len(value) and not lib.is_string_array(value, skipna=True):
+                raise TypeError(
+                    "Invalid value for dtype 'str'. Value should be a "
+                    "string or missing value (or array of those)."
+                )
+        return value
+
+    def __setitem__(self, key, value) -> None:
+        value = self._maybe_convert_setitem_value(value)
+
+        key = check_array_indexer(self, key)
+        scalar_key = lib.is_scalar(key)
+        scalar_value = lib.is_scalar(value)
+        if scalar_key and not scalar_value:
+            raise ValueError("setting an array element with a sequence.")
+
+        if not scalar_value:
+            if value.dtype == self.dtype:
+                value = value._ndarray
+            else:
+                value = np.asarray(value)
+                mask = isna(value)
+                if mask.any():
+                    value = value.copy()
+                    value[isna(value)] = self.dtype.na_value
+
+        super().__setitem__(key, value)
+
+    def _putmask(self, mask: npt.NDArray[np.bool_], value) -> None:
+        # the super() method NDArrayBackedExtensionArray._putmask uses
+        # np.putmask which doesn't properly handle None/pd.NA, so using the
+        # base class implementation that uses __setitem__
+        ExtensionArray._putmask(self, mask, value)
+
+    def _where(self, mask: npt.NDArray[np.bool_], value) -> Self:
+        # the super() method NDArrayBackedExtensionArray._where uses
+        # np.putmask which doesn't properly handle None/pd.NA, so using the
+        # base class implementation that uses __setitem__
+        return ExtensionArray._where(self, mask, value)
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        if isinstance(values, BaseStringArray) or (
+            isinstance(values, ExtensionArray) and is_string_dtype(values.dtype)
+        ):
+            values = values.astype(self.dtype, copy=False)
+        else:
+            if not lib.is_string_array(np.asarray(values), skipna=True):
+                values = np.array(
+                    [val for val in values if isinstance(val, str) or isna(val)],
+                    dtype=object,
+                )
+                if not len(values):
+                    return np.zeros(self.shape, dtype=bool)
+
+            values = self._from_sequence(values, dtype=self.dtype)
+
+        return isin(np.asarray(self), np.asarray(values))
+
+    def astype(self, dtype, copy: bool = True):
+        dtype = pandas_dtype(dtype)
+
+        if dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+
+        elif isinstance(dtype, IntegerDtype):
+            arr = self._ndarray.copy()
+            mask = self.isna()
+            arr[mask] = 0
+            values = arr.astype(dtype.numpy_dtype)
+            return IntegerArray(values, mask, copy=False)
+        elif isinstance(dtype, FloatingDtype):
+            arr_ea = self.copy()
+            mask = self.isna()
+            arr_ea[mask] = "0"
+            values = arr_ea.astype(dtype.numpy_dtype)
+            return FloatingArray(values, mask, copy=False)
+        elif isinstance(dtype, ExtensionDtype):
+            # Skip the NumpyExtensionArray.astype method
+            return ExtensionArray.astype(self, dtype, copy)
+        elif np.issubdtype(dtype, np.floating):
+            arr = self._ndarray.copy()
+            mask = self.isna()
+            arr[mask] = 0
+            values = arr.astype(dtype)
+            values[mask] = np.nan
+            return values
+
+        return super().astype(dtype, copy)
+
+    def _reduce(
+        self,
+        name: str,
+        *,
+        skipna: bool = True,
+        keepdims: bool = False,
+        axis: AxisInt | None = 0,
+        **kwargs,
+    ):
+        if self.dtype.na_value is np.nan and name in ["any", "all"]:
+            if name == "any":
+                return nanops.nanany(self._ndarray, skipna=skipna)
+            else:
+                return nanops.nanall(self._ndarray, skipna=skipna)
+
+        if name in ["min", "max", "argmin", "argmax", "sum"]:
+            result = getattr(self, name)(skipna=skipna, axis=axis, **kwargs)
+            if keepdims:
+                return self._from_sequence([result], dtype=self.dtype)
+            return result
+
+        raise TypeError(f"Cannot perform reduction '{name}' with string dtype")
+
+    def _accumulate(self, name: str, *, skipna: bool = True, **kwargs) -> StringArray:
+        """
+        Return an ExtensionArray performing an accumulation operation.
+
+        The underlying data type might change.
+
+        Parameters
+        ----------
+        name : str
+            Name of the function, supported values are:
+            - cummin
+            - cummax
+            - cumsum
+            - cumprod
+        skipna : bool, default True
+            If True, skip NA values.
+        **kwargs
+            Additional keyword arguments passed to the accumulation function.
+            Currently, there is no supported kwarg.
+
+        Returns
+        -------
+        array
+
+        Raises
+        ------
+        NotImplementedError : subclass does not define accumulations
+        """
+        if name == "cumprod":
+            msg = f"operation '{name}' not supported for dtype '{self.dtype}'"
+            raise TypeError(msg)
+
+        # We may need to strip out trailing NA values
+        tail: np.ndarray | None = None
+        na_mask: np.ndarray | None = None
+        ndarray = self._ndarray
+        np_func = {
+            "cumsum": np.cumsum,
+            "cummin": np.minimum.accumulate,
+            "cummax": np.maximum.accumulate,
+        }[name]
+
+        if self._hasna:
+            na_mask = cast("npt.NDArray[np.bool_]", isna(ndarray))
+            if np.all(na_mask):
+                return type(self)(ndarray)
+            if skipna:
+                if name == "cumsum":
+                    ndarray = np.where(na_mask, "", ndarray)
+                else:
+                    # We can retain the running min/max by forward/backward filling.
+                    ndarray = ndarray.copy()
+                    missing.pad_or_backfill_inplace(
+                        ndarray,
+                        method="pad",
+                        axis=0,
+                    )
+                    missing.pad_or_backfill_inplace(
+                        ndarray,
+                        method="backfill",
+                        axis=0,
+                    )
+            else:
+                # When not skipping NA values, the result should be null from
+                # the first NA value onward.
+                idx = np.argmax(na_mask)
+                tail = np.empty(len(ndarray) - idx, dtype="object")
+                tail[:] = self.dtype.na_value
+                ndarray = ndarray[:idx]
+
+        # mypy: Cannot call function of unknown type
+        np_result = np_func(ndarray)  # type: ignore[operator]
+
+        if tail is not None:
+            np_result = np.hstack((np_result, tail))
+        elif na_mask is not None:
+            # Argument 2 to "where" has incompatible type "NAType | float"
+            np_result = np.where(na_mask, self.dtype.na_value, np_result)  # type: ignore[arg-type]
+
+        result = type(self)(np_result)
+        return result
+
+    def _wrap_reduction_result(self, axis: AxisInt | None, result) -> Any:
+        if self.dtype.na_value is np.nan and result is libmissing.NA:
+            # the masked_reductions use pd.NA -> convert to np.nan
+            return np.nan
+        return super()._wrap_reduction_result(axis, result)
+
+    def min(self, axis=None, skipna: bool = True, **kwargs) -> Scalar:
+        nv.validate_min((), kwargs)
+        result = masked_reductions.min(
+            values=self.to_numpy(), mask=self.isna(), skipna=skipna
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def max(self, axis=None, skipna: bool = True, **kwargs) -> Scalar:
+        nv.validate_max((), kwargs)
+        result = masked_reductions.max(
+            values=self.to_numpy(), mask=self.isna(), skipna=skipna
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def sum(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        skipna: bool = True,
+        min_count: int = 0,
+        **kwargs,
+    ) -> Scalar:
+        nv.validate_sum((), kwargs)
+        result = masked_reductions.sum(
+            values=self._ndarray, mask=self.isna(), skipna=skipna
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        from pandas.core.algorithms import value_counts_internal as value_counts
+
+        result = value_counts(self._ndarray, sort=False, dropna=dropna)
+        result.index = result.index.astype(self.dtype)
+
+        if self.dtype.na_value is libmissing.NA:
+            result = result.astype("Int64")
+        return result
+
+    def memory_usage(self, deep: bool = False) -> int:
+        result = self._ndarray.nbytes
+        if deep:
+            return result + lib.memory_usage_of_objects(self._ndarray)
+        return result
+
+    @doc(ExtensionArray.searchsorted)
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if self._hasna:
+            raise ValueError(
+                "searchsorted requires array to be sorted, which is impossible "
+                "with NAs present."
+            )
+        return super().searchsorted(value=value, side=side, sorter=sorter)
+
+    def _cmp_method(self, other, op):
+        from pandas.arrays import BooleanArray
+
+        if isinstance(other, StringArray):
+            other = other._ndarray
+
+        mask = isna(self) | isna(other)
+        valid = ~mask
+
+        if not lib.is_scalar(other):
+            if len(other) != len(self):
+                # prevent improper broadcasting when other is 2D
+                raise ValueError(
+                    f"Lengths of operands do not match: {len(self)} != {len(other)}"
+                )
+
+            # for array-likes, first filter out NAs before converting to numpy
+            if not is_array_like(other):
+                other = np.asarray(other)
+            other = other[valid]
+
+        if op.__name__ in ops.ARITHMETIC_BINOPS:
+            result = np.empty_like(self._ndarray, dtype="object")
+            result[mask] = self.dtype.na_value
+            result[valid] = op(self._ndarray[valid], other)
+            return self._from_backing_data(result)
+        else:
+            # logical
+            result = np.zeros(len(self._ndarray), dtype="bool")
+            result[valid] = op(self._ndarray[valid], other)
+            res_arr = BooleanArray(result, mask)
+            if self.dtype.na_value is np.nan:
+                if op == operator.ne:
+                    return res_arr.to_numpy(np.bool_, na_value=True)
+                else:
+                    return res_arr.to_numpy(np.bool_, na_value=False)
+            return res_arr
+
+    _arith_method = _cmp_method
+
+
+class StringArrayNumpySemantics(StringArray):
+    _storage = "python"
+    _na_value = np.nan
+
+    def _validate(self) -> None:
+        """Validate that we only store NaN or strings."""
+        if len(self._ndarray) and not lib.is_string_array(self._ndarray, skipna=True):
+            raise ValueError(
+                "StringArrayNumpySemantics requires a sequence of strings or NaN"
+            )
+        if self._ndarray.dtype != "object":
+            raise ValueError(
+                "StringArrayNumpySemantics requires a sequence of strings or NaN. Got "
+                f"'{self._ndarray.dtype}' dtype instead."
+            )
+        # TODO validate or force NA/None to NaN
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        if dtype is None:
+            dtype = StringDtype(storage="python", na_value=np.nan)
+        return super()._from_sequence(scalars, dtype=dtype, copy=copy)
diff --git a/pandas/pandas/core/arrays/string_arrow.py b/pandas/pandas/core/arrays/string_arrow.py
new file mode 100644
index 0000000000000..d35083fd892a8
--- /dev/null
+++ b/pandas/pandas/core/arrays/string_arrow.py
@@ -0,0 +1,489 @@
+from __future__ import annotations
+
+import operator
+import re
+from typing import (
+    TYPE_CHECKING,
+    Union,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas.compat import (
+    pa_version_under10p1,
+    pa_version_under13p0,
+    pa_version_under16p0,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_scalar,
+    pandas_dtype,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.arrays._arrow_string_mixins import ArrowStringArrayMixin
+from pandas.core.arrays.arrow import ArrowExtensionArray
+from pandas.core.arrays.boolean import BooleanDtype
+from pandas.core.arrays.floating import Float64Dtype
+from pandas.core.arrays.integer import Int64Dtype
+from pandas.core.arrays.numeric import NumericDtype
+from pandas.core.arrays.string_ import (
+    BaseStringArray,
+    StringDtype,
+)
+from pandas.core.strings.object_array import ObjectStringArrayMixin
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+    import pyarrow.compute as pc
+
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        Dtype,
+        NpDtype,
+        Self,
+        npt,
+    )
+
+    from pandas.core.dtypes.dtypes import ExtensionDtype
+
+    from pandas import Series
+
+
+ArrowStringScalarOrNAT = Union[str, libmissing.NAType]
+
+
+def _chk_pyarrow_available() -> None:
+    if pa_version_under10p1:
+        msg = "pyarrow>=10.0.1 is required for PyArrow backed ArrowExtensionArray."
+        raise ImportError(msg)
+
+
+def _is_string_view(typ):
+    return not pa_version_under16p0 and pa.types.is_string_view(typ)
+
+
+# TODO: Inherit directly from BaseStringArrayMethods. Currently we inherit from
+# ObjectStringArrayMixin because we want to have the object-dtype based methods as
+# fallback for the ones that pyarrow doesn't yet support
+
+
+class ArrowStringArray(ObjectStringArrayMixin, ArrowExtensionArray, BaseStringArray):
+    """
+    Extension array for string data in a ``pyarrow.ChunkedArray``.
+
+    .. warning::
+
+       ArrowStringArray is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    Parameters
+    ----------
+    values : pyarrow.Array or pyarrow.ChunkedArray
+        The array of data.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    :func:`array`
+        The recommended function for creating a ArrowStringArray.
+    Series.str
+        The string methods are available on Series backed by
+        a ArrowStringArray.
+
+    Notes
+    -----
+    ArrowStringArray returns a BooleanArray for comparison methods.
+
+    Examples
+    --------
+    >>> pd.array(["This is", "some text", None, "data."], dtype="string[pyarrow]")
+    <ArrowStringArray>
+    ['This is', 'some text', <NA>, 'data.']
+    Length: 4, dtype: string
+    """
+
+    # error: Incompatible types in assignment (expression has type "StringDtype",
+    # base class "ArrowExtensionArray" defined the type as "ArrowDtype")
+    _dtype: StringDtype  # type: ignore[assignment]
+    _storage = "pyarrow"
+    _na_value: libmissing.NAType | float = libmissing.NA
+
+    def __init__(self, values) -> None:
+        _chk_pyarrow_available()
+        if isinstance(values, (pa.Array, pa.ChunkedArray)) and (
+            pa.types.is_string(values.type)
+            or _is_string_view(values.type)
+            or (
+                pa.types.is_dictionary(values.type)
+                and (
+                    pa.types.is_string(values.type.value_type)
+                    or pa.types.is_large_string(values.type.value_type)
+                    or _is_string_view(values.type.value_type)
+                )
+            )
+        ):
+            values = pc.cast(values, pa.large_string())
+
+        super().__init__(values)
+        self._dtype = StringDtype(storage=self._storage, na_value=self._na_value)
+
+        if not pa.types.is_large_string(self._pa_array.type):
+            raise ValueError(
+                "ArrowStringArray requires a PyArrow (chunked) array of "
+                "large_string type"
+            )
+
+    @classmethod
+    def _box_pa_scalar(cls, value, pa_type: pa.DataType | None = None) -> pa.Scalar:
+        pa_scalar = super()._box_pa_scalar(value, pa_type)
+        if pa.types.is_string(pa_scalar.type) and pa_type is None:
+            pa_scalar = pc.cast(pa_scalar, pa.large_string())
+        return pa_scalar
+
+    @classmethod
+    def _box_pa_array(
+        cls, value, pa_type: pa.DataType | None = None, copy: bool = False
+    ) -> pa.Array | pa.ChunkedArray:
+        pa_array = super()._box_pa_array(value, pa_type)
+        if pa.types.is_string(pa_array.type) and pa_type is None:
+            pa_array = pc.cast(pa_array, pa.large_string())
+        return pa_array
+
+    def __len__(self) -> int:
+        """
+        Length of this array.
+
+        Returns
+        -------
+        length : int
+        """
+        return len(self._pa_array)
+
+    @classmethod
+    def _from_sequence(
+        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
+    ) -> Self:
+        from pandas.core.arrays.masked import BaseMaskedArray
+
+        _chk_pyarrow_available()
+
+        if dtype and not (isinstance(dtype, str) and dtype == "string"):
+            dtype = pandas_dtype(dtype)
+            assert isinstance(dtype, StringDtype) and dtype.storage == "pyarrow"
+
+        if isinstance(scalars, BaseMaskedArray):
+            # avoid costly conversion to object dtype in ensure_string_array and
+            # numerical issues with Float32Dtype
+            na_values = scalars._mask
+            result = scalars._data
+            result = lib.ensure_string_array(result, copy=copy, convert_na_value=False)
+            return cls(pa.array(result, mask=na_values, type=pa.large_string()))
+        elif isinstance(scalars, (pa.Array, pa.ChunkedArray)):
+            return cls(pc.cast(scalars, pa.large_string()))
+
+        # convert non-na-likes to str
+        result = lib.ensure_string_array(scalars, copy=copy)
+        return cls(pa.array(result, type=pa.large_string(), from_pandas=True))
+
+    @classmethod
+    def _from_sequence_of_strings(
+        cls, strings, *, dtype: ExtensionDtype, copy: bool = False
+    ) -> Self:
+        return cls._from_sequence(strings, dtype=dtype, copy=copy)
+
+    @property
+    def dtype(self) -> StringDtype:  # type: ignore[override]
+        """
+        An instance of 'string[pyarrow]'.
+        """
+        return self._dtype
+
+    def insert(self, loc: int, item) -> ArrowStringArray:
+        if self.dtype.na_value is np.nan and item is np.nan:
+            item = libmissing.NA
+        if not isinstance(item, str) and item is not libmissing.NA:
+            raise TypeError(
+                f"Invalid value '{item}' for dtype 'str'. Value should be a "
+                f"string or missing value, got '{type(item).__name__}' instead."
+            )
+        return super().insert(loc, item)
+
+    def _convert_bool_result(self, values, na=lib.no_default, method_name=None):
+        if na is not lib.no_default and not isna(na) and not isinstance(na, bool):
+            # GH#59561
+            warnings.warn(
+                f"Allowing a non-bool 'na' in obj.str.{method_name} is deprecated "
+                "and will raise in a future version.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+            na = bool(na)
+
+        if self.dtype.na_value is np.nan:
+            if na is lib.no_default or isna(na):
+                # NaN propagates as False
+                values = values.fill_null(False)
+            else:
+                values = values.fill_null(na)
+            return values.to_numpy()
+        else:
+            if na is not lib.no_default and not isna(na):  # pyright: ignore [reportGeneralTypeIssues]
+                values = values.fill_null(na)
+        return BooleanDtype().__from_arrow__(values)
+
+    def _maybe_convert_setitem_value(self, value):
+        """Maybe convert value to be pyarrow compatible."""
+        if is_scalar(value):
+            if isna(value):
+                value = None
+            elif not isinstance(value, str):
+                raise TypeError(
+                    f"Invalid value '{value}' for dtype 'str'. Value should be a "
+                    f"string or missing value, got '{type(value).__name__}' instead."
+                )
+        else:
+            value = np.array(value, dtype=object, copy=True)
+            value[isna(value)] = None
+            for v in value:
+                if not (v is None or isinstance(v, str)):
+                    raise TypeError(
+                        "Invalid value for dtype 'str'. Value should be a "
+                        "string or missing value (or array of those)."
+                    )
+        return super()._maybe_convert_setitem_value(value)
+
+    def isin(self, values: ArrayLike) -> npt.NDArray[np.bool_]:
+        value_set = [
+            pa_scalar.as_py()
+            for pa_scalar in [pa.scalar(value, from_pandas=True) for value in values]
+            if pa_scalar.type in (pa.string(), pa.null(), pa.large_string())
+        ]
+
+        # short-circuit to return all False array.
+        if not len(value_set):
+            return np.zeros(len(self), dtype=bool)
+
+        result = pc.is_in(
+            self._pa_array, value_set=pa.array(value_set, type=self._pa_array.type)
+        )
+        # pyarrow 2.0.0 returned nulls, so we explicitly specify dtype to convert nulls
+        # to False
+        return np.array(result, dtype=np.bool_)
+
+    def astype(self, dtype, copy: bool = True):
+        dtype = pandas_dtype(dtype)
+
+        if dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+        elif isinstance(dtype, NumericDtype):
+            data = self._pa_array.cast(pa.from_numpy_dtype(dtype.numpy_dtype))
+            return dtype.__from_arrow__(data)
+        elif isinstance(dtype, np.dtype) and np.issubdtype(dtype, np.floating):
+            return self.to_numpy(dtype=dtype, na_value=np.nan)
+
+        return super().astype(dtype, copy=copy)
+
+    # ------------------------------------------------------------------------
+    # String methods interface
+
+    _str_isalnum = ArrowStringArrayMixin._str_isalnum
+    _str_isalpha = ArrowStringArrayMixin._str_isalpha
+    _str_isdecimal = ArrowStringArrayMixin._str_isdecimal
+    _str_isdigit = ArrowStringArrayMixin._str_isdigit
+    _str_islower = ArrowStringArrayMixin._str_islower
+    _str_isnumeric = ArrowStringArrayMixin._str_isnumeric
+    _str_isspace = ArrowStringArrayMixin._str_isspace
+    _str_istitle = ArrowStringArrayMixin._str_istitle
+    _str_isupper = ArrowStringArrayMixin._str_isupper
+
+    _str_map = BaseStringArray._str_map
+    _str_startswith = ArrowStringArrayMixin._str_startswith
+    _str_endswith = ArrowStringArrayMixin._str_endswith
+    _str_pad = ArrowStringArrayMixin._str_pad
+    _str_match = ArrowStringArrayMixin._str_match
+    _str_fullmatch = ArrowStringArrayMixin._str_fullmatch
+    _str_lower = ArrowStringArrayMixin._str_lower
+    _str_upper = ArrowStringArrayMixin._str_upper
+    _str_strip = ArrowStringArrayMixin._str_strip
+    _str_lstrip = ArrowStringArrayMixin._str_lstrip
+    _str_rstrip = ArrowStringArrayMixin._str_rstrip
+    _str_removesuffix = ArrowStringArrayMixin._str_removesuffix
+    _str_get = ArrowStringArrayMixin._str_get
+    _str_capitalize = ArrowStringArrayMixin._str_capitalize
+    _str_title = ArrowStringArrayMixin._str_title
+    _str_swapcase = ArrowStringArrayMixin._str_swapcase
+    _str_slice_replace = ArrowStringArrayMixin._str_slice_replace
+    _str_len = ArrowStringArrayMixin._str_len
+    _str_slice = ArrowStringArrayMixin._str_slice
+
+    def _str_contains(
+        self,
+        pat,
+        case: bool = True,
+        flags: int = 0,
+        na=lib.no_default,
+        regex: bool = True,
+    ):
+        if flags:
+            return super()._str_contains(pat, case, flags, na, regex)
+
+        return ArrowStringArrayMixin._str_contains(self, pat, case, flags, na, regex)
+
+    def _str_replace(
+        self,
+        pat: str | re.Pattern,
+        repl: str | Callable,
+        n: int = -1,
+        case: bool = True,
+        flags: int = 0,
+        regex: bool = True,
+    ):
+        if isinstance(pat, re.Pattern) or callable(repl) or not case or flags:
+            return super()._str_replace(pat, repl, n, case, flags, regex)
+
+        return ArrowStringArrayMixin._str_replace(
+            self, pat, repl, n, case, flags, regex
+        )
+
+    def _str_repeat(self, repeats: int | Sequence[int]):
+        if not isinstance(repeats, int):
+            return super()._str_repeat(repeats)
+        else:
+            return ArrowExtensionArray._str_repeat(self, repeats=repeats)
+
+    def _str_removeprefix(self, prefix: str):
+        if not pa_version_under13p0:
+            return ArrowStringArrayMixin._str_removeprefix(self, prefix)
+        return super()._str_removeprefix(prefix)
+
+    def _str_count(self, pat: str, flags: int = 0):
+        if flags:
+            return super()._str_count(pat, flags)
+        result = pc.count_substring_regex(self._pa_array, pat)
+        return self._convert_int_result(result)
+
+    def _str_find(self, sub: str, start: int = 0, end: int | None = None):
+        if (
+            pa_version_under13p0
+            and not (start != 0 and end is not None)
+            and not (start == 0 and end is None)
+        ):
+            # GH#59562
+            return super()._str_find(sub, start, end)
+        return ArrowStringArrayMixin._str_find(self, sub, start, end)
+
+    def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
+        if dtype is None:
+            dtype = np.int64
+        dummies_pa, labels = ArrowExtensionArray(self._pa_array)._str_get_dummies(
+            sep, dtype
+        )
+        if len(labels) == 0:
+            return np.empty(shape=(0, 0), dtype=dtype), labels
+        dummies = np.vstack(dummies_pa.to_numpy())
+        _dtype = pandas_dtype(dtype)
+        dummies_dtype: NpDtype
+        if isinstance(_dtype, np.dtype):
+            dummies_dtype = _dtype
+        else:
+            dummies_dtype = np.bool_
+        return dummies.astype(dummies_dtype, copy=False), labels
+
+    def _convert_int_result(self, result):
+        if self.dtype.na_value is np.nan:
+            if isinstance(result, pa.Array):
+                result = result.to_numpy(zero_copy_only=False)
+            else:
+                result = result.to_numpy()
+            if result.dtype == np.int32:
+                result = result.astype(np.int64)
+            return result
+
+        return Int64Dtype().__from_arrow__(result)
+
+    def _convert_rank_result(self, result):
+        if self.dtype.na_value is np.nan:
+            if isinstance(result, pa.Array):
+                result = result.to_numpy(zero_copy_only=False)
+            else:
+                result = result.to_numpy()
+            return result.astype("float64", copy=False)
+
+        return Float64Dtype().__from_arrow__(result)
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        if self.dtype.na_value is np.nan and name in ["any", "all"]:
+            if not skipna:
+                nas = pc.is_null(self._pa_array)
+                arr = pc.or_kleene(nas, pc.not_equal(self._pa_array, ""))
+            else:
+                arr = pc.not_equal(self._pa_array, "")
+            result = ArrowExtensionArray(arr)._reduce(
+                name, skipna=skipna, keepdims=keepdims, **kwargs
+            )
+            if keepdims:
+                # ArrowExtensionArray will return a length-1 bool[pyarrow] array
+                return result.astype(np.bool_)
+            return result
+
+        if name in ("min", "max", "sum", "argmin", "argmax"):
+            result = self._reduce_calc(name, skipna=skipna, keepdims=keepdims, **kwargs)
+        else:
+            raise TypeError(f"Cannot perform reduction '{name}' with string dtype")
+
+        if name in ("argmin", "argmax") and isinstance(result, pa.Array):
+            return self._convert_int_result(result)
+        elif isinstance(result, pa.Array):
+            return type(self)(result)
+        else:
+            return result
+
+    def value_counts(self, dropna: bool = True) -> Series:
+        result = super().value_counts(dropna=dropna)
+        if self.dtype.na_value is np.nan:
+            res_values = result._values.to_numpy()
+            return result._constructor(
+                res_values, index=result.index, name=result.name, copy=False
+            )
+        return result
+
+    def _cmp_method(self, other, op):
+        result = super()._cmp_method(other, op)
+        if self.dtype.na_value is np.nan:
+            if op == operator.ne:
+                return result.to_numpy(np.bool_, na_value=True)
+            else:
+                return result.to_numpy(np.bool_, na_value=False)
+        return result
+
+    def __pos__(self) -> Self:
+        raise TypeError(f"bad operand type for unary +: '{self.dtype}'")
+
+
+class ArrowStringArrayNumpySemantics(ArrowStringArray):
+    _na_value = np.nan
diff --git a/pandas/pandas/core/arrays/timedeltas.py b/pandas/pandas/core/arrays/timedeltas.py
new file mode 100644
index 0000000000000..c5b3129c506c8
--- /dev/null
+++ b/pandas/pandas/core/arrays/timedeltas.py
@@ -0,0 +1,1240 @@
+from __future__ import annotations
+
+from datetime import timedelta
+import operator
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    tslibs,
+)
+from pandas._libs.tslibs import (
+    NaT,
+    NaTType,
+    Tick,
+    Timedelta,
+    astype_overflowsafe,
+    get_supported_dtype,
+    iNaT,
+    is_supported_dtype,
+    periods_per_second,
+)
+from pandas._libs.tslibs.conversion import cast_from_unit_vectorized
+from pandas._libs.tslibs.fields import (
+    get_timedelta_days,
+    get_timedelta_field,
+)
+from pandas._libs.tslibs.timedeltas import (
+    array_to_timedelta64,
+    floordiv_object_array,
+    ints_to_pytimedelta,
+    parse_timedelta_unit,
+    truediv_object_array,
+)
+from pandas.compat.numpy import function as nv
+from pandas.util._validators import validate_endpoints
+
+from pandas.core.dtypes.common import (
+    TD64NS_DTYPE,
+    is_float_dtype,
+    is_integer_dtype,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import (
+    nanops,
+    roperator,
+)
+from pandas.core.array_algos import datetimelike_accumulations
+from pandas.core.arrays import datetimelike as dtl
+from pandas.core.arrays._ranges import generate_regular_range
+import pandas.core.common as com
+from pandas.core.ops.common import unpack_zerodim_and_defer
+
+if TYPE_CHECKING:
+    from collections.abc import Iterator
+
+    from pandas._typing import (
+        AxisInt,
+        DateTimeErrorChoices,
+        DtypeObj,
+        NpDtype,
+        Self,
+        npt,
+    )
+
+    from pandas import DataFrame
+
+import textwrap
+
+
+def _field_accessor(name: str, alias: str, docstring: str):
+    def f(self) -> np.ndarray:
+        values = self.asi8
+        if alias == "days":
+            result = get_timedelta_days(values, reso=self._creso)
+        else:
+            # error: Incompatible types in assignment (
+            # expression has type "ndarray[Any, dtype[signedinteger[_32Bit]]]",
+            # variable has type "ndarray[Any, dtype[signedinteger[_64Bit]]]
+            result = get_timedelta_field(values, alias, reso=self._creso)  # type: ignore[assignment]
+        if self._hasna:
+            result = self._maybe_mask_results(
+                result, fill_value=None, convert="float64"
+            )
+
+        return result
+
+    f.__name__ = name
+    f.__doc__ = f"\n{docstring}\n"
+    return property(f)
+
+
+class TimedeltaArray(dtl.TimelikeOps):
+    """
+    Pandas ExtensionArray for timedelta data.
+
+    .. warning::
+
+       TimedeltaArray is currently experimental, and its API may change
+       without warning. In particular, :attr:`TimedeltaArray.dtype` is
+       expected to change to be an instance of an ``ExtensionDtype``
+       subclass.
+
+    Parameters
+    ----------
+    data : array-like
+        The timedelta data.
+    dtype : numpy.dtype
+        Currently, only ``numpy.dtype("timedelta64[ns]")`` is accepted.
+    freq : Offset, optional
+        Frequency of the data.
+    copy : bool, default False
+        Whether to copy the underlying array of data.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    Timedelta : Represents a duration, the difference between two dates or times.
+    TimedeltaIndex : Immutable Index of timedelta64 data.
+    to_timedelta : Convert argument to timedelta.
+
+    Examples
+    --------
+    >>> pd.arrays.TimedeltaArray._from_sequence(pd.TimedeltaIndex(["1h", "2h"]))
+    <TimedeltaArray>
+    ['0 days 01:00:00', '0 days 02:00:00']
+    Length: 2, dtype: timedelta64[ns]
+    """
+
+    _typ = "timedeltaarray"
+    _internal_fill_value = np.timedelta64("NaT", "ns")
+    _recognized_scalars = (timedelta, np.timedelta64, Tick)
+    _is_recognized_dtype = lambda x: lib.is_np_dtype(x, "m")
+    _infer_matches = ("timedelta", "timedelta64")
+
+    @property
+    def _scalar_type(self) -> type[Timedelta]:
+        return Timedelta
+
+    __array_priority__ = 1000
+    # define my properties & methods for delegation
+    _other_ops: list[str] = []
+    _bool_ops: list[str] = []
+    _field_ops: list[str] = ["days", "seconds", "microseconds", "nanoseconds"]
+    _datetimelike_ops: list[str] = _field_ops + _bool_ops + ["unit", "freq"]
+    _datetimelike_methods: list[str] = [
+        "to_pytimedelta",
+        "total_seconds",
+        "round",
+        "floor",
+        "ceil",
+        "as_unit",
+    ]
+
+    # Note: ndim must be defined to ensure NaT.__richcmp__(TimedeltaArray)
+    #  operates pointwise.
+
+    def _box_func(self, x: np.timedelta64) -> Timedelta | NaTType:
+        y = x.view("i8")
+        if y == NaT._value:
+            return NaT
+        return Timedelta._from_value_and_reso(y, reso=self._creso)
+
+    @property
+    # error: Return type "dtype" of "dtype" incompatible with return type
+    # "ExtensionDtype" in supertype "ExtensionArray"
+    def dtype(self) -> np.dtype[np.timedelta64]:  # type: ignore[override]
+        """
+        The dtype for the TimedeltaArray.
+
+        .. warning::
+
+           A future version of pandas will change dtype to be an instance
+           of a :class:`pandas.api.extensions.ExtensionDtype` subclass,
+           not a ``numpy.dtype``.
+
+        Returns
+        -------
+        numpy.dtype
+        """
+        return self._ndarray.dtype
+
+    # ----------------------------------------------------------------
+    # Constructors
+
+    _freq = None
+
+    @classmethod
+    def _validate_dtype(cls, values, dtype):
+        # used in TimeLikeOps.__init__
+        dtype = _validate_td64_dtype(dtype)
+        _validate_td64_dtype(values.dtype)
+        if dtype != values.dtype:
+            raise ValueError("Values resolution does not match dtype.")
+        return dtype
+
+    # error: Signature of "_simple_new" incompatible with supertype "NDArrayBacked"
+    @classmethod
+    def _simple_new(  # type: ignore[override]
+        cls,
+        values: npt.NDArray[np.timedelta64],
+        freq: Tick | None = None,
+        dtype: np.dtype[np.timedelta64] = TD64NS_DTYPE,
+    ) -> Self:
+        # Require td64 dtype, not unit-less, matching values.dtype
+        assert lib.is_np_dtype(dtype, "m")
+        assert not tslibs.is_unitless(dtype)
+        assert isinstance(values, np.ndarray), type(values)
+        assert dtype == values.dtype
+        assert freq is None or isinstance(freq, Tick)
+
+        result = super()._simple_new(values=values, dtype=dtype)
+        result._freq = freq
+        return result
+
+    @classmethod
+    def _from_sequence(cls, data, *, dtype=None, copy: bool = False) -> Self:
+        if dtype:
+            dtype = _validate_td64_dtype(dtype)
+
+        data, freq = sequence_to_td64ns(data, copy=copy, unit=None)
+
+        if dtype is not None:
+            data = astype_overflowsafe(data, dtype=dtype, copy=False)
+
+        return cls._simple_new(data, dtype=data.dtype, freq=freq)
+
+    @classmethod
+    def _from_sequence_not_strict(
+        cls,
+        data,
+        *,
+        dtype=None,
+        copy: bool = False,
+        freq=lib.no_default,
+        unit=None,
+    ) -> Self:
+        """
+        _from_sequence_not_strict but without responsibility for finding the
+        result's `freq`.
+        """
+        if dtype:
+            dtype = _validate_td64_dtype(dtype)
+
+        assert unit not in ["Y", "y", "M"]  # caller is responsible for checking
+
+        data, inferred_freq = sequence_to_td64ns(data, copy=copy, unit=unit)
+
+        if dtype is not None:
+            data = astype_overflowsafe(data, dtype=dtype, copy=False)
+
+        result = cls._simple_new(data, dtype=data.dtype, freq=inferred_freq)
+
+        result._maybe_pin_freq(freq, {})
+        return result
+
+    @classmethod
+    def _generate_range(
+        cls, start, end, periods, freq, closed=None, *, unit: str | None = None
+    ) -> Self:
+        periods = dtl.validate_periods(periods)
+        if freq is None and any(x is None for x in [periods, start, end]):
+            raise ValueError("Must provide freq argument if no data is supplied")
+
+        if com.count_not_none(start, end, periods, freq) != 3:
+            raise ValueError(
+                "Of the four parameters: start, end, periods, "
+                "and freq, exactly three must be specified"
+            )
+
+        if start is not None:
+            start = Timedelta(start).as_unit("ns")
+
+        if end is not None:
+            end = Timedelta(end).as_unit("ns")
+
+        if unit is not None:
+            if unit not in ["s", "ms", "us", "ns"]:
+                raise ValueError("'unit' must be one of 's', 'ms', 'us', 'ns'")
+        else:
+            unit = "ns"
+
+        if start is not None and unit is not None:
+            start = start.as_unit(unit, round_ok=False)
+        if end is not None and unit is not None:
+            end = end.as_unit(unit, round_ok=False)
+
+        left_closed, right_closed = validate_endpoints(closed)
+
+        if freq is not None:
+            index = generate_regular_range(start, end, periods, freq, unit=unit)
+        else:
+            index = np.linspace(start._value, end._value, periods).astype("i8")
+
+        if not left_closed:
+            index = index[1:]
+        if not right_closed:
+            index = index[:-1]
+
+        td64values = index.view(f"m8[{unit}]")
+        return cls._simple_new(td64values, dtype=td64values.dtype, freq=freq)
+
+    # ----------------------------------------------------------------
+    # DatetimeLike Interface
+
+    def _unbox_scalar(self, value) -> np.timedelta64:
+        if not isinstance(value, self._scalar_type) and value is not NaT:
+            raise ValueError("'value' should be a Timedelta.")
+        self._check_compatible_with(value)
+        if value is NaT:
+            return np.timedelta64(value._value, self.unit)
+        else:
+            return value.as_unit(self.unit, round_ok=False).asm8
+
+    def _scalar_from_string(self, value) -> Timedelta | NaTType:
+        return Timedelta(value)
+
+    def _check_compatible_with(self, other) -> None:
+        # we don't have anything to validate.
+        pass
+
+    # ----------------------------------------------------------------
+    # Array-Like / EA-Interface Methods
+
+    def astype(self, dtype, copy: bool = True):
+        # We handle
+        #   --> timedelta64[ns]
+        #   --> timedelta64
+        # DatetimeLikeArrayMixin super call handles other cases
+        dtype = pandas_dtype(dtype)
+
+        if lib.is_np_dtype(dtype, "m"):
+            if dtype == self.dtype:
+                if copy:
+                    return self.copy()
+                return self
+
+            if is_supported_dtype(dtype):
+                # unit conversion e.g. timedelta64[s]
+                res_values = astype_overflowsafe(self._ndarray, dtype, copy=False)
+                return type(self)._simple_new(
+                    res_values, dtype=res_values.dtype, freq=self.freq
+                )
+            else:
+                raise ValueError(
+                    f"Cannot convert from {self.dtype} to {dtype}. "
+                    "Supported resolutions are 's', 'ms', 'us', 'ns'"
+                )
+
+        return dtl.DatetimeLikeArrayMixin.astype(self, dtype, copy=copy)
+
+    def __iter__(self) -> Iterator:
+        if self.ndim > 1:
+            for i in range(len(self)):
+                yield self[i]
+        else:
+            # convert in chunks of 10k for efficiency
+            data = self._ndarray
+            length = len(self)
+            chunksize = 10000
+            chunks = (length // chunksize) + 1
+            for i in range(chunks):
+                start_i = i * chunksize
+                end_i = min((i + 1) * chunksize, length)
+                converted = ints_to_pytimedelta(data[start_i:end_i], box=True)
+                yield from converted
+
+    # ----------------------------------------------------------------
+    # Reductions
+
+    def sum(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        keepdims: bool = False,
+        initial=None,
+        skipna: bool = True,
+        min_count: int = 0,
+    ):
+        nv.validate_sum(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims, "initial": initial}
+        )
+
+        result = nanops.nansum(
+            self._ndarray, axis=axis, skipna=skipna, min_count=min_count
+        )
+        return self._wrap_reduction_result(axis, result)
+
+    def std(
+        self,
+        *,
+        axis: AxisInt | None = None,
+        dtype: NpDtype | None = None,
+        out=None,
+        ddof: int = 1,
+        keepdims: bool = False,
+        skipna: bool = True,
+    ):
+        nv.validate_stat_ddof_func(
+            (), {"dtype": dtype, "out": out, "keepdims": keepdims}, fname="std"
+        )
+
+        result = nanops.nanstd(self._ndarray, axis=axis, skipna=skipna, ddof=ddof)
+        if axis is None or self.ndim == 1:
+            return self._box_func(result)
+        return self._from_backing_data(result)
+
+    # ----------------------------------------------------------------
+    # Accumulations
+
+    def _accumulate(self, name: str, *, skipna: bool = True, **kwargs):
+        if name == "cumsum":
+            op = getattr(datetimelike_accumulations, name)
+            result = op(self._ndarray.copy(), skipna=skipna, **kwargs)
+
+            return type(self)._simple_new(result, freq=None, dtype=self.dtype)
+        elif name == "cumprod":
+            raise TypeError("cumprod not supported for Timedelta.")
+
+        else:
+            return super()._accumulate(name, skipna=skipna, **kwargs)
+
+    # ----------------------------------------------------------------
+    # Rendering Methods
+
+    def _formatter(self, boxed: bool = False):
+        from pandas.io.formats.format import get_format_timedelta64
+
+        return get_format_timedelta64(self, box=True)
+
+    def _format_native_types(
+        self, *, na_rep: str | float = "NaT", date_format=None, **kwargs
+    ) -> npt.NDArray[np.object_]:
+        from pandas.io.formats.format import get_format_timedelta64
+
+        # Relies on TimeDelta._repr_base
+        formatter = get_format_timedelta64(self, na_rep)
+        # equiv: np.array([formatter(x) for x in self._ndarray])
+        #  but independent of dimension
+        return np.frompyfunc(formatter, 1, 1)(self._ndarray)
+
+    # ----------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _add_offset(self, other):
+        assert not isinstance(other, Tick)
+        raise TypeError(
+            f"cannot add the type {type(other).__name__} to a {type(self).__name__}"
+        )
+
+    @unpack_zerodim_and_defer("__mul__")
+    def __mul__(self, other) -> Self:
+        if is_scalar(other):
+            # numpy will accept float and int, raise TypeError for others
+            result = self._ndarray * other
+            if result.dtype.kind != "m":
+                # numpy >= 2.1 may not raise a TypeError
+                # and seems to dispatch to others.__rmul__?
+                raise TypeError(f"Cannot multiply with {type(other).__name__}")
+            freq = None
+            if self.freq is not None and not isna(other):
+                freq = self.freq * other
+                if freq.n == 0:
+                    # GH#51575 Better to have no freq than an incorrect one
+                    freq = None
+            return type(self)._simple_new(result, dtype=result.dtype, freq=freq)
+
+        if not hasattr(other, "dtype"):
+            # list, tuple
+            other = np.array(other)
+        if len(other) != len(self) and not lib.is_np_dtype(other.dtype, "m"):
+            # Exclude timedelta64 here so we correctly raise TypeError
+            #  for that instead of ValueError
+            raise ValueError("Cannot multiply with unequal lengths")
+
+        if is_object_dtype(other.dtype):
+            # this multiplication will succeed only if all elements of other
+            #  are int or float scalars, so we will end up with
+            #  timedelta64[ns]-dtyped result
+            arr = self._ndarray
+            result = [arr[n] * other[n] for n in range(len(self))]
+            result = np.array(result)
+            return type(self)._simple_new(result, dtype=result.dtype)
+
+        # numpy will accept float or int dtype, raise TypeError for others
+        result = self._ndarray * other
+        if result.dtype.kind != "m":
+            # numpy >= 2.1 may not raise a TypeError
+            # and seems to dispatch to others.__rmul__?
+            raise TypeError(f"Cannot multiply with {type(other).__name__}")
+        return type(self)._simple_new(result, dtype=result.dtype)
+
+    __rmul__ = __mul__
+
+    def _scalar_divlike_op(self, other, op):
+        """
+        Shared logic for __truediv__, __rtruediv__, __floordiv__, __rfloordiv__
+        with scalar 'other'.
+        """
+        if isinstance(other, self._recognized_scalars):
+            other = Timedelta(other)
+            # mypy assumes that __new__ returns an instance of the class
+            # github.com/python/mypy/issues/1020
+            if cast("Timedelta | NaTType", other) is NaT:
+                # specifically timedelta64-NaT
+                res = np.empty(self.shape, dtype=np.float64)
+                res.fill(np.nan)
+                return res
+
+            # otherwise, dispatch to Timedelta implementation
+            return op(self._ndarray, other)
+
+        else:
+            # caller is responsible for checking lib.is_scalar(other)
+            # assume other is numeric, otherwise numpy will raise
+
+            if op in [roperator.rtruediv, roperator.rfloordiv]:
+                raise TypeError(
+                    f"Cannot divide {type(other).__name__} by {type(self).__name__}"
+                )
+
+            result = op(self._ndarray, other)
+            freq = None
+
+            if self.freq is not None:
+                # Note: freq gets division, not floor-division, even if op
+                #  is floordiv.
+                freq = self.freq / other
+                if freq.nanos == 0 and self.freq.nanos != 0:
+                    # e.g. if self.freq is Nano(1) then dividing by 2
+                    #  rounds down to zero
+                    freq = None
+
+            return type(self)._simple_new(result, dtype=result.dtype, freq=freq)
+
+    def _cast_divlike_op(self, other):
+        if not hasattr(other, "dtype"):
+            # e.g. list, tuple
+            other = np.array(other)
+
+        if len(other) != len(self):
+            raise ValueError("Cannot divide vectors with unequal lengths")
+        return other
+
+    def _vector_divlike_op(self, other, op) -> np.ndarray | Self:
+        """
+        Shared logic for __truediv__, __floordiv__, and their reversed versions
+        with timedelta64-dtype ndarray other.
+        """
+        # Let numpy handle it
+        result = op(self._ndarray, np.asarray(other))
+
+        if (is_integer_dtype(other.dtype) or is_float_dtype(other.dtype)) and op in [
+            operator.truediv,
+            operator.floordiv,
+        ]:
+            return type(self)._simple_new(result, dtype=result.dtype)
+
+        if op in [operator.floordiv, roperator.rfloordiv]:
+            mask = self.isna() | isna(other)
+            if mask.any():
+                result = result.astype(np.float64)
+                np.putmask(result, mask, np.nan)
+
+        return result
+
+    @unpack_zerodim_and_defer("__truediv__")
+    def __truediv__(self, other):
+        # timedelta / X is well-defined for timedelta-like or numeric X
+        op = operator.truediv
+        if is_scalar(other):
+            return self._scalar_divlike_op(other, op)
+
+        other = self._cast_divlike_op(other)
+        if (
+            lib.is_np_dtype(other.dtype, "m")
+            or is_integer_dtype(other.dtype)
+            or is_float_dtype(other.dtype)
+        ):
+            return self._vector_divlike_op(other, op)
+
+        if is_object_dtype(other.dtype):
+            other = np.asarray(other)
+            if self.ndim > 1:
+                res_cols = [left / right for left, right in zip(self, other)]
+                res_cols2 = [x.reshape(1, -1) for x in res_cols]
+                result = np.concatenate(res_cols2, axis=0)
+            else:
+                result = truediv_object_array(self._ndarray, other)
+
+            return result
+
+        else:
+            return NotImplemented
+
+    @unpack_zerodim_and_defer("__rtruediv__")
+    def __rtruediv__(self, other):
+        # X / timedelta is defined only for timedelta-like X
+        op = roperator.rtruediv
+        if is_scalar(other):
+            return self._scalar_divlike_op(other, op)
+
+        other = self._cast_divlike_op(other)
+        if lib.is_np_dtype(other.dtype, "m"):
+            return self._vector_divlike_op(other, op)
+
+        elif is_object_dtype(other.dtype):
+            # Note: unlike in __truediv__, we do not _need_ to do type
+            #  inference on the result.  It does not raise, a numeric array
+            #  is returned.  GH#23829
+            result_list = [other[n] / self[n] for n in range(len(self))]
+            return np.array(result_list)
+
+        else:
+            return NotImplemented
+
+    @unpack_zerodim_and_defer("__floordiv__")
+    def __floordiv__(self, other):
+        op = operator.floordiv
+        if is_scalar(other):
+            return self._scalar_divlike_op(other, op)
+
+        other = self._cast_divlike_op(other)
+        if (
+            lib.is_np_dtype(other.dtype, "m")
+            or is_integer_dtype(other.dtype)
+            or is_float_dtype(other.dtype)
+        ):
+            return self._vector_divlike_op(other, op)
+
+        elif is_object_dtype(other.dtype):
+            other = np.asarray(other)
+            if self.ndim > 1:
+                res_cols = [left // right for left, right in zip(self, other)]
+                res_cols2 = [x.reshape(1, -1) for x in res_cols]
+                result = np.concatenate(res_cols2, axis=0)
+            else:
+                result = floordiv_object_array(self._ndarray, other)
+
+            assert result.dtype == object
+            return result
+
+        else:
+            return NotImplemented
+
+    @unpack_zerodim_and_defer("__rfloordiv__")
+    def __rfloordiv__(self, other):
+        op = roperator.rfloordiv
+        if is_scalar(other):
+            return self._scalar_divlike_op(other, op)
+
+        other = self._cast_divlike_op(other)
+        if lib.is_np_dtype(other.dtype, "m"):
+            return self._vector_divlike_op(other, op)
+
+        elif is_object_dtype(other.dtype):
+            result_list = [other[n] // self[n] for n in range(len(self))]
+            result = np.array(result_list)
+            return result
+
+        else:
+            return NotImplemented
+
+    @unpack_zerodim_and_defer("__mod__")
+    def __mod__(self, other):
+        # Note: This is a naive implementation, can likely be optimized
+        if isinstance(other, self._recognized_scalars):
+            other = Timedelta(other)
+        return self - (self // other) * other
+
+    @unpack_zerodim_and_defer("__rmod__")
+    def __rmod__(self, other):
+        # Note: This is a naive implementation, can likely be optimized
+        if isinstance(other, self._recognized_scalars):
+            other = Timedelta(other)
+        return other - (other // self) * self
+
+    @unpack_zerodim_and_defer("__divmod__")
+    def __divmod__(self, other):
+        # Note: This is a naive implementation, can likely be optimized
+        if isinstance(other, self._recognized_scalars):
+            other = Timedelta(other)
+
+        res1 = self // other
+        res2 = self - res1 * other
+        return res1, res2
+
+    @unpack_zerodim_and_defer("__rdivmod__")
+    def __rdivmod__(self, other):
+        # Note: This is a naive implementation, can likely be optimized
+        if isinstance(other, self._recognized_scalars):
+            other = Timedelta(other)
+
+        res1 = other // self
+        res2 = other - res1 * self
+        return res1, res2
+
+    def __neg__(self) -> TimedeltaArray:
+        freq = None
+        if self.freq is not None:
+            freq = -self.freq
+        return type(self)._simple_new(-self._ndarray, dtype=self.dtype, freq=freq)
+
+    def __pos__(self) -> TimedeltaArray:
+        return type(self)._simple_new(
+            self._ndarray.copy(), dtype=self.dtype, freq=self.freq
+        )
+
+    def __abs__(self) -> TimedeltaArray:
+        # Note: freq is not preserved
+        return type(self)._simple_new(np.abs(self._ndarray), dtype=self.dtype)
+
+    # ----------------------------------------------------------------
+    # Conversion Methods - Vectorized analogues of Timedelta methods
+
+    def total_seconds(self) -> npt.NDArray[np.float64]:
+        """
+        Return total duration of each element expressed in seconds.
+
+        This method is available directly on TimedeltaArray, TimedeltaIndex
+        and on Series containing timedelta values under the ``.dt`` namespace.
+
+        Returns
+        -------
+        ndarray, Index or Series
+            When the calling object is a TimedeltaArray, the return type
+            is ndarray.  When the calling object is a TimedeltaIndex,
+            the return type is an Index with a float64 dtype. When the calling object
+            is a Series, the return type is Series of type `float64` whose
+            index is the same as the original.
+
+        See Also
+        --------
+        datetime.timedelta.total_seconds : Standard library version
+            of this method.
+        TimedeltaIndex.components : Return a DataFrame with components of
+            each Timedelta.
+
+        Examples
+        --------
+        **Series**
+
+        >>> s = pd.Series(pd.to_timedelta(np.arange(5), unit="D"))
+        >>> s
+        0   0 days
+        1   1 days
+        2   2 days
+        3   3 days
+        4   4 days
+        dtype: timedelta64[ns]
+
+        >>> s.dt.total_seconds()
+        0         0.0
+        1     86400.0
+        2    172800.0
+        3    259200.0
+        4    345600.0
+        dtype: float64
+
+        **TimedeltaIndex**
+
+        >>> idx = pd.to_timedelta(np.arange(5), unit="D")
+        >>> idx
+        TimedeltaIndex(['0 days', '1 days', '2 days', '3 days', '4 days'],
+                       dtype='timedelta64[ns]', freq=None)
+
+        >>> idx.total_seconds()
+        Index([0.0, 86400.0, 172800.0, 259200.0, 345600.0], dtype='float64')
+        """
+        pps = periods_per_second(self._creso)
+        return self._maybe_mask_results(self.asi8 / pps, fill_value=None)
+
+    def to_pytimedelta(self) -> npt.NDArray[np.object_]:
+        """
+        Return an ndarray of datetime.timedelta objects.
+
+        Returns
+        -------
+        numpy.ndarray
+            A NumPy ``timedelta64`` object representing the same duration as the
+            original pandas ``Timedelta`` object. The precision of the resulting
+            object is in nanoseconds, which is the default
+            time resolution used by pandas for ``Timedelta`` objects, ensuring
+            high precision for time-based calculations.
+
+        See Also
+        --------
+        to_timedelta : Convert argument to timedelta format.
+        Timedelta : Represents a duration between two dates or times.
+        DatetimeIndex: Index of datetime64 data.
+        Timedelta.components : Return a components namedtuple-like
+                               of a single timedelta.
+
+        Examples
+        --------
+        >>> tdelta_idx = pd.to_timedelta([1, 2, 3], unit="D")
+        >>> tdelta_idx
+        TimedeltaIndex(['1 days', '2 days', '3 days'],
+                        dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.to_pytimedelta()
+        array([datetime.timedelta(days=1), datetime.timedelta(days=2),
+               datetime.timedelta(days=3)], dtype=object)
+
+        >>> tidx = pd.TimedeltaIndex(data=["1 days 02:30:45", "3 days 04:15:10"])
+        >>> tidx
+        TimedeltaIndex(['1 days 02:30:45', '3 days 04:15:10'],
+               dtype='timedelta64[ns]', freq=None)
+        >>> tidx.to_pytimedelta()
+        array([datetime.timedelta(days=1, seconds=9045),
+                datetime.timedelta(days=3, seconds=15310)], dtype=object)
+        """
+        return ints_to_pytimedelta(self._ndarray)
+
+    days_docstring = textwrap.dedent(
+        """Number of days for each element.
+
+    See Also
+    --------
+    Series.dt.seconds : Return number of seconds for each element.
+    Series.dt.microseconds : Return number of microseconds for each element.
+    Series.dt.nanoseconds : Return number of nanoseconds for each element.
+
+    Examples
+    --------
+    For Series:
+
+    >>> ser = pd.Series(pd.to_timedelta([1, 2, 3], unit='D'))
+    >>> ser
+    0   1 days
+    1   2 days
+    2   3 days
+    dtype: timedelta64[ns]
+    >>> ser.dt.days
+    0    1
+    1    2
+    2    3
+    dtype: int64
+
+    For TimedeltaIndex:
+
+    >>> tdelta_idx = pd.to_timedelta(["0 days", "10 days", "20 days"])
+    >>> tdelta_idx
+    TimedeltaIndex(['0 days', '10 days', '20 days'],
+                    dtype='timedelta64[ns]', freq=None)
+    >>> tdelta_idx.days
+    Index([0, 10, 20], dtype='int64')"""
+    )
+    days = _field_accessor("days", "days", days_docstring)
+
+    seconds_docstring = textwrap.dedent(
+        """Number of seconds (>= 0 and less than 1 day) for each element.
+
+    See Also
+    --------
+    Series.dt.seconds : Return number of seconds for each element.
+    Series.dt.nanoseconds : Return number of nanoseconds for each element.
+
+    Examples
+    --------
+    For Series:
+
+    >>> ser = pd.Series(pd.to_timedelta([1, 2, 3], unit='s'))
+    >>> ser
+    0   0 days 00:00:01
+    1   0 days 00:00:02
+    2   0 days 00:00:03
+    dtype: timedelta64[ns]
+    >>> ser.dt.seconds
+    0    1
+    1    2
+    2    3
+    dtype: int32
+
+    For TimedeltaIndex:
+
+    >>> tdelta_idx = pd.to_timedelta([1, 2, 3], unit='s')
+    >>> tdelta_idx
+    TimedeltaIndex(['0 days 00:00:01', '0 days 00:00:02', '0 days 00:00:03'],
+                   dtype='timedelta64[ns]', freq=None)
+    >>> tdelta_idx.seconds
+    Index([1, 2, 3], dtype='int32')"""
+    )
+    seconds = _field_accessor(
+        "seconds",
+        "seconds",
+        seconds_docstring,
+    )
+
+    microseconds_docstring = textwrap.dedent(
+        """Number of microseconds (>= 0 and less than 1 second) for each element.
+
+    See Also
+    --------
+    pd.Timedelta.microseconds : Number of microseconds (>= 0 and less than 1 second).
+    pd.Timedelta.to_pytimedelta.microseconds : Number of microseconds (>= 0 and less
+        than 1 second) of a datetime.timedelta.
+
+    Examples
+    --------
+    For Series:
+
+    >>> ser = pd.Series(pd.to_timedelta([1, 2, 3], unit='us'))
+    >>> ser
+    0   0 days 00:00:00.000001
+    1   0 days 00:00:00.000002
+    2   0 days 00:00:00.000003
+    dtype: timedelta64[ns]
+    >>> ser.dt.microseconds
+    0    1
+    1    2
+    2    3
+    dtype: int32
+
+    For TimedeltaIndex:
+
+    >>> tdelta_idx = pd.to_timedelta([1, 2, 3], unit='us')
+    >>> tdelta_idx
+    TimedeltaIndex(['0 days 00:00:00.000001', '0 days 00:00:00.000002',
+                    '0 days 00:00:00.000003'],
+                   dtype='timedelta64[ns]', freq=None)
+    >>> tdelta_idx.microseconds
+    Index([1, 2, 3], dtype='int32')"""
+    )
+    microseconds = _field_accessor(
+        "microseconds",
+        "microseconds",
+        microseconds_docstring,
+    )
+
+    nanoseconds_docstring = textwrap.dedent(
+        """Number of nanoseconds (>= 0 and less than 1 microsecond) for each element.
+
+    See Also
+    --------
+    Series.dt.seconds : Return number of seconds for each element.
+    Series.dt.microseconds : Return number of nanoseconds for each element.
+
+    Examples
+    --------
+    For Series:
+
+    >>> ser = pd.Series(pd.to_timedelta([1, 2, 3], unit='ns'))
+    >>> ser
+    0   0 days 00:00:00.000000001
+    1   0 days 00:00:00.000000002
+    2   0 days 00:00:00.000000003
+    dtype: timedelta64[ns]
+    >>> ser.dt.nanoseconds
+    0    1
+    1    2
+    2    3
+    dtype: int32
+
+    For TimedeltaIndex:
+
+    >>> tdelta_idx = pd.to_timedelta([1, 2, 3], unit='ns')
+    >>> tdelta_idx
+    TimedeltaIndex(['0 days 00:00:00.000000001', '0 days 00:00:00.000000002',
+                    '0 days 00:00:00.000000003'],
+                   dtype='timedelta64[ns]', freq=None)
+    >>> tdelta_idx.nanoseconds
+    Index([1, 2, 3], dtype='int32')"""
+    )
+    nanoseconds = _field_accessor(
+        "nanoseconds",
+        "nanoseconds",
+        nanoseconds_docstring,
+    )
+
+    @property
+    def components(self) -> DataFrame:
+        """
+        Return a DataFrame of the individual resolution components of the Timedeltas.
+
+        The components (days, hours, minutes seconds, milliseconds, microseconds,
+        nanoseconds) are returned as columns in a DataFrame.
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        TimedeltaIndex.total_seconds : Return total duration expressed in seconds.
+        Timedelta.components : Return a components namedtuple-like of a single
+            timedelta.
+
+        Examples
+        --------
+        >>> tdelta_idx = pd.to_timedelta(["1 day 3 min 2 us 42 ns"])
+        >>> tdelta_idx
+        TimedeltaIndex(['1 days 00:03:00.000002042'],
+                       dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.components
+           days  hours  minutes  seconds  milliseconds  microseconds  nanoseconds
+        0     1      0        3        0             0             2           42
+        """
+        from pandas import DataFrame
+
+        columns = [
+            "days",
+            "hours",
+            "minutes",
+            "seconds",
+            "milliseconds",
+            "microseconds",
+            "nanoseconds",
+        ]
+        hasnans = self._hasna
+        if hasnans:
+
+            def f(x):
+                if isna(x):
+                    return [np.nan] * len(columns)
+                return x.components
+
+        else:
+
+            def f(x):
+                return x.components
+
+        result = DataFrame([f(x) for x in self], columns=columns)
+        if not hasnans:
+            result = result.astype("int64")
+        return result
+
+
+# ---------------------------------------------------------------------
+# Constructor Helpers
+
+
+def sequence_to_td64ns(
+    data,
+    copy: bool = False,
+    unit=None,
+    errors: DateTimeErrorChoices = "raise",
+) -> tuple[np.ndarray, Tick | None]:
+    """
+    Parameters
+    ----------
+    data : list-like
+    copy : bool, default False
+    unit : str, optional
+        The timedelta unit to treat integers as multiples of. For numeric
+        data this defaults to ``'ns'``.
+        Must be un-specified if the data contains a str and ``errors=="raise"``.
+    errors : {"raise", "coerce", "ignore"}, default "raise"
+        How to handle elements that cannot be converted to timedelta64[ns].
+        See ``pandas.to_timedelta`` for details.
+
+    Returns
+    -------
+    converted : numpy.ndarray
+        The sequence converted to a numpy array with dtype ``timedelta64[ns]``.
+    inferred_freq : Tick or None
+        The inferred frequency of the sequence.
+
+    Raises
+    ------
+    ValueError : Data cannot be converted to timedelta64[ns].
+
+    Notes
+    -----
+    Unlike `pandas.to_timedelta`, if setting ``errors=ignore`` will not cause
+    errors to be ignored; they are caught and subsequently ignored at a
+    higher level.
+    """
+    assert unit not in ["Y", "y", "M"]  # caller is responsible for checking
+
+    inferred_freq = None
+    if unit is not None:
+        unit = parse_timedelta_unit(unit)
+
+    data, copy = dtl.ensure_arraylike_for_datetimelike(
+        data, copy, cls_name="TimedeltaArray"
+    )
+
+    if isinstance(data, TimedeltaArray):
+        inferred_freq = data.freq
+
+    # Convert whatever we have into timedelta64[ns] dtype
+    if data.dtype == object or is_string_dtype(data.dtype):
+        # no need to make a copy, need to convert if string-dtyped
+        data = _objects_to_td64ns(data, unit=unit, errors=errors)
+        copy = False
+
+    elif is_integer_dtype(data.dtype):
+        # treat as multiples of the given unit
+        data, copy_made = _ints_to_td64ns(data, unit=unit)
+        copy = copy and not copy_made
+
+    elif is_float_dtype(data.dtype):
+        # cast the unit, multiply base/frac separately
+        # to avoid precision issues from float -> int
+        if isinstance(data.dtype, ExtensionDtype):
+            mask = data._mask
+            data = data._data
+        else:
+            mask = np.isnan(data)
+
+        data = cast_from_unit_vectorized(data, unit or "ns")
+        data[mask] = iNaT
+        data = data.view("m8[ns]")
+        copy = False
+
+    elif lib.is_np_dtype(data.dtype, "m"):
+        if not is_supported_dtype(data.dtype):
+            # cast to closest supported unit, i.e. s or ns
+            new_dtype = get_supported_dtype(data.dtype)
+            data = astype_overflowsafe(data, dtype=new_dtype, copy=False)
+            copy = False
+
+    else:
+        # This includes datetime64-dtype, see GH#23539, GH#29794
+        raise TypeError(f"dtype {data.dtype} cannot be converted to timedelta64[ns]")
+
+    if not copy:
+        data = np.asarray(data)
+    else:
+        data = np.array(data, copy=copy)
+
+    assert data.dtype.kind == "m"
+    assert data.dtype != "m8"  # i.e. not unit-less
+
+    return data, inferred_freq
+
+
+def _ints_to_td64ns(data, unit: str = "ns") -> tuple[np.ndarray, bool]:
+    """
+    Convert an ndarray with integer-dtype to timedelta64[ns] dtype, treating
+    the integers as multiples of the given timedelta unit.
+
+    Parameters
+    ----------
+    data : numpy.ndarray with integer-dtype
+    unit : str, default "ns"
+        The timedelta unit to treat integers as multiples of.
+
+    Returns
+    -------
+    numpy.ndarray : timedelta64[ns] array converted from data
+    bool : whether a copy was made
+    """
+    copy_made = False
+    unit = unit if unit is not None else "ns"
+
+    if data.dtype != np.int64:
+        # converting to int64 makes a copy, so we can avoid
+        # re-copying later
+        data = data.astype(np.int64)
+        copy_made = True
+
+    if unit != "ns":
+        dtype_str = f"timedelta64[{unit}]"
+        data = data.view(dtype_str)
+
+        data = astype_overflowsafe(data, dtype=TD64NS_DTYPE)
+
+        # the astype conversion makes a copy, so we can avoid re-copying later
+        copy_made = True
+
+    else:
+        data = data.view("timedelta64[ns]")
+
+    return data, copy_made
+
+
+def _objects_to_td64ns(
+    data, unit=None, errors: DateTimeErrorChoices = "raise"
+) -> np.ndarray:
+    """
+    Convert a object-dtyped or string-dtyped array into an
+    timedelta64[ns]-dtyped array.
+
+    Parameters
+    ----------
+    data : ndarray or Index
+    unit : str, default "ns"
+        The timedelta unit to treat integers as multiples of.
+        Must not be specified if the data contains a str.
+    errors : {"raise", "coerce", "ignore"}, default "raise"
+        How to handle elements that cannot be converted to timedelta64[ns].
+        See ``pandas.to_timedelta`` for details.
+
+    Returns
+    -------
+    numpy.ndarray : timedelta64[ns] array converted from data
+
+    Raises
+    ------
+    ValueError : Data cannot be converted to timedelta64[ns].
+
+    Notes
+    -----
+    Unlike `pandas.to_timedelta`, if setting `errors=ignore` will not cause
+    errors to be ignored; they are caught and subsequently ignored at a
+    higher level.
+    """
+    # coerce Index to np.ndarray, converting string-dtype if necessary
+    values = np.asarray(data, dtype=np.object_)
+
+    result = array_to_timedelta64(values, unit=unit, errors=errors)
+    return result.view("timedelta64[ns]")
+
+
+def _validate_td64_dtype(dtype) -> DtypeObj:
+    dtype = pandas_dtype(dtype)
+    if dtype == np.dtype("m8"):
+        # no precision disallowed GH#24806
+        msg = (
+            "Passing in 'timedelta' dtype with no precision is not allowed. "
+            "Please pass in 'timedelta64[ns]' instead."
+        )
+        raise ValueError(msg)
+
+    if not lib.is_np_dtype(dtype, "m"):
+        raise ValueError(f"dtype '{dtype}' is invalid, should be np.timedelta64 dtype")
+    elif not is_supported_dtype(dtype):
+        raise ValueError("Supported timedelta64 resolutions are 's', 'ms', 'us', 'ns'")
+
+    return dtype
diff --git a/pandas/pandas/core/base.py b/pandas/pandas/core/base.py
new file mode 100644
index 0000000000000..a64cd8633c1db
--- /dev/null
+++ b/pandas/pandas/core/base.py
@@ -0,0 +1,1490 @@
+"""
+Base and utility classes for pandas objects.
+"""
+
+from __future__ import annotations
+
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Generic,
+    Literal,
+    cast,
+    final,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._typing import (
+    AxisInt,
+    DtypeObj,
+    IndexLabel,
+    NDFrameT,
+    Self,
+    Shape,
+    npt,
+)
+from pandas.compat import PYPY
+from pandas.compat.numpy import function as nv
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+)
+
+from pandas.core.dtypes.cast import can_hold_element
+from pandas.core.dtypes.common import (
+    is_object_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    remove_na_arraylike,
+)
+
+from pandas.core import (
+    algorithms,
+    nanops,
+    ops,
+)
+from pandas.core.accessor import DirNamesMixin
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays import ExtensionArray
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterator,
+    )
+
+    from pandas._typing import (
+        DropKeep,
+        NumpySorter,
+        NumpyValueArrayLike,
+        ScalarLike_co,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+
+_shared_docs: dict[str, str] = {}
+
+
+class PandasObject(DirNamesMixin):
+    """
+    Baseclass for various pandas objects.
+    """
+
+    # results from calls to methods decorated with cache_readonly get added to _cache
+    _cache: dict[str, Any]
+
+    @property
+    def _constructor(self) -> type[Self]:
+        """
+        Class constructor (for this class it's just `__class__`).
+        """
+        return type(self)
+
+    def __repr__(self) -> str:
+        """
+        Return a string representation for a particular object.
+        """
+        # Should be overwritten by base classes
+        return object.__repr__(self)
+
+    def _reset_cache(self, key: str | None = None) -> None:
+        """
+        Reset cached properties. If ``key`` is passed, only clears that key.
+        """
+        if not hasattr(self, "_cache"):
+            return
+        if key is None:
+            self._cache.clear()
+        else:
+            self._cache.pop(key, None)
+
+    def __sizeof__(self) -> int:
+        """
+        Generates the total memory usage for an object that returns
+        either a value or Series of values
+        """
+        memory_usage = getattr(self, "memory_usage", None)
+        if memory_usage:
+            mem = memory_usage(deep=True)
+            return int(mem if is_scalar(mem) else mem.sum())
+
+        # no memory_usage attribute, so fall back to object's 'sizeof'
+        return super().__sizeof__()
+
+
+class NoNewAttributesMixin:
+    """
+    Mixin which prevents adding new attributes.
+
+    Prevents additional attributes via xxx.attribute = "something" after a
+    call to `self.__freeze()`. Mainly used to prevent the user from using
+    wrong attributes on an accessor (`Series.cat/.str/.dt`).
+
+    If you really want to add a new attribute at a later time, you need to use
+    `object.__setattr__(self, key, value)`.
+    """
+
+    def _freeze(self) -> None:
+        """
+        Prevents setting additional attributes.
+        """
+        object.__setattr__(self, "__frozen", True)
+
+    # prevent adding any attribute via s.xxx.new_attribute = ...
+    def __setattr__(self, key: str, value) -> None:
+        # _cache is used by a decorator
+        # We need to check both 1.) cls.__dict__ and 2.) getattr(self, key)
+        # because
+        # 1.) getattr is false for attributes that raise errors
+        # 2.) cls.__dict__ doesn't traverse into base classes
+        if getattr(self, "__frozen", False) and not (
+            key == "_cache"
+            or key in type(self).__dict__
+            or getattr(self, key, None) is not None
+        ):
+            raise AttributeError(f"You cannot add any new attribute '{key}'")
+        object.__setattr__(self, key, value)
+
+
+class SelectionMixin(Generic[NDFrameT]):
+    """
+    mixin implementing the selection & aggregation interface on a group-like
+    object sub-classes need to define: obj, exclusions
+    """
+
+    obj: NDFrameT
+    _selection: IndexLabel | None = None
+    exclusions: frozenset[Hashable]
+    _internal_names = ["_cache", "__setstate__"]
+    _internal_names_set = set(_internal_names)
+
+    @final
+    @property
+    def _selection_list(self):
+        if not isinstance(
+            self._selection, (list, tuple, ABCSeries, ABCIndex, np.ndarray)
+        ):
+            return [self._selection]
+        return self._selection
+
+    @cache_readonly
+    def _selected_obj(self):
+        if self._selection is None or isinstance(self.obj, ABCSeries):
+            return self.obj
+        else:
+            return self.obj[self._selection]
+
+    @final
+    @cache_readonly
+    def ndim(self) -> int:
+        return self._selected_obj.ndim
+
+    @final
+    @cache_readonly
+    def _obj_with_exclusions(self):
+        if isinstance(self.obj, ABCSeries):
+            return self.obj
+
+        if self._selection is not None:
+            return self.obj[self._selection_list]
+
+        if len(self.exclusions) > 0:
+            # equivalent to `self.obj.drop(self.exclusions, axis=1)
+            #  but this avoids consolidating and making a copy
+            # TODO: following GH#45287 can we now use .drop directly without
+            #  making a copy?
+            return self.obj._drop_axis(self.exclusions, axis=1, only_slice=True)
+        else:
+            return self.obj
+
+    def __getitem__(self, key):
+        if self._selection is not None:
+            raise IndexError(f"Column(s) {self._selection} already selected")
+
+        if isinstance(key, (list, tuple, ABCSeries, ABCIndex, np.ndarray)):
+            if len(self.obj.columns.intersection(key)) != len(set(key)):
+                bad_keys = list(set(key).difference(self.obj.columns))
+                raise KeyError(f"Columns not found: {str(bad_keys)[1:-1]}")
+            return self._gotitem(list(key), ndim=2)
+
+        else:
+            if key not in self.obj:
+                raise KeyError(f"Column not found: {key}")
+            ndim = self.obj[key].ndim
+            return self._gotitem(key, ndim=ndim)
+
+    def _gotitem(self, key, ndim: int, subset=None):
+        """
+        sub-classes to define
+        return a sliced object
+
+        Parameters
+        ----------
+        key : str / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        raise AbstractMethodError(self)
+
+    @final
+    def _infer_selection(self, key, subset: Series | DataFrame):
+        """
+        Infer the `selection` to pass to our constructor in _gotitem.
+        """
+        # Shared by Rolling and Resample
+        selection = None
+        if subset.ndim == 2 and (
+            (lib.is_scalar(key) and key in subset) or lib.is_list_like(key)
+        ):
+            selection = key
+        elif subset.ndim == 1 and lib.is_scalar(key) and key == subset.name:
+            selection = key
+        return selection
+
+    def aggregate(self, func, *args, **kwargs):
+        raise AbstractMethodError(self)
+
+    agg = aggregate
+
+
+class IndexOpsMixin(OpsMixin):
+    """
+    Common ops mixin to support a unified interface / docs for Series / Index
+    """
+
+    # ndarray compatibility
+    __array_priority__ = 1000
+    _hidden_attrs: frozenset[str] = frozenset(
+        ["tolist"]  # tolist is not deprecated, just suppressed in the __dir__
+    )
+
+    @property
+    def dtype(self) -> DtypeObj:
+        # must be defined here as a property for mypy
+        raise AbstractMethodError(self)
+
+    @property
+    def _values(self) -> ExtensionArray | np.ndarray:
+        # must be defined here as a property for mypy
+        raise AbstractMethodError(self)
+
+    @final
+    def transpose(self, *args, **kwargs) -> Self:
+        """
+        Return the transpose, which is by definition self.
+
+        Returns
+        -------
+        %(klass)s
+        """
+        nv.validate_transpose(args, kwargs)
+        return self
+
+    T = property(
+        transpose,
+        doc="""
+        Return the transpose, which is by definition self.
+
+        See Also
+        --------
+        Index : Immutable sequence used for indexing and alignment.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(['Ant', 'Bear', 'Cow'])
+        >>> s
+        0     Ant
+        1    Bear
+        2     Cow
+        dtype: object
+        >>> s.T
+        0     Ant
+        1    Bear
+        2     Cow
+        dtype: object
+
+        For Index:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.T
+        Index([1, 2, 3], dtype='int64')
+        """,
+    )
+
+    @property
+    def shape(self) -> Shape:
+        """
+        Return a tuple of the shape of the underlying data.
+
+        See Also
+        --------
+        Series.ndim : Number of dimensions of the underlying data.
+        Series.size : Return the number of elements in the underlying data.
+        Series.nbytes : Return the number of bytes in the underlying data.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.shape
+        (3,)
+        """
+        return self._values.shape
+
+    def __len__(self) -> int:
+        # We need this defined here for mypy
+        raise AbstractMethodError(self)
+
+    # Temporarily avoid using `-> Literal[1]:` because of an IPython (jedi) bug
+    # https://github.com/ipython/ipython/issues/14412
+    # https://github.com/davidhalter/jedi/issues/1990
+    @property
+    def ndim(self) -> int:
+        """
+        Number of dimensions of the underlying data, by definition 1.
+
+        See Also
+        --------
+        Series.size: Return the number of elements in the underlying data.
+        Series.shape: Return a tuple of the shape of the underlying data.
+        Series.dtype: Return the dtype object of the underlying data.
+        Series.values: Return Series as ndarray or ndarray-like depending on the dtype.
+
+        Examples
+        --------
+        >>> s = pd.Series(["Ant", "Bear", "Cow"])
+        >>> s
+        0     Ant
+        1    Bear
+        2     Cow
+        dtype: object
+        >>> s.ndim
+        1
+
+        For Index:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.ndim
+        1
+        """
+        return 1
+
+    @final
+    def item(self):
+        """
+        Return the first element of the underlying data as a Python scalar.
+
+        Returns
+        -------
+        scalar
+            The first element of Series or Index.
+
+        Raises
+        ------
+        ValueError
+            If the data is not length = 1.
+
+        See Also
+        --------
+        Index.values : Returns an array representing the data in the Index.
+        Series.head : Returns the first `n` rows.
+
+        Examples
+        --------
+        >>> s = pd.Series([1])
+        >>> s.item()
+        1
+
+        For an index:
+
+        >>> s = pd.Series([1], index=["a"])
+        >>> s.index.item()
+        'a'
+        """
+        if len(self) == 1:
+            return next(iter(self))
+        raise ValueError("can only convert an array of size 1 to a Python scalar")
+
+    @property
+    def nbytes(self) -> int:
+        """
+        Return the number of bytes in the underlying data.
+
+        See Also
+        --------
+        Series.ndim : Number of dimensions of the underlying data.
+        Series.size : Return the number of elements in the underlying data.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["Ant", "Bear", "Cow"])
+        >>> s
+        0     Ant
+        1    Bear
+        2     Cow
+        dtype: object
+        >>> s.nbytes
+        24
+
+        For Index:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.nbytes
+        24
+        """
+        return self._values.nbytes
+
+    @property
+    def size(self) -> int:
+        """
+        Return the number of elements in the underlying data.
+
+        See Also
+        --------
+        Series.ndim: Number of dimensions of the underlying data, by definition 1.
+        Series.shape: Return a tuple of the shape of the underlying data.
+        Series.dtype: Return the dtype object of the underlying data.
+        Series.values: Return Series as ndarray or ndarray-like depending on the dtype.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series(["Ant", "Bear", "Cow"])
+        >>> s
+        0     Ant
+        1    Bear
+        2     Cow
+        dtype: object
+        >>> s.size
+        3
+
+        For Index:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.size
+        3
+        """
+        return len(self._values)
+
+    @property
+    def array(self) -> ExtensionArray:
+        """
+        The ExtensionArray of the data backing this Series or Index.
+
+        This property provides direct access to the underlying array data of a
+        Series or Index without requiring conversion to a NumPy array. It
+        returns an ExtensionArray, which is the native storage format for
+        pandas extension dtypes.
+
+        Returns
+        -------
+        ExtensionArray
+            An ExtensionArray of the values stored within. For extension
+            types, this is the actual array. For NumPy native types, this
+            is a thin (no copy) wrapper around :class:`numpy.ndarray`.
+
+            ``.array`` differs from ``.values``, which may require converting
+            the data to a different form.
+
+        See Also
+        --------
+        Index.to_numpy : Similar method that always returns a NumPy array.
+        Series.to_numpy : Similar method that always returns a NumPy array.
+
+        Notes
+        -----
+        This table lays out the different array types for each extension
+        dtype within pandas.
+
+        ================== =============================
+        dtype              array type
+        ================== =============================
+        category           Categorical
+        period             PeriodArray
+        interval           IntervalArray
+        IntegerNA          IntegerArray
+        string             StringArray
+        boolean            BooleanArray
+        datetime64[ns, tz] DatetimeArray
+        ================== =============================
+
+        For any 3rd-party extension types, the array type will be an
+        ExtensionArray.
+
+        For all remaining dtypes ``.array`` will be a
+        :class:`arrays.NumpyExtensionArray` wrapping the actual ndarray
+        stored within. If you absolutely need a NumPy array (possibly with
+        copying / coercing data), then use :meth:`Series.to_numpy` instead.
+
+        Examples
+        --------
+        For regular NumPy types like int, and float, a NumpyExtensionArray
+        is returned.
+
+        >>> pd.Series([1, 2, 3]).array
+        <NumpyExtensionArray>
+        [1, 2, 3]
+        Length: 3, dtype: int64
+
+        For extension types, like Categorical, the actual ExtensionArray
+        is returned
+
+        >>> ser = pd.Series(pd.Categorical(["a", "b", "a"]))
+        >>> ser.array
+        ['a', 'b', 'a']
+        Categories (2, object): ['a', 'b']
+        """
+        raise AbstractMethodError(self)
+
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+        **kwargs,
+    ) -> np.ndarray:
+        """
+        A NumPy ndarray representing the values in this Series or Index.
+
+        Parameters
+        ----------
+        dtype : str or numpy.dtype, optional
+            The dtype to pass to :meth:`numpy.asarray`.
+        copy : bool, default False
+            Whether to ensure that the returned value is not a view on
+            another array. Note that ``copy=False`` does not *ensure* that
+            ``to_numpy()`` is no-copy. Rather, ``copy=True`` ensure that
+            a copy is made, even if not strictly necessary.
+        na_value : Any, optional
+            The value to use for missing values. The default value depends
+            on `dtype` and the type of the array.
+        **kwargs
+            Additional keywords passed through to the ``to_numpy`` method
+            of the underlying array (for extension arrays).
+
+        Returns
+        -------
+        numpy.ndarray
+            The NumPy ndarray holding the values from this Series or Index.
+            The dtype of the array may differ. See Notes.
+
+        See Also
+        --------
+        Series.array : Get the actual data stored within.
+        Index.array : Get the actual data stored within.
+        DataFrame.to_numpy : Similar method for DataFrame.
+
+        Notes
+        -----
+        The returned array will be the same up to equality (values equal
+        in `self` will be equal in the returned array; likewise for values
+        that are not equal). When `self` contains an ExtensionArray, the
+        dtype may be different. For example, for a category-dtype Series,
+        ``to_numpy()`` will return a NumPy array and the categorical dtype
+        will be lost.
+
+        For NumPy dtypes, this will be a reference to the actual data stored
+        in this Series or Index (assuming ``copy=False``). Modifying the result
+        in place will modify the data stored in the Series or Index (not that
+        we recommend doing that).
+
+        For extension types, ``to_numpy()`` *may* require copying data and
+        coercing the result to a NumPy type (possibly object), which may be
+        expensive. When you need a no-copy reference to the underlying data,
+        :attr:`Series.array` should be used instead.
+
+        This table lays out the different dtypes and default return types of
+        ``to_numpy()`` for various dtypes within pandas.
+
+        ================== ================================
+        dtype              array type
+        ================== ================================
+        category[T]        ndarray[T] (same dtype as input)
+        period             ndarray[object] (Periods)
+        interval           ndarray[object] (Intervals)
+        IntegerNA          ndarray[object]
+        datetime64[ns]     datetime64[ns]
+        datetime64[ns, tz] ndarray[object] (Timestamps)
+        ================== ================================
+
+        Examples
+        --------
+        >>> ser = pd.Series(pd.Categorical(["a", "b", "a"]))
+        >>> ser.to_numpy()
+        array(['a', 'b', 'a'], dtype=object)
+
+        Specify the `dtype` to control how datetime-aware data is represented.
+        Use ``dtype=object`` to return an ndarray of pandas :class:`Timestamp`
+        objects, each with the correct ``tz``.
+
+        >>> ser = pd.Series(pd.date_range("2000", periods=2, tz="CET"))
+        >>> ser.to_numpy(dtype=object)
+        array([Timestamp('2000-01-01 00:00:00+0100', tz='CET'),
+               Timestamp('2000-01-02 00:00:00+0100', tz='CET')],
+              dtype=object)
+
+        Or ``dtype='datetime64[ns]'`` to return an ndarray of native
+        datetime64 values. The values are converted to UTC and the timezone
+        info is dropped.
+
+        >>> ser.to_numpy(dtype="datetime64[ns]")
+        ... # doctest: +ELLIPSIS
+        array(['1999-12-31T23:00:00.000000000', '2000-01-01T23:00:00...'],
+              dtype='datetime64[ns]')
+        """
+        if isinstance(self.dtype, ExtensionDtype):
+            return self.array.to_numpy(dtype, copy=copy, na_value=na_value, **kwargs)
+        elif kwargs:
+            bad_keys = next(iter(kwargs.keys()))
+            raise TypeError(
+                f"to_numpy() got an unexpected keyword argument '{bad_keys}'"
+            )
+
+        fillna = (
+            na_value is not lib.no_default
+            # no need to fillna with np.nan if we already have a float dtype
+            and not (na_value is np.nan and np.issubdtype(self.dtype, np.floating))
+        )
+
+        values = self._values
+        if fillna and self.hasnans:
+            if not can_hold_element(values, na_value):
+                # if we can't hold the na_value asarray either makes a copy or we
+                # error before modifying values. The asarray later on thus won't make
+                # another copy
+                values = np.asarray(values, dtype=dtype)
+            else:
+                values = values.copy()
+
+            values[np.asanyarray(isna(self))] = na_value
+
+        result = np.asarray(values, dtype=dtype)
+
+        if (copy and not fillna) or not copy:
+            if np.shares_memory(self._values[:2], result[:2]):
+                # Take slices to improve performance of check
+                if not copy:
+                    result = result.view()
+                    result.flags.writeable = False
+                else:
+                    result = result.copy()
+
+        return result
+
+    @final
+    @property
+    def empty(self) -> bool:
+        """
+        Indicator whether Index is empty.
+
+        An Index is considered empty if it has no elements. This property can be
+        useful for quickly checking the state of an Index, especially in data
+        processing and analysis workflows where handling of empty datasets might
+        be required.
+
+        Returns
+        -------
+        bool
+            If Index is empty, return True, if not return False.
+
+        See Also
+        --------
+        Index.size : Return the number of elements in the underlying data.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.empty
+        False
+
+        >>> idx_empty = pd.Index([])
+        >>> idx_empty
+        Index([], dtype='object')
+        >>> idx_empty.empty
+        True
+
+        If we only have NaNs in our DataFrame, it is not considered empty!
+
+        >>> idx = pd.Index([np.nan, np.nan])
+        >>> idx
+        Index([nan, nan], dtype='float64')
+        >>> idx.empty
+        False
+        """
+        return not self.size
+
+    @doc(op="max", oppose="min", value="largest")
+    def argmax(
+        self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs
+    ) -> int:
+        """
+        Return int position of the {value} value in the Series.
+
+        If the {op}imum is achieved in multiple locations,
+        the first row position is returned.
+
+        Parameters
+        ----------
+        axis : {{None}}
+            Unused. Parameter needed for compatibility with DataFrame.
+        skipna : bool, default True
+            Exclude NA/null values. If the entire Series is NA, or if ``skipna=False``
+            and there is an NA value, this method will raise a ``ValueError``.
+        *args, **kwargs
+            Additional arguments and keywords for compatibility with NumPy.
+
+        Returns
+        -------
+        int
+            Row position of the {op}imum value.
+
+        See Also
+        --------
+        Series.arg{op} : Return position of the {op}imum value.
+        Series.arg{oppose} : Return position of the {oppose}imum value.
+        numpy.ndarray.arg{op} : Equivalent method for numpy arrays.
+        Series.idxmax : Return index label of the maximum values.
+        Series.idxmin : Return index label of the minimum values.
+
+        Examples
+        --------
+        Consider dataset containing cereal calories
+
+        >>> s = pd.Series(
+        ...     [100.0, 110.0, 120.0, 110.0],
+        ...     index=[
+        ...         "Corn Flakes",
+        ...         "Almond Delight",
+        ...         "Cinnamon Toast Crunch",
+        ...         "Cocoa Puff",
+        ...     ],
+        ... )
+        >>> s
+        Corn Flakes              100.0
+        Almond Delight           110.0
+        Cinnamon Toast Crunch    120.0
+        Cocoa Puff               110.0
+        dtype: float64
+
+        >>> s.argmax()
+        2
+        >>> s.argmin()
+        0
+
+        The maximum cereal calories is the third element and
+        the minimum cereal calories is the first element,
+        since series is zero-indexed.
+        """
+        delegate = self._values
+        nv.validate_minmax_axis(axis)
+        skipna = nv.validate_argmax_with_skipna(skipna, args, kwargs)
+
+        if isinstance(delegate, ExtensionArray):
+            return delegate.argmax(skipna=skipna)
+        else:
+            result = nanops.nanargmax(delegate, skipna=skipna)
+            # error: Incompatible return value type (got "Union[int, ndarray]", expected
+            # "int")
+            return result  # type: ignore[return-value]
+
+    @doc(argmax, op="min", oppose="max", value="smallest")
+    def argmin(
+        self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs
+    ) -> int:
+        delegate = self._values
+        nv.validate_minmax_axis(axis)
+        skipna = nv.validate_argmax_with_skipna(skipna, args, kwargs)
+
+        if isinstance(delegate, ExtensionArray):
+            return delegate.argmin(skipna=skipna)
+        else:
+            result = nanops.nanargmin(delegate, skipna=skipna)
+            # error: Incompatible return value type (got "Union[int, ndarray]", expected
+            # "int")
+            return result  # type: ignore[return-value]
+
+    def tolist(self) -> list:
+        """
+        Return a list of the values.
+
+        These are each a scalar type, which is a Python scalar
+        (for str, int, float) or a pandas scalar
+        (for Timestamp/Timedelta/Interval/Period)
+
+        Returns
+        -------
+        list
+            List containing the values as Python or pandas scalers.
+
+        See Also
+        --------
+        numpy.ndarray.tolist : Return the array as an a.ndim-levels deep
+            nested list of Python scalars.
+
+        Examples
+        --------
+        For Series
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.to_list()
+        [1, 2, 3]
+
+        For Index:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+
+        >>> idx.to_list()
+        [1, 2, 3]
+        """
+        return self._values.tolist()
+
+    to_list = tolist
+
+    def __iter__(self) -> Iterator:
+        """
+        Return an iterator of the values.
+
+        These are each a scalar type, which is a Python scalar
+        (for str, int, float) or a pandas scalar
+        (for Timestamp/Timedelta/Interval/Period)
+
+        Returns
+        -------
+        iterator
+            An iterator yielding scalar values from the Series.
+
+        See Also
+        --------
+        Series.items : Lazily iterate over (index, value) tuples.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> for x in s:
+        ...     print(x)
+        1
+        2
+        3
+        """
+        # We are explicitly making element iterators.
+        if not isinstance(self._values, np.ndarray):
+            # Check type instead of dtype to catch DTA/TDA
+            return iter(self._values)
+        else:
+            return map(self._values.item, range(self._values.size))
+
+    @cache_readonly
+    def hasnans(self) -> bool:
+        """
+        Return True if there are any NaNs.
+
+        Enables various performance speedups.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Series.isna : Detect missing values.
+        Series.notna : Detect existing (non-missing) values.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, None])
+        >>> s
+        0    1.0
+        1    2.0
+        2    3.0
+        3    NaN
+        dtype: float64
+        >>> s.hasnans
+        True
+        """
+        # error: Item "bool" of "Union[bool, ndarray[Any, dtype[bool_]], NDFrame]"
+        # has no attribute "any"
+        return bool(isna(self).any())  # type: ignore[union-attr]
+
+    @final
+    def _map_values(self, mapper, na_action=None):
+        """
+        An internal function that maps values using the input
+        correspondence (which can be a dict, Series, or function).
+
+        Parameters
+        ----------
+        mapper : function, dict, or Series
+            The input correspondence object
+        na_action : {None, 'ignore'}
+            If 'ignore', propagate NA values, without passing them to the
+            mapping function
+
+        Returns
+        -------
+        Union[Index, MultiIndex], inferred
+            The output of the mapping function applied to the index.
+            If the function returns a tuple with more than one element
+            a MultiIndex will be returned.
+        """
+        arr = self._values
+
+        if isinstance(arr, ExtensionArray):
+            return arr.map(mapper, na_action=na_action)
+
+        return algorithms.map_array(arr, mapper, na_action=na_action)
+
+    def value_counts(
+        self,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        bins=None,
+        dropna: bool = True,
+    ) -> Series:
+        """
+        Return a Series containing counts of unique values.
+
+        The resulting object will be in descending order so that the
+        first element is the most frequently-occurring element.
+        Excludes NA values by default.
+
+        Parameters
+        ----------
+        normalize : bool, default False
+            If True then the object returned will contain the relative
+            frequencies of the unique values.
+        sort : bool, default True
+            Sort by frequencies when True. Preserve the order of the data when False.
+        ascending : bool, default False
+            Sort in ascending order.
+        bins : int, optional
+            Rather than count values, group them into half-open bins,
+            a convenience for ``pd.cut``, only works with numeric data.
+        dropna : bool, default True
+            Don't include counts of NaN.
+
+        Returns
+        -------
+        Series
+            Series containing counts of unique values.
+
+        See Also
+        --------
+        Series.count: Number of non-NA elements in a Series.
+        DataFrame.count: Number of non-NA elements in a DataFrame.
+        DataFrame.value_counts: Equivalent method on DataFrames.
+
+        Examples
+        --------
+        >>> index = pd.Index([3, 1, 2, 3, 4, np.nan])
+        >>> index.value_counts()
+        3.0    2
+        1.0    1
+        2.0    1
+        4.0    1
+        Name: count, dtype: int64
+
+        With `normalize` set to `True`, returns the relative frequency by
+        dividing all values by the sum of values.
+
+        >>> s = pd.Series([3, 1, 2, 3, 4, np.nan])
+        >>> s.value_counts(normalize=True)
+        3.0    0.4
+        1.0    0.2
+        2.0    0.2
+        4.0    0.2
+        Name: proportion, dtype: float64
+
+        **bins**
+
+        Bins can be useful for going from a continuous variable to a
+        categorical variable; instead of counting unique
+        apparitions of values, divide the index in the specified
+        number of half-open bins.
+
+        >>> s.value_counts(bins=3)
+        (0.996, 2.0]    2
+        (2.0, 3.0]      2
+        (3.0, 4.0]      1
+        Name: count, dtype: int64
+
+        **dropna**
+
+        With `dropna` set to `False` we can also see NaN index values.
+
+        >>> s.value_counts(dropna=False)
+        3.0    2
+        1.0    1
+        2.0    1
+        4.0    1
+        NaN    1
+        Name: count, dtype: int64
+
+        **Categorical Dtypes**
+
+        Rows with categorical type will be counted as one group
+        if they have same categories and order.
+        In the example below, even though ``a``, ``c``, and ``d``
+        all have the same data types of ``category``,
+        only ``c`` and ``d`` will be counted as one group
+        since ``a`` doesn't have the same categories.
+
+        >>> df = pd.DataFrame({"a": [1], "b": ["2"], "c": [3], "d": [3]})
+        >>> df = df.astype({"a": "category", "c": "category", "d": "category"})
+        >>> df
+           a  b  c  d
+        0  1  2  3  3
+
+        >>> df.dtypes
+        a    category
+        b      object
+        c    category
+        d    category
+        dtype: object
+
+        >>> df.dtypes.value_counts()
+        category    2
+        category    1
+        object      1
+        Name: count, dtype: int64
+        """
+        return algorithms.value_counts_internal(
+            self,
+            sort=sort,
+            ascending=ascending,
+            normalize=normalize,
+            bins=bins,
+            dropna=dropna,
+        )
+
+    def unique(self):
+        values = self._values
+        if not isinstance(values, np.ndarray):
+            # i.e. ExtensionArray
+            result = values.unique()
+        else:
+            result = algorithms.unique1d(values)
+        return result
+
+    @final
+    def nunique(self, dropna: bool = True) -> int:
+        """
+        Return number of unique elements in the object.
+
+        Excludes NA values by default.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include NaN in the count.
+
+        Returns
+        -------
+        int
+            A integer indicating the number of unique elements in the object.
+
+        See Also
+        --------
+        DataFrame.nunique: Method nunique for DataFrame.
+        Series.count: Count non-NA/null observations in the Series.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 3, 5, 7, 7])
+        >>> s
+        0    1
+        1    3
+        2    5
+        3    7
+        4    7
+        dtype: int64
+
+        >>> s.nunique()
+        4
+        """
+        uniqs = self.unique()
+        if dropna:
+            uniqs = remove_na_arraylike(uniqs)
+        return len(uniqs)
+
+    @property
+    def is_unique(self) -> bool:
+        """
+        Return True if values in the object are unique.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Series.unique : Return unique values of Series object.
+        Series.drop_duplicates : Return Series with duplicate values removed.
+        Series.duplicated : Indicate duplicate Series values.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.is_unique
+        True
+
+        >>> s = pd.Series([1, 2, 3, 1])
+        >>> s.is_unique
+        False
+        """
+        return self.nunique(dropna=False) == len(self)
+
+    @property
+    def is_monotonic_increasing(self) -> bool:
+        """
+        Return True if values in the object are monotonically increasing.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Series.is_monotonic_decreasing : Return boolean if values in the object are
+            monotonically decreasing.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 2])
+        >>> s.is_monotonic_increasing
+        True
+
+        >>> s = pd.Series([3, 2, 1])
+        >>> s.is_monotonic_increasing
+        False
+        """
+        from pandas import Index
+
+        return Index(self).is_monotonic_increasing
+
+    @property
+    def is_monotonic_decreasing(self) -> bool:
+        """
+        Return True if values in the object are monotonically decreasing.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Series.is_monotonic_increasing : Return boolean if values in the object are
+            monotonically increasing.
+
+        Examples
+        --------
+        >>> s = pd.Series([3, 2, 2, 1])
+        >>> s.is_monotonic_decreasing
+        True
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.is_monotonic_decreasing
+        False
+        """
+        from pandas import Index
+
+        return Index(self).is_monotonic_decreasing
+
+    @final
+    def _memory_usage(self, deep: bool = False) -> int:
+        """
+        Memory usage of the values.
+
+        Parameters
+        ----------
+        deep : bool, default False
+            Introspect the data deeply, interrogate
+            `object` dtypes for system-level memory consumption.
+
+        Returns
+        -------
+        bytes used
+            Returns memory usage of the values in the Index in bytes.
+
+        See Also
+        --------
+        numpy.ndarray.nbytes : Total bytes consumed by the elements of the
+            array.
+
+        Notes
+        -----
+        Memory usage does not include memory consumed by elements that
+        are not components of the array if deep=False or if used on PyPy
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.memory_usage()
+        24
+        """
+        if hasattr(self.array, "memory_usage"):
+            return self.array.memory_usage(  # pyright: ignore[reportAttributeAccessIssue]
+                deep=deep,
+            )
+
+        v = self.array.nbytes
+        if deep and is_object_dtype(self.dtype) and not PYPY:
+            values = cast(np.ndarray, self._values)
+            v += lib.memory_usage_of_objects(values)
+        return v
+
+    @doc(
+        algorithms.factorize,
+        values="",
+        order="",
+        size_hint="",
+        sort=textwrap.dedent(
+            """\
+            sort : bool, default False
+                Sort `uniques` and shuffle `codes` to maintain the
+                relationship.
+            """
+        ),
+    )
+    def factorize(
+        self,
+        sort: bool = False,
+        use_na_sentinel: bool = True,
+    ) -> tuple[npt.NDArray[np.intp], Index]:
+        codes, uniques = algorithms.factorize(
+            self._values, sort=sort, use_na_sentinel=use_na_sentinel
+        )
+        if uniques.dtype == np.float16:
+            uniques = uniques.astype(np.float32)
+
+        if isinstance(self, ABCMultiIndex):
+            # preserve MultiIndex
+            uniques = self._constructor(uniques)
+        else:
+            from pandas import Index
+
+            try:
+                uniques = Index(uniques, dtype=self.dtype)
+            except NotImplementedError:
+                # not all dtypes are supported in Index that are allowed for Series
+                # e.g. float16 or bytes
+                uniques = Index(uniques)
+        return codes, uniques
+
+    _shared_docs["searchsorted"] = """
+        Find indices where elements should be inserted to maintain order.
+
+        Find the indices into a sorted {klass} `self` such that, if the
+        corresponding elements in `value` were inserted before the indices,
+        the order of `self` would be preserved.
+
+        .. note::
+
+            The {klass} *must* be monotonically sorted, otherwise
+            wrong locations will likely be returned. Pandas does *not*
+            check this for you.
+
+        Parameters
+        ----------
+        value : array-like or scalar
+            Values to insert into `self`.
+        side : {{'left', 'right'}}, optional
+            If 'left', the index of the first suitable location found is given.
+            If 'right', return the last such index.  If there is no suitable
+            index, return either 0 or N (where N is the length of `self`).
+        sorter : 1-D array-like, optional
+            Optional array of integer indices that sort `self` into ascending
+            order. They are typically the result of ``np.argsort``.
+
+        Returns
+        -------
+        int or array of int
+            A scalar or array of insertion points with the
+            same shape as `value`.
+
+        See Also
+        --------
+        sort_values : Sort by the values along either axis.
+        numpy.searchsorted : Similar method from NumPy.
+
+        Notes
+        -----
+        Binary search is used to find the required insertion points.
+
+        Examples
+        --------
+        >>> ser = pd.Series([1, 2, 3])
+        >>> ser
+        0    1
+        1    2
+        2    3
+        dtype: int64
+
+        >>> ser.searchsorted(4)
+        3
+
+        >>> ser.searchsorted([0, 4])
+        array([0, 3])
+
+        >>> ser.searchsorted([1, 3], side='left')
+        array([0, 2])
+
+        >>> ser.searchsorted([1, 3], side='right')
+        array([1, 3])
+
+        >>> ser = pd.Series(pd.to_datetime(['3/11/2000', '3/12/2000', '3/13/2000']))
+        >>> ser
+        0   2000-03-11
+        1   2000-03-12
+        2   2000-03-13
+        dtype: datetime64[s]
+
+        >>> ser.searchsorted('3/14/2000')
+        3
+
+        >>> ser = pd.Categorical(
+        ...     ['apple', 'bread', 'bread', 'cheese', 'milk'], ordered=True
+        ... )
+        >>> ser
+        ['apple', 'bread', 'bread', 'cheese', 'milk']
+        Categories (4, object): ['apple' < 'bread' < 'cheese' < 'milk']
+
+        >>> ser.searchsorted('bread')
+        1
+
+        >>> ser.searchsorted(['bread'], side='right')
+        array([3])
+
+        If the values are not monotonically sorted, wrong locations
+        may be returned:
+
+        >>> ser = pd.Series([2, 1, 3])
+        >>> ser
+        0    2
+        1    1
+        2    3
+        dtype: int64
+
+        >>> ser.searchsorted(1)  # doctest: +SKIP
+        0  # wrong result, correct would be 1
+        """
+
+    # This overload is needed so that the call to searchsorted in
+    # pandas.core.resample.TimeGrouper._get_period_bins picks the correct result
+
+    # error: Overloaded function signatures 1 and 2 overlap with incompatible
+    # return types
+    @overload
+    def searchsorted(  # type: ignore[overload-overlap]
+        self,
+        value: ScalarLike_co,
+        side: Literal["left", "right"] = ...,
+        sorter: NumpySorter = ...,
+    ) -> np.intp: ...
+
+    @overload
+    def searchsorted(
+        self,
+        value: npt.ArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = ...,
+        sorter: NumpySorter = ...,
+    ) -> npt.NDArray[np.intp]: ...
+
+    @doc(_shared_docs["searchsorted"], klass="Index")
+    def searchsorted(
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if isinstance(value, ABCDataFrame):
+            msg = (
+                "Value must be 1-D array-like or scalar, "
+                f"{type(value).__name__} is not supported"
+            )
+            raise ValueError(msg)
+
+        values = self._values
+        if not isinstance(values, np.ndarray):
+            # Going through EA.searchsorted directly improves performance GH#38083
+            return values.searchsorted(value, side=side, sorter=sorter)
+
+        return algorithms.searchsorted(
+            values,
+            value,
+            side=side,
+            sorter=sorter,
+        )
+
+    def drop_duplicates(self, *, keep: DropKeep = "first") -> Self:
+        duplicated = self._duplicated(keep=keep)
+        # error: Value of type "IndexOpsMixin" is not indexable
+        return self[~duplicated]  # type: ignore[index]
+
+    @final
+    def _duplicated(self, keep: DropKeep = "first") -> npt.NDArray[np.bool_]:
+        arr = self._values
+        if isinstance(arr, ExtensionArray):
+            return arr.duplicated(keep=keep)
+        return algorithms.duplicated(arr, keep=keep)
+
+    def _arith_method(self, other, op):
+        res_name = ops.get_op_result_name(self, other)
+
+        lvalues = self._values
+        rvalues = extract_array(other, extract_numpy=True, extract_range=True)
+        rvalues = ops.maybe_prepare_scalar_for_op(rvalues, lvalues.shape)
+        rvalues = ensure_wrapped_if_datetimelike(rvalues)
+        if isinstance(rvalues, range):
+            rvalues = np.arange(rvalues.start, rvalues.stop, rvalues.step)
+
+        with np.errstate(all="ignore"):
+            result = ops.arithmetic_op(lvalues, rvalues, op)
+
+        return self._construct_result(result, name=res_name)
+
+    def _construct_result(self, result, name):
+        """
+        Construct an appropriately-wrapped result from the ArrayLike result
+        of an arithmetic-like operation.
+        """
+        raise AbstractMethodError(self)
diff --git a/pandas/pandas/core/common.py b/pandas/pandas/core/common.py
new file mode 100644
index 0000000000000..100ad312bd839
--- /dev/null
+++ b/pandas/pandas/core/common.py
@@ -0,0 +1,658 @@
+"""
+Misc tools for implementing data structures
+
+Note: pandas.core.common is *not* part of the public API.
+"""
+
+from __future__ import annotations
+
+import builtins
+from collections import (
+    abc,
+    defaultdict,
+)
+from collections.abc import (
+    Callable,
+    Collection,
+    Generator,
+    Hashable,
+    Iterable,
+    Sequence,
+)
+import contextlib
+from functools import partial
+import inspect
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypeVar,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.compat.numpy import np_version_gte1p24
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_integer,
+)
+from pandas.core.dtypes.generic import (
+    ABCExtensionArray,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import iterable_not_string
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        AnyArrayLike,
+        ArrayLike,
+        Concatenate,
+        NpDtype,
+        P,
+        RandomState,
+        T,
+    )
+
+    from pandas import Index
+
+
+def flatten(line):
+    """
+    Flatten an arbitrarily nested sequence.
+
+    Parameters
+    ----------
+    line : sequence
+        The non string sequence to flatten
+
+    Notes
+    -----
+    This doesn't consider strings sequences.
+
+    Returns
+    -------
+    flattened : generator
+    """
+    for element in line:
+        if iterable_not_string(element):
+            yield from flatten(element)
+        else:
+            yield element
+
+
+def consensus_name_attr(objs):
+    name = objs[0].name
+    for obj in objs[1:]:
+        try:
+            if obj.name != name:
+                name = None
+        except ValueError:
+            name = None
+    return name
+
+
+def is_bool_indexer(key: Any) -> bool:
+    """
+    Check whether `key` is a valid boolean indexer.
+
+    Parameters
+    ----------
+    key : Any
+        Only list-likes may be considered boolean indexers.
+        All other types are not considered a boolean indexer.
+        For array-like input, boolean ndarrays or ExtensionArrays
+        with ``_is_boolean`` set are considered boolean indexers.
+
+    Returns
+    -------
+    bool
+        Whether `key` is a valid boolean indexer.
+
+    Raises
+    ------
+    ValueError
+        When the array is an object-dtype ndarray or ExtensionArray
+        and contains missing values.
+
+    See Also
+    --------
+    check_array_indexer : Check that `key` is a valid array to index,
+        and convert to an ndarray.
+    """
+    if isinstance(
+        key, (ABCSeries, np.ndarray, ABCIndex, ABCExtensionArray)
+    ) and not isinstance(key, ABCMultiIndex):
+        if key.dtype == np.object_:
+            key_array = np.asarray(key)
+
+            if not lib.is_bool_array(key_array):
+                na_msg = "Cannot mask with non-boolean array containing NA / NaN values"
+                if lib.is_bool_array(key_array, skipna=True):
+                    # Don't raise on e.g. ["A", "B", np.nan], see
+                    #  test_loc_getitem_list_of_labels_categoricalindex_with_na
+                    raise ValueError(na_msg)
+                return False
+            return True
+        elif is_bool_dtype(key.dtype):
+            return True
+    elif isinstance(key, list):
+        # check if np.array(key).dtype would be bool
+        if len(key) > 0:
+            if type(key) is not list:
+                # GH#42461 cython will raise TypeError if we pass a subclass
+                key = list(key)
+            return lib.is_bool_list(key)
+
+    return False
+
+
+def cast_scalar_indexer(val):
+    """
+    Disallow indexing with a float key, even if that key is a round number.
+
+    Parameters
+    ----------
+    val : scalar
+
+    Returns
+    -------
+    outval : scalar
+    """
+    # assumes lib.is_scalar(val)
+    if lib.is_float(val) and val.is_integer():
+        raise IndexError(
+            # GH#34193
+            "Indexing with a float is no longer supported. Manually convert "
+            "to an integer key instead."
+        )
+    return val
+
+
+def not_none(*args):
+    """
+    Returns a generator consisting of the arguments that are not None.
+    """
+    return (arg for arg in args if arg is not None)
+
+
+def any_none(*args) -> bool:
+    """
+    Returns a boolean indicating if any argument is None.
+    """
+    return any(arg is None for arg in args)
+
+
+def all_none(*args) -> bool:
+    """
+    Returns a boolean indicating if all arguments are None.
+    """
+    return all(arg is None for arg in args)
+
+
+def any_not_none(*args) -> bool:
+    """
+    Returns a boolean indicating if any argument is not None.
+    """
+    return any(arg is not None for arg in args)
+
+
+def all_not_none(*args) -> bool:
+    """
+    Returns a boolean indicating if all arguments are not None.
+    """
+    return all(arg is not None for arg in args)
+
+
+def count_not_none(*args) -> int:
+    """
+    Returns the count of arguments that are not None.
+    """
+    return sum(x is not None for x in args)
+
+
+@overload
+def asarray_tuplesafe(
+    values: ArrayLike | list | tuple | zip, dtype: NpDtype | None = ...
+) -> np.ndarray:
+    # ExtensionArray can only be returned when values is an Index, all other iterables
+    # will return np.ndarray. Unfortunately "all other" cannot be encoded in a type
+    # signature, so instead we special-case some common types.
+    ...
+
+
+@overload
+def asarray_tuplesafe(values: Iterable, dtype: NpDtype | None = ...) -> ArrayLike: ...
+
+
+def asarray_tuplesafe(values: Iterable, dtype: NpDtype | None = None) -> ArrayLike:
+    if not (isinstance(values, (list, tuple)) or hasattr(values, "__array__")):
+        values = list(values)
+    elif isinstance(values, ABCIndex):
+        return values._values
+    elif isinstance(values, ABCSeries):
+        return values._values
+
+    if isinstance(values, list) and dtype in [np.object_, object]:
+        return construct_1d_object_array_from_listlike(values)
+
+    try:
+        with warnings.catch_warnings():
+            # Can remove warning filter once NumPy 1.24 is min version
+            if not np_version_gte1p24:
+                warnings.simplefilter("ignore", np.VisibleDeprecationWarning)
+            result = np.asarray(values, dtype=dtype)
+    except ValueError:
+        # Using try/except since it's more performant than checking is_list_like
+        # over each element
+        # error: Argument 1 to "construct_1d_object_array_from_listlike"
+        # has incompatible type "Iterable[Any]"; expected "Sized"
+        return construct_1d_object_array_from_listlike(values)  # type: ignore[arg-type]
+
+    if issubclass(result.dtype.type, str):
+        result = np.asarray(values, dtype=object)
+
+    if result.ndim == 2:
+        # Avoid building an array of arrays:
+        values = [tuple(x) for x in values]
+        result = construct_1d_object_array_from_listlike(values)
+
+    return result
+
+
+def index_labels_to_array(
+    labels: np.ndarray | Iterable, dtype: NpDtype | None = None
+) -> np.ndarray:
+    """
+    Transform label or iterable of labels to array, for use in Index.
+
+    Parameters
+    ----------
+    dtype : dtype
+        If specified, use as dtype of the resulting array, otherwise infer.
+
+    Returns
+    -------
+    array
+    """
+    if isinstance(labels, (str, tuple)):
+        labels = [labels]
+
+    if not isinstance(labels, (list, np.ndarray)):
+        try:
+            labels = list(labels)
+        except TypeError:  # non-iterable
+            labels = [labels]
+
+    labels = asarray_tuplesafe(labels, dtype=dtype)
+
+    return labels
+
+
+def maybe_make_list(obj):
+    if obj is not None and not isinstance(obj, (tuple, list)):
+        return [obj]
+    return obj
+
+
+def maybe_iterable_to_list(obj: Iterable[T] | T) -> Collection[T] | T:
+    """
+    If obj is Iterable but not list-like, consume into list.
+    """
+    if isinstance(obj, abc.Iterable) and not isinstance(obj, abc.Sized):
+        return list(obj)
+    obj = cast(Collection, obj)
+    return obj
+
+
+def is_null_slice(obj) -> bool:
+    """
+    We have a null slice.
+    """
+    return (
+        isinstance(obj, slice)
+        and obj.start is None
+        and obj.stop is None
+        and obj.step is None
+    )
+
+
+def is_empty_slice(obj) -> bool:
+    """
+    We have an empty slice, e.g. no values are selected.
+    """
+    return (
+        isinstance(obj, slice)
+        and obj.start is not None
+        and obj.stop is not None
+        and obj.start == obj.stop
+    )
+
+
+def is_true_slices(line: abc.Iterable) -> abc.Generator[bool, None, None]:
+    """
+    Find non-trivial slices in "line": yields a bool.
+    """
+    for k in line:
+        yield isinstance(k, slice) and not is_null_slice(k)
+
+
+# TODO: used only once in indexing; belongs elsewhere?
+def is_full_slice(obj, line: int) -> bool:
+    """
+    We have a full length slice.
+    """
+    return (
+        isinstance(obj, slice)
+        and obj.start == 0
+        and obj.stop == line
+        and obj.step is None
+    )
+
+
+def get_callable_name(obj):
+    # typical case has name
+    if hasattr(obj, "__name__"):
+        return obj.__name__
+    # some objects don't; could recurse
+    if isinstance(obj, partial):
+        return get_callable_name(obj.func)
+    # fall back to class name
+    if callable(obj):
+        return type(obj).__name__
+    # everything failed (probably because the argument
+    # wasn't actually callable); we return None
+    # instead of the empty string in this case to allow
+    # distinguishing between no name and a name of ''
+    return None
+
+
+def apply_if_callable(maybe_callable, obj, **kwargs):
+    """
+    Evaluate possibly callable input using obj and kwargs if it is callable,
+    otherwise return as it is.
+
+    Parameters
+    ----------
+    maybe_callable : possibly a callable
+    obj : NDFrame
+    **kwargs
+    """
+    if callable(maybe_callable):
+        return maybe_callable(obj, **kwargs)
+
+    return maybe_callable
+
+
+def standardize_mapping(into):
+    """
+    Helper function to standardize a supplied mapping.
+
+    Parameters
+    ----------
+    into : instance or subclass of collections.abc.Mapping
+        Must be a class, an initialized collections.defaultdict,
+        or an instance of a collections.abc.Mapping subclass.
+
+    Returns
+    -------
+    mapping : a collections.abc.Mapping subclass or other constructor
+        a callable object that can accept an iterator to create
+        the desired Mapping.
+
+    See Also
+    --------
+    DataFrame.to_dict
+    Series.to_dict
+    """
+    if not inspect.isclass(into):
+        if isinstance(into, defaultdict):
+            return partial(defaultdict, into.default_factory)
+        into = type(into)
+    if not issubclass(into, abc.Mapping):
+        raise TypeError(f"unsupported type: {into}")
+    if into == defaultdict:
+        raise TypeError("to_dict() only accepts initialized defaultdicts")
+    return into
+
+
+@overload
+def random_state(state: np.random.Generator) -> np.random.Generator: ...
+
+
+@overload
+def random_state(
+    state: int | np.ndarray | np.random.BitGenerator | np.random.RandomState | None,
+) -> np.random.RandomState: ...
+
+
+def random_state(state: RandomState | None = None):
+    """
+    Helper function for processing random_state arguments.
+
+    Parameters
+    ----------
+    state : int, array-like, BitGenerator, Generator, np.random.RandomState, None.
+        If receives an int, array-like, or BitGenerator, passes to
+        np.random.RandomState() as seed.
+        If receives an np.random RandomState or Generator, just returns that unchanged.
+        If receives `None`, returns np.random.
+        If receives anything else, raises an informative ValueError.
+
+        Default None.
+
+    Returns
+    -------
+    np.random.RandomState or np.random.Generator. If state is None, returns np.random
+
+    """
+    if is_integer(state) or isinstance(state, (np.ndarray, np.random.BitGenerator)):
+        return np.random.RandomState(state)
+    elif isinstance(state, np.random.RandomState):
+        return state
+    elif isinstance(state, np.random.Generator):
+        return state
+    elif state is None:
+        return np.random
+    else:
+        raise ValueError(
+            "random_state must be an integer, array-like, a BitGenerator, Generator, "
+            "a numpy RandomState, or None"
+        )
+
+
+_T = TypeVar("_T")  # Secondary TypeVar for use in pipe's type hints
+
+
+@overload
+def pipe(
+    obj: _T,
+    func: Callable[Concatenate[_T, P], T],
+    *args: P.args,
+    **kwargs: P.kwargs,
+) -> T: ...
+
+
+@overload
+def pipe(
+    obj: Any,
+    func: tuple[Callable[..., T], str],
+    *args: Any,
+    **kwargs: Any,
+) -> T: ...
+
+
+def pipe(
+    obj: _T,
+    func: Callable[Concatenate[_T, P], T] | tuple[Callable[..., T], str],
+    *args: Any,
+    **kwargs: Any,
+) -> T:
+    """
+    Apply a function ``func`` to object ``obj`` either by passing obj as the
+    first argument to the function or, in the case that the func is a tuple,
+    interpret the first element of the tuple as a function and pass the obj to
+    that function as a keyword argument whose key is the value of the second
+    element of the tuple.
+
+    Parameters
+    ----------
+    func : callable or tuple of (callable, str)
+        Function to apply to this object or, alternatively, a
+        ``(callable, data_keyword)`` tuple where ``data_keyword`` is a
+        string indicating the keyword of ``callable`` that expects the
+        object.
+    *args : iterable, optional
+        Positional arguments passed into ``func``.
+    **kwargs : dict, optional
+        A dictionary of keyword arguments passed into ``func``.
+
+    Returns
+    -------
+    object : the return type of ``func``.
+    """
+    if isinstance(func, tuple):
+        # Assigning to func_ so pyright understands that it's a callable
+        func_, target = func
+        if target in kwargs:
+            msg = f"{target} is both the pipe target and a keyword argument"
+            raise ValueError(msg)
+        kwargs[target] = obj
+        return func_(*args, **kwargs)
+    else:
+        return func(obj, *args, **kwargs)
+
+
+def get_rename_function(mapper):
+    """
+    Returns a function that will map names/labels, dependent if mapper
+    is a dict, Series or just a function.
+    """
+
+    def f(x):
+        if x in mapper:
+            return mapper[x]
+        else:
+            return x
+
+    return f if isinstance(mapper, (abc.Mapping, ABCSeries)) else mapper
+
+
+def convert_to_list_like(
+    values: Hashable | Iterable | AnyArrayLike,
+) -> list | AnyArrayLike:
+    """
+    Convert list-like or scalar input to list-like. List, numpy and pandas array-like
+    inputs are returned unmodified whereas others are converted to list.
+    """
+    if isinstance(values, (list, np.ndarray, ABCIndex, ABCSeries, ABCExtensionArray)):
+        return values
+    elif isinstance(values, abc.Iterable) and not isinstance(values, str):
+        return list(values)
+
+    return [values]
+
+
+@contextlib.contextmanager
+def temp_setattr(obj, attr: str, value, condition: bool = True) -> Generator[None]:
+    """
+    Temporarily set attribute on an object.
+
+    Parameters
+    ----------
+    obj : object
+        Object whose attribute will be modified.
+    attr : str
+        Attribute to modify.
+    value : Any
+        Value to temporarily set attribute to.
+    condition : bool, default True
+        Whether to set the attribute. Provided in order to not have to
+        conditionally use this context manager.
+
+    Yields
+    ------
+    object : obj with modified attribute.
+    """
+    if condition:
+        old_value = getattr(obj, attr)
+        setattr(obj, attr, value)
+    try:
+        yield obj
+    finally:
+        if condition:
+            setattr(obj, attr, old_value)
+
+
+def require_length_match(data, index: Index) -> None:
+    """
+    Check the length of data matches the length of the index.
+    """
+    if len(data) != len(index):
+        raise ValueError(
+            "Length of values "
+            f"({len(data)}) "
+            "does not match length of index "
+            f"({len(index)})"
+        )
+
+
+_cython_table = {
+    builtins.sum: "sum",
+    builtins.max: "max",
+    builtins.min: "min",
+    np.all: "all",
+    np.any: "any",
+    np.sum: "sum",
+    np.nansum: "sum",
+    np.mean: "mean",
+    np.nanmean: "mean",
+    np.prod: "prod",
+    np.nanprod: "prod",
+    np.std: "std",
+    np.nanstd: "std",
+    np.var: "var",
+    np.nanvar: "var",
+    np.median: "median",
+    np.nanmedian: "median",
+    np.max: "max",
+    np.nanmax: "max",
+    np.min: "min",
+    np.nanmin: "min",
+    np.cumprod: "cumprod",
+    np.nancumprod: "cumprod",
+    np.cumsum: "cumsum",
+    np.nancumsum: "cumsum",
+}
+
+
+def get_cython_func(arg: Callable) -> str | None:
+    """
+    if we define an internal function for this argument, return it
+    """
+    return _cython_table.get(arg)
+
+
+def fill_missing_names(names: Sequence[Hashable | None]) -> list[Hashable]:
+    """
+    If a name is missing then replace it by level_n, where n is the count
+
+    .. versionadded:: 1.4.0
+
+    Parameters
+    ----------
+    names : list-like
+        list of column names or None values.
+
+    Returns
+    -------
+    list
+        list of column names with the None values replaced.
+    """
+    return [f"level_{i}" if name is None else name for i, name in enumerate(names)]
diff --git a/pandas/pandas/core/computation/__init__.py b/pandas/pandas/core/computation/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/computation/align.py b/pandas/pandas/core/computation/align.py
new file mode 100644
index 0000000000000..6158c4f4d0539
--- /dev/null
+++ b/pandas/pandas/core/computation/align.py
@@ -0,0 +1,227 @@
+"""
+Core eval alignment algorithms.
+"""
+
+from __future__ import annotations
+
+from functools import (
+    partial,
+    wraps,
+)
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._config.config import get_option
+
+from pandas.errors import PerformanceWarning
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+from pandas.core.base import PandasObject
+import pandas.core.common as com
+from pandas.core.computation.common import result_type_many
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+
+    from pandas._typing import F
+
+    from pandas.core.generic import NDFrame
+    from pandas.core.indexes.api import Index
+
+
+def _align_core_single_unary_op(
+    term,
+) -> tuple[partial | type[NDFrame], dict[str, Index] | None]:
+    typ: partial | type[NDFrame]
+    axes: dict[str, Index] | None = None
+
+    if isinstance(term.value, np.ndarray):
+        typ = partial(np.asanyarray, dtype=term.value.dtype)
+    else:
+        typ = type(term.value)
+        if hasattr(term.value, "axes"):
+            axes = _zip_axes_from_type(typ, term.value.axes)
+
+    return typ, axes
+
+
+def _zip_axes_from_type(
+    typ: type[NDFrame], new_axes: Sequence[Index]
+) -> dict[str, Index]:
+    return {name: new_axes[i] for i, name in enumerate(typ._AXIS_ORDERS)}
+
+
+def _any_pandas_objects(terms) -> bool:
+    """
+    Check a sequence of terms for instances of PandasObject.
+    """
+    return any(isinstance(term.value, PandasObject) for term in terms)
+
+
+def _filter_special_cases(f) -> Callable[[F], F]:
+    @wraps(f)
+    def wrapper(terms):
+        # single unary operand
+        if len(terms) == 1:
+            return _align_core_single_unary_op(terms[0])
+
+        term_values = (term.value for term in terms)
+
+        # we don't have any pandas objects
+        if not _any_pandas_objects(terms):
+            return result_type_many(*term_values), None
+
+        return f(terms)
+
+    return wrapper
+
+
+@_filter_special_cases
+def _align_core(terms):
+    term_index = [i for i, term in enumerate(terms) if hasattr(term.value, "axes")]
+    term_dims = [terms[i].value.ndim for i in term_index]
+
+    from pandas import Series
+
+    ndims = Series(dict(zip(term_index, term_dims)))
+
+    # initial axes are the axes of the largest-axis'd term
+    biggest = terms[ndims.idxmax()].value
+    typ = biggest._constructor
+    axes = biggest.axes
+    naxes = len(axes)
+    gt_than_one_axis = naxes > 1
+
+    for value in (terms[i].value for i in term_index):
+        is_series = isinstance(value, ABCSeries)
+        is_series_and_gt_one_axis = is_series and gt_than_one_axis
+
+        for axis, items in enumerate(value.axes):
+            if is_series_and_gt_one_axis:
+                ax, itm = naxes - 1, value.index
+            else:
+                ax, itm = axis, items
+
+            if not axes[ax].is_(itm):
+                axes[ax] = axes[ax].union(itm)
+
+    for i, ndim in ndims.items():
+        for axis, items in zip(range(ndim), axes):
+            ti = terms[i].value
+
+            if hasattr(ti, "reindex"):
+                transpose = isinstance(ti, ABCSeries) and naxes > 1
+                reindexer = axes[naxes - 1] if transpose else items
+
+                term_axis_size = len(ti.axes[axis])
+                reindexer_size = len(reindexer)
+
+                ordm = np.log10(max(1, abs(reindexer_size - term_axis_size)))
+                if (
+                    get_option("performance_warnings")
+                    and ordm >= 1
+                    and reindexer_size >= 10000
+                ):
+                    w = (
+                        f"Alignment difference on axis {axis} is larger "
+                        f"than an order of magnitude on term {terms[i].name!r}, "
+                        f"by more than {ordm:.4g}; performance may suffer."
+                    )
+                    warnings.warn(
+                        w, category=PerformanceWarning, stacklevel=find_stack_level()
+                    )
+
+                obj = ti.reindex(reindexer, axis=axis)
+                terms[i].update(obj)
+
+        terms[i].update(terms[i].value.values)
+
+    return typ, _zip_axes_from_type(typ, axes)
+
+
+def align_terms(terms):
+    """
+    Align a set of terms.
+    """
+    try:
+        # flatten the parse tree (a nested list, really)
+        terms = list(com.flatten(terms))
+    except TypeError:
+        # can't iterate so it must just be a constant or single variable
+        if isinstance(terms.value, (ABCSeries, ABCDataFrame)):
+            typ = type(terms.value)
+            name = terms.value.name if isinstance(terms.value, ABCSeries) else None
+            return typ, _zip_axes_from_type(typ, terms.value.axes), name
+        return np.result_type(terms.type), None, None
+
+    # if all resolved variables are numeric scalars
+    if all(term.is_scalar for term in terms):
+        return result_type_many(*(term.value for term in terms)).type, None, None
+
+    # if all input series have a common name, propagate it to the returned series
+    names = {term.value.name for term in terms if isinstance(term.value, ABCSeries)}
+    name = names.pop() if len(names) == 1 else None
+
+    # perform the main alignment
+    typ, axes = _align_core(terms)
+    return typ, axes, name
+
+
+def reconstruct_object(typ, obj, axes, dtype, name):
+    """
+    Reconstruct an object given its type, raw value, and possibly empty
+    (None) axes.
+
+    Parameters
+    ----------
+    typ : object
+        A type
+    obj : object
+        The value to use in the type constructor
+    axes : dict
+        The axes to use to construct the resulting pandas object
+
+    Returns
+    -------
+    ret : typ
+        An object of type ``typ`` with the value `obj` and possible axes
+        `axes`.
+    """
+    try:
+        typ = typ.type
+    except AttributeError:
+        pass
+
+    res_t = np.result_type(obj.dtype, dtype)
+
+    if not isinstance(typ, partial) and issubclass(typ, PandasObject):
+        if name is None:
+            return typ(obj, dtype=res_t, **axes)
+        return typ(obj, dtype=res_t, name=name, **axes)
+
+    # special case for pathological things like ~True/~False
+    if hasattr(res_t, "type") and typ == np.bool_ and res_t != np.bool_:
+        ret_value = res_t.type(obj)
+    else:
+        ret_value = res_t.type(obj)
+        # The condition is to distinguish 0-dim array (returned in case of
+        # scalar) and 1 element array
+        # e.g. np.array(0) and np.array([0])
+        if (
+            len(obj.shape) == 1
+            and len(obj) == 1
+            and not isinstance(ret_value, np.ndarray)
+        ):
+            ret_value = np.array([ret_value]).astype(res_t)
+
+    return ret_value
diff --git a/pandas/pandas/core/computation/api.py b/pandas/pandas/core/computation/api.py
new file mode 100644
index 0000000000000..bd3be5b3f8c42
--- /dev/null
+++ b/pandas/pandas/core/computation/api.py
@@ -0,0 +1,2 @@
+__all__ = ["eval"]
+from pandas.core.computation.eval import eval
diff --git a/pandas/pandas/core/computation/check.py b/pandas/pandas/core/computation/check.py
new file mode 100644
index 0000000000000..caccf34f81111
--- /dev/null
+++ b/pandas/pandas/core/computation/check.py
@@ -0,0 +1,8 @@
+from __future__ import annotations
+
+from pandas.compat._optional import import_optional_dependency
+
+ne = import_optional_dependency("numexpr", errors="warn")
+NUMEXPR_INSTALLED = ne is not None
+
+__all__ = ["NUMEXPR_INSTALLED"]
diff --git a/pandas/pandas/core/computation/common.py b/pandas/pandas/core/computation/common.py
new file mode 100644
index 0000000000000..115191829f044
--- /dev/null
+++ b/pandas/pandas/core/computation/common.py
@@ -0,0 +1,48 @@
+from __future__ import annotations
+
+from functools import reduce
+
+import numpy as np
+
+from pandas._config import get_option
+
+
+def ensure_decoded(s) -> str:
+    """
+    If we have bytes, decode them to unicode.
+    """
+    if isinstance(s, (np.bytes_, bytes)):
+        s = s.decode(get_option("display.encoding"))
+    return s
+
+
+def result_type_many(*arrays_and_dtypes):
+    """
+    Wrapper around numpy.result_type which overcomes the NPY_MAXARGS (32)
+    argument limit.
+    """
+    try:
+        return np.result_type(*arrays_and_dtypes)
+    except ValueError:
+        # we have > NPY_MAXARGS terms in our expression
+        return reduce(np.result_type, arrays_and_dtypes)
+    except TypeError:
+        from pandas.core.dtypes.cast import find_common_type
+        from pandas.core.dtypes.common import is_extension_array_dtype
+
+        arr_and_dtypes = list(arrays_and_dtypes)
+        ea_dtypes, non_ea_dtypes = [], []
+        for arr_or_dtype in arr_and_dtypes:
+            if is_extension_array_dtype(arr_or_dtype):
+                ea_dtypes.append(arr_or_dtype)
+            else:
+                non_ea_dtypes.append(arr_or_dtype)
+
+        if non_ea_dtypes:
+            try:
+                np_dtype = np.result_type(*non_ea_dtypes)
+            except ValueError:
+                np_dtype = reduce(np.result_type, arrays_and_dtypes)
+            return find_common_type(ea_dtypes + [np_dtype])
+
+        return find_common_type(ea_dtypes)
diff --git a/pandas/pandas/core/computation/engines.py b/pandas/pandas/core/computation/engines.py
new file mode 100644
index 0000000000000..d2a181cbb3c36
--- /dev/null
+++ b/pandas/pandas/core/computation/engines.py
@@ -0,0 +1,151 @@
+"""
+Engine classes for :func:`~pandas.eval`
+"""
+
+from __future__ import annotations
+
+import abc
+from typing import TYPE_CHECKING
+
+from pandas.errors import NumExprClobberingError
+
+from pandas.core.computation.align import (
+    align_terms,
+    reconstruct_object,
+)
+from pandas.core.computation.ops import (
+    MATHOPS,
+    REDUCTIONS,
+)
+
+from pandas.io.formats import printing
+
+if TYPE_CHECKING:
+    from pandas.core.computation.expr import Expr
+
+_ne_builtins = frozenset(MATHOPS + REDUCTIONS)
+
+
+def _check_ne_builtin_clash(expr: Expr) -> None:
+    """
+    Attempt to prevent foot-shooting in a helpful way.
+
+    Parameters
+    ----------
+    expr : Expr
+        Terms can contain
+    """
+    names = expr.names
+    overlap = names & _ne_builtins
+
+    if overlap:
+        s = ", ".join([repr(x) for x in overlap])
+        raise NumExprClobberingError(
+            f'Variables in expression "{expr}" overlap with builtins: ({s})'
+        )
+
+
+class AbstractEngine(metaclass=abc.ABCMeta):
+    """Object serving as a base class for all engines."""
+
+    has_neg_frac = False
+
+    def __init__(self, expr) -> None:
+        self.expr = expr
+        self.aligned_axes = None
+        self.result_type = None
+        self.result_name = None
+
+    def convert(self) -> str:
+        """
+        Convert an expression for evaluation.
+
+        Defaults to return the expression as a string.
+        """
+        return printing.pprint_thing(self.expr)
+
+    def evaluate(self) -> object:
+        """
+        Run the engine on the expression.
+
+        This method performs alignment which is necessary no matter what engine
+        is being used, thus its implementation is in the base class.
+
+        Returns
+        -------
+        object
+            The result of the passed expression.
+        """
+        if not self._is_aligned:
+            self.result_type, self.aligned_axes, self.result_name = align_terms(
+                self.expr.terms
+            )
+
+        # make sure no names in resolvers and locals/globals clash
+        res = self._evaluate()
+        return reconstruct_object(
+            self.result_type,
+            res,
+            self.aligned_axes,
+            self.expr.terms.return_type,
+            self.result_name,
+        )
+
+    @property
+    def _is_aligned(self) -> bool:
+        return self.aligned_axes is not None and self.result_type is not None
+
+    @abc.abstractmethod
+    def _evaluate(self):
+        """
+        Return an evaluated expression.
+
+        Parameters
+        ----------
+        env : Scope
+            The local and global environment in which to evaluate an
+            expression.
+
+        Notes
+        -----
+        Must be implemented by subclasses.
+        """
+
+
+class NumExprEngine(AbstractEngine):
+    """NumExpr engine class"""
+
+    has_neg_frac = True
+
+    def _evaluate(self):
+        import numexpr as ne
+
+        # convert the expression to a valid numexpr expression
+        s = self.convert()
+
+        env = self.expr.env
+        scope = env.full_scope
+        _check_ne_builtin_clash(self.expr)
+        return ne.evaluate(s, local_dict=scope)
+
+
+class PythonEngine(AbstractEngine):
+    """
+    Evaluate an expression in Python space.
+
+    Mostly for testing purposes.
+    """
+
+    has_neg_frac = False
+
+    def evaluate(self):
+        return self.expr()
+
+    def _evaluate(self) -> None:
+        pass
+
+
+ENGINES: dict[str, type[AbstractEngine]] = {
+    "numexpr": NumExprEngine,
+    "python": PythonEngine,
+}
diff --git a/pandas/pandas/core/computation/eval.py b/pandas/pandas/core/computation/eval.py
new file mode 100644
index 0000000000000..f8e3200ef2ba0
--- /dev/null
+++ b/pandas/pandas/core/computation/eval.py
@@ -0,0 +1,451 @@
+"""
+Top level ``eval`` module.
+"""
+
+from __future__ import annotations
+
+import tokenize
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import warnings
+
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import validate_bool_kwarg
+
+from pandas.core.dtypes.common import (
+    is_extension_array_dtype,
+    is_string_dtype,
+)
+
+from pandas.core.computation.engines import ENGINES
+from pandas.core.computation.expr import (
+    PARSERS,
+    Expr,
+)
+from pandas.core.computation.parsing import tokenize_string
+from pandas.core.computation.scope import ensure_scope
+from pandas.core.generic import NDFrame
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from pandas.core.computation.ops import BinOp
+
+
+def _check_engine(engine: str | None) -> str:
+    """
+    Make sure a valid engine is passed.
+
+    Parameters
+    ----------
+    engine : str
+        String to validate.
+
+    Raises
+    ------
+    KeyError
+      * If an invalid engine is passed.
+    ImportError
+      * If numexpr was requested but doesn't exist.
+
+    Returns
+    -------
+    str
+        Engine name.
+    """
+    from pandas.core.computation.check import NUMEXPR_INSTALLED
+    from pandas.core.computation.expressions import USE_NUMEXPR
+
+    if engine is None:
+        engine = "numexpr" if USE_NUMEXPR else "python"
+
+    if engine not in ENGINES:
+        valid_engines = list(ENGINES.keys())
+        raise KeyError(
+            f"Invalid engine '{engine}' passed, valid engines are {valid_engines}"
+        )
+
+    # TODO: validate this in a more general way (thinking of future engines
+    # that won't necessarily be import-able)
+    # Could potentially be done on engine instantiation
+    if engine == "numexpr" and not NUMEXPR_INSTALLED:
+        raise ImportError(
+            "'numexpr' is not installed or an unsupported version. Cannot use "
+            "engine='numexpr' for query/eval if 'numexpr' is not installed"
+        )
+
+    return engine
+
+
+def _check_parser(parser: str) -> None:
+    """
+    Make sure a valid parser is passed.
+
+    Parameters
+    ----------
+    parser : str
+
+    Raises
+    ------
+    KeyError
+      * If an invalid parser is passed
+    """
+    if parser not in PARSERS:
+        raise KeyError(
+            f"Invalid parser '{parser}' passed, valid parsers are {PARSERS.keys()}"
+        )
+
+
+def _check_resolvers(resolvers) -> None:
+    if resolvers is not None:
+        for resolver in resolvers:
+            if not hasattr(resolver, "__getitem__"):
+                name = type(resolver).__name__
+                raise TypeError(
+                    f"Resolver of type '{name}' does not "
+                    "implement the __getitem__ method"
+                )
+
+
+def _check_expression(expr) -> None:
+    """
+    Make sure an expression is not an empty string
+
+    Parameters
+    ----------
+    expr : object
+        An object that can be converted to a string
+
+    Raises
+    ------
+    ValueError
+      * If expr is an empty string
+    """
+    if not expr:
+        raise ValueError("expr cannot be an empty string")
+
+
+def _convert_expression(expr) -> str:
+    """
+    Convert an object to an expression.
+
+    This function converts an object to an expression (a unicode string) and
+    checks to make sure it isn't empty after conversion. This is used to
+    convert operators to their string representation for recursive calls to
+    :func:`~pandas.eval`.
+
+    Parameters
+    ----------
+    expr : object
+        The object to be converted to a string.
+
+    Returns
+    -------
+    str
+        The string representation of an object.
+
+    Raises
+    ------
+    ValueError
+      * If the expression is empty.
+    """
+    s = pprint_thing(expr)
+    _check_expression(s)
+    return s
+
+
+def _check_for_locals(expr: str, stack_level: int, parser: str) -> None:
+    at_top_of_stack = stack_level == 0
+    not_pandas_parser = parser != "pandas"
+
+    if not_pandas_parser:
+        msg = "The '@' prefix is only supported by the pandas parser"
+    elif at_top_of_stack:
+        msg = (
+            "The '@' prefix is not allowed in top-level eval calls.\n"
+            "please refer to your variables by name without the '@' prefix."
+        )
+
+    if at_top_of_stack or not_pandas_parser:
+        for toknum, tokval in tokenize_string(expr):
+            if toknum == tokenize.OP and tokval == "@":
+                raise SyntaxError(msg)
+
+
+def eval(
+    expr: str | BinOp,  # we leave BinOp out of the docstr bc it isn't for users
+    parser: str = "pandas",
+    engine: str | None = None,
+    local_dict=None,
+    global_dict=None,
+    resolvers=(),
+    level: int = 0,
+    target=None,
+    inplace: bool = False,
+) -> Any:
+    """
+    Evaluate a Python expression as a string using various backends.
+
+    .. warning::
+
+        This function can run arbitrary code which can make you vulnerable to code
+        injection if you pass user input to this function.
+
+    Parameters
+    ----------
+    expr : str
+        The expression to evaluate. This string cannot contain any Python
+        `statements
+        <https://docs.python.org/3/reference/simple_stmts.html#simple-statements>`__,
+        only Python `expressions
+        <https://docs.python.org/3/reference/simple_stmts.html#expression-statements>`__.
+
+        By default, with the numexpr engine, the following operations are supported:
+
+        - Arithmetic operations: ``+``, ``-``, ``*``, ``/``, ``**``, ``%``
+        - Boolean operations: ``|`` (or), ``&`` (and), and ``~`` (not)
+        - Comparison operators: ``<``, ``<=``, ``==``, ``!=``, ``>=``, ``>``
+
+        Furthermore, the following mathematical functions are supported:
+
+        - Trigonometric: ``sin``, ``cos``, ``tan``, ``arcsin``, ``arccos``, \
+            ``arctan``, ``arctan2``, ``sinh``, ``cosh``, ``tanh``, ``arcsinh``, \
+            ``arccosh`` and ``arctanh``
+        - Logarithms: ``log`` natural, ``log10`` base 10, ``log1p`` log(1+x)
+        - Absolute Value ``abs``
+        - Square root ``sqrt``
+        - Exponential ``exp`` and Exponential minus one ``expm1``
+
+        See the numexpr engine `documentation
+        <https://numexpr.readthedocs.io/en/latest/user_guide.html#supported-functions>`__
+        for further function support details.
+
+        Using the ``'python'`` engine allows the use of native Python operators
+        such as floor division ``//``, in addition to built-in and user-defined
+        Python functions.
+
+        Additionally, the ``'pandas'`` parser allows the use of :keyword:`and`,
+        :keyword:`or`, and :keyword:`not` with the same semantics as the
+        corresponding bitwise operators.
+    parser : {'pandas', 'python'}, default 'pandas'
+        The parser to use to construct the syntax tree from the expression. The
+        default of ``'pandas'`` parses code slightly different than standard
+        Python. Alternatively, you can parse an expression using the
+        ``'python'`` parser to retain strict Python semantics.  See the
+        :ref:`enhancing performance <enhancingperf.eval>` documentation for
+        more details.
+    engine : {'python', 'numexpr'}, default 'numexpr'
+
+        The engine used to evaluate the expression. Supported engines are
+
+        - None : tries to use ``numexpr``, falls back to ``python``
+        - ``'numexpr'`` : This default engine evaluates pandas objects using
+          numexpr for large speed ups in complex expressions with large frames.
+        - ``'python'`` : Performs operations as if you had ``eval``'d in top
+          level python. This engine is generally not that useful.
+
+        More backends may be available in the future.
+    local_dict : dict or None, optional
+        A dictionary of local variables, taken from locals() by default.
+    global_dict : dict or None, optional
+        A dictionary of global variables, taken from globals() by default.
+    resolvers : list of dict-like or None, optional
+        A list of objects implementing the ``__getitem__`` special method that
+        you can use to inject an additional collection of namespaces to use for
+        variable lookup. For example, this is used in the
+        :meth:`~DataFrame.query` method to inject the
+        ``DataFrame.index`` and ``DataFrame.columns``
+        variables that refer to their respective :class:`~pandas.DataFrame`
+        instance attributes.
+    level : int, optional
+        The number of prior stack frames to traverse and add to the current
+        scope. Most users will **not** need to change this parameter.
+    target : object, optional, default None
+        This is the target object for assignment. It is used when there is
+        variable assignment in the expression. If so, then `target` must
+        support item assignment with string keys, and if a copy is being
+        returned, it must also support `.copy()`.
+    inplace : bool, default False
+        If `target` is provided, and the expression mutates `target`, whether
+        to modify `target` inplace. Otherwise, return a copy of `target` with
+        the mutation.
+
+    Returns
+    -------
+    ndarray, numeric scalar, DataFrame, Series, or None
+        The completion value of evaluating the given code or None if ``inplace=True``.
+
+    Raises
+    ------
+    ValueError
+        There are many instances where such an error can be raised:
+
+        - `target=None`, but the expression is multiline.
+        - The expression is multiline, but not all them have item assignment.
+          An example of such an arrangement is this:
+
+          a = b + 1
+          a + 2
+
+          Here, there are expressions on different lines, making it multiline,
+          but the last line has no variable assigned to the output of `a + 2`.
+        - `inplace=True`, but the expression is missing item assignment.
+        - Item assignment is provided, but the `target` does not support
+          string item assignment.
+        - Item assignment is provided and `inplace=False`, but the `target`
+          does not support the `.copy()` method
+
+    See Also
+    --------
+    DataFrame.query : Evaluates a boolean expression to query the columns
+            of a frame.
+    DataFrame.eval : Evaluate a string describing operations on
+            DataFrame columns.
+
+    Notes
+    -----
+    The ``dtype`` of any objects involved in an arithmetic ``%`` operation are
+    recursively cast to ``float64``.
+
+    See the :ref:`enhancing performance <enhancingperf.eval>` documentation for
+    more details.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"animal": ["dog", "pig"], "age": [10, 20]})
+    >>> df
+      animal  age
+    0    dog   10
+    1    pig   20
+
+    We can add a new column using ``pd.eval``:
+
+    >>> pd.eval("double_age = df.age * 2", target=df)
+      animal  age  double_age
+    0    dog   10          20
+    1    pig   20          40
+    """
+    inplace = validate_bool_kwarg(inplace, "inplace")
+
+    exprs: list[str | BinOp]
+    if isinstance(expr, str):
+        _check_expression(expr)
+        exprs = [e.strip() for e in expr.splitlines() if e.strip() != ""]
+    else:
+        # ops.BinOp; for internal compat, not intended to be passed by users
+        exprs = [expr]
+    multi_line = len(exprs) > 1
+
+    if multi_line and target is None:
+        raise ValueError(
+            "multi-line expressions are only valid in the "
+            "context of data, use DataFrame.eval"
+        )
+    engine = _check_engine(engine)
+    _check_parser(parser)
+    _check_resolvers(resolvers)
+
+    ret = None
+    first_expr = True
+    target_modified = False
+
+    for expr in exprs:
+        expr = _convert_expression(expr)
+        _check_for_locals(expr, level, parser)
+
+        # get our (possibly passed-in) scope
+        env = ensure_scope(
+            level + 1,
+            global_dict=global_dict,
+            local_dict=local_dict,
+            resolvers=resolvers,
+            target=target,
+        )
+
+        parsed_expr = Expr(expr, engine=engine, parser=parser, env=env)
+
+        if engine == "numexpr" and (
+            (
+                is_extension_array_dtype(parsed_expr.terms.return_type)
+                and not is_string_dtype(parsed_expr.terms.return_type)
+            )
+            or (
+                getattr(parsed_expr.terms, "operand_types", None) is not None
+                and any(
+                    (is_extension_array_dtype(elem) and not is_string_dtype(elem))
+                    for elem in parsed_expr.terms.operand_types
+                )
+            )
+        ):
+            warnings.warn(
+                "Engine has switched to 'python' because numexpr does not support "
+                "extension array dtypes. Please set your engine to python manually.",
+                RuntimeWarning,
+                stacklevel=find_stack_level(),
+            )
+            engine = "python"
+
+        # construct the engine and evaluate the parsed expression
+        eng = ENGINES[engine]
+        eng_inst = eng(parsed_expr)
+        ret = eng_inst.evaluate()
+
+        if parsed_expr.assigner is None:
+            if multi_line:
+                raise ValueError(
+                    "Multi-line expressions are only valid "
+                    "if all expressions contain an assignment"
+                )
+            if inplace:
+                raise ValueError("Cannot operate inplace if there is no assignment")
+
+        # assign if needed
+        assigner = parsed_expr.assigner
+        if env.target is not None and assigner is not None:
+            target_modified = True
+
+            # if returning a copy, copy only on the first assignment
+            if not inplace and first_expr:
+                try:
+                    target = env.target
+                    if isinstance(target, NDFrame):
+                        target = target.copy(deep=False)
+                    else:
+                        target = target.copy()
+                except AttributeError as err:
+                    raise ValueError("Cannot return a copy of the target") from err
+            else:
+                target = env.target
+
+            # TypeError is most commonly raised (e.g. int, list), but you
+            # get IndexError if you try to do this assignment on np.ndarray.
+            # we will ignore numpy warnings here; e.g. if trying
+            # to use a non-numeric indexer
+            try:
+                if inplace and isinstance(target, NDFrame):
+                    target.loc[:, assigner] = ret
+                else:
+                    target[assigner] = ret  # pyright: ignore[reportIndexIssue]
+            except (TypeError, IndexError) as err:
+                raise ValueError("Cannot assign expression output to target") from err
+
+            if not resolvers:
+                resolvers = ({assigner: ret},)
+            else:
+                # existing resolver needs updated to handle
+                # case of mutating existing column in copy
+                for resolver in resolvers:
+                    if assigner in resolver:
+                        resolver[assigner] = ret
+                        break
+                else:
+                    resolvers += ({assigner: ret},)
+
+            ret = None
+            first_expr = False
+
+    # We want to exclude `inplace=None` as being False.
+    if inplace is False:
+        return target if target_modified else ret
diff --git a/pandas/pandas/core/computation/expr.py b/pandas/pandas/core/computation/expr.py
new file mode 100644
index 0000000000000..14a393b02409c
--- /dev/null
+++ b/pandas/pandas/core/computation/expr.py
@@ -0,0 +1,842 @@
+"""
+:func:`~pandas.eval` parsers.
+"""
+
+from __future__ import annotations
+
+import ast
+from functools import (
+    partial,
+    reduce,
+)
+from keyword import iskeyword
+import tokenize
+from typing import (
+    TYPE_CHECKING,
+    ClassVar,
+    TypeVar,
+)
+
+import numpy as np
+
+from pandas.errors import UndefinedVariableError
+
+from pandas.core.dtypes.common import is_string_dtype
+
+import pandas.core.common as com
+from pandas.core.computation.ops import (
+    ARITH_OPS_SYMS,
+    BOOL_OPS_SYMS,
+    CMP_OPS_SYMS,
+    LOCAL_TAG,
+    MATHOPS,
+    REDUCTIONS,
+    UNARY_OPS_SYMS,
+    BinOp,
+    Constant,
+    FuncNode,
+    Op,
+    Term,
+    UnaryOp,
+    is_term,
+)
+from pandas.core.computation.parsing import (
+    clean_backtick_quoted_toks,
+    tokenize_string,
+)
+from pandas.core.computation.scope import Scope
+
+from pandas.io.formats import printing
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+
+def _rewrite_assign(tok: tuple[int, str]) -> tuple[int, str]:
+    """
+    Rewrite the assignment operator for PyTables expressions that use ``=``
+    as a substitute for ``==``.
+
+    Parameters
+    ----------
+    tok : tuple of int, str
+        ints correspond to the all caps constants in the tokenize module
+
+    Returns
+    -------
+    tuple of int, str
+        Either the input or token or the replacement values
+    """
+    toknum, tokval = tok
+    return toknum, "==" if tokval == "=" else tokval
+
+
+def _replace_booleans(tok: tuple[int, str]) -> tuple[int, str]:
+    """
+    Replace ``&`` with ``and`` and ``|`` with ``or`` so that bitwise
+    precedence is changed to boolean precedence.
+
+    Parameters
+    ----------
+    tok : tuple of int, str
+        ints correspond to the all caps constants in the tokenize module
+
+    Returns
+    -------
+    tuple of int, str
+        Either the input or token or the replacement values
+    """
+    toknum, tokval = tok
+    if toknum == tokenize.OP:
+        if tokval == "&":
+            return tokenize.NAME, "and"
+        elif tokval == "|":
+            return tokenize.NAME, "or"
+        return toknum, tokval
+    return toknum, tokval
+
+
+def _replace_locals(tok: tuple[int, str]) -> tuple[int, str]:
+    """
+    Replace local variables with a syntactically valid name.
+
+    Parameters
+    ----------
+    tok : tuple of int, str
+        ints correspond to the all caps constants in the tokenize module
+
+    Returns
+    -------
+    tuple of int, str
+        Either the input or token or the replacement values
+
+    Notes
+    -----
+    This is somewhat of a hack in that we rewrite a string such as ``'@a'`` as
+    ``'__pd_eval_local_a'`` by telling the tokenizer that ``__pd_eval_local_``
+    is a ``tokenize.OP`` and to replace the ``'@'`` symbol with it.
+    """
+    toknum, tokval = tok
+    if toknum == tokenize.OP and tokval == "@":
+        return tokenize.OP, LOCAL_TAG
+    return toknum, tokval
+
+
+def _compose2(f, g):
+    """
+    Compose 2 callables.
+    """
+    return lambda *args, **kwargs: f(g(*args, **kwargs))
+
+
+def _compose(*funcs):
+    """
+    Compose 2 or more callables.
+    """
+    assert len(funcs) > 1, "At least 2 callables must be passed to compose"
+    return reduce(_compose2, funcs)
+
+
+def _preparse(
+    source: str,
+    f=_compose(
+        _replace_locals, _replace_booleans, _rewrite_assign, clean_backtick_quoted_toks
+    ),
+) -> str:
+    """
+    Compose a collection of tokenization functions.
+
+    Parameters
+    ----------
+    source : str
+        A Python source code string
+    f : callable
+        This takes a tuple of (toknum, tokval) as its argument and returns a
+        tuple with the same structure but possibly different elements. Defaults
+        to the composition of ``_rewrite_assign``, ``_replace_booleans``, and
+        ``_replace_locals``.
+
+    Returns
+    -------
+    str
+        Valid Python source code
+
+    Notes
+    -----
+    The `f` parameter can be any callable that takes *and* returns input of the
+    form ``(toknum, tokval)``, where ``toknum`` is one of the constants from
+    the ``tokenize`` module and ``tokval`` is a string.
+    """
+    assert callable(f), "f must be callable"
+    return tokenize.untokenize(f(x) for x in tokenize_string(source))  # pyright: ignore[reportArgumentType]
+
+
+def _is_type(t):
+    """
+    Factory for a type checking function of type ``t`` or tuple of types.
+    """
+    return lambda x: isinstance(x.value, t)
+
+
+_is_list = _is_type(list)
+_is_str = _is_type(str)
+
+
+# partition all AST nodes
+_all_nodes = frozenset(
+    node
+    for node in (getattr(ast, name) for name in dir(ast))
+    if isinstance(node, type) and issubclass(node, ast.AST)
+)
+
+
+def _filter_nodes(superclass, all_nodes=_all_nodes):
+    """
+    Filter out AST nodes that are subclasses of ``superclass``.
+    """
+    node_names = (node.__name__ for node in all_nodes if issubclass(node, superclass))
+    return frozenset(node_names)
+
+
+_all_node_names = frozenset(x.__name__ for x in _all_nodes)
+_mod_nodes = _filter_nodes(ast.mod)
+_stmt_nodes = _filter_nodes(ast.stmt)
+_expr_nodes = _filter_nodes(ast.expr)
+_expr_context_nodes = _filter_nodes(ast.expr_context)
+_boolop_nodes = _filter_nodes(ast.boolop)
+_operator_nodes = _filter_nodes(ast.operator)
+_unary_op_nodes = _filter_nodes(ast.unaryop)
+_cmp_op_nodes = _filter_nodes(ast.cmpop)
+_comprehension_nodes = _filter_nodes(ast.comprehension)
+_handler_nodes = _filter_nodes(ast.excepthandler)
+_arguments_nodes = _filter_nodes(ast.arguments)
+_keyword_nodes = _filter_nodes(ast.keyword)
+_alias_nodes = _filter_nodes(ast.alias)
+
+
+# nodes that we don't support directly but are needed for parsing
+_hacked_nodes = frozenset(["Assign", "Module", "Expr"])
+
+
+_unsupported_expr_nodes = frozenset(
+    [
+        "Yield",
+        "GeneratorExp",
+        "IfExp",
+        "DictComp",
+        "SetComp",
+        "Repr",
+        "Lambda",
+        "Set",
+        "AST",
+        "Is",
+        "IsNot",
+    ]
+)
+
+# these nodes are low priority or won't ever be supported (e.g., AST)
+_unsupported_nodes = (
+    _stmt_nodes
+    | _mod_nodes
+    | _handler_nodes
+    | _arguments_nodes
+    | _keyword_nodes
+    | _alias_nodes
+    | _expr_context_nodes
+    | _unsupported_expr_nodes
+) - _hacked_nodes
+
+# we're adding a different assignment in some cases to be equality comparison
+# and we don't want `stmt` and friends in their so get only the class whose
+# names are capitalized
+_base_supported_nodes = (_all_node_names - _unsupported_nodes) | _hacked_nodes
+intersection = _unsupported_nodes & _base_supported_nodes
+_msg = f"cannot both support and not support {intersection}"
+assert not intersection, _msg
+
+
+def _node_not_implemented(node_name: str) -> Callable[..., None]:
+    """
+    Return a function that raises a NotImplementedError with a passed node name.
+    """
+
+    def f(self, *args, **kwargs):
+        raise NotImplementedError(f"'{node_name}' nodes are not implemented")
+
+    return f
+
+
+# should be bound by BaseExprVisitor but that creates a circular dependency:
+# _T is used in disallow, but disallow is used to define BaseExprVisitor
+# https://github.com/microsoft/pyright/issues/2315
+_T = TypeVar("_T")
+
+
+def disallow(nodes: set[str]) -> Callable[[type[_T]], type[_T]]:
+    """
+    Decorator to disallow certain nodes from parsing. Raises a
+    NotImplementedError instead.
+
+    Returns
+    -------
+    callable
+    """
+
+    def disallowed(cls: type[_T]) -> type[_T]:
+        # error: "Type[_T]" has no attribute "unsupported_nodes"
+        cls.unsupported_nodes = ()  # type: ignore[attr-defined]
+        for node in nodes:
+            new_method = _node_not_implemented(node)
+            name = f"visit_{node}"
+            # error: "Type[_T]" has no attribute "unsupported_nodes"
+            cls.unsupported_nodes += (name,)  # type: ignore[attr-defined]
+            setattr(cls, name, new_method)
+        return cls
+
+    return disallowed
+
+
+def _op_maker(op_class, op_symbol):
+    """
+    Return a function to create an op class with its symbol already passed.
+
+    Returns
+    -------
+    callable
+    """
+
+    def f(self, node, *args, **kwargs):
+        """
+        Return a partial function with an Op subclass with an operator already passed.
+
+        Returns
+        -------
+        callable
+        """
+        return partial(op_class, op_symbol, *args, **kwargs)
+
+    return f
+
+
+_op_classes = {"binary": BinOp, "unary": UnaryOp}
+
+
+def add_ops(op_classes):
+    """
+    Decorator to add default implementation of ops.
+    """
+
+    def f(cls):
+        for op_attr_name, op_class in op_classes.items():
+            ops = getattr(cls, f"{op_attr_name}_ops")
+            ops_map = getattr(cls, f"{op_attr_name}_op_nodes_map")
+            for op in ops:
+                op_node = ops_map[op]
+                if op_node is not None:
+                    made_op = _op_maker(op_class, op)
+                    setattr(cls, f"visit_{op_node}", made_op)
+        return cls
+
+    return f
+
+
+@disallow(_unsupported_nodes)
+@add_ops(_op_classes)
+class BaseExprVisitor(ast.NodeVisitor):
+    """
+    Custom ast walker. Parsers of other engines should subclass this class
+    if necessary.
+
+    Parameters
+    ----------
+    env : Scope
+    engine : str
+    parser : str
+    preparser : callable
+    """
+
+    const_type: ClassVar[type[Term]] = Constant
+    term_type: ClassVar[type[Term]] = Term
+
+    binary_ops = CMP_OPS_SYMS + BOOL_OPS_SYMS + ARITH_OPS_SYMS
+    binary_op_nodes = (
+        "Gt",
+        "Lt",
+        "GtE",
+        "LtE",
+        "Eq",
+        "NotEq",
+        "In",
+        "NotIn",
+        "BitAnd",
+        "BitOr",
+        "And",
+        "Or",
+        "Add",
+        "Sub",
+        "Mult",
+        "Div",
+        "Pow",
+        "FloorDiv",
+        "Mod",
+    )
+    binary_op_nodes_map = dict(zip(binary_ops, binary_op_nodes))
+
+    unary_ops = UNARY_OPS_SYMS
+    unary_op_nodes = "UAdd", "USub", "Invert", "Not"
+    unary_op_nodes_map = dict(zip(unary_ops, unary_op_nodes))
+
+    rewrite_map = {
+        ast.Eq: ast.In,
+        ast.NotEq: ast.NotIn,
+        ast.In: ast.In,
+        ast.NotIn: ast.NotIn,
+    }
+
+    unsupported_nodes: tuple[str, ...]
+
+    def __init__(self, env, engine, parser, preparser=_preparse) -> None:
+        self.env = env
+        self.engine = engine
+        self.parser = parser
+        self.preparser = preparser
+        self.assigner = None
+
+    def visit(self, node, **kwargs):
+        if isinstance(node, str):
+            clean = self.preparser(node)
+            try:
+                node = ast.fix_missing_locations(ast.parse(clean))
+            except SyntaxError as e:
+                if any(iskeyword(x) for x in clean.split()):
+                    e.msg = "Python keyword not valid identifier in numexpr query"
+                raise e
+
+        method = f"visit_{type(node).__name__}"
+        visitor = getattr(self, method)
+        return visitor(node, **kwargs)
+
+    def visit_Module(self, node, **kwargs):
+        if len(node.body) != 1:
+            raise SyntaxError("only a single expression is allowed")
+        expr = node.body[0]
+        return self.visit(expr, **kwargs)
+
+    def visit_Expr(self, node, **kwargs):
+        return self.visit(node.value, **kwargs)
+
+    def _rewrite_membership_op(self, node, left, right):
+        # the kind of the operator (is actually an instance)
+        op_instance = node.op
+        op_type = type(op_instance)
+
+        # must be two terms and the comparison operator must be ==/!=/in/not in
+        if is_term(left) and is_term(right) and op_type in self.rewrite_map:
+            left_list, right_list = map(_is_list, (left, right))
+            left_str, right_str = map(_is_str, (left, right))
+
+            # if there are any strings or lists in the expression
+            if left_list or right_list or left_str or right_str:
+                op_instance = self.rewrite_map[op_type]()
+
+            # pop the string variable out of locals and replace it with a list
+            # of one string, kind of a hack
+            if right_str:
+                name = self.env.add_tmp([right.value])
+                right = self.term_type(name, self.env)
+
+            if left_str:
+                name = self.env.add_tmp([left.value])
+                left = self.term_type(name, self.env)
+
+        op = self.visit(op_instance)
+        return op, op_instance, left, right
+
+    def _maybe_transform_eq_ne(self, node, left=None, right=None):
+        if left is None:
+            left = self.visit(node.left, side="left")
+        if right is None:
+            right = self.visit(node.right, side="right")
+        op, op_class, left, right = self._rewrite_membership_op(node, left, right)
+        return op, op_class, left, right
+
+    def _maybe_downcast_constants(self, left, right):
+        f32 = np.dtype(np.float32)
+        if (
+            left.is_scalar
+            and hasattr(left, "value")
+            and not right.is_scalar
+            and right.return_type == f32
+        ):
+            # right is a float32 array, left is a scalar
+            name = self.env.add_tmp(np.float32(left.value))
+            left = self.term_type(name, self.env)
+        if (
+            right.is_scalar
+            and hasattr(right, "value")
+            and not left.is_scalar
+            and left.return_type == f32
+        ):
+            # left is a float32 array, right is a scalar
+            name = self.env.add_tmp(np.float32(right.value))
+            right = self.term_type(name, self.env)
+
+        return left, right
+
+    def _maybe_eval(self, binop, eval_in_python):
+        # eval `in` and `not in` (for now) in "partial" python space
+        # things that can be evaluated in "eval" space will be turned into
+        # temporary variables. for example,
+        # [1,2] in a + 2 * b
+        # in that case a + 2 * b will be evaluated using numexpr, and the "in"
+        # call will be evaluated using isin (in python space)
+        return binop.evaluate(
+            self.env, self.engine, self.parser, self.term_type, eval_in_python
+        )
+
+    def _maybe_evaluate_binop(
+        self,
+        op,
+        op_class,
+        lhs,
+        rhs,
+        eval_in_python=("in", "not in"),
+        maybe_eval_in_python=("==", "!=", "<", ">", "<=", ">="),
+    ):
+        res = op(lhs, rhs)
+
+        if res.has_invalid_return_type:
+            raise TypeError(
+                f"unsupported operand type(s) for {res.op}: "
+                f"'{lhs.type}' and '{rhs.type}'"
+            )
+
+        if self.engine != "pytables" and (
+            (res.op in CMP_OPS_SYMS and getattr(lhs, "is_datetime", False))
+            or getattr(rhs, "is_datetime", False)
+        ):
+            # all date ops must be done in python bc numexpr doesn't work
+            # well with NaT
+            return self._maybe_eval(res, self.binary_ops)
+
+        if res.op in eval_in_python:
+            # "in"/"not in" ops are always evaluated in python
+            return self._maybe_eval(res, eval_in_python)
+        elif self.engine != "pytables":
+            if (
+                getattr(lhs, "return_type", None) == object
+                or is_string_dtype(getattr(lhs, "return_type", None))
+                or getattr(rhs, "return_type", None) == object
+                or is_string_dtype(getattr(rhs, "return_type", None))
+            ):
+                # evaluate "==" and "!=" in python if either of our operands
+                # has an object or string return type
+                return self._maybe_eval(res, eval_in_python + maybe_eval_in_python)
+        return res
+
+    def visit_BinOp(self, node, **kwargs):
+        op, op_class, left, right = self._maybe_transform_eq_ne(node)
+        left, right = self._maybe_downcast_constants(left, right)
+        return self._maybe_evaluate_binop(op, op_class, left, right)
+
+    def visit_UnaryOp(self, node, **kwargs):
+        op = self.visit(node.op)
+        operand = self.visit(node.operand)
+        return op(operand)
+
+    def visit_Name(self, node, **kwargs) -> Term:
+        return self.term_type(node.id, self.env, **kwargs)
+
+    # TODO(py314): deprecated since Python 3.8. Remove after Python 3.14 is min
+    def visit_NameConstant(self, node, **kwargs) -> Term:
+        return self.const_type(node.value, self.env)
+
+    # TODO(py314): deprecated since Python 3.8. Remove after Python 3.14 is min
+    def visit_Num(self, node, **kwargs) -> Term:
+        return self.const_type(node.value, self.env)
+
+    def visit_Constant(self, node, **kwargs) -> Term:
+        return self.const_type(node.value, self.env)
+
+    # TODO(py314): deprecated since Python 3.8. Remove after Python 3.14 is min
+    def visit_Str(self, node, **kwargs) -> Term:
+        name = self.env.add_tmp(node.s)
+        return self.term_type(name, self.env)
+
+    def visit_List(self, node, **kwargs) -> Term:
+        name = self.env.add_tmp([self.visit(e)(self.env) for e in node.elts])
+        return self.term_type(name, self.env)
+
+    visit_Tuple = visit_List
+
+    def visit_Index(self, node, **kwargs):
+        """df.index[4]"""
+        return self.visit(node.value)
+
+    def visit_Subscript(self, node, **kwargs) -> Term:
+        from pandas import eval as pd_eval
+
+        value = self.visit(node.value)
+        slobj = self.visit(node.slice)
+        result = pd_eval(
+            slobj, local_dict=self.env, engine=self.engine, parser=self.parser
+        )
+        try:
+            # a Term instance
+            v = value.value[result]
+        except AttributeError:
+            # an Op instance
+            lhs = pd_eval(
+                value, local_dict=self.env, engine=self.engine, parser=self.parser
+            )
+            v = lhs[result]
+        name = self.env.add_tmp(v)
+        return self.term_type(name, env=self.env)
+
+    def visit_Slice(self, node, **kwargs) -> slice:
+        """df.index[slice(4,6)]"""
+        lower = node.lower
+        if lower is not None:
+            lower = self.visit(lower).value
+        upper = node.upper
+        if upper is not None:
+            upper = self.visit(upper).value
+        step = node.step
+        if step is not None:
+            step = self.visit(step).value
+
+        return slice(lower, upper, step)
+
+    def visit_Assign(self, node, **kwargs):
+        """
+        support a single assignment node, like
+
+        c = a + b
+
+        set the assigner at the top level, must be a Name node which
+        might or might not exist in the resolvers
+
+        """
+        if len(node.targets) != 1:
+            raise SyntaxError("can only assign a single expression")
+        if not isinstance(node.targets[0], ast.Name):
+            raise SyntaxError("left hand side of an assignment must be a single name")
+        if self.env.target is None:
+            raise ValueError("cannot assign without a target object")
+
+        try:
+            assigner = self.visit(node.targets[0], **kwargs)
+        except UndefinedVariableError:
+            assigner = node.targets[0].id
+
+        self.assigner = getattr(assigner, "name", assigner)
+        if self.assigner is None:
+            raise SyntaxError(
+                "left hand side of an assignment must be a single resolvable name"
+            )
+
+        return self.visit(node.value, **kwargs)
+
+    def visit_Attribute(self, node, **kwargs):
+        attr = node.attr
+        value = node.value
+
+        ctx = node.ctx
+        if isinstance(ctx, ast.Load):
+            # resolve the value
+            resolved = self.visit(value).value
+            try:
+                v = getattr(resolved, attr)
+                name = self.env.add_tmp(v)
+                return self.term_type(name, self.env)
+            except AttributeError:
+                # something like datetime.datetime where scope is overridden
+                if isinstance(value, ast.Name) and value.id == attr:
+                    return resolved
+                raise
+
+        raise ValueError(f"Invalid Attribute context {type(ctx).__name__}")
+
+    def visit_Call(self, node, side=None, **kwargs):
+        if isinstance(node.func, ast.Attribute) and node.func.attr != "__call__":
+            res = self.visit_Attribute(node.func)
+        elif not isinstance(node.func, ast.Name):
+            raise TypeError("Only named functions are supported")
+        else:
+            try:
+                res = self.visit(node.func)
+            except UndefinedVariableError:
+                # Check if this is a supported function name
+                try:
+                    res = FuncNode(node.func.id)
+                except ValueError:
+                    # Raise original error
+                    raise
+
+        if res is None:
+            # error: "expr" has no attribute "id"
+            raise ValueError(
+                f"Invalid function call {node.func.id}"  # type: ignore[attr-defined]
+            )
+        if hasattr(res, "value"):
+            res = res.value
+
+        if isinstance(res, FuncNode):
+            new_args = [self.visit(arg) for arg in node.args]
+
+            if node.keywords:
+                raise TypeError(
+                    f'Function "{res.name}" does not support keyword arguments'
+                )
+
+            return res(*new_args)
+
+        else:
+            new_args = [self.visit(arg)(self.env) for arg in node.args]
+
+            for key in node.keywords:
+                if not isinstance(key, ast.keyword):
+                    # error: "expr" has no attribute "id"
+                    raise ValueError(
+                        f"keyword error in function call '{node.func.id}'"  # type: ignore[attr-defined]
+                    )
+
+                if key.arg:
+                    kwargs[key.arg] = self.visit(key.value)(self.env)
+
+            name = self.env.add_tmp(res(*new_args, **kwargs))
+            return self.term_type(name=name, env=self.env)
+
+    def translate_In(self, op):
+        return op
+
+    def visit_Compare(self, node, **kwargs):
+        ops = node.ops
+        comps = node.comparators
+
+        # base case: we have something like a CMP b
+        if len(comps) == 1:
+            op = self.translate_In(ops[0])
+            binop = ast.BinOp(op=op, left=node.left, right=comps[0])
+            return self.visit(binop)
+
+        # recursive case: we have a chained comparison, a CMP b CMP c, etc.
+        left = node.left
+        values = []
+        for op, comp in zip(ops, comps):
+            new_node = self.visit(
+                ast.Compare(comparators=[comp], left=left, ops=[self.translate_In(op)])
+            )
+            left = comp
+            values.append(new_node)
+        return self.visit(ast.BoolOp(op=ast.And(), values=values))
+
+    def _try_visit_binop(self, bop):
+        if isinstance(bop, (Op, Term)):
+            return bop
+        return self.visit(bop)
+
+    def visit_BoolOp(self, node, **kwargs):
+        def visitor(x, y):
+            lhs = self._try_visit_binop(x)
+            rhs = self._try_visit_binop(y)
+
+            op, op_class, lhs, rhs = self._maybe_transform_eq_ne(node, lhs, rhs)
+            return self._maybe_evaluate_binop(op, node.op, lhs, rhs)
+
+        operands = node.values
+        return reduce(visitor, operands)
+
+
+_python_not_supported = frozenset(["Dict", "BoolOp", "In", "NotIn"])
+_numexpr_supported_calls = frozenset(REDUCTIONS + MATHOPS)
+
+
+@disallow(
+    (_unsupported_nodes | _python_not_supported)
+    - (_boolop_nodes | frozenset(["BoolOp", "Attribute", "In", "NotIn", "Tuple"]))
+)
+class PandasExprVisitor(BaseExprVisitor):
+    def __init__(
+        self,
+        env,
+        engine,
+        parser,
+        preparser=partial(
+            _preparse,
+            f=_compose(_replace_locals, _replace_booleans, clean_backtick_quoted_toks),
+        ),
+    ) -> None:
+        super().__init__(env, engine, parser, preparser)
+
+
+@disallow(_unsupported_nodes | _python_not_supported | frozenset(["Not"]))
+class PythonExprVisitor(BaseExprVisitor):
+    def __init__(
+        self, env, engine, parser, preparser=lambda source, f=None: source
+    ) -> None:
+        super().__init__(env, engine, parser, preparser=preparser)
+
+
+class Expr:
+    """
+    Object encapsulating an expression.
+
+    Parameters
+    ----------
+    expr : str
+    engine : str, optional, default 'numexpr'
+    parser : str, optional, default 'pandas'
+    env : Scope, optional, default None
+    level : int, optional, default 2
+    """
+
+    env: Scope
+    engine: str
+    parser: str
+
+    def __init__(
+        self,
+        expr,
+        engine: str = "numexpr",
+        parser: str = "pandas",
+        env: Scope | None = None,
+        level: int = 0,
+    ) -> None:
+        self.expr = expr
+        self.env = env or Scope(level=level + 1)
+        self.engine = engine
+        self.parser = parser
+        self._visitor = PARSERS[parser](self.env, self.engine, self.parser)
+        self.terms = self.parse()
+
+    @property
+    def assigner(self):
+        return getattr(self._visitor, "assigner", None)
+
+    def __call__(self):
+        return self.terms(self.env)
+
+    def __repr__(self) -> str:
+        return printing.pprint_thing(self.terms)
+
+    def __len__(self) -> int:
+        return len(self.expr)
+
+    def parse(self):
+        """
+        Parse an expression.
+        """
+        return self._visitor.visit(self.expr)
+
+    @property
+    def names(self):
+        """
+        Get the names in an expression.
+        """
+        if is_term(self.terms):
+            return frozenset([self.terms.name])
+        return frozenset(term.name for term in com.flatten(self.terms))
+
+
+PARSERS = {"python": PythonExprVisitor, "pandas": PandasExprVisitor}
diff --git a/pandas/pandas/core/computation/expressions.py b/pandas/pandas/core/computation/expressions.py
new file mode 100644
index 0000000000000..5a5fad0d83d7a
--- /dev/null
+++ b/pandas/pandas/core/computation/expressions.py
@@ -0,0 +1,290 @@
+"""
+Expressions
+-----------
+
+Offer fast expression evaluation through numexpr
+
+"""
+
+from __future__ import annotations
+
+import operator
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core import roperator
+from pandas.core.computation.check import NUMEXPR_INSTALLED
+
+if NUMEXPR_INSTALLED:
+    import numexpr as ne
+
+if TYPE_CHECKING:
+    from pandas._typing import FuncType
+
+_TEST_MODE: bool | None = None
+_TEST_RESULT: list[bool] = []
+USE_NUMEXPR = NUMEXPR_INSTALLED
+_evaluate: FuncType | None = None
+_where: FuncType | None = None
+
+# the set of dtypes that we will allow pass to numexpr
+_ALLOWED_DTYPES = {
+    "evaluate": {"int64", "int32", "float64", "float32", "bool"},
+    "where": {"int64", "float64", "bool"},
+}
+
+# the minimum prod shape that we will use numexpr
+_MIN_ELEMENTS = 1_000_000
+
+
+def set_use_numexpr(v: bool = True) -> None:
+    # set/unset to use numexpr
+    global USE_NUMEXPR
+    if NUMEXPR_INSTALLED:
+        USE_NUMEXPR = v
+
+    # choose what we are going to do
+    global _evaluate, _where
+
+    _evaluate = _evaluate_numexpr if USE_NUMEXPR else _evaluate_standard
+    _where = _where_numexpr if USE_NUMEXPR else _where_standard
+
+
+def set_numexpr_threads(n=None) -> None:
+    # if we are using numexpr, set the threads to n
+    # otherwise reset
+    if NUMEXPR_INSTALLED and USE_NUMEXPR:
+        if n is None:
+            n = ne.detect_number_of_cores()
+        ne.set_num_threads(n)
+
+
+def _evaluate_standard(op, op_str, left_op, right_op):
+    """
+    Standard evaluation.
+    """
+    if _TEST_MODE:
+        _store_test_result(False)
+    return op(left_op, right_op)
+
+
+def _can_use_numexpr(op, op_str, left_op, right_op, dtype_check) -> bool:
+    """return left_op boolean if we WILL be using numexpr"""
+    if op_str is not None:
+        # required min elements (otherwise we are adding overhead)
+        if left_op.size > _MIN_ELEMENTS:
+            # check for dtype compatibility
+            dtypes: set[str] = set()
+            for o in [left_op, right_op]:
+                # ndarray and Series Case
+                if hasattr(o, "dtype"):
+                    dtypes |= {o.dtype.name}
+
+            # allowed are a superset
+            if not len(dtypes) or _ALLOWED_DTYPES[dtype_check] >= dtypes:
+                return True
+
+    return False
+
+
+def _evaluate_numexpr(op, op_str, left_op, right_op):
+    result = None
+
+    if _can_use_numexpr(op, op_str, left_op, right_op, "evaluate"):
+        is_reversed = op.__name__.strip("_").startswith("r")
+        if is_reversed:
+            # we were originally called by a reversed op method
+            left_op, right_op = right_op, left_op
+
+        left_value = left_op
+        right_value = right_op
+
+        try:
+            result = ne.evaluate(
+                f"left_value {op_str} right_value",
+                local_dict={"left_value": left_value, "right_value": right_value},
+                casting="safe",
+            )
+        except TypeError:
+            # numexpr raises eg for array ** array with integers
+            # (https://github.com/pydata/numexpr/issues/379)
+            pass
+        except NotImplementedError:
+            if _bool_arith_fallback(op_str, left_op, right_op):
+                pass
+            else:
+                raise
+
+        if is_reversed:
+            # reverse order to original for fallback
+            left_op, right_op = right_op, left_op
+
+    if _TEST_MODE:
+        _store_test_result(result is not None)
+
+    if result is None:
+        result = _evaluate_standard(op, op_str, left_op, right_op)
+
+    return result
+
+
+_op_str_mapping = {
+    operator.add: "+",
+    roperator.radd: "+",
+    operator.mul: "*",
+    roperator.rmul: "*",
+    operator.sub: "-",
+    roperator.rsub: "-",
+    operator.truediv: "/",
+    roperator.rtruediv: "/",
+    # floordiv not supported by numexpr 2.x
+    operator.floordiv: None,
+    roperator.rfloordiv: None,
+    # we require Python semantics for mod of negative for backwards compatibility
+    # see https://github.com/pydata/numexpr/issues/365
+    # so sticking with unaccelerated for now GH#36552
+    operator.mod: None,
+    roperator.rmod: None,
+    operator.pow: "**",
+    roperator.rpow: "**",
+    operator.eq: "==",
+    operator.ne: "!=",
+    operator.le: "<=",
+    operator.lt: "<",
+    operator.ge: ">=",
+    operator.gt: ">",
+    operator.and_: "&",
+    roperator.rand_: "&",
+    operator.or_: "|",
+    roperator.ror_: "|",
+    operator.xor: "^",
+    roperator.rxor: "^",
+    divmod: None,
+    roperator.rdivmod: None,
+}
+
+
+def _where_standard(cond, left_op, right_op):
+    # Caller is responsible for extracting ndarray if necessary
+    return np.where(cond, left_op, right_op)
+
+
+def _where_numexpr(cond, left_op, right_op):
+    # Caller is responsible for extracting ndarray if necessary
+    result = None
+
+    if _can_use_numexpr(None, "where", left_op, right_op, "where"):
+        result = ne.evaluate(
+            "where(cond_value, a_value, b_value)",
+            local_dict={"cond_value": cond, "a_value": left_op, "b_value": right_op},
+            casting="safe",
+        )
+
+    if result is None:
+        result = _where_standard(cond, left_op, right_op)
+
+    return result
+
+
+# turn myself on
+set_use_numexpr(get_option("compute.use_numexpr"))
+
+
+def _has_bool_dtype(x):
+    try:
+        return x.dtype == bool
+    except AttributeError:
+        return isinstance(x, (bool, np.bool_))
+
+
+_BOOL_OP_UNSUPPORTED = {"+": "|", "*": "&", "-": "^"}
+
+
+def _bool_arith_fallback(op_str, left_op, right_op) -> bool:
+    """
+    Check if we should fallback to the python `_evaluate_standard` in case
+    of an unsupported operation by numexpr, which is the case for some
+    boolean ops.
+    """
+    if _has_bool_dtype(left_op) and _has_bool_dtype(right_op):
+        if op_str in _BOOL_OP_UNSUPPORTED:
+            warnings.warn(
+                f"evaluating in Python space because the {op_str!r} "
+                "operator is not supported by numexpr for the bool dtype, "
+                f"use {_BOOL_OP_UNSUPPORTED[op_str]!r} instead.",
+                stacklevel=find_stack_level(),
+            )
+            return True
+    return False
+
+
+def evaluate(op, left_op, right_op, use_numexpr: bool = True):
+    """
+    Evaluate and return the expression of the op on left_op and right_op.
+
+    Parameters
+    ----------
+    op : the actual operand
+    left_op : left operand
+    right_op : right operand
+    use_numexpr : bool, default True
+        Whether to try to use numexpr.
+    """
+    op_str = _op_str_mapping[op]
+    if op_str is not None:
+        if use_numexpr:
+            # error: "None" not callable
+            return _evaluate(op, op_str, left_op, right_op)  # type: ignore[misc]
+    return _evaluate_standard(op, op_str, left_op, right_op)
+
+
+def where(cond, left_op, right_op, use_numexpr: bool = True):
+    """
+    Evaluate the where condition cond on left_op and right_op.
+
+    Parameters
+    ----------
+    cond : np.ndarray[bool]
+    left_op : return if cond is True
+    right_op : return if cond is False
+    use_numexpr : bool, default True
+        Whether to try to use numexpr.
+    """
+    assert _where is not None
+    if use_numexpr:
+        return _where(cond, left_op, right_op)
+    else:
+        return _where_standard(cond, left_op, right_op)
+
+
+def set_test_mode(v: bool = True) -> None:
+    """
+    Keeps track of whether numexpr was used.
+
+    Stores an additional ``True`` for every successful use of evaluate with
+    numexpr since the last ``get_test_result``.
+    """
+    global _TEST_MODE, _TEST_RESULT
+    _TEST_MODE = v
+    _TEST_RESULT = []
+
+
+def _store_test_result(used_numexpr: bool) -> None:
+    if used_numexpr:
+        _TEST_RESULT.append(used_numexpr)
+
+
+def get_test_result() -> list[bool]:
+    """
+    Get test result and reset test_results.
+    """
+    global _TEST_RESULT
+    res = _TEST_RESULT
+    _TEST_RESULT = []
+    return res
diff --git a/pandas/pandas/core/computation/ops.py b/pandas/pandas/core/computation/ops.py
new file mode 100644
index 0000000000000..f06ded6d9f98e
--- /dev/null
+++ b/pandas/pandas/core/computation/ops.py
@@ -0,0 +1,561 @@
+"""
+Operator classes for eval.
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+from functools import partial
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import Timestamp
+
+from pandas.core.dtypes.common import (
+    is_list_like,
+    is_scalar,
+)
+
+import pandas.core.common as com
+from pandas.core.computation.common import (
+    ensure_decoded,
+    result_type_many,
+)
+from pandas.core.computation.scope import DEFAULT_GLOBALS
+
+from pandas.io.formats.printing import (
+    pprint_thing,
+    pprint_thing_encoded,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Iterable,
+        Iterator,
+    )
+
+REDUCTIONS = ("sum", "prod", "min", "max")
+
+_unary_math_ops = (
+    "sin",
+    "cos",
+    "tan",
+    "exp",
+    "log",
+    "expm1",
+    "log1p",
+    "sqrt",
+    "sinh",
+    "cosh",
+    "tanh",
+    "arcsin",
+    "arccos",
+    "arctan",
+    "arccosh",
+    "arcsinh",
+    "arctanh",
+    "abs",
+    "log10",
+    "floor",
+    "ceil",
+)
+_binary_math_ops = ("arctan2",)
+
+MATHOPS = _unary_math_ops + _binary_math_ops
+
+
+LOCAL_TAG = "__pd_eval_local_"
+
+
+class Term:
+    def __new__(cls, name, env, side=None, encoding=None):
+        klass = Constant if not isinstance(name, str) else cls
+        supr_new = super(Term, klass).__new__
+        return supr_new(klass)
+
+    is_local: bool
+
+    def __init__(self, name, env, side=None, encoding=None) -> None:
+        # name is a str for Term, but may be something else for subclasses
+        self._name = name
+        self.env = env
+        self.side = side
+        tname = str(name)
+        self.is_local = tname.startswith(LOCAL_TAG) or tname in DEFAULT_GLOBALS
+        self._value = self._resolve_name()
+        self.encoding = encoding
+
+    @property
+    def local_name(self) -> str:
+        return self.name.replace(LOCAL_TAG, "")
+
+    def __repr__(self) -> str:
+        return pprint_thing(self.name)
+
+    def __call__(self, *args, **kwargs):
+        return self.value
+
+    def evaluate(self, *args, **kwargs) -> Term:
+        return self
+
+    def _resolve_name(self):
+        local_name = str(self.local_name)
+        is_local = self.is_local
+        if local_name in self.env.scope and isinstance(
+            self.env.scope[local_name], type
+        ):
+            is_local = False
+
+        res = self.env.resolve(local_name, is_local=is_local)
+        self.update(res)
+
+        if hasattr(res, "ndim") and isinstance(res.ndim, int) and res.ndim > 2:
+            raise NotImplementedError(
+                "N-dimensional objects, where N > 2, are not supported with eval"
+            )
+        return res
+
+    def update(self, value) -> None:
+        """
+        search order for local (i.e., @variable) variables:
+
+        scope, key_variable
+        [('locals', 'local_name'),
+         ('globals', 'local_name'),
+         ('locals', 'key'),
+         ('globals', 'key')]
+        """
+        key = self.name
+
+        # if it's a variable name (otherwise a constant)
+        if isinstance(key, str):
+            self.env.swapkey(self.local_name, key, new_value=value)
+
+        self.value = value
+
+    @property
+    def is_scalar(self) -> bool:
+        return is_scalar(self._value)
+
+    @property
+    def type(self):
+        try:
+            # potentially very slow for large, mixed dtype frames
+            return self._value.values.dtype
+        except AttributeError:
+            try:
+                # ndarray
+                return self._value.dtype
+            except AttributeError:
+                # scalar
+                return type(self._value)
+
+    return_type = type
+
+    @property
+    def raw(self) -> str:
+        return f"{type(self).__name__}(name={self.name!r}, type={self.type})"
+
+    @property
+    def is_datetime(self) -> bool:
+        try:
+            t = self.type.type
+        except AttributeError:
+            t = self.type
+
+        return issubclass(t, (datetime, np.datetime64))
+
+    @property
+    def value(self):
+        return self._value
+
+    @value.setter
+    def value(self, new_value) -> None:
+        self._value = new_value
+
+    @property
+    def name(self):
+        return self._name
+
+    @property
+    def ndim(self) -> int:
+        return self._value.ndim
+
+
+class Constant(Term):
+    def _resolve_name(self):
+        return self._name
+
+    @property
+    def name(self):
+        return self.value
+
+    def __repr__(self) -> str:
+        # in python 2 str() of float
+        # can truncate shorter than repr()
+        return repr(self.name)
+
+
+_bool_op_map = {"not": "~", "and": "&", "or": "|"}
+
+
+class Op:
+    """
+    Hold an operator of arbitrary arity.
+    """
+
+    op: str
+
+    def __init__(self, op: str, operands: Iterable[Term | Op], encoding=None) -> None:
+        self.op = _bool_op_map.get(op, op)
+        self.operands = operands
+        self.encoding = encoding
+
+    def __iter__(self) -> Iterator:
+        return iter(self.operands)
+
+    def __repr__(self) -> str:
+        """
+        Print a generic n-ary operator and its operands using infix notation.
+        """
+        # recurse over the operands
+        parened = (f"({pprint_thing(opr)})" for opr in self.operands)
+        return pprint_thing(f" {self.op} ".join(parened))
+
+    @property
+    def return_type(self):
+        # clobber types to bool if the op is a boolean operator
+        if self.op in (CMP_OPS_SYMS + BOOL_OPS_SYMS):
+            return np.bool_
+        return result_type_many(*(term.type for term in com.flatten(self)))
+
+    @property
+    def has_invalid_return_type(self) -> bool:
+        types = self.operand_types
+        obj_dtype_set = frozenset([np.dtype("object")])
+        return self.return_type == object and types - obj_dtype_set
+
+    @property
+    def operand_types(self):
+        return frozenset(term.type for term in com.flatten(self))
+
+    @property
+    def is_scalar(self) -> bool:
+        return all(operand.is_scalar for operand in self.operands)
+
+    @property
+    def is_datetime(self) -> bool:
+        try:
+            t = self.return_type.type
+        except AttributeError:
+            t = self.return_type
+
+        return issubclass(t, (datetime, np.datetime64))
+
+
+def _in(x, y):
+    """
+    Compute the vectorized membership of ``x in y`` if possible, otherwise
+    use Python.
+    """
+    try:
+        return x.isin(y)
+    except AttributeError:
+        if is_list_like(x):
+            try:
+                return y.isin(x)
+            except AttributeError:
+                pass
+        return x in y
+
+
+def _not_in(x, y):
+    """
+    Compute the vectorized membership of ``x not in y`` if possible,
+    otherwise use Python.
+    """
+    try:
+        return ~x.isin(y)
+    except AttributeError:
+        if is_list_like(x):
+            try:
+                return ~y.isin(x)
+            except AttributeError:
+                pass
+        return x not in y
+
+
+CMP_OPS_SYMS = (">", "<", ">=", "<=", "==", "!=", "in", "not in")
+_cmp_ops_funcs = (
+    operator.gt,
+    operator.lt,
+    operator.ge,
+    operator.le,
+    operator.eq,
+    operator.ne,
+    _in,
+    _not_in,
+)
+_cmp_ops_dict = dict(zip(CMP_OPS_SYMS, _cmp_ops_funcs))
+
+BOOL_OPS_SYMS = ("&", "|", "and", "or")
+_bool_ops_funcs = (operator.and_, operator.or_, operator.and_, operator.or_)
+_bool_ops_dict = dict(zip(BOOL_OPS_SYMS, _bool_ops_funcs))
+
+ARITH_OPS_SYMS = ("+", "-", "*", "/", "**", "//", "%")
+_arith_ops_funcs = (
+    operator.add,
+    operator.sub,
+    operator.mul,
+    operator.truediv,
+    operator.pow,
+    operator.floordiv,
+    operator.mod,
+)
+_arith_ops_dict = dict(zip(ARITH_OPS_SYMS, _arith_ops_funcs))
+
+_binary_ops_dict = {}
+
+for d in (_cmp_ops_dict, _bool_ops_dict, _arith_ops_dict):
+    _binary_ops_dict.update(d)
+
+
+def is_term(obj) -> bool:
+    return isinstance(obj, Term)
+
+
+class BinOp(Op):
+    """
+    Hold a binary operator and its operands.
+
+    Parameters
+    ----------
+    op : str
+    lhs : Term or Op
+    rhs : Term or Op
+    """
+
+    def __init__(self, op: str, lhs, rhs) -> None:
+        super().__init__(op, (lhs, rhs))
+        self.lhs = lhs
+        self.rhs = rhs
+
+        self._disallow_scalar_only_bool_ops()
+
+        self.convert_values()
+
+        try:
+            self.func = _binary_ops_dict[op]
+        except KeyError as err:
+            # has to be made a list for python3
+            keys = list(_binary_ops_dict.keys())
+            raise ValueError(
+                f"Invalid binary operator {op!r}, valid operators are {keys}"
+            ) from err
+
+    def __call__(self, env):
+        """
+        Recursively evaluate an expression in Python space.
+
+        Parameters
+        ----------
+        env : Scope
+
+        Returns
+        -------
+        object
+            The result of an evaluated expression.
+        """
+        # recurse over the left/right nodes
+        left = self.lhs(env)
+        right = self.rhs(env)
+
+        return self.func(left, right)
+
+    def evaluate(self, env, engine: str, parser, term_type, eval_in_python):
+        """
+        Evaluate a binary operation *before* being passed to the engine.
+
+        Parameters
+        ----------
+        env : Scope
+        engine : str
+        parser : str
+        term_type : type
+        eval_in_python : list
+
+        Returns
+        -------
+        term_type
+            The "pre-evaluated" expression as an instance of ``term_type``
+        """
+        if engine == "python":
+            res = self(env)
+        else:
+            # recurse over the left/right nodes
+
+            left = self.lhs.evaluate(
+                env,
+                engine=engine,
+                parser=parser,
+                term_type=term_type,
+                eval_in_python=eval_in_python,
+            )
+
+            right = self.rhs.evaluate(
+                env,
+                engine=engine,
+                parser=parser,
+                term_type=term_type,
+                eval_in_python=eval_in_python,
+            )
+
+            # base cases
+            if self.op in eval_in_python:
+                res = self.func(left.value, right.value)
+            else:
+                from pandas.core.computation.eval import eval
+
+                res = eval(self, local_dict=env, engine=engine, parser=parser)
+
+        name = env.add_tmp(res)
+        return term_type(name, env=env)
+
+    def convert_values(self) -> None:
+        """
+        Convert datetimes to a comparable value in an expression.
+        """
+
+        def stringify(value):
+            encoder: Callable
+            if self.encoding is not None:
+                encoder = partial(pprint_thing_encoded, encoding=self.encoding)
+            else:
+                encoder = pprint_thing
+            return encoder(value)
+
+        lhs, rhs = self.lhs, self.rhs
+
+        if is_term(lhs) and lhs.is_datetime and is_term(rhs) and rhs.is_scalar:
+            v = rhs.value
+            if isinstance(v, (int, float)):
+                v = stringify(v)
+            v = Timestamp(ensure_decoded(v))
+            if v.tz is not None:
+                v = v.tz_convert("UTC")
+            self.rhs.update(v)
+
+        if is_term(rhs) and rhs.is_datetime and is_term(lhs) and lhs.is_scalar:
+            v = lhs.value
+            if isinstance(v, (int, float)):
+                v = stringify(v)
+            v = Timestamp(ensure_decoded(v))
+            if v.tz is not None:
+                v = v.tz_convert("UTC")
+            self.lhs.update(v)
+
+    def _disallow_scalar_only_bool_ops(self) -> None:
+        rhs = self.rhs
+        lhs = self.lhs
+
+        # GH#24883 unwrap dtype if necessary to ensure we have a type object
+        rhs_rt = rhs.return_type
+        rhs_rt = getattr(rhs_rt, "type", rhs_rt)
+        lhs_rt = lhs.return_type
+        lhs_rt = getattr(lhs_rt, "type", lhs_rt)
+        if (
+            (lhs.is_scalar or rhs.is_scalar)
+            and self.op in _bool_ops_dict
+            and (
+                not (
+                    issubclass(rhs_rt, (bool, np.bool_))
+                    and issubclass(lhs_rt, (bool, np.bool_))
+                )
+            )
+        ):
+            raise NotImplementedError("cannot evaluate scalar only bool ops")
+
+
+UNARY_OPS_SYMS = ("+", "-", "~", "not")
+_unary_ops_funcs = (operator.pos, operator.neg, operator.invert, operator.invert)
+_unary_ops_dict = dict(zip(UNARY_OPS_SYMS, _unary_ops_funcs))
+
+
+class UnaryOp(Op):
+    """
+    Hold a unary operator and its operands.
+
+    Parameters
+    ----------
+    op : str
+        The token used to represent the operator.
+    operand : Term or Op
+        The Term or Op operand to the operator.
+
+    Raises
+    ------
+    ValueError
+        * If no function associated with the passed operator token is found.
+    """
+
+    def __init__(self, op: Literal["+", "-", "~", "not"], operand) -> None:
+        super().__init__(op, (operand,))
+        self.operand = operand
+
+        try:
+            self.func = _unary_ops_dict[op]
+        except KeyError as err:
+            raise ValueError(
+                f"Invalid unary operator {op!r}, valid operators are {UNARY_OPS_SYMS}"
+            ) from err
+
+    def __call__(self, env) -> MathCall:
+        operand = self.operand(env)
+        # error: Cannot call function of unknown type
+        return self.func(operand)  # type: ignore[operator]
+
+    def __repr__(self) -> str:
+        return pprint_thing(f"{self.op}({self.operand})")
+
+    @property
+    def return_type(self) -> np.dtype:
+        operand = self.operand
+        if operand.return_type == np.dtype("bool"):
+            return np.dtype("bool")
+        if isinstance(operand, Op) and (
+            operand.op in _cmp_ops_dict or operand.op in _bool_ops_dict
+        ):
+            return np.dtype("bool")
+        return np.dtype("int")
+
+
+class MathCall(Op):
+    def __init__(self, func, args) -> None:
+        super().__init__(func.name, args)
+        self.func = func
+
+    def __call__(self, env):
+        # error: "Op" not callable
+        operands = [op(env) for op in self.operands]  # type: ignore[operator]
+        return self.func.func(*operands)
+
+    def __repr__(self) -> str:
+        operands = map(str, self.operands)
+        return pprint_thing(f"{self.op}({','.join(operands)})")
+
+
+class FuncNode:
+    def __init__(self, name: str) -> None:
+        if name not in MATHOPS:
+            raise ValueError(f'"{name}" is not a supported function')
+        self.name = name
+        self.func = getattr(np, name)
+
+    def __call__(self, *args) -> MathCall:
+        return MathCall(self, args)
diff --git a/pandas/pandas/core/computation/parsing.py b/pandas/pandas/core/computation/parsing.py
new file mode 100644
index 0000000000000..8441941797a6e
--- /dev/null
+++ b/pandas/pandas/core/computation/parsing.py
@@ -0,0 +1,252 @@
+"""
+:func:`~pandas.eval` source string parsing functions
+"""
+
+from __future__ import annotations
+
+from enum import Enum
+from io import StringIO
+from keyword import iskeyword
+import token
+import tokenize
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterator,
+    )
+
+# A token value Python's tokenizer probably will never use.
+BACKTICK_QUOTED_STRING = 100
+
+
+def create_valid_python_identifier(name: str) -> str:
+    """
+    Create valid Python identifiers from any string.
+
+    Check if name contains any special characters. If it contains any
+    special characters, the special characters will be replaced by
+    a special string and a prefix is added.
+
+    Raises
+    ------
+    SyntaxError
+        If the returned name is not a Python valid identifier, raise an exception.
+    """
+    if name.isidentifier() and not iskeyword(name):
+        return name
+
+    # Escape characters that fall outside the ASCII range (U+0001..U+007F).
+    # GH 49633
+    gen = (
+        (c, "".join(chr(b) for b in c.encode("ascii", "backslashreplace")))
+        for c in name
+    )
+    name = "".join(
+        c_escaped.replace("\\", "_UNICODE_" if c != c_escaped else "_BACKSLASH_")
+        for c, c_escaped in gen
+    )
+
+    # Create a dict with the special characters and their replacement string.
+    # EXACT_TOKEN_TYPES contains these special characters
+    # token.tok_name contains a readable description of the replacement string.
+    special_characters_replacements = {
+        char: f"_{token.tok_name[tokval]}_"
+        for char, tokval in (tokenize.EXACT_TOKEN_TYPES.items())
+    }
+    special_characters_replacements.update(
+        {
+            " ": "_",
+            "?": "_QUESTIONMARK_",
+            "!": "_EXCLAMATIONMARK_",
+            "$": "_DOLLARSIGN_",
+            "€": "_EUROSIGN_",
+            "°": "_DEGREESIGN_",
+            "'": "_SINGLEQUOTE_",
+            '"': "_DOUBLEQUOTE_",
+            "#": "_HASH_",
+            "`": "_BACKTICK_",
+        }
+    )
+
+    name = "".join([special_characters_replacements.get(char, char) for char in name])
+    name = f"BACKTICK_QUOTED_STRING_{name}"
+
+    if not name.isidentifier():
+        raise SyntaxError(f"Could not convert '{name}' to a valid Python identifier.")
+
+    return name
+
+
+def clean_backtick_quoted_toks(tok: tuple[int, str]) -> tuple[int, str]:
+    """
+    Clean up a column name if surrounded by backticks.
+
+    Backtick quoted string are indicated by a certain tokval value. If a string
+    is a backtick quoted token it will processed by
+    :func:`_create_valid_python_identifier` so that the parser can find this
+    string when the query is executed.
+    In this case the tok will get the NAME tokval.
+
+    Parameters
+    ----------
+    tok : tuple of int, str
+        ints correspond to the all caps constants in the tokenize module
+
+    Returns
+    -------
+    tok : Tuple[int, str]
+        Either the input or token or the replacement values
+    """
+    toknum, tokval = tok
+    if toknum == BACKTICK_QUOTED_STRING:
+        return tokenize.NAME, create_valid_python_identifier(tokval)
+    return toknum, tokval
+
+
+def clean_column_name(name: Hashable) -> Hashable:
+    """
+    Function to emulate the cleaning of a backtick quoted name.
+
+    The purpose for this function is to see what happens to the name of
+    identifier if it goes to the process of being parsed a Python code
+    inside a backtick quoted string and than being cleaned
+    (removed of any special characters).
+
+    Parameters
+    ----------
+    name : hashable
+        Name to be cleaned.
+
+    Returns
+    -------
+    name : hashable
+        Returns the name after tokenizing and cleaning.
+    """
+    try:
+        # Escape backticks
+        name = name.replace("`", "``") if isinstance(name, str) else name
+
+        tokenized = tokenize_string(f"`{name}`")
+        tokval = next(tokenized)[1]
+        return create_valid_python_identifier(tokval)
+    except SyntaxError:
+        return name
+
+
+class ParseState(Enum):
+    DEFAULT = 0
+    IN_BACKTICK = 1
+    IN_SINGLE_QUOTE = 2
+    IN_DOUBLE_QUOTE = 3
+
+
+def _split_by_backtick(s: str) -> list[tuple[bool, str]]:
+    """
+    Splits a str into substrings along backtick characters (`).
+
+    Disregards backticks inside quotes.
+
+    Parameters
+    ----------
+    s : str
+        The Python source code string.
+
+    Returns
+    -------
+    substrings: list[tuple[bool, str]]
+        List of tuples, where each tuple has two elements:
+        The first is a boolean indicating if the substring is backtick-quoted.
+        The second is the actual substring.
+    """
+    substrings = []
+    substr: list[str] = []  # Will join into a string before adding to `substrings`
+    i = 0
+    parse_state = ParseState.DEFAULT
+    while i < len(s):
+        char = s[i]
+
+        match char:
+            case "`":
+                # start of a backtick-quoted string
+                if parse_state == ParseState.DEFAULT:
+                    if substr:
+                        substrings.append((False, "".join(substr)))
+
+                    substr = [char]
+                    i += 1
+                    parse_state = ParseState.IN_BACKTICK
+                    continue
+
+                elif parse_state == ParseState.IN_BACKTICK:
+                    # escaped backtick inside a backtick-quoted string
+                    next_char = s[i + 1] if (i != len(s) - 1) else None
+                    if next_char == "`":
+                        substr.append(char)
+                        substr.append(next_char)
+                        i += 2
+                        continue
+
+                    # end of the backtick-quoted string
+                    else:
+                        substr.append(char)
+                        substrings.append((True, "".join(substr)))
+
+                        substr = []
+                        i += 1
+                        parse_state = ParseState.DEFAULT
+                        continue
+            case "'":
+                # start of a single-quoted string
+                if parse_state == ParseState.DEFAULT:
+                    parse_state = ParseState.IN_SINGLE_QUOTE
+                # end of a single-quoted string
+                elif (parse_state == ParseState.IN_SINGLE_QUOTE) and (s[i - 1] != "\\"):
+                    parse_state = ParseState.DEFAULT
+            case '"':
+                # start of a double-quoted string
+                if parse_state == ParseState.DEFAULT:
+                    parse_state = ParseState.IN_DOUBLE_QUOTE
+                # end of a double-quoted string
+                elif (parse_state == ParseState.IN_DOUBLE_QUOTE) and (s[i - 1] != "\\"):
+                    parse_state = ParseState.DEFAULT
+        substr.append(char)
+        i += 1
+
+    if substr:
+        substrings.append((False, "".join(substr)))
+
+    return substrings
+
+
+def tokenize_string(source: str) -> Iterator[tuple[int, str]]:
+    """
+    Tokenize a Python source code string.
+
+    Parameters
+    ----------
+    source : str
+        The Python source code string.
+
+    Returns
+    -------
+    tok_generator : Iterator[Tuple[int, str]]
+        An iterator yielding all tokens with only toknum and tokval (Tuple[ing, str]).
+    """
+    # GH 59285
+    # Escape characters, including backticks
+    source = "".join(
+        (
+            create_valid_python_identifier(substring[1:-1])
+            if is_backtick_quoted
+            else substring
+        )
+        for is_backtick_quoted, substring in _split_by_backtick(source)
+    )
+
+    line_reader = StringIO(source).readline
+    token_generator = tokenize.generate_tokens(line_reader)
+
+    for toknum, tokval, _, _, _ in token_generator:
+        yield toknum, tokval
diff --git a/pandas/pandas/core/computation/pytables.py b/pandas/pandas/core/computation/pytables.py
new file mode 100644
index 0000000000000..166c9d47294cd
--- /dev/null
+++ b/pandas/pandas/core/computation/pytables.py
@@ -0,0 +1,668 @@
+"""manage PyTables query interface via Expressions"""
+
+from __future__ import annotations
+
+import ast
+from decimal import (
+    Decimal,
+    InvalidOperation,
+)
+from functools import partial
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    ClassVar,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import (
+    Timedelta,
+    Timestamp,
+)
+from pandas.errors import UndefinedVariableError
+
+from pandas.core.dtypes.common import is_list_like
+
+import pandas.core.common as com
+from pandas.core.computation import (
+    expr,
+    ops,
+    scope as _scope,
+)
+from pandas.core.computation.common import ensure_decoded
+from pandas.core.computation.expr import BaseExprVisitor
+from pandas.core.computation.ops import is_term
+from pandas.core.construction import extract_array
+from pandas.core.indexes.base import Index
+
+from pandas.io.formats.printing import (
+    pprint_thing,
+    pprint_thing_encoded,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        Self,
+        npt,
+    )
+
+
+class PyTablesScope(_scope.Scope):
+    __slots__ = ("queryables",)
+
+    queryables: dict[str, Any]
+
+    def __init__(
+        self,
+        level: int,
+        global_dict=None,
+        local_dict=None,
+        queryables: dict[str, Any] | None = None,
+    ) -> None:
+        super().__init__(level + 1, global_dict=global_dict, local_dict=local_dict)
+        self.queryables = queryables or {}
+
+
+class Term(ops.Term):
+    env: PyTablesScope
+
+    def __new__(cls, name, env, side=None, encoding=None):
+        if isinstance(name, str):
+            klass = cls
+        else:
+            klass = Constant
+        return object.__new__(klass)
+
+    def __init__(self, name, env: PyTablesScope, side=None, encoding=None) -> None:
+        super().__init__(name, env, side=side, encoding=encoding)
+
+    def _resolve_name(self):
+        # must be a queryables
+        if self.side == "left":
+            # Note: The behavior of __new__ ensures that self.name is a str here
+            if self.name not in self.env.queryables:
+                raise NameError(f"name {self.name!r} is not defined")
+            return self.name
+
+        # resolve the rhs (and allow it to be None)
+        try:
+            return self.env.resolve(self.name, is_local=False)
+        except UndefinedVariableError:
+            return self.name
+
+    # read-only property overwriting read/write property
+    @property  # type: ignore[misc]
+    def value(self):
+        return self._value
+
+
+class Constant(Term):
+    def __init__(self, name, env: PyTablesScope, side=None, encoding=None) -> None:
+        assert isinstance(env, PyTablesScope), type(env)
+        super().__init__(name, env, side=side, encoding=encoding)
+
+    def _resolve_name(self):
+        return self._name
+
+
+class BinOp(ops.BinOp):
+    _max_selectors = 31
+
+    op: str
+    queryables: dict[str, Any]
+    condition: str | None
+
+    def __init__(self, op: str, lhs, rhs, queryables: dict[str, Any], encoding) -> None:
+        super().__init__(op, lhs, rhs)
+        self.queryables = queryables
+        self.encoding = encoding
+        self.condition = None
+
+    def _disallow_scalar_only_bool_ops(self) -> None:
+        pass
+
+    def prune(self, klass):
+        def pr(left, right):
+            """create and return a new specialized BinOp from myself"""
+            if left is None:
+                return right
+            elif right is None:
+                return left
+
+            k = klass
+            if isinstance(left, ConditionBinOp):
+                if isinstance(right, ConditionBinOp):
+                    k = JointConditionBinOp
+                elif isinstance(left, k):
+                    return left
+                elif isinstance(right, k):
+                    return right
+
+            elif isinstance(left, FilterBinOp):
+                if isinstance(right, FilterBinOp):
+                    k = JointFilterBinOp
+                elif isinstance(left, k):
+                    return left
+                elif isinstance(right, k):
+                    return right
+
+            return k(
+                self.op, left, right, queryables=self.queryables, encoding=self.encoding
+            ).evaluate()
+
+        left, right = self.lhs, self.rhs
+
+        if is_term(left) and is_term(right):
+            res = pr(left.value, right.value)
+        elif not is_term(left) and is_term(right):
+            res = pr(left.prune(klass), right.value)
+        elif is_term(left) and not is_term(right):
+            res = pr(left.value, right.prune(klass))
+        elif not (is_term(left) or is_term(right)):
+            res = pr(left.prune(klass), right.prune(klass))
+
+        return res
+
+    def conform(self, rhs):
+        """inplace conform rhs"""
+        if not is_list_like(rhs):
+            rhs = [rhs]
+        if isinstance(rhs, np.ndarray):
+            rhs = rhs.ravel()
+        return rhs
+
+    @property
+    def is_valid(self) -> bool:
+        """return True if this is a valid field"""
+        return self.lhs in self.queryables
+
+    @property
+    def is_in_table(self) -> bool:
+        """
+        return True if this is a valid column name for generation (e.g. an
+        actual column in the table)
+        """
+        return self.queryables.get(self.lhs) is not None
+
+    @property
+    def kind(self):
+        """the kind of my field"""
+        return getattr(self.queryables.get(self.lhs), "kind", None)
+
+    @property
+    def meta(self):
+        """the meta of my field"""
+        return getattr(self.queryables.get(self.lhs), "meta", None)
+
+    @property
+    def metadata(self):
+        """the metadata of my field"""
+        return getattr(self.queryables.get(self.lhs), "metadata", None)
+
+    def generate(self, v) -> str:
+        """create and return the op string for this TermValue"""
+        val = v.tostring(self.encoding)
+        return f"({self.lhs} {self.op} {val})"
+
+    def convert_value(self, conv_val) -> TermValue:
+        """
+        convert the expression that is in the term to something that is
+        accepted by pytables
+        """
+
+        def stringify(value):
+            if self.encoding is not None:
+                return pprint_thing_encoded(value, encoding=self.encoding)
+            return pprint_thing(value)
+
+        kind = ensure_decoded(self.kind)
+        meta = ensure_decoded(self.meta)
+        if kind == "datetime" or (kind and kind.startswith("datetime64")):
+            if isinstance(conv_val, (int, float)):
+                conv_val = stringify(conv_val)
+            conv_val = ensure_decoded(conv_val)
+            conv_val = Timestamp(conv_val).as_unit("ns")
+            if conv_val.tz is not None:
+                conv_val = conv_val.tz_convert("UTC")
+            return TermValue(conv_val, conv_val._value, kind)
+        elif kind in ("timedelta64", "timedelta"):
+            if isinstance(conv_val, str):
+                conv_val = Timedelta(conv_val)
+            else:
+                conv_val = Timedelta(conv_val, unit="s")
+            conv_val = conv_val.as_unit("ns")._value
+            return TermValue(int(conv_val), conv_val, kind)
+        elif meta == "category":
+            metadata = extract_array(self.metadata, extract_numpy=True)
+            result: npt.NDArray[np.intp] | np.intp | int
+            if conv_val not in metadata:
+                result = -1
+            else:
+                result = metadata.searchsorted(conv_val, side="left")
+            return TermValue(result, result, "integer")
+        elif kind == "integer":
+            try:
+                v_dec = Decimal(conv_val)
+            except InvalidOperation:
+                # GH 54186
+                # convert v to float to raise float's ValueError
+                float(conv_val)
+            else:
+                conv_val = int(v_dec.to_integral_exact(rounding="ROUND_HALF_EVEN"))
+            return TermValue(conv_val, conv_val, kind)
+        elif kind == "float":
+            conv_val = float(conv_val)
+            return TermValue(conv_val, conv_val, kind)
+        elif kind == "bool":
+            if isinstance(conv_val, str):
+                conv_val = conv_val.strip().lower() not in [
+                    "false",
+                    "f",
+                    "no",
+                    "n",
+                    "none",
+                    "0",
+                    "[]",
+                    "{}",
+                    "",
+                ]
+            else:
+                conv_val = bool(conv_val)
+            return TermValue(conv_val, conv_val, kind)
+        elif isinstance(conv_val, str):
+            # string quoting
+            return TermValue(conv_val, stringify(conv_val), "string")
+        else:
+            raise TypeError(
+                f"Cannot compare {conv_val} of type {type(conv_val)} to {kind} column"
+            )
+
+    def convert_values(self) -> None:
+        pass
+
+
+class FilterBinOp(BinOp):
+    filter: tuple[Any, Any, Index] | None = None
+
+    def __repr__(self) -> str:
+        if self.filter is None:
+            return "Filter: Not Initialized"
+        return pprint_thing(f"[Filter : [{self.filter[0]}] -> [{self.filter[1]}]")
+
+    def invert(self) -> Self:
+        """invert the filter"""
+        if self.filter is not None:
+            self.filter = (
+                self.filter[0],
+                self.generate_filter_op(invert=True),
+                self.filter[2],
+            )
+        return self
+
+    def format(self):
+        """return the actual filter format"""
+        return [self.filter]
+
+    # error: Signature of "evaluate" incompatible with supertype "BinOp"
+    def evaluate(self) -> Self | None:  # type: ignore[override]
+        if not self.is_valid:
+            raise ValueError(f"query term is not valid [{self}]")
+
+        rhs = self.conform(self.rhs)
+        values = list(rhs)
+
+        if self.is_in_table:
+            # if too many values to create the expression, use a filter instead
+            if self.op in ["==", "!="] and len(values) > self._max_selectors:
+                filter_op = self.generate_filter_op()
+                self.filter = (self.lhs, filter_op, Index(values))
+
+                return self
+            return None
+
+        # equality conditions
+        if self.op in ["==", "!="]:
+            filter_op = self.generate_filter_op()
+            self.filter = (self.lhs, filter_op, Index(values))
+
+        else:
+            raise TypeError(
+                f"passing a filterable condition to a non-table indexer [{self}]"
+            )
+
+        return self
+
+    def generate_filter_op(self, invert: bool = False):
+        if (self.op == "!=" and not invert) or (self.op == "==" and invert):
+            return lambda axis, vals: ~axis.isin(vals)
+        else:
+            return lambda axis, vals: axis.isin(vals)
+
+
+class JointFilterBinOp(FilterBinOp):
+    def format(self):
+        raise NotImplementedError("unable to collapse Joint Filters")
+
+    # error: Signature of "evaluate" incompatible with supertype "BinOp"
+    def evaluate(self) -> Self:  # type: ignore[override]
+        return self
+
+
+class ConditionBinOp(BinOp):
+    def __repr__(self) -> str:
+        return pprint_thing(f"[Condition : [{self.condition}]]")
+
+    def invert(self):
+        """invert the condition"""
+        # if self.condition is not None:
+        #    self.condition = "~(%s)" % self.condition
+        # return self
+        raise NotImplementedError(
+            "cannot use an invert condition when passing to numexpr"
+        )
+
+    def format(self):
+        """return the actual ne format"""
+        return self.condition
+
+    # error: Signature of "evaluate" incompatible with supertype "BinOp"
+    def evaluate(self) -> Self | None:  # type: ignore[override]
+        if not self.is_valid:
+            raise ValueError(f"query term is not valid [{self}]")
+
+        # convert values if we are in the table
+        if not self.is_in_table:
+            return None
+
+        rhs = self.conform(self.rhs)
+        values = [self.convert_value(v) for v in rhs]
+
+        # equality conditions
+        if self.op in ["==", "!="]:
+            # too many values to create the expression?
+            if len(values) <= self._max_selectors:
+                vs = [self.generate(v) for v in values]
+                self.condition = f"({' | '.join(vs)})"
+
+            # use a filter after reading
+            else:
+                return None
+        else:
+            self.condition = self.generate(values[0])
+
+        return self
+
+
+class JointConditionBinOp(ConditionBinOp):
+    # error: Signature of "evaluate" incompatible with supertype "BinOp"
+    def evaluate(self) -> Self:  # type: ignore[override]
+        self.condition = f"({self.lhs.condition} {self.op} {self.rhs.condition})"
+        return self
+
+
+class UnaryOp(ops.UnaryOp):
+    def prune(self, klass):
+        if self.op != "~":
+            raise NotImplementedError("UnaryOp only support invert type ops")
+
+        operand = self.operand
+        operand = operand.prune(klass)
+
+        if operand is not None and (
+            (issubclass(klass, ConditionBinOp) and operand.condition is not None)
+            or (
+                not issubclass(klass, ConditionBinOp)
+                and issubclass(klass, FilterBinOp)
+                and operand.filter is not None
+            )
+        ):
+            return operand.invert()
+        return None
+
+
+class PyTablesExprVisitor(BaseExprVisitor):
+    const_type: ClassVar[type[ops.Term]] = Constant
+    term_type: ClassVar[type[Term]] = Term
+
+    def __init__(self, env, engine, parser, **kwargs) -> None:
+        super().__init__(env, engine, parser)
+        for bin_op in self.binary_ops:
+            bin_node = self.binary_op_nodes_map[bin_op]
+            setattr(
+                self,
+                f"visit_{bin_node}",
+                lambda node, bin_op=bin_op: partial(BinOp, bin_op, **kwargs),
+            )
+
+    def visit_UnaryOp(self, node, **kwargs) -> ops.Term | UnaryOp | None:
+        if isinstance(node.op, (ast.Not, ast.Invert)):
+            return UnaryOp("~", self.visit(node.operand))
+        elif isinstance(node.op, ast.USub):
+            return self.const_type(-self.visit(node.operand).value, self.env)
+        elif isinstance(node.op, ast.UAdd):
+            raise NotImplementedError("Unary addition not supported")
+        # TODO: return None might never be reached
+        return None
+
+    def visit_Index(self, node, **kwargs):
+        return self.visit(node.value).value
+
+    def visit_Assign(self, node, **kwargs):
+        cmpr = ast.Compare(
+            ops=[ast.Eq()], left=node.targets[0], comparators=[node.value]
+        )
+        return self.visit(cmpr)
+
+    def visit_Subscript(self, node, **kwargs) -> ops.Term:
+        # only allow simple subscripts
+
+        value = self.visit(node.value)
+        slobj = self.visit(node.slice)
+        try:
+            value = value.value
+        except AttributeError:
+            pass
+
+        if isinstance(slobj, Term):
+            # In py39 np.ndarray lookups with Term containing int raise
+            slobj = slobj.value
+
+        try:
+            return self.const_type(value[slobj], self.env)
+        except TypeError as err:
+            raise ValueError(f"cannot subscript {value!r} with {slobj!r}") from err
+
+    def visit_Attribute(self, node, **kwargs):
+        attr = node.attr
+        value = node.value
+
+        ctx = type(node.ctx)
+        if ctx == ast.Load:
+            # resolve the value
+            resolved = self.visit(value)
+
+            # try to get the value to see if we are another expression
+            try:
+                resolved = resolved.value
+            except AttributeError:
+                pass
+
+            try:
+                return self.term_type(getattr(resolved, attr), self.env)
+            except AttributeError:
+                # something like datetime.datetime where scope is overridden
+                if isinstance(value, ast.Name) and value.id == attr:
+                    return resolved
+
+        raise ValueError(f"Invalid Attribute context {ctx.__name__}")
+
+    def translate_In(self, op):
+        return ast.Eq() if isinstance(op, ast.In) else op
+
+    def _rewrite_membership_op(self, node, left, right):
+        return self.visit(node.op), node.op, left, right
+
+
+def _validate_where(w):
+    """
+    Validate that the where statement is of the right type.
+
+    The type may either be String, Expr, or list-like of Exprs.
+
+    Parameters
+    ----------
+    w : String term expression, Expr, or list-like of Exprs.
+
+    Returns
+    -------
+    where : The original where clause if the check was successful.
+
+    Raises
+    ------
+    TypeError : An invalid data type was passed in for w (e.g. dict).
+    """
+    if not (isinstance(w, (PyTablesExpr, str)) or is_list_like(w)):
+        raise TypeError(
+            "where must be passed as a string, PyTablesExpr, "
+            "or list-like of PyTablesExpr"
+        )
+
+    return w
+
+
+class PyTablesExpr(expr.Expr):
+    """
+    Hold a pytables-like expression, comprised of possibly multiple 'terms'.
+
+    Parameters
+    ----------
+    where : string term expression, PyTablesExpr, or list-like of PyTablesExprs
+    queryables : a "kinds" map (dict of column name -> kind), or None if column
+        is non-indexable
+    encoding : an encoding that will encode the query terms
+
+    Returns
+    -------
+    a PyTablesExpr object
+
+    Examples
+    --------
+    'index>=date'
+    "columns=['A', 'D']"
+    'columns=A'
+    'columns==A'
+    "~(columns=['A','B'])"
+    'index>df.index[3] & string="bar"'
+    '(index>df.index[3] & index<=df.index[6]) | string="bar"'
+    "ts>=Timestamp('2012-02-01')"
+    "major_axis>=20130101"
+    """
+
+    _visitor: PyTablesExprVisitor | None
+    env: PyTablesScope
+    expr: str
+
+    def __init__(
+        self,
+        where,
+        queryables: dict[str, Any] | None = None,
+        encoding=None,
+        scope_level: int = 0,
+    ) -> None:
+        where = _validate_where(where)
+
+        self.encoding = encoding
+        self.condition = None
+        self.filter = None
+        self.terms = None
+        self._visitor = None
+
+        # capture the environment if needed
+        local_dict: _scope.DeepChainMap[Any, Any] | None = None
+
+        if isinstance(where, PyTablesExpr):
+            local_dict = where.env.scope
+            _where = where.expr
+
+        elif is_list_like(where):
+            where = list(where)
+            for idx, w in enumerate(where):
+                if isinstance(w, PyTablesExpr):
+                    local_dict = w.env.scope
+                else:
+                    where[idx] = _validate_where(w)
+            _where = " & ".join([f"({w})" for w in com.flatten(where)])
+        else:
+            # _validate_where ensures we otherwise have a string
+            _where = where
+
+        self.expr = _where
+        self.env = PyTablesScope(scope_level + 1, local_dict=local_dict)
+
+        if queryables is not None and isinstance(self.expr, str):
+            self.env.queryables.update(queryables)
+            self._visitor = PyTablesExprVisitor(
+                self.env,
+                queryables=queryables,
+                parser="pytables",
+                engine="pytables",
+                encoding=encoding,
+            )
+            self.terms = self.parse()
+
+    def __repr__(self) -> str:
+        if self.terms is not None:
+            return pprint_thing(self.terms)
+        return pprint_thing(self.expr)
+
+    def evaluate(self):
+        """create and return the numexpr condition and filter"""
+        try:
+            self.condition = self.terms.prune(ConditionBinOp)
+        except AttributeError as err:
+            raise ValueError(
+                f"cannot process expression [{self.expr}], [{self}] "
+                "is not a valid condition"
+            ) from err
+        try:
+            self.filter = self.terms.prune(FilterBinOp)
+        except AttributeError as err:
+            raise ValueError(
+                f"cannot process expression [{self.expr}], [{self}] "
+                "is not a valid filter"
+            ) from err
+
+        return self.condition, self.filter
+
+
+class TermValue:
+    """hold a term value the we use to construct a condition/filter"""
+
+    def __init__(self, value, converted, kind: str) -> None:
+        assert isinstance(kind, str), kind
+        self.value = value
+        self.converted = converted
+        self.kind = kind
+
+    def tostring(self, encoding) -> str:
+        """quote the string if not encoded else encode and return"""
+        if self.kind == "string":
+            if encoding is not None:
+                return str(self.converted)
+            return f'"{self.converted}"'
+        elif self.kind == "float":
+            # python 2 str(float) is not always
+            # round-trippable so use repr()
+            return repr(self.converted)
+        return str(self.converted)
+
+
+def maybe_expression(s) -> bool:
+    """loose checking if s is a pytables-acceptable expression"""
+    if not isinstance(s, str):
+        return False
+    operations = PyTablesExprVisitor.binary_ops + PyTablesExprVisitor.unary_ops + ("=",)
+
+    # make sure we have an op at least
+    return any(op in s for op in operations)
diff --git a/pandas/pandas/core/computation/scope.py b/pandas/pandas/core/computation/scope.py
new file mode 100644
index 0000000000000..336d62b9d9579
--- /dev/null
+++ b/pandas/pandas/core/computation/scope.py
@@ -0,0 +1,356 @@
+"""
+Module for scope operations
+"""
+
+from __future__ import annotations
+
+from collections import ChainMap
+import datetime
+import inspect
+from io import StringIO
+import itertools
+import pprint
+import struct
+import sys
+from typing import TypeVar
+
+import numpy as np
+
+from pandas._libs.tslibs import Timestamp
+from pandas.errors import UndefinedVariableError
+
+_KT = TypeVar("_KT")
+_VT = TypeVar("_VT")
+
+
+# https://docs.python.org/3/library/collections.html#chainmap-examples-and-recipes
+class DeepChainMap(ChainMap[_KT, _VT]):
+    """
+    Variant of ChainMap that allows direct updates to inner scopes.
+
+    Only works when all passed mapping are mutable.
+    """
+
+    def __setitem__(self, key: _KT, value: _VT) -> None:
+        for mapping in self.maps:
+            if key in mapping:
+                mapping[key] = value
+                return
+        self.maps[0][key] = value
+
+    def __delitem__(self, key: _KT) -> None:
+        """
+        Raises
+        ------
+        KeyError
+            If `key` doesn't exist.
+        """
+        for mapping in self.maps:
+            if key in mapping:
+                del mapping[key]
+                return
+        raise KeyError(key)
+
+
+def ensure_scope(
+    level: int, global_dict=None, local_dict=None, resolvers=(), target=None
+) -> Scope:
+    """Ensure that we are grabbing the correct scope."""
+    return Scope(
+        level + 1,
+        global_dict=global_dict,
+        local_dict=local_dict,
+        resolvers=resolvers,
+        target=target,
+    )
+
+
+def _replacer(x) -> str:
+    """
+    Replace a number with its hexadecimal representation. Used to tag
+    temporary variables with their calling scope's id.
+    """
+    # get the hex repr of the binary char and remove 0x and pad by pad_size
+    # zeros
+    try:
+        hexin = ord(x)
+    except TypeError:
+        # bytes literals masquerade as ints when iterating in py3
+        hexin = x
+
+    return hex(hexin)
+
+
+def _raw_hex_id(obj) -> str:
+    """Return the padded hexadecimal id of ``obj``."""
+    # interpret as a pointer since that's what really what id returns
+    packed = struct.pack("@P", id(obj))
+    return "".join([_replacer(x) for x in packed])
+
+
+DEFAULT_GLOBALS = {
+    "Timestamp": Timestamp,
+    "datetime": datetime.datetime,
+    "True": True,
+    "False": False,
+    "list": list,
+    "tuple": tuple,
+    "inf": np.inf,
+    "Inf": np.inf,
+}
+
+
+def _get_pretty_string(obj) -> str:
+    """
+    Return a prettier version of obj.
+
+    Parameters
+    ----------
+    obj : object
+        Object to pretty print
+
+    Returns
+    -------
+    str
+        Pretty print object repr
+    """
+    sio = StringIO()
+    pprint.pprint(obj, stream=sio)
+    return sio.getvalue()
+
+
+class Scope:
+    """
+    Object to hold scope, with a few bells to deal with some custom syntax
+    and contexts added by pandas.
+
+    Parameters
+    ----------
+    level : int
+    global_dict : dict or None, optional, default None
+    local_dict : dict or Scope or None, optional, default None
+    resolvers : list-like or None, optional, default None
+    target : object
+
+    Attributes
+    ----------
+    level : int
+    scope : DeepChainMap
+    target : object
+    temps : dict
+    """
+
+    __slots__ = ["level", "resolvers", "scope", "target", "temps"]
+    level: int
+    scope: DeepChainMap
+    resolvers: DeepChainMap
+    temps: dict
+
+    def __init__(
+        self, level: int, global_dict=None, local_dict=None, resolvers=(), target=None
+    ) -> None:
+        self.level = level + 1
+
+        # shallow copy because we don't want to keep filling this up with what
+        # was there before if there are multiple calls to Scope/_ensure_scope
+        self.scope = DeepChainMap(DEFAULT_GLOBALS.copy())
+        self.target = target
+
+        if isinstance(local_dict, Scope):
+            self.scope.update(local_dict.scope)
+            if local_dict.target is not None:
+                self.target = local_dict.target
+            self._update(local_dict.level)
+
+        frame = sys._getframe(self.level)
+
+        try:
+            # shallow copy here because we don't want to replace what's in
+            # scope when we align terms (alignment accesses the underlying
+            # numpy array of pandas objects)
+            scope_global = self.scope.new_child(
+                (global_dict if global_dict is not None else frame.f_globals).copy()
+            )
+            self.scope = DeepChainMap(scope_global)
+            if not isinstance(local_dict, Scope):
+                scope_local = self.scope.new_child(
+                    (local_dict if local_dict is not None else frame.f_locals).copy()
+                )
+                self.scope = DeepChainMap(scope_local)
+        finally:
+            del frame
+
+        # assumes that resolvers are going from outermost scope to inner
+        if isinstance(local_dict, Scope):
+            resolvers += tuple(local_dict.resolvers.maps)
+        self.resolvers = DeepChainMap(*resolvers)
+        self.temps = {}
+
+    def __repr__(self) -> str:
+        scope_keys = _get_pretty_string(list(self.scope.keys()))
+        res_keys = _get_pretty_string(list(self.resolvers.keys()))
+        return f"{type(self).__name__}(scope={scope_keys}, resolvers={res_keys})"
+
+    @property
+    def has_resolvers(self) -> bool:
+        """
+        Return whether we have any extra scope.
+
+        For example, DataFrames pass Their columns as resolvers during calls to
+        ``DataFrame.eval()`` and ``DataFrame.query()``.
+
+        Returns
+        -------
+        hr : bool
+        """
+        return bool(len(self.resolvers))
+
+    def resolve(self, key: str, is_local: bool):
+        """
+        Resolve a variable name in a possibly local context.
+
+        Parameters
+        ----------
+        key : str
+            A variable name
+        is_local : bool
+            Flag indicating whether the variable is local or not (prefixed with
+            the '@' symbol)
+
+        Returns
+        -------
+        value : object
+            The value of a particular variable
+        """
+        try:
+            # only look for locals in outer scope
+            if is_local:
+                return self.scope[key]
+
+            # not a local variable so check in resolvers if we have them
+            if self.has_resolvers:
+                return self.resolvers[key]
+
+            # if we're here that means that we have no locals and we also have
+            # no resolvers
+            assert not is_local and not self.has_resolvers
+            return self.scope[key]
+        except KeyError:
+            try:
+                # last ditch effort we look in temporaries
+                # these are created when parsing indexing expressions
+                # e.g., df[df > 0]
+                return self.temps[key]
+            except KeyError as err:
+                raise UndefinedVariableError(key, is_local) from err
+
+    def swapkey(self, old_key: str, new_key: str, new_value=None) -> None:
+        """
+        Replace a variable name, with a potentially new value.
+
+        Parameters
+        ----------
+        old_key : str
+            Current variable name to replace
+        new_key : str
+            New variable name to replace `old_key` with
+        new_value : object
+            Value to be replaced along with the possible renaming
+        """
+        if self.has_resolvers:
+            maps = self.resolvers.maps + self.scope.maps
+        else:
+            maps = self.scope.maps
+
+        maps.append(self.temps)
+
+        for mapping in maps:
+            if old_key in mapping:
+                mapping[new_key] = new_value
+                return
+
+    def _get_vars(self, stack, scopes: list[str]) -> None:
+        """
+        Get specifically scoped variables from a list of stack frames.
+
+        Parameters
+        ----------
+        stack : list
+            A list of stack frames as returned by ``inspect.stack()``
+        scopes : sequence of strings
+            A sequence containing valid stack frame attribute names that
+            evaluate to a dictionary. For example, ('locals', 'globals')
+        """
+        variables = itertools.product(scopes, stack)
+        for scope, (frame, _, _, _, _, _) in variables:
+            try:
+                d = getattr(frame, f"f_{scope}")
+                self.scope = DeepChainMap(self.scope.new_child(d))
+            finally:
+                # won't remove it, but DECREF it
+                # in Py3 this probably isn't necessary since frame won't be
+                # scope after the loop
+                del frame
+
+    def _update(self, level: int) -> None:
+        """
+        Update the current scope by going back `level` levels.
+
+        Parameters
+        ----------
+        level : int
+        """
+        sl = level + 1
+
+        # add sl frames to the scope starting with the
+        # most distant and overwriting with more current
+        # makes sure that we can capture variable scope
+        stack = inspect.stack()
+
+        try:
+            self._get_vars(stack[:sl], scopes=["locals"])
+        finally:
+            del stack[:], stack
+
+    def add_tmp(self, value) -> str:
+        """
+        Add a temporary variable to the scope.
+
+        Parameters
+        ----------
+        value : object
+            An arbitrary object to be assigned to a temporary variable.
+
+        Returns
+        -------
+        str
+            The name of the temporary variable created.
+        """
+        name = f"{type(value).__name__}_{self.ntemps}_{_raw_hex_id(self)}"
+
+        # add to inner most scope
+        assert name not in self.temps
+        self.temps[name] = value
+        assert name in self.temps
+
+        # only increment if the variable gets put in the scope
+        return name
+
+    @property
+    def ntemps(self) -> int:
+        """The number of temporary variables in this scope"""
+        return len(self.temps)
+
+    @property
+    def full_scope(self) -> DeepChainMap:
+        """
+        Return the full scope for use with passing to engines transparently
+        as a mapping.
+
+        Returns
+        -------
+        vars : DeepChainMap
+            All variables in this scope.
+        """
+        maps = [self.temps] + self.resolvers.maps + self.scope.maps
+        return DeepChainMap(*maps)
diff --git a/pandas/pandas/core/config_init.py b/pandas/pandas/core/config_init.py
new file mode 100644
index 0000000000000..20fe8cbab1c9f
--- /dev/null
+++ b/pandas/pandas/core/config_init.py
@@ -0,0 +1,899 @@
+"""
+This module is imported from the pandas package __init__.py file
+in order to ensure that the core.config options registered here will
+be available as soon as the user loads the package. if register_option
+is invoked inside specific modules, they will not be registered until that
+module is imported, which may or may not be a problem.
+
+If you need to make sure options are available even before a certain
+module is imported, register them here rather than in the module.
+
+"""
+
+from __future__ import annotations
+
+from collections.abc import Callable
+import os
+from typing import Any
+
+import pandas._config.config as cf
+from pandas._config.config import (
+    is_bool,
+    is_callable,
+    is_instance_factory,
+    is_int,
+    is_nonnegative_int,
+    is_one_of_factory,
+    is_str,
+    is_text,
+)
+
+# compute
+
+use_bottleneck_doc = """
+: bool
+    Use the bottleneck library to accelerate if it is installed,
+    the default is True
+    Valid values: False,True
+"""
+
+
+def use_bottleneck_cb(key: str) -> None:
+    from pandas.core import nanops
+
+    nanops.set_use_bottleneck(cf.get_option(key))
+
+
+use_numexpr_doc = """
+: bool
+    Use the numexpr library to accelerate computation if it is installed,
+    the default is True
+    Valid values: False,True
+"""
+
+
+def use_numexpr_cb(key: str) -> None:
+    from pandas.core.computation import expressions
+
+    expressions.set_use_numexpr(cf.get_option(key))
+
+
+use_numba_doc = """
+: bool
+    Use the numba engine option for select operations if it is installed,
+    the default is False
+    Valid values: False,True
+"""
+
+
+def use_numba_cb(key: str) -> None:
+    from pandas.core.util import numba_
+
+    numba_.set_use_numba(cf.get_option(key))
+
+
+with cf.config_prefix("compute"):
+    cf.register_option(
+        "use_bottleneck",
+        True,
+        use_bottleneck_doc,
+        validator=is_bool,
+        cb=use_bottleneck_cb,
+    )
+    cf.register_option(
+        "use_numexpr", True, use_numexpr_doc, validator=is_bool, cb=use_numexpr_cb
+    )
+    cf.register_option(
+        "use_numba", False, use_numba_doc, validator=is_bool, cb=use_numba_cb
+    )
+#
+# options from the "display" namespace
+
+pc_precision_doc = """
+: int
+    Floating point output precision in terms of number of places after the
+    decimal, for regular formatting as well as scientific notation. Similar
+    to ``precision`` in :meth:`numpy.set_printoptions`.
+"""
+
+pc_max_rows_doc = """
+: int
+    If max_rows is exceeded, switch to truncate view. Depending on
+    `large_repr`, objects are either centrally truncated or printed as
+    a summary view.
+
+    'None' value means unlimited. Beware that printing a large number of rows
+    could cause your rendering environment (the browser, etc.) to crash.
+
+    In case python/IPython is running in a terminal and `large_repr`
+    equals 'truncate' this can be set to 0 and pandas will auto-detect
+    the height of the terminal and print a truncated object which fits
+    the screen height. The IPython notebook, IPython qtconsole, or
+    IDLE do not run in a terminal and hence it is not possible to do
+    correct auto-detection.
+"""
+
+pc_min_rows_doc = """
+: int
+    The numbers of rows to show in a truncated view (when `max_rows` is
+    exceeded). Ignored when `max_rows` is set to None or 0. When set to
+    None, follows the value of `max_rows`.
+"""
+
+pc_max_cols_doc = """
+: int
+    If max_cols is exceeded, switch to truncate view. Depending on
+    `large_repr`, objects are either centrally truncated or printed as
+    a summary view.
+
+    'None' value means unlimited. Beware that printing a large number of
+    columns could cause your rendering environment (the browser, etc.) to
+    crash.
+
+    In case python/IPython is running in a terminal and `large_repr`
+    equals 'truncate' this can be set to 0 or None and pandas will auto-detect
+    the width of the terminal and print a truncated object which fits
+    the screen width. The IPython notebook, IPython qtconsole, or IDLE
+    do not run in a terminal and hence it is not possible to do
+    correct auto-detection and defaults to 20.
+"""
+
+pc_max_categories_doc = """
+: int
+    This sets the maximum number of categories pandas should output when
+    printing out a `Categorical` or a Series of dtype "category".
+"""
+
+pc_max_info_cols_doc = """
+: int
+    max_info_columns is used in DataFrame.info method to decide if
+    per column information will be printed.
+"""
+
+pc_nb_repr_h_doc = """
+: boolean
+    When True, IPython notebook will use html representation for
+    pandas objects (if it is available).
+"""
+
+pc_pprint_nest_depth = """
+: int
+    Controls the number of nested levels to process when pretty-printing
+"""
+
+pc_multi_sparse_doc = """
+: boolean
+    "sparsify" MultiIndex display (don't display repeated
+    elements in outer levels within groups)
+"""
+
+float_format_doc = """
+: callable
+    The callable should accept a floating point number and return
+    a string with the desired format of the number. This is used
+    in some places like SeriesFormatter.
+    See formats.format.EngFormatter for an example.
+"""
+
+max_colwidth_doc = """
+: int or None
+    The maximum width in characters of a column in the repr of
+    a pandas data structure. When the column overflows, a "..."
+    placeholder is embedded in the output. A 'None' value means unlimited.
+"""
+
+colheader_justify_doc = """
+: 'left'/'right'
+    Controls the justification of column headers. used by DataFrameFormatter.
+"""
+
+pc_expand_repr_doc = """
+: boolean
+    Whether to print out the full DataFrame repr for wide DataFrames across
+    multiple lines, `max_columns` is still respected, but the output will
+    wrap-around across multiple "pages" if its width exceeds `display.width`.
+"""
+
+pc_show_dimensions_doc = """
+: boolean or 'truncate'
+    Whether to print out dimensions at the end of DataFrame repr.
+    If 'truncate' is specified, only print out the dimensions if the
+    frame is truncated (e.g. not display all rows and/or columns)
+"""
+
+pc_east_asian_width_doc = """
+: boolean
+    Whether to use the Unicode East Asian Width to calculate the display text
+    width.
+    Enabling this may affect to the performance (default: False)
+"""
+
+
+pc_table_schema_doc = """
+: boolean
+    Whether to publish a Table Schema representation for frontends
+    that support it.
+    (default: False)
+"""
+
+pc_html_border_doc = """
+: int
+    A ``border=value`` attribute is inserted in the ``<table>`` tag
+    for the DataFrame HTML repr.
+"""
+
+pc_html_use_mathjax_doc = """\
+: boolean
+    When True, Jupyter notebook will process table contents using MathJax,
+    rendering mathematical expressions enclosed by the dollar symbol.
+    (default: True)
+"""
+
+pc_max_dir_items = """\
+: int
+    The number of items that will be added to `dir(...)`. 'None' value means
+    unlimited. Because dir is cached, changing this option will not immediately
+    affect already existing dataframes until a column is deleted or added.
+
+    This is for instance used to suggest columns from a dataframe to tab
+    completion.
+"""
+
+pc_width_doc = """
+: int
+    Width of the display in characters. In case python/IPython is running in
+    a terminal this can be set to None and pandas will correctly auto-detect
+    the width.
+    Note that the IPython notebook, IPython qtconsole, or IDLE do not run in a
+    terminal and hence it is not possible to correctly detect the width.
+"""
+
+pc_chop_threshold_doc = """
+: float or None
+    if set to a float value, all float values smaller than the given threshold
+    will be displayed as exactly 0 by repr and friends.
+"""
+
+pc_max_seq_items = """
+: int or None
+    When pretty-printing a long sequence, no more then `max_seq_items`
+    will be printed. If items are omitted, they will be denoted by the
+    addition of "..." to the resulting string.
+
+    If set to None, the number of items to be printed is unlimited.
+"""
+
+pc_max_info_rows_doc = """
+: int
+    df.info() will usually show null-counts for each column.
+    For large frames this can be quite slow. max_info_rows and max_info_cols
+    limit this null check only to frames with smaller dimensions than
+    specified.
+"""
+
+pc_large_repr_doc = """
+: 'truncate'/'info'
+    For DataFrames exceeding max_rows/max_cols, the repr (and HTML repr) can
+    show a truncated table, or switch to the view from
+    df.info() (the behaviour in earlier versions of pandas).
+"""
+
+pc_memory_usage_doc = """
+: bool, string or None
+    This specifies if the memory usage of a DataFrame should be displayed when
+    df.info() is called. Valid values True,False,'deep'
+"""
+
+
+def table_schema_cb(key: str) -> None:
+    from pandas.io.formats.printing import enable_data_resource_formatter
+
+    enable_data_resource_formatter(cf.get_option(key))
+
+
+def is_terminal() -> bool:
+    """
+    Detect if Python is running in a terminal.
+
+    Returns True if Python is running in a terminal or False if not.
+    """
+    try:
+        # error: Name 'get_ipython' is not defined
+        ip = get_ipython()  # type: ignore[name-defined]
+    except NameError:  # assume standard Python interpreter in a terminal
+        return True
+    else:
+        if hasattr(ip, "kernel"):  # IPython as a Jupyter kernel
+            return False
+        else:  # IPython in a terminal
+            return True
+
+
+with cf.config_prefix("display"):
+    cf.register_option("precision", 6, pc_precision_doc, validator=is_nonnegative_int)
+    cf.register_option(
+        "float_format",
+        None,
+        float_format_doc,
+        validator=is_one_of_factory([None, is_callable]),
+    )
+    cf.register_option(
+        "max_info_rows",
+        1690785,
+        pc_max_info_rows_doc,
+        validator=is_int,
+    )
+    cf.register_option("max_rows", 60, pc_max_rows_doc, validator=is_nonnegative_int)
+    cf.register_option(
+        "min_rows",
+        10,
+        pc_min_rows_doc,
+        validator=is_instance_factory((type(None), int)),
+    )
+    cf.register_option("max_categories", 8, pc_max_categories_doc, validator=is_int)
+
+    cf.register_option(
+        "max_colwidth",
+        50,
+        max_colwidth_doc,
+        validator=is_nonnegative_int,
+    )
+    if is_terminal():
+        max_cols = 0  # automatically determine optimal number of columns
+    else:
+        max_cols = 20  # cannot determine optimal number of columns
+    cf.register_option(
+        "max_columns", max_cols, pc_max_cols_doc, validator=is_nonnegative_int
+    )
+    cf.register_option(
+        "large_repr",
+        "truncate",
+        pc_large_repr_doc,
+        validator=is_one_of_factory(["truncate", "info"]),
+    )
+    cf.register_option("max_info_columns", 100, pc_max_info_cols_doc, validator=is_int)
+    cf.register_option(
+        "colheader_justify", "right", colheader_justify_doc, validator=is_text
+    )
+    cf.register_option("notebook_repr_html", True, pc_nb_repr_h_doc, validator=is_bool)
+    cf.register_option("pprint_nest_depth", 3, pc_pprint_nest_depth, validator=is_int)
+    cf.register_option("multi_sparse", True, pc_multi_sparse_doc, validator=is_bool)
+    cf.register_option("expand_frame_repr", True, pc_expand_repr_doc)
+    cf.register_option(
+        "show_dimensions",
+        "truncate",
+        pc_show_dimensions_doc,
+        validator=is_one_of_factory([True, False, "truncate"]),
+    )
+    cf.register_option("chop_threshold", None, pc_chop_threshold_doc)
+    cf.register_option("max_seq_items", 100, pc_max_seq_items)
+    cf.register_option(
+        "width", 80, pc_width_doc, validator=is_instance_factory((type(None), int))
+    )
+    cf.register_option(
+        "memory_usage",
+        True,
+        pc_memory_usage_doc,
+        validator=is_one_of_factory([None, True, False, "deep"]),
+    )
+    cf.register_option(
+        "unicode.east_asian_width", False, pc_east_asian_width_doc, validator=is_bool
+    )
+    cf.register_option(
+        "unicode.ambiguous_as_wide", False, pc_east_asian_width_doc, validator=is_bool
+    )
+    cf.register_option(
+        "html.table_schema",
+        False,
+        pc_table_schema_doc,
+        validator=is_bool,
+        cb=table_schema_cb,
+    )
+    cf.register_option("html.border", 1, pc_html_border_doc, validator=is_int)
+    cf.register_option(
+        "html.use_mathjax", True, pc_html_use_mathjax_doc, validator=is_bool
+    )
+    cf.register_option(
+        "max_dir_items", 100, pc_max_dir_items, validator=is_nonnegative_int
+    )
+
+tc_sim_interactive_doc = """
+: boolean
+    Whether to simulate interactive mode for purposes of testing
+"""
+
+with cf.config_prefix("mode"):
+    cf.register_option("sim_interactive", False, tc_sim_interactive_doc)
+
+
+# TODO better name?
+copy_on_write_doc = """
+: bool
+    Use new copy-view behaviour using Copy-on-Write. Defaults to False,
+    unless overridden by the 'PANDAS_COPY_ON_WRITE' environment variable
+    (if set to "1" for True, needs to be set before pandas is imported).
+"""
+
+
+with cf.config_prefix("mode"):
+    cf.register_option(
+        "copy_on_write",
+        # Get the default from an environment variable, if set, otherwise defaults
+        # to False. This environment variable can be set for testing.
+        "warn"
+        if os.environ.get("PANDAS_COPY_ON_WRITE", "0") == "warn"
+        else os.environ.get("PANDAS_COPY_ON_WRITE", "0") == "1",
+        copy_on_write_doc,
+        validator=is_one_of_factory([True, False, "warn"]),
+    )
+
+
+# user warnings
+chained_assignment = """
+: string
+    Raise an exception, warn, or no action if trying to use chained assignment,
+    The default is warn
+"""
+
+with cf.config_prefix("mode"):
+    cf.register_option(
+        "chained_assignment",
+        "warn",
+        chained_assignment,
+        validator=is_one_of_factory([None, "warn", "raise"]),
+    )
+
+performance_warnings = """
+: boolean
+    Whether to show or hide PerformanceWarnings.
+"""
+
+with cf.config_prefix("mode"):
+    cf.register_option(
+        "performance_warnings",
+        True,
+        performance_warnings,
+        validator=is_bool,
+    )
+
+
+string_storage_doc = """
+: string
+    The default storage for StringDtype.
+"""
+
+
+def is_valid_string_storage(value: Any) -> None:
+    legal_values = ["auto", "python", "pyarrow"]
+    if value not in legal_values:
+        msg = "Value must be one of python|pyarrow"
+        if value == "pyarrow_numpy":
+            # TODO: we can remove extra message after 3.0
+            msg += (
+                ". 'pyarrow_numpy' was specified, but this option should be "
+                "enabled using pandas.options.future.infer_string instead"
+            )
+        raise ValueError(msg)
+
+
+with cf.config_prefix("mode"):
+    cf.register_option(
+        "string_storage",
+        "auto",
+        string_storage_doc,
+        # validator=is_one_of_factory(["python", "pyarrow"]),
+        validator=is_valid_string_storage,
+    )
+
+
+# Set up the io.excel specific reader configuration.
+reader_engine_doc = """
+: string
+    The default Excel reader engine for '{ext}' files. Available options:
+    auto, {others}.
+"""
+
+_xls_options = ["xlrd", "calamine"]
+_xlsm_options = ["xlrd", "openpyxl", "calamine"]
+_xlsx_options = ["xlrd", "openpyxl", "calamine"]
+_ods_options = ["odf", "calamine"]
+_xlsb_options = ["pyxlsb", "calamine"]
+
+
+with cf.config_prefix("io.excel.xls"):
+    cf.register_option(
+        "reader",
+        "auto",
+        reader_engine_doc.format(ext="xls", others=", ".join(_xls_options)),
+        validator=is_one_of_factory(_xls_options + ["auto"]),
+    )
+
+with cf.config_prefix("io.excel.xlsm"):
+    cf.register_option(
+        "reader",
+        "auto",
+        reader_engine_doc.format(ext="xlsm", others=", ".join(_xlsm_options)),
+        validator=is_one_of_factory(_xlsm_options + ["auto"]),
+    )
+
+
+with cf.config_prefix("io.excel.xlsx"):
+    cf.register_option(
+        "reader",
+        "auto",
+        reader_engine_doc.format(ext="xlsx", others=", ".join(_xlsx_options)),
+        validator=is_one_of_factory(_xlsx_options + ["auto"]),
+    )
+
+
+with cf.config_prefix("io.excel.ods"):
+    cf.register_option(
+        "reader",
+        "auto",
+        reader_engine_doc.format(ext="ods", others=", ".join(_ods_options)),
+        validator=is_one_of_factory(_ods_options + ["auto"]),
+    )
+
+with cf.config_prefix("io.excel.xlsb"):
+    cf.register_option(
+        "reader",
+        "auto",
+        reader_engine_doc.format(ext="xlsb", others=", ".join(_xlsb_options)),
+        validator=is_one_of_factory(_xlsb_options + ["auto"]),
+    )
+
+# Set up the io.excel specific writer configuration.
+writer_engine_doc = """
+: string
+    The default Excel writer engine for '{ext}' files. Available options:
+    auto, {others}.
+"""
+
+_xlsm_options = ["openpyxl"]
+_xlsx_options = ["openpyxl", "xlsxwriter"]
+_ods_options = ["odf"]
+
+
+with cf.config_prefix("io.excel.xlsm"):
+    cf.register_option(
+        "writer",
+        "auto",
+        writer_engine_doc.format(ext="xlsm", others=", ".join(_xlsm_options)),
+        validator=str,
+    )
+
+
+with cf.config_prefix("io.excel.xlsx"):
+    cf.register_option(
+        "writer",
+        "auto",
+        writer_engine_doc.format(ext="xlsx", others=", ".join(_xlsx_options)),
+        validator=str,
+    )
+
+
+with cf.config_prefix("io.excel.ods"):
+    cf.register_option(
+        "writer",
+        "auto",
+        writer_engine_doc.format(ext="ods", others=", ".join(_ods_options)),
+        validator=str,
+    )
+
+
+# Set up the io.parquet specific configuration.
+parquet_engine_doc = """
+: string
+    The default parquet reader/writer engine. Available options:
+    'auto', 'pyarrow', 'fastparquet', the default is 'auto'
+"""
+
+with cf.config_prefix("io.parquet"):
+    cf.register_option(
+        "engine",
+        "auto",
+        parquet_engine_doc,
+        validator=is_one_of_factory(["auto", "pyarrow", "fastparquet"]),
+    )
+
+
+# Set up the io.sql specific configuration.
+sql_engine_doc = """
+: string
+    The default sql reader/writer engine. Available options:
+    'auto', 'sqlalchemy', the default is 'auto'
+"""
+
+with cf.config_prefix("io.sql"):
+    cf.register_option(
+        "engine",
+        "auto",
+        sql_engine_doc,
+        validator=is_one_of_factory(["auto", "sqlalchemy"]),
+    )
+
+# --------
+# Plotting
+# ---------
+
+plotting_backend_doc = """
+: str
+    The plotting backend to use. The default value is "matplotlib", the
+    backend provided with pandas. Other backends can be specified by
+    providing the name of the module that implements the backend.
+"""
+
+
+def register_plotting_backend_cb(key: str | None) -> None:
+    if key == "matplotlib":
+        # We defer matplotlib validation, since it's the default
+        return
+    from pandas.plotting._core import _get_plot_backend
+
+    _get_plot_backend(key)
+
+
+with cf.config_prefix("plotting"):
+    cf.register_option(
+        "backend",
+        defval="matplotlib",
+        doc=plotting_backend_doc,
+        validator=register_plotting_backend_cb,  # type: ignore[arg-type]
+    )
+
+
+register_converter_doc = """
+: bool or 'auto'.
+    Whether to register converters with matplotlib's units registry for
+    dates, times, datetimes, and Periods. Toggling to False will remove
+    the converters, restoring any converters that pandas overwrote.
+"""
+
+
+def register_converter_cb(key: str) -> None:
+    from pandas.plotting import (
+        deregister_matplotlib_converters,
+        register_matplotlib_converters,
+    )
+
+    if cf.get_option(key):
+        register_matplotlib_converters()
+    else:
+        deregister_matplotlib_converters()
+
+
+with cf.config_prefix("plotting.matplotlib"):
+    cf.register_option(
+        "register_converters",
+        "auto",
+        register_converter_doc,
+        validator=is_one_of_factory(["auto", True, False]),
+        cb=register_converter_cb,
+    )
+
+# ------
+# Styler
+# ------
+
+styler_sparse_index_doc = """
+: bool
+    Whether to sparsify the display of a hierarchical index. Setting to False will
+    display each explicit level element in a hierarchical key for each row.
+"""
+
+styler_sparse_columns_doc = """
+: bool
+    Whether to sparsify the display of hierarchical columns. Setting to False will
+    display each explicit level element in a hierarchical key for each column.
+"""
+
+styler_render_repr = """
+: str
+    Determine which output to use in Jupyter Notebook in {"html", "latex"}.
+"""
+
+styler_max_elements = """
+: int
+    The maximum number of data-cell (<td>) elements that will be rendered before
+    trimming will occur over columns, rows or both if needed.
+"""
+
+styler_max_rows = """
+: int, optional
+    The maximum number of rows that will be rendered. May still be reduced to
+    satisfy ``max_elements``, which takes precedence.
+"""
+
+styler_max_columns = """
+: int, optional
+    The maximum number of columns that will be rendered. May still be reduced to
+    satisfy ``max_elements``, which takes precedence.
+"""
+
+styler_precision = """
+: int
+    The precision for floats and complex numbers.
+"""
+
+styler_decimal = """
+: str
+    The character representation for the decimal separator for floats and complex.
+"""
+
+styler_thousands = """
+: str, optional
+    The character representation for thousands separator for floats, int and complex.
+"""
+
+styler_na_rep = """
+: str, optional
+    The string representation for values identified as missing.
+"""
+
+styler_escape = """
+: str, optional
+    Whether to escape certain characters according to the given context; html or latex.
+"""
+
+styler_formatter = """
+: str, callable, dict, optional
+    A formatter object to be used as default within ``Styler.format``.
+"""
+
+styler_multirow_align = """
+: {"c", "t", "b"}
+    The specifier for vertical alignment of sparsified LaTeX multirows.
+"""
+
+styler_multicol_align = r"""
+: {"r", "c", "l", "naive-l", "naive-r"}
+    The specifier for horizontal alignment of sparsified LaTeX multicolumns. Pipe
+    decorators can also be added to non-naive values to draw vertical
+    rules, e.g. "\|r" will draw a rule on the left side of right aligned merged cells.
+"""
+
+styler_hrules = """
+: bool
+    Whether to add horizontal rules on top and bottom and below the headers.
+"""
+
+styler_environment = """
+: str
+    The environment to replace ``\\begin{table}``. If "longtable" is used results
+    in a specific longtable environment format.
+"""
+
+styler_encoding = """
+: str
+    The encoding used for output HTML and LaTeX files.
+"""
+
+styler_mathjax = """
+: bool
+    If False will render special CSS classes to table attributes that indicate Mathjax
+    will not be used in Jupyter Notebook.
+"""
+
+with cf.config_prefix("styler"):
+    cf.register_option("sparse.index", True, styler_sparse_index_doc, validator=is_bool)
+
+    cf.register_option(
+        "sparse.columns", True, styler_sparse_columns_doc, validator=is_bool
+    )
+
+    cf.register_option(
+        "render.repr",
+        "html",
+        styler_render_repr,
+        validator=is_one_of_factory(["html", "latex"]),
+    )
+
+    cf.register_option(
+        "render.max_elements",
+        2**18,
+        styler_max_elements,
+        validator=is_nonnegative_int,
+    )
+
+    cf.register_option(
+        "render.max_rows",
+        None,
+        styler_max_rows,
+        validator=is_nonnegative_int,
+    )
+
+    cf.register_option(
+        "render.max_columns",
+        None,
+        styler_max_columns,
+        validator=is_nonnegative_int,
+    )
+
+    cf.register_option("render.encoding", "utf-8", styler_encoding, validator=is_str)
+
+    cf.register_option("format.decimal", ".", styler_decimal, validator=is_str)
+
+    cf.register_option(
+        "format.precision", 6, styler_precision, validator=is_nonnegative_int
+    )
+
+    cf.register_option(
+        "format.thousands",
+        None,
+        styler_thousands,
+        validator=is_instance_factory((type(None), str)),
+    )
+
+    cf.register_option(
+        "format.na_rep",
+        None,
+        styler_na_rep,
+        validator=is_instance_factory((type(None), str)),
+    )
+
+    cf.register_option(
+        "format.escape",
+        None,
+        styler_escape,
+        validator=is_one_of_factory([None, "html", "latex", "latex-math"]),
+    )
+
+    # error: Argument 1 to "is_instance_factory" has incompatible type "tuple[
+    # ..., <typing special form>, ...]"; expected "type | tuple[type, ...]"
+    cf.register_option(
+        "format.formatter",
+        None,
+        styler_formatter,
+        validator=is_instance_factory(
+            (type(None), dict, Callable, str)  # type: ignore[arg-type]
+        ),
+    )
+
+    cf.register_option("html.mathjax", True, styler_mathjax, validator=is_bool)
+
+    cf.register_option(
+        "latex.multirow_align",
+        "c",
+        styler_multirow_align,
+        validator=is_one_of_factory(["c", "t", "b", "naive"]),
+    )
+
+    val_mca = ["r", "|r|", "|r", "r|", "c", "|c|", "|c", "c|", "l", "|l|", "|l", "l|"]
+    val_mca += ["naive-l", "naive-r"]
+    cf.register_option(
+        "latex.multicol_align",
+        "r",
+        styler_multicol_align,
+        validator=is_one_of_factory(val_mca),
+    )
+
+    cf.register_option("latex.hrules", False, styler_hrules, validator=is_bool)
+
+    cf.register_option(
+        "latex.environment",
+        None,
+        styler_environment,
+        validator=is_instance_factory((type(None), str)),
+    )
+
+
+with cf.config_prefix("future"):
+    cf.register_option(
+        "infer_string",
+        True if os.environ.get("PANDAS_FUTURE_INFER_STRING", "0") == "1" else False,
+        "Whether to infer sequence of str objects as pyarrow string "
+        "dtype, which will be the default in pandas 3.0 "
+        "(at which point this option will be deprecated).",
+        validator=is_one_of_factory([True, False]),
+    )
+
+    cf.register_option(
+        "no_silent_downcasting",
+        False,
+        "Whether to opt-in to the future behavior which will *not* silently "
+        "downcast results from Series and DataFrame `where`, `mask`, and `clip` "
+        "methods. "
+        "Silent downcasting will be removed in pandas 3.0 "
+        "(at which point this option will be deprecated).",
+        validator=is_one_of_factory([True, False]),
+    )
diff --git a/pandas/pandas/core/construction.py b/pandas/pandas/core/construction.py
new file mode 100644
index 0000000000000..ada492787a179
--- /dev/null
+++ b/pandas/pandas/core/construction.py
@@ -0,0 +1,835 @@
+"""
+Constructor functions intended to be shared by pd.array, Series.__init__,
+and Index.__new__.
+
+These should not depend on core.internals.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    cast,
+    overload,
+)
+
+import numpy as np
+from numpy import ma
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    get_supported_dtype,
+    is_supported_dtype,
+)
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.cast import (
+    construct_1d_arraylike_from_scalar,
+    construct_1d_object_array_from_listlike,
+    maybe_cast_to_datetime,
+    maybe_cast_to_integer_array,
+    maybe_convert_platform,
+    maybe_infer_to_datetimelike,
+    maybe_promote,
+)
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_list_like,
+    is_object_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import NumpyEADtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCExtensionArray,
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+import pandas.core.common as com
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from pandas._typing import (
+        AnyArrayLike,
+        ArrayLike,
+        Dtype,
+        DtypeObj,
+        T,
+    )
+
+    from pandas import (
+        Index,
+        Series,
+    )
+    from pandas.core.arrays import (
+        DatetimeArray,
+        ExtensionArray,
+        TimedeltaArray,
+    )
+
+
+def array(
+    data: Sequence[object] | AnyArrayLike,
+    dtype: Dtype | None = None,
+    copy: bool = True,
+) -> ExtensionArray:
+    """
+    Create an array.
+
+    This method constructs an array using pandas extension types when possible.
+    If `dtype` is specified, it determines the type of array returned. Otherwise,
+    pandas attempts to infer the appropriate dtype based on `data`.
+
+    Parameters
+    ----------
+    data : Sequence of objects
+        The scalars inside `data` should be instances of the
+        scalar type for `dtype`. It's expected that `data`
+        represents a 1-dimensional array of data.
+
+        When `data` is an Index or Series, the underlying array
+        will be extracted from `data`.
+
+    dtype : str, np.dtype, or ExtensionDtype, optional
+        The dtype to use for the array. This may be a NumPy
+        dtype or an extension type registered with pandas using
+        :meth:`pandas.api.extensions.register_extension_dtype`.
+
+        If not specified, there are two possibilities:
+
+        1. When `data` is a :class:`Series`, :class:`Index`, or
+           :class:`ExtensionArray`, the `dtype` will be taken
+           from the data.
+        2. Otherwise, pandas will attempt to infer the `dtype`
+           from the data.
+
+        Note that when `data` is a NumPy array, ``data.dtype`` is
+        *not* used for inferring the array type. This is because
+        NumPy cannot represent all the types of data that can be
+        held in extension arrays.
+
+        Currently, pandas will infer an extension dtype for sequences of
+
+        ============================== =======================================
+        Scalar Type                    Array Type
+        ============================== =======================================
+        :class:`pandas.Interval`       :class:`pandas.arrays.IntervalArray`
+        :class:`pandas.Period`         :class:`pandas.arrays.PeriodArray`
+        :class:`datetime.datetime`     :class:`pandas.arrays.DatetimeArray`
+        :class:`datetime.timedelta`    :class:`pandas.arrays.TimedeltaArray`
+        :class:`int`                   :class:`pandas.arrays.IntegerArray`
+        :class:`float`                 :class:`pandas.arrays.FloatingArray`
+        :class:`str`                   :class:`pandas.arrays.StringArray` or
+                                       :class:`pandas.arrays.ArrowStringArray`
+        :class:`bool`                  :class:`pandas.arrays.BooleanArray`
+        ============================== =======================================
+
+        The ExtensionArray created when the scalar type is :class:`str` is determined by
+        ``pd.options.mode.string_storage`` if the dtype is not explicitly given.
+
+        For all other cases, NumPy's usual inference rules will be used.
+    copy : bool, default True
+        Whether to copy the data, even if not necessary. Depending
+        on the type of `data`, creating the new array may require
+        copying data, even if ``copy=False``.
+
+    Returns
+    -------
+    ExtensionArray
+        The newly created array.
+
+    Raises
+    ------
+    ValueError
+        When `data` is not 1-dimensional.
+
+    See Also
+    --------
+    numpy.array : Construct a NumPy array.
+    Series : Construct a pandas Series.
+    Index : Construct a pandas Index.
+    arrays.NumpyExtensionArray : ExtensionArray wrapping a NumPy array.
+    Series.array : Extract the array stored within a Series.
+
+    Notes
+    -----
+    Omitting the `dtype` argument means pandas will attempt to infer the
+    best array type from the values in the data. As new array types are
+    added by pandas and 3rd party libraries, the "best" array type may
+    change. We recommend specifying `dtype` to ensure that
+
+    1. the correct array type for the data is returned
+    2. the returned array type doesn't change as new extension types
+       are added by pandas and third-party libraries
+
+    Additionally, if the underlying memory representation of the returned
+    array matters, we recommend specifying the `dtype` as a concrete object
+    rather than a string alias or allowing it to be inferred. For example,
+    a future version of pandas or a 3rd-party library may include a
+    dedicated ExtensionArray for string data. In this event, the following
+    would no longer return a :class:`arrays.NumpyExtensionArray` backed by a
+    NumPy array.
+
+    >>> pd.array(["a", "b"], dtype=str)
+    <NumpyExtensionArray>
+    ['a', 'b']
+    Length: 2, dtype: str32
+
+    This would instead return the new ExtensionArray dedicated for string
+    data. If you really need the new array to be backed by a  NumPy array,
+    specify that in the dtype.
+
+    >>> pd.array(["a", "b"], dtype=np.dtype("<U1"))
+    <NumpyExtensionArray>
+    ['a', 'b']
+    Length: 2, dtype: str32
+
+    Finally, Pandas has arrays that mostly overlap with NumPy
+
+      * :class:`arrays.DatetimeArray`
+      * :class:`arrays.TimedeltaArray`
+
+    When data with a ``datetime64[ns]`` or ``timedelta64[ns]`` dtype is
+    passed, pandas will always return a ``DatetimeArray`` or ``TimedeltaArray``
+    rather than a ``NumpyExtensionArray``. This is for symmetry with the case of
+    timezone-aware data, which NumPy does not natively support.
+
+    >>> pd.array(["2015", "2016"], dtype="datetime64[ns]")
+    <DatetimeArray>
+    ['2015-01-01 00:00:00', '2016-01-01 00:00:00']
+    Length: 2, dtype: datetime64[ns]
+
+    >>> pd.array(["1h", "2h"], dtype="timedelta64[ns]")
+    <TimedeltaArray>
+    ['0 days 01:00:00', '0 days 02:00:00']
+    Length: 2, dtype: timedelta64[ns]
+
+    Examples
+    --------
+    If a dtype is not specified, pandas will infer the best dtype from the values.
+    See the description of `dtype` for the types pandas infers for.
+
+    >>> pd.array([1, 2])
+    <IntegerArray>
+    [1, 2]
+    Length: 2, dtype: Int64
+
+    >>> pd.array([1, 2, np.nan])
+    <IntegerArray>
+    [1, 2, <NA>]
+    Length: 3, dtype: Int64
+
+    >>> pd.array([1.1, 2.2])
+    <FloatingArray>
+    [1.1, 2.2]
+    Length: 2, dtype: Float64
+
+    >>> pd.array(["a", None, "c"])
+    <StringArray>
+    ['a', <NA>, 'c']
+    Length: 3, dtype: string
+
+    >>> with pd.option_context("string_storage", "pyarrow"):
+    ...     arr = pd.array(["a", None, "c"])
+    >>> arr
+    <ArrowStringArray>
+    ['a', <NA>, 'c']
+    Length: 3, dtype: string
+
+    >>> pd.array([pd.Period("2000", freq="D"), pd.Period("2000", freq="D")])
+    <PeriodArray>
+    ['2000-01-01', '2000-01-01']
+    Length: 2, dtype: period[D]
+
+    You can use the string alias for `dtype`
+
+    >>> pd.array(["a", "b", "a"], dtype="category")
+    ['a', 'b', 'a']
+    Categories (2, object): ['a', 'b']
+
+    Or specify the actual dtype
+
+    >>> pd.array(
+    ...     ["a", "b", "a"], dtype=pd.CategoricalDtype(["a", "b", "c"], ordered=True)
+    ... )
+    ['a', 'b', 'a']
+    Categories (3, object): ['a' < 'b' < 'c']
+
+    If pandas does not infer a dedicated extension type a
+    :class:`arrays.NumpyExtensionArray` is returned.
+
+    >>> pd.array([1 + 1j, 3 + 2j])
+    <NumpyExtensionArray>
+    [(1+1j), (3+2j)]
+    Length: 2, dtype: complex128
+
+    As mentioned in the "Notes" section, new extension types may be added
+    in the future (by pandas or 3rd party libraries), causing the return
+    value to no longer be a :class:`arrays.NumpyExtensionArray`. Specify the
+    `dtype` as a NumPy dtype if you need to ensure there's no future change in
+    behavior.
+
+    >>> pd.array([1, 2], dtype=np.dtype("int32"))
+    <NumpyExtensionArray>
+    [1, 2]
+    Length: 2, dtype: int32
+
+    `data` must be 1-dimensional. A ValueError is raised when the input
+    has the wrong dimensionality.
+
+    >>> pd.array(1)
+    Traceback (most recent call last):
+      ...
+    ValueError: Cannot pass scalar '1' to 'pandas.array'.
+    """
+    from pandas.core.arrays import (
+        BooleanArray,
+        DatetimeArray,
+        ExtensionArray,
+        FloatingArray,
+        IntegerArray,
+        NumpyExtensionArray,
+        TimedeltaArray,
+    )
+    from pandas.core.arrays.string_ import StringDtype
+
+    if lib.is_scalar(data):
+        msg = f"Cannot pass scalar '{data}' to 'pandas.array'."
+        raise ValueError(msg)
+    elif isinstance(data, ABCDataFrame):
+        raise TypeError("Cannot pass DataFrame to 'pandas.array'")
+
+    if dtype is None and isinstance(data, (ABCSeries, ABCIndex, ExtensionArray)):
+        # Note: we exclude np.ndarray here, will do type inference on it
+        dtype = data.dtype
+
+    data = extract_array(data, extract_numpy=True)
+
+    # this returns None for not-found dtypes.
+    if dtype is not None:
+        dtype = pandas_dtype(dtype)
+
+    if isinstance(data, ExtensionArray) and (dtype is None or data.dtype == dtype):
+        # e.g. TimedeltaArray[s], avoid casting to NumpyExtensionArray
+        if copy:
+            return data.copy()
+        return data
+
+    if isinstance(dtype, ExtensionDtype):
+        cls = dtype.construct_array_type()
+        return cls._from_sequence(data, dtype=dtype, copy=copy)
+
+    if dtype is None:
+        was_ndarray = isinstance(data, np.ndarray)
+        # error: Item "Sequence[object]" of "Sequence[object] | ExtensionArray |
+        # ndarray[Any, Any]" has no attribute "dtype"
+        if not was_ndarray or data.dtype == object:  # type: ignore[union-attr]
+            result = lib.maybe_convert_objects(
+                ensure_object(data),
+                convert_non_numeric=True,
+                convert_to_nullable_dtype=True,
+                dtype_if_all_nat=None,
+            )
+            result = ensure_wrapped_if_datetimelike(result)
+            if isinstance(result, np.ndarray):
+                if len(result) == 0 and not was_ndarray:
+                    # e.g. empty list
+                    return FloatingArray._from_sequence(data, dtype="Float64")
+                return NumpyExtensionArray._from_sequence(
+                    data, dtype=result.dtype, copy=copy
+                )
+            if result is data and copy:
+                return result.copy()
+            return result
+
+        data = cast(np.ndarray, data)
+        result = ensure_wrapped_if_datetimelike(data)
+        if result is not data:
+            result = cast("DatetimeArray | TimedeltaArray", result)
+            if copy and result.dtype == data.dtype:
+                return result.copy()
+            return result
+
+        if data.dtype.kind in "SU":
+            # StringArray/ArrowStringArray depending on pd.options.mode.string_storage
+            dtype = StringDtype()
+            cls = dtype.construct_array_type()
+            return cls._from_sequence(data, dtype=dtype, copy=copy)
+
+        elif data.dtype.kind in "iu":
+            dtype = IntegerArray._dtype_cls._get_dtype_mapping()[data.dtype]
+            return IntegerArray._from_sequence(data, dtype=dtype, copy=copy)
+        elif data.dtype.kind == "f":
+            # GH#44715 Exclude np.float16 bc FloatingArray does not support it;
+            #  we will fall back to NumpyExtensionArray.
+            if data.dtype == np.float16:
+                return NumpyExtensionArray._from_sequence(
+                    data, dtype=data.dtype, copy=copy
+                )
+            dtype = FloatingArray._dtype_cls._get_dtype_mapping()[data.dtype]
+            return FloatingArray._from_sequence(data, dtype=dtype, copy=copy)
+
+        elif data.dtype.kind == "b":
+            return BooleanArray._from_sequence(data, dtype="boolean", copy=copy)
+        else:
+            # e.g. complex
+            return NumpyExtensionArray._from_sequence(data, dtype=data.dtype, copy=copy)
+
+    # Pandas overrides NumPy for
+    #   1. datetime64[ns,us,ms,s]
+    #   2. timedelta64[ns,us,ms,s]
+    # so that a DatetimeArray is returned.
+    if lib.is_np_dtype(dtype, "M") and is_supported_dtype(dtype):
+        return DatetimeArray._from_sequence(data, dtype=dtype, copy=copy)
+    if lib.is_np_dtype(dtype, "m") and is_supported_dtype(dtype):
+        return TimedeltaArray._from_sequence(data, dtype=dtype, copy=copy)
+
+    elif lib.is_np_dtype(dtype, "mM"):
+        raise ValueError(
+            # GH#53817
+            r"datetime64 and timedelta64 dtype resolutions other than "
+            r"'s', 'ms', 'us', and 'ns' are no longer supported."
+        )
+
+    return NumpyExtensionArray._from_sequence(data, dtype=dtype, copy=copy)
+
+
+_typs = frozenset(
+    {
+        "index",
+        "rangeindex",
+        "multiindex",
+        "datetimeindex",
+        "timedeltaindex",
+        "periodindex",
+        "categoricalindex",
+        "intervalindex",
+        "series",
+    }
+)
+
+
+@overload
+def extract_array(
+    obj: Series | Index, extract_numpy: bool = ..., extract_range: bool = ...
+) -> ArrayLike: ...
+
+
+@overload
+def extract_array(
+    obj: T, extract_numpy: bool = ..., extract_range: bool = ...
+) -> T | ArrayLike: ...
+
+
+def extract_array(
+    obj: T, extract_numpy: bool = False, extract_range: bool = False
+) -> T | ArrayLike:
+    """
+    Extract the ndarray or ExtensionArray from a Series or Index.
+
+    For all other types, `obj` is just returned as is.
+
+    Parameters
+    ----------
+    obj : object
+        For Series / Index, the underlying ExtensionArray is unboxed.
+
+    extract_numpy : bool, default False
+        Whether to extract the ndarray from a NumpyExtensionArray.
+
+    extract_range : bool, default False
+        If we have a RangeIndex, return range._values if True
+        (which is a materialized integer ndarray), otherwise return unchanged.
+
+    Returns
+    -------
+    arr : object
+
+    Examples
+    --------
+    >>> extract_array(pd.Series(["a", "b", "c"], dtype="category"))
+    ['a', 'b', 'c']
+    Categories (3, object): ['a', 'b', 'c']
+
+    Other objects like lists, arrays, and DataFrames are just passed through.
+
+    >>> extract_array([1, 2, 3])
+    [1, 2, 3]
+
+    For an ndarray-backed Series / Index the ndarray is returned.
+
+    >>> extract_array(pd.Series([1, 2, 3]))
+    array([1, 2, 3])
+
+    To extract all the way down to the ndarray, pass ``extract_numpy=True``.
+
+    >>> extract_array(pd.Series([1, 2, 3]), extract_numpy=True)
+    array([1, 2, 3])
+    """
+    typ = getattr(obj, "_typ", None)
+    if typ in _typs:
+        # i.e. isinstance(obj, (ABCIndex, ABCSeries))
+        if typ == "rangeindex":
+            if extract_range:
+                # error: "T" has no attribute "_values"
+                return obj._values  # type: ignore[attr-defined]
+            return obj
+
+        # error: "T" has no attribute "_values"
+        return obj._values  # type: ignore[attr-defined]
+
+    elif extract_numpy and typ == "npy_extension":
+        # i.e. isinstance(obj, ABCNumpyExtensionArray)
+        # error: "T" has no attribute "to_numpy"
+        return obj.to_numpy()  # type: ignore[attr-defined]
+
+    return obj
+
+
+def ensure_wrapped_if_datetimelike(arr):
+    """
+    Wrap datetime64 and timedelta64 ndarrays in DatetimeArray/TimedeltaArray.
+    """
+    if isinstance(arr, np.ndarray):
+        if arr.dtype.kind == "M":
+            from pandas.core.arrays import DatetimeArray
+
+            dtype = get_supported_dtype(arr.dtype)
+            return DatetimeArray._from_sequence(arr, dtype=dtype)
+
+        elif arr.dtype.kind == "m":
+            from pandas.core.arrays import TimedeltaArray
+
+            dtype = get_supported_dtype(arr.dtype)
+            return TimedeltaArray._from_sequence(arr, dtype=dtype)
+
+    return arr
+
+
+def sanitize_masked_array(data: ma.MaskedArray) -> np.ndarray:
+    """
+    Convert numpy MaskedArray to ensure mask is softened.
+    """
+    mask = ma.getmaskarray(data)
+    if mask.any():
+        dtype, fill_value = maybe_promote(data.dtype, np.nan)
+        dtype = cast(np.dtype, dtype)
+        data = ma.asarray(data.astype(dtype, copy=True))
+        data.soften_mask()  # set hardmask False if it was True
+        data[mask] = fill_value
+    else:
+        data = data.copy()
+    return data
+
+
+def sanitize_array(
+    data,
+    index: Index | None,
+    dtype: DtypeObj | None = None,
+    copy: bool = False,
+    *,
+    allow_2d: bool = False,
+) -> ArrayLike:
+    """
+    Sanitize input data to an ndarray or ExtensionArray, copy if specified,
+    coerce to the dtype if specified.
+
+    Parameters
+    ----------
+    data : Any
+    index : Index or None, default None
+    dtype : np.dtype, ExtensionDtype, or None, default None
+    copy : bool, default False
+    allow_2d : bool, default False
+        If False, raise if we have a 2D Arraylike.
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+    """
+    original_dtype = dtype
+    if isinstance(data, ma.MaskedArray):
+        data = sanitize_masked_array(data)
+
+    if isinstance(dtype, NumpyEADtype):
+        # Avoid ending up with a NumpyExtensionArray
+        dtype = dtype.numpy_dtype
+
+    infer_object = not isinstance(data, (ABCIndex, ABCSeries))
+
+    # extract ndarray or ExtensionArray, ensure we have no NumpyExtensionArray
+    data = extract_array(data, extract_numpy=True, extract_range=True)
+
+    if isinstance(data, np.ndarray) and data.ndim == 0:
+        if dtype is None:
+            dtype = data.dtype
+        data = lib.item_from_zerodim(data)
+    elif isinstance(data, range):
+        # GH#16804
+        data = range_to_ndarray(data)
+        copy = False
+
+    if not is_list_like(data):
+        if index is None:
+            raise ValueError("index must be specified when data is not list-like")
+        if isinstance(data, str) and using_string_dtype() and original_dtype is None:
+            from pandas.core.arrays.string_ import StringDtype
+
+            dtype = StringDtype(na_value=np.nan)
+        data = construct_1d_arraylike_from_scalar(data, len(index), dtype)
+
+        return data
+
+    elif isinstance(data, ABCExtensionArray):
+        # it is already ensured above this is not a NumpyExtensionArray
+        # Until GH#49309 is fixed this check needs to come before the
+        #  ExtensionDtype check
+        if dtype is not None:
+            subarr = data.astype(dtype, copy=copy)
+        elif copy:
+            subarr = data.copy()
+        else:
+            subarr = data
+
+    elif isinstance(dtype, ExtensionDtype):
+        # create an extension array from its dtype
+        _sanitize_non_ordered(data)
+        cls = dtype.construct_array_type()
+        if not hasattr(data, "__array__"):
+            data = list(data)
+        subarr = cls._from_sequence(data, dtype=dtype, copy=copy)
+
+    # GH#846
+    elif isinstance(data, np.ndarray):
+        if isinstance(data, np.matrix):
+            data = data.A
+
+        if dtype is None:
+            subarr = data
+            if data.dtype == object and infer_object:
+                subarr = maybe_infer_to_datetimelike(data)
+            elif data.dtype.kind == "U" and using_string_dtype():
+                from pandas.core.arrays.string_ import StringDtype
+
+                dtype = StringDtype(na_value=np.nan)
+                subarr = dtype.construct_array_type()._from_sequence(data, dtype=dtype)
+
+            if (
+                subarr is data
+                or (subarr.dtype == "str" and subarr.dtype.storage == "python")  # type: ignore[union-attr]
+            ) and copy:
+                subarr = subarr.copy()
+
+        else:
+            # we will try to copy by-definition here
+            subarr = _try_cast(data, dtype, copy)
+
+    elif hasattr(data, "__array__"):
+        # e.g. dask array GH#38645
+        if not copy:
+            data = np.asarray(data)
+        else:
+            data = np.array(data, copy=copy)
+        return sanitize_array(
+            data,
+            index=index,
+            dtype=dtype,
+            copy=False,
+            allow_2d=allow_2d,
+        )
+
+    else:
+        _sanitize_non_ordered(data)
+        # materialize e.g. generators, convert e.g. tuples, abc.ValueView
+        data = list(data)
+
+        if len(data) == 0 and dtype is None:
+            # We default to float64, matching numpy
+            subarr = np.array([], dtype=np.float64)
+
+        elif dtype is not None:
+            subarr = _try_cast(data, dtype, copy)
+
+        else:
+            subarr = maybe_convert_platform(data)
+            if subarr.dtype == object:
+                subarr = cast(np.ndarray, subarr)
+                subarr = maybe_infer_to_datetimelike(subarr)
+
+    subarr = _sanitize_ndim(subarr, data, dtype, index, allow_2d=allow_2d)
+
+    if isinstance(subarr, np.ndarray):
+        # at this point we should have dtype be None or subarr.dtype == dtype
+        dtype = cast(np.dtype, dtype)
+        subarr = _sanitize_str_dtypes(subarr, data, dtype, copy)
+
+    return subarr
+
+
+def range_to_ndarray(rng: range) -> np.ndarray:
+    """
+    Cast a range object to ndarray.
+    """
+    # GH#30171 perf avoid realizing range as a list in np.array
+    try:
+        arr = np.arange(rng.start, rng.stop, rng.step, dtype="int64")
+    except OverflowError:
+        # GH#30173 handling for ranges that overflow int64
+        if (rng.start >= 0 and rng.step > 0) or (rng.step < 0 <= rng.stop):
+            try:
+                arr = np.arange(rng.start, rng.stop, rng.step, dtype="uint64")
+            except OverflowError:
+                arr = construct_1d_object_array_from_listlike(list(rng))
+        else:
+            arr = construct_1d_object_array_from_listlike(list(rng))
+    return arr
+
+
+def _sanitize_non_ordered(data) -> None:
+    """
+    Raise only for unordered sets, e.g., not for dict_keys
+    """
+    if isinstance(data, (set, frozenset)):
+        raise TypeError(f"'{type(data).__name__}' type is unordered")
+
+
+def _sanitize_ndim(
+    result: ArrayLike,
+    data,
+    dtype: DtypeObj | None,
+    index: Index | None,
+    *,
+    allow_2d: bool = False,
+) -> ArrayLike:
+    """
+    Ensure we have a 1-dimensional result array.
+    """
+    if getattr(result, "ndim", 0) == 0:
+        raise ValueError("result should be arraylike with ndim > 0")
+
+    if result.ndim == 1:
+        # the result that we want
+        result = _maybe_repeat(result, index)
+
+    elif result.ndim > 1:
+        if isinstance(data, np.ndarray):
+            if allow_2d:
+                return result
+            raise ValueError(
+                f"Data must be 1-dimensional, got ndarray of shape {data.shape} instead"
+            )
+        if is_object_dtype(dtype) and isinstance(dtype, ExtensionDtype):
+            # i.e. NumpyEADtype("O")
+
+            result = com.asarray_tuplesafe(data, dtype=np.dtype("object"))
+            cls = dtype.construct_array_type()
+            result = cls._from_sequence(result, dtype=dtype)
+        else:
+            # error: Argument "dtype" to "asarray_tuplesafe" has incompatible type
+            # "Union[dtype[Any], ExtensionDtype, None]"; expected "Union[str,
+            # dtype[Any], None]"
+            result = com.asarray_tuplesafe(data, dtype=dtype)  # type: ignore[arg-type]
+    return result
+
+
+def _sanitize_str_dtypes(
+    result: np.ndarray, data, dtype: np.dtype | None, copy: bool
+) -> np.ndarray:
+    """
+    Ensure we have a dtype that is supported by pandas.
+    """
+
+    # This is to prevent mixed-type Series getting all casted to
+    # NumPy string type, e.g. NaN --> '-1#IND'.
+    if issubclass(result.dtype.type, str):
+        # GH#16605
+        # If not empty convert the data to dtype
+        # GH#19853: If data is a scalar, result has already the result
+        if not lib.is_scalar(data):
+            if not np.all(isna(data)):
+                data = np.asarray(data, dtype=dtype)
+            if not copy:
+                result = np.asarray(data, dtype=object)
+            else:
+                result = np.array(data, dtype=object, copy=copy)
+    return result
+
+
+def _maybe_repeat(arr: ArrayLike, index: Index | None) -> ArrayLike:
+    """
+    If we have a length-1 array and an index describing how long we expect
+    the result to be, repeat the array.
+    """
+    if index is not None:
+        if 1 == len(arr) != len(index):
+            arr = arr.repeat(len(index))
+    return arr
+
+
+def _try_cast(
+    arr: list | np.ndarray,
+    dtype: np.dtype,
+    copy: bool,
+) -> ArrayLike:
+    """
+    Convert input to numpy ndarray and optionally cast to a given dtype.
+
+    Parameters
+    ----------
+    arr : ndarray or list
+        Excludes: ExtensionArray, Series, Index.
+    dtype : np.dtype
+    copy : bool
+        If False, don't copy the data if not needed.
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+    """
+    is_ndarray = isinstance(arr, np.ndarray)
+
+    if dtype == object:
+        if not is_ndarray:
+            subarr = construct_1d_object_array_from_listlike(arr)
+            return subarr
+        return ensure_wrapped_if_datetimelike(arr).astype(dtype, copy=copy)
+
+    elif dtype.kind == "U":
+        # TODO: test cases with arr.dtype.kind in "mM"
+        if is_ndarray:
+            arr = cast(np.ndarray, arr)
+            shape = arr.shape
+            if arr.ndim > 1:
+                arr = arr.ravel()
+        else:
+            shape = (len(arr),)
+        return lib.ensure_string_array(arr, convert_na_value=False, copy=copy).reshape(
+            shape
+        )
+
+    elif dtype.kind in "mM":
+        if is_ndarray:
+            arr = cast(np.ndarray, arr)
+            if arr.ndim == 2 and arr.shape[1] == 1:
+                # GH#60081: DataFrame Constructor converts 1D data to array of
+                # shape (N, 1), but maybe_cast_to_datetime assumes 1D input
+                return maybe_cast_to_datetime(arr[:, 0], dtype).reshape(arr.shape)
+        return maybe_cast_to_datetime(arr, dtype)
+
+    # GH#15832: Check if we are requesting a numeric dtype and
+    # that we can convert the data to the requested dtype.
+    elif dtype.kind in "iu":
+        # this will raise if we have e.g. floats
+
+        subarr = maybe_cast_to_integer_array(arr, dtype)
+    elif not copy:
+        subarr = np.asarray(arr, dtype=dtype)
+    else:
+        subarr = np.array(arr, dtype=dtype, copy=copy)
+
+    return subarr
diff --git a/pandas/pandas/core/dtypes/__init__.py b/pandas/pandas/core/dtypes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/dtypes/api.py b/pandas/pandas/core/dtypes/api.py
new file mode 100644
index 0000000000000..e66104d6afcd9
--- /dev/null
+++ b/pandas/pandas/core/dtypes/api.py
@@ -0,0 +1,83 @@
+from pandas.core.dtypes.common import (
+    is_any_real_numeric_dtype,
+    is_array_like,
+    is_bool,
+    is_bool_dtype,
+    is_categorical_dtype,
+    is_complex,
+    is_complex_dtype,
+    is_datetime64_any_dtype,
+    is_datetime64_dtype,
+    is_datetime64_ns_dtype,
+    is_datetime64tz_dtype,
+    is_dict_like,
+    is_dtype_equal,
+    is_extension_array_dtype,
+    is_file_like,
+    is_float,
+    is_float_dtype,
+    is_hashable,
+    is_int64_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_interval_dtype,
+    is_iterator,
+    is_list_like,
+    is_named_tuple,
+    is_number,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_period_dtype,
+    is_re,
+    is_re_compilable,
+    is_scalar,
+    is_signed_integer_dtype,
+    is_sparse,
+    is_string_dtype,
+    is_timedelta64_dtype,
+    is_timedelta64_ns_dtype,
+    is_unsigned_integer_dtype,
+    pandas_dtype,
+)
+
+__all__ = [
+    "is_any_real_numeric_dtype",
+    "is_array_like",
+    "is_bool",
+    "is_bool_dtype",
+    "is_categorical_dtype",
+    "is_complex",
+    "is_complex_dtype",
+    "is_datetime64_any_dtype",
+    "is_datetime64_dtype",
+    "is_datetime64_ns_dtype",
+    "is_datetime64tz_dtype",
+    "is_dict_like",
+    "is_dtype_equal",
+    "is_extension_array_dtype",
+    "is_file_like",
+    "is_float",
+    "is_float_dtype",
+    "is_hashable",
+    "is_int64_dtype",
+    "is_integer",
+    "is_integer_dtype",
+    "is_interval_dtype",
+    "is_iterator",
+    "is_list_like",
+    "is_named_tuple",
+    "is_number",
+    "is_numeric_dtype",
+    "is_object_dtype",
+    "is_period_dtype",
+    "is_re",
+    "is_re_compilable",
+    "is_scalar",
+    "is_signed_integer_dtype",
+    "is_sparse",
+    "is_string_dtype",
+    "is_timedelta64_dtype",
+    "is_timedelta64_ns_dtype",
+    "is_unsigned_integer_dtype",
+    "pandas_dtype",
+]
diff --git a/pandas/pandas/core/dtypes/astype.py b/pandas/pandas/core/dtypes/astype.py
new file mode 100644
index 0000000000000..086f7d2da6640
--- /dev/null
+++ b/pandas/pandas/core/dtypes/astype.py
@@ -0,0 +1,303 @@
+"""
+Functions for implementing 'astype' methods according to pandas conventions,
+particularly ones that differ from numpy.
+"""
+
+from __future__ import annotations
+
+import inspect
+from typing import (
+    TYPE_CHECKING,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs.timedeltas import array_to_timedelta64
+from pandas.errors import IntCastingNaNError
+
+from pandas.core.dtypes.common import (
+    is_object_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ExtensionDtype,
+    NumpyEADtype,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        DtypeObj,
+        IgnoreRaise,
+    )
+
+    from pandas.core.arrays import ExtensionArray
+
+
+@overload
+def _astype_nansafe(
+    arr: np.ndarray, dtype: np.dtype, copy: bool = ..., skipna: bool = ...
+) -> np.ndarray: ...
+
+
+@overload
+def _astype_nansafe(
+    arr: np.ndarray, dtype: ExtensionDtype, copy: bool = ..., skipna: bool = ...
+) -> ExtensionArray: ...
+
+
+def _astype_nansafe(
+    arr: np.ndarray, dtype: DtypeObj, copy: bool = True, skipna: bool = False
+) -> ArrayLike:
+    """
+    Cast the elements of an array to a given dtype a nan-safe manner.
+
+    Parameters
+    ----------
+    arr : ndarray
+    dtype : np.dtype or ExtensionDtype
+    copy : bool, default True
+        If False, a view will be attempted but may fail, if
+        e.g. the item sizes don't align.
+    skipna: bool, default False
+        Whether or not we should skip NaN when casting as a string-type.
+
+    Raises
+    ------
+    ValueError
+        The dtype was a datetime64/timedelta64 dtype, but it had no unit.
+    """
+
+    # dispatch on extension dtype if needed
+    if isinstance(dtype, ExtensionDtype):
+        return dtype.construct_array_type()._from_sequence(arr, dtype=dtype, copy=copy)
+
+    elif not isinstance(dtype, np.dtype):  # pragma: no cover
+        raise ValueError("dtype must be np.dtype or ExtensionDtype")
+
+    if arr.dtype.kind in "mM":
+        from pandas.core.construction import ensure_wrapped_if_datetimelike
+
+        arr = ensure_wrapped_if_datetimelike(arr)
+        res = arr.astype(dtype, copy=copy)
+        return np.asarray(res)
+
+    if issubclass(dtype.type, str):
+        shape = arr.shape
+        if arr.ndim > 1:
+            arr = arr.ravel()
+        return lib.ensure_string_array(
+            arr, skipna=skipna, convert_na_value=False
+        ).reshape(shape)
+
+    elif np.issubdtype(arr.dtype, np.floating) and dtype.kind in "iu":
+        return _astype_float_to_int_nansafe(arr, dtype, copy)
+
+    elif arr.dtype == object:
+        # if we have a datetime/timedelta array of objects
+        # then coerce to datetime64[ns] and use DatetimeArray.astype
+
+        if lib.is_np_dtype(dtype, "M"):
+            from pandas.core.arrays import DatetimeArray
+
+            dta = DatetimeArray._from_sequence(arr, dtype=dtype)
+            return dta._ndarray
+
+        elif lib.is_np_dtype(dtype, "m"):
+            from pandas.core.construction import ensure_wrapped_if_datetimelike
+
+            # bc we know arr.dtype == object, this is equivalent to
+            #  `np.asarray(to_timedelta(arr))`, but using a lower-level API that
+            #  does not require a circular import.
+            tdvals = array_to_timedelta64(arr).view("m8[ns]")
+
+            tda = ensure_wrapped_if_datetimelike(tdvals)
+            return tda.astype(dtype, copy=False)._ndarray
+
+    if dtype.name in ("datetime64", "timedelta64"):
+        msg = (
+            f"The '{dtype.name}' dtype has no unit. Please pass in "
+            f"'{dtype.name}[ns]' instead."
+        )
+        raise ValueError(msg)
+
+    if copy or arr.dtype == object or dtype == object:
+        # Explicit copy, or required since NumPy can't view from / to object.
+        return arr.astype(dtype, copy=True)
+
+    return arr.astype(dtype, copy=copy)
+
+
+def _astype_float_to_int_nansafe(
+    values: np.ndarray, dtype: np.dtype, copy: bool
+) -> np.ndarray:
+    """
+    astype with a check preventing converting NaN to an meaningless integer value.
+    """
+    if not np.isfinite(values).all():
+        raise IntCastingNaNError(
+            "Cannot convert non-finite values (NA or inf) to integer"
+        )
+    if dtype.kind == "u":
+        # GH#45151
+        if not (values >= 0).all():
+            raise ValueError(f"Cannot losslessly cast from {values.dtype} to {dtype}")
+    with warnings.catch_warnings():
+        warnings.filterwarnings("ignore", category=RuntimeWarning)
+        return values.astype(dtype, copy=copy)
+
+
+def astype_array(values: ArrayLike, dtype: DtypeObj, copy: bool = False) -> ArrayLike:
+    """
+    Cast array (ndarray or ExtensionArray) to the new dtype.
+
+    Parameters
+    ----------
+    values : ndarray or ExtensionArray
+    dtype : dtype object
+    copy : bool, default False
+        copy if indicated
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+    if values.dtype == dtype:
+        if copy:
+            return values.copy()
+        return values
+
+    if not isinstance(values, np.ndarray):
+        # i.e. ExtensionArray
+        values = values.astype(dtype, copy=copy)
+
+    else:
+        values = _astype_nansafe(values, dtype, copy=copy)
+
+    # in pandas we don't store numpy str dtypes, so convert to object
+    if isinstance(dtype, np.dtype) and issubclass(values.dtype.type, str):
+        values = np.array(values, dtype=object)
+
+    return values
+
+
+def astype_array_safe(
+    values: ArrayLike, dtype, copy: bool = False, errors: IgnoreRaise = "raise"
+) -> ArrayLike:
+    """
+    Cast array (ndarray or ExtensionArray) to the new dtype.
+
+    This basically is the implementation for DataFrame/Series.astype and
+    includes all custom logic for pandas (NaN-safety, converting str to object,
+    not allowing )
+
+    Parameters
+    ----------
+    values : ndarray or ExtensionArray
+    dtype : str, dtype convertible
+    copy : bool, default False
+        copy if indicated
+    errors : str, {'raise', 'ignore'}, default 'raise'
+        - ``raise`` : allow exceptions to be raised
+        - ``ignore`` : suppress exceptions. On error return original object
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+    errors_legal_values = ("raise", "ignore")
+
+    if errors not in errors_legal_values:
+        invalid_arg = (
+            "Expected value of kwarg 'errors' to be one of "
+            f"{list(errors_legal_values)}. Supplied value is '{errors}'"
+        )
+        raise ValueError(invalid_arg)
+
+    if inspect.isclass(dtype) and issubclass(dtype, ExtensionDtype):
+        msg = (
+            f"Expected an instance of {dtype.__name__}, "
+            "but got the class instead. Try instantiating 'dtype'."
+        )
+        raise TypeError(msg)
+
+    dtype = pandas_dtype(dtype)
+    if isinstance(dtype, NumpyEADtype):
+        # Ensure we don't end up with a NumpyExtensionArray
+        dtype = dtype.numpy_dtype
+
+    try:
+        new_values = astype_array(values, dtype, copy=copy)
+    except (ValueError, TypeError):
+        # e.g. _astype_nansafe can fail on object-dtype of strings
+        #  trying to convert to float
+        if errors == "ignore":
+            new_values = values
+        else:
+            raise
+
+    return new_values
+
+
+def astype_is_view(dtype: DtypeObj, new_dtype: DtypeObj) -> bool:
+    """Checks if astype avoided copying the data.
+
+    Parameters
+    ----------
+    dtype : Original dtype
+    new_dtype : target dtype
+
+    Returns
+    -------
+    True if new data is a view or not guaranteed to be a copy, False otherwise
+    """
+    if dtype.kind in "iufb" and dtype.kind == new_dtype.kind:
+        # fastpath for numeric dtypes
+        if hasattr(dtype, "itemsize") and hasattr(new_dtype, "itemsize"):
+            return dtype.itemsize == new_dtype.itemsize  # pyright: ignore[reportAttributeAccessIssue]
+
+    if isinstance(dtype, np.dtype) and not isinstance(new_dtype, np.dtype):
+        new_dtype, dtype = dtype, new_dtype
+
+    if dtype == new_dtype:
+        return True
+
+    elif isinstance(dtype, np.dtype) and isinstance(new_dtype, np.dtype):
+        # Only equal numpy dtypes avoid a copy
+        return False
+
+    elif is_string_dtype(dtype) and is_string_dtype(new_dtype):
+        # Potentially! a view when converting from object to string
+        return True
+
+    elif is_object_dtype(dtype) and new_dtype.kind == "O":
+        # When the underlying array has dtype object, we don't have to make a copy
+        return True
+
+    elif dtype.kind in "mM" and new_dtype.kind in "mM":
+        dtype = getattr(dtype, "numpy_dtype", dtype)
+        new_dtype = getattr(new_dtype, "numpy_dtype", new_dtype)
+        return getattr(dtype, "unit", None) == getattr(new_dtype, "unit", None)
+
+    numpy_dtype = getattr(dtype, "numpy_dtype", None)
+    new_numpy_dtype = getattr(new_dtype, "numpy_dtype", None)
+
+    if numpy_dtype is None and isinstance(dtype, np.dtype):
+        numpy_dtype = dtype
+
+    if new_numpy_dtype is None and isinstance(new_dtype, np.dtype):
+        new_numpy_dtype = new_dtype
+
+    if numpy_dtype is not None and new_numpy_dtype is not None:
+        # if both have NumPy dtype or one of them is a numpy dtype
+        # they are only a view when the numpy dtypes are equal, e.g.
+        # int64 -> Int64 or int64[pyarrow]
+        # int64 -> Int32 copies
+        return numpy_dtype == new_numpy_dtype
+
+    # Assume this is a view since we don't know for sure if a copy was made
+    return True
diff --git a/pandas/pandas/core/dtypes/base.py b/pandas/pandas/core/dtypes/base.py
new file mode 100644
index 0000000000000..d8a42d83b6c54
--- /dev/null
+++ b/pandas/pandas/core/dtypes/base.py
@@ -0,0 +1,586 @@
+"""
+Extend pandas with custom array types.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypeVar,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import missing as libmissing
+from pandas._libs.hashtable import object_hash
+from pandas._libs.properties import cache_readonly
+from pandas.errors import AbstractMethodError
+
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        DtypeObj,
+        Self,
+        Shape,
+        npt,
+        type_t,
+    )
+
+    from pandas import Index
+    from pandas.core.arrays import ExtensionArray
+
+    # To parameterize on same ExtensionDtype
+    ExtensionDtypeT = TypeVar("ExtensionDtypeT", bound="ExtensionDtype")
+
+
+class ExtensionDtype:
+    """
+    A custom data type, to be paired with an ExtensionArray.
+
+    See Also
+    --------
+    extensions.register_extension_dtype: Register an ExtensionType
+        with pandas as class decorator.
+    extensions.ExtensionArray: Abstract base class for custom 1-D array types.
+
+    Notes
+    -----
+    The interface includes the following abstract methods that must
+    be implemented by subclasses:
+
+    * type
+    * name
+    * construct_array_type
+
+    The following attributes and methods influence the behavior of the dtype in
+    pandas operations
+
+    * _is_numeric
+    * _is_boolean
+    * _get_common_dtype
+
+    The `na_value` class attribute can be used to set the default NA value
+    for this type. :attr:`numpy.nan` is used by default.
+
+    ExtensionDtypes are required to be hashable. The base class provides
+    a default implementation, which relies on the ``_metadata`` class
+    attribute. ``_metadata`` should be a tuple containing the strings
+    that define your data type. For example, with ``PeriodDtype`` that's
+    the ``freq`` attribute.
+
+    **If you have a parametrized dtype you should set the ``_metadata``
+    class property**.
+
+    Ideally, the attributes in ``_metadata`` will match the
+    parameters to your ``ExtensionDtype.__init__`` (if any). If any of
+    the attributes in ``_metadata`` don't implement the standard
+    ``__eq__`` or ``__hash__``, the default implementations here will not
+    work.
+
+    Examples
+    --------
+
+    For interaction with Apache Arrow (pyarrow), a ``__from_arrow__`` method
+    can be implemented: this method receives a pyarrow Array or ChunkedArray
+    as only argument and is expected to return the appropriate pandas
+    ExtensionArray for this dtype and the passed values:
+
+    >>> import pyarrow
+    >>> from pandas.api.extensions import ExtensionArray
+    >>> class ExtensionDtype:
+    ...     def __from_arrow__(
+    ...         self, array: pyarrow.Array | pyarrow.ChunkedArray
+    ...     ) -> ExtensionArray: ...
+
+    This class does not inherit from 'abc.ABCMeta' for performance reasons.
+    Methods and properties required by the interface raise
+    ``pandas.errors.AbstractMethodError`` and no ``register`` method is
+    provided for registering virtual subclasses.
+    """
+
+    _metadata: tuple[str, ...] = ()
+
+    def __str__(self) -> str:
+        return self.name
+
+    def __eq__(self, other: object) -> bool:
+        """
+        Check whether 'other' is equal to self.
+
+        By default, 'other' is considered equal if either
+
+        * it's a string matching 'self.name'.
+        * it's an instance of this type and all of the attributes
+          in ``self._metadata`` are equal between `self` and `other`.
+
+        Parameters
+        ----------
+        other : Any
+
+        Returns
+        -------
+        bool
+        """
+        if isinstance(other, str):
+            try:
+                other = self.construct_from_string(other)
+            except TypeError:
+                return False
+        if isinstance(other, type(self)):
+            return all(
+                getattr(self, attr) == getattr(other, attr) for attr in self._metadata
+            )
+        return False
+
+    def __hash__(self) -> int:
+        # for python>=3.10, different nan objects have different hashes
+        # we need to avoid that and thus use hash function with old behavior
+        return object_hash(tuple(getattr(self, attr) for attr in self._metadata))
+
+    def __ne__(self, other: object) -> bool:
+        return not self.__eq__(other)
+
+    @property
+    def na_value(self) -> object:
+        """
+        Default NA value to use for this type.
+
+        This is used in e.g. ExtensionArray.take. This should be the
+        user-facing "boxed" version of the NA value, not the physical NA value
+        for storage.  e.g. for JSONArray, this is an empty dictionary.
+        """
+        return np.nan
+
+    @property
+    def type(self) -> type_t[Any]:
+        """
+        The scalar type for the array, e.g. ``int``
+
+        It's expected ``ExtensionArray[item]`` returns an instance
+        of ``ExtensionDtype.type`` for scalar ``item``, assuming
+        that value is valid (not NA). NA values do not need to be
+        instances of `type`.
+        """
+        raise AbstractMethodError(self)
+
+    @property
+    def kind(self) -> str:
+        """
+        A character code (one of 'biufcmMOSUV'), default 'O'
+
+        This should match the NumPy dtype used when the array is
+        converted to an ndarray, which is probably 'O' for object if
+        the extension type cannot be represented as a built-in NumPy
+        type.
+
+        See Also
+        --------
+        numpy.dtype.kind
+        """
+        return "O"
+
+    @property
+    def name(self) -> str:
+        """
+        A string identifying the data type.
+
+        Will be used for display in, e.g. ``Series.dtype``
+        """
+        raise AbstractMethodError(self)
+
+    @property
+    def names(self) -> list[str] | None:
+        """
+        Ordered list of field names, or None if there are no fields.
+
+        This is for compatibility with NumPy arrays, and may be removed in the
+        future.
+        """
+        return None
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[ExtensionArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        raise AbstractMethodError(cls)
+
+    def empty(self, shape: Shape) -> ExtensionArray:
+        """
+        Construct an ExtensionArray of this dtype with the given shape.
+
+        Analogous to numpy.empty.
+
+        Parameters
+        ----------
+        shape : int or tuple[int]
+
+        Returns
+        -------
+        ExtensionArray
+        """
+        cls = self.construct_array_type()
+        return cls._empty(shape, dtype=self)
+
+    @classmethod
+    def construct_from_string(cls, string: str) -> Self:
+        r"""
+        Construct this type from a string.
+
+        This is useful mainly for data types that accept parameters.
+        For example, a period dtype accepts a frequency parameter that
+        can be set as ``period[h]`` (where H means hourly frequency).
+
+        By default, in the abstract class, just the name of the type is
+        expected. But subclasses can overwrite this method to accept
+        parameters.
+
+        Parameters
+        ----------
+        string : str
+            The name of the type, for example ``category``.
+
+        Returns
+        -------
+        ExtensionDtype
+            Instance of the dtype.
+
+        Raises
+        ------
+        TypeError
+            If a class cannot be constructed from this 'string'.
+
+        Examples
+        --------
+        For extension dtypes with arguments the following may be an
+        adequate implementation.
+
+        >>> import re
+        >>> @classmethod
+        ... def construct_from_string(cls, string):
+        ...     pattern = re.compile(r"^my_type\[(?P<arg_name>.+)\]$")
+        ...     match = pattern.match(string)
+        ...     if match:
+        ...         return cls(**match.groupdict())
+        ...     else:
+        ...         raise TypeError(
+        ...             f"Cannot construct a '{cls.__name__}' from '{string}'"
+        ...         )
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+        # error: Non-overlapping equality check (left operand type: "str", right
+        #  operand type: "Callable[[ExtensionDtype], str]")  [comparison-overlap]
+        assert isinstance(cls.name, str), (cls, type(cls.name))
+        if string != cls.name:
+            raise TypeError(f"Cannot construct a '{cls.__name__}' from '{string}'")
+        return cls()
+
+    @classmethod
+    def is_dtype(cls, dtype: object) -> bool:
+        """
+        Check if we match 'dtype'.
+
+        Parameters
+        ----------
+        dtype : object
+            The object to check.
+
+        Returns
+        -------
+        bool
+
+        Notes
+        -----
+        The default implementation is True if
+
+        1. ``cls.construct_from_string(dtype)`` is an instance
+           of ``cls``.
+        2. ``dtype`` is an object and is an instance of ``cls``
+        3. ``dtype`` has a ``dtype`` attribute, and any of the above
+           conditions is true for ``dtype.dtype``.
+        """
+        dtype = getattr(dtype, "dtype", dtype)
+
+        if isinstance(dtype, (ABCSeries, ABCIndex, ABCDataFrame, np.dtype)):
+            # https://github.com/pandas-dev/pandas/issues/22960
+            # avoid passing data to `construct_from_string`. This could
+            # cause a FutureWarning from numpy about failing elementwise
+            # comparison from, e.g., comparing DataFrame == 'category'.
+            return False
+        elif dtype is None:
+            return False
+        elif isinstance(dtype, cls):
+            return True
+        if isinstance(dtype, str):
+            try:
+                return cls.construct_from_string(dtype) is not None
+            except TypeError:
+                return False
+        return False
+
+    @property
+    def _is_numeric(self) -> bool:
+        """
+        Whether columns with this dtype should be considered numeric.
+
+        By default ExtensionDtypes are assumed to be non-numeric.
+        They'll be excluded from operations that exclude non-numeric
+        columns, like (groupby) reductions, plotting, etc.
+        """
+        return False
+
+    @property
+    def _is_boolean(self) -> bool:
+        """
+        Whether this dtype should be considered boolean.
+
+        By default, ExtensionDtypes are assumed to be non-numeric.
+        Setting this to True will affect the behavior of several places,
+        e.g.
+
+        * is_bool
+        * boolean indexing
+
+        Returns
+        -------
+        bool
+        """
+        return False
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        """
+        Return the common dtype, if one exists.
+
+        Used in `find_common_type` implementation. This is for example used
+        to determine the resulting dtype in a concat operation.
+
+        If no common dtype exists, return None (which gives the other dtypes
+        the chance to determine a common dtype). If all dtypes in the list
+        return None, then the common dtype will be "object" dtype (this means
+        it is never needed to return "object" dtype from this method itself).
+
+        Parameters
+        ----------
+        dtypes : list of dtypes
+            The dtypes for which to determine a common dtype. This is a list
+            of np.dtype or ExtensionDtype instances.
+
+        Returns
+        -------
+        Common dtype (np.dtype or ExtensionDtype) or None
+        """
+        if len(set(dtypes)) == 1:
+            # only itself
+            return self
+        else:
+            return None
+
+    @property
+    def _can_hold_na(self) -> bool:
+        """
+        Can arrays of this dtype hold NA values?
+        """
+        return True
+
+    @property
+    def _is_immutable(self) -> bool:
+        """
+        Can arrays with this dtype be modified with __setitem__? If not, return
+        True.
+
+        Immutable arrays are expected to raise TypeError on __setitem__ calls.
+        """
+        return False
+
+    @cache_readonly
+    def index_class(self) -> type_t[Index]:
+        """
+        The Index subclass to return from Index.__new__ when this dtype is
+        encountered.
+        """
+        from pandas import Index
+
+        return Index
+
+    @property
+    def _supports_2d(self) -> bool:
+        """
+        Do ExtensionArrays with this dtype support 2D arrays?
+
+        Historically ExtensionArrays were limited to 1D. By returning True here,
+        authors can indicate that their arrays support 2D instances. This can
+        improve performance in some cases, particularly operations with `axis=1`.
+
+        Arrays that support 2D values should:
+
+            - implement Array.reshape
+            - subclass the Dim2CompatTests in tests.extension.base
+            - _concat_same_type should support `axis` keyword
+            - _reduce and reductions should support `axis` keyword
+        """
+        return False
+
+    @property
+    def _can_fast_transpose(self) -> bool:
+        """
+        Is transposing an array with this dtype zero-copy?
+
+        Only relevant for cases where _supports_2d is True.
+        """
+        return False
+
+
+class StorageExtensionDtype(ExtensionDtype):
+    """ExtensionDtype that may be backed by more than one implementation."""
+
+    name: str
+    _metadata = ("storage",)
+
+    def __init__(self, storage: str | None = None) -> None:
+        self.storage = storage
+
+    def __repr__(self) -> str:
+        return f"{self.name}[{self.storage}]"
+
+    def __str__(self) -> str:
+        return self.name
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, str) and other == self.name:
+            return True
+        return super().__eq__(other)
+
+    def __hash__(self) -> int:
+        # custom __eq__ so have to override __hash__
+        return super().__hash__()
+
+    @property
+    def na_value(self) -> libmissing.NAType:
+        return libmissing.NA
+
+
+def register_extension_dtype(cls: type_t[ExtensionDtypeT]) -> type_t[ExtensionDtypeT]:
+    """
+    Register an ExtensionType with pandas as class decorator.
+
+    This enables operations like ``.astype(name)`` for the name
+    of the ExtensionDtype.
+
+    Returns
+    -------
+    callable
+        A class decorator.
+
+    See Also
+    --------
+    api.extensions.ExtensionDtype : The base class for creating custom pandas
+        data types.
+    Series : One-dimensional array with axis labels.
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous
+        tabular data.
+
+    Examples
+    --------
+    >>> from pandas.api.extensions import register_extension_dtype, ExtensionDtype
+    >>> @register_extension_dtype
+    ... class MyExtensionDtype(ExtensionDtype):
+    ...     name = "myextension"
+    """
+    _registry.register(cls)
+    return cls
+
+
+class Registry:
+    """
+    Registry for dtype inference.
+
+    The registry allows one to map a string repr of a extension
+    dtype to an extension dtype. The string alias can be used in several
+    places, including
+
+    * Series and Index constructors
+    * :meth:`pandas.array`
+    * :meth:`pandas.Series.astype`
+
+    Multiple extension types can be registered.
+    These are tried in order.
+    """
+
+    def __init__(self) -> None:
+        self.dtypes: list[type_t[ExtensionDtype]] = []
+
+    def register(self, dtype: type_t[ExtensionDtype]) -> None:
+        """
+        Parameters
+        ----------
+        dtype : ExtensionDtype class
+        """
+        if not issubclass(dtype, ExtensionDtype):
+            raise ValueError("can only register pandas extension dtypes")
+
+        self.dtypes.append(dtype)
+
+    @overload
+    def find(self, dtype: type_t[ExtensionDtypeT]) -> type_t[ExtensionDtypeT]: ...
+
+    @overload
+    def find(self, dtype: ExtensionDtypeT) -> ExtensionDtypeT: ...
+
+    @overload
+    def find(self, dtype: str) -> ExtensionDtype | None: ...
+
+    @overload
+    def find(
+        self, dtype: npt.DTypeLike
+    ) -> type_t[ExtensionDtype] | ExtensionDtype | None: ...
+
+    def find(
+        self, dtype: type_t[ExtensionDtype] | ExtensionDtype | npt.DTypeLike
+    ) -> type_t[ExtensionDtype] | ExtensionDtype | None:
+        """
+        Parameters
+        ----------
+        dtype : ExtensionDtype class or instance or str or numpy dtype or python type
+
+        Returns
+        -------
+        return the first matching dtype, otherwise return None
+        """
+        if not isinstance(dtype, str):
+            dtype_type: type_t
+            if not isinstance(dtype, type):
+                dtype_type = type(dtype)
+            else:
+                dtype_type = dtype
+            if issubclass(dtype_type, ExtensionDtype):
+                # cast needed here as mypy doesn't know we have figured
+                # out it is an ExtensionDtype or type_t[ExtensionDtype]
+                return cast("ExtensionDtype | type_t[ExtensionDtype]", dtype)
+
+            return None
+
+        for dtype_type in self.dtypes:
+            try:
+                return dtype_type.construct_from_string(dtype)
+            except TypeError:
+                pass
+
+        return None
+
+
+_registry = Registry()
diff --git a/pandas/pandas/core/dtypes/cast.py b/pandas/pandas/core/dtypes/cast.py
new file mode 100644
index 0000000000000..f11aefeeaaa00
--- /dev/null
+++ b/pandas/pandas/core/dtypes/cast.py
@@ -0,0 +1,1980 @@
+"""
+Routines for casting.
+"""
+
+from __future__ import annotations
+
+import datetime as dt
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    TypeVar,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import (
+    Interval,
+    Period,
+    lib,
+)
+from pandas._libs.missing import (
+    NA,
+    NAType,
+    checknull,
+)
+from pandas._libs.tslibs import (
+    NaT,
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+    Timedelta,
+    Timestamp,
+    is_supported_dtype,
+)
+from pandas._libs.tslibs.timedeltas import array_to_timedelta64
+from pandas.errors import (
+    IntCastingNaNError,
+    LossySetitemError,
+)
+
+from pandas.core.dtypes.common import (
+    ensure_int8,
+    ensure_int16,
+    ensure_int32,
+    ensure_int64,
+    ensure_object,
+    ensure_str,
+    is_bool,
+    is_complex,
+    is_float,
+    is_integer,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype as pandas_dtype_func,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    BaseMaskedDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    PandasExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCExtensionArray,
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import is_list_like
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+    na_value_for_dtype,
+    notna,
+)
+
+from pandas.io._util import _arrow_dtype_mapping
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Collection,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        Dtype,
+        DtypeObj,
+        NumpyIndexT,
+        Scalar,
+        npt,
+    )
+
+    from pandas import Index
+    from pandas.core.arrays import (
+        Categorical,
+        DatetimeArray,
+        ExtensionArray,
+        IntervalArray,
+        PeriodArray,
+        TimedeltaArray,
+    )
+
+
+_int8_max = np.iinfo(np.int8).max
+_int16_max = np.iinfo(np.int16).max
+_int32_max = np.iinfo(np.int32).max
+
+_dtype_obj = np.dtype(object)
+
+NumpyArrayT = TypeVar("NumpyArrayT", bound=np.ndarray)
+
+
+def maybe_convert_platform(
+    values: list | tuple | range | np.ndarray | ExtensionArray,
+) -> ArrayLike:
+    """try to do platform conversion, allow ndarray or list here"""
+    arr: ArrayLike
+
+    if isinstance(values, (list, tuple, range)):
+        arr = construct_1d_object_array_from_listlike(values)
+    else:
+        # The caller is responsible for ensuring that we have np.ndarray
+        #  or ExtensionArray here.
+        arr = values
+
+    if arr.dtype == _dtype_obj:
+        arr = cast(np.ndarray, arr)
+        arr = lib.maybe_convert_objects(arr)
+
+    return arr
+
+
+def is_nested_object(obj) -> bool:
+    """
+    return a boolean if we have a nested object, e.g. a Series with 1 or
+    more Series elements
+
+    This may not be necessarily be performant.
+
+    """
+    return bool(
+        isinstance(obj, ABCSeries)
+        and is_object_dtype(obj.dtype)
+        and any(isinstance(v, ABCSeries) for v in obj._values)
+    )
+
+
+def maybe_box_datetimelike(value: Scalar, dtype: Dtype | None = None) -> Scalar:
+    """
+    Cast scalar to Timestamp or Timedelta if scalar is datetime-like
+    and dtype is not object.
+
+    Parameters
+    ----------
+    value : scalar
+    dtype : Dtype, optional
+
+    Returns
+    -------
+    scalar
+    """
+    if dtype == _dtype_obj:
+        pass
+    elif isinstance(value, (np.datetime64, dt.datetime)):
+        value = Timestamp(value)
+    elif isinstance(value, (np.timedelta64, dt.timedelta)):
+        value = Timedelta(value)
+
+    return value
+
+
+def maybe_box_native(value: Scalar | None | NAType) -> Scalar | None | NAType:
+    """
+    If passed a scalar cast the scalar to a python native type.
+
+    Parameters
+    ----------
+    value : scalar or Series
+
+    Returns
+    -------
+    scalar or Series
+    """
+    if is_float(value):
+        value = float(value)
+    elif is_integer(value):
+        value = int(value)
+    elif is_bool(value):
+        value = bool(value)
+    elif isinstance(value, (np.datetime64, np.timedelta64)):
+        value = maybe_box_datetimelike(value)
+    elif value is NA:
+        value = None
+    return value
+
+
+def _maybe_unbox_datetimelike(value: Scalar, dtype: DtypeObj) -> Scalar:
+    """
+    Convert a Timedelta or Timestamp to timedelta64 or datetime64 for setting
+    into a numpy array.  Failing to unbox would risk dropping nanoseconds.
+
+    Notes
+    -----
+    Caller is responsible for checking dtype.kind in "mM"
+    """
+    if is_valid_na_for_dtype(value, dtype):
+        # GH#36541: can't fill array directly with pd.NaT
+        # > np.empty(10, dtype="datetime64[ns]").fill(pd.NaT)
+        # ValueError: cannot convert float NaN to integer
+        value = dtype.type("NaT", "ns")
+    elif isinstance(value, Timestamp):
+        if value.tz is None:
+            value = value.to_datetime64()
+        elif not isinstance(dtype, DatetimeTZDtype):
+            raise TypeError("Cannot unbox tzaware Timestamp to tznaive dtype")
+    elif isinstance(value, Timedelta):
+        value = value.to_timedelta64()
+
+    _disallow_mismatched_datetimelike(value, dtype)
+    return value
+
+
+def _disallow_mismatched_datetimelike(value, dtype: DtypeObj) -> None:
+    """
+    numpy allows np.array(dt64values, dtype="timedelta64[ns]") and
+    vice-versa, but we do not want to allow this, so we need to
+    check explicitly
+    """
+    vdtype = getattr(value, "dtype", None)
+    if vdtype is None:
+        return
+    elif (vdtype.kind == "m" and dtype.kind == "M") or (
+        vdtype.kind == "M" and dtype.kind == "m"
+    ):
+        raise TypeError(f"Cannot cast {value!r} to {dtype}")
+
+
+@overload
+def maybe_downcast_to_dtype(
+    result: np.ndarray, dtype: str | np.dtype
+) -> np.ndarray: ...
+
+
+@overload
+def maybe_downcast_to_dtype(
+    result: ExtensionArray, dtype: str | np.dtype
+) -> ArrayLike: ...
+
+
+def maybe_downcast_to_dtype(result: ArrayLike, dtype: str | np.dtype) -> ArrayLike:
+    """
+    try to cast to the specified dtype (e.g. convert back to bool/int
+    or could be an astype of float64->float32
+    """
+    if isinstance(result, ABCSeries):
+        result = result._values
+    do_round = False
+
+    if isinstance(dtype, str):
+        if dtype == "infer":
+            inferred_type = lib.infer_dtype(result, skipna=False)
+            if inferred_type == "boolean":
+                dtype = "bool"
+            elif inferred_type == "integer":
+                dtype = "int64"
+            elif inferred_type == "datetime64":
+                dtype = "datetime64[ns]"
+            elif inferred_type in ["timedelta", "timedelta64"]:
+                dtype = "timedelta64[ns]"
+
+            # try to upcast here
+            elif inferred_type == "floating":
+                dtype = "int64"
+                if issubclass(result.dtype.type, np.number):
+                    do_round = True
+
+            else:
+                # TODO: complex?  what if result is already non-object?
+                dtype = "object"
+
+        dtype = np.dtype(dtype)
+
+    if not isinstance(dtype, np.dtype):
+        # enforce our signature annotation
+        raise TypeError(dtype)  # pragma: no cover
+
+    converted = maybe_downcast_numeric(result, dtype, do_round)
+    if converted is not result:
+        return converted
+
+    # a datetimelike
+    # GH12821, iNaT is cast to float
+    if dtype.kind in "mM" and result.dtype.kind in "if":
+        result = result.astype(dtype)
+
+    elif dtype.kind == "m" and result.dtype == _dtype_obj:
+        # test_where_downcast_to_td64
+        result = cast(np.ndarray, result)
+        result = array_to_timedelta64(result)
+
+    elif dtype == np.dtype("M8[ns]") and result.dtype == _dtype_obj:
+        result = cast(np.ndarray, result)
+        return np.asarray(maybe_cast_to_datetime(result, dtype=dtype))
+
+    return result
+
+
+@overload
+def maybe_downcast_numeric(
+    result: np.ndarray, dtype: np.dtype, do_round: bool = False
+) -> np.ndarray: ...
+
+
+@overload
+def maybe_downcast_numeric(
+    result: ExtensionArray, dtype: DtypeObj, do_round: bool = False
+) -> ArrayLike: ...
+
+
+def maybe_downcast_numeric(
+    result: ArrayLike, dtype: DtypeObj, do_round: bool = False
+) -> ArrayLike:
+    """
+    Subset of maybe_downcast_to_dtype restricted to numeric dtypes.
+
+    Parameters
+    ----------
+    result : ndarray or ExtensionArray
+    dtype : np.dtype or ExtensionDtype
+    do_round : bool
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+    if not isinstance(dtype, np.dtype) or not isinstance(result.dtype, np.dtype):
+        # e.g. SparseDtype has no itemsize attr
+        return result
+
+    def trans(x):
+        if do_round:
+            return x.round()
+        return x
+
+    if dtype.kind == result.dtype.kind:
+        # don't allow upcasts here (except if empty)
+        if result.dtype.itemsize <= dtype.itemsize and result.size:
+            return result
+
+    if dtype.kind in "biu":
+        if not result.size:
+            # if we don't have any elements, just astype it
+            return trans(result).astype(dtype)
+
+        if isinstance(result, np.ndarray):
+            element = result.item(0)
+        else:
+            element = result.iloc[0]
+        if not isinstance(element, (np.integer, np.floating, int, float, bool)):
+            # a comparable, e.g. a Decimal may slip in here
+            return result
+
+        if (
+            issubclass(result.dtype.type, (np.object_, np.number))
+            and notna(result).all()
+        ):
+            new_result = trans(result).astype(dtype)
+            if new_result.dtype.kind == "O" or result.dtype.kind == "O":
+                # np.allclose may raise TypeError on object-dtype
+                if (new_result == result).all():
+                    return new_result
+            else:
+                if np.allclose(new_result, result, rtol=0):
+                    return new_result
+
+    elif (
+        issubclass(dtype.type, np.floating)
+        and result.dtype.kind != "b"
+        and not is_string_dtype(result.dtype)
+    ):
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore", "overflow encountered in cast", RuntimeWarning
+            )
+            new_result = result.astype(dtype)
+
+        # Adjust tolerances based on floating point size
+        size_tols = {4: 5e-4, 8: 5e-8, 16: 5e-16}
+
+        atol = size_tols.get(new_result.dtype.itemsize, 0.0)
+
+        # Check downcast float values are still equal within 7 digits when
+        # converting from float64 to float32
+        if np.allclose(new_result, result, equal_nan=True, rtol=0.0, atol=atol):
+            return new_result
+
+    elif dtype.kind == result.dtype.kind == "c":
+        new_result = result.astype(dtype)
+
+        if np.array_equal(new_result, result, equal_nan=True):
+            # TODO: use tolerance like we do for float?
+            return new_result
+
+    return result
+
+
+def maybe_upcast_numeric_to_64bit(arr: NumpyIndexT) -> NumpyIndexT:
+    """
+    If array is a int/uint/float bit size lower than 64 bit, upcast it to 64 bit.
+
+    Parameters
+    ----------
+    arr : ndarray or ExtensionArray
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+    dtype = arr.dtype
+    if dtype.kind == "i" and dtype != np.int64:
+        return arr.astype(np.int64)
+    elif dtype.kind == "u" and dtype != np.uint64:
+        return arr.astype(np.uint64)
+    elif dtype.kind == "f" and dtype != np.float64:
+        return arr.astype(np.float64)
+    else:
+        return arr
+
+
+def maybe_cast_pointwise_result(
+    result: ArrayLike,
+    dtype: DtypeObj,
+    numeric_only: bool = False,
+    same_dtype: bool = True,
+) -> ArrayLike:
+    """
+    Try casting result of a pointwise operation back to the original dtype if
+    appropriate.
+
+    Parameters
+    ----------
+    result : array-like
+        Result to cast.
+    dtype : np.dtype or ExtensionDtype
+        Input Series from which result was calculated.
+    numeric_only : bool, default False
+        Whether to cast only numerics or datetimes as well.
+    same_dtype : bool, default True
+        Specify dtype when calling _from_sequence
+
+    Returns
+    -------
+    result : array-like
+        result maybe casted to the dtype.
+    """
+
+    if isinstance(dtype, ExtensionDtype):
+        cls = dtype.construct_array_type()
+        if same_dtype:
+            result = _maybe_cast_to_extension_array(cls, result, dtype=dtype)
+        else:
+            result = _maybe_cast_to_extension_array(cls, result)
+
+    elif (numeric_only and dtype.kind in "iufcb") or not numeric_only:
+        result = maybe_downcast_to_dtype(result, dtype)
+
+    return result
+
+
+def _maybe_cast_to_extension_array(
+    cls: type[ExtensionArray], obj: ArrayLike, dtype: ExtensionDtype | None = None
+) -> ArrayLike:
+    """
+    Call to `_from_sequence` that returns the object unchanged on Exception.
+
+    Parameters
+    ----------
+    cls : class, subclass of ExtensionArray
+    obj : arraylike
+        Values to pass to cls._from_sequence
+    dtype : ExtensionDtype, optional
+
+    Returns
+    -------
+    ExtensionArray or obj
+    """
+    result: ArrayLike
+
+    if dtype is not None:
+        try:
+            result = cls._from_scalars(obj, dtype=dtype)
+        except (TypeError, ValueError):
+            return obj
+        return result
+
+    try:
+        result = cls._from_sequence(obj, dtype=dtype)
+    except Exception:
+        # We can't predict what downstream EA constructors may raise
+        result = obj
+    return result
+
+
+@overload
+def ensure_dtype_can_hold_na(dtype: np.dtype) -> np.dtype: ...
+
+
+@overload
+def ensure_dtype_can_hold_na(dtype: ExtensionDtype) -> ExtensionDtype: ...
+
+
+def ensure_dtype_can_hold_na(dtype: DtypeObj) -> DtypeObj:
+    """
+    If we have a dtype that cannot hold NA values, find the best match that can.
+    """
+    if isinstance(dtype, ExtensionDtype):
+        if dtype._can_hold_na:
+            return dtype
+        elif isinstance(dtype, IntervalDtype):
+            # TODO(GH#45349): don't special-case IntervalDtype, allow
+            #  overriding instead of returning object below.
+            return IntervalDtype(np.float64, closed=dtype.closed)
+        return _dtype_obj
+    elif dtype.kind == "b":
+        return _dtype_obj
+    elif dtype.kind in "iu":
+        return np.dtype(np.float64)
+    return dtype
+
+
+_canonical_nans = {
+    np.datetime64: np.datetime64("NaT", "ns"),
+    np.timedelta64: np.timedelta64("NaT", "ns"),
+    type(np.nan): np.nan,
+}
+
+
+def maybe_promote(dtype: np.dtype, fill_value=np.nan):
+    """
+    Find the minimal dtype that can hold both the given dtype and fill_value.
+
+    Parameters
+    ----------
+    dtype : np.dtype
+    fill_value : scalar, default np.nan
+
+    Returns
+    -------
+    dtype
+        Upcasted from dtype argument if necessary.
+    fill_value
+        Upcasted from fill_value argument if necessary.
+
+    Raises
+    ------
+    ValueError
+        If fill_value is a non-scalar and dtype is not object.
+    """
+    orig = fill_value
+    orig_is_nat = False
+    if checknull(fill_value):
+        # https://github.com/pandas-dev/pandas/pull/39692#issuecomment-1441051740
+        #  avoid cache misses with NaN/NaT values that are not singletons
+        if fill_value is not NA:
+            try:
+                orig_is_nat = np.isnat(fill_value)
+            except TypeError:
+                pass
+
+        fill_value = _canonical_nans.get(type(fill_value), fill_value)
+
+    # for performance, we are using a cached version of the actual implementation
+    # of the function in _maybe_promote. However, this doesn't always work (in case
+    # of non-hashable arguments), so we fallback to the actual implementation if needed
+    try:
+        # error: Argument 3 to "__call__" of "_lru_cache_wrapper" has incompatible type
+        # "Type[Any]"; expected "Hashable"  [arg-type]
+        dtype, fill_value = _maybe_promote_cached(
+            dtype,
+            fill_value,
+            type(fill_value),  # type: ignore[arg-type]
+        )
+    except TypeError:
+        # if fill_value is not hashable (required for caching)
+        dtype, fill_value = _maybe_promote(dtype, fill_value)
+
+    if (dtype == _dtype_obj and orig is not None) or (
+        orig_is_nat and np.datetime_data(orig)[0] != "ns"
+    ):
+        # GH#51592,53497 restore our potentially non-canonical fill_value
+        fill_value = orig
+    return dtype, fill_value
+
+
+@functools.lru_cache
+def _maybe_promote_cached(dtype, fill_value, fill_value_type):
+    # The cached version of _maybe_promote below
+    # This also use fill_value_type as (unused) argument to use this in the
+    # cache lookup -> to differentiate 1 and True
+    return _maybe_promote(dtype, fill_value)
+
+
+def _maybe_promote(dtype: np.dtype, fill_value=np.nan):
+    # The actual implementation of the function, use `maybe_promote` above for
+    # a cached version.
+    if not is_scalar(fill_value):
+        # with object dtype there is nothing to promote, and the user can
+        #  pass pretty much any weird fill_value they like
+        if dtype != object:
+            # with object dtype there is nothing to promote, and the user can
+            #  pass pretty much any weird fill_value they like
+            raise ValueError("fill_value must be a scalar")
+        dtype = _dtype_obj
+        return dtype, fill_value
+
+    if is_valid_na_for_dtype(fill_value, dtype) and dtype.kind in "iufcmM":
+        dtype = ensure_dtype_can_hold_na(dtype)
+        fv = na_value_for_dtype(dtype)
+        return dtype, fv
+
+    elif isinstance(dtype, CategoricalDtype):
+        if fill_value in dtype.categories or isna(fill_value):
+            return dtype, fill_value
+        else:
+            return object, ensure_object(fill_value)
+
+    elif isna(fill_value):
+        dtype = _dtype_obj
+        if fill_value is None:
+            # but we retain e.g. pd.NA
+            fill_value = np.nan
+        return dtype, fill_value
+
+    # returns tuple of (dtype, fill_value)
+    if issubclass(dtype.type, np.datetime64):
+        inferred, fv = infer_dtype_from_scalar(fill_value)
+        if inferred == dtype:
+            return dtype, fv
+
+        from pandas.core.arrays import DatetimeArray
+
+        dta = DatetimeArray._from_sequence([], dtype="M8[ns]")
+        try:
+            fv = dta._validate_setitem_value(fill_value)
+            return dta.dtype, fv
+        except (ValueError, TypeError):
+            return _dtype_obj, fill_value
+
+    elif issubclass(dtype.type, np.timedelta64):
+        inferred, fv = infer_dtype_from_scalar(fill_value)
+        if inferred == dtype:
+            return dtype, fv
+
+        elif inferred.kind == "m":
+            # different unit, e.g. passed np.timedelta64(24, "h") with dtype=m8[ns]
+            # see if we can losslessly cast it to our dtype
+            unit = np.datetime_data(dtype)[0]
+            try:
+                td = Timedelta(fill_value).as_unit(unit, round_ok=False)
+            except OutOfBoundsTimedelta:
+                return _dtype_obj, fill_value
+            else:
+                return dtype, td.asm8
+
+        return _dtype_obj, fill_value
+
+    elif is_float(fill_value):
+        if issubclass(dtype.type, np.bool_):
+            dtype = np.dtype(np.object_)
+
+        elif issubclass(dtype.type, np.integer):
+            dtype = np.dtype(np.float64)
+
+        elif dtype.kind == "f":
+            mst = np.min_scalar_type(fill_value)
+            if mst > dtype:
+                # e.g. mst is np.float64 and dtype is np.float32
+                dtype = mst
+
+        elif dtype.kind == "c":
+            mst = np.min_scalar_type(fill_value)
+            dtype = np.promote_types(dtype, mst)
+
+    elif is_bool(fill_value):
+        if not issubclass(dtype.type, np.bool_):
+            dtype = np.dtype(np.object_)
+
+    elif is_integer(fill_value):
+        if issubclass(dtype.type, np.bool_):
+            dtype = np.dtype(np.object_)
+
+        elif issubclass(dtype.type, np.integer):
+            if not np_can_cast_scalar(fill_value, dtype):  # type: ignore[arg-type]
+                # upcast to prevent overflow
+                mst = np.min_scalar_type(fill_value)
+                dtype = np.promote_types(dtype, mst)
+                if dtype.kind == "f":
+                    # Case where we disagree with numpy
+                    dtype = np.dtype(np.object_)
+
+    elif is_complex(fill_value):
+        if issubclass(dtype.type, np.bool_):
+            dtype = np.dtype(np.object_)
+
+        elif issubclass(dtype.type, (np.integer, np.floating)):
+            mst = np.min_scalar_type(fill_value)
+            dtype = np.promote_types(dtype, mst)
+
+        elif dtype.kind == "c":
+            mst = np.min_scalar_type(fill_value)
+            if mst > dtype:
+                # e.g. mst is np.complex128 and dtype is np.complex64
+                dtype = mst
+
+    else:
+        dtype = np.dtype(np.object_)
+
+    # in case we have a string that looked like a number
+    if issubclass(dtype.type, (bytes, str)):
+        dtype = np.dtype(np.object_)
+
+    fill_value = _ensure_dtype_type(fill_value, dtype)
+    return dtype, fill_value
+
+
+def _ensure_dtype_type(value, dtype: np.dtype):
+    """
+    Ensure that the given value is an instance of the given dtype.
+
+    e.g. if out dtype is np.complex64_, we should have an instance of that
+    as opposed to a python complex object.
+
+    Parameters
+    ----------
+    value : object
+    dtype : np.dtype
+
+    Returns
+    -------
+    object
+    """
+    # Start with exceptions in which we do _not_ cast to numpy types
+
+    if dtype == _dtype_obj:
+        return value
+
+    # Note: before we get here we have already excluded isna(value)
+    return dtype.type(value)
+
+
+def infer_dtype_from(val) -> tuple[DtypeObj, Any]:
+    """
+    Interpret the dtype from a scalar or array.
+
+    Parameters
+    ----------
+    val : object
+    """
+    if not is_list_like(val):
+        return infer_dtype_from_scalar(val)
+    return infer_dtype_from_array(val)
+
+
+def infer_dtype_from_scalar(val) -> tuple[DtypeObj, Any]:
+    """
+    Interpret the dtype from a scalar.
+
+    Parameters
+    ----------
+    val : object
+    """
+    dtype: DtypeObj = _dtype_obj
+
+    # a 1-element ndarray
+    if isinstance(val, np.ndarray):
+        if val.ndim != 0:
+            msg = "invalid ndarray passed to infer_dtype_from_scalar"
+            raise ValueError(msg)
+
+        dtype = val.dtype
+        val = lib.item_from_zerodim(val)
+
+    elif isinstance(val, str):
+        # If we create an empty array using a string to infer
+        # the dtype, NumPy will only allocate one character per entry
+        # so this is kind of bad. Alternately we could use np.repeat
+        # instead of np.empty (but then you still don't want things
+        # coming out as np.str_!
+
+        dtype = _dtype_obj
+        if using_string_dtype():
+            from pandas.core.arrays.string_ import StringDtype
+
+            dtype = StringDtype(na_value=np.nan)
+
+    elif isinstance(val, (np.datetime64, dt.datetime)):
+        try:
+            val = Timestamp(val)
+        except OutOfBoundsDatetime:
+            return _dtype_obj, val
+
+        if val is NaT or val.tz is None:
+            val = val.to_datetime64()
+            dtype = val.dtype
+            # TODO: test with datetime(2920, 10, 1) based on test_replace_dtypes
+        else:
+            dtype = DatetimeTZDtype(unit=val.unit, tz=val.tz)
+
+    elif isinstance(val, (np.timedelta64, dt.timedelta)):
+        try:
+            val = Timedelta(val)
+        except (OutOfBoundsTimedelta, OverflowError):
+            dtype = _dtype_obj
+        else:
+            if val is NaT:
+                val = np.timedelta64("NaT", "ns")
+            else:
+                val = val.asm8
+            dtype = val.dtype
+
+    elif is_bool(val):
+        dtype = np.dtype(np.bool_)
+
+    elif is_integer(val):
+        if isinstance(val, np.integer):
+            dtype = np.dtype(type(val))
+        else:
+            dtype = np.dtype(np.int64)
+
+        try:
+            np.array(val, dtype=dtype)
+        except OverflowError:
+            dtype = np.array(val).dtype
+
+    elif is_float(val):
+        if isinstance(val, np.floating):
+            dtype = np.dtype(type(val))
+        else:
+            dtype = np.dtype(np.float64)
+
+    elif is_complex(val):
+        dtype = np.dtype(np.complex128)
+
+    if isinstance(val, Period):
+        dtype = PeriodDtype(freq=val.freq)
+    elif isinstance(val, Interval):
+        subtype = infer_dtype_from_scalar(val.left)[0]
+        dtype = IntervalDtype(subtype=subtype, closed=val.closed)
+
+    return dtype, val
+
+
+def dict_compat(d: dict[Scalar, Scalar]) -> dict[Scalar, Scalar]:
+    """
+    Convert datetimelike-keyed dicts to a Timestamp-keyed dict.
+
+    Parameters
+    ----------
+    d: dict-like object
+
+    Returns
+    -------
+    dict
+    """
+    return {maybe_box_datetimelike(key): value for key, value in d.items()}
+
+
+def infer_dtype_from_array(arr) -> tuple[DtypeObj, ArrayLike]:
+    """
+    Infer the dtype from an array.
+
+    Parameters
+    ----------
+    arr : array
+
+    Returns
+    -------
+    tuple (pandas-compat dtype, array)
+
+
+    Examples
+    --------
+    >>> np.asarray([1, "1"])
+    array(['1', '1'], dtype='<U21')
+
+    >>> infer_dtype_from_array([1, "1"])
+    (dtype('O'), [1, '1'])
+    """
+    if isinstance(arr, np.ndarray):
+        return arr.dtype, arr
+
+    if not is_list_like(arr):
+        raise TypeError("'arr' must be list-like")
+
+    arr_dtype = getattr(arr, "dtype", None)
+    if isinstance(arr_dtype, ExtensionDtype):
+        return arr.dtype, arr
+
+    elif isinstance(arr, ABCSeries):
+        return arr.dtype, np.asarray(arr)
+
+    # don't force numpy coerce with nan's
+    inferred = lib.infer_dtype(arr, skipna=False)
+    if inferred in ["string", "bytes", "mixed", "mixed-integer"]:
+        return (np.dtype(np.object_), arr)
+
+    arr = np.asarray(arr)
+    return arr.dtype, arr
+
+
+def _maybe_infer_dtype_type(element):
+    """
+    Try to infer an object's dtype, for use in arithmetic ops.
+
+    Uses `element.dtype` if that's available.
+    Objects implementing the iterator protocol are cast to a NumPy array,
+    and from there the array's type is used.
+
+    Parameters
+    ----------
+    element : object
+        Possibly has a `.dtype` attribute, and possibly the iterator
+        protocol.
+
+    Returns
+    -------
+    tipo : type
+
+    Examples
+    --------
+    >>> from collections import namedtuple
+    >>> Foo = namedtuple("Foo", "dtype")
+    >>> _maybe_infer_dtype_type(Foo(np.dtype("i8")))
+    dtype('int64')
+    """
+    tipo = None
+    if hasattr(element, "dtype"):
+        tipo = element.dtype
+    elif is_list_like(element):
+        element = np.asarray(element)
+        tipo = element.dtype
+    return tipo
+
+
+def invalidate_string_dtypes(dtype_set: set[DtypeObj]) -> None:
+    """
+    Change string like dtypes to object for
+    ``DataFrame.select_dtypes()``.
+    """
+    # error: Argument 1 to <set> has incompatible type "Type[generic]"; expected
+    # "Union[dtype[Any], ExtensionDtype, None]"
+    # error: Argument 2 to <set> has incompatible type "Type[generic]"; expected
+    # "Union[dtype[Any], ExtensionDtype, None]"
+    non_string_dtypes = dtype_set - {
+        np.dtype("S").type,  # type: ignore[arg-type]
+        np.dtype("<U").type,  # type: ignore[arg-type]
+    }
+    if non_string_dtypes != dtype_set:
+        raise TypeError("string dtypes are not allowed, use 'object' instead")
+
+
+def coerce_indexer_dtype(indexer, categories) -> np.ndarray:
+    """coerce the indexer input array to the smallest dtype possible"""
+    length = len(categories)
+    if length < _int8_max:
+        return ensure_int8(indexer)
+    elif length < _int16_max:
+        return ensure_int16(indexer)
+    elif length < _int32_max:
+        return ensure_int32(indexer)
+    return ensure_int64(indexer)
+
+
+def convert_dtypes(
+    input_array: ArrayLike,
+    convert_string: bool = True,
+    convert_integer: bool = True,
+    convert_boolean: bool = True,
+    convert_floating: bool = True,
+    infer_objects: bool = False,
+    dtype_backend: Literal["numpy_nullable", "pyarrow"] = "numpy_nullable",
+) -> DtypeObj:
+    """
+    Convert objects to best possible type, and optionally,
+    to types supporting ``pd.NA``.
+
+    Parameters
+    ----------
+    input_array : ExtensionArray or np.ndarray
+    convert_string : bool, default True
+        Whether object dtypes should be converted to ``StringDtype()``.
+    convert_integer : bool, default True
+        Whether, if possible, conversion can be done to integer extension types.
+    convert_boolean : bool, defaults True
+        Whether object dtypes should be converted to ``BooleanDtypes()``.
+    convert_floating : bool, defaults True
+        Whether, if possible, conversion can be done to floating extension types.
+        If `convert_integer` is also True, preference will be give to integer
+        dtypes if the floats can be faithfully casted to integers.
+    infer_objects : bool, defaults False
+        Whether to also infer objects to float/int if possible. Is only hit if the
+        object array contains pd.NA.
+    dtype_backend : {'numpy_nullable', 'pyarrow'}, default 'numpy_nullable'
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). Behaviour is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype
+        * ``"pyarrow"``: returns pyarrow-backed nullable :class:`ArrowDtype`
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    np.dtype, or ExtensionDtype
+    """
+    from pandas.core.arrays.string_ import StringDtype
+
+    inferred_dtype: str | DtypeObj
+
+    if (
+        convert_string or convert_integer or convert_boolean or convert_floating
+    ) and isinstance(input_array, np.ndarray):
+        if input_array.dtype == object:
+            inferred_dtype = lib.infer_dtype(input_array)
+        else:
+            inferred_dtype = input_array.dtype
+
+        if is_string_dtype(inferred_dtype):
+            if not convert_string or inferred_dtype == "bytes":
+                inferred_dtype = input_array.dtype
+            else:
+                inferred_dtype = pandas_dtype_func("string")
+
+        if convert_integer:
+            target_int_dtype = pandas_dtype_func("Int64")
+
+            if input_array.dtype.kind in "iu":
+                from pandas.core.arrays.integer import NUMPY_INT_TO_DTYPE
+
+                inferred_dtype = NUMPY_INT_TO_DTYPE.get(
+                    input_array.dtype, target_int_dtype
+                )
+            elif input_array.dtype.kind in "fcb":
+                # TODO: de-dup with maybe_cast_to_integer_array?
+                arr = input_array[notna(input_array)]
+                if (arr.astype(int) == arr).all():
+                    inferred_dtype = target_int_dtype
+                else:
+                    inferred_dtype = input_array.dtype
+            elif (
+                infer_objects
+                and input_array.dtype == object
+                and (isinstance(inferred_dtype, str) and inferred_dtype == "integer")
+            ):
+                inferred_dtype = target_int_dtype
+
+        if convert_floating:
+            if input_array.dtype.kind in "fcb":
+                # i.e. numeric but not integer
+                from pandas.core.arrays.floating import NUMPY_FLOAT_TO_DTYPE
+
+                inferred_float_dtype: DtypeObj = NUMPY_FLOAT_TO_DTYPE.get(
+                    input_array.dtype, pandas_dtype_func("Float64")
+                )
+                # if we could also convert to integer, check if all floats
+                # are actually integers
+                if convert_integer:
+                    # TODO: de-dup with maybe_cast_to_integer_array?
+                    arr = input_array[notna(input_array)]
+                    if (arr.astype(int) == arr).all():
+                        inferred_dtype = pandas_dtype_func("Int64")
+                    else:
+                        inferred_dtype = inferred_float_dtype
+                else:
+                    inferred_dtype = inferred_float_dtype
+            elif (
+                infer_objects
+                and input_array.dtype == object
+                and (
+                    isinstance(inferred_dtype, str)
+                    and inferred_dtype == "mixed-integer-float"
+                )
+            ):
+                inferred_dtype = pandas_dtype_func("Float64")
+
+        if convert_boolean:
+            if input_array.dtype.kind == "b":
+                inferred_dtype = pandas_dtype_func("boolean")
+            elif isinstance(inferred_dtype, str) and inferred_dtype == "boolean":
+                inferred_dtype = pandas_dtype_func("boolean")
+
+        if isinstance(inferred_dtype, str):
+            # If we couldn't do anything else, then we retain the dtype
+            inferred_dtype = input_array.dtype
+
+    elif (
+        convert_string
+        and isinstance(input_array.dtype, StringDtype)
+        and input_array.dtype.na_value is np.nan
+    ):
+        inferred_dtype = pandas_dtype_func("string")
+
+    else:
+        inferred_dtype = input_array.dtype
+
+    if dtype_backend == "pyarrow" and not isinstance(inferred_dtype, ArrowDtype):
+        from pandas.core.arrays.arrow.array import to_pyarrow_type
+        from pandas.core.arrays.string_ import StringDtype
+
+        assert not isinstance(inferred_dtype, str)
+
+        if (
+            (convert_integer and inferred_dtype.kind in "iu")
+            or (convert_floating and inferred_dtype.kind in "fc")
+            or (convert_boolean and inferred_dtype.kind == "b")
+            or (convert_string and isinstance(inferred_dtype, StringDtype))
+            or (
+                inferred_dtype.kind not in "iufcb"
+                and not isinstance(inferred_dtype, StringDtype)
+            )
+        ):
+            if isinstance(inferred_dtype, PandasExtensionDtype) and not isinstance(
+                inferred_dtype, DatetimeTZDtype
+            ):
+                base_dtype = inferred_dtype.base
+            elif isinstance(inferred_dtype, (BaseMaskedDtype, ArrowDtype)):
+                base_dtype = inferred_dtype.numpy_dtype
+            elif isinstance(inferred_dtype, StringDtype):
+                base_dtype = np.dtype(str)
+            else:
+                base_dtype = inferred_dtype
+            if (
+                base_dtype.kind == "O"  # type: ignore[union-attr]
+                and input_array.size > 0
+                and isna(input_array).all()
+            ):
+                import pyarrow as pa
+
+                pa_type = pa.null()
+            else:
+                pa_type = to_pyarrow_type(base_dtype)
+            if pa_type is not None:
+                inferred_dtype = ArrowDtype(pa_type)
+    elif dtype_backend == "numpy_nullable" and isinstance(inferred_dtype, ArrowDtype):
+        # GH 53648
+        inferred_dtype = _arrow_dtype_mapping()[inferred_dtype.pyarrow_dtype]
+
+    # error: Incompatible return value type (got "Union[str, Union[dtype[Any],
+    # ExtensionDtype]]", expected "Union[dtype[Any], ExtensionDtype]")
+    return inferred_dtype  # type: ignore[return-value]
+
+
+def maybe_infer_to_datetimelike(
+    value: npt.NDArray[np.object_],
+    convert_to_nullable_dtype: bool = False,
+) -> np.ndarray | DatetimeArray | TimedeltaArray | PeriodArray | IntervalArray:
+    """
+    we might have a array (or single object) that is datetime like,
+    and no dtype is passed don't change the value unless we find a
+    datetime/timedelta set
+
+    this is pretty strict in that a datetime/timedelta is REQUIRED
+    in addition to possible nulls/string likes
+
+    Parameters
+    ----------
+    value : np.ndarray[object]
+
+    Returns
+    -------
+    np.ndarray, DatetimeArray, TimedeltaArray, PeriodArray, or IntervalArray
+
+    """
+    if not isinstance(value, np.ndarray) or value.dtype != object:
+        # Caller is responsible for passing only ndarray[object]
+        raise TypeError(type(value))  # pragma: no cover
+    if value.ndim != 1:
+        # Caller is responsible
+        raise ValueError(value.ndim)  # pragma: no cover
+
+    if not len(value):
+        return value
+
+    # error: Incompatible return value type (got "Union[ExtensionArray,
+    # ndarray[Any, Any]]", expected "Union[ndarray[Any, Any], DatetimeArray,
+    # TimedeltaArray, PeriodArray, IntervalArray]")
+    return lib.maybe_convert_objects(  # type: ignore[return-value]
+        value,
+        # Here we do not convert numeric dtypes, as if we wanted that,
+        #  numpy would have done it for us.
+        convert_numeric=False,
+        convert_non_numeric=True,
+        convert_to_nullable_dtype=convert_to_nullable_dtype,
+        dtype_if_all_nat=np.dtype("M8[s]"),
+    )
+
+
+def maybe_cast_to_datetime(
+    value: np.ndarray | list, dtype: np.dtype
+) -> DatetimeArray | TimedeltaArray | np.ndarray:
+    """
+    try to cast the array/value to a datetimelike dtype, converting float
+    nan to iNaT
+
+    Caller is responsible for handling ExtensionDtype cases and non dt64/td64
+    cases.
+    """
+    from pandas.core.arrays.datetimes import DatetimeArray
+    from pandas.core.arrays.timedeltas import TimedeltaArray
+
+    assert dtype.kind in "mM"
+    if not is_list_like(value):
+        raise TypeError("value must be listlike")
+
+    # TODO: _from_sequence would raise ValueError in cases where
+    #  _ensure_nanosecond_dtype raises TypeError
+    _ensure_nanosecond_dtype(dtype)
+
+    if lib.is_np_dtype(dtype, "m"):
+        res = TimedeltaArray._from_sequence(value, dtype=dtype)
+        return res
+    else:
+        try:
+            dta = DatetimeArray._from_sequence(value, dtype=dtype)
+        except ValueError as err:
+            # We can give a Series-specific exception message.
+            if "cannot supply both a tz and a timezone-naive dtype" in str(err):
+                raise ValueError(
+                    "Cannot convert timezone-aware data to "
+                    "timezone-naive dtype. Use "
+                    "pd.Series(values).dt.tz_localize(None) instead."
+                ) from err
+            raise
+
+        return dta
+
+
+def _ensure_nanosecond_dtype(dtype: DtypeObj) -> None:
+    """
+    Convert dtypes with granularity less than nanosecond to nanosecond
+
+    >>> _ensure_nanosecond_dtype(np.dtype("M8[us]"))
+
+    >>> _ensure_nanosecond_dtype(np.dtype("M8[D]"))
+    Traceback (most recent call last):
+        ...
+    TypeError: dtype=datetime64[D] is not supported. Supported resolutions are 's', 'ms', 'us', and 'ns'
+
+    >>> _ensure_nanosecond_dtype(np.dtype("m8[ps]"))
+    Traceback (most recent call last):
+        ...
+    TypeError: dtype=timedelta64[ps] is not supported. Supported resolutions are 's', 'ms', 'us', and 'ns'
+    """  # noqa: E501
+    msg = (
+        f"The '{dtype.name}' dtype has no unit. "
+        f"Please pass in '{dtype.name}[ns]' instead."
+    )
+
+    # unpack e.g. SparseDtype
+    dtype = getattr(dtype, "subtype", dtype)
+
+    if not isinstance(dtype, np.dtype):
+        # i.e. datetime64tz
+        pass
+
+    elif dtype.kind in "mM":
+        if not is_supported_dtype(dtype):
+            # pre-2.0 we would silently swap in nanos for lower-resolutions,
+            #  raise for above-nano resolutions
+            if dtype.name in ["datetime64", "timedelta64"]:
+                raise ValueError(msg)
+            # TODO: ValueError or TypeError? existing test
+            #  test_constructor_generic_timestamp_bad_frequency expects TypeError
+            raise TypeError(
+                f"dtype={dtype} is not supported. Supported resolutions are 's', "
+                "'ms', 'us', and 'ns'"
+            )
+
+
+# TODO: other value-dependent functions to standardize here include
+#  Index._find_common_type_compat
+def find_result_type(left_dtype: DtypeObj, right: Any) -> DtypeObj:
+    """
+    Find the type/dtype for the result of an operation between objects.
+
+    This is similar to find_common_type, but looks at the right object instead
+    of just its dtype. This can be useful in particular when the right
+    object does not have a `dtype`.
+
+    Parameters
+    ----------
+    left_dtype : np.dtype or ExtensionDtype
+    right : Any
+
+    Returns
+    -------
+    np.dtype or ExtensionDtype
+
+    See also
+    --------
+    find_common_type
+    numpy.result_type
+    """
+    new_dtype: DtypeObj
+
+    if (
+        isinstance(left_dtype, np.dtype)
+        and left_dtype.kind in "iuc"
+        and (lib.is_integer(right) or lib.is_float(right))
+    ):
+        # e.g. with int8 dtype and right=512, we want to end up with
+        # np.int16, whereas infer_dtype_from(512) gives np.int64,
+        #  which will make us upcast too far.
+        if lib.is_float(right) and right.is_integer() and left_dtype.kind != "f":
+            right = int(right)
+        # After NEP 50, numpy won't inspect Python scalars
+        # TODO: do we need to recreate numpy's inspection logic for floats too
+        # (this breaks some tests)
+        if isinstance(right, int) and not isinstance(right, np.integer):
+            # This gives an unsigned type by default
+            # (if our number is positive)
+
+            # If our left dtype is signed, we might not want this since
+            # this might give us 1 dtype too big
+            # We should check if the corresponding int dtype (e.g. int64 for uint64)
+            # can hold the number
+            right_dtype = np.min_scalar_type(right)
+            if right == 0:
+                # Special case 0
+                right = left_dtype
+            elif (
+                not np.issubdtype(left_dtype, np.unsignedinteger)
+                and 0 < right <= np.iinfo(right_dtype).max
+            ):
+                # If left dtype isn't unsigned, check if it fits in the signed dtype
+                right = np.dtype(f"i{right_dtype.itemsize}")
+            else:
+                right = right_dtype
+
+        new_dtype = np.result_type(left_dtype, right)
+
+    elif is_valid_na_for_dtype(right, left_dtype):
+        # e.g. IntervalDtype[int] and None/np.nan
+        new_dtype = ensure_dtype_can_hold_na(left_dtype)
+
+    else:
+        dtype, _ = infer_dtype_from(right)
+        new_dtype = find_common_type([left_dtype, dtype])
+
+    return new_dtype
+
+
+def common_dtype_categorical_compat(
+    objs: Sequence[Index | ArrayLike], dtype: DtypeObj
+) -> DtypeObj:
+    """
+    Update the result of find_common_type to account for NAs in a Categorical.
+
+    Parameters
+    ----------
+    objs : list[np.ndarray | ExtensionArray | Index]
+    dtype : np.dtype or ExtensionDtype
+
+    Returns
+    -------
+    np.dtype or ExtensionDtype
+    """
+    # GH#38240
+
+    # TODO: more generally, could do `not can_hold_na(dtype)`
+    if lib.is_np_dtype(dtype, "iu"):
+        for obj in objs:
+            # We don't want to accidentally allow e.g. "categorical" str here
+            obj_dtype = getattr(obj, "dtype", None)
+            if isinstance(obj_dtype, CategoricalDtype):
+                if isinstance(obj, ABCIndex):
+                    # This check may already be cached
+                    hasnas = obj.hasnans
+                else:
+                    # Categorical
+                    hasnas = cast("Categorical", obj)._hasna
+
+                if hasnas:
+                    # see test_union_int_categorical_with_nan
+                    dtype = np.dtype(np.float64)
+                    break
+    return dtype
+
+
+def np_find_common_type(*dtypes: np.dtype) -> np.dtype:
+    """
+    np.find_common_type implementation pre-1.25 deprecation using np.result_type
+    https://github.com/pandas-dev/pandas/pull/49569#issuecomment-1308300065
+
+    Parameters
+    ----------
+    dtypes : np.dtypes
+
+    Returns
+    -------
+    np.dtype
+    """
+    try:
+        common_dtype = np.result_type(*dtypes)
+        if common_dtype.kind in "mMSU":
+            # NumPy promotion currently (1.25) misbehaves for for times and strings,
+            # so fall back to object (find_common_dtype did unless there
+            # was only one dtype)
+            common_dtype = np.dtype("O")
+
+    except TypeError:
+        common_dtype = np.dtype("O")
+    return common_dtype
+
+
+@overload
+def find_common_type(types: list[np.dtype]) -> np.dtype: ...
+
+
+@overload
+def find_common_type(types: list[ExtensionDtype]) -> DtypeObj: ...
+
+
+@overload
+def find_common_type(types: list[DtypeObj]) -> DtypeObj: ...
+
+
+def find_common_type(types):
+    """
+    Find a common data type among the given dtypes.
+
+    Parameters
+    ----------
+    types : list of dtypes
+
+    Returns
+    -------
+    pandas extension or numpy dtype
+
+    See Also
+    --------
+    numpy.find_common_type
+
+    """
+    if not types:
+        raise ValueError("no types given")
+
+    first = types[0]
+
+    # workaround for find_common_type([np.dtype('datetime64[ns]')] * 2)
+    # => object
+    if lib.dtypes_all_equal(list(types)):
+        return first
+
+    # get unique types (dict.fromkeys is used as order-preserving set())
+    types = list(dict.fromkeys(types).keys())
+
+    if any(isinstance(t, ExtensionDtype) for t in types):
+        for t in types:
+            if isinstance(t, ExtensionDtype):
+                res = t._get_common_dtype(types)
+                if res is not None:
+                    return res
+        return np.dtype("object")
+
+    # take lowest unit
+    if all(lib.is_np_dtype(t, "M") for t in types):
+        return np.dtype(max(types))
+    if all(lib.is_np_dtype(t, "m") for t in types):
+        return np.dtype(max(types))
+
+    # don't mix bool / int or float or complex
+    # this is different from numpy, which casts bool with float/int as int
+    has_bools = any(t.kind == "b" for t in types)
+    if has_bools:
+        for t in types:
+            if t.kind in "iufc":
+                return np.dtype("object")
+
+    return np_find_common_type(*types)
+
+
+def construct_2d_arraylike_from_scalar(
+    value: Scalar, length: int, width: int, dtype: np.dtype, copy: bool
+) -> np.ndarray:
+    shape = (length, width)
+
+    if dtype.kind in "mM":
+        value = _maybe_box_and_unbox_datetimelike(value, dtype)
+    elif dtype == _dtype_obj:
+        if isinstance(value, (np.timedelta64, np.datetime64)):
+            # calling np.array below would cast to pytimedelta/pydatetime
+            out = np.empty(shape, dtype=object)
+            out.fill(value)
+            return out
+
+    # Attempt to coerce to a numpy array
+    try:
+        if not copy:
+            arr = np.asarray(value, dtype=dtype)
+        else:
+            arr = np.array(value, dtype=dtype, copy=copy)
+    except (ValueError, TypeError) as err:
+        raise TypeError(
+            f"DataFrame constructor called with incompatible data and dtype: {err}"
+        ) from err
+
+    if arr.ndim != 0:
+        raise ValueError("DataFrame constructor not properly called!")
+
+    return np.full(shape, arr)
+
+
+def construct_1d_arraylike_from_scalar(
+    value: Scalar, length: int, dtype: DtypeObj | None
+) -> ArrayLike:
+    """
+    create a np.ndarray / pandas type of specified shape and dtype
+    filled with values
+
+    Parameters
+    ----------
+    value : scalar value
+    length : int
+    dtype : pandas_dtype or np.dtype
+
+    Returns
+    -------
+    np.ndarray / pandas type of length, filled with value
+
+    """
+
+    if dtype is None:
+        try:
+            dtype, value = infer_dtype_from_scalar(value)
+        except OutOfBoundsDatetime:
+            dtype = _dtype_obj
+
+    if isinstance(dtype, ExtensionDtype):
+        cls = dtype.construct_array_type()
+        seq = [] if length == 0 else [value]
+        return cls._from_sequence(seq, dtype=dtype).repeat(length)
+
+    if length and dtype.kind in "iu" and isna(value):
+        # coerce if we have nan for an integer dtype
+        dtype = np.dtype("float64")
+    elif lib.is_np_dtype(dtype, "US"):
+        # we need to coerce to object dtype to avoid
+        # to allow numpy to take our string as a scalar value
+        dtype = np.dtype("object")
+        if not isna(value):
+            value = ensure_str(value)
+    elif dtype.kind in "mM":
+        value = _maybe_box_and_unbox_datetimelike(value, dtype)
+
+    subarr = np.empty(length, dtype=dtype)
+    if length:
+        # GH 47391: numpy > 1.24 will raise filling np.nan into int dtypes
+        subarr.fill(value)
+
+    return subarr
+
+
+def _maybe_box_and_unbox_datetimelike(value: Scalar, dtype: DtypeObj):
+    # Caller is responsible for checking dtype.kind in "mM"
+
+    if isinstance(value, dt.datetime):
+        # we dont want to box dt64, in particular datetime64("NaT")
+        value = maybe_box_datetimelike(value, dtype)
+
+    return _maybe_unbox_datetimelike(value, dtype)
+
+
+def construct_1d_object_array_from_listlike(values: Collection) -> np.ndarray:
+    """
+    Transform any list-like object in a 1-dimensional numpy array of object
+    dtype.
+
+    Parameters
+    ----------
+    values : any iterable which has a len()
+
+    Raises
+    ------
+    TypeError
+        * If `values` does not have a len()
+
+    Returns
+    -------
+    1-dimensional numpy array of dtype object
+    """
+    # numpy will try to interpret nested lists as further dimensions in np.array(),
+    # hence explicitly making a 1D array using np.fromiter
+    return np.fromiter(values, dtype="object", count=len(values))
+
+
+def maybe_cast_to_integer_array(arr: list | np.ndarray, dtype: np.dtype) -> np.ndarray:
+    """
+    Takes any dtype and returns the casted version, raising for when data is
+    incompatible with integer/unsigned integer dtypes.
+
+    Parameters
+    ----------
+    arr : np.ndarray or list
+        The array to cast.
+    dtype : np.dtype
+        The integer dtype to cast the array to.
+
+    Returns
+    -------
+    ndarray
+        Array of integer or unsigned integer dtype.
+
+    Raises
+    ------
+    OverflowError : the dtype is incompatible with the data
+    ValueError : loss of precision has occurred during casting
+
+    Examples
+    --------
+    If you try to coerce negative values to unsigned integers, it raises:
+
+    >>> pd.Series([-1], dtype="uint64")
+    Traceback (most recent call last):
+        ...
+    OverflowError: Trying to coerce negative values to unsigned integers
+
+    Also, if you try to coerce float values to integers, it raises:
+
+    >>> maybe_cast_to_integer_array([1, 2, 3.5], dtype=np.dtype("int64"))
+    Traceback (most recent call last):
+        ...
+    ValueError: Trying to coerce float values to integers
+    """
+    assert dtype.kind in "iu"
+
+    try:
+        if not isinstance(arr, np.ndarray):
+            with warnings.catch_warnings():
+                # We already disallow dtype=uint w/ negative numbers
+                # (test_constructor_coercion_signed_to_unsigned) so safe to ignore.
+                warnings.filterwarnings(
+                    "ignore",
+                    "NumPy will stop allowing conversion of out-of-bound Python int",
+                    DeprecationWarning,
+                )
+                casted = np.asarray(arr, dtype=dtype)
+        else:
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore", category=RuntimeWarning)
+                casted = arr.astype(dtype, copy=False)
+    except OverflowError as err:
+        raise OverflowError(
+            "The elements provided in the data cannot all be "
+            f"casted to the dtype {dtype}"
+        ) from err
+
+    if isinstance(arr, np.ndarray) and arr.dtype == dtype:
+        # avoid expensive array_equal check
+        return casted
+
+    with warnings.catch_warnings():
+        warnings.filterwarnings("ignore", category=RuntimeWarning)
+        warnings.filterwarnings(
+            "ignore", "elementwise comparison failed", FutureWarning
+        )
+        if np.array_equal(arr, casted):
+            return casted
+
+    # We do this casting to allow for proper
+    # data and dtype checking.
+    #
+    # We didn't do this earlier because NumPy
+    # doesn't handle `uint64` correctly.
+    arr = np.asarray(arr)
+
+    if np.issubdtype(arr.dtype, str):
+        # TODO(numpy-2.0 min): This case will raise an OverflowError above
+        if (casted.astype(str) == arr).all():
+            return casted
+        raise ValueError(f"string values cannot be losslessly cast to {dtype}")
+
+    if dtype.kind == "u" and (arr < 0).any():
+        # TODO: can this be hit anymore after numpy 2.0?
+        raise OverflowError("Trying to coerce negative values to unsigned integers")
+
+    if arr.dtype.kind == "f":
+        if not np.isfinite(arr).all():
+            raise IntCastingNaNError(
+                "Cannot convert non-finite values (NA or inf) to integer"
+            )
+        raise ValueError("Trying to coerce float values to integers")
+    if arr.dtype == object:
+        raise ValueError("Trying to coerce object values to integers")
+
+    if casted.dtype < arr.dtype:
+        # TODO: Can this path be hit anymore with numpy > 2
+        # GH#41734 e.g. [1, 200, 923442] and dtype="int8" -> overflows
+        raise ValueError(
+            f"Values are too large to be losslessly converted to {dtype}. "
+            f"To cast anyway, use pd.Series(values).astype({dtype})"
+        )
+
+    if arr.dtype.kind in "mM":
+        # test_constructor_maskedarray_nonfloat
+        raise TypeError(
+            f"Constructing a Series or DataFrame from {arr.dtype} values and "
+            f"dtype={dtype} is not supported. Use values.view({dtype}) instead."
+        )
+
+    # No known cases that get here, but raising explicitly to cover our bases.
+    raise ValueError(f"values cannot be losslessly cast to {dtype}")
+
+
+def can_hold_element(arr: ArrayLike, element: Any) -> bool:
+    """
+    Can we do an inplace setitem with this element in an array with this dtype?
+
+    Parameters
+    ----------
+    arr : np.ndarray or ExtensionArray
+    element : Any
+
+    Returns
+    -------
+    bool
+    """
+    dtype = arr.dtype
+    if not isinstance(dtype, np.dtype) or dtype.kind in "mM":
+        if isinstance(dtype, (PeriodDtype, IntervalDtype, DatetimeTZDtype, np.dtype)):
+            # np.dtype here catches datetime64ns and timedelta64ns; we assume
+            #  in this case that we have DatetimeArray/TimedeltaArray
+            arr = cast(
+                "PeriodArray | DatetimeArray | TimedeltaArray | IntervalArray", arr
+            )
+            try:
+                arr._validate_setitem_value(element)
+                return True
+            except (ValueError, TypeError):
+                return False
+
+        if dtype == "string":
+            try:
+                arr._maybe_convert_setitem_value(element)  # type: ignore[union-attr]
+                return True
+            except (ValueError, TypeError):
+                return False
+
+        # This is technically incorrect, but maintains the behavior of
+        # ExtensionBlock._can_hold_element
+        return True
+
+    try:
+        np_can_hold_element(dtype, element)
+        return True
+    except (TypeError, LossySetitemError):
+        return False
+
+
+def np_can_hold_element(dtype: np.dtype, element: Any) -> Any:
+    """
+    Raise if we cannot losslessly set this element into an ndarray with this dtype.
+
+    Specifically about places where we disagree with numpy.  i.e. there are
+    cases where numpy will raise in doing the setitem that we do not check
+    for here, e.g. setting str "X" into a numeric ndarray.
+
+    Returns
+    -------
+    Any
+        The element, potentially cast to the dtype.
+
+    Raises
+    ------
+    ValueError : If we cannot losslessly store this element with this dtype.
+    """
+    if dtype == _dtype_obj:
+        return element
+
+    tipo = _maybe_infer_dtype_type(element)
+
+    if dtype.kind in "iu":
+        if isinstance(element, range):
+            if _dtype_can_hold_range(element, dtype):
+                return element
+            raise LossySetitemError
+
+        if is_integer(element) or (is_float(element) and element.is_integer()):
+            # e.g. test_setitem_series_int8 if we have a python int 1
+            #  tipo may be np.int32, despite the fact that it will fit
+            #  in smaller int dtypes.
+            info = np.iinfo(dtype)
+            if info.min <= element <= info.max:
+                return dtype.type(element)
+            raise LossySetitemError
+
+        if tipo is not None:
+            if tipo.kind not in "iu":
+                if isinstance(element, np.ndarray) and element.dtype.kind == "f":
+                    # If all can be losslessly cast to integers, then we can hold them
+                    with np.errstate(invalid="ignore"):
+                        # We check afterwards if cast was losslessly, so no need to show
+                        # the warning
+                        casted = element.astype(dtype)
+                    comp = casted == element
+                    if comp.all():
+                        # Return the casted values bc they can be passed to
+                        #  np.putmask, whereas the raw values cannot.
+                        #  see TestSetitemFloatNDarrayIntoIntegerSeries
+                        return casted
+                    raise LossySetitemError
+
+                elif isinstance(element, ABCExtensionArray) and isinstance(
+                    element.dtype, CategoricalDtype
+                ):
+                    # GH#52927 setting Categorical value into non-EA frame
+                    # TODO: general-case for EAs?
+                    try:
+                        casted = element.astype(dtype)
+                    except (ValueError, TypeError) as err:
+                        raise LossySetitemError from err
+                    # Check for cases of either
+                    #  a) lossy overflow/rounding or
+                    #  b) semantic changes like dt64->int64
+                    comp = casted == element
+                    if not comp.all():
+                        raise LossySetitemError
+                    return casted
+
+                # Anything other than integer we cannot hold
+                raise LossySetitemError
+            if (
+                dtype.kind == "u"
+                and isinstance(element, np.ndarray)
+                and element.dtype.kind == "i"
+            ):
+                # see test_where_uint64
+                casted = element.astype(dtype)
+                if (casted == element).all():
+                    # TODO: faster to check (element >=0).all()?  potential
+                    #  itemsize issues there?
+                    return casted
+                raise LossySetitemError
+            if dtype.itemsize < tipo.itemsize:
+                raise LossySetitemError
+            if not isinstance(tipo, np.dtype):
+                # i.e. nullable IntegerDtype; we can put this into an ndarray
+                #  losslessly iff it has no NAs
+                arr = element._values if isinstance(element, ABCSeries) else element
+                if arr._hasna:
+                    raise LossySetitemError
+                return element
+
+            return element
+
+        raise LossySetitemError
+
+    if dtype.kind == "f":
+        if lib.is_integer(element) or lib.is_float(element):
+            casted = dtype.type(element)
+            if np.isnan(casted) or casted == element:
+                return casted
+            # otherwise e.g. overflow see TestCoercionFloat32
+            raise LossySetitemError
+
+        if tipo is not None:
+            # TODO: itemsize check?
+            if tipo.kind not in "iuf":
+                # Anything other than float/integer we cannot hold
+                raise LossySetitemError
+            if not isinstance(tipo, np.dtype):
+                # i.e. nullable IntegerDtype or FloatingDtype;
+                #  we can put this into an ndarray losslessly iff it has no NAs
+                if element._hasna:
+                    raise LossySetitemError
+                return element
+            elif tipo.itemsize > dtype.itemsize or tipo.kind != dtype.kind:
+                if isinstance(element, np.ndarray):
+                    # e.g. TestDataFrameIndexingWhere::test_where_alignment
+                    casted = element.astype(dtype)
+                    if np.array_equal(casted, element, equal_nan=True):
+                        return casted
+                    raise LossySetitemError
+
+            return element
+
+        raise LossySetitemError
+
+    if dtype.kind == "c":
+        if lib.is_integer(element) or lib.is_complex(element) or lib.is_float(element):
+            if np.isnan(element):
+                # see test_where_complex GH#6345
+                return dtype.type(element)
+
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore")
+                casted = dtype.type(element)
+            if casted == element:
+                return casted
+            # otherwise e.g. overflow see test_32878_complex_itemsize
+            raise LossySetitemError
+
+        if tipo is not None:
+            if tipo.kind in "iufc":
+                return element
+            raise LossySetitemError
+        raise LossySetitemError
+
+    if dtype.kind == "b":
+        if tipo is not None:
+            if tipo.kind == "b":
+                if not isinstance(tipo, np.dtype):
+                    # i.e. we have a BooleanArray
+                    if element._hasna:
+                        # i.e. there are pd.NA elements
+                        raise LossySetitemError
+                return element
+            raise LossySetitemError
+        if lib.is_bool(element):
+            return element
+        raise LossySetitemError
+
+    if dtype.kind == "S":
+        # TODO: test tests.frame.methods.test_replace tests get here,
+        #  need more targeted tests.  xref phofl has a PR about this
+        if tipo is not None:
+            if tipo.kind == "S" and tipo.itemsize <= dtype.itemsize:
+                return element
+            raise LossySetitemError
+        if isinstance(element, bytes) and len(element) <= dtype.itemsize:
+            return element
+        raise LossySetitemError
+
+    if dtype.kind == "V":
+        # i.e. np.void, which cannot hold _anything_
+        raise LossySetitemError
+
+    raise NotImplementedError(dtype)
+
+
+def _dtype_can_hold_range(rng: range, dtype: np.dtype) -> bool:
+    """
+    _maybe_infer_dtype_type infers to int64 (and float64 for very large endpoints),
+    but in many cases a range can be held by a smaller integer dtype.
+    Check if this is one of those cases.
+    """
+    if not len(rng):
+        return True
+    return np_can_cast_scalar(rng.start, dtype) and np_can_cast_scalar(rng.stop, dtype)
+
+
+def np_can_cast_scalar(element: Scalar, dtype: np.dtype) -> bool:
+    """
+    np.can_cast pandas-equivalent for pre 2-0 behavior that allowed scalar
+    inference
+
+    Parameters
+    ----------
+    element : Scalar
+    dtype : np.dtype
+
+    Returns
+    -------
+    bool
+    """
+    try:
+        np_can_hold_element(dtype, element)
+        return True
+    except (LossySetitemError, NotImplementedError):
+        return False
diff --git a/pandas/pandas/core/dtypes/common.py b/pandas/pandas/core/dtypes/common.py
new file mode 100644
index 0000000000000..e92f2363b69f1
--- /dev/null
+++ b/pandas/pandas/core/dtypes/common.py
@@ -0,0 +1,1947 @@
+"""
+Common type operations.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import (
+    Interval,
+    Period,
+    algos,
+    lib,
+)
+from pandas._libs.tslibs import conversion
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import _registry as registry
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    PeriodDtype,
+    SparseDtype,
+)
+from pandas.core.dtypes.generic import ABCIndex
+from pandas.core.dtypes.inference import (
+    is_array_like,
+    is_bool,
+    is_complex,
+    is_dataclass,
+    is_decimal,
+    is_dict_like,
+    is_file_like,
+    is_float,
+    is_hashable,
+    is_integer,
+    is_iterator,
+    is_list_like,
+    is_named_tuple,
+    is_nested_list_like,
+    is_number,
+    is_re,
+    is_re_compilable,
+    is_scalar,
+    is_sequence,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        ArrayLike,
+        DtypeObj,
+    )
+
+DT64NS_DTYPE = conversion.DT64NS_DTYPE
+TD64NS_DTYPE = conversion.TD64NS_DTYPE
+INT64_DTYPE = np.dtype(np.int64)
+
+# oh the troubles to reduce import time
+_is_scipy_sparse = None
+
+ensure_float64 = algos.ensure_float64
+ensure_int64 = algos.ensure_int64
+ensure_int32 = algos.ensure_int32
+ensure_int16 = algos.ensure_int16
+ensure_int8 = algos.ensure_int8
+ensure_platform_int = algos.ensure_platform_int
+ensure_object = algos.ensure_object
+ensure_uint64 = algos.ensure_uint64
+
+
+def ensure_str(value: bytes | Any) -> str:
+    """
+    Ensure that bytes and non-strings get converted into ``str`` objects.
+    """
+    if isinstance(value, bytes):
+        value = value.decode("utf-8")
+    elif not isinstance(value, str):
+        value = str(value)
+    return value
+
+
+def ensure_python_int(value: int | np.integer) -> int:
+    """
+    Ensure that a value is a python int.
+
+    Parameters
+    ----------
+    value: int or numpy.integer
+
+    Returns
+    -------
+    int
+
+    Raises
+    ------
+    TypeError: if the value isn't an int or can't be converted to one.
+    """
+    if not (is_integer(value) or is_float(value)):
+        if not is_scalar(value):
+            raise TypeError(
+                f"Value needs to be a scalar value, was type {type(value).__name__}"
+            )
+        raise TypeError(f"Wrong type {type(value)} for value {value}")
+    try:
+        new_value = int(value)
+        assert new_value == value
+    except (TypeError, ValueError, AssertionError) as err:
+        raise TypeError(f"Wrong type {type(value)} for value {value}") from err
+    return new_value
+
+
+def classes(*klasses) -> Callable:
+    """Evaluate if the tipo is a subclass of the klasses."""
+    return lambda tipo: issubclass(tipo, klasses)
+
+
+def _classes_and_not_datetimelike(*klasses) -> Callable:
+    """
+    Evaluate if the tipo is a subclass of the klasses
+    and not a datetimelike.
+    """
+    return lambda tipo: (
+        issubclass(tipo, klasses)
+        and not issubclass(tipo, (np.datetime64, np.timedelta64))
+    )
+
+
+def is_object_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the object dtype.
+
+    This method examines the input to determine if it is of the
+    object data type. Object dtype is a generic data type that can
+    hold any Python objects, including strings, lists, and custom
+    objects.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the object dtype.
+
+    See Also
+    --------
+    api.types.is_numeric_dtype : Check whether the provided array or dtype is of a
+        numeric dtype.
+    api.types.is_string_dtype : Check whether the provided array or dtype is of
+        the string dtype.
+    api.types.is_bool_dtype : Check whether the provided array or dtype is of a
+        boolean dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_object_dtype
+    >>> is_object_dtype(object)
+    True
+    >>> is_object_dtype(int)
+    False
+    >>> is_object_dtype(np.array([], dtype=object))
+    True
+    >>> is_object_dtype(np.array([], dtype=int))
+    False
+    >>> is_object_dtype([1, 2, 3])
+    False
+    """
+    return _is_dtype_type(arr_or_dtype, classes(np.object_))
+
+
+def is_sparse(arr) -> bool:
+    """
+    Check whether an array-like is a 1-D pandas sparse array.
+
+    .. deprecated:: 2.1.0
+        Use isinstance(dtype, pd.SparseDtype) instead.
+
+    Check that the one-dimensional array-like is a pandas sparse array.
+    Returns True if it is a pandas sparse array, not another type of
+    sparse array.
+
+    Parameters
+    ----------
+    arr : array-like
+        Array-like to check.
+
+    Returns
+    -------
+    bool
+        Whether or not the array-like is a pandas sparse array.
+
+    See Also
+    --------
+    api.types.SparseDtype : The dtype object for pandas sparse arrays.
+
+    Examples
+    --------
+    Returns `True` if the parameter is a 1-D pandas sparse array.
+
+    >>> from pandas.api.types import is_sparse
+    >>> is_sparse(pd.arrays.SparseArray([0, 0, 1, 0]))
+    True
+    >>> is_sparse(pd.Series(pd.arrays.SparseArray([0, 0, 1, 0])))
+    True
+
+    Returns `False` if the parameter is not sparse.
+
+    >>> is_sparse(np.array([0, 0, 1, 0]))
+    False
+    >>> is_sparse(pd.Series([0, 1, 0, 0]))
+    False
+
+    Returns `False` if the parameter is not a pandas sparse array.
+
+    >>> from scipy.sparse import bsr_matrix
+    >>> is_sparse(bsr_matrix([0, 1, 0, 0]))
+    False
+
+    Returns `False` if the parameter has more than one dimension.
+    """
+    warnings.warn(
+        "is_sparse is deprecated and will be removed in a future "
+        "version. Check `isinstance(dtype, pd.SparseDtype)` instead.",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    dtype = getattr(arr, "dtype", arr)
+    return isinstance(dtype, SparseDtype)
+
+
+def is_scipy_sparse(arr) -> bool:
+    """
+    Check whether an array-like is a scipy.sparse.spmatrix instance.
+
+    Parameters
+    ----------
+    arr : array-like
+        The array-like to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like is a scipy.sparse.spmatrix instance.
+
+    Notes
+    -----
+    If scipy is not installed, this function will always return False.
+
+    Examples
+    --------
+    >>> from scipy.sparse import bsr_matrix
+    >>> is_scipy_sparse(bsr_matrix([1, 2, 3]))
+    True
+    >>> is_scipy_sparse(pd.arrays.SparseArray([1, 2, 3]))
+    False
+    """
+    global _is_scipy_sparse
+
+    if _is_scipy_sparse is None:
+        try:
+            from scipy.sparse import issparse as _is_scipy_sparse
+        except ImportError:
+            _is_scipy_sparse = lambda _: False
+
+    assert _is_scipy_sparse is not None
+    return _is_scipy_sparse(arr)
+
+
+def is_datetime64_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the datetime64 dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the datetime64 dtype.
+
+    See Also
+    --------
+    api.types.is_datetime64_ns_dtype: Check whether the provided array or
+                                        dtype is of the datetime64[ns] dtype.
+    api.types.is_datetime64_any_dtype: Check whether the provided array or
+                                        dtype is of the datetime64 dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_datetime64_dtype
+    >>> is_datetime64_dtype(object)
+    False
+    >>> is_datetime64_dtype(np.datetime64)
+    True
+    >>> is_datetime64_dtype(np.array([], dtype=int))
+    False
+    >>> is_datetime64_dtype(np.array([], dtype=np.datetime64))
+    True
+    >>> is_datetime64_dtype([1, 2, 3])
+    False
+    """
+    if isinstance(arr_or_dtype, np.dtype):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.kind == "M"
+    return _is_dtype_type(arr_or_dtype, classes(np.datetime64))
+
+
+def is_datetime64tz_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of a DatetimeTZDtype dtype.
+
+    .. deprecated:: 2.1.0
+        Use isinstance(dtype, pd.DatetimeTZDtype) instead.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of a DatetimeTZDtype dtype.
+
+    See Also
+    --------
+    api.types.is_datetime64_dtype: Check whether an array-like or
+                                        dtype is of the datetime64 dtype.
+    api.types.is_datetime64_any_dtype: Check whether the provided array or
+                                        dtype is of the datetime64 dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_datetime64tz_dtype
+    >>> is_datetime64tz_dtype(object)
+    False
+    >>> is_datetime64tz_dtype([1, 2, 3])
+    False
+    >>> is_datetime64tz_dtype(pd.DatetimeIndex([1, 2, 3]))  # tz-naive
+    False
+    >>> is_datetime64tz_dtype(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern"))
+    True
+
+    >>> from pandas.core.dtypes.dtypes import DatetimeTZDtype
+    >>> dtype = DatetimeTZDtype("ns", tz="US/Eastern")
+    >>> s = pd.Series([], dtype=dtype)
+    >>> is_datetime64tz_dtype(dtype)
+    True
+    >>> is_datetime64tz_dtype(s)
+    True
+    """
+    # GH#52607
+    warnings.warn(
+        "is_datetime64tz_dtype is deprecated and will be removed in a future "
+        "version. Check `isinstance(dtype, pd.DatetimeTZDtype)` instead.",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    if isinstance(arr_or_dtype, DatetimeTZDtype):
+        # GH#33400 fastpath for dtype object
+        # GH 34986
+        return True
+
+    if arr_or_dtype is None:
+        return False
+    return DatetimeTZDtype.is_dtype(arr_or_dtype)
+
+
+def is_timedelta64_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the timedelta64 dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the timedelta64 dtype.
+
+    See Also
+    --------
+    api.types.is_timedelta64_ns_dtype : Check whether the provided array or dtype is
+        of the timedelta64[ns] dtype.
+    api.types.is_period_dtype : Check whether an array-like or dtype is of the
+        Period dtype.
+
+    Examples
+    --------
+    >>> from pandas.core.dtypes.common import is_timedelta64_dtype
+    >>> is_timedelta64_dtype(object)
+    False
+    >>> is_timedelta64_dtype(np.timedelta64)
+    True
+    >>> is_timedelta64_dtype([1, 2, 3])
+    False
+    >>> is_timedelta64_dtype(pd.Series([], dtype="timedelta64[ns]"))
+    True
+    >>> is_timedelta64_dtype("0 days")
+    False
+    """
+    if isinstance(arr_or_dtype, np.dtype):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.kind == "m"
+
+    return _is_dtype_type(arr_or_dtype, classes(np.timedelta64))
+
+
+def is_period_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the Period dtype.
+
+    .. deprecated:: 2.2.0
+        Use isinstance(dtype, pd.PeriodDtype) instead.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the Period dtype.
+
+    See Also
+    --------
+    api.types.is_timedelta64_ns_dtype : Check whether the provided array or dtype is
+        of the timedelta64[ns] dtype.
+    api.types.is_timedelta64_dtype: Check whether an array-like or dtype
+        is of the timedelta64 dtype.
+
+    Examples
+    --------
+    >>> from pandas.core.dtypes.common import is_period_dtype
+    >>> is_period_dtype(object)
+    False
+    >>> is_period_dtype(pd.PeriodDtype(freq="D"))
+    True
+    >>> is_period_dtype([1, 2, 3])
+    False
+    >>> is_period_dtype(pd.Period("2017-01-01"))
+    False
+    >>> is_period_dtype(pd.PeriodIndex([], freq="Y"))
+    True
+    """
+    warnings.warn(
+        "is_period_dtype is deprecated and will be removed in a future version. "
+        "Use `isinstance(dtype, pd.PeriodDtype)` instead",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    if isinstance(arr_or_dtype, ExtensionDtype):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.type is Period
+
+    if arr_or_dtype is None:
+        return False
+    return PeriodDtype.is_dtype(arr_or_dtype)
+
+
+def is_interval_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the Interval dtype.
+
+    .. deprecated:: 2.2.0
+        Use isinstance(dtype, pd.IntervalDtype) instead.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the Interval dtype.
+
+    See Also
+    --------
+    api.types.is_object_dtype : Check whether an array-like or dtype is of the
+        object dtype.
+    api.types.is_numeric_dtype : Check whether the provided array or dtype is
+        of a numeric dtype.
+    api.types.is_categorical_dtype : Check whether an array-like or dtype is of
+        the Categorical dtype.
+
+    Examples
+    --------
+    >>> from pandas.core.dtypes.common import is_interval_dtype
+    >>> is_interval_dtype(object)
+    False
+    >>> is_interval_dtype(pd.IntervalDtype())
+    True
+    >>> is_interval_dtype([1, 2, 3])
+    False
+    >>>
+    >>> interval = pd.Interval(1, 2, closed="right")
+    >>> is_interval_dtype(interval)
+    False
+    >>> is_interval_dtype(pd.IntervalIndex([interval]))
+    True
+    """
+    # GH#52607
+    warnings.warn(
+        "is_interval_dtype is deprecated and will be removed in a future version. "
+        "Use `isinstance(dtype, pd.IntervalDtype)` instead",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    if isinstance(arr_or_dtype, ExtensionDtype):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.type is Interval
+
+    if arr_or_dtype is None:
+        return False
+    return IntervalDtype.is_dtype(arr_or_dtype)
+
+
+def is_categorical_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether an array-like or dtype is of the Categorical dtype.
+
+    .. deprecated:: 2.2.0
+        Use isinstance(dtype, pd.CategoricalDtype) instead.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array-like or dtype is of the Categorical dtype.
+
+    See Also
+    --------
+    api.types.is_list_like: Check if the object is list-like.
+    api.types.is_complex_dtype: Check whether the provided array or
+                                dtype is of a complex dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_categorical_dtype
+    >>> from pandas import CategoricalDtype
+    >>> is_categorical_dtype(object)
+    False
+    >>> is_categorical_dtype(CategoricalDtype())
+    True
+    >>> is_categorical_dtype([1, 2, 3])
+    False
+    >>> is_categorical_dtype(pd.Categorical([1, 2, 3]))
+    True
+    >>> is_categorical_dtype(pd.CategoricalIndex([1, 2, 3]))
+    True
+    """
+    # GH#52527
+    warnings.warn(
+        "is_categorical_dtype is deprecated and will be removed in a future "
+        "version. Use isinstance(dtype, pd.CategoricalDtype) instead",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    if isinstance(arr_or_dtype, ExtensionDtype):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.name == "category"
+
+    if arr_or_dtype is None:
+        return False
+    return CategoricalDtype.is_dtype(arr_or_dtype)
+
+
+def is_string_or_object_np_dtype(dtype: np.dtype) -> bool:
+    """
+    Faster alternative to is_string_dtype, assumes we have a np.dtype object.
+    """
+    return dtype == object or dtype.kind in "SU"
+
+
+def is_string_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of the string dtype.
+
+    If an array is passed with an object dtype, the elements must be
+    inferred as strings.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of the string dtype.
+
+    See Also
+    --------
+    api.types.is_string_dtype : Check whether the provided array or dtype
+                                is of the string dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_string_dtype
+    >>> is_string_dtype(str)
+    True
+    >>> is_string_dtype(object)
+    True
+    >>> is_string_dtype(int)
+    False
+    >>> is_string_dtype(np.array(["a", "b"]))
+    True
+    >>> is_string_dtype(pd.Series([1, 2]))
+    False
+    >>> is_string_dtype(pd.Series([1, 2], dtype=object))
+    False
+    """
+    if hasattr(arr_or_dtype, "dtype") and _get_dtype(arr_or_dtype).kind == "O":
+        return is_all_strings(arr_or_dtype)
+
+    def condition(dtype) -> bool:
+        if is_string_or_object_np_dtype(dtype):
+            return True
+        try:
+            return dtype == "string"
+        except TypeError:
+            return False
+
+    return _is_dtype(arr_or_dtype, condition)
+
+
+def is_dtype_equal(source, target) -> bool:
+    """
+    Check if two dtypes are equal.
+
+    Parameters
+    ----------
+    source : The first dtype to compare
+    target : The second dtype to compare
+
+    Returns
+    -------
+    boolean
+        Whether or not the two dtypes are equal.
+
+    Examples
+    --------
+    >>> is_dtype_equal(int, float)
+    False
+    >>> is_dtype_equal("int", int)
+    True
+    >>> is_dtype_equal(object, "category")
+    False
+    >>> is_dtype_equal(CategoricalDtype(), "category")
+    True
+    >>> is_dtype_equal(DatetimeTZDtype(tz="UTC"), "datetime64")
+    False
+    """
+    if isinstance(target, str):
+        if not isinstance(source, str):
+            # GH#38516 ensure we get the same behavior from
+            #  is_dtype_equal(CDT, "category") and CDT == "category"
+            try:
+                src = _get_dtype(source)
+                if isinstance(src, ExtensionDtype):
+                    return src == target
+            except (TypeError, AttributeError, ImportError):
+                return False
+    elif isinstance(source, str):
+        return is_dtype_equal(target, source)
+
+    try:
+        source = _get_dtype(source)
+        target = _get_dtype(target)
+        return source == target
+    except (TypeError, AttributeError, ImportError):
+        # invalid comparison
+        # object == category will hit this
+        return False
+
+
+def is_integer_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of an integer dtype.
+
+    Unlike in `is_any_int_dtype`, timedelta64 instances will return False.
+
+    The nullable Integer dtypes (e.g. pandas.Int64Dtype) are also considered
+    as integer by this function.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of an integer dtype and
+        not an instance of timedelta64.
+
+    See Also
+    --------
+    api.types.is_integer : Return True if given object is integer.
+    api.types.is_numeric_dtype : Check whether the provided array or dtype is of a
+        numeric dtype.
+    api.types.is_float_dtype : Check whether the provided array or dtype is of a
+        float dtype.
+    Int64Dtype : An ExtensionDtype for Int64Dtype integer data.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_integer_dtype
+    >>> is_integer_dtype(str)
+    False
+    >>> is_integer_dtype(int)
+    True
+    >>> is_integer_dtype(float)
+    False
+    >>> is_integer_dtype(np.uint64)
+    True
+    >>> is_integer_dtype("int8")
+    True
+    >>> is_integer_dtype("Int8")
+    True
+    >>> is_integer_dtype(pd.Int8Dtype)
+    True
+    >>> is_integer_dtype(np.datetime64)
+    False
+    >>> is_integer_dtype(np.timedelta64)
+    False
+    >>> is_integer_dtype(np.array(["a", "b"]))
+    False
+    >>> is_integer_dtype(pd.Series([1, 2]))
+    True
+    >>> is_integer_dtype(np.array([], dtype=np.timedelta64))
+    False
+    >>> is_integer_dtype(pd.Index([1, 2.0]))  # float
+    False
+    """
+    return _is_dtype_type(
+        arr_or_dtype, _classes_and_not_datetimelike(np.integer)
+    ) or _is_dtype(
+        arr_or_dtype, lambda typ: isinstance(typ, ExtensionDtype) and typ.kind in "iu"
+    )
+
+
+def is_signed_integer_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a signed integer dtype.
+
+    Unlike in `is_any_int_dtype`, timedelta64 instances will return False.
+
+    The nullable Integer dtypes (e.g. pandas.Int64Dtype) are also considered
+    as integer by this function.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a signed integer dtype
+        and not an instance of timedelta64.
+
+    See Also
+    --------
+    api.types.is_integer_dtype: Check whether the provided array or dtype
+        is of an integer dtype.
+    api.types.is_numeric_dtype: Check whether the provided array or dtype
+        is of a numeric dtype.
+    api.types.is_unsigned_integer_dtype: Check whether the provided array
+        or dtype is of an unsigned integer dtype.
+
+    Examples
+    --------
+    >>> from pandas.core.dtypes.common import is_signed_integer_dtype
+    >>> is_signed_integer_dtype(str)
+    False
+    >>> is_signed_integer_dtype(int)
+    True
+    >>> is_signed_integer_dtype(float)
+    False
+    >>> is_signed_integer_dtype(np.uint64)  # unsigned
+    False
+    >>> is_signed_integer_dtype("int8")
+    True
+    >>> is_signed_integer_dtype("Int8")
+    True
+    >>> is_signed_integer_dtype(pd.Int8Dtype)
+    True
+    >>> is_signed_integer_dtype(np.datetime64)
+    False
+    >>> is_signed_integer_dtype(np.timedelta64)
+    False
+    >>> is_signed_integer_dtype(np.array(["a", "b"]))
+    False
+    >>> is_signed_integer_dtype(pd.Series([1, 2]))
+    True
+    >>> is_signed_integer_dtype(np.array([], dtype=np.timedelta64))
+    False
+    >>> is_signed_integer_dtype(pd.Index([1, 2.0]))  # float
+    False
+    >>> is_signed_integer_dtype(np.array([1, 2], dtype=np.uint32))  # unsigned
+    False
+    """
+    return _is_dtype_type(
+        arr_or_dtype, _classes_and_not_datetimelike(np.signedinteger)
+    ) or _is_dtype(
+        arr_or_dtype, lambda typ: isinstance(typ, ExtensionDtype) and typ.kind == "i"
+    )
+
+
+def is_unsigned_integer_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of an unsigned integer dtype.
+
+    The nullable Integer dtypes (e.g. pandas.UInt64Dtype) are also
+    considered as integer by this function.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of an unsigned integer dtype.
+
+    See Also
+    --------
+    api.types.is_signed_integer_dtype : Check whether the provided array
+        or dtype is of an signed integer dtype.
+    api.types.is_integer_dtype : Check whether the provided array or dtype
+        is of an integer dtype.
+    api.types.is_numeric_dtype : Check whether the provided array or dtype
+        is of a numeric dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_unsigned_integer_dtype
+    >>> is_unsigned_integer_dtype(str)
+    False
+    >>> is_unsigned_integer_dtype(int)  # signed
+    False
+    >>> is_unsigned_integer_dtype(float)
+    False
+    >>> is_unsigned_integer_dtype(np.uint64)
+    True
+    >>> is_unsigned_integer_dtype("uint8")
+    True
+    >>> is_unsigned_integer_dtype("UInt8")
+    True
+    >>> is_unsigned_integer_dtype(pd.UInt8Dtype)
+    True
+    >>> is_unsigned_integer_dtype(np.array(["a", "b"]))
+    False
+    >>> is_unsigned_integer_dtype(pd.Series([1, 2]))  # signed
+    False
+    >>> is_unsigned_integer_dtype(pd.Index([1, 2.0]))  # float
+    False
+    >>> is_unsigned_integer_dtype(np.array([1, 2], dtype=np.uint32))
+    True
+    """
+    return _is_dtype_type(
+        arr_or_dtype, _classes_and_not_datetimelike(np.unsignedinteger)
+    ) or _is_dtype(
+        arr_or_dtype, lambda typ: isinstance(typ, ExtensionDtype) and typ.kind == "u"
+    )
+
+
+def is_int64_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of the int64 dtype.
+
+    .. deprecated:: 2.1.0
+
+       is_int64_dtype is deprecated and will be removed in a future
+       version. Use dtype == np.int64 instead.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of the int64 dtype.
+
+    See Also
+    --------
+    api.types.is_float_dtype : Check whether the provided array or dtype is of a
+        float dtype.
+    api.types.is_bool_dtype : Check whether the provided array or dtype is of a
+        boolean dtype.
+    api.types.is_object_dtype : Check whether an array-like or dtype is of the
+        object dtype.
+    numpy.int64 : Numpy's 64-bit integer type.
+
+    Notes
+    -----
+    Depending on system architecture, the return value of `is_int64_dtype(
+    int)` will be True if the OS uses 64-bit integers and False if the OS
+    uses 32-bit integers.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_int64_dtype
+    >>> is_int64_dtype(str)  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.int32)  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.int64)  # doctest: +SKIP
+    True
+    >>> is_int64_dtype("int8")  # doctest: +SKIP
+    False
+    >>> is_int64_dtype("Int8")  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(pd.Int64Dtype)  # doctest: +SKIP
+    True
+    >>> is_int64_dtype(float)  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.uint64)  # unsigned  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.array(["a", "b"]))  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.array([1, 2], dtype=np.int64))  # doctest: +SKIP
+    True
+    >>> is_int64_dtype(pd.Index([1, 2.0]))  # float  # doctest: +SKIP
+    False
+    >>> is_int64_dtype(np.array([1, 2], dtype=np.uint32))  # unsigned  # doctest: +SKIP
+    False
+    """
+    # GH#52564
+    warnings.warn(
+        "is_int64_dtype is deprecated and will be removed in a future "
+        "version. Use dtype == np.int64 instead.",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return _is_dtype_type(arr_or_dtype, classes(np.int64))
+
+
+def is_datetime64_any_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of the datetime64 dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    bool
+        Whether or not the array or dtype is of the datetime64 dtype.
+
+    See Also
+    --------
+    api.types.is_datetime64_dtype : Check whether an array-like or dtype is of the
+        datetime64 dtype.
+    api.is_datetime64_ns_dtype : Check whether the provided array or dtype is of the
+        datetime64[ns] dtype.
+    api.is_datetime64tz_dtype : Check whether an array-like or dtype is of a
+        DatetimeTZDtype dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_datetime64_any_dtype
+    >>> from pandas.core.dtypes.dtypes import DatetimeTZDtype
+    >>> is_datetime64_any_dtype(str)
+    False
+    >>> is_datetime64_any_dtype(int)
+    False
+    >>> is_datetime64_any_dtype(np.datetime64)  # can be tz-naive
+    True
+    >>> is_datetime64_any_dtype(DatetimeTZDtype("ns", "US/Eastern"))
+    True
+    >>> is_datetime64_any_dtype(np.array(["a", "b"]))
+    False
+    >>> is_datetime64_any_dtype(np.array([1, 2]))
+    False
+    >>> is_datetime64_any_dtype(np.array([], dtype="datetime64[ns]"))
+    True
+    >>> is_datetime64_any_dtype(pd.DatetimeIndex([1, 2, 3], dtype="datetime64[ns]"))
+    True
+    """
+    if isinstance(arr_or_dtype, (np.dtype, ExtensionDtype)):
+        # GH#33400 fastpath for dtype object
+        return arr_or_dtype.kind == "M"
+
+    if arr_or_dtype is None:
+        return False
+
+    try:
+        tipo = _get_dtype(arr_or_dtype)
+    except TypeError:
+        return False
+    return (
+        lib.is_np_dtype(tipo, "M")
+        or isinstance(tipo, DatetimeTZDtype)
+        or (isinstance(tipo, ExtensionDtype) and tipo.kind == "M")
+    )
+
+
+def is_datetime64_ns_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of the datetime64[ns] dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    bool
+        Whether or not the array or dtype is of the datetime64[ns] dtype.
+
+    See Also
+    --------
+    api.types.is_datetime64_dtype: Check whether an array-like or
+                                        dtype is of the datetime64 dtype.
+    api.types.is_datetime64_any_dtype: Check whether the provided array or
+                                        dtype is of the datetime64 dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_datetime64_ns_dtype
+    >>> from pandas.core.dtypes.dtypes import DatetimeTZDtype
+    >>> is_datetime64_ns_dtype(str)
+    False
+    >>> is_datetime64_ns_dtype(int)
+    False
+    >>> is_datetime64_ns_dtype(np.datetime64)  # no unit
+    False
+    >>> is_datetime64_ns_dtype(DatetimeTZDtype("ns", "US/Eastern"))
+    True
+    >>> is_datetime64_ns_dtype(np.array(["a", "b"]))
+    False
+    >>> is_datetime64_ns_dtype(np.array([1, 2]))
+    False
+    >>> is_datetime64_ns_dtype(np.array([], dtype="datetime64"))  # no unit
+    False
+    >>> is_datetime64_ns_dtype(np.array([], dtype="datetime64[ps]"))  # wrong unit
+    False
+    >>> is_datetime64_ns_dtype(pd.DatetimeIndex([1, 2, 3], dtype="datetime64[ns]"))
+    True
+    """
+    if arr_or_dtype is None:
+        return False
+    try:
+        tipo = _get_dtype(arr_or_dtype)
+    except TypeError:
+        return False
+    return tipo == DT64NS_DTYPE or (
+        isinstance(tipo, DatetimeTZDtype) and tipo.unit == "ns"
+    )
+
+
+def is_timedelta64_ns_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of the timedelta64[ns] dtype.
+
+    This is a very specific dtype, so generic ones like `np.timedelta64`
+    will return False if passed into this function.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of the timedelta64[ns] dtype.
+
+    See Also
+    --------
+    api.types.is_timedelta64_dtype: Check whether an array-like or dtype
+        is of the timedelta64 dtype.
+
+    Examples
+    --------
+    >>> from pandas.core.dtypes.common import is_timedelta64_ns_dtype
+    >>> is_timedelta64_ns_dtype(np.dtype("m8[ns]"))
+    True
+    >>> is_timedelta64_ns_dtype(np.dtype("m8[ps]"))  # Wrong frequency
+    False
+    >>> is_timedelta64_ns_dtype(np.array([1, 2], dtype="m8[ns]"))
+    True
+    >>> is_timedelta64_ns_dtype(np.array([1, 2], dtype=np.timedelta64))
+    False
+    """
+    return _is_dtype(arr_or_dtype, lambda dtype: dtype == TD64NS_DTYPE)
+
+
+# This exists to silence numpy deprecation warnings, see GH#29553
+def is_numeric_v_string_like(a: ArrayLike, b) -> bool:
+    """
+    Check if we are comparing a string-like object to a numeric ndarray.
+    NumPy doesn't like to compare such objects, especially numeric arrays
+    and scalar string-likes.
+
+    Parameters
+    ----------
+    a : array-like, scalar
+        The first object to check.
+    b : array-like, scalar
+        The second object to check.
+
+    Returns
+    -------
+    boolean
+        Whether we return a comparing a string-like object to a numeric array.
+
+    Examples
+    --------
+    >>> is_numeric_v_string_like(np.array([1]), "foo")
+    True
+    >>> is_numeric_v_string_like(np.array([1, 2]), np.array(["foo"]))
+    True
+    >>> is_numeric_v_string_like(np.array(["foo"]), np.array([1, 2]))
+    True
+    >>> is_numeric_v_string_like(np.array([1]), np.array([2]))
+    False
+    >>> is_numeric_v_string_like(np.array(["foo"]), np.array(["foo"]))
+    False
+    """
+    is_a_array = isinstance(a, np.ndarray)
+    is_b_array = isinstance(b, np.ndarray)
+
+    is_a_numeric_array = is_a_array and a.dtype.kind in ("u", "i", "f", "c", "b")
+    is_b_numeric_array = is_b_array and b.dtype.kind in ("u", "i", "f", "c", "b")
+    is_a_string_array = is_a_array and a.dtype.kind in ("S", "U")
+    is_b_string_array = is_b_array and b.dtype.kind in ("S", "U")
+
+    is_b_scalar_string_like = not is_b_array and isinstance(b, str)
+
+    return (
+        (is_a_numeric_array and is_b_scalar_string_like)
+        or (is_a_numeric_array and is_b_string_array)
+        or (is_b_numeric_array and is_a_string_array)
+    )
+
+
+def needs_i8_conversion(dtype: DtypeObj | None) -> bool:
+    """
+    Check whether the dtype should be converted to int64.
+
+    Dtype "needs" such a conversion if the dtype is of a datetime-like dtype
+
+    Parameters
+    ----------
+    dtype : np.dtype, ExtensionDtype, or None
+
+    Returns
+    -------
+    boolean
+        Whether or not the dtype should be converted to int64.
+
+    Examples
+    --------
+    >>> needs_i8_conversion(str)
+    False
+    >>> needs_i8_conversion(np.int64)
+    False
+    >>> needs_i8_conversion(np.datetime64)
+    False
+    >>> needs_i8_conversion(np.dtype(np.datetime64))
+    True
+    >>> needs_i8_conversion(np.array(["a", "b"]))
+    False
+    >>> needs_i8_conversion(pd.Series([1, 2]))
+    False
+    >>> needs_i8_conversion(pd.Series([], dtype="timedelta64[ns]"))
+    False
+    >>> needs_i8_conversion(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern"))
+    False
+    >>> needs_i8_conversion(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern").dtype)
+    True
+    """
+    if isinstance(dtype, np.dtype):
+        return dtype.kind in "mM"
+    return isinstance(dtype, (PeriodDtype, DatetimeTZDtype))
+
+
+def is_numeric_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a numeric dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a numeric dtype.
+
+    See Also
+    --------
+    api.types.is_integer_dtype: Check whether the provided array or dtype
+        is of an integer dtype.
+    api.types.is_unsigned_integer_dtype: Check whether the provided array
+        or dtype is of an unsigned integer dtype.
+    api.types.is_signed_integer_dtype: Check whether the provided array
+        or dtype is of an signed integer dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_numeric_dtype
+    >>> is_numeric_dtype(str)
+    False
+    >>> is_numeric_dtype(int)
+    True
+    >>> is_numeric_dtype(float)
+    True
+    >>> is_numeric_dtype(np.uint64)
+    True
+    >>> is_numeric_dtype(np.datetime64)
+    False
+    >>> is_numeric_dtype(np.timedelta64)
+    False
+    >>> is_numeric_dtype(np.array(["a", "b"]))
+    False
+    >>> is_numeric_dtype(pd.Series([1, 2]))
+    True
+    >>> is_numeric_dtype(pd.Index([1, 2.0]))
+    True
+    >>> is_numeric_dtype(np.array([], dtype=np.timedelta64))
+    False
+    """
+    return _is_dtype_type(
+        arr_or_dtype, _classes_and_not_datetimelike(np.number, np.bool_)
+    ) or _is_dtype(
+        arr_or_dtype, lambda typ: isinstance(typ, ExtensionDtype) and typ._is_numeric
+    )
+
+
+def is_any_real_numeric_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a real number dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a real number dtype.
+
+    See Also
+    --------
+    is_numeric_dtype : Check if a dtype is numeric.
+    is_complex_dtype : Check if a dtype is complex.
+    is_bool_dtype : Check if a dtype is boolean.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_any_real_numeric_dtype
+    >>> is_any_real_numeric_dtype(int)
+    True
+    >>> is_any_real_numeric_dtype(float)
+    True
+    >>> is_any_real_numeric_dtype(object)
+    False
+    >>> is_any_real_numeric_dtype(str)
+    False
+    >>> is_any_real_numeric_dtype(complex(1, 2))
+    False
+    >>> is_any_real_numeric_dtype(bool)
+    False
+    """
+    return (
+        is_numeric_dtype(arr_or_dtype)
+        and not is_complex_dtype(arr_or_dtype)
+        and not is_bool_dtype(arr_or_dtype)
+    )
+
+
+def is_float_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a float dtype.
+
+    The function checks for floating-point data types, which represent real numbers
+    that may have fractional components.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a float dtype.
+
+    See Also
+    --------
+    api.types.is_numeric_dtype : Check whether the provided array or dtype is of
+        a numeric dtype.
+    api.types.is_integer_dtype : Check whether the provided array or dtype is of
+        an integer dtype.
+    api.types.is_object_dtype : Check whether an array-like or dtype is of the
+        object dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_float_dtype
+    >>> is_float_dtype(str)
+    False
+    >>> is_float_dtype(int)
+    False
+    >>> is_float_dtype(float)
+    True
+    >>> is_float_dtype(np.array(["a", "b"]))
+    False
+    >>> is_float_dtype(pd.Series([1, 2]))
+    False
+    >>> is_float_dtype(pd.Index([1, 2.0]))
+    True
+    """
+    return _is_dtype_type(arr_or_dtype, classes(np.floating)) or _is_dtype(
+        arr_or_dtype, lambda typ: isinstance(typ, ExtensionDtype) and typ.kind in "f"
+    )
+
+
+def is_bool_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a boolean dtype.
+
+    This function verifies whether a given object is a boolean data type. The input
+    can be an array or a dtype object. Accepted array types include instances
+    of ``np.array``, ``pd.Series``, ``pd.Index``, and similar array-like structures.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a boolean dtype.
+
+    See Also
+    --------
+    api.types.is_bool : Check if an object is a boolean.
+
+    Notes
+    -----
+    An ExtensionArray is considered boolean when the ``_is_boolean``
+    attribute is set to True.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_bool_dtype
+    >>> is_bool_dtype(str)
+    False
+    >>> is_bool_dtype(int)
+    False
+    >>> is_bool_dtype(bool)
+    True
+    >>> is_bool_dtype(np.bool_)
+    True
+    >>> is_bool_dtype(np.array(["a", "b"]))
+    False
+    >>> is_bool_dtype(pd.Series([1, 2]))
+    False
+    >>> is_bool_dtype(np.array([True, False]))
+    True
+    >>> is_bool_dtype(pd.Categorical([True, False]))
+    True
+    >>> is_bool_dtype(pd.arrays.SparseArray([True, False]))
+    True
+    """
+    if arr_or_dtype is None:
+        return False
+    try:
+        dtype = _get_dtype(arr_or_dtype)
+    except (TypeError, ValueError):
+        return False
+
+    if isinstance(dtype, CategoricalDtype):
+        arr_or_dtype = dtype.categories
+        # now we use the special definition for Index
+
+    if isinstance(arr_or_dtype, ABCIndex):
+        # Allow Index[object] that is all-bools or Index["boolean"]
+        if arr_or_dtype.inferred_type == "boolean":
+            if not is_bool_dtype(arr_or_dtype.dtype):
+                # GH#52680
+                warnings.warn(
+                    "The behavior of is_bool_dtype with an object-dtype Index "
+                    "of bool objects is deprecated. In a future version, "
+                    "this will return False. Cast the Index to a bool dtype instead.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+            return True
+        return False
+    elif isinstance(dtype, ExtensionDtype):
+        return getattr(dtype, "_is_boolean", False)
+
+    return issubclass(dtype.type, np.bool_)
+
+
+def is_1d_only_ea_dtype(dtype: DtypeObj | None) -> bool:
+    """
+    Analogue to is_extension_array_dtype but excluding DatetimeTZDtype.
+    """
+    return isinstance(dtype, ExtensionDtype) and not dtype._supports_2d
+
+
+def is_extension_array_dtype(arr_or_dtype) -> bool:
+    """
+    Check if an object is a pandas extension array type.
+
+    See the :ref:`Use Guide <extending.extension-types>` for more.
+
+    Parameters
+    ----------
+    arr_or_dtype : object
+        For array-like input, the ``.dtype`` attribute will
+        be extracted.
+
+    Returns
+    -------
+    bool
+        Whether the `arr_or_dtype` is an extension array type.
+
+    See Also
+    --------
+    api.extensions.ExtensionArray : Abstract base class for pandas extension arrays.
+
+    Notes
+    -----
+    This checks whether an object implements the pandas extension
+    array interface. In pandas, this includes:
+
+    * Categorical
+    * Sparse
+    * Interval
+    * Period
+    * DatetimeArray
+    * TimedeltaArray
+
+    Third-party libraries may implement arrays or types satisfying
+    this interface as well.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_extension_array_dtype
+    >>> arr = pd.Categorical(["a", "b"])
+    >>> is_extension_array_dtype(arr)
+    True
+    >>> is_extension_array_dtype(arr.dtype)
+    True
+
+    >>> arr = np.array(["a", "b"])
+    >>> is_extension_array_dtype(arr.dtype)
+    False
+    """
+    dtype = getattr(arr_or_dtype, "dtype", arr_or_dtype)
+    if isinstance(dtype, ExtensionDtype):
+        return True
+    elif isinstance(dtype, np.dtype):
+        return False
+    else:
+        try:
+            with warnings.catch_warnings():
+                # pandas_dtype(..) can raise UserWarning for class input
+                warnings.simplefilter("ignore", UserWarning)
+                dtype = pandas_dtype(dtype)
+        except (TypeError, ValueError):
+            # np.dtype(..) can raise ValueError
+            return False
+        return isinstance(dtype, ExtensionDtype)
+
+
+def is_ea_or_datetimelike_dtype(dtype: DtypeObj | None) -> bool:
+    """
+    Check for ExtensionDtype, datetime64 dtype, or timedelta64 dtype.
+
+    Notes
+    -----
+    Checks only for dtype objects, not dtype-castable strings or types.
+    """
+    return isinstance(dtype, ExtensionDtype) or (lib.is_np_dtype(dtype, "mM"))
+
+
+def is_complex_dtype(arr_or_dtype) -> bool:
+    """
+    Check whether the provided array or dtype is of a complex dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array or dtype to check.
+
+    Returns
+    -------
+    boolean
+        Whether or not the array or dtype is of a complex dtype.
+
+    See Also
+    --------
+    api.types.is_complex: Return True if given object is complex.
+    api.types.is_numeric_dtype: Check whether the provided array or
+                                dtype is of a numeric dtype.
+    api.types.is_integer_dtype: Check whether the provided array or
+                                dtype is of an integer dtype.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_complex_dtype
+    >>> is_complex_dtype(str)
+    False
+    >>> is_complex_dtype(int)
+    False
+    >>> is_complex_dtype(np.complex128)
+    True
+    >>> is_complex_dtype(np.array(["a", "b"]))
+    False
+    >>> is_complex_dtype(pd.Series([1, 2]))
+    False
+    >>> is_complex_dtype(np.array([1 + 1j, 5]))
+    True
+    """
+    return _is_dtype_type(arr_or_dtype, classes(np.complexfloating))
+
+
+def _is_dtype(arr_or_dtype, condition) -> bool:
+    """
+    Return true if the condition is satisfied for the arr_or_dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like, str, np.dtype, or ExtensionArrayType
+        The array-like or dtype object whose dtype we want to extract.
+    condition : callable[Union[np.dtype, ExtensionDtype]]
+
+    Returns
+    -------
+    bool
+
+    """
+    if arr_or_dtype is None:
+        return False
+    try:
+        dtype = _get_dtype(arr_or_dtype)
+    except (TypeError, ValueError):
+        return False
+    return condition(dtype)
+
+
+def _get_dtype(arr_or_dtype) -> DtypeObj:
+    """
+    Get the dtype instance associated with an array
+    or dtype object.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype object whose dtype we want to extract.
+
+    Returns
+    -------
+    obj_dtype : The extract dtype instance from the
+                passed in array or dtype object.
+
+    Raises
+    ------
+    TypeError : The passed in object is None.
+    """
+    if arr_or_dtype is None:
+        raise TypeError("Cannot deduce dtype from null object")
+
+    # fastpath
+    if isinstance(arr_or_dtype, np.dtype):
+        return arr_or_dtype
+    elif isinstance(arr_or_dtype, type):
+        return np.dtype(arr_or_dtype)
+
+    # if we have an array-like
+    elif hasattr(arr_or_dtype, "dtype"):
+        arr_or_dtype = arr_or_dtype.dtype
+
+    return pandas_dtype(arr_or_dtype)
+
+
+def _is_dtype_type(arr_or_dtype, condition) -> bool:
+    """
+    Return true if the condition is satisfied for the arr_or_dtype.
+
+    Parameters
+    ----------
+    arr_or_dtype : array-like or dtype
+        The array-like or dtype object whose dtype we want to extract.
+    condition : callable[Union[np.dtype, ExtensionDtypeType]]
+
+    Returns
+    -------
+    bool : if the condition is satisfied for the arr_or_dtype
+    """
+    if arr_or_dtype is None:
+        return condition(type(None))
+
+    # fastpath
+    if isinstance(arr_or_dtype, np.dtype):
+        return condition(arr_or_dtype.type)
+    elif isinstance(arr_or_dtype, type):
+        if issubclass(arr_or_dtype, ExtensionDtype):
+            arr_or_dtype = arr_or_dtype.type
+        return condition(np.dtype(arr_or_dtype).type)
+
+    # if we have an array-like
+    if hasattr(arr_or_dtype, "dtype"):
+        arr_or_dtype = arr_or_dtype.dtype
+
+    # we are not possibly a dtype
+    elif is_list_like(arr_or_dtype):
+        return condition(type(None))
+
+    try:
+        tipo = pandas_dtype(arr_or_dtype).type
+    except (TypeError, ValueError):
+        if is_scalar(arr_or_dtype):
+            return condition(type(None))
+
+        return False
+
+    return condition(tipo)
+
+
+def infer_dtype_from_object(dtype) -> type:
+    """
+    Get a numpy dtype.type-style object for a dtype object.
+
+    This methods also includes handling of the datetime64[ns] and
+    datetime64[ns, TZ] objects.
+
+    If no dtype can be found, we return ``object``.
+
+    Parameters
+    ----------
+    dtype : dtype, type
+        The dtype object whose numpy dtype.type-style
+        object we want to extract.
+
+    Returns
+    -------
+    type
+    """
+    if isinstance(dtype, type) and issubclass(dtype, np.generic):
+        # Type object from a dtype
+
+        return dtype
+    elif isinstance(dtype, (np.dtype, ExtensionDtype)):
+        # dtype object
+        try:
+            _validate_date_like_dtype(dtype)
+        except TypeError:
+            # Should still pass if we don't have a date-like
+            pass
+        if hasattr(dtype, "numpy_dtype"):
+            # TODO: Implement this properly
+            # https://github.com/pandas-dev/pandas/issues/52576
+            return dtype.numpy_dtype.type
+        return dtype.type
+
+    try:
+        dtype = pandas_dtype(dtype)
+    except TypeError:
+        pass
+
+    if isinstance(dtype, ExtensionDtype):
+        return dtype.type
+    elif isinstance(dtype, str):
+        # TODO(jreback)
+        # should deprecate these
+        if dtype in ["datetimetz", "datetime64tz"]:
+            return DatetimeTZDtype.type
+        elif dtype in ["period"]:
+            raise NotImplementedError
+
+        if dtype in ["datetime", "timedelta"]:
+            dtype += "64"
+        try:
+            return infer_dtype_from_object(getattr(np, dtype))
+        except (AttributeError, TypeError):
+            # Handles cases like _get_dtype(int) i.e.,
+            # Python objects that are valid dtypes
+            # (unlike user-defined types, in general)
+            #
+            # TypeError handles the float16 type code of 'e'
+            # further handle internal types
+            pass
+
+    return infer_dtype_from_object(np.dtype(dtype))
+
+
+def _validate_date_like_dtype(dtype) -> None:
+    """
+    Check whether the dtype is a date-like dtype. Raises an error if invalid.
+
+    Parameters
+    ----------
+    dtype : dtype, type
+        The dtype to check.
+
+    Raises
+    ------
+    TypeError : The dtype could not be casted to a date-like dtype.
+    ValueError : The dtype is an illegal date-like dtype (e.g. the
+                 frequency provided is too specific)
+    """
+    try:
+        typ = np.datetime_data(dtype)[0]
+    except ValueError as e:
+        raise TypeError(e) from e
+    if typ not in ["generic", "ns"]:
+        raise ValueError(
+            f"{dtype.name!r} is too specific of a frequency, "
+            f"try passing {dtype.type.__name__!r}"
+        )
+
+
+def validate_all_hashable(*args, error_name: str | None = None) -> None:
+    """
+    Return None if all args are hashable, else raise a TypeError.
+
+    Parameters
+    ----------
+    *args
+        Arguments to validate.
+    error_name : str, optional
+        The name to use if error
+
+    Raises
+    ------
+    TypeError : If an argument is not hashable
+
+    Returns
+    -------
+    None
+    """
+    if not all(is_hashable(arg) for arg in args):
+        if error_name:
+            raise TypeError(f"{error_name} must be a hashable type")
+        raise TypeError("All elements must be hashable")
+
+
+def pandas_dtype(dtype) -> DtypeObj:
+    """
+    Convert input into a pandas only dtype object or a numpy dtype object.
+
+    Parameters
+    ----------
+    dtype : object
+        The object to be converted into a dtype.
+
+    Returns
+    -------
+    np.dtype or a pandas dtype
+        The converted dtype, which can be either a numpy dtype or a pandas dtype.
+
+    Raises
+    ------
+    TypeError if not a dtype
+
+    See Also
+    --------
+    api.types.is_dtype : Return true if the condition is satisfied for the arr_or_dtype.
+
+    Examples
+    --------
+    >>> pd.api.types.pandas_dtype(int)
+    dtype('int64')
+    """
+    # short-circuit
+    if isinstance(dtype, np.ndarray):
+        return dtype.dtype
+    elif isinstance(dtype, (np.dtype, ExtensionDtype)):
+        return dtype
+
+    # builtin aliases
+    if dtype is str and using_string_dtype():
+        from pandas.core.arrays.string_ import StringDtype
+
+        return StringDtype(na_value=np.nan)
+
+    # registered extension types
+    result = registry.find(dtype)
+    if result is not None:
+        if isinstance(result, type):
+            # GH 31356, GH 54592
+            warnings.warn(
+                f"Instantiating {result.__name__} without any arguments."
+                f"Pass a {result.__name__} instance to silence this warning.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+            result = result()
+        return result
+
+    # try a numpy dtype
+    # raise a consistent TypeError if failed
+    try:
+        with warnings.catch_warnings():
+            # TODO: warnings.catch_warnings can be removed when numpy>2.3.0
+            # is the minimum version
+            # GH#51523 - Series.astype(np.integer) doesn't show
+            # numpy deprecation warning of np.integer
+            # Hence enabling DeprecationWarning
+            warnings.simplefilter("always", DeprecationWarning)
+            npdtype = np.dtype(dtype)
+    except SyntaxError as err:
+        # np.dtype uses `eval` which can raise SyntaxError
+        raise TypeError(f"data type '{dtype}' not understood") from err
+
+    # Any invalid dtype (such as pd.Timestamp) should raise an error.
+    # np.dtype(invalid_type).kind = 0 for such objects. However, this will
+    # also catch some valid dtypes such as object, np.object_ and 'object'
+    # which we safeguard against by catching them earlier and returning
+    # np.dtype(valid_dtype) before this condition is evaluated.
+    if is_hashable(dtype) and dtype in [
+        object,
+        np.object_,
+        "object",
+        "O",
+        "object_",
+    ]:
+        # check hashability to avoid errors/DeprecationWarning when we get
+        # here and `dtype` is an array
+        return npdtype
+    elif npdtype.kind == "O":
+        raise TypeError(f"dtype '{dtype}' not understood")
+
+    return npdtype
+
+
+def is_all_strings(value: ArrayLike) -> bool:
+    """
+    Check if this is an array of strings that we should try parsing.
+
+    Includes object-dtype ndarray containing all-strings, StringArray,
+    and Categorical with all-string categories.
+    Does not include numpy string dtypes.
+    """
+    dtype = value.dtype
+
+    if isinstance(dtype, np.dtype):
+        if len(value) == 0:
+            return dtype == np.dtype("object")
+        else:
+            return dtype == np.dtype("object") and lib.is_string_array(
+                np.asarray(value), skipna=False
+            )
+    elif isinstance(dtype, CategoricalDtype):
+        return dtype.categories.inferred_type == "string"
+    return dtype == "string"
+
+
+__all__ = [
+    "DT64NS_DTYPE",
+    "INT64_DTYPE",
+    "TD64NS_DTYPE",
+    "classes",
+    "ensure_float64",
+    "ensure_python_int",
+    "ensure_str",
+    "infer_dtype_from_object",
+    "is_1d_only_ea_dtype",
+    "is_all_strings",
+    "is_any_real_numeric_dtype",
+    "is_array_like",
+    "is_bool",
+    "is_bool_dtype",
+    "is_categorical_dtype",
+    "is_complex",
+    "is_complex_dtype",
+    "is_dataclass",
+    "is_datetime64_any_dtype",
+    "is_datetime64_dtype",
+    "is_datetime64_ns_dtype",
+    "is_datetime64tz_dtype",
+    "is_decimal",
+    "is_dict_like",
+    "is_dtype_equal",
+    "is_ea_or_datetimelike_dtype",
+    "is_extension_array_dtype",
+    "is_file_like",
+    "is_float_dtype",
+    "is_int64_dtype",
+    "is_integer_dtype",
+    "is_interval_dtype",
+    "is_iterator",
+    "is_named_tuple",
+    "is_nested_list_like",
+    "is_number",
+    "is_numeric_dtype",
+    "is_object_dtype",
+    "is_period_dtype",
+    "is_re",
+    "is_re_compilable",
+    "is_scipy_sparse",
+    "is_sequence",
+    "is_signed_integer_dtype",
+    "is_sparse",
+    "is_string_dtype",
+    "is_string_or_object_np_dtype",
+    "is_timedelta64_dtype",
+    "is_timedelta64_ns_dtype",
+    "is_unsigned_integer_dtype",
+    "needs_i8_conversion",
+    "pandas_dtype",
+    "validate_all_hashable",
+]
diff --git a/pandas/pandas/core/dtypes/concat.py b/pandas/pandas/core/dtypes/concat.py
new file mode 100644
index 0000000000000..dcf8cb5c78536
--- /dev/null
+++ b/pandas/pandas/core/dtypes/concat.py
@@ -0,0 +1,335 @@
+"""
+Utility functions related to concat.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.astype import astype_array
+from pandas.core.dtypes.cast import (
+    common_dtype_categorical_compat,
+    find_common_type,
+    np_find_common_type,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+from pandas.core.dtypes.generic import (
+    ABCCategoricalIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        DtypeObj,
+    )
+
+    from pandas.core.arrays import (
+        Categorical,
+        ExtensionArray,
+    )
+
+
+def _is_nonempty(x: ArrayLike, axis: AxisInt) -> bool:
+    # filter empty arrays
+    # 1-d dtypes always are included here
+    if x.ndim <= axis:
+        return True
+    return x.shape[axis] > 0
+
+
+def concat_compat(
+    to_concat: Sequence[ArrayLike], axis: AxisInt = 0, ea_compat_axis: bool = False
+) -> ArrayLike:
+    """
+    provide concatenation of an array of arrays each of which is a single
+    'normalized' dtypes (in that for example, if it's object, then it is a
+    non-datetimelike and provide a combined dtype for the resulting array that
+    preserves the overall dtype if possible)
+
+    Parameters
+    ----------
+    to_concat : sequence of arrays
+    axis : axis to provide concatenation
+    ea_compat_axis : bool, default False
+        For ExtensionArray compat, behave as if axis == 1 when determining
+        whether to drop empty arrays.
+
+    Returns
+    -------
+    a single array, preserving the combined dtypes
+    """
+    if len(to_concat) and lib.dtypes_all_equal([obj.dtype for obj in to_concat]):
+        # fastpath!
+        obj = to_concat[0]
+        if isinstance(obj, np.ndarray):
+            to_concat_arrs = cast("Sequence[np.ndarray]", to_concat)
+            return np.concatenate(to_concat_arrs, axis=axis)
+
+        to_concat_eas = cast("Sequence[ExtensionArray]", to_concat)
+        if ea_compat_axis:
+            # We have 1D objects, that don't support axis keyword
+            return obj._concat_same_type(to_concat_eas)
+        elif axis == 0:
+            return obj._concat_same_type(to_concat_eas)
+        else:
+            # e.g. DatetimeArray
+            # NB: We are assuming here that ensure_wrapped_if_arraylike has
+            #  been called where relevant.
+            return obj._concat_same_type(
+                # error: Unexpected keyword argument "axis" for "_concat_same_type"
+                # of "ExtensionArray"
+                to_concat_eas,
+                axis=axis,  # type: ignore[call-arg]
+            )
+
+    # If all arrays are empty, there's nothing to convert, just short-cut to
+    # the concatenation, #3121.
+    #
+    # Creating an empty array directly is tempting, but the winnings would be
+    # marginal given that it would still require shape & dtype calculation and
+    # np.concatenate which has them both implemented is compiled.
+    non_empties = [x for x in to_concat if _is_nonempty(x, axis)]
+
+    any_ea, kinds, target_dtype = _get_result_dtype(to_concat, non_empties)
+
+    if target_dtype is not None:
+        to_concat = [astype_array(arr, target_dtype, copy=False) for arr in to_concat]
+
+    if not isinstance(to_concat[0], np.ndarray):
+        # i.e. isinstance(to_concat[0], ExtensionArray)
+        to_concat_eas = cast("Sequence[ExtensionArray]", to_concat)
+        cls = type(to_concat[0])
+        # GH#53640: eg. for datetime array, axis=1 but 0 is default
+        # However, class method `_concat_same_type()` for some classes
+        # may not support the `axis` keyword
+        if ea_compat_axis or axis == 0:
+            return cls._concat_same_type(to_concat_eas)
+        else:
+            return cls._concat_same_type(
+                to_concat_eas,
+                axis=axis,  # type: ignore[call-arg]
+            )
+    else:
+        to_concat_arrs = cast("Sequence[np.ndarray]", to_concat)
+        result = np.concatenate(to_concat_arrs, axis=axis)
+
+        if not any_ea and "b" in kinds and result.dtype.kind in "iuf":
+            # GH#39817 cast to object instead of casting bools to numeric
+            result = result.astype(object, copy=False)
+    return result
+
+
+def _get_result_dtype(
+    to_concat: Sequence[ArrayLike], non_empties: Sequence[ArrayLike]
+) -> tuple[bool, set[str], DtypeObj | None]:
+    target_dtype = None
+
+    dtypes = {obj.dtype for obj in to_concat}
+    kinds = {obj.dtype.kind for obj in to_concat}
+
+    any_ea = any(not isinstance(x, np.ndarray) for x in to_concat)
+    if any_ea:
+        # i.e. any ExtensionArrays
+
+        # we ignore axis here, as internally concatting with EAs is always
+        # for axis=0
+        if len(dtypes) != 1:
+            target_dtype = find_common_type([x.dtype for x in to_concat])
+            target_dtype = common_dtype_categorical_compat(to_concat, target_dtype)
+
+    elif not len(non_empties):
+        # we have all empties, but may need to coerce the result dtype to
+        # object if we have non-numeric type operands (numpy would otherwise
+        # cast this to float)
+        if len(kinds) != 1:
+            if not len(kinds - {"i", "u", "f"}) or not len(kinds - {"b", "i", "u"}):
+                # let numpy coerce
+                pass
+            else:
+                # coerce to object
+                target_dtype = np.dtype(object)
+                kinds = {"o"}
+    else:
+        # error: Argument 1 to "np_find_common_type" has incompatible type
+        # "*Set[Union[ExtensionDtype, Any]]"; expected "dtype[Any]"
+        target_dtype = np_find_common_type(*dtypes)  # type: ignore[arg-type]
+
+    return any_ea, kinds, target_dtype
+
+
+def union_categoricals(
+    to_union, sort_categories: bool = False, ignore_order: bool = False
+) -> Categorical:
+    """
+    Combine list-like of Categorical-like, unioning categories.
+
+    All categories must have the same dtype.
+
+    Parameters
+    ----------
+    to_union : list-like
+        Categorical, CategoricalIndex, or Series with dtype='category'.
+    sort_categories : bool, default False
+        If true, resulting categories will be lexsorted, otherwise
+        they will be ordered as they appear in the data.
+    ignore_order : bool, default False
+        If true, the ordered attribute of the Categoricals will be ignored.
+        Results in an unordered categorical.
+
+    Returns
+    -------
+    Categorical
+        The union of categories being combined.
+
+    Raises
+    ------
+    TypeError
+        - all inputs do not have the same dtype
+        - all inputs do not have the same ordered property
+        - all inputs are ordered and their categories are not identical
+        - sort_categories=True and Categoricals are ordered
+    ValueError
+        Empty list of categoricals passed
+
+    See Also
+    --------
+    CategoricalDtype : Type for categorical data with the categories and orderedness.
+    Categorical : Represent a categorical variable in classic R / S-plus fashion.
+
+    Notes
+    -----
+    To learn more about categories, see `link
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html#unioning>`__
+
+    Examples
+    --------
+    If you want to combine categoricals that do not necessarily have
+    the same categories, `union_categoricals` will combine a list-like
+    of categoricals. The new categories will be the union of the
+    categories being combined.
+
+    >>> a = pd.Categorical(["b", "c"])
+    >>> b = pd.Categorical(["a", "b"])
+    >>> pd.api.types.union_categoricals([a, b])
+    ['b', 'c', 'a', 'b']
+    Categories (3, object): ['b', 'c', 'a']
+
+    By default, the resulting categories will be ordered as they appear
+    in the `categories` of the data. If you want the categories to be
+    lexsorted, use `sort_categories=True` argument.
+
+    >>> pd.api.types.union_categoricals([a, b], sort_categories=True)
+    ['b', 'c', 'a', 'b']
+    Categories (3, object): ['a', 'b', 'c']
+
+    `union_categoricals` also works with the case of combining two
+    categoricals of the same categories and order information (e.g. what
+    you could also `append` for).
+
+    >>> a = pd.Categorical(["a", "b"], ordered=True)
+    >>> b = pd.Categorical(["a", "b", "a"], ordered=True)
+    >>> pd.api.types.union_categoricals([a, b])
+    ['a', 'b', 'a', 'b', 'a']
+    Categories (2, object): ['a' < 'b']
+
+    Raises `TypeError` because the categories are ordered and not identical.
+
+    >>> a = pd.Categorical(["a", "b"], ordered=True)
+    >>> b = pd.Categorical(["a", "b", "c"], ordered=True)
+    >>> pd.api.types.union_categoricals([a, b])
+    Traceback (most recent call last):
+        ...
+    TypeError: to union ordered Categoricals, all categories must be the same
+
+    Ordered categoricals with different categories or orderings can be
+    combined by using the `ignore_ordered=True` argument.
+
+    >>> a = pd.Categorical(["a", "b", "c"], ordered=True)
+    >>> b = pd.Categorical(["c", "b", "a"], ordered=True)
+    >>> pd.api.types.union_categoricals([a, b], ignore_order=True)
+    ['a', 'b', 'c', 'c', 'b', 'a']
+    Categories (3, object): ['a', 'b', 'c']
+
+    `union_categoricals` also works with a `CategoricalIndex`, or `Series`
+    containing categorical data, but note that the resulting array will
+    always be a plain `Categorical`
+
+    >>> a = pd.Series(["b", "c"], dtype="category")
+    >>> b = pd.Series(["a", "b"], dtype="category")
+    >>> pd.api.types.union_categoricals([a, b])
+    ['b', 'c', 'a', 'b']
+    Categories (3, object): ['b', 'c', 'a']
+    """
+    from pandas import Categorical
+    from pandas.core.arrays.categorical import recode_for_categories
+
+    if len(to_union) == 0:
+        raise ValueError("No Categoricals to union")
+
+    def _maybe_unwrap(x):
+        if isinstance(x, (ABCCategoricalIndex, ABCSeries)):
+            return x._values
+        elif isinstance(x, Categorical):
+            return x
+        else:
+            raise TypeError("all components to combine must be Categorical")
+
+    to_union = [_maybe_unwrap(x) for x in to_union]
+    first = to_union[0]
+
+    if not lib.dtypes_all_equal([obj.categories.dtype for obj in to_union]):
+        raise TypeError("dtype of categories must be the same")
+
+    ordered = False
+    if all(first._categories_match_up_to_permutation(other) for other in to_union[1:]):
+        # identical categories - fastpath
+        categories = first.categories
+        ordered = first.ordered
+
+        all_codes = [first._encode_with_my_categories(x)._codes for x in to_union]
+        new_codes = np.concatenate(all_codes)
+
+        if sort_categories and not ignore_order and ordered:
+            raise TypeError("Cannot use sort_categories=True with ordered Categoricals")
+
+        if sort_categories and not categories.is_monotonic_increasing:
+            categories = categories.sort_values()
+            indexer = categories.get_indexer(first.categories)
+
+            from pandas.core.algorithms import take_nd
+
+            new_codes = take_nd(indexer, new_codes, fill_value=-1)
+    elif ignore_order or all(not c.ordered for c in to_union):
+        # different categories - union and recode
+        cats = first.categories.append([c.categories for c in to_union[1:]])
+        categories = cats.unique()
+        if sort_categories:
+            categories = categories.sort_values()
+
+        new_codes = [
+            recode_for_categories(c.codes, c.categories, categories) for c in to_union
+        ]
+        new_codes = np.concatenate(new_codes)
+    else:
+        # ordered - to show a proper error message
+        if all(c.ordered for c in to_union):
+            msg = "to union ordered Categoricals, all categories must be the same"
+            raise TypeError(msg)
+        raise TypeError("Categorical.ordered must be the same")
+
+    if ignore_order:
+        ordered = False
+
+    dtype = CategoricalDtype(categories=categories, ordered=ordered)
+    return Categorical._simple_new(new_codes, dtype=dtype)
diff --git a/pandas/pandas/core/dtypes/dtypes.py b/pandas/pandas/core/dtypes/dtypes.py
new file mode 100644
index 0000000000000..a02a8b8b110bf
--- /dev/null
+++ b/pandas/pandas/core/dtypes/dtypes.py
@@ -0,0 +1,2451 @@
+"""
+Define extension dtypes.
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+from decimal import Decimal
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+import zoneinfo
+
+import numpy as np
+
+from pandas._config.config import get_option
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas._libs.interval import Interval
+from pandas._libs.properties import cache_readonly
+from pandas._libs.tslibs import (
+    BaseOffset,
+    NaT,
+    NaTType,
+    Period,
+    Timedelta,
+    Timestamp,
+    timezones,
+    to_offset,
+    tz_compare,
+)
+from pandas._libs.tslibs.dtypes import (
+    PeriodDtypeBase,
+    abbrev_to_npy_unit,
+)
+from pandas._libs.tslibs.offsets import BDay
+from pandas.compat import pa_version_under10p1
+from pandas.errors import PerformanceWarning
+from pandas.util._decorators import set_module
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import (
+    ExtensionDtype,
+    StorageExtensionDtype,
+    register_extension_dtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCCategoricalIndex,
+    ABCIndex,
+    ABCRangeIndex,
+)
+from pandas.core.dtypes.inference import (
+    is_bool,
+    is_list_like,
+)
+
+if not pa_version_under10p1:
+    import pyarrow as pa
+
+if TYPE_CHECKING:
+    from collections.abc import MutableMapping
+    from datetime import tzinfo
+
+    import pyarrow as pa  # noqa: TC004
+
+    from pandas._typing import (
+        Dtype,
+        DtypeObj,
+        IntervalClosedType,
+        Ordered,
+        Scalar,
+        Self,
+        npt,
+        type_t,
+    )
+
+    from pandas import (
+        Categorical,
+        CategoricalIndex,
+        DatetimeIndex,
+        Index,
+        IntervalIndex,
+        PeriodIndex,
+    )
+    from pandas.core.arrays import (
+        BaseMaskedArray,
+        DatetimeArray,
+        IntervalArray,
+        NumpyExtensionArray,
+        PeriodArray,
+        SparseArray,
+    )
+    from pandas.core.arrays.arrow import ArrowExtensionArray
+
+str_type = str
+
+
+class PandasExtensionDtype(ExtensionDtype):
+    """
+    A np.dtype duck-typed class, suitable for holding a custom dtype.
+
+    THIS IS NOT A REAL NUMPY DTYPE
+    """
+
+    type: Any
+    kind: Any
+    # The Any type annotations above are here only because mypy seems to have a
+    # problem dealing with multiple inheritance from PandasExtensionDtype
+    # and ExtensionDtype's @properties in the subclasses below. The kind and
+    # type variables in those subclasses are explicitly typed below.
+    subdtype = None
+    str: str_type
+    num = 100
+    shape: tuple[int, ...] = ()
+    itemsize = 8
+    base: DtypeObj | None = None
+    isbuiltin = 0
+    isnative = 0
+    _cache_dtypes: dict[str_type, PandasExtensionDtype] = {}
+
+    def __repr__(self) -> str_type:
+        """
+        Return a string representation for a particular object.
+        """
+        return str(self)
+
+    def __hash__(self) -> int:
+        raise NotImplementedError("sub-classes should implement an __hash__ method")
+
+    def __getstate__(self) -> dict[str_type, Any]:
+        # pickle support; we don't want to pickle the cache
+        return {k: getattr(self, k, None) for k in self._metadata}
+
+    @classmethod
+    def reset_cache(cls) -> None:
+        """clear the cache"""
+        cls._cache_dtypes = {}
+
+
+class CategoricalDtypeType(type):
+    """
+    the type of CategoricalDtype, this metaclass determines subclass ability
+    """
+
+
+@register_extension_dtype
+@set_module("pandas")
+class CategoricalDtype(PandasExtensionDtype, ExtensionDtype):
+    """
+    Type for categorical data with the categories and orderedness.
+
+    It is a dtype representation for categorical data, which allows users to define
+    a fixed set of values and optionally impose an ordering. This is particularly
+    useful for handling categorical variables efficiently, as it can significantly
+    reduce memory usage compared to using object dtypes.
+
+    Parameters
+    ----------
+    categories : sequence, optional
+        Must be unique, and must not contain any nulls.
+        The categories are stored in an Index,
+        and if an index is provided the dtype of that index will be used.
+    ordered : bool or None, default False
+        Whether or not this categorical is treated as a ordered categorical.
+        None can be used to maintain the ordered value of existing categoricals when
+        used in operations that combine categoricals, e.g. astype, and will resolve to
+        False if there is no existing ordered to maintain.
+
+    Attributes
+    ----------
+    categories
+    ordered
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    Categorical : Represent a categorical variable in classic R / S-plus fashion.
+
+    Notes
+    -----
+    This class is useful for specifying the type of a ``Categorical``
+    independent of the values. See :ref:`categorical.categoricaldtype`
+    for more.
+
+    Examples
+    --------
+    >>> t = pd.CategoricalDtype(categories=["b", "a"], ordered=True)
+    >>> pd.Series(["a", "b", "a", "c"], dtype=t)
+    0      a
+    1      b
+    2      a
+    3    NaN
+    dtype: category
+    Categories (2, object): ['b' < 'a']
+
+    An empty CategoricalDtype with a specific dtype can be created
+    by providing an empty index. As follows,
+
+    >>> pd.CategoricalDtype(pd.DatetimeIndex([])).categories.dtype
+    dtype('<M8[s]')
+    """
+
+    # TODO: Document public vs. private API
+    name = "category"
+    type: type[CategoricalDtypeType] = CategoricalDtypeType
+    kind: str_type = "O"
+    str = "|O08"
+    base = np.dtype("O")
+    _metadata = ("categories", "ordered")
+    _cache_dtypes: dict[str_type, PandasExtensionDtype] = {}
+    _supports_2d = False
+    _can_fast_transpose = False
+
+    def __init__(self, categories=None, ordered: Ordered = False) -> None:
+        self._finalize(categories, ordered, fastpath=False)
+
+    @classmethod
+    def _from_fastpath(
+        cls, categories=None, ordered: bool | None = None
+    ) -> CategoricalDtype:
+        self = cls.__new__(cls)
+        self._finalize(categories, ordered, fastpath=True)
+        return self
+
+    @classmethod
+    def _from_categorical_dtype(
+        cls, dtype: CategoricalDtype, categories=None, ordered: Ordered | None = None
+    ) -> CategoricalDtype:
+        if categories is ordered is None:
+            return dtype
+        if categories is None:
+            categories = dtype.categories
+        if ordered is None:
+            ordered = dtype.ordered
+        return cls(categories, ordered)
+
+    @classmethod
+    def _from_values_or_dtype(
+        cls,
+        values=None,
+        categories=None,
+        ordered: bool | None = None,
+        dtype: Dtype | None = None,
+    ) -> CategoricalDtype:
+        """
+        Construct dtype from the input parameters used in :class:`Categorical`.
+
+        This constructor method specifically does not do the factorization
+        step, if that is needed to find the categories. This constructor may
+        therefore return ``CategoricalDtype(categories=None, ordered=None)``,
+        which may not be useful. Additional steps may therefore have to be
+        taken to create the final dtype.
+
+        The return dtype is specified from the inputs in this prioritized
+        order:
+        1. if dtype is a CategoricalDtype, return dtype
+        2. if dtype is the string 'category', create a CategoricalDtype from
+           the supplied categories and ordered parameters, and return that.
+        3. if values is a categorical, use value.dtype, but override it with
+           categories and ordered if either/both of those are not None.
+        4. if dtype is None and values is not a categorical, construct the
+           dtype from categories and ordered, even if either of those is None.
+
+        Parameters
+        ----------
+        values : list-like, optional
+            The list-like must be 1-dimensional.
+        categories : list-like, optional
+            Categories for the CategoricalDtype.
+        ordered : bool, optional
+            Designating if the categories are ordered.
+        dtype : CategoricalDtype or the string "category", optional
+            If ``CategoricalDtype``, cannot be used together with
+            `categories` or `ordered`.
+
+        Returns
+        -------
+        CategoricalDtype
+
+        Examples
+        --------
+        >>> pd.CategoricalDtype._from_values_or_dtype()
+        CategoricalDtype(categories=None, ordered=None, categories_dtype=None)
+        >>> pd.CategoricalDtype._from_values_or_dtype(
+        ...     categories=["a", "b"], ordered=True
+        ... )
+        CategoricalDtype(categories=['a', 'b'], ordered=True, categories_dtype=object)
+        >>> dtype1 = pd.CategoricalDtype(["a", "b"], ordered=True)
+        >>> dtype2 = pd.CategoricalDtype(["x", "y"], ordered=False)
+        >>> c = pd.Categorical([0, 1], dtype=dtype1)
+        >>> pd.CategoricalDtype._from_values_or_dtype(
+        ...     c, ["x", "y"], ordered=True, dtype=dtype2
+        ... )
+        Traceback (most recent call last):
+            ...
+        ValueError: Cannot specify `categories` or `ordered` together with
+        `dtype`.
+
+        The supplied dtype takes precedence over values' dtype:
+
+        >>> pd.CategoricalDtype._from_values_or_dtype(c, dtype=dtype2)
+        CategoricalDtype(categories=['x', 'y'], ordered=False, categories_dtype=object)
+        """
+
+        if dtype is not None:
+            # The dtype argument takes precedence over values.dtype (if any)
+            if isinstance(dtype, str):
+                if dtype == "category":
+                    if ordered is None and cls.is_dtype(values):
+                        # GH#49309 preserve orderedness
+                        ordered = values.dtype.ordered
+
+                    dtype = CategoricalDtype(categories, ordered)
+                else:
+                    raise ValueError(f"Unknown dtype {dtype!r}")
+            elif categories is not None or ordered is not None:
+                raise ValueError(
+                    "Cannot specify `categories` or `ordered` together with `dtype`."
+                )
+            elif not isinstance(dtype, CategoricalDtype):
+                raise ValueError(f"Cannot not construct CategoricalDtype from {dtype}")
+        elif cls.is_dtype(values):
+            # If no "dtype" was passed, use the one from "values", but honor
+            # the "ordered" and "categories" arguments
+            dtype = values.dtype._from_categorical_dtype(
+                values.dtype, categories, ordered
+            )
+        else:
+            # If dtype=None and values is not categorical, create a new dtype.
+            # Note: This could potentially have categories=None and
+            # ordered=None.
+            dtype = CategoricalDtype(categories, ordered)
+
+        return cast(CategoricalDtype, dtype)
+
+    @classmethod
+    def construct_from_string(cls, string: str_type) -> CategoricalDtype:
+        """
+        Construct a CategoricalDtype from a string.
+
+        Parameters
+        ----------
+        string : str
+            Must be the string "category" in order to be successfully constructed.
+
+        Returns
+        -------
+        CategoricalDtype
+            Instance of the dtype.
+
+        Raises
+        ------
+        TypeError
+            If a CategoricalDtype cannot be constructed from the input.
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+        if string != cls.name:
+            raise TypeError(f"Cannot construct a 'CategoricalDtype' from '{string}'")
+
+        # need ordered=None to ensure that operations specifying dtype="category" don't
+        # override the ordered value for existing categoricals
+        return cls(ordered=None)
+
+    def _finalize(self, categories, ordered: Ordered, fastpath: bool = False) -> None:
+        if ordered is not None:
+            self.validate_ordered(ordered)
+
+        if categories is not None:
+            categories = self.validate_categories(categories, fastpath=fastpath)
+
+        self._categories = categories
+        self._ordered = ordered
+
+    def __setstate__(self, state: MutableMapping[str_type, Any]) -> None:
+        # for pickle compat. __get_state__ is defined in the
+        # PandasExtensionDtype superclass and uses the public properties to
+        # pickle -> need to set the settable private ones here (see GH26067)
+        self._categories = state.pop("categories", None)
+        self._ordered = state.pop("ordered", False)
+
+    def __hash__(self) -> int:
+        # _hash_categories returns a uint64, so use the negative
+        # space for when we have unknown categories to avoid a conflict
+        if self.categories is None:
+            if self.ordered:
+                return -1
+            else:
+                return -2
+        # We *do* want to include the real self.ordered here
+        return int(self._hash_categories)
+
+    def __eq__(self, other: object) -> bool:
+        """
+        Rules for CDT equality:
+        1) Any CDT is equal to the string 'category'
+        2) Any CDT is equal to itself
+        3) Any CDT is equal to a CDT with categories=None regardless of ordered
+        4) A CDT with ordered=True is only equal to another CDT with
+           ordered=True and identical categories in the same order
+        5) A CDT with ordered={False, None} is only equal to another CDT with
+           ordered={False, None} and identical categories, but same order is
+           not required. There is no distinction between False/None.
+        6) Any other comparison returns False
+        """
+        if isinstance(other, str):
+            return other == self.name
+        elif other is self:
+            return True
+        elif not (hasattr(other, "ordered") and hasattr(other, "categories")):
+            return False
+        elif self.categories is None or other.categories is None:
+            # For non-fully-initialized dtypes, these are only equal to
+            #  - the string "category" (handled above)
+            #  - other CategoricalDtype with categories=None
+            return self.categories is other.categories
+        elif self.ordered or other.ordered:
+            # At least one has ordered=True; equal if both have ordered=True
+            # and the same values for categories in the same order.
+            return (self.ordered == other.ordered) and self.categories.equals(
+                other.categories
+            )
+        else:
+            # Neither has ordered=True; equal if both have the same categories,
+            # but same order is not necessary.  There is no distinction between
+            # ordered=False and ordered=None: CDT(., False) and CDT(., None)
+            # will be equal if they have the same categories.
+            left = self.categories
+            right = other.categories
+
+            # GH#36280 the ordering of checks here is for performance
+            if not left.dtype == right.dtype:
+                return False
+
+            if len(left) != len(right):
+                return False
+
+            if self.categories.equals(other.categories):
+                # Check and see if they happen to be identical categories
+                return True
+
+            if left.dtype != object:
+                # Faster than calculating hash
+                indexer = left.get_indexer(right)
+                # Because left and right have the same length and are unique,
+                #  `indexer` not having any -1s implies that there is a
+                #  bijection between `left` and `right`.
+                return bool((indexer != -1).all())
+
+            # With object-dtype we need a comparison that identifies
+            #  e.g. int(2) as distinct from float(2)
+            return set(left) == set(right)
+
+    def __repr__(self) -> str_type:
+        if self.categories is None:
+            data = "None"
+            dtype = "None"
+        else:
+            data = self.categories._format_data(name=type(self).__name__)
+            if isinstance(self.categories, ABCRangeIndex):
+                data = str(self.categories._range)
+            data = data.rstrip(", ")
+            dtype = self.categories.dtype
+
+        return (
+            f"CategoricalDtype(categories={data}, ordered={self.ordered}, "
+            f"categories_dtype={dtype})"
+        )
+
+    @cache_readonly
+    def _hash_categories(self) -> int:
+        from pandas.core.util.hashing import (
+            combine_hash_arrays,
+            hash_array,
+            hash_tuples,
+        )
+
+        categories = self.categories
+        ordered = self.ordered
+
+        if len(categories) and isinstance(categories[0], tuple):
+            # assumes if any individual category is a tuple, then all our. ATM
+            # I don't really want to support just some of the categories being
+            # tuples.
+            cat_list = list(categories)  # breaks if a np.array of categories
+            cat_array = hash_tuples(cat_list)
+        else:
+            if categories.dtype == "O" and len({type(x) for x in categories}) != 1:
+                # TODO: hash_array doesn't handle mixed types. It casts
+                # everything to a str first, which means we treat
+                # {'1', '2'} the same as {'1', 2}
+                # find a better solution
+                hashed = hash((tuple(categories), ordered))
+                return hashed
+
+            if DatetimeTZDtype.is_dtype(categories.dtype):
+                # Avoid future warning.
+                categories = categories.view("datetime64[ns]")
+
+            cat_array = hash_array(np.asarray(categories), categorize=False)
+        if ordered:
+            cat_array = np.vstack(
+                [cat_array, np.arange(len(cat_array), dtype=cat_array.dtype)]
+            )
+        else:
+            cat_array = cat_array.reshape(1, len(cat_array))
+        combined_hashed = combine_hash_arrays(iter(cat_array), num_items=len(cat_array))
+        return np.bitwise_xor.reduce(combined_hashed)
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[Categorical]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas import Categorical
+
+        return Categorical
+
+    @staticmethod
+    def validate_ordered(ordered: Ordered) -> None:
+        """
+        Validates that we have a valid ordered parameter. If
+        it is not a boolean, a TypeError will be raised.
+
+        Parameters
+        ----------
+        ordered : object
+            The parameter to be verified.
+
+        Raises
+        ------
+        TypeError
+            If 'ordered' is not a boolean.
+        """
+        if not is_bool(ordered):
+            raise TypeError("'ordered' must either be 'True' or 'False'")
+
+    @staticmethod
+    def validate_categories(categories, fastpath: bool = False) -> Index:
+        """
+        Validates that we have good categories
+
+        Parameters
+        ----------
+        categories : array-like
+        fastpath : bool
+            Whether to skip nan and uniqueness checks
+
+        Returns
+        -------
+        categories : Index
+        """
+        from pandas.core.indexes.base import Index
+
+        if not fastpath and not is_list_like(categories):
+            raise TypeError(
+                f"Parameter 'categories' must be list-like, was {categories!r}"
+            )
+        if not isinstance(categories, ABCIndex):
+            categories = Index._with_infer(categories, tupleize_cols=False)
+
+        if not fastpath:
+            if categories.hasnans:
+                raise ValueError("Categorical categories cannot be null")
+
+            if not categories.is_unique:
+                raise ValueError("Categorical categories must be unique")
+
+        if isinstance(categories, ABCCategoricalIndex):
+            categories = categories.categories
+
+        return categories
+
+    def update_dtype(self, dtype: str_type | CategoricalDtype) -> CategoricalDtype:
+        """
+        Returns a CategoricalDtype with categories and ordered taken from dtype
+        if specified, otherwise falling back to self if unspecified
+
+        Parameters
+        ----------
+        dtype : CategoricalDtype
+
+        Returns
+        -------
+        new_dtype : CategoricalDtype
+        """
+        if isinstance(dtype, str) and dtype == "category":
+            # dtype='category' should not change anything
+            return self
+        elif not self.is_dtype(dtype):
+            raise ValueError(
+                f"a CategoricalDtype must be passed to perform an update, got {dtype!r}"
+            )
+        else:
+            # from here on, dtype is a CategoricalDtype
+            dtype = cast(CategoricalDtype, dtype)
+
+        # update categories/ordered unless they've been explicitly passed as None
+        if (
+            isinstance(dtype, CategoricalDtype)
+            and dtype.categories is not None
+            and dtype.ordered is not None
+        ):
+            # Avoid re-validation in CategoricalDtype constructor
+            return dtype
+        new_categories = (
+            dtype.categories if dtype.categories is not None else self.categories
+        )
+        new_ordered = dtype.ordered if dtype.ordered is not None else self.ordered
+
+        return CategoricalDtype(new_categories, new_ordered)
+
+    @property
+    def categories(self) -> Index:
+        """
+        An ``Index`` containing the unique categories allowed.
+
+        See Also
+        --------
+        ordered : Whether the categories have an ordered relationship.
+
+        Examples
+        --------
+        >>> cat_type = pd.CategoricalDtype(categories=["a", "b"], ordered=True)
+        >>> cat_type.categories
+        Index(['a', 'b'], dtype='object')
+        """
+        return self._categories
+
+    @property
+    def ordered(self) -> Ordered:
+        """
+        Whether the categories have an ordered relationship.
+
+        See Also
+        --------
+        categories : An Index containing the unique categories allowed.
+
+        Examples
+        --------
+        >>> cat_type = pd.CategoricalDtype(categories=["a", "b"], ordered=True)
+        >>> cat_type.ordered
+        True
+
+        >>> cat_type = pd.CategoricalDtype(categories=["a", "b"], ordered=False)
+        >>> cat_type.ordered
+        False
+        """
+        return self._ordered
+
+    @property
+    def _is_boolean(self) -> bool:
+        from pandas.core.dtypes.common import is_bool_dtype
+
+        return is_bool_dtype(self.categories)
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        # check if we have all categorical dtype with identical categories
+        if all(isinstance(x, CategoricalDtype) for x in dtypes):
+            first = dtypes[0]
+            if all(first == other for other in dtypes[1:]):
+                return first
+
+        # special case non-initialized categorical
+        # TODO we should figure out the expected return value in general
+        non_init_cats = [
+            isinstance(x, CategoricalDtype) and x.categories is None for x in dtypes
+        ]
+        if all(non_init_cats):
+            return self
+        elif any(non_init_cats):
+            return None
+
+        # categorical is aware of Sparse -> extract sparse subdtypes
+        subtypes = (x.subtype if isinstance(x, SparseDtype) else x for x in dtypes)
+        # extract the categories' dtype
+        non_cat_dtypes = [
+            x.categories.dtype if isinstance(x, CategoricalDtype) else x
+            for x in subtypes
+        ]
+        # TODO should categorical always give an answer?
+        from pandas.core.dtypes.cast import find_common_type
+
+        return find_common_type(non_cat_dtypes)
+
+    @cache_readonly
+    def index_class(self) -> type_t[CategoricalIndex]:
+        from pandas import CategoricalIndex
+
+        return CategoricalIndex
+
+
+@register_extension_dtype
+@set_module("pandas")
+class DatetimeTZDtype(PandasExtensionDtype):
+    """
+    An ExtensionDtype for timezone-aware datetime data.
+
+    **This is not an actual numpy dtype**, but a duck type.
+
+    Parameters
+    ----------
+    unit : str, default "ns"
+        The precision of the datetime data. Valid options are
+        ``"s"``, ``"ms"``, ``"us"``, ``"ns"``.
+    tz : str, int, or datetime.tzinfo
+        The timezone.
+
+    Attributes
+    ----------
+    unit
+    tz
+
+    Methods
+    -------
+    None
+
+    Raises
+    ------
+    ZoneInfoNotFoundError
+        When the requested timezone cannot be found.
+
+    See Also
+    --------
+    numpy.datetime64 : Numpy data type for datetime.
+    datetime.datetime : Python datetime object.
+
+    Examples
+    --------
+    >>> from zoneinfo import ZoneInfo
+    >>> pd.DatetimeTZDtype(tz=ZoneInfo("UTC"))
+    datetime64[ns, UTC]
+
+    >>> pd.DatetimeTZDtype(tz=ZoneInfo("Europe/Paris"))
+    datetime64[ns, Europe/Paris]
+    """
+
+    type: type[Timestamp] = Timestamp
+    kind: str_type = "M"
+    num = 101
+    _metadata = ("unit", "tz")
+    _match = re.compile(r"(datetime64|M8)\[(?P<unit>.+), (?P<tz>.+)\]")
+    _cache_dtypes: dict[str_type, PandasExtensionDtype] = {}
+    _supports_2d = True
+    _can_fast_transpose = True
+
+    @property
+    def na_value(self) -> NaTType:
+        return NaT
+
+    @cache_readonly
+    def base(self) -> DtypeObj:  # type: ignore[override]
+        return np.dtype(f"M8[{self.unit}]")
+
+    # error: Signature of "str" incompatible with supertype "PandasExtensionDtype"
+    @cache_readonly
+    def str(self) -> str:  # type: ignore[override]
+        return f"|M8[{self.unit}]"
+
+    def __init__(self, unit: str_type | DatetimeTZDtype = "ns", tz=None) -> None:
+        if isinstance(unit, DatetimeTZDtype):
+            # error: "str" has no attribute "tz"
+            unit, tz = unit.unit, unit.tz  # type: ignore[attr-defined]
+
+        if unit != "ns":
+            if isinstance(unit, str) and tz is None:
+                # maybe a string like datetime64[ns, tz], which we support for
+                # now.
+                result = type(self).construct_from_string(unit)
+                unit = result.unit
+                tz = result.tz
+                msg = (
+                    f"Passing a dtype alias like 'datetime64[ns, {tz}]' "
+                    "to DatetimeTZDtype is no longer supported. Use "
+                    "'DatetimeTZDtype.construct_from_string()' instead."
+                )
+                raise ValueError(msg)
+            if unit not in ["s", "ms", "us", "ns"]:
+                raise ValueError("DatetimeTZDtype only supports s, ms, us, ns units")
+
+        if tz:
+            tz = timezones.maybe_get_tz(tz)
+            tz = timezones.tz_standardize(tz)
+        elif tz is not None:
+            raise zoneinfo.ZoneInfoNotFoundError(tz)
+        if tz is None:
+            raise TypeError("A 'tz' is required.")
+
+        self._unit = unit
+        self._tz = tz
+
+    @cache_readonly
+    def _creso(self) -> int:
+        """
+        The NPY_DATETIMEUNIT corresponding to this dtype's resolution.
+        """
+        return abbrev_to_npy_unit(self.unit)
+
+    @property
+    def unit(self) -> str_type:
+        """
+        The precision of the datetime data.
+
+        See Also
+        --------
+        DatetimeTZDtype.tz : Retrieves the timezone.
+
+        Examples
+        --------
+        >>> from zoneinfo import ZoneInfo
+        >>> dtype = pd.DatetimeTZDtype(tz=ZoneInfo("America/Los_Angeles"))
+        >>> dtype.unit
+        'ns'
+        """
+        return self._unit
+
+    @property
+    def tz(self) -> tzinfo:
+        """
+        The timezone.
+
+        See Also
+        --------
+        DatetimeTZDtype.unit : Retrieves precision of the datetime data.
+
+        Examples
+        --------
+        >>> from zoneinfo import ZoneInfo
+        >>> dtype = pd.DatetimeTZDtype(tz=ZoneInfo("America/Los_Angeles"))
+        >>> dtype.tz
+        zoneinfo.ZoneInfo(key='America/Los_Angeles')
+        """
+        return self._tz
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[DatetimeArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays import DatetimeArray
+
+        return DatetimeArray
+
+    @classmethod
+    def construct_from_string(cls, string: str_type) -> DatetimeTZDtype:
+        """
+        Construct a DatetimeTZDtype from a string.
+
+        Parameters
+        ----------
+        string : str
+            The string alias for this DatetimeTZDtype.
+            Should be formatted like ``datetime64[ns, <tz>]``,
+            where ``<tz>`` is the timezone name.
+
+        Examples
+        --------
+        >>> DatetimeTZDtype.construct_from_string("datetime64[ns, UTC]")
+        datetime64[ns, UTC]
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+
+        msg = f"Cannot construct a 'DatetimeTZDtype' from '{string}'"
+        match = cls._match.match(string)
+        if match:
+            d = match.groupdict()
+            try:
+                return cls(unit=d["unit"], tz=d["tz"])
+            except (KeyError, TypeError, ValueError) as err:
+                # KeyError if maybe_get_tz tries and fails to get a
+                #  zoneinfo timezone (actually zoneinfo.ZoneInfoNotFoundError).
+                # TypeError if we pass a nonsense tz;
+                # ValueError if we pass a unit other than "ns"
+                raise TypeError(msg) from err
+        raise TypeError(msg)
+
+    def __str__(self) -> str_type:
+        return f"datetime64[{self.unit}, {self.tz}]"
+
+    @property
+    def name(self) -> str_type:
+        """A string representation of the dtype."""
+        return str(self)
+
+    def __hash__(self) -> int:
+        # make myself hashable
+        # TODO: update this.
+        return hash(str(self))
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, str):
+            if other.startswith("M8["):
+                other = f"datetime64[{other[3:]}"
+            return other == self.name
+
+        return (
+            isinstance(other, DatetimeTZDtype)
+            and self.unit == other.unit
+            and tz_compare(self.tz, other.tz)
+        )
+
+    def __from_arrow__(self, array: pa.Array | pa.ChunkedArray) -> DatetimeArray:
+        """
+        Construct DatetimeArray from pyarrow Array/ChunkedArray.
+
+        Note: If the units in the pyarrow Array are the same as this
+        DatetimeDtype, then values corresponding to the integer representation
+        of ``NaT`` (e.g. one nanosecond before :attr:`pandas.Timestamp.min`)
+        are converted to ``NaT``, regardless of the null indicator in the
+        pyarrow array.
+
+        Parameters
+        ----------
+        array : pyarrow.Array or pyarrow.ChunkedArray
+            The Arrow array to convert to DatetimeArray.
+
+        Returns
+        -------
+        extension array : DatetimeArray
+        """
+        import pyarrow
+
+        from pandas.core.arrays import DatetimeArray
+
+        array = array.cast(pyarrow.timestamp(unit=self._unit), safe=True)
+
+        if isinstance(array, pyarrow.Array):
+            np_arr = array.to_numpy(zero_copy_only=False)
+        else:
+            np_arr = array.to_numpy()
+
+        return DatetimeArray._simple_new(np_arr, dtype=self)
+
+    def __setstate__(self, state) -> None:
+        # for pickle compat. __get_state__ is defined in the
+        # PandasExtensionDtype superclass and uses the public properties to
+        # pickle -> need to set the settable private ones here (see GH26067)
+        self._tz = state["tz"]
+        self._unit = state["unit"]
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        if all(isinstance(t, DatetimeTZDtype) and t.tz == self.tz for t in dtypes):
+            np_dtype = np.max([cast(DatetimeTZDtype, t).base for t in [self, *dtypes]])
+            unit = np.datetime_data(np_dtype)[0]
+            return type(self)(unit=unit, tz=self.tz)
+        return super()._get_common_dtype(dtypes)
+
+    @cache_readonly
+    def index_class(self) -> type_t[DatetimeIndex]:
+        from pandas import DatetimeIndex
+
+        return DatetimeIndex
+
+
+@register_extension_dtype
+@set_module("pandas")
+class PeriodDtype(PeriodDtypeBase, PandasExtensionDtype):
+    """
+    An ExtensionDtype for Period data.
+
+    **This is not an actual numpy dtype**, but a duck type.
+
+    Parameters
+    ----------
+    freq : str or DateOffset
+        The frequency of this PeriodDtype.
+
+    Attributes
+    ----------
+    freq
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    Period : Represents a single time period.
+    PeriodIndex : Immutable index for period data.
+    date_range : Return a fixed frequency DatetimeIndex.
+    Series : One-dimensional array with axis labels.
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+
+    Examples
+    --------
+    >>> pd.PeriodDtype(freq="D")
+    period[D]
+
+    >>> pd.PeriodDtype(freq=pd.offsets.MonthEnd())
+    period[M]
+    """
+
+    type: type[Period] = Period
+    kind: str_type = "O"
+    str = "|O08"
+    base = np.dtype("O")
+    num = 102
+    _metadata = ("freq",)
+    _match = re.compile(r"(P|p)eriod\[(?P<freq>.+)\]")
+    # error: Incompatible types in assignment (expression has type
+    # "Dict[int, PandasExtensionDtype]", base class "PandasExtensionDtype"
+    # defined the type as "Dict[str, PandasExtensionDtype]")  [assignment]
+    _cache_dtypes: dict[BaseOffset, int] = {}  # type: ignore[assignment]
+    __hash__ = PeriodDtypeBase.__hash__
+    _freq: BaseOffset
+    _supports_2d = True
+    _can_fast_transpose = True
+
+    def __new__(cls, freq) -> PeriodDtype:  # noqa: PYI034
+        """
+        Parameters
+        ----------
+        freq : PeriodDtype, BaseOffset, or string
+        """
+        if isinstance(freq, PeriodDtype):
+            return freq
+
+        if not isinstance(freq, BaseOffset):
+            freq = cls._parse_dtype_strict(freq)
+
+        if isinstance(freq, BDay):
+            # GH#53446
+            # TODO(3.0): enforcing this will close GH#10575
+            warnings.warn(
+                "PeriodDtype[B] is deprecated and will be removed in a future "
+                "version. Use a DatetimeIndex with freq='B' instead",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        try:
+            dtype_code = cls._cache_dtypes[freq]
+        except KeyError:
+            dtype_code = freq._period_dtype_code
+            cls._cache_dtypes[freq] = dtype_code
+        u = PeriodDtypeBase.__new__(cls, dtype_code, freq.n)
+        u._freq = freq
+        return u
+
+    def __reduce__(self) -> tuple[type_t[Self], tuple[str_type]]:
+        return type(self), (self.name,)
+
+    @property
+    def freq(self) -> BaseOffset:
+        """
+        The frequency object of this PeriodDtype.
+
+        The `freq` property returns the `BaseOffset` object that represents the
+        frequency of the PeriodDtype. This frequency specifies the interval (e.g.,
+        daily, monthly, yearly) associated with the Period type. It is essential
+        for operations that depend on time-based calculations within a period index
+        or series.
+
+        See Also
+        --------
+        Period : Represents a period of time.
+        PeriodIndex : Immutable ndarray holding ordinal values indicating
+            regular periods.
+        PeriodDtype : An ExtensionDtype for Period data.
+        date_range : Return a fixed frequency range of dates.
+
+        Examples
+        --------
+        >>> dtype = pd.PeriodDtype(freq="D")
+        >>> dtype.freq
+        <Day>
+        """
+        return self._freq
+
+    @classmethod
+    def _parse_dtype_strict(cls, freq: str_type) -> BaseOffset:
+        if isinstance(freq, str):  # note: freq is already of type str!
+            if freq.startswith(("Period[", "period[")):
+                m = cls._match.search(freq)
+                if m is not None:
+                    freq = m.group("freq")
+
+            freq_offset = to_offset(freq, is_period=True)
+            if freq_offset is not None:
+                return freq_offset
+
+        raise TypeError(
+            "PeriodDtype argument should be string or BaseOffset, "
+            f"got {type(freq).__name__}"
+        )
+
+    @classmethod
+    def construct_from_string(cls, string: str_type) -> PeriodDtype:
+        """
+        Strict construction from a string, raise a TypeError if not
+        possible
+        """
+        if (
+            isinstance(string, str) and (string.startswith(("period[", "Period[")))
+        ) or isinstance(string, BaseOffset):
+            # do not parse string like U as period[U]
+            # avoid tuple to be regarded as freq
+            try:
+                return cls(freq=string)
+            except ValueError:
+                pass
+        if isinstance(string, str):
+            msg = f"Cannot construct a 'PeriodDtype' from '{string}'"
+        else:
+            msg = f"'construct_from_string' expects a string, got {type(string)}"
+        raise TypeError(msg)
+
+    def __str__(self) -> str_type:
+        return self.name
+
+    @property
+    def name(self) -> str_type:
+        return f"period[{self._freqstr}]"
+
+    @property
+    def na_value(self) -> NaTType:
+        return NaT
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, str):
+            return other[:1].lower() + other[1:] == self.name
+
+        return super().__eq__(other)
+
+    def __ne__(self, other: object) -> bool:
+        return not self.__eq__(other)
+
+    @classmethod
+    def is_dtype(cls, dtype: object) -> bool:
+        """
+        Return a boolean if the passed type is an actual dtype that we
+        can match (via string or type)
+        """
+        if isinstance(dtype, str):
+            # PeriodDtype can be instantiated from freq string like "U",
+            # but doesn't regard freq str like "U" as dtype.
+            if dtype.startswith(("period[", "Period[")):
+                try:
+                    return cls._parse_dtype_strict(dtype) is not None
+                except ValueError:
+                    return False
+            else:
+                return False
+        return super().is_dtype(dtype)
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[PeriodArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays import PeriodArray
+
+        return PeriodArray
+
+    def __from_arrow__(self, array: pa.Array | pa.ChunkedArray) -> PeriodArray:
+        """
+        Construct PeriodArray from pyarrow Array/ChunkedArray.
+        """
+        import pyarrow
+
+        from pandas.core.arrays import PeriodArray
+        from pandas.core.arrays.arrow._arrow_utils import (
+            pyarrow_array_to_numpy_and_mask,
+        )
+
+        if isinstance(array, pyarrow.Array):
+            chunks = [array]
+        else:
+            chunks = array.chunks
+
+        results = []
+        for arr in chunks:
+            data, mask = pyarrow_array_to_numpy_and_mask(arr, dtype=np.dtype(np.int64))
+            parr = PeriodArray(data.copy(), dtype=self, copy=False)
+            # error: Invalid index type "ndarray[Any, dtype[bool_]]" for "PeriodArray";
+            # expected type "Union[int, Sequence[int], Sequence[bool], slice]"
+            parr[~mask] = NaT  # type: ignore[index]
+            results.append(parr)
+
+        if not results:
+            return PeriodArray(np.array([], dtype="int64"), dtype=self, copy=False)
+        return PeriodArray._concat_same_type(results)
+
+    @cache_readonly
+    def index_class(self) -> type_t[PeriodIndex]:
+        from pandas import PeriodIndex
+
+        return PeriodIndex
+
+
+@register_extension_dtype
+@set_module("pandas")
+class IntervalDtype(PandasExtensionDtype):
+    """
+    An ExtensionDtype for Interval data.
+
+    **This is not an actual numpy dtype**, but a duck type.
+
+    Parameters
+    ----------
+    subtype : str, np.dtype
+        The dtype of the Interval bounds.
+    closed : {'right', 'left', 'both', 'neither'}, default 'right'
+        Whether the interval is closed on the left-side, right-side, both or
+        neither. See the Notes for more detailed explanation.
+
+    Attributes
+    ----------
+    subtype
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    PeriodDtype : An ExtensionDtype for Period data.
+
+    Examples
+    --------
+    >>> pd.IntervalDtype(subtype="int64", closed="both")
+    interval[int64, both]
+    """
+
+    name = "interval"
+    kind: str_type = "O"
+    str = "|O08"
+    base = np.dtype("O")
+    num = 103
+    _metadata = (
+        "subtype",
+        "closed",
+    )
+
+    _match = re.compile(
+        r"(I|i)nterval\[(?P<subtype>[^,]+(\[.+\])?)"
+        r"(, (?P<closed>(right|left|both|neither)))?\]"
+    )
+
+    _cache_dtypes: dict[str_type, PandasExtensionDtype] = {}
+    _subtype: None | np.dtype
+    _closed: IntervalClosedType | None
+
+    def __init__(self, subtype=None, closed: IntervalClosedType | None = None) -> None:
+        from pandas.core.dtypes.common import (
+            is_string_dtype,
+            pandas_dtype,
+        )
+
+        if closed is not None and closed not in {"right", "left", "both", "neither"}:
+            raise ValueError("closed must be one of 'right', 'left', 'both', 'neither'")
+
+        if isinstance(subtype, IntervalDtype):
+            if closed is not None and closed != subtype.closed:
+                raise ValueError(
+                    "dtype.closed and 'closed' do not match. "
+                    "Try IntervalDtype(dtype.subtype, closed) instead."
+                )
+            self._subtype = subtype._subtype
+            self._closed = subtype._closed
+        elif subtype is None:
+            # we are called as an empty constructor
+            # generally for pickle compat
+            self._subtype = None
+            self._closed = closed
+        elif isinstance(subtype, str) and subtype.lower() == "interval":
+            self._subtype = None
+            self._closed = closed
+        else:
+            if isinstance(subtype, str):
+                m = IntervalDtype._match.search(subtype)
+                if m is not None:
+                    gd = m.groupdict()
+                    subtype = gd["subtype"]
+                    if gd.get("closed", None) is not None:
+                        if closed is not None:
+                            if closed != gd["closed"]:
+                                raise ValueError(
+                                    "'closed' keyword does not match value "
+                                    "specified in dtype string"
+                                )
+                        closed = gd["closed"]  # type: ignore[assignment]
+
+            try:
+                subtype = pandas_dtype(subtype)
+            except TypeError as err:
+                raise TypeError("could not construct IntervalDtype") from err
+            if CategoricalDtype.is_dtype(subtype) or is_string_dtype(subtype):
+                # GH 19016
+                msg = (
+                    "category, object, and string subtypes are not supported "
+                    "for IntervalDtype"
+                )
+                raise TypeError(msg)
+            self._subtype = subtype
+            self._closed = closed
+
+    @cache_readonly
+    def _can_hold_na(self) -> bool:
+        subtype = self._subtype
+        if subtype is None:
+            # partially-initialized
+            raise NotImplementedError(
+                "_can_hold_na is not defined for partially-initialized IntervalDtype"
+            )
+        if subtype.kind in "iu":
+            return False
+        return True
+
+    @property
+    def closed(self) -> IntervalClosedType:
+        return self._closed  # type: ignore[return-value]
+
+    @property
+    def subtype(self):
+        """
+        The dtype of the Interval bounds.
+
+        See Also
+        --------
+        IntervalDtype: An ExtensionDtype for Interval data.
+
+        Examples
+        --------
+        >>> dtype = pd.IntervalDtype(subtype="int64", closed="both")
+        >>> dtype.subtype
+        dtype('int64')
+        """
+        return self._subtype
+
+    @classmethod
+    def construct_array_type(cls) -> type[IntervalArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays import IntervalArray
+
+        return IntervalArray
+
+    @classmethod
+    def construct_from_string(cls, string: str_type) -> IntervalDtype:
+        """
+        attempt to construct this type from a string, raise a TypeError
+        if its not possible
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+
+        if string.lower() == "interval" or cls._match.search(string) is not None:
+            return cls(string)
+
+        msg = (
+            f"Cannot construct a 'IntervalDtype' from '{string}'.\n\n"
+            "Incorrectly formatted string passed to constructor. "
+            "Valid formats include Interval or Interval[dtype] "
+            "where dtype is numeric, datetime, or timedelta"
+        )
+        raise TypeError(msg)
+
+    @property
+    def type(self) -> type[Interval]:
+        return Interval
+
+    def __str__(self) -> str_type:
+        if self.subtype is None:
+            return "interval"
+        if self.closed is None:
+            # Only partially initialized GH#38394
+            return f"interval[{self.subtype}]"
+        return f"interval[{self.subtype}, {self.closed}]"
+
+    def __hash__(self) -> int:
+        # make myself hashable
+        return hash(str(self))
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, str):
+            return other.lower() in (self.name.lower(), str(self).lower())
+        elif not isinstance(other, IntervalDtype):
+            return False
+        elif self.subtype is None or other.subtype is None:
+            # None should match any subtype
+            return True
+        elif self.closed != other.closed:
+            return False
+        else:
+            return self.subtype == other.subtype
+
+    def __setstate__(self, state) -> None:
+        # for pickle compat. __get_state__ is defined in the
+        # PandasExtensionDtype superclass and uses the public properties to
+        # pickle -> need to set the settable private ones here (see GH26067)
+        self._subtype = state["subtype"]
+
+        # backward-compat older pickles won't have "closed" key
+        self._closed = state.pop("closed", None)
+
+    @classmethod
+    def is_dtype(cls, dtype: object) -> bool:
+        """
+        Return a boolean if the passed type is an actual dtype that we
+        can match (via string or type)
+        """
+        if isinstance(dtype, str):
+            if dtype.lower().startswith("interval"):
+                try:
+                    return cls.construct_from_string(dtype) is not None
+                except (ValueError, TypeError):
+                    return False
+            else:
+                return False
+        return super().is_dtype(dtype)
+
+    def __from_arrow__(self, array: pa.Array | pa.ChunkedArray) -> IntervalArray:
+        """
+        Construct IntervalArray from pyarrow Array/ChunkedArray.
+        """
+        import pyarrow
+
+        from pandas.core.arrays import IntervalArray
+
+        if isinstance(array, pyarrow.Array):
+            chunks = [array]
+        else:
+            chunks = array.chunks
+
+        results = []
+        for arr in chunks:
+            if isinstance(arr, pyarrow.ExtensionArray):
+                arr = arr.storage
+            left = np.asarray(arr.field("left"), dtype=self.subtype)
+            right = np.asarray(arr.field("right"), dtype=self.subtype)
+            iarr = IntervalArray.from_arrays(left, right, closed=self.closed)
+            results.append(iarr)
+
+        if not results:
+            return IntervalArray.from_arrays(
+                np.array([], dtype=self.subtype),
+                np.array([], dtype=self.subtype),
+                closed=self.closed,
+            )
+        return IntervalArray._concat_same_type(results)
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        if not all(isinstance(x, IntervalDtype) for x in dtypes):
+            return None
+
+        closed = cast("IntervalDtype", dtypes[0]).closed
+        if not all(cast("IntervalDtype", x).closed == closed for x in dtypes):
+            return np.dtype(object)
+
+        from pandas.core.dtypes.cast import find_common_type
+
+        common = find_common_type([cast("IntervalDtype", x).subtype for x in dtypes])
+        if common == object:
+            return np.dtype(object)
+        return IntervalDtype(common, closed=closed)
+
+    @cache_readonly
+    def index_class(self) -> type_t[IntervalIndex]:
+        from pandas import IntervalIndex
+
+        return IntervalIndex
+
+
+class NumpyEADtype(ExtensionDtype):
+    """
+    A Pandas ExtensionDtype for NumPy dtypes.
+
+    This is mostly for internal compatibility, and is not especially
+    useful on its own.
+
+    Parameters
+    ----------
+    dtype : object
+        Object to be converted to a NumPy data type object.
+
+    See Also
+    --------
+    numpy.dtype
+    """
+
+    _metadata = ("_dtype",)
+    _supports_2d = False
+    _can_fast_transpose = False
+
+    def __init__(self, dtype: npt.DTypeLike | NumpyEADtype | None) -> None:
+        if isinstance(dtype, NumpyEADtype):
+            # make constructor idempotent
+            dtype = dtype.numpy_dtype
+        self._dtype = np.dtype(dtype)
+
+    def __repr__(self) -> str:
+        return f"NumpyEADtype({self.name!r})"
+
+    @property
+    def numpy_dtype(self) -> np.dtype:
+        """
+        The NumPy dtype this NumpyEADtype wraps.
+        """
+        return self._dtype
+
+    @property
+    def name(self) -> str:
+        """
+        A bit-width name for this data-type.
+        """
+        return self._dtype.name
+
+    @property
+    def type(self) -> type[np.generic]:
+        """
+        The type object used to instantiate a scalar of this NumPy data-type.
+        """
+        return self._dtype.type
+
+    @property
+    def _is_numeric(self) -> bool:
+        # exclude object, str, unicode, void.
+        return self.kind in set("biufc")
+
+    @property
+    def _is_boolean(self) -> bool:
+        return self.kind == "b"
+
+    @classmethod
+    def construct_from_string(cls, string: str) -> NumpyEADtype:
+        try:
+            dtype = np.dtype(string)
+        except TypeError as err:
+            if not isinstance(string, str):
+                msg = f"'construct_from_string' expects a string, got {type(string)}"
+            else:
+                msg = f"Cannot construct a 'NumpyEADtype' from '{string}'"
+            raise TypeError(msg) from err
+        return cls(dtype)
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[NumpyExtensionArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays import NumpyExtensionArray
+
+        return NumpyExtensionArray
+
+    @property
+    def kind(self) -> str:
+        """
+        A character code (one of 'biufcmMOSUV') identifying the general kind of data.
+        """
+        return self._dtype.kind
+
+    @property
+    def itemsize(self) -> int:
+        """
+        The element size of this data-type object.
+        """
+        return self._dtype.itemsize
+
+
+class BaseMaskedDtype(ExtensionDtype):
+    """
+    Base class for dtypes for BaseMaskedArray subclasses.
+    """
+
+    base = None
+    type: type
+    _internal_fill_value: Scalar
+
+    @property
+    def _truthy_value(self):
+        # Fill values used for 'any'
+        if self.kind == "f":
+            return 1.0
+        if self.kind in "iu":
+            return 1
+        return True
+
+    @property
+    def _falsey_value(self):
+        # Fill values used for 'all'
+        if self.kind == "f":
+            return 0.0
+        if self.kind in "iu":
+            return 0
+        return False
+
+    @property
+    def na_value(self) -> libmissing.NAType:
+        return libmissing.NA
+
+    @cache_readonly
+    def numpy_dtype(self) -> np.dtype:
+        """Return an instance of our numpy dtype"""
+        return np.dtype(self.type)
+
+    @cache_readonly
+    def kind(self) -> str:
+        return self.numpy_dtype.kind
+
+    @cache_readonly
+    def itemsize(self) -> int:
+        """Return the number of bytes in this dtype"""
+        return self.numpy_dtype.itemsize
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[BaseMaskedArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        raise NotImplementedError
+
+    @classmethod
+    def from_numpy_dtype(cls, dtype: np.dtype) -> BaseMaskedDtype:
+        """
+        Construct the MaskedDtype corresponding to the given numpy dtype.
+        """
+        if dtype.kind == "b":
+            from pandas.core.arrays.boolean import BooleanDtype
+
+            return BooleanDtype()
+        elif dtype.kind in "iu":
+            from pandas.core.arrays.integer import NUMPY_INT_TO_DTYPE
+
+            return NUMPY_INT_TO_DTYPE[dtype]
+        elif dtype.kind == "f":
+            from pandas.core.arrays.floating import NUMPY_FLOAT_TO_DTYPE
+
+            return NUMPY_FLOAT_TO_DTYPE[dtype]
+        else:
+            raise NotImplementedError(dtype)
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        # We unwrap any masked dtypes, find the common dtype we would use
+        #  for that, then re-mask the result.
+        from pandas.core.dtypes.cast import find_common_type
+
+        new_dtype = find_common_type(
+            [
+                dtype.numpy_dtype if isinstance(dtype, BaseMaskedDtype) else dtype
+                for dtype in dtypes
+            ]
+        )
+        if not isinstance(new_dtype, np.dtype):
+            # If we ever support e.g. Masked[DatetimeArray] then this will change
+            return None
+        try:
+            return type(self).from_numpy_dtype(new_dtype)
+        except (KeyError, NotImplementedError):
+            return None
+
+
+@register_extension_dtype
+@set_module("pandas")
+class SparseDtype(ExtensionDtype):
+    """
+    Dtype for data stored in :class:`SparseArray`.
+
+    ``SparseDtype`` is used as the data type for :class:`SparseArray`, enabling
+    more efficient storage of data that contains a significant number of
+    repetitive values typically represented by a fill value. It supports any
+    scalar dtype as the underlying data type of the non-fill values.
+
+    Parameters
+    ----------
+    dtype : str, ExtensionDtype, numpy.dtype, type, default numpy.float64
+        The dtype of the underlying array storing the non-fill value values.
+    fill_value : scalar, optional
+        The scalar value not stored in the SparseArray. By default, this
+        depends on ``dtype``.
+
+        =========== ==========
+        dtype       na_value
+        =========== ==========
+        float       ``np.nan``
+        complex     ``np.nan``
+        int         ``0``
+        bool        ``False``
+        datetime64  ``pd.NaT``
+        timedelta64 ``pd.NaT``
+        =========== ==========
+
+        The default value may be overridden by specifying a ``fill_value``.
+
+    Attributes
+    ----------
+    None
+
+    Methods
+    -------
+    None
+
+    See Also
+    --------
+    arrays.SparseArray : The array structure that uses SparseDtype
+        for data representation.
+
+    Examples
+    --------
+    >>> ser = pd.Series([1, 0, 0], dtype=pd.SparseDtype(dtype=int, fill_value=0))
+    >>> ser
+    0    1
+    1    0
+    2    0
+    dtype: Sparse[int64, 0]
+    >>> ser.sparse.density
+    0.3333333333333333
+    """
+
+    _is_immutable = True
+
+    # We include `_is_na_fill_value` in the metadata to avoid hash collisions
+    # between SparseDtype(float, 0.0) and SparseDtype(float, nan).
+    # Without is_na_fill_value in the comparison, those would be equal since
+    # hash(nan) is (sometimes?) 0.
+    _metadata = ("_dtype", "_fill_value", "_is_na_fill_value")
+
+    def __init__(self, dtype: Dtype = np.float64, fill_value: Any = None) -> None:
+        if isinstance(dtype, type(self)):
+            if fill_value is None:
+                fill_value = dtype.fill_value
+            dtype = dtype.subtype
+
+        from pandas.core.dtypes.common import (
+            is_string_dtype,
+            pandas_dtype,
+        )
+        from pandas.core.dtypes.missing import na_value_for_dtype
+
+        dtype = pandas_dtype(dtype)
+        if is_string_dtype(dtype):
+            dtype = np.dtype("object")
+        if not isinstance(dtype, np.dtype):
+            # GH#53160
+            raise TypeError("SparseDtype subtype must be a numpy dtype")
+
+        if fill_value is None:
+            fill_value = na_value_for_dtype(dtype)
+
+        self._dtype = dtype
+        self._fill_value = fill_value
+        self._check_fill_value()
+
+    def __hash__(self) -> int:
+        # Python3 doesn't inherit __hash__ when a base class overrides
+        # __eq__, so we explicitly do it here.
+        return super().__hash__()
+
+    def __eq__(self, other: object) -> bool:
+        # We have to override __eq__ to handle NA values in _metadata.
+        # The base class does simple == checks, which fail for NA.
+        if isinstance(other, str):
+            try:
+                other = self.construct_from_string(other)
+            except TypeError:
+                return False
+
+        if isinstance(other, type(self)):
+            subtype = self.subtype == other.subtype
+            if self._is_na_fill_value or other._is_na_fill_value:
+                # this case is complicated by two things:
+                # SparseDtype(float, float(nan)) == SparseDtype(float, np.nan)
+                # SparseDtype(float, np.nan)     != SparseDtype(float, pd.NaT)
+                # i.e. we want to treat any floating-point NaN as equal, but
+                # not a floating-point NaN and a datetime NaT.
+                fill_value = isinstance(
+                    self.fill_value, type(other.fill_value)
+                ) or isinstance(other.fill_value, type(self.fill_value))
+            else:
+                with warnings.catch_warnings():
+                    # Ignore spurious numpy warning
+                    warnings.filterwarnings(
+                        "ignore",
+                        "elementwise comparison failed",
+                        category=DeprecationWarning,
+                    )
+
+                    fill_value = self.fill_value == other.fill_value
+
+            return subtype and fill_value
+        return False
+
+    @property
+    def fill_value(self):
+        """
+        The fill value of the array.
+
+        Converting the SparseArray to a dense ndarray will fill the
+        array with this value.
+
+        .. warning::
+
+           It's possible to end up with a SparseArray that has ``fill_value``
+           values in ``sp_values``. This can occur, for example, when setting
+           ``SparseArray.fill_value`` directly.
+        """
+        return self._fill_value
+
+    def _check_fill_value(self) -> None:
+        if not lib.is_scalar(self._fill_value):
+            raise ValueError(
+                f"fill_value must be a scalar. Got {self._fill_value} instead"
+            )
+
+        from pandas.core.dtypes.cast import can_hold_element
+        from pandas.core.dtypes.missing import (
+            is_valid_na_for_dtype,
+            isna,
+        )
+
+        from pandas.core.construction import ensure_wrapped_if_datetimelike
+
+        # GH#23124 require fill_value and subtype to match
+        val = self._fill_value
+        if isna(val):
+            if not is_valid_na_for_dtype(val, self.subtype):
+                raise ValueError(
+                    # GH#53043
+                    "fill_value must be a valid value for the SparseDtype.subtype"
+                )
+        else:
+            dummy = np.empty(0, dtype=self.subtype)
+            dummy = ensure_wrapped_if_datetimelike(dummy)
+
+            if not can_hold_element(dummy, val):
+                raise ValueError(
+                    # GH#53043
+                    "fill_value must be a valid value for the SparseDtype.subtype"
+                )
+
+    @property
+    def _is_na_fill_value(self) -> bool:
+        from pandas import isna
+
+        return isna(self.fill_value)
+
+    @property
+    def _is_numeric(self) -> bool:
+        return not self.subtype == object
+
+    @property
+    def _is_boolean(self) -> bool:
+        return self.subtype.kind == "b"
+
+    @property
+    def kind(self) -> str:
+        """
+        The sparse kind. Either 'integer', or 'block'.
+        """
+        return self.subtype.kind
+
+    @property
+    def type(self):
+        return self.subtype.type
+
+    @property
+    def subtype(self):
+        return self._dtype
+
+    @property
+    def name(self) -> str:
+        return f"Sparse[{self.subtype.name}, {self.fill_value!r}]"
+
+    def __repr__(self) -> str:
+        return self.name
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[SparseArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays.sparse.array import SparseArray
+
+        return SparseArray
+
+    @classmethod
+    def construct_from_string(cls, string: str) -> SparseDtype:
+        """
+        Construct a SparseDtype from a string form.
+
+        Parameters
+        ----------
+        string : str
+            Can take the following forms.
+
+            string           dtype
+            ================ ============================
+            'int'            SparseDtype[np.int64, 0]
+            'Sparse'         SparseDtype[np.float64, nan]
+            'Sparse[int]'    SparseDtype[np.int64, 0]
+            'Sparse[int, 0]' SparseDtype[np.int64, 0]
+            ================ ============================
+
+            It is not possible to specify non-default fill values
+            with a string. An argument like ``'Sparse[int, 1]'``
+            will raise a ``TypeError`` because the default fill value
+            for integers is 0.
+
+        Returns
+        -------
+        SparseDtype
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+        msg = f"Cannot construct a 'SparseDtype' from '{string}'"
+        if string.startswith("Sparse"):
+            try:
+                sub_type, has_fill_value = cls._parse_subtype(string)
+            except ValueError as err:
+                raise TypeError(msg) from err
+            else:
+                result = SparseDtype(sub_type)
+                msg = (
+                    f"Cannot construct a 'SparseDtype' from '{string}'.\n\nIt "
+                    "looks like the fill_value in the string is not "
+                    "the default for the dtype. Non-default fill_values "
+                    "are not supported. Use the 'SparseDtype()' "
+                    "constructor instead."
+                )
+                if has_fill_value and str(result) != string:
+                    raise TypeError(msg)
+                return result
+        else:
+            raise TypeError(msg)
+
+    @staticmethod
+    def _parse_subtype(dtype: str) -> tuple[str, bool]:
+        """
+        Parse a string to get the subtype
+
+        Parameters
+        ----------
+        dtype : str
+            A string like
+
+            * Sparse[subtype]
+            * Sparse[subtype, fill_value]
+
+        Returns
+        -------
+        subtype : str
+
+        Raises
+        ------
+        ValueError
+            When the subtype cannot be extracted.
+        """
+        xpr = re.compile(r"Sparse\[(?P<subtype>[^,]*)(, )?(?P<fill_value>.*?)?\]$")
+        m = xpr.match(dtype)
+        has_fill_value = False
+        if m:
+            subtype = m.groupdict()["subtype"]
+            has_fill_value = bool(m.groupdict()["fill_value"])
+        elif dtype == "Sparse":
+            subtype = "float64"
+        else:
+            raise ValueError(f"Cannot parse {dtype}")
+        return subtype, has_fill_value
+
+    @classmethod
+    def is_dtype(cls, dtype: object) -> bool:
+        dtype = getattr(dtype, "dtype", dtype)
+        if isinstance(dtype, str) and dtype.startswith("Sparse"):
+            sub_type, _ = cls._parse_subtype(dtype)
+            dtype = np.dtype(sub_type)
+        elif isinstance(dtype, cls):
+            return True
+        return isinstance(dtype, np.dtype) or dtype == "Sparse"
+
+    def update_dtype(self, dtype) -> SparseDtype:
+        """
+        Convert the SparseDtype to a new dtype.
+
+        This takes care of converting the ``fill_value``.
+
+        Parameters
+        ----------
+        dtype : Union[str, numpy.dtype, SparseDtype]
+            The new dtype to use.
+
+            * For a SparseDtype, it is simply returned
+            * For a NumPy dtype (or str), the current fill value
+              is converted to the new dtype, and a SparseDtype
+              with `dtype` and the new fill value is returned.
+
+        Returns
+        -------
+        SparseDtype
+            A new SparseDtype with the correct `dtype` and fill value
+            for that `dtype`.
+
+        Raises
+        ------
+        ValueError
+            When the current fill value cannot be converted to the
+            new `dtype` (e.g. trying to convert ``np.nan`` to an
+            integer dtype).
+
+
+        Examples
+        --------
+        >>> SparseDtype(int, 0).update_dtype(float)
+        Sparse[float64, 0.0]
+
+        >>> SparseDtype(int, 1).update_dtype(SparseDtype(float, np.nan))
+        Sparse[float64, nan]
+        """
+        from pandas.core.dtypes.astype import astype_array
+        from pandas.core.dtypes.common import pandas_dtype
+
+        cls = type(self)
+        dtype = pandas_dtype(dtype)
+
+        if not isinstance(dtype, cls):
+            if not isinstance(dtype, np.dtype):
+                raise TypeError("sparse arrays of extension dtypes not supported")
+
+            fv_asarray = np.atleast_1d(np.array(self.fill_value))
+            fvarr = astype_array(fv_asarray, dtype)
+            # NB: not fv_0d.item(), as that casts dt64->int
+            fill_value = fvarr[0]
+            dtype = cls(dtype, fill_value=fill_value)
+
+        return dtype
+
+    @property
+    def _subtype_with_str(self):
+        """
+        Whether the SparseDtype's subtype should be considered ``str``.
+
+        Typically, pandas will store string data in an object-dtype array.
+        When converting values to a dtype, e.g. in ``.astype``, we need to
+        be more specific, we need the actual underlying type.
+
+        Returns
+        -------
+        >>> SparseDtype(int, 1)._subtype_with_str
+        dtype('int64')
+
+        >>> SparseDtype(object, 1)._subtype_with_str
+        dtype('O')
+
+        >>> dtype = SparseDtype(str, "")
+        >>> dtype.subtype
+        dtype('O')
+
+        >>> dtype._subtype_with_str
+        <class 'str'>
+        """
+        if isinstance(self.fill_value, str):
+            return type(self.fill_value)
+        return self.subtype
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        # TODO for now only handle SparseDtypes and numpy dtypes => extend
+        # with other compatible extension dtypes
+        from pandas.core.dtypes.cast import np_find_common_type
+
+        if any(
+            isinstance(x, ExtensionDtype) and not isinstance(x, SparseDtype)
+            for x in dtypes
+        ):
+            return None
+
+        fill_values = [x.fill_value for x in dtypes if isinstance(x, SparseDtype)]
+        fill_value = fill_values[0]
+
+        from pandas import isna
+
+        # np.nan isn't a singleton, so we may end up with multiple
+        # NaNs here, so we ignore the all NA case too.
+        if get_option("performance_warnings") and (
+            not (len(set(fill_values)) == 1 or isna(fill_values).all())
+        ):
+            warnings.warn(
+                "Concatenating sparse arrays with multiple fill "
+                f"values: '{fill_values}'. Picking the first and "
+                "converting the rest.",
+                PerformanceWarning,
+                stacklevel=find_stack_level(),
+            )
+        np_dtypes = (x.subtype if isinstance(x, SparseDtype) else x for x in dtypes)
+        # error: Argument 1 to "np_find_common_type" has incompatible type
+        # "*Generator[Any | dtype[Any] | ExtensionDtype, None, None]";
+        # expected "dtype[Any]"  [arg-type]
+        return SparseDtype(np_find_common_type(*np_dtypes), fill_value=fill_value)  # type: ignore [arg-type]
+
+
+@register_extension_dtype
+@set_module("pandas")
+class ArrowDtype(StorageExtensionDtype):
+    """
+    An ExtensionDtype for PyArrow data types.
+
+    .. warning::
+
+       ArrowDtype is considered experimental. The implementation and
+       parts of the API may change without warning.
+
+    While most ``dtype`` arguments can accept the "string"
+    constructor, e.g. ``"int64[pyarrow]"``, ArrowDtype is useful
+    if the data type contains parameters like ``pyarrow.timestamp``.
+
+    Parameters
+    ----------
+    pyarrow_dtype : pa.DataType
+        An instance of a `pyarrow.DataType <https://arrow.apache.org/docs/python/api/datatypes.html#factory-functions>`__.
+
+    Attributes
+    ----------
+    pyarrow_dtype
+
+    Methods
+    -------
+    None
+
+    Returns
+    -------
+    ArrowDtype
+
+    See Also
+    --------
+    DataFrame.convert_dtypes : Convert columns to the best possible dtypes.
+
+    Examples
+    --------
+    >>> import pyarrow as pa
+    >>> pd.ArrowDtype(pa.int64())
+    int64[pyarrow]
+
+    Types with parameters must be constructed with ArrowDtype.
+
+    >>> pd.ArrowDtype(pa.timestamp("s", tz="America/New_York"))
+    timestamp[s, tz=America/New_York][pyarrow]
+    >>> pd.ArrowDtype(pa.list_(pa.int64()))
+    list<item: int64>[pyarrow]
+    """
+
+    _metadata = ("storage", "pyarrow_dtype")  # type: ignore[assignment]
+
+    def __init__(self, pyarrow_dtype: pa.DataType) -> None:
+        super().__init__("pyarrow")
+        if pa_version_under10p1:
+            raise ImportError("pyarrow>=10.0.1 is required for ArrowDtype")
+        if not isinstance(pyarrow_dtype, pa.DataType):
+            raise ValueError(
+                f"pyarrow_dtype ({pyarrow_dtype}) must be an instance "
+                f"of a pyarrow.DataType. Got {type(pyarrow_dtype)} instead."
+            )
+        self.pyarrow_dtype = pyarrow_dtype
+
+    def __repr__(self) -> str:
+        return self.name
+
+    def __hash__(self) -> int:
+        # make myself hashable
+        return hash(str(self))
+
+    def __eq__(self, other: object) -> bool:
+        if not isinstance(other, type(self)):
+            return super().__eq__(other)
+        return self.pyarrow_dtype == other.pyarrow_dtype
+
+    @property
+    def type(self):
+        """
+        Returns associated scalar type.
+        """
+        pa_type = self.pyarrow_dtype
+        if pa.types.is_integer(pa_type):
+            return int
+        elif pa.types.is_floating(pa_type):
+            return float
+        elif pa.types.is_string(pa_type) or pa.types.is_large_string(pa_type):
+            return str
+        elif (
+            pa.types.is_binary(pa_type)
+            or pa.types.is_fixed_size_binary(pa_type)
+            or pa.types.is_large_binary(pa_type)
+        ):
+            return bytes
+        elif pa.types.is_boolean(pa_type):
+            return bool
+        elif pa.types.is_duration(pa_type):
+            if pa_type.unit == "ns":
+                return Timedelta
+            else:
+                return timedelta
+        elif pa.types.is_timestamp(pa_type):
+            if pa_type.unit == "ns":
+                return Timestamp
+            else:
+                return datetime
+        elif pa.types.is_date(pa_type):
+            return date
+        elif pa.types.is_time(pa_type):
+            return time
+        elif pa.types.is_decimal(pa_type):
+            return Decimal
+        elif pa.types.is_dictionary(pa_type):
+            # TODO: Potentially change this & CategoricalDtype.type to
+            #  something more representative of the scalar
+            return CategoricalDtypeType
+        elif pa.types.is_list(pa_type) or pa.types.is_large_list(pa_type):
+            return list
+        elif pa.types.is_fixed_size_list(pa_type):
+            return list
+        elif pa.types.is_map(pa_type):
+            return list
+        elif pa.types.is_struct(pa_type):
+            return dict
+        elif pa.types.is_null(pa_type):
+            # TODO: None? pd.NA? pa.null?
+            return type(pa_type)
+        elif isinstance(pa_type, pa.ExtensionType):
+            return type(self)(pa_type.storage_type).type
+        raise NotImplementedError(pa_type)
+
+    @property
+    def name(self) -> str:  # type: ignore[override]
+        """
+        A string identifying the data type.
+        """
+        return f"{self.pyarrow_dtype!s}[{self.storage}]"
+
+    @cache_readonly
+    def numpy_dtype(self) -> np.dtype:
+        """Return an instance of the related numpy dtype"""
+        if pa.types.is_timestamp(self.pyarrow_dtype):
+            # pa.timestamp(unit).to_pandas_dtype() returns ns units
+            # regardless of the pyarrow timestamp units.
+            # This can be removed if/when pyarrow addresses it:
+            # https://github.com/apache/arrow/issues/34462
+            return np.dtype(f"datetime64[{self.pyarrow_dtype.unit}]")
+        if pa.types.is_duration(self.pyarrow_dtype):
+            # pa.duration(unit).to_pandas_dtype() returns ns units
+            # regardless of the pyarrow duration units
+            # This can be removed if/when pyarrow addresses it:
+            # https://github.com/apache/arrow/issues/34462
+            return np.dtype(f"timedelta64[{self.pyarrow_dtype.unit}]")
+        if pa.types.is_string(self.pyarrow_dtype) or pa.types.is_large_string(
+            self.pyarrow_dtype
+        ):
+            # pa.string().to_pandas_dtype() = object which we don't want
+            return np.dtype(str)
+        try:
+            return np.dtype(self.pyarrow_dtype.to_pandas_dtype())
+        except (NotImplementedError, TypeError):
+            return np.dtype(object)
+
+    @cache_readonly
+    def kind(self) -> str:
+        if pa.types.is_timestamp(self.pyarrow_dtype):
+            # To mirror DatetimeTZDtype
+            return "M"
+        return self.numpy_dtype.kind
+
+    @cache_readonly
+    def itemsize(self) -> int:
+        """Return the number of bytes in this dtype"""
+        return self.numpy_dtype.itemsize
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[ArrowExtensionArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        from pandas.core.arrays.arrow import ArrowExtensionArray
+
+        return ArrowExtensionArray
+
+    @classmethod
+    def construct_from_string(cls, string: str) -> ArrowDtype:
+        """
+        Construct this type from a string.
+
+        Parameters
+        ----------
+        string : str
+            string should follow the format f"{pyarrow_type}[pyarrow]"
+            e.g. int64[pyarrow]
+        """
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+        if not string.endswith("[pyarrow]"):
+            raise TypeError(f"'{string}' must end with '[pyarrow]'")
+        if string in ("string[pyarrow]", "str[pyarrow]"):
+            # Ensure Registry.find skips ArrowDtype to use StringDtype instead
+            raise TypeError("string[pyarrow] should be constructed by StringDtype")
+        if pa_version_under10p1:
+            raise ImportError("pyarrow>=10.0.1 is required for ArrowDtype")
+
+        base_type = string[:-9]  # get rid of "[pyarrow]"
+        try:
+            pa_dtype = pa.type_for_alias(base_type)
+        except ValueError as err:
+            has_parameters = re.search(r"[\[\(].*[\]\)]", base_type)
+            if has_parameters:
+                # Fallback to try common temporal types
+                try:
+                    return cls._parse_temporal_dtype_string(base_type)
+                except (NotImplementedError, ValueError):
+                    # Fall through to raise with nice exception message below
+                    pass
+
+                raise NotImplementedError(
+                    "Passing pyarrow type specific parameters "
+                    f"({has_parameters.group()}) in the string is not supported. "
+                    "Please construct an ArrowDtype object with a pyarrow_dtype "
+                    "instance with specific parameters."
+                ) from err
+            raise TypeError(f"'{base_type}' is not a valid pyarrow data type.") from err
+        return cls(pa_dtype)
+
+    # TODO(arrow#33642): This can be removed once supported by pyarrow
+    @classmethod
+    def _parse_temporal_dtype_string(cls, string: str) -> ArrowDtype:
+        """
+        Construct a temporal ArrowDtype from string.
+        """
+        # we assume
+        #  1) "[pyarrow]" has already been stripped from the end of our string.
+        #  2) we know "[" is present
+        head, tail = string.split("[", 1)
+
+        if not tail.endswith("]"):
+            raise ValueError
+        tail = tail[:-1]
+
+        if head == "timestamp":
+            assert "," in tail  # otherwise type_for_alias should work
+            unit, tz = tail.split(",", 1)
+            unit = unit.strip()
+            tz = tz.strip()
+            if tz.startswith("tz="):
+                tz = tz[3:]
+
+            pa_type = pa.timestamp(unit, tz=tz)
+            dtype = cls(pa_type)
+            return dtype
+
+        raise NotImplementedError(string)
+
+    @property
+    def _is_numeric(self) -> bool:
+        """
+        Whether columns with this dtype should be considered numeric.
+        """
+        # TODO: pa.types.is_boolean?
+        return (
+            pa.types.is_integer(self.pyarrow_dtype)
+            or pa.types.is_floating(self.pyarrow_dtype)
+            or pa.types.is_decimal(self.pyarrow_dtype)
+        )
+
+    @property
+    def _is_boolean(self) -> bool:
+        """
+        Whether this dtype should be considered boolean.
+        """
+        return pa.types.is_boolean(self.pyarrow_dtype)
+
+    def _get_common_dtype(self, dtypes: list[DtypeObj]) -> DtypeObj | None:
+        # We unwrap any masked dtypes, find the common dtype we would use
+        #  for that, then re-mask the result.
+        # Mirrors BaseMaskedDtype
+        from pandas.core.dtypes.cast import find_common_type
+
+        null_dtype = type(self)(pa.null())
+
+        new_dtype = find_common_type(
+            [
+                dtype.numpy_dtype if isinstance(dtype, ArrowDtype) else dtype
+                for dtype in dtypes
+                if dtype != null_dtype
+            ]
+        )
+        if not isinstance(new_dtype, np.dtype):
+            return None
+        try:
+            pa_dtype = pa.from_numpy_dtype(new_dtype)
+            return type(self)(pa_dtype)
+        except NotImplementedError:
+            return None
+
+    def __from_arrow__(self, array: pa.Array | pa.ChunkedArray) -> ArrowExtensionArray:
+        """
+        Construct IntegerArray/FloatingArray from pyarrow Array/ChunkedArray.
+        """
+        array_class = self.construct_array_type()
+        arr = array.cast(self.pyarrow_dtype, safe=True)
+        return array_class(arr)
diff --git a/pandas/pandas/core/dtypes/generic.py b/pandas/pandas/core/dtypes/generic.py
new file mode 100644
index 0000000000000..8d3d86217dedf
--- /dev/null
+++ b/pandas/pandas/core/dtypes/generic.py
@@ -0,0 +1,148 @@
+"""define generic base classes for pandas objects"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Type,
+    cast,
+)
+
+if TYPE_CHECKING:
+    from pandas import (
+        Categorical,
+        CategoricalIndex,
+        DataFrame,
+        DatetimeIndex,
+        Index,
+        IntervalIndex,
+        MultiIndex,
+        PeriodIndex,
+        RangeIndex,
+        Series,
+        TimedeltaIndex,
+    )
+    from pandas.core.arrays import (
+        DatetimeArray,
+        ExtensionArray,
+        NumpyExtensionArray,
+        PeriodArray,
+        TimedeltaArray,
+    )
+    from pandas.core.generic import NDFrame
+
+
+# define abstract base classes to enable isinstance type checking on our
+# objects
+def create_pandas_abc_type(name, attr, comp) -> type:
+    def _check(inst) -> bool:
+        return getattr(inst, attr, "_typ") in comp
+
+    # https://github.com/python/mypy/issues/1006
+    # error: 'classmethod' used with a non-method
+    @classmethod  # type: ignore[misc]
+    def _instancecheck(cls, inst) -> bool:
+        return _check(inst) and not isinstance(inst, type)
+
+    @classmethod  # type: ignore[misc]
+    def _subclasscheck(cls, inst) -> bool:
+        # Raise instead of returning False
+        # This is consistent with default __subclasscheck__ behavior
+        if not isinstance(inst, type):
+            raise TypeError("issubclass() arg 1 must be a class")
+
+        return _check(inst)
+
+    dct = {"__instancecheck__": _instancecheck, "__subclasscheck__": _subclasscheck}
+    meta = type("ABCBase", (type,), dct)
+    return meta(name, (), dct)
+
+
+ABCRangeIndex = cast(
+    "Type[RangeIndex]",
+    create_pandas_abc_type("ABCRangeIndex", "_typ", ("rangeindex",)),
+)
+ABCMultiIndex = cast(
+    "Type[MultiIndex]",
+    create_pandas_abc_type("ABCMultiIndex", "_typ", ("multiindex",)),
+)
+ABCDatetimeIndex = cast(
+    "Type[DatetimeIndex]",
+    create_pandas_abc_type("ABCDatetimeIndex", "_typ", ("datetimeindex",)),
+)
+ABCTimedeltaIndex = cast(
+    "Type[TimedeltaIndex]",
+    create_pandas_abc_type("ABCTimedeltaIndex", "_typ", ("timedeltaindex",)),
+)
+ABCPeriodIndex = cast(
+    "Type[PeriodIndex]",
+    create_pandas_abc_type("ABCPeriodIndex", "_typ", ("periodindex",)),
+)
+ABCCategoricalIndex = cast(
+    "Type[CategoricalIndex]",
+    create_pandas_abc_type("ABCCategoricalIndex", "_typ", ("categoricalindex",)),
+)
+ABCIntervalIndex = cast(
+    "Type[IntervalIndex]",
+    create_pandas_abc_type("ABCIntervalIndex", "_typ", ("intervalindex",)),
+)
+ABCIndex = cast(
+    "Type[Index]",
+    create_pandas_abc_type(
+        "ABCIndex",
+        "_typ",
+        {
+            "index",
+            "rangeindex",
+            "multiindex",
+            "datetimeindex",
+            "timedeltaindex",
+            "periodindex",
+            "categoricalindex",
+            "intervalindex",
+        },
+    ),
+)
+
+
+ABCNDFrame = cast(
+    "Type[NDFrame]",
+    create_pandas_abc_type("ABCNDFrame", "_typ", ("series", "dataframe")),
+)
+ABCSeries = cast(
+    "Type[Series]",
+    create_pandas_abc_type("ABCSeries", "_typ", ("series",)),
+)
+ABCDataFrame = cast(
+    "Type[DataFrame]", create_pandas_abc_type("ABCDataFrame", "_typ", ("dataframe",))
+)
+
+ABCCategorical = cast(
+    "Type[Categorical]",
+    create_pandas_abc_type("ABCCategorical", "_typ", ("categorical")),
+)
+ABCDatetimeArray = cast(
+    "Type[DatetimeArray]",
+    create_pandas_abc_type("ABCDatetimeArray", "_typ", ("datetimearray")),
+)
+ABCTimedeltaArray = cast(
+    "Type[TimedeltaArray]",
+    create_pandas_abc_type("ABCTimedeltaArray", "_typ", ("timedeltaarray")),
+)
+ABCPeriodArray = cast(
+    "Type[PeriodArray]",
+    create_pandas_abc_type("ABCPeriodArray", "_typ", ("periodarray",)),
+)
+ABCExtensionArray = cast(
+    "Type[ExtensionArray]",
+    create_pandas_abc_type(
+        "ABCExtensionArray",
+        "_typ",
+        # Note: IntervalArray and SparseArray are included bc they have _typ="extension"
+        {"extension", "categorical", "periodarray", "datetimearray", "timedeltaarray"},
+    ),
+)
+ABCNumpyExtensionArray = cast(
+    "Type[NumpyExtensionArray]",
+    create_pandas_abc_type("ABCNumpyExtensionArray", "_typ", ("npy_extension",)),
+)
diff --git a/pandas/pandas/core/dtypes/inference.py b/pandas/pandas/core/dtypes/inference.py
new file mode 100644
index 0000000000000..918d107f2ce6c
--- /dev/null
+++ b/pandas/pandas/core/dtypes/inference.py
@@ -0,0 +1,498 @@
+"""basic inference routines"""
+
+from __future__ import annotations
+
+from collections import abc
+from numbers import Number
+import re
+from re import Pattern
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import lib
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import TypeGuard
+
+is_bool = lib.is_bool
+
+is_integer = lib.is_integer
+
+is_float = lib.is_float
+
+is_complex = lib.is_complex
+
+is_scalar = lib.is_scalar
+
+is_decimal = lib.is_decimal
+
+is_list_like = lib.is_list_like
+
+is_iterator = lib.is_iterator
+
+
+def is_number(obj: object) -> TypeGuard[Number | np.number]:
+    """
+    Check if the object is a number.
+
+    Returns True when the object is a number, and False if is not.
+
+    Parameters
+    ----------
+    obj : any type
+        The object to check if is a number.
+
+    Returns
+    -------
+    bool
+        Whether `obj` is a number or not.
+
+    See Also
+    --------
+    api.types.is_integer: Checks a subgroup of numbers.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_number
+    >>> is_number(1)
+    True
+    >>> is_number(7.15)
+    True
+
+    Booleans are valid because they are int subclass.
+
+    >>> is_number(False)
+    True
+
+    >>> is_number("foo")
+    False
+    >>> is_number("5")
+    False
+    """
+    return isinstance(obj, (Number, np.number))
+
+
+def iterable_not_string(obj: object) -> bool:
+    """
+    Check if the object is an iterable but not a string.
+
+    Parameters
+    ----------
+    obj : The object to check.
+
+    Returns
+    -------
+    is_iter_not_string : bool
+        Whether `obj` is a non-string iterable.
+
+    Examples
+    --------
+    >>> iterable_not_string([1, 2, 3])
+    True
+    >>> iterable_not_string("foo")
+    False
+    >>> iterable_not_string(1)
+    False
+    """
+    return isinstance(obj, abc.Iterable) and not isinstance(obj, str)
+
+
+def is_file_like(obj: object) -> bool:
+    """
+    Check if the object is a file-like object.
+
+    For objects to be considered file-like, they must
+    be an iterator AND have either a `read` and/or `write`
+    method as an attribute.
+
+    Note: file-like objects must be iterable, but
+    iterable objects need not be file-like.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check for file-like properties.
+        This can be any Python object, and the function will
+        check if it has attributes typically associated with
+        file-like objects (e.g., `read`, `write`, `__iter__`).
+
+    Returns
+    -------
+    bool
+        Whether `obj` has file-like properties.
+
+    See Also
+    --------
+    api.types.is_dict_like : Check if the object is dict-like.
+    api.types.is_hashable : Return True if hash(obj) will succeed, False otherwise.
+    api.types.is_named_tuple : Check if the object is a named tuple.
+    api.types.is_iterator : Check if the object is an iterator.
+
+    Examples
+    --------
+    >>> import io
+    >>> from pandas.api.types import is_file_like
+    >>> buffer = io.StringIO("data")
+    >>> is_file_like(buffer)
+    True
+    >>> is_file_like([1, 2, 3])
+    False
+    """
+    if not (hasattr(obj, "read") or hasattr(obj, "write")):
+        return False
+
+    return bool(hasattr(obj, "__iter__"))
+
+
+def is_re(obj: object) -> TypeGuard[Pattern]:
+    """
+    Check if the object is a regex pattern instance.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check for being a regex pattern. Typically,
+        this would be an object that you expect to be a compiled
+        pattern from the `re` module.
+
+    Returns
+    -------
+    bool
+        Whether `obj` is a regex pattern.
+
+    See Also
+    --------
+    api.types.is_float : Return True if given object is float.
+    api.types.is_iterator : Check if the object is an iterator.
+    api.types.is_integer : Return True if given object is integer.
+    api.types.is_re_compilable : Check if the object can be compiled
+                                into a regex pattern instance.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_re
+    >>> import re
+    >>> is_re(re.compile(".*"))
+    True
+    >>> is_re("foo")
+    False
+    """
+    return isinstance(obj, Pattern)
+
+
+def is_re_compilable(obj: object) -> bool:
+    """
+    Check if the object can be compiled into a regex pattern instance.
+
+    Parameters
+    ----------
+    obj : The object to check
+        The object to check if the object can be compiled into a regex pattern instance.
+
+    Returns
+    -------
+    bool
+        Whether `obj` can be compiled as a regex pattern.
+
+    See Also
+    --------
+    api.types.is_re : Check if the object is a regex pattern instance.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_re_compilable
+    >>> is_re_compilable(".*")
+    True
+    >>> is_re_compilable(1)
+    False
+    """
+    try:
+        re.compile(obj)  # type: ignore[call-overload]
+    except TypeError:
+        return False
+    else:
+        return True
+
+
+def is_array_like(obj: object) -> bool:
+    """
+    Check if the object is array-like.
+
+    For an object to be considered array-like, it must be list-like and
+    have a `dtype` attribute.
+
+    Parameters
+    ----------
+    obj : The object to check
+
+    Returns
+    -------
+    is_array_like : bool
+        Whether `obj` has array-like properties.
+
+    Examples
+    --------
+    >>> is_array_like(np.array([1, 2, 3]))
+    True
+    >>> is_array_like(pd.Series(["a", "b"]))
+    True
+    >>> is_array_like(pd.Index(["2016-01-01"]))
+    True
+    >>> is_array_like([1, 2, 3])
+    False
+    >>> is_array_like(("a", "b"))
+    False
+    """
+    return is_list_like(obj) and hasattr(obj, "dtype")
+
+
+def is_nested_list_like(obj: object) -> bool:
+    """
+    Check if the object is list-like, and that all of its elements
+    are also list-like.
+
+    Parameters
+    ----------
+    obj : The object to check
+
+    Returns
+    -------
+    is_list_like : bool
+        Whether `obj` has list-like properties.
+
+    Examples
+    --------
+    >>> is_nested_list_like([[1, 2, 3]])
+    True
+    >>> is_nested_list_like([{1, 2, 3}, {1, 2, 3}])
+    True
+    >>> is_nested_list_like(["foo"])
+    False
+    >>> is_nested_list_like([])
+    False
+    >>> is_nested_list_like([[1, 2, 3], 1])
+    False
+
+    Notes
+    -----
+    This won't reliably detect whether a consumable iterator (e. g.
+    a generator) is a nested-list-like without consuming the iterator.
+    To avoid consuming it, we always return False if the outer container
+    doesn't define `__len__`.
+
+    See Also
+    --------
+    is_list_like
+    """
+    return (
+        is_list_like(obj)
+        and hasattr(obj, "__len__")
+        # need PEP 724 to handle these typing errors
+        and len(obj) > 0  # pyright: ignore[reportArgumentType]
+        and all(is_list_like(item) for item in obj)  # type: ignore[attr-defined]
+    )
+
+
+def is_dict_like(obj: object) -> bool:
+    """
+    Check if the object is dict-like.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check. This can be any Python object,
+        and the function will determine whether it
+        behaves like a dictionary.
+
+    Returns
+    -------
+    bool
+        Whether `obj` has dict-like properties.
+
+    See Also
+    --------
+    api.types.is_list_like : Check if the object is list-like.
+    api.types.is_file_like : Check if the object is a file-like.
+    api.types.is_named_tuple : Check if the object is a named tuple.
+
+    Examples
+    --------
+    >>> from pandas.api.types import is_dict_like
+    >>> is_dict_like({1: 2})
+    True
+    >>> is_dict_like([1, 2, 3])
+    False
+    >>> is_dict_like(dict)
+    False
+    >>> is_dict_like(dict())
+    True
+    """
+    dict_like_attrs = ("__getitem__", "keys", "__contains__")
+    return (
+        all(hasattr(obj, attr) for attr in dict_like_attrs)
+        # [GH 25196] exclude classes
+        and not isinstance(obj, type)
+    )
+
+
+def is_named_tuple(obj: object) -> bool:
+    """
+    Check if the object is a named tuple.
+
+    Parameters
+    ----------
+    obj : object
+        The object that will be checked to determine
+        whether it is a named tuple.
+
+    Returns
+    -------
+    bool
+        Whether `obj` is a named tuple.
+
+    See Also
+    --------
+    api.types.is_dict_like: Check if the object is dict-like.
+    api.types.is_hashable: Return True if hash(obj)
+                                  will succeed, False otherwise.
+    api.types.is_categorical_dtype : Check if the dtype is categorical.
+
+    Examples
+    --------
+    >>> from collections import namedtuple
+    >>> from pandas.api.types import is_named_tuple
+    >>> Point = namedtuple("Point", ["x", "y"])
+    >>> p = Point(1, 2)
+    >>>
+    >>> is_named_tuple(p)
+    True
+    >>> is_named_tuple((1, 2))
+    False
+    """
+    return isinstance(obj, abc.Sequence) and hasattr(obj, "_fields")
+
+
+def is_hashable(obj: object) -> TypeGuard[Hashable]:
+    """
+    Return True if hash(obj) will succeed, False otherwise.
+
+    Some types will pass a test against collections.abc.Hashable but fail when
+    they are actually hashed with hash().
+
+    Distinguish between these and other types by trying the call to hash() and
+    seeing if they raise TypeError.
+
+    Parameters
+    ----------
+    obj : object
+        The object to check for hashability. Any Python object can be passed here.
+
+    Returns
+    -------
+    bool
+        True if object can be hashed (i.e., does not raise TypeError when
+        passed to hash()), and False otherwise (e.g., if object is mutable
+        like a list or dictionary).
+
+    See Also
+    --------
+    api.types.is_float : Return True if given object is float.
+    api.types.is_iterator : Check if the object is an iterator.
+    api.types.is_list_like : Check if the object is list-like.
+    api.types.is_dict_like : Check if the object is dict-like.
+
+    Examples
+    --------
+    >>> import collections
+    >>> from pandas.api.types import is_hashable
+    >>> a = ([],)
+    >>> isinstance(a, collections.abc.Hashable)
+    True
+    >>> is_hashable(a)
+    False
+    """
+    # Unfortunately, we can't use isinstance(obj, collections.abc.Hashable),
+    # which can be faster than calling hash. That is because numpy scalars
+    # fail this test.
+
+    # Reconsider this decision once this numpy bug is fixed:
+    # https://github.com/numpy/numpy/issues/5562
+
+    try:
+        hash(obj)
+    except TypeError:
+        return False
+    else:
+        return True
+
+
+def is_sequence(obj: object) -> bool:
+    """
+    Check if the object is a sequence of objects.
+    String types are not included as sequences here.
+
+    Parameters
+    ----------
+    obj : The object to check
+
+    Returns
+    -------
+    is_sequence : bool
+        Whether `obj` is a sequence of objects.
+
+    Examples
+    --------
+    >>> l = [1, 2, 3]
+    >>>
+    >>> is_sequence(l)
+    True
+    >>> is_sequence(iter(l))
+    False
+    """
+    try:
+        # Can iterate over it.
+        iter(obj)  # type: ignore[call-overload]
+        # Has a length associated with it.
+        len(obj)  # type: ignore[arg-type]
+        return not isinstance(obj, (str, bytes))
+    except (TypeError, AttributeError):
+        return False
+
+
+def is_dataclass(item: object) -> bool:
+    """
+    Checks if the object is a data-class instance
+
+    Parameters
+    ----------
+    item : object
+
+    Returns
+    --------
+    is_dataclass : bool
+        True if the item is an instance of a data-class,
+        will return false if you pass the data class itself
+
+    Examples
+    --------
+    >>> from dataclasses import dataclass
+    >>> @dataclass
+    ... class Point:
+    ...     x: int
+    ...     y: int
+
+    >>> is_dataclass(Point)
+    False
+    >>> is_dataclass(Point(0, 2))
+    True
+
+    """
+    try:
+        import dataclasses
+
+        return dataclasses.is_dataclass(item) and not isinstance(item, type)
+    except ImportError:
+        return False
diff --git a/pandas/pandas/core/dtypes/missing.py b/pandas/pandas/core/dtypes/missing.py
new file mode 100644
index 0000000000000..f20ca44728664
--- /dev/null
+++ b/pandas/pandas/core/dtypes/missing.py
@@ -0,0 +1,740 @@
+"""
+missing types & inference
+"""
+
+from __future__ import annotations
+
+from decimal import Decimal
+from typing import (
+    TYPE_CHECKING,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+import pandas._libs.missing as libmissing
+from pandas._libs.tslibs import (
+    NaT,
+    iNaT,
+)
+from pandas.util._decorators import set_module
+
+from pandas.core.dtypes.common import (
+    DT64NS_DTYPE,
+    TD64NS_DTYPE,
+    ensure_object,
+    is_scalar,
+    is_string_or_object_np_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCExtensionArray,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import is_list_like
+
+if TYPE_CHECKING:
+    from re import Pattern
+
+    from pandas._libs.missing import NAType
+    from pandas._libs.tslibs import NaTType
+    from pandas._typing import (
+        ArrayLike,
+        DtypeObj,
+        NDFrame,
+        NDFrameT,
+        Scalar,
+        npt,
+    )
+
+    from pandas import Series
+    from pandas.core.indexes.base import Index
+
+
+isposinf_scalar = libmissing.isposinf_scalar
+isneginf_scalar = libmissing.isneginf_scalar
+
+_dtype_object = np.dtype("object")
+_dtype_str = np.dtype(str)
+
+
+@overload
+def isna(obj: Scalar | Pattern | NAType | NaTType) -> bool: ...
+
+
+@overload
+def isna(
+    obj: ArrayLike | Index | list,
+) -> npt.NDArray[np.bool_]: ...
+
+
+@overload
+def isna(obj: NDFrameT) -> NDFrameT: ...
+
+
+# handle unions
+@overload
+def isna(
+    obj: NDFrameT | ArrayLike | Index | list,
+) -> NDFrameT | npt.NDArray[np.bool_]: ...
+
+
+@overload
+def isna(obj: object) -> bool | npt.NDArray[np.bool_] | NDFrame: ...
+
+
+@set_module("pandas")
+def isna(obj: object) -> bool | npt.NDArray[np.bool_] | NDFrame:
+    """
+    Detect missing values for an array-like object.
+
+    This function takes a scalar or array-like object and indicates
+    whether values are missing (``NaN`` in numeric arrays, ``None`` or ``NaN``
+    in object arrays, ``NaT`` in datetimelike).
+
+    Parameters
+    ----------
+    obj : scalar or array-like
+        Object to check for null or missing values.
+
+    Returns
+    -------
+    bool or array-like of bool
+        For scalar input, returns a scalar boolean.
+        For array input, returns an array of boolean indicating whether each
+        corresponding element is missing.
+
+    See Also
+    --------
+    notna : Boolean inverse of pandas.isna.
+    Series.isna : Detect missing values in a Series.
+    DataFrame.isna : Detect missing values in a DataFrame.
+    Index.isna : Detect missing values in an Index.
+
+    Examples
+    --------
+    Scalar arguments (including strings) result in a scalar boolean.
+
+    >>> pd.isna("dog")
+    False
+
+    >>> pd.isna(pd.NA)
+    True
+
+    >>> pd.isna(np.nan)
+    True
+
+    ndarrays result in an ndarray of booleans.
+
+    >>> array = np.array([[1, np.nan, 3], [4, 5, np.nan]])
+    >>> array
+    array([[ 1., nan,  3.],
+           [ 4.,  5., nan]])
+    >>> pd.isna(array)
+    array([[False,  True, False],
+           [False, False,  True]])
+
+    For indexes, an ndarray of booleans is returned.
+
+    >>> index = pd.DatetimeIndex(["2017-07-05", "2017-07-06", None, "2017-07-08"])
+    >>> index
+    DatetimeIndex(['2017-07-05', '2017-07-06', 'NaT', '2017-07-08'],
+                  dtype='datetime64[s]', freq=None)
+    >>> pd.isna(index)
+    array([False, False,  True, False])
+
+    For Series and DataFrame, the same type is returned, containing booleans.
+
+    >>> df = pd.DataFrame([["ant", "bee", "cat"], ["dog", None, "fly"]])
+    >>> df
+         0     1    2
+    0  ant   bee  cat
+    1  dog  None  fly
+    >>> pd.isna(df)
+           0      1      2
+    0  False  False  False
+    1  False   True  False
+
+    >>> pd.isna(df[1])
+    0    False
+    1     True
+    Name: 1, dtype: bool
+    """
+    return _isna(obj)
+
+
+isnull = isna
+
+
+def _isna(obj):
+    """
+    Detect missing values, treating None, NaN or NA as null.
+
+    Parameters
+    ----------
+    obj: ndarray or object value
+        Input array or scalar value.
+
+    Returns
+    -------
+    boolean ndarray or boolean
+    """
+    if is_scalar(obj):
+        return libmissing.checknull(obj)
+    elif isinstance(obj, ABCMultiIndex):
+        raise NotImplementedError("isna is not defined for MultiIndex")
+    elif isinstance(obj, type):
+        return False
+    elif isinstance(obj, (np.ndarray, ABCExtensionArray)):
+        return _isna_array(obj)
+    elif isinstance(obj, ABCIndex):
+        # Try to use cached isna, which also short-circuits for integer dtypes
+        #  and avoids materializing RangeIndex._values
+        if not obj._can_hold_na:
+            return obj.isna()
+        return _isna_array(obj._values)
+
+    elif isinstance(obj, ABCSeries):
+        result = _isna_array(obj._values)
+        # box
+        result = obj._constructor(result, index=obj.index, name=obj.name, copy=False)
+        return result
+    elif isinstance(obj, ABCDataFrame):
+        return obj.isna()
+    elif isinstance(obj, list):
+        return _isna_array(np.asarray(obj, dtype=object))
+    elif hasattr(obj, "__array__"):
+        return _isna_array(np.asarray(obj))
+    else:
+        return False
+
+
+def _isna_array(values: ArrayLike) -> npt.NDArray[np.bool_] | NDFrame:
+    """
+    Return an array indicating which values of the input array are NaN / NA.
+
+    Parameters
+    ----------
+    obj: ndarray or ExtensionArray
+        The input array whose elements are to be checked.
+
+    Returns
+    -------
+    array-like
+        Array of boolean values denoting the NA status of each element.
+    """
+    dtype = values.dtype
+    result: npt.NDArray[np.bool_] | NDFrame
+
+    if not isinstance(values, np.ndarray):
+        # i.e. ExtensionArray
+        # error: Incompatible types in assignment (expression has type
+        # "Union[ndarray[Any, Any], ExtensionArraySupportsAnyAll]", variable has
+        # type "ndarray[Any, dtype[bool_]]")
+        result = values.isna()  # type: ignore[assignment]
+    elif isinstance(values, np.rec.recarray):
+        # GH 48526
+        result = _isna_recarray_dtype(values)
+    elif is_string_or_object_np_dtype(values.dtype):
+        result = _isna_string_dtype(values)
+    elif dtype.kind in "mM":
+        # this is the NaT pattern
+        result = values.view("i8") == iNaT
+    else:
+        result = np.isnan(values)
+
+    return result
+
+
+def _isna_string_dtype(values: np.ndarray) -> npt.NDArray[np.bool_]:
+    # Working around NumPy ticket 1542
+    dtype = values.dtype
+
+    if dtype.kind in ("S", "U"):
+        result = np.zeros(values.shape, dtype=bool)
+    else:
+        if values.ndim in {1, 2}:
+            result = libmissing.isnaobj(values)
+        else:
+            # 0-D, reached via e.g. mask_missing
+            result = libmissing.isnaobj(values.ravel())
+            result = result.reshape(values.shape)
+
+    return result
+
+
+def _isna_recarray_dtype(values: np.rec.recarray) -> npt.NDArray[np.bool_]:
+    result = np.zeros(values.shape, dtype=bool)
+    for i, record in enumerate(values):
+        record_as_array = np.array(record.tolist())
+        does_record_contain_nan = isna_all(record_as_array)
+        result[i] = np.any(does_record_contain_nan)
+
+    return result
+
+
+@overload
+def notna(obj: Scalar | Pattern | NAType | NaTType) -> bool: ...
+
+
+@overload
+def notna(
+    obj: ArrayLike | Index | list,
+) -> npt.NDArray[np.bool_]: ...
+
+
+@overload
+def notna(obj: NDFrameT) -> NDFrameT: ...
+
+
+# handle unions
+@overload
+def notna(
+    obj: NDFrameT | ArrayLike | Index | list,
+) -> NDFrameT | npt.NDArray[np.bool_]: ...
+
+
+@overload
+def notna(obj: object) -> bool | npt.NDArray[np.bool_] | NDFrame: ...
+
+
+@set_module("pandas")
+def notna(obj: object) -> bool | npt.NDArray[np.bool_] | NDFrame:
+    """
+    Detect non-missing values for an array-like object.
+
+    This function takes a scalar or array-like object and indicates
+    whether values are valid (not missing, which is ``NaN`` in numeric
+    arrays, ``None`` or ``NaN`` in object arrays, ``NaT`` in datetimelike).
+
+    Parameters
+    ----------
+    obj : array-like or object value
+        Object to check for *not* null or *non*-missing values.
+
+    Returns
+    -------
+    bool or array-like of bool
+        For scalar input, returns a scalar boolean.
+        For array input, returns an array of boolean indicating whether each
+        corresponding element is valid.
+
+    See Also
+    --------
+    isna : Boolean inverse of pandas.notna.
+    Series.notna : Detect valid values in a Series.
+    DataFrame.notna : Detect valid values in a DataFrame.
+    Index.notna : Detect valid values in an Index.
+
+    Examples
+    --------
+    Scalar arguments (including strings) result in a scalar boolean.
+
+    >>> pd.notna("dog")
+    True
+
+    >>> pd.notna(pd.NA)
+    False
+
+    >>> pd.notna(np.nan)
+    False
+
+    ndarrays result in an ndarray of booleans.
+
+    >>> array = np.array([[1, np.nan, 3], [4, 5, np.nan]])
+    >>> array
+    array([[ 1., nan,  3.],
+           [ 4.,  5., nan]])
+    >>> pd.notna(array)
+    array([[ True, False,  True],
+           [ True,  True, False]])
+
+    For indexes, an ndarray of booleans is returned.
+
+    >>> index = pd.DatetimeIndex(["2017-07-05", "2017-07-06", None, "2017-07-08"])
+    >>> index
+    DatetimeIndex(['2017-07-05', '2017-07-06', 'NaT', '2017-07-08'],
+                  dtype='datetime64[s]', freq=None)
+    >>> pd.notna(index)
+    array([ True,  True, False,  True])
+
+    For Series and DataFrame, the same type is returned, containing booleans.
+
+    >>> df = pd.DataFrame([["ant", "bee", "cat"], ["dog", None, "fly"]])
+    >>> df
+         0     1    2
+    0  ant   bee  cat
+    1  dog  None  fly
+    >>> pd.notna(df)
+          0      1     2
+    0  True   True  True
+    1  True  False  True
+
+    >>> pd.notna(df[1])
+    0     True
+    1    False
+    Name: 1, dtype: bool
+    """
+    res = isna(obj)
+    if isinstance(res, bool):
+        return not res
+    return ~res
+
+
+notnull = notna
+
+
+def array_equivalent(
+    left,
+    right,
+    strict_nan: bool = False,
+    dtype_equal: bool = False,
+) -> bool:
+    """
+    True if two arrays, left and right, have equal non-NaN elements, and NaNs
+    in corresponding locations.  False otherwise. It is assumed that left and
+    right are NumPy arrays of the same dtype. The behavior of this function
+    (particularly with respect to NaNs) is not defined if the dtypes are
+    different.
+
+    Parameters
+    ----------
+    left, right : ndarrays
+    strict_nan : bool, default False
+        If True, consider NaN and None to be different.
+    dtype_equal : bool, default False
+        Whether `left` and `right` are known to have the same dtype
+        according to `is_dtype_equal`. Some methods like `BlockManager.equals`.
+        require that the dtypes match. Setting this to ``True`` can improve
+        performance, but will give different results for arrays that are
+        equal but different dtypes.
+
+    Returns
+    -------
+    b : bool
+        Returns True if the arrays are equivalent.
+
+    Examples
+    --------
+    >>> array_equivalent(np.array([1, 2, np.nan]), np.array([1, 2, np.nan]))
+    True
+    >>> array_equivalent(np.array([1, np.nan, 2]), np.array([1, 2, np.nan]))
+    False
+    """
+    left, right = np.asarray(left), np.asarray(right)
+
+    # shape compat
+    if left.shape != right.shape:
+        return False
+
+    if dtype_equal:
+        # fastpath when we require that the dtypes match (Block.equals)
+        if left.dtype.kind in "fc":
+            return _array_equivalent_float(left, right)
+        elif left.dtype.kind in "mM":
+            return _array_equivalent_datetimelike(left, right)
+        elif is_string_or_object_np_dtype(left.dtype):
+            # TODO: fastpath for pandas' StringDtype
+            return _array_equivalent_object(left, right, strict_nan)
+        else:
+            return np.array_equal(left, right)
+
+    # Slow path when we allow comparing different dtypes.
+    # Object arrays can contain None, NaN and NaT.
+    # string dtypes must be come to this path for NumPy 1.7.1 compat
+    if left.dtype.kind in "OSU" or right.dtype.kind in "OSU":
+        # Note: `in "OSU"` is non-trivially faster than `in ["O", "S", "U"]`
+        #  or `in ("O", "S", "U")`
+        return _array_equivalent_object(left, right, strict_nan)
+
+    # NaNs can occur in float and complex arrays.
+    if left.dtype.kind in "fc":
+        if not (left.size and right.size):
+            return True
+        return ((left == right) | (isna(left) & isna(right))).all()
+
+    elif left.dtype.kind in "mM" or right.dtype.kind in "mM":
+        # datetime64, timedelta64, Period
+        if left.dtype != right.dtype:
+            return False
+
+        left = left.view("i8")
+        right = right.view("i8")
+
+    # if we have structured dtypes, compare first
+    if (
+        left.dtype.type is np.void or right.dtype.type is np.void
+    ) and left.dtype != right.dtype:
+        return False
+
+    return np.array_equal(left, right)
+
+
+def _array_equivalent_float(left: np.ndarray, right: np.ndarray) -> bool:
+    return bool(((left == right) | (np.isnan(left) & np.isnan(right))).all())
+
+
+def _array_equivalent_datetimelike(left: np.ndarray, right: np.ndarray) -> bool:
+    return np.array_equal(left.view("i8"), right.view("i8"))
+
+
+def _array_equivalent_object(
+    left: np.ndarray, right: np.ndarray, strict_nan: bool
+) -> bool:
+    left = ensure_object(left)
+    right = ensure_object(right)
+
+    mask: npt.NDArray[np.bool_] | None = None
+    if strict_nan:
+        mask = isna(left) & isna(right)
+        if not mask.any():
+            mask = None
+
+    try:
+        if mask is None:
+            return lib.array_equivalent_object(left, right)
+        if not lib.array_equivalent_object(left[~mask], right[~mask]):
+            return False
+        left_remaining = left[mask]
+        right_remaining = right[mask]
+    except ValueError:
+        # can raise a ValueError if left and right cannot be
+        # compared (e.g. nested arrays)
+        left_remaining = left
+        right_remaining = right
+
+    for left_value, right_value in zip(left_remaining, right_remaining):
+        if left_value is NaT and right_value is not NaT:
+            return False
+
+        elif left_value is libmissing.NA and right_value is not libmissing.NA:
+            return False
+
+        elif isinstance(left_value, float) and np.isnan(left_value):
+            if not isinstance(right_value, float) or not np.isnan(right_value):
+                return False
+        else:
+            with warnings.catch_warnings():
+                # suppress numpy's "elementwise comparison failed"
+                warnings.simplefilter("ignore", DeprecationWarning)
+                try:
+                    if np.any(np.asarray(left_value != right_value)):
+                        return False
+                except TypeError as err:
+                    if "boolean value of NA is ambiguous" in str(err):
+                        return False
+                    raise
+                except ValueError:
+                    # numpy can raise a ValueError if left and right cannot be
+                    # compared (e.g. nested arrays)
+                    return False
+    return True
+
+
+def array_equals(left: ArrayLike, right: ArrayLike) -> bool:
+    """
+    ExtensionArray-compatible implementation of array_equivalent.
+    """
+    if left.dtype != right.dtype:
+        return False
+    elif isinstance(left, ABCExtensionArray):
+        return left.equals(right)
+    else:
+        return array_equivalent(left, right, dtype_equal=True)
+
+
+def infer_fill_value(val):
+    """
+    infer the fill value for the nan/NaT from the provided
+    scalar/ndarray/list-like if we are a NaT, return the correct dtyped
+    element to provide proper block construction
+    """
+    if not is_list_like(val):
+        val = [val]
+    val = np.asarray(val)
+    if val.dtype.kind in "mM":
+        return np.array("NaT", dtype=val.dtype)
+    elif val.dtype == object:
+        dtype = lib.infer_dtype(ensure_object(val), skipna=False)
+        if dtype in ["datetime", "datetime64"]:
+            return np.array("NaT", dtype=DT64NS_DTYPE)
+        elif dtype in ["timedelta", "timedelta64"]:
+            return np.array("NaT", dtype=TD64NS_DTYPE)
+        return np.array(np.nan, dtype=object)
+    elif val.dtype.kind == "U":
+        return np.array(np.nan, dtype=val.dtype)
+    return np.nan
+
+
+def construct_1d_array_from_inferred_fill_value(
+    value: object, length: int
+) -> ArrayLike:
+    # Find our empty_value dtype by constructing an array
+    #  from our value and doing a .take on it
+    from pandas.core.algorithms import take_nd
+    from pandas.core.construction import sanitize_array
+    from pandas.core.indexes.base import Index
+
+    arr = sanitize_array(value, Index(range(1)), copy=False)
+    taker = -1 * np.ones(length, dtype=np.intp)
+    return take_nd(arr, taker)
+
+
+def maybe_fill(arr: np.ndarray) -> np.ndarray:
+    """
+    Fill numpy.ndarray with NaN, unless we have a integer or boolean dtype.
+    """
+    if arr.dtype.kind not in "iub":
+        arr.fill(np.nan)
+    return arr
+
+
+def na_value_for_dtype(dtype: DtypeObj, compat: bool = True):
+    """
+    Return a dtype compat na value
+
+    Parameters
+    ----------
+    dtype : string / dtype
+    compat : bool, default True
+
+    Returns
+    -------
+    np.dtype or a pandas dtype
+
+    Examples
+    --------
+    >>> na_value_for_dtype(np.dtype("int64"))
+    0
+    >>> na_value_for_dtype(np.dtype("int64"), compat=False)
+    nan
+    >>> na_value_for_dtype(np.dtype("float64"))
+    nan
+    >>> na_value_for_dtype(np.dtype("complex128"))
+    nan
+    >>> na_value_for_dtype(np.dtype("bool"))
+    False
+    >>> na_value_for_dtype(np.dtype("datetime64[ns]"))
+    numpy.datetime64('NaT')
+    """
+
+    if isinstance(dtype, ExtensionDtype):
+        return dtype.na_value
+    elif dtype.kind in "mM":
+        unit = np.datetime_data(dtype)[0]
+        return dtype.type("NaT", unit)
+    elif dtype.kind in "fc":
+        return np.nan
+    elif dtype.kind in "iu":
+        if compat:
+            return 0
+        return np.nan
+    elif dtype.kind == "b":
+        if compat:
+            return False
+        return np.nan
+    return np.nan
+
+
+def remove_na_arraylike(arr: Series | Index | np.ndarray):
+    """
+    Return array-like containing only true/non-NaN values, possibly empty.
+    """
+    if isinstance(arr.dtype, ExtensionDtype):
+        return arr[notna(arr)]
+    else:
+        return arr[notna(np.asarray(arr))]
+
+
+def is_valid_na_for_dtype(obj, dtype: DtypeObj) -> bool:
+    """
+    isna check that excludes incompatible dtypes
+
+    Parameters
+    ----------
+    obj : object
+    dtype : np.datetime64, np.timedelta64, DatetimeTZDtype, or PeriodDtype
+
+    Returns
+    -------
+    bool
+    """
+    if not lib.is_scalar(obj) or not isna(obj):
+        return False
+    elif dtype.kind == "M":
+        if isinstance(dtype, np.dtype):
+            # i.e. not tzaware
+            return not isinstance(obj, (np.timedelta64, Decimal))
+        # we have to rule out tznaive dt64("NaT")
+        return not isinstance(obj, (np.timedelta64, np.datetime64, Decimal))
+    elif dtype.kind == "m":
+        return not isinstance(obj, (np.datetime64, Decimal))
+    elif dtype.kind in "iufc":
+        # Numeric
+        return obj is not NaT and not isinstance(obj, (np.datetime64, np.timedelta64))
+    elif dtype.kind == "b":
+        # We allow pd.NA, None, np.nan in BooleanArray (same as IntervalDtype)
+        return lib.is_float(obj) or obj is None or obj is libmissing.NA
+
+    elif dtype == _dtype_str:
+        # numpy string dtypes to avoid float np.nan
+        return not isinstance(obj, (np.datetime64, np.timedelta64, Decimal, float))
+
+    elif dtype == _dtype_object:
+        # This is needed for Categorical, but is kind of weird
+        return True
+
+    elif isinstance(dtype, PeriodDtype):
+        return not isinstance(obj, (np.datetime64, np.timedelta64, Decimal))
+
+    elif isinstance(dtype, IntervalDtype):
+        return lib.is_float(obj) or obj is None or obj is libmissing.NA
+
+    elif isinstance(dtype, CategoricalDtype):
+        return is_valid_na_for_dtype(obj, dtype.categories.dtype)
+
+    # fallback, default to allowing NaN, None, NA, NaT
+    return not isinstance(obj, (np.datetime64, np.timedelta64, Decimal))
+
+
+def isna_all(arr: ArrayLike) -> bool:
+    """
+    Optimized equivalent to isna(arr).all()
+    """
+    total_len = len(arr)
+
+    # Usually it's enough to check but a small fraction of values to see if
+    #  a block is NOT null, chunks should help in such cases.
+    #  parameters 1000 and 40 were chosen arbitrarily
+    chunk_len = max(total_len // 40, 1000)
+
+    dtype = arr.dtype
+    if lib.is_np_dtype(dtype, "f"):
+        checker = np.isnan
+
+    elif (lib.is_np_dtype(dtype, "mM")) or isinstance(
+        dtype, (DatetimeTZDtype, PeriodDtype)
+    ):
+        # error: Incompatible types in assignment (expression has type
+        # "Callable[[Any], Any]", variable has type "ufunc")
+        checker = lambda x: np.asarray(x.view("i8")) == iNaT  # type: ignore[assignment]
+
+    else:
+        # error: Incompatible types in assignment (expression has type "Callable[[Any],
+        # Any]", variable has type "ufunc")
+        checker = _isna_array  # type: ignore[assignment]
+
+    return all(
+        checker(arr[i : i + chunk_len]).all() for i in range(0, total_len, chunk_len)
+    )
diff --git a/pandas/pandas/core/flags.py b/pandas/pandas/core/flags.py
new file mode 100644
index 0000000000000..eceb86dc61d9f
--- /dev/null
+++ b/pandas/pandas/core/flags.py
@@ -0,0 +1,126 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+import weakref
+
+if TYPE_CHECKING:
+    from pandas.core.generic import NDFrame
+
+
+class Flags:
+    """
+    Flags that apply to pandas objects.
+
+    “Flags” differ from “metadata”. Flags reflect properties of the pandas
+    object (the Series or DataFrame). Metadata refer to properties of the
+    dataset, and should be stored in DataFrame.attrs.
+
+    Parameters
+    ----------
+    obj : Series or DataFrame
+        The object these flags are associated with.
+    allows_duplicate_labels : bool, default True
+        Whether to allow duplicate labels in this object. By default,
+        duplicate labels are permitted. Setting this to ``False`` will
+        cause an :class:`errors.DuplicateLabelError` to be raised when
+        `index` (or columns for DataFrame) is not unique, or any
+        subsequent operation on introduces duplicates.
+        See :ref:`duplicates.disallow` for more.
+
+        .. warning::
+
+           This is an experimental feature. Currently, many methods fail to
+           propagate the ``allows_duplicate_labels`` value. In future versions
+           it is expected that every method taking or returning one or more
+           DataFrame or Series objects will propagate ``allows_duplicate_labels``.
+
+    See Also
+    --------
+    DataFrame.attrs : Dictionary of global attributes of this dataset.
+    Series.attrs : Dictionary of global attributes of this dataset.
+
+    Examples
+    --------
+    Attributes can be set in two ways:
+
+    >>> df = pd.DataFrame()
+    >>> df.flags
+    <Flags(allows_duplicate_labels=True)>
+    >>> df.flags.allows_duplicate_labels = False
+    >>> df.flags
+    <Flags(allows_duplicate_labels=False)>
+
+    >>> df.flags["allows_duplicate_labels"] = True
+    >>> df.flags
+    <Flags(allows_duplicate_labels=True)>
+    """
+
+    _keys: set[str] = {"allows_duplicate_labels"}
+
+    def __init__(self, obj: NDFrame, *, allows_duplicate_labels: bool) -> None:
+        self._allows_duplicate_labels = allows_duplicate_labels
+        self._obj = weakref.ref(obj)
+
+    @property
+    def allows_duplicate_labels(self) -> bool:
+        """
+        Whether this object allows duplicate labels.
+
+        Setting ``allows_duplicate_labels=False`` ensures that the
+        index (and columns of a DataFrame) are unique. Most methods
+        that accept and return a Series or DataFrame will propagate
+        the value of ``allows_duplicate_labels``.
+
+        See :ref:`duplicates` for more.
+
+        See Also
+        --------
+        DataFrame.attrs : Set global metadata on this object.
+        DataFrame.set_flags : Set global flags on this object.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2]}, index=["a", "a"])
+        >>> df.flags.allows_duplicate_labels
+        True
+        >>> df.flags.allows_duplicate_labels = False
+        Traceback (most recent call last):
+            ...
+        pandas.errors.DuplicateLabelError: Index has duplicates.
+              positions
+        label
+        a        [0, 1]
+        """
+        return self._allows_duplicate_labels
+
+    @allows_duplicate_labels.setter
+    def allows_duplicate_labels(self, value: bool) -> None:
+        value = bool(value)
+        obj = self._obj()
+        if obj is None:
+            raise ValueError("This flag's object has been deleted.")
+
+        if not value:
+            for ax in obj.axes:
+                ax._maybe_check_unique()
+
+        self._allows_duplicate_labels = value
+
+    def __getitem__(self, key: str):
+        if key not in self._keys:
+            raise KeyError(key)
+
+        return getattr(self, key)
+
+    def __setitem__(self, key: str, value) -> None:
+        if key not in self._keys:
+            raise ValueError(f"Unknown flag {key}. Must be one of {self._keys}")
+        setattr(self, key, value)
+
+    def __repr__(self) -> str:
+        return f"<Flags(allows_duplicate_labels={self.allows_duplicate_labels})>"
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, type(self)):
+            return self.allows_duplicate_labels == other.allows_duplicate_labels
+        return False
diff --git a/pandas/pandas/core/frame.py b/pandas/pandas/core/frame.py
new file mode 100644
index 0000000000000..4cf595d167c46
--- /dev/null
+++ b/pandas/pandas/core/frame.py
@@ -0,0 +1,13867 @@
+"""
+DataFrame
+---------
+An efficient 2D container for potentially mixed-type time series or other
+labeled data series.
+
+Similar to its R counterpart, data.frame, except providing automatic data
+alignment and a host of useful data manipulation methods having to do with the
+labeling information
+"""
+
+from __future__ import annotations
+
+import collections
+from collections import abc
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    Iterator,
+    Mapping,
+    Sequence,
+)
+import functools
+from io import StringIO
+import itertools
+import operator
+import sys
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+from numpy import ma
+
+from pandas._config import get_option
+
+from pandas._libs import (
+    algos as libalgos,
+    lib,
+    properties,
+)
+from pandas._libs.hashtable import duplicated
+from pandas._libs.lib import is_range_indexer
+from pandas.compat import PYPY
+from pandas.compat._constants import REF_COUNT
+from pandas.compat._optional import import_optional_dependency
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    ChainedAssignmentError,
+    InvalidIndexError,
+)
+from pandas.errors.cow import (
+    _chained_assignment_method_msg,
+    _chained_assignment_msg,
+)
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    deprecate_nonkeyword_arguments,
+    doc,
+    set_module,
+)
+from pandas.util._exceptions import (
+    find_stack_level,
+    rewrite_warning,
+)
+from pandas.util._validators import (
+    validate_ascending,
+    validate_bool_kwarg,
+    validate_percentile,
+)
+
+from pandas.core.dtypes.cast import (
+    LossySetitemError,
+    can_hold_element,
+    construct_1d_arraylike_from_scalar,
+    construct_2d_arraylike_from_scalar,
+    find_common_type,
+    infer_dtype_from_scalar,
+    invalidate_string_dtypes,
+    maybe_downcast_to_dtype,
+)
+from pandas.core.dtypes.common import (
+    infer_dtype_from_object,
+    is_1d_only_ea_dtype,
+    is_array_like,
+    is_bool_dtype,
+    is_dataclass,
+    is_dict_like,
+    is_float,
+    is_float_dtype,
+    is_hashable,
+    is_integer,
+    is_integer_dtype,
+    is_iterator,
+    is_list_like,
+    is_scalar,
+    is_sequence,
+    needs_i8_conversion,
+    pandas_dtype,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    BaseMaskedDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    notna,
+)
+
+from pandas.core import (
+    algorithms,
+    common as com,
+    nanops,
+    ops,
+    roperator,
+)
+from pandas.core.accessor import Accessor
+from pandas.core.apply import reconstruct_and_relabel_result
+from pandas.core.array_algos.take import take_2d_multi
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays import (
+    BaseMaskedArray,
+    DatetimeArray,
+    ExtensionArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.sparse import SparseFrameAccessor
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    sanitize_array,
+    sanitize_masked_array,
+)
+from pandas.core.generic import (
+    NDFrame,
+    make_doc,
+)
+from pandas.core.indexers import check_key_length
+from pandas.core.indexes.api import (
+    DatetimeIndex,
+    Index,
+    PeriodIndex,
+    default_index,
+    ensure_index,
+    ensure_index_from_sequences,
+)
+from pandas.core.indexes.multi import (
+    MultiIndex,
+    maybe_droplevels,
+)
+from pandas.core.indexing import (
+    check_bool_indexer,
+    check_dict_or_set_indexers,
+)
+from pandas.core.internals import BlockManager
+from pandas.core.internals.construction import (
+    arrays_to_mgr,
+    dataclasses_to_dicts,
+    dict_to_mgr,
+    ndarray_to_mgr,
+    nested_data_to_arrays,
+    rec_array_to_mgr,
+    reorder_arrays,
+    to_arrays,
+    treat_as_nested,
+)
+from pandas.core.methods import selectn
+from pandas.core.reshape.melt import melt
+from pandas.core.series import Series
+from pandas.core.shared_docs import _shared_docs
+from pandas.core.sorting import (
+    get_group_index,
+    lexsort_indexer,
+    nargsort,
+)
+
+from pandas.io.common import get_handle
+from pandas.io.formats import (
+    console,
+    format as fmt,
+)
+from pandas.io.formats.info import (
+    INFO_DOCSTRING,
+    DataFrameInfo,
+    frame_sub_kwargs,
+)
+import pandas.plotting
+
+if TYPE_CHECKING:
+    import datetime
+
+    from pandas._libs.internals import BlockValuesRefs
+    from pandas._typing import (
+        AggFuncType,
+        AnyAll,
+        AnyArrayLike,
+        ArrayLike,
+        Axes,
+        Axis,
+        AxisInt,
+        ColspaceArgType,
+        CompressionOptions,
+        CorrelationMethod,
+        DropKeep,
+        Dtype,
+        DtypeObj,
+        FilePath,
+        FloatFormatType,
+        FormattersType,
+        Frequency,
+        FromDictOrient,
+        HashableT,
+        HashableT2,
+        IgnoreRaise,
+        IndexKeyFunc,
+        IndexLabel,
+        JoinValidate,
+        Level,
+        ListLike,
+        MergeHow,
+        MergeValidate,
+        MutableMappingT,
+        NaPosition,
+        NsmallestNlargestKeep,
+        PythonFuncType,
+        QuantileInterpolation,
+        ReadBuffer,
+        ReindexMethod,
+        Renamer,
+        Scalar,
+        Self,
+        SequenceNotStr,
+        SortKind,
+        StorageOptions,
+        Suffixes,
+        T,
+        ToStataByteorder,
+        ToTimestampHow,
+        UpdateJoin,
+        ValueKeyFunc,
+        WriteBuffer,
+        XMLParsers,
+        npt,
+    )
+
+    from pandas.core.groupby.generic import DataFrameGroupBy
+    from pandas.core.interchange.dataframe_protocol import DataFrame as DataFrameXchg
+    from pandas.core.internals.managers import SingleBlockManager
+
+    from pandas.io.formats.style import Styler
+
+# ---------------------------------------------------------------------
+# Docstring templates
+
+_shared_doc_kwargs = {
+    "axes": "index, columns",
+    "klass": "DataFrame",
+    "axes_single_arg": "{0 or 'index', 1 or 'columns'}",
+    "axis": """axis : {0 or 'index', 1 or 'columns'}, default 0
+        If 0 or 'index': apply function to each column.
+        If 1 or 'columns': apply function to each row.""",
+    "inplace": """
+    inplace : bool, default False
+        Whether to modify the DataFrame rather than creating a new one.""",
+    "optional_by": """
+by : str or list of str
+    Name or list of names to sort by.
+
+    - if `axis` is 0 or `'index'` then `by` may contain index
+      levels and/or column labels.
+    - if `axis` is 1 or `'columns'` then `by` may contain column
+      levels and/or index labels.""",
+    "optional_reindex": """
+labels : array-like, optional
+    New labels / index to conform the axis specified by 'axis' to.
+index : array-like, optional
+    New labels for the index. Preferably an Index object to avoid
+    duplicating data.
+columns : array-like, optional
+    New labels for the columns. Preferably an Index object to avoid
+    duplicating data.
+axis : int or str, optional
+    Axis to target. Can be either the axis name ('index', 'columns')
+    or number (0, 1).""",
+}
+
+_merge_doc = """
+Merge DataFrame or named Series objects with a database-style join.
+
+A named Series object is treated as a DataFrame with a single named column.
+
+The join is done on columns or indexes. If joining columns on
+columns, the DataFrame indexes *will be ignored*. Otherwise if joining indexes
+on indexes or indexes on a column or columns, the index will be passed on.
+When performing a cross merge, no column specifications to merge on are
+allowed.
+
+.. warning::
+
+    If both key columns contain rows where the key is a null value, those
+    rows will be matched against each other. This is different from usual SQL
+    join behaviour and can lead to unexpected results.
+
+Parameters
+----------%s
+right : DataFrame or named Series
+    Object to merge with.
+how : {'left', 'right', 'outer', 'inner', 'cross', 'left_anti', 'right_anti'},
+    default 'inner'
+    Type of merge to be performed.
+
+    * left: use only keys from left frame, similar to a SQL left outer join;
+      preserve key order.
+    * right: use only keys from right frame, similar to a SQL right outer join;
+      preserve key order.
+    * outer: use union of keys from both frames, similar to a SQL full outer
+      join; sort keys lexicographically.
+    * inner: use intersection of keys from both frames, similar to a SQL inner
+      join; preserve the order of the left keys.
+    * cross: creates the cartesian product from both frames, preserves the order
+      of the left keys.
+    * left_anti: use only keys from left frame that are not in right frame, similar
+      to SQL left anti join; preserve key order.
+    * right_anti: use only keys from right frame that are not in left frame, similar
+      to SQL right anti join; preserve key order.
+on : label or list
+    Column or index level names to join on. These must be found in both
+    DataFrames. If `on` is None and not merging on indexes then this defaults
+    to the intersection of the columns in both DataFrames.
+left_on : label or list, or array-like
+    Column or index level names to join on in the left DataFrame. Can also
+    be an array or list of arrays of the length of the left DataFrame.
+    These arrays are treated as if they are columns.
+right_on : label or list, or array-like
+    Column or index level names to join on in the right DataFrame. Can also
+    be an array or list of arrays of the length of the right DataFrame.
+    These arrays are treated as if they are columns.
+left_index : bool, default False
+    Use the index from the left DataFrame as the join key(s). If it is a
+    MultiIndex, the number of keys in the other DataFrame (either the index
+    or a number of columns) must match the number of levels.
+right_index : bool, default False
+    Use the index from the right DataFrame as the join key. Same caveats as
+    left_index.
+sort : bool, default False
+    Sort the join keys lexicographically in the result DataFrame. If False,
+    the order of the join keys depends on the join type (how keyword).
+suffixes : list-like, default is ("_x", "_y")
+    A length-2 sequence where each element is optionally a string
+    indicating the suffix to add to overlapping column names in
+    `left` and `right` respectively. Pass a value of `None` instead
+    of a string to indicate that the column name from `left` or
+    `right` should be left as-is, with no suffix. At least one of the
+    values must not be None.
+copy : bool, default False
+    If False, avoid copy if possible.
+
+    .. note::
+        The `copy` keyword will change behavior in pandas 3.0.
+        `Copy-on-Write
+        <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+        will be enabled by default, which means that all methods with a
+        `copy` keyword will use a lazy copy mechanism to defer the copy and
+        ignore the `copy` keyword. The `copy` keyword will be removed in a
+        future version of pandas.
+
+        You can already get the future behavior and improvements through
+        enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+    .. deprecated:: 3.0.0
+indicator : bool or str, default False
+    If True, adds a column to the output DataFrame called "_merge" with
+    information on the source of each row. The column can be given a different
+    name by providing a string argument. The column will have a Categorical
+    type with the value of "left_only" for observations whose merge key only
+    appears in the left DataFrame, "right_only" for observations
+    whose merge key only appears in the right DataFrame, and "both"
+    if the observation's merge key is found in both DataFrames.
+
+validate : str, optional
+    If specified, checks if merge is of specified type.
+
+    * "one_to_one" or "1:1": check if merge keys are unique in both
+      left and right datasets.
+    * "one_to_many" or "1:m": check if merge keys are unique in left
+      dataset.
+    * "many_to_one" or "m:1": check if merge keys are unique in right
+      dataset.
+    * "many_to_many" or "m:m": allowed, but does not result in checks.
+
+Returns
+-------
+DataFrame
+    A DataFrame of the two merged objects.
+
+See Also
+--------
+merge_ordered : Merge with optional filling/interpolation.
+merge_asof : Merge on nearest keys.
+DataFrame.join : Similar method using indices.
+
+Examples
+--------
+>>> df1 = pd.DataFrame({'lkey': ['foo', 'bar', 'baz', 'foo'],
+...                     'value': [1, 2, 3, 5]})
+>>> df2 = pd.DataFrame({'rkey': ['foo', 'bar', 'baz', 'foo'],
+...                     'value': [5, 6, 7, 8]})
+>>> df1
+    lkey value
+0   foo      1
+1   bar      2
+2   baz      3
+3   foo      5
+>>> df2
+    rkey value
+0   foo      5
+1   bar      6
+2   baz      7
+3   foo      8
+
+Merge df1 and df2 on the lkey and rkey columns. The value columns have
+the default suffixes, _x and _y, appended.
+
+>>> df1.merge(df2, left_on='lkey', right_on='rkey')
+  lkey  value_x rkey  value_y
+0  foo        1  foo        5
+1  foo        1  foo        8
+2  bar        2  bar        6
+3  baz        3  baz        7
+4  foo        5  foo        5
+5  foo        5  foo        8
+
+Merge DataFrames df1 and df2 with specified left and right suffixes
+appended to any overlapping columns.
+
+>>> df1.merge(df2, left_on='lkey', right_on='rkey',
+...           suffixes=('_left', '_right'))
+  lkey  value_left rkey  value_right
+0  foo           1  foo            5
+1  foo           1  foo            8
+2  bar           2  bar            6
+3  baz           3  baz            7
+4  foo           5  foo            5
+5  foo           5  foo            8
+
+Merge DataFrames df1 and df2, but raise an exception if the DataFrames have
+any overlapping columns.
+
+>>> df1.merge(df2, left_on='lkey', right_on='rkey', suffixes=(False, False))
+Traceback (most recent call last):
+...
+ValueError: columns overlap but no suffix specified:
+    Index(['value'], dtype='object')
+
+>>> df1 = pd.DataFrame({'a': ['foo', 'bar'], 'b': [1, 2]})
+>>> df2 = pd.DataFrame({'a': ['foo', 'baz'], 'c': [3, 4]})
+>>> df1
+      a  b
+0   foo  1
+1   bar  2
+>>> df2
+      a  c
+0   foo  3
+1   baz  4
+
+>>> df1.merge(df2, how='inner', on='a')
+      a  b  c
+0   foo  1  3
+
+>>> df1.merge(df2, how='left', on='a')
+      a  b  c
+0   foo  1  3.0
+1   bar  2  NaN
+
+>>> df1 = pd.DataFrame({'left': ['foo', 'bar']})
+>>> df2 = pd.DataFrame({'right': [7, 8]})
+>>> df1
+    left
+0   foo
+1   bar
+>>> df2
+    right
+0   7
+1   8
+
+>>> df1.merge(df2, how='cross')
+   left  right
+0   foo      7
+1   foo      8
+2   bar      7
+3   bar      8
+"""
+
+
+# -----------------------------------------------------------------------
+# DataFrame class
+
+
+@set_module("pandas")
+class DataFrame(NDFrame, OpsMixin):
+    """
+    Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+
+    Data structure also contains labeled axes (rows and columns).
+    Arithmetic operations align on both row and column labels. Can be
+    thought of as a dict-like container for Series objects. The primary
+    pandas data structure.
+
+    Parameters
+    ----------
+    data : ndarray (structured or homogeneous), Iterable, dict, or DataFrame
+        Dict can contain Series, arrays, constants, dataclass or list-like objects. If
+        data is a dict, column order follows insertion-order. If a dict contains Series
+        which have an index defined, it is aligned by its index. This alignment also
+        occurs if data is a Series or a DataFrame itself. Alignment is done on
+        Series/DataFrame inputs.
+
+        If data is a list of dicts, column order follows insertion-order.
+
+    index : Index or array-like
+        Index to use for resulting frame. Will default to RangeIndex if
+        no indexing information part of input data and no index provided.
+    columns : Index or array-like
+        Column labels to use for resulting frame when data does not have them,
+        defaulting to RangeIndex(0, 1, 2, ..., n). If data contains column labels,
+        will perform column selection instead.
+    dtype : dtype, default None
+        Data type to force. Only a single dtype is allowed. If None, infer.
+        If ``data`` is DataFrame then is ignored.
+    copy : bool or None, default None
+        Copy data from inputs.
+        For dict data, the default of None behaves like ``copy=True``.  For DataFrame
+        or 2d ndarray input, the default of None behaves like ``copy=False``.
+        If data is a dict containing one or more Series (possibly of different dtypes),
+        ``copy=False`` will ensure that these inputs are not copied.
+
+        .. versionchanged:: 1.3.0
+
+    See Also
+    --------
+    DataFrame.from_records : Constructor from tuples, also record arrays.
+    DataFrame.from_dict : From dicts of Series, arrays, or dicts.
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+    read_table : Read general delimited file into DataFrame.
+    read_clipboard : Read text from clipboard into DataFrame.
+
+    Notes
+    -----
+    Please reference the :ref:`User Guide <basics.dataframe>` for more information.
+
+    Examples
+    --------
+    Constructing DataFrame from a dictionary.
+
+    >>> d = {"col1": [1, 2], "col2": [3, 4]}
+    >>> df = pd.DataFrame(data=d)
+    >>> df
+       col1  col2
+    0     1     3
+    1     2     4
+
+    Notice that the inferred dtype is int64.
+
+    >>> df.dtypes
+    col1    int64
+    col2    int64
+    dtype: object
+
+    To enforce a single dtype:
+
+    >>> df = pd.DataFrame(data=d, dtype=np.int8)
+    >>> df.dtypes
+    col1    int8
+    col2    int8
+    dtype: object
+
+    Constructing DataFrame from a dictionary including Series:
+
+    >>> d = {"col1": [0, 1, 2, 3], "col2": pd.Series([2, 3], index=[2, 3])}
+    >>> pd.DataFrame(data=d, index=[0, 1, 2, 3])
+       col1  col2
+    0     0   NaN
+    1     1   NaN
+    2     2   2.0
+    3     3   3.0
+
+    Constructing DataFrame from numpy ndarray:
+
+    >>> df2 = pd.DataFrame(
+    ...     np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]]), columns=["a", "b", "c"]
+    ... )
+    >>> df2
+       a  b  c
+    0  1  2  3
+    1  4  5  6
+    2  7  8  9
+
+    Constructing DataFrame from a numpy ndarray that has labeled columns:
+
+    >>> data = np.array(
+    ...     [(1, 2, 3), (4, 5, 6), (7, 8, 9)],
+    ...     dtype=[("a", "i4"), ("b", "i4"), ("c", "i4")],
+    ... )
+    >>> df3 = pd.DataFrame(data, columns=["c", "a"])
+    >>> df3
+       c  a
+    0  3  1
+    1  6  4
+    2  9  7
+
+    Constructing DataFrame from dataclass:
+
+    >>> from dataclasses import make_dataclass
+    >>> Point = make_dataclass("Point", [("x", int), ("y", int)])
+    >>> pd.DataFrame([Point(0, 0), Point(0, 3), Point(2, 3)])
+       x  y
+    0  0  0
+    1  0  3
+    2  2  3
+
+    Constructing DataFrame from Series/DataFrame:
+
+    >>> ser = pd.Series([1, 2, 3], index=["a", "b", "c"])
+    >>> df = pd.DataFrame(data=ser, index=["a", "c"])
+    >>> df
+       0
+    a  1
+    c  3
+
+    >>> df1 = pd.DataFrame([1, 2, 3], index=["a", "b", "c"], columns=["x"])
+    >>> df2 = pd.DataFrame(data=df1, index=["a", "c"])
+    >>> df2
+       x
+    a  1
+    c  3
+    """
+
+    _internal_names_set = {"columns", "index"} | NDFrame._internal_names_set
+    _typ = "dataframe"
+    _HANDLED_TYPES = (Series, Index, ExtensionArray, np.ndarray)
+    _accessors: set[str] = {"sparse"}
+    _hidden_attrs: frozenset[str] = NDFrame._hidden_attrs | frozenset([])
+    _mgr: BlockManager
+
+    # similar to __array_priority__, positions DataFrame before Series, Index,
+    #  and ExtensionArray.  Should NOT be overridden by subclasses.
+    __pandas_priority__ = 4000
+
+    @property
+    def _constructor(self) -> type[DataFrame]:
+        return DataFrame
+
+    def _constructor_from_mgr(self, mgr, axes) -> DataFrame:
+        df = DataFrame._from_mgr(mgr, axes=axes)
+
+        if type(self) is DataFrame:
+            # This would also work `if self._constructor is DataFrame`, but
+            #  this check is slightly faster, benefiting the most-common case.
+            return df
+
+        elif type(self).__name__ == "GeoDataFrame":
+            # Shim until geopandas can override their _constructor_from_mgr
+            #  bc they have different behavior for Managers than for DataFrames
+            return self._constructor(mgr)
+
+        # We assume that the subclass __init__ knows how to handle a
+        #  pd.DataFrame object.
+        return self._constructor(df)
+
+    _constructor_sliced: Callable[..., Series] = Series
+
+    def _constructor_sliced_from_mgr(self, mgr, axes) -> Series:
+        ser = Series._from_mgr(mgr, axes)
+        ser._name = None  # caller is responsible for setting real name
+
+        if type(self) is DataFrame:
+            # This would also work `if self._constructor_sliced is Series`, but
+            #  this check is slightly faster, benefiting the most-common case.
+            return ser
+
+        # We assume that the subclass __init__ knows how to handle a
+        #  pd.Series object.
+        return self._constructor_sliced(ser)
+
+    # ----------------------------------------------------------------------
+    # Constructors
+
+    def __init__(
+        self,
+        data=None,
+        index: Axes | None = None,
+        columns: Axes | None = None,
+        dtype: Dtype | None = None,
+        copy: bool | None = None,
+    ) -> None:
+        allow_mgr = False
+        if dtype is not None:
+            dtype = self._validate_dtype(dtype)
+
+        if isinstance(data, DataFrame):
+            data = data._mgr
+            allow_mgr = True
+            if not copy:
+                # if not copying data, ensure to still return a shallow copy
+                # to avoid the result sharing the same Manager
+                data = data.copy(deep=False)
+
+        if isinstance(data, BlockManager):
+            if not allow_mgr:
+                # GH#52419
+                warnings.warn(
+                    f"Passing a {type(data).__name__} to {type(self).__name__} "
+                    "is deprecated and will raise in a future version. "
+                    "Use public APIs instead.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+
+            data = data.copy(deep=False)
+            # first check if a Manager is passed without any other arguments
+            # -> use fastpath (without checking Manager type)
+            if index is None and columns is None and dtype is None and not copy:
+                # GH#33357 fastpath
+                NDFrame.__init__(self, data)
+                return
+
+        # GH47215
+        if isinstance(index, set):
+            raise ValueError("index cannot be a set")
+        if isinstance(columns, set):
+            raise ValueError("columns cannot be a set")
+
+        if copy is None:
+            if isinstance(data, dict):
+                # retain pre-GH#38939 default behavior
+                copy = True
+            elif not isinstance(data, (Index, DataFrame, Series)):
+                copy = True
+            else:
+                copy = False
+
+        if data is None:
+            index = index if index is not None else default_index(0)
+            columns = columns if columns is not None else default_index(0)
+            dtype = dtype if dtype is not None else pandas_dtype(object)
+            data = []
+
+        if isinstance(data, BlockManager):
+            mgr = self._init_mgr(
+                data, axes={"index": index, "columns": columns}, dtype=dtype, copy=copy
+            )
+
+        elif isinstance(data, dict):
+            # GH#38939 de facto copy defaults to False only in non-dict cases
+            mgr = dict_to_mgr(data, index, columns, dtype=dtype, copy=copy)
+        elif isinstance(data, ma.MaskedArray):
+            from numpy.ma import mrecords
+
+            # masked recarray
+            if isinstance(data, mrecords.MaskedRecords):
+                raise TypeError(
+                    "MaskedRecords are not supported. Pass "
+                    "{name: data[name] for name in data.dtype.names} "
+                    "instead"
+                )
+
+            # a masked array
+            data = sanitize_masked_array(data)
+            mgr = ndarray_to_mgr(
+                data,
+                index,
+                columns,
+                dtype=dtype,
+                copy=copy,
+            )
+
+        elif isinstance(data, (np.ndarray, Series, Index, ExtensionArray)):
+            if data.dtype.names:
+                # i.e. numpy structured array
+                data = cast(np.ndarray, data)
+                mgr = rec_array_to_mgr(
+                    data,
+                    index,
+                    columns,
+                    dtype,
+                    copy,
+                )
+            elif getattr(data, "name", None) is not None:
+                # i.e. Series/Index with non-None name
+                mgr = dict_to_mgr(
+                    # error: Item "ndarray" of "Union[ndarray, Series, Index]" has no
+                    # attribute "name"
+                    {data.name: data},  # type: ignore[union-attr]
+                    index,
+                    columns,
+                    dtype=dtype,
+                    copy=copy,
+                )
+            else:
+                mgr = ndarray_to_mgr(
+                    data,
+                    index,
+                    columns,
+                    dtype=dtype,
+                    copy=copy,
+                )
+
+        # For data is list-like, or Iterable (will consume into list)
+        elif is_list_like(data):
+            if not isinstance(data, abc.Sequence):
+                if hasattr(data, "__array__"):
+                    # GH#44616 big perf improvement for e.g. pytorch tensor
+                    data = np.asarray(data)
+                else:
+                    data = list(data)
+            if len(data) > 0:
+                if is_dataclass(data[0]):
+                    data = dataclasses_to_dicts(data)
+                if not isinstance(data, np.ndarray) and treat_as_nested(data):
+                    # exclude ndarray as we may have cast it a few lines above
+                    if columns is not None:
+                        columns = ensure_index(columns)
+                    arrays, columns, index = nested_data_to_arrays(
+                        # error: Argument 3 to "nested_data_to_arrays" has incompatible
+                        # type "Optional[Collection[Any]]"; expected "Optional[Index]"
+                        data,
+                        columns,
+                        index,  # type: ignore[arg-type]
+                        dtype,
+                    )
+                    mgr = arrays_to_mgr(
+                        arrays,
+                        columns,
+                        index,
+                        dtype=dtype,
+                    )
+                else:
+                    mgr = ndarray_to_mgr(
+                        data,
+                        index,
+                        columns,
+                        dtype=dtype,
+                        copy=copy,
+                    )
+            else:
+                mgr = dict_to_mgr(
+                    {},
+                    index,
+                    columns if columns is not None else default_index(0),
+                    dtype=dtype,
+                )
+        # For data is scalar
+        else:
+            if index is None or columns is None:
+                raise ValueError("DataFrame constructor not properly called!")
+
+            index = ensure_index(index)
+            columns = ensure_index(columns)
+
+            if not dtype:
+                dtype, _ = infer_dtype_from_scalar(data)
+
+            # For data is a scalar extension dtype
+            if isinstance(dtype, ExtensionDtype):
+                # TODO(EA2D): special case not needed with 2D EAs
+
+                values = [
+                    construct_1d_arraylike_from_scalar(data, len(index), dtype)
+                    for _ in range(len(columns))
+                ]
+                mgr = arrays_to_mgr(values, columns, index, dtype=None)
+            else:
+                arr2d = construct_2d_arraylike_from_scalar(
+                    data,
+                    len(index),
+                    len(columns),
+                    dtype,
+                    copy,
+                )
+
+                mgr = ndarray_to_mgr(
+                    arr2d,
+                    index,
+                    columns,
+                    dtype=arr2d.dtype,
+                    copy=False,
+                )
+
+        NDFrame.__init__(self, mgr)
+
+    # ----------------------------------------------------------------------
+
+    def __dataframe__(
+        self, nan_as_null: bool = False, allow_copy: bool = True
+    ) -> DataFrameXchg:
+        """
+        Return the dataframe interchange object implementing the interchange protocol.
+
+        .. note::
+
+           For new development, we highly recommend using the Arrow C Data Interface
+           alongside the Arrow PyCapsule Interface instead of the interchange protocol
+
+        .. warning::
+
+            Due to severe implementation issues, we recommend only considering using the
+            interchange protocol in the following cases:
+
+            - converting to pandas: for pandas >= 2.0.3
+            - converting from pandas: for pandas >= 3.0.0
+
+        Parameters
+        ----------
+        nan_as_null : bool, default False
+            `nan_as_null` is DEPRECATED and has no effect. Please avoid using
+            it; it will be removed in a future release.
+        allow_copy : bool, default True
+            Whether to allow memory copying when exporting. If set to False
+            it would cause non-zero-copy exports to fail.
+
+        Returns
+        -------
+        DataFrame interchange object
+            The object which consuming library can use to ingress the dataframe.
+
+        See Also
+        --------
+        DataFrame.from_records : Constructor from tuples, also record arrays.
+        DataFrame.from_dict : From dicts of Series, arrays, or dicts.
+
+        Notes
+        -----
+        Details on the interchange protocol:
+        https://data-apis.org/dataframe-protocol/latest/index.html
+
+        Examples
+        --------
+        >>> df_not_necessarily_pandas = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> interchange_object = df_not_necessarily_pandas.__dataframe__()
+        >>> interchange_object.column_names()
+        Index(['A', 'B'], dtype='object')
+        >>> df_pandas = pd.api.interchange.from_dataframe(
+        ...     interchange_object.select_columns_by_name(["A"])
+        ... )
+        >>> df_pandas
+             A
+        0    1
+        1    2
+
+        These methods (``column_names``, ``select_columns_by_name``) should work
+        for any dataframe library which implements the interchange protocol.
+        """
+
+        from pandas.core.interchange.dataframe import PandasDataFrameXchg
+
+        return PandasDataFrameXchg(self, allow_copy=allow_copy)
+
+    def __arrow_c_stream__(self, requested_schema=None):
+        """
+        Export the pandas DataFrame as an Arrow C stream PyCapsule.
+
+        This relies on pyarrow to convert the pandas DataFrame to the Arrow
+        format (and follows the default behaviour of ``pyarrow.Table.from_pandas``
+        in its handling of the index, i.e. store the index as a column except
+        for RangeIndex).
+        This conversion is not necessarily zero-copy.
+
+        Parameters
+        ----------
+        requested_schema : PyCapsule, default None
+            The schema to which the dataframe should be casted, passed as a
+            PyCapsule containing a C ArrowSchema representation of the
+            requested schema.
+
+        Returns
+        -------
+        PyCapsule
+        """
+        pa = import_optional_dependency("pyarrow", min_version="14.0.0")
+        if requested_schema is not None:
+            requested_schema = pa.Schema._import_from_c_capsule(requested_schema)
+        table = pa.Table.from_pandas(self, schema=requested_schema)
+        return table.__arrow_c_stream__()
+
+    # ----------------------------------------------------------------------
+
+    @property
+    def axes(self) -> list[Index]:
+        """
+        Return a list representing the axes of the DataFrame.
+
+        It has the row axis labels and column axis labels as the only members.
+        They are returned in that order.
+
+        See Also
+        --------
+        DataFrame.index: The index (row labels) of the DataFrame.
+        DataFrame.columns: The column labels of the DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df.axes
+        [RangeIndex(start=0, stop=2, step=1), Index(['col1', 'col2'],
+        dtype='object')]
+        """
+        return [self.index, self.columns]
+
+    @property
+    def shape(self) -> tuple[int, int]:
+        """
+        Return a tuple representing the dimensionality of the DataFrame.
+
+        Unlike the `len()` method, which only returns the number of rows, `shape`
+        provides both row and column counts, making it a more informative method for
+        understanding dataset size.
+
+        See Also
+        --------
+        numpy.ndarray.shape : Tuple of array dimensions.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df.shape
+        (2, 2)
+
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4], "col3": [5, 6]})
+        >>> df.shape
+        (2, 3)
+        """
+        return len(self.index), len(self.columns)
+
+    @property
+    def _is_homogeneous_type(self) -> bool:
+        """
+        Whether all the columns in a DataFrame have the same type.
+
+        Returns
+        -------
+        bool
+
+        Examples
+        --------
+        >>> DataFrame({"A": [1, 2], "B": [3, 4]})._is_homogeneous_type
+        True
+        >>> DataFrame({"A": [1, 2], "B": [3.0, 4.0]})._is_homogeneous_type
+        False
+
+        Items with the same type but different sizes are considered
+        different types.
+
+        >>> DataFrame(
+        ...     {
+        ...         "A": np.array([1, 2], dtype=np.int32),
+        ...         "B": np.array([1, 2], dtype=np.int64),
+        ...     }
+        ... )._is_homogeneous_type
+        False
+        """
+        # The "<" part of "<=" here is for empty DataFrame cases
+        return len({block.values.dtype for block in self._mgr.blocks}) <= 1
+
+    @property
+    def _can_fast_transpose(self) -> bool:
+        """
+        Can we transpose this DataFrame without creating any new array objects.
+        """
+        blocks = self._mgr.blocks
+        if len(blocks) != 1:
+            return False
+
+        dtype = blocks[0].dtype
+        # TODO(EA2D) special case would be unnecessary with 2D EAs
+        return not is_1d_only_ea_dtype(dtype)
+
+    @property
+    def _values(self) -> np.ndarray | DatetimeArray | TimedeltaArray | PeriodArray:
+        """
+        Analogue to ._values that may return a 2D ExtensionArray.
+        """
+        mgr = self._mgr
+
+        blocks = mgr.blocks
+        if len(blocks) != 1:
+            return ensure_wrapped_if_datetimelike(self.values)
+
+        arr = blocks[0].values
+        if arr.ndim == 1:
+            # non-2D ExtensionArray
+            return self.values
+
+        # more generally, whatever we allow in NDArrayBackedExtensionBlock
+        arr = cast("np.ndarray | DatetimeArray | TimedeltaArray | PeriodArray", arr)
+        return arr.T
+
+    # ----------------------------------------------------------------------
+    # Rendering Methods
+
+    def _repr_fits_vertical_(self) -> bool:
+        """
+        Check length against max_rows.
+        """
+        max_rows = get_option("display.max_rows")
+        return len(self) <= max_rows
+
+    def _repr_fits_horizontal_(self) -> bool:
+        """
+        Check if full repr fits in horizontal boundaries imposed by the display
+        options width and max_columns.
+        """
+        width, height = console.get_console_size()
+        max_columns = get_option("display.max_columns")
+        nb_columns = len(self.columns)
+
+        # exceed max columns
+        if (max_columns and nb_columns > max_columns) or (
+            width and nb_columns > (width // 2)
+        ):
+            return False
+
+        # used by repr_html under IPython notebook or scripts ignore terminal
+        # dims
+        if width is None or not console.in_interactive_session():
+            return True
+
+        if get_option("display.width") is not None or console.in_ipython_frontend():
+            # check at least the column row for excessive width
+            max_rows = 1
+        else:
+            max_rows = get_option("display.max_rows")
+
+        # when auto-detecting, so width=None and not in ipython front end
+        # check whether repr fits horizontal by actually checking
+        # the width of the rendered repr
+        buf = StringIO()
+
+        # only care about the stuff we'll actually print out
+        # and to_string on entire frame may be expensive
+        d = self
+
+        if max_rows is not None:  # unlimited rows
+            # min of two, where one may be None
+            d = d.iloc[: min(max_rows, len(d))]
+        else:
+            return True
+
+        d.to_string(buf=buf)
+        value = buf.getvalue()
+        repr_width = max(len(line) for line in value.split("\n"))
+
+        return repr_width < width
+
+    def _info_repr(self) -> bool:
+        """
+        True if the repr should show the info view.
+        """
+        info_repr_option = get_option("display.large_repr") == "info"
+        return info_repr_option and not (
+            self._repr_fits_horizontal_() and self._repr_fits_vertical_()
+        )
+
+    def __repr__(self) -> str:
+        """
+        Return a string representation for a particular DataFrame.
+        """
+        if self._info_repr():
+            buf = StringIO()
+            self.info(buf=buf)
+            return buf.getvalue()
+
+        repr_params = fmt.get_dataframe_repr_params()
+        return self.to_string(**repr_params)
+
+    def _repr_html_(self) -> str | None:
+        """
+        Return a html representation for a particular DataFrame.
+
+        Mainly for IPython notebook.
+        """
+        if self._info_repr():
+            buf = StringIO()
+            self.info(buf=buf)
+            # need to escape the <class>, should be the first line.
+            val = buf.getvalue().replace("<", r"&lt;", 1)
+            val = val.replace(">", r"&gt;", 1)
+            return f"<pre>{val}</pre>"
+
+        if get_option("display.notebook_repr_html"):
+            max_rows = get_option("display.max_rows")
+            min_rows = get_option("display.min_rows")
+            max_cols = get_option("display.max_columns")
+            show_dimensions = get_option("display.show_dimensions")
+            show_floats = get_option("display.float_format")
+
+            formatter = fmt.DataFrameFormatter(
+                self,
+                columns=None,
+                col_space=None,
+                na_rep="NaN",
+                formatters=None,
+                float_format=show_floats,
+                sparsify=None,
+                justify=None,
+                index_names=True,
+                header=True,
+                index=True,
+                bold_rows=True,
+                escape=True,
+                max_rows=max_rows,
+                min_rows=min_rows,
+                max_cols=max_cols,
+                show_dimensions=show_dimensions,
+                decimal=".",
+            )
+            return fmt.DataFrameRenderer(formatter).to_html(notebook=True)
+        else:
+            return None
+
+    @overload
+    def to_string(
+        self,
+        buf: None = ...,
+        *,
+        columns: Axes | None = ...,
+        col_space: int | list[int] | dict[Hashable, int] | None = ...,
+        header: bool | SequenceNotStr[str] = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: fmt.FormattersType | None = ...,
+        float_format: fmt.FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        justify: str | None = ...,
+        max_rows: int | None = ...,
+        max_cols: int | None = ...,
+        show_dimensions: bool = ...,
+        decimal: str = ...,
+        line_width: int | None = ...,
+        min_rows: int | None = ...,
+        max_colwidth: int | None = ...,
+        encoding: str | None = ...,
+    ) -> str: ...
+
+    @overload
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        columns: Axes | None = ...,
+        col_space: int | list[int] | dict[Hashable, int] | None = ...,
+        header: bool | SequenceNotStr[str] = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: fmt.FormattersType | None = ...,
+        float_format: fmt.FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        justify: str | None = ...,
+        max_rows: int | None = ...,
+        max_cols: int | None = ...,
+        show_dimensions: bool = ...,
+        decimal: str = ...,
+        line_width: int | None = ...,
+        min_rows: int | None = ...,
+        max_colwidth: int | None = ...,
+        encoding: str | None = ...,
+    ) -> None: ...
+
+    @Substitution(
+        header_type="bool or list of str",
+        header="Write out the column names. If a list of columns "
+        "is given, it is assumed to be aliases for the "
+        "column names",
+        col_space_type="int, list or dict of int",
+        col_space="The minimum width of each column. If a list of ints is given "
+        "every integers corresponds with one column. If a dict is given, the key "
+        "references the column, while the value defines the space to use.",
+    )
+    @Substitution(shared_params=fmt.common_docstring, returns=fmt.return_docstring)
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        columns: Axes | None = None,
+        col_space: int | list[int] | dict[Hashable, int] | None = None,
+        header: bool | SequenceNotStr[str] = True,
+        index: bool = True,
+        na_rep: str = "NaN",
+        formatters: fmt.FormattersType | None = None,
+        float_format: fmt.FloatFormatType | None = None,
+        sparsify: bool | None = None,
+        index_names: bool = True,
+        justify: str | None = None,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        show_dimensions: bool = False,
+        decimal: str = ".",
+        line_width: int | None = None,
+        min_rows: int | None = None,
+        max_colwidth: int | None = None,
+        encoding: str | None = None,
+    ) -> str | None:
+        """
+        Render a DataFrame to a console-friendly tabular output.
+        %(shared_params)s
+        line_width : int, optional
+            Width to wrap a line in characters.
+        min_rows : int, optional
+            The number of rows to display in the console in a truncated repr
+            (when number of rows is above `max_rows`).
+        max_colwidth : int, optional
+            Max width to truncate each column in characters. By default, no limit.
+        encoding : str, default "utf-8"
+            Set character encoding.
+        %(returns)s
+        See Also
+        --------
+        to_html : Convert DataFrame to HTML.
+
+        Examples
+        --------
+        >>> d = {"col1": [1, 2, 3], "col2": [4, 5, 6]}
+        >>> df = pd.DataFrame(d)
+        >>> print(df.to_string())
+           col1  col2
+        0     1     4
+        1     2     5
+        2     3     6
+        """
+        from pandas import option_context
+
+        with option_context("display.max_colwidth", max_colwidth):
+            formatter = fmt.DataFrameFormatter(
+                self,
+                columns=columns,
+                col_space=col_space,
+                na_rep=na_rep,
+                formatters=formatters,
+                float_format=float_format,
+                sparsify=sparsify,
+                justify=justify,
+                index_names=index_names,
+                header=header,
+                index=index,
+                min_rows=min_rows,
+                max_rows=max_rows,
+                max_cols=max_cols,
+                show_dimensions=show_dimensions,
+                decimal=decimal,
+            )
+            return fmt.DataFrameRenderer(formatter).to_string(
+                buf=buf,
+                encoding=encoding,
+                line_width=line_width,
+            )
+
+    def _get_values_for_csv(
+        self,
+        *,
+        float_format: FloatFormatType | None,
+        date_format: str | None,
+        decimal: str,
+        na_rep: str,
+        quoting,  # int csv.QUOTE_FOO from stdlib
+    ) -> DataFrame:
+        # helper used by to_csv
+        mgr = self._mgr.get_values_for_csv(
+            float_format=float_format,
+            date_format=date_format,
+            decimal=decimal,
+            na_rep=na_rep,
+            quoting=quoting,
+        )
+        return self._constructor_from_mgr(mgr, axes=mgr.axes)
+
+    # ----------------------------------------------------------------------
+
+    @property
+    def style(self) -> Styler:
+        """
+        Returns a Styler object.
+
+        Contains methods for building a styled HTML representation of the DataFrame.
+
+        See Also
+        --------
+        io.formats.style.Styler : Helps style a DataFrame or Series according to the
+            data with HTML and CSS.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3]})
+        >>> df.style  # doctest: +SKIP
+
+        Please see
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        # Raise AttributeError so that inspect works even if jinja2 is not installed.
+        has_jinja2 = import_optional_dependency("jinja2", errors="ignore")
+        if not has_jinja2:
+            raise AttributeError("The '.style' accessor requires jinja2")
+
+        from pandas.io.formats.style import Styler
+
+        return Styler(self)
+
+    _shared_docs["items"] = r"""
+        Iterate over (column name, Series) pairs.
+
+        Iterates over the DataFrame columns, returning a tuple with
+        the column name and the content as a Series.
+
+        Yields
+        ------
+        label : object
+            The column names for the DataFrame being iterated over.
+        content : Series
+            The column entries belonging to each label, as a Series.
+
+        See Also
+        --------
+        DataFrame.iterrows : Iterate over DataFrame rows as
+            (index, Series) pairs.
+        DataFrame.itertuples : Iterate over DataFrame rows as namedtuples
+            of the values.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({'species': ['bear', 'bear', 'marsupial'],
+        ...                   'population': [1864, 22000, 80000]},
+        ...                   index=['panda', 'polar', 'koala'])
+        >>> df
+                species   population
+        panda   bear      1864
+        polar   bear      22000
+        koala   marsupial 80000
+        >>> for label, content in df.items():
+        ...     print(f'label: {label}')
+        ...     print(f'content: {content}', sep='\n')
+        ...
+        label: species
+        content:
+        panda         bear
+        polar         bear
+        koala    marsupial
+        Name: species, dtype: object
+        label: population
+        content:
+        panda     1864
+        polar    22000
+        koala    80000
+        Name: population, dtype: int64
+        """
+
+    @Appender(_shared_docs["items"])
+    def items(self) -> Iterable[tuple[Hashable, Series]]:
+        for i, k in enumerate(self.columns):
+            yield k, self._ixs(i, axis=1)
+
+    def iterrows(self) -> Iterable[tuple[Hashable, Series]]:
+        """
+        Iterate over DataFrame rows as (index, Series) pairs.
+
+        Yields
+        ------
+        index : label or tuple of label
+            The index of the row. A tuple for a `MultiIndex`.
+        data : Series
+            The data of the row as a Series.
+
+        See Also
+        --------
+        DataFrame.itertuples : Iterate over DataFrame rows as namedtuples of the values.
+        DataFrame.items : Iterate over (column name, Series) pairs.
+
+        Notes
+        -----
+        1. Because ``iterrows`` returns a Series for each row,
+           it does **not** preserve dtypes across the rows (dtypes are
+           preserved across columns for DataFrames).
+
+           To preserve dtypes while iterating over the rows, it is better
+           to use :meth:`itertuples` which returns namedtuples of the values
+           and which is generally faster than ``iterrows``.
+
+        2. You should **never modify** something you are iterating over.
+           This is not guaranteed to work in all cases. Depending on the
+           data types, the iterator returns a copy and not a view, and writing
+           to it will have no effect.
+
+        Examples
+        --------
+
+        >>> df = pd.DataFrame([[1, 1.5]], columns=["int", "float"])
+        >>> row = next(df.iterrows())[1]
+        >>> row
+        int      1.0
+        float    1.5
+        Name: 0, dtype: float64
+        >>> print(row["int"].dtype)
+        float64
+        >>> print(df["int"].dtype)
+        int64
+        """
+        columns = self.columns
+        klass = self._constructor_sliced
+        for k, v in zip(self.index, self.values):
+            s = klass(v, index=columns, name=k).__finalize__(self)
+            if self._mgr.is_single_block:
+                s._mgr.add_references(self._mgr)
+            yield k, s
+
+    def itertuples(
+        self, index: bool = True, name: str | None = "Pandas"
+    ) -> Iterable[tuple[Any, ...]]:
+        """
+        Iterate over DataFrame rows as namedtuples.
+
+        Parameters
+        ----------
+        index : bool, default True
+            If True, return the index as the first element of the tuple.
+        name : str or None, default "Pandas"
+            The name of the returned namedtuples or None to return regular
+            tuples.
+
+        Returns
+        -------
+        iterator
+            An object to iterate over namedtuples for each row in the
+            DataFrame with the first field possibly being the index and
+            following fields being the column values.
+
+        See Also
+        --------
+        DataFrame.iterrows : Iterate over DataFrame rows as (index, Series)
+            pairs.
+        DataFrame.items : Iterate over (column name, Series) pairs.
+
+        Notes
+        -----
+        The column names will be renamed to positional names if they are
+        invalid Python identifiers, repeated, or start with an underscore.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"num_legs": [4, 2], "num_wings": [0, 2]}, index=["dog", "hawk"]
+        ... )
+        >>> df
+              num_legs  num_wings
+        dog          4          0
+        hawk         2          2
+        >>> for row in df.itertuples():
+        ...     print(row)
+        Pandas(Index='dog', num_legs=4, num_wings=0)
+        Pandas(Index='hawk', num_legs=2, num_wings=2)
+
+        By setting the `index` parameter to False we can remove the index
+        as the first element of the tuple:
+
+        >>> for row in df.itertuples(index=False):
+        ...     print(row)
+        Pandas(num_legs=4, num_wings=0)
+        Pandas(num_legs=2, num_wings=2)
+
+        With the `name` parameter set we set a custom name for the yielded
+        namedtuples:
+
+        >>> for row in df.itertuples(name="Animal"):
+        ...     print(row)
+        Animal(Index='dog', num_legs=4, num_wings=0)
+        Animal(Index='hawk', num_legs=2, num_wings=2)
+        """
+        arrays = []
+        fields = list(self.columns)
+        if index:
+            arrays.append(self.index)
+            fields.insert(0, "Index")
+
+        # use integer indexing because of possible duplicate column names
+        arrays.extend(self.iloc[:, k] for k in range(len(self.columns)))
+
+        if name is not None:
+            # https://github.com/python/mypy/issues/9046
+            # error: namedtuple() expects a string literal as the first argument
+            itertuple = collections.namedtuple(  # type: ignore[misc]
+                name, fields, rename=True
+            )
+            return map(itertuple._make, zip(*arrays))
+
+        # fallback to regular tuples
+        return zip(*arrays)
+
+    def __len__(self) -> int:
+        """
+        Returns length of info axis, but here we use the index.
+        """
+        return len(self.index)
+
+    @overload
+    def dot(self, other: Series) -> Series: ...
+
+    @overload
+    def dot(self, other: DataFrame | Index | ArrayLike) -> DataFrame: ...
+
+    def dot(self, other: AnyArrayLike | DataFrame) -> DataFrame | Series:
+        """
+        Compute the matrix multiplication between the DataFrame and other.
+
+        This method computes the matrix product between the DataFrame and the
+        values of an other Series, DataFrame or a numpy array.
+
+        It can also be called using ``self @ other``.
+
+        Parameters
+        ----------
+        other : Series, DataFrame or array-like
+            The other object to compute the matrix product with.
+
+        Returns
+        -------
+        Series or DataFrame
+            If other is a Series, return the matrix product between self and
+            other as a Series. If other is a DataFrame or a numpy.array, return
+            the matrix product of self and other in a DataFrame of a np.array.
+
+        See Also
+        --------
+        Series.dot: Similar method for Series.
+
+        Notes
+        -----
+        The dimensions of DataFrame and other must be compatible in order to
+        compute the matrix multiplication. In addition, the column names of
+        DataFrame and the index of other must contain the same values, as they
+        will be aligned prior to the multiplication.
+
+        The dot method for Series computes the inner product, instead of the
+        matrix product here.
+
+        Examples
+        --------
+        Here we multiply a DataFrame with a Series.
+
+        >>> df = pd.DataFrame([[0, 1, -2, -1], [1, 1, 1, 1]])
+        >>> s = pd.Series([1, 1, 2, 1])
+        >>> df.dot(s)
+        0    -4
+        1     5
+        dtype: int64
+
+        Here we multiply a DataFrame with another DataFrame.
+
+        >>> other = pd.DataFrame([[0, 1], [1, 2], [-1, -1], [2, 0]])
+        >>> df.dot(other)
+            0   1
+        0   1   4
+        1   2   2
+
+        Note that the dot method give the same result as @
+
+        >>> df @ other
+            0   1
+        0   1   4
+        1   2   2
+
+        The dot method works also if other is an np.array.
+
+        >>> arr = np.array([[0, 1], [1, 2], [-1, -1], [2, 0]])
+        >>> df.dot(arr)
+            0   1
+        0   1   4
+        1   2   2
+
+        Note how shuffling of the objects does not change the result.
+
+        >>> s2 = s.reindex([1, 0, 2, 3])
+        >>> df.dot(s2)
+        0    -4
+        1     5
+        dtype: int64
+        """
+        if isinstance(other, (Series, DataFrame)):
+            common = self.columns.union(other.index)
+            if len(common) > len(self.columns) or len(common) > len(other.index):
+                raise ValueError("matrices are not aligned")
+
+            left = self.reindex(columns=common)
+            right = other.reindex(index=common)
+            lvals = left.values
+            rvals = right._values
+        else:
+            left = self
+            lvals = self.values
+            rvals = np.asarray(other)
+            if lvals.shape[1] != rvals.shape[0]:
+                raise ValueError(
+                    f"Dot product shape mismatch, {lvals.shape} vs {rvals.shape}"
+                )
+
+        if isinstance(other, DataFrame):
+            common_type = find_common_type(list(self.dtypes) + list(other.dtypes))
+            return self._constructor(
+                np.dot(lvals, rvals),
+                index=left.index,
+                columns=other.columns,
+                copy=False,
+                dtype=common_type,
+            )
+        elif isinstance(other, Series):
+            common_type = find_common_type(list(self.dtypes) + [other.dtypes])
+            return self._constructor_sliced(
+                np.dot(lvals, rvals), index=left.index, copy=False, dtype=common_type
+            )
+        elif isinstance(rvals, (np.ndarray, Index)):
+            result = np.dot(lvals, rvals)
+            if result.ndim == 2:
+                return self._constructor(result, index=left.index, copy=False)
+            else:
+                return self._constructor_sliced(result, index=left.index, copy=False)
+        else:  # pragma: no cover
+            raise TypeError(f"unsupported type: {type(other)}")
+
+    @overload
+    def __matmul__(self, other: Series) -> Series: ...
+
+    @overload
+    def __matmul__(self, other: AnyArrayLike | DataFrame) -> DataFrame | Series: ...
+
+    def __matmul__(self, other: AnyArrayLike | DataFrame) -> DataFrame | Series:
+        """
+        Matrix multiplication using binary `@` operator.
+        """
+        return self.dot(other)
+
+    def __rmatmul__(self, other) -> DataFrame:
+        """
+        Matrix multiplication using binary `@` operator.
+        """
+        try:
+            return self.T.dot(np.transpose(other)).T
+        except ValueError as err:
+            if "shape mismatch" not in str(err):
+                raise
+            # GH#21581 give exception message for original shapes
+            msg = f"shapes {np.shape(other)} and {self.shape} not aligned"
+            raise ValueError(msg) from err
+
+    # ----------------------------------------------------------------------
+    # IO methods (to / from other formats)
+
+    @classmethod
+    def from_dict(
+        cls,
+        data: dict,
+        orient: FromDictOrient = "columns",
+        dtype: Dtype | None = None,
+        columns: Axes | None = None,
+    ) -> DataFrame:
+        """
+        Construct DataFrame from dict of array-like or dicts.
+
+        Creates DataFrame object from dictionary by columns or by index
+        allowing dtype specification.
+
+        Parameters
+        ----------
+        data : dict
+            Of the form {field : array-like} or {field : dict}.
+        orient : {'columns', 'index', 'tight'}, default 'columns'
+            The "orientation" of the data. If the keys of the passed dict
+            should be the columns of the resulting DataFrame, pass 'columns'
+            (default). Otherwise if the keys should be rows, pass 'index'.
+            If 'tight', assume a dict with keys ['index', 'columns', 'data',
+            'index_names', 'column_names'].
+
+            .. versionadded:: 1.4.0
+               'tight' as an allowed value for the ``orient`` argument
+
+        dtype : dtype, default None
+            Data type to force after DataFrame construction, otherwise infer.
+        columns : list, default None
+            Column labels to use when ``orient='index'``. Raises a ValueError
+            if used with ``orient='columns'`` or ``orient='tight'``.
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        DataFrame.from_records : DataFrame from structured ndarray, sequence
+            of tuples or dicts, or DataFrame.
+        DataFrame : DataFrame object creation using constructor.
+        DataFrame.to_dict : Convert the DataFrame to a dictionary.
+
+        Examples
+        --------
+        By default the keys of the dict become the DataFrame columns:
+
+        >>> data = {"col_1": [3, 2, 1, 0], "col_2": ["a", "b", "c", "d"]}
+        >>> pd.DataFrame.from_dict(data)
+           col_1 col_2
+        0      3     a
+        1      2     b
+        2      1     c
+        3      0     d
+
+        Specify ``orient='index'`` to create the DataFrame using dictionary
+        keys as rows:
+
+        >>> data = {"row_1": [3, 2, 1, 0], "row_2": ["a", "b", "c", "d"]}
+        >>> pd.DataFrame.from_dict(data, orient="index")
+               0  1  2  3
+        row_1  3  2  1  0
+        row_2  a  b  c  d
+
+        When using the 'index' orientation, the column names can be
+        specified manually:
+
+        >>> pd.DataFrame.from_dict(data, orient="index", columns=["A", "B", "C", "D"])
+               A  B  C  D
+        row_1  3  2  1  0
+        row_2  a  b  c  d
+
+        Specify ``orient='tight'`` to create the DataFrame using a 'tight'
+        format:
+
+        >>> data = {
+        ...     "index": [("a", "b"), ("a", "c")],
+        ...     "columns": [("x", 1), ("y", 2)],
+        ...     "data": [[1, 3], [2, 4]],
+        ...     "index_names": ["n1", "n2"],
+        ...     "column_names": ["z1", "z2"],
+        ... }
+        >>> pd.DataFrame.from_dict(data, orient="tight")
+        z1     x  y
+        z2     1  2
+        n1 n2
+        a  b   1  3
+           c   2  4
+        """
+        index: list | Index | None = None
+        orient = orient.lower()  # type: ignore[assignment]
+        if orient == "index":
+            if len(data) > 0:
+                # TODO speed up Series case
+                if isinstance(next(iter(data.values())), (Series, dict)):
+                    data = _from_nested_dict(data)
+                else:
+                    index = list(data.keys())
+                    # error: Incompatible types in assignment (expression has type
+                    # "List[Any]", variable has type "Dict[Any, Any]")
+                    data = list(data.values())  # type: ignore[assignment]
+        elif orient in ("columns", "tight"):
+            if columns is not None:
+                raise ValueError(f"cannot use columns parameter with orient='{orient}'")
+        else:  # pragma: no cover
+            raise ValueError(
+                f"Expected 'index', 'columns' or 'tight' for orient parameter. "
+                f"Got '{orient}' instead"
+            )
+
+        if orient != "tight":
+            return cls(data, index=index, columns=columns, dtype=dtype)
+        else:
+            realdata = data["data"]
+
+            def create_index(indexlist, namelist) -> Index:
+                index: Index
+                if len(namelist) > 1:
+                    index = MultiIndex.from_tuples(indexlist, names=namelist)
+                else:
+                    index = Index(indexlist, name=namelist[0])
+                return index
+
+            index = create_index(data["index"], data["index_names"])
+            columns = create_index(data["columns"], data["column_names"])
+            return cls(realdata, index=index, columns=columns, dtype=dtype)
+
+    def to_numpy(
+        self,
+        dtype: npt.DTypeLike | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        """
+        Convert the DataFrame to a NumPy array.
+
+        By default, the dtype of the returned array will be the common NumPy
+        dtype of all types in the DataFrame. For example, if the dtypes are
+        ``float16`` and ``float32``, the results dtype will be ``float32``.
+        This may require copying data and coercing values, which may be
+        expensive.
+
+        Parameters
+        ----------
+        dtype : str or numpy.dtype, optional
+            The dtype to pass to :meth:`numpy.asarray`.
+        copy : bool, default False
+            Whether to ensure that the returned value is not a view on
+            another array. Note that ``copy=False`` does not *ensure* that
+            ``to_numpy()`` is no-copy. Rather, ``copy=True`` ensure that
+            a copy is made, even if not strictly necessary.
+        na_value : Any, optional
+            The value to use for missing values. The default value depends
+            on `dtype` and the dtypes of the DataFrame columns.
+
+        Returns
+        -------
+        numpy.ndarray
+            The NumPy array representing the values in the DataFrame.
+
+        See Also
+        --------
+        Series.to_numpy : Similar method for Series.
+
+        Examples
+        --------
+        >>> pd.DataFrame({"A": [1, 2], "B": [3, 4]}).to_numpy()
+        array([[1, 3],
+               [2, 4]])
+
+        With heterogeneous data, the lowest common type will have to
+        be used.
+
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3.0, 4.5]})
+        >>> df.to_numpy()
+        array([[1. , 3. ],
+               [2. , 4.5]])
+
+        For a mix of numeric and non-numeric types, the output array will
+        have object dtype.
+
+        >>> df["C"] = pd.date_range("2000", periods=2)
+        >>> df.to_numpy()
+        array([[1, 3.0, Timestamp('2000-01-01 00:00:00')],
+               [2, 4.5, Timestamp('2000-01-02 00:00:00')]], dtype=object)
+        """
+        if dtype is not None:
+            dtype = np.dtype(dtype)
+        result = self._mgr.as_array(dtype=dtype, copy=copy, na_value=na_value)
+        if result.dtype is not dtype:
+            result = np.asarray(result, dtype=dtype)
+
+        return result
+
+    @overload
+    def to_dict(
+        self,
+        orient: Literal["dict", "list", "series", "split", "tight", "index"] = ...,
+        *,
+        into: type[MutableMappingT] | MutableMappingT,
+        index: bool = ...,
+    ) -> MutableMappingT: ...
+
+    @overload
+    def to_dict(
+        self,
+        orient: Literal["records"],
+        *,
+        into: type[MutableMappingT] | MutableMappingT,
+        index: bool = ...,
+    ) -> list[MutableMappingT]: ...
+
+    @overload
+    def to_dict(
+        self,
+        orient: Literal["dict", "list", "series", "split", "tight", "index"] = ...,
+        *,
+        into: type[dict] = ...,
+        index: bool = ...,
+    ) -> dict: ...
+
+    @overload
+    def to_dict(
+        self,
+        orient: Literal["records"],
+        *,
+        into: type[dict] = ...,
+        index: bool = ...,
+    ) -> list[dict]: ...
+
+    # error: Incompatible default for argument "into" (default has type "type
+    # [dict[Any, Any]]", argument has type "type[MutableMappingT] | MutableMappingT")
+    def to_dict(
+        self,
+        orient: Literal[
+            "dict", "list", "series", "split", "tight", "records", "index"
+        ] = "dict",
+        *,
+        into: type[MutableMappingT] | MutableMappingT = dict,  # type: ignore[assignment]
+        index: bool = True,
+    ) -> MutableMappingT | list[MutableMappingT]:
+        """
+        Convert the DataFrame to a dictionary.
+
+        The type of the key-value pairs can be customized with the parameters
+        (see below).
+
+        Parameters
+        ----------
+        orient : str {'dict', 'list', 'series', 'split', 'tight', 'records', 'index'}
+            Determines the type of the values of the dictionary.
+
+            - 'dict' (default) : dict like {column -> {index -> value}}
+            - 'list' : dict like {column -> [values]}
+            - 'series' : dict like {column -> Series(values)}
+            - 'split' : dict like
+              {'index' -> [index], 'columns' -> [columns], 'data' -> [values]}
+            - 'tight' : dict like
+              {'index' -> [index], 'columns' -> [columns], 'data' -> [values],
+              'index_names' -> [index.names], 'column_names' -> [column.names]}
+            - 'records' : list like
+              [{column -> value}, ... , {column -> value}]
+            - 'index' : dict like {index -> {column -> value}}
+
+            .. versionadded:: 1.4.0
+                'tight' as an allowed value for the ``orient`` argument
+
+        into : class, default dict
+            The collections.abc.MutableMapping subclass used for all Mappings
+            in the return value.  Can be the actual class or an empty
+            instance of the mapping type you want.  If you want a
+            collections.defaultdict, you must pass it initialized.
+
+        index : bool, default True
+            Whether to include the index item (and index_names item if `orient`
+            is 'tight') in the returned dictionary. Can only be ``False``
+            when `orient` is 'split' or 'tight'. Note that when `orient` is
+            'records', this parameter does not take effect (index item always
+            not included).
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        dict, list or collections.abc.MutableMapping
+            Return a collections.abc.MutableMapping object representing the
+            DataFrame. The resulting transformation depends on the `orient`
+            parameter.
+
+        See Also
+        --------
+        DataFrame.from_dict: Create a DataFrame from a dictionary.
+        DataFrame.to_json: Convert a DataFrame to JSON format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"col1": [1, 2], "col2": [0.5, 0.75]}, index=["row1", "row2"]
+        ... )
+        >>> df
+              col1  col2
+        row1     1  0.50
+        row2     2  0.75
+        >>> df.to_dict()
+        {'col1': {'row1': 1, 'row2': 2}, 'col2': {'row1': 0.5, 'row2': 0.75}}
+
+        You can specify the return orientation.
+
+        >>> df.to_dict("series")
+        {'col1': row1    1
+                 row2    2
+        Name: col1, dtype: int64,
+        'col2': row1    0.50
+                row2    0.75
+        Name: col2, dtype: float64}
+
+        >>> df.to_dict("split")
+        {'index': ['row1', 'row2'], 'columns': ['col1', 'col2'],
+         'data': [[1, 0.5], [2, 0.75]]}
+
+        >>> df.to_dict("records")
+        [{'col1': 1, 'col2': 0.5}, {'col1': 2, 'col2': 0.75}]
+
+        >>> df.to_dict("index")
+        {'row1': {'col1': 1, 'col2': 0.5}, 'row2': {'col1': 2, 'col2': 0.75}}
+
+        >>> df.to_dict("tight")
+        {'index': ['row1', 'row2'], 'columns': ['col1', 'col2'],
+         'data': [[1, 0.5], [2, 0.75]], 'index_names': [None], 'column_names': [None]}
+
+        You can also specify the mapping type.
+
+        >>> from collections import OrderedDict, defaultdict
+        >>> df.to_dict(into=OrderedDict)
+        OrderedDict([('col1', OrderedDict([('row1', 1), ('row2', 2)])),
+                     ('col2', OrderedDict([('row1', 0.5), ('row2', 0.75)]))])
+
+        If you want a `defaultdict`, you need to initialize it:
+
+        >>> dd = defaultdict(list)
+        >>> df.to_dict("records", into=dd)
+        [defaultdict(<class 'list'>, {'col1': 1, 'col2': 0.5}),
+         defaultdict(<class 'list'>, {'col1': 2, 'col2': 0.75})]
+        """
+        from pandas.core.methods.to_dict import to_dict
+
+        return to_dict(self, orient, into=into, index=index)
+
+    @classmethod
+    def from_records(
+        cls,
+        data,
+        index=None,
+        exclude=None,
+        columns=None,
+        coerce_float: bool = False,
+        nrows: int | None = None,
+    ) -> DataFrame:
+        """
+        Convert structured or record ndarray to DataFrame.
+
+        Creates a DataFrame object from a structured ndarray, or sequence of
+        tuples or dicts.
+
+        Parameters
+        ----------
+        data : structured ndarray, sequence of tuples or dicts
+            Structured input data.
+        index : str, list of fields, array-like
+            Field of array to use as the index, alternately a specific set of
+            input labels to use.
+        exclude : sequence, default None
+            Columns or fields to exclude.
+        columns : sequence, default None
+            Column names to use. If the passed data do not have names
+            associated with them, this argument provides names for the
+            columns. Otherwise, this argument indicates the order of the columns
+            in the result (any names not found in the data will become all-NA
+            columns) and limits the data to these columns if not all column names
+            are provided.
+        coerce_float : bool, default False
+            Attempt to convert values of non-string, non-numeric objects (like
+            decimal.Decimal) to floating point, useful for SQL result sets.
+        nrows : int, default None
+            Number of rows to read if data is an iterator.
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        DataFrame.from_dict : DataFrame from dict of array-like or dicts.
+        DataFrame : DataFrame object creation using constructor.
+
+        Examples
+        --------
+        Data can be provided as a structured ndarray:
+
+        >>> data = np.array(
+        ...     [(3, "a"), (2, "b"), (1, "c"), (0, "d")],
+        ...     dtype=[("col_1", "i4"), ("col_2", "U1")],
+        ... )
+        >>> pd.DataFrame.from_records(data)
+           col_1 col_2
+        0      3     a
+        1      2     b
+        2      1     c
+        3      0     d
+
+        Data can be provided as a list of dicts:
+
+        >>> data = [
+        ...     {"col_1": 3, "col_2": "a"},
+        ...     {"col_1": 2, "col_2": "b"},
+        ...     {"col_1": 1, "col_2": "c"},
+        ...     {"col_1": 0, "col_2": "d"},
+        ... ]
+        >>> pd.DataFrame.from_records(data)
+           col_1 col_2
+        0      3     a
+        1      2     b
+        2      1     c
+        3      0     d
+
+        Data can be provided as a list of tuples with corresponding columns:
+
+        >>> data = [(3, "a"), (2, "b"), (1, "c"), (0, "d")]
+        >>> pd.DataFrame.from_records(data, columns=["col_1", "col_2"])
+           col_1 col_2
+        0      3     a
+        1      2     b
+        2      1     c
+        3      0     d
+        """
+        if isinstance(data, DataFrame):
+            raise TypeError(
+                "Passing a DataFrame to DataFrame.from_records is not supported. Use "
+                "set_index and/or drop to modify the DataFrame instead.",
+            )
+
+        result_index = None
+
+        # Make a copy of the input columns so we can modify it
+        if columns is not None:
+            columns = ensure_index(columns)
+
+        def maybe_reorder(
+            arrays: list[ArrayLike], arr_columns: Index, columns: Index, index
+        ) -> tuple[list[ArrayLike], Index, Index | None]:
+            """
+            If our desired 'columns' do not match the data's pre-existing 'arr_columns',
+            we re-order our arrays.  This is like a preemptive (cheap) reindex.
+            """
+            if len(arrays):
+                length = len(arrays[0])
+            else:
+                length = 0
+
+            result_index = None
+            if len(arrays) == 0 and index is None and length == 0:
+                result_index = default_index(0)
+
+            arrays, arr_columns = reorder_arrays(arrays, arr_columns, columns, length)
+            return arrays, arr_columns, result_index
+
+        if is_iterator(data):
+            if nrows == 0:
+                return cls()
+
+            try:
+                first_row = next(data)
+            except StopIteration:
+                return cls(index=index, columns=columns)
+
+            dtype = None
+            if hasattr(first_row, "dtype") and first_row.dtype.names:
+                dtype = first_row.dtype
+
+            values = [first_row]
+
+            if nrows is None:
+                values += data
+            else:
+                values.extend(itertools.islice(data, nrows - 1))
+
+            if dtype is not None:
+                data = np.array(values, dtype=dtype)
+            else:
+                data = values
+
+        if isinstance(data, dict):
+            if columns is None:
+                columns = arr_columns = ensure_index(sorted(data))
+                arrays = [data[k] for k in columns]
+            else:
+                arrays = []
+                arr_columns_list = []
+                for k, v in data.items():
+                    if k in columns:
+                        arr_columns_list.append(k)
+                        arrays.append(v)
+
+                arr_columns = Index(arr_columns_list)
+                arrays, arr_columns, result_index = maybe_reorder(
+                    arrays, arr_columns, columns, index
+                )
+
+        elif isinstance(data, np.ndarray):
+            arrays, columns = to_arrays(data, columns)
+            arr_columns = columns
+        else:
+            arrays, arr_columns = to_arrays(data, columns)
+            if coerce_float:
+                for i, arr in enumerate(arrays):
+                    if arr.dtype == object:
+                        # error: Argument 1 to "maybe_convert_objects" has
+                        # incompatible type "Union[ExtensionArray, ndarray]";
+                        # expected "ndarray"
+                        arrays[i] = lib.maybe_convert_objects(
+                            arr,  # type: ignore[arg-type]
+                            try_float=True,
+                        )
+
+            arr_columns = ensure_index(arr_columns)
+            if columns is None:
+                columns = arr_columns
+            else:
+                arrays, arr_columns, result_index = maybe_reorder(
+                    arrays, arr_columns, columns, index
+                )
+
+        if exclude is None:
+            exclude = set()
+        else:
+            exclude = set(exclude)
+
+        if index is not None:
+            if isinstance(index, str) or not hasattr(index, "__iter__"):
+                i = columns.get_loc(index)
+                exclude.add(index)
+                if len(arrays) > 0:
+                    result_index = Index(arrays[i], name=index)
+                else:
+                    result_index = Index([], name=index)
+            else:
+                try:
+                    index_data = [arrays[arr_columns.get_loc(field)] for field in index]
+                except (KeyError, TypeError):
+                    # raised by get_loc, see GH#29258
+                    result_index = index
+                else:
+                    result_index = ensure_index_from_sequences(index_data, names=index)
+                    exclude.update(index)
+
+        if any(exclude):
+            arr_exclude = (x for x in exclude if x in arr_columns)
+            to_remove = {arr_columns.get_loc(col) for col in arr_exclude}  # pyright: ignore[reportUnhashable]
+            arrays = [v for i, v in enumerate(arrays) if i not in to_remove]
+
+            columns = columns.drop(exclude)
+
+        mgr = arrays_to_mgr(arrays, columns, result_index)
+        df = DataFrame._from_mgr(mgr, axes=mgr.axes)
+        if cls is not DataFrame:
+            return cls(df, copy=False)
+        return df
+
+    def to_records(
+        self, index: bool = True, column_dtypes=None, index_dtypes=None
+    ) -> np.rec.recarray:
+        """
+        Convert DataFrame to a NumPy record array.
+
+        Index will be included as the first field of the record array if
+        requested.
+
+        Parameters
+        ----------
+        index : bool, default True
+            Include index in resulting record array, stored in 'index'
+            field or using the index label, if set.
+        column_dtypes : str, type, dict, default None
+            If a string or type, the data type to store all columns. If
+            a dictionary, a mapping of column names and indices (zero-indexed)
+            to specific data types.
+        index_dtypes : str, type, dict, default None
+            If a string or type, the data type to store all index levels. If
+            a dictionary, a mapping of index level names and indices
+            (zero-indexed) to specific data types.
+
+            This mapping is applied only if `index=True`.
+
+        Returns
+        -------
+        numpy.rec.recarray
+            NumPy ndarray with the DataFrame labels as fields and each row
+            of the DataFrame as entries.
+
+        See Also
+        --------
+        DataFrame.from_records: Convert structured or record ndarray
+            to DataFrame.
+        numpy.rec.recarray: An ndarray that allows field access using
+            attributes, analogous to typed columns in a
+            spreadsheet.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [0.5, 0.75]}, index=["a", "b"])
+        >>> df
+           A     B
+        a  1  0.50
+        b  2  0.75
+        >>> df.to_records()
+        rec.array([('a', 1, 0.5 ), ('b', 2, 0.75)],
+                  dtype=[('index', 'O'), ('A', '<i8'), ('B', '<f8')])
+
+        If the DataFrame index has no label then the recarray field name
+        is set to 'index'. If the index has a label then this is used as the
+        field name:
+
+        >>> df.index = df.index.rename("I")
+        >>> df.to_records()
+        rec.array([('a', 1, 0.5 ), ('b', 2, 0.75)],
+                  dtype=[('I', 'O'), ('A', '<i8'), ('B', '<f8')])
+
+        The index can be excluded from the record array:
+
+        >>> df.to_records(index=False)
+        rec.array([(1, 0.5 ), (2, 0.75)],
+                  dtype=[('A', '<i8'), ('B', '<f8')])
+
+        Data types can be specified for the columns:
+
+        >>> df.to_records(column_dtypes={"A": "int32"})
+        rec.array([('a', 1, 0.5 ), ('b', 2, 0.75)],
+                  dtype=[('I', 'O'), ('A', '<i4'), ('B', '<f8')])
+
+        As well as for the index:
+
+        >>> df.to_records(index_dtypes="<S2")
+        rec.array([(b'a', 1, 0.5 ), (b'b', 2, 0.75)],
+                  dtype=[('I', 'S2'), ('A', '<i8'), ('B', '<f8')])
+
+        >>> index_dtypes = f"<S{df.index.str.len().max()}"
+        >>> df.to_records(index_dtypes=index_dtypes)
+        rec.array([(b'a', 1, 0.5 ), (b'b', 2, 0.75)],
+                  dtype=[('I', 'S1'), ('A', '<i8'), ('B', '<f8')])
+        """
+        if index:
+            ix_vals = [
+                np.asarray(self.index.get_level_values(i))
+                for i in range(self.index.nlevels)
+            ]
+
+            arrays = ix_vals + [
+                np.asarray(self.iloc[:, i]) for i in range(len(self.columns))
+            ]
+
+            index_names = list(self.index.names)
+
+            if isinstance(self.index, MultiIndex):
+                index_names = com.fill_missing_names(index_names)
+            elif index_names[0] is None:
+                index_names = ["index"]
+
+            names = [str(name) for name in itertools.chain(index_names, self.columns)]
+        else:
+            arrays = [np.asarray(self.iloc[:, i]) for i in range(len(self.columns))]
+            names = [str(c) for c in self.columns]
+            index_names = []
+
+        index_len = len(index_names)
+        formats = []
+
+        for i, v in enumerate(arrays):
+            index_int = i
+
+            # When the names and arrays are collected, we
+            # first collect those in the DataFrame's index,
+            # followed by those in its columns.
+            #
+            # Thus, the total length of the array is:
+            # len(index_names) + len(DataFrame.columns).
+            #
+            # This check allows us to see whether we are
+            # handling a name / array in the index or column.
+            if index_int < index_len:
+                dtype_mapping = index_dtypes
+                name = index_names[index_int]
+            else:
+                index_int -= index_len
+                dtype_mapping = column_dtypes
+                name = self.columns[index_int]
+
+            # We have a dictionary, so we get the data type
+            # associated with the index or column (which can
+            # be denoted by its name in the DataFrame or its
+            # position in DataFrame's array of indices or
+            # columns, whichever is applicable.
+            if is_dict_like(dtype_mapping):
+                if name in dtype_mapping:
+                    dtype_mapping = dtype_mapping[name]
+                elif index_int in dtype_mapping:
+                    dtype_mapping = dtype_mapping[index_int]
+                else:
+                    dtype_mapping = None
+
+            # If no mapping can be found, use the array's
+            # dtype attribute for formatting.
+            #
+            # A valid dtype must either be a type or
+            # string naming a type.
+            if dtype_mapping is None:
+                formats.append(v.dtype)
+            elif isinstance(dtype_mapping, (type, np.dtype, str)):
+                # error: Argument 1 to "append" of "list" has incompatible
+                # type "Union[type, dtype[Any], str]"; expected "dtype[Any]"
+                formats.append(dtype_mapping)  # type: ignore[arg-type]
+            else:
+                element = "row" if i < index_len else "column"
+                msg = f"Invalid dtype {dtype_mapping} specified for {element} {name}"
+                raise ValueError(msg)
+
+        return np.rec.fromarrays(arrays, dtype={"names": names, "formats": formats})
+
+    @classmethod
+    def _from_arrays(
+        cls,
+        arrays,
+        columns,
+        index,
+        dtype: Dtype | None = None,
+        verify_integrity: bool = True,
+    ) -> Self:
+        """
+        Create DataFrame from a list of arrays corresponding to the columns.
+
+        Parameters
+        ----------
+        arrays : list-like of arrays
+            Each array in the list corresponds to one column, in order.
+        columns : list-like, Index
+            The column names for the resulting DataFrame.
+        index : list-like, Index
+            The rows labels for the resulting DataFrame.
+        dtype : dtype, optional
+            Optional dtype to enforce for all arrays.
+        verify_integrity : bool, default True
+            Validate and homogenize all input. If set to False, it is assumed
+            that all elements of `arrays` are actual arrays how they will be
+            stored in a block (numpy ndarray or ExtensionArray), have the same
+            length as and are aligned with the index, and that `columns` and
+            `index` are ensured to be an Index object.
+
+        Returns
+        -------
+        DataFrame
+        """
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        columns = ensure_index(columns)
+        if len(columns) != len(arrays):
+            raise ValueError("len(columns) must match len(arrays)")
+        mgr = arrays_to_mgr(
+            arrays,
+            columns,
+            index,
+            dtype=dtype,
+            verify_integrity=verify_integrity,
+        )
+        return cls._from_mgr(mgr, axes=mgr.axes)
+
+    @doc(
+        storage_options=_shared_docs["storage_options"],
+        compression_options=_shared_docs["compression_options"] % "path",
+    )
+    def to_stata(
+        self,
+        path: FilePath | WriteBuffer[bytes],
+        *,
+        convert_dates: dict[Hashable, str] | None = None,
+        write_index: bool = True,
+        byteorder: ToStataByteorder | None = None,
+        time_stamp: datetime.datetime | None = None,
+        data_label: str | None = None,
+        variable_labels: dict[Hashable, str] | None = None,
+        version: int | None = 114,
+        convert_strl: Sequence[Hashable] | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+        value_labels: dict[Hashable, dict[float, str]] | None = None,
+    ) -> None:
+        """
+        Export DataFrame object to Stata dta format.
+
+        Writes the DataFrame to a Stata dataset file.
+        "dta" files contain a Stata dataset.
+
+        Parameters
+        ----------
+        path : str, path object, or buffer
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a binary ``write()`` function.
+
+        convert_dates : dict
+            Dictionary mapping columns containing datetime types to stata
+            internal format to use when writing the dates. Options are 'tc',
+            'td', 'tm', 'tw', 'th', 'tq', 'ty'. Column can be either an integer
+            or a name. Datetime columns that do not have a conversion type
+            specified will be converted to 'tc'. Raises NotImplementedError if
+            a datetime column has timezone information.
+        write_index : bool
+            Write the index to Stata dataset.
+        byteorder : str
+            Can be ">", "<", "little", or "big". default is `sys.byteorder`.
+        time_stamp : datetime
+            A datetime to use as file creation date.  Default is the current
+            time.
+        data_label : str, optional
+            A label for the data set.  Must be 80 characters or smaller.
+        variable_labels : dict
+            Dictionary containing columns as keys and variable labels as
+            values. Each label must be 80 characters or smaller.
+        version : {{114, 117, 118, 119, None}}, default 114
+            Version to use in the output dta file. Set to None to let pandas
+            decide between 118 or 119 formats depending on the number of
+            columns in the frame. Version 114 can be read by Stata 10 and
+            later. Version 117 can be read by Stata 13 or later. Version 118
+            is supported in Stata 14 and later. Version 119 is supported in
+            Stata 15 and later. Version 114 limits string variables to 244
+            characters or fewer while versions 117 and later allow strings
+            with lengths up to 2,000,000 characters. Versions 118 and 119
+            support Unicode characters, and version 119 supports more than
+            32,767 variables.
+
+            Version 119 should usually only be used when the number of
+            variables exceeds the capacity of dta format 118. Exporting
+            smaller datasets in format 119 may have unintended consequences,
+            and, as of November 2020, Stata SE cannot read version 119 files.
+
+        convert_strl : list, optional
+            List of column names to convert to string columns to Stata StrL
+            format. Only available if version is 117.  Storing strings in the
+            StrL format can produce smaller dta files if strings have more than
+            8 characters and values are repeated.
+        {compression_options}
+
+            .. versionchanged:: 1.4.0 Zstandard support.
+
+        {storage_options}
+
+        value_labels : dict of dicts
+            Dictionary containing columns as keys and dictionaries of column value
+            to labels as values. Labels for a single variable must be 32,000
+            characters or smaller.
+
+            .. versionadded:: 1.4.0
+
+        Raises
+        ------
+        NotImplementedError
+            * If datetimes contain timezone information
+            * Column dtype is not representable in Stata
+        ValueError
+            * Columns listed in convert_dates are neither datetime64[ns]
+              or datetime.datetime
+            * Column listed in convert_dates is not in DataFrame
+            * Categorical label contains more than 32,000 characters
+
+        See Also
+        --------
+        read_stata : Import Stata data files.
+        io.stata.StataWriter : Low-level writer for Stata data files.
+        io.stata.StataWriter117 : Low-level writer for version 117 files.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [["falcon", 350], ["parrot", 18]], columns=["animal", "parrot"]
+        ... )
+        >>> df.to_stata("animals.dta")  # doctest: +SKIP
+        """
+        if version not in (114, 117, 118, 119, None):
+            raise ValueError("Only formats 114, 117, 118 and 119 are supported.")
+        if version == 114:
+            if convert_strl is not None:
+                raise ValueError("strl is not supported in format 114")
+            from pandas.io.stata import StataWriter as statawriter
+        elif version == 117:
+            # Incompatible import of "statawriter" (imported name has type
+            # "Type[StataWriter117]", local name has type "Type[StataWriter]")
+            from pandas.io.stata import (  # type: ignore[assignment]
+                StataWriter117 as statawriter,
+            )
+        else:  # versions 118 and 119
+            # Incompatible import of "statawriter" (imported name has type
+            # "Type[StataWriter117]", local name has type "Type[StataWriter]")
+            from pandas.io.stata import (  # type: ignore[assignment]
+                StataWriterUTF8 as statawriter,
+            )
+
+        kwargs: dict[str, Any] = {}
+        if version is None or version >= 117:
+            # strl conversion is only supported >= 117
+            kwargs["convert_strl"] = convert_strl
+        if version is None or version >= 118:
+            # Specifying the version is only supported for UTF8 (118 or 119)
+            kwargs["version"] = version
+
+        writer = statawriter(
+            path,
+            self,
+            convert_dates=convert_dates,
+            byteorder=byteorder,
+            time_stamp=time_stamp,
+            data_label=data_label,
+            write_index=write_index,
+            variable_labels=variable_labels,
+            compression=compression,
+            storage_options=storage_options,
+            value_labels=value_labels,
+            **kwargs,
+        )
+        writer.write_file()
+
+    def to_feather(self, path: FilePath | WriteBuffer[bytes], **kwargs) -> None:
+        """
+        Write a DataFrame to the binary Feather format.
+
+        Parameters
+        ----------
+        path : str, path object, file-like object
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a binary ``write()`` function. If a string or a path,
+            it will be used as Root Directory path when writing a partitioned dataset.
+        **kwargs :
+            Additional keywords passed to :func:`pyarrow.feather.write_feather`.
+            This includes the `compression`, `compression_level`, `chunksize`
+            and `version` keywords.
+
+        See Also
+        --------
+        DataFrame.to_parquet : Write a DataFrame to the binary parquet format.
+        DataFrame.to_excel : Write object to an Excel sheet.
+        DataFrame.to_sql : Write to a sql table.
+        DataFrame.to_csv : Write a csv file.
+        DataFrame.to_json : Convert the object to a JSON string.
+        DataFrame.to_html : Render a DataFrame as an HTML table.
+        DataFrame.to_string : Convert DataFrame to a string.
+
+        Notes
+        -----
+        This function writes the dataframe as a `feather file
+        <https://arrow.apache.org/docs/python/feather.html>`_. Requires a default
+        index. For saving the DataFrame with your custom index use a method that
+        supports custom indices e.g. `to_parquet`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2, 3], [4, 5, 6]])
+        >>> df.to_feather("file.feather")  # doctest: +SKIP
+        """
+        from pandas.io.feather_format import to_feather
+
+        to_feather(self, path, **kwargs)
+
+    @overload
+    def to_markdown(
+        self,
+        buf: None = ...,
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> str: ...
+
+    @overload
+    def to_markdown(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> None: ...
+
+    @overload
+    def to_markdown(
+        self,
+        buf: FilePath | WriteBuffer[str] | None,
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> str | None: ...
+
+    def to_markdown(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        mode: str = "wt",
+        index: bool = True,
+        storage_options: StorageOptions | None = None,
+        **kwargs,
+    ) -> str | None:
+        """
+        Print DataFrame in Markdown-friendly format.
+
+        Parameters
+        ----------
+        buf : str, Path or StringIO-like, optional, default None
+            Buffer to write to. If None, the output is returned as a string.
+        mode : str, optional
+            Mode in which file is opened, "wt" by default.
+        index : bool, optional, default True
+            Add index (row) labels.
+
+        storage_options : dict, optional
+            Extra options that make sense for a particular storage connection, e.g.
+            host, port, username, password, etc. For HTTP(S) URLs the key-value pairs
+            are forwarded to ``urllib.request.Request`` as header options. For other
+            URLs (e.g. starting with "s3://", and "gcs://") the key-value pairs are
+            forwarded to ``fsspec.open``. Please see ``fsspec`` and ``urllib`` for more
+            details, and for more examples on storage options refer `here
+            <https://pandas.pydata.org/docs/user_guide/io.html?
+            highlight=storage_options#reading-writing-remote-files>`_.
+
+        **kwargs
+            These parameters will be passed to `tabulate <https://pypi.org/project/tabulate>`_.
+
+        Returns
+        -------
+        str
+            DataFrame in Markdown-friendly format.
+
+        See Also
+        --------
+        DataFrame.to_html : Render DataFrame to HTML-formatted table.
+        DataFrame.to_latex : Render DataFrame to LaTeX-formatted table.
+
+        Notes
+        -----
+        Requires the `tabulate <https://pypi.org/project/tabulate>`_ package.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     data={"animal_1": ["elk", "pig"], "animal_2": ["dog", "quetzal"]}
+        ... )
+        >>> print(df.to_markdown())
+        |    | animal_1   | animal_2   |
+        |---:|:-----------|:-----------|
+        |  0 | elk        | dog        |
+        |  1 | pig        | quetzal    |
+
+        Output markdown with a tabulate option.
+
+        >>> print(df.to_markdown(tablefmt="grid"))
+        +----+------------+------------+
+        |    | animal_1   | animal_2   |
+        +====+============+============+
+        |  0 | elk        | dog        |
+        +----+------------+------------+
+        |  1 | pig        | quetzal    |
+        +----+------------+------------+
+        """
+        if "showindex" in kwargs:
+            raise ValueError("Pass 'index' instead of 'showindex")
+
+        kwargs.setdefault("headers", "keys")
+        kwargs.setdefault("tablefmt", "pipe")
+        kwargs.setdefault("showindex", index)
+        tabulate = import_optional_dependency("tabulate")
+        result = tabulate.tabulate(self, **kwargs)
+        if buf is None:
+            return result
+
+        with get_handle(buf, mode, storage_options=storage_options) as handles:
+            handles.handle.write(result)
+        return None
+
+    @overload
+    def to_parquet(
+        self,
+        path: None = ...,
+        *,
+        engine: Literal["auto", "pyarrow", "fastparquet"] = ...,
+        compression: str | None = ...,
+        index: bool | None = ...,
+        partition_cols: list[str] | None = ...,
+        storage_options: StorageOptions = ...,
+        **kwargs,
+    ) -> bytes: ...
+
+    @overload
+    def to_parquet(
+        self,
+        path: FilePath | WriteBuffer[bytes],
+        *,
+        engine: Literal["auto", "pyarrow", "fastparquet"] = ...,
+        compression: str | None = ...,
+        index: bool | None = ...,
+        partition_cols: list[str] | None = ...,
+        storage_options: StorageOptions = ...,
+        **kwargs,
+    ) -> None: ...
+
+    @doc(storage_options=_shared_docs["storage_options"])
+    def to_parquet(
+        self,
+        path: FilePath | WriteBuffer[bytes] | None = None,
+        *,
+        engine: Literal["auto", "pyarrow", "fastparquet"] = "auto",
+        compression: str | None = "snappy",
+        index: bool | None = None,
+        partition_cols: list[str] | None = None,
+        storage_options: StorageOptions | None = None,
+        **kwargs,
+    ) -> bytes | None:
+        """
+        Write a DataFrame to the binary parquet format.
+
+        This function writes the dataframe as a `parquet file
+        <https://parquet.apache.org/>`_. You can choose different parquet
+        backends, and have the option of compression. See
+        :ref:`the user guide <io.parquet>` for more details.
+
+        Parameters
+        ----------
+        path : str, path object, file-like object, or None, default None
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a binary ``write()`` function. If None, the result is
+            returned as bytes. If a string or path, it will be used as Root Directory
+            path when writing a partitioned dataset.
+        engine : {{'auto', 'pyarrow', 'fastparquet'}}, default 'auto'
+            Parquet library to use. If 'auto', then the option
+            ``io.parquet.engine`` is used. The default ``io.parquet.engine``
+            behavior is to try 'pyarrow', falling back to 'fastparquet' if
+            'pyarrow' is unavailable.
+        compression : str or None, default 'snappy'
+            Name of the compression to use. Use ``None`` for no compression.
+            Supported options: 'snappy', 'gzip', 'brotli', 'lz4', 'zstd'.
+        index : bool, default None
+            If ``True``, include the dataframe's index(es) in the file output.
+            If ``False``, they will not be written to the file.
+            If ``None``, similar to ``True`` the dataframe's index(es)
+            will be saved. However, instead of being saved as values,
+            the RangeIndex will be stored as a range in the metadata so it
+            doesn't require much space and is faster. Other indexes will
+            be included as columns in the file output.
+        partition_cols : list, optional, default None
+            Column names by which to partition the dataset.
+            Columns are partitioned in the order they are given.
+            Must be None if path is not a string.
+        {storage_options}
+
+        **kwargs
+            Additional arguments passed to the parquet library. See
+            :ref:`pandas io <io.parquet>` for more details.
+
+        Returns
+        -------
+        bytes if no path argument is provided else None
+            Returns the DataFrame converted to the binary parquet format as bytes if no
+            path argument. Returns None and writes the DataFrame to the specified
+            location in the Parquet format if the path argument is provided.
+
+        See Also
+        --------
+        read_parquet : Read a parquet file.
+        DataFrame.to_orc : Write an orc file.
+        DataFrame.to_csv : Write a csv file.
+        DataFrame.to_sql : Write to a sql table.
+        DataFrame.to_hdf : Write to hdf.
+
+        Notes
+        -----
+        * This function requires either the `fastparquet
+          <https://pypi.org/project/fastparquet>`_ or `pyarrow
+          <https://arrow.apache.org/docs/python/>`_ library.
+        * When saving a DataFrame with categorical columns to parquet,
+          the file size may increase due to the inclusion of all possible
+          categories, not just those present in the data. This behavior
+          is expected and consistent with pandas' handling of categorical data.
+          To manage file size and ensure a more predictable roundtrip process,
+          consider using :meth:`Categorical.remove_unused_categories` on the
+          DataFrame before saving.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(data={{"col1": [1, 2], "col2": [3, 4]}})
+        >>> df.to_parquet("df.parquet.gzip", compression="gzip")  # doctest: +SKIP
+        >>> pd.read_parquet("df.parquet.gzip")  # doctest: +SKIP
+           col1  col2
+        0     1     3
+        1     2     4
+
+        If you want to get a buffer to the parquet content you can use a io.BytesIO
+        object, as long as you don't use partition_cols, which creates multiple files.
+
+        >>> import io
+        >>> f = io.BytesIO()
+        >>> df.to_parquet(f)
+        >>> f.seek(0)
+        0
+        >>> content = f.read()
+        """
+        from pandas.io.parquet import to_parquet
+
+        return to_parquet(
+            self,
+            path,
+            engine,
+            compression=compression,
+            index=index,
+            partition_cols=partition_cols,
+            storage_options=storage_options,
+            **kwargs,
+        )
+
+    @overload
+    def to_orc(
+        self,
+        path: None = ...,
+        *,
+        engine: Literal["pyarrow"] = ...,
+        index: bool | None = ...,
+        engine_kwargs: dict[str, Any] | None = ...,
+    ) -> bytes: ...
+
+    @overload
+    def to_orc(
+        self,
+        path: FilePath | WriteBuffer[bytes],
+        *,
+        engine: Literal["pyarrow"] = ...,
+        index: bool | None = ...,
+        engine_kwargs: dict[str, Any] | None = ...,
+    ) -> None: ...
+
+    @overload
+    def to_orc(
+        self,
+        path: FilePath | WriteBuffer[bytes] | None,
+        *,
+        engine: Literal["pyarrow"] = ...,
+        index: bool | None = ...,
+        engine_kwargs: dict[str, Any] | None = ...,
+    ) -> bytes | None: ...
+
+    def to_orc(
+        self,
+        path: FilePath | WriteBuffer[bytes] | None = None,
+        *,
+        engine: Literal["pyarrow"] = "pyarrow",
+        index: bool | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+    ) -> bytes | None:
+        """
+        Write a DataFrame to the Optimized Row Columnar (ORC) format.
+
+        .. versionadded:: 1.5.0
+
+        Parameters
+        ----------
+        path : str, file-like object or None, default None
+            If a string, it will be used as Root Directory path
+            when writing a partitioned dataset. By file-like object,
+            we refer to objects with a write() method, such as a file handle
+            (e.g. via builtin open function). If path is None,
+            a bytes object is returned.
+        engine : {'pyarrow'}, default 'pyarrow'
+            ORC library to use.
+        index : bool, optional
+            If ``True``, include the dataframe's index(es) in the file output.
+            If ``False``, they will not be written to the file.
+            If ``None``, similar to ``infer`` the dataframe's index(es)
+            will be saved. However, instead of being saved as values,
+            the RangeIndex will be stored as a range in the metadata so it
+            doesn't require much space and is faster. Other indexes will
+            be included as columns in the file output.
+        engine_kwargs : dict[str, Any] or None, default None
+            Additional keyword arguments passed to :func:`pyarrow.orc.write_table`.
+
+        Returns
+        -------
+        bytes if no ``path`` argument is provided else None
+            Bytes object with DataFrame data if ``path`` is not specified else None.
+
+        Raises
+        ------
+        NotImplementedError
+            Dtype of one or more columns is category, unsigned integers, interval,
+            period or sparse.
+        ValueError
+            engine is not pyarrow.
+
+        See Also
+        --------
+        read_orc : Read a ORC file.
+        DataFrame.to_parquet : Write a parquet file.
+        DataFrame.to_csv : Write a csv file.
+        DataFrame.to_sql : Write to a sql table.
+        DataFrame.to_hdf : Write to hdf.
+
+        Notes
+        -----
+        * Find more information on ORC
+          `here <https://en.wikipedia.org/wiki/Apache_ORC>`__.
+        * Before using this function you should read the :ref:`user guide about
+          ORC <io.orc>` and :ref:`install optional dependencies <install.warn_orc>`.
+        * This function requires `pyarrow <https://arrow.apache.org/docs/python/>`_
+          library.
+        * For supported dtypes please refer to `supported ORC features in Arrow
+          <https://arrow.apache.org/docs/cpp/orc.html#data-types>`__.
+        * Currently timezones in datetime columns are not preserved when a
+          dataframe is converted into ORC files.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(data={"col1": [1, 2], "col2": [4, 3]})
+        >>> df.to_orc("df.orc")  # doctest: +SKIP
+        >>> pd.read_orc("df.orc")  # doctest: +SKIP
+           col1  col2
+        0     1     4
+        1     2     3
+
+        If you want to get a buffer to the orc content you can write it to io.BytesIO
+
+        >>> import io
+        >>> b = io.BytesIO(df.to_orc())  # doctest: +SKIP
+        >>> b.seek(0)  # doctest: +SKIP
+        0
+        >>> content = b.read()  # doctest: +SKIP
+        """
+        from pandas.io.orc import to_orc
+
+        return to_orc(
+            self, path, engine=engine, index=index, engine_kwargs=engine_kwargs
+        )
+
+    @overload
+    def to_html(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        columns: Axes | None = ...,
+        col_space: ColspaceArgType | None = ...,
+        header: bool = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: FormattersType | None = ...,
+        float_format: FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        justify: str | None = ...,
+        max_rows: int | None = ...,
+        max_cols: int | None = ...,
+        show_dimensions: bool | str = ...,
+        decimal: str = ...,
+        bold_rows: bool = ...,
+        classes: str | list | tuple | None = ...,
+        escape: bool = ...,
+        notebook: bool = ...,
+        border: int | bool | None = ...,
+        table_id: str | None = ...,
+        render_links: bool = ...,
+        encoding: str | None = ...,
+    ) -> None: ...
+
+    @overload
+    def to_html(
+        self,
+        buf: None = ...,
+        *,
+        columns: Axes | None = ...,
+        col_space: ColspaceArgType | None = ...,
+        header: bool = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: FormattersType | None = ...,
+        float_format: FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        justify: str | None = ...,
+        max_rows: int | None = ...,
+        max_cols: int | None = ...,
+        show_dimensions: bool | str = ...,
+        decimal: str = ...,
+        bold_rows: bool = ...,
+        classes: str | list | tuple | None = ...,
+        escape: bool = ...,
+        notebook: bool = ...,
+        border: int | bool | None = ...,
+        table_id: str | None = ...,
+        render_links: bool = ...,
+        encoding: str | None = ...,
+    ) -> str: ...
+
+    @Substitution(
+        header_type="bool",
+        header="Whether to print column labels, default True",
+        col_space_type="str or int, list or dict of int or str",
+        col_space="The minimum width of each column in CSS length "
+        "units.  An int is assumed to be px units.",
+    )
+    @Substitution(shared_params=fmt.common_docstring, returns=fmt.return_docstring)
+    def to_html(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        columns: Axes | None = None,
+        col_space: ColspaceArgType | None = None,
+        header: bool = True,
+        index: bool = True,
+        na_rep: str = "NaN",
+        formatters: FormattersType | None = None,
+        float_format: FloatFormatType | None = None,
+        sparsify: bool | None = None,
+        index_names: bool = True,
+        justify: str | None = None,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        show_dimensions: bool | str = False,
+        decimal: str = ".",
+        bold_rows: bool = True,
+        classes: str | list | tuple | None = None,
+        escape: bool = True,
+        notebook: bool = False,
+        border: int | bool | None = None,
+        table_id: str | None = None,
+        render_links: bool = False,
+        encoding: str | None = None,
+    ) -> str | None:
+        """
+        Render a DataFrame as an HTML table.
+        %(shared_params)s
+        bold_rows : bool, default True
+            Make the row labels bold in the output.
+        classes : str or list or tuple, default None
+            CSS class(es) to apply to the resulting html table.
+        escape : bool, default True
+            Convert the characters <, >, and & to HTML-safe sequences.
+        notebook : {True, False}, default False
+            Whether the generated HTML is for IPython Notebook.
+        border : int or bool
+            When an integer value is provided, it sets the border attribute in
+            the opening tag, specifying the thickness of the border.
+            If ``False`` or ``0`` is passed, the border attribute will not
+            be present in the ``<table>`` tag.
+            The default value for this parameter is governed by
+            ``pd.options.display.html.border``.
+        table_id : str, optional
+            A css id is included in the opening `<table>` tag if specified.
+        render_links : bool, default False
+            Convert URLs to HTML links.
+        encoding : str, default "utf-8"
+            Set character encoding.
+        %(returns)s
+        See Also
+        --------
+        to_string : Convert DataFrame to a string.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(data={"col1": [1, 2], "col2": [4, 3]})
+        >>> html_string = '''<table border="1" class="dataframe">
+        ...   <thead>
+        ...     <tr style="text-align: right;">
+        ...       <th></th>
+        ...       <th>col1</th>
+        ...       <th>col2</th>
+        ...     </tr>
+        ...   </thead>
+        ...   <tbody>
+        ...     <tr>
+        ...       <th>0</th>
+        ...       <td>1</td>
+        ...       <td>4</td>
+        ...     </tr>
+        ...     <tr>
+        ...       <th>1</th>
+        ...       <td>2</td>
+        ...       <td>3</td>
+        ...     </tr>
+        ...   </tbody>
+        ... </table>'''
+        >>> assert html_string == df.to_html()
+        """
+        if justify is not None and justify not in fmt.VALID_JUSTIFY_PARAMETERS:
+            raise ValueError("Invalid value for justify parameter")
+
+        formatter = fmt.DataFrameFormatter(
+            self,
+            columns=columns,
+            col_space=col_space,
+            na_rep=na_rep,
+            header=header,
+            index=index,
+            formatters=formatters,
+            float_format=float_format,
+            bold_rows=bold_rows,
+            sparsify=sparsify,
+            justify=justify,
+            index_names=index_names,
+            escape=escape,
+            decimal=decimal,
+            max_rows=max_rows,
+            max_cols=max_cols,
+            show_dimensions=show_dimensions,
+        )
+        # TODO: a generic formatter wld b in DataFrameFormatter
+        return fmt.DataFrameRenderer(formatter).to_html(
+            buf=buf,
+            classes=classes,
+            notebook=notebook,
+            border=border,
+            encoding=encoding,
+            table_id=table_id,
+            render_links=render_links,
+        )
+
+    @overload
+    def to_xml(
+        self,
+        path_or_buffer: None = ...,
+        *,
+        index: bool = ...,
+        root_name: str | None = ...,
+        row_name: str | None = ...,
+        na_rep: str | None = ...,
+        attr_cols: list[str] | None = ...,
+        elem_cols: list[str] | None = ...,
+        namespaces: dict[str | None, str] | None = ...,
+        prefix: str | None = ...,
+        encoding: str = ...,
+        xml_declaration: bool | None = ...,
+        pretty_print: bool | None = ...,
+        parser: XMLParsers | None = ...,
+        stylesheet: FilePath | ReadBuffer[str] | ReadBuffer[bytes] | None = ...,
+        compression: CompressionOptions = ...,
+        storage_options: StorageOptions | None = ...,
+    ) -> str: ...
+
+    @overload
+    def to_xml(
+        self,
+        path_or_buffer: FilePath | WriteBuffer[bytes] | WriteBuffer[str],
+        *,
+        index: bool = ...,
+        root_name: str | None = ...,
+        row_name: str | None = ...,
+        na_rep: str | None = ...,
+        attr_cols: list[str] | None = ...,
+        elem_cols: list[str] | None = ...,
+        namespaces: dict[str | None, str] | None = ...,
+        prefix: str | None = ...,
+        encoding: str = ...,
+        xml_declaration: bool | None = ...,
+        pretty_print: bool | None = ...,
+        parser: XMLParsers | None = ...,
+        stylesheet: FilePath | ReadBuffer[str] | ReadBuffer[bytes] | None = ...,
+        compression: CompressionOptions = ...,
+        storage_options: StorageOptions | None = ...,
+    ) -> None: ...
+
+    @doc(
+        storage_options=_shared_docs["storage_options"],
+        compression_options=_shared_docs["compression_options"] % "path_or_buffer",
+    )
+    def to_xml(
+        self,
+        path_or_buffer: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = None,
+        *,
+        index: bool = True,
+        root_name: str | None = "data",
+        row_name: str | None = "row",
+        na_rep: str | None = None,
+        attr_cols: list[str] | None = None,
+        elem_cols: list[str] | None = None,
+        namespaces: dict[str | None, str] | None = None,
+        prefix: str | None = None,
+        encoding: str = "utf-8",
+        xml_declaration: bool | None = True,
+        pretty_print: bool | None = True,
+        parser: XMLParsers | None = "lxml",
+        stylesheet: FilePath | ReadBuffer[str] | ReadBuffer[bytes] | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+    ) -> str | None:
+        """
+        Render a DataFrame to an XML document.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        path_or_buffer : str, path object, file-like object, or None, default None
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a ``write()`` function. If None, the result is returned
+            as a string.
+        index : bool, default True
+            Whether to include index in XML document.
+        root_name : str, default 'data'
+            The name of root element in XML document.
+        row_name : str, default 'row'
+            The name of row element in XML document.
+        na_rep : str, optional
+            Missing data representation.
+        attr_cols : list-like, optional
+            List of columns to write as attributes in row element.
+            Hierarchical columns will be flattened with underscore
+            delimiting the different levels.
+        elem_cols : list-like, optional
+            List of columns to write as children in row element. By default,
+            all columns output as children of row element. Hierarchical
+            columns will be flattened with underscore delimiting the
+            different levels.
+        namespaces : dict, optional
+            All namespaces to be defined in root element. Keys of dict
+            should be prefix names and values of dict corresponding URIs.
+            Default namespaces should be given empty string key. For
+            example, ::
+
+                namespaces = {{"": "https://example.com"}}
+
+        prefix : str, optional
+            Namespace prefix to be used for every element and/or attribute
+            in document. This should be one of the keys in ``namespaces``
+            dict.
+        encoding : str, default 'utf-8'
+            Encoding of the resulting document.
+        xml_declaration : bool, default True
+            Whether to include the XML declaration at start of document.
+        pretty_print : bool, default True
+            Whether output should be pretty printed with indentation and
+            line breaks.
+        parser : {{'lxml','etree'}}, default 'lxml'
+            Parser module to use for building of tree. Only 'lxml' and
+            'etree' are supported. With 'lxml', the ability to use XSLT
+            stylesheet is supported.
+        stylesheet : str, path object or file-like object, optional
+            A URL, file-like object, or a raw string containing an XSLT
+            script used to transform the raw XML output. Script should use
+            layout of elements and attributes from original output. This
+            argument requires ``lxml`` to be installed. Only XSLT 1.0
+            scripts and not later versions is currently supported.
+        {compression_options}
+
+            .. versionchanged:: 1.4.0 Zstandard support.
+
+        {storage_options}
+
+        Returns
+        -------
+        None or str
+            If ``io`` is None, returns the resulting XML format as a
+            string. Otherwise returns None.
+
+        See Also
+        --------
+        to_json : Convert the pandas object to a JSON string.
+        to_html : Convert DataFrame to a html.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [["square", 360, 4], ["circle", 360, np.nan], ["triangle", 180, 3]],
+        ...     columns=["shape", "degrees", "sides"],
+        ... )
+
+        >>> df.to_xml()  # doctest: +SKIP
+        <?xml version='1.0' encoding='utf-8'?>
+        <data>
+          <row>
+            <index>0</index>
+            <shape>square</shape>
+            <degrees>360</degrees>
+            <sides>4.0</sides>
+          </row>
+          <row>
+            <index>1</index>
+            <shape>circle</shape>
+            <degrees>360</degrees>
+            <sides/>
+          </row>
+          <row>
+            <index>2</index>
+            <shape>triangle</shape>
+            <degrees>180</degrees>
+            <sides>3.0</sides>
+          </row>
+        </data>
+
+        >>> df.to_xml(
+        ...     attr_cols=["index", "shape", "degrees", "sides"]
+        ... )  # doctest: +SKIP
+        <?xml version='1.0' encoding='utf-8'?>
+        <data>
+          <row index="0" shape="square" degrees="360" sides="4.0"/>
+          <row index="1" shape="circle" degrees="360"/>
+          <row index="2" shape="triangle" degrees="180" sides="3.0"/>
+        </data>
+
+        >>> df.to_xml(
+        ...     namespaces={{"doc": "https://example.com"}}, prefix="doc"
+        ... )  # doctest: +SKIP
+        <?xml version='1.0' encoding='utf-8'?>
+        <doc:data xmlns:doc="https://example.com">
+          <doc:row>
+            <doc:index>0</doc:index>
+            <doc:shape>square</doc:shape>
+            <doc:degrees>360</doc:degrees>
+            <doc:sides>4.0</doc:sides>
+          </doc:row>
+          <doc:row>
+            <doc:index>1</doc:index>
+            <doc:shape>circle</doc:shape>
+            <doc:degrees>360</doc:degrees>
+            <doc:sides/>
+          </doc:row>
+          <doc:row>
+            <doc:index>2</doc:index>
+            <doc:shape>triangle</doc:shape>
+            <doc:degrees>180</doc:degrees>
+            <doc:sides>3.0</doc:sides>
+          </doc:row>
+        </doc:data>
+        """
+
+        from pandas.io.formats.xml import (
+            EtreeXMLFormatter,
+            LxmlXMLFormatter,
+        )
+
+        lxml = import_optional_dependency("lxml.etree", errors="ignore")
+
+        TreeBuilder: type[EtreeXMLFormatter | LxmlXMLFormatter]
+
+        if parser == "lxml":
+            if lxml is not None:
+                TreeBuilder = LxmlXMLFormatter
+            else:
+                raise ImportError(
+                    "lxml not found, please install or use the etree parser."
+                )
+
+        elif parser == "etree":
+            TreeBuilder = EtreeXMLFormatter
+
+        else:
+            raise ValueError("Values for parser can only be lxml or etree.")
+
+        xml_formatter = TreeBuilder(
+            self,
+            path_or_buffer=path_or_buffer,
+            index=index,
+            root_name=root_name,
+            row_name=row_name,
+            na_rep=na_rep,
+            attr_cols=attr_cols,
+            elem_cols=elem_cols,
+            namespaces=namespaces,
+            prefix=prefix,
+            encoding=encoding,
+            xml_declaration=xml_declaration,
+            pretty_print=pretty_print,
+            stylesheet=stylesheet,
+            compression=compression,
+            storage_options=storage_options,
+        )
+
+        return xml_formatter.write_output()
+
+    # ----------------------------------------------------------------------
+    @doc(INFO_DOCSTRING, **frame_sub_kwargs)
+    def info(
+        self,
+        verbose: bool | None = None,
+        buf: WriteBuffer[str] | None = None,
+        max_cols: int | None = None,
+        memory_usage: bool | str | None = None,
+        show_counts: bool | None = None,
+    ) -> None:
+        info = DataFrameInfo(
+            data=self,
+            memory_usage=memory_usage,
+        )
+        info.render(
+            buf=buf,
+            max_cols=max_cols,
+            verbose=verbose,
+            show_counts=show_counts,
+        )
+
+    def memory_usage(self, index: bool = True, deep: bool = False) -> Series:
+        """
+        Return the memory usage of each column in bytes.
+
+        The memory usage can optionally include the contribution of
+        the index and elements of `object` dtype.
+
+        This value is displayed in `DataFrame.info` by default. This can be
+        suppressed by setting ``pandas.options.display.memory_usage`` to False.
+
+        Parameters
+        ----------
+        index : bool, default True
+            Specifies whether to include the memory usage of the DataFrame's
+            index in returned Series. If ``index=True``, the memory usage of
+            the index is the first item in the output.
+        deep : bool, default False
+            If True, introspect the data deeply by interrogating
+            `object` dtypes for system-level memory consumption, and include
+            it in the returned values.
+
+        Returns
+        -------
+        Series
+            A Series whose index is the original column names and whose values
+            is the memory usage of each column in bytes.
+
+        See Also
+        --------
+        numpy.ndarray.nbytes : Total bytes consumed by the elements of an
+            ndarray.
+        Series.memory_usage : Bytes consumed by a Series.
+        Categorical : Memory-efficient array for string values with
+            many repeated values.
+        DataFrame.info : Concise summary of a DataFrame.
+
+        Notes
+        -----
+        See the :ref:`Frequently Asked Questions <df-memory-usage>` for more
+        details.
+
+        Examples
+        --------
+        >>> dtypes = ["int64", "float64", "complex128", "object", "bool"]
+        >>> data = dict([(t, np.ones(shape=5000, dtype=int).astype(t)) for t in dtypes])
+        >>> df = pd.DataFrame(data)
+        >>> df.head()
+           int64  float64            complex128  object  bool
+        0      1      1.0              1.0+0.0j       1  True
+        1      1      1.0              1.0+0.0j       1  True
+        2      1      1.0              1.0+0.0j       1  True
+        3      1      1.0              1.0+0.0j       1  True
+        4      1      1.0              1.0+0.0j       1  True
+
+        >>> df.memory_usage()
+        Index           128
+        int64         40000
+        float64       40000
+        complex128    80000
+        object        40000
+        bool           5000
+        dtype: int64
+
+        >>> df.memory_usage(index=False)
+        int64         40000
+        float64       40000
+        complex128    80000
+        object        40000
+        bool           5000
+        dtype: int64
+
+        The memory footprint of `object` dtype columns is ignored by default:
+
+        >>> df.memory_usage(deep=True)
+        Index            128
+        int64          40000
+        float64        40000
+        complex128     80000
+        object        180000
+        bool            5000
+        dtype: int64
+
+        Use a Categorical for efficient storage of an object-dtype column with
+        many repeated values.
+
+        >>> df["object"].astype("category").memory_usage(deep=True)
+        5136
+        """
+        result = self._constructor_sliced(
+            [c.memory_usage(index=False, deep=deep) for col, c in self.items()],
+            index=self.columns,
+            dtype=np.intp,
+        )
+        if index:
+            index_memory_usage = self._constructor_sliced(
+                self.index.memory_usage(deep=deep), index=["Index"]
+            )
+            result = index_memory_usage._append(result)
+        return result
+
+    def transpose(
+        self,
+        *args,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> DataFrame:
+        """
+        Transpose index and columns.
+
+        Reflect the DataFrame over its main diagonal by writing rows as columns
+        and vice-versa. The property :attr:`.T` is an accessor to the method
+        :meth:`transpose`.
+
+        Parameters
+        ----------
+        *args : tuple, optional
+            Accepted for compatibility with NumPy.
+        copy : bool, default False
+            Whether to copy the data after transposing, even for DataFrames
+            with a single dtype.
+
+            Note that a copy is always required for mixed dtype DataFrames,
+            or for DataFrames with any extension types.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        DataFrame
+            The transposed DataFrame.
+
+        See Also
+        --------
+        numpy.transpose : Permute the dimensions of a given array.
+
+        Notes
+        -----
+        Transposing a DataFrame with mixed dtypes will result in a homogeneous
+        DataFrame with the `object` dtype. In such a case, a copy of the data
+        is always made.
+
+        Examples
+        --------
+        **Square DataFrame with homogeneous dtype**
+
+        >>> d1 = {"col1": [1, 2], "col2": [3, 4]}
+        >>> df1 = pd.DataFrame(data=d1)
+        >>> df1
+           col1  col2
+        0     1     3
+        1     2     4
+
+        >>> df1_transposed = df1.T  # or df1.transpose()
+        >>> df1_transposed
+              0  1
+        col1  1  2
+        col2  3  4
+
+        When the dtype is homogeneous in the original DataFrame, we get a
+        transposed DataFrame with the same dtype:
+
+        >>> df1.dtypes
+        col1    int64
+        col2    int64
+        dtype: object
+        >>> df1_transposed.dtypes
+        0    int64
+        1    int64
+        dtype: object
+
+        **Non-square DataFrame with mixed dtypes**
+
+        >>> d2 = {
+        ...     "name": ["Alice", "Bob"],
+        ...     "score": [9.5, 8],
+        ...     "employed": [False, True],
+        ...     "kids": [0, 0],
+        ... }
+        >>> df2 = pd.DataFrame(data=d2)
+        >>> df2
+            name  score  employed  kids
+        0  Alice    9.5     False     0
+        1    Bob    8.0      True     0
+
+        >>> df2_transposed = df2.T  # or df2.transpose()
+        >>> df2_transposed
+                      0     1
+        name      Alice   Bob
+        score       9.5   8.0
+        employed  False  True
+        kids          0     0
+
+        When the DataFrame has mixed dtypes, we get a transposed DataFrame with
+        the `object` dtype:
+
+        >>> df2.dtypes
+        name         object
+        score       float64
+        employed       bool
+        kids          int64
+        dtype: object
+        >>> df2_transposed.dtypes
+        0    object
+        1    object
+        dtype: object
+        """
+        self._check_copy_deprecation(copy)
+        nv.validate_transpose(args, {})
+        # construct the args
+
+        first_dtype = self.dtypes.iloc[0] if len(self.columns) else None
+
+        if self._can_fast_transpose:
+            # Note: tests pass without this, but this improves perf quite a bit.
+            new_vals = self._values.T
+
+            result = self._constructor(
+                new_vals,
+                index=self.columns,
+                columns=self.index,
+                copy=False,
+                dtype=new_vals.dtype,
+            )
+            if len(self) > 0:
+                result._mgr.add_references(self._mgr)
+
+        elif (
+            self._is_homogeneous_type
+            and first_dtype is not None
+            and isinstance(first_dtype, ExtensionDtype)
+        ):
+            new_values: list
+            if isinstance(first_dtype, BaseMaskedDtype):
+                # We have masked arrays with the same dtype. We can transpose faster.
+                from pandas.core.arrays.masked import (
+                    transpose_homogeneous_masked_arrays,
+                )
+
+                new_values = transpose_homogeneous_masked_arrays(
+                    cast(Sequence[BaseMaskedArray], self._iter_column_arrays())
+                )
+            elif isinstance(first_dtype, ArrowDtype):
+                # We have arrow EAs with the same dtype. We can transpose faster.
+                from pandas.core.arrays.arrow.array import (
+                    ArrowExtensionArray,
+                    transpose_homogeneous_pyarrow,
+                )
+
+                new_values = transpose_homogeneous_pyarrow(
+                    cast(Sequence[ArrowExtensionArray], self._iter_column_arrays())
+                )
+            else:
+                # We have other EAs with the same dtype. We preserve dtype in transpose.
+                arr_typ = first_dtype.construct_array_type()
+                values = self.values
+                new_values = [
+                    arr_typ._from_sequence(row, dtype=first_dtype) for row in values
+                ]
+
+            result = type(self)._from_arrays(
+                new_values,
+                index=self.columns,
+                columns=self.index,
+                verify_integrity=False,
+            )
+
+        else:
+            new_arr = self.values.T
+            result = self._constructor(
+                new_arr,
+                index=self.columns,
+                columns=self.index,
+                dtype=new_arr.dtype,
+                # We already made a copy (more than one block)
+                copy=False,
+            )
+
+        return result.__finalize__(self, method="transpose")
+
+    @property
+    def T(self) -> DataFrame:
+        """
+        The transpose of the DataFrame.
+
+        Returns
+        -------
+        DataFrame
+            The transposed DataFrame.
+
+        See Also
+        --------
+        DataFrame.transpose : Transpose index and columns.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df
+           col1  col2
+        0     1     3
+        1     2     4
+
+        >>> df.T
+              0  1
+        col1  1  2
+        col2  3  4
+        """
+        return self.transpose()
+
+    # ----------------------------------------------------------------------
+    # Indexing Methods
+
+    def _ixs(self, i: int, axis: AxisInt = 0) -> Series:
+        """
+        Parameters
+        ----------
+        i : int
+        axis : int
+
+        Returns
+        -------
+        Series
+        """
+        # irow
+        if axis == 0:
+            new_mgr = self._mgr.fast_xs(i)
+
+            result = self._constructor_sliced_from_mgr(new_mgr, axes=new_mgr.axes)
+            result._name = self.index[i]
+            return result.__finalize__(self)
+
+        # icol
+        else:
+            col_mgr = self._mgr.iget(i)
+            return self._box_col_values(col_mgr, i)
+
+    def _get_column_array(self, i: int) -> ArrayLike:
+        """
+        Get the values of the i'th column (ndarray or ExtensionArray, as stored
+        in the Block)
+
+        Warning! The returned array is a view but doesn't handle Copy-on-Write,
+        so this should be used with caution (for read-only purposes).
+        """
+        return self._mgr.iget_values(i)
+
+    def _iter_column_arrays(self) -> Iterator[ArrayLike]:
+        """
+        Iterate over the arrays of all columns in order.
+        This returns the values as stored in the Block (ndarray or ExtensionArray).
+
+        Warning! The returned array is a view but doesn't handle Copy-on-Write,
+        so this should be used with caution (for read-only purposes).
+        """
+        for i in range(len(self.columns)):
+            yield self._get_column_array(i)
+
+    def __getitem__(self, key):
+        check_dict_or_set_indexers(key)
+        key = lib.item_from_zerodim(key)
+        key = com.apply_if_callable(key, self)
+
+        if is_hashable(key) and not is_iterator(key) and not isinstance(key, slice):
+            # is_iterator to exclude generator e.g. test_getitem_listlike
+            # As of Python 3.12, slice is hashable which breaks MultiIndex (GH#57500)
+
+            # shortcut if the key is in columns
+            is_mi = isinstance(self.columns, MultiIndex)
+            # GH#45316 Return view if key is not duplicated
+            # Only use drop_duplicates with duplicates for performance
+            if not is_mi and (
+                (self.columns.is_unique and key in self.columns)
+                or key in self.columns.drop_duplicates(keep=False)
+            ):
+                return self._get_item(key)
+
+            elif is_mi and self.columns.is_unique and key in self.columns:
+                return self._getitem_multilevel(key)
+
+        # Do we have a slicer (on rows)?
+        if isinstance(key, slice):
+            return self._getitem_slice(key)
+
+        # Do we have a (boolean) DataFrame?
+        if isinstance(key, DataFrame):
+            return self.where(key)
+
+        # Do we have a (boolean) 1d indexer?
+        if com.is_bool_indexer(key):
+            return self._getitem_bool_array(key)
+
+        # We are left with two options: a single key, and a collection of keys,
+        # We interpret tuples as collections only for non-MultiIndex
+        is_single_key = isinstance(key, tuple) or not is_list_like(key)
+
+        if is_single_key:
+            if self.columns.nlevels > 1:
+                return self._getitem_multilevel(key)
+            indexer = self.columns.get_loc(key)
+            if is_integer(indexer):
+                indexer = [indexer]
+        else:
+            if is_iterator(key):
+                key = list(key)
+            indexer = self.columns._get_indexer_strict(key, "columns")[1]
+
+        # take() does not accept boolean indexers
+        if getattr(indexer, "dtype", None) == bool:
+            indexer = np.where(indexer)[0]
+
+        if isinstance(indexer, slice):
+            return self._slice(indexer, axis=1)
+
+        data = self.take(indexer, axis=1)
+
+        if is_single_key:
+            # What does looking for a single key in a non-unique index return?
+            # The behavior is inconsistent. It returns a Series, except when
+            # - the key itself is repeated (test on data.shape, #9519), or
+            # - we have a MultiIndex on columns (test on self.columns, #21309)
+            if data.shape[1] == 1 and not isinstance(self.columns, MultiIndex):
+                # GH#26490 using data[key] can cause RecursionError
+                return data._get_item(key)
+
+        return data
+
+    def _getitem_bool_array(self, key):
+        # also raises Exception if object array with NA values
+        # warning here just in case -- previously __setitem__ was
+        # reindexing but __getitem__ was not; it seems more reasonable to
+        # go with the __setitem__ behavior since that is more consistent
+        # with all other indexing behavior
+        if isinstance(key, Series) and not key.index.equals(self.index):
+            warnings.warn(
+                "Boolean Series key will be reindexed to match DataFrame index.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+        elif len(key) != len(self.index):
+            raise ValueError(
+                f"Item wrong length {len(key)} instead of {len(self.index)}."
+            )
+
+        # check_bool_indexer will throw exception if Series key cannot
+        # be reindexed to match DataFrame rows
+        key = check_bool_indexer(self.index, key)
+
+        if key.all():
+            return self.copy(deep=False)
+
+        indexer = key.nonzero()[0]
+        return self.take(indexer, axis=0)
+
+    def _getitem_multilevel(self, key):
+        # self.columns is a MultiIndex
+        loc = self.columns.get_loc(key)
+        if isinstance(loc, (slice, np.ndarray)):
+            new_columns = self.columns[loc]
+            result_columns = maybe_droplevels(new_columns, key)
+            result = self.iloc[:, loc]
+            result.columns = result_columns
+
+            # If there is only one column being returned, and its name is
+            # either an empty string, or a tuple with an empty string as its
+            # first element, then treat the empty string as a placeholder
+            # and return the column as if the user had provided that empty
+            # string in the key. If the result is a Series, exclude the
+            # implied empty string from its name.
+            if len(result.columns) == 1:
+                # e.g. test_frame_getitem_multicolumn_empty_level,
+                #  test_frame_mixed_depth_get, test_loc_setitem_single_column_slice
+                top = result.columns[0]
+                if isinstance(top, tuple):
+                    top = top[0]
+                if top == "":
+                    result = result[""]
+                    if isinstance(result, Series):
+                        result = self._constructor_sliced(
+                            result, index=self.index, name=key
+                        )
+
+            return result
+        else:
+            # loc is neither a slice nor ndarray, so must be an int
+            return self._ixs(loc, axis=1)
+
+    def _get_value(self, index, col, takeable: bool = False) -> Scalar:
+        """
+        Quickly retrieve single value at passed column and index.
+
+        Parameters
+        ----------
+        index : row label
+        col : column label
+        takeable : interpret the index/col as indexers, default False
+
+        Returns
+        -------
+        scalar
+
+        Notes
+        -----
+        Assumes that both `self.index._index_as_unique` and
+        `self.columns._index_as_unique`; Caller is responsible for checking.
+        """
+        if takeable:
+            series = self._ixs(col, axis=1)
+            return series._values[index]
+
+        series = self._get_item(col)
+
+        if not isinstance(self.index, MultiIndex):
+            # CategoricalIndex: Trying to use the engine fastpath may give incorrect
+            #  results if our categories are integers that dont match our codes
+            # IntervalIndex: IntervalTree has no get_loc
+            row = self.index.get_loc(index)
+            return series._values[row]
+
+        # For MultiIndex going through engine effectively restricts us to
+        #  same-length tuples; see test_get_set_value_no_partial_indexing
+        loc = self.index._engine.get_loc(index)
+        return series._values[loc]
+
+    def isetitem(self, loc, value) -> None:
+        """
+        Set the given value in the column with position `loc`.
+
+        This is a positional analogue to ``__setitem__``.
+
+        Parameters
+        ----------
+        loc : int or sequence of ints
+            Index position for the column.
+        value : scalar or arraylike
+            Value(s) for the column.
+
+        See Also
+        --------
+        DataFrame.iloc : Purely integer-location based indexing for selection by
+            position.
+
+        Notes
+        -----
+        ``frame.isetitem(loc, value)`` is an in-place method as it will
+        modify the DataFrame in place (not returning a new object). In contrast to
+        ``frame.iloc[:, i] = value`` which will try to update the existing values in
+        place, ``frame.isetitem(loc, value)`` will not update the values of the column
+        itself in place, it will instead insert a new array.
+
+        In cases where ``frame.columns`` is unique, this is equivalent to
+        ``frame[frame.columns[i]] = value``.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.isetitem(1, [5, 6])
+        >>> df
+              A  B
+        0     1  5
+        1     2  6
+        """
+        if isinstance(value, DataFrame):
+            if is_integer(loc):
+                loc = [loc]
+
+            if len(loc) != len(value.columns):
+                raise ValueError(
+                    f"Got {len(loc)} positions but value has {len(value.columns)} "
+                    f"columns."
+                )
+
+            for i, idx in enumerate(loc):
+                arraylike, refs = self._sanitize_column(value.iloc[:, i])
+                self._iset_item_mgr(idx, arraylike, inplace=False, refs=refs)
+            return
+
+        arraylike, refs = self._sanitize_column(value)
+        self._iset_item_mgr(loc, arraylike, inplace=False, refs=refs)
+
+    def __setitem__(self, key, value) -> None:
+        if not PYPY:
+            if sys.getrefcount(self) <= 3:
+                warnings.warn(
+                    _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
+                )
+
+        key = com.apply_if_callable(key, self)
+
+        # see if we can slice the rows
+        if isinstance(key, slice):
+            slc = self.index._convert_slice_indexer(key, kind="getitem")
+            return self._setitem_slice(slc, value)
+
+        if isinstance(key, DataFrame) or getattr(key, "ndim", None) == 2:
+            self._setitem_frame(key, value)
+        elif isinstance(key, (Series, np.ndarray, list, Index)):
+            self._setitem_array(key, value)
+        elif isinstance(value, DataFrame):
+            self._set_item_frame_value(key, value)
+        elif (
+            is_list_like(value)
+            and not self.columns.is_unique
+            and 1 < len(self.columns.get_indexer_for([key])) == len(value)
+        ):
+            # Column to set is duplicated
+            self._setitem_array([key], value)
+        else:
+            # set column
+            self._set_item(key, value)
+
+    def _setitem_slice(self, key: slice, value) -> None:
+        # NB: we can't just use self.loc[key] = value because that
+        #  operates on labels and we need to operate positional for
+        #  backwards-compat, xref GH#31469
+        self.iloc[key] = value
+
+    def _setitem_array(self, key, value) -> None:
+        # also raises Exception if object array with NA values
+        if com.is_bool_indexer(key):
+            # bool indexer is indexing along rows
+            if len(key) != len(self.index):
+                raise ValueError(
+                    f"Item wrong length {len(key)} instead of {len(self.index)}!"
+                )
+            key = check_bool_indexer(self.index, key)
+            indexer = key.nonzero()[0]
+            if isinstance(value, DataFrame):
+                # GH#39931 reindex since iloc does not align
+                value = value.reindex(self.index.take(indexer))
+            self.iloc[indexer] = value
+
+        else:
+            # Note: unlike self.iloc[:, indexer] = value, this will
+            #  never try to overwrite values inplace
+
+            if isinstance(value, DataFrame):
+                check_key_length(self.columns, key, value)
+                for k1, k2 in zip(key, value.columns):
+                    self[k1] = value[k2]
+
+            elif not is_list_like(value):
+                for col in key:
+                    self[col] = value
+
+            elif isinstance(value, np.ndarray) and value.ndim == 2:
+                self._iset_not_inplace(key, value)
+
+            elif np.ndim(value) > 1:
+                # list of lists
+                value = DataFrame(value).values
+                self._setitem_array(key, value)
+
+            else:
+                self._iset_not_inplace(key, value)
+
+    def _iset_not_inplace(self, key, value) -> None:
+        # GH#39510 when setting with df[key] = obj with a list-like key and
+        #  list-like value, we iterate over those listlikes and set columns
+        #  one at a time.  This is different from dispatching to
+        #  `self.loc[:, key]= value`  because loc.__setitem__ may overwrite
+        #  data inplace, whereas this will insert new arrays.
+
+        def igetitem(obj, i: int):
+            # Note: we catch DataFrame obj before getting here, but
+            #  hypothetically would return obj.iloc[:, i]
+            if isinstance(obj, np.ndarray):
+                return obj[..., i]
+            else:
+                return obj[i]
+
+        if self.columns.is_unique:
+            if np.shape(value)[-1] != len(key):
+                raise ValueError("Columns must be same length as key")
+
+            for i, col in enumerate(key):
+                self[col] = igetitem(value, i)
+
+        else:
+            ilocs = self.columns.get_indexer_non_unique(key)[0]
+            if (ilocs < 0).any():
+                # key entries not in self.columns
+                raise NotImplementedError
+
+            if np.shape(value)[-1] != len(ilocs):
+                raise ValueError("Columns must be same length as key")
+
+            assert np.ndim(value) <= 2
+
+            orig_columns = self.columns
+
+            # Using self.iloc[:, i] = ... may set values inplace, which
+            #  by convention we do not do in __setitem__
+            try:
+                self.columns = Index(range(len(self.columns)))
+                for i, iloc in enumerate(ilocs):
+                    self[iloc] = igetitem(value, i)
+            finally:
+                self.columns = orig_columns
+
+    def _setitem_frame(self, key, value) -> None:
+        # support boolean setting with DataFrame input, e.g.
+        # df[df > df2] = 0
+        if isinstance(key, np.ndarray):
+            if key.shape != self.shape:
+                raise ValueError("Array conditional must be same shape as self")
+            key = self._constructor(key, **self._construct_axes_dict(), copy=False)
+
+        if key.size and not all(is_bool_dtype(dtype) for dtype in key.dtypes):
+            raise TypeError(
+                "Must pass DataFrame or 2-d ndarray with boolean values only"
+            )
+
+        self._where(-key, value, inplace=True)
+
+    def _set_item_frame_value(self, key, value: DataFrame) -> None:
+        self._ensure_valid_index(value)
+
+        # align columns
+        if key in self.columns:
+            loc = self.columns.get_loc(key)
+            cols = self.columns[loc]
+            len_cols = 1 if is_scalar(cols) or isinstance(cols, tuple) else len(cols)
+            if len_cols != len(value.columns):
+                raise ValueError("Columns must be same length as key")
+
+            # align right-hand-side columns if self.columns
+            # is multi-index and self[key] is a sub-frame
+            if isinstance(self.columns, MultiIndex) and isinstance(
+                loc, (slice, Series, np.ndarray, Index)
+            ):
+                cols_droplevel = maybe_droplevels(cols, key)
+                if len(cols_droplevel) and not cols_droplevel.equals(value.columns):
+                    value = value.reindex(cols_droplevel, axis=1)
+
+                for col, col_droplevel in zip(cols, cols_droplevel):
+                    self[col] = value[col_droplevel]
+                return
+
+            if is_scalar(cols):
+                self[cols] = value[value.columns[0]]
+                return
+
+            locs: np.ndarray | list
+            if isinstance(loc, slice):
+                locs = np.arange(loc.start, loc.stop, loc.step)
+            elif is_scalar(loc):
+                locs = [loc]
+            else:
+                locs = loc.nonzero()[0]
+
+            return self.isetitem(locs, value)
+
+        if len(value.columns) > 1:
+            raise ValueError(
+                "Cannot set a DataFrame with multiple columns to the single "
+                f"column {key}"
+            )
+        elif len(value.columns) == 0:
+            raise ValueError(
+                f"Cannot set a DataFrame without columns to the column {key}"
+            )
+
+        self[key] = value[value.columns[0]]
+
+    def _iset_item_mgr(
+        self,
+        loc: int | slice | np.ndarray,
+        value,
+        inplace: bool = False,
+        refs: BlockValuesRefs | None = None,
+    ) -> None:
+        # when called from _set_item_mgr loc can be anything returned from get_loc
+        self._mgr.iset(loc, value, inplace=inplace, refs=refs)
+
+    def _set_item_mgr(
+        self, key, value: ArrayLike, refs: BlockValuesRefs | None = None
+    ) -> None:
+        try:
+            loc = self._info_axis.get_loc(key)
+        except KeyError:
+            # This item wasn't present, just insert at end
+            self._mgr.insert(len(self._info_axis), key, value, refs)
+        else:
+            self._iset_item_mgr(loc, value, refs=refs)
+
+    def _iset_item(self, loc: int, value: Series, inplace: bool = True) -> None:
+        # We are only called from _replace_columnwise which guarantees that
+        # no reindex is necessary
+        self._iset_item_mgr(loc, value._values, inplace=inplace, refs=value._references)
+
+    def _set_item(self, key, value) -> None:
+        """
+        Add series to DataFrame in specified column.
+
+        If series is a numpy-array (not a Series/TimeSeries), it must be the
+        same length as the DataFrames index or an error will be thrown.
+
+        Series/TimeSeries will be conformed to the DataFrames index to
+        ensure homogeneity.
+        """
+        value, refs = self._sanitize_column(value)
+
+        if (
+            key in self.columns
+            and value.ndim == 1
+            and not isinstance(value.dtype, ExtensionDtype)
+        ):
+            # broadcast across multiple columns if necessary
+            if not self.columns.is_unique or isinstance(self.columns, MultiIndex):
+                existing_piece = self[key]
+                if isinstance(existing_piece, DataFrame):
+                    value = np.tile(value, (len(existing_piece.columns), 1)).T
+                    refs = None
+
+        self._set_item_mgr(key, value, refs)
+
+    def _set_value(
+        self, index: IndexLabel, col, value: Scalar, takeable: bool = False
+    ) -> None:
+        """
+        Put single value at passed column and index.
+
+        Parameters
+        ----------
+        index : Label
+            row label
+        col : Label
+            column label
+        value : scalar
+        takeable : bool, default False
+            Sets whether or not index/col interpreted as indexers
+        """
+        try:
+            if takeable:
+                icol = col
+                iindex = cast(int, index)
+            else:
+                icol = self.columns.get_loc(col)
+                iindex = self.index.get_loc(index)
+            self._mgr.column_setitem(icol, iindex, value, inplace_only=True)
+
+        except (KeyError, TypeError, ValueError, LossySetitemError):
+            # get_loc might raise a KeyError for missing labels (falling back
+            #  to (i)loc will do expansion of the index)
+            # column_setitem will do validation that may raise TypeError,
+            #  ValueError, or LossySetitemError
+            # set using a non-recursive method & reset the cache
+            if takeable:
+                self.iloc[index, col] = value
+            else:
+                self.loc[index, col] = value
+
+        except InvalidIndexError as ii_err:
+            # GH48729: Seems like you are trying to assign a value to a
+            # row when only scalar options are permitted
+            raise InvalidIndexError(
+                f"You can only assign a scalar value not a {type(value)}"
+            ) from ii_err
+
+    def _ensure_valid_index(self, value) -> None:
+        """
+        Ensure that if we don't have an index, that we can create one from the
+        passed value.
+        """
+        # GH5632, make sure that we are a Series convertible
+        if not len(self.index) and is_list_like(value) and len(value):
+            if not isinstance(value, DataFrame):
+                try:
+                    value = Series(value)
+                except (ValueError, NotImplementedError, TypeError) as err:
+                    raise ValueError(
+                        "Cannot set a frame with no defined index "
+                        "and a value that cannot be converted to a Series"
+                    ) from err
+
+            # GH31368 preserve name of index
+            index_copy = value.index.copy()
+            if self.index.name is not None:
+                index_copy.name = self.index.name
+
+            self._mgr = self._mgr.reindex_axis(index_copy, axis=1, fill_value=np.nan)
+
+    def _box_col_values(self, values: SingleBlockManager, loc: int) -> Series:
+        """
+        Provide boxed values for a column.
+        """
+        # Lookup in columns so that if e.g. a str datetime was passed
+        #  we attach the Timestamp object as the name.
+        name = self.columns[loc]
+        # We get index=self.index bc values is a SingleBlockManager
+        obj = self._constructor_sliced_from_mgr(values, axes=values.axes)
+        obj._name = name
+        return obj.__finalize__(self)
+
+    def _get_item(self, item: Hashable) -> Series:
+        loc = self.columns.get_loc(item)
+        return self._ixs(loc, axis=1)
+
+    # ----------------------------------------------------------------------
+    # Unsorted
+
+    @overload
+    def query(
+        self, expr: str, *, inplace: Literal[False] = ..., **kwargs
+    ) -> DataFrame: ...
+
+    @overload
+    def query(self, expr: str, *, inplace: Literal[True], **kwargs) -> None: ...
+
+    @overload
+    def query(
+        self, expr: str, *, inplace: bool = ..., **kwargs
+    ) -> DataFrame | None: ...
+
+    def query(self, expr: str, *, inplace: bool = False, **kwargs) -> DataFrame | None:
+        """
+        Query the columns of a DataFrame with a boolean expression.
+
+        .. warning::
+
+            This method can run arbitrary code which can make you vulnerable to code
+            injection if you pass user input to this function.
+
+        Parameters
+        ----------
+        expr : str
+            The query string to evaluate.
+
+            See the documentation for :func:`eval` for details of
+            supported operations and functions in the query string.
+
+            See the documentation for :meth:`DataFrame.eval` for details on
+            referring to column names and variables in the query string.
+        inplace : bool
+            Whether to modify the DataFrame rather than creating a new one.
+        **kwargs
+            See the documentation for :func:`eval` for complete details
+            on the keyword arguments accepted by :meth:`DataFrame.query`.
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame resulting from the provided query expression or
+            None if ``inplace=True``.
+
+        See Also
+        --------
+        eval : Evaluate a string describing operations on
+            DataFrame columns.
+        DataFrame.eval : Evaluate a string describing operations on
+            DataFrame columns.
+
+        Notes
+        -----
+        The result of the evaluation of this expression is first passed to
+        :attr:`DataFrame.loc` and if that fails because of a
+        multidimensional key (e.g., a DataFrame) then the result will be passed
+        to :meth:`DataFrame.__getitem__`.
+
+        This method uses the top-level :func:`eval` function to
+        evaluate the passed query.
+
+        The :meth:`~pandas.DataFrame.query` method uses a slightly
+        modified Python syntax by default. For example, the ``&`` and ``|``
+        (bitwise) operators have the precedence of their boolean cousins,
+        :keyword:`and` and :keyword:`or`. This *is* syntactically valid Python,
+        however the semantics are different.
+
+        You can change the semantics of the expression by passing the keyword
+        argument ``parser='python'``. This enforces the same semantics as
+        evaluation in Python space. Likewise, you can pass ``engine='python'``
+        to evaluate an expression using Python itself as a backend. This is not
+        recommended as it is inefficient compared to using ``numexpr`` as the
+        engine.
+
+        The :attr:`DataFrame.index` and
+        :attr:`DataFrame.columns` attributes of the
+        :class:`~pandas.DataFrame` instance are placed in the query namespace
+        by default, which allows you to treat both the index and columns of the
+        frame as a column in the frame.
+        The identifier ``index`` is used for the frame index; you can also
+        use the name of the index to identify it in a query. Please note that
+        Python keywords may not be used as identifiers.
+
+        For further details and examples see the ``query`` documentation in
+        :ref:`indexing <indexing.query>`.
+
+        *Backtick quoted variables*
+
+        Backtick quoted variables are parsed as literal Python code and
+        are converted internally to a Python valid identifier.
+        This can lead to the following problems.
+
+        During parsing a number of disallowed characters inside the backtick
+        quoted string are replaced by strings that are allowed as a Python identifier.
+        These characters include all operators in Python, the space character, the
+        question mark, the exclamation mark, the dollar sign, and the euro sign.
+
+        A backtick can be escaped by double backticks.
+
+        See also the `Python documentation about lexical analysis
+        <https://docs.python.org/3/reference/lexical_analysis.html>`__
+        in combination with the source code in :mod:`pandas.core.computation.parsing`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"A": range(1, 6), "B": range(10, 0, -2), "C&C": range(10, 5, -1)}
+        ... )
+        >>> df
+           A   B  C&C
+        0  1  10   10
+        1  2   8    9
+        2  3   6    8
+        3  4   4    7
+        4  5   2    6
+        >>> df.query("A > B")
+           A  B  C&C
+        4  5  2    6
+
+        The previous expression is equivalent to
+
+        >>> df[df.A > df.B]
+           A  B  C&C
+        4  5  2    6
+
+        For columns with spaces in their name, you can use backtick quoting.
+
+        >>> df.query("B == `C&C`")
+           A   B  C&C
+        0  1  10   10
+
+        The previous expression is equivalent to
+
+        >>> df[df.B == df["C&C"]]
+           A   B  C&C
+        0  1  10   10
+
+        Using local variable:
+
+        >>> local_var = 2
+        >>> df.query("A <= @local_var")
+        A   B  C&C
+        0  1  10   10
+        1  2   8    9
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if not isinstance(expr, str):
+            msg = f"expr must be a string to be evaluated, {type(expr)} given"
+            raise ValueError(msg)
+        kwargs["level"] = kwargs.pop("level", 0) + 1
+        kwargs["target"] = None
+
+        res = self.eval(expr, **kwargs)
+
+        try:
+            result = self.loc[res]
+        except ValueError:
+            # when res is multi-dimensional loc raises, but this is sometimes a
+            # valid query
+            result = self[res]
+
+        if inplace:
+            self._update_inplace(result)
+            return None
+        else:
+            return result
+
+    @overload
+    def eval(self, expr: str, *, inplace: Literal[False] = ..., **kwargs) -> Any: ...
+
+    @overload
+    def eval(self, expr: str, *, inplace: Literal[True], **kwargs) -> None: ...
+
+    def eval(self, expr: str, *, inplace: bool = False, **kwargs) -> Any | None:
+        """
+        Evaluate a string describing operations on DataFrame columns.
+
+        .. warning::
+
+            This method can run arbitrary code which can make you vulnerable to code
+            injection if you pass user input to this function.
+
+        Operates on columns only, not specific rows or elements.  This allows
+        `eval` to run arbitrary code, which can make you vulnerable to code
+        injection if you pass user input to this function.
+
+        Parameters
+        ----------
+        expr : str
+            The expression string to evaluate.
+
+            You can refer to variables
+            in the environment by prefixing them with an '@' character like
+            ``@a + b``.
+
+            You can refer to column names that are not valid Python variable names
+            by surrounding them in backticks. Thus, column names containing spaces
+            or punctuation (besides underscores) or starting with digits must be
+            surrounded by backticks. (For example, a column named "Area (cm^2)" would
+            be referenced as ```Area (cm^2)```). Column names which are Python keywords
+            (like "if", "for", "import", etc) cannot be used.
+
+            For example, if one of your columns is called ``a a`` and you want
+            to sum it with ``b``, your query should be ```a a` + b``.
+
+            See the documentation for :func:`eval` for full details of
+            supported operations and functions in the expression string.
+        inplace : bool, default False
+            If the expression contains an assignment, whether to perform the
+            operation inplace and mutate the existing DataFrame. Otherwise,
+            a new DataFrame is returned.
+        **kwargs
+            See the documentation for :func:`eval` for complete details
+            on the keyword arguments accepted by
+            :meth:`~pandas.DataFrame.eval`.
+
+        Returns
+        -------
+        ndarray, scalar, pandas object, or None
+            The result of the evaluation or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.query : Evaluates a boolean expression to query the columns
+            of a frame.
+        DataFrame.assign : Can evaluate an expression or function to create new
+            values for a column.
+        eval : Evaluate a Python expression as a string using various
+            backends.
+
+        Notes
+        -----
+        For more details see the API documentation for :func:`~eval`.
+        For detailed examples see :ref:`enhancing performance with eval
+        <enhancingperf.eval>`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"A": range(1, 6), "B": range(10, 0, -2), "C&C": range(10, 5, -1)}
+        ... )
+        >>> df
+           A   B  C&C
+        0  1  10   10
+        1  2   8    9
+        2  3   6    8
+        3  4   4    7
+        4  5   2    6
+        >>> df.eval("A + B")
+        0    11
+        1    10
+        2     9
+        3     8
+        4     7
+        dtype: int64
+
+        Assignment is allowed though by default the original DataFrame is not
+        modified.
+
+        >>> df.eval("D = A + B")
+           A   B  C&C   D
+        0  1  10   10  11
+        1  2   8    9  10
+        2  3   6    8   9
+        3  4   4    7   8
+        4  5   2    6   7
+        >>> df
+           A   B  C&C
+        0  1  10   10
+        1  2   8    9
+        2  3   6    8
+        3  4   4    7
+        4  5   2    6
+
+        Multiple columns can be assigned to using multi-line expressions:
+
+        >>> df.eval(
+        ...     '''
+        ... D = A + B
+        ... E = A - B
+        ... '''
+        ... )
+           A   B  C&C   D  E
+        0  1  10   10  11 -9
+        1  2   8    9  10 -6
+        2  3   6    8   9 -3
+        3  4   4    7   8  0
+        4  5   2    6   7  3
+
+        For columns with spaces or other disallowed characters in their name, you can
+        use backtick quoting.
+
+        >>> df.eval("B * `C&C`")
+        0    100
+        1     72
+        2     48
+        3     28
+        4     12
+
+        Local variables shall be explicitly referenced using ``@``
+        character in front of the name:
+
+        >>> local_var = 2
+        >>> df.eval("@local_var * A")
+        0     2
+        1     4
+        2     6
+        3     8
+        4    10
+        """
+        from pandas.core.computation.eval import eval as _eval
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        kwargs["level"] = kwargs.pop("level", 0) + 1
+        index_resolvers = self._get_index_resolvers()
+        column_resolvers = self._get_cleaned_column_resolvers()
+        resolvers = column_resolvers, index_resolvers
+        if "target" not in kwargs:
+            kwargs["target"] = self
+        kwargs["resolvers"] = tuple(kwargs.get("resolvers", ())) + resolvers
+
+        return _eval(expr, inplace=inplace, **kwargs)
+
+    def select_dtypes(self, include=None, exclude=None) -> DataFrame:
+        """
+        Return a subset of the DataFrame's columns based on the column dtypes.
+
+        This method allows for filtering columns based on their data types.
+        It is useful when working with heterogeneous DataFrames where operations
+        need to be performed on a specific subset of data types.
+
+        Parameters
+        ----------
+        include, exclude : scalar or list-like
+            A selection of dtypes or strings to be included/excluded. At least
+            one of these parameters must be supplied.
+
+        Returns
+        -------
+        DataFrame
+            The subset of the frame including the dtypes in ``include`` and
+            excluding the dtypes in ``exclude``.
+
+        Raises
+        ------
+        ValueError
+            * If both of ``include`` and ``exclude`` are empty
+            * If ``include`` and ``exclude`` have overlapping elements
+        TypeError
+            * If any kind of string dtype is passed in.
+
+        See Also
+        --------
+        DataFrame.dtypes: Return Series with the data type of each column.
+
+        Notes
+        -----
+        * To select all *numeric* types, use ``np.number`` or ``'number'``
+        * To select strings you must use the ``object`` dtype, but note that
+          this will return *all* object dtype columns. With
+          ``pd.options.future.infer_string`` enabled, using ``"str"`` will
+          work to select all string columns.
+        * See the `numpy dtype hierarchy
+          <https://numpy.org/doc/stable/reference/arrays.scalars.html>`__
+        * To select datetimes, use ``np.datetime64``, ``'datetime'`` or
+          ``'datetime64'``
+        * To select timedeltas, use ``np.timedelta64``, ``'timedelta'`` or
+          ``'timedelta64'``
+        * To select Pandas categorical dtypes, use ``'category'``
+        * To select Pandas datetimetz dtypes, use ``'datetimetz'``
+          or ``'datetime64[ns, tz]'``
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"a": [1, 2] * 3, "b": [True, False] * 3, "c": [1.0, 2.0] * 3}
+        ... )
+        >>> df
+                a      b  c
+        0       1   True  1.0
+        1       2  False  2.0
+        2       1   True  1.0
+        3       2  False  2.0
+        4       1   True  1.0
+        5       2  False  2.0
+
+        >>> df.select_dtypes(include="bool")
+           b
+        0  True
+        1  False
+        2  True
+        3  False
+        4  True
+        5  False
+
+        >>> df.select_dtypes(include=["float64"])
+           c
+        0  1.0
+        1  2.0
+        2  1.0
+        3  2.0
+        4  1.0
+        5  2.0
+
+        >>> df.select_dtypes(exclude=["int64"])
+               b    c
+        0   True  1.0
+        1  False  2.0
+        2   True  1.0
+        3  False  2.0
+        4   True  1.0
+        5  False  2.0
+        """
+        if not is_list_like(include):
+            include = (include,) if include is not None else ()
+        if not is_list_like(exclude):
+            exclude = (exclude,) if exclude is not None else ()
+
+        selection = (frozenset(include), frozenset(exclude))
+
+        if not any(selection):
+            raise ValueError("at least one of include or exclude must be nonempty")
+
+        # convert the myriad valid dtypes object to a single representation
+        def check_int_infer_dtype(dtypes):
+            converted_dtypes: list[type] = []
+            for dtype in dtypes:
+                # Numpy maps int to different types (int32, in64) on Windows and Linux
+                # see https://github.com/numpy/numpy/issues/9464
+                if (isinstance(dtype, str) and dtype == "int") or (dtype is int):
+                    converted_dtypes.append(np.int32)
+                    converted_dtypes.append(np.int64)
+                elif dtype == "float" or dtype is float:
+                    # GH#42452 : np.dtype("float") coerces to np.float64 from Numpy 1.20
+                    converted_dtypes.extend([np.float64, np.float32])
+                else:
+                    converted_dtypes.append(infer_dtype_from_object(dtype))
+            return frozenset(converted_dtypes)
+
+        include = check_int_infer_dtype(include)
+        exclude = check_int_infer_dtype(exclude)
+
+        for dtypes in (include, exclude):
+            invalidate_string_dtypes(dtypes)
+
+        # can't both include AND exclude!
+        if not include.isdisjoint(exclude):
+            raise ValueError(f"include and exclude overlap on {(include & exclude)}")
+
+        def dtype_predicate(dtype: DtypeObj, dtypes_set) -> bool:
+            # GH 46870: BooleanDtype._is_numeric == True but should be excluded
+            dtype = dtype if not isinstance(dtype, ArrowDtype) else dtype.numpy_dtype
+            return issubclass(dtype.type, tuple(dtypes_set)) or (
+                np.number in dtypes_set
+                and getattr(dtype, "_is_numeric", False)
+                and not is_bool_dtype(dtype)
+            )
+
+        def predicate(arr: ArrayLike) -> bool:
+            dtype = arr.dtype
+            if include:
+                if not dtype_predicate(dtype, include):
+                    return False
+
+            if exclude:
+                if dtype_predicate(dtype, exclude):
+                    return False
+
+            return True
+
+        mgr = self._mgr._get_data_subset(predicate).copy(deep=False)
+        return self._constructor_from_mgr(mgr, axes=mgr.axes).__finalize__(self)
+
+    def insert(
+        self,
+        loc: int,
+        column: Hashable,
+        value: object,
+        allow_duplicates: bool | lib.NoDefault = lib.no_default,
+    ) -> None:
+        """
+        Insert column into DataFrame at specified location.
+
+        Raises a ValueError if `column` is already contained in the DataFrame,
+        unless `allow_duplicates` is set to True.
+
+        Parameters
+        ----------
+        loc : int
+            Insertion index. Must verify 0 <= loc <= len(columns).
+        column : str, number, or hashable object
+            Label of the inserted column.
+        value : Scalar, Series, or array-like
+            Content of the inserted column.
+        allow_duplicates : bool, optional, default lib.no_default
+            Allow duplicate column labels to be created.
+
+        See Also
+        --------
+        Index.insert : Insert new item by index.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df
+           col1  col2
+        0     1     3
+        1     2     4
+        >>> df.insert(1, "newcol", [99, 99])
+        >>> df
+           col1  newcol  col2
+        0     1      99     3
+        1     2      99     4
+        >>> df.insert(0, "col1", [100, 100], allow_duplicates=True)
+        >>> df
+           col1  col1  newcol  col2
+        0   100     1      99     3
+        1   100     2      99     4
+
+        Notice that pandas uses index alignment in case of `value` from type `Series`:
+
+        >>> df.insert(0, "col0", pd.Series([5, 6], index=[1, 2]))
+        >>> df
+           col0  col1  col1  newcol  col2
+        0   NaN   100     1      99     3
+        1   5.0   100     2      99     4
+        """
+        if allow_duplicates is lib.no_default:
+            allow_duplicates = False
+        if allow_duplicates and not self.flags.allows_duplicate_labels:
+            raise ValueError(
+                "Cannot specify 'allow_duplicates=True' when "
+                "'self.flags.allows_duplicate_labels' is False."
+            )
+        if not allow_duplicates and column in self.columns:
+            # Should this be a different kind of error??
+            raise ValueError(f"cannot insert {column}, already exists")
+        if not is_integer(loc):
+            raise TypeError("loc must be int")
+        # convert non stdlib ints to satisfy typing checks
+        loc = int(loc)
+        if isinstance(value, DataFrame) and len(value.columns) > 1:
+            raise ValueError(
+                f"Expected a one-dimensional object, got a DataFrame with "
+                f"{len(value.columns)} columns instead."
+            )
+        elif isinstance(value, DataFrame):
+            value = value.iloc[:, 0]
+
+        value, refs = self._sanitize_column(value)
+        self._mgr.insert(loc, column, value, refs=refs)
+
+    def assign(self, **kwargs) -> DataFrame:
+        r"""
+        Assign new columns to a DataFrame.
+
+        Returns a new object with all original columns in addition to new ones.
+        Existing columns that are re-assigned will be overwritten.
+
+        Parameters
+        ----------
+        **kwargs : callable or Series
+            The column names are keywords. If the values are
+            callable, they are computed on the DataFrame and
+            assigned to the new columns. The callable must not
+            change input DataFrame (though pandas doesn't check it).
+            If the values are not callable, (e.g. a Series, scalar, or array),
+            they are simply assigned.
+
+        Returns
+        -------
+        DataFrame
+            A new DataFrame with the new columns in addition to
+            all the existing columns.
+
+        See Also
+        --------
+        DataFrame.loc : Select a subset of a DataFrame by labels.
+        DataFrame.iloc : Select a subset of a DataFrame by positions.
+
+        Notes
+        -----
+        Assigning multiple columns within the same ``assign`` is possible.
+        Later items in '\*\*kwargs' may refer to newly created or modified
+        columns in 'df'; items are computed and assigned into 'df' in order.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"temp_c": [17.0, 25.0]}, index=["Portland", "Berkeley"])
+        >>> df
+                  temp_c
+        Portland    17.0
+        Berkeley    25.0
+
+        Where the value is a callable, evaluated on `df`:
+
+        >>> df.assign(temp_f=lambda x: x.temp_c * 9 / 5 + 32)
+                  temp_c  temp_f
+        Portland    17.0    62.6
+        Berkeley    25.0    77.0
+
+        Alternatively, the same behavior can be achieved by directly
+        referencing an existing Series or sequence:
+
+        >>> df.assign(temp_f=df["temp_c"] * 9 / 5 + 32)
+                  temp_c  temp_f
+        Portland    17.0    62.6
+        Berkeley    25.0    77.0
+
+        You can create multiple columns within the same assign where one
+        of the columns depends on another one defined within the same assign:
+
+        >>> df.assign(
+        ...     temp_f=lambda x: x["temp_c"] * 9 / 5 + 32,
+        ...     temp_k=lambda x: (x["temp_f"] + 459.67) * 5 / 9,
+        ... )
+                  temp_c  temp_f  temp_k
+        Portland    17.0    62.6  290.15
+        Berkeley    25.0    77.0  298.15
+        """
+        data = self.copy(deep=False)
+
+        for k, v in kwargs.items():
+            data[k] = com.apply_if_callable(v, data)
+        return data
+
+    def _sanitize_column(self, value) -> tuple[ArrayLike, BlockValuesRefs | None]:
+        """
+        Ensures new columns (which go into the BlockManager as new blocks) are
+        always copied (or a reference is being tracked to them under CoW)
+        and converted into an array.
+
+        Parameters
+        ----------
+        value : scalar, Series, or array-like
+
+        Returns
+        -------
+        tuple of numpy.ndarray or ExtensionArray and optional BlockValuesRefs
+        """
+        self._ensure_valid_index(value)
+
+        # Using a DataFrame would mean coercing values to one dtype
+        assert not isinstance(value, DataFrame)
+        if is_dict_like(value):
+            if not isinstance(value, Series):
+                value = Series(value)
+            return _reindex_for_setitem(value, self.index)
+
+        if is_list_like(value):
+            com.require_length_match(value, self.index)
+        return sanitize_array(value, self.index, copy=True, allow_2d=True), None
+
+    @property
+    def _series(self):
+        return {item: self._ixs(idx, axis=1) for idx, item in enumerate(self.columns)}
+
+    # ----------------------------------------------------------------------
+    # Reindexing and alignment
+
+    def _reindex_multi(self, axes: dict[str, Index], fill_value) -> DataFrame:
+        """
+        We are guaranteed non-Nones in the axes.
+        """
+
+        new_index, row_indexer = self.index.reindex(axes["index"])
+        new_columns, col_indexer = self.columns.reindex(axes["columns"])
+
+        if row_indexer is not None and col_indexer is not None:
+            # Fastpath. By doing two 'take's at once we avoid making an
+            #  unnecessary copy.
+            # We only get here with `self._can_fast_transpose`, which (almost)
+            #  ensures that self.values is cheap. It may be worth making this
+            #  condition more specific.
+            indexer = row_indexer, col_indexer
+            new_values = take_2d_multi(self.values, indexer, fill_value=fill_value)
+            return self._constructor(
+                new_values, index=new_index, columns=new_columns, copy=False
+            )
+        else:
+            return self._reindex_with_indexers(
+                {0: [new_index, row_indexer], 1: [new_columns, col_indexer]},
+                fill_value=fill_value,
+            )
+
+    @Appender(
+        """
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+
+        Change the row labels.
+
+        >>> df.set_axis(['a', 'b', 'c'], axis='index')
+           A  B
+        a  1  4
+        b  2  5
+        c  3  6
+
+        Change the column labels.
+
+        >>> df.set_axis(['I', 'II'], axis='columns')
+           I  II
+        0  1   4
+        1  2   5
+        2  3   6
+        """
+    )
+    @Substitution(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+        extended_summary_sub=" column or",
+        axis_description_sub=", and 1 identifies the columns",
+        see_also_sub=" or columns",
+    )
+    @Appender(NDFrame.set_axis.__doc__)
+    def set_axis(
+        self,
+        labels,
+        *,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> DataFrame:
+        return super().set_axis(labels, axis=axis, copy=copy)
+
+    @doc(
+        NDFrame.reindex,
+        klass=_shared_doc_kwargs["klass"],
+        optional_reindex=_shared_doc_kwargs["optional_reindex"],
+    )
+    def reindex(
+        self,
+        labels=None,
+        *,
+        index=None,
+        columns=None,
+        axis: Axis | None = None,
+        method: ReindexMethod | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        level: Level | None = None,
+        fill_value: Scalar | None = np.nan,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> DataFrame:
+        return super().reindex(
+            labels=labels,
+            index=index,
+            columns=columns,
+            axis=axis,
+            method=method,
+            level=level,
+            fill_value=fill_value,
+            limit=limit,
+            tolerance=tolerance,
+            copy=copy,
+        )
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level = ...,
+        inplace: Literal[True],
+        errors: IgnoreRaise = ...,
+    ) -> None: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level = ...,
+        inplace: Literal[False] = ...,
+        errors: IgnoreRaise = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level = ...,
+        inplace: bool = ...,
+        errors: IgnoreRaise = ...,
+    ) -> DataFrame | None: ...
+
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = None,
+        *,
+        axis: Axis = 0,
+        index: IndexLabel | ListLike = None,
+        columns: IndexLabel | ListLike = None,
+        level: Level | None = None,
+        inplace: bool = False,
+        errors: IgnoreRaise = "raise",
+    ) -> DataFrame | None:
+        """
+        Drop specified labels from rows or columns.
+
+        Remove rows or columns by specifying label names and corresponding
+        axis, or by directly specifying index or column names. When using a
+        multi-index, labels on different levels can be removed by specifying
+        the level. See the :ref:`user guide <advanced.shown_levels>`
+        for more information about the now unused levels.
+
+        Parameters
+        ----------
+        labels : single label or iterable of labels
+            Index or column labels to drop. A tuple will be used as a single
+            label and not treated as an iterable.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Whether to drop labels from the index (0 or 'index') or
+            columns (1 or 'columns').
+        index : single label or iterable of labels
+            Alternative to specifying axis (``labels, axis=0``
+            is equivalent to ``index=labels``).
+        columns : single label or iterable of labels
+            Alternative to specifying axis (``labels, axis=1``
+            is equivalent to ``columns=labels``).
+        level : int or level name, optional
+            For MultiIndex, level from which the labels will be removed.
+        inplace : bool, default False
+            If False, return a copy. Otherwise, do operation
+            in place and return None.
+        errors : {'ignore', 'raise'}, default 'raise'
+            If 'ignore', suppress error and only existing labels are
+            dropped.
+
+        Returns
+        -------
+        DataFrame or None
+            Returns DataFrame or None DataFrame with the specified
+            index or column labels removed or None if inplace=True.
+
+        Raises
+        ------
+        KeyError
+            If any of the labels is not found in the selected axis.
+
+        See Also
+        --------
+        DataFrame.loc : Label-location based indexer for selection by label.
+        DataFrame.dropna : Return DataFrame with labels on given axis omitted
+            where (all or any) data are missing.
+        DataFrame.drop_duplicates : Return DataFrame with duplicate rows
+            removed, optionally only considering certain columns.
+        Series.drop : Return Series with specified index labels removed.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(np.arange(12).reshape(3, 4), columns=["A", "B", "C", "D"])
+        >>> df
+           A  B   C   D
+        0  0  1   2   3
+        1  4  5   6   7
+        2  8  9  10  11
+
+        Drop columns
+
+        >>> df.drop(["B", "C"], axis=1)
+           A   D
+        0  0   3
+        1  4   7
+        2  8  11
+
+        >>> df.drop(columns=["B", "C"])
+           A   D
+        0  0   3
+        1  4   7
+        2  8  11
+
+        Drop a row by index
+
+        >>> df.drop([0, 1])
+           A  B   C   D
+        2  8  9  10  11
+
+        Drop columns and/or rows of MultiIndex DataFrame
+
+        >>> midx = pd.MultiIndex(
+        ...     levels=[["llama", "cow", "falcon"], ["speed", "weight", "length"]],
+        ...     codes=[[0, 0, 0, 1, 1, 1, 2, 2, 2], [0, 1, 2, 0, 1, 2, 0, 1, 2]],
+        ... )
+        >>> df = pd.DataFrame(
+        ...     index=midx,
+        ...     columns=["big", "small"],
+        ...     data=[
+        ...         [45, 30],
+        ...         [200, 100],
+        ...         [1.5, 1],
+        ...         [30, 20],
+        ...         [250, 150],
+        ...         [1.5, 0.8],
+        ...         [320, 250],
+        ...         [1, 0.8],
+        ...         [0.3, 0.2],
+        ...     ],
+        ... )
+        >>> df
+                        big     small
+        llama   speed   45.0    30.0
+                weight  200.0   100.0
+                length  1.5     1.0
+        cow     speed   30.0    20.0
+                weight  250.0   150.0
+                length  1.5     0.8
+        falcon  speed   320.0   250.0
+                weight  1.0     0.8
+                length  0.3     0.2
+
+        Drop a specific index combination from the MultiIndex
+        DataFrame, i.e., drop the combination ``'falcon'`` and
+        ``'weight'``, which deletes only the corresponding row
+
+        >>> df.drop(index=("falcon", "weight"))
+                        big     small
+        llama   speed   45.0    30.0
+                weight  200.0   100.0
+                length  1.5     1.0
+        cow     speed   30.0    20.0
+                weight  250.0   150.0
+                length  1.5     0.8
+        falcon  speed   320.0   250.0
+                length  0.3     0.2
+
+        >>> df.drop(index="cow", columns="small")
+                        big
+        llama   speed   45.0
+                weight  200.0
+                length  1.5
+        falcon  speed   320.0
+                weight  1.0
+                length  0.3
+
+        >>> df.drop(index="length", level=1)
+                        big     small
+        llama   speed   45.0    30.0
+                weight  200.0   100.0
+        cow     speed   30.0    20.0
+                weight  250.0   150.0
+        falcon  speed   320.0   250.0
+                weight  1.0     0.8
+        """
+        return super().drop(
+            labels=labels,
+            axis=axis,
+            index=index,
+            columns=columns,
+            level=level,
+            inplace=inplace,
+            errors=errors,
+        )
+
+    @overload
+    def rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: Literal[True],
+        level: Level = ...,
+        errors: IgnoreRaise = ...,
+    ) -> None: ...
+
+    @overload
+    def rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: Literal[False] = ...,
+        level: Level = ...,
+        errors: IgnoreRaise = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = ...,
+        level: Level = ...,
+        errors: IgnoreRaise = ...,
+    ) -> DataFrame | None: ...
+
+    def rename(
+        self,
+        mapper: Renamer | None = None,
+        *,
+        index: Renamer | None = None,
+        columns: Renamer | None = None,
+        axis: Axis | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = False,
+        level: Level | None = None,
+        errors: IgnoreRaise = "ignore",
+    ) -> DataFrame | None:
+        """
+        Rename columns or index labels.
+
+        Function / dict values must be unique (1-to-1). Labels not contained in
+        a dict / Series will be left as-is. Extra labels listed don't throw an
+        error.
+
+        See the :ref:`user guide <basics.rename>` for more.
+
+        Parameters
+        ----------
+        mapper : dict-like or function
+            Dict-like or function transformations to apply to
+            that axis' values. Use either ``mapper`` and ``axis`` to
+            specify the axis to target with ``mapper``, or ``index`` and
+            ``columns``.
+        index : dict-like or function
+            Alternative to specifying axis (``mapper, axis=0``
+            is equivalent to ``index=mapper``).
+        columns : dict-like or function
+            Alternative to specifying axis (``mapper, axis=1``
+            is equivalent to ``columns=mapper``).
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Axis to target with ``mapper``. Can be either the axis name
+            ('index', 'columns') or number (0, 1). The default is 'index'.
+        copy : bool, default False
+            Also copy underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        inplace : bool, default False
+            Whether to modify the DataFrame rather than creating a new one.
+            If True then value of copy is ignored.
+        level : int or level name, default None
+            In case of a MultiIndex, only rename labels in the specified
+            level.
+        errors : {'ignore', 'raise'}, default 'ignore'
+            If 'raise', raise a `KeyError` when a dict-like `mapper`, `index`,
+            or `columns` contains labels that are not present in the Index
+            being transformed.
+            If 'ignore', existing keys will be renamed and extra keys will be
+            ignored.
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with the renamed axis labels or None if ``inplace=True``.
+
+        Raises
+        ------
+        KeyError
+            If any of the labels is not found in the selected axis and
+            "errors='raise'".
+
+        See Also
+        --------
+        DataFrame.rename_axis : Set the name of the axis.
+
+        Examples
+        --------
+        ``DataFrame.rename`` supports two calling conventions
+
+        * ``(index=index_mapper, columns=columns_mapper, ...)``
+        * ``(mapper, axis={'index', 'columns'}, ...)``
+
+        We *highly* recommend using keyword arguments to clarify your
+        intent.
+
+        Rename columns using a mapping:
+
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        >>> df.rename(columns={"A": "a", "B": "c"})
+           a  c
+        0  1  4
+        1  2  5
+        2  3  6
+
+        Rename index using a mapping:
+
+        >>> df.rename(index={0: "x", 1: "y", 2: "z"})
+           A  B
+        x  1  4
+        y  2  5
+        z  3  6
+
+        Cast index labels to a different type:
+
+        >>> df.index
+        RangeIndex(start=0, stop=3, step=1)
+        >>> df.rename(index=str).index
+        Index(['0', '1', '2'], dtype='object')
+
+        >>> df.rename(columns={"A": "a", "B": "b", "C": "c"}, errors="raise")
+        Traceback (most recent call last):
+        KeyError: ['C'] not found in axis
+
+        Using axis-style parameters:
+
+        >>> df.rename(str.lower, axis="columns")
+           a  b
+        0  1  4
+        1  2  5
+        2  3  6
+
+        >>> df.rename({1: 2, 2: 4}, axis="index")
+           A  B
+        0  1  4
+        2  2  5
+        4  3  6
+        """
+        self._check_copy_deprecation(copy)
+        return super()._rename(
+            mapper=mapper,
+            index=index,
+            columns=columns,
+            axis=axis,
+            inplace=inplace,
+            level=level,
+            errors=errors,
+        )
+
+    def pop(self, item: Hashable) -> Series:
+        """
+        Return item and drop it from DataFrame. Raise KeyError if not found.
+
+        Parameters
+        ----------
+        item : label
+            Label of column to be popped.
+
+        Returns
+        -------
+        Series
+            Series representing the item that is dropped.
+
+        See Also
+        --------
+        DataFrame.drop: Drop specified labels from rows or columns.
+        DataFrame.drop_duplicates: Return DataFrame with duplicate rows removed.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("falcon", "bird", 389.0),
+        ...         ("parrot", "bird", 24.0),
+        ...         ("lion", "mammal", 80.5),
+        ...         ("monkey", "mammal", np.nan),
+        ...     ],
+        ...     columns=("name", "class", "max_speed"),
+        ... )
+        >>> df
+             name   class  max_speed
+        0  falcon    bird      389.0
+        1  parrot    bird       24.0
+        2    lion  mammal       80.5
+        3  monkey  mammal        NaN
+
+        >>> df.pop("class")
+        0      bird
+        1      bird
+        2    mammal
+        3    mammal
+        Name: class, dtype: object
+
+        >>> df
+             name  max_speed
+        0  falcon      389.0
+        1  parrot       24.0
+        2    lion       80.5
+        3  monkey        NaN
+        """
+        return super().pop(item=item)
+
+    @overload
+    def _replace_columnwise(
+        self, mapping: dict[Hashable, tuple[Any, Any]], inplace: Literal[True], regex
+    ) -> None: ...
+
+    @overload
+    def _replace_columnwise(
+        self, mapping: dict[Hashable, tuple[Any, Any]], inplace: Literal[False], regex
+    ) -> Self: ...
+
+    def _replace_columnwise(
+        self, mapping: dict[Hashable, tuple[Any, Any]], inplace: bool, regex
+    ) -> Self | None:
+        """
+        Dispatch to Series.replace column-wise.
+
+        Parameters
+        ----------
+        mapping : dict
+            of the form {col: (target, value)}
+        inplace : bool
+        regex : bool or same types as `to_replace` in DataFrame.replace
+
+        Returns
+        -------
+        DataFrame or None
+        """
+        # Operate column-wise
+        res = self if inplace else self.copy(deep=False)
+        ax = self.columns
+
+        for i, ax_value in enumerate(ax):
+            if ax_value in mapping:
+                ser = self.iloc[:, i]
+
+                target, value = mapping[ax_value]
+                newobj = ser.replace(target, value, regex=regex)
+
+                res._iset_item(i, newobj, inplace=inplace)
+
+        if inplace:
+            return None
+        return res.__finalize__(self)
+
+    @doc(NDFrame.shift, klass=_shared_doc_kwargs["klass"])
+    def shift(
+        self,
+        periods: int | Sequence[int] = 1,
+        freq: Frequency | None = None,
+        axis: Axis = 0,
+        fill_value: Hashable = lib.no_default,
+        suffix: str | None = None,
+    ) -> DataFrame:
+        if freq is not None and fill_value is not lib.no_default:
+            # GH#53832
+            raise ValueError(
+                "Passing a 'freq' together with a 'fill_value' is not allowed."
+            )
+
+        if self.empty and freq is None:
+            return self.copy()
+
+        axis = self._get_axis_number(axis)
+
+        if is_list_like(periods):
+            periods = cast(Sequence, periods)
+            if axis == 1:
+                raise ValueError(
+                    "If `periods` contains multiple shifts, `axis` cannot be 1."
+                )
+            if len(periods) == 0:
+                raise ValueError("If `periods` is an iterable, it cannot be empty.")
+            from pandas.core.reshape.concat import concat
+
+            shifted_dataframes = []
+            for period in periods:
+                if not is_integer(period):
+                    raise TypeError(
+                        f"Periods must be integer, but {period} is {type(period)}."
+                    )
+                period = cast(int, period)
+                shifted_dataframes.append(
+                    super()
+                    .shift(periods=period, freq=freq, axis=axis, fill_value=fill_value)
+                    .add_suffix(f"{suffix}_{period}" if suffix else f"_{period}")
+                )
+            return concat(shifted_dataframes, axis=1)
+        elif suffix:
+            raise ValueError("Cannot specify `suffix` if `periods` is an int.")
+        periods = cast(int, periods)
+
+        ncols = len(self.columns)
+        if axis == 1 and periods != 0 and ncols > 0 and freq is None:
+            if fill_value is lib.no_default:
+                # We will infer fill_value to match the closest column
+
+                # Use a column that we know is valid for our column's dtype GH#38434
+                label = self.columns[0]
+
+                if periods > 0:
+                    result = self.iloc[:, :-periods]
+                    for col in range(min(ncols, abs(periods))):
+                        # TODO(EA2D): doing this in a loop unnecessary with 2D EAs
+                        # Define filler inside loop so we get a copy
+                        filler = self.iloc[:, 0].shift(len(self))
+                        result.insert(0, label, filler, allow_duplicates=True)
+                else:
+                    result = self.iloc[:, -periods:]
+                    for col in range(min(ncols, abs(periods))):
+                        # Define filler inside loop so we get a copy
+                        filler = self.iloc[:, -1].shift(len(self))
+                        result.insert(
+                            len(result.columns), label, filler, allow_duplicates=True
+                        )
+
+                result.columns = self.columns.copy()
+                return result
+            elif len(self._mgr.blocks) > 1 or (
+                # If we only have one block and we know that we can't
+                #  keep the same dtype (i.e. the _can_hold_element check)
+                #  then we can go through the reindex_indexer path
+                #  (and avoid casting logic in the Block method).
+                not can_hold_element(self._mgr.blocks[0].values, fill_value)
+            ):
+                # GH#35488 we need to watch out for multi-block cases
+                # We only get here with fill_value not-lib.no_default
+                nper = abs(periods)
+                nper = min(nper, ncols)
+                if periods > 0:
+                    indexer = np.array(
+                        [-1] * nper + list(range(ncols - periods)), dtype=np.intp
+                    )
+                else:
+                    indexer = np.array(
+                        list(range(nper, ncols)) + [-1] * nper, dtype=np.intp
+                    )
+                mgr = self._mgr.reindex_indexer(
+                    self.columns,
+                    indexer,
+                    axis=0,
+                    fill_value=fill_value,
+                    allow_dups=True,
+                )
+                res_df = self._constructor_from_mgr(mgr, axes=mgr.axes)
+                return res_df.__finalize__(self, method="shift")
+            else:
+                return self.T.shift(periods=periods, fill_value=fill_value).T
+
+        return super().shift(
+            periods=periods, freq=freq, axis=axis, fill_value=fill_value
+        )
+
+    @overload
+    def set_index(
+        self,
+        keys,
+        *,
+        drop: bool = ...,
+        append: bool = ...,
+        inplace: Literal[False] = ...,
+        verify_integrity: bool = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def set_index(
+        self,
+        keys,
+        *,
+        drop: bool = ...,
+        append: bool = ...,
+        inplace: Literal[True],
+        verify_integrity: bool = ...,
+    ) -> None: ...
+
+    def set_index(
+        self,
+        keys,
+        *,
+        drop: bool = True,
+        append: bool = False,
+        inplace: bool = False,
+        verify_integrity: bool = False,
+    ) -> DataFrame | None:
+        """
+        Set the DataFrame index using existing columns.
+
+        Set the DataFrame index (row labels) using one or more existing
+        columns or arrays (of the correct length). The index can replace the
+        existing index or expand on it.
+
+        Parameters
+        ----------
+        keys : label or array-like or list of labels/arrays
+            This parameter can be either a single column key, a single array of
+            the same length as the calling DataFrame, or a list containing an
+            arbitrary combination of column keys and arrays. Here, "array"
+            encompasses :class:`Series`, :class:`Index`, ``np.ndarray``, and
+            instances of :class:`~collections.abc.Iterator`.
+        drop : bool, default True
+            Delete columns to be used as the new index.
+        append : bool, default False
+            Whether to append columns to existing index.
+        inplace : bool, default False
+            Whether to modify the DataFrame rather than creating a new one.
+        verify_integrity : bool, default False
+            Check the new index for duplicates. Otherwise defer the check until
+            necessary. Setting to False will improve the performance of this
+            method.
+
+        Returns
+        -------
+        DataFrame or None
+            Changed row labels or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.reset_index : Opposite of set_index.
+        DataFrame.reindex : Change to new indices or expand indices.
+        DataFrame.reindex_like : Change to same indices as other DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "month": [1, 4, 7, 10],
+        ...         "year": [2012, 2014, 2013, 2014],
+        ...         "sale": [55, 40, 84, 31],
+        ...     }
+        ... )
+        >>> df
+           month  year  sale
+        0      1  2012    55
+        1      4  2014    40
+        2      7  2013    84
+        3     10  2014    31
+
+        Set the index to become the 'month' column:
+
+        >>> df.set_index("month")
+               year  sale
+        month
+        1      2012    55
+        4      2014    40
+        7      2013    84
+        10     2014    31
+
+        Create a MultiIndex using columns 'year' and 'month':
+
+        >>> df.set_index(["year", "month"])
+                    sale
+        year  month
+        2012  1     55
+        2014  4     40
+        2013  7     84
+        2014  10    31
+
+        Create a MultiIndex using an Index and a column:
+
+        >>> df.set_index([pd.Index([1, 2, 3, 4]), "year"])
+                 month  sale
+           year
+        1  2012  1      55
+        2  2014  4      40
+        3  2013  7      84
+        4  2014  10     31
+
+        Create a MultiIndex using two Series:
+
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> df.set_index([s, s**2])
+              month  year  sale
+        1 1       1  2012    55
+        2 4       4  2014    40
+        3 9       7  2013    84
+        4 16     10  2014    31
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        self._check_inplace_and_allows_duplicate_labels(inplace)
+        if not isinstance(keys, list):
+            keys = [keys]
+
+        err_msg = (
+            'The parameter "keys" may be a column key, one-dimensional '
+            "array, or a list containing only valid column keys and "
+            "one-dimensional arrays."
+        )
+
+        missing: list[Hashable] = []
+        for col in keys:
+            if isinstance(col, (Index, Series, np.ndarray, list, abc.Iterator)):
+                # arrays are fine as long as they are one-dimensional
+                # iterators get converted to list below
+                if getattr(col, "ndim", 1) != 1:
+                    raise ValueError(err_msg)
+            else:
+                # everything else gets tried as a key; see GH 24969
+                try:
+                    found = col in self.columns
+                except TypeError as err:
+                    raise TypeError(
+                        f"{err_msg}. Received column of type {type(col)}"
+                    ) from err
+                else:
+                    if not found:
+                        missing.append(col)
+
+        if missing:
+            raise KeyError(f"None of {missing} are in the columns")
+
+        if inplace:
+            frame = self
+        else:
+            frame = self.copy(deep=False)
+
+        arrays: list[Index] = []
+        names: list[Hashable] = []
+        if append:
+            names = list(self.index.names)
+            if isinstance(self.index, MultiIndex):
+                arrays.extend(
+                    self.index._get_level_values(i) for i in range(self.index.nlevels)
+                )
+            else:
+                arrays.append(self.index)
+
+        to_remove: set[Hashable] = set()
+        for col in keys:
+            if isinstance(col, MultiIndex):
+                arrays.extend(col._get_level_values(n) for n in range(col.nlevels))
+                names.extend(col.names)
+            elif isinstance(col, (Index, Series)):
+                # if Index then not MultiIndex (treated above)
+
+                # error: Argument 1 to "append" of "list" has incompatible type
+                #  "Union[Index, Series]"; expected "Index"
+                arrays.append(col)  # type: ignore[arg-type]
+                names.append(col.name)
+            elif isinstance(col, (list, np.ndarray)):
+                # error: Argument 1 to "append" of "list" has incompatible type
+                # "Union[List[Any], ndarray]"; expected "Index"
+                arrays.append(col)  # type: ignore[arg-type]
+                names.append(None)
+            elif isinstance(col, abc.Iterator):
+                # error: Argument 1 to "append" of "list" has incompatible type
+                # "List[Any]"; expected "Index"
+                arrays.append(list(col))  # type: ignore[arg-type]
+                names.append(None)
+            # from here, col can only be a column label
+            else:
+                arrays.append(frame[col])
+                names.append(col)
+                if drop:
+                    to_remove.add(col)
+
+            if len(arrays[-1]) != len(self):
+                # check newest element against length of calling frame, since
+                # ensure_index_from_sequences would not raise for append=False.
+                raise ValueError(
+                    f"Length mismatch: Expected {len(self)} rows, "
+                    f"received array of length {len(arrays[-1])}"
+                )
+
+        index = ensure_index_from_sequences(arrays, names)
+
+        if verify_integrity and not index.is_unique:
+            duplicates = index[index.duplicated()].unique()
+            raise ValueError(f"Index has duplicate keys: {duplicates}")
+
+        # use set to handle duplicate column names gracefully in case of drop
+        for c in to_remove:
+            del frame[c]
+
+        # clear up memory usage
+        index._cleanup()
+
+        frame.index = index
+
+        if not inplace:
+            return frame
+        return None
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: bool = ...,
+        inplace: Literal[False] = ...,
+        col_level: Hashable = ...,
+        col_fill: Hashable = ...,
+        allow_duplicates: bool | lib.NoDefault = ...,
+        names: Hashable | Sequence[Hashable] | None = None,
+    ) -> DataFrame: ...
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: bool = ...,
+        inplace: Literal[True],
+        col_level: Hashable = ...,
+        col_fill: Hashable = ...,
+        allow_duplicates: bool | lib.NoDefault = ...,
+        names: Hashable | Sequence[Hashable] | None = None,
+    ) -> None: ...
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: bool = ...,
+        inplace: bool = ...,
+        col_level: Hashable = ...,
+        col_fill: Hashable = ...,
+        allow_duplicates: bool | lib.NoDefault = ...,
+        names: Hashable | Sequence[Hashable] | None = None,
+    ) -> DataFrame | None: ...
+
+    def reset_index(
+        self,
+        level: IndexLabel | None = None,
+        *,
+        drop: bool = False,
+        inplace: bool = False,
+        col_level: Hashable = 0,
+        col_fill: Hashable = "",
+        allow_duplicates: bool | lib.NoDefault = lib.no_default,
+        names: Hashable | Sequence[Hashable] | None = None,
+    ) -> DataFrame | None:
+        """
+        Reset the index, or a level of it.
+
+        Reset the index of the DataFrame, and use the default one instead.
+        If the DataFrame has a MultiIndex, this method can remove one or more
+        levels.
+
+        Parameters
+        ----------
+        level : int, str, tuple, or list, default None
+            Only remove the given levels from the index. Removes all levels by
+            default.
+        drop : bool, default False
+            Do not try to insert index into dataframe columns. This resets
+            the index to the default integer index.
+        inplace : bool, default False
+            Whether to modify the DataFrame rather than creating a new one.
+        col_level : int or str, default 0
+            If the columns have multiple levels, determines which level the
+            labels are inserted into. By default it is inserted into the first
+            level.
+        col_fill : object, default ''
+            If the columns have multiple levels, determines how the other
+            levels are named. If None then the index name is repeated.
+        allow_duplicates : bool, optional, default lib.no_default
+            Allow duplicate column labels to be created.
+
+            .. versionadded:: 1.5.0
+
+        names : int, str or 1-dimensional list, default None
+            Using the given string, rename the DataFrame column which contains the
+            index data. If the DataFrame has a MultiIndex, this has to be a list
+            with length equal to the number of levels.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with the new index or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.set_index : Opposite of reset_index.
+        DataFrame.reindex : Change to new indices or expand indices.
+        DataFrame.reindex_like : Change to same indices as other DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [("bird", 389.0), ("bird", 24.0), ("mammal", 80.5), ("mammal", np.nan)],
+        ...     index=["falcon", "parrot", "lion", "monkey"],
+        ...     columns=("class", "max_speed"),
+        ... )
+        >>> df
+                 class  max_speed
+        falcon    bird      389.0
+        parrot    bird       24.0
+        lion    mammal       80.5
+        monkey  mammal        NaN
+
+        When we reset the index, the old index is added as a column, and a
+        new sequential index is used:
+
+        >>> df.reset_index()
+            index   class  max_speed
+        0  falcon    bird      389.0
+        1  parrot    bird       24.0
+        2    lion  mammal       80.5
+        3  monkey  mammal        NaN
+
+        We can use the `drop` parameter to avoid the old index being added as
+        a column:
+
+        >>> df.reset_index(drop=True)
+            class  max_speed
+        0    bird      389.0
+        1    bird       24.0
+        2  mammal       80.5
+        3  mammal        NaN
+
+        You can also use `reset_index` with `MultiIndex`.
+
+        >>> index = pd.MultiIndex.from_tuples(
+        ...     [
+        ...         ("bird", "falcon"),
+        ...         ("bird", "parrot"),
+        ...         ("mammal", "lion"),
+        ...         ("mammal", "monkey"),
+        ...     ],
+        ...     names=["class", "name"],
+        ... )
+        >>> columns = pd.MultiIndex.from_tuples([("speed", "max"), ("species", "type")])
+        >>> df = pd.DataFrame(
+        ...     [(389.0, "fly"), (24.0, "fly"), (80.5, "run"), (np.nan, "jump")],
+        ...     index=index,
+        ...     columns=columns,
+        ... )
+        >>> df
+                       speed species
+                         max    type
+        class  name
+        bird   falcon  389.0     fly
+               parrot   24.0     fly
+        mammal lion     80.5     run
+               monkey    NaN    jump
+
+        Using the `names` parameter, choose a name for the index column:
+
+        >>> df.reset_index(names=["classes", "names"])
+          classes   names  speed species
+                             max    type
+        0    bird  falcon  389.0     fly
+        1    bird  parrot   24.0     fly
+        2  mammal    lion   80.5     run
+        3  mammal  monkey    NaN    jump
+
+        If the index has multiple levels, we can reset a subset of them:
+
+        >>> df.reset_index(level="class")
+                 class  speed species
+                          max    type
+        name
+        falcon    bird  389.0     fly
+        parrot    bird   24.0     fly
+        lion    mammal   80.5     run
+        monkey  mammal    NaN    jump
+
+        If we are not dropping the index, by default, it is placed in the top
+        level. We can place it in another level:
+
+        >>> df.reset_index(level="class", col_level=1)
+                        speed species
+                 class    max    type
+        name
+        falcon    bird  389.0     fly
+        parrot    bird   24.0     fly
+        lion    mammal   80.5     run
+        monkey  mammal    NaN    jump
+
+        When the index is inserted under another level, we can specify under
+        which one with the parameter `col_fill`:
+
+        >>> df.reset_index(level="class", col_level=1, col_fill="species")
+                      species  speed species
+                        class    max    type
+        name
+        falcon           bird  389.0     fly
+        parrot           bird   24.0     fly
+        lion           mammal   80.5     run
+        monkey         mammal    NaN    jump
+
+        If we specify a nonexistent level for `col_fill`, it is created:
+
+        >>> df.reset_index(level="class", col_level=1, col_fill="genus")
+                        genus  speed species
+                        class    max    type
+        name
+        falcon           bird  389.0     fly
+        parrot           bird   24.0     fly
+        lion           mammal   80.5     run
+        monkey         mammal    NaN    jump
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        self._check_inplace_and_allows_duplicate_labels(inplace)
+        if inplace:
+            new_obj = self
+        else:
+            new_obj = self.copy(deep=False)
+        if allow_duplicates is not lib.no_default:
+            allow_duplicates = validate_bool_kwarg(allow_duplicates, "allow_duplicates")
+
+        new_index = default_index(len(new_obj))
+        if level is not None:
+            if not isinstance(level, (tuple, list)):
+                level = [level]
+            level = [self.index._get_level_number(lev) for lev in level]
+            if len(level) < self.index.nlevels:
+                new_index = self.index.droplevel(level)
+
+        if not drop:
+            to_insert: Iterable[tuple[Any, Any | None]]
+
+            default = "index" if "index" not in self else "level_0"
+            names = self.index._get_default_index_names(names, default)
+
+            if isinstance(self.index, MultiIndex):
+                to_insert = zip(reversed(self.index.levels), reversed(self.index.codes))
+            else:
+                to_insert = ((self.index, None),)
+
+            multi_col = isinstance(self.columns, MultiIndex)
+            for j, (lev, lab) in enumerate(to_insert, start=1):
+                i = self.index.nlevels - j
+                if level is not None and i not in level:
+                    continue
+                name = names[i]
+                if multi_col:
+                    col_name = list(name) if isinstance(name, tuple) else [name]
+                    if col_fill is None:
+                        if len(col_name) not in (1, self.columns.nlevels):
+                            raise ValueError(
+                                "col_fill=None is incompatible "
+                                f"with incomplete column name {name}"
+                            )
+                        col_fill = col_name[0]
+
+                    lev_num = self.columns._get_level_number(col_level)
+                    name_lst = [col_fill] * lev_num + col_name
+                    missing = self.columns.nlevels - len(name_lst)
+                    name_lst += [col_fill] * missing
+                    name = tuple(name_lst)
+
+                # to ndarray and maybe infer different dtype
+                level_values = lev._values
+                if level_values.dtype == np.object_:
+                    level_values = lib.maybe_convert_objects(level_values)
+
+                if lab is not None:
+                    # if we have the codes, extract the values with a mask
+                    level_values = algorithms.take(
+                        level_values, lab, allow_fill=True, fill_value=lev._na_value
+                    )
+
+                new_obj.insert(
+                    0,
+                    name,
+                    level_values,
+                    allow_duplicates=allow_duplicates,
+                )
+
+        new_obj.index = new_index
+        if not inplace:
+            return new_obj
+
+        return None
+
+    # ----------------------------------------------------------------------
+    # Reindex-based selection methods
+
+    @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])
+    def isna(self) -> DataFrame:
+        res_mgr = self._mgr.isna(func=isna)
+        result = self._constructor_from_mgr(res_mgr, axes=res_mgr.axes)
+        return result.__finalize__(self, method="isna")
+
+    @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])
+    def isnull(self) -> DataFrame:
+        """
+        DataFrame.isnull is an alias for DataFrame.isna.
+        """
+        return self.isna()
+
+    @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])
+    def notna(self) -> DataFrame:
+        return ~self.isna()
+
+    @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])
+    def notnull(self) -> DataFrame:
+        """
+        DataFrame.notnull is an alias for DataFrame.notna.
+        """
+        return ~self.isna()
+
+    @overload
+    def dropna(
+        self,
+        *,
+        axis: Axis = ...,
+        how: AnyAll | lib.NoDefault = ...,
+        thresh: int | lib.NoDefault = ...,
+        subset: IndexLabel = ...,
+        inplace: Literal[False] = ...,
+        ignore_index: bool = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def dropna(
+        self,
+        *,
+        axis: Axis = ...,
+        how: AnyAll | lib.NoDefault = ...,
+        thresh: int | lib.NoDefault = ...,
+        subset: IndexLabel = ...,
+        inplace: Literal[True],
+        ignore_index: bool = ...,
+    ) -> None: ...
+
+    def dropna(
+        self,
+        *,
+        axis: Axis = 0,
+        how: AnyAll | lib.NoDefault = lib.no_default,
+        thresh: int | lib.NoDefault = lib.no_default,
+        subset: IndexLabel | AnyArrayLike | None = None,
+        inplace: bool = False,
+        ignore_index: bool = False,
+    ) -> DataFrame | None:
+        """
+        Remove missing values.
+
+        See the :ref:`User Guide <missing_data>` for more on which values are
+        considered missing, and how to work with missing data.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Determine if rows or columns which contain missing values are
+            removed.
+
+            * 0, or 'index' : Drop rows which contain missing values.
+            * 1, or 'columns' : Drop columns which contain missing value.
+
+            Only a single axis is allowed.
+
+        how : {'any', 'all'}, default 'any'
+            Determine if row or column is removed from DataFrame, when we have
+            at least one NA or all NA.
+
+            * 'any' : If any NA values are present, drop that row or column.
+            * 'all' : If all values are NA, drop that row or column.
+
+        thresh : int, optional
+            Require that many non-NA values. Cannot be combined with how.
+        subset : column label or iterable of labels, optional
+            Labels along other axis to consider, e.g. if you are dropping rows
+            these would be a list of columns to include.
+        inplace : bool, default False
+            Whether to modify the DataFrame rather than creating a new one.
+        ignore_index : bool, default ``False``
+            If ``True``, the resulting axis will be labeled 0, 1, …, n - 1.
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with NA entries dropped from it or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.isna: Indicate missing values.
+        DataFrame.notna : Indicate existing (non-missing) values.
+        DataFrame.fillna : Replace missing values.
+        Series.dropna : Drop missing values.
+        Index.dropna : Drop missing indices.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "name": ["Alfred", "Batman", "Catwoman"],
+        ...         "toy": [np.nan, "Batmobile", "Bullwhip"],
+        ...         "born": [pd.NaT, pd.Timestamp("1940-04-25"), pd.NaT],
+        ...     }
+        ... )
+        >>> df
+               name        toy       born
+        0    Alfred        NaN        NaT
+        1    Batman  Batmobile 1940-04-25
+        2  Catwoman   Bullwhip        NaT
+
+        Drop the rows where at least one element is missing.
+
+        >>> df.dropna()
+             name        toy       born
+        1  Batman  Batmobile 1940-04-25
+
+        Drop the columns where at least one element is missing.
+
+        >>> df.dropna(axis="columns")
+               name
+        0    Alfred
+        1    Batman
+        2  Catwoman
+
+        Drop the rows where all elements are missing.
+
+        >>> df.dropna(how="all")
+               name        toy       born
+        0    Alfred        NaN        NaT
+        1    Batman  Batmobile 1940-04-25
+        2  Catwoman   Bullwhip        NaT
+
+        Keep only the rows with at least 2 non-NA values.
+
+        >>> df.dropna(thresh=2)
+               name        toy       born
+        1    Batman  Batmobile 1940-04-25
+        2  Catwoman   Bullwhip        NaT
+
+        Define in which columns to look for missing values.
+
+        >>> df.dropna(subset=["name", "toy"])
+               name        toy       born
+        1    Batman  Batmobile 1940-04-25
+        2  Catwoman   Bullwhip        NaT
+        """
+        if (how is not lib.no_default) and (thresh is not lib.no_default):
+            raise TypeError(
+                "You cannot set both the how and thresh arguments at the same time."
+            )
+
+        if how is lib.no_default:
+            how = "any"
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if isinstance(axis, (tuple, list)):
+            # GH20987
+            raise TypeError("supplying multiple axes to axis is no longer supported.")
+
+        axis = self._get_axis_number(axis)
+        agg_axis = 1 - axis
+
+        agg_obj = self
+        if subset is not None:
+            # subset needs to be list
+            if not is_list_like(subset):
+                subset = [cast(Hashable, subset)]
+            ax = self._get_axis(agg_axis)
+            indices = ax.get_indexer_for(subset)
+            check = indices == -1
+            if check.any():
+                raise KeyError(np.array(subset)[check].tolist())
+            agg_obj = self.take(indices, axis=agg_axis)
+
+        if thresh is not lib.no_default:
+            count = agg_obj.count(axis=agg_axis)
+            mask = count >= thresh
+        elif how == "any":
+            # faster equivalent to 'agg_obj.count(agg_axis) == self.shape[agg_axis]'
+            mask = notna(agg_obj).all(axis=agg_axis, bool_only=False)
+        elif how == "all":
+            # faster equivalent to 'agg_obj.count(agg_axis) > 0'
+            mask = notna(agg_obj).any(axis=agg_axis, bool_only=False)
+        else:
+            raise ValueError(f"invalid how option: {how}")
+
+        if np.all(mask):
+            result = self.copy(deep=False)
+        else:
+            result = self.loc(axis=axis)[mask]
+
+        if ignore_index:
+            result.index = default_index(len(result))
+
+        if not inplace:
+            return result
+        self._update_inplace(result)
+        return None
+
+    @overload
+    def drop_duplicates(
+        self,
+        subset: Hashable | Iterable[Hashable] | None = ...,
+        *,
+        keep: DropKeep = ...,
+        inplace: Literal[True],
+        ignore_index: bool = ...,
+    ) -> None: ...
+
+    @overload
+    def drop_duplicates(
+        self,
+        subset: Hashable | Iterable[Hashable] | None = ...,
+        *,
+        keep: DropKeep = ...,
+        inplace: Literal[False] = ...,
+        ignore_index: bool = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def drop_duplicates(
+        self,
+        subset: Hashable | Iterable[Hashable] | None = ...,
+        *,
+        keep: DropKeep = ...,
+        inplace: bool = ...,
+        ignore_index: bool = ...,
+    ) -> DataFrame | None: ...
+
+    def drop_duplicates(
+        self,
+        subset: Hashable | Iterable[Hashable] | None = None,
+        *,
+        keep: DropKeep = "first",
+        inplace: bool = False,
+        ignore_index: bool = False,
+    ) -> DataFrame | None:
+        """
+        Return DataFrame with duplicate rows removed.
+
+        Considering certain columns is optional. Indexes, including time indexes
+        are ignored.
+
+        Parameters
+        ----------
+        subset : column label or iterable of labels, optional
+            Only consider certain columns for identifying duplicates, by
+            default use all of the columns.
+        keep : {'first', 'last', ``False``}, default 'first'
+            Determines which duplicates (if any) to keep.
+
+            - 'first' : Drop duplicates except for the first occurrence.
+            - 'last' : Drop duplicates except for the last occurrence.
+            - ``False`` : Drop all duplicates.
+
+        inplace : bool, default ``False``
+            Whether to modify the DataFrame rather than creating a new one.
+        ignore_index : bool, default ``False``
+            If ``True``, the resulting axis will be labeled 0, 1, …, n - 1.
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with duplicates removed or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.value_counts: Count unique combinations of columns.
+
+        Notes
+        -----
+        This method requires columns specified by ``subset`` to be of hashable type.
+        Passing unhashable columns will raise a ``TypeError``.
+
+        Examples
+        --------
+        Consider dataset containing ramen rating.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "brand": ["Yum Yum", "Yum Yum", "Indomie", "Indomie", "Indomie"],
+        ...         "style": ["cup", "cup", "cup", "pack", "pack"],
+        ...         "rating": [4, 4, 3.5, 15, 5],
+        ...     }
+        ... )
+        >>> df
+            brand style  rating
+        0  Yum Yum   cup     4.0
+        1  Yum Yum   cup     4.0
+        2  Indomie   cup     3.5
+        3  Indomie  pack    15.0
+        4  Indomie  pack     5.0
+
+        By default, it removes duplicate rows based on all columns.
+
+        >>> df.drop_duplicates()
+            brand style  rating
+        0  Yum Yum   cup     4.0
+        2  Indomie   cup     3.5
+        3  Indomie  pack    15.0
+        4  Indomie  pack     5.0
+
+        To remove duplicates on specific column(s), use ``subset``.
+
+        >>> df.drop_duplicates(subset=["brand"])
+            brand style  rating
+        0  Yum Yum   cup     4.0
+        2  Indomie   cup     3.5
+
+        To remove duplicates and keep last occurrences, use ``keep``.
+
+        >>> df.drop_duplicates(subset=["brand", "style"], keep="last")
+            brand style  rating
+        1  Yum Yum   cup     4.0
+        2  Indomie   cup     3.5
+        4  Indomie  pack     5.0
+        """
+        if self.empty:
+            return self.copy(deep=False)
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        ignore_index = validate_bool_kwarg(ignore_index, "ignore_index")
+
+        result = self[-self.duplicated(subset, keep=keep)]
+        if ignore_index:
+            result.index = default_index(len(result))
+
+        if inplace:
+            self._update_inplace(result)
+            return None
+        else:
+            return result
+
+    def duplicated(
+        self,
+        subset: Hashable | Iterable[Hashable] | None = None,
+        keep: DropKeep = "first",
+    ) -> Series:
+        """
+        Return boolean Series denoting duplicate rows.
+
+        Considering certain columns is optional.
+
+        Parameters
+        ----------
+        subset : column label or iterable of labels, optional
+            Only consider certain columns for identifying duplicates, by
+            default use all of the columns.
+        keep : {'first', 'last', False}, default 'first'
+            Determines which duplicates (if any) to mark.
+
+            - ``first`` : Mark duplicates as ``True`` except for the first occurrence.
+            - ``last`` : Mark duplicates as ``True`` except for the last occurrence.
+            - False : Mark all duplicates as ``True``.
+
+        Returns
+        -------
+        Series
+            Boolean series for each duplicated rows.
+
+        See Also
+        --------
+        Index.duplicated : Equivalent method on index.
+        Series.duplicated : Equivalent method on Series.
+        Series.drop_duplicates : Remove duplicate values from Series.
+        DataFrame.drop_duplicates : Remove duplicate values from DataFrame.
+
+        Examples
+        --------
+        Consider dataset containing ramen rating.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "brand": ["Yum Yum", "Yum Yum", "Indomie", "Indomie", "Indomie"],
+        ...         "style": ["cup", "cup", "cup", "pack", "pack"],
+        ...         "rating": [4, 4, 3.5, 15, 5],
+        ...     }
+        ... )
+        >>> df
+            brand style  rating
+        0  Yum Yum   cup     4.0
+        1  Yum Yum   cup     4.0
+        2  Indomie   cup     3.5
+        3  Indomie  pack    15.0
+        4  Indomie  pack     5.0
+
+        By default, for each set of duplicated values, the first occurrence
+        is set on False and all others on True.
+
+        >>> df.duplicated()
+        0    False
+        1     True
+        2    False
+        3    False
+        4    False
+        dtype: bool
+
+        By using 'last', the last occurrence of each set of duplicated values
+        is set on False and all others on True.
+
+        >>> df.duplicated(keep="last")
+        0     True
+        1    False
+        2    False
+        3    False
+        4    False
+        dtype: bool
+
+        By setting ``keep`` on False, all duplicates are True.
+
+        >>> df.duplicated(keep=False)
+        0     True
+        1     True
+        2    False
+        3    False
+        4    False
+        dtype: bool
+
+        To find duplicates on specific column(s), use ``subset``.
+
+        >>> df.duplicated(subset=["brand"])
+        0    False
+        1     True
+        2    False
+        3     True
+        4     True
+        dtype: bool
+        """
+
+        if self.empty:
+            return self._constructor_sliced(dtype=bool)
+
+        def f(vals) -> tuple[np.ndarray, int]:
+            labels, shape = algorithms.factorize(vals, size_hint=len(self))
+            return labels.astype("i8"), len(shape)
+
+        if subset is None:
+            subset = self.columns
+        elif (
+            not np.iterable(subset)
+            or isinstance(subset, str)
+            or (isinstance(subset, tuple) and subset in self.columns)
+        ):
+            subset = (subset,)
+
+        #  needed for mypy since can't narrow types using np.iterable
+        subset = cast(Sequence, subset)
+
+        # Verify all columns in subset exist in the queried dataframe
+        # Otherwise, raise a KeyError, same as if you try to __getitem__ with a
+        # key that doesn't exist.
+        diff = set(subset) - set(self.columns)
+        if diff:
+            raise KeyError(Index(diff))
+
+        if len(subset) == 1 and self.columns.is_unique:
+            # GH#45236 This is faster than get_group_index below
+            result = self[next(iter(subset))].duplicated(keep)
+            result.name = None
+        else:
+            vals = (col.values for name, col in self.items() if name in subset)
+            labels, shape = map(list, zip(*map(f, vals)))
+
+            ids = get_group_index(labels, tuple(shape), sort=False, xnull=False)
+            result = self._constructor_sliced(duplicated(ids, keep), index=self.index)
+        return result.__finalize__(self, method="duplicated")
+
+    # ----------------------------------------------------------------------
+    # Sorting
+    # error: Signature of "sort_values" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def sort_values(
+        self,
+        by: IndexLabel,
+        *,
+        axis: Axis = ...,
+        ascending=...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def sort_values(
+        self,
+        by: IndexLabel,
+        *,
+        axis: Axis = ...,
+        ascending=...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: str = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> None: ...
+
+    def sort_values(
+        self,
+        by: IndexLabel,
+        *,
+        axis: Axis = 0,
+        ascending: bool | list[bool] | tuple[bool, ...] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: str = "last",
+        ignore_index: bool = False,
+        key: ValueKeyFunc | None = None,
+    ) -> DataFrame | None:
+        """
+        Sort by the values along either axis.
+
+        Parameters
+        ----------
+        by : str or list of str
+            Name or list of names to sort by.
+
+            - if `axis` is 0 or `'index'` then `by` may contain index
+              levels and/or column labels.
+            - if `axis` is 1 or `'columns'` then `by` may contain column
+              levels and/or index labels.
+        axis : "{0 or 'index', 1 or 'columns'}", default 0
+             Axis to be sorted.
+        ascending : bool or list of bool, default True
+             Sort ascending vs. descending. Specify list for multiple sort
+             orders.  If this is a list of bools, must match the length of
+             the by.
+        inplace : bool, default False
+             If True, perform operation in-place.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+             Choice of sorting algorithm. See also :func:`numpy.sort` for more
+             information. `mergesort` and `stable` are the only stable algorithms. For
+             DataFrames, this option is only applied when sorting on a single
+             column or label.
+        na_position : {'first', 'last'}, default 'last'
+             Puts NaNs at the beginning if `first`; `last` puts NaNs at the
+             end.
+        ignore_index : bool, default False
+             If True, the resulting axis will be labeled 0, 1, …, n - 1.
+        key : callable, optional
+            Apply the key function to the values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect a
+            ``Series`` and return a Series with the same shape as the input.
+            It will be applied to each column in `by` independently. The values in the
+            returned Series will be used as the keys for sorting.
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with sorted values or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.sort_index : Sort a DataFrame by the index.
+        Series.sort_values : Similar method for a Series.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "col1": ["A", "A", "B", np.nan, "D", "C"],
+        ...         "col2": [2, 1, 9, 8, 7, 4],
+        ...         "col3": [0, 1, 9, 4, 2, 3],
+        ...         "col4": ["a", "B", "c", "D", "e", "F"],
+        ...     }
+        ... )
+        >>> df
+          col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+
+        **Sort by a single column**
+
+        In this case, we are sorting the rows according to values in ``col1``:
+
+        >>> df.sort_values(by=["col1"])
+          col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        5    C     4     3    F
+        4    D     7     2    e
+        3  NaN     8     4    D
+
+        **Sort by multiple columns**
+
+        You can also provide multiple columns to ``by`` argument, as shown below.
+        In this example, the rows are first sorted according to ``col1``, and then
+        the rows that have an identical value in ``col1`` are sorted according
+        to ``col2``.
+
+        >>> df.sort_values(by=["col1", "col2"])
+          col1  col2  col3 col4
+        1    A     1     1    B
+        0    A     2     0    a
+        2    B     9     9    c
+        5    C     4     3    F
+        4    D     7     2    e
+        3  NaN     8     4    D
+
+        **Sort in a descending order**
+
+        The sort order can be reversed using ``ascending`` argument, as shown below:
+
+        >>> df.sort_values(by="col1", ascending=False)
+          col1  col2  col3 col4
+        4    D     7     2    e
+        5    C     4     3    F
+        2    B     9     9    c
+        0    A     2     0    a
+        1    A     1     1    B
+        3  NaN     8     4    D
+
+        **Placing any** ``NA`` **first**
+
+        Note that in the above example, the rows that contain an ``NA`` value in their
+        ``col1`` are placed at the end of the dataframe. This behavior can be modified
+        via ``na_position`` argument, as shown below:
+
+        >>> df.sort_values(by="col1", ascending=False, na_position="first")
+          col1  col2  col3 col4
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+        2    B     9     9    c
+        0    A     2     0    a
+        1    A     1     1    B
+
+        **Customized sort order**
+
+        The ``key`` argument allows for a further customization of sorting behaviour.
+        For example, you may want
+        to ignore the `letter's case <https://en.wikipedia.org/wiki/Letter_case>`__
+        when sorting strings:
+
+        >>> df.sort_values(by="col4", key=lambda col: col.str.lower())
+           col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+
+        Another typical example is
+        `natural sorting <https://en.wikipedia.org/wiki/Natural_sort_order>`__.
+        This can be done using
+        ``natsort`` `package <https://github.com/SethMMorton/natsort>`__,
+        which provides sorted indices according
+        to their natural order, as shown below:
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "time": ["0hr", "128hr", "72hr", "48hr", "96hr"],
+        ...         "value": [10, 20, 30, 40, 50],
+        ...     }
+        ... )
+        >>> df
+            time  value
+        0    0hr     10
+        1  128hr     20
+        2   72hr     30
+        3   48hr     40
+        4   96hr     50
+        >>> from natsort import index_natsorted
+        >>> index_natsorted(df["time"])
+        [0, 3, 2, 4, 1]
+        >>> df.sort_values(
+        ...     by="time",
+        ...     key=lambda x: np.argsort(index_natsorted(x)),
+        ... )
+            time  value
+        0    0hr     10
+        3   48hr     40
+        2   72hr     30
+        4   96hr     50
+        1  128hr     20
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        axis = self._get_axis_number(axis)
+        ascending = validate_ascending(ascending)
+        if not isinstance(by, list):
+            by = [by]
+        # error: Argument 1 to "len" has incompatible type "Union[bool, List[bool]]";
+        # expected "Sized"
+        if is_sequence(ascending) and (
+            len(by) != len(ascending)  # type: ignore[arg-type]
+        ):
+            # error: Argument 1 to "len" has incompatible type "Union[bool,
+            # List[bool]]"; expected "Sized"
+            raise ValueError(
+                f"Length of ascending ({len(ascending)})"  # type: ignore[arg-type]
+                f" != length of by ({len(by)})"
+            )
+        if len(by) > 1:
+            keys = (self._get_label_or_level_values(x, axis=axis) for x in by)
+
+            # need to rewrap columns in Series to apply key function
+            if key is not None:
+                keys_data = [Series(k, name=name) for (k, name) in zip(keys, by)]
+            else:
+                # error: Argument 1 to "list" has incompatible type
+                # "Generator[ExtensionArray | ndarray[Any, Any], None, None]";
+                # expected "Iterable[Series]"
+                keys_data = list(keys)  # type: ignore[arg-type]
+
+            indexer = lexsort_indexer(
+                keys_data, orders=ascending, na_position=na_position, key=key
+            )
+        elif len(by):
+            # len(by) == 1
+
+            k = self._get_label_or_level_values(by[0], axis=axis)
+
+            # need to rewrap column in Series to apply key function
+            if key is not None:
+                # error: Incompatible types in assignment (expression has type
+                # "Series", variable has type "ndarray")
+                k = Series(k, name=by[0])  # type: ignore[assignment]
+
+            if isinstance(ascending, (tuple, list)):
+                ascending = ascending[0]
+
+            indexer = nargsort(
+                k, kind=kind, ascending=ascending, na_position=na_position, key=key
+            )
+        else:
+            if inplace:
+                return self._update_inplace(self)
+            else:
+                return self.copy(deep=False)
+
+        if is_range_indexer(indexer, len(indexer)):
+            result = self.copy(deep=False)
+            if ignore_index:
+                result.index = default_index(len(result))
+
+            if inplace:
+                return self._update_inplace(result)
+            else:
+                return result
+
+        new_data = self._mgr.take(
+            indexer, axis=self._get_block_manager_axis(axis), verify=False
+        )
+
+        if ignore_index:
+            new_data.set_axis(
+                self._get_block_manager_axis(axis), default_index(len(indexer))
+            )
+
+        result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="sort_values")
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> None: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: bool = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> DataFrame | None: ...
+
+    def sort_index(
+        self,
+        *,
+        axis: Axis = 0,
+        level: IndexLabel | None = None,
+        ascending: bool | Sequence[bool] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: NaPosition = "last",
+        sort_remaining: bool = True,
+        ignore_index: bool = False,
+        key: IndexKeyFunc | None = None,
+    ) -> DataFrame | None:
+        """
+        Sort object by labels (along an axis).
+
+        Returns a new DataFrame sorted by label if `inplace` argument is
+        ``False``, otherwise updates the original DataFrame and returns None.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis along which to sort.  The value 0 identifies the rows,
+            and 1 identifies the columns.
+        level : int or level name or list of ints or list of level names
+            If not None, sort on values in specified index level(s).
+        ascending : bool or list-like of bools, default True
+            Sort ascending vs. descending. When the index is a MultiIndex the
+            sort direction can be controlled for each level individually.
+        inplace : bool, default False
+            Whether to modify the DataFrame rather than creating a new one.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+            Choice of sorting algorithm. See also :func:`numpy.sort` for more
+            information. `mergesort` and `stable` are the only stable algorithms. For
+            DataFrames, this option is only applied when sorting on a single
+            column or label.
+        na_position : {'first', 'last'}, default 'last'
+            Puts NaNs at the beginning if `first`; `last` puts NaNs at the end.
+            Not implemented for MultiIndex.
+        sort_remaining : bool, default True
+            If True and sorting by level and index is multilevel, sort by other
+            levels too (in order) after sorting by specified level.
+        ignore_index : bool, default False
+            If True, the resulting axis will be labeled 0, 1, …, n - 1.
+        key : callable, optional
+            If not None, apply the key function to the index values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect an
+            ``Index`` and return an ``Index`` of the same shape. For MultiIndex
+            inputs, the key is applied *per level*.
+
+        Returns
+        -------
+        DataFrame or None
+            The original DataFrame sorted by the labels or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.sort_index : Sort Series by the index.
+        DataFrame.sort_values : Sort DataFrame by the value.
+        Series.sort_values : Sort Series by the value.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [1, 2, 3, 4, 5], index=[100, 29, 234, 1, 150], columns=["A"]
+        ... )
+        >>> df.sort_index()
+             A
+        1    4
+        29   2
+        100  1
+        150  5
+        234  3
+
+        By default, it sorts in ascending order, to sort in descending order,
+        use ``ascending=False``
+
+        >>> df.sort_index(ascending=False)
+             A
+        234  3
+        150  5
+        100  1
+        29   2
+        1    4
+
+        A key function can be specified which is applied to the index before
+        sorting. For a ``MultiIndex`` this is applied to each level separately.
+
+        >>> df = pd.DataFrame({"a": [1, 2, 3, 4]}, index=["A", "b", "C", "d"])
+        >>> df.sort_index(key=lambda x: x.str.lower())
+           a
+        A  1
+        b  2
+        C  3
+        d  4
+        """
+        return super().sort_index(
+            axis=axis,
+            level=level,
+            ascending=ascending,
+            inplace=inplace,
+            kind=kind,
+            na_position=na_position,
+            sort_remaining=sort_remaining,
+            ignore_index=ignore_index,
+            key=key,
+        )
+
+    def value_counts(
+        self,
+        subset: IndexLabel | None = None,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        dropna: bool = True,
+    ) -> Series:
+        """
+        Return a Series containing the frequency of each distinct row in the DataFrame.
+
+        Parameters
+        ----------
+        subset : label or list of labels, optional
+            Columns to use when counting unique combinations.
+        normalize : bool, default False
+            Return proportions rather than frequencies.
+        sort : bool, default True
+            Sort by frequencies when True. Preserve the order of the data when False.
+
+            .. versionchanged:: 3.0.0
+
+                Prior to 3.0.0, ``sort=False`` would sort by the columns values.
+        ascending : bool, default False
+            Sort in ascending order.
+        dropna : bool, default True
+            Do not include counts of rows that contain NA values.
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Series
+            Series containing the frequency of each distinct row in the DataFrame.
+
+        See Also
+        --------
+        Series.value_counts: Equivalent method on Series.
+
+        Notes
+        -----
+        The returned Series will have a MultiIndex with one level per input
+        column but an Index (non-multi) for a single label. By default, rows
+        that contain any NA values are omitted from the result. By default,
+        the resulting Series will be sorted by frequencies in descending order so that
+        the first element is the most frequently-occurring row.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        ...     index=["falcon", "dog", "cat", "ant"],
+        ... )
+        >>> df
+                num_legs  num_wings
+        falcon         2          2
+        dog            4          0
+        cat            4          0
+        ant            6          0
+
+        >>> df.value_counts()
+        num_legs  num_wings
+        4         0            2
+        2         2            1
+        6         0            1
+        Name: count, dtype: int64
+
+        >>> df.value_counts(sort=False)
+        num_legs  num_wings
+        2         2            1
+        4         0            2
+        6         0            1
+        Name: count, dtype: int64
+
+        >>> df.value_counts(ascending=True)
+        num_legs  num_wings
+        2         2            1
+        6         0            1
+        4         0            2
+        Name: count, dtype: int64
+
+        >>> df.value_counts(normalize=True)
+        num_legs  num_wings
+        4         0            0.50
+        2         2            0.25
+        6         0            0.25
+        Name: proportion, dtype: float64
+
+        With `dropna` set to `False` we can also count rows with NA values.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "first_name": ["John", "Anne", "John", "Beth"],
+        ...         "middle_name": ["Smith", pd.NA, pd.NA, "Louise"],
+        ...     }
+        ... )
+        >>> df
+          first_name middle_name
+        0       John       Smith
+        1       Anne        <NA>
+        2       John        <NA>
+        3       Beth      Louise
+
+        >>> df.value_counts()
+        first_name  middle_name
+        Beth        Louise         1
+        John        Smith          1
+        Name: count, dtype: int64
+
+        >>> df.value_counts(dropna=False)
+        first_name  middle_name
+        Anne        NaN            1
+        Beth        Louise         1
+        John        Smith          1
+                    NaN            1
+        Name: count, dtype: int64
+
+        >>> df.value_counts("first_name")
+        first_name
+        John    2
+        Anne    1
+        Beth    1
+        Name: count, dtype: int64
+        """
+        if subset is None:
+            subset = self.columns.tolist()
+
+        name = "proportion" if normalize else "count"
+        counts = self.groupby(
+            subset, sort=False, dropna=dropna, observed=False
+        )._grouper.size()
+        counts.name = name
+
+        if sort:
+            counts = counts.sort_values(ascending=ascending)
+        if normalize:
+            counts /= counts.sum()
+
+        # Force MultiIndex for a list_like subset with a single column
+        if is_list_like(subset) and len(subset) == 1:  # type: ignore[arg-type]
+            counts.index = MultiIndex.from_arrays(
+                [counts.index], names=[counts.index.name]
+            )
+
+        return counts
+
+    def nlargest(
+        self, n: int, columns: IndexLabel, keep: NsmallestNlargestKeep = "first"
+    ) -> DataFrame:
+        """
+        Return the first `n` rows ordered by `columns` in descending order.
+
+        Return the first `n` rows with the largest values in `columns`, in
+        descending order. The columns that are not specified are returned as
+        well, but not used for ordering.
+
+        This method is equivalent to
+        ``df.sort_values(columns, ascending=False).head(n)``, but more
+        performant.
+
+        Parameters
+        ----------
+        n : int
+            Number of rows to return.
+        columns : label or list of labels
+            Column label(s) to order by.
+        keep : {'first', 'last', 'all'}, default 'first'
+            Where there are duplicate values:
+
+            - ``first`` : prioritize the first occurrence(s)
+            - ``last`` : prioritize the last occurrence(s)
+            - ``all`` : keep all the ties of the smallest item even if it means
+              selecting more than ``n`` items.
+
+        Returns
+        -------
+        DataFrame
+            The first `n` rows ordered by the given columns in descending
+            order.
+
+        See Also
+        --------
+        DataFrame.nsmallest : Return the first `n` rows ordered by `columns` in
+            ascending order.
+        DataFrame.sort_values : Sort DataFrame by the values.
+        DataFrame.head : Return the first `n` rows without re-ordering.
+
+        Notes
+        -----
+        This function cannot be used with all column types. For example, when
+        specifying columns with `object` or `category` dtypes, ``TypeError`` is
+        raised.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "population": [
+        ...             59000000,
+        ...             65000000,
+        ...             434000,
+        ...             434000,
+        ...             434000,
+        ...             337000,
+        ...             11300,
+        ...             11300,
+        ...             11300,
+        ...         ],
+        ...         "GDP": [1937894, 2583560, 12011, 4520, 12128, 17036, 182, 38, 311],
+        ...         "alpha-2": ["IT", "FR", "MT", "MV", "BN", "IS", "NR", "TV", "AI"],
+        ...     },
+        ...     index=[
+        ...         "Italy",
+        ...         "France",
+        ...         "Malta",
+        ...         "Maldives",
+        ...         "Brunei",
+        ...         "Iceland",
+        ...         "Nauru",
+        ...         "Tuvalu",
+        ...         "Anguilla",
+        ...     ],
+        ... )
+        >>> df
+                  population      GDP alpha-2
+        Italy       59000000  1937894      IT
+        France      65000000  2583560      FR
+        Malta         434000    12011      MT
+        Maldives      434000     4520      MV
+        Brunei        434000    12128      BN
+        Iceland       337000    17036      IS
+        Nauru          11300      182      NR
+        Tuvalu         11300       38      TV
+        Anguilla       11300      311      AI
+
+        In the following example, we will use ``nlargest`` to select the three
+        rows having the largest values in column "population".
+
+        >>> df.nlargest(3, "population")
+                population      GDP alpha-2
+        France    65000000  2583560      FR
+        Italy     59000000  1937894      IT
+        Malta       434000    12011      MT
+
+        When using ``keep='last'``, ties are resolved in reverse order:
+
+        >>> df.nlargest(3, "population", keep="last")
+                population      GDP alpha-2
+        France    65000000  2583560      FR
+        Italy     59000000  1937894      IT
+        Brunei      434000    12128      BN
+
+        When using ``keep='all'``, the number of element kept can go beyond ``n``
+        if there are duplicate values for the smallest element, all the
+        ties are kept:
+
+        >>> df.nlargest(3, "population", keep="all")
+                  population      GDP alpha-2
+        France      65000000  2583560      FR
+        Italy       59000000  1937894      IT
+        Malta         434000    12011      MT
+        Maldives      434000     4520      MV
+        Brunei        434000    12128      BN
+
+        However, ``nlargest`` does not keep ``n`` distinct largest elements:
+
+        >>> df.nlargest(5, "population", keep="all")
+                  population      GDP alpha-2
+        France      65000000  2583560      FR
+        Italy       59000000  1937894      IT
+        Malta         434000    12011      MT
+        Maldives      434000     4520      MV
+        Brunei        434000    12128      BN
+
+        To order by the largest values in column "population" and then "GDP",
+        we can specify multiple columns like in the next example.
+
+        >>> df.nlargest(3, ["population", "GDP"])
+                population      GDP alpha-2
+        France    65000000  2583560      FR
+        Italy     59000000  1937894      IT
+        Brunei      434000    12128      BN
+        """
+        return selectn.SelectNFrame(self, n=n, keep=keep, columns=columns).nlargest()
+
+    def nsmallest(
+        self, n: int, columns: IndexLabel, keep: NsmallestNlargestKeep = "first"
+    ) -> DataFrame:
+        """
+        Return the first `n` rows ordered by `columns` in ascending order.
+
+        Return the first `n` rows with the smallest values in `columns`, in
+        ascending order. The columns that are not specified are returned as
+        well, but not used for ordering.
+
+        This method is equivalent to
+        ``df.sort_values(columns, ascending=True).head(n)``, but more
+        performant.
+
+        Parameters
+        ----------
+        n : int
+            Number of items to retrieve.
+        columns : list or str
+            Column name or names to order by.
+        keep : {'first', 'last', 'all'}, default 'first'
+            Where there are duplicate values:
+
+            - ``first`` : take the first occurrence.
+            - ``last`` : take the last occurrence.
+            - ``all`` : keep all the ties of the largest item even if it means
+              selecting more than ``n`` items.
+
+        Returns
+        -------
+        DataFrame
+            DataFrame with the first `n` rows ordered by `columns` in ascending order.
+
+        See Also
+        --------
+        DataFrame.nlargest : Return the first `n` rows ordered by `columns` in
+            descending order.
+        DataFrame.sort_values : Sort DataFrame by the values.
+        DataFrame.head : Return the first `n` rows without re-ordering.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "population": [
+        ...             59000000,
+        ...             65000000,
+        ...             434000,
+        ...             434000,
+        ...             434000,
+        ...             337000,
+        ...             337000,
+        ...             11300,
+        ...             11300,
+        ...         ],
+        ...         "GDP": [1937894, 2583560, 12011, 4520, 12128, 17036, 182, 38, 311],
+        ...         "alpha-2": ["IT", "FR", "MT", "MV", "BN", "IS", "NR", "TV", "AI"],
+        ...     },
+        ...     index=[
+        ...         "Italy",
+        ...         "France",
+        ...         "Malta",
+        ...         "Maldives",
+        ...         "Brunei",
+        ...         "Iceland",
+        ...         "Nauru",
+        ...         "Tuvalu",
+        ...         "Anguilla",
+        ...     ],
+        ... )
+        >>> df
+                  population      GDP alpha-2
+        Italy       59000000  1937894      IT
+        France      65000000  2583560      FR
+        Malta         434000    12011      MT
+        Maldives      434000     4520      MV
+        Brunei        434000    12128      BN
+        Iceland       337000    17036      IS
+        Nauru         337000      182      NR
+        Tuvalu         11300       38      TV
+        Anguilla       11300      311      AI
+
+        In the following example, we will use ``nsmallest`` to select the
+        three rows having the smallest values in column "population".
+
+        >>> df.nsmallest(3, "population")
+                  population    GDP alpha-2
+        Tuvalu         11300     38      TV
+        Anguilla       11300    311      AI
+        Iceland       337000  17036      IS
+
+        When using ``keep='last'``, ties are resolved in reverse order:
+
+        >>> df.nsmallest(3, "population", keep="last")
+                  population  GDP alpha-2
+        Anguilla       11300  311      AI
+        Tuvalu         11300   38      TV
+        Nauru         337000  182      NR
+
+        When using ``keep='all'``, the number of element kept can go beyond ``n``
+        if there are duplicate values for the largest element, all the
+        ties are kept.
+
+        >>> df.nsmallest(3, "population", keep="all")
+                  population    GDP alpha-2
+        Tuvalu         11300     38      TV
+        Anguilla       11300    311      AI
+        Iceland       337000  17036      IS
+        Nauru         337000    182      NR
+
+        However, ``nsmallest`` does not keep ``n`` distinct
+        smallest elements:
+
+        >>> df.nsmallest(4, "population", keep="all")
+                  population    GDP alpha-2
+        Tuvalu         11300     38      TV
+        Anguilla       11300    311      AI
+        Iceland       337000  17036      IS
+        Nauru         337000    182      NR
+
+        To order by the smallest values in column "population" and then "GDP", we can
+        specify multiple columns like in the next example.
+
+        >>> df.nsmallest(3, ["population", "GDP"])
+                  population  GDP alpha-2
+        Tuvalu         11300   38      TV
+        Anguilla       11300  311      AI
+        Nauru         337000  182      NR
+        """
+        return selectn.SelectNFrame(self, n=n, keep=keep, columns=columns).nsmallest()
+
+    def swaplevel(self, i: Axis = -2, j: Axis = -1, axis: Axis = 0) -> DataFrame:
+        """
+        Swap levels i and j in a :class:`MultiIndex`.
+
+        Default is to swap the two innermost levels of the index.
+
+        Parameters
+        ----------
+        i, j : int or str
+            Levels of the indices to be swapped. Can pass level name as string.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+                    The axis to swap levels on. 0 or 'index' for row-wise, 1 or
+                    'columns' for column-wise.
+
+        Returns
+        -------
+        DataFrame
+            DataFrame with levels swapped in MultiIndex.
+
+        See Also
+        --------
+        DataFrame.reorder_levels: Reorder levels of MultiIndex.
+        DataFrame.sort_index: Sort MultiIndex.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"Grade": ["A", "B", "A", "C"]},
+        ...     index=[
+        ...         ["Final exam", "Final exam", "Coursework", "Coursework"],
+        ...         ["History", "Geography", "History", "Geography"],
+        ...         ["January", "February", "March", "April"],
+        ...     ],
+        ... )
+        >>> df
+                                            Grade
+        Final exam  History     January      A
+                    Geography   February     B
+        Coursework  History     March        A
+                    Geography   April        C
+
+        In the following example, we will swap the levels of the indices.
+        Here, we will swap the levels column-wise, but levels can be swapped row-wise
+        in a similar manner. Note that column-wise is the default behaviour.
+        By not supplying any arguments for i and j, we swap the last and second to
+        last indices.
+
+        >>> df.swaplevel()
+                                            Grade
+        Final exam  January     History         A
+                    February    Geography       B
+        Coursework  March       History         A
+                    April       Geography       C
+
+        By supplying one argument, we can choose which index to swap the last
+        index with. We can for example swap the first index with the last one as
+        follows.
+
+        >>> df.swaplevel(0)
+                                            Grade
+        January     History     Final exam      A
+        February    Geography   Final exam      B
+        March       History     Coursework      A
+        April       Geography   Coursework      C
+
+        We can also define explicitly which indices we want to swap by supplying values
+        for both i and j. Here, we for example swap the first and second indices.
+
+        >>> df.swaplevel(0, 1)
+                                            Grade
+        History     Final exam  January         A
+        Geography   Final exam  February        B
+        History     Coursework  March           A
+        Geography   Coursework  April           C
+        """
+        result = self.copy(deep=False)
+
+        axis = self._get_axis_number(axis)
+
+        if not isinstance(result._get_axis(axis), MultiIndex):  # pragma: no cover
+            raise TypeError("Can only swap levels on a hierarchical axis.")
+
+        if axis == 0:
+            assert isinstance(result.index, MultiIndex)
+            result.index = result.index.swaplevel(i, j)
+        else:
+            assert isinstance(result.columns, MultiIndex)
+            result.columns = result.columns.swaplevel(i, j)
+        return result
+
+    def reorder_levels(self, order: Sequence[int | str], axis: Axis = 0) -> DataFrame:
+        """
+        Rearrange index or column levels using input ``order``.
+
+        May not drop or duplicate levels.
+
+        Parameters
+        ----------
+        order : list of int or list of str
+            List representing new level order. Reference level by number
+            (position) or by key (label).
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Where to reorder levels.
+
+        Returns
+        -------
+        DataFrame
+            DataFrame with indices or columns with reordered levels.
+
+        See Also
+        --------
+            DataFrame.swaplevel : Swap levels i and j in a MultiIndex.
+
+        Examples
+        --------
+        >>> data = {
+        ...     "class": ["Mammals", "Mammals", "Reptiles"],
+        ...     "diet": ["Omnivore", "Carnivore", "Carnivore"],
+        ...     "species": ["Humans", "Dogs", "Snakes"],
+        ... }
+        >>> df = pd.DataFrame(data, columns=["class", "diet", "species"])
+        >>> df = df.set_index(["class", "diet"])
+        >>> df
+                                          species
+        class      diet
+        Mammals    Omnivore                Humans
+                   Carnivore                 Dogs
+        Reptiles   Carnivore               Snakes
+
+        Let's reorder the levels of the index:
+
+        >>> df.reorder_levels(["diet", "class"])
+                                          species
+        diet      class
+        Omnivore  Mammals                  Humans
+        Carnivore Mammals                    Dogs
+                  Reptiles                 Snakes
+        """
+        axis = self._get_axis_number(axis)
+        if not isinstance(self._get_axis(axis), MultiIndex):  # pragma: no cover
+            raise TypeError("Can only reorder levels on a hierarchical axis.")
+
+        result = self.copy(deep=False)
+
+        if axis == 0:
+            assert isinstance(result.index, MultiIndex)
+            result.index = result.index.reorder_levels(order)
+        else:
+            assert isinstance(result.columns, MultiIndex)
+            result.columns = result.columns.reorder_levels(order)
+        return result
+
+    # ----------------------------------------------------------------------
+    # Arithmetic Methods
+
+    def _cmp_method(self, other, op):
+        axis: Literal[1] = 1  # only relevant for Series other case
+
+        self, other = self._align_for_op(other, axis, flex=False, level=None)
+
+        # See GH#4537 for discussion of scalar op behavior
+        new_data = self._dispatch_frame_op(other, op, axis=axis)
+        return self._construct_result(new_data)
+
+    def _arith_method(self, other, op):
+        if self._should_reindex_frame_op(other, op, 1, None, None):
+            return self._arith_method_with_reindex(other, op)
+
+        axis: Literal[1] = 1  # only relevant for Series other case
+        other = ops.maybe_prepare_scalar_for_op(other, (self.shape[axis],))
+
+        self, other = self._align_for_op(other, axis, flex=True, level=None)
+
+        with np.errstate(all="ignore"):
+            new_data = self._dispatch_frame_op(other, op, axis=axis)
+        return self._construct_result(new_data)
+
+    _logical_method = _arith_method
+
+    def _dispatch_frame_op(
+        self, right, func: Callable, axis: AxisInt | None = None
+    ) -> DataFrame:
+        """
+        Evaluate the frame operation func(left, right) by evaluating
+        column-by-column, dispatching to the Series implementation.
+
+        Parameters
+        ----------
+        right : scalar, Series, or DataFrame
+        func : arithmetic or comparison operator
+        axis : {None, 0, 1}
+
+        Returns
+        -------
+        DataFrame
+
+        Notes
+        -----
+        Caller is responsible for setting np.errstate where relevant.
+        """
+        # Get the appropriate array-op to apply to each column/block's values.
+        array_op = ops.get_array_op(func)
+
+        right = lib.item_from_zerodim(right)
+        if not is_list_like(right):
+            # i.e. scalar, faster than checking np.ndim(right) == 0
+            bm = self._mgr.apply(array_op, right=right)
+            return self._constructor_from_mgr(bm, axes=bm.axes)
+
+        elif isinstance(right, DataFrame):
+            assert self.index.equals(right.index)
+            assert self.columns.equals(right.columns)
+            # TODO: The previous assertion `assert right._indexed_same(self)`
+            #  fails in cases with empty columns reached via
+            #  _frame_arith_method_with_reindex
+
+            # TODO operate_blockwise expects a manager of the same type
+            bm = self._mgr.operate_blockwise(
+                right._mgr,
+                array_op,
+            )
+            return self._constructor_from_mgr(bm, axes=bm.axes)
+
+        elif isinstance(right, Series) and axis == 1:
+            # axis=1 means we want to operate row-by-row
+            assert right.index.equals(self.columns)
+
+            right = right._values
+            # maybe_align_as_frame ensures we do not have an ndarray here
+            assert not isinstance(right, np.ndarray)
+
+            arrays = [
+                array_op(_left, _right)
+                for _left, _right in zip(self._iter_column_arrays(), right)
+            ]
+
+        elif isinstance(right, Series):
+            assert right.index.equals(self.index)
+            right = right._values
+
+            arrays = [array_op(left, right) for left in self._iter_column_arrays()]
+
+        else:
+            raise NotImplementedError(right)
+
+        return type(self)._from_arrays(
+            arrays, self.columns, self.index, verify_integrity=False
+        )
+
+    def _combine_frame(self, other: DataFrame, func, fill_value=None):
+        # at this point we have `self._indexed_same(other)`
+
+        if fill_value is None:
+            # since _arith_op may be called in a loop, avoid function call
+            #  overhead if possible by doing this check once
+            _arith_op = func
+
+        else:
+
+            def _arith_op(left, right):
+                # for the mixed_type case where we iterate over columns,
+                # _arith_op(left, right) is equivalent to
+                # left._binop(right, func, fill_value=fill_value)
+                left, right = ops.fill_binop(left, right, fill_value)
+                return func(left, right)
+
+        new_data = self._dispatch_frame_op(other, _arith_op)
+        return new_data
+
+    def _arith_method_with_reindex(self, right: DataFrame, op) -> DataFrame:
+        """
+        For DataFrame-with-DataFrame operations that require reindexing,
+        operate only on shared columns, then reindex.
+
+        Parameters
+        ----------
+        right : DataFrame
+        op : binary operator
+
+        Returns
+        -------
+        DataFrame
+        """
+        left = self
+
+        # GH#31623, only operate on shared columns
+        cols, lcol_indexer, rcol_indexer = left.columns.join(
+            right.columns, how="inner", return_indexers=True
+        )
+
+        new_left = left if lcol_indexer is None else left.iloc[:, lcol_indexer]
+        new_right = right if rcol_indexer is None else right.iloc[:, rcol_indexer]
+
+        # GH#60498 For MultiIndex column alignment
+        if isinstance(cols, MultiIndex):
+            # When overwriting column names, make a shallow copy so as to not modify
+            # the input DFs
+            new_left = new_left.copy(deep=False)
+            new_right = new_right.copy(deep=False)
+            new_left.columns = cols
+            new_right.columns = cols
+
+        result = op(new_left, new_right)
+
+        # Do the join on the columns instead of using left._align_for_op
+        #  to avoid constructing two potentially large/sparse DataFrames
+        join_columns = left.columns.join(right.columns, how="outer")
+
+        if result.columns.has_duplicates:
+            # Avoid reindexing with a duplicate axis.
+            # https://github.com/pandas-dev/pandas/issues/35194
+            indexer, _ = result.columns.get_indexer_non_unique(join_columns)
+            indexer = algorithms.unique1d(indexer)
+            result = result._reindex_with_indexers(
+                {1: [join_columns, indexer]}, allow_dups=True
+            )
+        else:
+            result = result.reindex(join_columns, axis=1)
+
+        return result
+
+    def _should_reindex_frame_op(self, right, op, axis: int, fill_value, level) -> bool:
+        """
+        Check if this is an operation between DataFrames that will need to reindex.
+        """
+        if op is operator.pow or op is roperator.rpow:
+            # GH#32685 pow has special semantics for operating with null values
+            return False
+
+        if not isinstance(right, DataFrame):
+            return False
+
+        if (
+            (
+                isinstance(self.columns, MultiIndex)
+                or isinstance(right.columns, MultiIndex)
+            )
+            and not self.columns.equals(right.columns)
+            and fill_value is None
+        ):
+            # GH#60498 Reindex if MultiIndexe columns are not matching
+            # GH#60903 Don't reindex if fill_value is provided
+            return True
+
+        if fill_value is None and level is None and axis == 1:
+            # TODO: any other cases we should handle here?
+
+            # Intersection is always unique so we have to check the unique columns
+            left_uniques = self.columns.unique()
+            right_uniques = right.columns.unique()
+            cols = left_uniques.intersection(right_uniques)
+            if len(cols) and not (
+                len(cols) == len(left_uniques) and len(cols) == len(right_uniques)
+            ):
+                # TODO: is there a shortcut available when len(cols) == 0?
+                return True
+
+        return False
+
+    def _align_for_op(
+        self,
+        other,
+        axis: AxisInt,
+        flex: bool | None = False,
+        level: Level | None = None,
+    ):
+        """
+        Convert rhs to meet lhs dims if input is list, tuple or np.ndarray.
+
+        Parameters
+        ----------
+        left : DataFrame
+        right : Any
+        axis : int
+        flex : bool or None, default False
+            Whether this is a flex op, in which case we reindex.
+            None indicates not to check for alignment.
+        level : int or level name, default None
+
+        Returns
+        -------
+        left : DataFrame
+        right : Any
+        """
+        left, right = self, other
+
+        def to_series(right):
+            msg = (
+                "Unable to coerce to Series, "
+                "length must be {req_len}: given {given_len}"
+            )
+
+            # pass dtype to avoid doing inference, which would break consistency
+            #  with Index/Series ops
+            dtype = None
+            if getattr(right, "dtype", None) == object:
+                # can't pass right.dtype unconditionally as that would break on e.g.
+                #  datetime64[h] ndarray
+                dtype = object
+
+            if axis == 0:
+                if len(left.index) != len(right):
+                    raise ValueError(
+                        msg.format(req_len=len(left.index), given_len=len(right))
+                    )
+                right = left._constructor_sliced(right, index=left.index, dtype=dtype)
+            else:
+                if len(left.columns) != len(right):
+                    raise ValueError(
+                        msg.format(req_len=len(left.columns), given_len=len(right))
+                    )
+                right = left._constructor_sliced(right, index=left.columns, dtype=dtype)
+            return right
+
+        if isinstance(right, np.ndarray):
+            if right.ndim == 1:
+                right = to_series(right)
+
+            elif right.ndim == 2:
+                # We need to pass dtype=right.dtype to retain object dtype
+                #  otherwise we lose consistency with Index and array ops
+                dtype = None
+                if right.dtype == object:
+                    # can't pass right.dtype unconditionally as that would break on e.g.
+                    #  datetime64[h] ndarray
+                    dtype = object
+
+                if right.shape == left.shape:
+                    right = left._constructor(
+                        right, index=left.index, columns=left.columns, dtype=dtype
+                    )
+
+                elif right.shape[0] == left.shape[0] and right.shape[1] == 1:
+                    # Broadcast across columns
+                    right = np.broadcast_to(right, left.shape)
+                    right = left._constructor(
+                        right, index=left.index, columns=left.columns, dtype=dtype
+                    )
+
+                elif right.shape[1] == left.shape[1] and right.shape[0] == 1:
+                    # Broadcast along rows
+                    right = to_series(right[0, :])
+
+                else:
+                    raise ValueError(
+                        "Unable to coerce to DataFrame, shape "
+                        f"must be {left.shape}: given {right.shape}"
+                    )
+
+            elif right.ndim > 2:
+                raise ValueError(
+                    "Unable to coerce to Series/DataFrame, "
+                    f"dimension must be <= 2: {right.shape}"
+                )
+
+        elif is_list_like(right) and not isinstance(right, (Series, DataFrame)):
+            # GH#36702. Raise when attempting arithmetic with list of array-like.
+            if any(is_array_like(el) for el in right):
+                raise ValueError(
+                    f"Unable to coerce list of {type(right[0])} to Series/DataFrame"
+                )
+            # GH#17901
+            right = to_series(right)
+
+        if flex is not None and isinstance(right, DataFrame):
+            if not left._indexed_same(right):
+                if flex:
+                    left, right = left.align(right, join="outer", level=level)
+                else:
+                    raise ValueError(
+                        "Can only compare identically-labeled (both index and columns) "
+                        "DataFrame objects"
+                    )
+        elif isinstance(right, Series):
+            # axis=1 is default for DataFrame-with-Series op
+            axis = axis if axis is not None else 1
+            if not flex:
+                if not left.axes[axis].equals(right.index):
+                    raise ValueError(
+                        "Operands are not aligned. Do "
+                        "`left, right = left.align(right, axis=1)` "
+                        "before operating."
+                    )
+
+            left, right = left.align(
+                right,
+                join="outer",
+                axis=axis,
+                level=level,
+            )
+            right = left._maybe_align_series_as_frame(right, axis)
+
+        return left, right
+
+    def _maybe_align_series_as_frame(self, series: Series, axis: AxisInt):
+        """
+        If the Series operand is not EA-dtype, we can broadcast to 2D and operate
+        blockwise.
+        """
+        rvalues = series._values
+        if not isinstance(rvalues, np.ndarray):
+            # TODO(EA2D): no need to special-case with 2D EAs
+            if rvalues.dtype in ("datetime64[ns]", "timedelta64[ns]"):
+                # We can losslessly+cheaply cast to ndarray
+                rvalues = np.asarray(rvalues)
+            else:
+                return series
+
+        if axis == 0:
+            rvalues = rvalues.reshape(-1, 1)
+        else:
+            rvalues = rvalues.reshape(1, -1)
+
+        rvalues = np.broadcast_to(rvalues, self.shape)
+        # pass dtype to avoid doing inference
+        return self._constructor(
+            rvalues,
+            index=self.index,
+            columns=self.columns,
+            dtype=rvalues.dtype,
+        )
+
+    def _flex_arith_method(
+        self, other, op, *, axis: Axis = "columns", level=None, fill_value=None
+    ):
+        axis = self._get_axis_number(axis) if axis is not None else 1
+
+        if self._should_reindex_frame_op(other, op, axis, fill_value, level):
+            return self._arith_method_with_reindex(other, op)
+
+        if isinstance(other, Series) and fill_value is not None:
+            # TODO: We could allow this in cases where we end up going
+            #  through the DataFrame path
+            raise NotImplementedError(f"fill_value {fill_value} not supported.")
+
+        other = ops.maybe_prepare_scalar_for_op(other, self.shape)
+        self, other = self._align_for_op(other, axis, flex=True, level=level)
+
+        with np.errstate(all="ignore"):
+            if isinstance(other, DataFrame):
+                # Another DataFrame
+                new_data = self._combine_frame(other, op, fill_value)
+
+            elif isinstance(other, Series):
+                new_data = self._dispatch_frame_op(other, op, axis=axis)
+            else:
+                # in this case we always have `np.ndim(other) == 0`
+                if fill_value is not None:
+                    self = self.fillna(fill_value)
+
+                new_data = self._dispatch_frame_op(other, op)
+
+        return self._construct_result(new_data)
+
+    def _construct_result(self, result) -> DataFrame:
+        """
+        Wrap the result of an arithmetic, comparison, or logical operation.
+
+        Parameters
+        ----------
+        result : DataFrame
+
+        Returns
+        -------
+        DataFrame
+        """
+        out = self._constructor(result, copy=False).__finalize__(self)
+        # Pin columns instead of passing to constructor for compat with
+        #  non-unique columns case
+        out.columns = self.columns
+        out.index = self.index
+        return out
+
+    def __divmod__(self, other) -> tuple[DataFrame, DataFrame]:
+        # Naive implementation, room for optimization
+        div = self // other
+        mod = self - div * other
+        return div, mod
+
+    def __rdivmod__(self, other) -> tuple[DataFrame, DataFrame]:
+        # Naive implementation, room for optimization
+        div = other // self
+        mod = other - div * self
+        return div, mod
+
+    def _flex_cmp_method(self, other, op, *, axis: Axis = "columns", level=None):
+        axis = self._get_axis_number(axis) if axis is not None else 1
+
+        self, other = self._align_for_op(other, axis, flex=True, level=level)
+
+        new_data = self._dispatch_frame_op(other, op, axis=axis)
+        return self._construct_result(new_data)
+
+    @Appender(ops.make_flex_doc("eq", "dataframe"))
+    def eq(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.eq, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("ne", "dataframe"))
+    def ne(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.ne, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("le", "dataframe"))
+    def le(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.le, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("lt", "dataframe"))
+    def lt(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.lt, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("ge", "dataframe"))
+    def ge(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.ge, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("gt", "dataframe"))
+    def gt(self, other, axis: Axis = "columns", level=None) -> DataFrame:
+        return self._flex_cmp_method(other, operator.gt, axis=axis, level=level)
+
+    @Appender(ops.make_flex_doc("add", "dataframe"))
+    def add(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.add, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("radd", "dataframe"))
+    def radd(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.radd, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("sub", "dataframe"))
+    def sub(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.sub, level=level, fill_value=fill_value, axis=axis
+        )
+
+    subtract = sub
+
+    @Appender(ops.make_flex_doc("rsub", "dataframe"))
+    def rsub(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rsub, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("mul", "dataframe"))
+    def mul(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.mul, level=level, fill_value=fill_value, axis=axis
+        )
+
+    multiply = mul
+
+    @Appender(ops.make_flex_doc("rmul", "dataframe"))
+    def rmul(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rmul, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("truediv", "dataframe"))
+    def truediv(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.truediv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    div = truediv
+    divide = truediv
+
+    @Appender(ops.make_flex_doc("rtruediv", "dataframe"))
+    def rtruediv(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rtruediv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    rdiv = rtruediv
+
+    @Appender(ops.make_flex_doc("floordiv", "dataframe"))
+    def floordiv(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.floordiv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rfloordiv", "dataframe"))
+    def rfloordiv(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rfloordiv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("mod", "dataframe"))
+    def mod(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.mod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rmod", "dataframe"))
+    def rmod(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rmod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("pow", "dataframe"))
+    def pow(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, operator.pow, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rpow", "dataframe"))
+    def rpow(
+        self, other, axis: Axis = "columns", level=None, fill_value=None
+    ) -> DataFrame:
+        return self._flex_arith_method(
+            other, roperator.rpow, level=level, fill_value=fill_value, axis=axis
+        )
+
+    # ----------------------------------------------------------------------
+    # Combination-Related
+
+    @doc(
+        _shared_docs["compare"],
+        dedent(
+            """
+        Returns
+        -------
+        DataFrame
+            DataFrame that shows the differences stacked side by side.
+
+            The resulting index will be a MultiIndex with 'self' and 'other'
+            stacked alternately at the inner level.
+
+        Raises
+        ------
+        ValueError
+            When the two DataFrames don't have identical labels or shape.
+
+        See Also
+        --------
+        Series.compare : Compare with another Series and show differences.
+        DataFrame.equals : Test whether two objects contain the same elements.
+
+        Notes
+        -----
+        Matching NaNs will not appear as a difference.
+
+        Can only compare identically-labeled
+        (i.e. same shape, identical row and column labels) DataFrames
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {{
+        ...         "col1": ["a", "a", "b", "b", "a"],
+        ...         "col2": [1.0, 2.0, 3.0, np.nan, 5.0],
+        ...         "col3": [1.0, 2.0, 3.0, 4.0, 5.0]
+        ...     }},
+        ...     columns=["col1", "col2", "col3"],
+        ... )
+        >>> df
+          col1  col2  col3
+        0    a   1.0   1.0
+        1    a   2.0   2.0
+        2    b   3.0   3.0
+        3    b   NaN   4.0
+        4    a   5.0   5.0
+
+        >>> df2 = df.copy()
+        >>> df2.loc[0, 'col1'] = 'c'
+        >>> df2.loc[2, 'col3'] = 4.0
+        >>> df2
+          col1  col2  col3
+        0    c   1.0   1.0
+        1    a   2.0   2.0
+        2    b   3.0   4.0
+        3    b   NaN   4.0
+        4    a   5.0   5.0
+
+        Align the differences on columns
+
+        >>> df.compare(df2)
+          col1       col3
+          self other self other
+        0    a     c  NaN   NaN
+        2  NaN   NaN  3.0   4.0
+
+        Assign result_names
+
+        >>> df.compare(df2, result_names=("left", "right"))
+          col1       col3
+          left right left right
+        0    a     c  NaN   NaN
+        2  NaN   NaN  3.0   4.0
+
+        Stack the differences on rows
+
+        >>> df.compare(df2, align_axis=0)
+                col1  col3
+        0 self     a   NaN
+          other    c   NaN
+        2 self   NaN   3.0
+          other  NaN   4.0
+
+        Keep the equal values
+
+        >>> df.compare(df2, keep_equal=True)
+          col1       col3
+          self other self other
+        0    a     c  1.0   1.0
+        2    b     b  3.0   4.0
+
+        Keep all original rows and columns
+
+        >>> df.compare(df2, keep_shape=True)
+          col1       col2       col3
+          self other self other self other
+        0    a     c  NaN   NaN  NaN   NaN
+        1  NaN   NaN  NaN   NaN  NaN   NaN
+        2  NaN   NaN  NaN   NaN  3.0   4.0
+        3  NaN   NaN  NaN   NaN  NaN   NaN
+        4  NaN   NaN  NaN   NaN  NaN   NaN
+
+        Keep all original rows and columns and also all original values
+
+        >>> df.compare(df2, keep_shape=True, keep_equal=True)
+          col1       col2       col3
+          self other self other self other
+        0    a     c  1.0   1.0  1.0   1.0
+        1    a     a  2.0   2.0  2.0   2.0
+        2    b     b  3.0   3.0  3.0   4.0
+        3    b     b  NaN   NaN  4.0   4.0
+        4    a     a  5.0   5.0  5.0   5.0
+        """
+        ),
+        klass=_shared_doc_kwargs["klass"],
+    )
+    def compare(
+        self,
+        other: DataFrame,
+        align_axis: Axis = 1,
+        keep_shape: bool = False,
+        keep_equal: bool = False,
+        result_names: Suffixes = ("self", "other"),
+    ) -> DataFrame:
+        return super().compare(
+            other=other,
+            align_axis=align_axis,
+            keep_shape=keep_shape,
+            keep_equal=keep_equal,
+            result_names=result_names,
+        )
+
+    def combine(
+        self,
+        other: DataFrame,
+        func: Callable[[Series, Series], Series | Hashable],
+        fill_value=None,
+        overwrite: bool = True,
+    ) -> DataFrame:
+        """
+        Perform column-wise combine with another DataFrame.
+
+        Combines a DataFrame with `other` DataFrame using `func`
+        to element-wise combine columns. The row and column indexes of the
+        resulting DataFrame will be the union of the two.
+
+        Parameters
+        ----------
+        other : DataFrame
+            The DataFrame to merge column-wise.
+        func : function
+            Function that takes two series as inputs and return a Series or a
+            scalar. Used to merge the two dataframes column by columns.
+        fill_value : scalar value, default None
+            The value to fill NaNs with prior to passing any column to the
+            merge func.
+        overwrite : bool, default True
+            If True, columns in `self` that do not exist in `other` will be
+            overwritten with NaNs.
+
+        Returns
+        -------
+        DataFrame
+            Combination of the provided DataFrames.
+
+        See Also
+        --------
+        DataFrame.combine_first : Combine two DataFrame objects and default to
+            non-null values in frame calling the method.
+
+        Examples
+        --------
+        Combine using a simple function that chooses the smaller column.
+
+        >>> df1 = pd.DataFrame({"A": [0, 0], "B": [4, 4]})
+        >>> df2 = pd.DataFrame({"A": [1, 1], "B": [3, 3]})
+        >>> take_smaller = lambda s1, s2: s1 if s1.sum() < s2.sum() else s2
+        >>> df1.combine(df2, take_smaller)
+           A  B
+        0  0  3
+        1  0  3
+
+        Example using a true element-wise combine function.
+
+        >>> df1 = pd.DataFrame({"A": [5, 0], "B": [2, 4]})
+        >>> df2 = pd.DataFrame({"A": [1, 1], "B": [3, 3]})
+        >>> df1.combine(df2, np.minimum)
+           A  B
+        0  1  2
+        1  0  3
+
+        Using `fill_value` fills Nones prior to passing the column to the
+        merge function.
+
+        >>> df1 = pd.DataFrame({"A": [0, 0], "B": [None, 4]})
+        >>> df2 = pd.DataFrame({"A": [1, 1], "B": [3, 3]})
+        >>> df1.combine(df2, take_smaller, fill_value=-5)
+           A    B
+        0  0 -5.0
+        1  0  4.0
+
+        However, if the same element in both dataframes is None, that None
+        is preserved
+
+        >>> df1 = pd.DataFrame({"A": [0, 0], "B": [None, 4]})
+        >>> df2 = pd.DataFrame({"A": [1, 1], "B": [None, 3]})
+        >>> df1.combine(df2, take_smaller, fill_value=-5)
+            A    B
+        0  0 -5.0
+        1  0  3.0
+
+        Example that demonstrates the use of `overwrite` and behavior when
+        the axis differ between the dataframes.
+
+        >>> df1 = pd.DataFrame({"A": [0, 0], "B": [4, 4]})
+        >>> df2 = pd.DataFrame(
+        ...     {
+        ...         "B": [3, 3],
+        ...         "C": [-10, 1],
+        ...     },
+        ...     index=[1, 2],
+        ... )
+        >>> df1.combine(df2, take_smaller)
+             A    B     C
+        0  NaN  NaN   NaN
+        1  NaN  3.0 -10.0
+        2  NaN  3.0   1.0
+
+        >>> df1.combine(df2, take_smaller, overwrite=False)
+             A    B     C
+        0  0.0  NaN   NaN
+        1  0.0  3.0 -10.0
+        2  NaN  3.0   1.0
+
+        Demonstrating the preference of the passed in dataframe.
+
+        >>> df2 = pd.DataFrame(
+        ...     {
+        ...         "B": [3, 3],
+        ...         "C": [1, 1],
+        ...     },
+        ...     index=[1, 2],
+        ... )
+        >>> df2.combine(df1, take_smaller)
+           A    B   C
+        0  0.0  NaN NaN
+        1  0.0  3.0 NaN
+        2  NaN  3.0 NaN
+
+        >>> df2.combine(df1, take_smaller, overwrite=False)
+             A    B   C
+        0  0.0  NaN NaN
+        1  0.0  3.0 1.0
+        2  NaN  3.0 1.0
+        """
+        other_idxlen = len(other.index)  # save for compare
+        other_columns = other.columns
+
+        this, other = self.align(other)
+        new_index = this.index
+
+        if other.empty and len(new_index) == len(self.index):
+            return self.copy()
+
+        if self.empty and len(other) == other_idxlen:
+            return other.copy()
+
+        # preserve column order
+        new_columns = self.columns.union(other_columns, sort=False)
+        do_fill = fill_value is not None
+        result = {}
+        for col in new_columns:
+            series = this[col]
+            other_series = other[col]
+
+            this_dtype = series.dtype
+            other_dtype = other_series.dtype
+
+            this_mask = isna(series)
+            other_mask = isna(other_series)
+
+            # don't overwrite columns unnecessarily
+            # DO propagate if this column is not in the intersection
+            if not overwrite and other_mask.all():
+                result[col] = this[col].copy()
+                continue
+
+            if do_fill:
+                series = series.copy()
+                other_series = other_series.copy()
+                series[this_mask] = fill_value
+                other_series[other_mask] = fill_value
+
+            if col not in self.columns:
+                # If self DataFrame does not have col in other DataFrame,
+                # try to promote series, which is all NaN, as other_dtype.
+                new_dtype = other_dtype
+                try:
+                    series = series.astype(new_dtype)
+                except ValueError:
+                    # e.g. new_dtype is integer types
+                    pass
+            else:
+                # if we have different dtypes, possibly promote
+                new_dtype = find_common_type([this_dtype, other_dtype])
+                series = series.astype(new_dtype)
+                other_series = other_series.astype(new_dtype)
+
+            arr = func(series, other_series)
+            if isinstance(new_dtype, np.dtype):
+                # if new_dtype is an EA Dtype, then `func` is expected to return
+                # the correct dtype without any additional casting
+                # error: No overload variant of "maybe_downcast_to_dtype" matches
+                # argument types "Union[Series, Hashable]", "dtype[Any]"
+                arr = maybe_downcast_to_dtype(  # type: ignore[call-overload]
+                    arr, new_dtype
+                )
+
+            result[col] = arr
+
+        # convert_objects just in case
+        frame_result = self._constructor(result, index=new_index, columns=new_columns)
+        return frame_result.__finalize__(self, method="combine")
+
+    def combine_first(self, other: DataFrame) -> DataFrame:
+        """
+        Update null elements with value in the same location in `other`.
+
+        Combine two DataFrame objects by filling null values in one DataFrame
+        with non-null values from other DataFrame. The row and column indexes
+        of the resulting DataFrame will be the union of the two. The resulting
+        dataframe contains the 'first' dataframe values and overrides the
+        second one values where both first.loc[index, col] and
+        second.loc[index, col] are not missing values, upon calling
+        first.combine_first(second).
+
+        Parameters
+        ----------
+        other : DataFrame
+            Provided DataFrame to use to fill null values.
+
+        Returns
+        -------
+        DataFrame
+            The result of combining the provided DataFrame with the other object.
+
+        See Also
+        --------
+        DataFrame.combine : Perform series-wise operation on two DataFrames
+            using a given function.
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame({"A": [None, 0], "B": [None, 4]})
+        >>> df2 = pd.DataFrame({"A": [1, 1], "B": [3, 3]})
+        >>> df1.combine_first(df2)
+             A    B
+        0  1.0  3.0
+        1  0.0  4.0
+
+        Null values still persist if the location of that null value
+        does not exist in `other`
+
+        >>> df1 = pd.DataFrame({"A": [None, 0], "B": [4, None]})
+        >>> df2 = pd.DataFrame({"B": [3, 3], "C": [1, 1]}, index=[1, 2])
+        >>> df1.combine_first(df2)
+             A    B    C
+        0  NaN  4.0  NaN
+        1  0.0  3.0  1.0
+        2  NaN  3.0  1.0
+        """
+        from pandas.core.computation import expressions
+
+        def combiner(x: Series, y: Series):
+            mask = x.isna()._values
+
+            x_values = x._values
+            y_values = y._values
+
+            # If the column y in other DataFrame is not in first DataFrame,
+            # just return y_values.
+            if y.name not in self.columns:
+                return y_values
+
+            return expressions.where(mask, y_values, x_values)
+
+        if len(other) == 0:
+            combined = self.reindex(
+                self.columns.append(other.columns.difference(self.columns)), axis=1
+            )
+            combined = combined.astype(other.dtypes)
+        else:
+            combined = self.combine(other, combiner, overwrite=False)
+
+        dtypes = {
+            col: find_common_type([self.dtypes[col], other.dtypes[col]])
+            for col in self.columns.intersection(other.columns)
+            if combined.dtypes[col] != self.dtypes[col]
+        }
+
+        if dtypes:
+            combined = combined.astype(dtypes)
+
+        return combined.__finalize__(self, method="combine_first")
+
+    def update(
+        self,
+        other,
+        join: UpdateJoin = "left",
+        overwrite: bool = True,
+        filter_func=None,
+        errors: IgnoreRaise = "ignore",
+    ) -> None:
+        """
+        Modify in place using non-NA values from another DataFrame.
+
+        Aligns on indices. There is no return value.
+
+        Parameters
+        ----------
+        other : DataFrame, or object coercible into a DataFrame
+            Should have at least one matching index/column label
+            with the original DataFrame. If a Series is passed,
+            its name attribute must be set, and that will be
+            used as the column name to align with the original DataFrame.
+        join : {'left'}, default 'left'
+            Only left join is implemented, keeping the index and columns of the
+            original object.
+        overwrite : bool, default True
+            How to handle non-NA values for overlapping keys:
+
+            * True: overwrite original DataFrame's values
+              with values from `other`.
+            * False: only update values that are NA in
+              the original DataFrame.
+
+        filter_func : callable(1d-array) -> bool 1d-array, optional
+            Can choose to replace values other than NA. Return True for values
+            that should be updated.
+        errors : {'raise', 'ignore'}, default 'ignore'
+            If 'raise', will raise a ValueError if the DataFrame and `other`
+            both contain non-NA data in the same place.
+
+        Returns
+        -------
+        None
+            This method directly changes calling object.
+
+        Raises
+        ------
+        ValueError
+            * When `errors='raise'` and there's overlapping non-NA data.
+            * When `errors` is not either `'ignore'` or `'raise'`
+        NotImplementedError
+            * If `join != 'left'`
+
+        See Also
+        --------
+        dict.update : Similar method for dictionaries.
+        DataFrame.merge : For column(s)-on-column(s) operations.
+
+        Notes
+        -----
+        1. Duplicate indices on `other` are not supported and raises `ValueError`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [400, 500, 600]})
+        >>> new_df = pd.DataFrame({"B": [4, 5, 6], "C": [7, 8, 9]})
+        >>> df.update(new_df)
+        >>> df
+           A  B
+        0  1  4
+        1  2  5
+        2  3  6
+
+        The DataFrame's length does not increase as a result of the update,
+        only values at matching index/column labels are updated.
+
+        >>> df = pd.DataFrame({"A": ["a", "b", "c"], "B": ["x", "y", "z"]})
+        >>> new_df = pd.DataFrame({"B": ["d", "e", "f", "g", "h", "i"]})
+        >>> df.update(new_df)
+        >>> df
+           A  B
+        0  a  d
+        1  b  e
+        2  c  f
+
+        >>> df = pd.DataFrame({"A": ["a", "b", "c"], "B": ["x", "y", "z"]})
+        >>> new_df = pd.DataFrame({"B": ["d", "f"]}, index=[0, 2])
+        >>> df.update(new_df)
+        >>> df
+           A  B
+        0  a  d
+        1  b  y
+        2  c  f
+
+        For Series, its name attribute must be set.
+
+        >>> df = pd.DataFrame({"A": ["a", "b", "c"], "B": ["x", "y", "z"]})
+        >>> new_column = pd.Series(["d", "e", "f"], name="B")
+        >>> df.update(new_column)
+        >>> df
+           A  B
+        0  a  d
+        1  b  e
+        2  c  f
+
+        If `other` contains NaNs the corresponding values are not updated
+        in the original dataframe.
+
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [400.0, 500.0, 600.0]})
+        >>> new_df = pd.DataFrame({"B": [4, np.nan, 6]})
+        >>> df.update(new_df)
+        >>> df
+           A      B
+        0  1    4.0
+        1  2  500.0
+        2  3    6.0
+        """
+        if not PYPY:
+            if sys.getrefcount(self) <= REF_COUNT:
+                warnings.warn(
+                    _chained_assignment_method_msg,
+                    ChainedAssignmentError,
+                    stacklevel=2,
+                )
+
+        # TODO: Support other joins
+        if join != "left":  # pragma: no cover
+            raise NotImplementedError("Only left join is supported")
+        if errors not in ["ignore", "raise"]:
+            raise ValueError("The parameter errors must be either 'ignore' or 'raise'")
+
+        if not isinstance(other, DataFrame):
+            other = DataFrame(other)
+
+        if other.index.has_duplicates:
+            raise ValueError("Update not allowed with duplicate indexes on other.")
+
+        index_intersection = other.index.intersection(self.index)
+        if index_intersection.empty:
+            raise ValueError(
+                "Update not allowed when the index on `other` has no intersection "
+                "with this dataframe."
+            )
+
+        other = other.reindex(index_intersection)
+        this_data = self.loc[index_intersection]
+
+        for col in self.columns.intersection(other.columns):
+            this = this_data[col]
+            that = other[col]
+
+            if filter_func is not None:
+                mask = ~filter_func(this) | isna(that)
+            else:
+                if errors == "raise":
+                    mask_this = notna(that)
+                    mask_that = notna(this)
+                    if any(mask_this & mask_that):
+                        raise ValueError("Data overlaps.")
+
+                if overwrite:
+                    mask = isna(that)
+                else:
+                    mask = notna(this)
+
+            # don't overwrite columns unnecessarily
+            if mask.all():
+                continue
+
+            self.loc[index_intersection, col] = this.where(mask, that)
+
+    # ----------------------------------------------------------------------
+    # Data reshaping
+    @Appender(
+        dedent(
+            """
+        Examples
+        --------
+        >>> df = pd.DataFrame({'Animal': ['Falcon', 'Falcon',
+        ...                               'Parrot', 'Parrot'],
+        ...                    'Max Speed': [380., 370., 24., 26.]})
+        >>> df
+           Animal  Max Speed
+        0  Falcon      380.0
+        1  Falcon      370.0
+        2  Parrot       24.0
+        3  Parrot       26.0
+        >>> df.groupby(['Animal']).mean()
+                Max Speed
+        Animal
+        Falcon      375.0
+        Parrot       25.0
+
+        **Hierarchical Indexes**
+
+        We can groupby different levels of a hierarchical index
+        using the `level` parameter:
+
+        >>> arrays = [['Falcon', 'Falcon', 'Parrot', 'Parrot'],
+        ...           ['Captive', 'Wild', 'Captive', 'Wild']]
+        >>> index = pd.MultiIndex.from_arrays(arrays, names=('Animal', 'Type'))
+        >>> df = pd.DataFrame({'Max Speed': [390., 350., 30., 20.]},
+        ...                   index=index)
+        >>> df
+                        Max Speed
+        Animal Type
+        Falcon Captive      390.0
+               Wild         350.0
+        Parrot Captive       30.0
+               Wild          20.0
+        >>> df.groupby(level=0).mean()
+                Max Speed
+        Animal
+        Falcon      370.0
+        Parrot       25.0
+        >>> df.groupby(level="Type").mean()
+                 Max Speed
+        Type
+        Captive      210.0
+        Wild         185.0
+
+        We can also choose to include NA in group keys or not by setting
+        `dropna` parameter, the default setting is `True`.
+
+        >>> arr = [[1, 2, 3], [1, None, 4], [2, 1, 3], [1, 2, 2]]
+        >>> df = pd.DataFrame(arr, columns=["a", "b", "c"])
+
+        >>> df.groupby(by=["b"]).sum()
+            a   c
+        b
+        1.0 2   3
+        2.0 2   5
+
+        >>> df.groupby(by=["b"], dropna=False).sum()
+            a   c
+        b
+        1.0 2   3
+        2.0 2   5
+        NaN 1   4
+
+        >>> arr = [["a", 12, 12], [None, 12.3, 33.], ["b", 12.3, 123], ["a", 1, 1]]
+        >>> df = pd.DataFrame(arr, columns=["a", "b", "c"])
+
+        >>> df.groupby(by="a").sum()
+            b     c
+        a
+        a   13.0   13.0
+        b   12.3  123.0
+
+        >>> df.groupby(by="a", dropna=False).sum()
+            b     c
+        a
+        a   13.0   13.0
+        b   12.3  123.0
+        NaN 12.3   33.0
+
+        When using ``.apply()``, use ``group_keys`` to include or exclude the
+        group keys. The ``group_keys`` argument defaults to ``True`` (include).
+
+        >>> df = pd.DataFrame({'Animal': ['Falcon', 'Falcon',
+        ...                               'Parrot', 'Parrot'],
+        ...                    'Max Speed': [380., 370., 24., 26.]})
+        >>> df.groupby("Animal", group_keys=True)[['Max Speed']].apply(lambda x: x)
+                  Max Speed
+        Animal
+        Falcon 0      380.0
+               1      370.0
+        Parrot 2       24.0
+               3       26.0
+
+        >>> df.groupby("Animal", group_keys=False)[['Max Speed']].apply(lambda x: x)
+           Max Speed
+        0      380.0
+        1      370.0
+        2       24.0
+        3       26.0
+        """
+        )
+    )
+    @Appender(_shared_docs["groupby"] % _shared_doc_kwargs)
+    def groupby(
+        self,
+        by=None,
+        level: IndexLabel | None = None,
+        as_index: bool = True,
+        sort: bool = True,
+        group_keys: bool = True,
+        observed: bool = True,
+        dropna: bool = True,
+    ) -> DataFrameGroupBy:
+        from pandas.core.groupby.generic import DataFrameGroupBy
+
+        if level is None and by is None:
+            raise TypeError("You have to supply one of 'by' and 'level'")
+
+        return DataFrameGroupBy(
+            obj=self,
+            keys=by,
+            level=level,
+            as_index=as_index,
+            sort=sort,
+            group_keys=group_keys,
+            observed=observed,
+            dropna=dropna,
+        )
+
+    _shared_docs["pivot"] = """
+        Return reshaped DataFrame organized by given index / column values.
+
+        Reshape data (produce a "pivot" table) based on column values. Uses
+        unique values from specified `index` / `columns` to form axes of the
+        resulting DataFrame. This function does not support data
+        aggregation, multiple values will result in a MultiIndex in the
+        columns. See the :ref:`User Guide <reshaping>` for more on reshaping.
+
+        Parameters
+        ----------%s
+        columns : str or object or a list of str
+            Column to use to make new frame's columns.
+        index : str or object or a list of str, optional
+            Column to use to make new frame's index. If not given, uses existing index.
+        values : str, object or a list of the previous, optional
+            Column(s) to use for populating new frame's values. If not
+            specified, all remaining columns will be used and the result will
+            have hierarchically indexed columns.
+
+        Returns
+        -------
+        DataFrame
+            Returns reshaped DataFrame.
+
+        Raises
+        ------
+        ValueError:
+            When there are any `index`, `columns` combinations with multiple
+            values. `DataFrame.pivot_table` when you need to aggregate.
+
+        See Also
+        --------
+        DataFrame.pivot_table : Generalization of pivot that can handle
+            duplicate values for one index/column pair.
+        DataFrame.unstack : Pivot based on the index values instead of a
+            column.
+        wide_to_long : Wide panel to long format. Less flexible but more
+            user-friendly than melt.
+
+        Notes
+        -----
+        For finer-tuned control, see hierarchical indexing documentation along
+        with the related stack/unstack methods.
+
+        Reference :ref:`the user guide <reshaping.pivot>` for more examples.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({'foo': ['one', 'one', 'one', 'two', 'two',
+        ...                            'two'],
+        ...                    'bar': ['A', 'B', 'C', 'A', 'B', 'C'],
+        ...                    'baz': [1, 2, 3, 4, 5, 6],
+        ...                    'zoo': ['x', 'y', 'z', 'q', 'w', 't']})
+        >>> df
+            foo   bar  baz  zoo
+        0   one   A    1    x
+        1   one   B    2    y
+        2   one   C    3    z
+        3   two   A    4    q
+        4   two   B    5    w
+        5   two   C    6    t
+
+        >>> df.pivot(index='foo', columns='bar', values='baz')
+        bar  A   B   C
+        foo
+        one  1   2   3
+        two  4   5   6
+
+        >>> df.pivot(index='foo', columns='bar')['baz']
+        bar  A   B   C
+        foo
+        one  1   2   3
+        two  4   5   6
+
+        >>> df.pivot(index='foo', columns='bar', values=['baz', 'zoo'])
+              baz       zoo
+        bar   A  B  C   A  B  C
+        foo
+        one   1  2  3   x  y  z
+        two   4  5  6   q  w  t
+
+        You could also assign a list of column names or a list of index names.
+
+        >>> df = pd.DataFrame({
+        ...                   "lev1": [1, 1, 1, 2, 2, 2],
+        ...                   "lev2": [1, 1, 2, 1, 1, 2],
+        ...                   "lev3": [1, 2, 1, 2, 1, 2],
+        ...                   "lev4": [1, 2, 3, 4, 5, 6],
+        ...                   "values": [0, 1, 2, 3, 4, 5]})
+        >>> df
+            lev1 lev2 lev3 lev4 values
+        0   1    1    1    1    0
+        1   1    1    2    2    1
+        2   1    2    1    3    2
+        3   2    1    2    4    3
+        4   2    1    1    5    4
+        5   2    2    2    6    5
+
+        >>> df.pivot(index="lev1", columns=["lev2", "lev3"], values="values")
+        lev2    1         2
+        lev3    1    2    1    2
+        lev1
+        1     0.0  1.0  2.0  NaN
+        2     4.0  3.0  NaN  5.0
+
+        >>> df.pivot(index=["lev1", "lev2"], columns=["lev3"], values="values")
+              lev3    1    2
+        lev1  lev2
+           1     1  0.0  1.0
+                 2  2.0  NaN
+           2     1  4.0  3.0
+                 2  NaN  5.0
+
+        A ValueError is raised if there are any duplicates.
+
+        >>> df = pd.DataFrame({"foo": ['one', 'one', 'two', 'two'],
+        ...                    "bar": ['A', 'A', 'B', 'C'],
+        ...                    "baz": [1, 2, 3, 4]})
+        >>> df
+           foo bar  baz
+        0  one   A    1
+        1  one   A    2
+        2  two   B    3
+        3  two   C    4
+
+        Notice that the first two rows are the same for our `index`
+        and `columns` arguments.
+
+        >>> df.pivot(index='foo', columns='bar', values='baz')
+        Traceback (most recent call last):
+           ...
+        ValueError: Index contains duplicate entries, cannot reshape
+        """
+
+    @Substitution("")
+    @Appender(_shared_docs["pivot"])
+    def pivot(
+        self, *, columns, index=lib.no_default, values=lib.no_default
+    ) -> DataFrame:
+        from pandas.core.reshape.pivot import pivot
+
+        return pivot(self, index=index, columns=columns, values=values)
+
+    _shared_docs["pivot_table"] = """
+        Create a spreadsheet-style pivot table as a DataFrame.
+
+        The levels in the pivot table will be stored in MultiIndex objects
+        (hierarchical indexes) on the index and columns of the result DataFrame.
+
+        Parameters
+        ----------%s
+        values : list-like or scalar, optional
+            Column or columns to aggregate.
+        index : column, Grouper, array, or list of the previous
+            Keys to group by on the pivot table index. If a list is passed,
+            it can contain any of the other types (except list). If an array is
+            passed, it must be the same length as the data and will be used in
+            the same manner as column values.
+        columns : column, Grouper, array, or list of the previous
+            Keys to group by on the pivot table column. If a list is passed,
+            it can contain any of the other types (except list). If an array is
+            passed, it must be the same length as the data and will be used in
+            the same manner as column values.
+        aggfunc : function, list of functions, dict, default "mean"
+            If a list of functions is passed, the resulting pivot table will have
+            hierarchical columns whose top level are the function names
+            (inferred from the function objects themselves).
+            If a dict is passed, the key is column to aggregate and the value is
+            function or list of functions. If ``margin=True``, aggfunc will be
+            used to calculate the partial aggregates.
+        fill_value : scalar, default None
+            Value to replace missing values with (in the resulting pivot table,
+            after aggregation).
+        margins : bool, default False
+            If ``margins=True``, special ``All`` columns and rows
+            will be added with partial group aggregates across the categories
+            on the rows and columns.
+        dropna : bool, default True
+            Do not include columns whose entries are all NaN. If True,
+            rows with a NaN value in any column will be omitted before
+            computing margins.
+        margins_name : str, default 'All'
+            Name of the row / column that will contain the totals
+            when margins is True.
+        observed : bool, default False
+            This only applies if any of the groupers are Categoricals.
+            If True: only show observed values for categorical groupers.
+            If False: show all values for categorical groupers.
+
+            .. versionchanged:: 3.0.0
+
+                The default value is now ``True``.
+
+        sort : bool, default True
+            Specifies if the result should be sorted.
+
+            .. versionadded:: 1.3.0
+
+        **kwargs : dict
+            Optional keyword arguments to pass to ``aggfunc``.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        DataFrame
+            An Excel style pivot table.
+
+        See Also
+        --------
+        DataFrame.pivot : Pivot without aggregation that can handle
+            non-numeric data.
+        DataFrame.melt: Unpivot a DataFrame from wide to long format,
+            optionally leaving identifiers set.
+        wide_to_long : Wide panel to long format. Less flexible but more
+            user-friendly than melt.
+
+        Notes
+        -----
+        Reference :ref:`the user guide <reshaping.pivot>` for more examples.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": ["foo", "foo", "foo", "foo", "foo",
+        ...                          "bar", "bar", "bar", "bar"],
+        ...                    "B": ["one", "one", "one", "two", "two",
+        ...                          "one", "one", "two", "two"],
+        ...                    "C": ["small", "large", "large", "small",
+        ...                          "small", "large", "small", "small",
+        ...                          "large"],
+        ...                    "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+        ...                    "E": [2, 4, 5, 5, 6, 6, 8, 9, 9]})
+        >>> df
+             A    B      C  D  E
+        0  foo  one  small  1  2
+        1  foo  one  large  2  4
+        2  foo  one  large  2  5
+        3  foo  two  small  3  5
+        4  foo  two  small  3  6
+        5  bar  one  large  4  6
+        6  bar  one  small  5  8
+        7  bar  two  small  6  9
+        8  bar  two  large  7  9
+
+        This first example aggregates values by taking the sum.
+
+        >>> table = pd.pivot_table(df, values='D', index=['A', 'B'],
+        ...                        columns=['C'], aggfunc="sum")
+        >>> table
+        C        large  small
+        A   B
+        bar one    4.0    5.0
+            two    7.0    6.0
+        foo one    4.0    1.0
+            two    NaN    6.0
+
+        We can also fill missing values using the `fill_value` parameter.
+
+        >>> table = pd.pivot_table(df, values='D', index=['A', 'B'],
+        ...                        columns=['C'], aggfunc="sum", fill_value=0)
+        >>> table
+        C        large  small
+        A   B
+        bar one      4      5
+            two      7      6
+        foo one      4      1
+            two      0      6
+
+        The next example aggregates by taking the mean across multiple columns.
+
+        >>> table = pd.pivot_table(df, values=['D', 'E'], index=['A', 'C'],
+        ...                        aggfunc={'D': "mean", 'E': "mean"})
+        >>> table
+                        D         E
+        A   C
+        bar large  5.500000  7.500000
+            small  5.500000  8.500000
+        foo large  2.000000  4.500000
+            small  2.333333  4.333333
+
+        We can also calculate multiple types of aggregations for any given
+        value column.
+
+        >>> table = pd.pivot_table(df, values=['D', 'E'], index=['A', 'C'],
+        ...                        aggfunc={'D': "mean",
+        ...                                 'E': ["min", "max", "mean"]})
+        >>> table
+                          D   E
+                       mean max      mean  min
+        A   C
+        bar large  5.500000   9  7.500000    6
+            small  5.500000   9  8.500000    8
+        foo large  2.000000   5  4.500000    4
+            small  2.333333   6  4.333333    2
+        """
+
+    @Substitution("")
+    @Appender(_shared_docs["pivot_table"])
+    def pivot_table(
+        self,
+        values=None,
+        index=None,
+        columns=None,
+        aggfunc: AggFuncType = "mean",
+        fill_value=None,
+        margins: bool = False,
+        dropna: bool = True,
+        margins_name: Level = "All",
+        observed: bool = True,
+        sort: bool = True,
+        **kwargs,
+    ) -> DataFrame:
+        from pandas.core.reshape.pivot import pivot_table
+
+        return pivot_table(
+            self,
+            values=values,
+            index=index,
+            columns=columns,
+            aggfunc=aggfunc,
+            fill_value=fill_value,
+            margins=margins,
+            dropna=dropna,
+            margins_name=margins_name,
+            observed=observed,
+            sort=sort,
+            **kwargs,
+        )
+
+    def stack(
+        self,
+        level: IndexLabel = -1,
+        dropna: bool | lib.NoDefault = lib.no_default,
+        sort: bool | lib.NoDefault = lib.no_default,
+        future_stack: bool = True,
+    ):
+        """
+        Stack the prescribed level(s) from columns to index.
+
+        Return a reshaped DataFrame or Series having a multi-level
+        index with one or more new inner-most levels compared to the current
+        DataFrame. The new inner-most levels are created by pivoting the
+        columns of the current dataframe:
+
+        - if the columns have a single level, the output is a Series;
+        - if the columns have multiple levels, the new index level(s) is (are)
+          taken from the prescribed level(s) and the output is a DataFrame.
+
+        Parameters
+        ----------
+        level : int, str, list, default -1
+            Level(s) to stack from the column axis onto the index
+            axis, defined as one index or label, or a list of indices
+            or labels.
+        dropna : bool, default True
+            Whether to drop rows in the resulting Frame/Series with
+            missing values. Stacking a column level onto the index
+            axis can create combinations of index and column values
+            that are missing from the original dataframe. See Examples
+            section.
+        sort : bool, default True
+            Whether to sort the levels of the resulting MultiIndex.
+        future_stack : bool, default True
+            Whether to use the new implementation that will replace the current
+            implementation in pandas 3.0. When True, dropna and sort have no impact
+            on the result and must remain unspecified. See :ref:`pandas 2.1.0 Release
+            notes <whatsnew_210.enhancements.new_stack>` for more details.
+
+        Returns
+        -------
+        DataFrame or Series
+            Stacked dataframe or series.
+
+        See Also
+        --------
+        DataFrame.unstack : Unstack prescribed level(s) from index axis
+             onto column axis.
+        DataFrame.pivot : Reshape dataframe from long format to wide
+             format.
+        DataFrame.pivot_table : Create a spreadsheet-style pivot table
+             as a DataFrame.
+
+        Notes
+        -----
+        The function is named by analogy with a collection of books
+        being reorganized from being side by side on a horizontal
+        position (the columns of the dataframe) to being stacked
+        vertically on top of each other (in the index of the
+        dataframe).
+
+        Reference :ref:`the user guide <reshaping.stacking>` for more examples.
+
+        Examples
+        --------
+        **Single level columns**
+
+        >>> df_single_level_cols = pd.DataFrame(
+        ...     [[0, 1], [2, 3]], index=["cat", "dog"], columns=["weight", "height"]
+        ... )
+
+        Stacking a dataframe with a single level column axis returns a Series:
+
+        >>> df_single_level_cols
+             weight height
+        cat       0      1
+        dog       2      3
+        >>> df_single_level_cols.stack()
+        cat  weight    0
+             height    1
+        dog  weight    2
+             height    3
+        dtype: int64
+
+        **Multi level columns: simple case**
+
+        >>> multicol1 = pd.MultiIndex.from_tuples(
+        ...     [("weight", "kg"), ("weight", "pounds")]
+        ... )
+        >>> df_multi_level_cols1 = pd.DataFrame(
+        ...     [[1, 2], [2, 4]], index=["cat", "dog"], columns=multicol1
+        ... )
+
+        Stacking a dataframe with a multi-level column axis:
+
+        >>> df_multi_level_cols1
+             weight
+                 kg    pounds
+        cat       1        2
+        dog       2        4
+        >>> df_multi_level_cols1.stack()
+                    weight
+        cat kg           1
+            pounds       2
+        dog kg           2
+            pounds       4
+
+        **Missing values**
+
+        >>> multicol2 = pd.MultiIndex.from_tuples([("weight", "kg"), ("height", "m")])
+        >>> df_multi_level_cols2 = pd.DataFrame(
+        ...     [[1.0, 2.0], [3.0, 4.0]], index=["cat", "dog"], columns=multicol2
+        ... )
+
+        It is common to have missing values when stacking a dataframe
+        with multi-level columns, as the stacked dataframe typically
+        has more values than the original dataframe. Missing values
+        are filled with NaNs:
+
+        >>> df_multi_level_cols2
+            weight height
+                kg      m
+        cat    1.0    2.0
+        dog    3.0    4.0
+        >>> df_multi_level_cols2.stack()
+                weight  height
+        cat kg     1.0     NaN
+            m      NaN     2.0
+        dog kg     3.0     NaN
+            m      NaN     4.0
+
+        **Prescribing the level(s) to be stacked**
+
+        The first parameter controls which level or levels are stacked:
+
+        >>> df_multi_level_cols2.stack(0)
+                     kg    m
+        cat weight  1.0  NaN
+            height  NaN  2.0
+        dog weight  3.0  NaN
+            height  NaN  4.0
+        >>> df_multi_level_cols2.stack([0, 1])
+        cat  weight  kg    1.0
+             height  m     2.0
+        dog  weight  kg    3.0
+             height  m     4.0
+        dtype: float64
+        """
+        if not future_stack:
+            from pandas.core.reshape.reshape import (
+                stack,
+                stack_multiple,
+            )
+
+            warnings.warn(
+                "The previous implementation of stack is deprecated and will be "
+                "removed in a future version of pandas. See the What's New notes "
+                "for pandas 2.1.0 for details. Do not specify the future_stack "
+                "argument to adopt the new implementation and silence this warning.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+
+            if dropna is lib.no_default:
+                dropna = True
+            if sort is lib.no_default:
+                sort = True
+
+            if isinstance(level, (tuple, list)):
+                result = stack_multiple(self, level, dropna=dropna, sort=sort)
+            else:
+                result = stack(self, level, dropna=dropna, sort=sort)
+        else:
+            from pandas.core.reshape.reshape import stack_v3
+
+            if dropna is not lib.no_default:
+                raise ValueError(
+                    "dropna must be unspecified as the new "
+                    "implementation does not introduce rows of NA values. This "
+                    "argument will be removed in a future version of pandas."
+                )
+
+            if sort is not lib.no_default:
+                raise ValueError(
+                    "Cannot specify sort, this argument will be "
+                    "removed in a future version of pandas. Sort the result using "
+                    ".sort_index instead."
+                )
+
+            if (
+                isinstance(level, (tuple, list))
+                and not all(lev in self.columns.names for lev in level)
+                and not all(isinstance(lev, int) for lev in level)
+            ):
+                raise ValueError(
+                    "level should contain all level names or all level "
+                    "numbers, not a mixture of the two."
+                )
+
+            if not isinstance(level, (tuple, list)):
+                level = [level]
+            level = [self.columns._get_level_number(lev) for lev in level]
+            result = stack_v3(self, level)
+
+        return result.__finalize__(self, method="stack")
+
+    def explode(
+        self,
+        column: IndexLabel,
+        ignore_index: bool = False,
+    ) -> DataFrame:
+        """
+        Transform each element of a list-like to a row, replicating index values.
+
+        Parameters
+        ----------
+        column : IndexLabel
+            Column(s) to explode.
+            For multiple columns, specify a non-empty list with each element
+            be str or tuple, and all specified columns their list-like data
+            on same row of the frame must have matching length.
+
+            .. versionadded:: 1.3.0
+                Multi-column explode
+
+        ignore_index : bool, default False
+            If True, the resulting index will be labeled 0, 1, …, n - 1.
+
+        Returns
+        -------
+        DataFrame
+            Exploded lists to rows of the subset columns;
+            index will be duplicated for these rows.
+
+        Raises
+        ------
+        ValueError :
+            * If columns of the frame are not unique.
+            * If specified columns to explode is empty list.
+            * If specified columns to explode have not matching count of
+              elements rowwise in the frame.
+
+        See Also
+        --------
+        DataFrame.unstack : Pivot a level of the (necessarily hierarchical)
+            index labels.
+        DataFrame.melt : Unpivot a DataFrame from wide format to long format.
+        Series.explode : Explode a DataFrame from list-like columns to long format.
+
+        Notes
+        -----
+        This routine will explode list-likes including lists, tuples, sets,
+        Series, and np.ndarray. The result dtype of the subset rows will
+        be object. Scalars will be returned unchanged, and empty list-likes will
+        result in a np.nan for that row. In addition, the ordering of rows in the
+        output will be non-deterministic when exploding sets.
+
+        Reference :ref:`the user guide <reshaping.explode>` for more examples.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": [[0, 1, 2], "foo", [], [3, 4]],
+        ...         "B": 1,
+        ...         "C": [["a", "b", "c"], np.nan, [], ["d", "e"]],
+        ...     }
+        ... )
+        >>> df
+                   A  B          C
+        0  [0, 1, 2]  1  [a, b, c]
+        1        foo  1        NaN
+        2         []  1         []
+        3     [3, 4]  1     [d, e]
+
+        Single-column explode.
+
+        >>> df.explode("A")
+             A  B          C
+        0    0  1  [a, b, c]
+        0    1  1  [a, b, c]
+        0    2  1  [a, b, c]
+        1  foo  1        NaN
+        2  NaN  1         []
+        3    3  1     [d, e]
+        3    4  1     [d, e]
+
+        Multi-column explode.
+
+        >>> df.explode(list("AC"))
+             A  B    C
+        0    0  1    a
+        0    1  1    b
+        0    2  1    c
+        1  foo  1  NaN
+        2  NaN  1  NaN
+        3    3  1    d
+        3    4  1    e
+        """
+        if not self.columns.is_unique:
+            duplicate_cols = self.columns[self.columns.duplicated()].tolist()
+            raise ValueError(
+                f"DataFrame columns must be unique. Duplicate columns: {duplicate_cols}"
+            )
+
+        columns: list[Hashable]
+        if is_scalar(column) or isinstance(column, tuple):
+            columns = [column]
+        elif isinstance(column, list) and all(
+            is_scalar(c) or isinstance(c, tuple) for c in column
+        ):
+            if not column:
+                raise ValueError("column must be nonempty")
+            if len(column) > len(set(column)):
+                raise ValueError("column must be unique")
+            columns = column
+        else:
+            raise ValueError("column must be a scalar, tuple, or list thereof")
+
+        df = self.reset_index(drop=True)
+        if len(columns) == 1:
+            result = df[columns[0]].explode()
+        else:
+            mylen = lambda x: len(x) if (is_list_like(x) and len(x) > 0) else 1
+            counts0 = self[columns[0]].apply(mylen)
+            for c in columns[1:]:
+                if not all(counts0 == self[c].apply(mylen)):
+                    raise ValueError("columns must have matching element counts")
+            result = DataFrame({c: df[c].explode() for c in columns})
+        result = df.drop(columns, axis=1).join(result)
+        if ignore_index:
+            result.index = default_index(len(result))
+        else:
+            result.index = self.index.take(result.index)
+        result = result.reindex(columns=self.columns)
+
+        return result.__finalize__(self, method="explode")
+
+    def unstack(
+        self, level: IndexLabel = -1, fill_value=None, sort: bool = True
+    ) -> DataFrame | Series:
+        """
+        Pivot a level of the (necessarily hierarchical) index labels.
+
+        Returns a DataFrame having a new level of column labels whose inner-most level
+        consists of the pivoted index labels.
+
+        If the index is not a MultiIndex, the output will be a Series
+        (the analogue of stack when the columns are not a MultiIndex).
+
+        Parameters
+        ----------
+        level : int, str, or list of these, default -1 (last level)
+            Level(s) of index to unstack, can pass level name.
+        fill_value : int, str or dict
+            Replace NaN with this value if the unstack produces missing values.
+        sort : bool, default True
+            Sort the level(s) in the resulting MultiIndex columns.
+
+        Returns
+        -------
+        Series or DataFrame
+            If index is a MultiIndex: DataFrame with pivoted index labels as new
+            inner-most level column labels, else Series.
+
+        See Also
+        --------
+        DataFrame.pivot : Pivot a table based on column values.
+        DataFrame.stack : Pivot a level of the column labels (inverse operation
+            from `unstack`).
+
+        Notes
+        -----
+        Reference :ref:`the user guide <reshaping.stacking>` for more examples.
+
+        Examples
+        --------
+        >>> index = pd.MultiIndex.from_tuples(
+        ...     [("one", "a"), ("one", "b"), ("two", "a"), ("two", "b")]
+        ... )
+        >>> s = pd.Series(np.arange(1.0, 5.0), index=index)
+        >>> s
+        one  a   1.0
+             b   2.0
+        two  a   3.0
+             b   4.0
+        dtype: float64
+
+        >>> s.unstack(level=-1)
+             a   b
+        one  1.0  2.0
+        two  3.0  4.0
+
+        >>> s.unstack(level=0)
+           one  two
+        a  1.0   3.0
+        b  2.0   4.0
+
+        >>> df = s.unstack(level=0)
+        >>> df.unstack()
+        one  a  1.0
+             b  2.0
+        two  a  3.0
+             b  4.0
+        dtype: float64
+        """
+        from pandas.core.reshape.reshape import unstack
+
+        result = unstack(self, level, fill_value, sort)
+
+        return result.__finalize__(self, method="unstack")
+
+    def melt(
+        self,
+        id_vars=None,
+        value_vars=None,
+        var_name=None,
+        value_name: Hashable = "value",
+        col_level: Level | None = None,
+        ignore_index: bool = True,
+    ) -> DataFrame:
+        """
+        Unpivot DataFrame from wide to long format, optionally leaving identifiers set.
+
+        This function is useful to massage a DataFrame into a format where one
+        or more columns are identifier variables (`id_vars`), while all other
+        columns, considered measured variables (`value_vars`), are "unpivoted" to
+        the row axis, leaving just two non-identifier columns, 'variable' and
+        'value'.
+
+        Parameters
+        ----------
+        id_vars : scalar, tuple, list, or ndarray, optional
+            Column(s) to use as identifier variables.
+        value_vars : scalar, tuple, list, or ndarray, optional
+            Column(s) to unpivot. If not specified, uses all columns that
+            are not set as `id_vars`.
+        var_name : scalar, default None
+            Name to use for the 'variable' column. If None it uses
+            ``frame.columns.name`` or 'variable'.
+        value_name : scalar, default 'value'
+            Name to use for the 'value' column, can't be an existing column label.
+        col_level : scalar, optional
+            If columns are a MultiIndex then use this level to melt.
+        ignore_index : bool, default True
+            If True, original index is ignored. If False, original index is retained.
+            Index labels will be repeated as necessary.
+
+        Returns
+        -------
+        DataFrame
+            Unpivoted DataFrame.
+
+        See Also
+        --------
+        melt : Identical method.
+        pivot_table : Create a spreadsheet-style pivot table as a DataFrame.
+        DataFrame.pivot : Return reshaped DataFrame organized
+            by given index / column values.
+        DataFrame.explode : Explode a DataFrame from list-like
+                columns to long format.
+
+        Notes
+        -----
+        Reference :ref:`the user guide <reshaping.melt>` for more examples.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": {0: "a", 1: "b", 2: "c"},
+        ...         "B": {0: 1, 1: 3, 2: 5},
+        ...         "C": {0: 2, 1: 4, 2: 6},
+        ...     }
+        ... )
+        >>> df
+        A  B  C
+        0  a  1  2
+        1  b  3  4
+        2  c  5  6
+
+        >>> df.melt(id_vars=["A"], value_vars=["B"])
+        A variable  value
+        0  a        B      1
+        1  b        B      3
+        2  c        B      5
+
+        >>> df.melt(id_vars=["A"], value_vars=["B", "C"])
+        A variable  value
+        0  a        B      1
+        1  b        B      3
+        2  c        B      5
+        3  a        C      2
+        4  b        C      4
+        5  c        C      6
+
+        The names of 'variable' and 'value' columns can be customized:
+
+        >>> df.melt(
+        ...     id_vars=["A"],
+        ...     value_vars=["B"],
+        ...     var_name="myVarname",
+        ...     value_name="myValname",
+        ... )
+        A myVarname  myValname
+        0  a         B          1
+        1  b         B          3
+        2  c         B          5
+
+        Original index values can be kept around:
+
+        >>> df.melt(id_vars=["A"], value_vars=["B", "C"], ignore_index=False)
+        A variable  value
+        0  a        B      1
+        1  b        B      3
+        2  c        B      5
+        0  a        C      2
+        1  b        C      4
+        2  c        C      6
+
+        If you have multi-index columns:
+
+        >>> df.columns = [list("ABC"), list("DEF")]
+        >>> df
+        A  B  C
+        D  E  F
+        0  a  1  2
+        1  b  3  4
+        2  c  5  6
+
+        >>> df.melt(col_level=0, id_vars=["A"], value_vars=["B"])
+        A variable  value
+        0  a        B      1
+        1  b        B      3
+        2  c        B      5
+
+        >>> df.melt(id_vars=[("A", "D")], value_vars=[("B", "E")])
+        (A, D) variable_0 variable_1  value
+        0      a          B          E      1
+        1      b          B          E      3
+        2      c          B          E      5
+        """
+        return melt(
+            self,
+            id_vars=id_vars,
+            value_vars=value_vars,
+            var_name=var_name,
+            value_name=value_name,
+            col_level=col_level,
+            ignore_index=ignore_index,
+        ).__finalize__(self, method="melt")
+
+    # ----------------------------------------------------------------------
+    # Time series-related
+
+    @doc(
+        Series.diff,
+        klass="DataFrame",
+        extra_params="axis : {0 or 'index', 1 or 'columns'}, default 0\n    "
+        "Take difference over rows (0) or columns (1).\n",
+        other_klass="Series",
+        examples=dedent(
+            """
+        Difference with previous row
+
+        >>> df = pd.DataFrame({'a': [1, 2, 3, 4, 5, 6],
+        ...                    'b': [1, 1, 2, 3, 5, 8],
+        ...                    'c': [1, 4, 9, 16, 25, 36]})
+        >>> df
+           a  b   c
+        0  1  1   1
+        1  2  1   4
+        2  3  2   9
+        3  4  3  16
+        4  5  5  25
+        5  6  8  36
+
+        >>> df.diff()
+             a    b     c
+        0  NaN  NaN   NaN
+        1  1.0  0.0   3.0
+        2  1.0  1.0   5.0
+        3  1.0  1.0   7.0
+        4  1.0  2.0   9.0
+        5  1.0  3.0  11.0
+
+        Difference with previous column
+
+        >>> df.diff(axis=1)
+            a  b   c
+        0 NaN  0   0
+        1 NaN -1   3
+        2 NaN -1   7
+        3 NaN -1  13
+        4 NaN  0  20
+        5 NaN  2  28
+
+        Difference with 3rd previous row
+
+        >>> df.diff(periods=3)
+             a    b     c
+        0  NaN  NaN   NaN
+        1  NaN  NaN   NaN
+        2  NaN  NaN   NaN
+        3  3.0  2.0  15.0
+        4  3.0  4.0  21.0
+        5  3.0  6.0  27.0
+
+        Difference with following row
+
+        >>> df.diff(periods=-1)
+             a    b     c
+        0 -1.0  0.0  -3.0
+        1 -1.0 -1.0  -5.0
+        2 -1.0 -1.0  -7.0
+        3 -1.0 -2.0  -9.0
+        4 -1.0 -3.0 -11.0
+        5  NaN  NaN   NaN
+
+        Overflow in input dtype
+
+        >>> df = pd.DataFrame({'a': [1, 0]}, dtype=np.uint8)
+        >>> df.diff()
+               a
+        0    NaN
+        1  255.0"""
+        ),
+    )
+    def diff(self, periods: int = 1, axis: Axis = 0) -> DataFrame:
+        if not lib.is_integer(periods):
+            if not (is_float(periods) and periods.is_integer()):
+                raise ValueError("periods must be an integer")
+            periods = int(periods)
+
+        axis = self._get_axis_number(axis)
+        if axis == 1:
+            if periods != 0:
+                # in the periods == 0 case, this is equivalent diff of 0 periods
+                #  along axis=0, and the Manager method may be somewhat more
+                #  performant, so we dispatch in that case.
+                return self - self.shift(periods, axis=axis)
+            # With periods=0 this is equivalent to a diff with axis=0
+            axis = 0
+
+        new_data = self._mgr.diff(n=periods)
+        res_df = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        return res_df.__finalize__(self, "diff")
+
+    # ----------------------------------------------------------------------
+    # Function application
+
+    def _gotitem(
+        self,
+        key: IndexLabel,
+        ndim: int,
+        subset: DataFrame | Series | None = None,
+    ) -> DataFrame | Series:
+        """
+        Sub-classes to define. Return a sliced object.
+
+        Parameters
+        ----------
+        key : string / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        if subset is None:
+            subset = self
+        elif subset.ndim == 1:  # is Series
+            return subset
+
+        # TODO: _shallow_copy(subset)?
+        return subset[key]
+
+    _agg_see_also_doc = dedent(
+        """
+    See Also
+    --------
+    DataFrame.apply : Perform any type of operations.
+    DataFrame.transform : Perform transformation type operations.
+    DataFrame.groupby : Perform operations over groups.
+    DataFrame.resample : Perform operations over resampled bins.
+    DataFrame.rolling : Perform operations over rolling window.
+    DataFrame.expanding : Perform operations over expanding window.
+    core.window.ewm.ExponentialMovingWindow : Perform operation over exponential
+        weighted window.
+    """
+    )
+
+    _agg_examples_doc = dedent(
+        """
+    Examples
+    --------
+    >>> df = pd.DataFrame([[1, 2, 3],
+    ...                    [4, 5, 6],
+    ...                    [7, 8, 9],
+    ...                    [np.nan, np.nan, np.nan]],
+    ...                   columns=['A', 'B', 'C'])
+
+    Aggregate these functions over the rows.
+
+    >>> df.agg(['sum', 'min'])
+            A     B     C
+    sum  12.0  15.0  18.0
+    min   1.0   2.0   3.0
+
+    Different aggregations per column.
+
+    >>> df.agg({'A' : ['sum', 'min'], 'B' : ['min', 'max']})
+            A    B
+    sum  12.0  NaN
+    min   1.0  2.0
+    max   NaN  8.0
+
+    Aggregate different functions over the columns and rename the index of the resulting
+    DataFrame.
+
+    >>> df.agg(x=('A', 'max'), y=('B', 'min'), z=('C', 'mean'))
+         A    B    C
+    x  7.0  NaN  NaN
+    y  NaN  2.0  NaN
+    z  NaN  NaN  6.0
+
+    Aggregate over the columns.
+
+    >>> df.agg("mean", axis="columns")
+    0    2.0
+    1    5.0
+    2    8.0
+    3    NaN
+    dtype: float64
+    """
+    )
+
+    @doc(
+        _shared_docs["aggregate"],
+        klass=_shared_doc_kwargs["klass"],
+        axis=_shared_doc_kwargs["axis"],
+        see_also=_agg_see_also_doc,
+        examples=_agg_examples_doc,
+    )
+    def aggregate(self, func=None, axis: Axis = 0, *args, **kwargs):
+        from pandas.core.apply import frame_apply
+
+        axis = self._get_axis_number(axis)
+
+        op = frame_apply(self, func=func, axis=axis, args=args, kwargs=kwargs)
+        result = op.agg()
+        result = reconstruct_and_relabel_result(result, func, **kwargs)
+        return result
+
+    agg = aggregate
+
+    @doc(
+        _shared_docs["transform"],
+        klass=_shared_doc_kwargs["klass"],
+        axis=_shared_doc_kwargs["axis"],
+    )
+    def transform(
+        self, func: AggFuncType, axis: Axis = 0, *args, **kwargs
+    ) -> DataFrame:
+        from pandas.core.apply import frame_apply
+
+        op = frame_apply(self, func=func, axis=axis, args=args, kwargs=kwargs)
+        result = op.transform()
+        assert isinstance(result, DataFrame)
+        return result
+
+    def apply(
+        self,
+        func: AggFuncType,
+        axis: Axis = 0,
+        raw: bool = False,
+        result_type: Literal["expand", "reduce", "broadcast"] | None = None,
+        args=(),
+        by_row: Literal[False, "compat"] = "compat",
+        engine: Literal["python", "numba"] = "python",
+        engine_kwargs: dict[str, bool] | None = None,
+        **kwargs,
+    ):
+        """
+        Apply a function along an axis of the DataFrame.
+
+        Objects passed to the function are Series objects whose index is
+        either the DataFrame's index (``axis=0``) or the DataFrame's columns
+        (``axis=1``). By default (``result_type=None``), the final return type
+        is inferred from the return type of the applied function. Otherwise,
+        it depends on the `result_type` argument.
+
+        Parameters
+        ----------
+        func : function
+            Function to apply to each column or row.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Axis along which the function is applied:
+
+            * 0 or 'index': apply function to each column.
+            * 1 or 'columns': apply function to each row.
+
+        raw : bool, default False
+            Determines if row or column is passed as a Series or ndarray object:
+
+            * ``False`` : passes each row or column as a Series to the
+              function.
+            * ``True`` : the passed function will receive ndarray objects
+              instead.
+              If you are just applying a NumPy reduction function this will
+              achieve much better performance.
+
+        result_type : {'expand', 'reduce', 'broadcast', None}, default None
+            These only act when ``axis=1`` (columns):
+
+            * 'expand' : list-like results will be turned into columns.
+            * 'reduce' : returns a Series if possible rather than expanding
+              list-like results. This is the opposite of 'expand'.
+            * 'broadcast' : results will be broadcast to the original shape
+              of the DataFrame, the original index and columns will be
+              retained.
+
+            The default behaviour (None) depends on the return value of the
+            applied function: list-like results will be returned as a Series
+            of those. However if the apply function returns a Series these
+            are expanded to columns.
+        args : tuple
+            Positional arguments to pass to `func` in addition to the
+            array/series.
+        by_row : False or "compat", default "compat"
+            Only has an effect when ``func`` is a listlike or dictlike of funcs
+            and the func isn't a string.
+            If "compat", will if possible first translate the func into pandas
+            methods (e.g. ``Series().apply(np.sum)`` will be translated to
+            ``Series().sum()``). If that doesn't work, will try call to apply again with
+            ``by_row=True`` and if that fails, will call apply again with
+            ``by_row=False`` (backward compatible).
+            If False, the funcs will be passed the whole Series at once.
+
+            .. versionadded:: 2.1.0
+
+        engine : {'python', 'numba'}, default 'python'
+            Choose between the python (default) engine or the numba engine in apply.
+
+            The numba engine will attempt to JIT compile the passed function,
+            which may result in speedups for large DataFrames.
+            It also supports the following engine_kwargs :
+
+            - nopython (compile the function in nopython mode)
+            - nogil (release the GIL inside the JIT compiled function)
+            - parallel (try to apply the function in parallel over the DataFrame)
+
+              Note: Due to limitations within numba/how pandas interfaces with numba,
+              you should only use this if raw=True
+
+            Note: The numba compiler only supports a subset of
+            valid Python/numpy operations.
+
+            Please read more about the `supported python features
+            <https://numba.pydata.org/numba-doc/dev/reference/pysupported.html>`_
+            and `supported numpy features
+            <https://numba.pydata.org/numba-doc/dev/reference/numpysupported.html>`_
+            in numba to learn what you can or cannot use in the passed function.
+
+            .. versionadded:: 2.2.0
+
+        engine_kwargs : dict
+            Pass keyword arguments to the engine.
+            This is currently only used by the numba engine,
+            see the documentation for the engine argument for more information.
+        **kwargs
+            Additional keyword arguments to pass as keywords arguments to
+            `func`.
+
+        Returns
+        -------
+        Series or DataFrame
+            Result of applying ``func`` along the given axis of the
+            DataFrame.
+
+        See Also
+        --------
+        DataFrame.map: For elementwise operations.
+        DataFrame.aggregate: Only perform aggregating type operations.
+        DataFrame.transform: Only perform transforming type operations.
+
+        Notes
+        -----
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[4, 9]] * 3, columns=["A", "B"])
+        >>> df
+           A  B
+        0  4  9
+        1  4  9
+        2  4  9
+
+        Using a numpy universal function (in this case the same as
+        ``np.sqrt(df)``):
+
+        >>> df.apply(np.sqrt)
+             A    B
+        0  2.0  3.0
+        1  2.0  3.0
+        2  2.0  3.0
+
+        Using a reducing function on either axis
+
+        >>> df.apply(np.sum, axis=0)
+        A    12
+        B    27
+        dtype: int64
+
+        >>> df.apply(np.sum, axis=1)
+        0    13
+        1    13
+        2    13
+        dtype: int64
+
+        Returning a list-like will result in a Series
+
+        >>> df.apply(lambda x: [1, 2], axis=1)
+        0    [1, 2]
+        1    [1, 2]
+        2    [1, 2]
+        dtype: object
+
+        Passing ``result_type='expand'`` will expand list-like results
+        to columns of a Dataframe
+
+        >>> df.apply(lambda x: [1, 2], axis=1, result_type="expand")
+           0  1
+        0  1  2
+        1  1  2
+        2  1  2
+
+        Returning a Series inside the function is similar to passing
+        ``result_type='expand'``. The resulting column names
+        will be the Series index.
+
+        >>> df.apply(lambda x: pd.Series([1, 2], index=["foo", "bar"]), axis=1)
+           foo  bar
+        0    1    2
+        1    1    2
+        2    1    2
+
+        Passing ``result_type='broadcast'`` will ensure the same shape
+        result, whether list-like or scalar is returned by the function,
+        and broadcast it along the axis. The resulting column names will
+        be the originals.
+
+        >>> df.apply(lambda x: [1, 2], axis=1, result_type="broadcast")
+           A  B
+        0  1  2
+        1  1  2
+        2  1  2
+        """
+        from pandas.core.apply import frame_apply
+
+        op = frame_apply(
+            self,
+            func=func,
+            axis=axis,
+            raw=raw,
+            result_type=result_type,
+            by_row=by_row,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+            args=args,
+            kwargs=kwargs,
+        )
+        return op.apply().__finalize__(self, method="apply")
+
+    def map(
+        self, func: PythonFuncType, na_action: Literal["ignore"] | None = None, **kwargs
+    ) -> DataFrame:
+        """
+        Apply a function to a Dataframe elementwise.
+
+        .. versionadded:: 2.1.0
+
+           DataFrame.applymap was deprecated and renamed to DataFrame.map.
+
+        This method applies a function that accepts and returns a scalar
+        to every element of a DataFrame.
+
+        Parameters
+        ----------
+        func : callable
+            Python function, returns a single value from a single value.
+        na_action : {None, 'ignore'}, default None
+            If 'ignore', propagate NaN values, without passing them to func.
+        **kwargs
+            Additional keyword arguments to pass as keywords arguments to
+            `func`.
+
+        Returns
+        -------
+        DataFrame
+            Transformed DataFrame.
+
+        See Also
+        --------
+        DataFrame.apply : Apply a function along input axis of DataFrame.
+        DataFrame.replace: Replace values given in `to_replace` with `value`.
+        Series.map : Apply a function elementwise on a Series.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2.12], [3.356, 4.567]])
+        >>> df
+               0      1
+        0  1.000  2.120
+        1  3.356  4.567
+
+        >>> df.map(lambda x: len(str(x)))
+           0  1
+        0  3  4
+        1  5  5
+
+        Like Series.map, NA values can be ignored:
+
+        >>> df_copy = df.copy()
+        >>> df_copy.iloc[0, 0] = pd.NA
+        >>> df_copy.map(lambda x: len(str(x)), na_action="ignore")
+             0  1
+        0  NaN  4
+        1  5.0  5
+
+        It is also possible to use `map` with functions that are not
+        `lambda` functions:
+
+        >>> df.map(round, ndigits=1)
+             0    1
+        0  1.0  2.1
+        1  3.4  4.6
+
+        Note that a vectorized version of `func` often exists, which will
+        be much faster. You could square each number elementwise.
+
+        >>> df.map(lambda x: x**2)
+                   0          1
+        0   1.000000   4.494400
+        1  11.262736  20.857489
+
+        But it's better to avoid map in that case.
+
+        >>> df**2
+                   0          1
+        0   1.000000   4.494400
+        1  11.262736  20.857489
+        """
+        if na_action not in {"ignore", None}:
+            raise ValueError(f"na_action must be 'ignore' or None. Got {na_action!r}")
+
+        if self.empty:
+            return self.copy()
+
+        func = functools.partial(func, **kwargs)
+
+        def infer(x):
+            return x._map_values(func, na_action=na_action)
+
+        return self.apply(infer).__finalize__(self, "map")
+
+    # ----------------------------------------------------------------------
+    # Merging / joining methods
+
+    def _append(
+        self,
+        other,
+        ignore_index: bool = False,
+        verify_integrity: bool = False,
+        sort: bool = False,
+    ) -> DataFrame:
+        if isinstance(other, (Series, dict)):
+            if isinstance(other, dict):
+                if not ignore_index:
+                    raise TypeError("Can only append a dict if ignore_index=True")
+                other = Series(other)
+            if other.name is None and not ignore_index:
+                raise TypeError(
+                    "Can only append a Series if ignore_index=True "
+                    "or if the Series has a name"
+                )
+
+            index = Index(
+                [other.name],
+                name=self.index.names
+                if isinstance(self.index, MultiIndex)
+                else self.index.name,
+            )
+            row_df = other.to_frame().T
+            # infer_objects is needed for
+            #  test_append_empty_frame_to_series_with_dateutil_tz
+            other = row_df.infer_objects().rename_axis(index.names)
+        elif isinstance(other, list):
+            if not other:
+                pass
+            elif not isinstance(other[0], DataFrame):
+                other = DataFrame(other)
+                if self.index.name is not None and not ignore_index:
+                    other.index.name = self.index.name
+
+        from pandas.core.reshape.concat import concat
+
+        if isinstance(other, (list, tuple)):
+            to_concat = [self, *other]
+        else:
+            to_concat = [self, other]
+
+        result = concat(
+            to_concat,
+            ignore_index=ignore_index,
+            verify_integrity=verify_integrity,
+            sort=sort,
+        )
+        return result.__finalize__(self, method="append")
+
+    def join(
+        self,
+        other: DataFrame | Series | Iterable[DataFrame | Series],
+        on: IndexLabel | None = None,
+        how: MergeHow = "left",
+        lsuffix: str = "",
+        rsuffix: str = "",
+        sort: bool = False,
+        validate: JoinValidate | None = None,
+    ) -> DataFrame:
+        """
+        Join columns of another DataFrame.
+
+        Join columns with `other` DataFrame either on index or on a key
+        column. Efficiently join multiple DataFrame objects by index at once by
+        passing a list.
+
+        Parameters
+        ----------
+        other : DataFrame, Series, or a list containing any combination of them
+            Index should be similar to one of the columns in this one. If a
+            Series is passed, its name attribute must be set, and that will be
+            used as the column name in the resulting joined DataFrame.
+        on : str, list of str, or array-like, optional
+            Column or index level name(s) in the caller to join on the index
+            in `other`, otherwise joins index-on-index. If multiple
+            values given, the `other` DataFrame must have a MultiIndex. Can
+            pass an array as the join key if it is not already contained in
+            the calling DataFrame. Like an Excel VLOOKUP operation.
+        how : {'left', 'right', 'outer', 'inner', 'cross', 'left_anti', 'right_anti'},
+            default 'left'
+            How to handle the operation of the two objects.
+
+            * left: use calling frame's index (or column if on is specified)
+            * right: use `other`'s index.
+            * outer: form union of calling frame's index (or column if on is
+              specified) with `other`'s index, and sort it lexicographically.
+            * inner: form intersection of calling frame's index (or column if
+              on is specified) with `other`'s index, preserving the order
+              of the calling's one.
+            * cross: creates the cartesian product from both frames, preserves the order
+              of the left keys.
+            * left_anti: use set difference of calling frame's index and `other`'s
+              index.
+            * right_anti: use set difference of `other`'s index and calling frame's
+              index.
+        lsuffix : str, default ''
+            Suffix to use from left frame's overlapping columns.
+        rsuffix : str, default ''
+            Suffix to use from right frame's overlapping columns.
+        sort : bool, default False
+            Order result DataFrame lexicographically by the join key. If False,
+            the order of the join key depends on the join type (how keyword).
+        validate : str, optional
+            If specified, checks if join is of specified type.
+
+            * "one_to_one" or "1:1": check if join keys are unique in both left
+              and right datasets.
+            * "one_to_many" or "1:m": check if join keys are unique in left dataset.
+            * "many_to_one" or "m:1": check if join keys are unique in right dataset.
+            * "many_to_many" or "m:m": allowed, but does not result in checks.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        DataFrame
+            A dataframe containing columns from both the caller and `other`.
+
+        See Also
+        --------
+        DataFrame.merge : For column(s)-on-column(s) operations.
+
+        Notes
+        -----
+        Parameters `on`, `lsuffix`, and `rsuffix` are not supported when
+        passing a list of `DataFrame` objects.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "key": ["K0", "K1", "K2", "K3", "K4", "K5"],
+        ...         "A": ["A0", "A1", "A2", "A3", "A4", "A5"],
+        ...     }
+        ... )
+
+        >>> df
+          key   A
+        0  K0  A0
+        1  K1  A1
+        2  K2  A2
+        3  K3  A3
+        4  K4  A4
+        5  K5  A5
+
+        >>> other = pd.DataFrame({"key": ["K0", "K1", "K2"], "B": ["B0", "B1", "B2"]})
+
+        >>> other
+          key   B
+        0  K0  B0
+        1  K1  B1
+        2  K2  B2
+
+        Join DataFrames using their indexes.
+
+        >>> df.join(other, lsuffix="_caller", rsuffix="_other")
+          key_caller   A key_other    B
+        0         K0  A0        K0   B0
+        1         K1  A1        K1   B1
+        2         K2  A2        K2   B2
+        3         K3  A3       NaN  NaN
+        4         K4  A4       NaN  NaN
+        5         K5  A5       NaN  NaN
+
+        If we want to join using the key columns, we need to set key to be
+        the index in both `df` and `other`. The joined DataFrame will have
+        key as its index.
+
+        >>> df.set_index("key").join(other.set_index("key"))
+              A    B
+        key
+        K0   A0   B0
+        K1   A1   B1
+        K2   A2   B2
+        K3   A3  NaN
+        K4   A4  NaN
+        K5   A5  NaN
+
+        Another option to join using the key columns is to use the `on`
+        parameter. DataFrame.join always uses `other`'s index but we can use
+        any column in `df`. This method preserves the original DataFrame's
+        index in the result.
+
+        >>> df.join(other.set_index("key"), on="key")
+          key   A    B
+        0  K0  A0   B0
+        1  K1  A1   B1
+        2  K2  A2   B2
+        3  K3  A3  NaN
+        4  K4  A4  NaN
+        5  K5  A5  NaN
+
+        Using non-unique key values shows how they are matched.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "key": ["K0", "K1", "K1", "K3", "K0", "K1"],
+        ...         "A": ["A0", "A1", "A2", "A3", "A4", "A5"],
+        ...     }
+        ... )
+
+        >>> df
+          key   A
+        0  K0  A0
+        1  K1  A1
+        2  K1  A2
+        3  K3  A3
+        4  K0  A4
+        5  K1  A5
+
+        >>> df.join(other.set_index("key"), on="key", validate="m:1")
+          key   A    B
+        0  K0  A0   B0
+        1  K1  A1   B1
+        2  K1  A2   B1
+        3  K3  A3  NaN
+        4  K0  A4   B0
+        5  K1  A5   B1
+        """
+        from pandas.core.reshape.concat import concat
+        from pandas.core.reshape.merge import merge
+
+        if isinstance(other, Series):
+            if other.name is None:
+                raise ValueError("Other Series must have a name")
+            other = DataFrame({other.name: other})
+
+        if isinstance(other, DataFrame):
+            if how == "cross":
+                return merge(
+                    self,
+                    other,
+                    how=how,
+                    on=on,
+                    suffixes=(lsuffix, rsuffix),
+                    sort=sort,
+                    validate=validate,
+                )
+            return merge(
+                self,
+                other,
+                left_on=on,
+                how=how,
+                left_index=on is None,
+                right_index=True,
+                suffixes=(lsuffix, rsuffix),
+                sort=sort,
+                validate=validate,
+            )
+        else:
+            if on is not None:
+                raise ValueError(
+                    "Joining multiple DataFrames only supported for joining on index"
+                )
+
+            if rsuffix or lsuffix:
+                raise ValueError(
+                    "Suffixes not supported when joining multiple DataFrames"
+                )
+
+            # Mypy thinks the RHS is a
+            # "Union[DataFrame, Series, Iterable[Union[DataFrame, Series]]]" whereas
+            # the LHS is an "Iterable[DataFrame]", but in reality both types are
+            # "Iterable[Union[DataFrame, Series]]" due to the if statements
+            frames = [cast("DataFrame | Series", self)] + list(other)
+
+            can_concat = all(df.index.is_unique for df in frames)
+
+            # join indexes only using concat
+            if can_concat:
+                if how == "left":
+                    res = concat(
+                        frames, axis=1, join="outer", verify_integrity=True, sort=sort
+                    )
+                    return res.reindex(self.index)
+                else:
+                    return concat(
+                        frames, axis=1, join=how, verify_integrity=True, sort=sort
+                    )
+
+            joined = frames[0]
+
+            for frame in frames[1:]:
+                joined = merge(
+                    joined,
+                    frame,
+                    how=how,
+                    left_index=True,
+                    right_index=True,
+                    validate=validate,
+                )
+
+            return joined
+
+    @Substitution("")
+    @Appender(_merge_doc, indents=2)
+    def merge(
+        self,
+        right: DataFrame | Series,
+        how: MergeHow = "inner",
+        on: IndexLabel | AnyArrayLike | None = None,
+        left_on: IndexLabel | AnyArrayLike | None = None,
+        right_on: IndexLabel | AnyArrayLike | None = None,
+        left_index: bool = False,
+        right_index: bool = False,
+        sort: bool = False,
+        suffixes: Suffixes = ("_x", "_y"),
+        copy: bool | lib.NoDefault = lib.no_default,
+        indicator: str | bool = False,
+        validate: MergeValidate | None = None,
+    ) -> DataFrame:
+        self._check_copy_deprecation(copy)
+
+        from pandas.core.reshape.merge import merge
+
+        return merge(
+            self,
+            right,
+            how=how,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            left_index=left_index,
+            right_index=right_index,
+            sort=sort,
+            suffixes=suffixes,
+            indicator=indicator,
+            validate=validate,
+        )
+
+    def round(
+        self, decimals: int | dict[IndexLabel, int] | Series = 0, *args, **kwargs
+    ) -> DataFrame:
+        """
+        Round numeric columns in a DataFrame to a variable number of decimal places.
+
+        Parameters
+        ----------
+        decimals : int, dict, Series
+            Number of decimal places to round each column to. If an int is
+            given, round each column to the same number of places.
+            Otherwise dict and Series round to variable numbers of places.
+            Column names should be in the keys if `decimals` is a
+            dict-like, or in the index if `decimals` is a Series. Any
+            columns not included in `decimals` will be left as is. Elements
+            of `decimals` which are not columns of the input will be
+            ignored.
+        *args
+            Additional keywords have no effect but might be accepted for
+            compatibility with numpy.
+        **kwargs
+            Additional keywords have no effect but might be accepted for
+            compatibility with numpy.
+
+        Returns
+        -------
+        DataFrame
+            A DataFrame with the affected columns rounded to the specified
+            number of decimal places.
+
+        See Also
+        --------
+        numpy.around : Round a numpy array to the given number of decimals.
+        Series.round : Round a Series to the given number of decimals.
+
+        Notes
+        -----
+        For values exactly halfway between rounded decimal values, pandas rounds
+        to the nearest even value (e.g. -0.5 and 0.5 round to 0.0, 1.5 and 2.5
+        round to 2.0, etc.).
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [(0.21, 0.32), (0.01, 0.67), (0.66, 0.03), (0.21, 0.18)],
+        ...     columns=["dogs", "cats"],
+        ... )
+        >>> df
+            dogs  cats
+        0  0.21  0.32
+        1  0.01  0.67
+        2  0.66  0.03
+        3  0.21  0.18
+
+        By providing an integer each column is rounded to the same number
+        of decimal places
+
+        >>> df.round(1)
+            dogs  cats
+        0   0.2   0.3
+        1   0.0   0.7
+        2   0.7   0.0
+        3   0.2   0.2
+
+        With a dict, the number of places for specific columns can be
+        specified with the column names as key and the number of decimal
+        places as value
+
+        >>> df.round({"dogs": 1, "cats": 0})
+            dogs  cats
+        0   0.2   0.0
+        1   0.0   1.0
+        2   0.7   0.0
+        3   0.2   0.0
+
+        Using a Series, the number of places for specific columns can be
+        specified with the column names as index and the number of
+        decimal places as value
+
+        >>> decimals = pd.Series([0, 1], index=["cats", "dogs"])
+        >>> df.round(decimals)
+            dogs  cats
+        0   0.2   0.0
+        1   0.0   1.0
+        2   0.7   0.0
+        3   0.2   0.0
+        """
+        from pandas.core.reshape.concat import concat
+
+        def _dict_round(df: DataFrame, decimals) -> Iterator[Series]:
+            for col, vals in df.items():
+                try:
+                    yield _series_round(vals, decimals[col])
+                except KeyError:
+                    yield vals
+
+        def _series_round(ser: Series, decimals: int) -> Series:
+            if is_integer_dtype(ser.dtype) or is_float_dtype(ser.dtype):
+                return ser.round(decimals)
+            return ser
+
+        nv.validate_round(args, kwargs)
+
+        if isinstance(decimals, (dict, Series)):
+            if isinstance(decimals, Series) and not decimals.index.is_unique:
+                raise ValueError("Index of decimals must be unique")
+            if is_dict_like(decimals) and not all(
+                is_integer(value) for _, value in decimals.items()
+            ):
+                raise TypeError("Values in decimals must be integers")
+            new_cols = list(_dict_round(self, decimals))
+        elif is_integer(decimals):
+            # Dispatch to Block.round
+            # Argument "decimals" to "round" of "BaseBlockManager" has incompatible
+            # type "Union[int, integer[Any]]"; expected "int"
+            new_mgr = self._mgr.round(
+                decimals=decimals,  # type: ignore[arg-type]
+            )
+            return self._constructor_from_mgr(new_mgr, axes=new_mgr.axes).__finalize__(
+                self, method="round"
+            )
+        else:
+            raise TypeError("decimals must be an integer, a dict-like or a Series")
+
+        if new_cols is not None and len(new_cols) > 0:
+            return self._constructor(
+                concat(new_cols, axis=1), index=self.index, columns=self.columns
+            ).__finalize__(self, method="round")
+        else:
+            return self.copy(deep=False)
+
+    # ----------------------------------------------------------------------
+    # Statistical methods, etc.
+
+    def corr(
+        self,
+        method: CorrelationMethod = "pearson",
+        min_periods: int = 1,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        """
+        Compute pairwise correlation of columns, excluding NA/null values.
+
+        Parameters
+        ----------
+        method : {'pearson', 'kendall', 'spearman'} or callable
+            Method of correlation:
+
+            * pearson : standard correlation coefficient
+            * kendall : Kendall Tau correlation coefficient
+            * spearman : Spearman rank correlation
+            * callable: callable with input two 1d ndarrays
+                and returning a float. Note that the returned matrix from corr
+                will have 1 along the diagonals and will be symmetric
+                regardless of the callable's behavior.
+        min_periods : int, optional
+            Minimum number of observations required per pair of columns
+            to have a valid result. Currently only available for Pearson
+            and Spearman correlation.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        Returns
+        -------
+        DataFrame
+            Correlation matrix.
+
+        See Also
+        --------
+        DataFrame.corrwith : Compute pairwise correlation with another
+            DataFrame or Series.
+        Series.corr : Compute the correlation between two Series.
+
+        Notes
+        -----
+        Pearson, Kendall and Spearman correlation are currently computed using pairwise complete observations.
+
+        * `Pearson correlation coefficient <https://en.wikipedia.org/wiki/Pearson_correlation_coefficient>`_
+        * `Kendall rank correlation coefficient <https://en.wikipedia.org/wiki/Kendall_rank_correlation_coefficient>`_
+        * `Spearman's rank correlation coefficient <https://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient>`_
+
+        Examples
+        --------
+        >>> def histogram_intersection(a, b):
+        ...     v = np.minimum(a, b).sum().round(decimals=1)
+        ...     return v
+        >>> df = pd.DataFrame(
+        ...     [(0.2, 0.3), (0.0, 0.6), (0.6, 0.0), (0.2, 0.1)],
+        ...     columns=["dogs", "cats"],
+        ... )
+        >>> df.corr(method=histogram_intersection)
+              dogs  cats
+        dogs   1.0   0.3
+        cats   0.3   1.0
+
+        >>> df = pd.DataFrame(
+        ...     [(1, 1), (2, np.nan), (np.nan, 3), (4, 4)], columns=["dogs", "cats"]
+        ... )
+        >>> df.corr(min_periods=3)
+              dogs  cats
+        dogs   1.0   NaN
+        cats   NaN   1.0
+        """  # noqa: E501
+        data = self._get_numeric_data() if numeric_only else self
+        cols = data.columns
+        idx = cols.copy()
+        mat = data.to_numpy(dtype=float, na_value=np.nan, copy=False)
+
+        if method == "pearson":
+            correl = libalgos.nancorr(mat, minp=min_periods)
+        elif method == "spearman":
+            correl = libalgos.nancorr_spearman(mat, minp=min_periods)
+        elif method == "kendall" or callable(method):
+            if min_periods is None:
+                min_periods = 1
+            mat = mat.T
+            corrf = nanops.get_corr_func(method)
+            K = len(cols)
+            correl = np.empty((K, K), dtype=float)
+            mask = np.isfinite(mat)
+            for i, ac in enumerate(mat):
+                for j, bc in enumerate(mat):
+                    if i > j:
+                        continue
+
+                    valid = mask[i] & mask[j]
+                    if valid.sum() < min_periods:
+                        c = np.nan
+                    elif i == j:
+                        c = 1.0
+                    elif not valid.all():
+                        c = corrf(ac[valid], bc[valid])
+                    else:
+                        c = corrf(ac, bc)
+                    correl[i, j] = c
+                    correl[j, i] = c
+        else:
+            raise ValueError(
+                "method must be either 'pearson', "
+                "'spearman', 'kendall', or a callable, "
+                f"'{method}' was supplied"
+            )
+
+        result = self._constructor(correl, index=idx, columns=cols, copy=False)
+        return result.__finalize__(self, method="corr")
+
+    def cov(
+        self,
+        min_periods: int | None = None,
+        ddof: int | None = 1,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        """
+        Compute pairwise covariance of columns, excluding NA/null values.
+
+        Compute the pairwise covariance among the series of a DataFrame.
+        The returned data frame is the `covariance matrix
+        <https://en.wikipedia.org/wiki/Covariance_matrix>`__ of the columns
+        of the DataFrame.
+
+        Both NA and null values are automatically excluded from the
+        calculation. (See the note below about bias from missing values.)
+        A threshold can be set for the minimum number of
+        observations for each value created. Comparisons with observations
+        below this threshold will be returned as ``NaN``.
+
+        This method is generally used for the analysis of time series data to
+        understand the relationship between different measures
+        across time.
+
+        Parameters
+        ----------
+        min_periods : int, optional
+            Minimum number of observations required per pair of columns
+            to have a valid result.
+
+        ddof : int, default 1
+            Delta degrees of freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+            This argument is applicable only when no ``nan`` is in the dataframe.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        Returns
+        -------
+        DataFrame
+            The covariance matrix of the series of the DataFrame.
+
+        See Also
+        --------
+        Series.cov : Compute covariance with another Series.
+        core.window.ewm.ExponentialMovingWindow.cov : Exponential weighted sample
+            covariance.
+        core.window.expanding.Expanding.cov : Expanding sample covariance.
+        core.window.rolling.Rolling.cov : Rolling sample covariance.
+
+        Notes
+        -----
+        Returns the covariance matrix of the DataFrame's time series.
+        The covariance is normalized by N-ddof.
+
+        For DataFrames that have Series that are missing data (assuming that
+        data is `missing at random
+        <https://en.wikipedia.org/wiki/Missing_data#Missing_at_random>`__)
+        the returned covariance matrix will be an unbiased estimate
+        of the variance and covariance between the member Series.
+
+        However, for many applications this estimate may not be acceptable
+        because the estimate covariance matrix is not guaranteed to be positive
+        semi-definite. This could lead to estimate correlations having
+        absolute values which are greater than one, and/or a non-invertible
+        covariance matrix. See `Estimation of covariance matrices
+        <https://en.wikipedia.org/w/index.php?title=Estimation_of_covariance_
+        matrices>`__ for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [(1, 2), (0, 3), (2, 0), (1, 1)], columns=["dogs", "cats"]
+        ... )
+        >>> df.cov()
+                  dogs      cats
+        dogs  0.666667 -1.000000
+        cats -1.000000  1.666667
+
+        >>> np.random.seed(42)
+        >>> df = pd.DataFrame(
+        ...     np.random.randn(1000, 5), columns=["a", "b", "c", "d", "e"]
+        ... )
+        >>> df.cov()
+                  a         b         c         d         e
+        a  0.998438 -0.020161  0.059277 -0.008943  0.014144
+        b -0.020161  1.059352 -0.008543 -0.024738  0.009826
+        c  0.059277 -0.008543  1.010670 -0.001486 -0.000271
+        d -0.008943 -0.024738 -0.001486  0.921297 -0.013692
+        e  0.014144  0.009826 -0.000271 -0.013692  0.977795
+
+        **Minimum number of periods**
+
+        This method also supports an optional ``min_periods`` keyword
+        that specifies the required minimum number of non-NA observations for
+        each column pair in order to have a valid result:
+
+        >>> np.random.seed(42)
+        >>> df = pd.DataFrame(np.random.randn(20, 3), columns=["a", "b", "c"])
+        >>> df.loc[df.index[:5], "a"] = np.nan
+        >>> df.loc[df.index[5:10], "b"] = np.nan
+        >>> df.cov(min_periods=12)
+                  a         b         c
+        a  0.316741       NaN -0.150812
+        b       NaN  1.248003  0.191417
+        c -0.150812  0.191417  0.895202
+        """
+        data = self._get_numeric_data() if numeric_only else self
+        cols = data.columns
+        idx = cols.copy()
+        mat = data.to_numpy(dtype=float, na_value=np.nan, copy=False)
+
+        if notna(mat).all():
+            if min_periods is not None and min_periods > len(mat):
+                base_cov = np.empty((mat.shape[1], mat.shape[1]))
+                base_cov.fill(np.nan)
+            else:
+                base_cov = np.cov(mat.T, ddof=ddof)
+            base_cov = base_cov.reshape((len(cols), len(cols)))
+        else:
+            base_cov = libalgos.nancorr(mat, cov=True, minp=min_periods)
+
+        result = self._constructor(base_cov, index=idx, columns=cols, copy=False)
+        return result.__finalize__(self, method="cov")
+
+    def corrwith(
+        self,
+        other: DataFrame | Series,
+        axis: Axis = 0,
+        drop: bool = False,
+        method: CorrelationMethod = "pearson",
+        numeric_only: bool = False,
+        min_periods: int | None = None,
+    ) -> Series:
+        """
+        Compute pairwise correlation.
+
+        Pairwise correlation is computed between rows or columns of
+        DataFrame with rows or columns of Series or DataFrame. DataFrames
+        are first aligned along both axes before computing the
+        correlations.
+
+        Parameters
+        ----------
+        other : DataFrame, Series
+            Object with which to compute correlations.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to use. 0 or 'index' to compute row-wise, 1 or 'columns' for
+            column-wise.
+        drop : bool, default False
+            Drop missing indices from result.
+        method : {'pearson', 'kendall', 'spearman'} or callable
+            Method of correlation:
+
+            * pearson : standard correlation coefficient
+            * kendall : Kendall Tau correlation coefficient
+            * spearman : Spearman rank correlation
+            * callable: callable with input two 1d ndarrays
+                and returning a float.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+        min_periods : int, optional
+            Minimum number of observations needed to have a valid result.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        Returns
+        -------
+        Series
+            Pairwise correlations.
+
+        See Also
+        --------
+        DataFrame.corr : Compute pairwise correlation of columns.
+
+        Examples
+        --------
+        >>> index = ["a", "b", "c", "d", "e"]
+        >>> columns = ["one", "two", "three", "four"]
+        >>> df1 = pd.DataFrame(
+        ...     np.arange(20).reshape(5, 4), index=index, columns=columns
+        ... )
+        >>> df2 = pd.DataFrame(
+        ...     np.arange(16).reshape(4, 4), index=index[:4], columns=columns
+        ... )
+        >>> df1.corrwith(df2)
+        one      1.0
+        two      1.0
+        three    1.0
+        four     1.0
+        dtype: float64
+
+        >>> df2.corrwith(df1, axis=1)
+        a    1.0
+        b    1.0
+        c    1.0
+        d    1.0
+        e    NaN
+        dtype: float64
+        """
+        axis = self._get_axis_number(axis)
+        this = self._get_numeric_data() if numeric_only else self
+
+        if isinstance(other, Series):
+            return this.apply(
+                lambda x: other.corr(x, method=method, min_periods=min_periods),
+                axis=axis,
+            )
+
+        if numeric_only:
+            other = other._get_numeric_data()
+        left, right = this.align(other, join="inner")
+
+        if axis == 1:
+            left = left.T
+            right = right.T
+
+        if method == "pearson":
+            # mask missing values
+            left = left + right * 0
+            right = right + left * 0
+
+            # demeaned data
+            ldem = left - left.mean(numeric_only=numeric_only)
+            rdem = right - right.mean(numeric_only=numeric_only)
+
+            num = (ldem * rdem).sum()
+            dom = (
+                (left.count() - 1)
+                * left.std(numeric_only=numeric_only)
+                * right.std(numeric_only=numeric_only)
+            )
+
+            correl = num / dom
+
+        elif method in ["kendall", "spearman"] or callable(method):
+
+            def c(x):
+                return nanops.nancorr(x[0], x[1], method=method)
+
+            correl = self._constructor_sliced(
+                map(c, zip(left.values.T, right.values.T)),
+                index=left.columns,
+                copy=False,
+            )
+
+        else:
+            raise ValueError(
+                f"Invalid method {method} was passed, "
+                "valid methods are: 'pearson', 'kendall', "
+                "'spearman', or callable"
+            )
+
+        if not drop:
+            # Find non-matching labels along the given axis
+            # and append missing correlations (GH 22375)
+            raxis: AxisInt = 1 if axis == 0 else 0
+            result_index = this._get_axis(raxis).union(other._get_axis(raxis))
+            idx_diff = result_index.difference(correl.index)
+
+            if len(idx_diff) > 0:
+                correl = correl._append(
+                    Series([np.nan] * len(idx_diff), index=idx_diff)
+                )
+
+        return correl
+
+    # ----------------------------------------------------------------------
+    # ndarray-like stats methods
+
+    def count(self, axis: Axis = 0, numeric_only: bool = False) -> Series:
+        """
+        Count non-NA cells for each column or row.
+
+        The values `None`, `NaN`, `NaT`, ``pandas.NA`` are considered NA.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            If 0 or 'index' counts are generated for each column.
+            If 1 or 'columns' counts are generated for each row.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+        Returns
+        -------
+        Series
+            For each column/row the number of non-NA/null entries.
+
+        See Also
+        --------
+        Series.count: Number of non-NA elements in a Series.
+        DataFrame.value_counts: Count unique combinations of columns.
+        DataFrame.shape: Number of DataFrame rows and columns (including NA
+            elements).
+        DataFrame.isna: Boolean same-sized DataFrame showing places of NA
+            elements.
+
+        Examples
+        --------
+        Constructing DataFrame from a dictionary:
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "Person": ["John", "Myla", "Lewis", "John", "Myla"],
+        ...         "Age": [24.0, np.nan, 21.0, 33, 26],
+        ...         "Single": [False, True, True, True, False],
+        ...     }
+        ... )
+        >>> df
+           Person   Age  Single
+        0    John  24.0   False
+        1    Myla   NaN    True
+        2   Lewis  21.0    True
+        3    John  33.0    True
+        4    Myla  26.0   False
+
+        Notice the uncounted NA values:
+
+        >>> df.count()
+        Person    5
+        Age       4
+        Single    5
+        dtype: int64
+
+        Counts for each **row**:
+
+        >>> df.count(axis="columns")
+        0    3
+        1    2
+        2    3
+        3    3
+        4    3
+        dtype: int64
+        """
+        axis = self._get_axis_number(axis)
+
+        if numeric_only:
+            frame = self._get_numeric_data()
+        else:
+            frame = self
+
+        # GH #423
+        if len(frame._get_axis(axis)) == 0:
+            result = self._constructor_sliced(0, index=frame._get_agg_axis(axis))
+        else:
+            result = notna(frame).sum(axis=axis)
+
+        return result.astype("int64").__finalize__(self, method="count")
+
+    def _reduce(
+        self,
+        op,
+        name: str,
+        *,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        filter_type=None,
+        **kwds,
+    ):
+        assert filter_type is None or filter_type == "bool", filter_type
+        out_dtype = "bool" if filter_type == "bool" else None
+
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+
+        def func(values: np.ndarray):
+            # We only use this in the case that operates on self.values
+            return op(values, axis=axis, skipna=skipna, **kwds)
+
+        def blk_func(values, axis: Axis = 1):
+            if isinstance(values, ExtensionArray):
+                if not is_1d_only_ea_dtype(values.dtype):
+                    return values._reduce(name, axis=1, skipna=skipna, **kwds)
+                return values._reduce(name, skipna=skipna, keepdims=True, **kwds)
+            else:
+                return op(values, axis=axis, skipna=skipna, **kwds)
+
+        def _get_data() -> DataFrame:
+            if filter_type is None:
+                data = self._get_numeric_data()
+            else:
+                # GH#25101, GH#24434
+                assert filter_type == "bool"
+                data = self._get_bool_data()
+            return data
+
+        # Case with EAs see GH#35881
+        df = self
+        if numeric_only:
+            df = _get_data()
+        if axis is None:
+            dtype = find_common_type([block.values.dtype for block in df._mgr.blocks])
+            if isinstance(dtype, ExtensionDtype):
+                df = df.astype(dtype)
+                arr = concat_compat(list(df._iter_column_arrays()))
+                return arr._reduce(name, skipna=skipna, keepdims=False, **kwds)
+            return func(df.values)
+        elif axis == 1:
+            if len(df.index) == 0:
+                # Taking a transpose would result in no columns, losing the dtype.
+                # In the empty case, reducing along axis 0 or 1 gives the same
+                # result dtype, so reduce with axis=0 and ignore values
+                result = df._reduce(
+                    op,
+                    name,
+                    axis=0,
+                    skipna=skipna,
+                    numeric_only=False,
+                    filter_type=filter_type,
+                    **kwds,
+                ).iloc[:0]
+                result.index = df.index
+                return result
+
+            # kurtosis excluded since groupby does not implement it
+            if df.shape[1] and name != "kurt":
+                dtype = find_common_type(
+                    [block.values.dtype for block in df._mgr.blocks]
+                )
+                if isinstance(dtype, ExtensionDtype):
+                    # GH 54341: fastpath for EA-backed axis=1 reductions
+                    # This flattens the frame into a single 1D array while keeping
+                    # track of the row and column indices of the original frame. Once
+                    # flattened, grouping by the row indices and aggregating should
+                    # be equivalent to transposing the original frame and aggregating
+                    # with axis=0.
+                    name = {"argmax": "idxmax", "argmin": "idxmin"}.get(name, name)
+                    df = df.astype(dtype)
+                    arr = concat_compat(list(df._iter_column_arrays()))
+                    nrows, ncols = df.shape
+                    row_index = np.tile(np.arange(nrows), ncols)
+                    col_index = np.repeat(np.arange(ncols), nrows)
+                    ser = Series(arr, index=col_index, copy=False)
+                    # GroupBy will raise a warning with SeriesGroupBy as the object,
+                    # likely confusing users
+                    with rewrite_warning(
+                        target_message=(
+                            f"The behavior of SeriesGroupBy.{name} with all-NA values"
+                        ),
+                        target_category=FutureWarning,
+                        new_message=(
+                            f"The behavior of {type(self).__name__}.{name} with all-NA "
+                            "values, or any-NA and skipna=False, is deprecated. In "
+                            "a future version this will raise ValueError"
+                        ),
+                    ):
+                        result = ser.groupby(row_index).agg(name, **kwds)
+                    result.index = df.index
+                    if not skipna and name not in ("any", "all"):
+                        mask = df.isna().to_numpy(dtype=np.bool_).any(axis=1)
+                        other = -1 if name in ("idxmax", "idxmin") else lib.no_default
+                        result = result.mask(mask, other)
+                    return result
+
+            df = df.T
+
+        # After possibly _get_data and transposing, we are now in the
+        #  simple case where we can use BlockManager.reduce
+        res = df._mgr.reduce(blk_func)
+        out = df._constructor_from_mgr(res, axes=res.axes).iloc[0]
+        if out_dtype is not None and out.dtype != "boolean":
+            out = out.astype(out_dtype)
+        elif (df._mgr.get_dtypes() == object).any() and name not in ["any", "all"]:
+            out = out.astype(object)
+        elif len(self) == 0 and out.dtype == object and name in ("sum", "prod"):
+            # Even if we are object dtype, follow numpy and return
+            #  float64, see test_apply_funcs_over_empty
+            out = out.astype(np.float64)
+
+        return out
+
+    def _reduce_axis1(self, name: str, func, skipna: bool) -> Series:
+        """
+        Special case for _reduce to try to avoid a potentially-expensive transpose.
+
+        Apply the reduction block-wise along axis=1 and then reduce the resulting
+        1D arrays.
+        """
+        if name == "all":
+            result = np.ones(len(self), dtype=bool)
+            ufunc = np.logical_and
+        elif name == "any":
+            result = np.zeros(len(self), dtype=bool)
+            # error: Incompatible types in assignment
+            # (expression has type "_UFunc_Nin2_Nout1[Literal['logical_or'],
+            # Literal[20], Literal[False]]", variable has type
+            # "_UFunc_Nin2_Nout1[Literal['logical_and'], Literal[20],
+            # Literal[True]]")
+            ufunc = np.logical_or  # type: ignore[assignment]
+        else:
+            raise NotImplementedError(name)
+
+        for blocks in self._mgr.blocks:
+            middle = func(blocks.values, axis=0, skipna=skipna)
+            result = ufunc(result, middle)
+
+        res_ser = self._constructor_sliced(result, index=self.index, copy=False)
+        return res_ser
+
+    # error: Signature of "any" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def any(
+        self,
+        *,
+        axis: Axis = ...,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def any(
+        self,
+        *,
+        axis: None,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> bool: ...
+
+    @overload
+    def any(
+        self,
+        *,
+        axis: Axis | None,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> Series | bool: ...
+
+    @doc(make_doc("any", ndim=1))
+    def any(
+        self,
+        *,
+        axis: Axis | None = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> Series | bool:
+        result = self._logical_func(
+            "any", nanops.nanany, axis, bool_only, skipna, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="any")
+        return result
+
+    @overload
+    def all(
+        self,
+        *,
+        axis: Axis = ...,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def all(
+        self,
+        *,
+        axis: None,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> bool: ...
+
+    @overload
+    def all(
+        self,
+        *,
+        axis: Axis | None,
+        bool_only: bool = ...,
+        skipna: bool = ...,
+        **kwargs,
+    ) -> Series | bool: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="all")
+    @doc(make_doc("all", ndim=1))
+    def all(
+        self,
+        axis: Axis | None = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> Series | bool:
+        result = self._logical_func(
+            "all", nanops.nanall, axis, bool_only, skipna, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="all")
+        return result
+
+    # error: Signature of "min" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def min(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def min(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def min(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="min")
+    @doc(make_doc("min", ndim=2))
+    def min(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        result = super().min(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="min")
+        return result
+
+    # error: Signature of "max" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def max(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def max(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def max(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="max")
+    @doc(make_doc("max", ndim=2))
+    def max(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        result = super().max(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="max")
+        return result
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="sum")
+    def sum(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ) -> Series:
+        """
+        Return the sum of the values over the requested axis.
+
+        This is equivalent to the method ``numpy.sum``.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.sum with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer than
+            ``min_count`` non-NA values are present the result will be NA.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series or scalar
+            Sum over requested axis.
+
+        See Also
+        --------
+        Series.sum : Return the sum over Series values.
+        DataFrame.mean : Return the mean of the values over the requested axis.
+        DataFrame.median : Return the median of the values over the requested axis.
+        DataFrame.mode : Get the mode(s) of each element along the requested axis.
+        DataFrame.std : Return the standard deviation of the values over the
+            requested axis.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_arrays(
+        ...     [["warm", "warm", "cold", "cold"], ["dog", "falcon", "fish", "spider"]],
+        ...     names=["blooded", "animal"],
+        ... )
+        >>> s = pd.Series([4, 2, 0, 8], name="legs", index=idx)
+        >>> s
+        blooded  animal
+        warm     dog       4
+                 falcon    2
+        cold     fish      0
+                 spider    8
+        Name: legs, dtype: int64
+
+        >>> s.sum()
+        14
+
+        By default, the sum of an empty or all-NA Series is ``0``.
+
+        >>> pd.Series([], dtype="float64").sum()  # min_count=0 is the default
+        0.0
+
+        This can be controlled with the ``min_count`` parameter. For example, if
+        you'd like the sum of an empty series to be NaN, pass ``min_count=1``.
+
+        >>> pd.Series([], dtype="float64").sum(min_count=1)
+        nan
+
+        Thanks to the ``skipna`` parameter, ``min_count`` handles all-NA and
+        empty series identically.
+
+        >>> pd.Series([np.nan]).sum()
+        0.0
+
+        >>> pd.Series([np.nan]).sum(min_count=1)
+        nan
+        """
+        result = super().sum(
+            axis=axis,
+            skipna=skipna,
+            numeric_only=numeric_only,
+            min_count=min_count,
+            **kwargs,
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="sum")
+        return result
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="prod")
+    def prod(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ) -> Series:
+        """
+        Return the product of the values over the requested axis.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.prod with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer than
+            ``min_count`` non-NA values are present the result will be NA.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series or scalar
+            The product of the values over the requested axis.
+
+        See Also
+        --------
+        Series.sum : Return the sum.
+        Series.min : Return the minimum.
+        Series.max : Return the maximum.
+        Series.idxmin : Return the index of the minimum.
+        Series.idxmax : Return the index of the maximum.
+        DataFrame.sum : Return the sum over the requested axis.
+        DataFrame.min : Return the minimum over the requested axis.
+        DataFrame.max : Return the maximum over the requested axis.
+        DataFrame.idxmin : Return the index of the minimum over the requested axis.
+        DataFrame.idxmax : Return the index of the maximum over the requested axis.
+
+        Examples
+        --------
+        By default, the product of an empty or all-NA Series is ``1``
+
+        >>> pd.Series([], dtype="float64").prod()
+        1.0
+
+        This can be controlled with the ``min_count`` parameter
+
+        >>> pd.Series([], dtype="float64").prod(min_count=1)
+        nan
+
+        Thanks to the ``skipna`` parameter, ``min_count`` handles all-NA and
+        empty series identically.
+
+        >>> pd.Series([np.nan]).prod()
+        1.0
+
+        >>> pd.Series([np.nan]).prod(min_count=1)
+        nan
+        """
+        result = super().prod(
+            axis=axis,
+            skipna=skipna,
+            numeric_only=numeric_only,
+            min_count=min_count,
+            **kwargs,
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="prod")
+        return result
+
+    # error: Signature of "mean" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def mean(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def mean(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def mean(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="mean")
+    @doc(make_doc("mean", ndim=2))
+    def mean(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        result = super().mean(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="mean")
+        return result
+
+    # error: Signature of "median" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def median(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def median(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def median(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="median")
+    @doc(make_doc("median", ndim=2))
+    def median(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        result = super().median(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="median")
+        return result
+
+    # error: Signature of "sem" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def sem(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def sem(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def sem(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="sem")
+    def sem(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        """
+        Return unbiased standard error of the mean over requested axis.
+
+        Normalized by N-1 by default. This can be changed using the ddof argument
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.sem with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire row/column is NA, the result
+            will be NA.
+        ddof : int, default 1
+            Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+            where N represents the number of elements.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+        **kwargs :
+            Additional keywords passed.
+
+        Returns
+        -------
+        Series or DataFrame (if level specified)
+            Unbiased standard error of the mean over requested axis.
+
+        See Also
+        --------
+        DataFrame.var : Return unbiased variance over requested axis.
+        DataFrame.std : Returns sample standard deviation over requested axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.sem().round(6)
+        0.57735
+
+        With a DataFrame
+
+        >>> df = pd.DataFrame({"a": [1, 2], "b": [2, 3]}, index=["tiger", "zebra"])
+        >>> df
+               a   b
+        tiger  1   2
+        zebra  2   3
+        >>> df.sem()
+        a   0.5
+        b   0.5
+        dtype: float64
+
+        Using axis=1
+
+        >>> df.sem(axis=1)
+        tiger   0.5
+        zebra   0.5
+        dtype: float64
+
+        In this case, `numeric_only` should be set to `True`
+        to avoid getting an error.
+
+        >>> df = pd.DataFrame({"a": [1, 2], "b": ["T", "Z"]}, index=["tiger", "zebra"])
+        >>> df.sem(numeric_only=True)
+        a   0.5
+        dtype: float64
+        """
+        result = super().sem(
+            axis=axis, skipna=skipna, ddof=ddof, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="sem")
+        return result
+
+    # error: Signature of "var" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def var(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def var(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def var(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="var")
+    def var(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        """
+        Return unbiased variance over requested axis.
+
+        Normalized by N-1 by default. This can be changed using the ddof argument.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.var with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire row/column is NA, the result
+            will be NA.
+        ddof : int, default 1
+            Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+            where N represents the number of elements.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+        **kwargs :
+            Additional keywords passed.
+
+        Returns
+        -------
+        Series or scalaer
+            Unbiased variance over requested axis.
+
+        See Also
+        --------
+        numpy.var : Equivalent function in NumPy.
+        Series.var : Return unbiased variance over Series values.
+        Series.std : Return standard deviation over Series values.
+        DataFrame.std : Return standard deviation of the values over
+            the requested axis.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "person_id": [0, 1, 2, 3],
+        ...         "age": [21, 25, 62, 43],
+        ...         "height": [1.61, 1.87, 1.49, 2.01],
+        ...     }
+        ... ).set_index("person_id")
+        >>> df
+                   age  height
+        person_id
+        0           21    1.61
+        1           25    1.87
+        2           62    1.49
+        3           43    2.01
+
+        >>> df.var()
+        age       352.916667
+        height      0.056367
+        dtype: float64
+
+        Alternatively, ``ddof=0`` can be set to normalize by N instead of N-1:
+
+        >>> df.var(ddof=0)
+        age       264.687500
+        height      0.042275
+        dtype: float64
+        """
+        result = super().var(
+            axis=axis, skipna=skipna, ddof=ddof, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="var")
+        return result
+
+    # error: Signature of "std" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def std(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def std(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def std(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        ddof: int = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="std")
+    def std(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        """
+        Return sample standard deviation over requested axis.
+
+        Normalized by N-1 by default. This can be changed using the ddof argument.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.std with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire row/column is NA, the result
+            will be NA.
+        ddof : int, default 1
+            Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+            where N represents the number of elements.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+        **kwargs : dict
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series or scalar
+            Standard deviation over requested axis.
+
+        See Also
+        --------
+        Series.std : Return standard deviation over Series values.
+        DataFrame.mean : Return the mean of the values over the requested axis.
+        DataFrame.median : Return the median of the values over the requested axis.
+        DataFrame.mode : Get the mode(s) of each element along the requested axis.
+        DataFrame.sum : Return the sum of the values over the requested axis.
+
+        Notes
+        -----
+        To have the same behaviour as `numpy.std`, use `ddof=0` (instead of the
+        default `ddof=1`)
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "person_id": [0, 1, 2, 3],
+        ...         "age": [21, 25, 62, 43],
+        ...         "height": [1.61, 1.87, 1.49, 2.01],
+        ...     }
+        ... ).set_index("person_id")
+        >>> df
+                   age  height
+        person_id
+        0           21    1.61
+        1           25    1.87
+        2           62    1.49
+        3           43    2.01
+
+        The standard deviation of the columns can be found as follows:
+
+        >>> df.std()
+        age       18.786076
+        height     0.237417
+        dtype: float64
+
+        Alternatively, `ddof=0` can be set to normalize by N instead of N-1:
+
+        >>> df.std(ddof=0)
+        age       16.269219
+        height     0.205609
+        dtype: float64
+        """
+        result = super().std(
+            axis=axis, skipna=skipna, ddof=ddof, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="std")
+        return result
+
+    # error: Signature of "skew" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def skew(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def skew(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def skew(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="skew")
+    def skew(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        """
+        Return unbiased skew over requested axis.
+
+        Normalized by N-1.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series or scalar
+            Unbiased skew over requested axis.
+
+        See Also
+        --------
+        Dataframe.kurt : Returns unbiased kurtosis over requested axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.skew()
+        0.0
+
+        With a DataFrame
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [1, 2, 3], "b": [2, 3, 4], "c": [1, 3, 5]},
+        ...     index=["tiger", "zebra", "cow"],
+        ... )
+        >>> df
+                a   b   c
+        tiger   1   2   1
+        zebra   2   3   3
+        cow     3   4   5
+        >>> df.skew()
+        a   0.0
+        b   0.0
+        c   0.0
+        dtype: float64
+
+        Using axis=1
+
+        >>> df.skew(axis=1)
+        tiger   1.732051
+        zebra  -1.732051
+        cow     0.000000
+        dtype: float64
+
+        In this case, `numeric_only` should be set to `True` to avoid
+        getting an error.
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [1, 2, 3], "b": ["T", "Z", "X"]}, index=["tiger", "zebra", "cow"]
+        ... )
+        >>> df.skew(numeric_only=True)
+        a   0.0
+        dtype: float64
+        """
+        result = super().skew(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="skew")
+        return result
+
+    # error: Signature of "kurt" incompatible with supertype "NDFrame"
+    @overload  # type: ignore[override]
+    def kurt(
+        self,
+        *,
+        axis: Axis = ...,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series: ...
+
+    @overload
+    def kurt(
+        self,
+        *,
+        axis: None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Any: ...
+
+    @overload
+    def kurt(
+        self,
+        *,
+        axis: Axis | None,
+        skipna: bool = ...,
+        numeric_only: bool = ...,
+        **kwargs,
+    ) -> Series | Any: ...
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="kurt")
+    def kurt(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | Any:
+        """
+        Return unbiased kurtosis over requested axis.
+
+        Kurtosis obtained using Fisher's definition of
+        kurtosis (kurtosis of normal == 0.0). Normalized by N-1.
+
+        Parameters
+        ----------
+        axis : {index (0), columns (1)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series or scalar
+            Unbiased kurtosis over requested axis.
+
+        See Also
+        --------
+        Dataframe.kurtosis : Returns unbiased kurtosis over requested axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 2, 3], index=["cat", "dog", "dog", "mouse"])
+        >>> s
+        cat    1
+        dog    2
+        dog    2
+        mouse  3
+        dtype: int64
+        >>> s.kurt()
+        1.5
+
+        With a DataFrame
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [1, 2, 2, 3], "b": [3, 4, 4, 4]},
+        ...     index=["cat", "dog", "dog", "mouse"],
+        ... )
+        >>> df
+               a   b
+          cat  1   3
+          dog  2   4
+          dog  2   4
+        mouse  3   4
+        >>> df.kurt()
+        a   1.5
+        b   4.0
+        dtype: float64
+
+        With axis=None
+
+        >>> df.kurt(axis=None).round(6)
+        -0.988693
+
+        Using axis=1
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [1, 2], "b": [3, 4], "c": [3, 4], "d": [1, 2]},
+        ...     index=["cat", "dog"],
+        ... )
+        >>> df.kurt(axis=1)
+        cat   -6.0
+        dog   -6.0
+        dtype: float64
+        """
+        result = super().kurt(
+            axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+        if isinstance(result, Series):
+            result = result.__finalize__(self, method="kurt")
+        return result
+
+    # error: Incompatible types in assignment
+    kurtosis = kurt  # type: ignore[assignment]
+    product = prod
+
+    @doc(make_doc("cummin", ndim=2))
+    def cummin(
+        self,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        *args,
+        **kwargs,
+    ) -> Self:
+        data = self._get_numeric_data() if numeric_only else self
+        return NDFrame.cummin(data, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cummax", ndim=2))
+    def cummax(
+        self,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        *args,
+        **kwargs,
+    ) -> Self:
+        data = self._get_numeric_data() if numeric_only else self
+        return NDFrame.cummax(data, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cumsum", ndim=2))
+    def cumsum(
+        self,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        *args,
+        **kwargs,
+    ) -> Self:
+        data = self._get_numeric_data() if numeric_only else self
+        return NDFrame.cumsum(data, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cumprod", 2))
+    def cumprod(
+        self,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        *args,
+        **kwargs,
+    ) -> Self:
+        data = self._get_numeric_data() if numeric_only else self
+        return NDFrame.cumprod(data, axis, skipna, *args, **kwargs)
+
+    def nunique(self, axis: Axis = 0, dropna: bool = True) -> Series:
+        """
+        Count number of distinct elements in specified axis.
+
+        Return Series with number of distinct elements. Can ignore NaN
+        values.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to use. 0 or 'index' for row-wise, 1 or 'columns' for
+            column-wise.
+        dropna : bool, default True
+            Don't include NaN in the counts.
+
+        Returns
+        -------
+        Series
+            Series with counts of unique values per row or column, depending on `axis`.
+
+        See Also
+        --------
+        Series.nunique: Method nunique for Series.
+        DataFrame.count: Count non-NA cells for each column or row.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [4, 5, 6], "B": [4, 1, 1]})
+        >>> df.nunique()
+        A    3
+        B    2
+        dtype: int64
+
+        >>> df.nunique(axis=1)
+        0    1
+        1    2
+        2    2
+        dtype: int64
+        """
+        return self.apply(Series.nunique, axis=axis, dropna=dropna)
+
+    def idxmin(
+        self, axis: Axis = 0, skipna: bool = True, numeric_only: bool = False
+    ) -> Series:
+        """
+        Return index of first occurrence of minimum over requested axis.
+
+        NA/null values are excluded.
+
+        Parameters
+        ----------
+        axis : {{0 or 'index', 1 or 'columns'}}, default 0
+            The axis to use. 0 or 'index' for row-wise, 1 or 'columns' for column-wise.
+        skipna : bool, default True
+            Exclude NA/null values. If the entire DataFrame is NA,
+            or if ``skipna=False`` and there is an NA value, this method
+            will raise a ``ValueError``.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        Series
+            Indexes of minima along the specified axis.
+
+        Raises
+        ------
+        ValueError
+            * If the row/column is empty
+
+        See Also
+        --------
+        Series.idxmin : Return index of the minimum element.
+
+        Notes
+        -----
+        This method is the DataFrame version of ``ndarray.argmin``.
+
+        Examples
+        --------
+        Consider a dataset containing food consumption in Argentina.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         {
+        ...             "consumption": [10.51, 103.11, 55.48],
+        ...             "co2_emissions": [37.2, 19.66, 1712],
+        ...         }
+        ...     },
+        ...     index=["Pork", "Wheat Products", "Beef"],
+        ... )
+
+        >>> df
+                        consumption  co2_emissions
+        Pork                  10.51         37.20
+        Wheat Products       103.11         19.66
+        Beef                  55.48       1712.00
+
+        By default, it returns the index for the minimum value in each column.
+
+        >>> df.idxmin()
+        consumption                Pork
+        co2_emissions    Wheat Products
+        dtype: object
+
+        To return the index for the minimum value in each row, use ``axis="columns"``.
+
+        >>> df.idxmin(axis="columns")
+        Pork                consumption
+        Wheat Products    co2_emissions
+        Beef                consumption
+        dtype: object
+        """
+        axis = self._get_axis_number(axis)
+
+        if self.empty and len(self.axes[axis]):
+            axis_dtype = self.axes[axis].dtype
+            return self._constructor_sliced(dtype=axis_dtype)
+
+        if numeric_only:
+            data = self._get_numeric_data()
+        else:
+            data = self
+
+        res = data._reduce(
+            nanops.nanargmin, "argmin", axis=axis, skipna=skipna, numeric_only=False
+        )
+        indices = res._values
+        # indices will always be np.ndarray since axis is not N
+
+        if (indices == -1).any():
+            warnings.warn(
+                f"The behavior of {type(self).__name__}.idxmin with all-NA "
+                "values, or any-NA and skipna=False, is deprecated. In a future "
+                "version this will raise ValueError",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        index = data._get_axis(axis)
+        result = algorithms.take(
+            index._values, indices, allow_fill=True, fill_value=index._na_value
+        )
+        final_result = data._constructor_sliced(result, index=data._get_agg_axis(axis))
+        return final_result.__finalize__(self, method="idxmin")
+
+    def idxmax(
+        self, axis: Axis = 0, skipna: bool = True, numeric_only: bool = False
+    ) -> Series:
+        """
+        Return index of first occurrence of maximum over requested axis.
+
+        NA/null values are excluded.
+
+        Parameters
+        ----------
+        axis : {{0 or 'index', 1 or 'columns'}}, default 0
+            The axis to use. 0 or 'index' for row-wise, 1 or 'columns' for column-wise.
+        skipna : bool, default True
+            Exclude NA/null values. If the entire DataFrame is NA,
+            or if ``skipna=False`` and there is an NA value, this method
+            will raise a ``ValueError``.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        Series
+            Indexes of maxima along the specified axis.
+
+        Raises
+        ------
+        ValueError
+            * If the row/column is empty
+
+        See Also
+        --------
+        Series.idxmax : Return index of the maximum element.
+
+        Notes
+        -----
+        This method is the DataFrame version of ``ndarray.argmax``.
+
+        Examples
+        --------
+        Consider a dataset containing food consumption in Argentina.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         {
+        ...             "consumption": [10.51, 103.11, 55.48],
+        ...             "co2_emissions": [37.2, 19.66, 1712],
+        ...         }
+        ...     },
+        ...     index=["Pork", "Wheat Products", "Beef"],
+        ... )
+
+        >>> df
+                        consumption  co2_emissions
+        Pork                  10.51         37.20
+        Wheat Products       103.11         19.66
+        Beef                  55.48       1712.00
+
+        By default, it returns the index for the maximum value in each column.
+
+        >>> df.idxmax()
+        consumption     Wheat Products
+        co2_emissions             Beef
+        dtype: object
+
+        To return the index for the maximum value in each row, use ``axis="columns"``.
+
+        >>> df.idxmax(axis="columns")
+        Pork              co2_emissions
+        Wheat Products     consumption
+        Beef              co2_emissions
+        dtype: object
+        """
+        axis = self._get_axis_number(axis)
+
+        if self.empty and len(self.axes[axis]):
+            axis_dtype = self.axes[axis].dtype
+            return self._constructor_sliced(dtype=axis_dtype)
+
+        if numeric_only:
+            data = self._get_numeric_data()
+        else:
+            data = self
+
+        res = data._reduce(
+            nanops.nanargmax, "argmax", axis=axis, skipna=skipna, numeric_only=False
+        )
+        indices = res._values
+        # indices will always be 1d array since axis is not None
+
+        if (indices == -1).any():
+            warnings.warn(
+                f"The behavior of {type(self).__name__}.idxmax with all-NA "
+                "values, or any-NA and skipna=False, is deprecated. In a future "
+                "version this will raise ValueError",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        index = data._get_axis(axis)
+        result = algorithms.take(
+            index._values, indices, allow_fill=True, fill_value=index._na_value
+        )
+        final_result = data._constructor_sliced(result, index=data._get_agg_axis(axis))
+        return final_result.__finalize__(self, method="idxmax")
+
+    def _get_agg_axis(self, axis_num: int) -> Index:
+        """
+        Let's be explicit about this.
+        """
+        if axis_num == 0:
+            return self.columns
+        elif axis_num == 1:
+            return self.index
+        else:
+            raise ValueError(f"Axis must be 0 or 1 (got {axis_num!r})")
+
+    def mode(
+        self, axis: Axis = 0, numeric_only: bool = False, dropna: bool = True
+    ) -> DataFrame:
+        """
+        Get the mode(s) of each element along the selected axis.
+
+        The mode of a set of values is the value that appears most often.
+        It can be multiple values.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to iterate over while searching for the mode:
+
+            * 0 or 'index' : get mode of each column
+            * 1 or 'columns' : get mode of each row.
+
+        numeric_only : bool, default False
+            If True, only apply to numeric columns.
+        dropna : bool, default True
+            Don't consider counts of NaN/NaT.
+
+        Returns
+        -------
+        DataFrame
+            The modes of each column or row.
+
+        See Also
+        --------
+        Series.mode : Return the highest frequency value in a Series.
+        Series.value_counts : Return the counts of values in a Series.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("bird", 2, 2),
+        ...         ("mammal", 4, np.nan),
+        ...         ("arthropod", 8, 0),
+        ...         ("bird", 2, np.nan),
+        ...     ],
+        ...     index=("falcon", "horse", "spider", "ostrich"),
+        ...     columns=("species", "legs", "wings"),
+        ... )
+        >>> df
+                   species  legs  wings
+        falcon        bird     2    2.0
+        horse       mammal     4    NaN
+        spider   arthropod     8    0.0
+        ostrich       bird     2    NaN
+
+        By default, missing values are not considered, and the mode of wings
+        are both 0 and 2. Because the resulting DataFrame has two rows,
+        the second row of ``species`` and ``legs`` contains ``NaN``.
+
+        >>> df.mode()
+          species  legs  wings
+        0    bird   2.0    0.0
+        1     NaN   NaN    2.0
+
+        Setting ``dropna=False`` ``NaN`` values are considered and they can be
+        the mode (like for wings).
+
+        >>> df.mode(dropna=False)
+          species  legs  wings
+        0    bird     2    NaN
+
+        Setting ``numeric_only=True``, only the mode of numeric columns is
+        computed, and columns of other types are ignored.
+
+        >>> df.mode(numeric_only=True)
+           legs  wings
+        0   2.0    0.0
+        1   NaN    2.0
+
+        To compute the mode over columns and not rows, use the axis parameter:
+
+        >>> df.mode(axis="columns", numeric_only=True)
+                   0    1
+        falcon   2.0  NaN
+        horse    4.0  NaN
+        spider   0.0  8.0
+        ostrich  2.0  NaN
+        """
+        data = self if not numeric_only else self._get_numeric_data()
+
+        def f(s):
+            return s.mode(dropna=dropna)
+
+        data = data.apply(f, axis=axis)
+        # Ensure index is type stable (should always use int index)
+        if data.empty:
+            data.index = default_index(0)
+
+        return data
+
+    @overload
+    def quantile(
+        self,
+        q: float = ...,
+        axis: Axis = ...,
+        numeric_only: bool = ...,
+        interpolation: QuantileInterpolation = ...,
+        method: Literal["single", "table"] = ...,
+    ) -> Series: ...
+
+    @overload
+    def quantile(
+        self,
+        q: AnyArrayLike | Sequence[float],
+        axis: Axis = ...,
+        numeric_only: bool = ...,
+        interpolation: QuantileInterpolation = ...,
+        method: Literal["single", "table"] = ...,
+    ) -> Series | DataFrame: ...
+
+    @overload
+    def quantile(
+        self,
+        q: float | AnyArrayLike | Sequence[float] = ...,
+        axis: Axis = ...,
+        numeric_only: bool = ...,
+        interpolation: QuantileInterpolation = ...,
+        method: Literal["single", "table"] = ...,
+    ) -> Series | DataFrame: ...
+
+    def quantile(
+        self,
+        q: float | AnyArrayLike | Sequence[float] = 0.5,
+        axis: Axis = 0,
+        numeric_only: bool = False,
+        interpolation: QuantileInterpolation = "linear",
+        method: Literal["single", "table"] = "single",
+    ) -> Series | DataFrame:
+        """
+        Return values at the given quantile over requested axis.
+
+        Parameters
+        ----------
+        q : float or array-like, default 0.5 (50% quantile)
+            Value between 0 <= q <= 1, the quantile(s) to compute.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Equals 0 or 'index' for row-wise, 1 or 'columns' for column-wise.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        interpolation : {'linear', 'lower', 'higher', 'midpoint', 'nearest'}
+            This optional parameter specifies the interpolation method to use,
+            when the desired quantile lies between two data points `i` and `j`:
+
+            * linear: `i + (j - i) * fraction`, where `fraction` is the
+              fractional part of the index surrounded by `i` and `j`.
+            * lower: `i`.
+            * higher: `j`.
+            * nearest: `i` or `j` whichever is nearest.
+            * midpoint: (`i` + `j`) / 2.
+        method : {'single', 'table'}, default 'single'
+            Whether to compute quantiles per-column ('single') or over all columns
+            ('table'). When 'table', the only allowed interpolation methods are
+            'nearest', 'lower', and 'higher'.
+
+        Returns
+        -------
+        Series or DataFrame
+
+            If ``q`` is an array, a DataFrame will be returned where the
+              index is ``q``, the columns are the columns of self, and the
+              values are the quantiles.
+            If ``q`` is a float, a Series will be returned where the
+              index is the columns of self and the values are the quantiles.
+
+        See Also
+        --------
+        core.window.rolling.Rolling.quantile: Rolling quantile.
+        numpy.percentile: Numpy function to compute the percentile.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     np.array([[1, 1], [2, 10], [3, 100], [4, 100]]), columns=["a", "b"]
+        ... )
+        >>> df.quantile(0.1)
+        a    1.3
+        b    3.7
+        Name: 0.1, dtype: float64
+        >>> df.quantile([0.1, 0.5])
+               a     b
+        0.1  1.3   3.7
+        0.5  2.5  55.0
+
+        Specifying `method='table'` will compute the quantile over all columns.
+
+        >>> df.quantile(0.1, method="table", interpolation="nearest")
+        a    1
+        b    1
+        Name: 0.1, dtype: int64
+        >>> df.quantile([0.1, 0.5], method="table", interpolation="nearest")
+             a    b
+        0.1  1    1
+        0.5  3  100
+
+        Specifying `numeric_only=False` will also compute the quantile of
+        datetime and timedelta data.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": [1, 2],
+        ...         "B": [pd.Timestamp("2010"), pd.Timestamp("2011")],
+        ...         "C": [pd.Timedelta("1 days"), pd.Timedelta("2 days")],
+        ...     }
+        ... )
+        >>> df.quantile(0.5, numeric_only=False)
+        A                    1.5
+        B    2010-07-02 12:00:00
+        C        1 days 12:00:00
+        Name: 0.5, dtype: object
+        """
+        validate_percentile(q)
+        axis = self._get_axis_number(axis)
+
+        if not is_list_like(q):
+            # BlockManager.quantile expects listlike, so we wrap and unwrap here
+            # error: List item 0 has incompatible type "float | ExtensionArray |
+            # ndarray[Any, Any] | Index | Series | Sequence[float]"; expected "float"
+            res_df = self.quantile(
+                [q],  # type: ignore[list-item]
+                axis=axis,
+                numeric_only=numeric_only,
+                interpolation=interpolation,
+                method=method,
+            )
+            if method == "single":
+                res = res_df.iloc[0]
+            else:
+                # cannot directly iloc over sparse arrays
+                res = res_df.T.iloc[:, 0]
+            if axis == 1 and len(self) == 0:
+                # GH#41544 try to get an appropriate dtype
+                dtype = find_common_type(list(self.dtypes))
+                if needs_i8_conversion(dtype):
+                    return res.astype(dtype)
+            return res
+
+        q = Index(q, dtype=np.float64)
+        data = self._get_numeric_data() if numeric_only else self
+
+        if axis == 1:
+            data = data.T
+
+        if len(data.columns) == 0:
+            # GH#23925 _get_numeric_data may have dropped all columns
+            cols = self.columns[:0]
+
+            dtype = np.float64
+            if axis == 1:
+                # GH#41544 try to get an appropriate dtype
+                cdtype = find_common_type(list(self.dtypes))
+                if needs_i8_conversion(cdtype):
+                    dtype = cdtype
+
+            res = self._constructor([], index=q, columns=cols, dtype=dtype)
+            return res.__finalize__(self, method="quantile")
+
+        valid_method = {"single", "table"}
+        if method not in valid_method:
+            raise ValueError(
+                f"Invalid method: {method}. Method must be in {valid_method}."
+            )
+        if method == "single":
+            res = data._mgr.quantile(qs=q, interpolation=interpolation)
+        elif method == "table":
+            valid_interpolation = {"nearest", "lower", "higher"}
+            if interpolation not in valid_interpolation:
+                raise ValueError(
+                    f"Invalid interpolation: {interpolation}. "
+                    f"Interpolation must be in {valid_interpolation}"
+                )
+            # handle degenerate case
+            if len(data) == 0:
+                if data.ndim == 2:
+                    dtype = find_common_type(list(self.dtypes))
+                else:
+                    dtype = self.dtype
+                return self._constructor([], index=q, columns=data.columns, dtype=dtype)
+
+            q_idx = np.quantile(np.arange(len(data)), q, method=interpolation)
+
+            by = data.columns
+            if len(by) > 1:
+                keys = [data._get_label_or_level_values(x) for x in by]
+                indexer = lexsort_indexer(keys)
+            else:
+                k = data._get_label_or_level_values(by[0])
+                indexer = nargsort(k)
+
+            res = data._mgr.take(indexer[q_idx], verify=False)
+            res.axes[1] = q
+
+        result = self._constructor_from_mgr(res, axes=res.axes)
+        return result.__finalize__(self, method="quantile")
+
+    def to_timestamp(
+        self,
+        freq: Frequency | None = None,
+        how: ToTimestampHow = "start",
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> DataFrame:
+        """
+        Cast PeriodIndex to DatetimeIndex of timestamps, at *beginning* of period.
+
+        This can be changed to the *end* of the period, by specifying `how="e"`.
+
+        Parameters
+        ----------
+        freq : str, default frequency of PeriodIndex
+            Desired frequency.
+        how : {'s', 'e', 'start', 'end'}
+            Convention for converting period to timestamp; start of period
+            vs. end.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to convert (the index by default).
+        copy : bool, default False
+            If False then underlying input data is not copied.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        DataFrame with DatetimeIndex
+            DataFrame with the PeriodIndex cast to DatetimeIndex.
+
+        See Also
+        --------
+        DataFrame.to_period: Inverse method to cast DatetimeIndex to PeriodIndex.
+        Series.to_timestamp: Equivalent method for Series.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023", "2024"], freq="Y")
+        >>> d = {"col1": [1, 2], "col2": [3, 4]}
+        >>> df1 = pd.DataFrame(data=d, index=idx)
+        >>> df1
+              col1   col2
+        2023     1      3
+        2024	 2      4
+
+        The resulting timestamps will be at the beginning of the year in this case
+
+        >>> df1 = df1.to_timestamp()
+        >>> df1
+                    col1   col2
+        2023-01-01     1      3
+        2024-01-01     2      4
+        >>> df1.index
+        DatetimeIndex(['2023-01-01', '2024-01-01'], dtype='datetime64[ns]', freq=None)
+
+        Using `freq` which is the offset that the Timestamps will have
+
+        >>> df2 = pd.DataFrame(data=d, index=idx)
+        >>> df2 = df2.to_timestamp(freq="M")
+        >>> df2
+                    col1   col2
+        2023-01-31     1      3
+        2024-01-31     2      4
+        >>> df2.index
+        DatetimeIndex(['2023-01-31', '2024-01-31'], dtype='datetime64[ns]', freq=None)
+        """
+        self._check_copy_deprecation(copy)
+        new_obj = self.copy(deep=False)
+
+        axis_name = self._get_axis_name(axis)
+        old_ax = getattr(self, axis_name)
+        if not isinstance(old_ax, PeriodIndex):
+            raise TypeError(f"unsupported Type {type(old_ax).__name__}")
+
+        new_ax = old_ax.to_timestamp(freq=freq, how=how)
+
+        setattr(new_obj, axis_name, new_ax)
+        return new_obj
+
+    def to_period(
+        self,
+        freq: Frequency | None = None,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> DataFrame:
+        """
+        Convert DataFrame from DatetimeIndex to PeriodIndex.
+
+        Convert DataFrame from DatetimeIndex to PeriodIndex with desired
+        frequency (inferred from index if not passed). Either index of columns can be
+        converted, depending on `axis` argument.
+
+        Parameters
+        ----------
+        freq : str, default
+            Frequency of the PeriodIndex.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to convert (the index by default).
+        copy : bool, default False
+            If False then underlying input data is not copied.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        DataFrame
+            The DataFrame with the converted PeriodIndex.
+
+        See Also
+        --------
+        Series.to_period: Equivalent method for Series.
+        Series.dt.to_period: Convert DateTime column values.
+
+        Examples
+        --------
+        >>> idx = pd.to_datetime(
+        ...     [
+        ...         "2001-03-31 00:00:00",
+        ...         "2002-05-31 00:00:00",
+        ...         "2003-08-31 00:00:00",
+        ...     ]
+        ... )
+
+        >>> idx
+        DatetimeIndex(['2001-03-31', '2002-05-31', '2003-08-31'],
+        dtype='datetime64[s]', freq=None)
+
+        >>> idx.to_period("M")
+        PeriodIndex(['2001-03', '2002-05', '2003-08'], dtype='period[M]')
+
+        For the yearly frequency
+
+        >>> idx.to_period("Y")
+        PeriodIndex(['2001', '2002', '2003'], dtype='period[Y-DEC]')
+        """
+        self._check_copy_deprecation(copy)
+        new_obj = self.copy(deep=False)
+
+        axis_name = self._get_axis_name(axis)
+        old_ax = getattr(self, axis_name)
+        if not isinstance(old_ax, DatetimeIndex):
+            raise TypeError(f"unsupported Type {type(old_ax).__name__}")
+
+        new_ax = old_ax.to_period(freq=freq)
+
+        setattr(new_obj, axis_name, new_ax)
+        return new_obj
+
+    def isin(self, values: Series | DataFrame | Sequence | Mapping) -> DataFrame:
+        """
+        Whether each element in the DataFrame is contained in values.
+
+        Parameters
+        ----------
+        values : iterable, Series, DataFrame or dict
+            The result will only be true at a location if all the
+            labels match. If `values` is a Series, that's the index. If
+            `values` is a dict, the keys must be the column names,
+            which must match. If `values` is a DataFrame,
+            then both the index and column labels must match.
+
+        Returns
+        -------
+        DataFrame
+            DataFrame of booleans showing whether each element in the DataFrame
+            is contained in values.
+
+        See Also
+        --------
+        DataFrame.eq: Equality test for DataFrame.
+        Series.isin: Equivalent method on Series.
+        Series.str.contains: Test if pattern or regex is contained within a
+            string of a Series or Index.
+
+        Notes
+        -----
+            ``__iter__`` is used (and not ``__contains__``) to iterate over values
+            when checking if it contains the elements in DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"num_legs": [2, 4], "num_wings": [2, 0]}, index=["falcon", "dog"]
+        ... )
+        >>> df
+                num_legs  num_wings
+        falcon         2          2
+        dog            4          0
+
+        When ``values`` is a list check whether every value in the DataFrame
+        is present in the list (which animals have 0 or 2 legs or wings)
+
+        >>> df.isin([0, 2])
+                num_legs  num_wings
+        falcon      True       True
+        dog        False       True
+
+        To check if ``values`` is *not* in the DataFrame, use the ``~`` operator:
+
+        >>> ~df.isin([0, 2])
+                num_legs  num_wings
+        falcon     False      False
+        dog         True      False
+
+        When ``values`` is a dict, we can pass values to check for each
+        column separately:
+
+        >>> df.isin({"num_wings": [0, 3]})
+                num_legs  num_wings
+        falcon     False      False
+        dog        False       True
+
+        When ``values`` is a Series or DataFrame the index and column must
+        match. Note that 'falcon' does not match based on the number of legs
+        in other.
+
+        >>> other = pd.DataFrame(
+        ...     {"num_legs": [8, 3], "num_wings": [0, 2]}, index=["spider", "falcon"]
+        ... )
+        >>> df.isin(other)
+                num_legs  num_wings
+        falcon     False       True
+        dog        False      False
+        """
+        if isinstance(values, dict):
+            from pandas.core.reshape.concat import concat
+
+            values = collections.defaultdict(list, values)
+            result = concat(
+                (
+                    self.iloc[:, [i]].isin(values[col])
+                    for i, col in enumerate(self.columns)
+                ),
+                axis=1,
+            )
+        elif isinstance(values, Series):
+            if not values.index.is_unique:
+                raise ValueError("cannot compute isin with a duplicate axis.")
+            result = self.eq(values.reindex_like(self), axis="index")
+        elif isinstance(values, DataFrame):
+            if not (values.columns.is_unique and values.index.is_unique):
+                raise ValueError("cannot compute isin with a duplicate axis.")
+            result = self.eq(values.reindex_like(self))
+        else:
+            if not is_list_like(values):
+                raise TypeError(
+                    "only list-like or dict-like objects are allowed "
+                    "to be passed to DataFrame.isin(), "
+                    f"you passed a '{type(values).__name__}'"
+                )
+
+            def isin_(x):
+                # error: Argument 2 to "isin" has incompatible type "Union[Series,
+                # DataFrame, Sequence[Any], Mapping[Any, Any]]"; expected
+                # "Union[Union[Union[ExtensionArray, ndarray[Any, Any]], Index,
+                # Series], List[Any], range]"
+                result = algorithms.isin(
+                    x.ravel(),
+                    values,  # type: ignore[arg-type]
+                )
+                return result.reshape(x.shape)
+
+            res_mgr = self._mgr.apply(isin_)
+            result = self._constructor_from_mgr(
+                res_mgr,
+                axes=res_mgr.axes,
+            )
+        return result.__finalize__(self, method="isin")
+
+    # ----------------------------------------------------------------------
+    # Add index and columns
+    _AXIS_ORDERS: list[Literal["index", "columns"]] = ["index", "columns"]
+    _AXIS_TO_AXIS_NUMBER: dict[Axis, int] = {
+        **NDFrame._AXIS_TO_AXIS_NUMBER,
+        1: 1,
+        "columns": 1,
+    }
+    _AXIS_LEN = len(_AXIS_ORDERS)
+    _info_axis_number: Literal[1] = 1
+    _info_axis_name: Literal["columns"] = "columns"
+
+    index = properties.AxisProperty(
+        axis=1,
+        doc="""
+        The index (row labels) of the DataFrame.
+
+        The index of a DataFrame is a series of labels that identify each row.
+        The labels can be integers, strings, or any other hashable type. The index
+        is used for label-based access and alignment, and can be accessed or
+        modified using this attribute.
+
+        Returns
+        -------
+        pandas.Index
+            The index labels of the DataFrame.
+
+        See Also
+        --------
+        DataFrame.columns : The column labels of the DataFrame.
+        DataFrame.to_numpy : Convert the DataFrame to a NumPy array.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({'Name': ['Alice', 'Bob', 'Aritra'],
+        ...                    'Age': [25, 30, 35],
+        ...                    'Location': ['Seattle', 'New York', 'Kona']},
+        ...                   index=([10, 20, 30]))
+        >>> df.index
+        Index([10, 20, 30], dtype='int64')
+
+        In this example, we create a DataFrame with 3 rows and 3 columns,
+        including Name, Age, and Location information. We set the index labels to
+        be the integers 10, 20, and 30. We then access the `index` attribute of the
+        DataFrame, which returns an `Index` object containing the index labels.
+
+        >>> df.index = [100, 200, 300]
+        >>> df
+            Name  Age Location
+        100  Alice   25  Seattle
+        200    Bob   30 New York
+        300  Aritra  35    Kona
+
+        In this example, we modify the index labels of the DataFrame by assigning
+        a new list of labels to the `index` attribute. The DataFrame is then
+        updated with the new labels, and the output shows the modified DataFrame.
+        """,
+    )
+    columns = properties.AxisProperty(
+        axis=0,
+        doc="""
+        The column labels of the DataFrame.
+
+        This property holds the column names as a pandas ``Index`` object.
+        It provides an immutable sequence of column labels that can be
+        used for data selection, renaming, and alignment in DataFrame operations.
+
+        Returns
+        -------
+        pandas.Index
+            The column labels of the DataFrame.
+
+        See Also
+        --------
+        DataFrame.index: The index (row labels) of the DataFrame.
+        DataFrame.axes: Return a list representing the axes of the DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({'A': [1, 2], 'B': [3, 4]})
+        >>> df
+                A  B
+        0    1  3
+        1    2  4
+        >>> df.columns
+        Index(['A', 'B'], dtype='object')
+        """,
+    )
+
+    # ----------------------------------------------------------------------
+    # Add plotting methods to DataFrame
+    plot = Accessor("plot", pandas.plotting.PlotAccessor)
+    hist = pandas.plotting.hist_frame
+    boxplot = pandas.plotting.boxplot_frame
+    sparse = Accessor("sparse", SparseFrameAccessor)
+
+    # ----------------------------------------------------------------------
+    # Internal Interface Methods
+
+    def _to_dict_of_blocks(self):
+        """
+        Return a dict of dtype -> Constructor Types that
+        each is a homogeneous dtype.
+
+        Internal ONLY.
+        """
+        mgr = self._mgr
+        return {
+            k: self._constructor_from_mgr(v, axes=v.axes).__finalize__(self)
+            for k, v in mgr.to_iter_dict()
+        }
+
+    @property
+    def values(self) -> np.ndarray:
+        """
+        Return a Numpy representation of the DataFrame.
+
+        .. warning::
+
+           We recommend using :meth:`DataFrame.to_numpy` instead.
+
+        Only the values in the DataFrame will be returned, the axes labels
+        will be removed.
+
+        Returns
+        -------
+        numpy.ndarray
+            The values of the DataFrame.
+
+        See Also
+        --------
+        DataFrame.to_numpy : Recommended alternative to this method.
+        DataFrame.index : Retrieve the index labels.
+        DataFrame.columns : Retrieving the column names.
+
+        Notes
+        -----
+        The dtype will be a lower-common-denominator dtype (implicit
+        upcasting); that is to say if the dtypes (even of numeric types)
+        are mixed, the one that accommodates all will be chosen. Use this
+        with care if you are not dealing with the blocks.
+
+        e.g. If the dtypes are float16 and float32, dtype will be upcast to
+        float32.  If dtypes are int32 and uint8, dtype will be upcast to
+        int32. By :func:`numpy.find_common_type` convention, mixing int64
+        and uint64 will result in a float64 dtype.
+
+        Examples
+        --------
+        A DataFrame where all columns are the same type (e.g., int64) results
+        in an array of the same type.
+
+        >>> df = pd.DataFrame(
+        ...     {"age": [3, 29], "height": [94, 170], "weight": [31, 115]}
+        ... )
+        >>> df
+           age  height  weight
+        0    3      94      31
+        1   29     170     115
+        >>> df.dtypes
+        age       int64
+        height    int64
+        weight    int64
+        dtype: object
+        >>> df.values
+        array([[  3,  94,  31],
+               [ 29, 170, 115]])
+
+        A DataFrame with mixed type columns(e.g., str/object, int64, float32)
+        results in an ndarray of the broadest type that accommodates these
+        mixed types (e.g., object).
+
+        >>> df2 = pd.DataFrame(
+        ...     [
+        ...         ("parrot", 24.0, "second"),
+        ...         ("lion", 80.5, 1),
+        ...         ("monkey", np.nan, None),
+        ...     ],
+        ...     columns=("name", "max_speed", "rank"),
+        ... )
+        >>> df2.dtypes
+        name          object
+        max_speed    float64
+        rank          object
+        dtype: object
+        >>> df2.values
+        array([['parrot', 24.0, 'second'],
+               ['lion', 80.5, 1],
+               ['monkey', nan, None]], dtype=object)
+        """
+        return self._mgr.as_array()
+
+
+def _from_nested_dict(
+    data: Mapping[HashableT, Mapping[HashableT2, T]],
+) -> collections.defaultdict[HashableT2, dict[HashableT, T]]:
+    new_data: collections.defaultdict[HashableT2, dict[HashableT, T]] = (
+        collections.defaultdict(dict)
+    )
+    for index, s in data.items():
+        for col, v in s.items():
+            new_data[col][index] = v
+    return new_data
+
+
+def _reindex_for_setitem(
+    value: DataFrame | Series, index: Index
+) -> tuple[ArrayLike, BlockValuesRefs | None]:
+    # reindex if necessary
+
+    if value.index.equals(index) or not len(index):
+        if isinstance(value, Series):
+            return value._values, value._references
+        return value._values.copy(), None
+
+    # GH#4107
+    try:
+        reindexed_value = value.reindex(index)._values
+    except ValueError as err:
+        # raised in MultiIndex.from_tuples, see test_insert_error_msmgs
+        if not value.index.is_unique:
+            # duplicate axis
+            raise err
+
+        raise TypeError(
+            "incompatible index of inserted column with frame index"
+        ) from err
+    return reindexed_value, None
diff --git a/pandas/pandas/core/generic.py b/pandas/pandas/core/generic.py
new file mode 100644
index 0000000000000..80c43b76e53db
--- /dev/null
+++ b/pandas/pandas/core/generic.py
@@ -0,0 +1,13000 @@
+# pyright: reportPropertyTypeMismatch=false
+from __future__ import annotations
+
+import collections
+from copy import deepcopy
+import datetime as dt
+from functools import partial
+from json import loads
+import operator
+import pickle
+import re
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    ClassVar,
+    Literal,
+    NoReturn,
+    cast,
+    final,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import config
+
+from pandas._libs import lib
+from pandas._libs.lib import is_range_indexer
+from pandas._libs.tslibs import (
+    Period,
+    Timestamp,
+    to_offset,
+)
+from pandas._typing import (
+    AlignJoin,
+    AnyArrayLike,
+    ArrayLike,
+    Axes,
+    Axis,
+    AxisInt,
+    CompressionOptions,
+    Concatenate,
+    DtypeArg,
+    DtypeBackend,
+    DtypeObj,
+    FilePath,
+    FillnaOptions,
+    FloatFormatType,
+    FormattersType,
+    Frequency,
+    IgnoreRaise,
+    IndexKeyFunc,
+    IndexLabel,
+    InterpolateOptions,
+    IntervalClosedType,
+    JSONSerializable,
+    Level,
+    ListLike,
+    Manager,
+    NaPosition,
+    NDFrameT,
+    OpenFileErrors,
+    RandomState,
+    ReindexMethod,
+    Renamer,
+    Scalar,
+    Self,
+    SequenceNotStr,
+    SortKind,
+    StorageOptions,
+    Suffixes,
+    T,
+    TimeAmbiguous,
+    TimedeltaConvertibleTypes,
+    TimeNonexistent,
+    TimestampConvertibleTypes,
+    TimeUnit,
+    ValueKeyFunc,
+    WriteBuffer,
+    WriteExcelBuffer,
+    npt,
+)
+from pandas.compat import PYPY
+from pandas.compat._constants import REF_COUNT
+from pandas.compat._optional import import_optional_dependency
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    AbstractMethodError,
+    ChainedAssignmentError,
+    InvalidIndexError,
+)
+from pandas.errors.cow import _chained_assignment_method_msg
+from pandas.util._decorators import (
+    deprecate_kwarg,
+    doc,
+)
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import (
+    check_dtype_backend,
+    validate_ascending,
+    validate_bool_kwarg,
+    validate_inclusive,
+)
+
+from pandas.core.dtypes.astype import astype_is_view
+from pandas.core.dtypes.common import (
+    ensure_object,
+    ensure_platform_int,
+    ensure_str,
+    is_bool,
+    is_bool_dtype,
+    is_dict_like,
+    is_extension_array_dtype,
+    is_list_like,
+    is_number,
+    is_numeric_dtype,
+    is_re_compilable,
+    is_scalar,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import (
+    is_hashable,
+    is_nested_list_like,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    notna,
+)
+
+from pandas.core import (
+    algorithms as algos,
+    arraylike,
+    common,
+    indexing,
+    missing,
+    nanops,
+    sample,
+)
+from pandas.core.array_algos.replace import should_use_regex
+from pandas.core.arrays import ExtensionArray
+from pandas.core.base import PandasObject
+from pandas.core.construction import extract_array
+from pandas.core.flags import Flags
+from pandas.core.indexes.api import (
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    default_index,
+    ensure_index,
+)
+from pandas.core.internals import BlockManager
+from pandas.core.methods.describe import describe_ndframe
+from pandas.core.missing import (
+    clean_fill_method,
+    clean_reindex_fill_method,
+    find_valid_index,
+)
+from pandas.core.reshape.concat import concat
+from pandas.core.shared_docs import _shared_docs
+from pandas.core.sorting import get_indexer_indexer
+from pandas.core.window import (
+    Expanding,
+    ExponentialMovingWindow,
+    Rolling,
+    Window,
+)
+
+from pandas.io.formats.format import (
+    DataFrameFormatter,
+    DataFrameRenderer,
+)
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+    from collections.abc import (
+        Hashable,
+        Iterator,
+        Mapping,
+        Sequence,
+    )
+
+    from pandas._libs.tslibs import BaseOffset
+    from pandas._typing import P
+
+    from pandas import (
+        DataFrame,
+        ExcelWriter,
+        HDFStore,
+        Series,
+    )
+    from pandas.core.indexers.objects import BaseIndexer
+    from pandas.core.resample import Resampler
+
+import textwrap
+
+# goal is to be able to define the docs close to function, while still being
+# able to share
+_shared_docs = {**_shared_docs}
+_shared_doc_kwargs = {
+    "axes": "keywords for axes",
+    "klass": "Series/DataFrame",
+    "axes_single_arg": "{0 or 'index'} for Series, {0 or 'index', 1 or 'columns'} for DataFrame",  # noqa: E501
+    "inplace": """
+    inplace : bool, default False
+        If True, performs operation inplace and returns None.""",
+    "optional_by": """
+        by : str or list of str
+            Name or list of names to sort by""",
+}
+
+
+class NDFrame(PandasObject, indexing.IndexingMixin):
+    """
+    N-dimensional analogue of DataFrame. Store multi-dimensional in a
+    size-mutable, labeled data structure
+
+    Parameters
+    ----------
+    data : BlockManager
+    axes : list
+    copy : bool, default False
+    """
+
+    _internal_names: list[str] = [
+        "_mgr",
+        "_item_cache",
+        "_cache",
+        "_name",
+        "_metadata",
+        "_flags",
+    ]
+    _internal_names_set: set[str] = set(_internal_names)
+    _accessors: set[str] = set()
+    _hidden_attrs: frozenset[str] = frozenset([])
+    _metadata: list[str] = []
+    _mgr: Manager
+    _attrs: dict[Hashable, Any]
+    _typ: str
+
+    # ----------------------------------------------------------------------
+    # Constructors
+
+    def __init__(self, data: Manager) -> None:
+        object.__setattr__(self, "_mgr", data)
+        object.__setattr__(self, "_attrs", {})
+        object.__setattr__(self, "_flags", Flags(self, allows_duplicate_labels=True))
+
+    @final
+    @classmethod
+    def _init_mgr(
+        cls,
+        mgr: Manager,
+        axes: dict[Literal["index", "columns"], Axes | None],
+        dtype: DtypeObj | None = None,
+        copy: bool = False,
+    ) -> Manager:
+        """passed a manager and a axes dict"""
+        for a, axe in axes.items():
+            if axe is not None:
+                axe = ensure_index(axe)
+                bm_axis = cls._get_block_manager_axis(a)
+                mgr = mgr.reindex_axis(axe, axis=bm_axis)
+
+        # make a copy if explicitly requested
+        if copy:
+            mgr = mgr.copy()
+        if dtype is not None:
+            # avoid further copies if we can
+            if (
+                isinstance(mgr, BlockManager)
+                and len(mgr.blocks) == 1
+                and mgr.blocks[0].values.dtype == dtype
+            ):
+                pass
+            else:
+                mgr = mgr.astype(dtype=dtype)
+        return mgr
+
+    @final
+    @classmethod
+    def _from_mgr(cls, mgr: Manager, axes: list[Index]) -> Self:
+        """
+        Construct a new object of this type from a Manager object and axes.
+
+        Parameters
+        ----------
+        mgr : Manager
+            Must have the same ndim as cls.
+        axes : list[Index]
+
+        Notes
+        -----
+        The axes must match mgr.axes, but are required for future-proofing
+        in the event that axes are refactored out of the Manager objects.
+        """
+        obj = cls.__new__(cls)
+        NDFrame.__init__(obj, mgr)
+        return obj
+
+    # ----------------------------------------------------------------------
+    # attrs and flags
+
+    @property
+    def attrs(self) -> dict[Hashable, Any]:
+        """
+        Dictionary of global attributes of this dataset.
+
+        .. warning::
+
+           attrs is experimental and may change without warning.
+
+        See Also
+        --------
+        DataFrame.flags : Global flags applying to this object.
+
+        Notes
+        -----
+        Many operations that create new datasets will copy ``attrs``. Copies
+        are always deep so that changing ``attrs`` will only affect the
+        present dataset. ``pandas.concat`` copies ``attrs`` only if all input
+        datasets have the same ``attrs``.
+
+        Examples
+        --------
+        For Series:
+
+        >>> ser = pd.Series([1, 2, 3])
+        >>> ser.attrs = {"A": [10, 20, 30]}
+        >>> ser.attrs
+        {'A': [10, 20, 30]}
+
+        For DataFrame:
+
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.attrs = {"A": [10, 20, 30]}
+        >>> df.attrs
+        {'A': [10, 20, 30]}
+        """
+        return self._attrs
+
+    @attrs.setter
+    def attrs(self, value: Mapping[Hashable, Any]) -> None:
+        self._attrs = dict(value)
+
+    @final
+    @property
+    def flags(self) -> Flags:
+        """
+        Get the properties associated with this pandas object.
+
+        The available flags are
+
+        * :attr:`Flags.allows_duplicate_labels`
+
+        See Also
+        --------
+        Flags : Flags that apply to pandas objects.
+        DataFrame.attrs : Global metadata applying to this dataset.
+
+        Notes
+        -----
+        "Flags" differ from "metadata". Flags reflect properties of the
+        pandas object (the Series or DataFrame). Metadata refer to properties
+        of the dataset, and should be stored in :attr:`DataFrame.attrs`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2]})
+        >>> df.flags
+        <Flags(allows_duplicate_labels=True)>
+
+        Flags can be get or set using ``.``
+
+        >>> df.flags.allows_duplicate_labels
+        True
+        >>> df.flags.allows_duplicate_labels = False
+
+        Or by slicing with a key
+
+        >>> df.flags["allows_duplicate_labels"]
+        False
+        >>> df.flags["allows_duplicate_labels"] = True
+        """
+        return self._flags
+
+    @final
+    def set_flags(
+        self,
+        *,
+        copy: bool | lib.NoDefault = lib.no_default,
+        allows_duplicate_labels: bool | None = None,
+    ) -> Self:
+        """
+        Return a new object with updated flags.
+
+        This method creates a shallow copy of the original object, preserving its
+        underlying data while modifying its global flags. In particular, it allows
+        you to update properties such as whether duplicate labels are permitted. This
+        behavior is especially useful in method chains, where one wishes to
+        adjust DataFrame or Series characteristics without altering the original object.
+
+        Parameters
+        ----------
+        copy : bool, default False
+            Specify if a copy of the object should be made.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        allows_duplicate_labels : bool, optional
+            Whether the returned object allows duplicate labels.
+
+        Returns
+        -------
+        Series or DataFrame
+            The same type as the caller.
+
+        See Also
+        --------
+        DataFrame.attrs : Global metadata applying to this dataset.
+        DataFrame.flags : Global flags applying to this object.
+
+        Notes
+        -----
+        This method returns a new object that's a view on the same data
+        as the input. Mutating the input or the output values will be reflected
+        in the other.
+
+        This method is intended to be used in method chains.
+
+        "Flags" differ from "metadata". Flags reflect properties of the
+        pandas object (the Series or DataFrame). Metadata refer to properties
+        of the dataset, and should be stored in :attr:`DataFrame.attrs`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2]})
+        >>> df.flags.allows_duplicate_labels
+        True
+        >>> df2 = df.set_flags(allows_duplicate_labels=False)
+        >>> df2.flags.allows_duplicate_labels
+        False
+        """
+        self._check_copy_deprecation(copy)
+        df = self.copy(deep=False)
+        if allows_duplicate_labels is not None:
+            df.flags["allows_duplicate_labels"] = allows_duplicate_labels
+        return df
+
+    @final
+    @classmethod
+    def _validate_dtype(cls, dtype) -> DtypeObj | None:
+        """validate the passed dtype"""
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+            # a compound dtype
+            if dtype.kind == "V":
+                raise NotImplementedError(
+                    "compound dtypes are not implemented "
+                    f"in the {cls.__name__} constructor"
+                )
+
+        return dtype
+
+    # ----------------------------------------------------------------------
+    # Construction
+
+    # error: Signature of "_constructor" incompatible with supertype "PandasObject"
+    @property
+    def _constructor(self) -> Callable[..., Self]:  # type: ignore[override]
+        """
+        Used when a manipulation result has the same dimensions as the
+        original.
+        """
+        raise AbstractMethodError(self)
+
+    # ----------------------------------------------------------------------
+    # Axis
+    _AXIS_ORDERS: list[Literal["index", "columns"]]
+    _AXIS_TO_AXIS_NUMBER: dict[Axis, AxisInt] = {0: 0, "index": 0, "rows": 0}
+    _info_axis_number: int
+    _info_axis_name: Literal["index", "columns"]
+    _AXIS_LEN: int
+
+    @final
+    def _construct_axes_dict(
+        self, axes: Sequence[Axis] | None = None, **kwargs: AxisInt
+    ) -> dict:
+        """Return an axes dictionary for myself."""
+        d = {a: self._get_axis(a) for a in (axes or self._AXIS_ORDERS)}
+        # error: Argument 1 to "update" of "MutableMapping" has incompatible type
+        # "Dict[str, Any]"; expected "SupportsKeysAndGetItem[Union[int, str], Any]"
+        d.update(kwargs)  # type: ignore[arg-type]
+        return d
+
+    @final
+    @classmethod
+    def _get_axis_number(cls, axis: Axis) -> AxisInt:
+        try:
+            return cls._AXIS_TO_AXIS_NUMBER[axis]
+        except KeyError as err:
+            raise ValueError(
+                f"No axis named {axis} for object type {cls.__name__}"
+            ) from err
+
+    @final
+    @classmethod
+    def _get_axis_name(cls, axis: Axis) -> Literal["index", "columns"]:
+        axis_number = cls._get_axis_number(axis)
+        return cls._AXIS_ORDERS[axis_number]
+
+    @final
+    def _get_axis(self, axis: Axis) -> Index:
+        axis_number = self._get_axis_number(axis)
+        assert axis_number in {0, 1}
+        return self.index if axis_number == 0 else self.columns
+
+    @final
+    @classmethod
+    def _get_block_manager_axis(cls, axis: Axis) -> AxisInt:
+        """Map the axis to the block_manager axis."""
+        axis = cls._get_axis_number(axis)
+        ndim = cls._AXIS_LEN
+        if ndim == 2:
+            # i.e. DataFrame
+            return 1 - axis
+        return axis
+
+    @final
+    def _get_axis_resolvers(self, axis: str) -> dict[str, Series | MultiIndex]:
+        # index or columns
+        axis_index = getattr(self, axis)
+        d = {}
+        prefix = axis[0]
+
+        for i, name in enumerate(axis_index.names):
+            if name is not None:
+                key = level = name
+            else:
+                # prefix with 'i' or 'c' depending on the input axis
+                # e.g., you must do ilevel_0 for the 0th level of an unnamed
+                # multiiindex
+                key = f"{prefix}level_{i}"
+                level = i
+
+            level_values = axis_index.get_level_values(level)
+            s = level_values.to_series()
+            s.index = axis_index
+            d[key] = s
+
+        # put the index/columns itself in the dict
+        if isinstance(axis_index, MultiIndex):
+            dindex = axis_index
+        else:
+            dindex = axis_index.to_series()
+
+        d[axis] = dindex
+        return d
+
+    @final
+    def _get_index_resolvers(self) -> dict[Hashable, Series | MultiIndex]:
+        from pandas.core.computation.parsing import clean_column_name
+
+        d: dict[str, Series | MultiIndex] = {}
+        for axis_name in self._AXIS_ORDERS:
+            d.update(self._get_axis_resolvers(axis_name))
+
+        return {clean_column_name(k): v for k, v in d.items() if not isinstance(k, int)}
+
+    @final
+    def _get_cleaned_column_resolvers(self) -> dict[Hashable, Series]:
+        """
+        Return the special character free column resolvers of a DataFrame.
+
+        Column names with special characters are 'cleaned up' so that they can
+        be referred to by backtick quoting.
+        Used in :meth:`DataFrame.eval`.
+        """
+        from pandas.core.computation.parsing import clean_column_name
+        from pandas.core.series import Series
+
+        if isinstance(self, ABCSeries):
+            return {clean_column_name(self.name): self}
+
+        dtypes = self.dtypes
+        return {
+            clean_column_name(k): Series(
+                v, copy=False, index=self.index, name=k, dtype=dtype
+            ).__finalize__(self)
+            for k, v, dtype in zip(self.columns, self._iter_column_arrays(), dtypes)
+            if not isinstance(k, int)
+        }
+
+    @final
+    @property
+    def _info_axis(self) -> Index:
+        return getattr(self, self._info_axis_name)
+
+    @property
+    def shape(self) -> tuple[int, ...]:
+        """
+        Return a tuple of axis dimensions
+        """
+        return tuple(len(self._get_axis(a)) for a in self._AXIS_ORDERS)
+
+    @property
+    def axes(self) -> list[Index]:
+        """
+        Return index label(s) of the internal NDFrame
+        """
+        # we do it this way because if we have reversed axes, then
+        # the block manager shows then reversed
+        return [self._get_axis(a) for a in self._AXIS_ORDERS]
+
+    @final
+    @property
+    def ndim(self) -> int:
+        """
+        Return an int representing the number of axes / array dimensions.
+
+        Return 1 if Series. Otherwise return 2 if DataFrame.
+
+        See Also
+        --------
+        numpy.ndarray.ndim : Number of array dimensions.
+
+        Examples
+        --------
+        >>> s = pd.Series({"a": 1, "b": 2, "c": 3})
+        >>> s.ndim
+        1
+
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df.ndim
+        2
+        """
+        return self._mgr.ndim
+
+    @final
+    @property
+    def size(self) -> int:
+        """
+        Return an int representing the number of elements in this object.
+
+        Return the number of rows if Series. Otherwise return the number of
+        rows times number of columns if DataFrame.
+
+        See Also
+        --------
+        numpy.ndarray.size : Number of elements in the array.
+
+        Examples
+        --------
+        >>> s = pd.Series({"a": 1, "b": 2, "c": 3})
+        >>> s.size
+        3
+
+        >>> df = pd.DataFrame({"col1": [1, 2], "col2": [3, 4]})
+        >>> df.size
+        4
+        """
+
+        return int(np.prod(self.shape))
+
+    def set_axis(
+        self,
+        labels,
+        *,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Self:
+        """
+        Assign desired index to given axis.
+
+        Indexes for%(extended_summary_sub)s row labels can be changed by assigning
+        a list-like or Index.
+
+        Parameters
+        ----------
+        labels : list-like, Index
+            The values for the new index.
+
+        axis : %(axes_single_arg)s, default 0
+            The axis to update. The value 0 identifies the rows. For `Series`
+            this parameter is unused and defaults to 0.
+
+        copy : bool, default False
+            Whether to make a copy of the underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        %(klass)s
+            An object of type %(klass)s.
+
+        See Also
+        --------
+        %(klass)s.rename_axis : Alter the name of the index%(see_also_sub)s.
+        """
+        self._check_copy_deprecation(copy)
+        return self._set_axis_nocheck(labels, axis, inplace=False)
+
+    @overload
+    def _set_axis_nocheck(
+        self, labels, axis: Axis, inplace: Literal[False]
+    ) -> Self: ...
+
+    @overload
+    def _set_axis_nocheck(self, labels, axis: Axis, inplace: Literal[True]) -> None: ...
+
+    @overload
+    def _set_axis_nocheck(self, labels, axis: Axis, inplace: bool) -> Self | None: ...
+
+    @final
+    def _set_axis_nocheck(self, labels, axis: Axis, inplace: bool) -> Self | None:
+        if inplace:
+            setattr(self, self._get_axis_name(axis), labels)
+            return None
+        obj = self.copy(deep=False)
+        setattr(obj, obj._get_axis_name(axis), labels)
+        return obj
+
+    @final
+    def _set_axis(self, axis: AxisInt, labels: AnyArrayLike | list) -> None:
+        """
+        This is called from the cython code when we set the `index` attribute
+        directly, e.g. `series.index = [1, 2, 3]`.
+        """
+        labels = ensure_index(labels)
+        self._mgr.set_axis(axis, labels)
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def droplevel(self, level: IndexLabel, axis: Axis = 0) -> Self:
+        """
+        Return {klass} with requested index / column level(s) removed.
+
+        Parameters
+        ----------
+        level : int, str, or list-like
+            If a string is given, must be the name of a level
+            If list-like, elements must be names or positional indexes
+            of levels.
+
+        axis : {{0 or 'index', 1 or 'columns'}}, default 0
+            Axis along which the level(s) is removed:
+
+            * 0 or 'index': remove level(s) in column.
+            * 1 or 'columns': remove level(s) in row.
+
+            For `Series` this parameter is unused and defaults to 0.
+
+        Returns
+        -------
+        {klass}
+            {klass} with requested index / column level(s) removed.
+
+        See Also
+        --------
+        DataFrame.replace : Replace values given in `to_replace` with `value`.
+        DataFrame.pivot : Return reshaped DataFrame organized by given
+            index / column values.
+
+        Examples
+        --------
+        >>> df = (
+        ...     pd.DataFrame([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12]])
+        ...     .set_index([0, 1])
+        ...     .rename_axis(["a", "b"])
+        ... )
+
+        >>> df.columns = pd.MultiIndex.from_tuples(
+        ...     [("c", "e"), ("d", "f")], names=["level_1", "level_2"]
+        ... )
+
+        >>> df
+        level_1   c   d
+        level_2   e   f
+        a b
+        1 2      3   4
+        5 6      7   8
+        9 10    11  12
+
+        >>> df.droplevel("a")
+        level_1   c   d
+        level_2   e   f
+        b
+        2        3   4
+        6        7   8
+        10      11  12
+
+        >>> df.droplevel("level_2", axis=1)
+        level_1   c   d
+        a b
+        1 2      3   4
+        5 6      7   8
+        9 10    11  12
+        """
+        labels = self._get_axis(axis)
+        new_labels = labels.droplevel(level)
+        return self.set_axis(new_labels, axis=axis)
+
+    def pop(self, item: Hashable) -> Series | Any:
+        result = self[item]
+        del self[item]
+
+        return result
+
+    @final
+    def squeeze(self, axis: Axis | None = None) -> Scalar | Series | DataFrame:
+        """
+        Squeeze 1 dimensional axis objects into scalars.
+
+        Series or DataFrames with a single element are squeezed to a scalar.
+        DataFrames with a single column or a single row are squeezed to a
+        Series. Otherwise the object is unchanged.
+
+        This method is most useful when you don't know if your
+        object is a Series or DataFrame, but you do know it has just a single
+        column. In that case you can safely call `squeeze` to ensure you have a
+        Series.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns', None}, default None
+            A specific axis to squeeze. By default, all length-1 axes are
+            squeezed. For `Series` this parameter is unused and defaults to `None`.
+
+        Returns
+        -------
+        DataFrame, Series, or scalar
+            The projection after squeezing `axis` or all the axes.
+
+        See Also
+        --------
+        Series.iloc : Integer-location based indexing for selecting scalars.
+        DataFrame.iloc : Integer-location based indexing for selecting Series.
+        Series.to_frame : Inverse of DataFrame.squeeze for a
+            single-column DataFrame.
+
+        Examples
+        --------
+        >>> primes = pd.Series([2, 3, 5, 7])
+
+        Slicing might produce a Series with a single value:
+
+        >>> even_primes = primes[primes % 2 == 0]
+        >>> even_primes
+        0    2
+        dtype: int64
+
+        >>> even_primes.squeeze()
+        2
+
+        Squeezing objects with more than one value in every axis does nothing:
+
+        >>> odd_primes = primes[primes % 2 == 1]
+        >>> odd_primes
+        1    3
+        2    5
+        3    7
+        dtype: int64
+
+        >>> odd_primes.squeeze()
+        1    3
+        2    5
+        3    7
+        dtype: int64
+
+        Squeezing is even more effective when used with DataFrames.
+
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+        >>> df
+           a  b
+        0  1  2
+        1  3  4
+
+        Slicing a single column will produce a DataFrame with the columns
+        having only one value:
+
+        >>> df_a = df[["a"]]
+        >>> df_a
+           a
+        0  1
+        1  3
+
+        So the columns can be squeezed down, resulting in a Series:
+
+        >>> df_a.squeeze("columns")
+        0    1
+        1    3
+        Name: a, dtype: int64
+
+        Slicing a single row from a single column will produce a single
+        scalar DataFrame:
+
+        >>> df_0a = df.loc[df.index < 1, ["a"]]
+        >>> df_0a
+           a
+        0  1
+
+        Squeezing the rows produces a single scalar Series:
+
+        >>> df_0a.squeeze("rows")
+        a    1
+        Name: 0, dtype: int64
+
+        Squeezing all axes will project directly into a scalar:
+
+        >>> df_0a.squeeze()
+        1
+        """
+        axes = range(self._AXIS_LEN) if axis is None else (self._get_axis_number(axis),)
+        result = self.iloc[
+            tuple(
+                0 if i in axes and len(a) == 1 else slice(None)
+                for i, a in enumerate(self.axes)
+            )
+        ]
+        if isinstance(result, NDFrame):
+            result = result.__finalize__(self, method="squeeze")
+        return result
+
+    # ----------------------------------------------------------------------
+    # Rename
+
+    @overload
+    def _rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        inplace: Literal[False] = ...,
+        level: Level | None = ...,
+        errors: str = ...,
+    ) -> Self: ...
+
+    @overload
+    def _rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        inplace: Literal[True],
+        level: Level | None = ...,
+        errors: str = ...,
+    ) -> None: ...
+
+    @overload
+    def _rename(
+        self,
+        mapper: Renamer | None = ...,
+        *,
+        index: Renamer | None = ...,
+        columns: Renamer | None = ...,
+        axis: Axis | None = ...,
+        inplace: bool,
+        level: Level | None = ...,
+        errors: str = ...,
+    ) -> Self | None: ...
+
+    @final
+    def _rename(
+        self,
+        mapper: Renamer | None = None,
+        *,
+        index: Renamer | None = None,
+        columns: Renamer | None = None,
+        axis: Axis | None = None,
+        inplace: bool = False,
+        level: Level | None = None,
+        errors: str = "ignore",
+    ) -> Self | None:
+        # called by Series.rename and DataFrame.rename
+
+        if mapper is None and index is None and columns is None:
+            raise TypeError("must pass an index to rename")
+
+        if index is not None or columns is not None:
+            if axis is not None:
+                raise TypeError(
+                    "Cannot specify both 'axis' and any of 'index' or 'columns'"
+                )
+            if mapper is not None:
+                raise TypeError(
+                    "Cannot specify both 'mapper' and any of 'index' or 'columns'"
+                )
+        else:
+            # use the mapper argument
+            if axis and self._get_axis_number(axis) == 1:
+                columns = mapper
+            else:
+                index = mapper
+
+        self._check_inplace_and_allows_duplicate_labels(inplace)
+        result = self if inplace else self.copy(deep=False)
+
+        for axis_no, replacements in enumerate((index, columns)):
+            if replacements is None:
+                continue
+
+            ax = self._get_axis(axis_no)
+            f = common.get_rename_function(replacements)
+
+            if level is not None:
+                level = ax._get_level_number(level)
+
+            # GH 13473
+            if not callable(replacements):
+                if ax._is_multi and level is not None:
+                    indexer = ax.get_level_values(level).get_indexer_for(replacements)
+                else:
+                    indexer = ax.get_indexer_for(replacements)
+
+                if errors == "raise" and len(indexer[indexer == -1]):
+                    missing_labels = [
+                        label
+                        for index, label in enumerate(replacements)
+                        if indexer[index] == -1
+                    ]
+                    raise KeyError(f"{missing_labels} not found in axis")
+
+            new_index = ax._transform_index(f, level=level)
+            result._set_axis_nocheck(new_index, axis=axis_no, inplace=True)
+
+        if inplace:
+            self._update_inplace(result)
+            return None
+        else:
+            return result.__finalize__(self, method="rename")
+
+    @overload
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        columns=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: Literal[False] = ...,
+    ) -> Self: ...
+
+    @overload
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        columns=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: Literal[True],
+    ) -> None: ...
+
+    @overload
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        columns=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = ...,
+    ) -> Self | None: ...
+
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = lib.no_default,
+        *,
+        index=lib.no_default,
+        columns=lib.no_default,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = False,
+    ) -> Self | None:
+        """
+        Set the name of the axis for the index or columns.
+
+        Parameters
+        ----------
+        mapper : scalar, list-like, optional
+            Value to set the axis name attribute.
+
+            Use either ``mapper`` and ``axis`` to
+            specify the axis to target with ``mapper``, or ``index``
+            and/or ``columns``.
+        index : scalar, list-like, dict-like or function, optional
+            A scalar, list-like, dict-like or functions transformations to
+            apply to that axis' values.
+        columns : scalar, list-like, dict-like or function, optional
+            A scalar, list-like, dict-like or functions transformations to
+            apply to that axis' values.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to rename.
+        copy : bool, default False
+            Also copy underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        inplace : bool, default False
+            Modifies the object directly, instead of creating a new Series
+            or DataFrame.
+
+        Returns
+        -------
+        DataFrame, or None
+            The same type as the caller or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.rename : Alter Series index labels or name.
+        DataFrame.rename : Alter DataFrame index labels or name.
+        Index.rename : Set new names on index.
+
+        Notes
+        -----
+        ``DataFrame.rename_axis`` supports two calling conventions
+
+        * ``(index=index_mapper, columns=columns_mapper, ...)``
+        * ``(mapper, axis={'index', 'columns'}, ...)``
+
+        The first calling convention will only modify the names of
+        the index and/or the names of the Index object that is the columns.
+        In this case, the parameter ``copy`` is ignored.
+
+        The second calling convention will modify the names of the
+        corresponding index if mapper is a list or a scalar.
+        However, if mapper is dict-like or a function, it will use the
+        deprecated behavior of modifying the axis *labels*.
+
+        We *highly* recommend using keyword arguments to clarify your
+        intent.
+
+        Examples
+        --------
+        **DataFrame**
+
+        >>> df = pd.DataFrame(
+        ...     {"num_legs": [4, 4, 2], "num_arms": [0, 0, 2]}, ["dog", "cat", "monkey"]
+        ... )
+        >>> df
+                num_legs  num_arms
+        dog            4         0
+        cat            4         0
+        monkey         2         2
+        >>> df = df.rename_axis("animal")
+        >>> df
+                num_legs  num_arms
+        animal
+        dog            4         0
+        cat            4         0
+        monkey         2         2
+        >>> df = df.rename_axis("limbs", axis="columns")
+        >>> df
+        limbs   num_legs  num_arms
+        animal
+        dog            4         0
+        cat            4         0
+        monkey         2         2
+
+        **MultiIndex**
+
+        >>> df.index = pd.MultiIndex.from_product(
+        ...     [["mammal"], ["dog", "cat", "monkey"]], names=["type", "name"]
+        ... )
+        >>> df
+        limbs          num_legs  num_arms
+        type   name
+        mammal dog            4         0
+               cat            4         0
+               monkey         2         2
+
+        >>> df.rename_axis(index={"type": "class"})
+        limbs          num_legs  num_arms
+        class  name
+        mammal dog            4         0
+               cat            4         0
+               monkey         2         2
+
+        >>> df.rename_axis(columns=str.upper)
+        LIMBS          num_legs  num_arms
+        type   name
+        mammal dog            4         0
+               cat            4         0
+               monkey         2         2
+        """
+        self._check_copy_deprecation(copy)
+        axes = {"index": index, "columns": columns}
+
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if mapper is not lib.no_default:
+            # Use v0.23 behavior if a scalar or list
+            non_mapper = is_scalar(mapper) or (
+                is_list_like(mapper) and not is_dict_like(mapper)
+            )
+            if non_mapper:
+                return self._set_axis_name(mapper, axis=axis, inplace=inplace)
+            else:
+                raise ValueError("Use `.rename` to alter labels with a mapper.")
+        else:
+            # Use new behavior.  Means that index and/or columns
+            # is specified
+            result = self if inplace else self.copy(deep=False)
+
+            for axis in range(self._AXIS_LEN):
+                v = axes.get(self._get_axis_name(axis))
+                if v is lib.no_default:
+                    continue
+                non_mapper = is_scalar(v) or (is_list_like(v) and not is_dict_like(v))
+                if non_mapper:
+                    newnames = v
+                else:
+                    f = common.get_rename_function(v)
+                    curnames = self._get_axis(axis).names
+                    newnames = [f(name) for name in curnames]
+                result._set_axis_name(newnames, axis=axis, inplace=True)
+            if not inplace:
+                return result
+            return None
+
+    @overload
+    def _set_axis_name(
+        self, name, axis: Axis = ..., *, inplace: Literal[False] = ...
+    ) -> Self: ...
+
+    @overload
+    def _set_axis_name(
+        self, name, axis: Axis = ..., *, inplace: Literal[True]
+    ) -> None: ...
+
+    @overload
+    def _set_axis_name(
+        self, name, axis: Axis = ..., *, inplace: bool
+    ) -> Self | None: ...
+
+    @final
+    def _set_axis_name(
+        self, name, axis: Axis = 0, *, inplace: bool = False
+    ) -> Self | None:
+        """
+        Set the name(s) of the axis.
+
+        Parameters
+        ----------
+        name : str or list of str
+            Name(s) to set.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            The axis to set the label. The value 0 or 'index' specifies index,
+            and the value 1 or 'columns' specifies columns.
+        inplace : bool, default False
+            If `True`, do operation inplace and return None.
+
+        Returns
+        -------
+        Series, DataFrame, or None
+            The same type as the caller or `None` if `inplace` is `True`.
+
+        See Also
+        --------
+        DataFrame.rename : Alter the axis labels of :class:`DataFrame`.
+        Series.rename : Alter the index labels or set the index name
+            of :class:`Series`.
+        Index.rename : Set the name of :class:`Index` or :class:`MultiIndex`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"num_legs": [4, 4, 2]}, ["dog", "cat", "monkey"])
+        >>> df
+                num_legs
+        dog            4
+        cat            4
+        monkey         2
+        >>> df._set_axis_name("animal")
+                num_legs
+        animal
+        dog            4
+        cat            4
+        monkey         2
+        >>> df.index = pd.MultiIndex.from_product(
+        ...     [["mammal"], ["dog", "cat", "monkey"]]
+        ... )
+        >>> df._set_axis_name(["type", "name"])
+                       num_legs
+        type   name
+        mammal dog        4
+               cat        4
+               monkey     2
+        """
+        axis = self._get_axis_number(axis)
+        idx = self._get_axis(axis).set_names(name)
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        renamed = self if inplace else self.copy(deep=False)
+        if axis == 0:
+            renamed.index = idx
+        else:
+            renamed.columns = idx
+
+        if not inplace:
+            return renamed
+        return None
+
+    # ----------------------------------------------------------------------
+    # Comparison Methods
+
+    @final
+    def _indexed_same(self, other) -> bool:
+        return all(
+            self._get_axis(a).equals(other._get_axis(a)) for a in self._AXIS_ORDERS
+        )
+
+    @final
+    def equals(self, other: object) -> bool:
+        """
+        Test whether two objects contain the same elements.
+
+        This function allows two Series or DataFrames to be compared against
+        each other to see if they have the same shape and elements. NaNs in
+        the same location are considered equal.
+
+        The row/column index do not need to have the same type, as long
+        as the values are considered equal. Corresponding columns and
+        index must be of the same dtype.
+
+        Parameters
+        ----------
+        other : Series or DataFrame
+            The other Series or DataFrame to be compared with the first.
+
+        Returns
+        -------
+        bool
+            True if all elements are the same in both objects, False
+            otherwise.
+
+        See Also
+        --------
+        Series.eq : Compare two Series objects of the same length
+            and return a Series where each element is True if the element
+            in each Series is equal, False otherwise.
+        DataFrame.eq : Compare two DataFrame objects of the same shape and
+            return a DataFrame where each element is True if the respective
+            element in each DataFrame is equal, False otherwise.
+        testing.assert_series_equal : Raises an AssertionError if left and
+            right are not equal. Provides an easy interface to ignore
+            inequality in dtypes, indexes and precision among others.
+        testing.assert_frame_equal : Like assert_series_equal, but targets
+            DataFrames.
+        numpy.array_equal : Return True if two arrays have the same shape
+            and elements, False otherwise.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({1: [10], 2: [20]})
+        >>> df
+            1   2
+        0  10  20
+
+        DataFrames df and exactly_equal have the same types and values for
+        their elements and column labels, which will return True.
+
+        >>> exactly_equal = pd.DataFrame({1: [10], 2: [20]})
+        >>> exactly_equal
+            1   2
+        0  10  20
+        >>> df.equals(exactly_equal)
+        True
+
+        DataFrames df and different_column_type have the same element
+        types and values, but have different types for the column labels,
+        which will still return True.
+
+        >>> different_column_type = pd.DataFrame({1.0: [10], 2.0: [20]})
+        >>> different_column_type
+           1.0  2.0
+        0   10   20
+        >>> df.equals(different_column_type)
+        True
+
+        DataFrames df and different_data_type have different types for the
+        same values for their elements, and will return False even though
+        their column labels are the same values and types.
+
+        >>> different_data_type = pd.DataFrame({1: [10.0], 2: [20.0]})
+        >>> different_data_type
+              1     2
+        0  10.0  20.0
+        >>> df.equals(different_data_type)
+        False
+        """
+        if not (isinstance(other, type(self)) or isinstance(self, type(other))):
+            return False
+        other = cast(NDFrame, other)
+        return self._mgr.equals(other._mgr)
+
+    # -------------------------------------------------------------------------
+    # Unary Methods
+
+    @final
+    def __neg__(self) -> Self:
+        def blk_func(values: ArrayLike):
+            if is_bool_dtype(values.dtype):
+                # error: Argument 1 to "inv" has incompatible type "Union
+                # [ExtensionArray, ndarray[Any, Any]]"; expected
+                # "_SupportsInversion[ndarray[Any, dtype[bool_]]]"
+                return operator.inv(values)  # type: ignore[arg-type]
+            else:
+                # error: Argument 1 to "neg" has incompatible type "Union
+                # [ExtensionArray, ndarray[Any, Any]]"; expected
+                # "_SupportsNeg[ndarray[Any, dtype[Any]]]"
+                return operator.neg(values)  # type: ignore[arg-type]
+
+        new_data = self._mgr.apply(blk_func)
+        res = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        return res.__finalize__(self, method="__neg__")
+
+    @final
+    def __pos__(self) -> Self:
+        def blk_func(values: ArrayLike):
+            if is_bool_dtype(values.dtype):
+                return values.copy()
+            else:
+                # error: Argument 1 to "pos" has incompatible type "Union
+                # [ExtensionArray, ndarray[Any, Any]]"; expected
+                # "_SupportsPos[ndarray[Any, dtype[Any]]]"
+                return operator.pos(values)  # type: ignore[arg-type]
+
+        new_data = self._mgr.apply(blk_func)
+        res = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        return res.__finalize__(self, method="__pos__")
+
+    @final
+    def __invert__(self) -> Self:
+        if not self.size:
+            # inv fails with 0 len
+            return self.copy(deep=False)
+
+        new_data = self._mgr.apply(operator.invert)
+        res = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        return res.__finalize__(self, method="__invert__")
+
+    @final
+    def __bool__(self) -> NoReturn:
+        raise ValueError(
+            f"The truth value of a {type(self).__name__} is ambiguous. "
+            "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
+        )
+
+    @final
+    def abs(self) -> Self:
+        """
+        Return a Series/DataFrame with absolute numeric value of each element.
+
+        This function only applies to elements that are all numeric.
+
+        Returns
+        -------
+        abs
+            Series/DataFrame containing the absolute value of each element.
+
+        See Also
+        --------
+        numpy.absolute : Calculate the absolute value element-wise.
+
+        Notes
+        -----
+        For ``complex`` inputs, ``1.2 + 1j``, the absolute value is
+        :math:`\\sqrt{ a^2 + b^2 }`.
+
+        Examples
+        --------
+        Absolute numeric values in a Series.
+
+        >>> s = pd.Series([-1.10, 2, -3.33, 4])
+        >>> s.abs()
+        0    1.10
+        1    2.00
+        2    3.33
+        3    4.00
+        dtype: float64
+
+        Absolute numeric values in a Series with complex numbers.
+
+        >>> s = pd.Series([1.2 + 1j])
+        >>> s.abs()
+        0    1.56205
+        dtype: float64
+
+        Absolute numeric values in a Series with a Timedelta element.
+
+        >>> s = pd.Series([pd.Timedelta("1 days")])
+        >>> s.abs()
+        0   1 days
+        dtype: timedelta64[ns]
+
+        Select rows with data closest to certain value using argsort (from
+        `StackOverflow <https://stackoverflow.com/a/17758115>`__).
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [4, 5, 6, 7], "b": [10, 20, 30, 40], "c": [100, 50, -30, -50]}
+        ... )
+        >>> df
+             a    b    c
+        0    4   10  100
+        1    5   20   50
+        2    6   30  -30
+        3    7   40  -50
+        >>> df.loc[(df.c - 43).abs().argsort()]
+             a    b    c
+        1    5   20   50
+        0    4   10  100
+        2    6   30  -30
+        3    7   40  -50
+        """
+        res_mgr = self._mgr.apply(np.abs)
+        return self._constructor_from_mgr(res_mgr, axes=res_mgr.axes).__finalize__(
+            self, name="abs"
+        )
+
+    @final
+    def __abs__(self) -> Self:
+        return self.abs()
+
+    @final
+    def __round__(self, decimals: int = 0) -> Self:
+        return self.round(decimals).__finalize__(self, method="__round__")
+
+    # -------------------------------------------------------------------------
+    # Label or Level Combination Helpers
+    #
+    # A collection of helper methods for DataFrame/Series operations that
+    # accept a combination of column/index labels and levels.  All such
+    # operations should utilize/extend these methods when possible so that we
+    # have consistent precedence and validation logic throughout the library.
+
+    @final
+    def _is_level_reference(self, key: Level, axis: Axis = 0) -> bool:
+        """
+        Test whether a key is a level reference for a given axis.
+
+        To be considered a level reference, `key` must be a string that:
+          - (axis=0): Matches the name of an index level and does NOT match
+            a column label.
+          - (axis=1): Matches the name of a column level and does NOT match
+            an index label.
+
+        Parameters
+        ----------
+        key : Hashable
+            Potential level name for the given axis
+        axis : int, default 0
+            Axis that levels are associated with (0 for index, 1 for columns)
+
+        Returns
+        -------
+        is_level : bool
+        """
+        axis_int = self._get_axis_number(axis)
+
+        return (
+            key is not None
+            and is_hashable(key)
+            and key in self.axes[axis_int].names
+            and not self._is_label_reference(key, axis=axis_int)
+        )
+
+    @final
+    def _is_label_reference(self, key: Level, axis: Axis = 0) -> bool:
+        """
+        Test whether a key is a label reference for a given axis.
+
+        To be considered a label reference, `key` must be a string that:
+          - (axis=0): Matches a column label
+          - (axis=1): Matches an index label
+
+        Parameters
+        ----------
+        key : Hashable
+            Potential label name, i.e. Index entry.
+        axis : int, default 0
+            Axis perpendicular to the axis that labels are associated with
+            (0 means search for column labels, 1 means search for index labels)
+
+        Returns
+        -------
+        is_label: bool
+        """
+        axis_int = self._get_axis_number(axis)
+        other_axes = (ax for ax in range(self._AXIS_LEN) if ax != axis_int)
+
+        return (
+            key is not None
+            and is_hashable(key)
+            and any(key in self.axes[ax] for ax in other_axes)
+        )
+
+    @final
+    def _is_label_or_level_reference(self, key: Level, axis: AxisInt = 0) -> bool:
+        """
+        Test whether a key is a label or level reference for a given axis.
+
+        To be considered either a label or a level reference, `key` must be a
+        string that:
+          - (axis=0): Matches a column label or an index level
+          - (axis=1): Matches an index label or a column level
+
+        Parameters
+        ----------
+        key : Hashable
+            Potential label or level name
+        axis : int, default 0
+            Axis that levels are associated with (0 for index, 1 for columns)
+
+        Returns
+        -------
+        bool
+        """
+        return self._is_level_reference(key, axis=axis) or self._is_label_reference(
+            key, axis=axis
+        )
+
+    @final
+    def _check_label_or_level_ambiguity(self, key: Level, axis: Axis = 0) -> None:
+        """
+        Check whether `key` is ambiguous.
+
+        By ambiguous, we mean that it matches both a level of the input
+        `axis` and a label of the other axis.
+
+        Parameters
+        ----------
+        key : Hashable
+            Label or level name.
+        axis : int, default 0
+            Axis that levels are associated with (0 for index, 1 for columns).
+
+        Raises
+        ------
+        ValueError: `key` is ambiguous
+        """
+
+        axis_int = self._get_axis_number(axis)
+        other_axes = (ax for ax in range(self._AXIS_LEN) if ax != axis_int)
+
+        if (
+            key is not None
+            and is_hashable(key)
+            and key in self.axes[axis_int].names
+            and any(key in self.axes[ax] for ax in other_axes)
+        ):
+            # Build an informative and grammatical warning
+            level_article, level_type = (
+                ("an", "index") if axis_int == 0 else ("a", "column")
+            )
+
+            label_article, label_type = (
+                ("a", "column") if axis_int == 0 else ("an", "index")
+            )
+
+            msg = (
+                f"'{key}' is both {level_article} {level_type} level and "
+                f"{label_article} {label_type} label, which is ambiguous."
+            )
+            raise ValueError(msg)
+
+    @final
+    def _get_label_or_level_values(self, key: Level, axis: AxisInt = 0) -> ArrayLike:
+        """
+        Return a 1-D array of values associated with `key`, a label or level
+        from the given `axis`.
+
+        Retrieval logic:
+          - (axis=0): Return column values if `key` matches a column label.
+            Otherwise return index level values if `key` matches an index
+            level.
+          - (axis=1): Return row values if `key` matches an index label.
+            Otherwise return column level values if 'key' matches a column
+            level
+
+        Parameters
+        ----------
+        key : Hashable
+            Label or level name.
+        axis : int, default 0
+            Axis that levels are associated with (0 for index, 1 for columns)
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray
+
+        Raises
+        ------
+        KeyError
+            if `key` matches neither a label nor a level
+        ValueError
+            if `key` matches multiple labels
+        """
+        axis = self._get_axis_number(axis)
+        first_other_axes = next(
+            (ax for ax in range(self._AXIS_LEN) if ax != axis), None
+        )
+
+        if self._is_label_reference(key, axis=axis):
+            self._check_label_or_level_ambiguity(key, axis=axis)
+            if first_other_axes is None:
+                raise ValueError("axis matched all axes")
+            values = self.xs(key, axis=first_other_axes)._values
+        elif self._is_level_reference(key, axis=axis):
+            values = self.axes[axis].get_level_values(key)._values
+        else:
+            raise KeyError(key)
+
+        # Check for duplicates
+        if values.ndim > 1:
+            if first_other_axes is not None and isinstance(
+                self._get_axis(first_other_axes), MultiIndex
+            ):
+                multi_message = (
+                    "\n"
+                    "For a multi-index, the label must be a "
+                    "tuple with elements corresponding to each level."
+                )
+            else:
+                multi_message = ""
+
+            label_axis_name = "column" if axis == 0 else "index"
+            raise ValueError(
+                f"The {label_axis_name} label '{key}' is not unique.{multi_message}"
+            )
+
+        return values
+
+    @final
+    def _drop_labels_or_levels(self, keys, axis: AxisInt = 0):
+        """
+        Drop labels and/or levels for the given `axis`.
+
+        For each key in `keys`:
+          - (axis=0): If key matches a column label then drop the column.
+            Otherwise if key matches an index level then drop the level.
+          - (axis=1): If key matches an index label then drop the row.
+            Otherwise if key matches a column level then drop the level.
+
+        Parameters
+        ----------
+        keys : str or list of str
+            labels or levels to drop
+        axis : int, default 0
+            Axis that levels are associated with (0 for index, 1 for columns)
+
+        Returns
+        -------
+        dropped: DataFrame
+
+        Raises
+        ------
+        ValueError
+            if any `keys` match neither a label nor a level
+        """
+        axis = self._get_axis_number(axis)
+
+        # Validate keys
+        keys = common.maybe_make_list(keys)
+        invalid_keys = [
+            k for k in keys if not self._is_label_or_level_reference(k, axis=axis)
+        ]
+
+        if invalid_keys:
+            raise ValueError(
+                "The following keys are not valid labels or "
+                f"levels for axis {axis}: {invalid_keys}"
+            )
+
+        # Compute levels and labels to drop
+        levels_to_drop = [k for k in keys if self._is_level_reference(k, axis=axis)]
+
+        labels_to_drop = [k for k in keys if not self._is_level_reference(k, axis=axis)]
+
+        # Perform copy upfront and then use inplace operations below.
+        # This ensures that we always perform exactly one copy.
+        # ``copy`` and/or ``inplace`` options could be added in the future.
+        dropped = self.copy(deep=False)
+
+        if axis == 0:
+            # Handle dropping index levels
+            if levels_to_drop:
+                dropped.reset_index(levels_to_drop, drop=True, inplace=True)
+
+            # Handle dropping columns labels
+            if labels_to_drop:
+                dropped.drop(labels_to_drop, axis=1, inplace=True)
+        else:
+            # Handle dropping column levels
+            if levels_to_drop:
+                if isinstance(dropped.columns, MultiIndex):
+                    # Drop the specified levels from the MultiIndex
+                    dropped.columns = dropped.columns.droplevel(levels_to_drop)
+                else:
+                    # Drop the last level of Index by replacing with
+                    # a RangeIndex
+                    dropped.columns = default_index(dropped.columns.size)
+
+            # Handle dropping index labels
+            if labels_to_drop:
+                dropped.drop(labels_to_drop, axis=0, inplace=True)
+
+        return dropped
+
+    # ----------------------------------------------------------------------
+    # Iteration
+
+    # https://github.com/python/typeshed/issues/2148#issuecomment-520783318
+    # Incompatible types in assignment (expression has type "None", base class
+    # "object" defined the type as "Callable[[object], int]")
+    __hash__: ClassVar[None]  # type: ignore[assignment]
+
+    def __iter__(self) -> Iterator:
+        """
+        Iterate over info axis.
+
+        Returns
+        -------
+        iterator
+            Info axis as iterator.
+
+        See Also
+        --------
+        DataFrame.items : Iterate over (column name, Series) pairs.
+        DataFrame.itertuples : Iterate over DataFrame rows as namedtuples.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        >>> for x in df:
+        ...     print(x)
+        A
+        B
+        """
+        return iter(self._info_axis)
+
+    # can we get a better explanation of this?
+    def keys(self) -> Index:
+        """
+        Get the 'info axis' (see Indexing for more).
+
+        This is index for Series, columns for DataFrame.
+
+        Returns
+        -------
+        Index
+            Info axis.
+
+        See Also
+        --------
+        DataFrame.index : The index (row labels) of the DataFrame.
+        DataFrame.columns: The column labels of the DataFrame.
+
+        Examples
+        --------
+        >>> d = pd.DataFrame(
+        ...     data={"A": [1, 2, 3], "B": [0, 4, 8]}, index=["a", "b", "c"]
+        ... )
+        >>> d
+           A  B
+        a  1  0
+        b  2  4
+        c  3  8
+        >>> d.keys()
+        Index(['A', 'B'], dtype='object')
+        """
+        return self._info_axis
+
+    def items(self):
+        """
+        Iterate over (label, values) on info axis
+
+        This is index for Series and columns for DataFrame.
+
+        Returns
+        -------
+        Generator
+        """
+        for h in self._info_axis:
+            yield h, self[h]
+
+    def __len__(self) -> int:
+        """Returns length of info axis"""
+        return len(self._info_axis)
+
+    @final
+    def __contains__(self, key) -> bool:
+        """True if the key is in the info axis"""
+        return key in self._info_axis
+
+    @property
+    def empty(self) -> bool:
+        """
+        Indicator whether Series/DataFrame is empty.
+
+        True if Series/DataFrame is entirely empty (no items), meaning any of the
+        axes are of length 0.
+
+        Returns
+        -------
+        bool
+            If Series/DataFrame is empty, return True, if not return False.
+
+        See Also
+        --------
+        Series.dropna : Return series without null values.
+        DataFrame.dropna : Return DataFrame with labels on given axis omitted
+            where (all or any) data are missing.
+
+        Notes
+        -----
+        If Series/DataFrame contains only NaNs, it is still not considered empty. See
+        the example below.
+
+        Examples
+        --------
+        An example of an actual empty DataFrame. Notice the index is empty:
+
+        >>> df_empty = pd.DataFrame({"A": []})
+        >>> df_empty
+        Empty DataFrame
+        Columns: [A]
+        Index: []
+        >>> df_empty.empty
+        True
+
+        If we only have NaNs in our DataFrame, it is not considered empty! We
+        will need to drop the NaNs to make the DataFrame empty:
+
+        >>> df = pd.DataFrame({"A": [np.nan]})
+        >>> df
+            A
+        0 NaN
+        >>> df.empty
+        False
+        >>> df.dropna().empty
+        True
+
+        >>> ser_empty = pd.Series({"A": []})
+        >>> ser_empty
+        A    []
+        dtype: object
+        >>> ser_empty.empty
+        False
+        >>> ser_empty = pd.Series()
+        >>> ser_empty.empty
+        True
+        """
+        return any(len(self._get_axis(a)) == 0 for a in self._AXIS_ORDERS)
+
+    # ----------------------------------------------------------------------
+    # Array Interface
+
+    # This is also set in IndexOpsMixin
+    # GH#23114 Ensure ndarray.__op__(DataFrame) returns NotImplemented
+    __array_priority__: int = 1000
+
+    def __array__(
+        self, dtype: npt.DTypeLike | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        if copy is False and not self._mgr.is_single_block and not self.empty:
+            # check this manually, otherwise ._values will already return a copy
+            # and np.array(values, copy=False) will not raise an error
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+        values = self._values
+        if copy is None:
+            # Note: branch avoids `copy=None` for NumPy 1.x support
+            arr = np.asarray(values, dtype=dtype)
+        else:
+            arr = np.array(values, dtype=dtype, copy=copy)
+
+        if (
+            copy is not True
+            and astype_is_view(values.dtype, arr.dtype)
+            and self._mgr.is_single_block
+        ):
+            # Check if both conversions can be done without a copy
+            if astype_is_view(self.dtypes.iloc[0], values.dtype) and astype_is_view(
+                values.dtype, arr.dtype
+            ):
+                arr = arr.view()
+                arr.flags.writeable = False
+        return arr
+
+    @final
+    def __array_ufunc__(
+        self, ufunc: np.ufunc, method: str, *inputs: Any, **kwargs: Any
+    ):
+        return arraylike.array_ufunc(self, ufunc, method, *inputs, **kwargs)
+
+    # ----------------------------------------------------------------------
+    # Picklability
+
+    @final
+    def __getstate__(self) -> dict[str, Any]:
+        meta = {k: getattr(self, k, None) for k in self._metadata}
+        return {
+            "_mgr": self._mgr,
+            "_typ": self._typ,
+            "_metadata": self._metadata,
+            "attrs": self.attrs,
+            "_flags": {k: self.flags[k] for k in self.flags._keys},
+            **meta,
+        }
+
+    @final
+    def __setstate__(self, state) -> None:
+        if isinstance(state, BlockManager):
+            self._mgr = state
+        elif isinstance(state, dict):
+            if "_data" in state and "_mgr" not in state:
+                # compat for older pickles
+                state["_mgr"] = state.pop("_data")
+            typ = state.get("_typ")
+            if typ is not None:
+                attrs = state.get("_attrs", {})
+                if attrs is None:  # should not happen, but better be on the safe side
+                    attrs = {}
+                object.__setattr__(self, "_attrs", attrs)
+                flags = state.get("_flags", {"allows_duplicate_labels": True})
+                object.__setattr__(self, "_flags", Flags(self, **flags))
+
+                # set in the order of internal names
+                # to avoid definitional recursion
+                # e.g. say fill_value needing _mgr to be
+                # defined
+                meta = set(self._internal_names + self._metadata)
+                for k in meta:
+                    if k in state and k != "_flags":
+                        v = state[k]
+                        object.__setattr__(self, k, v)
+
+                for k, v in state.items():
+                    if k not in meta:
+                        object.__setattr__(self, k, v)
+
+            else:
+                raise NotImplementedError("Pre-0.12 pickles are no longer supported")
+        elif len(state) == 2:
+            raise NotImplementedError("Pre-0.12 pickles are no longer supported")
+
+    # ----------------------------------------------------------------------
+    # Rendering Methods
+
+    def __repr__(self) -> str:
+        # string representation based upon iterating over self
+        # (since, by definition, `PandasContainers` are iterable)
+        prepr = f"[{','.join(map(pprint_thing, self))}]"
+        return f"{type(self).__name__}({prepr})"
+
+    @final
+    def _repr_latex_(self):
+        """
+        Returns a LaTeX representation for a particular object.
+        Mainly for use with nbconvert (jupyter notebook conversion to pdf).
+        """
+        if config.get_option("styler.render.repr") == "latex":
+            return self.to_latex()
+        else:
+            return None
+
+    @final
+    def _repr_data_resource_(self):
+        """
+        Not a real Jupyter special repr method, but we use the same
+        naming convention.
+        """
+        if config.get_option("display.html.table_schema"):
+            data = self.head(config.get_option("display.max_rows"))
+
+            as_json = data.to_json(orient="table")
+            as_json = cast(str, as_json)
+            return loads(as_json, object_pairs_hook=collections.OrderedDict)
+
+    # ----------------------------------------------------------------------
+    # I/O Methods
+
+    @final
+    @doc(
+        klass="object",
+        storage_options=_shared_docs["storage_options"],
+        storage_options_versionadded="1.2.0",
+        encoding_parameter="",
+        verbose_parameter="",
+        extra_parameters=textwrap.dedent(
+            """\
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        ),
+    )
+    def to_excel(
+        self,
+        excel_writer: FilePath | WriteExcelBuffer | ExcelWriter,
+        *,
+        sheet_name: str = "Sheet1",
+        na_rep: str = "",
+        float_format: str | None = None,
+        columns: Sequence[Hashable] | None = None,
+        header: Sequence[Hashable] | bool = True,
+        index: bool = True,
+        index_label: IndexLabel | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        engine: Literal["openpyxl", "xlsxwriter"] | None = None,
+        merge_cells: bool = True,
+        inf_rep: str = "inf",
+        freeze_panes: tuple[int, int] | None = None,
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+    ) -> None:
+        """
+        Write {klass} to an Excel sheet.
+
+        To write a single {klass} to an Excel .xlsx file it is only necessary to
+        specify a target file name. To write to multiple sheets it is necessary to
+        create an `ExcelWriter` object with a target file name, and specify a sheet
+        in the file to write to.
+
+        Multiple sheets may be written to by specifying unique `sheet_name`.
+        With all data written to the file it is necessary to save the changes.
+        Note that creating an `ExcelWriter` object with a file name that already
+        exists will result in the contents of the existing file being erased.
+
+        Parameters
+        ----------
+        excel_writer : path-like, file-like, or ExcelWriter object
+            File path or existing ExcelWriter.
+        sheet_name : str, default 'Sheet1'
+            Name of sheet which will contain DataFrame.
+        na_rep : str, default ''
+            Missing data representation.
+        float_format : str, optional
+            Format string for floating point numbers. For example
+            ``float_format="%.2f"`` will format 0.1234 to 0.12.
+        columns : sequence or list of str, optional
+            Columns to write.
+        header : bool or list of str, default True
+            Write out the column names. If a list of string is given it is
+            assumed to be aliases for the column names.
+        index : bool, default True
+            Write row names (index).
+        index_label : str or sequence, optional
+            Column label for index column(s) if desired. If not specified, and
+            `header` and `index` are True, then the index names are used. A
+            sequence should be given if the DataFrame uses MultiIndex.
+        startrow : int, default 0
+            Upper left cell row to dump data frame.
+        startcol : int, default 0
+            Upper left cell column to dump data frame.
+        engine : str, optional
+            Write engine to use, 'openpyxl' or 'xlsxwriter'. You can also set this
+            via the options ``io.excel.xlsx.writer`` or
+            ``io.excel.xlsm.writer``.
+
+        merge_cells : bool or 'columns', default False
+            If True, write MultiIndex index and columns as merged cells.
+            If 'columns', merge MultiIndex column cells only.
+        {encoding_parameter}
+        inf_rep : str, default 'inf'
+            Representation for infinity (there is no native representation for
+            infinity in Excel).
+        {verbose_parameter}
+        freeze_panes : tuple of int (length 2), optional
+            Specifies the one-based bottommost row and rightmost column that
+            is to be frozen.
+        {storage_options}
+
+            .. versionadded:: {storage_options_versionadded}
+        {extra_parameters}
+        See Also
+        --------
+        to_csv : Write DataFrame to a comma-separated values (csv) file.
+        ExcelWriter : Class for writing DataFrame objects into excel sheets.
+        read_excel : Read an Excel file into a pandas DataFrame.
+        read_csv : Read a comma-separated values (csv) file into DataFrame.
+        io.formats.style.Styler.to_excel : Add styles to Excel sheet.
+
+        Notes
+        -----
+        For compatibility with :meth:`~DataFrame.to_csv`,
+        to_excel serializes lists and dicts to strings before writing.
+
+        Once a workbook has been saved it is not possible to write further
+        data without rewriting the whole workbook.
+
+        pandas will check the number of rows, columns,
+        and cell character count does not exceed Excel's limitations.
+        All other limitations must be checked by the user.
+
+        Examples
+        --------
+
+        Create, write to and save a workbook:
+
+        >>> df1 = pd.DataFrame(
+        ...     [["a", "b"], ["c", "d"]],
+        ...     index=["row 1", "row 2"],
+        ...     columns=["col 1", "col 2"],
+        ... )
+        >>> df1.to_excel("output.xlsx")  # doctest: +SKIP
+
+        To specify the sheet name:
+
+        >>> df1.to_excel("output.xlsx", sheet_name="Sheet_name_1")  # doctest: +SKIP
+
+        If you wish to write to more than one sheet in the workbook, it is
+        necessary to specify an ExcelWriter object:
+
+        >>> df2 = df1.copy()
+        >>> with pd.ExcelWriter("output.xlsx") as writer:  # doctest: +SKIP
+        ...     df1.to_excel(writer, sheet_name="Sheet_name_1")
+        ...     df2.to_excel(writer, sheet_name="Sheet_name_2")
+
+        ExcelWriter can also be used to append to an existing Excel file:
+
+        >>> with pd.ExcelWriter("output.xlsx", mode="a") as writer:  # doctest: +SKIP
+        ...     df1.to_excel(writer, sheet_name="Sheet_name_3")
+
+        To set the library that is used to write the Excel file,
+        you can pass the `engine` keyword (the default engine is
+        automatically chosen depending on the file extension):
+
+        >>> df1.to_excel("output1.xlsx", engine="xlsxwriter")  # doctest: +SKIP
+        """
+        if engine_kwargs is None:
+            engine_kwargs = {}
+
+        df = self if isinstance(self, ABCDataFrame) else self.to_frame()
+
+        from pandas.io.formats.excel import ExcelFormatter
+
+        formatter = ExcelFormatter(
+            df,
+            na_rep=na_rep,
+            cols=columns,
+            header=header,
+            float_format=float_format,
+            index=index,
+            index_label=index_label,
+            merge_cells=merge_cells,
+            inf_rep=inf_rep,
+        )
+        formatter.write(
+            excel_writer,
+            sheet_name=sheet_name,
+            startrow=startrow,
+            startcol=startcol,
+            freeze_panes=freeze_panes,
+            engine=engine,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @final
+    @doc(
+        storage_options=_shared_docs["storage_options"],
+        compression_options=_shared_docs["compression_options"] % "path_or_buf",
+    )
+    def to_json(
+        self,
+        path_or_buf: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = None,
+        *,
+        orient: Literal["split", "records", "index", "table", "columns", "values"]
+        | None = None,
+        date_format: str | None = None,
+        double_precision: int = 10,
+        force_ascii: bool = True,
+        date_unit: TimeUnit = "ms",
+        default_handler: Callable[[Any], JSONSerializable] | None = None,
+        lines: bool = False,
+        compression: CompressionOptions = "infer",
+        index: bool | None = None,
+        indent: int | None = None,
+        storage_options: StorageOptions | None = None,
+        mode: Literal["a", "w"] = "w",
+    ) -> str | None:
+        """
+        Convert the object to a JSON string.
+
+        Note NaN's and None will be converted to null and datetime objects
+        will be converted to UNIX timestamps.
+
+        Parameters
+        ----------
+        path_or_buf : str, path object, file-like object, or None, default None
+            String, path object (implementing os.PathLike[str]), or file-like
+            object implementing a write() function. If None, the result is
+            returned as a string.
+        orient : str
+            Indication of expected JSON string format.
+
+            * Series:
+
+                - default is 'index'
+                - allowed values are: {{'split', 'records', 'index', 'table'}}.
+
+            * DataFrame:
+
+                - default is 'columns'
+                - allowed values are: {{'split', 'records', 'index', 'columns',
+                  'values', 'table'}}.
+
+            * The format of the JSON string:
+
+                - 'split' : dict like {{'index' -> [index], 'columns' -> [columns],
+                  'data' -> [values]}}
+                - 'records' : list like [{{column -> value}}, ... , {{column -> value}}]
+                - 'index' : dict like {{index -> {{column -> value}}}}
+                - 'columns' : dict like {{column -> {{index -> value}}}}
+                - 'values' : just the values array
+                - 'table' : dict like {{'schema': {{schema}}, 'data': {{data}}}}
+
+                Describing the data, where data component is like ``orient='records'``.
+
+        date_format : {{None, 'epoch', 'iso'}}
+            Type of date conversion. 'epoch' = epoch milliseconds,
+            'iso' = ISO8601. The default depends on the `orient`. For
+            ``orient='table'``, the default is 'iso'. For all other orients,
+            the default is 'epoch'.
+
+            .. deprecated:: 3.0.0
+                'epoch' date format is deprecated and will be removed in a future
+                version, please use 'iso' instead.
+
+        double_precision : int, default 10
+            The number of decimal places to use when encoding
+            floating point values. The possible maximal value is 15.
+            Passing double_precision greater than 15 will raise a ValueError.
+        force_ascii : bool, default True
+            Force encoded string to be ASCII.
+        date_unit : str, default 'ms' (milliseconds)
+            The time unit to encode to, governs timestamp and ISO8601
+            precision.  One of 's', 'ms', 'us', 'ns' for second, millisecond,
+            microsecond, and nanosecond respectively.
+        default_handler : callable, default None
+            Handler to call if object cannot otherwise be converted to a
+            suitable format for JSON. Should receive a single argument which is
+            the object to convert and return a serialisable object.
+        lines : bool, default False
+            If 'orient' is 'records' write out line-delimited json format. Will
+            throw ValueError if incorrect 'orient' since others are not
+            list-like.
+        {compression_options}
+
+            .. versionchanged:: 1.4.0 Zstandard support.
+
+        index : bool or None, default None
+            The index is only used when 'orient' is 'split', 'index', 'column',
+            or 'table'. Of these, 'index' and 'column' do not support
+            `index=False`. The string 'index' as a column name with empty :class:`Index`
+            or if it is 'index' will raise a ``ValueError``.
+
+        indent : int, optional
+           Length of whitespace used to indent each record.
+
+        {storage_options}
+
+        mode : str, default 'w' (writing)
+            Specify the IO mode for output when supplying a path_or_buf.
+            Accepted args are 'w' (writing) and 'a' (append) only.
+            mode='a' is only supported when lines is True and orient is 'records'.
+
+        Returns
+        -------
+        None or str
+            If path_or_buf is None, returns the resulting json format as a
+            string. Otherwise returns None.
+
+        See Also
+        --------
+        read_json : Convert a JSON string to pandas object.
+
+        Notes
+        -----
+        The behavior of ``indent=0`` varies from the stdlib, which does not
+        indent the output but does insert newlines. Currently, ``indent=0``
+        and the default ``indent=None`` are equivalent in pandas, though this
+        may change in a future release.
+
+        ``orient='table'`` contains a 'pandas_version' field under 'schema'.
+        This stores the version of `pandas` used in the latest revision of the
+        schema.
+
+        Examples
+        --------
+        >>> from json import loads, dumps
+        >>> df = pd.DataFrame(
+        ...     [["a", "b"], ["c", "d"]],
+        ...     index=["row 1", "row 2"],
+        ...     columns=["col 1", "col 2"],
+        ... )
+
+        >>> result = df.to_json(orient="split")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        {{
+            "columns": [
+                "col 1",
+                "col 2"
+            ],
+            "index": [
+                "row 1",
+                "row 2"
+            ],
+            "data": [
+                [
+                    "a",
+                    "b"
+                ],
+                [
+                    "c",
+                    "d"
+                ]
+            ]
+        }}
+
+        Encoding/decoding a Dataframe using ``'records'`` formatted JSON.
+        Note that index labels are not preserved with this encoding.
+
+        >>> result = df.to_json(orient="records")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        [
+            {{
+                "col 1": "a",
+                "col 2": "b"
+            }},
+            {{
+                "col 1": "c",
+                "col 2": "d"
+            }}
+        ]
+
+        Encoding/decoding a Dataframe using ``'index'`` formatted JSON:
+
+        >>> result = df.to_json(orient="index")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        {{
+            "row 1": {{
+                "col 1": "a",
+                "col 2": "b"
+            }},
+            "row 2": {{
+                "col 1": "c",
+                "col 2": "d"
+            }}
+        }}
+
+        Encoding/decoding a Dataframe using ``'columns'`` formatted JSON:
+
+        >>> result = df.to_json(orient="columns")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        {{
+            "col 1": {{
+                "row 1": "a",
+                "row 2": "c"
+            }},
+            "col 2": {{
+                "row 1": "b",
+                "row 2": "d"
+            }}
+        }}
+
+        Encoding/decoding a Dataframe using ``'values'`` formatted JSON:
+
+        >>> result = df.to_json(orient="values")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        [
+            [
+                "a",
+                "b"
+            ],
+            [
+                "c",
+                "d"
+            ]
+        ]
+
+        Encoding with Table Schema:
+
+        >>> result = df.to_json(orient="table")
+        >>> parsed = loads(result)
+        >>> dumps(parsed, indent=4)  # doctest: +SKIP
+        {{
+            "schema": {{
+                "fields": [
+                    {{
+                        "name": "index",
+                        "type": "string"
+                    }},
+                    {{
+                        "name": "col 1",
+                        "type": "string"
+                    }},
+                    {{
+                        "name": "col 2",
+                        "type": "string"
+                    }}
+                ],
+                "primaryKey": [
+                    "index"
+                ],
+                "pandas_version": "1.4.0"
+            }},
+            "data": [
+                {{
+                    "index": "row 1",
+                    "col 1": "a",
+                    "col 2": "b"
+                }},
+                {{
+                    "index": "row 2",
+                    "col 1": "c",
+                    "col 2": "d"
+                }}
+            ]
+        }}
+        """
+        from pandas.io import json
+
+        if date_format is None and orient == "table":
+            date_format = "iso"
+        elif date_format is None:
+            date_format = "epoch"
+            dtypes = self.dtypes if self.ndim == 2 else [self.dtype]
+            if any(lib.is_np_dtype(dtype, "mM") for dtype in dtypes):
+                warnings.warn(
+                    "The default 'epoch' date format is deprecated and will be removed "
+                    "in a future version, please use 'iso' date format instead.",
+                    FutureWarning,
+                    stacklevel=find_stack_level(),
+                )
+        elif date_format == "epoch":
+            # GH#57063
+            warnings.warn(
+                "'epoch' date format is deprecated and will be removed in a future "
+                "version, please use 'iso' date format instead.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        config.is_nonnegative_int(indent)
+        indent = indent or 0
+
+        return json.to_json(
+            path_or_buf=path_or_buf,
+            obj=self,
+            orient=orient,
+            date_format=date_format,
+            double_precision=double_precision,
+            force_ascii=force_ascii,
+            date_unit=date_unit,
+            default_handler=default_handler,
+            lines=lines,
+            compression=compression,
+            index=index,
+            indent=indent,
+            storage_options=storage_options,
+            mode=mode,
+        )
+
+    @final
+    def to_hdf(
+        self,
+        path_or_buf: FilePath | HDFStore,
+        *,
+        key: str,
+        mode: Literal["a", "w", "r+"] = "a",
+        complevel: int | None = None,
+        complib: Literal["zlib", "lzo", "bzip2", "blosc"] | None = None,
+        append: bool = False,
+        format: Literal["fixed", "table"] | None = None,
+        index: bool = True,
+        min_itemsize: int | dict[str, int] | None = None,
+        nan_rep=None,
+        dropna: bool | None = None,
+        data_columns: Literal[True] | list[str] | None = None,
+        errors: OpenFileErrors = "strict",
+        encoding: str = "UTF-8",
+    ) -> None:
+        """
+        Write the contained data to an HDF5 file using HDFStore.
+
+        Hierarchical Data Format (HDF) is self-describing, allowing an
+        application to interpret the structure and contents of a file with
+        no outside information. One HDF file can hold a mix of related objects
+        which can be accessed as a group or as individual objects.
+
+        In order to add another DataFrame or Series to an existing HDF file
+        please use append mode and a different a key.
+
+        .. warning::
+
+           One can store a subclass of ``DataFrame`` or ``Series`` to HDF5,
+           but the type of the subclass is lost upon storing.
+
+        For more information see the :ref:`user guide <io.hdf5>`.
+
+        Parameters
+        ----------
+        path_or_buf : str or pandas.HDFStore
+            File path or HDFStore object.
+        key : str
+            Identifier for the group in the store.
+        mode : {'a', 'w', 'r+'}, default 'a'
+            Mode to open file:
+
+            - 'w': write, a new file is created (an existing file with
+              the same name would be deleted).
+            - 'a': append, an existing file is opened for reading and
+              writing, and if the file does not exist it is created.
+            - 'r+': similar to 'a', but the file must already exist.
+        complevel : {0-9}, default None
+            Specifies a compression level for data.
+            A value of 0 or None disables compression.
+        complib : {'zlib', 'lzo', 'bzip2', 'blosc'}, default 'zlib'
+            Specifies the compression library to be used.
+            These additional compressors for Blosc are supported
+            (default if no compressor specified: 'blosc:blosclz'):
+            {'blosc:blosclz', 'blosc:lz4', 'blosc:lz4hc', 'blosc:snappy',
+            'blosc:zlib', 'blosc:zstd'}.
+            Specifying a compression library which is not available issues
+            a ValueError.
+        append : bool, default False
+            For Table formats, append the input data to the existing.
+        format : {'fixed', 'table', None}, default 'fixed'
+            Possible values:
+
+            - 'fixed': Fixed format. Fast writing/reading. Not-appendable,
+              nor searchable.
+            - 'table': Table format. Write as a PyTables Table structure
+              which may perform worse but allow more flexible operations
+              like searching / selecting subsets of the data.
+            - If None, pd.get_option('io.hdf.default_format') is checked,
+              followed by fallback to "fixed".
+        index : bool, default True
+            Write DataFrame index as a column.
+        min_itemsize : dict or int, optional
+            Map column names to minimum string sizes for columns.
+        nan_rep : Any, optional
+            How to represent null values as str.
+            Not allowed with append=True.
+        dropna : bool, default False, optional
+            Remove missing values.
+        data_columns : list of columns or True, optional
+            List of columns to create as indexed data columns for on-disk
+            queries, or True to use all columns. By default only the axes
+            of the object are indexed. See
+            :ref:`Query via data columns<io.hdf5-query-data-columns>`. for
+            more information.
+            Applicable only to format='table'.
+        errors : str, default 'strict'
+            Specifies how encoding and decoding errors are to be handled.
+            See the errors argument for :func:`open` for a full list
+            of options.
+        encoding : str, default "UTF-8"
+            Set character encoding.
+
+        See Also
+        --------
+        read_hdf : Read from HDF file.
+        DataFrame.to_orc : Write a DataFrame to the binary orc format.
+        DataFrame.to_parquet : Write a DataFrame to the binary parquet format.
+        DataFrame.to_sql : Write to a SQL table.
+        DataFrame.to_feather : Write out feather-format for DataFrames.
+        DataFrame.to_csv : Write out to a csv file.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"A": [1, 2, 3], "B": [4, 5, 6]}, index=["a", "b", "c"]
+        ... )  # doctest: +SKIP
+        >>> df.to_hdf("data.h5", key="df", mode="w")  # doctest: +SKIP
+
+        We can add another object to the same file:
+
+        >>> s = pd.Series([1, 2, 3, 4])  # doctest: +SKIP
+        >>> s.to_hdf("data.h5", key="s")  # doctest: +SKIP
+
+        Reading from HDF file:
+
+        >>> pd.read_hdf("data.h5", "df")  # doctest: +SKIP
+        A  B
+        a  1  4
+        b  2  5
+        c  3  6
+        >>> pd.read_hdf("data.h5", "s")  # doctest: +SKIP
+        0    1
+        1    2
+        2    3
+        3    4
+        dtype: int64
+        """
+        from pandas.io import pytables
+
+        # Argument 3 to "to_hdf" has incompatible type "NDFrame"; expected
+        # "Union[DataFrame, Series]" [arg-type]
+        pytables.to_hdf(
+            path_or_buf,
+            key,
+            self,  # type: ignore[arg-type]
+            mode=mode,
+            complevel=complevel,
+            complib=complib,
+            append=append,
+            format=format,
+            index=index,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            dropna=dropna,
+            data_columns=data_columns,
+            errors=errors,
+            encoding=encoding,
+        )
+
+    @final
+    def to_sql(
+        self,
+        name: str,
+        con,
+        *,
+        schema: str | None = None,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        index: bool = True,
+        index_label: IndexLabel | None = None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        method: Literal["multi"] | Callable | None = None,
+    ) -> int | None:
+        """
+        Write records stored in a DataFrame to a SQL database.
+
+        Databases supported by SQLAlchemy [1]_ are supported. Tables can be
+        newly created, appended to, or overwritten.
+
+        .. warning::
+            The pandas library does not attempt to sanitize inputs provided via a to_sql call.
+            Please refer to the documentation for the underlying database driver to see if it
+            will properly prevent injection, or alternatively be advised of a security risk when
+            executing arbitrary commands in a to_sql call.
+
+        Parameters
+        ----------
+        name : str
+            Name of SQL table.
+        con : ADBC connection, sqlalchemy.engine.(Engine or Connection) or sqlite3.Connection
+            ADBC provides high performance I/O with native type support, where available.
+            Using SQLAlchemy makes it possible to use any DB supported by that
+            library. Legacy support is provided for sqlite3.Connection objects. The user
+            is responsible for engine disposal and connection closure for the SQLAlchemy
+            connectable. See `here \
+                <https://docs.sqlalchemy.org/en/20/core/connections.html>`_.
+            If passing a sqlalchemy.engine.Connection which is already in a transaction,
+            the transaction will not be committed.  If passing a sqlite3.Connection,
+            it will not be possible to roll back the record insertion.
+
+        schema : str, optional
+            Specify the schema (if database flavor supports this). If None, use
+            default schema.
+        if_exists : {'fail', 'replace', 'append', 'delete_rows'}, default 'fail'
+            How to behave if the table already exists.
+
+            * fail: Raise a ValueError.
+            * replace: Drop the table before inserting new values.
+            * append: Insert new values to the existing table.
+            * delete_rows: If a table exists, delete all records and insert data.
+
+        index : bool, default True
+            Write DataFrame index as a column. Uses `index_label` as the column
+            name in the table. Creates a table index for this column.
+        index_label : str or sequence, default None
+            Column label for index column(s). If None is given (default) and
+            `index` is True, then the index names are used.
+            A sequence should be given if the DataFrame uses MultiIndex.
+        chunksize : int, optional
+            Specify the number of rows in each batch to be written to the database connection at a time.
+            By default, all rows will be written at once. Also see the method keyword.
+        dtype : dict or scalar, optional
+            Specifying the datatype for columns. If a dictionary is used, the
+            keys should be the column names and the values should be the
+            SQLAlchemy types or strings for the sqlite3 legacy mode. If a
+            scalar is provided, it will be applied to all columns.
+        method : {None, 'multi', callable}, optional
+            Controls the SQL insertion clause used:
+
+            * None : Uses standard SQL ``INSERT`` clause (one per row).
+            * 'multi': Pass multiple values in a single ``INSERT`` clause.
+            * callable with signature ``(pd_table, conn, keys, data_iter)``.
+
+            Details and a sample callable implementation can be found in the
+            section :ref:`insert method <io.sql.method>`.
+
+        Returns
+        -------
+        None or int
+            Number of rows affected by to_sql. None is returned if the callable
+            passed into ``method`` does not return an integer number of rows.
+
+            The number of returned rows affected is the sum of the ``rowcount``
+            attribute of ``sqlite3.Cursor`` or SQLAlchemy connectable which may not
+            reflect the exact number of written rows as stipulated in the
+            `sqlite3 <https://docs.python.org/3/library/sqlite3.html#sqlite3.Cursor.rowcount>`__ or
+            `SQLAlchemy <https://docs.sqlalchemy.org/en/20/core/connections.html#sqlalchemy.engine.CursorResult.rowcount>`__.
+
+            .. versionadded:: 1.4.0
+
+        Raises
+        ------
+        ValueError
+            When the table already exists and `if_exists` is 'fail' (the
+            default).
+
+        See Also
+        --------
+        read_sql : Read a DataFrame from a table.
+
+        Notes
+        -----
+        Timezone aware datetime columns will be written as
+        ``Timestamp with timezone`` type with SQLAlchemy if supported by the
+        database. Otherwise, the datetimes will be stored as timezone unaware
+        timestamps local to the original timezone.
+
+        Not all datastores support ``method="multi"``. Oracle, for example,
+        does not support multi-value insert.
+
+        References
+        ----------
+        .. [1] https://docs.sqlalchemy.org
+        .. [2] https://www.python.org/dev/peps/pep-0249/
+
+        Examples
+        --------
+        Create an in-memory SQLite database.
+
+        >>> from sqlalchemy import create_engine
+        >>> engine = create_engine('sqlite://', echo=False)
+
+        Create a table from scratch with 3 rows.
+
+        >>> df = pd.DataFrame({'name' : ['User 1', 'User 2', 'User 3']})
+        >>> df
+             name
+        0  User 1
+        1  User 2
+        2  User 3
+
+        >>> df.to_sql(name='users', con=engine)
+        3
+        >>> from sqlalchemy import text
+        >>> with engine.connect() as conn:
+        ...     conn.execute(text("SELECT * FROM users")).fetchall()
+        [(0, 'User 1'), (1, 'User 2'), (2, 'User 3')]
+
+        An `sqlalchemy.engine.Connection` can also be passed to `con`:
+
+        >>> with engine.begin() as connection:
+        ...     df1 = pd.DataFrame({'name' : ['User 4', 'User 5']})
+        ...     df1.to_sql(name='users', con=connection, if_exists='append')
+        2
+
+        This is allowed to support operations that require that the same
+        DBAPI connection is used for the entire operation.
+
+        >>> df2 = pd.DataFrame({'name' : ['User 6', 'User 7']})
+        >>> df2.to_sql(name='users', con=engine, if_exists='append')
+        2
+        >>> with engine.connect() as conn:
+        ...     conn.execute(text("SELECT * FROM users")).fetchall()
+        [(0, 'User 1'), (1, 'User 2'), (2, 'User 3'),
+         (0, 'User 4'), (1, 'User 5'), (0, 'User 6'),
+         (1, 'User 7')]
+
+        Overwrite the table with just ``df2``.
+
+        >>> df2.to_sql(name='users', con=engine, if_exists='replace',
+        ...            index_label='id')
+        2
+        >>> with engine.connect() as conn:
+        ...     conn.execute(text("SELECT * FROM users")).fetchall()
+        [(0, 'User 6'), (1, 'User 7')]
+
+        Delete all rows before inserting new records with ``df3``
+
+        >>> df3 = pd.DataFrame({"name": ['User 8', 'User 9']})
+        >>> df3.to_sql(name='users', con=engine, if_exists='delete_rows',
+        ...            index_label='id')
+        2
+        >>> with engine.connect() as conn:
+        ...     conn.execute(text("SELECT * FROM users")).fetchall()
+        [(0, 'User 8'), (1, 'User 9')]
+
+        Use ``method`` to define a callable insertion method to do nothing
+        if there's a primary key conflict on a table in a PostgreSQL database.
+
+        >>> from sqlalchemy.dialects.postgresql import insert
+        >>> def insert_on_conflict_nothing(table, conn, keys, data_iter):
+        ...     # "a" is the primary key in "conflict_table"
+        ...     data = [dict(zip(keys, row)) for row in data_iter]
+        ...     stmt = insert(table.table).values(data).on_conflict_do_nothing(index_elements=["a"])
+        ...     result = conn.execute(stmt)
+        ...     return result.rowcount
+        >>> df_conflict.to_sql(name="conflict_table", con=conn, if_exists="append",  # noqa: F821
+        ...                    method=insert_on_conflict_nothing)  # doctest: +SKIP
+        0
+
+        For MySQL, a callable to update columns ``b`` and ``c`` if there's a conflict
+        on a primary key.
+
+        >>> from sqlalchemy.dialects.mysql import insert   # noqa: F811
+        >>> def insert_on_conflict_update(table, conn, keys, data_iter):
+        ...     # update columns "b" and "c" on primary key conflict
+        ...     data = [dict(zip(keys, row)) for row in data_iter]
+        ...     stmt = (
+        ...         insert(table.table)
+        ...         .values(data)
+        ...     )
+        ...     stmt = stmt.on_duplicate_key_update(b=stmt.inserted.b, c=stmt.inserted.c)
+        ...     result = conn.execute(stmt)
+        ...     return result.rowcount
+        >>> df_conflict.to_sql(name="conflict_table", con=conn, if_exists="append",  # noqa: F821
+        ...                    method=insert_on_conflict_update)  # doctest: +SKIP
+        2
+
+        Specify the dtype (especially useful for integers with missing values).
+        Notice that while pandas is forced to store the data as floating point,
+        the database supports nullable integers. When fetching the data with
+        Python, we get back integer scalars.
+
+        >>> df = pd.DataFrame({"A": [1, None, 2]})
+        >>> df
+             A
+        0  1.0
+        1  NaN
+        2  2.0
+
+        >>> from sqlalchemy.types import Integer
+        >>> df.to_sql(name='integers', con=engine, index=False,
+        ...           dtype={"A": Integer()})
+        3
+
+        >>> with engine.connect() as conn:
+        ...     conn.execute(text("SELECT * FROM integers")).fetchall()
+        [(1,), (None,), (2,)]
+
+        .. versionadded:: 2.2.0
+
+           pandas now supports writing via ADBC drivers
+
+        >>> df = pd.DataFrame({'name' : ['User 10', 'User 11', 'User 12']})
+        >>> df
+              name
+        0  User 10
+        1  User 11
+        2  User 12
+
+        >>> from adbc_driver_sqlite import dbapi  # doctest:+SKIP
+        >>> with dbapi.connect("sqlite://") as conn:  # doctest:+SKIP
+        ...     df.to_sql(name="users", con=conn)
+        3
+        """  # noqa: E501
+        from pandas.io import sql
+
+        return sql.to_sql(
+            self,
+            name,
+            con,
+            schema=schema,
+            if_exists=if_exists,
+            index=index,
+            index_label=index_label,
+            chunksize=chunksize,
+            dtype=dtype,
+            method=method,
+        )
+
+    @final
+    @doc(
+        storage_options=_shared_docs["storage_options"],
+        compression_options=_shared_docs["compression_options"] % "path",
+    )
+    def to_pickle(
+        self,
+        path: FilePath | WriteBuffer[bytes],
+        *,
+        compression: CompressionOptions = "infer",
+        protocol: int = pickle.HIGHEST_PROTOCOL,
+        storage_options: StorageOptions | None = None,
+    ) -> None:
+        """
+        Pickle (serialize) object to file.
+
+        Parameters
+        ----------
+        path : str, path object, or file-like object
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a binary ``write()`` function. File path where
+            the pickled object will be stored.
+        {compression_options}
+        protocol : int
+            Int which indicates which protocol should be used by the pickler,
+            default HIGHEST_PROTOCOL (see [1]_ paragraph 12.1.2). The possible
+            values are 0, 1, 2, 3, 4, 5. A negative value for the protocol
+            parameter is equivalent to setting its value to HIGHEST_PROTOCOL.
+
+            .. [1] https://docs.python.org/3/library/pickle.html.
+
+        {storage_options}
+
+        See Also
+        --------
+        read_pickle : Load pickled pandas object (or any object) from file.
+        DataFrame.to_hdf : Write DataFrame to an HDF5 file.
+        DataFrame.to_sql : Write DataFrame to a SQL database.
+        DataFrame.to_parquet : Write a DataFrame to the binary parquet format.
+
+        Examples
+        --------
+        >>> original_df = pd.DataFrame(
+        ...     {{"foo": range(5), "bar": range(5, 10)}}
+        ... )  # doctest: +SKIP
+        >>> original_df  # doctest: +SKIP
+           foo  bar
+        0    0    5
+        1    1    6
+        2    2    7
+        3    3    8
+        4    4    9
+        >>> original_df.to_pickle("./dummy.pkl")  # doctest: +SKIP
+
+        >>> unpickled_df = pd.read_pickle("./dummy.pkl")  # doctest: +SKIP
+        >>> unpickled_df  # doctest: +SKIP
+           foo  bar
+        0    0    5
+        1    1    6
+        2    2    7
+        3    3    8
+        4    4    9
+        """
+        from pandas.io.pickle import to_pickle
+
+        to_pickle(
+            self,
+            path,
+            compression=compression,
+            protocol=protocol,
+            storage_options=storage_options,
+        )
+
+    @final
+    def to_clipboard(
+        self, *, excel: bool = True, sep: str | None = None, **kwargs
+    ) -> None:
+        r"""
+        Copy object to the system clipboard.
+
+        Write a text representation of object to the system clipboard.
+        This can be pasted into Excel, for example.
+
+        Parameters
+        ----------
+        excel : bool, default True
+            Produce output in a csv format for easy pasting into excel.
+
+            - True, use the provided separator for csv pasting.
+            - False, write a string representation of the object to the clipboard.
+
+        sep : str, default ``'\t'``
+            Field delimiter.
+        **kwargs
+            These parameters will be passed to DataFrame.to_csv.
+
+        See Also
+        --------
+        DataFrame.to_csv : Write a DataFrame to a comma-separated values
+            (csv) file.
+        read_clipboard : Read text from clipboard and pass to read_csv.
+
+        Notes
+        -----
+        Requirements for your platform.
+
+          - Linux : `xclip`, or `xsel` (with `PyQt4` modules)
+          - Windows : none
+          - macOS : none
+
+        This method uses the processes developed for the package `pyperclip`. A
+        solution to render any output string format is given in the examples.
+
+        Examples
+        --------
+        Copy the contents of a DataFrame to the clipboard.
+
+        >>> df = pd.DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+
+        >>> df.to_clipboard(sep=",")  # doctest: +SKIP
+        ... # Wrote the following to the system clipboard:
+        ... # ,A,B,C
+        ... # 0,1,2,3
+        ... # 1,4,5,6
+
+        We can omit the index by passing the keyword `index` and setting
+        it to false.
+
+        >>> df.to_clipboard(sep=",", index=False)  # doctest: +SKIP
+        ... # Wrote the following to the system clipboard:
+        ... # A,B,C
+        ... # 1,2,3
+        ... # 4,5,6
+
+        Using the original `pyperclip` package for any string output format.
+
+        .. code-block:: python
+
+           import pyperclip
+
+           html = df.style.to_html()
+           pyperclip.copy(html)
+        """
+        from pandas.io import clipboards
+
+        clipboards.to_clipboard(self, excel=excel, sep=sep, **kwargs)
+
+    @final
+    def to_xarray(self):
+        """
+        Return an xarray object from the pandas object.
+
+        Returns
+        -------
+        xarray.DataArray or xarray.Dataset
+            Data in the pandas structure converted to Dataset if the object is
+            a DataFrame, or a DataArray if the object is a Series.
+
+        See Also
+        --------
+        DataFrame.to_hdf : Write DataFrame to an HDF5 file.
+        DataFrame.to_parquet : Write a DataFrame to the binary parquet format.
+
+        Notes
+        -----
+        See the `xarray docs <https://xarray.pydata.org/en/stable/>`__
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("falcon", "bird", 389.0, 2),
+        ...         ("parrot", "bird", 24.0, 2),
+        ...         ("lion", "mammal", 80.5, 4),
+        ...         ("monkey", "mammal", np.nan, 4),
+        ...     ],
+        ...     columns=["name", "class", "max_speed", "num_legs"],
+        ... )
+        >>> df
+             name   class  max_speed  num_legs
+        0  falcon    bird      389.0         2
+        1  parrot    bird       24.0         2
+        2    lion  mammal       80.5         4
+        3  monkey  mammal        NaN         4
+
+        >>> df.to_xarray()  # doctest: +SKIP
+        <xarray.Dataset>
+        Dimensions:    (index: 4)
+        Coordinates:
+          * index      (index) int64 32B 0 1 2 3
+        Data variables:
+            name       (index) object 32B 'falcon' 'parrot' 'lion' 'monkey'
+            class      (index) object 32B 'bird' 'bird' 'mammal' 'mammal'
+            max_speed  (index) float64 32B 389.0 24.0 80.5 nan
+            num_legs   (index) int64 32B 2 2 4 4
+
+        >>> df["max_speed"].to_xarray()  # doctest: +SKIP
+        <xarray.DataArray 'max_speed' (index: 4)>
+        array([389. ,  24. ,  80.5,   nan])
+        Coordinates:
+          * index    (index) int64 0 1 2 3
+
+        >>> dates = pd.to_datetime(
+        ...     ["2018-01-01", "2018-01-01", "2018-01-02", "2018-01-02"]
+        ... )
+        >>> df_multiindex = pd.DataFrame(
+        ...     {
+        ...         "date": dates,
+        ...         "animal": ["falcon", "parrot", "falcon", "parrot"],
+        ...         "speed": [350, 18, 361, 15],
+        ...     }
+        ... )
+        >>> df_multiindex = df_multiindex.set_index(["date", "animal"])
+
+        >>> df_multiindex
+                           speed
+        date       animal
+        2018-01-01 falcon    350
+                   parrot     18
+        2018-01-02 falcon    361
+                   parrot     15
+
+        >>> df_multiindex.to_xarray()  # doctest: +SKIP
+        <xarray.Dataset>
+        Dimensions:  (date: 2, animal: 2)
+        Coordinates:
+          * date     (date) datetime64[s] 2018-01-01 2018-01-02
+          * animal   (animal) object 'falcon' 'parrot'
+        Data variables:
+            speed    (date, animal) int64 350 18 361 15
+        """
+        xarray = import_optional_dependency("xarray")
+
+        if self.ndim == 1:
+            return xarray.DataArray.from_series(self)
+        else:
+            return xarray.Dataset.from_dataframe(self)
+
+    @overload
+    def to_latex(
+        self,
+        buf: None = ...,
+        *,
+        columns: Sequence[Hashable] | None = ...,
+        header: bool | SequenceNotStr[str] = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: FormattersType | None = ...,
+        float_format: FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        bold_rows: bool = ...,
+        column_format: str | None = ...,
+        longtable: bool | None = ...,
+        escape: bool | None = ...,
+        encoding: str | None = ...,
+        decimal: str = ...,
+        multicolumn: bool | None = ...,
+        multicolumn_format: str | None = ...,
+        multirow: bool | None = ...,
+        caption: str | tuple[str, str] | None = ...,
+        label: str | None = ...,
+        position: str | None = ...,
+    ) -> str: ...
+
+    @overload
+    def to_latex(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        columns: Sequence[Hashable] | None = ...,
+        header: bool | SequenceNotStr[str] = ...,
+        index: bool = ...,
+        na_rep: str = ...,
+        formatters: FormattersType | None = ...,
+        float_format: FloatFormatType | None = ...,
+        sparsify: bool | None = ...,
+        index_names: bool = ...,
+        bold_rows: bool = ...,
+        column_format: str | None = ...,
+        longtable: bool | None = ...,
+        escape: bool | None = ...,
+        encoding: str | None = ...,
+        decimal: str = ...,
+        multicolumn: bool | None = ...,
+        multicolumn_format: str | None = ...,
+        multirow: bool | None = ...,
+        caption: str | tuple[str, str] | None = ...,
+        label: str | None = ...,
+        position: str | None = ...,
+    ) -> None: ...
+
+    @final
+    def to_latex(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        columns: Sequence[Hashable] | None = None,
+        header: bool | SequenceNotStr[str] = True,
+        index: bool = True,
+        na_rep: str = "NaN",
+        formatters: FormattersType | None = None,
+        float_format: FloatFormatType | None = None,
+        sparsify: bool | None = None,
+        index_names: bool = True,
+        bold_rows: bool = False,
+        column_format: str | None = None,
+        longtable: bool | None = None,
+        escape: bool | None = None,
+        encoding: str | None = None,
+        decimal: str = ".",
+        multicolumn: bool | None = None,
+        multicolumn_format: str | None = None,
+        multirow: bool | None = None,
+        caption: str | tuple[str, str] | None = None,
+        label: str | None = None,
+        position: str | None = None,
+    ) -> str | None:
+        r"""
+        Render object to a LaTeX tabular, longtable, or nested table.
+
+        Requires ``\usepackage{booktabs}``.  The output can be copy/pasted
+        into a main LaTeX document or read from an external file
+        with ``\input{table.tex}``.
+
+        .. versionchanged:: 2.0.0
+           Refactored to use the Styler implementation via jinja2 templating.
+
+        Parameters
+        ----------
+        buf : str, Path or StringIO-like, optional, default None
+            Buffer to write to. If None, the output is returned as a string.
+        columns : list of label, optional
+            The subset of columns to write. Writes all columns by default.
+        header : bool or list of str, default True
+            Write out the column names. If a list of strings is given,
+            it is assumed to be aliases for the column names. Braces must be escaped.
+        index : bool, default True
+            Write row names (index).
+        na_rep : str, default 'NaN'
+            Missing data representation.
+        formatters : list of functions or dict of {str: function}, optional
+            Formatter functions to apply to columns' elements by position or
+            name. The result of each function must be a unicode string.
+            List must be of length equal to the number of columns.
+        float_format : one-parameter function or str, optional, default None
+            Formatter for floating point numbers. For example
+            ``float_format="%.2f"`` and ``float_format="{:0.2f}".format`` will
+            both result in 0.1234 being formatted as 0.12.
+        sparsify : bool, optional
+            Set to False for a DataFrame with a hierarchical index to print
+            every multiindex key at each row. By default, the value will be
+            read from the config module.
+        index_names : bool, default True
+            Prints the names of the indexes.
+        bold_rows : bool, default False
+            Make the row labels bold in the output.
+        column_format : str, optional
+            The columns format as specified in `LaTeX table format
+            <https://en.wikibooks.org/wiki/LaTeX/Tables>`__ e.g. 'rcl' for 3
+            columns. By default, 'l' will be used for all columns except
+            columns of numbers, which default to 'r'.
+        longtable : bool, optional
+            Use a longtable environment instead of tabular. Requires
+            adding a \usepackage{longtable} to your LaTeX preamble.
+            By default, the value will be read from the pandas config
+            module, and set to `True` if the option ``styler.latex.environment`` is
+            `"longtable"`.
+
+            .. versionchanged:: 2.0.0
+               The pandas option affecting this argument has changed.
+        escape : bool, optional
+            By default, the value will be read from the pandas config
+            module and set to `True` if the option ``styler.format.escape`` is
+            `"latex"`. When set to False prevents from escaping latex special
+            characters in column names.
+
+            .. versionchanged:: 2.0.0
+               The pandas option affecting this argument has changed, as has the
+               default value to `False`.
+        encoding : str, optional
+            A string representing the encoding to use in the output file,
+            defaults to 'utf-8'.
+        decimal : str, default '.'
+            Character recognized as decimal separator, e.g. ',' in Europe.
+        multicolumn : bool, default True
+            Use \multicolumn to enhance MultiIndex columns.
+            The default will be read from the config module, and is set
+            as the option ``styler.sparse.columns``.
+
+            .. versionchanged:: 2.0.0
+               The pandas option affecting this argument has changed.
+        multicolumn_format : str, default 'r'
+            The alignment for multicolumns, similar to `column_format`
+            The default will be read from the config module, and is set as the option
+            ``styler.latex.multicol_align``.
+
+            .. versionchanged:: 2.0.0
+               The pandas option affecting this argument has changed, as has the
+               default value to "r".
+        multirow : bool, default True
+            Use \multirow to enhance MultiIndex rows. Requires adding a
+            \usepackage{multirow} to your LaTeX preamble. Will print
+            centered labels (instead of top-aligned) across the contained
+            rows, separating groups via clines. The default will be read
+            from the pandas config module, and is set as the option
+            ``styler.sparse.index``.
+
+            .. versionchanged:: 2.0.0
+               The pandas option affecting this argument has changed, as has the
+               default value to `True`.
+        caption : str or tuple, optional
+            Tuple (full_caption, short_caption),
+            which results in ``\caption[short_caption]{full_caption}``;
+            if a single string is passed, no short caption will be set.
+        label : str, optional
+            The LaTeX label to be placed inside ``\label{}`` in the output.
+            This is used with ``\ref{}`` in the main ``.tex`` file.
+
+        position : str, optional
+            The LaTeX positional argument for tables, to be placed after
+            ``\begin{}`` in the output.
+
+        Returns
+        -------
+        str or None
+            If buf is None, returns the result as a string. Otherwise returns None.
+
+        See Also
+        --------
+        io.formats.style.Styler.to_latex : Render a DataFrame to LaTeX
+            with conditional formatting.
+        DataFrame.to_string : Render a DataFrame to a console-friendly
+            tabular output.
+        DataFrame.to_html : Render a DataFrame as an HTML table.
+
+        Notes
+        -----
+        As of v2.0.0 this method has changed to use the Styler implementation as
+        part of :meth:`.Styler.to_latex` via ``jinja2`` templating. This means
+        that ``jinja2`` is a requirement, and needs to be installed, for this method
+        to function. It is advised that users switch to using Styler, since that
+        implementation is more frequently updated and contains much more
+        flexibility with the output.
+
+        Examples
+        --------
+        Convert a general DataFrame to LaTeX with formatting:
+
+        >>> df = pd.DataFrame(dict(name=['Raphael', 'Donatello'],
+        ...                        age=[26, 45],
+        ...                        height=[181.23, 177.65]))
+        >>> print(df.to_latex(index=False,
+        ...                   formatters={"name": str.upper},
+        ...                   float_format="{:.1f}".format,
+        ...                   ))  # doctest: +SKIP
+        \begin{tabular}{lrr}
+        \toprule
+        name & age & height \\
+        \midrule
+        RAPHAEL & 26 & 181.2 \\
+        DONATELLO & 45 & 177.7 \\
+        \bottomrule
+        \end{tabular}
+        """
+        # Get defaults from the pandas config
+        if self.ndim == 1:
+            self = self.to_frame()
+        if longtable is None:
+            longtable = config.get_option("styler.latex.environment") == "longtable"
+        if escape is None:
+            escape = config.get_option("styler.format.escape") == "latex"
+        if multicolumn is None:
+            multicolumn = config.get_option("styler.sparse.columns")
+        if multicolumn_format is None:
+            multicolumn_format = config.get_option("styler.latex.multicol_align")
+        if multirow is None:
+            multirow = config.get_option("styler.sparse.index")
+
+        if column_format is not None and not isinstance(column_format, str):
+            raise ValueError("`column_format` must be str or unicode")
+        length = len(self.columns) if columns is None else len(columns)
+        if isinstance(header, (list, tuple)) and len(header) != length:
+            raise ValueError(f"Writing {length} cols but got {len(header)} aliases")
+
+        # Refactor formatters/float_format/decimal/na_rep/escape to Styler structure
+        base_format_ = {
+            "na_rep": na_rep,
+            "escape": "latex" if escape else None,
+            "decimal": decimal,
+        }
+        index_format_: dict[str, Any] = {"axis": 0, **base_format_}
+        column_format_: dict[str, Any] = {"axis": 1, **base_format_}
+
+        if isinstance(float_format, str):
+            float_format_: Callable | None = lambda x: float_format % x
+        else:
+            float_format_ = float_format
+
+        def _wrap(x, alt_format_):
+            if isinstance(x, (float, complex)) and float_format_ is not None:
+                return float_format_(x)
+            else:
+                return alt_format_(x)
+
+        formatters_: list | tuple | dict | Callable | None = None
+        if isinstance(formatters, list):
+            formatters_ = {
+                c: partial(_wrap, alt_format_=formatters[i])
+                for i, c in enumerate(self.columns)
+            }
+        elif isinstance(formatters, dict):
+            index_formatter = formatters.pop("__index__", None)
+            column_formatter = formatters.pop("__columns__", None)
+            if index_formatter is not None:
+                index_format_.update({"formatter": index_formatter})
+            if column_formatter is not None:
+                column_format_.update({"formatter": column_formatter})
+
+            formatters_ = formatters
+            float_columns = self.select_dtypes(include="float").columns
+            for col in float_columns:
+                if col not in formatters.keys():
+                    formatters_.update({col: float_format_})
+        elif formatters is None and float_format is not None:
+            formatters_ = partial(_wrap, alt_format_=lambda v: v)
+        format_index_ = [index_format_, column_format_]
+
+        # Deal with hiding indexes and relabelling column names
+        hide_: list[dict] = []
+        relabel_index_: list[dict] = []
+        if columns:
+            hide_.append(
+                {
+                    "subset": [c for c in self.columns if c not in columns],
+                    "axis": "columns",
+                }
+            )
+        if header is False:
+            hide_.append({"axis": "columns"})
+        elif isinstance(header, (list, tuple)):
+            relabel_index_.append({"labels": header, "axis": "columns"})
+            format_index_ = [index_format_]  # column_format is overwritten
+
+        if index is False:
+            hide_.append({"axis": "index"})
+        if index_names is False:
+            hide_.append({"names": True, "axis": "index"})
+
+        render_kwargs_ = {
+            "hrules": True,
+            "sparse_index": sparsify,
+            "sparse_columns": sparsify,
+            "environment": "longtable" if longtable else None,
+            "multicol_align": multicolumn_format
+            if multicolumn
+            else f"naive-{multicolumn_format}",
+            "multirow_align": "t" if multirow else "naive",
+            "encoding": encoding,
+            "caption": caption,
+            "label": label,
+            "position": position,
+            "column_format": column_format,
+            "clines": "skip-last;data"
+            if (multirow and isinstance(self.index, MultiIndex))
+            else None,
+            "bold_rows": bold_rows,
+        }
+
+        return self._to_latex_via_styler(
+            buf,
+            hide=hide_,
+            relabel_index=relabel_index_,
+            format={"formatter": formatters_, **base_format_},
+            format_index=format_index_,
+            render_kwargs=render_kwargs_,
+        )
+
+    @final
+    def _to_latex_via_styler(
+        self,
+        buf=None,
+        *,
+        hide: dict | list[dict] | None = None,
+        relabel_index: dict | list[dict] | None = None,
+        format: dict | list[dict] | None = None,
+        format_index: dict | list[dict] | None = None,
+        render_kwargs: dict | None = None,
+    ):
+        """
+        Render object to a LaTeX tabular, longtable, or nested table.
+
+        Uses the ``Styler`` implementation with the following, ordered, method chaining:
+
+        .. code-block:: python
+           styler = Styler(DataFrame)
+           styler.hide(**hide)
+           styler.relabel_index(**relabel_index)
+           styler.format(**format)
+           styler.format_index(**format_index)
+           styler.to_latex(buf=buf, **render_kwargs)
+
+        Parameters
+        ----------
+        buf : str, Path or StringIO-like, optional, default None
+            Buffer to write to. If None, the output is returned as a string.
+        hide : dict, list of dict
+            Keyword args to pass to the method call of ``Styler.hide``. If a list will
+            call the method numerous times.
+        relabel_index : dict, list of dict
+            Keyword args to pass to the method of ``Styler.relabel_index``. If a list
+            will call the method numerous times.
+        format : dict, list of dict
+            Keyword args to pass to the method call of ``Styler.format``. If a list will
+            call the method numerous times.
+        format_index : dict, list of dict
+            Keyword args to pass to the method call of ``Styler.format_index``. If a
+            list will call the method numerous times.
+        render_kwargs : dict
+            Keyword args to pass to the method call of ``Styler.to_latex``.
+
+        Returns
+        -------
+        str or None
+            If buf is None, returns the result as a string. Otherwise returns None.
+        """
+        from pandas.io.formats.style import Styler
+
+        self = cast("DataFrame", self)
+        styler = Styler(self, uuid="")
+
+        for kw_name in ["hide", "relabel_index", "format", "format_index"]:
+            kw = vars()[kw_name]
+            if isinstance(kw, dict):
+                getattr(styler, kw_name)(**kw)
+            elif isinstance(kw, list):
+                for sub_kw in kw:
+                    getattr(styler, kw_name)(**sub_kw)
+
+        # bold_rows is not a direct kwarg of Styler.to_latex
+        render_kwargs = {} if render_kwargs is None else render_kwargs
+        if render_kwargs.pop("bold_rows"):
+            styler.map_index(lambda v: "textbf:--rwrap;")
+
+        return styler.to_latex(buf=buf, **render_kwargs)
+
+    @overload
+    def to_csv(
+        self,
+        path_or_buf: None = ...,
+        *,
+        sep: str = ...,
+        na_rep: str = ...,
+        float_format: str | Callable | None = ...,
+        columns: Sequence[Hashable] | None = ...,
+        header: bool | list[str] = ...,
+        index: bool = ...,
+        index_label: IndexLabel | None = ...,
+        mode: str = ...,
+        encoding: str | None = ...,
+        compression: CompressionOptions = ...,
+        quoting: int | None = ...,
+        quotechar: str = ...,
+        lineterminator: str | None = ...,
+        chunksize: int | None = ...,
+        date_format: str | None = ...,
+        doublequote: bool = ...,
+        escapechar: str | None = ...,
+        decimal: str = ...,
+        errors: OpenFileErrors = ...,
+        storage_options: StorageOptions = ...,
+    ) -> str: ...
+
+    @overload
+    def to_csv(
+        self,
+        path_or_buf: FilePath | WriteBuffer[bytes] | WriteBuffer[str],
+        *,
+        sep: str = ...,
+        na_rep: str = ...,
+        float_format: str | Callable | None = ...,
+        columns: Sequence[Hashable] | None = ...,
+        header: bool | list[str] = ...,
+        index: bool = ...,
+        index_label: IndexLabel | None = ...,
+        mode: str = ...,
+        encoding: str | None = ...,
+        compression: CompressionOptions = ...,
+        quoting: int | None = ...,
+        quotechar: str = ...,
+        lineterminator: str | None = ...,
+        chunksize: int | None = ...,
+        date_format: str | None = ...,
+        doublequote: bool = ...,
+        escapechar: str | None = ...,
+        decimal: str = ...,
+        errors: OpenFileErrors = ...,
+        storage_options: StorageOptions = ...,
+    ) -> None: ...
+
+    @final
+    @doc(
+        storage_options=_shared_docs["storage_options"],
+        compression_options=_shared_docs["compression_options"] % "path_or_buf",
+    )
+    def to_csv(
+        self,
+        path_or_buf: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = None,
+        *,
+        sep: str = ",",
+        na_rep: str = "",
+        float_format: str | Callable | None = None,
+        columns: Sequence[Hashable] | None = None,
+        header: bool | list[str] = True,
+        index: bool = True,
+        index_label: IndexLabel | None = None,
+        mode: str = "w",
+        encoding: str | None = None,
+        compression: CompressionOptions = "infer",
+        quoting: int | None = None,
+        quotechar: str = '"',
+        lineterminator: str | None = None,
+        chunksize: int | None = None,
+        date_format: str | None = None,
+        doublequote: bool = True,
+        escapechar: str | None = None,
+        decimal: str = ".",
+        errors: OpenFileErrors = "strict",
+        storage_options: StorageOptions | None = None,
+    ) -> str | None:
+        r"""
+        Write object to a comma-separated values (csv) file.
+
+        Parameters
+        ----------
+        path_or_buf : str, path object, file-like object, or None, default None
+            String, path object (implementing os.PathLike[str]), or file-like
+            object implementing a write() function. If None, the result is
+            returned as a string. If a non-binary file object is passed, it should
+            be opened with `newline=''`, disabling universal newlines. If a binary
+            file object is passed, `mode` might need to contain a `'b'`.
+        sep : str, default ','
+            String of length 1. Field delimiter for the output file.
+        na_rep : str, default ''
+            Missing data representation.
+        float_format : str, Callable, default None
+            Format string for floating point numbers. If a Callable is given, it takes
+            precedence over other numeric formatting parameters, like decimal.
+        columns : sequence, optional
+            Columns to write.
+        header : bool or list of str, default True
+            Write out the column names. If a list of strings is given it is
+            assumed to be aliases for the column names.
+        index : bool, default True
+            Write row names (index).
+        index_label : str or sequence, or False, default None
+            Column label for index column(s) if desired. If None is given, and
+            `header` and `index` are True, then the index names are used. A
+            sequence should be given if the object uses MultiIndex. If
+            False do not print fields for index names. Use index_label=False
+            for easier importing in R.
+        mode : {{'w', 'x', 'a'}}, default 'w'
+            Forwarded to either `open(mode=)` or `fsspec.open(mode=)` to control
+            the file opening. Typical values include:
+
+            - 'w', truncate the file first.
+            - 'x', exclusive creation, failing if the file already exists.
+            - 'a', append to the end of file if it exists.
+
+        encoding : str, optional
+            A string representing the encoding to use in the output file,
+            defaults to 'utf-8'. `encoding` is not supported if `path_or_buf`
+            is a non-binary file object.
+        {compression_options}
+
+               May be a dict with key 'method' as compression mode
+               and other entries as additional compression options if
+               compression mode is 'zip'.
+
+               Passing compression options as keys in dict is
+               supported for compression modes 'gzip', 'bz2', 'zstd', and 'zip'.
+        quoting : optional constant from csv module
+            Defaults to csv.QUOTE_MINIMAL. If you have set a `float_format`
+            then floats are converted to strings and thus csv.QUOTE_NONNUMERIC
+            will treat them as non-numeric.
+        quotechar : str, default '\"'
+            String of length 1. Character used to quote fields.
+        lineterminator : str, optional
+            The newline character or character sequence to use in the output
+            file. Defaults to `os.linesep`, which depends on the OS in which
+            this method is called ('\\n' for linux, '\\r\\n' for Windows, i.e.).
+
+            .. versionchanged:: 1.5.0
+
+                Previously was line_terminator, changed for consistency with
+                read_csv and the standard library 'csv' module.
+
+        chunksize : int or None
+            Rows to write at a time.
+        date_format : str, default None
+            Format string for datetime objects.
+        doublequote : bool, default True
+            Control quoting of `quotechar` inside a field.
+        escapechar : str, default None
+            String of length 1. Character used to escape `sep` and `quotechar`
+            when appropriate.
+        decimal : str, default '.'
+            Character recognized as decimal separator. E.g. use ',' for
+            European data.
+        errors : str, default 'strict'
+            Specifies how encoding and decoding errors are to be handled.
+            See the errors argument for :func:`open` for a full list
+            of options.
+
+        {storage_options}
+
+        Returns
+        -------
+        None or str
+            If path_or_buf is None, returns the resulting csv format as a
+            string. Otherwise returns None.
+
+        See Also
+        --------
+        read_csv : Load a CSV file into a DataFrame.
+        to_excel : Write DataFrame to an Excel file.
+
+        Examples
+        --------
+        Create 'out.csv' containing 'df' without indices
+
+        >>> df = pd.DataFrame(
+        ...     [["Raphael", "red", "sai"], ["Donatello", "purple", "bo staff"]],
+        ...     columns=["name", "mask", "weapon"],
+        ... )
+        >>> df.to_csv("out.csv", index=False)  # doctest: +SKIP
+
+        Create 'out.zip' containing 'out.csv'
+
+        >>> df.to_csv(index=False)
+        'name,mask,weapon\nRaphael,red,sai\nDonatello,purple,bo staff\n'
+        >>> compression_opts = dict(
+        ...     method="zip", archive_name="out.csv"
+        ... )  # doctest: +SKIP
+        >>> df.to_csv(
+        ...     "out.zip", index=False, compression=compression_opts
+        ... )  # doctest: +SKIP
+
+        To write a csv file to a new folder or nested folder you will first
+        need to create it using either Pathlib or os:
+
+        >>> from pathlib import Path  # doctest: +SKIP
+        >>> filepath = Path("folder/subfolder/out.csv")  # doctest: +SKIP
+        >>> filepath.parent.mkdir(parents=True, exist_ok=True)  # doctest: +SKIP
+        >>> df.to_csv(filepath)  # doctest: +SKIP
+
+        >>> import os  # doctest: +SKIP
+        >>> os.makedirs("folder/subfolder", exist_ok=True)  # doctest: +SKIP
+        >>> df.to_csv("folder/subfolder/out.csv")  # doctest: +SKIP
+
+        Format floats to two decimal places:
+
+        >>> df.to_csv("out1.csv", float_format="%.2f")  # doctest: +SKIP
+
+        Format floats using scientific notation:
+
+        >>> df.to_csv("out2.csv", float_format="{{:.2e}}".format)  # doctest: +SKIP
+        """
+        df = self if isinstance(self, ABCDataFrame) else self.to_frame()
+
+        formatter = DataFrameFormatter(
+            frame=df,
+            header=header,
+            index=index,
+            na_rep=na_rep,
+            float_format=float_format,
+            decimal=decimal,
+        )
+
+        return DataFrameRenderer(formatter).to_csv(
+            path_or_buf,
+            lineterminator=lineterminator,
+            sep=sep,
+            encoding=encoding,
+            errors=errors,
+            compression=compression,
+            quoting=quoting,
+            columns=columns,
+            index_label=index_label,
+            mode=mode,
+            chunksize=chunksize,
+            quotechar=quotechar,
+            date_format=date_format,
+            doublequote=doublequote,
+            escapechar=escapechar,
+            storage_options=storage_options,
+        )
+
+    # ----------------------------------------------------------------------
+    # Indexing Methods
+
+    @final
+    def take(self, indices, axis: Axis = 0, **kwargs) -> Self:
+        """
+        Return the elements in the given *positional* indices along an axis.
+
+        This means that we are not indexing according to actual values in
+        the index attribute of the object. We are indexing according to the
+        actual position of the element in the object.
+
+        Parameters
+        ----------
+        indices : array-like
+            An array of ints indicating which positions to take.
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            The axis on which to select elements. ``0`` means that we are
+            selecting rows, ``1`` means that we are selecting columns.
+            For `Series` this parameter is unused and defaults to 0.
+        **kwargs
+            For compatibility with :meth:`numpy.take`. Has no effect on the
+            output.
+
+        Returns
+        -------
+        same type as caller
+            An array-like containing the elements taken from the object.
+
+        See Also
+        --------
+        DataFrame.loc : Select a subset of a DataFrame by labels.
+        DataFrame.iloc : Select a subset of a DataFrame by positions.
+        numpy.take : Take elements from an array along an axis.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("falcon", "bird", 389.0),
+        ...         ("parrot", "bird", 24.0),
+        ...         ("lion", "mammal", 80.5),
+        ...         ("monkey", "mammal", np.nan),
+        ...     ],
+        ...     columns=["name", "class", "max_speed"],
+        ...     index=[0, 2, 3, 1],
+        ... )
+        >>> df
+             name   class  max_speed
+        0  falcon    bird      389.0
+        2  parrot    bird       24.0
+        3    lion  mammal       80.5
+        1  monkey  mammal        NaN
+
+        Take elements at positions 0 and 3 along the axis 0 (default).
+
+        Note how the actual indices selected (0 and 1) do not correspond to
+        our selected indices 0 and 3. That's because we are selecting the 0th
+        and 3rd rows, not rows whose indices equal 0 and 3.
+
+        >>> df.take([0, 3])
+             name   class  max_speed
+        0  falcon    bird      389.0
+        1  monkey  mammal        NaN
+
+        Take elements at indices 1 and 2 along the axis 1 (column selection).
+
+        >>> df.take([1, 2], axis=1)
+            class  max_speed
+        0    bird      389.0
+        2    bird       24.0
+        3  mammal       80.5
+        1  mammal        NaN
+
+        We may take elements using negative integers for positive indices,
+        starting from the end of the object, just like with Python lists.
+
+        >>> df.take([-1, -2])
+             name   class  max_speed
+        1  monkey  mammal        NaN
+        3    lion  mammal       80.5
+        """
+
+        nv.validate_take((), kwargs)
+
+        if isinstance(indices, slice):
+            raise TypeError(
+                f"{type(self).__name__}.take requires a sequence of integers, "
+                "not slice."
+            )
+        indices = np.asarray(indices, dtype=np.intp)
+        if axis == 0 and indices.ndim == 1 and is_range_indexer(indices, len(self)):
+            return self.copy(deep=False)
+
+        new_data = self._mgr.take(
+            indices,
+            axis=self._get_block_manager_axis(axis),
+            verify=True,
+        )
+        return self._constructor_from_mgr(new_data, axes=new_data.axes).__finalize__(
+            self, method="take"
+        )
+
+    @final
+    def xs(
+        self,
+        key: IndexLabel,
+        axis: Axis = 0,
+        level: IndexLabel | None = None,
+        drop_level: bool = True,
+    ) -> Self:
+        """
+        Return cross-section from the Series/DataFrame.
+
+        This method takes a `key` argument to select data at a particular
+        level of a MultiIndex.
+
+        Parameters
+        ----------
+        key : label or tuple of label
+            Label contained in the index, or partially in a MultiIndex.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Axis to retrieve cross-section on.
+        level : object, defaults to first n levels (n=1 or len(key))
+            In case of a key partially contained in a MultiIndex, indicate
+            which levels are used. Levels can be referred by label or position.
+        drop_level : bool, default True
+            If False, returns object with same levels as self.
+
+        Returns
+        -------
+        Series or DataFrame
+            Cross-section from the original Series or DataFrame
+            corresponding to the selected index levels.
+
+        See Also
+        --------
+        DataFrame.loc : Access a group of rows and columns
+            by label(s) or a boolean array.
+        DataFrame.iloc : Purely integer-location based indexing
+            for selection by position.
+
+        Notes
+        -----
+        `xs` can not be used to set values.
+
+        MultiIndex Slicers is a generic way to get/set values on
+        any level or levels.
+        It is a superset of `xs` functionality, see
+        :ref:`MultiIndex Slicers <advanced.mi_slicers>`.
+
+        Examples
+        --------
+        >>> d = {
+        ...     "num_legs": [4, 4, 2, 2],
+        ...     "num_wings": [0, 0, 2, 2],
+        ...     "class": ["mammal", "mammal", "mammal", "bird"],
+        ...     "animal": ["cat", "dog", "bat", "penguin"],
+        ...     "locomotion": ["walks", "walks", "flies", "walks"],
+        ... }
+        >>> df = pd.DataFrame(data=d)
+        >>> df = df.set_index(["class", "animal", "locomotion"])
+        >>> df
+                                   num_legs  num_wings
+        class  animal  locomotion
+        mammal cat     walks              4          0
+               dog     walks              4          0
+               bat     flies              2          2
+        bird   penguin walks              2          2
+
+        Get values at specified index
+
+        >>> df.xs("mammal")
+                           num_legs  num_wings
+        animal locomotion
+        cat    walks              4          0
+        dog    walks              4          0
+        bat    flies              2          2
+
+        Get values at several indexes
+
+        >>> df.xs(("mammal", "dog", "walks"))
+        num_legs     4
+        num_wings    0
+        Name: (mammal, dog, walks), dtype: int64
+
+        Get values at specified index and level
+
+        >>> df.xs("cat", level=1)
+                           num_legs  num_wings
+        class  locomotion
+        mammal walks              4          0
+
+        Get values at several indexes and levels
+
+        >>> df.xs(("bird", "walks"), level=[0, "locomotion"])
+                 num_legs  num_wings
+        animal
+        penguin         2          2
+
+        Get values at specified column and axis
+
+        >>> df.xs("num_wings", axis=1)
+        class   animal   locomotion
+        mammal  cat      walks         0
+                dog      walks         0
+                bat      flies         2
+        bird    penguin  walks         2
+        Name: num_wings, dtype: int64
+        """
+        axis = self._get_axis_number(axis)
+        labels = self._get_axis(axis)
+
+        if isinstance(key, list):
+            raise TypeError("list keys are not supported in xs, pass a tuple instead")
+
+        if level is not None:
+            if not isinstance(labels, MultiIndex):
+                raise TypeError("Index must be a MultiIndex")
+            loc, new_ax = labels.get_loc_level(key, level=level, drop_level=drop_level)
+
+            # create the tuple of the indexer
+            _indexer = [slice(None)] * self.ndim
+            _indexer[axis] = loc
+            indexer = tuple(_indexer)
+
+            result = self.iloc[indexer]
+            setattr(result, result._get_axis_name(axis), new_ax)
+            return result
+
+        if axis == 1:
+            if drop_level:
+                return self[key]
+            index = self.columns
+        else:
+            index = self.index
+
+        if isinstance(index, MultiIndex):
+            loc, new_index = index._get_loc_level(key, level=0)
+            if not drop_level:
+                if lib.is_integer(loc):
+                    # Slice index must be an integer or None
+                    new_index = index[loc : loc + 1]
+                else:
+                    new_index = index[loc]
+        else:
+            loc = index.get_loc(key)
+
+            if isinstance(loc, np.ndarray):
+                if loc.dtype == np.bool_:
+                    (inds,) = loc.nonzero()
+                    return self.take(inds, axis=axis)
+                else:
+                    return self.take(loc, axis=axis)
+
+            if not is_scalar(loc):
+                new_index = index[loc]
+
+        if is_scalar(loc) and axis == 0:
+            # In this case loc should be an integer
+            if self.ndim == 1:
+                # if we encounter an array-like and we only have 1 dim
+                # that means that their are list/ndarrays inside the Series!
+                # so just return them (GH 6394)
+                return self._values[loc]
+
+            new_mgr = self._mgr.fast_xs(loc)
+
+            result = self._constructor_sliced_from_mgr(new_mgr, axes=new_mgr.axes)
+            result._name = self.index[loc]
+            result = result.__finalize__(self)
+        elif is_scalar(loc):
+            result = self.iloc[:, slice(loc, loc + 1)]
+        elif axis == 1:
+            result = self.iloc[:, loc]
+        else:
+            result = self.iloc[loc]
+            result.index = new_index
+
+        return result
+
+    def __getitem__(self, item):
+        raise AbstractMethodError(self)
+
+    @final
+    def _getitem_slice(self, key: slice) -> Self:
+        """
+        __getitem__ for the case where the key is a slice object.
+        """
+        # _convert_slice_indexer to determine if this slice is positional
+        #  or label based, and if the latter, convert to positional
+        slobj = self.index._convert_slice_indexer(key, kind="getitem")
+        if isinstance(slobj, np.ndarray):
+            # reachable with DatetimeIndex
+            indexer = lib.maybe_indices_to_slice(slobj.astype(np.intp), len(self))
+            if isinstance(indexer, np.ndarray):
+                # GH#43223 If we can not convert, use take
+                return self.take(indexer, axis=0)
+            slobj = indexer
+        return self._slice(slobj)
+
+    def _slice(self, slobj: slice, axis: AxisInt = 0) -> Self:
+        """
+        Construct a slice of this container.
+
+        Slicing with this method is *always* positional.
+        """
+        assert isinstance(slobj, slice), type(slobj)
+        axis = self._get_block_manager_axis(axis)
+        new_mgr = self._mgr.get_slice(slobj, axis=axis)
+        result = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+        result = result.__finalize__(self)
+        return result
+
+    @final
+    def __delitem__(self, key) -> None:
+        """
+        Delete item
+        """
+        deleted = False
+
+        maybe_shortcut = False
+        if self.ndim == 2 and isinstance(self.columns, MultiIndex):
+            try:
+                # By using engine's __contains__ we effectively
+                # restrict to same-length tuples
+                maybe_shortcut = key not in self.columns._engine
+            except TypeError:
+                pass
+
+        if maybe_shortcut:
+            # Allow shorthand to delete all columns whose first len(key)
+            # elements match key:
+            if not isinstance(key, tuple):
+                key = (key,)
+            for col in self.columns:
+                if isinstance(col, tuple) and col[: len(key)] == key:
+                    del self[col]
+                    deleted = True
+        if not deleted:
+            # If the above loop ran and didn't delete anything because
+            # there was no match, this call should raise the appropriate
+            # exception:
+            loc = self.axes[-1].get_loc(key)
+            self._mgr = self._mgr.idelete(loc)
+
+    # ----------------------------------------------------------------------
+    # Unsorted
+
+    @final
+    def _check_inplace_and_allows_duplicate_labels(self, inplace: bool) -> None:
+        if inplace and not self.flags.allows_duplicate_labels:
+            raise ValueError(
+                "Cannot specify 'inplace=True' when "
+                "'self.flags.allows_duplicate_labels' is False."
+            )
+
+    @final
+    def get(self, key, default=None):
+        """
+        Get item from object for given key (ex: DataFrame column).
+
+        Returns ``default`` value if not found.
+
+        Parameters
+        ----------
+        key : object
+            Key for which item should be returned.
+        default : object, default None
+            Default value to return if key is not found.
+
+        Returns
+        -------
+        same type as items contained in object
+            Item for given key or ``default`` value, if key is not found.
+
+        See Also
+        --------
+        DataFrame.get : Get item from object for given key (ex: DataFrame column).
+        Series.get : Get item from object for given key (ex: DataFrame column).
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         [24.3, 75.7, "high"],
+        ...         [31, 87.8, "high"],
+        ...         [22, 71.6, "medium"],
+        ...         [35, 95, "medium"],
+        ...     ],
+        ...     columns=["temp_celsius", "temp_fahrenheit", "windspeed"],
+        ...     index=pd.date_range(start="2014-02-12", end="2014-02-15", freq="D"),
+        ... )
+
+        >>> df
+                    temp_celsius  temp_fahrenheit windspeed
+        2014-02-12          24.3             75.7      high
+        2014-02-13          31.0             87.8      high
+        2014-02-14          22.0             71.6    medium
+        2014-02-15          35.0             95.0    medium
+
+        >>> df.get(["temp_celsius", "windspeed"])
+                    temp_celsius windspeed
+        2014-02-12          24.3      high
+        2014-02-13          31.0      high
+        2014-02-14          22.0    medium
+        2014-02-15          35.0    medium
+
+        >>> ser = df["windspeed"]
+        >>> ser.get("2014-02-13")
+        'high'
+
+        If the key isn't found, the default value will be used.
+
+        >>> df.get(["temp_celsius", "temp_kelvin"], default="default_value")
+        'default_value'
+
+        >>> ser.get("2014-02-10", "[unknown]")
+        '[unknown]'
+        """
+        try:
+            return self[key]
+        except (KeyError, ValueError, IndexError):
+            return default
+
+    @staticmethod
+    def _check_copy_deprecation(copy):
+        if copy is not lib.no_default:
+            warnings.warn(
+                "The copy keyword is deprecated and will be removed in a future "
+                "version. Copy-on-Write is active in pandas since 3.0 which utilizes "
+                "a lazy copy mechanism that defers copies until necessary. Use "
+                ".copy() to make an eager copy if necessary.",
+                DeprecationWarning,
+                stacklevel=find_stack_level(),
+            )
+
+    # issue 58667
+    @deprecate_kwarg("method", None)
+    @final
+    def reindex_like(
+        self,
+        other,
+        method: Literal["backfill", "bfill", "pad", "ffill", "nearest"] | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> Self:
+        """
+        Return an object with matching indices as other object.
+
+        Conform the object to the same index on all axes. Optional
+        filling logic, placing NaN in locations having no value
+        in the previous index. A new object is produced unless the
+        new index is equivalent to the current one and copy=False.
+
+        Parameters
+        ----------
+        other : Object of the same data type
+            Its row and column indices are used to define the new indices
+            of this object.
+        method : {None, 'backfill'/'bfill', 'pad'/'ffill', 'nearest'}
+            Method to use for filling holes in reindexed DataFrame.
+            Please note: this is only applicable to DataFrames/Series with a
+            monotonically increasing/decreasing index.
+
+            .. deprecated:: 3.0.0
+
+            * None (default): don't fill gaps
+            * pad / ffill: propagate last valid observation forward to next
+              valid
+            * backfill / bfill: use next valid observation to fill gap
+            * nearest: use nearest valid observations to fill gap.
+
+        copy : bool, default False
+            Return a new object, even if the passed indexes are the same.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        limit : int, default None
+            Maximum number of consecutive labels to fill for inexact matches.
+        tolerance : optional
+            Maximum distance between original and new labels for inexact
+            matches. The values of the index at the matching locations must
+            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
+
+            Tolerance may be a scalar value, which applies the same tolerance
+            to all values, or list-like, which applies variable tolerance per
+            element. List-like includes list, tuple, array, Series, and must be
+            the same size as the index and its dtype must exactly match the
+            index's type.
+
+        Returns
+        -------
+        Series or DataFrame
+            Same type as caller, but with changed indices on each axis.
+
+        See Also
+        --------
+        DataFrame.set_index : Set row labels.
+        DataFrame.reset_index : Remove row labels or move them to new columns.
+        DataFrame.reindex : Change to new indices or expand indices.
+
+        Notes
+        -----
+        Same as calling
+        ``.reindex(index=other.index, columns=other.columns,...)``.
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame(
+        ...     [
+        ...         [24.3, 75.7, "high"],
+        ...         [31, 87.8, "high"],
+        ...         [22, 71.6, "medium"],
+        ...         [35, 95, "medium"],
+        ...     ],
+        ...     columns=["temp_celsius", "temp_fahrenheit", "windspeed"],
+        ...     index=pd.date_range(start="2014-02-12", end="2014-02-15", freq="D"),
+        ... )
+
+        >>> df1
+                    temp_celsius  temp_fahrenheit windspeed
+        2014-02-12          24.3             75.7      high
+        2014-02-13          31.0             87.8      high
+        2014-02-14          22.0             71.6    medium
+        2014-02-15          35.0             95.0    medium
+
+        >>> df2 = pd.DataFrame(
+        ...     [[28, "low"], [30, "low"], [35.1, "medium"]],
+        ...     columns=["temp_celsius", "windspeed"],
+        ...     index=pd.DatetimeIndex(["2014-02-12", "2014-02-13", "2014-02-15"]),
+        ... )
+
+        >>> df2
+                    temp_celsius windspeed
+        2014-02-12          28.0       low
+        2014-02-13          30.0       low
+        2014-02-15          35.1    medium
+
+        >>> df2.reindex_like(df1)
+                    temp_celsius  temp_fahrenheit windspeed
+        2014-02-12          28.0              NaN       low
+        2014-02-13          30.0              NaN       low
+        2014-02-14           NaN              NaN       NaN
+        2014-02-15          35.1              NaN    medium
+        """
+        self._check_copy_deprecation(copy)
+        d = other._construct_axes_dict(
+            axes=self._AXIS_ORDERS,
+            method=method,
+            limit=limit,
+            tolerance=tolerance,
+        )
+
+        return self.reindex(**d)
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: Literal[True],
+        errors: IgnoreRaise = ...,
+    ) -> None: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: Literal[False] = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Self: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: bool = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Self | None: ...
+
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = None,
+        *,
+        axis: Axis = 0,
+        index: IndexLabel | ListLike = None,
+        columns: IndexLabel | ListLike = None,
+        level: Level | None = None,
+        inplace: bool = False,
+        errors: IgnoreRaise = "raise",
+    ) -> Self | None:
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if labels is not None:
+            if index is not None or columns is not None:
+                raise ValueError("Cannot specify both 'labels' and 'index'/'columns'")
+            axis_name = self._get_axis_name(axis)
+            axes = {axis_name: labels}
+        elif index is not None or columns is not None:
+            axes = {"index": index}
+            if self.ndim == 2:
+                axes["columns"] = columns
+        else:
+            raise ValueError(
+                "Need to specify at least one of 'labels', 'index' or 'columns'"
+            )
+
+        obj = self
+
+        for axis, labels in axes.items():
+            if labels is not None:
+                obj = obj._drop_axis(labels, axis, level=level, errors=errors)
+
+        if inplace:
+            self._update_inplace(obj)
+            return None
+        else:
+            return obj
+
+    @final
+    def _drop_axis(
+        self,
+        labels,
+        axis,
+        level=None,
+        errors: IgnoreRaise = "raise",
+        only_slice: bool = False,
+    ) -> Self:
+        """
+        Drop labels from specified axis. Used in the ``drop`` method
+        internally.
+
+        Parameters
+        ----------
+        labels : single label or list-like
+        axis : int or axis name
+        level : int or level name, default None
+            For MultiIndex
+        errors : {'ignore', 'raise'}, default 'raise'
+            If 'ignore', suppress error and existing labels are dropped.
+        only_slice : bool, default False
+            Whether indexing along columns should be view-only.
+
+        """
+        axis_num = self._get_axis_number(axis)
+        axis = self._get_axis(axis)
+
+        if axis.is_unique:
+            if level is not None:
+                if not isinstance(axis, MultiIndex):
+                    raise AssertionError("axis must be a MultiIndex")
+                new_axis = axis.drop(labels, level=level, errors=errors)
+            else:
+                new_axis = axis.drop(labels, errors=errors)
+            indexer = axis.get_indexer(new_axis)
+
+        # Case for non-unique axis
+        else:
+            is_tuple_labels = is_nested_list_like(labels) or isinstance(labels, tuple)
+            labels = ensure_object(common.index_labels_to_array(labels))
+            if level is not None:
+                if not isinstance(axis, MultiIndex):
+                    raise AssertionError("axis must be a MultiIndex")
+                mask = ~axis.get_level_values(level).isin(labels)
+
+                # GH 18561 MultiIndex.drop should raise if label is absent
+                if errors == "raise" and mask.all():
+                    raise KeyError(f"{labels} not found in axis")
+            elif (
+                isinstance(axis, MultiIndex)
+                and labels.dtype == "object"
+                and not is_tuple_labels
+            ):
+                # Set level to zero in case of MultiIndex and label is string,
+                #  because isin can't handle strings for MultiIndexes GH#36293
+                # In case of tuples we get dtype object but have to use isin GH#42771
+                mask = ~axis.get_level_values(0).isin(labels)
+            else:
+                mask = ~axis.isin(labels)
+                # Check if label doesn't exist along axis
+                labels_missing = (axis.get_indexer_for(labels) == -1).any()
+                if errors == "raise" and labels_missing:
+                    raise KeyError(f"{labels} not found in axis")
+
+            if isinstance(mask.dtype, ExtensionDtype):
+                # GH#45860
+                mask = mask.to_numpy(dtype=bool)
+
+            indexer = mask.nonzero()[0]
+            new_axis = axis.take(indexer)
+
+        bm_axis = self.ndim - axis_num - 1
+        new_mgr = self._mgr.reindex_indexer(
+            new_axis,
+            indexer,
+            axis=bm_axis,
+            allow_dups=True,
+            only_slice=only_slice,
+        )
+        result = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+        if self.ndim == 1:
+            result._name = self.name
+
+        return result.__finalize__(self)
+
+    @final
+    def _update_inplace(self, result) -> None:
+        """
+        Replace self internals with result.
+
+        Parameters
+        ----------
+        result : same type as self
+        """
+        # NOTE: This does *not* call __finalize__ and that's an explicit
+        # decision that we may revisit in the future.
+        self._mgr = result._mgr
+
+    @final
+    def add_prefix(self, prefix: str, axis: Axis | None = None) -> Self:
+        """
+        Prefix labels with string `prefix`.
+
+        For Series, the row labels are prefixed.
+        For DataFrame, the column labels are prefixed.
+
+        Parameters
+        ----------
+        prefix : str
+            The string to add before each label.
+        axis : {0 or 'index', 1 or 'columns', None}, default None
+            Axis to add prefix on
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            New Series or DataFrame with updated labels.
+
+        See Also
+        --------
+        Series.add_suffix: Suffix row labels with string `suffix`.
+        DataFrame.add_suffix: Suffix column labels with string `suffix`.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s
+        0    1
+        1    2
+        2    3
+        3    4
+        dtype: int64
+
+        >>> s.add_prefix("item_")
+        item_0    1
+        item_1    2
+        item_2    3
+        item_3    4
+        dtype: int64
+
+        >>> df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [3, 4, 5, 6]})
+        >>> df
+           A  B
+        0  1  3
+        1  2  4
+        2  3  5
+        3  4  6
+
+        >>> df.add_prefix("col_")
+             col_A  col_B
+        0       1       3
+        1       2       4
+        2       3       5
+        3       4       6
+        """
+        f = lambda x: f"{prefix}{x}"
+
+        axis_name = self._info_axis_name
+        if axis is not None:
+            axis_name = self._get_axis_name(axis)
+
+        mapper = {axis_name: f}
+
+        # error: Keywords must be strings
+        # error: No overload variant of "_rename" of "NDFrame" matches
+        # argument type "dict[Literal['index', 'columns'], Callable[[Any], str]]"
+        return self._rename(**mapper)  # type: ignore[call-overload, misc]
+
+    @final
+    def add_suffix(self, suffix: str, axis: Axis | None = None) -> Self:
+        """
+        Suffix labels with string `suffix`.
+
+        For Series, the row labels are suffixed.
+        For DataFrame, the column labels are suffixed.
+
+        Parameters
+        ----------
+        suffix : str
+            The string to add after each label.
+        axis : {0 or 'index', 1 or 'columns', None}, default None
+            Axis to add suffix on
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            New Series or DataFrame with updated labels.
+
+        See Also
+        --------
+        Series.add_prefix: Prefix row labels with string `prefix`.
+        DataFrame.add_prefix: Prefix column labels with string `prefix`.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s
+        0    1
+        1    2
+        2    3
+        3    4
+        dtype: int64
+
+        >>> s.add_suffix("_item")
+        0_item    1
+        1_item    2
+        2_item    3
+        3_item    4
+        dtype: int64
+
+        >>> df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [3, 4, 5, 6]})
+        >>> df
+           A  B
+        0  1  3
+        1  2  4
+        2  3  5
+        3  4  6
+
+        >>> df.add_suffix("_col")
+             A_col  B_col
+        0       1       3
+        1       2       4
+        2       3       5
+        3       4       6
+        """
+        f = lambda x: f"{x}{suffix}"
+
+        axis_name = self._info_axis_name
+        if axis is not None:
+            axis_name = self._get_axis_name(axis)
+
+        mapper = {axis_name: f}
+        # error: Keywords must be strings
+        # error: No overload variant of "_rename" of "NDFrame" matches argument
+        # type "dict[Literal['index', 'columns'], Callable[[Any], str]]"
+        return self._rename(**mapper)  # type: ignore[call-overload, misc]
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> Self: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> None: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: bool = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> Self | None: ...
+
+    def sort_values(
+        self,
+        *,
+        axis: Axis = 0,
+        ascending: bool | Sequence[bool] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: NaPosition = "last",
+        ignore_index: bool = False,
+        key: ValueKeyFunc | None = None,
+    ) -> Self | None:
+        """
+        Sort by the values along either axis.
+
+        Parameters
+        ----------%(optional_by)s
+        axis : %(axes_single_arg)s, default 0
+             Axis to be sorted.
+        ascending : bool or list of bool, default True
+             Sort ascending vs. descending. Specify list for multiple sort
+             orders.  If this is a list of bools, must match the length of
+             the by.
+        inplace : bool, default False
+             If True, perform operation in-place.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+             Choice of sorting algorithm. See also :func:`numpy.sort` for more
+             information. `mergesort` and `stable` are the only stable algorithms. For
+             DataFrames, this option is only applied when sorting on a single
+             column or label.
+        na_position : {'first', 'last'}, default 'last'
+             Puts NaNs at the beginning if `first`; `last` puts NaNs at the
+             end.
+        ignore_index : bool, default False
+             If True, the resulting axis will be labeled 0, 1, …, n - 1.
+        key : callable, optional
+            Apply the key function to the values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect a
+            ``Series`` and return a Series with the same shape as the input.
+            It will be applied to each column in `by` independently. The values in the
+            returned Series will be used as the keys for sorting.
+
+        Returns
+        -------
+        DataFrame or None
+            DataFrame with sorted values or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.sort_index : Sort a DataFrame by the index.
+        Series.sort_values : Similar method for a Series.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "col1": ["A", "A", "B", np.nan, "D", "C"],
+        ...         "col2": [2, 1, 9, 8, 7, 4],
+        ...         "col3": [0, 1, 9, 4, 2, 3],
+        ...         "col4": ["a", "B", "c", "D", "e", "F"],
+        ...     }
+        ... )
+        >>> df
+          col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+
+        Sort by col1
+
+        >>> df.sort_values(by=["col1"])
+          col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        5    C     4     3    F
+        4    D     7     2    e
+        3  NaN     8     4    D
+
+        Sort by multiple columns
+
+        >>> df.sort_values(by=["col1", "col2"])
+          col1  col2  col3 col4
+        1    A     1     1    B
+        0    A     2     0    a
+        2    B     9     9    c
+        5    C     4     3    F
+        4    D     7     2    e
+        3  NaN     8     4    D
+
+        Sort Descending
+
+        >>> df.sort_values(by="col1", ascending=False)
+          col1  col2  col3 col4
+        4    D     7     2    e
+        5    C     4     3    F
+        2    B     9     9    c
+        0    A     2     0    a
+        1    A     1     1    B
+        3  NaN     8     4    D
+
+        Putting NAs first
+
+        >>> df.sort_values(by="col1", ascending=False, na_position="first")
+          col1  col2  col3 col4
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+        2    B     9     9    c
+        0    A     2     0    a
+        1    A     1     1    B
+
+        Sorting with a key function
+
+        >>> df.sort_values(by="col4", key=lambda col: col.str.lower())
+           col1  col2  col3 col4
+        0    A     2     0    a
+        1    A     1     1    B
+        2    B     9     9    c
+        3  NaN     8     4    D
+        4    D     7     2    e
+        5    C     4     3    F
+
+        Natural sort with the key argument,
+        using the `natsort <https://github.com/SethMMorton/natsort>` package.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "time": ["0hr", "128hr", "72hr", "48hr", "96hr"],
+        ...         "value": [10, 20, 30, 40, 50],
+        ...     }
+        ... )
+        >>> df
+            time  value
+        0    0hr     10
+        1  128hr     20
+        2   72hr     30
+        3   48hr     40
+        4   96hr     50
+        >>> from natsort import index_natsorted
+        >>> df.sort_values(
+        ...     by="time", key=lambda x: np.argsort(index_natsorted(df["time"]))
+        ... )
+            time  value
+        0    0hr     10
+        3   48hr     40
+        2   72hr     30
+        4   96hr     50
+        1  128hr     20
+        """
+        raise AbstractMethodError(self)
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> None: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> Self: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: bool = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> Self | None: ...
+
+    def sort_index(
+        self,
+        *,
+        axis: Axis = 0,
+        level: IndexLabel | None = None,
+        ascending: bool | Sequence[bool] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: NaPosition = "last",
+        sort_remaining: bool = True,
+        ignore_index: bool = False,
+        key: IndexKeyFunc | None = None,
+    ) -> Self | None:
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        axis = self._get_axis_number(axis)
+        ascending = validate_ascending(ascending)
+
+        target = self._get_axis(axis)
+
+        indexer = get_indexer_indexer(
+            target, level, ascending, kind, na_position, sort_remaining, key
+        )
+
+        if indexer is None:
+            if inplace:
+                result = self
+            else:
+                result = self.copy(deep=False)
+
+            if ignore_index:
+                if axis == 1:
+                    result.columns = default_index(len(self.columns))
+                else:
+                    result.index = default_index(len(self))
+            if inplace:
+                return None
+            else:
+                return result
+
+        baxis = self._get_block_manager_axis(axis)
+        new_data = self._mgr.take(indexer, axis=baxis, verify=False)
+
+        # reconstruct axis if needed
+        if not ignore_index:
+            new_axis = new_data.axes[baxis]._sort_levels_monotonic()
+        else:
+            new_axis = default_index(len(indexer))
+        new_data.set_axis(baxis, new_axis)
+
+        result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="sort_index")
+
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        optional_reindex="",
+    )
+    def reindex(
+        self,
+        labels=None,
+        *,
+        index=None,
+        columns=None,
+        axis: Axis | None = None,
+        method: ReindexMethod | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        level: Level | None = None,
+        fill_value: Scalar | None = np.nan,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> Self:
+        """
+        Conform {klass} to new index with optional filling logic.
+
+        Places NA/NaN in locations having no value in the previous index. A new object
+        is produced unless the new index is equivalent to the current one and
+        ``copy=False``.
+
+        Parameters
+        ----------
+        {optional_reindex}
+        method : {{None, 'backfill'/'bfill', 'pad'/'ffill', 'nearest'}}
+            Method to use for filling holes in reindexed DataFrame.
+            Please note: this is only applicable to DataFrames/Series with a
+            monotonically increasing/decreasing index.
+
+            * None (default): don't fill gaps
+            * pad / ffill: Propagate last valid observation forward to next
+              valid.
+            * backfill / bfill: Use next valid observation to fill gap.
+            * nearest: Use nearest valid observations to fill gap.
+
+        copy : bool, default False
+            Return a new object, even if the passed indexes are the same.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : scalar, default np.nan
+            Value to use for missing values. Defaults to NaN, but can be any
+            "compatible" value.
+        limit : int, default None
+            Maximum number of consecutive elements to forward or backward fill.
+        tolerance : optional
+            Maximum distance between original and new labels for inexact
+            matches. The values of the index at the matching locations most
+            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
+
+            Tolerance may be a scalar value, which applies the same tolerance
+            to all values, or list-like, which applies variable tolerance per
+            element. List-like includes list, tuple, array, Series, and must be
+            the same size as the index and its dtype must exactly match the
+            index's type.
+
+        Returns
+        -------
+        {klass}
+            {klass} with changed index.
+
+        See Also
+        --------
+        DataFrame.set_index : Set row labels.
+        DataFrame.reset_index : Remove row labels or move them to new columns.
+        DataFrame.reindex_like : Change to same indices as other DataFrame.
+
+        Examples
+        --------
+        ``DataFrame.reindex`` supports two calling conventions
+
+        * ``(index=index_labels, columns=column_labels, ...)``
+        * ``(labels, axis={{'index', 'columns'}}, ...)``
+
+        We *highly* recommend using keyword arguments to clarify your
+        intent.
+
+        Create a DataFrame with some fictional data.
+
+        >>> index = ["Firefox", "Chrome", "Safari", "IE10", "Konqueror"]
+        >>> columns = ["http_status", "response_time"]
+        >>> df = pd.DataFrame(
+        ...     [[200, 0.04], [200, 0.02], [404, 0.07], [404, 0.08], [301, 1.0]],
+        ...     columns=columns,
+        ...     index=index,
+        ... )
+        >>> df
+                   http_status  response_time
+        Firefox            200           0.04
+        Chrome             200           0.02
+        Safari             404           0.07
+        IE10               404           0.08
+        Konqueror          301           1.00
+
+        Create a new index and reindex the DataFrame. By default
+        values in the new index that do not have corresponding
+        records in the DataFrame are assigned ``NaN``.
+
+        >>> new_index = ["Safari", "Iceweasel", "Comodo Dragon", "IE10", "Chrome"]
+        >>> df.reindex(new_index)
+                       http_status  response_time
+        Safari               404.0           0.07
+        Iceweasel              NaN            NaN
+        Comodo Dragon          NaN            NaN
+        IE10                 404.0           0.08
+        Chrome               200.0           0.02
+
+        We can fill in the missing values by passing a value to
+        the keyword ``fill_value``. Because the index is not monotonically
+        increasing or decreasing, we cannot use arguments to the keyword
+        ``method`` to fill the ``NaN`` values.
+
+        >>> df.reindex(new_index, fill_value=0)
+                       http_status  response_time
+        Safari                 404           0.07
+        Iceweasel                0           0.00
+        Comodo Dragon            0           0.00
+        IE10                   404           0.08
+        Chrome                 200           0.02
+
+        >>> df.reindex(new_index, fill_value="missing")
+                      http_status response_time
+        Safari                404          0.07
+        Iceweasel         missing       missing
+        Comodo Dragon     missing       missing
+        IE10                  404          0.08
+        Chrome                200          0.02
+
+        We can also reindex the columns.
+
+        >>> df.reindex(columns=["http_status", "user_agent"])
+                   http_status  user_agent
+        Firefox            200         NaN
+        Chrome             200         NaN
+        Safari             404         NaN
+        IE10               404         NaN
+        Konqueror          301         NaN
+
+        Or we can use "axis-style" keyword arguments
+
+        >>> df.reindex(["http_status", "user_agent"], axis="columns")
+                   http_status  user_agent
+        Firefox            200         NaN
+        Chrome             200         NaN
+        Safari             404         NaN
+        IE10               404         NaN
+        Konqueror          301         NaN
+
+        To further illustrate the filling functionality in
+        ``reindex``, we will create a DataFrame with a
+        monotonically increasing index (for example, a sequence
+        of dates).
+
+        >>> date_index = pd.date_range("1/1/2010", periods=6, freq="D")
+        >>> df2 = pd.DataFrame(
+        ...     {{"prices": [100, 101, np.nan, 100, 89, 88]}}, index=date_index
+        ... )
+        >>> df2
+                    prices
+        2010-01-01   100.0
+        2010-01-02   101.0
+        2010-01-03     NaN
+        2010-01-04   100.0
+        2010-01-05    89.0
+        2010-01-06    88.0
+
+        Suppose we decide to expand the DataFrame to cover a wider
+        date range.
+
+        >>> date_index2 = pd.date_range("12/29/2009", periods=10, freq="D")
+        >>> df2.reindex(date_index2)
+                    prices
+        2009-12-29     NaN
+        2009-12-30     NaN
+        2009-12-31     NaN
+        2010-01-01   100.0
+        2010-01-02   101.0
+        2010-01-03     NaN
+        2010-01-04   100.0
+        2010-01-05    89.0
+        2010-01-06    88.0
+        2010-01-07     NaN
+
+        The index entries that did not have a value in the original data frame
+        (for example, '2009-12-29') are by default filled with ``NaN``.
+        If desired, we can fill in the missing values using one of several
+        options.
+
+        For example, to back-propagate the last valid value to fill the ``NaN``
+        values, pass ``bfill`` as an argument to the ``method`` keyword.
+
+        >>> df2.reindex(date_index2, method="bfill")
+                    prices
+        2009-12-29   100.0
+        2009-12-30   100.0
+        2009-12-31   100.0
+        2010-01-01   100.0
+        2010-01-02   101.0
+        2010-01-03     NaN
+        2010-01-04   100.0
+        2010-01-05    89.0
+        2010-01-06    88.0
+        2010-01-07     NaN
+
+        Please note that the ``NaN`` value present in the original DataFrame
+        (at index value 2010-01-03) will not be filled by any of the
+        value propagation schemes. This is because filling while reindexing
+        does not look at DataFrame values, but only compares the original and
+        desired indexes. If you do want to fill in the ``NaN`` values present
+        in the original DataFrame, use the ``fillna()`` method.
+
+        See the :ref:`user guide <basics.reindexing>` for more.
+        """
+        # TODO: Decide if we care about having different examples for different
+        # kinds
+        self._check_copy_deprecation(copy)
+
+        if index is not None and columns is not None and labels is not None:
+            raise TypeError("Cannot specify all of 'labels', 'index', 'columns'.")
+        elif index is not None or columns is not None:
+            if axis is not None:
+                raise TypeError(
+                    "Cannot specify both 'axis' and any of 'index' or 'columns'"
+                )
+            if labels is not None:
+                if index is not None:
+                    columns = labels
+                else:
+                    index = labels
+        else:
+            if axis and self._get_axis_number(axis) == 1:
+                columns = labels
+            else:
+                index = labels
+        axes: dict[Literal["index", "columns"], Any] = {
+            "index": index,
+            "columns": columns,
+        }
+        method = clean_reindex_fill_method(method)
+
+        # if all axes that are requested to reindex are equal, then only copy
+        # if indicated must have index names equal here as well as values
+        if all(
+            self._get_axis(axis_name).identical(ax)
+            for axis_name, ax in axes.items()
+            if ax is not None
+        ):
+            return self.copy(deep=False)
+
+        # check if we are a multi reindex
+        if self._needs_reindex_multi(axes, method, level):
+            return self._reindex_multi(axes, fill_value)
+
+        # perform the reindex on the axes
+        return self._reindex_axes(
+            axes, level, limit, tolerance, method, fill_value
+        ).__finalize__(self, method="reindex")
+
+    @final
+    def _reindex_axes(
+        self,
+        axes,
+        level: Level | None,
+        limit: int | None,
+        tolerance,
+        method,
+        fill_value: Scalar | None,
+    ) -> Self:
+        """Perform the reindex for all the axes."""
+        obj = self
+        for a in self._AXIS_ORDERS:
+            labels = axes[a]
+            if labels is None:
+                continue
+
+            ax = self._get_axis(a)
+            new_index, indexer = ax.reindex(
+                labels, level=level, limit=limit, tolerance=tolerance, method=method
+            )
+
+            axis = self._get_axis_number(a)
+            obj = obj._reindex_with_indexers(
+                {axis: [new_index, indexer]},
+                fill_value=fill_value,
+                allow_dups=False,
+            )
+
+        return obj
+
+    def _needs_reindex_multi(self, axes, method, level: Level | None) -> bool:
+        """Check if we do need a multi reindex."""
+        return (
+            (common.count_not_none(*axes.values()) == self._AXIS_LEN)
+            and method is None
+            and level is None
+            # reindex_multi calls self.values, so we only want to go
+            #  down that path when doing so is cheap.
+            and self._can_fast_transpose
+        )
+
+    def _reindex_multi(self, axes, fill_value):
+        raise AbstractMethodError(self)
+
+    @final
+    def _reindex_with_indexers(
+        self,
+        reindexers,
+        fill_value=None,
+        allow_dups: bool = False,
+    ) -> Self:
+        """allow_dups indicates an internal call here"""
+        # reindex doing multiple operations on different axes if indicated
+        new_data = self._mgr
+        for axis in sorted(reindexers.keys()):
+            index, indexer = reindexers[axis]
+            baxis = self._get_block_manager_axis(axis)
+
+            if index is None:
+                continue
+
+            index = ensure_index(index)
+            if indexer is not None:
+                indexer = ensure_platform_int(indexer)
+
+            # TODO: speed up on homogeneous DataFrame objects (see _reindex_multi)
+            new_data = new_data.reindex_indexer(
+                index,
+                indexer,
+                axis=baxis,
+                fill_value=fill_value,
+                allow_dups=allow_dups,
+            )
+
+        if new_data is self._mgr:
+            new_data = new_data.copy(deep=False)
+
+        return self._constructor_from_mgr(new_data, axes=new_data.axes).__finalize__(
+            self
+        )
+
+    def filter(
+        self,
+        items=None,
+        like: str | None = None,
+        regex: str | None = None,
+        axis: Axis | None = None,
+    ) -> Self:
+        """
+        Subset the DataFrame or Series according to the specified index labels.
+
+        For DataFrame, filter rows or columns depending on ``axis`` argument.
+        Note that this routine does not filter based on content.
+        The filter is applied to the labels of the index.
+
+        Parameters
+        ----------
+        items : list-like
+            Keep labels from axis which are in items.
+        like : str
+            Keep labels from axis for which "like in label == True".
+        regex : str (regular expression)
+            Keep labels from axis for which re.search(regex, label) == True.
+        axis : {0 or 'index', 1 or 'columns', None}, default None
+            The axis to filter on, expressed either as an index (int)
+            or axis name (str). By default this is the info axis, 'columns' for
+            ``DataFrame``. For ``Series`` this parameter is unused and defaults to
+            ``None``.
+
+        Returns
+        -------
+        Same type as caller
+            The filtered subset of the DataFrame or Series.
+
+        See Also
+        --------
+        DataFrame.loc : Access a group of rows and columns
+            by label(s) or a boolean array.
+
+        Notes
+        -----
+        The ``items``, ``like``, and ``regex`` parameters are
+        enforced to be mutually exclusive.
+
+        ``axis`` defaults to the info axis that is used when indexing
+        with ``[]``.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     np.array(([1, 2, 3], [4, 5, 6])),
+        ...     index=["mouse", "rabbit"],
+        ...     columns=["one", "two", "three"],
+        ... )
+        >>> df
+                one  two  three
+        mouse     1    2      3
+        rabbit    4    5      6
+
+        >>> # select columns by name
+        >>> df.filter(items=["one", "three"])
+                 one  three
+        mouse     1      3
+        rabbit    4      6
+
+        >>> # select columns by regular expression
+        >>> df.filter(regex="e$", axis=1)
+                 one  three
+        mouse     1      3
+        rabbit    4      6
+
+        >>> # select rows containing 'bbi'
+        >>> df.filter(like="bbi", axis=0)
+                 one  two  three
+        rabbit    4    5      6
+        """
+        nkw = common.count_not_none(items, like, regex)
+        if nkw > 1:
+            raise TypeError(
+                "Keyword arguments `items`, `like`, or `regex` are mutually exclusive"
+            )
+
+        if axis is None:
+            axis = self._info_axis_name
+        labels = self._get_axis(axis)
+
+        if items is not None:
+            name = self._get_axis_name(axis)
+            items = Index(items).intersection(labels)
+            if len(items) == 0:
+                # Keep the dtype of labels when we are empty
+                items = items.astype(labels.dtype)
+            # error: Keywords must be strings
+            return self.reindex(**{name: items})  # type: ignore[misc]
+        elif like:
+
+            def f(x) -> bool:
+                assert like is not None  # needed for mypy
+                return like in ensure_str(x)
+
+            values = labels.map(f)
+            return self.loc(axis=axis)[values]
+        elif regex:
+
+            def f(x) -> bool:
+                return matcher.search(ensure_str(x)) is not None
+
+            matcher = re.compile(regex)
+            values = labels.map(f)
+            return self.loc(axis=axis)[values]
+        else:
+            raise TypeError("Must pass either `items`, `like`, or `regex`")
+
+    @final
+    def head(self, n: int = 5) -> Self:
+        """
+        Return the first `n` rows.
+
+        This function exhibits the same behavior as ``df[:n]``, returning the
+        first ``n`` rows based on position. It is useful for quickly checking
+        if your object has the right type of data in it.
+
+        When ``n`` is positive, it returns the first ``n`` rows. For ``n`` equal to 0,
+        it returns an empty object. When ``n`` is negative, it returns
+        all rows except the last ``|n|`` rows, mirroring the behavior of ``df[:n]``.
+
+        If ``n`` is larger than the number of rows, this function returns all rows.
+
+        Parameters
+        ----------
+        n : int, default 5
+            Number of rows to select.
+
+        Returns
+        -------
+        same type as caller
+            The first `n` rows of the caller object.
+
+        See Also
+        --------
+        DataFrame.tail: Returns the last `n` rows.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "animal": [
+        ...             "alligator",
+        ...             "bee",
+        ...             "falcon",
+        ...             "lion",
+        ...             "monkey",
+        ...             "parrot",
+        ...             "shark",
+        ...             "whale",
+        ...             "zebra",
+        ...         ]
+        ...     }
+        ... )
+        >>> df
+              animal
+        0  alligator
+        1        bee
+        2     falcon
+        3       lion
+        4     monkey
+        5     parrot
+        6      shark
+        7      whale
+        8      zebra
+
+        Viewing the first 5 lines
+
+        >>> df.head()
+              animal
+        0  alligator
+        1        bee
+        2     falcon
+        3       lion
+        4     monkey
+
+        Viewing the first `n` lines (three in this case)
+
+        >>> df.head(3)
+              animal
+        0  alligator
+        1        bee
+        2     falcon
+
+        For negative values of `n`
+
+        >>> df.head(-3)
+              animal
+        0  alligator
+        1        bee
+        2     falcon
+        3       lion
+        4     monkey
+        5     parrot
+        """
+        return self.iloc[:n].copy()
+
+    @final
+    def tail(self, n: int = 5) -> Self:
+        """
+        Return the last `n` rows.
+
+        This function returns last `n` rows from the object based on
+        position. It is useful for quickly verifying data, for example,
+        after sorting or appending rows.
+
+        For negative values of `n`, this function returns all rows except
+        the first `|n|` rows, equivalent to ``df[|n|:]``.
+
+        If ``n`` is larger than the number of rows, this function returns all rows.
+
+        Parameters
+        ----------
+        n : int, default 5
+            Number of rows to select.
+
+        Returns
+        -------
+        type of caller
+            The last `n` rows of the caller object.
+
+        See Also
+        --------
+        DataFrame.head : The first `n` rows of the caller object.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "animal": [
+        ...             "alligator",
+        ...             "bee",
+        ...             "falcon",
+        ...             "lion",
+        ...             "monkey",
+        ...             "parrot",
+        ...             "shark",
+        ...             "whale",
+        ...             "zebra",
+        ...         ]
+        ...     }
+        ... )
+        >>> df
+              animal
+        0  alligator
+        1        bee
+        2     falcon
+        3       lion
+        4     monkey
+        5     parrot
+        6      shark
+        7      whale
+        8      zebra
+
+        Viewing the last 5 lines
+
+        >>> df.tail()
+           animal
+        4  monkey
+        5  parrot
+        6   shark
+        7   whale
+        8   zebra
+
+        Viewing the last `n` lines (three in this case)
+
+        >>> df.tail(3)
+          animal
+        6  shark
+        7  whale
+        8  zebra
+
+        For negative values of `n`
+
+        >>> df.tail(-3)
+           animal
+        3    lion
+        4  monkey
+        5  parrot
+        6   shark
+        7   whale
+        8   zebra
+        """
+        if n == 0:
+            return self.iloc[0:0].copy()
+        return self.iloc[-n:].copy()
+
+    @final
+    def sample(
+        self,
+        n: int | None = None,
+        frac: float | None = None,
+        replace: bool = False,
+        weights=None,
+        random_state: RandomState | None = None,
+        axis: Axis | None = None,
+        ignore_index: bool = False,
+    ) -> Self:
+        """
+        Return a random sample of items from an axis of object.
+
+        You can use `random_state` for reproducibility.
+
+        Parameters
+        ----------
+        n : int, optional
+            Number of items from axis to return. Cannot be used with `frac`.
+            Default = 1 if `frac` = None.
+        frac : float, optional
+            Fraction of axis items to return. Cannot be used with `n`.
+        replace : bool, default False
+            Allow or disallow sampling of the same row more than once.
+        weights : str or ndarray-like, optional
+            Default ``None`` results in equal probability weighting.
+            If passed a Series, will align with target object on index. Index
+            values in weights not found in sampled object will be ignored and
+            index values in sampled object not in weights will be assigned
+            weights of zero.
+            If called on a DataFrame, will accept the name of a column
+            when axis = 0.
+            Unless weights are a Series, weights must be same length as axis
+            being sampled.
+            If weights do not sum to 1, they will be normalized to sum to 1.
+            Missing values in the weights column will be treated as zero.
+            Infinite values not allowed.
+        random_state : int, array-like, BitGenerator, np.random.RandomState, np.random.Generator, optional
+            If int, array-like, or BitGenerator, seed for random number generator.
+            If np.random.RandomState or np.random.Generator, use as given.
+            Default ``None`` results in sampling with the current state of np.random.
+
+            .. versionchanged:: 1.4.0
+
+                np.random.Generator objects now accepted
+
+        axis : {0 or 'index', 1 or 'columns', None}, default None
+            Axis to sample. Accepts axis number or name. Default is stat axis
+            for given data type. For `Series` this parameter is unused and defaults to `None`.
+        ignore_index : bool, default False
+            If True, the resulting index will be labeled 0, 1, …, n - 1.
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Series or DataFrame
+            A new object of same type as caller containing `n` items randomly
+            sampled from the caller object.
+
+        See Also
+        --------
+        DataFrameGroupBy.sample: Generates random samples from each group of a
+            DataFrame object.
+        SeriesGroupBy.sample: Generates random samples from each group of a
+            Series object.
+        numpy.random.choice: Generates a random sample from a given 1-D numpy
+            array.
+
+        Notes
+        -----
+        If `frac` > 1, `replacement` should be set to `True`.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "num_legs": [2, 4, 8, 0],
+        ...         "num_wings": [2, 0, 0, 0],
+        ...         "num_specimen_seen": [10, 2, 1, 8],
+        ...     },
+        ...     index=["falcon", "dog", "spider", "fish"],
+        ... )
+        >>> df
+                num_legs  num_wings  num_specimen_seen
+        falcon         2          2                 10
+        dog            4          0                  2
+        spider         8          0                  1
+        fish           0          0                  8
+
+        Extract 3 random elements from the ``Series`` ``df['num_legs']``:
+        Note that we use `random_state` to ensure the reproducibility of
+        the examples.
+
+        >>> df["num_legs"].sample(n=3, random_state=1)
+        fish      0
+        spider    8
+        falcon    2
+        Name: num_legs, dtype: int64
+
+        A random 50% sample of the ``DataFrame`` with replacement:
+
+        >>> df.sample(frac=0.5, replace=True, random_state=1)
+              num_legs  num_wings  num_specimen_seen
+        dog          4          0                  2
+        fish         0          0                  8
+
+        An upsample sample of the ``DataFrame`` with replacement:
+        Note that `replace` parameter has to be `True` for `frac` parameter > 1.
+
+        >>> df.sample(frac=2, replace=True, random_state=1)
+                num_legs  num_wings  num_specimen_seen
+        dog            4          0                  2
+        fish           0          0                  8
+        falcon         2          2                 10
+        falcon         2          2                 10
+        fish           0          0                  8
+        dog            4          0                  2
+        fish           0          0                  8
+        dog            4          0                  2
+
+        Using a DataFrame column as weights. Rows with larger value in the
+        `num_specimen_seen` column are more likely to be sampled.
+
+        >>> df.sample(n=2, weights="num_specimen_seen", random_state=1)
+                num_legs  num_wings  num_specimen_seen
+        falcon         2          2                 10
+        fish           0          0                  8
+        """  # noqa: E501
+        if axis is None:
+            axis = 0
+
+        axis = self._get_axis_number(axis)
+        obj_len = self.shape[axis]
+
+        # Process random_state argument
+        rs = common.random_state(random_state)
+
+        size = sample.process_sampling_size(n, frac, replace)
+        if size is None:
+            assert frac is not None
+            size = round(frac * obj_len)
+
+        if weights is not None:
+            weights = sample.preprocess_weights(self, weights, axis)
+
+        sampled_indices = sample.sample(obj_len, size, replace, weights, rs)
+        result = self.take(sampled_indices, axis=axis)
+
+        if ignore_index:
+            result.index = default_index(len(result))
+
+        return result
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        r"""
+        Apply chainable functions that expect Series or DataFrames.
+
+        Parameters
+        ----------
+        func : function
+            Function to apply to the {klass}.
+            ``args``, and ``kwargs`` are passed into ``func``.
+            Alternatively a ``(callable, data_keyword)`` tuple where
+            ``data_keyword`` is a string indicating the keyword of
+            ``callable`` that expects the {klass}.
+        *args : iterable, optional
+            Positional arguments passed into ``func``.
+        **kwargs : mapping, optional
+            A dictionary of keyword arguments passed into ``func``.
+
+        Returns
+        -------
+        The return type of ``func``.
+            The result of applying ``func`` to the Series or DataFrame.
+
+        See Also
+        --------
+        DataFrame.apply : Apply a function along input axis of DataFrame.
+        DataFrame.map : Apply a function elementwise on a whole DataFrame.
+        Series.map : Apply a mapping correspondence on a
+            :class:`~pandas.Series`.
+
+        Notes
+        -----
+        Use ``.pipe`` when chaining together functions that expect
+        Series, DataFrames or GroupBy objects.
+
+        Examples
+        --------
+        Constructing an income DataFrame from a dictionary.
+
+        >>> data = [[8000, 1000], [9500, np.nan], [5000, 2000]]
+        >>> df = pd.DataFrame(data, columns=["Salary", "Others"])
+        >>> df
+           Salary  Others
+        0    8000  1000.0
+        1    9500     NaN
+        2    5000  2000.0
+
+        Functions that perform tax reductions on an income DataFrame.
+
+        >>> def subtract_federal_tax(df):
+        ...     return df * 0.9
+        >>> def subtract_state_tax(df, rate):
+        ...     return df * (1 - rate)
+        >>> def subtract_national_insurance(df, rate, rate_increase):
+        ...     new_rate = rate + rate_increase
+        ...     return df * (1 - new_rate)
+
+        Instead of writing
+
+        >>> subtract_national_insurance(
+        ...     subtract_state_tax(subtract_federal_tax(df), rate=0.12),
+        ...     rate=0.05,
+        ...     rate_increase=0.02,
+        ... )  # doctest: +SKIP
+
+        You can write
+
+        >>> (
+        ...     df.pipe(subtract_federal_tax)
+        ...     .pipe(subtract_state_tax, rate=0.12)
+        ...     .pipe(subtract_national_insurance, rate=0.05, rate_increase=0.02)
+        ... )
+            Salary   Others
+        0  5892.48   736.56
+        1  6997.32      NaN
+        2  3682.80  1473.12
+
+        If you have a function that takes the data as (say) the second
+        argument, pass a tuple indicating which keyword expects the
+        data. For example, suppose ``national_insurance`` takes its data as ``df``
+        in the second argument:
+
+        >>> def subtract_national_insurance(rate, df, rate_increase):
+        ...     new_rate = rate + rate_increase
+        ...     return df * (1 - new_rate)
+        >>> (
+        ...     df.pipe(subtract_federal_tax)
+        ...     .pipe(subtract_state_tax, rate=0.12)
+        ...     .pipe(
+        ...         (subtract_national_insurance, "df"), rate=0.05, rate_increase=0.02
+        ...     )
+        ... )
+            Salary   Others
+        0  5892.48   736.56
+        1  6997.32      NaN
+        2  3682.80  1473.12
+        """
+        return common.pipe(self.copy(deep=False), func, *args, **kwargs)
+
+    # ----------------------------------------------------------------------
+    # Attribute access
+
+    @final
+    def __finalize__(self, other, method: str | None = None, **kwargs) -> Self:
+        """
+        Propagate metadata from other to self.
+
+        Parameters
+        ----------
+        other : the object from which to get the attributes that we are going
+            to propagate
+        method : str, optional
+            A passed method name providing context on where ``__finalize__``
+            was called.
+
+            .. warning::
+
+               The value passed as `method` are not currently considered
+               stable across pandas releases.
+        """
+        if isinstance(other, NDFrame):
+            if other.attrs:
+                # We want attrs propagation to have minimal performance
+                # impact if attrs are not used; i.e. attrs is an empty dict.
+                # One could make the deepcopy unconditionally, but a deepcopy
+                # of an empty dict is 50x more expensive than the empty check.
+                self.attrs = deepcopy(other.attrs)
+
+            self.flags.allows_duplicate_labels = other.flags.allows_duplicate_labels
+            # For subclasses using _metadata.
+            for name in set(self._metadata) & set(other._metadata):
+                assert isinstance(name, str)
+                object.__setattr__(self, name, getattr(other, name, None))
+
+        if method == "concat":
+            objs = other.objs
+            # propagate attrs only if all concat arguments have the same attrs
+            if all(bool(obj.attrs) for obj in objs):
+                # all concatenate arguments have non-empty attrs
+                attrs = objs[0].attrs
+                have_same_attrs = all(obj.attrs == attrs for obj in objs[1:])
+                if have_same_attrs:
+                    self.attrs = deepcopy(attrs)
+
+            allows_duplicate_labels = all(x.flags.allows_duplicate_labels for x in objs)
+            self.flags.allows_duplicate_labels = allows_duplicate_labels
+
+        return self
+
+    @final
+    def __getattr__(self, name: str):
+        """
+        After regular attribute access, try looking up the name
+        This allows simpler access to columns for interactive use.
+        """
+        # Note: obj.x will always call obj.__getattribute__('x') prior to
+        # calling obj.__getattr__('x').
+        if (
+            name not in self._internal_names_set
+            and name not in self._metadata
+            and name not in self._accessors
+            and self._info_axis._can_hold_identifiers_and_holds_name(name)
+        ):
+            return self[name]
+        return object.__getattribute__(self, name)
+
+    @final
+    def __setattr__(self, name: str, value) -> None:
+        """
+        After regular attribute access, try setting the name
+        This allows simpler access to columns for interactive use.
+        """
+        # first try regular attribute access via __getattribute__, so that
+        # e.g. ``obj.x`` and ``obj.x = 4`` will always reference/modify
+        # the same attribute.
+
+        try:
+            object.__getattribute__(self, name)
+            return object.__setattr__(self, name, value)
+        except AttributeError:
+            pass
+
+        # if this fails, go on to more involved attribute setting
+        # (note that this matches __getattr__, above).
+        if name in self._internal_names_set:
+            object.__setattr__(self, name, value)
+        elif name in self._metadata:
+            object.__setattr__(self, name, value)
+        else:
+            try:
+                existing = getattr(self, name)
+                if isinstance(existing, Index):
+                    object.__setattr__(self, name, value)
+                elif name in self._info_axis:
+                    self[name] = value
+                else:
+                    object.__setattr__(self, name, value)
+            except (AttributeError, TypeError):
+                if isinstance(self, ABCDataFrame) and (is_list_like(value)):
+                    warnings.warn(
+                        "Pandas doesn't allow columns to be "
+                        "created via a new attribute name - see "
+                        "https://pandas.pydata.org/pandas-docs/"
+                        "stable/indexing.html#attribute-access",
+                        stacklevel=find_stack_level(),
+                    )
+                object.__setattr__(self, name, value)
+
+    @final
+    def _dir_additions(self) -> set[str]:
+        """
+        add the string-like attributes from the info_axis.
+        If info_axis is a MultiIndex, its first level values are used.
+        """
+        additions = super()._dir_additions()
+        if self._info_axis._can_hold_strings:
+            additions.update(self._info_axis._dir_additions_for_owner)
+        return additions
+
+    # ----------------------------------------------------------------------
+    # Consolidation of internals
+
+    @final
+    def _consolidate_inplace(self) -> None:
+        """Consolidate data in place and return None"""
+
+        self._mgr = self._mgr.consolidate()
+
+    @final
+    def _consolidate(self):
+        """
+        Compute NDFrame with "consolidated" internals (data of each dtype
+        grouped together in a single ndarray).
+
+        Returns
+        -------
+        consolidated : same type as caller
+        """
+        cons_data = self._mgr.consolidate()
+        return self._constructor_from_mgr(cons_data, axes=cons_data.axes).__finalize__(
+            self
+        )
+
+    @final
+    @property
+    def _is_mixed_type(self) -> bool:
+        if self._mgr.is_single_block:
+            # Includes all Series cases
+            return False
+
+        if self._mgr.any_extension_types:
+            # Even if they have the same dtype, we can't consolidate them,
+            #  so we pretend this is "mixed'"
+            return True
+
+        return self.dtypes.nunique() > 1
+
+    @final
+    def _get_numeric_data(self) -> Self:
+        new_mgr = self._mgr.get_numeric_data()
+        return self._constructor_from_mgr(new_mgr, axes=new_mgr.axes).__finalize__(self)
+
+    @final
+    def _get_bool_data(self):
+        new_mgr = self._mgr.get_bool_data()
+        return self._constructor_from_mgr(new_mgr, axes=new_mgr.axes).__finalize__(self)
+
+    # ----------------------------------------------------------------------
+    # Internal Interface Methods
+
+    @property
+    def values(self):
+        raise AbstractMethodError(self)
+
+    @property
+    def _values(self) -> ArrayLike:
+        """internal implementation"""
+        raise AbstractMethodError(self)
+
+    @property
+    def dtypes(self):
+        """
+        Return the dtypes in the DataFrame.
+
+        This returns a Series with the data type of each column.
+        The result's index is the original DataFrame's columns. Columns
+        with mixed types are stored with the ``object`` dtype. See
+        :ref:`the User Guide <basics.dtypes>` for more.
+
+        Returns
+        -------
+        pandas.Series
+            The data type of each column.
+
+        See Also
+        --------
+        Series.dtypes : Return the dtype object of the underlying data.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "float": [1.0],
+        ...         "int": [1],
+        ...         "datetime": [pd.Timestamp("20180310")],
+        ...         "string": ["foo"],
+        ...     }
+        ... )
+        >>> df.dtypes
+        float              float64
+        int                  int64
+        datetime    datetime64[s]
+        string              object
+        dtype: object
+        """
+        data = self._mgr.get_dtypes()
+        return self._constructor_sliced(data, index=self._info_axis, dtype=np.object_)
+
+    @final
+    def astype(
+        self,
+        dtype,
+        copy: bool | lib.NoDefault = lib.no_default,
+        errors: IgnoreRaise = "raise",
+    ) -> Self:
+        """
+        Cast a pandas object to a specified dtype ``dtype``.
+
+        This method allows the conversion of the data types of pandas objects,
+        including DataFrames and Series, to the specified dtype. It supports casting
+        entire objects to a single data type or applying different data types to
+        individual columns using a mapping.
+
+        Parameters
+        ----------
+        dtype : str, data type, Series or Mapping of column name -> data type
+            Use a str, numpy.dtype, pandas.ExtensionDtype or Python type to
+            cast entire pandas object to the same type. Alternatively, use a
+            mapping, e.g. {col: dtype, ...}, where col is a column label and dtype is
+            a numpy.dtype or Python type to cast one or more of the DataFrame's
+            columns to column-specific types.
+        copy : bool, default False
+            Return a copy when ``copy=True`` (be very careful setting
+            ``copy=False`` as changes to values then may propagate to other
+            pandas objects).
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        errors : {'raise', 'ignore'}, default 'raise'
+            Control raising of exceptions on invalid data for provided dtype.
+
+            - ``raise`` : allow exceptions to be raised
+            - ``ignore`` : suppress exceptions. On error return original object.
+
+        Returns
+        -------
+        same type as caller
+            The pandas object casted to the specified ``dtype``.
+
+        See Also
+        --------
+        to_datetime : Convert argument to datetime.
+        to_timedelta : Convert argument to timedelta.
+        to_numeric : Convert argument to a numeric type.
+        numpy.ndarray.astype : Cast a numpy array to a specified type.
+
+        Notes
+        -----
+        .. versionchanged:: 2.0.0
+
+            Using ``astype`` to convert from timezone-naive dtype to
+            timezone-aware dtype will raise an exception.
+            Use :meth:`Series.dt.tz_localize` instead.
+
+        Examples
+        --------
+        Create a DataFrame:
+
+        >>> d = {"col1": [1, 2], "col2": [3, 4]}
+        >>> df = pd.DataFrame(data=d)
+        >>> df.dtypes
+        col1    int64
+        col2    int64
+        dtype: object
+
+        Cast all columns to int32:
+
+        >>> df.astype("int32").dtypes
+        col1    int32
+        col2    int32
+        dtype: object
+
+        Cast col1 to int32 using a dictionary:
+
+        >>> df.astype({"col1": "int32"}).dtypes
+        col1    int32
+        col2    int64
+        dtype: object
+
+        Create a series:
+
+        >>> ser = pd.Series([1, 2], dtype="int32")
+        >>> ser
+        0    1
+        1    2
+        dtype: int32
+        >>> ser.astype("int64")
+        0    1
+        1    2
+        dtype: int64
+
+        Convert to categorical type:
+
+        >>> ser.astype("category")
+        0    1
+        1    2
+        dtype: category
+        Categories (2, int32): [1, 2]
+
+        Convert to ordered categorical type with custom ordering:
+
+        >>> from pandas.api.types import CategoricalDtype
+        >>> cat_dtype = CategoricalDtype(categories=[2, 1], ordered=True)
+        >>> ser.astype(cat_dtype)
+        0    1
+        1    2
+        dtype: category
+        Categories (2, int64): [2 < 1]
+
+        Create a series of dates:
+
+        >>> ser_date = pd.Series(pd.date_range("20200101", periods=3))
+        >>> ser_date
+        0   2020-01-01
+        1   2020-01-02
+        2   2020-01-03
+        dtype: datetime64[ns]
+        """
+        self._check_copy_deprecation(copy)
+        if is_dict_like(dtype):
+            if self.ndim == 1:  # i.e. Series
+                if len(dtype) > 1 or self.name not in dtype:
+                    raise KeyError(
+                        "Only the Series name can be used for "
+                        "the key in Series dtype mappings."
+                    )
+                new_type = dtype[self.name]
+                return self.astype(new_type, errors=errors)
+
+            # GH#44417 cast to Series so we can use .iat below, which will be
+            #  robust in case we
+            from pandas import Series
+
+            dtype_ser = Series(dtype, dtype=object)
+
+            for col_name in dtype_ser.index:
+                if col_name not in self:
+                    raise KeyError(
+                        "Only a column name can be used for the "
+                        "key in a dtype mappings argument. "
+                        f"'{col_name}' not found in columns."
+                    )
+
+            dtype_ser = dtype_ser.reindex(self.columns, fill_value=None)
+
+            results = []
+            for i, (col_name, col) in enumerate(self.items()):
+                cdt = dtype_ser.iat[i]
+                if isna(cdt):
+                    res_col = col.copy(deep=False)
+                else:
+                    try:
+                        res_col = col.astype(dtype=cdt, errors=errors)
+                    except ValueError as ex:
+                        ex.args = (
+                            f"{ex}: Error while type casting for column '{col_name}'",
+                        )
+                        raise
+                results.append(res_col)
+
+        elif is_extension_array_dtype(dtype) and self.ndim > 1:
+            # TODO(EA2D): special case not needed with 2D EAs
+            dtype = pandas_dtype(dtype)
+            if isinstance(dtype, ExtensionDtype) and all(
+                block.values.dtype == dtype for block in self._mgr.blocks
+            ):
+                return self.copy(deep=False)
+            # GH 18099/22869: columnwise conversion to extension dtype
+            # GH 24704: self.items handles duplicate column names
+            results = [ser.astype(dtype, errors=errors) for _, ser in self.items()]
+
+        else:
+            # else, only a single dtype is given
+            new_data = self._mgr.astype(dtype=dtype, errors=errors)
+            res = self._constructor_from_mgr(new_data, axes=new_data.axes)
+            return res.__finalize__(self, method="astype")
+
+        # GH 33113: handle empty frame or series
+        if not results:
+            return self.copy(deep=False)
+
+        # GH 19920: retain column metadata after concat
+        result = concat(results, axis=1)
+        # GH#40810 retain subclass
+        # error: Incompatible types in assignment
+        # (expression has type "Self", variable has type "DataFrame")
+        result = self._constructor(result)  # type: ignore[assignment]
+        result.columns = self.columns
+        result = result.__finalize__(self, method="astype")
+        # https://github.com/python/mypy/issues/8354
+        return cast(Self, result)
+
+    @final
+    def copy(self, deep: bool = True) -> Self:
+        """
+        Make a copy of this object's indices and data.
+
+        When ``deep=True`` (default), a new object will be created with a
+        copy of the calling object's data and indices. Modifications to
+        the data or indices of the copy will not be reflected in the
+        original object (see notes below).
+
+        When ``deep=False``, a new object will be created without copying
+        the calling object's data or index (only references to the data
+        and index are copied). Any changes to the data of the original
+        will be reflected in the shallow copy (and vice versa).
+
+        .. note::
+            The ``deep=False`` behaviour as described above will change
+            in pandas 3.0. `Copy-on-Write
+            <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+            will be enabled by default, which means that the "shallow" copy
+            is that is returned with ``deep=False`` will still avoid making
+            an eager copy, but changes to the data of the original will *no*
+            longer be reflected in the shallow copy (or vice versa). Instead,
+            it makes use of a lazy (deferred) copy mechanism that will copy
+            the data only when any changes to the original or shallow copy is
+            made.
+
+            You can already get the future behavior and improvements through
+            enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+        Parameters
+        ----------
+        deep : bool, default True
+            Make a deep copy, including a copy of the data and the indices.
+            With ``deep=False`` neither the indices nor the data are copied.
+
+        Returns
+        -------
+        Series or DataFrame
+            Object type matches caller.
+
+        See Also
+        --------
+        copy.copy : Return a shallow copy of an object.
+        copy.deepcopy : Return a deep copy of an object.
+
+        Notes
+        -----
+        When ``deep=True``, data is copied but actual Python objects
+        will not be copied recursively, only the reference to the object.
+        This is in contrast to `copy.deepcopy` in the Standard Library,
+        which recursively copies object data (see examples below).
+
+        While ``Index`` objects are copied when ``deep=True``, the underlying
+        numpy array is not copied for performance reasons. Since ``Index`` is
+        immutable, the underlying data can be safely shared and a copy
+        is not needed.
+
+        Since pandas is not thread safe, see the
+        :ref:`gotchas <gotchas.thread-safety>` when copying in a threading
+        environment.
+
+        Copy-on-Write protects shallow copies against accidental modifications.
+        This means that any changes to the copied data would make a new copy
+        of the data upon write (and vice versa). Changes made to either the
+        original or copied variable would not be reflected in the counterpart.
+        See :ref:`Copy_on_Write <copy_on_write>` for more information.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2], index=["a", "b"])
+        >>> s
+        a    1
+        b    2
+        dtype: int64
+
+        >>> s_copy = s.copy()
+        >>> s_copy
+        a    1
+        b    2
+        dtype: int64
+
+        **Shallow copy versus default (deep) copy:**
+
+        >>> s = pd.Series([1, 2], index=["a", "b"])
+        >>> deep = s.copy()
+        >>> shallow = s.copy(deep=False)
+
+        Shallow copy shares index with original, the data is a
+        view of the original.
+
+        >>> s is shallow
+        False
+        >>> s.values is shallow.values
+        False
+        >>> s.index is shallow.index
+        False
+
+        Deep copy has own copy of data and index.
+
+        >>> s is deep
+        False
+        >>> s.values is deep.values or s.index is deep.index
+        False
+
+        The shallow copy is protected against updating the original object
+        as well. Thus, updates will only reflect in one of both objects.
+
+        >>> s.iloc[0] = 3
+        >>> shallow.iloc[1] = 4
+        >>> s
+        a    3
+        b    2
+        dtype: int64
+        >>> shallow
+        a    1
+        b    4
+        dtype: int64
+        >>> deep
+        a    1
+        b    2
+        dtype: int64
+
+        Note that when copying an object containing Python objects, a deep copy
+        will copy the data, but will not do so recursively. Updating a nested
+        data object will be reflected in the deep copy.
+
+        >>> s = pd.Series([[1, 2], [3, 4]])
+        >>> deep = s.copy()
+        >>> s[0][0] = 10
+        >>> s
+        0    [10, 2]
+        1     [3, 4]
+        dtype: object
+        >>> deep
+        0    [10, 2]
+        1     [3, 4]
+        dtype: object
+        """
+        data = self._mgr.copy(deep=deep)
+        return self._constructor_from_mgr(data, axes=data.axes).__finalize__(
+            self, method="copy"
+        )
+
+    @final
+    def __copy__(self, deep: bool = True) -> Self:
+        return self.copy(deep=deep)
+
+    @final
+    def __deepcopy__(self, memo=None) -> Self:
+        """
+        Parameters
+        ----------
+        memo, default None
+            Standard signature. Unused
+        """
+        return self.copy(deep=True)
+
+    @final
+    def infer_objects(self, copy: bool | lib.NoDefault = lib.no_default) -> Self:
+        """
+        Attempt to infer better dtypes for object columns.
+
+        Attempts soft conversion of object-dtyped
+        columns, leaving non-object and unconvertible
+        columns unchanged. The inference rules are the
+        same as during normal Series/DataFrame construction.
+
+        Parameters
+        ----------
+        copy : bool, default False
+            Whether to make a copy for non-object or non-inferable columns
+            or Series.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        same type as input object
+            Returns an object of the same type as the input object.
+
+        See Also
+        --------
+        to_datetime : Convert argument to datetime.
+        to_timedelta : Convert argument to timedelta.
+        to_numeric : Convert argument to numeric type.
+        convert_dtypes : Convert argument to best possible dtype.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": ["a", 1, 2, 3]})
+        >>> df = df.iloc[1:]
+        >>> df
+           A
+        1  1
+        2  2
+        3  3
+
+        >>> df.dtypes
+        A    object
+        dtype: object
+
+        >>> df.infer_objects().dtypes
+        A    int64
+        dtype: object
+        """
+        self._check_copy_deprecation(copy)
+        new_mgr = self._mgr.convert()
+        res = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+        return res.__finalize__(self, method="infer_objects")
+
+    @final
+    def convert_dtypes(
+        self,
+        infer_objects: bool = True,
+        convert_string: bool = True,
+        convert_integer: bool = True,
+        convert_boolean: bool = True,
+        convert_floating: bool = True,
+        dtype_backend: DtypeBackend = "numpy_nullable",
+    ) -> Self:
+        """
+        Convert columns from numpy dtypes to the best dtypes that support ``pd.NA``.
+
+        Parameters
+        ----------
+        infer_objects : bool, default True
+            Whether object dtypes should be converted to the best possible types.
+        convert_string : bool, default True
+            Whether object dtypes should be converted to ``StringDtype()``.
+        convert_integer : bool, default True
+            Whether, if possible, conversion can be done to integer extension types.
+        convert_boolean : bool, defaults True
+            Whether object dtypes should be converted to ``BooleanDtypes()``.
+        convert_floating : bool, defaults True
+            Whether, if possible, conversion can be done to floating extension types.
+            If `convert_integer` is also True, preference will be give to integer
+            dtypes if the floats can be faithfully casted to integers.
+        dtype_backend : {'numpy_nullable', 'pyarrow'}, default 'numpy_nullable'
+            Back-end data type applied to the resultant :class:`DataFrame` or
+            :class:`Series` (still experimental). Behaviour is as follows:
+
+            * ``"numpy_nullable"``: returns nullable-dtype-backed
+              :class:`DataFrame` or :class:`Serires`.
+            * ``"pyarrow"``: returns pyarrow-backed nullable :class:`ArrowDtype`
+              :class:`DataFrame` or :class:`Series`.
+
+            .. versionadded:: 2.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Copy of input object with new dtype.
+
+        See Also
+        --------
+        infer_objects : Infer dtypes of objects.
+        to_datetime : Convert argument to datetime.
+        to_timedelta : Convert argument to timedelta.
+        to_numeric : Convert argument to a numeric type.
+
+        Notes
+        -----
+        By default, ``convert_dtypes`` will attempt to convert a Series (or each
+        Series in a DataFrame) to dtypes that support ``pd.NA``. By using the options
+        ``convert_string``, ``convert_integer``, ``convert_boolean`` and
+        ``convert_floating``, it is possible to turn off individual conversions
+        to ``StringDtype``, the integer extension types, ``BooleanDtype``
+        or floating extension types, respectively.
+
+        For object-dtyped columns, if ``infer_objects`` is ``True``, use the inference
+        rules as during normal Series/DataFrame construction.  Then, if possible,
+        convert to ``StringDtype``, ``BooleanDtype`` or an appropriate integer
+        or floating extension type, otherwise leave as ``object``.
+
+        If the dtype is integer, convert to an appropriate integer extension type.
+
+        If the dtype is numeric, and consists of all integers, convert to an
+        appropriate integer extension type. Otherwise, convert to an
+        appropriate floating extension type.
+
+        In the future, as new dtypes are added that support ``pd.NA``, the results
+        of this method will change to support those new dtypes.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "a": pd.Series([1, 2, 3], dtype=np.dtype("int32")),
+        ...         "b": pd.Series(["x", "y", "z"], dtype=np.dtype("O")),
+        ...         "c": pd.Series([True, False, np.nan], dtype=np.dtype("O")),
+        ...         "d": pd.Series(["h", "i", np.nan], dtype=np.dtype("O")),
+        ...         "e": pd.Series([10, np.nan, 20], dtype=np.dtype("float")),
+        ...         "f": pd.Series([np.nan, 100.5, 200], dtype=np.dtype("float")),
+        ...     }
+        ... )
+
+        Start with a DataFrame with default dtypes.
+
+        >>> df
+           a  b      c    d     e      f
+        0  1  x   True    h  10.0    NaN
+        1  2  y  False    i   NaN  100.5
+        2  3  z    NaN  NaN  20.0  200.0
+
+        >>> df.dtypes
+        a      int32
+        b     object
+        c     object
+        d     object
+        e    float64
+        f    float64
+        dtype: object
+
+        Convert the DataFrame to use best possible dtypes.
+
+        >>> dfn = df.convert_dtypes()
+        >>> dfn
+           a  b      c     d     e      f
+        0  1  x   True     h    10   <NA>
+        1  2  y  False     i  <NA>  100.5
+        2  3  z   <NA>  <NA>    20  200.0
+
+        >>> dfn.dtypes
+        a             Int32
+        b    string[python]
+        c           boolean
+        d    string[python]
+        e             Int64
+        f           Float64
+        dtype: object
+
+        Start with a Series of strings and missing data represented by ``np.nan``.
+
+        >>> s = pd.Series(["a", "b", np.nan])
+        >>> s
+        0      a
+        1      b
+        2    NaN
+        dtype: object
+
+        Obtain a Series with dtype ``StringDtype``.
+
+        >>> s.convert_dtypes()
+        0       a
+        1       b
+        2    <NA>
+        dtype: string
+        """
+        check_dtype_backend(dtype_backend)
+        new_mgr = self._mgr.convert_dtypes(
+            infer_objects=infer_objects,
+            convert_string=convert_string,
+            convert_integer=convert_integer,
+            convert_boolean=convert_boolean,
+            convert_floating=convert_floating,
+            dtype_backend=dtype_backend,
+        )
+        res = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+        return res.__finalize__(self, method="convert_dtypes")
+
+    # ----------------------------------------------------------------------
+    # Filling NA's
+
+    @final
+    def _pad_or_backfill(
+        self,
+        method: Literal["ffill", "bfill", "pad", "backfill"],
+        *,
+        axis: None | Axis = None,
+        inplace: bool = False,
+        limit: None | int = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+    ):
+        if axis is None:
+            axis = 0
+        axis = self._get_axis_number(axis)
+        method = clean_fill_method(method)
+
+        if axis == 1:
+            if not self._mgr.is_single_block and inplace:
+                raise NotImplementedError
+            # e.g. test_align_fill_method
+            result = self.T._pad_or_backfill(
+                method=method, limit=limit, limit_area=limit_area
+            ).T
+
+            return result
+
+        new_mgr = self._mgr.pad_or_backfill(
+            method=method,
+            limit=limit,
+            limit_area=limit_area,
+            inplace=inplace,
+        )
+        result = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="fillna")
+
+    @overload
+    def fillna(
+        self,
+        value: Hashable | Mapping | Series | DataFrame,
+        *,
+        axis: Axis | None = ...,
+        inplace: Literal[False] = ...,
+        limit: int | None = ...,
+    ) -> Self: ...
+
+    @overload
+    def fillna(
+        self,
+        value: Hashable | Mapping | Series | DataFrame,
+        *,
+        axis: Axis | None = ...,
+        inplace: Literal[True],
+        limit: int | None = ...,
+    ) -> None: ...
+
+    @overload
+    def fillna(
+        self,
+        value: Hashable | Mapping | Series | DataFrame,
+        *,
+        axis: Axis | None = ...,
+        inplace: bool = ...,
+        limit: int | None = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+    )
+    def fillna(
+        self,
+        value: Hashable | Mapping | Series | DataFrame,
+        *,
+        axis: Axis | None = None,
+        inplace: bool = False,
+        limit: int | None = None,
+    ) -> Self | None:
+        """
+        Fill NA/NaN values with `value`.
+
+        Parameters
+        ----------
+        value : scalar, dict, Series, or DataFrame
+            Value to use to fill holes (e.g. 0), alternately a
+            dict/Series/DataFrame of values specifying which value to use for
+            each index (for a Series) or column (for a DataFrame).  Values not
+            in the dict/Series/DataFrame will not be filled. This value cannot
+            be a list.
+        axis : {axes_single_arg}
+            Axis along which to fill missing values. For `Series`
+            this parameter is unused and defaults to 0.
+        inplace : bool, default False
+            If True, fill in-place. Note: this will modify any
+            other views on this object (e.g., a no-copy slice for a column in a
+            DataFrame).
+        limit : int, default None
+            This is the maximum number of entries along the entire axis
+            where NaNs will be filled. Must be greater than 0 if not None.
+
+        Returns
+        -------
+        {klass} or None
+            Object with missing values filled or None if ``inplace=True``.
+
+        See Also
+        --------
+        ffill : Fill values by propagating the last valid observation to next valid.
+        bfill : Fill values by using the next valid observation to fill the gap.
+        interpolate : Fill NaN values using interpolation.
+        reindex : Conform object to new index.
+        asfreq : Convert TimeSeries to specified frequency.
+
+        Notes
+        -----
+        For non-object dtype, ``value=None`` will use the NA value of the dtype.
+        See more details in the :ref:`Filling missing data<missing_data.fillna>`
+        section.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         [np.nan, 2, np.nan, 0],
+        ...         [3, 4, np.nan, 1],
+        ...         [np.nan, np.nan, np.nan, np.nan],
+        ...         [np.nan, 3, np.nan, 4],
+        ...     ],
+        ...     columns=list("ABCD"),
+        ... )
+        >>> df
+             A    B   C    D
+        0  NaN  2.0 NaN  0.0
+        1  3.0  4.0 NaN  1.0
+        2  NaN  NaN NaN  NaN
+        3  NaN  3.0 NaN  4.0
+
+        Replace all NaN elements with 0s.
+
+        >>> df.fillna(0)
+             A    B    C    D
+        0  0.0  2.0  0.0  0.0
+        1  3.0  4.0  0.0  1.0
+        2  0.0  0.0  0.0  0.0
+        3  0.0  3.0  0.0  4.0
+
+        Replace all NaN elements in column 'A', 'B', 'C', and 'D', with 0, 1,
+        2, and 3 respectively.
+
+        >>> values = {{"A": 0, "B": 1, "C": 2, "D": 3}}
+        >>> df.fillna(value=values)
+             A    B    C    D
+        0  0.0  2.0  2.0  0.0
+        1  3.0  4.0  2.0  1.0
+        2  0.0  1.0  2.0  3.0
+        3  0.0  3.0  2.0  4.0
+
+        Only replace the first NaN element.
+
+        >>> df.fillna(value=values, limit=1)
+             A    B    C    D
+        0  0.0  2.0  2.0  0.0
+        1  3.0  4.0  NaN  1.0
+        2  NaN  1.0  NaN  3.0
+        3  NaN  3.0  NaN  4.0
+
+        When filling using a DataFrame, replacement happens along
+        the same column names and same indices
+
+        >>> df2 = pd.DataFrame(np.zeros((4, 4)), columns=list("ABCE"))
+        >>> df.fillna(df2)
+             A    B    C    D
+        0  0.0  2.0  0.0  0.0
+        1  3.0  4.0  0.0  1.0
+        2  0.0  0.0  0.0  NaN
+        3  0.0  3.0  0.0  4.0
+
+        Note that column D is not affected since it is not present in df2.
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        if isinstance(value, (list, tuple)):
+            raise TypeError(
+                '"value" parameter must be a scalar or dict, but '
+                f'you passed a "{type(value).__name__}"'
+            )
+
+        # set the default here, so functions examining the signature
+        # can detect if something was set (e.g. in groupby) (GH9221)
+        if axis is None:
+            axis = 0
+        axis = self._get_axis_number(axis)
+
+        if self.ndim == 1:
+            if isinstance(value, (dict, ABCSeries)):
+                if not len(value):
+                    # test_fillna_nonscalar
+                    if inplace:
+                        return None
+                    return self.copy(deep=False)
+                from pandas import Series
+
+                value = Series(value)
+                value = value.reindex(self.index)
+                value = value._values
+            elif not is_list_like(value):
+                pass
+            else:
+                raise TypeError(
+                    '"value" parameter must be a scalar, dict '
+                    "or Series, but you passed a "
+                    f'"{type(value).__name__}"'
+                )
+
+            new_data = self._mgr.fillna(value=value, limit=limit, inplace=inplace)
+
+        elif isinstance(value, (dict, ABCSeries)):
+            if axis == 1:
+                raise NotImplementedError(
+                    "Currently only can fill with dict/Series column by column"
+                )
+            result = self if inplace else self.copy(deep=False)
+            for k, v in value.items():
+                if k not in result:
+                    continue
+
+                res_k = result[k].fillna(v, limit=limit)
+
+                if not inplace:
+                    result[k] = res_k
+                else:
+                    # We can write into our existing column(s) iff dtype
+                    #  was preserved.
+                    if isinstance(res_k, ABCSeries):
+                        # i.e. 'k' only shows up once in self.columns
+                        if res_k.dtype == result[k].dtype:
+                            result.loc[:, k] = res_k
+                        else:
+                            # Different dtype -> no way to do inplace.
+                            result[k] = res_k
+                    else:
+                        # see test_fillna_dict_inplace_nonunique_columns
+                        locs = result.columns.get_loc(k)
+                        if isinstance(locs, slice):
+                            locs = range(self.shape[1])[locs]
+                        elif isinstance(locs, np.ndarray) and locs.dtype.kind == "b":
+                            locs = locs.nonzero()[0]
+                        elif not (
+                            isinstance(locs, np.ndarray) and locs.dtype.kind == "i"
+                        ):
+                            # Should never be reached, but let's cover our bases
+                            raise NotImplementedError(
+                                "Unexpected get_loc result, please report a bug at "
+                                "https://github.com/pandas-dev/pandas"
+                            )
+
+                        for i, loc in enumerate(locs):
+                            res_loc = res_k.iloc[:, i]
+                            target = self.iloc[:, loc]
+
+                            if res_loc.dtype == target.dtype:
+                                result.iloc[:, loc] = res_loc
+                            else:
+                                result.isetitem(loc, res_loc)
+            if inplace:
+                return self._update_inplace(result)
+            else:
+                return result
+
+        elif not is_list_like(value):
+            if axis == 1:
+                result = self.T.fillna(value=value, limit=limit).T
+                new_data = result._mgr
+            else:
+                new_data = self._mgr.fillna(value=value, limit=limit, inplace=inplace)
+        elif isinstance(value, ABCDataFrame) and self.ndim == 2:
+            new_data = self.where(self.notna(), value)._mgr
+        else:
+            raise ValueError(f"invalid fill value with a {type(value)}")
+
+        result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="fillna")
+
+    @overload
+    def ffill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: Literal[False] = ...,
+        limit: None | int = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+    ) -> Self: ...
+
+    @overload
+    def ffill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: Literal[True],
+        limit: None | int = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+    ) -> None: ...
+
+    @overload
+    def ffill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: bool = ...,
+        limit: None | int = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+    )
+    def ffill(
+        self,
+        *,
+        axis: None | Axis = None,
+        inplace: bool = False,
+        limit: None | int = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+    ) -> Self | None:
+        """
+        Fill NA/NaN values by propagating the last valid observation to next valid.
+
+        Parameters
+        ----------
+        axis : {axes_single_arg}
+            Axis along which to fill missing values. For `Series`
+            this parameter is unused and defaults to 0.
+        inplace : bool, default False
+            If True, fill in-place. Note: this will modify any
+            other views on this object (e.g., a no-copy slice for a column in a
+            DataFrame).
+        limit : int, default None
+            If method is specified, this is the maximum number of consecutive
+            NaN values to forward/backward fill. In other words, if there is
+            a gap with more than this number of consecutive NaNs, it will only
+            be partially filled. If method is not specified, this is the
+            maximum number of entries along the entire axis where NaNs will be
+            filled. Must be greater than 0 if not None.
+        limit_area : {{`None`, 'inside', 'outside'}}, default None
+            If limit is specified, consecutive NaNs will be filled with this
+            restriction.
+
+            * ``None``: No fill restriction.
+            * 'inside': Only fill NaNs surrounded by valid values
+              (interpolate).
+            * 'outside': Only fill NaNs outside valid values (extrapolate).
+
+            .. versionadded:: 2.2.0
+
+        Returns
+        -------
+        {klass} or None
+            Object with missing values filled or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.bfill : Fill NA/NaN values by using the next valid observation
+            to fill the gap.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         [np.nan, 2, np.nan, 0],
+        ...         [3, 4, np.nan, 1],
+        ...         [np.nan, np.nan, np.nan, np.nan],
+        ...         [np.nan, 3, np.nan, 4],
+        ...     ],
+        ...     columns=list("ABCD"),
+        ... )
+        >>> df
+             A    B   C    D
+        0  NaN  2.0 NaN  0.0
+        1  3.0  4.0 NaN  1.0
+        2  NaN  NaN NaN  NaN
+        3  NaN  3.0 NaN  4.0
+
+        >>> df.ffill()
+             A    B   C    D
+        0  NaN  2.0 NaN  0.0
+        1  3.0  4.0 NaN  1.0
+        2  3.0  4.0 NaN  1.0
+        3  3.0  3.0 NaN  4.0
+
+        >>> ser = pd.Series([1, np.nan, 2, 3])
+        >>> ser.ffill()
+        0   1.0
+        1   1.0
+        2   2.0
+        3   3.0
+        dtype: float64
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        return self._pad_or_backfill(
+            "ffill",
+            axis=axis,
+            inplace=inplace,
+            limit=limit,
+            limit_area=limit_area,
+        )
+
+    @overload
+    def bfill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: Literal[False] = ...,
+        limit: None | int = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+    ) -> Self: ...
+
+    @overload
+    def bfill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: Literal[True],
+        limit: None | int = ...,
+    ) -> None: ...
+
+    @overload
+    def bfill(
+        self,
+        *,
+        axis: None | Axis = ...,
+        inplace: bool = ...,
+        limit: None | int = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+    )
+    def bfill(
+        self,
+        *,
+        axis: None | Axis = None,
+        inplace: bool = False,
+        limit: None | int = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+    ) -> Self | None:
+        """
+        Fill NA/NaN values by using the next valid observation to fill the gap.
+
+        Parameters
+        ----------
+        axis : {axes_single_arg}
+            Axis along which to fill missing values. For `Series`
+            this parameter is unused and defaults to 0.
+        inplace : bool, default False
+            If True, fill in-place. Note: this will modify any
+            other views on this object (e.g., a no-copy slice for a column in a
+            DataFrame).
+        limit : int, default None
+            If method is specified, this is the maximum number of consecutive
+            NaN values to forward/backward fill. In other words, if there is
+            a gap with more than this number of consecutive NaNs, it will only
+            be partially filled. If method is not specified, this is the
+            maximum number of entries along the entire axis where NaNs will be
+            filled. Must be greater than 0 if not None.
+        limit_area : {{`None`, 'inside', 'outside'}}, default None
+            If limit is specified, consecutive NaNs will be filled with this
+            restriction.
+
+            * ``None``: No fill restriction.
+            * 'inside': Only fill NaNs surrounded by valid values
+              (interpolate).
+            * 'outside': Only fill NaNs outside valid values (extrapolate).
+
+            .. versionadded:: 2.2.0
+
+        Returns
+        -------
+        {klass} or None
+            Object with missing values filled or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.ffill : Fill NA/NaN values by propagating the last valid
+            observation to next valid.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series([1, None, None, 2])
+        >>> s.bfill()
+        0    1.0
+        1    2.0
+        2    2.0
+        3    2.0
+        dtype: float64
+        >>> s.bfill(limit=1)
+        0    1.0
+        1    NaN
+        2    2.0
+        3    2.0
+        dtype: float64
+
+        With DataFrame:
+
+        >>> df = pd.DataFrame({{"A": [1, None, None, 4], "B": [None, 5, None, 7]}})
+        >>> df
+              A     B
+        0   1.0	  NaN
+        1   NaN	  5.0
+        2   NaN   NaN
+        3   4.0   7.0
+        >>> df.bfill()
+              A     B
+        0   1.0   5.0
+        1   4.0   5.0
+        2   4.0   7.0
+        3   4.0   7.0
+        >>> df.bfill(limit=1)
+              A     B
+        0   1.0   5.0
+        1   NaN   5.0
+        2   4.0   7.0
+        3   4.0   7.0
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        return self._pad_or_backfill(
+            "bfill",
+            axis=axis,
+            inplace=inplace,
+            limit=limit,
+            limit_area=limit_area,
+        )
+
+    @overload
+    def replace(
+        self,
+        to_replace=...,
+        value=...,
+        *,
+        inplace: Literal[False] = ...,
+        regex: bool = ...,
+    ) -> Self: ...
+
+    @overload
+    def replace(
+        self,
+        to_replace=...,
+        value=...,
+        *,
+        inplace: Literal[True],
+        regex: bool = ...,
+    ) -> None: ...
+
+    @overload
+    def replace(
+        self,
+        to_replace=...,
+        value=...,
+        *,
+        inplace: bool = ...,
+        regex: bool = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        _shared_docs["replace"],
+        klass=_shared_doc_kwargs["klass"],
+        inplace=_shared_doc_kwargs["inplace"],
+    )
+    def replace(
+        self,
+        to_replace=None,
+        value=lib.no_default,
+        *,
+        inplace: bool = False,
+        regex: bool = False,
+    ) -> Self | None:
+        if not is_bool(regex) and to_replace is not None:
+            raise ValueError("'to_replace' must be 'None' if 'regex' is not a bool")
+
+        if not (
+            is_scalar(to_replace)
+            or is_re_compilable(to_replace)
+            or is_list_like(to_replace)
+        ):
+            raise TypeError(
+                "Expecting 'to_replace' to be either a scalar, array-like, "
+                "dict or None, got invalid type "
+                f"{type(to_replace).__name__!r}"
+            )
+
+        if value is lib.no_default and not (
+            is_dict_like(to_replace) or is_dict_like(regex)
+        ):
+            raise ValueError(
+                # GH#33302
+                f"{type(self).__name__}.replace must specify either 'value', "
+                "a dict-like 'to_replace', or dict-like 'regex'."
+            )
+
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        if value is lib.no_default:
+            if not is_dict_like(to_replace):
+                # In this case we have checked above that
+                #  1) regex is dict-like and 2) to_replace is None
+                to_replace = regex
+                regex = True
+
+            items = list(to_replace.items())
+            if items:
+                keys, values = zip(*items)
+            else:
+                keys, values = ([], [])  # type: ignore[assignment]
+
+            are_mappings = [is_dict_like(v) for v in values]
+
+            if any(are_mappings):
+                if not all(are_mappings):
+                    raise TypeError(
+                        "If a nested mapping is passed, all values "
+                        "of the top level mapping must be mappings"
+                    )
+                # passed a nested dict/Series
+                to_rep_dict = {}
+                value_dict = {}
+
+                for k, v in items:
+                    # error: Incompatible types in assignment (expression has type
+                    # "list[Never]", variable has type "tuple[Any, ...]")
+                    keys, values = list(zip(*v.items())) or (  # type: ignore[assignment]
+                        [],
+                        [],
+                    )
+
+                    to_rep_dict[k] = list(keys)
+                    value_dict[k] = list(values)
+
+                to_replace, value = to_rep_dict, value_dict
+            else:
+                to_replace, value = keys, values
+
+            return self.replace(to_replace, value, inplace=inplace, regex=regex)
+        else:
+            # need a non-zero len on all axes
+            if not self.size:
+                if inplace:
+                    return None
+                return self.copy(deep=False)
+            if is_dict_like(to_replace):
+                if is_dict_like(value):  # {'A' : NA} -> {'A' : 0}
+                    if isinstance(self, ABCSeries):
+                        raise ValueError(
+                            "to_replace and value cannot be dict-like for "
+                            "Series.replace"
+                        )
+                    # Note: Checking below for `in foo.keys()` instead of
+                    #  `in foo` is needed for when we have a Series and not dict
+                    mapping = {
+                        col: (to_replace[col], value[col])
+                        for col in to_replace.keys()
+                        if col in value.keys() and col in self
+                    }
+                    return self._replace_columnwise(mapping, inplace, regex)
+
+                # {'A': NA} -> 0
+                elif not is_list_like(value):
+                    # Operate column-wise
+                    if self.ndim == 1:
+                        raise ValueError(
+                            "Series.replace cannot use dict-like to_replace "
+                            "and non-None value"
+                        )
+                    mapping = {
+                        col: (to_rep, value) for col, to_rep in to_replace.items()
+                    }
+                    return self._replace_columnwise(mapping, inplace, regex)
+                else:
+                    raise TypeError("value argument must be scalar, dict, or Series")
+
+            elif is_list_like(to_replace):
+                if not is_list_like(value):
+                    # e.g. to_replace = [NA, ''] and value is 0,
+                    #  so we replace NA with 0 and then replace '' with 0
+                    value = [value] * len(to_replace)
+
+                # e.g. we have to_replace = [NA, ''] and value = [0, 'missing']
+                if len(to_replace) != len(value):
+                    raise ValueError(
+                        f"Replacement lists must match in length. "
+                        f"Expecting {len(to_replace)} got {len(value)} "
+                    )
+                new_data = self._mgr.replace_list(
+                    src_list=to_replace,
+                    dest_list=value,
+                    inplace=inplace,
+                    regex=regex,
+                )
+
+            elif to_replace is None:
+                if not (
+                    is_re_compilable(regex)
+                    or is_list_like(regex)
+                    or is_dict_like(regex)
+                ):
+                    raise TypeError(
+                        f"'regex' must be a string or a compiled regular expression "
+                        f"or a list or dict of strings or regular expressions, "
+                        f"you passed a {type(regex).__name__!r}"
+                    )
+                return self.replace(regex, value, inplace=inplace, regex=True)
+            else:
+                # dest iterable dict-like
+                if is_dict_like(value):  # NA -> {'A' : 0, 'B' : -1}
+                    # Operate column-wise
+                    if self.ndim == 1:
+                        raise ValueError(
+                            "Series.replace cannot use dict-value and "
+                            "non-None to_replace"
+                        )
+                    mapping = {col: (to_replace, val) for col, val in value.items()}
+                    return self._replace_columnwise(mapping, inplace, regex)
+
+                elif not is_list_like(value):  # NA -> 0
+                    regex = should_use_regex(regex, to_replace)
+                    if regex:
+                        new_data = self._mgr.replace_regex(
+                            to_replace=to_replace,
+                            value=value,
+                            inplace=inplace,
+                        )
+                    else:
+                        new_data = self._mgr.replace(
+                            to_replace=to_replace, value=value, inplace=inplace
+                        )
+                else:
+                    raise TypeError(
+                        f'Invalid "to_replace" type: {type(to_replace).__name__!r}'
+                    )
+
+        result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="replace")
+
+    @overload
+    def interpolate(
+        self,
+        method: InterpolateOptions = ...,
+        *,
+        axis: Axis = ...,
+        limit: int | None = ...,
+        inplace: Literal[False] = ...,
+        limit_direction: Literal["forward", "backward", "both"] | None = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+        **kwargs,
+    ) -> Self: ...
+
+    @overload
+    def interpolate(
+        self,
+        method: InterpolateOptions = ...,
+        *,
+        axis: Axis = ...,
+        limit: int | None = ...,
+        inplace: Literal[True],
+        limit_direction: Literal["forward", "backward", "both"] | None = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+        **kwargs,
+    ) -> None: ...
+
+    @overload
+    def interpolate(
+        self,
+        method: InterpolateOptions = ...,
+        *,
+        axis: Axis = ...,
+        limit: int | None = ...,
+        inplace: bool = ...,
+        limit_direction: Literal["forward", "backward", "both"] | None = ...,
+        limit_area: Literal["inside", "outside"] | None = ...,
+        **kwargs,
+    ) -> Self | None: ...
+
+    @final
+    def interpolate(
+        self,
+        method: InterpolateOptions = "linear",
+        *,
+        axis: Axis = 0,
+        limit: int | None = None,
+        inplace: bool = False,
+        limit_direction: Literal["forward", "backward", "both"] | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        **kwargs,
+    ) -> Self | None:
+        """
+        Fill NaN values using an interpolation method.
+
+        Please note that only ``method='linear'`` is supported for
+        DataFrame/Series with a MultiIndex.
+
+        Parameters
+        ----------
+        method : str, default 'linear'
+            Interpolation technique to use. One of:
+
+            * 'linear': Ignore the index and treat the values as equally
+              spaced. This is the only method supported on MultiIndexes.
+            * 'time': Works on daily and higher resolution data to interpolate
+              given length of interval. This interpolates values based on
+              time interval between observations.
+            * 'index': The interpolation uses the numerical values
+              of the DataFrame's index to linearly calculate missing values.
+            * 'values': Interpolation based on the numerical values
+              in the DataFrame, treating them as equally spaced along the index.
+            * 'nearest', 'zero', 'slinear', 'quadratic', 'cubic',
+              'barycentric', 'polynomial': Passed to
+              `scipy.interpolate.interp1d`, whereas 'spline' is passed to
+              `scipy.interpolate.UnivariateSpline`. These methods use the numerical
+              values of the index.  Both 'polynomial' and 'spline' require that
+              you also specify an `order` (int), e.g.
+              ``df.interpolate(method='polynomial', order=5)``. Note that,
+              `slinear` method in Pandas refers to the Scipy first order `spline`
+              instead of Pandas first order `spline`.
+            * 'krogh', 'piecewise_polynomial', 'spline', 'pchip', 'akima',
+              'cubicspline': Wrappers around the SciPy interpolation methods of
+              similar names. See `Notes`.
+            * 'from_derivatives': Refers to
+              `scipy.interpolate.BPoly.from_derivatives`.
+
+        axis : {{0 or 'index', 1 or 'columns', None}}, default None
+            Axis to interpolate along. For `Series` this parameter is unused
+            and defaults to 0.
+        limit : int, optional
+            Maximum number of consecutive NaNs to fill. Must be greater than
+            0.
+        inplace : bool, default False
+            Update the data in place if possible.
+        limit_direction : {{'forward', 'backward', 'both'}}, optional, default 'forward'
+            Consecutive NaNs will be filled in this direction.
+
+        limit_area : {{`None`, 'inside', 'outside'}}, default None
+            If limit is specified, consecutive NaNs will be filled with this
+            restriction.
+
+            * ``None``: No fill restriction.
+            * 'inside': Only fill NaNs surrounded by valid values
+              (interpolate).
+            * 'outside': Only fill NaNs outside valid values (extrapolate).
+
+        **kwargs : optional
+            Keyword arguments to pass on to the interpolating function.
+
+        Returns
+        -------
+        Series or DataFrame or None
+            Returns the same object type as the caller, interpolated at
+            some or all ``NaN`` values or None if ``inplace=True``.
+
+        See Also
+        --------
+        fillna : Fill missing values using different methods.
+        scipy.interpolate.Akima1DInterpolator : Piecewise cubic polynomials
+            (Akima interpolator).
+        scipy.interpolate.BPoly.from_derivatives : Piecewise polynomial in the
+            Bernstein basis.
+        scipy.interpolate.interp1d : Interpolate a 1-D function.
+        scipy.interpolate.KroghInterpolator : Interpolate polynomial (Krogh
+            interpolator).
+        scipy.interpolate.PchipInterpolator : PCHIP 1-d monotonic cubic
+            interpolation.
+        scipy.interpolate.CubicSpline : Cubic spline data interpolator.
+
+        Notes
+        -----
+        The 'krogh', 'piecewise_polynomial', 'spline', 'pchip' and 'akima'
+        methods are wrappers around the respective SciPy implementations of
+        similar names. These use the actual numerical values of the index.
+        For more information on their behavior, see the
+        `SciPy documentation
+        <https://docs.scipy.org/doc/scipy/reference/interpolate.html#univariate-interpolation>`__.
+
+        Examples
+        --------
+        Filling in ``NaN`` in a :class:`~pandas.Series` via linear
+        interpolation.
+
+        >>> s = pd.Series([0, 1, np.nan, 3])
+        >>> s
+        0    0.0
+        1    1.0
+        2    NaN
+        3    3.0
+        dtype: float64
+        >>> s.interpolate()
+        0    0.0
+        1    1.0
+        2    2.0
+        3    3.0
+        dtype: float64
+
+        Filling in ``NaN`` in a Series via polynomial interpolation or splines:
+        Both 'polynomial' and 'spline' methods require that you also specify
+        an ``order`` (int).
+
+        >>> s = pd.Series([0, 2, np.nan, 8])
+        >>> s.interpolate(method="polynomial", order=2)
+        0    0.000000
+        1    2.000000
+        2    4.666667
+        3    8.000000
+        dtype: float64
+
+        Fill the DataFrame forward (that is, going down) along each column
+        using linear interpolation.
+
+        Note how the last entry in column 'a' is interpolated differently,
+        because there is no entry after it to use for interpolation.
+        Note how the first entry in column 'b' remains ``NaN``, because there
+        is no entry before it to use for interpolation.
+
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         (0.0, np.nan, -1.0, 1.0),
+        ...         (np.nan, 2.0, np.nan, np.nan),
+        ...         (2.0, 3.0, np.nan, 9.0),
+        ...         (np.nan, 4.0, -4.0, 16.0),
+        ...     ],
+        ...     columns=list("abcd"),
+        ... )
+        >>> df
+             a    b    c     d
+        0  0.0  NaN -1.0   1.0
+        1  NaN  2.0  NaN   NaN
+        2  2.0  3.0  NaN   9.0
+        3  NaN  4.0 -4.0  16.0
+        >>> df.interpolate(method="linear", limit_direction="forward", axis=0)
+             a    b    c     d
+        0  0.0  NaN -1.0   1.0
+        1  1.0  2.0 -2.0   5.0
+        2  2.0  3.0 -3.0   9.0
+        3  2.0  4.0 -4.0  16.0
+
+        Using polynomial interpolation.
+
+        >>> df["d"].interpolate(method="polynomial", order=2)
+        0     1.0
+        1     4.0
+        2     9.0
+        3    16.0
+        Name: d, dtype: float64
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        axis = self._get_axis_number(axis)
+
+        if self.empty:
+            if inplace:
+                return None
+            return self.copy()
+
+        if not isinstance(method, str):
+            raise ValueError("'method' should be a string, not None.")
+
+        obj, should_transpose = (self.T, True) if axis == 1 else (self, False)
+
+        if isinstance(obj.index, MultiIndex) and method != "linear":
+            raise ValueError(
+                "Only `method=linear` interpolation is supported on MultiIndexes."
+            )
+
+        limit_direction = missing.infer_limit_direction(limit_direction, method)
+
+        index = missing.get_interp_index(method, obj.index)
+        new_data = obj._mgr.interpolate(
+            method=method,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            inplace=inplace,
+            **kwargs,
+        )
+
+        result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+        if should_transpose:
+            result = result.T
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result.__finalize__(self, method="interpolate")
+
+    # ----------------------------------------------------------------------
+    # Timeseries methods Methods
+
+    @final
+    def asof(self, where, subset=None):
+        """
+        Return the last row(s) without any NaNs before `where`.
+
+        The last row (for each element in `where`, if list) without any
+        NaN is taken.
+        In case of a :class:`~pandas.DataFrame`, the last row without NaN
+        considering only the subset of columns (if not `None`)
+
+        If there is no good value, NaN is returned for a Series or
+        a Series of NaN values for a DataFrame
+
+        Parameters
+        ----------
+        where : date or array-like of dates
+            Date(s) before which the last row(s) are returned.
+        subset : str or array-like of str, default `None`
+            For DataFrame, if not `None`, only use these columns to
+            check for NaNs.
+
+        Returns
+        -------
+        scalar, Series, or DataFrame
+
+            The return can be:
+
+            * scalar : when `self` is a Series and `where` is a scalar
+            * Series: when `self` is a Series and `where` is an array-like,
+              or when `self` is a DataFrame and `where` is a scalar
+            * DataFrame : when `self` is a DataFrame and `where` is an
+              array-like
+
+        See Also
+        --------
+        merge_asof : Perform an asof merge. Similar to left join.
+
+        Notes
+        -----
+        Dates are assumed to be sorted. Raises if this is not the case.
+
+        Examples
+        --------
+        A Series and a scalar `where`.
+
+        >>> s = pd.Series([1, 2, np.nan, 4], index=[10, 20, 30, 40])
+        >>> s
+        10    1.0
+        20    2.0
+        30    NaN
+        40    4.0
+        dtype: float64
+
+        >>> s.asof(20)
+        2.0
+
+        For a sequence `where`, a Series is returned. The first value is
+        NaN, because the first element of `where` is before the first
+        index value.
+
+        >>> s.asof([5, 20])
+        5     NaN
+        20    2.0
+        dtype: float64
+
+        Missing values are not considered. The following is ``2.0``, not
+        NaN, even though NaN is at the index location for ``30``.
+
+        >>> s.asof(30)
+        2.0
+
+        Take all columns into consideration
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "a": [10.0, 20.0, 30.0, 40.0, 50.0],
+        ...         "b": [None, None, None, None, 500],
+        ...     },
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2018-02-27 09:01:00",
+        ...             "2018-02-27 09:02:00",
+        ...             "2018-02-27 09:03:00",
+        ...             "2018-02-27 09:04:00",
+        ...             "2018-02-27 09:05:00",
+        ...         ]
+        ...     ),
+        ... )
+        >>> df.asof(pd.DatetimeIndex(["2018-02-27 09:03:30", "2018-02-27 09:04:30"]))
+                              a   b
+        2018-02-27 09:03:30 NaN NaN
+        2018-02-27 09:04:30 NaN NaN
+
+        Take a single column into consideration
+
+        >>> df.asof(
+        ...     pd.DatetimeIndex(["2018-02-27 09:03:30", "2018-02-27 09:04:30"]),
+        ...     subset=["a"],
+        ... )
+                                a   b
+        2018-02-27 09:03:30  30.0 NaN
+        2018-02-27 09:04:30  40.0 NaN
+        """
+        if isinstance(where, str):
+            where = Timestamp(where)
+
+        if not self.index.is_monotonic_increasing:
+            raise ValueError("asof requires a sorted index")
+
+        is_series = isinstance(self, ABCSeries)
+        if is_series:
+            if subset is not None:
+                raise ValueError("subset is not valid for Series")
+        else:
+            if subset is None:
+                subset = self.columns
+            if not is_list_like(subset):
+                subset = [subset]
+
+        is_list = is_list_like(where)
+        if not is_list:
+            start = self.index[0]
+            if isinstance(self.index, PeriodIndex):
+                where = Period(where, freq=self.index.freq)
+
+            if where < start:
+                if not is_series:
+                    return self._constructor_sliced(
+                        index=self.columns, name=where, dtype=np.float64
+                    )
+                return np.nan
+
+            # It's always much faster to use a *while* loop here for
+            # Series than pre-computing all the NAs. However a
+            # *while* loop is extremely expensive for DataFrame
+            # so we later pre-compute all the NAs and use the same
+            # code path whether *where* is a scalar or list.
+            # See PR: https://github.com/pandas-dev/pandas/pull/14476
+            if is_series:
+                loc = self.index.searchsorted(where, side="right")
+                if loc > 0:
+                    loc -= 1
+
+                values = self._values
+                while loc > 0 and isna(values[loc]):
+                    loc -= 1
+                return values[loc]
+
+        if not isinstance(where, Index):
+            where = Index(where) if is_list else Index([where])
+
+        nulls = self.isna() if is_series else self[subset].isna().any(axis=1)
+        if nulls.all():
+            if is_series:
+                self = cast("Series", self)
+                return self._constructor(np.nan, index=where, name=self.name)
+            elif is_list:
+                self = cast("DataFrame", self)
+                return self._constructor(np.nan, index=where, columns=self.columns)
+            else:
+                self = cast("DataFrame", self)
+                return self._constructor_sliced(
+                    np.nan, index=self.columns, name=where[0]
+                )
+
+        # error: Unsupported operand type for
+        # ~ ("ExtensionArray | ndarray[Any, Any] | Any")
+        locs = self.index.asof_locs(where, ~nulls._values)  # type: ignore[operator]
+
+        # mask the missing
+        mask = locs == -1
+        data = self.take(locs)
+        data.index = where
+        if mask.any():
+            # GH#16063 only do this setting when necessary, otherwise
+            #  we'd cast e.g. bools to floats
+            data.loc[mask] = np.nan
+        return data if is_list else data.iloc[-1]
+
+    # ----------------------------------------------------------------------
+    # Action Methods
+
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def isna(self) -> Self:
+        """
+        Detect missing values.
+
+        Return a boolean same-sized object indicating if the values are NA.
+        NA values, such as None or :attr:`numpy.NaN`, gets mapped to True
+        values.
+        Everything else gets mapped to False values. Characters such as empty
+        strings ``''`` or :attr:`numpy.inf` are not considered NA values.
+
+        Returns
+        -------
+        {klass}
+            Mask of bool values for each element in {klass} that
+            indicates whether an element is an NA value.
+
+        See Also
+        --------
+        {klass}.isnull : Alias of isna.
+        {klass}.notna : Boolean inverse of isna.
+        {klass}.dropna : Omit axes labels with missing values.
+        isna : Top-level isna.
+
+        Examples
+        --------
+        Show which entries in a DataFrame are NA.
+
+        >>> df = pd.DataFrame(
+        ...     dict(
+        ...         age=[5, 6, np.nan],
+        ...         born=[
+        ...             pd.NaT,
+        ...             pd.Timestamp("1939-05-27"),
+        ...             pd.Timestamp("1940-04-25"),
+        ...         ],
+        ...         name=["Alfred", "Batman", ""],
+        ...         toy=[None, "Batmobile", "Joker"],
+        ...     )
+        ... )
+        >>> df
+           age       born    name        toy
+        0  5.0        NaT  Alfred       None
+        1  6.0 1939-05-27  Batman  Batmobile
+        2  NaN 1940-04-25              Joker
+
+        >>> df.isna()
+             age   born   name    toy
+        0  False   True  False   True
+        1  False  False  False  False
+        2   True  False  False  False
+
+        Show which entries in a Series are NA.
+
+        >>> ser = pd.Series([5, 6, np.nan])
+        >>> ser
+        0    5.0
+        1    6.0
+        2    NaN
+        dtype: float64
+
+        >>> ser.isna()
+        0    False
+        1    False
+        2     True
+        dtype: bool
+        """
+        return isna(self).__finalize__(self, method="isna")
+
+    @doc(isna, klass=_shared_doc_kwargs["klass"])
+    def isnull(self) -> Self:
+        return isna(self).__finalize__(self, method="isnull")
+
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def notna(self) -> Self:
+        """
+        Detect existing (non-missing) values.
+
+        Return a boolean same-sized object indicating if the values are not NA.
+        Non-missing values get mapped to True. Characters such as empty
+        strings ``''`` or :attr:`numpy.inf` are not considered NA values.
+        NA values, such as None or :attr:`numpy.NaN`, get mapped to False
+        values.
+
+        Returns
+        -------
+        {klass}
+            Mask of bool values for each element in {klass} that
+            indicates whether an element is not an NA value.
+
+        See Also
+        --------
+        {klass}.notnull : Alias of notna.
+        {klass}.isna : Boolean inverse of notna.
+        {klass}.dropna : Omit axes labels with missing values.
+        notna : Top-level notna.
+
+        Examples
+        --------
+        Show which entries in a DataFrame are not NA.
+
+        >>> df = pd.DataFrame(
+        ...     dict(
+        ...         age=[5, 6, np.nan],
+        ...         born=[
+        ...             pd.NaT,
+        ...             pd.Timestamp("1939-05-27"),
+        ...             pd.Timestamp("1940-04-25"),
+        ...         ],
+        ...         name=["Alfred", "Batman", ""],
+        ...         toy=[None, "Batmobile", "Joker"],
+        ...     )
+        ... )
+        >>> df
+           age       born    name        toy
+        0  5.0        NaT  Alfred       None
+        1  6.0 1939-05-27  Batman  Batmobile
+        2  NaN 1940-04-25              Joker
+
+        >>> df.notna()
+             age   born  name    toy
+        0   True  False  True  False
+        1   True   True  True   True
+        2  False   True  True   True
+
+        Show which entries in a Series are not NA.
+
+        >>> ser = pd.Series([5, 6, np.nan])
+        >>> ser
+        0    5.0
+        1    6.0
+        2    NaN
+        dtype: float64
+
+        >>> ser.notna()
+        0     True
+        1     True
+        2    False
+        dtype: bool
+        """
+        return notna(self).__finalize__(self, method="notna")
+
+    @doc(notna, klass=_shared_doc_kwargs["klass"])
+    def notnull(self) -> Self:
+        return notna(self).__finalize__(self, method="notnull")
+
+    @final
+    def _clip_with_scalar(self, lower, upper, inplace: bool = False):
+        if (lower is not None and np.any(isna(lower))) or (
+            upper is not None and np.any(isna(upper))
+        ):
+            raise ValueError("Cannot use an NA value as a clip threshold")
+
+        result = self
+        mask = self.isna()
+
+        if lower is not None:
+            cond = mask | (self >= lower)
+            result = result.where(cond, lower, inplace=inplace)  # type: ignore[assignment]
+        if upper is not None:
+            cond = mask | (self <= upper)
+            result = self if inplace else result
+            result = result.where(cond, upper, inplace=inplace)  # type: ignore[assignment]
+
+        return result
+
+    @final
+    def _clip_with_one_bound(self, threshold, method, axis, inplace):
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+
+        # method is self.le for upper bound and self.ge for lower bound
+        if is_scalar(threshold) and is_number(threshold):
+            if method.__name__ == "le":
+                return self._clip_with_scalar(None, threshold, inplace=inplace)
+            return self._clip_with_scalar(threshold, None, inplace=inplace)
+
+        # GH #15390
+        # In order for where method to work, the threshold must
+        # be transformed to NDFrame from other array like structure.
+        if (not isinstance(threshold, ABCSeries)) and is_list_like(threshold):
+            if isinstance(self, ABCSeries):
+                threshold = self._constructor(threshold, index=self.index)
+            else:
+                threshold = self._align_for_op(threshold, axis, flex=None)[1]
+
+        # GH 40420
+        # Treat missing thresholds as no bounds, not clipping the values
+        if is_list_like(threshold):
+            fill_value = np.inf if method.__name__ == "le" else -np.inf
+            threshold_inf = threshold.fillna(fill_value)
+        else:
+            threshold_inf = threshold
+
+        subset = method(threshold_inf, axis=axis) | isna(self)
+
+        # GH 40420
+        return self.where(subset, threshold, axis=axis, inplace=inplace)
+
+    @overload
+    def clip(
+        self,
+        lower=...,
+        upper=...,
+        *,
+        axis: Axis | None = ...,
+        inplace: Literal[False] = ...,
+        **kwargs,
+    ) -> Self: ...
+
+    @overload
+    def clip(
+        self,
+        lower=...,
+        upper=...,
+        *,
+        axis: Axis | None = ...,
+        inplace: Literal[True],
+        **kwargs,
+    ) -> None: ...
+
+    @overload
+    def clip(
+        self,
+        lower=...,
+        upper=...,
+        *,
+        axis: Axis | None = ...,
+        inplace: bool = ...,
+        **kwargs,
+    ) -> Self | None: ...
+
+    @final
+    def clip(
+        self,
+        lower=None,
+        upper=None,
+        *,
+        axis: Axis | None = None,
+        inplace: bool = False,
+        **kwargs,
+    ) -> Self | None:
+        """
+        Trim values at input threshold(s).
+
+        Assigns values outside boundary to boundary values. Thresholds
+        can be singular values or array like, and in the latter case
+        the clipping is performed element-wise in the specified axis.
+
+        Parameters
+        ----------
+        lower : float or array-like, default None
+            Minimum threshold value. All values below this
+            threshold will be set to it. A missing
+            threshold (e.g `NA`) will not clip the value.
+        upper : float or array-like, default None
+            Maximum threshold value. All values above this
+            threshold will be set to it. A missing
+            threshold (e.g `NA`) will not clip the value.
+        axis : {{0 or 'index', 1 or 'columns', None}}, default None
+            Align object with lower and upper along the given axis.
+            For `Series` this parameter is unused and defaults to `None`.
+        inplace : bool, default False
+            Whether to perform the operation in place on the data.
+        **kwargs
+            Additional keywords have no effect but might be accepted
+            for compatibility with numpy.
+
+        Returns
+        -------
+        Series or DataFrame or None
+            Same type as calling object with the values outside the
+            clip boundaries replaced or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.clip : Trim values at input threshold in series.
+        DataFrame.clip : Trim values at input threshold in DataFrame.
+        numpy.clip : Clip (limit) the values in an array.
+
+        Examples
+        --------
+        >>> data = {"col_0": [9, -3, 0, -1, 5], "col_1": [-2, -7, 6, 8, -5]}
+        >>> df = pd.DataFrame(data)
+        >>> df
+           col_0  col_1
+        0      9     -2
+        1     -3     -7
+        2      0      6
+        3     -1      8
+        4      5     -5
+
+        Clips per column using lower and upper thresholds:
+
+        >>> df.clip(-4, 6)
+           col_0  col_1
+        0      6     -2
+        1     -3     -4
+        2      0      6
+        3     -1      6
+        4      5     -4
+
+        Clips using specific lower and upper thresholds per column:
+
+        >>> df.clip([-2, -1], [4, 5])
+            col_0  col_1
+        0      4     -1
+        1     -2     -1
+        2      0      5
+        3     -1      5
+        4      4     -1
+
+        Clips using specific lower and upper thresholds per column element:
+
+        >>> t = pd.Series([2, -4, -1, 6, 3])
+        >>> t
+        0    2
+        1   -4
+        2   -1
+        3    6
+        4    3
+        dtype: int64
+
+        >>> df.clip(t, t + 4, axis=0)
+           col_0  col_1
+        0      6      2
+        1     -3     -4
+        2      0      3
+        3      6      8
+        4      5      3
+
+        Clips using specific lower threshold per column element, with missing values:
+
+        >>> t = pd.Series([2, -4, np.nan, 6, 3])
+        >>> t
+        0    2.0
+        1   -4.0
+        2    NaN
+        3    6.0
+        4    3.0
+        dtype: float64
+
+        >>> df.clip(t, axis=0)
+        col_0  col_1
+        0      9.0    2.0
+        1     -3.0   -4.0
+        2      0.0    6.0
+        3      6.0    8.0
+        4      5.0    3.0
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        axis = nv.validate_clip_with_axis(axis, (), kwargs)
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+
+        # GH 17276
+        # numpy doesn't like NaN as a clip value
+        # so ignore
+        # GH 19992
+        # numpy doesn't drop a list-like bound containing NaN
+        isna_lower = isna(lower)
+        if not is_list_like(lower):
+            if np.any(isna_lower):
+                lower = None
+        elif np.all(isna_lower):
+            lower = None
+        isna_upper = isna(upper)
+        if not is_list_like(upper):
+            if np.any(isna_upper):
+                upper = None
+        elif np.all(isna_upper):
+            upper = None
+
+        # GH 2747 (arguments were reversed)
+        if (
+            lower is not None
+            and upper is not None
+            and is_scalar(lower)
+            and is_scalar(upper)
+        ):
+            lower, upper = min(lower, upper), max(lower, upper)
+
+        # fast-path for scalars
+        if (lower is None or is_number(lower)) and (upper is None or is_number(upper)):
+            return self._clip_with_scalar(lower, upper, inplace=inplace)
+
+        result = self
+        if lower is not None:
+            result = result._clip_with_one_bound(
+                lower, method=self.ge, axis=axis, inplace=inplace
+            )
+        if upper is not None:
+            if inplace:
+                result = self
+            result = result._clip_with_one_bound(
+                upper, method=self.le, axis=axis, inplace=inplace
+            )
+
+        return result
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def asfreq(
+        self,
+        freq: Frequency,
+        method: FillnaOptions | None = None,
+        how: Literal["start", "end"] | None = None,
+        normalize: bool = False,
+        fill_value: Hashable | None = None,
+    ) -> Self:
+        """
+        Convert time series to specified frequency.
+
+        Returns the original data conformed to a new index with the specified
+        frequency.
+
+        If the index of this {klass} is a :class:`~pandas.PeriodIndex`, the new index
+        is the result of transforming the original index with
+        :meth:`PeriodIndex.asfreq <pandas.PeriodIndex.asfreq>` (so the original index
+        will map one-to-one to the new index).
+
+        Otherwise, the new index will be equivalent to ``pd.date_range(start, end,
+        freq=freq)`` where ``start`` and ``end`` are, respectively, the min and
+        max entries in the original index (see :func:`pandas.date_range`). The
+        values corresponding to any timesteps in the new index which were not present
+        in the original index will be null (``NaN``), unless a method for filling
+        such unknowns is provided (see the ``method`` parameter below).
+
+        The :meth:`resample` method is more appropriate if an operation on each group of
+        timesteps (such as an aggregate) is necessary to represent the data at the new
+        frequency.
+
+        Parameters
+        ----------
+        freq : DateOffset or str
+            Frequency DateOffset or string.
+        method : {{'backfill'/'bfill', 'pad'/'ffill'}}, default None
+            Method to use for filling holes in reindexed Series (note this
+            does not fill NaNs that already were present):
+
+            * 'pad' / 'ffill': propagate last valid observation forward to next
+              valid based on the order of the index
+            * 'backfill' / 'bfill': use NEXT valid observation to fill.
+        how : {{'start', 'end'}}, default end
+            For PeriodIndex only (see PeriodIndex.asfreq).
+        normalize : bool, default False
+            Whether to reset output index to midnight.
+        fill_value : scalar, optional
+            Value to use for missing values, applied during upsampling (note
+            this does not fill NaNs that already were present).
+
+        Returns
+        -------
+        {klass}
+            {klass} object reindexed to the specified frequency.
+
+        See Also
+        --------
+        reindex : Conform DataFrame to new index with optional filling logic.
+
+        Notes
+        -----
+        To learn more about the frequency strings, please see
+        :ref:`this link<timeseries.offset_aliases>`.
+
+        Examples
+        --------
+        Start by creating a series with 4 one minute timestamps.
+
+        >>> index = pd.date_range("1/1/2000", periods=4, freq="min")
+        >>> series = pd.Series([0.0, None, 2.0, 3.0], index=index)
+        >>> df = pd.DataFrame({{"s": series}})
+        >>> df
+                               s
+        2000-01-01 00:00:00    0.0
+        2000-01-01 00:01:00    NaN
+        2000-01-01 00:02:00    2.0
+        2000-01-01 00:03:00    3.0
+
+        Upsample the series into 30 second bins.
+
+        >>> df.asfreq(freq="30s")
+                               s
+        2000-01-01 00:00:00    0.0
+        2000-01-01 00:00:30    NaN
+        2000-01-01 00:01:00    NaN
+        2000-01-01 00:01:30    NaN
+        2000-01-01 00:02:00    2.0
+        2000-01-01 00:02:30    NaN
+        2000-01-01 00:03:00    3.0
+
+        Upsample again, providing a ``fill value``.
+
+        >>> df.asfreq(freq="30s", fill_value=9.0)
+                               s
+        2000-01-01 00:00:00    0.0
+        2000-01-01 00:00:30    9.0
+        2000-01-01 00:01:00    NaN
+        2000-01-01 00:01:30    9.0
+        2000-01-01 00:02:00    2.0
+        2000-01-01 00:02:30    9.0
+        2000-01-01 00:03:00    3.0
+
+        Upsample again, providing a ``method``.
+
+        >>> df.asfreq(freq="30s", method="bfill")
+                               s
+        2000-01-01 00:00:00    0.0
+        2000-01-01 00:00:30    NaN
+        2000-01-01 00:01:00    NaN
+        2000-01-01 00:01:30    2.0
+        2000-01-01 00:02:00    2.0
+        2000-01-01 00:02:30    3.0
+        2000-01-01 00:03:00    3.0
+        """
+        from pandas.core.resample import asfreq
+
+        return asfreq(
+            self,
+            freq,
+            method=method,
+            how=how,
+            normalize=normalize,
+            fill_value=fill_value,
+        )
+
+    @final
+    def at_time(self, time, asof: bool = False, axis: Axis | None = None) -> Self:
+        """
+        Select values at particular time of day (e.g., 9:30AM).
+
+        Parameters
+        ----------
+        time : datetime.time or str
+            The values to select.
+        asof : bool, default False
+            This parameter is currently not supported.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            For `Series` this parameter is unused and defaults to 0.
+
+        Returns
+        -------
+        Series or DataFrame
+            The values with the specified time.
+
+        Raises
+        ------
+        TypeError
+            If the index is not  a :class:`DatetimeIndex`
+
+        See Also
+        --------
+        between_time : Select values between particular times of the day.
+        first : Select initial periods of time series based on a date offset.
+        last : Select final periods of time series based on a date offset.
+        DatetimeIndex.indexer_at_time : Get just the index locations for
+            values at particular time of the day.
+
+        Examples
+        --------
+        >>> i = pd.date_range("2018-04-09", periods=4, freq="12h")
+        >>> ts = pd.DataFrame({"A": [1, 2, 3, 4]}, index=i)
+        >>> ts
+                             A
+        2018-04-09 00:00:00  1
+        2018-04-09 12:00:00  2
+        2018-04-10 00:00:00  3
+        2018-04-10 12:00:00  4
+
+        >>> ts.at_time("12:00")
+                             A
+        2018-04-09 12:00:00  2
+        2018-04-10 12:00:00  4
+        """
+        if axis is None:
+            axis = 0
+        axis = self._get_axis_number(axis)
+
+        index = self._get_axis(axis)
+
+        if not isinstance(index, DatetimeIndex):
+            raise TypeError("Index must be DatetimeIndex")
+
+        indexer = index.indexer_at_time(time, asof=asof)
+        return self.take(indexer, axis=axis)
+
+    @final
+    def between_time(
+        self,
+        start_time,
+        end_time,
+        inclusive: IntervalClosedType = "both",
+        axis: Axis | None = None,
+    ) -> Self:
+        """
+        Select values between particular times of the day (e.g., 9:00-9:30 AM).
+
+        By setting ``start_time`` to be later than ``end_time``,
+        you can get the times that are *not* between the two times.
+
+        Parameters
+        ----------
+        start_time : datetime.time or str
+            Initial time as a time filter limit.
+        end_time : datetime.time or str
+            End time as a time filter limit.
+        inclusive : {"both", "neither", "left", "right"}, default "both"
+            Include boundaries; whether to set each bound as closed or open.
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Determine range time on index or columns value.
+            For `Series` this parameter is unused and defaults to 0.
+
+        Returns
+        -------
+        Series or DataFrame
+            Data from the original object filtered to the specified dates range.
+
+        Raises
+        ------
+        TypeError
+            If the index is not  a :class:`DatetimeIndex`
+
+        See Also
+        --------
+        at_time : Select values at a particular time of the day.
+        first : Select initial periods of time series based on a date offset.
+        last : Select final periods of time series based on a date offset.
+        DatetimeIndex.indexer_between_time : Get just the index locations for
+            values between particular times of the day.
+
+        Examples
+        --------
+        >>> i = pd.date_range("2018-04-09", periods=4, freq="1D20min")
+        >>> ts = pd.DataFrame({"A": [1, 2, 3, 4]}, index=i)
+        >>> ts
+                             A
+        2018-04-09 00:00:00  1
+        2018-04-10 00:20:00  2
+        2018-04-11 00:40:00  3
+        2018-04-12 01:00:00  4
+
+        >>> ts.between_time("0:15", "0:45")
+                             A
+        2018-04-10 00:20:00  2
+        2018-04-11 00:40:00  3
+
+        You get the times that are *not* between two times by setting
+        ``start_time`` later than ``end_time``:
+
+        >>> ts.between_time("0:45", "0:15")
+                             A
+        2018-04-09 00:00:00  1
+        2018-04-12 01:00:00  4
+        """
+        if axis is None:
+            axis = 0
+        axis = self._get_axis_number(axis)
+
+        index = self._get_axis(axis)
+        if not isinstance(index, DatetimeIndex):
+            raise TypeError("Index must be DatetimeIndex")
+
+        left_inclusive, right_inclusive = validate_inclusive(inclusive)
+        indexer = index.indexer_between_time(
+            start_time,
+            end_time,
+            include_start=left_inclusive,
+            include_end=right_inclusive,
+        )
+        return self.take(indexer, axis=axis)
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def resample(
+        self,
+        rule,
+        closed: Literal["right", "left"] | None = None,
+        label: Literal["right", "left"] | None = None,
+        convention: Literal["start", "end", "s", "e"] | lib.NoDefault = lib.no_default,
+        on: Level | None = None,
+        level: Level | None = None,
+        origin: str | TimestampConvertibleTypes = "start_day",
+        offset: TimedeltaConvertibleTypes | None = None,
+        group_keys: bool = False,
+    ) -> Resampler:
+        """
+        Resample time-series data.
+
+        Convenience method for frequency conversion and resampling of time series.
+        The object must have a datetime-like index (`DatetimeIndex`, `PeriodIndex`,
+        or `TimedeltaIndex`), or the caller must pass the label of a datetime-like
+        series/index to the ``on``/``level`` keyword parameter.
+
+        Parameters
+        ----------
+        rule : DateOffset, Timedelta or str
+            The offset string or object representing target conversion.
+        closed : {{'right', 'left'}}, default None
+            Which side of bin interval is closed. The default is 'left'
+            for all frequency offsets except for 'ME', 'YE', 'QE', 'BME',
+            'BA', 'BQE', and 'W' which all have a default of 'right'.
+        label : {{'right', 'left'}}, default None
+            Which bin edge label to label bucket with. The default is 'left'
+            for all frequency offsets except for 'ME', 'YE', 'QE', 'BME',
+            'BA', 'BQE', and 'W' which all have a default of 'right'.
+        convention : {{'start', 'end', 's', 'e'}}, default 'start'
+            For `PeriodIndex` only, controls whether to use the start or
+            end of `rule`.
+
+            .. deprecated:: 2.2.0
+                Convert PeriodIndex to DatetimeIndex before resampling instead.
+        on : str, optional
+            For a DataFrame, column to use instead of index for resampling.
+            Column must be datetime-like.
+        level : str or int, optional
+            For a MultiIndex, level (name or number) to use for
+            resampling. `level` must be datetime-like.
+        origin : Timestamp or str, default 'start_day'
+            The timestamp on which to adjust the grouping. The timezone of origin
+            must match the timezone of the index.
+            If string, must be Timestamp convertible or one of the following:
+
+            - 'epoch': `origin` is 1970-01-01
+            - 'start': `origin` is the first value of the timeseries
+            - 'start_day': `origin` is the first day at midnight of the timeseries
+
+            - 'end': `origin` is the last value of the timeseries
+            - 'end_day': `origin` is the ceiling midnight of the last day
+
+            .. versionadded:: 1.3.0
+
+            .. note::
+
+                Only takes effect for Tick-frequencies (i.e. fixed frequencies like
+                days, hours, and minutes, rather than months or quarters).
+        offset : Timedelta or str, default is None
+            An offset timedelta added to the origin.
+
+        group_keys : bool, default False
+            Whether to include the group keys in the result index when using
+            ``.apply()`` on the resampled object.
+
+            .. versionadded:: 1.5.0
+
+                Not specifying ``group_keys`` will retain values-dependent behavior
+                from pandas 1.4 and earlier (see :ref:`pandas 1.5.0 Release notes
+                <whatsnew_150.enhancements.resample_group_keys>` for examples).
+
+            .. versionchanged:: 2.0.0
+
+                ``group_keys`` now defaults to ``False``.
+
+        Returns
+        -------
+        pandas.api.typing.Resampler
+            :class:`~pandas.core.Resampler` object.
+
+        See Also
+        --------
+        Series.resample : Resample a Series.
+        DataFrame.resample : Resample a DataFrame.
+        groupby : Group {klass} by mapping, function, label, or list of labels.
+        asfreq : Reindex a {klass} with the given frequency without grouping.
+
+        Notes
+        -----
+        See the `user guide
+        <https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#resampling>`__
+        for more.
+
+        To learn more about the offset strings, please see `this link
+        <https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#dateoffset-objects>`__.
+
+        Examples
+        --------
+        Start by creating a series with 9 one minute timestamps.
+
+        >>> index = pd.date_range("1/1/2000", periods=9, freq="min")
+        >>> series = pd.Series(range(9), index=index)
+        >>> series
+        2000-01-01 00:00:00    0
+        2000-01-01 00:01:00    1
+        2000-01-01 00:02:00    2
+        2000-01-01 00:03:00    3
+        2000-01-01 00:04:00    4
+        2000-01-01 00:05:00    5
+        2000-01-01 00:06:00    6
+        2000-01-01 00:07:00    7
+        2000-01-01 00:08:00    8
+        Freq: min, dtype: int64
+
+        Downsample the series into 3 minute bins and sum the values
+        of the timestamps falling into a bin.
+
+        >>> series.resample("3min").sum()
+        2000-01-01 00:00:00     3
+        2000-01-01 00:03:00    12
+        2000-01-01 00:06:00    21
+        Freq: 3min, dtype: int64
+
+        Downsample the series into 3 minute bins as above, but label each
+        bin using the right edge instead of the left. Please note that the
+        value in the bucket used as the label is not included in the bucket,
+        which it labels. For example, in the original series the
+        bucket ``2000-01-01 00:03:00`` contains the value 3, but the summed
+        value in the resampled bucket with the label ``2000-01-01 00:03:00``
+        does not include 3 (if it did, the summed value would be 6, not 3).
+
+        >>> series.resample("3min", label="right").sum()
+        2000-01-01 00:03:00     3
+        2000-01-01 00:06:00    12
+        2000-01-01 00:09:00    21
+        Freq: 3min, dtype: int64
+
+        To include this value close the right side of the bin interval,
+        as shown below.
+
+        >>> series.resample("3min", label="right", closed="right").sum()
+        2000-01-01 00:00:00     0
+        2000-01-01 00:03:00     6
+        2000-01-01 00:06:00    15
+        2000-01-01 00:09:00    15
+        Freq: 3min, dtype: int64
+
+        Upsample the series into 30 second bins.
+
+        >>> series.resample("30s").asfreq()[0:5]  # Select first 5 rows
+        2000-01-01 00:00:00   0.0
+        2000-01-01 00:00:30   NaN
+        2000-01-01 00:01:00   1.0
+        2000-01-01 00:01:30   NaN
+        2000-01-01 00:02:00   2.0
+        Freq: 30s, dtype: float64
+
+        Upsample the series into 30 second bins and fill the ``NaN``
+        values using the ``ffill`` method.
+
+        >>> series.resample("30s").ffill()[0:5]
+        2000-01-01 00:00:00    0
+        2000-01-01 00:00:30    0
+        2000-01-01 00:01:00    1
+        2000-01-01 00:01:30    1
+        2000-01-01 00:02:00    2
+        Freq: 30s, dtype: int64
+
+        Upsample the series into 30 second bins and fill the
+        ``NaN`` values using the ``bfill`` method.
+
+        >>> series.resample("30s").bfill()[0:5]
+        2000-01-01 00:00:00    0
+        2000-01-01 00:00:30    1
+        2000-01-01 00:01:00    1
+        2000-01-01 00:01:30    2
+        2000-01-01 00:02:00    2
+        Freq: 30s, dtype: int64
+
+        Pass a custom function via ``apply``
+
+        >>> def custom_resampler(arraylike):
+        ...     return np.sum(arraylike) + 5
+        >>> series.resample("3min").apply(custom_resampler)
+        2000-01-01 00:00:00     8
+        2000-01-01 00:03:00    17
+        2000-01-01 00:06:00    26
+        Freq: 3min, dtype: int64
+
+        For DataFrame objects, the keyword `on` can be used to specify the
+        column instead of the index for resampling.
+
+        >>> df = pd.DataFrame([10, 11, 9, 13, 14, 18, 17, 19], columns=["price"])
+        >>> df["volume"] = [50, 60, 40, 100, 50, 100, 40, 50]
+        >>> df["week_starting"] = pd.date_range("01/01/2018", periods=8, freq="W")
+        >>> df
+           price  volume week_starting
+        0     10      50    2018-01-07
+        1     11      60    2018-01-14
+        2      9      40    2018-01-21
+        3     13     100    2018-01-28
+        4     14      50    2018-02-04
+        5     18     100    2018-02-11
+        6     17      40    2018-02-18
+        7     19      50    2018-02-25
+        >>> df.resample("ME", on="week_starting").mean()
+                       price  volume
+        week_starting
+        2018-01-31     10.75    62.5
+        2018-02-28     17.00    60.0
+
+        For a DataFrame with MultiIndex, the keyword `level` can be used to
+        specify on which level the resampling needs to take place.
+
+        >>> days = pd.date_range("1/1/2000", periods=4, freq="D")
+        >>> df2 = pd.DataFrame(
+        ...     [
+        ...         [10, 50],
+        ...         [11, 60],
+        ...         [9, 40],
+        ...         [13, 100],
+        ...         [14, 50],
+        ...         [18, 100],
+        ...         [17, 40],
+        ...         [19, 50],
+        ...     ],
+        ...     columns=["price", "volume"],
+        ...     index=pd.MultiIndex.from_product([days, ["morning", "afternoon"]]),
+        ... )
+        >>> df2
+                              price  volume
+        2000-01-01 morning       10      50
+                   afternoon     11      60
+        2000-01-02 morning        9      40
+                   afternoon     13     100
+        2000-01-03 morning       14      50
+                   afternoon     18     100
+        2000-01-04 morning       17      40
+                   afternoon     19      50
+        >>> df2.resample("D", level=0).sum()
+                    price  volume
+        2000-01-01     21     110
+        2000-01-02     22     140
+        2000-01-03     32     150
+        2000-01-04     36      90
+
+        If you want to adjust the start of the bins based on a fixed timestamp:
+
+        >>> start, end = "2000-10-01 23:30:00", "2000-10-02 00:30:00"
+        >>> rng = pd.date_range(start, end, freq="7min")
+        >>> ts = pd.Series(np.arange(len(rng)) * 3, index=rng)
+        >>> ts
+        2000-10-01 23:30:00     0
+        2000-10-01 23:37:00     3
+        2000-10-01 23:44:00     6
+        2000-10-01 23:51:00     9
+        2000-10-01 23:58:00    12
+        2000-10-02 00:05:00    15
+        2000-10-02 00:12:00    18
+        2000-10-02 00:19:00    21
+        2000-10-02 00:26:00    24
+        Freq: 7min, dtype: int64
+
+        >>> ts.resample("17min").sum()
+        2000-10-01 23:14:00     0
+        2000-10-01 23:31:00     9
+        2000-10-01 23:48:00    21
+        2000-10-02 00:05:00    54
+        2000-10-02 00:22:00    24
+        Freq: 17min, dtype: int64
+
+        >>> ts.resample("17min", origin="epoch").sum()
+        2000-10-01 23:18:00     0
+        2000-10-01 23:35:00    18
+        2000-10-01 23:52:00    27
+        2000-10-02 00:09:00    39
+        2000-10-02 00:26:00    24
+        Freq: 17min, dtype: int64
+
+        >>> ts.resample("17min", origin="2000-01-01").sum()
+        2000-10-01 23:24:00     3
+        2000-10-01 23:41:00    15
+        2000-10-01 23:58:00    45
+        2000-10-02 00:15:00    45
+        Freq: 17min, dtype: int64
+
+        If you want to adjust the start of the bins with an `offset` Timedelta, the two
+        following lines are equivalent:
+
+        >>> ts.resample("17min", origin="start").sum()
+        2000-10-01 23:30:00     9
+        2000-10-01 23:47:00    21
+        2000-10-02 00:04:00    54
+        2000-10-02 00:21:00    24
+        Freq: 17min, dtype: int64
+
+        >>> ts.resample("17min", offset="23h30min").sum()
+        2000-10-01 23:30:00     9
+        2000-10-01 23:47:00    21
+        2000-10-02 00:04:00    54
+        2000-10-02 00:21:00    24
+        Freq: 17min, dtype: int64
+
+        If you want to take the largest Timestamp as the end of the bins:
+
+        >>> ts.resample("17min", origin="end").sum()
+        2000-10-01 23:35:00     0
+        2000-10-01 23:52:00    18
+        2000-10-02 00:09:00    27
+        2000-10-02 00:26:00    63
+        Freq: 17min, dtype: int64
+
+        In contrast with the `start_day`, you can use `end_day` to take the ceiling
+        midnight of the largest Timestamp as the end of the bins and drop the bins
+        not containing data:
+
+        >>> ts.resample("17min", origin="end_day").sum()
+        2000-10-01 23:38:00     3
+        2000-10-01 23:55:00    15
+        2000-10-02 00:12:00    45
+        2000-10-02 00:29:00    45
+        Freq: 17min, dtype: int64
+        """
+        from pandas.core.resample import get_resampler
+
+        if convention is not lib.no_default:
+            warnings.warn(
+                f"The 'convention' keyword in {type(self).__name__}.resample is "
+                "deprecated and will be removed in a future version. "
+                "Explicitly cast PeriodIndex to DatetimeIndex before resampling "
+                "instead.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        else:
+            convention = "start"
+
+        return get_resampler(
+            cast("Series | DataFrame", self),
+            freq=rule,
+            label=label,
+            closed=closed,
+            convention=convention,
+            key=on,
+            level=level,
+            origin=origin,
+            offset=offset,
+            group_keys=group_keys,
+        )
+
+    @final
+    def rank(
+        self,
+        axis: Axis = 0,
+        method: Literal["average", "min", "max", "first", "dense"] = "average",
+        numeric_only: bool = False,
+        na_option: Literal["keep", "top", "bottom"] = "keep",
+        ascending: bool = True,
+        pct: bool = False,
+    ) -> Self:
+        """
+        Compute numerical data ranks (1 through n) along axis.
+
+        By default, equal values are assigned a rank that is the average of the
+        ranks of those values.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Index to direct ranking.
+            For `Series` this parameter is unused and defaults to 0.
+        method : {'average', 'min', 'max', 'first', 'dense'}, default 'average'
+            How to rank the group of records that have the same value (i.e. ties):
+
+            * average: average rank of the group
+            * min: lowest rank in the group
+            * max: highest rank in the group
+            * first: ranks assigned in order they appear in the array
+            * dense: like 'min', but rank always increases by 1 between groups.
+
+        numeric_only : bool, default False
+            For DataFrame objects, rank only numeric columns if set to True.
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        na_option : {'keep', 'top', 'bottom'}, default 'keep'
+            How to rank NaN values:
+
+            * keep: assign NaN rank to NaN values
+            * top: assign lowest rank to NaN values
+            * bottom: assign highest rank to NaN values
+
+        ascending : bool, default True
+            Whether or not the elements should be ranked in ascending order.
+        pct : bool, default False
+            Whether or not to display the returned rankings in percentile
+            form.
+
+        Returns
+        -------
+        same type as caller
+            Return a Series or DataFrame with data ranks as values.
+
+        See Also
+        --------
+        core.groupby.DataFrameGroupBy.rank : Rank of values within each group.
+        core.groupby.SeriesGroupBy.rank : Rank of values within each group.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     data={
+        ...         "Animal": ["cat", "penguin", "dog", "spider", "snake"],
+        ...         "Number_legs": [4, 2, 4, 8, np.nan],
+        ...     }
+        ... )
+        >>> df
+            Animal  Number_legs
+        0      cat          4.0
+        1  penguin          2.0
+        2      dog          4.0
+        3   spider          8.0
+        4    snake          NaN
+
+        Ties are assigned the mean of the ranks (by default) for the group.
+
+        >>> s = pd.Series(range(5), index=list("abcde"))
+        >>> s["d"] = s["b"]
+        >>> s.rank()
+        a    1.0
+        b    2.5
+        c    4.0
+        d    2.5
+        e    5.0
+        dtype: float64
+
+        The following example shows how the method behaves with the above
+        parameters:
+
+        * default_rank: this is the default behaviour obtained without using
+          any parameter.
+        * max_rank: setting ``method = 'max'`` the records that have the
+          same values are ranked using the highest rank (e.g.: since 'cat'
+          and 'dog' are both in the 2nd and 3rd position, rank 3 is assigned.)
+        * NA_bottom: choosing ``na_option = 'bottom'``, if there are records
+          with NaN values they are placed at the bottom of the ranking.
+        * pct_rank: when setting ``pct = True``, the ranking is expressed as
+          percentile rank.
+
+        >>> df["default_rank"] = df["Number_legs"].rank()
+        >>> df["max_rank"] = df["Number_legs"].rank(method="max")
+        >>> df["NA_bottom"] = df["Number_legs"].rank(na_option="bottom")
+        >>> df["pct_rank"] = df["Number_legs"].rank(pct=True)
+        >>> df
+            Animal  Number_legs  default_rank  max_rank  NA_bottom  pct_rank
+        0      cat          4.0           2.5       3.0        2.5     0.625
+        1  penguin          2.0           1.0       1.0        1.0     0.250
+        2      dog          4.0           2.5       3.0        2.5     0.625
+        3   spider          8.0           4.0       4.0        4.0     1.000
+        4    snake          NaN           NaN       NaN        5.0       NaN
+        """
+        axis_int = self._get_axis_number(axis)
+
+        if na_option not in {"keep", "top", "bottom"}:
+            msg = "na_option must be one of 'keep', 'top', or 'bottom'"
+            raise ValueError(msg)
+
+        def ranker(data):
+            if data.ndim == 2:
+                # i.e. DataFrame, we cast to ndarray
+                values = data.values
+            else:
+                # i.e. Series, can dispatch to EA
+                values = data._values
+
+            if isinstance(values, ExtensionArray):
+                ranks = values._rank(
+                    axis=axis_int,
+                    method=method,
+                    ascending=ascending,
+                    na_option=na_option,
+                    pct=pct,
+                )
+            else:
+                ranks = algos.rank(
+                    values,
+                    axis=axis_int,
+                    method=method,
+                    ascending=ascending,
+                    na_option=na_option,
+                    pct=pct,
+                )
+
+            ranks_obj = self._constructor(ranks, **data._construct_axes_dict())
+            return ranks_obj.__finalize__(self, method="rank")
+
+        if numeric_only:
+            if self.ndim == 1 and not is_numeric_dtype(self.dtype):
+                # GH#47500
+                raise TypeError(
+                    "Series.rank does not allow numeric_only=True with "
+                    "non-numeric dtype."
+                )
+            data = self._get_numeric_data()
+        else:
+            data = self
+
+        return ranker(data)
+
+    @doc(_shared_docs["compare"], klass=_shared_doc_kwargs["klass"])
+    def compare(
+        self,
+        other: Self,
+        align_axis: Axis = 1,
+        keep_shape: bool = False,
+        keep_equal: bool = False,
+        result_names: Suffixes = ("self", "other"),
+    ):
+        if type(self) is not type(other):
+            cls_self, cls_other = type(self).__name__, type(other).__name__
+            raise TypeError(
+                f"can only compare '{cls_self}' (not '{cls_other}') with '{cls_self}'"
+            )
+
+        # error: Unsupported left operand type for & ("Self")
+        mask = ~((self == other) | (self.isna() & other.isna()))  # type: ignore[operator]
+        mask.fillna(True, inplace=True)
+
+        if not keep_equal:
+            self = self.where(mask)
+            other = other.where(mask)
+
+        if not keep_shape:
+            if isinstance(self, ABCDataFrame):
+                cmask = mask.any()
+                rmask = mask.any(axis=1)
+                self = self.loc[rmask, cmask]
+                other = other.loc[rmask, cmask]
+            else:
+                self = self[mask]
+                other = other[mask]
+        if not isinstance(result_names, tuple):
+            raise TypeError(
+                f"Passing 'result_names' as a {type(result_names)} is not "
+                "supported. Provide 'result_names' as a tuple instead."
+            )
+
+        if align_axis in (1, "columns"):  # This is needed for Series
+            axis = 1
+        else:
+            axis = self._get_axis_number(align_axis)
+
+        # error: List item 0 has incompatible type "NDFrame"; expected
+        #  "Union[Series, DataFrame]"
+        diff = concat(
+            [self, other],  # type: ignore[list-item]
+            axis=axis,
+            keys=result_names,
+        )
+
+        if axis >= self.ndim:
+            # No need to reorganize data if stacking on new axis
+            # This currently applies for stacking two Series on columns
+            return diff
+
+        ax = diff._get_axis(axis)
+        ax_names = np.array(ax.names)
+
+        # set index names to positions to avoid confusion
+        ax.names = np.arange(len(ax_names))
+
+        # bring self-other to inner level
+        order = list(range(1, ax.nlevels)) + [0]
+        if isinstance(diff, ABCDataFrame):
+            diff = diff.reorder_levels(order, axis=axis)
+        else:
+            diff = diff.reorder_levels(order)
+
+        # restore the index names in order
+        diff._get_axis(axis=axis).names = ax_names[order]
+
+        # reorder axis to keep things organized
+        indices = (
+            np.arange(diff.shape[axis])
+            .reshape([2, diff.shape[axis] // 2])
+            .T.reshape(-1)
+        )
+        diff = diff.take(indices, axis=axis)
+
+        return diff
+
+    @final
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+    )
+    def align(
+        self,
+        other: NDFrameT,
+        join: AlignJoin = "outer",
+        axis: Axis | None = None,
+        level: Level | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        fill_value: Hashable | None = None,
+    ) -> tuple[Self, NDFrameT]:
+        """
+        Align two objects on their axes with the specified join method.
+
+        Join method is specified for each axis Index.
+
+        Parameters
+        ----------
+        other : DataFrame or Series
+            The object to align with.
+        join : {{'outer', 'inner', 'left', 'right'}}, default 'outer'
+            Type of alignment to be performed.
+
+            * left: use only keys from left frame, preserve key order.
+            * right: use only keys from right frame, preserve key order.
+            * outer: use union of keys from both frames, sort keys lexicographically.
+            * inner: use intersection of keys from both frames,
+              preserve the order of the left keys.
+
+        axis : allowed axis of the other object, default None
+            Align on index (0), columns (1), or both (None).
+        level : int or level name, default None
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        copy : bool, default False
+            Always returns new objects. If copy=False and no reindexing is
+            required then original objects are returned.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        fill_value : scalar, default np.nan
+            Value to use for missing values. Defaults to NaN, but can be any
+            "compatible" value.
+
+        Returns
+        -------
+        tuple of ({klass}, type of other)
+            Aligned objects.
+
+        See Also
+        --------
+        Series.align : Align two objects on their axes with specified join method.
+        DataFrame.align : Align two objects on their axes with specified join method.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [[1, 2, 3, 4], [6, 7, 8, 9]], columns=["D", "B", "E", "A"], index=[1, 2]
+        ... )
+        >>> other = pd.DataFrame(
+        ...     [[10, 20, 30, 40], [60, 70, 80, 90], [600, 700, 800, 900]],
+        ...     columns=["A", "B", "C", "D"],
+        ...     index=[2, 3, 4],
+        ... )
+        >>> df
+           D  B  E  A
+        1  1  2  3  4
+        2  6  7  8  9
+        >>> other
+            A    B    C    D
+        2   10   20   30   40
+        3   60   70   80   90
+        4  600  700  800  900
+
+        Align on columns:
+
+        >>> left, right = df.align(other, join="outer", axis=1)
+        >>> left
+           A  B   C  D  E
+        1  4  2 NaN  1  3
+        2  9  7 NaN  6  8
+        >>> right
+            A    B    C    D   E
+        2   10   20   30   40 NaN
+        3   60   70   80   90 NaN
+        4  600  700  800  900 NaN
+
+        We can also align on the index:
+
+        >>> left, right = df.align(other, join="outer", axis=0)
+        >>> left
+            D    B    E    A
+        1  1.0  2.0  3.0  4.0
+        2  6.0  7.0  8.0  9.0
+        3  NaN  NaN  NaN  NaN
+        4  NaN  NaN  NaN  NaN
+        >>> right
+            A      B      C      D
+        1    NaN    NaN    NaN    NaN
+        2   10.0   20.0   30.0   40.0
+        3   60.0   70.0   80.0   90.0
+        4  600.0  700.0  800.0  900.0
+
+        Finally, the default `axis=None` will align on both index and columns:
+
+        >>> left, right = df.align(other, join="outer", axis=None)
+        >>> left
+             A    B   C    D    E
+        1  4.0  2.0 NaN  1.0  3.0
+        2  9.0  7.0 NaN  6.0  8.0
+        3  NaN  NaN NaN  NaN  NaN
+        4  NaN  NaN NaN  NaN  NaN
+        >>> right
+               A      B      C      D   E
+        1    NaN    NaN    NaN    NaN NaN
+        2   10.0   20.0   30.0   40.0 NaN
+        3   60.0   70.0   80.0   90.0 NaN
+        4  600.0  700.0  800.0  900.0 NaN
+        """
+        self._check_copy_deprecation(copy)
+
+        _right: DataFrame | Series
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+        if isinstance(other, ABCDataFrame):
+            left, _right, join_index = self._align_frame(
+                other,
+                join=join,
+                axis=axis,
+                level=level,
+                fill_value=fill_value,
+            )
+
+        elif isinstance(other, ABCSeries):
+            left, _right, join_index = self._align_series(
+                other,
+                join=join,
+                axis=axis,
+                level=level,
+                fill_value=fill_value,
+            )
+        else:  # pragma: no cover
+            raise TypeError(f"unsupported type: {type(other)}")
+
+        right = cast(NDFrameT, _right)
+        if self.ndim == 1 or axis == 0:
+            # If we are aligning timezone-aware DatetimeIndexes and the timezones
+            #  do not match, convert both to UTC.
+            if isinstance(left.index.dtype, DatetimeTZDtype):
+                if left.index.tz != right.index.tz:
+                    if join_index is not None:
+                        # GH#33671 copy to ensure we don't change the index on
+                        #  our original Series
+                        left = left.copy(deep=False)
+                        right = right.copy(deep=False)
+                        left.index = join_index
+                        right.index = join_index
+
+        left = left.__finalize__(self)
+        right = right.__finalize__(other)
+        return left, right
+
+    @final
+    def _align_frame(
+        self,
+        other: DataFrame,
+        join: AlignJoin = "outer",
+        axis: Axis | None = None,
+        level=None,
+        fill_value=None,
+    ) -> tuple[Self, DataFrame, Index | None]:
+        # defaults
+        join_index, join_columns = None, None
+        ilidx, iridx = None, None
+        clidx, cridx = None, None
+
+        is_series = isinstance(self, ABCSeries)
+
+        if (axis is None or axis == 0) and not self.index.equals(other.index):
+            join_index, ilidx, iridx = self.index.join(
+                other.index, how=join, level=level, return_indexers=True
+            )
+
+        if (
+            (axis is None or axis == 1)
+            and not is_series
+            and not self.columns.equals(other.columns)
+        ):
+            join_columns, clidx, cridx = self.columns.join(
+                other.columns, how=join, level=level, return_indexers=True
+            )
+
+        if is_series:
+            reindexers = {0: [join_index, ilidx]}
+        else:
+            reindexers = {0: [join_index, ilidx], 1: [join_columns, clidx]}
+
+        left = self._reindex_with_indexers(
+            reindexers, fill_value=fill_value, allow_dups=True
+        )
+        # other must be always DataFrame
+        right = other._reindex_with_indexers(
+            {0: [join_index, iridx], 1: [join_columns, cridx]},
+            fill_value=fill_value,
+            allow_dups=True,
+        )
+        return left, right, join_index
+
+    @final
+    def _align_series(
+        self,
+        other: Series,
+        join: AlignJoin = "outer",
+        axis: Axis | None = None,
+        level=None,
+        fill_value=None,
+    ) -> tuple[Self, Series, Index | None]:
+        is_series = isinstance(self, ABCSeries)
+
+        if (not is_series and axis is None) or axis not in [None, 0, 1]:
+            raise ValueError("Must specify axis=0 or 1")
+
+        if is_series and axis == 1:
+            raise ValueError("cannot align series to a series other than axis 0")
+
+        # series/series compat, other must always be a Series
+        if not axis:
+            # equal
+            if self.index.equals(other.index):
+                join_index, lidx, ridx = None, None, None
+            else:
+                join_index, lidx, ridx = self.index.join(
+                    other.index, how=join, level=level, return_indexers=True
+                )
+
+            if is_series:
+                left = self._reindex_indexer(join_index, lidx)
+            elif lidx is None or join_index is None:
+                left = self.copy(deep=False)
+            else:
+                new_mgr = self._mgr.reindex_indexer(join_index, lidx, axis=1)
+                left = self._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+
+            right = other._reindex_indexer(join_index, ridx)
+
+        else:
+            # one has > 1 ndim
+            fdata = self._mgr
+            join_index = self.axes[1]
+            lidx, ridx = None, None
+            if not join_index.equals(other.index):
+                join_index, lidx, ridx = join_index.join(
+                    other.index, how=join, level=level, return_indexers=True
+                )
+
+            if lidx is not None:
+                bm_axis = self._get_block_manager_axis(1)
+                fdata = fdata.reindex_indexer(join_index, lidx, axis=bm_axis)
+
+            left = self._constructor_from_mgr(fdata, axes=fdata.axes)
+
+            if ridx is None:
+                right = other.copy(deep=False)
+            else:
+                right = other.reindex(join_index, level=level)
+
+        # fill
+        fill_na = notna(fill_value)
+        if fill_na:
+            left = left.fillna(fill_value)
+            right = right.fillna(fill_value)
+
+        return left, right, join_index
+
+    @overload
+    def _where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[False] = ...,
+        axis: Axis | None = ...,
+        level=...,
+    ) -> Self: ...
+
+    @overload
+    def _where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[True],
+        axis: Axis | None = ...,
+        level=...,
+    ) -> None: ...
+
+    @overload
+    def _where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: bool,
+        axis: Axis | None = ...,
+        level=...,
+    ) -> Self | None: ...
+
+    @final
+    def _where(
+        self,
+        cond,
+        other=lib.no_default,
+        *,
+        inplace: bool = False,
+        axis: Axis | None = None,
+        level=None,
+    ) -> Self | None:
+        """
+        Equivalent to public method `where`, except that `other` is not
+        applied as a function even if callable. Used in __setitem__.
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if axis is not None:
+            axis = self._get_axis_number(axis)
+
+        # align the cond to same shape as myself
+        cond = common.apply_if_callable(cond, self)
+        if isinstance(cond, NDFrame):
+            # CoW: Make sure reference is not kept alive
+            if cond.ndim == 1 and self.ndim == 2:
+                cond = cond._constructor_expanddim(
+                    {i: cond for i in range(len(self.columns))},
+                    copy=False,
+                )
+                cond.columns = self.columns
+            cond = cond.align(self, join="right")[0]
+        else:
+            if not hasattr(cond, "shape"):
+                cond = np.asanyarray(cond)
+            if cond.shape != self.shape:
+                raise ValueError("Array conditional must be same shape as self")
+            cond = self._constructor(cond, **self._construct_axes_dict(), copy=False)
+
+        # make sure we are boolean
+        fill_value = bool(inplace)
+        cond = cond.fillna(fill_value)
+        cond = cond.infer_objects()
+
+        msg = "Boolean array expected for the condition, not {dtype}"
+
+        if not cond.empty:
+            if not isinstance(cond, ABCDataFrame):
+                # This is a single-dimensional object.
+                if not is_bool_dtype(cond):
+                    raise TypeError(msg.format(dtype=cond.dtype))
+            else:
+                for _dt in cond.dtypes:
+                    if not is_bool_dtype(_dt):
+                        raise TypeError(msg.format(dtype=_dt))
+                if cond._mgr.any_extension_types:
+                    # GH51574: avoid object ndarray conversion later on
+                    cond = cond._constructor(
+                        cond.to_numpy(dtype=bool, na_value=fill_value),
+                        **cond._construct_axes_dict(),
+                    )
+        else:
+            # GH#21947 we have an empty DataFrame/Series, could be object-dtype
+            cond = cond.astype(bool)
+
+        cond = -cond if inplace else cond
+        cond = cond.reindex(self._info_axis, axis=self._info_axis_number)
+
+        # try to align with other
+        if isinstance(other, NDFrame):
+            # align with me
+            if other.ndim <= self.ndim:
+                # CoW: Make sure reference is not kept alive
+                other = self.align(
+                    other,
+                    join="left",
+                    axis=axis,
+                    level=level,
+                    fill_value=None,
+                )[1]
+
+                # if we are NOT aligned, raise as we cannot where index
+                if axis is None and not other._indexed_same(self):
+                    raise InvalidIndexError
+
+                if other.ndim < self.ndim:
+                    # TODO(EA2D): avoid object-dtype cast in EA case GH#38729
+                    other = other._values
+                    if axis == 0:
+                        other = np.reshape(other, (-1, 1))
+                    elif axis == 1:
+                        other = np.reshape(other, (1, -1))
+
+                    other = np.broadcast_to(other, self.shape)
+
+            # slice me out of the other
+            else:
+                raise NotImplementedError(
+                    "cannot align with a higher dimensional NDFrame"
+                )
+
+        elif not isinstance(other, (MultiIndex, NDFrame)):
+            # mainly just catching Index here
+            other = extract_array(other, extract_numpy=True)
+
+        if isinstance(other, (np.ndarray, ExtensionArray)):
+            if other.shape != self.shape:
+                if self.ndim != 1:
+                    # In the ndim == 1 case we may have
+                    #  other length 1, which we treat as scalar (GH#2745, GH#4192)
+                    #  or len(other) == icond.sum(), which we treat like
+                    #  __setitem__ (GH#3235)
+                    raise ValueError(
+                        "other must be the same shape as self when an ndarray"
+                    )
+
+            # we are the same shape, so create an actual object for alignment
+            else:
+                other = self._constructor(
+                    other, **self._construct_axes_dict(), copy=False
+                )
+
+        if axis is None:
+            axis = 0
+
+        if self.ndim == getattr(other, "ndim", 0):
+            align = True
+        else:
+            align = self._get_axis_number(axis) == 1
+
+        if inplace:
+            # we may have different type blocks come out of putmask, so
+            # reconstruct the block manager
+
+            new_data = self._mgr.putmask(mask=cond, new=other, align=align)
+            result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+            return self._update_inplace(result)
+
+        else:
+            new_data = self._mgr.where(
+                other=other,
+                cond=cond,
+                align=align,
+            )
+            result = self._constructor_from_mgr(new_data, axes=new_data.axes)
+            return result.__finalize__(self)
+
+    @overload
+    def where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[False] = ...,
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> Self: ...
+
+    @overload
+    def where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[True],
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> None: ...
+
+    @overload
+    def where(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: bool = ...,
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        cond="True",
+        cond_rev="False",
+        name="where",
+        name_other="mask",
+    )
+    def where(
+        self,
+        cond,
+        other=np.nan,
+        *,
+        inplace: bool = False,
+        axis: Axis | None = None,
+        level: Level | None = None,
+    ) -> Self | None:
+        """
+        Replace values where the condition is {cond_rev}.
+
+        Parameters
+        ----------
+        cond : bool {klass}, array-like, or callable
+            Where `cond` is {cond}, keep the original value. Where
+            {cond_rev}, replace with corresponding value from `other`.
+            If `cond` is callable, it is computed on the {klass} and
+            should return boolean {klass} or array. The callable must
+            not change input {klass} (though pandas doesn't check it).
+        other : scalar, {klass}, or callable
+            Entries where `cond` is {cond_rev} are replaced with
+            corresponding value from `other`.
+            If other is callable, it is computed on the {klass} and
+            should return scalar or {klass}. The callable must not
+            change input {klass} (though pandas doesn't check it).
+            If not specified, entries will be filled with the corresponding
+            NULL value (``np.nan`` for numpy dtypes, ``pd.NA`` for extension
+            dtypes).
+        inplace : bool, default False
+            Whether to perform the operation in place on the data.
+        axis : int, default None
+            Alignment axis if needed. For `Series` this parameter is
+            unused and defaults to 0.
+        level : int, default None
+            Alignment level if needed.
+
+        Returns
+        -------
+        Series or DataFrame or None
+            When applied to a Series, the function will return a Series,
+            and when applied to a DataFrame, it will return a DataFrame;
+            if ``inplace=True``, it will return None.
+
+        See Also
+        --------
+        :func:`DataFrame.{name_other}` : Return an object of same shape as
+            caller.
+        :func:`Series.{name_other}` : Return an object of same shape as
+            caller.
+
+        Notes
+        -----
+        The {name} method is an application of the if-then idiom. For each
+        element in the caller, if ``cond`` is ``{cond}`` the
+        element is used; otherwise the corresponding element from
+        ``other`` is used. If the axis of ``other`` does not align with axis of
+        ``cond`` {klass}, the values of ``cond`` on misaligned index positions
+        will be filled with {cond_rev}.
+
+        The signature for :func:`Series.where` or
+        :func:`DataFrame.where` differs from :func:`numpy.where`.
+        Roughly ``df1.where(m, df2)`` is equivalent to ``np.where(m, df1, df2)``.
+
+        For further details and examples see the ``{name}`` documentation in
+        :ref:`indexing <indexing.where_mask>`.
+
+        The dtype of the object takes precedence. The fill value is casted to
+        the object's dtype, if this can be done losslessly.
+
+        Examples
+        --------
+        >>> s = pd.Series(range(5))
+        >>> s.where(s > 0)
+        0    NaN
+        1    1.0
+        2    2.0
+        3    3.0
+        4    4.0
+        dtype: float64
+        >>> s.mask(s > 0)
+        0    0.0
+        1    NaN
+        2    NaN
+        3    NaN
+        4    NaN
+        dtype: float64
+
+        >>> s = pd.Series(range(5))
+        >>> t = pd.Series([True, False])
+        >>> s.where(t, 99)
+        0     0
+        1    99
+        2    99
+        3    99
+        4    99
+        dtype: int64
+        >>> s.mask(t, 99)
+        0    99
+        1     1
+        2    99
+        3    99
+        4    99
+        dtype: int64
+
+        >>> s.where(s > 1, 10)
+        0    10
+        1    10
+        2    2
+        3    3
+        4    4
+        dtype: int64
+        >>> s.mask(s > 1, 10)
+        0     0
+        1     1
+        2    10
+        3    10
+        4    10
+        dtype: int64
+
+        >>> df = pd.DataFrame(np.arange(10).reshape(-1, 2), columns=["A", "B"])
+        >>> df
+           A  B
+        0  0  1
+        1  2  3
+        2  4  5
+        3  6  7
+        4  8  9
+        >>> m = df % 3 == 0
+        >>> df.where(m, -df)
+           A  B
+        0  0 -1
+        1 -2  3
+        2 -4 -5
+        3  6 -7
+        4 -8  9
+        >>> df.where(m, -df) == np.where(m, df, -df)
+              A     B
+        0  True  True
+        1  True  True
+        2  True  True
+        3  True  True
+        4  True  True
+        >>> df.where(m, -df) == df.mask(~m, -df)
+              A     B
+        0  True  True
+        1  True  True
+        2  True  True
+        3  True  True
+        4  True  True
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        other = common.apply_if_callable(other, self)
+        return self._where(cond, other, inplace=inplace, axis=axis, level=level)
+
+    @overload
+    def mask(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[False] = ...,
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> Self: ...
+
+    @overload
+    def mask(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: Literal[True],
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> None: ...
+
+    @overload
+    def mask(
+        self,
+        cond,
+        other=...,
+        *,
+        inplace: bool = ...,
+        axis: Axis | None = ...,
+        level: Level = ...,
+    ) -> Self | None: ...
+
+    @final
+    @doc(
+        where,
+        klass=_shared_doc_kwargs["klass"],
+        cond="False",
+        cond_rev="True",
+        name="mask",
+        name_other="where",
+    )
+    def mask(
+        self,
+        cond,
+        other=lib.no_default,
+        *,
+        inplace: bool = False,
+        axis: Axis | None = None,
+        level: Level | None = None,
+    ) -> Self | None:
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if inplace:
+            if not PYPY:
+                if sys.getrefcount(self) <= REF_COUNT:
+                    warnings.warn(
+                        _chained_assignment_method_msg,
+                        ChainedAssignmentError,
+                        stacklevel=2,
+                    )
+
+        cond = common.apply_if_callable(cond, self)
+        other = common.apply_if_callable(other, self)
+
+        # see gh-21891
+        if not hasattr(cond, "__invert__"):
+            cond = np.array(cond)
+
+        return self._where(
+            ~cond,
+            other=other,
+            inplace=inplace,
+            axis=axis,
+            level=level,
+        )
+
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def shift(
+        self,
+        periods: int | Sequence[int] = 1,
+        freq=None,
+        axis: Axis = 0,
+        fill_value: Hashable = lib.no_default,
+        suffix: str | None = None,
+    ) -> Self | DataFrame:
+        """
+        Shift index by desired number of periods with an optional time `freq`.
+
+        When `freq` is not passed, shift the index without realigning the data.
+        If `freq` is passed (in this case, the index must be date or datetime,
+        or it will raise a `NotImplementedError`), the index will be
+        increased using the periods and the `freq`. `freq` can be inferred
+        when specified as "infer" as long as either freq or inferred_freq
+        attribute is set in the index.
+
+        Parameters
+        ----------
+        periods : int or Sequence
+            Number of periods to shift. Can be positive or negative.
+            If an iterable of ints, the data will be shifted once by each int.
+            This is equivalent to shifting by one value at a time and
+            concatenating all resulting frames. The resulting columns will have
+            the shift suffixed to their column names. For multiple periods,
+            axis must not be 1.
+        freq : DateOffset, tseries.offsets, timedelta, or str, optional
+            Offset to use from the tseries module or time rule (e.g. 'EOM').
+            If `freq` is specified then the index values are shifted but the
+            data is not realigned. That is, use `freq` if you would like to
+            extend the index when shifting and preserve the original data.
+            If `freq` is specified as "infer" then it will be inferred from
+            the freq or inferred_freq attributes of the index. If neither of
+            those attributes exist, a ValueError is thrown.
+        axis : {{0 or 'index', 1 or 'columns', None}}, default None
+            Shift direction. For `Series` this parameter is unused and defaults to 0.
+        fill_value : object, optional
+            The scalar value to use for newly introduced missing values.
+            the default depends on the dtype of `self`.
+            For Boolean and numeric NumPy data types, ``np.nan`` is used.
+            For datetime, timedelta, or period data, etc. :attr:`NaT` is used.
+            For extension dtypes, ``self.dtype.na_value`` is used.
+        suffix : str, optional
+            If str and periods is an iterable, this is added after the column
+            name and before the shift value for each shifted column name.
+
+        Returns
+        -------
+        {klass}
+            Copy of input object, shifted.
+
+        See Also
+        --------
+        Index.shift : Shift values of Index.
+        DatetimeIndex.shift : Shift values of DatetimeIndex.
+        PeriodIndex.shift : Shift values of PeriodIndex.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [[10, 13, 17], [20, 23, 27], [15, 18, 22], [30, 33, 37], [45, 48, 52]],
+        ...     columns=["Col1", "Col2", "Col3"],
+        ...     index=pd.date_range("2020-01-01", "2020-01-05"),
+        ... )
+        >>> df
+                    Col1  Col2  Col3
+        2020-01-01    10    13    17
+        2020-01-02    20    23    27
+        2020-01-03    15    18    22
+        2020-01-04    30    33    37
+        2020-01-05    45    48    52
+
+        >>> df.shift(periods=3)
+                    Col1  Col2  Col3
+        2020-01-01   NaN   NaN   NaN
+        2020-01-02   NaN   NaN   NaN
+        2020-01-03   NaN   NaN   NaN
+        2020-01-04  10.0  13.0  17.0
+        2020-01-05  20.0  23.0  27.0
+
+        >>> df.shift(periods=1, axis="columns")
+                    Col1  Col2  Col3
+        2020-01-01   NaN    10    13
+        2020-01-02   NaN    20    23
+        2020-01-03   NaN    15    18
+        2020-01-04   NaN    30    33
+        2020-01-05   NaN    45    48
+
+        >>> df.shift(periods=3, fill_value=0)
+                    Col1  Col2  Col3
+        2020-01-01     0     0     0
+        2020-01-02     0     0     0
+        2020-01-03     0     0     0
+        2020-01-04    10    13    17
+        2020-01-05    20    23    27
+
+        >>> df.shift(periods=3, freq="D")
+                    Col1  Col2  Col3
+        2020-01-04    10    13    17
+        2020-01-05    20    23    27
+        2020-01-06    15    18    22
+        2020-01-07    30    33    37
+        2020-01-08    45    48    52
+
+        >>> df.shift(periods=3, freq="infer")
+                    Col1  Col2  Col3
+        2020-01-04    10    13    17
+        2020-01-05    20    23    27
+        2020-01-06    15    18    22
+        2020-01-07    30    33    37
+        2020-01-08    45    48    52
+
+        >>> df["Col1"].shift(periods=[0, 1, 2])
+                    Col1_0  Col1_1  Col1_2
+        2020-01-01      10     NaN     NaN
+        2020-01-02      20    10.0     NaN
+        2020-01-03      15    20.0    10.0
+        2020-01-04      30    15.0    20.0
+        2020-01-05      45    30.0    15.0
+        """
+        axis = self._get_axis_number(axis)
+
+        if freq is not None and fill_value is not lib.no_default:
+            # GH#53832
+            raise ValueError(
+                "Passing a 'freq' together with a 'fill_value' is not allowed."
+            )
+
+        if periods == 0:
+            return self.copy(deep=False)
+
+        if is_list_like(periods) and isinstance(self, ABCSeries):
+            return self.to_frame().shift(
+                periods=periods, freq=freq, axis=axis, fill_value=fill_value
+            )
+        periods = cast(int, periods)
+
+        if freq is None:
+            # when freq is None, data is shifted, index is not
+            axis = self._get_axis_number(axis)
+            assert axis == 0  # axis == 1 cases handled in DataFrame.shift
+            new_data = self._mgr.shift(periods=periods, fill_value=fill_value)
+            return self._constructor_from_mgr(
+                new_data, axes=new_data.axes
+            ).__finalize__(self, method="shift")
+
+        return self._shift_with_freq(periods, axis, freq)
+
+    @final
+    def _shift_with_freq(self, periods: int, axis: int, freq) -> Self:
+        # see shift.__doc__
+        # when freq is given, index is shifted, data is not
+        index = self._get_axis(axis)
+
+        if freq == "infer":
+            freq = getattr(index, "freq", None)
+
+            if freq is None:
+                freq = getattr(index, "inferred_freq", None)
+
+            if freq is None:
+                msg = "Freq was not set in the index hence cannot be inferred"
+                raise ValueError(msg)
+
+        elif isinstance(freq, str):
+            is_period = isinstance(index, PeriodIndex)
+            freq = to_offset(freq, is_period=is_period)
+
+        if isinstance(index, PeriodIndex):
+            orig_freq = to_offset(index.freq)
+            if freq != orig_freq:
+                assert orig_freq is not None  # for mypy
+                raise ValueError(
+                    f"Given freq {PeriodDtype(freq)._freqstr} "
+                    f"does not match PeriodIndex freq "
+                    f"{PeriodDtype(orig_freq)._freqstr}"
+                )
+            new_ax: Index = index.shift(periods)
+        else:
+            new_ax = index.shift(periods, freq)
+
+        result = self.set_axis(new_ax, axis=axis)
+        return result.__finalize__(self, method="shift")
+
+    @final
+    def truncate(
+        self,
+        before=None,
+        after=None,
+        axis: Axis | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Self:
+        """
+        Truncate a Series or DataFrame before and after some index value.
+
+        This is a useful shorthand for boolean indexing based on index
+        values above or below certain thresholds.
+
+        Parameters
+        ----------
+        before : date, str, int
+            Truncate all rows before this index value.
+        after : date, str, int
+            Truncate all rows after this index value.
+        axis : {0 or 'index', 1 or 'columns'}, optional
+            Axis to truncate. Truncates the index (rows) by default.
+            For `Series` this parameter is unused and defaults to 0.
+        copy : bool, default is False,
+            Return a copy of the truncated section.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        type of caller
+            The truncated Series or DataFrame.
+
+        See Also
+        --------
+        DataFrame.loc : Select a subset of a DataFrame by label.
+        DataFrame.iloc : Select a subset of a DataFrame by position.
+
+        Notes
+        -----
+        If the index being truncated contains only datetime values,
+        `before` and `after` may be specified as strings instead of
+        Timestamps.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": ["a", "b", "c", "d", "e"],
+        ...         "B": ["f", "g", "h", "i", "j"],
+        ...         "C": ["k", "l", "m", "n", "o"],
+        ...     },
+        ...     index=[1, 2, 3, 4, 5],
+        ... )
+        >>> df
+           A  B  C
+        1  a  f  k
+        2  b  g  l
+        3  c  h  m
+        4  d  i  n
+        5  e  j  o
+
+        >>> df.truncate(before=2, after=4)
+           A  B  C
+        2  b  g  l
+        3  c  h  m
+        4  d  i  n
+
+        The columns of a DataFrame can be truncated.
+
+        >>> df.truncate(before="A", after="B", axis="columns")
+           A  B
+        1  a  f
+        2  b  g
+        3  c  h
+        4  d  i
+        5  e  j
+
+        For Series, only rows can be truncated.
+
+        >>> df["A"].truncate(before=2, after=4)
+        2    b
+        3    c
+        4    d
+        Name: A, dtype: object
+
+        The index values in ``truncate`` can be datetimes or string
+        dates.
+
+        >>> dates = pd.date_range("2016-01-01", "2016-02-01", freq="s")
+        >>> df = pd.DataFrame(index=dates, data={"A": 1})
+        >>> df.tail()
+                             A
+        2016-01-31 23:59:56  1
+        2016-01-31 23:59:57  1
+        2016-01-31 23:59:58  1
+        2016-01-31 23:59:59  1
+        2016-02-01 00:00:00  1
+
+        >>> df.truncate(
+        ...     before=pd.Timestamp("2016-01-05"), after=pd.Timestamp("2016-01-10")
+        ... ).tail()
+                             A
+        2016-01-09 23:59:56  1
+        2016-01-09 23:59:57  1
+        2016-01-09 23:59:58  1
+        2016-01-09 23:59:59  1
+        2016-01-10 00:00:00  1
+
+        Because the index is a DatetimeIndex containing only dates, we can
+        specify `before` and `after` as strings. They will be coerced to
+        Timestamps before truncation.
+
+        >>> df.truncate("2016-01-05", "2016-01-10").tail()
+                             A
+        2016-01-09 23:59:56  1
+        2016-01-09 23:59:57  1
+        2016-01-09 23:59:58  1
+        2016-01-09 23:59:59  1
+        2016-01-10 00:00:00  1
+
+        Note that ``truncate`` assumes a 0 value for any unspecified time
+        component (midnight). This differs from partial string slicing, which
+        returns any partially matching dates.
+
+        >>> df.loc["2016-01-05":"2016-01-10", :].tail()
+                             A
+        2016-01-10 23:59:55  1
+        2016-01-10 23:59:56  1
+        2016-01-10 23:59:57  1
+        2016-01-10 23:59:58  1
+        2016-01-10 23:59:59  1
+        """
+        self._check_copy_deprecation(copy)
+
+        if axis is None:
+            axis = 0
+        axis = self._get_axis_number(axis)
+        ax = self._get_axis(axis)
+
+        # GH 17935
+        # Check that index is sorted
+        if not ax.is_monotonic_increasing and not ax.is_monotonic_decreasing:
+            raise ValueError("truncate requires a sorted index")
+
+        # if we have a date index, convert to dates, otherwise
+        # treat like a slice
+        if ax._is_all_dates:
+            from pandas.core.tools.datetimes import to_datetime
+
+            before = to_datetime(before)
+            after = to_datetime(after)
+
+        if before is not None and after is not None and before > after:
+            raise ValueError(f"Truncate: {after} must be after {before}")
+
+        if len(ax) > 1 and ax.is_monotonic_decreasing and ax.nunique() > 1:
+            before, after = after, before
+
+        slicer = [slice(None, None)] * self._AXIS_LEN
+        slicer[axis] = slice(before, after)
+        result = self.loc[tuple(slicer)]
+
+        if isinstance(ax, MultiIndex):
+            setattr(result, self._get_axis_name(axis), ax.truncate(before, after))
+
+        result = result.copy(deep=False)
+
+        return result
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def tz_convert(
+        self,
+        tz,
+        axis: Axis = 0,
+        level=None,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Self:
+        """
+        Convert tz-aware axis to target time zone.
+
+        Parameters
+        ----------
+        tz : str or tzinfo object or None
+            Target time zone. Passing ``None`` will convert to
+            UTC and remove the timezone information.
+        axis : {{0 or 'index', 1 or 'columns'}}, default 0
+            The axis to convert
+        level : int, str, default None
+            If axis is a MultiIndex, convert a specific level. Otherwise
+            must be None.
+        copy : bool, default False
+            Also make a copy of the underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        {klass}
+            Object with time zone converted axis.
+
+        Raises
+        ------
+        TypeError
+            If the axis is tz-naive.
+
+        See Also
+        --------
+        DataFrame.tz_localize: Localize tz-naive index of DataFrame to target time zone.
+        Series.tz_localize: Localize tz-naive index of Series to target time zone.
+
+        Examples
+        --------
+        Change to another time zone:
+
+        >>> s = pd.Series(
+        ...     [1],
+        ...     index=pd.DatetimeIndex(["2018-09-15 01:30:00+02:00"]),
+        ... )
+        >>> s.tz_convert("Asia/Shanghai")
+        2018-09-15 07:30:00+08:00    1
+        dtype: int64
+
+        Pass None to convert to UTC and get a tz-naive index:
+
+        >>> s = pd.Series([1], index=pd.DatetimeIndex(["2018-09-15 01:30:00+02:00"]))
+        >>> s.tz_convert(None)
+        2018-09-14 23:30:00    1
+        dtype: int64
+        """
+        self._check_copy_deprecation(copy)
+        axis = self._get_axis_number(axis)
+        ax = self._get_axis(axis)
+
+        def _tz_convert(ax, tz):
+            if not hasattr(ax, "tz_convert"):
+                if len(ax) > 0:
+                    ax_name = self._get_axis_name(axis)
+                    raise TypeError(
+                        f"{ax_name} is not a valid DatetimeIndex or PeriodIndex"
+                    )
+                ax = DatetimeIndex([], tz=tz)
+            else:
+                ax = ax.tz_convert(tz)
+            return ax
+
+        # if a level is given it must be a MultiIndex level or
+        # equivalent to the axis name
+        if isinstance(ax, MultiIndex):
+            level = ax._get_level_number(level)
+            new_level = _tz_convert(ax.levels[level], tz)
+            ax = ax.set_levels(new_level, level=level)
+        else:
+            if level not in (None, 0, ax.name):
+                raise ValueError(f"The level {level} is not valid")
+            ax = _tz_convert(ax, tz)
+
+        result = self.copy(deep=False)
+        result = result.set_axis(ax, axis=axis)
+        return result.__finalize__(self, method="tz_convert")
+
+    @final
+    @doc(klass=_shared_doc_kwargs["klass"])
+    def tz_localize(
+        self,
+        tz,
+        axis: Axis = 0,
+        level=None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        """
+        Localize time zone naive index of a Series or DataFrame to target time zone.
+
+        This operation localizes the Index. To localize the values in a
+        time zone naive Series, use :meth:`Series.dt.tz_localize`.
+
+        Parameters
+        ----------
+        tz : str or tzinfo or None
+            Time zone to localize. Passing ``None`` will remove the
+            time zone information and preserve local time.
+        axis : {{0 or 'index', 1 or 'columns'}}, default 0
+            The axis to localize
+        level : int, str, default None
+            If axis ia a MultiIndex, localize a specific level. Otherwise
+            must be None.
+        copy : bool, default False
+            Also make a copy of the underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        ambiguous : 'infer', bool, bool-ndarray, 'NaT', default 'raise'
+            When clocks moved backward due to DST, ambiguous times may arise.
+            For example in Central European Time (UTC+01), when going from
+            03:00 DST to 02:00 non-DST, 02:30:00 local time occurs both at
+            00:30:00 UTC and at 01:30:00 UTC. In such a situation, the
+            `ambiguous` parameter dictates how ambiguous times should be
+            handled.
+
+            - 'infer' will attempt to infer fall dst-transition hours based on
+              order
+            - bool (or bool-ndarray) where True signifies a DST time, False designates
+              a non-DST time (note that this flag is only applicable for
+              ambiguous times)
+            - 'NaT' will return NaT where there are ambiguous times
+            - 'raise' will raise a ValueError if there are ambiguous
+              times.
+        nonexistent : str, default 'raise'
+            A nonexistent time does not exist in a particular timezone
+            where clocks moved forward due to DST. Valid values are:
+
+            - 'shift_forward' will shift the nonexistent time forward to the
+              closest existing time
+            - 'shift_backward' will shift the nonexistent time backward to the
+              closest existing time
+            - 'NaT' will return NaT where there are nonexistent times
+            - timedelta objects will shift nonexistent times by the timedelta
+            - 'raise' will raise a ValueError if there are
+              nonexistent times.
+
+        Returns
+        -------
+        {klass}
+            Same type as the input, with time zone naive or aware index, depending on
+            ``tz``.
+
+        Raises
+        ------
+        TypeError
+            If the TimeSeries is tz-aware and tz is not None.
+
+        See Also
+        --------
+        Series.dt.tz_localize: Localize the values in a time zone naive Series.
+        Timestamp.tz_localize: Localize the Timestamp to a timezone.
+
+        Examples
+        --------
+        Localize local times:
+
+        >>> s = pd.Series(
+        ...     [1],
+        ...     index=pd.DatetimeIndex(["2018-09-15 01:30:00"]),
+        ... )
+        >>> s.tz_localize("CET")
+        2018-09-15 01:30:00+02:00    1
+        dtype: int64
+
+        Pass None to convert to tz-naive index and preserve local time:
+
+        >>> s = pd.Series([1], index=pd.DatetimeIndex(["2018-09-15 01:30:00+02:00"]))
+        >>> s.tz_localize(None)
+        2018-09-15 01:30:00    1
+        dtype: int64
+
+        Be careful with DST changes. When there is sequential data, pandas
+        can infer the DST time:
+
+        >>> s = pd.Series(
+        ...     range(7),
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2018-10-28 01:30:00",
+        ...             "2018-10-28 02:00:00",
+        ...             "2018-10-28 02:30:00",
+        ...             "2018-10-28 02:00:00",
+        ...             "2018-10-28 02:30:00",
+        ...             "2018-10-28 03:00:00",
+        ...             "2018-10-28 03:30:00",
+        ...         ]
+        ...     ),
+        ... )
+        >>> s.tz_localize("CET", ambiguous="infer")
+        2018-10-28 01:30:00+02:00    0
+        2018-10-28 02:00:00+02:00    1
+        2018-10-28 02:30:00+02:00    2
+        2018-10-28 02:00:00+01:00    3
+        2018-10-28 02:30:00+01:00    4
+        2018-10-28 03:00:00+01:00    5
+        2018-10-28 03:30:00+01:00    6
+        dtype: int64
+
+        In some cases, inferring the DST is impossible. In such cases, you can
+        pass an ndarray to the ambiguous parameter to set the DST explicitly
+
+        >>> s = pd.Series(
+        ...     range(3),
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2018-10-28 01:20:00",
+        ...             "2018-10-28 02:36:00",
+        ...             "2018-10-28 03:46:00",
+        ...         ]
+        ...     ),
+        ... )
+        >>> s.tz_localize("CET", ambiguous=np.array([True, True, False]))
+        2018-10-28 01:20:00+02:00    0
+        2018-10-28 02:36:00+02:00    1
+        2018-10-28 03:46:00+01:00    2
+        dtype: int64
+
+        If the DST transition causes nonexistent times, you can shift these
+        dates forward or backward with a timedelta object or `'shift_forward'`
+        or `'shift_backward'`.
+
+        >>> dti = pd.DatetimeIndex(
+        ...     ["2015-03-29 02:30:00", "2015-03-29 03:30:00"], dtype="M8[ns]"
+        ... )
+        >>> s = pd.Series(range(2), index=dti)
+        >>> s.tz_localize("Europe/Warsaw", nonexistent="shift_forward")
+        2015-03-29 03:00:00+02:00    0
+        2015-03-29 03:30:00+02:00    1
+        dtype: int64
+        >>> s.tz_localize("Europe/Warsaw", nonexistent="shift_backward")
+        2015-03-29 01:59:59.999999999+01:00    0
+        2015-03-29 03:30:00+02:00              1
+        dtype: int64
+        >>> s.tz_localize("Europe/Warsaw", nonexistent=pd.Timedelta("1h"))
+        2015-03-29 03:30:00+02:00    0
+        2015-03-29 03:30:00+02:00    1
+        dtype: int64
+        """
+        self._check_copy_deprecation(copy)
+        nonexistent_options = ("raise", "NaT", "shift_forward", "shift_backward")
+        if nonexistent not in nonexistent_options and not isinstance(
+            nonexistent, dt.timedelta
+        ):
+            raise ValueError(
+                "The nonexistent argument must be one of 'raise', "
+                "'NaT', 'shift_forward', 'shift_backward' or "
+                "a timedelta object"
+            )
+
+        axis = self._get_axis_number(axis)
+        ax = self._get_axis(axis)
+
+        def _tz_localize(ax, tz, ambiguous, nonexistent):
+            if not hasattr(ax, "tz_localize"):
+                if len(ax) > 0:
+                    ax_name = self._get_axis_name(axis)
+                    raise TypeError(
+                        f"{ax_name} is not a valid DatetimeIndex or PeriodIndex"
+                    )
+                ax = DatetimeIndex([], tz=tz)
+            else:
+                ax = ax.tz_localize(tz, ambiguous=ambiguous, nonexistent=nonexistent)
+            return ax
+
+        # if a level is given it must be a MultiIndex level or
+        # equivalent to the axis name
+        if isinstance(ax, MultiIndex):
+            level = ax._get_level_number(level)
+            new_level = _tz_localize(ax.levels[level], tz, ambiguous, nonexistent)
+            ax = ax.set_levels(new_level, level=level)
+        else:
+            if level not in (None, 0, ax.name):
+                raise ValueError(f"The level {level} is not valid")
+            ax = _tz_localize(ax, tz, ambiguous, nonexistent)
+
+        result = self.copy(deep=False)
+        result = result.set_axis(ax, axis=axis)
+        return result.__finalize__(self, method="tz_localize")
+
+    # ----------------------------------------------------------------------
+    # Numeric Methods
+
+    @final
+    def describe(
+        self,
+        percentiles=None,
+        include=None,
+        exclude=None,
+    ) -> Self:
+        """
+        Generate descriptive statistics.
+
+        Descriptive statistics include those that summarize the central
+        tendency, dispersion and shape of a
+        dataset's distribution, excluding ``NaN`` values.
+
+        Analyzes both numeric and object series, as well
+        as ``DataFrame`` column sets of mixed data types. The output
+        will vary depending on what is provided. Refer to the notes
+        below for more detail.
+
+        Parameters
+        ----------
+        percentiles : list-like of numbers, optional
+            The percentiles to include in the output. All should
+            fall between 0 and 1. The default is
+            ``[.25, .5, .75]``, which returns the 25th, 50th, and
+            75th percentiles.
+        include : 'all', list-like of dtypes or None (default), optional
+            A white list of data types to include in the result. Ignored
+            for ``Series``. Here are the options:
+
+            - 'all' : All columns of the input will be included in the output.
+            - A list-like of dtypes : Limits the results to the
+              provided data types.
+              To limit the result to numeric types submit
+              ``numpy.number``. To limit it instead to object columns submit
+              the ``numpy.object`` data type. Strings
+              can also be used in the style of
+              ``select_dtypes`` (e.g. ``df.describe(include=['O'])``). To
+              select pandas categorical columns, use ``'category'``
+            - None (default) : The result will include all numeric columns.
+        exclude : list-like of dtypes or None (default), optional,
+            A black list of data types to omit from the result. Ignored
+            for ``Series``. Here are the options:
+
+            - A list-like of dtypes : Excludes the provided data types
+              from the result. To exclude numeric types submit
+              ``numpy.number``. To exclude object columns submit the data
+              type ``numpy.object``. Strings can also be used in the style of
+              ``select_dtypes`` (e.g. ``df.describe(exclude=['O'])``). To
+              exclude pandas categorical columns, use ``'category'``
+            - None (default) : The result will exclude nothing.
+
+        Returns
+        -------
+        Series or DataFrame
+            Summary statistics of the Series or Dataframe provided.
+
+        See Also
+        --------
+        DataFrame.count: Count number of non-NA/null observations.
+        DataFrame.max: Maximum of the values in the object.
+        DataFrame.min: Minimum of the values in the object.
+        DataFrame.mean: Mean of the values.
+        DataFrame.std: Standard deviation of the observations.
+        DataFrame.select_dtypes: Subset of a DataFrame including/excluding
+            columns based on their dtype.
+
+        Notes
+        -----
+        For numeric data, the result's index will include ``count``,
+        ``mean``, ``std``, ``min``, ``max`` as well as lower, ``50`` and
+        upper percentiles. By default the lower percentile is ``25`` and the
+        upper percentile is ``75``. The ``50`` percentile is the
+        same as the median.
+
+        For object data (e.g. strings), the result's index
+        will include ``count``, ``unique``, ``top``, and ``freq``. The ``top``
+        is the most common value. The ``freq`` is the most common value's
+        frequency.
+
+        If multiple object values have the highest count, then the
+        ``count`` and ``top`` results will be arbitrarily chosen from
+        among those with the highest count.
+
+        For mixed data types provided via a ``DataFrame``, the default is to
+        return only an analysis of numeric columns. If the DataFrame consists
+        only of object and categorical data without any numeric columns, the
+        default is to return an analysis of both the object and categorical
+        columns. If ``include='all'`` is provided as an option, the result
+        will include a union of attributes of each type.
+
+        The `include` and `exclude` parameters can be used to limit
+        which columns in a ``DataFrame`` are analyzed for the output.
+        The parameters are ignored when analyzing a ``Series``.
+
+        Examples
+        --------
+        Describing a numeric ``Series``.
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.describe()
+        count    3.0
+        mean     2.0
+        std      1.0
+        min      1.0
+        25%      1.5
+        50%      2.0
+        75%      2.5
+        max      3.0
+        dtype: float64
+
+        Describing a categorical ``Series``.
+
+        >>> s = pd.Series(["a", "a", "b", "c"])
+        >>> s.describe()
+        count     4
+        unique    3
+        top       a
+        freq      2
+        dtype: object
+
+        Describing a timestamp ``Series``.
+
+        >>> s = pd.Series(
+        ...     [
+        ...         np.datetime64("2000-01-01"),
+        ...         np.datetime64("2010-01-01"),
+        ...         np.datetime64("2010-01-01"),
+        ...     ]
+        ... )
+        >>> s.describe()
+        count                      3
+        mean     2006-09-01 08:00:00
+        min      2000-01-01 00:00:00
+        25%      2004-12-31 12:00:00
+        50%      2010-01-01 00:00:00
+        75%      2010-01-01 00:00:00
+        max      2010-01-01 00:00:00
+        dtype: object
+
+        Describing a ``DataFrame``. By default only numeric fields
+        are returned.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "categorical": pd.Categorical(["d", "e", "f"]),
+        ...         "numeric": [1, 2, 3],
+        ...         "object": ["a", "b", "c"],
+        ...     }
+        ... )
+        >>> df.describe()
+               numeric
+        count      3.0
+        mean       2.0
+        std        1.0
+        min        1.0
+        25%        1.5
+        50%        2.0
+        75%        2.5
+        max        3.0
+
+        Describing all columns of a ``DataFrame`` regardless of data type.
+
+        >>> df.describe(include="all")  # doctest: +SKIP
+               categorical  numeric object
+        count            3      3.0      3
+        unique           3      NaN      3
+        top              f      NaN      a
+        freq             1      NaN      1
+        mean           NaN      2.0    NaN
+        std            NaN      1.0    NaN
+        min            NaN      1.0    NaN
+        25%            NaN      1.5    NaN
+        50%            NaN      2.0    NaN
+        75%            NaN      2.5    NaN
+        max            NaN      3.0    NaN
+
+        Describing a column from a ``DataFrame`` by accessing it as
+        an attribute.
+
+        >>> df.numeric.describe()
+        count    3.0
+        mean     2.0
+        std      1.0
+        min      1.0
+        25%      1.5
+        50%      2.0
+        75%      2.5
+        max      3.0
+        Name: numeric, dtype: float64
+
+        Including only numeric columns in a ``DataFrame`` description.
+
+        >>> df.describe(include=[np.number])
+               numeric
+        count      3.0
+        mean       2.0
+        std        1.0
+        min        1.0
+        25%        1.5
+        50%        2.0
+        75%        2.5
+        max        3.0
+
+        Including only string columns in a ``DataFrame`` description.
+
+        >>> df.describe(include=[object])  # doctest: +SKIP
+               object
+        count       3
+        unique      3
+        top         a
+        freq        1
+
+        Including only categorical columns from a ``DataFrame`` description.
+
+        >>> df.describe(include=["category"])
+               categorical
+        count            3
+        unique           3
+        top              d
+        freq             1
+
+        Excluding numeric columns from a ``DataFrame`` description.
+
+        >>> df.describe(exclude=[np.number])  # doctest: +SKIP
+               categorical object
+        count            3      3
+        unique           3      3
+        top              f      a
+        freq             1      1
+
+        Excluding object columns from a ``DataFrame`` description.
+
+        >>> df.describe(exclude=[object])  # doctest: +SKIP
+               categorical  numeric
+        count            3      3.0
+        unique           3      NaN
+        top              f      NaN
+        freq             1      NaN
+        mean           NaN      2.0
+        std            NaN      1.0
+        min            NaN      1.0
+        25%            NaN      1.5
+        50%            NaN      2.0
+        75%            NaN      2.5
+        max            NaN      3.0
+        """
+        return describe_ndframe(
+            obj=self,
+            include=include,
+            exclude=exclude,
+            percentiles=percentiles,
+        ).__finalize__(self, method="describe")
+
+    @final
+    def pct_change(
+        self,
+        periods: int = 1,
+        fill_method: None = None,
+        freq=None,
+        **kwargs,
+    ) -> Self:
+        """
+        Fractional change between the current and a prior element.
+
+        Computes the fractional change from the immediately previous row by
+        default. This is useful in comparing the fraction of change in a time
+        series of elements.
+
+        .. note::
+
+            Despite the name of this method, it calculates fractional change
+            (also known as per unit change or relative change) and not
+            percentage change. If you need the percentage change, multiply
+            these values by 100.
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Periods to shift for forming percent change.
+        fill_method : None
+            Must be None. This argument will be removed in a future version of pandas.
+
+            .. deprecated:: 2.1
+                All options of `fill_method` are deprecated except `fill_method=None`.
+
+        freq : DateOffset, timedelta, or str, optional
+            Increment to use from time series API (e.g. 'ME' or BDay()).
+        **kwargs
+            Additional keyword arguments are passed into
+            `DataFrame.shift` or `Series.shift`.
+
+        Returns
+        -------
+        Series or DataFrame
+            The same type as the calling object.
+
+        See Also
+        --------
+        Series.diff : Compute the difference of two elements in a Series.
+        DataFrame.diff : Compute the difference of two elements in a DataFrame.
+        Series.shift : Shift the index by some number of periods.
+        DataFrame.shift : Shift the index by some number of periods.
+
+        Examples
+        --------
+        **Series**
+
+        >>> s = pd.Series([90, 91, 85])
+        >>> s
+        0    90
+        1    91
+        2    85
+        dtype: int64
+
+        >>> s.pct_change()
+        0         NaN
+        1    0.011111
+        2   -0.065934
+        dtype: float64
+
+        >>> s.pct_change(periods=2)
+        0         NaN
+        1         NaN
+        2   -0.055556
+        dtype: float64
+
+        See the percentage change in a Series where filling NAs with last
+        valid observation forward to next valid.
+
+        >>> s = pd.Series([90, 91, None, 85])
+        >>> s
+        0    90.0
+        1    91.0
+        2     NaN
+        3    85.0
+        dtype: float64
+
+        >>> s.ffill().pct_change()
+        0         NaN
+        1    0.011111
+        2    0.000000
+        3   -0.065934
+        dtype: float64
+
+        **DataFrame**
+
+        Percentage change in French franc, Deutsche Mark, and Italian lira from
+        1980-01-01 to 1980-03-01.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "FR": [4.0405, 4.0963, 4.3149],
+        ...         "GR": [1.7246, 1.7482, 1.8519],
+        ...         "IT": [804.74, 810.01, 860.13],
+        ...     },
+        ...     index=["1980-01-01", "1980-02-01", "1980-03-01"],
+        ... )
+        >>> df
+                        FR      GR      IT
+        1980-01-01  4.0405  1.7246  804.74
+        1980-02-01  4.0963  1.7482  810.01
+        1980-03-01  4.3149  1.8519  860.13
+
+        >>> df.pct_change()
+                          FR        GR        IT
+        1980-01-01       NaN       NaN       NaN
+        1980-02-01  0.013810  0.013684  0.006549
+        1980-03-01  0.053365  0.059318  0.061876
+
+        Percentage of change in GOOG and APPL stock volume. Shows computing
+        the percentage change between columns.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "2016": [1769950, 30586265],
+        ...         "2015": [1500923, 40912316],
+        ...         "2014": [1371819, 41403351],
+        ...     },
+        ...     index=["GOOG", "APPL"],
+        ... )
+        >>> df
+                  2016      2015      2014
+        GOOG   1769950   1500923   1371819
+        APPL  30586265  40912316  41403351
+
+        >>> df.pct_change(axis="columns", periods=-1)
+                  2016      2015  2014
+        GOOG  0.179241  0.094112   NaN
+        APPL -0.252395 -0.011860   NaN
+        """
+        # GH#53491
+        if fill_method is not None:
+            raise ValueError(f"fill_method must be None; got {fill_method=}.")
+
+        axis = self._get_axis_number(kwargs.pop("axis", "index"))
+        shifted = self.shift(periods=periods, freq=freq, axis=axis, **kwargs)
+        # Unsupported left operand type for / ("Self")
+        rs = self / shifted - 1  # type: ignore[operator]
+        if freq is not None:
+            # Shift method is implemented differently when freq is not None
+            # We want to restore the original index
+            rs = rs.loc[~rs.index.duplicated()]
+            rs = rs.reindex_like(self)
+        return rs.__finalize__(self, method="pct_change")
+
+    @final
+    def _logical_func(
+        self,
+        name: str,
+        func,
+        axis: Axis | None = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> Series | bool:
+        nv.validate_logical_func((), kwargs, fname=name)
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+
+        if self.ndim > 1 and axis is None:
+            # Reduce along one dimension then the other, to simplify DataFrame._reduce
+            res = self._logical_func(
+                name, func, axis=0, bool_only=bool_only, skipna=skipna, **kwargs
+            )
+            # error: Item "bool" of "Series | bool" has no attribute "_logical_func"
+            return res._logical_func(  # type: ignore[union-attr]
+                name, func, skipna=skipna, **kwargs
+            )
+        elif axis is None:
+            axis = 0
+
+        if (
+            self.ndim > 1
+            and axis == 1
+            and len(self._mgr.blocks) > 1
+            # TODO(EA2D): special-case not needed
+            and all(block.values.ndim == 2 for block in self._mgr.blocks)
+            and not kwargs
+        ):
+            # Fastpath avoiding potentially expensive transpose
+            obj = self
+            if bool_only:
+                obj = self._get_bool_data()
+            return obj._reduce_axis1(name, func, skipna=skipna)
+
+        return self._reduce(
+            func,
+            name=name,
+            axis=axis,
+            skipna=skipna,
+            numeric_only=bool_only,
+            filter_type="bool",
+        )
+
+    def any(
+        self,
+        *,
+        axis: Axis | None = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> Series | bool:
+        return self._logical_func(
+            "any", nanops.nanany, axis, bool_only, skipna, **kwargs
+        )
+
+    def all(
+        self,
+        *,
+        axis: Axis = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> Series | bool:
+        return self._logical_func(
+            "all", nanops.nanall, axis, bool_only, skipna, **kwargs
+        )
+
+    @final
+    def _accum_func(
+        self,
+        name: str,
+        func,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        *args,
+        **kwargs,
+    ):
+        skipna = nv.validate_cum_func_with_skipna(skipna, args, kwargs, name)
+        if axis is None:
+            axis = 0
+        else:
+            axis = self._get_axis_number(axis)
+
+        if axis == 1:
+            return self.T._accum_func(
+                name,
+                func,
+                axis=0,
+                skipna=skipna,
+                *args,  # noqa: B026
+                **kwargs,
+            ).T
+
+        def block_accum_func(blk_values):
+            values = blk_values.T if hasattr(blk_values, "T") else blk_values
+
+            result: np.ndarray | ExtensionArray
+            if isinstance(values, ExtensionArray):
+                result = values._accumulate(name, skipna=skipna, **kwargs)
+            else:
+                result = nanops.na_accum_func(values, func, skipna=skipna)
+
+            result = result.T if hasattr(result, "T") else result
+            return result
+
+        result = self._mgr.apply(block_accum_func)
+
+        return self._constructor_from_mgr(result, axes=result.axes).__finalize__(
+            self, method=name
+        )
+
+    def cummax(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return self._accum_func(
+            "cummax", np.maximum.accumulate, axis, skipna, *args, **kwargs
+        )
+
+    def cummin(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return self._accum_func(
+            "cummin", np.minimum.accumulate, axis, skipna, *args, **kwargs
+        )
+
+    def cumsum(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return self._accum_func("cumsum", np.cumsum, axis, skipna, *args, **kwargs)
+
+    def cumprod(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return self._accum_func("cumprod", np.cumprod, axis, skipna, *args, **kwargs)
+
+    @final
+    def _stat_function_ddof(
+        self,
+        name: str,
+        func,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        nv.validate_stat_ddof_func((), kwargs, fname=name)
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+
+        return self._reduce(
+            func, name, axis=axis, numeric_only=numeric_only, skipna=skipna, ddof=ddof
+        )
+
+    def sem(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function_ddof(
+            "sem", nanops.nansem, axis, skipna, ddof, numeric_only, **kwargs
+        )
+
+    def var(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function_ddof(
+            "var", nanops.nanvar, axis, skipna, ddof, numeric_only, **kwargs
+        )
+
+    def std(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function_ddof(
+            "std", nanops.nanstd, axis, skipna, ddof, numeric_only, **kwargs
+        )
+
+    @final
+    def _stat_function(
+        self,
+        name: str,
+        func,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        assert name in ["median", "mean", "min", "max", "kurt", "skew"], name
+        nv.validate_func(name, (), kwargs)
+
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+
+        return self._reduce(
+            func, name=name, axis=axis, skipna=skipna, numeric_only=numeric_only
+        )
+
+    def min(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        return self._stat_function(
+            "min",
+            nanops.nanmin,
+            axis,
+            skipna,
+            numeric_only,
+            **kwargs,
+        )
+
+    def max(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        return self._stat_function(
+            "max",
+            nanops.nanmax,
+            axis,
+            skipna,
+            numeric_only,
+            **kwargs,
+        )
+
+    def mean(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function(
+            "mean", nanops.nanmean, axis, skipna, numeric_only, **kwargs
+        )
+
+    def median(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function(
+            "median", nanops.nanmedian, axis, skipna, numeric_only, **kwargs
+        )
+
+    def skew(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function(
+            "skew", nanops.nanskew, axis, skipna, numeric_only, **kwargs
+        )
+
+    def kurt(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series | float:
+        return self._stat_function(
+            "kurt", nanops.nankurt, axis, skipna, numeric_only, **kwargs
+        )
+
+    kurtosis = kurt
+
+    @final
+    def _min_count_stat_function(
+        self,
+        name: str,
+        func,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ):
+        assert name in ["sum", "prod"], name
+        nv.validate_func(name, (), kwargs)
+
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+
+        return self._reduce(
+            func,
+            name=name,
+            axis=axis,
+            skipna=skipna,
+            numeric_only=numeric_only,
+            min_count=min_count,
+        )
+
+    def sum(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ):
+        return self._min_count_stat_function(
+            "sum", nanops.nansum, axis, skipna, numeric_only, min_count, **kwargs
+        )
+
+    def prod(
+        self,
+        *,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ):
+        return self._min_count_stat_function(
+            "prod",
+            nanops.nanprod,
+            axis,
+            skipna,
+            numeric_only,
+            min_count,
+            **kwargs,
+        )
+
+    product = prod
+
+    @final
+    @doc(Rolling)
+    def rolling(
+        self,
+        window: int | dt.timedelta | str | BaseOffset | BaseIndexer,
+        min_periods: int | None = None,
+        center: bool = False,
+        win_type: str | None = None,
+        on: str | None = None,
+        closed: IntervalClosedType | None = None,
+        step: int | None = None,
+        method: str = "single",
+    ) -> Window | Rolling:
+        if win_type is not None:
+            return Window(
+                self,
+                window=window,
+                min_periods=min_periods,
+                center=center,
+                win_type=win_type,
+                on=on,
+                closed=closed,
+                step=step,
+                method=method,
+            )
+
+        return Rolling(
+            self,
+            window=window,
+            min_periods=min_periods,
+            center=center,
+            win_type=win_type,
+            on=on,
+            closed=closed,
+            step=step,
+            method=method,
+        )
+
+    @final
+    @doc(Expanding)
+    def expanding(
+        self,
+        min_periods: int = 1,
+        method: Literal["single", "table"] = "single",
+    ) -> Expanding:
+        return Expanding(self, min_periods=min_periods, method=method)
+
+    @final
+    @doc(ExponentialMovingWindow)
+    def ewm(
+        self,
+        com: float | None = None,
+        span: float | None = None,
+        halflife: float | TimedeltaConvertibleTypes | None = None,
+        alpha: float | None = None,
+        min_periods: int | None = 0,
+        adjust: bool = True,
+        ignore_na: bool = False,
+        times: np.ndarray | DataFrame | Series | None = None,
+        method: Literal["single", "table"] = "single",
+    ) -> ExponentialMovingWindow:
+        return ExponentialMovingWindow(
+            self,
+            com=com,
+            span=span,
+            halflife=halflife,
+            alpha=alpha,
+            min_periods=min_periods,
+            adjust=adjust,
+            ignore_na=ignore_na,
+            times=times,
+            method=method,
+        )
+
+    # ----------------------------------------------------------------------
+    # Arithmetic Methods
+
+    @final
+    def _inplace_method(self, other, op) -> Self:
+        """
+        Wrap arithmetic method to operate inplace.
+        """
+        result = op(self, other)
+
+        # this makes sure that we are aligned like the input
+        # we are updating inplace
+        self._update_inplace(result.reindex_like(self))
+        return self
+
+    @final
+    def __iadd__(self, other) -> Self:
+        # error: Unsupported left operand type for + ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__add__)  # type: ignore[operator]
+
+    @final
+    def __isub__(self, other) -> Self:
+        # error: Unsupported left operand type for - ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__sub__)  # type: ignore[operator]
+
+    @final
+    def __imul__(self, other) -> Self:
+        # error: Unsupported left operand type for * ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__mul__)  # type: ignore[operator]
+
+    @final
+    def __itruediv__(self, other) -> Self:
+        # error: Unsupported left operand type for / ("Type[NDFrame]")
+        return self._inplace_method(
+            other,
+            type(self).__truediv__,  # type: ignore[operator]
+        )
+
+    @final
+    def __ifloordiv__(self, other) -> Self:
+        # error: Unsupported left operand type for // ("Type[NDFrame]")
+        return self._inplace_method(
+            other,
+            type(self).__floordiv__,  # type: ignore[operator]
+        )
+
+    @final
+    def __imod__(self, other) -> Self:
+        # error: Unsupported left operand type for % ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__mod__)  # type: ignore[operator]
+
+    @final
+    def __ipow__(self, other) -> Self:
+        # error: Unsupported left operand type for ** ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__pow__)  # type: ignore[operator]
+
+    @final
+    def __iand__(self, other) -> Self:
+        # error: Unsupported left operand type for & ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__and__)  # type: ignore[operator]
+
+    @final
+    def __ior__(self, other) -> Self:
+        return self._inplace_method(other, type(self).__or__)
+
+    @final
+    def __ixor__(self, other) -> Self:
+        # error: Unsupported left operand type for ^ ("Type[NDFrame]")
+        return self._inplace_method(other, type(self).__xor__)  # type: ignore[operator]
+
+    # ----------------------------------------------------------------------
+    # Misc methods
+
+    @final
+    def _find_valid_index(self, *, how: str) -> Hashable:
+        """
+        Retrieves the index of the first valid value.
+
+        Parameters
+        ----------
+        how : {'first', 'last'}
+            Use this parameter to change between the first or last valid index.
+
+        Returns
+        -------
+        idx_first_valid : type of index
+        """
+        is_valid = self.notna().values
+        idxpos = find_valid_index(how=how, is_valid=is_valid)
+        if idxpos is None:
+            return None
+        return self.index[idxpos]
+
+    @final
+    @doc(position="first", klass=_shared_doc_kwargs["klass"])
+    def first_valid_index(self) -> Hashable:
+        """
+        Return index for {position} non-missing value or None, if no value is found.
+
+        See the :ref:`User Guide <missing_data>` for more information
+        on which values are considered missing.
+
+        Returns
+        -------
+        type of index
+            Index of {position} non-missing value.
+
+        See Also
+        --------
+        DataFrame.last_valid_index : Return index for last non-NA value or None, if
+            no non-NA value is found.
+        Series.last_valid_index : Return index for last non-NA value or None, if no
+            non-NA value is found.
+        DataFrame.isna : Detect missing values.
+
+        Examples
+        --------
+        For Series:
+
+        >>> s = pd.Series([None, 3, 4])
+        >>> s.first_valid_index()
+        1
+        >>> s.last_valid_index()
+        2
+
+        >>> s = pd.Series([None, None])
+        >>> print(s.first_valid_index())
+        None
+        >>> print(s.last_valid_index())
+        None
+
+        If all elements in Series are NA/null, returns None.
+
+        >>> s = pd.Series()
+        >>> print(s.first_valid_index())
+        None
+        >>> print(s.last_valid_index())
+        None
+
+        If Series is empty, returns None.
+
+        For DataFrame:
+
+        >>> df = pd.DataFrame({{"A": [None, None, 2], "B": [None, 3, 4]}})
+        >>> df
+             A      B
+        0  NaN    NaN
+        1  NaN    3.0
+        2  2.0    4.0
+        >>> df.first_valid_index()
+        1
+        >>> df.last_valid_index()
+        2
+
+        >>> df = pd.DataFrame({{"A": [None, None, None], "B": [None, None, None]}})
+        >>> df
+             A      B
+        0  None   None
+        1  None   None
+        2  None   None
+        >>> print(df.first_valid_index())
+        None
+        >>> print(df.last_valid_index())
+        None
+
+        If all elements in DataFrame are NA/null, returns None.
+
+        >>> df = pd.DataFrame()
+        >>> df
+        Empty DataFrame
+        Columns: []
+        Index: []
+        >>> print(df.first_valid_index())
+        None
+        >>> print(df.last_valid_index())
+        None
+
+        If DataFrame is empty, returns None.
+        """
+        return self._find_valid_index(how="first")
+
+    @final
+    @doc(first_valid_index, position="last", klass=_shared_doc_kwargs["klass"])
+    def last_valid_index(self) -> Hashable:
+        return self._find_valid_index(how="last")
+
+
+_num_doc = """
+{desc}
+
+Parameters
+----------
+axis : {axis_descr}
+    Axis for the function to be applied on.
+    For `Series` this parameter is unused and defaults to 0.
+
+    For DataFrames, specifying ``axis=None`` will apply the aggregation
+    across both axes.
+
+    .. versionadded:: 2.0.0
+
+skipna : bool, default True
+    Exclude NA/null values when computing the result.
+numeric_only : bool, default False
+    Include only float, int, boolean columns.
+
+{min_count}\
+**kwargs
+    Additional keyword arguments to be passed to the function.
+
+Returns
+-------
+{name1} or scalar\
+
+    Value containing the calculation referenced in the description.\
+{see_also}\
+{examples}
+"""
+
+_sum_prod_doc = """
+{desc}
+
+Parameters
+----------
+axis : {axis_descr}
+    Axis for the function to be applied on.
+    For `Series` this parameter is unused and defaults to 0.
+
+    .. warning::
+
+        The behavior of DataFrame.{name} with ``axis=None`` is deprecated,
+        in a future version this will reduce over both axes and return a scalar
+        To retain the old behavior, pass axis=0 (or do not pass axis).
+
+    .. versionadded:: 2.0.0
+
+skipna : bool, default True
+    Exclude NA/null values when computing the result.
+numeric_only : bool, default False
+    Include only float, int, boolean columns. Not implemented for Series.
+
+{min_count}\
+**kwargs
+    Additional keyword arguments to be passed to the function.
+
+Returns
+-------
+{name1} or scalar\
+
+    Value containing the calculation referenced in the description.\
+{see_also}\
+{examples}
+"""
+
+_num_ddof_doc = """
+{desc}
+
+Parameters
+----------
+axis : {axis_descr}
+    For `Series` this parameter is unused and defaults to 0.
+
+    .. warning::
+
+        The behavior of DataFrame.{name} with ``axis=None`` is deprecated,
+        in a future version this will reduce over both axes and return a scalar
+        To retain the old behavior, pass axis=0 (or do not pass axis).
+
+skipna : bool, default True
+    Exclude NA/null values. If an entire row/column is NA, the result
+    will be NA.
+ddof : int, default 1
+    Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+    where N represents the number of elements.
+numeric_only : bool, default False
+    Include only float, int, boolean columns. Not implemented for Series.
+**kwargs :
+    Additional keywords have no effect but might be accepted
+    for compatibility with NumPy.
+
+Returns
+-------
+{name1} or {name2} (if level specified)
+    {return_desc}
+
+See Also
+--------
+{see_also}\
+{notes}\
+{examples}
+"""
+
+_sem_see_also = """\
+scipy.stats.sem : Compute standard error of the mean.
+{name2}.std : Return sample standard deviation over requested axis.
+{name2}.var : Return unbiased variance over requested axis.
+{name2}.mean : Return the mean of the values over the requested axis.
+{name2}.median : Return the median of the values over the requested axis.
+{name2}.mode : Return the mode(s) of the Series."""
+
+_sem_return_desc = """\
+Unbiased standard error of the mean over requested axis."""
+
+_std_see_also = """\
+numpy.std : Compute the standard deviation along the specified axis.
+{name2}.var : Return unbiased variance over requested axis.
+{name2}.sem : Return unbiased standard error of the mean over requested axis.
+{name2}.mean : Return the mean of the values over the requested axis.
+{name2}.median : Return the median of the values over the requested axis.
+{name2}.mode : Return the mode(s) of the Series."""
+
+_std_return_desc = """\
+Standard deviation over requested axis."""
+
+_std_notes = """
+
+Notes
+-----
+To have the same behaviour as `numpy.std`, use `ddof=0` (instead of the
+default `ddof=1`)"""
+
+_std_examples = """
+
+Examples
+--------
+>>> df = pd.DataFrame({'person_id': [0, 1, 2, 3],
+...                    'age': [21, 25, 62, 43],
+...                    'height': [1.61, 1.87, 1.49, 2.01]}
+...                   ).set_index('person_id')
+>>> df
+           age  height
+person_id
+0           21    1.61
+1           25    1.87
+2           62    1.49
+3           43    2.01
+
+The standard deviation of the columns can be found as follows:
+
+>>> df.std()
+age       18.786076
+height     0.237417
+dtype: float64
+
+Alternatively, `ddof=0` can be set to normalize by N instead of N-1:
+
+>>> df.std(ddof=0)
+age       16.269219
+height     0.205609
+dtype: float64"""
+
+_var_examples = """
+
+Examples
+--------
+>>> df = pd.DataFrame({'person_id': [0, 1, 2, 3],
+...                    'age': [21, 25, 62, 43],
+...                    'height': [1.61, 1.87, 1.49, 2.01]}
+...                   ).set_index('person_id')
+>>> df
+           age  height
+person_id
+0           21    1.61
+1           25    1.87
+2           62    1.49
+3           43    2.01
+
+>>> df.var()
+age       352.916667
+height      0.056367
+dtype: float64
+
+Alternatively, ``ddof=0`` can be set to normalize by N instead of N-1:
+
+>>> df.var(ddof=0)
+age       264.687500
+height      0.042275
+dtype: float64"""
+
+_bool_doc = """
+{desc}
+
+Parameters
+----------
+axis : {{0 or 'index', 1 or 'columns', None}}, default 0
+    Indicate which axis or axes should be reduced. For `Series` this parameter
+    is unused and defaults to 0.
+
+    * 0 / 'index' : reduce the index, return a Series whose index is the
+      original column labels.
+    * 1 / 'columns' : reduce the columns, return a Series whose index is the
+      original index.
+    * None : reduce all axes, return a scalar.
+
+bool_only : bool, default False
+    Include only boolean columns. Not implemented for Series.
+skipna : bool, default True
+    Exclude NA/null values. If the entire row/column is NA and skipna is
+    True, then the result will be {empty_value}, as for an empty row/column.
+    If skipna is False, then NA are treated as True, because these are not
+    equal to zero.
+**kwargs : any, default None
+    Additional keywords have no effect but might be accepted for
+    compatibility with NumPy.
+
+Returns
+-------
+{name2} or {name1}
+    If axis=None, then a scalar boolean is returned.
+    Otherwise a Series is returned with index matching the index argument.
+
+{see_also}
+{examples}"""
+
+_all_desc = """\
+Return whether all elements are True, potentially over an axis.
+
+Returns True unless there at least one element within a series or
+along a Dataframe axis that is False or equivalent (e.g. zero or
+empty)."""
+
+_all_examples = """\
+Examples
+--------
+**Series**
+
+>>> pd.Series([True, True]).all()
+True
+>>> pd.Series([True, False]).all()
+False
+>>> pd.Series([], dtype="float64").all()
+True
+>>> pd.Series([np.nan]).all()
+True
+>>> pd.Series([np.nan]).all(skipna=False)
+True
+
+**DataFrames**
+
+Create a DataFrame from a dictionary.
+
+>>> df = pd.DataFrame({'col1': [True, True], 'col2': [True, False]})
+>>> df
+   col1   col2
+0  True   True
+1  True  False
+
+Default behaviour checks if values in each column all return True.
+
+>>> df.all()
+col1     True
+col2    False
+dtype: bool
+
+Specify ``axis='columns'`` to check if values in each row all return True.
+
+>>> df.all(axis='columns')
+0     True
+1    False
+dtype: bool
+
+Or ``axis=None`` for whether every value is True.
+
+>>> df.all(axis=None)
+False
+"""
+
+_all_see_also = """\
+See Also
+--------
+Series.all : Return True if all elements are True.
+DataFrame.any : Return True if one (or more) elements are True.
+"""
+
+_cnum_pd_doc = """
+Return cumulative {desc} over a DataFrame or Series axis.
+
+Returns a DataFrame or Series of the same size containing the cumulative
+{desc}.
+
+Parameters
+----------
+axis : {{0 or 'index', 1 or 'columns'}}, default 0
+    The index or the name of the axis. 0 is equivalent to None or 'index'.
+    For `Series` this parameter is unused and defaults to 0.
+skipna : bool, default True
+    Exclude NA/null values. If an entire row/column is NA, the result
+    will be NA.
+numeric_only : bool, default False
+    Include only float, int, boolean columns.
+*args, **kwargs
+    Additional keywords have no effect but might be accepted for
+    compatibility with NumPy.
+
+Returns
+-------
+{name1} or {name2}
+    Return cumulative {desc} of {name1} or {name2}.
+
+See Also
+--------
+core.window.expanding.Expanding.{accum_func_name} : Similar functionality
+    but ignores ``NaN`` values.
+{name2}.{accum_func_name} : Return the {desc} over
+    {name2} axis.
+{name2}.cummax : Return cumulative maximum over {name2} axis.
+{name2}.cummin : Return cumulative minimum over {name2} axis.
+{name2}.cumsum : Return cumulative sum over {name2} axis.
+{name2}.cumprod : Return cumulative product over {name2} axis.
+
+{examples}"""
+
+_cnum_series_doc = """
+Return cumulative {desc} over a DataFrame or Series axis.
+
+Returns a DataFrame or Series of the same size containing the cumulative
+{desc}.
+
+Parameters
+----------
+axis : {{0 or 'index', 1 or 'columns'}}, default 0
+    The index or the name of the axis. 0 is equivalent to None or 'index'.
+    For `Series` this parameter is unused and defaults to 0.
+skipna : bool, default True
+    Exclude NA/null values. If an entire row/column is NA, the result
+    will be NA.
+*args, **kwargs
+    Additional keywords have no effect but might be accepted for
+    compatibility with NumPy.
+
+Returns
+-------
+{name1} or {name2}
+    Return cumulative {desc} of {name1} or {name2}.
+
+See Also
+--------
+core.window.expanding.Expanding.{accum_func_name} : Similar functionality
+    but ignores ``NaN`` values.
+{name2}.{accum_func_name} : Return the {desc} over
+    {name2} axis.
+{name2}.cummax : Return cumulative maximum over {name2} axis.
+{name2}.cummin : Return cumulative minimum over {name2} axis.
+{name2}.cumsum : Return cumulative sum over {name2} axis.
+{name2}.cumprod : Return cumulative product over {name2} axis.
+
+{examples}"""
+
+_cummin_examples = """\
+Examples
+--------
+**Series**
+
+>>> s = pd.Series([2, np.nan, 5, -1, 0])
+>>> s
+0    2.0
+1    NaN
+2    5.0
+3   -1.0
+4    0.0
+dtype: float64
+
+By default, NA values are ignored.
+
+>>> s.cummin()
+0    2.0
+1    NaN
+2    2.0
+3   -1.0
+4   -1.0
+dtype: float64
+
+To include NA values in the operation, use ``skipna=False``
+
+>>> s.cummin(skipna=False)
+0    2.0
+1    NaN
+2    NaN
+3    NaN
+4    NaN
+dtype: float64
+
+**DataFrame**
+
+>>> df = pd.DataFrame([[2.0, 1.0],
+...                    [3.0, np.nan],
+...                    [1.0, 0.0]],
+...                   columns=list('AB'))
+>>> df
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  1.0  0.0
+
+By default, iterates over rows and finds the minimum
+in each column. This is equivalent to ``axis=None`` or ``axis='index'``.
+
+>>> df.cummin()
+     A    B
+0  2.0  1.0
+1  2.0  NaN
+2  1.0  0.0
+
+To iterate over columns and find the minimum in each row,
+use ``axis=1``
+
+>>> df.cummin(axis=1)
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  1.0  0.0
+"""
+
+_cumsum_examples = """\
+Examples
+--------
+**Series**
+
+>>> s = pd.Series([2, np.nan, 5, -1, 0])
+>>> s
+0    2.0
+1    NaN
+2    5.0
+3   -1.0
+4    0.0
+dtype: float64
+
+By default, NA values are ignored.
+
+>>> s.cumsum()
+0    2.0
+1    NaN
+2    7.0
+3    6.0
+4    6.0
+dtype: float64
+
+To include NA values in the operation, use ``skipna=False``
+
+>>> s.cumsum(skipna=False)
+0    2.0
+1    NaN
+2    NaN
+3    NaN
+4    NaN
+dtype: float64
+
+**DataFrame**
+
+>>> df = pd.DataFrame([[2.0, 1.0],
+...                    [3.0, np.nan],
+...                    [1.0, 0.0]],
+...                   columns=list('AB'))
+>>> df
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  1.0  0.0
+
+By default, iterates over rows and finds the sum
+in each column. This is equivalent to ``axis=None`` or ``axis='index'``.
+
+>>> df.cumsum()
+     A    B
+0  2.0  1.0
+1  5.0  NaN
+2  6.0  1.0
+
+To iterate over columns and find the sum in each row,
+use ``axis=1``
+
+>>> df.cumsum(axis=1)
+     A    B
+0  2.0  3.0
+1  3.0  NaN
+2  1.0  1.0
+"""
+
+_cumprod_examples = """\
+Examples
+--------
+**Series**
+
+>>> s = pd.Series([2, np.nan, 5, -1, 0])
+>>> s
+0    2.0
+1    NaN
+2    5.0
+3   -1.0
+4    0.0
+dtype: float64
+
+By default, NA values are ignored.
+
+>>> s.cumprod()
+0     2.0
+1     NaN
+2    10.0
+3   -10.0
+4    -0.0
+dtype: float64
+
+To include NA values in the operation, use ``skipna=False``
+
+>>> s.cumprod(skipna=False)
+0    2.0
+1    NaN
+2    NaN
+3    NaN
+4    NaN
+dtype: float64
+
+**DataFrame**
+
+>>> df = pd.DataFrame([[2.0, 1.0],
+...                    [3.0, np.nan],
+...                    [1.0, 0.0]],
+...                   columns=list('AB'))
+>>> df
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  1.0  0.0
+
+By default, iterates over rows and finds the product
+in each column. This is equivalent to ``axis=None`` or ``axis='index'``.
+
+>>> df.cumprod()
+     A    B
+0  2.0  1.0
+1  6.0  NaN
+2  6.0  0.0
+
+To iterate over columns and find the product in each row,
+use ``axis=1``
+
+>>> df.cumprod(axis=1)
+     A    B
+0  2.0  2.0
+1  3.0  NaN
+2  1.0  0.0
+"""
+
+_cummax_examples = """\
+Examples
+--------
+**Series**
+
+>>> s = pd.Series([2, np.nan, 5, -1, 0])
+>>> s
+0    2.0
+1    NaN
+2    5.0
+3   -1.0
+4    0.0
+dtype: float64
+
+By default, NA values are ignored.
+
+>>> s.cummax()
+0    2.0
+1    NaN
+2    5.0
+3    5.0
+4    5.0
+dtype: float64
+
+To include NA values in the operation, use ``skipna=False``
+
+>>> s.cummax(skipna=False)
+0    2.0
+1    NaN
+2    NaN
+3    NaN
+4    NaN
+dtype: float64
+
+**DataFrame**
+
+>>> df = pd.DataFrame([[2.0, 1.0],
+...                    [3.0, np.nan],
+...                    [1.0, 0.0]],
+...                   columns=list('AB'))
+>>> df
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  1.0  0.0
+
+By default, iterates over rows and finds the maximum
+in each column. This is equivalent to ``axis=None`` or ``axis='index'``.
+
+>>> df.cummax()
+     A    B
+0  2.0  1.0
+1  3.0  NaN
+2  3.0  1.0
+
+To iterate over columns and find the maximum in each row,
+use ``axis=1``
+
+>>> df.cummax(axis=1)
+     A    B
+0  2.0  2.0
+1  3.0  NaN
+2  1.0  1.0
+"""
+
+_any_see_also = """\
+See Also
+--------
+numpy.any : Numpy version of this method.
+Series.any : Return whether any element is True.
+Series.all : Return whether all elements are True.
+DataFrame.any : Return whether any element is True over requested axis.
+DataFrame.all : Return whether all elements are True over requested axis.
+"""
+
+_any_desc = """\
+Return whether any element is True, potentially over an axis.
+
+Returns False unless there is at least one element within a series or
+along a Dataframe axis that is True or equivalent (e.g. non-zero or
+non-empty)."""
+
+_any_examples = """\
+Examples
+--------
+**Series**
+
+For Series input, the output is a scalar indicating whether any element
+is True.
+
+>>> pd.Series([False, False]).any()
+False
+>>> pd.Series([True, False]).any()
+True
+>>> pd.Series([], dtype="float64").any()
+False
+>>> pd.Series([np.nan]).any()
+False
+>>> pd.Series([np.nan]).any(skipna=False)
+True
+
+**DataFrame**
+
+Whether each column contains at least one True element (the default).
+
+>>> df = pd.DataFrame({"A": [1, 2], "B": [0, 2], "C": [0, 0]})
+>>> df
+   A  B  C
+0  1  0  0
+1  2  2  0
+
+>>> df.any()
+A     True
+B     True
+C    False
+dtype: bool
+
+Aggregating over the columns.
+
+>>> df = pd.DataFrame({"A": [True, False], "B": [1, 2]})
+>>> df
+       A  B
+0   True  1
+1  False  2
+
+>>> df.any(axis='columns')
+0    True
+1    True
+dtype: bool
+
+>>> df = pd.DataFrame({"A": [True, False], "B": [1, 0]})
+>>> df
+       A  B
+0   True  1
+1  False  0
+
+>>> df.any(axis='columns')
+0    True
+1    False
+dtype: bool
+
+Aggregating over the entire DataFrame with ``axis=None``.
+
+>>> df.any(axis=None)
+True
+
+`any` for an empty DataFrame is an empty Series.
+
+>>> pd.DataFrame([]).any()
+Series([], dtype: bool)
+"""
+
+_shared_docs["stat_func_example"] = """
+
+Examples
+--------
+>>> idx = pd.MultiIndex.from_arrays([
+...     ['warm', 'warm', 'cold', 'cold'],
+...     ['dog', 'falcon', 'fish', 'spider']],
+...     names=['blooded', 'animal'])
+>>> s = pd.Series([4, 2, 0, 8], name='legs', index=idx)
+>>> s
+blooded  animal
+warm     dog       4
+         falcon    2
+cold     fish      0
+         spider    8
+Name: legs, dtype: int64
+
+>>> s.{stat_func}()
+{default_output}"""
+
+_sum_examples = _shared_docs["stat_func_example"].format(
+    stat_func="sum", verb="Sum", default_output=14, level_output_0=6, level_output_1=8
+)
+
+_sum_examples += """
+
+By default, the sum of an empty or all-NA Series is ``0``.
+
+>>> pd.Series([], dtype="float64").sum()  # min_count=0 is the default
+0.0
+
+This can be controlled with the ``min_count`` parameter. For example, if
+you'd like the sum of an empty series to be NaN, pass ``min_count=1``.
+
+>>> pd.Series([], dtype="float64").sum(min_count=1)
+nan
+
+Thanks to the ``skipna`` parameter, ``min_count`` handles all-NA and
+empty series identically.
+
+>>> pd.Series([np.nan]).sum()
+0.0
+
+>>> pd.Series([np.nan]).sum(min_count=1)
+nan"""
+
+_max_examples: str = _shared_docs["stat_func_example"].format(
+    stat_func="max", verb="Max", default_output=8, level_output_0=4, level_output_1=8
+)
+
+_min_examples: str = _shared_docs["stat_func_example"].format(
+    stat_func="min", verb="Min", default_output=0, level_output_0=2, level_output_1=0
+)
+
+_skew_see_also = """
+
+See Also
+--------
+Series.skew : Return unbiased skew over requested axis.
+Series.var : Return unbiased variance over requested axis.
+Series.std : Return unbiased standard deviation over requested axis."""
+
+_stat_func_see_also = """
+
+See Also
+--------
+Series.sum : Return the sum.
+Series.min : Return the minimum.
+Series.max : Return the maximum.
+Series.idxmin : Return the index of the minimum.
+Series.idxmax : Return the index of the maximum.
+DataFrame.sum : Return the sum over the requested axis.
+DataFrame.min : Return the minimum over the requested axis.
+DataFrame.max : Return the maximum over the requested axis.
+DataFrame.idxmin : Return the index of the minimum over the requested axis.
+DataFrame.idxmax : Return the index of the maximum over the requested axis."""
+
+_prod_examples = """
+
+Examples
+--------
+By default, the product of an empty or all-NA Series is ``1``
+
+>>> pd.Series([], dtype="float64").prod()
+1.0
+
+This can be controlled with the ``min_count`` parameter
+
+>>> pd.Series([], dtype="float64").prod(min_count=1)
+nan
+
+Thanks to the ``skipna`` parameter, ``min_count`` handles all-NA and
+empty series identically.
+
+>>> pd.Series([np.nan]).prod()
+1.0
+
+>>> pd.Series([np.nan]).prod(min_count=1)
+nan"""
+
+_min_count_stub = """\
+min_count : int, default 0
+    The required number of valid values to perform the operation. If fewer than
+    ``min_count`` non-NA values are present the result will be NA.
+"""
+
+
+def make_doc(name: str, ndim: int) -> str:
+    """
+    Generate the docstring for a Series/DataFrame reduction.
+    """
+    if ndim == 1:
+        name1 = "scalar"
+        name2 = "Series"
+        axis_descr = "{index (0)}"
+    else:
+        name1 = "Series"
+        name2 = "DataFrame"
+        axis_descr = "{index (0), columns (1)}"
+
+    if name == "any":
+        base_doc = _bool_doc
+        desc = _any_desc
+        see_also = _any_see_also
+        examples = _any_examples
+        kwargs = {"empty_value": "False"}
+    elif name == "all":
+        base_doc = _bool_doc
+        desc = _all_desc
+        see_also = _all_see_also
+        examples = _all_examples
+        kwargs = {"empty_value": "True"}
+    elif name == "min":
+        base_doc = _num_doc
+        desc = (
+            "Return the minimum of the values over the requested axis.\n\n"
+            "If you want the *index* of the minimum, use ``idxmin``. This is "
+            "the equivalent of the ``numpy.ndarray`` method ``argmin``."
+        )
+        see_also = _stat_func_see_also
+        examples = _min_examples
+        kwargs = {"min_count": ""}
+    elif name == "max":
+        base_doc = _num_doc
+        desc = (
+            "Return the maximum of the values over the requested axis.\n\n"
+            "If you want the *index* of the maximum, use ``idxmax``. This is "
+            "the equivalent of the ``numpy.ndarray`` method ``argmax``."
+        )
+        see_also = _stat_func_see_also
+        examples = _max_examples
+        kwargs = {"min_count": ""}
+
+    elif name == "sum":
+        base_doc = _sum_prod_doc
+        desc = (
+            "Return the sum of the values over the requested axis.\n\n"
+            "This is equivalent to the method ``numpy.sum``."
+        )
+        see_also = _stat_func_see_also
+        examples = _sum_examples
+        kwargs = {"min_count": _min_count_stub}
+
+    elif name == "prod":
+        base_doc = _sum_prod_doc
+        desc = "Return the product of the values over the requested axis."
+        see_also = _stat_func_see_also
+        examples = _prod_examples
+        kwargs = {"min_count": _min_count_stub}
+
+    elif name == "median":
+        base_doc = _num_doc
+        desc = "Return the median of the values over the requested axis."
+        see_also = _stat_func_see_also
+        examples = """
+
+            Examples
+            --------
+            >>> s = pd.Series([1, 2, 3])
+            >>> s.median()
+            2.0
+
+            With a DataFrame
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': [2, 3]}, index=['tiger', 'zebra'])
+            >>> df
+                   a   b
+            tiger  1   2
+            zebra  2   3
+            >>> df.median()
+            a   1.5
+            b   2.5
+            dtype: float64
+
+            Using axis=1
+
+            >>> df.median(axis=1)
+            tiger   1.5
+            zebra   2.5
+            dtype: float64
+
+            In this case, `numeric_only` should be set to `True`
+            to avoid getting an error.
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': ['T', 'Z']},
+            ...                   index=['tiger', 'zebra'])
+            >>> df.median(numeric_only=True)
+            a   1.5
+            dtype: float64"""
+        kwargs = {"min_count": ""}
+
+    elif name == "mean":
+        base_doc = _num_doc
+        desc = "Return the mean of the values over the requested axis."
+        see_also = _stat_func_see_also
+        examples = """
+
+            Examples
+            --------
+            >>> s = pd.Series([1, 2, 3])
+            >>> s.mean()
+            2.0
+
+            With a DataFrame
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': [2, 3]}, index=['tiger', 'zebra'])
+            >>> df
+                   a   b
+            tiger  1   2
+            zebra  2   3
+            >>> df.mean()
+            a   1.5
+            b   2.5
+            dtype: float64
+
+            Using axis=1
+
+            >>> df.mean(axis=1)
+            tiger   1.5
+            zebra   2.5
+            dtype: float64
+
+            In this case, `numeric_only` should be set to `True` to avoid
+            getting an error.
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': ['T', 'Z']},
+            ...                   index=['tiger', 'zebra'])
+            >>> df.mean(numeric_only=True)
+            a   1.5
+            dtype: float64"""
+        kwargs = {"min_count": ""}
+
+    elif name == "var":
+        base_doc = _num_ddof_doc
+        desc = (
+            "Return unbiased variance over requested axis.\n\nNormalized by "
+            "N-1 by default. This can be changed using the ddof argument."
+        )
+        examples = _var_examples
+        see_also = ""
+        kwargs = {"notes": ""}
+
+    elif name == "std":
+        base_doc = _num_ddof_doc
+        desc = (
+            "Return sample standard deviation over requested axis."
+            "\n\nNormalized by N-1 by default. This can be changed using the "
+            "ddof argument."
+        )
+        examples = _std_examples
+        see_also = _std_see_also.format(name2=name2)
+        kwargs = {"notes": "", "return_desc": _std_return_desc}
+
+    elif name == "sem":
+        base_doc = _num_ddof_doc
+        desc = (
+            "Return unbiased standard error of the mean over requested "
+            "axis.\n\nNormalized by N-1 by default. This can be changed "
+            "using the ddof argument"
+        )
+        examples = """
+
+            Examples
+            --------
+            >>> s = pd.Series([1, 2, 3])
+            >>> s.sem().round(6)
+            0.57735
+
+            With a DataFrame
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': [2, 3]}, index=['tiger', 'zebra'])
+            >>> df
+                   a   b
+            tiger  1   2
+            zebra  2   3
+            >>> df.sem()
+            a   0.5
+            b   0.5
+            dtype: float64
+
+            Using axis=1
+
+            >>> df.sem(axis=1)
+            tiger   0.5
+            zebra   0.5
+            dtype: float64
+
+            In this case, `numeric_only` should be set to `True`
+            to avoid getting an error.
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': ['T', 'Z']},
+            ...                   index=['tiger', 'zebra'])
+            >>> df.sem(numeric_only=True)
+            a   0.5
+            dtype: float64"""
+        see_also = _sem_see_also.format(name2=name2)
+        kwargs = {"notes": "", "return_desc": _sem_return_desc}
+
+    elif name == "skew":
+        base_doc = _num_doc
+        desc = "Return unbiased skew over requested axis.\n\nNormalized by N-1."
+        see_also = _skew_see_also
+        examples = """
+
+            Examples
+            --------
+            >>> s = pd.Series([1, 2, 3])
+            >>> s.skew()
+            0.0
+
+            With a DataFrame
+
+            >>> df = pd.DataFrame({'a': [1, 2, 3], 'b': [2, 3, 4], 'c': [1, 3, 5]},
+            ...                   index=['tiger', 'zebra', 'cow'])
+            >>> df
+                    a   b   c
+            tiger   1   2   1
+            zebra   2   3   3
+            cow     3   4   5
+            >>> df.skew()
+            a   0.0
+            b   0.0
+            c   0.0
+            dtype: float64
+
+            Using axis=1
+
+            >>> df.skew(axis=1)
+            tiger   1.732051
+            zebra  -1.732051
+            cow     0.000000
+            dtype: float64
+
+            In this case, `numeric_only` should be set to `True` to avoid
+            getting an error.
+
+            >>> df = pd.DataFrame({'a': [1, 2, 3], 'b': ['T', 'Z', 'X']},
+            ...                   index=['tiger', 'zebra', 'cow'])
+            >>> df.skew(numeric_only=True)
+            a   0.0
+            dtype: float64"""
+        kwargs = {"min_count": ""}
+
+    elif name == "kurt":
+        base_doc = _num_doc
+        desc = (
+            "Return unbiased kurtosis over requested axis.\n\n"
+            "Kurtosis obtained using Fisher's definition of\n"
+            "kurtosis (kurtosis of normal == 0.0). Normalized "
+            "by N-1."
+        )
+        see_also = ""
+        examples = """
+
+            Examples
+            --------
+            >>> s = pd.Series([1, 2, 2, 3], index=['cat', 'dog', 'dog', 'mouse'])
+            >>> s
+            cat    1
+            dog    2
+            dog    2
+            mouse  3
+            dtype: int64
+            >>> s.kurt()
+            1.5
+
+            With a DataFrame
+
+            >>> df = pd.DataFrame({'a': [1, 2, 2, 3], 'b': [3, 4, 4, 4]},
+            ...                   index=['cat', 'dog', 'dog', 'mouse'])
+            >>> df
+                   a   b
+              cat  1   3
+              dog  2   4
+              dog  2   4
+            mouse  3   4
+            >>> df.kurt()
+            a   1.5
+            b   4.0
+            dtype: float64
+
+            With axis=None
+
+            >>> df.kurt(axis=None).round(6)
+            -0.988693
+
+            Using axis=1
+
+            >>> df = pd.DataFrame({'a': [1, 2], 'b': [3, 4], 'c': [3, 4], 'd': [1, 2]},
+            ...                   index=['cat', 'dog'])
+            >>> df.kurt(axis=1)
+            cat   -6.0
+            dog   -6.0
+            dtype: float64"""
+        kwargs = {"min_count": ""}
+
+    elif name == "cumsum":
+        if ndim == 1:
+            base_doc = _cnum_series_doc
+        else:
+            base_doc = _cnum_pd_doc
+
+        desc = "sum"
+        see_also = ""
+        examples = _cumsum_examples
+        kwargs = {"accum_func_name": "sum"}
+
+    elif name == "cumprod":
+        if ndim == 1:
+            base_doc = _cnum_series_doc
+        else:
+            base_doc = _cnum_pd_doc
+
+        desc = "product"
+        see_also = ""
+        examples = _cumprod_examples
+        kwargs = {"accum_func_name": "prod"}
+
+    elif name == "cummin":
+        if ndim == 1:
+            base_doc = _cnum_series_doc
+        else:
+            base_doc = _cnum_pd_doc
+
+        desc = "minimum"
+        see_also = ""
+        examples = _cummin_examples
+        kwargs = {"accum_func_name": "min"}
+
+    elif name == "cummax":
+        if ndim == 1:
+            base_doc = _cnum_series_doc
+        else:
+            base_doc = _cnum_pd_doc
+
+        desc = "maximum"
+        see_also = ""
+        examples = _cummax_examples
+        kwargs = {"accum_func_name": "max"}
+
+    else:
+        raise NotImplementedError
+
+    docstr = base_doc.format(
+        desc=desc,
+        name=name,
+        name1=name1,
+        name2=name2,
+        axis_descr=axis_descr,
+        see_also=see_also,
+        examples=examples,
+        **kwargs,
+    )
+    return docstr
diff --git a/pandas/pandas/core/groupby/__init__.py b/pandas/pandas/core/groupby/__init__.py
new file mode 100644
index 0000000000000..ec477626a098f
--- /dev/null
+++ b/pandas/pandas/core/groupby/__init__.py
@@ -0,0 +1,15 @@
+from pandas.core.groupby.generic import (
+    DataFrameGroupBy,
+    NamedAgg,
+    SeriesGroupBy,
+)
+from pandas.core.groupby.groupby import GroupBy
+from pandas.core.groupby.grouper import Grouper
+
+__all__ = [
+    "DataFrameGroupBy",
+    "GroupBy",
+    "Grouper",
+    "NamedAgg",
+    "SeriesGroupBy",
+]
diff --git a/pandas/pandas/core/groupby/base.py b/pandas/pandas/core/groupby/base.py
new file mode 100644
index 0000000000000..7699fb3d0f864
--- /dev/null
+++ b/pandas/pandas/core/groupby/base.py
@@ -0,0 +1,121 @@
+"""
+Provide basic components for groupby.
+"""
+
+from __future__ import annotations
+
+import dataclasses
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+
+@dataclasses.dataclass(order=True, frozen=True)
+class OutputKey:
+    label: Hashable
+    position: int
+
+
+# special case to prevent duplicate plots when catching exceptions when
+# forwarding methods from NDFrames
+plotting_methods = frozenset(["plot", "hist"])
+
+# cythonized transformations or canned "agg+broadcast", which do not
+# require postprocessing of the result by transform.
+cythonized_kernels = frozenset(["cumprod", "cumsum", "shift", "cummin", "cummax"])
+
+# List of aggregation/reduction functions.
+# These map each group to a single numeric value
+reduction_kernels = frozenset(
+    [
+        "all",
+        "any",
+        "corrwith",
+        "count",
+        "first",
+        "idxmax",
+        "idxmin",
+        "last",
+        "max",
+        "mean",
+        "median",
+        "min",
+        "nunique",
+        "prod",
+        # as long as `quantile`'s signature accepts only
+        # a single quantile value, it's a reduction.
+        # GH#27526 might change that.
+        "quantile",
+        "sem",
+        "size",
+        "skew",
+        "kurt",
+        "std",
+        "sum",
+        "var",
+    ]
+)
+
+# List of transformation functions.
+# a transformation is a function that, for each group,
+# produces a result that has the same shape as the group.
+
+
+transformation_kernels = frozenset(
+    [
+        "bfill",
+        "cumcount",
+        "cummax",
+        "cummin",
+        "cumprod",
+        "cumsum",
+        "diff",
+        "ffill",
+        "ngroup",
+        "pct_change",
+        "rank",
+        "shift",
+    ]
+)
+
+# these are all the public methods on Grouper which don't belong
+# in either of the above lists
+groupby_other_methods = frozenset(
+    [
+        "agg",
+        "aggregate",
+        "apply",
+        "boxplot",
+        # corr and cov return ngroups*ncolumns rows, so they
+        # are neither a transformation nor a reduction
+        "corr",
+        "cov",
+        "describe",
+        "expanding",
+        "ewm",
+        "filter",
+        "get_group",
+        "groups",
+        "head",
+        "hist",
+        "indices",
+        "ndim",
+        "ngroups",
+        "nth",
+        "ohlc",
+        "pipe",
+        "plot",
+        "resample",
+        "rolling",
+        "tail",
+        "take",
+        "transform",
+        "sample",
+        "value_counts",
+    ]
+)
+# Valid values  of `name` for `groupby.transform(name)`
+# NOTE: do NOT edit this directly. New additions should be inserted
+# into the appropriate list above.
+transform_kernel_allowlist = reduction_kernels | transformation_kernels
diff --git a/pandas/pandas/core/groupby/categorical.py b/pandas/pandas/core/groupby/categorical.py
new file mode 100644
index 0000000000000..90cd8e3ffa1c7
--- /dev/null
+++ b/pandas/pandas/core/groupby/categorical.py
@@ -0,0 +1,83 @@
+from __future__ import annotations
+
+import numpy as np
+
+from pandas.core.algorithms import unique1d
+from pandas.core.arrays.categorical import (
+    Categorical,
+    CategoricalDtype,
+    recode_for_categories,
+)
+
+
+def recode_for_groupby(c: Categorical, sort: bool, observed: bool) -> Categorical:
+    """
+    Code the categories to ensure we can groupby for categoricals.
+
+    If observed=True, we return a new Categorical with the observed
+    categories only.
+
+    If sort=False, return a copy of self, coded with categories as
+    returned by .unique(), followed by any categories not appearing in
+    the data. If sort=True, return self.
+
+    This method is needed solely to ensure the categorical index of the
+    GroupBy result has categories in the order of appearance in the data
+    (GH-8868).
+
+    Parameters
+    ----------
+    c : Categorical
+    sort : bool
+        The value of the sort parameter groupby was called with.
+    observed : bool
+        Account only for the observed values
+
+    Returns
+    -------
+    Categorical
+        If sort=False, the new categories are set to the order of
+        appearance in codes (unless ordered=True, in which case the
+        original order is preserved), followed by any unrepresented
+        categories in the original order.
+    """
+    # we only care about observed values
+    if observed:
+        # In cases with c.ordered, this is equivalent to
+        #  return c.remove_unused_categories(), c
+
+        take_codes = unique1d(c.codes[c.codes != -1])  # type: ignore[no-untyped-call]
+
+        if sort:
+            take_codes = np.sort(take_codes)
+
+        # we recode according to the uniques
+        categories = c.categories.take(take_codes)
+        codes = recode_for_categories(c.codes, c.categories, categories)
+
+        # return a new categorical that maps our new codes
+        # and categories
+        dtype = CategoricalDtype(categories, ordered=c.ordered)
+        return Categorical._simple_new(codes, dtype=dtype)
+
+    # Already sorted according to c.categories; all is fine
+    if sort:
+        return c
+
+    # sort=False should order groups in as-encountered order (GH-8868)
+
+    # GH:46909: Re-ordering codes faster than using (set|add|reorder)_categories
+    # GH 38140: exclude nan from indexer for categories
+    unique_notnan_codes = unique1d(c.codes[c.codes != -1])  # type: ignore[no-untyped-call]
+    if sort:
+        unique_notnan_codes = np.sort(unique_notnan_codes)
+    if (num_cat := len(c.categories)) > len(unique_notnan_codes):
+        # GH 13179: All categories need to be present, even if missing from the data
+        missing_codes = np.setdiff1d(
+            np.arange(num_cat), unique_notnan_codes, assume_unique=True
+        )
+        take_codes = np.concatenate((unique_notnan_codes, missing_codes))
+    else:
+        take_codes = unique_notnan_codes
+
+    return Categorical(c, c.categories.take(take_codes))
diff --git a/pandas/pandas/core/groupby/generic.py b/pandas/pandas/core/groupby/generic.py
new file mode 100644
index 0000000000000..1251403db6ff3
--- /dev/null
+++ b/pandas/pandas/core/groupby/generic.py
@@ -0,0 +1,3370 @@
+"""
+Define the SeriesGroupBy and DataFrameGroupBy
+classes that hold the groupby interfaces (and some implementations).
+
+These are user facing as the result of the ``df.groupby(...)`` operations,
+which here returns a DataFrameGroupBy object.
+"""
+
+from __future__ import annotations
+
+from collections import abc
+from collections.abc import Callable
+from functools import partial
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    NamedTuple,
+    TypeVar,
+    Union,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import Interval
+from pandas._libs.hashtable import duplicated
+from pandas.errors import SpecificationError
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    doc,
+    set_module,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    is_bool,
+    is_dict_like,
+    is_integer_dtype,
+    is_list_like,
+    is_numeric_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    IntervalDtype,
+)
+from pandas.core.dtypes.inference import is_hashable
+from pandas.core.dtypes.missing import (
+    isna,
+    notna,
+)
+
+from pandas.core import algorithms
+from pandas.core.apply import (
+    GroupByApply,
+    maybe_mangle_lambdas,
+    reconstruct_func,
+    validate_func_kwargs,
+)
+import pandas.core.common as com
+from pandas.core.frame import DataFrame
+from pandas.core.groupby import base
+from pandas.core.groupby.groupby import (
+    GroupBy,
+    GroupByPlot,
+    _transform_template,
+)
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    all_indexes_same,
+    default_index,
+)
+from pandas.core.series import Series
+from pandas.core.sorting import get_group_index
+from pandas.core.util.numba_ import maybe_use_numba
+
+from pandas.plotting import boxplot_frame_groupby
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        BlockManager,
+        CorrelationMethod,
+        IndexLabel,
+        Manager,
+        SingleBlockManager,
+        TakeIndexer,
+    )
+
+    from pandas import Categorical
+    from pandas.core.generic import NDFrame
+
+# TODO(typing) the return value on this callable should be any *scalar*.
+AggScalar = Union[str, Callable[..., Any]]
+# TODO: validate types on ScalarResult and move to _typing
+# Blocked from using by https://github.com/python/mypy/issues/1484
+# See note at _mangle_lambda_list
+ScalarResult = TypeVar("ScalarResult")
+
+
+@set_module("pandas")
+class NamedAgg(NamedTuple):
+    """
+    Helper for column specific aggregation with control over output column names.
+
+    Subclass of typing.NamedTuple.
+
+    Parameters
+    ----------
+    column : Hashable
+        Column label in the DataFrame to apply aggfunc.
+    aggfunc : function or str
+        Function to apply to the provided column. If string, the name of a built-in
+        pandas function.
+
+    See Also
+    --------
+    DataFrame.groupby : Group DataFrame using a mapper or by a Series of columns.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"key": [1, 1, 2], "a": [-1, 0, 1], 1: [10, 11, 12]})
+    >>> agg_a = pd.NamedAgg(column="a", aggfunc="min")
+    >>> agg_1 = pd.NamedAgg(column=1, aggfunc=lambda x: np.mean(x))
+    >>> df.groupby("key").agg(result_a=agg_a, result_1=agg_1)
+         result_a  result_1
+    key
+    1          -1      10.5
+    2           1      12.0
+    """
+
+    column: Hashable
+    aggfunc: AggScalar
+
+
+@set_module("pandas.api.typing")
+class SeriesGroupBy(GroupBy[Series]):
+    def _wrap_agged_manager(self, mgr: Manager) -> Series:
+        out = self.obj._constructor_from_mgr(mgr, axes=mgr.axes)
+        out._name = self.obj.name
+        return out
+
+    def _get_data_to_aggregate(
+        self, *, numeric_only: bool = False, name: str | None = None
+    ) -> SingleBlockManager:
+        ser = self._obj_with_exclusions
+        single = ser._mgr
+        if numeric_only and not is_numeric_dtype(ser.dtype):
+            # GH#41291 match Series behavior
+            kwd_name = "numeric_only"
+            raise TypeError(
+                f"Cannot use {kwd_name}=True with "
+                f"{type(self).__name__}.{name} and non-numeric dtypes."
+            )
+        return single
+
+    _agg_examples_doc = dedent(
+        """
+    Examples
+    --------
+    >>> s = pd.Series([1, 2, 3, 4])
+
+    >>> s
+    0    1
+    1    2
+    2    3
+    3    4
+    dtype: int64
+
+    >>> s.groupby([1, 1, 2, 2]).min()
+    1    1
+    2    3
+    dtype: int64
+
+    >>> s.groupby([1, 1, 2, 2]).agg('min')
+    1    1
+    2    3
+    dtype: int64
+
+    >>> s.groupby([1, 1, 2, 2]).agg(['min', 'max'])
+       min  max
+    1    1    2
+    2    3    4
+
+    The output column names can be controlled by passing
+    the desired column names and aggregations as keyword arguments.
+
+    >>> s.groupby([1, 1, 2, 2]).agg(
+    ...     minimum='min',
+    ...     maximum='max',
+    ... )
+       minimum  maximum
+    1        1        2
+    2        3        4
+
+    .. versionchanged:: 1.3.0
+
+        The resulting dtype will reflect the return value of the aggregating function.
+
+    >>> s.groupby([1, 1, 2, 2]).agg(lambda x: x.astype(float).min())
+    1    1.0
+    2    3.0
+    dtype: float64
+    """
+    )
+
+    def apply(self, func, *args, **kwargs) -> Series:
+        """
+        Apply function ``func`` group-wise and combine the results together.
+
+        The function passed to ``apply`` must take a series as its first
+        argument and return a DataFrame, Series or scalar. ``apply`` will
+        then take care of combining the results back together into a single
+        dataframe or series. ``apply`` is therefore a highly flexible
+        grouping method.
+
+        While ``apply`` is a very flexible method, its downside is that
+        using it can be quite a bit slower than using more specific methods
+        like ``agg`` or ``transform``. Pandas offers a wide range of method that will
+        be much faster than using ``apply`` for their specific purposes, so try to
+        use them before reaching for ``apply``.
+
+        Parameters
+        ----------
+        func : callable
+            A callable that takes a series as its first argument, and
+            returns a dataframe, a series or a scalar. In addition the
+            callable may take positional and keyword arguments.
+
+        *args : tuple
+            Optional positional arguments to pass to ``func``.
+
+        **kwargs : dict
+            Optional keyword arguments to pass to ``func``.
+
+        Returns
+        -------
+        Series or DataFrame
+            A pandas object with the result of applying ``func`` to each group.
+
+        See Also
+        --------
+        pipe : Apply function to the full GroupBy object instead of to each
+            group.
+        aggregate : Apply aggregate function to the GroupBy object.
+        transform : Apply function column-by-column to the GroupBy object.
+        Series.apply : Apply a function to a Series.
+        DataFrame.apply : Apply a function to each row or column of a DataFrame.
+
+        Notes
+        -----
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``,
+            see the examples below.
+
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        >>> s = pd.Series([0, 1, 2], index="a a b".split())
+        >>> g1 = s.groupby(s.index, group_keys=False)
+        >>> g2 = s.groupby(s.index, group_keys=True)
+
+        From ``s`` above we can see that ``g`` has two groups, ``a`` and ``b``.
+        Notice that ``g1`` have ``g2`` have two groups, ``a`` and ``b``, and only
+        differ in their ``group_keys`` argument. Calling `apply` in various ways,
+        we can get different grouping results:
+
+        Example 1: The function passed to `apply` takes a Series as
+        its argument and returns a Series.  `apply` combines the result for
+        each group together into a new Series.
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``.
+
+        >>> g1.apply(lambda x: x * 2 if x.name == "a" else x / 2)
+        a    0.0
+        a    2.0
+        b    1.0
+        dtype: float64
+
+        In the above, the groups are not part of the index. We can have them included
+        by using ``g2`` where ``group_keys=True``:
+
+        >>> g2.apply(lambda x: x * 2 if x.name == "a" else x / 2)
+        a  a    0.0
+           a    2.0
+        b  b    1.0
+        dtype: float64
+
+        Example 2: The function passed to `apply` takes a Series as
+        its argument and returns a scalar. `apply` combines the result for
+        each group together into a Series, including setting the index as
+        appropriate:
+
+        >>> g1.apply(lambda x: x.max() - x.min())
+        a    1
+        b    0
+        dtype: int64
+
+        The ``group_keys`` argument has no effect here because the result is not
+        like-indexed (i.e. :ref:`a transform <groupby.transform>`) when compared
+        to the input.
+
+        >>> g2.apply(lambda x: x.max() - x.min())
+        a    1
+        b    0
+        dtype: int64
+        """
+        return super().apply(func, *args, **kwargs)
+
+    def aggregate(self, func=None, *args, engine=None, engine_kwargs=None, **kwargs):
+        """
+        Aggregate using one or more operations.
+
+        The ``aggregate`` method enables flexible and efficient aggregation of grouped
+        data using a variety of functions, including built-in, user-defined, and
+        optimized JIT-compiled functions.
+
+        Parameters
+        ----------
+        func : function, str, list, dict or None
+            Function to use for aggregating the data. If a function, must either
+            work when passed a Series or when passed to Series.apply.
+
+            Accepted combinations are:
+
+            - function
+            - string function name
+            - list of functions and/or function names, e.g. ``[np.sum, 'mean']``
+            - None, in which case ``**kwargs`` are used with Named Aggregation. Here
+              the output has one column for each element in ``**kwargs``. The name of
+              the column is keyword, whereas the value determines the aggregation
+              used to compute the values in the column.
+
+              Can also accept a Numba JIT function with
+              ``engine='numba'`` specified. Only passing a single function is supported
+              with this engine.
+
+              If the ``'numba'`` engine is chosen, the function must be
+              a user defined function with ``values`` and ``index`` as the
+              first and second arguments respectively in the function signature.
+              Each group's index will be passed to the user defined function
+              and optionally available for use.
+
+            .. deprecated:: 2.1.0
+
+                Passing a dictionary is deprecated and will raise in a future version
+                of pandas. Pass a list of aggregations instead.
+        *args
+            Positional arguments to pass to func.
+        engine : str, default None
+            * ``'cython'`` : Runs the function through C-extensions from cython.
+            * ``'numba'`` : Runs the function through JIT compiled code from numba.
+            * ``None`` : Defaults to ``'cython'`` or globally setting
+                ``compute.use_numba``
+
+        engine_kwargs : dict, default None
+            * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{'nopython': True, 'nogil': False, 'parallel': False}`` and will be
+              applied to the function
+
+        **kwargs
+            * If ``func`` is None, ``**kwargs`` are used to define the output names and
+              aggregations via Named Aggregation. See ``func`` entry.
+            * Otherwise, keyword arguments to be passed into func.
+
+        Returns
+        -------
+        Series
+            Aggregated Series based on the grouping and the applied aggregation
+            functions.
+
+        See Also
+        --------
+        SeriesGroupBy.apply : Apply function func group-wise
+            and combine the results together.
+        SeriesGroupBy.transform : Transforms the Series on each group
+            based on the given function.
+        Series.aggregate : Aggregate using one or more operations.
+
+        Notes
+        -----
+        When using ``engine='numba'``, there will be no "fall back" behavior internally.
+        The group data and group index will be passed as numpy arrays to the JITed
+        user defined function, and no alternative execution attempts will be tried.
+
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``,
+            see the examples below.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, 4])
+
+        >>> s
+        0    1
+        1    2
+        2    3
+        3    4
+        dtype: int64
+
+        >>> s.groupby([1, 1, 2, 2]).min()
+        1    1
+        2    3
+        dtype: int64
+
+        >>> s.groupby([1, 1, 2, 2]).agg("min")
+        1    1
+        2    3
+        dtype: int64
+
+        >>> s.groupby([1, 1, 2, 2]).agg(["min", "max"])
+           min  max
+        1    1    2
+        2    3    4
+
+        The output column names can be controlled by passing
+        the desired column names and aggregations as keyword arguments.
+
+        >>> s.groupby([1, 1, 2, 2]).agg(
+        ...     minimum="min",
+        ...     maximum="max",
+        ... )
+           minimum  maximum
+        1        1        2
+        2        3        4
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the aggregating
+            function.
+
+        >>> s.groupby([1, 1, 2, 2]).agg(lambda x: x.astype(float).min())
+        1    1.0
+        2    3.0
+        dtype: float64
+        """
+        relabeling = func is None
+        columns = None
+        if relabeling:
+            columns, func = validate_func_kwargs(kwargs)
+            kwargs = {}
+
+        if isinstance(func, str):
+            if maybe_use_numba(engine) and engine is not None:
+                # Not all agg functions support numba, only propagate numba kwargs
+                # if user asks for numba, and engine is not None
+                # (if engine is None, the called function will handle the case where
+                # numba is requested via the global option)
+                kwargs["engine"] = engine
+            if engine_kwargs is not None:
+                kwargs["engine_kwargs"] = engine_kwargs
+            return getattr(self, func)(*args, **kwargs)
+
+        elif isinstance(func, abc.Iterable):
+            # Catch instances of lists / tuples
+            # but not the class list / tuple itself.
+            func = maybe_mangle_lambdas(func)
+            kwargs["engine"] = engine
+            kwargs["engine_kwargs"] = engine_kwargs
+            ret = self._aggregate_multiple_funcs(func, *args, **kwargs)
+            if relabeling:
+                # columns is not narrowed by mypy from relabeling flag
+                assert columns is not None  # for mypy
+                ret.columns = columns
+            if not self.as_index:
+                ret = ret.reset_index()
+            return ret
+
+        else:
+            if maybe_use_numba(engine):
+                return self._aggregate_with_numba(
+                    func, *args, engine_kwargs=engine_kwargs, **kwargs
+                )
+
+            if self.ngroups == 0:
+                # e.g. test_evaluate_with_empty_groups without any groups to
+                #  iterate over, we have no output on which to do dtype
+                #  inference. We default to using the existing dtype.
+                #  xref GH#51445
+                obj = self._obj_with_exclusions
+                return self.obj._constructor(
+                    [],
+                    name=self.obj.name,
+                    index=self._grouper.result_index,
+                    dtype=obj.dtype,
+                )
+            return self._python_agg_general(func, *args, **kwargs)
+
+    agg = aggregate
+
+    def _python_agg_general(self, func, *args, **kwargs):
+        f = lambda x: func(x, *args, **kwargs)
+
+        obj = self._obj_with_exclusions
+        result = self._grouper.agg_series(obj, f)
+        res = obj._constructor(result, name=obj.name)
+        return self._wrap_aggregated_output(res)
+
+    def _aggregate_multiple_funcs(self, arg, *args, **kwargs) -> DataFrame:
+        if isinstance(arg, dict):
+            raise SpecificationError("nested renamer is not supported")
+
+        if any(isinstance(x, (tuple, list)) for x in arg):
+            arg = ((x, x) if not isinstance(x, (tuple, list)) else x for x in arg)
+        else:
+            # list of functions / function names
+            columns = (com.get_callable_name(f) or f for f in arg)
+            arg = zip(columns, arg)
+
+        results: dict[base.OutputKey, DataFrame | Series] = {}
+        with com.temp_setattr(self, "as_index", True):
+            # Combine results using the index, need to adjust index after
+            # if as_index=False (GH#50724)
+            for idx, (name, func) in enumerate(arg):
+                key = base.OutputKey(label=name, position=idx)
+                results[key] = self.aggregate(func, *args, **kwargs)
+
+        if any(isinstance(x, DataFrame) for x in results.values()):
+            from pandas import concat
+
+            res_df = concat(
+                results.values(), axis=1, keys=[key.label for key in results]
+            )
+            return res_df
+
+        indexed_output = {key.position: val for key, val in results.items()}
+        output = self.obj._constructor_expanddim(indexed_output, index=None)
+        output.columns = Index(key.label for key in results)
+
+        return output
+
+    def _wrap_applied_output(
+        self,
+        data: Series,
+        values: list[Any],
+        not_indexed_same: bool = False,
+        is_transform: bool = False,
+    ) -> DataFrame | Series:
+        """
+        Wrap the output of SeriesGroupBy.apply into the expected result.
+
+        Parameters
+        ----------
+        data : Series
+            Input data for groupby operation.
+        values : List[Any]
+            Applied output for each group.
+        not_indexed_same : bool, default False
+            Whether the applied outputs are not indexed the same as the group axes.
+
+        Returns
+        -------
+        DataFrame or Series
+        """
+        if len(values) == 0:
+            # GH #6265
+            if is_transform:
+                # GH#47787 see test_group_on_empty_multiindex
+                res_index = data.index
+            elif not self.group_keys:
+                res_index = None
+            else:
+                res_index = self._grouper.result_index
+
+            return self.obj._constructor(
+                [],
+                name=self.obj.name,
+                index=res_index,
+                dtype=data.dtype,
+            )
+        assert values is not None
+
+        if isinstance(values[0], dict):
+            # GH #823 #24880
+            index = self._grouper.result_index
+            res_df = self.obj._constructor_expanddim(values, index=index)
+            # if self.observed is False,
+            # keep all-NaN rows created while re-indexing
+            res_ser = res_df.stack()
+            res_ser.name = self.obj.name
+            return res_ser
+        elif isinstance(values[0], (Series, DataFrame)):
+            result = self._concat_objects(
+                values,
+                not_indexed_same=not_indexed_same,
+                is_transform=is_transform,
+            )
+            if isinstance(result, Series):
+                result.name = self.obj.name
+            if not self.as_index and not_indexed_same:
+                result = self._insert_inaxis_grouper(result)
+                result.index = default_index(len(result))
+            return result
+        else:
+            # GH #6265 #24880
+            result = self.obj._constructor(
+                data=values, index=self._grouper.result_index, name=self.obj.name
+            )
+            if not self.as_index:
+                result = self._insert_inaxis_grouper(result)
+                result.index = default_index(len(result))
+            return result
+
+    __examples_series_doc = dedent(
+        """
+    >>> ser = pd.Series([390.0, 350.0, 30.0, 20.0],
+    ...                 index=["Falcon", "Falcon", "Parrot", "Parrot"],
+    ...                 name="Max Speed")
+    >>> grouped = ser.groupby([1, 1, 2, 2])
+    >>> grouped.transform(lambda x: (x - x.mean()) / x.std())
+        Falcon    0.707107
+        Falcon   -0.707107
+        Parrot    0.707107
+        Parrot   -0.707107
+        Name: Max Speed, dtype: float64
+
+    Broadcast result of the transformation
+
+    >>> grouped.transform(lambda x: x.max() - x.min())
+    Falcon    40.0
+    Falcon    40.0
+    Parrot    10.0
+    Parrot    10.0
+    Name: Max Speed, dtype: float64
+
+    >>> grouped.transform("mean")
+    Falcon    370.0
+    Falcon    370.0
+    Parrot     25.0
+    Parrot     25.0
+    Name: Max Speed, dtype: float64
+
+    .. versionchanged:: 1.3.0
+
+    The resulting dtype will reflect the return value of the passed ``func``,
+    for example:
+
+    >>> grouped.transform(lambda x: x.astype(int).max())
+    Falcon    390
+    Falcon    390
+    Parrot     30
+    Parrot     30
+    Name: Max Speed, dtype: int64
+    """
+    )
+
+    @Substitution(klass="Series", example=__examples_series_doc)
+    @Appender(_transform_template)
+    def transform(self, func, *args, engine=None, engine_kwargs=None, **kwargs):
+        return self._transform(
+            func, *args, engine=engine, engine_kwargs=engine_kwargs, **kwargs
+        )
+
+    def _cython_transform(self, how: str, numeric_only: bool = False, **kwargs):
+        obj = self._obj_with_exclusions
+
+        try:
+            result = self._grouper._cython_operation(
+                "transform", obj._values, how, 0, **kwargs
+            )
+        except NotImplementedError as err:
+            # e.g. test_groupby_raises_string
+            raise TypeError(f"{how} is not supported for {obj.dtype} dtype") from err
+
+        return obj._constructor(result, index=self.obj.index, name=obj.name)
+
+    def _transform_general(
+        self, func: Callable, engine, engine_kwargs, *args, **kwargs
+    ) -> Series:
+        """
+        Transform with a callable `func`.
+        """
+        if maybe_use_numba(engine):
+            return self._transform_with_numba(
+                func, *args, engine_kwargs=engine_kwargs, **kwargs
+            )
+        assert callable(func)
+        klass = type(self.obj)
+
+        results = []
+        for name, group in self._grouper.get_iterator(
+            self._obj_with_exclusions,
+        ):
+            # this setattr is needed for test_transform_lambda_with_datetimetz
+            object.__setattr__(group, "name", name)
+            res = func(group, *args, **kwargs)
+
+            results.append(klass(res, index=group.index))
+
+        # check for empty "results" to avoid concat ValueError
+        if results:
+            from pandas.core.reshape.concat import concat
+
+            concatenated = concat(results, ignore_index=True)
+            result = self._set_result_index_ordered(concatenated)
+        else:
+            result = self.obj._constructor(dtype=np.float64)
+
+        result.name = self.obj.name
+        return result
+
+    def filter(self, func, dropna: bool = True, *args, **kwargs):
+        """
+        Filter elements from groups that don't satisfy a criterion.
+
+        Elements from groups are filtered if they do not satisfy the
+        boolean criterion specified by func.
+
+        Parameters
+        ----------
+        func : function
+            Criterion to apply to each group. Should return True or False.
+        dropna : bool, optional
+            Drop groups that do not pass the filter. True by default; if False,
+            groups that evaluate False are filled with NaNs.
+        *args : tuple
+            Optional positional arguments to pass to `func`.
+        **kwargs : dict
+            Optional keyword arguments to pass to `func`.
+
+        Returns
+        -------
+        Series
+            The filtered subset of the original Series.
+
+        See Also
+        --------
+        Series.filter: Filter elements of ungrouped Series.
+        DataFrameGroupBy.filter : Filter elements from groups base on criterion.
+
+        Notes
+        -----
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": ["foo", "bar", "foo", "bar", "foo", "bar"],
+        ...         "B": [1, 2, 3, 4, 5, 6],
+        ...         "C": [2.0, 5.0, 8.0, 1.0, 2.0, 9.0],
+        ...     }
+        ... )
+        >>> grouped = df.groupby("A")
+        >>> df.groupby("A").B.filter(lambda x: x.mean() > 3.0)
+        1    2
+        3    4
+        5    6
+        Name: B, dtype: int64
+        """
+        if isinstance(func, str):
+            wrapper = lambda x: getattr(x, func)(*args, **kwargs)
+        else:
+            wrapper = lambda x: func(x, *args, **kwargs)
+
+        # Interpret np.nan as False.
+        def true_and_notna(x) -> bool:
+            b = wrapper(x)
+            return notna(b) and b
+
+        try:
+            indices = [
+                self._get_index(name)
+                for name, group in self._grouper.get_iterator(self._obj_with_exclusions)
+                if true_and_notna(group)
+            ]
+        except (ValueError, TypeError) as err:
+            raise TypeError("the filter must return a boolean result") from err
+
+        filtered = self._apply_filter(indices, dropna)
+        return filtered
+
+    def nunique(self, dropna: bool = True) -> Series | DataFrame:
+        """
+        Return number of unique elements in the group.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include NaN in the counts.
+
+        Returns
+        -------
+        Series
+            Number of unique values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.nunique : Method nunique for Resampler.
+
+        Examples
+        --------
+        >>> lst = ["a", "a", "b", "b"]
+        >>> ser = pd.Series([1, 2, 3, 3], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    3
+        dtype: int64
+        >>> ser.groupby(level=0).nunique()
+        a    2
+        b    1
+        dtype: int64
+        """
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+        val = self.obj._values
+        codes, uniques = algorithms.factorize(val, use_na_sentinel=dropna, sort=False)
+
+        if self._grouper.has_dropped_na:
+            mask = ids >= 0
+            ids = ids[mask]
+            codes = codes[mask]
+
+        group_index = get_group_index(
+            labels=[ids, codes],
+            shape=(ngroups, len(uniques)),
+            sort=False,
+            xnull=dropna,
+        )
+
+        if dropna:
+            mask = group_index >= 0
+            if (~mask).any():
+                ids = ids[mask]
+                group_index = group_index[mask]
+
+        mask = duplicated(group_index, "first")
+        res = np.bincount(ids[~mask], minlength=ngroups)
+        res = ensure_int64(res)
+
+        ri = self._grouper.result_index
+        result: Series | DataFrame = self.obj._constructor(
+            res, index=ri, name=self.obj.name
+        )
+        if not self.as_index:
+            result = self._insert_inaxis_grouper(result)
+            result.index = default_index(len(result))
+        return result
+
+    @doc(Series.describe)
+    def describe(self, percentiles=None, include=None, exclude=None) -> Series:
+        return super().describe(
+            percentiles=percentiles, include=include, exclude=exclude
+        )
+
+    def value_counts(
+        self,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        bins=None,
+        dropna: bool = True,
+    ) -> Series | DataFrame:
+        """
+        Return a Series or DataFrame containing counts of unique rows.
+
+        .. versionadded:: 1.4.0
+
+        Parameters
+        ----------
+        normalize : bool, default False
+            Return proportions rather than frequencies.
+        sort : bool, default True
+            Sort by frequencies.
+        ascending : bool, default False
+            Sort in ascending order.
+        bins : int or list of ints, optional
+            Rather than count values, group them into half-open bins,
+            a convenience for pd.cut, only works with numeric data.
+        dropna : bool, default True
+            Don't include counts of rows that contain NA values.
+
+        Returns
+        -------
+        Series or DataFrame
+            Series if the groupby ``as_index`` is True, otherwise DataFrame.
+
+        See Also
+        --------
+        Series.value_counts: Equivalent method on Series.
+        DataFrame.value_counts: Equivalent method on DataFrame.
+        DataFrameGroupBy.value_counts: Equivalent method on DataFrameGroupBy.
+
+        Notes
+        -----
+        - If the groupby ``as_index`` is True then the returned Series will have a
+          MultiIndex with one level per input column.
+        - If the groupby ``as_index`` is False then the returned DataFrame will have an
+          additional column with the value_counts. The column is labelled 'count' or
+          'proportion', depending on the ``normalize`` parameter.
+
+        By default, rows that contain any NA values are omitted from
+        the result.
+
+        By default, the result will be in descending order so that the
+        first element of each group is the most frequently-occurring row.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     [1, 1, 2, 3, 2, 3, 3, 1, 1, 3, 3, 3],
+        ...     index=["A", "A", "A", "A", "A", "A", "B", "B", "B", "B", "B", "B"],
+        ... )
+        >>> s
+        A    1
+        A    1
+        A    2
+        A    3
+        A    2
+        A    3
+        B    3
+        B    1
+        B    1
+        B    3
+        B    3
+        B    3
+        dtype: int64
+        >>> g1 = s.groupby(s.index)
+        >>> g1.value_counts(bins=2)
+        A  (0.997, 2.0]    4
+           (2.0, 3.0]      2
+        B  (2.0, 3.0]      4
+           (0.997, 2.0]    2
+        Name: count, dtype: int64
+        >>> g1.value_counts(normalize=True)
+        A  1    0.333333
+           2    0.333333
+           3    0.333333
+        B  3    0.666667
+           1    0.333333
+        Name: proportion, dtype: float64
+        """
+        name = "proportion" if normalize else "count"
+
+        if bins is None:
+            result = self._value_counts(
+                normalize=normalize, sort=sort, ascending=ascending, dropna=dropna
+            )
+            result.name = name
+            return result
+
+        from pandas.core.reshape.merge import get_join_indexers
+        from pandas.core.reshape.tile import cut
+
+        ids = self._grouper.ids
+        val = self.obj._values
+
+        index_names = self._grouper.names + [self.obj.name]
+
+        if isinstance(val.dtype, CategoricalDtype) or (
+            bins is not None and not np.iterable(bins)
+        ):
+            # scalar bins cannot be done at top level
+            # in a backward compatible way
+            # GH38672 relates to categorical dtype
+            ser = self.apply(
+                Series.value_counts,
+                normalize=normalize,
+                sort=sort,
+                ascending=ascending,
+                bins=bins,
+            )
+            ser.name = name
+            ser.index.names = index_names
+            return ser
+
+        # groupby removes null keys from groupings
+        mask = ids != -1
+        ids, val = ids[mask], val[mask]
+
+        lab: Index | np.ndarray
+        if bins is None:
+            lab, lev = algorithms.factorize(val, sort=True)
+            llab = lambda lab, inc: lab[inc]
+        else:
+            # lab is a Categorical with categories an IntervalIndex
+            cat_ser = cut(Series(val, copy=False), bins, include_lowest=True)
+            cat_obj = cast("Categorical", cat_ser._values)
+            lev = cat_obj.categories
+            lab = lev.take(
+                cat_obj.codes,
+                allow_fill=True,
+                fill_value=lev._na_value,
+            )
+            llab = lambda lab, inc: lab[inc]._multiindex.codes[-1]
+
+        if isinstance(lab.dtype, IntervalDtype):
+            # TODO: should we do this inside II?
+            lab_interval = cast(Interval, lab)
+
+            sorter = np.lexsort((lab_interval.left, lab_interval.right, ids))
+        else:
+            sorter = np.lexsort((lab, ids))
+
+        ids, lab = ids[sorter], lab[sorter]
+
+        # group boundaries are where group ids change
+        idchanges = 1 + np.nonzero(ids[1:] != ids[:-1])[0]
+        idx = np.r_[0, idchanges]
+        if not len(ids):
+            idx = idchanges
+
+        # new values are where sorted labels change
+        lchanges = llab(lab, slice(1, None)) != llab(lab, slice(None, -1))
+        inc = np.r_[True, lchanges]
+        if not len(val):
+            inc = lchanges
+        inc[idx] = True  # group boundaries are also new values
+        out = np.diff(np.nonzero(np.r_[inc, True])[0])  # value counts
+
+        # num. of times each group should be repeated
+        rep = partial(np.repeat, repeats=np.add.reduceat(inc, idx))
+
+        # multi-index components
+        if isinstance(self._grouper.result_index, MultiIndex):
+            codes = list(self._grouper.result_index.codes)
+        else:
+            codes = [
+                algorithms.factorize(
+                    self._grouper.result_index,
+                    sort=self._grouper._sort,
+                    use_na_sentinel=self._grouper.dropna,
+                )[0]
+            ]
+        codes = [rep(level_codes) for level_codes in codes] + [llab(lab, inc)]
+        levels = self._grouper.levels + [lev]
+
+        if dropna:
+            mask = codes[-1] != -1
+            if mask.all():
+                dropna = False
+            else:
+                out, codes = out[mask], [level_codes[mask] for level_codes in codes]
+
+        if normalize:
+            out = out.astype("float")
+            d = np.diff(np.r_[idx, len(ids)])
+            if dropna:
+                m = ids[lab == -1]
+                np.add.at(d, m, -1)
+                acc = rep(d)[mask]
+            else:
+                acc = rep(d)
+            out /= acc
+
+        if sort and bins is None:
+            cat = ids[inc][mask] if dropna else ids[inc]
+            sorter = np.lexsort((out if ascending else -out, cat))
+            out, codes[-1] = out[sorter], codes[-1][sorter]
+
+        if bins is not None:
+            # for compat. with libgroupby.value_counts need to ensure every
+            # bin is present at every index level, null filled with zeros
+            diff = np.zeros(len(out), dtype="bool")
+            for level_codes in codes[:-1]:
+                diff |= np.r_[True, level_codes[1:] != level_codes[:-1]]
+
+            ncat, nbin = diff.sum(), len(levels[-1])
+
+            left = [np.repeat(np.arange(ncat), nbin), np.tile(np.arange(nbin), ncat)]
+
+            right = [diff.cumsum() - 1, codes[-1]]
+
+            # error: Argument 1 to "get_join_indexers" has incompatible type
+            # "List[ndarray[Any, Any]]"; expected "List[Union[Union[ExtensionArray,
+            # ndarray[Any, Any]], Index, Series]]
+            _, idx = get_join_indexers(
+                left,  # type: ignore[arg-type]
+                right,
+                sort=False,
+                how="left",
+            )
+            if idx is not None:
+                out = np.where(idx != -1, out[idx], 0)
+
+            if sort:
+                sorter = np.lexsort((out if ascending else -out, left[0]))
+                out, left[-1] = out[sorter], left[-1][sorter]
+
+            # build the multi-index w/ full levels
+            def build_codes(lev_codes: np.ndarray) -> np.ndarray:
+                return np.repeat(lev_codes[diff], nbin)
+
+            codes = [build_codes(lev_codes) for lev_codes in codes[:-1]]
+            codes.append(left[-1])
+
+        mi = MultiIndex(
+            levels=levels, codes=codes, names=index_names, verify_integrity=False
+        )
+
+        if is_integer_dtype(out.dtype):
+            out = ensure_int64(out)
+        result = self.obj._constructor(out, index=mi, name=name)
+        if not self.as_index:
+            result = result.reset_index()
+        return result
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        **kwargs,
+    ) -> Series:
+        """
+        Return the elements in the given *positional* indices in each group.
+
+        This means that we are not indexing according to actual values in
+        the index attribute of the object. We are indexing according to the
+        actual position of the element in the object.
+
+        If a requested index does not exist for some group, this method will raise.
+        To get similar behavior that ignores indices that don't exist, see
+        :meth:`.SeriesGroupBy.nth`.
+
+        Parameters
+        ----------
+        indices : array-like
+            An array of ints indicating which positions to take in each group.
+
+        **kwargs
+            For compatibility with :meth:`numpy.take`. Has no effect on the
+            output.
+
+        Returns
+        -------
+        Series
+            A Series containing the elements taken from each group.
+
+        See Also
+        --------
+        Series.take : Take elements from a Series along an axis.
+        Series.loc : Select a subset of a DataFrame by labels.
+        Series.iloc : Select a subset of a DataFrame by positions.
+        numpy.take : Take elements from an array along an axis.
+        SeriesGroupBy.nth : Similar to take, won't raise if indices don't exist.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("falcon", "bird", 389.0),
+        ...         ("parrot", "bird", 24.0),
+        ...         ("lion", "mammal", 80.5),
+        ...         ("monkey", "mammal", np.nan),
+        ...         ("rabbit", "mammal", 15.0),
+        ...     ],
+        ...     columns=["name", "class", "max_speed"],
+        ...     index=[4, 3, 2, 1, 0],
+        ... )
+        >>> df
+             name   class  max_speed
+        4  falcon    bird      389.0
+        3  parrot    bird       24.0
+        2    lion  mammal       80.5
+        1  monkey  mammal        NaN
+        0  rabbit  mammal       15.0
+        >>> gb = df["name"].groupby([1, 1, 2, 2, 2])
+
+        Take elements at rows 0 and 1 in each group.
+
+        >>> gb.take([0, 1])
+        1  4    falcon
+           3    parrot
+        2  2      lion
+           1    monkey
+        Name: name, dtype: object
+
+        We may take elements using negative integers for positive indices,
+        starting from the end of the object, just like with Python lists.
+
+        >>> gb.take([-1, -2])
+        1  3    parrot
+           4    falcon
+        2  0    rabbit
+           1    monkey
+        Name: name, dtype: object
+        """
+        result = self._op_via_apply("take", indices=indices, **kwargs)
+        return result
+
+    def skew(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series:
+        """
+        Return unbiased skew within groups.
+
+        Normalized by N-1.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series
+            Unbiased skew within groups.
+
+        See Also
+        --------
+        Series.skew : Return unbiased skew over requested axis.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [390.0, 350.0, 357.0, np.nan, 22.0, 20.0, 30.0],
+        ...     index=[
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...     ],
+        ...     name="Max Speed",
+        ... )
+        >>> ser
+        Falcon    390.0
+        Falcon    350.0
+        Falcon    357.0
+        Falcon      NaN
+        Parrot     22.0
+        Parrot     20.0
+        Parrot     30.0
+        Name: Max Speed, dtype: float64
+        >>> ser.groupby(level=0).skew()
+        Falcon    1.525174
+        Parrot    1.457863
+        Name: Max Speed, dtype: float64
+        >>> ser.groupby(level=0).skew(skipna=False)
+        Falcon         NaN
+        Parrot    1.457863
+        Name: Max Speed, dtype: float64
+        """
+
+        return self._cython_agg_general(
+            "skew", alt=None, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    def kurt(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Series:
+        """
+        Return unbiased kurtosis within groups.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        Series
+            Unbiased kurtosis within groups.
+
+        See Also
+        --------
+        Series.kurt : Return unbiased kurtosis over requested axis.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [390.0, 350.0, 357.0, 333.0, np.nan, 22.0, 20.0, 30.0, 40.0, 41.0],
+        ...     index=[
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Falcon",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...         "Parrot",
+        ...     ],
+        ...     name="Max Speed",
+        ... )
+        >>> ser
+        Falcon    390.0
+        Falcon    350.0
+        Falcon    357.0
+        Falcon    333.0
+        Falcon      NaN
+        Parrot     22.0
+        Parrot     20.0
+        Parrot     30.0
+        Parrot     40.0
+        Parrot     41.0
+        Name: Max Speed, dtype: float64
+        >>> ser.groupby(level=0).kurt()
+        Falcon    1.622109
+        Parrot   -2.878714
+        Name: Max Speed, dtype: float64
+        >>> ser.groupby(level=0).kurt(skipna=False)
+        Falcon         NaN
+        Parrot   -2.878714
+        Name: Max Speed, dtype: float64
+        """
+
+        def alt(obj):
+            # This should not be reached since the cython path should raise
+            #  TypeError and not NotImplementedError.
+            raise TypeError(f"'kurt' is not supported for dtype={obj.dtype}")
+
+        return self._cython_agg_general(
+            "kurt", alt=alt, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @property
+    @doc(Series.plot.__doc__)
+    def plot(self) -> GroupByPlot:
+        result = GroupByPlot(self)
+        return result
+
+    @doc(Series.nlargest.__doc__)
+    def nlargest(
+        self, n: int = 5, keep: Literal["first", "last", "all"] = "first"
+    ) -> Series:
+        f = partial(Series.nlargest, n=n, keep=keep)
+        data = self._obj_with_exclusions
+        # Don't change behavior if result index happens to be the same, i.e.
+        # already ordered and n >= all group sizes.
+        result = self._python_apply_general(f, data, not_indexed_same=True)
+        return result
+
+    @doc(Series.nsmallest.__doc__)
+    def nsmallest(
+        self, n: int = 5, keep: Literal["first", "last", "all"] = "first"
+    ) -> Series:
+        f = partial(Series.nsmallest, n=n, keep=keep)
+        data = self._obj_with_exclusions
+        # Don't change behavior if result index happens to be the same, i.e.
+        # already ordered and n >= all group sizes.
+        result = self._python_apply_general(f, data, not_indexed_same=True)
+        return result
+
+    def idxmin(self, skipna: bool = True) -> Series:
+        """
+        Return the row label of the minimum value.
+
+        If multiple values equal the minimum, the first row label with that
+        value is returned.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values.
+
+        Returns
+        -------
+        Series
+            Indexes of minima in each group.
+
+        Raises
+        ------
+        ValueError
+            If the Series is empty or skipna=False and any value is NA.
+
+        See Also
+        --------
+        numpy.argmin : Return indices of the minimum values
+            along the given axis.
+        DataFrame.idxmin : Return index of first occurrence of minimum
+            over requested axis.
+        Series.idxmax : Return index *label* of the first occurrence
+            of maximum of values.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+
+        >>> ser.groupby(["a", "a", "b", "b"]).idxmin()
+        a   2023-01-01
+        b   2023-02-01
+        dtype: datetime64[s]
+        """
+        return self._idxmax_idxmin("idxmin", skipna=skipna)
+
+    def idxmax(self, skipna: bool = True) -> Series:
+        """
+        Return the row label of the maximum value.
+
+        If multiple values equal the maximum, the first row label with that
+        value is returned.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values.
+
+        Returns
+        -------
+        Series
+            Indexes of maxima in each group.
+
+        Raises
+        ------
+        ValueError
+            If the Series is empty or skipna=False and any value is NA.
+
+        See Also
+        --------
+        numpy.argmax : Return indices of the maximum values
+            along the given axis.
+        DataFrame.idxmax : Return index of first occurrence of maximum
+            over requested axis.
+        Series.idxmin : Return index *label* of the first occurrence
+            of minimum of values.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+
+        >>> ser.groupby(["a", "a", "b", "b"]).idxmax()
+        a   2023-01-15
+        b   2023-02-15
+        dtype: datetime64[s]
+        """
+        return self._idxmax_idxmin("idxmax", skipna=skipna)
+
+    @doc(Series.corr.__doc__)
+    def corr(
+        self,
+        other: Series,
+        method: CorrelationMethod = "pearson",
+        min_periods: int | None = None,
+    ) -> Series:
+        result = self._op_via_apply(
+            "corr", other=other, method=method, min_periods=min_periods
+        )
+        return result
+
+    @doc(Series.cov.__doc__)
+    def cov(
+        self, other: Series, min_periods: int | None = None, ddof: int | None = 1
+    ) -> Series:
+        result = self._op_via_apply(
+            "cov", other=other, min_periods=min_periods, ddof=ddof
+        )
+        return result
+
+    @property
+    def is_monotonic_increasing(self) -> Series:
+        """
+        Return whether each group's values are monotonically increasing.
+
+        Returns
+        -------
+        Series
+
+        See Also
+        --------
+        SeriesGroupBy.is_monotonic_decreasing : Return whether each group's values
+            are monotonically decreasing.
+
+        Examples
+        --------
+        >>> s = pd.Series([2, 1, 3, 4], index=["Falcon", "Falcon", "Parrot", "Parrot"])
+        >>> s.groupby(level=0).is_monotonic_increasing
+        Falcon    False
+        Parrot     True
+        dtype: bool
+        """
+        return self.apply(lambda ser: ser.is_monotonic_increasing)
+
+    @property
+    def is_monotonic_decreasing(self) -> Series:
+        """
+        Return whether each group's values are monotonically decreasing.
+
+        Returns
+        -------
+        Series
+
+        See Also
+        --------
+        SeriesGroupBy.is_monotonic_increasing : Return whether each group's values
+            are monotonically increasing.
+
+        Examples
+        --------
+        >>> s = pd.Series([2, 1, 3, 4], index=["Falcon", "Falcon", "Parrot", "Parrot"])
+        >>> s.groupby(level=0).is_monotonic_decreasing
+        Falcon     True
+        Parrot    False
+        dtype: bool
+        """
+        return self.apply(lambda ser: ser.is_monotonic_decreasing)
+
+    @doc(Series.hist.__doc__)
+    def hist(
+        self,
+        by=None,
+        ax=None,
+        grid: bool = True,
+        xlabelsize: int | None = None,
+        xrot: float | None = None,
+        ylabelsize: int | None = None,
+        yrot: float | None = None,
+        figsize: tuple[float, float] | None = None,
+        bins: int | Sequence[int] = 10,
+        backend: str | None = None,
+        legend: bool = False,
+        **kwargs,
+    ):
+        result = self._op_via_apply(
+            "hist",
+            by=by,
+            ax=ax,
+            grid=grid,
+            xlabelsize=xlabelsize,
+            xrot=xrot,
+            ylabelsize=ylabelsize,
+            yrot=yrot,
+            figsize=figsize,
+            bins=bins,
+            backend=backend,
+            legend=legend,
+            **kwargs,
+        )
+        return result
+
+    @property
+    @doc(Series.dtype.__doc__)
+    def dtype(self) -> Series:
+        return self.apply(lambda ser: ser.dtype)
+
+    def unique(self) -> Series:
+        """
+        Return unique values for each group.
+
+        It returns unique values for each of the grouped values. Returned in
+        order of appearance. Hash table-based unique, therefore does NOT sort.
+
+        Returns
+        -------
+        Series
+            Unique values for each of the grouped values.
+
+        See Also
+        --------
+        Series.unique : Return unique values of Series object.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("Chihuahua", "dog", 6.1),
+        ...         ("Beagle", "dog", 15.2),
+        ...         ("Chihuahua", "dog", 6.9),
+        ...         ("Persian", "cat", 9.2),
+        ...         ("Chihuahua", "dog", 7),
+        ...         ("Persian", "cat", 8.8),
+        ...     ],
+        ...     columns=["breed", "animal", "height_in"],
+        ... )
+        >>> df
+               breed     animal   height_in
+        0  Chihuahua        dog         6.1
+        1     Beagle        dog        15.2
+        2  Chihuahua        dog         6.9
+        3    Persian        cat         9.2
+        4  Chihuahua        dog         7.0
+        5    Persian        cat         8.8
+        >>> ser = df.groupby("animal")["breed"].unique()
+        >>> ser
+        animal
+        cat              [Persian]
+        dog    [Chihuahua, Beagle]
+        Name: breed, dtype: object
+        """
+        result = self._op_via_apply("unique")
+        return result
+
+
+@set_module("pandas.api.typing")
+class DataFrameGroupBy(GroupBy[DataFrame]):
+    _agg_examples_doc = dedent(
+        """
+    Examples
+    --------
+    >>> data = {"A": [1, 1, 2, 2],
+    ...         "B": [1, 2, 3, 4],
+    ...         "C": [0.362838, 0.227877, 1.267767, -0.562860]}
+    >>> df = pd.DataFrame(data)
+    >>> df
+       A  B         C
+    0  1  1  0.362838
+    1  1  2  0.227877
+    2  2  3  1.267767
+    3  2  4 -0.562860
+
+    The aggregation is for each column.
+
+    >>> df.groupby('A').agg('min')
+       B         C
+    A
+    1  1  0.227877
+    2  3 -0.562860
+
+    Multiple aggregations
+
+    >>> df.groupby('A').agg(['min', 'max'])
+        B             C
+      min max       min       max
+    A
+    1   1   2  0.227877  0.362838
+    2   3   4 -0.562860  1.267767
+
+    Select a column for aggregation
+
+    >>> df.groupby('A').B.agg(['min', 'max'])
+       min  max
+    A
+    1    1    2
+    2    3    4
+
+    User-defined function for aggregation
+
+    >>> df.groupby('A').agg(lambda x: sum(x) + 2)
+        B	       C
+    A
+    1	5	2.590715
+    2	9	2.704907
+
+    Different aggregations per column
+
+    >>> df.groupby('A').agg({'B': ['min', 'max'], 'C': 'sum'})
+        B             C
+      min max       sum
+    A
+    1   1   2  0.590715
+    2   3   4  0.704907
+
+    To control the output names with different aggregations per column,
+    pandas supports "named aggregation"
+
+    >>> df.groupby("A").agg(
+    ...     b_min=pd.NamedAgg(column="B", aggfunc="min"),
+    ...     c_sum=pd.NamedAgg(column="C", aggfunc="sum")
+    ... )
+       b_min     c_sum
+    A
+    1      1  0.590715
+    2      3  0.704907
+
+    - The keywords are the *output* column names
+    - The values are tuples whose first element is the column to select
+      and the second element is the aggregation to apply to that column.
+      Pandas provides the ``pandas.NamedAgg`` namedtuple with the fields
+      ``['column', 'aggfunc']`` to make it clearer what the arguments are.
+      As usual, the aggregation can be a callable or a string alias.
+
+    See :ref:`groupby.aggregate.named` for more.
+
+    .. versionchanged:: 1.3.0
+
+        The resulting dtype will reflect the return value of the aggregating function.
+
+    >>> df.groupby("A")[["B"]].agg(lambda x: x.astype(float).min())
+          B
+    A
+    1   1.0
+    2   3.0
+    """
+    )
+
+    def aggregate(self, func=None, *args, engine=None, engine_kwargs=None, **kwargs):
+        """
+        Aggregate using one or more operations.
+
+        The ``aggregate`` function allows the application of one or more aggregation
+        operations on groups of data within a DataFrameGroupBy object. It supports
+        various aggregation methods, including user-defined functions and predefined
+        functions such as 'sum', 'mean', etc.
+
+        Parameters
+        ----------
+        func : function, str, list, dict or None
+            Function to use for aggregating the data. If a function, must either
+            work when passed a DataFrame or when passed to DataFrame.apply.
+
+            Accepted combinations are:
+
+            - function
+            - string function name
+            - list of functions and/or function names, e.g. ``[np.sum, 'mean']``
+            - dict of index labels -> functions, function names or list of such.
+            - None, in which case ``**kwargs`` are used with Named Aggregation. Here the
+              output has one column for each element in ``**kwargs``. The name of the
+              column is keyword, whereas the value determines the aggregation used to
+              compute the values in the column.
+
+              Can also accept a Numba JIT function with
+              ``engine='numba'`` specified. Only passing a single function is supported
+              with this engine.
+
+              If the ``'numba'`` engine is chosen, the function must be
+              a user defined function with ``values`` and ``index`` as the
+              first and second arguments respectively in the function signature.
+              Each group's index will be passed to the user defined function
+              and optionally available for use.
+
+        *args
+            Positional arguments to pass to func.
+        engine : str, default None
+            * ``'cython'`` : Runs the function through C-extensions from cython.
+            * ``'numba'`` : Runs the function through JIT compiled code from numba.
+            * ``None`` : Defaults to ``'cython'`` or globally setting
+                ``compute.use_numba``
+
+        engine_kwargs : dict, default None
+            * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{'nopython': True, 'nogil': False, 'parallel': False}`` and will be
+              applied to the function
+
+        **kwargs
+            * If ``func`` is None, ``**kwargs`` are used to define the output names and
+              aggregations via Named Aggregation. See ``func`` entry.
+            * Otherwise, keyword arguments to be passed into func.
+
+        Returns
+        -------
+        DataFrame
+            Aggregated DataFrame based on the grouping and the applied aggregation
+            functions.
+
+        See Also
+        --------
+        DataFrame.groupby.apply : Apply function func group-wise
+            and combine the results together.
+        DataFrame.groupby.transform : Transforms the Series on each group
+            based on the given function.
+        DataFrame.aggregate : Aggregate using one or more operations.
+
+        Notes
+        -----
+        When using ``engine='numba'``, there will be no "fall back" behavior internally.
+        The group data and group index will be passed as numpy arrays to the JITed
+        user defined function, and no alternative execution attempts will be tried.
+
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``,
+            see the examples below.
+
+        Examples
+        --------
+        >>> data = {
+        ...     "A": [1, 1, 2, 2],
+        ...     "B": [1, 2, 3, 4],
+        ...     "C": [0.362838, 0.227877, 1.267767, -0.562860],
+        ... }
+        >>> df = pd.DataFrame(data)
+        >>> df
+           A  B         C
+        0  1  1  0.362838
+        1  1  2  0.227877
+        2  2  3  1.267767
+        3  2  4 -0.562860
+
+        The aggregation is for each column.
+
+        >>> df.groupby("A").agg("min")
+           B         C
+        A
+        1  1  0.227877
+        2  3 -0.562860
+
+        Multiple aggregations
+
+        >>> df.groupby("A").agg(["min", "max"])
+            B             C
+          min max       min       max
+        A
+        1   1   2  0.227877  0.362838
+        2   3   4 -0.562860  1.267767
+
+        Select a column for aggregation
+
+        >>> df.groupby("A").B.agg(["min", "max"])
+           min  max
+        A
+        1    1    2
+        2    3    4
+
+        User-defined function for aggregation
+
+        >>> df.groupby("A").agg(lambda x: sum(x) + 2)
+            B          C
+        A
+        1       5       2.590715
+        2       9       2.704907
+
+        Different aggregations per column
+
+        >>> df.groupby("A").agg({"B": ["min", "max"], "C": "sum"})
+            B             C
+          min max       sum
+        A
+        1   1   2  0.590715
+        2   3   4  0.704907
+
+        To control the output names with different aggregations per column,
+        pandas supports "named aggregation"
+
+        >>> df.groupby("A").agg(
+        ...     b_min=pd.NamedAgg(column="B", aggfunc="min"),
+        ...     c_sum=pd.NamedAgg(column="C", aggfunc="sum"),
+        ... )
+           b_min     c_sum
+        A
+        1      1  0.590715
+        2      3  0.704907
+
+        - The keywords are the *output* column names
+        - The values are tuples whose first element is the column to select
+          and the second element is the aggregation to apply to that column.
+          Pandas provides the ``pandas.NamedAgg`` namedtuple with the fields
+          ``['column', 'aggfunc']`` to make it clearer what the arguments are.
+          As usual, the aggregation can be a callable or a string alias.
+
+        See :ref:`groupby.aggregate.named` for more.
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the aggregating
+            function.
+
+        >>> df.groupby("A")[["B"]].agg(lambda x: x.astype(float).min())
+              B
+        A
+        1   1.0
+        2   3.0
+        """
+        relabeling, func, columns, order = reconstruct_func(func, **kwargs)
+        func = maybe_mangle_lambdas(func)
+
+        if maybe_use_numba(engine):
+            # Not all agg functions support numba, only propagate numba kwargs
+            # if user asks for numba
+            kwargs["engine"] = engine
+            kwargs["engine_kwargs"] = engine_kwargs
+
+        op = GroupByApply(self, func, args=args, kwargs=kwargs)
+        result = op.agg()
+        if not is_dict_like(func) and result is not None:
+            # GH #52849
+            if not self.as_index and is_list_like(func):
+                return result.reset_index()
+            else:
+                return result
+        elif relabeling:
+            # this should be the only (non-raising) case with relabeling
+            # used reordered index of columns
+            result = cast(DataFrame, result)
+            result = result.iloc[:, order]
+            result = cast(DataFrame, result)
+            # error: Incompatible types in assignment (expression has type
+            # "Optional[List[str]]", variable has type
+            # "Union[Union[Union[ExtensionArray, ndarray[Any, Any]],
+            # Index, Series], Sequence[Any]]")
+            result.columns = columns  # type: ignore[assignment]
+
+        if result is None:
+            # Remove the kwargs we inserted
+            # (already stored in engine, engine_kwargs arguments)
+            if "engine" in kwargs:
+                del kwargs["engine"]
+                del kwargs["engine_kwargs"]
+            # at this point func is not a str, list-like, dict-like,
+            # or a known callable(e.g. sum)
+            if maybe_use_numba(engine):
+                return self._aggregate_with_numba(
+                    func, *args, engine_kwargs=engine_kwargs, **kwargs
+                )
+            # grouper specific aggregations
+            if self._grouper.nkeys > 1:
+                # test_groupby_as_index_series_scalar gets here with 'not self.as_index'
+                return self._python_agg_general(func, *args, **kwargs)
+            elif args or kwargs:
+                # test_pass_args_kwargs gets here (with and without as_index)
+                # can't return early
+                result = self._aggregate_frame(func, *args, **kwargs)
+
+            else:
+                # try to treat as if we are passing a list
+                gba = GroupByApply(self, [func], args=(), kwargs={})
+                try:
+                    result = gba.agg()
+
+                except ValueError as err:
+                    if "No objects to concatenate" not in str(err):
+                        raise
+                    # _aggregate_frame can fail with e.g. func=Series.mode,
+                    # where it expects 1D values but would be getting 2D values
+                    # In other tests, using aggregate_frame instead of GroupByApply
+                    #  would give correct values but incorrect dtypes
+                    #  object vs float64 in test_cython_agg_empty_buckets
+                    #  float64 vs int64 in test_category_order_apply
+                    result = self._aggregate_frame(func)
+
+                else:
+                    # GH#32040, GH#35246
+                    # e.g. test_groupby_as_index_select_column_sum_empty_df
+                    result = cast(DataFrame, result)
+                    result.columns = self._obj_with_exclusions.columns.copy()
+
+        if not self.as_index:
+            result = self._insert_inaxis_grouper(result)
+            result.index = default_index(len(result))
+
+        return result
+
+    agg = aggregate
+
+    def _python_agg_general(self, func, *args, **kwargs):
+        f = lambda x: func(x, *args, **kwargs)
+
+        if self.ngroups == 0:
+            # e.g. test_evaluate_with_empty_groups different path gets different
+            #  result dtype in empty case.
+            return self._python_apply_general(f, self._selected_obj, is_agg=True)
+
+        obj = self._obj_with_exclusions
+
+        if not len(obj.columns):
+            # e.g. test_margins_no_values_no_cols
+            return self._python_apply_general(f, self._selected_obj)
+
+        output: dict[int, ArrayLike] = {}
+        for idx, (name, ser) in enumerate(obj.items()):
+            result = self._grouper.agg_series(ser, f)
+            output[idx] = result
+
+        res = self.obj._constructor(output)
+        res.columns = obj.columns.copy(deep=False)
+        return self._wrap_aggregated_output(res)
+
+    def _aggregate_frame(self, func, *args, **kwargs) -> DataFrame:
+        if self._grouper.nkeys != 1:
+            raise AssertionError("Number of keys must be 1")
+
+        obj = self._obj_with_exclusions
+
+        result: dict[Hashable, NDFrame | np.ndarray] = {}
+        for name, grp_df in self._grouper.get_iterator(obj):
+            fres = func(grp_df, *args, **kwargs)
+            result[name] = fres
+
+        result_index = self._grouper.result_index
+        out = self.obj._constructor(result, index=obj.columns, columns=result_index)
+        out = out.T
+
+        return out
+
+    def _wrap_applied_output(
+        self,
+        data: DataFrame,
+        values: list,
+        not_indexed_same: bool = False,
+        is_transform: bool = False,
+    ):
+        if len(values) == 0:
+            if is_transform:
+                # GH#47787 see test_group_on_empty_multiindex
+                res_index = data.index
+            elif not self.group_keys:
+                res_index = None
+            else:
+                res_index = self._grouper.result_index
+
+            result = self.obj._constructor(index=res_index, columns=data.columns)
+            result = result.astype(data.dtypes)
+            return result
+
+        # GH12824
+        # using values[0] here breaks test_groupby_apply_none_first
+        first_not_none = next(com.not_none(*values), None)
+
+        if first_not_none is None:
+            # GH9684 - All values are None, return an empty frame
+            # GH57775 - Ensure that columns and dtypes from original frame are kept.
+            result = self.obj._constructor(columns=data.columns)
+            result = result.astype(data.dtypes)
+            return result
+        elif isinstance(first_not_none, DataFrame):
+            return self._concat_objects(
+                values,
+                not_indexed_same=not_indexed_same,
+                is_transform=is_transform,
+            )
+
+        key_index = self._grouper.result_index if self.as_index else None
+
+        if isinstance(first_not_none, (np.ndarray, Index)):
+            # GH#1738: values is list of arrays of unequal lengths
+            #  fall through to the outer else clause
+            # TODO: sure this is right?  we used to do this
+            #  after raising AttributeError above
+            # GH 18930
+            if not is_hashable(self._selection):
+                # error: Need type annotation for "name"
+                name = tuple(self._selection)  # type: ignore[var-annotated, arg-type]
+            else:
+                # error: Incompatible types in assignment
+                # (expression has type "Hashable", variable
+                # has type "Tuple[Any, ...]")
+                name = self._selection  # type: ignore[assignment]
+            return self.obj._constructor_sliced(values, index=key_index, name=name)
+        elif not isinstance(first_not_none, Series):
+            # values are not series or array-like but scalars
+            # self._selection not passed through to Series as the
+            # result should not take the name of original selection
+            # of columns
+            if self.as_index:
+                return self.obj._constructor_sliced(values, index=key_index)
+            else:
+                result = self.obj._constructor(values, columns=[self._selection])
+                result = self._insert_inaxis_grouper(result)
+                return result
+        else:
+            # values are Series
+            return self._wrap_applied_output_series(
+                values,
+                not_indexed_same,
+                first_not_none,
+                key_index,
+                is_transform,
+            )
+
+    def _wrap_applied_output_series(
+        self,
+        values: list[Series],
+        not_indexed_same: bool,
+        first_not_none,
+        key_index: Index | None,
+        is_transform: bool,
+    ) -> DataFrame | Series:
+        kwargs = first_not_none._construct_axes_dict()
+        backup = Series(**kwargs)
+        values = [x if (x is not None) else backup for x in values]
+
+        all_indexed_same = all_indexes_same(x.index for x in values)
+
+        if not all_indexed_same:
+            # GH 8467
+            return self._concat_objects(
+                values,
+                not_indexed_same=True,
+                is_transform=is_transform,
+            )
+
+        # Combine values
+        # vstack+constructor is faster than concat and handles MI-columns
+        stacked_values = np.vstack([np.asarray(v) for v in values])
+
+        index = key_index
+        columns = first_not_none.index.copy()
+        if columns.name is None:
+            # GH6124 - propagate name of Series when it's consistent
+            names = {v.name for v in values}
+            if len(names) == 1:
+                columns.name = next(iter(names))
+
+        if stacked_values.dtype == object:
+            # We'll have the DataFrame constructor do inference
+            stacked_values = stacked_values.tolist()
+        result = self.obj._constructor(stacked_values, index=index, columns=columns)
+
+        if not self.as_index:
+            result = self._insert_inaxis_grouper(result)
+
+        return result
+
+    def _cython_transform(
+        self,
+        how: str,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> DataFrame:
+        # We have multi-block tests
+        #  e.g. test_rank_min_int, test_cython_transform_frame
+        #  test_transform_numeric_ret
+        mgr: BlockManager = self._get_data_to_aggregate(
+            numeric_only=numeric_only, name=how
+        )
+
+        def arr_func(bvalues: ArrayLike) -> ArrayLike:
+            return self._grouper._cython_operation(
+                "transform", bvalues, how, 1, **kwargs
+            )
+
+        res_mgr = mgr.apply(arr_func)
+
+        res_df = self.obj._constructor_from_mgr(res_mgr, axes=res_mgr.axes)
+        return res_df
+
+    def _transform_general(self, func, engine, engine_kwargs, *args, **kwargs):
+        if maybe_use_numba(engine):
+            return self._transform_with_numba(
+                func, *args, engine_kwargs=engine_kwargs, **kwargs
+            )
+        from pandas.core.reshape.concat import concat
+
+        applied = []
+        obj = self._obj_with_exclusions
+        gen = self._grouper.get_iterator(obj)
+        fast_path, slow_path = self._define_paths(func, *args, **kwargs)
+
+        # Determine whether to use slow or fast path by evaluating on the first group.
+        # Need to handle the case of an empty generator and process the result so that
+        # it does not need to be computed again.
+        try:
+            name, group = next(gen)
+        except StopIteration:
+            pass
+        else:
+            # 2023-02-27 No tests broken by disabling this pinning
+            object.__setattr__(group, "name", name)
+            try:
+                path, res = self._choose_path(fast_path, slow_path, group)
+            except ValueError as err:
+                # e.g. test_transform_with_non_scalar_group
+                msg = "transform must return a scalar value for each group"
+                raise ValueError(msg) from err
+            if group.size > 0:
+                res = _wrap_transform_general_frame(self.obj, group, res)
+                applied.append(res)
+
+        # Compute and process with the remaining groups
+        for name, group in gen:
+            if group.size == 0:
+                continue
+            # 2023-02-27 No tests broken by disabling this pinning
+            object.__setattr__(group, "name", name)
+            res = path(group)
+
+            res = _wrap_transform_general_frame(self.obj, group, res)
+            applied.append(res)
+
+        concat_index = obj.columns
+        concatenated = concat(
+            applied, axis=0, verify_integrity=False, ignore_index=True
+        )
+        concatenated = concatenated.reindex(concat_index, axis=1)
+        return self._set_result_index_ordered(concatenated)
+
+    __examples_dataframe_doc = dedent(
+        """
+    >>> df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
+    ...                           'foo', 'bar'],
+    ...                    'B' : ['one', 'one', 'two', 'three',
+    ...                           'two', 'two'],
+    ...                    'C' : [1, 5, 5, 2, 5, 5],
+    ...                    'D' : [2.0, 5., 8., 1., 2., 9.]})
+    >>> grouped = df.groupby('A')[['C', 'D']]
+    >>> grouped.transform(lambda x: (x - x.mean()) / x.std())
+            C         D
+    0 -1.154701 -0.577350
+    1  0.577350  0.000000
+    2  0.577350  1.154701
+    3 -1.154701 -1.000000
+    4  0.577350 -0.577350
+    5  0.577350  1.000000
+
+    Broadcast result of the transformation
+
+    >>> grouped.transform(lambda x: x.max() - x.min())
+        C    D
+    0  4.0  6.0
+    1  3.0  8.0
+    2  4.0  6.0
+    3  3.0  8.0
+    4  4.0  6.0
+    5  3.0  8.0
+
+    >>> grouped.transform("mean")
+        C    D
+    0  3.666667  4.0
+    1  4.000000  5.0
+    2  3.666667  4.0
+    3  4.000000  5.0
+    4  3.666667  4.0
+    5  4.000000  5.0
+
+    .. versionchanged:: 1.3.0
+
+    The resulting dtype will reflect the return value of the passed ``func``,
+    for example:
+
+    >>> grouped.transform(lambda x: x.astype(int).max())
+    C  D
+    0  5  8
+    1  5  9
+    2  5  8
+    3  5  9
+    4  5  8
+    5  5  9
+    """
+    )
+
+    @Substitution(klass="DataFrame", example=__examples_dataframe_doc)
+    @Appender(_transform_template)
+    def transform(self, func, *args, engine=None, engine_kwargs=None, **kwargs):
+        return self._transform(
+            func, *args, engine=engine, engine_kwargs=engine_kwargs, **kwargs
+        )
+
+    def _define_paths(self, func, *args, **kwargs):
+        if isinstance(func, str):
+            fast_path = lambda group: getattr(group, func)(*args, **kwargs)
+            slow_path = lambda group: group.apply(
+                lambda x: getattr(x, func)(*args, **kwargs), axis=0
+            )
+        else:
+            fast_path = lambda group: func(group, *args, **kwargs)
+            slow_path = lambda group: group.apply(
+                lambda x: func(x, *args, **kwargs), axis=0
+            )
+        return fast_path, slow_path
+
+    def _choose_path(self, fast_path: Callable, slow_path: Callable, group: DataFrame):
+        path = slow_path
+        res = slow_path(group)
+
+        if self.ngroups == 1:
+            # no need to evaluate multiple paths when only
+            # a single group exists
+            return path, res
+
+        # if we make it here, test if we can use the fast path
+        try:
+            res_fast = fast_path(group)
+        except AssertionError:
+            raise  # pragma: no cover
+        except Exception:
+            # GH#29631 For user-defined function, we can't predict what may be
+            #  raised; see test_transform.test_transform_fastpath_raises
+            return path, res
+
+        # verify fast path returns either:
+        # a DataFrame with columns equal to group.columns
+        # OR a Series with index equal to group.columns
+        if isinstance(res_fast, DataFrame):
+            if not res_fast.columns.equals(group.columns):
+                return path, res
+        elif isinstance(res_fast, Series):
+            if not res_fast.index.equals(group.columns):
+                return path, res
+        else:
+            return path, res
+
+        if res_fast.equals(res):
+            path = fast_path
+
+        return path, res
+
+    def filter(self, func, dropna: bool = True, *args, **kwargs) -> DataFrame:
+        """
+        Filter elements from groups that don't satisfy a criterion.
+
+        Elements from groups are filtered if they do not satisfy the
+        boolean criterion specified by func.
+
+        Parameters
+        ----------
+        func : function
+            Criterion to apply to each group. Should return True or False.
+        dropna : bool
+            Drop groups that do not pass the filter. True by default; if False,
+            groups that evaluate False are filled with NaNs.
+        *args : tuple
+            Additional positional arguments to pass to `func`.
+        **kwargs : dict
+            Additional keyword arguments to pass to `func`.
+
+        Returns
+        -------
+        DataFrame
+            The filtered subset of the original DataFrame.
+
+        See Also
+        --------
+        DataFrame.filter: Filter elements of ungrouped DataFrame.
+        SeriesGroupBy.filter : Filter elements from groups base on criterion.
+
+        Notes
+        -----
+        Each subframe is endowed the attribute 'name' in case you need to know
+        which group you are working on.
+
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": ["foo", "bar", "foo", "bar", "foo", "bar"],
+        ...         "B": [1, 2, 3, 4, 5, 6],
+        ...         "C": [2.0, 5.0, 8.0, 1.0, 2.0, 9.0],
+        ...     }
+        ... )
+        >>> grouped = df.groupby("A")
+        >>> grouped.filter(lambda x: x["B"].mean() > 3.0)
+             A  B    C
+        1  bar  2  5.0
+        3  bar  4  1.0
+        5  bar  6  9.0
+        """
+        indices = []
+
+        obj = self._selected_obj
+        gen = self._grouper.get_iterator(obj)
+
+        for name, group in gen:
+            # 2023-02-27 no tests are broken this pinning, but it is documented in the
+            #  docstring above.
+            object.__setattr__(group, "name", name)
+
+            res = func(group, *args, **kwargs)
+
+            try:
+                res = res.squeeze()
+            except AttributeError:  # allow e.g., scalars and frames to pass
+                pass
+
+            # interpret the result of the filter
+            if is_bool(res) or (is_scalar(res) and isna(res)):
+                if notna(res) and res:
+                    indices.append(self._get_index(name))
+            else:
+                # non scalars aren't allowed
+                raise TypeError(
+                    f"filter function returned a {type(res).__name__}, "
+                    "but expected a scalar bool"
+                )
+
+        return self._apply_filter(indices, dropna)
+
+    def __getitem__(self, key) -> DataFrameGroupBy | SeriesGroupBy:
+        # per GH 23566
+        if isinstance(key, tuple) and len(key) > 1:
+            # if len == 1, then it becomes a SeriesGroupBy and this is actually
+            # valid syntax, so don't raise
+            raise ValueError(
+                "Cannot subset columns with a tuple with more than one element. "
+                "Use a list instead."
+            )
+        return super().__getitem__(key)
+
+    def _gotitem(self, key, ndim: int, subset=None):
+        """
+        sub-classes to define
+        return a sliced object
+
+        Parameters
+        ----------
+        key : string / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        if ndim == 2:
+            if subset is None:
+                subset = self.obj
+            return DataFrameGroupBy(
+                subset,
+                self.keys,
+                level=self.level,
+                grouper=self._grouper,
+                exclusions=self.exclusions,
+                selection=key,
+                as_index=self.as_index,
+                sort=self.sort,
+                group_keys=self.group_keys,
+                observed=self.observed,
+                dropna=self.dropna,
+            )
+        elif ndim == 1:
+            if subset is None:
+                subset = self.obj[key]
+            return SeriesGroupBy(
+                subset,
+                self.keys,
+                level=self.level,
+                grouper=self._grouper,
+                exclusions=self.exclusions,
+                selection=key,
+                as_index=self.as_index,
+                sort=self.sort,
+                group_keys=self.group_keys,
+                observed=self.observed,
+                dropna=self.dropna,
+            )
+
+        raise AssertionError("invalid ndim for _gotitem")
+
+    def _get_data_to_aggregate(
+        self, *, numeric_only: bool = False, name: str | None = None
+    ) -> BlockManager:
+        obj = self._obj_with_exclusions
+        mgr = obj._mgr
+        if numeric_only:
+            mgr = mgr.get_numeric_data()
+        return mgr
+
+    def _wrap_agged_manager(self, mgr: BlockManager) -> DataFrame:
+        return self.obj._constructor_from_mgr(mgr, axes=mgr.axes)
+
+    def _apply_to_column_groupbys(self, func) -> DataFrame:
+        from pandas.core.reshape.concat import concat
+
+        obj = self._obj_with_exclusions
+        columns = obj.columns
+        sgbs = (
+            SeriesGroupBy(
+                obj.iloc[:, i],
+                selection=colname,
+                grouper=self._grouper,
+                exclusions=self.exclusions,
+                observed=self.observed,
+            )
+            for i, colname in enumerate(obj.columns)
+        )
+        results = [func(sgb) for sgb in sgbs]
+
+        if not len(results):
+            # concat would raise
+            res_df = DataFrame([], columns=columns, index=self._grouper.result_index)
+        else:
+            res_df = concat(results, keys=columns, axis=1)
+
+        if not self.as_index:
+            res_df.index = default_index(len(res_df))
+            res_df = self._insert_inaxis_grouper(res_df)
+        return res_df
+
+    def nunique(self, dropna: bool = True) -> DataFrame:
+        """
+        Return DataFrame with counts of unique elements in each position.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't include NaN in the counts.
+
+        Returns
+        -------
+        nunique: DataFrame
+            Counts of unique elements in each position.
+
+        See Also
+        --------
+        DataFrame.nunique : Count number of distinct elements in specified axis.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "id": ["spam", "egg", "egg", "spam", "ham", "ham"],
+        ...         "value1": [1, 5, 5, 2, 5, 5],
+        ...         "value2": list("abbaxy"),
+        ...     }
+        ... )
+        >>> df
+             id  value1 value2
+        0  spam       1      a
+        1   egg       5      b
+        2   egg       5      b
+        3  spam       2      a
+        4   ham       5      x
+        5   ham       5      y
+
+        >>> df.groupby("id").nunique()
+              value1  value2
+        id
+        egg        1       1
+        ham        1       2
+        spam       2       1
+
+        Check for rows with the same id but conflicting values:
+
+        >>> df.groupby("id").filter(lambda g: (g.nunique() > 1).any())
+             id  value1 value2
+        0  spam       1      a
+        3  spam       2      a
+        4   ham       5      x
+        5   ham       5      y
+        """
+        return self._apply_to_column_groupbys(lambda sgb: sgb.nunique(dropna))
+
+    def idxmax(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        """
+        Return index of first occurrence of maximum in each group.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        DataFrame
+            Indexes of maxima in each column according to the group.
+
+        Raises
+        ------
+        ValueError
+            * If a column is empty or skipna=False and any value is NA.
+
+        See Also
+        --------
+        Series.idxmax : Return index of the maximum element.
+        DataFrame.idxmax : Indexes of maxima along the specified axis.
+
+        Notes
+        -----
+        This method is the DataFrame version of ``ndarray.argmax``.
+
+        Examples
+        --------
+        Consider a dataset containing food consumption in Argentina.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "consumption": [10.51, 103.11, 55.48],
+        ...         "co2_emissions": [37.2, 19.66, 1712],
+        ...         "food_type": ["meat", "plant", "meat"],
+        ...     },
+        ...     index=["Pork", "Wheat Products", "Beef"],
+        ... )
+
+        >>> df
+                        consumption  co2_emissions
+        Pork                  10.51         37.20
+        Wheat Products       103.11         19.66
+        Beef                  55.48       1712.00
+
+        By default, it returns the index for the maximum value in each column
+        according to the group.
+
+        >>> df.groupby("food_type").idxmax()
+                        consumption   co2_emissions
+        food_type
+        animal                 Beef            Beef
+        plant        Wheat Products  Wheat Products
+        """
+        return self._idxmax_idxmin("idxmax", numeric_only=numeric_only, skipna=skipna)
+
+    def idxmin(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        """
+        Return index of first occurrence of minimum in each group.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA values.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        DataFrame
+            Indexes of minima in each column according to the group.
+
+        Raises
+        ------
+        ValueError
+            * If a column is empty or skipna=False and any value is NA.
+
+        See Also
+        --------
+        Series.idxmin : Return index of the minimum element.
+        DataFrame.idxmin : Indexes of minima along the specified axis.
+
+        Notes
+        -----
+        This method is the DataFrame version of ``ndarray.argmin``.
+
+        Examples
+        --------
+        Consider a dataset containing food consumption in Argentina.
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "consumption": [10.51, 103.11, 55.48],
+        ...         "co2_emissions": [37.2, 19.66, 1712],
+        ...         "food_type": ["meat", "plant", "meat"],
+        ...     },
+        ...     index=["Pork", "Wheat Products", "Beef"],
+        ... )
+
+        >>> df
+                        consumption  co2_emissions
+        Pork                  10.51         37.20
+        Wheat Products       103.11         19.66
+        Beef                  55.48       1712.00
+
+        By default, it returns the index for the minimum value in each column
+        according to the group.
+
+        >>> df.groupby("food_type").idxmin()
+                        consumption   co2_emissions
+        food_type
+        animal                 Pork            Pork
+        plant        Wheat Products  Wheat Products
+        """
+        return self._idxmax_idxmin("idxmin", numeric_only=numeric_only, skipna=skipna)
+
+    boxplot = boxplot_frame_groupby
+
+    def value_counts(
+        self,
+        subset: Sequence[Hashable] | None = None,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        dropna: bool = True,
+    ) -> DataFrame | Series:
+        """
+        Return a Series or DataFrame containing counts of unique rows.
+
+        .. versionadded:: 1.4.0
+
+        Parameters
+        ----------
+        subset : list-like, optional
+            Columns to use when counting unique combinations.
+        normalize : bool, default False
+            Return proportions rather than frequencies.
+        sort : bool, default True
+            Sort by frequencies when True. When False, non-grouping columns will appear
+            in the order they occur in within groups.
+
+            .. versionchanged:: 3.0.0
+
+                In prior versions, ``sort=False`` would sort the non-grouping columns
+                by label.
+        ascending : bool, default False
+            Sort in ascending order.
+        dropna : bool, default True
+            Don't include counts of rows that contain NA values.
+
+        Returns
+        -------
+        Series or DataFrame
+            Series if the groupby ``as_index`` is True, otherwise DataFrame.
+
+        See Also
+        --------
+        Series.value_counts: Equivalent method on Series.
+        DataFrame.value_counts: Equivalent method on DataFrame.
+        SeriesGroupBy.value_counts: Equivalent method on SeriesGroupBy.
+
+        Notes
+        -----
+        - If the groupby ``as_index`` is True then the returned Series will have a
+          MultiIndex with one level per input column.
+        - If the groupby ``as_index`` is False then the returned DataFrame will have an
+          additional column with the value_counts. The column is labelled 'count' or
+          'proportion', depending on the ``normalize`` parameter.
+
+        By default, rows that contain any NA values are omitted from
+        the result.
+
+        By default, the result will be in descending order so that the
+        first element of each group is the most frequently-occurring row.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "gender": ["male", "male", "female", "male", "female", "male"],
+        ...         "education": ["low", "medium", "high", "low", "high", "low"],
+        ...         "country": ["US", "FR", "US", "FR", "FR", "FR"],
+        ...     }
+        ... )
+
+        >>> df
+                gender  education   country
+        0       male    low         US
+        1       male    medium      FR
+        2       female  high        US
+        3       male    low         FR
+        4       female  high        FR
+        5       male    low         FR
+
+        >>> df.groupby("gender").value_counts()
+        gender  education  country
+        female  high       US         1
+                           FR         1
+        male    low        FR         2
+                           US         1
+                medium     FR         1
+        Name: count, dtype: int64
+
+        >>> df.groupby("gender").value_counts(ascending=True)
+        gender  education  country
+        female  high       US         1
+                           FR         1
+        male    low        US         1
+                medium     FR         1
+                low        FR         2
+        Name: count, dtype: int64
+
+        >>> df.groupby("gender").value_counts(normalize=True)
+        gender  education  country
+        female  high       US         0.50
+                           FR         0.50
+        male    low        FR         0.50
+                           US         0.25
+                medium     FR         0.25
+        Name: proportion, dtype: float64
+
+        >>> df.groupby("gender", as_index=False).value_counts()
+           gender education country  count
+        0  female      high      US      1
+        1  female      high      FR      1
+        2    male       low      FR      2
+        3    male       low      US      1
+        4    male    medium      FR      1
+
+        >>> df.groupby("gender", as_index=False).value_counts(normalize=True)
+           gender education country  proportion
+        0  female      high      US        0.50
+        1  female      high      FR        0.50
+        2    male       low      FR        0.50
+        3    male       low      US        0.25
+        4    male    medium      FR        0.25
+        """
+        return self._value_counts(subset, normalize, sort, ascending, dropna)
+
+    def take(
+        self,
+        indices: TakeIndexer,
+        **kwargs,
+    ) -> DataFrame:
+        """
+        Return the elements in the given *positional* indices in each group.
+
+        This means that we are not indexing according to actual values in
+        the index attribute of the object. We are indexing according to the
+        actual position of the element in the object.
+
+        If a requested index does not exist for some group, this method will raise.
+        To get similar behavior that ignores indices that don't exist, see
+        :meth:`.DataFrameGroupBy.nth`.
+
+        Parameters
+        ----------
+        indices : array-like
+            An array of ints indicating which positions to take.
+
+        **kwargs
+            For compatibility with :meth:`numpy.take`. Has no effect on the
+            output.
+
+        Returns
+        -------
+        DataFrame
+            An DataFrame containing the elements taken from each group.
+
+        See Also
+        --------
+        DataFrame.take : Take elements from a Series along an axis.
+        DataFrame.loc : Select a subset of a DataFrame by labels.
+        DataFrame.iloc : Select a subset of a DataFrame by positions.
+        numpy.take : Take elements from an array along an axis.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [
+        ...         ("falcon", "bird", 389.0),
+        ...         ("parrot", "bird", 24.0),
+        ...         ("lion", "mammal", 80.5),
+        ...         ("monkey", "mammal", np.nan),
+        ...         ("rabbit", "mammal", 15.0),
+        ...     ],
+        ...     columns=["name", "class", "max_speed"],
+        ...     index=[4, 3, 2, 1, 0],
+        ... )
+        >>> df
+             name   class  max_speed
+        4  falcon    bird      389.0
+        3  parrot    bird       24.0
+        2    lion  mammal       80.5
+        1  monkey  mammal        NaN
+        0  rabbit  mammal       15.0
+        >>> gb = df.groupby([1, 1, 2, 2, 2])
+
+        Take elements at rows 0 and 1.
+
+        Note how the indices selected in the result do not correspond to
+        our input indices 0 and 1. That's because we are selecting the 0th
+        and 1st rows, not rows whose indices equal 0 and 1.
+
+        >>> gb.take([0, 1])
+               name   class  max_speed
+        1 4  falcon    bird      389.0
+          3  parrot    bird       24.0
+        2 2    lion  mammal       80.5
+          1  monkey  mammal        NaN
+
+        The order of the specified indices influences the order in the result.
+        Here, the order is swapped from the previous example.
+
+        >>> gb.take([1, 0])
+               name   class  max_speed
+        1 3  parrot    bird       24.0
+          4  falcon    bird      389.0
+        2 1  monkey  mammal        NaN
+          2    lion  mammal       80.5
+
+        We may take elements using negative integers for positive indices,
+        starting from the end of the object, just like with Python lists.
+
+        >>> gb.take([-1, -2])
+               name   class  max_speed
+        1 3  parrot    bird       24.0
+          4  falcon    bird      389.0
+        2 0  rabbit  mammal       15.0
+          1  monkey  mammal        NaN
+        """
+        result = self._op_via_apply("take", indices=indices, **kwargs)
+        return result
+
+    def skew(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> DataFrame:
+        """
+        Return unbiased skew within groups.
+
+        Normalized by N-1.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        DataFrame
+            Unbiased skew within groups.
+
+        See Also
+        --------
+        DataFrame.skew : Return unbiased skew over requested axis.
+
+        Examples
+        --------
+        >>> arrays = [
+        ...     ["falcon", "parrot", "cockatoo", "kiwi", "lion", "monkey", "rabbit"],
+        ...     ["bird", "bird", "bird", "bird", "mammal", "mammal", "mammal"],
+        ... ]
+        >>> index = pd.MultiIndex.from_arrays(arrays, names=("name", "class"))
+        >>> df = pd.DataFrame(
+        ...     {"max_speed": [389.0, 24.0, 70.0, np.nan, 80.5, 21.5, 15.0]},
+        ...     index=index,
+        ... )
+        >>> df
+                        max_speed
+        name     class
+        falcon   bird        389.0
+        parrot   bird         24.0
+        cockatoo bird         70.0
+        kiwi     bird          NaN
+        lion     mammal       80.5
+        monkey   mammal       21.5
+        rabbit   mammal       15.0
+        >>> gb = df.groupby(["class"])
+        >>> gb.skew()
+                max_speed
+        class
+        bird     1.628296
+        mammal   1.669046
+        >>> gb.skew(skipna=False)
+                max_speed
+        class
+        bird          NaN
+        mammal   1.669046
+        """
+
+        def alt(obj):
+            # This should not be reached since the cython path should raise
+            #  TypeError and not NotImplementedError.
+            raise TypeError(f"'skew' is not supported for dtype={obj.dtype}")
+
+        return self._cython_agg_general(
+            "skew", alt=alt, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    def kurt(
+        self,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> DataFrame:
+        """
+        Return unbiased kurtosis within groups.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        DataFrame
+            Unbiased kurtosis within groups.
+
+        See Also
+        --------
+        DataFrame.kurt : Return unbiased kurtosis over requested axis.
+
+        Examples
+        --------
+        >>> arrays = [
+        ...     [
+        ...         "falcon",
+        ...         "parrot",
+        ...         "cockatoo",
+        ...         "kiwi",
+        ...         "eagle",
+        ...         "lion",
+        ...         "monkey",
+        ...         "rabbit",
+        ...         "dog",
+        ...         "wolf",
+        ...     ],
+        ...     [
+        ...         "bird",
+        ...         "bird",
+        ...         "bird",
+        ...         "bird",
+        ...         "bird",
+        ...         "mammal",
+        ...         "mammal",
+        ...         "mammal",
+        ...         "mammal",
+        ...         "mammal",
+        ...     ],
+        ... ]
+        >>> index = pd.MultiIndex.from_arrays(arrays, names=("name", "class"))
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "max_speed": [
+        ...             389.0,
+        ...             24.0,
+        ...             70.0,
+        ...             np.nan,
+        ...             350.0,
+        ...             80.5,
+        ...             21.5,
+        ...             15.0,
+        ...             40.0,
+        ...             50.0,
+        ...         ]
+        ...     },
+        ...     index=index,
+        ... )
+        >>> df
+                         max_speed
+        name     class
+        falcon   bird        389.0
+        parrot   bird         24.0
+        cockatoo bird         70.0
+        kiwi     bird          NaN
+        eagle    bird        350.0
+        lion     mammal       80.5
+        monkey   mammal       21.5
+        rabbit   mammal       15.0
+        dog      mammal       40.0
+        wolf     mammal       50.0
+        >>> gb = df.groupby(["class"])
+        >>> gb.kurt()
+                max_speed
+        class
+        bird    -5.493277
+        mammal   0.204125
+        >>> gb.kurt(skipna=False)
+                max_speed
+        class
+        bird          NaN
+        mammal   0.204125
+        """
+
+        return self._cython_agg_general(
+            "kurt", alt=None, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @property
+    @doc(DataFrame.plot.__doc__)
+    def plot(self) -> GroupByPlot:
+        result = GroupByPlot(self)
+        return result
+
+    @doc(DataFrame.corr.__doc__)
+    def corr(
+        self,
+        method: str | Callable[[np.ndarray, np.ndarray], float] = "pearson",
+        min_periods: int = 1,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        result = self._op_via_apply(
+            "corr", method=method, min_periods=min_periods, numeric_only=numeric_only
+        )
+        return result
+
+    @doc(DataFrame.cov.__doc__)
+    def cov(
+        self,
+        min_periods: int | None = None,
+        ddof: int | None = 1,
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        result = self._op_via_apply(
+            "cov", min_periods=min_periods, ddof=ddof, numeric_only=numeric_only
+        )
+        return result
+
+    def hist(
+        self,
+        column: IndexLabel | None = None,
+        by=None,
+        grid: bool = True,
+        xlabelsize: int | None = None,
+        xrot: float | None = None,
+        ylabelsize: int | None = None,
+        yrot: float | None = None,
+        ax=None,
+        sharex: bool = False,
+        sharey: bool = False,
+        figsize: tuple[float, float] | None = None,
+        layout: tuple[int, int] | None = None,
+        bins: int | Sequence[int] = 10,
+        backend: str | None = None,
+        legend: bool = False,
+        **kwargs,
+    ):
+        """
+        Make a histogram of the DataFrame's columns.
+
+        A `histogram`_ is a representation of the distribution of data.
+        This function calls :meth:`matplotlib.pyplot.hist`, on each series in
+        the DataFrame, resulting in one histogram per column.
+
+        .. _histogram: https://en.wikipedia.org/wiki/Histogram
+
+        Parameters
+        ----------
+        column : str or sequence, optional
+            If passed, will be used to limit data to a subset of columns.
+        by : object, optional
+            If passed, then used to form histograms for separate groups.
+        grid : bool, default True
+            Whether to show axis grid lines.
+        xlabelsize : int, default None
+            If specified changes the x-axis label size.
+        xrot : float, default None
+            Rotation of x axis labels. For example, a value of 90 displays the
+            x labels rotated 90 degrees clockwise.
+        ylabelsize : int, default None
+            If specified changes the y-axis label size.
+        yrot : float, default None
+            Rotation of y axis labels. For example, a value of 90 displays the
+            y labels rotated 90 degrees clockwise.
+        ax : Matplotlib axes object, default None
+            The axes to plot the histogram on.
+        sharex : bool, default True if ax is None else False
+            In case subplots=True, share x axis and set some x axis labels to
+            invisible; defaults to True if ax is None otherwise False if an ax
+            is passed in.
+            Note that passing in both an ax and sharex=True will alter all x axis
+            labels for all subplots in a figure.
+        sharey : bool, default False
+            In case subplots=True, share y axis and set some y axis labels to
+            invisible.
+        figsize : tuple, optional
+            The size in inches of the figure to create. Uses the value in
+            `matplotlib.rcParams` by default.
+        layout : tuple, optional
+            Tuple of (rows, columns) for the layout of the histograms.
+        bins : int or sequence, default 10
+            Number of histogram bins to be used. If an integer is given, bins + 1
+            bin edges are calculated and returned. If bins is a sequence, gives
+            bin edges, including left edge of first bin and right edge of last
+            bin. In this case, bins is returned unmodified.
+
+        backend : str, default None
+            Backend to use instead of the backend specified in the option
+            ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+            specify the ``plotting.backend`` for the whole session, set
+            ``pd.options.plotting.backend``.
+
+        legend : bool, default False
+            Whether to show the legend.
+
+        **kwargs
+            All other plotting keyword arguments to be passed to
+            :meth:`matplotlib.pyplot.hist`.
+
+        Returns
+        -------
+        matplotlib.Axes or numpy.ndarray
+            A ``matplotlib.Axes`` object or an array of ``Axes`` objects, depending on
+            the layout and grouping.
+
+        See Also
+        --------
+        matplotlib.pyplot.hist : Plot a histogram using matplotlib.
+
+        Examples
+        --------
+        This example draws a histogram based on the length and width of
+        some animals, displayed in three bins
+
+        .. plot::
+            :context: close-figs
+
+            >>> data = {
+            ...     "length": [1.5, 0.5, 1.2, 0.9, 3],
+            ...     "width": [0.7, 0.2, 0.15, 0.2, 1.1],
+            ... }
+            >>> index = ["pig", "rabbit", "duck", "chicken", "horse"]
+            >>> df = pd.DataFrame(data, index=index)
+            >>> hist = df.groupby("length").hist(bins=3)
+        """
+        result = self._op_via_apply(
+            "hist",
+            column=column,
+            by=by,
+            grid=grid,
+            xlabelsize=xlabelsize,
+            xrot=xrot,
+            ylabelsize=ylabelsize,
+            yrot=yrot,
+            ax=ax,
+            sharex=sharex,
+            sharey=sharey,
+            figsize=figsize,
+            layout=layout,
+            bins=bins,
+            backend=backend,
+            legend=legend,
+            **kwargs,
+        )
+        return result
+
+    def corrwith(
+        self,
+        other: DataFrame | Series,
+        drop: bool = False,
+        method: CorrelationMethod = "pearson",
+        numeric_only: bool = False,
+    ) -> DataFrame:
+        """
+        Compute pairwise correlation.
+
+        .. deprecated:: 3.0.0
+
+        Pairwise correlation is computed between rows or columns of
+        DataFrame with rows or columns of Series or DataFrame. DataFrames
+        are first aligned along both axes before computing the
+        correlations.
+
+        Parameters
+        ----------
+        other : DataFrame, Series
+            Object with which to compute correlations.
+        drop : bool, default False
+            Drop missing indices from result.
+        method : {'pearson', 'kendall', 'spearman'} or callable
+            Method of correlation:
+
+            * pearson : standard correlation coefficient
+            * kendall : Kendall Tau correlation coefficient
+            * spearman : Spearman rank correlation
+            * callable: callable with input two 1d ndarrays
+                and returning a float.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+                The default value of ``numeric_only`` is now ``False``.
+
+        Returns
+        -------
+        Series
+            Pairwise correlations.
+
+        See Also
+        --------
+        DataFrame.corr : Compute pairwise correlation of columns.
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame(
+        ...     {
+        ...         "Day": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+        ...         "Data": [6, 6, 8, 5, 4, 2, 7, 3, 9],
+        ...     }
+        ... )
+        >>> df2 = pd.DataFrame(
+        ...     {
+        ...         "Day": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+        ...         "Data": [5, 3, 8, 3, 1, 1, 2, 3, 6],
+        ...     }
+        ... )
+
+        >>> df1.groupby("Day").corrwith(df2)
+                 Data  Day
+        Day
+        1    0.917663  NaN
+        2    0.755929  NaN
+        3    0.576557  NaN
+        """
+        warnings.warn(
+            "DataFrameGroupBy.corrwith is deprecated",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        result = self._op_via_apply(
+            "corrwith",
+            other=other,
+            drop=drop,
+            method=method,
+            numeric_only=numeric_only,
+        )
+        return result
+
+
+def _wrap_transform_general_frame(
+    obj: DataFrame, group: DataFrame, res: DataFrame | Series
+) -> DataFrame:
+    from pandas import concat
+
+    if isinstance(res, Series):
+        # we need to broadcast across the
+        # other dimension; this will preserve dtypes
+        # GH14457
+        if res.index.is_(obj.index):
+            res_frame = concat([res] * len(group.columns), axis=1, ignore_index=True)
+            res_frame.columns = group.columns
+            res_frame.index = group.index
+        else:
+            res_frame = obj._constructor(
+                np.tile(res.values, (len(group.index), 1)),
+                columns=group.columns,
+                index=group.index,
+            )
+        assert isinstance(res_frame, DataFrame)
+        return res_frame
+    elif isinstance(res, DataFrame) and not res.index.is_(group.index):
+        return res._align_frame(group)[0]
+    else:
+        return res
diff --git a/pandas/pandas/core/groupby/groupby.py b/pandas/pandas/core/groupby/groupby.py
new file mode 100644
index 0000000000000..d0c0ed29b6d44
--- /dev/null
+++ b/pandas/pandas/core/groupby/groupby.py
@@ -0,0 +1,5661 @@
+"""
+Provide the groupby split-apply-combine paradigm. Define the GroupBy
+class providing the base-class of operations.
+
+The SeriesGroupBy and DataFrameGroupBy sub-class
+(defined in pandas.core.groupby.generic)
+expose these user-facing objects to provide specific functionality.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    Iterator,
+    Mapping,
+    Sequence,
+)
+import datetime
+from functools import (
+    partial,
+    wraps,
+)
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    TypeVar,
+    Union,
+    cast,
+    final,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    Timestamp,
+    lib,
+)
+from pandas._libs.algos import rank_1d
+import pandas._libs.groupby as libgroupby
+from pandas._libs.missing import NA
+from pandas._typing import (
+    AnyArrayLike,
+    ArrayLike,
+    DtypeObj,
+    IndexLabel,
+    IntervalClosedType,
+    NDFrameT,
+    PositionalIndexer,
+    RandomState,
+    npt,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    AbstractMethodError,
+    DataError,
+)
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    cache_readonly,
+    doc,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import (
+    coerce_indexer_dtype,
+    ensure_dtype_can_hold_na,
+)
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_float_dtype,
+    is_hashable,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_scalar,
+    needs_i8_conversion,
+    pandas_dtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+    notna,
+)
+
+from pandas.core import (
+    algorithms,
+    sample,
+)
+from pandas.core._numba import executor
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BaseMaskedArray,
+    ExtensionArray,
+    FloatingArray,
+    IntegerArray,
+    SparseArray,
+)
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.arrays.string_arrow import (
+    ArrowStringArray,
+    ArrowStringArrayNumpySemantics,
+)
+from pandas.core.base import (
+    PandasObject,
+    SelectionMixin,
+)
+import pandas.core.common as com
+from pandas.core.frame import DataFrame
+from pandas.core.generic import NDFrame
+from pandas.core.groupby import (
+    base,
+    numba_,
+    ops,
+)
+from pandas.core.groupby.grouper import get_grouper
+from pandas.core.groupby.indexing import (
+    GroupByIndexingMixin,
+    GroupByNthSelector,
+)
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    default_index,
+)
+from pandas.core.internals.blocks import ensure_block_shape
+from pandas.core.series import Series
+from pandas.core.sorting import get_group_index_sorter
+from pandas.core.util.numba_ import (
+    get_jit_arguments,
+    maybe_use_numba,
+    prepare_function_arguments,
+)
+
+if TYPE_CHECKING:
+    from pandas._libs.tslibs import BaseOffset
+    from pandas._typing import (
+        Any,
+        Concatenate,
+        P,
+        Self,
+        T,
+    )
+
+    from pandas.core.indexers.objects import BaseIndexer
+    from pandas.core.resample import Resampler
+    from pandas.core.window import (
+        ExpandingGroupby,
+        ExponentialMovingWindowGroupby,
+        RollingGroupby,
+    )
+
+_common_see_also = """
+        See Also
+        --------
+        Series.%(name)s : Apply a function %(name)s to a Series.
+        DataFrame.%(name)s : Apply a function %(name)s
+            to each row or column of a DataFrame.
+"""
+
+_groupby_agg_method_engine_template = """
+Compute {fname} of group values.
+
+Parameters
+----------
+numeric_only : bool, default {no}
+    Include only float, int, boolean columns.
+
+    .. versionchanged:: 2.0.0
+
+        numeric_only no longer accepts ``None``.
+
+min_count : int, default {mc}
+    The required number of valid values to perform the operation. If fewer
+    than ``min_count`` non-NA values are present the result will be NA.
+
+engine : str, default None {e}
+    * ``'cython'`` : Runs rolling apply through C-extensions from cython.
+    * ``'numba'`` : Runs rolling apply through JIT compiled code from numba.
+        Only available when ``raw`` is set to ``True``.
+    * ``None`` : Defaults to ``'cython'`` or globally setting ``compute.use_numba``
+
+engine_kwargs : dict, default None {ek}
+    * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+    * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+        and ``parallel`` dictionary keys. The values must either be ``True`` or
+        ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+        ``{{'nopython': True, 'nogil': False, 'parallel': False}}`` and will be
+        applied to both the ``func`` and the ``apply`` groupby aggregation.
+
+Returns
+-------
+Series or DataFrame
+    Computed {fname} of values within each group.
+
+See Also
+--------
+SeriesGroupBy.min : Return the min of the group values.
+DataFrameGroupBy.min : Return the min of the group values.
+SeriesGroupBy.max : Return the max of the group values.
+DataFrameGroupBy.max : Return the max of the group values.
+SeriesGroupBy.sum : Return the sum of the group values.
+DataFrameGroupBy.sum : Return the sum of the group values.
+
+Examples
+--------
+{example}
+"""
+
+_groupby_agg_method_skipna_engine_template = """
+Compute {fname} of group values.
+
+Parameters
+----------
+numeric_only : bool, default {no}
+    Include only float, int, boolean columns.
+
+    .. versionchanged:: 2.0.0
+
+        numeric_only no longer accepts ``None``.
+
+min_count : int, default {mc}
+    The required number of valid values to perform the operation. If fewer
+    than ``min_count`` non-NA values are present the result will be NA.
+
+skipna : bool, default {s}
+    Exclude NA/null values. If the entire group is NA and ``skipna`` is
+    ``True``, the result will be NA.
+
+    .. versionchanged:: 3.0.0
+
+engine : str, default None {e}
+    * ``'cython'`` : Runs rolling apply through C-extensions from cython.
+    * ``'numba'`` : Runs rolling apply through JIT compiled code from numba.
+        Only available when ``raw`` is set to ``True``.
+    * ``None`` : Defaults to ``'cython'`` or globally setting ``compute.use_numba``
+
+engine_kwargs : dict, default None {ek}
+    * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+    * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+        and ``parallel`` dictionary keys. The values must either be ``True`` or
+        ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+        ``{{'nopython': True, 'nogil': False, 'parallel': False}}`` and will be
+        applied to both the ``func`` and the ``apply`` groupby aggregation.
+
+Returns
+-------
+Series or DataFrame
+    Computed {fname} of values within each group.
+
+See Also
+--------
+SeriesGroupBy.min : Return the min of the group values.
+DataFrameGroupBy.min : Return the min of the group values.
+SeriesGroupBy.max : Return the max of the group values.
+DataFrameGroupBy.max : Return the max of the group values.
+SeriesGroupBy.sum : Return the sum of the group values.
+DataFrameGroupBy.sum : Return the sum of the group values.
+
+Examples
+--------
+{example}
+"""
+
+_pipe_template = """
+Apply a ``func`` with arguments to this %(klass)s object and return its result.
+
+Use `.pipe` when you want to improve readability by chaining together
+functions that expect Series, DataFrames, GroupBy or Resampler objects.
+Instead of writing
+
+>>> h = lambda x, arg2, arg3: x + 1 - arg2 * arg3
+>>> g = lambda x, arg1: x * 5 / arg1
+>>> f = lambda x: x ** 4
+>>> df = pd.DataFrame([["a", 4], ["b", 5]], columns=["group", "value"])
+>>> h(g(f(df.groupby('group')), arg1=1), arg2=2, arg3=3)  # doctest: +SKIP
+
+You can write
+
+>>> (df.groupby('group')
+...    .pipe(f)
+...    .pipe(g, arg1=1)
+...    .pipe(h, arg2=2, arg3=3))  # doctest: +SKIP
+
+which is much more readable.
+
+Parameters
+----------
+func : callable or tuple of (callable, str)
+    Function to apply to this %(klass)s object or, alternatively,
+    a `(callable, data_keyword)` tuple where `data_keyword` is a
+    string indicating the keyword of `callable` that expects the
+    %(klass)s object.
+*args : iterable, optional
+       Positional arguments passed into `func`.
+**kwargs : dict, optional
+         A dictionary of keyword arguments passed into `func`.
+
+Returns
+-------
+%(klass)s
+    The original object with the function `func` applied.
+
+See Also
+--------
+Series.pipe : Apply a function with arguments to a series.
+DataFrame.pipe: Apply a function with arguments to a dataframe.
+apply : Apply function to each group instead of to the
+    full %(klass)s object.
+
+Notes
+-----
+See more `here
+<https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html#piping-function-calls>`_
+
+Examples
+--------
+%(examples)s
+"""
+
+_transform_template = """
+Call function producing a same-indexed %(klass)s on each group.
+
+Returns a %(klass)s having the same indexes as the original object
+filled with the transformed values.
+
+Parameters
+----------
+func : function, str
+    Function to apply to each group. See the Notes section below for requirements.
+
+    Accepted inputs are:
+
+    - String
+    - Python function
+    - Numba JIT function with ``engine='numba'`` specified.
+
+    Only passing a single function is supported with this engine.
+    If the ``'numba'`` engine is chosen, the function must be
+    a user defined function with ``values`` and ``index`` as the
+    first and second arguments respectively in the function signature.
+    Each group's index will be passed to the user defined function
+    and optionally available for use.
+
+    If a string is chosen, then it needs to be the name
+    of the groupby method you want to use.
+*args
+    Positional arguments to pass to func.
+engine : str, default None
+    * ``'cython'`` : Runs the function through C-extensions from cython.
+    * ``'numba'`` : Runs the function through JIT compiled code from numba.
+    * ``None`` : Defaults to ``'cython'`` or the global setting ``compute.use_numba``
+
+engine_kwargs : dict, default None
+    * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+    * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+      and ``parallel`` dictionary keys. The values must either be ``True`` or
+      ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+      ``{'nopython': True, 'nogil': False, 'parallel': False}`` and will be
+      applied to the function
+
+**kwargs
+    Keyword arguments to be passed into func.
+
+Returns
+-------
+%(klass)s
+    %(klass)s with the same indexes as the original object filled
+    with transformed values.
+
+See Also
+--------
+%(klass)s.groupby.apply : Apply function ``func`` group-wise and combine
+    the results together.
+%(klass)s.groupby.aggregate : Aggregate using one or more operations.
+%(klass)s.transform : Call ``func`` on self producing a %(klass)s with the
+    same axis shape as self.
+
+Notes
+-----
+Each group is endowed the attribute 'name' in case you need to know
+which group you are working on.
+
+The current implementation imposes three requirements on f:
+
+* f must return a value that either has the same shape as the input
+  subframe or can be broadcast to the shape of the input subframe.
+  For example, if `f` returns a scalar it will be broadcast to have the
+  same shape as the input subframe.
+* if this is a DataFrame, f must support application column-by-column
+  in the subframe. If f also supports application to the entire subframe,
+  then a fast path is used starting from the second chunk.
+* f must not mutate groups. Mutation is not supported and may
+  produce unexpected results. See :ref:`gotchas.udf-mutation` for more details.
+
+When using ``engine='numba'``, there will be no "fall back" behavior internally.
+The group data and group index will be passed as numpy arrays to the JITed
+user defined function, and no alternative execution attempts will be tried.
+
+.. versionchanged:: 1.3.0
+
+    The resulting dtype will reflect the return value of the passed ``func``,
+    see the examples below.
+
+.. versionchanged:: 2.0.0
+
+    When using ``.transform`` on a grouped DataFrame and the transformation function
+    returns a DataFrame, pandas now aligns the result's index
+    with the input's index. You can call ``.to_numpy()`` on the
+    result of the transformation function to avoid alignment.
+
+Examples
+--------
+%(example)s"""
+
+
+@final
+class GroupByPlot(PandasObject):
+    """
+    Class implementing the .plot attribute for groupby objects.
+    """
+
+    def __init__(self, groupby: GroupBy) -> None:
+        self._groupby = groupby
+
+    def __call__(self, *args, **kwargs):
+        def f(self):
+            return self.plot(*args, **kwargs)
+
+        f.__name__ = "plot"
+        return self._groupby._python_apply_general(f, self._groupby._selected_obj)
+
+    def __getattr__(self, name: str):
+        def attr(*args, **kwargs):
+            def f(self):
+                return getattr(self.plot, name)(*args, **kwargs)
+
+            return self._groupby._python_apply_general(f, self._groupby._selected_obj)
+
+        return attr
+
+
+_KeysArgType = Union[
+    Hashable,
+    list[Hashable],
+    Callable[[Hashable], Hashable],
+    list[Callable[[Hashable], Hashable]],
+    Mapping[Hashable, Hashable],
+]
+
+
+class BaseGroupBy(PandasObject, SelectionMixin[NDFrameT], GroupByIndexingMixin):
+    _hidden_attrs = PandasObject._hidden_attrs | {
+        "as_index",
+        "dropna",
+        "exclusions",
+        "grouper",
+        "group_keys",
+        "keys",
+        "level",
+        "obj",
+        "observed",
+        "sort",
+    }
+
+    _grouper: ops.BaseGrouper
+    keys: _KeysArgType | None = None
+    level: IndexLabel | None = None
+    group_keys: bool
+
+    @final
+    def __len__(self) -> int:
+        return self._grouper.ngroups
+
+    @final
+    def __repr__(self) -> str:
+        # TODO: Better repr for GroupBy object
+        return object.__repr__(self)
+
+    @final
+    @property
+    def groups(self) -> dict[Hashable, Index]:
+        """
+        Dict {group name -> group labels}.
+
+        This property provides a dictionary representation of the groupings formed
+        during a groupby operation, where each key represents a unique group value from
+        the specified column(s), and each value is a list of index labels
+        that belong to that group.
+
+        See Also
+        --------
+        core.groupby.DataFrameGroupBy.get_group : Retrieve group from a
+            ``DataFrameGroupBy`` object with provided name.
+        core.groupby.SeriesGroupBy.get_group : Retrieve group from a
+            ``SeriesGroupBy`` object with provided name.
+        core.resample.Resampler.get_group : Retrieve group from a
+            ``Resampler`` object with provided name.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 3], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        dtype: int64
+        >>> ser.groupby(level=0).groups
+        {'a': ['a', 'a'], 'b': ['b']}
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(data, columns=["a", "b", "c"])
+        >>> df
+           a  b  c
+        0  1  2  3
+        1  1  5  6
+        2  7  8  9
+        >>> df.groupby(by="a").groups
+        {1: [0, 1], 7: [2]}
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").groups
+        {Timestamp('2023-01-01 00:00:00'): 2, Timestamp('2023-02-01 00:00:00'): 4}
+        """
+        if isinstance(self.keys, list) and len(self.keys) == 1:
+            warnings.warn(
+                "`groups` by one element list returns scalar is deprecated "
+                "and will be removed. In a future version `groups` by one element "
+                "list will return tuple. Use ``df.groupby(by='a').groups`` "
+                "instead of ``df.groupby(by=['a']).groups`` to avoid this warning",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        return self._grouper.groups
+
+    @final
+    @property
+    def ngroups(self) -> int:
+        return self._grouper.ngroups
+
+    @final
+    @property
+    def indices(self) -> dict[Hashable, npt.NDArray[np.intp]]:
+        """
+        Dict {group name -> group indices}.
+
+        The dictionary keys represent the group labels (e.g., timestamps for a
+        time-based resampling operation), and the values are arrays of integer
+        positions indicating where the elements of each group are located in the
+        original data. This property is particularly useful when working with
+        resampled data, as it provides insight into how the original time-series data
+        has been grouped.
+
+        See Also
+        --------
+        core.groupby.DataFrameGroupBy.indices : Provides a mapping of group rows to
+            positions of the elements.
+        core.groupby.SeriesGroupBy.indices : Provides a mapping of group rows to
+            positions of the elements.
+        core.resample.Resampler.indices : Provides a mapping of group rows to
+            positions of the elements.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 3], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        dtype: int64
+        >>> ser.groupby(level=0).indices
+        {'a': array([0, 1]), 'b': array([2])}
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["owl", "toucan", "eagle"]
+        ... )
+        >>> df
+                a  b  c
+        owl     1  2  3
+        toucan  1  5  6
+        eagle   7  8  9
+        >>> df.groupby(by=["a"]).indices
+        {1: array([0, 1]), 7: array([2])}
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").indices
+        defaultdict(<class 'list'>, {Timestamp('2023-01-01 00:00:00'): [0, 1],
+        Timestamp('2023-02-01 00:00:00'): [2, 3]})
+        """
+        return self._grouper.indices
+
+    @final
+    def _get_indices(self, names):
+        """
+        Safe get multiple indices, translate keys for
+        datelike to underlying repr.
+        """
+
+        def get_converter(s):
+            # possibly convert to the actual key types
+            # in the indices, could be a Timestamp or a np.datetime64
+            if isinstance(s, datetime.datetime):
+                return lambda key: Timestamp(key)
+            elif isinstance(s, np.datetime64):
+                return lambda key: Timestamp(key).asm8
+            else:
+                return lambda key: key
+
+        if len(names) == 0:
+            return []
+
+        if len(self.indices) > 0:
+            index_sample = next(iter(self.indices))
+        else:
+            index_sample = None  # Dummy sample
+
+        name_sample = names[0]
+        if isinstance(index_sample, tuple):
+            if not isinstance(name_sample, tuple):
+                msg = "must supply a tuple to get_group with multiple grouping keys"
+                raise ValueError(msg)
+            if not len(name_sample) == len(index_sample):
+                try:
+                    # If the original grouper was a tuple
+                    return [self.indices[name] for name in names]
+                except KeyError as err:
+                    # turns out it wasn't a tuple
+                    msg = (
+                        "must supply a same-length tuple to get_group "
+                        "with multiple grouping keys"
+                    )
+                    raise ValueError(msg) from err
+
+            converters = (get_converter(s) for s in index_sample)
+            names = (tuple(f(n) for f, n in zip(converters, name)) for name in names)
+
+        else:
+            converter = get_converter(index_sample)
+            names = (converter(name) for name in names)
+
+        return [self.indices.get(name, []) for name in names]
+
+    @final
+    def _get_index(self, name):
+        """
+        Safe get index, translate keys for datelike to underlying repr.
+        """
+        return self._get_indices([name])[0]
+
+    @final
+    @cache_readonly
+    def _selected_obj(self):
+        # Note: _selected_obj is always just `self.obj` for SeriesGroupBy
+        if isinstance(self.obj, Series):
+            return self.obj
+
+        if self._selection is not None:
+            if is_hashable(self._selection):
+                # i.e. a single key, so selecting it will return a Series.
+                #  In this case, _obj_with_exclusions would wrap the key
+                #  in a list and return a single-column DataFrame.
+                return self.obj[self._selection]
+
+            # Otherwise _selection is equivalent to _selection_list, so
+            #  _selected_obj matches _obj_with_exclusions, so we can reuse
+            #  that and avoid making a copy.
+            return self._obj_with_exclusions
+
+        return self.obj
+
+    @final
+    def _dir_additions(self) -> set[str]:
+        return self.obj._dir_additions()
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    @Substitution(
+        klass="GroupBy",
+        examples=dedent(
+            """\
+        >>> df = pd.DataFrame({'A': 'a b a b'.split(), 'B': [1, 2, 3, 4]})
+        >>> df
+           A  B
+        0  a  1
+        1  b  2
+        2  a  3
+        3  b  4
+
+        To get the difference between each groups maximum and minimum value in one
+        pass, you can do
+
+        >>> df.groupby('A').pipe(lambda x: x.max() - x.min())
+           B
+        A
+        a  2
+        b  2"""
+        ),
+    )
+    @Appender(_pipe_template)
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        return com.pipe(self, func, *args, **kwargs)
+
+    @final
+    def get_group(self, name) -> DataFrame | Series:
+        """
+        Construct DataFrame from group with provided name.
+
+        Parameters
+        ----------
+        name : object
+            The name of the group to get as a DataFrame.
+
+        Returns
+        -------
+        Series or DataFrame
+            Get the respective Series or DataFrame corresponding to the group provided.
+
+        See Also
+        --------
+        DataFrameGroupBy.groups: Dictionary representation of the groupings formed
+            during a groupby operation.
+        DataFrameGroupBy.indices: Provides a mapping of group rows to positions
+            of the elements.
+        SeriesGroupBy.groups: Dictionary representation of the groupings formed
+            during a groupby operation.
+        SeriesGroupBy.indices: Provides a mapping of group rows to positions
+            of the elements.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 3], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        dtype: int64
+        >>> ser.groupby(level=0).get_group("a")
+        a    1
+        a    2
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["owl", "toucan", "eagle"]
+        ... )
+        >>> df
+                a  b  c
+        owl     1  2  3
+        toucan  1  5  6
+        eagle   7  8  9
+        >>> df.groupby(by=["a"]).get_group((1,))
+                a  b  c
+        owl     1  2  3
+        toucan  1  5  6
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").get_group("2023-01-01")
+        2023-01-01    1
+        2023-01-15    2
+        dtype: int64
+        """
+        keys = self.keys
+        level = self.level
+        # mypy doesn't recognize level/keys as being sized when passed to len
+        if (is_list_like(level) and len(level) == 1) or (  # type: ignore[arg-type]
+            is_list_like(keys) and len(keys) == 1  # type: ignore[arg-type]
+        ):
+            # GH#25971
+            if isinstance(name, tuple) and len(name) == 1:
+                name = name[0]
+            else:
+                raise KeyError(name)
+
+        inds = self._get_index(name)
+        if not len(inds):
+            raise KeyError(name)
+        return self._selected_obj.iloc[inds]
+
+    @final
+    def __iter__(self) -> Iterator[tuple[Hashable, NDFrameT]]:
+        """
+        Groupby iterator.
+
+        This method provides an iterator over the groups created by the ``resample``
+        or ``groupby`` operation on the object. The method yields tuples where
+        the first element is the label (group key) corresponding to each group or
+        resampled bin, and the second element is the subset of the data that falls
+        within that group or bin.
+
+        Returns
+        -------
+        Iterator
+            Generator yielding a sequence of (name, subsetted object)
+            for each group.
+
+        See Also
+        --------
+        Series.groupby : Group data by a specific key or column.
+        DataFrame.groupby : Group DataFrame using mapper or by columns.
+        DataFrame.resample : Resample a DataFrame.
+        Series.resample : Resample a Series.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 3], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        dtype: int64
+        >>> for x, y in ser.groupby(level=0):
+        ...     print(f"{x}\\n{y}\\n")
+        a
+        a    1
+        a    2
+        dtype: int64
+        b
+        b    3
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(data, columns=["a", "b", "c"])
+        >>> df
+           a  b  c
+        0  1  2  3
+        1  1  5  6
+        2  7  8  9
+        >>> for x, y in df.groupby(by=["a"]):
+        ...     print(f"{x}\\n{y}\\n")
+        (1,)
+           a  b  c
+        0  1  2  3
+        1  1  5  6
+        (7,)
+           a  b  c
+        2  7  8  9
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> for x, y in ser.resample("MS"):
+        ...     print(f"{x}\\n{y}\\n")
+        2023-01-01 00:00:00
+        2023-01-01    1
+        2023-01-15    2
+        dtype: int64
+        2023-02-01 00:00:00
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        """
+        keys = self.keys
+        level = self.level
+        result = self._grouper.get_iterator(self._selected_obj)
+        # mypy: Argument 1 to "len" has incompatible type "Hashable"; expected "Sized"
+        if (
+            (is_list_like(level) and len(level) == 1)  # type: ignore[arg-type]
+            or (isinstance(keys, list) and len(keys) == 1)
+        ):
+            # GH#42795 - when keys is a list, return tuples even when length is 1
+            result = (((key,), group) for key, group in result)
+        return result
+
+
+# To track operations that expand dimensions, like ohlc
+OutputFrameOrSeries = TypeVar("OutputFrameOrSeries", bound=NDFrame)
+
+
+class GroupBy(BaseGroupBy[NDFrameT]):
+    """
+    Class for grouping and aggregating relational data.
+
+    See aggregate, transform, and apply functions on this object.
+
+    It's easiest to use obj.groupby(...) to use GroupBy, but you can also do:
+
+    ::
+
+        grouped = groupby(obj, ...)
+
+    Parameters
+    ----------
+    obj : pandas object
+    level : int, default None
+        Level of MultiIndex
+    groupings : list of Grouping objects
+        Most users should ignore this
+    exclusions : array-like, optional
+        List of columns to exclude
+    name : str
+        Most users should ignore this
+
+    Returns
+    -------
+    **Attributes**
+    groups : dict
+        {group name -> group labels}
+    len(grouped) : int
+        Number of groups
+
+    Notes
+    -----
+    After grouping, see aggregate, apply, and transform functions. Here are
+    some other brief notes about usage. When grouping by multiple groups, the
+    result index will be a MultiIndex (hierarchical) by default.
+
+    Iteration produces (key, group) tuples, i.e. chunking the data by group. So
+    you can write code like:
+
+    ::
+
+        grouped = obj.groupby(keys)
+        for key, group in grouped:
+            # do something with the data
+
+    Function calls on GroupBy, if not specially implemented, "dispatch" to the
+    grouped data. So if you group a DataFrame and wish to invoke the std()
+    method on each group, you can simply do:
+
+    ::
+
+        df.groupby(mapper).std()
+
+    rather than
+
+    ::
+
+        df.groupby(mapper).aggregate(np.std)
+
+    You can pass arguments to these "wrapped" functions, too.
+
+    See the online documentation for full exposition on these topics and much
+    more
+    """
+
+    _grouper: ops.BaseGrouper
+    as_index: bool
+
+    @final
+    def __init__(
+        self,
+        obj: NDFrameT,
+        keys: _KeysArgType | None = None,
+        level: IndexLabel | None = None,
+        grouper: ops.BaseGrouper | None = None,
+        exclusions: frozenset[Hashable] | None = None,
+        selection: IndexLabel | None = None,
+        as_index: bool = True,
+        sort: bool = True,
+        group_keys: bool = True,
+        observed: bool = False,
+        dropna: bool = True,
+    ) -> None:
+        self._selection = selection
+
+        assert isinstance(obj, NDFrame), type(obj)
+
+        self.level = level
+        self.as_index = as_index
+        self.keys = keys
+        self.sort = sort
+        self.group_keys = group_keys
+        self.dropna = dropna
+
+        if grouper is None:
+            grouper, exclusions, obj = get_grouper(
+                obj,
+                keys,
+                level=level,
+                sort=sort,
+                observed=observed,
+                dropna=self.dropna,
+            )
+
+        self.observed = observed
+        self.obj = obj
+        self._grouper = grouper
+        self.exclusions = frozenset(exclusions) if exclusions else frozenset()
+
+    def __getattr__(self, attr: str):
+        if attr in self._internal_names_set:
+            return object.__getattribute__(self, attr)
+        if attr in self.obj:
+            return self[attr]
+
+        raise AttributeError(
+            f"'{type(self).__name__}' object has no attribute '{attr}'"
+        )
+
+    @final
+    def _op_via_apply(self, name: str, *args, **kwargs):
+        """Compute the result of an operation by using GroupBy's apply."""
+        f = getattr(type(self._obj_with_exclusions), name)
+
+        def curried(x):
+            return f(x, *args, **kwargs)
+
+        # preserve the name so we can detect it when calling plot methods,
+        # to avoid duplicates
+        curried.__name__ = name
+
+        # special case otherwise extra plots are created when catching the
+        # exception below
+        if name in base.plotting_methods:
+            return self._python_apply_general(curried, self._selected_obj)
+
+        is_transform = name in base.transformation_kernels
+        result = self._python_apply_general(
+            curried,
+            self._obj_with_exclusions,
+            is_transform=is_transform,
+            not_indexed_same=not is_transform,
+        )
+
+        if self._grouper.has_dropped_na and is_transform:
+            # result will have dropped rows due to nans, fill with null
+            # and ensure index is ordered same as the input
+            result = self._set_result_index_ordered(result)
+        return result
+
+    # -----------------------------------------------------------------
+    # Dispatch/Wrapping
+
+    @final
+    def _concat_objects(
+        self,
+        values,
+        not_indexed_same: bool = False,
+        is_transform: bool = False,
+    ):
+        from pandas.core.reshape.concat import concat
+
+        if self.group_keys and not is_transform:
+            if self.as_index:
+                # possible MI return case
+                group_keys = self._grouper.result_index
+                group_levels = self._grouper.levels
+                group_names = self._grouper.names
+
+                result = concat(
+                    values,
+                    axis=0,
+                    keys=group_keys,
+                    levels=group_levels,
+                    names=group_names,
+                    sort=False,
+                )
+            else:
+                result = concat(values, axis=0)
+
+        elif not not_indexed_same:
+            result = concat(values, axis=0)
+
+            ax = self._selected_obj.index
+            if self.dropna:
+                labels = self._grouper.ids
+                mask = labels != -1
+                ax = ax[mask]
+
+            # this is a very unfortunate situation
+            # we can't use reindex to restore the original order
+            # when the ax has duplicates
+            # so we resort to this
+            # GH 14776, 30667
+            # TODO: can we reuse e.g. _reindex_non_unique?
+            if ax.has_duplicates and not result.axes[0].equals(ax):
+                # e.g. test_category_order_transformer
+                target = algorithms.unique1d(ax._values)
+                indexer, _ = result.index.get_indexer_non_unique(target)
+                result = result.take(indexer, axis=0)
+            else:
+                result = result.reindex(ax, axis=0)
+
+        else:
+            result = concat(values, axis=0)
+
+        if self.obj.ndim == 1:
+            name = self.obj.name
+        elif is_hashable(self._selection):
+            name = self._selection
+        else:
+            name = None
+
+        if isinstance(result, Series) and name is not None:
+            result.name = name
+
+        return result
+
+    @final
+    def _set_result_index_ordered(
+        self, result: OutputFrameOrSeries
+    ) -> OutputFrameOrSeries:
+        # set the result index on the passed values object and
+        # return the new object, xref 8046
+
+        index = self.obj.index
+
+        if self._grouper.is_monotonic and not self._grouper.has_dropped_na:
+            # shortcut if we have an already ordered grouper
+            result = result.set_axis(index, axis=0)
+            return result
+
+        # row order is scrambled => sort the rows by position in original index
+        original_positions = Index(self._grouper.result_ilocs)
+        result = result.set_axis(original_positions, axis=0)
+        result = result.sort_index(axis=0)
+        if self._grouper.has_dropped_na:
+            # Add back in any missing rows due to dropna - index here is integral
+            # with values referring to the row of the input so can use RangeIndex
+            result = result.reindex(default_index(len(index)), axis=0)
+        result = result.set_axis(index, axis=0)
+
+        return result
+
+    @final
+    def _insert_inaxis_grouper(
+        self, result: Series | DataFrame, qs: npt.NDArray[np.float64] | None = None
+    ) -> DataFrame:
+        if isinstance(result, Series):
+            result = result.to_frame()
+
+        n_groupings = len(self._grouper.groupings)
+
+        if qs is not None:
+            result.insert(
+                0, f"level_{n_groupings}", np.tile(qs, len(result) // len(qs))
+            )
+
+        # zip in reverse so we can always insert at loc 0
+        for level, (name, lev) in enumerate(
+            zip(
+                reversed(self._grouper.names),
+                self._grouper.get_group_levels(),
+            )
+        ):
+            if name is None:
+                # Behave the same as .reset_index() when a level is unnamed
+                name = (
+                    "index"
+                    if n_groupings == 1 and qs is None
+                    else f"level_{n_groupings - level - 1}"
+                )
+
+            # GH #28549
+            # When using .apply(-), name will be in columns already
+            if name not in result.columns:
+                # if in_axis:
+                if qs is None:
+                    result.insert(0, name, lev)
+                else:
+                    result.insert(0, name, Index(np.repeat(lev, len(qs))))
+
+        return result
+
+    @final
+    def _wrap_aggregated_output(
+        self,
+        result: Series | DataFrame,
+        qs: npt.NDArray[np.float64] | None = None,
+    ):
+        """
+        Wraps the output of GroupBy aggregations into the expected result.
+
+        Parameters
+        ----------
+        result : Series, DataFrame
+
+        Returns
+        -------
+        Series or DataFrame
+        """
+        # ATM we do not get here for SeriesGroupBy; when we do, we will
+        #  need to require that result.name already match self.obj.name
+
+        if not self.as_index:
+            # `not self.as_index` is only relevant for DataFrameGroupBy,
+            #   enforced in __init__
+            result = self._insert_inaxis_grouper(result, qs=qs)
+            result = result._consolidate()
+            result.index = default_index(len(result))
+
+        else:
+            index = self._grouper.result_index
+            if qs is not None:
+                # We get here with len(qs) != 1 and not self.as_index
+                #  in test_pass_args_kwargs
+                index = _insert_quantile_level(index, qs)
+            result.index = index
+
+        return result
+
+    def _wrap_applied_output(
+        self,
+        data,
+        values: list,
+        not_indexed_same: bool = False,
+        is_transform: bool = False,
+    ):
+        raise AbstractMethodError(self)
+
+    # -----------------------------------------------------------------
+    # numba
+
+    @final
+    def _numba_prep(self, data: DataFrame):
+        ngroups = self._grouper.ngroups
+        sorted_index = self._grouper.result_ilocs
+        sorted_ids = self._grouper._sorted_ids
+
+        sorted_data = data.take(sorted_index, axis=0).to_numpy()
+        # GH 46867
+        index_data = data.index
+        if isinstance(index_data, MultiIndex):
+            if len(self._grouper.groupings) > 1:
+                raise NotImplementedError(
+                    "Grouping with more than 1 grouping labels and "
+                    "a MultiIndex is not supported with engine='numba'"
+                )
+            group_key = self._grouper.groupings[0].name
+            index_data = index_data.get_level_values(group_key)
+        sorted_index_data = index_data.take(sorted_index).to_numpy()
+
+        starts, ends = lib.generate_slices(sorted_ids, ngroups)
+        return (
+            starts,
+            ends,
+            sorted_index_data,
+            sorted_data,
+        )
+
+    def _numba_agg_general(
+        self,
+        func: Callable,
+        dtype_mapping: dict[np.dtype, Any],
+        engine_kwargs: dict[str, bool] | None,
+        **aggregator_kwargs,
+    ):
+        """
+        Perform groupby with a standard numerical aggregation function (e.g. mean)
+        with Numba.
+        """
+        if not self.as_index:
+            raise NotImplementedError(
+                "as_index=False is not supported. Use .reset_index() instead."
+            )
+
+        data = self._obj_with_exclusions
+        df = data if data.ndim == 2 else data.to_frame()
+
+        aggregator = executor.generate_shared_aggregator(
+            func,
+            dtype_mapping,
+            True,  # is_grouped_kernel
+            **get_jit_arguments(engine_kwargs),
+        )
+        # Pass group ids to kernel directly if it can handle it
+        # (This is faster since it doesn't require a sort)
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+
+        res_mgr = df._mgr.apply(
+            aggregator, labels=ids, ngroups=ngroups, **aggregator_kwargs
+        )
+        res_mgr.axes[1] = self._grouper.result_index
+        result = df._constructor_from_mgr(res_mgr, axes=res_mgr.axes)
+
+        if data.ndim == 1:
+            result = result.squeeze("columns")
+            result.name = data.name
+        else:
+            result.columns = data.columns
+        return result
+
+    @final
+    def _transform_with_numba(self, func, *args, engine_kwargs=None, **kwargs):
+        """
+        Perform groupby transform routine with the numba engine.
+
+        This routine mimics the data splitting routine of the DataSplitter class
+        to generate the indices of each group in the sorted data and then passes the
+        data and indices into a Numba jitted function.
+        """
+        data = self._obj_with_exclusions
+        index_sorting = self._grouper.result_ilocs
+        df = data if data.ndim == 2 else data.to_frame()
+
+        starts, ends, sorted_index, sorted_data = self._numba_prep(df)
+        numba_.validate_udf(func)
+        args, kwargs = prepare_function_arguments(
+            func, args, kwargs, num_required_args=2
+        )
+        numba_transform_func = numba_.generate_numba_transform_func(
+            func, **get_jit_arguments(engine_kwargs)
+        )
+        result = numba_transform_func(
+            sorted_data,
+            sorted_index,
+            starts,
+            ends,
+            len(df.columns),
+            *args,
+        )
+        # result values needs to be resorted to their original positions since we
+        # evaluated the data sorted by group
+        result = result.take(np.argsort(index_sorting), axis=0)
+        index = data.index
+        if data.ndim == 1:
+            result_kwargs = {"name": data.name}
+            result = result.ravel()
+        else:
+            result_kwargs = {"columns": data.columns}
+        return data._constructor(result, index=index, **result_kwargs)
+
+    @final
+    def _aggregate_with_numba(self, func, *args, engine_kwargs=None, **kwargs):
+        """
+        Perform groupby aggregation routine with the numba engine.
+
+        This routine mimics the data splitting routine of the DataSplitter class
+        to generate the indices of each group in the sorted data and then passes the
+        data and indices into a Numba jitted function.
+        """
+        data = self._obj_with_exclusions
+        df = data if data.ndim == 2 else data.to_frame()
+
+        starts, ends, sorted_index, sorted_data = self._numba_prep(df)
+        numba_.validate_udf(func)
+        args, kwargs = prepare_function_arguments(
+            func, args, kwargs, num_required_args=2
+        )
+        numba_agg_func = numba_.generate_numba_agg_func(
+            func, **get_jit_arguments(engine_kwargs)
+        )
+        result = numba_agg_func(
+            sorted_data,
+            sorted_index,
+            starts,
+            ends,
+            len(df.columns),
+            *args,
+        )
+        index = self._grouper.result_index
+        if data.ndim == 1:
+            result_kwargs = {"name": data.name}
+            result = result.ravel()
+        else:
+            result_kwargs = {"columns": data.columns}
+        res = data._constructor(result, index=index, **result_kwargs)
+        if not self.as_index:
+            res = self._insert_inaxis_grouper(res)
+            res.index = default_index(len(res))
+        return res
+
+    # -----------------------------------------------------------------
+    # apply/agg/transform
+
+    def apply(self, func, *args, include_groups: bool = False, **kwargs) -> NDFrameT:
+        """
+        Apply function ``func`` group-wise and combine the results together.
+
+        The function passed to ``apply`` must take a dataframe as its first
+        argument and return a DataFrame, Series or scalar. ``apply`` will
+        then take care of combining the results back together into a single
+        dataframe or series. ``apply`` is therefore a highly flexible
+        grouping method.
+
+        While ``apply`` is a very flexible method, its downside is that
+        using it can be quite a bit slower than using more specific methods
+        like ``agg`` or ``transform``. Pandas offers a wide range of method that will
+        be much faster than using ``apply`` for their specific purposes, so try to
+        use them before reaching for ``apply``.
+
+        Parameters
+        ----------
+        func : callable
+            A callable that takes a dataframe as its first argument, and
+            returns a dataframe, a series or a scalar. In addition the
+            callable may take positional and keyword arguments.
+
+        *args : tuple
+            Optional positional arguments to pass to ``func``.
+
+        include_groups : bool, default False
+            When True, will attempt to apply ``func`` to the groupings in
+            the case that they are columns of the DataFrame. If this raises a
+            TypeError, the result will be computed with the groupings excluded.
+            When False, the groupings will be excluded when applying ``func``.
+
+            .. versionadded:: 2.2.0
+
+            .. versionchanged:: 3.0.0
+
+            The default changed from True to False, and True is no longer allowed.
+
+        **kwargs : dict
+            Optional keyword arguments to pass to ``func``.
+
+        Returns
+        -------
+        Series or DataFrame
+            A pandas object with the result of applying ``func`` to each group.
+
+        See Also
+        --------
+        pipe : Apply function to the full GroupBy object instead of to each
+            group.
+        aggregate : Apply aggregate function to the GroupBy object.
+        transform : Apply function column-by-column to the GroupBy object.
+        Series.apply : Apply a function to a Series.
+        DataFrame.apply : Apply a function to each row or column of a DataFrame.
+
+        Notes
+        -----
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``,
+            see the examples below.
+
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": "a a b".split(), "B": [1, 2, 3], "C": [4, 6, 5]})
+        >>> g1 = df.groupby("A", group_keys=False)
+        >>> g2 = df.groupby("A", group_keys=True)
+
+        Notice that ``g1`` and ``g2`` have two groups, ``a`` and ``b``, and only
+        differ in their ``group_keys`` argument. Calling `apply` in various ways,
+        we can get different grouping results:
+
+        Example 1: below the function passed to `apply` takes a DataFrame as
+        its argument and returns a DataFrame. `apply` combines the result for
+        each group together into a new DataFrame:
+
+        >>> g1[["B", "C"]].apply(lambda x: x / x.sum())
+                  B    C
+        0  0.333333  0.4
+        1  0.666667  0.6
+        2  1.000000  1.0
+
+        In the above, the groups are not part of the index. We can have them included
+        by using ``g2`` where ``group_keys=True``:
+
+        >>> g2[["B", "C"]].apply(lambda x: x / x.sum())
+                    B    C
+        A
+        a 0  0.333333  0.4
+          1  0.666667  0.6
+        b 2  1.000000  1.0
+
+        Example 2: The function passed to `apply` takes a DataFrame as
+        its argument and returns a Series.  `apply` combines the result for
+        each group together into a new DataFrame.
+
+        .. versionchanged:: 1.3.0
+
+            The resulting dtype will reflect the return value of the passed ``func``.
+
+        >>> g1[["B", "C"]].apply(lambda x: x.astype(float).max() - x.min())
+             B    C
+        A
+        a  1.0  2.0
+        b  0.0  0.0
+
+        >>> g2[["B", "C"]].apply(lambda x: x.astype(float).max() - x.min())
+             B    C
+        A
+        a  1.0  2.0
+        b  0.0  0.0
+
+        The ``group_keys`` argument has no effect here because the result is not
+        like-indexed (i.e. :ref:`a transform <groupby.transform>`) when compared
+        to the input.
+
+        Example 3: The function passed to `apply` takes a DataFrame as
+        its argument and returns a scalar. `apply` combines the result for
+        each group together into a Series, including setting the index as
+        appropriate:
+
+        >>> g1.apply(lambda x: x.C.max() - x.B.min())
+        A
+        a    5
+        b    2
+        dtype: int64
+
+        Example 4: The function passed to ``apply`` returns ``None`` for one of the
+        group. This group is filtered from the result:
+
+        >>> g1.apply(lambda x: None if x.iloc[0, 0] == 3 else x)
+           B  C
+        0  1  4
+        1  2  6
+        """
+        if include_groups:
+            raise ValueError("include_groups=True is no longer allowed.")
+        if isinstance(func, str):
+            if hasattr(self, func):
+                res = getattr(self, func)
+                if callable(res):
+                    return res(*args, **kwargs)
+                elif args or kwargs:
+                    raise ValueError(f"Cannot pass arguments to property {func}")
+                return res
+
+            else:
+                raise TypeError(f"apply func should be callable, not '{func}'")
+
+        elif args or kwargs:
+            if callable(func):
+
+                @wraps(func)
+                def f(g):
+                    return func(g, *args, **kwargs)
+
+            else:
+                raise ValueError(
+                    "func must be a callable if args or kwargs are supplied"
+                )
+        else:
+            f = func
+
+        return self._python_apply_general(f, self._obj_with_exclusions)
+
+    @final
+    def _python_apply_general(
+        self,
+        f: Callable,
+        data: DataFrame | Series,
+        not_indexed_same: bool | None = None,
+        is_transform: bool = False,
+        is_agg: bool = False,
+    ) -> NDFrameT:
+        """
+        Apply function f in python space
+
+        Parameters
+        ----------
+        f : callable
+            Function to apply
+        data : Series or DataFrame
+            Data to apply f to
+        not_indexed_same: bool, optional
+            When specified, overrides the value of not_indexed_same. Apply behaves
+            differently when the result index is equal to the input index, but
+            this can be coincidental leading to value-dependent behavior.
+        is_transform : bool, default False
+            Indicator for whether the function is actually a transform
+            and should not have group keys prepended.
+        is_agg : bool, default False
+            Indicator for whether the function is an aggregation. When the
+            result is empty, we don't want to warn for this case.
+            See _GroupBy._python_agg_general.
+
+        Returns
+        -------
+        Series or DataFrame
+            data after applying f
+        """
+        values, mutated = self._grouper.apply_groupwise(f, data)
+        if not_indexed_same is None:
+            not_indexed_same = mutated
+
+        return self._wrap_applied_output(
+            data,
+            values,
+            not_indexed_same,
+            is_transform,
+        )
+
+    @final
+    def _agg_general(
+        self,
+        numeric_only: bool = False,
+        min_count: int = -1,
+        *,
+        alias: str,
+        npfunc: Callable | None = None,
+        **kwargs,
+    ):
+        result = self._cython_agg_general(
+            how=alias,
+            alt=npfunc,
+            numeric_only=numeric_only,
+            min_count=min_count,
+            **kwargs,
+        )
+        return result.__finalize__(self.obj, method="groupby")
+
+    def _agg_py_fallback(
+        self, how: str, values: ArrayLike, ndim: int, alt: Callable
+    ) -> ArrayLike:
+        """
+        Fallback to pure-python aggregation if _cython_operation raises
+        NotImplementedError.
+        """
+        # We get here with a) EADtypes and b) object dtype
+        assert alt is not None
+
+        if values.ndim == 1:
+            # For DataFrameGroupBy we only get here with ExtensionArray
+            ser = Series(values, copy=False)
+        else:
+            # We only get here with values.dtype == object
+            df = DataFrame(values.T, dtype=values.dtype)
+            # bc we split object blocks in grouped_reduce, we have only 1 col
+            # otherwise we'd have to worry about block-splitting GH#39329
+            assert df.shape[1] == 1
+            # Avoid call to self.values that can occur in DataFrame
+            #  reductions; see GH#28949
+            ser = df.iloc[:, 0]
+
+        # We do not get here with UDFs, so we know that our dtype
+        #  should always be preserved by the implemented aggregations
+        # TODO: Is this exactly right; see WrappedCythonOp get_result_dtype?
+        try:
+            res_values = self._grouper.agg_series(ser, alt, preserve_dtype=True)
+        except Exception as err:
+            msg = f"agg function failed [how->{how},dtype->{ser.dtype}]"
+            # preserve the kind of exception that raised
+            raise type(err)(msg) from err
+
+        if ser.dtype == object:
+            res_values = res_values.astype(object, copy=False)
+
+        # If we are DataFrameGroupBy and went through a SeriesGroupByPath
+        # then we need to reshape
+        # GH#32223 includes case with IntegerArray values, ndarray res_values
+        # test_groupby_duplicate_columns with object dtype values
+        return ensure_block_shape(res_values, ndim=ndim)
+
+    @final
+    def _cython_agg_general(
+        self,
+        how: str,
+        alt: Callable | None = None,
+        numeric_only: bool = False,
+        min_count: int = -1,
+        **kwargs,
+    ):
+        # Note: we never get here with how="ohlc" for DataFrameGroupBy;
+        #  that goes through SeriesGroupBy
+
+        data = self._get_data_to_aggregate(numeric_only=numeric_only, name=how)
+
+        def array_func(values: ArrayLike) -> ArrayLike:
+            try:
+                result = self._grouper._cython_operation(
+                    "aggregate",
+                    values,
+                    how,
+                    axis=data.ndim - 1,
+                    min_count=min_count,
+                    **kwargs,
+                )
+            except NotImplementedError:
+                # generally if we have numeric_only=False
+                # and non-applicable functions
+                # try to python agg
+                # TODO: shouldn't min_count matter?
+                # TODO: avoid special casing SparseArray here
+                if how in ["any", "all"] and isinstance(values, SparseArray):
+                    pass
+                elif alt is None or how in ["any", "all", "std", "sem"]:
+                    raise  # TODO: re-raise as TypeError?  should not be reached
+            else:
+                return result
+
+            assert alt is not None
+            result = self._agg_py_fallback(how, values, ndim=data.ndim, alt=alt)
+            return result
+
+        new_mgr = data.grouped_reduce(array_func)
+        res = self._wrap_agged_manager(new_mgr)
+        if how in ["idxmin", "idxmax"]:
+            res = self._wrap_idxmax_idxmin(res)
+        out = self._wrap_aggregated_output(res)
+        return out
+
+    def _cython_transform(self, how: str, numeric_only: bool = False, **kwargs):
+        raise AbstractMethodError(self)
+
+    @final
+    def _transform(self, func, *args, engine=None, engine_kwargs=None, **kwargs):
+        # optimized transforms
+        if not isinstance(func, str):
+            return self._transform_general(func, engine, engine_kwargs, *args, **kwargs)
+
+        elif func not in base.transform_kernel_allowlist:
+            msg = f"'{func}' is not a valid function name for transform(name)"
+            raise ValueError(msg)
+        elif func in base.cythonized_kernels or func in base.transformation_kernels:
+            # cythonized transform or canned "agg+broadcast"
+            if engine is not None:
+                kwargs["engine"] = engine
+                kwargs["engine_kwargs"] = engine_kwargs
+            return getattr(self, func)(*args, **kwargs)
+
+        else:
+            # i.e. func in base.reduction_kernels
+            if self.observed:
+                return self._reduction_kernel_transform(
+                    func, *args, engine=engine, engine_kwargs=engine_kwargs, **kwargs
+                )
+
+            with (
+                com.temp_setattr(self, "observed", True),
+                com.temp_setattr(self, "_grouper", self._grouper.observed_grouper),
+            ):
+                return self._reduction_kernel_transform(
+                    func, *args, engine=engine, engine_kwargs=engine_kwargs, **kwargs
+                )
+
+    @final
+    def _reduction_kernel_transform(
+        self, func, *args, engine=None, engine_kwargs=None, **kwargs
+    ):
+        # GH#30918 Use _transform_fast only when we know func is an aggregation
+        # If func is a reduction, we need to broadcast the
+        # result to the whole group. Compute func result
+        # and deal with possible broadcasting below.
+        with com.temp_setattr(self, "as_index", True):
+            # GH#49834 - result needs groups in the index for
+            # _wrap_transform_fast_result
+            if func in ["idxmin", "idxmax"]:
+                func = cast(Literal["idxmin", "idxmax"], func)
+                result = self._idxmax_idxmin(func, True, *args, **kwargs)
+            else:
+                if engine is not None:
+                    kwargs["engine"] = engine
+                    kwargs["engine_kwargs"] = engine_kwargs
+                result = getattr(self, func)(*args, **kwargs)
+
+        return self._wrap_transform_fast_result(result)
+
+    @final
+    def _wrap_transform_fast_result(self, result: NDFrameT) -> NDFrameT:
+        """
+        Fast transform path for aggregations.
+        """
+        obj = self._obj_with_exclusions
+
+        # for each col, reshape to size of original frame by take operation
+        ids = self._grouper.ids
+        result = result.reindex(self._grouper.result_index, axis=0)
+
+        if self.obj.ndim == 1:
+            # i.e. SeriesGroupBy
+            out = algorithms.take_nd(result._values, ids)
+            output = obj._constructor(out, index=obj.index, name=obj.name)
+        else:
+            # `.size()` gives Series output on DataFrame input, need axis 0
+            # GH#46209
+            # Don't convert indices: negative indices need to give rise
+            # to null values in the result
+            new_ax = result.index.take(ids)
+            output = result._reindex_with_indexers({0: (new_ax, ids)}, allow_dups=True)
+            output = output.set_axis(obj.index, axis=0)
+        return output
+
+    # -----------------------------------------------------------------
+    # Utilities
+
+    @final
+    def _apply_filter(self, indices, dropna):
+        if len(indices) == 0:
+            indices = np.array([], dtype="int64")
+        else:
+            indices = np.sort(np.concatenate(indices))
+        if dropna:
+            filtered = self._selected_obj.take(indices, axis=0)
+        else:
+            mask = np.empty(len(self._selected_obj.index), dtype=bool)
+            mask.fill(False)
+            mask[indices.astype(int)] = True
+            # mask fails to broadcast when passed to where; broadcast manually.
+            mask = np.tile(mask, list(self._selected_obj.shape[1:]) + [1]).T
+            filtered = self._selected_obj.where(mask)  # Fill with NaNs.
+        return filtered
+
+    @final
+    def _cumcount_array(self, ascending: bool = True) -> np.ndarray:
+        """
+        Parameters
+        ----------
+        ascending : bool, default True
+            If False, number in reverse, from length of group - 1 to 0.
+
+        Notes
+        -----
+        this is currently implementing sort=False
+        (though the default is sort=True) for groupby in general
+        """
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+        sorter = get_group_index_sorter(ids, ngroups)
+        ids, count = ids[sorter], len(ids)
+
+        if count == 0:
+            return np.empty(0, dtype=np.int64)
+
+        run = np.r_[True, ids[:-1] != ids[1:]]
+        rep = np.diff(np.r_[np.nonzero(run)[0], count])
+        out = (~run).cumsum()
+
+        if ascending:
+            out -= np.repeat(out[run], rep)
+        else:
+            out = np.repeat(out[np.r_[run[1:], True]], rep) - out
+
+        if self._grouper.has_dropped_na:
+            out = np.where(ids == -1, np.nan, out.astype(np.float64, copy=False))
+        else:
+            out = out.astype(np.int64, copy=False)
+
+        rev = np.empty(count, dtype=np.intp)
+        rev[sorter] = np.arange(count, dtype=np.intp)
+        return out[rev]
+
+    # -----------------------------------------------------------------
+
+    @final
+    @property
+    def _obj_1d_constructor(self) -> Callable:
+        # GH28330 preserve subclassed Series/DataFrames
+        if isinstance(self.obj, DataFrame):
+            return self.obj._constructor_sliced
+        assert isinstance(self.obj, Series)
+        return self.obj._constructor
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def any(self, skipna: bool = True) -> NDFrameT:
+        """
+        Return True if any value in the group is truthful, else False.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Flag to ignore nan values during truth testing.
+
+        Returns
+        -------
+        Series or DataFrame
+            DataFrame or Series of boolean values, where a value is True if any element
+            is True within its respective group, False otherwise.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 0], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    0
+        dtype: int64
+        >>> ser.groupby(level=0).any()
+        a     True
+        b    False
+        dtype: bool
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 0, 3], [1, 0, 6], [7, 1, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["ostrich", "penguin", "parrot"]
+        ... )
+        >>> df
+                 a  b  c
+        ostrich  1  0  3
+        penguin  1  0  6
+        parrot   7  1  9
+        >>> df.groupby(by=["a"]).any()
+               b      c
+        a
+        1  False   True
+        7   True   True
+        """
+        return self._cython_agg_general(
+            "any",
+            alt=lambda x: Series(x, copy=False).any(skipna=skipna),
+            skipna=skipna,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def all(self, skipna: bool = True) -> NDFrameT:
+        """
+        Return True if all values in the group are truthful, else False.
+
+        Parameters
+        ----------
+        skipna : bool, default True
+            Flag to ignore nan values during truth testing.
+
+        Returns
+        -------
+        Series or DataFrame
+            DataFrame or Series of boolean values, where a value is True if all elements
+            are True within its respective group, False otherwise.
+        %(see_also)s
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 0], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    0
+        dtype: int64
+        >>> ser.groupby(level=0).all()
+        a     True
+        b    False
+        dtype: bool
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 0, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["ostrich", "penguin", "parrot"]
+        ... )
+        >>> df
+                 a  b  c
+        ostrich  1  0  3
+        penguin  1  5  6
+        parrot   7  8  9
+        >>> df.groupby(by=["a"]).all()
+               b      c
+        a
+        1  False   True
+        7   True   True
+        """
+        return self._cython_agg_general(
+            "all",
+            alt=lambda x: Series(x, copy=False).all(skipna=skipna),
+            skipna=skipna,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def count(self) -> NDFrameT:
+        """
+        Compute count of group, excluding missing values.
+
+        Returns
+        -------
+        Series or DataFrame
+            Count of values within each group.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, np.nan], index=lst)
+        >>> ser
+        a    1.0
+        a    2.0
+        b    NaN
+        dtype: float64
+        >>> ser.groupby(level=0).count()
+        a    2
+        b    0
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, np.nan, 3], [1, np.nan, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["cow", "horse", "bull"]
+        ... )
+        >>> df
+                a	  b	c
+        cow     1	NaN	3
+        horse	1	NaN	6
+        bull	7	8.0	9
+        >>> df.groupby("a").count()
+            b   c
+        a
+        1   0   2
+        7   1   1
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").count()
+        2023-01-01    2
+        2023-02-01    2
+        Freq: MS, dtype: int64
+        """
+        data = self._get_data_to_aggregate()
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+        mask = ids != -1
+
+        is_series = data.ndim == 1
+
+        def hfunc(bvalues: ArrayLike) -> ArrayLike:
+            # TODO(EA2D): reshape would not be necessary with 2D EAs
+            if bvalues.ndim == 1:
+                # EA
+                masked = mask & ~isna(bvalues).reshape(1, -1)
+            else:
+                masked = mask & ~isna(bvalues)
+
+            counted = lib.count_level_2d(masked, labels=ids, max_bin=ngroups)
+            if isinstance(bvalues, BaseMaskedArray):
+                return IntegerArray(
+                    counted[0], mask=np.zeros(counted.shape[1], dtype=np.bool_)
+                )
+            elif isinstance(bvalues, ArrowExtensionArray) and not isinstance(
+                bvalues.dtype, StringDtype
+            ):
+                dtype = pandas_dtype("int64[pyarrow]")
+                return type(bvalues)._from_sequence(counted[0], dtype=dtype)
+            if is_series:
+                assert counted.ndim == 2
+                assert counted.shape[0] == 1
+                return counted[0]
+            return counted
+
+        new_mgr = data.grouped_reduce(hfunc)
+        new_obj = self._wrap_agged_manager(new_mgr)
+        result = self._wrap_aggregated_output(new_obj)
+
+        return result
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def mean(
+        self,
+        numeric_only: bool = False,
+        skipna: bool = True,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        """
+        Compute mean of groups, excluding missing values.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None`` and defaults to ``False``.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        engine : str, default None
+            * ``'cython'`` : Runs the operation through C-extensions from cython.
+            * ``'numba'`` : Runs the operation through JIT compiled code from numba.
+            * ``None`` : Defaults to ``'cython'`` or globally setting
+              ``compute.use_numba``
+
+            .. versionadded:: 1.4.0
+
+        engine_kwargs : dict, default None
+            * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{{'nopython': True, 'nogil': False, 'parallel': False}}``
+
+            .. versionadded:: 1.4.0
+
+        Returns
+        -------
+        pandas.Series or pandas.DataFrame
+            Mean of values within each group. Same object type as the caller.
+        %(see_also)s
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"A": [1, 1, 2, 1, 2], "B": [np.nan, 2, 3, 4, 5], "C": [1, 2, 1, 1, 2]},
+        ...     columns=["A", "B", "C"],
+        ... )
+
+        Groupby one column and return the mean of the remaining columns in
+        each group.
+
+        >>> df.groupby("A").mean()
+             B         C
+        A
+        1  3.0  1.333333
+        2  4.0  1.500000
+
+        Groupby two columns and return the mean of the remaining column.
+
+        >>> df.groupby(["A", "B"]).mean()
+                 C
+        A B
+        1 2.0  2.0
+          4.0  1.0
+        2 3.0  1.0
+          5.0  2.0
+
+        Groupby one column and return the mean of only particular column in
+        the group.
+
+        >>> df.groupby("A")["B"].mean()
+        A
+        1    3.0
+        2    4.0
+        Name: B, dtype: float64
+        """
+
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_mean
+
+            return self._numba_agg_general(
+                grouped_mean,
+                executor.float_dtype_mapping,
+                engine_kwargs,
+                min_periods=0,
+                skipna=skipna,
+            )
+        else:
+            result = self._cython_agg_general(
+                "mean",
+                alt=lambda x: Series(x, copy=False).mean(
+                    numeric_only=numeric_only, skipna=skipna
+                ),
+                numeric_only=numeric_only,
+                skipna=skipna,
+            )
+            return result.__finalize__(self.obj, method="groupby")
+
+    @final
+    def median(self, numeric_only: bool = False, skipna: bool = True) -> NDFrameT:
+        """
+        Compute median of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None`` and defaults to False.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Median of values within each group.
+
+        See Also
+        --------
+        Series.groupby : Apply a function groupby to a Series.
+        DataFrame.groupby : Apply a function groupby to each row or column of a
+            DataFrame.
+
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([7, 2, 8, 4, 3, 3], index=lst)
+        >>> ser
+        a     7
+        a     2
+        a     8
+        b     4
+        b     3
+        b     3
+        dtype: int64
+        >>> ser.groupby(level=0).median()
+        a    7.0
+        b    3.0
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = {"a": [1, 3, 5, 7, 7, 8, 3], "b": [1, 4, 8, 4, 4, 2, 1]}
+        >>> df = pd.DataFrame(
+        ...     data, index=["dog", "dog", "dog", "mouse", "mouse", "mouse", "mouse"]
+        ... )
+        >>> df
+                 a  b
+          dog    1  1
+          dog    3  4
+          dog    5  8
+        mouse    7  4
+        mouse    7  4
+        mouse    8  2
+        mouse    3  1
+        >>> df.groupby(level=0).median()
+                 a    b
+        dog    3.0  4.0
+        mouse  7.0  3.0
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 3, 4, 5],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").median()
+        2023-01-01    2.0
+        2023-02-01    4.0
+        Freq: MS, dtype: float64
+        """
+        result = self._cython_agg_general(
+            "median",
+            alt=lambda x: Series(x, copy=False).median(
+                numeric_only=numeric_only, skipna=skipna
+            ),
+            numeric_only=numeric_only,
+            skipna=skipna,
+        )
+        return result.__finalize__(self.obj, method="groupby")
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def std(
+        self,
+        ddof: int = 1,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        numeric_only: bool = False,
+        skipna: bool = True,
+    ):
+        """
+        Compute standard deviation of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Delta Degrees of Freedom. The divisor used in calculations is ``N - ddof``,
+            where ``N`` represents the number of elements.
+
+        engine : str, default None
+            * ``'cython'`` : Runs the operation through C-extensions from cython.
+            * ``'numba'`` : Runs the operation through JIT compiled code from numba.
+            * ``None`` : Defaults to ``'cython'`` or globally setting
+              ``compute.use_numba``
+
+            .. versionadded:: 1.4.0
+
+        engine_kwargs : dict, default None
+            * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{{'nopython': True, 'nogil': False, 'parallel': False}}``
+
+            .. versionadded:: 1.4.0
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Standard deviation of values within each group.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([7, 2, 8, 4, 3, 3], index=lst)
+        >>> ser
+        a     7
+        a     2
+        a     8
+        b     4
+        b     3
+        b     3
+        dtype: int64
+        >>> ser.groupby(level=0).std()
+        a    3.21455
+        b    0.57735
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = {"a": [1, 3, 5, 7, 7, 8, 3], "b": [1, 4, 8, 4, 4, 2, 1]}
+        >>> df = pd.DataFrame(
+        ...     data, index=["dog", "dog", "dog", "mouse", "mouse", "mouse", "mouse"]
+        ... )
+        >>> df
+                 a  b
+          dog    1  1
+          dog    3  4
+          dog    5  8
+        mouse    7  4
+        mouse    7  4
+        mouse    8  2
+        mouse    3  1
+        >>> df.groupby(level=0).std()
+                      a         b
+        dog    2.000000  3.511885
+        mouse  2.217356  1.500000
+        """
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_var
+
+            return np.sqrt(
+                self._numba_agg_general(
+                    grouped_var,
+                    executor.float_dtype_mapping,
+                    engine_kwargs,
+                    min_periods=0,
+                    ddof=ddof,
+                    skipna=skipna,
+                )
+            )
+        else:
+            return self._cython_agg_general(
+                "std",
+                alt=lambda x: Series(x, copy=False).std(ddof=ddof, skipna=skipna),
+                numeric_only=numeric_only,
+                ddof=ddof,
+                skipna=skipna,
+            )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def var(
+        self,
+        ddof: int = 1,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        numeric_only: bool = False,
+        skipna: bool = True,
+    ):
+        """
+        Compute variance of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Degrees of freedom.
+
+        engine : str, default None
+            * ``'cython'`` : Runs the operation through C-extensions from cython.
+            * ``'numba'`` : Runs the operation through JIT compiled code from numba.
+            * ``None`` : Defaults to ``'cython'`` or globally setting
+              ``compute.use_numba``
+
+            .. versionadded:: 1.4.0
+
+        engine_kwargs : dict, default None
+            * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{{'nopython': True, 'nogil': False, 'parallel': False}}``
+
+            .. versionadded:: 1.4.0
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Variance of values within each group.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([7, 2, 8, 4, 3, 3], index=lst)
+        >>> ser
+        a     7
+        a     2
+        a     8
+        b     4
+        b     3
+        b     3
+        dtype: int64
+        >>> ser.groupby(level=0).var()
+        a    10.333333
+        b     0.333333
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = {"a": [1, 3, 5, 7, 7, 8, 3], "b": [1, 4, 8, 4, 4, 2, 1]}
+        >>> df = pd.DataFrame(
+        ...     data, index=["dog", "dog", "dog", "mouse", "mouse", "mouse", "mouse"]
+        ... )
+        >>> df
+                 a  b
+          dog    1  1
+          dog    3  4
+          dog    5  8
+        mouse    7  4
+        mouse    7  4
+        mouse    8  2
+        mouse    3  1
+        >>> df.groupby(level=0).var()
+                      a          b
+        dog    4.000000  12.333333
+        mouse  4.916667   2.250000
+        """
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_var
+
+            return self._numba_agg_general(
+                grouped_var,
+                executor.float_dtype_mapping,
+                engine_kwargs,
+                min_periods=0,
+                ddof=ddof,
+                skipna=skipna,
+            )
+        else:
+            return self._cython_agg_general(
+                "var",
+                alt=lambda x: Series(x, copy=False).var(ddof=ddof, skipna=skipna),
+                numeric_only=numeric_only,
+                ddof=ddof,
+                skipna=skipna,
+            )
+
+    @final
+    def _value_counts(
+        self,
+        subset: Sequence[Hashable] | None = None,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        dropna: bool = True,
+    ) -> DataFrame | Series:
+        """
+        Shared implementation of value_counts for SeriesGroupBy and DataFrameGroupBy.
+
+        SeriesGroupBy additionally supports a bins argument. See the docstring of
+        DataFrameGroupBy.value_counts for a description of arguments.
+        """
+        name = "proportion" if normalize else "count"
+
+        df = self.obj
+        obj = self._obj_with_exclusions
+
+        in_axis_names = {
+            grouping.name for grouping in self._grouper.groupings if grouping.in_axis
+        }
+        if isinstance(obj, Series):
+            _name = obj.name
+            keys: Iterable[Series] = [] if _name in in_axis_names else [obj]
+        else:
+            unique_cols = set(obj.columns)
+            if subset is not None:
+                subsetted = set(subset)
+                clashing = subsetted & set(in_axis_names)
+                if clashing:
+                    raise ValueError(
+                        f"Keys {clashing} in subset cannot be in "
+                        "the groupby column keys."
+                    )
+                doesnt_exist = subsetted - unique_cols
+                if doesnt_exist:
+                    raise ValueError(
+                        f"Keys {doesnt_exist} in subset do not exist in the DataFrame."
+                    )
+            else:
+                subsetted = unique_cols
+
+            keys = (
+                # Can't use .values because the column label needs to be preserved
+                obj.iloc[:, idx]
+                for idx, _name in enumerate(obj.columns)
+                if _name not in in_axis_names and _name in subsetted
+            )
+
+        groupings = list(self._grouper.groupings)
+        for key in keys:
+            grouper, _, _ = get_grouper(
+                df,
+                key=key,
+                sort=False,
+                observed=False,
+                dropna=dropna,
+            )
+            groupings += list(grouper.groupings)
+
+        # Take the size of the overall columns
+        gb = df.groupby(
+            groupings,
+            sort=False,
+            observed=self.observed,
+            dropna=self.dropna,
+        )
+        result_series = cast(Series, gb.size())
+        result_series.name = name
+
+        if sort:
+            # Sort by the values
+            result_series = result_series.sort_values(
+                ascending=ascending, kind="stable"
+            )
+        if self.sort:
+            # Sort by the groupings
+            names = result_series.index.names
+            # GH#55951 - Temporarily replace names in case they are integers
+            result_series.index.names = range(len(names))
+            index_level = range(len(self._grouper.groupings))
+            result_series = result_series.sort_index(
+                level=index_level, sort_remaining=False
+            )
+            result_series.index.names = names
+
+        if normalize:
+            # Normalize the results by dividing by the original group sizes.
+            # We are guaranteed to have the first N levels be the
+            # user-requested grouping.
+            levels = list(
+                range(len(self._grouper.groupings), result_series.index.nlevels)
+            )
+            indexed_group_size = result_series.groupby(
+                result_series.index.droplevel(levels),
+                sort=self.sort,
+                dropna=self.dropna,
+                # GH#43999 - deprecation of observed=False
+                observed=False,
+            ).transform("sum")
+            result_series /= indexed_group_size
+
+            # Handle groups of non-observed categories
+            result_series = result_series.fillna(0.0)
+
+        result: Series | DataFrame
+        if self.as_index:
+            result = result_series
+        else:
+            # Convert to frame
+            index = result_series.index
+            columns = com.fill_missing_names(index.names)
+            if name in columns:
+                raise ValueError(f"Column label '{name}' is duplicate of result column")
+            result_series.name = name
+            result_series.index = index.set_names(range(len(columns)))
+            result_frame = result_series.reset_index()
+            orig_dtype = self._grouper.groupings[0].obj.columns.dtype  # type: ignore[union-attr]
+            cols = Index(columns, dtype=orig_dtype).insert(len(columns), name)
+            result_frame.columns = cols
+            result = result_frame
+        return result.__finalize__(self.obj, method="value_counts")
+
+    @final
+    def sem(
+        self, ddof: int = 1, numeric_only: bool = False, skipna: bool = True
+    ) -> NDFrameT:
+        """
+        Compute standard error of the mean of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Degrees of freedom.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Standard error of the mean of values within each group.
+
+        See Also
+        --------
+        DataFrame.sem : Return unbiased standard error of the mean over requested axis.
+        Series.sem : Return unbiased standard error of the mean over requested axis.
+
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b", "b"]
+        >>> ser = pd.Series([5, 10, 8, 14], index=lst)
+        >>> ser
+        a     5
+        a    10
+        b     8
+        b    14
+        dtype: int64
+        >>> ser.groupby(level=0).sem()
+        a    2.5
+        b    3.0
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 12, 11], [1, 15, 2], [2, 5, 8], [2, 6, 12]]
+        >>> df = pd.DataFrame(
+        ...     data,
+        ...     columns=["a", "b", "c"],
+        ...     index=["tuna", "salmon", "catfish", "goldfish"],
+        ... )
+        >>> df
+                   a   b   c
+            tuna   1  12  11
+          salmon   1  15   2
+         catfish   2   5   8
+        goldfish   2   6  12
+        >>> df.groupby("a").sem()
+              b  c
+        a
+        1    1.5  4.5
+        2    0.5  2.0
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 8],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").sem()
+        2023-01-01    0.577350
+        2023-02-01    1.527525
+        Freq: MS, dtype: float64
+        """
+        if numeric_only and self.obj.ndim == 1 and not is_numeric_dtype(self.obj.dtype):
+            raise TypeError(
+                f"{type(self).__name__}.sem called with "
+                f"numeric_only={numeric_only} and dtype {self.obj.dtype}"
+            )
+        return self._cython_agg_general(
+            "sem",
+            alt=lambda x: Series(x, copy=False).sem(ddof=ddof, skipna=skipna),
+            numeric_only=numeric_only,
+            ddof=ddof,
+            skipna=skipna,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def size(self) -> DataFrame | Series:
+        """
+        Compute group sizes.
+
+        Returns
+        -------
+        DataFrame or Series
+            Number of rows in each group as a Series if as_index is True
+            or a DataFrame if as_index is False.
+        %(see_also)s
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([1, 2, 3], index=lst)
+        >>> ser
+        a     1
+        a     2
+        b     3
+        dtype: int64
+        >>> ser.groupby(level=0).size()
+        a    2
+        b    1
+        dtype: int64
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [7, 8, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["owl", "toucan", "eagle"]
+        ... )
+        >>> df
+                a  b  c
+        owl     1  2  3
+        toucan  1  5  6
+        eagle   7  8  9
+        >>> df.groupby("a").size()
+        a
+        1    2
+        7    1
+        dtype: int64
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3],
+        ...     index=pd.DatetimeIndex(["2023-01-01", "2023-01-15", "2023-02-01"]),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        dtype: int64
+        >>> ser.resample("MS").size()
+        2023-01-01    2
+        2023-02-01    1
+        Freq: MS, dtype: int64
+        """
+        result = self._grouper.size()
+        dtype_backend: None | Literal["pyarrow", "numpy_nullable"] = None
+        if isinstance(self.obj, Series):
+            if isinstance(self.obj.array, ArrowExtensionArray):
+                if isinstance(self.obj.array, ArrowStringArrayNumpySemantics):
+                    dtype_backend = None
+                elif isinstance(self.obj.array, ArrowStringArray):
+                    dtype_backend = "numpy_nullable"
+                else:
+                    dtype_backend = "pyarrow"
+            elif isinstance(self.obj.array, BaseMaskedArray):
+                dtype_backend = "numpy_nullable"
+        # TODO: For DataFrames what if columns are mixed arrow/numpy/masked?
+
+        # GH28330 preserve subclassed Series/DataFrames through calls
+        if isinstance(self.obj, Series):
+            result = self._obj_1d_constructor(result, name=self.obj.name)
+        else:
+            result = self._obj_1d_constructor(result)
+
+        if dtype_backend is not None:
+            result = result.convert_dtypes(
+                infer_objects=False,
+                convert_string=False,
+                convert_boolean=False,
+                convert_floating=False,
+                dtype_backend=dtype_backend,
+            )
+
+        if not self.as_index:
+            # error: Incompatible types in assignment (expression has
+            # type "DataFrame", variable has type "Series")
+            result = result.rename("size").reset_index()  # type: ignore[assignment]
+        return result
+
+    @final
+    @doc(
+        _groupby_agg_method_skipna_engine_template,
+        fname="sum",
+        no=False,
+        mc=0,
+        s=True,
+        e=None,
+        ek=None,
+        example=dedent(
+            """\
+        For SeriesGroupBy:
+
+        >>> lst = ['a', 'a', 'b', 'b']
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).sum()
+        a    3
+        b    7
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(data, columns=["a", "b", "c"],
+        ...                   index=["tiger", "leopard", "cheetah", "lion"])
+        >>> df
+                  a  b  c
+          tiger   1  8  2
+        leopard   1  2  5
+        cheetah   2  5  8
+           lion   2  6  9
+        >>> df.groupby("a").sum()
+             b   c
+        a
+        1   10   7
+        2   11  17"""
+        ),
+    )
+    def sum(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        skipna: bool = True,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_sum
+
+            return self._numba_agg_general(
+                grouped_sum,
+                executor.default_dtype_mapping,
+                engine_kwargs,
+                min_periods=min_count,
+                skipna=skipna,
+            )
+        else:
+            # If we are grouping on categoricals we want unobserved categories to
+            # return zero, rather than the default of NaN which the reindexing in
+            # _agg_general() returns. GH #31422
+            with com.temp_setattr(self, "observed", True):
+                result = self._agg_general(
+                    numeric_only=numeric_only,
+                    min_count=min_count,
+                    alias="sum",
+                    npfunc=np.sum,
+                    skipna=skipna,
+                )
+
+            return result
+
+    @final
+    def prod(
+        self, numeric_only: bool = False, min_count: int = 0, skipna: bool = True
+    ) -> NDFrameT:
+        """
+        Compute prod of group values.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None``.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` non-NA values are present the result will be NA.
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series or DataFrame
+            Computed prod of values within each group.
+
+        See Also
+        --------
+        Series.prod : Return the product of the values over the requested axis.
+        DataFrame.prod : Return the product of the values over the requested axis.
+
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b", "b"]
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).prod()
+        a    2
+        b   12
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data,
+        ...     columns=["a", "b", "c"],
+        ...     index=["tiger", "leopard", "cheetah", "lion"],
+        ... )
+        >>> df
+                  a  b  c
+          tiger   1  8  2
+        leopard   1  2  5
+        cheetah   2  5  8
+           lion   2  6  9
+        >>> df.groupby("a").prod()
+             b    c
+        a
+        1   16   10
+        2   30   72
+        """
+        return self._agg_general(
+            numeric_only=numeric_only,
+            min_count=min_count,
+            skipna=skipna,
+            alias="prod",
+            npfunc=np.prod,
+        )
+
+    @final
+    @doc(
+        _groupby_agg_method_skipna_engine_template,
+        fname="min",
+        no=False,
+        mc=-1,
+        e=None,
+        ek=None,
+        s=True,
+        example=dedent(
+            """\
+        For SeriesGroupBy:
+
+        >>> lst = ['a', 'a', 'b', 'b']
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).min()
+        a    1
+        b    3
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(data, columns=["a", "b", "c"],
+        ...                   index=["tiger", "leopard", "cheetah", "lion"])
+        >>> df
+                  a  b  c
+          tiger   1  8  2
+        leopard   1  2  5
+        cheetah   2  5  8
+           lion   2  6  9
+        >>> df.groupby("a").min()
+            b  c
+        a
+        1   2  2
+        2   5  8"""
+        ),
+    )
+    def min(
+        self,
+        numeric_only: bool = False,
+        min_count: int = -1,
+        skipna: bool = True,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_min_max
+
+            return self._numba_agg_general(
+                grouped_min_max,
+                executor.identity_dtype_mapping,
+                engine_kwargs,
+                min_periods=min_count,
+                is_max=False,
+                skipna=skipna,
+            )
+        else:
+            return self._agg_general(
+                numeric_only=numeric_only,
+                min_count=min_count,
+                skipna=skipna,
+                alias="min",
+                npfunc=np.min,
+            )
+
+    @final
+    @doc(
+        _groupby_agg_method_skipna_engine_template,
+        fname="max",
+        no=False,
+        mc=-1,
+        e=None,
+        ek=None,
+        s=True,
+        example=dedent(
+            """\
+        For SeriesGroupBy:
+
+        >>> lst = ['a', 'a', 'b', 'b']
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).max()
+        a    2
+        b    4
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(data, columns=["a", "b", "c"],
+        ...                   index=["tiger", "leopard", "cheetah", "lion"])
+        >>> df
+                  a  b  c
+          tiger   1  8  2
+        leopard   1  2  5
+        cheetah   2  5  8
+           lion   2  6  9
+        >>> df.groupby("a").max()
+            b  c
+        a
+        1   8  5
+        2   6  9"""
+        ),
+    )
+    def max(
+        self,
+        numeric_only: bool = False,
+        min_count: int = -1,
+        skipna: bool = True,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            from pandas.core._numba.kernels import grouped_min_max
+
+            return self._numba_agg_general(
+                grouped_min_max,
+                executor.identity_dtype_mapping,
+                engine_kwargs,
+                min_periods=min_count,
+                is_max=True,
+                skipna=skipna,
+            )
+        else:
+            return self._agg_general(
+                numeric_only=numeric_only,
+                min_count=min_count,
+                skipna=skipna,
+                alias="max",
+                npfunc=np.max,
+            )
+
+    @final
+    def first(
+        self, numeric_only: bool = False, min_count: int = -1, skipna: bool = True
+    ) -> NDFrameT:
+        """
+        Compute the first entry of each column within each group.
+
+        Defaults to skipping NA elements.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        min_count : int, default -1
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` valid values are present the result will be NA.
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 2.2.1
+
+        Returns
+        -------
+        Series or DataFrame
+            First values within each group.
+
+        See Also
+        --------
+        DataFrame.groupby : Apply a function groupby to each row or column of a
+            DataFrame.
+        core.groupby.DataFrameGroupBy.last : Compute the last non-null entry
+            of each column.
+        core.groupby.DataFrameGroupBy.nth : Take the nth row from each group.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     dict(
+        ...         A=[1, 1, 3],
+        ...         B=[None, 5, 6],
+        ...         C=[1, 2, 3],
+        ...         D=["3/11/2000", "3/12/2000", "3/13/2000"],
+        ...     )
+        ... )
+        >>> df["D"] = pd.to_datetime(df["D"])
+        >>> df.groupby("A").first()
+             B  C          D
+        A
+        1  5.0  1 2000-03-11
+        3  6.0  3 2000-03-13
+        >>> df.groupby("A").first(min_count=2)
+            B    C          D
+        A
+        1 NaN  1.0 2000-03-11
+        3 NaN  NaN        NaT
+        >>> df.groupby("A").first(numeric_only=True)
+             B  C
+        A
+        1  5.0  1
+        3  6.0  3
+        """
+
+        def first_compat(obj: NDFrameT):
+            def first(x: Series):
+                """Helper function for first item that isn't NA."""
+                arr = x.array[notna(x.array)]
+                if not len(arr):
+                    return x.array.dtype.na_value
+                return arr[0]
+
+            if isinstance(obj, DataFrame):
+                return obj.apply(first)
+            elif isinstance(obj, Series):
+                return first(obj)
+            else:  # pragma: no cover
+                raise TypeError(type(obj))
+
+        return self._agg_general(
+            numeric_only=numeric_only,
+            min_count=min_count,
+            alias="first",
+            npfunc=first_compat,
+            skipna=skipna,
+        )
+
+    @final
+    def last(
+        self, numeric_only: bool = False, min_count: int = -1, skipna: bool = True
+    ) -> NDFrameT:
+        """
+        Compute the last entry of each column within each group.
+
+        Defaults to skipping NA elements.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. If None, will attempt to use
+            everything, then use only numeric data.
+        min_count : int, default -1
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` valid values are present the result will be NA.
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+
+            .. versionadded:: 2.2.1
+
+        Returns
+        -------
+        Series or DataFrame
+            Last of values within each group.
+
+        See Also
+        --------
+        DataFrame.groupby : Apply a function groupby to each row or column of a
+            DataFrame.
+        core.groupby.DataFrameGroupBy.first : Compute the first non-null entry
+            of each column.
+        core.groupby.DataFrameGroupBy.nth : Take the nth row from each group.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(dict(A=[1, 1, 3], B=[5, None, 6], C=[1, 2, 3]))
+        >>> df.groupby("A").last()
+             B  C
+        A
+        1  5.0  2
+        3  6.0  3
+        """
+
+        def last_compat(obj: NDFrameT):
+            def last(x: Series):
+                """Helper function for last item that isn't NA."""
+                arr = x.array[notna(x.array)]
+                if not len(arr):
+                    return x.array.dtype.na_value
+                return arr[-1]
+
+            if isinstance(obj, DataFrame):
+                return obj.apply(last)
+            elif isinstance(obj, Series):
+                return last(obj)
+            else:  # pragma: no cover
+                raise TypeError(type(obj))
+
+        return self._agg_general(
+            numeric_only=numeric_only,
+            min_count=min_count,
+            alias="last",
+            npfunc=last_compat,
+            skipna=skipna,
+        )
+
+    @final
+    def ohlc(self) -> DataFrame:
+        """
+        Compute open, high, low and close values of a group, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex
+
+        Returns
+        -------
+        DataFrame
+            Open, high, low and close values within each group.
+
+        See Also
+        --------
+        DataFrame.agg : Aggregate using one or more operations over the specified axis.
+        DataFrame.resample : Resample time-series data.
+        DataFrame.groupby : Group DataFrame using a mapper or by a Series of columns.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = [
+        ...     "SPX",
+        ...     "CAC",
+        ...     "SPX",
+        ...     "CAC",
+        ...     "SPX",
+        ...     "CAC",
+        ...     "SPX",
+        ...     "CAC",
+        ... ]
+        >>> ser = pd.Series([3.4, 9.0, 7.2, 5.2, 8.8, 9.4, 0.1, 0.5], index=lst)
+        >>> ser
+        SPX     3.4
+        CAC     9.0
+        SPX     7.2
+        CAC     5.2
+        SPX     8.8
+        CAC     9.4
+        SPX     0.1
+        CAC     0.5
+        dtype: float64
+        >>> ser.groupby(level=0).ohlc()
+             open  high  low  close
+        CAC   9.0   9.4  0.5    0.5
+        SPX   3.4   8.8  0.1    0.1
+
+        For DataFrameGroupBy:
+
+        >>> data = {
+        ...     2022: [1.2, 2.3, 8.9, 4.5, 4.4, 3, 2, 1],
+        ...     2023: [3.4, 9.0, 7.2, 5.2, 8.8, 9.4, 8.2, 1.0],
+        ... }
+        >>> df = pd.DataFrame(
+        ...     data, index=["SPX", "CAC", "SPX", "CAC", "SPX", "CAC", "SPX", "CAC"]
+        ... )
+        >>> df
+             2022  2023
+        SPX   1.2   3.4
+        CAC   2.3   9.0
+        SPX   8.9   7.2
+        CAC   4.5   5.2
+        SPX   4.4   8.8
+        CAC   3.0   9.4
+        SPX   2.0   8.2
+        CAC   1.0   1.0
+        >>> df.groupby(level=0).ohlc()
+            2022                 2023
+            open high  low close open high  low close
+        CAC  2.3  4.5  1.0   1.0  9.0  9.4  1.0   1.0
+        SPX  1.2  8.9  1.2   2.0  3.4  8.8  3.4   8.2
+
+        For Resampler:
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 5],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").ohlc()
+                    open  high  low  close
+        2023-01-01     1     3    1      2
+        2023-02-01     4     5    3      5
+        """
+        if self.obj.ndim == 1:
+            obj = self._selected_obj
+
+            is_numeric = is_numeric_dtype(obj.dtype)
+            if not is_numeric:
+                raise DataError("No numeric types to aggregate")
+
+            res_values = self._grouper._cython_operation(
+                "aggregate", obj._values, "ohlc", axis=0, min_count=-1
+            )
+
+            agg_names = ["open", "high", "low", "close"]
+            result = self.obj._constructor_expanddim(
+                res_values, index=self._grouper.result_index, columns=agg_names
+            )
+            return result
+
+        result = self._apply_to_column_groupbys(lambda sgb: sgb.ohlc())
+        return result
+
+    @doc(DataFrame.describe)
+    def describe(
+        self,
+        percentiles=None,
+        include=None,
+        exclude=None,
+    ) -> NDFrameT:
+        obj = self._obj_with_exclusions
+
+        if len(obj) == 0:
+            described = obj.describe(
+                percentiles=percentiles, include=include, exclude=exclude
+            )
+            if obj.ndim == 1:
+                result = described
+            else:
+                result = described.unstack()
+            return result.to_frame().T.iloc[:0]
+
+        with com.temp_setattr(self, "as_index", True):
+            result = self._python_apply_general(
+                lambda x: x.describe(
+                    percentiles=percentiles, include=include, exclude=exclude
+                ),
+                obj,
+                not_indexed_same=True,
+            )
+
+        # GH#49256 - properly handle the grouping column(s)
+        result = result.unstack()
+        if not self.as_index:
+            result = self._insert_inaxis_grouper(result)
+            result.index = default_index(len(result))
+
+        return result
+
+    @final
+    def resample(
+        self, rule, *args, include_groups: bool = False, **kwargs
+    ) -> Resampler:
+        """
+        Provide resampling when using a TimeGrouper.
+
+        Given a grouper, the function resamples it according to a string
+        "string" -> "frequency".
+
+        See the :ref:`frequency aliases <timeseries.offset_aliases>`
+        documentation for more details.
+
+        Parameters
+        ----------
+        rule : str or DateOffset
+            The offset string or object representing target grouper conversion.
+        *args
+            Possible arguments are `how`, `fill_method`, `limit`, `kind` and
+            `on`, and other arguments of `TimeGrouper`.
+        include_groups : bool, default True
+            When True, will attempt to include the groupings in the operation in
+            the case that they are columns of the DataFrame. If this raises a
+            TypeError, the result will be computed with the groupings excluded.
+            When False, the groupings will be excluded when applying ``func``.
+
+            .. versionadded:: 2.2.0
+
+            .. versionchanged:: 3.0
+
+               The default was changed to False, and True is no longer allowed.
+
+        **kwargs
+            Possible arguments are `how`, `fill_method`, `limit`, `kind` and
+            `on`, and other arguments of `TimeGrouper`.
+
+        Returns
+        -------
+        DatetimeIndexResampler, PeriodIndexResampler or TimdeltaResampler
+            Resampler object for the type of the index.
+
+        See Also
+        --------
+        Grouper : Specify a frequency to resample with when
+            grouping by a key.
+        DatetimeIndex.resample : Frequency conversion and resampling of
+            time series.
+
+        Examples
+        --------
+        >>> idx = pd.date_range("1/1/2000", periods=4, freq="min")
+        >>> df = pd.DataFrame(data=4 * [range(2)], index=idx, columns=["a", "b"])
+        >>> df.iloc[2, 0] = 5
+        >>> df
+                            a  b
+        2000-01-01 00:00:00  0  1
+        2000-01-01 00:01:00  0  1
+        2000-01-01 00:02:00  5  1
+        2000-01-01 00:03:00  0  1
+
+        Downsample the DataFrame into 3 minute bins and sum the values of
+        the timestamps falling into a bin.
+
+        >>> df.groupby("a").resample("3min").sum()
+                                 b
+        a
+        0   2000-01-01 00:00:00  2
+            2000-01-01 00:03:00  1
+        5   2000-01-01 00:00:00  1
+
+        Upsample the series into 30 second bins.
+
+        >>> df.groupby("a").resample("30s").sum()
+                            b
+        a
+        0   2000-01-01 00:00:00  1
+            2000-01-01 00:00:30  0
+            2000-01-01 00:01:00  1
+            2000-01-01 00:01:30  0
+            2000-01-01 00:02:00  0
+            2000-01-01 00:02:30  0
+            2000-01-01 00:03:00  1
+        5   2000-01-01 00:02:00  1
+
+        Resample by month. Values are assigned to the month of the period.
+
+        >>> df.groupby("a").resample("ME").sum()
+                    b
+        a
+        0   2000-01-31  3
+        5   2000-01-31  1
+
+        Downsample the series into 3 minute bins as above, but close the right
+        side of the bin interval.
+
+        >>> (df.groupby("a").resample("3min", closed="right").sum())
+                                 b
+        a
+        0   1999-12-31 23:57:00  1
+            2000-01-01 00:00:00  2
+        5   2000-01-01 00:00:00  1
+
+        Downsample the series into 3 minute bins and close the right side of
+        the bin interval, but label each bin using the right edge instead of
+        the left.
+
+        >>> (df.groupby("a").resample("3min", closed="right", label="right").sum())
+                                 b
+        a
+        0   2000-01-01 00:00:00  1
+            2000-01-01 00:03:00  2
+        5   2000-01-01 00:03:00  1
+        """
+        from pandas.core.resample import get_resampler_for_grouping
+
+        if include_groups:
+            raise ValueError("include_groups=True is no longer allowed.")
+
+        return get_resampler_for_grouping(self, rule, *args, **kwargs)
+
+    @final
+    def rolling(
+        self,
+        window: int | datetime.timedelta | str | BaseOffset | BaseIndexer,
+        min_periods: int | None = None,
+        center: bool = False,
+        win_type: str | None = None,
+        on: str | None = None,
+        closed: IntervalClosedType | None = None,
+        method: str = "single",
+    ) -> RollingGroupby:
+        """
+        Return a rolling grouper, providing rolling functionality per group.
+
+        Parameters
+        ----------
+        window : int, timedelta, str, offset, or BaseIndexer subclass
+            Interval of the moving window.
+
+            If an integer, the delta between the start and end of each window.
+            The number of points in the window depends on the ``closed`` argument.
+
+            If a timedelta, str, or offset, the time period of each window. Each
+            window will be a variable sized based on the observations included in
+            the time-period. This is only valid for datetimelike indexes.
+            To learn more about the offsets & frequency strings, please see
+            :ref:`this link<timeseries.offset_aliases>`.
+
+            If a BaseIndexer subclass, the window boundaries
+            based on the defined ``get_window_bounds`` method. Additional rolling
+            keyword arguments, namely ``min_periods``, ``center``, ``closed`` and
+            ``step`` will be passed to ``get_window_bounds``.
+
+        min_periods : int, default None
+            Minimum number of observations in window required to have a value;
+            otherwise, result is ``np.nan``.
+
+            For a window that is specified by an offset,
+            ``min_periods`` will default to 1.
+
+            For a window that is specified by an integer, ``min_periods`` will default
+            to the size of the window.
+
+        center : bool, default False
+            If False, set the window labels as the right edge of the window index.
+
+            If True, set the window labels as the center of the window index.
+
+        win_type : str, default None
+            If ``None``, all points are evenly weighted.
+
+            If a string, it must be a valid `scipy.signal window function
+            <https://docs.scipy.org/doc/scipy/reference/signal.windows.html#module-scipy.signal.windows>`__.
+
+            Certain Scipy window types require additional parameters to be passed
+            in the aggregation function. The additional parameters must match
+            the keywords specified in the Scipy window type method signature.
+
+        on : str, optional
+            For a DataFrame, a column label or Index level on which
+            to calculate the rolling window, rather than the DataFrame's index.
+
+            Provided integer column is ignored and excluded from result since
+            an integer index is not used to calculate the rolling window.
+
+        closed : str, default None
+            Determines the inclusivity of points in the window
+
+            If ``'right'``, uses the window (first, last] meaning the last point
+            is included in the calculations.
+
+            If ``'left'``, uses the window [first, last) meaning the first point
+            is included in the calculations.
+
+            If ``'both'``, uses the window [first, last] meaning all points in
+            the window are included in the calculations.
+
+            If ``'neither'``, uses the window (first, last) meaning the first
+            and last points in the window are excluded from calculations.
+
+            () and [] are referencing open and closed set
+            notation respetively.
+
+            Default ``None`` (``'right'``).
+
+        method : str {'single', 'table'}, default 'single'
+            Execute the rolling operation per single column or row (``'single'``)
+            or over the entire object (``'table'``).
+
+            This argument is only implemented when specifying ``engine='numba'``
+            in the method call.
+
+        Returns
+        -------
+        pandas.api.typing.RollingGroupby
+            Return a new grouper with our rolling appended.
+
+        See Also
+        --------
+        Series.rolling : Calling object with Series data.
+        DataFrame.rolling : Calling object with DataFrames.
+        Series.groupby : Apply a function groupby to a Series.
+        DataFrame.groupby : Apply a function groupby.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "A": [1, 1, 2, 2],
+        ...         "B": [1, 2, 3, 4],
+        ...         "C": [0.362, 0.227, 1.267, -0.562],
+        ...     }
+        ... )
+        >>> df
+              A  B      C
+        0     1  1  0.362
+        1     1  2  0.227
+        2     2  3  1.267
+        3     2  4 -0.562
+
+        >>> df.groupby("A").rolling(2).sum()
+            B      C
+        A
+        1 0  NaN    NaN
+          1  3.0  0.589
+        2 2  NaN    NaN
+          3  7.0  0.705
+
+        >>> df.groupby("A").rolling(2, min_periods=1).sum()
+            B      C
+        A
+        1 0  1.0  0.362
+          1  3.0  0.589
+        2 2  3.0  1.267
+          3  7.0  0.705
+
+        >>> df.groupby("A").rolling(2, on="B").sum()
+            B      C
+        A
+        1 0  1    NaN
+          1  2  0.589
+        2 2  3    NaN
+          3  4  0.705
+        """
+        from pandas.core.window import RollingGroupby
+
+        return RollingGroupby(
+            self._selected_obj,
+            window=window,
+            min_periods=min_periods,
+            center=center,
+            win_type=win_type,
+            on=on,
+            closed=closed,
+            method=method,
+            _grouper=self._grouper,
+            _as_index=self.as_index,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Appender(_common_see_also)
+    def expanding(self, *args, **kwargs) -> ExpandingGroupby:
+        """
+        Return an expanding grouper, providing expanding
+        functionality per group.
+
+        Returns
+        -------
+        pandas.api.typing.ExpandingGroupby
+        """
+        from pandas.core.window import ExpandingGroupby
+
+        return ExpandingGroupby(
+            self._selected_obj,
+            *args,
+            _grouper=self._grouper,
+            **kwargs,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Appender(_common_see_also)
+    def ewm(self, *args, **kwargs) -> ExponentialMovingWindowGroupby:
+        """
+        Return an ewm grouper, providing ewm functionality per group.
+
+        Returns
+        -------
+        pandas.api.typing.ExponentialMovingWindowGroupby
+        """
+        from pandas.core.window import ExponentialMovingWindowGroupby
+
+        return ExponentialMovingWindowGroupby(
+            self._selected_obj,
+            *args,
+            _grouper=self._grouper,
+            **kwargs,
+        )
+
+    @final
+    def _fill(self, direction: Literal["ffill", "bfill"], limit: int | None = None):
+        """
+        Shared function for `pad` and `backfill` to call Cython method.
+
+        Parameters
+        ----------
+        direction : {'ffill', 'bfill'}
+            Direction passed to underlying Cython function. `bfill` will cause
+            values to be filled backwards. `ffill` and any other values will
+            default to a forward fill
+        limit : int, default None
+            Maximum number of consecutive values to fill. If `None`, this
+            method will convert to -1 prior to passing to Cython
+
+        Returns
+        -------
+        `Series` or `DataFrame` with filled values
+
+        See Also
+        --------
+        pad : Returns Series with minimum number of char in object.
+        backfill : Backward fill the missing values in the dataset.
+        """
+        # Need int value for Cython
+        if limit is None:
+            limit = -1
+
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+
+        col_func = partial(
+            libgroupby.group_fillna_indexer,
+            labels=ids,
+            limit=limit,
+            compute_ffill=(direction == "ffill"),
+            ngroups=ngroups,
+        )
+
+        def blk_func(values: ArrayLike) -> ArrayLike:
+            mask = isna(values)
+            if values.ndim == 1:
+                indexer = np.empty(values.shape, dtype=np.intp)
+                col_func(out=indexer, mask=mask)  # type: ignore[arg-type]
+                return algorithms.take_nd(values, indexer)
+
+            else:
+                # We broadcast algorithms.take_nd analogous to
+                #  np.take_along_axis
+                if isinstance(values, np.ndarray):
+                    dtype = values.dtype
+                    if self._grouper.has_dropped_na:
+                        # dropped null groups give rise to nan in the result
+                        dtype = ensure_dtype_can_hold_na(values.dtype)
+                    out = np.empty(values.shape, dtype=dtype)
+                else:
+                    # Note: we only get here with backfill/pad,
+                    #  so if we have a dtype that cannot hold NAs,
+                    #  then there will be no -1s in indexer, so we can use
+                    #  the original dtype (no need to ensure_dtype_can_hold_na)
+                    out = type(values)._empty(values.shape, dtype=values.dtype)
+
+                for i, value_element in enumerate(values):
+                    # call group_fillna_indexer column-wise
+                    indexer = np.empty(values.shape[1], dtype=np.intp)
+                    col_func(out=indexer, mask=mask[i])
+                    out[i, :] = algorithms.take_nd(value_element, indexer)
+                return out
+
+        mgr = self._get_data_to_aggregate()
+        res_mgr = mgr.apply(blk_func)
+
+        new_obj = self._wrap_agged_manager(res_mgr)
+        new_obj.index = self.obj.index
+        return new_obj
+
+    @final
+    @Substitution(name="groupby")
+    def ffill(self, limit: int | None = None):
+        """
+        Forward fill the values.
+
+        Parameters
+        ----------
+        limit : int, optional
+            Limit of how many values to fill.
+
+        Returns
+        -------
+        Series or DataFrame
+            Object with missing values filled.
+
+        See Also
+        --------
+        Series.ffill: Returns Series with minimum number of char in object.
+        DataFrame.ffill: Object with missing values filled or None if inplace=True.
+        Series.fillna: Fill NaN values of a Series.
+        DataFrame.fillna: Fill NaN values of a DataFrame.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> key = [0, 0, 1, 1]
+        >>> ser = pd.Series([np.nan, 2, 3, np.nan], index=key)
+        >>> ser
+        0    NaN
+        0    2.0
+        1    3.0
+        1    NaN
+        dtype: float64
+        >>> ser.groupby(level=0).ffill()
+        0    NaN
+        0    2.0
+        1    3.0
+        1    3.0
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "key": [0, 0, 1, 1, 1],
+        ...         "A": [np.nan, 2, np.nan, 3, np.nan],
+        ...         "B": [2, 3, np.nan, np.nan, np.nan],
+        ...         "C": [np.nan, np.nan, 2, np.nan, np.nan],
+        ...     }
+        ... )
+        >>> df
+           key    A    B   C
+        0    0  NaN  2.0 NaN
+        1    0  2.0  3.0 NaN
+        2    1  NaN  NaN 2.0
+        3    1  3.0  NaN NaN
+        4    1  NaN  NaN NaN
+
+        Propagate non-null values forward or backward within each group along columns.
+
+        >>> df.groupby("key").ffill()
+             A    B   C
+        0  NaN  2.0 NaN
+        1  2.0  3.0 NaN
+        2  NaN  NaN 2.0
+        3  3.0  NaN 2.0
+        4  3.0  NaN 2.0
+
+        Propagate non-null values forward or backward within each group along rows.
+
+        >>> df.T.groupby(np.array([0, 0, 1, 1])).ffill().T
+           key    A    B    C
+        0  0.0  0.0  2.0  2.0
+        1  0.0  2.0  3.0  3.0
+        2  1.0  1.0  NaN  2.0
+        3  1.0  3.0  NaN  NaN
+        4  1.0  1.0  NaN  NaN
+
+        Only replace the first NaN element within a group along columns.
+
+        >>> df.groupby("key").ffill(limit=1)
+             A    B    C
+        0  NaN  2.0  NaN
+        1  2.0  3.0  NaN
+        2  NaN  NaN  2.0
+        3  3.0  NaN  2.0
+        4  3.0  NaN  NaN
+        """
+        return self._fill("ffill", limit=limit)
+
+    @final
+    @Substitution(name="groupby")
+    def bfill(self, limit: int | None = None):
+        """
+        Backward fill the values.
+
+        Parameters
+        ----------
+        limit : int, optional
+            Limit of how many values to fill.
+
+        Returns
+        -------
+        Series or DataFrame
+            Object with missing values filled.
+
+        See Also
+        --------
+        Series.bfill :  Backward fill the missing values in the dataset.
+        DataFrame.bfill:  Backward fill the missing values in the dataset.
+        Series.fillna: Fill NaN values of a Series.
+        DataFrame.fillna: Fill NaN values of a DataFrame.
+
+        Examples
+        --------
+
+        With Series:
+
+        >>> index = ["Falcon", "Falcon", "Parrot", "Parrot", "Parrot"]
+        >>> s = pd.Series([None, 1, None, None, 3], index=index)
+        >>> s
+        Falcon    NaN
+        Falcon    1.0
+        Parrot    NaN
+        Parrot    NaN
+        Parrot    3.0
+        dtype: float64
+        >>> s.groupby(level=0).bfill()
+        Falcon    1.0
+        Falcon    1.0
+        Parrot    3.0
+        Parrot    3.0
+        Parrot    3.0
+        dtype: float64
+        >>> s.groupby(level=0).bfill(limit=1)
+        Falcon    1.0
+        Falcon    1.0
+        Parrot    NaN
+        Parrot    3.0
+        Parrot    3.0
+        dtype: float64
+
+        With DataFrame:
+
+        >>> df = pd.DataFrame(
+        ...     {"A": [1, None, None, None, 4], "B": [None, None, 5, None, 7]},
+        ...     index=index,
+        ... )
+        >>> df
+                  A	    B
+        Falcon	1.0	  NaN
+        Falcon	NaN	  NaN
+        Parrot	NaN	  5.0
+        Parrot	NaN	  NaN
+        Parrot	4.0	  7.0
+        >>> df.groupby(level=0).bfill()
+                  A	    B
+        Falcon	1.0	  NaN
+        Falcon	NaN	  NaN
+        Parrot	4.0	  5.0
+        Parrot	4.0	  7.0
+        Parrot	4.0	  7.0
+        >>> df.groupby(level=0).bfill(limit=1)
+                  A	    B
+        Falcon	1.0	  NaN
+        Falcon	NaN	  NaN
+        Parrot	NaN	  5.0
+        Parrot	4.0	  7.0
+        Parrot	4.0	  7.0
+        """
+        return self._fill("bfill", limit=limit)
+
+    @final
+    @property
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def nth(self) -> GroupByNthSelector:
+        """
+        Take the nth row from each group if n is an int, otherwise a subset of rows.
+
+        Can be either a call or an index. dropna is not available with index notation.
+        Index notation accepts a comma separated list of integers and slices.
+
+        If dropna, will take the nth non-null row, dropna is either
+        'all' or 'any'; this is equivalent to calling dropna(how=dropna)
+        before the groupby.
+
+        Returns
+        -------
+        Series or DataFrame
+            N-th value within each group.
+        %(see_also)s
+        Examples
+        --------
+
+        >>> df = pd.DataFrame(
+        ...     {"A": [1, 1, 2, 1, 2], "B": [np.nan, 2, 3, 4, 5]}, columns=["A", "B"]
+        ... )
+        >>> g = df.groupby("A")
+        >>> g.nth(0)
+           A   B
+        0  1 NaN
+        2  2 3.0
+        >>> g.nth(1)
+           A   B
+        1  1 2.0
+        4  2 5.0
+        >>> g.nth(-1)
+           A   B
+        3  1 4.0
+        4  2 5.0
+        >>> g.nth([0, 1])
+           A   B
+        0  1 NaN
+        1  1 2.0
+        2  2 3.0
+        4  2 5.0
+        >>> g.nth(slice(None, -1))
+           A   B
+        0  1 NaN
+        1  1 2.0
+        2  2 3.0
+
+        Index notation may also be used
+
+        >>> g.nth[0, 1]
+           A   B
+        0  1 NaN
+        1  1 2.0
+        2  2 3.0
+        4  2 5.0
+        >>> g.nth[:-1]
+           A   B
+        0  1 NaN
+        1  1 2.0
+        2  2 3.0
+
+        Specifying `dropna` allows ignoring ``NaN`` values
+
+        >>> g.nth(0, dropna="any")
+           A   B
+        1  1 2.0
+        2  2 3.0
+
+        When the specified ``n`` is larger than any of the groups, an
+        empty DataFrame is returned
+
+        >>> g.nth(3, dropna="any")
+        Empty DataFrame
+        Columns: [A, B]
+        Index: []
+        """
+        return GroupByNthSelector(self)
+
+    def _nth(
+        self,
+        n: PositionalIndexer | tuple,
+        dropna: Literal["any", "all", None] = None,
+    ) -> NDFrameT:
+        if not dropna:
+            mask = self._make_mask_from_positional_indexer(n)
+
+            ids = self._grouper.ids
+
+            # Drop NA values in grouping
+            mask = mask & (ids != -1)
+
+            out = self._mask_selected_obj(mask)
+            return out
+
+        # dropna is truthy
+        if not is_integer(n):
+            raise ValueError("dropna option only supported for an integer argument")
+
+        if dropna not in ["any", "all"]:
+            # Note: when agg-ing picker doesn't raise this, just returns NaN
+            raise ValueError(
+                "For a DataFrame or Series groupby.nth, dropna must be "
+                "either None, 'any' or 'all', "
+                f"(was passed {dropna})."
+            )
+
+        # old behaviour, but with all and any support for DataFrames.
+        # modified in GH 7559 to have better perf
+        n = cast(int, n)
+        dropped = self._selected_obj.dropna(how=dropna, axis=0)
+
+        # get a new grouper for our dropped obj
+        grouper: np.ndarray | Index | ops.BaseGrouper
+        if len(dropped) == len(self._selected_obj):
+            # Nothing was dropped, can use the same grouper
+            grouper = self._grouper
+        else:
+            # we don't have the grouper info available
+            # (e.g. we have selected out
+            # a column that is not in the current object)
+            axis = self._grouper.axis
+            grouper = self._grouper.codes_info[axis.isin(dropped.index)]
+            if self._grouper.has_dropped_na:
+                # Null groups need to still be encoded as -1 when passed to groupby
+                nulls = grouper == -1
+                # error: No overload variant of "where" matches argument types
+                #        "Any", "NAType", "Any"
+                values = np.where(nulls, NA, grouper)  # type: ignore[call-overload]
+                grouper = Index(values, dtype="Int64")
+
+        grb = dropped.groupby(grouper, as_index=self.as_index, sort=self.sort)
+        return grb.nth(n)
+
+    @final
+    def quantile(
+        self,
+        q: float | AnyArrayLike = 0.5,
+        interpolation: Literal[
+            "linear", "lower", "higher", "nearest", "midpoint"
+        ] = "linear",
+        numeric_only: bool = False,
+    ):
+        """
+        Return group values at the given quantile, a la numpy.percentile.
+
+        Parameters
+        ----------
+        q : float or array-like, default 0.5 (50% quantile)
+            Value(s) between 0 and 1 providing the quantile(s) to compute.
+        interpolation : {'linear', 'lower', 'higher', 'midpoint', 'nearest'}
+            Method to use when the desired quantile falls between two points.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        Returns
+        -------
+        Series or DataFrame
+            Return type determined by caller of GroupBy object.
+
+        See Also
+        --------
+        Series.quantile : Similar method for Series.
+        DataFrame.quantile : Similar method for DataFrame.
+        numpy.percentile : NumPy method to compute qth percentile.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [["a", 1], ["a", 2], ["a", 3], ["b", 1], ["b", 3], ["b", 5]],
+        ...     columns=["key", "val"],
+        ... )
+        >>> df.groupby("key").quantile()
+            val
+        key
+        a    2.0
+        b    3.0
+        """
+        mgr = self._get_data_to_aggregate(numeric_only=numeric_only, name="quantile")
+        obj = self._wrap_agged_manager(mgr)
+        splitter = self._grouper._get_splitter(obj)
+        sdata = splitter._sorted_data
+
+        starts, ends = lib.generate_slices(splitter._slabels, splitter.ngroups)
+
+        def pre_processor(vals: ArrayLike) -> tuple[np.ndarray, DtypeObj | None]:
+            if isinstance(vals.dtype, StringDtype) or is_object_dtype(vals.dtype):
+                raise TypeError(
+                    f"dtype '{vals.dtype}' does not support operation 'quantile'"
+                )
+
+            inference: DtypeObj | None = None
+            if isinstance(vals, BaseMaskedArray) and is_numeric_dtype(vals.dtype):
+                out = vals.to_numpy(dtype=float, na_value=np.nan)
+                inference = vals.dtype
+            elif is_integer_dtype(vals.dtype):
+                if isinstance(vals, ExtensionArray):
+                    out = vals.to_numpy(dtype=float, na_value=np.nan)
+                else:
+                    out = vals
+                inference = np.dtype(np.int64)
+            elif is_bool_dtype(vals.dtype) and isinstance(vals, ExtensionArray):
+                out = vals.to_numpy(dtype=float, na_value=np.nan)
+            elif is_bool_dtype(vals.dtype):
+                # GH#51424 remove to match Series/DataFrame behavior
+                raise TypeError("Cannot use quantile with bool dtype")
+            elif needs_i8_conversion(vals.dtype):
+                inference = vals.dtype
+                # In this case we need to delay the casting until after the
+                #  np.lexsort below.
+                # error: Incompatible return value type (got
+                # "Tuple[Union[ExtensionArray, ndarray[Any, Any]], Union[Any,
+                # ExtensionDtype]]", expected "Tuple[ndarray[Any, Any],
+                # Optional[Union[dtype[Any], ExtensionDtype]]]")
+                return vals, inference  # type: ignore[return-value]
+            elif isinstance(vals, ExtensionArray) and is_float_dtype(vals.dtype):
+                inference = np.dtype(np.float64)
+                out = vals.to_numpy(dtype=float, na_value=np.nan)
+            else:
+                out = np.asarray(vals)
+
+            return out, inference
+
+        def post_processor(
+            vals: np.ndarray,
+            inference: DtypeObj | None,
+            result_mask: np.ndarray | None,
+            orig_vals: ArrayLike,
+        ) -> ArrayLike:
+            if inference:
+                # Check for edge case
+                if isinstance(orig_vals, BaseMaskedArray):
+                    assert result_mask is not None  # for mypy
+
+                    if interpolation in {"linear", "midpoint"} and not is_float_dtype(
+                        orig_vals
+                    ):
+                        return FloatingArray(vals, result_mask)
+                    else:
+                        # Item "ExtensionDtype" of "Union[ExtensionDtype, str,
+                        # dtype[Any], Type[object]]" has no attribute "numpy_dtype"
+                        # [union-attr]
+                        with warnings.catch_warnings():
+                            # vals.astype with nan can warn with numpy >1.24
+                            warnings.filterwarnings("ignore", category=RuntimeWarning)
+                            return type(orig_vals)(
+                                vals.astype(
+                                    inference.numpy_dtype  # type: ignore[union-attr]
+                                ),
+                                result_mask,
+                            )
+
+                elif not (
+                    is_integer_dtype(inference)
+                    and interpolation in {"linear", "midpoint"}
+                ):
+                    if needs_i8_conversion(inference):
+                        # error: Item "ExtensionArray" of "Union[ExtensionArray,
+                        # ndarray[Any, Any]]" has no attribute "_ndarray"
+                        vals = vals.astype("i8").view(
+                            orig_vals._ndarray.dtype  # type: ignore[union-attr]
+                        )
+                        # error: Item "ExtensionArray" of "Union[ExtensionArray,
+                        # ndarray[Any, Any]]" has no attribute "_from_backing_data"
+                        return orig_vals._from_backing_data(  # type: ignore[union-attr]
+                            vals
+                        )
+
+                    assert isinstance(inference, np.dtype)  # for mypy
+                    return vals.astype(inference)
+
+            return vals
+
+        if is_scalar(q):
+            qs = np.array([q], dtype=np.float64)
+            pass_qs: None | np.ndarray = None
+        else:
+            qs = np.asarray(q, dtype=np.float64)
+            pass_qs = qs
+
+        ids = self._grouper.ids
+        ngroups = self._grouper.ngroups
+        if self.dropna:
+            # splitter drops NA groups, we need to do the same
+            ids = ids[ids >= 0]
+        nqs = len(qs)
+
+        func = partial(
+            libgroupby.group_quantile,
+            labels=ids,
+            qs=qs,
+            interpolation=interpolation,
+            starts=starts,
+            ends=ends,
+        )
+
+        def blk_func(values: ArrayLike) -> ArrayLike:
+            orig_vals = values
+            if isinstance(values, BaseMaskedArray):
+                mask = values._mask
+                result_mask = np.zeros((ngroups, nqs), dtype=np.bool_)
+            else:
+                mask = isna(values)
+                result_mask = None
+
+            is_datetimelike = needs_i8_conversion(values.dtype)
+
+            vals, inference = pre_processor(values)
+
+            ncols = 1
+            if vals.ndim == 2:
+                ncols = vals.shape[0]
+
+            out = np.empty((ncols, ngroups, nqs), dtype=np.float64)
+
+            if is_datetimelike:
+                vals = vals.view("i8")
+
+            if vals.ndim == 1:
+                # EA is always 1d
+                func(
+                    out[0],
+                    values=vals,
+                    mask=mask,  # type: ignore[arg-type]
+                    result_mask=result_mask,
+                    is_datetimelike=is_datetimelike,
+                )
+            else:
+                for i in range(ncols):
+                    func(
+                        out[i],
+                        values=vals[i],
+                        mask=mask[i],
+                        result_mask=None,
+                        is_datetimelike=is_datetimelike,
+                    )
+
+            if vals.ndim == 1:
+                out = out.ravel("K")
+                if result_mask is not None:
+                    result_mask = result_mask.ravel("K")
+            else:
+                out = out.reshape(ncols, ngroups * nqs)
+
+            return post_processor(out, inference, result_mask, orig_vals)
+
+        res_mgr = sdata._mgr.grouped_reduce(blk_func)
+
+        res = self._wrap_agged_manager(res_mgr)
+        return self._wrap_aggregated_output(res, qs=pass_qs)
+
+    @final
+    @Substitution(name="groupby")
+    def ngroup(self, ascending: bool = True):
+        """
+        Number each group from 0 to the number of groups - 1.
+
+        This is the enumerative complement of cumcount.  Note that the
+        numbers given to the groups match the order in which the groups
+        would be seen when iterating over the groupby object, not the
+        order they are first observed.
+
+        Groups with missing keys (where `pd.isna()` is True) will be labeled with `NaN`
+        and will be skipped from the count.
+
+        Parameters
+        ----------
+        ascending : bool, default True
+            If False, number in reverse, from number of group - 1 to 0.
+
+        Returns
+        -------
+        Series
+            Unique numbers for each group.
+
+        See Also
+        --------
+        .cumcount : Number the rows in each group.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"color": ["red", None, "red", "blue", "blue", "red"]})
+        >>> df
+           color
+        0    red
+        1   None
+        2    red
+        3   blue
+        4   blue
+        5    red
+        >>> df.groupby("color").ngroup()
+        0    1.0
+        1    NaN
+        2    1.0
+        3    0.0
+        4    0.0
+        5    1.0
+        dtype: float64
+        >>> df.groupby("color", dropna=False).ngroup()
+        0    1
+        1    2
+        2    1
+        3    0
+        4    0
+        5    1
+        dtype: int64
+        >>> df.groupby("color", dropna=False).ngroup(ascending=False)
+        0    1
+        1    0
+        2    1
+        3    2
+        4    2
+        5    1
+        dtype: int64
+        """
+        obj = self._obj_with_exclusions
+        index = obj.index
+        comp_ids = self._grouper.ids
+
+        dtype: type
+        if self._grouper.has_dropped_na:
+            comp_ids = np.where(comp_ids == -1, np.nan, comp_ids)
+            dtype = np.float64
+        else:
+            dtype = np.int64
+
+        if any(ping._passed_categorical for ping in self._grouper.groupings):
+            # comp_ids reflect non-observed groups, we need only observed
+            comp_ids = rank_1d(comp_ids, ties_method="dense") - 1
+
+        result = self._obj_1d_constructor(comp_ids, index, dtype=dtype)
+        if not ascending:
+            result = self.ngroups - 1 - result
+        return result
+
+    @final
+    @Substitution(name="groupby")
+    def cumcount(self, ascending: bool = True):
+        """
+        Number each item in each group from 0 to the length of that group - 1.
+
+        Essentially this is equivalent to
+
+        .. code-block:: python
+
+            self.apply(lambda x: pd.Series(np.arange(len(x)), x.index))
+
+        Parameters
+        ----------
+        ascending : bool, default True
+            If False, number in reverse, from length of group - 1 to 0.
+
+        Returns
+        -------
+        Series
+            Sequence number of each element within each group.
+
+        See Also
+        --------
+        .ngroup : Number the groups themselves.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([["a"], ["a"], ["a"], ["b"], ["b"], ["a"]], columns=["A"])
+        >>> df
+           A
+        0  a
+        1  a
+        2  a
+        3  b
+        4  b
+        5  a
+        >>> df.groupby("A").cumcount()
+        0    0
+        1    1
+        2    2
+        3    0
+        4    1
+        5    3
+        dtype: int64
+        >>> df.groupby("A").cumcount(ascending=False)
+        0    3
+        1    2
+        2    1
+        3    1
+        4    0
+        5    0
+        dtype: int64
+        """
+        index = self._obj_with_exclusions.index
+        cumcounts = self._cumcount_array(ascending=ascending)
+        return self._obj_1d_constructor(cumcounts, index)
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def rank(
+        self,
+        method: str = "average",
+        ascending: bool = True,
+        na_option: str = "keep",
+        pct: bool = False,
+    ) -> NDFrameT:
+        """
+        Provide the rank of values within each group.
+
+        Parameters
+        ----------
+        method : {'average', 'min', 'max', 'first', 'dense'}, default 'average'
+            * average: average rank of group.
+            * min: lowest rank in group.
+            * max: highest rank in group.
+            * first: ranks assigned in order they appear in the array.
+            * dense: like 'min', but rank always increases by 1 between groups.
+        ascending : bool, default True
+            False for ranks by high (1) to low (N).
+        na_option : {'keep', 'top', 'bottom'}, default 'keep'
+            * keep: leave NA values where they are.
+            * top: smallest rank if ascending.
+            * bottom: smallest rank if descending.
+        pct : bool, default False
+            Compute percentage rank of data within each group.
+
+        Returns
+        -------
+        DataFrame
+            The ranking of values within each group.
+        %(see_also)s
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "group": ["a", "a", "a", "a", "a", "b", "b", "b", "b", "b"],
+        ...         "value": [2, 4, 2, 3, 5, 1, 2, 4, 1, 5],
+        ...     }
+        ... )
+        >>> df
+          group  value
+        0     a      2
+        1     a      4
+        2     a      2
+        3     a      3
+        4     a      5
+        5     b      1
+        6     b      2
+        7     b      4
+        8     b      1
+        9     b      5
+        >>> for method in ["average", "min", "max", "dense", "first"]:
+        ...     df[f"{method}_rank"] = df.groupby("group")["value"].rank(method)
+        >>> df
+          group  value  average_rank  min_rank  max_rank  dense_rank  first_rank
+        0     a      2           1.5       1.0       2.0         1.0         1.0
+        1     a      4           4.0       4.0       4.0         3.0         4.0
+        2     a      2           1.5       1.0       2.0         1.0         2.0
+        3     a      3           3.0       3.0       3.0         2.0         3.0
+        4     a      5           5.0       5.0       5.0         4.0         5.0
+        5     b      1           1.5       1.0       2.0         1.0         1.0
+        6     b      2           3.0       3.0       3.0         2.0         3.0
+        7     b      4           4.0       4.0       4.0         3.0         4.0
+        8     b      1           1.5       1.0       2.0         1.0         2.0
+        9     b      5           5.0       5.0       5.0         4.0         5.0
+        """
+        if na_option not in {"keep", "top", "bottom"}:
+            msg = "na_option must be one of 'keep', 'top', or 'bottom'"
+            raise ValueError(msg)
+
+        kwargs = {
+            "ties_method": method,
+            "ascending": ascending,
+            "na_option": na_option,
+            "pct": pct,
+        }
+
+        return self._cython_transform(
+            "rank",
+            numeric_only=False,
+            **kwargs,
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def cumprod(self, numeric_only: bool = False, *args, **kwargs) -> NDFrameT:
+        """
+        Cumulative product for each group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        *args : tuple
+            Positional arguments to be passed to `func`.
+        **kwargs : dict
+            Additional/specific keyword arguments to be passed to the function,
+            such as `numeric_only` and `skipna`.
+
+        Returns
+        -------
+        Series or DataFrame
+            Cumulative product for each group. Same object type as the caller.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([6, 2, 0], index=lst)
+        >>> ser
+        a    6
+        a    2
+        b    0
+        dtype: int64
+        >>> ser.groupby(level=0).cumprod()
+        a    6
+        a   12
+        b    0
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["cow", "horse", "bull"]
+        ... )
+        >>> df
+                a   b   c
+        cow     1   8   2
+        horse   1   2   5
+        bull    2   6   9
+        >>> df.groupby("a").groups
+        {1: ['cow', 'horse'], 2: ['bull']}
+        >>> df.groupby("a").cumprod()
+                b   c
+        cow     8   2
+        horse  16  10
+        bull    6   9
+        """
+        nv.validate_groupby_func("cumprod", args, kwargs, ["skipna"])
+        return self._cython_transform("cumprod", numeric_only, **kwargs)
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def cumsum(self, numeric_only: bool = False, *args, **kwargs) -> NDFrameT:
+        """
+        Cumulative sum for each group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        *args : tuple
+            Positional arguments to be passed to `func`.
+        **kwargs : dict
+            Additional/specific keyword arguments to be passed to the function,
+            such as `numeric_only` and `skipna`.
+
+        Returns
+        -------
+        Series or DataFrame
+            Cumulative sum for each group. Same object type as the caller.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b"]
+        >>> ser = pd.Series([6, 2, 0], index=lst)
+        >>> ser
+        a    6
+        a    2
+        b    0
+        dtype: int64
+        >>> ser.groupby(level=0).cumsum()
+        a    6
+        a    8
+        b    0
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 2, 5], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["fox", "gorilla", "lion"]
+        ... )
+        >>> df
+                  a   b   c
+        fox       1   8   2
+        gorilla   1   2   5
+        lion      2   6   9
+        >>> df.groupby("a").groups
+        {1: ['fox', 'gorilla'], 2: ['lion']}
+        >>> df.groupby("a").cumsum()
+                  b   c
+        fox       8   2
+        gorilla  10   7
+        lion      6   9
+        """
+        nv.validate_groupby_func("cumsum", args, kwargs, ["skipna"])
+        return self._cython_transform("cumsum", numeric_only, **kwargs)
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def cummin(
+        self,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> NDFrameT:
+        """
+        Cumulative min for each group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+        **kwargs : dict, optional
+            Additional keyword arguments to be passed to the function, such as `skipna`,
+            to control whether NA/null values are ignored.
+
+        Returns
+        -------
+        Series or DataFrame
+            Cumulative min for each group. Same object type as the caller.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([1, 6, 2, 3, 0, 4], index=lst)
+        >>> ser
+        a    1
+        a    6
+        a    2
+        b    3
+        b    0
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).cummin()
+        a    1
+        a    1
+        a    1
+        b    3
+        b    0
+        b    0
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 0, 2], [1, 1, 5], [6, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["snake", "rabbit", "turtle"]
+        ... )
+        >>> df
+                a   b   c
+        snake   1   0   2
+        rabbit  1   1   5
+        turtle  6   6   9
+        >>> df.groupby("a").groups
+        {1: ['snake', 'rabbit'], 6: ['turtle']}
+        >>> df.groupby("a").cummin()
+                b   c
+        snake   0   2
+        rabbit  0   2
+        turtle  6   9
+        """
+        skipna = kwargs.get("skipna", True)
+        return self._cython_transform(
+            "cummin", numeric_only=numeric_only, skipna=skipna
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def cummax(
+        self,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> NDFrameT:
+        """
+        Cumulative max for each group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+        **kwargs : dict, optional
+            Additional keyword arguments to be passed to the function, such as `skipna`,
+            to control whether NA/null values are ignored.
+
+        Returns
+        -------
+        Series or DataFrame
+            Cumulative max for each group. Same object type as the caller.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([1, 6, 2, 3, 1, 4], index=lst)
+        >>> ser
+        a    1
+        a    6
+        a    2
+        b    3
+        b    1
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).cummax()
+        a    1
+        a    6
+        a    6
+        b    3
+        b    3
+        b    4
+        dtype: int64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 8, 2], [1, 1, 0], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data, columns=["a", "b", "c"], index=["cow", "horse", "bull"]
+        ... )
+        >>> df
+                a   b   c
+        cow     1   8   2
+        horse   1   1   0
+        bull    2   6   9
+        >>> df.groupby("a").groups
+        {1: ['cow', 'horse'], 2: ['bull']}
+        >>> df.groupby("a").cummax()
+                b   c
+        cow     8   2
+        horse   8   2
+        bull    6   9
+        """
+        skipna = kwargs.get("skipna", True)
+        return self._cython_transform(
+            "cummax", numeric_only=numeric_only, skipna=skipna
+        )
+
+    @final
+    @Substitution(name="groupby")
+    def shift(
+        self,
+        periods: int | Sequence[int] = 1,
+        freq=None,
+        fill_value=lib.no_default,
+        suffix: str | None = None,
+    ):
+        """
+        Shift each group by periods observations.
+
+        If freq is passed, the index will be increased using the periods and the freq.
+
+        Parameters
+        ----------
+        periods : int | Sequence[int], default 1
+            Number of periods to shift. If a list of values, shift each group by
+            each period.
+        freq : str, optional
+            Frequency string.
+        fill_value : optional
+            The scalar value to use for newly introduced missing values.
+
+            .. versionchanged:: 2.1.0
+                Will raise a ``ValueError`` if ``freq`` is provided too.
+
+        suffix : str, optional
+            A string to add to each shifted column if there are multiple periods.
+            Ignored otherwise.
+
+        Returns
+        -------
+        Series or DataFrame
+            Object shifted within each group.
+
+        See Also
+        --------
+        Index.shift : Shift values of Index.
+
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b", "b"]
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).shift(1)
+        a    NaN
+        a    1.0
+        b    NaN
+        b    3.0
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data,
+        ...     columns=["a", "b", "c"],
+        ...     index=["tuna", "salmon", "catfish", "goldfish"],
+        ... )
+        >>> df
+                   a  b  c
+            tuna   1  2  3
+          salmon   1  5  6
+         catfish   2  5  8
+        goldfish   2  6  9
+        >>> df.groupby("a").shift(1)
+                      b    c
+            tuna    NaN  NaN
+          salmon    2.0  3.0
+         catfish    NaN  NaN
+        goldfish    5.0  8.0
+        """
+        if is_list_like(periods):
+            periods = cast(Sequence, periods)
+            if len(periods) == 0:
+                raise ValueError("If `periods` is an iterable, it cannot be empty.")
+            from pandas.core.reshape.concat import concat
+
+            add_suffix = True
+        else:
+            if not is_integer(periods):
+                raise TypeError(
+                    f"Periods must be integer, but {periods} is {type(periods)}."
+                )
+            if suffix:
+                raise ValueError("Cannot specify `suffix` if `periods` is an int.")
+            periods = [cast(int, periods)]
+            add_suffix = False
+
+        shifted_dataframes = []
+        for period in periods:
+            if not is_integer(period):
+                raise TypeError(
+                    f"Periods must be integer, but {period} is {type(period)}."
+                )
+            period = cast(int, period)
+            if freq is not None:
+                f = lambda x: x.shift(
+                    period,
+                    freq,
+                    0,  # axis
+                    fill_value,
+                )
+                shifted = self._python_apply_general(
+                    f, self._selected_obj, is_transform=True
+                )
+            else:
+                if fill_value is lib.no_default:
+                    fill_value = None
+                ids = self._grouper.ids
+                ngroups = self._grouper.ngroups
+                res_indexer = np.zeros(len(ids), dtype=np.int64)
+
+                libgroupby.group_shift_indexer(res_indexer, ids, ngroups, period)
+
+                obj = self._obj_with_exclusions
+
+                shifted = obj._reindex_with_indexers(
+                    {0: (obj.index, res_indexer)},
+                    fill_value=fill_value,
+                    allow_dups=True,
+                )
+
+            if add_suffix:
+                if isinstance(shifted, Series):
+                    shifted = cast(NDFrameT, shifted.to_frame())
+                shifted = shifted.add_suffix(
+                    f"{suffix}_{period}" if suffix else f"_{period}"
+                )
+            shifted_dataframes.append(cast(Union[Series, DataFrame], shifted))
+
+        return (
+            shifted_dataframes[0]
+            if len(shifted_dataframes) == 1
+            else concat(shifted_dataframes, axis=1)
+        )
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def diff(
+        self,
+        periods: int = 1,
+    ) -> NDFrameT:
+        """
+        First discrete difference of element.
+
+        Calculates the difference of each element compared with another
+        element in the group (default is element in previous row).
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Periods to shift for calculating difference, accepts negative values.
+
+        Returns
+        -------
+        Series or DataFrame
+            First differences.
+        %(see_also)s
+        Examples
+        --------
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([7, 2, 8, 4, 3, 3], index=lst)
+        >>> ser
+        a     7
+        a     2
+        a     8
+        b     4
+        b     3
+        b     3
+        dtype: int64
+        >>> ser.groupby(level=0).diff()
+        a    NaN
+        a   -5.0
+        a    6.0
+        b    NaN
+        b   -1.0
+        b    0.0
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = {"a": [1, 3, 5, 7, 7, 8, 3], "b": [1, 4, 8, 4, 4, 2, 1]}
+        >>> df = pd.DataFrame(
+        ...     data, index=["dog", "dog", "dog", "mouse", "mouse", "mouse", "mouse"]
+        ... )
+        >>> df
+                 a  b
+          dog    1  1
+          dog    3  4
+          dog    5  8
+        mouse    7  4
+        mouse    7  4
+        mouse    8  2
+        mouse    3  1
+        >>> df.groupby(level=0).diff()
+                 a    b
+          dog  NaN  NaN
+          dog  2.0  3.0
+          dog  2.0  4.0
+        mouse  NaN  NaN
+        mouse  0.0  0.0
+        mouse  1.0 -2.0
+        mouse -5.0 -1.0
+        """
+        obj = self._obj_with_exclusions
+        shifted = self.shift(periods=periods)
+
+        # GH45562 - to retain existing behavior and match behavior of Series.diff(),
+        # int8 and int16 are coerced to float32 rather than float64.
+        dtypes_to_f32 = ["int8", "int16"]
+        if obj.ndim == 1:
+            if obj.dtype in dtypes_to_f32:
+                shifted = shifted.astype("float32")
+        else:
+            to_coerce = [c for c, dtype in obj.dtypes.items() if dtype in dtypes_to_f32]
+            if len(to_coerce):
+                shifted = shifted.astype({c: "float32" for c in to_coerce})
+
+        return obj - shifted
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def pct_change(
+        self,
+        periods: int = 1,
+        fill_method: None = None,
+        freq=None,
+    ):
+        """
+        Calculate pct_change of each value to previous entry in group.
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Periods to shift for calculating percentage change. Comparing with
+            a period of 1 means adjacent elements are compared, whereas a period
+            of 2 compares every other element.
+
+        fill_method : None
+            Must be None. This argument will be removed in a future version of pandas.
+
+            .. deprecated:: 2.1
+                All options of `fill_method` are deprecated except `fill_method=None`.
+
+        freq : str, pandas offset object, or None, default None
+            The frequency increment for time series data (e.g., 'M' for month-end).
+            If None, the frequency is inferred from the index. Relevant for time
+            series data only.
+
+        Returns
+        -------
+        Series or DataFrame
+            Percentage changes within each group.
+        %(see_also)s
+        Examples
+        --------
+
+        For SeriesGroupBy:
+
+        >>> lst = ["a", "a", "b", "b"]
+        >>> ser = pd.Series([1, 2, 3, 4], index=lst)
+        >>> ser
+        a    1
+        a    2
+        b    3
+        b    4
+        dtype: int64
+        >>> ser.groupby(level=0).pct_change()
+        a         NaN
+        a    1.000000
+        b         NaN
+        b    0.333333
+        dtype: float64
+
+        For DataFrameGroupBy:
+
+        >>> data = [[1, 2, 3], [1, 5, 6], [2, 5, 8], [2, 6, 9]]
+        >>> df = pd.DataFrame(
+        ...     data,
+        ...     columns=["a", "b", "c"],
+        ...     index=["tuna", "salmon", "catfish", "goldfish"],
+        ... )
+        >>> df
+                   a  b  c
+            tuna   1  2  3
+          salmon   1  5  6
+         catfish   2  5  8
+        goldfish   2  6  9
+        >>> df.groupby("a").pct_change()
+                    b  c
+            tuna    NaN    NaN
+          salmon    1.5  1.000
+         catfish    NaN    NaN
+        goldfish    0.2  0.125
+        """
+        # GH#53491
+        if fill_method is not None:
+            raise ValueError(f"fill_method must be None; got {fill_method=}.")
+
+        # TODO(GH#23918): Remove this conditional for SeriesGroupBy when
+        #  GH#23918 is fixed
+        if freq is not None:
+            f = lambda x: x.pct_change(
+                periods=periods,
+                freq=freq,
+                axis=0,
+            )
+            return self._python_apply_general(f, self._selected_obj, is_transform=True)
+
+        if fill_method is None:  # GH30463
+            op = "ffill"
+        else:
+            op = fill_method
+        filled = getattr(self, op)(limit=0)
+        fill_grp = filled.groupby(self._grouper.codes, group_keys=self.group_keys)
+        shifted = fill_grp.shift(periods=periods, freq=freq)
+        return (filled / shifted) - 1
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def head(self, n: int = 5) -> NDFrameT:
+        """
+        Return first n rows of each group.
+
+        Similar to ``.apply(lambda x: x.head(n))``, but it returns a subset of rows
+        from the original DataFrame with original index and order preserved
+        (``as_index`` flag is ignored).
+
+        Parameters
+        ----------
+        n : int
+            If positive: number of entries to include from start of each group.
+            If negative: number of entries to exclude from end of each group.
+
+        Returns
+        -------
+        Series or DataFrame
+            Subset of original Series or DataFrame as determined by n.
+        %(see_also)s
+        Examples
+        --------
+
+        >>> df = pd.DataFrame([[1, 2], [1, 4], [5, 6]], columns=["A", "B"])
+        >>> df.groupby("A").head(1)
+           A  B
+        0  1  2
+        2  5  6
+        >>> df.groupby("A").head(-1)
+           A  B
+        0  1  2
+        """
+        mask = self._make_mask_from_positional_indexer(slice(None, n))
+        return self._mask_selected_obj(mask)
+
+    @final
+    @Substitution(name="groupby")
+    @Substitution(see_also=_common_see_also)
+    def tail(self, n: int = 5) -> NDFrameT:
+        """
+        Return last n rows of each group.
+
+        Similar to ``.apply(lambda x: x.tail(n))``, but it returns a subset of rows
+        from the original DataFrame with original index and order preserved
+        (``as_index`` flag is ignored).
+
+        Parameters
+        ----------
+        n : int
+            If positive: number of entries to include from end of each group.
+            If negative: number of entries to exclude from start of each group.
+
+        Returns
+        -------
+        Series or DataFrame
+            Subset of original Series or DataFrame as determined by n.
+        %(see_also)s
+        Examples
+        --------
+
+        >>> df = pd.DataFrame(
+        ...     [["a", 1], ["a", 2], ["b", 1], ["b", 2]], columns=["A", "B"]
+        ... )
+        >>> df.groupby("A").tail(1)
+           A  B
+        1  a  2
+        3  b  2
+        >>> df.groupby("A").tail(-1)
+           A  B
+        1  a  2
+        3  b  2
+        """
+        if n:
+            mask = self._make_mask_from_positional_indexer(slice(-n, None))
+        else:
+            mask = self._make_mask_from_positional_indexer([])
+
+        return self._mask_selected_obj(mask)
+
+    @final
+    def _mask_selected_obj(self, mask: npt.NDArray[np.bool_]) -> NDFrameT:
+        """
+        Return _selected_obj with mask applied.
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+            Boolean mask to apply.
+
+        Returns
+        -------
+        Series or DataFrame
+            Filtered _selected_obj.
+        """
+        ids = self._grouper.ids
+        mask = mask & (ids != -1)
+        return self._selected_obj[mask]
+
+    @final
+    def sample(
+        self,
+        n: int | None = None,
+        frac: float | None = None,
+        replace: bool = False,
+        weights: Sequence | Series | None = None,
+        random_state: RandomState | None = None,
+    ):
+        """
+        Return a random sample of items from each group.
+
+        You can use `random_state` for reproducibility.
+
+        Parameters
+        ----------
+        n : int, optional
+            Number of items to return for each group. Cannot be used with
+            `frac` and must be no larger than the smallest group unless
+            `replace` is True. Default is one if `frac` is None.
+        frac : float, optional
+            Fraction of items to return. Cannot be used with `n`.
+        replace : bool, default False
+            Allow or disallow sampling of the same row more than once.
+        weights : list-like, optional
+            Default None results in equal probability weighting.
+            If passed a list-like then values must have the same length as
+            the underlying DataFrame or Series object and will be used as
+            sampling probabilities after normalization within each group.
+            Values must be non-negative with at least one positive element
+            within each group.
+        random_state : int, array-like, BitGenerator, np.random.RandomState, np.random.Generator, optional
+            If int, array-like, or BitGenerator, seed for random number generator.
+            If np.random.RandomState or np.random.Generator, use as given.
+            Default ``None`` results in sampling with the current state of np.random.
+
+            .. versionchanged:: 1.4.0
+
+                np.random.Generator objects now accepted
+
+        Returns
+        -------
+        Series or DataFrame
+            A new object of same type as caller containing items randomly
+            sampled within each group from the caller object.
+
+        See Also
+        --------
+        DataFrame.sample: Generate random samples from a DataFrame object.
+        Series.sample: Generate random samples from a Series object.
+        numpy.random.choice: Generate a random sample from a given 1-D numpy
+            array.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {"a": ["red"] * 2 + ["blue"] * 2 + ["black"] * 2, "b": range(6)}
+        ... )
+        >>> df
+               a  b
+        0    red  0
+        1    red  1
+        2   blue  2
+        3   blue  3
+        4  black  4
+        5  black  5
+
+        Select one row at random for each distinct value in column a. The
+        `random_state` argument can be used to guarantee reproducibility:
+
+        >>> df.groupby("a").sample(n=1, random_state=1)
+               a  b
+        4  black  4
+        2   blue  2
+        1    red  1
+
+        Set `frac` to sample fixed proportions rather than counts:
+
+        >>> df.groupby("a")["b"].sample(frac=0.5, random_state=2)
+        5    5
+        2    2
+        0    0
+        Name: b, dtype: int64
+
+        Control sample probabilities within groups by setting weights:
+
+        >>> df.groupby("a").sample(
+        ...     n=1,
+        ...     weights=[1, 1, 1, 0, 0, 1],
+        ...     random_state=1,
+        ... )
+               a  b
+        5  black  5
+        2   blue  2
+        0    red  0
+        """  # noqa: E501
+        if self._selected_obj.empty:
+            # GH48459 prevent ValueError when object is empty
+            return self._selected_obj
+        size = sample.process_sampling_size(n, frac, replace)
+        if weights is not None:
+            weights_arr = sample.preprocess_weights(self._selected_obj, weights, axis=0)
+
+        random_state = com.random_state(random_state)
+
+        group_iterator = self._grouper.get_iterator(self._selected_obj)
+
+        sampled_indices = []
+        for labels, obj in group_iterator:
+            grp_indices = self.indices[labels]
+            group_size = len(grp_indices)
+            if size is not None:
+                sample_size = size
+            else:
+                assert frac is not None
+                sample_size = round(frac * group_size)
+
+            grp_sample = sample.sample(
+                group_size,
+                size=sample_size,
+                replace=replace,
+                weights=None if weights is None else weights_arr[grp_indices],
+                random_state=random_state,
+            )
+            sampled_indices.append(grp_indices[grp_sample])
+
+        sampled_indices = np.concatenate(sampled_indices)
+        return self._selected_obj.take(sampled_indices, axis=0)
+
+    def _idxmax_idxmin(
+        self,
+        how: Literal["idxmax", "idxmin"],
+        ignore_unobserved: bool = False,
+        skipna: bool = True,
+        numeric_only: bool = False,
+    ) -> NDFrameT:
+        """Compute idxmax/idxmin.
+
+        Parameters
+        ----------
+        how : {'idxmin', 'idxmax'}
+            Whether to compute idxmin or idxmax.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        skipna : bool, default True
+            Exclude NA/null values. If an entire group is NA, the result will be NA.
+        ignore_unobserved : bool, default False
+            When True and an unobserved group is encountered, do not raise. This used
+            for transform where unobserved groups do not play an impact on the result.
+
+        Returns
+        -------
+        Series or DataFrame
+            idxmax or idxmin for the groupby operation.
+        """
+        if not self.observed and any(
+            ping._passed_categorical for ping in self._grouper.groupings
+        ):
+            expected_len = len(self._grouper.result_index)
+            # TODO: Better way to find # of observed groups?
+            group_sizes = self._grouper.size()
+            result_len = group_sizes[group_sizes > 0].shape[0]
+            assert result_len <= expected_len
+            has_unobserved = result_len < expected_len
+
+            raise_err: bool | np.bool_ = not ignore_unobserved and has_unobserved
+            # Only raise an error if there are columns to compute; otherwise we return
+            # an empty DataFrame with an index (possibly including unobserved) but no
+            # columns
+            data = self._obj_with_exclusions
+            if raise_err and isinstance(data, DataFrame):
+                if numeric_only:
+                    data = data._get_numeric_data()
+                raise_err = len(data.columns) > 0
+
+            if raise_err:
+                raise ValueError(
+                    f"Can't get {how} of an empty group due to unobserved categories. "
+                    "Specify observed=True in groupby instead."
+                )
+        elif not skipna and self._obj_with_exclusions.isna().any(axis=None):
+            raise ValueError(
+                f"{type(self).__name__}.{how} with skipna=False encountered an NA "
+                f"value."
+            )
+
+        result = self._agg_general(
+            numeric_only=numeric_only,
+            min_count=1,
+            alias=how,
+            skipna=skipna,
+        )
+        return result
+
+    def _wrap_idxmax_idxmin(self, res: NDFrameT) -> NDFrameT:
+        index = self.obj.index
+        if res.size == 0:
+            result = res.astype(index.dtype)
+        else:
+            if isinstance(index, MultiIndex):
+                index = index.to_flat_index()
+            values = res._values
+            assert isinstance(values, np.ndarray)
+            na_value = na_value_for_dtype(index.dtype, compat=False)
+            if isinstance(res, Series):
+                # mypy: expression has type "Series", variable has type "NDFrameT"
+                result = res._constructor(  # type: ignore[assignment]
+                    index.array.take(values, allow_fill=True, fill_value=na_value),
+                    index=res.index,
+                    name=res.name,
+                )
+            else:
+                data = {}
+                for k, column_values in enumerate(values.T):
+                    data[k] = index.array.take(
+                        column_values, allow_fill=True, fill_value=na_value
+                    )
+                result = self.obj._constructor(data, index=res.index)
+                result.columns = res.columns
+        return result
+
+
+@doc(GroupBy)
+def get_groupby(
+    obj: NDFrame,
+    by: _KeysArgType | None = None,
+    grouper: ops.BaseGrouper | None = None,
+    group_keys: bool = True,
+) -> GroupBy:
+    klass: type[GroupBy]
+    if isinstance(obj, Series):
+        from pandas.core.groupby.generic import SeriesGroupBy
+
+        klass = SeriesGroupBy
+    elif isinstance(obj, DataFrame):
+        from pandas.core.groupby.generic import DataFrameGroupBy
+
+        klass = DataFrameGroupBy
+    else:  # pragma: no cover
+        raise TypeError(f"invalid type: {obj}")
+
+    return klass(
+        obj=obj,
+        keys=by,
+        grouper=grouper,
+        group_keys=group_keys,
+    )
+
+
+def _insert_quantile_level(idx: Index, qs: npt.NDArray[np.float64]) -> MultiIndex:
+    """
+    Insert the sequence 'qs' of quantiles as the inner-most level of a MultiIndex.
+
+    The quantile level in the MultiIndex is a repeated copy of 'qs'.
+
+    Parameters
+    ----------
+    idx : Index
+    qs : np.ndarray[float64]
+
+    Returns
+    -------
+    MultiIndex
+    """
+    nqs = len(qs)
+    lev_codes, lev = Index(qs).factorize()
+    lev_codes = coerce_indexer_dtype(lev_codes, lev)
+
+    if idx._is_multi:
+        idx = cast(MultiIndex, idx)
+        levels = list(idx.levels) + [lev]
+        codes = [np.repeat(x, nqs) for x in idx.codes] + [np.tile(lev_codes, len(idx))]
+        mi = MultiIndex(levels=levels, codes=codes, names=idx.names + [None])
+    else:
+        nidx = len(idx)
+        idx_codes = coerce_indexer_dtype(np.arange(nidx), idx)
+        levels = [idx, lev]
+        codes = [np.repeat(idx_codes, nqs), np.tile(lev_codes, nidx)]
+        mi = MultiIndex(levels=levels, codes=codes, names=[idx.name, None])
+
+    return mi
diff --git a/pandas/pandas/core/groupby/grouper.py b/pandas/pandas/core/groupby/grouper.py
new file mode 100644
index 0000000000000..c9d874fc08dbe
--- /dev/null
+++ b/pandas/pandas/core/groupby/grouper.py
@@ -0,0 +1,936 @@
+"""
+Provide user facing operators for doing the split part of the
+split-apply-combine paradigm.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    final,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import OutOfBoundsDatetime
+from pandas.errors import InvalidIndexError
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.common import (
+    is_list_like,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+from pandas.core import algorithms
+from pandas.core.arrays import (
+    Categorical,
+    ExtensionArray,
+)
+import pandas.core.common as com
+from pandas.core.frame import DataFrame
+from pandas.core.groupby import ops
+from pandas.core.groupby.categorical import recode_for_groupby
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    default_index,
+)
+from pandas.core.series import Series
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterator,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        NDFrameT,
+        npt,
+    )
+
+    from pandas.core.generic import NDFrame
+
+
+class Grouper:
+    """
+    A Grouper allows the user to specify a groupby instruction for an object.
+
+    This specification will select a column via the key parameter, or if the
+    level parameter is given, a level of the index of the target
+    object.
+
+    If ``level`` is passed as a keyword to both `Grouper` and
+    `groupby`, the values passed to `Grouper` take precedence.
+
+    Parameters
+    ----------
+    *args
+        Currently unused, reserved for future use.
+    **kwargs
+        Dictionary of the keyword arguments to pass to Grouper.
+
+    Attributes
+    ----------
+    key : str, defaults to None
+        Groupby key, which selects the grouping column of the target.
+    level : name/number, defaults to None
+        The level for the target index.
+    freq : str / frequency object, defaults to None
+        This will groupby the specified frequency if the target selection
+        (via key or level) is a datetime-like object. For full specification
+        of available frequencies, please see :ref:`here<timeseries.offset_aliases>`.
+    sort : bool, default to False
+        Whether to sort the resulting labels.
+    closed : {'left' or 'right'}
+        Closed end of interval. Only when `freq` parameter is passed.
+    label : {'left' or 'right'}
+        Interval boundary to use for labeling.
+        Only when `freq` parameter is passed.
+    convention : {'start', 'end', 'e', 's'}
+        If grouper is PeriodIndex and `freq` parameter is passed.
+
+    origin : Timestamp or str, default 'start_day'
+        The timestamp on which to adjust the grouping. The timezone of origin must
+        match the timezone of the index.
+        If string, must be one of the following:
+
+        - 'epoch': `origin` is 1970-01-01
+        - 'start': `origin` is the first value of the timeseries
+        - 'start_day': `origin` is the first day at midnight of the timeseries
+
+        - 'end': `origin` is the last value of the timeseries
+        - 'end_day': `origin` is the ceiling midnight of the last day
+
+        .. versionadded:: 1.3.0
+
+    offset : Timedelta or str, default is None
+        An offset timedelta added to the origin.
+
+    dropna : bool, default True
+        If True, and if group keys contain NA values, NA values together with
+        row/column will be dropped. If False, NA values will also be treated as
+        the key in groups.
+
+    Returns
+    -------
+    Grouper or pandas.api.typing.TimeGrouper
+        A TimeGrouper is returned if ``freq`` is not ``None``. Otherwise, a Grouper
+        is returned.
+
+    See Also
+    --------
+    Series.groupby : Apply a function groupby to a Series.
+    DataFrame.groupby : Apply a function groupby.
+
+    Examples
+    --------
+    ``df.groupby(pd.Grouper(key="Animal"))`` is equivalent to ``df.groupby('Animal')``
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "Animal": ["Falcon", "Parrot", "Falcon", "Falcon", "Parrot"],
+    ...         "Speed": [100, 5, 200, 300, 15],
+    ...     }
+    ... )
+    >>> df
+       Animal  Speed
+    0  Falcon    100
+    1  Parrot      5
+    2  Falcon    200
+    3  Falcon    300
+    4  Parrot     15
+    >>> df.groupby(pd.Grouper(key="Animal")).mean()
+            Speed
+    Animal
+    Falcon  200.0
+    Parrot   10.0
+
+    Specify a resample operation on the column 'Publish date'
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "Publish date": [
+    ...             pd.Timestamp("2000-01-02"),
+    ...             pd.Timestamp("2000-01-02"),
+    ...             pd.Timestamp("2000-01-09"),
+    ...             pd.Timestamp("2000-01-16"),
+    ...         ],
+    ...         "ID": [0, 1, 2, 3],
+    ...         "Price": [10, 20, 30, 40],
+    ...     }
+    ... )
+    >>> df
+      Publish date  ID  Price
+    0   2000-01-02   0     10
+    1   2000-01-02   1     20
+    2   2000-01-09   2     30
+    3   2000-01-16   3     40
+    >>> df.groupby(pd.Grouper(key="Publish date", freq="1W")).mean()
+                   ID  Price
+    Publish date
+    2000-01-02    0.5   15.0
+    2000-01-09    2.0   30.0
+    2000-01-16    3.0   40.0
+
+    If you want to adjust the start of the bins based on a fixed timestamp:
+
+    >>> start, end = "2000-10-01 23:30:00", "2000-10-02 00:30:00"
+    >>> rng = pd.date_range(start, end, freq="7min")
+    >>> ts = pd.Series(np.arange(len(rng)) * 3, index=rng)
+    >>> ts
+    2000-10-01 23:30:00     0
+    2000-10-01 23:37:00     3
+    2000-10-01 23:44:00     6
+    2000-10-01 23:51:00     9
+    2000-10-01 23:58:00    12
+    2000-10-02 00:05:00    15
+    2000-10-02 00:12:00    18
+    2000-10-02 00:19:00    21
+    2000-10-02 00:26:00    24
+    Freq: 7min, dtype: int64
+
+    >>> ts.groupby(pd.Grouper(freq="17min")).sum()
+    2000-10-01 23:14:00     0
+    2000-10-01 23:31:00     9
+    2000-10-01 23:48:00    21
+    2000-10-02 00:05:00    54
+    2000-10-02 00:22:00    24
+    Freq: 17min, dtype: int64
+
+    >>> ts.groupby(pd.Grouper(freq="17min", origin="epoch")).sum()
+    2000-10-01 23:18:00     0
+    2000-10-01 23:35:00    18
+    2000-10-01 23:52:00    27
+    2000-10-02 00:09:00    39
+    2000-10-02 00:26:00    24
+    Freq: 17min, dtype: int64
+
+    >>> ts.groupby(pd.Grouper(freq="17min", origin="2000-01-01")).sum()
+    2000-10-01 23:24:00     3
+    2000-10-01 23:41:00    15
+    2000-10-01 23:58:00    45
+    2000-10-02 00:15:00    45
+    Freq: 17min, dtype: int64
+
+    If you want to adjust the start of the bins with an `offset` Timedelta, the two
+    following lines are equivalent:
+
+    >>> ts.groupby(pd.Grouper(freq="17min", origin="start")).sum()
+    2000-10-01 23:30:00     9
+    2000-10-01 23:47:00    21
+    2000-10-02 00:04:00    54
+    2000-10-02 00:21:00    24
+    Freq: 17min, dtype: int64
+
+    >>> ts.groupby(pd.Grouper(freq="17min", offset="23h30min")).sum()
+    2000-10-01 23:30:00     9
+    2000-10-01 23:47:00    21
+    2000-10-02 00:04:00    54
+    2000-10-02 00:21:00    24
+    Freq: 17min, dtype: int64
+
+    To replace the use of the deprecated `base` argument, you can now use `offset`,
+    in this example it is equivalent to have `base=2`:
+
+    >>> ts.groupby(pd.Grouper(freq="17min", offset="2min")).sum()
+    2000-10-01 23:16:00     0
+    2000-10-01 23:33:00     9
+    2000-10-01 23:50:00    36
+    2000-10-02 00:07:00    39
+    2000-10-02 00:24:00    24
+    Freq: 17min, dtype: int64
+    """
+
+    sort: bool
+    dropna: bool
+    _grouper: Index | None
+
+    _attributes: tuple[str, ...] = ("key", "level", "freq", "sort", "dropna")
+
+    def __new__(cls, *args, **kwargs):
+        if kwargs.get("freq") is not None:
+            from pandas.core.resample import TimeGrouper
+
+            cls = TimeGrouper
+        return super().__new__(cls)
+
+    def __init__(
+        self,
+        key=None,
+        level=None,
+        freq=None,
+        sort: bool = False,
+        dropna: bool = True,
+    ) -> None:
+        self.key = key
+        self.level = level
+        self.freq = freq
+        self.sort = sort
+        self.dropna = dropna
+
+        self._indexer_deprecated: npt.NDArray[np.intp] | None = None
+        self.binner = None
+        self._grouper = None
+        self._indexer: npt.NDArray[np.intp] | None = None
+
+    def _get_grouper(
+        self, obj: NDFrameT, validate: bool = True
+    ) -> tuple[ops.BaseGrouper, NDFrameT]:
+        """
+        Parameters
+        ----------
+        obj : Series or DataFrame
+        validate : bool, default True
+            if True, validate the grouper
+
+        Returns
+        -------
+        a tuple of grouper, obj (possibly sorted)
+        """
+        obj, _, _ = self._set_grouper(obj)
+        grouper, _, obj = get_grouper(
+            obj,
+            [self.key],
+            level=self.level,
+            sort=self.sort,
+            validate=validate,
+            dropna=self.dropna,
+        )
+
+        return grouper, obj
+
+    def _set_grouper(
+        self, obj: NDFrameT, sort: bool = False, *, gpr_index: Index | None = None
+    ) -> tuple[NDFrameT, Index, npt.NDArray[np.intp] | None]:
+        """
+        given an object and the specifications, setup the internal grouper
+        for this particular specification
+
+        Parameters
+        ----------
+        obj : Series or DataFrame
+        sort : bool, default False
+            whether the resulting grouper should be sorted
+        gpr_index : Index or None, default None
+
+        Returns
+        -------
+        NDFrame
+        Index
+        np.ndarray[np.intp] | None
+        """
+        assert obj is not None
+
+        if self.key is not None and self.level is not None:
+            raise ValueError("The Grouper cannot specify both a key and a level!")
+
+        # Keep self._grouper value before overriding
+        if self._grouper is None:
+            # TODO: What are we assuming about subsequent calls?
+            self._grouper = gpr_index
+            self._indexer = self._indexer_deprecated
+
+        # the key must be a valid info item
+        if self.key is not None:
+            key = self.key
+            # The 'on' is already defined
+            if getattr(gpr_index, "name", None) == key and isinstance(obj, Series):
+                # Sometimes self._grouper will have been resorted while
+                # obj has not. In this case there is a mismatch when we
+                # call self._grouper.take(obj.index) so we need to undo the sorting
+                # before we call _grouper.take.
+                assert self._grouper is not None
+                if self._indexer is not None:
+                    reverse_indexer = self._indexer.argsort()
+                    unsorted_ax = self._grouper.take(reverse_indexer)
+                    ax = unsorted_ax.take(obj.index)
+                else:
+                    ax = self._grouper.take(obj.index)
+            else:
+                if key not in obj._info_axis:
+                    raise KeyError(f"The grouper name {key} is not found")
+                ax = Index(obj[key], name=key)
+
+        else:
+            ax = obj.index
+            if self.level is not None:
+                level = self.level
+
+                # if a level is given it must be a mi level or
+                # equivalent to the axis name
+                if isinstance(ax, MultiIndex):
+                    level = ax._get_level_number(level)
+                    ax = Index(ax._get_level_values(level), name=ax.names[level])
+
+                else:
+                    if level not in (0, ax.name):
+                        raise ValueError(f"The level {level} is not valid")
+
+        # possibly sort
+        indexer: npt.NDArray[np.intp] | None = None
+        if (self.sort or sort) and not ax.is_monotonic_increasing:
+            # use stable sort to support first, last, nth
+            # TODO: why does putting na_position="first" fix datetimelike cases?
+            indexer = self._indexer_deprecated = ax.array.argsort(
+                kind="mergesort", na_position="first"
+            )
+            ax = ax.take(indexer)
+            obj = obj.take(indexer, axis=0)
+
+        return obj, ax, indexer
+
+    @final
+    def __repr__(self) -> str:
+        attrs_list = (
+            f"{attr_name}={getattr(self, attr_name)!r}"
+            for attr_name in self._attributes
+            if getattr(self, attr_name) is not None
+        )
+        attrs = ", ".join(attrs_list)
+        cls_name = type(self).__name__
+        return f"{cls_name}({attrs})"
+
+
+@final
+class Grouping:
+    """
+    Holds the grouping information for a single key
+
+    Parameters
+    ----------
+    index : Index
+    grouper :
+    obj : DataFrame or Series
+    name : Label
+    level :
+    observed : bool, default False
+        If we are a Categorical, use the observed values
+    in_axis : if the Grouping is a column in self.obj and hence among
+        Groupby.exclusions list
+    dropna : bool, default True
+        Whether to drop NA groups.
+    uniques : Array-like, optional
+        When specified, will be used for unique values. Enables including empty groups
+        in the result for a BinGrouper. Must not contain duplicates.
+
+    Attributes
+    -------
+    indices : dict
+        Mapping of {group -> index_list}
+    codes : ndarray
+        Group codes
+    group_index : Index or None
+        unique groups
+    groups : dict
+        Mapping of {group -> label_list}
+    """
+
+    _codes: npt.NDArray[np.signedinteger] | None = None
+    _orig_cats: Index | None
+    _index: Index
+
+    def __init__(
+        self,
+        index: Index,
+        grouper=None,
+        obj: NDFrame | None = None,
+        level=None,
+        sort: bool = True,
+        observed: bool = False,
+        in_axis: bool = False,
+        dropna: bool = True,
+        uniques: ArrayLike | None = None,
+    ) -> None:
+        self.level = level
+        self._orig_grouper = grouper
+        grouping_vector = _convert_grouper(index, grouper)
+        self._orig_cats = None
+        self._index = index
+        self._sort = sort
+        self.obj = obj
+        self._observed = observed
+        self.in_axis = in_axis
+        self._dropna = dropna
+        self._uniques = uniques
+
+        # we have a single grouper which may be a myriad of things,
+        # some of which are dependent on the passing in level
+
+        ilevel = self._ilevel
+        if ilevel is not None:
+            # In extant tests, the new self.grouping_vector matches
+            #  `index.get_level_values(ilevel)` whenever
+            #  mapper is None and isinstance(index, MultiIndex)
+            if isinstance(index, MultiIndex):
+                index_level = index.get_level_values(ilevel)
+            else:
+                index_level = index
+
+            if grouping_vector is None:
+                grouping_vector = index_level
+            else:
+                mapper = grouping_vector
+                grouping_vector = index_level.map(mapper)
+
+        # a passed Grouper like, directly get the grouper in the same way
+        # as single grouper groupby, use the group_info to get codes
+        elif isinstance(grouping_vector, Grouper):
+            # get the new grouper; we already have disambiguated
+            # what key/level refer to exactly, don't need to
+            # check again as we have by this point converted these
+            # to an actual value (rather than a pd.Grouper)
+            assert self.obj is not None  # for mypy
+            newgrouper, newobj = grouping_vector._get_grouper(self.obj, validate=False)
+            self.obj = newobj
+
+            if isinstance(newgrouper, ops.BinGrouper):
+                # TODO: can we unwrap this and get a tighter typing
+                #  for self.grouping_vector?
+                grouping_vector = newgrouper
+            else:
+                # ops.BaseGrouper
+                # TODO: 2023-02-03 no test cases with len(newgrouper.groupings) > 1.
+                #  If that were to occur, would we be throwing out information?
+                # error: Cannot determine type of "grouping_vector"  [has-type]
+                ng = newgrouper.groupings[0].grouping_vector  # type: ignore[has-type]
+                # use Index instead of ndarray so we can recover the name
+                grouping_vector = Index(ng, name=newgrouper.result_index.name)
+
+        elif not isinstance(
+            grouping_vector, (Series, Index, ExtensionArray, np.ndarray)
+        ):
+            # no level passed
+            if getattr(grouping_vector, "ndim", 1) != 1:
+                t = str(type(grouping_vector))
+                raise ValueError(f"Grouper for '{t}' not 1-dimensional")
+
+            grouping_vector = index.map(grouping_vector)
+
+            if not (
+                hasattr(grouping_vector, "__len__")
+                and len(grouping_vector) == len(index)
+            ):
+                grper = pprint_thing(grouping_vector)
+                errmsg = (
+                    f"Grouper result violates len(labels) == len(data)\nresult: {grper}"
+                )
+                raise AssertionError(errmsg)
+
+        if isinstance(grouping_vector, np.ndarray):
+            if grouping_vector.dtype.kind in "mM":
+                # if we have a date/time-like grouper, make sure that we have
+                # Timestamps like
+                # TODO 2022-10-08 we only have one test that gets here and
+                #  values are already in nanoseconds in that case.
+                grouping_vector = Series(grouping_vector).to_numpy()
+        elif isinstance(getattr(grouping_vector, "dtype", None), CategoricalDtype):
+            # a passed Categorical
+            self._orig_cats = grouping_vector.categories
+            grouping_vector = recode_for_groupby(grouping_vector, sort, observed)
+
+        self.grouping_vector = grouping_vector
+
+    def __repr__(self) -> str:
+        return f"Grouping({self.name})"
+
+    def __iter__(self) -> Iterator:
+        return iter(self.indices)
+
+    @cache_readonly
+    def _passed_categorical(self) -> bool:
+        dtype = getattr(self.grouping_vector, "dtype", None)
+        return isinstance(dtype, CategoricalDtype)
+
+    @cache_readonly
+    def name(self) -> Hashable:
+        ilevel = self._ilevel
+        if ilevel is not None:
+            return self._index.names[ilevel]
+
+        if isinstance(self._orig_grouper, (Index, Series)):
+            return self._orig_grouper.name
+
+        elif isinstance(self.grouping_vector, ops.BaseGrouper):
+            return self.grouping_vector.result_index.name
+
+        elif isinstance(self.grouping_vector, Index):
+            return self.grouping_vector.name
+
+        # otherwise we have ndarray or ExtensionArray -> no name
+        return None
+
+    @cache_readonly
+    def _ilevel(self) -> int | None:
+        """
+        If necessary, converted index level name to index level position.
+        """
+        level = self.level
+        if level is None:
+            return None
+        if not isinstance(level, int):
+            index = self._index
+            if level not in index.names:
+                raise AssertionError(f"Level {level} not in index")
+            return index.names.index(level)
+        return level
+
+    @property
+    def ngroups(self) -> int:
+        return len(self.uniques)
+
+    @cache_readonly
+    def indices(self) -> dict[Hashable, npt.NDArray[np.intp]]:
+        # we have a list of groupers
+        if isinstance(self.grouping_vector, ops.BaseGrouper):
+            return self.grouping_vector.indices
+
+        values = Categorical(self.grouping_vector)
+        return values._reverse_indexer()
+
+    @property
+    def codes(self) -> npt.NDArray[np.signedinteger]:
+        return self._codes_and_uniques[0]
+
+    @property
+    def uniques(self) -> ArrayLike:
+        return self._codes_and_uniques[1]
+
+    @cache_readonly
+    def _codes_and_uniques(self) -> tuple[npt.NDArray[np.signedinteger], ArrayLike]:
+        uniques: ArrayLike
+        if self._passed_categorical:
+            # we make a CategoricalIndex out of the cat grouper
+            # preserving the categories / ordered attributes;
+            # doesn't (yet - GH#46909) handle dropna=False
+            cat = self.grouping_vector
+            categories = cat.categories
+
+            if self._observed:
+                ucodes = algorithms.unique1d(cat.codes)
+                ucodes = ucodes[ucodes != -1]
+                if self._sort:
+                    ucodes = np.sort(ucodes)
+            else:
+                ucodes = np.arange(len(categories))
+
+            has_dropped_na = False
+            if not self._dropna:
+                na_mask = cat.isna()
+                if np.any(na_mask):
+                    has_dropped_na = True
+                    if self._sort:
+                        # NA goes at the end, gets `largest non-NA code + 1`
+                        na_code = len(categories)
+                    else:
+                        # Insert NA in result based on first appearance, need
+                        # the number of unique codes prior
+                        na_idx = na_mask.argmax()
+                        na_code = algorithms.nunique_ints(cat.codes[:na_idx])
+                    ucodes = np.insert(ucodes, na_code, -1)
+
+            uniques = Categorical.from_codes(
+                codes=ucodes, categories=categories, ordered=cat.ordered, validate=False
+            )
+            codes = cat.codes
+
+            if has_dropped_na:
+                if not self._sort:
+                    # NA code is based on first appearance, increment higher codes
+                    codes = np.where(codes >= na_code, codes + 1, codes)
+                codes = np.where(na_mask, na_code, codes)
+
+            return codes, uniques
+
+        elif isinstance(self.grouping_vector, ops.BaseGrouper):
+            # we have a list of groupers
+            codes = self.grouping_vector.codes_info
+            uniques = self.grouping_vector.result_index._values
+        elif self._uniques is not None:
+            # GH#50486 Code grouping_vector using _uniques; allows
+            # including uniques that are not present in grouping_vector.
+            cat = Categorical(self.grouping_vector, categories=self._uniques)
+            codes = cat.codes
+            uniques = self._uniques
+        else:
+            # GH35667, replace dropna=False with use_na_sentinel=False
+            # error: Incompatible types in assignment (expression has type "Union[
+            # ndarray[Any, Any], Index]", variable has type "Categorical")
+            codes, uniques = algorithms.factorize(  # type: ignore[assignment]
+                self.grouping_vector, sort=self._sort, use_na_sentinel=self._dropna
+            )
+        return codes, uniques
+
+    @cache_readonly
+    def groups(self) -> dict[Hashable, Index]:
+        codes, uniques = self._codes_and_uniques
+        uniques = Index._with_infer(uniques, name=self.name)
+        cats = Categorical.from_codes(codes, uniques, validate=False)
+        return self._index.groupby(cats)
+
+    @property
+    def observed_grouping(self) -> Grouping:
+        if self._observed:
+            return self
+
+        return self._observed_grouping
+
+    @cache_readonly
+    def _observed_grouping(self) -> Grouping:
+        grouping = Grouping(
+            self._index,
+            self._orig_grouper,
+            obj=self.obj,
+            level=self.level,
+            sort=self._sort,
+            observed=True,
+            in_axis=self.in_axis,
+            dropna=self._dropna,
+            uniques=self._uniques,
+        )
+        return grouping
+
+
+def get_grouper(
+    obj: NDFrameT,
+    key=None,
+    level=None,
+    sort: bool = True,
+    observed: bool = False,
+    validate: bool = True,
+    dropna: bool = True,
+) -> tuple[ops.BaseGrouper, frozenset[Hashable], NDFrameT]:
+    """
+    Create and return a BaseGrouper, which is an internal
+    mapping of how to create the grouper indexers.
+    This may be composed of multiple Grouping objects, indicating
+    multiple groupers
+
+    Groupers are ultimately index mappings. They can originate as:
+    index mappings, keys to columns, functions, or Groupers
+
+    Groupers enable local references to level,sort, while
+    the passed in level, and sort are 'global'.
+
+    This routine tries to figure out what the passing in references
+    are and then creates a Grouping for each one, combined into
+    a BaseGrouper.
+
+    If observed & we have a categorical grouper, only show the observed
+    values.
+
+    If validate, then check for key/level overlaps.
+
+    """
+    group_axis = obj.index
+
+    # validate that the passed single level is compatible with the passed
+    # index of the object
+    if level is not None:
+        # TODO: These if-block and else-block are almost same.
+        # MultiIndex instance check is removable, but it seems that there are
+        # some processes only for non-MultiIndex in else-block,
+        # eg. `obj.index.name != level`. We have to consider carefully whether
+        # these are applicable for MultiIndex. Even if these are applicable,
+        # we need to check if it makes no side effect to subsequent processes
+        # on the outside of this condition.
+        # (GH 17621)
+        if isinstance(group_axis, MultiIndex):
+            if is_list_like(level) and len(level) == 1:
+                level = level[0]
+
+            if key is None and is_scalar(level):
+                # Get the level values from group_axis
+                key = group_axis.get_level_values(level)
+                level = None
+
+        else:
+            # allow level to be a length-one list-like object
+            # (e.g., level=[0])
+            # GH 13901
+            if is_list_like(level):
+                nlevels = len(level)
+                if nlevels == 1:
+                    level = level[0]
+                elif nlevels == 0:
+                    raise ValueError("No group keys passed!")
+                else:
+                    raise ValueError("multiple levels only valid with MultiIndex")
+
+            if isinstance(level, str):
+                if obj.index.name != level:
+                    raise ValueError(f"level name {level} is not the name of the index")
+            elif level > 0 or level < -1:
+                raise ValueError("level > 0 or level < -1 only valid with MultiIndex")
+
+            # NOTE: `group_axis` and `group_axis.get_level_values(level)`
+            # are same in this section.
+            level = None
+            key = group_axis
+
+    # a passed-in Grouper, directly convert
+    if isinstance(key, Grouper):
+        grouper, obj = key._get_grouper(obj, validate=False)
+        if key.key is None:
+            return grouper, frozenset(), obj
+        else:
+            return grouper, frozenset({key.key}), obj
+
+    # already have a BaseGrouper, just return it
+    elif isinstance(key, ops.BaseGrouper):
+        return key, frozenset(), obj
+
+    if not isinstance(key, list):
+        keys = [key]
+        match_axis_length = False
+    else:
+        keys = key
+        match_axis_length = len(keys) == len(group_axis)
+
+    # what are we after, exactly?
+    any_callable = any(callable(g) or isinstance(g, dict) for g in keys)
+    any_groupers = any(isinstance(g, (Grouper, Grouping)) for g in keys)
+    any_arraylike = any(
+        isinstance(g, (list, tuple, Series, Index, np.ndarray)) for g in keys
+    )
+
+    # is this an index replacement?
+    if (
+        not any_callable
+        and not any_arraylike
+        and not any_groupers
+        and match_axis_length
+        and level is None
+    ):
+        if isinstance(obj, DataFrame):
+            all_in_columns_index = all(
+                g in obj.columns or g in obj.index.names for g in keys
+            )
+        else:
+            assert isinstance(obj, Series)
+            all_in_columns_index = all(g in obj.index.names for g in keys)
+
+        if not all_in_columns_index:
+            keys = [com.asarray_tuplesafe(keys)]
+
+    if isinstance(level, (tuple, list)):
+        if key is None:
+            keys = [None] * len(level)
+        levels = level
+    else:
+        levels = [level] * len(keys)
+
+    groupings: list[Grouping] = []
+    exclusions: set[Hashable] = set()
+
+    # if the actual grouper should be obj[key]
+    def is_in_axis(key) -> bool:
+        if not _is_label_like(key):
+            if obj.ndim == 1:
+                return False
+
+            # items -> .columns for DataFrame, .index for Series
+            items = obj.axes[-1]
+            try:
+                items.get_loc(key)
+            except (KeyError, TypeError, InvalidIndexError):
+                # TypeError shows up here if we pass e.g. an Index
+                return False
+
+        return True
+
+    # if the grouper is obj[name]
+    def is_in_obj(gpr) -> bool:
+        if not hasattr(gpr, "name"):
+            return False
+        # We check the references to determine if the
+        # series is part of the object
+        try:
+            obj_gpr_column = obj[gpr.name]
+        except (KeyError, IndexError, InvalidIndexError, OutOfBoundsDatetime):
+            return False
+        if isinstance(gpr, Series) and isinstance(obj_gpr_column, Series):
+            return gpr._mgr.references_same_values(obj_gpr_column._mgr, 0)
+        return False
+
+    for gpr, level in zip(keys, levels):
+        if is_in_obj(gpr):  # df.groupby(df['name'])
+            in_axis = True
+            exclusions.add(gpr.name)
+
+        elif is_in_axis(gpr):  # df.groupby('name')
+            if obj.ndim != 1 and gpr in obj:
+                if validate:
+                    obj._check_label_or_level_ambiguity(gpr, axis=0)
+                in_axis, name, gpr = True, gpr, obj[gpr]
+                if gpr.ndim != 1:
+                    # non-unique columns; raise here to get the name in the
+                    # exception message
+                    raise ValueError(f"Grouper for '{name}' not 1-dimensional")
+                exclusions.add(name)
+            elif obj._is_level_reference(gpr, axis=0):
+                in_axis, level, gpr = False, gpr, None
+            else:
+                raise KeyError(gpr)
+        elif isinstance(gpr, Grouper) and gpr.key is not None:
+            # Add key to exclusions
+            exclusions.add(gpr.key)
+            in_axis = True
+        else:
+            in_axis = False
+
+        # create the Grouping
+        # allow us to passing the actual Grouping as the gpr
+        ping = (
+            Grouping(
+                group_axis,
+                gpr,
+                obj=obj,
+                level=level,
+                sort=sort,
+                observed=observed,
+                in_axis=in_axis,
+                dropna=dropna,
+            )
+            if not isinstance(gpr, Grouping)
+            else gpr
+        )
+
+        groupings.append(ping)
+
+    if len(groupings) == 0 and len(obj):
+        raise ValueError("No group keys passed!")
+    if len(groupings) == 0:
+        groupings.append(Grouping(default_index(0), np.array([], dtype=np.intp)))
+
+    # create the internals grouper
+    grouper = ops.BaseGrouper(group_axis, groupings, sort=sort, dropna=dropna)
+    return grouper, frozenset(exclusions), obj
+
+
+def _is_label_like(val) -> bool:
+    return isinstance(val, (str, tuple)) or (val is not None and is_scalar(val))
+
+
+def _convert_grouper(axis: Index, grouper):
+    if isinstance(grouper, dict):
+        return grouper.get
+    elif isinstance(grouper, Series):
+        if grouper.index.equals(axis):
+            return grouper._values
+        else:
+            return grouper.reindex(axis)._values
+    elif isinstance(grouper, MultiIndex):
+        return grouper._values
+    elif isinstance(grouper, (list, tuple, Index, Categorical, np.ndarray)):
+        if len(grouper) != len(axis):
+            raise ValueError("Grouper and axis must be same length")
+
+        if isinstance(grouper, (list, tuple)):
+            grouper = com.asarray_tuplesafe(grouper)
+        return grouper
+    else:
+        return grouper
diff --git a/pandas/pandas/core/groupby/indexing.py b/pandas/pandas/core/groupby/indexing.py
new file mode 100644
index 0000000000000..c658f625d5ea9
--- /dev/null
+++ b/pandas/pandas/core/groupby/indexing.py
@@ -0,0 +1,304 @@
+from __future__ import annotations
+
+from collections.abc import Iterable
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+
+import numpy as np
+
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+)
+
+from pandas.core.dtypes.common import (
+    is_integer,
+    is_list_like,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import PositionalIndexer
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+    from pandas.core.groupby import groupby
+
+
+class GroupByIndexingMixin:
+    """
+    Mixin for adding ._positional_selector to GroupBy.
+    """
+
+    @cache_readonly
+    def _positional_selector(self) -> GroupByPositionalSelector:
+        """
+        Return positional selection for each group.
+
+        ``groupby._positional_selector[i:j]`` is similar to
+        ``groupby.apply(lambda x: x.iloc[i:j])``
+        but much faster and preserves the original index and order.
+
+        ``_positional_selector[]`` is compatible with and extends :meth:`~GroupBy.head`
+        and :meth:`~GroupBy.tail`. For example:
+
+        - ``head(5)``
+        - ``_positional_selector[5:-5]``
+        - ``tail(5)``
+
+        together return all the rows.
+
+        Allowed inputs for the index are:
+
+        - An integer valued iterable, e.g. ``range(2, 4)``.
+        - A comma separated list of integers and slices, e.g. ``5``, ``2, 4``, ``2:4``.
+
+        The output format is the same as :meth:`~GroupBy.head` and
+        :meth:`~GroupBy.tail`, namely
+        a subset of the ``DataFrame`` or ``Series`` with the index and order preserved.
+
+        Returns
+        -------
+        Series
+            The filtered subset of the original Series.
+        DataFrame
+            The filtered subset of the original DataFrame.
+
+        See Also
+        --------
+        DataFrame.iloc : Purely integer-location based indexing for selection by
+            position.
+        GroupBy.head : Return first n rows of each group.
+        GroupBy.tail : Return last n rows of each group.
+        GroupBy.nth : Take the nth row from each group if n is an int, or a
+            subset of rows, if n is a list of ints.
+
+        Notes
+        -----
+        - The slice step cannot be negative.
+        - If the index specification results in overlaps, the item is not duplicated.
+        - If the index specification changes the order of items, then
+          they are returned in their original order.
+          By contrast, ``DataFrame.iloc`` can change the row order.
+        - ``groupby()`` parameters such as as_index and dropna are ignored.
+
+        The differences between ``_positional_selector[]`` and :meth:`~GroupBy.nth`
+        with ``as_index=False`` are:
+
+        - Input to ``_positional_selector`` can include
+          one or more slices whereas ``nth``
+          just handles an integer or a list of integers.
+        - ``_positional_selector`` can  accept a slice relative to the
+          last row of each group.
+        - ``_positional_selector`` does not have an equivalent to the
+          ``nth()`` ``dropna`` parameter.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [["a", 1], ["a", 2], ["a", 3], ["b", 4], ["b", 5]], columns=["A", "B"]
+        ... )
+        >>> df.groupby("A")._positional_selector[1:2]
+           A  B
+        1  a  2
+        4  b  5
+
+        >>> df.groupby("A")._positional_selector[1, -1]
+           A  B
+        1  a  2
+        2  a  3
+        4  b  5
+        """
+        if TYPE_CHECKING:
+            groupby_self = cast(groupby.GroupBy, self)
+        else:
+            groupby_self = self
+
+        return GroupByPositionalSelector(groupby_self)
+
+    def _make_mask_from_positional_indexer(
+        self,
+        arg: PositionalIndexer | tuple,
+    ) -> np.ndarray:
+        if is_list_like(arg):
+            if all(is_integer(i) for i in cast(Iterable, arg)):
+                mask = self._make_mask_from_list(cast(Iterable[int], arg))
+            else:
+                mask = self._make_mask_from_tuple(cast(tuple, arg))
+
+        elif isinstance(arg, slice):
+            mask = self._make_mask_from_slice(arg)
+        elif is_integer(arg):
+            mask = self._make_mask_from_int(cast(int, arg))
+        else:
+            raise TypeError(
+                f"Invalid index {type(arg)}. "
+                "Must be integer, list-like, slice or a tuple of "
+                "integers and slices"
+            )
+
+        if isinstance(mask, bool):
+            if mask:
+                mask = self._ascending_count >= 0
+            else:
+                mask = self._ascending_count < 0
+
+        return cast(np.ndarray, mask)
+
+    def _make_mask_from_int(self, arg: int) -> np.ndarray:
+        if arg >= 0:
+            return self._ascending_count == arg
+        else:
+            return self._descending_count == (-arg - 1)
+
+    def _make_mask_from_list(self, args: Iterable[int]) -> bool | np.ndarray:
+        positive = [arg for arg in args if arg >= 0]
+        negative = [-arg - 1 for arg in args if arg < 0]
+
+        mask: bool | np.ndarray = False
+
+        if positive:
+            mask |= np.isin(self._ascending_count, positive)
+
+        if negative:
+            mask |= np.isin(self._descending_count, negative)
+
+        return mask
+
+    def _make_mask_from_tuple(self, args: tuple) -> bool | np.ndarray:
+        mask: bool | np.ndarray = False
+
+        for arg in args:
+            if is_integer(arg):
+                mask |= self._make_mask_from_int(cast(int, arg))
+            elif isinstance(arg, slice):
+                mask |= self._make_mask_from_slice(arg)
+            else:
+                raise ValueError(
+                    f"Invalid argument {type(arg)}. Should be int or slice."
+                )
+
+        return mask
+
+    def _make_mask_from_slice(self, arg: slice) -> bool | np.ndarray:
+        start = arg.start
+        stop = arg.stop
+        step = arg.step
+
+        if step is not None and step < 0:
+            raise ValueError(f"Invalid step {step}. Must be non-negative")
+
+        mask: bool | np.ndarray = True
+
+        if step is None:
+            step = 1
+
+        if start is None:
+            if step > 1:
+                mask &= self._ascending_count % step == 0
+
+        elif start >= 0:
+            mask &= self._ascending_count >= start
+
+            if step > 1:
+                mask &= (self._ascending_count - start) % step == 0
+
+        else:
+            mask &= self._descending_count < -start
+
+            offset_array = self._descending_count + start + 1
+            limit_array = (
+                self._ascending_count + self._descending_count + (start + 1)
+            ) < 0
+            offset_array = np.where(limit_array, self._ascending_count, offset_array)
+
+            mask &= offset_array % step == 0
+
+        if stop is not None:
+            if stop >= 0:
+                mask &= self._ascending_count < stop
+            else:
+                mask &= self._descending_count >= -stop
+
+        return mask
+
+    @cache_readonly
+    def _ascending_count(self) -> np.ndarray:
+        if TYPE_CHECKING:
+            groupby_self = cast(groupby.GroupBy, self)
+        else:
+            groupby_self = self
+
+        return groupby_self._cumcount_array()
+
+    @cache_readonly
+    def _descending_count(self) -> np.ndarray:
+        if TYPE_CHECKING:
+            groupby_self = cast(groupby.GroupBy, self)
+        else:
+            groupby_self = self
+
+        return groupby_self._cumcount_array(ascending=False)
+
+
+@doc(GroupByIndexingMixin._positional_selector)
+class GroupByPositionalSelector:
+    def __init__(self, groupby_object: groupby.GroupBy) -> None:
+        self.groupby_object = groupby_object
+
+    def __getitem__(self, arg: PositionalIndexer | tuple) -> DataFrame | Series:
+        """
+        Select by positional index per group.
+
+        Implements GroupBy._positional_selector
+
+        Parameters
+        ----------
+        arg : PositionalIndexer | tuple
+            Allowed values are:
+            - int
+            - int valued iterable such as list or range
+            - slice with step either None or positive
+            - tuple of integers and slices
+
+        Returns
+        -------
+        Series
+            The filtered subset of the original groupby Series.
+        DataFrame
+            The filtered subset of the original groupby DataFrame.
+
+        See Also
+        --------
+        DataFrame.iloc : Integer-location based indexing for selection by position.
+        GroupBy.head : Return first n rows of each group.
+        GroupBy.tail : Return last n rows of each group.
+        GroupBy._positional_selector : Return positional selection for each group.
+        GroupBy.nth : Take the nth row from each group if n is an int, or a
+            subset of rows, if n is a list of ints.
+        """
+        mask = self.groupby_object._make_mask_from_positional_indexer(arg)
+        return self.groupby_object._mask_selected_obj(mask)
+
+
+class GroupByNthSelector:
+    """
+    Dynamically substituted for GroupBy.nth to enable both call and index
+    """
+
+    def __init__(self, groupby_object: groupby.GroupBy) -> None:
+        self.groupby_object = groupby_object
+
+    def __call__(
+        self,
+        n: PositionalIndexer | tuple,
+        dropna: Literal["any", "all", None] = None,
+    ) -> DataFrame | Series:
+        return self.groupby_object._nth(n, dropna)
+
+    def __getitem__(self, n: PositionalIndexer | tuple) -> DataFrame | Series:
+        return self.groupby_object._nth(n)
diff --git a/pandas/pandas/core/groupby/numba_.py b/pandas/pandas/core/groupby/numba_.py
new file mode 100644
index 0000000000000..73b681c64c3a3
--- /dev/null
+++ b/pandas/pandas/core/groupby/numba_.py
@@ -0,0 +1,183 @@
+"""Common utilities for Numba operations with groupby ops"""
+
+from __future__ import annotations
+
+import functools
+import inspect
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.util.numba_ import (
+    NumbaUtilError,
+    jit_user_function,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import Scalar
+
+
+def validate_udf(func: Callable) -> None:
+    """
+    Validate user defined function for ops when using Numba with groupby ops.
+
+    The first signature arguments should include:
+
+    def f(values, index, ...):
+        ...
+
+    Parameters
+    ----------
+    func : function, default False
+        user defined function
+
+    Returns
+    -------
+    None
+
+    Raises
+    ------
+    NumbaUtilError
+    """
+    if not callable(func):
+        raise NotImplementedError(
+            "Numba engine can only be used with a single function."
+        )
+    udf_signature = list(inspect.signature(func).parameters.keys())
+    expected_args = ["values", "index"]
+    min_number_args = len(expected_args)
+    if (
+        len(udf_signature) < min_number_args
+        or udf_signature[:min_number_args] != expected_args
+    ):
+        raise NumbaUtilError(
+            f"The first {min_number_args} arguments to {func.__name__} must be "
+            f"{expected_args}"
+        )
+
+
+@functools.cache
+def generate_numba_agg_func(
+    func: Callable[..., Scalar],
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+) -> Callable[[np.ndarray, np.ndarray, np.ndarray, np.ndarray, int, Any], np.ndarray]:
+    """
+    Generate a numba jitted agg function specified by values from engine_kwargs.
+
+    1. jit the user's function
+    2. Return a groupby agg function with the jitted function inline
+
+    Configurations specified in engine_kwargs apply to both the user's
+    function _AND_ the groupby evaluation loop.
+
+    Parameters
+    ----------
+    func : function
+        function to be applied to each group and will be JITed
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+    numba_func = jit_user_function(func)
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def group_agg(
+        values: np.ndarray,
+        index: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        num_columns: int,
+        *args: Any,
+    ) -> np.ndarray:
+        assert len(begin) == len(end)
+        num_groups = len(begin)
+
+        result = np.empty((num_groups, num_columns))
+        for i in numba.prange(num_groups):
+            group_index = index[begin[i] : end[i]]
+            for j in numba.prange(num_columns):
+                group = values[begin[i] : end[i], j]
+                result[i, j] = numba_func(group, group_index, *args)
+        return result
+
+    return group_agg
+
+
+@functools.cache
+def generate_numba_transform_func(
+    func: Callable[..., np.ndarray],
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+) -> Callable[[np.ndarray, np.ndarray, np.ndarray, np.ndarray, int, Any], np.ndarray]:
+    """
+    Generate a numba jitted transform function specified by values from engine_kwargs.
+
+    1. jit the user's function
+    2. Return a groupby transform function with the jitted function inline
+
+    Configurations specified in engine_kwargs apply to both the user's
+    function _AND_ the groupby evaluation loop.
+
+    Parameters
+    ----------
+    func : function
+        function to be applied to each window and will be JITed
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+    numba_func = jit_user_function(func)
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def group_transform(
+        values: np.ndarray,
+        index: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        num_columns: int,
+        *args: Any,
+    ) -> np.ndarray:
+        assert len(begin) == len(end)
+        num_groups = len(begin)
+
+        result = np.empty((len(values), num_columns))
+        for i in numba.prange(num_groups):
+            group_index = index[begin[i] : end[i]]
+            for j in numba.prange(num_columns):
+                group = values[begin[i] : end[i], j]
+                result[begin[i] : end[i], j] = numba_func(group, group_index, *args)
+        return result
+
+    return group_transform
diff --git a/pandas/pandas/core/groupby/ops.py b/pandas/pandas/core/groupby/ops.py
new file mode 100644
index 0000000000000..c4c7f73ee166c
--- /dev/null
+++ b/pandas/pandas/core/groupby/ops.py
@@ -0,0 +1,1264 @@
+"""
+Provide classes to perform the groupby aggregate operations.
+
+These are not exposed to the user and provide implementations of the grouping
+operations, primarily in cython. These classes (BaseGrouper and BinGrouper)
+are contained *in* the SeriesGroupBy and DataFrameGroupBy objects.
+"""
+
+from __future__ import annotations
+
+import collections
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Generic,
+    final,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    lib,
+)
+import pandas._libs.groupby as libgroupby
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    NDFrameT,
+    Shape,
+    npt,
+)
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.cast import (
+    maybe_cast_pointwise_result,
+    maybe_downcast_to_dtype,
+)
+from pandas.core.dtypes.common import (
+    ensure_float64,
+    ensure_int64,
+    ensure_platform_int,
+    ensure_uint64,
+    is_1d_only_ea_dtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    maybe_fill,
+)
+
+from pandas.core.arrays import Categorical
+from pandas.core.frame import DataFrame
+from pandas.core.groupby import grouper
+from pandas.core.indexes.api import (
+    CategoricalIndex,
+    Index,
+    MultiIndex,
+    ensure_index,
+)
+from pandas.core.series import Series
+from pandas.core.sorting import (
+    compress_group_index,
+    decons_obs_group_ids,
+    get_group_index,
+    get_group_index_sorter,
+    get_indexer_dict,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Generator,
+        Hashable,
+        Iterator,
+    )
+
+    from pandas.core.generic import NDFrame
+
+
+def check_result_array(obj, dtype) -> None:
+    # Our operation is supposed to be an aggregation/reduction. If
+    #  it returns an ndarray, this likely means an invalid operation has
+    #  been passed. See test_apply_without_aggregation, test_agg_must_agg
+    if isinstance(obj, np.ndarray):
+        if dtype != object:
+            # If it is object dtype, the function can be a reduction/aggregation
+            #  and still return an ndarray e.g. test_agg_over_numpy_arrays
+            raise ValueError("Must produce aggregated value")
+
+
+def extract_result(res):
+    """
+    Extract the result object, it might be a 0-dim ndarray
+    or a len-1 0-dim, or a scalar
+    """
+    if hasattr(res, "_values"):
+        # Preserve EA
+        res = res._values
+        if res.ndim == 1 and len(res) == 1:
+            # see test_agg_lambda_with_timezone, test_resampler_grouper.py::test_apply
+            res = res[0]
+    return res
+
+
+class WrappedCythonOp:
+    """
+    Dispatch logic for functions defined in _libs.groupby
+
+    Parameters
+    ----------
+    kind: str
+        Whether the operation is an aggregate or transform.
+    how: str
+        Operation name, e.g. "mean".
+    has_dropped_na: bool
+        True precisely when dropna=True and the grouper contains a null value.
+    """
+
+    # Functions for which we do _not_ attempt to cast the cython result
+    #  back to the original dtype.
+    cast_blocklist = frozenset(
+        ["any", "all", "rank", "count", "size", "idxmin", "idxmax"]
+    )
+
+    def __init__(self, kind: str, how: str, has_dropped_na: bool) -> None:
+        self.kind = kind
+        self.how = how
+        self.has_dropped_na = has_dropped_na
+
+    _CYTHON_FUNCTIONS: dict[str, dict] = {
+        "aggregate": {
+            "any": functools.partial(libgroupby.group_any_all, val_test="any"),
+            "all": functools.partial(libgroupby.group_any_all, val_test="all"),
+            "sum": "group_sum",
+            "prod": "group_prod",
+            "idxmin": functools.partial(libgroupby.group_idxmin_idxmax, name="idxmin"),
+            "idxmax": functools.partial(libgroupby.group_idxmin_idxmax, name="idxmax"),
+            "min": "group_min",
+            "max": "group_max",
+            "mean": "group_mean",
+            "median": "group_median_float64",
+            "var": "group_var",
+            "std": functools.partial(libgroupby.group_var, name="std"),
+            "sem": functools.partial(libgroupby.group_var, name="sem"),
+            "skew": "group_skew",
+            "kurt": "group_kurt",
+            "first": "group_nth",
+            "last": "group_last",
+            "ohlc": "group_ohlc",
+        },
+        "transform": {
+            "cumprod": "group_cumprod",
+            "cumsum": "group_cumsum",
+            "cummin": "group_cummin",
+            "cummax": "group_cummax",
+            "rank": "group_rank",
+        },
+    }
+
+    _cython_arity = {"ohlc": 4}  # OHLC
+
+    @classmethod
+    def get_kind_from_how(cls, how: str) -> str:
+        if how in cls._CYTHON_FUNCTIONS["aggregate"]:
+            return "aggregate"
+        return "transform"
+
+    # Note: we make this a classmethod and pass kind+how so that caching
+    #  works at the class level and not the instance level
+    @classmethod
+    @functools.cache
+    def _get_cython_function(
+        cls, kind: str, how: str, dtype: np.dtype, is_numeric: bool
+    ):
+        dtype_str = dtype.name
+        ftype = cls._CYTHON_FUNCTIONS[kind][how]
+
+        # see if there is a fused-type version of function
+        # only valid for numeric
+        if callable(ftype):
+            f = ftype
+        else:
+            f = getattr(libgroupby, ftype)
+        if is_numeric:
+            return f
+        elif dtype == np.dtype(object):
+            if how in ["median", "cumprod"]:
+                # no fused types -> no __signatures__
+                raise NotImplementedError(
+                    f"function is not implemented for this dtype: "
+                    f"[how->{how},dtype->{dtype_str}]"
+                )
+            elif how in ["std", "sem", "idxmin", "idxmax"]:
+                # We have a partial object that does not have __signatures__
+                return f
+            elif how in ["skew", "kurt"]:
+                # _get_cython_vals will convert to float64
+                pass
+            elif "object" not in f.__signatures__:
+                # raise NotImplementedError here rather than TypeError later
+                raise NotImplementedError(
+                    f"function is not implemented for this dtype: "
+                    f"[how->{how},dtype->{dtype_str}]"
+                )
+            return f
+        else:
+            raise NotImplementedError(
+                "This should not be reached. Please report a bug at "
+                "github.com/pandas-dev/pandas/",
+                dtype,
+            )
+
+    def _get_cython_vals(self, values: np.ndarray) -> np.ndarray:
+        """
+        Cast numeric dtypes to float64 for functions that only support that.
+
+        Parameters
+        ----------
+        values : np.ndarray
+
+        Returns
+        -------
+        values : np.ndarray
+        """
+        how = self.how
+
+        if how in ["median", "std", "sem", "skew", "kurt"]:
+            # median only has a float64 implementation
+            # We should only get here with is_numeric, as non-numeric cases
+            #  should raise in _get_cython_function
+            values = ensure_float64(values)
+
+        elif values.dtype.kind in "iu":
+            if how in ["var", "mean"] or (
+                self.kind == "transform" and self.has_dropped_na
+            ):
+                # has_dropped_na check need for test_null_group_str_transformer
+                # result may still include NaN, so we have to cast
+                values = ensure_float64(values)
+
+            elif how in ["sum", "ohlc", "prod", "cumsum", "cumprod"]:
+                # Avoid overflow during group op
+                if values.dtype.kind == "i":
+                    values = ensure_int64(values)
+                else:
+                    values = ensure_uint64(values)
+
+        return values
+
+    def _get_output_shape(self, ngroups: int, values: np.ndarray) -> Shape:
+        how = self.how
+        kind = self.kind
+
+        arity = self._cython_arity.get(how, 1)
+
+        out_shape: Shape
+        if how == "ohlc":
+            out_shape = (ngroups, arity)
+        elif arity > 1:
+            raise NotImplementedError(
+                "arity of more than 1 is not supported for the 'how' argument"
+            )
+        elif kind == "transform":
+            out_shape = values.shape
+        else:
+            out_shape = (ngroups,) + values.shape[1:]
+        return out_shape
+
+    def _get_out_dtype(self, dtype: np.dtype) -> np.dtype:
+        how = self.how
+
+        if how == "rank":
+            out_dtype = "float64"
+        elif how in ["idxmin", "idxmax"]:
+            # The Cython implementation only produces the row number; we'll take
+            # from the index using this in post processing
+            out_dtype = "intp"
+        else:
+            if dtype.kind in "iufcb":
+                out_dtype = f"{dtype.kind}{dtype.itemsize}"
+            else:
+                out_dtype = "object"
+        return np.dtype(out_dtype)
+
+    def _get_result_dtype(self, dtype: np.dtype) -> np.dtype:
+        """
+        Get the desired dtype of a result based on the
+        input dtype and how it was computed.
+
+        Parameters
+        ----------
+        dtype : np.dtype
+
+        Returns
+        -------
+        np.dtype
+            The desired dtype of the result.
+        """
+        how = self.how
+
+        if how in ["sum", "cumsum", "sum", "prod", "cumprod"]:
+            if dtype == np.dtype(bool):
+                return np.dtype(np.int64)
+        elif how in ["mean", "median", "var", "std", "sem"]:
+            if dtype.kind in "fc":
+                return dtype
+            elif dtype.kind in "iub":
+                return np.dtype(np.float64)
+        return dtype
+
+    @final
+    def _cython_op_ndim_compat(
+        self,
+        values: np.ndarray,
+        *,
+        min_count: int,
+        ngroups: int,
+        comp_ids: np.ndarray,
+        mask: npt.NDArray[np.bool_] | None = None,
+        result_mask: npt.NDArray[np.bool_] | None = None,
+        **kwargs,
+    ) -> np.ndarray:
+        if values.ndim == 1:
+            # expand to 2d, dispatch, then squeeze if appropriate
+            values2d = values[None, :]
+            if mask is not None:
+                mask = mask[None, :]
+            if result_mask is not None:
+                result_mask = result_mask[None, :]
+            res = self._call_cython_op(
+                values2d,
+                min_count=min_count,
+                ngroups=ngroups,
+                comp_ids=comp_ids,
+                mask=mask,
+                result_mask=result_mask,
+                **kwargs,
+            )
+            if res.shape[0] == 1:
+                return res[0]
+
+            # otherwise we have OHLC
+            return res.T
+
+        return self._call_cython_op(
+            values,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=comp_ids,
+            mask=mask,
+            result_mask=result_mask,
+            **kwargs,
+        )
+
+    @final
+    def _call_cython_op(
+        self,
+        values: np.ndarray,  # np.ndarray[ndim=2]
+        *,
+        min_count: int,
+        ngroups: int,
+        comp_ids: np.ndarray,
+        mask: npt.NDArray[np.bool_] | None,
+        result_mask: npt.NDArray[np.bool_] | None,
+        **kwargs,
+    ) -> np.ndarray:  # np.ndarray[ndim=2]
+        orig_values = values
+
+        dtype = values.dtype
+        is_numeric = dtype.kind in "iufcb"
+
+        is_datetimelike = dtype.kind in "mM"
+
+        if self.how in ["any", "all"]:
+            if mask is None:
+                mask = isna(values)
+
+        if is_datetimelike:
+            values = values.view("int64")
+            is_numeric = True
+        elif dtype.kind == "b":
+            values = values.view("uint8")
+        if values.dtype == "float16":
+            values = values.astype(np.float32)
+
+        if self.how in ["any", "all"]:
+            if dtype == object:
+                if kwargs["skipna"]:
+                    # GH#37501: don't raise on pd.NA when skipna=True
+                    if mask is not None and mask.any():
+                        # mask on original values computed separately
+                        values = values.copy()
+                        values[mask] = True
+            values = values.astype(bool, copy=False).view(np.int8)
+            is_numeric = True
+
+        values = values.T
+        if mask is not None:
+            mask = mask.T
+            if result_mask is not None:
+                result_mask = result_mask.T
+
+        out_shape = self._get_output_shape(ngroups, values)
+        func = self._get_cython_function(self.kind, self.how, values.dtype, is_numeric)
+        values = self._get_cython_vals(values)
+        out_dtype = self._get_out_dtype(values.dtype)
+
+        result = maybe_fill(np.empty(out_shape, dtype=out_dtype))
+        if self.kind == "aggregate":
+            counts = np.zeros(ngroups, dtype=np.int64)
+            if self.how in [
+                "idxmin",
+                "idxmax",
+                "min",
+                "max",
+                "mean",
+                "last",
+                "first",
+                "sum",
+                "median",
+            ]:
+                func(
+                    out=result,
+                    counts=counts,
+                    values=values,
+                    labels=comp_ids,
+                    min_count=min_count,
+                    mask=mask,
+                    result_mask=result_mask,
+                    is_datetimelike=is_datetimelike,
+                    **kwargs,
+                )
+            elif self.how in ["sem", "std", "var", "ohlc", "prod"]:
+                if self.how in ["std", "sem"]:
+                    kwargs["is_datetimelike"] = is_datetimelike
+                func(
+                    result,
+                    counts,
+                    values,
+                    comp_ids,
+                    min_count=min_count,
+                    mask=mask,
+                    result_mask=result_mask,
+                    **kwargs,
+                )
+            elif self.how in ["any", "all"]:
+                func(
+                    out=result,
+                    values=values,
+                    labels=comp_ids,
+                    mask=mask,
+                    result_mask=result_mask,
+                    **kwargs,
+                )
+                result = result.astype(bool, copy=False)
+            elif self.how in ["skew", "kurt"]:
+                func(
+                    out=result,
+                    counts=counts,
+                    values=values,
+                    labels=comp_ids,
+                    mask=mask,
+                    result_mask=result_mask,
+                    **kwargs,
+                )
+                if dtype == object:
+                    result = result.astype(object)
+
+            else:
+                raise NotImplementedError(f"{self.how} is not implemented")
+        else:
+            # TODO: min_count
+            if self.how != "rank":
+                # TODO: should rank take result_mask?
+                kwargs["result_mask"] = result_mask
+            func(
+                out=result,
+                values=values,
+                labels=comp_ids,
+                ngroups=ngroups,
+                is_datetimelike=is_datetimelike,
+                mask=mask,
+                **kwargs,
+            )
+
+        if self.kind == "aggregate" and self.how not in ["idxmin", "idxmax"]:
+            # i.e. counts is defined.  Locations where count<min_count
+            # need to have the result set to np.nan, which may require casting,
+            # see GH#40767. For idxmin/idxmax is handled specially via post-processing
+            if result.dtype.kind in "iu" and not is_datetimelike:
+                # if the op keeps the int dtypes, we have to use 0
+                cutoff = max(0 if self.how in ["sum", "prod"] else 1, min_count)
+                empty_groups = counts < cutoff
+                if empty_groups.any():
+                    if result_mask is not None:
+                        assert result_mask[empty_groups].all()
+                    else:
+                        # Note: this conversion could be lossy, see GH#40767
+                        result = result.astype("float64")
+                        result[empty_groups] = np.nan
+
+        result = result.T
+
+        if self.how not in self.cast_blocklist:
+            # e.g. if we are int64 and need to restore to datetime64/timedelta64
+            # "rank" is the only member of cast_blocklist we get here
+            # Casting only needed for float16, bool, datetimelike,
+            #  and self.how in ["sum", "prod", "ohlc", "cumprod"]
+            res_dtype = self._get_result_dtype(orig_values.dtype)
+            op_result = maybe_downcast_to_dtype(result, res_dtype)
+        else:
+            op_result = result
+
+        return op_result
+
+    @final
+    def _validate_axis(self, axis: AxisInt, values: ArrayLike) -> None:
+        if values.ndim > 2:
+            raise NotImplementedError("number of dimensions is currently limited to 2")
+        if values.ndim == 2:
+            assert axis == 1, axis
+        elif not is_1d_only_ea_dtype(values.dtype):
+            # Note: it is *not* the case that axis is always 0 for 1-dim values,
+            #  as we can have 1D ExtensionArrays that we need to treat as 2D
+            assert axis == 0
+
+    @final
+    def cython_operation(
+        self,
+        *,
+        values: ArrayLike,
+        axis: AxisInt,
+        min_count: int = -1,
+        comp_ids: np.ndarray,
+        ngroups: int,
+        **kwargs,
+    ) -> ArrayLike:
+        """
+        Call our cython function, with appropriate pre- and post- processing.
+        """
+        self._validate_axis(axis, values)
+
+        if not isinstance(values, np.ndarray):
+            # i.e. ExtensionArray
+            return values._groupby_op(
+                how=self.how,
+                has_dropped_na=self.has_dropped_na,
+                min_count=min_count,
+                ngroups=ngroups,
+                ids=comp_ids,
+                **kwargs,
+            )
+
+        return self._cython_op_ndim_compat(
+            values,
+            min_count=min_count,
+            ngroups=ngroups,
+            comp_ids=comp_ids,
+            mask=None,
+            **kwargs,
+        )
+
+
+class BaseGrouper:
+    """
+    This is an internal Grouper class, which actually holds
+    the generated groups
+
+    Parameters
+    ----------
+    axis : Index
+    groupings : Sequence[Grouping]
+        all the grouping instances to handle in this grouper
+        for example for grouper list to groupby, need to pass the list
+    sort : bool, default True
+        whether this grouper will give sorted result or not
+
+    """
+
+    axis: Index
+
+    def __init__(
+        self,
+        axis: Index,
+        groupings: list[grouper.Grouping],
+        sort: bool = True,
+        dropna: bool = True,
+    ) -> None:
+        assert isinstance(axis, Index), axis
+
+        self.axis = axis
+        self._groupings = groupings
+        self._sort = sort
+        self.dropna = dropna
+
+    @property
+    def groupings(self) -> list[grouper.Grouping]:
+        return self._groupings
+
+    def __iter__(self) -> Iterator[Hashable]:
+        return iter(self.indices)
+
+    @property
+    def nkeys(self) -> int:
+        return len(self.groupings)
+
+    def get_iterator(self, data: NDFrameT) -> Iterator[tuple[Hashable, NDFrameT]]:
+        """
+        Groupby iterator
+
+        Returns
+        -------
+        Generator yielding sequence of (name, subsetted object)
+        for each group
+        """
+        splitter = self._get_splitter(data)
+        # TODO: Would be more efficient to skip unobserved for transforms
+        keys = self.result_index
+        yield from zip(keys, splitter)
+
+    @final
+    def _get_splitter(self, data: NDFrame) -> DataSplitter:
+        """
+        Returns
+        -------
+        Generator yielding subsetted objects
+        """
+        if isinstance(data, Series):
+            klass: type[DataSplitter] = SeriesSplitter
+        else:
+            # i.e. DataFrame
+            klass = FrameSplitter
+
+        return klass(
+            data,
+            self.ngroups,
+            sorted_ids=self._sorted_ids,
+            sort_idx=self.result_ilocs,
+        )
+
+    @cache_readonly
+    def indices(self) -> dict[Hashable, npt.NDArray[np.intp]]:
+        """dict {group name -> group indices}"""
+        if len(self.groupings) == 1 and isinstance(self.result_index, CategoricalIndex):
+            # This shows unused categories in indices GH#38642
+            return self.groupings[0].indices
+        codes_list = [ping.codes for ping in self.groupings]
+        return get_indexer_dict(codes_list, self.levels)
+
+    @final
+    @cache_readonly
+    def result_ilocs(self) -> npt.NDArray[np.intp]:
+        """
+        Get the original integer locations of result_index in the input.
+        """
+        # Original indices are where group_index would go via sorting.
+        # But when dropna is true, we need to remove null values while accounting for
+        # any gaps that then occur because of them.
+        ids = self.ids
+
+        if self.has_dropped_na:
+            mask = np.where(ids >= 0)
+            # Count how many gaps are caused by previous null values for each position
+            null_gaps = np.cumsum(ids == -1)[mask]
+            ids = ids[mask]
+
+        result = get_group_index_sorter(ids, self.ngroups)
+
+        if self.has_dropped_na:
+            # Shift by the number of prior null gaps
+            result += np.take(null_gaps, result)
+
+        return result
+
+    @property
+    def codes(self) -> list[npt.NDArray[np.signedinteger]]:
+        return [ping.codes for ping in self.groupings]
+
+    @property
+    def levels(self) -> list[Index]:
+        if len(self.groupings) > 1:
+            # mypy doesn't know result_index must be a MultiIndex
+            return list(self.result_index.levels)  # type: ignore[attr-defined]
+        else:
+            return [self.result_index]
+
+    @property
+    def names(self) -> list[Hashable]:
+        return [ping.name for ping in self.groupings]
+
+    @final
+    def size(self) -> Series:
+        """
+        Compute group sizes.
+        """
+        ids = self.ids
+        ngroups = self.ngroups
+        out: np.ndarray | list
+        if ngroups:
+            out = np.bincount(ids[ids != -1], minlength=ngroups)
+        else:
+            out = []
+        return Series(out, index=self.result_index, dtype="int64", copy=False)
+
+    @cache_readonly
+    def groups(self) -> dict[Hashable, Index]:
+        """dict {group name -> group labels}"""
+        if len(self.groupings) == 1:
+            return self.groupings[0].groups
+        result_index, ids = self.result_index_and_ids
+        values = result_index._values
+        categories = Categorical(ids, categories=range(len(result_index)))
+        result = {
+            # mypy is not aware that group has to be an integer
+            values[group]: self.axis.take(axis_ilocs)  # type: ignore[call-overload]
+            for group, axis_ilocs in categories._reverse_indexer().items()
+        }
+        return result
+
+    @final
+    @cache_readonly
+    def is_monotonic(self) -> bool:
+        # return if my group orderings are monotonic
+        return Index(self.ids).is_monotonic_increasing
+
+    @final
+    @cache_readonly
+    def has_dropped_na(self) -> bool:
+        """
+        Whether grouper has null value(s) that are dropped.
+        """
+        return bool((self.ids < 0).any())
+
+    @cache_readonly
+    def codes_info(self) -> npt.NDArray[np.intp]:
+        # return the codes of items in original grouped axis
+        return self.ids
+
+    @final
+    @cache_readonly
+    def ngroups(self) -> int:
+        return len(self.result_index)
+
+    @property
+    def result_index(self) -> Index:
+        return self.result_index_and_ids[0]
+
+    @property
+    def ids(self) -> npt.NDArray[np.intp]:
+        return self.result_index_and_ids[1]
+
+    @cache_readonly
+    def result_index_and_ids(self) -> tuple[Index, npt.NDArray[np.intp]]:
+        levels = [Index._with_infer(ping.uniques) for ping in self.groupings]
+        obs = [
+            ping._observed or not ping._passed_categorical for ping in self.groupings
+        ]
+        sorts = [ping._sort for ping in self.groupings]
+        # When passed a categorical grouping, keep all categories
+        for k, (ping, level) in enumerate(zip(self.groupings, levels)):
+            if ping._passed_categorical:
+                levels[k] = level.set_categories(ping._orig_cats)
+
+        if len(self.groupings) == 1:
+            result_index = levels[0]
+            result_index.name = self.names[0]
+            ids = ensure_platform_int(self.codes[0])
+        elif all(obs):
+            result_index, ids = self._ob_index_and_ids(
+                levels, self.codes, self.names, sorts
+            )
+        elif not any(obs):
+            result_index, ids = self._unob_index_and_ids(levels, self.codes, self.names)
+        else:
+            # Combine unobserved and observed parts
+            names = self.names
+            codes = [ping.codes for ping in self.groupings]
+            ob_indices = [idx for idx, ob in enumerate(obs) if ob]
+            unob_indices = [idx for idx, ob in enumerate(obs) if not ob]
+            ob_index, ob_ids = self._ob_index_and_ids(
+                levels=[levels[idx] for idx in ob_indices],
+                codes=[codes[idx] for idx in ob_indices],
+                names=[names[idx] for idx in ob_indices],
+                sorts=[sorts[idx] for idx in ob_indices],
+            )
+            unob_index, unob_ids = self._unob_index_and_ids(
+                levels=[levels[idx] for idx in unob_indices],
+                codes=[codes[idx] for idx in unob_indices],
+                names=[names[idx] for idx in unob_indices],
+            )
+
+            result_index_codes = np.concatenate(
+                [
+                    np.tile(unob_index.codes, len(ob_index)),
+                    np.repeat(ob_index.codes, len(unob_index), axis=1),
+                ],
+                axis=0,
+            )
+            _, index = np.unique(unob_indices + ob_indices, return_index=True)
+            result_index = MultiIndex(
+                levels=list(unob_index.levels) + list(ob_index.levels),
+                codes=result_index_codes,
+                names=list(unob_index.names) + list(ob_index.names),
+            ).reorder_levels(index)
+            ids = len(unob_index) * ob_ids + unob_ids
+
+            if any(sorts):
+                # Sort result_index and recode ids using the new order
+                n_levels = len(sorts)
+                drop_levels = [
+                    n_levels - idx
+                    for idx, sort in enumerate(reversed(sorts), 1)
+                    if not sort
+                ]
+                if len(drop_levels) > 0:
+                    sorter = result_index._drop_level_numbers(drop_levels).argsort()
+                else:
+                    sorter = result_index.argsort()
+                result_index = result_index.take(sorter)
+                _, index = np.unique(sorter, return_index=True)
+                ids = ensure_platform_int(ids)
+                ids = index.take(ids)
+            else:
+                # Recode ids and reorder result_index with observed groups up front,
+                # unobserved at the end
+                ids, uniques = compress_group_index(ids, sort=False)
+                ids = ensure_platform_int(ids)
+                taker = np.concatenate(
+                    [uniques, np.delete(np.arange(len(result_index)), uniques)]
+                )
+                result_index = result_index.take(taker)
+
+        return result_index, ids
+
+    @property
+    def observed_grouper(self) -> BaseGrouper:
+        if all(ping._observed for ping in self.groupings):
+            return self
+
+        return self._observed_grouper
+
+    @cache_readonly
+    def _observed_grouper(self) -> BaseGrouper:
+        groupings = [ping.observed_grouping for ping in self.groupings]
+        grouper = BaseGrouper(self.axis, groupings, sort=self._sort, dropna=self.dropna)
+        return grouper
+
+    def _ob_index_and_ids(
+        self,
+        levels: list[Index],
+        codes: list[npt.NDArray[np.intp]],
+        names: list[Hashable],
+        sorts: list[bool],
+    ) -> tuple[MultiIndex, npt.NDArray[np.intp]]:
+        consistent_sorting = all(sorts[0] == sort for sort in sorts[1:])
+        sort_in_compress = sorts[0] if consistent_sorting else False
+        shape = tuple(len(level) for level in levels)
+        group_index = get_group_index(codes, shape, sort=True, xnull=True)
+        ob_ids, obs_group_ids = compress_group_index(group_index, sort=sort_in_compress)
+        ob_ids = ensure_platform_int(ob_ids)
+        ob_index_codes = decons_obs_group_ids(
+            ob_ids, obs_group_ids, shape, codes, xnull=True
+        )
+        ob_index = MultiIndex(
+            levels=levels,
+            codes=ob_index_codes,
+            names=names,
+            verify_integrity=False,
+        )
+        if not consistent_sorting and len(ob_index) > 0:
+            # Sort by the levels where the corresponding sort argument is True
+            n_levels = len(sorts)
+            drop_levels = [
+                n_levels - idx
+                for idx, sort in enumerate(reversed(sorts), 1)
+                if not sort
+            ]
+            if len(drop_levels) > 0:
+                sorter = ob_index._drop_level_numbers(drop_levels).argsort()
+            else:
+                sorter = ob_index.argsort()
+            ob_index = ob_index.take(sorter)
+            _, index = np.unique(sorter, return_index=True)
+            ob_ids = np.where(ob_ids == -1, -1, index.take(ob_ids))
+        ob_ids = ensure_platform_int(ob_ids)
+        return ob_index, ob_ids
+
+    def _unob_index_and_ids(
+        self,
+        levels: list[Index],
+        codes: list[npt.NDArray[np.intp]],
+        names: list[Hashable],
+    ) -> tuple[MultiIndex, npt.NDArray[np.intp]]:
+        shape = tuple(len(level) for level in levels)
+        unob_ids = get_group_index(codes, shape, sort=True, xnull=True)
+        unob_index = MultiIndex.from_product(levels, names=names)
+        unob_ids = ensure_platform_int(unob_ids)
+        return unob_index, unob_ids
+
+    @final
+    def get_group_levels(self) -> Generator[Index]:
+        # Note: only called from _insert_inaxis_grouper, which
+        #  is only called for BaseGrouper, never for BinGrouper
+        result_index = self.result_index
+        if len(self.groupings) == 1:
+            yield result_index
+        else:
+            for level in range(result_index.nlevels - 1, -1, -1):
+                yield result_index.get_level_values(level)
+
+    # ------------------------------------------------------------
+    # Aggregation functions
+
+    @final
+    def _cython_operation(
+        self,
+        kind: str,
+        values,
+        how: str,
+        axis: AxisInt,
+        min_count: int = -1,
+        **kwargs,
+    ) -> ArrayLike:
+        """
+        Returns the values of a cython operation.
+        """
+        assert kind in ["transform", "aggregate"]
+
+        cy_op = WrappedCythonOp(kind=kind, how=how, has_dropped_na=self.has_dropped_na)
+
+        return cy_op.cython_operation(
+            values=values,
+            axis=axis,
+            min_count=min_count,
+            comp_ids=self.ids,
+            ngroups=self.ngroups,
+            **kwargs,
+        )
+
+    @final
+    def agg_series(
+        self, obj: Series, func: Callable, preserve_dtype: bool = False
+    ) -> ArrayLike:
+        """
+        Parameters
+        ----------
+        obj : Series
+        func : function taking a Series and returning a scalar-like
+        preserve_dtype : bool
+            Whether the aggregation is known to be dtype-preserving.
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray
+        """
+
+        if not isinstance(obj._values, np.ndarray):
+            # we can preserve a little bit more aggressively with EA dtype
+            #  because maybe_cast_pointwise_result will do a try/except
+            #  with _from_sequence.  NB we are assuming here that _from_sequence
+            #  is sufficiently strict that it casts appropriately.
+            preserve_dtype = True
+
+        result = self._aggregate_series_pure_python(obj, func)
+
+        npvalues = lib.maybe_convert_objects(result, try_float=False)
+        if preserve_dtype:
+            out = maybe_cast_pointwise_result(npvalues, obj.dtype, numeric_only=True)
+        else:
+            out = npvalues
+        return out
+
+    @final
+    def _aggregate_series_pure_python(
+        self, obj: Series, func: Callable
+    ) -> npt.NDArray[np.object_]:
+        result = np.empty(self.ngroups, dtype="O")
+        initialized = False
+
+        splitter = self._get_splitter(obj)
+
+        for i, group in enumerate(splitter):
+            res = func(group)
+            res = extract_result(res)
+
+            if not initialized:
+                # We only do this validation on the first iteration
+                check_result_array(res, group.dtype)
+                initialized = True
+
+            result[i] = res
+
+        return result
+
+    @final
+    def apply_groupwise(
+        self, f: Callable, data: DataFrame | Series
+    ) -> tuple[list, bool]:
+        mutated = False
+        splitter = self._get_splitter(data)
+        group_keys = self.result_index
+        result_values = []
+
+        # This calls DataSplitter.__iter__
+        zipped = zip(group_keys, splitter)
+
+        for key, group in zipped:
+            # Pinning name is needed for
+            #  test_group_apply_once_per_group,
+            #  test_inconsistent_return_type, test_set_group_name,
+            #  test_group_name_available_in_inference_pass,
+            #  test_groupby_multi_timezone
+            object.__setattr__(group, "name", key)
+
+            # group might be modified
+            group_axes = group.axes
+            res = f(group)
+            if not mutated and not _is_indexed_like(res, group_axes):
+                mutated = True
+            result_values.append(res)
+        # getattr pattern for __name__ is needed for functools.partial objects
+        if len(group_keys) == 0 and getattr(f, "__name__", None) in [
+            "skew",
+            "kurt",
+            "sum",
+            "prod",
+        ]:
+            #  If group_keys is empty, then no function calls have been made,
+            #  so we will not have raised even if this is an invalid dtype.
+            #  So do one dummy call here to raise appropriate TypeError.
+            f(data.iloc[:0])
+
+        return result_values, mutated
+
+    # ------------------------------------------------------------
+    # Methods for sorting subsets of our GroupBy's object
+
+    @final
+    @cache_readonly
+    def _sorted_ids(self) -> npt.NDArray[np.intp]:
+        result = self.ids.take(self.result_ilocs)
+        if getattr(self, "dropna", True):
+            # BinGrouper has no dropna
+            result = result[result >= 0]
+        return result
+
+
+class BinGrouper(BaseGrouper):
+    """
+    This is an internal Grouper class
+
+    Parameters
+    ----------
+    bins : the split index of binlabels to group the item of axis
+    binlabels : the label list
+    indexer : np.ndarray[np.intp], optional
+        the indexer created by Grouper
+        some groupers (TimeGrouper) will sort its axis and its
+        group_info is also sorted, so need the indexer to reorder
+
+    Examples
+    --------
+    bins: [2, 4, 6, 8, 10]
+    binlabels: DatetimeIndex(['2005-01-01', '2005-01-03',
+        '2005-01-05', '2005-01-07', '2005-01-09'],
+        dtype='datetime64[ns]', freq='2D')
+
+    the group_info, which contains the label of each item in grouped
+    axis, the index of label in label list, group number, is
+
+    (array([0, 0, 1, 1, 2, 2, 3, 3, 4, 4]), array([0, 1, 2, 3, 4]), 5)
+
+    means that, the grouped axis has 10 items, can be grouped into 5
+    labels, the first and second items belong to the first label, the
+    third and forth items belong to the second label, and so on
+
+    """
+
+    bins: npt.NDArray[np.int64]
+    binlabels: Index
+
+    def __init__(
+        self,
+        bins,
+        binlabels,
+        indexer=None,
+    ) -> None:
+        self.bins = ensure_int64(bins)
+        self.binlabels = ensure_index(binlabels)
+        self.indexer = indexer
+
+        # These lengths must match, otherwise we could call agg_series
+        #  with empty self.bins, which would raise later.
+        assert len(self.binlabels) == len(self.bins)
+
+    @cache_readonly
+    def groups(self):
+        """dict {group name -> group labels}"""
+        # this is mainly for compat
+        # GH 3881
+        result = {
+            key: value
+            for key, value in zip(self.binlabels, self.bins)
+            if key is not NaT
+        }
+        return result
+
+    @property
+    def nkeys(self) -> int:
+        # still matches len(self.groupings), but we can hard-code
+        return 1
+
+    @cache_readonly
+    def codes_info(self) -> npt.NDArray[np.intp]:
+        # return the codes of items in original grouped axis
+        ids = self.ids
+        if self.indexer is not None:
+            sorter = np.lexsort((ids, self.indexer))
+            ids = ids[sorter]
+        return ids
+
+    def get_iterator(self, data: NDFrame):
+        """
+        Groupby iterator
+
+        Returns
+        -------
+        Generator yielding sequence of (name, subsetted object)
+        for each group
+        """
+        slicer = lambda start, edge: data.iloc[start:edge]
+
+        start = 0
+        for edge, label in zip(self.bins, self.binlabels):
+            if label is not NaT:
+                yield label, slicer(start, edge)
+            start = edge
+
+        if start < len(data):
+            yield self.binlabels[-1], slicer(start, None)
+
+    @cache_readonly
+    def indices(self):
+        indices = collections.defaultdict(list)
+
+        i = 0
+        for label, bin in zip(self.binlabels, self.bins):
+            if i < bin:
+                if label is not NaT:
+                    indices[label] = list(range(i, bin))
+                i = bin
+        return indices
+
+    @cache_readonly
+    def codes(self) -> list[npt.NDArray[np.intp]]:
+        return [self.ids]
+
+    @cache_readonly
+    def result_index_and_ids(self):
+        result_index = self.binlabels
+        if len(self.binlabels) != 0 and isna(self.binlabels[0]):
+            result_index = result_index[1:]
+
+        ngroups = len(result_index)
+        rep = np.diff(np.r_[0, self.bins])
+
+        rep = ensure_platform_int(rep)
+        if ngroups == len(self.bins):
+            ids = np.repeat(np.arange(ngroups), rep)
+        else:
+            ids = np.repeat(np.r_[-1, np.arange(ngroups)], rep)
+        ids = ensure_platform_int(ids)
+
+        return result_index, ids
+
+    @property
+    def levels(self) -> list[Index]:
+        return [self.binlabels]
+
+    @property
+    def names(self) -> list[Hashable]:
+        return [self.binlabels.name]
+
+    @property
+    def groupings(self) -> list[grouper.Grouping]:
+        lev = self.binlabels
+        codes = self.ids
+        labels = lev.take(codes)
+        ping = grouper.Grouping(
+            labels, labels, in_axis=False, level=None, uniques=lev._values
+        )
+        return [ping]
+
+    @property
+    def observed_grouper(self) -> BinGrouper:
+        return self
+
+
+def _is_indexed_like(obj, axes) -> bool:
+    if isinstance(obj, Series):
+        if len(axes) > 1:
+            return False
+        return obj.index.equals(axes[0])
+    elif isinstance(obj, DataFrame):
+        return obj.index.equals(axes[0])
+
+    return False
+
+
+# ----------------------------------------------------------------------
+# Splitting / application
+
+
+class DataSplitter(Generic[NDFrameT]):
+    def __init__(
+        self,
+        data: NDFrameT,
+        ngroups: int,
+        *,
+        sort_idx: npt.NDArray[np.intp],
+        sorted_ids: npt.NDArray[np.intp],
+    ) -> None:
+        self.data = data
+        self.ngroups = ngroups
+
+        self._slabels = sorted_ids
+        self._sort_idx = sort_idx
+
+    def __iter__(self) -> Iterator:
+        if self.ngroups == 0:
+            # we are inside a generator, rather than raise StopIteration
+            # we merely return signal the end
+            return
+
+        starts, ends = lib.generate_slices(self._slabels, self.ngroups)
+        sdata = self._sorted_data
+        for start, end in zip(starts, ends):
+            yield self._chop(sdata, slice(start, end))
+
+    @cache_readonly
+    def _sorted_data(self) -> NDFrameT:
+        return self.data.take(self._sort_idx, axis=0)
+
+    def _chop(self, sdata, slice_obj: slice) -> NDFrame:
+        raise AbstractMethodError(self)
+
+
+class SeriesSplitter(DataSplitter):
+    def _chop(self, sdata: Series, slice_obj: slice) -> Series:
+        # fastpath equivalent to `sdata.iloc[slice_obj]`
+        mgr = sdata._mgr.get_slice(slice_obj)
+        ser = sdata._constructor_from_mgr(mgr, axes=mgr.axes)
+        ser._name = sdata.name
+        return ser.__finalize__(sdata, method="groupby")
+
+
+class FrameSplitter(DataSplitter):
+    def _chop(self, sdata: DataFrame, slice_obj: slice) -> DataFrame:
+        # Fastpath equivalent to:
+        # return sdata.iloc[slice_obj]
+        mgr = sdata._mgr.get_slice(slice_obj, axis=1)
+        df = sdata._constructor_from_mgr(mgr, axes=mgr.axes)
+        return df.__finalize__(sdata, method="groupby")
diff --git a/pandas/pandas/core/indexers/__init__.py b/pandas/pandas/core/indexers/__init__.py
new file mode 100644
index 0000000000000..036b32b3feac2
--- /dev/null
+++ b/pandas/pandas/core/indexers/__init__.py
@@ -0,0 +1,31 @@
+from pandas.core.indexers.utils import (
+    check_array_indexer,
+    check_key_length,
+    check_setitem_lengths,
+    disallow_ndim_indexing,
+    is_empty_indexer,
+    is_list_like_indexer,
+    is_scalar_indexer,
+    is_valid_positional_slice,
+    length_of_indexer,
+    maybe_convert_indices,
+    unpack_1tuple,
+    unpack_tuple_and_ellipses,
+    validate_indices,
+)
+
+__all__ = [
+    "check_array_indexer",
+    "check_key_length",
+    "check_setitem_lengths",
+    "disallow_ndim_indexing",
+    "is_empty_indexer",
+    "is_list_like_indexer",
+    "is_scalar_indexer",
+    "is_valid_positional_slice",
+    "length_of_indexer",
+    "maybe_convert_indices",
+    "unpack_1tuple",
+    "unpack_tuple_and_ellipses",
+    "validate_indices",
+]
diff --git a/pandas/pandas/core/indexers/objects.py b/pandas/pandas/core/indexers/objects.py
new file mode 100644
index 0000000000000..88379164534f2
--- /dev/null
+++ b/pandas/pandas/core/indexers/objects.py
@@ -0,0 +1,515 @@
+"""Indexer objects for computing start/end window bounds for rolling operations"""
+
+from __future__ import annotations
+
+from datetime import timedelta
+
+import numpy as np
+
+from pandas._libs.tslibs import BaseOffset
+from pandas._libs.window.indexers import calculate_variable_window_bounds
+from pandas.util._decorators import Appender
+
+from pandas.core.dtypes.common import ensure_platform_int
+
+from pandas.core.indexes.datetimes import DatetimeIndex
+
+from pandas.tseries.offsets import Nano
+
+get_window_bounds_doc = """
+Computes the bounds of a window.
+
+Parameters
+----------
+num_values : int, default 0
+    number of values that will be aggregated over
+window_size : int, default 0
+    the number of rows in a window
+min_periods : int, default None
+    min_periods passed from the top level rolling API
+center : bool, default None
+    center passed from the top level rolling API
+closed : str, default None
+    closed passed from the top level rolling API
+step : int, default None
+    step passed from the top level rolling API
+    .. versionadded:: 1.5
+win_type : str, default None
+    win_type passed from the top level rolling API
+
+Returns
+-------
+A tuple of ndarray[int64]s, indicating the boundaries of each
+window
+"""
+
+
+class BaseIndexer:
+    """
+    Base class for window bounds calculations.
+
+    Parameters
+    ----------
+    index_array : np.ndarray, default None
+        Array-like structure representing the indices for the data points.
+        If None, the default indices are assumed. This can be useful for
+        handling non-uniform indices in data, such as in time series
+        with irregular timestamps.
+    window_size : int, default 0
+        Size of the moving window. This is the number of observations used
+        for calculating the statistic. The default is to consider all
+        observations within the window.
+    **kwargs
+        Additional keyword arguments passed to the subclass's methods.
+
+    See Also
+    --------
+    DataFrame.rolling : Provides rolling window calculations on dataframe.
+    Series.rolling : Provides rolling window calculations on series.
+
+    Examples
+    --------
+    >>> from pandas.api.indexers import BaseIndexer
+    >>> class CustomIndexer(BaseIndexer):
+    ...     def get_window_bounds(self, num_values, min_periods, center, closed, step):
+    ...         start = np.arange(num_values, dtype=np.int64)
+    ...         end = np.arange(num_values, dtype=np.int64) + self.window_size
+    ...         return start, end
+    >>> df = pd.DataFrame({"values": range(5)})
+    >>> indexer = CustomIndexer(window_size=2)
+    >>> df.rolling(indexer).sum()
+        values
+    0	1.0
+    1	3.0
+    2	5.0
+    3	7.0
+    4	4.0
+    """
+
+    def __init__(
+        self, index_array: np.ndarray | None = None, window_size: int = 0, **kwargs
+    ) -> None:
+        self.index_array = index_array
+        self.window_size = window_size
+        # Set user defined kwargs as attributes that can be used in get_window_bounds
+        for key, value in kwargs.items():
+            setattr(self, key, value)
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        raise NotImplementedError
+
+
+class FixedWindowIndexer(BaseIndexer):
+    """Creates window boundaries that are of fixed length."""
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        if center or self.window_size == 0:
+            offset = (self.window_size - 1) // 2
+        else:
+            offset = 0
+
+        end = np.arange(1 + offset, num_values + 1 + offset, step, dtype="int64")
+        start = end - self.window_size
+        if closed in ["left", "both"]:
+            start -= 1
+        if closed in ["left", "neither"]:
+            end -= 1
+
+        end = np.clip(end, 0, num_values)
+        start = np.clip(start, 0, num_values)
+
+        return start, end
+
+
+class VariableWindowIndexer(BaseIndexer):
+    """Creates window boundaries that are of variable length, namely for time series."""
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        # error: Argument 4 to "calculate_variable_window_bounds" has incompatible
+        # type "Optional[bool]"; expected "bool"
+        # error: Argument 6 to "calculate_variable_window_bounds" has incompatible
+        # type "Optional[ndarray]"; expected "ndarray"
+        return calculate_variable_window_bounds(
+            num_values,
+            self.window_size,
+            min_periods,
+            center,  # type: ignore[arg-type]
+            closed,
+            self.index_array,  # type: ignore[arg-type]
+        )
+
+
+class VariableOffsetWindowIndexer(BaseIndexer):
+    """
+    Calculate window boundaries based on a non-fixed offset such as a BusinessDay.
+
+    Parameters
+    ----------
+    index_array : np.ndarray, default 0
+        Array-like structure specifying the indices for data points.
+        This parameter is currently not used.
+
+    window_size : int, optional, default 0
+        Specifies the number of data points in each window.
+        This parameter is currently not used.
+
+    index : DatetimeIndex, optional
+        ``DatetimeIndex`` of the labels of each observation.
+
+    offset : BaseOffset, optional
+        ``DateOffset`` representing the size of the window.
+
+    **kwargs
+        Additional keyword arguments passed to the parent class ``BaseIndexer``.
+
+    See Also
+    --------
+    api.indexers.BaseIndexer : Base class for all indexers.
+    DataFrame.rolling : Rolling window calculations on DataFrames.
+    offsets : Module providing various time offset classes.
+
+    Examples
+    --------
+    >>> from pandas.api.indexers import VariableOffsetWindowIndexer
+    >>> df = pd.DataFrame(range(10), index=pd.date_range("2020", periods=10))
+    >>> offset = pd.offsets.BDay(1)
+    >>> indexer = VariableOffsetWindowIndexer(index=df.index, offset=offset)
+    >>> df
+                0
+    2020-01-01  0
+    2020-01-02  1
+    2020-01-03  2
+    2020-01-04  3
+    2020-01-05  4
+    2020-01-06  5
+    2020-01-07  6
+    2020-01-08  7
+    2020-01-09  8
+    2020-01-10  9
+    >>> df.rolling(indexer).sum()
+                   0
+    2020-01-01   0.0
+    2020-01-02   1.0
+    2020-01-03   2.0
+    2020-01-04   3.0
+    2020-01-05   7.0
+    2020-01-06  12.0
+    2020-01-07   6.0
+    2020-01-08   7.0
+    2020-01-09   8.0
+    2020-01-10   9.0
+    """
+
+    def __init__(
+        self,
+        index_array: np.ndarray | None = None,
+        window_size: int = 0,
+        index: DatetimeIndex | None = None,
+        offset: BaseOffset | None = None,
+        **kwargs,
+    ) -> None:
+        super().__init__(index_array, window_size, **kwargs)
+        if not isinstance(index, DatetimeIndex):
+            raise ValueError("index must be a DatetimeIndex.")
+        self.index = index
+        if not isinstance(offset, BaseOffset):
+            raise ValueError("offset must be a DateOffset-like object.")
+        self.offset = offset
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        if step is not None:
+            raise NotImplementedError("step not implemented for variable offset window")
+        if num_values <= 0:
+            return np.empty(0, dtype="int64"), np.empty(0, dtype="int64")
+
+        # if windows is variable, default is 'right', otherwise default is 'both'
+        if closed is None:
+            closed = "right" if self.index is not None else "both"
+
+        right_closed = closed in ["right", "both"]
+        left_closed = closed in ["left", "both"]
+
+        if self.index[num_values - 1] < self.index[0]:
+            index_growth_sign = -1
+        else:
+            index_growth_sign = 1
+        offset_diff = index_growth_sign * self.offset
+
+        start = np.empty(num_values, dtype="int64")
+        start.fill(-1)
+        end = np.empty(num_values, dtype="int64")
+        end.fill(-1)
+
+        start[0] = 0
+
+        # right endpoint is closed
+        if right_closed:
+            end[0] = 1
+        # right endpoint is open
+        else:
+            end[0] = 0
+
+        zero = timedelta(0)
+        # start is start of slice interval (including)
+        # end is end of slice interval (not including)
+        for i in range(1, num_values):
+            end_bound = self.index[i]
+            start_bound = end_bound - offset_diff
+
+            # left endpoint is closed
+            if left_closed:
+                start_bound -= Nano(1)
+
+            # advance the start bound until we are
+            # within the constraint
+            start[i] = i
+            for j in range(start[i - 1], i):
+                start_diff = (self.index[j] - start_bound) * index_growth_sign
+                if start_diff > zero:
+                    start[i] = j
+                    break
+
+            # end bound is previous end
+            # or current index
+            end_diff = (self.index[end[i - 1]] - end_bound) * index_growth_sign
+            if end_diff == zero and not right_closed:
+                end[i] = end[i - 1] + 1
+            elif end_diff <= zero:
+                end[i] = i + 1
+            else:
+                end[i] = end[i - 1]
+
+            # right endpoint is open
+            if not right_closed:
+                end[i] -= 1
+
+        return start, end
+
+
+class ExpandingIndexer(BaseIndexer):
+    """Calculate expanding window bounds, mimicking df.expanding()"""
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        return (
+            np.zeros(num_values, dtype=np.int64),
+            np.arange(1, num_values + 1, dtype=np.int64),
+        )
+
+
+class FixedForwardWindowIndexer(BaseIndexer):
+    """
+    Creates window boundaries for fixed-length windows that include the current row.
+
+    Parameters
+    ----------
+    index_array : np.ndarray, default None
+        Array-like structure representing the indices for the data points.
+        If None, the default indices are assumed. This can be useful for
+        handling non-uniform indices in data, such as in time series
+        with irregular timestamps.
+    window_size : int, default 0
+        Size of the moving window. This is the number of observations used
+        for calculating the statistic. The default is to consider all
+        observations within the window.
+    **kwargs
+        Additional keyword arguments passed to the subclass's methods.
+
+    See Also
+    --------
+    DataFrame.rolling : Provides rolling window calculations.
+    api.indexers.VariableWindowIndexer : Calculate window bounds based on
+        variable-sized windows.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    >>> df
+         B
+    0  0.0
+    1  1.0
+    2  2.0
+    3  NaN
+    4  4.0
+
+    >>> indexer = pd.api.indexers.FixedForwardWindowIndexer(window_size=2)
+    >>> df.rolling(window=indexer, min_periods=1).sum()
+         B
+    0  1.0
+    1  3.0
+    2  2.0
+    3  4.0
+    4  4.0
+    """
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        if center:
+            raise ValueError("Forward-looking windows can't have center=True")
+        if closed is not None:
+            raise ValueError(
+                "Forward-looking windows don't support setting the closed argument"
+            )
+        if step is None:
+            step = 1
+
+        start = np.arange(0, num_values, step, dtype="int64")
+        end = start + self.window_size
+        if self.window_size:
+            end = np.clip(end, 0, num_values)
+
+        return start, end
+
+
+class GroupbyIndexer(BaseIndexer):
+    """Calculate bounds to compute groupby rolling, mimicking df.groupby().rolling()"""
+
+    def __init__(
+        self,
+        index_array: np.ndarray | None = None,
+        window_size: int | BaseIndexer = 0,
+        groupby_indices: dict | None = None,
+        window_indexer: type[BaseIndexer] = BaseIndexer,
+        indexer_kwargs: dict | None = None,
+        **kwargs,
+    ) -> None:
+        """
+        Parameters
+        ----------
+        index_array : np.ndarray or None
+            np.ndarray of the index of the original object that we are performing
+            a chained groupby operation over. This index has been pre-sorted relative to
+            the groups
+        window_size : int or BaseIndexer
+            window size during the windowing operation
+        groupby_indices : dict or None
+            dict of {group label: [positional index of rows belonging to the group]}
+        window_indexer : BaseIndexer
+            BaseIndexer class determining the start and end bounds of each group
+        indexer_kwargs : dict or None
+            Custom kwargs to be passed to window_indexer
+        **kwargs :
+            keyword arguments that will be available when get_window_bounds is called
+        """
+        self.groupby_indices = groupby_indices or {}
+        self.window_indexer = window_indexer
+        self.indexer_kwargs = indexer_kwargs.copy() if indexer_kwargs else {}
+        super().__init__(
+            index_array=index_array,
+            window_size=self.indexer_kwargs.pop("window_size", window_size),
+            **kwargs,
+        )
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        # 1) For each group, get the indices that belong to the group
+        # 2) Use the indices to calculate the start & end bounds of the window
+        # 3) Append the window bounds in group order
+        start_arrays = []
+        end_arrays = []
+        window_indices_start = 0
+        for indices in self.groupby_indices.values():
+            index_array: np.ndarray | None
+
+            if self.index_array is not None:
+                index_array = self.index_array.take(ensure_platform_int(indices))
+            else:
+                index_array = self.index_array
+            indexer = self.window_indexer(
+                index_array=index_array,
+                window_size=self.window_size,
+                **self.indexer_kwargs,
+            )
+            start, end = indexer.get_window_bounds(
+                len(indices), min_periods, center, closed, step
+            )
+            start = start.astype(np.int64)
+            end = end.astype(np.int64)
+            assert len(start) == len(end), (
+                "these should be equal in length from get_window_bounds"
+            )
+            # Cannot use groupby_indices as they might not be monotonic with the object
+            # we're rolling over
+            window_indices = np.arange(
+                window_indices_start, window_indices_start + len(indices)
+            )
+            window_indices_start += len(indices)
+            # Extend as we'll be slicing window like [start, end)
+            window_indices = np.append(window_indices, [window_indices[-1] + 1]).astype(
+                np.int64, copy=False
+            )
+            start_arrays.append(window_indices.take(ensure_platform_int(start)))
+            end_arrays.append(window_indices.take(ensure_platform_int(end)))
+        if len(start_arrays) == 0:
+            return np.array([], dtype=np.int64), np.array([], dtype=np.int64)
+        start = np.concatenate(start_arrays)
+        end = np.concatenate(end_arrays)
+        return start, end
+
+
+class ExponentialMovingWindowIndexer(BaseIndexer):
+    """Calculate ewm window bounds (the entire window)"""
+
+    @Appender(get_window_bounds_doc)
+    def get_window_bounds(
+        self,
+        num_values: int = 0,
+        min_periods: int | None = None,
+        center: bool | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]:
+        return np.array([0], dtype=np.int64), np.array([num_values], dtype=np.int64)
diff --git a/pandas/pandas/core/indexers/utils.py b/pandas/pandas/core/indexers/utils.py
new file mode 100644
index 0000000000000..b089be3469d87
--- /dev/null
+++ b/pandas/pandas/core/indexers/utils.py
@@ -0,0 +1,554 @@
+"""
+Low-dependency indexing utilities.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.common import (
+    is_array_like,
+    is_bool_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import AnyArrayLike
+
+    from pandas.core.frame import DataFrame
+    from pandas.core.indexes.base import Index
+
+# -----------------------------------------------------------
+# Indexer Identification
+
+
+def is_valid_positional_slice(slc: slice) -> bool:
+    """
+    Check if a slice object can be interpreted as a positional indexer.
+
+    Parameters
+    ----------
+    slc : slice
+
+    Returns
+    -------
+    bool
+
+    Notes
+    -----
+    A valid positional slice may also be interpreted as a label-based slice
+    depending on the index being sliced.
+    """
+    return (
+        lib.is_int_or_none(slc.start)
+        and lib.is_int_or_none(slc.stop)
+        and lib.is_int_or_none(slc.step)
+    )
+
+
+def is_list_like_indexer(key) -> bool:
+    """
+    Check if we have a list-like indexer that is *not* a NamedTuple.
+
+    Parameters
+    ----------
+    key : object
+
+    Returns
+    -------
+    bool
+    """
+    # allow a list_like, but exclude NamedTuples which can be indexers
+    return is_list_like(key) and not (isinstance(key, tuple) and type(key) is not tuple)
+
+
+def is_scalar_indexer(indexer, ndim: int) -> bool:
+    """
+    Return True if we are all scalar indexers.
+
+    Parameters
+    ----------
+    indexer : object
+    ndim : int
+        Number of dimensions in the object being indexed.
+
+    Returns
+    -------
+    bool
+    """
+    if ndim == 1 and is_integer(indexer):
+        # GH37748: allow indexer to be an integer for Series
+        return True
+    if isinstance(indexer, tuple) and len(indexer) == ndim:
+        return all(is_integer(x) for x in indexer)
+    return False
+
+
+def is_empty_indexer(indexer) -> bool:
+    """
+    Check if we have an empty indexer.
+
+    Parameters
+    ----------
+    indexer : object
+
+    Returns
+    -------
+    bool
+    """
+    if is_list_like(indexer) and not len(indexer):
+        return True
+    if not isinstance(indexer, tuple):
+        indexer = (indexer,)
+    return any(isinstance(idx, np.ndarray) and len(idx) == 0 for idx in indexer)
+
+
+# -----------------------------------------------------------
+# Indexer Validation
+
+
+def check_setitem_lengths(indexer, value, values) -> bool:
+    """
+    Validate that value and indexer are the same length.
+
+    An special-case is allowed for when the indexer is a boolean array
+    and the number of true values equals the length of ``value``. In
+    this case, no exception is raised.
+
+    Parameters
+    ----------
+    indexer : sequence
+        Key for the setitem.
+    value : array-like
+        Value for the setitem.
+    values : array-like
+        Values being set into.
+
+    Returns
+    -------
+    bool
+        Whether this is an empty listlike setting which is a no-op.
+
+    Raises
+    ------
+    ValueError
+        When the indexer is an ndarray or list and the lengths don't match.
+    """
+    no_op = False
+
+    if isinstance(indexer, (np.ndarray, list)):
+        # We can ignore other listlikes because they are either
+        #  a) not necessarily 1-D indexers, e.g. tuple
+        #  b) boolean indexers e.g. BoolArray
+        if is_list_like(value):
+            if len(indexer) != len(value) and values.ndim == 1:
+                # boolean with truth values == len of the value is ok too
+                if isinstance(indexer, list):
+                    indexer = np.array(indexer)
+                if not (
+                    isinstance(indexer, np.ndarray)
+                    and indexer.dtype == np.bool_
+                    and indexer.sum() == len(value)
+                ):
+                    raise ValueError(
+                        "cannot set using a list-like indexer "
+                        "with a different length than the value"
+                    )
+            if not len(indexer):
+                no_op = True
+
+    elif isinstance(indexer, slice):
+        if is_list_like(value):
+            if len(value) != length_of_indexer(indexer, values) and values.ndim == 1:
+                # In case of two dimensional value is used row-wise and broadcasted
+                raise ValueError(
+                    "cannot set using a slice indexer with a "
+                    "different length than the value"
+                )
+            if not len(value):
+                no_op = True
+
+    return no_op
+
+
+def validate_indices(indices: np.ndarray, n: int) -> None:
+    """
+    Perform bounds-checking for an indexer.
+
+    -1 is allowed for indicating missing values.
+
+    Parameters
+    ----------
+    indices : ndarray
+    n : int
+        Length of the array being indexed.
+
+    Raises
+    ------
+    ValueError
+
+    Examples
+    --------
+    >>> validate_indices(np.array([1, 2]), 3)  # OK
+
+    >>> validate_indices(np.array([1, -2]), 3)
+    Traceback (most recent call last):
+        ...
+    ValueError: negative dimensions are not allowed
+
+    >>> validate_indices(np.array([1, 2, 3]), 3)
+    Traceback (most recent call last):
+        ...
+    IndexError: indices are out-of-bounds
+
+    >>> validate_indices(np.array([-1, -1]), 0)  # OK
+
+    >>> validate_indices(np.array([0, 1]), 0)
+    Traceback (most recent call last):
+        ...
+    IndexError: indices are out-of-bounds
+    """
+    if len(indices):
+        min_idx = indices.min()
+        if min_idx < -1:
+            msg = f"'indices' contains values less than allowed ({min_idx} < -1)"
+            raise ValueError(msg)
+
+        max_idx = indices.max()
+        if max_idx >= n:
+            raise IndexError("indices are out-of-bounds")
+
+
+# -----------------------------------------------------------
+# Indexer Conversion
+
+
+def maybe_convert_indices(indices, n: int, verify: bool = True) -> np.ndarray:
+    """
+    Attempt to convert indices into valid, positive indices.
+
+    If we have negative indices, translate to positive here.
+    If we have indices that are out-of-bounds, raise an IndexError.
+
+    Parameters
+    ----------
+    indices : array-like
+        Array of indices that we are to convert.
+    n : int
+        Number of elements in the array that we are indexing.
+    verify : bool, default True
+        Check that all entries are between 0 and n - 1, inclusive.
+
+    Returns
+    -------
+    array-like
+        An array-like of positive indices that correspond to the ones
+        that were passed in initially to this function.
+
+    Raises
+    ------
+    IndexError
+        One of the converted indices either exceeded the number of,
+        elements (specified by `n`), or was still negative.
+    """
+    if isinstance(indices, list):
+        indices = np.array(indices)
+        if len(indices) == 0:
+            # If `indices` is empty, np.array will return a float,
+            # and will cause indexing errors.
+            return np.empty(0, dtype=np.intp)
+
+    mask = indices < 0
+    if mask.any():
+        indices = indices.copy()
+        indices[mask] += n
+
+    if verify:
+        mask = (indices >= n) | (indices < 0)
+        if mask.any():
+            raise IndexError("indices are out-of-bounds")
+    return indices
+
+
+# -----------------------------------------------------------
+# Unsorted
+
+
+def length_of_indexer(indexer, target=None) -> int:
+    """
+    Return the expected length of target[indexer]
+
+    Returns
+    -------
+    int
+    """
+    if target is not None and isinstance(indexer, slice):
+        target_len = len(target)
+        start = indexer.start
+        stop = indexer.stop
+        step = indexer.step
+        if start is None:
+            start = 0
+        elif start < 0:
+            start += target_len
+        if stop is None or stop > target_len:
+            stop = target_len
+        elif stop < 0:
+            stop += target_len
+        if step is None:
+            step = 1
+        elif step < 0:
+            start, stop = stop + 1, start + 1
+            step = -step
+        return (stop - start + step - 1) // step
+    elif isinstance(indexer, (ABCSeries, ABCIndex, np.ndarray, list)):
+        if isinstance(indexer, list):
+            indexer = np.array(indexer)
+
+        if indexer.dtype == bool:
+            # GH#25774
+            return indexer.sum()
+        return len(indexer)
+    elif isinstance(indexer, range):
+        return (indexer.stop - indexer.start) // indexer.step
+    elif not is_list_like_indexer(indexer):
+        return 1
+    raise AssertionError("cannot find the length of the indexer")
+
+
+def disallow_ndim_indexing(result) -> None:
+    """
+    Helper function to disallow multi-dimensional indexing on 1D Series/Index.
+
+    GH#27125 indexer like idx[:, None] expands dim, but we cannot do that
+    and keep an index, so we used to return ndarray, which was deprecated
+    in GH#30588.
+    """
+    if np.ndim(result) > 1:
+        raise ValueError(
+            "Multi-dimensional indexing (e.g. `obj[:, None]`) is no longer "
+            "supported. Convert to a numpy array before indexing instead."
+        )
+
+
+def unpack_1tuple(tup):
+    """
+    If we have a length-1 tuple/list that contains a slice, unpack to just
+    the slice.
+
+    Notes
+    -----
+    The list case is deprecated.
+    """
+    if len(tup) == 1 and isinstance(tup[0], slice):
+        # if we don't have a MultiIndex, we may still be able to handle
+        #  a 1-tuple.  see test_1tuple_without_multiindex
+
+        if isinstance(tup, list):
+            # GH#31299
+            raise ValueError(
+                "Indexing with a single-item list containing a "
+                "slice is not allowed. Pass a tuple instead.",
+            )
+
+        return tup[0]
+    return tup
+
+
+def check_key_length(columns: Index, key, value: DataFrame) -> None:
+    """
+    Checks if a key used as indexer has the same length as the columns it is
+    associated with.
+
+    Parameters
+    ----------
+    columns : Index The columns of the DataFrame to index.
+    key : A list-like of keys to index with.
+    value : DataFrame The value to set for the keys.
+
+    Raises
+    ------
+    ValueError: If the length of key is not equal to the number of columns in value
+                or if the number of columns referenced by key is not equal to number
+                of columns.
+    """
+    if columns.is_unique:
+        if len(value.columns) != len(key):
+            raise ValueError("Columns must be same length as key")
+    else:
+        # Missing keys in columns are represented as -1
+        if len(columns.get_indexer_non_unique(key)[0]) != len(value.columns):
+            raise ValueError("Columns must be same length as key")
+
+
+def unpack_tuple_and_ellipses(item: tuple):
+    """
+    Possibly unpack arr[..., n] to arr[n]
+    """
+    if len(item) > 1:
+        # Note: we are assuming this indexing is being done on a 1D arraylike
+        if item[0] is Ellipsis:
+            item = item[1:]
+        elif item[-1] is Ellipsis:
+            item = item[:-1]
+
+    if len(item) > 1:
+        raise IndexError("too many indices for array.")
+
+    item = item[0]
+    return item
+
+
+# -----------------------------------------------------------
+# Public indexer validation
+
+
+def check_array_indexer(array: AnyArrayLike, indexer: Any) -> Any:
+    """
+    Check if `indexer` is a valid array indexer for `array`.
+
+    For a boolean mask, `array` and `indexer` are checked to have the same
+    length. The dtype is validated, and if it is an integer or boolean
+    ExtensionArray, it is checked if there are missing values present, and
+    it is converted to the appropriate numpy array. Other dtypes will raise
+    an error.
+
+    Non-array indexers (integer, slice, Ellipsis, tuples, ..) are passed
+    through as is.
+
+    Parameters
+    ----------
+    array : array-like
+        The array that is being indexed (only used for the length).
+    indexer : array-like or list-like
+        The array-like that's used to index. List-like input that is not yet
+        a numpy array or an ExtensionArray is converted to one. Other input
+        types are passed through as is.
+
+    Returns
+    -------
+    numpy.ndarray
+        The validated indexer as a numpy array that can be used to index.
+
+    Raises
+    ------
+    IndexError
+        When the lengths don't match.
+    ValueError
+        When `indexer` cannot be converted to a numpy ndarray to index
+        (e.g. presence of missing values).
+
+    See Also
+    --------
+    api.types.is_bool_dtype : Check if `key` is of boolean dtype.
+
+    Examples
+    --------
+    When checking a boolean mask, a boolean ndarray is returned when the
+    arguments are all valid.
+
+    >>> mask = pd.array([True, False])
+    >>> arr = pd.array([1, 2])
+    >>> pd.api.indexers.check_array_indexer(arr, mask)
+    array([ True, False])
+
+    An IndexError is raised when the lengths don't match.
+
+    >>> mask = pd.array([True, False, True])
+    >>> pd.api.indexers.check_array_indexer(arr, mask)
+    Traceback (most recent call last):
+    ...
+    IndexError: Boolean index has wrong length: 3 instead of 2.
+
+    NA values in a boolean array are treated as False.
+
+    >>> mask = pd.array([True, pd.NA])
+    >>> pd.api.indexers.check_array_indexer(arr, mask)
+    array([ True, False])
+
+    A numpy boolean mask will get passed through (if the length is correct):
+
+    >>> mask = np.array([True, False])
+    >>> pd.api.indexers.check_array_indexer(arr, mask)
+    array([ True, False])
+
+    Similarly for integer indexers, an integer ndarray is returned when it is
+    a valid indexer, otherwise an error is  (for integer indexers, a matching
+    length is not required):
+
+    >>> indexer = pd.array([0, 2], dtype="Int64")
+    >>> arr = pd.array([1, 2, 3])
+    >>> pd.api.indexers.check_array_indexer(arr, indexer)
+    array([0, 2])
+
+    >>> indexer = pd.array([0, pd.NA], dtype="Int64")
+    >>> pd.api.indexers.check_array_indexer(arr, indexer)
+    Traceback (most recent call last):
+    ...
+    ValueError: Cannot index with an integer indexer containing NA values
+
+    For non-integer/boolean dtypes, an appropriate error is raised:
+
+    >>> indexer = np.array([0.0, 2.0], dtype="float64")
+    >>> pd.api.indexers.check_array_indexer(arr, indexer)
+    Traceback (most recent call last):
+    ...
+    IndexError: arrays used as indices must be of integer or boolean type
+    """
+    from pandas.core.construction import array as pd_array
+
+    # whatever is not an array-like is returned as-is (possible valid array
+    # indexers that are not array-like: integer, slice, Ellipsis, None)
+    # In this context, tuples are not considered as array-like, as they have
+    # a specific meaning in indexing (multi-dimensional indexing)
+    if is_list_like(indexer):
+        if isinstance(indexer, tuple):
+            return indexer
+    else:
+        return indexer
+
+    # convert list-likes to array
+    if not is_array_like(indexer):
+        indexer = pd_array(indexer)
+        if len(indexer) == 0:
+            # empty list is converted to float array by pd.array
+            indexer = np.array([], dtype=np.intp)
+
+    dtype = indexer.dtype
+    if is_bool_dtype(dtype):
+        if isinstance(dtype, ExtensionDtype):
+            indexer = indexer.to_numpy(dtype=bool, na_value=False)
+        else:
+            indexer = np.asarray(indexer, dtype=bool)
+
+        # GH26658
+        if len(indexer) != len(array):
+            raise IndexError(
+                f"Boolean index has wrong length: "
+                f"{len(indexer)} instead of {len(array)}"
+            )
+    elif is_integer_dtype(dtype):
+        try:
+            indexer = np.asarray(indexer, dtype=np.intp)
+        except ValueError as err:
+            raise ValueError(
+                "Cannot index with an integer indexer containing NA values"
+            ) from err
+    else:
+        raise IndexError("arrays used as indices must be of integer or boolean type")
+
+    return indexer
diff --git a/pandas/pandas/core/indexes/__init__.py b/pandas/pandas/core/indexes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/indexes/accessors.py b/pandas/pandas/core/indexes/accessors.py
new file mode 100644
index 0000000000000..c404323a1168c
--- /dev/null
+++ b/pandas/pandas/core/indexes/accessors.py
@@ -0,0 +1,697 @@
+"""
+datetimelike delegation
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    NoReturn,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_list_like,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import ABCSeries
+
+from pandas.core.accessor import (
+    PandasDelegate,
+    delegate_names,
+)
+from pandas.core.arrays import (
+    DatetimeArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.arrow.array import ArrowExtensionArray
+from pandas.core.base import (
+    NoNewAttributesMixin,
+    PandasObject,
+)
+from pandas.core.indexes.datetimes import DatetimeIndex
+from pandas.core.indexes.timedeltas import TimedeltaIndex
+
+if TYPE_CHECKING:
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+class Properties(PandasDelegate, PandasObject, NoNewAttributesMixin):
+    _hidden_attrs = PandasObject._hidden_attrs | {
+        "orig",
+        "name",
+    }
+
+    def __init__(self, data: Series, orig) -> None:
+        if not isinstance(data, ABCSeries):
+            raise TypeError(
+                f"cannot convert an object of type {type(data)} to a datetimelike index"
+            )
+
+        self._parent = data
+        self.orig = orig
+        self.name = getattr(data, "name", None)
+        self._freeze()
+
+    def _get_values(self):
+        data = self._parent
+        if lib.is_np_dtype(data.dtype, "M"):
+            return DatetimeIndex(data, copy=False, name=self.name)
+
+        elif isinstance(data.dtype, DatetimeTZDtype):
+            return DatetimeIndex(data, copy=False, name=self.name)
+
+        elif lib.is_np_dtype(data.dtype, "m"):
+            return TimedeltaIndex(data, copy=False, name=self.name)
+
+        elif isinstance(data.dtype, PeriodDtype):
+            return PeriodArray(data, copy=False)
+
+        raise TypeError(
+            f"cannot convert an object of type {type(data)} to a datetimelike index"
+        )
+
+    def _delegate_property_get(self, name: str):
+        from pandas import Series
+
+        values = self._get_values()
+
+        result = getattr(values, name)
+
+        # maybe need to upcast (ints)
+        if isinstance(result, np.ndarray):
+            if is_integer_dtype(result):
+                result = result.astype("int64")
+        elif not is_list_like(result):
+            return result
+
+        result = np.asarray(result)
+
+        if self.orig is not None:
+            index = self.orig.index
+        else:
+            index = self._parent.index
+        # return the result as a Series
+        return Series(result, index=index, name=self.name).__finalize__(self._parent)
+
+    def _delegate_property_set(self, name: str, value, *args, **kwargs) -> NoReturn:
+        raise ValueError(
+            "modifications to a property of a datetimelike object are not supported. "
+            "Change values on the original."
+        )
+
+    def _delegate_method(self, name: str, *args, **kwargs):
+        from pandas import Series
+
+        values = self._get_values()
+
+        method = getattr(values, name)
+        result = method(*args, **kwargs)
+
+        if not is_list_like(result):
+            return result
+
+        return Series(result, index=self._parent.index, name=self.name).__finalize__(
+            self._parent
+        )
+
+
+@delegate_names(
+    delegate=ArrowExtensionArray,
+    accessors=TimedeltaArray._datetimelike_ops,
+    typ="property",
+    accessor_mapping=lambda x: f"_dt_{x}",
+    raise_on_missing=False,
+)
+@delegate_names(
+    delegate=ArrowExtensionArray,
+    accessors=TimedeltaArray._datetimelike_methods,
+    typ="method",
+    accessor_mapping=lambda x: f"_dt_{x}",
+    raise_on_missing=False,
+)
+@delegate_names(
+    delegate=ArrowExtensionArray,
+    accessors=DatetimeArray._datetimelike_ops,
+    typ="property",
+    accessor_mapping=lambda x: f"_dt_{x}",
+    raise_on_missing=False,
+)
+@delegate_names(
+    delegate=ArrowExtensionArray,
+    accessors=DatetimeArray._datetimelike_methods,
+    typ="method",
+    accessor_mapping=lambda x: f"_dt_{x}",
+    raise_on_missing=False,
+)
+class ArrowTemporalProperties(PandasDelegate, PandasObject, NoNewAttributesMixin):
+    def __init__(self, data: Series, orig) -> None:
+        if not isinstance(data, ABCSeries):
+            raise TypeError(
+                f"cannot convert an object of type {type(data)} to a datetimelike index"
+            )
+
+        self._parent = data
+        self._orig = orig
+        self._freeze()
+
+    def _delegate_property_get(self, name: str):
+        if not hasattr(self._parent.array, f"_dt_{name}"):
+            raise NotImplementedError(
+                f"dt.{name} is not supported for {self._parent.dtype}"
+            )
+        result = getattr(self._parent.array, f"_dt_{name}")
+
+        if not is_list_like(result):
+            return result
+
+        if self._orig is not None:
+            index = self._orig.index
+        else:
+            index = self._parent.index
+        # return the result as a Series, which is by definition a copy
+        result = type(self._parent)(
+            result, index=index, name=self._parent.name
+        ).__finalize__(self._parent)
+
+        return result
+
+    def _delegate_method(self, name: str, *args, **kwargs):
+        if not hasattr(self._parent.array, f"_dt_{name}"):
+            raise NotImplementedError(
+                f"dt.{name} is not supported for {self._parent.dtype}"
+            )
+
+        result = getattr(self._parent.array, f"_dt_{name}")(*args, **kwargs)
+
+        if self._orig is not None:
+            index = self._orig.index
+        else:
+            index = self._parent.index
+        # return the result as a Series, which is by definition a copy
+        result = type(self._parent)(
+            result, index=index, name=self._parent.name
+        ).__finalize__(self._parent)
+
+        return result
+
+    def to_pytimedelta(self):
+        # GH 57463
+        warnings.warn(
+            f"The behavior of {type(self).__name__}.to_pytimedelta is deprecated, "
+            "in a future version this will return a Series containing python "
+            "datetime.timedelta objects instead of an ndarray. To retain the "
+            "old behavior, call `np.array` on the result",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        return cast(ArrowExtensionArray, self._parent.array)._dt_to_pytimedelta()
+
+    def to_pydatetime(self) -> Series:
+        # GH#20306
+        return cast(ArrowExtensionArray, self._parent.array)._dt_to_pydatetime()
+
+    def isocalendar(self) -> DataFrame:
+        from pandas import DataFrame
+
+        result = (
+            cast(ArrowExtensionArray, self._parent.array)
+            ._dt_isocalendar()
+            ._pa_array.combine_chunks()
+        )
+        iso_calendar_df = DataFrame(
+            {
+                col: type(self._parent.array)(result.field(i))  # type: ignore[call-arg]
+                for i, col in enumerate(["year", "week", "day"])
+            }
+        )
+        return iso_calendar_df
+
+    @property
+    def components(self) -> DataFrame:
+        from pandas import DataFrame
+
+        components_df = DataFrame(
+            {
+                col: getattr(self._parent.array, f"_dt_{col}")
+                for col in [
+                    "days",
+                    "hours",
+                    "minutes",
+                    "seconds",
+                    "milliseconds",
+                    "microseconds",
+                    "nanoseconds",
+                ]
+            }
+        )
+        return components_df
+
+
+@delegate_names(
+    delegate=DatetimeArray,
+    accessors=DatetimeArray._datetimelike_ops + ["unit"],
+    typ="property",
+)
+@delegate_names(
+    delegate=DatetimeArray,
+    accessors=DatetimeArray._datetimelike_methods + ["as_unit"],
+    typ="method",
+)
+class DatetimeProperties(Properties):
+    """
+    Accessor object for datetimelike properties of the Series values.
+
+    Examples
+    --------
+    >>> seconds_series = pd.Series(pd.date_range("2000-01-01", periods=3, freq="s"))
+    >>> seconds_series
+    0   2000-01-01 00:00:00
+    1   2000-01-01 00:00:01
+    2   2000-01-01 00:00:02
+    dtype: datetime64[ns]
+    >>> seconds_series.dt.second
+    0    0
+    1    1
+    2    2
+    dtype: int32
+
+    >>> hours_series = pd.Series(pd.date_range("2000-01-01", periods=3, freq="h"))
+    >>> hours_series
+    0   2000-01-01 00:00:00
+    1   2000-01-01 01:00:00
+    2   2000-01-01 02:00:00
+    dtype: datetime64[ns]
+    >>> hours_series.dt.hour
+    0    0
+    1    1
+    2    2
+    dtype: int32
+
+    >>> quarters_series = pd.Series(pd.date_range("2000-01-01", periods=3, freq="QE"))
+    >>> quarters_series
+    0   2000-03-31
+    1   2000-06-30
+    2   2000-09-30
+    dtype: datetime64[ns]
+    >>> quarters_series.dt.quarter
+    0    1
+    1    2
+    2    3
+    dtype: int32
+
+    Returns a Series indexed like the original Series.
+    Raises TypeError if the Series does not contain datetimelike values.
+    """
+
+    def to_pydatetime(self) -> Series:
+        """
+        Return the data as a Series of :class:`datetime.datetime` objects.
+
+        Timezone information is retained if present.
+
+        .. warning::
+
+           Python's datetime uses microsecond resolution, which is lower than
+           pandas (nanosecond). The values are truncated.
+
+        Returns
+        -------
+        numpy.ndarray
+            Object dtype array containing native Python datetime objects.
+
+        See Also
+        --------
+        datetime.datetime : Standard library value for a datetime.
+
+        Examples
+        --------
+        >>> s = pd.Series(pd.date_range("20180310", periods=2))
+        >>> s
+        0   2018-03-10
+        1   2018-03-11
+        dtype: datetime64[ns]
+
+        >>> s.dt.to_pydatetime()
+        0    2018-03-10 00:00:00
+        1    2018-03-11 00:00:00
+        dtype: object
+
+        pandas' nanosecond precision is truncated to microseconds.
+
+        >>> s = pd.Series(pd.date_range("20180310", periods=2, freq="ns"))
+        >>> s
+        0   2018-03-10 00:00:00.000000000
+        1   2018-03-10 00:00:00.000000001
+        dtype: datetime64[ns]
+
+        >>> s.dt.to_pydatetime()
+        0    2018-03-10 00:00:00
+        1    2018-03-10 00:00:00
+        dtype: object
+        """
+        # GH#20306
+        from pandas import Series
+
+        return Series(self._get_values().to_pydatetime(), dtype=object)
+
+    @property
+    def freq(self):
+        """
+        Tries to return a string representing a frequency generated by infer_freq.
+
+        Returns None if it can't autodetect the frequency.
+
+        See Also
+        --------
+        Series.dt.to_period : Cast to PeriodArray/PeriodIndex at a particular
+            frequency.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["2024-01-01", "2024-01-02", "2024-01-03", "2024-01-04"])
+        >>> ser = pd.to_datetime(ser)
+        >>> ser.dt.freq
+        'D'
+
+        >>> ser = pd.Series(["2022-01-01", "2024-01-01", "2026-01-01", "2028-01-01"])
+        >>> ser = pd.to_datetime(ser)
+        >>> ser.dt.freq
+        '2YS-JAN'
+        """
+        return self._get_values().inferred_freq
+
+    def isocalendar(self) -> DataFrame:
+        """
+        Calculate year, week, and day according to the ISO 8601 standard.
+
+        Returns
+        -------
+        DataFrame
+            With columns year, week and day.
+
+        See Also
+        --------
+        Timestamp.isocalendar : Function return a 3-tuple containing ISO year,
+            week number, and weekday for the given Timestamp object.
+        datetime.date.isocalendar : Return a named tuple object with
+            three components: year, week and weekday.
+
+        Examples
+        --------
+        >>> ser = pd.to_datetime(pd.Series(["2010-01-01", pd.NaT]))
+        >>> ser.dt.isocalendar()
+           year  week  day
+        0  2009    53     5
+        1  <NA>  <NA>  <NA>
+        >>> ser.dt.isocalendar().week
+        0      53
+        1    <NA>
+        Name: week, dtype: UInt32
+        """
+        return self._get_values().isocalendar().set_index(self._parent.index)
+
+
+@delegate_names(
+    delegate=TimedeltaArray, accessors=TimedeltaArray._datetimelike_ops, typ="property"
+)
+@delegate_names(
+    delegate=TimedeltaArray,
+    accessors=TimedeltaArray._datetimelike_methods,
+    typ="method",
+)
+class TimedeltaProperties(Properties):
+    """
+    Accessor object for datetimelike properties of the Series values.
+
+    Returns a Series indexed like the original Series.
+    Raises TypeError if the Series does not contain datetimelike values.
+
+    Examples
+    --------
+    >>> seconds_series = pd.Series(
+    ...     pd.timedelta_range(start="1 second", periods=3, freq="s")
+    ... )
+    >>> seconds_series
+    0   0 days 00:00:01
+    1   0 days 00:00:02
+    2   0 days 00:00:03
+    dtype: timedelta64[ns]
+    >>> seconds_series.dt.seconds
+    0    1
+    1    2
+    2    3
+    dtype: int32
+    """
+
+    def to_pytimedelta(self) -> np.ndarray:
+        """
+        Return an array of native :class:`datetime.timedelta` objects.
+
+        Python's standard `datetime` library uses a different representation
+        timedelta's. This method converts a Series of pandas Timedeltas
+        to `datetime.timedelta` format with the same length as the original
+        Series.
+
+        Returns
+        -------
+        numpy.ndarray
+            Array of 1D containing data with `datetime.timedelta` type.
+
+        See Also
+        --------
+        datetime.timedelta : A duration expressing the difference
+            between two date, time, or datetime.
+
+        Examples
+        --------
+        >>> s = pd.Series(pd.to_timedelta(np.arange(5), unit="D"))
+        >>> s
+        0   0 days
+        1   1 days
+        2   2 days
+        3   3 days
+        4   4 days
+        dtype: timedelta64[ns]
+
+        >>> s.dt.to_pytimedelta()
+        array([datetime.timedelta(0), datetime.timedelta(days=1),
+        datetime.timedelta(days=2), datetime.timedelta(days=3),
+        datetime.timedelta(days=4)], dtype=object)
+        """
+        # GH 57463
+        warnings.warn(
+            f"The behavior of {type(self).__name__}.to_pytimedelta is deprecated, "
+            "in a future version this will return a Series containing python "
+            "datetime.timedelta objects instead of an ndarray. To retain the "
+            "old behavior, call `np.array` on the result",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        return self._get_values().to_pytimedelta()
+
+    @property
+    def components(self) -> DataFrame:
+        """
+        Return a Dataframe of the components of the Timedeltas.
+
+        Each row of the DataFrame corresponds to a Timedelta in the original
+        Series and contains the individual components (days, hours, minutes,
+        seconds, milliseconds, microseconds, nanoseconds) of the Timedelta.
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        TimedeltaIndex.components : Return a DataFrame of the individual resolution
+            components of the Timedeltas.
+        Series.dt.total_seconds : Return the total number of seconds in the duration.
+
+        Examples
+        --------
+        >>> s = pd.Series(pd.to_timedelta(np.arange(5), unit="s"))
+        >>> s
+        0   0 days 00:00:00
+        1   0 days 00:00:01
+        2   0 days 00:00:02
+        3   0 days 00:00:03
+        4   0 days 00:00:04
+        dtype: timedelta64[ns]
+        >>> s.dt.components
+           days  hours  minutes  seconds  milliseconds  microseconds  nanoseconds
+        0     0      0        0        0             0             0            0
+        1     0      0        0        1             0             0            0
+        2     0      0        0        2             0             0            0
+        3     0      0        0        3             0             0            0
+        4     0      0        0        4             0             0            0
+        """
+        return (
+            self._get_values()
+            .components.set_index(self._parent.index)
+            .__finalize__(self._parent)
+        )
+
+    @property
+    def freq(self):
+        return self._get_values().inferred_freq
+
+
+@delegate_names(
+    delegate=PeriodArray, accessors=PeriodArray._datetimelike_ops, typ="property"
+)
+@delegate_names(
+    delegate=PeriodArray, accessors=PeriodArray._datetimelike_methods, typ="method"
+)
+class PeriodProperties(Properties):
+    """
+    Accessor object for datetimelike properties of the Series values.
+
+    Returns a Series indexed like the original Series.
+    Raises TypeError if the Series does not contain datetimelike values.
+
+    Examples
+    --------
+    >>> seconds_series = pd.Series(
+    ...     pd.period_range(
+    ...         start="2000-01-01 00:00:00", end="2000-01-01 00:00:03", freq="s"
+    ...     )
+    ... )
+    >>> seconds_series
+    0    2000-01-01 00:00:00
+    1    2000-01-01 00:00:01
+    2    2000-01-01 00:00:02
+    3    2000-01-01 00:00:03
+    dtype: period[s]
+    >>> seconds_series.dt.second
+    0    0
+    1    1
+    2    2
+    3    3
+    dtype: int64
+
+    >>> hours_series = pd.Series(
+    ...     pd.period_range(start="2000-01-01 00:00", end="2000-01-01 03:00", freq="h")
+    ... )
+    >>> hours_series
+    0    2000-01-01 00:00
+    1    2000-01-01 01:00
+    2    2000-01-01 02:00
+    3    2000-01-01 03:00
+    dtype: period[h]
+    >>> hours_series.dt.hour
+    0    0
+    1    1
+    2    2
+    3    3
+    dtype: int64
+
+    >>> quarters_series = pd.Series(
+    ...     pd.period_range(start="2000-01-01", end="2000-12-31", freq="Q-DEC")
+    ... )
+    >>> quarters_series
+    0    2000Q1
+    1    2000Q2
+    2    2000Q3
+    3    2000Q4
+    dtype: period[Q-DEC]
+    >>> quarters_series.dt.quarter
+    0    1
+    1    2
+    2    3
+    3    4
+    dtype: int64
+    """
+
+
+class CombinedDatetimelikeProperties(
+    DatetimeProperties, TimedeltaProperties, PeriodProperties
+):
+    """
+    Accessor object for Series values' datetime-like, timedelta and period properties.
+
+    See Also
+    --------
+    DatetimeIndex : Index of datetime64 data.
+
+    Examples
+    --------
+    >>> dates = pd.Series(
+    ...     ["2024-01-01", "2024-01-15", "2024-02-5"], dtype="datetime64[ns]"
+    ... )
+    >>> dates.dt.day
+    0     1
+    1    15
+    2     5
+    dtype: int32
+    >>> dates.dt.month
+    0    1
+    1    1
+    2    2
+    dtype: int32
+
+    >>> dates = pd.Series(
+    ...     ["2024-01-01", "2024-01-15", "2024-02-5"], dtype="datetime64[ns, UTC]"
+    ... )
+    >>> dates.dt.day
+    0     1
+    1    15
+    2     5
+    dtype: int32
+    >>> dates.dt.month
+    0    1
+    1    1
+    2    2
+    dtype: int32
+    """
+
+    def __new__(cls, data: Series):  # pyright: ignore[reportInconsistentConstructor]
+        # CombinedDatetimelikeProperties isn't really instantiated. Instead
+        # we need to choose which parent (datetime or timedelta) is
+        # appropriate. Since we're checking the dtypes anyway, we'll just
+        # do all the validation here.
+
+        if not isinstance(data, ABCSeries):
+            raise TypeError(
+                f"cannot convert an object of type {type(data)} to a datetimelike index"
+            )
+
+        orig = data if isinstance(data.dtype, CategoricalDtype) else None
+        if orig is not None:
+            data = data._constructor(
+                orig.array,
+                name=orig.name,
+                copy=False,
+                dtype=orig._values.categories.dtype,
+                index=orig.index,
+            )
+
+        if isinstance(data.dtype, ArrowDtype) and data.dtype.kind in "Mm":
+            return ArrowTemporalProperties(data, orig)
+        if lib.is_np_dtype(data.dtype, "M"):
+            return DatetimeProperties(data, orig)
+        elif isinstance(data.dtype, DatetimeTZDtype):
+            return DatetimeProperties(data, orig)
+        elif lib.is_np_dtype(data.dtype, "m"):
+            return TimedeltaProperties(data, orig)
+        elif isinstance(data.dtype, PeriodDtype):
+            return PeriodProperties(data, orig)
+
+        raise AttributeError("Can only use .dt accessor with datetimelike values")
diff --git a/pandas/pandas/core/indexes/api.py b/pandas/pandas/core/indexes/api.py
new file mode 100644
index 0000000000000..058e584336905
--- /dev/null
+++ b/pandas/pandas/core/indexes/api.py
@@ -0,0 +1,338 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    lib,
+)
+from pandas.errors import InvalidIndexError
+
+from pandas.core.dtypes.cast import find_common_type
+
+from pandas.core.algorithms import safe_sort
+from pandas.core.indexes.base import (
+    Index,
+    _new_Index,
+    ensure_index,
+    ensure_index_from_sequences,
+    get_unanimous_names,
+    maybe_sequence_to_range,
+)
+from pandas.core.indexes.category import CategoricalIndex
+from pandas.core.indexes.datetimes import DatetimeIndex
+from pandas.core.indexes.interval import IntervalIndex
+from pandas.core.indexes.multi import MultiIndex
+from pandas.core.indexes.period import PeriodIndex
+from pandas.core.indexes.range import RangeIndex
+from pandas.core.indexes.timedeltas import TimedeltaIndex
+
+if TYPE_CHECKING:
+    from pandas._typing import Axis
+
+
+__all__ = [
+    "CategoricalIndex",
+    "DatetimeIndex",
+    "Index",
+    "IntervalIndex",
+    "InvalidIndexError",
+    "MultiIndex",
+    "NaT",
+    "PeriodIndex",
+    "RangeIndex",
+    "TimedeltaIndex",
+    "_new_Index",
+    "all_indexes_same",
+    "default_index",
+    "ensure_index",
+    "ensure_index_from_sequences",
+    "get_objs_combined_axis",
+    "get_unanimous_names",
+    "maybe_sequence_to_range",
+    "safe_sort_index",
+    "union_indexes",
+]
+
+
+def get_objs_combined_axis(
+    objs,
+    intersect: bool = False,
+    axis: Axis = 0,
+    sort: bool = True,
+) -> Index:
+    """
+    Extract combined index: return intersection or union (depending on the
+    value of "intersect") of indexes on given axis, or None if all objects
+    lack indexes (e.g. they are numpy arrays).
+
+    Parameters
+    ----------
+    objs : list
+        Series or DataFrame objects, may be mix of the two.
+    intersect : bool, default False
+        If True, calculate the intersection between indexes. Otherwise,
+        calculate the union.
+    axis : {0 or 'index', 1 or 'outer'}, default 0
+        The axis to extract indexes from.
+    sort : bool, default True
+        Whether the result index should come out sorted or not.
+
+    Returns
+    -------
+    Index
+    """
+    obs_idxes = [obj._get_axis(axis) for obj in objs]
+    return _get_combined_index(obs_idxes, intersect=intersect, sort=sort)
+
+
+def _get_distinct_objs(objs: list[Index]) -> list[Index]:
+    """
+    Return a list with distinct elements of "objs" (different ids).
+    Preserves order.
+    """
+    ids: set[int] = set()
+    res = []
+    for obj in objs:
+        if id(obj) not in ids:
+            ids.add(id(obj))
+            res.append(obj)
+    return res
+
+
+def _get_combined_index(
+    indexes: list[Index],
+    intersect: bool = False,
+    sort: bool = False,
+) -> Index:
+    """
+    Return the union or intersection of indexes.
+
+    Parameters
+    ----------
+    indexes : list of Index or list objects
+        When intersect=True, do not accept list of lists.
+    intersect : bool, default False
+        If True, calculate the intersection between indexes. Otherwise,
+        calculate the union.
+    sort : bool, default False
+        Whether the result index should come out sorted or not.
+
+    Returns
+    -------
+    Index
+    """
+    # TODO: handle index names!
+    indexes = _get_distinct_objs(indexes)
+    if len(indexes) == 0:
+        index: Index = default_index(0)
+    elif len(indexes) == 1:
+        index = indexes[0]
+    elif intersect:
+        index = indexes[0]
+        for other in indexes[1:]:
+            index = index.intersection(other)
+    else:
+        index = union_indexes(indexes, sort=False)
+        index = ensure_index(index)
+
+    if sort:
+        index = safe_sort_index(index)
+    return index
+
+
+def safe_sort_index(index: Index) -> Index:
+    """
+    Returns the sorted index
+
+    We keep the dtypes and the name attributes.
+
+    Parameters
+    ----------
+    index : an Index
+
+    Returns
+    -------
+    Index
+    """
+    if index.is_monotonic_increasing:
+        return index
+
+    try:
+        array_sorted = safe_sort(index)
+    except TypeError:
+        pass
+    else:
+        if isinstance(array_sorted, Index):
+            return array_sorted
+
+        array_sorted = cast(np.ndarray, array_sorted)
+        if isinstance(index, MultiIndex):
+            index = MultiIndex.from_tuples(array_sorted, names=index.names)
+        else:
+            index = Index(array_sorted, name=index.name, dtype=index.dtype)
+
+    return index
+
+
+def union_indexes(indexes, sort: bool | None = True) -> Index:
+    """
+    Return the union of indexes.
+
+    The behavior of sort and names is not consistent.
+
+    Parameters
+    ----------
+    indexes : list of Index or list objects
+    sort : bool, default True
+        Whether the result index should come out sorted or not.
+
+    Returns
+    -------
+    Index
+    """
+    if len(indexes) == 0:
+        raise AssertionError("Must have at least 1 Index to union")
+    if len(indexes) == 1:
+        result = indexes[0]
+        if isinstance(result, list):
+            if not sort:
+                result = Index(result)
+            else:
+                result = Index(sorted(result))
+        return result
+
+    indexes, kind = _sanitize_and_check(indexes)
+
+    if kind == "special":
+        result = indexes[0]
+
+        num_dtis = 0
+        num_dti_tzs = 0
+        for idx in indexes:
+            if isinstance(idx, DatetimeIndex):
+                num_dtis += 1
+                if idx.tz is not None:
+                    num_dti_tzs += 1
+        if num_dti_tzs not in [0, num_dtis]:
+            # TODO: this behavior is not tested (so may not be desired),
+            #  but is kept in order to keep behavior the same when
+            #  deprecating union_many
+            # test_frame_from_dict_with_mixed_indexes
+            raise TypeError("Cannot join tz-naive with tz-aware DatetimeIndex")
+
+        if num_dtis == len(indexes):
+            sort = True
+            result = indexes[0]
+
+        elif num_dtis > 1:
+            # If we have mixed timezones, our casting behavior may depend on
+            #  the order of indexes, which we don't want.
+            sort = False
+
+            # TODO: what about Categorical[dt64]?
+            # test_frame_from_dict_with_mixed_indexes
+            indexes = [x.astype(object, copy=False) for x in indexes]
+            result = indexes[0]
+
+        for other in indexes[1:]:
+            result = result.union(other, sort=None if sort else False)
+        return result
+
+    elif kind == "array":
+        if not all_indexes_same(indexes):
+            dtype = find_common_type([idx.dtype for idx in indexes])
+            inds = [ind.astype(dtype, copy=False) for ind in indexes]
+            index = inds[0].unique()
+            other = inds[1].append(inds[2:])
+            diff = other[index.get_indexer_for(other) == -1]
+            if len(diff):
+                index = index.append(diff.unique())
+            if sort:
+                index = index.sort_values()
+        else:
+            index = indexes[0]
+
+        name = get_unanimous_names(*indexes)[0]
+        if name != index.name:
+            index = index.rename(name)
+        return index
+    elif kind == "list":
+        dtypes = [idx.dtype for idx in indexes if isinstance(idx, Index)]
+        if dtypes:
+            dtype = find_common_type(dtypes)
+        else:
+            dtype = None
+        all_lists = (idx.tolist() if isinstance(idx, Index) else idx for idx in indexes)
+        return Index(
+            lib.fast_unique_multiple_list_gen(all_lists, sort=bool(sort)),
+            dtype=dtype,
+        )
+    else:
+        raise ValueError(f"{kind=} must be 'special', 'array' or 'list'.")
+
+
+def _sanitize_and_check(indexes):
+    """
+    Verify the type of indexes and convert lists to Index.
+
+    Cases:
+
+    - [list, list, ...]: Return ([list, list, ...], 'list')
+    - [list, Index, ...]: Return _sanitize_and_check([Index, Index, ...])
+        Lists are sorted and converted to Index.
+    - [Index, Index, ...]: Return ([Index, Index, ...], TYPE)
+        TYPE = 'special' if at least one special type, 'array' otherwise.
+
+    Parameters
+    ----------
+    indexes : list of Index or list objects
+
+    Returns
+    -------
+    sanitized_indexes : list of Index or list objects
+    type : {'list', 'array', 'special'}
+    """
+    kinds = {type(index) for index in indexes}
+
+    if list in kinds:
+        if len(kinds) > 1:
+            indexes = [
+                Index(list(x)) if not isinstance(x, Index) else x for x in indexes
+            ]
+            kinds -= {list}
+        else:
+            return indexes, "list"
+
+    if len(kinds) > 1 or Index not in kinds:
+        return indexes, "special"
+    else:
+        return indexes, "array"
+
+
+def all_indexes_same(indexes) -> bool:
+    """
+    Determine if all indexes contain the same elements.
+
+    Parameters
+    ----------
+    indexes : iterable of Index objects
+
+    Returns
+    -------
+    bool
+        True if all indexes contain the same elements, False otherwise.
+    """
+    itr = iter(indexes)
+    first = next(itr)
+    return all(first.equals(index) for index in itr)
+
+
+def default_index(n: int) -> RangeIndex:
+    rng = range(n)
+    return RangeIndex._simple_new(rng, name=None)
diff --git a/pandas/pandas/core/indexes/base.py b/pandas/pandas/core/indexes/base.py
new file mode 100644
index 0000000000000..c17c8c1d9a172
--- /dev/null
+++ b/pandas/pandas/core/indexes/base.py
@@ -0,0 +1,7831 @@
+from __future__ import annotations
+
+from collections import abc
+from datetime import datetime
+import functools
+from itertools import zip_longest
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    ClassVar,
+    Literal,
+    NoReturn,
+    cast,
+    final,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import (
+    get_option,
+    using_string_dtype,
+)
+
+from pandas._libs import (
+    NaT,
+    algos as libalgos,
+    index as libindex,
+    lib,
+    writers,
+)
+from pandas._libs.internals import BlockValuesRefs
+import pandas._libs.join as libjoin
+from pandas._libs.lib import (
+    is_datetime_array,
+    no_default,
+)
+from pandas._libs.tslibs import (
+    IncompatibleFrequency,
+    OutOfBoundsDatetime,
+    Timestamp,
+    tz_compare,
+)
+from pandas._typing import (
+    AnyAll,
+    ArrayLike,
+    Axes,
+    Axis,
+    AxisInt,
+    DropKeep,
+    Dtype,
+    DtypeObj,
+    F,
+    IgnoreRaise,
+    IndexLabel,
+    IndexT,
+    JoinHow,
+    Level,
+    NaPosition,
+    ReindexMethod,
+    Self,
+    Shape,
+    SliceType,
+    npt,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    DuplicateLabelError,
+    InvalidIndexError,
+)
+from pandas.util._decorators import (
+    Appender,
+    cache_readonly,
+    doc,
+    set_module,
+)
+from pandas.util._exceptions import (
+    find_stack_level,
+    rewrite_exception,
+)
+
+from pandas.core.dtypes.astype import (
+    astype_array,
+    astype_is_view,
+)
+from pandas.core.dtypes.cast import (
+    LossySetitemError,
+    can_hold_element,
+    common_dtype_categorical_compat,
+    find_result_type,
+    infer_dtype_from,
+    maybe_cast_pointwise_result,
+    np_can_hold_element,
+)
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    ensure_object,
+    ensure_platform_int,
+    is_any_real_numeric_dtype,
+    is_bool_dtype,
+    is_ea_or_datetimelike_dtype,
+    is_float,
+    is_hashable,
+    is_integer,
+    is_iterator,
+    is_list_like,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_scalar,
+    is_signed_integer_dtype,
+    is_string_dtype,
+    needs_i8_conversion,
+    pandas_dtype,
+    validate_all_hashable,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    PeriodDtype,
+    SparseDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCCategoricalIndex,
+    ABCDataFrame,
+    ABCDatetimeIndex,
+    ABCIntervalIndex,
+    ABCMultiIndex,
+    ABCPeriodIndex,
+    ABCRangeIndex,
+    ABCSeries,
+    ABCTimedeltaIndex,
+)
+from pandas.core.dtypes.inference import is_dict_like
+from pandas.core.dtypes.missing import (
+    array_equivalent,
+    is_valid_na_for_dtype,
+    isna,
+)
+
+from pandas.core import (
+    arraylike,
+    nanops,
+    ops,
+)
+from pandas.core.accessor import Accessor
+import pandas.core.algorithms as algos
+from pandas.core.array_algos.putmask import (
+    setitem_datetimelike_compat,
+    validate_putmask,
+)
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BaseMaskedArray,
+    Categorical,
+    DatetimeArray,
+    ExtensionArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.string_ import (
+    StringArray,
+    StringDtype,
+)
+from pandas.core.base import (
+    IndexOpsMixin,
+    PandasObject,
+)
+import pandas.core.common as com
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+    sanitize_array,
+)
+from pandas.core.indexers import (
+    disallow_ndim_indexing,
+    is_valid_positional_slice,
+)
+from pandas.core.indexes.frozen import FrozenList
+from pandas.core.missing import clean_reindex_fill_method
+from pandas.core.ops import get_op_result_name
+from pandas.core.sorting import (
+    ensure_key_mapped,
+    get_group_index_sorter,
+    nargsort,
+)
+from pandas.core.strings.accessor import StringMethods
+
+from pandas.io.formats.printing import (
+    PrettyDict,
+    default_pprint,
+    format_object_summary,
+    pprint_thing,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterable,
+        Sequence,
+    )
+
+    from pandas import (
+        CategoricalIndex,
+        DataFrame,
+        MultiIndex,
+        Series,
+    )
+    from pandas.core.arrays import (
+        IntervalArray,
+        PeriodArray,
+    )
+
+__all__ = ["Index"]
+
+_unsortable_types = frozenset(("mixed", "mixed-integer"))
+
+_index_doc_kwargs: dict[str, str] = {
+    "klass": "Index",
+    "inplace": "",
+    "target_klass": "Index",
+    "raises_section": "",
+    "unique": "Index",
+    "duplicated": "np.ndarray",
+}
+_index_shared_docs: dict[str, str] = {}
+str_t = str
+
+_dtype_obj = np.dtype("object")
+
+_masked_engines = {
+    "Complex128": libindex.MaskedComplex128Engine,
+    "Complex64": libindex.MaskedComplex64Engine,
+    "Float64": libindex.MaskedFloat64Engine,
+    "Float32": libindex.MaskedFloat32Engine,
+    "UInt64": libindex.MaskedUInt64Engine,
+    "UInt32": libindex.MaskedUInt32Engine,
+    "UInt16": libindex.MaskedUInt16Engine,
+    "UInt8": libindex.MaskedUInt8Engine,
+    "Int64": libindex.MaskedInt64Engine,
+    "Int32": libindex.MaskedInt32Engine,
+    "Int16": libindex.MaskedInt16Engine,
+    "Int8": libindex.MaskedInt8Engine,
+    "boolean": libindex.MaskedBoolEngine,
+    "double[pyarrow]": libindex.MaskedFloat64Engine,
+    "float64[pyarrow]": libindex.MaskedFloat64Engine,
+    "float32[pyarrow]": libindex.MaskedFloat32Engine,
+    "float[pyarrow]": libindex.MaskedFloat32Engine,
+    "uint64[pyarrow]": libindex.MaskedUInt64Engine,
+    "uint32[pyarrow]": libindex.MaskedUInt32Engine,
+    "uint16[pyarrow]": libindex.MaskedUInt16Engine,
+    "uint8[pyarrow]": libindex.MaskedUInt8Engine,
+    "int64[pyarrow]": libindex.MaskedInt64Engine,
+    "int32[pyarrow]": libindex.MaskedInt32Engine,
+    "int16[pyarrow]": libindex.MaskedInt16Engine,
+    "int8[pyarrow]": libindex.MaskedInt8Engine,
+    "bool[pyarrow]": libindex.MaskedBoolEngine,
+}
+
+
+def _maybe_return_indexers(meth: F) -> F:
+    """
+    Decorator to simplify 'return_indexers' checks in Index.join.
+    """
+
+    @functools.wraps(meth)
+    def join(
+        self,
+        other: Index,
+        *,
+        how: JoinHow = "left",
+        level=None,
+        return_indexers: bool = False,
+        sort: bool = False,
+    ):
+        join_index, lidx, ridx = meth(self, other, how=how, level=level, sort=sort)
+        if not return_indexers:
+            return join_index
+
+        if lidx is not None:
+            lidx = ensure_platform_int(lidx)
+        if ridx is not None:
+            ridx = ensure_platform_int(ridx)
+        return join_index, lidx, ridx
+
+    return cast(F, join)
+
+
+def _new_Index(cls, d):
+    """
+    This is called upon unpickling, rather than the default which doesn't
+    have arguments and breaks __new__.
+    """
+    # required for backward compat, because PI can't be instantiated with
+    # ordinals through __new__ GH #13277
+    if issubclass(cls, ABCPeriodIndex):
+        from pandas.core.indexes.period import _new_PeriodIndex
+
+        return _new_PeriodIndex(cls, **d)
+
+    if issubclass(cls, ABCMultiIndex):
+        if "labels" in d and "codes" not in d:
+            # GH#23752 "labels" kwarg has been replaced with "codes"
+            d["codes"] = d.pop("labels")
+
+        # Since this was a valid MultiIndex at pickle-time, we don't need to
+        #  check validty at un-pickle time.
+        d["verify_integrity"] = False
+
+    elif "dtype" not in d and "data" in d:
+        # Prevent Index.__new__ from conducting inference;
+        #  "data" key not in RangeIndex
+        d["dtype"] = d["data"].dtype
+    return cls.__new__(cls, **d)
+
+
+@set_module("pandas")
+class Index(IndexOpsMixin, PandasObject):
+    """
+    Immutable sequence used for indexing and alignment.
+
+    The basic object storing axis labels for all pandas objects.
+
+    .. versionchanged:: 2.0.0
+
+       Index can hold all numpy numeric dtypes (except float16). Previously only
+       int64/uint64/float64 dtypes were accepted.
+
+    Parameters
+    ----------
+    data : array-like (1-dimensional)
+        An array-like structure containing the data for the index. This could be a
+        Python list, a NumPy array, or a pandas Series.
+    dtype : str, numpy.dtype, or ExtensionDtype, optional
+        Data type for the output Index. If not specified, this will be
+        inferred from `data`.
+        See the :ref:`user guide <basics.dtypes>` for more usages.
+    copy : bool, default False
+        Copy input data.
+    name : object
+        Name to be stored in the index.
+    tupleize_cols : bool (default: True)
+        When True, attempt to create a MultiIndex if possible.
+
+    See Also
+    --------
+    RangeIndex : Index implementing a monotonic integer range.
+    CategoricalIndex : Index of :class:`Categorical` s.
+    MultiIndex : A multi-level, or hierarchical Index.
+    IntervalIndex : An Index of :class:`Interval` s.
+    DatetimeIndex : Index of datetime64 data.
+    TimedeltaIndex : Index of timedelta64 data.
+    PeriodIndex : Index of Period data.
+
+    Notes
+    -----
+    An Index instance can **only** contain hashable objects.
+    An Index instance *can not* hold numpy float16 dtype.
+
+    Examples
+    --------
+    >>> pd.Index([1, 2, 3])
+    Index([1, 2, 3], dtype='int64')
+
+    >>> pd.Index(list("abc"))
+    Index(['a', 'b', 'c'], dtype='object')
+
+    >>> pd.Index([1, 2, 3], dtype="uint8")
+    Index([1, 2, 3], dtype='uint8')
+    """
+
+    # similar to __array_priority__, positions Index after Series and DataFrame
+    #  but before ExtensionArray.  Should NOT be overridden by subclasses.
+    __pandas_priority__ = 2000
+
+    # Cython methods; see github.com/cython/cython/issues/2647
+    #  for why we need to wrap these instead of making them class attributes
+    # Moreover, cython will choose the appropriate-dtyped sub-function
+    #  given the dtypes of the passed arguments
+
+    @final
+    def _left_indexer_unique(self, other: Self) -> npt.NDArray[np.intp]:
+        # Caller is responsible for ensuring other.dtype == self.dtype
+        sv = self._get_join_target()
+        ov = other._get_join_target()
+        # similar but not identical to ov.searchsorted(sv)
+        return libjoin.left_join_indexer_unique(sv, ov)
+
+    @final
+    def _left_indexer(
+        self, other: Self
+    ) -> tuple[ArrayLike, npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        # Caller is responsible for ensuring other.dtype == self.dtype
+        sv = self._get_join_target()
+        ov = other._get_join_target()
+        joined_ndarray, lidx, ridx = libjoin.left_join_indexer(sv, ov)
+        joined = self._from_join_target(joined_ndarray)
+        return joined, lidx, ridx
+
+    @final
+    def _inner_indexer(
+        self, other: Self
+    ) -> tuple[ArrayLike, npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        # Caller is responsible for ensuring other.dtype == self.dtype
+        sv = self._get_join_target()
+        ov = other._get_join_target()
+        joined_ndarray, lidx, ridx = libjoin.inner_join_indexer(sv, ov)
+        joined = self._from_join_target(joined_ndarray)
+        return joined, lidx, ridx
+
+    @final
+    def _outer_indexer(
+        self, other: Self
+    ) -> tuple[ArrayLike, npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        # Caller is responsible for ensuring other.dtype == self.dtype
+        sv = self._get_join_target()
+        ov = other._get_join_target()
+        joined_ndarray, lidx, ridx = libjoin.outer_join_indexer(sv, ov)
+        joined = self._from_join_target(joined_ndarray)
+        return joined, lidx, ridx
+
+    _typ: str = "index"
+    _data: ExtensionArray | np.ndarray
+    _data_cls: type[ExtensionArray] | tuple[type[np.ndarray], type[ExtensionArray]] = (
+        np.ndarray,
+        ExtensionArray,
+    )
+    _id: object | None = None
+    _name: Hashable = None
+    # MultiIndex.levels previously allowed setting the index name. We
+    # don't allow this anymore, and raise if it happens rather than
+    # failing silently.
+    _no_setting_name: bool = False
+    _comparables: list[str] = ["name"]
+    _attributes: list[str] = ["name"]
+
+    @cache_readonly
+    def _can_hold_strings(self) -> bool:
+        return not is_numeric_dtype(self.dtype)
+
+    _engine_types: dict[np.dtype | ExtensionDtype, type[libindex.IndexEngine]] = {
+        np.dtype(np.int8): libindex.Int8Engine,
+        np.dtype(np.int16): libindex.Int16Engine,
+        np.dtype(np.int32): libindex.Int32Engine,
+        np.dtype(np.int64): libindex.Int64Engine,
+        np.dtype(np.uint8): libindex.UInt8Engine,
+        np.dtype(np.uint16): libindex.UInt16Engine,
+        np.dtype(np.uint32): libindex.UInt32Engine,
+        np.dtype(np.uint64): libindex.UInt64Engine,
+        np.dtype(np.float32): libindex.Float32Engine,
+        np.dtype(np.float64): libindex.Float64Engine,
+        np.dtype(np.complex64): libindex.Complex64Engine,
+        np.dtype(np.complex128): libindex.Complex128Engine,
+    }
+
+    @property
+    def _engine_type(
+        self,
+    ) -> type[libindex.IndexEngine | libindex.ExtensionEngine]:
+        return self._engine_types.get(self.dtype, libindex.ObjectEngine)
+
+    # whether we support partial string indexing. Overridden
+    # in DatetimeIndex and PeriodIndex
+    _supports_partial_string_indexing = False
+
+    _accessors = {"str"}
+
+    str = Accessor("str", StringMethods)
+
+    _references = None
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data=None,
+        dtype=None,
+        copy: bool = False,
+        name=None,
+        tupleize_cols: bool = True,
+    ) -> Self:
+        from pandas.core.indexes.range import RangeIndex
+
+        name = maybe_extract_name(name, data, cls)
+
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        data_dtype = getattr(data, "dtype", None)
+
+        refs = None
+        if not copy and isinstance(data, (ABCSeries, Index)):
+            refs = data._references
+
+        # range
+        if isinstance(data, (range, RangeIndex)):
+            result = RangeIndex(start=data, copy=copy, name=name)
+            if dtype is not None:
+                return result.astype(dtype, copy=False)
+            # error: Incompatible return value type (got "MultiIndex",
+            # expected "Self")
+            return result  # type: ignore[return-value]
+
+        elif is_ea_or_datetimelike_dtype(dtype):
+            # non-EA dtype indexes have special casting logic, so we punt here
+            if isinstance(data, (set, frozenset)):
+                data = list(data)
+
+        elif is_ea_or_datetimelike_dtype(data_dtype):
+            pass
+
+        elif isinstance(data, (np.ndarray, ABCMultiIndex)):
+            if isinstance(data, ABCMultiIndex):
+                data = data._values
+
+            if data.dtype.kind not in "iufcbmM":
+                # GH#11836 we need to avoid having numpy coerce
+                # things that look like ints/floats to ints unless
+                # they are actually ints, e.g. '0' and 0.0
+                # should not be coerced
+                data = com.asarray_tuplesafe(data, dtype=_dtype_obj)
+        elif isinstance(data, (ABCSeries, Index)):
+            # GH 56244: Avoid potential inference on object types
+            pass
+        elif is_scalar(data):
+            raise cls._raise_scalar_data_error(data)
+        elif hasattr(data, "__array__"):
+            return cls(np.asarray(data), dtype=dtype, copy=copy, name=name)
+        elif not is_list_like(data) and not isinstance(data, memoryview):
+            # 2022-11-16 the memoryview check is only necessary on some CI
+            #  builds, not clear why
+            raise cls._raise_scalar_data_error(data)
+
+        else:
+            if tupleize_cols:
+                # GH21470: convert iterable to list before determining if empty
+                if is_iterator(data):
+                    data = list(data)
+
+                if data and all(isinstance(e, tuple) for e in data):
+                    # we must be all tuples, otherwise don't construct
+                    # 10697
+                    from pandas.core.indexes.multi import MultiIndex
+
+                    # error: Incompatible return value type (got "MultiIndex",
+                    # expected "Self")
+                    return MultiIndex.from_tuples(  # type: ignore[return-value]
+                        data, names=name
+                    )
+            # other iterable of some kind
+
+            if not isinstance(data, (list, tuple)):
+                # we allow set/frozenset, which Series/sanitize_array does not, so
+                #  cast to list here
+                data = list(data)
+            if len(data) == 0:
+                # unlike Series, we default to object dtype:
+                data = np.array(data, dtype=object)
+
+            if len(data) and isinstance(data[0], tuple):
+                # Ensure we get 1-D array of tuples instead of 2D array.
+                data = com.asarray_tuplesafe(data, dtype=_dtype_obj)
+
+        try:
+            arr = sanitize_array(data, None, dtype=dtype, copy=copy)
+        except ValueError as err:
+            if "index must be specified when data is not list-like" in str(err):
+                raise cls._raise_scalar_data_error(data) from err
+            if "Data must be 1-dimensional" in str(err):
+                raise ValueError("Index data must be 1-dimensional") from err
+            raise
+        arr = ensure_wrapped_if_datetimelike(arr)
+
+        klass = cls._dtype_to_subclass(arr.dtype)
+
+        arr = klass._ensure_array(arr, arr.dtype, copy=False)
+        return klass._simple_new(arr, name, refs=refs)
+
+    @classmethod
+    def _ensure_array(cls, data, dtype, copy: bool):
+        """
+        Ensure we have a valid array to pass to _simple_new.
+        """
+        if data.ndim > 1:
+            # GH#13601, GH#20285, GH#27125
+            raise ValueError("Index data must be 1-dimensional")
+        elif dtype == np.float16:
+            # float16 not supported (no indexing engine)
+            raise NotImplementedError("float16 indexes are not supported")
+
+        if copy:
+            # asarray_tuplesafe does not always copy underlying data,
+            #  so need to make sure that this happens
+            data = data.copy()
+        return data
+
+    @final
+    @classmethod
+    def _dtype_to_subclass(cls, dtype: DtypeObj):
+        # Delay import for perf. https://github.com/pandas-dev/pandas/pull/31423
+
+        if isinstance(dtype, ExtensionDtype):
+            return dtype.index_class
+
+        if dtype.kind == "M":
+            from pandas import DatetimeIndex
+
+            return DatetimeIndex
+
+        elif dtype.kind == "m":
+            from pandas import TimedeltaIndex
+
+            return TimedeltaIndex
+
+        elif dtype.kind == "O":
+            # NB: assuming away MultiIndex
+            return Index
+
+        elif issubclass(dtype.type, str) or is_numeric_dtype(dtype):
+            return Index
+
+        raise NotImplementedError(dtype)
+
+    # NOTE for new Index creation:
+
+    # - _simple_new: It returns new Index with the same type as the caller.
+    #   All metadata (such as name) must be provided by caller's responsibility.
+    #   Using _shallow_copy is recommended because it fills these metadata
+    #   otherwise specified.
+
+    # - _shallow_copy: It returns new Index with the same type (using
+    #   _simple_new), but fills caller's metadata otherwise specified. Passed
+    #   kwargs will overwrite corresponding metadata.
+
+    # See each method's docstring.
+
+    @classmethod
+    def _simple_new(
+        cls, values: ArrayLike, name: Hashable | None = None, refs=None
+    ) -> Self:
+        """
+        We require that we have a dtype compat for the values. If we are passed
+        a non-dtype compat, then coerce using the constructor.
+
+        Must be careful not to recurse.
+        """
+        assert isinstance(values, cls._data_cls), type(values)
+
+        result = object.__new__(cls)
+        result._data = values
+        result._name = name
+        result._cache = {}
+        result._reset_identity()
+        if refs is not None:
+            result._references = refs
+        else:
+            result._references = BlockValuesRefs()
+        result._references.add_index_reference(result)
+
+        return result
+
+    @classmethod
+    def _with_infer(cls, *args, **kwargs):
+        """
+        Constructor that uses the 1.0.x behavior inferring numeric dtypes
+        for ndarray[object] inputs.
+        """
+        result = cls(*args, **kwargs)
+
+        if result.dtype == _dtype_obj and not result._is_multi:
+            # error: Argument 1 to "maybe_convert_objects" has incompatible type
+            # "Union[ExtensionArray, ndarray[Any, Any]]"; expected
+            # "ndarray[Any, Any]"
+            values = lib.maybe_convert_objects(result._values)  # type: ignore[arg-type]
+            if values.dtype.kind in "iufb":
+                return Index(values, name=result.name)
+
+        return result
+
+    @cache_readonly
+    def _constructor(self) -> type[Self]:
+        return type(self)
+
+    @final
+    def _maybe_check_unique(self) -> None:
+        """
+        Check that an Index has no duplicates.
+
+        This is typically only called via
+        `NDFrame.flags.allows_duplicate_labels.setter` when it's set to
+        True (duplicates aren't allowed).
+
+        Raises
+        ------
+        DuplicateLabelError
+            When the index is not unique.
+        """
+        if not self.is_unique:
+            msg = """Index has duplicates."""
+            duplicates = self._format_duplicate_message()
+            msg += f"\n{duplicates}"
+
+            raise DuplicateLabelError(msg)
+
+    @final
+    def _format_duplicate_message(self) -> DataFrame:
+        """
+        Construct the DataFrame for a DuplicateLabelError.
+
+        This returns a DataFrame indicating the labels and positions
+        of duplicates in an index. This should only be called when it's
+        already known that duplicates are present.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["a", "b", "a"])
+        >>> idx._format_duplicate_message()
+            positions
+        label
+        a        [0, 2]
+        """
+        from pandas import Series
+
+        duplicates = self[self.duplicated(keep="first")].unique()
+        assert len(duplicates)
+
+        out = (
+            Series(np.arange(len(self)), copy=False)
+            .groupby(self, observed=False)
+            .agg(list)[duplicates]
+        )
+        if self._is_multi:
+            # test_format_duplicate_labels_message_multi
+            # error: "Type[Index]" has no attribute "from_tuples"  [attr-defined]
+            out.index = type(self).from_tuples(out.index)  # type: ignore[attr-defined]
+
+        if self.nlevels == 1:
+            out = out.rename_axis("label")
+        return out.to_frame(name="positions")
+
+    # --------------------------------------------------------------------
+    # Index Internals Methods
+
+    def _shallow_copy(self, values, name: Hashable = no_default) -> Self:
+        """
+        Create a new Index with the same class as the caller, don't copy the
+        data, use the same object attributes with passed in attributes taking
+        precedence.
+
+        *this is an internal non-public method*
+
+        Parameters
+        ----------
+        values : the values to create the new Index, optional
+        name : Label, defaults to self.name
+        """
+        name = self._name if name is no_default else name
+
+        return self._simple_new(values, name=name, refs=self._references)
+
+    def _view(self) -> Self:
+        """
+        fastpath to make a shallow copy, i.e. new object with same data.
+        """
+        result = self._simple_new(self._values, name=self._name, refs=self._references)
+
+        result._cache = self._cache
+        return result
+
+    @final
+    def _rename(self, name: Hashable) -> Self:
+        """
+        fastpath for rename if new name is already validated.
+        """
+        result = self._view()
+        result._name = name
+        return result
+
+    @final
+    def is_(self, other) -> bool:
+        """
+        More flexible, faster check like ``is`` but that works through views.
+
+        Note: this is *not* the same as ``Index.identical()``, which checks
+        that metadata is also the same.
+
+        Parameters
+        ----------
+        other : object
+            Other object to compare against.
+
+        Returns
+        -------
+        bool
+            True if both have same underlying data, False otherwise.
+
+        See Also
+        --------
+        Index.identical : Works like ``Index.is_`` but also checks metadata.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index(["1", "2", "3"])
+        >>> idx1.is_(idx1.view())
+        True
+
+        >>> idx1.is_(idx1.copy())
+        False
+        """
+        if self is other:
+            return True
+        elif not hasattr(other, "_id"):
+            return False
+        elif self._id is None or other._id is None:
+            return False
+        else:
+            return self._id is other._id
+
+    @final
+    def _reset_identity(self) -> None:
+        """
+        Initializes or resets ``_id`` attribute with new object.
+        """
+        self._id = object()
+
+    @final
+    def _cleanup(self) -> None:
+        if "_engine" in self._cache:
+            self._engine.clear_mapping()
+
+    @cache_readonly
+    def _engine(
+        self,
+    ) -> libindex.IndexEngine | libindex.ExtensionEngine | libindex.MaskedIndexEngine:
+        # For base class (object dtype) we get ObjectEngine
+        target_values = self._get_engine_target()
+
+        if isinstance(self._values, ArrowExtensionArray) and self.dtype.kind in "Mm":
+            import pyarrow as pa
+
+            pa_type = self._values._pa_array.type
+            if pa.types.is_timestamp(pa_type):
+                target_values = self._values._to_datetimearray()
+                return libindex.DatetimeEngine(target_values._ndarray)
+            elif pa.types.is_duration(pa_type):
+                target_values = self._values._to_timedeltaarray()
+                return libindex.TimedeltaEngine(target_values._ndarray)
+
+        if isinstance(target_values, ExtensionArray):
+            if isinstance(target_values, (BaseMaskedArray, ArrowExtensionArray)):
+                try:
+                    return _masked_engines[target_values.dtype.name](target_values)
+                except KeyError:
+                    # Not supported yet e.g. decimal
+                    pass
+            elif self._engine_type is libindex.ObjectEngine:
+                return libindex.ExtensionEngine(target_values)
+
+        target_values = cast(np.ndarray, target_values)
+        # to avoid a reference cycle, bind `target_values` to a local variable, so
+        # `self` is not passed into the lambda.
+        if target_values.dtype == bool:
+            return libindex.BoolEngine(target_values)
+        elif target_values.dtype == np.complex64:
+            return libindex.Complex64Engine(target_values)
+        elif target_values.dtype == np.complex128:
+            return libindex.Complex128Engine(target_values)
+        elif needs_i8_conversion(self.dtype):
+            # We need to keep M8/m8 dtype when initializing the Engine,
+            #  but don't want to change _get_engine_target bc it is used
+            #  elsewhere
+            # error: Item "ExtensionArray" of "Union[ExtensionArray,
+            # ndarray[Any, Any]]" has no attribute "_ndarray"  [union-attr]
+            target_values = self._data._ndarray  # type: ignore[union-attr]
+        elif is_string_dtype(self.dtype) and not is_object_dtype(self.dtype):
+            return libindex.StringObjectEngine(target_values, self.dtype.na_value)  # type: ignore[union-attr]
+
+        # error: Argument 1 to "ExtensionEngine" has incompatible type
+        # "ndarray[Any, Any]"; expected "ExtensionArray"
+        return self._engine_type(target_values)  # type: ignore[arg-type]
+
+    @final
+    @cache_readonly
+    def _dir_additions_for_owner(self) -> set[str_t]:
+        """
+        Add the string-like labels to the owner dataframe/series dir output.
+
+        If this is a MultiIndex, it's first level values are used.
+        """
+        return {
+            c
+            for c in self.unique(level=0)[: get_option("display.max_dir_items")]
+            if isinstance(c, str) and c.isidentifier()
+        }
+
+    # --------------------------------------------------------------------
+    # Array-Like Methods
+
+    # ndarray compat
+    def __len__(self) -> int:
+        """
+        Return the length of the Index.
+        """
+        return len(self._data)
+
+    def __array__(self, dtype=None, copy=None) -> np.ndarray:
+        """
+        The array interface, return my values.
+        """
+        if copy is None:
+            # Note, that the if branch exists for NumPy 1.x support
+            return np.asarray(self._data, dtype=dtype)
+
+        return np.array(self._data, dtype=dtype, copy=copy)
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str_t, *inputs, **kwargs):
+        if any(isinstance(other, (ABCSeries, ABCDataFrame)) for other in inputs):
+            return NotImplemented
+
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            return result
+
+        if "out" in kwargs:
+            # e.g. test_dti_isub_tdi
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                return result
+
+        new_inputs = [x if x is not self else x._values for x in inputs]
+        result = getattr(ufunc, method)(*new_inputs, **kwargs)
+        if ufunc.nout == 2:
+            # i.e. np.divmod, np.modf, np.frexp
+            return tuple(self.__array_wrap__(x) for x in result)
+        elif method == "reduce":
+            result = lib.item_from_zerodim(result)
+            return result
+        elif is_scalar(result):
+            # e.g. matmul
+            return result
+
+        if result.dtype == np.float16:
+            result = result.astype(np.float32)
+
+        return self.__array_wrap__(result)
+
+    @final
+    def __array_wrap__(self, result, context=None, return_scalar=False):
+        """
+        Gets called after a ufunc and other functions e.g. np.split.
+        """
+        result = lib.item_from_zerodim(result)
+        if (not isinstance(result, Index) and is_bool_dtype(result.dtype)) or np.ndim(
+            result
+        ) > 1:
+            # exclude Index to avoid warning from is_bool_dtype deprecation;
+            #  in the Index case it doesn't matter which path we go down.
+            # reached in plotting tests with e.g. np.nonzero(index)
+            return result
+
+        return Index(result, name=self.name)
+
+    @cache_readonly
+    def dtype(self) -> DtypeObj:
+        """
+        Return the dtype object of the underlying data.
+
+        See Also
+        --------
+        Index.inferred_type: Return a string of the type inferred from the values.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.dtype
+        dtype('int64')
+        """
+        return self._data.dtype
+
+    @final
+    def ravel(self, order: str_t = "C") -> Self:
+        """
+        Return a view on self.
+
+        Parameters
+        ----------
+        order : {'K', 'A', 'C', 'F'}, default 'C'
+            Specify the memory layout of the view. This parameter is not
+            implemented currently.
+
+        Returns
+        -------
+        Index
+            A view on self.
+
+        See Also
+        --------
+        numpy.ndarray.ravel : Return a flattened array.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3], index=["a", "b", "c"])
+        >>> s.index.ravel()
+        Index(['a', 'b', 'c'], dtype='object')
+        """
+        return self[:]
+
+    def view(self, cls=None):
+        """
+        Return a view of the Index with the specified dtype or a new Index instance.
+
+        This method returns a view of the calling Index object if no arguments are
+        provided. If a dtype is specified through the `cls` argument, it attempts
+        to return a view of the Index with the specified dtype. Note that viewing
+        the Index as a different dtype reinterprets the underlying data, which can
+        lead to unexpected results for non-numeric or incompatible dtype conversions.
+
+        Parameters
+        ----------
+        cls : data-type or ndarray sub-class, optional
+            Data-type descriptor of the returned view, e.g., float32 or int16.
+            Omitting it results in the view having the same data-type as `self`.
+            This argument can also be specified as an ndarray sub-class,
+            e.g., np.int64 or np.float32 which then specifies the type of
+            the returned object.
+
+        Returns
+        -------
+        Index or ndarray
+            A view of the Index. If `cls` is None, the returned object is an Index
+            view with the same dtype as the calling object. If a numeric `cls` is
+            specified an ndarray view with the new dtype is returned.
+
+        Raises
+        ------
+        ValueError
+            If attempting to change to a dtype in a way that is not compatible with
+            the original dtype's memory layout, for example, viewing an 'int64' Index
+            as 'str'.
+
+        See Also
+        --------
+        Index.copy : Returns a copy of the Index.
+        numpy.ndarray.view : Returns a new view of array with the same data.
+
+        Examples
+        --------
+        >>> idx = pd.Index([-1, 0, 1])
+        >>> idx.view()
+        Index([-1, 0, 1], dtype='int64')
+
+        >>> idx.view(np.uint64)
+        array([18446744073709551615,                    0,                    1],
+          dtype=uint64)
+
+        Viewing as 'int32' or 'float32' reinterprets the memory, which may lead to
+        unexpected behavior:
+
+        >>> idx.view("float32")
+        array([   nan,    nan, 0.e+00, 0.e+00, 1.e-45, 0.e+00], dtype=float32)
+        """
+        # we need to see if we are subclassing an
+        # index type here
+        if cls is not None:
+            dtype = cls
+            if isinstance(cls, str):
+                dtype = pandas_dtype(cls)
+
+            if needs_i8_conversion(dtype):
+                idx_cls = self._dtype_to_subclass(dtype)
+                arr = self.array.view(dtype)
+                if isinstance(arr, ExtensionArray):
+                    # here we exclude non-supported dt64/td64 dtypes
+                    return idx_cls._simple_new(
+                        arr, name=self.name, refs=self._references
+                    )
+                return arr
+
+            result = self._data.view(cls)
+        else:
+            result = self._view()
+        if isinstance(result, Index):
+            result._id = self._id
+        return result
+
+    def astype(self, dtype: Dtype, copy: bool = True):
+        """
+        Create an Index with values cast to dtypes.
+
+        The class of a new Index is determined by dtype. When conversion is
+        impossible, a TypeError exception is raised.
+
+        Parameters
+        ----------
+        dtype : numpy dtype or pandas type
+            Note that any signed integer `dtype` is treated as ``'int64'``,
+            and any unsigned integer `dtype` is treated as ``'uint64'``,
+            regardless of the size.
+        copy : bool, default True
+            By default, astype always returns a newly allocated object.
+            If copy is set to False and internal requirements on dtype are
+            satisfied, the original data is used to create a new Index
+            or the original Index is returned.
+
+        Returns
+        -------
+        Index
+            Index with values cast to specified dtype.
+
+        See Also
+        --------
+        Index.dtype: Return the dtype object of the underlying data.
+        Index.dtypes: Return the dtype object of the underlying data.
+        Index.convert_dtypes: Convert columns to the best possible dtypes.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.astype("float")
+        Index([1.0, 2.0, 3.0], dtype='float64')
+        """
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        if self.dtype == dtype:
+            # Ensure that self.astype(self.dtype) is self
+            return self.copy() if copy else self
+
+        values = self._data
+        if isinstance(values, ExtensionArray):
+            with rewrite_exception(type(values).__name__, type(self).__name__):
+                new_values = values.astype(dtype, copy=copy)
+
+        elif isinstance(dtype, ExtensionDtype):
+            cls = dtype.construct_array_type()
+            # Note: for RangeIndex and CategoricalDtype self vs self._values
+            #  behaves differently here.
+            new_values = cls._from_sequence(self, dtype=dtype, copy=copy)
+
+        else:
+            # GH#13149 specifically use astype_array instead of astype
+            new_values = astype_array(values, dtype=dtype, copy=copy)
+
+        # pass copy=False because any copying will be done in the astype above
+        result = Index(new_values, name=self.name, dtype=new_values.dtype, copy=False)
+        if (
+            not copy
+            and self._references is not None
+            and astype_is_view(self.dtype, dtype)
+        ):
+            result._references = self._references
+            result._references.add_index_reference(result)
+        return result
+
+    _index_shared_docs["take"] = """
+        Return a new %(klass)s of the values selected by the indices.
+
+        For internal compatibility with numpy arrays.
+
+        Parameters
+        ----------
+        indices : array-like
+            Indices to be taken.
+        axis : int, optional
+            The axis over which to select values, always 0.
+        allow_fill : bool, default True
+            How to handle negative values in `indices`.
+
+            * False: negative values in `indices` indicate positional indices
+              from the right (the default). This is similar to
+              :func:`numpy.take`.
+
+            * True: negative values in `indices` indicate
+              missing values. These values are set to `fill_value`. Any other
+              other negative values raise a ``ValueError``.
+
+        fill_value : scalar, default None
+            If allow_fill=True and fill_value is not None, indices specified by
+            -1 are regarded as NA. If Index doesn't hold NA, raise ValueError.
+        **kwargs
+            Required for compatibility with numpy.
+
+        Returns
+        -------
+        Index
+            An index formed of elements at the given indices. Will be the same
+            type as self, except for RangeIndex.
+
+        See Also
+        --------
+        numpy.ndarray.take: Return an array formed from the
+            elements of a at the given indices.
+
+        Examples
+        --------
+        >>> idx = pd.Index(['a', 'b', 'c'])
+        >>> idx.take([2, 2, 1, 2])
+        Index(['c', 'c', 'b', 'c'], dtype='object')
+        """
+
+    @Appender(_index_shared_docs["take"] % _index_doc_kwargs)
+    def take(
+        self,
+        indices,
+        axis: Axis = 0,
+        allow_fill: bool = True,
+        fill_value=None,
+        **kwargs,
+    ) -> Self:
+        if kwargs:
+            nv.validate_take((), kwargs)
+        if is_scalar(indices):
+            raise TypeError("Expected indices to be array-like")
+        indices = ensure_platform_int(indices)
+        allow_fill = self._maybe_disallow_fill(allow_fill, fill_value, indices)
+
+        if indices.ndim == 1 and lib.is_range_indexer(indices, len(self)):
+            return self.copy()
+
+        # Note: we discard fill_value and use self._na_value, only relevant
+        #  in the case where allow_fill is True and fill_value is not None
+        values = self._values
+        if isinstance(values, np.ndarray):
+            taken = algos.take(
+                values, indices, allow_fill=allow_fill, fill_value=self._na_value
+            )
+        else:
+            # algos.take passes 'axis' keyword which not all EAs accept
+            taken = values.take(
+                indices, allow_fill=allow_fill, fill_value=self._na_value
+            )
+        return self._constructor._simple_new(taken, name=self.name)
+
+    @final
+    def _maybe_disallow_fill(self, allow_fill: bool, fill_value, indices) -> bool:
+        """
+        We only use pandas-style take when allow_fill is True _and_
+        fill_value is not None.
+        """
+        if allow_fill and fill_value is not None:
+            # only fill if we are passing a non-None fill_value
+            if self._can_hold_na:
+                if (indices < -1).any():
+                    raise ValueError(
+                        "When allow_fill=True and fill_value is not None, "
+                        "all indices must be >= -1"
+                    )
+            else:
+                cls_name = type(self).__name__
+                raise ValueError(
+                    f"Unable to fill values because {cls_name} cannot contain NA"
+                )
+        else:
+            allow_fill = False
+        return allow_fill
+
+    _index_shared_docs["repeat"] = """
+        Repeat elements of a %(klass)s.
+
+        Returns a new %(klass)s where each element of the current %(klass)s
+        is repeated consecutively a given number of times.
+
+        Parameters
+        ----------
+        repeats : int or array of ints
+            The number of repetitions for each element. This should be a
+            non-negative integer. Repeating 0 times will return an empty
+            %(klass)s.
+        axis : None
+            Must be ``None``. Has no effect but is accepted for compatibility
+            with numpy.
+
+        Returns
+        -------
+        %(klass)s
+            Newly created %(klass)s with repeated elements.
+
+        See Also
+        --------
+        Series.repeat : Equivalent function for Series.
+        numpy.repeat : Similar method for :class:`numpy.ndarray`.
+
+        Examples
+        --------
+        >>> idx = pd.Index(['a', 'b', 'c'])
+        >>> idx
+        Index(['a', 'b', 'c'], dtype='object')
+        >>> idx.repeat(2)
+        Index(['a', 'a', 'b', 'b', 'c', 'c'], dtype='object')
+        >>> idx.repeat([1, 2, 3])
+        Index(['a', 'b', 'b', 'c', 'c', 'c'], dtype='object')
+        """
+
+    @Appender(_index_shared_docs["repeat"] % _index_doc_kwargs)
+    def repeat(self, repeats, axis: None = None) -> Self:
+        repeats = ensure_platform_int(repeats)
+        nv.validate_repeat((), {"axis": axis})
+        res_values = self._values.repeat(repeats)
+
+        # _constructor so RangeIndex-> Index with an int64 dtype
+        return self._constructor._simple_new(res_values, name=self.name)
+
+    # --------------------------------------------------------------------
+    # Copying Methods
+
+    def copy(
+        self,
+        name: Hashable | None = None,
+        deep: bool = False,
+    ) -> Self:
+        """
+        Make a copy of this object.
+
+        Name is set on the new object.
+
+        Parameters
+        ----------
+        name : Label, optional
+            Set name for new object.
+        deep : bool, default False
+            If True attempts to make a deep copy of the Index.
+                Else makes a shallow copy.
+
+        Returns
+        -------
+        Index
+            Index refer to new object which is a copy of this object.
+
+        See Also
+        --------
+        Index.delete: Make new Index with passed location(-s) deleted.
+        Index.drop: Make new Index with passed list of labels deleted.
+
+        Notes
+        -----
+        In most cases, there should be no functional difference from using
+        ``deep``, but if ``deep`` is passed it will attempt to deepcopy.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> new_idx = idx.copy()
+        >>> idx is new_idx
+        False
+        """
+
+        name = self._validate_names(name=name, deep=deep)[0]
+        if deep:
+            new_data = self._data.copy()
+            new_index = type(self)._simple_new(new_data, name=name)
+        else:
+            new_index = self._rename(name=name)
+        return new_index
+
+    @final
+    def __copy__(self, **kwargs) -> Self:
+        return self.copy(**kwargs)
+
+    @final
+    def __deepcopy__(self, memo=None) -> Self:
+        """
+        Parameters
+        ----------
+        memo, default None
+            Standard signature. Unused
+        """
+        return self.copy(deep=True)
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+
+    @final
+    def __repr__(self) -> str_t:
+        """
+        Return a string representation for this object.
+        """
+        klass_name = type(self).__name__
+        data = self._format_data()
+        attrs = self._format_attrs()
+        attrs_str = [f"{k}={v}" for k, v in attrs]
+        prepr = ", ".join(attrs_str)
+
+        return f"{klass_name}({data}{prepr})"
+
+    @property
+    def _formatter_func(self):
+        """
+        Return the formatter function.
+        """
+        return default_pprint
+
+    @final
+    def _format_data(self, name=None) -> str_t:
+        """
+        Return the formatted data as a unicode string.
+        """
+        # do we want to justify (only do so for non-objects)
+        is_justify = True
+
+        if self.inferred_type == "string":
+            is_justify = False
+        elif isinstance(self.dtype, CategoricalDtype):
+            self = cast("CategoricalIndex", self)
+            if is_object_dtype(self.categories.dtype):
+                is_justify = False
+        elif isinstance(self, ABCRangeIndex):
+            # We will do the relevant formatting via attrs
+            return ""
+
+        return format_object_summary(
+            self,
+            self._formatter_func,
+            is_justify=is_justify,
+            name=name,
+            line_break_each_value=self._is_multi,
+        )
+
+    def _format_attrs(self) -> list[tuple[str_t, str_t | int | bool | None]]:
+        """
+        Return a list of tuples of the (attr,formatted_value).
+        """
+        attrs: list[tuple[str_t, str_t | int | bool | None]] = []
+
+        if not self._is_multi:
+            attrs.append(("dtype", f"'{self.dtype}'"))
+
+        if self.name is not None:
+            attrs.append(("name", default_pprint(self.name)))
+        elif self._is_multi and any(x is not None for x in self.names):
+            attrs.append(("names", default_pprint(self.names)))
+
+        max_seq_items = get_option("display.max_seq_items") or len(self)
+        if len(self) > max_seq_items:
+            attrs.append(("length", len(self)))
+        return attrs
+
+    @final
+    def _get_level_names(self) -> range | Sequence[Hashable]:
+        """
+        Return a name or list of names with None replaced by the level number.
+        """
+        if self._is_multi:
+            return maybe_sequence_to_range(
+                [
+                    level if name is None else name
+                    for level, name in enumerate(self.names)
+                ]
+            )
+        else:
+            return range(1) if self.name is None else [self.name]
+
+    @final
+    def _mpl_repr(self) -> np.ndarray:
+        # how to represent ourselves to matplotlib
+        if isinstance(self.dtype, np.dtype) and self.dtype.kind != "M":
+            return cast(np.ndarray, self.values)
+        return self.astype(object, copy=False)._values
+
+    _default_na_rep = "NaN"
+
+    @final
+    def _format_flat(
+        self,
+        *,
+        include_name: bool,
+        formatter: Callable | None = None,
+    ) -> list[str_t]:
+        """
+        Render a string representation of the Index.
+        """
+        header = []
+        if include_name:
+            header.append(
+                pprint_thing(self.name, escape_chars=("\t", "\r", "\n"))
+                if self.name is not None
+                else ""
+            )
+
+        if formatter is not None:
+            return header + list(self.map(formatter))
+
+        return self._format_with_header(header=header, na_rep=self._default_na_rep)
+
+    def _format_with_header(self, *, header: list[str_t], na_rep: str_t) -> list[str_t]:
+        from pandas.io.formats.format import format_array
+
+        values = self._values
+
+        if (
+            is_object_dtype(values.dtype)
+            or is_string_dtype(values.dtype)
+            or isinstance(self.dtype, (IntervalDtype, CategoricalDtype))
+        ):
+            # TODO: why do we need different justify for these cases?
+            justify = "all"
+        else:
+            justify = "left"
+        # passing leading_space=False breaks test_format_missing,
+        #  test_index_repr_in_frame_with_nan, but would otherwise make
+        #  trim_front unnecessary
+        formatted = format_array(values, None, justify=justify)
+        result = trim_front(formatted)
+        return header + result
+
+    def _get_values_for_csv(
+        self,
+        *,
+        na_rep: str_t = "",
+        decimal: str_t = ".",
+        float_format=None,
+        date_format=None,
+        quoting=None,
+    ) -> npt.NDArray[np.object_]:
+        return get_values_for_csv(
+            self._values,
+            na_rep=na_rep,
+            decimal=decimal,
+            float_format=float_format,
+            date_format=date_format,
+            quoting=quoting,
+        )
+
+    def _summary(self, name=None) -> str_t:
+        """
+        Return a summarized representation.
+
+        Parameters
+        ----------
+        name : str
+            name to use in the summary representation
+
+        Returns
+        -------
+        String with a summarized representation of the index
+        """
+        if len(self) > 0:
+            head = self[0]
+            if hasattr(head, "format") and not isinstance(head, str):
+                head = head.format()
+            elif needs_i8_conversion(self.dtype):
+                # e.g. Timedelta, display as values, not quoted
+                head = self._formatter_func(head).replace("'", "")
+            tail = self[-1]
+            if hasattr(tail, "format") and not isinstance(tail, str):
+                tail = tail.format()
+            elif needs_i8_conversion(self.dtype):
+                # e.g. Timedelta, display as values, not quoted
+                tail = self._formatter_func(tail).replace("'", "")
+
+            index_summary = f", {head} to {tail}"
+        else:
+            index_summary = ""
+
+        if name is None:
+            name = type(self).__name__
+        return f"{name}: {len(self)} entries{index_summary}"
+
+    # --------------------------------------------------------------------
+    # Conversion Methods
+
+    def to_flat_index(self) -> Self:
+        """
+        Identity method.
+
+        This is implemented for compatibility with subclass implementations
+        when chaining.
+
+        Returns
+        -------
+        pd.Index
+            Caller.
+
+        See Also
+        --------
+        MultiIndex.to_flat_index : Subclass implementation.
+        """
+        return self
+
+    @final
+    def to_series(self, index=None, name: Hashable | None = None) -> Series:
+        """
+        Create a Series with both index and values equal to the index keys.
+
+        Useful with map for returning an indexer based on an index.
+
+        Parameters
+        ----------
+        index : Index, optional
+            Index of resulting Series. If None, defaults to original index.
+        name : str, optional
+            Name of resulting Series. If None, defaults to name of original
+            index.
+
+        Returns
+        -------
+        Series
+            The dtype will be based on the type of the Index values.
+
+        See Also
+        --------
+        Index.to_frame : Convert an Index to a DataFrame.
+        Series.to_frame : Convert Series to DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["Ant", "Bear", "Cow"], name="animal")
+
+        By default, the original index and original name is reused.
+
+        >>> idx.to_series()
+        animal
+        Ant      Ant
+        Bear    Bear
+        Cow      Cow
+        Name: animal, dtype: object
+
+        To enforce a new index, specify new labels to ``index``:
+
+        >>> idx.to_series(index=[0, 1, 2])
+        0     Ant
+        1    Bear
+        2     Cow
+        Name: animal, dtype: object
+
+        To override the name of the resulting column, specify ``name``:
+
+        >>> idx.to_series(name="zoo")
+        animal
+        Ant      Ant
+        Bear    Bear
+        Cow      Cow
+        Name: zoo, dtype: object
+        """
+        from pandas import Series
+
+        if index is None:
+            index = self._view()
+        if name is None:
+            name = self.name
+
+        return Series(self._values.copy(), index=index, name=name)
+
+    def to_frame(
+        self, index: bool = True, name: Hashable = lib.no_default
+    ) -> DataFrame:
+        """
+        Create a DataFrame with a column containing the Index.
+
+        Parameters
+        ----------
+        index : bool, default True
+            Set the index of the returned DataFrame as the original Index.
+
+        name : object, defaults to index.name
+            The passed name should substitute for the index name (if it has
+            one).
+
+        Returns
+        -------
+        DataFrame
+            DataFrame containing the original Index data.
+
+        See Also
+        --------
+        Index.to_series : Convert an Index to a Series.
+        Series.to_frame : Convert Series to DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["Ant", "Bear", "Cow"], name="animal")
+        >>> idx.to_frame()
+               animal
+        animal
+        Ant       Ant
+        Bear     Bear
+        Cow       Cow
+
+        By default, the original Index is reused. To enforce a new Index:
+
+        >>> idx.to_frame(index=False)
+            animal
+        0   Ant
+        1  Bear
+        2   Cow
+
+        To override the name of the resulting column, specify `name`:
+
+        >>> idx.to_frame(index=False, name="zoo")
+            zoo
+        0   Ant
+        1  Bear
+        2   Cow
+        """
+        from pandas import DataFrame
+
+        if name is lib.no_default:
+            result_name = self._get_level_names()
+        else:
+            result_name = Index([name])  # type: ignore[assignment]
+        result = DataFrame(self, copy=False)
+        result.columns = result_name
+
+        if index:
+            result.index = self
+        return result
+
+    # --------------------------------------------------------------------
+    # Name-Centric Methods
+
+    @property
+    def name(self) -> Hashable:
+        """
+        Return Index or MultiIndex name.
+
+        See Also
+        --------
+        Index.set_names: Able to set new names partially and by level.
+        Index.rename: Able to set new names partially and by level.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3], name="x")
+        >>> idx
+        Index([1, 2, 3], dtype='int64',  name='x')
+        >>> idx.name
+        'x'
+        """
+        return self._name
+
+    @name.setter
+    def name(self, value: Hashable) -> None:
+        if self._no_setting_name:
+            # Used in MultiIndex.levels to avoid silently ignoring name updates.
+            raise RuntimeError(
+                "Cannot set name on a level of a MultiIndex. Use "
+                "'MultiIndex.set_names' instead."
+            )
+        maybe_extract_name(value, None, type(self))
+        self._name = value
+
+    @final
+    def _validate_names(
+        self, name=None, names=None, deep: bool = False
+    ) -> list[Hashable]:
+        """
+        Handles the quirks of having a singular 'name' parameter for general
+        Index and plural 'names' parameter for MultiIndex.
+        """
+        from copy import deepcopy
+
+        if names is not None and name is not None:
+            raise TypeError("Can only provide one of `names` and `name`")
+        if names is None and name is None:
+            new_names = deepcopy(self.names) if deep else self.names
+        elif names is not None:
+            if not is_list_like(names):
+                raise TypeError("Must pass list-like as `names`.")
+            new_names = names
+        elif not is_list_like(name):
+            new_names = [name]
+        else:
+            new_names = name
+
+        if len(new_names) != len(self.names):
+            raise ValueError(
+                f"Length of new names must be {len(self.names)}, got {len(new_names)}"
+            )
+
+        # All items in 'new_names' need to be hashable
+        validate_all_hashable(*new_names, error_name=f"{type(self).__name__}.name")
+
+        return new_names
+
+    def _get_default_index_names(
+        self, names: Hashable | Sequence[Hashable] | None = None, default=None
+    ) -> list[Hashable]:
+        """
+        Get names of index.
+
+        Parameters
+        ----------
+        names : int, str or 1-dimensional list, default None
+            Index names to set.
+        default : str
+            Default name of index.
+
+        Raises
+        ------
+        TypeError
+            if names not str or list-like
+        """
+        from pandas.core.indexes.multi import MultiIndex
+
+        if names is not None:
+            if isinstance(names, (int, str)):
+                names = [names]
+
+        if not isinstance(names, list) and names is not None:
+            raise ValueError("Index names must be str or 1-dimensional list")
+
+        if not names:
+            if isinstance(self, MultiIndex):
+                names = com.fill_missing_names(self.names)
+            else:
+                names = [default] if self.name is None else [self.name]
+
+        return names
+
+    def _get_names(self) -> FrozenList:
+        """
+        Get names on index.
+
+        This method returns a FrozenList containing the names of the object.
+        It's primarily intended for internal use.
+
+        Returns
+        -------
+        FrozenList
+            A FrozenList containing the object's names, contains None if the object
+            does not have a name.
+
+        See Also
+        --------
+        Index.name : Index name as a string, or None for MultiIndex.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3], name="x")
+        >>> idx.names
+        FrozenList(['x'])
+
+        >>> idx = pd.Index([1, 2, 3], name=("x", "y"))
+        >>> idx.names
+        FrozenList([('x', 'y')])
+
+        If the index does not have a name set:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.names
+        FrozenList([None])
+        """
+        return FrozenList((self.name,))
+
+    def _set_names(self, values, *, level=None) -> None:
+        """
+        Set new names on index. Each name has to be a hashable type.
+
+        Parameters
+        ----------
+        values : str or sequence
+            name(s) to set
+        level : int, level name, or sequence of int/level names (default None)
+            If the index is a MultiIndex (hierarchical), level(s) to set (None
+            for all levels).  Otherwise level must be None
+
+        Raises
+        ------
+        TypeError if each name is not hashable.
+        """
+        if not is_list_like(values):
+            raise ValueError("Names must be a list-like")
+        if len(values) != 1:
+            raise ValueError(f"Length of new names must be 1, got {len(values)}")
+
+        # GH 20527
+        # All items in 'name' need to be hashable:
+        validate_all_hashable(*values, error_name=f"{type(self).__name__}.name")
+
+        self._name = values[0]
+
+    names = property(fset=_set_names, fget=_get_names)
+
+    @overload
+    def set_names(self, names, *, level=..., inplace: Literal[False] = ...) -> Self: ...
+
+    @overload
+    def set_names(self, names, *, level=..., inplace: Literal[True]) -> None: ...
+
+    @overload
+    def set_names(self, names, *, level=..., inplace: bool = ...) -> Self | None: ...
+
+    def set_names(self, names, *, level=None, inplace: bool = False) -> Self | None:
+        """
+        Set Index or MultiIndex name.
+
+        Able to set new names partially and by level.
+
+        Parameters
+        ----------
+
+        names : label or list of label or dict-like for MultiIndex
+            Name(s) to set.
+
+            .. versionchanged:: 1.3.0
+
+        level : int, label or list of int or label, optional
+            If the index is a MultiIndex and names is not dict-like, level(s) to set
+            (None for all levels). Otherwise level must be None.
+
+            .. versionchanged:: 1.3.0
+
+        inplace : bool, default False
+            Modifies the object directly, instead of creating a new Index or
+            MultiIndex.
+
+        Returns
+        -------
+        Index or None
+            The same type as the caller or None if ``inplace=True``.
+
+        See Also
+        --------
+        Index.rename : Able to set new names without level.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3, 4])
+        >>> idx
+        Index([1, 2, 3, 4], dtype='int64')
+        >>> idx.set_names("quarter")
+        Index([1, 2, 3, 4], dtype='int64', name='quarter')
+
+        >>> idx = pd.MultiIndex.from_product([["python", "cobra"], [2018, 2019]])
+        >>> idx
+        MultiIndex([('python', 2018),
+                    ('python', 2019),
+                    ( 'cobra', 2018),
+                    ( 'cobra', 2019)],
+                   )
+        >>> idx = idx.set_names(["kind", "year"])
+        >>> idx.set_names("species", level=0)
+        MultiIndex([('python', 2018),
+                    ('python', 2019),
+                    ( 'cobra', 2018),
+                    ( 'cobra', 2019)],
+                   names=['species', 'year'])
+
+        When renaming levels with a dict, levels can not be passed.
+
+        >>> idx.set_names({"kind": "snake"})
+        MultiIndex([('python', 2018),
+                    ('python', 2019),
+                    ( 'cobra', 2018),
+                    ( 'cobra', 2019)],
+                   names=['snake', 'year'])
+        """
+        if level is not None and not isinstance(self, ABCMultiIndex):
+            raise ValueError("Level must be None for non-MultiIndex")
+
+        if level is not None and not is_list_like(level) and is_list_like(names):
+            raise TypeError("Names must be a string when a single level is provided.")
+
+        if not is_list_like(names) and level is None and self.nlevels > 1:
+            raise TypeError("Must pass list-like as `names`.")
+
+        if is_dict_like(names) and not isinstance(self, ABCMultiIndex):
+            raise TypeError("Can only pass dict-like as `names` for MultiIndex.")
+
+        if is_dict_like(names) and level is not None:
+            raise TypeError("Can not pass level for dictlike `names`.")
+
+        if isinstance(self, ABCMultiIndex) and is_dict_like(names) and level is None:
+            # Transform dict to list of new names and corresponding levels
+            level, names_adjusted = [], []
+            for i, name in enumerate(self.names):
+                if name in names.keys():
+                    level.append(i)
+                    names_adjusted.append(names[name])
+            names = names_adjusted
+
+        if not is_list_like(names):
+            names = [names]
+        if level is not None and not is_list_like(level):
+            level = [level]
+
+        if inplace:
+            idx = self
+        else:
+            idx = self._view()
+
+        idx._set_names(names, level=level)
+        if not inplace:
+            return idx
+        return None
+
+    @overload
+    def rename(self, name, *, inplace: Literal[False] = ...) -> Self: ...
+
+    @overload
+    def rename(self, name, *, inplace: Literal[True]) -> None: ...
+
+    def rename(self, name, *, inplace: bool = False) -> Self | None:
+        """
+        Alter Index or MultiIndex name.
+
+        Able to set new names without level. Defaults to returning new index.
+        Length of names must match number of levels in MultiIndex.
+
+        Parameters
+        ----------
+        name : label or list of labels
+            Name(s) to set.
+        inplace : bool, default False
+            Modifies the object directly, instead of creating a new Index or
+            MultiIndex.
+
+        Returns
+        -------
+        Index or None
+            The same type as the caller or None if ``inplace=True``.
+
+        See Also
+        --------
+        Index.set_names : Able to set new names partially and by level.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["A", "C", "A", "B"], name="score")
+        >>> idx.rename("grade")
+        Index(['A', 'C', 'A', 'B'], dtype='object', name='grade')
+
+        >>> idx = pd.MultiIndex.from_product(
+        ...     [["python", "cobra"], [2018, 2019]], names=["kind", "year"]
+        ... )
+        >>> idx
+        MultiIndex([('python', 2018),
+                    ('python', 2019),
+                    ( 'cobra', 2018),
+                    ( 'cobra', 2019)],
+                   names=['kind', 'year'])
+        >>> idx.rename(["species", "year"])
+        MultiIndex([('python', 2018),
+                    ('python', 2019),
+                    ( 'cobra', 2018),
+                    ( 'cobra', 2019)],
+                   names=['species', 'year'])
+        >>> idx.rename("species")
+        Traceback (most recent call last):
+        TypeError: Must pass list-like as `names`.
+        """
+        return self.set_names([name], inplace=inplace)
+
+    # --------------------------------------------------------------------
+    # Level-Centric Methods
+
+    @property
+    def nlevels(self) -> int:
+        """
+        Number of levels.
+        """
+        return 1
+
+    def _sort_levels_monotonic(self) -> Self:
+        """
+        Compat with MultiIndex.
+        """
+        return self
+
+    @final
+    def _validate_index_level(self, level) -> None:
+        """
+        Validate index level.
+
+        For single-level Index getting level number is a no-op, but some
+        verification must be done like in MultiIndex.
+
+        """
+        if isinstance(level, int):
+            if level < 0 and level != -1:
+                raise IndexError(
+                    "Too many levels: Index has only 1 level, "
+                    f"{level} is not a valid level number"
+                )
+            if level > 0:
+                raise IndexError(
+                    f"Too many levels: Index has only 1 level, not {level + 1}"
+                )
+        elif level != self.name:
+            raise KeyError(
+                f"Requested level ({level}) does not match index name ({self.name})"
+            )
+
+    def _get_level_number(self, level) -> int:
+        self._validate_index_level(level)
+        return 0
+
+    def sortlevel(
+        self,
+        level=None,
+        ascending: bool | list[bool] = True,
+        sort_remaining=None,
+        na_position: NaPosition = "first",
+    ) -> tuple[Self, np.ndarray]:
+        """
+        For internal compatibility with the Index API.
+
+        Sort the Index. This is for compat with MultiIndex
+
+        Parameters
+        ----------
+        ascending : bool, default True
+            False to sort in descending order
+        na_position : {'first' or 'last'}, default 'first'
+            Argument 'first' puts NaNs at the beginning, 'last' puts NaNs at
+            the end.
+
+            .. versionadded:: 2.1.0
+
+        level, sort_remaining are compat parameters
+
+        Returns
+        -------
+        Index
+        """
+        if not isinstance(ascending, (list, bool)):
+            raise TypeError(
+                "ascending must be a single bool value or"
+                "a list of bool values of length 1"
+            )
+
+        if isinstance(ascending, list):
+            if len(ascending) != 1:
+                raise TypeError("ascending must be a list of bool values of length 1")
+            ascending = ascending[0]
+
+        if not isinstance(ascending, bool):
+            raise TypeError("ascending must be a bool value")
+
+        return self.sort_values(
+            return_indexer=True, ascending=ascending, na_position=na_position
+        )
+
+    def _get_level_values(self, level) -> Index:
+        """
+        Return an Index of values for requested level.
+
+        This is primarily useful to get an individual level of values from a
+        MultiIndex, but is provided on Index as well for compatibility.
+
+        Parameters
+        ----------
+        level : int or str
+            It is either the integer position or the name of the level.
+
+        Returns
+        -------
+        Index
+            Calling object, as there is only one level in the Index.
+
+        See Also
+        --------
+        MultiIndex.get_level_values : Get values for a level of a MultiIndex.
+
+        Notes
+        -----
+        For Index, level should be 0, since there are no multiple levels.
+
+        Examples
+        --------
+        >>> idx = pd.Index(list("abc"))
+        >>> idx
+        Index(['a', 'b', 'c'], dtype='object')
+
+        Get level values by supplying `level` as integer:
+
+        >>> idx.get_level_values(0)
+        Index(['a', 'b', 'c'], dtype='object')
+        """
+        self._validate_index_level(level)
+        return self
+
+    get_level_values = _get_level_values
+
+    @final
+    def droplevel(self, level: IndexLabel = 0):
+        """
+        Return index with requested level(s) removed.
+
+        If resulting index has only 1 level left, the result will be
+        of Index type, not MultiIndex. The original index is not modified inplace.
+
+        Parameters
+        ----------
+        level : int, str, or list-like, default 0
+            If a string is given, must be the name of a level
+            If list-like, elements must be names or indexes of levels.
+
+        Returns
+        -------
+        Index or MultiIndex
+            Returns an Index or MultiIndex object, depending on the resulting index
+            after removing the requested level(s).
+
+        See Also
+        --------
+        Index.dropna : Return Index without NA/NaN values.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays(
+        ...     [[1, 2], [3, 4], [5, 6]], names=["x", "y", "z"]
+        ... )
+        >>> mi
+        MultiIndex([(1, 3, 5),
+                    (2, 4, 6)],
+                   names=['x', 'y', 'z'])
+
+        >>> mi.droplevel()
+        MultiIndex([(3, 5),
+                    (4, 6)],
+                   names=['y', 'z'])
+
+        >>> mi.droplevel(2)
+        MultiIndex([(1, 3),
+                    (2, 4)],
+                   names=['x', 'y'])
+
+        >>> mi.droplevel("z")
+        MultiIndex([(1, 3),
+                    (2, 4)],
+                   names=['x', 'y'])
+
+        >>> mi.droplevel(["x", "y"])
+        Index([5, 6], dtype='int64', name='z')
+        """
+        if not isinstance(level, (tuple, list)):
+            level = [level]
+
+        levnums = sorted((self._get_level_number(lev) for lev in level), reverse=True)
+
+        return self._drop_level_numbers(levnums)
+
+    @final
+    def _drop_level_numbers(self, levnums: list[int]):
+        """
+        Drop MultiIndex levels by level _number_, not name.
+        """
+
+        if not levnums and not isinstance(self, ABCMultiIndex):
+            return self
+        if len(levnums) >= self.nlevels:
+            raise ValueError(
+                f"Cannot remove {len(levnums)} levels from an index with "
+                f"{self.nlevels} levels: at least one level must be left."
+            )
+        # The two checks above guarantee that here self is a MultiIndex
+        self = cast("MultiIndex", self)
+
+        new_levels = list(self.levels)
+        new_codes = list(self.codes)
+        new_names = list(self.names)
+
+        for i in levnums:
+            new_levels.pop(i)
+            new_codes.pop(i)
+            new_names.pop(i)
+
+        if len(new_levels) == 1:
+            lev = new_levels[0]
+
+            if len(lev) == 0:
+                # If lev is empty, lev.take will fail GH#42055
+                if len(new_codes[0]) == 0:
+                    # GH#45230 preserve RangeIndex here
+                    #  see test_reset_index_empty_rangeindex
+                    result = lev[:0]
+                else:
+                    res_values = algos.take(lev._values, new_codes[0], allow_fill=True)
+                    # _constructor instead of type(lev) for RangeIndex compat GH#35230
+                    result = lev._constructor._simple_new(res_values, name=new_names[0])
+            else:
+                # set nan if needed
+                mask = new_codes[0] == -1
+                result = new_levels[0].take(new_codes[0])
+                if mask.any():
+                    result = result.putmask(mask, np.nan)
+
+                result._name = new_names[0]
+
+            return result
+        else:
+            from pandas.core.indexes.multi import MultiIndex
+
+            return MultiIndex(
+                levels=new_levels,
+                codes=new_codes,
+                names=new_names,
+                verify_integrity=False,
+            )
+
+    # --------------------------------------------------------------------
+    # Introspection Methods
+
+    @cache_readonly
+    @final
+    def _can_hold_na(self) -> bool:
+        if isinstance(self.dtype, ExtensionDtype):
+            return self.dtype._can_hold_na
+        if self.dtype.kind in "iub":
+            return False
+        return True
+
+    @property
+    def is_monotonic_increasing(self) -> bool:
+        """
+        Return a boolean if the values are equal or increasing.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Index.is_monotonic_decreasing : Check if the values are equal or decreasing.
+
+        Examples
+        --------
+        >>> pd.Index([1, 2, 3]).is_monotonic_increasing
+        True
+        >>> pd.Index([1, 2, 2]).is_monotonic_increasing
+        True
+        >>> pd.Index([1, 3, 2]).is_monotonic_increasing
+        False
+        """
+        return self._engine.is_monotonic_increasing
+
+    @property
+    def is_monotonic_decreasing(self) -> bool:
+        """
+        Return a boolean if the values are equal or decreasing.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Index.is_monotonic_increasing : Check if the values are equal or increasing.
+
+        Examples
+        --------
+        >>> pd.Index([3, 2, 1]).is_monotonic_decreasing
+        True
+        >>> pd.Index([3, 2, 2]).is_monotonic_decreasing
+        True
+        >>> pd.Index([3, 1, 2]).is_monotonic_decreasing
+        False
+        """
+        return self._engine.is_monotonic_decreasing
+
+    @final
+    @property
+    def _is_strictly_monotonic_increasing(self) -> bool:
+        """
+        Return if the index is strictly monotonic increasing
+        (only increasing) values.
+
+        Examples
+        --------
+        >>> Index([1, 2, 3])._is_strictly_monotonic_increasing
+        True
+        >>> Index([1, 2, 2])._is_strictly_monotonic_increasing
+        False
+        >>> Index([1, 3, 2])._is_strictly_monotonic_increasing
+        False
+        """
+        return self.is_unique and self.is_monotonic_increasing
+
+    @final
+    @property
+    def _is_strictly_monotonic_decreasing(self) -> bool:
+        """
+        Return if the index is strictly monotonic decreasing
+        (only decreasing) values.
+
+        Examples
+        --------
+        >>> Index([3, 2, 1])._is_strictly_monotonic_decreasing
+        True
+        >>> Index([3, 2, 2])._is_strictly_monotonic_decreasing
+        False
+        >>> Index([3, 1, 2])._is_strictly_monotonic_decreasing
+        False
+        """
+        return self.is_unique and self.is_monotonic_decreasing
+
+    @cache_readonly
+    def is_unique(self) -> bool:
+        """
+        Return if the index has unique values.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Index.has_duplicates : Inverse method that checks if it has duplicate values.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 5, 7, 7])
+        >>> idx.is_unique
+        False
+
+        >>> idx = pd.Index([1, 5, 7])
+        >>> idx.is_unique
+        True
+
+        >>> idx = pd.Index(["Watermelon", "Orange", "Apple", "Watermelon"]).astype(
+        ...     "category"
+        ... )
+        >>> idx.is_unique
+        False
+
+        >>> idx = pd.Index(["Orange", "Apple", "Watermelon"]).astype("category")
+        >>> idx.is_unique
+        True
+        """
+        return self._engine.is_unique
+
+    @final
+    @property
+    def has_duplicates(self) -> bool:
+        """
+        Check if the Index has duplicate values.
+
+        Returns
+        -------
+        bool
+            Whether or not the Index has duplicate values.
+
+        See Also
+        --------
+        Index.is_unique : Inverse method that checks if it has unique values.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 5, 7, 7])
+        >>> idx.has_duplicates
+        True
+
+        >>> idx = pd.Index([1, 5, 7])
+        >>> idx.has_duplicates
+        False
+
+        >>> idx = pd.Index(["Watermelon", "Orange", "Apple", "Watermelon"]).astype(
+        ...     "category"
+        ... )
+        >>> idx.has_duplicates
+        True
+
+        >>> idx = pd.Index(["Orange", "Apple", "Watermelon"]).astype("category")
+        >>> idx.has_duplicates
+        False
+        """
+        return not self.is_unique
+
+    @cache_readonly
+    def inferred_type(self) -> str_t:
+        """
+        Return a string of the type inferred from the values.
+
+        See Also
+        --------
+        Index.dtype : Return the dtype object of the underlying data.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.inferred_type
+        'integer'
+        """
+        return lib.infer_dtype(self._values, skipna=False)
+
+    @cache_readonly
+    @final
+    def _is_all_dates(self) -> bool:
+        """
+        Whether or not the index values only consist of dates.
+        """
+        if needs_i8_conversion(self.dtype):
+            return True
+        elif self.dtype != _dtype_obj:
+            # TODO(ExtensionIndex): 3rd party EA might override?
+            # Note: this includes IntervalIndex, even when the left/right
+            #  contain datetime-like objects.
+            return False
+        elif self._is_multi:
+            return False
+        return is_datetime_array(ensure_object(self._values))
+
+    @final
+    @cache_readonly
+    def _is_multi(self) -> bool:
+        """
+        Cached check equivalent to isinstance(self, MultiIndex)
+        """
+        return isinstance(self, ABCMultiIndex)
+
+    # --------------------------------------------------------------------
+    # Pickle Methods
+
+    def __reduce__(self):
+        d = {"data": self._data, "name": self.name}
+        return _new_Index, (type(self), d), None
+
+    # --------------------------------------------------------------------
+    # Null Handling Methods
+
+    @cache_readonly
+    def _na_value(self):
+        """The expected NA value to use with this index."""
+        dtype = self.dtype
+        if isinstance(dtype, np.dtype):
+            if dtype.kind in "mM":
+                return NaT
+            return np.nan
+        return dtype.na_value
+
+    @cache_readonly
+    def _isnan(self) -> npt.NDArray[np.bool_]:
+        """
+        Return if each value is NaN.
+        """
+        if self._can_hold_na:
+            return isna(self)
+        else:
+            # shouldn't reach to this condition by checking hasnans beforehand
+            values = np.empty(len(self), dtype=np.bool_)
+            values.fill(False)
+            return values
+
+    @cache_readonly
+    def hasnans(self) -> bool:
+        """
+        Return True if there are any NaNs.
+
+        Enables various performance speedups.
+
+        Returns
+        -------
+        bool
+
+        See Also
+        --------
+        Index.isna : Detect missing values.
+        Index.dropna : Return Index without NA/NaN values.
+        Index.fillna : Fill NA/NaN values with the specified value.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3], index=["a", "b", None])
+        >>> s
+        a    1
+        b    2
+        None 3
+        dtype: int64
+        >>> s.index.hasnans
+        True
+        """
+        if self._can_hold_na:
+            return bool(self._isnan.any())
+        else:
+            return False
+
+    @final
+    def isna(self) -> npt.NDArray[np.bool_]:
+        """
+        Detect missing values.
+
+        Return a boolean same-sized object indicating if the values are NA.
+        NA values, such as ``None``, :attr:`numpy.NaN` or :attr:`pd.NaT`, get
+        mapped to ``True`` values.
+        Everything else get mapped to ``False`` values. Characters such as
+        empty strings `''` or :attr:`numpy.inf` are not considered NA values.
+
+        Returns
+        -------
+        numpy.ndarray[bool]
+            A boolean array of whether my values are NA.
+
+        See Also
+        --------
+        Index.notna : Boolean inverse of isna.
+        Index.dropna : Omit entries with missing values.
+        isna : Top-level isna.
+        Series.isna : Detect missing values in Series object.
+
+        Examples
+        --------
+        Show which entries in a pandas.Index are NA. The result is an
+        array.
+
+        >>> idx = pd.Index([5.2, 6.0, np.nan])
+        >>> idx
+        Index([5.2, 6.0, nan], dtype='float64')
+        >>> idx.isna()
+        array([False, False,  True])
+
+        Empty strings are not considered NA values. None is considered an NA
+        value.
+
+        >>> idx = pd.Index(["black", "", "red", None])
+        >>> idx
+        Index(['black', '', 'red', None], dtype='object')
+        >>> idx.isna()
+        array([False, False, False,  True])
+
+        For datetimes, `NaT` (Not a Time) is considered as an NA value.
+
+        >>> idx = pd.DatetimeIndex(
+        ...     [pd.Timestamp("1940-04-25"), pd.Timestamp(""), None, pd.NaT]
+        ... )
+        >>> idx
+        DatetimeIndex(['1940-04-25', 'NaT', 'NaT', 'NaT'],
+                      dtype='datetime64[s]', freq=None)
+        >>> idx.isna()
+        array([False,  True,  True,  True])
+        """
+        return self._isnan
+
+    isnull = isna
+
+    @final
+    def notna(self) -> npt.NDArray[np.bool_]:
+        """
+        Detect existing (non-missing) values.
+
+        Return a boolean same-sized object indicating if the values are not NA.
+        Non-missing values get mapped to ``True``. Characters such as empty
+        strings ``''`` or :attr:`numpy.inf` are not considered NA values.
+        NA values, such as None or :attr:`numpy.NaN`, get mapped to ``False``
+        values.
+
+        Returns
+        -------
+        numpy.ndarray[bool]
+            Boolean array to indicate which entries are not NA.
+
+        See Also
+        --------
+        Index.notnull : Alias of notna.
+        Index.isna: Inverse of notna.
+        notna : Top-level notna.
+
+        Examples
+        --------
+        Show which entries in an Index are not NA. The result is an
+        array.
+
+        >>> idx = pd.Index([5.2, 6.0, np.nan])
+        >>> idx
+        Index([5.2, 6.0, nan], dtype='float64')
+        >>> idx.notna()
+        array([ True,  True, False])
+
+        Empty strings are not considered NA values. None is considered a NA
+        value.
+
+        >>> idx = pd.Index(["black", "", "red", None])
+        >>> idx
+        Index(['black', '', 'red', None], dtype='object')
+        >>> idx.notna()
+        array([ True,  True,  True, False])
+        """
+        return ~self.isna()
+
+    notnull = notna
+
+    def fillna(self, value):
+        """
+        Fill NA/NaN values with the specified value.
+
+        Parameters
+        ----------
+        value : scalar
+            Scalar value to use to fill holes (e.g. 0).
+            This value cannot be a list-likes.
+
+        Returns
+        -------
+        Index
+           NA/NaN values replaced with `value`.
+
+        See Also
+        --------
+        DataFrame.fillna : Fill NaN values of a DataFrame.
+        Series.fillna : Fill NaN Values of a Series.
+
+        Examples
+        --------
+        >>> idx = pd.Index([np.nan, np.nan, 3])
+        >>> idx.fillna(0)
+        Index([0.0, 0.0, 3.0], dtype='float64')
+        """
+        if not is_scalar(value):
+            raise TypeError(f"'value' must be a scalar, passed: {type(value).__name__}")
+
+        if self.hasnans:
+            result = self.putmask(self._isnan, value)
+            # no need to care metadata other than name
+            # because it can't have freq if it has NaTs
+            # _with_infer needed for test_fillna_categorical
+            return Index._with_infer(result, name=self.name)
+        return self._view()
+
+    def dropna(self, how: AnyAll = "any") -> Self:
+        """
+        Return Index without NA/NaN values.
+
+        Parameters
+        ----------
+        how : {'any', 'all'}, default 'any'
+            If the Index is a MultiIndex, drop the value when any or all levels
+            are NaN.
+
+        Returns
+        -------
+        Index
+            Returns an Index object after removing NA/NaN values.
+
+        See Also
+        --------
+        Index.fillna : Fill NA/NaN values with the specified value.
+        Index.isna : Detect missing values.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, np.nan, 3])
+        >>> idx.dropna()
+        Index([1.0, 3.0], dtype='float64')
+        """
+        if how not in ("any", "all"):
+            raise ValueError(f"invalid how option: {how}")
+
+        if self.hasnans:
+            res_values = self._values[~self._isnan]
+            return type(self)._simple_new(res_values, name=self.name)
+        return self._view()
+
+    # --------------------------------------------------------------------
+    # Uniqueness Methods
+
+    def unique(self, level: Hashable | None = None) -> Self:
+        """
+        Return unique values in the index.
+
+        Unique values are returned in order of appearance, this does NOT sort.
+
+        Parameters
+        ----------
+        level : int or hashable, optional
+            Only return values from specified level (for MultiIndex).
+            If int, gets the level by integer position, else by level name.
+
+        Returns
+        -------
+        Index
+            Unique values in the index.
+
+        See Also
+        --------
+        unique : Numpy array of unique values in that column.
+        Series.unique : Return unique values of Series object.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 1, 2, 3, 3])
+        >>> idx.unique()
+        Index([1, 2, 3], dtype='int64')
+        """
+        if level is not None:
+            self._validate_index_level(level)
+
+        if self.is_unique:
+            return self._view()
+
+        result = super().unique()
+        return self._shallow_copy(result)
+
+    def drop_duplicates(self, *, keep: DropKeep = "first") -> Self:
+        """
+        Return Index with duplicate values removed.
+
+        Parameters
+        ----------
+        keep : {'first', 'last', ``False``}, default 'first'
+            - 'first' : Drop duplicates except for the first occurrence.
+            - 'last' : Drop duplicates except for the last occurrence.
+            - ``False`` : Drop all duplicates.
+
+        Returns
+        -------
+        Index
+            A new Index object with the duplicate values removed.
+
+        See Also
+        --------
+        Series.drop_duplicates : Equivalent method on Series.
+        DataFrame.drop_duplicates : Equivalent method on DataFrame.
+        Index.duplicated : Related method on Index, indicating duplicate
+            Index values.
+
+        Examples
+        --------
+        Generate an pandas.Index with duplicate values.
+
+        >>> idx = pd.Index(["llama", "cow", "llama", "beetle", "llama", "hippo"])
+
+        The `keep` parameter controls  which duplicate values are removed.
+        The value 'first' keeps the first occurrence for each
+        set of duplicated entries. The default value of keep is 'first'.
+
+        >>> idx.drop_duplicates(keep="first")
+        Index(['llama', 'cow', 'beetle', 'hippo'], dtype='object')
+
+        The value 'last' keeps the last occurrence for each set of duplicated
+        entries.
+
+        >>> idx.drop_duplicates(keep="last")
+        Index(['cow', 'beetle', 'llama', 'hippo'], dtype='object')
+
+        The value ``False`` discards all sets of duplicated entries.
+
+        >>> idx.drop_duplicates(keep=False)
+        Index(['cow', 'beetle', 'hippo'], dtype='object')
+        """
+        if self.is_unique:
+            return self._view()
+
+        return super().drop_duplicates(keep=keep)
+
+    def duplicated(self, keep: DropKeep = "first") -> npt.NDArray[np.bool_]:
+        """
+        Indicate duplicate index values.
+
+        Duplicated values are indicated as ``True`` values in the resulting
+        array. Either all duplicates, all except the first, or all except the
+        last occurrence of duplicates can be indicated.
+
+        Parameters
+        ----------
+        keep : {'first', 'last', False}, default 'first'
+            The value or values in a set of duplicates to mark as missing.
+
+            - 'first' : Mark duplicates as ``True`` except for the first
+              occurrence.
+            - 'last' : Mark duplicates as ``True`` except for the last
+              occurrence.
+            - ``False`` : Mark all duplicates as ``True``.
+
+        Returns
+        -------
+        np.ndarray[bool]
+            A numpy array of boolean values indicating duplicate index values.
+
+        See Also
+        --------
+        Series.duplicated : Equivalent method on pandas.Series.
+        DataFrame.duplicated : Equivalent method on pandas.DataFrame.
+        Index.drop_duplicates : Remove duplicate values from Index.
+
+        Examples
+        --------
+        By default, for each set of duplicated values, the first occurrence is
+        set to False and all others to True:
+
+        >>> idx = pd.Index(["llama", "cow", "llama", "beetle", "llama"])
+        >>> idx.duplicated()
+        array([False, False,  True, False,  True])
+
+        which is equivalent to
+
+        >>> idx.duplicated(keep="first")
+        array([False, False,  True, False,  True])
+
+        By using 'last', the last occurrence of each set of duplicated values
+        is set on False and all others on True:
+
+        >>> idx.duplicated(keep="last")
+        array([ True, False,  True, False, False])
+
+        By setting keep on ``False``, all duplicates are True:
+
+        >>> idx.duplicated(keep=False)
+        array([ True, False,  True, False,  True])
+        """
+        if self.is_unique:
+            # fastpath available bc we are immutable
+            return np.zeros(len(self), dtype=bool)
+        return self._duplicated(keep=keep)
+
+    # --------------------------------------------------------------------
+    # Arithmetic & Logical Methods
+
+    def __iadd__(self, other):
+        # alias for __add__
+        return self + other
+
+    @final
+    def __bool__(self) -> NoReturn:
+        raise ValueError(
+            f"The truth value of a {type(self).__name__} is ambiguous. "
+            "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
+        )
+
+    # --------------------------------------------------------------------
+    # Set Operation Methods
+
+    def _get_reconciled_name_object(self, other):
+        """
+        If the result of a set operation will be self,
+        return self, unless the name changes, in which
+        case make a shallow copy of self.
+        """
+        name = get_op_result_name(self, other)
+        if self.name is not name:
+            return self.rename(name)
+        return self
+
+    @final
+    def _validate_sort_keyword(self, sort) -> None:
+        if sort not in [None, False, True]:
+            raise ValueError(
+                "The 'sort' keyword only takes the values of "
+                f"None, True, or False; {sort} was passed."
+            )
+
+    @final
+    def _dti_setop_align_tzs(self, other: Index, setop: str_t) -> tuple[Index, Index]:
+        """
+        With mismatched timezones, cast both to UTC.
+        """
+        # Caller is responsible for checking
+        #  `self.dtype != other.dtype`
+        if (
+            isinstance(self, ABCDatetimeIndex)
+            and isinstance(other, ABCDatetimeIndex)
+            and self.tz is not None
+            and other.tz is not None
+        ):
+            # GH#39328, GH#45357
+            left = self.tz_convert("UTC")
+            right = other.tz_convert("UTC")
+            return left, right
+        return self, other
+
+    @final
+    def union(self, other, sort: bool | None = None):
+        """
+        Form the union of two Index objects.
+
+        If the Index objects are incompatible, both Index objects will be
+        cast to dtype('object') first.
+
+        Parameters
+        ----------
+        other : Index or array-like
+            Index or an array-like object containing elements to form the union
+            with the original Index.
+        sort : bool or None, default None
+            Whether to sort the resulting Index.
+
+            * None : Sort the result, except when
+
+              1. `self` and `other` are equal.
+              2. `self` or `other` has length 0.
+              3. Some values in `self` or `other` cannot be compared.
+                 A RuntimeWarning is issued in this case.
+
+            * False : do not sort the result.
+            * True : Sort the result (which may raise TypeError).
+
+        Returns
+        -------
+        Index
+            Returns a new Index object with all unique elements from both the original
+            Index and the `other` Index.
+
+        See Also
+        --------
+        Index.unique : Return unique values in the index.
+        Index.intersection : Form the intersection of two Index objects.
+        Index.difference : Return a new Index with elements of index not in `other`.
+
+        Examples
+        --------
+        Union matching dtypes
+
+        >>> idx1 = pd.Index([1, 2, 3, 4])
+        >>> idx2 = pd.Index([3, 4, 5, 6])
+        >>> idx1.union(idx2)
+        Index([1, 2, 3, 4, 5, 6], dtype='int64')
+
+        Union mismatched dtypes
+
+        >>> idx1 = pd.Index(["a", "b", "c", "d"])
+        >>> idx2 = pd.Index([1, 2, 3, 4])
+        >>> idx1.union(idx2)
+        Index(['a', 'b', 'c', 'd', 1, 2, 3, 4], dtype='object')
+
+        MultiIndex case
+
+        >>> idx1 = pd.MultiIndex.from_arrays(
+        ...     [[1, 1, 2, 2], ["Red", "Blue", "Red", "Blue"]]
+        ... )
+        >>> idx1
+        MultiIndex([(1,  'Red'),
+            (1, 'Blue'),
+            (2,  'Red'),
+            (2, 'Blue')],
+           )
+        >>> idx2 = pd.MultiIndex.from_arrays(
+        ...     [[3, 3, 2, 2], ["Red", "Green", "Red", "Green"]]
+        ... )
+        >>> idx2
+        MultiIndex([(3,   'Red'),
+            (3, 'Green'),
+            (2,   'Red'),
+            (2, 'Green')],
+           )
+        >>> idx1.union(idx2)
+        MultiIndex([(1,  'Blue'),
+            (1,   'Red'),
+            (2,  'Blue'),
+            (2, 'Green'),
+            (2,   'Red'),
+            (3, 'Green'),
+            (3,   'Red')],
+           )
+        >>> idx1.union(idx2, sort=False)
+        MultiIndex([(1,   'Red'),
+            (1,  'Blue'),
+            (2,   'Red'),
+            (2,  'Blue'),
+            (3,   'Red'),
+            (3, 'Green'),
+            (2, 'Green')],
+           )
+        """
+        self._validate_sort_keyword(sort)
+        self._assert_can_do_setop(other)
+        other, result_name = self._convert_can_do_setop(other)
+
+        if self.dtype != other.dtype:
+            if (
+                isinstance(self, ABCMultiIndex)
+                and not is_object_dtype(_unpack_nested_dtype(other))
+                and len(other) > 0
+            ):
+                raise NotImplementedError(
+                    "Can only union MultiIndex with MultiIndex or Index of tuples, "
+                    "try mi.to_flat_index().union(other) instead."
+                )
+            self, other = self._dti_setop_align_tzs(other, "union")
+
+            dtype = self._find_common_type_compat(other)
+            left = self.astype(dtype, copy=False)
+            right = other.astype(dtype, copy=False)
+            return left.union(right, sort=sort)
+
+        elif not len(other) or self.equals(other):
+            # NB: whether this (and the `if not len(self)` check below) come before
+            #  or after the dtype equality check above affects the returned dtype
+            result = self._get_reconciled_name_object(other)
+            if sort is True:
+                return result.sort_values()
+            return result
+
+        elif not len(self):
+            result = other._get_reconciled_name_object(self)
+            if sort is True:
+                return result.sort_values()
+            return result
+
+        result = self._union(other, sort=sort)
+
+        return self._wrap_setop_result(other, result)
+
+    def _union(self, other: Index, sort: bool | None):
+        """
+        Specific union logic should go here. In subclasses, union behavior
+        should be overwritten here rather than in `self.union`.
+
+        Parameters
+        ----------
+        other : Index or array-like
+        sort : False or None, default False
+            Whether to sort the resulting index.
+
+            * True : sort the result
+            * False : do not sort the result.
+            * None : sort the result, except when `self` and `other` are equal
+              or when the values cannot be compared.
+
+        Returns
+        -------
+        Index
+        """
+        lvals = self._values
+        rvals = other._values
+
+        if (
+            sort in (None, True)
+            and (self.is_unique or other.is_unique)
+            and self._can_use_libjoin
+            and other._can_use_libjoin
+        ):
+            # Both are monotonic and at least one is unique, so can use outer join
+            #  (actually don't need either unique, but without this restriction
+            #  test_union_same_value_duplicated_in_both fails)
+            try:
+                return self._outer_indexer(other)[0]
+            except (TypeError, IncompatibleFrequency):
+                # incomparable objects; should only be for object dtype
+                value_list = list(lvals)
+
+                # worth making this faster? a very unusual case
+                value_set = set(lvals)
+                value_list.extend(x for x in rvals if x not in value_set)
+                # If objects are unorderable, we must have object dtype.
+                return np.array(value_list, dtype=object)
+
+        elif not other.is_unique:
+            # other has duplicates
+            result_dups = algos.union_with_duplicates(self, other)
+            return _maybe_try_sort(result_dups, sort)
+
+        # The rest of this method is analogous to Index._intersection_via_get_indexer
+
+        # Self may have duplicates; other already checked as unique
+        # find indexes of things in "other" that are not in "self"
+        if self._index_as_unique:
+            indexer = self.get_indexer(other)
+            missing = (indexer == -1).nonzero()[0]
+        else:
+            missing = algos.unique1d(self.get_indexer_non_unique(other)[1])
+
+        result: Index | MultiIndex | ArrayLike
+        if self._is_multi:
+            # Preserve MultiIndex to avoid losing dtypes
+            result = self.append(other.take(missing))
+
+        else:
+            if len(missing) > 0:
+                other_diff = rvals.take(missing)
+                result = concat_compat((lvals, other_diff))
+            else:
+                result = lvals
+
+        if not self.is_monotonic_increasing or not other.is_monotonic_increasing:
+            # if both are monotonic then result should already be sorted
+            result = _maybe_try_sort(result, sort)
+
+        return result
+
+    @final
+    def _wrap_setop_result(self, other: Index, result) -> Index:
+        name = get_op_result_name(self, other)
+        if isinstance(result, Index):
+            if result.name != name:
+                result = result.rename(name)
+        else:
+            result = self._shallow_copy(result, name=name)
+        return result
+
+    @final
+    def intersection(self, other, sort: bool = False):
+        # default sort keyword is different here from other setops intentionally
+        #  done in GH#25063
+        """
+        Form the intersection of two Index objects.
+
+        This returns a new Index with elements common to the index and `other`.
+
+        Parameters
+        ----------
+        other : Index or array-like
+            An Index or an array-like object containing elements to form the
+            intersection with the original Index.
+        sort : True, False or None, default False
+            Whether to sort the resulting index.
+
+            * None : sort the result, except when `self` and `other` are equal
+              or when the values cannot be compared.
+            * False : do not sort the result.
+            * True : Sort the result (which may raise TypeError).
+
+        Returns
+        -------
+        Index
+            Returns a new Index object with elements common to both the original Index
+            and the `other` Index.
+
+        See Also
+        --------
+        Index.union : Form the union of two Index objects.
+        Index.difference : Return a new Index with elements of index not in other.
+        Index.isin : Return a boolean array where the index values are in values.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([1, 2, 3, 4])
+        >>> idx2 = pd.Index([3, 4, 5, 6])
+        >>> idx1.intersection(idx2)
+        Index([3, 4], dtype='int64')
+        """
+        self._validate_sort_keyword(sort)
+        self._assert_can_do_setop(other)
+        other, result_name = self._convert_can_do_setop(other)
+
+        if self.dtype != other.dtype:
+            self, other = self._dti_setop_align_tzs(other, "intersection")
+
+        if self.equals(other):
+            if not self.is_unique:
+                result = self.unique()._get_reconciled_name_object(other)
+            else:
+                result = self._get_reconciled_name_object(other)
+            if sort is True:
+                result = result.sort_values()
+            return result
+
+        if len(self) == 0 or len(other) == 0:
+            # fastpath; we need to be careful about having commutativity
+
+            if self._is_multi or other._is_multi:
+                # _convert_can_do_setop ensures that we have both or neither
+                # We retain self.levels
+                return self[:0].rename(result_name)
+
+            dtype = self._find_common_type_compat(other)
+            if self.dtype == dtype:
+                # Slicing allows us to retain DTI/TDI.freq, RangeIndex
+
+                # Note: self[:0] vs other[:0] affects
+                #  1) which index's `freq` we get in DTI/TDI cases
+                #     This may be a historical artifact, i.e. no documented
+                #     reason for this choice.
+                #  2) The `step` we get in RangeIndex cases
+                if len(self) == 0:
+                    return self[:0].rename(result_name)
+                else:
+                    return other[:0].rename(result_name)
+
+            return Index([], dtype=dtype, name=result_name)
+
+        elif not self._should_compare(other):
+            # We can infer that the intersection is empty.
+            if isinstance(self, ABCMultiIndex):
+                return self[:0].rename(result_name)
+            return Index([], name=result_name)
+
+        elif self.dtype != other.dtype:
+            dtype = self._find_common_type_compat(other)
+            this = self.astype(dtype, copy=False)
+            other = other.astype(dtype, copy=False)
+            return this.intersection(other, sort=sort)
+
+        result = self._intersection(other, sort=sort)
+        return self._wrap_intersection_result(other, result)
+
+    def _intersection(self, other: Index, sort: bool = False):
+        """
+        intersection specialized to the case with matching dtypes.
+        """
+        if self._can_use_libjoin and other._can_use_libjoin:
+            try:
+                res_indexer, indexer, _ = self._inner_indexer(other)
+            except TypeError:
+                # non-comparable; should only be for object dtype
+                pass
+            else:
+                # TODO: algos.unique1d should preserve DTA/TDA
+                if is_numeric_dtype(self.dtype):
+                    # This is faster, because Index.unique() checks for uniqueness
+                    # before calculating the unique values.
+                    res = algos.unique1d(res_indexer)
+                else:
+                    result = self.take(indexer)
+                    res = result.drop_duplicates()
+                return ensure_wrapped_if_datetimelike(res)
+
+        res_values = self._intersection_via_get_indexer(other, sort=sort)
+        res_values = _maybe_try_sort(res_values, sort)
+        return res_values
+
+    def _wrap_intersection_result(self, other, result):
+        # We will override for MultiIndex to handle empty results
+        return self._wrap_setop_result(other, result)
+
+    @final
+    def _intersection_via_get_indexer(
+        self, other: Index | MultiIndex, sort
+    ) -> ArrayLike | MultiIndex:
+        """
+        Find the intersection of two Indexes using get_indexer.
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray or MultiIndex
+            The returned array will be unique.
+        """
+        left_unique = self.unique()
+        right_unique = other.unique()
+
+        # even though we are unique, we need get_indexer_for for IntervalIndex
+        indexer = left_unique.get_indexer_for(right_unique)
+
+        mask = indexer != -1
+
+        taker = indexer.take(mask.nonzero()[0])
+        if sort is False:
+            # sort bc we want the elements in the same order they are in self
+            # unnecessary in the case with sort=None bc we will sort later
+            taker = np.sort(taker)
+
+        result: MultiIndex | ExtensionArray | np.ndarray
+        if isinstance(left_unique, ABCMultiIndex):
+            result = left_unique.take(taker)
+        else:
+            result = left_unique.take(taker)._values
+        return result
+
+    @final
+    def difference(self, other, sort: bool | None = None):
+        """
+        Return a new Index with elements of index not in `other`.
+
+        This is the set difference of two Index objects.
+
+        Parameters
+        ----------
+        other : Index or array-like
+            Index object or an array-like object containing elements to be compared
+            with the elements of the original Index.
+        sort : bool or None, default None
+            Whether to sort the resulting index. By default, the
+            values are attempted to be sorted, but any TypeError from
+            incomparable elements is caught by pandas.
+
+            * None : Attempt to sort the result, but catch any TypeErrors
+              from comparing incomparable elements.
+            * False : Do not sort the result.
+            * True : Sort the result (which may raise TypeError).
+
+        Returns
+        -------
+        Index
+            Returns a new Index object containing elements that are in the original
+            Index but not in the `other` Index.
+
+        See Also
+        --------
+        Index.symmetric_difference : Compute the symmetric difference of two Index
+            objects.
+        Index.intersection : Form the intersection of two Index objects.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([2, 1, 3, 4])
+        >>> idx2 = pd.Index([3, 4, 5, 6])
+        >>> idx1.difference(idx2)
+        Index([1, 2], dtype='int64')
+        >>> idx1.difference(idx2, sort=False)
+        Index([2, 1], dtype='int64')
+        """
+        self._validate_sort_keyword(sort)
+        self._assert_can_do_setop(other)
+        other, result_name = self._convert_can_do_setop(other)
+
+        # Note: we do NOT call _dti_setop_align_tzs here, as there
+        #  is no requirement that .difference be commutative, so it does
+        #  not cast to object.
+
+        if self.equals(other):
+            # Note: we do not (yet) sort even if sort=None GH#24959
+            return self[:0].rename(result_name)
+
+        if len(other) == 0:
+            # Note: we do not (yet) sort even if sort=None GH#24959
+            result = self.unique().rename(result_name)
+            if sort is True:
+                return result.sort_values()
+            return result
+
+        if not self._should_compare(other):
+            # Nothing matches -> difference is everything
+            result = self.unique().rename(result_name)
+            if sort is True:
+                return result.sort_values()
+            return result
+
+        result = self._difference(other, sort=sort)
+        return self._wrap_difference_result(other, result)
+
+    def _difference(self, other, sort):
+        # overridden by RangeIndex
+        this = self
+        if isinstance(self, ABCCategoricalIndex) and self.hasnans and other.hasnans:
+            this = this.dropna()
+        other = other.unique()
+        the_diff = this[other.get_indexer_for(this) == -1]
+        the_diff = the_diff if this.is_unique else the_diff.unique()
+        the_diff = _maybe_try_sort(the_diff, sort)
+        return the_diff
+
+    def _wrap_difference_result(self, other, result):
+        # We will override for MultiIndex to handle empty results
+        return self._wrap_setop_result(other, result)
+
+    def symmetric_difference(
+        self,
+        other,
+        result_name: abc.Hashable | None = None,
+        sort: bool | None = None,
+    ):
+        """
+        Compute the symmetric difference of two Index objects.
+
+        Parameters
+        ----------
+        other : Index or array-like
+            Index or an array-like object with elements to compute the symmetric
+            difference with the original Index.
+        result_name : str
+            A string representing the name of the resulting Index, if desired.
+        sort : bool or None, default None
+            Whether to sort the resulting index. By default, the
+            values are attempted to be sorted, but any TypeError from
+            incomparable elements is caught by pandas.
+
+            * None : Attempt to sort the result, but catch any TypeErrors
+              from comparing incomparable elements.
+            * False : Do not sort the result.
+            * True : Sort the result (which may raise TypeError).
+
+        Returns
+        -------
+        Index
+            Returns a new Index object containing elements that appear in either the
+            original Index or the `other` Index, but not both.
+
+        See Also
+        --------
+        Index.difference : Return a new Index with elements of index not in other.
+        Index.union : Form the union of two Index objects.
+        Index.intersection : Form the intersection of two Index objects.
+
+        Notes
+        -----
+        ``symmetric_difference`` contains elements that appear in either
+        ``idx1`` or ``idx2`` but not both. Equivalent to the Index created by
+        ``idx1.difference(idx2) | idx2.difference(idx1)`` with duplicates
+        dropped.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([1, 2, 3, 4])
+        >>> idx2 = pd.Index([2, 3, 4, 5])
+        >>> idx1.symmetric_difference(idx2)
+        Index([1, 5], dtype='int64')
+        """
+        self._validate_sort_keyword(sort)
+        self._assert_can_do_setop(other)
+        other, result_name_update = self._convert_can_do_setop(other)
+        if result_name is None:
+            result_name = result_name_update
+
+        if self.dtype != other.dtype:
+            self, other = self._dti_setop_align_tzs(other, "symmetric_difference")
+
+        if not self._should_compare(other):
+            return self.union(other, sort=sort).rename(result_name)
+
+        elif self.dtype != other.dtype:
+            dtype = self._find_common_type_compat(other)
+            this = self.astype(dtype, copy=False)
+            that = other.astype(dtype, copy=False)
+            return this.symmetric_difference(that, sort=sort).rename(result_name)
+
+        this = self.unique()
+        other = other.unique()
+        indexer = this.get_indexer_for(other)
+
+        # {this} minus {other}
+        common_indexer = indexer.take((indexer != -1).nonzero()[0])
+        left_indexer = np.setdiff1d(
+            np.arange(this.size), common_indexer, assume_unique=True
+        )
+        left_diff = this.take(left_indexer)
+
+        # {other} minus {this}
+        right_indexer = (indexer == -1).nonzero()[0]
+        right_diff = other.take(right_indexer)
+
+        res_values = left_diff.append(right_diff)
+        result = _maybe_try_sort(res_values, sort)
+
+        if not self._is_multi:
+            return Index(result, name=result_name, dtype=res_values.dtype)
+        else:
+            left_diff = cast("MultiIndex", left_diff)
+            if len(result) == 0:
+                # result might be an Index, if other was an Index
+                return left_diff.remove_unused_levels().set_names(result_name)
+            return result.set_names(result_name)
+
+    @final
+    def _assert_can_do_setop(self, other) -> bool:
+        if not is_list_like(other):
+            raise TypeError("Input must be Index or array-like")
+        return True
+
+    def _convert_can_do_setop(self, other) -> tuple[Index, Hashable]:
+        if not isinstance(other, Index):
+            other = Index(other, name=self.name)
+            result_name = self.name
+        else:
+            result_name = get_op_result_name(self, other)
+        return other, result_name
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    def get_loc(self, key):
+        """
+        Get integer location, slice or boolean mask for requested label.
+
+        Parameters
+        ----------
+        key : label
+            The key to check its location if it is present in the index.
+
+        Returns
+        -------
+        int if unique index, slice if monotonic index, else mask
+            Integer location, slice or boolean mask.
+
+        See Also
+        --------
+        Index.get_slice_bound : Calculate slice bound that corresponds to
+            given label.
+        Index.get_indexer : Computes indexer and mask for new index given
+            the current index.
+        Index.get_non_unique : Returns indexer and masks for new index given
+            the current index.
+        Index.get_indexer_for : Returns an indexer even when non-unique.
+
+        Examples
+        --------
+        >>> unique_index = pd.Index(list("abc"))
+        >>> unique_index.get_loc("b")
+        1
+
+        >>> monotonic_index = pd.Index(list("abbc"))
+        >>> monotonic_index.get_loc("b")
+        slice(1, 3, None)
+
+        >>> non_monotonic_index = pd.Index(list("abcb"))
+        >>> non_monotonic_index.get_loc("b")
+        array([False,  True, False,  True])
+        """
+        casted_key = self._maybe_cast_indexer(key)
+        try:
+            return self._engine.get_loc(casted_key)
+        except KeyError as err:
+            if isinstance(casted_key, slice) or (
+                isinstance(casted_key, abc.Iterable)
+                and any(isinstance(x, slice) for x in casted_key)
+            ):
+                raise InvalidIndexError(key) from err
+            raise KeyError(key) from err
+        except TypeError:
+            # If we have a listlike key, _check_indexing_error will raise
+            #  InvalidIndexError. Otherwise we fall through and re-raise
+            #  the TypeError.
+            self._check_indexing_error(key)
+            raise
+
+    @final
+    def get_indexer(
+        self,
+        target,
+        method: ReindexMethod | None = None,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> npt.NDArray[np.intp]:
+        """
+        Compute indexer and mask for new index given the current index.
+
+        The indexer should be then used as an input to ndarray.take to align the
+        current data to the new index.
+
+        Parameters
+        ----------
+        target : Index
+            An iterable containing the values to be used for computing indexer.
+        method : {None, 'pad'/'ffill', 'backfill'/'bfill', 'nearest'}, optional
+            * default: exact matches only.
+            * pad / ffill: find the PREVIOUS index value if no exact match.
+            * backfill / bfill: use NEXT index value if no exact match
+            * nearest: use the NEAREST index value if no exact match. Tied
+              distances are broken by preferring the larger index value.
+        limit : int, optional
+            Maximum number of consecutive labels in ``target`` to match for
+            inexact matches.
+        tolerance : optional
+            Maximum distance between original and new labels for inexact
+            matches. The values of the index at the matching locations must
+            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
+
+            Tolerance may be a scalar value, which applies the same tolerance
+            to all values, or list-like, which applies variable tolerance per
+            element. List-like includes list, tuple, array, Series, and must be
+            the same size as the index and its dtype must exactly match the
+            index's type.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            Integers from 0 to n - 1 indicating that the index at these
+            positions matches the corresponding target values. Missing values
+            in the target are marked by -1.
+
+        See Also
+        --------
+        Index.get_indexer_for : Returns an indexer even when non-unique.
+        Index.get_non_unique : Returns indexer and masks for new index given
+            the current index.
+
+        Notes
+        -----
+        Returns -1 for unmatched values, for further explanation see the
+        example below.
+
+        Examples
+        --------
+        >>> index = pd.Index(["c", "a", "b"])
+        >>> index.get_indexer(["a", "b", "x"])
+        array([ 1,  2, -1])
+
+        Notice that the return value is an array of locations in ``index``
+        and ``x`` is marked by -1, as it is not in ``index``.
+        """
+        method = clean_reindex_fill_method(method)
+        orig_target = target
+        target = self._maybe_cast_listlike_indexer(target)
+
+        self._check_indexing_method(method, limit, tolerance)
+
+        if not self._index_as_unique:
+            raise InvalidIndexError(self._requires_unique_msg)
+
+        if len(target) == 0:
+            return np.array([], dtype=np.intp)
+
+        if not self._should_compare(target) and not self._should_partial_index(target):
+            # IntervalIndex get special treatment bc numeric scalars can be
+            #  matched to Interval scalars
+            return self._get_indexer_non_comparable(target, method=method, unique=True)
+
+        if isinstance(self.dtype, CategoricalDtype):
+            # _maybe_cast_listlike_indexer ensures target has our dtype
+            #  (could improve perf by doing _should_compare check earlier?)
+            assert self.dtype == target.dtype
+
+            indexer = self._engine.get_indexer(target.codes)
+            if self.hasnans and target.hasnans:
+                # After _maybe_cast_listlike_indexer, target elements which do not
+                # belong to some category are changed to NaNs
+                # Mask to track actual NaN values compared to inserted NaN values
+                # GH#45361
+                target_nans = isna(orig_target)
+                loc = self.get_loc(np.nan)
+                mask = target.isna()
+                indexer[target_nans] = loc
+                indexer[mask & ~target_nans] = -1
+            return indexer
+
+        if isinstance(target.dtype, CategoricalDtype):
+            # potential fastpath
+            # get an indexer for unique categories then propagate to codes via take_nd
+            # get_indexer instead of _get_indexer needed for MultiIndex cases
+            #  e.g. test_append_different_columns_types
+            categories_indexer = self.get_indexer(target.categories)
+
+            indexer = algos.take_nd(categories_indexer, target.codes, fill_value=-1)
+
+            if (not self._is_multi and self.hasnans) and target.hasnans:
+                # Exclude MultiIndex because hasnans raises NotImplementedError
+                # we should only get here if we are unique, so loc is an integer
+                # GH#41934
+                loc = self.get_loc(np.nan)
+                mask = target.isna()
+                indexer[mask] = loc
+
+            return ensure_platform_int(indexer)
+
+        pself, ptarget = self._maybe_downcast_for_indexing(target)
+        if pself is not self or ptarget is not target:
+            return pself.get_indexer(
+                ptarget, method=method, limit=limit, tolerance=tolerance
+            )
+
+        if self.dtype == target.dtype and self.equals(target):
+            # Only call equals if we have same dtype to avoid inference/casting
+            return np.arange(len(target), dtype=np.intp)
+
+        if self.dtype != target.dtype and not self._should_partial_index(target):
+            # _should_partial_index e.g. IntervalIndex with numeric scalars
+            #  that can be matched to Interval scalars.
+            dtype = self._find_common_type_compat(target)
+
+            this = self.astype(dtype, copy=False)
+            target = target.astype(dtype, copy=False)
+            return this._get_indexer(
+                target, method=method, limit=limit, tolerance=tolerance
+            )
+
+        return self._get_indexer(target, method, limit, tolerance)
+
+    def _get_indexer(
+        self,
+        target: Index,
+        method: str_t | None = None,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> npt.NDArray[np.intp]:
+        if tolerance is not None:
+            tolerance = self._convert_tolerance(tolerance, target)
+
+        if method in ["pad", "backfill"]:
+            indexer = self._get_fill_indexer(target, method, limit, tolerance)
+        elif method == "nearest":
+            indexer = self._get_nearest_indexer(target, limit, tolerance)
+        else:
+            if target._is_multi and self._is_multi:
+                engine = self._engine
+                # error: Item "IndexEngine" of "Union[IndexEngine, ExtensionEngine]"
+                # has no attribute "_extract_level_codes"
+                tgt_values = engine._extract_level_codes(  # type: ignore[union-attr]
+                    target
+                )
+            else:
+                tgt_values = target._get_engine_target()
+
+            indexer = self._engine.get_indexer(tgt_values)
+
+        return ensure_platform_int(indexer)
+
+    @final
+    def _should_partial_index(self, target: Index) -> bool:
+        """
+        Should we attempt partial-matching indexing?
+        """
+        if isinstance(self.dtype, IntervalDtype):
+            if isinstance(target.dtype, IntervalDtype):
+                return False
+            # "Index" has no attribute "left"
+            return self.left._should_compare(target)  # type: ignore[attr-defined]
+        return False
+
+    @final
+    def _check_indexing_method(
+        self,
+        method: str_t | None,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> None:
+        """
+        Raise if we have a get_indexer `method` that is not supported or valid.
+        """
+        if method not in [None, "bfill", "backfill", "pad", "ffill", "nearest"]:
+            # in practice the clean_reindex_fill_method call would raise
+            #  before we get here
+            raise ValueError("Invalid fill method")  # pragma: no cover
+
+        if self._is_multi:
+            if method == "nearest":
+                raise NotImplementedError(
+                    "method='nearest' not implemented yet "
+                    "for MultiIndex; see GitHub issue 9365"
+                )
+            if method in ("pad", "backfill"):
+                if tolerance is not None:
+                    raise NotImplementedError(
+                        "tolerance not implemented yet for MultiIndex"
+                    )
+
+        if isinstance(self.dtype, (IntervalDtype, CategoricalDtype)):
+            # GH#37871 for now this is only for IntervalIndex and CategoricalIndex
+            if method is not None:
+                raise NotImplementedError(
+                    f"method {method} not yet implemented for {type(self).__name__}"
+                )
+
+        if method is None:
+            if tolerance is not None:
+                raise ValueError(
+                    "tolerance argument only valid if doing pad, "
+                    "backfill or nearest reindexing"
+                )
+            if limit is not None:
+                raise ValueError(
+                    "limit argument only valid if doing pad, "
+                    "backfill or nearest reindexing"
+                )
+
+    def _convert_tolerance(self, tolerance, target: np.ndarray | Index) -> np.ndarray:
+        # override this method on subclasses
+        tolerance = np.asarray(tolerance)
+        if target.size != tolerance.size and tolerance.size > 1:
+            raise ValueError("list-like tolerance size must match target index size")
+        elif is_numeric_dtype(self) and not np.issubdtype(tolerance.dtype, np.number):
+            if tolerance.ndim > 0:
+                raise ValueError(
+                    f"tolerance argument for {type(self).__name__} with dtype "
+                    f"{self.dtype} must contain numeric elements if it is list type"
+                )
+
+            raise ValueError(
+                f"tolerance argument for {type(self).__name__} with dtype {self.dtype} "
+                f"must be numeric if it is a scalar: {tolerance!r}"
+            )
+        return tolerance
+
+    @final
+    def _get_fill_indexer(
+        self, target: Index, method: str_t, limit: int | None = None, tolerance=None
+    ) -> npt.NDArray[np.intp]:
+        if self._is_multi:
+            if not (self.is_monotonic_increasing or self.is_monotonic_decreasing):
+                raise ValueError("index must be monotonic increasing or decreasing")
+            encoded = self.append(target)._engine.values  # type: ignore[union-attr]
+            self_encoded = Index(encoded[: len(self)])
+            target_encoded = Index(encoded[len(self) :])
+            return self_encoded._get_fill_indexer(
+                target_encoded, method, limit, tolerance
+            )
+
+        if self.is_monotonic_increasing and target.is_monotonic_increasing:
+            target_values = target._get_engine_target()
+            own_values = self._get_engine_target()
+            if not isinstance(target_values, np.ndarray) or not isinstance(
+                own_values, np.ndarray
+            ):
+                raise NotImplementedError
+
+            if method == "pad":
+                indexer = libalgos.pad(own_values, target_values, limit=limit)
+            else:
+                # i.e. "backfill"
+                indexer = libalgos.backfill(own_values, target_values, limit=limit)
+        else:
+            indexer = self._get_fill_indexer_searchsorted(target, method, limit)
+        if tolerance is not None and len(self):
+            indexer = self._filter_indexer_tolerance(target, indexer, tolerance)
+        return indexer
+
+    @final
+    def _get_fill_indexer_searchsorted(
+        self, target: Index, method: str_t, limit: int | None = None
+    ) -> npt.NDArray[np.intp]:
+        """
+        Fallback pad/backfill get_indexer that works for monotonic decreasing
+        indexes and non-monotonic targets.
+        """
+        if limit is not None:
+            raise ValueError(
+                f"limit argument for {method!r} method only well-defined "
+                "if index and target are monotonic"
+            )
+
+        side: Literal["left", "right"] = "left" if method == "pad" else "right"
+
+        # find exact matches first (this simplifies the algorithm)
+        indexer = self.get_indexer(target)
+        nonexact = indexer == -1
+        indexer[nonexact] = self._searchsorted_monotonic(target[nonexact], side)
+        if side == "left":
+            # searchsorted returns "indices into a sorted array such that,
+            # if the corresponding elements in v were inserted before the
+            # indices, the order of a would be preserved".
+            # Thus, we need to subtract 1 to find values to the left.
+            indexer[nonexact] -= 1
+            # This also mapped not found values (values of 0 from
+            # np.searchsorted) to -1, which conveniently is also our
+            # sentinel for missing values
+        else:
+            # Mark indices to the right of the largest value as not found
+            indexer[indexer == len(self)] = -1
+        return indexer
+
+    @final
+    def _get_nearest_indexer(
+        self, target: Index, limit: int | None, tolerance
+    ) -> npt.NDArray[np.intp]:
+        """
+        Get the indexer for the nearest index labels; requires an index with
+        values that can be subtracted from each other (e.g., not strings or
+        tuples).
+        """
+        if not len(self):
+            return self._get_fill_indexer(target, "pad")
+
+        left_indexer = self.get_indexer(target, "pad", limit=limit)
+        right_indexer = self.get_indexer(target, "backfill", limit=limit)
+
+        left_distances = self._difference_compat(target, left_indexer)
+        right_distances = self._difference_compat(target, right_indexer)
+
+        op = operator.lt if self.is_monotonic_increasing else operator.le
+        indexer = np.where(
+            # error: Argument 1&2 has incompatible type "Union[ExtensionArray,
+            # ndarray[Any, Any]]"; expected "Union[SupportsDunderLE,
+            # SupportsDunderGE, SupportsDunderGT, SupportsDunderLT]"
+            op(left_distances, right_distances)  # type: ignore[arg-type]
+            | (right_indexer == -1),
+            left_indexer,
+            right_indexer,
+        )
+        if tolerance is not None:
+            indexer = self._filter_indexer_tolerance(target, indexer, tolerance)
+        return indexer
+
+    @final
+    def _filter_indexer_tolerance(
+        self,
+        target: Index,
+        indexer: npt.NDArray[np.intp],
+        tolerance,
+    ) -> npt.NDArray[np.intp]:
+        distance = self._difference_compat(target, indexer)
+
+        return np.where(distance <= tolerance, indexer, -1)
+
+    @final
+    def _difference_compat(
+        self, target: Index, indexer: npt.NDArray[np.intp]
+    ) -> ArrayLike:
+        # Compatibility for PeriodArray, for which __sub__ returns an ndarray[object]
+        #  of DateOffset objects, which do not support __abs__ (and would be slow
+        #  if they did)
+
+        if isinstance(self.dtype, PeriodDtype):
+            # Note: we only get here with matching dtypes
+            own_values = cast("PeriodArray", self._data)._ndarray
+            target_values = cast("PeriodArray", target._data)._ndarray
+            diff = own_values[indexer] - target_values
+        else:
+            # error: Unsupported left operand type for - ("ExtensionArray")
+            diff = self._values[indexer] - target._values  # type: ignore[operator]
+        return abs(diff)
+
+    # --------------------------------------------------------------------
+    # Indexer Conversion Methods
+
+    @final
+    def _validate_positional_slice(self, key: slice) -> None:
+        """
+        For positional indexing, a slice must have either int or None
+        for each of start, stop, and step.
+        """
+        self._validate_indexer("positional", key.start, "iloc")
+        self._validate_indexer("positional", key.stop, "iloc")
+        self._validate_indexer("positional", key.step, "iloc")
+
+    def _convert_slice_indexer(self, key: slice, kind: Literal["loc", "getitem"]):
+        """
+        Convert a slice indexer.
+
+        By definition, these are labels unless 'iloc' is passed in.
+        Floats are not allowed as the start, step, or stop of the slice.
+
+        Parameters
+        ----------
+        key : label of the slice bound
+        kind : {'loc', 'getitem'}
+        """
+
+        # potentially cast the bounds to integers
+        start, stop, step = key.start, key.stop, key.step
+
+        # figure out if this is a positional indexer
+        is_index_slice = is_valid_positional_slice(key)
+
+        # TODO(GH#50617): once Series.__[gs]etitem__ is removed we should be able
+        #  to simplify this.
+        if kind == "getitem":
+            # called from the getitem slicers, validate that we are in fact integers
+            if is_index_slice:
+                # In this case the _validate_indexer checks below are redundant
+                return key
+            elif self.dtype.kind in "iu":
+                # Note: these checks are redundant if we know is_index_slice
+                self._validate_indexer("slice", key.start, "getitem")
+                self._validate_indexer("slice", key.stop, "getitem")
+                self._validate_indexer("slice", key.step, "getitem")
+                return key
+
+        # convert the slice to an indexer here; checking that the user didn't
+        #  pass a positional slice to loc
+        is_positional = is_index_slice and self._should_fallback_to_positional
+
+        # if we are mixed and have integers
+        if is_positional:
+            try:
+                # Validate start & stop
+                if start is not None:
+                    self.get_loc(start)
+                if stop is not None:
+                    self.get_loc(stop)
+                is_positional = False
+            except KeyError:
+                pass
+
+        if com.is_null_slice(key):
+            # It doesn't matter if we are positional or label based
+            indexer = key
+        elif is_positional:
+            if kind == "loc":
+                # GH#16121, GH#24612, GH#31810
+                raise TypeError(
+                    "Slicing a positional slice with .loc is not allowed, "
+                    "Use .loc with labels or .iloc with positions instead.",
+                )
+            indexer = key
+        else:
+            indexer = self.slice_indexer(start, stop, step)
+
+        return indexer
+
+    @final
+    def _raise_invalid_indexer(
+        self,
+        form: Literal["slice", "positional"],
+        key,
+        reraise: lib.NoDefault | None | Exception = lib.no_default,
+    ) -> None:
+        """
+        Raise consistent invalid indexer message.
+        """
+        msg = (
+            f"cannot do {form} indexing on {type(self).__name__} with these "
+            f"indexers [{key}] of type {type(key).__name__}"
+        )
+        if reraise is not lib.no_default:
+            raise TypeError(msg) from reraise
+        raise TypeError(msg)
+
+    # --------------------------------------------------------------------
+    # Reindex Methods
+
+    @final
+    def _validate_can_reindex(self, indexer: np.ndarray) -> None:
+        """
+        Check if we are allowing reindexing with this particular indexer.
+
+        Parameters
+        ----------
+        indexer : an integer ndarray
+
+        Raises
+        ------
+        ValueError if its a duplicate axis
+        """
+        # trying to reindex on an axis with duplicates
+        if not self._index_as_unique and len(indexer):
+            raise ValueError("cannot reindex on an axis with duplicate labels")
+
+    def reindex(
+        self,
+        target,
+        method: ReindexMethod | None = None,
+        level=None,
+        limit: int | None = None,
+        tolerance: float | None = None,
+    ) -> tuple[Index, npt.NDArray[np.intp] | None]:
+        """
+        Create index with target's values.
+
+        Parameters
+        ----------
+        target : an iterable
+            An iterable containing the values to be used for creating the new index.
+        method : {None, 'pad'/'ffill', 'backfill'/'bfill', 'nearest'}, optional
+            * default: exact matches only.
+            * pad / ffill: find the PREVIOUS index value if no exact match.
+            * backfill / bfill: use NEXT index value if no exact match
+            * nearest: use the NEAREST index value if no exact match. Tied
+              distances are broken by preferring the larger index value.
+        level : int, optional
+            Level of multiindex.
+        limit : int, optional
+            Maximum number of consecutive labels in ``target`` to match for
+            inexact matches.
+        tolerance : int or float, optional
+            Maximum distance between original and new labels for inexact
+            matches. The values of the index at the matching locations must
+            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
+
+            Tolerance may be a scalar value, which applies the same tolerance
+            to all values, or list-like, which applies variable tolerance per
+            element. List-like includes list, tuple, array, Series, and must be
+            the same size as the index and its dtype must exactly match the
+            index's type.
+
+        Returns
+        -------
+        new_index : pd.Index
+            Resulting index.
+        indexer : np.ndarray[np.intp] or None
+            Indices of output values in original index.
+
+        Raises
+        ------
+        TypeError
+            If ``method`` passed along with ``level``.
+        ValueError
+            If non-unique multi-index
+        ValueError
+            If non-unique index and ``method`` or ``limit`` passed.
+
+        See Also
+        --------
+        Series.reindex : Conform Series to new index with optional filling logic.
+        DataFrame.reindex : Conform DataFrame to new index with optional filling logic.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["car", "bike", "train", "tractor"])
+        >>> idx
+        Index(['car', 'bike', 'train', 'tractor'], dtype='object')
+        >>> idx.reindex(["car", "bike"])
+        (Index(['car', 'bike'], dtype='object'), array([0, 1]))
+        """
+        # GH6552: preserve names when reindexing to non-named target
+        # (i.e. neither Index nor Series).
+        preserve_names = not hasattr(target, "name")
+
+        # GH7774: preserve dtype/tz if target is empty and not an Index.
+        if is_iterator(target):
+            target = list(target)
+
+        if not isinstance(target, Index) and len(target) == 0:
+            if level is not None and self._is_multi:
+                # "Index" has no attribute "levels"; maybe "nlevels"?
+                idx = self.levels[level]  # type: ignore[attr-defined]
+            else:
+                idx = self
+            target = idx[:0]
+        else:
+            target = ensure_index(target)
+
+        if level is not None and (
+            isinstance(self, ABCMultiIndex) or isinstance(target, ABCMultiIndex)
+        ):
+            if method is not None:
+                raise TypeError("Fill method not supported if level passed")
+
+            # TODO: tests where passing `keep_order=not self._is_multi`
+            #  makes a difference for non-MultiIndex case
+            target, indexer, _ = self._join_level(
+                target, level, how="right", keep_order=not self._is_multi
+            )
+
+        else:
+            if self.equals(target):
+                indexer = None
+            else:
+                if self._index_as_unique:
+                    indexer = self.get_indexer(
+                        target, method=method, limit=limit, tolerance=tolerance
+                    )
+                elif self._is_multi:
+                    raise ValueError("cannot handle a non-unique multi-index!")
+                elif not self.is_unique:
+                    # GH#42568
+                    raise ValueError("cannot reindex on an axis with duplicate labels")
+                else:
+                    indexer, _ = self.get_indexer_non_unique(target)
+
+        target = self._wrap_reindex_result(target, indexer, preserve_names)
+        return target, indexer
+
+    def _wrap_reindex_result(self, target, indexer, preserve_names: bool):
+        target = self._maybe_preserve_names(target, preserve_names)
+        return target
+
+    def _maybe_preserve_names(self, target: IndexT, preserve_names: bool) -> IndexT:
+        if preserve_names and target.nlevels == 1 and target.name != self.name:
+            target = target.copy(deep=False)
+            target.name = self.name
+        return target
+
+    @final
+    def _reindex_non_unique(
+        self, target: Index
+    ) -> tuple[Index, npt.NDArray[np.intp], npt.NDArray[np.intp] | None]:
+        """
+        Create a new index with target's values (move/add/delete values as
+        necessary) use with non-unique Index and a possibly non-unique target.
+
+        Parameters
+        ----------
+        target : an iterable
+
+        Returns
+        -------
+        new_index : pd.Index
+            Resulting index.
+        indexer : np.ndarray[np.intp]
+            Indices of output values in original index.
+        new_indexer : np.ndarray[np.intp] or None
+
+        """
+        target = ensure_index(target)
+        if len(target) == 0:
+            # GH#13691
+            return self[:0], np.array([], dtype=np.intp), None
+
+        indexer, missing = self.get_indexer_non_unique(target)
+        check = indexer != -1
+        new_labels: Index | np.ndarray = self.take(indexer[check])
+        new_indexer = None
+
+        if len(missing):
+            length = np.arange(len(indexer), dtype=np.intp)
+
+            missing = ensure_platform_int(missing)
+            missing_labels = target.take(missing)
+            missing_indexer = length[~check]
+            cur_labels = self.take(indexer[check]).values
+            cur_indexer = length[check]
+
+            # Index constructor below will do inference
+            new_labels = np.empty((len(indexer),), dtype=object)
+            new_labels[cur_indexer] = cur_labels
+            new_labels[missing_indexer] = missing_labels
+
+            # GH#38906
+            if not len(self):
+                new_indexer = np.arange(0, dtype=np.intp)
+
+            # a unique indexer
+            elif target.is_unique:
+                # see GH5553, make sure we use the right indexer
+                new_indexer = np.arange(len(indexer), dtype=np.intp)
+                new_indexer[cur_indexer] = np.arange(len(cur_labels))
+                new_indexer[missing_indexer] = -1
+
+            # we have a non_unique selector, need to use the original
+            # indexer here
+            else:
+                # need to retake to have the same size as the indexer
+                indexer[~check] = -1
+
+                # reset the new indexer to account for the new size
+                new_indexer = np.arange(len(self.take(indexer)), dtype=np.intp)
+                new_indexer[~check] = -1
+
+        if not isinstance(self, ABCMultiIndex):
+            new_index = Index(new_labels, name=self.name)
+        else:
+            new_index = type(self).from_tuples(new_labels, names=self.names)
+        return new_index, indexer, new_indexer
+
+    # --------------------------------------------------------------------
+    # Join Methods
+
+    @overload
+    def join(
+        self,
+        other: Index,
+        *,
+        how: JoinHow = ...,
+        level: Level = ...,
+        return_indexers: Literal[True],
+        sort: bool = ...,
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]: ...
+
+    @overload
+    def join(
+        self,
+        other: Index,
+        *,
+        how: JoinHow = ...,
+        level: Level = ...,
+        return_indexers: Literal[False] = ...,
+        sort: bool = ...,
+    ) -> Index: ...
+
+    @overload
+    def join(
+        self,
+        other: Index,
+        *,
+        how: JoinHow = ...,
+        level: Level = ...,
+        return_indexers: bool = ...,
+        sort: bool = ...,
+    ) -> (
+        Index | tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]
+    ): ...
+
+    @final
+    @_maybe_return_indexers
+    def join(
+        self,
+        other: Index,
+        *,
+        how: JoinHow = "left",
+        level: Level | None = None,
+        return_indexers: bool = False,
+        sort: bool = False,
+    ) -> Index | tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        """
+        Compute join_index and indexers to conform data structures to the new index.
+
+        Parameters
+        ----------
+        other : Index
+            The other index on which join is performed.
+        how : {'left', 'right', 'inner', 'outer'}
+        level : int or level name, default None
+            It is either the integer position or the name of the level.
+        return_indexers : bool, default False
+            Whether to return the indexers or not for both the index objects.
+        sort : bool, default False
+            Sort the join keys lexicographically in the result Index. If False,
+            the order of the join keys depends on the join type (how keyword).
+
+        Returns
+        -------
+        join_index, (left_indexer, right_indexer)
+            The new index.
+
+        See Also
+        --------
+        DataFrame.join : Join columns with `other` DataFrame either on index
+            or on a key.
+        DataFrame.merge : Merge DataFrame or named Series objects with a
+            database-style join.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([1, 2, 3])
+        >>> idx2 = pd.Index([4, 5, 6])
+        >>> idx1.join(idx2, how="outer")
+        Index([1, 2, 3, 4, 5, 6], dtype='int64')
+        >>> idx1.join(other=idx2, how="outer", return_indexers=True)
+        (Index([1, 2, 3, 4, 5, 6], dtype='int64'),
+        array([ 0,  1,  2, -1, -1, -1]), array([-1, -1, -1,  0,  1,  2]))
+        """
+        other = ensure_index(other)
+        sort = sort or how == "outer"
+
+        if isinstance(self, ABCDatetimeIndex) and isinstance(other, ABCDatetimeIndex):
+            if (self.tz is None) ^ (other.tz is None):
+                # Raise instead of casting to object below.
+                raise TypeError("Cannot join tz-naive with tz-aware DatetimeIndex")
+
+        if not self._is_multi and not other._is_multi:
+            # We have specific handling for MultiIndex below
+            pself, pother = self._maybe_downcast_for_indexing(other)
+            if pself is not self or pother is not other:
+                return pself.join(
+                    pother, how=how, level=level, return_indexers=True, sort=sort
+                )
+
+        # try to figure out the join level
+        # GH3662
+        if level is None and (self._is_multi or other._is_multi):
+            # have the same levels/names so a simple join
+            if self.names == other.names:
+                pass
+            else:
+                return self._join_multi(other, how=how)
+
+        # join on the level
+        if level is not None and (self._is_multi or other._is_multi):
+            return self._join_level(other, level, how=how)
+
+        if len(self) == 0 or len(other) == 0:
+            try:
+                return self._join_empty(other, how, sort)
+            except TypeError:
+                # object dtype; non-comparable objects
+                pass
+
+        if self.dtype != other.dtype:
+            dtype = self._find_common_type_compat(other)
+            this = self.astype(dtype, copy=False)
+            other = other.astype(dtype, copy=False)
+            return this.join(other, how=how, return_indexers=True)
+        elif (
+            isinstance(self, ABCCategoricalIndex)
+            and isinstance(other, ABCCategoricalIndex)
+            and not self.ordered
+            and not self.categories.equals(other.categories)
+        ):
+            # dtypes are "equal" but categories are in different order
+            other = Index(other._values.reorder_categories(self.categories))
+
+        _validate_join_method(how)
+
+        if (
+            self.is_monotonic_increasing
+            and other.is_monotonic_increasing
+            and self._can_use_libjoin
+            and other._can_use_libjoin
+            and (self.is_unique or other.is_unique)
+        ):
+            try:
+                return self._join_monotonic(other, how=how)
+            except TypeError:
+                # object dtype; non-comparable objects
+                pass
+        elif not self.is_unique or not other.is_unique:
+            return self._join_non_unique(other, how=how, sort=sort)
+
+        return self._join_via_get_indexer(other, how, sort)
+
+    def _join_empty(
+        self, other: Index, how: JoinHow, sort: bool
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        assert len(self) == 0 or len(other) == 0
+        _validate_join_method(how)
+
+        lidx: np.ndarray | None
+        ridx: np.ndarray | None
+
+        if len(other):
+            how = cast(JoinHow, {"left": "right", "right": "left"}.get(how, how))
+            join_index, ridx, lidx = other._join_empty(self, how, sort)
+        elif how in ["left", "outer"]:
+            if sort and not self.is_monotonic_increasing:
+                lidx = self.argsort()
+                join_index = self.take(lidx)
+            else:
+                lidx = None
+                join_index = self._view()
+            ridx = np.broadcast_to(np.intp(-1), len(join_index))
+        else:
+            join_index = other._view()
+            lidx = np.array([], dtype=np.intp)
+            ridx = None
+        return join_index, lidx, ridx
+
+    @final
+    def _join_via_get_indexer(
+        self, other: Index, how: JoinHow, sort: bool
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        # Fallback if we do not have any fastpaths available based on
+        #  uniqueness/monotonicity
+
+        # Note: at this point we have checked matching dtypes
+        lindexer: npt.NDArray[np.intp] | None
+        rindexer: npt.NDArray[np.intp] | None
+
+        if how == "left":
+            if sort:
+                join_index, lindexer = self.sort_values(return_indexer=True)
+                rindexer = other.get_indexer_for(join_index)
+                return join_index, lindexer, rindexer
+            else:
+                join_index = self
+        elif how == "right":
+            if sort:
+                join_index, rindexer = other.sort_values(return_indexer=True)
+                lindexer = self.get_indexer_for(join_index)
+                return join_index, lindexer, rindexer
+            else:
+                join_index = other
+        elif how == "inner":
+            join_index = self.intersection(other, sort=sort)
+        elif how == "outer":
+            try:
+                join_index = self.union(other, sort=sort)
+            except TypeError:
+                join_index = self.union(other)
+                try:
+                    join_index = _maybe_try_sort(join_index, sort)
+                except TypeError:
+                    pass
+
+        names = other.names if how == "right" else self.names
+        if join_index.names != names:
+            join_index = join_index.set_names(names)
+
+        if join_index is self:
+            lindexer = None
+        else:
+            lindexer = self.get_indexer_for(join_index)
+        if join_index is other:
+            rindexer = None
+        else:
+            rindexer = other.get_indexer_for(join_index)
+        return join_index, lindexer, rindexer
+
+    @final
+    def _join_multi(self, other: Index, how: JoinHow):
+        from pandas.core.indexes.multi import MultiIndex
+        from pandas.core.reshape.merge import restore_dropped_levels_multijoin
+
+        # figure out join names
+        self_names_list = list(self.names)
+        other_names_list = list(other.names)
+        self_names_order = self_names_list.index
+        other_names_order = other_names_list.index
+        self_names = set(self_names_list)
+        other_names = set(other_names_list)
+        overlap = self_names & other_names
+
+        # need at least 1 in common
+        if not overlap:
+            raise ValueError("cannot join with no overlapping index names")
+
+        if isinstance(self, MultiIndex) and isinstance(other, MultiIndex):
+            # Drop the non-matching levels from left and right respectively
+            ldrop_names = sorted(self_names - overlap, key=self_names_order)
+            rdrop_names = sorted(other_names - overlap, key=other_names_order)
+
+            # if only the order differs
+            if not len(ldrop_names + rdrop_names):
+                self_jnlevels = self
+                other_jnlevels = other.reorder_levels(self.names)
+            else:
+                self_jnlevels = self.droplevel(ldrop_names)
+                other_jnlevels = other.droplevel(rdrop_names)
+
+            # Join left and right
+            # Join on same leveled multi-index frames is supported
+            join_idx, lidx, ridx = self_jnlevels.join(
+                other_jnlevels, how=how, return_indexers=True
+            )
+
+            # Restore the dropped levels
+            # Returned index level order is
+            # common levels, ldrop_names, rdrop_names
+            dropped_names = ldrop_names + rdrop_names
+
+            # error: Argument 5/6 to "restore_dropped_levels_multijoin" has
+            # incompatible type "Optional[ndarray[Any, dtype[signedinteger[Any
+            # ]]]]"; expected "ndarray[Any, dtype[signedinteger[Any]]]"
+            levels, codes, names = restore_dropped_levels_multijoin(
+                self,
+                other,
+                dropped_names,
+                join_idx,
+                lidx,  # type: ignore[arg-type]
+                ridx,  # type: ignore[arg-type]
+            )
+
+            # Re-create the multi-index
+            multi_join_idx = MultiIndex(
+                levels=levels, codes=codes, names=names, verify_integrity=False
+            )
+
+            multi_join_idx = multi_join_idx.remove_unused_levels()
+
+            # maintain the order of the index levels
+            if how == "right":
+                level_order = other_names_list + ldrop_names
+            else:
+                level_order = self_names_list + rdrop_names
+            multi_join_idx = multi_join_idx.reorder_levels(level_order)
+
+            return multi_join_idx, lidx, ridx
+
+        jl = next(iter(overlap))
+
+        # Case where only one index is multi
+        # make the indices into mi's that match
+        flip_order = False
+        if isinstance(self, MultiIndex):
+            self, other = other, self
+            flip_order = True
+            # flip if join method is right or left
+            flip: dict[JoinHow, JoinHow] = {"right": "left", "left": "right"}
+            how = flip.get(how, how)
+
+        level = other.names.index(jl)
+        result = self._join_level(other, level, how=how)
+
+        if flip_order:
+            return result[0], result[2], result[1]
+        return result
+
+    @final
+    def _join_non_unique(
+        self, other: Index, how: JoinHow = "left", sort: bool = False
+    ) -> tuple[Index, npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        from pandas.core.reshape.merge import get_join_indexers_non_unique
+
+        # We only get here if dtypes match
+        assert self.dtype == other.dtype
+
+        left_idx, right_idx = get_join_indexers_non_unique(
+            self._values, other._values, how=how, sort=sort
+        )
+
+        if how == "right":
+            join_index = other.take(right_idx)
+        else:
+            join_index = self.take(left_idx)
+
+        if how == "outer":
+            mask = left_idx == -1
+            if mask.any():
+                right = other.take(right_idx)
+                join_index = join_index.putmask(mask, right)
+
+        if isinstance(join_index, ABCMultiIndex) and how == "outer":
+            # test_join_index_levels
+            join_index = join_index._sort_levels_monotonic()
+        return join_index, left_idx, right_idx
+
+    @final
+    def _join_level(
+        self, other: Index, level, how: JoinHow = "left", keep_order: bool = True
+    ) -> tuple[MultiIndex, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        """
+        The join method *only* affects the level of the resulting
+        MultiIndex. Otherwise it just exactly aligns the Index data to the
+        labels of the level in the MultiIndex.
+
+        If ```keep_order == True```, the order of the data indexed by the
+        MultiIndex will not be changed; otherwise, it will tie out
+        with `other`.
+        """
+        from pandas.core.indexes.multi import MultiIndex
+
+        def _get_leaf_sorter(labels: list[np.ndarray]) -> npt.NDArray[np.intp]:
+            """
+            Returns sorter for the inner most level while preserving the
+            order of higher levels.
+
+            Parameters
+            ----------
+            labels : list[np.ndarray]
+                Each ndarray has signed integer dtype, not necessarily identical.
+
+            Returns
+            -------
+            np.ndarray[np.intp]
+            """
+            if labels[0].size == 0:
+                return np.empty(0, dtype=np.intp)
+
+            if len(labels) == 1:
+                return get_group_index_sorter(ensure_platform_int(labels[0]))
+
+            # find indexers of beginning of each set of
+            # same-key labels w.r.t all but last level
+            tic = labels[0][:-1] != labels[0][1:]
+            for lab in labels[1:-1]:
+                tic |= lab[:-1] != lab[1:]
+
+            starts = np.hstack(([True], tic, [True])).nonzero()[0]
+            lab = ensure_int64(labels[-1])
+            return lib.get_level_sorter(lab, ensure_platform_int(starts))
+
+        if isinstance(self, MultiIndex) and isinstance(other, MultiIndex):
+            raise TypeError("Join on level between two MultiIndex objects is ambiguous")
+
+        left, right = self, other
+
+        flip_order = not isinstance(self, MultiIndex)
+        if flip_order:
+            left, right = right, left
+            flip: dict[JoinHow, JoinHow] = {"right": "left", "left": "right"}
+            how = flip.get(how, how)
+
+        assert isinstance(left, MultiIndex)
+
+        level = left._get_level_number(level)
+        old_level = left.levels[level]
+
+        if not right.is_unique:
+            raise NotImplementedError(
+                "Index._join_level on non-unique index is not implemented"
+            )
+
+        new_level, left_lev_indexer, right_lev_indexer = old_level.join(
+            right, how=how, return_indexers=True
+        )
+
+        if left_lev_indexer is None:
+            if keep_order or len(left) == 0:
+                left_indexer = None
+                join_index = left
+            else:  # sort the leaves
+                left_indexer = _get_leaf_sorter(left.codes[: level + 1])
+                join_index = left[left_indexer]
+
+        else:
+            left_lev_indexer = ensure_platform_int(left_lev_indexer)
+            rev_indexer = lib.get_reverse_indexer(left_lev_indexer, len(old_level))
+            old_codes = left.codes[level]
+
+            taker = old_codes[old_codes != -1]
+            new_lev_codes = rev_indexer.take(taker)
+
+            new_codes = list(left.codes)
+            new_codes[level] = new_lev_codes
+
+            new_levels = list(left.levels)
+            new_levels[level] = new_level
+
+            if keep_order:  # just drop missing values. o.w. keep order
+                left_indexer = np.arange(len(left), dtype=np.intp)
+                left_indexer = cast(np.ndarray, left_indexer)
+                mask = new_lev_codes != -1
+                if not mask.all():
+                    new_codes = [lab[mask] for lab in new_codes]
+                    left_indexer = left_indexer[mask]
+
+            else:  # tie out the order with other
+                if level == 0:  # outer most level, take the fast route
+                    max_new_lev = 0 if len(new_lev_codes) == 0 else new_lev_codes.max()
+                    ngroups = 1 + max_new_lev
+                    left_indexer, counts = libalgos.groupsort_indexer(
+                        new_lev_codes, ngroups
+                    )
+
+                    # missing values are placed first; drop them!
+                    left_indexer = left_indexer[counts[0] :]
+                    new_codes = [lab[left_indexer] for lab in new_codes]
+
+                else:  # sort the leaves
+                    mask = new_lev_codes != -1
+                    mask_all = mask.all()
+                    if not mask_all:
+                        new_codes = [lab[mask] for lab in new_codes]
+
+                    left_indexer = _get_leaf_sorter(new_codes[: level + 1])
+                    new_codes = [lab[left_indexer] for lab in new_codes]
+
+                    # left_indexers are w.r.t masked frame.
+                    # reverse to original frame!
+                    if not mask_all:
+                        left_indexer = mask.nonzero()[0][left_indexer]
+
+            join_index = MultiIndex(
+                levels=new_levels,
+                codes=new_codes,
+                names=left.names,
+                verify_integrity=False,
+            )
+
+        if right_lev_indexer is not None:
+            right_indexer = right_lev_indexer.take(join_index.codes[level])
+        else:
+            right_indexer = join_index.codes[level]
+
+        if flip_order:
+            left_indexer, right_indexer = right_indexer, left_indexer
+
+        left_indexer = (
+            None if left_indexer is None else ensure_platform_int(left_indexer)
+        )
+        right_indexer = (
+            None if right_indexer is None else ensure_platform_int(right_indexer)
+        )
+        return join_index, left_indexer, right_indexer
+
+    def _join_monotonic(
+        self, other: Index, how: JoinHow = "left"
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        # We only get here with (caller is responsible for ensuring):
+        #  1) matching dtypes
+        #  2) both monotonic increasing
+        #  3) other.is_unique or self.is_unique
+        assert other.dtype == self.dtype
+        assert self._can_use_libjoin and other._can_use_libjoin
+
+        if self.equals(other):
+            # This is a convenient place for this check, but its correctness
+            #  does not depend on monotonicity, so it could go earlier
+            #  in the calling method.
+            ret_index = other if how == "right" else self
+            return ret_index, None, None
+
+        ridx: npt.NDArray[np.intp] | None
+        lidx: npt.NDArray[np.intp] | None
+
+        if how == "left":
+            if other.is_unique:
+                # We can perform much better than the general case
+                join_index = self
+                lidx = None
+                ridx = self._left_indexer_unique(other)
+            else:
+                join_array, lidx, ridx = self._left_indexer(other)
+                join_index, lidx, ridx = self._wrap_join_result(
+                    join_array, other, lidx, ridx, how
+                )
+        elif how == "right":
+            if self.is_unique:
+                # We can perform much better than the general case
+                join_index = other
+                lidx = other._left_indexer_unique(self)
+                ridx = None
+            else:
+                join_array, ridx, lidx = other._left_indexer(self)
+                join_index, lidx, ridx = self._wrap_join_result(
+                    join_array, other, lidx, ridx, how
+                )
+        elif how == "inner":
+            join_array, lidx, ridx = self._inner_indexer(other)
+            join_index, lidx, ridx = self._wrap_join_result(
+                join_array, other, lidx, ridx, how
+            )
+        elif how == "outer":
+            join_array, lidx, ridx = self._outer_indexer(other)
+            join_index, lidx, ridx = self._wrap_join_result(
+                join_array, other, lidx, ridx, how
+            )
+
+        lidx = None if lidx is None else ensure_platform_int(lidx)
+        ridx = None if ridx is None else ensure_platform_int(ridx)
+        return join_index, lidx, ridx
+
+    def _wrap_join_result(
+        self,
+        joined: ArrayLike,
+        other: Self,
+        lidx: npt.NDArray[np.intp] | None,
+        ridx: npt.NDArray[np.intp] | None,
+        how: JoinHow,
+    ) -> tuple[Self, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        assert other.dtype == self.dtype
+
+        if lidx is not None and lib.is_range_indexer(lidx, len(self)):
+            lidx = None
+        if ridx is not None and lib.is_range_indexer(ridx, len(other)):
+            ridx = None
+
+        # return self or other if possible to maintain cached attributes
+        if lidx is None:
+            join_index = self
+        elif ridx is None:
+            join_index = other
+        else:
+            join_index = self._constructor._with_infer(joined, dtype=self.dtype)
+
+        names = other.names if how == "right" else self.names
+        if join_index.names != names:
+            join_index = join_index.set_names(names)
+
+        return join_index, lidx, ridx
+
+    @final
+    @cache_readonly
+    def _can_use_libjoin(self) -> bool:
+        """
+        Whether we can use the fastpaths implemented in _libs.join.
+
+        This is driven by whether (in monotonic increasing cases that are
+        guaranteed not to have NAs) we can convert to a np.ndarray without
+        making a copy. If we cannot, this negates the performance benefit
+        of using libjoin.
+        """
+        if not self.is_monotonic_increasing:
+            # The libjoin functions all assume monotonicity.
+            return False
+
+        if type(self) is Index:
+            # excludes EAs, but include masks, we get here with monotonic
+            # values only, meaning no NA
+            return (
+                isinstance(self.dtype, np.dtype)
+                or isinstance(self._values, (ArrowExtensionArray, BaseMaskedArray))
+                or (
+                    isinstance(self.dtype, StringDtype)
+                    and self.dtype.storage == "python"
+                )
+            )
+        # Exclude index types where the conversion to numpy converts to object dtype,
+        #  which negates the performance benefit of libjoin
+        # Subclasses should override to return False if _get_join_target is
+        #  not zero-copy.
+        # TODO: exclude RangeIndex (which allocates memory)?
+        #  Doing so seems to break test_concat_datetime_timezone
+        return not isinstance(self, (ABCIntervalIndex, ABCMultiIndex))
+
+    # --------------------------------------------------------------------
+    # Uncategorized Methods
+
+    @property
+    def values(self) -> ArrayLike:
+        """
+        Return an array representing the data in the Index.
+
+        .. warning::
+
+           We recommend using :attr:`Index.array` or
+           :meth:`Index.to_numpy`, depending on whether you need
+           a reference to the underlying data or a NumPy array.
+
+        Returns
+        -------
+        array: numpy.ndarray or ExtensionArray
+
+        See Also
+        --------
+        Index.array : Reference to the underlying data.
+        Index.to_numpy : A NumPy array representing the underlying data.
+
+        Examples
+        --------
+        For :class:`pandas.Index`:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.values
+        array([1, 2, 3])
+
+        For :class:`pandas.IntervalIndex`:
+
+        >>> idx = pd.interval_range(start=0, end=5)
+        >>> idx.values
+        <IntervalArray>
+        [(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]]
+        Length: 5, dtype: interval[int64, right]
+        """
+        data = self._data
+        if isinstance(data, np.ndarray):
+            data = data.view()
+            data.flags.writeable = False
+        return data
+
+    @cache_readonly
+    @doc(IndexOpsMixin.array)
+    def array(self) -> ExtensionArray:
+        array = self._data
+        if isinstance(array, np.ndarray):
+            from pandas.core.arrays.numpy_ import NumpyExtensionArray
+
+            array = NumpyExtensionArray(array)
+        return array
+
+    @property
+    def _values(self) -> ExtensionArray | np.ndarray:
+        """
+        The best array representation.
+
+        This is an ndarray or ExtensionArray.
+
+        ``_values`` are consistent between ``Series`` and ``Index``.
+
+        It may differ from the public '.values' method.
+
+        index             | values          | _values       |
+        ----------------- | --------------- | ------------- |
+        Index             | ndarray         | ndarray       |
+        CategoricalIndex  | Categorical     | Categorical   |
+        DatetimeIndex     | ndarray[M8ns]   | DatetimeArray |
+        DatetimeIndex[tz] | ndarray[M8ns]   | DatetimeArray |
+        PeriodIndex       | ndarray[object] | PeriodArray   |
+        IntervalIndex     | IntervalArray   | IntervalArray |
+
+        See Also
+        --------
+        values : Values
+        """
+        return self._data
+
+    def _get_engine_target(self) -> ArrayLike:
+        """
+        Get the ndarray or ExtensionArray that we can pass to the IndexEngine
+        constructor.
+        """
+        vals = self._values
+        if isinstance(vals, StringArray):
+            # GH#45652 much more performant than ExtensionEngine
+            return vals._ndarray
+        if isinstance(vals, ArrowExtensionArray) and self.dtype.kind in "Mm":
+            import pyarrow as pa
+
+            pa_type = vals._pa_array.type
+            if pa.types.is_timestamp(pa_type):
+                vals = vals._to_datetimearray()
+                return vals._ndarray.view("i8")
+            elif pa.types.is_duration(pa_type):
+                vals = vals._to_timedeltaarray()
+                return vals._ndarray.view("i8")
+        if (
+            type(self) is Index
+            and isinstance(self._values, ExtensionArray)
+            and not isinstance(self._values, BaseMaskedArray)
+            and not (
+                isinstance(self._values, ArrowExtensionArray)
+                and is_numeric_dtype(self.dtype)
+                # Exclude decimal
+                and self.dtype.kind != "O"
+            )
+        ):
+            # TODO(ExtensionIndex): remove special-case, just use self._values
+            return self._values.astype(object)
+        return vals
+
+    @final
+    def _get_join_target(self) -> np.ndarray:
+        """
+        Get the ndarray or ExtensionArray that we can pass to the join
+        functions.
+        """
+        if isinstance(self._values, BaseMaskedArray):
+            # This is only used if our array is monotonic, so no NAs present
+            return self._values._data
+        elif isinstance(self._values, ArrowExtensionArray):
+            # This is only used if our array is monotonic, so no missing values
+            # present
+            return self._values.to_numpy()
+
+        # TODO: exclude ABCRangeIndex case here as it copies
+        target = self._get_engine_target()
+        if not isinstance(target, np.ndarray):
+            raise ValueError("_can_use_libjoin should return False.")
+        return target
+
+    def _from_join_target(self, result: np.ndarray) -> ArrayLike:
+        """
+        Cast the ndarray returned from one of the libjoin.foo_indexer functions
+        back to type(self._data).
+        """
+        if isinstance(self.values, BaseMaskedArray):
+            return type(self.values)(result, np.zeros(result.shape, dtype=np.bool_))
+        elif isinstance(self.values, (ArrowExtensionArray, StringArray)):
+            return type(self.values)._from_sequence(result, dtype=self.dtype)
+        return result
+
+    @doc(IndexOpsMixin._memory_usage)
+    def memory_usage(self, deep: bool = False) -> int:
+        result = self._memory_usage(deep=deep)
+
+        # include our engine hashtable, only if it's already cached
+        if "_engine" in self._cache:
+            result += self._engine.sizeof(deep=deep)
+        return result
+
+    @final
+    def where(self, cond, other=None) -> Index:
+        """
+        Replace values where the condition is False.
+
+        The replacement is taken from other.
+
+        Parameters
+        ----------
+        cond : bool array-like with the same length as self
+            Condition to select the values on.
+        other : scalar, or array-like, default None
+            Replacement if the condition is False.
+
+        Returns
+        -------
+        pandas.Index
+            A copy of self with values replaced from other
+            where the condition is False.
+
+        See Also
+        --------
+        Series.where : Same method for Series.
+        DataFrame.where : Same method for DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["car", "bike", "train", "tractor"])
+        >>> idx
+        Index(['car', 'bike', 'train', 'tractor'], dtype='object')
+        >>> idx.where(idx.isin(["car", "train"]), "other")
+        Index(['car', 'other', 'train', 'other'], dtype='object')
+        """
+        if isinstance(self, ABCMultiIndex):
+            raise NotImplementedError(
+                ".where is not supported for MultiIndex operations"
+            )
+        cond = np.asarray(cond, dtype=bool)
+        return self.putmask(~cond, other)
+
+    # construction helpers
+    @final
+    @classmethod
+    def _raise_scalar_data_error(cls, data):
+        # We return the TypeError so that we can raise it from the constructor
+        #  in order to keep mypy happy
+        raise TypeError(
+            f"{cls.__name__}(...) must be called with a collection of some "
+            f"kind, {repr(data) if not isinstance(data, np.generic) else str(data)} "
+            "was passed"
+        )
+
+    def _validate_fill_value(self, value):
+        """
+        Check if the value can be inserted into our array without casting,
+        and convert it to an appropriate native type if necessary.
+
+        Raises
+        ------
+        TypeError
+            If the value cannot be inserted into an array of this dtype.
+        """
+        dtype = self.dtype
+        if isinstance(dtype, np.dtype) and dtype.kind not in "mM":
+            # return np_can_hold_element(dtype, value)
+            try:
+                return np_can_hold_element(dtype, value)
+            except LossySetitemError as err:
+                # re-raise as TypeError for consistency
+                raise TypeError from err
+        elif not can_hold_element(self._values, value):
+            raise TypeError
+        return value
+
+    @cache_readonly
+    def _is_memory_usage_qualified(self) -> bool:
+        """
+        Return a boolean if we need a qualified .info display.
+        """
+        return is_object_dtype(self.dtype) or (
+            is_string_dtype(self.dtype) and self.dtype.storage == "python"  # type: ignore[union-attr]
+        )
+
+    def __contains__(self, key: Any) -> bool:
+        """
+        Return a boolean indicating whether the provided key is in the index.
+
+        Parameters
+        ----------
+        key : label
+            The key to check if it is present in the index.
+
+        Returns
+        -------
+        bool
+            Whether the key search is in the index.
+
+        Raises
+        ------
+        TypeError
+            If the key is not hashable.
+
+        See Also
+        --------
+        Index.isin : Returns an ndarray of boolean dtype indicating whether the
+            list-like key is in the index.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3, 4])
+        >>> idx
+        Index([1, 2, 3, 4], dtype='int64')
+
+        >>> 2 in idx
+        True
+        >>> 6 in idx
+        False
+        """
+        hash(key)
+        try:
+            return key in self._engine
+        except (OverflowError, TypeError, ValueError):
+            return False
+
+    # https://github.com/python/typeshed/issues/2148#issuecomment-520783318
+    # Incompatible types in assignment (expression has type "None", base class
+    # "object" defined the type as "Callable[[object], int]")
+    __hash__: ClassVar[None]  # type: ignore[assignment]
+
+    @final
+    def __setitem__(self, key, value) -> None:
+        raise TypeError("Index does not support mutable operations")
+
+    def __getitem__(self, key):
+        """
+        Override numpy.ndarray's __getitem__ method to work as desired.
+
+        This function adds lists and Series as valid boolean indexers
+        (ndarrays only supports ndarray with dtype=bool).
+
+        If resulting ndim != 1, plain ndarray is returned instead of
+        corresponding `Index` subclass.
+
+        """
+        getitem = self._data.__getitem__
+
+        if is_integer(key) or is_float(key):
+            # GH#44051 exclude bool, which would return a 2d ndarray
+            key = com.cast_scalar_indexer(key)
+            return getitem(key)
+
+        if isinstance(key, slice):
+            # This case is separated from the conditional above to avoid
+            # pessimization com.is_bool_indexer and ndim checks.
+            return self._getitem_slice(key)
+
+        if com.is_bool_indexer(key):
+            # if we have list[bools, length=1e5] then doing this check+convert
+            #  takes 166 µs + 2.1 ms and cuts the ndarray.__getitem__
+            #  time below from 3.8 ms to 496 µs
+            # if we already have ndarray[bool], the overhead is 1.4 µs or .25%
+            if isinstance(getattr(key, "dtype", None), ExtensionDtype):
+                key = key.to_numpy(dtype=bool, na_value=False)
+            else:
+                key = np.asarray(key, dtype=bool)
+
+            if not isinstance(self.dtype, ExtensionDtype):
+                if len(key) == 0 and len(key) != len(self):
+                    raise ValueError(
+                        "The length of the boolean indexer cannot be 0 "
+                        "when the Index has length greater than 0."
+                    )
+
+        result = getitem(key)
+        # Because we ruled out integer above, we always get an arraylike here
+        if result.ndim > 1:
+            disallow_ndim_indexing(result)
+
+        # NB: Using _constructor._simple_new would break if MultiIndex
+        #  didn't override __getitem__
+        return self._constructor._simple_new(result, name=self._name)
+
+    def _getitem_slice(self, slobj: slice) -> Self:
+        """
+        Fastpath for __getitem__ when we know we have a slice.
+        """
+        res = self._data[slobj]
+        result = type(self)._simple_new(res, name=self._name, refs=self._references)
+        if "_engine" in self._cache:
+            reverse = slobj.step is not None and slobj.step < 0
+            result._engine._update_from_sliced(self._engine, reverse=reverse)  # type: ignore[union-attr]
+
+        return result
+
+    @final
+    def _can_hold_identifiers_and_holds_name(self, name) -> bool:
+        """
+        Faster check for ``name in self`` when we know `name` is a Python
+        identifier (e.g. in NDFrame.__getattr__, which hits this to support
+        . key lookup). For indexes that can't hold identifiers (everything
+        but object & categorical) we just return False.
+
+        https://github.com/pandas-dev/pandas/issues/19764
+        """
+        if (
+            is_object_dtype(self.dtype)
+            or is_string_dtype(self.dtype)
+            or isinstance(self.dtype, CategoricalDtype)
+        ):
+            return name in self
+        return False
+
+    def append(self, other: Index | Sequence[Index]) -> Index:
+        """
+        Append a collection of Index options together.
+
+        Parameters
+        ----------
+        other : Index or list/tuple of indices
+            Single Index or a collection of indices, which can be either a list or a
+            tuple.
+
+        Returns
+        -------
+        Index
+            Returns a new Index object resulting from appending the provided other
+            indices to the original Index.
+
+        See Also
+        --------
+        Index.insert : Make new Index inserting new item at location.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.append(pd.Index([4]))
+        Index([1, 2, 3, 4], dtype='int64')
+        """
+        to_concat = [self]
+
+        if isinstance(other, (list, tuple)):
+            to_concat += list(other)
+        else:
+            # error: Argument 1 to "append" of "list" has incompatible type
+            # "Union[Index, Sequence[Index]]"; expected "Index"
+            to_concat.append(other)  # type: ignore[arg-type]
+
+        for obj in to_concat:
+            if not isinstance(obj, Index):
+                raise TypeError("all inputs must be Index")
+
+        names = {obj.name for obj in to_concat}
+        name = None if len(names) > 1 else self.name
+
+        return self._concat(to_concat, name)
+
+    def _concat(self, to_concat: list[Index], name: Hashable) -> Index:
+        """
+        Concatenate multiple Index objects.
+        """
+        to_concat_vals = [x._values for x in to_concat]
+
+        result = concat_compat(to_concat_vals)
+
+        return Index._with_infer(result, name=name)
+
+    def putmask(self, mask, value) -> Index:
+        """
+        Return a new Index of the values set with the mask.
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool]
+            Array of booleans denoting where values in the original
+            data are not ``NA``.
+        value : scalar
+            Scalar value to use to fill holes (e.g. 0).
+            This value cannot be a list-likes.
+
+        Returns
+        -------
+        Index
+            A new Index of the values set with the mask.
+
+        See Also
+        --------
+        numpy.putmask : Changes elements of an array
+            based on conditional and input values.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([1, 2, 3])
+        >>> idx2 = pd.Index([5, 6, 7])
+        >>> idx1.putmask([True, False, False], idx2)
+        Index([5, 2, 3], dtype='int64')
+        """
+        mask, noop = validate_putmask(self._values, mask)
+        if noop:
+            return self.copy()
+
+        if self.dtype != object and is_valid_na_for_dtype(value, self.dtype):
+            # e.g. None -> np.nan, see also Block._standardize_fill_value
+            value = self._na_value
+
+        try:
+            converted = self._validate_fill_value(value)
+        except (LossySetitemError, ValueError, TypeError) as err:
+            if is_object_dtype(self.dtype):  # pragma: no cover
+                raise err
+
+            # See also: Block.coerce_to_target_dtype
+            dtype = self._find_common_type_compat(value)
+            return self.astype(dtype).putmask(mask, value)
+
+        values = self._values.copy()
+
+        if isinstance(values, np.ndarray):
+            converted = setitem_datetimelike_compat(values, mask.sum(), converted)
+            np.putmask(values, mask, converted)
+
+        else:
+            # Note: we use the original value here, not converted, as
+            #  _validate_fill_value is not idempotent
+            values._putmask(mask, value)
+
+        return self._shallow_copy(values)
+
+    def equals(self, other: Any) -> bool:
+        """
+        Determine if two Index object are equal.
+
+        The things that are being compared are:
+
+        * The elements inside the Index object.
+        * The order of the elements inside the Index object.
+
+        Parameters
+        ----------
+        other : Any
+            The other object to compare against.
+
+        Returns
+        -------
+        bool
+            True if "other" is an Index and it has the same elements and order
+            as the calling index; False otherwise.
+
+        See Also
+        --------
+        Index.identical: Checks that object attributes and types are also equal.
+        Index.has_duplicates: Check if the Index has duplicate values.
+        Index.is_unique: Return if the index has unique values.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index([1, 2, 3])
+        >>> idx1
+        Index([1, 2, 3], dtype='int64')
+        >>> idx1.equals(pd.Index([1, 2, 3]))
+        True
+
+        The elements inside are compared
+
+        >>> idx2 = pd.Index(["1", "2", "3"])
+        >>> idx2
+        Index(['1', '2', '3'], dtype='object')
+
+        >>> idx1.equals(idx2)
+        False
+
+        The order is compared
+
+        >>> ascending_idx = pd.Index([1, 2, 3])
+        >>> ascending_idx
+        Index([1, 2, 3], dtype='int64')
+        >>> descending_idx = pd.Index([3, 2, 1])
+        >>> descending_idx
+        Index([3, 2, 1], dtype='int64')
+        >>> ascending_idx.equals(descending_idx)
+        False
+
+        The dtype is *not* compared
+
+        >>> int64_idx = pd.Index([1, 2, 3], dtype="int64")
+        >>> int64_idx
+        Index([1, 2, 3], dtype='int64')
+        >>> uint64_idx = pd.Index([1, 2, 3], dtype="uint64")
+        >>> uint64_idx
+        Index([1, 2, 3], dtype='uint64')
+        >>> int64_idx.equals(uint64_idx)
+        True
+        """
+        if self.is_(other):
+            return True
+
+        if not isinstance(other, Index):
+            return False
+
+        if len(self) != len(other):
+            # quickly return if the lengths are different
+            return False
+
+        if (
+            isinstance(self.dtype, StringDtype)
+            and self.dtype.na_value is np.nan
+            and other.dtype != self.dtype
+        ):
+            # TODO(infer_string) can we avoid this special case?
+            # special case for object behavior
+            return other.equals(self.astype(object))
+
+        if is_object_dtype(self.dtype) and not is_object_dtype(other.dtype):
+            # if other is not object, use other's logic for coercion
+            return other.equals(self)
+
+        if isinstance(other, ABCMultiIndex):
+            # d-level MultiIndex can equal d-tuple Index
+            return other.equals(self)
+
+        if isinstance(self._values, ExtensionArray):
+            # Dispatch to the ExtensionArray's .equals method.
+            if not isinstance(other, type(self)):
+                return False
+
+            earr = cast(ExtensionArray, self._data)
+            return earr.equals(other._data)
+
+        if isinstance(other.dtype, ExtensionDtype):
+            # All EA-backed Index subclasses override equals
+            return other.equals(self)
+
+        return array_equivalent(self._values, other._values)
+
+    @final
+    def identical(self, other) -> bool:
+        """
+        Similar to equals, but checks that object attributes and types are also equal.
+
+        Parameters
+        ----------
+        other : Index
+            The Index object you want to compare with the current Index object.
+
+        Returns
+        -------
+        bool
+            If two Index objects have equal elements and same type True,
+            otherwise False.
+
+        See Also
+        --------
+        Index.equals: Determine if two Index object are equal.
+        Index.has_duplicates: Check if the Index has duplicate values.
+        Index.is_unique: Return if the index has unique values.
+
+        Examples
+        --------
+        >>> idx1 = pd.Index(["1", "2", "3"])
+        >>> idx2 = pd.Index(["1", "2", "3"])
+        >>> idx2.identical(idx1)
+        True
+
+        >>> idx1 = pd.Index(["1", "2", "3"], name="A")
+        >>> idx2 = pd.Index(["1", "2", "3"], name="B")
+        >>> idx2.identical(idx1)
+        False
+        """
+        return (
+            self.equals(other)
+            and all(
+                getattr(self, c, None) == getattr(other, c, None)
+                for c in self._comparables
+            )
+            and type(self) == type(other)
+            and self.dtype == other.dtype
+        )
+
+    @final
+    def asof(self, label):
+        """
+        Return the label from the index, or, if not present, the previous one.
+
+        Assuming that the index is sorted, return the passed index label if it
+        is in the index, or return the previous index label if the passed one
+        is not in the index.
+
+        Parameters
+        ----------
+        label : object
+            The label up to which the method returns the latest index label.
+
+        Returns
+        -------
+        object
+            The passed label if it is in the index. The previous label if the
+            passed label is not in the sorted index or `NaN` if there is no
+            such label.
+
+        See Also
+        --------
+        Series.asof : Return the latest value in a Series up to the
+            passed index.
+        merge_asof : Perform an asof merge (similar to left join but it
+            matches on nearest key rather than equal key).
+        Index.get_loc : An `asof` is a thin wrapper around `get_loc`
+            with method='pad'.
+
+        Examples
+        --------
+        `Index.asof` returns the latest index label up to the passed label.
+
+        >>> idx = pd.Index(["2013-12-31", "2014-01-02", "2014-01-03"])
+        >>> idx.asof("2014-01-01")
+        '2013-12-31'
+
+        If the label is in the index, the method returns the passed label.
+
+        >>> idx.asof("2014-01-02")
+        '2014-01-02'
+
+        If all of the labels in the index are later than the passed label,
+        NaN is returned.
+
+        >>> idx.asof("1999-01-02")
+        nan
+
+        If the index is not sorted, an error is raised.
+
+        >>> idx_not_sorted = pd.Index(["2013-12-31", "2015-01-02", "2014-01-03"])
+        >>> idx_not_sorted.asof("2013-12-31")
+        Traceback (most recent call last):
+        ValueError: index must be monotonic increasing or decreasing
+        """
+        self._searchsorted_monotonic(label)  # validate sortedness
+        try:
+            loc = self.get_loc(label)
+        except (KeyError, TypeError) as err:
+            # KeyError -> No exact match, try for padded
+            # TypeError -> passed e.g. non-hashable, fall through to get
+            #  the tested exception message
+            indexer = self.get_indexer([label], method="pad")
+            if indexer.ndim > 1 or indexer.size > 1:
+                raise TypeError("asof requires scalar valued input") from err
+            loc = indexer.item()
+            if loc == -1:
+                return self._na_value
+        else:
+            if isinstance(loc, slice):
+                loc = loc.indices(len(self))[-1]
+
+        return self[loc]
+
+    def asof_locs(
+        self, where: Index, mask: npt.NDArray[np.bool_]
+    ) -> npt.NDArray[np.intp]:
+        """
+        Return the locations (indices) of labels in the index.
+
+        As in the :meth:`pandas.Index.asof`, if the label (a particular entry in
+        ``where``) is not in the index, the latest index label up to the
+        passed label is chosen and its index returned.
+
+        If all of the labels in the index are later than a label in ``where``,
+        -1 is returned.
+
+        ``mask`` is used to ignore ``NA`` values in the index during calculation.
+
+        Parameters
+        ----------
+        where : Index
+            An Index consisting of an array of timestamps.
+        mask : np.ndarray[bool]
+            Array of booleans denoting where values in the original
+            data are not ``NA``.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            An array of locations (indices) of the labels from the index
+            which correspond to the return values of :meth:`pandas.Index.asof`
+            for every element in ``where``.
+
+        See Also
+        --------
+        Index.asof : Return the label from the index, or, if not present, the
+            previous one.
+
+        Examples
+        --------
+        >>> idx = pd.date_range("2023-06-01", periods=3, freq="D")
+        >>> where = pd.DatetimeIndex(
+        ...     ["2023-05-30 00:12:00", "2023-06-01 00:00:00", "2023-06-02 23:59:59"]
+        ... )
+        >>> mask = np.ones(3, dtype=bool)
+        >>> idx.asof_locs(where, mask)
+        array([-1,  0,  1])
+
+        We can use ``mask`` to ignore certain values in the index during calculation.
+
+        >>> mask[1] = False
+        >>> idx.asof_locs(where, mask)
+        array([-1,  0,  0])
+        """
+        # error: No overload variant of "searchsorted" of "ndarray" matches argument
+        # types "Union[ExtensionArray, ndarray[Any, Any]]", "str"
+        # TODO: will be fixed when ExtensionArray.searchsorted() is fixed
+        locs = self._values[mask].searchsorted(
+            where._values,
+            side="right",  # type: ignore[call-overload]
+        )
+        locs = np.where(locs > 0, locs - 1, 0)
+
+        result = np.arange(len(self), dtype=np.intp)[mask].take(locs)
+
+        first_value = self._values[mask.argmax()]
+        result[(locs == 0) & (where._values < first_value)] = -1
+
+        return result
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        return_indexer: Literal[False] = ...,
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> Self: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        return_indexer: Literal[True],
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> tuple[Self, np.ndarray]: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        return_indexer: bool = ...,
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> Self | tuple[Self, np.ndarray]: ...
+
+    def sort_values(
+        self,
+        *,
+        return_indexer: bool = False,
+        ascending: bool = True,
+        na_position: NaPosition = "last",
+        key: Callable | None = None,
+    ) -> Self | tuple[Self, np.ndarray]:
+        """
+        Return a sorted copy of the index.
+
+        Return a sorted copy of the index, and optionally return the indices
+        that sorted the index itself.
+
+        Parameters
+        ----------
+        return_indexer : bool, default False
+            Should the indices that would sort the index be returned.
+        ascending : bool, default True
+            Should the index values be sorted in an ascending order.
+        na_position : {'first' or 'last'}, default 'last'
+            Argument 'first' puts NaNs at the beginning, 'last' puts NaNs at
+            the end.
+        key : callable, optional
+            If not None, apply the key function to the index values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect an
+            ``Index`` and return an ``Index`` of the same shape.
+
+        Returns
+        -------
+        sorted_index : pandas.Index
+            Sorted copy of the index.
+        indexer : numpy.ndarray, optional
+            The indices that the index itself was sorted by.
+
+        See Also
+        --------
+        Series.sort_values : Sort values of a Series.
+        DataFrame.sort_values : Sort values in a DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index([10, 100, 1, 1000])
+        >>> idx
+        Index([10, 100, 1, 1000], dtype='int64')
+
+        Sort values in ascending order (default behavior).
+
+        >>> idx.sort_values()
+        Index([1, 10, 100, 1000], dtype='int64')
+
+        Sort values in descending order, and also get the indices `idx` was
+        sorted by.
+
+        >>> idx.sort_values(ascending=False, return_indexer=True)
+        (Index([1000, 100, 10, 1], dtype='int64'), array([3, 1, 0, 2]))
+        """
+        if key is None and (
+            (ascending and self.is_monotonic_increasing)
+            or (not ascending and self.is_monotonic_decreasing)
+        ):
+            if return_indexer:
+                indexer = np.arange(len(self), dtype=np.intp)
+                return self.copy(), indexer
+            else:
+                return self.copy()
+
+        # GH 35584. Sort missing values according to na_position kwarg
+        # ignore na_position for MultiIndex
+        if not isinstance(self, ABCMultiIndex):
+            _as = nargsort(
+                items=self, ascending=ascending, na_position=na_position, key=key
+            )
+        else:
+            idx = cast(Index, ensure_key_mapped(self, key))
+            _as = idx.argsort(na_position=na_position)
+            if not ascending:
+                _as = _as[::-1]
+
+        sorted_index = self.take(_as)
+
+        if return_indexer:
+            return sorted_index, _as
+        else:
+            return sorted_index
+
+    def shift(self, periods: int = 1, freq=None) -> Self:
+        """
+        Shift index by desired number of time frequency increments.
+
+        This method is for shifting the values of datetime-like indexes
+        by a specified time increment a given number of times.
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Number of periods (or increments) to shift by,
+            can be positive or negative.
+        freq : pandas.DateOffset, pandas.Timedelta or str, optional
+            Frequency increment to shift by.
+            If None, the index is shifted by its own `freq` attribute.
+            Offset aliases are valid strings, e.g., 'D', 'W', 'M' etc.
+
+        Returns
+        -------
+        pandas.Index
+            Shifted index.
+
+        See Also
+        --------
+        Series.shift : Shift values of Series.
+
+        Notes
+        -----
+        This method is only implemented for datetime-like index classes,
+        i.e., DatetimeIndex, PeriodIndex and TimedeltaIndex.
+
+        Examples
+        --------
+        Put the first 5 month starts of 2011 into an index.
+
+        >>> month_starts = pd.date_range("1/1/2011", periods=5, freq="MS")
+        >>> month_starts
+        DatetimeIndex(['2011-01-01', '2011-02-01', '2011-03-01', '2011-04-01',
+                       '2011-05-01'],
+                      dtype='datetime64[ns]', freq='MS')
+
+        Shift the index by 10 days.
+
+        >>> month_starts.shift(10, freq="D")
+        DatetimeIndex(['2011-01-11', '2011-02-11', '2011-03-11', '2011-04-11',
+                       '2011-05-11'],
+                      dtype='datetime64[ns]', freq=None)
+
+        The default value of `freq` is the `freq` attribute of the index,
+        which is 'MS' (month start) in this example.
+
+        >>> month_starts.shift(10)
+        DatetimeIndex(['2011-11-01', '2011-12-01', '2012-01-01', '2012-02-01',
+                       '2012-03-01'],
+                      dtype='datetime64[ns]', freq='MS')
+        """
+        raise NotImplementedError(
+            f"This method is only implemented for DatetimeIndex, PeriodIndex and "
+            f"TimedeltaIndex; Got type {type(self).__name__}"
+        )
+
+    def argsort(self, *args, **kwargs) -> npt.NDArray[np.intp]:
+        """
+        Return the integer indices that would sort the index.
+
+        Parameters
+        ----------
+        *args
+            Passed to `numpy.ndarray.argsort`.
+        **kwargs
+            Passed to `numpy.ndarray.argsort`.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            Integer indices that would sort the index if used as
+            an indexer.
+
+        See Also
+        --------
+        numpy.argsort : Similar method for NumPy arrays.
+        Index.sort_values : Return sorted copy of Index.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["b", "a", "d", "c"])
+        >>> idx
+        Index(['b', 'a', 'd', 'c'], dtype='object')
+
+        >>> order = idx.argsort()
+        >>> order
+        array([1, 0, 3, 2])
+
+        >>> idx[order]
+        Index(['a', 'b', 'c', 'd'], dtype='object')
+        """
+        # This works for either ndarray or EA, is overridden
+        #  by RangeIndex, MultIIndex
+        return self._data.argsort(*args, **kwargs)
+
+    def _check_indexing_error(self, key) -> None:
+        if not is_scalar(key):
+            # if key is not a scalar, directly raise an error (the code below
+            # would convert to numpy arrays and raise later any way) - GH29926
+            raise InvalidIndexError(key)
+
+    @cache_readonly
+    def _should_fallback_to_positional(self) -> bool:
+        """
+        Should an integer key be treated as positional?
+        """
+        return self.inferred_type not in {
+            "integer",
+            "mixed-integer",
+            "floating",
+            "complex",
+        }
+
+    _index_shared_docs["get_indexer_non_unique"] = """
+        Compute indexer and mask for new index given the current index.
+
+        The indexer should be then used as an input to ndarray.take to align the
+        current data to the new index.
+
+        Parameters
+        ----------
+        target : %(target_klass)s
+            An iterable containing the values to be used for computing indexer.
+
+        Returns
+        -------
+        indexer : np.ndarray[np.intp]
+            Integers from 0 to n - 1 indicating that the index at these
+            positions matches the corresponding target values. Missing values
+            in the target are marked by -1.
+        missing : np.ndarray[np.intp]
+            An indexer into the target of the values not found.
+            These correspond to the -1 in the indexer array.
+
+        See Also
+        --------
+        Index.get_indexer : Computes indexer and mask for new index given
+            the current index.
+        Index.get_indexer_for : Returns an indexer even when non-unique.
+
+        Examples
+        --------
+        >>> index = pd.Index(['c', 'b', 'a', 'b', 'b'])
+        >>> index.get_indexer_non_unique(['b', 'b'])
+        (array([1, 3, 4, 1, 3, 4]), array([], dtype=int64))
+
+        In the example below there are no matched values.
+
+        >>> index = pd.Index(['c', 'b', 'a', 'b', 'b'])
+        >>> index.get_indexer_non_unique(['q', 'r', 't'])
+        (array([-1, -1, -1]), array([0, 1, 2]))
+
+        For this reason, the returned ``indexer`` contains only integers equal to -1.
+        It demonstrates that there's no match between the index and the ``target``
+        values at these positions. The mask [0, 1, 2] in the return value shows that
+        the first, second, and third elements are missing.
+
+        Notice that the return value is a tuple contains two items. In the example
+        below the first item is an array of locations in ``index``. The second
+        item is a mask shows that the first and third elements are missing.
+
+        >>> index = pd.Index(['c', 'b', 'a', 'b', 'b'])
+        >>> index.get_indexer_non_unique(['f', 'b', 's'])
+        (array([-1,  1,  3,  4, -1]), array([0, 2]))
+        """
+
+    @Appender(_index_shared_docs["get_indexer_non_unique"] % _index_doc_kwargs)
+    def get_indexer_non_unique(
+        self, target
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        target = self._maybe_cast_listlike_indexer(target)
+
+        if not self._should_compare(target) and not self._should_partial_index(target):
+            # _should_partial_index e.g. IntervalIndex with numeric scalars
+            #  that can be matched to Interval scalars.
+            return self._get_indexer_non_comparable(target, method=None, unique=False)
+
+        pself, ptarget = self._maybe_downcast_for_indexing(target)
+        if pself is not self or ptarget is not target:
+            return pself.get_indexer_non_unique(ptarget)
+
+        if self.dtype != target.dtype:
+            # TODO: if object, could use infer_dtype to preempt costly
+            #  conversion if still non-comparable?
+            dtype = self._find_common_type_compat(target)
+
+            this = self.astype(dtype, copy=False)
+            that = target.astype(dtype, copy=False)
+            return this.get_indexer_non_unique(that)
+
+        # TODO: get_indexer has fastpaths for both Categorical-self and
+        #  Categorical-target. Can we do something similar here?
+
+        # Note: _maybe_downcast_for_indexing ensures we never get here
+        #  with MultiIndex self and non-Multi target
+        if self._is_multi and target._is_multi:
+            engine = self._engine
+            # Item "IndexEngine" of "Union[IndexEngine, ExtensionEngine]" has
+            # no attribute "_extract_level_codes"
+            tgt_values = engine._extract_level_codes(target)  # type: ignore[union-attr]
+        else:
+            tgt_values = target._get_engine_target()
+
+        indexer, missing = self._engine.get_indexer_non_unique(tgt_values)
+        return ensure_platform_int(indexer), ensure_platform_int(missing)
+
+    @final
+    def get_indexer_for(self, target) -> npt.NDArray[np.intp]:
+        """
+        Guaranteed return of an indexer even when non-unique.
+
+        This dispatches to get_indexer or get_indexer_non_unique
+        as appropriate.
+
+        Parameters
+        ----------
+        target : Index
+            An iterable containing the values to be used for computing indexer.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            List of indices.
+
+        See Also
+        --------
+        Index.get_indexer : Computes indexer and mask for new index given
+            the current index.
+        Index.get_non_unique : Returns indexer and masks for new index given
+            the current index.
+
+        Examples
+        --------
+        >>> idx = pd.Index([np.nan, "var1", np.nan])
+        >>> idx.get_indexer_for([np.nan])
+        array([0, 2])
+        """
+        if self._index_as_unique:
+            return self.get_indexer(target)
+        indexer, _ = self.get_indexer_non_unique(target)
+        return indexer
+
+    def _get_indexer_strict(self, key, axis_name: str_t) -> tuple[Index, np.ndarray]:
+        """
+        Analogue to get_indexer that raises if any elements are missing.
+        """
+        keyarr = key
+        if not isinstance(keyarr, Index):
+            keyarr = com.asarray_tuplesafe(keyarr)
+
+        if self._index_as_unique:
+            indexer = self.get_indexer_for(keyarr)
+            keyarr = self.reindex(keyarr)[0]
+        else:
+            keyarr, indexer, new_indexer = self._reindex_non_unique(keyarr)
+
+        self._raise_if_missing(keyarr, indexer, axis_name)
+
+        keyarr = self.take(indexer)
+        if isinstance(key, Index):
+            # GH 42790 - Preserve name from an Index
+            keyarr.name = key.name
+        if lib.is_np_dtype(keyarr.dtype, "mM") or isinstance(
+            keyarr.dtype, DatetimeTZDtype
+        ):
+            # DTI/TDI.take can infer a freq in some cases when we dont want one
+            if isinstance(key, list) or (
+                isinstance(key, type(self))
+                # "Index" has no attribute "freq"
+                and key.freq is None  # type: ignore[attr-defined]
+            ):
+                keyarr = keyarr._with_freq(None)
+
+        return keyarr, indexer
+
+    def _raise_if_missing(self, key, indexer, axis_name: str_t) -> None:
+        """
+        Check that indexer can be used to return a result.
+
+        e.g. at least one element was found,
+        unless the list of keys was actually empty.
+
+        Parameters
+        ----------
+        key : list-like
+            Targeted labels (only used to show correct error message).
+        indexer: array-like of booleans
+            Indices corresponding to the key,
+            (with -1 indicating not found).
+        axis_name : str
+
+        Raises
+        ------
+        KeyError
+            If at least one key was requested but none was found.
+        """
+        if len(key) == 0:
+            return
+
+        # Count missing values
+        missing_mask = indexer < 0
+        nmissing = missing_mask.sum()
+
+        if nmissing:
+            if nmissing == len(indexer):
+                raise KeyError(f"None of [{key}] are in the [{axis_name}]")
+
+            not_found = list(ensure_index(key)[missing_mask.nonzero()[0]].unique())
+            raise KeyError(f"{not_found} not in index")
+
+    @overload
+    def _get_indexer_non_comparable(
+        self, target: Index, method, unique: Literal[True] = ...
+    ) -> npt.NDArray[np.intp]: ...
+
+    @overload
+    def _get_indexer_non_comparable(
+        self, target: Index, method, unique: Literal[False]
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+
+    @overload
+    def _get_indexer_non_comparable(
+        self, target: Index, method, unique: bool = True
+    ) -> npt.NDArray[np.intp] | tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]: ...
+
+    @final
+    def _get_indexer_non_comparable(
+        self, target: Index, method, unique: bool = True
+    ) -> npt.NDArray[np.intp] | tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        """
+        Called from get_indexer or get_indexer_non_unique when the target
+        is of a non-comparable dtype.
+
+        For get_indexer lookups with method=None, get_indexer is an _equality_
+        check, so non-comparable dtypes mean we will always have no matches.
+
+        For get_indexer lookups with a method, get_indexer is an _inequality_
+        check, so non-comparable dtypes mean we will always raise TypeError.
+
+        Parameters
+        ----------
+        target : Index
+        method : str or None
+        unique : bool, default True
+            * True if called from get_indexer.
+            * False if called from get_indexer_non_unique.
+
+        Raises
+        ------
+        TypeError
+            If doing an inequality check, i.e. method is not None.
+        """
+        if method is not None:
+            other_dtype = _unpack_nested_dtype(target)
+            raise TypeError(f"Cannot compare dtypes {self.dtype} and {other_dtype}")
+
+        no_matches = -1 * np.ones(target.shape, dtype=np.intp)
+        if unique:
+            # This is for get_indexer
+            return no_matches
+        else:
+            # This is for get_indexer_non_unique
+            missing = np.arange(len(target), dtype=np.intp)
+            return no_matches, missing
+
+    @property
+    def _index_as_unique(self) -> bool:
+        """
+        Whether we should treat this as unique for the sake of
+        get_indexer vs get_indexer_non_unique.
+
+        For IntervalIndex compat.
+        """
+        return self.is_unique
+
+    _requires_unique_msg = "Reindexing only valid with uniquely valued Index objects"
+
+    @final
+    def _maybe_downcast_for_indexing(self, other: Index) -> tuple[Index, Index]:
+        """
+        When dealing with an object-dtype Index and a non-object Index, see
+        if we can upcast the object-dtype one to improve performance.
+        """
+
+        if isinstance(self, ABCDatetimeIndex) and isinstance(other, ABCDatetimeIndex):
+            if (
+                self.tz is not None
+                and other.tz is not None
+                and not tz_compare(self.tz, other.tz)
+            ):
+                # standardize on UTC
+                return self.tz_convert("UTC"), other.tz_convert("UTC")
+
+        elif self.inferred_type == "date" and isinstance(other, ABCDatetimeIndex):
+            try:
+                return type(other)(self), other
+            except OutOfBoundsDatetime:
+                return self, other
+        elif self.inferred_type == "timedelta" and isinstance(other, ABCTimedeltaIndex):
+            # TODO: we dont have tests that get here
+            return type(other)(self), other
+
+        elif self.dtype.kind == "u" and other.dtype.kind == "i":
+            # GH#41873
+            if other.min() >= 0:
+                # lookup min as it may be cached
+                # TODO: may need itemsize check if we have non-64-bit Indexes
+                return self, other.astype(self.dtype)
+
+        elif self._is_multi and not other._is_multi:
+            try:
+                # "Type[Index]" has no attribute "from_tuples"
+                other = type(self).from_tuples(other)  # type: ignore[attr-defined]
+            except (TypeError, ValueError):
+                # let's instead try with a straight Index
+                self = Index(self._values)
+
+        if not is_object_dtype(self.dtype) and is_object_dtype(other.dtype):
+            # Reverse op so we dont need to re-implement on the subclasses
+            other, self = other._maybe_downcast_for_indexing(self)
+
+        return self, other
+
+    @final
+    def _find_common_type_compat(self, target) -> DtypeObj:
+        """
+        Implementation of find_common_type that adjusts for Index-specific
+        special cases.
+        """
+        target_dtype, _ = infer_dtype_from(target)
+
+        if using_string_dtype():
+            # special case: if left or right is a zero-length RangeIndex or
+            # Index[object], those can be created by the default empty constructors
+            # -> for that case ignore this dtype and always return the other
+            # (https://github.com/pandas-dev/pandas/pull/60797)
+            from pandas.core.indexes.range import RangeIndex
+
+            if len(self) == 0 and (
+                isinstance(self, RangeIndex) or self.dtype == np.object_
+            ):
+                return target_dtype
+            if (
+                isinstance(target, Index)
+                and len(target) == 0
+                and (isinstance(target, RangeIndex) or target_dtype == np.object_)
+            ):
+                return self.dtype
+
+        # special case: if one dtype is uint64 and the other a signed int, return object
+        # See https://github.com/pandas-dev/pandas/issues/26778 for discussion
+        # Now it's:
+        # * float | [u]int -> float
+        # * uint64 | signed int  -> object
+        # We may change union(float | [u]int) to go to object.
+        if self.dtype == "uint64" or target_dtype == "uint64":
+            if is_signed_integer_dtype(self.dtype) or is_signed_integer_dtype(
+                target_dtype
+            ):
+                return _dtype_obj
+
+        dtype = find_result_type(self.dtype, target)
+        dtype = common_dtype_categorical_compat([self, target], dtype)
+        return dtype
+
+    @final
+    def _should_compare(self, other: Index) -> bool:
+        """
+        Check if `self == other` can ever have non-False entries.
+        """
+
+        # NB: we use inferred_type rather than is_bool_dtype to catch
+        #  object_dtype_of_bool and categorical[object_dtype_of_bool] cases
+        if (
+            other.inferred_type == "boolean" and is_any_real_numeric_dtype(self.dtype)
+        ) or (
+            self.inferred_type == "boolean" and is_any_real_numeric_dtype(other.dtype)
+        ):
+            # GH#16877 Treat boolean labels passed to a numeric index as not
+            #  found. Without this fix False and True would be treated as 0 and 1
+            #  respectively.
+            return False
+
+        dtype = _unpack_nested_dtype(other)
+        return (
+            self._is_comparable_dtype(dtype)
+            or is_object_dtype(dtype)
+            or is_string_dtype(dtype)
+        )
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        """
+        Can we compare values of the given dtype to our own?
+        """
+        if self.dtype.kind == "b":
+            return dtype.kind == "b"
+        elif is_numeric_dtype(self.dtype):
+            return is_numeric_dtype(dtype)
+        # TODO: this was written assuming we only get here with object-dtype,
+        #  which is no longer correct. Can we specialize for EA?
+        return True
+
+    @final
+    def groupby(self, values) -> PrettyDict[Hashable, Index]:
+        """
+        Group the index labels by a given array of values.
+
+        Parameters
+        ----------
+        values : array
+            Values used to determine the groups.
+
+        Returns
+        -------
+        dict
+            {group name -> group labels}
+        """
+        # TODO: if we are a MultiIndex, we can do better
+        # that converting to tuples
+        if isinstance(values, ABCMultiIndex):
+            values = values._values
+        values = Categorical(values)
+        result = values._reverse_indexer()
+
+        # map to the label
+        result = {k: self.take(v) for k, v in result.items()}
+
+        return PrettyDict(result)
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        """
+        Map values using an input mapping or function.
+
+        Parameters
+        ----------
+        mapper : function, dict, or Series
+            Mapping correspondence.
+        na_action : {None, 'ignore'}
+            If 'ignore', propagate NA values, without passing them to the
+            mapping correspondence.
+
+        Returns
+        -------
+        Union[Index, MultiIndex]
+            The output of the mapping function applied to the index.
+            If the function returns a tuple with more than one element
+            a MultiIndex will be returned.
+
+        See Also
+        --------
+        Index.where : Replace values where the condition is False.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.map({1: "a", 2: "b", 3: "c"})
+        Index(['a', 'b', 'c'], dtype='object')
+
+        Using `map` with a function:
+
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx.map("I am a {}".format)
+        Index(['I am a 1', 'I am a 2', 'I am a 3'], dtype='object')
+
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> idx.map(lambda x: x.upper())
+        Index(['A', 'B', 'C'], dtype='object')
+        """
+        from pandas.core.indexes.multi import MultiIndex
+
+        new_values = self._map_values(mapper, na_action=na_action)
+
+        # we can return a MultiIndex
+        if new_values.size and isinstance(new_values[0], tuple):
+            if isinstance(self, MultiIndex):
+                names = self.names
+            elif self.name:
+                names = [self.name] * len(new_values[0])
+            else:
+                names = None
+            return MultiIndex.from_tuples(new_values, names=names)
+
+        dtype = None
+        if not new_values.size:
+            # empty
+            dtype = self.dtype
+
+        # e.g. if we are floating and new_values is all ints, then we
+        #  don't want to cast back to floating.  But if we are UInt64
+        #  and new_values is all ints, we want to try.
+        same_dtype = lib.infer_dtype(new_values, skipna=False) == self.inferred_type
+        if same_dtype:
+            new_values = maybe_cast_pointwise_result(
+                new_values, self.dtype, same_dtype=same_dtype
+            )
+
+        return Index._with_infer(new_values, dtype=dtype, copy=False, name=self.name)
+
+    # TODO: De-duplicate with map, xref GH#32349
+    @final
+    def _transform_index(self, func, *, level=None) -> Index:
+        """
+        Apply function to all values found in index.
+
+        This includes transforming multiindex entries separately.
+        Only apply function to one level of the MultiIndex if level is specified.
+        """
+        if isinstance(self, ABCMultiIndex):
+            values = [
+                self.get_level_values(i).map(func)
+                if i == level or level is None
+                else self.get_level_values(i)
+                for i in range(self.nlevels)
+            ]
+            return type(self).from_arrays(values)
+        else:
+            items = [func(x) for x in self]
+            return Index(items, name=self.name, tupleize_cols=False)
+
+    def isin(self, values, level: str_t | int | None = None) -> npt.NDArray[np.bool_]:
+        """
+        Return a boolean array where the index values are in `values`.
+
+        Compute boolean array of whether each index value is found in the
+        passed set of values. The length of the returned boolean array matches
+        the length of the index.
+
+        Parameters
+        ----------
+        values : set or list-like
+            Sought values.
+        level : str or int, optional
+            Name or position of the index level to use (if the index is a
+            `MultiIndex`).
+
+        Returns
+        -------
+        np.ndarray[bool]
+            NumPy array of boolean values.
+
+        See Also
+        --------
+        Series.isin : Same for Series.
+        DataFrame.isin : Same method for DataFrames.
+
+        Notes
+        -----
+        In the case of `MultiIndex` you must either specify `values` as a
+        list-like object containing tuples that are the same length as the
+        number of levels, or specify `level`. Otherwise it will raise a
+        ``ValueError``.
+
+        If `level` is specified:
+
+        - if it is the name of one *and only one* index level, use that level;
+        - otherwise it should be a number indicating level position.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+
+        Check whether each index value in a list of values.
+
+        >>> idx.isin([1, 4])
+        array([ True, False, False])
+
+        >>> midx = pd.MultiIndex.from_arrays(
+        ...     [[1, 2, 3], ["red", "blue", "green"]], names=["number", "color"]
+        ... )
+        >>> midx
+        MultiIndex([(1,   'red'),
+                    (2,  'blue'),
+                    (3, 'green')],
+                   names=['number', 'color'])
+
+        Check whether the strings in the 'color' level of the MultiIndex
+        are in a list of colors.
+
+        >>> midx.isin(["red", "orange", "yellow"], level="color")
+        array([ True, False, False])
+
+        To check across the levels of a MultiIndex, pass a list of tuples:
+
+        >>> midx.isin([(1, "red"), (3, "red")])
+        array([ True, False, False])
+        """
+        if level is not None:
+            self._validate_index_level(level)
+        return algos.isin(self._values, values)
+
+    def _get_string_slice(self, key: str_t):
+        # this is for partial string indexing,
+        # overridden in DatetimeIndex, TimedeltaIndex and PeriodIndex
+        raise NotImplementedError
+
+    def slice_indexer(
+        self,
+        start: Hashable | None = None,
+        end: Hashable | None = None,
+        step: int | None = None,
+    ) -> slice:
+        """
+        Compute the slice indexer for input labels and step.
+
+        Index needs to be ordered and unique.
+
+        Parameters
+        ----------
+        start : label, default None
+            If None, defaults to the beginning.
+        end : label, default None
+            If None, defaults to the end.
+        step : int, default None
+            If None, defaults to 1.
+
+        Returns
+        -------
+        slice
+            A slice object.
+
+        Raises
+        ------
+        KeyError : If key does not exist, or key is not unique and index is
+            not ordered.
+
+        See Also
+        --------
+        Index.slice_locs : Computes slice locations for input labels.
+        Index.get_slice_bound : Retrieves slice bound that corresponds to given label.
+
+        Notes
+        -----
+        This function assumes that the data is sorted, so use at your own peril.
+
+        Examples
+        --------
+        This is a method on all index types. For example you can do:
+
+        >>> idx = pd.Index(list("abcd"))
+        >>> idx.slice_indexer(start="b", end="c")
+        slice(1, 3, None)
+
+        >>> idx = pd.MultiIndex.from_arrays([list("abcd"), list("efgh")])
+        >>> idx.slice_indexer(start="b", end=("c", "g"))
+        slice(1, 3, None)
+        """
+        start_slice, end_slice = self.slice_locs(start, end, step=step)
+
+        # return a slice
+        if not is_scalar(start_slice):
+            raise AssertionError("Start slice bound is non-scalar")
+        if not is_scalar(end_slice):
+            raise AssertionError("End slice bound is non-scalar")
+
+        return slice(start_slice, end_slice, step)
+
+    def _maybe_cast_indexer(self, key):
+        """
+        If we have a float key and are not a floating index, then try to cast
+        to an int if equivalent.
+        """
+        return key
+
+    def _maybe_cast_listlike_indexer(self, target) -> Index:
+        """
+        Analogue to maybe_cast_indexer for get_indexer instead of get_loc.
+        """
+        target_index = ensure_index(target)
+        if (
+            not hasattr(target, "dtype")
+            and self.dtype == object
+            and target_index.dtype == "string"
+        ):
+            # If we started with a list-like, avoid inference to string dtype if self
+            # is object dtype (coercing to string dtype will alter the missing values)
+            target_index = Index(target, dtype=self.dtype)
+        return target_index
+
+    @final
+    def _validate_indexer(
+        self,
+        form: Literal["positional", "slice"],
+        key,
+        kind: Literal["getitem", "iloc"],
+    ) -> None:
+        """
+        If we are positional indexer, validate that we have appropriate
+        typed bounds must be an integer.
+        """
+        if not lib.is_int_or_none(key):
+            self._raise_invalid_indexer(form, key)
+
+    def _maybe_cast_slice_bound(self, label, side: str_t):
+        """
+        This function should be overloaded in subclasses that allow non-trivial
+        casting on label-slice bounds, e.g. datetime-like indices allowing
+        strings containing formatted datetimes.
+
+        Parameters
+        ----------
+        label : object
+        side : {'left', 'right'}
+
+        Returns
+        -------
+        label : object
+
+        Notes
+        -----
+        Value of `side` parameter should be validated in caller.
+        """
+
+        # We are a plain index here (sub-class override this method if they
+        # wish to have special treatment for floats/ints, e.g. datetimelike Indexes
+
+        if is_numeric_dtype(self.dtype):
+            return self._maybe_cast_indexer(label)
+
+        # reject them, if index does not contain label
+        if (is_float(label) or is_integer(label)) and label not in self:
+            self._raise_invalid_indexer("slice", label)
+
+        return label
+
+    def _searchsorted_monotonic(self, label, side: Literal["left", "right"] = "left"):
+        if self.is_monotonic_increasing:
+            return self.searchsorted(label, side=side)
+        elif self.is_monotonic_decreasing:
+            # np.searchsorted expects ascending sort order, have to reverse
+            # everything for it to work (element ordering, search side and
+            # resulting value).
+            pos = self[::-1].searchsorted(
+                label, side="right" if side == "left" else "left"
+            )
+            return len(self) - pos
+
+        raise ValueError("index must be monotonic increasing or decreasing")
+
+    def get_slice_bound(self, label, side: Literal["left", "right"]) -> int:
+        """
+        Calculate slice bound that corresponds to given label.
+
+        Returns leftmost (one-past-the-rightmost if ``side=='right'``) position
+        of given label.
+
+        Parameters
+        ----------
+        label : object
+            The label for which to calculate the slice bound.
+        side : {'left', 'right'}
+            if 'left' return leftmost position of given label.
+            if 'right' return one-past-the-rightmost position of given label.
+
+        Returns
+        -------
+        int
+            Index of label.
+
+        See Also
+        --------
+        Index.get_loc : Get integer location, slice or boolean mask for requested
+            label.
+
+        Examples
+        --------
+        >>> idx = pd.RangeIndex(5)
+        >>> idx.get_slice_bound(3, "left")
+        3
+
+        >>> idx.get_slice_bound(3, "right")
+        4
+
+        If ``label`` is non-unique in the index, an error will be raised.
+
+        >>> idx_duplicate = pd.Index(["a", "b", "a", "c", "d"])
+        >>> idx_duplicate.get_slice_bound("a", "left")
+        Traceback (most recent call last):
+        KeyError: Cannot get left slice bound for non-unique label: 'a'
+        """
+
+        if side not in ("left", "right"):
+            raise ValueError(
+                "Invalid value for side kwarg, must be either "
+                f"'left' or 'right': {side}"
+            )
+
+        original_label = label
+
+        # For datetime indices label may be a string that has to be converted
+        # to datetime boundary according to its resolution.
+        label = self._maybe_cast_slice_bound(label, side)
+
+        # we need to look up the label
+        try:
+            slc = self.get_loc(label)
+        except KeyError as err:
+            try:
+                return self._searchsorted_monotonic(label, side)
+            except ValueError:
+                # raise the original KeyError
+                raise err from None
+
+        if isinstance(slc, np.ndarray):
+            # get_loc may return a boolean array, which
+            # is OK as long as they are representable by a slice.
+            assert is_bool_dtype(slc.dtype)
+            slc = lib.maybe_booleans_to_slice(slc.view("u1"))
+            if isinstance(slc, np.ndarray):
+                raise KeyError(
+                    f"Cannot get {side} slice bound for non-unique "
+                    f"label: {original_label!r}"
+                )
+
+        if isinstance(slc, slice):
+            if side == "left":
+                return slc.start
+            else:
+                return slc.stop
+        else:
+            if side == "right":
+                return slc + 1
+            else:
+                return slc
+
+    def slice_locs(
+        self,
+        start: SliceType = None,
+        end: SliceType = None,
+        step: int | None = None,
+    ) -> tuple[int, int]:
+        """
+        Compute slice locations for input labels.
+
+        Parameters
+        ----------
+        start : label, default None
+            If None, defaults to the beginning.
+        end : label, default None
+            If None, defaults to the end.
+        step : int, defaults None
+            If None, defaults to 1.
+
+        Returns
+        -------
+        tuple[int, int]
+            Returns a tuple of two integers representing the slice locations for the
+            input labels within the index.
+
+        See Also
+        --------
+        Index.get_loc : Get location for a single label.
+
+        Notes
+        -----
+        This method only works if the index is monotonic or unique.
+
+        Examples
+        --------
+        >>> idx = pd.Index(list("abcd"))
+        >>> idx.slice_locs(start="b", end="c")
+        (1, 3)
+
+        >>> idx = pd.Index(list("bcde"))
+        >>> idx.slice_locs(start="a", end="c")
+        (0, 2)
+        """
+        inc = step is None or step >= 0
+
+        if not inc:
+            # If it's a reverse slice, temporarily swap bounds.
+            start, end = end, start
+
+        # GH 16785: If start and end happen to be date strings with UTC offsets
+        # attempt to parse and check that the offsets are the same
+        if isinstance(start, (str, datetime)) and isinstance(end, (str, datetime)):
+            try:
+                ts_start = Timestamp(start)
+                ts_end = Timestamp(end)
+            except (ValueError, TypeError):
+                pass
+            else:
+                if not tz_compare(ts_start.tzinfo, ts_end.tzinfo):
+                    raise ValueError("Both dates must have the same UTC offset")
+
+        start_slice = None
+        if start is not None:
+            start_slice = self.get_slice_bound(start, "left")
+        if start_slice is None:
+            start_slice = 0
+
+        end_slice = None
+        if end is not None:
+            end_slice = self.get_slice_bound(end, "right")
+        if end_slice is None:
+            end_slice = len(self)
+
+        if not inc:
+            # Bounds at this moment are swapped, swap them back and shift by 1.
+            #
+            # slice_locs('B', 'A', step=-1): s='B', e='A'
+            #
+            #              s='A'                 e='B'
+            # AFTER SWAP:    |                     |
+            #                v ------------------> V
+            #           -----------------------------------
+            #           | | |A|A|A|A| | | | | |B|B| | | | |
+            #           -----------------------------------
+            #              ^ <------------------ ^
+            # SHOULD BE:   |                     |
+            #           end=s-1              start=e-1
+            #
+            end_slice, start_slice = start_slice - 1, end_slice - 1
+
+            # i == -1 triggers ``len(self) + i`` selection that points to the
+            # last element, not before-the-first one, subtracting len(self)
+            # compensates that.
+            if end_slice == -1:
+                end_slice -= len(self)
+            if start_slice == -1:
+                start_slice -= len(self)
+
+        return start_slice, end_slice
+
+    def delete(
+        self, loc: int | np.integer | list[int] | npt.NDArray[np.integer]
+    ) -> Self:
+        """
+        Make new Index with passed location(-s) deleted.
+
+        Parameters
+        ----------
+        loc : int or list of int
+            Location of item(-s) which will be deleted.
+            Use a list of locations to delete more than one value at the same time.
+
+        Returns
+        -------
+        Index
+            Will be same type as self, except for RangeIndex.
+
+        See Also
+        --------
+        numpy.delete : Delete any rows and column from NumPy array (ndarray).
+
+        Examples
+        --------
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> idx.delete(1)
+        Index(['a', 'c'], dtype='object')
+
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> idx.delete([0, 2])
+        Index(['b'], dtype='object')
+        """
+        values = self._values
+        res_values: ArrayLike
+        if isinstance(values, np.ndarray):
+            # TODO(__array_function__): special casing will be unnecessary
+            res_values = np.delete(values, loc)
+        else:
+            res_values = values.delete(loc)
+
+        # _constructor so RangeIndex-> Index with an int64 dtype
+        return self._constructor._simple_new(res_values, name=self.name)
+
+    def insert(self, loc: int, item) -> Index:
+        """
+        Make new Index inserting new item at location.
+
+        Follows Python numpy.insert semantics for negative values.
+
+        Parameters
+        ----------
+        loc : int
+            The integer location where the new item will be inserted.
+        item : object
+            The new item to be inserted into the Index.
+
+        Returns
+        -------
+        Index
+            Returns a new Index object resulting from inserting the specified item at
+            the specified location within the original Index.
+
+        See Also
+        --------
+        Index.append : Append a collection of Indexes together.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> idx.insert(1, "x")
+        Index(['a', 'x', 'b', 'c'], dtype='object')
+        """
+        item = lib.item_from_zerodim(item)
+        if is_valid_na_for_dtype(item, self.dtype) and self.dtype != object:
+            item = self._na_value
+
+        arr = self._values
+
+        if using_string_dtype() and len(self) == 0 and self.dtype == np.object_:
+            # special case: if we are an empty object-dtype Index, also
+            # take into account the inserted item for the resulting dtype
+            # (https://github.com/pandas-dev/pandas/pull/60797)
+            dtype = self._find_common_type_compat(item)
+            if dtype != self.dtype:
+                return self.astype(dtype).insert(loc, item)
+
+        try:
+            if isinstance(arr, ExtensionArray):
+                res_values = arr.insert(loc, item)
+                return type(self)._simple_new(res_values, name=self.name)
+            else:
+                item = self._validate_fill_value(item)
+        except (TypeError, ValueError, LossySetitemError):
+            # e.g. trying to insert an integer into a DatetimeIndex
+            #  We cannot keep the same dtype, so cast to the (often object)
+            #  minimal shared dtype before doing the insert.
+            dtype = self._find_common_type_compat(item)
+            if dtype == self.dtype:
+                # EA's might run into recursion errors if loc is invalid
+                raise
+            return self.astype(dtype).insert(loc, item)
+
+        if arr.dtype != object or not isinstance(
+            item, (tuple, np.datetime64, np.timedelta64)
+        ):
+            # with object-dtype we need to worry about numpy incorrectly casting
+            # dt64/td64 to integer, also about treating tuples as sequences
+            # special-casing dt64/td64 https://github.com/numpy/numpy/issues/12550
+            casted = arr.dtype.type(item)
+            new_values = np.insert(arr, loc, casted)
+
+        else:
+            # error: No overload variant of "insert" matches argument types
+            # "ndarray[Any, Any]", "int", "None"
+            new_values = np.insert(arr, loc, None)  # type: ignore[call-overload]
+            loc = loc if loc >= 0 else loc - 1
+            new_values[loc] = item
+
+        # GH#51363 stopped doing dtype inference here
+        out = Index(new_values, dtype=new_values.dtype, name=self.name)
+        return out
+
+    def drop(
+        self,
+        labels: Index | np.ndarray | Iterable[Hashable],
+        errors: IgnoreRaise = "raise",
+    ) -> Index:
+        """
+        Make new Index with passed list of labels deleted.
+
+        Parameters
+        ----------
+        labels : array-like or scalar
+            Array-like object or a scalar value, representing the labels to be removed
+            from the Index.
+        errors : {'ignore', 'raise'}, default 'raise'
+            If 'ignore', suppress error and existing labels are dropped.
+
+        Returns
+        -------
+        Index
+            Will be same type as self, except for RangeIndex.
+
+        Raises
+        ------
+        KeyError
+            If not all of the labels are found in the selected axis
+
+        See Also
+        --------
+        Index.dropna : Return Index without NA/NaN values.
+        Index.drop_duplicates : Return Index with duplicate values removed.
+
+        Examples
+        --------
+        >>> idx = pd.Index(["a", "b", "c"])
+        >>> idx.drop(["a"])
+        Index(['b', 'c'], dtype='object')
+        """
+        if not isinstance(labels, Index):
+            # avoid materializing e.g. RangeIndex
+            arr_dtype = "object" if self.dtype == "object" else None
+            labels = com.index_labels_to_array(labels, dtype=arr_dtype)
+
+        indexer = self.get_indexer_for(labels)
+        mask = indexer == -1
+        if mask.any():
+            if errors != "ignore":
+                raise KeyError(f"{labels[mask].tolist()} not found in axis")
+            indexer = indexer[~mask]
+        return self.delete(indexer)
+
+    @final
+    def infer_objects(self, copy: bool = True) -> Index:
+        """
+        If we have an object dtype, try to infer a non-object dtype.
+
+        Parameters
+        ----------
+        copy : bool, default True
+            Whether to make a copy in cases where no inference occurs.
+        """
+        if self._is_multi:
+            raise NotImplementedError(
+                "infer_objects is not implemented for MultiIndex. "
+                "Use index.to_frame().infer_objects() instead."
+            )
+        if self.dtype != object:
+            return self.copy() if copy else self
+
+        values = self._values
+        values = cast("npt.NDArray[np.object_]", values)
+        res_values = lib.maybe_convert_objects(
+            values,
+            convert_non_numeric=True,
+        )
+        if copy and res_values is values:
+            return self.copy()
+        result = Index(res_values, name=self.name)
+        if not copy and res_values is values and self._references is not None:
+            result._references = self._references
+            result._references.add_index_reference(result)
+        return result
+
+    @final
+    def diff(self, periods: int = 1) -> Index:
+        """
+        Computes the difference between consecutive values in the Index object.
+
+        If periods is greater than 1, computes the difference between values that
+        are `periods` number of positions apart.
+
+        Parameters
+        ----------
+        periods : int, optional
+            The number of positions between the current and previous
+            value to compute the difference with. Default is 1.
+
+        Returns
+        -------
+        Index
+            A new Index object with the computed differences.
+
+        Examples
+        --------
+        >>> import pandas as pd
+        >>> idx = pd.Index([10, 20, 30, 40, 50])
+        >>> idx.diff()
+        Index([nan, 10.0, 10.0, 10.0, 10.0], dtype='float64')
+
+        """
+        return Index(self.to_series().diff(periods))
+
+    def round(self, decimals: int = 0) -> Self:
+        """
+        Round each value in the Index to the given number of decimals.
+
+        Parameters
+        ----------
+        decimals : int, optional
+            Number of decimal places to round to. If decimals is negative,
+            it specifies the number of positions to the left of the decimal point.
+
+        Returns
+        -------
+        Index
+            A new Index with the rounded values.
+
+        Examples
+        --------
+        >>> import pandas as pd
+        >>> idx = pd.Index([10.1234, 20.5678, 30.9123, 40.4567, 50.7890])
+        >>> idx.round(decimals=2)
+        Index([10.12, 20.57, 30.91, 40.46, 50.79], dtype='float64')
+
+        """
+        return self._constructor(self.to_series().round(decimals))
+
+    # --------------------------------------------------------------------
+    # Generated Arithmetic, Comparison, and Unary Methods
+
+    def _cmp_method(self, other, op):
+        """
+        Wrapper used to dispatch comparison operations.
+        """
+        if self.is_(other):
+            # fastpath
+            if op in {operator.eq, operator.le, operator.ge}:
+                arr = np.ones(len(self), dtype=bool)
+                if self._can_hold_na and not isinstance(self, ABCMultiIndex):
+                    # TODO: should set MultiIndex._can_hold_na = False?
+                    arr[self.isna()] = False
+                return arr
+            elif op is operator.ne:
+                arr = np.zeros(len(self), dtype=bool)
+                if self._can_hold_na and not isinstance(self, ABCMultiIndex):
+                    arr[self.isna()] = True
+                return arr
+
+        if isinstance(other, (np.ndarray, Index, ABCSeries, ExtensionArray)) and len(
+            self
+        ) != len(other):
+            raise ValueError("Lengths must match to compare")
+
+        if not isinstance(other, ABCMultiIndex):
+            other = extract_array(other, extract_numpy=True)
+        else:
+            other = np.asarray(other)
+
+        if is_object_dtype(self.dtype) and isinstance(other, ExtensionArray):
+            # e.g. PeriodArray, Categorical
+            result = op(self._values, other)
+
+        elif isinstance(self._values, ExtensionArray):
+            result = op(self._values, other)
+
+        elif is_object_dtype(self.dtype) and not isinstance(self, ABCMultiIndex):
+            # don't pass MultiIndex
+            result = ops.comp_method_OBJECT_ARRAY(op, self._values, other)
+
+        else:
+            result = ops.comparison_op(self._values, other, op)
+
+        return result
+
+    @final
+    def _logical_method(self, other, op):
+        res_name = ops.get_op_result_name(self, other)
+
+        lvalues = self._values
+        rvalues = extract_array(other, extract_numpy=True, extract_range=True)
+
+        res_values = ops.logical_op(lvalues, rvalues, op)
+        return self._construct_result(res_values, name=res_name)
+
+    @final
+    def _construct_result(self, result, name):
+        if isinstance(result, tuple):
+            return (
+                Index(result[0], name=name, dtype=result[0].dtype),
+                Index(result[1], name=name, dtype=result[1].dtype),
+            )
+        return Index(result, name=name, dtype=result.dtype)
+
+    def _arith_method(self, other, op):
+        if (
+            isinstance(other, Index)
+            and is_object_dtype(other.dtype)
+            and type(other) is not Index
+        ):
+            # We return NotImplemented for object-dtype index *subclasses* so they have
+            # a chance to implement ops before we unwrap them.
+            # See https://github.com/pandas-dev/pandas/issues/31109
+            return NotImplemented
+
+        return super()._arith_method(other, op)
+
+    @final
+    def _unary_method(self, op):
+        result = op(self._values)
+        return Index(result, name=self.name)
+
+    def __abs__(self) -> Index:
+        return self._unary_method(operator.abs)
+
+    def __neg__(self) -> Index:
+        return self._unary_method(operator.neg)
+
+    def __pos__(self) -> Index:
+        return self._unary_method(operator.pos)
+
+    def __invert__(self) -> Index:
+        # GH#8875
+        return self._unary_method(operator.inv)
+
+    # --------------------------------------------------------------------
+    # Reductions
+
+    def any(self, *args, **kwargs):
+        """
+        Return whether any element is Truthy.
+
+        Parameters
+        ----------
+        *args
+            Required for compatibility with numpy.
+        **kwargs
+            Required for compatibility with numpy.
+
+        Returns
+        -------
+        bool or array-like (if axis is specified)
+            A single element array-like may be converted to bool.
+
+        See Also
+        --------
+        Index.all : Return whether all elements are True.
+        Series.all : Return whether all elements are True.
+
+        Notes
+        -----
+        Not a Number (NaN), positive infinity and negative infinity
+        evaluate to True because these are not equal to zero.
+
+        Examples
+        --------
+        >>> index = pd.Index([0, 1, 2])
+        >>> index.any()
+        True
+
+        >>> index = pd.Index([0, 0, 0])
+        >>> index.any()
+        False
+        """
+        nv.validate_any(args, kwargs)
+        self._maybe_disable_logical_methods("any")
+        vals = self._values
+        if not isinstance(vals, np.ndarray):
+            # i.e. EA, call _reduce instead of "any" to get TypeError instead
+            #  of AttributeError
+            return vals._reduce("any")
+        return np.any(vals)
+
+    def all(self, *args, **kwargs):
+        """
+        Return whether all elements are Truthy.
+
+        Parameters
+        ----------
+        *args
+            Required for compatibility with numpy.
+        **kwargs
+            Required for compatibility with numpy.
+
+        Returns
+        -------
+        bool or array-like (if axis is specified)
+            A single element array-like may be converted to bool.
+
+        See Also
+        --------
+        Index.any : Return whether any element in an Index is True.
+        Series.any : Return whether any element in a Series is True.
+        Series.all : Return whether all elements in a Series are True.
+
+        Notes
+        -----
+        Not a Number (NaN), positive infinity and negative infinity
+        evaluate to True because these are not equal to zero.
+
+        Examples
+        --------
+        True, because nonzero integers are considered True.
+
+        >>> pd.Index([1, 2, 3]).all()
+        True
+
+        False, because ``0`` is considered False.
+
+        >>> pd.Index([0, 1, 2]).all()
+        False
+        """
+        nv.validate_all(args, kwargs)
+        self._maybe_disable_logical_methods("all")
+        vals = self._values
+        if not isinstance(vals, np.ndarray):
+            # i.e. EA, call _reduce instead of "all" to get TypeError instead
+            #  of AttributeError
+            return vals._reduce("all")
+        return np.all(vals)
+
+    @final
+    def _maybe_disable_logical_methods(self, opname: str_t) -> None:
+        """
+        raise if this Index subclass does not support any or all.
+        """
+        if isinstance(self, ABCMultiIndex):
+            raise TypeError(f"cannot perform {opname} with {type(self).__name__}")
+
+    @Appender(IndexOpsMixin.argmin.__doc__)
+    def argmin(
+        self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs
+    ) -> int:
+        nv.validate_argmin(args, kwargs)
+        nv.validate_minmax_axis(axis)
+
+        if not self._is_multi and self.hasnans:
+            if not skipna:
+                raise ValueError("Encountered an NA value with skipna=False")
+            elif self._isnan.all():
+                raise ValueError("Encountered all NA values")
+
+        return super().argmin(skipna=skipna)
+
+    @Appender(IndexOpsMixin.argmax.__doc__)
+    def argmax(
+        self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs
+    ) -> int:
+        nv.validate_argmax(args, kwargs)
+        nv.validate_minmax_axis(axis)
+
+        if not self._is_multi and self.hasnans:
+            if not skipna:
+                raise ValueError("Encountered an NA value with skipna=False")
+            elif self._isnan.all():
+                raise ValueError("Encountered all NA values")
+        return super().argmax(skipna=skipna)
+
+    def min(self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs):
+        """
+        Return the minimum value of the Index.
+
+        Parameters
+        ----------
+        axis : {None}
+            Dummy argument for consistency with Series.
+        skipna : bool, default True
+            Exclude NA/null values when showing the result.
+        *args, **kwargs
+            Additional arguments and keywords for compatibility with NumPy.
+
+        Returns
+        -------
+        scalar
+            Minimum value.
+
+        See Also
+        --------
+        Index.max : Return the maximum value of the object.
+        Series.min : Return the minimum value in a Series.
+        DataFrame.min : Return the minimum values in a DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index([3, 2, 1])
+        >>> idx.min()
+        1
+
+        >>> idx = pd.Index(["c", "b", "a"])
+        >>> idx.min()
+        'a'
+
+        For a MultiIndex, the minimum is determined lexicographically.
+
+        >>> idx = pd.MultiIndex.from_product([("a", "b"), (2, 1)])
+        >>> idx.min()
+        ('a', 1)
+        """
+        nv.validate_min(args, kwargs)
+        nv.validate_minmax_axis(axis)
+
+        if not len(self):
+            return self._na_value
+
+        if len(self) and self.is_monotonic_increasing:
+            # quick check
+            first = self[0]
+            if not isna(first):
+                return first
+
+        if not self._is_multi and self.hasnans:
+            # Take advantage of cache
+            mask = self._isnan
+            if not skipna or mask.all():
+                return self._na_value
+
+        if not self._is_multi and not isinstance(self._values, np.ndarray):
+            return self._values._reduce(name="min", skipna=skipna)
+
+        return nanops.nanmin(self._values, skipna=skipna)
+
+    def max(self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs):
+        """
+        Return the maximum value of the Index.
+
+        Parameters
+        ----------
+        axis : int, optional
+            For compatibility with NumPy. Only 0 or None are allowed.
+        skipna : bool, default True
+            Exclude NA/null values when showing the result.
+        *args, **kwargs
+            Additional arguments and keywords for compatibility with NumPy.
+
+        Returns
+        -------
+        scalar
+            Maximum value.
+
+        See Also
+        --------
+        Index.min : Return the minimum value in an Index.
+        Series.max : Return the maximum value in a Series.
+        DataFrame.max : Return the maximum values in a DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.Index([3, 2, 1])
+        >>> idx.max()
+        3
+
+        >>> idx = pd.Index(["c", "b", "a"])
+        >>> idx.max()
+        'c'
+
+        For a MultiIndex, the maximum is determined lexicographically.
+
+        >>> idx = pd.MultiIndex.from_product([("a", "b"), (2, 1)])
+        >>> idx.max()
+        ('b', 2)
+        """
+
+        nv.validate_max(args, kwargs)
+        nv.validate_minmax_axis(axis)
+
+        if not len(self):
+            return self._na_value
+
+        if len(self) and self.is_monotonic_increasing:
+            # quick check
+            last = self[-1]
+            if not isna(last):
+                return last
+
+        if not self._is_multi and self.hasnans:
+            # Take advantage of cache
+            mask = self._isnan
+            if not skipna or mask.all():
+                return self._na_value
+
+        if not self._is_multi and not isinstance(self._values, np.ndarray):
+            return self._values._reduce(name="max", skipna=skipna)
+
+        return nanops.nanmax(self._values, skipna=skipna)
+
+    # --------------------------------------------------------------------
+
+    @final
+    @property
+    def shape(self) -> Shape:
+        """
+        Return a tuple of the shape of the underlying data.
+
+        See Also
+        --------
+        Index.size: Return the number of elements in the underlying data.
+        Index.ndim: Number of dimensions of the underlying data, by definition 1.
+        Index.dtype: Return the dtype object of the underlying data.
+        Index.values: Return an array representing the data in the Index.
+
+        Examples
+        --------
+        >>> idx = pd.Index([1, 2, 3])
+        >>> idx
+        Index([1, 2, 3], dtype='int64')
+        >>> idx.shape
+        (3,)
+        """
+        # See GH#27775, GH#27384 for history/reasoning in how this is defined.
+        return (len(self),)
+
+
+def maybe_sequence_to_range(sequence) -> Any | range:
+    """
+    Convert a 1D, non-pandas sequence to a range if possible.
+
+    Returns the input if not possible.
+
+    Parameters
+    ----------
+    sequence : 1D sequence
+    names : sequence of str
+
+    Returns
+    -------
+    Any : input or range
+    """
+    if isinstance(sequence, (range, ExtensionArray)):
+        return sequence
+    elif len(sequence) == 1 or lib.infer_dtype(sequence, skipna=False) != "integer":
+        return sequence
+    elif isinstance(sequence, (ABCSeries, Index)) and not (
+        isinstance(sequence.dtype, np.dtype) and sequence.dtype.kind == "i"
+    ):
+        return sequence
+    if len(sequence) == 0:
+        return range(0)
+    try:
+        np_sequence = np.asarray(sequence, dtype=np.int64)
+    except OverflowError:
+        return sequence
+    diff = np_sequence[1] - np_sequence[0]
+    if diff == 0:
+        return sequence
+    elif len(sequence) == 2 or lib.is_sequence_range(np_sequence, diff):
+        return range(np_sequence[0], np_sequence[-1] + diff, diff)
+    else:
+        return sequence
+
+
+def ensure_index_from_sequences(sequences, names=None) -> Index:
+    """
+    Construct an index from sequences of data.
+
+    A single sequence returns an Index. Many sequences returns a
+    MultiIndex.
+
+    Parameters
+    ----------
+    sequences : sequence of sequences
+    names : sequence of str
+
+    Returns
+    -------
+    index : Index or MultiIndex
+
+    Examples
+    --------
+    >>> ensure_index_from_sequences([[1, 2, 4]], names=["name"])
+    Index([1, 2, 4], dtype='int64', name='name')
+
+    >>> ensure_index_from_sequences([["a", "a"], ["a", "b"]], names=["L1", "L2"])
+    MultiIndex([('a', 'a'),
+                ('a', 'b')],
+               names=['L1', 'L2'])
+
+    See Also
+    --------
+    ensure_index
+    """
+    from pandas.core.indexes.api import default_index
+    from pandas.core.indexes.multi import MultiIndex
+
+    if len(sequences) == 0:
+        return default_index(0)
+    elif len(sequences) == 1:
+        if names is not None:
+            names = names[0]
+        return Index(maybe_sequence_to_range(sequences[0]), name=names)
+    else:
+        # TODO: Apply maybe_sequence_to_range to sequences?
+        return MultiIndex.from_arrays(sequences, names=names)
+
+
+def ensure_index(index_like: Axes, copy: bool = False) -> Index:
+    """
+    Ensure that we have an index from some index-like object.
+
+    Parameters
+    ----------
+    index_like : sequence
+        An Index or other sequence
+    copy : bool, default False
+
+    Returns
+    -------
+    index : Index or MultiIndex
+
+    See Also
+    --------
+    ensure_index_from_sequences
+
+    Examples
+    --------
+    >>> ensure_index(["a", "b"])
+    Index(['a', 'b'], dtype='object')
+
+    >>> ensure_index([("a", "a"), ("b", "c")])
+    Index([('a', 'a'), ('b', 'c')], dtype='object')
+
+    >>> ensure_index([["a", "a"], ["b", "c"]])
+    MultiIndex([('a', 'b'),
+            ('a', 'c')],
+           )
+    """
+    if isinstance(index_like, Index):
+        if copy:
+            index_like = index_like.copy()
+        return index_like
+
+    if isinstance(index_like, ABCSeries):
+        name = index_like.name
+        return Index(index_like, name=name, copy=copy)
+
+    if is_iterator(index_like):
+        index_like = list(index_like)
+
+    if isinstance(index_like, list):
+        if type(index_like) is not list:
+            # must check for exactly list here because of strict type
+            # check in clean_index_list
+            index_like = list(index_like)
+
+        if len(index_like) and lib.is_all_arraylike(index_like):
+            from pandas.core.indexes.multi import MultiIndex
+
+            return MultiIndex.from_arrays(index_like)
+        else:
+            return Index(index_like, copy=copy, tupleize_cols=False)
+    else:
+        return Index(index_like, copy=copy)
+
+
+def trim_front(strings: list[str]) -> list[str]:
+    """
+    Trims leading spaces evenly among all strings.
+
+    Examples
+    --------
+    >>> trim_front([" a", " b"])
+    ['a', 'b']
+
+    >>> trim_front([" a", " "])
+    ['a', '']
+    """
+    if not strings:
+        return strings
+    smallest_leading_space = min(len(x) - len(x.lstrip()) for x in strings)
+    if smallest_leading_space > 0:
+        strings = [x[smallest_leading_space:] for x in strings]
+    return strings
+
+
+def _validate_join_method(method: str) -> None:
+    if method not in ["left", "right", "inner", "outer"]:
+        raise ValueError(f"do not recognize join method {method}")
+
+
+def maybe_extract_name(name, obj, cls) -> Hashable:
+    """
+    If no name is passed, then extract it from data, validating hashability.
+    """
+    if name is None and isinstance(obj, (Index, ABCSeries)):
+        # Note we don't just check for "name" attribute since that would
+        #  pick up e.g. dtype.name
+        name = obj.name
+
+    # GH#29069
+    if not is_hashable(name):
+        raise TypeError(f"{cls.__name__}.name must be a hashable type")
+
+    return name
+
+
+def get_unanimous_names(*indexes: Index) -> tuple[Hashable, ...]:
+    """
+    Return common name if all indices agree, otherwise None (level-by-level).
+
+    Parameters
+    ----------
+    indexes : list of Index objects
+
+    Returns
+    -------
+    list
+        A list representing the unanimous 'names' found.
+    """
+    name_tups = (tuple(i.names) for i in indexes)
+    name_sets = ({*ns} for ns in zip_longest(*name_tups))
+    names = tuple(ns.pop() if len(ns) == 1 else None for ns in name_sets)
+    return names
+
+
+def _unpack_nested_dtype(other: Index) -> DtypeObj:
+    """
+    When checking if our dtype is comparable with another, we need
+    to unpack CategoricalDtype to look at its categories.dtype.
+
+    Parameters
+    ----------
+    other : Index
+
+    Returns
+    -------
+    np.dtype or ExtensionDtype
+    """
+    dtype = other.dtype
+    if isinstance(dtype, CategoricalDtype):
+        # If there is ever a SparseIndex, this could get dispatched
+        #  here too.
+        return dtype.categories.dtype
+    elif isinstance(dtype, ArrowDtype):
+        # GH 53617
+        import pyarrow as pa
+
+        if pa.types.is_dictionary(dtype.pyarrow_dtype):
+            other = other[:0].astype(ArrowDtype(dtype.pyarrow_dtype.value_type))
+    return other.dtype
+
+
+def _maybe_try_sort(result: Index | ArrayLike, sort: bool | None):
+    if sort is not False:
+        try:
+            # error: Incompatible types in assignment (expression has type
+            # "Union[ExtensionArray, ndarray[Any, Any], Index, Series,
+            # Tuple[Union[Union[ExtensionArray, ndarray[Any, Any]], Index, Series],
+            # ndarray[Any, Any]]]", variable has type "Union[Index,
+            # Union[ExtensionArray, ndarray[Any, Any]]]")
+            result = algos.safe_sort(result)  # type: ignore[assignment]
+        except TypeError as err:
+            if sort is True:
+                raise
+            warnings.warn(
+                f"{err}, sort order is undefined for incomparable objects.",
+                RuntimeWarning,
+                stacklevel=find_stack_level(),
+            )
+    return result
+
+
+def get_values_for_csv(
+    values: ArrayLike,
+    *,
+    date_format,
+    na_rep: str = "nan",
+    quoting=None,
+    float_format=None,
+    decimal: str = ".",
+) -> npt.NDArray[np.object_]:
+    """
+    Convert to types which can be consumed by the standard library's
+    csv.writer.writerows.
+    """
+    if isinstance(values, Categorical) and values.categories.dtype.kind in "Mm":
+        # GH#40754 Convert categorical datetimes to datetime array
+        values = algos.take_nd(
+            values.categories._values,
+            ensure_platform_int(values._codes),
+            fill_value=na_rep,
+        )
+
+    values = ensure_wrapped_if_datetimelike(values)
+
+    if isinstance(values, (DatetimeArray, TimedeltaArray)):
+        if values.ndim == 1:
+            result = values._format_native_types(na_rep=na_rep, date_format=date_format)
+            result = result.astype(object, copy=False)
+            return result
+
+        # GH#21734 Process every column separately, they might have different formats
+        results_converted = []
+        for i in range(len(values)):
+            result = values[i, :]._format_native_types(
+                na_rep=na_rep, date_format=date_format
+            )
+            results_converted.append(result.astype(object, copy=False))
+        return np.vstack(results_converted)
+
+    elif isinstance(values.dtype, PeriodDtype):
+        # TODO: tests that get here in column path
+        values = cast("PeriodArray", values)
+        res = values._format_native_types(na_rep=na_rep, date_format=date_format)
+        return res
+
+    elif isinstance(values.dtype, IntervalDtype):
+        # TODO: tests that get here in column path
+        values = cast("IntervalArray", values)
+        mask = values.isna()
+        if not quoting:
+            result = np.asarray(values).astype(str)
+        else:
+            result = np.array(values, dtype=object, copy=True)
+
+        result[mask] = na_rep
+        return result
+
+    elif values.dtype.kind == "f" and not isinstance(values.dtype, SparseDtype):
+        # see GH#13418: no special formatting is desired at the
+        # output (important for appropriate 'quoting' behaviour),
+        # so do not pass it through the FloatArrayFormatter
+        if float_format is None and decimal == ".":
+            mask = isna(values)
+
+            if not quoting:
+                values = values.astype(str)
+            else:
+                values = np.array(values, dtype="object")
+
+            values[mask] = na_rep
+            values = values.astype(object, copy=False)
+            return values
+
+        from pandas.io.formats.format import FloatArrayFormatter
+
+        formatter = FloatArrayFormatter(
+            values,
+            na_rep=na_rep,
+            float_format=float_format,
+            decimal=decimal,
+            quoting=quoting,
+            fixed_width=False,
+        )
+        res = formatter.get_result_as_array()
+        res = res.astype(object, copy=False)
+        return res
+
+    elif isinstance(values, ExtensionArray):
+        mask = isna(values)
+
+        new_values = np.asarray(values.astype(object))
+        new_values[mask] = na_rep
+        return new_values
+
+    else:
+        mask = isna(values)
+        itemsize = writers.word_len(na_rep)
+
+        if values.dtype != _dtype_obj and not quoting and itemsize:
+            values = values.astype(str)
+            if values.dtype.itemsize / np.dtype("U1").itemsize < itemsize:
+                # enlarge for the na_rep
+                values = values.astype(f"<U{itemsize}")
+        else:
+            values = np.array(values, dtype="object")
+
+        values[mask] = na_rep
+        values = values.astype(object, copy=False)
+        return values
diff --git a/pandas/pandas/core/indexes/category.py b/pandas/pandas/core/indexes/category.py
new file mode 100644
index 0000000000000..d20a84449fb85
--- /dev/null
+++ b/pandas/pandas/core/indexes/category.py
@@ -0,0 +1,536 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import index as libindex
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+    set_module,
+)
+
+from pandas.core.dtypes.common import is_scalar
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import CategoricalDtype
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+)
+
+from pandas.core.arrays.categorical import (
+    Categorical,
+    contains,
+)
+from pandas.core.construction import extract_array
+from pandas.core.indexes.base import (
+    Index,
+    maybe_extract_name,
+)
+from pandas.core.indexes.extension import (
+    NDArrayBackedExtensionIndex,
+    inherit_names,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import (
+        Dtype,
+        DtypeObj,
+        Self,
+        npt,
+    )
+
+
+@inherit_names(
+    [
+        "argsort",
+        "tolist",
+        "codes",
+        "categories",
+        "ordered",
+        "_reverse_indexer",
+        "searchsorted",
+        "min",
+        "max",
+    ],
+    Categorical,
+)
+@inherit_names(
+    [
+        "rename_categories",
+        "reorder_categories",
+        "add_categories",
+        "remove_categories",
+        "remove_unused_categories",
+        "set_categories",
+        "as_ordered",
+        "as_unordered",
+    ],
+    Categorical,
+    wrap=True,
+)
+@set_module("pandas")
+class CategoricalIndex(NDArrayBackedExtensionIndex):
+    """
+    Index based on an underlying :class:`Categorical`.
+
+    CategoricalIndex, like Categorical, can only take on a limited,
+    and usually fixed, number of possible values (`categories`). Also,
+    like Categorical, it might have an order, but numerical operations
+    (additions, divisions, ...) are not possible.
+
+    Parameters
+    ----------
+    data : array-like (1-dimensional)
+        The values of the categorical. If `categories` are given, values not in
+        `categories` will be replaced with NaN.
+    categories : index-like, optional
+        The categories for the categorical. Items need to be unique.
+        If the categories are not given here (and also not in `dtype`), they
+        will be inferred from the `data`.
+    ordered : bool, optional
+        Whether or not this categorical is treated as an ordered
+        categorical. If not given here or in `dtype`, the resulting
+        categorical will be unordered.
+    dtype : CategoricalDtype or "category", optional
+        If :class:`CategoricalDtype`, cannot be used together with
+        `categories` or `ordered`.
+    copy : bool, default False
+        Make a copy of input ndarray.
+    name : object, optional
+        Name to be stored in the index.
+
+    Attributes
+    ----------
+    codes
+    categories
+    ordered
+
+    Methods
+    -------
+    rename_categories
+    reorder_categories
+    add_categories
+    remove_categories
+    remove_unused_categories
+    set_categories
+    as_ordered
+    as_unordered
+    map
+
+    Raises
+    ------
+    ValueError
+        If the categories do not validate.
+    TypeError
+        If an explicit ``ordered=True`` is given but no `categories` and the
+        `values` are not sortable.
+
+    See Also
+    --------
+    Index : The base pandas Index type.
+    Categorical : A categorical array.
+    CategoricalDtype : Type for categorical data.
+
+    Notes
+    -----
+    See the `user guide
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/advanced.html#categoricalindex>`__
+    for more.
+
+    Examples
+    --------
+    >>> pd.CategoricalIndex(["a", "b", "c", "a", "b", "c"])
+    CategoricalIndex(['a', 'b', 'c', 'a', 'b', 'c'],
+                     categories=['a', 'b', 'c'], ordered=False, dtype='category')
+
+    ``CategoricalIndex`` can also be instantiated from a ``Categorical``:
+
+    >>> c = pd.Categorical(["a", "b", "c", "a", "b", "c"])
+    >>> pd.CategoricalIndex(c)
+    CategoricalIndex(['a', 'b', 'c', 'a', 'b', 'c'],
+                     categories=['a', 'b', 'c'], ordered=False, dtype='category')
+
+    Ordered ``CategoricalIndex`` can have a min and max value.
+
+    >>> ci = pd.CategoricalIndex(
+    ...     ["a", "b", "c", "a", "b", "c"], ordered=True, categories=["c", "b", "a"]
+    ... )
+    >>> ci
+    CategoricalIndex(['a', 'b', 'c', 'a', 'b', 'c'],
+                     categories=['c', 'b', 'a'], ordered=True, dtype='category')
+    >>> ci.min()
+    'c'
+    """
+
+    _typ = "categoricalindex"
+    _data_cls = Categorical
+
+    @property
+    def _can_hold_strings(self):
+        return self.categories._can_hold_strings
+
+    @cache_readonly
+    def _should_fallback_to_positional(self) -> bool:
+        return self.categories._should_fallback_to_positional
+
+    codes: np.ndarray
+    categories: Index
+    ordered: bool | None
+    _data: Categorical
+    _values: Categorical
+
+    @property
+    def _engine_type(self) -> type[libindex.IndexEngine]:
+        # self.codes can have dtype int8, int16, int32 or int64, so we need
+        # to return the corresponding engine type (libindex.Int8Engine, etc.).
+        return {
+            np.int8: libindex.Int8Engine,
+            np.int16: libindex.Int16Engine,
+            np.int32: libindex.Int32Engine,
+            np.int64: libindex.Int64Engine,
+        }[self.codes.dtype.type]
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data=None,
+        categories=None,
+        ordered=None,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        name: Hashable | None = None,
+    ) -> Self:
+        name = maybe_extract_name(name, data, cls)
+
+        if is_scalar(data):
+            # GH#38944 include None here, which pre-2.0 subbed in []
+            cls._raise_scalar_data_error(data)
+
+        data = Categorical(
+            data, categories=categories, ordered=ordered, dtype=dtype, copy=copy
+        )
+
+        return cls._simple_new(data, name=name)
+
+    # --------------------------------------------------------------------
+
+    def _is_dtype_compat(self, other: Index) -> Categorical:
+        """
+        *this is an internal non-public method*
+
+        provide a comparison between the dtype of self and other (coercing if
+        needed)
+
+        Parameters
+        ----------
+        other : Index
+
+        Returns
+        -------
+        Categorical
+
+        Raises
+        ------
+        TypeError if the dtypes are not compatible
+        """
+        if isinstance(other.dtype, CategoricalDtype):
+            cat = extract_array(other)
+            cat = cast(Categorical, cat)
+            if not cat._categories_match_up_to_permutation(self._values):
+                raise TypeError(
+                    "categories must match existing categories when appending"
+                )
+
+        elif other._is_multi:
+            # preempt raising NotImplementedError in isna call
+            raise TypeError("MultiIndex is not dtype-compatible with CategoricalIndex")
+        else:
+            values = other
+
+            cat = Categorical(other, dtype=self.dtype)
+            other = CategoricalIndex(cat)
+            if not other.isin(values).all():
+                raise TypeError(
+                    "cannot append a non-category item to a CategoricalIndex"
+                )
+            cat = other._values
+
+            if not ((cat == values) | (isna(cat) & isna(values))).all():
+                # GH#37667 see test_equals_non_category
+                raise TypeError(
+                    "categories must match existing categories when appending"
+                )
+
+        return cat
+
+    def equals(self, other: object) -> bool:
+        """
+        Determine if two CategoricalIndex objects contain the same elements.
+
+        The order and orderedness of elements matters. The categories matter,
+        but the order of the categories matters only when ``ordered=True``.
+
+        Parameters
+        ----------
+        other : object
+            The CategoricalIndex object to compare with.
+
+        Returns
+        -------
+        bool
+            ``True`` if two :class:`pandas.CategoricalIndex` objects have equal
+            elements, ``False`` otherwise.
+
+        See Also
+        --------
+        Categorical.equals : Returns True if categorical arrays are equal.
+
+        Examples
+        --------
+        >>> ci = pd.CategoricalIndex(["a", "b", "c", "a", "b", "c"])
+        >>> ci2 = pd.CategoricalIndex(pd.Categorical(["a", "b", "c", "a", "b", "c"]))
+        >>> ci.equals(ci2)
+        True
+
+        The order of elements matters.
+
+        >>> ci3 = pd.CategoricalIndex(["c", "b", "a", "a", "b", "c"])
+        >>> ci.equals(ci3)
+        False
+
+        The orderedness also matters.
+
+        >>> ci4 = ci.as_ordered()
+        >>> ci.equals(ci4)
+        False
+
+        The categories matter, but the order of the categories matters only when
+        ``ordered=True``.
+
+        >>> ci5 = ci.set_categories(["a", "b", "c", "d"])
+        >>> ci.equals(ci5)
+        False
+
+        >>> ci6 = ci.set_categories(["b", "c", "a"])
+        >>> ci.equals(ci6)
+        True
+        >>> ci_ordered = pd.CategoricalIndex(
+        ...     ["a", "b", "c", "a", "b", "c"], ordered=True
+        ... )
+        >>> ci2_ordered = ci_ordered.set_categories(["b", "c", "a"])
+        >>> ci_ordered.equals(ci2_ordered)
+        False
+        """
+        if self.is_(other):
+            return True
+
+        if not isinstance(other, Index):
+            return False
+
+        try:
+            other = self._is_dtype_compat(other)
+        except (TypeError, ValueError):
+            return False
+
+        return self._data.equals(other)
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+
+    @property
+    def _formatter_func(self):
+        return self.categories._formatter_func
+
+    def _format_attrs(self):
+        """
+        Return a list of tuples of the (attr,formatted_value)
+        """
+        attrs: list[tuple[str, str | int | bool | None]]
+
+        attrs = [
+            (
+                "categories",
+                f"[{', '.join(self._data._repr_categories())}]",
+            ),
+            ("ordered", self.ordered),
+        ]
+        extra = super()._format_attrs()
+        return attrs + extra
+
+    # --------------------------------------------------------------------
+
+    @property
+    def inferred_type(self) -> str:
+        return "categorical"
+
+    @doc(Index.__contains__)
+    def __contains__(self, key: Any) -> bool:
+        # if key is a NaN, check if any NaN is in self.
+        if is_valid_na_for_dtype(key, self.categories.dtype):
+            return self.hasnans
+        if self.categories._typ == "rangeindex":
+            container: Index | libindex.IndexEngine | libindex.ExtensionEngine = (
+                self.categories
+            )
+        else:
+            container = self._engine
+        return contains(self, key, container=container)
+
+    def reindex(
+        self, target, method=None, level=None, limit: int | None = None, tolerance=None
+    ) -> tuple[Index, npt.NDArray[np.intp] | None]:
+        """
+        Create index with target's values (move/add/delete values as necessary)
+
+        Returns
+        -------
+        new_index : pd.Index
+            Resulting index
+        indexer : np.ndarray[np.intp] or None
+            Indices of output values in original index
+
+        """
+        if method is not None:
+            raise NotImplementedError(
+                "argument method is not implemented for CategoricalIndex.reindex"
+            )
+        if level is not None:
+            raise NotImplementedError(
+                "argument level is not implemented for CategoricalIndex.reindex"
+            )
+        if limit is not None:
+            raise NotImplementedError(
+                "argument limit is not implemented for CategoricalIndex.reindex"
+            )
+        return super().reindex(target)
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    def _maybe_cast_indexer(self, key) -> int:
+        # GH#41933: we have to do this instead of self._data._validate_scalar
+        #  because this will correctly get partial-indexing on Interval categories
+        try:
+            return self._data._unbox_scalar(key)
+        except KeyError:
+            if is_valid_na_for_dtype(key, self.categories.dtype):
+                return -1
+            raise
+
+    def _maybe_cast_listlike_indexer(self, values) -> CategoricalIndex:
+        if isinstance(values, CategoricalIndex):
+            values = values._data
+        if isinstance(values, Categorical):
+            # Indexing on codes is more efficient if categories are the same,
+            #  so we can apply some optimizations based on the degree of
+            #  dtype-matching.
+            cat = self._data._encode_with_my_categories(values)
+            codes = cat._codes
+        else:
+            codes = self.categories.get_indexer(values)
+            codes = codes.astype(self.codes.dtype, copy=False)
+            cat = self._data._from_backing_data(codes)
+        return type(self)._simple_new(cat)
+
+    # --------------------------------------------------------------------
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        return self.categories._is_comparable_dtype(dtype)
+
+    def map(self, mapper, na_action: Literal["ignore"] | None = None):
+        """
+        Map values using input an input mapping or function.
+
+        Maps the values (their categories, not the codes) of the index to new
+        categories. If the mapping correspondence is one-to-one the result is a
+        :class:`~pandas.CategoricalIndex` which has the same order property as
+        the original, otherwise an :class:`~pandas.Index` is returned.
+
+        If a `dict` or :class:`~pandas.Series` is used any unmapped category is
+        mapped to `NaN`. Note that if this happens an :class:`~pandas.Index`
+        will be returned.
+
+        Parameters
+        ----------
+        mapper : function, dict, or Series
+            Mapping correspondence.
+        na_action : {None, 'ignore'}, default 'ignore'
+            If 'ignore', propagate NaN values, without passing them to
+            the mapping correspondence.
+
+        Returns
+        -------
+        pandas.CategoricalIndex or pandas.Index
+            Mapped index.
+
+        See Also
+        --------
+        Index.map : Apply a mapping correspondence on an
+            :class:`~pandas.Index`.
+        Series.map : Apply a mapping correspondence on a
+            :class:`~pandas.Series`.
+        Series.apply : Apply more complex functions on a
+            :class:`~pandas.Series`.
+
+        Examples
+        --------
+        >>> idx = pd.CategoricalIndex(["a", "b", "c"])
+        >>> idx
+        CategoricalIndex(['a', 'b', 'c'], categories=['a', 'b', 'c'],
+                          ordered=False, dtype='category')
+        >>> idx.map(lambda x: x.upper())
+        CategoricalIndex(['A', 'B', 'C'], categories=['A', 'B', 'C'],
+                         ordered=False, dtype='category')
+        >>> idx.map({"a": "first", "b": "second", "c": "third"})
+        CategoricalIndex(['first', 'second', 'third'], categories=['first',
+                         'second', 'third'], ordered=False, dtype='category')
+
+        If the mapping is one-to-one the ordering of the categories is
+        preserved:
+
+        >>> idx = pd.CategoricalIndex(["a", "b", "c"], ordered=True)
+        >>> idx
+        CategoricalIndex(['a', 'b', 'c'], categories=['a', 'b', 'c'],
+                         ordered=True, dtype='category')
+        >>> idx.map({"a": 3, "b": 2, "c": 1})
+        CategoricalIndex([3, 2, 1], categories=[3, 2, 1], ordered=True,
+                         dtype='category')
+
+        If the mapping is not one-to-one an :class:`~pandas.Index` is returned:
+
+        >>> idx.map({"a": "first", "b": "second", "c": "first"})
+        Index(['first', 'second', 'first'], dtype='object')
+
+        If a `dict` is used, all unmapped categories are mapped to `NaN` and
+        the result is an :class:`~pandas.Index`:
+
+        >>> idx.map({"a": "first", "b": "second"})
+        Index(['first', 'second', nan], dtype='object')
+        """
+        mapped = self._values.map(mapper, na_action=na_action)
+        return Index(mapped, name=self.name)
+
+    def _concat(self, to_concat: list[Index], name: Hashable) -> Index:
+        # if calling index is category, don't check dtype of others
+        try:
+            cat = Categorical._concat_same_type(
+                [self._is_dtype_compat(c) for c in to_concat]
+            )
+        except TypeError:
+            # not all to_concat elements are among our categories (or NA)
+
+            res = concat_compat([x._values for x in to_concat])
+            return Index(res, name=name)
+        else:
+            return type(self)._simple_new(cat, name=name)
diff --git a/pandas/pandas/core/indexes/datetimelike.py b/pandas/pandas/core/indexes/datetimelike.py
new file mode 100644
index 0000000000000..8b316de30662c
--- /dev/null
+++ b/pandas/pandas/core/indexes/datetimelike.py
@@ -0,0 +1,854 @@
+"""
+Base and utility classes for tseries type pandas objects.
+"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+    final,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    Timedelta,
+    lib,
+)
+from pandas._libs.tslibs import (
+    BaseOffset,
+    Resolution,
+    Tick,
+    parsing,
+    to_offset,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    InvalidIndexError,
+    NullFrequencyError,
+)
+from pandas.util._decorators import (
+    Appender,
+    cache_readonly,
+    doc,
+)
+
+from pandas.core.dtypes.common import (
+    is_integer,
+    is_list_like,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    PeriodDtype,
+)
+
+from pandas.core.arrays import (
+    DatetimeArray,
+    ExtensionArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.datetimelike import DatetimeLikeArrayMixin
+import pandas.core.common as com
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.base import (
+    Index,
+    _index_shared_docs,
+)
+from pandas.core.indexes.extension import NDArrayBackedExtensionIndex
+from pandas.core.indexes.range import RangeIndex
+from pandas.core.tools.timedeltas import to_timedelta
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+    from datetime import datetime
+
+    from pandas._typing import (
+        Axis,
+        JoinHow,
+        Self,
+        npt,
+    )
+
+    from pandas import CategoricalIndex
+
+_index_doc_kwargs = dict(ibase._index_doc_kwargs)
+
+
+class DatetimeIndexOpsMixin(NDArrayBackedExtensionIndex, ABC):
+    """
+    Common ops mixin to support a unified interface datetimelike Index.
+    """
+
+    _can_hold_strings = False
+    _data: DatetimeArray | TimedeltaArray | PeriodArray
+
+    @doc(DatetimeLikeArrayMixin.mean)
+    def mean(self, *, skipna: bool = True, axis: int | None = 0):
+        return self._data.mean(skipna=skipna, axis=axis)
+
+    @property
+    def freq(self) -> BaseOffset | None:
+        """
+        Return the frequency object if it is set, otherwise None.
+
+        To learn more about the frequency strings, please see
+        :ref:`this link<timeseries.offset_aliases>`.
+
+        See Also
+        --------
+        DatetimeIndex.freq : Return the frequency object if it is set, otherwise None.
+        PeriodIndex.freq : Return the frequency object if it is set, otherwise None.
+
+        Examples
+        --------
+        >>> datetimeindex = pd.date_range(
+        ...     "2022-02-22 02:22:22", periods=10, tz="America/Chicago", freq="h"
+        ... )
+        >>> datetimeindex
+        DatetimeIndex(['2022-02-22 02:22:22-06:00', '2022-02-22 03:22:22-06:00',
+                       '2022-02-22 04:22:22-06:00', '2022-02-22 05:22:22-06:00',
+                       '2022-02-22 06:22:22-06:00', '2022-02-22 07:22:22-06:00',
+                       '2022-02-22 08:22:22-06:00', '2022-02-22 09:22:22-06:00',
+                       '2022-02-22 10:22:22-06:00', '2022-02-22 11:22:22-06:00'],
+                      dtype='datetime64[ns, America/Chicago]', freq='h')
+        >>> datetimeindex.freq
+        <Hour>
+        """
+        return self._data.freq
+
+    @freq.setter
+    def freq(self, value) -> None:
+        # error: Property "freq" defined in "PeriodArray" is read-only  [misc]
+        self._data.freq = value  # type: ignore[misc]
+
+    @property
+    def asi8(self) -> npt.NDArray[np.int64]:
+        return self._data.asi8
+
+    @property
+    @doc(DatetimeLikeArrayMixin.freqstr)
+    def freqstr(self) -> str:
+        from pandas import PeriodIndex
+
+        if self._data.freqstr is not None and isinstance(
+            self._data, (PeriodArray, PeriodIndex)
+        ):
+            freq = PeriodDtype(self._data.freq)._freqstr
+            return freq
+        else:
+            return self._data.freqstr  # type: ignore[return-value]
+
+    @cache_readonly
+    @abstractmethod
+    def _resolution_obj(self) -> Resolution: ...
+
+    @cache_readonly
+    @doc(DatetimeLikeArrayMixin.resolution)
+    def resolution(self) -> str:
+        return self._data.resolution
+
+    # ------------------------------------------------------------------------
+
+    @cache_readonly
+    def hasnans(self) -> bool:
+        return self._data._hasna
+
+    def equals(self, other: Any) -> bool:
+        """
+        Determines if two Index objects contain the same elements.
+        """
+        if self.is_(other):
+            return True
+
+        if not isinstance(other, Index):
+            return False
+        elif other.dtype.kind in "iufc":
+            return False
+        elif not isinstance(other, type(self)):
+            should_try = False
+            inferable = self._data._infer_matches
+            if other.dtype == object:
+                should_try = other.inferred_type in inferable
+            elif isinstance(other.dtype, CategoricalDtype):
+                other = cast("CategoricalIndex", other)
+                should_try = other.categories.inferred_type in inferable
+
+            if should_try:
+                try:
+                    other = type(self)(other)
+                except (ValueError, TypeError, OverflowError):
+                    # e.g.
+                    #  ValueError -> cannot parse str entry, or OutOfBoundsDatetime
+                    #  TypeError  -> trying to convert IntervalIndex to DatetimeIndex
+                    #  OverflowError -> Index([very_large_timedeltas])
+                    return False
+
+        if self.dtype != other.dtype:
+            # have different timezone
+            return False
+
+        return np.array_equal(self.asi8, other.asi8)
+
+    @Appender(Index.__contains__.__doc__)
+    def __contains__(self, key: Any) -> bool:
+        hash(key)
+        try:
+            self.get_loc(key)
+        except (KeyError, TypeError, ValueError, InvalidIndexError):
+            return False
+        return True
+
+    def _convert_tolerance(self, tolerance, target):
+        tolerance = np.asarray(to_timedelta(tolerance).to_numpy())
+        return super()._convert_tolerance(tolerance, target)
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+    _default_na_rep = "NaT"
+
+    def _format_with_header(
+        self, *, header: list[str], na_rep: str, date_format: str | None = None
+    ) -> list[str]:
+        # TODO: not reached in tests 2023-10-11
+        # matches base class except for whitespace padding and date_format
+        return header + list(
+            self._get_values_for_csv(na_rep=na_rep, date_format=date_format)
+        )
+
+    @property
+    def _formatter_func(self):
+        return self._data._formatter()
+
+    def _format_attrs(self):
+        """
+        Return a list of tuples of the (attr,formatted_value).
+        """
+        attrs = super()._format_attrs()
+        for attrib in self._attributes:
+            # iterating over _attributes prevents us from doing this for PeriodIndex
+            if attrib == "freq":
+                freq = self.freqstr
+                if freq is not None:
+                    freq = repr(freq)  # e.g. D -> 'D'
+                attrs.append(("freq", freq))
+        return attrs
+
+    @Appender(Index._summary.__doc__)
+    def _summary(self, name=None) -> str:
+        result = super()._summary(name=name)
+        if self.freq:
+            result += f"\nFreq: {self.freqstr}"
+
+        return result
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    @final
+    def _can_partial_date_slice(self, reso: Resolution) -> bool:
+        # e.g. test_getitem_setitem_periodindex
+        # History of conversation GH#3452, GH#3931, GH#2369, GH#14826
+        return reso > self._resolution_obj
+        # NB: for DTI/PI, not TDI
+
+    def _parsed_string_to_bounds(self, reso: Resolution, parsed):
+        raise NotImplementedError
+
+    def _parse_with_reso(self, label: str) -> tuple[datetime, Resolution]:
+        # overridden by TimedeltaIndex
+        try:
+            if self.freq is None or hasattr(self.freq, "rule_code"):
+                freq = self.freq
+        except NotImplementedError:
+            freq = getattr(self, "freqstr", getattr(self, "inferred_freq", None))
+
+        freqstr: str | None
+        if freq is not None and not isinstance(freq, str):
+            freqstr = freq.rule_code
+        else:
+            freqstr = freq
+
+        if isinstance(label, np.str_):
+            # GH#45580
+            label = str(label)
+
+        parsed, reso_str = parsing.parse_datetime_string_with_reso(label, freqstr)
+        reso = Resolution.from_attrname(reso_str)
+        return parsed, reso
+
+    def _get_string_slice(self, key: str) -> slice | npt.NDArray[np.intp]:
+        # overridden by TimedeltaIndex
+        parsed, reso = self._parse_with_reso(key)
+        try:
+            return self._partial_date_slice(reso, parsed)
+        except KeyError as err:
+            raise KeyError(key) from err
+
+    @final
+    def _partial_date_slice(
+        self,
+        reso: Resolution,
+        parsed: datetime,
+    ) -> slice | npt.NDArray[np.intp]:
+        """
+        Parameters
+        ----------
+        reso : Resolution
+        parsed : datetime
+
+        Returns
+        -------
+        slice or ndarray[intp]
+        """
+        if not self._can_partial_date_slice(reso):
+            raise ValueError
+
+        t1, t2 = self._parsed_string_to_bounds(reso, parsed)
+        vals = self._data._ndarray
+        unbox = self._data._unbox
+
+        if self.is_monotonic_increasing:
+            if len(self) and (
+                (t1 < self[0] and t2 < self[0]) or (t1 > self[-1] and t2 > self[-1])
+            ):
+                # we are out of range
+                raise KeyError
+
+            # TODO: does this depend on being monotonic _increasing_?
+
+            # a monotonic (sorted) series can be sliced
+            left = vals.searchsorted(unbox(t1), side="left")
+            right = vals.searchsorted(unbox(t2), side="right")
+            return slice(left, right)
+
+        else:
+            lhs_mask = vals >= unbox(t1)
+            rhs_mask = vals <= unbox(t2)
+
+            # try to find the dates
+            return (lhs_mask & rhs_mask).nonzero()[0]
+
+    def _maybe_cast_slice_bound(self, label, side: str):
+        """
+        If label is a string, cast it to scalar type according to resolution.
+
+        Parameters
+        ----------
+        label : object
+        side : {'left', 'right'}
+
+        Returns
+        -------
+        label : object
+
+        Notes
+        -----
+        Value of `side` parameter should be validated in caller.
+        """
+        if isinstance(label, str):
+            try:
+                parsed, reso = self._parse_with_reso(label)
+            except ValueError as err:
+                # DTI -> parsing.DateParseError
+                # TDI -> 'unit abbreviation w/o a number'
+                # PI -> string cannot be parsed as datetime-like
+                self._raise_invalid_indexer("slice", label, err)
+
+            lower, upper = self._parsed_string_to_bounds(reso, parsed)
+            return lower if side == "left" else upper
+        elif not isinstance(label, self._data._recognized_scalars):
+            self._raise_invalid_indexer("slice", label)
+
+        return label
+
+    # --------------------------------------------------------------------
+    # Arithmetic Methods
+
+    def shift(self, periods: int = 1, freq=None) -> Self:
+        """
+        Shift index by desired number of time frequency increments.
+
+        This method is for shifting the values of datetime-like indexes
+        by a specified time increment a given number of times.
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Number of periods (or increments) to shift by,
+            can be positive or negative.
+        freq : pandas.DateOffset, pandas.Timedelta or string, optional
+            Frequency increment to shift by.
+            If None, the index is shifted by its own `freq` attribute.
+            Offset aliases are valid strings, e.g., 'D', 'W', 'M' etc.
+
+        Returns
+        -------
+        pandas.DatetimeIndex
+            Shifted index.
+
+        See Also
+        --------
+        Index.shift : Shift values of Index.
+        PeriodIndex.shift : Shift values of PeriodIndex.
+        """
+        raise NotImplementedError
+
+    # --------------------------------------------------------------------
+
+    @doc(Index._maybe_cast_listlike_indexer)
+    def _maybe_cast_listlike_indexer(self, keyarr):
+        try:
+            res = self._data._validate_listlike(keyarr, allow_object=True)
+        except (ValueError, TypeError):
+            if not isinstance(keyarr, ExtensionArray):
+                # e.g. we don't want to cast DTA to ndarray[object]
+                res = com.asarray_tuplesafe(keyarr)
+                # TODO: com.asarray_tuplesafe shouldn't cast e.g. DatetimeArray
+            else:
+                res = keyarr
+        return Index(res, dtype=res.dtype)
+
+
+class DatetimeTimedeltaMixin(DatetimeIndexOpsMixin, ABC):
+    """
+    Mixin class for methods shared by DatetimeIndex and TimedeltaIndex,
+    but not PeriodIndex
+    """
+
+    _data: DatetimeArray | TimedeltaArray
+    _comparables = ["name", "freq"]
+    _attributes = ["name", "freq"]
+
+    # Compat for frequency inference, see GH#23789
+    _is_monotonic_increasing = Index.is_monotonic_increasing
+    _is_monotonic_decreasing = Index.is_monotonic_decreasing
+    _is_unique = Index.is_unique
+
+    @property
+    def unit(self) -> str:
+        return self._data.unit
+
+    def as_unit(self, unit: str) -> Self:
+        """
+        Convert to a dtype with the given unit resolution.
+
+        This method is for converting the dtype of a ``DatetimeIndex`` or
+        ``TimedeltaIndex`` to a new dtype with the given unit
+        resolution/precision.
+
+        Parameters
+        ----------
+        unit : {'s', 'ms', 'us', 'ns'}
+
+        Returns
+        -------
+        same type as self
+            Converted to the specified unit.
+
+        See Also
+        --------
+        Timestamp.as_unit : Convert to the given unit.
+        Timedelta.as_unit : Convert to the given unit.
+        DatetimeIndex.as_unit : Convert to the given unit.
+        TimedeltaIndex.as_unit : Convert to the given unit.
+
+        Examples
+        --------
+        For :class:`pandas.DatetimeIndex`:
+
+        >>> idx = pd.DatetimeIndex(["2020-01-02 01:02:03.004005006"])
+        >>> idx
+        DatetimeIndex(['2020-01-02 01:02:03.004005006'],
+                      dtype='datetime64[ns]', freq=None)
+        >>> idx.as_unit("s")
+        DatetimeIndex(['2020-01-02 01:02:03'], dtype='datetime64[s]', freq=None)
+
+        For :class:`pandas.TimedeltaIndex`:
+
+        >>> tdelta_idx = pd.to_timedelta(["1 day 3 min 2 us 42 ns"])
+        >>> tdelta_idx
+        TimedeltaIndex(['1 days 00:03:00.000002042'],
+                        dtype='timedelta64[ns]', freq=None)
+        >>> tdelta_idx.as_unit("s")
+        TimedeltaIndex(['1 days 00:03:00'], dtype='timedelta64[s]', freq=None)
+        """
+        arr = self._data.as_unit(unit)
+        return type(self)._simple_new(arr, name=self.name)
+
+    def _with_freq(self, freq):
+        arr = self._data._with_freq(freq)
+        return type(self)._simple_new(arr, name=self._name)
+
+    @property
+    def values(self) -> np.ndarray:
+        # NB: For Datetime64TZ this is lossy
+        data = self._data._ndarray
+        data = data.view()
+        data.flags.writeable = False
+        return data
+
+    @doc(DatetimeIndexOpsMixin.shift)
+    def shift(self, periods: int = 1, freq=None) -> Self:
+        if freq is not None and freq != self.freq:
+            if isinstance(freq, str):
+                freq = to_offset(freq)
+            offset = periods * freq
+            return self + offset
+
+        if periods == 0 or len(self) == 0:
+            # GH#14811 empty case
+            return self.copy()
+
+        if self.freq is None:
+            raise NullFrequencyError("Cannot shift with no freq")
+
+        start = self[0] + periods * self.freq
+        end = self[-1] + periods * self.freq
+
+        # Note: in the DatetimeTZ case, _generate_range will infer the
+        #  appropriate timezone from `start` and `end`, so tz does not need
+        #  to be passed explicitly.
+        result = self._data._generate_range(
+            start=start, end=end, periods=None, freq=self.freq, unit=self.unit
+        )
+        return type(self)._simple_new(result, name=self.name)
+
+    @cache_readonly
+    @doc(DatetimeLikeArrayMixin.inferred_freq)
+    def inferred_freq(self) -> str | None:
+        return self._data.inferred_freq
+
+    # --------------------------------------------------------------------
+    # Set Operation Methods
+
+    @cache_readonly
+    def _as_range_index(self) -> RangeIndex:
+        # Convert our i8 representations to RangeIndex
+        # Caller is responsible for checking isinstance(self.freq, Tick)
+        freq = cast(Tick, self.freq)
+        tick = Timedelta(freq).as_unit(self.unit)._value
+        rng = range(self[0]._value, self[-1]._value + tick, tick)
+        return RangeIndex(rng)
+
+    def _can_range_setop(self, other) -> bool:
+        return isinstance(self.freq, Tick) and isinstance(other.freq, Tick)
+
+    def _wrap_range_setop(self, other, res_i8) -> Self:
+        new_freq = None
+        if not len(res_i8):
+            # RangeIndex defaults to step=1, which we don't want.
+            new_freq = self.freq
+        elif isinstance(res_i8, RangeIndex):
+            new_freq = to_offset(
+                Timedelta(res_i8.step, unit=self.unit).as_unit(self.unit)
+            )
+
+        # TODO(GH#41493): we cannot just do
+        #  type(self._data)(res_i8.values, dtype=self.dtype, freq=new_freq)
+        # because test_setops_preserve_freq fails with _validate_frequency raising.
+        # This raising is incorrect, as 'on_freq' is incorrect. This will
+        # be fixed by GH#41493
+        res_values = res_i8.values.view(self._data._ndarray.dtype)
+        result = type(self._data)._simple_new(
+            # error: Argument "dtype" to "_simple_new" of "DatetimeArray" has
+            # incompatible type "Union[dtype[Any], ExtensionDtype]"; expected
+            # "Union[dtype[datetime64], DatetimeTZDtype]"
+            res_values,
+            dtype=self.dtype,  # type: ignore[arg-type]
+            freq=new_freq,  # type: ignore[arg-type]
+        )
+        return cast("Self", self._wrap_setop_result(other, result))
+
+    def _range_intersect(self, other, sort) -> Self:
+        # Dispatch to RangeIndex intersection logic.
+        left = self._as_range_index
+        right = other._as_range_index
+        res_i8 = left.intersection(right, sort=sort)
+        return self._wrap_range_setop(other, res_i8)
+
+    def _range_union(self, other, sort) -> Self:
+        # Dispatch to RangeIndex union logic.
+        left = self._as_range_index
+        right = other._as_range_index
+        res_i8 = left.union(right, sort=sort)
+        return self._wrap_range_setop(other, res_i8)
+
+    def _intersection(self, other: Index, sort: bool = False) -> Index:
+        """
+        intersection specialized to the case with matching dtypes and both non-empty.
+        """
+        other = cast("DatetimeTimedeltaMixin", other)
+
+        if self._can_range_setop(other):
+            return self._range_intersect(other, sort=sort)
+
+        if not self._can_fast_intersect(other):
+            result = Index._intersection(self, other, sort=sort)
+            # We need to invalidate the freq because Index._intersection
+            #  uses _shallow_copy on a view of self._data, which will preserve
+            #  self.freq if we're not careful.
+            # At this point we should have result.dtype == self.dtype
+            #  and type(result) is type(self._data)
+            result = self._wrap_setop_result(other, result)
+            return result._with_freq(None)._with_freq("infer")
+
+        else:
+            return self._fast_intersect(other, sort)
+
+    def _fast_intersect(self, other, sort):
+        # to make our life easier, "sort" the two ranges
+        if self[0] <= other[0]:
+            left, right = self, other
+        else:
+            left, right = other, self
+
+        # after sorting, the intersection always starts with the right index
+        # and ends with the index of which the last elements is smallest
+        end = min(left[-1], right[-1])
+        start = right[0]
+
+        if end < start:
+            result = self[:0]
+        else:
+            lslice = slice(*left.slice_locs(start, end))
+            result = left._values[lslice]
+
+        return result
+
+    def _can_fast_intersect(self, other: Self) -> bool:
+        # Note: we only get here with len(self) > 0 and len(other) > 0
+        if self.freq is None:
+            return False
+
+        elif other.freq != self.freq:
+            return False
+
+        elif not self.is_monotonic_increasing:
+            # Because freq is not None, we must then be monotonic decreasing
+            return False
+
+        # this along with matching freqs ensure that we "line up",
+        #  so intersection will preserve freq
+        # Note we are assuming away Ticks, as those go through _range_intersect
+        # GH#42104
+        return self.freq.n == 1
+
+    def _can_fast_union(self, other: Self) -> bool:
+        # Assumes that type(self) == type(other), as per the annotation
+        # The ability to fast_union also implies that `freq` should be
+        #  retained on union.
+        freq = self.freq
+
+        if freq is None or freq != other.freq:
+            return False
+
+        if not self.is_monotonic_increasing:
+            # Because freq is not None, we must then be monotonic decreasing
+            # TODO: do union on the reversed indexes?
+            return False
+
+        if len(self) == 0 or len(other) == 0:
+            # only reached via union_many
+            return True
+
+        # to make our life easier, "sort" the two ranges
+        if self[0] <= other[0]:
+            left, right = self, other
+        else:
+            left, right = other, self
+
+        right_start = right[0]
+        left_end = left[-1]
+
+        # Only need to "adjoin", not overlap
+        return (right_start == left_end + freq) or right_start in left
+
+    def _fast_union(self, other: Self, sort=None) -> Self:
+        # Caller is responsible for ensuring self and other are non-empty
+
+        # to make our life easier, "sort" the two ranges
+        if self[0] <= other[0]:
+            left, right = self, other
+        elif sort is False:
+            # TDIs are not in the "correct" order and we don't want
+            #  to sort but want to remove overlaps
+            left, right = self, other
+            left_start = left[0]
+            loc = right.searchsorted(left_start, side="left")
+            right_chunk = right._values[:loc]
+            dates = concat_compat((left._values, right_chunk))
+            result = type(self)._simple_new(dates, name=self.name)
+            return result
+        else:
+            left, right = other, self
+
+        left_end = left[-1]
+        right_end = right[-1]
+
+        # concatenate
+        if left_end < right_end:
+            loc = right.searchsorted(left_end, side="right")
+            right_chunk = right._values[loc:]
+            dates = concat_compat([left._values, right_chunk])
+            # The can_fast_union check ensures that the result.freq
+            #  should match self.freq
+            assert isinstance(dates, type(self._data))
+            # error: Item "ExtensionArray" of "ExtensionArray |
+            # ndarray[Any, Any]" has no attribute "_freq"
+            assert dates._freq == self.freq  # type: ignore[union-attr]
+            result = type(self)._simple_new(dates)
+            return result
+        else:
+            return left
+
+    def _union(self, other, sort):
+        # We are called by `union`, which is responsible for this validation
+        assert isinstance(other, type(self))
+        assert self.dtype == other.dtype
+
+        if self._can_range_setop(other):
+            return self._range_union(other, sort=sort)
+
+        if self._can_fast_union(other):
+            result = self._fast_union(other, sort=sort)
+            # in the case with sort=None, the _can_fast_union check ensures
+            #  that result.freq == self.freq
+            return result
+        else:
+            return super()._union(other, sort)._with_freq("infer")
+
+    # --------------------------------------------------------------------
+    # Join Methods
+
+    def _get_join_freq(self, other):
+        """
+        Get the freq to attach to the result of a join operation.
+        """
+        freq = None
+        if self._can_fast_union(other):
+            freq = self.freq
+        return freq
+
+    def _wrap_join_result(
+        self,
+        joined,
+        other,
+        lidx: npt.NDArray[np.intp] | None,
+        ridx: npt.NDArray[np.intp] | None,
+        how: JoinHow,
+    ) -> tuple[Self, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        assert other.dtype == self.dtype, (other.dtype, self.dtype)
+        join_index, lidx, ridx = super()._wrap_join_result(
+            joined, other, lidx, ridx, how
+        )
+        join_index._data._freq = self._get_join_freq(other)
+        return join_index, lidx, ridx
+
+    def _get_engine_target(self) -> np.ndarray:
+        # engine methods and libjoin methods need dt64/td64 values cast to i8
+        return self._data._ndarray.view("i8")
+
+    def _from_join_target(self, result: np.ndarray):
+        # view e.g. i8 back to M8[ns]
+        result = result.view(self._data._ndarray.dtype)
+        return self._data._from_backing_data(result)
+
+    # --------------------------------------------------------------------
+    # List-like Methods
+
+    def _get_delete_freq(self, loc: int | slice | Sequence[int]):
+        """
+        Find the `freq` for self.delete(loc).
+        """
+        freq = None
+        if self.freq is not None:
+            if is_integer(loc):
+                if loc in (0, -len(self), -1, len(self) - 1):
+                    freq = self.freq
+            else:
+                if is_list_like(loc):
+                    # error: Incompatible types in assignment (expression has
+                    # type "Union[slice, ndarray]", variable has type
+                    # "Union[int, slice, Sequence[int]]")
+                    loc = lib.maybe_indices_to_slice(  # type: ignore[assignment]
+                        np.asarray(loc, dtype=np.intp), len(self)
+                    )
+                if isinstance(loc, slice) and loc.step in (1, None):
+                    if loc.start in (0, None) or loc.stop in (len(self), None):
+                        freq = self.freq
+        return freq
+
+    def _get_insert_freq(self, loc: int, item):
+        """
+        Find the `freq` for self.insert(loc, item).
+        """
+        value = self._data._validate_scalar(item)
+        item = self._data._box_func(value)
+
+        freq = None
+        if self.freq is not None:
+            # freq can be preserved on edge cases
+            if self.size:
+                if item is NaT:
+                    pass
+                elif loc in (0, -len(self)) and item + self.freq == self[0]:
+                    freq = self.freq
+                elif (loc == len(self)) and item - self.freq == self[-1]:
+                    freq = self.freq
+            else:
+                # Adding a single item to an empty index may preserve freq
+                if isinstance(self.freq, Tick):
+                    # all TimedeltaIndex cases go through here; is_on_offset
+                    #  would raise TypeError
+                    freq = self.freq
+                elif self.freq.is_on_offset(item):
+                    freq = self.freq
+        return freq
+
+    @doc(NDArrayBackedExtensionIndex.delete)
+    def delete(self, loc) -> Self:
+        result = super().delete(loc)
+        result._data._freq = self._get_delete_freq(loc)
+        return result
+
+    @doc(NDArrayBackedExtensionIndex.insert)
+    def insert(self, loc: int, item):
+        result = super().insert(loc, item)
+        if isinstance(result, type(self)):
+            # i.e. parent class method did not cast
+            result._data._freq = self._get_insert_freq(loc, item)
+        return result
+
+    # --------------------------------------------------------------------
+    # NDArray-Like Methods
+
+    @Appender(_index_shared_docs["take"] % _index_doc_kwargs)
+    def take(
+        self,
+        indices,
+        axis: Axis = 0,
+        allow_fill: bool = True,
+        fill_value=None,
+        **kwargs,
+    ) -> Self:
+        nv.validate_take((), kwargs)
+        indices = np.asarray(indices, dtype=np.intp)
+
+        result = NDArrayBackedExtensionIndex.take(
+            self, indices, axis, allow_fill, fill_value, **kwargs
+        )
+
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(self))
+        if isinstance(maybe_slice, slice):
+            freq = self._data._get_getitem_freq(maybe_slice)
+            result._data._freq = freq
+        return result
diff --git a/pandas/pandas/core/indexes/datetimes.py b/pandas/pandas/core/indexes/datetimes.py
new file mode 100644
index 0000000000000..9adbaadbdcdc8
--- /dev/null
+++ b/pandas/pandas/core/indexes/datetimes.py
@@ -0,0 +1,1137 @@
+from __future__ import annotations
+
+import datetime as dt
+import operator
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    Period,
+    Timestamp,
+    index as libindex,
+    lib,
+)
+from pandas._libs.tslibs import (
+    Resolution,
+    Tick,
+    Timedelta,
+    periods_per_day,
+    timezones,
+    to_offset,
+)
+from pandas._libs.tslibs.offsets import prefix_mapping
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+    set_module,
+)
+
+from pandas.core.dtypes.common import is_scalar
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.generic import ABCSeries
+from pandas.core.dtypes.missing import is_valid_na_for_dtype
+
+from pandas.core.arrays.datetimes import (
+    DatetimeArray,
+    tz_to_dtype,
+)
+import pandas.core.common as com
+from pandas.core.indexes.base import (
+    Index,
+    maybe_extract_name,
+)
+from pandas.core.indexes.datetimelike import DatetimeTimedeltaMixin
+from pandas.core.indexes.extension import inherit_names
+from pandas.core.tools.times import to_time
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import (
+        Dtype,
+        DtypeObj,
+        Frequency,
+        IntervalClosedType,
+        Self,
+        TimeAmbiguous,
+        TimeNonexistent,
+        npt,
+    )
+
+    from pandas.core.api import (
+        DataFrame,
+        PeriodIndex,
+    )
+
+from pandas._libs.tslibs.dtypes import OFFSET_TO_PERIOD_FREQSTR
+
+
+def _new_DatetimeIndex(cls, d):
+    """
+    This is called upon unpickling, rather than the default which doesn't
+    have arguments and breaks __new__
+    """
+    if "data" in d and not isinstance(d["data"], DatetimeIndex):
+        # Avoid need to verify integrity by calling simple_new directly
+        data = d.pop("data")
+        if not isinstance(data, DatetimeArray):
+            # For backward compat with older pickles, we may need to construct
+            #  a DatetimeArray to adapt to the newer _simple_new signature
+            tz = d.pop("tz")
+            freq = d.pop("freq")
+            dta = DatetimeArray._simple_new(data, dtype=tz_to_dtype(tz), freq=freq)
+        else:
+            dta = data
+            for key in ["tz", "freq"]:
+                # These are already stored in our DatetimeArray; if they are
+                #  also in the pickle and don't match, we have a problem.
+                if key in d:
+                    assert d[key] == getattr(dta, key)
+                    d.pop(key)
+        result = cls._simple_new(dta, **d)
+    else:
+        with warnings.catch_warnings():
+            # TODO: If we knew what was going in to **d, we might be able to
+            #  go through _simple_new instead
+            warnings.simplefilter("ignore")
+            result = cls.__new__(cls, **d)
+
+    return result
+
+
+@inherit_names(
+    DatetimeArray._field_ops
+    + [
+        method
+        for method in DatetimeArray._datetimelike_methods
+        if method not in ("tz_localize", "tz_convert", "strftime")
+    ],
+    DatetimeArray,
+    wrap=True,
+)
+@inherit_names(["is_normalized"], DatetimeArray, cache=True)
+@inherit_names(
+    [
+        "tz",
+        "tzinfo",
+        "dtype",
+        "to_pydatetime",
+        "date",
+        "time",
+        "timetz",
+        "std",
+    ]
+    + DatetimeArray._bool_ops,
+    DatetimeArray,
+)
+@set_module("pandas")
+class DatetimeIndex(DatetimeTimedeltaMixin):
+    """
+    Immutable ndarray-like of datetime64 data.
+
+    Represented internally as int64, and which can be boxed to Timestamp objects
+    that are subclasses of datetime and carry metadata.
+
+    .. versionchanged:: 2.0.0
+        The various numeric date/time attributes (:attr:`~DatetimeIndex.day`,
+        :attr:`~DatetimeIndex.month`, :attr:`~DatetimeIndex.year` etc.) now have dtype
+        ``int32``. Previously they had dtype ``int64``.
+
+    Parameters
+    ----------
+    data : array-like (1-dimensional)
+        Datetime-like data to construct index with.
+    freq : str or pandas offset object, optional
+        One of pandas date offset strings or corresponding objects. The string
+        'infer' can be passed in order to set the frequency of the index as the
+        inferred frequency upon creation.
+    tz : zoneinfo.ZoneInfo, pytz.timezone, dateutil.tz.tzfile, datetime.tzinfo or str
+        Set the Timezone of the data.
+    ambiguous : 'infer', bool-ndarray, 'NaT', default 'raise'
+        When clocks moved backward due to DST, ambiguous times may arise.
+        For example in Central European Time (UTC+01), when going from 03:00
+        DST to 02:00 non-DST, 02:30:00 local time occurs both at 00:30:00 UTC
+        and at 01:30:00 UTC. In such a situation, the `ambiguous` parameter
+        dictates how ambiguous times should be handled.
+
+        - 'infer' will attempt to infer fall dst-transition hours based on
+          order
+        - bool-ndarray where True signifies a DST time, False signifies a
+          non-DST time (note that this flag is only applicable for ambiguous
+          times)
+        - 'NaT' will return NaT where there are ambiguous times
+        - 'raise' will raise a ValueError if there are ambiguous times.
+    dayfirst : bool, default False
+        If True, parse dates in `data` with the day first order.
+    yearfirst : bool, default False
+        If True parse dates in `data` with the year first order.
+    dtype : numpy.dtype or DatetimeTZDtype or str, default None
+        Note that the only NumPy dtype allowed is `datetime64[ns]`.
+    copy : bool, default False
+        Make a copy of input ndarray.
+    name : label, default None
+        Name to be stored in the index.
+
+    Attributes
+    ----------
+    year
+    month
+    day
+    hour
+    minute
+    second
+    microsecond
+    nanosecond
+    date
+    time
+    timetz
+    dayofyear
+    day_of_year
+    dayofweek
+    day_of_week
+    weekday
+    quarter
+    tz
+    freq
+    freqstr
+    is_month_start
+    is_month_end
+    is_quarter_start
+    is_quarter_end
+    is_year_start
+    is_year_end
+    is_leap_year
+    inferred_freq
+
+    Methods
+    -------
+    normalize
+    strftime
+    snap
+    tz_convert
+    tz_localize
+    round
+    floor
+    ceil
+    to_period
+    to_pydatetime
+    to_series
+    to_frame
+    month_name
+    day_name
+    mean
+    std
+
+    See Also
+    --------
+    Index : The base pandas Index type.
+    TimedeltaIndex : Index of timedelta64 data.
+    PeriodIndex : Index of Period data.
+    to_datetime : Convert argument to datetime.
+    date_range : Create a fixed-frequency DatetimeIndex.
+
+    Notes
+    -----
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    >>> idx = pd.DatetimeIndex(["1/1/2020 10:00:00+00:00", "2/1/2020 11:00:00+00:00"])
+    >>> idx
+    DatetimeIndex(['2020-01-01 10:00:00+00:00', '2020-02-01 11:00:00+00:00'],
+    dtype='datetime64[s, UTC]', freq=None)
+    """
+
+    _typ = "datetimeindex"
+
+    _data_cls = DatetimeArray
+    _supports_partial_string_indexing = True
+
+    @property
+    def _engine_type(self) -> type[libindex.DatetimeEngine]:
+        return libindex.DatetimeEngine
+
+    _data: DatetimeArray
+    _values: DatetimeArray
+    tz: dt.tzinfo | None
+
+    # --------------------------------------------------------------------
+    # methods that dispatch to DatetimeArray and wrap result
+
+    @doc(DatetimeArray.strftime)
+    def strftime(self, date_format) -> Index:
+        arr = self._data.strftime(date_format)
+        return Index(arr, name=self.name, dtype=arr.dtype)
+
+    @doc(DatetimeArray.tz_convert)
+    def tz_convert(self, tz) -> Self:
+        arr = self._data.tz_convert(tz)
+        return type(self)._simple_new(arr, name=self.name, refs=self._references)
+
+    @doc(DatetimeArray.tz_localize)
+    def tz_localize(
+        self,
+        tz,
+        ambiguous: TimeAmbiguous = "raise",
+        nonexistent: TimeNonexistent = "raise",
+    ) -> Self:
+        arr = self._data.tz_localize(tz, ambiguous, nonexistent)
+        return type(self)._simple_new(arr, name=self.name)
+
+    @doc(DatetimeArray.to_period)
+    def to_period(self, freq=None) -> PeriodIndex:
+        from pandas.core.indexes.api import PeriodIndex
+
+        arr = self._data.to_period(freq)
+        return PeriodIndex._simple_new(arr, name=self.name)
+
+    @doc(DatetimeArray.to_julian_date)
+    def to_julian_date(self) -> Index:
+        arr = self._data.to_julian_date()
+        return Index._simple_new(arr, name=self.name)
+
+    @doc(DatetimeArray.isocalendar)
+    def isocalendar(self) -> DataFrame:
+        df = self._data.isocalendar()
+        return df.set_index(self)
+
+    @cache_readonly
+    def _resolution_obj(self) -> Resolution:
+        return self._data._resolution_obj
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data=None,
+        freq: Frequency | lib.NoDefault = lib.no_default,
+        tz=lib.no_default,
+        ambiguous: TimeAmbiguous = "raise",
+        dayfirst: bool = False,
+        yearfirst: bool = False,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        name: Hashable | None = None,
+    ) -> Self:
+        if is_scalar(data):
+            cls._raise_scalar_data_error(data)
+
+        # - Cases checked above all return/raise before reaching here - #
+
+        name = maybe_extract_name(name, data, cls)
+
+        if (
+            isinstance(data, DatetimeArray)
+            and freq is lib.no_default
+            and tz is lib.no_default
+            and dtype is None
+        ):
+            # fastpath, similar logic in TimedeltaIndex.__new__;
+            # Note in this particular case we retain non-nano.
+            if copy:
+                data = data.copy()
+            return cls._simple_new(data, name=name)
+
+        dtarr = DatetimeArray._from_sequence_not_strict(
+            data,
+            dtype=dtype,
+            copy=copy,
+            tz=tz,
+            freq=freq,
+            dayfirst=dayfirst,
+            yearfirst=yearfirst,
+            ambiguous=ambiguous,
+        )
+        refs = None
+        if not copy and isinstance(data, (Index, ABCSeries)):
+            refs = data._references
+
+        subarr = cls._simple_new(dtarr, name=name, refs=refs)
+        return subarr
+
+    # --------------------------------------------------------------------
+
+    @cache_readonly
+    def _is_dates_only(self) -> bool:
+        """
+        Return a boolean if we are only dates (and don't have a timezone)
+
+        Returns
+        -------
+        bool
+        """
+        if isinstance(self.freq, Tick):
+            delta = Timedelta(self.freq)
+
+            if delta % dt.timedelta(days=1) != dt.timedelta(days=0):
+                return False
+
+        return self._values._is_dates_only
+
+    def __reduce__(self):
+        d = {"data": self._data, "name": self.name}
+        return _new_DatetimeIndex, (type(self), d), None
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        """
+        Can we compare values of the given dtype to our own?
+        """
+        if self.tz is not None:
+            # If we have tz, we can compare to tzaware
+            return isinstance(dtype, DatetimeTZDtype)
+        # if we dont have tz, we can only compare to tznaive
+        return lib.is_np_dtype(dtype, "M")
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+
+    @cache_readonly
+    def _formatter_func(self):
+        # Note this is equivalent to the DatetimeIndexOpsMixin method but
+        #  uses the maybe-cached self._is_dates_only instead of re-computing it.
+        from pandas.io.formats.format import get_format_datetime64
+
+        formatter = get_format_datetime64(is_dates_only=self._is_dates_only)
+        return lambda x: f"'{formatter(x)}'"
+
+    # --------------------------------------------------------------------
+    # Set Operation Methods
+
+    def _can_range_setop(self, other) -> bool:
+        # GH 46702: If self or other have non-UTC tzs, DST transitions prevent
+        # range representation due to no singular step
+        if (
+            self.tz is not None
+            and not timezones.is_utc(self.tz)
+            and not timezones.is_fixed_offset(self.tz)
+        ):
+            return False
+        if (
+            other.tz is not None
+            and not timezones.is_utc(other.tz)
+            and not timezones.is_fixed_offset(other.tz)
+        ):
+            return False
+        return super()._can_range_setop(other)
+
+    # --------------------------------------------------------------------
+
+    def _get_time_micros(self) -> npt.NDArray[np.int64]:
+        """
+        Return the number of microseconds since midnight.
+
+        Returns
+        -------
+        ndarray[int64_t]
+        """
+        values = self._data._local_timestamps()
+
+        ppd = periods_per_day(self._data._creso)
+
+        frac = values % ppd
+        if self.unit == "ns":
+            micros = frac // 1000
+        elif self.unit == "us":
+            micros = frac
+        elif self.unit == "ms":
+            micros = frac * 1000
+        elif self.unit == "s":
+            micros = frac * 1_000_000
+        else:  # pragma: no cover
+            raise NotImplementedError(self.unit)
+
+        micros[self._isnan] = -1
+        return micros
+
+    def snap(self, freq: Frequency = "S") -> DatetimeIndex:
+        """
+        Snap time stamps to nearest occurring frequency.
+
+        Parameters
+        ----------
+        freq : str, Timedelta, datetime.timedelta, or DateOffset, default 'S'
+            Frequency strings can have multiples, e.g. '5h'. See
+            :ref:`here <timeseries.offset_aliases>` for a list of
+            frequency aliases.
+
+        Returns
+        -------
+        DatetimeIndex
+            Time stamps to nearest occurring `freq`.
+
+        See Also
+        --------
+        DatetimeIndex.round : Perform round operation on the data to the
+            specified `freq`.
+        DatetimeIndex.floor : Perform floor operation on the data to the
+            specified `freq`.
+
+        Examples
+        --------
+        >>> idx = pd.DatetimeIndex(
+        ...     ["2023-01-01", "2023-01-02", "2023-02-01", "2023-02-02"],
+        ...     dtype="M8[ns]",
+        ... )
+        >>> idx
+        DatetimeIndex(['2023-01-01', '2023-01-02', '2023-02-01', '2023-02-02'],
+        dtype='datetime64[ns]', freq=None)
+        >>> idx.snap("MS")
+        DatetimeIndex(['2023-01-01', '2023-01-01', '2023-02-01', '2023-02-01'],
+        dtype='datetime64[ns]', freq=None)
+        """
+        # Superdumb, punting on any optimizing
+        freq = to_offset(freq)
+
+        dta = self._data.copy()
+
+        for i, v in enumerate(self):
+            s = v
+            if not freq.is_on_offset(s):
+                t0 = freq.rollback(s)
+                t1 = freq.rollforward(s)
+                if abs(s - t0) < abs(t1 - s):
+                    s = t0
+                else:
+                    s = t1
+            dta[i] = s
+
+        return DatetimeIndex._simple_new(dta, name=self.name)
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    def _parsed_string_to_bounds(
+        self, reso: Resolution, parsed: dt.datetime
+    ) -> tuple[Timestamp, Timestamp]:
+        """
+        Calculate datetime bounds for parsed time string and its resolution.
+
+        Parameters
+        ----------
+        reso : Resolution
+            Resolution provided by parsed string.
+        parsed : datetime
+            Datetime from parsed string.
+
+        Returns
+        -------
+        lower, upper: pd.Timestamp
+        """
+        freq = OFFSET_TO_PERIOD_FREQSTR.get(reso.attr_abbrev, reso.attr_abbrev)
+        per = Period(parsed, freq=freq)
+        start, end = per.start_time, per.end_time
+        start = start.as_unit(self.unit)
+        end = end.as_unit(self.unit)
+
+        # GH 24076
+        # If an incoming date string contained a UTC offset, need to localize
+        # the parsed date to this offset first before aligning with the index's
+        # timezone
+        start = start.tz_localize(parsed.tzinfo)
+        end = end.tz_localize(parsed.tzinfo)
+
+        if parsed.tzinfo is not None:
+            if self.tz is None:
+                raise ValueError(
+                    "The index must be timezone aware when indexing "
+                    "with a date string with a UTC offset"
+                )
+        # The flipped case with parsed.tz is None and self.tz is not None
+        #  is ruled out bc parsed and reso are produced by _parse_with_reso,
+        #  which localizes parsed.
+        return start, end
+
+    def _parse_with_reso(self, label: str) -> tuple[Timestamp, Resolution]:
+        parsed, reso = super()._parse_with_reso(label)
+
+        parsed = Timestamp(parsed)
+
+        if self.tz is not None and parsed.tzinfo is None:
+            # we special-case timezone-naive strings and timezone-aware
+            #  DatetimeIndex
+            # https://github.com/pandas-dev/pandas/pull/36148#issuecomment-687883081
+            parsed = parsed.tz_localize(self.tz)
+
+        return parsed, reso
+
+    def _disallow_mismatched_indexing(self, key) -> None:
+        """
+        Check for mismatched-tzawareness indexing and re-raise as KeyError.
+        """
+        # we get here with isinstance(key, self._data._recognized_scalars)
+        try:
+            # GH#36148
+            self._data._assert_tzawareness_compat(key)
+        except TypeError as err:
+            raise KeyError(key) from err
+
+    def get_loc(self, key):
+        """
+        Get integer location for requested label
+
+        Returns
+        -------
+        loc : int
+        """
+        self._check_indexing_error(key)
+
+        orig_key = key
+        if is_valid_na_for_dtype(key, self.dtype):
+            key = NaT
+
+        if isinstance(key, self._data._recognized_scalars):
+            # needed to localize naive datetimes
+            self._disallow_mismatched_indexing(key)
+            key = Timestamp(key)
+
+        elif isinstance(key, str):
+            try:
+                parsed, reso = self._parse_with_reso(key)
+            except ValueError as err:
+                raise KeyError(key) from err
+            self._disallow_mismatched_indexing(parsed)
+
+            if self._can_partial_date_slice(reso):
+                try:
+                    return self._partial_date_slice(reso, parsed)
+                except KeyError as err:
+                    raise KeyError(key) from err
+
+            key = parsed
+
+        elif isinstance(key, dt.timedelta):
+            # GH#20464
+            raise TypeError(
+                f"Cannot index {type(self).__name__} with {type(key).__name__}"
+            )
+
+        elif isinstance(key, dt.time):
+            return self.indexer_at_time(key)
+
+        else:
+            # unrecognized type
+            raise KeyError(key)
+
+        try:
+            return Index.get_loc(self, key)
+        except KeyError as err:
+            raise KeyError(orig_key) from err
+
+    @doc(DatetimeTimedeltaMixin._maybe_cast_slice_bound)
+    def _maybe_cast_slice_bound(self, label, side: str):
+        # GH#42855 handle date here instead of get_slice_bound
+        if isinstance(label, dt.date) and not isinstance(label, dt.datetime):
+            # Pandas supports slicing with dates, treated as datetimes at midnight.
+            # https://github.com/pandas-dev/pandas/issues/31501
+            label = Timestamp(label).to_pydatetime()
+
+        label = super()._maybe_cast_slice_bound(label, side)
+        self._data._assert_tzawareness_compat(label)
+        return Timestamp(label)
+
+    def slice_indexer(self, start=None, end=None, step=None):
+        """
+        Return indexer for specified label slice.
+        Index.slice_indexer, customized to handle time slicing.
+
+        In addition to functionality provided by Index.slice_indexer, does the
+        following:
+
+        - if both `start` and `end` are instances of `datetime.time`, it
+          invokes `indexer_between_time`
+        - if `start` and `end` are both either string or None perform
+          value-based selection in non-monotonic cases.
+
+        """
+        # For historical reasons DatetimeIndex supports slices between two
+        # instances of datetime.time as if it were applying a slice mask to
+        # an array of (self.hour, self.minute, self.seconds, self.microsecond).
+        if isinstance(start, dt.time) and isinstance(end, dt.time):
+            if step is not None and step != 1:
+                raise ValueError("Must have step size of 1 with time slices")
+            return self.indexer_between_time(start, end)
+
+        if isinstance(start, dt.time) or isinstance(end, dt.time):
+            raise KeyError("Cannot mix time and non-time slice keys")
+
+        def check_str_or_none(point) -> bool:
+            return point is not None and not isinstance(point, str)
+
+        # GH#33146 if start and end are combinations of str and None and Index is not
+        # monotonic, we can not use Index.slice_indexer because it does not honor the
+        # actual elements, is only searching for start and end
+        if (
+            check_str_or_none(start)
+            or check_str_or_none(end)
+            or self.is_monotonic_increasing
+        ):
+            return Index.slice_indexer(self, start, end, step)
+
+        mask = np.array(True)
+        in_index = True
+        if start is not None:
+            start_casted = self._maybe_cast_slice_bound(start, "left")
+            mask = start_casted <= self
+            in_index &= (start_casted == self).any()
+
+        if end is not None:
+            end_casted = self._maybe_cast_slice_bound(end, "right")
+            mask = (self <= end_casted) & mask
+            in_index &= (end_casted == self).any()
+
+        if not in_index:
+            raise KeyError(
+                "Value based partial slicing on non-monotonic DatetimeIndexes "
+                "with non-existing keys is not allowed.",
+            )
+        indexer = mask.nonzero()[0][::step]
+        if len(indexer) == len(self):
+            return slice(None)
+        else:
+            return indexer
+
+    # --------------------------------------------------------------------
+
+    @property
+    def inferred_type(self) -> str:
+        # b/c datetime is represented as microseconds since the epoch, make
+        # sure we can't have ambiguous indexing
+        return "datetime64"
+
+    def indexer_at_time(self, time, asof: bool = False) -> npt.NDArray[np.intp]:
+        """
+        Return index locations of values at particular time of day.
+
+        Parameters
+        ----------
+        time : datetime.time or str
+            Time passed in either as object (datetime.time) or as string in
+            appropriate format ("%H:%M", "%H%M", "%I:%M%p", "%I%M%p",
+            "%H:%M:%S", "%H%M%S", "%I:%M:%S%p", "%I%M%S%p").
+        asof : bool, default False
+            This parameter is currently not supported.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            Index locations of values at given `time` of day.
+
+        See Also
+        --------
+        indexer_between_time : Get index locations of values between particular
+            times of day.
+        DataFrame.at_time : Select values at particular time of day.
+
+        Examples
+        --------
+        >>> idx = pd.DatetimeIndex(
+        ...     ["1/1/2020 10:00", "2/1/2020 11:00", "3/1/2020 10:00"]
+        ... )
+        >>> idx.indexer_at_time("10:00")
+        array([0, 2])
+        """
+        if asof:
+            raise NotImplementedError("'asof' argument is not supported")
+
+        if isinstance(time, str):
+            from dateutil.parser import parse
+
+            time = parse(time).time()
+
+        if time.tzinfo:
+            if self.tz is None:
+                raise ValueError("Index must be timezone aware.")
+            time_micros = self.tz_convert(time.tzinfo)._get_time_micros()
+        else:
+            time_micros = self._get_time_micros()
+        micros = _time_to_micros(time)
+        return (time_micros == micros).nonzero()[0]
+
+    def indexer_between_time(
+        self, start_time, end_time, include_start: bool = True, include_end: bool = True
+    ) -> npt.NDArray[np.intp]:
+        """
+        Return index locations of values between particular times of day.
+
+        Parameters
+        ----------
+        start_time, end_time : datetime.time, str
+            Time passed either as object (datetime.time) or as string in
+            appropriate format ("%H:%M", "%H%M", "%I:%M%p", "%I%M%p",
+            "%H:%M:%S", "%H%M%S", "%I:%M:%S%p","%I%M%S%p").
+        include_start : bool, default True
+            Include boundaries; whether to set start bound as closed or open.
+        include_end : bool, default True
+            Include boundaries; whether to set end bound as closed or open.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+            Index locations of values between particular times of day.
+
+        See Also
+        --------
+        indexer_at_time : Get index locations of values at particular time of day.
+        DataFrame.between_time : Select values between particular times of day.
+
+        Examples
+        --------
+        >>> idx = pd.date_range("2023-01-01", periods=4, freq="h")
+        >>> idx
+        DatetimeIndex(['2023-01-01 00:00:00', '2023-01-01 01:00:00',
+                           '2023-01-01 02:00:00', '2023-01-01 03:00:00'],
+                          dtype='datetime64[ns]', freq='h')
+        >>> idx.indexer_between_time("00:00", "2:00", include_end=False)
+        array([0, 1])
+        """
+        start_time = to_time(start_time)
+        end_time = to_time(end_time)
+        time_micros = self._get_time_micros()
+        start_micros = _time_to_micros(start_time)
+        end_micros = _time_to_micros(end_time)
+
+        if include_start and include_end:
+            lop = rop = operator.le
+        elif include_start:
+            lop = operator.le
+            rop = operator.lt
+        elif include_end:
+            lop = operator.lt
+            rop = operator.le
+        else:
+            lop = rop = operator.lt
+
+        if start_time <= end_time:
+            join_op = operator.and_
+        else:
+            join_op = operator.or_
+
+        mask = join_op(lop(start_micros, time_micros), rop(time_micros, end_micros))
+
+        return mask.nonzero()[0]
+
+
+@set_module("pandas")
+def date_range(
+    start=None,
+    end=None,
+    periods=None,
+    freq=None,
+    tz=None,
+    normalize: bool = False,
+    name: Hashable | None = None,
+    inclusive: IntervalClosedType = "both",
+    *,
+    unit: str | None = None,
+    **kwargs,
+) -> DatetimeIndex:
+    """
+    Return a fixed frequency DatetimeIndex.
+
+    Returns the range of equally spaced time points (where the difference between any
+    two adjacent points is specified by the given frequency) such that they fall in the
+    range `[start, end]` , where the first one and the last one are, resp., the first
+    and last time points in that range that fall on the boundary of ``freq`` (if given
+    as a frequency string) or that are valid for ``freq`` (if given as a
+    :class:`pandas.tseries.offsets.DateOffset`). If ``freq`` is positive, the points
+    satisfy `start <[=] x <[=] end`, and if ``freq`` is negative, the points satisfy
+    `end <[=] x <[=] start`. (If exactly one of ``start``, ``end``, or ``freq`` is *not*
+    specified, this missing parameter can be computed given ``periods``, the number of
+    timesteps in the range. See the note below.)
+
+    Parameters
+    ----------
+    start : str or datetime-like, optional
+        Left bound for generating dates.
+    end : str or datetime-like, optional
+        Right bound for generating dates.
+    periods : int, optional
+        Number of periods to generate.
+    freq : str, Timedelta, datetime.timedelta, or DateOffset, default 'D'
+        Frequency strings can have multiples, e.g. '5h'. See
+        :ref:`here <timeseries.offset_aliases>` for a list of
+        frequency aliases.
+    tz : str or tzinfo, optional
+        Time zone name for returning localized DatetimeIndex, for example
+        'Asia/Hong_Kong'. By default, the resulting DatetimeIndex is
+        timezone-naive unless timezone-aware datetime-likes are passed.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    name : str, default None
+        Name of the resulting DatetimeIndex.
+    inclusive : {"both", "neither", "left", "right"}, default "both"
+        Include boundaries; Whether to set each bound as closed or open.
+
+        .. versionadded:: 1.4.0
+    unit : str, default None
+        Specify the desired resolution of the result.
+
+        .. versionadded:: 2.0.0
+    **kwargs
+        For compatibility. Has no effect on the result.
+
+    Returns
+    -------
+    DatetimeIndex
+        A DatetimeIndex object of the generated dates.
+
+    See Also
+    --------
+    DatetimeIndex : An immutable container for datetimes.
+    timedelta_range : Return a fixed frequency TimedeltaIndex.
+    period_range : Return a fixed frequency PeriodIndex.
+    interval_range : Return a fixed frequency IntervalIndex.
+
+    Notes
+    -----
+    Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
+    exactly three must be specified. If ``freq`` is omitted, the resulting
+    ``DatetimeIndex`` will have ``periods`` linearly spaced elements between
+    ``start`` and ``end`` (closed on both sides).
+
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    **Specifying the values**
+
+    The next four examples generate the same `DatetimeIndex`, but vary
+    the combination of `start`, `end` and `periods`.
+
+    Specify `start` and `end`, with the default daily frequency.
+
+    >>> pd.date_range(start="1/1/2018", end="1/08/2018")
+    DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
+                   '2018-01-05', '2018-01-06', '2018-01-07', '2018-01-08'],
+                  dtype='datetime64[ns]', freq='D')
+
+    Specify timezone-aware `start` and `end`, with the default daily frequency.
+
+    >>> pd.date_range(
+    ...     start=pd.to_datetime("1/1/2018").tz_localize("Europe/Berlin"),
+    ...     end=pd.to_datetime("1/08/2018").tz_localize("Europe/Berlin"),
+    ... )
+    DatetimeIndex(['2018-01-01 00:00:00+01:00', '2018-01-02 00:00:00+01:00',
+                   '2018-01-03 00:00:00+01:00', '2018-01-04 00:00:00+01:00',
+                   '2018-01-05 00:00:00+01:00', '2018-01-06 00:00:00+01:00',
+                   '2018-01-07 00:00:00+01:00', '2018-01-08 00:00:00+01:00'],
+                  dtype='datetime64[ns, Europe/Berlin]', freq='D')
+
+    Specify `start` and `periods`, the number of periods (days).
+
+    >>> pd.date_range(start="1/1/2018", periods=8)
+    DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
+                   '2018-01-05', '2018-01-06', '2018-01-07', '2018-01-08'],
+                  dtype='datetime64[ns]', freq='D')
+
+    Specify `end` and `periods`, the number of periods (days).
+
+    >>> pd.date_range(end="1/1/2018", periods=8)
+    DatetimeIndex(['2017-12-25', '2017-12-26', '2017-12-27', '2017-12-28',
+                   '2017-12-29', '2017-12-30', '2017-12-31', '2018-01-01'],
+                  dtype='datetime64[ns]', freq='D')
+
+    Specify `start`, `end`, and `periods`; the frequency is generated
+    automatically (linearly spaced).
+
+    >>> pd.date_range(start="2018-04-24", end="2018-04-27", periods=3)
+    DatetimeIndex(['2018-04-24 00:00:00', '2018-04-25 12:00:00',
+                   '2018-04-27 00:00:00'],
+                  dtype='datetime64[ns]', freq=None)
+
+    **Other Parameters**
+
+    Changed the `freq` (frequency) to ``'ME'`` (month end frequency).
+
+    >>> pd.date_range(start="1/1/2018", periods=5, freq="ME")
+    DatetimeIndex(['2018-01-31', '2018-02-28', '2018-03-31', '2018-04-30',
+                   '2018-05-31'],
+                  dtype='datetime64[ns]', freq='ME')
+
+    Multiples are allowed
+
+    >>> pd.date_range(start="1/1/2018", periods=5, freq="3ME")
+    DatetimeIndex(['2018-01-31', '2018-04-30', '2018-07-31', '2018-10-31',
+                   '2019-01-31'],
+                  dtype='datetime64[ns]', freq='3ME')
+
+    `freq` can also be specified as an Offset object.
+
+    >>> pd.date_range(start="1/1/2018", periods=5, freq=pd.offsets.MonthEnd(3))
+    DatetimeIndex(['2018-01-31', '2018-04-30', '2018-07-31', '2018-10-31',
+                   '2019-01-31'],
+                  dtype='datetime64[ns]', freq='3ME')
+
+    Specify `tz` to set the timezone.
+
+    >>> pd.date_range(start="1/1/2018", periods=5, tz="Asia/Tokyo")
+    DatetimeIndex(['2018-01-01 00:00:00+09:00', '2018-01-02 00:00:00+09:00',
+                   '2018-01-03 00:00:00+09:00', '2018-01-04 00:00:00+09:00',
+                   '2018-01-05 00:00:00+09:00'],
+                  dtype='datetime64[ns, Asia/Tokyo]', freq='D')
+
+    `inclusive` controls whether to include `start` and `end` that are on the
+    boundary. The default, "both", includes boundary points on either end.
+
+    >>> pd.date_range(start="2017-01-01", end="2017-01-04", inclusive="both")
+    DatetimeIndex(['2017-01-01', '2017-01-02', '2017-01-03', '2017-01-04'],
+                  dtype='datetime64[ns]', freq='D')
+
+    Use ``inclusive='left'`` to exclude `end` if it falls on the boundary.
+
+    >>> pd.date_range(start="2017-01-01", end="2017-01-04", inclusive="left")
+    DatetimeIndex(['2017-01-01', '2017-01-02', '2017-01-03'],
+                  dtype='datetime64[ns]', freq='D')
+
+    Use ``inclusive='right'`` to exclude `start` if it falls on the boundary, and
+    similarly ``inclusive='neither'`` will exclude both `start` and `end`.
+
+    >>> pd.date_range(start="2017-01-01", end="2017-01-04", inclusive="right")
+    DatetimeIndex(['2017-01-02', '2017-01-03', '2017-01-04'],
+                  dtype='datetime64[ns]', freq='D')
+
+    **Specify a unit**
+
+    >>> pd.date_range(start="2017-01-01", periods=10, freq="100YS", unit="s")
+    DatetimeIndex(['2017-01-01', '2117-01-01', '2217-01-01', '2317-01-01',
+                   '2417-01-01', '2517-01-01', '2617-01-01', '2717-01-01',
+                   '2817-01-01', '2917-01-01'],
+                  dtype='datetime64[s]', freq='100YS-JAN')
+    """
+    if freq is None and com.any_none(periods, start, end):
+        freq = "D"
+
+    dtarr = DatetimeArray._generate_range(
+        start=start,
+        end=end,
+        periods=periods,
+        freq=freq,
+        tz=tz,
+        normalize=normalize,
+        inclusive=inclusive,
+        unit=unit,
+        **kwargs,
+    )
+    return DatetimeIndex._simple_new(dtarr, name=name)
+
+
+@set_module("pandas")
+def bdate_range(
+    start=None,
+    end=None,
+    periods: int | None = None,
+    freq: Frequency | dt.timedelta = "B",
+    tz=None,
+    normalize: bool = True,
+    name: Hashable | None = None,
+    weekmask=None,
+    holidays=None,
+    inclusive: IntervalClosedType = "both",
+    **kwargs,
+) -> DatetimeIndex:
+    """
+    Return a fixed frequency DatetimeIndex with business day as the default.
+
+    Parameters
+    ----------
+    start : str or datetime-like, default None
+        Left bound for generating dates.
+    end : str or datetime-like, default None
+        Right bound for generating dates.
+    periods : int, default None
+        Number of periods to generate.
+    freq : str, Timedelta, datetime.timedelta, or DateOffset, default 'B'
+        Frequency strings can have multiples, e.g. '5h'. The default is
+        business daily ('B').
+    tz : str or None
+        Time zone name for returning localized DatetimeIndex, for example
+        Asia/Beijing.
+    normalize : bool, default False
+        Normalize start/end dates to midnight before generating date range.
+    name : str, default None
+        Name of the resulting DatetimeIndex.
+    weekmask : str or None, default None
+        Weekmask of valid business days, passed to ``numpy.busdaycalendar``,
+        only used when custom frequency strings are passed.  The default
+        value None is equivalent to 'Mon Tue Wed Thu Fri'.
+    holidays : list-like or None, default None
+        Dates to exclude from the set of valid business days, passed to
+        ``numpy.busdaycalendar``, only used when custom frequency strings
+        are passed.
+    inclusive : {"both", "neither", "left", "right"}, default "both"
+        Include boundaries; Whether to set each bound as closed or open.
+
+        .. versionadded:: 1.4.0
+    **kwargs
+        For compatibility. Has no effect on the result.
+
+    Returns
+    -------
+    DatetimeIndex
+        Fixed frequency DatetimeIndex.
+
+    See Also
+    --------
+    date_range : Return a fixed frequency DatetimeIndex.
+    period_range : Return a fixed frequency PeriodIndex.
+    timedelta_range : Return a fixed frequency TimedeltaIndex.
+
+    Notes
+    -----
+    Of the four parameters: ``start``, ``end``, ``periods``, and ``freq``,
+    exactly three must be specified.  Specifying ``freq`` is a requirement
+    for ``bdate_range``.  Use ``date_range`` if specifying ``freq`` is not
+    desired.
+
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    Note how the two weekend days are skipped in the result.
+
+    >>> pd.bdate_range(start="1/1/2018", end="1/08/2018")
+    DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
+               '2018-01-05', '2018-01-08'],
+              dtype='datetime64[ns]', freq='B')
+    """
+    if freq is None:
+        msg = "freq must be specified for bdate_range; use date_range instead"
+        raise TypeError(msg)
+
+    if isinstance(freq, str) and freq.startswith("C"):
+        try:
+            weekmask = weekmask or "Mon Tue Wed Thu Fri"
+            freq = prefix_mapping[freq](holidays=holidays, weekmask=weekmask)
+        except (KeyError, TypeError) as err:
+            msg = f"invalid custom frequency string: {freq}"
+            raise ValueError(msg) from err
+    elif holidays or weekmask:
+        msg = (
+            "a custom frequency string is required when holidays or "
+            f"weekmask are passed, got frequency {freq}"
+        )
+        raise ValueError(msg)
+
+    return date_range(
+        start=start,
+        end=end,
+        periods=periods,
+        freq=freq,
+        tz=tz,
+        normalize=normalize,
+        name=name,
+        inclusive=inclusive,
+        **kwargs,
+    )
+
+
+def _time_to_micros(time_obj: dt.time) -> int:
+    seconds = time_obj.hour * 60 * 60 + 60 * time_obj.minute + time_obj.second
+    return 1_000_000 * seconds + time_obj.microsecond
diff --git a/pandas/pandas/core/indexes/extension.py b/pandas/pandas/core/indexes/extension.py
new file mode 100644
index 0000000000000..2eeacfb769be4
--- /dev/null
+++ b/pandas/pandas/core/indexes/extension.py
@@ -0,0 +1,176 @@
+"""
+Shared methods for Index subclasses backed by ExtensionArray.
+"""
+
+from __future__ import annotations
+
+from inspect import signature
+from typing import (
+    TYPE_CHECKING,
+    TypeVar,
+)
+
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.generic import ABCDataFrame
+
+from pandas.core.indexes.base import Index
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    import numpy as np
+
+    from pandas._typing import (
+        ArrayLike,
+        npt,
+    )
+
+    from pandas.core.arrays import IntervalArray
+    from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+
+_ExtensionIndexT = TypeVar("_ExtensionIndexT", bound="ExtensionIndex")
+
+
+def _inherit_from_data(
+    name: str, delegate: type, cache: bool = False, wrap: bool = False
+):
+    """
+    Make an alias for a method of the underlying ExtensionArray.
+
+    Parameters
+    ----------
+    name : str
+        Name of an attribute the class should inherit from its EA parent.
+    delegate : class
+    cache : bool, default False
+        Whether to convert wrapped properties into cache_readonly
+    wrap : bool, default False
+        Whether to wrap the inherited result in an Index.
+
+    Returns
+    -------
+    attribute, method, property, or cache_readonly
+    """
+    attr = getattr(delegate, name)
+
+    if isinstance(attr, property) or type(attr).__name__ == "getset_descriptor":
+        # getset_descriptor i.e. property defined in cython class
+        if cache:
+
+            def cached(self):
+                return getattr(self._data, name)
+
+            cached.__name__ = name
+            cached.__doc__ = attr.__doc__
+            method = cache_readonly(cached)
+
+        else:
+
+            def fget(self):
+                result = getattr(self._data, name)
+                if wrap:
+                    if isinstance(result, type(self._data)):
+                        return type(self)._simple_new(result, name=self.name)
+                    elif isinstance(result, ABCDataFrame):
+                        return result.set_index(self)
+                    return Index(result, name=self.name, dtype=result.dtype)
+                return result
+
+            def fset(self, value) -> None:
+                setattr(self._data, name, value)
+
+            fget.__name__ = name
+            fget.__doc__ = attr.__doc__
+
+            method = property(fget, fset)
+
+    elif not callable(attr):
+        # just a normal attribute, no wrapping
+        method = attr
+
+    else:
+        # error: Incompatible redefinition (redefinition with type "Callable[[Any,
+        # VarArg(Any), KwArg(Any)], Any]", original type "property")
+        def method(self, *args, **kwargs):  # type: ignore[misc]
+            if "inplace" in kwargs:
+                raise ValueError(f"cannot use inplace with {type(self).__name__}")
+            result = attr(self._data, *args, **kwargs)
+            if wrap:
+                if isinstance(result, type(self._data)):
+                    return type(self)._simple_new(result, name=self.name)
+                elif isinstance(result, ABCDataFrame):
+                    return result.set_index(self)
+                return Index(result, name=self.name, dtype=result.dtype)
+            return result
+
+        # error: "property" has no attribute "__name__"
+        method.__name__ = name  # type: ignore[attr-defined]
+        method.__doc__ = attr.__doc__
+        method.__signature__ = signature(attr)  # type: ignore[attr-defined]
+    return method
+
+
+def inherit_names(
+    names: list[str], delegate: type, cache: bool = False, wrap: bool = False
+) -> Callable[[type[_ExtensionIndexT]], type[_ExtensionIndexT]]:
+    """
+    Class decorator to pin attributes from an ExtensionArray to a Index subclass.
+
+    Parameters
+    ----------
+    names : List[str]
+    delegate : class
+    cache : bool, default False
+    wrap : bool, default False
+        Whether to wrap the inherited result in an Index.
+    """
+
+    def wrapper(cls: type[_ExtensionIndexT]) -> type[_ExtensionIndexT]:
+        for name in names:
+            meth = _inherit_from_data(name, delegate, cache=cache, wrap=wrap)
+            setattr(cls, name, meth)
+
+        return cls
+
+    return wrapper
+
+
+class ExtensionIndex(Index):
+    """
+    Index subclass for indexes backed by ExtensionArray.
+    """
+
+    # The base class already passes through to _data:
+    #  size, __len__, dtype
+
+    _data: IntervalArray | NDArrayBackedExtensionArray
+
+    # ---------------------------------------------------------------------
+
+    def _validate_fill_value(self, value):
+        """
+        Convert value to be insertable to underlying array.
+        """
+        return self._data._validate_setitem_value(value)
+
+    @cache_readonly
+    def _isnan(self) -> npt.NDArray[np.bool_]:
+        # error: Incompatible return value type (got "ExtensionArray", expected
+        # "ndarray")
+        return self._data.isna()  # type: ignore[return-value]
+
+
+class NDArrayBackedExtensionIndex(ExtensionIndex):
+    """
+    Index subclass for indexes backed by NDArrayBackedExtensionArray.
+    """
+
+    _data: NDArrayBackedExtensionArray
+
+    def _get_engine_target(self) -> np.ndarray:
+        return self._data._ndarray
+
+    def _from_join_target(self, result: np.ndarray) -> ArrayLike:
+        assert result.dtype == self._data._ndarray.dtype
+        return self._data._from_backing_data(result)
diff --git a/pandas/pandas/core/indexes/frozen.py b/pandas/pandas/core/indexes/frozen.py
new file mode 100644
index 0000000000000..254bd71ade209
--- /dev/null
+++ b/pandas/pandas/core/indexes/frozen.py
@@ -0,0 +1,123 @@
+"""
+frozen (immutable) data structures to support MultiIndexing
+
+These are used for:
+
+- .names (FrozenList)
+
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    NoReturn,
+)
+
+from pandas.core.base import PandasObject
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from pandas._typing import Self
+
+
+class FrozenList(PandasObject, list):
+    """
+    Container that doesn't allow setting item *but*
+    because it's technically hashable, will be used
+    for lookups, appropriately, etc.
+    """
+
+    # Side note: This has to be of type list. Otherwise,
+    #            it messes up PyTables type checks.
+
+    def union(self, other) -> FrozenList:
+        """
+        Returns a FrozenList with other concatenated to the end of self.
+
+        Parameters
+        ----------
+        other : array-like
+            The array-like whose elements we are concatenating.
+
+        Returns
+        -------
+        FrozenList
+            The collection difference between self and other.
+        """
+        if isinstance(other, tuple):
+            other = list(other)
+        return type(self)(super().__add__(other))
+
+    def difference(self, other) -> FrozenList:
+        """
+        Returns a FrozenList with elements from other removed from self.
+
+        Parameters
+        ----------
+        other : array-like
+            The array-like whose elements we are removing self.
+
+        Returns
+        -------
+        FrozenList
+            The collection difference between self and other.
+        """
+        other = set(other)
+        temp = [x for x in self if x not in other]
+        return type(self)(temp)
+
+    # TODO: Consider deprecating these in favor of `union` (xref gh-15506)
+    # error: Incompatible types in assignment (expression has type
+    # "Callable[[FrozenList, Any], FrozenList]", base class "list" defined the
+    # type as overloaded function)
+    __add__ = __iadd__ = union  # type: ignore[assignment]
+
+    def __getitem__(self, n):
+        if isinstance(n, slice):
+            return type(self)(super().__getitem__(n))
+        return super().__getitem__(n)
+
+    def __radd__(self, other) -> Self:
+        if isinstance(other, tuple):
+            other = list(other)
+        return type(self)(other + list(self))
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, (tuple, FrozenList)):
+            other = list(other)
+        return super().__eq__(other)
+
+    __req__ = __eq__
+
+    def __mul__(self, other) -> Self:
+        return type(self)(super().__mul__(other))
+
+    __imul__ = __mul__
+
+    def __reduce__(self):
+        return type(self), (list(self),)
+
+    # error: Signature of "__hash__" incompatible with supertype "list"
+    def __hash__(self) -> int:  # type: ignore[override]
+        return hash(tuple(self))
+
+    def _disabled(self, *args, **kwargs) -> NoReturn:
+        """
+        This method will not function because object is immutable.
+        """
+        raise TypeError(f"'{type(self).__name__}' does not support mutable operations.")
+
+    def __str__(self) -> str:
+        return pprint_thing(
+            self, quote_strings=True, escape_chars=("\t", "\r", "\n", "'")
+        )
+
+    def __repr__(self) -> str:
+        return f"{type(self).__name__}({self!s})"
+
+    __setitem__ = __setslice__ = _disabled  # type: ignore[assignment]
+    __delitem__ = __delslice__ = _disabled
+    pop = append = extend = _disabled
+    remove = sort = insert = _disabled  # type: ignore[assignment]
diff --git a/pandas/pandas/core/indexes/interval.py b/pandas/pandas/core/indexes/interval.py
new file mode 100644
index 0000000000000..13811c28e6c1e
--- /dev/null
+++ b/pandas/pandas/core/indexes/interval.py
@@ -0,0 +1,1302 @@
+"""define the IntervalIndex"""
+
+from __future__ import annotations
+
+from operator import (
+    le,
+    lt,
+)
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.interval import (
+    Interval,
+    IntervalMixin,
+    IntervalTree,
+)
+from pandas._libs.tslibs import (
+    BaseOffset,
+    Period,
+    Timedelta,
+    Timestamp,
+    to_offset,
+)
+from pandas.errors import InvalidIndexError
+from pandas.util._decorators import (
+    Appender,
+    cache_readonly,
+    set_module,
+)
+from pandas.util._exceptions import rewrite_exception
+
+from pandas.core.dtypes.cast import (
+    find_common_type,
+    infer_dtype_from_scalar,
+    maybe_box_datetimelike,
+    maybe_downcast_numeric,
+    maybe_upcast_numeric_to_64bit,
+)
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_number,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    IntervalDtype,
+)
+from pandas.core.dtypes.missing import is_valid_na_for_dtype
+
+from pandas.core.algorithms import unique
+from pandas.core.arrays.datetimelike import validate_periods
+from pandas.core.arrays.interval import (
+    IntervalArray,
+    _interval_shared_docs,
+)
+import pandas.core.common as com
+from pandas.core.indexers import is_valid_positional_slice
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.base import (
+    Index,
+    _index_shared_docs,
+    ensure_index,
+    maybe_extract_name,
+)
+from pandas.core.indexes.datetimes import (
+    DatetimeIndex,
+    date_range,
+)
+from pandas.core.indexes.extension import (
+    ExtensionIndex,
+    inherit_names,
+)
+from pandas.core.indexes.multi import MultiIndex
+from pandas.core.indexes.timedeltas import (
+    TimedeltaIndex,
+    timedelta_range,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import (
+        Dtype,
+        DtypeObj,
+        IntervalClosedType,
+        Self,
+        npt,
+    )
+_index_doc_kwargs = dict(ibase._index_doc_kwargs)
+
+_index_doc_kwargs.update(
+    {
+        "klass": "IntervalIndex",
+        "qualname": "IntervalIndex",
+        "target_klass": "IntervalIndex or list of Intervals",
+        "name": textwrap.dedent(
+            """\
+         name : object, optional
+              Name to be stored in the index.
+         """
+        ),
+    }
+)
+
+
+def _get_next_label(label):
+    # see test_slice_locs_with_ints_and_floats_succeeds
+    dtype = getattr(label, "dtype", type(label))
+    if isinstance(label, (Timestamp, Timedelta)):
+        dtype = "datetime64[ns]"
+    dtype = pandas_dtype(dtype)
+
+    if lib.is_np_dtype(dtype, "mM") or isinstance(dtype, DatetimeTZDtype):
+        return label + np.timedelta64(1, "ns")
+    elif is_integer_dtype(dtype):
+        return label + 1
+    elif is_float_dtype(dtype):
+        return np.nextafter(label, np.inf)
+    else:
+        raise TypeError(f"cannot determine next label for type {type(label)!r}")
+
+
+def _get_prev_label(label):
+    # see test_slice_locs_with_ints_and_floats_succeeds
+    dtype = getattr(label, "dtype", type(label))
+    if isinstance(label, (Timestamp, Timedelta)):
+        dtype = "datetime64[ns]"
+    dtype = pandas_dtype(dtype)
+
+    if lib.is_np_dtype(dtype, "mM") or isinstance(dtype, DatetimeTZDtype):
+        return label - np.timedelta64(1, "ns")
+    elif is_integer_dtype(dtype):
+        return label - 1
+    elif is_float_dtype(dtype):
+        return np.nextafter(label, -np.inf)
+    else:
+        raise TypeError(f"cannot determine next label for type {type(label)!r}")
+
+
+def _new_IntervalIndex(cls, d):
+    """
+    This is called upon unpickling, rather than the default which doesn't have
+    arguments and breaks __new__.
+    """
+    return cls.from_arrays(**d)
+
+
+@Appender(
+    _interval_shared_docs["class"]
+    % {
+        "klass": "IntervalIndex",
+        "summary": "Immutable index of intervals that are closed on the same side.",
+        "name": _index_doc_kwargs["name"],
+        "extra_attributes": "is_overlapping\nvalues\n",
+        "extra_methods": "",
+        "examples": textwrap.dedent(
+            """\
+    Examples
+    --------
+    A new ``IntervalIndex`` is typically constructed using
+    :func:`interval_range`:
+
+    >>> pd.interval_range(start=0, end=5)
+    IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]],
+                  dtype='interval[int64, right]')
+
+    It may also be constructed using one of the constructor
+    methods: :meth:`IntervalIndex.from_arrays`,
+    :meth:`IntervalIndex.from_breaks`, and :meth:`IntervalIndex.from_tuples`.
+
+    See further examples in the doc strings of ``interval_range`` and the
+    mentioned constructor methods.
+    """
+        ),
+    }
+)
+@inherit_names(["set_closed", "to_tuples"], IntervalArray, wrap=True)
+@inherit_names(
+    [
+        "__array__",
+        "overlaps",
+        "contains",
+        "closed_left",
+        "closed_right",
+        "open_left",
+        "open_right",
+        "is_empty",
+    ],
+    IntervalArray,
+)
+@inherit_names(["is_non_overlapping_monotonic", "closed"], IntervalArray, cache=True)
+@set_module("pandas")
+class IntervalIndex(ExtensionIndex):
+    _typ = "intervalindex"
+
+    # annotate properties pinned via inherit_names
+    closed: IntervalClosedType
+    is_non_overlapping_monotonic: bool
+    closed_left: bool
+    closed_right: bool
+    open_left: bool
+    open_right: bool
+
+    _data: IntervalArray
+    _values: IntervalArray
+    _can_hold_strings = False
+    _data_cls = IntervalArray
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data,
+        closed: IntervalClosedType | None = None,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        name: Hashable | None = None,
+        verify_integrity: bool = True,
+    ) -> Self:
+        name = maybe_extract_name(name, data, cls)
+
+        with rewrite_exception("IntervalArray", cls.__name__):
+            array = IntervalArray(
+                data,
+                closed=closed,
+                copy=copy,
+                dtype=dtype,
+                verify_integrity=verify_integrity,
+            )
+
+        return cls._simple_new(array, name)
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_breaks"]
+        % {
+            "klass": "IntervalIndex",
+            "name": textwrap.dedent(
+                """
+             name : str, optional
+                  Name of the resulting IntervalIndex."""
+            ),
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.IntervalIndex.from_breaks([0, 1, 2, 3])
+        IntervalIndex([(0, 1], (1, 2], (2, 3]],
+                      dtype='interval[int64, right]')
+        """
+            ),
+        }
+    )
+    def from_breaks(
+        cls,
+        breaks,
+        closed: IntervalClosedType | None = "right",
+        name: Hashable | None = None,
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> IntervalIndex:
+        with rewrite_exception("IntervalArray", cls.__name__):
+            array = IntervalArray.from_breaks(
+                breaks, closed=closed, copy=copy, dtype=dtype
+            )
+        return cls._simple_new(array, name=name)
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_arrays"]
+        % {
+            "klass": "IntervalIndex",
+            "name": textwrap.dedent(
+                """
+             name : str, optional
+                  Name of the resulting IntervalIndex."""
+            ),
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.IntervalIndex.from_arrays([0, 1, 2], [1, 2, 3])
+        IntervalIndex([(0, 1], (1, 2], (2, 3]],
+                      dtype='interval[int64, right]')
+        """
+            ),
+        }
+    )
+    def from_arrays(
+        cls,
+        left,
+        right,
+        closed: IntervalClosedType = "right",
+        name: Hashable | None = None,
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> IntervalIndex:
+        with rewrite_exception("IntervalArray", cls.__name__):
+            array = IntervalArray.from_arrays(
+                left, right, closed, copy=copy, dtype=dtype
+            )
+        return cls._simple_new(array, name=name)
+
+    @classmethod
+    @Appender(
+        _interval_shared_docs["from_tuples"]
+        % {
+            "klass": "IntervalIndex",
+            "name": textwrap.dedent(
+                """
+             name : str, optional
+                  Name of the resulting IntervalIndex."""
+            ),
+            "examples": textwrap.dedent(
+                """\
+        Examples
+        --------
+        >>> pd.IntervalIndex.from_tuples([(0, 1), (1, 2)])
+        IntervalIndex([(0, 1], (1, 2]],
+                       dtype='interval[int64, right]')
+        """
+            ),
+        }
+    )
+    def from_tuples(
+        cls,
+        data,
+        closed: IntervalClosedType = "right",
+        name: Hashable | None = None,
+        copy: bool = False,
+        dtype: Dtype | None = None,
+    ) -> IntervalIndex:
+        with rewrite_exception("IntervalArray", cls.__name__):
+            arr = IntervalArray.from_tuples(data, closed=closed, copy=copy, dtype=dtype)
+        return cls._simple_new(arr, name=name)
+
+    # --------------------------------------------------------------------
+    # error: Return type "IntervalTree" of "_engine" incompatible with return type
+    # "Union[IndexEngine, ExtensionEngine]" in supertype "Index"
+    @cache_readonly
+    def _engine(self) -> IntervalTree:  # type: ignore[override]
+        # IntervalTree does not supports numpy array unless they are 64 bit
+        left = self._maybe_convert_i8(self.left)
+        left = maybe_upcast_numeric_to_64bit(left)
+        right = self._maybe_convert_i8(self.right)
+        right = maybe_upcast_numeric_to_64bit(right)
+        return IntervalTree(left, right, closed=self.closed)
+
+    def __contains__(self, key: Any) -> bool:
+        """
+        return a boolean if this key is IN the index
+        We *only* accept an Interval
+
+        Parameters
+        ----------
+        key : Interval
+
+        Returns
+        -------
+        bool
+        """
+        hash(key)
+        if not isinstance(key, Interval):
+            if is_valid_na_for_dtype(key, self.dtype):
+                return self.hasnans
+            return False
+
+        try:
+            self.get_loc(key)
+            return True
+        except KeyError:
+            return False
+
+    def _getitem_slice(self, slobj: slice) -> IntervalIndex:
+        """
+        Fastpath for __getitem__ when we know we have a slice.
+        """
+        res = self._data[slobj]
+        return type(self)._simple_new(res, name=self._name)
+
+    @cache_readonly
+    def _multiindex(self) -> MultiIndex:
+        return MultiIndex.from_arrays([self.left, self.right], names=["left", "right"])
+
+    def __reduce__(self):
+        d = {
+            "left": self.left,
+            "right": self.right,
+            "closed": self.closed,
+            "name": self.name,
+        }
+        return _new_IntervalIndex, (type(self), d), None
+
+    @property
+    def inferred_type(self) -> str:
+        """Return a string of the type inferred from the values"""
+        return "interval"
+
+    # Cannot determine type of "memory_usage"
+    @Appender(Index.memory_usage.__doc__)  # type: ignore[has-type]
+    def memory_usage(self, deep: bool = False) -> int:
+        # we don't use an explicit engine
+        # so return the bytes here
+        return self.left.memory_usage(deep=deep) + self.right.memory_usage(deep=deep)
+
+    # IntervalTree doesn't have a is_monotonic_decreasing, so have to override
+    #  the Index implementation
+    @cache_readonly
+    def is_monotonic_decreasing(self) -> bool:
+        """
+        Return True if the IntervalIndex is monotonic decreasing (only equal or
+        decreasing values), else False
+        """
+        return self[::-1].is_monotonic_increasing
+
+    @cache_readonly
+    def is_unique(self) -> bool:
+        """
+        Return True if the IntervalIndex contains unique elements, else False.
+        """
+        left = self.left
+        right = self.right
+
+        if self.isna().sum() > 1:
+            return False
+
+        if left.is_unique or right.is_unique:
+            return True
+
+        seen_pairs = set()
+        check_idx = np.where(left.duplicated(keep=False))[0]
+        for idx in check_idx:
+            pair = (left[idx], right[idx])
+            if pair in seen_pairs:
+                return False
+            seen_pairs.add(pair)
+
+        return True
+
+    @property
+    def is_overlapping(self) -> bool:
+        """
+        Return True if the IntervalIndex has overlapping intervals, else False.
+
+        Two intervals overlap if they share a common point, including closed
+        endpoints. Intervals that only have an open endpoint in common do not
+        overlap.
+
+        Returns
+        -------
+        bool
+            Boolean indicating if the IntervalIndex has overlapping intervals.
+
+        See Also
+        --------
+        Interval.overlaps : Check whether two Interval objects overlap.
+        IntervalIndex.overlaps : Check an IntervalIndex elementwise for
+            overlaps.
+
+        Examples
+        --------
+        >>> index = pd.IntervalIndex.from_tuples([(0, 2), (1, 3), (4, 5)])
+        >>> index
+        IntervalIndex([(0, 2], (1, 3], (4, 5]],
+              dtype='interval[int64, right]')
+        >>> index.is_overlapping
+        True
+
+        Intervals that share closed endpoints overlap:
+
+        >>> index = pd.interval_range(0, 3, closed="both")
+        >>> index
+        IntervalIndex([[0, 1], [1, 2], [2, 3]],
+              dtype='interval[int64, both]')
+        >>> index.is_overlapping
+        True
+
+        Intervals that only have an open endpoint in common do not overlap:
+
+        >>> index = pd.interval_range(0, 3, closed="left")
+        >>> index
+        IntervalIndex([[0, 1), [1, 2), [2, 3)],
+              dtype='interval[int64, left]')
+        >>> index.is_overlapping
+        False
+        """
+        # GH 23309
+        return self._engine.is_overlapping
+
+    def _needs_i8_conversion(self, key) -> bool:
+        """
+        Check if a given key needs i8 conversion. Conversion is necessary for
+        Timestamp, Timedelta, DatetimeIndex, and TimedeltaIndex keys. An
+        Interval-like requires conversion if its endpoints are one of the
+        aforementioned types.
+
+        Assumes that any list-like data has already been cast to an Index.
+
+        Parameters
+        ----------
+        key : scalar or Index-like
+            The key that should be checked for i8 conversion
+
+        Returns
+        -------
+        bool
+        """
+        key_dtype = getattr(key, "dtype", None)
+        if isinstance(key_dtype, IntervalDtype) or isinstance(key, Interval):
+            return self._needs_i8_conversion(key.left)
+
+        i8_types = (Timestamp, Timedelta, DatetimeIndex, TimedeltaIndex)
+        return isinstance(key, i8_types)
+
+    def _maybe_convert_i8(self, key):
+        """
+        Maybe convert a given key to its equivalent i8 value(s). Used as a
+        preprocessing step prior to IntervalTree queries (self._engine), which
+        expects numeric data.
+
+        Parameters
+        ----------
+        key : scalar or list-like
+            The key that should maybe be converted to i8.
+
+        Returns
+        -------
+        scalar or list-like
+            The original key if no conversion occurred, int if converted scalar,
+            Index with an int64 dtype if converted list-like.
+        """
+        if is_list_like(key):
+            key = ensure_index(key)
+            key = maybe_upcast_numeric_to_64bit(key)
+
+        if not self._needs_i8_conversion(key):
+            return key
+
+        scalar = is_scalar(key)
+        key_dtype = getattr(key, "dtype", None)
+        if isinstance(key_dtype, IntervalDtype) or isinstance(key, Interval):
+            # convert left/right and reconstruct
+            left = self._maybe_convert_i8(key.left)
+            right = self._maybe_convert_i8(key.right)
+            constructor = Interval if scalar else IntervalIndex.from_arrays
+            return constructor(left, right, closed=self.closed)
+
+        if scalar:
+            # Timestamp/Timedelta
+            key_dtype, key_i8 = infer_dtype_from_scalar(key)
+            if isinstance(key, Period):
+                key_i8 = key.ordinal
+            elif isinstance(key_i8, Timestamp):
+                key_i8 = key_i8._value
+            elif isinstance(key_i8, (np.datetime64, np.timedelta64)):
+                key_i8 = key_i8.view("i8")
+        else:
+            # DatetimeIndex/TimedeltaIndex
+            key_dtype, key_i8 = key.dtype, Index(key.asi8)
+            if key.hasnans:
+                # convert NaT from its i8 value to np.nan so it's not viewed
+                # as a valid value, maybe causing errors (e.g. is_overlapping)
+                key_i8 = key_i8.where(~key._isnan)
+
+        # ensure consistency with IntervalIndex subtype
+        # error: Item "ExtensionDtype"/"dtype[Any]" of "Union[dtype[Any],
+        # ExtensionDtype]" has no attribute "subtype"
+        subtype = self.dtype.subtype  # type: ignore[union-attr]
+
+        if subtype != key_dtype:
+            raise ValueError(
+                f"Cannot index an IntervalIndex of subtype {subtype} with "
+                f"values of dtype {key_dtype}"
+            )
+
+        return key_i8
+
+    def _searchsorted_monotonic(self, label, side: Literal["left", "right"] = "left"):
+        if not self.is_non_overlapping_monotonic:
+            raise KeyError(
+                "can only get slices from an IntervalIndex if bounds are "
+                "non-overlapping and all monotonic increasing or decreasing"
+            )
+
+        if isinstance(label, (IntervalMixin, IntervalIndex)):
+            raise NotImplementedError("Interval objects are not currently supported")
+
+        # GH 20921: "not is_monotonic_increasing" for the second condition
+        # instead of "is_monotonic_decreasing" to account for single element
+        # indexes being both increasing and decreasing
+        if (side == "left" and self.left.is_monotonic_increasing) or (
+            side == "right" and not self.left.is_monotonic_increasing
+        ):
+            sub_idx = self.right
+            if self.open_right:
+                label = _get_next_label(label)
+        else:
+            sub_idx = self.left
+            if self.open_left:
+                label = _get_prev_label(label)
+
+        return sub_idx._searchsorted_monotonic(label, side)
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    def get_loc(self, key) -> int | slice | np.ndarray:
+        """
+        Get integer location, slice or boolean mask for requested label.
+
+        The `get_loc` method is used to retrieve the integer index, a slice for
+        slicing objects, or a boolean mask indicating the presence of the label
+        in the `IntervalIndex`.
+
+        Parameters
+        ----------
+        key : label
+            The value or range to find in the IntervalIndex.
+
+        Returns
+        -------
+        int if unique index, slice if monotonic index, else mask
+            The position or positions found. This could be a single
+            number, a range, or an array of true/false values
+            indicating the position(s) of the label.
+
+        See Also
+        --------
+        IntervalIndex.get_indexer_non_unique : Compute indexer and
+            mask for new index given the current index.
+        Index.get_loc : Similar method in the base Index class.
+
+        Examples
+        --------
+        >>> i1, i2 = pd.Interval(0, 1), pd.Interval(1, 2)
+        >>> index = pd.IntervalIndex([i1, i2])
+        >>> index.get_loc(1)
+        0
+
+        You can also supply a point inside an interval.
+
+        >>> index.get_loc(1.5)
+        1
+
+        If a label is in several intervals, you get the locations of all the
+        relevant intervals.
+
+        >>> i3 = pd.Interval(0, 2)
+        >>> overlapping_index = pd.IntervalIndex([i1, i2, i3])
+        >>> overlapping_index.get_loc(0.5)
+        array([ True, False,  True])
+
+        Only exact matches will be returned if an interval is provided.
+
+        >>> index.get_loc(pd.Interval(0, 1))
+        0
+        """
+        self._check_indexing_error(key)
+
+        if isinstance(key, Interval):
+            if self.closed != key.closed:
+                raise KeyError(key)
+            mask = (self.left == key.left) & (self.right == key.right)
+        elif is_valid_na_for_dtype(key, self.dtype):
+            mask = self.isna()
+        else:
+            # assume scalar
+            op_left = le if self.closed_left else lt
+            op_right = le if self.closed_right else lt
+            try:
+                mask = op_left(self.left, key) & op_right(key, self.right)
+            except TypeError as err:
+                # scalar is not comparable to II subtype --> invalid label
+                raise KeyError(key) from err
+
+        matches = mask.sum()
+        if matches == 0:
+            raise KeyError(key)
+        if matches == 1:
+            return mask.argmax()
+
+        res = lib.maybe_booleans_to_slice(mask.view("u1"))
+        if isinstance(res, slice) and res.stop is None:
+            # TODO: DO this in maybe_booleans_to_slice?
+            res = slice(res.start, len(self), res.step)
+        return res
+
+    def _get_indexer(
+        self,
+        target: Index,
+        method: str | None = None,
+        limit: int | None = None,
+        tolerance: Any | None = None,
+    ) -> npt.NDArray[np.intp]:
+        if isinstance(target, IntervalIndex):
+            # We only get here with not self.is_overlapping
+            # -> at most one match per interval in target
+            # want exact matches -> need both left/right to match, so defer to
+            # left/right get_indexer, compare elementwise, equality -> match
+            indexer = self._get_indexer_unique_sides(target)
+
+        elif not (is_object_dtype(target.dtype) or is_string_dtype(target.dtype)):
+            # homogeneous scalar index: use IntervalTree
+            # we should always have self._should_partial_index(target) here
+            target = self._maybe_convert_i8(target)
+            indexer = self._engine.get_indexer(target.values)
+        else:
+            # heterogeneous scalar index: defer elementwise to get_loc
+            # we should always have self._should_partial_index(target) here
+            return self._get_indexer_pointwise(target)[0]
+
+        return ensure_platform_int(indexer)
+
+    @Appender(_index_shared_docs["get_indexer_non_unique"] % _index_doc_kwargs)
+    def get_indexer_non_unique(
+        self, target: Index
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        target = ensure_index(target)
+
+        if not self._should_compare(target) and not self._should_partial_index(target):
+            # e.g. IntervalIndex with different closed or incompatible subtype
+            #  -> no matches
+            return self._get_indexer_non_comparable(target, None, unique=False)
+
+        elif isinstance(target, IntervalIndex):
+            if self.left.is_unique and self.right.is_unique:
+                # fastpath available even if we don't have self._index_as_unique
+                indexer = self._get_indexer_unique_sides(target)
+                missing = (indexer == -1).nonzero()[0]
+            else:
+                return self._get_indexer_pointwise(target)
+
+        elif is_object_dtype(target.dtype) or not self._should_partial_index(target):
+            # target might contain intervals: defer elementwise to get_loc
+            return self._get_indexer_pointwise(target)
+
+        else:
+            # Note: this case behaves differently from other Index subclasses
+            #  because IntervalIndex does partial-int indexing
+            target = self._maybe_convert_i8(target)
+            indexer, missing = self._engine.get_indexer_non_unique(target.values)
+
+        return ensure_platform_int(indexer), ensure_platform_int(missing)
+
+    def _get_indexer_unique_sides(self, target: IntervalIndex) -> npt.NDArray[np.intp]:
+        """
+        _get_indexer specialized to the case where both of our sides are unique.
+        """
+        # Caller is responsible for checking
+        #  `self.left.is_unique and self.right.is_unique`
+
+        left_indexer = self.left.get_indexer(target.left)
+        right_indexer = self.right.get_indexer(target.right)
+        indexer = np.where(left_indexer == right_indexer, left_indexer, -1)
+        return indexer
+
+    def _get_indexer_pointwise(
+        self, target: Index
+    ) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        """
+        pointwise implementation for get_indexer and get_indexer_non_unique.
+        """
+        indexer, missing = [], []
+        for i, key in enumerate(target):
+            try:
+                locs = self.get_loc(key)
+                if isinstance(locs, slice):
+                    # Only needed for get_indexer_non_unique
+                    locs = np.arange(locs.start, locs.stop, locs.step, dtype="intp")
+                elif lib.is_integer(locs):
+                    locs = np.array(locs, ndmin=1)
+                else:
+                    # otherwise we have ndarray[bool]
+                    locs = np.where(locs)[0]
+            except KeyError:
+                missing.append(i)
+                locs = np.array([-1])
+            except InvalidIndexError:
+                # i.e. non-scalar key e.g. a tuple.
+                # see test_append_different_columns_types_raises
+                missing.append(i)
+                locs = np.array([-1])
+
+            indexer.append(locs)
+
+        indexer = np.concatenate(indexer)
+        return ensure_platform_int(indexer), ensure_platform_int(missing)
+
+    @cache_readonly
+    def _index_as_unique(self) -> bool:
+        return not self.is_overlapping and self._engine._na_count < 2
+
+    _requires_unique_msg = (
+        "cannot handle overlapping indices; use IntervalIndex.get_indexer_non_unique"
+    )
+
+    def _convert_slice_indexer(self, key: slice, kind: Literal["loc", "getitem"]):
+        if not (key.step is None or key.step == 1):
+            # GH#31658 if label-based, we require step == 1,
+            #  if positional, we disallow float start/stop
+            msg = "label-based slicing with step!=1 is not supported for IntervalIndex"
+            if kind == "loc":
+                raise ValueError(msg)
+            if kind == "getitem":
+                if not is_valid_positional_slice(key):
+                    # i.e. this cannot be interpreted as a positional slice
+                    raise ValueError(msg)
+
+        return super()._convert_slice_indexer(key, kind)
+
+    @cache_readonly
+    def _should_fallback_to_positional(self) -> bool:
+        # integer lookups in Series.__getitem__ are unambiguously
+        #  positional in this case
+        # error: Item "ExtensionDtype"/"dtype[Any]" of "Union[dtype[Any],
+        # ExtensionDtype]" has no attribute "subtype"
+        return self.dtype.subtype.kind in "mM"  # type: ignore[union-attr]
+
+    def _maybe_cast_slice_bound(self, label, side: str):
+        return getattr(self, side)._maybe_cast_slice_bound(label, side)
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        if not isinstance(dtype, IntervalDtype):
+            return False
+        common_subtype = find_common_type([self.dtype, dtype])
+        return not is_object_dtype(common_subtype)
+
+    # --------------------------------------------------------------------
+
+    @cache_readonly
+    def left(self) -> Index:
+        """
+        Return left bounds of the intervals in the IntervalIndex.
+
+        The left bounds of each interval in the IntervalIndex are
+        returned as an Index. The datatype of the left bounds is the
+        same as the datatype of the endpoints of the intervals.
+
+        Returns
+        -------
+        Index
+            An Index containing the left bounds of the intervals.
+
+        See Also
+        --------
+        IntervalIndex.right : Return the right bounds of the intervals
+            in the IntervalIndex.
+        IntervalIndex.mid : Return the mid-point of the intervals in
+            the IntervalIndex.
+        IntervalIndex.length : Return the length of the intervals in
+            the IntervalIndex.
+
+        Examples
+        --------
+        >>> iv_idx = pd.IntervalIndex.from_arrays([1, 2, 3], [4, 5, 6], closed="right")
+        >>> iv_idx.left
+        Index([1, 2, 3], dtype='int64')
+
+        >>> iv_idx = pd.IntervalIndex.from_tuples(
+        ...     [(1, 4), (2, 5), (3, 6)], closed="left"
+        ... )
+        >>> iv_idx.left
+        Index([1, 2, 3], dtype='int64')
+        """
+        return Index(self._data.left, copy=False)
+
+    @cache_readonly
+    def right(self) -> Index:
+        """
+        Return right bounds of the intervals in the IntervalIndex.
+
+        The right bounds of each interval in the IntervalIndex are
+        returned as an Index. The datatype of the right bounds is the
+        same as the datatype of the endpoints of the intervals.
+
+        Returns
+        -------
+        Index
+            An Index containing the right bounds of the intervals.
+
+        See Also
+        --------
+        IntervalIndex.left : Return the left bounds of the intervals
+            in the IntervalIndex.
+        IntervalIndex.mid : Return the mid-point of the intervals in
+            the IntervalIndex.
+        IntervalIndex.length : Return the length of the intervals in
+            the IntervalIndex.
+
+        Examples
+        --------
+        >>> iv_idx = pd.IntervalIndex.from_arrays([1, 2, 3], [4, 5, 6], closed="right")
+        >>> iv_idx.right
+        Index([4, 5, 6], dtype='int64')
+
+        >>> iv_idx = pd.IntervalIndex.from_tuples(
+        ...     [(1, 4), (2, 5), (3, 6)], closed="left"
+        ... )
+        >>> iv_idx.right
+        Index([4, 5, 6], dtype='int64')
+        """
+        return Index(self._data.right, copy=False)
+
+    @cache_readonly
+    def mid(self) -> Index:
+        """
+        Return the midpoint of each interval in the IntervalIndex as an Index.
+
+        Each midpoint is calculated as the average of the left and right bounds
+        of each interval. The midpoints are returned as a pandas Index object.
+
+        Returns
+        -------
+        pandas.Index
+            An Index containing the midpoints of each interval.
+
+        See Also
+        --------
+        IntervalIndex.left : Return the left bounds of the intervals
+            in the IntervalIndex.
+        IntervalIndex.right : Return the right bounds of the intervals
+            in the IntervalIndex.
+        IntervalIndex.length : Return the length of the intervals in
+            the IntervalIndex.
+
+        Notes
+        -----
+        The midpoint is the average of the interval bounds, potentially resulting
+        in a floating-point number even if bounds are integers. The returned Index
+        will have a dtype that accurately holds the midpoints. This computation is
+        the same regardless of whether intervals are open or closed.
+
+        Examples
+        --------
+        >>> iv_idx = pd.IntervalIndex.from_arrays([1, 2, 3], [4, 5, 6])
+        >>> iv_idx.mid
+        Index([2.5, 3.5, 4.5], dtype='float64')
+
+        >>> iv_idx = pd.IntervalIndex.from_tuples([(1, 4), (2, 5), (3, 6)])
+        >>> iv_idx.mid
+        Index([2.5, 3.5, 4.5], dtype='float64')
+        """
+        return Index(self._data.mid, copy=False)
+
+    @property
+    def length(self) -> Index:
+        """
+        Calculate the length of each interval in the IntervalIndex.
+
+        This method returns a new Index containing the lengths of each interval
+        in the IntervalIndex. The length of an interval is defined as the difference
+        between its end and its start.
+
+        Returns
+        -------
+        Index
+            An Index containing the lengths of each interval.
+
+        See Also
+        --------
+        Interval.length : Return the length of the Interval.
+
+        Examples
+        --------
+        >>> intervals = pd.IntervalIndex.from_arrays(
+        ...     [1, 2, 3], [4, 5, 6], closed="right"
+        ... )
+        >>> intervals.length
+        Index([3, 3, 3], dtype='int64')
+
+        >>> intervals = pd.IntervalIndex.from_tuples([(1, 5), (6, 10), (11, 15)])
+        >>> intervals.length
+        Index([4, 4, 4], dtype='int64')
+        """
+        return Index(self._data.length, copy=False)
+
+    # --------------------------------------------------------------------
+    # Set Operations
+
+    def _intersection(self, other, sort: bool = False):
+        """
+        intersection specialized to the case with matching dtypes.
+        """
+        # For IntervalIndex we also know other.closed == self.closed
+        if self.left.is_unique and self.right.is_unique:
+            taken = self._intersection_unique(other)
+        elif other.left.is_unique and other.right.is_unique and self.isna().sum() <= 1:
+            # Swap other/self if other is unique and self does not have
+            # multiple NaNs
+            taken = other._intersection_unique(self)
+        else:
+            # duplicates
+            taken = self._intersection_non_unique(other)
+
+        if sort:
+            taken = taken.sort_values()
+
+        return taken
+
+    def _intersection_unique(self, other: IntervalIndex) -> IntervalIndex:
+        """
+        Used when the IntervalIndex does not have any common endpoint,
+        no matter left or right.
+        Return the intersection with another IntervalIndex.
+        Parameters
+        ----------
+        other : IntervalIndex
+        Returns
+        -------
+        IntervalIndex
+        """
+        # Note: this is much more performant than super()._intersection(other)
+        lindexer = self.left.get_indexer(other.left)
+        rindexer = self.right.get_indexer(other.right)
+
+        match = (lindexer == rindexer) & (lindexer != -1)
+        indexer = lindexer.take(match.nonzero()[0])
+        indexer = unique(indexer)
+
+        return self.take(indexer)
+
+    def _intersection_non_unique(self, other: IntervalIndex) -> IntervalIndex:
+        """
+        Used when the IntervalIndex does have some common endpoints,
+        on either sides.
+        Return the intersection with another IntervalIndex.
+
+        Parameters
+        ----------
+        other : IntervalIndex
+
+        Returns
+        -------
+        IntervalIndex
+        """
+        # Note: this is about 3.25x faster than super()._intersection(other)
+        #  in IntervalIndexMethod.time_intersection_both_duplicate(1000)
+        mask = np.zeros(len(self), dtype=bool)
+
+        if self.hasnans and other.hasnans:
+            first_nan_loc = np.arange(len(self))[self.isna()][0]
+            mask[first_nan_loc] = True
+
+        other_tups = set(zip(other.left, other.right))
+        for i, tup in enumerate(zip(self.left, self.right)):
+            if tup in other_tups:
+                mask[i] = True
+
+        return self[mask]
+
+    # --------------------------------------------------------------------
+
+    def _get_engine_target(self) -> np.ndarray:
+        # Note: we _could_ use libjoin functions by either casting to object
+        #  dtype or constructing tuples (faster than constructing Intervals)
+        #  but the libjoin fastpaths are no longer fast in these cases.
+        raise NotImplementedError(
+            "IntervalIndex does not use libjoin fastpaths or pass values to "
+            "IndexEngine objects"
+        )
+
+    def _from_join_target(self, result):
+        raise NotImplementedError("IntervalIndex does not use libjoin fastpaths")
+
+    # TODO: arithmetic operations
+
+
+def _is_valid_endpoint(endpoint) -> bool:
+    """
+    Helper for interval_range to check if start/end are valid types.
+    """
+    return any(
+        [
+            is_number(endpoint),
+            isinstance(endpoint, Timestamp),
+            isinstance(endpoint, Timedelta),
+            endpoint is None,
+        ]
+    )
+
+
+def _is_type_compatible(a, b) -> bool:
+    """
+    Helper for interval_range to check type compat of start/end/freq.
+    """
+    is_ts_compat = lambda x: isinstance(x, (Timestamp, BaseOffset))
+    is_td_compat = lambda x: isinstance(x, (Timedelta, BaseOffset))
+    return (
+        (is_number(a) and is_number(b))
+        or (is_ts_compat(a) and is_ts_compat(b))
+        or (is_td_compat(a) and is_td_compat(b))
+        or com.any_none(a, b)
+    )
+
+
+def interval_range(
+    start=None,
+    end=None,
+    periods=None,
+    freq=None,
+    name: Hashable | None = None,
+    closed: IntervalClosedType = "right",
+) -> IntervalIndex:
+    """
+    Return a fixed frequency IntervalIndex.
+
+    Parameters
+    ----------
+    start : numeric or datetime-like, default None
+        Left bound for generating intervals.
+    end : numeric or datetime-like, default None
+        Right bound for generating intervals.
+    periods : int, default None
+        Number of periods to generate.
+    freq : numeric, str, Timedelta, datetime.timedelta, or DateOffset, default None
+        The length of each interval. Must be consistent with the type of start
+        and end, e.g. 2 for numeric, or '5H' for datetime-like.  Default is 1
+        for numeric and 'D' for datetime-like.
+    name : str, default None
+        Name of the resulting IntervalIndex.
+    closed : {'left', 'right', 'both', 'neither'}, default 'right'
+        Whether the intervals are closed on the left-side, right-side, both
+        or neither.
+
+    Returns
+    -------
+    IntervalIndex
+        Object with a fixed frequency.
+
+    See Also
+    --------
+    IntervalIndex : An Index of intervals that are all closed on the same side.
+
+    Notes
+    -----
+    Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
+    exactly three must be specified. If ``freq`` is omitted, the resulting
+    ``IntervalIndex`` will have ``periods`` linearly spaced elements between
+    ``start`` and ``end``, inclusively.
+
+    To learn more about datetime-like frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    Numeric ``start`` and  ``end`` is supported.
+
+    >>> pd.interval_range(start=0, end=5)
+    IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]],
+                  dtype='interval[int64, right]')
+
+    Additionally, datetime-like input is also supported.
+
+    >>> pd.interval_range(
+    ...     start=pd.Timestamp("2017-01-01"), end=pd.Timestamp("2017-01-04")
+    ... )
+    IntervalIndex([(2017-01-01 00:00:00, 2017-01-02 00:00:00],
+                   (2017-01-02 00:00:00, 2017-01-03 00:00:00],
+                   (2017-01-03 00:00:00, 2017-01-04 00:00:00]],
+                  dtype='interval[datetime64[ns], right]')
+
+    The ``freq`` parameter specifies the frequency between the left and right.
+    endpoints of the individual intervals within the ``IntervalIndex``.  For
+    numeric ``start`` and ``end``, the frequency must also be numeric.
+
+    >>> pd.interval_range(start=0, periods=4, freq=1.5)
+    IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
+                  dtype='interval[float64, right]')
+
+    Similarly, for datetime-like ``start`` and ``end``, the frequency must be
+    convertible to a DateOffset.
+
+    >>> pd.interval_range(start=pd.Timestamp("2017-01-01"), periods=3, freq="MS")
+    IntervalIndex([(2017-01-01 00:00:00, 2017-02-01 00:00:00],
+                   (2017-02-01 00:00:00, 2017-03-01 00:00:00],
+                   (2017-03-01 00:00:00, 2017-04-01 00:00:00]],
+                  dtype='interval[datetime64[ns], right]')
+
+    Specify ``start``, ``end``, and ``periods``; the frequency is generated
+    automatically (linearly spaced).
+
+    >>> pd.interval_range(start=0, end=6, periods=4)
+    IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
+              dtype='interval[float64, right]')
+
+    The ``closed`` parameter specifies which endpoints of the individual
+    intervals within the ``IntervalIndex`` are closed.
+
+    >>> pd.interval_range(end=5, periods=4, closed="both")
+    IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]],
+                  dtype='interval[int64, both]')
+    """
+    start = maybe_box_datetimelike(start)
+    end = maybe_box_datetimelike(end)
+    endpoint = start if start is not None else end
+
+    if freq is None and com.any_none(periods, start, end):
+        freq = 1 if is_number(endpoint) else "D"
+
+    if com.count_not_none(start, end, periods, freq) != 3:
+        raise ValueError(
+            "Of the four parameters: start, end, periods, and "
+            "freq, exactly three must be specified"
+        )
+
+    if not _is_valid_endpoint(start):
+        raise ValueError(f"start must be numeric or datetime-like, got {start}")
+    if not _is_valid_endpoint(end):
+        raise ValueError(f"end must be numeric or datetime-like, got {end}")
+
+    periods = validate_periods(periods)
+
+    if freq is not None and not is_number(freq):
+        try:
+            freq = to_offset(freq)
+        except ValueError as err:
+            raise ValueError(
+                f"freq must be numeric or convertible to DateOffset, got {freq}"
+            ) from err
+
+    # verify type compatibility
+    if not all(
+        [
+            _is_type_compatible(start, end),
+            _is_type_compatible(start, freq),
+            _is_type_compatible(end, freq),
+        ]
+    ):
+        raise TypeError("start, end, freq need to be type compatible")
+
+    # +1 to convert interval count to breaks count (n breaks = n-1 intervals)
+    if periods is not None:
+        periods += 1
+
+    breaks: np.ndarray | TimedeltaIndex | DatetimeIndex
+
+    if is_number(endpoint):
+        dtype: np.dtype = np.dtype("int64")
+        if com.all_not_none(start, end, freq):
+            if (
+                isinstance(start, (float, np.float16))
+                or isinstance(end, (float, np.float16))
+                or isinstance(freq, (float, np.float16))
+            ):
+                dtype = np.dtype("float64")
+            elif (
+                isinstance(start, (np.integer, np.floating))
+                and isinstance(end, (np.integer, np.floating))
+                and start.dtype == end.dtype
+            ):
+                dtype = start.dtype
+            # 0.1 ensures we capture end
+            breaks = np.arange(start, end + (freq * 0.1), freq)
+            breaks = maybe_downcast_numeric(breaks, dtype)
+        else:
+            # compute the period/start/end if unspecified (at most one)
+            if periods is None:
+                periods = int((end - start) // freq) + 1
+            elif start is None:
+                start = end - (periods - 1) * freq
+            elif end is None:
+                end = start + (periods - 1) * freq
+
+            breaks = np.linspace(start, end, periods)
+        if all(is_integer(x) for x in com.not_none(start, end, freq)):
+            # np.linspace always produces float output
+
+            # error: Argument 1 to "maybe_downcast_numeric" has incompatible type
+            # "Union[ndarray[Any, Any], TimedeltaIndex, DatetimeIndex]";
+            # expected "ndarray[Any, Any]"  [
+            breaks = maybe_downcast_numeric(
+                breaks,  # type: ignore[arg-type]
+                dtype,
+            )
+    else:
+        # delegate to the appropriate range function
+        if isinstance(endpoint, Timestamp):
+            breaks = date_range(start=start, end=end, periods=periods, freq=freq)
+        else:
+            breaks = timedelta_range(start=start, end=end, periods=periods, freq=freq)
+
+    return IntervalIndex.from_breaks(
+        breaks,
+        name=name,
+        closed=closed,
+        dtype=IntervalDtype(subtype=breaks.dtype, closed=closed),
+    )
diff --git a/pandas/pandas/core/indexes/multi.py b/pandas/pandas/core/indexes/multi.py
new file mode 100644
index 0000000000000..79eb1b693d866
--- /dev/null
+++ b/pandas/pandas/core/indexes/multi.py
@@ -0,0 +1,4338 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Collection,
+    Generator,
+    Hashable,
+    Iterable,
+    Sequence,
+)
+from functools import wraps
+from sys import getsizeof
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs import (
+    algos as libalgos,
+    index as libindex,
+    lib,
+)
+from pandas._libs.hashtable import duplicated
+from pandas._typing import (
+    AnyAll,
+    AnyArrayLike,
+    Axis,
+    DropKeep,
+    DtypeObj,
+    F,
+    IgnoreRaise,
+    IndexLabel,
+    IndexT,
+    Scalar,
+    Self,
+    Shape,
+    npt,
+)
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    InvalidIndexError,
+    PerformanceWarning,
+    UnsortedIndexError,
+)
+from pandas.util._decorators import (
+    Appender,
+    cache_readonly,
+    doc,
+    set_module,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import coerce_indexer_dtype
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    ensure_platform_int,
+    is_hashable,
+    is_integer,
+    is_iterator,
+    is_list_like,
+    is_object_dtype,
+    is_scalar,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import is_array_like
+from pandas.core.dtypes.missing import (
+    array_equivalent,
+    isna,
+)
+
+import pandas.core.algorithms as algos
+from pandas.core.array_algos.putmask import validate_putmask
+from pandas.core.arrays import (
+    Categorical,
+    ExtensionArray,
+)
+from pandas.core.arrays.categorical import (
+    factorize_from_iterables,
+    recode_for_categories,
+)
+import pandas.core.common as com
+from pandas.core.construction import sanitize_array
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.base import (
+    Index,
+    _index_shared_docs,
+    ensure_index,
+    get_unanimous_names,
+)
+from pandas.core.indexes.frozen import FrozenList
+from pandas.core.ops.invalid import make_invalid_op
+from pandas.core.sorting import (
+    get_group_index,
+    lexsort_indexer,
+)
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from pandas import (
+        CategoricalIndex,
+        DataFrame,
+        Series,
+    )
+
+_index_doc_kwargs = dict(ibase._index_doc_kwargs)
+_index_doc_kwargs.update(
+    {"klass": "MultiIndex", "target_klass": "MultiIndex or list of tuples"}
+)
+
+
+class MultiIndexUInt64Engine(libindex.BaseMultiIndexCodesEngine, libindex.UInt64Engine):
+    """Manages a MultiIndex by mapping label combinations to positive integers.
+
+    The number of possible label combinations must not overflow the 64 bits integers.
+    """
+
+    _base = libindex.UInt64Engine
+    _codes_dtype = "uint64"
+
+
+class MultiIndexUInt32Engine(libindex.BaseMultiIndexCodesEngine, libindex.UInt32Engine):
+    """Manages a MultiIndex by mapping label combinations to positive integers.
+
+    The number of possible label combinations must not overflow the 32 bits integers.
+    """
+
+    _base = libindex.UInt32Engine
+    _codes_dtype = "uint32"
+
+
+class MultiIndexUInt16Engine(libindex.BaseMultiIndexCodesEngine, libindex.UInt16Engine):
+    """Manages a MultiIndex by mapping label combinations to positive integers.
+
+    The number of possible label combinations must not overflow the 16 bits integers.
+    """
+
+    _base = libindex.UInt16Engine
+    _codes_dtype = "uint16"
+
+
+class MultiIndexUInt8Engine(libindex.BaseMultiIndexCodesEngine, libindex.UInt8Engine):
+    """Manages a MultiIndex by mapping label combinations to positive integers.
+
+    The number of possible label combinations must not overflow the 8 bits integers.
+    """
+
+    _base = libindex.UInt8Engine
+    _codes_dtype = "uint8"
+
+
+class MultiIndexPyIntEngine(libindex.BaseMultiIndexCodesEngine, libindex.ObjectEngine):
+    """Manages a MultiIndex by mapping label combinations to positive integers.
+
+    This class manages those (extreme) cases in which the number of possible
+    label combinations overflows the 64 bits integers, and uses an ObjectEngine
+    containing Python integers.
+    """
+
+    _base = libindex.ObjectEngine
+    _codes_dtype = "object"
+
+
+def names_compat(meth: F) -> F:
+    """
+    A decorator to allow either `name` or `names` keyword but not both.
+
+    This makes it easier to share code with base class.
+    """
+
+    @wraps(meth)
+    def new_meth(self_or_cls, *args, **kwargs):
+        if "name" in kwargs and "names" in kwargs:
+            raise TypeError("Can only provide one of `names` and `name`")
+        if "name" in kwargs:
+            kwargs["names"] = kwargs.pop("name")
+
+        return meth(self_or_cls, *args, **kwargs)
+
+    return cast(F, new_meth)
+
+
+@set_module("pandas")
+class MultiIndex(Index):
+    """
+    A multi-level, or hierarchical, index object for pandas objects.
+
+    Parameters
+    ----------
+    levels : sequence of arrays
+        The unique labels for each level.
+    codes : sequence of arrays
+        Integers for each level designating which label at each location.
+    sortorder : optional int
+        Level of sortedness (must be lexicographically sorted by that
+        level).
+    names : optional sequence of objects
+        Names for each of the index levels. (name is accepted for compat).
+    copy : bool, default False
+        Copy the meta-data.
+    name : Label
+        Kept for compatibility with 1-dimensional Index. Should not be used.
+    verify_integrity : bool, default True
+        Check that the levels/codes are consistent and valid.
+
+    Attributes
+    ----------
+    names
+    levels
+    codes
+    nlevels
+    levshape
+    dtypes
+
+    Methods
+    -------
+    from_arrays
+    from_tuples
+    from_product
+    from_frame
+    set_levels
+    set_codes
+    to_frame
+    to_flat_index
+    sortlevel
+    droplevel
+    swaplevel
+    reorder_levels
+    remove_unused_levels
+    get_level_values
+    get_indexer
+    get_loc
+    get_locs
+    get_loc_level
+    drop
+
+    See Also
+    --------
+    MultiIndex.from_arrays  : Convert list of arrays to MultiIndex.
+    MultiIndex.from_product : Create a MultiIndex from the cartesian product
+                              of iterables.
+    MultiIndex.from_tuples  : Convert list of tuples to a MultiIndex.
+    MultiIndex.from_frame   : Make a MultiIndex from a DataFrame.
+    Index : The base pandas Index type.
+
+    Notes
+    -----
+    See the `user guide
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/advanced.html>`__
+    for more.
+
+    Examples
+    --------
+    A new ``MultiIndex`` is typically constructed using one of the helper
+    methods :meth:`MultiIndex.from_arrays`, :meth:`MultiIndex.from_product`
+    and :meth:`MultiIndex.from_tuples`. For example (using ``.from_arrays``):
+
+    >>> arrays = [[1, 1, 2, 2], ["red", "blue", "red", "blue"]]
+    >>> pd.MultiIndex.from_arrays(arrays, names=("number", "color"))
+    MultiIndex([(1,  'red'),
+                (1, 'blue'),
+                (2,  'red'),
+                (2, 'blue')],
+               names=['number', 'color'])
+
+    See further examples for how to construct a MultiIndex in the doc strings
+    of the mentioned helper methods.
+    """
+
+    _hidden_attrs = Index._hidden_attrs | frozenset()
+
+    # initialize to zero-length tuples to make everything work
+    _typ = "multiindex"
+    _names: list[Hashable | None] = []
+    _levels = FrozenList()
+    _codes = FrozenList()
+    _comparables = ["names"]
+
+    sortorder: int | None
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        levels=None,
+        codes=None,
+        sortorder=None,
+        names=None,
+        copy: bool = False,
+        name=None,
+        verify_integrity: bool = True,
+    ) -> Self:
+        # compat with Index
+        if name is not None:
+            names = name
+        if levels is None or codes is None:
+            raise TypeError("Must pass both levels and codes")
+        if len(levels) != len(codes):
+            raise ValueError("Length of levels and codes must be the same.")
+        if len(levels) == 0:
+            raise ValueError("Must pass non-zero number of levels/codes")
+
+        result = object.__new__(cls)
+        result._cache = {}
+
+        # we've already validated levels and codes, so shortcut here
+        result._set_levels(levels, copy=copy, validate=False)
+        result._set_codes(codes, copy=copy, validate=False)
+
+        result._names = [None] * len(levels)
+        if names is not None:
+            # handles name validation
+            result._set_names(names)
+
+        if sortorder is not None:
+            result.sortorder = int(sortorder)
+        else:
+            result.sortorder = sortorder
+
+        if verify_integrity:
+            new_codes = result._verify_integrity()
+            result._codes = new_codes
+
+        result._reset_identity()
+        result._references = None
+
+        return result
+
+    def _validate_codes(self, level: Index, code: np.ndarray) -> np.ndarray:
+        """
+        Reassign code values as -1 if their corresponding levels are NaN.
+
+        Parameters
+        ----------
+        code : Index
+            Code to reassign.
+        level : np.ndarray
+            Level to check for missing values (NaN, NaT, None).
+
+        Returns
+        -------
+        new code where code value = -1 if it corresponds
+        to a level with missing values (NaN, NaT, None).
+        """
+        null_mask = isna(level)
+        if np.any(null_mask):
+            code = np.where(null_mask[code], -1, code)
+        return code
+
+    def _verify_integrity(
+        self,
+        codes: list | None = None,
+        levels: list | None = None,
+        levels_to_verify: list[int] | range | None = None,
+    ) -> FrozenList:
+        """
+        Parameters
+        ----------
+        codes : optional list
+            Codes to check for validity. Defaults to current codes.
+        levels : optional list
+            Levels to check for validity. Defaults to current levels.
+        levels_to_validate: optional list
+            Specifies the levels to verify.
+
+        Raises
+        ------
+        ValueError
+            If length of levels and codes don't match, if the codes for any
+            level would exceed level bounds, or there are any duplicate levels.
+
+        Returns
+        -------
+        new codes where code value = -1 if it corresponds to a
+        NaN level.
+        """
+        # NOTE: Currently does not check, among other things, that cached
+        # nlevels matches nor that sortorder matches actually sortorder.
+        codes = codes or self.codes
+        levels = levels or self.levels
+        if levels_to_verify is None:
+            levels_to_verify = range(len(levels))
+
+        if len(levels) != len(codes):
+            raise ValueError(
+                "Length of levels and codes must match. NOTE: "
+                "this index is in an inconsistent state."
+            )
+        codes_length = len(codes[0])
+        for i in levels_to_verify:
+            level = levels[i]
+            level_codes = codes[i]
+
+            if len(level_codes) != codes_length:
+                raise ValueError(
+                    f"Unequal code lengths: {[len(code_) for code_ in codes]}"
+                )
+            if len(level_codes) and level_codes.max() >= len(level):
+                raise ValueError(
+                    f"On level {i}, code max ({level_codes.max()}) >= length of "
+                    f"level ({len(level)}). NOTE: this index is in an "
+                    "inconsistent state"
+                )
+            if len(level_codes) and level_codes.min() < -1:
+                raise ValueError(f"On level {i}, code value ({level_codes.min()}) < -1")
+            if not level.is_unique:
+                raise ValueError(
+                    f"Level values must be unique: {list(level)} on level {i}"
+                )
+        if self.sortorder is not None:
+            if self.sortorder > _lexsort_depth(self.codes, self.nlevels):
+                raise ValueError(
+                    "Value for sortorder must be inferior or equal to actual "
+                    f"lexsort_depth: sortorder {self.sortorder} "
+                    f"with lexsort_depth {_lexsort_depth(self.codes, self.nlevels)}"
+                )
+
+        result_codes = []
+        for i in range(len(levels)):
+            if i in levels_to_verify:
+                result_codes.append(self._validate_codes(levels[i], codes[i]))
+            else:
+                result_codes.append(codes[i])
+
+        new_codes = FrozenList(result_codes)
+        return new_codes
+
+    @classmethod
+    def from_arrays(
+        cls,
+        arrays,
+        sortorder: int | None = None,
+        names: Sequence[Hashable] | Hashable | lib.NoDefault = lib.no_default,
+    ) -> MultiIndex:
+        """
+        Convert arrays to MultiIndex.
+
+        Parameters
+        ----------
+        arrays : list / sequence of array-likes
+            Each array-like gives one level's value for each data point.
+            len(arrays) is the number of levels.
+        sortorder : int or None
+            Level of sortedness (must be lexicographically sorted by that
+            level).
+        names : list / sequence of str, optional
+            Names for the levels in the index.
+
+        Returns
+        -------
+        MultiIndex
+
+        See Also
+        --------
+        MultiIndex.from_tuples : Convert list of tuples to MultiIndex.
+        MultiIndex.from_product : Make a MultiIndex from cartesian product
+                                  of iterables.
+        MultiIndex.from_frame : Make a MultiIndex from a DataFrame.
+
+        Examples
+        --------
+        >>> arrays = [[1, 1, 2, 2], ["red", "blue", "red", "blue"]]
+        >>> pd.MultiIndex.from_arrays(arrays, names=("number", "color"))
+        MultiIndex([(1,  'red'),
+                    (1, 'blue'),
+                    (2,  'red'),
+                    (2, 'blue')],
+                   names=['number', 'color'])
+        """
+        error_msg = "Input must be a list / sequence of array-likes."
+        if not is_list_like(arrays):
+            raise TypeError(error_msg)
+        if is_iterator(arrays):
+            arrays = list(arrays)
+
+        # Check if elements of array are list-like
+        for array in arrays:
+            if not is_list_like(array):
+                raise TypeError(error_msg)
+
+        # Check if lengths of all arrays are equal or not,
+        # raise ValueError, if not
+        for i in range(1, len(arrays)):
+            if len(arrays[i]) != len(arrays[i - 1]):
+                raise ValueError("all arrays must be same length")
+
+        codes, levels = factorize_from_iterables(arrays)
+        if names is lib.no_default:
+            names = [getattr(arr, "name", None) for arr in arrays]
+
+        return cls(
+            levels=levels,
+            codes=codes,
+            sortorder=sortorder,
+            names=names,
+            verify_integrity=False,
+        )
+
+    @classmethod
+    @names_compat
+    def from_tuples(
+        cls,
+        tuples: Iterable[tuple[Hashable, ...]],
+        sortorder: int | None = None,
+        names: Sequence[Hashable] | Hashable | None = None,
+    ) -> MultiIndex:
+        """
+        Convert list of tuples to MultiIndex.
+
+        Parameters
+        ----------
+        tuples : list / sequence of tuple-likes
+            Each tuple is the index of one row/column.
+        sortorder : int or None
+            Level of sortedness (must be lexicographically sorted by that
+            level).
+        names : list / sequence of str, optional
+            Names for the levels in the index.
+
+        Returns
+        -------
+        MultiIndex
+
+        See Also
+        --------
+        MultiIndex.from_arrays : Convert list of arrays to MultiIndex.
+        MultiIndex.from_product : Make a MultiIndex from cartesian product
+                                  of iterables.
+        MultiIndex.from_frame : Make a MultiIndex from a DataFrame.
+
+        Examples
+        --------
+        >>> tuples = [(1, "red"), (1, "blue"), (2, "red"), (2, "blue")]
+        >>> pd.MultiIndex.from_tuples(tuples, names=("number", "color"))
+        MultiIndex([(1,  'red'),
+                    (1, 'blue'),
+                    (2,  'red'),
+                    (2, 'blue')],
+                   names=['number', 'color'])
+        """
+        if not is_list_like(tuples):
+            raise TypeError("Input must be a list / sequence of tuple-likes.")
+        if is_iterator(tuples):
+            tuples = list(tuples)
+        tuples = cast(Collection[tuple[Hashable, ...]], tuples)
+
+        # handling the empty tuple cases
+        if len(tuples) and all(isinstance(e, tuple) and not e for e in tuples):
+            codes = [np.zeros(len(tuples))]
+            levels = [Index(com.asarray_tuplesafe(tuples, dtype=np.dtype("object")))]
+            return cls(
+                levels=levels,
+                codes=codes,
+                sortorder=sortorder,
+                names=names,
+                verify_integrity=False,
+            )
+
+        arrays: list[Sequence[Hashable]]
+        if len(tuples) == 0:
+            if names is None:
+                raise TypeError("Cannot infer number of levels from empty list")
+            # error: Argument 1 to "len" has incompatible type "Hashable";
+            # expected "Sized"
+            arrays = [[]] * len(names)  # type: ignore[arg-type]
+        elif isinstance(tuples, (np.ndarray, Index)):
+            if isinstance(tuples, Index):
+                tuples = np.asarray(tuples._values)
+
+            arrays = list(lib.tuples_to_object_array(tuples).T)
+        elif isinstance(tuples, list):
+            arrays = list(lib.to_object_array_tuples(tuples).T)
+        else:
+            arrs = zip(*tuples)
+            arrays = cast(list[Sequence[Hashable]], arrs)
+
+        return cls.from_arrays(arrays, sortorder=sortorder, names=names)
+
+    @classmethod
+    def from_product(
+        cls,
+        iterables: Sequence[Iterable[Hashable]],
+        sortorder: int | None = None,
+        names: Sequence[Hashable] | Hashable | lib.NoDefault = lib.no_default,
+    ) -> MultiIndex:
+        """
+        Make a MultiIndex from the cartesian product of multiple iterables.
+
+        Parameters
+        ----------
+        iterables : list / sequence of iterables
+            Each iterable has unique labels for each level of the index.
+        sortorder : int or None
+            Level of sortedness (must be lexicographically sorted by that
+            level).
+        names : list / sequence of str, optional
+            Names for the levels in the index.
+            If not explicitly provided, names will be inferred from the
+            elements of iterables if an element has a name attribute.
+
+        Returns
+        -------
+        MultiIndex
+
+        See Also
+        --------
+        MultiIndex.from_arrays : Convert list of arrays to MultiIndex.
+        MultiIndex.from_tuples : Convert list of tuples to MultiIndex.
+        MultiIndex.from_frame : Make a MultiIndex from a DataFrame.
+
+        Examples
+        --------
+        >>> numbers = [0, 1, 2]
+        >>> colors = ["green", "purple"]
+        >>> pd.MultiIndex.from_product([numbers, colors], names=["number", "color"])
+        MultiIndex([(0,  'green'),
+                    (0, 'purple'),
+                    (1,  'green'),
+                    (1, 'purple'),
+                    (2,  'green'),
+                    (2, 'purple')],
+                   names=['number', 'color'])
+        """
+
+        if not is_list_like(iterables):
+            raise TypeError("Input must be a list / sequence of iterables.")
+        if is_iterator(iterables):
+            iterables = list(iterables)
+
+        codes, levels = factorize_from_iterables(iterables)
+        if names is lib.no_default:
+            names = [getattr(it, "name", None) for it in iterables]
+
+        # codes are all ndarrays, so cartesian_product is lossless
+        codes = cartesian_product(codes)
+        return cls(levels, codes, sortorder=sortorder, names=names)
+
+    @classmethod
+    def from_frame(
+        cls,
+        df: DataFrame,
+        sortorder: int | None = None,
+        names: Sequence[Hashable] | Hashable | None = None,
+    ) -> MultiIndex:
+        """
+        Make a MultiIndex from a DataFrame.
+
+        Parameters
+        ----------
+        df : DataFrame
+            DataFrame to be converted to MultiIndex.
+        sortorder : int, optional
+            Level of sortedness (must be lexicographically sorted by that
+            level).
+        names : list-like, optional
+            If no names are provided, use the column names, or tuple of column
+            names if the columns is a MultiIndex. If a sequence, overwrite
+            names with the given sequence.
+
+        Returns
+        -------
+        MultiIndex
+            The MultiIndex representation of the given DataFrame.
+
+        See Also
+        --------
+        MultiIndex.from_arrays : Convert list of arrays to MultiIndex.
+        MultiIndex.from_tuples : Convert list of tuples to MultiIndex.
+        MultiIndex.from_product : Make a MultiIndex from cartesian product
+                                  of iterables.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [["HI", "Temp"], ["HI", "Precip"], ["NJ", "Temp"], ["NJ", "Precip"]],
+        ...     columns=["a", "b"],
+        ... )
+        >>> df
+              a       b
+        0    HI    Temp
+        1    HI  Precip
+        2    NJ    Temp
+        3    NJ  Precip
+
+        >>> pd.MultiIndex.from_frame(df)
+        MultiIndex([('HI',   'Temp'),
+                    ('HI', 'Precip'),
+                    ('NJ',   'Temp'),
+                    ('NJ', 'Precip')],
+                   names=['a', 'b'])
+
+        Using explicit names, instead of the column names
+
+        >>> pd.MultiIndex.from_frame(df, names=["state", "observation"])
+        MultiIndex([('HI',   'Temp'),
+                    ('HI', 'Precip'),
+                    ('NJ',   'Temp'),
+                    ('NJ', 'Precip')],
+                   names=['state', 'observation'])
+        """
+        if not isinstance(df, ABCDataFrame):
+            raise TypeError("Input must be a DataFrame")
+
+        column_names, columns = zip(*df.items())
+        names = column_names if names is None else names
+        return cls.from_arrays(columns, sortorder=sortorder, names=names)
+
+    # --------------------------------------------------------------------
+
+    @cache_readonly
+    def _values(self) -> np.ndarray:
+        # We override here, since our parent uses _data, which we don't use.
+        values = []
+
+        for i in range(self.nlevels):
+            index = self.levels[i]
+            codes = self.codes[i]
+
+            vals = index
+            if isinstance(vals.dtype, CategoricalDtype):
+                vals = cast("CategoricalIndex", vals)
+                vals = vals._data._internal_get_values()
+
+            if isinstance(vals.dtype, ExtensionDtype) or lib.is_np_dtype(
+                vals.dtype, "mM"
+            ):
+                vals = vals.astype(object)
+
+            array_vals = np.asarray(vals)
+            array_vals = algos.take_nd(array_vals, codes, fill_value=index._na_value)
+            values.append(array_vals)
+
+        arr = lib.fast_zip(values)
+        return arr
+
+    @property
+    def values(self) -> np.ndarray:
+        return self._values
+
+    @property
+    def array(self):
+        """
+        Raises a ValueError for `MultiIndex` because there's no single
+        array backing a MultiIndex.
+
+        Raises
+        ------
+        ValueError
+        """
+        raise ValueError(
+            "MultiIndex has no single backing array. Use "
+            "'MultiIndex.to_numpy()' to get a NumPy array of tuples."
+        )
+
+    @cache_readonly
+    def dtypes(self) -> Series:
+        """
+        Return the dtypes as a Series for the underlying MultiIndex.
+
+        See Also
+        --------
+        Index.dtype : Return the dtype object of the underlying data.
+        Series.dtypes : Return the data type of the underlying Series.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_product(
+        ...     [(0, 1, 2), ("green", "purple")], names=["number", "color"]
+        ... )
+        >>> idx
+        MultiIndex([(0,  'green'),
+                    (0, 'purple'),
+                    (1,  'green'),
+                    (1, 'purple'),
+                    (2,  'green'),
+                    (2, 'purple')],
+                   names=['number', 'color'])
+        >>> idx.dtypes
+        number     int64
+        color     object
+        dtype: object
+        """
+        from pandas import Series
+
+        names = com.fill_missing_names(self.names)
+        return Series([level.dtype for level in self.levels], index=Index(names))
+
+    def __len__(self) -> int:
+        return len(self.codes[0])
+
+    @property
+    def size(self) -> int:
+        """
+        Return the number of elements in the underlying data.
+        """
+        # override Index.size to avoid materializing _values
+        return len(self)
+
+    # --------------------------------------------------------------------
+    # Levels Methods
+
+    @cache_readonly
+    def levels(self) -> FrozenList:
+        """
+        Levels of the MultiIndex.
+
+        Levels refer to the different hierarchical levels or layers in a MultiIndex.
+        In a MultiIndex, each level represents a distinct dimension or category of
+        the index.
+
+        To access the levels, you can use the levels attribute of the MultiIndex,
+        which returns a tuple of Index objects. Each Index object represents a
+        level in the MultiIndex and contains the unique values found in that
+        specific level.
+
+        If a MultiIndex is created with levels A, B, C, and the DataFrame using
+        it filters out all rows of the level C, MultiIndex.levels will still
+        return A, B, C.
+
+        See Also
+        --------
+        MultiIndex.codes : The codes of the levels in the MultiIndex.
+        MultiIndex.get_level_values : Return vector of label values for requested
+            level.
+
+        Examples
+        --------
+        >>> index = pd.MultiIndex.from_product(
+        ...     [["mammal"], ("goat", "human", "cat", "dog")],
+        ...     names=["Category", "Animals"],
+        ... )
+        >>> leg_num = pd.DataFrame(data=(4, 2, 4, 4), index=index, columns=["Legs"])
+        >>> leg_num
+                          Legs
+        Category Animals
+        mammal   goat        4
+                 human       2
+                 cat         4
+                 dog         4
+
+        >>> leg_num.index.levels
+        FrozenList([['mammal'], ['cat', 'dog', 'goat', 'human']])
+
+        MultiIndex levels will not change even if the DataFrame using the MultiIndex
+        does not contain all them anymore.
+        See how "human" is not in the DataFrame, but it is still in levels:
+
+        >>> large_leg_num = leg_num[leg_num.Legs > 2]
+        >>> large_leg_num
+                          Legs
+        Category Animals
+        mammal   goat        4
+                 cat         4
+                 dog         4
+
+        >>> large_leg_num.index.levels
+        FrozenList([['mammal'], ['cat', 'dog', 'goat', 'human']])
+        """
+        # Use cache_readonly to ensure that self.get_locs doesn't repeatedly
+        # create new IndexEngine
+        # https://github.com/pandas-dev/pandas/issues/31648
+        result = [x._rename(name=name) for x, name in zip(self._levels, self._names)]
+        for level in result:
+            # disallow midx.levels[0].name = "foo"
+            level._no_setting_name = True
+        return FrozenList(result)
+
+    def _set_levels(
+        self,
+        levels,
+        *,
+        level=None,
+        copy: bool = False,
+        validate: bool = True,
+        verify_integrity: bool = False,
+    ) -> None:
+        # This is NOT part of the levels property because it should be
+        # externally not allowed to set levels. User beware if you change
+        # _levels directly
+        if validate:
+            if len(levels) == 0:
+                raise ValueError("Must set non-zero number of levels.")
+            if level is None and len(levels) != self.nlevels:
+                raise ValueError("Length of levels must match number of levels.")
+            if level is not None and len(levels) != len(level):
+                raise ValueError("Length of levels must match length of level.")
+
+        if level is None:
+            new_levels = FrozenList(
+                ensure_index(lev, copy=copy)._view() for lev in levels
+            )
+            level_numbers: range | list[int] = range(len(new_levels))
+        else:
+            level_numbers = [self._get_level_number(lev) for lev in level]
+            new_levels_list = list(self._levels)
+            for lev_num, lev in zip(level_numbers, levels):
+                new_levels_list[lev_num] = ensure_index(lev, copy=copy)._view()
+            new_levels = FrozenList(new_levels_list)
+
+        if verify_integrity:
+            new_codes = self._verify_integrity(
+                levels=new_levels, levels_to_verify=level_numbers
+            )
+            self._codes = new_codes
+
+        names = self.names
+        self._levels = new_levels
+        if any(names):
+            self._set_names(names)
+
+        self._reset_cache()
+
+    def set_levels(
+        self, levels, *, level=None, verify_integrity: bool = True
+    ) -> MultiIndex:
+        """
+        Set new levels on MultiIndex. Defaults to returning new index.
+
+        The `set_levels` method provides a flexible way to change the levels of a
+        `MultiIndex`. This is particularly useful when you need to update the
+        index structure of your DataFrame without altering the data. The method
+        returns a new `MultiIndex` unless the operation is performed in-place,
+        ensuring that the original index remains unchanged unless explicitly
+        modified.
+
+        The method checks the integrity of the new levels against the existing
+        codes by default, but this can be disabled if you are confident that
+        your levels are consistent with the underlying data. This can be useful
+        when you want to perform optimizations or make specific adjustments to
+        the index levels that do not strictly adhere to the original structure.
+
+        Parameters
+        ----------
+        levels : sequence or list of sequence
+            New level(s) to apply.
+        level : int, level name, or sequence of int/level names (default None)
+            Level(s) to set (None for all levels).
+        verify_integrity : bool, default True
+            If True, checks that levels and codes are compatible.
+
+        Returns
+        -------
+        MultiIndex
+            A new `MultiIndex` with the updated levels.
+
+        See Also
+        --------
+        MultiIndex.set_codes : Set new codes on the existing `MultiIndex`.
+        MultiIndex.remove_unused_levels : Create new MultiIndex from current that
+            removes unused levels.
+        Index.set_names : Set Index or MultiIndex name.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_tuples(
+        ...     [
+        ...         (1, "one"),
+        ...         (1, "two"),
+        ...         (2, "one"),
+        ...         (2, "two"),
+        ...         (3, "one"),
+        ...         (3, "two"),
+        ...     ],
+        ...     names=["foo", "bar"],
+        ... )
+        >>> idx
+        MultiIndex([(1, 'one'),
+            (1, 'two'),
+            (2, 'one'),
+            (2, 'two'),
+            (3, 'one'),
+            (3, 'two')],
+           names=['foo', 'bar'])
+
+        >>> idx.set_levels([["a", "b", "c"], [1, 2]])
+        MultiIndex([('a', 1),
+                    ('a', 2),
+                    ('b', 1),
+                    ('b', 2),
+                    ('c', 1),
+                    ('c', 2)],
+                   names=['foo', 'bar'])
+        >>> idx.set_levels(["a", "b", "c"], level=0)
+        MultiIndex([('a', 'one'),
+                    ('a', 'two'),
+                    ('b', 'one'),
+                    ('b', 'two'),
+                    ('c', 'one'),
+                    ('c', 'two')],
+                   names=['foo', 'bar'])
+        >>> idx.set_levels(["a", "b"], level="bar")
+        MultiIndex([(1, 'a'),
+                    (1, 'b'),
+                    (2, 'a'),
+                    (2, 'b'),
+                    (3, 'a'),
+                    (3, 'b')],
+                   names=['foo', 'bar'])
+
+        If any of the levels passed to ``set_levels()`` exceeds the
+        existing length, all of the values from that argument will
+        be stored in the MultiIndex levels, though the values will
+        be truncated in the MultiIndex output.
+
+        >>> idx.set_levels([["a", "b", "c"], [1, 2, 3, 4]], level=[0, 1])
+        MultiIndex([('a', 1),
+            ('a', 2),
+            ('b', 1),
+            ('b', 2),
+            ('c', 1),
+            ('c', 2)],
+           names=['foo', 'bar'])
+        >>> idx.set_levels([["a", "b", "c"], [1, 2, 3, 4]], level=[0, 1]).levels
+        FrozenList([['a', 'b', 'c'], [1, 2, 3, 4]])
+        """
+
+        if isinstance(levels, Index):
+            pass
+        elif is_array_like(levels):
+            levels = Index(levels)
+        elif is_list_like(levels):
+            levels = list(levels)
+
+        level, levels = _require_listlike(level, levels, "Levels")
+        idx = self._view()
+        idx._reset_identity()
+        idx._set_levels(
+            levels, level=level, validate=True, verify_integrity=verify_integrity
+        )
+        return idx
+
+    @property
+    def nlevels(self) -> int:
+        """
+        Integer number of levels in this MultiIndex.
+
+        See Also
+        --------
+        MultiIndex.levels : Get the levels of the MultiIndex.
+        MultiIndex.codes : Get the codes of the MultiIndex.
+        MultiIndex.from_arrays : Convert arrays to MultiIndex.
+        MultiIndex.from_tuples : Convert list of tuples to MultiIndex.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a"], ["b"], ["c"]])
+        >>> mi
+        MultiIndex([('a', 'b', 'c')],
+                   )
+        >>> mi.nlevels
+        3
+        """
+        return len(self._levels)
+
+    @property
+    def levshape(self) -> Shape:
+        """
+        A tuple representing the length of each level in the MultiIndex.
+
+        In a `MultiIndex`, each level can contain multiple unique values. The
+        `levshape` property provides a quick way to assess the size of each
+        level by returning a tuple where each entry represents the number of
+        unique values in that specific level. This is particularly useful in
+        scenarios where you need to understand the structure and distribution
+        of your index levels, such as when working with multidimensional data.
+
+        See Also
+        --------
+        MultiIndex.shape : Return a tuple of the shape of the MultiIndex.
+        MultiIndex.levels : Returns the levels of the MultiIndex.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a"], ["b"], ["c"]])
+        >>> mi
+        MultiIndex([('a', 'b', 'c')],
+                   )
+        >>> mi.levshape
+        (1, 1, 1)
+        """
+        return tuple(len(x) for x in self.levels)
+
+    # --------------------------------------------------------------------
+    # Codes Methods
+
+    @property
+    def codes(self) -> FrozenList:
+        """
+        Codes of the MultiIndex.
+
+        Codes are the position of the index value in the list of level values
+        for each level.
+
+        Returns
+        -------
+        tuple of numpy.ndarray
+            The codes of the MultiIndex. Each array in the tuple corresponds
+            to a level in the MultiIndex.
+
+        See Also
+        --------
+        MultiIndex.set_codes : Set new codes on MultiIndex.
+
+        Examples
+        --------
+        >>> arrays = [[1, 1, 2, 2], ["red", "blue", "red", "blue"]]
+        >>> mi = pd.MultiIndex.from_arrays(arrays, names=("number", "color"))
+        >>> mi.codes
+        FrozenList([[0, 0, 1, 1], [1, 0, 1, 0]])
+        """
+        return self._codes
+
+    def _set_codes(
+        self,
+        codes,
+        *,
+        level=None,
+        copy: bool = False,
+        validate: bool = True,
+        verify_integrity: bool = False,
+    ) -> None:
+        if validate:
+            if level is None and len(codes) != self.nlevels:
+                raise ValueError("Length of codes must match number of levels")
+            if level is not None and len(codes) != len(level):
+                raise ValueError("Length of codes must match length of levels.")
+
+        level_numbers: list[int] | range
+        if level is None:
+            new_codes = FrozenList(
+                _coerce_indexer_frozen(level_codes, lev, copy=copy).view()
+                for lev, level_codes in zip(self._levels, codes)
+            )
+            level_numbers = range(len(new_codes))
+        else:
+            level_numbers = [self._get_level_number(lev) for lev in level]
+            new_codes_list = list(self._codes)
+            for lev_num, level_codes in zip(level_numbers, codes):
+                lev = self.levels[lev_num]
+                new_codes_list[lev_num] = _coerce_indexer_frozen(
+                    level_codes, lev, copy=copy
+                )
+            new_codes = FrozenList(new_codes_list)
+
+        if verify_integrity:
+            new_codes = self._verify_integrity(
+                codes=new_codes, levels_to_verify=level_numbers
+            )
+
+        self._codes = new_codes
+
+        self._reset_cache()
+
+    def set_codes(
+        self, codes, *, level=None, verify_integrity: bool = True
+    ) -> MultiIndex:
+        """
+        Set new codes on MultiIndex. Defaults to returning new index.
+
+        Parameters
+        ----------
+        codes : sequence or list of sequence
+            New codes to apply.
+        level : int, level name, or sequence of int/level names (default None)
+            Level(s) to set (None for all levels).
+        verify_integrity : bool, default True
+            If True, checks that levels and codes are compatible.
+
+        Returns
+        -------
+        new index (of same type and class...etc) or None
+            The same type as the caller or None if ``inplace=True``.
+
+        See Also
+        --------
+        MultiIndex.set_levels : Set new levels on MultiIndex.
+        MultiIndex.codes : Get the codes of the levels in the MultiIndex.
+        MultiIndex.levels : Get the levels of the MultiIndex.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_tuples(
+        ...     [(1, "one"), (1, "two"), (2, "one"), (2, "two")], names=["foo", "bar"]
+        ... )
+        >>> idx
+        MultiIndex([(1, 'one'),
+            (1, 'two'),
+            (2, 'one'),
+            (2, 'two')],
+           names=['foo', 'bar'])
+
+        >>> idx.set_codes([[1, 0, 1, 0], [0, 0, 1, 1]])
+        MultiIndex([(2, 'one'),
+                    (1, 'one'),
+                    (2, 'two'),
+                    (1, 'two')],
+                   names=['foo', 'bar'])
+        >>> idx.set_codes([1, 0, 1, 0], level=0)
+        MultiIndex([(2, 'one'),
+                    (1, 'two'),
+                    (2, 'one'),
+                    (1, 'two')],
+                   names=['foo', 'bar'])
+        >>> idx.set_codes([0, 0, 1, 1], level="bar")
+        MultiIndex([(1, 'one'),
+                    (1, 'one'),
+                    (2, 'two'),
+                    (2, 'two')],
+                   names=['foo', 'bar'])
+        >>> idx.set_codes([[1, 0, 1, 0], [0, 0, 1, 1]], level=[0, 1])
+        MultiIndex([(2, 'one'),
+                    (1, 'one'),
+                    (2, 'two'),
+                    (1, 'two')],
+                   names=['foo', 'bar'])
+        """
+
+        level, codes = _require_listlike(level, codes, "Codes")
+        idx = self._view()
+        idx._reset_identity()
+        idx._set_codes(codes, level=level, verify_integrity=verify_integrity)
+        return idx
+
+    # --------------------------------------------------------------------
+    # Index Internals
+
+    @cache_readonly
+    def _engine(self):
+        # Calculate the number of bits needed to represent labels in each
+        # level, as log2 of their sizes:
+        # NaN values are shifted to 1 and missing values in other while
+        # calculating the indexer are shifted to 0
+        sizes = np.ceil(
+            np.log2(
+                [len(level) + libindex.multiindex_nulls_shift for level in self.levels]
+            )
+        )
+
+        # Sum bit counts, starting from the _right_....
+        lev_bits = np.cumsum(sizes[::-1])[::-1]
+
+        # ... in order to obtain offsets such that sorting the combination of
+        # shifted codes (one for each level, resulting in a unique integer) is
+        # equivalent to sorting lexicographically the codes themselves. Notice
+        # that each level needs to be shifted by the number of bits needed to
+        # represent the _previous_ ones:
+        offsets = np.concatenate([lev_bits[1:], [0]])
+        # Downcast the type if possible, to prevent upcasting when shifting codes:
+        offsets = offsets.astype(np.min_scalar_type(int(offsets[0])))
+
+        # Check the total number of bits needed for our representation:
+        if lev_bits[0] > 64:
+            # The levels would overflow a 64 bit uint - use Python integers:
+            return MultiIndexPyIntEngine(self.levels, self.codes, offsets)
+        if lev_bits[0] > 32:
+            # The levels would overflow a 32 bit uint - use uint64
+            return MultiIndexUInt64Engine(self.levels, self.codes, offsets)
+        if lev_bits[0] > 16:
+            # The levels would overflow a 16 bit uint - use uint8
+            return MultiIndexUInt32Engine(self.levels, self.codes, offsets)
+        if lev_bits[0] > 8:
+            # The levels would overflow a 8 bit uint - use uint16
+            return MultiIndexUInt16Engine(self.levels, self.codes, offsets)
+        # The levels fit in an 8 bit uint - use uint8
+        return MultiIndexUInt8Engine(self.levels, self.codes, offsets)
+
+    # Return type "Callable[..., MultiIndex]" of "_constructor" incompatible with return
+    # type "Type[MultiIndex]" in supertype "Index"
+    @property
+    def _constructor(self) -> Callable[..., MultiIndex]:  # type: ignore[override]
+        return type(self).from_tuples
+
+    @doc(Index._shallow_copy)
+    def _shallow_copy(self, values: np.ndarray, name=lib.no_default) -> MultiIndex:
+        names = name if name is not lib.no_default else self.names
+
+        return type(self).from_tuples(values, sortorder=None, names=names)
+
+    def _view(self) -> MultiIndex:
+        result = type(self)(
+            levels=self.levels,
+            codes=self.codes,
+            sortorder=self.sortorder,
+            names=self.names,
+            verify_integrity=False,
+        )
+        result._cache = self._cache.copy()
+        result._reset_cache("levels")  # GH32669
+        return result
+
+    # --------------------------------------------------------------------
+
+    # error: Signature of "copy" incompatible with supertype "Index"
+    def copy(  # type: ignore[override]
+        self,
+        names=None,
+        deep: bool = False,
+        name=None,
+    ) -> Self:
+        """
+        Make a copy of this object. Names, dtype, levels and codes can be passed and \
+        will be set on new copy.
+
+        The `copy` method provides a mechanism to create a duplicate of an
+        existing MultiIndex object. This is particularly useful in scenarios where
+        modifications are required on an index, but the original MultiIndex should
+        remain unchanged. By specifying the `deep` parameter, users can control
+        whether the copy should be a deep or shallow copy, providing flexibility
+        depending on the size and complexity of the MultiIndex.
+
+        Parameters
+        ----------
+        names : sequence, optional
+            Names to set on the new MultiIndex object.
+        deep : bool, default False
+            If False, the new object will be a shallow copy. If True, a deep copy
+            will be attempted. Deep copying can be potentially expensive for large
+            MultiIndex objects.
+        name : Label
+            Kept for compatibility with 1-dimensional Index. Should not be used.
+
+        Returns
+        -------
+        MultiIndex
+            A new MultiIndex object with the specified modifications.
+
+        See Also
+        --------
+        MultiIndex.from_arrays : Convert arrays to MultiIndex.
+        MultiIndex.from_tuples : Convert list of tuples to MultiIndex.
+        MultiIndex.from_frame : Convert DataFrame to MultiIndex.
+
+        Notes
+        -----
+        In most cases, there should be no functional difference from using
+        ``deep``, but if ``deep`` is passed it will attempt to deepcopy.
+        This could be potentially expensive on large MultiIndex objects.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a"], ["b"], ["c"]])
+        >>> mi
+        MultiIndex([('a', 'b', 'c')],
+                   )
+        >>> mi.copy()
+        MultiIndex([('a', 'b', 'c')],
+                   )
+        """
+        names = self._validate_names(name=name, names=names, deep=deep)
+        keep_id = not deep
+        levels, codes = None, None
+
+        if deep:
+            from copy import deepcopy
+
+            levels = deepcopy(self.levels)
+            codes = deepcopy(self.codes)
+
+        levels = levels if levels is not None else self.levels
+        codes = codes if codes is not None else self.codes
+
+        new_index = type(self)(
+            levels=levels,
+            codes=codes,
+            sortorder=self.sortorder,
+            names=names,
+            verify_integrity=False,
+        )
+        new_index._cache = self._cache.copy()
+        new_index._reset_cache("levels")  # GH32669
+        if keep_id:
+            new_index._id = self._id
+        return new_index
+
+    def __array__(self, dtype=None, copy=None) -> np.ndarray:
+        """the array interface, return my values"""
+        if copy is False:
+            # self.values is always a newly construct array, so raise.
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+        if copy is True:
+            # explicit np.array call to ensure a copy is made and unique objects
+            # are returned, because self.values is cached
+            return np.array(self.values, dtype=dtype)
+        return self.values
+
+    def view(self, cls=None) -> Self:
+        """this is defined as a copy with the same identity"""
+        result = self.copy()
+        result._id = self._id
+        return result
+
+    @doc(Index.__contains__)
+    def __contains__(self, key: Any) -> bool:
+        hash(key)
+        try:
+            self.get_loc(key)
+            return True
+        except (LookupError, TypeError, ValueError):
+            return False
+
+    @cache_readonly
+    def dtype(self) -> np.dtype:
+        return np.dtype("O")
+
+    @cache_readonly
+    def _is_memory_usage_qualified(self) -> bool:
+        """return a boolean if we need a qualified .info display"""
+
+        def f(dtype) -> bool:
+            return is_object_dtype(dtype) or (
+                is_string_dtype(dtype) and dtype.storage == "python"
+            )
+
+        return any(f(level.dtype) for level in self.levels)
+
+    # Cannot determine type of "memory_usage"
+    @doc(Index.memory_usage)  # type: ignore[has-type]
+    def memory_usage(self, deep: bool = False) -> int:
+        # we are overwriting our base class to avoid
+        # computing .values here which could materialize
+        # a tuple representation unnecessarily
+        return self._nbytes(deep)
+
+    @cache_readonly
+    def nbytes(self) -> int:
+        """return the number of bytes in the underlying data"""
+        return self._nbytes(False)
+
+    def _nbytes(self, deep: bool = False) -> int:
+        """
+        return the number of bytes in the underlying data
+        deeply introspect the level data if deep=True
+
+        include the engine hashtable
+
+        *this is in internal routine*
+
+        """
+        # for implementations with no useful getsizeof (PyPy)
+        objsize = 24
+
+        level_nbytes = sum(i.memory_usage(deep=deep) for i in self.levels)
+        label_nbytes = sum(i.nbytes for i in self.codes)
+        names_nbytes = sum(getsizeof(i, objsize) for i in self.names)
+        result = level_nbytes + label_nbytes + names_nbytes
+
+        # include our engine hashtable, only if it's already cached
+        if "_engine" in self._cache:
+            result += self._engine.sizeof(deep=deep)
+        return result
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+
+    def _formatter_func(self, tup):
+        """
+        Formats each item in tup according to its level's formatter function.
+        """
+        formatter_funcs = (level._formatter_func for level in self.levels)
+        return tuple(func(val) for func, val in zip(formatter_funcs, tup))
+
+    def _get_values_for_csv(
+        self, *, na_rep: str = "nan", **kwargs
+    ) -> npt.NDArray[np.object_]:
+        new_levels = []
+        new_codes = []
+
+        # go through the levels and format them
+        for level, level_codes in zip(self.levels, self.codes):
+            level_strs = level._get_values_for_csv(na_rep=na_rep, **kwargs)
+            # add nan values, if there are any
+            mask = level_codes == -1
+            if mask.any():
+                nan_index = len(level_strs)
+                # numpy 1.21 deprecated implicit string casting
+                level_strs = level_strs.astype(str)
+                level_strs = np.append(level_strs, na_rep)
+                assert not level_codes.flags.writeable  # i.e. copy is needed
+                level_codes = level_codes.copy()  # make writeable
+                level_codes[mask] = nan_index
+            new_levels.append(level_strs)
+            new_codes.append(level_codes)
+
+        if len(new_levels) == 1:
+            # a single-level multi-index
+            return Index(new_levels[0].take(new_codes[0]))._get_values_for_csv()
+        else:
+            # reconstruct the multi-index
+            mi = MultiIndex(
+                levels=new_levels,
+                codes=new_codes,
+                names=self.names,
+                sortorder=self.sortorder,
+                verify_integrity=False,
+            )
+            return mi._values
+
+    def _format_multi(
+        self,
+        *,
+        include_names: bool,
+        sparsify: bool | None | lib.NoDefault,
+        formatter: Callable | None = None,
+    ) -> list:
+        if len(self) == 0:
+            return []
+
+        stringified_levels = []
+        for lev, level_codes in zip(self.levels, self.codes):
+            na = _get_na_rep(lev.dtype)
+
+            if len(lev) > 0:
+                taken = formatted = lev.take(level_codes)
+                formatted = taken._format_flat(include_name=False, formatter=formatter)
+
+                # we have some NA
+                mask = level_codes == -1
+                if mask.any():
+                    formatted = np.array(formatted, dtype=object)
+                    formatted[mask] = na
+                    formatted = formatted.tolist()
+
+            else:
+                # weird all NA case
+                formatted = [
+                    pprint_thing(na if isna(x) else x, escape_chars=("\t", "\r", "\n"))
+                    for x in algos.take_nd(lev._values, level_codes)
+                ]
+            stringified_levels.append(formatted)
+
+        result_levels = []
+        for lev, lev_name in zip(stringified_levels, self.names):
+            level = []
+
+            if include_names:
+                level.append(
+                    pprint_thing(lev_name, escape_chars=("\t", "\r", "\n"))
+                    if lev_name is not None
+                    else ""
+                )
+
+            level.extend(np.array(lev, dtype=object))
+            result_levels.append(level)
+
+        if sparsify is None:
+            sparsify = get_option("display.multi_sparse")
+
+        if sparsify:
+            sentinel: Literal[""] | bool | lib.NoDefault = ""
+            # GH3547 use value of sparsify as sentinel if it's "Falsey"
+            assert isinstance(sparsify, bool) or sparsify is lib.no_default
+            if sparsify is lib.no_default:
+                sentinel = sparsify
+            # little bit of a kludge job for #1217
+            result_levels = sparsify_labels(
+                result_levels, start=int(include_names), sentinel=sentinel
+            )
+
+        return result_levels
+
+    # --------------------------------------------------------------------
+    # Names Methods
+
+    def _get_names(self) -> FrozenList:
+        return FrozenList(self._names)
+
+    def _set_names(self, names, *, level=None) -> None:
+        """
+        Set new names on index. Each name has to be a hashable type.
+
+        Parameters
+        ----------
+        values : str or sequence
+            name(s) to set
+        level : int, level name, or sequence of int/level names (default None)
+            If the index is a MultiIndex (hierarchical), level(s) to set (None
+            for all levels).  Otherwise level must be None
+
+        Raises
+        ------
+        TypeError if each name is not hashable.
+
+        Notes
+        -----
+        sets names on levels. WARNING: mutates!
+
+        Note that you generally want to set this *after* changing levels, so
+        that it only acts on copies
+        """
+        # GH 15110
+        # Don't allow a single string for names in a MultiIndex
+        if names is not None and not is_list_like(names):
+            raise ValueError("Names should be list-like for a MultiIndex")
+        names = list(names)
+
+        if level is not None and len(names) != len(level):
+            raise ValueError("Length of names must match length of level.")
+        if level is None and len(names) != self.nlevels:
+            raise ValueError(
+                "Length of names must match number of levels in MultiIndex."
+            )
+
+        if level is None:
+            level = range(self.nlevels)
+        else:
+            level = (self._get_level_number(lev) for lev in level)
+
+        # set the name
+        for lev, name in zip(level, names):
+            if name is not None:
+                # GH 20527
+                # All items in 'names' need to be hashable:
+                if not is_hashable(name):
+                    raise TypeError(
+                        f"{type(self).__name__}.name must be a hashable type"
+                    )
+            self._names[lev] = name
+
+        # If .levels has been accessed, the .name of each level in our cache
+        # will be stale.
+        self._reset_cache("levels")
+
+    names = property(
+        fset=_set_names,
+        fget=_get_names,
+        doc="""
+        Names of levels in MultiIndex.
+
+        This attribute provides access to the names of the levels in a `MultiIndex`.
+        The names are stored as a `FrozenList`, which is an immutable list-like
+        container. Each name corresponds to a level in the `MultiIndex`, and can be
+        used to identify or manipulate the levels individually.
+
+        See Also
+        --------
+        MultiIndex.set_names : Set Index or MultiIndex name.
+        MultiIndex.rename : Rename specific levels in a MultiIndex.
+        Index.names : Get names on index.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays(
+        ...     [[1, 2], [3, 4], [5, 6]], names=['x', 'y', 'z']
+        ... )
+        >>> mi
+        MultiIndex([(1, 3, 5),
+                    (2, 4, 6)],
+                   names=['x', 'y', 'z'])
+        >>> mi.names
+        FrozenList(['x', 'y', 'z'])
+        """,
+    )
+
+    # --------------------------------------------------------------------
+
+    @cache_readonly
+    def inferred_type(self) -> str:
+        return "mixed"
+
+    def _get_level_number(self, level) -> int:
+        count = self.names.count(level)
+        if (count > 1) and not is_integer(level):
+            raise ValueError(
+                f"The name {level} occurs multiple times, use a level number"
+            )
+        try:
+            level = self.names.index(level)
+        except ValueError as err:
+            if not is_integer(level):
+                raise KeyError(f"Level {level} not found") from err
+            if level < 0:
+                level += self.nlevels
+                if level < 0:
+                    orig_level = level - self.nlevels
+                    raise IndexError(
+                        f"Too many levels: Index has only {self.nlevels} levels, "
+                        f"{orig_level} is not a valid level number"
+                    ) from err
+            # Note: levels are zero-based
+            elif level >= self.nlevels:
+                raise IndexError(
+                    f"Too many levels: Index has only {self.nlevels} levels, "
+                    f"not {level + 1}"
+                ) from err
+        return level
+
+    @cache_readonly
+    def is_monotonic_increasing(self) -> bool:
+        """
+        Return a boolean if the values are equal or increasing.
+        """
+        if any(-1 in code for code in self.codes):
+            return False
+
+        if all(level.is_monotonic_increasing for level in self.levels):
+            # If each level is sorted, we can operate on the codes directly. GH27495
+            return libalgos.is_lexsorted(
+                [x.astype("int64", copy=False) for x in self.codes]
+            )
+
+        # reversed() because lexsort() wants the most significant key last.
+        values = [
+            self._get_level_values(i)._values for i in reversed(range(len(self.levels)))
+        ]
+        try:
+            # error: Argument 1 to "lexsort" has incompatible type
+            # "List[Union[ExtensionArray, ndarray[Any, Any]]]";
+            # expected "Union[_SupportsArray[dtype[Any]],
+            # _NestedSequence[_SupportsArray[dtype[Any]]], bool,
+            # int, float, complex, str, bytes, _NestedSequence[Union
+            # [bool, int, float, complex, str, bytes]]]"
+            sort_order = np.lexsort(values)  # type: ignore[arg-type]
+            return Index(sort_order).is_monotonic_increasing
+        except TypeError:
+            # we have mixed types and np.lexsort is not happy
+            return Index(self._values).is_monotonic_increasing
+
+    @cache_readonly
+    def is_monotonic_decreasing(self) -> bool:
+        """
+        Return a boolean if the values are equal or decreasing.
+        """
+        # monotonic decreasing if and only if reverse is monotonic increasing
+        return self[::-1].is_monotonic_increasing
+
+    @doc(Index.duplicated)
+    def duplicated(self, keep: DropKeep = "first") -> npt.NDArray[np.bool_]:
+        shape = tuple(len(lev) for lev in self.levels)
+        ids = get_group_index(self.codes, shape, sort=False, xnull=False)
+
+        return duplicated(ids, keep)
+
+    # error: Cannot override final attribute "_duplicated"
+    # (previously declared in base class "IndexOpsMixin")
+    _duplicated = duplicated  # type: ignore[misc]
+
+    def fillna(self, value):
+        """
+        fillna is not implemented for MultiIndex
+        """
+        raise NotImplementedError("fillna is not defined for MultiIndex")
+
+    @doc(Index.dropna)
+    def dropna(self, how: AnyAll = "any") -> MultiIndex:
+        nans = [level_codes == -1 for level_codes in self.codes]
+        if how == "any":
+            indexer = np.any(nans, axis=0)
+        elif how == "all":
+            indexer = np.all(nans, axis=0)
+        else:
+            raise ValueError(f"invalid how option: {how}")
+
+        new_codes = [level_codes[~indexer] for level_codes in self.codes]
+        return self.set_codes(codes=new_codes)
+
+    def _get_level_values(self, level: int, unique: bool = False) -> Index:
+        """
+        Return vector of label values for requested level,
+        equal to the length of the index
+
+        **this is an internal method**
+
+        Parameters
+        ----------
+        level : int
+        unique : bool, default False
+            if True, drop duplicated values
+
+        Returns
+        -------
+        Index
+        """
+        lev = self.levels[level]
+        level_codes = self.codes[level]
+        name = self._names[level]
+        if unique:
+            level_codes = algos.unique(level_codes)
+        filled = algos.take_nd(lev._values, level_codes, fill_value=lev._na_value)
+        return lev._shallow_copy(filled, name=name)
+
+    # error: Signature of "get_level_values" incompatible with supertype "Index"
+    def get_level_values(self, level) -> Index:  # type: ignore[override]
+        """
+        Return vector of label values for requested level.
+
+        Length of returned vector is equal to the length of the index.
+        The `get_level_values` method is a crucial utility for extracting
+        specific level values from a `MultiIndex`. This function is particularly
+        useful when working with multi-level data, allowing you to isolate
+        and manipulate individual levels without having to deal with the
+        complexity of the entire `MultiIndex` structure. It seamlessly handles
+        both integer and string-based level access, providing flexibility in
+        how you can interact with the data. Additionally, this method ensures
+        that the returned `Index` maintains the integrity of the original data,
+        even when missing values are present, by appropriately casting the
+        result to a suitable data type.
+
+        Parameters
+        ----------
+        level : int or str
+            ``level`` is either the integer position of the level in the
+            MultiIndex, or the name of the level.
+
+        Returns
+        -------
+        Index
+            Values is a level of this MultiIndex converted to
+            a single :class:`Index` (or subclass thereof).
+
+        See Also
+        --------
+        MultiIndex : A multi-level, or hierarchical, index object for pandas objects.
+        Index : Immutable sequence used for indexing and alignment.
+        MultiIndex.remove_unused_levels : Create new MultiIndex from current that
+            removes unused levels.
+
+        Notes
+        -----
+        If the level contains missing values, the result may be casted to
+        ``float`` with missing values specified as ``NaN``. This is because
+        the level is converted to a regular ``Index``.
+
+        Examples
+        --------
+        Create a MultiIndex:
+
+        >>> mi = pd.MultiIndex.from_arrays((list("abc"), list("def")))
+        >>> mi.names = ["level_1", "level_2"]
+
+        Get level values by supplying level as either integer or name:
+
+        >>> mi.get_level_values(0)
+        Index(['a', 'b', 'c'], dtype='object', name='level_1')
+        >>> mi.get_level_values("level_2")
+        Index(['d', 'e', 'f'], dtype='object', name='level_2')
+
+        If a level contains missing values, the return type of the level
+        may be cast to ``float``.
+
+        >>> pd.MultiIndex.from_arrays([[1, None, 2], [3, 4, 5]]).dtypes
+        level_0    int64
+        level_1    int64
+        dtype: object
+        >>> pd.MultiIndex.from_arrays([[1, None, 2], [3, 4, 5]]).get_level_values(0)
+        Index([1.0, nan, 2.0], dtype='float64')
+        """
+        level = self._get_level_number(level)
+        values = self._get_level_values(level)
+        return values
+
+    @doc(Index.unique)
+    def unique(self, level=None):
+        if level is None:
+            return self.drop_duplicates()
+        else:
+            level = self._get_level_number(level)
+            return self._get_level_values(level=level, unique=True)
+
+    def to_frame(
+        self,
+        index: bool = True,
+        name=lib.no_default,
+        allow_duplicates: bool = False,
+    ) -> DataFrame:
+        """
+        Create a DataFrame with the levels of the MultiIndex as columns.
+
+        Column ordering is determined by the DataFrame constructor with data as
+        a dict.
+
+        Parameters
+        ----------
+        index : bool, default True
+            Set the index of the returned DataFrame as the original MultiIndex.
+
+        name : list / sequence of str, optional
+            The passed names should substitute index level names.
+
+        allow_duplicates : bool, optional default False
+            Allow duplicate column labels to be created.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        DataFrame
+            DataFrame representation of the MultiIndex, with levels as columns.
+
+        See Also
+        --------
+        DataFrame : Two-dimensional, size-mutable, potentially heterogeneous
+            tabular data.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a", "b"], ["c", "d"]])
+        >>> mi
+        MultiIndex([('a', 'c'),
+                    ('b', 'd')],
+                   )
+
+        >>> df = mi.to_frame()
+        >>> df
+             0  1
+        a c  a  c
+        b d  b  d
+
+        >>> df = mi.to_frame(index=False)
+        >>> df
+           0  1
+        0  a  c
+        1  b  d
+
+        >>> df = mi.to_frame(name=["x", "y"])
+        >>> df
+             x  y
+        a c  a  c
+        b d  b  d
+        """
+        from pandas import DataFrame
+
+        if name is not lib.no_default:
+            if not is_list_like(name):
+                raise TypeError("'name' must be a list / sequence of column names.")
+
+            if len(name) != len(self.levels):
+                raise ValueError(
+                    "'name' should have same length as number of levels on index."
+                )
+            idx_names = name
+        else:
+            idx_names = self._get_level_names()
+
+        if not allow_duplicates and len(set(idx_names)) != len(idx_names):
+            raise ValueError(
+                "Cannot create duplicate column labels if allow_duplicates is False"
+            )
+
+        # Guarantee resulting column order - PY36+ dict maintains insertion order
+        result = DataFrame(
+            {level: self._get_level_values(level) for level in range(len(self.levels))},
+            copy=False,
+        )
+        result.columns = idx_names
+
+        if index:
+            result.index = self
+        return result
+
+    # error: Return type "Index" of "to_flat_index" incompatible with return type
+    # "MultiIndex" in supertype "Index"
+    def to_flat_index(self) -> Index:  # type: ignore[override]
+        """
+        Convert a MultiIndex to an Index of Tuples containing the level values.
+
+        Returns
+        -------
+        pd.Index
+            Index with the MultiIndex data represented in Tuples.
+
+        See Also
+        --------
+        MultiIndex.from_tuples : Convert flat index back to MultiIndex.
+
+        Notes
+        -----
+        This method will simply return the caller if called by anything other
+        than a MultiIndex.
+
+        Examples
+        --------
+        >>> index = pd.MultiIndex.from_product(
+        ...     [["foo", "bar"], ["baz", "qux"]], names=["a", "b"]
+        ... )
+        >>> index.to_flat_index()
+        Index([('foo', 'baz'), ('foo', 'qux'),
+               ('bar', 'baz'), ('bar', 'qux')],
+              dtype='object')
+        """
+        return Index(self._values, tupleize_cols=False)
+
+    def _is_lexsorted(self) -> bool:
+        """
+        Return True if the codes are lexicographically sorted.
+
+        Returns
+        -------
+        bool
+
+        Examples
+        --------
+        In the below examples, the first level of the MultiIndex is sorted because
+        a<b<c, so there is no need to look at the next level.
+
+        >>> pd.MultiIndex.from_arrays(
+        ...     [["a", "b", "c"], ["d", "e", "f"]]
+        ... )._is_lexsorted()
+        True
+        >>> pd.MultiIndex.from_arrays(
+        ...     [["a", "b", "c"], ["d", "f", "e"]]
+        ... )._is_lexsorted()
+        True
+
+        In case there is a tie, the lexicographical sorting looks
+        at the next level of the MultiIndex.
+
+        >>> pd.MultiIndex.from_arrays([[0, 1, 1], ["a", "b", "c"]])._is_lexsorted()
+        True
+        >>> pd.MultiIndex.from_arrays([[0, 1, 1], ["a", "c", "b"]])._is_lexsorted()
+        False
+        >>> pd.MultiIndex.from_arrays(
+        ...     [["a", "a", "b", "b"], ["aa", "bb", "aa", "bb"]]
+        ... )._is_lexsorted()
+        True
+        >>> pd.MultiIndex.from_arrays(
+        ...     [["a", "a", "b", "b"], ["bb", "aa", "aa", "bb"]]
+        ... )._is_lexsorted()
+        False
+        """
+        return self._lexsort_depth == self.nlevels
+
+    @cache_readonly
+    def _lexsort_depth(self) -> int:
+        """
+        Compute and return the lexsort_depth, the number of levels of the
+        MultiIndex that are sorted lexically
+
+        Returns
+        -------
+        int
+        """
+        if self.sortorder is not None:
+            return self.sortorder
+        return _lexsort_depth(self.codes, self.nlevels)
+
+    def _sort_levels_monotonic(self, raise_if_incomparable: bool = False) -> MultiIndex:
+        """
+        This is an *internal* function.
+
+        Create a new MultiIndex from the current to monotonically sorted
+        items IN the levels. This does not actually make the entire MultiIndex
+        monotonic, JUST the levels.
+
+        The resulting MultiIndex will have the same outward
+        appearance, meaning the same .values and ordering. It will also
+        be .equals() to the original.
+
+        Returns
+        -------
+        MultiIndex
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex(
+        ...     levels=[["a", "b"], ["bb", "aa"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+        ... )
+        >>> mi
+        MultiIndex([('a', 'bb'),
+                    ('a', 'aa'),
+                    ('b', 'bb'),
+                    ('b', 'aa')],
+                   )
+
+        >>> mi.sort_values()
+        MultiIndex([('a', 'aa'),
+                    ('a', 'bb'),
+                    ('b', 'aa'),
+                    ('b', 'bb')],
+                   )
+        """
+        if self._is_lexsorted() and self.is_monotonic_increasing:
+            return self
+
+        new_levels = []
+        new_codes = []
+
+        for lev, level_codes in zip(self.levels, self.codes):
+            if not lev.is_monotonic_increasing:
+                try:
+                    # indexer to reorder the levels
+                    indexer = lev.argsort()
+                except TypeError:
+                    if raise_if_incomparable:
+                        raise
+                else:
+                    lev = lev.take(indexer)
+
+                    # indexer to reorder the level codes
+                    indexer = ensure_platform_int(indexer)
+                    ri = lib.get_reverse_indexer(indexer, len(indexer))
+                    level_codes = algos.take_nd(ri, level_codes, fill_value=-1)
+
+            new_levels.append(lev)
+            new_codes.append(level_codes)
+
+        return MultiIndex(
+            new_levels,
+            new_codes,
+            names=self.names,
+            sortorder=self.sortorder,
+            verify_integrity=False,
+        )
+
+    def remove_unused_levels(self) -> MultiIndex:
+        """
+        Create new MultiIndex from current that removes unused levels.
+
+        Unused level(s) means levels that are not expressed in the
+        labels. The resulting MultiIndex will have the same outward
+        appearance, meaning the same .values and ordering. It will
+        also be .equals() to the original.
+
+        The `remove_unused_levels` method is useful in cases where you have a
+        MultiIndex with hierarchical levels, but some of these levels are no
+        longer needed due to filtering or subsetting operations. By removing
+        the unused levels, the resulting MultiIndex becomes more compact and
+        efficient, which can improve performance in subsequent operations.
+
+        Returns
+        -------
+        MultiIndex
+            A new MultiIndex with unused levels removed.
+
+        See Also
+        --------
+        MultiIndex.droplevel : Remove specified levels from a MultiIndex.
+        MultiIndex.reorder_levels : Rearrange levels of a MultiIndex.
+        MultiIndex.set_levels : Set new levels on a MultiIndex.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_product([range(2), list("ab")])
+        >>> mi
+        MultiIndex([(0, 'a'),
+                    (0, 'b'),
+                    (1, 'a'),
+                    (1, 'b')],
+                   )
+
+        >>> mi[2:]
+        MultiIndex([(1, 'a'),
+                    (1, 'b')],
+                   )
+
+        The 0 from the first level is not represented
+        and can be removed
+
+        >>> mi2 = mi[2:].remove_unused_levels()
+        >>> mi2.levels
+        FrozenList([[1], ['a', 'b']])
+        """
+        new_levels = []
+        new_codes = []
+
+        changed = False
+        for lev, level_codes in zip(self.levels, self.codes):
+            # Since few levels are typically unused, bincount() is more
+            # efficient than unique() - however it only accepts positive values
+            # (and drops order):
+            uniques = np.where(np.bincount(level_codes + 1) > 0)[0] - 1
+            has_na = int(len(uniques) and (uniques[0] == -1))
+
+            if len(uniques) != len(lev) + has_na:
+                if lev.isna().any() and len(uniques) == len(lev):
+                    break
+                # We have unused levels
+                changed = True
+
+                # Recalculate uniques, now preserving order.
+                # Can easily be cythonized by exploiting the already existing
+                # "uniques" and stop parsing "level_codes" when all items
+                # are found:
+                uniques = algos.unique(level_codes)
+                if has_na:
+                    na_idx = np.where(uniques == -1)[0]
+                    # Just ensure that -1 is in first position:
+                    uniques[[0, na_idx[0]]] = uniques[[na_idx[0], 0]]
+
+                # codes get mapped from uniques to 0:len(uniques)
+                # -1 (if present) is mapped to last position
+                code_mapping = np.zeros(len(lev) + has_na)
+                # ... and reassigned value -1:
+                code_mapping[uniques] = np.arange(len(uniques)) - has_na
+
+                level_codes = code_mapping[level_codes]
+
+                # new levels are simple
+                lev = lev.take(uniques[has_na:])
+
+            new_levels.append(lev)
+            new_codes.append(level_codes)
+
+        result = self.view()
+
+        if changed:
+            result._reset_identity()
+            result._set_levels(new_levels, validate=False)
+            result._set_codes(new_codes, validate=False)
+
+        return result
+
+    # --------------------------------------------------------------------
+    # Pickling Methods
+
+    def __reduce__(self):
+        """Necessary for making this object picklable"""
+        d = {
+            "levels": list(self.levels),
+            "codes": list(self.codes),
+            "sortorder": self.sortorder,
+            "names": list(self.names),
+        }
+        return ibase._new_Index, (type(self), d), None
+
+    # --------------------------------------------------------------------
+
+    def __getitem__(self, key):
+        if is_scalar(key):
+            key = com.cast_scalar_indexer(key)
+
+            retval = []
+            for lev, level_codes in zip(self.levels, self.codes):
+                if level_codes[key] == -1:
+                    retval.append(np.nan)
+                else:
+                    retval.append(lev[level_codes[key]])
+
+            return tuple(retval)
+        else:
+            # in general cannot be sure whether the result will be sorted
+            sortorder = None
+            if com.is_bool_indexer(key):
+                key = np.asarray(key, dtype=bool)
+                sortorder = self.sortorder
+            elif isinstance(key, slice):
+                if key.step is None or key.step > 0:
+                    sortorder = self.sortorder
+            elif isinstance(key, Index):
+                key = np.asarray(key)
+
+            new_codes = [level_codes[key] for level_codes in self.codes]
+
+            return MultiIndex(
+                levels=self.levels,
+                codes=new_codes,
+                names=self.names,
+                sortorder=sortorder,
+                verify_integrity=False,
+            )
+
+    def _getitem_slice(self: MultiIndex, slobj: slice) -> MultiIndex:
+        """
+        Fastpath for __getitem__ when we know we have a slice.
+        """
+        sortorder = None
+        if slobj.step is None or slobj.step > 0:
+            sortorder = self.sortorder
+
+        new_codes = [level_codes[slobj] for level_codes in self.codes]
+
+        return type(self)(
+            levels=self.levels,
+            codes=new_codes,
+            names=self._names,
+            sortorder=sortorder,
+            verify_integrity=False,
+        )
+
+    @Appender(_index_shared_docs["take"] % _index_doc_kwargs)
+    def take(
+        self: MultiIndex,
+        indices,
+        axis: Axis = 0,
+        allow_fill: bool = True,
+        fill_value=None,
+        **kwargs,
+    ) -> MultiIndex:
+        nv.validate_take((), kwargs)
+        indices = ensure_platform_int(indices)
+
+        # only fill if we are passing a non-None fill_value
+        allow_fill = self._maybe_disallow_fill(allow_fill, fill_value, indices)
+
+        if indices.ndim == 1 and lib.is_range_indexer(indices, len(self)):
+            return self.copy()
+
+        na_value = -1
+
+        taken = [lab.take(indices) for lab in self.codes]
+        if allow_fill:
+            mask = indices == -1
+            if mask.any():
+                masked = []
+                for new_label in taken:
+                    label_values = new_label
+                    label_values[mask] = na_value
+                    masked.append(np.asarray(label_values))
+                taken = masked
+
+        return MultiIndex(
+            levels=self.levels, codes=taken, names=self.names, verify_integrity=False
+        )
+
+    def append(self, other):
+        """
+        Append a collection of Index options together.
+
+        The `append` method is used to combine multiple `Index` objects into a single
+        `Index`. This is particularly useful when dealing with multi-level indexing
+        (MultiIndex) where you might need to concatenate different levels of indices.
+        The method handles the alignment of the levels and codes of the indices being
+        appended to ensure consistency in the resulting `MultiIndex`.
+
+        Parameters
+        ----------
+        other : Index or list/tuple of indices
+            Index or list/tuple of Index objects to be appended.
+
+        Returns
+        -------
+        Index
+            The combined index.
+
+        See Also
+        --------
+        MultiIndex: A multi-level, or hierarchical, index object for pandas objects.
+        Index.append : Append a collection of Index options together.
+        concat : Concatenate pandas objects along a particular axis.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a"], ["b"]])
+        >>> mi
+        MultiIndex([('a', 'b')],
+                   )
+        >>> mi.append(mi)
+        MultiIndex([('a', 'b'), ('a', 'b')],
+                   )
+        """
+        if not isinstance(other, (list, tuple)):
+            other = [other]
+
+        if all(
+            (isinstance(o, MultiIndex) and o.nlevels >= self.nlevels) for o in other
+        ):
+            codes = []
+            levels = []
+            names = []
+            for i in range(self.nlevels):
+                level_values = self.levels[i]
+                for mi in other:
+                    level_values = level_values.union(mi.levels[i])
+                level_codes = [
+                    recode_for_categories(
+                        mi.codes[i], mi.levels[i], level_values, copy=False
+                    )
+                    for mi in ([self, *other])
+                ]
+                level_name = self.names[i]
+                if any(mi.names[i] != level_name for mi in other):
+                    level_name = None
+                codes.append(np.concatenate(level_codes))
+                levels.append(level_values)
+                names.append(level_name)
+            return MultiIndex(
+                codes=codes, levels=levels, names=names, verify_integrity=False
+            )
+
+        to_concat = (self._values,) + tuple(k._values for k in other)
+        new_tuples = np.concatenate(to_concat)
+
+        # if all(isinstance(x, MultiIndex) for x in other):
+        try:
+            # We only get here if other contains at least one index with tuples,
+            # setting names to None automatically
+            return MultiIndex.from_tuples(new_tuples)
+        except (TypeError, IndexError):
+            return Index(new_tuples)
+
+    def argsort(
+        self, *args, na_position: str = "last", **kwargs
+    ) -> npt.NDArray[np.intp]:
+        target = self._sort_levels_monotonic(raise_if_incomparable=True)
+        keys = [lev.codes for lev in target._get_codes_for_sorting()]
+        return lexsort_indexer(keys, na_position=na_position, codes_given=True)
+
+    @Appender(_index_shared_docs["repeat"] % _index_doc_kwargs)
+    def repeat(self, repeats: int, axis=None) -> MultiIndex:
+        nv.validate_repeat((), {"axis": axis})
+        # error: Incompatible types in assignment (expression has type "ndarray",
+        # variable has type "int")
+        repeats = ensure_platform_int(repeats)  # type: ignore[assignment]
+        return MultiIndex(
+            levels=self.levels,
+            codes=[
+                level_codes.view(np.ndarray).astype(np.intp, copy=False).repeat(repeats)
+                for level_codes in self.codes
+            ],
+            names=self.names,
+            sortorder=self.sortorder,
+            verify_integrity=False,
+        )
+
+    # error: Signature of "drop" incompatible with supertype "Index"
+    def drop(  # type: ignore[override]
+        self,
+        codes,
+        level: Index | np.ndarray | Iterable[Hashable] | None = None,
+        errors: IgnoreRaise = "raise",
+    ) -> MultiIndex:
+        """
+        Make a new :class:`pandas.MultiIndex` with the passed list of codes deleted.
+
+        This method allows for the removal of specified labels from a MultiIndex.
+        The labels to be removed can be provided as a list of tuples if no level
+        is specified, or as a list of labels from a specific level if the level
+        parameter is provided. This can be useful for refining the structure of a
+        MultiIndex to fit specific requirements.
+
+        Parameters
+        ----------
+        codes : array-like
+            Must be a list of tuples when ``level`` is not specified.
+        level : int or level name, default None
+            Level from which the labels will be dropped.
+        errors : str, default 'raise'
+            If 'ignore', suppress error and existing labels are dropped.
+
+        Returns
+        -------
+        MultiIndex
+            A new MultiIndex with the specified labels removed.
+
+        See Also
+        --------
+        MultiIndex.remove_unused_levels : Create new MultiIndex from current that
+            removes unused levels.
+        MultiIndex.reorder_levels : Rearrange levels using input order.
+        MultiIndex.rename : Rename levels in a MultiIndex.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_product(
+        ...     [(0, 1, 2), ("green", "purple")], names=["number", "color"]
+        ... )
+        >>> idx
+        MultiIndex([(0,  'green'),
+                    (0, 'purple'),
+                    (1,  'green'),
+                    (1, 'purple'),
+                    (2,  'green'),
+                    (2, 'purple')],
+                   names=['number', 'color'])
+        >>> idx.drop([(1, "green"), (2, "purple")])
+        MultiIndex([(0,  'green'),
+                    (0, 'purple'),
+                    (1, 'purple'),
+                    (2,  'green')],
+                   names=['number', 'color'])
+
+        We can also drop from a specific level.
+
+        >>> idx.drop("green", level="color")
+        MultiIndex([(0, 'purple'),
+                    (1, 'purple'),
+                    (2, 'purple')],
+                   names=['number', 'color'])
+
+        >>> idx.drop([1, 2], level=0)
+        MultiIndex([(0,  'green'),
+                    (0, 'purple')],
+                   names=['number', 'color'])
+        """
+        if level is not None:
+            return self._drop_from_level(codes, level, errors)
+
+        if not isinstance(codes, (np.ndarray, Index)):
+            try:
+                codes = com.index_labels_to_array(codes, dtype=np.dtype("object"))
+            except ValueError:
+                pass
+
+        inds = []
+        for level_codes in codes:
+            try:
+                loc = self.get_loc(level_codes)
+                # get_loc returns either an integer, a slice, or a boolean
+                # mask
+                if isinstance(loc, int):
+                    inds.append(loc)
+                elif isinstance(loc, slice):
+                    step = loc.step if loc.step is not None else 1
+                    inds.extend(range(loc.start, loc.stop, step))
+                elif com.is_bool_indexer(loc):
+                    if get_option("performance_warnings") and self._lexsort_depth == 0:
+                        warnings.warn(
+                            "dropping on a non-lexsorted multi-index "
+                            "without a level parameter may impact performance.",
+                            PerformanceWarning,
+                            stacklevel=find_stack_level(),
+                        )
+                    loc = loc.nonzero()[0]
+                    inds.extend(loc)
+                else:
+                    msg = f"unsupported indexer of type {type(loc)}"
+                    raise AssertionError(msg)
+            except KeyError:
+                if errors != "ignore":
+                    raise
+
+        return self.delete(inds)
+
+    def _drop_from_level(
+        self, codes, level, errors: IgnoreRaise = "raise"
+    ) -> MultiIndex:
+        codes = com.index_labels_to_array(codes)
+        i = self._get_level_number(level)
+        index = self.levels[i]
+        values = index.get_indexer(codes)
+        # If nan should be dropped it will equal -1 here. We have to check which values
+        # are not nan and equal -1, this means they are missing in the index
+        nan_codes = isna(codes)
+        values[(np.equal(nan_codes, False)) & (values == -1)] = -2
+        if index.shape[0] == self.shape[0]:
+            values[np.equal(nan_codes, True)] = -2
+
+        not_found = codes[values == -2]
+        if len(not_found) != 0 and errors != "ignore":
+            raise KeyError(f"labels {not_found} not found in level")
+        mask = ~algos.isin(self.codes[i], values)
+
+        return self[mask]
+
+    def swaplevel(self, i=-2, j=-1) -> MultiIndex:
+        """
+        Swap level i with level j.
+
+        Calling this method does not change the ordering of the values.
+
+        Parameters
+        ----------
+        i : int, str, default -2
+            First level of index to be swapped. Can pass level name as string.
+            Type of parameters can be mixed.
+        j : int, str, default -1
+            Second level of index to be swapped. Can pass level name as string.
+            Type of parameters can be mixed.
+
+        Returns
+        -------
+        MultiIndex
+            A new MultiIndex.
+
+        See Also
+        --------
+        Series.swaplevel : Swap levels i and j in a MultiIndex.
+        DataFrame.swaplevel : Swap levels i and j in a MultiIndex on a
+            particular axis.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex(
+        ...     levels=[["a", "b"], ["bb", "aa"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+        ... )
+        >>> mi
+        MultiIndex([('a', 'bb'),
+                    ('a', 'aa'),
+                    ('b', 'bb'),
+                    ('b', 'aa')],
+                   )
+        >>> mi.swaplevel(0, 1)
+        MultiIndex([('bb', 'a'),
+                    ('aa', 'a'),
+                    ('bb', 'b'),
+                    ('aa', 'b')],
+                   )
+        """
+        new_levels = list(self.levels)
+        new_codes = list(self.codes)
+        new_names = list(self.names)
+
+        i = self._get_level_number(i)
+        j = self._get_level_number(j)
+
+        new_levels[i], new_levels[j] = new_levels[j], new_levels[i]
+        new_codes[i], new_codes[j] = new_codes[j], new_codes[i]
+        new_names[i], new_names[j] = new_names[j], new_names[i]
+
+        return MultiIndex(
+            levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+        )
+
+    def reorder_levels(self, order) -> MultiIndex:
+        """
+        Rearrange levels using input order. May not drop or duplicate levels.
+
+        `reorder_levels` is useful when you need to change the order of levels in
+        a MultiIndex, such as when reordering levels for hierarchical indexing. It
+        maintains the integrity of the MultiIndex, ensuring that all existing levels
+        are present and no levels are duplicated. This method is helpful for aligning
+        the index structure with other data structures or for optimizing the order
+        for specific data operations.
+
+        Parameters
+        ----------
+        order : list of int or list of str
+            List representing new level order. Reference level by number
+            (position) or by key (label).
+
+        Returns
+        -------
+        MultiIndex
+            A new MultiIndex with levels rearranged according to the specified order.
+
+        See Also
+        --------
+        MultiIndex.swaplevel : Swap two levels of the MultiIndex.
+        MultiIndex.set_names : Set names for the MultiIndex levels.
+        DataFrame.reorder_levels : Reorder levels in a DataFrame with a MultiIndex.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([[1, 2], [3, 4]], names=["x", "y"])
+        >>> mi
+        MultiIndex([(1, 3),
+                    (2, 4)],
+                   names=['x', 'y'])
+
+        >>> mi.reorder_levels(order=[1, 0])
+        MultiIndex([(3, 1),
+                    (4, 2)],
+                   names=['y', 'x'])
+
+        >>> mi.reorder_levels(order=["y", "x"])
+        MultiIndex([(3, 1),
+                    (4, 2)],
+                   names=['y', 'x'])
+        """
+        order = [self._get_level_number(i) for i in order]
+        result = self._reorder_ilevels(order)
+        return result
+
+    def _reorder_ilevels(self, order) -> MultiIndex:
+        if len(order) != self.nlevels:
+            raise AssertionError(
+                f"Length of order must be same as number of levels ({self.nlevels}), "
+                f"got {len(order)}"
+            )
+        new_levels = [self.levels[i] for i in order]
+        new_codes = [self.codes[i] for i in order]
+        new_names = [self.names[i] for i in order]
+
+        return MultiIndex(
+            levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+        )
+
+    def _recode_for_new_levels(
+        self, new_levels, copy: bool = True
+    ) -> Generator[np.ndarray]:
+        if len(new_levels) > self.nlevels:
+            raise AssertionError(
+                f"Length of new_levels ({len(new_levels)}) "
+                f"must be <= self.nlevels ({self.nlevels})"
+            )
+        for i in range(len(new_levels)):
+            yield recode_for_categories(
+                self.codes[i], self.levels[i], new_levels[i], copy=copy
+            )
+
+    def _get_codes_for_sorting(self) -> list[Categorical]:
+        """
+        we are categorizing our codes by using the
+        available categories (all, not just observed)
+        excluding any missing ones (-1); this is in preparation
+        for sorting, where we need to disambiguate that -1 is not
+        a valid valid
+        """
+
+        def cats(level_codes: np.ndarray) -> np.ndarray:
+            return np.arange(
+                level_codes.max() + 1 if len(level_codes) else 0,
+                dtype=level_codes.dtype,
+            )
+
+        return [
+            Categorical.from_codes(level_codes, cats(level_codes), True, validate=False)
+            for level_codes in self.codes
+        ]
+
+    def sortlevel(
+        self,
+        level: IndexLabel = 0,
+        ascending: bool | list[bool] = True,
+        sort_remaining: bool = True,
+        na_position: str = "first",
+    ) -> tuple[MultiIndex, npt.NDArray[np.intp]]:
+        """
+        Sort MultiIndex at the requested level.
+
+        This method is useful when dealing with MultiIndex objects, allowing for
+        sorting at a specific level of the index. The function preserves the
+        relative ordering of data within the same level while sorting
+        the overall MultiIndex. The method provides flexibility with the `ascending`
+        parameter to define the sort order and with the `sort_remaining` parameter to
+        control whether the remaining levels should also be sorted. Sorting a
+        MultiIndex can be crucial when performing operations that require ordered
+        indices, such as grouping or merging datasets. The `na_position` argument is
+        important in handling missing values consistently across different levels.
+
+        Parameters
+        ----------
+        level : list-like, int or str, default 0
+            If a string is given, must be a name of the level.
+            If list-like must be names or ints of levels.
+        ascending : bool, default True
+            False to sort in descending order.
+            Can also be a list to specify a directed ordering.
+        sort_remaining : bool, default True
+            If True, sorts by the remaining levels after sorting by the specified
+            `level`.
+        na_position : {'first' or 'last'}, default 'first'
+            Argument 'first' puts NaNs at the beginning, 'last' puts NaNs at
+            the end.
+
+            .. versionadded:: 2.1.0
+
+        Returns
+        -------
+        sorted_index : pd.MultiIndex
+            Resulting index.
+        indexer : np.ndarray[np.intp]
+            Indices of output values in original index.
+
+        See Also
+        --------
+        MultiIndex : A multi-level, or hierarchical, index object for pandas objects.
+        Index.sort_values : Sort Index values.
+        DataFrame.sort_index : Sort DataFrame by the index.
+        Series.sort_index : Sort Series by the index.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([[0, 0], [2, 1]])
+        >>> mi
+        MultiIndex([(0, 2),
+                    (0, 1)],
+                   )
+
+        >>> mi.sortlevel()
+        (MultiIndex([(0, 1),
+                    (0, 2)],
+                   ), array([1, 0]))
+
+        >>> mi.sortlevel(sort_remaining=False)
+        (MultiIndex([(0, 2),
+                    (0, 1)],
+                   ), array([0, 1]))
+
+        >>> mi.sortlevel(1)
+        (MultiIndex([(0, 1),
+                    (0, 2)],
+                   ), array([1, 0]))
+
+        >>> mi.sortlevel(1, ascending=False)
+        (MultiIndex([(0, 2),
+                    (0, 1)],
+                   ), array([0, 1]))
+        """
+        if not is_list_like(level):
+            level = [level]
+        # error: Item "Hashable" of "Union[Hashable, Sequence[Hashable]]" has
+        # no attribute "__iter__" (not iterable)
+        level = [
+            self._get_level_number(lev)
+            for lev in level  # type: ignore[union-attr]
+        ]
+        sortorder = None
+
+        codes = [self.codes[lev] for lev in level]
+        # we have a directed ordering via ascending
+        if isinstance(ascending, list):
+            if not len(level) == len(ascending):
+                raise ValueError("level must have same length as ascending")
+        elif sort_remaining:
+            codes.extend(
+                [self.codes[lev] for lev in range(len(self.levels)) if lev not in level]
+            )
+        else:
+            sortorder = level[0]
+
+        indexer = lexsort_indexer(
+            codes, orders=ascending, na_position=na_position, codes_given=True
+        )
+
+        indexer = ensure_platform_int(indexer)
+        new_codes = [level_codes.take(indexer) for level_codes in self.codes]
+
+        new_index = MultiIndex(
+            codes=new_codes,
+            levels=self.levels,
+            names=self.names,
+            sortorder=sortorder,
+            verify_integrity=False,
+        )
+
+        return new_index, indexer
+
+    def _wrap_reindex_result(self, target, indexer, preserve_names: bool):
+        if not isinstance(target, MultiIndex):
+            if indexer is None:
+                target = self
+            elif (indexer >= 0).all():
+                target = self.take(indexer)
+            else:
+                try:
+                    target = MultiIndex.from_tuples(target)
+                except TypeError:
+                    # not all tuples, see test_constructor_dict_multiindex_reindex_flat
+                    return target
+
+        target = self._maybe_preserve_names(target, preserve_names)
+        return target
+
+    def _maybe_preserve_names(self, target: IndexT, preserve_names: bool) -> IndexT:
+        if (
+            preserve_names
+            and target.nlevels == self.nlevels
+            and target.names != self.names
+        ):
+            target = target.copy(deep=False)
+            target.names = self.names
+        return target
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    def _check_indexing_error(self, key) -> None:
+        if not is_hashable(key) or is_iterator(key):
+            # We allow tuples if they are hashable, whereas other Index
+            #  subclasses require scalar.
+            # We have to explicitly exclude generators, as these are hashable.
+            raise InvalidIndexError(key)
+
+    @cache_readonly
+    def _should_fallback_to_positional(self) -> bool:
+        """
+        Should integer key(s) be treated as positional?
+        """
+        # GH#33355
+        return self.levels[0]._should_fallback_to_positional
+
+    def _get_indexer_strict(
+        self, key, axis_name: str
+    ) -> tuple[Index, npt.NDArray[np.intp]]:
+        keyarr = key
+        if not isinstance(keyarr, Index):
+            keyarr = com.asarray_tuplesafe(keyarr)
+
+        if len(keyarr) and not isinstance(keyarr[0], tuple):
+            indexer = self._get_indexer_level_0(keyarr)
+
+            self._raise_if_missing(key, indexer, axis_name)
+            return self[indexer], indexer
+
+        return super()._get_indexer_strict(key, axis_name)
+
+    def _raise_if_missing(self, key, indexer, axis_name: str) -> None:
+        keyarr = key
+        if not isinstance(key, Index):
+            keyarr = com.asarray_tuplesafe(key)
+
+        if len(keyarr) and not isinstance(keyarr[0], tuple):
+            # i.e. same condition for special case in MultiIndex._get_indexer_strict
+
+            mask = indexer == -1
+            if mask.any():
+                check = self.levels[0].get_indexer(keyarr)
+                cmask = check == -1
+                if cmask.any():
+                    raise KeyError(f"{keyarr[cmask]} not in index")
+                # We get here when levels still contain values which are not
+                # actually in Index anymore
+                raise KeyError(f"{keyarr} not in index")
+        else:
+            return super()._raise_if_missing(key, indexer, axis_name)
+
+    def _get_indexer_level_0(self, target) -> npt.NDArray[np.intp]:
+        """
+        Optimized equivalent to `self.get_level_values(0).get_indexer_for(target)`.
+        """
+        lev = self.levels[0]
+        codes = self._codes[0]
+        cat = Categorical.from_codes(codes=codes, categories=lev, validate=False)
+        ci = Index(cat)
+        return ci.get_indexer_for(target)
+
+    def get_slice_bound(
+        self,
+        label: Hashable | Sequence[Hashable],
+        side: Literal["left", "right"],
+    ) -> int:
+        """
+        For an ordered MultiIndex, compute slice bound
+        that corresponds to given label.
+
+        Returns leftmost (one-past-the-rightmost if `side=='right') position
+        of given label.
+
+        Parameters
+        ----------
+        label : object or tuple of objects
+        side : {'left', 'right'}
+
+        Returns
+        -------
+        int
+            Index of label.
+
+        Notes
+        -----
+        This method only works if level 0 index of the MultiIndex is lexsorted.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([list("abbc"), list("gefd")])
+
+        Get the locations from the leftmost 'b' in the first level
+        until the end of the multiindex:
+
+        >>> mi.get_slice_bound("b", side="left")
+        1
+
+        Like above, but if you get the locations from the rightmost
+        'b' in the first level and 'f' in the second level:
+
+        >>> mi.get_slice_bound(("b", "f"), side="right")
+        3
+
+        See Also
+        --------
+        MultiIndex.get_loc : Get location for a label or a tuple of labels.
+        MultiIndex.get_locs : Get location for a label/slice/list/mask or a
+                              sequence of such.
+        """
+        if not isinstance(label, tuple):
+            label = (label,)
+        return self._partial_tup_index(label, side=side)
+
+    def slice_locs(self, start=None, end=None, step=None) -> tuple[int, int]:
+        """
+        For an ordered MultiIndex, compute the slice locations for input
+        labels.
+
+        The input labels can be tuples representing partial levels, e.g. for a
+        MultiIndex with 3 levels, you can pass a single value (corresponding to
+        the first level), or a 1-, 2-, or 3-tuple.
+
+        Parameters
+        ----------
+        start : label or tuple, default None
+            If None, defaults to the beginning
+        end : label or tuple
+            If None, defaults to the end
+        step : int or None
+            Slice step
+
+        Returns
+        -------
+        (start, end) : (int, int)
+
+        Notes
+        -----
+        This method only works if the MultiIndex is properly lexsorted. So,
+        if only the first 2 levels of a 3-level MultiIndex are lexsorted,
+        you can only pass two levels to ``.slice_locs``.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays(
+        ...     [list("abbd"), list("deff")], names=["A", "B"]
+        ... )
+
+        Get the slice locations from the beginning of 'b' in the first level
+        until the end of the multiindex:
+
+        >>> mi.slice_locs(start="b")
+        (1, 4)
+
+        Like above, but stop at the end of 'b' in the first level and 'f' in
+        the second level:
+
+        >>> mi.slice_locs(start="b", end=("b", "f"))
+        (1, 3)
+
+        See Also
+        --------
+        MultiIndex.get_loc : Get location for a label or a tuple of labels.
+        MultiIndex.get_locs : Get location for a label/slice/list/mask or a
+                              sequence of such.
+        """
+        # This function adds nothing to its parent implementation (the magic
+        # happens in get_slice_bound method), but it adds meaningful doc.
+        return super().slice_locs(start, end, step)
+
+    def _partial_tup_index(self, tup: tuple, side: Literal["left", "right"] = "left"):
+        if len(tup) > self._lexsort_depth:
+            raise UnsortedIndexError(
+                f"Key length ({len(tup)}) was greater than MultiIndex lexsort depth "
+                f"({self._lexsort_depth})"
+            )
+
+        n = len(tup)
+        start, end = 0, len(self)
+        zipped = zip(tup, self.levels, self.codes)
+        for k, (lab, lev, level_codes) in enumerate(zipped):
+            section = level_codes[start:end]
+
+            loc: npt.NDArray[np.intp] | np.intp | int
+            if lab not in lev and not isna(lab):
+                # short circuit
+                try:
+                    loc = algos.searchsorted(lev, lab, side=side)
+                except TypeError as err:
+                    # non-comparable e.g. test_slice_locs_with_type_mismatch
+                    raise TypeError(f"Level type mismatch: {lab}") from err
+                if not is_integer(loc):
+                    # non-comparable level, e.g. test_groupby_example
+                    raise TypeError(f"Level type mismatch: {lab}")
+                if side == "right" and loc >= 0:
+                    loc -= 1
+                return start + algos.searchsorted(section, loc, side=side)
+
+            idx = self._get_loc_single_level_index(lev, lab)
+            if isinstance(idx, slice) and k < n - 1:
+                # Get start and end value from slice, necessary when a non-integer
+                # interval is given as input GH#37707
+                start = idx.start
+                end = idx.stop
+            elif k < n - 1:
+                # error: Incompatible types in assignment (expression has type
+                # "Union[ndarray[Any, dtype[signedinteger[Any]]]
+                end = start + algos.searchsorted(  # type: ignore[assignment]
+                    section, idx, side="right"
+                )
+                # error: Incompatible types in assignment (expression has type
+                # "Union[ndarray[Any, dtype[signedinteger[Any]]]
+                start = start + algos.searchsorted(  # type: ignore[assignment]
+                    section, idx, side="left"
+                )
+            elif isinstance(idx, slice):
+                idx = idx.start
+                return start + algos.searchsorted(section, idx, side=side)
+            else:
+                return start + algos.searchsorted(section, idx, side=side)
+
+    def _get_loc_single_level_index(self, level_index: Index, key: Hashable) -> int:
+        """
+        If key is NA value, location of index unify as -1.
+
+        Parameters
+        ----------
+        level_index: Index
+        key : label
+
+        Returns
+        -------
+        loc : int
+            If key is NA value, loc is -1
+            Else, location of key in index.
+
+        See Also
+        --------
+        Index.get_loc : The get_loc method for (single-level) index.
+        """
+        if is_scalar(key) and isna(key):
+            # TODO: need is_valid_na_for_dtype(key, level_index.dtype)
+            return -1
+        else:
+            return level_index.get_loc(key)
+
+    def get_loc(self, key):
+        """
+        Get location for a label or a tuple of labels. The location is returned \
+        as an integer/slice or boolean mask.
+
+        This method returns the integer location, slice object, or boolean mask
+        corresponding to the specified key, which can be a single label or a tuple
+        of labels. The key represents a position in the MultiIndex, and the location
+        indicates where the key is found within the index.
+
+        Parameters
+        ----------
+        key : label or tuple of labels (one for each level)
+            A label or tuple of labels that correspond to the levels of the MultiIndex.
+            The key must match the structure of the MultiIndex.
+
+        Returns
+        -------
+        int, slice object or boolean mask
+            If the key is past the lexsort depth, the return may be a
+            boolean mask array, otherwise it is always a slice or int.
+
+        See Also
+        --------
+        Index.get_loc : The get_loc method for (single-level) index.
+        MultiIndex.slice_locs : Get slice location given start label(s) and
+                                end label(s).
+        MultiIndex.get_locs : Get location for a label/slice/list/mask or a
+                              sequence of such.
+
+        Notes
+        -----
+        The key cannot be a slice, list of same-level labels, a boolean mask,
+        or a sequence of such. If you want to use those, use
+        :meth:`MultiIndex.get_locs` instead.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([list("abb"), list("def")])
+
+        >>> mi.get_loc("b")
+        slice(1, 3, None)
+
+        >>> mi.get_loc(("b", "e"))
+        1
+        """
+        self._check_indexing_error(key)
+
+        def _maybe_to_slice(loc):
+            """convert integer indexer to boolean mask or slice if possible"""
+            if not isinstance(loc, np.ndarray) or loc.dtype != np.intp:
+                return loc
+
+            loc = lib.maybe_indices_to_slice(loc, len(self))
+            if isinstance(loc, slice):
+                return loc
+
+            mask = np.empty(len(self), dtype="bool")
+            mask.fill(False)
+            mask[loc] = True
+            return mask
+
+        if not isinstance(key, tuple):
+            loc = self._get_level_indexer(key, level=0)
+            return _maybe_to_slice(loc)
+
+        keylen = len(key)
+        if self.nlevels < keylen:
+            raise KeyError(
+                f"Key length ({keylen}) exceeds index depth ({self.nlevels})"
+            )
+
+        if keylen == self.nlevels and self.is_unique:
+            # TODO: what if we have an IntervalIndex level?
+            #  i.e. do we need _index_as_unique on that level?
+            try:
+                return self._engine.get_loc(key)
+            except KeyError as err:
+                raise KeyError(key) from err
+            except TypeError:
+                # e.g. test_partial_slicing_with_multiindex partial string slicing
+                loc, _ = self.get_loc_level(key, range(self.nlevels))
+                return loc
+
+        # -- partial selection or non-unique index
+        # break the key into 2 parts based on the lexsort_depth of the index;
+        # the first part returns a continuous slice of the index; the 2nd part
+        # needs linear search within the slice
+        i = self._lexsort_depth
+        lead_key, follow_key = key[:i], key[i:]
+
+        if not lead_key:
+            start = 0
+            stop = len(self)
+        else:
+            try:
+                start, stop = self.slice_locs(lead_key, lead_key)
+            except TypeError as err:
+                # e.g. test_groupby_example key = ((0, 0, 1, 2), "new_col")
+                #  when self has 5 integer levels
+                raise KeyError(key) from err
+
+        if start == stop:
+            raise KeyError(key)
+
+        if not follow_key:
+            return slice(start, stop)
+
+        if get_option("performance_warnings"):
+            warnings.warn(
+                "indexing past lexsort depth may impact performance.",
+                PerformanceWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        loc = np.arange(start, stop, dtype=np.intp)
+
+        for i, k in enumerate(follow_key, len(lead_key)):
+            mask = self.codes[i][loc] == self._get_loc_single_level_index(
+                self.levels[i], k
+            )
+            if not mask.all():
+                loc = loc[mask]
+            if not len(loc):
+                raise KeyError(key)
+
+        return _maybe_to_slice(loc) if len(loc) != stop - start else slice(start, stop)
+
+    def get_loc_level(self, key, level: IndexLabel = 0, drop_level: bool = True):
+        """
+        Get location and sliced index for requested label(s)/level(s).
+
+        The `get_loc_level` method is a more advanced form of `get_loc`, allowing
+        users to specify not just a label or sequence of labels, but also the level(s)
+        in which to search. This method is useful when you need to isolate particular
+        sections of a MultiIndex, either for further analysis or for slicing and
+        dicing the data. The method provides flexibility in terms of maintaining
+        or dropping levels from the resulting index based on the `drop_level`
+        parameter.
+
+        Parameters
+        ----------
+        key : label or sequence of labels
+            The label(s) for which to get the location.
+        level : int/level name or list thereof, optional
+            The level(s) in the MultiIndex to consider. If not provided, defaults
+            to the first level.
+        drop_level : bool, default True
+            If ``False``, the resulting index will not drop any level.
+
+        Returns
+        -------
+        tuple
+            A 2-tuple where the elements :
+
+            Element 0: int, slice object or boolean array.
+
+            Element 1: The resulting sliced multiindex/index. If the key
+            contains all levels, this will be ``None``.
+
+        See Also
+        --------
+        MultiIndex.get_loc  : Get location for a label or a tuple of labels.
+        MultiIndex.get_locs : Get location for a label/slice/list/mask or a
+                              sequence of such.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([list("abb"), list("def")], names=["A", "B"])
+
+        >>> mi.get_loc_level("b")
+        (slice(1, 3, None), Index(['e', 'f'], dtype='object', name='B'))
+
+        >>> mi.get_loc_level("e", level="B")
+        (array([False,  True, False]), Index(['b'], dtype='object', name='A'))
+
+        >>> mi.get_loc_level(["b", "e"])
+        (1, None)
+        """
+        if not isinstance(level, (range, list, tuple)):
+            level = self._get_level_number(level)
+        else:
+            level = [self._get_level_number(lev) for lev in level]
+
+        loc, mi = self._get_loc_level(key, level=level)
+        if not drop_level:
+            if lib.is_integer(loc):
+                # Slice index must be an integer or None
+                mi = self[loc : loc + 1]
+            else:
+                mi = self[loc]
+        return loc, mi
+
+    def _get_loc_level(self, key, level: int | list[int] = 0):
+        """
+        get_loc_level but with `level` known to be positional, not name-based.
+        """
+
+        # different name to distinguish from maybe_droplevels
+        def maybe_mi_droplevels(indexer, levels):
+            """
+            If level does not exist or all levels were dropped, the exception
+            has to be handled outside.
+            """
+            new_index = self[indexer]
+
+            for i in sorted(levels, reverse=True):
+                new_index = new_index._drop_level_numbers([i])
+
+            return new_index
+
+        if isinstance(level, (tuple, list)):
+            if len(key) != len(level):
+                raise AssertionError(
+                    "Key for location must have same length as number of levels"
+                )
+            result = None
+            for lev, k in zip(level, key):
+                loc, new_index = self._get_loc_level(k, level=lev)
+                if isinstance(loc, slice):
+                    mask = np.zeros(len(self), dtype=bool)
+                    mask[loc] = True
+                    loc = mask
+                result = loc if result is None else result & loc
+
+            try:
+                # FIXME: we should be only dropping levels on which we are
+                #  scalar-indexing
+                mi = maybe_mi_droplevels(result, level)
+            except ValueError:
+                # droplevel failed because we tried to drop all levels,
+                #  i.e. len(level) == self.nlevels
+                mi = self[result]
+
+            return result, mi
+
+        # kludge for #1796
+        if isinstance(key, list):
+            key = tuple(key)
+
+        if isinstance(key, tuple) and level == 0:
+            try:
+                # Check if this tuple is a single key in our first level
+                if key in self.levels[0]:
+                    indexer = self._get_level_indexer(key, level=level)
+                    new_index = maybe_mi_droplevels(indexer, [0])
+                    return indexer, new_index
+            except (TypeError, InvalidIndexError):
+                pass
+
+            if not any(isinstance(k, slice) for k in key):
+                if len(key) == self.nlevels and self.is_unique:
+                    # Complete key in unique index -> standard get_loc
+                    try:
+                        return (self._engine.get_loc(key), None)
+                    except KeyError as err:
+                        raise KeyError(key) from err
+                    except TypeError:
+                        # e.g. partial string indexing
+                        #  test_partial_string_timestamp_multiindex
+                        pass
+
+                # partial selection
+                indexer = self.get_loc(key)
+                ilevels = [i for i in range(len(key)) if key[i] != slice(None, None)]
+                if len(ilevels) == self.nlevels:
+                    if is_integer(indexer):
+                        # we are dropping all levels
+                        return indexer, None
+
+                    # TODO: in some cases we still need to drop some levels,
+                    #  e.g. test_multiindex_perf_warn
+                    # test_partial_string_timestamp_multiindex
+                    ilevels = [
+                        i
+                        for i in range(len(key))
+                        if (
+                            not isinstance(key[i], str)
+                            or not self.levels[i]._supports_partial_string_indexing
+                        )
+                        and key[i] != slice(None, None)
+                    ]
+                    if len(ilevels) == self.nlevels:
+                        # TODO: why?
+                        ilevels = []
+                return indexer, maybe_mi_droplevels(indexer, ilevels)
+
+            else:
+                indexer = None
+                for i, k in enumerate(key):
+                    if not isinstance(k, slice):
+                        loc_level = self._get_level_indexer(k, level=i)
+                        if isinstance(loc_level, slice):
+                            if com.is_null_slice(loc_level) or com.is_full_slice(
+                                loc_level, len(self)
+                            ):
+                                # everything
+                                continue
+
+                            # e.g. test_xs_IndexSlice_argument_not_implemented
+                            k_index = np.zeros(len(self), dtype=bool)
+                            k_index[loc_level] = True
+
+                        else:
+                            k_index = loc_level
+
+                    elif com.is_null_slice(k):
+                        # taking everything, does not affect `indexer` below
+                        continue
+
+                    else:
+                        # FIXME: this message can be inaccurate, e.g.
+                        #  test_series_varied_multiindex_alignment
+                        raise TypeError(f"Expected label or tuple of labels, got {key}")
+
+                    if indexer is None:
+                        indexer = k_index
+                    else:
+                        indexer &= k_index
+                if indexer is None:
+                    indexer = slice(None, None)
+                ilevels = [i for i in range(len(key)) if key[i] != slice(None, None)]
+                return indexer, maybe_mi_droplevels(indexer, ilevels)
+        else:
+            indexer = self._get_level_indexer(key, level=level)
+            if (
+                isinstance(key, str)
+                and self.levels[level]._supports_partial_string_indexing
+            ):
+                # check to see if we did an exact lookup vs sliced
+                check = self.levels[level].get_loc(key)
+                if not is_integer(check):
+                    # e.g. test_partial_string_timestamp_multiindex
+                    return indexer, self[indexer]
+
+            try:
+                result_index = maybe_mi_droplevels(indexer, [level])
+            except ValueError:
+                result_index = self[indexer]
+
+            return indexer, result_index
+
+    def _get_level_indexer(
+        self, key, level: int = 0, indexer: npt.NDArray[np.bool_] | None = None
+    ):
+        # `level` kwarg is _always_ positional, never name
+        # return a boolean array or slice showing where the key is
+        # in the totality of values
+        # if the indexer is provided, then use this
+
+        level_index = self.levels[level]
+        level_codes = self.codes[level]
+
+        def convert_indexer(start, stop, step, indexer=indexer, codes=level_codes):
+            # Compute a bool indexer to identify the positions to take.
+            # If we have an existing indexer, we only need to examine the
+            # subset of positions where the existing indexer is True.
+            if indexer is not None:
+                # we only need to look at the subset of codes where the
+                # existing indexer equals True
+                codes = codes[indexer]
+
+            if step is None or step == 1:
+                new_indexer = (codes >= start) & (codes < stop)
+            else:
+                r = np.arange(start, stop, step, dtype=codes.dtype)
+                new_indexer = algos.isin(codes, r)
+
+            if indexer is None:
+                return new_indexer
+
+            indexer = indexer.copy()
+            indexer[indexer] = new_indexer
+            return indexer
+
+        if isinstance(key, slice):
+            # handle a slice, returning a slice if we can
+            # otherwise a boolean indexer
+            step = key.step
+            is_negative_step = step is not None and step < 0
+
+            try:
+                if key.start is not None:
+                    start = level_index.get_loc(key.start)
+                elif is_negative_step:
+                    start = len(level_index) - 1
+                else:
+                    start = 0
+
+                if key.stop is not None:
+                    stop = level_index.get_loc(key.stop)
+                elif is_negative_step:
+                    stop = 0
+                elif isinstance(start, slice):
+                    stop = len(level_index)
+                else:
+                    stop = len(level_index) - 1
+            except KeyError:
+                # we have a partial slice (like looking up a partial date
+                # string)
+                start = stop = level_index.slice_indexer(key.start, key.stop, key.step)
+                step = start.step
+
+            if isinstance(start, slice) or isinstance(stop, slice):
+                # we have a slice for start and/or stop
+                # a partial date slicer on a DatetimeIndex generates a slice
+                # note that the stop ALREADY includes the stopped point (if
+                # it was a string sliced)
+                start = getattr(start, "start", start)
+                stop = getattr(stop, "stop", stop)
+                return convert_indexer(start, stop, step)
+
+            elif level > 0 or self._lexsort_depth == 0 or step is not None:
+                # need to have like semantics here to right
+                # searching as when we are using a slice
+                # so adjust the stop by 1 (so we include stop)
+                stop = (stop - 1) if is_negative_step else (stop + 1)
+                return convert_indexer(start, stop, step)
+            else:
+                # sorted, so can return slice object -> view
+                i = algos.searchsorted(level_codes, start, side="left")
+                j = algos.searchsorted(level_codes, stop, side="right")
+                return slice(i, j, step)
+
+        else:
+            idx = self._get_loc_single_level_index(level_index, key)
+
+            if level > 0 or self._lexsort_depth == 0:
+                # Desired level is not sorted
+                if isinstance(idx, slice):
+                    # test_get_loc_partial_timestamp_multiindex
+                    locs = (level_codes >= idx.start) & (level_codes < idx.stop)
+                    return locs
+
+                locs = np.asarray(level_codes == idx, dtype=bool)
+
+                if not locs.any():
+                    # The label is present in self.levels[level] but unused:
+                    raise KeyError(key)
+                return locs
+
+            if isinstance(idx, slice):
+                # e.g. test_partial_string_timestamp_multiindex
+                start = algos.searchsorted(level_codes, idx.start, side="left")
+                # NB: "left" here bc of slice semantics
+                end = algos.searchsorted(level_codes, idx.stop, side="left")
+            else:
+                start = algos.searchsorted(level_codes, idx, side="left")
+                end = algos.searchsorted(level_codes, idx, side="right")
+
+            if start == end:
+                # The label is present in self.levels[level] but unused:
+                raise KeyError(key)
+            return slice(start, end)
+
+    def get_locs(self, seq) -> npt.NDArray[np.intp]:
+        """
+        Get location for a sequence of labels.
+
+        Parameters
+        ----------
+        seq : label, slice, list, mask or a sequence of such
+           You should use one of the above for each level.
+           If a level should not be used, set it to ``slice(None)``.
+
+        Returns
+        -------
+        numpy.ndarray
+            NumPy array of integers suitable for passing to iloc.
+
+        See Also
+        --------
+        MultiIndex.get_loc : Get location for a label or a tuple of labels.
+        MultiIndex.slice_locs : Get slice location given start label(s) and
+                                end label(s).
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([list("abb"), list("def")])
+
+        >>> mi.get_locs("b")  # doctest: +SKIP
+        array([1, 2], dtype=int64)
+
+        >>> mi.get_locs([slice(None), ["e", "f"]])  # doctest: +SKIP
+        array([1, 2], dtype=int64)
+
+        >>> mi.get_locs([[True, False, True], slice("e", "f")])  # doctest: +SKIP
+        array([2], dtype=int64)
+        """
+
+        # must be lexsorted to at least as many levels
+        true_slices = [i for (i, s) in enumerate(com.is_true_slices(seq)) if s]
+        if true_slices and true_slices[-1] >= self._lexsort_depth:
+            raise UnsortedIndexError(
+                "MultiIndex slicing requires the index to be lexsorted: slicing "
+                f"on levels {true_slices}, lexsort depth {self._lexsort_depth}"
+            )
+
+        if any(x is Ellipsis for x in seq):
+            raise NotImplementedError(
+                "MultiIndex does not support indexing with Ellipsis"
+            )
+
+        n = len(self)
+
+        def _to_bool_indexer(indexer) -> npt.NDArray[np.bool_]:
+            if isinstance(indexer, slice):
+                new_indexer = np.zeros(n, dtype=np.bool_)
+                new_indexer[indexer] = True
+                return new_indexer
+            return indexer
+
+        # a bool indexer for the positions we want to take
+        indexer: npt.NDArray[np.bool_] | None = None
+
+        for i, k in enumerate(seq):
+            lvl_indexer: npt.NDArray[np.bool_] | slice | None = None
+
+            if com.is_bool_indexer(k):
+                if len(k) != n:
+                    raise ValueError(
+                        "cannot index with a boolean indexer that "
+                        "is not the same length as the index"
+                    )
+                if isinstance(k, (ABCSeries, Index)):
+                    k = k._values
+                lvl_indexer = np.asarray(k)
+                if indexer is None:
+                    lvl_indexer = lvl_indexer.copy()
+
+            elif is_list_like(k):
+                # a collection of labels to include from this level (these are or'd)
+
+                # GH#27591 check if this is a single tuple key in the level
+                try:
+                    lvl_indexer = self._get_level_indexer(k, level=i, indexer=indexer)
+                except (InvalidIndexError, TypeError, KeyError) as err:
+                    # InvalidIndexError e.g. non-hashable, fall back to treating
+                    #  this as a sequence of labels
+                    # KeyError it can be ambiguous if this is a label or sequence
+                    #  of labels
+                    #  github.com/pandas-dev/pandas/issues/39424#issuecomment-871626708
+                    for x in k:
+                        if not is_hashable(x):
+                            # e.g. slice
+                            raise err
+                        # GH 39424: Ignore not founds
+                        # GH 42351: No longer ignore not founds & enforced in 2.0
+                        # TODO: how to handle IntervalIndex level? (no test cases)
+                        item_indexer = self._get_level_indexer(
+                            x, level=i, indexer=indexer
+                        )
+                        if lvl_indexer is None:
+                            lvl_indexer = _to_bool_indexer(item_indexer)
+                        elif isinstance(item_indexer, slice):
+                            lvl_indexer[item_indexer] = True  # type: ignore[index]
+                        else:
+                            lvl_indexer |= item_indexer
+
+                if lvl_indexer is None:
+                    # no matches we are done
+                    # test_loc_getitem_duplicates_multiindex_empty_indexer
+                    return np.array([], dtype=np.intp)
+
+            elif com.is_null_slice(k):
+                # empty slice
+                if indexer is None and i == len(seq) - 1:
+                    return np.arange(n, dtype=np.intp)
+                continue
+
+            else:
+                # a slice or a single label
+                lvl_indexer = self._get_level_indexer(k, level=i, indexer=indexer)
+
+            # update indexer
+            lvl_indexer = _to_bool_indexer(lvl_indexer)
+            if indexer is None:
+                indexer = lvl_indexer
+            else:
+                indexer &= lvl_indexer
+                if not np.any(indexer) and np.any(lvl_indexer):
+                    raise KeyError(seq)
+
+        # empty indexer
+        if indexer is None:
+            return np.array([], dtype=np.intp)
+
+        pos_indexer = indexer.nonzero()[0]
+        return self._reorder_indexer(seq, pos_indexer)
+
+    # --------------------------------------------------------------------
+
+    def _reorder_indexer(
+        self,
+        seq: tuple[Scalar | Iterable | AnyArrayLike, ...],
+        indexer: npt.NDArray[np.intp],
+    ) -> npt.NDArray[np.intp]:
+        """
+        Reorder an indexer of a MultiIndex (self) so that the labels are in the
+        same order as given in seq
+
+        Parameters
+        ----------
+        seq : label/slice/list/mask or a sequence of such
+        indexer: a position indexer of self
+
+        Returns
+        -------
+        indexer : a sorted position indexer of self ordered as seq
+        """
+
+        # check if sorting is necessary
+        need_sort = False
+        for i, k in enumerate(seq):
+            if com.is_null_slice(k) or com.is_bool_indexer(k) or is_scalar(k):
+                pass
+            elif is_list_like(k):
+                if len(k) <= 1:  # type: ignore[arg-type]
+                    pass
+                elif self._is_lexsorted():
+                    # If the index is lexsorted and the list_like label
+                    # in seq are sorted then we do not need to sort
+                    k_codes = self.levels[i].get_indexer(k)
+                    k_codes = k_codes[k_codes >= 0]  # Filter absent keys
+                    # True if the given codes are not ordered
+                    need_sort = (k_codes[:-1] > k_codes[1:]).any()
+                else:
+                    need_sort = True
+            elif isinstance(k, slice):
+                if self._is_lexsorted():
+                    need_sort = k.step is not None and k.step < 0
+                else:
+                    need_sort = True
+            else:
+                need_sort = True
+            if need_sort:
+                break
+        if not need_sort:
+            return indexer
+
+        n = len(self)
+        keys: tuple[np.ndarray, ...] = ()
+        # For each level of the sequence in seq, map the level codes with the
+        # order they appears in a list-like sequence
+        # This mapping is then use to reorder the indexer
+        for i, k in enumerate(seq):
+            if is_scalar(k):
+                # GH#34603 we want to treat a scalar the same as an all equal list
+                k = [k]
+            if com.is_bool_indexer(k):
+                new_order = np.arange(n)[indexer]
+            elif is_list_like(k):
+                # Generate a map with all level codes as sorted initially
+                if not isinstance(k, (np.ndarray, ExtensionArray, Index, ABCSeries)):
+                    k = sanitize_array(k, None)
+                k = algos.unique(k)
+                key_order_map = np.ones(len(self.levels[i]), dtype=np.uint64) * len(
+                    self.levels[i]
+                )
+                # Set order as given in the indexer list
+                level_indexer = self.levels[i].get_indexer(k)
+                level_indexer = level_indexer[level_indexer >= 0]  # Filter absent keys
+                key_order_map[level_indexer] = np.arange(len(level_indexer))
+
+                new_order = key_order_map[self.codes[i][indexer]]
+            elif isinstance(k, slice) and k.step is not None and k.step < 0:
+                # flip order for negative step
+                new_order = np.arange(n - 1, -1, -1)[indexer]
+            elif isinstance(k, slice) and k.start is None and k.stop is None:
+                # slice(None) should not determine order GH#31330
+                new_order = np.ones((n,), dtype=np.intp)[indexer]
+            else:
+                # For all other case, use the same order as the level
+                new_order = np.arange(n)[indexer]
+            keys = (new_order,) + keys
+
+        # Find the reordering using lexsort on the keys mapping
+        ind = np.lexsort(keys)
+        return indexer[ind]
+
+    def truncate(self, before=None, after=None) -> MultiIndex:
+        """
+        Slice index between two labels / tuples, return new MultiIndex.
+
+        Parameters
+        ----------
+        before : label or tuple, can be partial. Default None
+            None defaults to start.
+        after : label or tuple, can be partial. Default None
+            None defaults to end.
+
+        Returns
+        -------
+        MultiIndex
+            The truncated MultiIndex.
+
+        See Also
+        --------
+        DataFrame.truncate : Truncate a DataFrame before and after some index values.
+        Series.truncate : Truncate a Series before and after some index values.
+
+        Examples
+        --------
+        >>> mi = pd.MultiIndex.from_arrays([["a", "b", "c"], ["x", "y", "z"]])
+        >>> mi
+        MultiIndex([('a', 'x'), ('b', 'y'), ('c', 'z')],
+                   )
+        >>> mi.truncate(before="a", after="b")
+        MultiIndex([('a', 'x'), ('b', 'y')],
+                   )
+        """
+        if after and before and after < before:
+            raise ValueError("after < before")
+
+        i, j = self.levels[0].slice_locs(before, after)
+        left, right = self.slice_locs(before, after)
+
+        new_levels = list(self.levels)
+        new_levels[0] = new_levels[0][i:j]
+
+        new_codes = [level_codes[left:right] for level_codes in self.codes]
+        new_codes[0] = new_codes[0] - i
+
+        return MultiIndex(
+            levels=new_levels,
+            codes=new_codes,
+            names=self._names,
+            verify_integrity=False,
+        )
+
+    def equals(self, other: object) -> bool:
+        """
+        Determines if two MultiIndex objects have the same labeling information
+        (the levels themselves do not necessarily have to be the same)
+
+        See Also
+        --------
+        equal_levels
+        """
+        if self.is_(other):
+            return True
+
+        if not isinstance(other, Index):
+            return False
+
+        if len(self) != len(other):
+            return False
+
+        if not isinstance(other, MultiIndex):
+            # d-level MultiIndex can equal d-tuple Index
+            if not self._should_compare(other):
+                # object Index or Categorical[object] may contain tuples
+                return False
+            return array_equivalent(self._values, other._values)
+
+        if self.nlevels != other.nlevels:
+            return False
+
+        for i in range(self.nlevels):
+            self_codes = self.codes[i]
+            other_codes = other.codes[i]
+            self_mask = self_codes == -1
+            other_mask = other_codes == -1
+            if not np.array_equal(self_mask, other_mask):
+                return False
+            self_level = self.levels[i]
+            other_level = other.levels[i]
+            new_codes = recode_for_categories(
+                other_codes, other_level, self_level, copy=False
+            )
+            if not np.array_equal(self_codes, new_codes):
+                return False
+            if not self_level[:0].equals(other_level[:0]):
+                # e.g. Int64 != int64
+                return False
+        return True
+
+    def equal_levels(self, other: MultiIndex) -> bool:
+        """
+        Return True if the levels of both MultiIndex objects are the same
+
+        """
+        if self.nlevels != other.nlevels:
+            return False
+
+        for i in range(self.nlevels):
+            if not self.levels[i].equals(other.levels[i]):
+                return False
+        return True
+
+    # --------------------------------------------------------------------
+    # Set Methods
+
+    def _union(self, other, sort) -> MultiIndex:
+        other, result_names = self._convert_can_do_setop(other)
+        if other.has_duplicates:
+            # This is only necessary if other has dupes,
+            # otherwise difference is faster
+            result = super()._union(other, sort)
+
+            if isinstance(result, MultiIndex):
+                return result
+            return MultiIndex.from_arrays(
+                zip(*result), sortorder=None, names=result_names
+            )
+
+        else:
+            right_missing = other.difference(self, sort=False)
+            if len(right_missing):
+                result = self.append(right_missing)
+            else:
+                result = self._get_reconciled_name_object(other)
+
+            if sort is not False:
+                try:
+                    result = result.sort_values()
+                except TypeError:
+                    if sort is True:
+                        raise
+                    warnings.warn(
+                        "The values in the array are unorderable. "
+                        "Pass `sort=False` to suppress this warning.",
+                        RuntimeWarning,
+                        stacklevel=find_stack_level(),
+                    )
+            return result
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        return is_object_dtype(dtype)
+
+    def _get_reconciled_name_object(self, other) -> MultiIndex:
+        """
+        If the result of a set operation will be self,
+        return self, unless the names change, in which
+        case make a shallow copy of self.
+        """
+        names = self._maybe_match_names(other)
+        if self.names != names:
+            # error: Cannot determine type of "rename"
+            return self.rename(names)  # type: ignore[has-type]
+        return self
+
+    def _maybe_match_names(self, other):
+        """
+        Try to find common names to attach to the result of an operation between
+        a and b. Return a consensus list of names if they match at least partly
+        or list of None if they have completely different names.
+        """
+        if len(self.names) != len(other.names):
+            return [None] * len(self.names)
+        names = []
+        for a_name, b_name in zip(self.names, other.names):
+            if a_name == b_name:
+                names.append(a_name)
+            else:
+                # TODO: what if they both have np.nan for their names?
+                names.append(None)
+        return names
+
+    def _wrap_intersection_result(self, other, result) -> MultiIndex:
+        _, result_names = self._convert_can_do_setop(other)
+        return result.set_names(result_names)
+
+    def _wrap_difference_result(self, other, result: MultiIndex) -> MultiIndex:
+        _, result_names = self._convert_can_do_setop(other)
+
+        if len(result) == 0:
+            return result.remove_unused_levels().set_names(result_names)
+        else:
+            return result.set_names(result_names)
+
+    def _convert_can_do_setop(self, other):
+        result_names = self.names
+
+        if not isinstance(other, Index):
+            if len(other) == 0:
+                return self[:0], self.names
+            else:
+                msg = "other must be a MultiIndex or a list of tuples"
+                try:
+                    other = MultiIndex.from_tuples(other, names=self.names)
+                except (ValueError, TypeError) as err:
+                    # ValueError raised by tuples_to_object_array if we
+                    #  have non-object dtype
+                    raise TypeError(msg) from err
+        else:
+            result_names = get_unanimous_names(self, other)
+
+        return other, result_names
+
+    # --------------------------------------------------------------------
+
+    @doc(Index.astype)
+    def astype(self, dtype, copy: bool = True):
+        dtype = pandas_dtype(dtype)
+        if isinstance(dtype, CategoricalDtype):
+            msg = "> 1 ndim Categorical are not supported at this time"
+            raise NotImplementedError(msg)
+        if not is_object_dtype(dtype):
+            raise TypeError(
+                "Setting a MultiIndex dtype to anything other than object "
+                "is not supported"
+            )
+        if copy is True:
+            return self._view()
+        return self
+
+    def _validate_fill_value(self, item):
+        if isinstance(item, MultiIndex):
+            # GH#43212
+            if item.nlevels != self.nlevels:
+                raise ValueError("Item must have length equal to number of levels.")
+            return item._values
+        elif not isinstance(item, tuple):
+            # Pad the key with empty strings if lower levels of the key
+            # aren't specified:
+            item = (item,) + ("",) * (self.nlevels - 1)
+        elif len(item) != self.nlevels:
+            raise ValueError("Item must have length equal to number of levels.")
+        return item
+
+    def putmask(self, mask, value: MultiIndex) -> MultiIndex:
+        """
+        Return a new MultiIndex of the values set with the mask.
+
+        Parameters
+        ----------
+        mask : array like
+        value : MultiIndex
+            Must either be the same length as self or length one
+
+        Returns
+        -------
+        MultiIndex
+        """
+        mask, noop = validate_putmask(self, mask)
+        if noop:
+            return self.copy()
+
+        if len(mask) == len(value):
+            subset = value[mask].remove_unused_levels()
+        else:
+            subset = value.remove_unused_levels()
+
+        new_levels = []
+        new_codes = []
+
+        for i, (value_level, level, level_codes) in enumerate(
+            zip(subset.levels, self.levels, self.codes)
+        ):
+            new_level = level.union(value_level, sort=False)
+            value_codes = new_level.get_indexer_for(subset.get_level_values(i))
+            new_code = ensure_int64(level_codes)
+            new_code[mask] = value_codes
+            new_levels.append(new_level)
+            new_codes.append(new_code)
+
+        return MultiIndex(
+            levels=new_levels, codes=new_codes, names=self.names, verify_integrity=False
+        )
+
+    def insert(self, loc: int, item) -> MultiIndex:
+        """
+        Make new MultiIndex inserting new item at location
+
+        Parameters
+        ----------
+        loc : int
+        item : tuple
+            Must be same length as number of levels in the MultiIndex
+
+        Returns
+        -------
+        new_index : Index
+        """
+        item = self._validate_fill_value(item)
+
+        new_levels = []
+        new_codes = []
+        for k, level, level_codes in zip(item, self.levels, self.codes):
+            if k not in level:
+                # have to insert into level
+                # must insert at end otherwise you have to recompute all the
+                # other codes
+                lev_loc = len(level)
+                level = level.insert(lev_loc, k)
+                if isna(level[lev_loc]):  # GH 59003, 60388
+                    lev_loc = -1
+            else:
+                lev_loc = level.get_loc(k)
+
+            new_levels.append(level)
+            new_codes.append(np.insert(ensure_int64(level_codes), loc, lev_loc))
+
+        return MultiIndex(
+            levels=new_levels, codes=new_codes, names=self.names, verify_integrity=False
+        )
+
+    def delete(self, loc) -> MultiIndex:
+        """
+        Make new index with passed location deleted
+
+        Returns
+        -------
+        new_index : MultiIndex
+        """
+        new_codes = [np.delete(level_codes, loc) for level_codes in self.codes]
+        return MultiIndex(
+            levels=self.levels,
+            codes=new_codes,
+            names=self.names,
+            verify_integrity=False,
+        )
+
+    @doc(Index.isin)
+    def isin(self, values, level=None) -> npt.NDArray[np.bool_]:
+        if isinstance(values, Generator):
+            values = list(values)
+
+        if level is None:
+            if len(values) == 0:
+                return np.zeros((len(self),), dtype=np.bool_)
+            if not isinstance(values, MultiIndex):
+                values = MultiIndex.from_tuples(values)
+            return values.unique().get_indexer_for(self) != -1
+        else:
+            num = self._get_level_number(level)
+            levs = self.get_level_values(num)
+
+            if levs.size == 0:
+                return np.zeros(len(levs), dtype=np.bool_)
+            return levs.isin(values)
+
+    # error: Incompatible types in assignment (expression has type overloaded function,
+    # base class "Index" defined the type as "Callable[[Index, Any, bool], Any]")
+    rename = Index.set_names  # type: ignore[assignment]
+
+    # ---------------------------------------------------------------
+    # Arithmetic/Numeric Methods - Disabled
+
+    __add__ = make_invalid_op("__add__")
+    __radd__ = make_invalid_op("__radd__")
+    __iadd__ = make_invalid_op("__iadd__")
+    __sub__ = make_invalid_op("__sub__")
+    __rsub__ = make_invalid_op("__rsub__")
+    __isub__ = make_invalid_op("__isub__")
+    __pow__ = make_invalid_op("__pow__")
+    __rpow__ = make_invalid_op("__rpow__")
+    __mul__ = make_invalid_op("__mul__")
+    __rmul__ = make_invalid_op("__rmul__")
+    __floordiv__ = make_invalid_op("__floordiv__")
+    __rfloordiv__ = make_invalid_op("__rfloordiv__")
+    __truediv__ = make_invalid_op("__truediv__")
+    __rtruediv__ = make_invalid_op("__rtruediv__")
+    __mod__ = make_invalid_op("__mod__")
+    __rmod__ = make_invalid_op("__rmod__")
+    __divmod__ = make_invalid_op("__divmod__")
+    __rdivmod__ = make_invalid_op("__rdivmod__")
+    # Unary methods disabled
+    __neg__ = make_invalid_op("__neg__")
+    __pos__ = make_invalid_op("__pos__")
+    __abs__ = make_invalid_op("__abs__")
+    __invert__ = make_invalid_op("__invert__")
+
+
+def _lexsort_depth(codes: list[np.ndarray], nlevels: int) -> int:
+    """Count depth (up to a maximum of `nlevels`) with which codes are lexsorted."""
+    int64_codes = [ensure_int64(level_codes) for level_codes in codes]
+    for k in range(nlevels, 0, -1):
+        if libalgos.is_lexsorted(int64_codes[:k]):
+            return k
+    return 0
+
+
+def sparsify_labels(label_list, start: int = 0, sentinel: object = ""):
+    pivoted = list(zip(*label_list))
+    k = len(label_list)
+
+    result = pivoted[: start + 1]
+    prev = pivoted[start]
+
+    for cur in pivoted[start + 1 :]:
+        sparse_cur = []
+
+        for i, (p, t) in enumerate(zip(prev, cur)):
+            if i == k - 1:
+                sparse_cur.append(t)
+                result.append(sparse_cur)  # type: ignore[arg-type]
+                break
+
+            if p == t:
+                sparse_cur.append(sentinel)
+            else:
+                sparse_cur.extend(cur[i:])
+                result.append(sparse_cur)  # type: ignore[arg-type]
+                break
+
+        prev = cur
+
+    return list(zip(*result))
+
+
+def _get_na_rep(dtype: DtypeObj) -> str:
+    if isinstance(dtype, ExtensionDtype):
+        return f"{dtype.na_value}"
+    else:
+        dtype_type = dtype.type
+
+    return {np.datetime64: "NaT", np.timedelta64: "NaT"}.get(dtype_type, "NaN")
+
+
+def maybe_droplevels(index: Index, key) -> Index:
+    """
+    Attempt to drop level or levels from the given index.
+
+    Parameters
+    ----------
+    index: Index
+    key : scalar or tuple
+
+    Returns
+    -------
+    Index
+    """
+    # drop levels
+    original_index = index
+    if isinstance(key, tuple):
+        # Caller is responsible for ensuring the key is not an entry in the first
+        #  level of the MultiIndex.
+        for _ in key:
+            try:
+                index = index._drop_level_numbers([0])
+            except ValueError:
+                # we have dropped too much, so back out
+                return original_index
+    else:
+        try:
+            index = index._drop_level_numbers([0])
+        except ValueError:
+            pass
+
+    return index
+
+
+def _coerce_indexer_frozen(array_like, categories, copy: bool = False) -> np.ndarray:
+    """
+    Coerce the array-like indexer to the smallest integer dtype that can encode all
+    of the given categories.
+
+    Parameters
+    ----------
+    array_like : array-like
+    categories : array-like
+    copy : bool
+
+    Returns
+    -------
+    np.ndarray
+        Non-writeable.
+    """
+    array_like = coerce_indexer_dtype(array_like, categories)
+    if copy:
+        array_like = array_like.copy()
+    array_like.flags.writeable = False
+    return array_like
+
+
+def _require_listlike(level, arr, arrname: str):
+    """
+    Ensure that level is either None or listlike, and arr is list-of-listlike.
+    """
+    if level is not None and not is_list_like(level):
+        if not is_list_like(arr):
+            raise TypeError(f"{arrname} must be list-like")
+        if len(arr) > 0 and is_list_like(arr[0]):
+            raise TypeError(f"{arrname} must be list-like")
+        level = [level]
+        arr = [arr]
+    elif level is None or is_list_like(level):
+        if not is_list_like(arr) or not is_list_like(arr[0]):
+            raise TypeError(f"{arrname} must be list of lists-like")
+    return level, arr
+
+
+def cartesian_product(X: list[np.ndarray]) -> list[np.ndarray]:
+    """
+    Numpy version of itertools.product.
+    Sometimes faster (for large inputs)...
+
+    Parameters
+    ----------
+    X : list-like of list-likes
+
+    Returns
+    -------
+    product : list of ndarrays
+
+    Examples
+    --------
+    >>> cartesian_product([list("ABC"), [1, 2]])
+    [array(['A', 'A', 'B', 'B', 'C', 'C'], dtype='<U1'), array([1, 2, 1, 2, 1, 2])]
+
+    See Also
+    --------
+    itertools.product : Cartesian product of input iterables.  Equivalent to
+        nested for-loops.
+    """
+    msg = "Input must be a list-like of list-likes"
+    if not is_list_like(X):
+        raise TypeError(msg)
+    for x in X:
+        if not is_list_like(x):
+            raise TypeError(msg)
+
+    if len(X) == 0:
+        return []
+
+    lenX = np.fromiter((len(x) for x in X), dtype=np.intp)
+    cumprodX = np.cumprod(lenX)
+
+    if np.any(cumprodX < 0):
+        raise ValueError("Product space too large to allocate arrays!")
+
+    a = np.roll(cumprodX, 1)
+    a[0] = 1
+
+    if cumprodX[-1] != 0:
+        b = cumprodX[-1] / cumprodX
+    else:
+        # if any factor is empty, the cartesian product is empty
+        b = np.zeros_like(cumprodX)
+
+    return [
+        np.tile(
+            np.repeat(x, b[i]),
+            np.prod(a[i]),
+        )
+        for i, x in enumerate(X)
+    ]
diff --git a/pandas/pandas/core/indexes/period.py b/pandas/pandas/core/indexes/period.py
new file mode 100644
index 0000000000000..0a7a0319bed3a
--- /dev/null
+++ b/pandas/pandas/core/indexes/period.py
@@ -0,0 +1,616 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import index as libindex
+from pandas._libs.tslibs import (
+    BaseOffset,
+    NaT,
+    Period,
+    Resolution,
+    Tick,
+)
+from pandas._libs.tslibs.dtypes import OFFSET_TO_PERIOD_FREQSTR
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+    set_module,
+)
+
+from pandas.core.dtypes.common import is_integer
+from pandas.core.dtypes.dtypes import PeriodDtype
+from pandas.core.dtypes.generic import ABCSeries
+from pandas.core.dtypes.missing import is_valid_na_for_dtype
+
+from pandas.core.arrays.period import (
+    PeriodArray,
+    period_array,
+    raise_on_incompatible,
+    validate_dtype_freq,
+)
+import pandas.core.common as com
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.base import maybe_extract_name
+from pandas.core.indexes.datetimelike import DatetimeIndexOpsMixin
+from pandas.core.indexes.datetimes import (
+    DatetimeIndex,
+    Index,
+)
+from pandas.core.indexes.extension import inherit_names
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import (
+        Dtype,
+        DtypeObj,
+        Self,
+        npt,
+    )
+
+
+_index_doc_kwargs = dict(ibase._index_doc_kwargs)
+_index_doc_kwargs.update({"target_klass": "PeriodIndex or list of Periods"})
+_shared_doc_kwargs = {
+    "klass": "PeriodArray",
+}
+
+# --- Period index sketch
+
+
+def _new_PeriodIndex(cls, **d):
+    # GH13277 for unpickling
+    values = d.pop("data")
+    if values.dtype == "int64":
+        freq = d.pop("freq", None)
+        dtype = PeriodDtype(freq)
+        values = PeriodArray(values, dtype=dtype)
+        return cls._simple_new(values, **d)
+    else:
+        return cls(values, **d)
+
+
+@inherit_names(
+    ["strftime", "start_time", "end_time"] + PeriodArray._field_ops,
+    PeriodArray,
+    wrap=True,
+)
+@inherit_names(["is_leap_year"], PeriodArray)
+@set_module("pandas")
+class PeriodIndex(DatetimeIndexOpsMixin):
+    """
+    Immutable ndarray holding ordinal values indicating regular periods in time.
+
+    Index keys are boxed to Period objects which carries the metadata (eg,
+    frequency information).
+
+    Parameters
+    ----------
+    data : array-like (1d int np.ndarray or PeriodArray), optional
+        Optional period-like data to construct index with.
+    freq : str or period object, optional
+        One of pandas period strings or corresponding objects.
+    dtype : str or PeriodDtype, default None
+        A dtype from which to extract a freq.
+    copy : bool
+        Make a copy of input ndarray.
+    name : str, default None
+        Name of the resulting PeriodIndex.
+
+    Attributes
+    ----------
+    day
+    dayofweek
+    day_of_week
+    dayofyear
+    day_of_year
+    days_in_month
+    daysinmonth
+    end_time
+    freq
+    freqstr
+    hour
+    is_leap_year
+    minute
+    month
+    quarter
+    qyear
+    second
+    start_time
+    week
+    weekday
+    weekofyear
+    year
+
+    Methods
+    -------
+    asfreq
+    strftime
+    to_timestamp
+    from_fields
+    from_ordinals
+
+    Raises
+    ------
+    ValueError
+        Passing the parameter data as a list without specifying either freq or
+        dtype will raise a ValueError: "freq not specified and cannot be inferred"
+
+    See Also
+    --------
+    Index : The base pandas Index type.
+    Period : Represents a period of time.
+    DatetimeIndex : Index with datetime64 data.
+    TimedeltaIndex : Index of timedelta64 data.
+    period_range : Create a fixed-frequency PeriodIndex.
+
+    Examples
+    --------
+    >>> idx = pd.PeriodIndex(data=["2000Q1", "2002Q3"], freq="Q")
+    >>> idx
+    PeriodIndex(['2000Q1', '2002Q3'], dtype='period[Q-DEC]')
+    """
+
+    _typ = "periodindex"
+
+    _data: PeriodArray
+    freq: BaseOffset
+    dtype: PeriodDtype
+
+    _data_cls = PeriodArray
+    _supports_partial_string_indexing = True
+
+    @property
+    def _engine_type(self) -> type[libindex.PeriodEngine]:
+        return libindex.PeriodEngine
+
+    @cache_readonly
+    def _resolution_obj(self) -> Resolution:
+        # for compat with DatetimeIndex
+        return self.dtype._resolution_obj
+
+    # --------------------------------------------------------------------
+    # methods that dispatch to array and wrap result in Index
+    # These are defined here instead of via inherit_names for mypy
+
+    @doc(
+        PeriodArray.asfreq,
+        other="arrays.PeriodArray",
+        other_name="PeriodArray",
+        **_shared_doc_kwargs,
+    )
+    def asfreq(self, freq=None, how: str = "E") -> Self:
+        arr = self._data.asfreq(freq, how)
+        return type(self)._simple_new(arr, name=self.name)
+
+    @doc(PeriodArray.to_timestamp)
+    def to_timestamp(self, freq=None, how: str = "start") -> DatetimeIndex:
+        arr = self._data.to_timestamp(freq, how)
+        return DatetimeIndex._simple_new(arr, name=self.name)
+
+    @property
+    @doc(PeriodArray.hour.fget)
+    def hour(self) -> Index:
+        return Index(self._data.hour, name=self.name)
+
+    @property
+    @doc(PeriodArray.minute.fget)
+    def minute(self) -> Index:
+        return Index(self._data.minute, name=self.name)
+
+    @property
+    @doc(PeriodArray.second.fget)
+    def second(self) -> Index:
+        return Index(self._data.second, name=self.name)
+
+    # ------------------------------------------------------------------------
+    # Index Constructors
+
+    def __new__(
+        cls,
+        data=None,
+        freq=None,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        name: Hashable | None = None,
+    ) -> Self:
+        refs = None
+        if not copy and isinstance(data, (Index, ABCSeries)):
+            refs = data._references
+
+        name = maybe_extract_name(name, data, cls)
+
+        freq = validate_dtype_freq(dtype, freq)
+
+        # PeriodIndex allow PeriodIndex(period_index, freq=different)
+        # Let's not encourage that kind of behavior in PeriodArray.
+
+        if freq and isinstance(data, cls) and data.freq != freq:
+            # TODO: We can do some of these with no-copy / coercion?
+            # e.g. D -> 2D seems to be OK
+            data = data.asfreq(freq)
+
+        # don't pass copy here, since we copy later.
+        data = period_array(data=data, freq=freq)
+
+        if copy:
+            data = data.copy()
+
+        return cls._simple_new(data, name=name, refs=refs)
+
+    @classmethod
+    def from_fields(
+        cls,
+        *,
+        year=None,
+        quarter=None,
+        month=None,
+        day=None,
+        hour=None,
+        minute=None,
+        second=None,
+        freq=None,
+    ) -> Self:
+        """
+        Construct a PeriodIndex from fields (year, month, day, etc.).
+
+        Parameters
+        ----------
+        year : int, array, or Series, default None
+            Year for the PeriodIndex.
+        quarter : int, array, or Series, default None
+            Quarter for the PeriodIndex.
+        month : int, array, or Series, default None
+            Month for the PeriodIndex.
+        day : int, array, or Series, default None
+            Day for the PeriodIndex.
+        hour : int, array, or Series, default None
+            Hour for the PeriodIndex.
+        minute : int, array, or Series, default None
+            Minute for the PeriodIndex.
+        second : int, array, or Series, default None
+            Second for the PeriodIndex.
+        freq : str or period object, optional
+            One of pandas period strings or corresponding objects.
+
+        Returns
+        -------
+        PeriodIndex
+
+        See Also
+        --------
+        PeriodIndex.from_ordinals : Construct a PeriodIndex from ordinals.
+        PeriodIndex.to_timestamp : Cast to DatetimeArray/Index.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex.from_fields(year=[2000, 2002], quarter=[1, 3])
+        >>> idx
+        PeriodIndex(['2000Q1', '2002Q3'], dtype='period[Q-DEC]')
+        """
+        fields = {
+            "year": year,
+            "quarter": quarter,
+            "month": month,
+            "day": day,
+            "hour": hour,
+            "minute": minute,
+            "second": second,
+        }
+        fields = {key: value for key, value in fields.items() if value is not None}
+        arr = PeriodArray._from_fields(fields=fields, freq=freq)
+        return cls._simple_new(arr)
+
+    @classmethod
+    def from_ordinals(cls, ordinals, *, freq, name=None) -> Self:
+        """
+        Construct a PeriodIndex from ordinals.
+
+        Parameters
+        ----------
+        ordinals : array-like of int
+            The period offsets from the proleptic Gregorian epoch.
+        freq : str or period object
+            One of pandas period strings or corresponding objects.
+        name : str, default None
+            Name of the resulting PeriodIndex.
+
+        Returns
+        -------
+        PeriodIndex
+
+        See Also
+        --------
+        PeriodIndex.from_fields : Construct a PeriodIndex from fields
+            (year, month, day, etc.).
+        PeriodIndex.to_timestamp : Cast to DatetimeArray/Index.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex.from_ordinals([-1, 0, 1], freq="Q")
+        >>> idx
+        PeriodIndex(['1969Q4', '1970Q1', '1970Q2'], dtype='period[Q-DEC]')
+        """
+        ordinals = np.asarray(ordinals, dtype=np.int64)
+        dtype = PeriodDtype(freq)
+        data = PeriodArray._simple_new(ordinals, dtype=dtype)
+        return cls._simple_new(data, name=name)
+
+    # ------------------------------------------------------------------------
+    # Data
+
+    @property
+    def values(self) -> npt.NDArray[np.object_]:
+        return np.asarray(self, dtype=object)
+
+    def _maybe_convert_timedelta(self, other) -> int | npt.NDArray[np.int64]:
+        """
+        Convert timedelta-like input to an integer multiple of self.freq
+
+        Parameters
+        ----------
+        other : timedelta, np.timedelta64, DateOffset, int, np.ndarray
+
+        Returns
+        -------
+        converted : int, np.ndarray[int64]
+
+        Raises
+        ------
+        IncompatibleFrequency : if the input cannot be written as a multiple
+            of self.freq.  Note IncompatibleFrequency subclasses ValueError.
+        """
+        if isinstance(other, (timedelta, np.timedelta64, Tick, np.ndarray)):
+            if isinstance(self.freq, Tick):
+                # _check_timedeltalike_freq_compat will raise if incompatible
+                delta = self._data._check_timedeltalike_freq_compat(other)
+                return delta
+        elif isinstance(other, BaseOffset):
+            if other.base == self.freq.base:
+                return other.n
+
+            raise raise_on_incompatible(self, other)
+        elif is_integer(other):
+            assert isinstance(other, int)
+            return other
+
+        # raise when input doesn't have freq
+        raise raise_on_incompatible(self, None)
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        """
+        Can we compare values of the given dtype to our own?
+        """
+        return self.dtype == dtype
+
+    # ------------------------------------------------------------------------
+    # Index Methods
+
+    def asof_locs(self, where: Index, mask: npt.NDArray[np.bool_]) -> np.ndarray:
+        """
+        where : array of timestamps
+        mask : np.ndarray[bool]
+            Array of booleans where data is not NA.
+        """
+        if isinstance(where, DatetimeIndex):
+            where = PeriodIndex(where._values, freq=self.freq)
+        elif not isinstance(where, PeriodIndex):
+            raise TypeError("asof_locs `where` must be DatetimeIndex or PeriodIndex")
+
+        return super().asof_locs(where, mask)
+
+    @property
+    def is_full(self) -> bool:
+        """
+        Returns True if this PeriodIndex is range-like in that all Periods
+        between start and end are present, in order.
+        """
+        if len(self) == 0:
+            return True
+        if not self.is_monotonic_increasing:
+            raise ValueError("Index is not monotonic")
+        values = self.asi8
+        return bool(((values[1:] - values[:-1]) < 2).all())
+
+    @property
+    def inferred_type(self) -> str:
+        # b/c data is represented as ints make sure we can't have ambiguous
+        # indexing
+        return "period"
+
+    # ------------------------------------------------------------------------
+    # Indexing Methods
+
+    def _convert_tolerance(self, tolerance, target):
+        # Returned tolerance must be in dtype/units so that
+        #  `|self._get_engine_target() - target._engine_target()| <= tolerance`
+        #  is meaningful.  Since PeriodIndex returns int64 for engine_target,
+        #  we may need to convert timedelta64 tolerance to int64.
+        tolerance = super()._convert_tolerance(tolerance, target)
+
+        if self.dtype == target.dtype:
+            # convert tolerance to i8
+            tolerance = self._maybe_convert_timedelta(tolerance)
+
+        return tolerance
+
+    def get_loc(self, key):
+        """
+        Get integer location for requested label.
+
+        Parameters
+        ----------
+        key : Period, NaT, str, or datetime
+            String or datetime key must be parsable as Period.
+
+        Returns
+        -------
+        loc : int or ndarray[int64]
+
+        Raises
+        ------
+        KeyError
+            Key is not present in the index.
+        TypeError
+            If key is listlike or otherwise not hashable.
+        """
+        orig_key = key
+
+        self._check_indexing_error(key)
+
+        if is_valid_na_for_dtype(key, self.dtype):
+            key = NaT
+
+        elif isinstance(key, str):
+            try:
+                parsed, reso = self._parse_with_reso(key)
+            except ValueError as err:
+                # A string with invalid format
+                raise KeyError(f"Cannot interpret '{key}' as period") from err
+
+            if self._can_partial_date_slice(reso):
+                try:
+                    return self._partial_date_slice(reso, parsed)
+                except KeyError as err:
+                    raise KeyError(key) from err
+
+            if reso == self._resolution_obj:
+                # the reso < self._resolution_obj case goes
+                #  through _get_string_slice
+                key = self._cast_partial_indexing_scalar(parsed)
+            else:
+                raise KeyError(key)
+
+        elif isinstance(key, Period):
+            self._disallow_mismatched_indexing(key)
+
+        elif isinstance(key, datetime):
+            key = self._cast_partial_indexing_scalar(key)
+
+        else:
+            # in particular integer, which Period constructor would cast to string
+            raise KeyError(key)
+
+        try:
+            return Index.get_loc(self, key)
+        except KeyError as err:
+            raise KeyError(orig_key) from err
+
+    def _disallow_mismatched_indexing(self, key: Period) -> None:
+        if key._dtype != self.dtype:
+            raise KeyError(key)
+
+    def _cast_partial_indexing_scalar(self, label: datetime) -> Period:
+        try:
+            period = Period(label, freq=self.freq)
+        except ValueError as err:
+            # we cannot construct the Period
+            raise KeyError(label) from err
+        return period
+
+    @doc(DatetimeIndexOpsMixin._maybe_cast_slice_bound)
+    def _maybe_cast_slice_bound(self, label, side: str):
+        if isinstance(label, datetime):
+            label = self._cast_partial_indexing_scalar(label)
+
+        return super()._maybe_cast_slice_bound(label, side)
+
+    def _parsed_string_to_bounds(self, reso: Resolution, parsed: datetime):
+        freq = OFFSET_TO_PERIOD_FREQSTR.get(reso.attr_abbrev, reso.attr_abbrev)
+        iv = Period(parsed, freq=freq)
+        return (iv.asfreq(self.freq, how="start"), iv.asfreq(self.freq, how="end"))
+
+    @doc(DatetimeIndexOpsMixin.shift)
+    def shift(self, periods: int = 1, freq=None) -> Self:
+        if freq is not None:
+            raise TypeError(
+                f"`freq` argument is not supported for {type(self).__name__}.shift"
+            )
+        return self + periods
+
+
+def period_range(
+    start=None,
+    end=None,
+    periods: int | None = None,
+    freq=None,
+    name: Hashable | None = None,
+) -> PeriodIndex:
+    """
+    Return a fixed frequency PeriodIndex.
+
+    The day (calendar) is the default frequency.
+
+    Parameters
+    ----------
+    start : str, datetime, date, pandas.Timestamp, or period-like, default None
+        Left bound for generating periods.
+    end : str, datetime, date, pandas.Timestamp, or period-like, default None
+        Right bound for generating periods.
+    periods : int, default None
+        Number of periods to generate.
+    freq : str or DateOffset, optional
+        Frequency alias. By default the freq is taken from `start` or `end`
+        if those are Period objects. Otherwise, the default is ``"D"`` for
+        daily frequency.
+    name : str, default None
+        Name of the resulting PeriodIndex.
+
+    Returns
+    -------
+    PeriodIndex
+        A PeriodIndex of fixed frequency periods.
+
+    See Also
+    --------
+    date_range : Returns a fixed frequency DatetimeIndex.
+    Period : Represents a period of time.
+    PeriodIndex : Immutable ndarray holding ordinal values indicating regular periods
+        in time.
+
+    Notes
+    -----
+    Of the three parameters: ``start``, ``end``, and ``periods``, exactly two
+    must be specified.
+
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    >>> pd.period_range(start="2017-01-01", end="2018-01-01", freq="M")
+    PeriodIndex(['2017-01', '2017-02', '2017-03', '2017-04', '2017-05', '2017-06',
+             '2017-07', '2017-08', '2017-09', '2017-10', '2017-11', '2017-12',
+             '2018-01'],
+            dtype='period[M]')
+
+    If ``start`` or ``end`` are ``Period`` objects, they will be used as anchor
+    endpoints for a ``PeriodIndex`` with frequency matching that of the
+    ``period_range`` constructor.
+
+    >>> pd.period_range(
+    ...     start=pd.Period("2017Q1", freq="Q"),
+    ...     end=pd.Period("2017Q2", freq="Q"),
+    ...     freq="M",
+    ... )
+    PeriodIndex(['2017-03', '2017-04', '2017-05', '2017-06'],
+                dtype='period[M]')
+    """
+    if com.count_not_none(start, end, periods) != 2:
+        raise ValueError(
+            "Of the three parameters: start, end, and periods, "
+            "exactly two must be specified"
+        )
+    if freq is None and (not isinstance(start, Period) and not isinstance(end, Period)):
+        freq = "D"
+
+    data, freq = PeriodArray._generate_range(start, end, periods, freq)
+    dtype = PeriodDtype(freq)
+    data = PeriodArray(data, dtype=dtype)
+    return PeriodIndex(data, name=name)
diff --git a/pandas/pandas/core/indexes/range.py b/pandas/pandas/core/indexes/range.py
new file mode 100644
index 0000000000000..2db50bbbdfa37
--- /dev/null
+++ b/pandas/pandas/core/indexes/range.py
@@ -0,0 +1,1481 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterator,
+)
+from datetime import timedelta
+import operator
+from sys import getsizeof
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    index as libindex,
+    lib,
+)
+from pandas._libs.lib import no_default
+from pandas.compat.numpy import function as nv
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+    set_module,
+)
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    ensure_python_int,
+    is_float,
+    is_integer,
+    is_scalar,
+    is_signed_integer_dtype,
+)
+from pandas.core.dtypes.generic import ABCTimedeltaIndex
+
+from pandas.core import ops
+import pandas.core.common as com
+from pandas.core.construction import extract_array
+from pandas.core.indexers import check_array_indexer
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.base import (
+    Index,
+    maybe_extract_name,
+)
+from pandas.core.ops.common import unpack_zerodim_and_defer
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        Axis,
+        Dtype,
+        JoinHow,
+        NaPosition,
+        NumpySorter,
+        Self,
+        npt,
+    )
+
+    from pandas import Series
+
+_empty_range = range(0)
+_dtype_int64 = np.dtype(np.int64)
+
+
+def min_fitting_element(start: int, step: int, lower_limit: int) -> int:
+    """Returns the smallest element greater than or equal to the limit"""
+    no_steps = -(-(lower_limit - start) // abs(step))
+    return start + abs(step) * no_steps
+
+
+@set_module("pandas")
+class RangeIndex(Index):
+    """
+    Immutable Index implementing a monotonic integer range.
+
+    RangeIndex is a memory-saving special case of an Index limited to representing
+    monotonic ranges with a 64-bit dtype. Using RangeIndex may in some instances
+    improve computing speed.
+
+    This is the default index type used
+    by DataFrame and Series when no explicit index is provided by the user.
+
+    Parameters
+    ----------
+    start : int (default: 0), range, or other RangeIndex instance
+        If int and "stop" is not given, interpreted as "stop" instead.
+    stop : int (default: 0)
+        The end value of the range (exclusive).
+    step : int (default: 1)
+        The step size of the range.
+    dtype : np.int64
+        Unused, accepted for homogeneity with other index types.
+    copy : bool, default False
+        Unused, accepted for homogeneity with other index types.
+    name : object, optional
+        Name to be stored in the index.
+
+    Attributes
+    ----------
+    start
+    stop
+    step
+
+    Methods
+    -------
+    from_range
+
+    See Also
+    --------
+    Index : The base pandas Index type.
+
+    Examples
+    --------
+    >>> list(pd.RangeIndex(5))
+    [0, 1, 2, 3, 4]
+
+    >>> list(pd.RangeIndex(-2, 4))
+    [-2, -1, 0, 1, 2, 3]
+
+    >>> list(pd.RangeIndex(0, 10, 2))
+    [0, 2, 4, 6, 8]
+
+    >>> list(pd.RangeIndex(2, -10, -3))
+    [2, -1, -4, -7]
+
+    >>> list(pd.RangeIndex(0))
+    []
+
+    >>> list(pd.RangeIndex(1, 0))
+    []
+    """
+
+    _typ = "rangeindex"
+    _dtype_validation_metadata = (is_signed_integer_dtype, "signed integer")
+    _range: range
+    _values: np.ndarray
+
+    @property
+    def _engine_type(self) -> type[libindex.Int64Engine]:
+        return libindex.Int64Engine
+
+    # --------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        start=None,
+        stop=None,
+        step=None,
+        dtype: Dtype | None = None,
+        copy: bool = False,
+        name: Hashable | None = None,
+    ) -> Self:
+        cls._validate_dtype(dtype)
+        name = maybe_extract_name(name, start, cls)
+
+        # RangeIndex
+        if isinstance(start, cls):
+            return start.copy(name=name)
+        elif isinstance(start, range):
+            return cls._simple_new(start, name=name)
+
+        # validate the arguments
+        if com.all_none(start, stop, step):
+            raise TypeError("RangeIndex(...) must be called with integers")
+
+        start = ensure_python_int(start) if start is not None else 0
+
+        if stop is None:
+            start, stop = 0, start
+        else:
+            stop = ensure_python_int(stop)
+
+        step = ensure_python_int(step) if step is not None else 1
+        if step == 0:
+            raise ValueError("Step must not be zero")
+
+        rng = range(start, stop, step)
+        return cls._simple_new(rng, name=name)
+
+    @classmethod
+    def from_range(cls, data: range, name=None, dtype: Dtype | None = None) -> Self:
+        """
+        Create :class:`pandas.RangeIndex` from a ``range`` object.
+
+        This method provides a way to create a :class:`pandas.RangeIndex` directly
+        from a Python ``range`` object. The resulting :class:`RangeIndex` will have
+        the same start, stop, and step values as the input ``range`` object.
+        It is particularly useful for constructing indices in an efficient and
+        memory-friendly manner.
+
+        Parameters
+        ----------
+        data : range
+            The range object to be converted into a RangeIndex.
+        name : str, default None
+            Name to be stored in the index.
+        dtype : Dtype or None
+            Data type for the RangeIndex. If None, the default integer type will
+            be used.
+
+        Returns
+        -------
+        RangeIndex
+
+        See Also
+        --------
+        RangeIndex : Immutable Index implementing a monotonic integer range.
+        Index : Immutable sequence used for indexing and alignment.
+
+        Examples
+        --------
+        >>> pd.RangeIndex.from_range(range(5))
+        RangeIndex(start=0, stop=5, step=1)
+
+        >>> pd.RangeIndex.from_range(range(2, -10, -3))
+        RangeIndex(start=2, stop=-10, step=-3)
+        """
+        if not isinstance(data, range):
+            raise TypeError(
+                f"{cls.__name__}(...) must be called with object coercible to a "
+                f"range, {data!r} was passed"
+            )
+        cls._validate_dtype(dtype)
+        return cls._simple_new(data, name=name)
+
+    #  error: Argument 1 of "_simple_new" is incompatible with supertype "Index";
+    #  supertype defines the argument type as
+    #  "Union[ExtensionArray, ndarray[Any, Any]]"  [override]
+    @classmethod
+    def _simple_new(  # type: ignore[override]
+        cls, values: range, name: Hashable | None = None
+    ) -> Self:
+        result = object.__new__(cls)
+
+        assert isinstance(values, range)
+
+        result._range = values
+        result._name = name
+        result._cache = {}
+        result._reset_identity()
+        result._references = None
+        return result
+
+    @classmethod
+    def _validate_dtype(cls, dtype: Dtype | None) -> None:
+        if dtype is None:
+            return
+
+        validation_func, expected = cls._dtype_validation_metadata
+        if not validation_func(dtype):
+            raise ValueError(
+                f"Incorrect `dtype` passed: expected {expected}, received {dtype}"
+            )
+
+    # --------------------------------------------------------------------
+
+    # error: Return type "Type[Index]" of "_constructor" incompatible with return
+    # type "Type[RangeIndex]" in supertype "Index"
+    @cache_readonly
+    def _constructor(self) -> type[Index]:  # type: ignore[override]
+        """return the class to use for construction"""
+        return Index
+
+    # error: Signature of "_data" incompatible with supertype "Index"
+    @cache_readonly
+    def _data(self) -> np.ndarray:  # type: ignore[override]
+        """
+        An int array that for performance reasons is created only when needed.
+
+        The constructed array is saved in ``_cache``.
+        """
+        return np.arange(self.start, self.stop, self.step, dtype=np.int64)
+
+    def _get_data_as_items(self) -> list[tuple[str, int]]:
+        """return a list of tuples of start, stop, step"""
+        rng = self._range
+        return [("start", rng.start), ("stop", rng.stop), ("step", rng.step)]
+
+    def __reduce__(self):
+        d = {"name": self._name}
+        d.update(dict(self._get_data_as_items()))
+        return ibase._new_Index, (type(self), d), None
+
+    # --------------------------------------------------------------------
+    # Rendering Methods
+
+    def _format_attrs(self):
+        """
+        Return a list of tuples of the (attr, formatted_value)
+        """
+        attrs = cast("list[tuple[str, str | int]]", self._get_data_as_items())
+        if self._name is not None:
+            attrs.append(("name", ibase.default_pprint(self._name)))
+        return attrs
+
+    def _format_with_header(self, *, header: list[str], na_rep: str) -> list[str]:
+        # Equivalent to Index implementation, but faster
+        if not len(self._range):
+            return header
+        first_val_str = str(self._range[0])
+        last_val_str = str(self._range[-1])
+        max_length = max(len(first_val_str), len(last_val_str))
+
+        return header + [f"{x:<{max_length}}" for x in self._range]
+
+    # --------------------------------------------------------------------
+
+    @property
+    def start(self) -> int:
+        """
+        The value of the `start` parameter (``0`` if this was not supplied).
+
+        This property returns the starting value of the `RangeIndex`. If the `start`
+        value is not explicitly provided during the creation of the `RangeIndex`,
+        it defaults to 0.
+
+        See Also
+        --------
+        RangeIndex : Immutable index implementing a range-based index.
+        RangeIndex.stop : Returns the stop value of the `RangeIndex`.
+        RangeIndex.step : Returns the step value of the `RangeIndex`.
+
+        Examples
+        --------
+        >>> idx = pd.RangeIndex(5)
+        >>> idx.start
+        0
+
+        >>> idx = pd.RangeIndex(2, -10, -3)
+        >>> idx.start
+        2
+        """
+        # GH 25710
+        return self._range.start
+
+    @property
+    def stop(self) -> int:
+        """
+        The value of the `stop` parameter.
+
+        This property returns the `stop` value of the RangeIndex, which defines the
+        upper (or lower, in case of negative steps) bound of the index range. The
+        `stop` value is exclusive, meaning the RangeIndex includes values up to but
+        not including this value.
+
+        See Also
+        --------
+        RangeIndex : Immutable index representing a range of integers.
+        RangeIndex.start : The start value of the RangeIndex.
+        RangeIndex.step : The step size between elements in the RangeIndex.
+
+        Examples
+        --------
+        >>> idx = pd.RangeIndex(5)
+        >>> idx.stop
+        5
+
+        >>> idx = pd.RangeIndex(2, -10, -3)
+        >>> idx.stop
+        -10
+        """
+        return self._range.stop
+
+    @property
+    def step(self) -> int:
+        """
+        The value of the `step` parameter (``1`` if this was not supplied).
+
+        The ``step`` parameter determines the increment (or decrement in the case
+        of negative values) between consecutive elements in the ``RangeIndex``.
+
+        See Also
+        --------
+        RangeIndex : Immutable index implementing a range-based index.
+        RangeIndex.stop : Returns the stop value of the RangeIndex.
+        RangeIndex.start : Returns the start value of the RangeIndex.
+
+        Examples
+        --------
+        >>> idx = pd.RangeIndex(5)
+        >>> idx.step
+        1
+
+        >>> idx = pd.RangeIndex(2, -10, -3)
+        >>> idx.step
+        -3
+
+        Even if :class:`pandas.RangeIndex` is empty, ``step`` is still ``1`` if
+        not supplied.
+
+        >>> idx = pd.RangeIndex(1, 0)
+        >>> idx.step
+        1
+        """
+        # GH 25710
+        return self._range.step
+
+    @cache_readonly
+    def nbytes(self) -> int:
+        """
+        Return the number of bytes in the underlying data.
+        """
+        rng = self._range
+        return getsizeof(rng) + sum(
+            getsizeof(getattr(rng, attr_name))
+            for attr_name in ["start", "stop", "step"]
+        )
+
+    def memory_usage(self, deep: bool = False) -> int:
+        """
+        Memory usage of my values
+
+        Parameters
+        ----------
+        deep : bool
+            Introspect the data deeply, interrogate
+            `object` dtypes for system-level memory consumption
+
+        Returns
+        -------
+        bytes used
+
+        Notes
+        -----
+        Memory usage does not include memory consumed by elements that
+        are not components of the array if deep=False
+
+        See Also
+        --------
+        numpy.ndarray.nbytes
+        """
+        return self.nbytes
+
+    @property
+    def dtype(self) -> np.dtype:
+        return _dtype_int64
+
+    @property
+    def is_unique(self) -> bool:
+        """return if the index has unique values"""
+        return True
+
+    @cache_readonly
+    def is_monotonic_increasing(self) -> bool:
+        return self._range.step > 0 or len(self) <= 1
+
+    @cache_readonly
+    def is_monotonic_decreasing(self) -> bool:
+        return self._range.step < 0 or len(self) <= 1
+
+    def __contains__(self, key: Any) -> bool:
+        hash(key)
+        try:
+            key = ensure_python_int(key)
+        except (TypeError, OverflowError):
+            return False
+        return key in self._range
+
+    @property
+    def inferred_type(self) -> str:
+        return "integer"
+
+    # --------------------------------------------------------------------
+    # Indexing Methods
+
+    @doc(Index.get_loc)
+    def get_loc(self, key) -> int:
+        if is_integer(key) or (is_float(key) and key.is_integer()):
+            new_key = int(key)
+            try:
+                return self._range.index(new_key)
+            except ValueError as err:
+                raise KeyError(key) from err
+        if isinstance(key, Hashable):
+            raise KeyError(key)
+        self._check_indexing_error(key)
+        raise KeyError(key)
+
+    def _get_indexer(
+        self,
+        target: Index,
+        method: str | None = None,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> npt.NDArray[np.intp]:
+        if com.any_not_none(method, tolerance, limit):
+            return super()._get_indexer(
+                target, method=method, tolerance=tolerance, limit=limit
+            )
+
+        if self.step > 0:
+            start, stop, step = self.start, self.stop, self.step
+        else:
+            # GH 28678: work on reversed range for simplicity
+            reverse = self._range[::-1]
+            start, stop, step = reverse.start, reverse.stop, reverse.step
+
+        target_array = np.asarray(target)
+        locs = target_array - start
+        valid = (locs % step == 0) & (locs >= 0) & (target_array < stop)
+        locs[~valid] = -1
+        locs[valid] = locs[valid] / step
+
+        if step != self.step:
+            # We reversed this range: transform to original locs
+            locs[valid] = len(self) - 1 - locs[valid]
+        return ensure_platform_int(locs)
+
+    @cache_readonly
+    def _should_fallback_to_positional(self) -> bool:
+        """
+        Should an integer key be treated as positional?
+        """
+        return False
+
+    # --------------------------------------------------------------------
+
+    def tolist(self) -> list[int]:
+        return list(self._range)
+
+    @doc(Index.__iter__)
+    def __iter__(self) -> Iterator[int]:
+        yield from self._range
+
+    @doc(Index._shallow_copy)
+    def _shallow_copy(self, values, name: Hashable = no_default):
+        name = self._name if name is no_default else name
+
+        if values.dtype.kind == "f":
+            return Index(values, name=name, dtype=np.float64)
+        if values.dtype.kind == "i" and values.ndim == 1:
+            # GH 46675 & 43885: If values is equally spaced, return a
+            # more memory-compact RangeIndex instead of Index with 64-bit dtype
+            if len(values) == 1:
+                start = values[0]
+                new_range = range(start, start + self.step, self.step)
+                return type(self)._simple_new(new_range, name=name)
+            maybe_range = ibase.maybe_sequence_to_range(values)
+            if isinstance(maybe_range, range):
+                return type(self)._simple_new(maybe_range, name=name)
+        return self._constructor._simple_new(values, name=name)
+
+    def _view(self) -> Self:
+        result = type(self)._simple_new(self._range, name=self._name)
+        result._cache = self._cache
+        return result
+
+    def _wrap_reindex_result(self, target, indexer, preserve_names: bool):
+        if not isinstance(target, type(self)) and target.dtype.kind == "i":
+            target = self._shallow_copy(target._values, name=target.name)
+        return super()._wrap_reindex_result(target, indexer, preserve_names)
+
+    @doc(Index.copy)
+    def copy(self, name: Hashable | None = None, deep: bool = False) -> Self:
+        name = self._validate_names(name=name, deep=deep)[0]
+        new_index = self._rename(name=name)
+        return new_index
+
+    def _minmax(self, meth: Literal["min", "max"]) -> int | float:
+        no_steps = len(self) - 1
+        if no_steps == -1:
+            return np.nan
+        elif (meth == "min" and self.step > 0) or (meth == "max" and self.step < 0):
+            return self.start
+
+        return self.start + self.step * no_steps
+
+    def min(self, axis=None, skipna: bool = True, *args, **kwargs) -> int | float:
+        """The minimum value of the RangeIndex"""
+        nv.validate_minmax_axis(axis)
+        nv.validate_min(args, kwargs)
+        return self._minmax("min")
+
+    def max(self, axis=None, skipna: bool = True, *args, **kwargs) -> int | float:
+        """The maximum value of the RangeIndex"""
+        nv.validate_minmax_axis(axis)
+        nv.validate_max(args, kwargs)
+        return self._minmax("max")
+
+    def _argminmax(
+        self,
+        meth: Literal["min", "max"],
+        axis=None,
+        skipna: bool = True,
+    ) -> int:
+        nv.validate_minmax_axis(axis)
+        if len(self) == 0:
+            return getattr(super(), f"arg{meth}")(
+                axis=axis,
+                skipna=skipna,
+            )
+        elif meth == "min":
+            if self.step > 0:
+                return 0
+            else:
+                return len(self) - 1
+        elif meth == "max":
+            if self.step > 0:
+                return len(self) - 1
+            else:
+                return 0
+        else:
+            raise ValueError(f"{meth=} must be max or min")
+
+    def argmin(self, axis=None, skipna: bool = True, *args, **kwargs) -> int:
+        nv.validate_argmin(args, kwargs)
+        return self._argminmax("min", axis=axis, skipna=skipna)
+
+    def argmax(self, axis=None, skipna: bool = True, *args, **kwargs) -> int:
+        nv.validate_argmax(args, kwargs)
+        return self._argminmax("max", axis=axis, skipna=skipna)
+
+    def argsort(self, *args, **kwargs) -> npt.NDArray[np.intp]:
+        """
+        Returns the indices that would sort the index and its
+        underlying data.
+
+        Returns
+        -------
+        np.ndarray[np.intp]
+
+        See Also
+        --------
+        numpy.ndarray.argsort
+        """
+        ascending = kwargs.pop("ascending", True)  # EA compat
+        kwargs.pop("kind", None)  # e.g. "mergesort" is irrelevant
+        nv.validate_argsort(args, kwargs)
+
+        start, stop, step = None, None, None
+        if self._range.step > 0:
+            if ascending:
+                start = len(self)
+            else:
+                start, stop, step = len(self) - 1, -1, -1
+        elif ascending:
+            start, stop, step = len(self) - 1, -1, -1
+        else:
+            start = len(self)
+
+        return np.arange(start, stop, step, dtype=np.intp)
+
+    def factorize(
+        self,
+        sort: bool = False,
+        use_na_sentinel: bool = True,
+    ) -> tuple[npt.NDArray[np.intp], RangeIndex]:
+        if sort and self.step < 0:
+            codes = np.arange(len(self) - 1, -1, -1, dtype=np.intp)
+            uniques = self[::-1]
+        else:
+            codes = np.arange(len(self), dtype=np.intp)
+            uniques = self
+        return codes, uniques
+
+    def equals(self, other: object) -> bool:
+        """
+        Determines if two Index objects contain the same elements.
+        """
+        if isinstance(other, RangeIndex):
+            return self._range == other._range
+        return super().equals(other)
+
+    # error: Signature of "sort_values" incompatible with supertype "Index"
+    @overload  # type: ignore[override]
+    def sort_values(
+        self,
+        *,
+        return_indexer: Literal[False] = ...,
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> Self: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        return_indexer: Literal[True],
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> tuple[Self, np.ndarray | RangeIndex]: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        return_indexer: bool = ...,
+        ascending: bool = ...,
+        na_position: NaPosition = ...,
+        key: Callable | None = ...,
+    ) -> Self | tuple[Self, np.ndarray | RangeIndex]: ...
+
+    def sort_values(
+        self,
+        *,
+        return_indexer: bool = False,
+        ascending: bool = True,
+        na_position: NaPosition = "last",
+        key: Callable | None = None,
+    ) -> Self | tuple[Self, np.ndarray | RangeIndex]:
+        if key is not None:
+            return super().sort_values(
+                return_indexer=return_indexer,
+                ascending=ascending,
+                na_position=na_position,
+                key=key,
+            )
+        else:
+            sorted_index = self
+            inverse_indexer = False
+            if ascending:
+                if self.step < 0:
+                    sorted_index = self[::-1]
+                    inverse_indexer = True
+            else:
+                if self.step > 0:
+                    sorted_index = self[::-1]
+                    inverse_indexer = True
+
+        if return_indexer:
+            if inverse_indexer:
+                rng = range(len(self) - 1, -1, -1)
+            else:
+                rng = range(len(self))
+            return sorted_index, RangeIndex(rng)
+        else:
+            return sorted_index
+
+    # --------------------------------------------------------------------
+    # Set Operations
+
+    def _intersection(self, other: Index, sort: bool = False):
+        # caller is responsible for checking self and other are both non-empty
+
+        if not isinstance(other, RangeIndex):
+            return super()._intersection(other, sort=sort)
+
+        first = self._range[::-1] if self.step < 0 else self._range
+        second = other._range[::-1] if other.step < 0 else other._range
+
+        # check whether intervals intersect
+        # deals with in- and decreasing ranges
+        int_low = max(first.start, second.start)
+        int_high = min(first.stop, second.stop)
+        if int_high <= int_low:
+            return self._simple_new(_empty_range)
+
+        # Method hint: linear Diophantine equation
+        # solve intersection problem
+        # performance hint: for identical step sizes, could use
+        # cheaper alternative
+        gcd, s, _ = self._extended_gcd(first.step, second.step)
+
+        # check whether element sets intersect
+        if (first.start - second.start) % gcd:
+            return self._simple_new(_empty_range)
+
+        # calculate parameters for the RangeIndex describing the
+        # intersection disregarding the lower bounds
+        tmp_start = first.start + (second.start - first.start) * first.step // gcd * s
+        new_step = first.step * second.step // gcd
+
+        # adjust index to limiting interval
+        new_start = min_fitting_element(tmp_start, new_step, int_low)
+        new_range = range(new_start, int_high, new_step)
+
+        if (self.step < 0 and other.step < 0) is not (new_range.step < 0):
+            new_range = new_range[::-1]
+
+        return self._simple_new(new_range)
+
+    def _extended_gcd(self, a: int, b: int) -> tuple[int, int, int]:
+        """
+        Extended Euclidean algorithms to solve Bezout's identity:
+           a*x + b*y = gcd(x, y)
+        Finds one particular solution for x, y: s, t
+        Returns: gcd, s, t
+        """
+        s, old_s = 0, 1
+        t, old_t = 1, 0
+        r, old_r = b, a
+        while r:
+            quotient = old_r // r
+            old_r, r = r, old_r - quotient * r
+            old_s, s = s, old_s - quotient * s
+            old_t, t = t, old_t - quotient * t
+        return old_r, old_s, old_t
+
+    def _range_in_self(self, other: range) -> bool:
+        """Check if other range is contained in self"""
+        # https://stackoverflow.com/a/32481015
+        if not other:
+            return True
+        if not self._range:
+            return False
+        if len(other) > 1 and other.step % self._range.step:
+            return False
+        return other.start in self._range and other[-1] in self._range
+
+    def _union(self, other: Index, sort: bool | None):
+        """
+        Form the union of two Index objects and sorts if possible
+
+        Parameters
+        ----------
+        other : Index or array-like
+
+        sort : bool or None, default None
+            Whether to sort (monotonically increasing) the resulting index.
+            ``sort=None|True`` returns a ``RangeIndex`` if possible or a sorted
+            ``Index`` with a int64 dtype if not.
+            ``sort=False`` can return a ``RangeIndex`` if self is monotonically
+            increasing and other is fully contained in self. Otherwise, returns
+            an unsorted ``Index`` with an int64 dtype.
+
+        Returns
+        -------
+        union : Index
+        """
+        if isinstance(other, RangeIndex):
+            if sort in (None, True) or (
+                sort is False and self.step > 0 and self._range_in_self(other._range)
+            ):
+                # GH 47557: Can still return a RangeIndex
+                # if other range in self and sort=False
+                start_s, step_s = self.start, self.step
+                end_s = self.start + self.step * (len(self) - 1)
+                start_o, step_o = other.start, other.step
+                end_o = other.start + other.step * (len(other) - 1)
+                if self.step < 0:
+                    start_s, step_s, end_s = end_s, -step_s, start_s
+                if other.step < 0:
+                    start_o, step_o, end_o = end_o, -step_o, start_o
+                if len(self) == 1 and len(other) == 1:
+                    step_s = step_o = abs(self.start - other.start)
+                elif len(self) == 1:
+                    step_s = step_o
+                elif len(other) == 1:
+                    step_o = step_s
+                start_r = min(start_s, start_o)
+                end_r = max(end_s, end_o)
+                if step_o == step_s:
+                    if (
+                        (start_s - start_o) % step_s == 0
+                        and (start_s - end_o) <= step_s
+                        and (start_o - end_s) <= step_s
+                    ):
+                        return type(self)(start_r, end_r + step_s, step_s)
+                    if (
+                        (step_s % 2 == 0)
+                        and (abs(start_s - start_o) == step_s / 2)
+                        and (abs(end_s - end_o) == step_s / 2)
+                    ):
+                        # e.g. range(0, 10, 2) and range(1, 11, 2)
+                        #  but not range(0, 20, 4) and range(1, 21, 4) GH#44019
+                        return type(self)(start_r, end_r + step_s / 2, step_s / 2)
+
+                elif step_o % step_s == 0:
+                    if (
+                        (start_o - start_s) % step_s == 0
+                        and (start_o + step_s >= start_s)
+                        and (end_o - step_s <= end_s)
+                    ):
+                        return type(self)(start_r, end_r + step_s, step_s)
+                elif step_s % step_o == 0:
+                    if (
+                        (start_s - start_o) % step_o == 0
+                        and (start_s + step_o >= start_o)
+                        and (end_s - step_o <= end_o)
+                    ):
+                        return type(self)(start_r, end_r + step_o, step_o)
+
+        return super()._union(other, sort=sort)
+
+    def _difference(self, other, sort=None):
+        # optimized set operation if we have another RangeIndex
+        self._validate_sort_keyword(sort)
+        self._assert_can_do_setop(other)
+        other, result_name = self._convert_can_do_setop(other)
+
+        if not isinstance(other, RangeIndex):
+            return super()._difference(other, sort=sort)
+
+        if sort is not False and self.step < 0:
+            return self[::-1]._difference(other)
+
+        res_name = ops.get_op_result_name(self, other)
+
+        first = self._range[::-1] if self.step < 0 else self._range
+        overlap = self.intersection(other)
+        if overlap.step < 0:
+            overlap = overlap[::-1]
+
+        if len(overlap) == 0:
+            return self.rename(name=res_name)
+        if len(overlap) == len(self):
+            return self[:0].rename(res_name)
+
+        # overlap.step will always be a multiple of self.step (see _intersection)
+
+        if len(overlap) == 1:
+            if overlap[0] == self[0]:
+                return self[1:]
+
+            elif overlap[0] == self[-1]:
+                return self[:-1]
+
+            elif len(self) == 3 and overlap[0] == self[1]:
+                return self[::2]
+
+            else:
+                return super()._difference(other, sort=sort)
+
+        elif len(overlap) == 2 and overlap[0] == first[0] and overlap[-1] == first[-1]:
+            # e.g. range(-8, 20, 7) and range(13, -9, -3)
+            return self[1:-1]
+
+        if overlap.step == first.step:
+            if overlap[0] == first.start:
+                # The difference is everything after the intersection
+                new_rng = range(overlap[-1] + first.step, first.stop, first.step)
+            elif overlap[-1] == first[-1]:
+                # The difference is everything before the intersection
+                new_rng = range(first.start, overlap[0], first.step)
+            elif overlap._range == first[1:-1]:
+                # e.g. range(4) and range(1, 3)
+                step = len(first) - 1
+                new_rng = first[::step]
+            else:
+                # The difference is not range-like
+                # e.g. range(1, 10, 1) and range(3, 7, 1)
+                return super()._difference(other, sort=sort)
+
+        else:
+            # We must have len(self) > 1, bc we ruled out above
+            #  len(overlap) == 0 and len(overlap) == len(self)
+            assert len(self) > 1
+
+            if overlap.step == first.step * 2:
+                if overlap[0] == first[0] and overlap[-1] in (first[-1], first[-2]):
+                    # e.g. range(1, 10, 1) and range(1, 10, 2)
+                    new_rng = first[1::2]
+
+                elif overlap[0] == first[1] and overlap[-1] in (first[-1], first[-2]):
+                    # e.g. range(1, 10, 1) and range(2, 10, 2)
+                    new_rng = first[::2]
+
+                else:
+                    # We can get here with  e.g. range(20) and range(0, 10, 2)
+                    return super()._difference(other, sort=sort)
+
+            else:
+                # e.g. range(10) and range(0, 10, 3)
+                return super()._difference(other, sort=sort)
+
+        if first is not self._range:
+            new_rng = new_rng[::-1]
+        new_index = type(self)._simple_new(new_rng, name=res_name)
+
+        return new_index
+
+    def symmetric_difference(
+        self, other, result_name: Hashable | None = None, sort=None
+    ) -> Index:
+        if not isinstance(other, RangeIndex) or sort is not None:
+            return super().symmetric_difference(other, result_name, sort)
+
+        left = self.difference(other)
+        right = other.difference(self)
+        result = left.union(right)
+
+        if result_name is not None:
+            result = result.rename(result_name)
+        return result
+
+    def _join_empty(
+        self, other: Index, how: JoinHow, sort: bool
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        if not isinstance(other, RangeIndex) and other.dtype.kind == "i":
+            other = self._shallow_copy(other._values, name=other.name)
+        return super()._join_empty(other, how=how, sort=sort)
+
+    def _join_monotonic(
+        self, other: Index, how: JoinHow = "left"
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        # This currently only gets called for the monotonic increasing case
+        if not isinstance(other, type(self)):
+            maybe_ri = self._shallow_copy(other._values, name=other.name)
+            if not isinstance(maybe_ri, type(self)):
+                return super()._join_monotonic(other, how=how)
+            other = maybe_ri
+
+        if self.equals(other):
+            ret_index = other if how == "right" else self
+            return ret_index, None, None
+
+        if how == "left":
+            join_index = self
+            lidx = None
+            ridx = other.get_indexer(join_index)
+        elif how == "right":
+            join_index = other
+            lidx = self.get_indexer(join_index)
+            ridx = None
+        elif how == "inner":
+            join_index = self.intersection(other)
+            lidx = self.get_indexer(join_index)
+            ridx = other.get_indexer(join_index)
+        elif how == "outer":
+            join_index = self.union(other)
+            lidx = self.get_indexer(join_index)
+            ridx = other.get_indexer(join_index)
+
+        lidx = None if lidx is None else ensure_platform_int(lidx)
+        ridx = None if ridx is None else ensure_platform_int(ridx)
+        return join_index, lidx, ridx
+
+    # --------------------------------------------------------------------
+
+    # error: Return type "Index" of "delete" incompatible with return type
+    #  "RangeIndex" in supertype "Index"
+    def delete(self, loc) -> Index:  # type: ignore[override]
+        # In some cases we can retain RangeIndex, see also
+        #  DatetimeTimedeltaMixin._get_delete_Freq
+        if is_integer(loc):
+            if loc in (0, -len(self)):
+                return self[1:]
+            if loc in (-1, len(self) - 1):
+                return self[:-1]
+            if len(self) == 3 and loc in (1, -2):
+                return self[::2]
+
+        elif lib.is_list_like(loc):
+            slc = lib.maybe_indices_to_slice(np.asarray(loc, dtype=np.intp), len(self))
+
+            if isinstance(slc, slice):
+                # defer to RangeIndex._difference, which is optimized to return
+                #  a RangeIndex whenever possible
+                other = self[slc]
+                return self.difference(other, sort=False)
+
+        return super().delete(loc)
+
+    def insert(self, loc: int, item) -> Index:
+        if is_integer(item) or is_float(item):
+            # We can retain RangeIndex is inserting at the beginning or end,
+            #  or right in the middle.
+            if len(self) == 0 and loc == 0 and is_integer(item):
+                new_rng = range(item, item + self.step, self.step)
+                return type(self)._simple_new(new_rng, name=self._name)
+            elif len(self):
+                rng = self._range
+                if loc == 0 and item == self[0] - self.step:
+                    new_rng = range(rng.start - rng.step, rng.stop, rng.step)
+                    return type(self)._simple_new(new_rng, name=self._name)
+
+                elif loc == len(self) and item == self[-1] + self.step:
+                    new_rng = range(rng.start, rng.stop + rng.step, rng.step)
+                    return type(self)._simple_new(new_rng, name=self._name)
+
+                elif len(self) == 2 and item == self[0] + self.step / 2:
+                    # e.g. inserting 1 into [0, 2]
+                    step = int(self.step / 2)
+                    new_rng = range(self.start, self.stop, step)
+                    return type(self)._simple_new(new_rng, name=self._name)
+
+        return super().insert(loc, item)
+
+    def _concat(self, indexes: list[Index], name: Hashable) -> Index:
+        """
+        Overriding parent method for the case of all RangeIndex instances.
+
+        When all members of "indexes" are of type RangeIndex: result will be
+        RangeIndex if possible, Index with a int64 dtype otherwise. E.g.:
+        indexes = [RangeIndex(3), RangeIndex(3, 6)] -> RangeIndex(6)
+        indexes = [RangeIndex(3), RangeIndex(4, 6)] -> Index([0,1,2,4,5], dtype='int64')
+        """
+        if not all(isinstance(x, RangeIndex) for x in indexes):
+            result = super()._concat(indexes, name)
+            if result.dtype.kind == "i":
+                return self._shallow_copy(result._values)
+            return result
+
+        elif len(indexes) == 1:
+            return indexes[0]
+
+        rng_indexes = cast(list[RangeIndex], indexes)
+
+        start = step = next_ = None
+
+        # Filter the empty indexes
+        non_empty_indexes = []
+        all_same_index = True
+        prev: RangeIndex | None = None
+        for obj in rng_indexes:
+            if len(obj):
+                non_empty_indexes.append(obj)
+                if all_same_index:
+                    if prev is not None:
+                        all_same_index = prev.equals(obj)
+                    else:
+                        prev = obj
+
+        for obj in non_empty_indexes:
+            rng = obj._range
+
+            if start is None:
+                # This is set by the first non-empty index
+                start = rng.start
+                if step is None and len(rng) > 1:
+                    step = rng.step
+            elif step is None:
+                # First non-empty index had only one element
+                if rng.start == start:
+                    if all_same_index:
+                        values = np.tile(
+                            non_empty_indexes[0]._values, len(non_empty_indexes)
+                        )
+                    else:
+                        values = np.concatenate([x._values for x in rng_indexes])
+                    result = self._constructor(values)
+                    return result.rename(name)
+
+                step = rng.start - start
+
+            non_consecutive = (step != rng.step and len(rng) > 1) or (
+                next_ is not None and rng.start != next_
+            )
+            if non_consecutive:
+                if all_same_index:
+                    values = np.tile(
+                        non_empty_indexes[0]._values, len(non_empty_indexes)
+                    )
+                else:
+                    values = np.concatenate([x._values for x in rng_indexes])
+                result = self._constructor(values)
+                return result.rename(name)
+
+            if step is not None:
+                next_ = rng[-1] + step
+
+        if non_empty_indexes:
+            # Get the stop value from "next" or alternatively
+            # from the last non-empty index
+            stop = non_empty_indexes[-1].stop if next_ is None else next_
+            if len(non_empty_indexes) == 1:
+                step = non_empty_indexes[0].step
+            return RangeIndex(start, stop, step, name=name)
+
+        # Here all "indexes" had 0 length, i.e. were empty.
+        # In this case return an empty range index.
+        return RangeIndex(_empty_range, name=name)
+
+    def __len__(self) -> int:
+        """
+        return the length of the RangeIndex
+        """
+        return len(self._range)
+
+    @property
+    def size(self) -> int:
+        return len(self)
+
+    def __getitem__(self, key):
+        """
+        Conserve RangeIndex type for scalar and slice keys.
+        """
+        if key is Ellipsis:
+            key = slice(None)
+        if isinstance(key, slice):
+            return self._getitem_slice(key)
+        elif is_integer(key):
+            new_key = int(key)
+            try:
+                return self._range[new_key]
+            except IndexError as err:
+                raise IndexError(
+                    f"index {key} is out of bounds for axis 0 with size {len(self)}"
+                ) from err
+        elif is_scalar(key):
+            raise IndexError(
+                "only integers, slices (`:`), "
+                "ellipsis (`...`), numpy.newaxis (`None`) "
+                "and integer or boolean "
+                "arrays are valid indices"
+            )
+        elif com.is_bool_indexer(key):
+            if isinstance(getattr(key, "dtype", None), ExtensionDtype):
+                key = key.to_numpy(dtype=bool, na_value=False)
+            else:
+                key = np.asarray(key, dtype=bool)
+            check_array_indexer(self._range, key)  # type: ignore[arg-type]
+            key = np.flatnonzero(key)
+        try:
+            return self.take(key)
+        except (TypeError, ValueError):
+            return super().__getitem__(key)
+
+    def _getitem_slice(self, slobj: slice) -> Self:
+        """
+        Fastpath for __getitem__ when we know we have a slice.
+        """
+        res = self._range[slobj]
+        return type(self)._simple_new(res, name=self._name)
+
+    @unpack_zerodim_and_defer("__floordiv__")
+    def __floordiv__(self, other):
+        if is_integer(other) and other != 0:
+            if len(self) == 0 or (self.start % other == 0 and self.step % other == 0):
+                start = self.start // other
+                step = self.step // other
+                stop = start + len(self) * step
+                new_range = range(start, stop, step or 1)
+                return self._simple_new(new_range, name=self._name)
+            if len(self) == 1:
+                start = self.start // other
+                new_range = range(start, start + 1, 1)
+                return self._simple_new(new_range, name=self._name)
+
+        return super().__floordiv__(other)
+
+    # --------------------------------------------------------------------
+    # Reductions
+
+    def all(self, *args, **kwargs) -> bool:
+        return 0 not in self._range
+
+    def any(self, *args, **kwargs) -> bool:
+        return any(self._range)
+
+    # --------------------------------------------------------------------
+
+    # error: Return type "RangeIndex | Index" of "round" incompatible with
+    # return type "RangeIndex" in supertype "Index"
+    def round(self, decimals: int = 0) -> Self | Index:  # type: ignore[override]
+        """
+        Round each value in the Index to the given number of decimals.
+
+        Parameters
+        ----------
+        decimals : int, optional
+            Number of decimal places to round to. If decimals is negative,
+            it specifies the number of positions to the left of the decimal point
+            e.g. ``round(11.0, -1) == 10.0``.
+
+        Returns
+        -------
+        Index or RangeIndex
+            A new Index with the rounded values.
+
+        Examples
+        --------
+        >>> import pandas as pd
+        >>> idx = pd.RangeIndex(10, 30, 10)
+        >>> idx.round(decimals=-1)
+        RangeIndex(start=10, stop=30, step=10)
+        >>> idx = pd.RangeIndex(10, 15, 1)
+        >>> idx.round(decimals=-1)
+        Index([10, 10, 10, 10, 10], dtype='int64')
+        """
+        if decimals >= 0:
+            return self.copy()
+        elif self.start % 10**-decimals == 0 and self.step % 10**-decimals == 0:
+            # e.g. RangeIndex(10, 30, 10).round(-1) doesn't need rounding
+            return self.copy()
+        else:
+            return super().round(decimals=decimals)
+
+    def _cmp_method(self, other, op):
+        if isinstance(other, RangeIndex) and self._range == other._range:
+            # Both are immutable so if ._range attr. are equal, shortcut is possible
+            return super()._cmp_method(self, op)
+        return super()._cmp_method(other, op)
+
+    def _arith_method(self, other, op):
+        """
+        Parameters
+        ----------
+        other : Any
+        op : callable that accepts 2 params
+            perform the binary op
+        """
+
+        if isinstance(other, ABCTimedeltaIndex):
+            # Defer to TimedeltaIndex implementation
+            return NotImplemented
+        elif isinstance(other, (timedelta, np.timedelta64)):
+            # GH#19333 is_integer evaluated True on timedelta64,
+            # so we need to catch these explicitly
+            return super()._arith_method(other, op)
+        elif lib.is_np_dtype(getattr(other, "dtype", None), "m"):
+            # Must be an np.ndarray; GH#22390
+            return super()._arith_method(other, op)
+
+        if op in [
+            operator.pow,
+            ops.rpow,
+            operator.mod,
+            ops.rmod,
+            operator.floordiv,
+            ops.rfloordiv,
+            divmod,
+            ops.rdivmod,
+        ]:
+            return super()._arith_method(other, op)
+
+        step: Callable | None = None
+        if op in [operator.mul, ops.rmul, operator.truediv, ops.rtruediv]:
+            step = op
+
+        # TODO: if other is a RangeIndex we may have more efficient options
+        right = extract_array(other, extract_numpy=True, extract_range=True)
+        left = self
+
+        try:
+            # apply if we have an override
+            if step:
+                with np.errstate(all="ignore"):
+                    rstep = step(left.step, right)
+
+                # we don't have a representable op
+                # so return a base index
+                if not is_integer(rstep) or not rstep:
+                    raise ValueError
+
+            # GH#53255
+            else:
+                rstep = -left.step if op == ops.rsub else left.step
+
+            with np.errstate(all="ignore"):
+                rstart = op(left.start, right)
+                rstop = op(left.stop, right)
+
+            res_name = ops.get_op_result_name(self, other)
+            result = type(self)(rstart, rstop, rstep, name=res_name)
+
+            # for compat with numpy / Index with int64 dtype
+            # even if we can represent as a RangeIndex, return
+            # as a float64 Index if we have float-like descriptors
+            if not all(is_integer(x) for x in [rstart, rstop, rstep]):
+                result = result.astype("float64")
+
+            return result
+
+        except (ValueError, TypeError, ZeroDivisionError):
+            # test_arithmetic_explicit_conversions
+            return super()._arith_method(other, op)
+
+    def __abs__(self) -> Self | Index:
+        if len(self) == 0 or self.min() >= 0:
+            return self.copy()
+        elif self.max() <= 0:
+            return -self
+        else:
+            return super().__abs__()
+
+    def __neg__(self) -> Self:
+        rng = range(-self.start, -self.stop, -self.step)
+        return self._simple_new(rng, name=self.name)
+
+    def __pos__(self) -> Self:
+        return self.copy()
+
+    def __invert__(self) -> Self:
+        if len(self) == 0:
+            return self.copy()
+        rng = range(~self.start, ~self.stop, -self.step)
+        return self._simple_new(rng, name=self.name)
+
+    # error: Return type "Index" of "take" incompatible with return type
+    # "RangeIndex" in supertype "Index"
+    def take(  # type: ignore[override]
+        self,
+        indices,
+        axis: Axis = 0,
+        allow_fill: bool = True,
+        fill_value=None,
+        **kwargs,
+    ) -> Self | Index:
+        if kwargs:
+            nv.validate_take((), kwargs)
+        if is_scalar(indices):
+            raise TypeError("Expected indices to be array-like")
+        indices = ensure_platform_int(indices)
+
+        # raise an exception if allow_fill is True and fill_value is not None
+        self._maybe_disallow_fill(allow_fill, fill_value, indices)
+
+        if len(indices) == 0:
+            return type(self)(_empty_range, name=self.name)
+        else:
+            ind_max = indices.max()
+            if ind_max >= len(self):
+                raise IndexError(
+                    f"index {ind_max} is out of bounds for axis 0 with size {len(self)}"
+                )
+            ind_min = indices.min()
+            if ind_min < -len(self):
+                raise IndexError(
+                    f"index {ind_min} is out of bounds for axis 0 with size {len(self)}"
+                )
+            taken = indices.astype(self.dtype, casting="safe")
+            if ind_min < 0:
+                taken %= len(self)
+            if self.step != 1:
+                taken *= self.step
+            if self.start != 0:
+                taken += self.start
+
+        return self._shallow_copy(taken, name=self.name)
+
+    def value_counts(
+        self,
+        normalize: bool = False,
+        sort: bool = True,
+        ascending: bool = False,
+        bins=None,
+        dropna: bool = True,
+    ) -> Series:
+        from pandas import Series
+
+        if bins is not None:
+            return super().value_counts(
+                normalize=normalize,
+                sort=sort,
+                ascending=ascending,
+                bins=bins,
+                dropna=dropna,
+            )
+        name = "proportion" if normalize else "count"
+        data: npt.NDArray[np.floating] | npt.NDArray[np.signedinteger] = np.ones(
+            len(self), dtype=np.int64
+        )
+        if normalize:
+            data = data / len(self)
+        return Series(data, index=self.copy(), name=name)
+
+    def searchsorted(  # type: ignore[override]
+        self,
+        value,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        if side not in {"left", "right"} or sorter is not None:
+            return super().searchsorted(value=value, side=side, sorter=sorter)
+
+        was_scalar = False
+        if is_scalar(value):
+            was_scalar = True
+            array_value = np.array([value])
+        else:
+            array_value = np.asarray(value)
+        if array_value.dtype.kind not in "iu":
+            return super().searchsorted(value=value, side=side, sorter=sorter)
+
+        if flip := (self.step < 0):
+            rng = self._range[::-1]
+            start = rng.start
+            step = rng.step
+            shift = side == "right"
+        else:
+            start = self.start
+            step = self.step
+            shift = side == "left"
+        result = (array_value - start - int(shift)) // step + 1
+        if flip:
+            result = len(self) - result
+        result = np.maximum(np.minimum(result, len(self)), 0)
+        if was_scalar:
+            return np.intp(result.item())
+        return result.astype(np.intp, copy=False)
diff --git a/pandas/pandas/core/indexes/timedeltas.py b/pandas/pandas/core/indexes/timedeltas.py
new file mode 100644
index 0000000000000..fa3de46621643
--- /dev/null
+++ b/pandas/pandas/core/indexes/timedeltas.py
@@ -0,0 +1,337 @@
+"""implement the TimedeltaIndex"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas._libs import (
+    index as libindex,
+    lib,
+)
+from pandas._libs.tslibs import (
+    Resolution,
+    Timedelta,
+    to_offset,
+)
+from pandas.util._decorators import set_module
+
+from pandas.core.dtypes.common import (
+    is_scalar,
+    pandas_dtype,
+)
+from pandas.core.dtypes.generic import ABCSeries
+
+from pandas.core.arrays.timedeltas import TimedeltaArray
+import pandas.core.common as com
+from pandas.core.indexes.base import (
+    Index,
+    maybe_extract_name,
+)
+from pandas.core.indexes.datetimelike import DatetimeTimedeltaMixin
+from pandas.core.indexes.extension import inherit_names
+
+if TYPE_CHECKING:
+    from pandas._libs import NaTType
+    from pandas._typing import DtypeObj
+
+
+@inherit_names(
+    ["__neg__", "__pos__", "__abs__", "total_seconds", "round", "floor", "ceil"]
+    + TimedeltaArray._field_ops,
+    TimedeltaArray,
+    wrap=True,
+)
+@inherit_names(
+    [
+        "components",
+        "to_pytimedelta",
+        "sum",
+        "std",
+        "median",
+    ],
+    TimedeltaArray,
+)
+@set_module("pandas")
+class TimedeltaIndex(DatetimeTimedeltaMixin):
+    """
+    Immutable Index of timedelta64 data.
+
+    Represented internally as int64, and scalars returned Timedelta objects.
+
+    Parameters
+    ----------
+    data : array-like (1-dimensional), optional
+        Optional timedelta-like data to construct index with.
+    freq : str or pandas offset object, optional
+        One of pandas date offset strings or corresponding objects. The string
+        ``'infer'`` can be passed in order to set the frequency of the index as
+        the inferred frequency upon creation.
+    dtype : numpy.dtype or str, default None
+        Valid ``numpy`` dtypes are ``timedelta64[ns]``, ``timedelta64[us]``,
+        ``timedelta64[ms]``, and ``timedelta64[s]``.
+    copy : bool
+        Make a copy of input array.
+    name : object
+        Name to be stored in the index.
+
+    Attributes
+    ----------
+    days
+    seconds
+    microseconds
+    nanoseconds
+    components
+    inferred_freq
+
+    Methods
+    -------
+    to_pytimedelta
+    to_series
+    round
+    floor
+    ceil
+    to_frame
+    mean
+
+    See Also
+    --------
+    Index : The base pandas Index type.
+    Timedelta : Represents a duration between two dates or times.
+    DatetimeIndex : Index of datetime64 data.
+    PeriodIndex : Index of Period data.
+    timedelta_range : Create a fixed-frequency TimedeltaIndex.
+
+    Notes
+    -----
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    >>> pd.TimedeltaIndex(["0 days", "1 days", "2 days", "3 days", "4 days"])
+    TimedeltaIndex(['0 days', '1 days', '2 days', '3 days', '4 days'],
+                   dtype='timedelta64[ns]', freq=None)
+
+    We can also let pandas infer the frequency when possible.
+
+    >>> pd.TimedeltaIndex(np.arange(5) * 24 * 3600 * 1e9, freq="infer")
+    TimedeltaIndex(['0 days', '1 days', '2 days', '3 days', '4 days'],
+                   dtype='timedelta64[ns]', freq='D')
+    """
+
+    _typ = "timedeltaindex"
+
+    _data_cls = TimedeltaArray
+
+    @property
+    def _engine_type(self) -> type[libindex.TimedeltaEngine]:
+        return libindex.TimedeltaEngine
+
+    _data: TimedeltaArray
+
+    # Use base class method instead of DatetimeTimedeltaMixin._get_string_slice
+    _get_string_slice = Index._get_string_slice
+
+    # error: Signature of "_resolution_obj" incompatible with supertype
+    # "DatetimeIndexOpsMixin"
+    @property
+    def _resolution_obj(self) -> Resolution | None:  # type: ignore[override]
+        return self._data._resolution_obj
+
+    # -------------------------------------------------------------------
+    # Constructors
+
+    def __new__(
+        cls,
+        data=None,
+        freq=lib.no_default,
+        dtype=None,
+        copy: bool = False,
+        name=None,
+    ):
+        name = maybe_extract_name(name, data, cls)
+
+        if is_scalar(data):
+            cls._raise_scalar_data_error(data)
+
+        if dtype is not None:
+            dtype = pandas_dtype(dtype)
+
+        if (
+            isinstance(data, TimedeltaArray)
+            and freq is lib.no_default
+            and (dtype is None or dtype == data.dtype)
+        ):
+            if copy:
+                data = data.copy()
+            return cls._simple_new(data, name=name)
+
+        if (
+            isinstance(data, TimedeltaIndex)
+            and freq is lib.no_default
+            and name is None
+            and (dtype is None or dtype == data.dtype)
+        ):
+            if copy:
+                return data.copy()
+            else:
+                return data._view()
+
+        # - Cases checked above all return/raise before reaching here - #
+
+        tdarr = TimedeltaArray._from_sequence_not_strict(
+            data, freq=freq, unit=None, dtype=dtype, copy=copy
+        )
+        refs = None
+        if not copy and isinstance(data, (ABCSeries, Index)):
+            refs = data._references
+
+        return cls._simple_new(tdarr, name=name, refs=refs)
+
+    # -------------------------------------------------------------------
+
+    def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
+        """
+        Can we compare values of the given dtype to our own?
+        """
+        return lib.is_np_dtype(dtype, "m")  # aka self._data._is_recognized_dtype
+
+    # -------------------------------------------------------------------
+    # Indexing Methods
+
+    def get_loc(self, key):
+        """
+        Get integer location for requested label
+
+        Returns
+        -------
+        loc : int, slice, or ndarray[int]
+        """
+        self._check_indexing_error(key)
+
+        try:
+            key = self._data._validate_scalar(key, unbox=False)
+        except TypeError as err:
+            raise KeyError(key) from err
+
+        return Index.get_loc(self, key)
+
+    # error: Return type "tuple[Timedelta | NaTType, None]" of "_parse_with_reso"
+    # incompatible with return type "tuple[datetime, Resolution]" in supertype
+    # "DatetimeIndexOpsMixin"
+    def _parse_with_reso(self, label: str) -> tuple[Timedelta | NaTType, None]:  # type: ignore[override]
+        # the "with_reso" is a no-op for TimedeltaIndex
+        parsed = Timedelta(label)
+        return parsed, None
+
+    def _parsed_string_to_bounds(self, reso, parsed: Timedelta):
+        # reso is unused, included to match signature of DTI/PI
+        lbound = parsed.round(parsed.resolution_string)
+        rbound = lbound + to_offset(parsed.resolution_string) - Timedelta(1, "ns")
+        return lbound, rbound
+
+    # -------------------------------------------------------------------
+
+    @property
+    def inferred_type(self) -> str:
+        return "timedelta64"
+
+
+@set_module("pandas")
+def timedelta_range(
+    start=None,
+    end=None,
+    periods: int | None = None,
+    freq=None,
+    name=None,
+    closed=None,
+    *,
+    unit: str | None = None,
+) -> TimedeltaIndex:
+    """
+    Return a fixed frequency TimedeltaIndex with day as the default.
+
+    Parameters
+    ----------
+    start : str or timedelta-like, default None
+        Left bound for generating timedeltas.
+    end : str or timedelta-like, default None
+        Right bound for generating timedeltas.
+    periods : int, default None
+        Number of periods to generate.
+    freq : str, Timedelta, datetime.timedelta, or DateOffset, default 'D'
+        Frequency strings can have multiples, e.g. '5h'.
+    name : str, default None
+        Name of the resulting TimedeltaIndex.
+    closed : str, default None
+        Make the interval closed with respect to the given frequency to
+        the 'left', 'right', or both sides (None).
+    unit : str, default None
+        Specify the desired resolution of the result.
+
+        .. versionadded:: 2.0.0
+
+    Returns
+    -------
+    TimedeltaIndex
+        Fixed frequency, with day as the default.
+
+    See Also
+    --------
+    date_range : Return a fixed frequency DatetimeIndex.
+    period_range : Return a fixed frequency PeriodIndex.
+
+    Notes
+    -----
+    Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
+    exactly three must be specified. If ``freq`` is omitted, the resulting
+    ``TimedeltaIndex`` will have ``periods`` linearly spaced elements between
+    ``start`` and ``end`` (closed on both sides).
+
+    To learn more about the frequency strings, please see
+    :ref:`this link<timeseries.offset_aliases>`.
+
+    Examples
+    --------
+    >>> pd.timedelta_range(start="1 day", periods=4)
+    TimedeltaIndex(['1 days', '2 days', '3 days', '4 days'],
+                   dtype='timedelta64[ns]', freq='D')
+
+    The ``closed`` parameter specifies which endpoint is included.  The default
+    behavior is to include both endpoints.
+
+    >>> pd.timedelta_range(start="1 day", periods=4, closed="right")
+    TimedeltaIndex(['2 days', '3 days', '4 days'],
+                   dtype='timedelta64[ns]', freq='D')
+
+    The ``freq`` parameter specifies the frequency of the TimedeltaIndex.
+    Only fixed frequencies can be passed, non-fixed frequencies such as
+    'M' (month end) will raise.
+
+    >>> pd.timedelta_range(start="1 day", end="2 days", freq="6h")
+    TimedeltaIndex(['1 days 00:00:00', '1 days 06:00:00', '1 days 12:00:00',
+                    '1 days 18:00:00', '2 days 00:00:00'],
+                   dtype='timedelta64[ns]', freq='6h')
+
+    Specify ``start``, ``end``, and ``periods``; the frequency is generated
+    automatically (linearly spaced).
+
+    >>> pd.timedelta_range(start="1 day", end="5 days", periods=4)
+    TimedeltaIndex(['1 days 00:00:00', '2 days 08:00:00', '3 days 16:00:00',
+                    '5 days 00:00:00'],
+                   dtype='timedelta64[ns]', freq=None)
+
+    **Specify a unit**
+
+    >>> pd.timedelta_range("1 Day", periods=3, freq="100000D", unit="s")
+    TimedeltaIndex(['1 days', '100001 days', '200001 days'],
+                   dtype='timedelta64[s]', freq='100000D')
+    """
+    if freq is None and com.any_none(periods, start, end):
+        freq = "D"
+
+    freq = to_offset(freq)
+    tdarr = TimedeltaArray._generate_range(
+        start, end, periods, freq, closed=closed, unit=unit
+    )
+    return TimedeltaIndex._simple_new(tdarr, name=name)
diff --git a/pandas/pandas/core/indexing.py b/pandas/pandas/core/indexing.py
new file mode 100644
index 0000000000000..8a493fef54d3b
--- /dev/null
+++ b/pandas/pandas/core/indexing.py
@@ -0,0 +1,2757 @@
+from __future__ import annotations
+
+from contextlib import suppress
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypeVar,
+    cast,
+    final,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs.indexing import NDFrameIndexerBase
+from pandas._libs.lib import item_from_zerodim
+from pandas.compat import PYPY
+from pandas.errors import (
+    AbstractMethodError,
+    ChainedAssignmentError,
+    IndexingError,
+    InvalidIndexError,
+    LossySetitemError,
+)
+from pandas.errors.cow import _chained_assignment_msg
+from pandas.util._decorators import doc
+
+from pandas.core.dtypes.cast import (
+    can_hold_element,
+    maybe_promote,
+)
+from pandas.core.dtypes.common import (
+    is_array_like,
+    is_bool_dtype,
+    is_hashable,
+    is_integer,
+    is_iterator,
+    is_list_like,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_scalar,
+    is_sequence,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    construct_1d_array_from_inferred_fill_value,
+    infer_fill_value,
+    is_valid_na_for_dtype,
+    isna,
+    na_value_for_dtype,
+)
+
+from pandas.core import algorithms as algos
+import pandas.core.common as com
+from pandas.core.construction import (
+    array as pd_array,
+    extract_array,
+)
+from pandas.core.indexers import (
+    check_array_indexer,
+    is_list_like_indexer,
+    is_scalar_indexer,
+    length_of_indexer,
+)
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        Axis,
+        AxisInt,
+        Self,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+T = TypeVar("T")
+# "null slice"
+_NS = slice(None, None)
+_one_ellipsis_message = "indexer may only contain one '...' entry"
+
+
+# the public IndexSlicerMaker
+class _IndexSlice:
+    """
+    Create an object to more easily perform multi-index slicing.
+
+    See Also
+    --------
+    MultiIndex.remove_unused_levels : New MultiIndex with no unused levels.
+
+    Notes
+    -----
+    See :ref:`Defined Levels <advanced.shown_levels>`
+    for further info on slicing a MultiIndex.
+
+    Examples
+    --------
+    >>> midx = pd.MultiIndex.from_product([["A0", "A1"], ["B0", "B1", "B2", "B3"]])
+    >>> columns = ["foo", "bar"]
+    >>> dfmi = pd.DataFrame(
+    ...     np.arange(16).reshape((len(midx), len(columns))),
+    ...     index=midx,
+    ...     columns=columns,
+    ... )
+
+    Using the default slice command:
+
+    >>> dfmi.loc[(slice(None), slice("B0", "B1")), :]
+               foo  bar
+        A0 B0    0    1
+           B1    2    3
+        A1 B0    8    9
+           B1   10   11
+
+    Using the IndexSlice class for a more intuitive command:
+
+    >>> idx = pd.IndexSlice
+    >>> dfmi.loc[idx[:, "B0":"B1"], :]
+               foo  bar
+        A0 B0    0    1
+           B1    2    3
+        A1 B0    8    9
+           B1   10   11
+    """
+
+    def __getitem__(self, arg):
+        return arg
+
+
+IndexSlice = _IndexSlice()
+
+
+class IndexingMixin:
+    """
+    Mixin for adding .loc/.iloc/.at/.iat to Dataframes and Series.
+    """
+
+    @property
+    def iloc(self) -> _iLocIndexer:
+        """
+        Purely integer-location based indexing for selection by position.
+
+        .. versionchanged:: 3.0
+
+           Callables which return a tuple are deprecated as input.
+
+        ``.iloc[]`` is primarily integer position based (from ``0`` to
+        ``length-1`` of the axis), but may also be used with a boolean
+        array.
+
+        Allowed inputs are:
+
+        - An integer, e.g. ``5``.
+        - A list or array of integers, e.g. ``[4, 3, 0]``.
+        - A slice object with ints, e.g. ``1:7``.
+        - A boolean array.
+        - A ``callable`` function with one argument (the calling Series or
+          DataFrame) and that returns valid output for indexing (one of the above).
+          This is useful in method chains, when you don't have a reference to the
+          calling object, but would like to base your selection on
+          some value.
+        - A tuple of row and column indexes. The tuple elements consist of one of the
+          above inputs, e.g. ``(0, 1)``.
+
+        ``.iloc`` will raise ``IndexError`` if a requested indexer is
+        out-of-bounds, except *slice* indexers which allow out-of-bounds
+        indexing (this conforms with python/numpy *slice* semantics).
+
+        See more at :ref:`Selection by Position <indexing.integer>`.
+
+        See Also
+        --------
+        DataFrame.iat : Fast integer location scalar accessor.
+        DataFrame.loc : Purely label-location based indexer for selection by label.
+        Series.iloc : Purely integer-location based indexing for
+                       selection by position.
+
+        Examples
+        --------
+        >>> mydict = [
+        ...     {"a": 1, "b": 2, "c": 3, "d": 4},
+        ...     {"a": 100, "b": 200, "c": 300, "d": 400},
+        ...     {"a": 1000, "b": 2000, "c": 3000, "d": 4000},
+        ... ]
+        >>> df = pd.DataFrame(mydict)
+        >>> df
+              a     b     c     d
+        0     1     2     3     4
+        1   100   200   300   400
+        2  1000  2000  3000  4000
+
+        **Indexing just the rows**
+
+        With a scalar integer.
+
+        >>> type(df.iloc[0])
+        <class 'pandas.Series'>
+        >>> df.iloc[0]
+        a    1
+        b    2
+        c    3
+        d    4
+        Name: 0, dtype: int64
+
+        With a list of integers.
+
+        >>> df.iloc[[0]]
+           a  b  c  d
+        0  1  2  3  4
+        >>> type(df.iloc[[0]])
+        <class 'pandas.DataFrame'>
+
+        >>> df.iloc[[0, 1]]
+             a    b    c    d
+        0    1    2    3    4
+        1  100  200  300  400
+
+        With a `slice` object.
+
+        >>> df.iloc[:3]
+              a     b     c     d
+        0     1     2     3     4
+        1   100   200   300   400
+        2  1000  2000  3000  4000
+
+        With a boolean mask the same length as the index.
+
+        >>> df.iloc[[True, False, True]]
+              a     b     c     d
+        0     1     2     3     4
+        2  1000  2000  3000  4000
+
+        With a callable, useful in method chains. The `x` passed
+        to the ``lambda`` is the DataFrame being sliced. This selects
+        the rows whose index label even.
+
+        >>> df.iloc[lambda x: x.index % 2 == 0]
+              a     b     c     d
+        0     1     2     3     4
+        2  1000  2000  3000  4000
+
+        **Indexing both axes**
+
+        You can mix the indexer types for the index and columns. Use ``:`` to
+        select the entire axis.
+
+        With scalar integers.
+
+        >>> df.iloc[0, 1]
+        2
+
+        With lists of integers.
+
+        >>> df.iloc[[0, 2], [1, 3]]
+              b     d
+        0     2     4
+        2  2000  4000
+
+        With `slice` objects.
+
+        >>> df.iloc[1:3, 0:3]
+              a     b     c
+        1   100   200   300
+        2  1000  2000  3000
+
+        With a boolean array whose length matches the columns.
+
+        >>> df.iloc[:, [True, False, True, False]]
+              a     c
+        0     1     3
+        1   100   300
+        2  1000  3000
+
+        With a callable function that expects the Series or DataFrame.
+
+        >>> df.iloc[:, lambda df: [0, 2]]
+              a     c
+        0     1     3
+        1   100   300
+        2  1000  3000
+        """
+        return _iLocIndexer("iloc", self)
+
+    @property
+    def loc(self) -> _LocIndexer:
+        """
+        Access a group of rows and columns by label(s) or a boolean array.
+
+        ``.loc[]`` is primarily label based, but may also be used with a
+        boolean array.
+
+        Allowed inputs are:
+
+        - A single label, e.g. ``5`` or ``'a'``, (note that ``5`` is
+          interpreted as a *label* of the index, and **never** as an
+          integer position along the index).
+        - A list or array of labels, e.g. ``['a', 'b', 'c']``.
+        - A slice object with labels, e.g. ``'a':'f'``.
+
+          .. warning:: Note that contrary to usual python slices, **both** the
+              start and the stop are included
+
+        - A boolean array of the same length as the axis being sliced,
+          e.g. ``[True, False, True]``.
+        - An alignable boolean Series. The index of the key will be aligned before
+          masking.
+        - An alignable Index. The Index of the returned selection will be the input.
+        - A ``callable`` function with one argument (the calling Series or
+          DataFrame) and that returns valid output for indexing (one of the above)
+
+        See more at :ref:`Selection by Label <indexing.label>`.
+
+        Raises
+        ------
+        KeyError
+            If any items are not found.
+        IndexingError
+            If an indexed key is passed and its index is unalignable to the frame index.
+
+        See Also
+        --------
+        DataFrame.at : Access a single value for a row/column label pair.
+        DataFrame.iloc : Access group of rows and columns by integer position(s).
+        DataFrame.xs : Returns a cross-section (row(s) or column(s)) from the
+                       Series/DataFrame.
+        Series.loc : Access group of values using labels.
+
+        Examples
+        --------
+        **Getting values**
+
+        >>> df = pd.DataFrame(
+        ...     [[1, 2], [4, 5], [7, 8]],
+        ...     index=["cobra", "viper", "sidewinder"],
+        ...     columns=["max_speed", "shield"],
+        ... )
+        >>> df
+                    max_speed  shield
+        cobra               1       2
+        viper               4       5
+        sidewinder          7       8
+
+        Single label. Note this returns the row as a Series.
+
+        >>> df.loc["viper"]
+        max_speed    4
+        shield       5
+        Name: viper, dtype: int64
+
+        List of labels. Note using ``[[]]`` returns a DataFrame.
+
+        >>> df.loc[["viper", "sidewinder"]]
+                    max_speed  shield
+        viper               4       5
+        sidewinder          7       8
+
+        Single label for row and column
+
+        >>> df.loc["cobra", "shield"]
+        2
+
+        Slice with labels for row and single label for column. As mentioned
+        above, note that both the start and stop of the slice are included.
+
+        >>> df.loc["cobra":"viper", "max_speed"]
+        cobra    1
+        viper    4
+        Name: max_speed, dtype: int64
+
+        Boolean list with the same length as the row axis
+
+        >>> df.loc[[False, False, True]]
+                    max_speed  shield
+        sidewinder          7       8
+
+        Alignable boolean Series:
+
+        >>> df.loc[
+        ...     pd.Series([False, True, False], index=["viper", "sidewinder", "cobra"])
+        ... ]
+                             max_speed  shield
+        sidewinder          7       8
+
+        Index (same behavior as ``df.reindex``)
+
+        >>> df.loc[pd.Index(["cobra", "viper"], name="foo")]
+               max_speed  shield
+        foo
+        cobra          1       2
+        viper          4       5
+
+        Conditional that returns a boolean Series
+
+        >>> df.loc[df["shield"] > 6]
+                    max_speed  shield
+        sidewinder          7       8
+
+        Conditional that returns a boolean Series with column labels specified
+
+        >>> df.loc[df["shield"] > 6, ["max_speed"]]
+                    max_speed
+        sidewinder          7
+
+        Multiple conditional using ``&`` that returns a boolean Series
+
+        >>> df.loc[(df["max_speed"] > 1) & (df["shield"] < 8)]
+                    max_speed  shield
+        viper          4       5
+
+        Multiple conditional using ``|`` that returns a boolean Series
+
+        >>> df.loc[(df["max_speed"] > 4) | (df["shield"] < 5)]
+                    max_speed  shield
+        cobra               1       2
+        sidewinder          7       8
+
+        Please ensure that each condition is wrapped in parentheses ``()``.
+        See the :ref:`user guide<indexing.boolean>`
+        for more details and explanations of Boolean indexing.
+
+        .. note::
+            If you find yourself using 3 or more conditionals in ``.loc[]``,
+            consider using :ref:`advanced indexing<advanced.advanced_hierarchical>`.
+
+            See below for using ``.loc[]`` on MultiIndex DataFrames.
+
+        Callable that returns a boolean Series
+
+        >>> df.loc[lambda df: df["shield"] == 8]
+                    max_speed  shield
+        sidewinder          7       8
+
+        **Setting values**
+
+        Set value for all items matching the list of labels
+
+        >>> df.loc[["viper", "sidewinder"], ["shield"]] = 50
+        >>> df
+                    max_speed  shield
+        cobra               1       2
+        viper               4      50
+        sidewinder          7      50
+
+        Set value for an entire row
+
+        >>> df.loc["cobra"] = 10
+        >>> df
+                    max_speed  shield
+        cobra              10      10
+        viper               4      50
+        sidewinder          7      50
+
+        Set value for an entire column
+
+        >>> df.loc[:, "max_speed"] = 30
+        >>> df
+                    max_speed  shield
+        cobra              30      10
+        viper              30      50
+        sidewinder         30      50
+
+        Set value for rows matching callable condition
+
+        >>> df.loc[df["shield"] > 35] = 0
+        >>> df
+                    max_speed  shield
+        cobra              30      10
+        viper               0       0
+        sidewinder          0       0
+
+        Add value matching location
+
+        >>> df.loc["viper", "shield"] += 5
+        >>> df
+                    max_speed  shield
+        cobra              30      10
+        viper               0       5
+        sidewinder          0       0
+
+        Setting using a ``Series`` or a ``DataFrame`` sets the values matching the
+        index labels, not the index positions.
+
+        >>> shuffled_df = df.loc[["viper", "cobra", "sidewinder"]]
+        >>> df.loc[:] += shuffled_df
+        >>> df
+                    max_speed  shield
+        cobra              60      20
+        viper               0      10
+        sidewinder          0       0
+
+        **Getting values on a DataFrame with an index that has integer labels**
+
+        Another example using integers for the index
+
+        >>> df = pd.DataFrame(
+        ...     [[1, 2], [4, 5], [7, 8]],
+        ...     index=[7, 8, 9],
+        ...     columns=["max_speed", "shield"],
+        ... )
+        >>> df
+           max_speed  shield
+        7          1       2
+        8          4       5
+        9          7       8
+
+        Slice with integer labels for rows. As mentioned above, note that both
+        the start and stop of the slice are included.
+
+        >>> df.loc[7:9]
+           max_speed  shield
+        7          1       2
+        8          4       5
+        9          7       8
+
+        **Getting values with a MultiIndex**
+
+        A number of examples using a DataFrame with a MultiIndex
+
+        >>> tuples = [
+        ...     ("cobra", "mark i"),
+        ...     ("cobra", "mark ii"),
+        ...     ("sidewinder", "mark i"),
+        ...     ("sidewinder", "mark ii"),
+        ...     ("viper", "mark ii"),
+        ...     ("viper", "mark iii"),
+        ... ]
+        >>> index = pd.MultiIndex.from_tuples(tuples)
+        >>> values = [[12, 2], [0, 4], [10, 20], [1, 4], [7, 1], [16, 36]]
+        >>> df = pd.DataFrame(values, columns=["max_speed", "shield"], index=index)
+        >>> df
+                             max_speed  shield
+        cobra      mark i           12       2
+                   mark ii           0       4
+        sidewinder mark i           10      20
+                   mark ii           1       4
+        viper      mark ii           7       1
+                   mark iii         16      36
+
+        Single label. Note this returns a DataFrame with a single index.
+
+        >>> df.loc["cobra"]
+                 max_speed  shield
+        mark i          12       2
+        mark ii          0       4
+
+        Single index tuple. Note this returns a Series.
+
+        >>> df.loc[("cobra", "mark ii")]
+        max_speed    0
+        shield       4
+        Name: (cobra, mark ii), dtype: int64
+
+        Single label for row and column. Similar to passing in a tuple, this
+        returns a Series.
+
+        >>> df.loc["cobra", "mark i"]
+        max_speed    12
+        shield        2
+        Name: (cobra, mark i), dtype: int64
+
+        Single tuple. Note using ``[[]]`` returns a DataFrame.
+
+        >>> df.loc[[("cobra", "mark ii")]]
+                       max_speed  shield
+        cobra mark ii          0       4
+
+        Single tuple for the index with a single label for the column
+
+        >>> df.loc[("cobra", "mark i"), "shield"]
+        2
+
+        Slice from index tuple to single label
+
+        >>> df.loc[("cobra", "mark i") : "viper"]
+                             max_speed  shield
+        cobra      mark i           12       2
+                   mark ii           0       4
+        sidewinder mark i           10      20
+                   mark ii           1       4
+        viper      mark ii           7       1
+                   mark iii         16      36
+
+        Slice from index tuple to index tuple
+
+        >>> df.loc[("cobra", "mark i") : ("viper", "mark ii")]
+                            max_speed  shield
+        cobra      mark i          12       2
+                   mark ii          0       4
+        sidewinder mark i          10      20
+                   mark ii          1       4
+        viper      mark ii          7       1
+
+        Please see the :ref:`user guide<advanced.advanced_hierarchical>`
+        for more details and explanations of advanced indexing.
+        """
+        return _LocIndexer("loc", self)
+
+    @property
+    def at(self) -> _AtIndexer:
+        """
+        Access a single value for a row/column label pair.
+
+        Similar to ``loc``, in that both provide label-based lookups. Use
+        ``at`` if you only need to get or set a single value in a DataFrame
+        or Series.
+
+        Raises
+        ------
+        KeyError
+            If getting a value and 'label' does not exist in a DataFrame or Series.
+
+        ValueError
+            If row/column label pair is not a tuple or if any label
+            from the pair is not a scalar for DataFrame.
+            If label is list-like (*excluding* NamedTuple) for Series.
+
+        See Also
+        --------
+        DataFrame.at : Access a single value for a row/column pair by label.
+        DataFrame.iat : Access a single value for a row/column pair by integer
+            position.
+        DataFrame.loc : Access a group of rows and columns by label(s).
+        DataFrame.iloc : Access a group of rows and columns by integer
+            position(s).
+        Series.at : Access a single value by label.
+        Series.iat : Access a single value by integer position.
+        Series.loc : Access a group of rows by label(s).
+        Series.iloc : Access a group of rows by integer position(s).
+
+        Notes
+        -----
+        See :ref:`Fast scalar value getting and setting <indexing.basics.get_value>`
+        for more details.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [[0, 2, 3], [0, 4, 1], [10, 20, 30]],
+        ...     index=[4, 5, 6],
+        ...     columns=["A", "B", "C"],
+        ... )
+        >>> df
+            A   B   C
+        4   0   2   3
+        5   0   4   1
+        6  10  20  30
+
+        Get value at specified row/column pair
+
+        >>> df.at[4, "B"]
+        2
+
+        Set value at specified row/column pair
+
+        >>> df.at[4, "B"] = 10
+        >>> df.at[4, "B"]
+        10
+
+        Get value within a Series
+
+        >>> df.loc[5].at["B"]
+        4
+        """
+        return _AtIndexer("at", self)
+
+    @property
+    def iat(self) -> _iAtIndexer:
+        """
+        Access a single value for a row/column pair by integer position.
+
+        Similar to ``iloc``, in that both provide integer-based lookups. Use
+        ``iat`` if you only need to get or set a single value in a DataFrame
+        or Series.
+
+        Raises
+        ------
+        IndexError
+            When integer position is out of bounds.
+
+        See Also
+        --------
+        DataFrame.at : Access a single value for a row/column label pair.
+        DataFrame.loc : Access a group of rows and columns by label(s).
+        DataFrame.iloc : Access a group of rows and columns by integer position(s).
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [[0, 2, 3], [0, 4, 1], [10, 20, 30]], columns=["A", "B", "C"]
+        ... )
+        >>> df
+            A   B   C
+        0   0   2   3
+        1   0   4   1
+        2  10  20  30
+
+        Get value at specified row/column pair
+
+        >>> df.iat[1, 2]
+        1
+
+        Set value at specified row/column pair
+
+        >>> df.iat[1, 2] = 10
+        >>> df.iat[1, 2]
+        10
+
+        Get value within a series
+
+        >>> df.loc[0].iat[1]
+        2
+        """
+        return _iAtIndexer("iat", self)
+
+
+class _LocationIndexer(NDFrameIndexerBase):
+    _valid_types: str
+    axis: AxisInt | None = None
+
+    # sub-classes need to set _takeable
+    _takeable: bool
+
+    @final
+    def __call__(self, axis: Axis | None = None) -> Self:
+        # we need to return a copy of ourselves
+        new_self = type(self)(self.name, self.obj)
+
+        if axis is not None:
+            axis_int_none = self.obj._get_axis_number(axis)
+        else:
+            axis_int_none = axis
+        new_self.axis = axis_int_none
+        return new_self
+
+    def _get_setitem_indexer(self, key):
+        """
+        Convert a potentially-label-based key into a positional indexer.
+        """
+        if self.name == "loc":
+            # always holds here bc iloc overrides _get_setitem_indexer
+            self._ensure_listlike_indexer(key, axis=self.axis)
+
+        if isinstance(key, tuple):
+            for x in key:
+                check_dict_or_set_indexers(x)
+
+        if self.axis is not None:
+            key = _tupleize_axis_indexer(self.ndim, self.axis, key)
+
+        ax = self.obj._get_axis(0)
+
+        if (
+            isinstance(ax, MultiIndex)
+            and self.name != "iloc"
+            and is_hashable(key)
+            and not isinstance(key, slice)
+        ):
+            with suppress(KeyError, InvalidIndexError):
+                # TypeError e.g. passed a bool
+                return ax.get_loc(key)
+
+        if isinstance(key, tuple):
+            with suppress(IndexingError):
+                # suppress "Too many indexers"
+                return self._convert_tuple(key)
+
+        if isinstance(key, range):
+            # GH#45479 test_loc_setitem_range_key
+            key = list(key)
+
+        return self._convert_to_indexer(key, axis=0)
+
+    @final
+    def _maybe_mask_setitem_value(self, indexer, value):
+        """
+        If we have obj.iloc[mask] = series_or_frame and series_or_frame has the
+        same length as obj, we treat this as obj.iloc[mask] = series_or_frame[mask],
+        similar to Series.__setitem__.
+
+        Note this is only for loc, not iloc.
+        """
+
+        if (
+            isinstance(indexer, tuple)
+            and len(indexer) == 2
+            and isinstance(value, (ABCSeries, ABCDataFrame))
+        ):
+            pi, icols = indexer
+            ndim = value.ndim
+            if com.is_bool_indexer(pi) and len(value) == len(pi):
+                newkey = pi.nonzero()[0]
+
+                if is_scalar_indexer(icols, self.ndim - 1) and ndim == 1:
+                    # e.g. test_loc_setitem_boolean_mask_allfalse
+                    if len(newkey) == 0:
+                        value = value.iloc[:0]
+                    else:
+                        # test_loc_setitem_ndframe_values_alignment
+                        value = self.obj.iloc._align_series(indexer, value)
+                    indexer = (newkey, icols)
+
+                elif (
+                    isinstance(icols, np.ndarray)
+                    and icols.dtype.kind == "i"
+                    and len(icols) == 1
+                ):
+                    if ndim == 1:
+                        # We implicitly broadcast, though numpy does not, see
+                        # github.com/pandas-dev/pandas/pull/45501#discussion_r789071825
+                        # test_loc_setitem_ndframe_values_alignment
+                        value = self.obj.iloc._align_series(indexer, value)
+                        indexer = (newkey, icols)
+
+                    elif ndim == 2 and value.shape[1] == 1:
+                        if len(newkey) == 0:
+                            value = value.iloc[:0]
+                        else:
+                            # test_loc_setitem_ndframe_values_alignment
+                            value = self.obj.iloc._align_frame(indexer, value)
+                        indexer = (newkey, icols)
+        elif com.is_bool_indexer(indexer):
+            indexer = indexer.nonzero()[0]
+
+        return indexer, value
+
+    @final
+    def _ensure_listlike_indexer(self, key, axis=None, value=None) -> None:
+        """
+        Ensure that a list-like of column labels are all present by adding them if
+        they do not already exist.
+
+        Parameters
+        ----------
+        key : list-like of column labels
+            Target labels.
+        axis : key axis if known
+        """
+        column_axis = 1
+
+        # column only exists in 2-dimensional DataFrame
+        if self.ndim != 2:
+            return
+
+        if isinstance(key, tuple) and len(key) > 1:
+            # key may be a tuple if we are .loc
+            # if length of key is > 1 set key to column part
+            # unless axis is already specified, then go with that
+            if axis is None:
+                axis = column_axis
+            key = key[axis]
+
+        if (
+            axis == column_axis
+            and not isinstance(self.obj.columns, MultiIndex)
+            and is_list_like_indexer(key)
+            and not com.is_bool_indexer(key)
+            and all(is_hashable(k) for k in key)
+        ):
+            # GH#38148
+            keys = self.obj.columns.union(key, sort=False)
+            diff = Index(key).difference(self.obj.columns, sort=False)
+
+            if len(diff):
+                # e.g. if we are doing df.loc[:, ["A", "B"]] = 7 and "B"
+                #  is a new column, add the new columns with dtype=np.void
+                #  so that later when we go through setitem_single_column
+                #  we will use isetitem. Without this, the reindex_axis
+                #  below would create float64 columns in this example, which
+                #  would successfully hold 7, so we would end up with the wrong
+                #  dtype.
+                indexer = np.arange(len(keys), dtype=np.intp)
+                indexer[len(self.obj.columns) :] = -1
+                new_mgr = self.obj._mgr.reindex_indexer(
+                    keys, indexer=indexer, axis=0, only_slice=True, use_na_proxy=True
+                )
+                self.obj._mgr = new_mgr
+                return
+
+            self.obj._mgr = self.obj._mgr.reindex_axis(keys, axis=0, only_slice=True)
+
+    @final
+    def __setitem__(self, key, value) -> None:
+        if not PYPY:
+            if sys.getrefcount(self.obj) <= 2:
+                warnings.warn(
+                    _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
+                )
+
+        check_dict_or_set_indexers(key)
+        if isinstance(key, tuple):
+            key = (list(x) if is_iterator(x) else x for x in key)
+            key = tuple(com.apply_if_callable(x, self.obj) for x in key)
+        else:
+            maybe_callable = com.apply_if_callable(key, self.obj)
+            key = self._raise_callable_usage(key, maybe_callable)
+        indexer = self._get_setitem_indexer(key)
+        self._has_valid_setitem_indexer(key)
+
+        iloc: _iLocIndexer = (
+            cast("_iLocIndexer", self) if self.name == "iloc" else self.obj.iloc
+        )
+        iloc._setitem_with_indexer(indexer, value, self.name)
+
+    def _validate_key(self, key, axis: AxisInt) -> None:
+        """
+        Ensure that key is valid for current indexer.
+
+        Parameters
+        ----------
+        key : scalar, slice or list-like
+            Key requested.
+        axis : int
+            Dimension on which the indexing is being made.
+
+        Raises
+        ------
+        TypeError
+            If the key (or some element of it) has wrong type.
+        IndexError
+            If the key (or some element of it) is out of bounds.
+        KeyError
+            If the key was not found.
+        """
+        raise AbstractMethodError(self)
+
+    @final
+    def _expand_ellipsis(self, tup: tuple) -> tuple:
+        """
+        If a tuple key includes an Ellipsis, replace it with an appropriate
+        number of null slices.
+        """
+        if any(x is Ellipsis for x in tup):
+            if tup.count(Ellipsis) > 1:
+                raise IndexingError(_one_ellipsis_message)
+
+            if len(tup) == self.ndim:
+                # It is unambiguous what axis this Ellipsis is indexing,
+                #  treat as a single null slice.
+                i = tup.index(Ellipsis)
+                # FIXME: this assumes only one Ellipsis
+                new_key = tup[:i] + (_NS,) + tup[i + 1 :]
+                return new_key
+
+            # TODO: other cases?  only one test gets here, and that is covered
+            #  by _validate_key_length
+        return tup
+
+    @final
+    def _validate_tuple_indexer(self, key: tuple) -> tuple:
+        """
+        Check the key for valid keys across my indexer.
+        """
+        key = self._validate_key_length(key)
+        key = self._expand_ellipsis(key)
+        for i, k in enumerate(key):
+            try:
+                self._validate_key(k, i)
+            except ValueError as err:
+                raise ValueError(
+                    f"Location based indexing can only have [{self._valid_types}] types"
+                ) from err
+        return key
+
+    @final
+    def _is_nested_tuple_indexer(self, tup: tuple) -> bool:
+        """
+        Returns
+        -------
+        bool
+        """
+        if any(isinstance(ax, MultiIndex) for ax in self.obj.axes):
+            return any(is_nested_tuple(tup, ax) for ax in self.obj.axes)
+        return False
+
+    @final
+    def _convert_tuple(self, key: tuple) -> tuple:
+        # Note: we assume _tupleize_axis_indexer has been called, if necessary.
+        self._validate_key_length(key)
+        keyidx = [self._convert_to_indexer(k, axis=i) for i, k in enumerate(key)]
+        return tuple(keyidx)
+
+    @final
+    def _validate_key_length(self, key: tuple) -> tuple:
+        if len(key) > self.ndim:
+            if key[0] is Ellipsis:
+                # e.g. Series.iloc[..., 3] reduces to just Series.iloc[3]
+                key = key[1:]
+                if Ellipsis in key:
+                    raise IndexingError(_one_ellipsis_message)
+                return self._validate_key_length(key)
+            raise IndexingError("Too many indexers")
+        return key
+
+    @final
+    def _getitem_tuple_same_dim(self, tup: tuple):
+        """
+        Index with indexers that should return an object of the same dimension
+        as self.obj.
+
+        This is only called after a failed call to _getitem_lowerdim.
+        """
+        retval = self.obj
+        # Selecting columns before rows is significantly faster
+        start_val = (self.ndim - len(tup)) + 1
+        for i, key in enumerate(reversed(tup)):
+            i = self.ndim - i - start_val
+            if com.is_null_slice(key):
+                continue
+
+            retval = getattr(retval, self.name)._getitem_axis(key, axis=i)
+            # We should never have retval.ndim < self.ndim, as that should
+            #  be handled by the _getitem_lowerdim call above.
+            assert retval.ndim == self.ndim
+
+        if retval is self.obj:
+            # if all axes were a null slice (`df.loc[:, :]`), ensure we still
+            # return a new object (https://github.com/pandas-dev/pandas/pull/49469)
+            retval = retval.copy(deep=False)
+
+        return retval
+
+    @final
+    def _getitem_lowerdim(self, tup: tuple):
+        # we can directly get the axis result since the axis is specified
+        if self.axis is not None:
+            axis = self.obj._get_axis_number(self.axis)
+            return self._getitem_axis(tup, axis=axis)
+
+        # we may have a nested tuples indexer here
+        if self._is_nested_tuple_indexer(tup):
+            return self._getitem_nested_tuple(tup)
+
+        # we maybe be using a tuple to represent multiple dimensions here
+        ax0 = self.obj._get_axis(0)
+        # ...but iloc should handle the tuple as simple integer-location
+        # instead of checking it as multiindex representation (GH 13797)
+        if (
+            isinstance(ax0, MultiIndex)
+            and self.name != "iloc"
+            and not any(isinstance(x, slice) for x in tup)
+        ):
+            # Note: in all extant test cases, replacing the slice condition with
+            #  `all(is_hashable(x) or com.is_null_slice(x) for x in tup)`
+            #  is equivalent.
+            #  (see the other place where we call _handle_lowerdim_multi_index_axis0)
+            with suppress(IndexingError):
+                return cast(_LocIndexer, self)._handle_lowerdim_multi_index_axis0(tup)
+
+        tup = self._validate_key_length(tup)
+
+        for i, key in enumerate(tup):
+            if is_label_like(key):
+                # We don't need to check for tuples here because those are
+                #  caught by the _is_nested_tuple_indexer check above.
+                section = self._getitem_axis(key, axis=i)
+
+                # We should never have a scalar section here, because
+                #  _getitem_lowerdim is only called after a check for
+                #  is_scalar_access, which that would be.
+                if section.ndim == self.ndim:
+                    # we're in the middle of slicing through a MultiIndex
+                    # revise the key wrt to `section` by inserting an _NS
+                    new_key = tup[:i] + (_NS,) + tup[i + 1 :]
+
+                else:
+                    # Note: the section.ndim == self.ndim check above
+                    #  rules out having DataFrame here, so we dont need to worry
+                    #  about transposing.
+                    new_key = tup[:i] + tup[i + 1 :]
+
+                    if len(new_key) == 1:
+                        new_key = new_key[0]
+
+                # Slices should return views, but calling iloc/loc with a null
+                # slice returns a new object.
+                if com.is_null_slice(new_key):
+                    return section
+                # This is an elided recursive call to iloc/loc
+                return getattr(section, self.name)[new_key]
+
+        raise IndexingError("not applicable")
+
+    @final
+    def _getitem_nested_tuple(self, tup: tuple):
+        # we have a nested tuple so have at least 1 multi-index level
+        # we should be able to match up the dimensionality here
+
+        def _contains_slice(x: object) -> bool:
+            # Check if object is a slice or a tuple containing a slice
+            if isinstance(x, tuple):
+                return any(isinstance(v, slice) for v in x)
+            elif isinstance(x, slice):
+                return True
+            return False
+
+        for key in tup:
+            check_dict_or_set_indexers(key)
+
+        # we have too many indexers for our dim, but have at least 1
+        # multi-index dimension, try to see if we have something like
+        # a tuple passed to a series with a multi-index
+        if len(tup) > self.ndim:
+            if self.name != "loc":
+                # This should never be reached, but let's be explicit about it
+                raise ValueError("Too many indices")  # pragma: no cover
+            if all(
+                (is_hashable(x) and not _contains_slice(x)) or com.is_null_slice(x)
+                for x in tup
+            ):
+                # GH#10521 Series should reduce MultiIndex dimensions instead of
+                #  DataFrame, IndexingError is not raised when slice(None,None,None)
+                #  with one row.
+                with suppress(IndexingError):
+                    return cast(_LocIndexer, self)._handle_lowerdim_multi_index_axis0(
+                        tup
+                    )
+            elif isinstance(self.obj, ABCSeries) and any(
+                isinstance(k, tuple) for k in tup
+            ):
+                # GH#35349 Raise if tuple in tuple for series
+                # Do this after the all-hashable-or-null-slice check so that
+                #  we are only getting non-hashable tuples, in particular ones
+                #  that themselves contain a slice entry
+                # See test_loc_series_getitem_too_many_dimensions
+                raise IndexingError("Too many indexers")
+
+            # this is a series with a multi-index specified a tuple of
+            # selectors
+            axis = self.axis or 0
+            return self._getitem_axis(tup, axis=axis)
+
+        # handle the multi-axis by taking sections and reducing
+        # this is iterative
+        obj = self.obj
+        # GH#41369 Loop in reverse order ensures indexing along columns before rows
+        # which selects only necessary blocks which avoids dtype conversion if possible
+        axis = len(tup) - 1
+        for key in reversed(tup):
+            if com.is_null_slice(key):
+                axis -= 1
+                continue
+
+            obj = getattr(obj, self.name)._getitem_axis(key, axis=axis)
+            axis -= 1
+
+            # if we have a scalar, we are done
+            if is_scalar(obj) or not hasattr(obj, "ndim"):
+                break
+
+        return obj
+
+    def _convert_to_indexer(self, key, axis: AxisInt):
+        raise AbstractMethodError(self)
+
+    def _raise_callable_usage(self, key: Any, maybe_callable: T) -> T:
+        # GH53533
+        if self.name == "iloc" and callable(key) and isinstance(maybe_callable, tuple):
+            raise ValueError(
+                "Returning a tuple from a callable with iloc is not allowed.",
+            )
+        return maybe_callable
+
+    @final
+    def __getitem__(self, key):
+        check_dict_or_set_indexers(key)
+        if type(key) is tuple:
+            key = (list(x) if is_iterator(x) else x for x in key)
+            key = tuple(com.apply_if_callable(x, self.obj) for x in key)
+            if self._is_scalar_access(key):
+                return self.obj._get_value(*key, takeable=self._takeable)
+            return self._getitem_tuple(key)
+        else:
+            # we by definition only have the 0th axis
+            axis = self.axis or 0
+
+            maybe_callable = com.apply_if_callable(key, self.obj)
+            maybe_callable = self._raise_callable_usage(key, maybe_callable)
+            return self._getitem_axis(maybe_callable, axis=axis)
+
+    def _is_scalar_access(self, key: tuple):
+        raise NotImplementedError
+
+    def _getitem_tuple(self, tup: tuple):
+        raise AbstractMethodError(self)
+
+    def _getitem_axis(self, key, axis: AxisInt):
+        raise NotImplementedError
+
+    def _has_valid_setitem_indexer(self, indexer) -> bool:
+        raise AbstractMethodError(self)
+
+    @final
+    def _getbool_axis(self, key, axis: AxisInt):
+        # caller is responsible for ensuring non-None axis
+        labels = self.obj._get_axis(axis)
+        key = check_bool_indexer(labels, key)
+        inds = key.nonzero()[0]
+        return self.obj.take(inds, axis=axis)
+
+
+@doc(IndexingMixin.loc)
+class _LocIndexer(_LocationIndexer):
+    _takeable: bool = False
+    _valid_types = (
+        "labels (MUST BE IN THE INDEX), slices of labels (BOTH "
+        "endpoints included! Can be slices of integers if the "
+        "index is integers), listlike of labels, boolean"
+    )
+
+    # -------------------------------------------------------------------
+    # Key Checks
+
+    @doc(_LocationIndexer._validate_key)
+    def _validate_key(self, key, axis: Axis) -> None:
+        # valid for a collection of labels (we check their presence later)
+        # slice of labels (where start-end in labels)
+        # slice of integers (only if in the labels)
+        # boolean not in slice and with boolean index
+        ax = self.obj._get_axis(axis)
+        if isinstance(key, bool) and not (
+            is_bool_dtype(ax.dtype)
+            or ax.dtype.name == "boolean"
+            or (
+                isinstance(ax, MultiIndex)
+                and is_bool_dtype(ax.get_level_values(0).dtype)
+            )
+        ):
+            raise KeyError(
+                f"{key}: boolean label can not be used without a boolean index"
+            )
+
+        if isinstance(key, slice) and (
+            isinstance(key.start, bool) or isinstance(key.stop, bool)
+        ):
+            raise TypeError(f"{key}: boolean values can not be used in a slice")
+
+    def _has_valid_setitem_indexer(self, indexer) -> bool:
+        return True
+
+    def _is_scalar_access(self, key: tuple) -> bool:
+        """
+        Returns
+        -------
+        bool
+        """
+        # this is a shortcut accessor to both .loc and .iloc
+        # that provide the equivalent access of .at and .iat
+        # a) avoid getting things via sections and (to minimize dtype changes)
+        # b) provide a performant path
+        if len(key) != self.ndim:
+            return False
+
+        for i, k in enumerate(key):
+            if not is_scalar(k):
+                return False
+
+            ax = self.obj.axes[i]
+            if isinstance(ax, MultiIndex):
+                return False
+
+            if isinstance(k, str) and ax._supports_partial_string_indexing:
+                # partial string indexing, df.loc['2000', 'A']
+                # should not be considered scalar
+                return False
+
+            if not ax._index_as_unique:
+                return False
+
+        return True
+
+    # -------------------------------------------------------------------
+    # MultiIndex Handling
+
+    def _multi_take_opportunity(self, tup: tuple) -> bool:
+        """
+        Check whether there is the possibility to use ``_multi_take``.
+
+        Currently the limit is that all axes being indexed, must be indexed with
+        list-likes.
+
+        Parameters
+        ----------
+        tup : tuple
+            Tuple of indexers, one per axis.
+
+        Returns
+        -------
+        bool
+            Whether the current indexing,
+            can be passed through `_multi_take`.
+        """
+        if not all(is_list_like_indexer(x) for x in tup):
+            return False
+
+        # just too complicated
+        return not any(com.is_bool_indexer(x) for x in tup)
+
+    def _multi_take(self, tup: tuple):
+        """
+        Create the indexers for the passed tuple of keys, and
+        executes the take operation. This allows the take operation to be
+        executed all at once, rather than once for each dimension.
+        Improving efficiency.
+
+        Parameters
+        ----------
+        tup : tuple
+            Tuple of indexers, one per axis.
+
+        Returns
+        -------
+        values: same type as the object being indexed
+        """
+        # GH 836
+        d = {
+            axis: self._get_listlike_indexer(key, axis)
+            for (key, axis) in zip(tup, self.obj._AXIS_ORDERS)
+        }
+        return self.obj._reindex_with_indexers(d, allow_dups=True)
+
+    # -------------------------------------------------------------------
+
+    def _getitem_iterable(self, key, axis: AxisInt):
+        """
+        Index current object with an iterable collection of keys.
+
+        Parameters
+        ----------
+        key : iterable
+            Targeted labels.
+        axis : int
+            Dimension on which the indexing is being made.
+
+        Raises
+        ------
+        KeyError
+            If no key was found. Will change in the future to raise if not all
+            keys were found.
+
+        Returns
+        -------
+        scalar, DataFrame, or Series: indexed value(s).
+        """
+        # we assume that not com.is_bool_indexer(key), as that is
+        #  handled before we get here.
+        self._validate_key(key, axis)
+
+        # A collection of keys
+        keyarr, indexer = self._get_listlike_indexer(key, axis)
+        return self.obj._reindex_with_indexers(
+            {axis: [keyarr, indexer]}, allow_dups=True
+        )
+
+    def _getitem_tuple(self, tup: tuple):
+        with suppress(IndexingError):
+            tup = self._expand_ellipsis(tup)
+            return self._getitem_lowerdim(tup)
+
+        # no multi-index, so validate all of the indexers
+        tup = self._validate_tuple_indexer(tup)
+
+        # ugly hack for GH #836
+        if self._multi_take_opportunity(tup):
+            return self._multi_take(tup)
+
+        return self._getitem_tuple_same_dim(tup)
+
+    def _get_label(self, label, axis: AxisInt):
+        # GH#5567 this will fail if the label is not present in the axis.
+        return self.obj.xs(label, axis=axis)
+
+    def _handle_lowerdim_multi_index_axis0(self, tup: tuple):
+        # we have an axis0 multi-index, handle or raise
+        axis = self.axis or 0
+        try:
+            # fast path for series or for tup devoid of slices
+            return self._get_label(tup, axis=axis)
+
+        except KeyError as ek:
+            # raise KeyError if number of indexers match
+            # else IndexingError will be raised
+            if self.ndim < len(tup) <= self.obj.index.nlevels:
+                raise ek
+            raise IndexingError("No label returned") from ek
+
+    def _getitem_axis(self, key, axis: AxisInt):
+        key = item_from_zerodim(key)
+        if is_iterator(key):
+            key = list(key)
+        if key is Ellipsis:
+            key = slice(None)
+
+        labels = self.obj._get_axis(axis)
+
+        if isinstance(key, tuple) and isinstance(labels, MultiIndex):
+            key = tuple(key)
+
+        if isinstance(key, slice):
+            self._validate_key(key, axis)
+            return self._get_slice_axis(key, axis=axis)
+        elif com.is_bool_indexer(key):
+            return self._getbool_axis(key, axis=axis)
+        elif is_list_like_indexer(key):
+            # an iterable multi-selection
+            if not (isinstance(key, tuple) and isinstance(labels, MultiIndex)):
+                if hasattr(key, "ndim") and key.ndim > 1:
+                    raise ValueError("Cannot index with multidimensional key")
+
+                return self._getitem_iterable(key, axis=axis)
+
+            # nested tuple slicing
+            if is_nested_tuple(key, labels):
+                locs = labels.get_locs(key)
+                indexer: list[slice | npt.NDArray[np.intp]] = [slice(None)] * self.ndim
+                indexer[axis] = locs
+                return self.obj.iloc[tuple(indexer)]
+
+        # fall thru to straight lookup
+        self._validate_key(key, axis)
+        return self._get_label(key, axis=axis)
+
+    def _get_slice_axis(self, slice_obj: slice, axis: AxisInt):
+        """
+        This is pretty simple as we just have to deal with labels.
+        """
+        # caller is responsible for ensuring non-None axis
+        obj = self.obj
+        if not need_slice(slice_obj):
+            return obj.copy(deep=False)
+
+        labels = obj._get_axis(axis)
+        indexer = labels.slice_indexer(slice_obj.start, slice_obj.stop, slice_obj.step)
+
+        if isinstance(indexer, slice):
+            return self.obj._slice(indexer, axis=axis)
+        else:
+            # DatetimeIndex overrides Index.slice_indexer and may
+            #  return a DatetimeIndex instead of a slice object.
+            return self.obj.take(indexer, axis=axis)
+
+    def _convert_to_indexer(self, key, axis: AxisInt):
+        """
+        Convert indexing key into something we can use to do actual fancy
+        indexing on a ndarray.
+
+        Examples
+        ix[:5] -> slice(0, 5)
+        ix[[1,2,3]] -> [1,2,3]
+        ix[['foo', 'bar', 'baz']] -> [i, j, k] (indices of foo, bar, baz)
+
+        Going by Zen of Python?
+        'In the face of ambiguity, refuse the temptation to guess.'
+        raise AmbiguousIndexError with integer labels?
+        - No, prefer label-based indexing
+        """
+        labels = self.obj._get_axis(axis)
+
+        if isinstance(key, slice):
+            return labels._convert_slice_indexer(key, kind="loc")
+
+        if (
+            isinstance(key, tuple)
+            and not isinstance(labels, MultiIndex)
+            and self.ndim < 2
+            and len(key) > 1
+        ):
+            raise IndexingError("Too many indexers")
+
+        # Slices are not valid keys passed in by the user,
+        # even though they are hashable in Python 3.12
+        contains_slice = False
+        if isinstance(key, tuple):
+            contains_slice = any(isinstance(v, slice) for v in key)
+
+        if is_scalar(key) or (
+            isinstance(labels, MultiIndex) and is_hashable(key) and not contains_slice
+        ):
+            # Otherwise get_loc will raise InvalidIndexError
+
+            # if we are a label return me
+            try:
+                return labels.get_loc(key)
+            except LookupError:
+                if isinstance(key, tuple) and isinstance(labels, MultiIndex):
+                    if len(key) == labels.nlevels:
+                        return {"key": key}
+                    raise
+            except InvalidIndexError:
+                # GH35015, using datetime as column indices raises exception
+                if not isinstance(labels, MultiIndex):
+                    raise
+            except ValueError:
+                if not is_integer(key):
+                    raise
+                return {"key": key}
+
+        if is_nested_tuple(key, labels):
+            if self.ndim == 1 and any(isinstance(k, tuple) for k in key):
+                # GH#35349 Raise if tuple in tuple for series
+                raise IndexingError("Too many indexers")
+            return labels.get_locs(key)
+
+        elif is_list_like_indexer(key):
+            if is_iterator(key):
+                key = list(key)
+
+            if com.is_bool_indexer(key):
+                key = check_bool_indexer(labels, key)
+                return key
+            else:
+                return self._get_listlike_indexer(key, axis)[1]
+        else:
+            try:
+                return labels.get_loc(key)
+            except LookupError:
+                # allow a not found key only if we are a setter
+                if not is_list_like_indexer(key):
+                    return {"key": key}
+                raise
+
+    def _get_listlike_indexer(self, key, axis: AxisInt):
+        """
+        Transform a list-like of keys into a new index and an indexer.
+
+        Parameters
+        ----------
+        key : list-like
+            Targeted labels.
+        axis:  int
+            Dimension on which the indexing is being made.
+
+        Raises
+        ------
+        KeyError
+            If at least one key was requested but none was found.
+
+        Returns
+        -------
+        keyarr: Index
+            New index (coinciding with 'key' if the axis is unique).
+        values : array-like
+            Indexer for the return object, -1 denotes keys not found.
+        """
+        ax = self.obj._get_axis(axis)
+        axis_name = self.obj._get_axis_name(axis)
+
+        keyarr, indexer = ax._get_indexer_strict(key, axis_name)
+
+        return keyarr, indexer
+
+
+@doc(IndexingMixin.iloc)
+class _iLocIndexer(_LocationIndexer):
+    _valid_types = (
+        "integer, integer slice (START point is INCLUDED, END "
+        "point is EXCLUDED), listlike of integers, boolean array"
+    )
+    _takeable = True
+
+    # -------------------------------------------------------------------
+    # Key Checks
+
+    def _validate_key(self, key, axis: AxisInt) -> None:
+        if com.is_bool_indexer(key):
+            if hasattr(key, "index") and isinstance(key.index, Index):
+                if key.index.inferred_type == "integer":
+                    raise NotImplementedError(
+                        "iLocation based boolean "
+                        "indexing on an integer type "
+                        "is not available"
+                    )
+                raise ValueError(
+                    "iLocation based boolean indexing cannot use an indexable as a mask"
+                )
+            return
+
+        if isinstance(key, slice):
+            return
+        elif is_integer(key):
+            self._validate_integer(key, axis)
+        elif isinstance(key, tuple):
+            # a tuple should already have been caught by this point
+            # so don't treat a tuple as a valid indexer
+            raise IndexingError("Too many indexers")
+        elif is_list_like_indexer(key):
+            if isinstance(key, ABCSeries):
+                arr = key._values
+            elif is_array_like(key):
+                arr = key
+            else:
+                arr = np.array(key)
+            len_axis = len(self.obj._get_axis(axis))
+
+            # check that the key has a numeric dtype
+            if not is_numeric_dtype(arr.dtype):
+                raise IndexError(f".iloc requires numeric indexers, got {arr}")
+
+            # check that the key does not exceed the maximum size of the index
+            if len(arr) and (arr.max() >= len_axis or arr.min() < -len_axis):
+                raise IndexError("positional indexers are out-of-bounds")
+        else:
+            raise ValueError(f"Can only index by location with a [{self._valid_types}]")
+
+    def _has_valid_setitem_indexer(self, indexer) -> bool:
+        """
+        Validate that a positional indexer cannot enlarge its target
+        will raise if needed, does not modify the indexer externally.
+
+        Returns
+        -------
+        bool
+        """
+        if isinstance(indexer, dict):
+            raise IndexError("iloc cannot enlarge its target object")
+
+        if isinstance(indexer, ABCDataFrame):
+            raise TypeError(
+                "DataFrame indexer for .iloc is not supported. "
+                "Consider using .loc with a DataFrame indexer for automatic alignment.",
+            )
+
+        if not isinstance(indexer, tuple):
+            indexer = _tuplify(self.ndim, indexer)
+
+        for ax, i in zip(self.obj.axes, indexer):
+            if isinstance(i, slice):
+                # should check the stop slice?
+                pass
+            elif is_list_like_indexer(i):
+                # should check the elements?
+                pass
+            elif is_integer(i):
+                if i >= len(ax):
+                    raise IndexError("iloc cannot enlarge its target object")
+            elif isinstance(i, dict):
+                raise IndexError("iloc cannot enlarge its target object")
+
+        return True
+
+    def _is_scalar_access(self, key: tuple) -> bool:
+        """
+        Returns
+        -------
+        bool
+        """
+        # this is a shortcut accessor to both .loc and .iloc
+        # that provide the equivalent access of .at and .iat
+        # a) avoid getting things via sections and (to minimize dtype changes)
+        # b) provide a performant path
+        if len(key) != self.ndim:
+            return False
+
+        return all(is_integer(k) for k in key)
+
+    def _validate_integer(self, key: int | np.integer, axis: AxisInt) -> None:
+        """
+        Check that 'key' is a valid position in the desired axis.
+
+        Parameters
+        ----------
+        key : int
+            Requested position.
+        axis : int
+            Desired axis.
+
+        Raises
+        ------
+        IndexError
+            If 'key' is not a valid position in axis 'axis'.
+        """
+        len_axis = len(self.obj._get_axis(axis))
+        if key >= len_axis or key < -len_axis:
+            raise IndexError("single positional indexer is out-of-bounds")
+
+    # -------------------------------------------------------------------
+
+    def _getitem_tuple(self, tup: tuple):
+        tup = self._validate_tuple_indexer(tup)
+        with suppress(IndexingError):
+            return self._getitem_lowerdim(tup)
+
+        return self._getitem_tuple_same_dim(tup)
+
+    def _get_list_axis(self, key, axis: AxisInt):
+        """
+        Return Series values by list or array of integers.
+
+        Parameters
+        ----------
+        key : list-like positional indexer
+        axis : int
+
+        Returns
+        -------
+        Series object
+
+        Notes
+        -----
+        `axis` can only be zero.
+        """
+        try:
+            return self.obj.take(key, axis=axis)
+        except IndexError as err:
+            # re-raise with different error message, e.g. test_getitem_ndarray_3d
+            raise IndexError("positional indexers are out-of-bounds") from err
+
+    def _getitem_axis(self, key, axis: AxisInt):
+        if key is Ellipsis:
+            key = slice(None)
+        elif isinstance(key, ABCDataFrame):
+            raise IndexError(
+                "DataFrame indexer is not allowed for .iloc\n"
+                "Consider using .loc for automatic alignment."
+            )
+
+        if isinstance(key, slice):
+            return self._get_slice_axis(key, axis=axis)
+
+        if is_iterator(key):
+            key = list(key)
+
+        if isinstance(key, list):
+            key = np.asarray(key)
+
+        if com.is_bool_indexer(key):
+            self._validate_key(key, axis)
+            return self._getbool_axis(key, axis=axis)
+
+        # a list of integers
+        elif is_list_like_indexer(key):
+            return self._get_list_axis(key, axis=axis)
+
+        # a single integer
+        else:
+            key = item_from_zerodim(key)
+            if not is_integer(key):
+                raise TypeError("Cannot index by location index with a non-integer key")
+
+            # validate the location
+            self._validate_integer(key, axis)
+
+            return self.obj._ixs(key, axis=axis)
+
+    def _get_slice_axis(self, slice_obj: slice, axis: AxisInt):
+        # caller is responsible for ensuring non-None axis
+        obj = self.obj
+
+        if not need_slice(slice_obj):
+            return obj.copy(deep=False)
+
+        labels = obj._get_axis(axis)
+        labels._validate_positional_slice(slice_obj)
+        return self.obj._slice(slice_obj, axis=axis)
+
+    def _convert_to_indexer(self, key: T, axis: AxisInt) -> T:
+        """
+        Much simpler as we only have to deal with our valid types.
+        """
+        return key
+
+    def _get_setitem_indexer(self, key):
+        # GH#32257 Fall through to let numpy do validation
+        if is_iterator(key):
+            key = list(key)
+
+        if self.axis is not None:
+            key = _tupleize_axis_indexer(self.ndim, self.axis, key)
+
+        return key
+
+    # -------------------------------------------------------------------
+
+    def _setitem_with_indexer(self, indexer, value, name: str = "iloc") -> None:
+        """
+        _setitem_with_indexer is for setting values on a Series/DataFrame
+        using positional indexers.
+
+        If the relevant keys are not present, the Series/DataFrame may be
+        expanded.
+
+        This method is currently broken when dealing with non-unique Indexes,
+        since it goes from positional indexers back to labels when calling
+        BlockManager methods, see GH#12991, GH#22046, GH#15686.
+        """
+        info_axis = self.obj._info_axis_number
+
+        # maybe partial set
+        take_split_path = not self.obj._mgr.is_single_block
+
+        if not take_split_path and isinstance(value, ABCDataFrame):
+            # Avoid cast of values
+            take_split_path = not value._mgr.is_single_block
+
+        # if there is only one block/type, still have to take split path
+        # unless the block is one-dimensional or it can hold the value
+        if not take_split_path and len(self.obj._mgr.blocks) and self.ndim > 1:
+            # in case of dict, keys are indices
+            val = list(value.values()) if isinstance(value, dict) else value
+            arr = self.obj._mgr.blocks[0].values
+            take_split_path = not can_hold_element(
+                arr, extract_array(val, extract_numpy=True)
+            )
+
+        # if we have any multi-indexes that have non-trivial slices
+        # (not null slices) then we must take the split path, xref
+        # GH 10360, GH 27841
+        if isinstance(indexer, tuple) and len(indexer) == len(self.obj.axes):
+            for i, ax in zip(indexer, self.obj.axes):
+                if isinstance(ax, MultiIndex) and not (
+                    is_integer(i) or com.is_null_slice(i)
+                ):
+                    take_split_path = True
+                    break
+
+        if isinstance(indexer, tuple):
+            nindexer = []
+            for i, idx in enumerate(indexer):
+                if isinstance(idx, dict):
+                    # reindex the axis to the new value
+                    # and set inplace
+                    key, _ = convert_missing_indexer(idx)
+
+                    # if this is the items axes, then take the main missing
+                    # path first
+                    # this correctly sets the dtype and avoids cache issues
+                    # essentially this separates out the block that is needed
+                    # to possibly be modified
+                    if self.ndim > 1 and i == info_axis:
+                        # add the new item, and set the value
+                        # must have all defined axes if we have a scalar
+                        # or a list-like on the non-info axes if we have a
+                        # list-like
+                        if not len(self.obj):
+                            if not is_list_like_indexer(value):
+                                raise ValueError(
+                                    "cannot set a frame with no "
+                                    "defined index and a scalar"
+                                )
+                            self.obj[key] = value
+                            return
+
+                        # add a new item with the dtype setup
+                        if com.is_null_slice(indexer[0]):
+                            # We are setting an entire column
+                            self.obj[key] = value
+                            return
+                        elif is_array_like(value):
+                            # GH#42099
+                            arr = extract_array(value, extract_numpy=True)
+                            taker = -1 * np.ones(len(self.obj), dtype=np.intp)
+                            empty_value = algos.take_nd(arr, taker)
+                            if not isinstance(value, ABCSeries):
+                                # if not Series (in which case we need to align),
+                                #  we can short-circuit
+                                if (
+                                    isinstance(arr, np.ndarray)
+                                    and arr.ndim == 1
+                                    and len(arr) == 1
+                                ):
+                                    # NumPy 1.25 deprecation: https://github.com/numpy/numpy/pull/10615
+                                    arr = arr[0, ...]
+                                empty_value[indexer[0]] = arr
+                                self.obj[key] = empty_value
+                                return
+
+                            self.obj[key] = empty_value
+                        elif not is_list_like(value):
+                            self.obj[key] = construct_1d_array_from_inferred_fill_value(
+                                value, len(self.obj)
+                            )
+                        else:
+                            # FIXME: GH#42099#issuecomment-864326014
+                            self.obj[key] = infer_fill_value(value)
+
+                        new_indexer = convert_from_missing_indexer_tuple(
+                            indexer, self.obj.axes
+                        )
+                        self._setitem_with_indexer(new_indexer, value, name)
+
+                        return
+
+                    # reindex the axis
+                    # make sure to clear the cache because we are
+                    # just replacing the block manager here
+                    # so the object is the same
+                    index = self.obj._get_axis(i)
+                    labels = index.insert(len(index), key)
+
+                    # We are expanding the Series/DataFrame values to match
+                    #  the length of thenew index `labels`.  GH#40096 ensure
+                    #  this is valid even if the index has duplicates.
+                    taker = np.arange(len(index) + 1, dtype=np.intp)
+                    taker[-1] = -1
+                    reindexers = {i: (labels, taker)}
+                    new_obj = self.obj._reindex_with_indexers(
+                        reindexers, allow_dups=True
+                    )
+                    self.obj._mgr = new_obj._mgr
+
+                    nindexer.append(labels.get_loc(key))
+
+                else:
+                    nindexer.append(idx)
+
+            indexer = tuple(nindexer)
+        else:
+            indexer, missing = convert_missing_indexer(indexer)
+
+            if missing:
+                self._setitem_with_indexer_missing(indexer, value)
+                return
+
+        if name == "loc":
+            # must come after setting of missing
+            indexer, value = self._maybe_mask_setitem_value(indexer, value)
+
+        # align and set the values
+        if take_split_path:
+            # We have to operate column-wise
+            self._setitem_with_indexer_split_path(indexer, value, name)
+        else:
+            self._setitem_single_block(indexer, value, name)
+
+    def _setitem_with_indexer_split_path(self, indexer, value, name: str):
+        """
+        Setitem column-wise.
+        """
+        # Above we only set take_split_path to True for 2D cases
+        assert self.ndim == 2
+
+        if not isinstance(indexer, tuple):
+            indexer = _tuplify(self.ndim, indexer)
+        if len(indexer) > self.ndim:
+            raise IndexError("too many indices for array")
+        if isinstance(indexer[0], np.ndarray) and indexer[0].ndim > 2:
+            raise ValueError(r"Cannot set values with ndim > 2")
+
+        if (isinstance(value, ABCSeries) and name != "iloc") or isinstance(value, dict):
+            from pandas import Series
+
+            value = self._align_series(indexer, Series(value))
+
+        # Ensure we have something we can iterate over
+        info_axis = indexer[1]
+        ilocs = self._ensure_iterable_column_indexer(info_axis)
+
+        pi = indexer[0]
+        lplane_indexer = length_of_indexer(pi, self.obj.index)
+        # lplane_indexer gives the expected length of obj[indexer[0]]
+
+        # we need an iterable, with a ndim of at least 1
+        # eg. don't pass through np.array(0)
+        if is_list_like_indexer(value) and getattr(value, "ndim", 1) > 0:
+            if isinstance(value, ABCDataFrame):
+                self._setitem_with_indexer_frame_value(indexer, value, name)
+
+            elif np.ndim(value) == 2:
+                # TODO: avoid np.ndim call in case it isn't an ndarray, since
+                #  that will construct an ndarray, which will be wasteful
+                self._setitem_with_indexer_2d_value(indexer, value)
+
+            elif len(ilocs) == 1 and lplane_indexer == len(value) and not is_scalar(pi):
+                # We are setting multiple rows in a single column.
+                self._setitem_single_column(ilocs[0], value, pi)
+
+            elif len(ilocs) == 1 and 0 != lplane_indexer != len(value):
+                # We are trying to set N values into M entries of a single
+                #  column, which is invalid for N != M
+                # Exclude zero-len for e.g. boolean masking that is all-false
+
+                if len(value) == 1 and not is_integer(info_axis):
+                    # This is a case like df.iloc[:3, [1]] = [0]
+                    #  where we treat as df.iloc[:3, 1] = 0
+                    return self._setitem_with_indexer((pi, info_axis[0]), value[0])
+
+                raise ValueError(
+                    "Must have equal len keys and value when setting with an iterable"
+                )
+
+            elif lplane_indexer == 0 and len(value) == len(self.obj.index):
+                # We get here in one case via .loc with a all-False mask
+                pass
+
+            elif self._is_scalar_access(indexer) and is_object_dtype(
+                self.obj.dtypes._values[ilocs[0]]
+            ):
+                # We are setting nested data, only possible for object dtype data
+                self._setitem_single_column(indexer[1], value, pi)
+
+            elif len(ilocs) == len(value):
+                # We are setting multiple columns in a single row.
+                for loc, v in zip(ilocs, value):
+                    self._setitem_single_column(loc, v, pi)
+
+            elif len(ilocs) == 1 and com.is_null_slice(pi) and len(self.obj) == 0:
+                # This is a setitem-with-expansion, see
+                #  test_loc_setitem_empty_append_expands_rows_mixed_dtype
+                # e.g. df = DataFrame(columns=["x", "y"])
+                #  df["x"] = df["x"].astype(np.int64)
+                #  df.loc[:, "x"] = [1, 2, 3]
+                self._setitem_single_column(ilocs[0], value, pi)
+
+            else:
+                raise ValueError(
+                    "Must have equal len keys and value when setting with an iterable"
+                )
+
+        else:
+            # scalar value
+            for loc in ilocs:
+                self._setitem_single_column(loc, value, pi)
+
+    def _setitem_with_indexer_2d_value(self, indexer, value) -> None:
+        # We get here with np.ndim(value) == 2, excluding DataFrame,
+        #  which goes through _setitem_with_indexer_frame_value
+        pi = indexer[0]
+
+        ilocs = self._ensure_iterable_column_indexer(indexer[1])
+
+        if not is_array_like(value):
+            # cast lists to array
+            value = np.array(value, dtype=object)
+        if len(ilocs) != value.shape[1]:
+            raise ValueError(
+                "Must have equal len keys and value when setting with an ndarray"
+            )
+
+        for i, loc in enumerate(ilocs):
+            value_col = value[:, i]
+            if is_object_dtype(value_col.dtype):
+                # casting to list so that we do type inference in setitem_single_column
+                value_col = value_col.tolist()
+            self._setitem_single_column(loc, value_col, pi)
+
+    def _setitem_with_indexer_frame_value(
+        self, indexer, value: DataFrame, name: str
+    ) -> None:
+        ilocs = self._ensure_iterable_column_indexer(indexer[1])
+
+        sub_indexer = list(indexer)
+        pi = indexer[0]
+
+        multiindex_indexer = isinstance(self.obj.columns, MultiIndex)
+
+        unique_cols = value.columns.is_unique
+
+        # We do not want to align the value in case of iloc GH#37728
+        if name == "iloc":
+            for i, loc in enumerate(ilocs):
+                val = value.iloc[:, i]
+                self._setitem_single_column(loc, val, pi)
+
+        elif not unique_cols and value.columns.equals(self.obj.columns):
+            # We assume we are already aligned, see
+            # test_iloc_setitem_frame_duplicate_columns_multiple_blocks
+            for loc in ilocs:
+                item = self.obj.columns[loc]
+                if item in value:
+                    sub_indexer[1] = item
+                    val = self._align_series(
+                        tuple(sub_indexer),
+                        value.iloc[:, loc],
+                        multiindex_indexer,
+                    )
+                else:
+                    val = np.nan
+
+                self._setitem_single_column(loc, val, pi)
+
+        elif not unique_cols:
+            raise ValueError("Setting with non-unique columns is not allowed.")
+
+        else:
+            for loc in ilocs:
+                item = self.obj.columns[loc]
+                if item in value:
+                    sub_indexer[1] = item
+                    val = self._align_series(
+                        tuple(sub_indexer),
+                        value[item],
+                        multiindex_indexer,
+                        using_cow=True,
+                    )
+                else:
+                    val = np.nan
+
+                self._setitem_single_column(loc, val, pi)
+
+    def _setitem_single_column(self, loc: int, value, plane_indexer) -> None:
+        """
+
+        Parameters
+        ----------
+        loc : int
+            Indexer for column position
+        plane_indexer : int, slice, listlike[int]
+            The indexer we use for setitem along axis=0.
+        """
+        pi = plane_indexer
+
+        is_full_setter = com.is_null_slice(pi) or com.is_full_slice(pi, len(self.obj))
+
+        is_null_setter = com.is_empty_slice(pi) or (is_array_like(pi) and len(pi) == 0)
+
+        if is_null_setter:
+            # no-op, don't cast dtype later
+            return
+
+        elif is_full_setter:
+            try:
+                self.obj._mgr.column_setitem(
+                    loc, plane_indexer, value, inplace_only=True
+                )
+            except (ValueError, TypeError, LossySetitemError) as exc:
+                # If we're setting an entire column and we can't do it inplace,
+                #  then we can use value's dtype (or inferred dtype)
+                #  instead of object
+                dtype = self.obj.dtypes.iloc[loc]
+                if dtype not in (np.void, object) and not self.obj.empty:
+                    # - Exclude np.void, as that is a special case for expansion.
+                    #   We want to raise for
+                    #       df = pd.DataFrame({'a': [1, 2]})
+                    #       df.loc[:, 'a'] = .3
+                    #   but not for
+                    #       df = pd.DataFrame({'a': [1, 2]})
+                    #       df.loc[:, 'b'] = .3
+                    # - Exclude `object`, as then no upcasting happens.
+                    # - Exclude empty initial object with enlargement,
+                    #   as then there's nothing to be inconsistent with.
+                    raise TypeError(
+                        f"Invalid value '{value}' for dtype '{dtype}'"
+                    ) from exc
+                self.obj.isetitem(loc, value)
+        else:
+            # set value into the column (first attempting to operate inplace, then
+            #  falling back to casting if necessary)
+            dtype = self.obj.dtypes.iloc[loc]
+            if dtype == np.void:
+                # This means we're expanding, with multiple columns, e.g.
+                #     df = pd.DataFrame({'A': [1,2,3], 'B': [4,5,6]})
+                #     df.loc[df.index <= 2, ['F', 'G']] = (1, 'abc')
+                # Columns F and G will initially be set to np.void.
+                # Here, we replace those temporary `np.void` columns with
+                # columns of the appropriate dtype, based on `value`.
+                self.obj.iloc[:, loc] = construct_1d_array_from_inferred_fill_value(
+                    value, len(self.obj)
+                )
+            self.obj._mgr.column_setitem(loc, plane_indexer, value)
+
+    def _setitem_single_block(self, indexer, value, name: str) -> None:
+        """
+        _setitem_with_indexer for the case when we have a single Block.
+        """
+        from pandas import Series
+
+        if (isinstance(value, ABCSeries) and name != "iloc") or isinstance(value, dict):
+            # TODO(EA): ExtensionBlock.setitem this causes issues with
+            # setting for extensionarrays that store dicts. Need to decide
+            # if it's worth supporting that.
+            value = self._align_series(indexer, Series(value))
+
+        info_axis = self.obj._info_axis_number
+        item_labels = self.obj._get_axis(info_axis)
+        if isinstance(indexer, tuple):
+            # if we are setting on the info axis ONLY
+            # set using those methods to avoid block-splitting
+            # logic here
+            if (
+                self.ndim == len(indexer) == 2
+                and is_integer(indexer[1])
+                and com.is_null_slice(indexer[0])
+            ):
+                col = item_labels[indexer[info_axis]]
+                if len(item_labels.get_indexer_for([col])) == 1:
+                    # e.g. test_loc_setitem_empty_append_expands_rows
+                    loc = item_labels.get_loc(col)
+                    self._setitem_single_column(loc, value, indexer[0])
+                    return
+
+            indexer = maybe_convert_ix(*indexer)  # e.g. test_setitem_frame_align
+
+        if isinstance(value, ABCDataFrame) and name != "iloc":
+            value = self._align_frame(indexer, value)._values
+
+        # actually do the set
+        self.obj._mgr = self.obj._mgr.setitem(indexer=indexer, value=value)
+
+    def _setitem_with_indexer_missing(self, indexer, value):
+        """
+        Insert new row(s) or column(s) into the Series or DataFrame.
+        """
+        from pandas import Series
+
+        # reindex the axis to the new value
+        # and set inplace
+        if self.ndim == 1:
+            index = self.obj.index
+            new_index = index.insert(len(index), indexer)
+
+            # we have a coerced indexer, e.g. a float
+            # that matches in an int64 Index, so
+            # we will not create a duplicate index, rather
+            # index to that element
+            # e.g. 0.0 -> 0
+            # GH#12246
+            if index.is_unique:
+                # pass new_index[-1:] instead if [new_index[-1]]
+                #  so that we retain dtype
+                new_indexer = index.get_indexer(new_index[-1:])
+                if (new_indexer != -1).any():
+                    # We get only here with loc, so can hard code
+                    return self._setitem_with_indexer(new_indexer, value, "loc")
+
+            # this preserves dtype of the value and of the object
+            if not is_scalar(value):
+                new_dtype = None
+
+            elif is_valid_na_for_dtype(value, self.obj.dtype):
+                if not is_object_dtype(self.obj.dtype):
+                    # Every NA value is suitable for object, no conversion needed
+                    value = na_value_for_dtype(self.obj.dtype, compat=False)
+
+                new_dtype = maybe_promote(self.obj.dtype, value)[0]
+
+            elif isna(value):
+                new_dtype = None
+            elif not self.obj.empty and not is_object_dtype(self.obj.dtype):
+                # We should not cast, if we have object dtype because we can
+                # set timedeltas into object series
+                curr_dtype = self.obj.dtype
+                curr_dtype = getattr(curr_dtype, "numpy_dtype", curr_dtype)
+                new_dtype = maybe_promote(curr_dtype, value)[0]
+            else:
+                new_dtype = None
+
+            new_values = Series([value], dtype=new_dtype)._values
+
+            if len(self.obj._values):
+                # GH#22717 handle casting compatibility that np.concatenate
+                #  does incorrectly
+                new_values = concat_compat([self.obj._values, new_values])
+            self.obj._mgr = self.obj._constructor(
+                new_values, index=new_index, name=self.obj.name
+            )._mgr
+
+        elif self.ndim == 2:
+            if not len(self.obj.columns):
+                # no columns and scalar
+                raise ValueError("cannot set a frame with no defined columns")
+
+            has_dtype = hasattr(value, "dtype")
+            if isinstance(value, ABCSeries):
+                # append a Series
+                value = value.reindex(index=self.obj.columns)
+                value.name = indexer
+            elif isinstance(value, dict):
+                value = Series(
+                    value, index=self.obj.columns, name=indexer, dtype=object
+                )
+            else:
+                # a list-list
+                if is_list_like_indexer(value):
+                    # must have conforming columns
+                    if len(value) != len(self.obj.columns):
+                        raise ValueError("cannot set a row with mismatched columns")
+
+                value = Series(value, index=self.obj.columns, name=indexer)
+
+            if not len(self.obj):
+                # We will ignore the existing dtypes instead of using
+                #  internals.concat logic
+                df = value.to_frame().T
+
+                idx = self.obj.index
+                if isinstance(idx, MultiIndex):
+                    name = idx.names
+                else:
+                    name = idx.name
+
+                df.index = Index([indexer], name=name)
+                if not has_dtype:
+                    # i.e. if we already had a Series or ndarray, keep that
+                    #  dtype.  But if we had a list or dict, then do inference
+                    df = df.infer_objects()
+                self.obj._mgr = df._mgr
+            else:
+                self.obj._mgr = self.obj._append(value)._mgr
+
+    def _ensure_iterable_column_indexer(self, column_indexer):
+        """
+        Ensure that our column indexer is something that can be iterated over.
+        """
+        ilocs: Sequence[int | np.integer] | np.ndarray | range
+        if is_integer(column_indexer):
+            ilocs = [column_indexer]
+        elif isinstance(column_indexer, slice):
+            ilocs = range(len(self.obj.columns))[column_indexer]
+        elif (
+            isinstance(column_indexer, np.ndarray) and column_indexer.dtype.kind == "b"
+        ):
+            ilocs = np.arange(len(column_indexer))[column_indexer]
+        else:
+            ilocs = column_indexer
+        return ilocs
+
+    def _align_series(
+        self,
+        indexer,
+        ser: Series,
+        multiindex_indexer: bool = False,
+        using_cow: bool = False,
+    ):
+        """
+        Parameters
+        ----------
+        indexer : tuple, slice, scalar
+            Indexer used to get the locations that will be set to `ser`.
+        ser : pd.Series
+            Values to assign to the locations specified by `indexer`.
+        multiindex_indexer : bool, optional
+            Defaults to False. Should be set to True if `indexer` was from
+            a `pd.MultiIndex`, to avoid unnecessary broadcasting.
+
+        Returns
+        -------
+        `np.array` of `ser` broadcast to the appropriate shape for assignment
+        to the locations selected by `indexer`
+        """
+        if isinstance(indexer, (slice, np.ndarray, list, Index)):
+            indexer = (indexer,)
+
+        if isinstance(indexer, tuple):
+            # flatten np.ndarray indexers
+            def ravel(i):
+                return i.ravel() if isinstance(i, np.ndarray) else i
+
+            indexer = tuple(map(ravel, indexer))
+
+            aligners = [not com.is_null_slice(idx) for idx in indexer]
+            sum_aligners = sum(aligners)
+            single_aligner = sum_aligners == 1
+            is_frame = self.ndim == 2
+            obj = self.obj
+
+            # are we a single alignable value on a non-primary
+            # dim (e.g. panel: 1,2, or frame: 0) ?
+            # hence need to align to a single axis dimension
+            # rather that find all valid dims
+
+            # frame
+            if is_frame:
+                single_aligner = single_aligner and aligners[0]
+
+            # we have a frame, with multiple indexers on both axes; and a
+            # series, so need to broadcast (see GH5206)
+            if sum_aligners == self.ndim and all(is_sequence(_) for _ in indexer):
+                ser_values = ser.reindex(obj.axes[0][indexer[0]])._values
+
+                # single indexer
+                if len(indexer) > 1 and not multiindex_indexer:
+                    len_indexer = len(indexer[1])
+                    ser_values = (
+                        np.tile(ser_values, len_indexer).reshape(len_indexer, -1).T
+                    )
+
+                return ser_values
+
+            for i, idx in enumerate(indexer):
+                ax = obj.axes[i]
+
+                # multiple aligners (or null slices)
+                if is_sequence(idx) or isinstance(idx, slice):
+                    if single_aligner and com.is_null_slice(idx):
+                        continue
+                    new_ix = ax[idx]
+                    if not is_list_like_indexer(new_ix):
+                        new_ix = Index([new_ix])
+                    else:
+                        new_ix = Index(new_ix)
+                    if not len(new_ix) or ser.index.equals(new_ix):
+                        if using_cow:
+                            return ser
+                        return ser._values.copy()
+
+                    return ser.reindex(new_ix)._values
+
+                # 2 dims
+                elif single_aligner:
+                    # reindex along index
+                    ax = self.obj.axes[1]
+                    if ser.index.equals(ax) or not len(ax):
+                        return ser._values.copy()
+                    return ser.reindex(ax)._values
+
+        elif is_integer(indexer) and self.ndim == 1:
+            if is_object_dtype(self.obj.dtype):
+                return ser
+            ax = self.obj._get_axis(0)
+
+            if ser.index.equals(ax):
+                return ser._values.copy()
+
+            return ser.reindex(ax)._values[indexer]
+
+        elif is_integer(indexer):
+            ax = self.obj._get_axis(1)
+
+            if ser.index.equals(ax):
+                return ser._values.copy()
+
+            return ser.reindex(ax)._values
+
+        raise ValueError("Incompatible indexer with Series")
+
+    def _align_frame(self, indexer, df: DataFrame) -> DataFrame:
+        is_frame = self.ndim == 2
+
+        if isinstance(indexer, tuple):
+            idx, cols = None, None
+            sindexers = []
+            for i, ix in enumerate(indexer):
+                ax = self.obj.axes[i]
+                if is_sequence(ix) or isinstance(ix, slice):
+                    if isinstance(ix, np.ndarray):
+                        ix = ix.reshape(-1)
+                    if idx is None:
+                        idx = ax[ix]
+                    elif cols is None:
+                        cols = ax[ix]
+                    else:
+                        break
+                else:
+                    sindexers.append(i)
+
+            if idx is not None and cols is not None:
+                if df.index.equals(idx) and df.columns.equals(cols):
+                    val = df.copy()
+                else:
+                    val = df.reindex(idx, columns=cols)
+                return val
+
+        elif (isinstance(indexer, slice) or is_list_like_indexer(indexer)) and is_frame:
+            ax = self.obj.index[indexer]
+            if df.index.equals(ax):
+                val = df.copy()
+            else:
+                # we have a multi-index and are trying to align
+                # with a particular, level GH3738
+                if (
+                    isinstance(ax, MultiIndex)
+                    and isinstance(df.index, MultiIndex)
+                    and ax.nlevels != df.index.nlevels
+                ):
+                    raise TypeError(
+                        "cannot align on a multi-index with out "
+                        "specifying the join levels"
+                    )
+
+                val = df.reindex(index=ax)
+            return val
+
+        raise ValueError("Incompatible indexer with DataFrame")
+
+
+class _ScalarAccessIndexer(NDFrameIndexerBase):
+    """
+    Access scalars quickly.
+    """
+
+    # sub-classes need to set _takeable
+    _takeable: bool
+
+    def _convert_key(self, key):
+        raise AbstractMethodError(self)
+
+    def __getitem__(self, key):
+        if not isinstance(key, tuple):
+            # we could have a convertible item here (e.g. Timestamp)
+            if not is_list_like_indexer(key):
+                key = (key,)
+            else:
+                raise ValueError("Invalid call for scalar access (getting)!")
+
+        key = self._convert_key(key)
+        return self.obj._get_value(*key, takeable=self._takeable)
+
+    def __setitem__(self, key, value) -> None:
+        if isinstance(key, tuple):
+            key = tuple(com.apply_if_callable(x, self.obj) for x in key)
+        else:
+            # scalar callable may return tuple
+            key = com.apply_if_callable(key, self.obj)
+
+        if not isinstance(key, tuple):
+            key = _tuplify(self.ndim, key)
+        key = list(self._convert_key(key))
+        if len(key) != self.ndim:
+            raise ValueError("Not enough indexers for scalar access (setting)!")
+
+        self.obj._set_value(*key, value=value, takeable=self._takeable)
+
+
+@doc(IndexingMixin.at)
+class _AtIndexer(_ScalarAccessIndexer):
+    _takeable = False
+
+    def _convert_key(self, key):
+        """
+        Require they keys to be the same type as the index. (so we don't
+        fallback)
+        """
+        # GH 26989
+        # For series, unpacking key needs to result in the label.
+        # This is already the case for len(key) == 1; e.g. (1,)
+        if self.ndim == 1 and len(key) > 1:
+            key = (key,)
+
+        return key
+
+    @property
+    def _axes_are_unique(self) -> bool:
+        # Only relevant for self.ndim == 2
+        assert self.ndim == 2
+        return self.obj.index.is_unique and self.obj.columns.is_unique
+
+    def __getitem__(self, key):
+        if self.ndim == 2 and not self._axes_are_unique:
+            # GH#33041 fall back to .loc
+            if not isinstance(key, tuple) or not all(is_scalar(x) for x in key):
+                raise ValueError("Invalid call for scalar access (getting)!")
+            return self.obj.loc[key]
+
+        return super().__getitem__(key)
+
+    def __setitem__(self, key, value) -> None:
+        if self.ndim == 2 and not self._axes_are_unique:
+            # GH#33041 fall back to .loc
+            if not isinstance(key, tuple) or not all(is_scalar(x) for x in key):
+                raise ValueError("Invalid call for scalar access (setting)!")
+
+            self.obj.loc[key] = value
+            return
+
+        return super().__setitem__(key, value)
+
+
+@doc(IndexingMixin.iat)
+class _iAtIndexer(_ScalarAccessIndexer):
+    _takeable = True
+
+    def _convert_key(self, key):
+        """
+        Require integer args. (and convert to label arguments)
+        """
+        for i in key:
+            if not is_integer(i):
+                raise ValueError("iAt based indexing can only have integer indexers")
+        return key
+
+
+def _tuplify(ndim: int, loc: Hashable) -> tuple[Hashable | slice, ...]:
+    """
+    Given an indexer for the first dimension, create an equivalent tuple
+    for indexing over all dimensions.
+
+    Parameters
+    ----------
+    ndim : int
+    loc : object
+
+    Returns
+    -------
+    tuple
+    """
+    _tup: list[Hashable | slice]
+    _tup = [slice(None, None) for _ in range(ndim)]
+    _tup[0] = loc
+    return tuple(_tup)
+
+
+def _tupleize_axis_indexer(ndim: int, axis: AxisInt, key) -> tuple:
+    """
+    If we have an axis, adapt the given key to be axis-independent.
+    """
+    new_key = [slice(None)] * ndim
+    new_key[axis] = key
+    return tuple(new_key)
+
+
+def check_bool_indexer(index: Index, key) -> np.ndarray:
+    """
+    Check if key is a valid boolean indexer for an object with such index and
+    perform reindexing or conversion if needed.
+
+    This function assumes that is_bool_indexer(key) == True.
+
+    Parameters
+    ----------
+    index : Index
+        Index of the object on which the indexing is done.
+    key : list-like
+        Boolean indexer to check.
+
+    Returns
+    -------
+    np.array
+        Resulting key.
+
+    Raises
+    ------
+    IndexError
+        If the key does not have the same length as index.
+    IndexingError
+        If the index of the key is unalignable to index.
+    """
+    result = key
+    if isinstance(key, ABCSeries) and not key.index.equals(index):
+        indexer = result.index.get_indexer_for(index)
+        if -1 in indexer:
+            raise IndexingError(
+                "Unalignable boolean Series provided as "
+                "indexer (index of the boolean Series and of "
+                "the indexed object do not match)."
+            )
+
+        result = result.take(indexer)
+
+        # fall through for boolean
+        if not isinstance(result.dtype, ExtensionDtype):
+            return result.astype(bool)._values
+
+    if is_object_dtype(key):
+        # key might be object-dtype bool, check_array_indexer needs bool array
+        result = np.asarray(result, dtype=bool)
+    elif not is_array_like(result):
+        # GH 33924
+        # key may contain nan elements, check_array_indexer needs bool array
+        result = pd_array(result, dtype=bool)
+    return check_array_indexer(index, result)
+
+
+def convert_missing_indexer(indexer):
+    """
+    Reverse convert a missing indexer, which is a dict
+    return the scalar indexer and a boolean indicating if we converted
+    """
+    if isinstance(indexer, dict):
+        # a missing key (but not a tuple indexer)
+        indexer = indexer["key"]
+
+        if isinstance(indexer, bool):
+            raise KeyError("cannot use a single bool to index into setitem")
+        return indexer, True
+
+    return indexer, False
+
+
+def convert_from_missing_indexer_tuple(indexer, axes):
+    """
+    Create a filtered indexer that doesn't have any missing indexers.
+    """
+
+    def get_indexer(_i, _idx):
+        return axes[_i].get_loc(_idx["key"]) if isinstance(_idx, dict) else _idx
+
+    return tuple(get_indexer(_i, _idx) for _i, _idx in enumerate(indexer))
+
+
+def maybe_convert_ix(*args):
+    """
+    We likely want to take the cross-product.
+    """
+    for arg in args:
+        if not isinstance(arg, (np.ndarray, list, ABCSeries, Index)):
+            return args
+    return np.ix_(*args)
+
+
+def is_nested_tuple(tup, labels) -> bool:
+    """
+    Returns
+    -------
+    bool
+    """
+    # check for a compatible nested tuple and multiindexes among the axes
+    if not isinstance(tup, tuple):
+        return False
+
+    for k in tup:
+        if is_list_like(k) or isinstance(k, slice):
+            return isinstance(labels, MultiIndex)
+
+    return False
+
+
+def is_label_like(key) -> bool:
+    """
+    Returns
+    -------
+    bool
+    """
+    # select a label or row
+    return (
+        not isinstance(key, slice)
+        and not is_list_like_indexer(key)
+        and key is not Ellipsis
+    )
+
+
+def need_slice(obj: slice) -> bool:
+    """
+    Returns
+    -------
+    bool
+    """
+    return (
+        obj.start is not None
+        or obj.stop is not None
+        or (obj.step is not None and obj.step != 1)
+    )
+
+
+def check_dict_or_set_indexers(key) -> None:
+    """
+    Check if the indexer is or contains a dict or set, which is no longer allowed.
+    """
+    if isinstance(key, set) or (
+        isinstance(key, tuple) and any(isinstance(x, set) for x in key)
+    ):
+        raise TypeError(
+            "Passing a set as an indexer is not supported. Use a list instead."
+        )
+
+    if isinstance(key, dict) or (
+        isinstance(key, tuple) and any(isinstance(x, dict) for x in key)
+    ):
+        raise TypeError(
+            "Passing a dict as an indexer is not supported. Use a list instead."
+        )
diff --git a/pandas/pandas/core/interchange/__init__.py b/pandas/pandas/core/interchange/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/interchange/buffer.py b/pandas/pandas/core/interchange/buffer.py
new file mode 100644
index 0000000000000..8953360a91c8e
--- /dev/null
+++ b/pandas/pandas/core/interchange/buffer.py
@@ -0,0 +1,135 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas.core.interchange.dataframe_protocol import (
+    Buffer,
+    DlpackDeviceType,
+)
+
+if TYPE_CHECKING:
+    import numpy as np
+    import pyarrow as pa
+
+
+class PandasBuffer(Buffer):
+    """
+    Data in the buffer is guaranteed to be contiguous in memory.
+    """
+
+    def __init__(self, x: np.ndarray, allow_copy: bool = True) -> None:
+        """
+        Handle only regular columns (= numpy arrays) for now.
+        """
+        if x.strides[0] and not x.strides == (x.dtype.itemsize,):
+            # The protocol does not support strided buffers, so a copy is
+            # necessary. If that's not allowed, we need to raise an exception.
+            if allow_copy:
+                x = x.copy()
+            else:
+                raise RuntimeError(
+                    "Exports cannot be zero-copy in the case of a non-contiguous buffer"
+                )
+
+        # Store the numpy array in which the data resides as a private
+        # attribute, so we can use it to retrieve the public attributes
+        self._x = x
+
+    @property
+    def bufsize(self) -> int:
+        """
+        Buffer size in bytes.
+        """
+        return self._x.size * self._x.dtype.itemsize
+
+    @property
+    def ptr(self) -> int:
+        """
+        Pointer to start of the buffer as an integer.
+        """
+        return self._x.__array_interface__["data"][0]
+
+    def __dlpack__(self) -> Any:
+        """
+        Represent this structure as DLPack interface.
+        """
+        return self._x.__dlpack__()
+
+    def __dlpack_device__(self) -> tuple[DlpackDeviceType, int | None]:
+        """
+        Device type and device ID for where the data in the buffer resides.
+        """
+        return (DlpackDeviceType.CPU, None)
+
+    def __repr__(self) -> str:
+        return (
+            "PandasBuffer("
+            + str(
+                {
+                    "bufsize": self.bufsize,
+                    "ptr": self.ptr,
+                    "device": self.__dlpack_device__()[0].name,
+                }
+            )
+            + ")"
+        )
+
+
+class PandasBufferPyarrow(Buffer):
+    """
+    Data in the buffer is guaranteed to be contiguous in memory.
+    """
+
+    def __init__(
+        self,
+        buffer: pa.Buffer,
+        *,
+        length: int,
+    ) -> None:
+        """
+        Handle pyarrow chunked arrays.
+        """
+        self._buffer = buffer
+        self._length = length
+
+    @property
+    def bufsize(self) -> int:
+        """
+        Buffer size in bytes.
+        """
+        return self._buffer.size
+
+    @property
+    def ptr(self) -> int:
+        """
+        Pointer to start of the buffer as an integer.
+        """
+        return self._buffer.address
+
+    def __dlpack__(self) -> Any:
+        """
+        Represent this structure as DLPack interface.
+        """
+        raise NotImplementedError
+
+    def __dlpack_device__(self) -> tuple[DlpackDeviceType, int | None]:
+        """
+        Device type and device ID for where the data in the buffer resides.
+        """
+        return (DlpackDeviceType.CPU, None)
+
+    def __repr__(self) -> str:
+        return (
+            "PandasBuffer[pyarrow]("
+            + str(
+                {
+                    "bufsize": self.bufsize,
+                    "ptr": self.ptr,
+                    "device": "CPU",
+                }
+            )
+            + ")"
+        )
diff --git a/pandas/pandas/core/interchange/column.py b/pandas/pandas/core/interchange/column.py
new file mode 100644
index 0000000000000..c27a9d8141712
--- /dev/null
+++ b/pandas/pandas/core/interchange/column.py
@@ -0,0 +1,474 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs.lib import infer_dtype
+from pandas._libs.tslibs import iNaT
+from pandas.errors import NoBufferPresent
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.dtypes import BaseMaskedDtype
+
+import pandas as pd
+from pandas import (
+    ArrowDtype,
+    DatetimeTZDtype,
+)
+from pandas.api.types import is_string_dtype
+from pandas.core.interchange.buffer import (
+    PandasBuffer,
+    PandasBufferPyarrow,
+)
+from pandas.core.interchange.dataframe_protocol import (
+    Column,
+    ColumnBuffers,
+    ColumnNullType,
+    DtypeKind,
+)
+from pandas.core.interchange.utils import (
+    ArrowCTypes,
+    Endianness,
+    dtype_to_arrow_c_fmt,
+)
+
+if TYPE_CHECKING:
+    from pandas.core.interchange.dataframe_protocol import Buffer
+
+_NP_KINDS = {
+    "i": DtypeKind.INT,
+    "u": DtypeKind.UINT,
+    "f": DtypeKind.FLOAT,
+    "b": DtypeKind.BOOL,
+    "U": DtypeKind.STRING,
+    "M": DtypeKind.DATETIME,
+    "m": DtypeKind.DATETIME,
+}
+
+_NULL_DESCRIPTION = {
+    DtypeKind.FLOAT: (ColumnNullType.USE_NAN, None),
+    DtypeKind.DATETIME: (ColumnNullType.USE_SENTINEL, iNaT),
+    DtypeKind.INT: (ColumnNullType.NON_NULLABLE, None),
+    DtypeKind.UINT: (ColumnNullType.NON_NULLABLE, None),
+    DtypeKind.BOOL: (ColumnNullType.NON_NULLABLE, None),
+    # Null values for categoricals are stored as `-1` sentinel values
+    # in the category date (e.g., `col.values.codes` is int8 np.ndarray)
+    DtypeKind.CATEGORICAL: (ColumnNullType.USE_SENTINEL, -1),
+    # follow Arrow in using 1 as valid value and 0 for missing/null value
+    DtypeKind.STRING: (ColumnNullType.USE_BYTEMASK, 0),
+}
+
+_NO_VALIDITY_BUFFER = {
+    ColumnNullType.NON_NULLABLE: "This column is non-nullable",
+    ColumnNullType.USE_NAN: "This column uses NaN as null",
+    ColumnNullType.USE_SENTINEL: "This column uses a sentinel value",
+}
+
+
+class PandasColumn(Column):
+    """
+    A column object, with only the methods and properties required by the
+    interchange protocol defined.
+    A column can contain one or more chunks. Each chunk can contain up to three
+    buffers - a data buffer, a mask buffer (depending on null representation),
+    and an offsets buffer (if variable-size binary; e.g., variable-length
+    strings).
+    Note: this Column object can only be produced by ``__dataframe__``, so
+          doesn't need its own version or ``__column__`` protocol.
+    """
+
+    def __init__(self, column: pd.Series, allow_copy: bool = True) -> None:
+        """
+        Note: doesn't deal with extension arrays yet, just assume a regular
+        Series/ndarray for now.
+        """
+        if isinstance(column, pd.DataFrame):
+            raise TypeError(
+                "Expected a Series, got a DataFrame. This likely happened "
+                "because you called __dataframe__ on a DataFrame which, "
+                "after converting column names to string, resulted in duplicated "
+                f"names: {column.columns}. Please rename these columns before "
+                "using the interchange protocol."
+            )
+        if not isinstance(column, pd.Series):
+            raise NotImplementedError(f"Columns of type {type(column)} not handled yet")
+
+        # Store the column as a private attribute
+        self._col = column
+        self._allow_copy = allow_copy
+
+    def size(self) -> int:
+        """
+        Size of the column, in elements.
+        """
+        return self._col.size
+
+    @property
+    def offset(self) -> int:
+        """
+        Offset of first element. Always zero.
+        """
+        # TODO: chunks are implemented now, probably this should return something
+        return 0
+
+    @cache_readonly
+    def dtype(self) -> tuple[DtypeKind, int, str, str]:
+        dtype = self._col.dtype
+
+        if isinstance(dtype, pd.CategoricalDtype):
+            codes = self._col.values.codes
+            (
+                _,
+                bitwidth,
+                c_arrow_dtype_f_str,
+                _,
+            ) = self._dtype_from_pandasdtype(codes.dtype)
+            return (
+                DtypeKind.CATEGORICAL,
+                bitwidth,
+                c_arrow_dtype_f_str,
+                Endianness.NATIVE,
+            )
+        elif is_string_dtype(dtype):
+            if infer_dtype(self._col) in ("string", "empty"):
+                return (
+                    DtypeKind.STRING,
+                    8,
+                    dtype_to_arrow_c_fmt(dtype),
+                    Endianness.NATIVE,
+                )
+            raise NotImplementedError("Non-string object dtypes are not supported yet")
+        else:
+            return self._dtype_from_pandasdtype(dtype)
+
+    def _dtype_from_pandasdtype(self, dtype) -> tuple[DtypeKind, int, str, str]:
+        """
+        See `self.dtype` for details.
+        """
+        # Note: 'c' (complex) not handled yet (not in array spec v1).
+        #       'b', 'B' (bytes), 'S', 'a', (old-style string) 'V' (void) not handled
+        #       datetime and timedelta both map to datetime (is timedelta handled?)
+
+        kind = _NP_KINDS.get(dtype.kind, None)
+        if kind is None:
+            # Not a NumPy dtype. Check if it's a categorical maybe
+            raise ValueError(f"Data type {dtype} not supported by interchange protocol")
+        if isinstance(dtype, ArrowDtype):
+            byteorder = dtype.numpy_dtype.byteorder
+        elif isinstance(dtype, DatetimeTZDtype):
+            byteorder = dtype.base.byteorder  # type: ignore[union-attr]
+        elif isinstance(dtype, BaseMaskedDtype):
+            byteorder = dtype.numpy_dtype.byteorder
+        else:
+            byteorder = dtype.byteorder
+
+        if dtype == "bool[pyarrow]":
+            # return early to avoid the `* 8` below, as this is a bitmask
+            # rather than a bytemask
+            return (
+                kind,
+                dtype.itemsize,  # pyright: ignore[reportAttributeAccessIssue]
+                ArrowCTypes.BOOL,
+                byteorder,
+            )
+
+        return kind, dtype.itemsize * 8, dtype_to_arrow_c_fmt(dtype), byteorder
+
+    @property
+    def describe_categorical(self):
+        """
+        If the dtype is categorical, there are two options:
+        - There are only values in the data buffer.
+        - There is a separate non-categorical Column encoding for categorical values.
+
+        Raises TypeError if the dtype is not categorical
+
+        Content of returned dict:
+            - "is_ordered" : bool, whether the ordering of dictionary indices is
+                             semantically meaningful.
+            - "is_dictionary" : bool, whether a dictionary-style mapping of
+                                categorical values to other objects exists
+            - "categories" : Column representing the (implicit) mapping of indices to
+                             category values (e.g. an array of cat1, cat2, ...).
+                             None if not a dictionary-style categorical.
+        """
+        if not self.dtype[0] == DtypeKind.CATEGORICAL:
+            raise TypeError(
+                "describe_categorical only works on a column with categorical dtype!"
+            )
+
+        return {
+            "is_ordered": self._col.cat.ordered,
+            "is_dictionary": True,
+            "categories": PandasColumn(pd.Series(self._col.cat.categories)),
+        }
+
+    @property
+    def describe_null(self):
+        if isinstance(self._col.dtype, BaseMaskedDtype):
+            column_null_dtype = ColumnNullType.USE_BYTEMASK
+            null_value = 1
+            return column_null_dtype, null_value
+        if isinstance(self._col.dtype, ArrowDtype):
+            # We already rechunk (if necessary / allowed) upon initialization, so this
+            # is already single-chunk by the time we get here.
+            if self._col.array._pa_array.chunks[0].buffers()[0] is None:  # type: ignore[attr-defined]
+                return ColumnNullType.NON_NULLABLE, None
+            return ColumnNullType.USE_BITMASK, 0
+        kind = self.dtype[0]
+        try:
+            null, value = _NULL_DESCRIPTION[kind]
+        except KeyError as err:
+            raise NotImplementedError(f"Data type {kind} not yet supported") from err
+
+        return null, value
+
+    @cache_readonly
+    def null_count(self) -> int:
+        """
+        Number of null elements. Should always be known.
+        """
+        return self._col.isna().sum().item()
+
+    @property
+    def metadata(self) -> dict[str, pd.Index]:
+        """
+        Store specific metadata of the column.
+        """
+        return {"pandas.index": self._col.index}
+
+    def num_chunks(self) -> int:
+        """
+        Return the number of chunks the column consists of.
+        """
+        return 1
+
+    def get_chunks(self, n_chunks: int | None = None):
+        """
+        Return an iterator yielding the chunks.
+        See `DataFrame.get_chunks` for details on ``n_chunks``.
+        """
+        if n_chunks and n_chunks > 1:
+            size = len(self._col)
+            step = size // n_chunks
+            if size % n_chunks != 0:
+                step += 1
+            for start in range(0, step * n_chunks, step):
+                yield PandasColumn(
+                    self._col.iloc[start : start + step], self._allow_copy
+                )
+        else:
+            yield self
+
+    def get_buffers(self) -> ColumnBuffers:
+        """
+        Return a dictionary containing the underlying buffers.
+        The returned dictionary has the following contents:
+            - "data": a two-element tuple whose first element is a buffer
+                      containing the data and whose second element is the data
+                      buffer's associated dtype.
+            - "validity": a two-element tuple whose first element is a buffer
+                          containing mask values indicating missing data and
+                          whose second element is the mask value buffer's
+                          associated dtype. None if the null representation is
+                          not a bit or byte mask.
+            - "offsets": a two-element tuple whose first element is a buffer
+                         containing the offset values for variable-size binary
+                         data (e.g., variable-length strings) and whose second
+                         element is the offsets buffer's associated dtype. None
+                         if the data buffer does not have an associated offsets
+                         buffer.
+        """
+        buffers: ColumnBuffers = {
+            "data": self._get_data_buffer(),
+            "validity": None,
+            "offsets": None,
+        }
+
+        try:
+            buffers["validity"] = self._get_validity_buffer()
+        except NoBufferPresent:
+            pass
+
+        try:
+            buffers["offsets"] = self._get_offsets_buffer()
+        except NoBufferPresent:
+            pass
+
+        return buffers
+
+    def _get_data_buffer(
+        self,
+    ) -> tuple[Buffer, tuple[DtypeKind, int, str, str]]:
+        """
+        Return the buffer containing the data and the buffer's associated dtype.
+        """
+        buffer: Buffer
+        if self.dtype[0] == DtypeKind.DATETIME:
+            # self.dtype[2] is an ArrowCTypes.TIMESTAMP where the tz will make
+            # it longer than 4 characters
+            if len(self.dtype[2]) > 4:
+                np_arr = self._col.dt.tz_convert(None).to_numpy()
+            else:
+                np_arr = self._col.to_numpy()
+            buffer = PandasBuffer(np_arr, allow_copy=self._allow_copy)
+            dtype = (
+                DtypeKind.INT,
+                64,
+                ArrowCTypes.INT64,
+                Endianness.NATIVE,
+            )
+        elif self.dtype[0] in (
+            DtypeKind.INT,
+            DtypeKind.UINT,
+            DtypeKind.FLOAT,
+            DtypeKind.BOOL,
+        ):
+            dtype = self.dtype
+            arr = self._col.array
+            if isinstance(self._col.dtype, ArrowDtype):
+                # We already rechunk (if necessary / allowed) upon initialization, so
+                # this is already single-chunk by the time we get here.
+                arr = arr._pa_array.chunks[0]  # type: ignore[attr-defined]
+                buffer = PandasBufferPyarrow(
+                    arr.buffers()[1],  # type: ignore[attr-defined]
+                    length=len(arr),
+                )
+                return buffer, dtype
+            if isinstance(self._col.dtype, BaseMaskedDtype):
+                np_arr = arr._data  # type: ignore[attr-defined]
+            else:
+                np_arr = arr._ndarray  # type: ignore[attr-defined]
+            buffer = PandasBuffer(np_arr, allow_copy=self._allow_copy)
+        elif self.dtype[0] == DtypeKind.CATEGORICAL:
+            codes = self._col.values._codes
+            buffer = PandasBuffer(codes, allow_copy=self._allow_copy)
+            dtype = self._dtype_from_pandasdtype(codes.dtype)
+        elif self.dtype[0] == DtypeKind.STRING:
+            # Marshal the strings from a NumPy object array into a byte array
+            buf = self._col.to_numpy()
+            b = bytearray()
+
+            # TODO: this for-loop is slow; can be implemented in Cython/C/C++ later
+            for obj in buf:
+                if isinstance(obj, str):
+                    b.extend(obj.encode(encoding="utf-8"))
+
+            # Convert the byte array to a Pandas "buffer" using
+            # a NumPy array as the backing store
+            buffer = PandasBuffer(np.frombuffer(b, dtype="uint8"))
+
+            # Define the dtype for the returned buffer
+            # TODO: this will need correcting
+            # https://github.com/pandas-dev/pandas/issues/54781
+            dtype = (
+                DtypeKind.UINT,
+                8,
+                ArrowCTypes.UINT8,
+                Endianness.NATIVE,
+            )  # note: currently only support native endianness
+        else:
+            raise NotImplementedError(f"Data type {self._col.dtype} not handled yet")
+
+        return buffer, dtype
+
+    def _get_validity_buffer(self) -> tuple[Buffer, Any] | None:
+        """
+        Return the buffer containing the mask values indicating missing data and
+        the buffer's associated dtype.
+        Raises NoBufferPresent if null representation is not a bit or byte mask.
+        """
+        null, invalid = self.describe_null
+        buffer: Buffer
+        if isinstance(self._col.dtype, ArrowDtype):
+            # We already rechunk (if necessary / allowed) upon initialization, so this
+            # is already single-chunk by the time we get here.
+            arr = self._col.array._pa_array.chunks[0]  # type: ignore[attr-defined]
+            dtype = (DtypeKind.BOOL, 1, ArrowCTypes.BOOL, Endianness.NATIVE)
+            if arr.buffers()[0] is None:
+                return None
+            buffer = PandasBufferPyarrow(
+                arr.buffers()[0],
+                length=len(arr),
+            )
+            return buffer, dtype
+
+        if isinstance(self._col.dtype, BaseMaskedDtype):
+            mask = self._col.array._mask  # type: ignore[attr-defined]
+            buffer = PandasBuffer(mask)
+            dtype = (DtypeKind.BOOL, 8, ArrowCTypes.BOOL, Endianness.NATIVE)
+            return buffer, dtype
+
+        if self.dtype[0] == DtypeKind.STRING:
+            # For now, use byte array as the mask.
+            # TODO: maybe store as bit array to save space?..
+            buf = self._col.to_numpy()
+
+            # Determine the encoding for valid values
+            valid = invalid == 0
+            invalid = not valid
+
+            mask = np.zeros(shape=(len(buf),), dtype=np.bool_)
+            for i, obj in enumerate(buf):
+                mask[i] = valid if isinstance(obj, str) else invalid
+
+            # Convert the mask array to a Pandas "buffer" using
+            # a NumPy array as the backing store
+            buffer = PandasBuffer(mask)
+
+            # Define the dtype of the returned buffer
+            dtype = (DtypeKind.BOOL, 8, ArrowCTypes.BOOL, Endianness.NATIVE)
+
+            return buffer, dtype
+
+        try:
+            msg = f"{_NO_VALIDITY_BUFFER[null]} so does not have a separate mask"
+        except KeyError as err:
+            # TODO: implement for other bit/byte masks?
+            raise NotImplementedError("See self.describe_null") from err
+
+        raise NoBufferPresent(msg)
+
+    def _get_offsets_buffer(self) -> tuple[PandasBuffer, Any]:
+        """
+        Return the buffer containing the offset values for variable-size binary
+        data (e.g., variable-length strings) and the buffer's associated dtype.
+        Raises NoBufferPresent if the data buffer does not have an associated
+        offsets buffer.
+        """
+        if self.dtype[0] == DtypeKind.STRING:
+            # For each string, we need to manually determine the next offset
+            values = self._col.to_numpy()
+            ptr = 0
+            offsets = np.zeros(shape=(len(values) + 1,), dtype=np.int64)
+            for i, v in enumerate(values):
+                # For missing values (in this case, `np.nan` values)
+                # we don't increment the pointer
+                if isinstance(v, str):
+                    b = v.encode(encoding="utf-8")
+                    ptr += len(b)
+
+                offsets[i + 1] = ptr
+
+            # Convert the offsets to a Pandas "buffer" using
+            # the NumPy array as the backing store
+            buffer = PandasBuffer(offsets)
+
+            # Assemble the buffer dtype info
+            dtype = (
+                DtypeKind.INT,
+                64,
+                ArrowCTypes.INT64,
+                Endianness.NATIVE,
+            )  # note: currently only support native endianness
+        else:
+            raise NoBufferPresent(
+                "This column has a fixed-length dtype so "
+                "it does not have an offsets buffer"
+            )
+
+        return buffer, dtype
diff --git a/pandas/pandas/core/interchange/dataframe.py b/pandas/pandas/core/interchange/dataframe.py
new file mode 100644
index 0000000000000..0a116af567e59
--- /dev/null
+++ b/pandas/pandas/core/interchange/dataframe.py
@@ -0,0 +1,113 @@
+from __future__ import annotations
+
+from collections import abc
+from typing import TYPE_CHECKING
+
+from pandas.core.interchange.column import PandasColumn
+from pandas.core.interchange.dataframe_protocol import DataFrame as DataFrameXchg
+from pandas.core.interchange.utils import maybe_rechunk
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Iterable,
+        Sequence,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+    )
+
+
+class PandasDataFrameXchg(DataFrameXchg):
+    """
+    A data frame class, with only the methods required by the interchange
+    protocol defined.
+    Instances of this (private) class are returned from
+    ``pd.DataFrame.__dataframe__`` as objects with the methods and
+    attributes defined on this class.
+    """
+
+    def __init__(self, df: DataFrame, allow_copy: bool = True) -> None:
+        """
+        Constructor - an instance of this (private) class is returned from
+        `pd.DataFrame.__dataframe__`.
+        """
+        self._df = df.rename(columns=str)
+        self._allow_copy = allow_copy
+        for i, _col in enumerate(self._df.columns):
+            rechunked = maybe_rechunk(self._df.iloc[:, i], allow_copy=allow_copy)
+            if rechunked is not None:
+                self._df.isetitem(i, rechunked)
+
+    def __dataframe__(
+        self, nan_as_null: bool = False, allow_copy: bool = True
+    ) -> PandasDataFrameXchg:
+        # `nan_as_null` can be removed here once it's removed from
+        # Dataframe.__dataframe__
+        return PandasDataFrameXchg(self._df, allow_copy)
+
+    @property
+    def metadata(self) -> dict[str, Index]:
+        # `index` isn't a regular column, and the protocol doesn't support row
+        # labels - so we export it as Pandas-specific metadata here.
+        return {"pandas.index": self._df.index}
+
+    def num_columns(self) -> int:
+        return len(self._df.columns)
+
+    def num_rows(self) -> int:
+        return len(self._df)
+
+    def num_chunks(self) -> int:
+        return 1
+
+    def column_names(self) -> Index:
+        return self._df.columns
+
+    def get_column(self, i: int) -> PandasColumn:
+        return PandasColumn(self._df.iloc[:, i], allow_copy=self._allow_copy)
+
+    def get_column_by_name(self, name: str) -> PandasColumn:
+        return PandasColumn(self._df[name], allow_copy=self._allow_copy)
+
+    def get_columns(self) -> list[PandasColumn]:
+        return [
+            PandasColumn(self._df[name], allow_copy=self._allow_copy)
+            for name in self._df.columns
+        ]
+
+    def select_columns(self, indices: Sequence[int]) -> PandasDataFrameXchg:
+        if not isinstance(indices, abc.Sequence):
+            raise ValueError("`indices` is not a sequence")
+        if not isinstance(indices, list):
+            indices = list(indices)
+
+        return PandasDataFrameXchg(
+            self._df.iloc[:, indices], allow_copy=self._allow_copy
+        )
+
+    def select_columns_by_name(self, names: list[str]) -> PandasDataFrameXchg:  # type: ignore[override]
+        if not isinstance(names, abc.Sequence):
+            raise ValueError("`names` is not a sequence")
+        if not isinstance(names, list):
+            names = list(names)
+
+        return PandasDataFrameXchg(self._df.loc[:, names], allow_copy=self._allow_copy)
+
+    def get_chunks(self, n_chunks: int | None = None) -> Iterable[PandasDataFrameXchg]:
+        """
+        Return an iterator yielding the chunks.
+        """
+        if n_chunks and n_chunks > 1:
+            size = len(self._df)
+            step = size // n_chunks
+            if size % n_chunks != 0:
+                step += 1
+            for start in range(0, step * n_chunks, step):
+                yield PandasDataFrameXchg(
+                    self._df.iloc[start : start + step, :],
+                    allow_copy=self._allow_copy,
+                )
+        else:
+            yield self
diff --git a/pandas/pandas/core/interchange/dataframe_protocol.py b/pandas/pandas/core/interchange/dataframe_protocol.py
new file mode 100644
index 0000000000000..95e7b6a26f93a
--- /dev/null
+++ b/pandas/pandas/core/interchange/dataframe_protocol.py
@@ -0,0 +1,465 @@
+"""
+A verbatim copy (vendored) of the spec from https://github.com/data-apis/dataframe-api
+"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+import enum
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypedDict,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Iterable,
+        Sequence,
+    )
+
+
+class DlpackDeviceType(enum.IntEnum):
+    """Integer enum for device type codes matching DLPack."""
+
+    CPU = 1
+    CUDA = 2
+    CPU_PINNED = 3
+    OPENCL = 4
+    VULKAN = 7
+    METAL = 8
+    VPI = 9
+    ROCM = 10
+
+
+class DtypeKind(enum.IntEnum):
+    """
+    Integer enum for data types.
+
+    Attributes
+    ----------
+    INT : int
+        Matches to signed integer data type.
+    UINT : int
+        Matches to unsigned integer data type.
+    FLOAT : int
+        Matches to floating point data type.
+    BOOL : int
+        Matches to boolean data type.
+    STRING : int
+        Matches to string data type (UTF-8 encoded).
+    DATETIME : int
+        Matches to datetime data type.
+    CATEGORICAL : int
+        Matches to categorical data type.
+    """
+
+    INT = 0
+    UINT = 1
+    FLOAT = 2
+    BOOL = 20
+    STRING = 21  # UTF-8
+    DATETIME = 22
+    CATEGORICAL = 23
+
+
+class ColumnNullType(enum.IntEnum):
+    """
+    Integer enum for null type representation.
+
+    Attributes
+    ----------
+    NON_NULLABLE : int
+        Non-nullable column.
+    USE_NAN : int
+        Use explicit float NaN value.
+    USE_SENTINEL : int
+        Sentinel value besides NaN/NaT.
+    USE_BITMASK : int
+        The bit is set/unset representing a null on a certain position.
+    USE_BYTEMASK : int
+        The byte is set/unset representing a null on a certain position.
+    """
+
+    NON_NULLABLE = 0
+    USE_NAN = 1
+    USE_SENTINEL = 2
+    USE_BITMASK = 3
+    USE_BYTEMASK = 4
+
+
+class ColumnBuffers(TypedDict):
+    # first element is a buffer containing the column data;
+    # second element is the data buffer's associated dtype
+    data: tuple[Buffer, Any]
+
+    # first element is a buffer containing mask values indicating missing data;
+    # second element is the mask value buffer's associated dtype.
+    # None if the null representation is not a bit or byte mask
+    validity: tuple[Buffer, Any] | None
+
+    # first element is a buffer containing the offset values for
+    # variable-size binary data (e.g., variable-length strings);
+    # second element is the offsets buffer's associated dtype.
+    # None if the data buffer does not have an associated offsets buffer
+    offsets: tuple[Buffer, Any] | None
+
+
+class CategoricalDescription(TypedDict):
+    # whether the ordering of dictionary indices is semantically meaningful
+    is_ordered: bool
+    # whether a dictionary-style mapping of categorical values to other objects exists
+    is_dictionary: bool
+    # Python-level only (e.g. ``{int: str}``).
+    # None if not a dictionary-style categorical.
+    categories: Column | None
+
+
+class Buffer(ABC):
+    """
+    Data in the buffer is guaranteed to be contiguous in memory.
+
+    Note that there is no dtype attribute present, a buffer can be thought of
+    as simply a block of memory. However, if the column that the buffer is
+    attached to has a dtype that's supported by DLPack and ``__dlpack__`` is
+    implemented, then that dtype information will be contained in the return
+    value from ``__dlpack__``.
+
+    This distinction is useful to support both data exchange via DLPack on a
+    buffer and (b) dtypes like variable-length strings which do not have a
+    fixed number of bytes per element.
+    """
+
+    @property
+    @abstractmethod
+    def bufsize(self) -> int:
+        """
+        Buffer size in bytes.
+        """
+
+    @property
+    @abstractmethod
+    def ptr(self) -> int:
+        """
+        Pointer to start of the buffer as an integer.
+        """
+
+    @abstractmethod
+    def __dlpack__(self):
+        """
+        Produce DLPack capsule (see array API standard).
+
+        Raises:
+
+            - TypeError : if the buffer contains unsupported dtypes.
+            - NotImplementedError : if DLPack support is not implemented
+
+        Useful to have to connect to array libraries. Support optional because
+        it's not completely trivial to implement for a Python-only library.
+        """
+        raise NotImplementedError("__dlpack__")
+
+    @abstractmethod
+    def __dlpack_device__(self) -> tuple[DlpackDeviceType, int | None]:
+        """
+        Device type and device ID for where the data in the buffer resides.
+        Uses device type codes matching DLPack.
+        Note: must be implemented even if ``__dlpack__`` is not.
+        """
+
+
+class Column(ABC):
+    """
+    A column object, with only the methods and properties required by the
+    interchange protocol defined.
+
+    A column can contain one or more chunks. Each chunk can contain up to three
+    buffers - a data buffer, a mask buffer (depending on null representation),
+    and an offsets buffer (if variable-size binary; e.g., variable-length
+    strings).
+
+    TBD: Arrow has a separate "null" dtype, and has no separate mask concept.
+         Instead, it seems to use "children" for both columns with a bit mask,
+         and for nested dtypes. Unclear whether this is elegant or confusing.
+         This design requires checking the null representation explicitly.
+
+         The Arrow design requires checking:
+         1. the ARROW_FLAG_NULLABLE (for sentinel values)
+         2. if a column has two children, combined with one of those children
+            having a null dtype.
+
+         Making the mask concept explicit seems useful. One null dtype would
+         not be enough to cover both bit and byte masks, so that would mean
+         even more checking if we did it the Arrow way.
+
+    TBD: there's also the "chunk" concept here, which is implicit in Arrow as
+         multiple buffers per array (= column here). Semantically it may make
+         sense to have both: chunks were meant for example for lazy evaluation
+         of data which doesn't fit in memory, while multiple buffers per column
+         could also come from doing a selection operation on a single
+         contiguous buffer.
+
+         Given these concepts, one would expect chunks to be all of the same
+         size (say a 10,000 row dataframe could have 10 chunks of 1,000 rows),
+         while multiple buffers could have data-dependent lengths. Not an issue
+         in pandas if one column is backed by a single NumPy array, but in
+         Arrow it seems possible.
+         Are multiple chunks *and* multiple buffers per column necessary for
+         the purposes of this interchange protocol, or must producers either
+         reuse the chunk concept for this or copy the data?
+
+    Note: this Column object can only be produced by ``__dataframe__``, so
+          doesn't need its own version or ``__column__`` protocol.
+    """
+
+    @abstractmethod
+    def size(self) -> int:
+        """
+        Size of the column, in elements.
+
+        Corresponds to DataFrame.num_rows() if column is a single chunk;
+        equal to size of this current chunk otherwise.
+        """
+
+    @property
+    @abstractmethod
+    def offset(self) -> int:
+        """
+        Offset of first element.
+
+        May be > 0 if using chunks; for example for a column with N chunks of
+        equal size M (only the last chunk may be shorter),
+        ``offset = n * M``, ``n = 0 .. N-1``.
+        """
+
+    @property
+    @abstractmethod
+    def dtype(self) -> tuple[DtypeKind, int, str, str]:
+        """
+        Dtype description as a tuple ``(kind, bit-width, format string, endianness)``.
+
+        Bit-width : the number of bits as an integer
+        Format string : data type description format string in Apache Arrow C
+                        Data Interface format.
+        Endianness : current only native endianness (``=``) is supported
+
+        Notes:
+            - Kind specifiers are aligned with DLPack where possible (hence the
+              jump to 20, leave enough room for future extension)
+            - Masks must be specified as boolean with either bit width 1 (for bit
+              masks) or 8 (for byte masks).
+            - Dtype width in bits was preferred over bytes
+            - Endianness isn't too useful, but included now in case in the future
+              we need to support non-native endianness
+            - Went with Apache Arrow format strings over NumPy format strings
+              because they're more complete from a dataframe perspective
+            - Format strings are mostly useful for datetime specification, and
+              for categoricals.
+            - For categoricals, the format string describes the type of the
+              categorical in the data buffer. In case of a separate encoding of
+              the categorical (e.g. an integer to string mapping), this can
+              be derived from ``self.describe_categorical``.
+            - Data types not included: complex, Arrow-style null, binary, decimal,
+              and nested (list, struct, map, union) dtypes.
+        """
+
+    @property
+    @abstractmethod
+    def describe_categorical(self) -> CategoricalDescription:
+        """
+        If the dtype is categorical, there are two options:
+        - There are only values in the data buffer.
+        - There is a separate non-categorical Column encoding for categorical values.
+
+        Raises TypeError if the dtype is not categorical
+
+        Returns the dictionary with description on how to interpret the data buffer:
+            - "is_ordered" : bool, whether the ordering of dictionary indices is
+                             semantically meaningful.
+            - "is_dictionary" : bool, whether a mapping of
+                                categorical values to other objects exists
+            - "categories" : Column representing the (implicit) mapping of indices to
+                             category values (e.g. an array of cat1, cat2, ...).
+                             None if not a dictionary-style categorical.
+
+        TBD: are there any other in-memory representations that are needed?
+        """
+
+    @property
+    @abstractmethod
+    def describe_null(self) -> tuple[ColumnNullType, Any]:
+        """
+        Return the missing value (or "null") representation the column dtype
+        uses, as a tuple ``(kind, value)``.
+
+        Value : if kind is "sentinel value", the actual value. If kind is a bit
+        mask or a byte mask, the value (0 or 1) indicating a missing value. None
+        otherwise.
+        """
+
+    @property
+    @abstractmethod
+    def null_count(self) -> int | None:
+        """
+        Number of null elements, if known.
+
+        Note: Arrow uses -1 to indicate "unknown", but None seems cleaner.
+        """
+
+    @property
+    @abstractmethod
+    def metadata(self) -> dict[str, Any]:
+        """
+        The metadata for the column. See `DataFrame.metadata` for more details.
+        """
+
+    @abstractmethod
+    def num_chunks(self) -> int:
+        """
+        Return the number of chunks the column consists of.
+        """
+
+    @abstractmethod
+    def get_chunks(self, n_chunks: int | None = None) -> Iterable[Column]:
+        """
+        Return an iterator yielding the chunks.
+
+        See `DataFrame.get_chunks` for details on ``n_chunks``.
+        """
+
+    @abstractmethod
+    def get_buffers(self) -> ColumnBuffers:
+        """
+        Return a dictionary containing the underlying buffers.
+
+        The returned dictionary has the following contents:
+
+            - "data": a two-element tuple whose first element is a buffer
+                      containing the data and whose second element is the data
+                      buffer's associated dtype.
+            - "validity": a two-element tuple whose first element is a buffer
+                          containing mask values indicating missing data and
+                          whose second element is the mask value buffer's
+                          associated dtype. None if the null representation is
+                          not a bit or byte mask.
+            - "offsets": a two-element tuple whose first element is a buffer
+                         containing the offset values for variable-size binary
+                         data (e.g., variable-length strings) and whose second
+                         element is the offsets buffer's associated dtype. None
+                         if the data buffer does not have an associated offsets
+                         buffer.
+        """
+
+
+#    def get_children(self) -> Iterable[Column]:
+#        """
+#        Children columns underneath the column, each object in this iterator
+#        must adhere to the column specification.
+#        """
+#        pass
+
+
+class DataFrame(ABC):
+    """
+    A data frame class, with only the methods required by the interchange
+    protocol defined.
+
+    A "data frame" represents an ordered collection of named columns.
+    A column's "name" must be a unique string.
+    Columns may be accessed by name or by position.
+
+    This could be a public data frame class, or an object with the methods and
+    attributes defined on this DataFrame class could be returned from the
+    ``__dataframe__`` method of a public data frame class in a library adhering
+    to the dataframe interchange protocol specification.
+    """
+
+    version = 0  # version of the protocol
+
+    @abstractmethod
+    def __dataframe__(self, nan_as_null: bool = False, allow_copy: bool = True):
+        """Construct a new interchange object, potentially changing the parameters."""
+
+    @property
+    @abstractmethod
+    def metadata(self) -> dict[str, Any]:
+        """
+        The metadata for the data frame, as a dictionary with string keys. The
+        contents of `metadata` may be anything, they are meant for a library
+        to store information that it needs to, e.g., roundtrip losslessly or
+        for two implementations to share data that is not (yet) part of the
+        interchange protocol specification. For avoiding collisions with other
+        entries, please add name the keys with the name of the library
+        followed by a period and the desired name, e.g, ``pandas.indexcol``.
+        """
+
+    @abstractmethod
+    def num_columns(self) -> int:
+        """
+        Return the number of columns in the DataFrame.
+        """
+
+    @abstractmethod
+    def num_rows(self) -> int | None:
+        # TODO: not happy with Optional, but need to flag it may be expensive
+        #       why include it if it may be None - what do we expect consumers
+        #       to do here?
+        """
+        Return the number of rows in the DataFrame, if available.
+        """
+
+    @abstractmethod
+    def num_chunks(self) -> int:
+        """
+        Return the number of chunks the DataFrame consists of.
+        """
+
+    @abstractmethod
+    def column_names(self) -> Iterable[str]:
+        """
+        Return an iterator yielding the column names.
+        """
+
+    @abstractmethod
+    def get_column(self, i: int) -> Column:
+        """
+        Return the column at the indicated position.
+        """
+
+    @abstractmethod
+    def get_column_by_name(self, name: str) -> Column:
+        """
+        Return the column whose name is the indicated name.
+        """
+
+    @abstractmethod
+    def get_columns(self) -> Iterable[Column]:
+        """
+        Return an iterator yielding the columns.
+        """
+
+    @abstractmethod
+    def select_columns(self, indices: Sequence[int]) -> DataFrame:
+        """
+        Create a new DataFrame by selecting a subset of columns by index.
+        """
+
+    @abstractmethod
+    def select_columns_by_name(self, names: Sequence[str]) -> DataFrame:
+        """
+        Create a new DataFrame by selecting a subset of columns by name.
+        """
+
+    @abstractmethod
+    def get_chunks(self, n_chunks: int | None = None) -> Iterable[DataFrame]:
+        """
+        Return an iterator yielding the chunks.
+
+        By default (None), yields the chunks that the data is stored as by the
+        producer. If given, ``n_chunks`` must be a multiple of
+        ``self.num_chunks()``, meaning the producer must subdivide each chunk
+        before yielding it.
+        """
diff --git a/pandas/pandas/core/interchange/from_dataframe.py b/pandas/pandas/core/interchange/from_dataframe.py
new file mode 100644
index 0000000000000..b990eca39b3dd
--- /dev/null
+++ b/pandas/pandas/core/interchange/from_dataframe.py
@@ -0,0 +1,590 @@
+from __future__ import annotations
+
+import ctypes
+import re
+from typing import (
+    Any,
+    overload,
+)
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+
+from pandas.compat._optional import import_optional_dependency
+
+import pandas as pd
+from pandas.core.interchange.dataframe_protocol import (
+    Buffer,
+    Column,
+    ColumnNullType,
+    DataFrame as DataFrameXchg,
+    DtypeKind,
+)
+from pandas.core.interchange.utils import (
+    ArrowCTypes,
+    Endianness,
+)
+
+_NP_DTYPES: dict[DtypeKind, dict[int, Any]] = {
+    DtypeKind.INT: {8: np.int8, 16: np.int16, 32: np.int32, 64: np.int64},
+    DtypeKind.UINT: {8: np.uint8, 16: np.uint16, 32: np.uint32, 64: np.uint64},
+    DtypeKind.FLOAT: {32: np.float32, 64: np.float64},
+    DtypeKind.BOOL: {1: bool, 8: bool},
+}
+
+
+def from_dataframe(df, allow_copy: bool = True) -> pd.DataFrame:
+    """
+    Build a ``pd.DataFrame`` from any DataFrame supporting the interchange protocol.
+
+    .. note::
+
+       For new development, we highly recommend using the Arrow C Data Interface
+       alongside the Arrow PyCapsule Interface instead of the interchange protocol.
+       From pandas 3.0 onwards, `from_dataframe` uses the PyCapsule Interface,
+       only falling back to the interchange protocol if that fails.
+
+    .. warning::
+
+        Due to severe implementation issues, we recommend only considering using the
+        interchange protocol in the following cases:
+
+        - converting to pandas: for pandas >= 2.0.3
+        - converting from pandas: for pandas >= 3.0.0
+
+    Parameters
+    ----------
+    df : DataFrameXchg
+        Object supporting the interchange protocol, i.e. `__dataframe__` method.
+    allow_copy : bool, default: True
+        Whether to allow copying the memory to perform the conversion
+        (if false then zero-copy approach is requested).
+
+    Returns
+    -------
+    pd.DataFrame
+        A pandas DataFrame built from the provided interchange
+        protocol object.
+
+    See Also
+    --------
+    pd.DataFrame : DataFrame class which can be created from various input data
+        formats, including objects that support the interchange protocol.
+
+    Examples
+    --------
+    >>> df_not_necessarily_pandas = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+    >>> interchange_object = df_not_necessarily_pandas.__dataframe__()
+    >>> interchange_object.column_names()
+    Index(['A', 'B'], dtype='object')
+    >>> df_pandas = pd.api.interchange.from_dataframe(
+    ...     interchange_object.select_columns_by_name(["A"])
+    ... )
+    >>> df_pandas
+         A
+    0    1
+    1    2
+
+    These methods (``column_names``, ``select_columns_by_name``) should work
+    for any dataframe library which implements the interchange protocol.
+    """
+    if isinstance(df, pd.DataFrame):
+        return df
+
+    if hasattr(df, "__arrow_c_stream__"):
+        try:
+            pa = import_optional_dependency("pyarrow", min_version="14.0.0")
+        except ImportError:
+            # fallback to _from_dataframe
+            pass
+        else:
+            try:
+                return pa.table(df).to_pandas(zero_copy_only=not allow_copy)
+            except pa.ArrowInvalid as e:
+                raise RuntimeError(e) from e
+
+    if not hasattr(df, "__dataframe__"):
+        raise ValueError("`df` does not support __dataframe__")
+
+    return _from_dataframe(
+        df.__dataframe__(allow_copy=allow_copy), allow_copy=allow_copy
+    )
+
+
+def _from_dataframe(df: DataFrameXchg, allow_copy: bool = True) -> pd.DataFrame:
+    """
+    Build a ``pd.DataFrame`` from the DataFrame interchange object.
+
+    Parameters
+    ----------
+    df : DataFrameXchg
+        Object supporting the interchange protocol, i.e. `__dataframe__` method.
+    allow_copy : bool, default: True
+        Whether to allow copying the memory to perform the conversion
+        (if false then zero-copy approach is requested).
+
+    Returns
+    -------
+    pd.DataFrame
+    """
+    pandas_dfs = []
+    for chunk in df.get_chunks():
+        pandas_df = protocol_df_chunk_to_pandas(chunk)
+        pandas_dfs.append(pandas_df)
+
+    if not allow_copy and len(pandas_dfs) > 1:
+        raise RuntimeError(
+            "To join chunks a copy is required which is forbidden by allow_copy=False"
+        )
+    if not pandas_dfs:
+        pandas_df = protocol_df_chunk_to_pandas(df)
+    elif len(pandas_dfs) == 1:
+        pandas_df = pandas_dfs[0]
+    else:
+        pandas_df = pd.concat(pandas_dfs, axis=0, ignore_index=True, copy=False)
+
+    index_obj = df.metadata.get("pandas.index", None)
+    if index_obj is not None:
+        pandas_df.index = index_obj
+
+    return pandas_df
+
+
+def protocol_df_chunk_to_pandas(df: DataFrameXchg) -> pd.DataFrame:
+    """
+    Convert interchange protocol chunk to ``pd.DataFrame``.
+
+    Parameters
+    ----------
+    df : DataFrameXchg
+
+    Returns
+    -------
+    pd.DataFrame
+    """
+    columns: dict[str, Any] = {}
+    buffers = []  # hold on to buffers, keeps memory alive
+    for name in df.column_names():
+        if not isinstance(name, str):
+            raise ValueError(f"Column {name} is not a string")
+        if name in columns:
+            raise ValueError(f"Column {name} is not unique")
+        col = df.get_column_by_name(name)
+        dtype = col.dtype[0]
+        if dtype in (
+            DtypeKind.INT,
+            DtypeKind.UINT,
+            DtypeKind.FLOAT,
+            DtypeKind.BOOL,
+        ):
+            columns[name], buf = primitive_column_to_ndarray(col)
+        elif dtype == DtypeKind.CATEGORICAL:
+            columns[name], buf = categorical_column_to_series(col)
+        elif dtype == DtypeKind.STRING:
+            columns[name], buf = string_column_to_ndarray(col)
+        elif dtype == DtypeKind.DATETIME:
+            columns[name], buf = datetime_column_to_ndarray(col)
+        else:
+            raise NotImplementedError(f"Data type {dtype} not handled yet")
+
+        buffers.append(buf)
+
+    pandas_df = pd.DataFrame(columns)
+    pandas_df.attrs["_INTERCHANGE_PROTOCOL_BUFFERS"] = buffers
+    return pandas_df
+
+
+def primitive_column_to_ndarray(col: Column) -> tuple[np.ndarray, Any]:
+    """
+    Convert a column holding one of the primitive dtypes to a NumPy array.
+
+    A primitive type is one of: int, uint, float, bool.
+
+    Parameters
+    ----------
+    col : Column
+
+    Returns
+    -------
+    tuple
+        Tuple of np.ndarray holding the data and the memory owner object
+        that keeps the memory alive.
+    """
+    buffers = col.get_buffers()
+
+    data_buff, data_dtype = buffers["data"]
+    data = buffer_to_ndarray(
+        data_buff, data_dtype, offset=col.offset, length=col.size()
+    )
+
+    data = set_nulls(data, col, buffers["validity"])
+    return data, buffers
+
+
+def categorical_column_to_series(col: Column) -> tuple[pd.Series, Any]:
+    """
+    Convert a column holding categorical data to a pandas Series.
+
+    Parameters
+    ----------
+    col : Column
+
+    Returns
+    -------
+    tuple
+        Tuple of pd.Series holding the data and the memory owner object
+        that keeps the memory alive.
+    """
+    categorical = col.describe_categorical
+
+    if not categorical["is_dictionary"]:
+        raise NotImplementedError("Non-dictionary categoricals not supported yet")
+
+    cat_column = categorical["categories"]
+    if hasattr(cat_column, "_col"):
+        # Item "Column" of "Optional[Column]" has no attribute "_col"
+        # Item "None" of "Optional[Column]" has no attribute "_col"
+        categories = np.array(cat_column._col)  # type: ignore[union-attr]
+    else:
+        raise NotImplementedError(
+            "Interchanging categorical columns isn't supported yet, and our "
+            "fallback of using the `col._col` attribute (a ndarray) failed."
+        )
+    buffers = col.get_buffers()
+
+    codes_buff, codes_dtype = buffers["data"]
+    codes = buffer_to_ndarray(
+        codes_buff, codes_dtype, offset=col.offset, length=col.size()
+    )
+
+    # Doing module in order to not get ``IndexError`` for
+    # out-of-bounds sentinel values in `codes`
+    if len(categories) > 0:
+        values = categories[codes % len(categories)]
+    else:
+        values = codes
+
+    cat = pd.Categorical(
+        values, categories=categories, ordered=categorical["is_ordered"]
+    )
+    data = pd.Series(cat)
+
+    data = set_nulls(data, col, buffers["validity"])
+    return data, buffers
+
+
+def string_column_to_ndarray(col: Column) -> tuple[np.ndarray, Any]:
+    """
+    Convert a column holding string data to a NumPy array.
+
+    Parameters
+    ----------
+    col : Column
+
+    Returns
+    -------
+    tuple
+        Tuple of np.ndarray holding the data and the memory owner object
+        that keeps the memory alive.
+    """
+    null_kind, sentinel_val = col.describe_null
+
+    if null_kind not in (
+        ColumnNullType.NON_NULLABLE,
+        ColumnNullType.USE_BITMASK,
+        ColumnNullType.USE_BYTEMASK,
+    ):
+        raise NotImplementedError(
+            f"{null_kind} null kind is not yet supported for string columns."
+        )
+
+    buffers = col.get_buffers()
+
+    assert buffers["offsets"], "String buffers must contain offsets"
+    # Retrieve the data buffer containing the UTF-8 code units
+    data_buff, _ = buffers["data"]
+    # We're going to reinterpret the buffer as uint8, so make sure we can do it safely
+    assert col.dtype[2] in (
+        ArrowCTypes.STRING,
+        ArrowCTypes.LARGE_STRING,
+    )  # format_str == utf-8
+    # Convert the buffers to NumPy arrays. In order to go from STRING to
+    # an equivalent ndarray, we claim that the buffer is uint8 (i.e., a byte array)
+    data_dtype = (
+        DtypeKind.UINT,
+        8,
+        ArrowCTypes.UINT8,
+        Endianness.NATIVE,
+    )
+    # Specify zero offset as we don't want to chunk the string data
+    data = buffer_to_ndarray(data_buff, data_dtype, offset=0, length=data_buff.bufsize)
+
+    # Retrieve the offsets buffer containing the index offsets demarcating
+    # the beginning and the ending of each string
+    offset_buff, offset_dtype = buffers["offsets"]
+    # Offsets buffer contains start-stop positions of strings in the data buffer,
+    # meaning that it has more elements than in the data buffer, do `col.size() + 1`
+    # here to pass a proper offsets buffer size
+    offsets = buffer_to_ndarray(
+        offset_buff, offset_dtype, offset=col.offset, length=col.size() + 1
+    )
+
+    null_pos = None
+    if null_kind in (ColumnNullType.USE_BITMASK, ColumnNullType.USE_BYTEMASK):
+        validity = buffers["validity"]
+        if validity is not None:
+            valid_buff, valid_dtype = validity
+            null_pos = buffer_to_ndarray(
+                valid_buff, valid_dtype, offset=col.offset, length=col.size()
+            )
+            if sentinel_val == 0:
+                null_pos = ~null_pos
+
+    # Assemble the strings from the code units
+    str_list: list[None | float | str] = [None] * col.size()
+    for i in range(col.size()):
+        # Check for missing values
+        if null_pos is not None and null_pos[i]:
+            str_list[i] = np.nan
+            continue
+
+        # Extract a range of code units
+        units = data[offsets[i] : offsets[i + 1]]
+
+        # Convert the list of code units to bytes
+        str_bytes = bytes(units)
+
+        # Create the string
+        string = str_bytes.decode(encoding="utf-8")
+
+        # Add to our list of strings
+        str_list[i] = string
+
+    if using_string_dtype():
+        res = pd.Series(str_list, dtype="str")
+    else:
+        res = np.asarray(str_list, dtype="object")  # type: ignore[assignment]
+
+    return res, buffers  # type: ignore[return-value]
+
+
+def parse_datetime_format_str(format_str, data) -> pd.Series | np.ndarray:
+    """Parse datetime `format_str` to interpret the `data`."""
+    # timestamp 'ts{unit}:tz'
+    timestamp_meta = re.match(r"ts([smun]):(.*)", format_str)
+    if timestamp_meta:
+        unit, tz = timestamp_meta.group(1), timestamp_meta.group(2)
+        if unit != "s":
+            # the format string describes only a first letter of the unit, so
+            # add one extra letter to convert the unit to numpy-style:
+            # 'm' -> 'ms', 'u' -> 'us', 'n' -> 'ns'
+            unit += "s"
+        data = data.astype(f"datetime64[{unit}]")
+        if tz != "":
+            data = pd.Series(data).dt.tz_localize("UTC").dt.tz_convert(tz)
+        return data
+
+    # date 'td{Days/Ms}'
+    date_meta = re.match(r"td([Dm])", format_str)
+    if date_meta:
+        unit = date_meta.group(1)
+        if unit == "D":
+            # NumPy doesn't support DAY unit, so converting days to seconds
+            # (converting to uint64 to avoid overflow)
+            data = (data.astype(np.uint64) * (24 * 60 * 60)).astype("datetime64[s]")
+        elif unit == "m":
+            data = data.astype("datetime64[ms]")
+        else:
+            raise NotImplementedError(f"Date unit is not supported: {unit}")
+        return data
+
+    raise NotImplementedError(f"DateTime kind is not supported: {format_str}")
+
+
+def datetime_column_to_ndarray(col: Column) -> tuple[np.ndarray | pd.Series, Any]:
+    """
+    Convert a column holding DateTime data to a NumPy array.
+
+    Parameters
+    ----------
+    col : Column
+
+    Returns
+    -------
+    tuple
+        Tuple of np.ndarray holding the data and the memory owner object
+        that keeps the memory alive.
+    """
+    buffers = col.get_buffers()
+
+    _, col_bit_width, format_str, _ = col.dtype
+    dbuf, _ = buffers["data"]
+    # Consider dtype being `uint` to get number of units passed since the 01.01.1970
+
+    data = buffer_to_ndarray(
+        dbuf,
+        (
+            DtypeKind.INT,
+            col_bit_width,
+            getattr(ArrowCTypes, f"INT{col_bit_width}"),
+            Endianness.NATIVE,
+        ),
+        offset=col.offset,
+        length=col.size(),
+    )
+
+    data = parse_datetime_format_str(format_str, data)  # type: ignore[assignment]
+    data = set_nulls(data, col, buffers["validity"])
+    return data, buffers
+
+
+def buffer_to_ndarray(
+    buffer: Buffer,
+    dtype: tuple[DtypeKind, int, str, str],
+    *,
+    length: int,
+    offset: int = 0,
+) -> np.ndarray:
+    """
+    Build a NumPy array from the passed buffer.
+
+    Parameters
+    ----------
+    buffer : Buffer
+        Buffer to build a NumPy array from.
+    dtype : tuple
+        Data type of the buffer conforming protocol dtypes format.
+    offset : int, default: 0
+        Number of elements to offset from the start of the buffer.
+    length : int, optional
+        If the buffer is a bit-mask, specifies a number of bits to read
+        from the buffer. Has no effect otherwise.
+
+    Returns
+    -------
+    np.ndarray
+
+    Notes
+    -----
+    The returned array doesn't own the memory. The caller of this function is
+    responsible for keeping the memory owner object alive as long as
+    the returned NumPy array is being used.
+    """
+    kind, bit_width, _, _ = dtype
+
+    column_dtype = _NP_DTYPES.get(kind, {}).get(bit_width, None)
+    if column_dtype is None:
+        raise NotImplementedError(f"Conversion for {dtype} is not yet supported.")
+
+    # TODO: No DLPack yet, so need to construct a new ndarray from the data pointer
+    # and size in the buffer plus the dtype on the column. Use DLPack as NumPy supports
+    # it since https://github.com/numpy/numpy/pull/19083
+    ctypes_type = np.ctypeslib.as_ctypes_type(column_dtype)
+
+    if bit_width == 1:
+        assert length is not None, "`length` must be specified for a bit-mask buffer."
+        pa = import_optional_dependency("pyarrow")
+        arr = pa.BooleanArray.from_buffers(
+            pa.bool_(),
+            length,
+            [None, pa.foreign_buffer(buffer.ptr, length)],
+            offset=offset,
+        )
+        return np.asarray(arr)
+    else:
+        data_pointer = ctypes.cast(
+            buffer.ptr + (offset * bit_width // 8), ctypes.POINTER(ctypes_type)
+        )
+        if length > 0:
+            return np.ctypeslib.as_array(data_pointer, shape=(length,))
+        return np.array([], dtype=ctypes_type)
+
+
+@overload
+def set_nulls(
+    data: np.ndarray,
+    col: Column,
+    validity: tuple[Buffer, tuple[DtypeKind, int, str, str]] | None,
+    allow_modify_inplace: bool = ...,
+) -> np.ndarray: ...
+
+
+@overload
+def set_nulls(
+    data: pd.Series,
+    col: Column,
+    validity: tuple[Buffer, tuple[DtypeKind, int, str, str]] | None,
+    allow_modify_inplace: bool = ...,
+) -> pd.Series: ...
+
+
+@overload
+def set_nulls(
+    data: np.ndarray | pd.Series,
+    col: Column,
+    validity: tuple[Buffer, tuple[DtypeKind, int, str, str]] | None,
+    allow_modify_inplace: bool = ...,
+) -> np.ndarray | pd.Series: ...
+
+
+def set_nulls(
+    data: np.ndarray | pd.Series,
+    col: Column,
+    validity: tuple[Buffer, tuple[DtypeKind, int, str, str]] | None,
+    allow_modify_inplace: bool = True,
+) -> np.ndarray | pd.Series:
+    """
+    Set null values for the data according to the column null kind.
+
+    Parameters
+    ----------
+    data : np.ndarray or pd.Series
+        Data to set nulls in.
+    col : Column
+        Column object that describes the `data`.
+    validity : tuple(Buffer, dtype) or None
+        The return value of ``col.buffers()``. We do not access the ``col.buffers()``
+        here to not take the ownership of the memory of buffer objects.
+    allow_modify_inplace : bool, default: True
+        Whether to modify the `data` inplace when zero-copy is possible (True) or always
+        modify a copy of the `data` (False).
+
+    Returns
+    -------
+    np.ndarray or pd.Series
+        Data with the nulls being set.
+    """
+    if validity is None:
+        return data
+    null_kind, sentinel_val = col.describe_null
+    null_pos = None
+
+    if null_kind == ColumnNullType.USE_SENTINEL:
+        null_pos = pd.Series(data) == sentinel_val
+    elif null_kind in (ColumnNullType.USE_BITMASK, ColumnNullType.USE_BYTEMASK):
+        assert validity, "Expected to have a validity buffer for the mask"
+        valid_buff, valid_dtype = validity
+        null_pos = buffer_to_ndarray(
+            valid_buff, valid_dtype, offset=col.offset, length=col.size()
+        )
+        if sentinel_val == 0:
+            null_pos = ~null_pos
+    elif null_kind in (ColumnNullType.NON_NULLABLE, ColumnNullType.USE_NAN):
+        pass
+    else:
+        raise NotImplementedError(f"Null kind {null_kind} is not yet supported.")
+
+    if null_pos is not None and np.any(null_pos):
+        if not allow_modify_inplace:
+            data = data.copy()
+        try:
+            data[null_pos] = None
+        except TypeError:
+            # TypeError happens if the `data` dtype appears to be non-nullable
+            # in numpy notation (bool, int, uint). If this happens,
+            # cast the `data` to nullable float dtype.
+            data = data.astype(float)
+            data[null_pos] = None
+
+    return data
diff --git a/pandas/pandas/core/interchange/utils.py b/pandas/pandas/core/interchange/utils.py
new file mode 100644
index 0000000000000..035a1f8abdbc5
--- /dev/null
+++ b/pandas/pandas/core/interchange/utils.py
@@ -0,0 +1,183 @@
+"""
+Utility functions and objects for implementing the interchange API.
+"""
+
+from __future__ import annotations
+
+import typing
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+    DatetimeTZDtype,
+)
+
+import pandas as pd
+
+if typing.TYPE_CHECKING:
+    from pandas._typing import DtypeObj
+
+
+# Maps str(pyarrow.DataType) = C type format string
+# Currently, no pyarrow API for this
+PYARROW_CTYPES = {
+    "null": "n",
+    "bool": "b",
+    "uint8": "C",
+    "uint16": "S",
+    "uint32": "I",
+    "uint64": "L",
+    "int8": "c",
+    "int16": "S",
+    "int32": "i",
+    "int64": "l",
+    "halffloat": "e",  # float16
+    "float": "f",  # float32
+    "double": "g",  # float64
+    "string": "u",
+    "large_string": "U",
+    "binary": "z",
+    "time32[s]": "tts",
+    "time32[ms]": "ttm",
+    "time64[us]": "ttu",
+    "time64[ns]": "ttn",
+    "date32[day]": "tdD",
+    "date64[ms]": "tdm",
+    "timestamp[s]": "tss:",
+    "timestamp[ms]": "tsm:",
+    "timestamp[us]": "tsu:",
+    "timestamp[ns]": "tsn:",
+    "duration[s]": "tDs",
+    "duration[ms]": "tDm",
+    "duration[us]": "tDu",
+    "duration[ns]": "tDn",
+}
+
+
+class ArrowCTypes:
+    """
+    Enum for Apache Arrow C type format strings.
+
+    The Arrow C data interface:
+    https://arrow.apache.org/docs/format/CDataInterface.html#data-type-description-format-strings
+    """
+
+    NULL = "n"
+    BOOL = "b"
+    INT8 = "c"
+    UINT8 = "C"
+    INT16 = "s"
+    UINT16 = "S"
+    INT32 = "i"
+    UINT32 = "I"
+    INT64 = "l"
+    UINT64 = "L"
+    FLOAT16 = "e"
+    FLOAT32 = "f"
+    FLOAT64 = "g"
+    STRING = "u"  # utf-8
+    LARGE_STRING = "U"  # utf-8
+    DATE32 = "tdD"
+    DATE64 = "tdm"
+    # Resoulution:
+    #   - seconds -> 's'
+    #   - milliseconds -> 'm'
+    #   - microseconds -> 'u'
+    #   - nanoseconds -> 'n'
+    TIMESTAMP = "ts{resolution}:{tz}"
+    TIME = "tt{resolution}"
+
+
+class Endianness:
+    """Enum indicating the byte-order of a data-type."""
+
+    LITTLE = "<"
+    BIG = ">"
+    NATIVE = "="
+    NA = "|"
+
+
+def dtype_to_arrow_c_fmt(dtype: DtypeObj) -> str:
+    """
+    Represent pandas `dtype` as a format string in Apache Arrow C notation.
+
+    Parameters
+    ----------
+    dtype : np.dtype
+        Datatype of pandas DataFrame to represent.
+
+    Returns
+    -------
+    str
+        Format string in Apache Arrow C notation of the given `dtype`.
+    """
+    if isinstance(dtype, CategoricalDtype):
+        return ArrowCTypes.INT64
+    elif dtype == np.dtype("O"):
+        return ArrowCTypes.STRING
+    elif isinstance(dtype, ArrowDtype):
+        import pyarrow as pa
+
+        pa_type = dtype.pyarrow_dtype
+        if pa.types.is_decimal(pa_type):
+            return f"d:{pa_type.precision},{pa_type.scale}"
+        elif pa.types.is_timestamp(pa_type) and pa_type.tz is not None:
+            return f"ts{pa_type.unit[0]}:{pa_type.tz}"
+        format_str = PYARROW_CTYPES.get(str(pa_type), None)
+        if format_str is not None:
+            return format_str
+
+    format_str = getattr(ArrowCTypes, dtype.name.upper(), None)
+    if format_str is not None:
+        return format_str
+
+    if isinstance(dtype, pd.StringDtype):
+        # TODO(infer_string) this should be LARGE_STRING for pyarrow storage,
+        # but current tests don't cover this distinction
+        return ArrowCTypes.STRING
+
+    elif lib.is_np_dtype(dtype, "M"):
+        # Selecting the first char of resolution string:
+        # dtype.str -> '<M8[ns]' -> 'n'
+        resolution = np.datetime_data(dtype)[0][0]
+        return ArrowCTypes.TIMESTAMP.format(resolution=resolution, tz="")
+
+    elif isinstance(dtype, DatetimeTZDtype):
+        return ArrowCTypes.TIMESTAMP.format(resolution=dtype.unit[0], tz=dtype.tz)
+
+    elif isinstance(dtype, pd.BooleanDtype):
+        return ArrowCTypes.BOOL
+
+    raise NotImplementedError(
+        f"Conversion of {dtype} to Arrow C format string is not implemented."
+    )
+
+
+def maybe_rechunk(series: pd.Series, *, allow_copy: bool) -> pd.Series | None:
+    """
+    Rechunk a multi-chunk pyarrow array into a single-chunk array, if necessary.
+
+    - Returns `None` if the input series is not backed by a multi-chunk pyarrow array
+      (and so doesn't need rechunking)
+    - Returns a single-chunk-backed-Series if the input is backed by a multi-chunk
+      pyarrow array and `allow_copy` is `True`.
+    - Raises a `RuntimeError` if `allow_copy` is `False` and input is a
+      based by a multi-chunk pyarrow array.
+    """
+    if not isinstance(series.dtype, pd.ArrowDtype):
+        return None
+    chunked_array = series.array._pa_array  # type: ignore[attr-defined]
+    if len(chunked_array.chunks) == 1:
+        return None
+    if not allow_copy:
+        raise RuntimeError(
+            "Found multi-chunk pyarrow array, but `allow_copy` is False. "
+            "Please rechunk the array before calling this function, or set "
+            "`allow_copy=True`."
+        )
+    arr = chunked_array.combine_chunks()
+    return pd.Series(arr, dtype=series.dtype, name=series.name, index=series.index)
diff --git a/pandas/pandas/core/internals/__init__.py b/pandas/pandas/core/internals/__init__.py
new file mode 100644
index 0000000000000..202bebde88c2c
--- /dev/null
+++ b/pandas/pandas/core/internals/__init__.py
@@ -0,0 +1,57 @@
+from pandas.core.internals.api import make_block  # 2023-09-18 pyarrow uses this
+from pandas.core.internals.concat import concatenate_managers
+from pandas.core.internals.managers import (
+    BlockManager,
+    SingleBlockManager,
+)
+
+__all__ = [
+    "Block",
+    "BlockManager",
+    "ExtensionBlock",
+    "SingleBlockManager",
+    "concatenate_managers",
+    "make_block",
+]
+
+
+def __getattr__(name: str):
+    # GH#55139
+    import warnings
+
+    if name == "create_block_manager_from_blocks":
+        # GH#33892
+        warnings.warn(
+            f"{name} is deprecated and will be removed in a future version. "
+            "Use public APIs instead.",
+            FutureWarning,
+            # https://github.com/pandas-dev/pandas/pull/55139#pullrequestreview-1720690758
+            # on hard-coding stacklevel
+            stacklevel=2,
+        )
+        from pandas.core.internals.managers import create_block_manager_from_blocks
+
+        return create_block_manager_from_blocks
+
+    if name in [
+        "Block",
+        "ExtensionBlock",
+    ]:
+        warnings.warn(
+            f"{name} is deprecated and will be removed in a future version. "
+            "Use public APIs instead.",
+            FutureWarning,
+            # https://github.com/pandas-dev/pandas/pull/55139#pullrequestreview-1720690758
+            # on hard-coding stacklevel
+            stacklevel=2,
+        )
+        if name == "ExtensionBlock":
+            from pandas.core.internals.blocks import ExtensionBlock
+
+            return ExtensionBlock
+        else:
+            from pandas.core.internals.blocks import Block
+
+            return Block
+
+    raise AttributeError(f"module 'pandas.core.internals' has no attribute '{name}'")
diff --git a/pandas/pandas/core/internals/api.py b/pandas/pandas/core/internals/api.py
new file mode 100644
index 0000000000000..04944db2ebd9c
--- /dev/null
+++ b/pandas/pandas/core/internals/api.py
@@ -0,0 +1,145 @@
+"""
+This is a pseudo-public API for downstream libraries.  We ask that downstream
+authors
+
+1) Try to avoid using internals directly altogether, and failing that,
+2) Use only functions exposed here (or in core.internals)
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._libs.internals import BlockPlacement
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+
+from pandas.core.arrays import (
+    DatetimeArray,
+    TimedeltaArray,
+)
+from pandas.core.construction import extract_array
+from pandas.core.internals.blocks import (
+    check_ndim,
+    ensure_block_shape,
+    extract_pandas_array,
+    get_block_type,
+    maybe_coerce_values,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Dtype,
+    )
+
+    from pandas.core.internals.blocks import Block
+
+
+def _make_block(values: ArrayLike, placement: np.ndarray) -> Block:
+    """
+    This is an analogue to blocks.new_block(_2d) that ensures:
+    1) correct dimension for EAs that support 2D (`ensure_block_shape`), and
+    2) correct EA class for datetime64/timedelta64 (`maybe_coerce_values`).
+
+    The input `values` is assumed to be either numpy array or ExtensionArray:
+    - In case of a numpy array, it is assumed to already be in the expected
+      shape for Blocks (2D, (cols, rows)).
+    - In case of an ExtensionArray the input can be 1D, also for EAs that are
+      internally stored as 2D.
+
+    For the rest no preprocessing or validation is done, except for those dtypes
+    that are internally stored as EAs but have an exact numpy equivalent (and at
+    the moment use that numpy dtype), i.e. datetime64/timedelta64.
+    """
+    dtype = values.dtype
+    klass = get_block_type(dtype)
+    placement_obj = BlockPlacement(placement)
+
+    if (isinstance(dtype, ExtensionDtype) and dtype._supports_2d) or isinstance(
+        values, (DatetimeArray, TimedeltaArray)
+    ):
+        values = ensure_block_shape(values, ndim=2)
+
+    values = maybe_coerce_values(values)
+    return klass(values, ndim=2, placement=placement_obj)
+
+
+def make_block(
+    values, placement, klass=None, ndim=None, dtype: Dtype | None = None
+) -> Block:
+    """
+    This is a pseudo-public analogue to blocks.new_block.
+
+    We ask that downstream libraries use this rather than any fully-internal
+    APIs, including but not limited to:
+
+    - core.internals.blocks.make_block
+    - Block.make_block
+    - Block.make_block_same_class
+    - Block.__init__
+    """
+    warnings.warn(
+        # GH#56815
+        "make_block is deprecated and will be removed in a future version. "
+        "Use pd.api.internals.create_dataframe_from_blocks or "
+        "(recommended) higher-level public APIs instead.",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if dtype is not None:
+        dtype = pandas_dtype(dtype)
+
+    values, dtype = extract_pandas_array(values, dtype, ndim)
+
+    from pandas.core.internals.blocks import ExtensionBlock
+
+    if klass is ExtensionBlock and isinstance(values.dtype, PeriodDtype):
+        # GH-44681 changed PeriodArray to be stored in the 2D
+        # NDArrayBackedExtensionBlock instead of ExtensionBlock
+        # -> still allow ExtensionBlock to be passed in this case for back compat
+        klass = None
+
+    if klass is None:
+        dtype = dtype or values.dtype
+        klass = get_block_type(dtype)
+
+    if not isinstance(placement, BlockPlacement):
+        placement = BlockPlacement(placement)
+
+    ndim = maybe_infer_ndim(values, placement, ndim)
+    if isinstance(values.dtype, (PeriodDtype, DatetimeTZDtype)):
+        # GH#41168 ensure we can pass 1D dt64tz values
+        # More generally, any EA dtype that isn't is_1d_only_ea_dtype
+        values = extract_array(values, extract_numpy=True)
+        values = ensure_block_shape(values, ndim)
+
+    check_ndim(values, placement, ndim)
+    values = maybe_coerce_values(values)
+    return klass(values, ndim=ndim, placement=placement)
+
+
+def maybe_infer_ndim(values, placement: BlockPlacement, ndim: int | None) -> int:
+    """
+    If `ndim` is not provided, infer it from placement and values.
+    """
+    if ndim is None:
+        # GH#38134 Block constructor now assumes ndim is not None
+        if not isinstance(values.dtype, np.dtype):
+            if len(placement) != 1:
+                ndim = 1
+            else:
+                ndim = 2
+        else:
+            ndim = values.ndim
+    return ndim
diff --git a/pandas/pandas/core/internals/blocks.py b/pandas/pandas/core/internals/blocks.py
new file mode 100644
index 0000000000000..d1a9081b234de
--- /dev/null
+++ b/pandas/pandas/core/internals/blocks.py
@@ -0,0 +1,2364 @@
+from __future__ import annotations
+
+import inspect
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    final,
+)
+import warnings
+import weakref
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    internals as libinternals,
+    lib,
+)
+from pandas._libs.internals import (
+    BlockPlacement,
+    BlockValuesRefs,
+)
+from pandas._libs.missing import NA
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    DtypeBackend,
+    DtypeObj,
+    FillnaOptions,
+    IgnoreRaise,
+    InterpolateOptions,
+    QuantileInterpolation,
+    Self,
+    Shape,
+    npt,
+)
+from pandas.errors import (
+    AbstractMethodError,
+    OutOfBoundsDatetime,
+)
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import validate_bool_kwarg
+
+from pandas.core.dtypes.astype import (
+    astype_array_safe,
+    astype_is_view,
+)
+from pandas.core.dtypes.cast import (
+    LossySetitemError,
+    can_hold_element,
+    convert_dtypes,
+    find_result_type,
+    np_can_hold_element,
+)
+from pandas.core.dtypes.common import (
+    is_1d_only_ea_dtype,
+    is_float_dtype,
+    is_integer_dtype,
+    is_list_like,
+    is_scalar,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    NumpyEADtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCNumpyExtensionArray,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import is_re
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+    na_value_for_dtype,
+)
+
+from pandas.core import missing
+import pandas.core.algorithms as algos
+from pandas.core.array_algos.putmask import (
+    extract_bool_array,
+    putmask_inplace,
+    putmask_without_repeat,
+    setitem_datetimelike_compat,
+    validate_putmask,
+)
+from pandas.core.array_algos.quantile import quantile_compat
+from pandas.core.array_algos.replace import (
+    compare_or_regex_search,
+    replace_regex,
+    should_use_regex,
+)
+from pandas.core.array_algos.transforms import shift
+from pandas.core.arrays import (
+    DatetimeArray,
+    ExtensionArray,
+    IntervalArray,
+    NumpyExtensionArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.base import PandasObject
+import pandas.core.common as com
+from pandas.core.computation import expressions
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import check_setitem_lengths
+from pandas.core.indexes.base import get_values_for_csv
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Generator,
+        Iterable,
+        Sequence,
+    )
+
+    from pandas.core.api import Index
+    from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+
+# comparison is faster than is_object_dtype
+_dtype_obj = np.dtype("object")
+
+
+class Block(PandasObject, libinternals.Block):
+    """
+    Canonical n-dimensional unit of homogeneous dtype contained in a pandas
+    data structure
+
+    Index-ignorant; let the container take care of that
+    """
+
+    values: np.ndarray | ExtensionArray
+    ndim: int
+    refs: BlockValuesRefs
+    __init__: Callable
+
+    __slots__ = ()
+    is_numeric = False
+
+    @final
+    @cache_readonly
+    def _validate_ndim(self) -> bool:
+        """
+        We validate dimension for blocks that can hold 2D values, which for now
+        means numpy dtypes or DatetimeTZDtype.
+        """
+        dtype = self.dtype
+        return not isinstance(dtype, ExtensionDtype) or isinstance(
+            dtype, DatetimeTZDtype
+        )
+
+    @final
+    @cache_readonly
+    def is_object(self) -> bool:
+        return self.values.dtype == _dtype_obj
+
+    @final
+    @cache_readonly
+    def is_extension(self) -> bool:
+        return not lib.is_np_dtype(self.values.dtype)
+
+    @final
+    @cache_readonly
+    def _can_consolidate(self) -> bool:
+        # We _could_ consolidate for DatetimeTZDtype but don't for now.
+        return not self.is_extension
+
+    @final
+    @cache_readonly
+    def _consolidate_key(self):
+        return self._can_consolidate, self.dtype.name
+
+    @final
+    @cache_readonly
+    def _can_hold_na(self) -> bool:
+        """
+        Can we store NA values in this Block?
+        """
+        dtype = self.dtype
+        if isinstance(dtype, np.dtype):
+            return dtype.kind not in "iub"
+        return dtype._can_hold_na
+
+    @final
+    @property
+    def is_bool(self) -> bool:
+        """
+        We can be bool if a) we are bool dtype or b) object dtype with bool objects.
+        """
+        return self.values.dtype == np.dtype(bool)
+
+    @final
+    def external_values(self):
+        return external_values(self.values)
+
+    @final
+    @cache_readonly
+    def fill_value(self):
+        # Used in reindex_indexer
+        return na_value_for_dtype(self.dtype, compat=False)
+
+    @final
+    def _standardize_fill_value(self, value):
+        # if we are passed a scalar None, convert it here
+        if self.dtype != _dtype_obj and is_valid_na_for_dtype(value, self.dtype):
+            value = self.fill_value
+        return value
+
+    @property
+    def mgr_locs(self) -> BlockPlacement:
+        return self._mgr_locs
+
+    @mgr_locs.setter
+    def mgr_locs(self, new_mgr_locs: BlockPlacement) -> None:
+        self._mgr_locs = new_mgr_locs
+
+    @final
+    def make_block(
+        self,
+        values,
+        placement: BlockPlacement | None = None,
+        refs: BlockValuesRefs | None = None,
+    ) -> Block:
+        """
+        Create a new block, with type inference propagate any values that are
+        not specified
+        """
+        if placement is None:
+            placement = self._mgr_locs
+        if self.is_extension:
+            values = ensure_block_shape(values, ndim=self.ndim)
+
+        return new_block(values, placement=placement, ndim=self.ndim, refs=refs)
+
+    @final
+    def make_block_same_class(
+        self,
+        values,
+        placement: BlockPlacement | None = None,
+        refs: BlockValuesRefs | None = None,
+    ) -> Self:
+        """Wrap given values in a block of same type as self."""
+        # Pre-2.0 we called ensure_wrapped_if_datetimelike because fastparquet
+        #  relied on it, as of 2.0 the caller is responsible for this.
+        if placement is None:
+            placement = self._mgr_locs
+
+        # We assume maybe_coerce_values has already been called
+        return type(self)(values, placement=placement, ndim=self.ndim, refs=refs)
+
+    @final
+    def __repr__(self) -> str:
+        # don't want to print out all of the items here
+        name = type(self).__name__
+        if self.ndim == 1:
+            result = f"{name}: {len(self)} dtype: {self.dtype}"
+        else:
+            shape = " x ".join([str(s) for s in self.shape])
+            result = f"{name}: {self.mgr_locs.indexer}, {shape}, dtype: {self.dtype}"
+
+        return result
+
+    @final
+    def __len__(self) -> int:
+        return len(self.values)
+
+    @final
+    def slice_block_columns(self, slc: slice) -> Self:
+        """
+        Perform __getitem__-like, return result as block.
+        """
+        new_mgr_locs = self._mgr_locs[slc]
+
+        new_values = self._slice(slc)
+        refs = self.refs
+        return type(self)(new_values, new_mgr_locs, self.ndim, refs=refs)
+
+    @final
+    def take_block_columns(self, indices: npt.NDArray[np.intp]) -> Self:
+        """
+        Perform __getitem__-like, return result as block.
+
+        Only supports slices that preserve dimensionality.
+        """
+        # Note: only called from is from internals.concat, and we can verify
+        #  that never happens with 1-column blocks, i.e. never for ExtensionBlock.
+
+        new_mgr_locs = self._mgr_locs[indices]
+
+        new_values = self._slice(indices)
+        return type(self)(new_values, new_mgr_locs, self.ndim, refs=None)
+
+    @final
+    def getitem_block_columns(
+        self, slicer: slice, new_mgr_locs: BlockPlacement, ref_inplace_op: bool = False
+    ) -> Self:
+        """
+        Perform __getitem__-like, return result as block.
+
+        Only supports slices that preserve dimensionality.
+        """
+        new_values = self._slice(slicer)
+        refs = self.refs if not ref_inplace_op or self.refs.has_reference() else None
+        return type(self)(new_values, new_mgr_locs, self.ndim, refs=refs)
+
+    @final
+    def _can_hold_element(self, element: Any) -> bool:
+        """require the same dtype as ourselves"""
+        element = extract_array(element, extract_numpy=True)
+        return can_hold_element(self.values, element)
+
+    @final
+    def should_store(self, value: ArrayLike) -> bool:
+        """
+        Should we set self.values[indexer] = value inplace or do we need to cast?
+
+        Parameters
+        ----------
+        value : np.ndarray or ExtensionArray
+
+        Returns
+        -------
+        bool
+        """
+        return value.dtype == self.dtype
+
+    # ---------------------------------------------------------------------
+    # Apply/Reduce and Helpers
+
+    @final
+    def apply(self, func, **kwargs) -> list[Block]:
+        """
+        apply the function to my values; return a block if we are not
+        one
+        """
+        result = func(self.values, **kwargs)
+
+        result = maybe_coerce_values(result)
+        return self._split_op_result(result)
+
+    @final
+    def reduce(self, func) -> list[Block]:
+        # We will apply the function and reshape the result into a single-row
+        #  Block with the same mgr_locs; squeezing will be done at a higher level
+        assert self.ndim == 2
+
+        result = func(self.values)
+
+        if self.values.ndim == 1:
+            res_values = result
+        else:
+            res_values = result.reshape(-1, 1)
+
+        nb = self.make_block(res_values)
+        return [nb]
+
+    @final
+    def _split_op_result(self, result: ArrayLike) -> list[Block]:
+        # See also: split_and_operate
+        if result.ndim > 1 and isinstance(result.dtype, ExtensionDtype):
+            # TODO(EA2D): unnecessary with 2D EAs
+            # if we get a 2D ExtensionArray, we need to split it into 1D pieces
+            nbs = []
+            for i, loc in enumerate(self._mgr_locs):
+                if not is_1d_only_ea_dtype(result.dtype):
+                    vals = result[i : i + 1]
+                else:
+                    vals = result[i]
+
+                bp = BlockPlacement(loc)
+                block = self.make_block(values=vals, placement=bp)
+                nbs.append(block)
+            return nbs
+
+        nb = self.make_block(result)
+
+        return [nb]
+
+    @final
+    def _split(self) -> Generator[Block]:
+        """
+        Split a block into a list of single-column blocks.
+        """
+        assert self.ndim == 2
+
+        for i, ref_loc in enumerate(self._mgr_locs):
+            vals = self.values[slice(i, i + 1)]
+
+            bp = BlockPlacement(ref_loc)
+            nb = type(self)(vals, placement=bp, ndim=2, refs=self.refs)
+            yield nb
+
+    @final
+    def split_and_operate(self, func, *args, **kwargs) -> list[Block]:
+        """
+        Split the block and apply func column-by-column.
+
+        Parameters
+        ----------
+        func : Block method
+        *args
+        **kwargs
+
+        Returns
+        -------
+        List[Block]
+        """
+        assert self.ndim == 2 and self.shape[0] != 1
+
+        res_blocks = []
+        for nb in self._split():
+            rbs = func(nb, *args, **kwargs)
+            res_blocks.extend(rbs)
+        return res_blocks
+
+    # ---------------------------------------------------------------------
+    # Up/Down-casting
+
+    @final
+    def coerce_to_target_dtype(self, other, raise_on_upcast: bool) -> Block:
+        """
+        coerce the current block to a dtype compat for other
+        we will return a block, possibly object, and not raise
+
+        we can also safely try to coerce to the same dtype
+        and will receive the same block
+        """
+        new_dtype = find_result_type(self.values.dtype, other)
+        if new_dtype == self.dtype:
+            # GH#52927 avoid RecursionError
+            raise AssertionError(
+                "Something has gone wrong, please report a bug at "
+                "https://github.com/pandas-dev/pandas/issues"
+            )
+
+        # In a future version of pandas, the default will be that
+        # setting `nan` into an integer series won't raise.
+        if (
+            is_scalar(other)
+            and is_integer_dtype(self.values.dtype)
+            and isna(other)
+            and other is not NaT
+            and not (
+                isinstance(other, (np.datetime64, np.timedelta64)) and np.isnat(other)
+            )
+        ):
+            raise_on_upcast = False
+        elif (
+            isinstance(other, np.ndarray)
+            and other.ndim == 1
+            and is_integer_dtype(self.values.dtype)
+            and is_float_dtype(other.dtype)
+            and lib.has_only_ints_or_nan(other)
+        ):
+            raise_on_upcast = False
+
+        if raise_on_upcast:
+            raise TypeError(f"Invalid value '{other}' for dtype '{self.values.dtype}'")
+        if self.values.dtype == new_dtype:
+            raise AssertionError(
+                f"Did not expect new dtype {new_dtype} to equal self.dtype "
+                f"{self.values.dtype}. Please report a bug at "
+                "https://github.com/pandas-dev/pandas/issues."
+            )
+        try:
+            return self.astype(new_dtype)
+        except OutOfBoundsDatetime as err:
+            # e.g. GH#56419 if self.dtype is a low-resolution dt64 and we try to
+            #  upcast to a higher-resolution dt64, we may have entries that are
+            #  out of bounds for the higher resolution.
+            #  Re-raise with a more informative message.
+            raise OutOfBoundsDatetime(
+                f"Incompatible (high-resolution) value for dtype='{self.dtype}'. "
+                "Explicitly cast before operating."
+            ) from err
+
+    @final
+    def convert(self) -> list[Block]:
+        """
+        Attempt to coerce any object types to better types. Return a copy
+        of the block (if copy = True).
+        """
+        if not self.is_object:
+            return [self.copy(deep=False)]
+
+        if self.ndim != 1 and self.shape[0] != 1:
+            blocks = self.split_and_operate(Block.convert)
+            if all(blk.dtype.kind == "O" for blk in blocks):
+                # Avoid fragmenting the block if convert is a no-op
+                return [self.copy(deep=False)]
+            return blocks
+
+        values = self.values
+        if values.ndim == 2:
+            # the check above ensures we only get here with values.shape[0] == 1,
+            # avoid doing .ravel as that might make a copy
+            values = values[0]
+
+        res_values = lib.maybe_convert_objects(
+            values,  # type: ignore[arg-type]
+            convert_non_numeric=True,
+        )
+        refs = None
+        if res_values is values or (
+            isinstance(res_values, NumpyExtensionArray)
+            and res_values._ndarray is values
+        ):
+            refs = self.refs
+
+        res_values = ensure_block_shape(res_values, self.ndim)
+        res_values = maybe_coerce_values(res_values)
+        return [self.make_block(res_values, refs=refs)]
+
+    def convert_dtypes(
+        self,
+        infer_objects: bool = True,
+        convert_string: bool = True,
+        convert_integer: bool = True,
+        convert_boolean: bool = True,
+        convert_floating: bool = True,
+        dtype_backend: DtypeBackend = "numpy_nullable",
+    ) -> list[Block]:
+        if infer_objects and self.is_object:
+            blks = self.convert()
+        else:
+            blks = [self]
+
+        if not any(
+            [convert_floating, convert_integer, convert_boolean, convert_string]
+        ):
+            return [b.copy(deep=False) for b in blks]
+
+        rbs = []
+        for blk in blks:
+            # Determine dtype column by column
+            sub_blks = (
+                [blk] if blk.ndim == 1 or self.shape[0] == 1 else list(blk._split())
+            )
+            dtypes = [
+                convert_dtypes(
+                    b.values,
+                    convert_string,
+                    convert_integer,
+                    convert_boolean,
+                    convert_floating,
+                    infer_objects,
+                    dtype_backend,
+                )
+                for b in sub_blks
+            ]
+            if all(dtype == self.dtype for dtype in dtypes):
+                # Avoid block splitting if no dtype changes
+                rbs.append(blk.copy(deep=False))
+                continue
+
+            for dtype, b in zip(dtypes, sub_blks):
+                rbs.append(b.astype(dtype=dtype, squeeze=b.ndim != 1))
+        return rbs
+
+    # ---------------------------------------------------------------------
+    # Array-Like Methods
+
+    @final
+    @cache_readonly
+    def dtype(self) -> DtypeObj:
+        return self.values.dtype
+
+    @final
+    def astype(
+        self,
+        dtype: DtypeObj,
+        errors: IgnoreRaise = "raise",
+        squeeze: bool = False,
+    ) -> Block:
+        """
+        Coerce to the new dtype.
+
+        Parameters
+        ----------
+        dtype : np.dtype or ExtensionDtype
+        errors : str, {'raise', 'ignore'}, default 'raise'
+            - ``raise`` : allow exceptions to be raised
+            - ``ignore`` : suppress exceptions. On error return original object
+        squeeze : bool, default False
+            squeeze values to ndim=1 if only one column is given
+
+        Returns
+        -------
+        Block
+        """
+        values = self.values
+        if squeeze and values.ndim == 2 and is_1d_only_ea_dtype(dtype):
+            if values.shape[0] != 1:
+                raise ValueError("Can not squeeze with more than one column.")
+            values = values[0, :]  # type: ignore[call-overload]
+
+        new_values = astype_array_safe(values, dtype, errors=errors)
+
+        new_values = maybe_coerce_values(new_values)
+
+        refs = None
+        if astype_is_view(values.dtype, new_values.dtype):
+            refs = self.refs
+
+        newb = self.make_block(new_values, refs=refs)
+        if newb.shape != self.shape:
+            raise TypeError(
+                f"cannot set astype for dtype "
+                f"({self.dtype.name} [{self.shape}]) to different shape "
+                f"({newb.dtype.name} [{newb.shape}])"
+            )
+        return newb
+
+    @final
+    def get_values_for_csv(
+        self, *, float_format, date_format, decimal, na_rep: str = "nan", quoting=None
+    ) -> Block:
+        """convert to our native types format"""
+        result = get_values_for_csv(
+            self.values,
+            na_rep=na_rep,
+            quoting=quoting,
+            float_format=float_format,
+            date_format=date_format,
+            decimal=decimal,
+        )
+        return self.make_block(result)
+
+    @final
+    def copy(self, deep: bool = True) -> Self:
+        """copy constructor"""
+        values = self.values
+        refs: BlockValuesRefs | None
+        if deep:
+            values = values.copy()
+            refs = None
+        else:
+            refs = self.refs
+        return type(self)(values, placement=self._mgr_locs, ndim=self.ndim, refs=refs)
+
+    # ---------------------------------------------------------------------
+    # Copy-on-Write Helpers
+
+    def _maybe_copy(self, inplace: bool) -> Self:
+        if inplace:
+            deep = self.refs.has_reference()
+            return self.copy(deep=deep)
+        return self.copy()
+
+    @final
+    def _get_refs_and_copy(self, inplace: bool):
+        refs = None
+        copy = not inplace
+        if inplace:
+            if self.refs.has_reference():
+                copy = True
+            else:
+                refs = self.refs
+        return copy, refs
+
+    # ---------------------------------------------------------------------
+    # Replace
+
+    @final
+    def replace(
+        self,
+        to_replace,
+        value,
+        inplace: bool = False,
+        # mask may be pre-computed if we're called from replace_list
+        mask: npt.NDArray[np.bool_] | None = None,
+    ) -> list[Block]:
+        """
+        replace the to_replace value with value, possible to create new
+        blocks here this is just a call to putmask.
+        """
+
+        # Note: the checks we do in NDFrame.replace ensure we never get
+        #  here with listlike to_replace or value, as those cases
+        #  go through replace_list
+        values = self.values
+
+        if not self._can_hold_element(to_replace):
+            # We cannot hold `to_replace`, so we know immediately that
+            #  replacing it is a no-op.
+            # Note: If to_replace were a list, NDFrame.replace would call
+            #  replace_list instead of replace.
+            return [self.copy(deep=False)]
+
+        if mask is None:
+            mask = missing.mask_missing(values, to_replace)
+        if not mask.any():
+            # Note: we get here with test_replace_extension_other incorrectly
+            #  bc _can_hold_element is incorrect.
+            return [self.copy(deep=False)]
+
+        elif self._can_hold_element(value) or (self.dtype == "string" and is_re(value)):
+            # TODO(CoW): Maybe split here as well into columns where mask has True
+            # and rest?
+            blk = self._maybe_copy(inplace)
+            putmask_inplace(blk.values, mask, value)
+            return [blk]
+
+        elif self.ndim == 1 or self.shape[0] == 1:
+            if value is None or value is NA:
+                blk = self.astype(np.dtype(object))
+            else:
+                blk = self.coerce_to_target_dtype(value, raise_on_upcast=False)
+            return blk.replace(
+                to_replace=to_replace,
+                value=value,
+                inplace=True,
+                mask=mask,
+            )
+
+        else:
+            # split so that we only upcast where necessary
+            blocks = []
+            for i, nb in enumerate(self._split()):
+                blocks.extend(
+                    type(self).replace(
+                        nb,
+                        to_replace=to_replace,
+                        value=value,
+                        inplace=True,
+                        mask=mask[i : i + 1],
+                    )
+                )
+            return blocks
+
+    @final
+    def _replace_regex(
+        self,
+        to_replace,
+        value,
+        inplace: bool = False,
+        mask=None,
+    ) -> list[Block]:
+        """
+        Replace elements by the given value.
+
+        Parameters
+        ----------
+        to_replace : object or pattern
+            Scalar to replace or regular expression to match.
+        value : object
+            Replacement object.
+        inplace : bool, default False
+            Perform inplace modification.
+        mask : array-like of bool, optional
+            True indicate corresponding element is ignored.
+
+        Returns
+        -------
+        List[Block]
+        """
+        if not is_re(to_replace) and not self._can_hold_element(to_replace):
+            # i.e. only if self.is_object is True, but could in principle include a
+            #  String ExtensionBlock
+            return [self.copy(deep=False)]
+
+        if is_re(to_replace) and self.dtype not in [object, "string"]:
+            # only object or string dtype can hold strings, and a regex object
+            # will only match strings
+            return [self.copy(deep=False)]
+
+        if not (
+            self._can_hold_element(value) or (self.dtype == "string" and is_re(value))
+        ):
+            block = self.astype(np.dtype(object))
+        else:
+            block = self._maybe_copy(inplace)
+
+        rx = re.compile(to_replace)
+
+        replace_regex(block.values, rx, value, mask)
+        return [block]
+
+    @final
+    def replace_list(
+        self,
+        src_list: Iterable[Any],
+        dest_list: Sequence[Any],
+        inplace: bool = False,
+        regex: bool = False,
+    ) -> list[Block]:
+        """
+        See BlockManager.replace_list docstring.
+        """
+        values = self.values
+
+        # Exclude anything that we know we won't contain
+        pairs = [
+            (x, y)
+            for x, y in zip(src_list, dest_list)
+            if (self._can_hold_element(x) or (self.dtype == "string" and is_re(x)))
+        ]
+        if not len(pairs):
+            return [self.copy(deep=False)]
+
+        src_len = len(pairs) - 1
+
+        if is_string_dtype(values.dtype):
+            # Calculate the mask once, prior to the call of comp
+            # in order to avoid repeating the same computations
+            na_mask = ~isna(values)
+            masks: Iterable[npt.NDArray[np.bool_]] = (
+                extract_bool_array(
+                    cast(
+                        ArrayLike,
+                        compare_or_regex_search(
+                            values, s[0], regex=regex, mask=na_mask
+                        ),
+                    )
+                )
+                for s in pairs
+            )
+        else:
+            # GH#38086 faster if we know we dont need to check for regex
+            masks = (missing.mask_missing(values, s[0]) for s in pairs)
+        # Materialize if inplace = True, since the masks can change
+        # as we replace
+        if inplace:
+            masks = list(masks)
+
+        # Don't set up refs here, otherwise we will think that we have
+        # references when we check again later
+        rb = [self]
+
+        for i, ((src, dest), mask) in enumerate(zip(pairs, masks)):
+            new_rb: list[Block] = []
+
+            # GH-39338: _replace_coerce can split a block into
+            # single-column blocks, so track the index so we know
+            # where to index into the mask
+            for blk_num, blk in enumerate(rb):
+                if len(rb) == 1:
+                    m = mask
+                else:
+                    mib = mask
+                    assert not isinstance(mib, bool)
+                    m = mib[blk_num : blk_num + 1]
+
+                # error: Argument "mask" to "_replace_coerce" of "Block" has
+                # incompatible type "Union[ExtensionArray, ndarray[Any, Any], bool]";
+                # expected "ndarray[Any, dtype[bool_]]"
+                result = blk._replace_coerce(
+                    to_replace=src,
+                    value=dest,
+                    mask=m,
+                    inplace=inplace,
+                    regex=regex,
+                )
+
+                if i != src_len:
+                    # This is ugly, but we have to get rid of intermediate refs
+                    # that did not go out of scope yet, otherwise we will trigger
+                    # many unnecessary copies
+                    for b in result:
+                        ref = weakref.ref(b)
+                        b.refs.referenced_blocks.pop(
+                            b.refs.referenced_blocks.index(ref)
+                        )
+                new_rb.extend(result)
+            rb = new_rb
+        return rb
+
+    @final
+    def _replace_coerce(
+        self,
+        to_replace,
+        value,
+        mask: npt.NDArray[np.bool_],
+        inplace: bool = True,
+        regex: bool = False,
+    ) -> list[Block]:
+        """
+        Replace value corresponding to the given boolean array with another
+        value.
+
+        Parameters
+        ----------
+        to_replace : object or pattern
+            Scalar to replace or regular expression to match.
+        value : object
+            Replacement object.
+        mask : np.ndarray[bool]
+            True indicate corresponding element is ignored.
+        inplace : bool, default True
+            Perform inplace modification.
+        regex : bool, default False
+            If true, perform regular expression substitution.
+
+        Returns
+        -------
+        List[Block]
+        """
+        if should_use_regex(regex, to_replace):
+            return self._replace_regex(
+                to_replace,
+                value,
+                inplace=inplace,
+                mask=mask,
+            )
+        else:
+            if value is None:
+                # gh-45601, gh-45836, gh-46634
+                if mask.any():
+                    has_ref = self.refs.has_reference()
+                    nb = self.astype(np.dtype(object))
+                    if not inplace:
+                        nb = nb.copy()
+                    elif inplace and has_ref and nb.refs.has_reference():
+                        # no copy in astype and we had refs before
+                        nb = nb.copy()
+                    putmask_inplace(nb.values, mask, value)
+                    return [nb]
+                return [self.copy(deep=False)]
+            return self.replace(
+                to_replace=to_replace,
+                value=value,
+                inplace=inplace,
+                mask=mask,
+            )
+
+    # ---------------------------------------------------------------------
+    # 2D Methods - Shared by NumpyBlock and NDArrayBackedExtensionBlock
+    #  but not ExtensionBlock
+
+    def _maybe_squeeze_arg(self, arg: np.ndarray) -> np.ndarray:
+        """
+        For compatibility with 1D-only ExtensionArrays.
+        """
+        return arg
+
+    def _unwrap_setitem_indexer(self, indexer):
+        """
+        For compatibility with 1D-only ExtensionArrays.
+        """
+        return indexer
+
+    # NB: this cannot be made cache_readonly because in mgr.set_values we pin
+    #  new .values that can have different shape GH#42631
+    @property
+    def shape(self) -> Shape:
+        return self.values.shape
+
+    def iget(self, i: int | tuple[int, int] | tuple[slice, int]) -> np.ndarray:
+        # In the case where we have a tuple[slice, int], the slice will always
+        #  be slice(None)
+        # Note: only reached with self.ndim == 2
+        # Invalid index type "Union[int, Tuple[int, int], Tuple[slice, int]]"
+        # for "Union[ndarray[Any, Any], ExtensionArray]"; expected type
+        # "Union[int, integer[Any]]"
+        return self.values[i]  # type: ignore[index]
+
+    def _slice(
+        self, slicer: slice | npt.NDArray[np.bool_] | npt.NDArray[np.intp]
+    ) -> ArrayLike:
+        """return a slice of my values"""
+
+        return self.values[slicer]
+
+    def set_inplace(self, locs, values: ArrayLike, copy: bool = False) -> None:
+        """
+        Modify block values in-place with new item value.
+
+        If copy=True, first copy the underlying values in place before modifying
+        (for Copy-on-Write).
+
+        Notes
+        -----
+        `set_inplace` never creates a new array or new Block, whereas `setitem`
+        _may_ create a new array and always creates a new Block.
+
+        Caller is responsible for checking values.dtype == self.dtype.
+        """
+        if copy:
+            self.values = self.values.copy()
+        self.values[locs] = values
+
+    @final
+    def take_nd(
+        self,
+        indexer: npt.NDArray[np.intp],
+        axis: AxisInt,
+        new_mgr_locs: BlockPlacement | None = None,
+        fill_value=lib.no_default,
+    ) -> Block:
+        """
+        Take values according to indexer and return them as a block.
+        """
+        values = self.values
+
+        if fill_value is lib.no_default:
+            fill_value = self.fill_value
+            allow_fill = False
+        else:
+            allow_fill = True
+
+        # Note: algos.take_nd has upcast logic similar to coerce_to_target_dtype
+        new_values = algos.take_nd(
+            values, indexer, axis=axis, allow_fill=allow_fill, fill_value=fill_value
+        )
+
+        # Called from three places in managers, all of which satisfy
+        #  these assertions
+        if isinstance(self, ExtensionBlock):
+            # NB: in this case, the 'axis' kwarg will be ignored in the
+            #  algos.take_nd call above.
+            assert not (self.ndim == 1 and new_mgr_locs is None)
+        assert not (axis == 0 and new_mgr_locs is None)
+
+        if new_mgr_locs is None:
+            new_mgr_locs = self._mgr_locs
+
+        if new_values.dtype != self.dtype:
+            return self.make_block(new_values, new_mgr_locs)
+        else:
+            return self.make_block_same_class(new_values, new_mgr_locs)
+
+    def _unstack(
+        self,
+        unstacker,
+        fill_value,
+        new_placement: npt.NDArray[np.intp],
+        needs_masking: npt.NDArray[np.bool_],
+    ):
+        """
+        Return a list of unstacked blocks of self
+
+        Parameters
+        ----------
+        unstacker : reshape._Unstacker
+        fill_value : int
+            Only used in ExtensionBlock._unstack
+        new_placement : np.ndarray[np.intp]
+        allow_fill : bool
+        needs_masking : np.ndarray[bool]
+
+        Returns
+        -------
+        blocks : list of Block
+            New blocks of unstacked values.
+        mask : array-like of bool
+            The mask of columns of `blocks` we should keep.
+        """
+        new_values, mask = unstacker.get_new_values(
+            self.values.T, fill_value=fill_value
+        )
+
+        mask = mask.any(0)
+        # TODO: in all tests we have mask.all(); can we rely on that?
+
+        # Note: these next two lines ensure that
+        #  mask.sum() == sum(len(nb.mgr_locs) for nb in blocks)
+        #  which the calling function needs in order to pass verify_integrity=False
+        #  to the BlockManager constructor
+        new_values = new_values.T[mask]
+        new_placement = new_placement[mask]
+
+        bp = BlockPlacement(new_placement)
+        blocks = [new_block_2d(new_values, placement=bp)]
+        return blocks, mask
+
+    # ---------------------------------------------------------------------
+
+    def setitem(self, indexer, value) -> Block:
+        """
+        Attempt self.values[indexer] = value, possibly creating a new array.
+
+        Parameters
+        ----------
+        indexer : tuple, list-like, array-like, slice, int
+            The subset of self.values to set
+        value : object
+            The value being set
+
+        Returns
+        -------
+        Block
+
+        Notes
+        -----
+        `indexer` is a direct slice/positional indexer. `value` must
+        be a compatible shape.
+        """
+
+        value = self._standardize_fill_value(value)
+
+        values = cast(np.ndarray, self.values)
+        if self.ndim == 2:
+            values = values.T
+
+        # length checking
+        check_setitem_lengths(indexer, value, values)
+
+        if self.dtype != _dtype_obj:
+            # GH48933: extract_array would convert a pd.Series value to np.ndarray
+            value = extract_array(value, extract_numpy=True)
+        try:
+            casted = np_can_hold_element(values.dtype, value)
+        except LossySetitemError:
+            # current dtype cannot store value, coerce to common dtype
+            nb = self.coerce_to_target_dtype(value, raise_on_upcast=True)
+            return nb.setitem(indexer, value)
+        else:
+            if self.dtype == _dtype_obj:
+                # TODO: avoid having to construct values[indexer]
+                vi = values[indexer]
+                if lib.is_list_like(vi):
+                    # checking lib.is_scalar here fails on
+                    #  test_iloc_setitem_custom_object
+                    casted = setitem_datetimelike_compat(values, len(vi), casted)
+
+            self = self._maybe_copy(inplace=True)
+            values = cast(np.ndarray, self.values.T)
+            if isinstance(casted, np.ndarray) and casted.ndim == 1 and len(casted) == 1:
+                # NumPy 1.25 deprecation: https://github.com/numpy/numpy/pull/10615
+                casted = casted[0, ...]
+            try:
+                values[indexer] = casted
+            except (TypeError, ValueError) as err:
+                if is_list_like(casted):
+                    raise ValueError(
+                        "setting an array element with a sequence."
+                    ) from err
+                raise
+        return self
+
+    def putmask(self, mask, new) -> list[Block]:
+        """
+        putmask the data to the block; it is possible that we may create a
+        new dtype of block
+
+        Return the resulting block(s).
+
+        Parameters
+        ----------
+        mask : np.ndarray[bool], SparseArray[bool], or BooleanArray
+        new : a ndarray/object
+
+        Returns
+        -------
+        List[Block]
+        """
+        orig_mask = mask
+        values = cast(np.ndarray, self.values)
+        mask, noop = validate_putmask(values.T, mask)
+        assert not isinstance(new, (ABCIndex, ABCSeries, ABCDataFrame))
+
+        if new is lib.no_default:
+            new = self.fill_value
+
+        new = self._standardize_fill_value(new)
+        new = extract_array(new, extract_numpy=True)
+
+        if noop:
+            return [self.copy(deep=False)]
+
+        try:
+            casted = np_can_hold_element(values.dtype, new)
+
+            self = self._maybe_copy(inplace=True)
+            values = cast(np.ndarray, self.values)
+
+            putmask_without_repeat(values.T, mask, casted)
+            return [self]
+        except LossySetitemError:
+            if self.ndim == 1 or self.shape[0] == 1:
+                # no need to split columns
+
+                if not is_list_like(new):
+                    # using just new[indexer] can't save us the need to cast
+                    return self.coerce_to_target_dtype(
+                        new, raise_on_upcast=True
+                    ).putmask(mask, new)
+                else:
+                    indexer = mask.nonzero()[0]
+                    nb = self.setitem(indexer, new[indexer])
+                    return [nb]
+
+            else:
+                is_array = isinstance(new, np.ndarray)
+
+                res_blocks = []
+                for i, nb in enumerate(self._split()):
+                    n = new
+                    if is_array:
+                        # we have a different value per-column
+                        n = new[:, i : i + 1]
+
+                    submask = orig_mask[:, i : i + 1]
+                    rbs = nb.putmask(submask, n)
+                    res_blocks.extend(rbs)
+                return res_blocks
+
+    def where(self, other, cond) -> list[Block]:
+        """
+        evaluate the block; return result block(s) from the result
+
+        Parameters
+        ----------
+        other : a ndarray/object
+        cond : np.ndarray[bool], SparseArray[bool], or BooleanArray
+
+        Returns
+        -------
+        List[Block]
+        """
+        assert cond.ndim == self.ndim
+        assert not isinstance(other, (ABCIndex, ABCSeries, ABCDataFrame))
+
+        transpose = self.ndim == 2
+
+        cond = extract_bool_array(cond)
+
+        # EABlocks override where
+        values = cast(np.ndarray, self.values)
+        orig_other = other
+        if transpose:
+            values = values.T
+
+        icond, noop = validate_putmask(values, ~cond)
+        if noop:
+            return [self.copy(deep=False)]
+
+        if other is lib.no_default:
+            other = self.fill_value
+
+        other = self._standardize_fill_value(other)
+
+        try:
+            # try/except here is equivalent to a self._can_hold_element check,
+            #  but this gets us back 'casted' which we will reuse below;
+            #  without using 'casted', expressions.where may do unwanted upcasts.
+            casted = np_can_hold_element(values.dtype, other)
+        except (ValueError, TypeError, LossySetitemError):
+            # we cannot coerce, return a compat dtype
+
+            if self.ndim == 1 or self.shape[0] == 1:
+                # no need to split columns
+
+                block = self.coerce_to_target_dtype(other, raise_on_upcast=False)
+                return block.where(orig_other, cond)
+
+            else:
+                is_array = isinstance(other, (np.ndarray, ExtensionArray))
+
+                res_blocks = []
+                for i, nb in enumerate(self._split()):
+                    oth = other
+                    if is_array:
+                        # we have a different value per-column
+                        oth = other[:, i : i + 1]
+
+                    submask = cond[:, i : i + 1]
+                    rbs = nb.where(oth, submask)
+                    res_blocks.extend(rbs)
+                return res_blocks
+
+        else:
+            other = casted
+            alt = setitem_datetimelike_compat(values, icond.sum(), other)
+            if alt is not other:
+                if is_list_like(other) and len(other) < len(values):
+                    # call np.where with other to get the appropriate ValueError
+                    np.where(~icond, values, other)
+                    raise NotImplementedError(
+                        "This should not be reached; call to np.where above is "
+                        "expected to raise ValueError. Please report a bug at "
+                        "github.com/pandas-dev/pandas"
+                    )
+                result = values.copy()
+                np.putmask(result, icond, alt)
+            else:
+                # By the time we get here, we should have all Series/Index
+                #  args extracted to ndarray
+                if (
+                    is_list_like(other)
+                    and not isinstance(other, np.ndarray)
+                    and len(other) == self.shape[-1]
+                ):
+                    # If we don't do this broadcasting here, then expressions.where
+                    #  will broadcast a 1D other to be row-like instead of
+                    #  column-like.
+                    other = np.array(other).reshape(values.shape)
+                    # If lengths don't match (or len(other)==1), we will raise
+                    #  inside expressions.where, see test_series_where
+
+                # Note: expressions.where may upcast.
+                result = expressions.where(~icond, values, other)
+                # The np_can_hold_element check _should_ ensure that we always
+                #  have result.dtype == self.dtype here.
+
+        if transpose:
+            result = result.T
+
+        return [self.make_block(result)]
+
+    def fillna(
+        self,
+        value,
+        limit: int | None = None,
+        inplace: bool = False,
+    ) -> list[Block]:
+        """
+        fillna on the block with the value. If we fail, then convert to
+        block to hold objects instead and try again
+        """
+        # Caller is responsible for validating limit; if int it is strictly positive
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        if not self._can_hold_na:
+            # can short-circuit the isna call
+            noop = True
+        else:
+            mask = isna(self.values)
+            mask, noop = validate_putmask(self.values, mask)
+
+        if noop:
+            # we can't process the value, but nothing to do
+            return [self.copy(deep=False)]
+
+        if limit is not None:
+            mask[mask.cumsum(self.values.ndim - 1) > limit] = False
+
+        if inplace:
+            nbs = self.putmask(mask.T, value)
+        else:
+            nbs = self.where(value, ~mask.T)
+        return extend_blocks(nbs)
+
+    def pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        inplace: bool = False,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+    ) -> list[Block]:
+        if not self._can_hold_na:
+            # If there are no NAs, then interpolate is a no-op
+            return [self.copy(deep=False)]
+
+        copy, refs = self._get_refs_and_copy(inplace)
+
+        # Dispatch to the NumpyExtensionArray method.
+        # We know self.array_values is a NumpyExtensionArray bc EABlock overrides
+        vals = cast(NumpyExtensionArray, self.array_values)
+        new_values = vals.T._pad_or_backfill(
+            method=method,
+            limit=limit,
+            limit_area=limit_area,
+            copy=copy,
+        ).T
+
+        data = extract_array(new_values, extract_numpy=True)
+        return [self.make_block_same_class(data, refs=refs)]
+
+    @final
+    def interpolate(
+        self,
+        *,
+        method: InterpolateOptions,
+        index: Index,
+        inplace: bool = False,
+        limit: int | None = None,
+        limit_direction: Literal["forward", "backward", "both"] = "forward",
+        limit_area: Literal["inside", "outside"] | None = None,
+        **kwargs,
+    ) -> list[Block]:
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        # error: Non-overlapping equality check [...]
+        if method == "asfreq":  # type: ignore[comparison-overlap]
+            # clean_fill_method used to allow this
+            missing.clean_fill_method(method)
+
+        if not self._can_hold_na:
+            # If there are no NAs, then interpolate is a no-op
+            return [self.copy(deep=False)]
+
+        if self.dtype == _dtype_obj:
+            # GH#53631
+            name = {1: "Series", 2: "DataFrame"}[self.ndim]
+            raise TypeError(f"{name} cannot interpolate with object dtype.")
+
+        copy, refs = self._get_refs_and_copy(inplace)
+
+        # Dispatch to the EA method.
+        new_values = self.array_values.interpolate(
+            method=method,
+            axis=self.ndim - 1,
+            index=index,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            copy=copy,
+            **kwargs,
+        )
+        data = extract_array(new_values, extract_numpy=True)
+        return [self.make_block_same_class(data, refs=refs)]
+
+    @final
+    def diff(self, n: int) -> list[Block]:
+        """return block for the diff of the values"""
+        # only reached with ndim == 2
+        # TODO(EA2D): transpose will be unnecessary with 2D EAs
+        new_values = algos.diff(self.values.T, n, axis=0).T
+        return [self.make_block(values=new_values)]
+
+    def shift(self, periods: int, fill_value: Any = None) -> list[Block]:
+        """shift the block by periods, possibly upcast"""
+        # convert integer to float if necessary. need to do a lot more than
+        # that, handle boolean etc also
+        axis = self.ndim - 1
+
+        # Note: periods is never 0 here, as that is handled at the top of
+        #  NDFrame.shift.  If that ever changes, we can do a check for periods=0
+        #  and possibly avoid coercing.
+
+        if not lib.is_scalar(fill_value) and self.dtype != _dtype_obj:
+            # with object dtype there is nothing to promote, and the user can
+            #  pass pretty much any weird fill_value they like
+            # see test_shift_object_non_scalar_fill
+            raise ValueError("fill_value must be a scalar")
+
+        fill_value = self._standardize_fill_value(fill_value)
+
+        try:
+            # error: Argument 1 to "np_can_hold_element" has incompatible type
+            # "Union[dtype[Any], ExtensionDtype]"; expected "dtype[Any]"
+            casted = np_can_hold_element(
+                self.dtype,  # type: ignore[arg-type]
+                fill_value,
+            )
+        except LossySetitemError:
+            nb = self.coerce_to_target_dtype(fill_value, raise_on_upcast=False)
+            return nb.shift(periods, fill_value=fill_value)
+
+        else:
+            values = cast(np.ndarray, self.values)
+            new_values = shift(values, periods, axis, casted)
+            return [self.make_block_same_class(new_values)]
+
+    @final
+    def quantile(
+        self,
+        qs: Index,  # with dtype float64
+        interpolation: QuantileInterpolation = "linear",
+    ) -> Block:
+        """
+        compute the quantiles of the
+
+        Parameters
+        ----------
+        qs : Index
+            The quantiles to be computed in float64.
+        interpolation : str, default 'linear'
+            Type of interpolation.
+
+        Returns
+        -------
+        Block
+        """
+        # We should always have ndim == 2 because Series dispatches to DataFrame
+        assert self.ndim == 2
+        assert is_list_like(qs)  # caller is responsible for this
+
+        result = quantile_compat(self.values, np.asarray(qs._values), interpolation)
+        # ensure_block_shape needed for cases where we start with EA and result
+        #  is ndarray, e.g. IntegerArray, SparseArray
+        result = ensure_block_shape(result, ndim=2)
+        return new_block_2d(result, placement=self._mgr_locs)
+
+    @final
+    def round(self, decimals: int) -> Self:
+        """
+        Rounds the values.
+        If the block is not of an integer or float dtype, nothing happens.
+        This is consistent with DataFrame.round behavior.
+        (Note: Series.round would raise)
+
+        Parameters
+        ----------
+        decimals: int,
+            Number of decimal places to round to.
+            Caller is responsible for validating this
+        """
+        if not self.is_numeric or self.is_bool:
+            return self.copy(deep=False)
+        # TODO: round only defined on BaseMaskedArray
+        # Series also does this, so would need to fix both places
+        # error: Item "ExtensionArray" of "Union[ndarray[Any, Any], ExtensionArray]"
+        # has no attribute "round"
+        values = self.values.round(decimals)  # type: ignore[union-attr]
+
+        refs = None
+        if values is self.values:
+            refs = self.refs
+
+        return self.make_block_same_class(values, refs=refs)
+
+    # ---------------------------------------------------------------------
+    # Abstract Methods Overridden By EABackedBlock and NumpyBlock
+
+    def delete(self, loc) -> list[Block]:
+        """Deletes the locs from the block.
+
+        We split the block to avoid copying the underlying data. We create new
+        blocks for every connected segment of the initial block that is not deleted.
+        The new blocks point to the initial array.
+        """
+        if not is_list_like(loc):
+            loc = [loc]
+
+        if self.ndim == 1:
+            values = cast(np.ndarray, self.values)
+            values = np.delete(values, loc)
+            mgr_locs = self._mgr_locs.delete(loc)
+            return [type(self)(values, placement=mgr_locs, ndim=self.ndim)]
+
+        if np.max(loc) >= self.values.shape[0]:
+            raise IndexError
+
+        # Add one out-of-bounds indexer as maximum to collect
+        # all columns after our last indexer if any
+        loc = np.concatenate([loc, [self.values.shape[0]]])
+        mgr_locs_arr = self._mgr_locs.as_array
+        new_blocks: list[Block] = []
+
+        previous_loc = -1
+        # TODO(CoW): This is tricky, if parent block goes out of scope
+        # all split blocks are referencing each other even though they
+        # don't share data
+        refs = self.refs if self.refs.has_reference() else None
+        for idx in loc:
+            if idx == previous_loc + 1:
+                # There is no column between current and last idx
+                pass
+            else:
+                # No overload variant of "__getitem__" of "ExtensionArray" matches
+                # argument type "Tuple[slice, slice]"
+                values = self.values[previous_loc + 1 : idx, :]  # type: ignore[call-overload]
+                locs = mgr_locs_arr[previous_loc + 1 : idx]
+                nb = type(self)(
+                    values, placement=BlockPlacement(locs), ndim=self.ndim, refs=refs
+                )
+                new_blocks.append(nb)
+
+            previous_loc = idx
+
+        return new_blocks
+
+    @property
+    def is_view(self) -> bool:
+        """return a boolean if I am possibly a view"""
+        raise AbstractMethodError(self)
+
+    @property
+    def array_values(self) -> ExtensionArray:
+        """
+        The array that Series.array returns. Always an ExtensionArray.
+        """
+        raise AbstractMethodError(self)
+
+    def get_values(self, dtype: DtypeObj | None = None) -> np.ndarray:
+        """
+        return an internal format, currently just the ndarray
+        this is often overridden to handle to_dense like operations
+        """
+        raise AbstractMethodError(self)
+
+
+class EABackedBlock(Block):
+    """
+    Mixin for Block subclasses backed by ExtensionArray.
+    """
+
+    values: ExtensionArray
+
+    @final
+    def shift(self, periods: int, fill_value: Any = None) -> list[Block]:
+        """
+        Shift the block by `periods`.
+
+        Dispatches to underlying ExtensionArray and re-boxes in an
+        ExtensionBlock.
+        """
+        # Transpose since EA.shift is always along axis=0, while we want to shift
+        #  along rows.
+        new_values = self.values.T.shift(periods=periods, fill_value=fill_value).T
+        return [self.make_block_same_class(new_values)]
+
+    @final
+    def setitem(self, indexer, value):
+        """
+        Attempt self.values[indexer] = value, possibly creating a new array.
+
+        This differs from Block.setitem by not allowing setitem to change
+        the dtype of the Block.
+
+        Parameters
+        ----------
+        indexer : tuple, list-like, array-like, slice, int
+            The subset of self.values to set
+        value : object
+            The value being set
+
+        Returns
+        -------
+        Block
+
+        Notes
+        -----
+        `indexer` is a direct slice/positional indexer. `value` must
+        be a compatible shape.
+        """
+        orig_indexer = indexer
+        orig_value = value
+
+        indexer = self._unwrap_setitem_indexer(indexer)
+        value = self._maybe_squeeze_arg(value)
+
+        values = self.values
+        if values.ndim == 2:
+            # TODO(GH#45419): string[pyarrow] tests break if we transpose
+            #  unconditionally
+            values = values.T
+        check_setitem_lengths(indexer, value, values)
+
+        try:
+            values[indexer] = value
+        except (ValueError, TypeError):
+            if isinstance(self.dtype, IntervalDtype):
+                # see TestSetitemFloatIntervalWithIntIntervalValues
+                nb = self.coerce_to_target_dtype(orig_value, raise_on_upcast=True)
+                return nb.setitem(orig_indexer, orig_value)
+
+            elif isinstance(self, NDArrayBackedExtensionBlock):
+                nb = self.coerce_to_target_dtype(orig_value, raise_on_upcast=True)
+                return nb.setitem(orig_indexer, orig_value)
+
+            else:
+                raise
+
+        else:
+            return self
+
+    @final
+    def where(self, other, cond) -> list[Block]:
+        arr = self.values.T
+
+        cond = extract_bool_array(cond)
+
+        orig_other = other
+        orig_cond = cond
+        other = self._maybe_squeeze_arg(other)
+        cond = self._maybe_squeeze_arg(cond)
+
+        if other is lib.no_default:
+            other = self.fill_value
+
+        icond, noop = validate_putmask(arr, ~cond)
+        if noop:
+            # GH#44181, GH#45135
+            # Avoid a) raising for Interval/PeriodDtype and b) unnecessary object upcast
+            return [self.copy(deep=False)]
+
+        try:
+            res_values = arr._where(cond, other).T
+        except (ValueError, TypeError):
+            if self.ndim == 1 or self.shape[0] == 1:
+                if isinstance(self.dtype, (IntervalDtype, StringDtype)):
+                    # TestSetitemFloatIntervalWithIntIntervalValues
+                    blk = self.coerce_to_target_dtype(orig_other, raise_on_upcast=False)
+                    if (
+                        self.ndim == 2
+                        and isinstance(orig_cond, np.ndarray)
+                        and orig_cond.ndim == 1
+                        and not is_1d_only_ea_dtype(blk.dtype)
+                    ):
+                        orig_cond = orig_cond[:, None]
+                    return blk.where(orig_other, orig_cond)
+
+                elif isinstance(self, NDArrayBackedExtensionBlock):
+                    # NB: not (yet) the same as
+                    #  isinstance(values, NDArrayBackedExtensionArray)
+                    blk = self.coerce_to_target_dtype(orig_other, raise_on_upcast=False)
+                    return blk.where(orig_other, orig_cond)
+
+                else:
+                    raise
+
+            else:
+                # Same pattern we use in Block.putmask
+                is_array = isinstance(orig_other, (np.ndarray, ExtensionArray))
+
+                res_blocks = []
+                for i, nb in enumerate(self._split()):
+                    n = orig_other
+                    if is_array:
+                        # we have a different value per-column
+                        n = orig_other[:, i : i + 1]
+
+                    submask = orig_cond[:, i : i + 1]
+                    rbs = nb.where(n, submask)
+                    res_blocks.extend(rbs)
+                return res_blocks
+
+        nb = self.make_block_same_class(res_values)
+        return [nb]
+
+    @final
+    def putmask(self, mask, new) -> list[Block]:
+        """
+        See Block.putmask.__doc__
+        """
+        mask = extract_bool_array(mask)
+        if new is lib.no_default:
+            new = self.fill_value
+
+        orig_new = new
+        orig_mask = mask
+        new = self._maybe_squeeze_arg(new)
+        mask = self._maybe_squeeze_arg(mask)
+
+        if not mask.any():
+            return [self.copy(deep=False)]
+
+        self = self._maybe_copy(inplace=True)
+        values = self.values
+        if values.ndim == 2:
+            values = values.T
+
+        try:
+            # Caller is responsible for ensuring matching lengths
+            values._putmask(mask, new)
+        except (TypeError, ValueError):
+            if self.ndim == 1 or self.shape[0] == 1:
+                if isinstance(self.dtype, IntervalDtype):
+                    # Discussion about what we want to support in the general
+                    #  case GH#39584
+                    blk = self.coerce_to_target_dtype(orig_new, raise_on_upcast=True)
+                    return blk.putmask(orig_mask, orig_new)
+
+                elif isinstance(self, NDArrayBackedExtensionBlock):
+                    # NB: not (yet) the same as
+                    #  isinstance(values, NDArrayBackedExtensionArray)
+                    blk = self.coerce_to_target_dtype(orig_new, raise_on_upcast=True)
+                    return blk.putmask(orig_mask, orig_new)
+
+                else:
+                    raise
+
+            else:
+                # Same pattern we use in Block.putmask
+                is_array = isinstance(orig_new, (np.ndarray, ExtensionArray))
+
+                res_blocks = []
+                for i, nb in enumerate(self._split()):
+                    n = orig_new
+                    if is_array:
+                        # we have a different value per-column
+                        n = orig_new[:, i : i + 1]
+
+                    submask = orig_mask[:, i : i + 1]
+                    rbs = nb.putmask(submask, n)
+                    res_blocks.extend(rbs)
+                return res_blocks
+
+        return [self]
+
+    @final
+    def delete(self, loc) -> list[Block]:
+        # This will be unnecessary if/when __array_function__ is implemented
+        if self.ndim == 1:
+            values = self.values.delete(loc)
+            mgr_locs = self._mgr_locs.delete(loc)
+            return [type(self)(values, placement=mgr_locs, ndim=self.ndim)]
+        elif self.values.ndim == 1:
+            # We get here through to_stata
+            return []
+        return super().delete(loc)
+
+    @final
+    @cache_readonly
+    def array_values(self) -> ExtensionArray:
+        return self.values
+
+    @final
+    def get_values(self, dtype: DtypeObj | None = None) -> np.ndarray:
+        """
+        return object dtype as boxed values, such as Timestamps/Timedelta
+        """
+        values: ArrayLike = self.values
+        if dtype == _dtype_obj:
+            values = values.astype(object)
+        # TODO(EA2D): reshape not needed with 2D EAs
+        return np.asarray(values).reshape(self.shape)
+
+    @final
+    def pad_or_backfill(
+        self,
+        *,
+        method: FillnaOptions,
+        inplace: bool = False,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+    ) -> list[Block]:
+        values = self.values
+
+        kwargs: dict[str, Any] = {"method": method, "limit": limit}
+        if "limit_area" in inspect.signature(values._pad_or_backfill).parameters:
+            kwargs["limit_area"] = limit_area
+        elif limit_area is not None:
+            raise NotImplementedError(
+                f"{type(values).__name__} does not implement limit_area "
+                "(added in pandas 2.2). 3rd-party ExtensionArray authors "
+                "need to add this argument to _pad_or_backfill."
+            )
+
+        if values.ndim == 2:
+            # NDArrayBackedExtensionArray.fillna assumes axis=0
+            new_values = values.T._pad_or_backfill(**kwargs).T
+        else:
+            new_values = values._pad_or_backfill(**kwargs)
+        return [self.make_block_same_class(new_values)]
+
+
+class ExtensionBlock(EABackedBlock):
+    """
+    Block for holding extension types.
+
+    Notes
+    -----
+    This holds all 3rd-party extension array types. It's also the immediate
+    parent class for our internal extension types' blocks.
+
+    ExtensionArrays are limited to 1-D.
+    """
+
+    values: ExtensionArray
+
+    def fillna(
+        self,
+        value,
+        limit: int | None = None,
+        inplace: bool = False,
+    ) -> list[Block]:
+        if isinstance(self.dtype, (IntervalDtype, StringDtype)):
+            # Block.fillna handles coercion (test_fillna_interval)
+            if isinstance(self.dtype, IntervalDtype) and limit is not None:
+                raise ValueError("limit must be None")
+            return super().fillna(
+                value=value,
+                limit=limit,
+                inplace=inplace,
+            )
+        if self._can_hold_na and not self.values._hasna:
+            refs = self.refs
+            new_values = self.values
+        else:
+            copy, refs = self._get_refs_and_copy(inplace)
+
+            try:
+                new_values = self.values.fillna(value=value, limit=limit, copy=copy)
+            except TypeError:
+                # 3rd party EA that has not implemented copy keyword yet
+                refs = None
+                new_values = self.values.fillna(value=value, limit=limit)
+                # issue the warning *after* retrying, in case the TypeError
+                #  was caused by an invalid fill_value
+                warnings.warn(
+                    # GH#53278
+                    "ExtensionArray.fillna added a 'copy' keyword in pandas "
+                    "2.1.0. In a future version, ExtensionArray subclasses will "
+                    "need to implement this keyword or an exception will be "
+                    "raised. In the interim, the keyword is ignored by "
+                    f"{type(self.values).__name__}.",
+                    DeprecationWarning,
+                    stacklevel=find_stack_level(),
+                )
+
+        return [self.make_block_same_class(new_values, refs=refs)]
+
+    @cache_readonly
+    def shape(self) -> Shape:
+        # TODO(EA2D): override unnecessary with 2D EAs
+        if self.ndim == 1:
+            return (len(self.values),)
+        return len(self._mgr_locs), len(self.values)
+
+    def iget(self, i: int | tuple[int, int] | tuple[slice, int]):
+        # In the case where we have a tuple[slice, int], the slice will always
+        #  be slice(None)
+        # We _could_ make the annotation more specific, but mypy would
+        #  complain about override mismatch:
+        #  Literal[0] | tuple[Literal[0], int] | tuple[slice, int]
+
+        # Note: only reached with self.ndim == 2
+
+        if isinstance(i, tuple):
+            # TODO(EA2D): unnecessary with 2D EAs
+            col, loc = i
+            if not com.is_null_slice(col) and col != 0:
+                raise IndexError(f"{self} only contains one item")
+            if isinstance(col, slice):
+                # the is_null_slice check above assures that col is slice(None)
+                #  so what we want is a view on all our columns and row loc
+                if loc < 0:
+                    loc += len(self.values)
+                # Note: loc:loc+1 vs [[loc]] makes a difference when called
+                #  from fast_xs because we want to get a view back.
+                return self.values[loc : loc + 1]
+            return self.values[loc]
+        else:
+            if i != 0:
+                raise IndexError(f"{self} only contains one item")
+            return self.values
+
+    def set_inplace(self, locs, values: ArrayLike, copy: bool = False) -> None:
+        # When an ndarray, we should have locs.tolist() == [0]
+        # When a BlockPlacement we should have list(locs) == [0]
+        if copy:
+            self.values = self.values.copy()
+        self.values[:] = values
+
+    def _maybe_squeeze_arg(self, arg):
+        """
+        If necessary, squeeze a (N, 1) ndarray to (N,)
+        """
+        # e.g. if we are passed a 2D mask for putmask
+        if (
+            isinstance(arg, (np.ndarray, ExtensionArray))
+            and arg.ndim == self.values.ndim + 1
+        ):
+            # TODO(EA2D): unnecessary with 2D EAs
+            assert arg.shape[1] == 1
+            # error: No overload variant of "__getitem__" of "ExtensionArray"
+            # matches argument type "Tuple[slice, int]"
+            arg = arg[:, 0]  # type: ignore[call-overload]
+        elif isinstance(arg, ABCDataFrame):
+            # 2022-01-06 only reached for setitem
+            # TODO: should we avoid getting here with DataFrame?
+            assert arg.shape[1] == 1
+            arg = arg._ixs(0, axis=1)._values
+
+        return arg
+
+    def _unwrap_setitem_indexer(self, indexer):
+        """
+        Adapt a 2D-indexer to our 1D values.
+
+        This is intended for 'setitem', not 'iget' or '_slice'.
+        """
+        # TODO: ATM this doesn't work for iget/_slice, can we change that?
+
+        if isinstance(indexer, tuple) and len(indexer) == 2:
+            # TODO(EA2D): not needed with 2D EAs
+            #  Should never have length > 2.  Caller is responsible for checking.
+            #  Length 1 is reached vis setitem_single_block and setitem_single_column
+            #  each of which pass indexer=(pi,)
+            if all(isinstance(x, np.ndarray) and x.ndim == 2 for x in indexer):
+                # GH#44703 went through indexing.maybe_convert_ix
+                first, second = indexer
+                if not (
+                    second.size == 1 and (second == 0).all() and first.shape[1] == 1
+                ):
+                    raise NotImplementedError(
+                        "This should not be reached. Please report a bug at "
+                        "github.com/pandas-dev/pandas/"
+                    )
+                indexer = first[:, 0]
+
+            elif lib.is_integer(indexer[1]) and indexer[1] == 0:
+                # reached via setitem_single_block passing the whole indexer
+                indexer = indexer[0]
+
+            elif com.is_null_slice(indexer[1]):
+                indexer = indexer[0]
+
+            elif is_list_like(indexer[1]) and indexer[1][0] == 0:
+                indexer = indexer[0]
+
+            else:
+                raise NotImplementedError(
+                    "This should not be reached. Please report a bug at "
+                    "github.com/pandas-dev/pandas/"
+                )
+        return indexer
+
+    @property
+    def is_view(self) -> bool:
+        """Extension arrays are never treated as views."""
+        return False
+
+    # error: Cannot override writeable attribute with read-only property
+    @cache_readonly
+    def is_numeric(self) -> bool:  # type: ignore[override]
+        return self.values.dtype._is_numeric
+
+    def _slice(
+        self, slicer: slice | npt.NDArray[np.bool_] | npt.NDArray[np.intp]
+    ) -> ExtensionArray:
+        """
+        Return a slice of my values.
+
+        Parameters
+        ----------
+        slicer : slice, ndarray[int], or ndarray[bool]
+            Valid (non-reducing) indexer for self.values.
+
+        Returns
+        -------
+        ExtensionArray
+        """
+        # Notes: ndarray[bool] is only reachable when via get_rows_with_mask, which
+        #  is only for Series, i.e. self.ndim == 1.
+
+        # return same dims as we currently have
+        if self.ndim == 2:
+            # reached via getitem_block via _slice_take_blocks_ax0
+            # TODO(EA2D): won't be necessary with 2D EAs
+
+            if not isinstance(slicer, slice):
+                raise AssertionError(
+                    "invalid slicing for a 1-ndim ExtensionArray", slicer
+                )
+            # GH#32959 only full-slicers along fake-dim0 are valid
+            # TODO(EA2D): won't be necessary with 2D EAs
+            # range(1) instead of self._mgr_locs to avoid exception on [::-1]
+            #  see test_iloc_getitem_slice_negative_step_ea_block
+            new_locs = range(1)[slicer]
+            if not len(new_locs):
+                raise AssertionError(
+                    "invalid slicing for a 1-ndim ExtensionArray", slicer
+                )
+            slicer = slice(None)
+
+        return self.values[slicer]
+
+    @final
+    def slice_block_rows(self, slicer: slice) -> Self:
+        """
+        Perform __getitem__-like specialized to slicing along index.
+        """
+        # GH#42787 in principle this is equivalent to values[..., slicer], but we don't
+        # require subclasses of ExtensionArray to support that form (for now).
+        new_values = self.values[slicer]
+        return type(self)(new_values, self._mgr_locs, ndim=self.ndim, refs=self.refs)
+
+    def _unstack(
+        self,
+        unstacker,
+        fill_value,
+        new_placement: npt.NDArray[np.intp],
+        needs_masking: npt.NDArray[np.bool_],
+    ):
+        # ExtensionArray-safe unstack.
+        # We override Block._unstack, which unstacks directly on the
+        # values of the array. For EA-backed blocks, this would require
+        # converting to a 2-D ndarray of objects.
+        # Instead, we unstack an ndarray of integer positions, followed by
+        # a `take` on the actual values.
+
+        # Caller is responsible for ensuring self.shape[-1] == len(unstacker.index)
+        new_values, mask = unstacker.arange_result
+
+        # Note: these next two lines ensure that
+        #  mask.sum() == sum(len(nb.mgr_locs) for nb in blocks)
+        #  which the calling function needs in order to pass verify_integrity=False
+        #  to the BlockManager constructor
+        new_values = new_values.T[mask]
+        new_placement = new_placement[mask]
+
+        # needs_masking[i] calculated once in BlockManager.unstack tells
+        #  us if there are any -1s in the relevant indices.  When False,
+        #  that allows us to go through a faster path in 'take', among
+        #  other things avoiding e.g. Categorical._validate_scalar.
+        blocks = [
+            # TODO: could cast to object depending on fill_value?
+            type(self)(
+                self.values.take(
+                    indices, allow_fill=needs_masking[i], fill_value=fill_value
+                ),
+                BlockPlacement(place),
+                ndim=2,
+            )
+            for i, (indices, place) in enumerate(zip(new_values, new_placement))
+        ]
+        return blocks, mask
+
+
+class NumpyBlock(Block):
+    values: np.ndarray
+    __slots__ = ()
+
+    @property
+    def is_view(self) -> bool:
+        """return a boolean if I am possibly a view"""
+        return self.values.base is not None
+
+    @property
+    def array_values(self) -> ExtensionArray:
+        return NumpyExtensionArray(self.values)
+
+    def get_values(self, dtype: DtypeObj | None = None) -> np.ndarray:
+        if dtype == _dtype_obj:
+            return self.values.astype(_dtype_obj)
+        return self.values
+
+    @cache_readonly
+    def is_numeric(self) -> bool:  # type: ignore[override]
+        dtype = self.values.dtype
+        kind = dtype.kind
+
+        return kind in "fciub"
+
+
+class NDArrayBackedExtensionBlock(EABackedBlock):
+    """
+    Block backed by an NDArrayBackedExtensionArray
+    """
+
+    values: NDArrayBackedExtensionArray
+
+    @property
+    def is_view(self) -> bool:
+        """return a boolean if I am possibly a view"""
+        # check the ndarray values of the DatetimeIndex values
+        return self.values._ndarray.base is not None
+
+
+class DatetimeLikeBlock(NDArrayBackedExtensionBlock):
+    """Block for datetime64[ns], timedelta64[ns]."""
+
+    __slots__ = ()
+    is_numeric = False
+    values: DatetimeArray | TimedeltaArray
+
+
+# -----------------------------------------------------------------
+# Constructor Helpers
+
+
+def maybe_coerce_values(values: ArrayLike) -> ArrayLike:
+    """
+    Input validation for values passed to __init__. Ensure that
+    any datetime64/timedelta64 dtypes are in nanoseconds.  Ensure
+    that we do not have string dtypes.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+
+    Returns
+    -------
+    values : np.ndarray or ExtensionArray
+    """
+    # Caller is responsible for ensuring NumpyExtensionArray is already extracted.
+
+    if isinstance(values, np.ndarray):
+        values = ensure_wrapped_if_datetimelike(values)
+
+        if issubclass(values.dtype.type, str):
+            values = np.array(values, dtype=object)
+
+    if isinstance(values, (DatetimeArray, TimedeltaArray)) and values.freq is not None:
+        # freq is only stored in DatetimeIndex/TimedeltaIndex, not in Series/DataFrame
+        values = values._with_freq(None)
+
+    return values
+
+
+def get_block_type(dtype: DtypeObj) -> type[Block]:
+    """
+    Find the appropriate Block subclass to use for the given values and dtype.
+
+    Parameters
+    ----------
+    dtype : numpy or pandas dtype
+
+    Returns
+    -------
+    cls : class, subclass of Block
+    """
+    if isinstance(dtype, DatetimeTZDtype):
+        return DatetimeLikeBlock
+    elif isinstance(dtype, PeriodDtype):
+        return NDArrayBackedExtensionBlock
+    elif isinstance(dtype, ExtensionDtype):
+        # Note: need to be sure NumpyExtensionArray is unwrapped before we get here
+        return ExtensionBlock
+
+    # We use kind checks because it is much more performant
+    #  than is_foo_dtype
+    kind = dtype.kind
+    if kind in "Mm":
+        return DatetimeLikeBlock
+
+    return NumpyBlock
+
+
+def new_block_2d(
+    values: ArrayLike, placement: BlockPlacement, refs: BlockValuesRefs | None = None
+) -> Block:
+    # new_block specialized to case with
+    #  ndim=2
+    #  isinstance(placement, BlockPlacement)
+    #  check_ndim/ensure_block_shape already checked
+    klass = get_block_type(values.dtype)
+
+    values = maybe_coerce_values(values)
+    return klass(values, ndim=2, placement=placement, refs=refs)
+
+
+def new_block(
+    values,
+    placement: BlockPlacement,
+    *,
+    ndim: int,
+    refs: BlockValuesRefs | None = None,
+) -> Block:
+    # caller is responsible for ensuring:
+    # - values is NOT a NumpyExtensionArray
+    # - check_ndim/ensure_block_shape already checked
+    # - maybe_coerce_values already called/unnecessary
+    klass = get_block_type(values.dtype)
+    return klass(values, ndim=ndim, placement=placement, refs=refs)
+
+
+def check_ndim(values, placement: BlockPlacement, ndim: int) -> None:
+    """
+    ndim inference and validation.
+
+    Validates that values.ndim and ndim are consistent.
+    Validates that len(values) and len(placement) are consistent.
+
+    Parameters
+    ----------
+    values : array-like
+    placement : BlockPlacement
+    ndim : int
+
+    Raises
+    ------
+    ValueError : the number of dimensions do not match
+    """
+
+    if values.ndim > ndim:
+        # Check for both np.ndarray and ExtensionArray
+        raise ValueError(
+            f"Wrong number of dimensions. values.ndim > ndim [{values.ndim} > {ndim}]"
+        )
+
+    if not is_1d_only_ea_dtype(values.dtype):
+        # TODO(EA2D): special case not needed with 2D EAs
+        if values.ndim != ndim:
+            raise ValueError(
+                "Wrong number of dimensions. "
+                f"values.ndim != ndim [{values.ndim} != {ndim}]"
+            )
+        if len(placement) != len(values):
+            raise ValueError(
+                f"Wrong number of items passed {len(values)}, "
+                f"placement implies {len(placement)}"
+            )
+    elif ndim == 2 and len(placement) != 1:
+        # TODO(EA2D): special case unnecessary with 2D EAs
+        raise ValueError("need to split")
+
+
+def extract_pandas_array(
+    values: ArrayLike, dtype: DtypeObj | None, ndim: int
+) -> tuple[ArrayLike, DtypeObj | None]:
+    """
+    Ensure that we don't allow NumpyExtensionArray / NumpyEADtype in internals.
+    """
+    # For now, blocks should be backed by ndarrays when possible.
+    if isinstance(values, ABCNumpyExtensionArray):
+        values = values.to_numpy()
+        if ndim and ndim > 1:
+            # TODO(EA2D): special case not needed with 2D EAs
+            values = np.atleast_2d(values)
+
+    if isinstance(dtype, NumpyEADtype):
+        dtype = dtype.numpy_dtype
+
+    return values, dtype
+
+
+# -----------------------------------------------------------------
+
+
+def extend_blocks(result, blocks=None) -> list[Block]:
+    """return a new extended blocks, given the result"""
+    if blocks is None:
+        blocks = []
+    if isinstance(result, list):
+        for r in result:
+            if isinstance(r, list):
+                blocks.extend(r)
+            else:
+                blocks.append(r)
+    else:
+        assert isinstance(result, Block), type(result)
+        blocks.append(result)
+    return blocks
+
+
+def ensure_block_shape(values: ArrayLike, ndim: int = 1) -> ArrayLike:
+    """
+    Reshape if possible to have values.ndim == ndim.
+    """
+
+    if values.ndim < ndim:
+        if not is_1d_only_ea_dtype(values.dtype):
+            # TODO(EA2D): https://github.com/pandas-dev/pandas/issues/23023
+            # block.shape is incorrect for "2D" ExtensionArrays
+            # We can't, and don't need to, reshape.
+            values = cast("np.ndarray | DatetimeArray | TimedeltaArray", values)
+            values = values.reshape(1, -1)
+
+    return values
+
+
+def external_values(values: ArrayLike) -> ArrayLike:
+    """
+    The array that Series.values returns (public attribute).
+
+    This has some historical constraints, and is overridden in block
+    subclasses to return the correct array (e.g. period returns
+    object ndarray and datetimetz a datetime64[ns] ndarray instead of
+    proper extension array).
+    """
+    if isinstance(values, (PeriodArray, IntervalArray)):
+        return values.astype(object)
+    elif isinstance(values, (DatetimeArray, TimedeltaArray)):
+        # NB: for datetime64tz this is different from np.asarray(values), since
+        #  that returns an object-dtype ndarray of Timestamps.
+        # Avoid raising in .astype in casting from dt64tz to dt64
+        values = values._ndarray
+
+    if isinstance(values, np.ndarray):
+        values = values.view()
+        values.flags.writeable = False
+
+    # TODO(CoW) we should also mark our ExtensionArrays as read-only
+
+    return values
diff --git a/pandas/pandas/core/internals/concat.py b/pandas/pandas/core/internals/concat.py
new file mode 100644
index 0000000000000..2ee7d3948a70f
--- /dev/null
+++ b/pandas/pandas/core/internals/concat.py
@@ -0,0 +1,479 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    algos as libalgos,
+    internals as libinternals,
+    lib,
+)
+from pandas._libs.missing import NA
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.cast import (
+    ensure_dtype_can_hold_na,
+    find_common_type,
+)
+from pandas.core.dtypes.common import (
+    is_1d_only_ea_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.missing import is_valid_na_for_dtype
+
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+from pandas.core.internals.blocks import (
+    ensure_block_shape,
+    new_block_2d,
+)
+from pandas.core.internals.managers import (
+    BlockManager,
+    make_na_array,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        DtypeObj,
+        Shape,
+    )
+
+    from pandas import Index
+    from pandas.core.internals.blocks import (
+        Block,
+        BlockPlacement,
+    )
+
+
+def concatenate_managers(
+    mgrs_indexers, axes: list[Index], concat_axis: AxisInt, copy: bool
+) -> BlockManager:
+    """
+    Concatenate block managers into one.
+
+    Parameters
+    ----------
+    mgrs_indexers : list of (BlockManager, {axis: indexer,...}) tuples
+    axes : list of Index
+    concat_axis : int
+    copy : bool
+
+    Returns
+    -------
+    BlockManager
+    """
+
+    needs_copy = copy and concat_axis == 0
+
+    # Assertions disabled for performance
+    # for tup in mgrs_indexers:
+    #    # caller is responsible for ensuring this
+    #    indexers = tup[1]
+    #    assert concat_axis not in indexers
+
+    if concat_axis == 0:
+        mgrs = _maybe_reindex_columns_na_proxy(axes, mgrs_indexers, needs_copy)
+        return mgrs[0].concat_horizontal(mgrs, axes)
+
+    if len(mgrs_indexers) > 0 and mgrs_indexers[0][0].nblocks > 0:
+        first_dtype = mgrs_indexers[0][0].blocks[0].dtype
+        if first_dtype in [np.float64, np.float32]:
+            # TODO: support more dtypes here.  This will be simpler once
+            #  JoinUnit.is_na behavior is deprecated.
+            #  (update 2024-04-13 that deprecation has been enforced)
+            if (
+                all(_is_homogeneous_mgr(mgr, first_dtype) for mgr, _ in mgrs_indexers)
+                and len(mgrs_indexers) > 1
+            ):
+                # Fastpath!
+                # Length restriction is just to avoid having to worry about 'copy'
+                shape = tuple(len(x) for x in axes)
+                nb = _concat_homogeneous_fastpath(mgrs_indexers, shape, first_dtype)
+                return BlockManager((nb,), axes)
+
+    mgrs = _maybe_reindex_columns_na_proxy(axes, mgrs_indexers, needs_copy)
+
+    if len(mgrs) == 1:
+        mgr = mgrs[0]
+        out = mgr.copy(deep=False)
+        out.axes = axes
+        return out
+
+    blocks = []
+    values: ArrayLike
+
+    for placement, join_units in _get_combined_plan(mgrs):
+        unit = join_units[0]
+        blk = unit.block
+
+        if _is_uniform_join_units(join_units):
+            vals = [ju.block.values for ju in join_units]
+
+            if not blk.is_extension:
+                # _is_uniform_join_units ensures a single dtype, so
+                #  we can use np.concatenate, which is more performant
+                #  than concat_compat
+                # error: Argument 1 to "concatenate" has incompatible type
+                # "List[Union[ndarray[Any, Any], ExtensionArray]]";
+                # expected "Union[_SupportsArray[dtype[Any]],
+                # _NestedSequence[_SupportsArray[dtype[Any]]]]"
+                values = np.concatenate(vals, axis=1)  # type: ignore[arg-type]
+            elif is_1d_only_ea_dtype(blk.dtype):
+                # TODO(EA2D): special-casing not needed with 2D EAs
+                values = concat_compat(vals, axis=0, ea_compat_axis=True)
+                values = ensure_block_shape(values, ndim=2)
+            else:
+                values = concat_compat(vals, axis=1)
+
+            values = ensure_wrapped_if_datetimelike(values)
+
+            fastpath = blk.values.dtype == values.dtype
+        else:
+            values = _concatenate_join_units(join_units, copy=copy)
+            fastpath = False
+
+        if fastpath:
+            b = blk.make_block_same_class(values, placement=placement)
+        else:
+            b = new_block_2d(values, placement=placement)
+
+        blocks.append(b)
+
+    return BlockManager(tuple(blocks), axes)
+
+
+def _maybe_reindex_columns_na_proxy(
+    axes: list[Index],
+    mgrs_indexers: list[tuple[BlockManager, dict[int, np.ndarray]]],
+    needs_copy: bool,
+) -> list[BlockManager]:
+    """
+    Reindex along columns so that all of the BlockManagers being concatenated
+    have matching columns.
+
+    Columns added in this reindexing have dtype=np.void, indicating they
+    should be ignored when choosing a column's final dtype.
+    """
+    new_mgrs = []
+
+    for mgr, indexers in mgrs_indexers:
+        # For axis=0 (i.e. columns) we use_na_proxy and only_slice, so this
+        #  is a cheap reindexing.
+        for i, indexer in indexers.items():
+            mgr = mgr.reindex_indexer(
+                axes[i],
+                indexers[i],
+                axis=i,
+                only_slice=True,  # only relevant for i==0
+                allow_dups=True,
+                use_na_proxy=True,  # only relevant for i==0
+            )
+        if needs_copy and not indexers:
+            mgr = mgr.copy()
+
+        new_mgrs.append(mgr)
+    return new_mgrs
+
+
+def _is_homogeneous_mgr(mgr: BlockManager, first_dtype: DtypeObj) -> bool:
+    """
+    Check if this Manager can be treated as a single ndarray.
+    """
+    if mgr.nblocks != 1:
+        return False
+    blk = mgr.blocks[0]
+    if not (blk.mgr_locs.is_slice_like and blk.mgr_locs.as_slice.step == 1):
+        return False
+
+    return blk.dtype == first_dtype
+
+
+def _concat_homogeneous_fastpath(
+    mgrs_indexers, shape: Shape, first_dtype: np.dtype
+) -> Block:
+    """
+    With single-Block managers with homogeneous dtypes (that can already hold nan),
+    we avoid [...]
+    """
+    # assumes
+    #  all(_is_homogeneous_mgr(mgr, first_dtype) for mgr, _ in in mgrs_indexers)
+
+    if all(not indexers for _, indexers in mgrs_indexers):
+        # https://github.com/pandas-dev/pandas/pull/52685#issuecomment-1523287739
+        arrs = [mgr.blocks[0].values.T for mgr, _ in mgrs_indexers]
+        arr = np.concatenate(arrs).T
+        bp = libinternals.BlockPlacement(slice(shape[0]))
+        nb = new_block_2d(arr, bp)
+        return nb
+
+    arr = np.empty(shape, dtype=first_dtype)
+
+    if first_dtype == np.float64:
+        take_func = libalgos.take_2d_axis0_float64_float64
+    else:
+        take_func = libalgos.take_2d_axis0_float32_float32
+
+    start = 0
+    for mgr, indexers in mgrs_indexers:
+        mgr_len = mgr.shape[1]
+        end = start + mgr_len
+
+        if 0 in indexers:
+            take_func(
+                mgr.blocks[0].values,
+                indexers[0],
+                arr[:, start:end],
+            )
+        else:
+            # No reindexing necessary, we can copy values directly
+            arr[:, start:end] = mgr.blocks[0].values
+
+        start += mgr_len
+
+    bp = libinternals.BlockPlacement(slice(shape[0]))
+    nb = new_block_2d(arr, bp)
+    return nb
+
+
+def _get_combined_plan(
+    mgrs: list[BlockManager],
+) -> Generator[tuple[BlockPlacement, list[JoinUnit]]]:
+    max_len = mgrs[0].shape[0]
+
+    blknos_list = [mgr.blknos for mgr in mgrs]
+    pairs = libinternals.get_concat_blkno_indexers(blknos_list)
+    for blknos, bp in pairs:
+        # assert bp.is_slice_like
+        # assert len(bp) > 0
+
+        units_for_bp = []
+        for k, mgr in enumerate(mgrs):
+            blkno = blknos[k]
+
+            nb = _get_block_for_concat_plan(mgr, bp, blkno, max_len=max_len)
+            unit = JoinUnit(nb)
+            units_for_bp.append(unit)
+
+        yield bp, units_for_bp
+
+
+def _get_block_for_concat_plan(
+    mgr: BlockManager, bp: BlockPlacement, blkno: int, *, max_len: int
+) -> Block:
+    blk = mgr.blocks[blkno]
+    # Assertions disabled for performance:
+    #  assert bp.is_slice_like
+    #  assert blkno != -1
+    #  assert (mgr.blknos[bp] == blkno).all()
+
+    if len(bp) == len(blk.mgr_locs) and (
+        blk.mgr_locs.is_slice_like and blk.mgr_locs.as_slice.step == 1
+    ):
+        nb = blk
+    else:
+        ax0_blk_indexer = mgr.blklocs[bp.indexer]
+
+        slc = lib.maybe_indices_to_slice(ax0_blk_indexer, max_len)
+        # TODO: in all extant test cases 2023-04-08 we have a slice here.
+        #  Will this always be the case?
+        if isinstance(slc, slice):
+            nb = blk.slice_block_columns(slc)
+        else:
+            nb = blk.take_block_columns(slc)
+
+    # assert nb.shape == (len(bp), mgr.shape[1])
+    return nb
+
+
+class JoinUnit:
+    def __init__(self, block: Block) -> None:
+        self.block = block
+
+    def __repr__(self) -> str:
+        return f"{type(self).__name__}({self.block!r})"
+
+    def _is_valid_na_for(self, dtype: DtypeObj) -> bool:
+        """
+        Check that we are all-NA of a type/dtype that is compatible with this dtype.
+        Augments `self.is_na` with an additional check of the type of NA values.
+        """
+        if not self.is_na:
+            return False
+
+        blk = self.block
+        if blk.dtype.kind == "V":
+            return True
+
+        if blk.dtype == object:
+            values = blk.values
+            return all(is_valid_na_for_dtype(x, dtype) for x in values.ravel(order="K"))
+
+        na_value = blk.fill_value
+        if na_value is NaT and blk.dtype != dtype:
+            # e.g. we are dt64 and other is td64
+            # fill_values match but we should not cast blk.values to dtype
+            # TODO: this will need updating if we ever have non-nano dt64/td64
+            return False
+
+        if na_value is NA and needs_i8_conversion(dtype):
+            # FIXME: kludge; test_append_empty_frame_with_timedelta64ns_nat
+            #  e.g. blk.dtype == "Int64" and dtype is td64, we dont want
+            #  to consider these as matching
+            return False
+
+        # TODO: better to use can_hold_element?
+        return is_valid_na_for_dtype(na_value, dtype)
+
+    @cache_readonly
+    def is_na(self) -> bool:
+        blk = self.block
+        if blk.dtype.kind == "V":
+            return True
+        return False
+
+    def get_reindexed_values(self, empty_dtype: DtypeObj, upcasted_na) -> ArrayLike:
+        values: ArrayLike
+
+        if upcasted_na is None and self.block.dtype.kind != "V":
+            # No upcasting is necessary
+            return self.block.values
+        else:
+            fill_value = upcasted_na
+
+            if self._is_valid_na_for(empty_dtype):
+                # note: always holds when self.block.dtype.kind == "V"
+                blk_dtype = self.block.dtype
+
+                if blk_dtype == np.dtype("object"):
+                    # we want to avoid filling with np.nan if we are
+                    # using None; we already know that we are all
+                    # nulls
+                    values = cast(np.ndarray, self.block.values)
+                    if values.size and values[0, 0] is None:
+                        fill_value = None
+
+                return make_na_array(empty_dtype, self.block.shape, fill_value)
+
+            return self.block.values
+
+
+def _concatenate_join_units(join_units: list[JoinUnit], copy: bool) -> ArrayLike:
+    """
+    Concatenate values from several join units along axis=1.
+    """
+    empty_dtype = _get_empty_dtype(join_units)
+
+    has_none_blocks = any(unit.block.dtype.kind == "V" for unit in join_units)
+    upcasted_na = _dtype_to_na_value(empty_dtype, has_none_blocks)
+
+    to_concat = [
+        ju.get_reindexed_values(empty_dtype=empty_dtype, upcasted_na=upcasted_na)
+        for ju in join_units
+    ]
+
+    if any(is_1d_only_ea_dtype(t.dtype) for t in to_concat):
+        # TODO(EA2D): special case not needed if all EAs used HybridBlocks
+
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
+        to_concat = [
+            t if is_1d_only_ea_dtype(t.dtype) else t[0, :]  # type: ignore[call-overload]
+            for t in to_concat
+        ]
+        concat_values = concat_compat(to_concat, axis=0, ea_compat_axis=True)
+        concat_values = ensure_block_shape(concat_values, 2)
+
+    else:
+        concat_values = concat_compat(to_concat, axis=1)
+
+    return concat_values
+
+
+def _dtype_to_na_value(dtype: DtypeObj, has_none_blocks: bool):
+    """
+    Find the NA value to go with this dtype.
+    """
+    if isinstance(dtype, ExtensionDtype):
+        return dtype.na_value
+    elif dtype.kind in "mM":
+        return dtype.type("NaT")
+    elif dtype.kind in "fc":
+        return dtype.type("NaN")
+    elif dtype.kind == "b":
+        # different from missing.na_value_for_dtype
+        return None
+    elif dtype.kind in "iu":
+        if not has_none_blocks:
+            # different from missing.na_value_for_dtype
+            return None
+        return np.nan
+    elif dtype.kind == "O":
+        return np.nan
+    raise NotImplementedError
+
+
+def _get_empty_dtype(join_units: Sequence[JoinUnit]) -> DtypeObj:
+    """
+    Return dtype and N/A values to use when concatenating specified units.
+
+    Returned N/A value may be None which means there was no casting involved.
+
+    Returns
+    -------
+    dtype
+    """
+    if lib.dtypes_all_equal([ju.block.dtype for ju in join_units]):
+        empty_dtype = join_units[0].block.dtype
+        return empty_dtype
+
+    has_none_blocks = any(unit.block.dtype.kind == "V" for unit in join_units)
+
+    dtypes = [unit.block.dtype for unit in join_units if not unit.is_na]
+
+    dtype = find_common_type(dtypes)
+    if has_none_blocks:
+        dtype = ensure_dtype_can_hold_na(dtype)
+
+    return dtype
+
+
+def _is_uniform_join_units(join_units: list[JoinUnit]) -> bool:
+    """
+    Check if the join units consist of blocks of uniform type that can
+    be concatenated using Block.concat_same_type instead of the generic
+    _concatenate_join_units (which uses `concat_compat`).
+
+    """
+    first = join_units[0].block
+    if first.dtype.kind == "V":
+        return False
+    return (
+        # exclude cases where a) ju.block is None or b) we have e.g. Int64+int64
+        all(type(ju.block) is type(first) for ju in join_units)
+        and
+        # e.g. DatetimeLikeBlock can be dt64 or td64, but these are not uniform
+        all(
+            ju.block.dtype == first.dtype
+            # GH#42092 we only want the dtype_equal check for non-numeric blocks
+            #  (for now, may change but that would need a deprecation)
+            or ju.block.dtype.kind in "iub"
+            for ju in join_units
+        )
+        and
+        # no blocks that would get missing values (can lead to type upcasts)
+        # unless we're an extension dtype.
+        all(not ju.is_na or ju.block.is_extension for ju in join_units)
+    )
diff --git a/pandas/pandas/core/internals/construction.py b/pandas/pandas/core/internals/construction.py
new file mode 100644
index 0000000000000..69da2be0306f6
--- /dev/null
+++ b/pandas/pandas/core/internals/construction.py
@@ -0,0 +1,995 @@
+"""
+Functions for preparing various inputs passed to the DataFrame or Series
+constructors before passing them to a BlockManager.
+"""
+
+from __future__ import annotations
+
+from collections import abc
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+from numpy import ma
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.astype import astype_is_view
+from pandas.core.dtypes.cast import (
+    construct_1d_arraylike_from_scalar,
+    dict_compat,
+    maybe_cast_to_datetime,
+    maybe_convert_platform,
+    maybe_infer_to_datetimelike,
+)
+from pandas.core.dtypes.common import (
+    is_1d_only_ea_dtype,
+    is_integer_dtype,
+    is_list_like,
+    is_named_tuple,
+    is_object_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import (
+    algorithms,
+    common as com,
+)
+from pandas.core.arrays import ExtensionArray
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.construction import (
+    array as pd_array,
+    extract_array,
+    range_to_ndarray,
+    sanitize_array,
+)
+from pandas.core.indexes.api import (
+    DatetimeIndex,
+    Index,
+    TimedeltaIndex,
+    default_index,
+    ensure_index,
+    get_objs_combined_axis,
+    maybe_sequence_to_range,
+    union_indexes,
+)
+from pandas.core.internals.blocks import (
+    BlockPlacement,
+    ensure_block_shape,
+    new_block,
+    new_block_2d,
+)
+from pandas.core.internals.managers import (
+    create_block_manager_from_blocks,
+    create_block_manager_from_column_arrays,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        DtypeObj,
+        Manager,
+        npt,
+    )
+# ---------------------------------------------------------------------
+# BlockManager Interface
+
+
+def arrays_to_mgr(
+    arrays,
+    columns: Index,
+    index,
+    *,
+    dtype: DtypeObj | None = None,
+    verify_integrity: bool = True,
+    consolidate: bool = True,
+) -> Manager:
+    """
+    Segregate Series based on type and coerce into matrices.
+
+    Needs to handle a lot of exceptional cases.
+    """
+    if verify_integrity:
+        # figure out the index, if necessary
+        if index is None:
+            index = _extract_index(arrays)
+        else:
+            index = ensure_index(index)
+
+        # don't force copy because getting jammed in an ndarray anyway
+        arrays, refs = _homogenize(arrays, index, dtype)
+        # _homogenize ensures
+        #  - all(len(x) == len(index) for x in arrays)
+        #  - all(x.ndim == 1 for x in arrays)
+        #  - all(isinstance(x, (np.ndarray, ExtensionArray)) for x in arrays)
+        #  - all(type(x) is not NumpyExtensionArray for x in arrays)
+
+    else:
+        index = ensure_index(index)
+        arrays = [extract_array(x, extract_numpy=True) for x in arrays]
+        # with _from_arrays, the passed arrays should never be Series objects
+        refs = [None] * len(arrays)
+
+        # Reached via DataFrame._from_arrays; we do minimal validation here
+        for arr in arrays:
+            if (
+                not isinstance(arr, (np.ndarray, ExtensionArray))
+                or arr.ndim != 1
+                or len(arr) != len(index)
+            ):
+                raise ValueError(
+                    "Arrays must be 1-dimensional np.ndarray or ExtensionArray "
+                    "with length matching len(index)"
+                )
+
+    columns = ensure_index(columns)
+    if len(columns) != len(arrays):
+        raise ValueError("len(arrays) must match len(columns)")
+
+    # from BlockManager perspective
+    axes = [columns, index]
+
+    return create_block_manager_from_column_arrays(
+        arrays, axes, consolidate=consolidate, refs=refs
+    )
+
+
+def rec_array_to_mgr(
+    data: np.rec.recarray | np.ndarray,
+    index,
+    columns,
+    dtype: DtypeObj | None,
+    copy: bool,
+) -> Manager:
+    """
+    Extract from a masked rec array and create the manager.
+    """
+    # essentially process a record array then fill it
+    fdata = ma.getdata(data)
+    if index is None:
+        index = default_index(len(fdata))
+    else:
+        index = ensure_index(index)
+
+    if columns is not None:
+        columns = ensure_index(columns)
+    arrays, arr_columns = to_arrays(fdata, columns)
+
+    # create the manager
+
+    arrays, arr_columns = reorder_arrays(arrays, arr_columns, columns, len(index))
+    if columns is None:
+        columns = arr_columns
+
+    mgr = arrays_to_mgr(arrays, columns, index, dtype=dtype)
+
+    if copy:
+        mgr = mgr.copy()
+    return mgr
+
+
+# ---------------------------------------------------------------------
+# DataFrame Constructor Interface
+
+
+def ndarray_to_mgr(
+    values, index, columns, dtype: DtypeObj | None, copy: bool
+) -> Manager:
+    # used in DataFrame.__init__
+    # input must be a ndarray, list, Series, Index, ExtensionArray
+    infer_object = not isinstance(values, (ABCSeries, Index, ExtensionArray))
+
+    if isinstance(values, ABCSeries):
+        if columns is None:
+            if values.name is not None:
+                columns = Index([values.name])
+        if index is None:
+            index = values.index
+        else:
+            values = values.reindex(index)
+
+        # zero len case (GH #2234)
+        if not len(values) and columns is not None and len(columns):
+            values = np.empty((0, 1), dtype=object)
+
+    vdtype = getattr(values, "dtype", None)
+    refs = None
+    if is_1d_only_ea_dtype(vdtype) or is_1d_only_ea_dtype(dtype):
+        # GH#19157
+
+        if isinstance(values, (np.ndarray, ExtensionArray)) and values.ndim > 1:
+            # GH#12513 a EA dtype passed with a 2D array, split into
+            #  multiple EAs that view the values
+            # error: No overload variant of "__getitem__" of "ExtensionArray"
+            # matches argument type "Tuple[slice, int]"
+            values = [
+                values[:, n]  # type: ignore[call-overload]
+                for n in range(values.shape[1])
+            ]
+        else:
+            values = [values]
+
+        if columns is None:
+            columns = Index(range(len(values)))
+        else:
+            columns = ensure_index(columns)
+
+        return arrays_to_mgr(values, columns, index, dtype=dtype)
+
+    elif isinstance(vdtype, ExtensionDtype):
+        # i.e. Datetime64TZ, PeriodDtype; cases with is_1d_only_ea_dtype(vdtype)
+        #  are already caught above
+        values = extract_array(values, extract_numpy=True)
+        if copy:
+            values = values.copy()
+        if values.ndim == 1:
+            values = values.reshape(-1, 1)
+
+    elif isinstance(values, (ABCSeries, Index)):
+        if not copy and (dtype is None or astype_is_view(values.dtype, dtype)):
+            refs = values._references
+
+        if copy:
+            values = values._values.copy()
+        else:
+            values = values._values
+
+        values = _ensure_2d(values)
+
+    elif isinstance(values, (np.ndarray, ExtensionArray)):
+        # drop subclass info
+        if copy and (dtype is None or astype_is_view(values.dtype, dtype)):
+            # only force a copy now if copy=True was requested
+            # and a subsequent `astype` will not already result in a copy
+            values = np.array(values, copy=True, order="F")
+        else:
+            values = np.asarray(values)
+        values = _ensure_2d(values)
+
+    else:
+        # by definition an array here
+        # the dtypes will be coerced to a single dtype
+        values = _prep_ndarraylike(values, copy=copy)
+
+    if dtype is not None and values.dtype != dtype:
+        # GH#40110 see similar check inside sanitize_array
+        values = sanitize_array(
+            values,
+            None,
+            dtype=dtype,
+            copy=copy,
+            allow_2d=True,
+        )
+
+    # _prep_ndarraylike ensures that values.ndim == 2 at this point
+    index, columns = _get_axes(
+        values.shape[0], values.shape[1], index=index, columns=columns
+    )
+
+    _check_values_indices_shape_match(values, index, columns)
+
+    values = values.T
+
+    # if we don't have a dtype specified, then try to convert objects
+    # on the entire block; this is to convert if we have datetimelike's
+    # embedded in an object type
+    if dtype is None and infer_object and is_object_dtype(values.dtype):
+        obj_columns = list(values)
+        maybe_datetime = [maybe_infer_to_datetimelike(x) for x in obj_columns]
+        # don't convert (and copy) the objects if no type inference occurs
+        if any(x is not y for x, y in zip(obj_columns, maybe_datetime)):
+            block_values = [
+                new_block_2d(ensure_block_shape(dval, 2), placement=BlockPlacement(n))
+                for n, dval in enumerate(maybe_datetime)
+            ]
+        else:
+            bp = BlockPlacement(slice(len(columns)))
+            nb = new_block_2d(values, placement=bp, refs=refs)
+            block_values = [nb]
+    elif dtype is None and values.dtype.kind == "U" and using_string_dtype():
+        dtype = StringDtype(na_value=np.nan)
+
+        obj_columns = list(values)
+        block_values = [
+            new_block(
+                dtype.construct_array_type()._from_sequence(data, dtype=dtype),
+                BlockPlacement(slice(i, i + 1)),
+                ndim=2,
+            )
+            for i, data in enumerate(obj_columns)
+        ]
+
+    else:
+        bp = BlockPlacement(slice(len(columns)))
+        nb = new_block_2d(values, placement=bp, refs=refs)
+        block_values = [nb]
+
+    if len(columns) == 0:
+        # TODO: check len(values) == 0?
+        block_values = []
+
+    return create_block_manager_from_blocks(
+        block_values, [columns, index], verify_integrity=False
+    )
+
+
+def _check_values_indices_shape_match(
+    values: np.ndarray, index: Index, columns: Index
+) -> None:
+    """
+    Check that the shape implied by our axes matches the actual shape of the
+    data.
+    """
+    if values.shape[1] != len(columns) or values.shape[0] != len(index):
+        # Could let this raise in Block constructor, but we get a more
+        #  helpful exception message this way.
+        if values.shape[0] == 0 < len(index):
+            raise ValueError("Empty data passed with indices specified.")
+
+        passed = values.shape
+        implied = (len(index), len(columns))
+        raise ValueError(f"Shape of passed values is {passed}, indices imply {implied}")
+
+
+def dict_to_mgr(
+    data: dict,
+    index,
+    columns,
+    *,
+    dtype: DtypeObj | None = None,
+    copy: bool = True,
+) -> Manager:
+    """
+    Segregate Series based on type and coerce into matrices.
+    Needs to handle a lot of exceptional cases.
+
+    Used in DataFrame.__init__
+    """
+    arrays: Sequence[Any]
+
+    if columns is not None:
+        columns = ensure_index(columns)
+        arrays = [np.nan] * len(columns)
+        midxs = set()
+        data_keys = ensure_index(data.keys())  # type: ignore[arg-type]
+        data_values = list(data.values())
+
+        for i, column in enumerate(columns):
+            try:
+                idx = data_keys.get_loc(column)
+            except KeyError:
+                midxs.add(i)
+                continue
+            array = data_values[idx]
+            arrays[i] = array
+            if is_scalar(array) and isna(array):
+                midxs.add(i)
+
+        if index is None:
+            # GH10856
+            # raise ValueError if only scalars in dict
+            if midxs:
+                index = _extract_index(
+                    [array for i, array in enumerate(arrays) if i not in midxs]
+                )
+            else:
+                index = _extract_index(arrays)
+        else:
+            index = ensure_index(index)
+
+        # no obvious "empty" int column
+        if midxs and not is_integer_dtype(dtype):
+            # GH#1783
+            for i in midxs:
+                arr = construct_1d_arraylike_from_scalar(
+                    arrays[i],
+                    len(index),
+                    dtype if dtype is not None else np.dtype("object"),
+                )
+                arrays[i] = arr
+
+    else:
+        keys = maybe_sequence_to_range(list(data.keys()))
+        columns = Index(keys) if keys else default_index(0)
+        arrays = [com.maybe_iterable_to_list(data[k]) for k in keys]
+
+    if copy:
+        # We only need to copy arrays that will not get consolidated, i.e.
+        #  only EA arrays
+        arrays = [
+            x.copy()
+            if isinstance(x, ExtensionArray)
+            else x.copy(deep=True)
+            if (
+                isinstance(x, Index)
+                or (isinstance(x, ABCSeries) and is_1d_only_ea_dtype(x.dtype))
+            )
+            else x
+            for x in arrays
+        ]
+
+    return arrays_to_mgr(arrays, columns, index, dtype=dtype, consolidate=copy)
+
+
+def nested_data_to_arrays(
+    data: Sequence,
+    columns: Index | None,
+    index: Index | None,
+    dtype: DtypeObj | None,
+) -> tuple[list[ArrayLike], Index, Index]:
+    """
+    Convert a single sequence of arrays to multiple arrays.
+    """
+    # By the time we get here we have already checked treat_as_nested(data)
+
+    if is_named_tuple(data[0]) and columns is None:
+        columns = ensure_index(data[0]._fields)
+
+    arrays, columns = to_arrays(data, columns, dtype=dtype)
+    columns = ensure_index(columns)
+
+    if index is None:
+        if isinstance(data[0], ABCSeries):
+            index = _get_names_from_index(data)
+        else:
+            index = default_index(len(data))
+
+    return arrays, columns, index
+
+
+def treat_as_nested(data) -> bool:
+    """
+    Check if we should use nested_data_to_arrays.
+    """
+    return (
+        len(data) > 0
+        and is_list_like(data[0])
+        and getattr(data[0], "ndim", 1) == 1
+        and not (isinstance(data, ExtensionArray) and data.ndim == 2)
+    )
+
+
+# ---------------------------------------------------------------------
+
+
+def _prep_ndarraylike(values, copy: bool = True) -> np.ndarray:
+    # values is specifically _not_ ndarray, EA, Index, or Series
+    # We only get here with `not treat_as_nested(values)`
+
+    if len(values) == 0:
+        # TODO: check for length-zero range, in which case return int64 dtype?
+        # TODO: reuse anything in try_cast?
+        return np.empty((0, 0), dtype=object)
+    elif isinstance(values, range):
+        arr = range_to_ndarray(values)
+        return arr[..., np.newaxis]
+
+    def convert(v):
+        if not is_list_like(v) or isinstance(v, ABCDataFrame):
+            return v
+
+        v = extract_array(v, extract_numpy=True)
+        res = maybe_convert_platform(v)
+        # We don't do maybe_infer_to_datetimelike here bc we will end up doing
+        #  it column-by-column in ndarray_to_mgr
+        return res
+
+    # we could have a 1-dim or 2-dim list here
+    # this is equiv of np.asarray, but does object conversion
+    # and platform dtype preservation
+    # does not convert e.g. [1, "a", True] to ["1", "a", "True"] like
+    #  np.asarray would
+    if is_list_like(values[0]):
+        values = np.array([convert(v) for v in values])
+    elif isinstance(values[0], np.ndarray) and values[0].ndim == 0:
+        # GH#21861 see test_constructor_list_of_lists
+        values = np.array([convert(v) for v in values])
+    else:
+        values = convert(values)
+
+    return _ensure_2d(values)
+
+
+def _ensure_2d(values: np.ndarray) -> np.ndarray:
+    """
+    Reshape 1D values, raise on anything else other than 2D.
+    """
+    if values.ndim == 1:
+        values = values.reshape((values.shape[0], 1))
+    elif values.ndim != 2:
+        raise ValueError(f"Must pass 2-d input. shape={values.shape}")
+    return values
+
+
+def _homogenize(
+    data, index: Index, dtype: DtypeObj | None
+) -> tuple[list[ArrayLike], list[Any]]:
+    oindex = None
+    homogenized = []
+    # if the original array-like in `data` is a Series, keep track of this Series' refs
+    refs: list[Any] = []
+
+    for val in data:
+        if isinstance(val, (ABCSeries, Index)):
+            if dtype is not None:
+                val = val.astype(dtype)
+            if isinstance(val, ABCSeries) and val.index is not index:
+                # Forces alignment. No need to copy data since we
+                # are putting it into an ndarray later
+                val = val.reindex(index)
+            refs.append(val._references)
+            val = val._values
+        else:
+            if isinstance(val, dict):
+                # GH#41785 this _should_ be equivalent to (but faster than)
+                #  val = Series(val, index=index)._values
+                if oindex is None:
+                    oindex = index.astype("O")
+
+                if isinstance(index, (DatetimeIndex, TimedeltaIndex)):
+                    # see test_constructor_dict_datetime64_index
+                    val = dict_compat(val)
+                else:
+                    # see test_constructor_subclass_dict
+                    val = dict(val)
+                val = lib.fast_multiget(val, oindex._values, default=np.nan)
+
+            val = sanitize_array(val, index, dtype=dtype, copy=False)
+            com.require_length_match(val, index)
+            refs.append(None)
+
+        homogenized.append(val)
+
+    return homogenized, refs
+
+
+def _extract_index(data) -> Index:
+    """
+    Try to infer an Index from the passed data, raise ValueError on failure.
+    """
+    index: Index
+    if len(data) == 0:
+        return default_index(0)
+
+    raw_lengths = set()
+    indexes: list[list[Hashable] | Index] = []
+
+    have_raw_arrays = False
+    have_series = False
+    have_dicts = False
+
+    for val in data:
+        if isinstance(val, ABCSeries):
+            have_series = True
+            indexes.append(val.index)
+        elif isinstance(val, dict):
+            have_dicts = True
+            indexes.append(list(val.keys()))
+        elif is_list_like(val) and getattr(val, "ndim", 1) == 1:
+            have_raw_arrays = True
+            raw_lengths.add(len(val))
+        elif isinstance(val, np.ndarray) and val.ndim > 1:
+            raise ValueError("Per-column arrays must each be 1-dimensional")
+
+    if not indexes and not raw_lengths:
+        raise ValueError("If using all scalar values, you must pass an index")
+
+    if have_series:
+        index = union_indexes(indexes)
+    elif have_dicts:
+        index = union_indexes(indexes, sort=False)
+
+    if have_raw_arrays:
+        if len(raw_lengths) > 1:
+            raise ValueError("All arrays must be of the same length")
+
+        if have_dicts:
+            raise ValueError(
+                "Mixing dicts with non-Series may lead to ambiguous ordering."
+            )
+        raw_length = raw_lengths.pop()
+        if have_series:
+            if raw_length != len(index):
+                msg = (
+                    f"array length {raw_length} does not match index "
+                    f"length {len(index)}"
+                )
+                raise ValueError(msg)
+        else:
+            index = default_index(raw_length)
+
+    return ensure_index(index)
+
+
+def reorder_arrays(
+    arrays: list[ArrayLike], arr_columns: Index, columns: Index | None, length: int
+) -> tuple[list[ArrayLike], Index]:
+    """
+    Preemptively (cheaply) reindex arrays with new columns.
+    """
+    # reorder according to the columns
+    if columns is not None:
+        if not columns.equals(arr_columns):
+            # if they are equal, there is nothing to do
+            new_arrays: list[ArrayLike] = []
+            indexer = arr_columns.get_indexer(columns)
+            for i, k in enumerate(indexer):
+                if k == -1:
+                    # by convention default is all-NaN object dtype
+                    arr = np.empty(length, dtype=object)
+                    arr.fill(np.nan)
+                else:
+                    arr = arrays[k]
+                new_arrays.append(arr)
+
+            arrays = new_arrays
+            arr_columns = columns
+
+    return arrays, arr_columns
+
+
+def _get_names_from_index(data) -> Index:
+    has_some_name = any(getattr(s, "name", None) is not None for s in data)
+    if not has_some_name:
+        return default_index(len(data))
+
+    index: list[Hashable] = list(range(len(data)))
+    count = 0
+    for i, s in enumerate(data):
+        n = getattr(s, "name", None)
+        if n is not None:
+            index[i] = n
+        else:
+            index[i] = f"Unnamed {count}"
+            count += 1
+
+    return Index(index)
+
+
+def _get_axes(
+    N: int, K: int, index: Index | None, columns: Index | None
+) -> tuple[Index, Index]:
+    # helper to create the axes as indexes
+    # return axes or defaults
+
+    if index is None:
+        index = default_index(N)
+    else:
+        index = ensure_index(index)
+
+    if columns is None:
+        columns = default_index(K)
+    else:
+        columns = ensure_index(columns)
+    return index, columns
+
+
+def dataclasses_to_dicts(data):
+    """
+    Converts a list of dataclass instances to a list of dictionaries.
+
+    Parameters
+    ----------
+    data : List[Type[dataclass]]
+
+    Returns
+    --------
+    list_dict : List[dict]
+
+    Examples
+    --------
+    >>> from dataclasses import dataclass
+    >>> @dataclass
+    ... class Point:
+    ...     x: int
+    ...     y: int
+
+    >>> dataclasses_to_dicts([Point(1, 2), Point(2, 3)])
+    [{'x': 1, 'y': 2}, {'x': 2, 'y': 3}]
+
+    """
+    from dataclasses import asdict
+
+    return list(map(asdict, data))
+
+
+# ---------------------------------------------------------------------
+# Conversion of Inputs to Arrays
+
+
+def to_arrays(
+    data, columns: Index | None, dtype: DtypeObj | None = None
+) -> tuple[list[ArrayLike], Index]:
+    """
+    Return list of arrays, columns.
+
+    Returns
+    -------
+    list[ArrayLike]
+        These will become columns in a DataFrame.
+    Index
+        This will become frame.columns.
+
+    Notes
+    -----
+    Ensures that len(result_arrays) == len(result_index).
+    """
+
+    if not len(data):
+        if isinstance(data, np.ndarray):
+            if data.dtype.names is not None:
+                # i.e. numpy structured array
+                columns = ensure_index(data.dtype.names)
+                arrays = [data[name] for name in columns]
+
+                if len(data) == 0:
+                    # GH#42456 the indexing above results in list of 2D ndarrays
+                    # TODO: is that an issue with numpy?
+                    for i, arr in enumerate(arrays):
+                        if arr.ndim == 2:
+                            arrays[i] = arr[:, 0]
+
+                return arrays, columns
+        return [], ensure_index([])
+
+    elif isinstance(data, np.ndarray) and data.dtype.names is not None:
+        # e.g. recarray
+        if columns is None:
+            columns = Index(data.dtype.names)
+        arrays = [data[k] for k in columns]
+        return arrays, columns
+
+    if isinstance(data[0], (list, tuple)):
+        arr = _list_to_arrays(data)
+    elif isinstance(data[0], abc.Mapping):
+        arr, columns = _list_of_dict_to_arrays(data, columns)
+    elif isinstance(data[0], ABCSeries):
+        arr, columns = _list_of_series_to_arrays(data, columns)
+    else:
+        # last ditch effort
+        data = [tuple(x) for x in data]
+        arr = _list_to_arrays(data)
+
+    content, columns = _finalize_columns_and_data(arr, columns, dtype)
+    return content, columns
+
+
+def _list_to_arrays(data: list[tuple | list]) -> np.ndarray:
+    # Returned np.ndarray has ndim = 2
+    # Note: we already check len(data) > 0 before getting hre
+    if isinstance(data[0], tuple):
+        content = lib.to_object_array_tuples(data)
+    else:
+        # list of lists
+        content = lib.to_object_array(data)
+    return content
+
+
+def _list_of_series_to_arrays(
+    data: list,
+    columns: Index | None,
+) -> tuple[np.ndarray, Index]:
+    # returned np.ndarray has ndim == 2
+
+    if columns is None:
+        # We know pass_data is non-empty because data[0] is a Series
+        pass_data = [x for x in data if isinstance(x, (ABCSeries, ABCDataFrame))]
+        columns = get_objs_combined_axis(pass_data, sort=False)
+
+    indexer_cache: dict[int, np.ndarray] = {}
+
+    aligned_values = []
+    for s in data:
+        index = getattr(s, "index", None)
+        if index is None:
+            index = default_index(len(s))
+
+        if id(index) in indexer_cache:
+            indexer = indexer_cache[id(index)]
+        else:
+            indexer = indexer_cache[id(index)] = index.get_indexer(columns)
+
+        values = extract_array(s, extract_numpy=True)
+        aligned_values.append(algorithms.take_nd(values, indexer))
+
+    content = np.vstack(aligned_values)
+    return content, columns
+
+
+def _list_of_dict_to_arrays(
+    data: list[dict],
+    columns: Index | None,
+) -> tuple[np.ndarray, Index]:
+    """
+    Convert list of dicts to numpy arrays
+
+    if `columns` is not passed, column names are inferred from the records
+    - for OrderedDict and dicts, the column names match
+      the key insertion-order from the first record to the last.
+    - For other kinds of dict-likes, the keys are lexically sorted.
+
+    Parameters
+    ----------
+    data : iterable
+        collection of records (OrderedDict, dict)
+    columns: iterables or None
+
+    Returns
+    -------
+    content : np.ndarray[object, ndim=2]
+    columns : Index
+    """
+    if columns is None:
+        gen = (list(x.keys()) for x in data)
+        sort = not any(isinstance(d, dict) for d in data)
+        pre_cols = lib.fast_unique_multiple_list_gen(gen, sort=sort)
+        columns = ensure_index(pre_cols)
+
+    # assure that they are of the base dict class and not of derived
+    # classes
+    data = [d if type(d) is dict else dict(d) for d in data]
+
+    content = lib.dicts_to_array(data, list(columns))
+    return content, columns
+
+
+def _finalize_columns_and_data(
+    content: np.ndarray,  # ndim == 2
+    columns: Index | None,
+    dtype: DtypeObj | None,
+) -> tuple[list[ArrayLike], Index]:
+    """
+    Ensure we have valid columns, cast object dtypes if possible.
+    """
+    contents = list(content.T)
+
+    try:
+        columns = _validate_or_indexify_columns(contents, columns)
+    except AssertionError as err:
+        # GH#26429 do not raise user-facing AssertionError
+        raise ValueError(err) from err
+
+    if len(contents) and contents[0].dtype == np.object_:
+        contents = convert_object_array(contents, dtype=dtype)
+
+    return contents, columns
+
+
+def _validate_or_indexify_columns(
+    content: list[np.ndarray], columns: Index | None
+) -> Index:
+    """
+    If columns is None, make numbers as column names; Otherwise, validate that
+    columns have valid length.
+
+    Parameters
+    ----------
+    content : list of np.ndarrays
+    columns : Index or None
+
+    Returns
+    -------
+    Index
+        If columns is None, assign positional column index value as columns.
+
+    Raises
+    ------
+    1. AssertionError when content is not composed of list of lists, and if
+        length of columns is not equal to length of content.
+    2. ValueError when content is list of lists, but length of each sub-list
+        is not equal
+    3. ValueError when content is list of lists, but length of sub-list is
+        not equal to length of content
+    """
+    if columns is None:
+        columns = default_index(len(content))
+    else:
+        # Add mask for data which is composed of list of lists
+        is_mi_list = isinstance(columns, list) and all(
+            isinstance(col, list) for col in columns
+        )
+
+        if not is_mi_list and len(columns) != len(content):  # pragma: no cover
+            # caller's responsibility to check for this...
+            raise AssertionError(
+                f"{len(columns)} columns passed, passed data had {len(content)} columns"
+            )
+        if is_mi_list:
+            # check if nested list column, length of each sub-list should be equal
+            if len({len(col) for col in columns}) > 1:
+                raise ValueError(
+                    "Length of columns passed for MultiIndex columns is different"
+                )
+
+            # if columns is not empty and length of sublist is not equal to content
+            if columns and len(columns[0]) != len(content):
+                raise ValueError(
+                    f"{len(columns[0])} columns passed, passed data had "
+                    f"{len(content)} columns"
+                )
+    return columns
+
+
+def convert_object_array(
+    content: list[npt.NDArray[np.object_]],
+    dtype: DtypeObj | None,
+    dtype_backend: str = "numpy",
+    coerce_float: bool = False,
+) -> list[ArrayLike]:
+    """
+    Internal function to convert object array.
+
+    Parameters
+    ----------
+    content: List[np.ndarray]
+    dtype: np.dtype or ExtensionDtype
+    dtype_backend: Controls if nullable/pyarrow dtypes are returned.
+    coerce_float: Cast floats that are integers to int.
+
+    Returns
+    -------
+    List[ArrayLike]
+    """
+    # provide soft conversion of object dtypes
+
+    def convert(arr):
+        if dtype != np.dtype("O"):
+            arr = lib.maybe_convert_objects(
+                arr,
+                try_float=coerce_float,
+                convert_to_nullable_dtype=dtype_backend != "numpy",
+            )
+            # Notes on cases that get here 2023-02-15
+            # 1) we DO get here when arr is all Timestamps and dtype=None
+            # 2) disabling this doesn't break the world, so this must be
+            #    getting caught at a higher level
+            # 3) passing convert_non_numeric to maybe_convert_objects get this right
+            # 4) convert_non_numeric?
+
+            if dtype is None:
+                if arr.dtype == np.dtype("O"):
+                    # i.e. maybe_convert_objects didn't convert
+                    convert_to_nullable_dtype = dtype_backend != "numpy"
+                    arr = maybe_infer_to_datetimelike(arr, convert_to_nullable_dtype)
+                    if convert_to_nullable_dtype and arr.dtype == np.dtype("O"):
+                        new_dtype = StringDtype()
+                        arr_cls = new_dtype.construct_array_type()
+                        arr = arr_cls._from_sequence(arr, dtype=new_dtype)
+                elif dtype_backend != "numpy" and isinstance(arr, np.ndarray):
+                    if arr.dtype.kind in "iufb":
+                        arr = pd_array(arr, copy=False)
+
+            elif isinstance(dtype, ExtensionDtype):
+                # TODO: test(s) that get here
+                # TODO: try to de-duplicate this convert function with
+                #  core.construction functions
+                cls = dtype.construct_array_type()
+                arr = cls._from_sequence(arr, dtype=dtype, copy=False)
+            elif dtype.kind in "mM":
+                # This restriction is harmless bc these are the only cases
+                #  where maybe_cast_to_datetime is not a no-op.
+                # Here we know:
+                #  1) dtype.kind in "mM" and
+                #  2) arr is either object or numeric dtype
+                arr = maybe_cast_to_datetime(arr, dtype)
+
+        return arr
+
+    arrays = [convert(arr) for arr in content]
+
+    return arrays
diff --git a/pandas/pandas/core/internals/managers.py b/pandas/pandas/core/internals/managers.py
new file mode 100644
index 0000000000000..a3738bb25f56c
--- /dev/null
+++ b/pandas/pandas/core/internals/managers.py
@@ -0,0 +1,2501 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Sequence,
+)
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    NoReturn,
+    cast,
+    final,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config.config import get_option
+
+from pandas._libs import (
+    algos as libalgos,
+    internals as libinternals,
+    lib,
+)
+from pandas._libs.internals import (
+    BlockPlacement,
+    BlockValuesRefs,
+)
+from pandas._libs.tslibs import Timestamp
+from pandas.errors import (
+    AbstractMethodError,
+    PerformanceWarning,
+)
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import validate_bool_kwarg
+
+from pandas.core.dtypes.cast import (
+    find_common_type,
+    infer_dtype_from_scalar,
+    np_can_hold_element,
+)
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    is_1d_only_ea_dtype,
+    is_list_like,
+)
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    ExtensionDtype,
+    SparseDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    array_equals,
+    isna,
+)
+
+import pandas.core.algorithms as algos
+from pandas.core.arrays import DatetimeArray
+from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+from pandas.core.base import PandasObject
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexers import maybe_convert_indices
+from pandas.core.indexes.api import (
+    Index,
+    default_index,
+    ensure_index,
+)
+from pandas.core.internals.blocks import (
+    Block,
+    NumpyBlock,
+    ensure_block_shape,
+    extend_blocks,
+    get_block_type,
+    maybe_coerce_values,
+    new_block,
+    new_block_2d,
+)
+from pandas.core.internals.ops import (
+    blockwise_all,
+    operate_blockwise,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        DtypeObj,
+        QuantileInterpolation,
+        Self,
+        Shape,
+        npt,
+    )
+
+    from pandas.api.extensions import ExtensionArray
+
+
+def interleaved_dtype(dtypes: list[DtypeObj]) -> DtypeObj | None:
+    """
+    Find the common dtype for `blocks`.
+
+    Parameters
+    ----------
+    blocks : List[DtypeObj]
+
+    Returns
+    -------
+    dtype : np.dtype, ExtensionDtype, or None
+        None is returned when `blocks` is empty.
+    """
+    if not len(dtypes):
+        return None
+
+    return find_common_type(dtypes)
+
+
+def ensure_np_dtype(dtype: DtypeObj) -> np.dtype:
+    # TODO: https://github.com/pandas-dev/pandas/issues/22791
+    # Give EAs some input on what happens here. Sparse needs this.
+    if isinstance(dtype, SparseDtype):
+        dtype = dtype.subtype
+        dtype = cast(np.dtype, dtype)
+    elif isinstance(dtype, ExtensionDtype):
+        dtype = np.dtype("object")
+    elif dtype == np.dtype(str):
+        dtype = np.dtype("object")
+    return dtype
+
+
+class BaseBlockManager(PandasObject):
+    """
+    Core internal data structure to implement DataFrame, Series, etc.
+
+    Manage a bunch of labeled 2D mixed-type ndarrays. Essentially it's a
+    lightweight blocked set of labeled data to be manipulated by the DataFrame
+    public API class
+
+    Attributes
+    ----------
+    shape
+    ndim
+    axes
+    values
+    items
+
+    Methods
+    -------
+    set_axis(axis, new_labels)
+    copy(deep=True)
+
+    get_dtypes
+
+    apply(func, axes, block_filter_fn)
+
+    get_bool_data
+    get_numeric_data
+
+    get_slice(slice_like, axis)
+    get(label)
+    iget(loc)
+
+    take(indexer, axis)
+    reindex_axis(new_labels, axis)
+    reindex_indexer(new_labels, indexer, axis)
+
+    delete(label)
+    insert(loc, label, value)
+    set(label, value)
+
+    Parameters
+    ----------
+    blocks: Sequence of Block
+    axes: Sequence of Index
+    verify_integrity: bool, default True
+
+    Notes
+    -----
+    This is *not* a public API class
+    """
+
+    __slots__ = ()
+
+    _blknos: npt.NDArray[np.intp]
+    _blklocs: npt.NDArray[np.intp]
+    blocks: tuple[Block, ...]
+    axes: list[Index]
+
+    @property
+    def ndim(self) -> int:
+        raise NotImplementedError
+
+    _known_consolidated: bool
+    _is_consolidated: bool
+
+    def __init__(self, blocks, axes, verify_integrity: bool = True) -> None:
+        raise NotImplementedError
+
+    @final
+    def __len__(self) -> int:
+        return len(self.items)
+
+    @property
+    def shape(self) -> Shape:
+        return tuple(len(ax) for ax in self.axes)
+
+    @classmethod
+    def from_blocks(cls, blocks: list[Block], axes: list[Index]) -> Self:
+        raise NotImplementedError
+
+    @property
+    def blknos(self) -> npt.NDArray[np.intp]:
+        """
+        Suppose we want to find the array corresponding to our i'th column.
+
+        blknos[i] identifies the block from self.blocks that contains this column.
+
+        blklocs[i] identifies the column of interest within
+        self.blocks[self.blknos[i]]
+        """
+        if self._blknos is None:
+            # Note: these can be altered by other BlockManager methods.
+            self._rebuild_blknos_and_blklocs()
+
+        return self._blknos
+
+    @property
+    def blklocs(self) -> npt.NDArray[np.intp]:
+        """
+        See blknos.__doc__
+        """
+        if self._blklocs is None:
+            # Note: these can be altered by other BlockManager methods.
+            self._rebuild_blknos_and_blklocs()
+
+        return self._blklocs
+
+    def make_empty(self, axes=None) -> Self:
+        """return an empty BlockManager with the items axis of len 0"""
+        if axes is None:
+            axes = [default_index(0)] + self.axes[1:]
+
+        # preserve dtype if possible
+        if self.ndim == 1:
+            assert isinstance(self, SingleBlockManager)  # for mypy
+            blk = self.blocks[0]
+            arr = blk.values[:0]
+            bp = BlockPlacement(slice(0, 0))
+            nb = blk.make_block_same_class(arr, placement=bp)
+            blocks = [nb]
+        else:
+            blocks = []
+        return type(self).from_blocks(blocks, axes)
+
+    def __bool__(self) -> bool:
+        return True
+
+    def set_axis(self, axis: AxisInt, new_labels: Index) -> None:
+        # Caller is responsible for ensuring we have an Index object.
+        self._validate_set_axis(axis, new_labels)
+        self.axes[axis] = new_labels
+
+    @final
+    def _validate_set_axis(self, axis: AxisInt, new_labels: Index) -> None:
+        # Caller is responsible for ensuring we have an Index object.
+        old_len = len(self.axes[axis])
+        new_len = len(new_labels)
+
+        if axis == 1 and len(self.items) == 0:
+            # If we are setting the index on a DataFrame with no columns,
+            #  it is OK to change the length.
+            pass
+
+        elif new_len != old_len:
+            raise ValueError(
+                f"Length mismatch: Expected axis has {old_len} elements, new "
+                f"values have {new_len} elements"
+            )
+
+    @property
+    def is_single_block(self) -> bool:
+        # Assumes we are 2D; overridden by SingleBlockManager
+        return len(self.blocks) == 1
+
+    @property
+    def items(self) -> Index:
+        return self.axes[0]
+
+    def _has_no_reference(self, i: int) -> bool:
+        """
+        Check for column `i` if it has references.
+        (whether it references another array or is itself being referenced)
+        Returns True if the column has no references.
+        """
+        blkno = self.blknos[i]
+        return self._has_no_reference_block(blkno)
+
+    def _has_no_reference_block(self, blkno: int) -> bool:
+        """
+        Check for block `i` if it has references.
+        (whether it references another array or is itself being referenced)
+        Returns True if the block has no references.
+        """
+        return not self.blocks[blkno].refs.has_reference()
+
+    def add_references(self, mgr: BaseBlockManager) -> None:
+        """
+        Adds the references from one manager to another. We assume that both
+        managers have the same block structure.
+        """
+        if len(self.blocks) != len(mgr.blocks):
+            # If block structure changes, then we made a copy
+            return
+        for i, blk in enumerate(self.blocks):
+            blk.refs = mgr.blocks[i].refs
+            blk.refs.add_reference(blk)
+
+    def references_same_values(self, mgr: BaseBlockManager, blkno: int) -> bool:
+        """
+        Checks if two blocks from two different block managers reference the
+        same underlying values.
+        """
+        blk = self.blocks[blkno]
+        return any(blk is ref() for ref in mgr.blocks[blkno].refs.referenced_blocks)
+
+    def get_dtypes(self) -> npt.NDArray[np.object_]:
+        dtypes = np.array([blk.dtype for blk in self.blocks], dtype=object)
+        return dtypes.take(self.blknos)
+
+    @property
+    def arrays(self) -> list[ArrayLike]:
+        """
+        Quick access to the backing arrays of the Blocks.
+
+        Only for compatibility with ArrayManager for testing convenience.
+        Not to be used in actual code, and return value is not the same as the
+        ArrayManager method (list of 1D arrays vs iterator of 2D ndarrays / 1D EAs).
+
+        Warning! The returned arrays don't handle Copy-on-Write, so this should
+        be used with caution (only in read-mode).
+        """
+        # TODO: Deprecate, usage in Dask
+        # https://github.com/dask/dask/blob/484fc3f1136827308db133cd256ba74df7a38d8c/dask/base.py#L1312
+        return [blk.values for blk in self.blocks]
+
+    def __repr__(self) -> str:
+        output = type(self).__name__
+        for i, ax in enumerate(self.axes):
+            if i == 0:
+                output += f"\nItems: {ax}"
+            else:
+                output += f"\nAxis {i}: {ax}"
+
+        for block in self.blocks:
+            output += f"\n{block}"
+        return output
+
+    def _equal_values(self, other: Self) -> bool:
+        """
+        To be implemented by the subclasses. Only check the column values
+        assuming shape and indexes have already been checked.
+        """
+        raise AbstractMethodError(self)
+
+    @final
+    def equals(self, other: object) -> bool:
+        """
+        Implementation for DataFrame.equals
+        """
+        if not isinstance(other, type(self)):
+            return False
+
+        self_axes, other_axes = self.axes, other.axes
+        if len(self_axes) != len(other_axes):
+            return False
+        if not all(ax1.equals(ax2) for ax1, ax2 in zip(self_axes, other_axes)):
+            return False
+
+        return self._equal_values(other)
+
+    def apply(
+        self,
+        f,
+        align_keys: list[str] | None = None,
+        **kwargs,
+    ) -> Self:
+        """
+        Iterate over the blocks, collect and create a new BlockManager.
+
+        Parameters
+        ----------
+        f : str or callable
+            Name of the Block method to apply.
+        align_keys: List[str] or None, default None
+        **kwargs
+            Keywords to pass to `f`
+
+        Returns
+        -------
+        BlockManager
+        """
+        assert "filter" not in kwargs
+
+        align_keys = align_keys or []
+        result_blocks: list[Block] = []
+        # fillna: Series/DataFrame is responsible for making sure value is aligned
+
+        aligned_args = {k: kwargs[k] for k in align_keys}
+
+        for b in self.blocks:
+            if aligned_args:
+                for k, obj in aligned_args.items():
+                    if isinstance(obj, (ABCSeries, ABCDataFrame)):
+                        # The caller is responsible for ensuring that
+                        #  obj.axes[-1].equals(self.items)
+                        if obj.ndim == 1:
+                            kwargs[k] = obj.iloc[b.mgr_locs.indexer]._values
+                        else:
+                            kwargs[k] = obj.iloc[:, b.mgr_locs.indexer]._values
+                    else:
+                        # otherwise we have an ndarray
+                        kwargs[k] = obj[b.mgr_locs.indexer]
+
+            if callable(f):
+                applied = b.apply(f, **kwargs)
+            else:
+                applied = getattr(b, f)(**kwargs)
+            result_blocks = extend_blocks(applied, result_blocks)
+
+        out = type(self).from_blocks(result_blocks, self.axes)
+        return out
+
+    @final
+    def isna(self, func) -> Self:
+        return self.apply("apply", func=func)
+
+    @final
+    def fillna(self, value, limit: int | None, inplace: bool) -> Self:
+        if limit is not None:
+            # Do this validation even if we go through one of the no-op paths
+            limit = libalgos.validate_limit(None, limit=limit)
+
+        return self.apply(
+            "fillna",
+            value=value,
+            limit=limit,
+            inplace=inplace,
+        )
+
+    @final
+    def where(self, other, cond, align: bool) -> Self:
+        if align:
+            align_keys = ["other", "cond"]
+        else:
+            align_keys = ["cond"]
+            other = extract_array(other, extract_numpy=True)
+
+        return self.apply(
+            "where",
+            align_keys=align_keys,
+            other=other,
+            cond=cond,
+        )
+
+    @final
+    def putmask(self, mask, new, align: bool = True) -> Self:
+        if align:
+            align_keys = ["new", "mask"]
+        else:
+            align_keys = ["mask"]
+            new = extract_array(new, extract_numpy=True)
+
+        return self.apply(
+            "putmask",
+            align_keys=align_keys,
+            mask=mask,
+            new=new,
+        )
+
+    @final
+    def round(self, decimals: int) -> Self:
+        return self.apply("round", decimals=decimals)
+
+    @final
+    def replace(self, to_replace, value, inplace: bool) -> Self:
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        # NDFrame.replace ensures the not-is_list_likes here
+        assert not lib.is_list_like(to_replace)
+        assert not lib.is_list_like(value)
+        return self.apply(
+            "replace",
+            to_replace=to_replace,
+            value=value,
+            inplace=inplace,
+        )
+
+    @final
+    def replace_regex(self, **kwargs) -> Self:
+        return self.apply("_replace_regex", **kwargs)
+
+    @final
+    def replace_list(
+        self,
+        src_list: list[Any],
+        dest_list: list[Any],
+        inplace: bool = False,
+        regex: bool = False,
+    ) -> Self:
+        """do a list replace"""
+        inplace = validate_bool_kwarg(inplace, "inplace")
+
+        bm = self.apply(
+            "replace_list",
+            src_list=src_list,
+            dest_list=dest_list,
+            inplace=inplace,
+            regex=regex,
+        )
+        bm._consolidate_inplace()
+        return bm
+
+    def interpolate(self, inplace: bool, **kwargs) -> Self:
+        return self.apply("interpolate", inplace=inplace, **kwargs)
+
+    def pad_or_backfill(self, inplace: bool, **kwargs) -> Self:
+        return self.apply("pad_or_backfill", inplace=inplace, **kwargs)
+
+    def shift(self, periods: int, fill_value) -> Self:
+        if fill_value is lib.no_default:
+            fill_value = None
+
+        return self.apply("shift", periods=periods, fill_value=fill_value)
+
+    def setitem(self, indexer, value) -> Self:
+        """
+        Set values with indexer.
+
+        For SingleBlockManager, this backs s[indexer] = value
+        """
+        if isinstance(indexer, np.ndarray) and indexer.ndim > self.ndim:
+            raise ValueError(f"Cannot set values with ndim > {self.ndim}")
+
+        if not self._has_no_reference(0):
+            # this method is only called if there is a single block -> hardcoded 0
+            # Split blocks to only copy the columns we want to modify
+            if self.ndim == 2 and isinstance(indexer, tuple):
+                blk_loc = self.blklocs[indexer[1]]
+                if is_list_like(blk_loc) and blk_loc.ndim == 2:
+                    blk_loc = np.squeeze(blk_loc, axis=0)
+                elif not is_list_like(blk_loc):
+                    # Keep dimension and copy data later
+                    blk_loc = [blk_loc]  # type: ignore[assignment]
+                if len(blk_loc) == 0:
+                    return self.copy(deep=False)
+
+                values = self.blocks[0].values
+                if values.ndim == 2:
+                    values = values[blk_loc]
+                    # "T" has no attribute "_iset_split_block"
+                    self._iset_split_block(  # type: ignore[attr-defined]
+                        0, blk_loc, values
+                    )
+                    # first block equals values
+                    self.blocks[0].setitem((indexer[0], np.arange(len(blk_loc))), value)
+                    return self
+            # No need to split if we either set all columns or on a single block
+            # manager
+            self = self.copy()
+
+        return self.apply("setitem", indexer=indexer, value=value)
+
+    def diff(self, n: int) -> Self:
+        # only reached with self.ndim == 2
+        return self.apply("diff", n=n)
+
+    def astype(self, dtype, errors: str = "raise") -> Self:
+        return self.apply("astype", dtype=dtype, errors=errors)
+
+    def convert(self) -> Self:
+        return self.apply("convert")
+
+    def convert_dtypes(self, **kwargs):
+        return self.apply("convert_dtypes", **kwargs)
+
+    def get_values_for_csv(
+        self, *, float_format, date_format, decimal, na_rep: str = "nan", quoting=None
+    ) -> Self:
+        """
+        Convert values to native types (strings / python objects) that are used
+        in formatting (repr / csv).
+        """
+        return self.apply(
+            "get_values_for_csv",
+            na_rep=na_rep,
+            quoting=quoting,
+            float_format=float_format,
+            date_format=date_format,
+            decimal=decimal,
+        )
+
+    @property
+    def any_extension_types(self) -> bool:
+        """Whether any of the blocks in this manager are extension blocks"""
+        return any(block.is_extension for block in self.blocks)
+
+    @property
+    def is_view(self) -> bool:
+        """return a boolean if we are a single block and are a view"""
+        if len(self.blocks) == 1:
+            return self.blocks[0].is_view
+
+        # It is technically possible to figure out which blocks are views
+        # e.g. [ b.values.base is not None for b in self.blocks ]
+        # but then we have the case of possibly some blocks being a view
+        # and some blocks not. setting in theory is possible on the non-view
+        # blocks. But this is a bit
+        # complicated
+
+        return False
+
+    def _get_data_subset(self, predicate: Callable) -> Self:
+        blocks = [blk for blk in self.blocks if predicate(blk.values)]
+        return self._combine(blocks)
+
+    def get_bool_data(self) -> Self:
+        """
+        Select blocks that are bool-dtype and columns from object-dtype blocks
+        that are all-bool.
+        """
+
+        new_blocks = []
+
+        for blk in self.blocks:
+            if blk.dtype == bool:
+                new_blocks.append(blk)
+
+            elif blk.is_object:
+                new_blocks.extend(nb for nb in blk._split() if nb.is_bool)
+
+        return self._combine(new_blocks)
+
+    def get_numeric_data(self) -> Self:
+        numeric_blocks = [blk for blk in self.blocks if blk.is_numeric]
+        if len(numeric_blocks) == len(self.blocks):
+            # Avoid somewhat expensive _combine
+            return self
+        return self._combine(numeric_blocks)
+
+    def _combine(self, blocks: list[Block], index: Index | None = None) -> Self:
+        """return a new manager with the blocks"""
+        if len(blocks) == 0:
+            if self.ndim == 2:
+                # retain our own Index dtype
+                if index is not None:
+                    axes = [self.items[:0], index]
+                else:
+                    axes = [self.items[:0]] + self.axes[1:]
+                return self.make_empty(axes)
+            return self.make_empty()
+
+        # FIXME: optimization potential
+        indexer = np.sort(np.concatenate([b.mgr_locs.as_array for b in blocks]))
+        inv_indexer = lib.get_reverse_indexer(indexer, self.shape[0])
+
+        new_blocks: list[Block] = []
+        for b in blocks:
+            nb = b.copy(deep=False)
+            nb.mgr_locs = BlockPlacement(inv_indexer[nb.mgr_locs.indexer])
+            new_blocks.append(nb)
+
+        axes = list(self.axes)
+        if index is not None:
+            axes[-1] = index
+        axes[0] = self.items.take(indexer)
+
+        return type(self).from_blocks(new_blocks, axes)
+
+    @property
+    def nblocks(self) -> int:
+        return len(self.blocks)
+
+    def copy(self, deep: bool | Literal["all"] = True) -> Self:
+        """
+        Make deep or shallow copy of BlockManager
+
+        Parameters
+        ----------
+        deep : bool, string or None, default True
+            If False or None, return a shallow copy (do not copy data)
+            If 'all', copy data and a deep copy of the index
+
+        Returns
+        -------
+        BlockManager
+        """
+        # this preserves the notion of view copying of axes
+        if deep:
+            # hit in e.g. tests.io.json.test_pandas
+
+            def copy_func(ax):
+                return ax.copy(deep=True) if deep == "all" else ax.view()
+
+            new_axes = [copy_func(ax) for ax in self.axes]
+        else:
+            new_axes = [ax.view() for ax in self.axes]
+
+        res = self.apply("copy", deep=deep)
+        res.axes = new_axes
+
+        if self.ndim > 1:
+            # Avoid needing to re-compute these
+            blknos = self._blknos
+            if blknos is not None:
+                res._blknos = blknos.copy()
+                res._blklocs = self._blklocs.copy()
+
+        if deep:
+            res._consolidate_inplace()
+        return res
+
+    def is_consolidated(self) -> bool:
+        return True
+
+    def consolidate(self) -> Self:
+        """
+        Join together blocks having same dtype
+
+        Returns
+        -------
+        y : BlockManager
+        """
+        if self.is_consolidated():
+            return self
+
+        bm = type(self)(self.blocks, self.axes, verify_integrity=False)
+        bm._is_consolidated = False
+        bm._consolidate_inplace()
+        return bm
+
+    def _consolidate_inplace(self) -> None:
+        return
+
+    @final
+    def reindex_axis(
+        self,
+        new_index: Index,
+        axis: AxisInt,
+        fill_value=None,
+        only_slice: bool = False,
+    ) -> Self:
+        """
+        Conform data manager to new index.
+        """
+        new_index, indexer = self.axes[axis].reindex(new_index)
+
+        return self.reindex_indexer(
+            new_index,
+            indexer,
+            axis=axis,
+            fill_value=fill_value,
+            only_slice=only_slice,
+        )
+
+    def reindex_indexer(
+        self,
+        new_axis: Index,
+        indexer: npt.NDArray[np.intp] | None,
+        axis: AxisInt,
+        fill_value=None,
+        allow_dups: bool = False,
+        only_slice: bool = False,
+        *,
+        use_na_proxy: bool = False,
+    ) -> Self:
+        """
+        Parameters
+        ----------
+        new_axis : Index
+        indexer : ndarray[intp] or None
+        axis : int
+        fill_value : object, default None
+        allow_dups : bool, default False
+        only_slice : bool, default False
+            Whether to take views, not copies, along columns.
+        use_na_proxy : bool, default False
+            Whether to use a np.void ndarray for newly introduced columns.
+
+        pandas-indexer with -1's only.
+        """
+        if indexer is None:
+            if new_axis is self.axes[axis]:
+                return self
+
+            result = self.copy(deep=False)
+            result.axes = list(self.axes)
+            result.axes[axis] = new_axis
+            return result
+
+        # Should be intp, but in some cases we get int64 on 32bit builds
+        assert isinstance(indexer, np.ndarray)
+
+        # some axes don't allow reindexing with dups
+        if not allow_dups:
+            self.axes[axis]._validate_can_reindex(indexer)
+
+        if axis >= self.ndim:
+            raise IndexError("Requested axis not found in manager")
+
+        if axis == 0:
+            new_blocks = list(
+                self._slice_take_blocks_ax0(
+                    indexer,
+                    fill_value=fill_value,
+                    only_slice=only_slice,
+                    use_na_proxy=use_na_proxy,
+                )
+            )
+        else:
+            new_blocks = [
+                blk.take_nd(
+                    indexer,
+                    axis=1,
+                    fill_value=(
+                        fill_value if fill_value is not None else blk.fill_value
+                    ),
+                )
+                for blk in self.blocks
+            ]
+
+        new_axes = list(self.axes)
+        new_axes[axis] = new_axis
+
+        new_mgr = type(self).from_blocks(new_blocks, new_axes)
+        if axis == 1:
+            # We can avoid the need to rebuild these
+            new_mgr._blknos = self.blknos.copy()
+            new_mgr._blklocs = self.blklocs.copy()
+        return new_mgr
+
+    def _slice_take_blocks_ax0(
+        self,
+        slice_or_indexer: slice | np.ndarray,
+        fill_value=lib.no_default,
+        only_slice: bool = False,
+        *,
+        use_na_proxy: bool = False,
+        ref_inplace_op: bool = False,
+    ) -> Generator[Block]:
+        """
+        Slice/take blocks along axis=0.
+
+        Overloaded for SingleBlock
+
+        Parameters
+        ----------
+        slice_or_indexer : slice or np.ndarray[int64]
+        fill_value : scalar, default lib.no_default
+        only_slice : bool, default False
+            If True, we always return views on existing arrays, never copies.
+            This is used when called from ops.blockwise.operate_blockwise.
+        use_na_proxy : bool, default False
+            Whether to use a np.void ndarray for newly introduced columns.
+        ref_inplace_op: bool, default False
+            Don't track refs if True because we operate inplace
+
+        Yields
+        ------
+        Block : New Block
+        """
+        allow_fill = fill_value is not lib.no_default
+
+        sl_type, slobj, sllen = _preprocess_slice_or_indexer(
+            slice_or_indexer, self.shape[0], allow_fill=allow_fill
+        )
+
+        if self.is_single_block:
+            blk = self.blocks[0]
+
+            if sl_type == "slice":
+                # GH#32959 EABlock would fail since we can't make 0-width
+                # TODO(EA2D): special casing unnecessary with 2D EAs
+                if sllen == 0:
+                    return
+                bp = BlockPlacement(slice(0, sllen))
+                yield blk.getitem_block_columns(slobj, new_mgr_locs=bp)
+                return
+            elif not allow_fill or self.ndim == 1:
+                if allow_fill and fill_value is None:
+                    fill_value = blk.fill_value
+
+                if not allow_fill and only_slice:
+                    # GH#33597 slice instead of take, so we get
+                    #  views instead of copies
+                    for i, ml in enumerate(slobj):
+                        yield blk.getitem_block_columns(
+                            slice(ml, ml + 1),
+                            new_mgr_locs=BlockPlacement(i),
+                            ref_inplace_op=ref_inplace_op,
+                        )
+                else:
+                    bp = BlockPlacement(slice(0, sllen))
+                    yield blk.take_nd(
+                        slobj,
+                        axis=0,
+                        new_mgr_locs=bp,
+                        fill_value=fill_value,
+                    )
+                return
+
+        if sl_type == "slice":
+            blknos = self.blknos[slobj]
+            blklocs = self.blklocs[slobj]
+        else:
+            blknos = algos.take_nd(
+                self.blknos, slobj, fill_value=-1, allow_fill=allow_fill
+            )
+            blklocs = algos.take_nd(
+                self.blklocs, slobj, fill_value=-1, allow_fill=allow_fill
+            )
+
+        # When filling blknos, make sure blknos is updated before appending to
+        # blocks list, that way new blkno is exactly len(blocks).
+        group = not only_slice
+        for blkno, mgr_locs in libinternals.get_blkno_placements(blknos, group=group):
+            if blkno == -1:
+                # If we've got here, fill_value was not lib.no_default
+
+                yield self._make_na_block(
+                    placement=mgr_locs,
+                    fill_value=fill_value,
+                    use_na_proxy=use_na_proxy,
+                )
+            else:
+                blk = self.blocks[blkno]
+
+                # Otherwise, slicing along items axis is necessary.
+                if not blk._can_consolidate and not blk._validate_ndim:
+                    # i.e. we dont go through here for DatetimeTZBlock
+                    # A non-consolidatable block, it's easy, because there's
+                    # only one item and each mgr loc is a copy of that single
+                    # item.
+                    deep = False
+                    for mgr_loc in mgr_locs:
+                        newblk = blk.copy(deep=deep)
+                        newblk.mgr_locs = BlockPlacement(slice(mgr_loc, mgr_loc + 1))
+                        yield newblk
+
+                else:
+                    # GH#32779 to avoid the performance penalty of copying,
+                    #  we may try to only slice
+                    taker = blklocs[mgr_locs.indexer]
+                    max_len = max(len(mgr_locs), taker.max() + 1)
+                    taker = lib.maybe_indices_to_slice(taker, max_len)
+
+                    if isinstance(taker, slice):
+                        nb = blk.getitem_block_columns(taker, new_mgr_locs=mgr_locs)
+                        yield nb
+                    elif only_slice:
+                        # GH#33597 slice instead of take, so we get
+                        #  views instead of copies
+                        for i, ml in zip(taker, mgr_locs):
+                            slc = slice(i, i + 1)
+                            bp = BlockPlacement(ml)
+                            nb = blk.getitem_block_columns(slc, new_mgr_locs=bp)
+                            # We have np.shares_memory(nb.values, blk.values)
+                            yield nb
+                    else:
+                        nb = blk.take_nd(taker, axis=0, new_mgr_locs=mgr_locs)
+                        yield nb
+
+    def _make_na_block(
+        self, placement: BlockPlacement, fill_value=None, use_na_proxy: bool = False
+    ) -> Block:
+        # Note: we only get here with self.ndim == 2
+
+        if use_na_proxy:
+            assert fill_value is None
+            shape = (len(placement), self.shape[1])
+            vals = np.empty(shape, dtype=np.void)
+            nb = NumpyBlock(vals, placement, ndim=2)
+            return nb
+
+        if fill_value is None or fill_value is np.nan:
+            fill_value = np.nan
+            # GH45857 avoid unnecessary upcasting
+            dtype = interleaved_dtype([blk.dtype for blk in self.blocks])
+            if dtype is not None and np.issubdtype(dtype.type, np.floating):
+                fill_value = dtype.type(fill_value)
+
+        shape = (len(placement), self.shape[1])
+
+        dtype, fill_value = infer_dtype_from_scalar(fill_value)
+        block_values = make_na_array(dtype, shape, fill_value)
+        return new_block_2d(block_values, placement=placement)
+
+    def take(
+        self,
+        indexer: npt.NDArray[np.intp],
+        axis: AxisInt = 1,
+        verify: bool = True,
+    ) -> Self:
+        """
+        Take items along any axis.
+
+        indexer : np.ndarray[np.intp]
+        axis : int, default 1
+        verify : bool, default True
+            Check that all entries are between 0 and len(self) - 1, inclusive.
+            Pass verify=False if this check has been done by the caller.
+
+        Returns
+        -------
+        BlockManager
+        """
+        # Caller is responsible for ensuring indexer annotation is accurate
+
+        n = self.shape[axis]
+        indexer = maybe_convert_indices(indexer, n, verify=verify)
+
+        new_labels = self.axes[axis].take(indexer)
+        return self.reindex_indexer(
+            new_axis=new_labels,
+            indexer=indexer,
+            axis=axis,
+            allow_dups=True,
+        )
+
+
+class BlockManager(libinternals.BlockManager, BaseBlockManager):
+    """
+    BaseBlockManager that holds 2D blocks.
+    """
+
+    ndim = 2
+
+    # ----------------------------------------------------------------
+    # Constructors
+
+    def __init__(
+        self,
+        blocks: Sequence[Block],
+        axes: Sequence[Index],
+        verify_integrity: bool = True,
+    ) -> None:
+        if verify_integrity:
+            # Assertion disabled for performance
+            # assert all(isinstance(x, Index) for x in axes)
+
+            for block in blocks:
+                if self.ndim != block.ndim:
+                    raise AssertionError(
+                        f"Number of Block dimensions ({block.ndim}) must equal "
+                        f"number of axes ({self.ndim})"
+                    )
+                # As of 2.0, the caller is responsible for ensuring that
+                #  DatetimeTZBlock with block.ndim == 2 has block.values.ndim ==2;
+                #  previously there was a special check for fastparquet compat.
+
+            self._verify_integrity()
+
+    def _verify_integrity(self) -> None:
+        mgr_shape = self.shape
+        tot_items = sum(len(x.mgr_locs) for x in self.blocks)
+        for block in self.blocks:
+            if block.shape[1:] != mgr_shape[1:]:
+                raise_construction_error(tot_items, block.shape[1:], self.axes)
+        if len(self.items) != tot_items:
+            raise AssertionError(
+                "Number of manager items must equal union of "
+                f"block items\n# manager items: {len(self.items)}, # "
+                f"tot_items: {tot_items}"
+            )
+
+    @classmethod
+    def from_blocks(cls, blocks: list[Block], axes: list[Index]) -> Self:
+        """
+        Constructor for BlockManager and SingleBlockManager with same signature.
+        """
+        return cls(blocks, axes, verify_integrity=False)
+
+    # ----------------------------------------------------------------
+    # Indexing
+
+    def fast_xs(self, loc: int) -> SingleBlockManager:
+        """
+        Return the array corresponding to `frame.iloc[loc]`.
+
+        Parameters
+        ----------
+        loc : int
+
+        Returns
+        -------
+        np.ndarray or ExtensionArray
+        """
+        if len(self.blocks) == 1:
+            # TODO: this could be wrong if blk.mgr_locs is not slice(None)-like;
+            #  is this ruled out in the general case?
+            result: np.ndarray | ExtensionArray = self.blocks[0].iget(
+                (slice(None), loc)
+            )
+            # in the case of a single block, the new block is a view
+            bp = BlockPlacement(slice(0, len(result)))
+            block = new_block(
+                result,
+                placement=bp,
+                ndim=1,
+                refs=self.blocks[0].refs,
+            )
+            return SingleBlockManager(block, self.axes[0])
+
+        dtype = interleaved_dtype([blk.dtype for blk in self.blocks])
+
+        n = len(self)
+
+        if isinstance(dtype, ExtensionDtype):
+            # TODO: use object dtype as workaround for non-performant
+            #  EA.__setitem__ methods. (primarily ArrowExtensionArray.__setitem__
+            #  when iteratively setting individual values)
+            #  https://github.com/pandas-dev/pandas/pull/54508#issuecomment-1675827918
+            result = np.empty(n, dtype=object)
+        else:
+            result = np.empty(n, dtype=dtype)
+            result = ensure_wrapped_if_datetimelike(result)
+
+        for blk in self.blocks:
+            # Such assignment may incorrectly coerce NaT to None
+            # result[blk.mgr_locs] = blk._slice((slice(None), loc))
+            for i, rl in enumerate(blk.mgr_locs):
+                result[rl] = blk.iget((i, loc))
+
+        if isinstance(dtype, ExtensionDtype):
+            cls = dtype.construct_array_type()
+            result = cls._from_sequence(result, dtype=dtype)
+
+        bp = BlockPlacement(slice(0, len(result)))
+        block = new_block(result, placement=bp, ndim=1)
+        return SingleBlockManager(block, self.axes[0])
+
+    def iget(self, i: int, track_ref: bool = True) -> SingleBlockManager:
+        """
+        Return the data as a SingleBlockManager.
+        """
+        block = self.blocks[self.blknos[i]]
+        values = block.iget(self.blklocs[i])
+
+        # shortcut for select a single-dim from a 2-dim BM
+        bp = BlockPlacement(slice(0, len(values)))
+        nb = type(block)(
+            values, placement=bp, ndim=1, refs=block.refs if track_ref else None
+        )
+        return SingleBlockManager(nb, self.axes[1])
+
+    def iget_values(self, i: int) -> ArrayLike:
+        """
+        Return the data for column i as the values (ndarray or ExtensionArray).
+
+        Warning! The returned array is a view but doesn't handle Copy-on-Write,
+        so this should be used with caution.
+        """
+        # TODO(CoW) making the arrays read-only might make this safer to use?
+        block = self.blocks[self.blknos[i]]
+        values = block.iget(self.blklocs[i])
+        return values
+
+    @property
+    def column_arrays(self) -> list[np.ndarray]:
+        """
+        Used in the JSON C code to access column arrays.
+        This optimizes compared to using `iget_values` by converting each
+
+        Warning! This doesn't handle Copy-on-Write, so should be used with
+        caution (current use case of consuming this in the JSON code is fine).
+        """
+        # This is an optimized equivalent to
+        #  result = [self.iget_values(i) for i in range(len(self.items))]
+        result: list[np.ndarray | None] = [None] * len(self.items)
+
+        for blk in self.blocks:
+            mgr_locs = blk._mgr_locs
+            values = blk.array_values._values_for_json()
+            if values.ndim == 1:
+                # TODO(EA2D): special casing not needed with 2D EAs
+                result[mgr_locs[0]] = values
+
+            else:
+                for i, loc in enumerate(mgr_locs):
+                    result[loc] = values[i]
+
+        # error: Incompatible return value type (got "List[None]",
+        # expected "List[ndarray[Any, Any]]")
+        return result  # type: ignore[return-value]
+
+    def iset(
+        self,
+        loc: int | slice | np.ndarray,
+        value: ArrayLike,
+        inplace: bool = False,
+        refs: BlockValuesRefs | None = None,
+    ) -> None:
+        """
+        Set new item in-place. Does not consolidate. Adds new Block if not
+        contained in the current set of items
+        """
+
+        # FIXME: refactor, clearly separate broadcasting & zip-like assignment
+        #        can prob also fix the various if tests for sparse/categorical
+        if self._blklocs is None and self.ndim > 1:
+            self._rebuild_blknos_and_blklocs()
+
+        # Note: we exclude DTA/TDA here
+        value_is_extension_type = is_1d_only_ea_dtype(value.dtype)
+        if not value_is_extension_type:
+            if value.ndim == 2:
+                value = value.T
+            else:
+                value = ensure_block_shape(value, ndim=2)
+
+            if value.shape[1:] != self.shape[1:]:
+                raise AssertionError(
+                    "Shape of new values must be compatible with manager shape"
+                )
+
+        if lib.is_integer(loc):
+            # We have 6 tests where loc is _not_ an int.
+            # In this case, get_blkno_placements will yield only one tuple,
+            #  containing (self._blknos[loc], BlockPlacement(slice(0, 1, 1)))
+
+            # Check if we can use _iset_single fastpath
+            loc = cast(int, loc)
+            blkno = self.blknos[loc]
+            blk = self.blocks[blkno]
+            if len(blk._mgr_locs) == 1:  # TODO: fastest way to check this?
+                return self._iset_single(
+                    loc,
+                    value,
+                    inplace=inplace,
+                    blkno=blkno,
+                    blk=blk,
+                    refs=refs,
+                )
+
+            # error: Incompatible types in assignment (expression has type
+            # "List[Union[int, slice, ndarray]]", variable has type "Union[int,
+            # slice, ndarray]")
+            loc = [loc]  # type: ignore[assignment]
+
+        # categorical/sparse/datetimetz
+        if value_is_extension_type:
+
+            def value_getitem(placement):
+                return value
+
+        else:
+
+            def value_getitem(placement):
+                return value[placement.indexer]
+
+        # Accessing public blknos ensures the public versions are initialized
+        blknos = self.blknos[loc]
+        blklocs = self.blklocs[loc].copy()
+
+        unfit_mgr_locs = []
+        unfit_val_locs = []
+        removed_blknos = []
+        for blkno_l, val_locs in libinternals.get_blkno_placements(blknos, group=True):
+            blk = self.blocks[blkno_l]
+            blk_locs = blklocs[val_locs.indexer]
+            if inplace and blk.should_store(value):
+                # Updating inplace -> check if we need to do Copy-on-Write
+                if not self._has_no_reference_block(blkno_l):
+                    self._iset_split_block(
+                        blkno_l, blk_locs, value_getitem(val_locs), refs=refs
+                    )
+                else:
+                    blk.set_inplace(blk_locs, value_getitem(val_locs))
+                    continue
+            else:
+                unfit_mgr_locs.append(blk.mgr_locs.as_array[blk_locs])
+                unfit_val_locs.append(val_locs)
+
+                # If all block items are unfit, schedule the block for removal.
+                if len(val_locs) == len(blk.mgr_locs):
+                    removed_blknos.append(blkno_l)
+                    continue
+                else:
+                    # Defer setting the new values to enable consolidation
+                    self._iset_split_block(blkno_l, blk_locs, refs=refs)
+
+        if len(removed_blknos):
+            # Remove blocks & update blknos accordingly
+            is_deleted = np.zeros(self.nblocks, dtype=np.bool_)
+            is_deleted[removed_blknos] = True
+
+            new_blknos = np.empty(self.nblocks, dtype=np.intp)
+            new_blknos.fill(-1)
+            new_blknos[~is_deleted] = np.arange(self.nblocks - len(removed_blknos))
+            self._blknos = new_blknos[self._blknos]
+            self.blocks = tuple(
+                blk for i, blk in enumerate(self.blocks) if i not in set(removed_blknos)
+            )
+
+        if unfit_val_locs:
+            unfit_idxr = np.concatenate(unfit_mgr_locs)
+            unfit_count = len(unfit_idxr)
+
+            new_blocks: list[Block] = []
+            if value_is_extension_type:
+                # This code (ab-)uses the fact that EA blocks contain only
+                # one item.
+                # TODO(EA2D): special casing unnecessary with 2D EAs
+                new_blocks.extend(
+                    new_block_2d(
+                        values=value,
+                        placement=BlockPlacement(slice(mgr_loc, mgr_loc + 1)),
+                        refs=refs,
+                    )
+                    for mgr_loc in unfit_idxr
+                )
+
+                self._blknos[unfit_idxr] = np.arange(unfit_count) + len(self.blocks)
+                self._blklocs[unfit_idxr] = 0
+
+            else:
+                # unfit_val_locs contains BlockPlacement objects
+                unfit_val_items = unfit_val_locs[0].append(unfit_val_locs[1:])
+
+                new_blocks.append(
+                    new_block_2d(
+                        values=value_getitem(unfit_val_items),
+                        placement=BlockPlacement(unfit_idxr),
+                        refs=refs,
+                    )
+                )
+
+                self._blknos[unfit_idxr] = len(self.blocks)
+                self._blklocs[unfit_idxr] = np.arange(unfit_count)
+
+            self.blocks += tuple(new_blocks)
+
+            # Newly created block's dtype may already be present.
+            self._known_consolidated = False
+
+    def _iset_split_block(
+        self,
+        blkno_l: int,
+        blk_locs: np.ndarray | list[int],
+        value: ArrayLike | None = None,
+        refs: BlockValuesRefs | None = None,
+    ) -> None:
+        """Removes columns from a block by splitting the block.
+
+        Avoids copying the whole block through slicing and updates the manager
+        after determining the new block structure. Optionally adds a new block,
+        otherwise has to be done by the caller.
+
+        Parameters
+        ----------
+        blkno_l: The block number to operate on, relevant for updating the manager
+        blk_locs: The locations of our block that should be deleted.
+        value: The value to set as a replacement.
+        refs: The reference tracking object of the value to set.
+        """
+        blk = self.blocks[blkno_l]
+
+        if self._blklocs is None:
+            self._rebuild_blknos_and_blklocs()
+
+        nbs_tup = tuple(blk.delete(blk_locs))
+        if value is not None:
+            locs = blk.mgr_locs.as_array[blk_locs]
+            first_nb = new_block_2d(value, BlockPlacement(locs), refs=refs)
+        else:
+            first_nb = nbs_tup[0]
+            nbs_tup = tuple(nbs_tup[1:])
+
+        nr_blocks = len(self.blocks)
+        blocks_tup = (
+            self.blocks[:blkno_l] + (first_nb,) + self.blocks[blkno_l + 1 :] + nbs_tup
+        )
+        self.blocks = blocks_tup
+
+        if not nbs_tup and value is not None:
+            # No need to update anything if split did not happen
+            return
+
+        self._blklocs[first_nb.mgr_locs.indexer] = np.arange(len(first_nb))
+
+        for i, nb in enumerate(nbs_tup):
+            self._blklocs[nb.mgr_locs.indexer] = np.arange(len(nb))
+            self._blknos[nb.mgr_locs.indexer] = i + nr_blocks
+
+    def _iset_single(
+        self,
+        loc: int,
+        value: ArrayLike,
+        inplace: bool,
+        blkno: int,
+        blk: Block,
+        refs: BlockValuesRefs | None = None,
+    ) -> None:
+        """
+        Fastpath for iset when we are only setting a single position and
+        the Block currently in that position is itself single-column.
+
+        In this case we can swap out the entire Block and blklocs and blknos
+        are unaffected.
+        """
+        # Caller is responsible for verifying value.shape
+
+        if inplace and blk.should_store(value):
+            copy = not self._has_no_reference_block(blkno)
+            iloc = self.blklocs[loc]
+            blk.set_inplace(slice(iloc, iloc + 1), value, copy=copy)
+            return
+
+        nb = new_block_2d(value, placement=blk._mgr_locs, refs=refs)
+        old_blocks = self.blocks
+        new_blocks = old_blocks[:blkno] + (nb,) + old_blocks[blkno + 1 :]
+        self.blocks = new_blocks
+        return
+
+    def column_setitem(
+        self, loc: int, idx: int | slice | np.ndarray, value, inplace_only: bool = False
+    ) -> None:
+        """
+        Set values ("setitem") into a single column (not setting the full column).
+
+        This is a method on the BlockManager level, to avoid creating an
+        intermediate Series at the DataFrame level (`s = df[loc]; s[idx] = value`)
+        """
+        if not self._has_no_reference(loc):
+            blkno = self.blknos[loc]
+            # Split blocks to only copy the column we want to modify
+            blk_loc = self.blklocs[loc]
+            # Copy our values
+            values = self.blocks[blkno].values
+            if values.ndim == 1:
+                values = values.copy()
+            else:
+                # Use [blk_loc] as indexer to keep ndim=2, this already results in a
+                # copy
+                values = values[[blk_loc]]
+            self._iset_split_block(blkno, [blk_loc], values)
+
+        # this manager is only created temporarily to mutate the values in place
+        # so don't track references, otherwise the `setitem` would perform CoW again
+        col_mgr = self.iget(loc, track_ref=False)
+        if inplace_only:
+            col_mgr.setitem_inplace(idx, value)
+        else:
+            new_mgr = col_mgr.setitem((idx,), value)
+            self.iset(loc, new_mgr._block.values, inplace=True)
+
+    def insert(self, loc: int, item: Hashable, value: ArrayLike, refs=None) -> None:
+        """
+        Insert item at selected position.
+
+        Parameters
+        ----------
+        loc : int
+        item : hashable
+        value : np.ndarray or ExtensionArray
+        refs : The reference tracking object of the value to set.
+        """
+        new_axis = self.items.insert(loc, item)
+
+        if value.ndim == 2:
+            value = value.T
+            if len(value) > 1:
+                raise ValueError(
+                    f"Expected a 1D array, got an array with shape {value.T.shape}"
+                )
+        else:
+            value = ensure_block_shape(value, ndim=self.ndim)
+
+        bp = BlockPlacement(slice(loc, loc + 1))
+        block = new_block_2d(values=value, placement=bp, refs=refs)
+
+        if not len(self.blocks):
+            # Fastpath
+            self._blklocs = np.array([0], dtype=np.intp)
+            self._blknos = np.array([0], dtype=np.intp)
+        else:
+            self._insert_update_mgr_locs(loc)
+            self._insert_update_blklocs_and_blknos(loc)
+
+        self.axes[0] = new_axis
+        self.blocks += (block,)
+
+        self._known_consolidated = False
+
+        if (
+            get_option("performance_warnings")
+            and sum(not block.is_extension for block in self.blocks) > 100
+        ):
+            warnings.warn(
+                "DataFrame is highly fragmented.  This is usually the result "
+                "of calling `frame.insert` many times, which has poor performance.  "
+                "Consider joining all columns at once using pd.concat(axis=1) "
+                "instead. To get a de-fragmented frame, use `newframe = frame.copy()`",
+                PerformanceWarning,
+                stacklevel=find_stack_level(),
+            )
+
+    def _insert_update_mgr_locs(self, loc) -> None:
+        """
+        When inserting a new Block at location 'loc', we increment
+        all of the mgr_locs of blocks above that by one.
+        """
+        # Faster version of set(arr) for sequences of small numbers
+        blknos = np.bincount(self.blknos[loc:]).nonzero()[0]
+        for blkno in blknos:
+            # .620 this way, .326 of which is in increment_above
+            blk = self.blocks[blkno]
+            blk._mgr_locs = blk._mgr_locs.increment_above(loc)
+
+    def _insert_update_blklocs_and_blknos(self, loc) -> None:
+        """
+        When inserting a new Block at location 'loc', we update our
+        _blklocs and _blknos.
+        """
+
+        # Accessing public blklocs ensures the public versions are initialized
+        if loc == self.blklocs.shape[0]:
+            # np.append is a lot faster, let's use it if we can.
+            self._blklocs = np.append(self._blklocs, 0)
+            self._blknos = np.append(self._blknos, len(self.blocks))
+        elif loc == 0:
+            # As of numpy 1.26.4, np.concatenate faster than np.append
+            self._blklocs = np.concatenate([[0], self._blklocs])
+            self._blknos = np.concatenate([[len(self.blocks)], self._blknos])
+        else:
+            new_blklocs, new_blknos = libinternals.update_blklocs_and_blknos(
+                self.blklocs, self.blknos, loc, len(self.blocks)
+            )
+            self._blklocs = new_blklocs
+            self._blknos = new_blknos
+
+    def idelete(self, indexer) -> BlockManager:
+        """
+        Delete selected locations, returning a new BlockManager.
+        """
+        is_deleted = np.zeros(self.shape[0], dtype=np.bool_)
+        is_deleted[indexer] = True
+        taker = (~is_deleted).nonzero()[0]
+
+        nbs = self._slice_take_blocks_ax0(taker, only_slice=True, ref_inplace_op=True)
+        new_columns = self.items[~is_deleted]
+        axes = [new_columns, self.axes[1]]
+        return type(self)(tuple(nbs), axes, verify_integrity=False)
+
+    # ----------------------------------------------------------------
+    # Block-wise Operation
+
+    def grouped_reduce(self, func: Callable) -> Self:
+        """
+        Apply grouped reduction function blockwise, returning a new BlockManager.
+
+        Parameters
+        ----------
+        func : grouped reduction function
+
+        Returns
+        -------
+        BlockManager
+        """
+        result_blocks: list[Block] = []
+
+        for blk in self.blocks:
+            if blk.is_object:
+                # split on object-dtype blocks bc some columns may raise
+                #  while others do not.
+                for sb in blk._split():
+                    applied = sb.apply(func)
+                    result_blocks = extend_blocks(applied, result_blocks)
+            else:
+                applied = blk.apply(func)
+                result_blocks = extend_blocks(applied, result_blocks)
+
+        if len(result_blocks) == 0:
+            nrows = 0
+        else:
+            nrows = result_blocks[0].values.shape[-1]
+        index = default_index(nrows)
+
+        return type(self).from_blocks(result_blocks, [self.axes[0], index])
+
+    def reduce(self, func: Callable) -> Self:
+        """
+        Apply reduction function blockwise, returning a single-row BlockManager.
+
+        Parameters
+        ----------
+        func : reduction function
+
+        Returns
+        -------
+        BlockManager
+        """
+        # If 2D, we assume that we're operating column-wise
+        assert self.ndim == 2
+
+        res_blocks: list[Block] = []
+        for blk in self.blocks:
+            nbs = blk.reduce(func)
+            res_blocks.extend(nbs)
+
+        index = Index([None])  # placeholder
+        new_mgr = type(self).from_blocks(res_blocks, [self.items, index])
+        return new_mgr
+
+    def operate_blockwise(self, other: BlockManager, array_op) -> BlockManager:
+        """
+        Apply array_op blockwise with another (aligned) BlockManager.
+        """
+        return operate_blockwise(self, other, array_op)
+
+    def _equal_values(self: BlockManager, other: BlockManager) -> bool:
+        """
+        Used in .equals defined in base class. Only check the column values
+        assuming shape and indexes have already been checked.
+        """
+        return blockwise_all(self, other, array_equals)
+
+    def quantile(
+        self,
+        *,
+        qs: Index,  # with dtype float 64
+        interpolation: QuantileInterpolation = "linear",
+    ) -> Self:
+        """
+        Iterate over blocks applying quantile reduction.
+        This routine is intended for reduction type operations and
+        will do inference on the generated blocks.
+
+        Parameters
+        ----------
+        interpolation : type of interpolation, default 'linear'
+        qs : list of the quantiles to be computed
+
+        Returns
+        -------
+        BlockManager
+        """
+        # Series dispatches to DataFrame for quantile, which allows us to
+        #  simplify some of the code here and in the blocks
+        assert self.ndim >= 2
+        assert is_list_like(qs)  # caller is responsible for this
+
+        new_axes = list(self.axes)
+        new_axes[1] = Index(qs, dtype=np.float64)
+
+        blocks = [
+            blk.quantile(qs=qs, interpolation=interpolation) for blk in self.blocks
+        ]
+
+        return type(self)(blocks, new_axes)
+
+    # ----------------------------------------------------------------
+
+    def unstack(self, unstacker, fill_value) -> BlockManager:
+        """
+        Return a BlockManager with all blocks unstacked.
+
+        Parameters
+        ----------
+        unstacker : reshape._Unstacker
+        fill_value : Any
+            fill_value for newly introduced missing values.
+
+        Returns
+        -------
+        unstacked : BlockManager
+        """
+        new_columns = unstacker.get_new_columns(self.items)
+        new_index = unstacker.new_index
+
+        allow_fill = not unstacker.mask_all
+        if allow_fill:
+            # calculating the full mask once and passing it to Block._unstack is
+            #  faster than letting calculating it in each repeated call
+            new_mask2D = (~unstacker.mask).reshape(*unstacker.full_shape)
+            needs_masking = new_mask2D.any(axis=0)
+        else:
+            needs_masking = np.zeros(unstacker.full_shape[1], dtype=bool)
+
+        new_blocks: list[Block] = []
+        columns_mask: list[np.ndarray] = []
+
+        if len(self.items) == 0:
+            factor = 1
+        else:
+            fac = len(new_columns) / len(self.items)
+            assert fac == int(fac)
+            factor = int(fac)
+
+        for blk in self.blocks:
+            mgr_locs = blk.mgr_locs
+            new_placement = mgr_locs.tile_for_unstack(factor)
+
+            blocks, mask = blk._unstack(
+                unstacker,
+                fill_value,
+                new_placement=new_placement,
+                needs_masking=needs_masking,
+            )
+
+            new_blocks.extend(blocks)
+            columns_mask.extend(mask)
+
+            # Block._unstack should ensure this holds,
+            assert mask.sum() == sum(len(nb._mgr_locs) for nb in blocks)
+            # In turn this ensures that in the BlockManager call below
+            #  we have len(new_columns) == sum(x.shape[0] for x in new_blocks)
+            #  which suffices to allow us to pass verify_inegrity=False
+
+        new_columns = new_columns[columns_mask]
+
+        bm = BlockManager(new_blocks, [new_columns, new_index], verify_integrity=False)
+        return bm
+
+    def to_iter_dict(self) -> Generator[tuple[str, Self]]:
+        """
+        Yield a tuple of (str(dtype), BlockManager)
+
+        Returns
+        -------
+        values : a tuple of (str(dtype), BlockManager)
+        """
+        key = lambda block: str(block.dtype)
+        for dtype, blocks in itertools.groupby(sorted(self.blocks, key=key), key=key):
+            # TODO(EA2D): the combine will be unnecessary with 2D EAs
+            yield dtype, self._combine(list(blocks))
+
+    def as_array(
+        self,
+        dtype: np.dtype | None = None,
+        copy: bool = False,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        """
+        Convert the blockmanager data into an numpy array.
+
+        Parameters
+        ----------
+        dtype : np.dtype or None, default None
+            Data type of the return array.
+        copy : bool, default False
+            If True then guarantee that a copy is returned. A value of
+            False does not guarantee that the underlying data is not
+            copied.
+        na_value : object, default lib.no_default
+            Value to be used as the missing value sentinel.
+
+        Returns
+        -------
+        arr : ndarray
+        """
+        passed_nan = lib.is_float(na_value) and isna(na_value)
+
+        if len(self.blocks) == 0:
+            arr = np.empty(self.shape, dtype=float)
+            return arr.transpose()
+
+        if self.is_single_block:
+            blk = self.blocks[0]
+
+            if na_value is not lib.no_default:
+                # We want to copy when na_value is provided to avoid
+                # mutating the original object
+                if lib.is_np_dtype(blk.dtype, "f") and passed_nan:
+                    # We are already numpy-float and na_value=np.nan
+                    pass
+                else:
+                    copy = True
+
+            if blk.is_extension:
+                # Avoid implicit conversion of extension blocks to object
+
+                # error: Item "ndarray" of "Union[ndarray, ExtensionArray]" has no
+                # attribute "to_numpy"
+                arr = blk.values.to_numpy(  # type: ignore[union-attr]
+                    dtype=dtype,
+                    na_value=na_value,
+                    copy=copy,
+                ).reshape(blk.shape)
+            elif not copy:
+                arr = np.asarray(blk.values, dtype=dtype)
+            else:
+                arr = np.array(blk.values, dtype=dtype, copy=copy)
+
+            if not copy:
+                arr = arr.view()
+                arr.flags.writeable = False
+        else:
+            arr = self._interleave(dtype=dtype, na_value=na_value)
+            # The underlying data was copied within _interleave, so no need
+            # to further copy if copy=True or setting na_value
+
+        if na_value is lib.no_default:
+            pass
+        elif arr.dtype.kind == "f" and passed_nan:
+            pass
+        else:
+            arr[isna(arr)] = na_value
+
+        return arr.transpose()
+
+    def _interleave(
+        self,
+        dtype: np.dtype | None = None,
+        na_value: object = lib.no_default,
+    ) -> np.ndarray:
+        """
+        Return ndarray from blocks with specified item order
+        Items must be contained in the blocks
+        """
+        if not dtype:
+            # Incompatible types in assignment (expression has type
+            # "Optional[Union[dtype[Any], ExtensionDtype]]", variable has
+            # type "Optional[dtype[Any]]")
+            dtype = interleaved_dtype(  # type: ignore[assignment]
+                [blk.dtype for blk in self.blocks]
+            )
+
+        # error: Argument 1 to "ensure_np_dtype" has incompatible type
+        # "Optional[dtype[Any]]"; expected "Union[dtype[Any], ExtensionDtype]"
+        dtype = ensure_np_dtype(dtype)  # type: ignore[arg-type]
+        result = np.empty(self.shape, dtype=dtype)
+
+        itemmask = np.zeros(self.shape[0])
+
+        if dtype == np.dtype("object") and na_value is lib.no_default:
+            # much more performant than using to_numpy below
+            for blk in self.blocks:
+                rl = blk.mgr_locs
+                arr = blk.get_values(dtype)
+                result[rl.indexer] = arr
+                itemmask[rl.indexer] = 1
+            return result
+
+        for blk in self.blocks:
+            rl = blk.mgr_locs
+            if blk.is_extension:
+                # Avoid implicit conversion of extension blocks to object
+
+                # error: Item "ndarray" of "Union[ndarray, ExtensionArray]" has no
+                # attribute "to_numpy"
+                arr = blk.values.to_numpy(  # type: ignore[union-attr]
+                    dtype=dtype,
+                    na_value=na_value,
+                )
+            else:
+                arr = blk.get_values(dtype)
+            result[rl.indexer] = arr
+            itemmask[rl.indexer] = 1
+
+        if not itemmask.all():
+            raise AssertionError("Some items were not contained in blocks")
+
+        return result
+
+    # ----------------------------------------------------------------
+    # Consolidation
+
+    def is_consolidated(self) -> bool:
+        """
+        Return True if more than one block with the same dtype
+        """
+        if not self._known_consolidated:
+            self._consolidate_check()
+        return self._is_consolidated
+
+    def _consolidate_check(self) -> None:
+        if len(self.blocks) == 1:
+            # fastpath
+            self._is_consolidated = True
+            self._known_consolidated = True
+            return
+        dtypes = [blk.dtype for blk in self.blocks if blk._can_consolidate]
+        self._is_consolidated = len(dtypes) == len(set(dtypes))
+        self._known_consolidated = True
+
+    def _consolidate_inplace(self) -> None:
+        # In general, _consolidate_inplace should only be called via
+        #  DataFrame._consolidate_inplace, otherwise we will fail to invalidate
+        #  the DataFrame's _item_cache. The exception is for newly-created
+        #  BlockManager objects not yet attached to a DataFrame.
+        if not self.is_consolidated():
+            self.blocks = _consolidate(self.blocks)
+            self._is_consolidated = True
+            self._known_consolidated = True
+            self._rebuild_blknos_and_blklocs()
+
+    # ----------------------------------------------------------------
+    # Concatenation
+
+    @classmethod
+    def concat_horizontal(cls, mgrs: list[Self], axes: list[Index]) -> Self:
+        """
+        Concatenate uniformly-indexed BlockManagers horizontally.
+        """
+        offset = 0
+        blocks: list[Block] = []
+        for mgr in mgrs:
+            for blk in mgr.blocks:
+                # We need to do getitem_block here otherwise we would be altering
+                #  blk.mgr_locs in place, which would render it invalid. This is only
+                #  relevant in the copy=False case.
+                nb = blk.slice_block_columns(slice(None))
+                nb._mgr_locs = nb._mgr_locs.add(offset)
+                blocks.append(nb)
+
+            offset += len(mgr.items)
+
+        new_mgr = cls(tuple(blocks), axes)
+        return new_mgr
+
+    @classmethod
+    def concat_vertical(cls, mgrs: list[Self], axes: list[Index]) -> Self:
+        """
+        Concatenate uniformly-indexed BlockManagers vertically.
+        """
+        raise NotImplementedError("This logic lives (for now) in internals.concat")
+
+
+class SingleBlockManager(BaseBlockManager):
+    """manage a single block with"""
+
+    @property
+    def ndim(self) -> Literal[1]:
+        return 1
+
+    _is_consolidated = True
+    _known_consolidated = True
+    __slots__ = ()
+    is_single_block = True
+
+    def __init__(
+        self,
+        block: Block,
+        axis: Index,
+        verify_integrity: bool = False,
+    ) -> None:
+        # Assertions disabled for performance
+        # assert isinstance(block, Block), type(block)
+        # assert isinstance(axis, Index), type(axis)
+
+        self.axes = [axis]
+        self.blocks = (block,)
+
+    @classmethod
+    def from_blocks(
+        cls,
+        blocks: list[Block],
+        axes: list[Index],
+    ) -> Self:
+        """
+        Constructor for BlockManager and SingleBlockManager with same signature.
+        """
+        assert len(blocks) == 1
+        assert len(axes) == 1
+        return cls(blocks[0], axes[0], verify_integrity=False)
+
+    @classmethod
+    def from_array(
+        cls, array: ArrayLike, index: Index, refs: BlockValuesRefs | None = None
+    ) -> SingleBlockManager:
+        """
+        Constructor for if we have an array that is not yet a Block.
+        """
+        array = maybe_coerce_values(array)
+        bp = BlockPlacement(slice(0, len(index)))
+        block = new_block(array, placement=bp, ndim=1, refs=refs)
+        return cls(block, index)
+
+    def to_2d_mgr(self, columns: Index) -> BlockManager:
+        """
+        Manager analogue of Series.to_frame
+        """
+        blk = self.blocks[0]
+        arr = ensure_block_shape(blk.values, ndim=2)
+        bp = BlockPlacement(0)
+        new_blk = type(blk)(arr, placement=bp, ndim=2, refs=blk.refs)
+        axes = [columns, self.axes[0]]
+        return BlockManager([new_blk], axes=axes, verify_integrity=False)
+
+    def _has_no_reference(self, i: int = 0) -> bool:
+        """
+        Check for column `i` if it has references.
+        (whether it references another array or is itself being referenced)
+        Returns True if the column has no references.
+        """
+        return not self.blocks[0].refs.has_reference()
+
+    def __getstate__(self):
+        block_values = [b.values for b in self.blocks]
+        block_items = [self.items[b.mgr_locs.indexer] for b in self.blocks]
+        axes_array = list(self.axes)
+
+        extra_state = {
+            "0.14.1": {
+                "axes": axes_array,
+                "blocks": [
+                    {"values": b.values, "mgr_locs": b.mgr_locs.indexer}
+                    for b in self.blocks
+                ],
+            }
+        }
+
+        # First three elements of the state are to maintain forward
+        # compatibility with 0.13.1.
+        return axes_array, block_values, block_items, extra_state
+
+    def __setstate__(self, state) -> None:
+        def unpickle_block(values, mgr_locs, ndim: int) -> Block:
+            # TODO(EA2D): ndim would be unnecessary with 2D EAs
+            # older pickles may store e.g. DatetimeIndex instead of DatetimeArray
+            values = extract_array(values, extract_numpy=True)
+            if not isinstance(mgr_locs, BlockPlacement):
+                mgr_locs = BlockPlacement(mgr_locs)
+
+            values = maybe_coerce_values(values)
+            return new_block(values, placement=mgr_locs, ndim=ndim)
+
+        if isinstance(state, tuple) and len(state) >= 4 and "0.14.1" in state[3]:
+            state = state[3]["0.14.1"]
+            self.axes = [ensure_index(ax) for ax in state["axes"]]
+            ndim = len(self.axes)
+            self.blocks = tuple(
+                unpickle_block(b["values"], b["mgr_locs"], ndim=ndim)
+                for b in state["blocks"]
+            )
+        else:
+            raise NotImplementedError("pre-0.14.1 pickles are no longer supported")
+
+        self._post_setstate()
+
+    def _post_setstate(self) -> None:
+        pass
+
+    @cache_readonly
+    def _block(self) -> Block:
+        return self.blocks[0]
+
+    @final
+    @property
+    def array(self) -> ArrayLike:
+        """
+        Quick access to the backing array of the Block.
+        """
+        return self.blocks[0].values
+
+    # error: Cannot override writeable attribute with read-only property
+    @property
+    def _blknos(self) -> None:  # type: ignore[override]
+        """compat with BlockManager"""
+        return None
+
+    # error: Cannot override writeable attribute with read-only property
+    @property
+    def _blklocs(self) -> None:  # type: ignore[override]
+        """compat with BlockManager"""
+        return None
+
+    def get_rows_with_mask(self, indexer: npt.NDArray[np.bool_]) -> Self:
+        # similar to get_slice, but not restricted to slice indexer
+        blk = self._block
+        if len(indexer) > 0 and indexer.all():
+            return type(self)(blk.copy(deep=False), self.index)
+        array = blk.values[indexer]
+
+        if isinstance(indexer, np.ndarray) and indexer.dtype.kind == "b":
+            # boolean indexing always gives a copy with numpy
+            refs = None
+        else:
+            # TODO(CoW) in theory only need to track reference if new_array is a view
+            refs = blk.refs
+
+        bp = BlockPlacement(slice(0, len(array)))
+        block = type(blk)(array, placement=bp, ndim=1, refs=refs)
+
+        new_idx = self.index[indexer]
+        return type(self)(block, new_idx)
+
+    def get_slice(self, slobj: slice, axis: AxisInt = 0) -> SingleBlockManager:
+        # Assertion disabled for performance
+        # assert isinstance(slobj, slice), type(slobj)
+        if axis >= self.ndim:
+            raise IndexError("Requested axis not found in manager")
+
+        blk = self._block
+        array = blk.values[slobj]
+        bp = BlockPlacement(slice(0, len(array)))
+        # TODO this method is only used in groupby SeriesSplitter at the moment,
+        # so passing refs is not yet covered by the tests
+        block = type(blk)(array, placement=bp, ndim=1, refs=blk.refs)
+        new_index = self.index._getitem_slice(slobj)
+        return type(self)(block, new_index)
+
+    @property
+    def index(self) -> Index:
+        return self.axes[0]
+
+    @property
+    def dtype(self) -> DtypeObj:
+        return self._block.dtype
+
+    def get_dtypes(self) -> npt.NDArray[np.object_]:
+        return np.array([self._block.dtype], dtype=object)
+
+    def external_values(self):
+        """The array that Series.values returns"""
+        return self._block.external_values()
+
+    def internal_values(self):
+        """The array that Series._values returns"""
+        return self._block.values
+
+    def array_values(self) -> ExtensionArray:
+        """The array that Series.array returns"""
+        return self._block.array_values
+
+    def get_numeric_data(self) -> Self:
+        if self._block.is_numeric:
+            return self.copy(deep=False)
+        return self.make_empty()
+
+    @property
+    def _can_hold_na(self) -> bool:
+        return self._block._can_hold_na
+
+    def setitem_inplace(self, indexer, value) -> None:
+        """
+        Set values with indexer.
+
+        For SingleBlockManager, this backs s[indexer] = value
+
+        This is an inplace version of `setitem()`, mutating the manager/values
+        in place, not returning a new Manager (and Block), and thus never changing
+        the dtype.
+        """
+        if not self._has_no_reference(0):
+            self.blocks = (self._block.copy(),)
+            self._reset_cache()
+
+        arr = self.array
+
+        # EAs will do this validation in their own __setitem__ methods.
+        if isinstance(arr, np.ndarray):
+            # Note: checking for ndarray instead of np.dtype means we exclude
+            #  dt64/td64, which do their own validation.
+            value = np_can_hold_element(arr.dtype, value)
+
+        if isinstance(value, np.ndarray) and value.ndim == 1 and len(value) == 1:
+            # NumPy 1.25 deprecation: https://github.com/numpy/numpy/pull/10615
+            value = value[0, ...]
+
+        arr[indexer] = value
+
+    def idelete(self, indexer) -> SingleBlockManager:
+        """
+        Delete single location from SingleBlockManager.
+
+        Ensures that self.blocks doesn't become empty.
+        """
+        nb = self._block.delete(indexer)[0]
+        self.blocks = (nb,)
+        self.axes[0] = self.axes[0].delete(indexer)
+        self._reset_cache()
+        return self
+
+    def fast_xs(self, loc):
+        """
+        fast path for getting a cross-section
+        return a view of the data
+        """
+        raise NotImplementedError("Use series._values[loc] instead")
+
+    def set_values(self, values: ArrayLike) -> None:
+        """
+        Set the values of the single block in place.
+
+        Use at your own risk! This does not check if the passed values are
+        valid for the current Block/SingleBlockManager (length, dtype, etc),
+        and this does not properly keep track of references.
+        """
+        # NOTE(CoW) Currently this is only used for FrameColumnApply.series_generator
+        # which handles CoW by setting the refs manually if necessary
+        self.blocks[0].values = values
+        self.blocks[0]._mgr_locs = BlockPlacement(slice(len(values)))
+
+    def _equal_values(self, other: Self) -> bool:
+        """
+        Used in .equals defined in base class. Only check the column values
+        assuming shape and indexes have already been checked.
+        """
+        # For SingleBlockManager (i.e.Series)
+        if other.ndim != 1:
+            return False
+        left = self.blocks[0].values
+        right = other.blocks[0].values
+        return array_equals(left, right)
+
+    def grouped_reduce(self, func):
+        arr = self.array
+        res = func(arr)
+        index = default_index(len(res))
+
+        mgr = type(self).from_array(res, index)
+        return mgr
+
+
+# --------------------------------------------------------------------
+# Constructor Helpers
+
+
+def create_block_manager_from_blocks(
+    blocks: list[Block],
+    axes: list[Index],
+    consolidate: bool = True,
+    verify_integrity: bool = True,
+) -> BlockManager:
+    # If verify_integrity=False, then caller is responsible for checking
+    #  all(x.shape[-1] == len(axes[1]) for x in blocks)
+    #  sum(x.shape[0] for x in blocks) == len(axes[0])
+    #  set(x for blk in blocks for x in blk.mgr_locs) == set(range(len(axes[0])))
+    #  all(blk.ndim == 2 for blk in blocks)
+    # This allows us to safely pass verify_integrity=False
+
+    try:
+        mgr = BlockManager(blocks, axes, verify_integrity=verify_integrity)
+
+    except ValueError as err:
+        arrays = [blk.values for blk in blocks]
+        tot_items = sum(arr.shape[0] for arr in arrays)
+        raise_construction_error(tot_items, arrays[0].shape[1:], axes, err)
+
+    if consolidate:
+        mgr._consolidate_inplace()
+    return mgr
+
+
+def create_block_manager_from_column_arrays(
+    arrays: list[ArrayLike],
+    axes: list[Index],
+    consolidate: bool,
+    refs: list,
+) -> BlockManager:
+    # Assertions disabled for performance (caller is responsible for verifying)
+    # assert isinstance(axes, list)
+    # assert all(isinstance(x, Index) for x in axes)
+    # assert all(isinstance(x, (np.ndarray, ExtensionArray)) for x in arrays)
+    # assert all(type(x) is not NumpyExtensionArray for x in arrays)
+    # assert all(x.ndim == 1 for x in arrays)
+    # assert all(len(x) == len(axes[1]) for x in arrays)
+    # assert len(arrays) == len(axes[0])
+    # These last three are sufficient to allow us to safely pass
+    #  verify_integrity=False below.
+
+    try:
+        blocks = _form_blocks(arrays, consolidate, refs)
+        mgr = BlockManager(blocks, axes, verify_integrity=False)
+    except ValueError as e:
+        raise_construction_error(len(arrays), arrays[0].shape, axes, e)
+    if consolidate:
+        mgr._consolidate_inplace()
+    return mgr
+
+
+def raise_construction_error(
+    tot_items: int,
+    block_shape: Shape,
+    axes: list[Index],
+    e: ValueError | None = None,
+) -> NoReturn:
+    """raise a helpful message about our construction"""
+    passed = tuple(map(int, [tot_items] + list(block_shape)))
+    # Correcting the user facing error message during dataframe construction
+    if len(passed) <= 2:
+        passed = passed[::-1]
+
+    implied = tuple(len(ax) for ax in axes)
+    # Correcting the user facing error message during dataframe construction
+    if len(implied) <= 2:
+        implied = implied[::-1]
+
+    # We return the exception object instead of raising it so that we
+    #  can raise it in the caller; mypy plays better with that
+    if passed == implied and e is not None:
+        raise e
+    if block_shape[0] == 0:
+        raise ValueError("Empty data passed with indices specified.")
+    raise ValueError(f"Shape of passed values is {passed}, indices imply {implied}")
+
+
+# -----------------------------------------------------------------------
+
+
+def _grouping_func(tup: tuple[int, ArrayLike]) -> tuple[int, DtypeObj]:
+    dtype = tup[1].dtype
+
+    if is_1d_only_ea_dtype(dtype):
+        # We know these won't be consolidated, so don't need to group these.
+        # This avoids expensive comparisons of CategoricalDtype objects
+        sep = id(dtype)
+    else:
+        sep = 0
+
+    return sep, dtype
+
+
+def _form_blocks(arrays: list[ArrayLike], consolidate: bool, refs: list) -> list[Block]:
+    tuples = enumerate(arrays)
+
+    if not consolidate:
+        return _tuples_to_blocks_no_consolidate(tuples, refs)
+
+    # when consolidating, we can ignore refs (either stacking always copies,
+    # or the EA is already copied in the calling dict_to_mgr)
+
+    # group by dtype
+    grouper = itertools.groupby(tuples, _grouping_func)
+
+    nbs: list[Block] = []
+    for (_, dtype), tup_block in grouper:
+        block_type = get_block_type(dtype)
+
+        if isinstance(dtype, np.dtype):
+            is_dtlike = dtype.kind in "mM"
+
+            if issubclass(dtype.type, (str, bytes)):
+                dtype = np.dtype(object)
+
+            values, placement = _stack_arrays(tup_block, dtype)
+            if is_dtlike:
+                values = ensure_wrapped_if_datetimelike(values)
+            blk = block_type(values, placement=BlockPlacement(placement), ndim=2)
+            nbs.append(blk)
+
+        elif is_1d_only_ea_dtype(dtype):
+            dtype_blocks = [
+                block_type(x[1], placement=BlockPlacement(x[0]), ndim=2)
+                for x in tup_block
+            ]
+            nbs.extend(dtype_blocks)
+
+        else:
+            dtype_blocks = [
+                block_type(
+                    ensure_block_shape(x[1], 2), placement=BlockPlacement(x[0]), ndim=2
+                )
+                for x in tup_block
+            ]
+            nbs.extend(dtype_blocks)
+    return nbs
+
+
+def _tuples_to_blocks_no_consolidate(tuples, refs) -> list[Block]:
+    # tuples produced within _form_blocks are of the form (placement, array)
+    return [
+        new_block_2d(
+            ensure_block_shape(arr, ndim=2), placement=BlockPlacement(i), refs=ref
+        )
+        for ((i, arr), ref) in zip(tuples, refs)
+    ]
+
+
+def _stack_arrays(tuples, dtype: np.dtype):
+    placement, arrays = zip(*tuples)
+
+    first = arrays[0]
+    shape = (len(arrays),) + first.shape
+
+    stacked = np.empty(shape, dtype=dtype)
+    for i, arr in enumerate(arrays):
+        stacked[i] = arr
+
+    return stacked, placement
+
+
+def _consolidate(blocks: tuple[Block, ...]) -> tuple[Block, ...]:
+    """
+    Merge blocks having same dtype, exclude non-consolidating blocks
+    """
+    # sort by _can_consolidate, dtype
+    gkey = lambda x: x._consolidate_key
+    grouper = itertools.groupby(sorted(blocks, key=gkey), gkey)
+
+    new_blocks: list[Block] = []
+    for (_can_consolidate, dtype), group_blocks in grouper:
+        merged_blocks, _ = _merge_blocks(
+            list(group_blocks), dtype=dtype, can_consolidate=_can_consolidate
+        )
+        new_blocks = extend_blocks(merged_blocks, new_blocks)
+    return tuple(new_blocks)
+
+
+def _merge_blocks(
+    blocks: list[Block], dtype: DtypeObj, can_consolidate: bool
+) -> tuple[list[Block], bool]:
+    if len(blocks) == 1:
+        return blocks, False
+
+    if can_consolidate:
+        # TODO: optimization potential in case all mgrs contain slices and
+        # combination of those slices is a slice, too.
+        new_mgr_locs = np.concatenate([b.mgr_locs.as_array for b in blocks])
+
+        new_values: ArrayLike
+
+        if isinstance(blocks[0].dtype, np.dtype):
+            # error: List comprehension has incompatible type List[Union[ndarray,
+            # ExtensionArray]]; expected List[Union[complex, generic,
+            # Sequence[Union[int, float, complex, str, bytes, generic]],
+            # Sequence[Sequence[Any]], SupportsArray]]
+            new_values = np.vstack([b.values for b in blocks])  # type: ignore[misc]
+        else:
+            bvals = [blk.values for blk in blocks]
+            bvals2 = cast(Sequence[NDArrayBackedExtensionArray], bvals)
+            new_values = bvals2[0]._concat_same_type(bvals2, axis=0)
+
+        argsort = np.argsort(new_mgr_locs)
+        new_values = new_values[argsort]
+        new_mgr_locs = new_mgr_locs[argsort]
+
+        bp = BlockPlacement(new_mgr_locs)
+        return [new_block_2d(new_values, placement=bp)], True
+
+    # can't consolidate --> no merge
+    return blocks, False
+
+
+def _preprocess_slice_or_indexer(
+    slice_or_indexer: slice | np.ndarray, length: int, allow_fill: bool
+):
+    if isinstance(slice_or_indexer, slice):
+        return (
+            "slice",
+            slice_or_indexer,
+            libinternals.slice_len(slice_or_indexer, length),
+        )
+    else:
+        if (
+            not isinstance(slice_or_indexer, np.ndarray)
+            or slice_or_indexer.dtype.kind != "i"
+        ):
+            dtype = getattr(slice_or_indexer, "dtype", None)
+            raise TypeError(type(slice_or_indexer), dtype)
+
+        indexer = ensure_platform_int(slice_or_indexer)
+        if not allow_fill:
+            indexer = maybe_convert_indices(indexer, length)
+        return "fancy", indexer, len(indexer)
+
+
+def make_na_array(dtype: DtypeObj, shape: Shape, fill_value) -> ArrayLike:
+    if isinstance(dtype, DatetimeTZDtype):
+        # NB: exclude e.g. pyarrow[dt64tz] dtypes
+        ts = Timestamp(fill_value).as_unit(dtype.unit)
+        i8values = np.full(shape, ts._value)
+        dt64values = i8values.view(f"M8[{dtype.unit}]")
+        return DatetimeArray._simple_new(dt64values, dtype=dtype)
+
+    elif is_1d_only_ea_dtype(dtype):
+        dtype = cast(ExtensionDtype, dtype)
+        cls = dtype.construct_array_type()
+
+        missing_arr = cls._from_sequence([], dtype=dtype)
+        ncols, nrows = shape
+        assert ncols == 1, ncols
+        empty_arr = -1 * np.ones((nrows,), dtype=np.intp)
+        return missing_arr.take(empty_arr, allow_fill=True, fill_value=fill_value)
+    elif isinstance(dtype, ExtensionDtype):
+        # TODO: no tests get here, a handful would if we disabled
+        #  the dt64tz special-case above (which is faster)
+        cls = dtype.construct_array_type()
+        missing_arr = cls._empty(shape=shape, dtype=dtype)
+        missing_arr[:] = fill_value
+        return missing_arr
+    else:
+        # NB: we should never get here with dtype integer or bool;
+        #  if we did, the missing_arr.fill would cast to gibberish
+        missing_arr_np = np.empty(shape, dtype=dtype)
+        missing_arr_np.fill(fill_value)
+
+        if dtype.kind in "mM":
+            missing_arr_np = ensure_wrapped_if_datetimelike(missing_arr_np)
+        return missing_arr_np
diff --git a/pandas/pandas/core/internals/ops.py b/pandas/pandas/core/internals/ops.py
new file mode 100644
index 0000000000000..cf9466c0bdf0b
--- /dev/null
+++ b/pandas/pandas/core/internals/ops.py
@@ -0,0 +1,154 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    NamedTuple,
+)
+
+from pandas.core.dtypes.common import is_1d_only_ea_dtype
+
+if TYPE_CHECKING:
+    from collections.abc import Iterator
+
+    from pandas._libs.internals import BlockPlacement
+    from pandas._typing import ArrayLike
+
+    from pandas.core.internals.blocks import Block
+    from pandas.core.internals.managers import BlockManager
+
+
+class BlockPairInfo(NamedTuple):
+    lvals: ArrayLike
+    rvals: ArrayLike
+    locs: BlockPlacement
+    left_ea: bool
+    right_ea: bool
+    rblk: Block
+
+
+def _iter_block_pairs(
+    left: BlockManager, right: BlockManager
+) -> Iterator[BlockPairInfo]:
+    # At this point we have already checked the parent DataFrames for
+    #  assert rframe._indexed_same(lframe)
+
+    for blk in left.blocks:
+        locs = blk.mgr_locs
+        blk_vals = blk.values
+
+        left_ea = blk_vals.ndim == 1
+
+        rblks = right._slice_take_blocks_ax0(locs.indexer, only_slice=True)
+
+        # Assertions are disabled for performance, but should hold:
+        # if left_ea:
+        #    assert len(locs) == 1, locs
+        #    assert len(rblks) == 1, rblks
+        #    assert rblks[0].shape[0] == 1, rblks[0].shape
+
+        for rblk in rblks:
+            right_ea = rblk.values.ndim == 1
+
+            lvals, rvals = _get_same_shape_values(blk, rblk, left_ea, right_ea)
+            info = BlockPairInfo(lvals, rvals, locs, left_ea, right_ea, rblk)
+            yield info
+
+
+def operate_blockwise(
+    left: BlockManager, right: BlockManager, array_op
+) -> BlockManager:
+    # At this point we have already checked the parent DataFrames for
+    #  assert rframe._indexed_same(lframe)
+
+    res_blks: list[Block] = []
+    for lvals, rvals, locs, left_ea, right_ea, rblk in _iter_block_pairs(left, right):
+        res_values = array_op(lvals, rvals)
+        if (
+            left_ea
+            and not right_ea
+            and hasattr(res_values, "reshape")
+            and not is_1d_only_ea_dtype(res_values.dtype)
+        ):
+            res_values = res_values.reshape(1, -1)
+        nbs = rblk._split_op_result(res_values)
+
+        # Assertions are disabled for performance, but should hold:
+        # if right_ea or left_ea:
+        #    assert len(nbs) == 1
+        # else:
+        #    assert res_values.shape == lvals.shape, (res_values.shape, lvals.shape)
+
+        _reset_block_mgr_locs(nbs, locs)
+
+        res_blks.extend(nbs)
+
+    # Assertions are disabled for performance, but should hold:
+    #  slocs = {y for nb in res_blks for y in nb.mgr_locs.as_array}
+    #  nlocs = sum(len(nb.mgr_locs.as_array) for nb in res_blks)
+    #  assert nlocs == len(left.items), (nlocs, len(left.items))
+    #  assert len(slocs) == nlocs, (len(slocs), nlocs)
+    #  assert slocs == set(range(nlocs)), slocs
+
+    new_mgr = type(right)(tuple(res_blks), axes=right.axes, verify_integrity=False)
+    return new_mgr
+
+
+def _reset_block_mgr_locs(nbs: list[Block], locs) -> None:
+    """
+    Reset mgr_locs to correspond to our original DataFrame.
+    """
+    for nb in nbs:
+        nblocs = locs[nb.mgr_locs.indexer]
+        nb.mgr_locs = nblocs
+        # Assertions are disabled for performance, but should hold:
+        #  assert len(nblocs) == nb.shape[0], (len(nblocs), nb.shape)
+        #  assert all(x in locs.as_array for x in nb.mgr_locs.as_array)
+
+
+def _get_same_shape_values(
+    lblk: Block, rblk: Block, left_ea: bool, right_ea: bool
+) -> tuple[ArrayLike, ArrayLike]:
+    """
+    Slice lblk.values to align with rblk.  Squeeze if we have EAs.
+    """
+    lvals = lblk.values
+    rvals = rblk.values
+
+    # Require that the indexing into lvals be slice-like
+    assert rblk.mgr_locs.is_slice_like, rblk.mgr_locs
+
+    # TODO(EA2D): with 2D EAs only this first clause would be needed
+    if not (left_ea or right_ea):
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[Union[ndarray, slice], slice]"
+        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[call-overload]
+        assert lvals.shape == rvals.shape, (lvals.shape, rvals.shape)
+    elif left_ea and right_ea:
+        assert lvals.shape == rvals.shape, (lvals.shape, rvals.shape)
+    elif right_ea:
+        # lvals are 2D, rvals are 1D
+
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[Union[ndarray, slice], slice]"
+        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[call-overload]
+        assert lvals.shape[0] == 1, lvals.shape
+        lvals = lvals[0, :]
+    else:
+        # lvals are 1D, rvals are 2D
+        assert rvals.shape[0] == 1, rvals.shape
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
+        rvals = rvals[0, :]  # type: ignore[call-overload]
+
+    return lvals, rvals
+
+
+def blockwise_all(left: BlockManager, right: BlockManager, op) -> bool:
+    """
+    Blockwise `all` reduction.
+    """
+    for info in _iter_block_pairs(left, right):
+        res = op(info.lvals, info.rvals)
+        if not res:
+            return False
+    return True
diff --git a/pandas/pandas/core/methods/__init__.py b/pandas/pandas/core/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/methods/describe.py b/pandas/pandas/core/methods/describe.py
new file mode 100644
index 0000000000000..6637cd8492062
--- /dev/null
+++ b/pandas/pandas/core/methods/describe.py
@@ -0,0 +1,374 @@
+"""
+Module responsible for execution of NDFrame.describe() method.
+
+Method NDFrame.describe() delegates actual execution to function describe_ndframe().
+"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._typing import (
+    DtypeObj,
+    NDFrameT,
+    npt,
+)
+from pandas.util._validators import validate_percentile
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_numeric_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+)
+
+from pandas.core.arrays.floating import Float64Dtype
+from pandas.core.reshape.concat import concat
+
+from pandas.io.formats.format import format_percentiles
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Sequence,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+def describe_ndframe(
+    *,
+    obj: NDFrameT,
+    include: str | Sequence[str] | None,
+    exclude: str | Sequence[str] | None,
+    percentiles: Sequence[float] | np.ndarray | None,
+) -> NDFrameT:
+    """Describe series or dataframe.
+
+    Called from pandas.core.generic.NDFrame.describe()
+
+    Parameters
+    ----------
+    obj: DataFrame or Series
+        Either dataframe or series to be described.
+    include : 'all', list-like of dtypes or None (default), optional
+        A white list of data types to include in the result. Ignored for ``Series``.
+    exclude : list-like of dtypes or None (default), optional,
+        A black list of data types to omit from the result. Ignored for ``Series``.
+    percentiles : list-like of numbers, optional
+        The percentiles to include in the output. All should fall between 0 and 1.
+        The default is ``[.25, .5, .75]``, which returns the 25th, 50th, and
+        75th percentiles.
+
+    Returns
+    -------
+    Dataframe or series description.
+    """
+    percentiles = _refine_percentiles(percentiles)
+
+    describer: NDFrameDescriberAbstract
+    if obj.ndim == 1:
+        describer = SeriesDescriber(
+            obj=cast("Series", obj),
+        )
+    else:
+        describer = DataFrameDescriber(
+            obj=cast("DataFrame", obj),
+            include=include,
+            exclude=exclude,
+        )
+
+    result = describer.describe(percentiles=percentiles)
+    return cast(NDFrameT, result)
+
+
+class NDFrameDescriberAbstract(ABC):
+    """Abstract class for describing dataframe or series.
+
+    Parameters
+    ----------
+    obj : Series or DataFrame
+        Object to be described.
+    """
+
+    def __init__(self, obj: DataFrame | Series) -> None:
+        self.obj = obj
+
+    @abstractmethod
+    def describe(self, percentiles: Sequence[float] | np.ndarray) -> DataFrame | Series:
+        """Do describe either series or dataframe.
+
+        Parameters
+        ----------
+        percentiles : list-like of numbers
+            The percentiles to include in the output.
+        """
+
+
+class SeriesDescriber(NDFrameDescriberAbstract):
+    """Class responsible for creating series description."""
+
+    obj: Series
+
+    def describe(self, percentiles: Sequence[float] | np.ndarray) -> Series:
+        describe_func = select_describe_func(
+            self.obj,
+        )
+        return describe_func(self.obj, percentiles)
+
+
+class DataFrameDescriber(NDFrameDescriberAbstract):
+    """Class responsible for creating dataobj description.
+
+    Parameters
+    ----------
+    obj : DataFrame
+        DataFrame to be described.
+    include : 'all', list-like of dtypes or None
+        A white list of data types to include in the result.
+    exclude : list-like of dtypes or None
+        A black list of data types to omit from the result.
+    """
+
+    obj: DataFrame
+
+    def __init__(
+        self,
+        obj: DataFrame,
+        *,
+        include: str | Sequence[str] | None,
+        exclude: str | Sequence[str] | None,
+    ) -> None:
+        self.include = include
+        self.exclude = exclude
+
+        if obj.ndim == 2 and obj.columns.size == 0:
+            raise ValueError("Cannot describe a DataFrame without columns")
+
+        super().__init__(obj)
+
+    def describe(self, percentiles: Sequence[float] | np.ndarray) -> DataFrame:
+        data = self._select_data()
+
+        ldesc: list[Series] = []
+        for _, series in data.items():
+            describe_func = select_describe_func(series)
+            ldesc.append(describe_func(series, percentiles))
+
+        col_names = reorder_columns(ldesc)
+        d = concat(
+            [x.reindex(col_names) for x in ldesc],
+            axis=1,
+            ignore_index=True,
+            sort=False,
+        )
+        d.columns = data.columns.copy()
+        return d
+
+    def _select_data(self) -> DataFrame:
+        """Select columns to be described."""
+        if (self.include is None) and (self.exclude is None):
+            # when some numerics are found, keep only numerics
+            default_include: list[npt.DTypeLike] = [np.number, "datetime"]
+            data = self.obj.select_dtypes(include=default_include)
+            if len(data.columns) == 0:
+                data = self.obj
+        elif self.include == "all":
+            if self.exclude is not None:
+                msg = "exclude must be None when include is 'all'"
+                raise ValueError(msg)
+            data = self.obj
+        else:
+            data = self.obj.select_dtypes(
+                include=self.include,
+                exclude=self.exclude,
+            )
+        return data
+
+
+def reorder_columns(ldesc: Sequence[Series]) -> list[Hashable]:
+    """Set a convenient order for rows for display."""
+    names: list[Hashable] = []
+    seen_names: set[Hashable] = set()
+    ldesc_indexes = sorted((x.index for x in ldesc), key=len)
+    for idxnames in ldesc_indexes:
+        for name in idxnames:
+            if name not in seen_names:
+                seen_names.add(name)
+                names.append(name)
+    return names
+
+
+def describe_numeric_1d(series: Series, percentiles: Sequence[float]) -> Series:
+    """Describe series containing numerical data.
+
+    Parameters
+    ----------
+    series : Series
+        Series to be described.
+    percentiles : list-like of numbers
+        The percentiles to include in the output.
+    """
+    from pandas import Series
+
+    formatted_percentiles = format_percentiles(percentiles)
+
+    stat_index = ["count", "mean", "std", "min"] + formatted_percentiles + ["max"]
+    d = (
+        [series.count(), series.mean(), series.std(), series.min()]
+        + series.quantile(percentiles).tolist()
+        + [series.max()]
+    )
+    # GH#48340 - always return float on non-complex numeric data
+    dtype: DtypeObj | None
+    if isinstance(series.dtype, ExtensionDtype):
+        if isinstance(series.dtype, ArrowDtype):
+            if series.dtype.kind == "m":
+                # GH53001: describe timedeltas with object dtype
+                dtype = None
+            else:
+                import pyarrow as pa
+
+                dtype = ArrowDtype(pa.float64())
+        else:
+            dtype = Float64Dtype()
+    elif series.dtype.kind in "iufb":
+        # i.e. numeric but exclude complex dtype
+        dtype = np.dtype("float")
+    else:
+        dtype = None
+    return Series(d, index=stat_index, name=series.name, dtype=dtype)
+
+
+def describe_categorical_1d(
+    data: Series,
+    percentiles_ignored: Sequence[float],
+) -> Series:
+    """Describe series containing categorical data.
+
+    Parameters
+    ----------
+    data : Series
+        Series to be described.
+    percentiles_ignored : list-like of numbers
+        Ignored, but in place to unify interface.
+    """
+    names = ["count", "unique", "top", "freq"]
+    objcounts = data.value_counts()
+    count_unique = len(objcounts[objcounts != 0])
+    if count_unique > 0:
+        top, freq = objcounts.index[0], objcounts.iloc[0]
+        dtype = None
+    else:
+        # If the DataFrame is empty, set 'top' and 'freq' to None
+        # to maintain output shape consistency
+        top, freq = np.nan, np.nan
+        dtype = "object"
+
+    result = [data.count(), count_unique, top, freq]
+
+    from pandas import Series
+
+    return Series(result, index=names, name=data.name, dtype=dtype)
+
+
+def describe_timestamp_1d(data: Series, percentiles: Sequence[float]) -> Series:
+    """Describe series containing datetime64 dtype.
+
+    Parameters
+    ----------
+    data : Series
+        Series to be described.
+    percentiles : list-like of numbers
+        The percentiles to include in the output.
+    """
+    # GH-30164
+    from pandas import Series
+
+    formatted_percentiles = format_percentiles(percentiles)
+
+    stat_index = ["count", "mean", "min"] + formatted_percentiles + ["max"]
+    d = (
+        [data.count(), data.mean(), data.min()]
+        + data.quantile(percentiles).tolist()
+        + [data.max()]
+    )
+    return Series(d, index=stat_index, name=data.name)
+
+
+def select_describe_func(
+    data: Series,
+) -> Callable:
+    """Select proper function for describing series based on data type.
+
+    Parameters
+    ----------
+    data : Series
+        Series to be described.
+    """
+    if is_bool_dtype(data.dtype):
+        return describe_categorical_1d
+    elif is_numeric_dtype(data):
+        return describe_numeric_1d
+    elif data.dtype.kind == "M" or isinstance(data.dtype, DatetimeTZDtype):
+        return describe_timestamp_1d
+    elif data.dtype.kind == "m":
+        return describe_numeric_1d
+    else:
+        return describe_categorical_1d
+
+
+def _refine_percentiles(
+    percentiles: Sequence[float] | np.ndarray | None,
+) -> npt.NDArray[np.float64]:
+    """
+    Ensure that percentiles are unique and sorted.
+
+    Parameters
+    ----------
+    percentiles : list-like of numbers, optional
+        The percentiles to include in the output.
+    """
+    if percentiles is None:
+        return np.array([0.25, 0.5, 0.75])
+
+    # explicit conversion of `percentiles` to list
+    percentiles = list(percentiles)
+
+    # get them all to be in [0, 1]
+    validate_percentile(percentiles)
+
+    # median should always be included
+    # if 0.5 not in percentiles:
+    #     percentiles.append(0.5) # Removing this line as a fix for issue : 60550
+
+    if percentiles == []:
+        percentiles.append(
+            0.5
+        )  # by default if percentiles is empty then appending 50th percentile.
+
+    percentiles = np.asarray(percentiles)
+
+    # sort and check for duplicates
+    unique_pcts = np.unique(percentiles)
+    assert percentiles is not None
+    if len(unique_pcts) < len(percentiles):
+        raise ValueError("percentiles cannot contain duplicates")
+
+    return unique_pcts
diff --git a/pandas/pandas/core/methods/selectn.py b/pandas/pandas/core/methods/selectn.py
new file mode 100644
index 0000000000000..02e7445f1d275
--- /dev/null
+++ b/pandas/pandas/core/methods/selectn.py
@@ -0,0 +1,280 @@
+"""
+Implementation of nlargest and nsmallest.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Hashable,
+    Sequence,
+)
+from typing import (
+    TYPE_CHECKING,
+    Generic,
+    cast,
+    final,
+)
+
+import numpy as np
+
+from pandas._libs import algos as libalgos
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_complex_dtype,
+    is_integer_dtype,
+    is_list_like,
+    is_numeric_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import BaseMaskedDtype
+
+from pandas.core.indexes.api import default_index
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        DtypeObj,
+        IndexLabel,
+        NDFrameT,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+else:
+    # Generic[...] requires a non-str, provide it with a plain TypeVar at
+    # runtime to avoid circular imports
+    from pandas._typing import T
+
+    NDFrameT = T
+    DataFrame = T
+    Series = T
+
+
+class SelectN(Generic[NDFrameT]):
+    def __init__(self, obj: NDFrameT, n: int, keep: str) -> None:
+        self.obj = obj
+        self.n = n
+        self.keep = keep
+
+        if self.keep not in ("first", "last", "all"):
+            raise ValueError('keep must be either "first", "last" or "all"')
+
+    def compute(self, method: str) -> NDFrameT:
+        raise NotImplementedError
+
+    @final
+    def nlargest(self) -> NDFrameT:
+        return self.compute("nlargest")
+
+    @final
+    def nsmallest(self) -> NDFrameT:
+        return self.compute("nsmallest")
+
+    @final
+    @staticmethod
+    def is_valid_dtype_n_method(dtype: DtypeObj) -> bool:
+        """
+        Helper function to determine if dtype is valid for
+        nsmallest/nlargest methods
+        """
+        if is_numeric_dtype(dtype):
+            return not is_complex_dtype(dtype)
+        return needs_i8_conversion(dtype)
+
+
+class SelectNSeries(SelectN[Series]):
+    """
+    Implement n largest/smallest for Series
+
+    Parameters
+    ----------
+    obj : Series
+    n : int
+    keep : {'first', 'last'}, default 'first'
+
+    Returns
+    -------
+    nordered : Series
+    """
+
+    def compute(self, method: str) -> Series:
+        from pandas.core.reshape.concat import concat
+
+        n = self.n
+        dtype = self.obj.dtype
+        if not self.is_valid_dtype_n_method(dtype):
+            raise TypeError(f"Cannot use method '{method}' with dtype {dtype}")
+
+        if n <= 0:
+            return self.obj[[]]
+
+        dropped = self.obj.dropna()
+        nan_index = self.obj.drop(dropped.index)
+
+        # slow method
+        if n >= len(self.obj):
+            ascending = method == "nsmallest"
+            return self.obj.sort_values(ascending=ascending).head(n)
+
+        # fast method
+        new_dtype = dropped.dtype
+
+        # Similar to algorithms._ensure_data
+        arr = dropped._values
+        if needs_i8_conversion(arr.dtype):
+            arr = arr.view("i8")
+        elif isinstance(arr.dtype, BaseMaskedDtype):
+            arr = arr._data
+        else:
+            arr = np.asarray(arr)
+        if arr.dtype.kind == "b":
+            arr = arr.view(np.uint8)
+
+        if method == "nlargest":
+            arr = -arr
+            if is_integer_dtype(new_dtype):
+                # GH 21426: ensure reverse ordering at boundaries
+                arr -= 1
+
+            elif is_bool_dtype(new_dtype):
+                # GH 26154: ensure False is smaller than True
+                arr = 1 - (-arr)
+
+        if self.keep == "last":
+            arr = arr[::-1]
+
+        nbase = n
+        narr = len(arr)
+        n = min(n, narr)
+
+        # arr passed into kth_smallest must be contiguous. We copy
+        # here because kth_smallest will modify its input
+        # avoid OOB access with kth_smallest_c when n <= 0
+        if len(arr) > 0:
+            kth_val = libalgos.kth_smallest(arr.copy(order="C"), n - 1)
+        else:
+            kth_val = np.nan
+        (ns,) = np.nonzero(arr <= kth_val)
+        inds = ns[arr[ns].argsort(kind="mergesort")]
+
+        if self.keep != "all":
+            inds = inds[:n]
+            findex = nbase
+        else:
+            if len(inds) < nbase <= len(nan_index) + len(inds):
+                findex = len(nan_index) + len(inds)
+            else:
+                findex = len(inds)
+
+        if self.keep == "last":
+            # reverse indices
+            inds = narr - 1 - inds
+
+        return concat([dropped.iloc[inds], nan_index]).iloc[:findex]
+
+
+class SelectNFrame(SelectN[DataFrame]):
+    """
+    Implement n largest/smallest for DataFrame
+
+    Parameters
+    ----------
+    obj : DataFrame
+    n : int
+    keep : {'first', 'last'}, default 'first'
+    columns : list or str
+
+    Returns
+    -------
+    nordered : DataFrame
+    """
+
+    def __init__(self, obj: DataFrame, n: int, keep: str, columns: IndexLabel) -> None:
+        super().__init__(obj, n, keep)
+        if not is_list_like(columns) or isinstance(columns, tuple):
+            columns = [columns]
+
+        columns = cast(Sequence[Hashable], columns)
+        columns = list(columns)
+        self.columns = columns
+
+    def compute(self, method: str) -> DataFrame:
+        n = self.n
+        frame = self.obj
+        columns = self.columns
+
+        for column in columns:
+            dtype = frame[column].dtype
+            if not self.is_valid_dtype_n_method(dtype):
+                raise TypeError(
+                    f"Column {column!r} has dtype {dtype}, "
+                    f"cannot use method {method!r} with this dtype"
+                )
+
+        def get_indexer(current_indexer: Index, other_indexer: Index) -> Index:
+            """
+            Helper function to concat `current_indexer` and `other_indexer`
+            depending on `method`
+            """
+            if method == "nsmallest":
+                return current_indexer.append(other_indexer)
+            else:
+                return other_indexer.append(current_indexer)
+
+        # Below we save and reset the index in case index contains duplicates
+        original_index = frame.index
+        cur_frame = frame = frame.reset_index(drop=True)
+        cur_n = n
+        indexer: Index = default_index(0)
+
+        for i, column in enumerate(columns):
+            # For each column we apply method to cur_frame[column].
+            # If it's the last column or if we have the number of
+            # results desired we are done.
+            # Otherwise there are duplicates of the largest/smallest
+            # value and we need to look at the rest of the columns
+            # to determine which of the rows with the largest/smallest
+            # value in the column to keep.
+            series = cur_frame[column]
+            is_last_column = len(columns) - 1 == i
+            values = getattr(series, method)(
+                cur_n, keep=self.keep if is_last_column else "all"
+            )
+
+            if is_last_column or len(values) <= cur_n:
+                indexer = get_indexer(indexer, values.index)
+                break
+
+            # Now find all values which are equal to
+            # the (nsmallest: largest)/(nlargest: smallest)
+            # from our series.
+            border_value = values == values[values.index[-1]]
+
+            # Some of these values are among the top-n
+            # some aren't.
+            unsafe_values = values[border_value]
+
+            # These values are definitely among the top-n
+            safe_values = values[~border_value]
+            indexer = get_indexer(indexer, safe_values.index)
+
+            # Go on and separate the unsafe_values on the remaining
+            # columns.
+            cur_frame = cur_frame.loc[unsafe_values.index]
+            cur_n = n - len(indexer)
+
+        frame = frame.take(indexer)
+
+        # Restore the index on frame
+        frame.index = original_index.take(indexer)
+
+        # If there is only one column, the frame is already sorted.
+        if len(columns) == 1:
+            return frame
+
+        ascending = method == "nsmallest"
+
+        return frame.sort_values(columns, ascending=ascending, kind="mergesort")
diff --git a/pandas/pandas/core/methods/to_dict.py b/pandas/pandas/core/methods/to_dict.py
new file mode 100644
index 0000000000000..aea95e4684573
--- /dev/null
+++ b/pandas/pandas/core/methods/to_dict.py
@@ -0,0 +1,286 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import maybe_box_native
+from pandas.core.dtypes.dtypes import (
+    BaseMaskedDtype,
+    ExtensionDtype,
+)
+
+from pandas.core import common as com
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+    from pandas._typing import MutableMappingT
+
+    from pandas import DataFrame
+
+
+def create_data_for_split(
+    df: DataFrame, are_all_object_dtype_cols: bool, object_dtype_indices: list[int]
+) -> Generator[list]:
+    """
+    Simple helper method to create data for to ``to_dict(orient="split")``
+    to create the main output data
+    """
+    if are_all_object_dtype_cols:
+        for tup in df.itertuples(index=False, name=None):
+            yield list(map(maybe_box_native, tup))
+    else:
+        for tup in df.itertuples(index=False, name=None):
+            data = list(tup)
+            if object_dtype_indices:
+                # If we have object_dtype_cols, apply maybe_box_naive after
+                # for perf
+                for i in object_dtype_indices:
+                    data[i] = maybe_box_native(data[i])
+            yield data
+
+
+@overload
+def to_dict(
+    df: DataFrame,
+    orient: Literal["dict", "list", "series", "split", "tight", "index"] = ...,
+    *,
+    into: type[MutableMappingT] | MutableMappingT,
+    index: bool = ...,
+) -> MutableMappingT: ...
+
+
+@overload
+def to_dict(
+    df: DataFrame,
+    orient: Literal["records"],
+    *,
+    into: type[MutableMappingT] | MutableMappingT,
+    index: bool = ...,
+) -> list[MutableMappingT]: ...
+
+
+@overload
+def to_dict(
+    df: DataFrame,
+    orient: Literal["dict", "list", "series", "split", "tight", "index"] = ...,
+    *,
+    into: type[dict] = ...,
+    index: bool = ...,
+) -> dict: ...
+
+
+@overload
+def to_dict(
+    df: DataFrame,
+    orient: Literal["records"],
+    *,
+    into: type[dict] = ...,
+    index: bool = ...,
+) -> list[dict]: ...
+
+
+# error: Incompatible default for argument "into" (default has type "type[dict
+# [Any, Any]]", argument has type "type[MutableMappingT] | MutableMappingT")
+def to_dict(
+    df: DataFrame,
+    orient: Literal[
+        "dict", "list", "series", "split", "tight", "records", "index"
+    ] = "dict",
+    *,
+    into: type[MutableMappingT] | MutableMappingT = dict,  # type: ignore[assignment]
+    index: bool = True,
+) -> MutableMappingT | list[MutableMappingT]:
+    """
+    Convert the DataFrame to a dictionary.
+
+    The type of the key-value pairs can be customized with the parameters
+    (see below).
+
+    Parameters
+    ----------
+    orient : str {'dict', 'list', 'series', 'split', 'tight', 'records', 'index'}
+        Determines the type of the values of the dictionary.
+
+        - 'dict' (default) : dict like {column -> {index -> value}}
+        - 'list' : dict like {column -> [values]}
+        - 'series' : dict like {column -> Series(values)}
+        - 'split' : dict like
+          {'index' -> [index], 'columns' -> [columns], 'data' -> [values]}
+        - 'tight' : dict like
+          {'index' -> [index], 'columns' -> [columns], 'data' -> [values],
+          'index_names' -> [index.names], 'column_names' -> [column.names]}
+        - 'records' : list like
+          [{column -> value}, ... , {column -> value}]
+        - 'index' : dict like {index -> {column -> value}}
+
+        .. versionadded:: 1.4.0
+            'tight' as an allowed value for the ``orient`` argument
+
+    into : class, default dict
+        The collections.abc.MutableMapping subclass used for all Mappings
+        in the return value.  Can be the actual class or an empty
+        instance of the mapping type you want.  If you want a
+        collections.defaultdict, you must pass it initialized.
+
+    index : bool, default True
+        Whether to include the index item (and index_names item if `orient`
+        is 'tight') in the returned dictionary. Can only be ``False``
+        when `orient` is 'split' or 'tight'.
+
+        .. versionadded:: 2.0.0
+
+    Returns
+    -------
+    dict, list or collections.abc.Mapping
+        Return a collections.abc.MutableMapping object representing the
+        DataFrame. The resulting transformation depends on the `orient` parameter.
+    """
+    if orient != "tight" and not df.columns.is_unique:
+        warnings.warn(
+            "DataFrame columns are not unique, some columns will be omitted.",
+            UserWarning,
+            stacklevel=find_stack_level(),
+        )
+    # GH16122
+    # error: Call to untyped function "standardize_mapping" in typed context
+    into_c = com.standardize_mapping(into)  # type: ignore[no-untyped-call]
+
+    #  error: Incompatible types in assignment (expression has type "str",
+    # variable has type "Literal['dict', 'list', 'series', 'split', 'tight',
+    # 'records', 'index']")
+    orient = orient.lower()  # type: ignore[assignment]
+
+    if not index and orient not in ["split", "tight"]:
+        raise ValueError(
+            "'index=False' is only valid when 'orient' is 'split' or 'tight'"
+        )
+
+    if orient == "series":
+        # GH46470 Return quickly if orient series to avoid creating dtype objects
+        return into_c((k, v) for k, v in df.items())
+
+    if orient == "dict":
+        return into_c((k, v.to_dict(into=into)) for k, v in df.items())
+
+    box_native_indices = [
+        i
+        for i, col_dtype in enumerate(df.dtypes.values)
+        if col_dtype == np.dtype(object) or isinstance(col_dtype, ExtensionDtype)
+    ]
+
+    are_all_object_dtype_cols = len(box_native_indices) == len(df.dtypes)
+
+    if orient == "list":
+        object_dtype_indices_as_set: set[int] = set(box_native_indices)
+        box_na_values = (
+            lib.no_default
+            if not isinstance(col_dtype, BaseMaskedDtype)
+            else libmissing.NA
+            for col_dtype in df.dtypes.values
+        )
+        return into_c(
+            (
+                k,
+                list(map(maybe_box_native, v.to_numpy(na_value=box_na_value)))
+                if i in object_dtype_indices_as_set
+                else list(map(maybe_box_native, v.to_numpy())),
+            )
+            for i, (box_na_value, (k, v)) in enumerate(zip(box_na_values, df.items()))
+        )
+
+    elif orient == "split":
+        data = list(
+            create_data_for_split(df, are_all_object_dtype_cols, box_native_indices)
+        )
+
+        return into_c(
+            ((("index", df.index.tolist()),) if index else ())
+            + (
+                ("columns", df.columns.tolist()),
+                ("data", data),
+            )
+        )
+
+    elif orient == "tight":
+        return into_c(
+            ((("index", df.index.tolist()),) if index else ())
+            + (
+                ("columns", df.columns.tolist()),
+                (
+                    "data",
+                    [
+                        list(map(maybe_box_native, t))
+                        for t in df.itertuples(index=False, name=None)
+                    ],
+                ),
+            )
+            + ((("index_names", list(df.index.names)),) if index else ())
+            + (("column_names", list(df.columns.names)),)
+        )
+
+    elif orient == "records":
+        columns = df.columns.tolist()
+        if are_all_object_dtype_cols:
+            return [
+                into_c(zip(columns, map(maybe_box_native, row)))
+                for row in df.itertuples(index=False, name=None)
+            ]
+        else:
+            data = [
+                into_c(zip(columns, t)) for t in df.itertuples(index=False, name=None)
+            ]
+            if box_native_indices:
+                object_dtype_indices_as_set = set(box_native_indices)
+                object_dtype_cols = {
+                    col
+                    for i, col in enumerate(df.columns)
+                    if i in object_dtype_indices_as_set
+                }
+                for row in data:
+                    for col in object_dtype_cols:
+                        row[col] = maybe_box_native(row[col])
+            return data  # type: ignore[return-value]
+
+    elif orient == "index":
+        if not df.index.is_unique:
+            raise ValueError("DataFrame index must be unique for orient='index'.")
+        columns = df.columns.tolist()
+        if are_all_object_dtype_cols:
+            return into_c(
+                (t[0], dict(zip(df.columns, map(maybe_box_native, t[1:]))))
+                for t in df.itertuples(name=None)
+            )
+        elif box_native_indices:
+            object_dtype_indices_as_set = set(box_native_indices)
+            return into_c(
+                (
+                    t[0],
+                    {
+                        column: maybe_box_native(v)
+                        if i in object_dtype_indices_as_set
+                        else v
+                        for i, (column, v) in enumerate(zip(columns, t[1:]))
+                    },
+                )
+                for t in df.itertuples(name=None)
+            )
+        else:
+            return into_c(
+                (t[0], dict(zip(columns, t[1:]))) for t in df.itertuples(name=None)
+            )
+
+    else:
+        raise ValueError(f"orient '{orient}' not understood")
diff --git a/pandas/pandas/core/missing.py b/pandas/pandas/core/missing.py
new file mode 100644
index 0000000000000..ff2daae002731
--- /dev/null
+++ b/pandas/pandas/core/missing.py
@@ -0,0 +1,1099 @@
+"""
+Routines for filling missing data.
+"""
+
+from __future__ import annotations
+
+from functools import wraps
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    algos,
+    lib,
+)
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    F,
+    ReindexMethod,
+    npt,
+)
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.dtypes.cast import infer_dtype_from
+from pandas.core.dtypes.common import (
+    is_array_like,
+    is_bool_dtype,
+    is_numeric_dtype,
+    is_numeric_v_string_like,
+    is_object_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.missing import (
+    is_valid_na_for_dtype,
+    isna,
+    na_value_for_dtype,
+)
+
+if TYPE_CHECKING:
+    from pandas import Index
+
+
+def check_value_size(value, mask: npt.NDArray[np.bool_], length: int):
+    """
+    Validate the size of the values passed to ExtensionArray.fillna.
+    """
+    if is_array_like(value):
+        if len(value) != length:
+            raise ValueError(
+                f"Length of 'value' does not match. Got ({len(value)}) "
+                f" expected {length}"
+            )
+        value = value[mask]
+
+    return value
+
+
+def mask_missing(arr: ArrayLike, values_to_mask) -> npt.NDArray[np.bool_]:
+    """
+    Return a masking array of same size/shape as arr
+    with entries equaling any member of values_to_mask set to True
+
+    Parameters
+    ----------
+    arr : ArrayLike
+    values_to_mask: list, tuple, or scalar
+
+    Returns
+    -------
+    np.ndarray[bool]
+    """
+    # When called from Block.replace/replace_list, values_to_mask is a scalar
+    #  known to be holdable by arr.
+    # When called from Series._single_replace, values_to_mask is tuple or list
+    dtype, values_to_mask = infer_dtype_from(values_to_mask)
+
+    if isinstance(dtype, np.dtype):
+        values_to_mask = np.array(values_to_mask, dtype=dtype)
+    else:
+        cls = dtype.construct_array_type()
+        if not lib.is_list_like(values_to_mask):
+            values_to_mask = [values_to_mask]
+        values_to_mask = cls._from_sequence(values_to_mask, dtype=dtype, copy=False)
+
+    potential_na = False
+    if is_object_dtype(arr.dtype):
+        # pre-compute mask to avoid comparison to NA
+        potential_na = True
+        arr_mask = ~isna(arr)
+
+    na_mask = isna(values_to_mask)
+    nonna = values_to_mask[~na_mask]
+
+    # GH 21977
+    mask = np.zeros(arr.shape, dtype=bool)
+    if (
+        is_numeric_dtype(arr.dtype)
+        and not is_bool_dtype(arr.dtype)
+        and is_bool_dtype(nonna.dtype)
+    ):
+        pass
+    elif (
+        is_bool_dtype(arr.dtype)
+        and is_numeric_dtype(nonna.dtype)
+        and not is_bool_dtype(nonna.dtype)
+    ):
+        pass
+    else:
+        for x in nonna:
+            if is_numeric_v_string_like(arr, x):
+                # GH#29553 prevent numpy deprecation warnings
+                pass
+            else:
+                if potential_na:
+                    new_mask = np.zeros(arr.shape, dtype=np.bool_)
+                    new_mask[arr_mask] = arr[arr_mask] == x
+                else:
+                    new_mask = arr == x
+
+                    if not isinstance(new_mask, np.ndarray):
+                        # usually BooleanArray
+                        new_mask = new_mask.to_numpy(dtype=bool, na_value=False)
+                mask |= new_mask
+
+    if na_mask.any():
+        mask |= isna(arr)
+
+    return mask
+
+
+@overload
+def clean_fill_method(
+    method: Literal["ffill", "pad", "bfill", "backfill"],
+    *,
+    allow_nearest: Literal[False] = ...,
+) -> Literal["pad", "backfill"]: ...
+
+
+@overload
+def clean_fill_method(
+    method: Literal["ffill", "pad", "bfill", "backfill", "nearest"],
+    *,
+    allow_nearest: Literal[True],
+) -> Literal["pad", "backfill", "nearest"]: ...
+
+
+def clean_fill_method(
+    method: Literal["ffill", "pad", "bfill", "backfill", "nearest"],
+    *,
+    allow_nearest: bool = False,
+) -> Literal["pad", "backfill", "nearest"]:
+    if isinstance(method, str):
+        # error: Incompatible types in assignment (expression has type "str", variable
+        # has type "Literal['ffill', 'pad', 'bfill', 'backfill', 'nearest']")
+        method = method.lower()  # type: ignore[assignment]
+        if method == "ffill":
+            method = "pad"
+        elif method == "bfill":
+            method = "backfill"
+
+    valid_methods = ["pad", "backfill"]
+    expecting = "pad (ffill) or backfill (bfill)"
+    if allow_nearest:
+        valid_methods.append("nearest")
+        expecting = "pad (ffill), backfill (bfill) or nearest"
+    if method not in valid_methods:
+        raise ValueError(f"Invalid fill method. Expecting {expecting}. Got {method}")
+    return method
+
+
+# interpolation methods that dispatch to np.interp
+
+NP_METHODS = ["linear", "time", "index", "values"]
+
+# interpolation methods that dispatch to _interpolate_scipy_wrapper
+
+SP_METHODS = [
+    "nearest",
+    "zero",
+    "slinear",
+    "quadratic",
+    "cubic",
+    "barycentric",
+    "krogh",
+    "spline",
+    "polynomial",
+    "from_derivatives",
+    "piecewise_polynomial",
+    "pchip",
+    "akima",
+    "cubicspline",
+]
+
+
+def clean_interp_method(method: str, index: Index, **kwargs) -> str:
+    order = kwargs.get("order")
+
+    if method in ("spline", "polynomial") and order is None:
+        raise ValueError("You must specify the order of the spline or polynomial.")
+
+    valid = NP_METHODS + SP_METHODS
+    if method not in valid:
+        raise ValueError(f"method must be one of {valid}. Got '{method}' instead.")
+
+    if method in ("krogh", "piecewise_polynomial", "pchip"):
+        if not index.is_monotonic_increasing:
+            raise ValueError(
+                f"{method} interpolation requires that the index be monotonic."
+            )
+
+    return method
+
+
+def find_valid_index(how: str, is_valid: npt.NDArray[np.bool_]) -> int | None:
+    """
+    Retrieves the positional index of the first valid value.
+
+    Parameters
+    ----------
+    how : {'first', 'last'}
+        Use this parameter to change between the first or last valid index.
+    is_valid: np.ndarray
+        Mask to find na_values.
+
+    Returns
+    -------
+    int or None
+    """
+    assert how in ["first", "last"]
+
+    if len(is_valid) == 0:  # early stop
+        return None
+
+    if is_valid.ndim == 2:
+        is_valid = is_valid.any(axis=1)  # reduce axis 1
+
+    if how == "first":
+        idxpos = is_valid[::].argmax()
+
+    elif how == "last":
+        idxpos = len(is_valid) - 1 - is_valid[::-1].argmax()
+
+    chk_notna = is_valid[idxpos]
+
+    if not chk_notna:
+        return None
+    # Incompatible return value type (got "signedinteger[Any]",
+    # expected "Optional[int]")
+    return idxpos  # type: ignore[return-value]
+
+
+def validate_limit_direction(
+    limit_direction: str,
+) -> Literal["forward", "backward", "both"]:
+    valid_limit_directions = ["forward", "backward", "both"]
+    limit_direction = limit_direction.lower()
+    if limit_direction not in valid_limit_directions:
+        raise ValueError(
+            "Invalid limit_direction: expecting one of "
+            f"{valid_limit_directions}, got '{limit_direction}'."
+        )
+    # error: Incompatible return value type (got "str", expected
+    # "Literal['forward', 'backward', 'both']")
+    return limit_direction  # type: ignore[return-value]
+
+
+def validate_limit_area(limit_area: str | None) -> Literal["inside", "outside"] | None:
+    if limit_area is not None:
+        valid_limit_areas = ["inside", "outside"]
+        limit_area = limit_area.lower()
+        if limit_area not in valid_limit_areas:
+            raise ValueError(
+                f"Invalid limit_area: expecting one of {valid_limit_areas}, got "
+                f"{limit_area}."
+            )
+    # error: Incompatible return value type (got "Optional[str]", expected
+    # "Optional[Literal['inside', 'outside']]")
+    return limit_area  # type: ignore[return-value]
+
+
+def infer_limit_direction(
+    limit_direction: Literal["backward", "forward", "both"] | None, method: str
+) -> Literal["backward", "forward", "both"]:
+    # Set `limit_direction` depending on `method`
+    if limit_direction is None:
+        if method in ("backfill", "bfill"):
+            limit_direction = "backward"
+        else:
+            limit_direction = "forward"
+    else:
+        if method in ("pad", "ffill") and limit_direction != "forward":
+            raise ValueError(
+                f"`limit_direction` must be 'forward' for method `{method}`"
+            )
+        if method in ("backfill", "bfill") and limit_direction != "backward":
+            raise ValueError(
+                f"`limit_direction` must be 'backward' for method `{method}`"
+            )
+    return limit_direction
+
+
+def get_interp_index(method, index: Index) -> Index:
+    # create/use the index
+    if method == "linear":
+        # prior default
+        from pandas import Index
+
+        if isinstance(index.dtype, DatetimeTZDtype) or lib.is_np_dtype(
+            index.dtype, "mM"
+        ):
+            # Convert datetime-like indexes to int64
+            index = Index(index.view("i8"))
+
+        elif not is_numeric_dtype(index.dtype):
+            # We keep behavior consistent with prior versions of pandas for
+            # non-numeric, non-datetime indexes
+            index = Index(range(len(index)))
+    else:
+        methods = {"index", "values", "nearest", "time"}
+        is_numeric_or_datetime = (
+            is_numeric_dtype(index.dtype)
+            or isinstance(index.dtype, DatetimeTZDtype)
+            or lib.is_np_dtype(index.dtype, "mM")
+        )
+        valid = NP_METHODS + SP_METHODS
+        if method in valid:
+            if method not in methods and not is_numeric_or_datetime:
+                raise ValueError(
+                    "Index column must be numeric or datetime type when "
+                    f"using {method} method other than linear. "
+                    "Try setting a numeric or datetime index column before "
+                    "interpolating."
+                )
+        else:
+            raise ValueError(f"Can not interpolate with method={method}.")
+
+    if isna(index).any():
+        raise NotImplementedError(
+            "Interpolation with NaNs in the index "
+            "has not been implemented. Try filling "
+            "those NaNs before interpolating."
+        )
+    return index
+
+
+def interpolate_2d_inplace(
+    data: np.ndarray,  # floating dtype
+    index: Index,
+    axis: AxisInt,
+    method: str = "linear",
+    limit: int | None = None,
+    limit_direction: str = "forward",
+    limit_area: str | None = None,
+    fill_value: Any | None = None,
+    mask=None,
+    **kwargs,
+) -> None:
+    """
+    Column-wise application of _interpolate_1d.
+
+    Notes
+    -----
+    Alters 'data' in-place.
+
+    The signature does differ from _interpolate_1d because it only
+    includes what is needed for Block.interpolate.
+    """
+    # validate the interp method
+    clean_interp_method(method, index, **kwargs)
+
+    if is_valid_na_for_dtype(fill_value, data.dtype):
+        fill_value = na_value_for_dtype(data.dtype, compat=False)
+
+    if method == "time":
+        if not needs_i8_conversion(index.dtype):
+            raise ValueError(
+                "time-weighted interpolation only works "
+                "on Series or DataFrames with a "
+                "DatetimeIndex"
+            )
+        method = "values"
+
+    limit_direction = validate_limit_direction(limit_direction)
+    limit_area_validated = validate_limit_area(limit_area)
+
+    # default limit is unlimited GH #16282
+    limit = algos.validate_limit(nobs=None, limit=limit)
+
+    indices = _index_to_interp_indices(index, method)
+
+    def func(yvalues: np.ndarray) -> None:
+        # process 1-d slices in the axis direction
+
+        _interpolate_1d(
+            indices=indices,
+            yvalues=yvalues,
+            method=method,
+            limit=limit,
+            limit_direction=limit_direction,
+            limit_area=limit_area_validated,
+            fill_value=fill_value,
+            bounds_error=False,
+            mask=mask,
+            **kwargs,
+        )
+
+    # error: No overload variant of "apply_along_axis" matches
+    # argument types "Callable[[ndarray[Any, Any]], None]",
+    # "int", "ndarray[Any, Any]"
+    np.apply_along_axis(func, axis, data)  # type: ignore[call-overload]
+
+
+def _index_to_interp_indices(index: Index, method: str) -> np.ndarray:
+    """
+    Convert Index to ndarray of indices to pass to NumPy/SciPy.
+    """
+    xarr = index._values
+    if needs_i8_conversion(xarr.dtype):
+        # GH#1646 for dt64tz
+        xarr = xarr.view("i8")
+
+    if method == "linear":
+        inds = xarr
+        inds = cast(np.ndarray, inds)
+    else:
+        inds = np.asarray(xarr)
+
+        if method in ("values", "index"):
+            if inds.dtype == np.object_:
+                inds = lib.maybe_convert_objects(inds)
+
+    return inds
+
+
+def _interpolate_1d(
+    indices: np.ndarray,
+    yvalues: np.ndarray,
+    method: str = "linear",
+    limit: int | None = None,
+    limit_direction: str = "forward",
+    limit_area: Literal["inside", "outside"] | None = None,
+    fill_value: Any | None = None,
+    bounds_error: bool = False,
+    order: int | None = None,
+    mask=None,
+    **kwargs,
+) -> None:
+    """
+    Logic for the 1-d interpolation.  The input
+    indices and yvalues will each be 1-d arrays of the same length.
+
+    Bounds_error is currently hardcoded to False since non-scipy ones don't
+    take it as an argument.
+
+    Notes
+    -----
+    Fills 'yvalues' in-place.
+    """
+    if mask is not None:
+        invalid = mask
+    else:
+        invalid = isna(yvalues)
+    valid = ~invalid
+
+    if not valid.any():
+        return
+
+    if valid.all():
+        return
+
+    # These index pointers to invalid values... i.e. {0, 1, etc...
+    all_nans = np.flatnonzero(invalid)
+
+    first_valid_index = find_valid_index(how="first", is_valid=valid)
+    if first_valid_index is None:  # no nan found in start
+        first_valid_index = 0
+    start_nans = np.arange(first_valid_index)
+
+    last_valid_index = find_valid_index(how="last", is_valid=valid)
+    if last_valid_index is None:  # no nan found in end
+        last_valid_index = len(yvalues)
+    end_nans = np.arange(1 + last_valid_index, len(valid))
+
+    # preserve_nans contains indices of invalid values,
+    # but in this case, it is the final set of indices that need to be
+    # preserved as NaN after the interpolation.
+
+    # For example if limit_direction='forward' then preserve_nans will
+    # contain indices of NaNs at the beginning of the series, and NaNs that
+    # are more than 'limit' away from the prior non-NaN.
+
+    # set preserve_nans based on direction using _interp_limit
+    if limit_direction == "forward":
+        preserve_nans = np.union1d(start_nans, _interp_limit(invalid, limit, 0))
+    elif limit_direction == "backward":
+        preserve_nans = np.union1d(end_nans, _interp_limit(invalid, 0, limit))
+    else:
+        # both directions... just use _interp_limit
+        preserve_nans = np.unique(_interp_limit(invalid, limit, limit))
+
+    # if limit_area is set, add either mid or outside indices
+    # to preserve_nans GH #16284
+    if limit_area == "inside":
+        # preserve NaNs on the outside
+        preserve_nans = np.union1d(preserve_nans, start_nans)
+        preserve_nans = np.union1d(preserve_nans, end_nans)
+    elif limit_area == "outside":
+        # preserve NaNs on the inside
+        mid_nans = np.setdiff1d(all_nans, start_nans, assume_unique=True)
+        mid_nans = np.setdiff1d(mid_nans, end_nans, assume_unique=True)
+        preserve_nans = np.union1d(preserve_nans, mid_nans)
+
+    is_datetimelike = yvalues.dtype.kind in "mM"
+
+    if is_datetimelike:
+        yvalues = yvalues.view("i8")
+
+    if method in NP_METHODS:
+        # np.interp requires sorted X values, #21037
+
+        indexer = np.argsort(indices[valid])
+        yvalues[invalid] = np.interp(
+            indices[invalid], indices[valid][indexer], yvalues[valid][indexer]
+        )
+    else:
+        yvalues[invalid] = _interpolate_scipy_wrapper(
+            indices[valid],
+            yvalues[valid],
+            indices[invalid],
+            method=method,
+            fill_value=fill_value,
+            bounds_error=bounds_error,
+            order=order,
+            **kwargs,
+        )
+
+    if mask is not None:
+        mask[:] = False
+        mask[preserve_nans] = True
+    elif is_datetimelike:
+        yvalues[preserve_nans] = NaT.value
+    else:
+        yvalues[preserve_nans] = np.nan
+    return
+
+
+def _interpolate_scipy_wrapper(
+    x: np.ndarray,
+    y: np.ndarray,
+    new_x: np.ndarray,
+    method: str,
+    fill_value=None,
+    bounds_error: bool = False,
+    order=None,
+    **kwargs,
+):
+    """
+    Passed off to scipy.interpolate.interp1d. method is scipy's kind.
+    Returns an array interpolated at new_x.  Add any new methods to
+    the list in _clean_interp_method.
+    """
+    extra = f"{method} interpolation requires SciPy."
+    import_optional_dependency("scipy", extra=extra)
+    from scipy import interpolate
+
+    new_x = np.asarray(new_x)
+
+    # ignores some kwargs that could be passed along.
+    alt_methods = {
+        "barycentric": interpolate.barycentric_interpolate,
+        "krogh": interpolate.krogh_interpolate,
+        "from_derivatives": _from_derivatives,
+        "piecewise_polynomial": _from_derivatives,
+        "cubicspline": _cubicspline_interpolate,
+        "akima": _akima_interpolate,
+        "pchip": interpolate.pchip_interpolate,
+    }
+
+    interp1d_methods = [
+        "nearest",
+        "zero",
+        "slinear",
+        "quadratic",
+        "cubic",
+        "polynomial",
+    ]
+    if method in interp1d_methods:
+        if method == "polynomial":
+            kind = order
+        else:
+            kind = method
+        terp = interpolate.interp1d(
+            x, y, kind=kind, fill_value=fill_value, bounds_error=bounds_error
+        )
+        new_y = terp(new_x)
+    elif method == "spline":
+        # GH #10633, #24014
+        if isna(order) or (order <= 0):
+            raise ValueError(
+                f"order needs to be specified and greater than 0; got order: {order}"
+            )
+        terp = interpolate.UnivariateSpline(x, y, k=order, **kwargs)
+        new_y = terp(new_x)
+    else:
+        # GH 7295: need to be able to write for some reason
+        # in some circumstances: check all three
+        if not x.flags.writeable:
+            x = x.copy()
+        if not y.flags.writeable:
+            y = y.copy()
+        if not new_x.flags.writeable:
+            new_x = new_x.copy()
+        terp = alt_methods.get(method, None)
+        if terp is None:
+            raise ValueError(f"Can not interpolate with method={method}.")
+
+        # Make sure downcast is not in kwargs for alt methods
+        kwargs.pop("downcast", None)
+        new_y = terp(x, y, new_x, **kwargs)
+    return new_y
+
+
+def _from_derivatives(
+    xi: np.ndarray,
+    yi: np.ndarray,
+    x: np.ndarray,
+    order=None,
+    der: int | list[int] | None = 0,
+    extrapolate: bool = False,
+):
+    """
+    Convenience function for interpolate.BPoly.from_derivatives.
+
+    Construct a piecewise polynomial in the Bernstein basis, compatible
+    with the specified values and derivatives at breakpoints.
+
+    Parameters
+    ----------
+    xi : array-like
+        sorted 1D array of x-coordinates
+    yi : array-like or list of array-likes
+        yi[i][j] is the j-th derivative known at xi[i]
+    order: None or int or array-like of ints. Default: None.
+        Specifies the degree of local polynomials. If not None, some
+        derivatives are ignored.
+    der : int or list
+        How many derivatives to extract; None for all potentially nonzero
+        derivatives (that is a number equal to the number of points), or a
+        list of derivatives to extract. This number includes the function
+        value as 0th derivative.
+     extrapolate : bool, optional
+        Whether to extrapolate to ouf-of-bounds points based on first and last
+        intervals, or to return NaNs. Default: True.
+
+    See Also
+    --------
+    scipy.interpolate.BPoly.from_derivatives
+
+    Returns
+    -------
+    y : scalar or array-like
+        The result, of length R or length M or M by R.
+    """
+    from scipy import interpolate
+
+    # return the method for compat with scipy version & backwards compat
+    method = interpolate.BPoly.from_derivatives
+    m = method(xi, yi.reshape(-1, 1), orders=order, extrapolate=extrapolate)
+
+    return m(x)
+
+
+def _akima_interpolate(
+    xi: np.ndarray,
+    yi: np.ndarray,
+    x: np.ndarray,
+    der: int | list[int] | None = 0,
+    axis: AxisInt = 0,
+):
+    """
+    Convenience function for akima interpolation.
+    xi and yi are arrays of values used to approximate some function f,
+    with ``yi = f(xi)``.
+
+    See `Akima1DInterpolator` for details.
+
+    Parameters
+    ----------
+    xi : np.ndarray
+        A sorted list of x-coordinates, of length N.
+    yi : np.ndarray
+        A 1-D array of real values.  `yi`'s length along the interpolation
+        axis must be equal to the length of `xi`. If N-D array, use axis
+        parameter to select correct axis.
+    x : np.ndarray
+        Of length M.
+    der : int, optional
+        How many derivatives to extract; None for all potentially
+        nonzero derivatives (that is a number equal to the number
+        of points), or a list of derivatives to extract. This number
+        includes the function value as 0th derivative.
+    axis : int, optional
+        Axis in the yi array corresponding to the x-coordinate values.
+
+    See Also
+    --------
+    scipy.interpolate.Akima1DInterpolator
+
+    Returns
+    -------
+    y : scalar or array-like
+        The result, of length R or length M or M by R,
+
+    """
+    from scipy import interpolate
+
+    P = interpolate.Akima1DInterpolator(xi, yi, axis=axis)
+
+    return P(x, nu=der)
+
+
+def _cubicspline_interpolate(
+    xi: np.ndarray,
+    yi: np.ndarray,
+    x: np.ndarray,
+    axis: AxisInt = 0,
+    bc_type: str | tuple[Any, Any] = "not-a-knot",
+    extrapolate=None,
+):
+    """
+    Convenience function for cubic spline data interpolator.
+
+    See `scipy.interpolate.CubicSpline` for details.
+
+    Parameters
+    ----------
+    xi : np.ndarray, shape (n,)
+        1-d array containing values of the independent variable.
+        Values must be real, finite and in strictly increasing order.
+    yi : np.ndarray
+        Array containing values of the dependent variable. It can have
+        arbitrary number of dimensions, but the length along ``axis``
+        (see below) must match the length of ``x``. Values must be finite.
+    x : np.ndarray, shape (m,)
+    axis : int, optional
+        Axis along which `y` is assumed to be varying. Meaning that for
+        ``x[i]`` the corresponding values are ``np.take(y, i, axis=axis)``.
+        Default is 0.
+    bc_type : string or 2-tuple, optional
+        Boundary condition type. Two additional equations, given by the
+        boundary conditions, are required to determine all coefficients of
+        polynomials on each segment [2]_.
+        If `bc_type` is a string, then the specified condition will be applied
+        at both ends of a spline. Available conditions are:
+        * 'not-a-knot' (default): The first and second segment at a curve end
+          are the same polynomial. It is a good default when there is no
+          information on boundary conditions.
+        * 'periodic': The interpolated functions is assumed to be periodic
+          of period ``x[-1] - x[0]``. The first and last value of `y` must be
+          identical: ``y[0] == y[-1]``. This boundary condition will result in
+          ``y'[0] == y'[-1]`` and ``y''[0] == y''[-1]``.
+        * 'clamped': The first derivative at curves ends are zero. Assuming
+          a 1D `y`, ``bc_type=((1, 0.0), (1, 0.0))`` is the same condition.
+        * 'natural': The second derivative at curve ends are zero. Assuming
+          a 1D `y`, ``bc_type=((2, 0.0), (2, 0.0))`` is the same condition.
+        If `bc_type` is a 2-tuple, the first and the second value will be
+        applied at the curve start and end respectively. The tuple values can
+        be one of the previously mentioned strings (except 'periodic') or a
+        tuple `(order, deriv_values)` allowing to specify arbitrary
+        derivatives at curve ends:
+        * `order`: the derivative order, 1 or 2.
+        * `deriv_value`: array-like containing derivative values, shape must
+          be the same as `y`, excluding ``axis`` dimension. For example, if
+          `y` is 1D, then `deriv_value` must be a scalar. If `y` is 3D with
+          the shape (n0, n1, n2) and axis=2, then `deriv_value` must be 2D
+          and have the shape (n0, n1).
+    extrapolate : {bool, 'periodic', None}, optional
+        If bool, determines whether to extrapolate to out-of-bounds points
+        based on first and last intervals, or to return NaNs. If 'periodic',
+        periodic extrapolation is used. If None (default), ``extrapolate`` is
+        set to 'periodic' for ``bc_type='periodic'`` and to True otherwise.
+
+    See Also
+    --------
+    scipy.interpolate.CubicHermiteSpline
+
+    Returns
+    -------
+    y : scalar or array-like
+        The result, of shape (m,)
+
+    References
+    ----------
+    .. [1] `Cubic Spline Interpolation
+            <https://en.wikiversity.org/wiki/Cubic_Spline_Interpolation>`_
+            on Wikiversity.
+    .. [2] Carl de Boor, "A Practical Guide to Splines", Springer-Verlag, 1978.
+    """
+    from scipy import interpolate
+
+    P = interpolate.CubicSpline(
+        xi, yi, axis=axis, bc_type=bc_type, extrapolate=extrapolate
+    )
+
+    return P(x)
+
+
+def pad_or_backfill_inplace(
+    values: np.ndarray,
+    method: Literal["pad", "backfill"] = "pad",
+    axis: AxisInt = 0,
+    limit: int | None = None,
+    limit_area: Literal["inside", "outside"] | None = None,
+) -> None:
+    """
+    Perform an actual interpolation of values, values will be make 2-d if
+    needed fills inplace, returns the result.
+
+    Parameters
+    ----------
+    values: np.ndarray
+        Input array.
+    method: str, default "pad"
+        Interpolation method. Could be "bfill" or "pad"
+    axis: 0 or 1
+        Interpolation axis
+    limit: int, optional
+        Index limit on interpolation.
+    limit_area: str, optional
+        Limit area for interpolation. Can be "inside" or "outside"
+
+    Notes
+    -----
+    Modifies values in-place.
+    """
+    transf = (lambda x: x) if axis == 0 else (lambda x: x.T)
+
+    # reshape a 1 dim if needed
+    if values.ndim == 1:
+        if axis != 0:  # pragma: no cover
+            raise AssertionError("cannot interpolate on a ndim == 1 with axis != 0")
+        values = values.reshape(tuple((1,) + values.shape))
+
+    method = clean_fill_method(method)
+    tvalues = transf(values)
+
+    func = get_fill_func(method, ndim=2)
+    # _pad_2d and _backfill_2d both modify tvalues inplace
+    func(tvalues, limit=limit, limit_area=limit_area)
+
+
+def _fillna_prep(
+    values, mask: npt.NDArray[np.bool_] | None = None
+) -> npt.NDArray[np.bool_]:
+    # boilerplate for _pad_1d, _backfill_1d, _pad_2d, _backfill_2d
+
+    if mask is None:
+        mask = isna(values)
+
+    return mask
+
+
+def _datetimelike_compat(func: F) -> F:
+    """
+    Wrapper to handle datetime64 and timedelta64 dtypes.
+    """
+
+    @wraps(func)
+    def new_func(
+        values,
+        limit: int | None = None,
+        limit_area: Literal["inside", "outside"] | None = None,
+        mask=None,
+    ):
+        if needs_i8_conversion(values.dtype):
+            if mask is None:
+                # This needs to occur before casting to int64
+                mask = isna(values)
+
+            result, mask = func(
+                values.view("i8"), limit=limit, limit_area=limit_area, mask=mask
+            )
+            return result.view(values.dtype), mask
+
+        return func(values, limit=limit, limit_area=limit_area, mask=mask)
+
+    return cast(F, new_func)
+
+
+@_datetimelike_compat
+def _pad_1d(
+    values: np.ndarray,
+    limit: int | None = None,
+    limit_area: Literal["inside", "outside"] | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    mask = _fillna_prep(values, mask)
+    if limit_area is not None and not mask.all():
+        _fill_limit_area_1d(mask, limit_area)
+    algos.pad_inplace(values, mask, limit=limit)
+    return values, mask
+
+
+@_datetimelike_compat
+def _backfill_1d(
+    values: np.ndarray,
+    limit: int | None = None,
+    limit_area: Literal["inside", "outside"] | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    mask = _fillna_prep(values, mask)
+    if limit_area is not None and not mask.all():
+        _fill_limit_area_1d(mask, limit_area)
+    algos.backfill_inplace(values, mask, limit=limit)
+    return values, mask
+
+
+@_datetimelike_compat
+def _pad_2d(
+    values: np.ndarray,
+    limit: int | None = None,
+    limit_area: Literal["inside", "outside"] | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]]:
+    mask = _fillna_prep(values, mask)
+    if limit_area is not None:
+        _fill_limit_area_2d(mask, limit_area)
+
+    if values.size:
+        algos.pad_2d_inplace(values, mask, limit=limit)
+    return values, mask
+
+
+@_datetimelike_compat
+def _backfill_2d(
+    values,
+    limit: int | None = None,
+    limit_area: Literal["inside", "outside"] | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+):
+    mask = _fillna_prep(values, mask)
+    if limit_area is not None:
+        _fill_limit_area_2d(mask, limit_area)
+
+    if values.size:
+        algos.backfill_2d_inplace(values, mask, limit=limit)
+    else:
+        # for test coverage
+        pass
+    return values, mask
+
+
+def _fill_limit_area_1d(
+    mask: npt.NDArray[np.bool_], limit_area: Literal["outside", "inside"]
+) -> None:
+    """Prepare 1d mask for ffill/bfill with limit_area.
+
+    Caller is responsible for checking at least one value of mask is False.
+    When called, mask will no longer faithfully represent when
+    the corresponding are NA or not.
+
+    Parameters
+    ----------
+    mask : np.ndarray[bool, ndim=1]
+        Mask representing NA values when filling.
+    limit_area : { "outside", "inside" }
+        Whether to limit filling to outside or inside the outer most non-NA value.
+    """
+    neg_mask = ~mask
+    first = neg_mask.argmax()
+    last = len(neg_mask) - neg_mask[::-1].argmax() - 1
+    if limit_area == "inside":
+        mask[:first] = False
+        mask[last + 1 :] = False
+    elif limit_area == "outside":
+        mask[first + 1 : last] = False
+
+
+def _fill_limit_area_2d(
+    mask: npt.NDArray[np.bool_], limit_area: Literal["outside", "inside"]
+) -> None:
+    """Prepare 2d mask for ffill/bfill with limit_area.
+
+    When called, mask will no longer faithfully represent when
+    the corresponding are NA or not.
+
+    Parameters
+    ----------
+    mask : np.ndarray[bool, ndim=1]
+        Mask representing NA values when filling.
+    limit_area : { "outside", "inside" }
+        Whether to limit filling to outside or inside the outer most non-NA value.
+    """
+    neg_mask = ~mask.T
+    if limit_area == "outside":
+        # Identify inside
+        la_mask = (
+            np.maximum.accumulate(neg_mask, axis=0)
+            & np.maximum.accumulate(neg_mask[::-1], axis=0)[::-1]
+        )
+    else:
+        # Identify outside
+        la_mask = (
+            ~np.maximum.accumulate(neg_mask, axis=0)
+            | ~np.maximum.accumulate(neg_mask[::-1], axis=0)[::-1]
+        )
+    mask[la_mask.T] = False
+
+
+_fill_methods = {"pad": _pad_1d, "backfill": _backfill_1d}
+
+
+def get_fill_func(method, ndim: int = 1):
+    method = clean_fill_method(method)
+    if ndim == 1:
+        return _fill_methods[method]
+    return {"pad": _pad_2d, "backfill": _backfill_2d}[method]
+
+
+def clean_reindex_fill_method(method) -> ReindexMethod | None:
+    if method is None:
+        return None
+    return clean_fill_method(method, allow_nearest=True)
+
+
+def _interp_limit(
+    invalid: npt.NDArray[np.bool_], fw_limit: int | None, bw_limit: int | None
+) -> np.ndarray:
+    """
+    Get indexers of values that won't be filled
+    because they exceed the limits.
+
+    Parameters
+    ----------
+    invalid : np.ndarray[bool]
+    fw_limit : int or None
+        forward limit to index
+    bw_limit : int or None
+        backward limit to index
+
+    Returns
+    -------
+    set of indexers
+
+    Notes
+    -----
+    This is equivalent to the more readable, but slower
+
+    .. code-block:: python
+
+        def _interp_limit(invalid, fw_limit, bw_limit):
+            for x in np.where(invalid)[0]:
+                if invalid[max(0, x - fw_limit) : x + bw_limit + 1].all():
+                    yield x
+    """
+    # handle forward first; the backward direction is the same except
+    # 1. operate on the reversed array
+    # 2. subtract the returned indices from N - 1
+    N = len(invalid)
+    f_idx = np.array([], dtype=np.int64)
+    b_idx = np.array([], dtype=np.int64)
+    assume_unique = True
+
+    def inner(invalid, limit: int):
+        limit = min(limit, N)
+        windowed = np.lib.stride_tricks.sliding_window_view(invalid, limit + 1).all(1)
+        idx = np.union1d(
+            np.where(windowed)[0] + limit,
+            np.where((~invalid[: limit + 1]).cumsum() == 0)[0],
+        )
+        return idx
+
+    if fw_limit is not None:
+        if fw_limit == 0:
+            f_idx = np.where(invalid)[0]
+            assume_unique = False
+        else:
+            f_idx = inner(invalid, fw_limit)
+
+    if bw_limit is not None:
+        if bw_limit == 0:
+            # then we don't even need to care about backwards
+            # just use forwards
+            return f_idx
+        else:
+            b_idx = N - 1 - inner(invalid[::-1], bw_limit)
+            if fw_limit == 0:
+                return b_idx
+
+    return np.intersect1d(f_idx, b_idx, assume_unique=assume_unique)
diff --git a/pandas/pandas/core/nanops.py b/pandas/pandas/core/nanops.py
new file mode 100644
index 0000000000000..d1dc0ff809497
--- /dev/null
+++ b/pandas/pandas/core/nanops.py
@@ -0,0 +1,1752 @@
+from __future__ import annotations
+
+import functools
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs import (
+    NaT,
+    NaTType,
+    iNaT,
+    lib,
+)
+from pandas._typing import (
+    ArrayLike,
+    AxisInt,
+    CorrelationMethod,
+    Dtype,
+    DtypeObj,
+    F,
+    Scalar,
+    Shape,
+    npt,
+)
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.dtypes.common import (
+    is_complex,
+    is_float,
+    is_float_dtype,
+    is_integer,
+    is_numeric_dtype,
+    is_object_dtype,
+    needs_i8_conversion,
+    pandas_dtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+    notna,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+bn = import_optional_dependency("bottleneck", errors="warn")
+_BOTTLENECK_INSTALLED = bn is not None
+_USE_BOTTLENECK = False
+
+
+def set_use_bottleneck(v: bool = True) -> None:
+    # set/unset to use bottleneck
+    global _USE_BOTTLENECK
+    if _BOTTLENECK_INSTALLED:
+        _USE_BOTTLENECK = v
+
+
+set_use_bottleneck(get_option("compute.use_bottleneck"))
+
+
+class disallow:
+    def __init__(self, *dtypes: Dtype) -> None:
+        super().__init__()
+        self.dtypes = tuple(pandas_dtype(dtype).type for dtype in dtypes)
+
+    def check(self, obj) -> bool:
+        return hasattr(obj, "dtype") and issubclass(obj.dtype.type, self.dtypes)
+
+    def __call__(self, f: F) -> F:
+        @functools.wraps(f)
+        def _f(*args, **kwargs):
+            obj_iter = itertools.chain(args, kwargs.values())
+            if any(self.check(obj) for obj in obj_iter):
+                f_name = f.__name__.replace("nan", "")
+                raise TypeError(
+                    f"reduction operation '{f_name}' not allowed for this dtype"
+                )
+            try:
+                return f(*args, **kwargs)
+            except ValueError as e:
+                # we want to transform an object array
+                # ValueError message to the more typical TypeError
+                # e.g. this is normally a disallowed function on
+                # object arrays that contain strings
+                if is_object_dtype(args[0]):
+                    raise TypeError(e) from e
+                raise
+
+        return cast(F, _f)
+
+
+class bottleneck_switch:
+    def __init__(self, name=None, **kwargs) -> None:
+        self.name = name
+        self.kwargs = kwargs
+
+    def __call__(self, alt: F) -> F:
+        bn_name = self.name or alt.__name__
+
+        try:
+            bn_func = getattr(bn, bn_name)
+        except (AttributeError, NameError):  # pragma: no cover
+            bn_func = None
+
+        @functools.wraps(alt)
+        def f(
+            values: np.ndarray,
+            *,
+            axis: AxisInt | None = None,
+            skipna: bool = True,
+            **kwds,
+        ):
+            if len(self.kwargs) > 0:
+                for k, v in self.kwargs.items():
+                    if k not in kwds:
+                        kwds[k] = v
+
+            if values.size == 0 and kwds.get("min_count") is None:
+                # We are empty, returning NA for our type
+                # Only applies for the default `min_count` of None
+                # since that affects how empty arrays are handled.
+                # TODO(GH-18976) update all the nanops methods to
+                # correctly handle empty inputs and remove this check.
+                # It *may* just be `var`
+                return _na_for_min_count(values, axis)
+
+            if _USE_BOTTLENECK and skipna and _bn_ok_dtype(values.dtype, bn_name):
+                if kwds.get("mask", None) is None:
+                    # `mask` is not recognised by bottleneck, would raise
+                    #  TypeError if called
+                    kwds.pop("mask", None)
+                    result = bn_func(values, axis=axis, **kwds)
+
+                    # prefer to treat inf/-inf as NA, but must compute the func
+                    # twice :(
+                    if _has_infs(result):
+                        result = alt(values, axis=axis, skipna=skipna, **kwds)
+                else:
+                    result = alt(values, axis=axis, skipna=skipna, **kwds)
+            else:
+                result = alt(values, axis=axis, skipna=skipna, **kwds)
+
+            return result
+
+        return cast(F, f)
+
+
+def _bn_ok_dtype(dtype: DtypeObj, name: str) -> bool:
+    # Bottleneck chokes on datetime64, PeriodDtype (or and EA)
+    if dtype != object and not needs_i8_conversion(dtype):
+        # GH 42878
+        # Bottleneck uses naive summation leading to O(n) loss of precision
+        # unlike numpy which implements pairwise summation, which has O(log(n)) loss
+        # crossref: https://github.com/pydata/bottleneck/issues/379
+
+        # GH 15507
+        # bottleneck does not properly upcast during the sum
+        # so can overflow
+
+        # GH 9422
+        # further we also want to preserve NaN when all elements
+        # are NaN, unlike bottleneck/numpy which consider this
+        # to be 0
+        return name not in ["nansum", "nanprod", "nanmean"]
+    return False
+
+
+def _has_infs(result) -> bool:
+    if isinstance(result, np.ndarray):
+        if result.dtype in ("f8", "f4"):
+            # Note: outside of an nanops-specific test, we always have
+            #  result.ndim == 1, so there is no risk of this ravel making a copy.
+            return lib.has_infs(result.ravel("K"))
+    try:
+        return np.isinf(result).any()
+    except (TypeError, NotImplementedError):
+        # if it doesn't support infs, then it can't have infs
+        return False
+
+
+def _get_fill_value(
+    dtype: DtypeObj, fill_value: Scalar | None = None, fill_value_typ=None
+):
+    """return the correct fill value for the dtype of the values"""
+    if fill_value is not None:
+        return fill_value
+    if _na_ok_dtype(dtype):
+        if fill_value_typ is None:
+            return np.nan
+        else:
+            if fill_value_typ == "+inf":
+                return np.inf
+            else:
+                return -np.inf
+    else:
+        if fill_value_typ == "+inf":
+            # need the max int here
+            return lib.i8max
+        else:
+            return iNaT
+
+
+def _maybe_get_mask(
+    values: np.ndarray, skipna: bool, mask: npt.NDArray[np.bool_] | None
+) -> npt.NDArray[np.bool_] | None:
+    """
+    Compute a mask if and only if necessary.
+
+    This function will compute a mask iff it is necessary. Otherwise,
+    return the provided mask (potentially None) when a mask does not need to be
+    computed.
+
+    A mask is never necessary if the values array is of boolean or integer
+    dtypes, as these are incapable of storing NaNs. If passing a NaN-capable
+    dtype that is interpretable as either boolean or integer data (eg,
+    timedelta64), a mask must be provided.
+
+    If the skipna parameter is False, a new mask will not be computed.
+
+    The mask is computed using isna() by default. Setting invert=True selects
+    notna() as the masking function.
+
+    Parameters
+    ----------
+    values : ndarray
+        input array to potentially compute mask for
+    skipna : bool
+        boolean for whether NaNs should be skipped
+    mask : Optional[ndarray]
+        nan-mask if known
+
+    Returns
+    -------
+    Optional[np.ndarray[bool]]
+    """
+    if mask is None:
+        if values.dtype.kind in "biu":
+            # Boolean data cannot contain nulls, so signal via mask being None
+            return None
+
+        if skipna or values.dtype.kind in "mM":
+            mask = isna(values)
+
+    return mask
+
+
+def _get_values(
+    values: np.ndarray,
+    skipna: bool,
+    fill_value: Any = None,
+    fill_value_typ: str | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> tuple[np.ndarray, npt.NDArray[np.bool_] | None]:
+    """
+    Utility to get the values view, mask, dtype, dtype_max, and fill_value.
+
+    If both mask and fill_value/fill_value_typ are not None and skipna is True,
+    the values array will be copied.
+
+    For input arrays of boolean or integer dtypes, copies will only occur if a
+    precomputed mask, a fill_value/fill_value_typ, and skipna=True are
+    provided.
+
+    Parameters
+    ----------
+    values : ndarray
+        input array to potentially compute mask for
+    skipna : bool
+        boolean for whether NaNs should be skipped
+    fill_value : Any
+        value to fill NaNs with
+    fill_value_typ : str
+        Set to '+inf' or '-inf' to handle dtype-specific infinities
+    mask : Optional[np.ndarray[bool]]
+        nan-mask if known
+
+    Returns
+    -------
+    values : ndarray
+        Potential copy of input value array
+    mask : Optional[ndarray[bool]]
+        Mask for values, if deemed necessary to compute
+    """
+    # In _get_values is only called from within nanops, and in all cases
+    #  with scalar fill_value.  This guarantee is important for the
+    #  np.where call below
+
+    mask = _maybe_get_mask(values, skipna, mask)
+
+    dtype = values.dtype
+
+    datetimelike = False
+    if values.dtype.kind in "mM":
+        # changing timedelta64/datetime64 to int64 needs to happen after
+        #  finding `mask` above
+        values = np.asarray(values.view("i8"))
+        datetimelike = True
+
+    if skipna and (mask is not None):
+        # get our fill value (in case we need to provide an alternative
+        # dtype for it)
+        fill_value = _get_fill_value(
+            dtype, fill_value=fill_value, fill_value_typ=fill_value_typ
+        )
+
+        if fill_value is not None:
+            if mask.any():
+                if datetimelike or _na_ok_dtype(dtype):
+                    values = values.copy()
+                    np.putmask(values, mask, fill_value)
+                else:
+                    # np.where will promote if needed
+                    values = np.where(~mask, values, fill_value)
+
+    return values, mask
+
+
+def _get_dtype_max(dtype: np.dtype) -> np.dtype:
+    # return a platform independent precision dtype
+    dtype_max = dtype
+    if dtype.kind in "bi":
+        dtype_max = np.dtype(np.int64)
+    elif dtype.kind == "u":
+        dtype_max = np.dtype(np.uint64)
+    elif dtype.kind == "f":
+        dtype_max = np.dtype(np.float64)
+    return dtype_max
+
+
+def _na_ok_dtype(dtype: DtypeObj) -> bool:
+    if needs_i8_conversion(dtype):
+        return False
+    return not issubclass(dtype.type, np.integer)
+
+
+def _wrap_results(result, dtype: np.dtype, fill_value=None):
+    """wrap our results if needed"""
+    if result is NaT:
+        pass
+
+    elif dtype.kind == "M":
+        if fill_value is None:
+            # GH#24293
+            fill_value = iNaT
+        if not isinstance(result, np.ndarray):
+            assert not isna(fill_value), "Expected non-null fill_value"
+            if result == fill_value:
+                result = np.nan
+
+            if isna(result):
+                result = np.datetime64("NaT", "ns").astype(dtype)
+            else:
+                result = np.int64(result).view(dtype)
+            # retain original unit
+            result = result.astype(dtype, copy=False)
+        else:
+            # If we have float dtype, taking a view will give the wrong result
+            result = result.astype(dtype)
+    elif dtype.kind == "m":
+        if not isinstance(result, np.ndarray):
+            if result == fill_value or np.isnan(result):
+                result = np.timedelta64("NaT").astype(dtype)
+
+            elif np.fabs(result) > lib.i8max:
+                # raise if we have a timedelta64[ns] which is too large
+                raise ValueError("overflow in timedelta operation")
+            else:
+                # return a timedelta64 with the original unit
+                result = np.int64(result).astype(dtype, copy=False)
+
+        else:
+            result = result.astype("m8[ns]").view(dtype)
+
+    return result
+
+
+def _datetimelike_compat(func: F) -> F:
+    """
+    If we have datetime64 or timedelta64 values, ensure we have a correct
+    mask before calling the wrapped function, then cast back afterwards.
+    """
+
+    @functools.wraps(func)
+    def new_func(
+        values: np.ndarray,
+        *,
+        axis: AxisInt | None = None,
+        skipna: bool = True,
+        mask: npt.NDArray[np.bool_] | None = None,
+        **kwargs,
+    ):
+        orig_values = values
+
+        datetimelike = values.dtype.kind in "mM"
+        if datetimelike and mask is None:
+            mask = isna(values)
+
+        result = func(values, axis=axis, skipna=skipna, mask=mask, **kwargs)
+
+        if datetimelike:
+            result = _wrap_results(result, orig_values.dtype, fill_value=iNaT)
+            if not skipna:
+                assert mask is not None  # checked above
+                result = _mask_datetimelike_result(result, axis, mask, orig_values)
+
+        return result
+
+    return cast(F, new_func)
+
+
+def _na_for_min_count(values: np.ndarray, axis: AxisInt | None) -> Scalar | np.ndarray:
+    """
+    Return the missing value for `values`.
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int or None
+        axis for the reduction, required if values.ndim > 1.
+
+    Returns
+    -------
+    result : scalar or ndarray
+        For 1-D values, returns a scalar of the correct missing type.
+        For 2-D values, returns a 1-D array where each element is missing.
+    """
+    # we either return np.nan or pd.NaT
+    if values.dtype.kind in "iufcb":
+        values = values.astype("float64")
+    fill_value = na_value_for_dtype(values.dtype)
+
+    if values.ndim == 1:
+        return fill_value
+    elif axis is None:
+        return fill_value
+    else:
+        result_shape = values.shape[:axis] + values.shape[axis + 1 :]
+
+        return np.full(result_shape, fill_value, dtype=values.dtype)
+
+
+def maybe_operate_rowwise(func: F) -> F:
+    """
+    NumPy operations on C-contiguous ndarrays with axis=1 can be
+    very slow if axis 1 >> axis 0.
+    Operate row-by-row and concatenate the results.
+    """
+
+    @functools.wraps(func)
+    def newfunc(values: np.ndarray, *, axis: AxisInt | None = None, **kwargs):
+        if (
+            axis == 1
+            and values.ndim == 2
+            and values.flags["C_CONTIGUOUS"]
+            # only takes this path for wide arrays (long dataframes), for threshold see
+            # https://github.com/pandas-dev/pandas/pull/43311#issuecomment-974891737
+            and (values.shape[1] / 1000) > values.shape[0]
+            and values.dtype != object
+            and values.dtype != bool
+        ):
+            arrs = list(values)
+            if kwargs.get("mask") is not None:
+                mask = kwargs.pop("mask")
+                results = [
+                    func(arrs[i], mask=mask[i], **kwargs) for i in range(len(arrs))
+                ]
+            else:
+                results = [func(x, **kwargs) for x in arrs]
+            return np.array(results)
+
+        return func(values, axis=axis, **kwargs)
+
+    return cast(F, newfunc)
+
+
+def nanany(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> bool:
+    """
+    Check if any elements along an axis evaluate to True.
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : bool
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 2])
+    >>> nanops.nanany(s.values)
+    True
+
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([np.nan])
+    >>> nanops.nanany(s.values)
+    False
+    """
+    if values.dtype.kind in "iub" and mask is None:
+        # GH#26032 fastpath
+        # error: Incompatible return value type (got "Union[bool_, ndarray]",
+        # expected "bool")
+        return values.any(axis)  # type: ignore[return-value]
+
+    if values.dtype.kind == "M":
+        # GH#34479
+        raise TypeError("datetime64 type does not support operation 'any'")
+
+    values, _ = _get_values(values, skipna, fill_value=False, mask=mask)
+
+    # For object type, any won't necessarily return
+    # boolean values (numpy/numpy#4352)
+    if values.dtype == object:
+        values = values.astype(bool)
+
+    # error: Incompatible return value type (got "Union[bool_, ndarray]", expected
+    # "bool")
+    return values.any(axis)  # type: ignore[return-value]
+
+
+def nanall(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> bool:
+    """
+    Check if all elements along an axis evaluate to True.
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : bool
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 2, np.nan])
+    >>> nanops.nanall(s.values)
+    True
+
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 0])
+    >>> nanops.nanall(s.values)
+    False
+    """
+    if values.dtype.kind in "iub" and mask is None:
+        # GH#26032 fastpath
+        # error: Incompatible return value type (got "Union[bool_, ndarray]",
+        # expected "bool")
+        return values.all(axis)  # type: ignore[return-value]
+
+    if values.dtype.kind == "M":
+        # GH#34479
+        raise TypeError("datetime64 type does not support operation 'all'")
+
+    values, _ = _get_values(values, skipna, fill_value=True, mask=mask)
+
+    # For object type, all won't necessarily return
+    # boolean values (numpy/numpy#4352)
+    if values.dtype == object:
+        values = values.astype(bool)
+
+    # error: Incompatible return value type (got "Union[bool_, ndarray]", expected
+    # "bool")
+    return values.all(axis)  # type: ignore[return-value]
+
+
+@disallow("M8")
+@_datetimelike_compat
+@maybe_operate_rowwise
+def nansum(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    min_count: int = 0,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Sum the elements along an axis ignoring NaNs
+
+    Parameters
+    ----------
+    values : ndarray[dtype]
+    axis : int, optional
+    skipna : bool, default True
+    min_count: int, default 0
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : dtype
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 2, np.nan])
+    >>> nanops.nansum(s.values)
+    3.0
+    """
+    dtype = values.dtype
+    values, mask = _get_values(values, skipna, fill_value=0, mask=mask)
+    dtype_sum = _get_dtype_max(dtype)
+    if dtype.kind == "f":
+        dtype_sum = dtype
+    elif dtype.kind == "m":
+        dtype_sum = np.dtype(np.float64)
+
+    the_sum = values.sum(axis, dtype=dtype_sum)
+    the_sum = _maybe_null_out(the_sum, axis, mask, values.shape, min_count=min_count)
+
+    return the_sum
+
+
+def _mask_datetimelike_result(
+    result: np.ndarray | np.datetime64 | np.timedelta64,
+    axis: AxisInt | None,
+    mask: npt.NDArray[np.bool_],
+    orig_values: np.ndarray,
+) -> np.ndarray | np.datetime64 | np.timedelta64 | NaTType:
+    if isinstance(result, np.ndarray):
+        # we need to apply the mask
+        result = result.astype("i8").view(orig_values.dtype)
+        axis_mask = mask.any(axis=axis)
+        # error: Unsupported target for indexed assignment ("Union[ndarray[Any, Any],
+        # datetime64, timedelta64]")
+        result[axis_mask] = iNaT  # type: ignore[index]
+    else:
+        if mask.any():
+            return np.int64(iNaT).view(orig_values.dtype)
+    return result
+
+
+@bottleneck_switch()
+@_datetimelike_compat
+def nanmean(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Compute the mean of the element along an axis ignoring NaNs
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    float
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 2, np.nan])
+    >>> nanops.nanmean(s.values)
+    1.5
+    """
+    dtype = values.dtype
+    values, mask = _get_values(values, skipna, fill_value=0, mask=mask)
+    dtype_sum = _get_dtype_max(dtype)
+    dtype_count = np.dtype(np.float64)
+
+    # not using needs_i8_conversion because that includes period
+    if dtype.kind in "mM":
+        dtype_sum = np.dtype(np.float64)
+    elif dtype.kind in "iu":
+        dtype_sum = np.dtype(np.float64)
+    elif dtype.kind == "f":
+        dtype_sum = dtype
+        dtype_count = dtype
+
+    count = _get_counts(values.shape, mask, axis, dtype=dtype_count)
+    the_sum = values.sum(axis, dtype=dtype_sum)
+    the_sum = _ensure_numeric(the_sum)
+
+    if axis is not None and getattr(the_sum, "ndim", False):
+        count = cast(np.ndarray, count)
+        with np.errstate(all="ignore"):
+            # suppress division by zero warnings
+            the_mean = the_sum / count
+        ct_mask = count == 0
+        if ct_mask.any():
+            the_mean[ct_mask] = np.nan
+    else:
+        the_mean = the_sum / count if count > 0 else np.nan
+
+    return the_mean
+
+
+@bottleneck_switch()
+def nanmedian(
+    values: np.ndarray, *, axis: AxisInt | None = None, skipna: bool = True, mask=None
+) -> float | np.ndarray:
+    """
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float | ndarray
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 2, 2])
+    >>> nanops.nanmedian(s.values)
+    2.0
+
+    >>> s = pd.Series([np.nan, np.nan, np.nan])
+    >>> nanops.nanmedian(s.values)
+    nan
+    """
+    # for floats without mask, the data already uses NaN as missing value
+    # indicator, and `mask` will be calculated from that below -> in those
+    # cases we never need to set NaN to the masked values
+    using_nan_sentinel = values.dtype.kind == "f" and mask is None
+
+    def get_median(x: np.ndarray, _mask=None):
+        if _mask is None:
+            _mask = notna(x)
+        else:
+            _mask = ~_mask
+        if not skipna and not _mask.all():
+            return np.nan
+        with warnings.catch_warnings():
+            # Suppress RuntimeWarning about All-NaN slice
+            warnings.filterwarnings(
+                "ignore", "All-NaN slice encountered", RuntimeWarning
+            )
+            warnings.filterwarnings("ignore", "Mean of empty slice", RuntimeWarning)
+            res = np.nanmedian(x[_mask])
+        return res
+
+    dtype = values.dtype
+    values, mask = _get_values(values, skipna, mask=mask, fill_value=None)
+    if values.dtype.kind != "f":
+        if values.dtype == object:
+            # GH#34671 avoid casting strings to numeric
+            inferred = lib.infer_dtype(values)
+            if inferred in ["string", "mixed"]:
+                raise TypeError(f"Cannot convert {values} to numeric")
+        try:
+            values = values.astype("f8")
+        except ValueError as err:
+            # e.g. "could not convert string to float: 'a'"
+            raise TypeError(str(err)) from err
+    if not using_nan_sentinel and mask is not None:
+        if not values.flags.writeable:
+            values = values.copy()
+        values[mask] = np.nan
+
+    notempty = values.size
+
+    res: float | np.ndarray
+
+    # an array from a frame
+    if values.ndim > 1 and axis is not None:
+        # there's a non-empty array to apply over otherwise numpy raises
+        if notempty:
+            if not skipna:
+                res = np.apply_along_axis(get_median, axis, values)
+
+            else:
+                # fastpath for the skipna case
+                with warnings.catch_warnings():
+                    # Suppress RuntimeWarning about All-NaN slice
+                    warnings.filterwarnings(
+                        "ignore", "All-NaN slice encountered", RuntimeWarning
+                    )
+                    if (values.shape[1] == 1 and axis == 0) or (
+                        values.shape[0] == 1 and axis == 1
+                    ):
+                        # GH52788: fastpath when squeezable, nanmedian for 2D array slow
+                        res = np.nanmedian(np.squeeze(values), keepdims=True)
+                    else:
+                        res = np.nanmedian(values, axis=axis)
+
+        else:
+            # must return the correct shape, but median is not defined for the
+            # empty set so return nans of shape "everything but the passed axis"
+            # since "axis" is where the reduction would occur if we had a nonempty
+            # array
+            res = _get_empty_reduction_result(values.shape, axis)
+
+    else:
+        # otherwise return a scalar value
+        res = get_median(values, mask) if notempty else np.nan
+    return _wrap_results(res, dtype)
+
+
+def _get_empty_reduction_result(
+    shape: Shape,
+    axis: AxisInt,
+) -> np.ndarray:
+    """
+    The result from a reduction on an empty ndarray.
+
+    Parameters
+    ----------
+    shape : Tuple[int, ...]
+    axis : int
+
+    Returns
+    -------
+    np.ndarray
+    """
+    shp = np.array(shape)
+    dims = np.arange(len(shape))
+    ret = np.empty(shp[dims != axis], dtype=np.float64)
+    ret.fill(np.nan)
+    return ret
+
+
+def _get_counts_nanvar(
+    values_shape: Shape,
+    mask: npt.NDArray[np.bool_] | None,
+    axis: AxisInt | None,
+    ddof: int,
+    dtype: np.dtype = np.dtype(np.float64),
+) -> tuple[float | np.ndarray, float | np.ndarray]:
+    """
+    Get the count of non-null values along an axis, accounting
+    for degrees of freedom.
+
+    Parameters
+    ----------
+    values_shape : Tuple[int, ...]
+        shape tuple from values ndarray, used if mask is None
+    mask : Optional[ndarray[bool]]
+        locations in values that should be considered missing
+    axis : Optional[int]
+        axis to count along
+    ddof : int
+        degrees of freedom
+    dtype : type, optional
+        type to use for count
+
+    Returns
+    -------
+    count : int, np.nan or np.ndarray
+    d : int, np.nan or np.ndarray
+    """
+    count = _get_counts(values_shape, mask, axis, dtype=dtype)
+    d = count - dtype.type(ddof)
+
+    # always return NaN, never inf
+    if is_float(count):
+        if count <= ddof:
+            # error: Incompatible types in assignment (expression has type
+            # "float", variable has type "Union[floating[Any], ndarray[Any,
+            # dtype[floating[Any]]]]")
+            count = np.nan  # type: ignore[assignment]
+            d = np.nan
+    else:
+        # count is not narrowed by is_float check
+        count = cast(np.ndarray, count)
+        mask = count <= ddof
+        if mask.any():
+            np.putmask(d, mask, np.nan)
+            np.putmask(count, mask, np.nan)
+    return count, d
+
+
+@bottleneck_switch(ddof=1)
+def nanstd(
+    values,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    ddof: int = 1,
+    mask=None,
+):
+    """
+    Compute the standard deviation along given axis while ignoring NaNs
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    ddof : int, default 1
+        Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+        where N represents the number of elements.
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 2, 3])
+    >>> nanops.nanstd(s.values)
+    1.0
+    """
+    if values.dtype == "M8[ns]":
+        values = values.view("m8[ns]")
+
+    orig_dtype = values.dtype
+    values, mask = _get_values(values, skipna, mask=mask)
+
+    result = np.sqrt(nanvar(values, axis=axis, skipna=skipna, ddof=ddof, mask=mask))
+    return _wrap_results(result, orig_dtype)
+
+
+@disallow("M8", "m8")
+@bottleneck_switch(ddof=1)
+def nanvar(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    ddof: int = 1,
+    mask=None,
+):
+    """
+    Compute the variance along given axis while ignoring NaNs
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    ddof : int, default 1
+        Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+        where N represents the number of elements.
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 2, 3])
+    >>> nanops.nanvar(s.values)
+    1.0
+    """
+    dtype = values.dtype
+    mask = _maybe_get_mask(values, skipna, mask)
+    if dtype.kind in "iu":
+        values = values.astype("f8")
+        if mask is not None:
+            values[mask] = np.nan
+
+    if values.dtype.kind == "f":
+        count, d = _get_counts_nanvar(values.shape, mask, axis, ddof, values.dtype)
+    else:
+        count, d = _get_counts_nanvar(values.shape, mask, axis, ddof)
+
+    if skipna and mask is not None:
+        values = values.copy()
+        np.putmask(values, mask, 0)
+
+    # xref GH10242
+    # Compute variance via two-pass algorithm, which is stable against
+    # cancellation errors and relatively accurate for small numbers of
+    # observations.
+    #
+    # See https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+    avg = _ensure_numeric(values.sum(axis=axis, dtype=np.float64)) / count
+    if axis is not None:
+        avg = np.expand_dims(avg, axis)
+    sqr = _ensure_numeric((avg - values) ** 2)
+    if mask is not None:
+        np.putmask(sqr, mask, 0)
+    result = sqr.sum(axis=axis, dtype=np.float64) / d
+
+    # Return variance as np.float64 (the datatype used in the accumulator),
+    # unless we were dealing with a float array, in which case use the same
+    # precision as the original values array.
+    if dtype.kind == "f":
+        result = result.astype(dtype, copy=False)
+    return result
+
+
+@disallow("M8", "m8")
+def nansem(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    ddof: int = 1,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Compute the standard error in the mean along given axis while ignoring NaNs
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    ddof : int, default 1
+        Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+        where N represents the number of elements.
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float64
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 2, 3])
+    >>> nanops.nansem(s.values)
+     0.5773502691896258
+    """
+    # This checks if non-numeric-like data is passed with numeric_only=False
+    # and raises a TypeError otherwise
+    nanvar(values, axis=axis, skipna=skipna, ddof=ddof, mask=mask)
+
+    mask = _maybe_get_mask(values, skipna, mask)
+    if values.dtype.kind != "f":
+        values = values.astype("f8")
+
+    if not skipna and mask is not None and mask.any():
+        return np.nan
+
+    count, _ = _get_counts_nanvar(values.shape, mask, axis, ddof, values.dtype)
+    var = nanvar(values, axis=axis, skipna=skipna, ddof=ddof, mask=mask)
+
+    return np.sqrt(var) / np.sqrt(count)
+
+
+def _nanminmax(meth, fill_value_typ):
+    @bottleneck_switch(name=f"nan{meth}")
+    @_datetimelike_compat
+    def reduction(
+        values: np.ndarray,
+        *,
+        axis: AxisInt | None = None,
+        skipna: bool = True,
+        mask: npt.NDArray[np.bool_] | None = None,
+    ):
+        if values.size == 0:
+            return _na_for_min_count(values, axis)
+
+        dtype = values.dtype
+        values, mask = _get_values(
+            values, skipna, fill_value_typ=fill_value_typ, mask=mask
+        )
+        result = getattr(values, meth)(axis)
+        result = _maybe_null_out(
+            result, axis, mask, values.shape, datetimelike=dtype.kind in "mM"
+        )
+        return result
+
+    return reduction
+
+
+nanmin = _nanminmax("min", fill_value_typ="+inf")
+nanmax = _nanminmax("max", fill_value_typ="-inf")
+
+
+def nanargmax(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> int | np.ndarray:
+    """
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : int or ndarray[int]
+        The index/indices  of max value in specified axis or -1 in the NA case
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> arr = np.array([1, 2, 3, np.nan, 4])
+    >>> nanops.nanargmax(arr)
+    4
+
+    >>> arr = np.array(range(12), dtype=np.float64).reshape(4, 3)
+    >>> arr[2:, 2] = np.nan
+    >>> arr
+    array([[ 0.,  1.,  2.],
+           [ 3.,  4.,  5.],
+           [ 6.,  7., nan],
+           [ 9., 10., nan]])
+    >>> nanops.nanargmax(arr, axis=1)
+    array([2, 2, 1, 1])
+    """
+    values, mask = _get_values(values, True, fill_value_typ="-inf", mask=mask)
+    result = values.argmax(axis)
+    # error: Argument 1 to "_maybe_arg_null_out" has incompatible type "Any |
+    # signedinteger[Any]"; expected "ndarray[Any, Any]"
+    result = _maybe_arg_null_out(result, axis, mask, skipna)  # type: ignore[arg-type]
+    return result
+
+
+def nanargmin(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> int | np.ndarray:
+    """
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : int or ndarray[int]
+        The index/indices of min value in specified axis or -1 in the NA case
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> arr = np.array([1, 2, 3, np.nan, 4])
+    >>> nanops.nanargmin(arr)
+    0
+
+    >>> arr = np.array(range(12), dtype=np.float64).reshape(4, 3)
+    >>> arr[2:, 0] = np.nan
+    >>> arr
+    array([[ 0.,  1.,  2.],
+           [ 3.,  4.,  5.],
+           [nan,  7.,  8.],
+           [nan, 10., 11.]])
+    >>> nanops.nanargmin(arr, axis=1)
+    array([0, 0, 1, 1])
+    """
+    values, mask = _get_values(values, True, fill_value_typ="+inf", mask=mask)
+    result = values.argmin(axis)
+    # error: Argument 1 to "_maybe_arg_null_out" has incompatible type "Any |
+    # signedinteger[Any]"; expected "ndarray[Any, Any]"
+    result = _maybe_arg_null_out(result, axis, mask, skipna)  # type: ignore[arg-type]
+    return result
+
+
+@disallow("M8", "m8")
+@maybe_operate_rowwise
+def nanskew(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Compute the sample skewness.
+
+    The statistic computed here is the adjusted Fisher-Pearson standardized
+    moment coefficient G1. The algorithm computes this coefficient directly
+    from the second and third central moment.
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float64
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 1, 2])
+    >>> nanops.nanskew(s.values)
+    1.7320508075688787
+    """
+    mask = _maybe_get_mask(values, skipna, mask)
+    if values.dtype.kind != "f":
+        values = values.astype("f8")
+        count = _get_counts(values.shape, mask, axis)
+    else:
+        count = _get_counts(values.shape, mask, axis, dtype=values.dtype)
+
+    if skipna and mask is not None:
+        values = values.copy()
+        np.putmask(values, mask, 0)
+    elif not skipna and mask is not None and mask.any():
+        return np.nan
+
+    with np.errstate(invalid="ignore", divide="ignore"):
+        mean = values.sum(axis, dtype=np.float64) / count
+    if axis is not None:
+        mean = np.expand_dims(mean, axis)
+
+    adjusted = values - mean
+    if skipna and mask is not None:
+        np.putmask(adjusted, mask, 0)
+    adjusted2 = adjusted**2
+    adjusted3 = adjusted2 * adjusted
+    m2 = adjusted2.sum(axis, dtype=np.float64)
+    m3 = adjusted3.sum(axis, dtype=np.float64)
+
+    # floating point error
+    #
+    # #18044 in _libs/windows.pyx calc_skew follow this behavior
+    # to fix the fperr to treat m2 <1e-14 as zero
+    m2 = _zero_out_fperr(m2)
+    m3 = _zero_out_fperr(m3)
+
+    with np.errstate(invalid="ignore", divide="ignore"):
+        result = (count * (count - 1) ** 0.5 / (count - 2)) * (m3 / m2**1.5)
+
+    dtype = values.dtype
+    if dtype.kind == "f":
+        result = result.astype(dtype, copy=False)
+
+    if isinstance(result, np.ndarray):
+        result = np.where(m2 == 0, 0, result)
+        result[count < 3] = np.nan
+    else:
+        result = dtype.type(0) if m2 == 0 else result
+        if count < 3:
+            return np.nan
+
+    return result
+
+
+@disallow("M8", "m8")
+@maybe_operate_rowwise
+def nankurt(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Compute the sample excess kurtosis
+
+    The statistic computed here is the adjusted Fisher-Pearson standardized
+    moment coefficient G2, computed directly from the second and fourth
+    central moment.
+
+    Parameters
+    ----------
+    values : ndarray
+    axis : int, optional
+    skipna : bool, default True
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    result : float64
+        Unless input is a float array, in which case use the same
+        precision as the input array.
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, np.nan, 1, 3, 2])
+    >>> nanops.nankurt(s.values)
+    -1.2892561983471076
+    """
+    mask = _maybe_get_mask(values, skipna, mask)
+    if values.dtype.kind != "f":
+        values = values.astype("f8")
+        count = _get_counts(values.shape, mask, axis)
+    else:
+        count = _get_counts(values.shape, mask, axis, dtype=values.dtype)
+
+    if skipna and mask is not None:
+        values = values.copy()
+        np.putmask(values, mask, 0)
+    elif not skipna and mask is not None and mask.any():
+        return np.nan
+
+    with np.errstate(invalid="ignore", divide="ignore"):
+        mean = values.sum(axis, dtype=np.float64) / count
+    if axis is not None:
+        mean = np.expand_dims(mean, axis)
+
+    adjusted = values - mean
+    if skipna and mask is not None:
+        np.putmask(adjusted, mask, 0)
+    adjusted2 = adjusted**2
+    adjusted4 = adjusted2**2
+    m2 = adjusted2.sum(axis, dtype=np.float64)
+    m4 = adjusted4.sum(axis, dtype=np.float64)
+
+    with np.errstate(invalid="ignore", divide="ignore"):
+        adj = 3 * (count - 1) ** 2 / ((count - 2) * (count - 3))
+        numerator = count * (count + 1) * (count - 1) * m4
+        denominator = (count - 2) * (count - 3) * m2**2
+
+    # floating point error
+    #
+    # #18044 in _libs/windows.pyx calc_kurt follow this behavior
+    # to fix the fperr to treat denom <1e-14 as zero
+    numerator = _zero_out_fperr(numerator)
+    denominator = _zero_out_fperr(denominator)
+
+    if not isinstance(denominator, np.ndarray):
+        # if ``denom`` is a scalar, check these corner cases first before
+        # doing division
+        if count < 4:
+            return np.nan
+        if denominator == 0:
+            return values.dtype.type(0)
+
+    with np.errstate(invalid="ignore", divide="ignore"):
+        result = numerator / denominator - adj
+
+    dtype = values.dtype
+    if dtype.kind == "f":
+        result = result.astype(dtype, copy=False)
+
+    if isinstance(result, np.ndarray):
+        result = np.where(denominator == 0, 0, result)
+        result[count < 4] = np.nan
+
+    return result
+
+
+@disallow("M8", "m8")
+@maybe_operate_rowwise
+def nanprod(
+    values: np.ndarray,
+    *,
+    axis: AxisInt | None = None,
+    skipna: bool = True,
+    min_count: int = 0,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> float:
+    """
+    Parameters
+    ----------
+    values : ndarray[dtype]
+    axis : int, optional
+    skipna : bool, default True
+    min_count: int, default 0
+    mask : ndarray[bool], optional
+        nan-mask if known
+
+    Returns
+    -------
+    Dtype
+        The product of all elements on a given axis. ( NaNs are treated as 1)
+
+    Examples
+    --------
+    >>> from pandas.core import nanops
+    >>> s = pd.Series([1, 2, 3, np.nan])
+    >>> nanops.nanprod(s.values)
+    6.0
+    """
+    mask = _maybe_get_mask(values, skipna, mask)
+
+    if skipna and mask is not None:
+        values = values.copy()
+        values[mask] = 1
+    result = values.prod(axis)
+    # error: Incompatible return value type (got "Union[ndarray, float]", expected
+    # "float")
+    return _maybe_null_out(  # type: ignore[return-value]
+        result, axis, mask, values.shape, min_count=min_count
+    )
+
+
+def _maybe_arg_null_out(
+    result: np.ndarray,
+    axis: AxisInt | None,
+    mask: npt.NDArray[np.bool_] | None,
+    skipna: bool,
+) -> np.ndarray | int:
+    # helper function for nanargmin/nanargmax
+    if mask is None:
+        return result
+
+    if axis is None or not getattr(result, "ndim", False):
+        if skipna and mask.all():
+            raise ValueError("Encountered all NA values")
+        elif not skipna and mask.any():
+            raise ValueError("Encountered an NA value with skipna=False")
+    else:
+        if skipna and mask.all(axis).any():
+            raise ValueError("Encountered all NA values")
+        elif not skipna and mask.any(axis).any():
+            raise ValueError("Encountered an NA value with skipna=False")
+    return result
+
+
+def _get_counts(
+    values_shape: Shape,
+    mask: npt.NDArray[np.bool_] | None,
+    axis: AxisInt | None,
+    dtype: np.dtype[np.floating] = np.dtype(np.float64),
+) -> np.floating | npt.NDArray[np.floating]:
+    """
+    Get the count of non-null values along an axis
+
+    Parameters
+    ----------
+    values_shape : tuple of int
+        shape tuple from values ndarray, used if mask is None
+    mask : Optional[ndarray[bool]]
+        locations in values that should be considered missing
+    axis : Optional[int]
+        axis to count along
+    dtype : type, optional
+        type to use for count
+
+    Returns
+    -------
+    count : scalar or array
+    """
+    if axis is None:
+        if mask is not None:
+            n = mask.size - mask.sum()
+        else:
+            n = np.prod(values_shape)
+        return dtype.type(n)
+
+    if mask is not None:
+        count = mask.shape[axis] - mask.sum(axis)
+    else:
+        count = values_shape[axis]
+
+    if is_integer(count):
+        return dtype.type(count)
+    return count.astype(dtype, copy=False)
+
+
+def _maybe_null_out(
+    result: np.ndarray | float | NaTType,
+    axis: AxisInt | None,
+    mask: npt.NDArray[np.bool_] | None,
+    shape: tuple[int, ...],
+    min_count: int = 1,
+    datetimelike: bool = False,
+) -> np.ndarray | float | NaTType:
+    """
+    Returns
+    -------
+    Dtype
+        The product of all elements on a given axis. ( NaNs are treated as 1)
+    """
+    if mask is None and min_count == 0:
+        # nothing to check; short-circuit
+        return result
+
+    if axis is not None and isinstance(result, np.ndarray):
+        if mask is not None:
+            null_mask = (mask.shape[axis] - mask.sum(axis) - min_count) < 0
+        else:
+            # we have no nulls, kept mask=None in _maybe_get_mask
+            below_count = shape[axis] - min_count < 0
+            new_shape = shape[:axis] + shape[axis + 1 :]
+            null_mask = np.broadcast_to(below_count, new_shape)
+
+        if np.any(null_mask):
+            if datetimelike:
+                # GH#60646 For datetimelike, no need to cast to float
+                result[null_mask] = iNaT
+            elif is_numeric_dtype(result):
+                if np.iscomplexobj(result):
+                    result = result.astype("c16")
+                elif not is_float_dtype(result):
+                    result = result.astype("f8", copy=False)
+                result[null_mask] = np.nan
+            else:
+                # GH12941, use None to auto cast null
+                result[null_mask] = None
+    elif result is not NaT:
+        if check_below_min_count(shape, mask, min_count):
+            result_dtype = getattr(result, "dtype", None)
+            if is_float_dtype(result_dtype):
+                # error: Item "None" of "Optional[Any]" has no attribute "type"
+                result = result_dtype.type("nan")  # type: ignore[union-attr]
+            else:
+                result = np.nan
+
+    return result
+
+
+def check_below_min_count(
+    shape: tuple[int, ...], mask: npt.NDArray[np.bool_] | None, min_count: int
+) -> bool:
+    """
+    Check for the `min_count` keyword. Returns True if below `min_count` (when
+    missing value should be returned from the reduction).
+
+    Parameters
+    ----------
+    shape : tuple
+        The shape of the values (`values.shape`).
+    mask : ndarray[bool] or None
+        Boolean numpy array (typically of same shape as `shape`) or None.
+    min_count : int
+        Keyword passed through from sum/prod call.
+
+    Returns
+    -------
+    bool
+    """
+    if min_count > 0:
+        if mask is None:
+            # no missing values, only check size
+            non_nulls = np.prod(shape)
+        else:
+            non_nulls = mask.size - mask.sum()
+        if non_nulls < min_count:
+            return True
+    return False
+
+
+def _zero_out_fperr(arg):
+    # #18044 reference this behavior to fix rolling skew/kurt issue
+    if isinstance(arg, np.ndarray):
+        return np.where(np.abs(arg) < 1e-14, 0, arg)
+    else:
+        return arg.dtype.type(0) if np.abs(arg) < 1e-14 else arg
+
+
+@disallow("M8", "m8")
+def nancorr(
+    a: np.ndarray,
+    b: np.ndarray,
+    *,
+    method: CorrelationMethod = "pearson",
+    min_periods: int | None = None,
+) -> float:
+    """
+    a, b: ndarrays
+    """
+    if len(a) != len(b):
+        raise AssertionError("Operands to nancorr must have same size")
+
+    if min_periods is None:
+        min_periods = 1
+
+    valid = notna(a) & notna(b)
+    if not valid.all():
+        a = a[valid]
+        b = b[valid]
+
+    if len(a) < min_periods:
+        return np.nan
+
+    a = _ensure_numeric(a)
+    b = _ensure_numeric(b)
+
+    f = get_corr_func(method)
+    return f(a, b)
+
+
+def get_corr_func(
+    method: CorrelationMethod,
+) -> Callable[[np.ndarray, np.ndarray], float]:
+    if method == "kendall":
+        from scipy.stats import kendalltau
+
+        def func(a, b):
+            return kendalltau(a, b)[0]
+
+        return func
+    elif method == "spearman":
+        from scipy.stats import spearmanr
+
+        def func(a, b):
+            return spearmanr(a, b)[0]
+
+        return func
+    elif method == "pearson":
+
+        def func(a, b):
+            return np.corrcoef(a, b)[0, 1]
+
+        return func
+    elif callable(method):
+        return method
+
+    raise ValueError(
+        f"Unknown method '{method}', expected one of "
+        "'kendall', 'spearman', 'pearson', or callable"
+    )
+
+
+@disallow("M8", "m8")
+def nancov(
+    a: np.ndarray,
+    b: np.ndarray,
+    *,
+    min_periods: int | None = None,
+    ddof: int | None = 1,
+) -> float:
+    if len(a) != len(b):
+        raise AssertionError("Operands to nancov must have same size")
+
+    if min_periods is None:
+        min_periods = 1
+
+    valid = notna(a) & notna(b)
+    if not valid.all():
+        a = a[valid]
+        b = b[valid]
+
+    if len(a) < min_periods:
+        return np.nan
+
+    a = _ensure_numeric(a)
+    b = _ensure_numeric(b)
+
+    return np.cov(a, b, ddof=ddof)[0, 1]
+
+
+def _ensure_numeric(x):
+    if isinstance(x, np.ndarray):
+        if x.dtype.kind in "biu":
+            x = x.astype(np.float64)
+        elif x.dtype == object:
+            inferred = lib.infer_dtype(x)
+            if inferred in ["string", "mixed"]:
+                # GH#44008, GH#36703 avoid casting e.g. strings to numeric
+                raise TypeError(f"Could not convert {x} to numeric")
+            try:
+                x = x.astype(np.complex128)
+            except (TypeError, ValueError):
+                try:
+                    x = x.astype(np.float64)
+                except ValueError as err:
+                    # GH#29941 we get here with object arrays containing strs
+                    raise TypeError(f"Could not convert {x} to numeric") from err
+            else:
+                if not np.any(np.imag(x)):
+                    x = x.real
+    elif not (is_float(x) or is_integer(x) or is_complex(x)):
+        if isinstance(x, str):
+            # GH#44008, GH#36703 avoid casting e.g. strings to numeric
+            raise TypeError(f"Could not convert string '{x}' to numeric")
+        try:
+            x = float(x)
+        except (TypeError, ValueError):
+            # e.g. "1+1j" or "foo"
+            try:
+                x = complex(x)
+            except ValueError as err:
+                # e.g. "foo"
+                raise TypeError(f"Could not convert {x} to numeric") from err
+    return x
+
+
+def na_accum_func(values: ArrayLike, accum_func, *, skipna: bool) -> ArrayLike:
+    """
+    Cumulative function with skipna support.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+    accum_func : {np.cumprod, np.maximum.accumulate, np.cumsum, np.minimum.accumulate}
+    skipna : bool
+
+    Returns
+    -------
+    np.ndarray or ExtensionArray
+    """
+    mask_a, mask_b = {
+        np.cumprod: (1.0, np.nan),
+        np.maximum.accumulate: (-np.inf, np.nan),
+        np.cumsum: (0.0, np.nan),
+        np.minimum.accumulate: (np.inf, np.nan),
+    }[accum_func]
+
+    # This should go through ea interface
+    assert values.dtype.kind not in "mM"
+
+    # We will be applying this function to block values
+    if skipna and not issubclass(values.dtype.type, (np.integer, np.bool_)):
+        vals = values.copy()
+        mask = isna(vals)
+        vals[mask] = mask_a
+        result = accum_func(vals, axis=0)
+        result[mask] = mask_b
+    else:
+        result = accum_func(values, axis=0)
+
+    return result
diff --git a/pandas/pandas/core/ops/__init__.py b/pandas/pandas/core/ops/__init__.py
new file mode 100644
index 0000000000000..9f9d69a182f72
--- /dev/null
+++ b/pandas/pandas/core/ops/__init__.py
@@ -0,0 +1,94 @@
+"""
+Arithmetic operations for PandasObjects
+
+This is not a public API.
+"""
+
+from __future__ import annotations
+
+from pandas.core.ops.array_ops import (
+    arithmetic_op,
+    comp_method_OBJECT_ARRAY,
+    comparison_op,
+    fill_binop,
+    get_array_op,
+    logical_op,
+    maybe_prepare_scalar_for_op,
+)
+from pandas.core.ops.common import (
+    get_op_result_name,
+    unpack_zerodim_and_defer,
+)
+from pandas.core.ops.docstrings import make_flex_doc
+from pandas.core.ops.invalid import invalid_comparison
+from pandas.core.ops.mask_ops import (
+    kleene_and,
+    kleene_or,
+    kleene_xor,
+)
+from pandas.core.roperator import (
+    radd,
+    rand_,
+    rdiv,
+    rdivmod,
+    rfloordiv,
+    rmod,
+    rmul,
+    ror_,
+    rpow,
+    rsub,
+    rtruediv,
+    rxor,
+)
+
+# -----------------------------------------------------------------------------
+# constants
+ARITHMETIC_BINOPS: set[str] = {
+    "add",
+    "sub",
+    "mul",
+    "pow",
+    "mod",
+    "floordiv",
+    "truediv",
+    "divmod",
+    "radd",
+    "rsub",
+    "rmul",
+    "rpow",
+    "rmod",
+    "rfloordiv",
+    "rtruediv",
+    "rdivmod",
+}
+
+
+__all__ = [
+    "ARITHMETIC_BINOPS",
+    "arithmetic_op",
+    "comp_method_OBJECT_ARRAY",
+    "comparison_op",
+    "fill_binop",
+    "get_array_op",
+    "get_op_result_name",
+    "invalid_comparison",
+    "kleene_and",
+    "kleene_or",
+    "kleene_xor",
+    "logical_op",
+    "make_flex_doc",
+    "maybe_prepare_scalar_for_op",
+    "radd",
+    "rand_",
+    "rdiv",
+    "rdivmod",
+    "rfloordiv",
+    "rmod",
+    "rmul",
+    "ror_",
+    "rpow",
+    "rsub",
+    "rtruediv",
+    "rxor",
+    "unpack_zerodim_and_defer",
+]
diff --git a/pandas/pandas/core/ops/array_ops.py b/pandas/pandas/core/ops/array_ops.py
new file mode 100644
index 0000000000000..3a466b6fc7fc8
--- /dev/null
+++ b/pandas/pandas/core/ops/array_ops.py
@@ -0,0 +1,601 @@
+"""
+Functions for arithmetic and comparison operations on NumPy arrays and
+ExtensionArrays.
+"""
+
+from __future__ import annotations
+
+import datetime
+from functools import partial
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    NaT,
+    Timedelta,
+    Timestamp,
+    lib,
+    ops as libops,
+)
+from pandas._libs.tslibs import (
+    BaseOffset,
+    get_supported_dtype,
+    is_supported_dtype,
+    is_unitless,
+)
+
+from pandas.core.dtypes.cast import (
+    construct_1d_object_array_from_listlike,
+    find_common_type,
+)
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_bool_dtype,
+    is_list_like,
+    is_numeric_v_string_like,
+    is_object_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.generic import (
+    ABCExtensionArray,
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    notna,
+)
+
+from pandas.core import roperator
+from pandas.core.computation import expressions
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+from pandas.core.ops import missing
+from pandas.core.ops.dispatch import should_extension_dispatch
+from pandas.core.ops.invalid import invalid_comparison
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Shape,
+    )
+
+# -----------------------------------------------------------------------------
+# Masking NA values and fallbacks for operations numpy does not support
+
+
+def fill_binop(left, right, fill_value):
+    """
+    If a non-None fill_value is given, replace null entries in left and right
+    with this value, but only in positions where _one_ of left/right is null,
+    not both.
+
+    Parameters
+    ----------
+    left : array-like
+    right : array-like
+    fill_value : object
+
+    Returns
+    -------
+    left : array-like
+    right : array-like
+
+    Notes
+    -----
+    Makes copies if fill_value is not None and NAs are present.
+    """
+    if fill_value is not None:
+        left_mask = isna(left)
+        right_mask = isna(right)
+
+        # one but not both
+        mask = left_mask ^ right_mask
+
+        if left_mask.any():
+            # Avoid making a copy if we can
+            left = left.copy()
+            left[left_mask & mask] = fill_value
+
+        if right_mask.any():
+            # Avoid making a copy if we can
+            right = right.copy()
+            right[right_mask & mask] = fill_value
+
+    return left, right
+
+
+def comp_method_OBJECT_ARRAY(op, x, y):
+    if isinstance(y, list):
+        # e.g. test_tuple_categories
+        y = construct_1d_object_array_from_listlike(y)
+
+    if isinstance(y, (np.ndarray, ABCSeries, ABCIndex)):
+        if not is_object_dtype(y.dtype):
+            y = y.astype(np.object_)
+
+        if isinstance(y, (ABCSeries, ABCIndex)):
+            y = y._values
+
+        if x.shape != y.shape:
+            raise ValueError("Shapes must match", x.shape, y.shape)
+        result = libops.vec_compare(x.ravel(), y.ravel(), op)
+    else:
+        result = libops.scalar_compare(x.ravel(), y, op)
+    return result.reshape(x.shape)
+
+
+def _masked_arith_op(x: np.ndarray, y, op) -> np.ndarray:
+    """
+    If the given arithmetic operation fails, attempt it again on
+    only the non-null elements of the input array(s).
+
+    Parameters
+    ----------
+    x : np.ndarray
+    y : np.ndarray, Series, Index
+    op : binary operator
+    """
+    # For Series `x` is 1D so ravel() is a no-op; calling it anyway makes
+    # the logic valid for both Series and DataFrame ops.
+    xrav = x.ravel()
+
+    if isinstance(y, np.ndarray):
+        dtype = find_common_type([x.dtype, y.dtype])
+        result = np.empty(x.size, dtype=dtype)
+
+        if len(x) != len(y):
+            raise ValueError(x.shape, y.shape)
+        ymask = notna(y)
+
+        # NB: ravel() is only safe since y is ndarray; for e.g. PeriodIndex
+        #  we would get int64 dtype, see GH#19956
+        yrav = y.ravel()
+        mask = notna(xrav) & ymask.ravel()
+
+        # See GH#5284, GH#5035, GH#19448 for historical reference
+        if mask.any():
+            result[mask] = op(xrav[mask], yrav[mask])
+
+    else:
+        if not is_scalar(y):
+            raise TypeError(
+                f"Cannot broadcast np.ndarray with operand of type {type(y)}"
+            )
+
+        # mask is only meaningful for x
+        result = np.empty(x.size, dtype=x.dtype)
+        mask = notna(xrav)
+
+        # 1 ** np.nan is 1. So we have to unmask those.
+        if op is pow:
+            mask = np.where(x == 1, False, mask)
+        elif op is roperator.rpow:
+            mask = np.where(y == 1, False, mask)
+
+        if mask.any():
+            result[mask] = op(xrav[mask], y)
+
+    np.putmask(result, ~mask, np.nan)
+    result = result.reshape(x.shape)  # 2D compat
+    return result
+
+
+def _na_arithmetic_op(left: np.ndarray, right, op, is_cmp: bool = False):
+    """
+    Return the result of evaluating op on the passed in values.
+
+    If native types are not compatible, try coercion to object dtype.
+
+    Parameters
+    ----------
+    left : np.ndarray
+    right : np.ndarray or scalar
+        Excludes DataFrame, Series, Index, ExtensionArray.
+    is_cmp : bool, default False
+        If this a comparison operation.
+
+    Returns
+    -------
+    array-like
+
+    Raises
+    ------
+    TypeError : invalid operation
+    """
+    if isinstance(right, str):
+        # can never use numexpr
+        func = op
+    else:
+        func = partial(expressions.evaluate, op)
+
+    try:
+        result = func(left, right)
+    except TypeError:
+        if not is_cmp and (
+            left.dtype == object or getattr(right, "dtype", None) == object
+        ):
+            # For object dtype, fallback to a masked operation (only operating
+            #  on the non-missing values)
+            # Don't do this for comparisons, as that will handle complex numbers
+            #  incorrectly, see GH#32047
+            result = _masked_arith_op(left, right, op)
+        else:
+            raise
+
+    if is_cmp and (is_scalar(result) or result is NotImplemented):
+        # numpy returned a scalar instead of operating element-wise
+        # e.g. numeric array vs str
+        # TODO: can remove this after dropping some future numpy version?
+        return invalid_comparison(left, right, op)
+
+    return missing.dispatch_fill_zeros(op, left, right, result)
+
+
+def arithmetic_op(left: ArrayLike, right: Any, op):
+    """
+    Evaluate an arithmetic operation `+`, `-`, `*`, `/`, `//`, `%`, `**`, ...
+
+    Note: the caller is responsible for ensuring that numpy warnings are
+    suppressed (with np.errstate(all="ignore")) if needed.
+
+    Parameters
+    ----------
+    left : np.ndarray or ExtensionArray
+    right : object
+        Cannot be a DataFrame or Index.  Series is *not* excluded.
+    op : {operator.add, operator.sub, ...}
+        Or one of the reversed variants from roperator.
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+        Or a 2-tuple of these in the case of divmod or rdivmod.
+    """
+    # NB: We assume that extract_array and ensure_wrapped_if_datetimelike
+    #  have already been called on `left` and `right`,
+    #  and `maybe_prepare_scalar_for_op` has already been called on `right`
+    # We need to special-case datetime64/timedelta64 dtypes (e.g. because numpy
+    # casts integer dtypes to timedelta64 when operating with timedelta64 - GH#22390)
+
+    if (
+        should_extension_dispatch(left, right)
+        or isinstance(right, (Timedelta, BaseOffset, Timestamp))
+        or right is NaT
+    ):
+        # Timedelta/Timestamp and other custom scalars are included in the check
+        # because numexpr will fail on it, see GH#31457
+        res_values = op(left, right)
+    else:
+        # TODO we should handle EAs consistently and move this check before the if/else
+        # (https://github.com/pandas-dev/pandas/issues/41165)
+        # error: Argument 2 to "_bool_arith_check" has incompatible type
+        # "Union[ExtensionArray, ndarray[Any, Any]]"; expected "ndarray[Any, Any]"
+        _bool_arith_check(op, left, right)  # type: ignore[arg-type]
+
+        # error: Argument 1 to "_na_arithmetic_op" has incompatible type
+        # "Union[ExtensionArray, ndarray[Any, Any]]"; expected "ndarray[Any, Any]"
+        res_values = _na_arithmetic_op(left, right, op)  # type: ignore[arg-type]
+
+    return res_values
+
+
+def comparison_op(left: ArrayLike, right: Any, op) -> ArrayLike:
+    """
+    Evaluate a comparison operation `=`, `!=`, `>=`, `>`, `<=`, or `<`.
+
+    Note: the caller is responsible for ensuring that numpy warnings are
+    suppressed (with np.errstate(all="ignore")) if needed.
+
+    Parameters
+    ----------
+    left : np.ndarray or ExtensionArray
+    right : object
+        Cannot be a DataFrame, Series, or Index.
+    op : {operator.eq, operator.ne, operator.gt, operator.ge, operator.lt, operator.le}
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+    # NB: We assume extract_array has already been called on left and right
+    lvalues = ensure_wrapped_if_datetimelike(left)
+    rvalues = ensure_wrapped_if_datetimelike(right)
+
+    rvalues = lib.item_from_zerodim(rvalues)
+    if isinstance(rvalues, list):
+        # We don't catch tuple here bc we may be comparing e.g. MultiIndex
+        #  to a tuple that represents a single entry, see test_compare_tuple_strs
+        rvalues = np.asarray(rvalues)
+
+    if isinstance(rvalues, (np.ndarray, ABCExtensionArray)):
+        # TODO: make this treatment consistent across ops and classes.
+        #  We are not catching all listlikes here (e.g. frozenset, tuple)
+        #  The ambiguous case is object-dtype.  See GH#27803
+        if len(lvalues) != len(rvalues):
+            raise ValueError(
+                "Lengths must match to compare", lvalues.shape, rvalues.shape
+            )
+
+    if should_extension_dispatch(lvalues, rvalues) or (
+        (isinstance(rvalues, (Timedelta, BaseOffset, Timestamp)) or right is NaT)
+        and lvalues.dtype != object
+    ):
+        # Call the method on lvalues
+        res_values = op(lvalues, rvalues)
+
+    elif is_scalar(rvalues) and isna(rvalues):  # TODO: but not pd.NA?
+        # numpy does not like comparisons vs None
+        if op is operator.ne:
+            res_values = np.ones(lvalues.shape, dtype=bool)
+        else:
+            res_values = np.zeros(lvalues.shape, dtype=bool)
+
+    elif is_numeric_v_string_like(lvalues, rvalues):
+        # GH#36377 going through the numexpr path would incorrectly raise
+        return invalid_comparison(lvalues, rvalues, op)
+
+    elif lvalues.dtype == object or isinstance(rvalues, str):
+        res_values = comp_method_OBJECT_ARRAY(op, lvalues, rvalues)
+
+    else:
+        res_values = _na_arithmetic_op(lvalues, rvalues, op, is_cmp=True)
+
+    return res_values
+
+
+def na_logical_op(x: np.ndarray, y, op):
+    try:
+        # For exposition, write:
+        #  yarr = isinstance(y, np.ndarray)
+        #  yint = is_integer(y) or (yarr and y.dtype.kind == "i")
+        #  ybool = is_bool(y) or (yarr and y.dtype.kind == "b")
+        #  xint = x.dtype.kind == "i"
+        #  xbool = x.dtype.kind == "b"
+        # Then Cases where this goes through without raising include:
+        #  (xint or xbool) and (yint or bool)
+        result = op(x, y)
+    except TypeError:
+        if isinstance(y, np.ndarray):
+            # bool-bool dtype operations should be OK, should not get here
+            assert not (x.dtype.kind == "b" and y.dtype.kind == "b")
+            x = ensure_object(x)
+            y = ensure_object(y)
+            result = libops.vec_binop(x.ravel(), y.ravel(), op)
+        else:
+            # let null fall thru
+            assert lib.is_scalar(y)
+            if not isna(y):
+                y = bool(y)
+            try:
+                result = libops.scalar_binop(x, y, op)
+            except (
+                TypeError,
+                ValueError,
+                AttributeError,
+                OverflowError,
+                NotImplementedError,
+            ) as err:
+                typ = type(y).__name__
+                raise TypeError(
+                    f"Cannot perform '{op.__name__}' with a dtyped [{x.dtype}] array "
+                    f"and scalar of type [{typ}]"
+                ) from err
+
+    return result.reshape(x.shape)
+
+
+def logical_op(left: ArrayLike, right: Any, op) -> ArrayLike:
+    """
+    Evaluate a logical operation `|`, `&`, or `^`.
+
+    Parameters
+    ----------
+    left : np.ndarray or ExtensionArray
+    right : object
+        Cannot be a DataFrame, Series, or Index.
+    op : {operator.and_, operator.or_, operator.xor}
+        Or one of the reversed variants from roperator.
+
+    Returns
+    -------
+    ndarray or ExtensionArray
+    """
+
+    def fill_bool(x, left=None):
+        # if `left` is specifically not-boolean, we do not cast to bool
+        if x.dtype.kind in "cfO":
+            # dtypes that can hold NA
+            mask = isna(x)
+            if mask.any():
+                x = x.astype(object)
+                x[mask] = False
+
+        if left is None or left.dtype.kind == "b":
+            x = x.astype(bool)
+        return x
+
+    right = lib.item_from_zerodim(right)
+    if is_list_like(right) and not hasattr(right, "dtype"):
+        # e.g. list, tuple
+        raise TypeError(
+            # GH#52264
+            "Logical ops (and, or, xor) between Pandas objects and dtype-less "
+            "sequences (e.g. list, tuple) are no longer supported. "
+            "Wrap the object in a Series, Index, or np.array "
+            "before operating instead.",
+        )
+
+    # NB: We assume extract_array has already been called on left and right
+    lvalues = ensure_wrapped_if_datetimelike(left)
+    rvalues = right
+
+    if should_extension_dispatch(lvalues, rvalues):
+        # Call the method on lvalues
+        res_values = op(lvalues, rvalues)
+
+    else:
+        if isinstance(rvalues, np.ndarray):
+            is_other_int_dtype = rvalues.dtype.kind in "iu"
+            if not is_other_int_dtype:
+                rvalues = fill_bool(rvalues, lvalues)
+
+        else:
+            # i.e. scalar
+            is_other_int_dtype = lib.is_integer(rvalues)
+
+        res_values = na_logical_op(lvalues, rvalues, op)
+
+        # For int vs int `^`, `|`, `&` are bitwise operators and return
+        #   integer dtypes.  Otherwise these are boolean ops
+        if not (left.dtype.kind in "iu" and is_other_int_dtype):
+            res_values = fill_bool(res_values)
+
+    return res_values
+
+
+def get_array_op(op):
+    """
+    Return a binary array operation corresponding to the given operator op.
+
+    Parameters
+    ----------
+    op : function
+        Binary operator from operator or roperator module.
+
+    Returns
+    -------
+    functools.partial
+    """
+    if isinstance(op, partial):
+        # We get here via dispatch_to_series in DataFrame case
+        # e.g. test_rolling_consistency_var_debiasing_factors
+        return op
+
+    op_name = op.__name__.strip("_").lstrip("r")
+    if op_name == "arith_op":
+        # Reached via DataFrame._combine_frame i.e. flex methods
+        # e.g. test_df_add_flex_filled_mixed_dtypes
+        return op
+
+    if op_name in {"eq", "ne", "lt", "le", "gt", "ge"}:
+        return partial(comparison_op, op=op)
+    elif op_name in {"and", "or", "xor", "rand", "ror", "rxor"}:
+        return partial(logical_op, op=op)
+    elif op_name in {
+        "add",
+        "sub",
+        "mul",
+        "truediv",
+        "floordiv",
+        "mod",
+        "divmod",
+        "pow",
+    }:
+        return partial(arithmetic_op, op=op)
+    else:
+        raise NotImplementedError(op_name)
+
+
+def maybe_prepare_scalar_for_op(obj, shape: Shape):
+    """
+    Cast non-pandas objects to pandas types to unify behavior of arithmetic
+    and comparison operations.
+
+    Parameters
+    ----------
+    obj: object
+    shape : tuple[int]
+
+    Returns
+    -------
+    out : object
+
+    Notes
+    -----
+    Be careful to call this *after* determining the `name` attribute to be
+    attached to the result of the arithmetic operation.
+    """
+    if type(obj) is datetime.timedelta:
+        # GH#22390  cast up to Timedelta to rely on Timedelta
+        # implementation; otherwise operation against numeric-dtype
+        # raises TypeError
+        return Timedelta(obj)
+    elif type(obj) is datetime.datetime:
+        # cast up to Timestamp to rely on Timestamp implementation, see Timedelta above
+        return Timestamp(obj)
+    elif isinstance(obj, np.datetime64):
+        # GH#28080 numpy casts integer-dtype to datetime64 when doing
+        #  array[int] + datetime64, which we do not allow
+        if isna(obj):
+            from pandas.core.arrays import DatetimeArray
+
+            # Avoid possible ambiguities with pd.NaT
+            # GH 52295
+            if is_unitless(obj.dtype):
+                obj = obj.astype("datetime64[ns]")
+            elif not is_supported_dtype(obj.dtype):
+                new_dtype = get_supported_dtype(obj.dtype)
+                obj = obj.astype(new_dtype)
+            right = np.broadcast_to(obj, shape)
+            return DatetimeArray._simple_new(right, dtype=right.dtype)
+
+        return Timestamp(obj)
+
+    elif isinstance(obj, np.timedelta64):
+        if isna(obj):
+            from pandas.core.arrays import TimedeltaArray
+
+            # wrapping timedelta64("NaT") in Timedelta returns NaT,
+            #  which would incorrectly be treated as a datetime-NaT, so
+            #  we broadcast and wrap in a TimedeltaArray
+            # GH 52295
+            if is_unitless(obj.dtype):
+                obj = obj.astype("timedelta64[ns]")
+            elif not is_supported_dtype(obj.dtype):
+                new_dtype = get_supported_dtype(obj.dtype)
+                obj = obj.astype(new_dtype)
+            right = np.broadcast_to(obj, shape)
+            return TimedeltaArray._simple_new(right, dtype=right.dtype)
+
+        # In particular non-nanosecond timedelta64 needs to be cast to
+        #  nanoseconds, or else we get undesired behavior like
+        #  np.timedelta64(3, 'D') / 2 == np.timedelta64(1, 'D')
+        return Timedelta(obj)
+
+    # We want NumPy numeric scalars to behave like Python scalars
+    # post NEP 50
+    elif isinstance(obj, np.integer):
+        return int(obj)
+
+    elif isinstance(obj, np.floating):
+        return float(obj)
+
+    return obj
+
+
+_BOOL_OP_NOT_ALLOWED = {
+    operator.truediv,
+    roperator.rtruediv,
+    operator.floordiv,
+    roperator.rfloordiv,
+    operator.pow,
+    roperator.rpow,
+}
+
+
+def _bool_arith_check(op, a: np.ndarray, b) -> None:
+    """
+    In contrast to numpy, pandas raises an error for certain operations
+    with booleans.
+    """
+    if op in _BOOL_OP_NOT_ALLOWED:
+        if a.dtype.kind == "b" and (is_bool_dtype(b) or lib.is_bool(b)):
+            op_name = op.__name__.strip("_").lstrip("r")
+            raise NotImplementedError(
+                f"operator '{op_name}' not implemented for bool dtypes"
+            )
diff --git a/pandas/pandas/core/ops/common.py b/pandas/pandas/core/ops/common.py
new file mode 100644
index 0000000000000..5cbe1c421e05a
--- /dev/null
+++ b/pandas/pandas/core/ops/common.py
@@ -0,0 +1,148 @@
+"""
+Boilerplate functions used in defining binary operations.
+"""
+
+from __future__ import annotations
+
+from functools import wraps
+from typing import TYPE_CHECKING
+
+from pandas._libs.lib import item_from_zerodim
+from pandas._libs.missing import is_matching_na
+
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import F
+
+
+def unpack_zerodim_and_defer(name: str) -> Callable[[F], F]:
+    """
+    Boilerplate for pandas conventions in arithmetic and comparison methods.
+
+    Parameters
+    ----------
+    name : str
+
+    Returns
+    -------
+    decorator
+    """
+
+    def wrapper(method: F) -> F:
+        return _unpack_zerodim_and_defer(method, name)
+
+    return wrapper
+
+
+def _unpack_zerodim_and_defer(method: F, name: str) -> F:
+    """
+    Boilerplate for pandas conventions in arithmetic and comparison methods.
+
+    Ensure method returns NotImplemented when operating against "senior"
+    classes.  Ensure zero-dimensional ndarrays are always unpacked.
+
+    Parameters
+    ----------
+    method : binary method
+    name : str
+
+    Returns
+    -------
+    method
+    """
+    stripped_name = name.removeprefix("__").removesuffix("__")
+    is_cmp = stripped_name in {"eq", "ne", "lt", "le", "gt", "ge"}
+
+    @wraps(method)
+    def new_method(self, other):
+        if is_cmp and isinstance(self, ABCIndex) and isinstance(other, ABCSeries):
+            # For comparison ops, Index does *not* defer to Series
+            pass
+        else:
+            prio = getattr(other, "__pandas_priority__", None)
+            if prio is not None:
+                if prio > self.__pandas_priority__:
+                    # e.g. other is DataFrame while self is Index/Series/EA
+                    return NotImplemented
+
+        other = item_from_zerodim(other)
+
+        return method(self, other)
+
+    # error: Incompatible return value type (got "Callable[[Any, Any], Any]",
+    # expected "F")
+    return new_method  # type: ignore[return-value]
+
+
+def get_op_result_name(left, right):
+    """
+    Find the appropriate name to pin to an operation result.  This result
+    should always be either an Index or a Series.
+
+    Parameters
+    ----------
+    left : {Series, Index}
+    right : object
+
+    Returns
+    -------
+    name : object
+        Usually a string
+    """
+    if isinstance(right, (ABCSeries, ABCIndex)):
+        name = _maybe_match_name(left, right)
+    else:
+        name = left.name
+    return name
+
+
+def _maybe_match_name(a, b):
+    """
+    Try to find a name to attach to the result of an operation between
+    a and b.  If only one of these has a `name` attribute, return that
+    name.  Otherwise return a consensus name if they match or None if
+    they have different names.
+
+    Parameters
+    ----------
+    a : object
+    b : object
+
+    Returns
+    -------
+    name : str or None
+
+    See Also
+    --------
+    pandas.core.common.consensus_name_attr
+    """
+    a_has = hasattr(a, "name")
+    b_has = hasattr(b, "name")
+    if a_has and b_has:
+        try:
+            if a.name == b.name:
+                return a.name
+            elif is_matching_na(a.name, b.name):
+                # e.g. both are np.nan
+                return a.name
+            else:
+                return None
+        except TypeError:
+            # pd.NA
+            if is_matching_na(a.name, b.name):
+                return a.name
+            return None
+        except ValueError:
+            # e.g. np.int64(1) vs (np.int64(1), np.int64(2))
+            return None
+    elif a_has:
+        return a.name
+    elif b_has:
+        return b.name
+    return None
diff --git a/pandas/pandas/core/ops/dispatch.py b/pandas/pandas/core/ops/dispatch.py
new file mode 100644
index 0000000000000..ebafc432dd89b
--- /dev/null
+++ b/pandas/pandas/core/ops/dispatch.py
@@ -0,0 +1,31 @@
+"""
+Functions for defining unary operations.
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas.core.dtypes.generic import ABCExtensionArray
+
+if TYPE_CHECKING:
+    from pandas._typing import ArrayLike
+
+
+def should_extension_dispatch(left: ArrayLike, right: Any) -> bool:
+    """
+    Identify cases where Series operation should dispatch to ExtensionArray method.
+
+    Parameters
+    ----------
+    left : np.ndarray or ExtensionArray
+    right : object
+
+    Returns
+    -------
+    bool
+    """
+    return isinstance(left, ABCExtensionArray) or isinstance(right, ABCExtensionArray)
diff --git a/pandas/pandas/core/ops/docstrings.py b/pandas/pandas/core/ops/docstrings.py
new file mode 100644
index 0000000000000..5ce0a2da86f31
--- /dev/null
+++ b/pandas/pandas/core/ops/docstrings.py
@@ -0,0 +1,783 @@
+"""
+Templating for ops docstrings
+"""
+
+from __future__ import annotations
+
+
+def make_flex_doc(op_name: str, typ: str) -> str:
+    """
+    Make the appropriate substitutions for the given operation and class-typ
+    into either _flex_doc_SERIES or _flex_doc_FRAME to return the docstring
+    to attach to a generated method.
+
+    Parameters
+    ----------
+    op_name : str {'__add__', '__sub__', ... '__eq__', '__ne__', ...}
+    typ : str {series, 'dataframe']}
+
+    Returns
+    -------
+    doc : str
+    """
+    op_name = op_name.replace("__", "")
+    op_desc = _op_descriptions[op_name]
+
+    op_desc_op = op_desc["op"]
+    assert op_desc_op is not None  # for mypy
+    if op_name.startswith("r"):
+        equiv = f"other {op_desc_op} {typ}"
+    elif op_name == "divmod":
+        equiv = f"{op_name}({typ}, other)"
+    else:
+        equiv = f"{typ} {op_desc_op} other"
+
+    if typ == "series":
+        base_doc = _flex_doc_SERIES
+        if op_desc["reverse"]:
+            base_doc += _see_also_reverse_SERIES.format(
+                reverse=op_desc["reverse"], see_also_desc=op_desc["see_also_desc"]
+            )
+        doc_no_examples = base_doc.format(
+            desc=op_desc["desc"],
+            op_name=op_name,
+            equiv=equiv,
+            series_returns=op_desc["series_returns"],
+        )
+        ser_example = op_desc["series_examples"]
+        if ser_example:
+            doc = doc_no_examples + ser_example
+        else:
+            doc = doc_no_examples
+    elif typ == "dataframe":
+        if op_name in ["eq", "ne", "le", "lt", "ge", "gt"]:
+            base_doc = _flex_comp_doc_FRAME
+            doc = _flex_comp_doc_FRAME.format(
+                op_name=op_name,
+                desc=op_desc["desc"],
+            )
+        else:
+            base_doc = _flex_doc_FRAME
+            doc = base_doc.format(
+                desc=op_desc["desc"],
+                op_name=op_name,
+                equiv=equiv,
+                reverse=op_desc["reverse"],
+            )
+    else:
+        raise AssertionError("Invalid typ argument.")
+    return doc
+
+
+_common_examples_algebra_SERIES = """
+Examples
+--------
+>>> a = pd.Series([1, 1, 1, np.nan], index=['a', 'b', 'c', 'd'])
+>>> a
+a    1.0
+b    1.0
+c    1.0
+d    NaN
+dtype: float64
+>>> b = pd.Series([1, np.nan, 1, np.nan], index=['a', 'b', 'd', 'e'])
+>>> b
+a    1.0
+b    NaN
+d    1.0
+e    NaN
+dtype: float64"""
+
+_common_examples_comparison_SERIES = """
+Examples
+--------
+>>> a = pd.Series([1, 1, 1, np.nan, 1], index=['a', 'b', 'c', 'd', 'e'])
+>>> a
+a    1.0
+b    1.0
+c    1.0
+d    NaN
+e    1.0
+dtype: float64
+>>> b = pd.Series([0, 1, 2, np.nan, 1], index=['a', 'b', 'c', 'd', 'f'])
+>>> b
+a    0.0
+b    1.0
+c    2.0
+d    NaN
+f    1.0
+dtype: float64"""
+
+_add_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.add(b, fill_value=0)
+a    2.0
+b    1.0
+c    1.0
+d    1.0
+e    NaN
+dtype: float64
+"""
+)
+
+_sub_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.subtract(b, fill_value=0)
+a    0.0
+b    1.0
+c    1.0
+d   -1.0
+e    NaN
+dtype: float64
+"""
+)
+
+_mul_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.multiply(b, fill_value=0)
+a    1.0
+b    0.0
+c    0.0
+d    0.0
+e    NaN
+dtype: float64
+"""
+)
+
+_div_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.divide(b, fill_value=0)
+a    1.0
+b    inf
+c    inf
+d    0.0
+e    NaN
+dtype: float64
+"""
+)
+
+_floordiv_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.floordiv(b, fill_value=0)
+a    1.0
+b    inf
+c    inf
+d    0.0
+e    NaN
+dtype: float64
+"""
+)
+
+_divmod_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.divmod(b, fill_value=0)
+(a    1.0
+ b    inf
+ c    inf
+ d    0.0
+ e    NaN
+ dtype: float64,
+ a    0.0
+ b    NaN
+ c    NaN
+ d    0.0
+ e    NaN
+ dtype: float64)
+"""
+)
+
+_mod_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.mod(b, fill_value=0)
+a    0.0
+b    NaN
+c    NaN
+d    0.0
+e    NaN
+dtype: float64
+"""
+)
+_pow_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.pow(b, fill_value=0)
+a    1.0
+b    1.0
+c    1.0
+d    0.0
+e    NaN
+dtype: float64
+"""
+)
+
+_ne_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.ne(b, fill_value=0)
+a    False
+b     True
+c     True
+d     True
+e     True
+dtype: bool
+"""
+)
+
+_eq_example_SERIES = (
+    _common_examples_algebra_SERIES
+    + """
+>>> a.eq(b, fill_value=0)
+a     True
+b    False
+c    False
+d    False
+e    False
+dtype: bool
+"""
+)
+
+_lt_example_SERIES = (
+    _common_examples_comparison_SERIES
+    + """
+>>> a.lt(b, fill_value=0)
+a    False
+b    False
+c     True
+d    False
+e    False
+f     True
+dtype: bool
+"""
+)
+
+_le_example_SERIES = (
+    _common_examples_comparison_SERIES
+    + """
+>>> a.le(b, fill_value=0)
+a    False
+b     True
+c     True
+d    False
+e    False
+f     True
+dtype: bool
+"""
+)
+
+_gt_example_SERIES = (
+    _common_examples_comparison_SERIES
+    + """
+>>> a.gt(b, fill_value=0)
+a     True
+b    False
+c    False
+d    False
+e     True
+f    False
+dtype: bool
+"""
+)
+
+_ge_example_SERIES = (
+    _common_examples_comparison_SERIES
+    + """
+>>> a.ge(b, fill_value=0)
+a     True
+b     True
+c    False
+d    False
+e     True
+f    False
+dtype: bool
+"""
+)
+
+_returns_series = """Series\n    The result of the operation."""
+
+_returns_tuple = """2-Tuple of Series\n    The result of the operation."""
+
+_op_descriptions: dict[str, dict[str, str | None]] = {
+    # Arithmetic Operators
+    "add": {
+        "op": "+",
+        "desc": "Addition",
+        "reverse": "radd",
+        "series_examples": _add_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "sub": {
+        "op": "-",
+        "desc": "Subtraction",
+        "reverse": "rsub",
+        "series_examples": _sub_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "mul": {
+        "op": "*",
+        "desc": "Multiplication",
+        "reverse": "rmul",
+        "series_examples": _mul_example_SERIES,
+        "series_returns": _returns_series,
+        "df_examples": None,
+    },
+    "mod": {
+        "op": "%",
+        "desc": "Modulo",
+        "reverse": "rmod",
+        "series_examples": _mod_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "pow": {
+        "op": "**",
+        "desc": "Exponential power",
+        "reverse": "rpow",
+        "series_examples": _pow_example_SERIES,
+        "series_returns": _returns_series,
+        "df_examples": None,
+    },
+    "truediv": {
+        "op": "/",
+        "desc": "Floating division",
+        "reverse": "rtruediv",
+        "series_examples": _div_example_SERIES,
+        "series_returns": _returns_series,
+        "df_examples": None,
+    },
+    "floordiv": {
+        "op": "//",
+        "desc": "Integer division",
+        "reverse": "rfloordiv",
+        "series_examples": _floordiv_example_SERIES,
+        "series_returns": _returns_series,
+        "df_examples": None,
+    },
+    "divmod": {
+        "op": "divmod",
+        "desc": "Integer division and modulo",
+        "reverse": "rdivmod",
+        "series_examples": _divmod_example_SERIES,
+        "series_returns": _returns_tuple,
+        "df_examples": None,
+    },
+    # Comparison Operators
+    "eq": {
+        "op": "==",
+        "desc": "Equal to",
+        "reverse": None,
+        "series_examples": _eq_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "ne": {
+        "op": "!=",
+        "desc": "Not equal to",
+        "reverse": "eq",
+        "series_examples": _ne_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "lt": {
+        "op": "<",
+        "desc": "Less than",
+        "reverse": None,
+        "series_examples": _lt_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "le": {
+        "op": "<=",
+        "desc": "Less than or equal to",
+        "reverse": None,
+        "series_examples": _le_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "gt": {
+        "op": ">",
+        "desc": "Greater than",
+        "reverse": "lt",
+        "series_examples": _gt_example_SERIES,
+        "series_returns": _returns_series,
+    },
+    "ge": {
+        "op": ">=",
+        "desc": "Greater than or equal to",
+        "reverse": "le",
+        "series_examples": _ge_example_SERIES,
+        "series_returns": _returns_series,
+    },
+}
+
+_py_num_ref = """see
+    `Python documentation
+    <https://docs.python.org/3/reference/datamodel.html#emulating-numeric-types>`_
+    for more details"""
+_op_names = list(_op_descriptions.keys())
+for key in _op_names:
+    reverse_op = _op_descriptions[key]["reverse"]
+    if reverse_op is not None:
+        _op_descriptions[reverse_op] = _op_descriptions[key].copy()
+        _op_descriptions[reverse_op]["reverse"] = key
+        _op_descriptions[key]["see_also_desc"] = (
+            f"Reverse of the {_op_descriptions[key]['desc']} operator, {_py_num_ref}"
+        )
+        _op_descriptions[reverse_op]["see_also_desc"] = (
+            f"Element-wise {_op_descriptions[key]['desc']}, {_py_num_ref}"
+        )
+
+_flex_doc_SERIES = """
+Return {desc} of series and other, element-wise (binary operator `{op_name}`).
+
+Equivalent to ``{equiv}``, but with support to substitute a fill_value for
+missing data in either one of the inputs.
+
+Parameters
+----------
+other : Series or scalar value
+    The second operand in this operation.
+level : int or name
+    Broadcast across a level, matching Index values on the
+    passed MultiIndex level.
+fill_value : None or float value, default None (NaN)
+    Fill existing missing (NaN) values, and any new element needed for
+    successful Series alignment, with this value before computation.
+    If data in both corresponding Series locations is missing
+    the result of filling (at that location) will be missing.
+axis : {{0 or 'index'}}
+    Unused. Parameter needed for compatibility with DataFrame.
+
+Returns
+-------
+{series_returns}
+"""
+
+_see_also_reverse_SERIES = """
+See Also
+--------
+Series.{reverse} : {see_also_desc}.
+"""
+
+_flex_doc_FRAME = """
+Get {desc} of dataframe and other, element-wise (binary operator `{op_name}`).
+
+Equivalent to ``{equiv}``, but with support to substitute a fill_value
+for missing data in one of the inputs. With reverse version, `{reverse}`.
+
+Among flexible wrappers (`add`, `sub`, `mul`, `div`, `floordiv`, `mod`, `pow`) to
+arithmetic operators: `+`, `-`, `*`, `/`, `//`, `%`, `**`.
+
+Parameters
+----------
+other : scalar, sequence, Series, dict or DataFrame
+    Any single or multiple element data structure, or list-like object.
+axis : {{0 or 'index', 1 or 'columns'}}
+    Whether to compare by the index (0 or 'index') or columns.
+    (1 or 'columns'). For Series input, axis to match Series index on.
+level : int or label
+    Broadcast across a level, matching Index values on the
+    passed MultiIndex level.
+fill_value : float or None, default None
+    Fill existing missing (NaN) values, and any new element needed for
+    successful DataFrame alignment, with this value before computation.
+    If data in both corresponding DataFrame locations is missing
+    the result will be missing.
+
+Returns
+-------
+DataFrame
+    Result of the arithmetic operation.
+
+See Also
+--------
+DataFrame.add : Add DataFrames.
+DataFrame.sub : Subtract DataFrames.
+DataFrame.mul : Multiply DataFrames.
+DataFrame.div : Divide DataFrames (float division).
+DataFrame.truediv : Divide DataFrames (float division).
+DataFrame.floordiv : Divide DataFrames (integer division).
+DataFrame.mod : Calculate modulo (remainder after division).
+DataFrame.pow : Calculate exponential power.
+
+Notes
+-----
+Mismatched indices will be unioned together.
+
+Examples
+--------
+>>> df = pd.DataFrame({{'angles': [0, 3, 4],
+...                    'degrees': [360, 180, 360]}},
+...                   index=['circle', 'triangle', 'rectangle'])
+>>> df
+           angles  degrees
+circle          0      360
+triangle        3      180
+rectangle       4      360
+
+Add a scalar with operator version which return the same
+results.
+
+>>> df + 1
+           angles  degrees
+circle          1      361
+triangle        4      181
+rectangle       5      361
+
+>>> df.add(1)
+           angles  degrees
+circle          1      361
+triangle        4      181
+rectangle       5      361
+
+Divide by constant with reverse version.
+
+>>> df.div(10)
+           angles  degrees
+circle        0.0     36.0
+triangle      0.3     18.0
+rectangle     0.4     36.0
+
+>>> df.rdiv(10)
+             angles   degrees
+circle          inf  0.027778
+triangle   3.333333  0.055556
+rectangle  2.500000  0.027778
+
+Subtract a list and Series by axis with operator version.
+
+>>> df - [1, 2]
+           angles  degrees
+circle         -1      358
+triangle        2      178
+rectangle       3      358
+
+>>> df.sub([1, 2], axis='columns')
+           angles  degrees
+circle         -1      358
+triangle        2      178
+rectangle       3      358
+
+>>> df.sub(pd.Series([1, 1, 1], index=['circle', 'triangle', 'rectangle']),
+...        axis='index')
+           angles  degrees
+circle         -1      359
+triangle        2      179
+rectangle       3      359
+
+Multiply a dictionary by axis.
+
+>>> df.mul({{'angles': 0, 'degrees': 2}})
+            angles  degrees
+circle           0      720
+triangle         0      360
+rectangle        0      720
+
+>>> df.mul({{'circle': 0, 'triangle': 2, 'rectangle': 3}}, axis='index')
+            angles  degrees
+circle           0        0
+triangle         6      360
+rectangle       12     1080
+
+Multiply a DataFrame of different shape with operator version.
+
+>>> other = pd.DataFrame({{'angles': [0, 3, 4]}},
+...                      index=['circle', 'triangle', 'rectangle'])
+>>> other
+           angles
+circle          0
+triangle        3
+rectangle       4
+
+>>> df * other
+           angles  degrees
+circle          0      NaN
+triangle        9      NaN
+rectangle      16      NaN
+
+>>> df.mul(other, fill_value=0)
+           angles  degrees
+circle          0      0.0
+triangle        9      0.0
+rectangle      16      0.0
+
+Divide by a MultiIndex by level.
+
+>>> df_multindex = pd.DataFrame({{'angles': [0, 3, 4, 4, 5, 6],
+...                              'degrees': [360, 180, 360, 360, 540, 720]}},
+...                             index=[['A', 'A', 'A', 'B', 'B', 'B'],
+...                                    ['circle', 'triangle', 'rectangle',
+...                                     'square', 'pentagon', 'hexagon']])
+>>> df_multindex
+             angles  degrees
+A circle          0      360
+  triangle        3      180
+  rectangle       4      360
+B square          4      360
+  pentagon        5      540
+  hexagon         6      720
+
+>>> df.div(df_multindex, level=1, fill_value=0)
+             angles  degrees
+A circle        NaN      1.0
+  triangle      1.0      1.0
+  rectangle     1.0      1.0
+B square        0.0      0.0
+  pentagon      0.0      0.0
+  hexagon       0.0      0.0
+
+>>> df_pow = pd.DataFrame({{'A': [2, 3, 4, 5],
+...                        'B': [6, 7, 8, 9]}})
+>>> df_pow.pow(2)
+    A   B
+0   4  36
+1   9  49
+2  16  64
+3  25  81
+"""
+
+_flex_comp_doc_FRAME = """
+Get {desc} of dataframe and other, element-wise (binary operator `{op_name}`).
+
+Among flexible wrappers (`eq`, `ne`, `le`, `lt`, `ge`, `gt`) to comparison
+operators.
+
+Equivalent to `==`, `!=`, `<=`, `<`, `>=`, `>` with support to choose axis
+(rows or columns) and level for comparison.
+
+Parameters
+----------
+other : scalar, sequence, Series, or DataFrame
+    Any single or multiple element data structure, or list-like object.
+axis : {{0 or 'index', 1 or 'columns'}}, default 'columns'
+    Whether to compare by the index (0 or 'index') or columns
+    (1 or 'columns').
+level : int or label
+    Broadcast across a level, matching Index values on the passed
+    MultiIndex level.
+
+Returns
+-------
+DataFrame of bool
+    Result of the comparison.
+
+See Also
+--------
+DataFrame.eq : Compare DataFrames for equality elementwise.
+DataFrame.ne : Compare DataFrames for inequality elementwise.
+DataFrame.le : Compare DataFrames for less than inequality
+    or equality elementwise.
+DataFrame.lt : Compare DataFrames for strictly less than
+    inequality elementwise.
+DataFrame.ge : Compare DataFrames for greater than inequality
+    or equality elementwise.
+DataFrame.gt : Compare DataFrames for strictly greater than
+    inequality elementwise.
+
+Notes
+-----
+Mismatched indices will be unioned together.
+`NaN` values are considered different (i.e. `NaN` != `NaN`).
+
+Examples
+--------
+>>> df = pd.DataFrame({{'cost': [250, 150, 100],
+...                    'revenue': [100, 250, 300]}},
+...                   index=['A', 'B', 'C'])
+>>> df
+   cost  revenue
+A   250      100
+B   150      250
+C   100      300
+
+Comparison with a scalar, using either the operator or method:
+
+>>> df == 100
+    cost  revenue
+A  False     True
+B  False    False
+C   True    False
+
+>>> df.eq(100)
+    cost  revenue
+A  False     True
+B  False    False
+C   True    False
+
+When `other` is a :class:`Series`, the columns of a DataFrame are aligned
+with the index of `other` and broadcast:
+
+>>> df != pd.Series([100, 250], index=["cost", "revenue"])
+    cost  revenue
+A   True     True
+B   True    False
+C  False     True
+
+Use the method to control the broadcast axis:
+
+>>> df.ne(pd.Series([100, 300], index=["A", "D"]), axis='index')
+   cost  revenue
+A  True    False
+B  True     True
+C  True     True
+D  True     True
+
+When comparing to an arbitrary sequence, the number of columns must
+match the number elements in `other`:
+
+>>> df == [250, 100]
+    cost  revenue
+A   True     True
+B  False    False
+C  False    False
+
+Use the method to control the axis:
+
+>>> df.eq([250, 250, 100], axis='index')
+    cost  revenue
+A   True    False
+B  False     True
+C   True    False
+
+Compare to a DataFrame of different shape.
+
+>>> other = pd.DataFrame({{'revenue': [300, 250, 100, 150]}},
+...                      index=['A', 'B', 'C', 'D'])
+>>> other
+   revenue
+A      300
+B      250
+C      100
+D      150
+
+>>> df.gt(other)
+    cost  revenue
+A  False    False
+B  False    False
+C  False     True
+D  False    False
+
+Compare to a MultiIndex by level.
+
+>>> df_multindex = pd.DataFrame({{'cost': [250, 150, 100, 150, 300, 220],
+...                              'revenue': [100, 250, 300, 200, 175, 225]}},
+...                             index=[['Q1', 'Q1', 'Q1', 'Q2', 'Q2', 'Q2'],
+...                                    ['A', 'B', 'C', 'A', 'B', 'C']])
+>>> df_multindex
+      cost  revenue
+Q1 A   250      100
+   B   150      250
+   C   100      300
+Q2 A   150      200
+   B   300      175
+   C   220      225
+
+>>> df.le(df_multindex, level=1)
+       cost  revenue
+Q1 A   True     True
+   B   True     True
+   C   True     True
+Q2 A  False     True
+   B   True    False
+   C   True    False
+"""
diff --git a/pandas/pandas/core/ops/invalid.py b/pandas/pandas/core/ops/invalid.py
new file mode 100644
index 0000000000000..395db1617cb63
--- /dev/null
+++ b/pandas/pandas/core/ops/invalid.py
@@ -0,0 +1,77 @@
+"""
+Templates for invalid operations.
+"""
+
+from __future__ import annotations
+
+import operator
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    NoReturn,
+)
+
+import numpy as np
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        ArrayLike,
+        Scalar,
+        npt,
+    )
+
+
+def invalid_comparison(
+    left: ArrayLike,
+    right: ArrayLike | Scalar,
+    op: Callable[[Any, Any], bool],
+) -> npt.NDArray[np.bool_]:
+    """
+    If a comparison has mismatched types and is not necessarily meaningful,
+    follow python3 conventions by:
+
+        - returning all-False for equality
+        - returning all-True for inequality
+        - raising TypeError otherwise
+
+    Parameters
+    ----------
+    left : array-like
+    right : scalar, array-like
+    op : operator.{eq, ne, lt, le, gt}
+
+    Raises
+    ------
+    TypeError : on inequality comparisons
+    """
+    if op is operator.eq:
+        res_values = np.zeros(left.shape, dtype=bool)
+    elif op is operator.ne:
+        res_values = np.ones(left.shape, dtype=bool)
+    else:
+        typ = type(right).__name__
+        raise TypeError(f"Invalid comparison between dtype={left.dtype} and {typ}")
+    return res_values
+
+
+def make_invalid_op(name: str) -> Callable[..., NoReturn]:
+    """
+    Return a binary method that always raises a TypeError.
+
+    Parameters
+    ----------
+    name : str
+
+    Returns
+    -------
+    invalid_op : function
+    """
+
+    def invalid_op(self: object, other: object = None) -> NoReturn:
+        typ = type(self).__name__
+        raise TypeError(f"cannot perform {name} with this index type: {typ}")
+
+    invalid_op.__name__ = name
+    return invalid_op
diff --git a/pandas/pandas/core/ops/mask_ops.py b/pandas/pandas/core/ops/mask_ops.py
new file mode 100644
index 0000000000000..86a93e5a3ca2b
--- /dev/null
+++ b/pandas/pandas/core/ops/mask_ops.py
@@ -0,0 +1,195 @@
+"""
+Ops for masked arrays.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+
+def kleene_or(
+    left: bool | np.ndarray | libmissing.NAType,
+    right: bool | np.ndarray | libmissing.NAType,
+    left_mask: np.ndarray | None,
+    right_mask: np.ndarray | None,
+) -> tuple[npt.NDArray[np.bool_], npt.NDArray[np.bool_]]:
+    """
+    Boolean ``or`` using Kleene logic.
+
+    Values are NA where we have ``NA | NA`` or ``NA | False``.
+    ``NA | True`` is considered True.
+
+    Parameters
+    ----------
+    left, right : ndarray, NA, or bool
+        The values of the array.
+    left_mask, right_mask : ndarray, optional
+        The masks. Only one of these may be None, which implies that
+        the associated `left` or `right` value is a scalar.
+
+    Returns
+    -------
+    result, mask: ndarray[bool]
+        The result of the logical or, and the new mask.
+    """
+    # To reduce the number of cases, we ensure that `left` & `left_mask`
+    # always come from an array, not a scalar. This is safe, since
+    # A | B == B | A
+    if left_mask is None:
+        return kleene_or(right, left, right_mask, left_mask)
+
+    if not isinstance(left, np.ndarray):
+        raise TypeError("Either `left` or `right` need to be a np.ndarray.")
+
+    raise_for_nan(right, method="or")
+
+    if right is libmissing.NA:
+        result = left.copy()
+    else:
+        result = left | right
+
+    if right_mask is not None:
+        # output is unknown where (False & NA), (NA & False), (NA & NA)
+        left_false = ~(left | left_mask)
+        right_false = ~(right | right_mask)
+        mask = (
+            (left_false & right_mask)
+            | (right_false & left_mask)
+            | (left_mask & right_mask)
+        )
+    else:
+        if right is True:
+            mask = np.zeros_like(left_mask)
+        elif right is libmissing.NA:
+            mask = (~left & ~left_mask) | left_mask
+        else:
+            # False
+            mask = left_mask.copy()
+
+    return result, mask
+
+
+def kleene_xor(
+    left: bool | np.ndarray | libmissing.NAType,
+    right: bool | np.ndarray | libmissing.NAType,
+    left_mask: np.ndarray | None,
+    right_mask: np.ndarray | None,
+) -> tuple[npt.NDArray[np.bool_], npt.NDArray[np.bool_]]:
+    """
+    Boolean ``xor`` using Kleene logic.
+
+    This is the same as ``or``, with the following adjustments
+
+    * True, True -> False
+    * True, NA   -> NA
+
+    Parameters
+    ----------
+    left, right : ndarray, NA, or bool
+        The values of the array.
+    left_mask, right_mask : ndarray, optional
+        The masks. Only one of these may be None, which implies that
+        the associated `left` or `right` value is a scalar.
+
+    Returns
+    -------
+    result, mask: ndarray[bool]
+        The result of the logical xor, and the new mask.
+    """
+    # To reduce the number of cases, we ensure that `left` & `left_mask`
+    # always come from an array, not a scalar. This is safe, since
+    # A ^ B == B ^ A
+    if left_mask is None:
+        return kleene_xor(right, left, right_mask, left_mask)
+
+    if not isinstance(left, np.ndarray):
+        raise TypeError("Either `left` or `right` need to be a np.ndarray.")
+
+    raise_for_nan(right, method="xor")
+    if right is libmissing.NA:
+        result = np.zeros_like(left)
+    else:
+        result = left ^ right
+
+    if right_mask is None:
+        if right is libmissing.NA:
+            mask = np.ones_like(left_mask)
+        else:
+            mask = left_mask.copy()
+    else:
+        mask = left_mask | right_mask
+
+    return result, mask
+
+
+def kleene_and(
+    left: bool | libmissing.NAType | np.ndarray,
+    right: bool | libmissing.NAType | np.ndarray,
+    left_mask: np.ndarray | None,
+    right_mask: np.ndarray | None,
+) -> tuple[npt.NDArray[np.bool_], npt.NDArray[np.bool_]]:
+    """
+    Boolean ``and`` using Kleene logic.
+
+    Values are ``NA`` for ``NA & NA`` or ``True & NA``.
+
+    Parameters
+    ----------
+    left, right : ndarray, NA, or bool
+        The values of the array.
+    left_mask, right_mask : ndarray, optional
+        The masks. Only one of these may be None, which implies that
+        the associated `left` or `right` value is a scalar.
+
+    Returns
+    -------
+    result, mask: ndarray[bool]
+        The result of the logical xor, and the new mask.
+    """
+    # To reduce the number of cases, we ensure that `left` & `left_mask`
+    # always come from an array, not a scalar. This is safe, since
+    # A & B == B & A
+    if left_mask is None:
+        return kleene_and(right, left, right_mask, left_mask)
+
+    if not isinstance(left, np.ndarray):
+        raise TypeError("Either `left` or `right` need to be a np.ndarray.")
+    raise_for_nan(right, method="and")
+
+    if right is libmissing.NA:
+        result = np.zeros_like(left)
+    else:
+        result = left & right
+
+    if right_mask is None:
+        # Scalar `right`
+        if right is libmissing.NA:
+            mask = (left & ~left_mask) | left_mask
+
+        else:
+            mask = left_mask.copy()
+            if right is False:
+                # unmask everything
+                mask[:] = False
+    else:
+        # unmask where either left or right is False
+        left_false = ~(left | left_mask)
+        right_false = ~(right | right_mask)
+        mask = (left_mask & ~right_false) | (right_mask & ~left_false)
+
+    return result, mask
+
+
+def raise_for_nan(value: object, method: str) -> None:
+    if lib.is_float(value) and np.isnan(value):
+        raise ValueError(f"Cannot perform logical '{method}' with floating NaN")
diff --git a/pandas/pandas/core/ops/missing.py b/pandas/pandas/core/ops/missing.py
new file mode 100644
index 0000000000000..a4e9e5305f74d
--- /dev/null
+++ b/pandas/pandas/core/ops/missing.py
@@ -0,0 +1,177 @@
+"""
+Missing data handling for arithmetic operations.
+
+In particular, pandas conventions regarding division by zero differ
+from numpy in the following ways:
+    1) np.array([-1, 0, 1], dtype=dtype1) // np.array([0, 0, 0], dtype=dtype2)
+       gives [nan, nan, nan] for most dtype combinations, and [0, 0, 0] for
+       the remaining pairs
+       (the remaining being dtype1==dtype2==intN and dtype==dtype2==uintN).
+
+       pandas convention is to return [-inf, nan, inf] for all dtype
+       combinations.
+
+       Note: the numpy behavior described here is py3-specific.
+
+    2) np.array([-1, 0, 1], dtype=dtype1) % np.array([0, 0, 0], dtype=dtype2)
+       gives precisely the same results as the // operation.
+
+       pandas convention is to return [nan, nan, nan] for all dtype
+       combinations.
+
+    3) divmod behavior consistent with 1) and 2).
+"""
+
+from __future__ import annotations
+
+import operator
+
+import numpy as np
+
+from pandas.core import roperator
+
+
+def _fill_zeros(result: np.ndarray, x, y) -> np.ndarray:
+    """
+    If this is a reversed op, then flip x,y
+
+    If we have an integer value (or array in y)
+    and we have 0's, fill them with np.nan,
+    return the result.
+
+    Mask the nan's from x.
+    """
+    if result.dtype.kind == "f":
+        return result
+
+    is_variable_type = hasattr(y, "dtype")
+    is_scalar_type = not isinstance(y, np.ndarray)
+
+    if not is_variable_type and not is_scalar_type:
+        # e.g. test_series_ops_name_retention with mod we get here with list/tuple
+        return result
+
+    if is_scalar_type:
+        y = np.array(y)
+
+    if y.dtype.kind in "iu":
+        ymask = y == 0
+        if ymask.any():
+            # GH#7325, mask and nans must be broadcastable
+            mask = ymask & ~np.isnan(result)
+
+            # GH#9308 doing ravel on result and mask can improve putmask perf,
+            #  but can also make unwanted copies.
+            result = result.astype("float64", copy=False)
+
+            np.putmask(result, mask, np.nan)
+
+    return result
+
+
+def mask_zero_div_zero(x, y, result: np.ndarray) -> np.ndarray:
+    """
+    Set results of  0 // 0 to np.nan, regardless of the dtypes
+    of the numerator or the denominator.
+
+    Parameters
+    ----------
+    x : ndarray
+    y : ndarray
+    result : ndarray
+
+    Returns
+    -------
+    ndarray
+        The filled result.
+
+    Examples
+    --------
+    >>> x = np.array([1, 0, -1], dtype=np.int64)
+    >>> x
+    array([ 1,  0, -1])
+    >>> y = 0  # int 0; numpy behavior is different with float
+    >>> result = x // y
+    >>> result  # raw numpy result does not fill division by zero
+    array([0, 0, 0])
+    >>> mask_zero_div_zero(x, y, result)
+    array([ inf,  nan, -inf])
+    """
+
+    if not hasattr(y, "dtype"):
+        # e.g. scalar, tuple
+        y = np.array(y)
+    if not hasattr(x, "dtype"):
+        # e.g scalar, tuple
+        x = np.array(x)
+
+    zmask = y == 0
+
+    if zmask.any():
+        # Flip sign if necessary for -0.0
+        zneg_mask = zmask & np.signbit(y)
+        zpos_mask = zmask & ~zneg_mask
+
+        x_lt0 = x < 0
+        x_gt0 = x > 0
+        nan_mask = zmask & (x == 0)
+        neginf_mask = (zpos_mask & x_lt0) | (zneg_mask & x_gt0)
+        posinf_mask = (zpos_mask & x_gt0) | (zneg_mask & x_lt0)
+
+        if nan_mask.any() or neginf_mask.any() or posinf_mask.any():
+            # Fill negative/0 with -inf, positive/0 with +inf, 0/0 with NaN
+            result = result.astype("float64", copy=False)
+
+            result[nan_mask] = np.nan
+            result[posinf_mask] = np.inf
+            result[neginf_mask] = -np.inf
+
+    return result
+
+
+def dispatch_fill_zeros(op, left, right, result):
+    """
+    Call _fill_zeros with the appropriate fill value depending on the operation,
+    with special logic for divmod and rdivmod.
+
+    Parameters
+    ----------
+    op : function (operator.add, operator.div, ...)
+    left : object (np.ndarray for non-reversed ops)
+        We have excluded ExtensionArrays here
+    right : object (np.ndarray for reversed ops)
+        We have excluded ExtensionArrays here
+    result : ndarray
+
+    Returns
+    -------
+    result : np.ndarray
+
+    Notes
+    -----
+    For divmod and rdivmod, the `result` parameter and returned `result`
+    is a 2-tuple of ndarray objects.
+    """
+    if op is divmod:
+        result = (
+            mask_zero_div_zero(left, right, result[0]),
+            _fill_zeros(result[1], left, right),
+        )
+    elif op is roperator.rdivmod:
+        result = (
+            mask_zero_div_zero(right, left, result[0]),
+            _fill_zeros(result[1], right, left),
+        )
+    elif op is operator.floordiv:
+        # Note: no need to do this for truediv; in py3 numpy behaves the way
+        #  we want.
+        result = mask_zero_div_zero(left, right, result)
+    elif op is roperator.rfloordiv:
+        # Note: no need to do this for rtruediv; in py3 numpy behaves the way
+        #  we want.
+        result = mask_zero_div_zero(right, left, result)
+    elif op is operator.mod:
+        result = _fill_zeros(result, left, right)
+    elif op is roperator.rmod:
+        result = _fill_zeros(result, right, left)
+    return result
diff --git a/pandas/pandas/core/resample.py b/pandas/pandas/core/resample.py
new file mode 100644
index 0000000000000..1cfc75ea11725
--- /dev/null
+++ b/pandas/pandas/core/resample.py
@@ -0,0 +1,2900 @@
+from __future__ import annotations
+
+import copy
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+    final,
+    no_type_check,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    BaseOffset,
+    IncompatibleFrequency,
+    NaT,
+    Period,
+    Timedelta,
+    Timestamp,
+    to_offset,
+)
+from pandas._typing import NDFrameT
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    doc,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+import pandas.core.algorithms as algos
+from pandas.core.apply import ResamplerWindowApply
+from pandas.core.arrays import ArrowExtensionArray
+from pandas.core.base import (
+    PandasObject,
+    SelectionMixin,
+)
+from pandas.core.generic import (
+    NDFrame,
+    _shared_docs,
+)
+from pandas.core.groupby.groupby import (
+    BaseGroupBy,
+    GroupBy,
+    _pipe_template,
+    get_groupby,
+)
+from pandas.core.groupby.grouper import Grouper
+from pandas.core.groupby.ops import BinGrouper
+from pandas.core.indexes.api import MultiIndex
+from pandas.core.indexes.base import Index
+from pandas.core.indexes.datetimes import (
+    DatetimeIndex,
+    date_range,
+)
+from pandas.core.indexes.period import (
+    PeriodIndex,
+    period_range,
+)
+from pandas.core.indexes.timedeltas import (
+    TimedeltaIndex,
+    timedelta_range,
+)
+from pandas.core.reshape.concat import concat
+
+from pandas.tseries.frequencies import (
+    is_subperiod,
+    is_superperiod,
+)
+from pandas.tseries.offsets import (
+    Day,
+    Tick,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+    )
+
+    from pandas._typing import (
+        Any,
+        AnyArrayLike,
+        Axis,
+        Concatenate,
+        FreqIndexT,
+        Frequency,
+        IndexLabel,
+        InterpolateOptions,
+        P,
+        Self,
+        T,
+        TimedeltaConvertibleTypes,
+        TimeGrouperOrigin,
+        TimestampConvertibleTypes,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+_shared_docs_kwargs: dict[str, str] = {}
+
+
+class Resampler(BaseGroupBy, PandasObject):
+    """
+    Class for resampling datetimelike data, a groupby-like operation.
+    See aggregate, transform, and apply functions on this object.
+
+    It's easiest to use obj.resample(...) to use Resampler.
+
+    Parameters
+    ----------
+    obj : Series or DataFrame
+    groupby : TimeGrouper
+
+    Returns
+    -------
+    a Resampler of the appropriate type
+
+    Notes
+    -----
+    After resampling, see aggregate, apply, and transform functions.
+    """
+
+    _grouper: BinGrouper
+    _timegrouper: TimeGrouper
+    binner: DatetimeIndex | TimedeltaIndex | PeriodIndex  # depends on subclass
+    exclusions: frozenset[Hashable] = frozenset()  # for SelectionMixin compat
+    _internal_names_set = set({"obj", "ax", "_indexer"})
+
+    # to the groupby descriptor
+    _attributes = [
+        "freq",
+        "closed",
+        "label",
+        "convention",
+        "origin",
+        "offset",
+    ]
+
+    def __init__(
+        self,
+        obj: NDFrame,
+        timegrouper: TimeGrouper,
+        *,
+        gpr_index: Index,
+        group_keys: bool = False,
+        selection=None,
+        include_groups: bool = False,
+    ) -> None:
+        if include_groups:
+            raise ValueError("include_groups=True is no longer allowed.")
+        self._timegrouper = timegrouper
+        self.keys = None
+        self.sort = True
+        self.group_keys = group_keys
+        self.as_index = True
+
+        self.obj, self.ax, self._indexer = self._timegrouper._set_grouper(
+            self._convert_obj(obj), sort=True, gpr_index=gpr_index
+        )
+        self.binner, self._grouper = self._get_binner()
+        self._selection = selection
+        if self._timegrouper.key is not None:
+            self.exclusions = frozenset([self._timegrouper.key])
+        else:
+            self.exclusions = frozenset()
+
+    @final
+    def __str__(self) -> str:
+        """
+        Provide a nice str repr of our rolling object.
+        """
+        attrs = (
+            f"{k}={getattr(self._timegrouper, k)}"
+            for k in self._attributes
+            if getattr(self._timegrouper, k, None) is not None
+        )
+        return f"{type(self).__name__} [{', '.join(attrs)}]"
+
+    @final
+    def __getattr__(self, attr: str):
+        if attr in self._internal_names_set:
+            return object.__getattribute__(self, attr)
+        if attr in self._attributes:
+            return getattr(self._timegrouper, attr)
+        if attr in self.obj:
+            return self[attr]
+
+        return object.__getattribute__(self, attr)
+
+    @final
+    @property
+    def _from_selection(self) -> bool:
+        """
+        Is the resampling from a DataFrame column or MultiIndex level.
+        """
+        # upsampling and PeriodIndex resampling do not work
+        # with selection, this state used to catch and raise an error
+        return self._timegrouper is not None and (
+            self._timegrouper.key is not None or self._timegrouper.level is not None
+        )
+
+    def _convert_obj(self, obj: NDFrameT) -> NDFrameT:
+        """
+        Provide any conversions for the object in order to correctly handle.
+
+        Parameters
+        ----------
+        obj : Series or DataFrame
+
+        Returns
+        -------
+        Series or DataFrame
+        """
+        return obj._consolidate()
+
+    def _get_binner_for_time(self):
+        raise AbstractMethodError(self)
+
+    @final
+    def _get_binner(self):
+        """
+        Create the BinGrouper, assume that self.set_grouper(obj)
+        has already been called.
+        """
+        binner, bins, binlabels = self._get_binner_for_time()
+        assert len(bins) == len(binlabels)
+        bin_grouper = BinGrouper(bins, binlabels, indexer=self._indexer)
+        return binner, bin_grouper
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    @final
+    @Substitution(
+        klass="Resampler",
+        examples="""
+    >>> df = pd.DataFrame({'A': [1, 2, 3, 4]},
+    ...                   index=pd.date_range('2012-08-02', periods=4))
+    >>> df
+                A
+    2012-08-02  1
+    2012-08-03  2
+    2012-08-04  3
+    2012-08-05  4
+
+    To get the difference between each 2-day period's maximum and minimum
+    value in one pass, you can do
+
+    >>> df.resample('2D').pipe(lambda x: x.max() - x.min())
+                A
+    2012-08-02  1
+    2012-08-04  1""",
+    )
+    @Appender(_pipe_template)
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        return super().pipe(func, *args, **kwargs)
+
+    _agg_see_also_doc = dedent(
+        """
+    See Also
+    --------
+    DataFrame.groupby.aggregate : Aggregate using callable, string, dict,
+        or list of string/callables.
+    DataFrame.resample.transform : Transforms the Series on each group
+        based on the given function.
+    DataFrame.aggregate: Aggregate using one or more
+        operations over the specified axis.
+    """
+    )
+
+    _agg_examples_doc = dedent(
+        """
+    Examples
+    --------
+    >>> s = pd.Series([1, 2, 3, 4, 5],
+    ...               index=pd.date_range('20130101', periods=5, freq='s'))
+    >>> s
+    2013-01-01 00:00:00    1
+    2013-01-01 00:00:01    2
+    2013-01-01 00:00:02    3
+    2013-01-01 00:00:03    4
+    2013-01-01 00:00:04    5
+    Freq: s, dtype: int64
+
+    >>> r = s.resample('2s')
+
+    >>> r.agg("sum")
+    2013-01-01 00:00:00    3
+    2013-01-01 00:00:02    7
+    2013-01-01 00:00:04    5
+    Freq: 2s, dtype: int64
+
+    >>> r.agg(['sum', 'mean', 'max'])
+                         sum  mean  max
+    2013-01-01 00:00:00    3   1.5    2
+    2013-01-01 00:00:02    7   3.5    4
+    2013-01-01 00:00:04    5   5.0    5
+
+    >>> r.agg({'result': lambda x: x.mean() / x.std(),
+    ...        'total': "sum"})
+                           result  total
+    2013-01-01 00:00:00  2.121320      3
+    2013-01-01 00:00:02  4.949747      7
+    2013-01-01 00:00:04       NaN      5
+
+    >>> r.agg(average="mean", total="sum")
+                             average  total
+    2013-01-01 00:00:00      1.5      3
+    2013-01-01 00:00:02      3.5      7
+    2013-01-01 00:00:04      5.0      5
+    """
+    )
+
+    @final
+    @doc(
+        _shared_docs["aggregate"],
+        see_also=_agg_see_also_doc,
+        examples=_agg_examples_doc,
+        klass="DataFrame",
+        axis="",
+    )
+    def aggregate(self, func=None, *args, **kwargs):
+        result = ResamplerWindowApply(self, func, args=args, kwargs=kwargs).agg()
+        if result is None:
+            how = func
+            result = self._groupby_and_aggregate(how, *args, **kwargs)
+
+        return result
+
+    agg = aggregate
+    apply = aggregate
+
+    @final
+    def transform(self, arg, *args, **kwargs):
+        """
+        Call function producing a like-indexed Series on each group.
+
+        Return a Series with the transformed values.
+
+        Parameters
+        ----------
+        arg : function
+            To apply to each group. Should return a Series with the same index.
+        *args, **kwargs
+            Additional arguments and keywords.
+
+        Returns
+        -------
+        Series
+            A Series with the transformed values, maintaining the same index as
+            the original object.
+
+        See Also
+        --------
+        core.resample.Resampler.apply : Apply a function along each group.
+        core.resample.Resampler.aggregate : Aggregate using one or more operations
+            over the specified axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2], index=pd.date_range("20180101", periods=2, freq="1h"))
+        >>> s
+        2018-01-01 00:00:00    1
+        2018-01-01 01:00:00    2
+        Freq: h, dtype: int64
+
+        >>> resampled = s.resample("15min")
+        >>> resampled.transform(lambda x: (x - x.mean()) / x.std())
+        2018-01-01 00:00:00   NaN
+        2018-01-01 01:00:00   NaN
+        Freq: h, dtype: float64
+        """
+        return self._selected_obj.groupby(self._timegrouper).transform(
+            arg, *args, **kwargs
+        )
+
+    def _downsample(self, how, **kwargs):
+        raise AbstractMethodError(self)
+
+    def _upsample(self, f, limit: int | None = None, fill_value=None):
+        raise AbstractMethodError(self)
+
+    def _gotitem(self, key, ndim: int, subset=None):
+        """
+        Sub-classes to define. Return a sliced object.
+
+        Parameters
+        ----------
+        key : string / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        grouper = self._grouper
+        if subset is None:
+            subset = self.obj
+            if key is not None:
+                subset = subset[key]
+            else:
+                # reached via Apply.agg_dict_like with selection=None and ndim=1
+                assert subset.ndim == 1
+        if ndim == 1:
+            assert subset.ndim == 1
+
+        grouped = get_groupby(
+            subset, by=None, grouper=grouper, group_keys=self.group_keys
+        )
+        return grouped
+
+    def _groupby_and_aggregate(self, how, *args, **kwargs):
+        """
+        Re-evaluate the obj with a groupby aggregation.
+        """
+        grouper = self._grouper
+
+        # Excludes `on` column when provided
+        obj = self._obj_with_exclusions
+
+        grouped = get_groupby(obj, by=None, grouper=grouper, group_keys=self.group_keys)
+
+        try:
+            if callable(how):
+                # TODO: test_resample_apply_with_additional_args fails if we go
+                #  through the non-lambda path, not clear that it should.
+                func = lambda x: how(x, *args, **kwargs)
+                result = grouped.aggregate(func)
+            else:
+                result = grouped.aggregate(how, *args, **kwargs)
+        except (AttributeError, KeyError):
+            # we have a non-reducing function; try to evaluate
+            # alternatively we want to evaluate only a column of the input
+
+            # test_apply_to_one_column_of_df the function being applied references
+            #  a DataFrame column, but aggregate_item_by_item operates column-wise
+            #  on Series, raising AttributeError or KeyError
+            #  (depending on whether the column lookup uses getattr/__getitem__)
+            result = grouped.apply(how, *args, **kwargs)
+
+        except ValueError as err:
+            if "Must produce aggregated value" in str(err):
+                # raised in _aggregate_named
+                # see test_apply_without_aggregation, test_apply_with_mutated_index
+                pass
+            else:
+                raise
+
+            # we have a non-reducing function
+            # try to evaluate
+            result = grouped.apply(how, *args, **kwargs)
+
+        return self._wrap_result(result)
+
+    @final
+    def _get_resampler_for_grouping(
+        self,
+        groupby: GroupBy,
+        key,
+    ):
+        """
+        Return the correct class for resampling with groupby.
+        """
+        return self._resampler_for_grouping(
+            groupby=groupby,
+            key=key,
+            parent=self,
+        )
+
+    def _wrap_result(self, result):
+        """
+        Potentially wrap any results.
+        """
+        # GH 47705
+        obj = self.obj
+        if (
+            isinstance(result, ABCDataFrame)
+            and len(result) == 0
+            and not isinstance(result.index, PeriodIndex)
+        ):
+            result = result.set_index(
+                _asfreq_compat(obj.index[:0], freq=self.freq), append=True
+            )
+
+        if isinstance(result, ABCSeries) and self._selection is not None:
+            result.name = self._selection
+
+        if isinstance(result, ABCSeries) and result.empty:
+            # When index is all NaT, result is empty but index is not
+            result.index = _asfreq_compat(obj.index[:0], freq=self.freq)
+            result.name = getattr(obj, "name", None)
+
+        if self._timegrouper._arrow_dtype is not None:
+            result.index = result.index.astype(self._timegrouper._arrow_dtype)
+
+        return result
+
+    @final
+    def ffill(self, limit: int | None = None):
+        """
+        Forward fill the values.
+
+        This method fills missing values by propagating the last valid
+        observation forward, up to the next valid observation. It is commonly
+        used in time series analysis when resampling data to a higher frequency
+        (upsampling) and filling gaps in the resampled output.
+
+        Parameters
+        ----------
+        limit : int, optional
+            Limit of how many values to fill.
+
+        Returns
+        -------
+        Series
+            The resampled data with missing values filled forward.
+
+        See Also
+        --------
+        Series.fillna: Fill NA/NaN values using the specified method.
+        DataFrame.fillna: Fill NA/NaN values using the specified method.
+
+        Examples
+        --------
+        Here we only create a ``Series``.
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+
+        Example for ``ffill`` with downsampling (we have fewer dates after resampling):
+
+        >>> ser.resample("MS").ffill()
+        2023-01-01    1
+        2023-02-01    3
+        Freq: MS, dtype: int64
+
+        Example for ``ffill`` with upsampling (fill the new dates with
+        the previous value):
+
+        >>> ser.resample("W").ffill()
+        2023-01-01    1
+        2023-01-08    1
+        2023-01-15    2
+        2023-01-22    2
+        2023-01-29    2
+        2023-02-05    3
+        2023-02-12    3
+        2023-02-19    4
+        Freq: W-SUN, dtype: int64
+
+        With upsampling and limiting (only fill the first new date with the
+        previous value):
+
+        >>> ser.resample("W").ffill(limit=1)
+        2023-01-01    1.0
+        2023-01-08    1.0
+        2023-01-15    2.0
+        2023-01-22    2.0
+        2023-01-29    NaN
+        2023-02-05    3.0
+        2023-02-12    NaN
+        2023-02-19    4.0
+        Freq: W-SUN, dtype: float64
+        """
+        return self._upsample("ffill", limit=limit)
+
+    @final
+    def nearest(self, limit: int | None = None):
+        """
+        Resample by using the nearest value.
+
+        When resampling data, missing values may appear (e.g., when the
+        resampling frequency is higher than the original frequency).
+        The `nearest` method will replace ``NaN`` values that appeared in
+        the resampled data with the value from the nearest member of the
+        sequence, based on the index value.
+        Missing values that existed in the original data will not be modified.
+        If `limit` is given, fill only this many values in each direction for
+        each of the original values.
+
+        Parameters
+        ----------
+        limit : int, optional
+            Limit of how many values to fill.
+
+        Returns
+        -------
+        Series or DataFrame
+            An upsampled Series or DataFrame with ``NaN`` values filled with
+            their nearest value.
+
+        See Also
+        --------
+        bfill : Backward fill the new missing values in the resampled data.
+        ffill : Forward fill ``NaN`` values.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2], index=pd.date_range("20180101", periods=2, freq="1h"))
+        >>> s
+        2018-01-01 00:00:00    1
+        2018-01-01 01:00:00    2
+        Freq: h, dtype: int64
+
+        >>> s.resample("15min").nearest()
+        2018-01-01 00:00:00    1
+        2018-01-01 00:15:00    1
+        2018-01-01 00:30:00    2
+        2018-01-01 00:45:00    2
+        2018-01-01 01:00:00    2
+        Freq: 15min, dtype: int64
+
+        Limit the number of upsampled values imputed by the nearest:
+
+        >>> s.resample("15min").nearest(limit=1)
+        2018-01-01 00:00:00    1.0
+        2018-01-01 00:15:00    1.0
+        2018-01-01 00:30:00    NaN
+        2018-01-01 00:45:00    2.0
+        2018-01-01 01:00:00    2.0
+        Freq: 15min, dtype: float64
+        """
+        return self._upsample("nearest", limit=limit)
+
+    @final
+    def bfill(self, limit: int | None = None):
+        """
+        Backward fill the new missing values in the resampled data.
+
+        In statistics, imputation is the process of replacing missing data with
+        substituted values [1]_. When resampling data, missing values may
+        appear (e.g., when the resampling frequency is higher than the original
+        frequency). The backward fill will replace NaN values that appeared in
+        the resampled data with the next value in the original sequence.
+        Missing values that existed in the original data will not be modified.
+
+        Parameters
+        ----------
+        limit : int, optional
+            Limit of how many values to fill.
+
+        Returns
+        -------
+        Series, DataFrame
+            An upsampled Series or DataFrame with backward filled NaN values.
+
+        See Also
+        --------
+        nearest : Fill NaN values with nearest neighbor starting from center.
+        ffill : Forward fill NaN values.
+        Series.fillna : Fill NaN values in the Series using the
+            specified method, which can be 'backfill'.
+        DataFrame.fillna : Fill NaN values in the DataFrame using the
+            specified method, which can be 'backfill'.
+
+        References
+        ----------
+        .. [1] https://en.wikipedia.org/wiki/Imputation_%28statistics%29
+
+        Examples
+        --------
+        Resampling a Series:
+
+        >>> s = pd.Series(
+        ...     [1, 2, 3], index=pd.date_range("20180101", periods=3, freq="h")
+        ... )
+        >>> s
+        2018-01-01 00:00:00    1
+        2018-01-01 01:00:00    2
+        2018-01-01 02:00:00    3
+        Freq: h, dtype: int64
+
+        >>> s.resample("30min").bfill()
+        2018-01-01 00:00:00    1
+        2018-01-01 00:30:00    2
+        2018-01-01 01:00:00    2
+        2018-01-01 01:30:00    3
+        2018-01-01 02:00:00    3
+        Freq: 30min, dtype: int64
+
+        >>> s.resample("15min").bfill(limit=2)
+        2018-01-01 00:00:00    1.0
+        2018-01-01 00:15:00    NaN
+        2018-01-01 00:30:00    2.0
+        2018-01-01 00:45:00    2.0
+        2018-01-01 01:00:00    2.0
+        2018-01-01 01:15:00    NaN
+        2018-01-01 01:30:00    3.0
+        2018-01-01 01:45:00    3.0
+        2018-01-01 02:00:00    3.0
+        Freq: 15min, dtype: float64
+
+        Resampling a DataFrame that has missing values:
+
+        >>> df = pd.DataFrame(
+        ...     {"a": [2, np.nan, 6], "b": [1, 3, 5]},
+        ...     index=pd.date_range("20180101", periods=3, freq="h"),
+        ... )
+        >>> df
+                               a  b
+        2018-01-01 00:00:00  2.0  1
+        2018-01-01 01:00:00  NaN  3
+        2018-01-01 02:00:00  6.0  5
+
+        >>> df.resample("30min").bfill()
+                               a  b
+        2018-01-01 00:00:00  2.0  1
+        2018-01-01 00:30:00  NaN  3
+        2018-01-01 01:00:00  NaN  3
+        2018-01-01 01:30:00  6.0  5
+        2018-01-01 02:00:00  6.0  5
+
+        >>> df.resample("15min").bfill(limit=2)
+                               a    b
+        2018-01-01 00:00:00  2.0  1.0
+        2018-01-01 00:15:00  NaN  NaN
+        2018-01-01 00:30:00  NaN  3.0
+        2018-01-01 00:45:00  NaN  3.0
+        2018-01-01 01:00:00  NaN  3.0
+        2018-01-01 01:15:00  NaN  NaN
+        2018-01-01 01:30:00  6.0  5.0
+        2018-01-01 01:45:00  6.0  5.0
+        2018-01-01 02:00:00  6.0  5.0
+        """
+        return self._upsample("bfill", limit=limit)
+
+    @final
+    def interpolate(
+        self,
+        method: InterpolateOptions = "linear",
+        *,
+        axis: Axis = 0,
+        limit: int | None = None,
+        inplace: bool = False,
+        limit_direction: Literal["forward", "backward", "both"] = "forward",
+        limit_area=None,
+        downcast=lib.no_default,
+        **kwargs,
+    ):
+        """
+        Interpolate values between target timestamps according to different methods.
+
+        The original index is first reindexed to target timestamps
+        (see :meth:`core.resample.Resampler.asfreq`),
+        then the interpolation of ``NaN`` values via :meth:`DataFrame.interpolate`
+        happens.
+
+        Parameters
+        ----------
+        method : str, default 'linear'
+            Interpolation technique to use. One of:
+
+            * 'linear': Ignore the index and treat the values as equally
+              spaced. This is the only method supported on MultiIndexes.
+            * 'time': Works on daily and higher resolution data to interpolate
+              given length of interval.
+            * 'index', 'values': use the actual numerical values of the index.
+            * 'pad': Fill in NaNs using existing values.
+            * 'nearest', 'zero', 'slinear', 'quadratic', 'cubic',
+              'barycentric', 'polynomial': Passed to
+              `scipy.interpolate.interp1d`, whereas 'spline' is passed to
+              `scipy.interpolate.UnivariateSpline`. These methods use the numerical
+              values of the index.  Both 'polynomial' and 'spline' require that
+              you also specify an `order` (int), e.g.
+              ``df.interpolate(method='polynomial', order=5)``. Note that,
+              `slinear` method in Pandas refers to the Scipy first order `spline`
+              instead of Pandas first order `spline`.
+            * 'krogh', 'piecewise_polynomial', 'spline', 'pchip', 'akima',
+              'cubicspline': Wrappers around the SciPy interpolation methods of
+              similar names. See `Notes`.
+            * 'from_derivatives': Refers to
+              `scipy.interpolate.BPoly.from_derivatives`.
+
+        axis : {{0 or 'index', 1 or 'columns', None}}, default None
+            Axis to interpolate along. For `Series` this parameter is unused
+            and defaults to 0.
+        limit : int, optional
+            Maximum number of consecutive NaNs to fill. Must be greater than
+            0.
+        inplace : bool, default False
+            Update the data in place if possible.
+        limit_direction : {{'forward', 'backward', 'both'}}, Optional
+            Consecutive NaNs will be filled in this direction.
+
+        limit_area : {{`None`, 'inside', 'outside'}}, default None
+            If limit is specified, consecutive NaNs will be filled with this
+            restriction.
+
+            * ``None``: No fill restriction.
+            * 'inside': Only fill NaNs surrounded by valid values
+              (interpolate).
+            * 'outside': Only fill NaNs outside valid values (extrapolate).
+
+        downcast : optional, 'infer' or None, defaults to None
+            Downcast dtypes if possible.
+
+            .. deprecated:: 2.1.0
+
+        **kwargs : optional
+            Keyword arguments to pass on to the interpolating function.
+
+        Returns
+        -------
+        DataFrame or Series
+            Interpolated values at the specified freq.
+
+        See Also
+        --------
+        core.resample.Resampler.asfreq: Return the values at the new freq,
+            essentially a reindex.
+        DataFrame.interpolate: Fill NaN values using an interpolation method.
+        DataFrame.bfill : Backward fill NaN values in the resampled data.
+        DataFrame.ffill : Forward fill NaN values.
+
+        Notes
+        -----
+        For high-frequent or non-equidistant time-series with timestamps
+        the reindexing followed by interpolation may lead to information loss
+        as shown in the last example.
+
+        Examples
+        --------
+
+        >>> start = "2023-03-01T07:00:00"
+        >>> timesteps = pd.date_range(start, periods=5, freq="s")
+        >>> series = pd.Series(data=[1, -1, 2, 1, 3], index=timesteps)
+        >>> series
+        2023-03-01 07:00:00    1
+        2023-03-01 07:00:01   -1
+        2023-03-01 07:00:02    2
+        2023-03-01 07:00:03    1
+        2023-03-01 07:00:04    3
+        Freq: s, dtype: int64
+
+        Downsample the dataframe to 0.5Hz by providing the period time of 2s.
+
+        >>> series.resample("2s").interpolate("linear")
+        2023-03-01 07:00:00    1
+        2023-03-01 07:00:02    2
+        2023-03-01 07:00:04    3
+        Freq: 2s, dtype: int64
+
+        Upsample the dataframe to 2Hz by providing the period time of 500ms.
+
+        >>> series.resample("500ms").interpolate("linear")
+        2023-03-01 07:00:00.000    1.0
+        2023-03-01 07:00:00.500    0.0
+        2023-03-01 07:00:01.000   -1.0
+        2023-03-01 07:00:01.500    0.5
+        2023-03-01 07:00:02.000    2.0
+        2023-03-01 07:00:02.500    1.5
+        2023-03-01 07:00:03.000    1.0
+        2023-03-01 07:00:03.500    2.0
+        2023-03-01 07:00:04.000    3.0
+        Freq: 500ms, dtype: float64
+
+        Internal reindexing with ``asfreq()`` prior to interpolation leads to
+        an interpolated timeseries on the basis of the reindexed timestamps
+        (anchors). It is assured that all available datapoints from original
+        series become anchors, so it also works for resampling-cases that lead
+        to non-aligned timestamps, as in the following example:
+
+        >>> series.resample("400ms").interpolate("linear")
+        2023-03-01 07:00:00.000    1.0
+        2023-03-01 07:00:00.400    0.2
+        2023-03-01 07:00:00.800   -0.6
+        2023-03-01 07:00:01.200   -0.4
+        2023-03-01 07:00:01.600    0.8
+        2023-03-01 07:00:02.000    2.0
+        2023-03-01 07:00:02.400    1.6
+        2023-03-01 07:00:02.800    1.2
+        2023-03-01 07:00:03.200    1.4
+        2023-03-01 07:00:03.600    2.2
+        2023-03-01 07:00:04.000    3.0
+        Freq: 400ms, dtype: float64
+
+        Note that the series correctly decreases between two anchors
+        ``07:00:00`` and ``07:00:02``.
+        """
+        assert downcast is lib.no_default  # just checking coverage
+        result = self._upsample("asfreq")
+
+        # If the original data has timestamps which are not aligned with the
+        # target timestamps, we need to add those points back to the data frame
+        # that is supposed to be interpolated. This does not work with
+        # PeriodIndex, so we skip this case. GH#21351
+        obj = self._selected_obj
+        is_period_index = isinstance(obj.index, PeriodIndex)
+
+        # Skip this step for PeriodIndex
+        if not is_period_index:
+            final_index = result.index
+            if isinstance(final_index, MultiIndex):
+                raise NotImplementedError(
+                    "Direct interpolation of MultiIndex data frames is not "
+                    "supported. If you tried to resample and interpolate on a "
+                    "grouped data frame, please use:\n"
+                    "`df.groupby(...).apply(lambda x: x.resample(...)."
+                    "interpolate(...))`"
+                    "\ninstead, as resampling and interpolation has to be "
+                    "performed for each group independently."
+                )
+
+            missing_data_points_index = obj.index.difference(final_index)
+            if len(missing_data_points_index) > 0:
+                result = concat(
+                    [result, obj.loc[missing_data_points_index]]
+                ).sort_index()
+
+        result_interpolated = result.interpolate(
+            method=method,
+            axis=axis,
+            limit=limit,
+            inplace=inplace,
+            limit_direction=limit_direction,
+            limit_area=limit_area,
+            downcast=downcast,
+            **kwargs,
+        )
+
+        # No further steps if the original data has a PeriodIndex
+        if is_period_index:
+            return result_interpolated
+
+        # Make sure that original data points which do not align with the
+        # resampled index are removed
+        result_interpolated = result_interpolated.loc[final_index]
+
+        # Make sure frequency indexes are preserved
+        result_interpolated.index = final_index
+        return result_interpolated
+
+    @final
+    def asfreq(self, fill_value=None):
+        """
+        Return the values at the new freq, essentially a reindex.
+
+        Parameters
+        ----------
+        fill_value : scalar, optional
+            Value to use for missing values, applied during upsampling (note
+            this does not fill NaNs that already were present).
+
+        Returns
+        -------
+        DataFrame or Series
+            Values at the specified freq.
+
+        See Also
+        --------
+        Series.asfreq: Convert TimeSeries to specified frequency.
+        DataFrame.asfreq: Convert TimeSeries to specified frequency.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-31", "2023-02-01", "2023-02-28"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-31    2
+        2023-02-01    3
+        2023-02-28    4
+        dtype: int64
+        >>> ser.resample("MS").asfreq()
+        2023-01-01    1
+        2023-02-01    3
+        Freq: MS, dtype: int64
+        """
+        return self._upsample("asfreq", fill_value=fill_value)
+
+    @final
+    def sum(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+    ):
+        """
+        Compute sum of group values.
+
+        This method provides a simple way to compute the sum of values within each
+        resampled group, particularly useful for aggregating time-based data into
+        daily, monthly, or yearly sums.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None``.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` non-NA values are present the result will be NA.
+
+        Returns
+        -------
+        Series or DataFrame
+            Computed sum of values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.count : Compute count of group, excluding missing
+            values.
+        DataFrame.resample : Resample time-series data.
+        Series.sum : Return the sum of the values over the requested axis.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").sum()
+        2023-01-01    3
+        2023-02-01    7
+        Freq: MS, dtype: int64
+        """
+        return self._downsample("sum", numeric_only=numeric_only, min_count=min_count)
+
+    @final
+    def prod(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+    ):
+        """
+        Compute prod of group values.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None``.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` non-NA values are present the result will be NA.
+
+        Returns
+        -------
+        Series or DataFrame
+            Computed prod of values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.sum : Compute sum of groups, excluding missing values.
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").prod()
+        2023-01-01    2
+        2023-02-01   12
+        Freq: MS, dtype: int64
+        """
+        return self._downsample("prod", numeric_only=numeric_only, min_count=min_count)
+
+    @final
+    def min(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+    ):
+        """
+        Compute min value of group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None``.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` non-NA values are present the result will be NA.
+
+        Returns
+        -------
+        Series or DataFrame
+            Compute the minimum value in the given Series or DataFrame.
+
+        See Also
+        --------
+        core.resample.Resampler.max : Compute max value of group.
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").min()
+        2023-01-01    1
+        2023-02-01    3
+        Freq: MS, dtype: int64
+        """
+        return self._downsample("min", numeric_only=numeric_only, min_count=min_count)
+
+    @final
+    def max(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+    ):
+        """
+        Compute max value of group.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None``.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer
+            than ``min_count`` non-NA values are present the result will be NA.
+
+        Returns
+        -------
+        Series or DataFrame
+            Computes the maximum value in the given Series or Dataframe.
+
+        See Also
+        --------
+        core.resample.Resampler.min : Compute min value of group.
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").max()
+        2023-01-01    2
+        2023-02-01    4
+        Freq: MS, dtype: int64
+        """
+        return self._downsample("max", numeric_only=numeric_only, min_count=min_count)
+
+    @final
+    @doc(GroupBy.first)
+    def first(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        skipna: bool = True,
+    ):
+        return self._downsample(
+            "first", numeric_only=numeric_only, min_count=min_count, skipna=skipna
+        )
+
+    @final
+    @doc(GroupBy.last)
+    def last(
+        self,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        skipna: bool = True,
+    ):
+        return self._downsample(
+            "last", numeric_only=numeric_only, min_count=min_count, skipna=skipna
+        )
+
+    @final
+    def median(self, numeric_only: bool = False):
+        """
+        Compute median of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only no longer accepts ``None`` and defaults to False.
+
+        Returns
+        -------
+        Series or DataFrame
+            Median of values within each group.
+
+        See Also
+        --------
+        Series.groupby : Apply a function groupby to a Series.
+        DataFrame.groupby : Apply a function groupby to each row or column of a
+            DataFrame.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 3, 4, 5],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").median()
+        2023-01-01    2.0
+        2023-02-01    4.0
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("median", numeric_only=numeric_only)
+
+    @final
+    def mean(
+        self,
+        numeric_only: bool = False,
+    ):
+        """
+        Compute mean of groups, excluding missing values.
+
+        Parameters
+        ----------
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        Returns
+        -------
+        DataFrame or Series
+            Mean of values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+        core.resample.Resampler.sum : Compute sum of groups, excluding missing values.
+        core.resample.Resampler.std : Compute standard deviation of groups, excluding
+            missing values.
+        core.resample.Resampler.var : Compute variance of groups, excluding missing
+            values.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    4
+        dtype: int64
+        >>> ser.resample("MS").mean()
+        2023-01-01    1.5
+        2023-02-01    3.5
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("mean", numeric_only=numeric_only)
+
+    @final
+    def std(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        """
+        Compute standard deviation of groups, excluding missing values.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Degrees of freedom.
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        Returns
+        -------
+        DataFrame or Series
+            Standard deviation of values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+        core.resample.Resampler.var : Compute variance of groups, excluding missing
+            values.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 8],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").std()
+        2023-01-01    1.000000
+        2023-02-01    2.645751
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("std", ddof=ddof, numeric_only=numeric_only)
+
+    @final
+    def var(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        """
+        Compute variance of groups, excluding missing values.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Degrees of freedom.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        Returns
+        -------
+        DataFrame or Series
+            Variance of values within each group.
+
+        See Also
+        --------
+        core.resample.Resampler.std : Compute standard deviation of groups, excluding
+            missing values.
+        core.resample.Resampler.mean : Compute mean of groups, excluding missing values.
+        core.resample.Resampler.median : Compute median of groups, excluding missing
+            values.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 8],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").var()
+        2023-01-01    1.0
+        2023-02-01    7.0
+        Freq: MS, dtype: float64
+
+        >>> ser.resample("MS").var(ddof=0)
+        2023-01-01    0.666667
+        2023-02-01    4.666667
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("var", ddof=ddof, numeric_only=numeric_only)
+
+    @final
+    def sem(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        """
+        Compute standard error of the mean of groups, excluding missing values.
+
+        For multiple groupings, the result index will be a MultiIndex.
+
+        Parameters
+        ----------
+        ddof : int, default 1
+            Degrees of freedom.
+
+        numeric_only : bool, default False
+            Include only `float`, `int` or `boolean` data.
+
+            .. versionadded:: 1.5.0
+
+            .. versionchanged:: 2.0.0
+
+                numeric_only now defaults to ``False``.
+
+        Returns
+        -------
+        Series or DataFrame
+            Standard error of the mean of values within each group.
+
+        See Also
+        --------
+        DataFrame.sem : Return unbiased standard error of the mean over requested axis.
+        Series.sem : Return unbiased standard error of the mean over requested axis.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 8],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").sem()
+        2023-01-01    0.577350
+        2023-02-01    1.527525
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("sem", ddof=ddof, numeric_only=numeric_only)
+
+    @final
+    @doc(GroupBy.ohlc)
+    def ohlc(self):
+        ax = self.ax
+        obj = self._obj_with_exclusions
+        if len(ax) == 0:
+            # GH#42902
+            obj = obj.copy()
+            obj.index = _asfreq_compat(obj.index, self.freq)
+            if obj.ndim == 1:
+                obj = obj.to_frame()
+                obj = obj.reindex(["open", "high", "low", "close"], axis=1)
+            else:
+                mi = MultiIndex.from_product(
+                    [obj.columns, ["open", "high", "low", "close"]]
+                )
+                obj = obj.reindex(mi, axis=1)
+            return obj
+
+        return self._downsample("ohlc")
+
+    @final
+    def nunique(self):
+        """
+        Return number of unique elements in the group.
+
+        Returns
+        -------
+        Series
+            Number of unique values within each group.
+
+        See Also
+        --------
+        core.groupby.SeriesGroupBy.nunique : Method nunique for SeriesGroupBy.
+
+        Examples
+        --------
+        >>> ser = pd.Series(
+        ...     [1, 2, 3, 3],
+        ...     index=pd.DatetimeIndex(
+        ...         ["2023-01-01", "2023-01-15", "2023-02-01", "2023-02-15"]
+        ...     ),
+        ... )
+        >>> ser
+        2023-01-01    1
+        2023-01-15    2
+        2023-02-01    3
+        2023-02-15    3
+        dtype: int64
+        >>> ser.resample("MS").nunique()
+        2023-01-01    2
+        2023-02-01    1
+        Freq: MS, dtype: int64
+        """
+        return self._downsample("nunique")
+
+    @final
+    @doc(GroupBy.size)
+    def size(self):
+        result = self._downsample("size")
+
+        # If the result is a non-empty DataFrame we stack to get a Series
+        # GH 46826
+        if isinstance(result, ABCDataFrame) and not result.empty:
+            result = result.stack()
+
+        if not len(self.ax):
+            from pandas import Series
+
+            if self._selected_obj.ndim == 1:
+                name = self._selected_obj.name
+            else:
+                name = None
+            result = Series([], index=result.index, dtype="int64", name=name)
+        return result
+
+    @final
+    @doc(GroupBy.count)
+    def count(self):
+        result = self._downsample("count")
+        if not len(self.ax):
+            if self._selected_obj.ndim == 1:
+                result = type(self._selected_obj)(
+                    [], index=result.index, dtype="int64", name=self._selected_obj.name
+                )
+            else:
+                from pandas import DataFrame
+
+                result = DataFrame(
+                    [], index=result.index, columns=result.columns, dtype="int64"
+                )
+
+        return result
+
+    @final
+    def quantile(self, q: float | list[float] | AnyArrayLike = 0.5, **kwargs):
+        """
+        Return value at the given quantile.
+
+        Parameters
+        ----------
+        q : float or array-like, default 0.5 (50% quantile)
+
+        Returns
+        -------
+        DataFrame or Series
+            Quantile of values within each group.
+
+        See Also
+        --------
+        Series.quantile
+            Return a series, where the index is q and the values are the quantiles.
+        DataFrame.quantile
+            Return a DataFrame, where the columns are the columns of self,
+            and the values are the quantiles.
+        DataFrameGroupBy.quantile
+            Return a DataFrame, where the columns are groupby columns,
+            and the values are its quantiles.
+
+        Examples
+        --------
+
+        >>> ser = pd.Series(
+        ...     [1, 3, 2, 4, 3, 8],
+        ...     index=pd.DatetimeIndex(
+        ...         [
+        ...             "2023-01-01",
+        ...             "2023-01-10",
+        ...             "2023-01-15",
+        ...             "2023-02-01",
+        ...             "2023-02-10",
+        ...             "2023-02-15",
+        ...         ]
+        ...     ),
+        ... )
+        >>> ser.resample("MS").quantile()
+        2023-01-01    2.0
+        2023-02-01    4.0
+        Freq: MS, dtype: float64
+
+        >>> ser.resample("MS").quantile(0.25)
+        2023-01-01    1.5
+        2023-02-01    3.5
+        Freq: MS, dtype: float64
+        """
+        return self._downsample("quantile", q=q, **kwargs)
+
+
+class _GroupByMixin(PandasObject, SelectionMixin):
+    """
+    Provide the groupby facilities.
+    """
+
+    _attributes: list[str]  # in practice the same as Resampler._attributes
+    _selection: IndexLabel | None = None
+    _groupby: GroupBy
+    _timegrouper: TimeGrouper
+
+    def __init__(
+        self,
+        *,
+        parent: Resampler,
+        groupby: GroupBy,
+        key=None,
+        selection: IndexLabel | None = None,
+    ) -> None:
+        # reached via ._gotitem and _get_resampler_for_grouping
+
+        assert isinstance(groupby, GroupBy), type(groupby)
+
+        # parent is always a Resampler, sometimes a _GroupByMixin
+        assert isinstance(parent, Resampler), type(parent)
+
+        # initialize our GroupByMixin object with
+        # the resampler attributes
+        for attr in self._attributes:
+            setattr(self, attr, getattr(parent, attr))
+        self._selection = selection
+
+        self.binner = parent.binner
+        self.key = key
+
+        self._groupby = groupby
+        self._timegrouper = copy.copy(parent._timegrouper)
+
+        self.ax = parent.ax
+        self.obj = parent.obj
+
+    @no_type_check
+    def _apply(self, f, *args, **kwargs):
+        """
+        Dispatch to _upsample; we are stripping all of the _upsample kwargs and
+        performing the original function call on the grouped object.
+        """
+
+        def func(x):
+            x = self._resampler_cls(x, timegrouper=self._timegrouper, gpr_index=self.ax)
+
+            if isinstance(f, str):
+                return getattr(x, f)(**kwargs)
+
+            return x.apply(f, *args, **kwargs)
+
+        result = self._groupby.apply(func)
+        return self._wrap_result(result)
+
+    _upsample = _apply
+    _downsample = _apply
+    _groupby_and_aggregate = _apply
+
+    @final
+    def _gotitem(self, key, ndim, subset=None):
+        """
+        Sub-classes to define. Return a sliced object.
+
+        Parameters
+        ----------
+        key : string / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        # create a new object to prevent aliasing
+        if subset is None:
+            subset = self.obj
+            if key is not None:
+                subset = subset[key]
+            else:
+                # reached via Apply.agg_dict_like with selection=None, ndim=1
+                assert subset.ndim == 1
+
+        # Try to select from a DataFrame, falling back to a Series
+        try:
+            if isinstance(key, list) and self.key not in key and self.key is not None:
+                key.append(self.key)
+            groupby = self._groupby[key]
+        except IndexError:
+            groupby = self._groupby
+
+        selection = self._infer_selection(key, subset)
+
+        new_rs = type(self)(
+            groupby=groupby,
+            parent=cast(Resampler, self),
+            selection=selection,
+        )
+        return new_rs
+
+
+class DatetimeIndexResampler(Resampler):
+    ax: DatetimeIndex
+
+    @property
+    def _resampler_for_grouping(self) -> type[DatetimeIndexResamplerGroupby]:
+        return DatetimeIndexResamplerGroupby
+
+    def _get_binner_for_time(self):
+        # this is how we are actually creating the bins
+        if isinstance(self.ax, PeriodIndex):
+            return self._timegrouper._get_time_period_bins(self.ax)
+        return self._timegrouper._get_time_bins(self.ax)
+
+    def _downsample(self, how, **kwargs):
+        """
+        Downsample the cython defined function.
+
+        Parameters
+        ----------
+        how : string / cython mapped function
+        **kwargs : kw args passed to how function
+        """
+        ax = self.ax
+
+        # Excludes `on` column when provided
+        obj = self._obj_with_exclusions
+
+        if not len(ax):
+            # reset to the new freq
+            obj = obj.copy()
+            obj.index = obj.index._with_freq(self.freq)
+            assert obj.index.freq == self.freq, (obj.index.freq, self.freq)
+            return obj
+
+        # do we have a regular frequency
+
+        # error: Item "None" of "Optional[Any]" has no attribute "binlabels"
+        if (
+            (ax.freq is not None or ax.inferred_freq is not None)
+            and len(self._grouper.binlabels) > len(ax)
+            and how is None
+        ):
+            # let's do an asfreq
+            return self.asfreq()
+
+        # we are downsampling
+        # we want to call the actual grouper method here
+        result = obj.groupby(self._grouper).aggregate(how, **kwargs)
+        return self._wrap_result(result)
+
+    def _adjust_binner_for_upsample(self, binner):
+        """
+        Adjust our binner when upsampling.
+
+        The range of a new index should not be outside specified range
+        """
+        if self.closed == "right":
+            binner = binner[1:]
+        else:
+            binner = binner[:-1]
+        return binner
+
+    def _upsample(self, method, limit: int | None = None, fill_value=None):
+        """
+        Parameters
+        ----------
+        method : string {'backfill', 'bfill', 'pad',
+            'ffill', 'asfreq'} method for upsampling
+        limit : int, default None
+            Maximum size gap to fill when reindexing
+        fill_value : scalar, default None
+            Value to use for missing values
+        """
+        if self._from_selection:
+            raise ValueError(
+                "Upsampling from level= or on= selection "
+                "is not supported, use .set_index(...) "
+                "to explicitly set index to datetime-like"
+            )
+
+        ax = self.ax
+        obj = self._selected_obj
+        binner = self.binner
+        res_index = self._adjust_binner_for_upsample(binner)
+
+        # if we have the same frequency as our axis, then we are equal sampling
+        if (
+            limit is None
+            and to_offset(ax.inferred_freq) == self.freq
+            and len(obj) == len(res_index)
+        ):
+            result = obj.copy()
+            result.index = res_index
+        else:
+            if method == "asfreq":
+                method = None
+            result = obj.reindex(
+                res_index, method=method, limit=limit, fill_value=fill_value
+            )
+
+        return self._wrap_result(result)
+
+    def _wrap_result(self, result):
+        result = super()._wrap_result(result)
+
+        # we may have a different kind that we were asked originally
+        # convert if needed
+        if isinstance(self.ax, PeriodIndex) and not isinstance(
+            result.index, PeriodIndex
+        ):
+            if isinstance(result.index, MultiIndex):
+                # GH 24103 - e.g. groupby resample
+                if not isinstance(result.index.levels[-1], PeriodIndex):
+                    new_level = result.index.levels[-1].to_period(self.freq)
+                    result.index = result.index.set_levels(new_level, level=-1)
+            else:
+                result.index = result.index.to_period(self.freq)
+        return result
+
+
+# error: Definition of "ax" in base class "_GroupByMixin" is incompatible
+# with definition in base class "DatetimeIndexResampler"
+class DatetimeIndexResamplerGroupby(  # type: ignore[misc]
+    _GroupByMixin, DatetimeIndexResampler
+):
+    """
+    Provides a resample of a groupby implementation
+    """
+
+    @property
+    def _resampler_cls(self):
+        return DatetimeIndexResampler
+
+
+class PeriodIndexResampler(DatetimeIndexResampler):
+    # error: Incompatible types in assignment (expression has type "PeriodIndex", base
+    # class "DatetimeIndexResampler" defined the type as "DatetimeIndex")
+    ax: PeriodIndex  # type: ignore[assignment]
+
+    @property
+    def _resampler_for_grouping(self):
+        warnings.warn(
+            "Resampling a groupby with a PeriodIndex is deprecated. "
+            "Cast to DatetimeIndex before resampling instead.",
+            FutureWarning,
+            stacklevel=find_stack_level(),
+        )
+        return PeriodIndexResamplerGroupby
+
+    def _get_binner_for_time(self):
+        if isinstance(self.ax, DatetimeIndex):
+            return super()._get_binner_for_time()
+        return self._timegrouper._get_period_bins(self.ax)
+
+    def _convert_obj(self, obj: NDFrameT) -> NDFrameT:
+        obj = super()._convert_obj(obj)
+
+        if self._from_selection:
+            # see GH 14008, GH 12871
+            msg = (
+                "Resampling from level= or on= selection "
+                "with a PeriodIndex is not currently supported, "
+                "use .set_index(...) to explicitly set index"
+            )
+            raise NotImplementedError(msg)
+
+        # convert to timestamp
+        if isinstance(obj, DatetimeIndex):
+            obj = obj.to_timestamp(how=self.convention)
+
+        return obj
+
+    def _downsample(self, how, **kwargs):
+        """
+        Downsample the cython defined function.
+
+        Parameters
+        ----------
+        how : string / cython mapped function
+        **kwargs : kw args passed to how function
+        """
+        # we may need to actually resample as if we are timestamps
+        if isinstance(self.ax, DatetimeIndex):
+            return super()._downsample(how, **kwargs)
+
+        ax = self.ax
+
+        if is_subperiod(ax.freq, self.freq):
+            # Downsampling
+            return self._groupby_and_aggregate(how, **kwargs)
+        elif is_superperiod(ax.freq, self.freq):
+            if how == "ohlc":
+                # GH #13083
+                # upsampling to subperiods is handled as an asfreq, which works
+                # for pure aggregating/reducing methods
+                # OHLC reduces along the time dimension, but creates multiple
+                # values for each period -> handle by _groupby_and_aggregate()
+                return self._groupby_and_aggregate(how)
+            return self.asfreq()
+        elif ax.freq == self.freq:
+            return self.asfreq()
+
+        raise IncompatibleFrequency(
+            f"Frequency {ax.freq} cannot be resampled to {self.freq}, "
+            "as they are not sub or super periods"
+        )
+
+    def _upsample(self, method, limit: int | None = None, fill_value=None):
+        """
+        Parameters
+        ----------
+        method : {'backfill', 'bfill', 'pad', 'ffill'}
+            Method for upsampling.
+        limit : int, default None
+            Maximum size gap to fill when reindexing.
+        fill_value : scalar, default None
+            Value to use for missing values.
+        """
+        # we may need to actually resample as if we are timestamps
+        if isinstance(self.ax, DatetimeIndex):
+            return super()._upsample(method, limit=limit, fill_value=fill_value)
+
+        ax = self.ax
+        obj = self.obj
+        new_index = self.binner
+
+        # Start vs. end of period
+        memb = ax.asfreq(self.freq, how=self.convention)
+
+        # Get the fill indexer
+        if method == "asfreq":
+            method = None
+        indexer = memb.get_indexer(new_index, method=method, limit=limit)
+        new_obj = _take_new_index(
+            obj,
+            indexer,
+            new_index,
+        )
+        return self._wrap_result(new_obj)
+
+
+# error: Definition of "ax" in base class "_GroupByMixin" is incompatible with
+# definition in base class "PeriodIndexResampler"
+class PeriodIndexResamplerGroupby(  # type: ignore[misc]
+    _GroupByMixin, PeriodIndexResampler
+):
+    """
+    Provides a resample of a groupby implementation.
+    """
+
+    @property
+    def _resampler_cls(self):
+        return PeriodIndexResampler
+
+
+class TimedeltaIndexResampler(DatetimeIndexResampler):
+    # error: Incompatible types in assignment (expression has type "TimedeltaIndex",
+    # base class "DatetimeIndexResampler" defined the type as "DatetimeIndex")
+    ax: TimedeltaIndex  # type: ignore[assignment]
+
+    @property
+    def _resampler_for_grouping(self):
+        return TimedeltaIndexResamplerGroupby
+
+    def _get_binner_for_time(self):
+        return self._timegrouper._get_time_delta_bins(self.ax)
+
+    def _adjust_binner_for_upsample(self, binner):
+        """
+        Adjust our binner when upsampling.
+
+        The range of a new index is allowed to be greater than original range
+        so we don't need to change the length of a binner, GH 13022
+        """
+        return binner
+
+
+# error: Definition of "ax" in base class "_GroupByMixin" is incompatible with
+# definition in base class "DatetimeIndexResampler"
+class TimedeltaIndexResamplerGroupby(  # type: ignore[misc]
+    _GroupByMixin, TimedeltaIndexResampler
+):
+    """
+    Provides a resample of a groupby implementation.
+    """
+
+    @property
+    def _resampler_cls(self):
+        return TimedeltaIndexResampler
+
+
+def get_resampler(obj: Series | DataFrame, **kwds) -> Resampler:
+    """
+    Create a TimeGrouper and return our resampler.
+    """
+    tg = TimeGrouper(obj, **kwds)  # type: ignore[arg-type]
+    return tg._get_resampler(obj)
+
+
+get_resampler.__doc__ = Resampler.__doc__
+
+
+def get_resampler_for_grouping(
+    groupby: GroupBy,
+    rule,
+    how=None,
+    fill_method=None,
+    limit: int | None = None,
+    on=None,
+    **kwargs,
+) -> Resampler:
+    """
+    Return our appropriate resampler when grouping as well.
+    """
+    # .resample uses 'on' similar to how .groupby uses 'key'
+    tg = TimeGrouper(freq=rule, key=on, **kwargs)
+    resampler = tg._get_resampler(groupby.obj)
+    return resampler._get_resampler_for_grouping(groupby=groupby, key=tg.key)
+
+
+class TimeGrouper(Grouper):
+    """
+    Custom groupby class for time-interval grouping.
+
+    Parameters
+    ----------
+    freq : pandas date offset or offset alias for identifying bin edges
+    closed : closed end of interval; 'left' or 'right'
+    label : interval boundary to use for labeling; 'left' or 'right'
+    convention : {'start', 'end', 'e', 's'}
+        If axis is PeriodIndex
+    """
+
+    _attributes = Grouper._attributes + (
+        "closed",
+        "label",
+        "how",
+        "convention",
+        "origin",
+        "offset",
+    )
+
+    origin: TimeGrouperOrigin
+
+    def __init__(
+        self,
+        obj: Grouper | None = None,
+        freq: Frequency = "Min",
+        key: str | None = None,
+        closed: Literal["left", "right"] | None = None,
+        label: Literal["left", "right"] | None = None,
+        how: str = "mean",
+        fill_method=None,
+        limit: int | None = None,
+        convention: Literal["start", "end", "e", "s"] | None = None,
+        origin: Literal["epoch", "start", "start_day", "end", "end_day"]
+        | TimestampConvertibleTypes = "start_day",
+        offset: TimedeltaConvertibleTypes | None = None,
+        group_keys: bool = False,
+        **kwargs,
+    ) -> None:
+        # Check for correctness of the keyword arguments which would
+        # otherwise silently use the default if misspelled
+        if label not in {None, "left", "right"}:
+            raise ValueError(f"Unsupported value {label} for `label`")
+        if closed not in {None, "left", "right"}:
+            raise ValueError(f"Unsupported value {closed} for `closed`")
+        if convention not in {None, "start", "end", "e", "s"}:
+            raise ValueError(f"Unsupported value {convention} for `convention`")
+
+        if (
+            (key is None and obj is not None and isinstance(obj.index, PeriodIndex))  # type: ignore[attr-defined]
+            or (
+                key is not None
+                and obj is not None
+                and getattr(obj[key], "dtype", None) == "period"  # type: ignore[index]
+            )
+        ):
+            freq = to_offset(freq, is_period=True)
+        else:
+            freq = to_offset(freq)
+
+        end_types = {"ME", "YE", "QE", "BME", "BYE", "BQE", "W"}
+        rule = freq.rule_code
+        if rule in end_types or ("-" in rule and rule[: rule.find("-")] in end_types):
+            if closed is None:
+                closed = "right"
+            if label is None:
+                label = "right"
+        else:
+            # The backward resample sets ``closed`` to ``'right'`` by default
+            # since the last value should be considered as the edge point for
+            # the last bin. When origin in "end" or "end_day", the value for a
+            # specific ``Timestamp`` index stands for the resample result from
+            # the current ``Timestamp`` minus ``freq`` to the current
+            # ``Timestamp`` with a right close.
+            if origin in ["end", "end_day"]:
+                if closed is None:
+                    closed = "right"
+                if label is None:
+                    label = "right"
+            else:
+                if closed is None:
+                    closed = "left"
+                if label is None:
+                    label = "left"
+
+        self.closed = closed
+        self.label = label
+        self.convention = convention if convention is not None else "e"
+        self.how = how
+        self.fill_method = fill_method
+        self.limit = limit
+        self.group_keys = group_keys
+        self._arrow_dtype: ArrowDtype | None = None
+
+        if origin in ("epoch", "start", "start_day", "end", "end_day"):
+            # error: Incompatible types in assignment (expression has type "Union[Union[
+            # Timestamp, datetime, datetime64, signedinteger[_64Bit], float, str],
+            # Literal['epoch', 'start', 'start_day', 'end', 'end_day']]", variable has
+            # type "Union[Timestamp, Literal['epoch', 'start', 'start_day', 'end',
+            # 'end_day']]")
+            self.origin = origin  # type: ignore[assignment]
+        else:
+            try:
+                self.origin = Timestamp(origin)
+            except (ValueError, TypeError) as err:
+                raise ValueError(
+                    "'origin' should be equal to 'epoch', 'start', 'start_day', "
+                    "'end', 'end_day' or "
+                    f"should be a Timestamp convertible type. Got '{origin}' instead."
+                ) from err
+
+        try:
+            self.offset = Timedelta(offset) if offset is not None else None
+        except (ValueError, TypeError) as err:
+            raise ValueError(
+                "'offset' should be a Timedelta convertible type. "
+                f"Got '{offset}' instead."
+            ) from err
+
+        # always sort time groupers
+        kwargs["sort"] = True
+
+        super().__init__(freq=freq, key=key, **kwargs)
+
+    def _get_resampler(self, obj: NDFrame) -> Resampler:
+        """
+        Return my resampler or raise if we have an invalid axis.
+
+        Parameters
+        ----------
+        obj : Series or DataFrame
+
+        Returns
+        -------
+        Resampler
+
+        Raises
+        ------
+        TypeError if incompatible axis
+
+        """
+        _, ax, _ = self._set_grouper(obj, gpr_index=None)
+        if isinstance(ax, DatetimeIndex):
+            return DatetimeIndexResampler(
+                obj,
+                timegrouper=self,
+                group_keys=self.group_keys,
+                gpr_index=ax,
+            )
+        elif isinstance(ax, PeriodIndex):
+            if isinstance(ax, PeriodIndex):
+                # GH#53481
+                warnings.warn(
+                    "Resampling with a PeriodIndex is deprecated. "
+                    "Cast index to DatetimeIndex before resampling instead.",
+                    FutureWarning,
+                    stacklevel=find_stack_level(),
+                )
+            return PeriodIndexResampler(
+                obj,
+                timegrouper=self,
+                group_keys=self.group_keys,
+                gpr_index=ax,
+            )
+        elif isinstance(ax, TimedeltaIndex):
+            return TimedeltaIndexResampler(
+                obj,
+                timegrouper=self,
+                group_keys=self.group_keys,
+                gpr_index=ax,
+            )
+
+        raise TypeError(
+            "Only valid with DatetimeIndex, "
+            "TimedeltaIndex or PeriodIndex, "
+            f"but got an instance of '{type(ax).__name__}'"
+        )
+
+    def _get_grouper(
+        self, obj: NDFrameT, validate: bool = True
+    ) -> tuple[BinGrouper, NDFrameT]:
+        # create the resampler and return our binner
+        r = self._get_resampler(obj)
+        return r._grouper, cast(NDFrameT, r.obj)
+
+    def _get_time_bins(self, ax: DatetimeIndex):
+        if not isinstance(ax, DatetimeIndex):
+            raise TypeError(
+                "axis must be a DatetimeIndex, but got "
+                f"an instance of {type(ax).__name__}"
+            )
+
+        if len(ax) == 0:
+            binner = labels = DatetimeIndex(
+                data=[], freq=self.freq, name=ax.name, dtype=ax.dtype
+            )
+            return binner, [], labels
+
+        first, last = _get_timestamp_range_edges(
+            ax.min(),
+            ax.max(),
+            self.freq,
+            unit=ax.unit,
+            closed=self.closed,
+            origin=self.origin,
+            offset=self.offset,
+        )
+        # GH #12037
+        # use first/last directly instead of call replace() on them
+        # because replace() will swallow the nanosecond part
+        # thus last bin maybe slightly before the end if the end contains
+        # nanosecond part and lead to `Values falls after last bin` error
+        # GH 25758: If DST lands at midnight (e.g. 'America/Havana'), user feedback
+        # has noted that ambiguous=True provides the most sensible result
+        binner = labels = date_range(
+            freq=self.freq,
+            start=first,
+            end=last,
+            tz=ax.tz,
+            name=ax.name,
+            ambiguous=True,
+            nonexistent="shift_forward",
+            unit=ax.unit,
+        )
+
+        ax_values = ax.asi8
+        binner, bin_edges = self._adjust_bin_edges(binner, ax_values)
+
+        # general version, knowing nothing about relative frequencies
+        bins = lib.generate_bins_dt64(
+            ax_values, bin_edges, self.closed, hasnans=ax.hasnans
+        )
+
+        if self.closed == "right":
+            labels = binner
+            if self.label == "right":
+                labels = labels[1:]
+        elif self.label == "right":
+            labels = labels[1:]
+
+        if ax.hasnans:
+            binner = binner.insert(0, NaT)
+            labels = labels.insert(0, NaT)
+
+        # if we end up with more labels than bins
+        # adjust the labels
+        # GH4076
+        if len(bins) < len(labels):
+            labels = labels[: len(bins)]
+
+        return binner, bins, labels
+
+    def _adjust_bin_edges(
+        self, binner: DatetimeIndex, ax_values: npt.NDArray[np.int64]
+    ) -> tuple[DatetimeIndex, npt.NDArray[np.int64]]:
+        # Some hacks for > daily data, see #1471, #1458, #1483
+
+        if self.freq.name in ("BME", "ME", "W") or self.freq.name.split("-")[0] in (
+            "BQE",
+            "BYE",
+            "QE",
+            "YE",
+            "W",
+        ):
+            # If the right end-point is on the last day of the month, roll forwards
+            # until the last moment of that day. Note that we only do this for offsets
+            # which correspond to the end of a super-daily period - "month start", for
+            # example, is excluded.
+            if self.closed == "right":
+                # GH 21459, GH 9119: Adjust the bins relative to the wall time
+                edges_dti = binner.tz_localize(None)
+                edges_dti = (
+                    edges_dti
+                    + Timedelta(days=1, unit=edges_dti.unit).as_unit(edges_dti.unit)
+                    - Timedelta(1, unit=edges_dti.unit).as_unit(edges_dti.unit)
+                )
+                bin_edges = edges_dti.tz_localize(binner.tz).asi8
+            else:
+                bin_edges = binner.asi8
+
+            # intraday values on last day
+            if bin_edges[-2] > ax_values.max():
+                bin_edges = bin_edges[:-1]
+                binner = binner[:-1]
+        else:
+            bin_edges = binner.asi8
+        return binner, bin_edges
+
+    def _get_time_delta_bins(self, ax: TimedeltaIndex):
+        if not isinstance(ax, TimedeltaIndex):
+            raise TypeError(
+                "axis must be a TimedeltaIndex, but got "
+                f"an instance of {type(ax).__name__}"
+            )
+
+        if not isinstance(self.freq, Tick):
+            # GH#51896
+            raise ValueError(
+                "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+                f"e.g. '24h' or '3D', not {self.freq}"
+            )
+
+        if not len(ax):
+            binner = labels = TimedeltaIndex(data=[], freq=self.freq, name=ax.name)
+            return binner, [], labels
+
+        start, end = ax.min(), ax.max()
+
+        if self.closed == "right":
+            end += self.freq
+
+        labels = binner = timedelta_range(
+            start=start, end=end, freq=self.freq, name=ax.name
+        )
+
+        end_stamps = labels
+        if self.closed == "left":
+            end_stamps += self.freq
+
+        bins = ax.searchsorted(end_stamps, side=self.closed)
+
+        if self.offset:
+            # GH 10530 & 31809
+            labels += self.offset
+
+        return binner, bins, labels
+
+    def _get_time_period_bins(self, ax: DatetimeIndex):
+        if not isinstance(ax, DatetimeIndex):
+            raise TypeError(
+                "axis must be a DatetimeIndex, but got "
+                f"an instance of {type(ax).__name__}"
+            )
+
+        freq = self.freq
+
+        if len(ax) == 0:
+            binner = labels = PeriodIndex(
+                data=[], freq=freq, name=ax.name, dtype=ax.dtype
+            )
+            return binner, [], labels
+
+        labels = binner = period_range(start=ax[0], end=ax[-1], freq=freq, name=ax.name)
+
+        end_stamps = (labels + freq).asfreq(freq, "s").to_timestamp()
+        if ax.tz:
+            end_stamps = end_stamps.tz_localize(ax.tz)
+        bins = ax.searchsorted(end_stamps, side="left")
+
+        return binner, bins, labels
+
+    def _get_period_bins(self, ax: PeriodIndex):
+        if not isinstance(ax, PeriodIndex):
+            raise TypeError(
+                "axis must be a PeriodIndex, but got "
+                f"an instance of {type(ax).__name__}"
+            )
+
+        memb = ax.asfreq(self.freq, how=self.convention)
+
+        # NaT handling as in pandas._lib.lib.generate_bins_dt64()
+        nat_count = 0
+        if memb.hasnans:
+            # error: Incompatible types in assignment (expression has type
+            # "bool_", variable has type "int")  [assignment]
+            nat_count = np.sum(memb._isnan)  # type: ignore[assignment]
+            memb = memb[~memb._isnan]
+
+        if not len(memb):
+            # index contains no valid (non-NaT) values
+            bins = np.array([], dtype=np.int64)
+            binner = labels = PeriodIndex(data=[], freq=self.freq, name=ax.name)
+            if len(ax) > 0:
+                # index is all NaT
+                binner, bins, labels = _insert_nat_bin(binner, bins, labels, len(ax))
+            return binner, bins, labels
+
+        freq_mult = self.freq.n
+
+        start = ax.min().asfreq(self.freq, how=self.convention)
+        end = ax.max().asfreq(self.freq, how="end")
+        bin_shift = 0
+
+        if isinstance(self.freq, Tick):
+            # GH 23882 & 31809: get adjusted bin edge labels with 'origin'
+            # and 'origin' support. This call only makes sense if the freq is a
+            # Tick since offset and origin are only used in those cases.
+            # Not doing this check could create an extra empty bin.
+            p_start, end = _get_period_range_edges(
+                start,
+                end,
+                self.freq,
+                closed=self.closed,
+                origin=self.origin,
+                offset=self.offset,
+            )
+
+            # Get offset for bin edge (not label edge) adjustment
+            start_offset = Period(start, self.freq) - Period(p_start, self.freq)
+            # error: Item "Period" of "Union[Period, Any]" has no attribute "n"
+            bin_shift = start_offset.n % freq_mult  # type: ignore[union-attr]
+            start = p_start
+
+        labels = binner = period_range(
+            start=start, end=end, freq=self.freq, name=ax.name
+        )
+
+        i8 = memb.asi8
+
+        # when upsampling to subperiods, we need to generate enough bins
+        expected_bins_count = len(binner) * freq_mult
+        i8_extend = expected_bins_count - (i8[-1] - i8[0])
+        rng = np.arange(i8[0], i8[-1] + i8_extend, freq_mult)
+        rng += freq_mult
+        # adjust bin edge indexes to account for base
+        rng -= bin_shift
+
+        # Wrap in PeriodArray for PeriodArray.searchsorted
+        prng = type(memb._data)(rng, dtype=memb.dtype)
+        bins = memb.searchsorted(prng, side="left")
+
+        if nat_count > 0:
+            binner, bins, labels = _insert_nat_bin(binner, bins, labels, nat_count)
+
+        return binner, bins, labels
+
+    def _set_grouper(
+        self, obj: NDFrameT, sort: bool = False, *, gpr_index: Index | None = None
+    ) -> tuple[NDFrameT, Index, npt.NDArray[np.intp] | None]:
+        obj, ax, indexer = super()._set_grouper(obj, sort, gpr_index=gpr_index)
+        if isinstance(ax.dtype, ArrowDtype) and ax.dtype.kind in "Mm":
+            self._arrow_dtype = ax.dtype
+            ax = Index(
+                cast(ArrowExtensionArray, ax.array)._maybe_convert_datelike_array()
+            )
+        return obj, ax, indexer
+
+
+@overload
+def _take_new_index(
+    obj: DataFrame, indexer: npt.NDArray[np.intp], new_index: Index
+) -> DataFrame: ...
+
+
+@overload
+def _take_new_index(
+    obj: Series, indexer: npt.NDArray[np.intp], new_index: Index
+) -> Series: ...
+
+
+def _take_new_index(
+    obj: DataFrame | Series,
+    indexer: npt.NDArray[np.intp],
+    new_index: Index,
+) -> DataFrame | Series:
+    if isinstance(obj, ABCSeries):
+        new_values = algos.take_nd(obj._values, indexer)
+        return obj._constructor(new_values, index=new_index, name=obj.name)
+    elif isinstance(obj, ABCDataFrame):
+        new_mgr = obj._mgr.reindex_indexer(new_axis=new_index, indexer=indexer, axis=1)
+        return obj._constructor_from_mgr(new_mgr, axes=new_mgr.axes)
+    else:
+        raise ValueError("'obj' should be either a Series or a DataFrame")
+
+
+def _get_timestamp_range_edges(
+    first: Timestamp,
+    last: Timestamp,
+    freq: BaseOffset,
+    unit: str,
+    closed: Literal["right", "left"] = "left",
+    origin: TimeGrouperOrigin = "start_day",
+    offset: Timedelta | None = None,
+) -> tuple[Timestamp, Timestamp]:
+    """
+    Adjust the `first` Timestamp to the preceding Timestamp that resides on
+    the provided offset. Adjust the `last` Timestamp to the following
+    Timestamp that resides on the provided offset. Input Timestamps that
+    already reside on the offset will be adjusted depending on the type of
+    offset and the `closed` parameter.
+
+    Parameters
+    ----------
+    first : pd.Timestamp
+        The beginning Timestamp of the range to be adjusted.
+    last : pd.Timestamp
+        The ending Timestamp of the range to be adjusted.
+    freq : pd.DateOffset
+        The dateoffset to which the Timestamps will be adjusted.
+    closed : {'right', 'left'}, default "left"
+        Which side of bin interval is closed.
+    origin : {'epoch', 'start', 'start_day'} or Timestamp, default 'start_day'
+        The timestamp on which to adjust the grouping. The timezone of origin must
+        match the timezone of the index.
+        If a timestamp is not used, these values are also supported:
+
+        - 'epoch': `origin` is 1970-01-01
+        - 'start': `origin` is the first value of the timeseries
+        - 'start_day': `origin` is the first day at midnight of the timeseries
+    offset : pd.Timedelta, default is None
+        An offset timedelta added to the origin.
+
+    Returns
+    -------
+    A tuple of length 2, containing the adjusted pd.Timestamp objects.
+    """
+    if isinstance(freq, Tick):
+        index_tz = first.tz
+        if isinstance(origin, Timestamp) and (origin.tz is None) != (index_tz is None):
+            raise ValueError("The origin must have the same timezone as the index.")
+        if origin == "epoch":
+            # set the epoch based on the timezone to have similar bins results when
+            # resampling on the same kind of indexes on different timezones
+            origin = Timestamp("1970-01-01", tz=index_tz)
+
+        if isinstance(freq, Day):
+            # _adjust_dates_anchored assumes 'D' means 24h, but first/last
+            # might contain a DST transition (23h, 24h, or 25h).
+            # So "pretend" the dates are naive when adjusting the endpoints
+            first = first.tz_localize(None)
+            last = last.tz_localize(None)
+            if isinstance(origin, Timestamp):
+                origin = origin.tz_localize(None)
+
+        first, last = _adjust_dates_anchored(
+            first, last, freq, closed=closed, origin=origin, offset=offset, unit=unit
+        )
+        if isinstance(freq, Day):
+            first = first.tz_localize(index_tz)
+            last = last.tz_localize(index_tz, nonexistent="shift_forward")
+    else:
+        first = first.normalize()
+        last = last.normalize()
+
+        if closed == "left":
+            first = Timestamp(freq.rollback(first))
+        else:
+            first = Timestamp(first - freq)
+
+        last = Timestamp(last + freq)
+
+    return first, last
+
+
+def _get_period_range_edges(
+    first: Period,
+    last: Period,
+    freq: BaseOffset,
+    closed: Literal["right", "left"] = "left",
+    origin: TimeGrouperOrigin = "start_day",
+    offset: Timedelta | None = None,
+) -> tuple[Period, Period]:
+    """
+    Adjust the provided `first` and `last` Periods to the respective Period of
+    the given offset that encompasses them.
+
+    Parameters
+    ----------
+    first : pd.Period
+        The beginning Period of the range to be adjusted.
+    last : pd.Period
+        The ending Period of the range to be adjusted.
+    freq : pd.DateOffset
+        The freq to which the Periods will be adjusted.
+    closed : {'right', 'left'}, default "left"
+        Which side of bin interval is closed.
+    origin : {'epoch', 'start', 'start_day'}, Timestamp, default 'start_day'
+        The timestamp on which to adjust the grouping. The timezone of origin must
+        match the timezone of the index.
+
+        If a timestamp is not used, these values are also supported:
+
+        - 'epoch': `origin` is 1970-01-01
+        - 'start': `origin` is the first value of the timeseries
+        - 'start_day': `origin` is the first day at midnight of the timeseries
+    offset : pd.Timedelta, default is None
+        An offset timedelta added to the origin.
+
+    Returns
+    -------
+    A tuple of length 2, containing the adjusted pd.Period objects.
+    """
+    if not all(isinstance(obj, Period) for obj in [first, last]):
+        raise TypeError("'first' and 'last' must be instances of type Period")
+
+    # GH 23882
+    first_ts = first.to_timestamp()
+    last_ts = last.to_timestamp()
+    adjust_first = not freq.is_on_offset(first_ts)
+    adjust_last = freq.is_on_offset(last_ts)
+
+    first_ts, last_ts = _get_timestamp_range_edges(
+        first_ts, last_ts, freq, unit="ns", closed=closed, origin=origin, offset=offset
+    )
+
+    first = (first_ts + int(adjust_first) * freq).to_period(freq)
+    last = (last_ts - int(adjust_last) * freq).to_period(freq)
+    return first, last
+
+
+def _insert_nat_bin(
+    binner: PeriodIndex, bins: np.ndarray, labels: PeriodIndex, nat_count: int
+) -> tuple[PeriodIndex, np.ndarray, PeriodIndex]:
+    # NaT handling as in pandas._lib.lib.generate_bins_dt64()
+    # shift bins by the number of NaT
+    assert nat_count > 0
+    bins += nat_count
+    bins = np.insert(bins, 0, nat_count)
+
+    # Incompatible types in assignment (expression has type "Index", variable
+    # has type "PeriodIndex")
+    binner = binner.insert(0, NaT)  # type: ignore[assignment]
+    # Incompatible types in assignment (expression has type "Index", variable
+    # has type "PeriodIndex")
+    labels = labels.insert(0, NaT)  # type: ignore[assignment]
+    return binner, bins, labels
+
+
+def _adjust_dates_anchored(
+    first: Timestamp,
+    last: Timestamp,
+    freq: Tick,
+    closed: Literal["right", "left"] = "right",
+    origin: TimeGrouperOrigin = "start_day",
+    offset: Timedelta | None = None,
+    unit: str = "ns",
+) -> tuple[Timestamp, Timestamp]:
+    # First and last offsets should be calculated from the start day to fix an
+    # error cause by resampling across multiple days when a one day period is
+    # not a multiple of the frequency. See GH 8683
+    # To handle frequencies that are not multiple or divisible by a day we let
+    # the possibility to define a fixed origin timestamp. See GH 31809
+    first = first.as_unit(unit)
+    last = last.as_unit(unit)
+    if offset is not None:
+        offset = offset.as_unit(unit)
+
+    freq_value = Timedelta(freq).as_unit(unit)._value
+
+    origin_timestamp = 0  # origin == "epoch"
+    if origin == "start_day":
+        origin_timestamp = first.normalize()._value
+    elif origin == "start":
+        origin_timestamp = first._value
+    elif isinstance(origin, Timestamp):
+        origin_timestamp = origin.as_unit(unit)._value
+    elif origin in ["end", "end_day"]:
+        origin_last = last if origin == "end" else last.ceil("D")
+        sub_freq_times = (origin_last._value - first._value) // freq_value
+        if closed == "left":
+            sub_freq_times += 1
+        first = origin_last - sub_freq_times * freq
+        origin_timestamp = first._value
+    origin_timestamp += offset._value if offset else 0
+
+    # GH 10117 & GH 19375. If first and last contain timezone information,
+    # Perform the calculation in UTC in order to avoid localizing on an
+    # Ambiguous or Nonexistent time.
+    first_tzinfo = first.tzinfo
+    last_tzinfo = last.tzinfo
+    if first_tzinfo is not None:
+        first = first.tz_convert("UTC")
+    if last_tzinfo is not None:
+        last = last.tz_convert("UTC")
+
+    foffset = (first._value - origin_timestamp) % freq_value
+    loffset = (last._value - origin_timestamp) % freq_value
+
+    if closed == "right":
+        if foffset > 0:
+            # roll back
+            fresult_int = first._value - foffset
+        else:
+            fresult_int = first._value - freq_value
+
+        if loffset > 0:
+            # roll forward
+            lresult_int = last._value + (freq_value - loffset)
+        else:
+            # already the end of the road
+            lresult_int = last._value
+    else:  # closed == 'left'
+        if foffset > 0:
+            fresult_int = first._value - foffset
+        else:
+            # start of the road
+            fresult_int = first._value
+
+        if loffset > 0:
+            # roll forward
+            lresult_int = last._value + (freq_value - loffset)
+        else:
+            lresult_int = last._value + freq_value
+    fresult = Timestamp(fresult_int, unit=unit)
+    lresult = Timestamp(lresult_int, unit=unit)
+    if first_tzinfo is not None:
+        fresult = fresult.tz_localize("UTC").tz_convert(first_tzinfo)
+    if last_tzinfo is not None:
+        lresult = lresult.tz_localize("UTC").tz_convert(last_tzinfo)
+    return fresult, lresult
+
+
+def asfreq(
+    obj: NDFrameT,
+    freq,
+    method=None,
+    how=None,
+    normalize: bool = False,
+    fill_value=None,
+) -> NDFrameT:
+    """
+    Utility frequency conversion method for Series/DataFrame.
+
+    See :meth:`pandas.NDFrame.asfreq` for full documentation.
+    """
+    if isinstance(obj.index, PeriodIndex):
+        if method is not None:
+            raise NotImplementedError("'method' argument is not supported")
+
+        if how is None:
+            how = "E"
+
+        if isinstance(freq, BaseOffset):
+            if hasattr(freq, "_period_dtype_code"):
+                freq = PeriodDtype(freq)._freqstr
+
+        new_obj = obj.copy()
+        new_obj.index = obj.index.asfreq(freq, how=how)
+
+    elif len(obj.index) == 0:
+        new_obj = obj.copy()
+
+        new_obj.index = _asfreq_compat(obj.index, freq)
+    else:
+        unit = None
+        if isinstance(obj.index, DatetimeIndex):
+            # TODO: should we disallow non-DatetimeIndex?
+            unit = obj.index.unit
+        dti = date_range(obj.index.min(), obj.index.max(), freq=freq, unit=unit)
+        dti.name = obj.index.name
+        new_obj = obj.reindex(dti, method=method, fill_value=fill_value)
+        if normalize:
+            new_obj.index = new_obj.index.normalize()
+
+    return new_obj
+
+
+def _asfreq_compat(index: FreqIndexT, freq) -> FreqIndexT:
+    """
+    Helper to mimic asfreq on (empty) DatetimeIndex and TimedeltaIndex.
+
+    Parameters
+    ----------
+    index : PeriodIndex, DatetimeIndex, or TimedeltaIndex
+    freq : DateOffset
+
+    Returns
+    -------
+    same type as index
+    """
+    if len(index) != 0:
+        # This should never be reached, always checked by the caller
+        raise ValueError(
+            "Can only set arbitrary freq for empty DatetimeIndex or TimedeltaIndex"
+        )
+    if isinstance(index, PeriodIndex):
+        new_index = index.asfreq(freq=freq)
+    elif isinstance(index, DatetimeIndex):
+        new_index = DatetimeIndex([], dtype=index.dtype, freq=freq, name=index.name)
+    elif isinstance(index, TimedeltaIndex):
+        new_index = TimedeltaIndex([], dtype=index.dtype, freq=freq, name=index.name)
+    else:  # pragma: no cover
+        raise TypeError(type(index))
+    return new_index
diff --git a/pandas/pandas/core/reshape/__init__.py b/pandas/pandas/core/reshape/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/reshape/api.py b/pandas/pandas/core/reshape/api.py
new file mode 100644
index 0000000000000..b1884c497f0ad
--- /dev/null
+++ b/pandas/pandas/core/reshape/api.py
@@ -0,0 +1,41 @@
+from pandas.core.reshape.concat import concat
+from pandas.core.reshape.encoding import (
+    from_dummies,
+    get_dummies,
+)
+from pandas.core.reshape.melt import (
+    lreshape,
+    melt,
+    wide_to_long,
+)
+from pandas.core.reshape.merge import (
+    merge,
+    merge_asof,
+    merge_ordered,
+)
+from pandas.core.reshape.pivot import (
+    crosstab,
+    pivot,
+    pivot_table,
+)
+from pandas.core.reshape.tile import (
+    cut,
+    qcut,
+)
+
+__all__ = [
+    "concat",
+    "crosstab",
+    "cut",
+    "from_dummies",
+    "get_dummies",
+    "lreshape",
+    "melt",
+    "merge",
+    "merge_asof",
+    "merge_ordered",
+    "pivot",
+    "pivot_table",
+    "qcut",
+    "wide_to_long",
+]
diff --git a/pandas/pandas/core/reshape/concat.py b/pandas/pandas/core/reshape/concat.py
new file mode 100644
index 0000000000000..e7cb7069bbc26
--- /dev/null
+++ b/pandas/pandas/core/reshape/concat.py
@@ -0,0 +1,942 @@
+"""
+Concat routines.
+"""
+
+from __future__ import annotations
+
+from collections import abc
+import types
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.util._decorators import set_module
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_scalar,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.arrays.categorical import (
+    factorize_from_iterable,
+    factorize_from_iterables,
+)
+import pandas.core.common as com
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    all_indexes_same,
+    default_index,
+    ensure_index,
+    get_objs_combined_axis,
+    get_unanimous_names,
+)
+from pandas.core.internals import concatenate_managers
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterable,
+        Mapping,
+    )
+
+    from pandas._typing import (
+        Axis,
+        AxisInt,
+        HashableT,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+# ---------------------------------------------------------------------
+# Concatenate DataFrame objects
+
+
+@overload
+def concat(
+    objs: Iterable[DataFrame] | Mapping[HashableT, DataFrame],
+    *,
+    axis: Literal[0, "index"] = ...,
+    join: str = ...,
+    ignore_index: bool = ...,
+    keys: Iterable[Hashable] | None = ...,
+    levels=...,
+    names: list[HashableT] | None = ...,
+    verify_integrity: bool = ...,
+    sort: bool = ...,
+    copy: bool | lib.NoDefault = ...,
+) -> DataFrame: ...
+
+
+@overload
+def concat(
+    objs: Iterable[Series] | Mapping[HashableT, Series],
+    *,
+    axis: Literal[0, "index"] = ...,
+    join: str = ...,
+    ignore_index: bool = ...,
+    keys: Iterable[Hashable] | None = ...,
+    levels=...,
+    names: list[HashableT] | None = ...,
+    verify_integrity: bool = ...,
+    sort: bool = ...,
+    copy: bool | lib.NoDefault = ...,
+) -> Series: ...
+
+
+@overload
+def concat(
+    objs: Iterable[Series | DataFrame] | Mapping[HashableT, Series | DataFrame],
+    *,
+    axis: Literal[0, "index"] = ...,
+    join: str = ...,
+    ignore_index: bool = ...,
+    keys: Iterable[Hashable] | None = ...,
+    levels=...,
+    names: list[HashableT] | None = ...,
+    verify_integrity: bool = ...,
+    sort: bool = ...,
+    copy: bool | lib.NoDefault = ...,
+) -> DataFrame | Series: ...
+
+
+@overload
+def concat(
+    objs: Iterable[Series | DataFrame] | Mapping[HashableT, Series | DataFrame],
+    *,
+    axis: Literal[1, "columns"],
+    join: str = ...,
+    ignore_index: bool = ...,
+    keys: Iterable[Hashable] | None = ...,
+    levels=...,
+    names: list[HashableT] | None = ...,
+    verify_integrity: bool = ...,
+    sort: bool = ...,
+    copy: bool | lib.NoDefault = ...,
+) -> DataFrame: ...
+
+
+@overload
+def concat(
+    objs: Iterable[Series | DataFrame] | Mapping[HashableT, Series | DataFrame],
+    *,
+    axis: Axis = ...,
+    join: str = ...,
+    ignore_index: bool = ...,
+    keys: Iterable[Hashable] | None = ...,
+    levels=...,
+    names: list[HashableT] | None = ...,
+    verify_integrity: bool = ...,
+    sort: bool = ...,
+    copy: bool | lib.NoDefault = ...,
+) -> DataFrame | Series: ...
+
+
+@set_module("pandas")
+def concat(
+    objs: Iterable[Series | DataFrame] | Mapping[HashableT, Series | DataFrame],
+    *,
+    axis: Axis = 0,
+    join: str = "outer",
+    ignore_index: bool = False,
+    keys: Iterable[Hashable] | None = None,
+    levels=None,
+    names: list[HashableT] | None = None,
+    verify_integrity: bool = False,
+    sort: bool = False,
+    copy: bool | lib.NoDefault = lib.no_default,
+) -> DataFrame | Series:
+    """
+    Concatenate pandas objects along a particular axis.
+
+    Allows optional set logic along the other axes.
+
+    Can also add a layer of hierarchical indexing on the concatenation axis,
+    which may be useful if the labels are the same (or overlapping) on
+    the passed axis number.
+
+    Parameters
+    ----------
+    objs : an iterable or mapping of Series or DataFrame objects
+        If a mapping is passed, the keys will be used as the `keys`
+        argument, unless it is passed, in which case the values will be
+        selected (see below). Any None objects will be dropped silently unless
+        they are all None in which case a ValueError will be raised.
+    axis : {0/'index', 1/'columns'}, default 0
+        The axis to concatenate along.
+    join : {'inner', 'outer'}, default 'outer'
+        How to handle indexes on other axis (or axes).
+    ignore_index : bool, default False
+        If True, do not use the index values along the concatenation axis. The
+        resulting axis will be labeled 0, ..., n - 1. This is useful if you are
+        concatenating objects where the concatenation axis does not have
+        meaningful indexing information. Note the index values on the other
+        axes are still respected in the join.
+    keys : sequence, default None
+        If multiple levels passed, should contain tuples. Construct
+        hierarchical index using the passed keys as the outermost level.
+    levels : list of sequences, default None
+        Specific levels (unique values) to use for constructing a
+        MultiIndex. Otherwise they will be inferred from the keys.
+    names : list, default None
+        Names for the levels in the resulting hierarchical index.
+    verify_integrity : bool, default False
+        Check whether the new concatenated axis contains duplicates. This can
+        be very expensive relative to the actual data concatenation.
+    sort : bool, default False
+        Sort non-concatenation axis. One exception to this is when the
+        non-concatenation axis is a DatetimeIndex and join='outer' and the axis is
+        not already aligned. In that case, the non-concatenation axis is always
+        sorted lexicographically.
+    copy : bool, default False
+        If False, do not copy data unnecessarily.
+
+        .. note::
+            The `copy` keyword will change behavior in pandas 3.0.
+            `Copy-on-Write
+            <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+            will be enabled by default, which means that all methods with a
+            `copy` keyword will use a lazy copy mechanism to defer the copy and
+            ignore the `copy` keyword. The `copy` keyword will be removed in a
+            future version of pandas.
+
+            You can already get the future behavior and improvements through
+            enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+        .. deprecated:: 3.0.0
+
+    Returns
+    -------
+    object, type of objs
+        When concatenating all ``Series`` along the index (axis=0), a
+        ``Series`` is returned. When ``objs`` contains at least one
+        ``DataFrame``, a ``DataFrame`` is returned. When concatenating along
+        the columns (axis=1), a ``DataFrame`` is returned.
+
+    See Also
+    --------
+    DataFrame.join : Join DataFrames using indexes.
+    DataFrame.merge : Merge DataFrames by indexes or columns.
+
+    Notes
+    -----
+    The keys, levels, and names arguments are all optional.
+
+    A walkthrough of how this method fits in with other tools for combining
+    pandas objects can be found `here
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/merging.html>`__.
+
+    It is not recommended to build DataFrames by adding single rows in a
+    for loop. Build a list of rows and make a DataFrame in a single concat.
+
+    Examples
+    --------
+    Combine two ``Series``.
+
+    >>> s1 = pd.Series(["a", "b"])
+    >>> s2 = pd.Series(["c", "d"])
+    >>> pd.concat([s1, s2])
+    0    a
+    1    b
+    0    c
+    1    d
+    dtype: object
+
+    Clear the existing index and reset it in the result
+    by setting the ``ignore_index`` option to ``True``.
+
+    >>> pd.concat([s1, s2], ignore_index=True)
+    0    a
+    1    b
+    2    c
+    3    d
+    dtype: object
+
+    Add a hierarchical index at the outermost level of
+    the data with the ``keys`` option.
+
+    >>> pd.concat([s1, s2], keys=["s1", "s2"])
+    s1  0    a
+        1    b
+    s2  0    c
+        1    d
+    dtype: object
+
+    Label the index keys you create with the ``names`` option.
+
+    >>> pd.concat([s1, s2], keys=["s1", "s2"], names=["Series name", "Row ID"])
+    Series name  Row ID
+    s1           0         a
+                 1         b
+    s2           0         c
+                 1         d
+    dtype: object
+
+    Combine two ``DataFrame`` objects with identical columns.
+
+    >>> df1 = pd.DataFrame([["a", 1], ["b", 2]], columns=["letter", "number"])
+    >>> df1
+      letter  number
+    0      a       1
+    1      b       2
+    >>> df2 = pd.DataFrame([["c", 3], ["d", 4]], columns=["letter", "number"])
+    >>> df2
+      letter  number
+    0      c       3
+    1      d       4
+    >>> pd.concat([df1, df2])
+      letter  number
+    0      a       1
+    1      b       2
+    0      c       3
+    1      d       4
+
+    Combine ``DataFrame`` objects with overlapping columns
+    and return everything. Columns outside the intersection will
+    be filled with ``NaN`` values.
+
+    >>> df3 = pd.DataFrame(
+    ...     [["c", 3, "cat"], ["d", 4, "dog"]], columns=["letter", "number", "animal"]
+    ... )
+    >>> df3
+      letter  number animal
+    0      c       3    cat
+    1      d       4    dog
+    >>> pd.concat([df1, df3], sort=False)
+      letter  number animal
+    0      a       1    NaN
+    1      b       2    NaN
+    0      c       3    cat
+    1      d       4    dog
+
+    Combine ``DataFrame`` objects with overlapping columns
+    and return only those that are shared by passing ``inner`` to
+    the ``join`` keyword argument.
+
+    >>> pd.concat([df1, df3], join="inner")
+      letter  number
+    0      a       1
+    1      b       2
+    0      c       3
+    1      d       4
+
+    Combine ``DataFrame`` objects horizontally along the x axis by
+    passing in ``axis=1``.
+
+    >>> df4 = pd.DataFrame(
+    ...     [["bird", "polly"], ["monkey", "george"]], columns=["animal", "name"]
+    ... )
+    >>> pd.concat([df1, df4], axis=1)
+      letter  number  animal    name
+    0      a       1    bird   polly
+    1      b       2  monkey  george
+
+    Prevent the result from including duplicate index values with the
+    ``verify_integrity`` option.
+
+    >>> df5 = pd.DataFrame([1], index=["a"])
+    >>> df5
+       0
+    a  1
+    >>> df6 = pd.DataFrame([2], index=["a"])
+    >>> df6
+       0
+    a  2
+    >>> pd.concat([df5, df6], verify_integrity=True)
+    Traceback (most recent call last):
+        ...
+    ValueError: Indexes have overlapping values: ['a']
+
+    Append a single row to the end of a ``DataFrame`` object.
+
+    >>> df7 = pd.DataFrame({"a": 1, "b": 2}, index=[0])
+    >>> df7
+        a   b
+    0   1   2
+    >>> new_row = pd.Series({"a": 3, "b": 4})
+    >>> new_row
+    a    3
+    b    4
+    dtype: int64
+    >>> pd.concat([df7, new_row.to_frame().T], ignore_index=True)
+        a   b
+    0   1   2
+    1   3   4
+    """
+    if ignore_index and keys is not None:
+        raise ValueError(
+            f"Cannot set {ignore_index=} and specify keys. Either should be used."
+        )
+
+    if copy is not lib.no_default:
+        warnings.warn(
+            "The copy keyword is deprecated and will be removed in a future "
+            "version. Copy-on-Write is active in pandas since 3.0 which utilizes "
+            "a lazy copy mechanism that defers copies until necessary. Use "
+            ".copy() to make an eager copy if necessary.",
+            DeprecationWarning,
+            stacklevel=find_stack_level(),
+        )
+    if join == "outer":
+        intersect = False
+    elif join == "inner":
+        intersect = True
+    else:  # pragma: no cover
+        raise ValueError(
+            "Only can inner (intersect) or outer (union) join the other axis"
+        )
+
+    if not is_bool(sort):
+        raise ValueError(
+            f"The 'sort' keyword only accepts boolean values; {sort} was passed."
+        )
+    sort = bool(sort)
+
+    objs, keys, ndims = _clean_keys_and_objs(objs, keys)
+
+    # select an object to be our result reference
+    sample, objs = _get_sample_object(objs, ndims, keys, names, levels, intersect)
+
+    # Standardize axis parameter to int
+    if sample.ndim == 1:
+        from pandas import DataFrame
+
+        bm_axis = DataFrame._get_axis_number(axis)
+        is_frame = False
+        is_series = True
+    else:
+        bm_axis = sample._get_axis_number(axis)
+        is_frame = True
+        is_series = False
+
+        # Need to flip BlockManager axis in the DataFrame special case
+        bm_axis = sample._get_block_manager_axis(bm_axis)
+
+    # if we have mixed ndims, then convert to highest ndim
+    # creating column numbers as needed
+    if len(ndims) > 1:
+        objs = _sanitize_mixed_ndim(objs, sample, ignore_index, bm_axis)
+
+    axis = 1 - bm_axis if is_frame else 0
+    names = names or getattr(keys, "names", None)
+    return _get_result(
+        objs,
+        is_series,
+        bm_axis,
+        ignore_index,
+        intersect,
+        sort,
+        keys,
+        levels,
+        verify_integrity,
+        names,
+        axis,
+    )
+
+
+def _sanitize_mixed_ndim(
+    objs: list[Series | DataFrame],
+    sample: Series | DataFrame,
+    ignore_index: bool,
+    axis: AxisInt,
+) -> list[Series | DataFrame]:
+    # if we have mixed ndims, then convert to highest ndim
+    # creating column numbers as needed
+
+    new_objs = []
+
+    current_column = 0
+    max_ndim = sample.ndim
+    for obj in objs:
+        ndim = obj.ndim
+        if ndim == max_ndim:
+            pass
+
+        elif ndim != max_ndim - 1:
+            raise ValueError(
+                "cannot concatenate unaligned mixed dimensional NDFrame objects"
+            )
+
+        else:
+            name = getattr(obj, "name", None)
+            if ignore_index or name is None:
+                if axis == 1:
+                    # doing a row-wise concatenation so need everything
+                    # to line up
+                    name = 0
+                else:
+                    # doing a column-wise concatenation so need series
+                    # to have unique names
+                    name = current_column
+                    current_column += 1
+                obj = sample._constructor(obj, copy=False)
+                if isinstance(obj, ABCDataFrame):
+                    obj.columns = range(name, name + 1, 1)
+            else:
+                obj = sample._constructor({name: obj}, copy=False)
+
+        new_objs.append(obj)
+
+    return new_objs
+
+
+def _get_result(
+    objs: list[Series | DataFrame],
+    is_series: bool,
+    bm_axis: AxisInt,
+    ignore_index: bool,
+    intersect: bool,
+    sort: bool,
+    keys: Iterable[Hashable] | None,
+    levels,
+    verify_integrity: bool,
+    names: list[HashableT] | None,
+    axis: AxisInt,
+):
+    cons: Callable[..., DataFrame | Series]
+    sample: DataFrame | Series
+
+    # series only
+    if is_series:
+        sample = cast("Series", objs[0])
+
+        # stack blocks
+        if bm_axis == 0:
+            name = com.consensus_name_attr(objs)
+            cons = sample._constructor
+
+            arrs = [ser._values for ser in objs]
+
+            res = concat_compat(arrs, axis=0)
+
+            if ignore_index:
+                new_index: Index = default_index(len(res))
+            else:
+                new_index = _get_concat_axis_series(
+                    objs,
+                    ignore_index,
+                    bm_axis,
+                    keys,
+                    levels,
+                    verify_integrity,
+                    names,
+                )
+
+            mgr = type(sample._mgr).from_array(res, index=new_index)
+
+            result = sample._constructor_from_mgr(mgr, axes=mgr.axes)
+            result._name = name
+            return result.__finalize__(
+                types.SimpleNamespace(objs=objs), method="concat"
+            )
+
+        # combine as columns in a frame
+        else:
+            data = dict(enumerate(objs))
+
+            # GH28330 Preserves subclassed objects through concat
+            cons = sample._constructor_expanddim
+
+            index = get_objs_combined_axis(
+                objs,
+                axis=objs[0]._get_block_manager_axis(0),
+                intersect=intersect,
+                sort=sort,
+            )
+            columns = _get_concat_axis_series(
+                objs, ignore_index, bm_axis, keys, levels, verify_integrity, names
+            )
+            df = cons(data, index=index, copy=False)
+            df.columns = columns
+            return df.__finalize__(types.SimpleNamespace(objs=objs), method="concat")
+
+    # combine block managers
+    else:
+        sample = cast("DataFrame", objs[0])
+
+        mgrs_indexers = []
+        result_axes = new_axes(
+            objs,
+            bm_axis,
+            intersect,
+            sort,
+            keys,
+            names,
+            axis,
+            levels,
+            verify_integrity,
+            ignore_index,
+        )
+        for obj in objs:
+            indexers = {}
+            for ax, new_labels in enumerate(result_axes):
+                # ::-1 to convert BlockManager ax to DataFrame ax
+                if ax == bm_axis:
+                    # Suppress reindexing on concat axis
+                    continue
+
+                # 1-ax to convert BlockManager axis to DataFrame axis
+                obj_labels = obj.axes[1 - ax]
+                if not new_labels.equals(obj_labels):
+                    indexers[ax] = obj_labels.get_indexer(new_labels)
+
+            mgrs_indexers.append((obj._mgr, indexers))
+
+        new_data = concatenate_managers(
+            mgrs_indexers, result_axes, concat_axis=bm_axis, copy=False
+        )
+
+        out = sample._constructor_from_mgr(new_data, axes=new_data.axes)
+        return out.__finalize__(types.SimpleNamespace(objs=objs), method="concat")
+
+
+def new_axes(
+    objs: list[Series | DataFrame],
+    bm_axis: AxisInt,
+    intersect: bool,
+    sort: bool,
+    keys: Iterable[Hashable] | None,
+    names: list[HashableT] | None,
+    axis: AxisInt,
+    levels,
+    verify_integrity: bool,
+    ignore_index: bool,
+) -> list[Index]:
+    """Return the new [index, column] result for concat."""
+    return [
+        _get_concat_axis_dataframe(
+            objs,
+            axis,
+            ignore_index,
+            keys,
+            names,
+            levels,
+            verify_integrity,
+        )
+        if i == bm_axis
+        else get_objs_combined_axis(
+            objs,
+            axis=objs[0]._get_block_manager_axis(i),
+            intersect=intersect,
+            sort=sort,
+        )
+        for i in range(2)
+    ]
+
+
+def _get_concat_axis_series(
+    objs: list[Series | DataFrame],
+    ignore_index: bool,
+    bm_axis: AxisInt,
+    keys: Iterable[Hashable] | None,
+    levels,
+    verify_integrity: bool,
+    names: list[HashableT] | None,
+) -> Index:
+    """Return result concat axis when concatenating Series objects."""
+    if ignore_index:
+        return default_index(len(objs))
+    elif bm_axis == 0:
+        indexes = [x.index for x in objs]
+        if keys is None:
+            if levels is not None:
+                raise ValueError("levels supported only when keys is not None")
+            concat_axis = _concat_indexes(indexes)
+        else:
+            concat_axis = _make_concat_multiindex(indexes, keys, levels, names)
+        if verify_integrity and not concat_axis.is_unique:
+            overlap = concat_axis[concat_axis.duplicated()].unique()
+            raise ValueError(f"Indexes have overlapping values: {overlap}")
+        return concat_axis
+    elif keys is None:
+        result_names: list[Hashable] = [None] * len(objs)
+        num = 0
+        has_names = False
+        for i, x in enumerate(objs):
+            if x.ndim != 1:
+                raise TypeError(
+                    f"Cannot concatenate type 'Series' with "
+                    f"object of type '{type(x).__name__}'"
+                )
+            if x.name is not None:
+                result_names[i] = x.name
+                has_names = True
+            else:
+                result_names[i] = num
+                num += 1
+        if has_names:
+            return Index(result_names)
+        else:
+            return default_index(len(objs))
+    else:
+        return ensure_index(keys).set_names(names)  # type: ignore[arg-type]
+
+
+def _get_concat_axis_dataframe(
+    objs: list[Series | DataFrame],
+    axis: AxisInt,
+    ignore_index: bool,
+    keys: Iterable[Hashable] | None,
+    names: list[HashableT] | None,
+    levels,
+    verify_integrity: bool,
+) -> Index:
+    """Return result concat axis when concatenating DataFrame objects."""
+    indexes_gen = (x.axes[axis] for x in objs)
+
+    if ignore_index:
+        return default_index(sum(len(i) for i in indexes_gen))
+    else:
+        indexes = list(indexes_gen)
+
+    if keys is None:
+        if levels is not None:
+            raise ValueError("levels supported only when keys is not None")
+        concat_axis = _concat_indexes(indexes)
+    else:
+        concat_axis = _make_concat_multiindex(indexes, keys, levels, names)
+
+    if verify_integrity and not concat_axis.is_unique:
+        overlap = concat_axis[concat_axis.duplicated()].unique()
+        raise ValueError(f"Indexes have overlapping values: {overlap}")
+
+    return concat_axis
+
+
+def _clean_keys_and_objs(
+    objs: Iterable[Series | DataFrame] | Mapping[HashableT, Series | DataFrame],
+    keys,
+) -> tuple[list[Series | DataFrame], Index | None, set[int]]:
+    """
+    Returns
+    -------
+    clean_objs : list[Series | DataFrame]
+        List of DataFrame and Series with Nones removed.
+    keys : Index | None
+        None if keys was None
+        Index if objs was a Mapping or keys was not None. Filtered where objs was None.
+    ndim : set[int]
+        Unique .ndim attribute of obj encountered.
+    """
+    if isinstance(objs, abc.Mapping):
+        if keys is None:
+            keys = objs.keys()
+        objs = [objs[k] for k in keys]
+    elif isinstance(objs, (ABCSeries, ABCDataFrame)) or is_scalar(objs):
+        raise TypeError(
+            "first argument must be an iterable of pandas "
+            f'objects, you passed an object of type "{type(objs).__name__}"'
+        )
+    elif not isinstance(objs, abc.Sized):
+        objs = list(objs)
+
+    if len(objs) == 0:
+        raise ValueError("No objects to concatenate")
+
+    if keys is not None:
+        if not isinstance(keys, Index):
+            keys = Index(keys)
+        if len(keys) != len(objs):
+            # GH#43485
+            raise ValueError(
+                f"The length of the keys ({len(keys)}) must match "
+                f"the length of the objects to concatenate ({len(objs)})"
+            )
+
+    # GH#1649
+    key_indices = []
+    clean_objs = []
+    ndims = set()
+    for i, obj in enumerate(objs):
+        if obj is None:
+            continue
+        elif isinstance(obj, (ABCSeries, ABCDataFrame)):
+            key_indices.append(i)
+            clean_objs.append(obj)
+            ndims.add(obj.ndim)
+        else:
+            msg = (
+                f"cannot concatenate object of type '{type(obj)}'; "
+                "only Series and DataFrame objs are valid"
+            )
+            raise TypeError(msg)
+
+    if keys is not None and len(key_indices) < len(keys):
+        keys = keys.take(key_indices)
+
+    if len(clean_objs) == 0:
+        raise ValueError("All objects passed were None")
+
+    return clean_objs, keys, ndims
+
+
+def _get_sample_object(
+    objs: list[Series | DataFrame],
+    ndims: set[int],
+    keys,
+    names,
+    levels,
+    intersect: bool,
+) -> tuple[Series | DataFrame, list[Series | DataFrame]]:
+    # get the sample
+    # want the highest ndim that we have, and must be non-empty
+    # unless all objs are empty
+    if len(ndims) > 1:
+        max_ndim = max(ndims)
+        for obj in objs:
+            if obj.ndim == max_ndim and sum(obj.shape):  # type: ignore[arg-type]
+                return obj, objs
+    elif keys is None and names is None and levels is None and not intersect:
+        # filter out the empties if we have not multi-index possibilities
+        # note to keep empty Series as it affect to result columns / name
+        if ndims.pop() == 2:
+            non_empties = [obj for obj in objs if sum(obj.shape)]
+        else:
+            non_empties = objs
+
+        if len(non_empties):
+            return non_empties[0], non_empties
+
+    return objs[0], objs
+
+
+def _concat_indexes(indexes) -> Index:
+    return indexes[0].append(indexes[1:])
+
+
+def validate_unique_levels(levels: list[Index]) -> None:
+    for level in levels:
+        if not level.is_unique:
+            raise ValueError(f"Level values not unique: {level.tolist()}")
+
+
+def _make_concat_multiindex(indexes, keys, levels=None, names=None) -> MultiIndex:
+    if (levels is None and isinstance(keys[0], tuple)) or (
+        levels is not None and len(levels) > 1
+    ):
+        zipped = list(zip(*keys))
+        if names is None:
+            names = [None] * len(zipped)
+
+        if levels is None:
+            _, levels = factorize_from_iterables(zipped)
+        else:
+            levels = [ensure_index(x) for x in levels]
+            validate_unique_levels(levels)
+    else:
+        zipped = [keys]
+        if names is None:
+            names = [None]
+
+        if levels is None:
+            levels = [ensure_index(keys).unique()]
+        else:
+            levels = [ensure_index(x) for x in levels]
+            validate_unique_levels(levels)
+
+    if not all_indexes_same(indexes):
+        codes_list = []
+
+        # things are potentially different sizes, so compute the exact codes
+        # for each level and pass those to MultiIndex.from_arrays
+
+        for hlevel, level in zip(zipped, levels):
+            to_concat = []
+            if isinstance(hlevel, Index) and hlevel.equals(level):
+                lens = [len(idx) for idx in indexes]
+                codes_list.append(np.repeat(np.arange(len(hlevel)), lens))
+            else:
+                for key, index in zip(hlevel, indexes):
+                    # Find matching codes, include matching nan values as equal.
+                    mask = (isna(level) & isna(key)) | (level == key)
+                    if not mask.any():
+                        raise ValueError(f"Key {key} not in level {level}")
+                    i = np.nonzero(mask)[0][0]
+
+                    to_concat.append(np.repeat(i, len(index)))
+                codes_list.append(np.concatenate(to_concat))
+
+        concat_index = _concat_indexes(indexes)
+
+        # these go at the end
+        if isinstance(concat_index, MultiIndex):
+            levels.extend(concat_index.levels)
+            codes_list.extend(concat_index.codes)
+        else:
+            codes, categories = factorize_from_iterable(concat_index)
+            levels.append(categories)
+            codes_list.append(codes)
+
+        if len(names) == len(levels):
+            names = list(names)
+        else:
+            # make sure that all of the passed indices have the same nlevels
+            if not len({idx.nlevels for idx in indexes}) == 1:
+                raise AssertionError(
+                    "Cannot concat indices that do not have the same number of levels"
+                )
+
+            # also copies
+            names = list(names) + list(get_unanimous_names(*indexes))
+
+        return MultiIndex(
+            levels=levels, codes=codes_list, names=names, verify_integrity=False
+        )
+
+    new_index = indexes[0]
+    n = len(new_index)
+    kpieces = len(indexes)
+
+    # also copies
+    new_names = list(names)
+    new_levels = list(levels)
+
+    # construct codes
+    new_codes = []
+
+    # do something a bit more speedy
+
+    for hlevel, level in zip(zipped, levels):
+        hlevel_index = ensure_index(hlevel)
+        mapped = level.get_indexer(hlevel_index)
+
+        mask = mapped == -1
+        if mask.any():
+            raise ValueError(
+                f"Values not found in passed level: {hlevel_index[mask]!s}"
+            )
+
+        new_codes.append(np.repeat(mapped, n))
+
+    if isinstance(new_index, MultiIndex):
+        new_levels.extend(new_index.levels)
+        new_codes.extend(np.tile(lab, kpieces) for lab in new_index.codes)
+    else:
+        new_levels.append(new_index.unique())
+        single_codes = new_index.unique().get_indexer(new_index)
+        new_codes.append(np.tile(single_codes, kpieces))
+
+    if len(new_names) < len(new_levels):
+        new_names.extend(new_index.names)
+
+    return MultiIndex(
+        levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+    )
diff --git a/pandas/pandas/core/reshape/encoding.py b/pandas/pandas/core/reshape/encoding.py
new file mode 100644
index 0000000000000..6a590ee5b227e
--- /dev/null
+++ b/pandas/pandas/core/reshape/encoding.py
@@ -0,0 +1,578 @@
+from __future__ import annotations
+
+from collections import defaultdict
+from collections.abc import (
+    Hashable,
+    Iterable,
+)
+import itertools
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import missing as libmissing
+from pandas._libs.sparse import IntIndex
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_list_like,
+    is_object_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+)
+
+from pandas.core.arrays import SparseArray
+from pandas.core.arrays.categorical import factorize_from_iterable
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.frame import DataFrame
+from pandas.core.indexes.api import (
+    Index,
+    default_index,
+)
+from pandas.core.series import Series
+
+if TYPE_CHECKING:
+    from pandas._typing import NpDtype
+
+
+def get_dummies(
+    data,
+    prefix=None,
+    prefix_sep: str | Iterable[str] | dict[str, str] = "_",
+    dummy_na: bool = False,
+    columns=None,
+    sparse: bool = False,
+    drop_first: bool = False,
+    dtype: NpDtype | None = None,
+) -> DataFrame:
+    """
+    Convert categorical variable into dummy/indicator variables.
+
+    Each variable is converted in as many 0/1 variables as there are different
+    values. Columns in the output are each named after a value; if the input is
+    a DataFrame, the name of the original variable is prepended to the value.
+
+    Parameters
+    ----------
+    data : array-like, Series, or DataFrame
+        Data of which to get dummy indicators.
+    prefix : str, list of str, or dict of str, default None
+        String to append DataFrame column names.
+        Pass a list with length equal to the number of columns
+        when calling get_dummies on a DataFrame. Alternatively, `prefix`
+        can be a dictionary mapping column names to prefixes.
+    prefix_sep : str, default '_'
+        If appending prefix, separator/delimiter to use. Or pass a
+        list or dictionary as with `prefix`.
+    dummy_na : bool, default False
+        If True, a NaN indicator column will be added even if no NaN values are present.
+        If False, NA values are encoded as all zero.
+    columns : list-like, default None
+        Column names in the DataFrame to be encoded.
+        If `columns` is None then all the columns with
+        `object`, `string`, or `category` dtype will be converted.
+    sparse : bool, default False
+        Whether the dummy-encoded columns should be backed by
+        a :class:`SparseArray` (True) or a regular NumPy array (False).
+    drop_first : bool, default False
+        Whether to get k-1 dummies out of k categorical levels by removing the
+        first level.
+    dtype : dtype, default bool
+        Data type for new columns. Only a single dtype is allowed.
+
+    Returns
+    -------
+    DataFrame
+        Dummy-coded data. If `data` contains other columns than the
+        dummy-coded one(s), these will be prepended, unaltered, to the result.
+
+    See Also
+    --------
+    Series.str.get_dummies : Convert Series of strings to dummy codes.
+    :func:`~pandas.from_dummies` : Convert dummy codes to categorical ``DataFrame``.
+
+    Notes
+    -----
+    Reference :ref:`the user guide <reshaping.dummies>` for more examples.
+
+    Examples
+    --------
+    >>> s = pd.Series(list("abca"))
+
+    >>> pd.get_dummies(s)
+           a      b      c
+    0   True  False  False
+    1  False   True  False
+    2  False  False   True
+    3   True  False  False
+
+    >>> s1 = ["a", "b", np.nan]
+
+    >>> pd.get_dummies(s1)
+           a      b
+    0   True  False
+    1  False   True
+    2  False  False
+
+    >>> pd.get_dummies(s1, dummy_na=True)
+           a      b    NaN
+    0   True  False  False
+    1  False   True  False
+    2  False  False   True
+
+    >>> df = pd.DataFrame({"A": ["a", "b", "a"], "B": ["b", "a", "c"], "C": [1, 2, 3]})
+
+    >>> pd.get_dummies(df, prefix=["col1", "col2"])
+       C  col1_a  col1_b  col2_a  col2_b  col2_c
+    0  1    True   False   False    True   False
+    1  2   False    True    True   False   False
+    2  3    True   False   False   False    True
+
+    >>> pd.get_dummies(pd.Series(list("abcaa")))
+           a      b      c
+    0   True  False  False
+    1  False   True  False
+    2  False  False   True
+    3   True  False  False
+    4   True  False  False
+
+    >>> pd.get_dummies(pd.Series(list("abcaa")), drop_first=True)
+           b      c
+    0  False  False
+    1   True  False
+    2  False   True
+    3  False  False
+    4  False  False
+
+    >>> pd.get_dummies(pd.Series(list("abc")), dtype=float)
+         a    b    c
+    0  1.0  0.0  0.0
+    1  0.0  1.0  0.0
+    2  0.0  0.0  1.0
+    """
+    from pandas.core.reshape.concat import concat
+
+    dtypes_to_encode = ["object", "string", "category"]
+
+    if isinstance(data, DataFrame):
+        # determine columns being encoded
+        if columns is None:
+            data_to_encode = data.select_dtypes(include=dtypes_to_encode)
+        elif not is_list_like(columns):
+            raise TypeError("Input must be a list-like for parameter `columns`")
+        else:
+            data_to_encode = data[columns]
+
+        # validate prefixes and separator to avoid silently dropping cols
+        def check_len(item, name: str) -> None:
+            if is_list_like(item):
+                if not len(item) == data_to_encode.shape[1]:
+                    len_msg = (
+                        f"Length of '{name}' ({len(item)}) did not match the "
+                        "length of the columns being encoded "
+                        f"({data_to_encode.shape[1]})."
+                    )
+                    raise ValueError(len_msg)
+
+        check_len(prefix, "prefix")
+        check_len(prefix_sep, "prefix_sep")
+
+        if isinstance(prefix, str):
+            prefix = itertools.cycle([prefix])
+        if isinstance(prefix, dict):
+            prefix = [prefix[col] for col in data_to_encode.columns]
+
+        if prefix is None:
+            prefix = data_to_encode.columns
+
+        # validate separators
+        if isinstance(prefix_sep, str):
+            prefix_sep = itertools.cycle([prefix_sep])
+        elif isinstance(prefix_sep, dict):
+            prefix_sep = [prefix_sep[col] for col in data_to_encode.columns]
+
+        with_dummies: list[DataFrame]
+        if data_to_encode.shape == data.shape:
+            # Encoding the entire df, do not prepend any dropped columns
+            with_dummies = []
+        elif columns is not None:
+            # Encoding only cols specified in columns. Get all cols not in
+            # columns to prepend to result.
+            with_dummies = [data.drop(columns, axis=1)]
+        else:
+            # Encoding only object and category dtype columns. Get remaining
+            # columns to prepend to result.
+            with_dummies = [data.select_dtypes(exclude=dtypes_to_encode)]
+
+        for col, pre, sep in zip(data_to_encode.items(), prefix, prefix_sep):
+            # col is (column_name, column), use just column data here
+            dummy = _get_dummies_1d(
+                col[1],
+                prefix=pre,
+                prefix_sep=sep,
+                dummy_na=dummy_na,
+                sparse=sparse,
+                drop_first=drop_first,
+                dtype=dtype,
+            )
+            with_dummies.append(dummy)
+        result = concat(with_dummies, axis=1)
+    else:
+        result = _get_dummies_1d(
+            data,
+            prefix,
+            prefix_sep,
+            dummy_na,
+            sparse=sparse,
+            drop_first=drop_first,
+            dtype=dtype,
+        )
+    return result
+
+
+def _get_dummies_1d(
+    data,
+    prefix,
+    prefix_sep: str | Iterable[str] | dict[str, str] = "_",
+    dummy_na: bool = False,
+    sparse: bool = False,
+    drop_first: bool = False,
+    dtype: NpDtype | None = None,
+) -> DataFrame:
+    from pandas.core.reshape.concat import concat
+
+    # Series avoids inconsistent NaN handling
+    codes, levels = factorize_from_iterable(Series(data, copy=False))
+
+    if dtype is None and hasattr(data, "dtype"):
+        input_dtype = data.dtype
+        if isinstance(input_dtype, CategoricalDtype):
+            input_dtype = input_dtype.categories.dtype
+
+        if isinstance(input_dtype, ArrowDtype):
+            import pyarrow as pa
+
+            dtype = ArrowDtype(pa.bool_())  # type: ignore[assignment]
+        elif (
+            isinstance(input_dtype, StringDtype)
+            and input_dtype.na_value is libmissing.NA
+        ):
+            dtype = pandas_dtype("boolean")  # type: ignore[assignment]
+        else:
+            dtype = np.dtype(bool)
+    elif dtype is None:
+        dtype = np.dtype(bool)
+
+    _dtype = pandas_dtype(dtype)
+
+    if is_object_dtype(_dtype):
+        raise ValueError("dtype=object is not a valid dtype for get_dummies")
+
+    def get_empty_frame(data) -> DataFrame:
+        index: Index | np.ndarray
+        if isinstance(data, Series):
+            index = data.index
+        else:
+            index = default_index(len(data))
+        return DataFrame(index=index)
+
+    # if all NaN
+    if not dummy_na and len(levels) == 0:
+        return get_empty_frame(data)
+
+    codes = codes.copy()
+    if dummy_na:
+        codes[codes == -1] = len(levels)
+        levels = levels.insert(len(levels), np.nan)
+
+    # if dummy_na, we just fake a nan level. drop_first will drop it again
+    if drop_first and len(levels) == 1:
+        return get_empty_frame(data)
+
+    number_of_cols = len(levels)
+
+    if prefix is None:
+        dummy_cols = levels
+    else:
+        dummy_cols = Index([f"{prefix}{prefix_sep}{level}" for level in levels])
+
+    index: Index | None
+    if isinstance(data, Series):
+        index = data.index
+    else:
+        index = None
+
+    if sparse:
+        fill_value: bool | float
+        if is_integer_dtype(dtype):
+            fill_value = 0
+        elif dtype == np.dtype(bool):
+            fill_value = False
+        else:
+            fill_value = 0.0
+
+        sparse_series = []
+        N = len(data)
+        sp_indices: list[list] = [[] for _ in range(len(dummy_cols))]
+        mask = codes != -1
+        codes = codes[mask]
+        n_idx = np.arange(N)[mask]
+
+        for ndx, code in zip(n_idx, codes):
+            sp_indices[code].append(ndx)
+
+        if drop_first:
+            # remove first categorical level to avoid perfect collinearity
+            # GH12042
+            sp_indices = sp_indices[1:]
+            dummy_cols = dummy_cols[1:]
+        for col, ixs in zip(dummy_cols, sp_indices):
+            sarr = SparseArray(
+                np.ones(len(ixs), dtype=dtype),
+                sparse_index=IntIndex(N, ixs),
+                fill_value=fill_value,
+                dtype=dtype,
+            )
+            sparse_series.append(Series(data=sarr, index=index, name=col, copy=False))
+
+        return concat(sparse_series, axis=1)
+
+    else:
+        # ensure ndarray layout is column-major
+        shape = len(codes), number_of_cols
+        dummy_dtype: NpDtype
+        if isinstance(_dtype, np.dtype):
+            dummy_dtype = _dtype
+        else:
+            dummy_dtype = np.bool_
+        dummy_mat = np.zeros(shape=shape, dtype=dummy_dtype, order="F")
+        dummy_mat[np.arange(len(codes)), codes] = 1
+
+        if not dummy_na:
+            # reset NaN GH4446
+            dummy_mat[codes == -1] = 0
+
+        if drop_first:
+            # remove first GH12042
+            dummy_mat = dummy_mat[:, 1:]
+            dummy_cols = dummy_cols[1:]
+        return DataFrame(dummy_mat, index=index, columns=dummy_cols, dtype=_dtype)
+
+
+def from_dummies(
+    data: DataFrame,
+    sep: None | str = None,
+    default_category: None | Hashable | dict[str, Hashable] = None,
+) -> DataFrame:
+    """
+    Create a categorical ``DataFrame`` from a ``DataFrame`` of dummy variables.
+
+    Inverts the operation performed by :func:`~pandas.get_dummies`.
+
+    .. versionadded:: 1.5.0
+
+    Parameters
+    ----------
+    data : DataFrame
+        Data which contains dummy-coded variables in form of integer columns of
+        1's and 0's.
+    sep : str, default None
+        Separator used in the column names of the dummy categories they are
+        character indicating the separation of the categorical names from the prefixes.
+        For example, if your column names are 'prefix_A' and 'prefix_B',
+        you can strip the underscore by specifying sep='_'.
+    default_category : None, Hashable or dict of Hashables, default None
+        The default category is the implied category when a value has none of the
+        listed categories specified with a one, i.e. if all dummies in a row are
+        zero. Can be a single value for all variables or a dict directly mapping
+        the default categories to a prefix of a variable.
+
+    Returns
+    -------
+    DataFrame
+        Categorical data decoded from the dummy input-data.
+
+    Raises
+    ------
+    ValueError
+        * When the input ``DataFrame`` ``data`` contains NA values.
+        * When the input ``DataFrame`` ``data`` contains column names with separators
+          that do not match the separator specified with ``sep``.
+        * When a ``dict`` passed to ``default_category`` does not include an implied
+          category for each prefix.
+        * When a value in ``data`` has more than one category assigned to it.
+        * When ``default_category=None`` and a value in ``data`` has no category
+          assigned to it.
+    TypeError
+        * When the input ``data`` is not of type ``DataFrame``.
+        * When the input ``DataFrame`` ``data`` contains non-dummy data.
+        * When the passed ``sep`` is of a wrong data type.
+        * When the passed ``default_category`` is of a wrong data type.
+
+    See Also
+    --------
+    :func:`~pandas.get_dummies` : Convert ``Series`` or ``DataFrame`` to dummy codes.
+    :class:`~pandas.Categorical` : Represent a categorical variable in classic.
+
+    Notes
+    -----
+    The columns of the passed dummy data should only include 1's and 0's,
+    or boolean values.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"a": [1, 0, 0, 1], "b": [0, 1, 0, 0], "c": [0, 0, 1, 0]})
+
+    >>> df
+       a  b  c
+    0  1  0  0
+    1  0  1  0
+    2  0  0  1
+    3  1  0  0
+
+    >>> pd.from_dummies(df)
+    0     a
+    1     b
+    2     c
+    3     a
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "col1_a": [1, 0, 1],
+    ...         "col1_b": [0, 1, 0],
+    ...         "col2_a": [0, 1, 0],
+    ...         "col2_b": [1, 0, 0],
+    ...         "col2_c": [0, 0, 1],
+    ...     }
+    ... )
+
+    >>> df
+          col1_a  col1_b  col2_a  col2_b  col2_c
+    0       1       0       0       1       0
+    1       0       1       1       0       0
+    2       1       0       0       0       1
+
+    >>> pd.from_dummies(df, sep="_")
+        col1    col2
+    0    a       b
+    1    b       a
+    2    a       c
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "col1_a": [1, 0, 0],
+    ...         "col1_b": [0, 1, 0],
+    ...         "col2_a": [0, 1, 0],
+    ...         "col2_b": [1, 0, 0],
+    ...         "col2_c": [0, 0, 0],
+    ...     }
+    ... )
+
+    >>> df
+          col1_a  col1_b  col2_a  col2_b  col2_c
+    0       1       0       0       1       0
+    1       0       1       1       0       0
+    2       0       0       0       0       0
+
+    >>> pd.from_dummies(df, sep="_", default_category={"col1": "d", "col2": "e"})
+        col1    col2
+    0    a       b
+    1    b       a
+    2    d       e
+    """
+    from pandas.core.reshape.concat import concat
+
+    if not isinstance(data, DataFrame):
+        raise TypeError(
+            "Expected 'data' to be a 'DataFrame'; "
+            f"Received 'data' of type: {type(data).__name__}"
+        )
+
+    col_isna_mask = data.isna().any()
+
+    if col_isna_mask.any():
+        raise ValueError(
+            f"Dummy DataFrame contains NA value in column: '{col_isna_mask.idxmax()}'"
+        )
+
+    # index data with a list of all columns that are dummies
+    try:
+        data_to_decode = data.astype("boolean")
+    except TypeError as err:
+        raise TypeError("Passed DataFrame contains non-dummy data") from err
+
+    # collect prefixes and get lists to slice data for each prefix
+    variables_slice = defaultdict(list)
+    if sep is None:
+        variables_slice[""] = list(data.columns)
+    elif isinstance(sep, str):
+        for col in data_to_decode.columns:
+            prefix = col.split(sep)[0]
+            if len(prefix) == len(col):
+                raise ValueError(f"Separator not specified for column: {col}")
+            variables_slice[prefix].append(col)
+    else:
+        raise TypeError(
+            "Expected 'sep' to be of type 'str' or 'None'; "
+            f"Received 'sep' of type: {type(sep).__name__}"
+        )
+
+    if default_category is not None:
+        if isinstance(default_category, dict):
+            if not len(default_category) == len(variables_slice):
+                len_msg = (
+                    f"Length of 'default_category' ({len(default_category)}) "
+                    f"did not match the length of the columns being encoded "
+                    f"({len(variables_slice)})"
+                )
+                raise ValueError(len_msg)
+        elif isinstance(default_category, Hashable):
+            default_category = dict(
+                zip(variables_slice, [default_category] * len(variables_slice))
+            )
+        else:
+            raise TypeError(
+                "Expected 'default_category' to be of type "
+                "'None', 'Hashable', or 'dict'; "
+                "Received 'default_category' of type: "
+                f"{type(default_category).__name__}"
+            )
+
+    cat_data = {}
+    for prefix, prefix_slice in variables_slice.items():
+        if sep is None:
+            cats = prefix_slice.copy()
+        else:
+            cats = [col[len(prefix + sep) :] for col in prefix_slice]
+        assigned = data_to_decode.loc[:, prefix_slice].sum(axis=1)
+        if any(assigned > 1):
+            raise ValueError(
+                "Dummy DataFrame contains multi-assignment(s); "
+                f"First instance in row: {assigned.idxmax()}"
+            )
+        if any(assigned == 0):
+            if isinstance(default_category, dict):
+                cats.append(default_category[prefix])
+            else:
+                raise ValueError(
+                    "Dummy DataFrame contains unassigned value(s); "
+                    f"First instance in row: {assigned.idxmin()}"
+                )
+            data_slice = concat(
+                (data_to_decode.loc[:, prefix_slice], assigned == 0), axis=1
+            )
+        else:
+            data_slice = data_to_decode.loc[:, prefix_slice]
+        cats_array = data._constructor_sliced(cats, dtype=data.columns.dtype)
+        # get indices of True entries along axis=1
+        true_values = data_slice.idxmax(axis=1)
+        indexer = data_slice.columns.get_indexer_for(true_values)
+        cat_data[prefix] = cats_array.take(indexer).set_axis(data.index)
+
+    result = DataFrame(cat_data)
+    if sep is not None:
+        result.columns = result.columns.astype(data.columns.dtype)
+    return result
diff --git a/pandas/pandas/core/reshape/melt.py b/pandas/pandas/core/reshape/melt.py
new file mode 100644
index 0000000000000..f4cb82816bbcf
--- /dev/null
+++ b/pandas/pandas/core/reshape/melt.py
@@ -0,0 +1,667 @@
+from __future__ import annotations
+
+import re
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.common import (
+    is_iterator,
+    is_list_like,
+)
+from pandas.core.dtypes.concat import concat_compat
+from pandas.core.dtypes.missing import notna
+
+import pandas.core.algorithms as algos
+from pandas.core.indexes.api import MultiIndex
+from pandas.core.reshape.concat import concat
+from pandas.core.tools.numeric import to_numeric
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import AnyArrayLike
+
+    from pandas import DataFrame
+
+
+def ensure_list_vars(arg_vars, variable: str, columns) -> list:
+    if arg_vars is not None:
+        if not is_list_like(arg_vars):
+            return [arg_vars]
+        elif isinstance(columns, MultiIndex) and not isinstance(arg_vars, list):
+            raise ValueError(
+                f"{variable} must be a list of tuples when columns are a MultiIndex"
+            )
+        else:
+            return list(arg_vars)
+    else:
+        return []
+
+
+def melt(
+    frame: DataFrame,
+    id_vars=None,
+    value_vars=None,
+    var_name=None,
+    value_name: Hashable = "value",
+    col_level=None,
+    ignore_index: bool = True,
+) -> DataFrame:
+    """
+    Unpivot a DataFrame from wide to long format, optionally leaving identifiers set.
+
+    This function is useful to reshape a DataFrame into a format where one
+    or more columns are identifier variables (`id_vars`), while all other
+    columns are considered measured variables (`value_vars`), and are "unpivoted" to
+    the row axis, leaving just two non-identifier columns, 'variable' and
+    'value'.
+
+    Parameters
+    ----------
+    frame : DataFrame
+        The DataFrame to unpivot.
+    id_vars : scalar, tuple, list, or ndarray, optional
+        Column(s) to use as identifier variables.
+    value_vars : scalar, tuple, list, or ndarray, optional
+        Column(s) to unpivot. If not specified, uses all columns that
+        are not set as `id_vars`.
+    var_name : scalar, tuple, list, or ndarray, optional
+        Name to use for the 'variable' column. If None it uses
+        ``frame.columns.name`` or 'variable'. Must be a scalar if columns are a
+        MultiIndex.
+    value_name : scalar, default 'value'
+        Name to use for the 'value' column, can't be an existing column label.
+    col_level : scalar, optional
+        If columns are a MultiIndex then use this level to melt.
+    ignore_index : bool, default True
+        If True, original index is ignored. If False, the original index is retained.
+        Index labels will be repeated as necessary.
+
+    Returns
+    -------
+    DataFrame
+        Unpivoted DataFrame.
+
+    See Also
+    --------
+    DataFrame.melt : Identical method.
+    pivot_table : Create a spreadsheet-style pivot table as a DataFrame.
+    DataFrame.pivot : Return reshaped DataFrame organized
+        by given index / column values.
+    DataFrame.explode : Explode a DataFrame from list-like
+            columns to long format.
+
+    Notes
+    -----
+    Reference :ref:`the user guide <reshaping.melt>` for more examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "A": {0: "a", 1: "b", 2: "c"},
+    ...         "B": {0: 1, 1: 3, 2: 5},
+    ...         "C": {0: 2, 1: 4, 2: 6},
+    ...     }
+    ... )
+    >>> df
+    A  B  C
+    0  a  1  2
+    1  b  3  4
+    2  c  5  6
+
+    >>> pd.melt(df, id_vars=["A"], value_vars=["B"])
+    A variable  value
+    0  a        B      1
+    1  b        B      3
+    2  c        B      5
+
+    >>> pd.melt(df, id_vars=["A"], value_vars=["B", "C"])
+    A variable  value
+    0  a        B      1
+    1  b        B      3
+    2  c        B      5
+    3  a        C      2
+    4  b        C      4
+    5  c        C      6
+
+    The names of 'variable' and 'value' columns can be customized:
+
+    >>> pd.melt(
+    ...     df,
+    ...     id_vars=["A"],
+    ...     value_vars=["B"],
+    ...     var_name="myVarname",
+    ...     value_name="myValname",
+    ... )
+    A myVarname  myValname
+    0  a         B          1
+    1  b         B          3
+    2  c         B          5
+
+    Original index values can be kept around:
+
+    >>> pd.melt(df, id_vars=["A"], value_vars=["B", "C"], ignore_index=False)
+    A variable  value
+    0  a        B      1
+    1  b        B      3
+    2  c        B      5
+    0  a        C      2
+    1  b        C      4
+    2  c        C      6
+
+    If you have multi-index columns:
+
+    >>> df.columns = [list("ABC"), list("DEF")]
+    >>> df
+    A  B  C
+    D  E  F
+    0  a  1  2
+    1  b  3  4
+    2  c  5  6
+
+    >>> pd.melt(df, col_level=0, id_vars=["A"], value_vars=["B"])
+    A variable  value
+    0  a        B      1
+    1  b        B      3
+    2  c        B      5
+
+    >>> pd.melt(df, id_vars=[("A", "D")], value_vars=[("B", "E")])
+    (A, D) variable_0 variable_1  value
+    0      a          B          E      1
+    1      b          B          E      3
+    2      c          B          E      5
+    """
+    if value_name in frame.columns:
+        raise ValueError(
+            f"value_name ({value_name}) cannot match an element in "
+            "the DataFrame columns."
+        )
+    id_vars = ensure_list_vars(id_vars, "id_vars", frame.columns)
+    value_vars_was_not_none = value_vars is not None
+    value_vars = ensure_list_vars(value_vars, "value_vars", frame.columns)
+
+    if id_vars or value_vars:
+        if col_level is not None:
+            level = frame.columns.get_level_values(col_level)
+        else:
+            level = frame.columns
+        labels = id_vars + value_vars
+        idx = level.get_indexer_for(labels)
+        missing = idx == -1
+        if missing.any():
+            missing_labels = [
+                lab for lab, not_found in zip(labels, missing) if not_found
+            ]
+            raise KeyError(
+                "The following id_vars or value_vars are not present in "
+                f"the DataFrame: {missing_labels}"
+            )
+        if value_vars_was_not_none:
+            frame = frame.iloc[:, algos.unique(idx)]
+        else:
+            frame = frame.copy(deep=False)
+    else:
+        frame = frame.copy(deep=False)
+
+    if col_level is not None:  # allow list or other?
+        # frame is a copy
+        frame.columns = frame.columns.get_level_values(col_level)
+
+    if var_name is None:
+        if isinstance(frame.columns, MultiIndex):
+            if len(frame.columns.names) == len(set(frame.columns.names)):
+                var_name = frame.columns.names
+            else:
+                var_name = [f"variable_{i}" for i in range(len(frame.columns.names))]
+        else:
+            var_name = [
+                frame.columns.name if frame.columns.name is not None else "variable"
+            ]
+    elif is_list_like(var_name):
+        if isinstance(frame.columns, MultiIndex):
+            if is_iterator(var_name):
+                var_name = list(var_name)
+            if len(var_name) > len(frame.columns):
+                raise ValueError(
+                    f"{var_name=} has {len(var_name)} items, "
+                    f"but the dataframe columns only have {len(frame.columns)} levels."
+                )
+        else:
+            raise ValueError(f"{var_name=} must be a scalar.")
+    else:
+        var_name = [var_name]
+
+    num_rows, K = frame.shape
+    num_cols_adjusted = K - len(id_vars)
+
+    mdata: dict[Hashable, AnyArrayLike] = {}
+    for col in id_vars:
+        id_data = frame.pop(col)
+        if not isinstance(id_data.dtype, np.dtype):
+            # i.e. ExtensionDtype
+            if num_cols_adjusted > 0:
+                mdata[col] = concat([id_data] * num_cols_adjusted, ignore_index=True)
+            else:
+                # We can't concat empty list. (GH 46044)
+                mdata[col] = type(id_data)([], name=id_data.name, dtype=id_data.dtype)
+        else:
+            mdata[col] = np.tile(id_data._values, num_cols_adjusted)
+
+    mcolumns = id_vars + var_name + [value_name]
+
+    if frame.shape[1] > 0 and not any(
+        not isinstance(dt, np.dtype) and dt._supports_2d for dt in frame.dtypes
+    ):
+        mdata[value_name] = concat(
+            [frame.iloc[:, i] for i in range(frame.shape[1])], ignore_index=True
+        ).values
+    else:
+        mdata[value_name] = frame._values.ravel("F")
+    for i, col in enumerate(var_name):
+        mdata[col] = frame.columns._get_level_values(i).repeat(num_rows)
+
+    result = frame._constructor(mdata, columns=mcolumns)
+
+    if not ignore_index:
+        taker = np.tile(np.arange(len(frame)), num_cols_adjusted)
+        result.index = frame.index.take(taker)
+
+    return result
+
+
+def lreshape(data: DataFrame, groups: dict, dropna: bool = True) -> DataFrame:
+    """
+    Reshape wide-format data to long. Generalized inverse of DataFrame.pivot.
+
+    Accepts a dictionary, ``groups``, in which each key is a new column name
+    and each value is a list of old column names that will be "melted" under
+    the new column name as part of the reshape.
+
+    Parameters
+    ----------
+    data : DataFrame
+        The wide-format DataFrame.
+    groups : dict
+        {new_name : list_of_columns}.
+    dropna : bool, default True
+        Do not include columns whose entries are all NaN.
+
+    Returns
+    -------
+    DataFrame
+        Reshaped DataFrame.
+
+    See Also
+    --------
+    melt : Unpivot a DataFrame from wide to long format, optionally leaving
+        identifiers set.
+    pivot : Create a spreadsheet-style pivot table as a DataFrame.
+    DataFrame.pivot : Pivot without aggregation that can handle
+        non-numeric data.
+    DataFrame.pivot_table : Generalization of pivot that can handle
+        duplicate values for one index/column pair.
+    DataFrame.unstack : Pivot based on the index values instead of a
+        column.
+    wide_to_long : Wide panel to long format. Less flexible but more
+        user-friendly than melt.
+
+    Examples
+    --------
+    >>> data = pd.DataFrame(
+    ...     {
+    ...         "hr1": [514, 573],
+    ...         "hr2": [545, 526],
+    ...         "team": ["Red Sox", "Yankees"],
+    ...         "year1": [2007, 2007],
+    ...         "year2": [2008, 2008],
+    ...     }
+    ... )
+    >>> data
+       hr1  hr2     team  year1  year2
+    0  514  545  Red Sox   2007   2008
+    1  573  526  Yankees   2007   2008
+
+    >>> pd.lreshape(data, {"year": ["year1", "year2"], "hr": ["hr1", "hr2"]})
+          team  year   hr
+    0  Red Sox  2007  514
+    1  Yankees  2007  573
+    2  Red Sox  2008  545
+    3  Yankees  2008  526
+    """
+    mdata = {}
+    pivot_cols = []
+    all_cols: set[Hashable] = set()
+    K = len(next(iter(groups.values())))
+    for target, names in groups.items():
+        if len(names) != K:
+            raise ValueError("All column lists must be same length")
+        to_concat = [data[col]._values for col in names]
+
+        mdata[target] = concat_compat(to_concat)
+        pivot_cols.append(target)
+        all_cols = all_cols.union(names)
+
+    id_cols = list(data.columns.difference(all_cols))
+    for col in id_cols:
+        mdata[col] = np.tile(data[col]._values, K)
+
+    if dropna:
+        mask = np.ones(len(mdata[pivot_cols[0]]), dtype=bool)
+        for c in pivot_cols:
+            mask &= notna(mdata[c])
+        if not mask.all():
+            mdata = {k: v[mask] for k, v in mdata.items()}
+
+    return data._constructor(mdata, columns=id_cols + pivot_cols)
+
+
+def wide_to_long(
+    df: DataFrame, stubnames, i, j, sep: str = "", suffix: str = r"\d+"
+) -> DataFrame:
+    r"""
+    Unpivot a DataFrame from wide to long format.
+
+    Less flexible but more user-friendly than melt.
+
+    With stubnames ['A', 'B'], this function expects to find one or more
+    group of columns with format
+    A-suffix1, A-suffix2,..., B-suffix1, B-suffix2,...
+    You specify what you want to call this suffix in the resulting long format
+    with `j` (for example `j='year'`)
+
+    Each row of these wide variables are assumed to be uniquely identified by
+    `i` (can be a single column name or a list of column names)
+
+    All remaining variables in the data frame are left intact.
+
+    Parameters
+    ----------
+    df : DataFrame
+        The wide-format DataFrame.
+    stubnames : str or list-like
+        The stub name(s). The wide format variables are assumed to
+        start with the stub names.
+    i : str or list-like
+        Column(s) to use as id variable(s).
+    j : str
+        The name of the sub-observation variable. What you wish to name your
+        suffix in the long format.
+    sep : str, default ""
+        A character indicating the separation of the variable names
+        in the wide format, to be stripped from the names in the long format.
+        For example, if your column names are A-suffix1, A-suffix2, you
+        can strip the hyphen by specifying `sep='-'`.
+    suffix : str, default '\\d+'
+        A regular expression capturing the wanted suffixes. '\\d+' captures
+        numeric suffixes. Suffixes with no numbers could be specified with the
+        negated character class '\\D+'. You can also further disambiguate
+        suffixes, for example, if your wide variables are of the form A-one,
+        B-two,.., and you have an unrelated column A-rating, you can ignore the
+        last one by specifying `suffix='(!?one|two)'`. When all suffixes are
+        numeric, they are cast to int64/float64.
+
+    Returns
+    -------
+    DataFrame
+        A DataFrame that contains each stub name as a variable, with new index
+        (i, j).
+
+    See Also
+    --------
+    melt : Unpivot a DataFrame from wide to long format, optionally leaving
+        identifiers set.
+    pivot : Create a spreadsheet-style pivot table as a DataFrame.
+    DataFrame.pivot : Pivot without aggregation that can handle
+        non-numeric data.
+    DataFrame.pivot_table : Generalization of pivot that can handle
+        duplicate values for one index/column pair.
+    DataFrame.unstack : Pivot based on the index values instead of a
+        column.
+
+    Notes
+    -----
+    All extra variables are left untouched. This simply uses
+    `pandas.melt` under the hood, but is hard-coded to "do the right thing"
+    in a typical case.
+
+    Examples
+    --------
+    >>> np.random.seed(123)
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "A1970": {0: "a", 1: "b", 2: "c"},
+    ...         "A1980": {0: "d", 1: "e", 2: "f"},
+    ...         "B1970": {0: 2.5, 1: 1.2, 2: 0.7},
+    ...         "B1980": {0: 3.2, 1: 1.3, 2: 0.1},
+    ...         "X": dict(zip(range(3), np.random.randn(3))),
+    ...     }
+    ... )
+    >>> df["id"] = df.index
+    >>> df
+      A1970 A1980  B1970  B1980         X  id
+    0     a     d    2.5    3.2 -1.085631   0
+    1     b     e    1.2    1.3  0.997345   1
+    2     c     f    0.7    0.1  0.282978   2
+    >>> pd.wide_to_long(df, ["A", "B"], i="id", j="year")
+    ... # doctest: +NORMALIZE_WHITESPACE
+                    X  A    B
+    id year
+    0  1970 -1.085631  a  2.5
+    1  1970  0.997345  b  1.2
+    2  1970  0.282978  c  0.7
+    0  1980 -1.085631  d  3.2
+    1  1980  0.997345  e  1.3
+    2  1980  0.282978  f  0.1
+
+    With multiple id columns
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "famid": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+    ...         "birth": [1, 2, 3, 1, 2, 3, 1, 2, 3],
+    ...         "ht1": [2.8, 2.9, 2.2, 2, 1.8, 1.9, 2.2, 2.3, 2.1],
+    ...         "ht2": [3.4, 3.8, 2.9, 3.2, 2.8, 2.4, 3.3, 3.4, 2.9],
+    ...     }
+    ... )
+    >>> df
+       famid  birth  ht1  ht2
+    0      1      1  2.8  3.4
+    1      1      2  2.9  3.8
+    2      1      3  2.2  2.9
+    3      2      1  2.0  3.2
+    4      2      2  1.8  2.8
+    5      2      3  1.9  2.4
+    6      3      1  2.2  3.3
+    7      3      2  2.3  3.4
+    8      3      3  2.1  2.9
+    >>> long_format = pd.wide_to_long(df, stubnames="ht", i=["famid", "birth"], j="age")
+    >>> long_format
+    ... # doctest: +NORMALIZE_WHITESPACE
+                      ht
+    famid birth age
+    1     1     1    2.8
+                2    3.4
+          2     1    2.9
+                2    3.8
+          3     1    2.2
+                2    2.9
+    2     1     1    2.0
+                2    3.2
+          2     1    1.8
+                2    2.8
+          3     1    1.9
+                2    2.4
+    3     1     1    2.2
+                2    3.3
+          2     1    2.3
+                2    3.4
+          3     1    2.1
+                2    2.9
+
+    Going from long back to wide just takes some creative use of `unstack`
+
+    >>> wide_format = long_format.unstack()
+    >>> wide_format.columns = wide_format.columns.map("{0[0]}{0[1]}".format)
+    >>> wide_format.reset_index()
+       famid  birth  ht1  ht2
+    0      1      1  2.8  3.4
+    1      1      2  2.9  3.8
+    2      1      3  2.2  2.9
+    3      2      1  2.0  3.2
+    4      2      2  1.8  2.8
+    5      2      3  1.9  2.4
+    6      3      1  2.2  3.3
+    7      3      2  2.3  3.4
+    8      3      3  2.1  2.9
+
+    Less wieldy column names are also handled
+
+    >>> np.random.seed(0)
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "A(weekly)-2010": np.random.rand(3),
+    ...         "A(weekly)-2011": np.random.rand(3),
+    ...         "B(weekly)-2010": np.random.rand(3),
+    ...         "B(weekly)-2011": np.random.rand(3),
+    ...         "X": np.random.randint(3, size=3),
+    ...     }
+    ... )
+    >>> df["id"] = df.index
+    >>> df  # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
+       A(weekly)-2010  A(weekly)-2011  B(weekly)-2010  B(weekly)-2011  X  id
+    0        0.548814        0.544883        0.437587        0.383442  0   0
+    1        0.715189        0.423655        0.891773        0.791725  1   1
+    2        0.602763        0.645894        0.963663        0.528895  1   2
+
+    >>> pd.wide_to_long(df, ["A(weekly)", "B(weekly)"], i="id", j="year", sep="-")
+    ... # doctest: +NORMALIZE_WHITESPACE
+             X  A(weekly)  B(weekly)
+    id year
+    0  2010  0   0.548814   0.437587
+    1  2010  1   0.715189   0.891773
+    2  2010  1   0.602763   0.963663
+    0  2011  0   0.544883   0.383442
+    1  2011  1   0.423655   0.791725
+    2  2011  1   0.645894   0.528895
+
+    If we have many columns, we could also use a regex to find our
+    stubnames and pass that list on to wide_to_long
+
+    >>> stubnames = sorted(
+    ...     set(
+    ...         [
+    ...             match[0]
+    ...             for match in df.columns.str.findall(r"[A-B]\(.*\)").values
+    ...             if match != []
+    ...         ]
+    ...     )
+    ... )
+    >>> list(stubnames)
+    ['A(weekly)', 'B(weekly)']
+
+    All of the above examples have integers as suffixes. It is possible to
+    have non-integers as suffixes.
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "famid": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+    ...         "birth": [1, 2, 3, 1, 2, 3, 1, 2, 3],
+    ...         "ht_one": [2.8, 2.9, 2.2, 2, 1.8, 1.9, 2.2, 2.3, 2.1],
+    ...         "ht_two": [3.4, 3.8, 2.9, 3.2, 2.8, 2.4, 3.3, 3.4, 2.9],
+    ...     }
+    ... )
+    >>> df
+       famid  birth  ht_one  ht_two
+    0      1      1     2.8     3.4
+    1      1      2     2.9     3.8
+    2      1      3     2.2     2.9
+    3      2      1     2.0     3.2
+    4      2      2     1.8     2.8
+    5      2      3     1.9     2.4
+    6      3      1     2.2     3.3
+    7      3      2     2.3     3.4
+    8      3      3     2.1     2.9
+
+    >>> long_format = pd.wide_to_long(
+    ...     df, stubnames="ht", i=["famid", "birth"], j="age", sep="_", suffix=r"\w+"
+    ... )
+    >>> long_format
+    ... # doctest: +NORMALIZE_WHITESPACE
+                      ht
+    famid birth age
+    1     1     one  2.8
+                two  3.4
+          2     one  2.9
+                two  3.8
+          3     one  2.2
+                two  2.9
+    2     1     one  2.0
+                two  3.2
+          2     one  1.8
+                two  2.8
+          3     one  1.9
+                two  2.4
+    3     1     one  2.2
+                two  3.3
+          2     one  2.3
+                two  3.4
+          3     one  2.1
+                two  2.9
+    """
+
+    def get_var_names(df, stub: str, sep: str, suffix: str):
+        regex = rf"^{re.escape(stub)}{re.escape(sep)}{suffix}$"
+        return df.columns[df.columns.str.match(regex)]
+
+    def melt_stub(df, stub: str, i, j, value_vars, sep: str):
+        newdf = melt(
+            df,
+            id_vars=i,
+            value_vars=value_vars,
+            value_name=stub.rstrip(sep),
+            var_name=j,
+        )
+        newdf[j] = newdf[j].str.replace(re.escape(stub + sep), "", regex=True)
+
+        # GH17627 Cast numerics suffixes to int/float
+        try:
+            newdf[j] = to_numeric(newdf[j])
+        except (TypeError, ValueError, OverflowError):
+            # TODO: anything else to catch?
+            pass
+
+        return newdf.set_index(i + [j])
+
+    if not is_list_like(stubnames):
+        stubnames = [stubnames]
+    else:
+        stubnames = list(stubnames)
+
+    if df.columns.isin(stubnames).any():
+        raise ValueError("stubname can't be identical to a column name")
+
+    if not is_list_like(i):
+        i = [i]
+    else:
+        i = list(i)
+
+    if df[i].duplicated().any():
+        raise ValueError("the id variables need to uniquely identify each row")
+
+    _melted = []
+    value_vars_flattened = []
+    for stub in stubnames:
+        value_var = get_var_names(df, stub, sep, suffix)
+        value_vars_flattened.extend(value_var)
+        _melted.append(melt_stub(df, stub, i, j, value_var, sep))
+
+    melted = concat(_melted, axis=1)
+    id_vars = df.columns.difference(value_vars_flattened)
+    new = df[id_vars]
+
+    if len(i) == 1:
+        return new.set_index(i).join(melted)
+    else:
+        return new.merge(melted.reset_index(), on=i).set_index(i + [j])
diff --git a/pandas/pandas/core/reshape/merge.py b/pandas/pandas/core/reshape/merge.py
new file mode 100644
index 0000000000000..09be82c59a5c6
--- /dev/null
+++ b/pandas/pandas/core/reshape/merge.py
@@ -0,0 +1,3076 @@
+"""
+SQL-style merge routines
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Hashable,
+    Sequence,
+)
+import datetime
+from functools import partial
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+    final,
+)
+import uuid
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    Timedelta,
+    hashtable as libhashtable,
+    join as libjoin,
+    lib,
+)
+from pandas._libs.lib import is_range_indexer
+from pandas._typing import (
+    AnyArrayLike,
+    ArrayLike,
+    IndexLabel,
+    JoinHow,
+    MergeHow,
+    Shape,
+    Suffixes,
+    npt,
+)
+from pandas.errors import MergeError
+from pandas.util._decorators import (
+    cache_readonly,
+    set_module,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.cast import find_common_type
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    ensure_object,
+    is_bool,
+    is_bool_dtype,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_number,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_string_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+)
+
+from pandas import (
+    ArrowDtype,
+    Categorical,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas.core.algorithms as algos
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BaseMaskedArray,
+    ExtensionArray,
+)
+from pandas.core.arrays.string_ import StringDtype
+import pandas.core.common as com
+from pandas.core.construction import (
+    ensure_wrapped_if_datetimelike,
+    extract_array,
+)
+from pandas.core.indexes.api import default_index
+from pandas.core.sorting import (
+    get_group_index,
+    is_int64_overflow_possible,
+)
+
+if TYPE_CHECKING:
+    from pandas import DataFrame
+    from pandas.core import groupby
+    from pandas.core.arrays import DatetimeArray
+    from pandas.core.indexes.frozen import FrozenList
+
+_factorizers = {
+    np.int64: libhashtable.Int64Factorizer,
+    np.longlong: libhashtable.Int64Factorizer,
+    np.int32: libhashtable.Int32Factorizer,
+    np.int16: libhashtable.Int16Factorizer,
+    np.int8: libhashtable.Int8Factorizer,
+    np.uint64: libhashtable.UInt64Factorizer,
+    np.uint32: libhashtable.UInt32Factorizer,
+    np.uint16: libhashtable.UInt16Factorizer,
+    np.uint8: libhashtable.UInt8Factorizer,
+    np.bool_: libhashtable.UInt8Factorizer,
+    np.float64: libhashtable.Float64Factorizer,
+    np.float32: libhashtable.Float32Factorizer,
+    np.complex64: libhashtable.Complex64Factorizer,
+    np.complex128: libhashtable.Complex128Factorizer,
+    np.object_: libhashtable.ObjectFactorizer,
+}
+
+# See https://github.com/pandas-dev/pandas/issues/52451
+if np.intc is not np.int32:
+    if np.dtype(np.intc).itemsize == 4:
+        _factorizers[np.intc] = libhashtable.Int32Factorizer
+    else:
+        _factorizers[np.intc] = libhashtable.Int64Factorizer
+
+if np.uintc is not np.uint32:
+    if np.dtype(np.uintc).itemsize == 4:
+        _factorizers[np.uintc] = libhashtable.UInt32Factorizer
+    else:
+        _factorizers[np.uintc] = libhashtable.UInt64Factorizer
+
+
+_known = (np.ndarray, ExtensionArray, Index, ABCSeries)
+
+
+@set_module("pandas")
+def merge(
+    left: DataFrame | Series,
+    right: DataFrame | Series,
+    how: MergeHow = "inner",
+    on: IndexLabel | AnyArrayLike | None = None,
+    left_on: IndexLabel | AnyArrayLike | None = None,
+    right_on: IndexLabel | AnyArrayLike | None = None,
+    left_index: bool = False,
+    right_index: bool = False,
+    sort: bool = False,
+    suffixes: Suffixes = ("_x", "_y"),
+    copy: bool | lib.NoDefault = lib.no_default,
+    indicator: str | bool = False,
+    validate: str | None = None,
+) -> DataFrame:
+    """
+    Merge DataFrame or named Series objects with a database-style join.
+
+    A named Series object is treated as a DataFrame with a single named column.
+
+    The join is done on columns or indexes. If joining columns on
+    columns, the DataFrame indexes *will be ignored*. Otherwise if joining indexes
+    on indexes or indexes on a column or columns, the index will be passed on.
+    When performing a cross merge, no column specifications to merge on are
+    allowed.
+
+    .. warning::
+
+        If both key columns contain rows where the key is a null value, those
+        rows will be matched against each other. This is different from usual SQL
+        join behaviour and can lead to unexpected results.
+
+    Parameters
+    ----------
+    left : DataFrame or named Series
+        First pandas object to merge.
+    right : DataFrame or named Series
+        Second pandas object to merge.
+    how : {'left', 'right', 'outer', 'inner', 'cross', 'left_anti', 'right_anti},
+        default 'inner'
+        Type of merge to be performed.
+
+        * left: use only keys from left frame, similar to a SQL left outer join;
+          preserve key order.
+        * right: use only keys from right frame, similar to a SQL right outer join;
+          preserve key order.
+        * outer: use union of keys from both frames, similar to a SQL full outer
+          join; sort keys lexicographically.
+        * inner: use intersection of keys from both frames, similar to a SQL inner
+          join; preserve the order of the left keys.
+        * cross: creates the cartesian product from both frames, preserves the order
+          of the left keys.
+        * left_anti: use only keys from left frame that are not in right frame, similar
+          to SQL left anti join; preserve key order.
+        * right_anti: use only keys from right frame that are not in left frame, similar
+          to SQL right anti join; preserve key order.
+    on : label or list
+        Column or index level names to join on. These must be found in both
+        DataFrames. If `on` is None and not merging on indexes then this defaults
+        to the intersection of the columns in both DataFrames.
+    left_on : label or list, or array-like
+        Column or index level names to join on in the left DataFrame. Can also
+        be an array or list of arrays of the length of the left DataFrame.
+        These arrays are treated as if they are columns.
+    right_on : label or list, or array-like
+        Column or index level names to join on in the right DataFrame. Can also
+        be an array or list of arrays of the length of the right DataFrame.
+        These arrays are treated as if they are columns.
+    left_index : bool, default False
+        Use the index from the left DataFrame as the join key(s). If it is a
+        MultiIndex, the number of keys in the other DataFrame (either the index
+        or a number of columns) must match the number of levels.
+    right_index : bool, default False
+        Use the index from the right DataFrame as the join key. Same caveats as
+        left_index.
+    sort : bool, default False
+        Sort the join keys lexicographically in the result DataFrame. If False,
+        the order of the join keys depends on the join type (how keyword).
+    suffixes : list-like, default is ("_x", "_y")
+        A length-2 sequence where each element is optionally a string
+        indicating the suffix to add to overlapping column names in
+        `left` and `right` respectively. Pass a value of `None` instead
+        of a string to indicate that the column name from `left` or
+        `right` should be left as-is, with no suffix. At least one of the
+        values must not be None.
+    copy : bool, default False
+        If False, avoid copy if possible.
+
+        .. note::
+            The `copy` keyword will change behavior in pandas 3.0.
+            `Copy-on-Write
+            <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+            will be enabled by default, which means that all methods with a
+            `copy` keyword will use a lazy copy mechanism to defer the copy and
+            ignore the `copy` keyword. The `copy` keyword will be removed in a
+            future version of pandas.
+
+            You can already get the future behavior and improvements through
+            enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+        .. deprecated:: 3.0.0
+    indicator : bool or str, default False
+        If True, adds a column to the output DataFrame called "_merge" with
+        information on the source of each row. The column can be given a different
+        name by providing a string argument. The column will have a Categorical
+        type with the value of "left_only" for observations whose merge key only
+        appears in the left DataFrame, "right_only" for observations
+        whose merge key only appears in the right DataFrame, and "both"
+        if the observation's merge key is found in both DataFrames.
+
+    validate : str, optional
+        If specified, checks if merge is of specified type.
+
+        * "one_to_one" or "1:1": check if merge keys are unique in both
+          left and right datasets.
+        * "one_to_many" or "1:m": check if merge keys are unique in left
+          dataset.
+        * "many_to_one" or "m:1": check if merge keys are unique in right
+          dataset.
+        * "many_to_many" or "m:m": allowed, but does not result in checks.
+
+    Returns
+    -------
+    DataFrame
+        A DataFrame of the two merged objects.
+
+    See Also
+    --------
+    merge_ordered : Merge with optional filling/interpolation.
+    merge_asof : Merge on nearest keys.
+    DataFrame.join : Similar method using indices.
+
+    Examples
+    --------
+    >>> df1 = pd.DataFrame(
+    ...     {"lkey": ["foo", "bar", "baz", "foo"], "value": [1, 2, 3, 5]}
+    ... )
+    >>> df2 = pd.DataFrame(
+    ...     {"rkey": ["foo", "bar", "baz", "foo"], "value": [5, 6, 7, 8]}
+    ... )
+    >>> df1
+        lkey value
+    0   foo      1
+    1   bar      2
+    2   baz      3
+    3   foo      5
+    >>> df2
+        rkey value
+    0   foo      5
+    1   bar      6
+    2   baz      7
+    3   foo      8
+
+    Merge df1 and df2 on the lkey and rkey columns. The value columns have
+    the default suffixes, _x and _y, appended.
+
+    >>> df1.merge(df2, left_on="lkey", right_on="rkey")
+      lkey  value_x rkey  value_y
+    0  foo        1  foo        5
+    1  foo        1  foo        8
+    2  bar        2  bar        6
+    3  baz        3  baz        7
+    4  foo        5  foo        5
+    5  foo        5  foo        8
+
+    Merge DataFrames df1 and df2 with specified left and right suffixes
+    appended to any overlapping columns.
+
+    >>> df1.merge(df2, left_on="lkey", right_on="rkey", suffixes=("_left", "_right"))
+      lkey  value_left rkey  value_right
+    0  foo           1  foo            5
+    1  foo           1  foo            8
+    2  bar           2  bar            6
+    3  baz           3  baz            7
+    4  foo           5  foo            5
+    5  foo           5  foo            8
+
+    Merge DataFrames df1 and df2, but raise an exception if the DataFrames have
+    any overlapping columns.
+
+    >>> df1.merge(df2, left_on="lkey", right_on="rkey", suffixes=(False, False))
+    Traceback (most recent call last):
+    ...
+    ValueError: columns overlap but no suffix specified:
+        Index(['value'], dtype='object')
+
+    >>> df1 = pd.DataFrame({"a": ["foo", "bar"], "b": [1, 2]})
+    >>> df2 = pd.DataFrame({"a": ["foo", "baz"], "c": [3, 4]})
+    >>> df1
+          a  b
+    0   foo  1
+    1   bar  2
+    >>> df2
+          a  c
+    0   foo  3
+    1   baz  4
+
+    >>> df1.merge(df2, how="inner", on="a")
+          a  b  c
+    0   foo  1  3
+
+    >>> df1.merge(df2, how="left", on="a")
+          a  b  c
+    0   foo  1  3.0
+    1   bar  2  NaN
+
+    >>> df1 = pd.DataFrame({"left": ["foo", "bar"]})
+    >>> df2 = pd.DataFrame({"right": [7, 8]})
+    >>> df1
+        left
+    0   foo
+    1   bar
+    >>> df2
+        right
+    0   7
+    1   8
+
+    >>> df1.merge(df2, how="cross")
+       left  right
+    0   foo      7
+    1   foo      8
+    2   bar      7
+    3   bar      8
+    """
+    left_df = _validate_operand(left)
+    left._check_copy_deprecation(copy)
+    right_df = _validate_operand(right)
+    if how == "cross":
+        return _cross_merge(
+            left_df,
+            right_df,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            left_index=left_index,
+            right_index=right_index,
+            sort=sort,
+            suffixes=suffixes,
+            indicator=indicator,
+            validate=validate,
+        )
+    else:
+        op = _MergeOperation(
+            left_df,
+            right_df,
+            how=how,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            left_index=left_index,
+            right_index=right_index,
+            sort=sort,
+            suffixes=suffixes,
+            indicator=indicator,
+            validate=validate,
+        )
+        return op.get_result()
+
+
+def _cross_merge(
+    left: DataFrame,
+    right: DataFrame,
+    on: IndexLabel | AnyArrayLike | None = None,
+    left_on: IndexLabel | AnyArrayLike | None = None,
+    right_on: IndexLabel | AnyArrayLike | None = None,
+    left_index: bool = False,
+    right_index: bool = False,
+    sort: bool = False,
+    suffixes: Suffixes = ("_x", "_y"),
+    indicator: str | bool = False,
+    validate: str | None = None,
+) -> DataFrame:
+    """
+    See merge.__doc__ with how='cross'
+    """
+
+    if (
+        left_index
+        or right_index
+        or right_on is not None
+        or left_on is not None
+        or on is not None
+    ):
+        raise MergeError(
+            "Can not pass on, right_on, left_on or set right_index=True or "
+            "left_index=True"
+        )
+
+    cross_col = f"_cross_{uuid.uuid4()}"
+    left = left.assign(**{cross_col: 1})
+    right = right.assign(**{cross_col: 1})
+
+    left_on = right_on = [cross_col]
+
+    res = merge(
+        left,
+        right,
+        how="inner",
+        on=on,
+        left_on=left_on,
+        right_on=right_on,
+        left_index=left_index,
+        right_index=right_index,
+        sort=sort,
+        suffixes=suffixes,
+        indicator=indicator,
+        validate=validate,
+    )
+    del res[cross_col]
+    return res
+
+
+def _groupby_and_merge(
+    by, left: DataFrame | Series, right: DataFrame | Series, merge_pieces
+):
+    """
+    groupby & merge; we are always performing a left-by type operation
+
+    Parameters
+    ----------
+    by: field to group
+    left: DataFrame
+    right: DataFrame
+    merge_pieces: function for merging
+    """
+    pieces = []
+    if not isinstance(by, (list, tuple)):
+        by = [by]
+
+    lby = left.groupby(by, sort=False)
+    rby: groupby.DataFrameGroupBy | groupby.SeriesGroupBy | None = None
+
+    # if we can groupby the rhs
+    # then we can get vastly better perf
+    if all(item in right.columns for item in by):
+        rby = right.groupby(by, sort=False)
+
+    for key, lhs in lby._grouper.get_iterator(lby._selected_obj):
+        if rby is None:
+            rhs = right
+        else:
+            try:
+                rhs = right.take(rby.indices[key])
+            except KeyError:
+                # key doesn't exist in left
+                lcols = lhs.columns.tolist()
+                cols = lcols + [r for r in right.columns if r not in set(lcols)]
+                merged = lhs.reindex(columns=cols)
+                merged.index = range(len(merged))
+                pieces.append(merged)
+                continue
+
+        merged = merge_pieces(lhs, rhs)
+
+        # make sure join keys are in the merged
+        # TODO, should merge_pieces do this?
+        merged[by] = key
+
+        pieces.append(merged)
+
+    # preserve the original order
+    # if we have a missing piece this can be reset
+    from pandas.core.reshape.concat import concat
+
+    result = concat(pieces, ignore_index=True)
+    result = result.reindex(columns=pieces[0].columns)
+    return result, lby
+
+
+@set_module("pandas")
+def merge_ordered(
+    left: DataFrame | Series,
+    right: DataFrame | Series,
+    on: IndexLabel | None = None,
+    left_on: IndexLabel | None = None,
+    right_on: IndexLabel | None = None,
+    left_by=None,
+    right_by=None,
+    fill_method: str | None = None,
+    suffixes: Suffixes = ("_x", "_y"),
+    how: JoinHow = "outer",
+) -> DataFrame:
+    """
+    Perform a merge for ordered data with optional filling/interpolation.
+
+    Designed for ordered data like time series data. Optionally
+    perform group-wise merge (see examples).
+
+    Parameters
+    ----------
+    left : DataFrame or named Series
+        First pandas object to merge.
+    right : DataFrame or named Series
+        Second pandas object to merge.
+    on : label or list
+        Field names to join on. Must be found in both DataFrames.
+    left_on : label or list, or array-like
+        Field names to join on in left DataFrame. Can be a vector or list of
+        vectors of the length of the DataFrame to use a particular vector as
+        the join key instead of columns.
+    right_on : label or list, or array-like
+        Field names to join on in right DataFrame or vector/list of vectors per
+        left_on docs.
+    left_by : column name or list of column names
+        Group left DataFrame by group columns and merge piece by piece with
+        right DataFrame. Must be None if either left or right are a Series.
+    right_by : column name or list of column names
+        Group right DataFrame by group columns and merge piece by piece with
+        left DataFrame. Must be None if either left or right are a Series.
+    fill_method : {'ffill', None}, default None
+        Interpolation method for data.
+    suffixes : list-like, default is ("_x", "_y")
+        A length-2 sequence where each element is optionally a string
+        indicating the suffix to add to overlapping column names in
+        `left` and `right` respectively. Pass a value of `None` instead
+        of a string to indicate that the column name from `left` or
+        `right` should be left as-is, with no suffix. At least one of the
+        values must not be None.
+
+    how : {'left', 'right', 'outer', 'inner'}, default 'outer'
+        * left: use only keys from left frame (SQL: left outer join)
+        * right: use only keys from right frame (SQL: right outer join)
+        * outer: use union of keys from both frames (SQL: full outer join)
+        * inner: use intersection of keys from both frames (SQL: inner join).
+
+    Returns
+    -------
+    DataFrame
+        The merged DataFrame output type will be the same as
+        'left', if it is a subclass of DataFrame.
+
+    See Also
+    --------
+    merge : Merge with a database-style join.
+    merge_asof : Merge on nearest keys.
+
+    Examples
+    --------
+    >>> from pandas import merge_ordered
+    >>> df1 = pd.DataFrame(
+    ...     {
+    ...         "key": ["a", "c", "e", "a", "c", "e"],
+    ...         "lvalue": [1, 2, 3, 1, 2, 3],
+    ...         "group": ["a", "a", "a", "b", "b", "b"],
+    ...     }
+    ... )
+    >>> df1
+      key  lvalue group
+    0   a       1     a
+    1   c       2     a
+    2   e       3     a
+    3   a       1     b
+    4   c       2     b
+    5   e       3     b
+
+    >>> df2 = pd.DataFrame({"key": ["b", "c", "d"], "rvalue": [1, 2, 3]})
+    >>> df2
+      key  rvalue
+    0   b       1
+    1   c       2
+    2   d       3
+
+    >>> merge_ordered(df1, df2, fill_method="ffill", left_by="group")
+      key  lvalue group  rvalue
+    0   a       1     a     NaN
+    1   b       1     a     1.0
+    2   c       2     a     2.0
+    3   d       2     a     3.0
+    4   e       3     a     3.0
+    5   a       1     b     NaN
+    6   b       1     b     1.0
+    7   c       2     b     2.0
+    8   d       2     b     3.0
+    9   e       3     b     3.0
+    """
+
+    def _merger(x, y) -> DataFrame:
+        # perform the ordered merge operation
+        op = _OrderedMerge(
+            x,
+            y,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            suffixes=suffixes,
+            fill_method=fill_method,
+            how=how,
+        )
+        return op.get_result()
+
+    if left_by is not None and right_by is not None:
+        raise ValueError("Can only group either left or right frames")
+    if left_by is not None:
+        if isinstance(left_by, str):
+            left_by = [left_by]
+        check = set(left_by).difference(left.columns)
+        if len(check) != 0:
+            raise KeyError(f"{check} not found in left columns")
+        result, _ = _groupby_and_merge(left_by, left, right, lambda x, y: _merger(x, y))
+    elif right_by is not None:
+        if isinstance(right_by, str):
+            right_by = [right_by]
+        check = set(right_by).difference(right.columns)
+        if len(check) != 0:
+            raise KeyError(f"{check} not found in right columns")
+        result, _ = _groupby_and_merge(
+            right_by, right, left, lambda x, y: _merger(y, x)
+        )
+    else:
+        result = _merger(left, right)
+    return result
+
+
+@set_module("pandas")
+def merge_asof(
+    left: DataFrame | Series,
+    right: DataFrame | Series,
+    on: IndexLabel | None = None,
+    left_on: IndexLabel | None = None,
+    right_on: IndexLabel | None = None,
+    left_index: bool = False,
+    right_index: bool = False,
+    by=None,
+    left_by=None,
+    right_by=None,
+    suffixes: Suffixes = ("_x", "_y"),
+    tolerance: int | datetime.timedelta | None = None,
+    allow_exact_matches: bool = True,
+    direction: str = "backward",
+) -> DataFrame:
+    """
+    Perform a merge by key distance.
+
+    This is similar to a left-join except that we match on nearest
+    key rather than equal keys. Both DataFrames must be sorted by the key.
+
+    For each row in the left DataFrame:
+
+      - A "backward" search selects the last row in the right DataFrame whose
+        'on' key is less than or equal to the left's key.
+
+      - A "forward" search selects the first row in the right DataFrame whose
+        'on' key is greater than or equal to the left's key.
+
+      - A "nearest" search selects the row in the right DataFrame whose 'on'
+        key is closest in absolute distance to the left's key.
+
+    Optionally match on equivalent keys with 'by' before searching with 'on'.
+
+    Parameters
+    ----------
+    left : DataFrame or named Series
+        First pandas object to merge.
+    right : DataFrame or named Series
+        Second pandas object to merge.
+    on : label
+        Field name to join on. Must be found in both DataFrames.
+        The data MUST be ordered. Furthermore this must be a numeric column,
+        such as datetimelike, integer, or float. On or left_on/right_on
+        must be given.
+    left_on : label
+        Field name to join on in left DataFrame.
+    right_on : label
+        Field name to join on in right DataFrame.
+    left_index : bool
+        Use the index of the left DataFrame as the join key.
+    right_index : bool
+        Use the index of the right DataFrame as the join key.
+    by : column name or list of column names
+        Match on these columns before performing merge operation.
+    left_by : column name
+        Field names to match on in the left DataFrame.
+    right_by : column name
+        Field names to match on in the right DataFrame.
+    suffixes : 2-length sequence (tuple, list, ...)
+        Suffix to apply to overlapping column names in the left and right
+        side, respectively.
+    tolerance : int or timedelta, optional, default None
+        Select asof tolerance within this range; must be compatible
+        with the merge index.
+    allow_exact_matches : bool, default True
+
+        - If True, allow matching with the same 'on' value
+          (i.e. less-than-or-equal-to / greater-than-or-equal-to)
+        - If False, don't match the same 'on' value
+          (i.e., strictly less-than / strictly greater-than).
+
+    direction : 'backward' (default), 'forward', or 'nearest'
+        Whether to search for prior, subsequent, or closest matches.
+
+    Returns
+    -------
+    DataFrame
+        A DataFrame of the two merged objects.
+
+    See Also
+    --------
+    merge : Merge with a database-style join.
+    merge_ordered : Merge with optional filling/interpolation.
+
+    Examples
+    --------
+    >>> left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+    >>> left
+        a left_val
+    0   1        a
+    1   5        b
+    2  10        c
+
+    >>> right = pd.DataFrame({"a": [1, 2, 3, 6, 7], "right_val": [1, 2, 3, 6, 7]})
+    >>> right
+       a  right_val
+    0  1          1
+    1  2          2
+    2  3          3
+    3  6          6
+    4  7          7
+
+    >>> pd.merge_asof(left, right, on="a")
+        a left_val  right_val
+    0   1        a          1
+    1   5        b          3
+    2  10        c          7
+
+    >>> pd.merge_asof(left, right, on="a", allow_exact_matches=False)
+        a left_val  right_val
+    0   1        a        NaN
+    1   5        b        3.0
+    2  10        c        7.0
+
+    >>> pd.merge_asof(left, right, on="a", direction="forward")
+        a left_val  right_val
+    0   1        a        1.0
+    1   5        b        6.0
+    2  10        c        NaN
+
+    >>> pd.merge_asof(left, right, on="a", direction="nearest")
+        a left_val  right_val
+    0   1        a          1
+    1   5        b          6
+    2  10        c          7
+
+    We can use indexed DataFrames as well.
+
+    >>> left = pd.DataFrame({"left_val": ["a", "b", "c"]}, index=[1, 5, 10])
+    >>> left
+       left_val
+    1         a
+    5         b
+    10        c
+
+    >>> right = pd.DataFrame({"right_val": [1, 2, 3, 6, 7]}, index=[1, 2, 3, 6, 7])
+    >>> right
+       right_val
+    1          1
+    2          2
+    3          3
+    6          6
+    7          7
+
+    >>> pd.merge_asof(left, right, left_index=True, right_index=True)
+       left_val  right_val
+    1         a          1
+    5         b          3
+    10        c          7
+
+    Here is a real-world times-series example
+
+    >>> quotes = pd.DataFrame(
+    ...     {
+    ...         "time": [
+    ...             pd.Timestamp("2016-05-25 13:30:00.023"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.023"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.030"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.041"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.048"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.049"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.072"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.075"),
+    ...         ],
+    ...         "ticker": [
+    ...             "GOOG",
+    ...             "MSFT",
+    ...             "MSFT",
+    ...             "MSFT",
+    ...             "GOOG",
+    ...             "AAPL",
+    ...             "GOOG",
+    ...             "MSFT",
+    ...         ],
+    ...         "bid": [720.50, 51.95, 51.97, 51.99, 720.50, 97.99, 720.50, 52.01],
+    ...         "ask": [720.93, 51.96, 51.98, 52.00, 720.93, 98.01, 720.88, 52.03],
+    ...     }
+    ... )
+    >>> quotes
+                         time ticker     bid     ask
+    0 2016-05-25 13:30:00.023   GOOG  720.50  720.93
+    1 2016-05-25 13:30:00.023   MSFT   51.95   51.96
+    2 2016-05-25 13:30:00.030   MSFT   51.97   51.98
+    3 2016-05-25 13:30:00.041   MSFT   51.99   52.00
+    4 2016-05-25 13:30:00.048   GOOG  720.50  720.93
+    5 2016-05-25 13:30:00.049   AAPL   97.99   98.01
+    6 2016-05-25 13:30:00.072   GOOG  720.50  720.88
+    7 2016-05-25 13:30:00.075   MSFT   52.01   52.03
+
+    >>> trades = pd.DataFrame(
+    ...     {
+    ...         "time": [
+    ...             pd.Timestamp("2016-05-25 13:30:00.023"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.038"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.048"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.048"),
+    ...             pd.Timestamp("2016-05-25 13:30:00.048"),
+    ...         ],
+    ...         "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+    ...         "price": [51.95, 51.95, 720.77, 720.92, 98.0],
+    ...         "quantity": [75, 155, 100, 100, 100],
+    ...     }
+    ... )
+    >>> trades
+                         time ticker   price  quantity
+    0 2016-05-25 13:30:00.023   MSFT   51.95        75
+    1 2016-05-25 13:30:00.038   MSFT   51.95       155
+    2 2016-05-25 13:30:00.048   GOOG  720.77       100
+    3 2016-05-25 13:30:00.048   GOOG  720.92       100
+    4 2016-05-25 13:30:00.048   AAPL   98.00       100
+
+    By default we are taking the asof of the quotes
+
+    >>> pd.merge_asof(trades, quotes, on="time", by="ticker")
+                         time ticker   price  quantity     bid     ask
+    0 2016-05-25 13:30:00.023   MSFT   51.95        75   51.95   51.96
+    1 2016-05-25 13:30:00.038   MSFT   51.95       155   51.97   51.98
+    2 2016-05-25 13:30:00.048   GOOG  720.77       100  720.50  720.93
+    3 2016-05-25 13:30:00.048   GOOG  720.92       100  720.50  720.93
+    4 2016-05-25 13:30:00.048   AAPL   98.00       100     NaN     NaN
+
+    We only asof within 2ms between the quote time and the trade time
+
+    >>> pd.merge_asof(
+    ...     trades, quotes, on="time", by="ticker", tolerance=pd.Timedelta("2ms")
+    ... )
+                         time ticker   price  quantity     bid     ask
+    0 2016-05-25 13:30:00.023   MSFT   51.95        75   51.95   51.96
+    1 2016-05-25 13:30:00.038   MSFT   51.95       155     NaN     NaN
+    2 2016-05-25 13:30:00.048   GOOG  720.77       100  720.50  720.93
+    3 2016-05-25 13:30:00.048   GOOG  720.92       100  720.50  720.93
+    4 2016-05-25 13:30:00.048   AAPL   98.00       100     NaN     NaN
+
+    We only asof within 10ms between the quote time and the trade time
+    and we exclude exact matches on time. However *prior* data will
+    propagate forward
+
+    >>> pd.merge_asof(
+    ...     trades,
+    ...     quotes,
+    ...     on="time",
+    ...     by="ticker",
+    ...     tolerance=pd.Timedelta("10ms"),
+    ...     allow_exact_matches=False,
+    ... )
+                         time ticker   price  quantity     bid     ask
+    0 2016-05-25 13:30:00.023   MSFT   51.95        75     NaN     NaN
+    1 2016-05-25 13:30:00.038   MSFT   51.95       155   51.97   51.98
+    2 2016-05-25 13:30:00.048   GOOG  720.77       100     NaN     NaN
+    3 2016-05-25 13:30:00.048   GOOG  720.92       100     NaN     NaN
+    4 2016-05-25 13:30:00.048   AAPL   98.00       100     NaN     NaN
+    """
+    op = _AsOfMerge(
+        left,
+        right,
+        on=on,
+        left_on=left_on,
+        right_on=right_on,
+        left_index=left_index,
+        right_index=right_index,
+        by=by,
+        left_by=left_by,
+        right_by=right_by,
+        suffixes=suffixes,
+        how="asof",
+        tolerance=tolerance,
+        allow_exact_matches=allow_exact_matches,
+        direction=direction,
+    )
+    return op.get_result()
+
+
+# TODO: transformations??
+class _MergeOperation:
+    """
+    Perform a database (SQL) merge operation between two DataFrame or Series
+    objects using either columns as keys or their row indexes
+    """
+
+    _merge_type = "merge"
+    how: JoinHow | Literal["asof"]
+    on: IndexLabel | None
+    # left_on/right_on may be None when passed, but in validate_specification
+    #  get replaced with non-None.
+    left_on: Sequence[Hashable | AnyArrayLike]
+    right_on: Sequence[Hashable | AnyArrayLike]
+    left_index: bool
+    right_index: bool
+    sort: bool
+    suffixes: Suffixes
+    indicator: str | bool
+    validate: str | None
+    join_names: list[Hashable]
+    right_join_keys: list[ArrayLike]
+    left_join_keys: list[ArrayLike]
+
+    def __init__(
+        self,
+        left: DataFrame | Series,
+        right: DataFrame | Series,
+        how: JoinHow | Literal["left_anti", "right_anti", "asof"] = "inner",
+        on: IndexLabel | AnyArrayLike | None = None,
+        left_on: IndexLabel | AnyArrayLike | None = None,
+        right_on: IndexLabel | AnyArrayLike | None = None,
+        left_index: bool = False,
+        right_index: bool = False,
+        sort: bool = True,
+        suffixes: Suffixes = ("_x", "_y"),
+        indicator: str | bool = False,
+        validate: str | None = None,
+    ) -> None:
+        _left = _validate_operand(left)
+        _right = _validate_operand(right)
+        self.left = self.orig_left = _left
+        self.right = self.orig_right = _right
+        self.how, self.anti_join = self._validate_how(how)
+
+        self.on = com.maybe_make_list(on)
+
+        self.suffixes = suffixes
+        self.sort = sort or how == "outer"
+
+        self.left_index = left_index
+        self.right_index = right_index
+
+        self.indicator = indicator
+
+        if not is_bool(left_index):
+            raise ValueError(
+                f"left_index parameter must be of type bool, not {type(left_index)}"
+            )
+        if not is_bool(right_index):
+            raise ValueError(
+                f"right_index parameter must be of type bool, not {type(right_index)}"
+            )
+
+        # GH 40993: raise when merging between different levels; enforced in 2.0
+        if _left.columns.nlevels != _right.columns.nlevels:
+            msg = (
+                "Not allowed to merge between different levels. "
+                f"({_left.columns.nlevels} levels on the left, "
+                f"{_right.columns.nlevels} on the right)"
+            )
+            raise MergeError(msg)
+
+        self.left_on, self.right_on = self._validate_left_right_on(left_on, right_on)
+
+        (
+            self.left_join_keys,
+            self.right_join_keys,
+            self.join_names,
+            left_drop,
+            right_drop,
+        ) = self._get_merge_keys()
+
+        if left_drop:
+            self.left = self.left._drop_labels_or_levels(left_drop)
+
+        if right_drop:
+            self.right = self.right._drop_labels_or_levels(right_drop)
+
+        self._maybe_require_matching_dtypes(self.left_join_keys, self.right_join_keys)
+        self._validate_tolerance(self.left_join_keys)
+
+        # validate the merge keys dtypes. We may need to coerce
+        # to avoid incompatible dtypes
+        self._maybe_coerce_merge_keys()
+
+        # If argument passed to validate,
+        # check if columns specified as unique
+        # are in fact unique.
+        if validate is not None:
+            self._validate_validate_kwd(validate)
+
+    @final
+    def _validate_how(
+        self, how: JoinHow | Literal["left_anti", "right_anti", "asof"]
+    ) -> tuple[JoinHow | Literal["asof"], bool]:
+        """
+        Validate the 'how' parameter and return the actual join type and whether
+        this is an anti join.
+        """
+        # GH 59435: raise when "how" is not a valid Merge type
+        merge_type = {
+            "left",
+            "right",
+            "inner",
+            "outer",
+            "left_anti",
+            "right_anti",
+            "cross",
+            "asof",
+        }
+        if how not in merge_type:
+            raise ValueError(
+                f"'{how}' is not a valid Merge type: "
+                f"left, right, inner, outer, left_anti, right_anti, cross, asof"
+            )
+        anti_join = False
+        if how in {"left_anti", "right_anti"}:
+            how = how.split("_")[0]  # type: ignore[assignment]
+            anti_join = True
+        how = cast(JoinHow | Literal["asof"], how)
+        return how, anti_join
+
+    def _maybe_require_matching_dtypes(
+        self, left_join_keys: list[ArrayLike], right_join_keys: list[ArrayLike]
+    ) -> None:
+        # Overridden by AsOfMerge
+        pass
+
+    def _validate_tolerance(self, left_join_keys: list[ArrayLike]) -> None:
+        # Overridden by AsOfMerge
+        pass
+
+    @final
+    def _reindex_and_concat(
+        self,
+        join_index: Index,
+        left_indexer: npt.NDArray[np.intp] | None,
+        right_indexer: npt.NDArray[np.intp] | None,
+    ) -> DataFrame:
+        """
+        reindex along index and concat along columns.
+        """
+        # Take views so we do not alter the originals
+        left = self.left[:]
+        right = self.right[:]
+
+        llabels, rlabels = _items_overlap_with_suffix(
+            self.left._info_axis, self.right._info_axis, self.suffixes
+        )
+
+        if left_indexer is not None and not is_range_indexer(left_indexer, len(left)):
+            # Pinning the index here (and in the right code just below) is not
+            #  necessary, but makes the `.take` more performant if we have e.g.
+            #  a MultiIndex for left.index.
+            lmgr = left._mgr.reindex_indexer(
+                join_index,
+                left_indexer,
+                axis=1,
+                only_slice=True,
+                allow_dups=True,
+                use_na_proxy=True,
+            )
+            left = left._constructor_from_mgr(lmgr, axes=lmgr.axes)
+        left.index = join_index
+
+        if right_indexer is not None and not is_range_indexer(
+            right_indexer, len(right)
+        ):
+            rmgr = right._mgr.reindex_indexer(
+                join_index,
+                right_indexer,
+                axis=1,
+                only_slice=True,
+                allow_dups=True,
+                use_na_proxy=True,
+            )
+            right = right._constructor_from_mgr(rmgr, axes=rmgr.axes)
+        right.index = join_index
+
+        from pandas import concat
+
+        left.columns = llabels
+        right.columns = rlabels
+        result = concat([left, right], axis=1)
+        return result
+
+    def get_result(self) -> DataFrame:
+        if self.indicator:
+            self.left, self.right = self._indicator_pre_merge(self.left, self.right)
+
+        join_index, left_indexer, right_indexer = self._get_join_info()
+
+        result = self._reindex_and_concat(join_index, left_indexer, right_indexer)
+        result = result.__finalize__(self, method=self._merge_type)
+
+        if self.indicator:
+            result = self._indicator_post_merge(result)
+
+        self._maybe_add_join_keys(result, left_indexer, right_indexer)
+
+        self._maybe_restore_index_levels(result)
+
+        return result.__finalize__(self, method="merge")
+
+    @final
+    @cache_readonly
+    def _indicator_name(self) -> str | None:
+        if isinstance(self.indicator, str):
+            return self.indicator
+        elif isinstance(self.indicator, bool):
+            return "_merge" if self.indicator else None
+        else:
+            raise ValueError(
+                "indicator option can only accept boolean or string arguments"
+            )
+
+    @final
+    def _indicator_pre_merge(
+        self, left: DataFrame, right: DataFrame
+    ) -> tuple[DataFrame, DataFrame]:
+        columns = left.columns.union(right.columns)
+
+        for i in ["_left_indicator", "_right_indicator"]:
+            if i in columns:
+                raise ValueError(
+                    "Cannot use `indicator=True` option when "
+                    f"data contains a column named {i}"
+                )
+        if self._indicator_name in columns:
+            raise ValueError(
+                "Cannot use name of an existing column for indicator column"
+            )
+
+        left = left.copy()
+        right = right.copy()
+
+        left["_left_indicator"] = 1
+        left["_left_indicator"] = left["_left_indicator"].astype("int8")
+
+        right["_right_indicator"] = 2
+        right["_right_indicator"] = right["_right_indicator"].astype("int8")
+
+        return left, right
+
+    @final
+    def _indicator_post_merge(self, result: DataFrame) -> DataFrame:
+        result["_left_indicator"] = result["_left_indicator"].fillna(0)
+        result["_right_indicator"] = result["_right_indicator"].fillna(0)
+
+        result[self._indicator_name] = Categorical(
+            (result["_left_indicator"] + result["_right_indicator"]),
+            categories=[1, 2, 3],
+        )
+        result[self._indicator_name] = result[
+            self._indicator_name
+        ].cat.rename_categories(["left_only", "right_only", "both"])
+
+        result = result.drop(labels=["_left_indicator", "_right_indicator"], axis=1)
+        return result
+
+    @final
+    def _maybe_restore_index_levels(self, result: DataFrame) -> None:
+        """
+        Restore index levels specified as `on` parameters
+
+        Here we check for cases where `self.left_on` and `self.right_on` pairs
+        each reference an index level in their respective DataFrames. The
+        joined columns corresponding to these pairs are then restored to the
+        index of `result`.
+
+        **Note:** This method has side effects. It modifies `result` in-place
+
+        Parameters
+        ----------
+        result: DataFrame
+            merge result
+
+        Returns
+        -------
+        None
+        """
+        names_to_restore = []
+        for name, left_key, right_key in zip(
+            self.join_names, self.left_on, self.right_on
+        ):
+            if (
+                # Argument 1 to "_is_level_reference" of "NDFrame" has incompatible
+                # type "Union[Hashable, ExtensionArray, Index, Series]"; expected
+                # "Hashable"
+                self.orig_left._is_level_reference(left_key)  # type: ignore[arg-type]
+                # Argument 1 to "_is_level_reference" of "NDFrame" has incompatible
+                # type "Union[Hashable, ExtensionArray, Index, Series]"; expected
+                # "Hashable"
+                and self.orig_right._is_level_reference(
+                    right_key  # type: ignore[arg-type]
+                )
+                and left_key == right_key
+                and name not in result.index.names
+            ):
+                names_to_restore.append(name)
+
+        if names_to_restore:
+            result.set_index(names_to_restore, inplace=True)
+
+    @final
+    def _maybe_add_join_keys(
+        self,
+        result: DataFrame,
+        left_indexer: npt.NDArray[np.intp] | None,
+        right_indexer: npt.NDArray[np.intp] | None,
+    ) -> None:
+        left_has_missing = None
+        right_has_missing = None
+
+        assert all(isinstance(x, _known) for x in self.left_join_keys)
+
+        keys = zip(self.join_names, self.left_on, self.right_on)
+        for i, (name, lname, rname) in enumerate(keys):
+            if not _should_fill(lname, rname):
+                continue
+
+            take_left, take_right = None, None
+
+            if name in result:
+                if left_indexer is not None or right_indexer is not None:
+                    if name in self.left:
+                        if left_has_missing is None:
+                            left_has_missing = (
+                                False
+                                if left_indexer is None
+                                else (left_indexer == -1).any()
+                            )
+
+                        if left_has_missing:
+                            take_right = self.right_join_keys[i]
+
+                            if result[name].dtype != self.left[name].dtype:
+                                take_left = self.left[name]._values
+
+                    elif name in self.right:
+                        if right_has_missing is None:
+                            right_has_missing = (
+                                False
+                                if right_indexer is None
+                                else (right_indexer == -1).any()
+                            )
+
+                        if right_has_missing:
+                            take_left = self.left_join_keys[i]
+
+                            if result[name].dtype != self.right[name].dtype:
+                                take_right = self.right[name]._values
+
+            else:
+                take_left = self.left_join_keys[i]
+                take_right = self.right_join_keys[i]
+
+            if take_left is not None or take_right is not None:
+                if take_left is None:
+                    lvals = result[name]._values
+                elif left_indexer is None:
+                    lvals = take_left
+                else:
+                    # TODO: can we pin down take_left's type earlier?
+                    take_left = extract_array(take_left, extract_numpy=True)
+                    lfill = na_value_for_dtype(take_left.dtype)
+                    lvals = algos.take_nd(take_left, left_indexer, fill_value=lfill)
+
+                if take_right is None:
+                    rvals = result[name]._values
+                elif right_indexer is None:
+                    rvals = take_right
+                else:
+                    # TODO: can we pin down take_right's type earlier?
+                    taker = extract_array(take_right, extract_numpy=True)
+                    rfill = na_value_for_dtype(taker.dtype)
+                    rvals = algos.take_nd(taker, right_indexer, fill_value=rfill)
+
+                # if we have an all missing left_indexer
+                # make sure to just use the right values or vice-versa
+                if left_indexer is not None and (left_indexer == -1).all():
+                    key_col = Index(rvals)
+                    result_dtype = rvals.dtype
+                elif right_indexer is not None and (right_indexer == -1).all():
+                    key_col = Index(lvals)
+                    result_dtype = lvals.dtype
+                else:
+                    key_col = Index(lvals)
+                    if left_indexer is not None:
+                        mask_left = left_indexer == -1
+                        key_col = key_col.where(~mask_left, rvals)
+                    result_dtype = find_common_type([lvals.dtype, rvals.dtype])
+                    if (
+                        lvals.dtype.kind == "M"
+                        and rvals.dtype.kind == "M"
+                        and result_dtype.kind == "O"
+                    ):
+                        # TODO(non-nano) Workaround for common_type not dealing
+                        # with different resolutions
+                        result_dtype = key_col.dtype
+
+                if result._is_label_reference(name):
+                    result[name] = result._constructor_sliced(
+                        key_col, dtype=result_dtype, index=result.index
+                    )
+                elif result._is_level_reference(name):
+                    if isinstance(result.index, MultiIndex):
+                        key_col.name = name
+                        idx_list = [
+                            result.index.get_level_values(level_name)
+                            if level_name != name
+                            else key_col
+                            for level_name in result.index.names
+                        ]
+
+                        result.set_index(idx_list, inplace=True)
+                    else:
+                        result.index = Index(key_col, name=name)
+                else:
+                    result.insert(i, name or f"key_{i}", key_col)
+
+    def _get_join_indexers(
+        self,
+    ) -> tuple[npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        """return the join indexers"""
+        # make mypy happy
+        assert self.how != "asof"
+        return get_join_indexers(
+            self.left_join_keys, self.right_join_keys, sort=self.sort, how=self.how
+        )
+
+    @final
+    def _get_join_info(
+        self,
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        left_ax = self.left.index
+        right_ax = self.right.index
+
+        if self.left_index and self.right_index and self.how != "asof":
+            join_index, left_indexer, right_indexer = left_ax.join(
+                right_ax, how=self.how, return_indexers=True, sort=self.sort
+            )
+
+        elif self.right_index and self.how == "left":
+            join_index, left_indexer, right_indexer = _left_join_on_index(
+                left_ax, right_ax, self.left_join_keys, sort=self.sort
+            )
+
+        elif self.left_index and self.how == "right":
+            join_index, right_indexer, left_indexer = _left_join_on_index(
+                right_ax, left_ax, self.right_join_keys, sort=self.sort
+            )
+        else:
+            (left_indexer, right_indexer) = self._get_join_indexers()
+
+            if self.right_index:
+                if len(self.left) > 0:
+                    join_index = self._create_join_index(
+                        left_ax,
+                        right_ax,
+                        left_indexer,
+                        how="right",
+                    )
+                elif right_indexer is None:
+                    join_index = right_ax.copy()
+                else:
+                    join_index = right_ax.take(right_indexer)
+            elif self.left_index:
+                if self.how == "asof":
+                    # GH#33463 asof should always behave like a left merge
+                    join_index = self._create_join_index(
+                        left_ax,
+                        right_ax,
+                        left_indexer,
+                        how="left",
+                    )
+
+                elif len(self.right) > 0:
+                    join_index = self._create_join_index(
+                        right_ax,
+                        left_ax,
+                        right_indexer,
+                        how="left",
+                    )
+                elif left_indexer is None:
+                    join_index = left_ax.copy()
+                else:
+                    join_index = left_ax.take(left_indexer)
+            else:
+                n = len(left_ax) if left_indexer is None else len(left_indexer)
+                join_index = default_index(n)
+
+        if self.anti_join:
+            join_index, left_indexer, right_indexer = self._handle_anti_join(
+                join_index, left_indexer, right_indexer
+            )
+
+        return join_index, left_indexer, right_indexer
+
+    @final
+    def _create_join_index(
+        self,
+        index: Index,
+        other_index: Index,
+        indexer: npt.NDArray[np.intp] | None,
+        how: JoinHow = "left",
+    ) -> Index:
+        """
+        Create a join index by rearranging one index to match another
+
+        Parameters
+        ----------
+        index : Index
+            index being rearranged
+        other_index : Index
+            used to supply values not found in index
+        indexer : np.ndarray[np.intp] or None
+            how to rearrange index
+        how : str
+            Replacement is only necessary if indexer based on other_index.
+
+        Returns
+        -------
+        Index
+        """
+        if self.how in (how, "outer") and not isinstance(other_index, MultiIndex):
+            # if final index requires values in other_index but not target
+            # index, indexer may hold missing (-1) values, causing Index.take
+            # to take the final value in target index. So, we set the last
+            # element to be the desired fill value. We do not use allow_fill
+            # and fill_value because it throws a ValueError on integer indices
+            mask = indexer == -1
+            if np.any(mask):
+                fill_value = na_value_for_dtype(index.dtype, compat=False)
+                index = index.append(Index([fill_value]))
+        if indexer is None:
+            return index.copy()
+        return index.take(indexer)
+
+    @final
+    def _handle_anti_join(
+        self,
+        join_index: Index,
+        left_indexer: npt.NDArray[np.intp] | None,
+        right_indexer: npt.NDArray[np.intp] | None,
+    ) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+        """
+        Handle anti join by returning the correct join index and indexers
+
+        Parameters
+        ----------
+        join_index : Index
+            join index
+        left_indexer : np.ndarray[np.intp] or None
+            left indexer
+        right_indexer : np.ndarray[np.intp] or None
+            right indexer
+
+        Returns
+        -------
+        Index, np.ndarray[np.intp] or None, np.ndarray[np.intp] or None
+        """
+        # Make sure indexers are not None
+        if left_indexer is None:
+            left_indexer = np.arange(len(self.left))
+        if right_indexer is None:
+            right_indexer = np.arange(len(self.right))
+
+        assert self.how in {"left", "right"}
+        if self.how == "left":
+            # Filter to rows where left keys are not in right keys
+            filt = right_indexer == -1
+        else:
+            # Filter to rows where right keys are not in left keys
+            filt = left_indexer == -1
+        join_index = join_index[filt]
+        left_indexer = left_indexer[filt]
+        right_indexer = right_indexer[filt]
+
+        return join_index, left_indexer, right_indexer
+
+    @final
+    def _get_merge_keys(
+        self,
+    ) -> tuple[
+        list[ArrayLike],
+        list[ArrayLike],
+        list[Hashable],
+        list[Hashable],
+        list[Hashable],
+    ]:
+        """
+        Returns
+        -------
+        left_keys, right_keys, join_names, left_drop, right_drop
+        """
+        left_keys: list[ArrayLike] = []
+        right_keys: list[ArrayLike] = []
+        join_names: list[Hashable] = []
+        right_drop: list[Hashable] = []
+        left_drop: list[Hashable] = []
+
+        left, right = self.left, self.right
+
+        is_lkey = lambda x: isinstance(x, _known) and len(x) == len(left)
+        is_rkey = lambda x: isinstance(x, _known) and len(x) == len(right)
+
+        # Note that pd.merge_asof() has separate 'on' and 'by' parameters. A
+        # user could, for example, request 'left_index' and 'left_by'. In a
+        # regular pd.merge(), users cannot specify both 'left_index' and
+        # 'left_on'. (Instead, users have a MultiIndex). That means the
+        # self.left_on in this function is always empty in a pd.merge(), but
+        # a pd.merge_asof(left_index=True, left_by=...) will result in a
+        # self.left_on array with a None in the middle of it. This requires
+        # a work-around as designated in the code below.
+        # See _validate_left_right_on() for where this happens.
+
+        # ugh, spaghetti re #733
+        if _any(self.left_on) and _any(self.right_on):
+            for lk, rk in zip(self.left_on, self.right_on):
+                lk = extract_array(lk, extract_numpy=True)
+                rk = extract_array(rk, extract_numpy=True)
+                if is_lkey(lk):
+                    lk = cast(ArrayLike, lk)
+                    left_keys.append(lk)
+                    if is_rkey(rk):
+                        rk = cast(ArrayLike, rk)
+                        right_keys.append(rk)
+                        join_names.append(None)  # what to do?
+                    else:
+                        # Then we're either Hashable or a wrong-length arraylike,
+                        #  the latter of which will raise
+                        rk = cast(Hashable, rk)
+                        if rk is not None:
+                            right_keys.append(right._get_label_or_level_values(rk))
+                            join_names.append(rk)
+                        else:
+                            # work-around for merge_asof(right_index=True)
+                            right_keys.append(right.index._values)
+                            join_names.append(right.index.name)
+                else:
+                    if not is_rkey(rk):
+                        # Then we're either Hashable or a wrong-length arraylike,
+                        #  the latter of which will raise
+                        rk = cast(Hashable, rk)
+                        if rk is not None:
+                            right_keys.append(right._get_label_or_level_values(rk))
+                        else:
+                            # work-around for merge_asof(right_index=True)
+                            right_keys.append(right.index._values)
+                        if lk is not None and lk == rk:  # FIXME: what about other NAs?
+                            right_drop.append(rk)
+                    else:
+                        rk = cast(ArrayLike, rk)
+                        right_keys.append(rk)
+                    if lk is not None:
+                        # Then we're either Hashable or a wrong-length arraylike,
+                        #  the latter of which will raise
+                        lk = cast(Hashable, lk)
+                        left_keys.append(left._get_label_or_level_values(lk))
+                        join_names.append(lk)
+                    else:
+                        # work-around for merge_asof(left_index=True)
+                        left_keys.append(left.index._values)
+                        join_names.append(left.index.name)
+        elif _any(self.left_on):
+            for k in self.left_on:
+                if is_lkey(k):
+                    k = extract_array(k, extract_numpy=True)
+                    k = cast(ArrayLike, k)
+                    left_keys.append(k)
+                    join_names.append(None)
+                else:
+                    # Then we're either Hashable or a wrong-length arraylike,
+                    #  the latter of which will raise
+                    k = cast(Hashable, k)
+                    left_keys.append(left._get_label_or_level_values(k))
+                    join_names.append(k)
+            if isinstance(self.right.index, MultiIndex):
+                right_keys = [
+                    lev._values.take(lev_codes)
+                    for lev, lev_codes in zip(
+                        self.right.index.levels, self.right.index.codes
+                    )
+                ]
+            else:
+                right_keys = [self.right.index._values]
+        elif _any(self.right_on):
+            for k in self.right_on:
+                k = extract_array(k, extract_numpy=True)
+                if is_rkey(k):
+                    k = cast(ArrayLike, k)
+                    right_keys.append(k)
+                    join_names.append(None)
+                else:
+                    # Then we're either Hashable or a wrong-length arraylike,
+                    #  the latter of which will raise
+                    k = cast(Hashable, k)
+                    right_keys.append(right._get_label_or_level_values(k))
+                    join_names.append(k)
+            if isinstance(self.left.index, MultiIndex):
+                left_keys = [
+                    lev._values.take(lev_codes)
+                    for lev, lev_codes in zip(
+                        self.left.index.levels, self.left.index.codes
+                    )
+                ]
+            else:
+                left_keys = [self.left.index._values]
+
+        return left_keys, right_keys, join_names, left_drop, right_drop
+
+    @final
+    def _maybe_coerce_merge_keys(self) -> None:
+        # we have valid merges but we may have to further
+        # coerce these if they are originally incompatible types
+        #
+        # for example if these are categorical, but are not dtype_equal
+        # or if we have object and integer dtypes
+
+        for lk, rk, name in zip(
+            self.left_join_keys, self.right_join_keys, self.join_names
+        ):
+            if (len(lk) and not len(rk)) or (not len(lk) and len(rk)):
+                continue
+
+            lk = extract_array(lk, extract_numpy=True)
+            rk = extract_array(rk, extract_numpy=True)
+
+            lk_is_cat = isinstance(lk.dtype, CategoricalDtype)
+            rk_is_cat = isinstance(rk.dtype, CategoricalDtype)
+            lk_is_object_or_string = is_object_dtype(lk.dtype) or is_string_dtype(
+                lk.dtype
+            )
+            rk_is_object_or_string = is_object_dtype(rk.dtype) or is_string_dtype(
+                rk.dtype
+            )
+
+            # if either left or right is a categorical
+            # then the must match exactly in categories & ordered
+            if lk_is_cat and rk_is_cat:
+                lk = cast(Categorical, lk)
+                rk = cast(Categorical, rk)
+                if lk._categories_match_up_to_permutation(rk):
+                    continue
+
+            elif lk_is_cat or rk_is_cat:
+                pass
+
+            elif lk.dtype == rk.dtype:
+                continue
+
+            msg = (
+                f"You are trying to merge on {lk.dtype} and {rk.dtype} columns "
+                f"for key '{name}'. If you wish to proceed you should use pd.concat"
+            )
+
+            # if we are numeric, then allow differing
+            # kinds to proceed, eg. int64 and int8, int and float
+            # further if we are object, but we infer to
+            # the same, then proceed
+            if is_numeric_dtype(lk.dtype) and is_numeric_dtype(rk.dtype):
+                if lk.dtype.kind == rk.dtype.kind:
+                    continue
+
+                if isinstance(lk.dtype, ExtensionDtype) and not isinstance(
+                    rk.dtype, ExtensionDtype
+                ):
+                    ct = find_common_type([lk.dtype, rk.dtype])
+                    if isinstance(ct, ExtensionDtype):
+                        com_cls = ct.construct_array_type()
+                        rk = com_cls._from_sequence(rk, dtype=ct, copy=False)
+                    else:
+                        rk = rk.astype(ct)
+                elif isinstance(rk.dtype, ExtensionDtype):
+                    ct = find_common_type([lk.dtype, rk.dtype])
+                    if isinstance(ct, ExtensionDtype):
+                        com_cls = ct.construct_array_type()
+                        lk = com_cls._from_sequence(lk, dtype=ct, copy=False)
+                    else:
+                        lk = lk.astype(ct)
+
+                # check whether ints and floats
+                if is_integer_dtype(rk.dtype) and is_float_dtype(lk.dtype):
+                    # GH 47391 numpy > 1.24 will raise a RuntimeError for nan -> int
+                    with np.errstate(invalid="ignore"):
+                        # error: Argument 1 to "astype" of "ndarray" has incompatible
+                        # type "Union[ExtensionDtype, Any, dtype[Any]]"; expected
+                        # "Union[dtype[Any], Type[Any], _SupportsDType[dtype[Any]]]"
+                        casted = lk.astype(rk.dtype)  # type: ignore[arg-type]
+
+                    mask = ~np.isnan(lk)
+                    match = lk == casted
+                    if not match[mask].all():
+                        warnings.warn(
+                            "You are merging on int and float "
+                            "columns where the float values "
+                            "are not equal to their int representation.",
+                            UserWarning,
+                            stacklevel=find_stack_level(),
+                        )
+                    continue
+
+                if is_float_dtype(rk.dtype) and is_integer_dtype(lk.dtype):
+                    # GH 47391 numpy > 1.24 will raise a RuntimeError for nan -> int
+                    with np.errstate(invalid="ignore"):
+                        # error: Argument 1 to "astype" of "ndarray" has incompatible
+                        # type "Union[ExtensionDtype, Any, dtype[Any]]"; expected
+                        # "Union[dtype[Any], Type[Any], _SupportsDType[dtype[Any]]]"
+                        casted = rk.astype(lk.dtype)  # type: ignore[arg-type]
+
+                    mask = ~np.isnan(rk)
+                    match = rk == casted
+                    if not match[mask].all():
+                        warnings.warn(
+                            "You are merging on int and float "
+                            "columns where the float values "
+                            "are not equal to their int representation.",
+                            UserWarning,
+                            stacklevel=find_stack_level(),
+                        )
+                    continue
+
+                # let's infer and see if we are ok
+                if lib.infer_dtype(lk, skipna=False) == lib.infer_dtype(
+                    rk, skipna=False
+                ):
+                    continue
+
+            # Check if we are trying to merge on obviously
+            # incompatible dtypes GH 9780, GH 15800
+
+            # bool values are coerced to object
+            elif (lk_is_object_or_string and is_bool_dtype(rk.dtype)) or (
+                is_bool_dtype(lk.dtype) and rk_is_object_or_string
+            ):
+                pass
+
+            # object values are allowed to be merged
+            elif (lk_is_object_or_string and is_numeric_dtype(rk.dtype)) or (
+                is_numeric_dtype(lk.dtype) and rk_is_object_or_string
+            ):
+                inferred_left = lib.infer_dtype(lk, skipna=False)
+                inferred_right = lib.infer_dtype(rk, skipna=False)
+                bool_types = ["integer", "mixed-integer", "boolean", "empty"]
+                string_types = ["string", "unicode", "mixed", "bytes", "empty"]
+
+                # inferred bool
+                if inferred_left in bool_types and inferred_right in bool_types:
+                    pass
+
+                # unless we are merging non-string-like with string-like
+                elif (
+                    inferred_left in string_types and inferred_right not in string_types
+                ) or (
+                    inferred_right in string_types and inferred_left not in string_types
+                ):
+                    raise ValueError(msg)
+
+            # datetimelikes must match exactly
+            elif needs_i8_conversion(lk.dtype) and not needs_i8_conversion(rk.dtype):
+                raise ValueError(msg)
+            elif not needs_i8_conversion(lk.dtype) and needs_i8_conversion(rk.dtype):
+                raise ValueError(msg)
+            elif isinstance(lk.dtype, DatetimeTZDtype) and not isinstance(
+                rk.dtype, DatetimeTZDtype
+            ):
+                raise ValueError(msg)
+            elif not isinstance(lk.dtype, DatetimeTZDtype) and isinstance(
+                rk.dtype, DatetimeTZDtype
+            ):
+                raise ValueError(msg)
+            elif (
+                isinstance(lk.dtype, DatetimeTZDtype)
+                and isinstance(rk.dtype, DatetimeTZDtype)
+            ) or (lk.dtype.kind == "M" and rk.dtype.kind == "M"):
+                # allows datetime with different resolutions
+                continue
+            # datetime and timedelta not allowed
+            elif lk.dtype.kind == "M" and rk.dtype.kind == "m":
+                raise ValueError(msg)
+            elif lk.dtype.kind == "m" and rk.dtype.kind == "M":
+                raise ValueError(msg)
+
+            elif is_object_dtype(lk.dtype) and is_object_dtype(rk.dtype):
+                continue
+
+            # Houston, we have a problem!
+            # let's coerce to object if the dtypes aren't
+            # categorical, otherwise coerce to the category
+            # dtype. If we coerced categories to object,
+            # then we would lose type information on some
+            # columns, and end up trying to merge
+            # incompatible dtypes. See GH 16900.
+            if name in self.left.columns:
+                typ = cast(Categorical, lk).categories.dtype if lk_is_cat else object
+                self.left = self.left.copy()
+                self.left[name] = self.left[name].astype(typ)
+            if name in self.right.columns:
+                typ = cast(Categorical, rk).categories.dtype if rk_is_cat else object
+                self.right = self.right.copy()
+                self.right[name] = self.right[name].astype(typ)
+
+    def _validate_left_right_on(self, left_on, right_on):
+        left_on = com.maybe_make_list(left_on)
+        right_on = com.maybe_make_list(right_on)
+
+        # Hm, any way to make this logic less complicated??
+        if self.on is None and left_on is None and right_on is None:
+            if self.left_index and self.right_index:
+                left_on, right_on = (), ()
+            elif self.left_index:
+                raise MergeError("Must pass right_on or right_index=True")
+            elif self.right_index:
+                raise MergeError("Must pass left_on or left_index=True")
+            else:
+                # use the common columns
+                left_cols = self.left.columns
+                right_cols = self.right.columns
+                common_cols = left_cols.intersection(right_cols)
+                if len(common_cols) == 0:
+                    raise MergeError(
+                        "No common columns to perform merge on. "
+                        f"Merge options: left_on={left_on}, "
+                        f"right_on={right_on}, "
+                        f"left_index={self.left_index}, "
+                        f"right_index={self.right_index}"
+                    )
+                if (
+                    not left_cols.join(common_cols, how="inner").is_unique
+                    or not right_cols.join(common_cols, how="inner").is_unique
+                ):
+                    raise MergeError(f"Data columns not unique: {common_cols!r}")
+                left_on = right_on = common_cols
+        elif self.on is not None:
+            if left_on is not None or right_on is not None:
+                raise MergeError(
+                    'Can only pass argument "on" OR "left_on" '
+                    'and "right_on", not a combination of both.'
+                )
+            if self.left_index or self.right_index:
+                raise MergeError(
+                    'Can only pass argument "on" OR "left_index" '
+                    'and "right_index", not a combination of both.'
+                )
+            left_on = right_on = self.on
+        elif left_on is not None:
+            if self.left_index:
+                raise MergeError(
+                    'Can only pass argument "left_on" OR "left_index" not both.'
+                )
+            if not self.right_index and right_on is None:
+                raise MergeError('Must pass "right_on" OR "right_index".')
+            n = len(left_on)
+            if self.right_index:
+                if len(left_on) != self.right.index.nlevels:
+                    raise ValueError(
+                        "len(left_on) must equal the number "
+                        'of levels in the index of "right"'
+                    )
+                right_on = [None] * n
+        elif right_on is not None:
+            if self.right_index:
+                raise MergeError(
+                    'Can only pass argument "right_on" OR "right_index" not both.'
+                )
+            if not self.left_index and left_on is None:
+                raise MergeError('Must pass "left_on" OR "left_index".')
+            n = len(right_on)
+            if self.left_index:
+                if len(right_on) != self.left.index.nlevels:
+                    raise ValueError(
+                        "len(right_on) must equal the number "
+                        'of levels in the index of "left"'
+                    )
+                left_on = [None] * n
+        if len(right_on) != len(left_on):
+            raise ValueError("len(right_on) must equal len(left_on)")
+
+        return left_on, right_on
+
+    @final
+    def _validate_validate_kwd(self, validate: str) -> None:
+        # Check uniqueness of each
+        if self.left_index:
+            left_unique = self.orig_left.index.is_unique
+        else:
+            left_unique = MultiIndex.from_arrays(self.left_join_keys).is_unique
+
+        if self.right_index:
+            right_unique = self.orig_right.index.is_unique
+        else:
+            right_unique = MultiIndex.from_arrays(self.right_join_keys).is_unique
+
+        # Check data integrity
+        if validate in ["one_to_one", "1:1"]:
+            if not left_unique and not right_unique:
+                raise MergeError(
+                    "Merge keys are not unique in either left "
+                    "or right dataset; not a one-to-one merge"
+                )
+            if not left_unique:
+                raise MergeError(
+                    "Merge keys are not unique in left dataset; not a one-to-one merge"
+                )
+            if not right_unique:
+                raise MergeError(
+                    "Merge keys are not unique in right dataset; not a one-to-one merge"
+                )
+
+        elif validate in ["one_to_many", "1:m"]:
+            if not left_unique:
+                raise MergeError(
+                    "Merge keys are not unique in left dataset; not a one-to-many merge"
+                )
+
+        elif validate in ["many_to_one", "m:1"]:
+            if not right_unique:
+                raise MergeError(
+                    "Merge keys are not unique in right dataset; "
+                    "not a many-to-one merge"
+                )
+
+        elif validate in ["many_to_many", "m:m"]:
+            pass
+
+        else:
+            raise ValueError(
+                f'"{validate}" is not a valid argument. '
+                "Valid arguments are:\n"
+                '- "1:1"\n'
+                '- "1:m"\n'
+                '- "m:1"\n'
+                '- "m:m"\n'
+                '- "one_to_one"\n'
+                '- "one_to_many"\n'
+                '- "many_to_one"\n'
+                '- "many_to_many"'
+            )
+
+
+def get_join_indexers(
+    left_keys: list[ArrayLike],
+    right_keys: list[ArrayLike],
+    sort: bool = False,
+    how: JoinHow = "inner",
+) -> tuple[npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
+    """
+
+    Parameters
+    ----------
+    left_keys : list[ndarray, ExtensionArray, Index, Series]
+    right_keys : list[ndarray, ExtensionArray, Index, Series]
+    sort : bool, default False
+    how : {'inner', 'outer', 'left', 'right'}, default 'inner'
+
+    Returns
+    -------
+    np.ndarray[np.intp] or None
+        Indexer into the left_keys.
+    np.ndarray[np.intp] or None
+        Indexer into the right_keys.
+    """
+    assert len(left_keys) == len(right_keys), (
+        "left_keys and right_keys must be the same length"
+    )
+
+    # fast-path for empty left/right
+    left_n = len(left_keys[0])
+    right_n = len(right_keys[0])
+    if left_n == 0:
+        if how in ["left", "inner"]:
+            return _get_empty_indexer()
+        elif not sort and how in ["right", "outer"]:
+            return _get_no_sort_one_missing_indexer(right_n, True)
+    elif right_n == 0:
+        if how in ["right", "inner"]:
+            return _get_empty_indexer()
+        elif not sort and how in ["left", "outer"]:
+            return _get_no_sort_one_missing_indexer(left_n, False)
+
+    lkey: ArrayLike
+    rkey: ArrayLike
+    if len(left_keys) > 1:
+        # get left & right join labels and num. of levels at each location
+        mapped = (
+            _factorize_keys(left_keys[n], right_keys[n], sort=sort)
+            for n in range(len(left_keys))
+        )
+        zipped = zip(*mapped)
+        llab, rlab, shape = (list(x) for x in zipped)
+
+        # get flat i8 keys from label lists
+        lkey, rkey = _get_join_keys(llab, rlab, tuple(shape), sort)
+    else:
+        lkey = left_keys[0]
+        rkey = right_keys[0]
+
+    left = Index(lkey)
+    right = Index(rkey)
+
+    if (
+        left.is_monotonic_increasing
+        and right.is_monotonic_increasing
+        and (left.is_unique or right.is_unique)
+    ):
+        _, lidx, ridx = left.join(right, how=how, return_indexers=True, sort=sort)
+    else:
+        lidx, ridx = get_join_indexers_non_unique(
+            left._values, right._values, sort, how
+        )
+
+    if lidx is not None and is_range_indexer(lidx, len(left)):
+        lidx = None
+    if ridx is not None and is_range_indexer(ridx, len(right)):
+        ridx = None
+    return lidx, ridx
+
+
+def get_join_indexers_non_unique(
+    left: ArrayLike,
+    right: ArrayLike,
+    sort: bool = False,
+    how: JoinHow = "inner",
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+    """
+    Get join indexers for left and right.
+
+    Parameters
+    ----------
+    left : ArrayLike
+    right : ArrayLike
+    sort : bool, default False
+    how : {'inner', 'outer', 'left', 'right'}, default 'inner'
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+        Indexer into left.
+    np.ndarray[np.intp]
+        Indexer into right.
+    """
+    lkey, rkey, count = _factorize_keys(left, right, sort=sort, how=how)
+    if count == -1:
+        # hash join
+        return lkey, rkey
+    if how == "left":
+        lidx, ridx = libjoin.left_outer_join(lkey, rkey, count, sort=sort)
+    elif how == "right":
+        ridx, lidx = libjoin.left_outer_join(rkey, lkey, count, sort=sort)
+    elif how == "inner":
+        lidx, ridx = libjoin.inner_join(lkey, rkey, count, sort=sort)
+    elif how == "outer":
+        lidx, ridx = libjoin.full_outer_join(lkey, rkey, count)
+    return lidx, ridx
+
+
+def restore_dropped_levels_multijoin(
+    left: MultiIndex,
+    right: MultiIndex,
+    dropped_level_names,
+    join_index: Index,
+    lindexer: npt.NDArray[np.intp],
+    rindexer: npt.NDArray[np.intp],
+) -> tuple[FrozenList, FrozenList, FrozenList]:
+    """
+    *this is an internal non-public method*
+
+    Returns the levels, labels and names of a multi-index to multi-index join.
+    Depending on the type of join, this method restores the appropriate
+    dropped levels of the joined multi-index.
+    The method relies on lindexer, rindexer which hold the index positions of
+    left and right, where a join was feasible
+
+    Parameters
+    ----------
+    left : MultiIndex
+        left index
+    right : MultiIndex
+        right index
+    dropped_level_names : str array
+        list of non-common level names
+    join_index : Index
+        the index of the join between the
+        common levels of left and right
+    lindexer : np.ndarray[np.intp]
+        left indexer
+    rindexer : np.ndarray[np.intp]
+        right indexer
+
+    Returns
+    -------
+    levels : list of Index
+        levels of combined multiindexes
+    labels : np.ndarray[np.intp]
+        labels of combined multiindexes
+    names : List[Hashable]
+        names of combined multiindex levels
+
+    """
+
+    def _convert_to_multiindex(index: Index) -> MultiIndex:
+        if isinstance(index, MultiIndex):
+            return index
+        else:
+            return MultiIndex.from_arrays([index._values], names=[index.name])
+
+    # For multi-multi joins with one overlapping level,
+    # the returned index if of type Index
+    # Assure that join_index is of type MultiIndex
+    # so that dropped levels can be appended
+    join_index = _convert_to_multiindex(join_index)
+
+    join_levels = join_index.levels
+    join_codes = join_index.codes
+    join_names = join_index.names
+
+    # Iterate through the levels that must be restored
+    for dropped_level_name in dropped_level_names:
+        if dropped_level_name in left.names:
+            idx = left
+            indexer = lindexer
+        else:
+            idx = right
+            indexer = rindexer
+
+        # The index of the level name to be restored
+        name_idx = idx.names.index(dropped_level_name)
+
+        restore_levels = idx.levels[name_idx]
+        # Inject -1 in the codes list where a join was not possible
+        # IOW indexer[i]=-1
+        codes = idx.codes[name_idx]
+        if indexer is None:
+            restore_codes = codes
+        else:
+            restore_codes = algos.take_nd(codes, indexer, fill_value=-1)
+
+        # error: Cannot determine type of "__add__"
+        join_levels = join_levels + [restore_levels]  # type: ignore[has-type]
+        join_codes = join_codes + [restore_codes]  # type: ignore[has-type]
+        join_names = join_names + [dropped_level_name]
+
+    return join_levels, join_codes, join_names
+
+
+class _OrderedMerge(_MergeOperation):
+    _merge_type = "ordered_merge"
+
+    def __init__(
+        self,
+        left: DataFrame | Series,
+        right: DataFrame | Series,
+        on: IndexLabel | None = None,
+        left_on: IndexLabel | None = None,
+        right_on: IndexLabel | None = None,
+        left_index: bool = False,
+        right_index: bool = False,
+        suffixes: Suffixes = ("_x", "_y"),
+        fill_method: str | None = None,
+        how: JoinHow | Literal["asof"] = "outer",
+    ) -> None:
+        self.fill_method = fill_method
+        _MergeOperation.__init__(
+            self,
+            left,
+            right,
+            on=on,
+            left_on=left_on,
+            left_index=left_index,
+            right_index=right_index,
+            right_on=right_on,
+            how=how,
+            suffixes=suffixes,
+            sort=True,  # factorize sorts
+        )
+
+    def get_result(self) -> DataFrame:
+        join_index, left_indexer, right_indexer = self._get_join_info()
+
+        left_join_indexer: npt.NDArray[np.intp] | None
+        right_join_indexer: npt.NDArray[np.intp] | None
+
+        if self.fill_method == "ffill":
+            if left_indexer is None:
+                left_join_indexer = None
+            else:
+                left_join_indexer = libjoin.ffill_indexer(left_indexer)
+            if right_indexer is None:
+                right_join_indexer = None
+            else:
+                right_join_indexer = libjoin.ffill_indexer(right_indexer)
+        elif self.fill_method is None:
+            left_join_indexer = left_indexer
+            right_join_indexer = right_indexer
+        else:
+            raise ValueError("fill_method must be 'ffill' or None")
+
+        result = self._reindex_and_concat(
+            join_index, left_join_indexer, right_join_indexer
+        )
+        self._maybe_add_join_keys(result, left_indexer, right_indexer)
+
+        return result
+
+
+def _asof_by_function(direction: str):
+    name = f"asof_join_{direction}_on_X_by_Y"
+    return getattr(libjoin, name, None)
+
+
+class _AsOfMerge(_OrderedMerge):
+    _merge_type = "asof_merge"
+
+    def __init__(
+        self,
+        left: DataFrame | Series,
+        right: DataFrame | Series,
+        on: IndexLabel | None = None,
+        left_on: IndexLabel | None = None,
+        right_on: IndexLabel | None = None,
+        left_index: bool = False,
+        right_index: bool = False,
+        by=None,
+        left_by=None,
+        right_by=None,
+        suffixes: Suffixes = ("_x", "_y"),
+        how: Literal["asof"] = "asof",
+        tolerance=None,
+        allow_exact_matches: bool = True,
+        direction: str = "backward",
+    ) -> None:
+        self.by = by
+        self.left_by = left_by
+        self.right_by = right_by
+        self.tolerance = tolerance
+        self.allow_exact_matches = allow_exact_matches
+        self.direction = direction
+
+        # check 'direction' is valid
+        if self.direction not in ["backward", "forward", "nearest"]:
+            raise MergeError(f"direction invalid: {self.direction}")
+
+        # validate allow_exact_matches
+        if not is_bool(self.allow_exact_matches):
+            msg = (
+                "allow_exact_matches must be boolean, "
+                f"passed {self.allow_exact_matches}"
+            )
+            raise MergeError(msg)
+
+        _OrderedMerge.__init__(
+            self,
+            left,
+            right,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            left_index=left_index,
+            right_index=right_index,
+            how=how,
+            suffixes=suffixes,
+            fill_method=None,
+        )
+
+    def _validate_left_right_on(self, left_on, right_on):
+        left_on, right_on = super()._validate_left_right_on(left_on, right_on)
+
+        # we only allow on to be a single item for on
+        if len(left_on) != 1 and not self.left_index:
+            raise MergeError("can only asof on a key for left")
+
+        if len(right_on) != 1 and not self.right_index:
+            raise MergeError("can only asof on a key for right")
+
+        if self.left_index and isinstance(self.left.index, MultiIndex):
+            raise MergeError("left can only have one index")
+
+        if self.right_index and isinstance(self.right.index, MultiIndex):
+            raise MergeError("right can only have one index")
+
+        # set 'by' columns
+        if self.by is not None:
+            if self.left_by is not None or self.right_by is not None:
+                raise MergeError("Can only pass by OR left_by and right_by")
+            self.left_by = self.right_by = self.by
+        if self.left_by is None and self.right_by is not None:
+            raise MergeError("missing left_by")
+        if self.left_by is not None and self.right_by is None:
+            raise MergeError("missing right_by")
+
+        # GH#29130 Check that merge keys do not have dtype object
+        if not self.left_index:
+            left_on_0 = left_on[0]
+            if isinstance(left_on_0, _known):
+                lo_dtype = left_on_0.dtype
+            else:
+                lo_dtype = (
+                    self.left._get_label_or_level_values(left_on_0).dtype
+                    if left_on_0 in self.left.columns
+                    else self.left.index.get_level_values(left_on_0)
+                )
+        else:
+            lo_dtype = self.left.index.dtype
+
+        if not self.right_index:
+            right_on_0 = right_on[0]
+            if isinstance(right_on_0, _known):
+                ro_dtype = right_on_0.dtype
+            else:
+                ro_dtype = (
+                    self.right._get_label_or_level_values(right_on_0).dtype
+                    if right_on_0 in self.right.columns
+                    else self.right.index.get_level_values(right_on_0)
+                )
+        else:
+            ro_dtype = self.right.index.dtype
+
+        if (
+            is_object_dtype(lo_dtype)
+            or is_object_dtype(ro_dtype)
+            or is_string_dtype(lo_dtype)
+            or is_string_dtype(ro_dtype)
+        ):
+            raise MergeError(
+                f"Incompatible merge dtype, {lo_dtype!r} and "
+                f"{ro_dtype!r}, both sides must have numeric dtype"
+            )
+
+        # add 'by' to our key-list so we can have it in the
+        # output as a key
+        if self.left_by is not None:
+            if not is_list_like(self.left_by):
+                self.left_by = [self.left_by]
+            if not is_list_like(self.right_by):
+                self.right_by = [self.right_by]
+
+            if len(self.left_by) != len(self.right_by):
+                raise MergeError("left_by and right_by must be the same length")
+
+            left_on = self.left_by + list(left_on)
+            right_on = self.right_by + list(right_on)
+
+        return left_on, right_on
+
+    def _maybe_require_matching_dtypes(
+        self, left_join_keys: list[ArrayLike], right_join_keys: list[ArrayLike]
+    ) -> None:
+        # TODO: why do we do this for AsOfMerge but not the others?
+
+        def _check_dtype_match(left: ArrayLike, right: ArrayLike, i: int) -> None:
+            if left.dtype != right.dtype:
+                if isinstance(left.dtype, CategoricalDtype) and isinstance(
+                    right.dtype, CategoricalDtype
+                ):
+                    # The generic error message is confusing for categoricals.
+                    #
+                    # In this function, the join keys include both the original
+                    # ones of the merge_asof() call, and also the keys passed
+                    # to its by= argument. Unordered but equal categories
+                    # are not supported for the former, but will fail
+                    # later with a ValueError, so we don't *need* to check
+                    # for them here.
+                    msg = (
+                        f"incompatible merge keys [{i}] {left.dtype!r} and "
+                        f"{right.dtype!r}, both sides category, but not equal ones"
+                    )
+                else:
+                    msg = (
+                        f"incompatible merge keys [{i}] {left.dtype!r} and "
+                        f"{right.dtype!r}, must be the same type"
+                    )
+                raise MergeError(msg)
+
+        # validate index types are the same
+        for i, (lk, rk) in enumerate(zip(left_join_keys, right_join_keys)):
+            _check_dtype_match(lk, rk, i)
+
+        if self.left_index:
+            lt = self.left.index._values
+        else:
+            lt = left_join_keys[-1]
+
+        if self.right_index:
+            rt = self.right.index._values
+        else:
+            rt = right_join_keys[-1]
+
+        _check_dtype_match(lt, rt, 0)
+
+    def _validate_tolerance(self, left_join_keys: list[ArrayLike]) -> None:
+        # validate tolerance; datetime.timedelta or Timedelta if we have a DTI
+        if self.tolerance is not None:
+            if self.left_index:
+                lt = self.left.index._values
+            else:
+                lt = left_join_keys[-1]
+
+            msg = (
+                f"incompatible tolerance {self.tolerance}, must be compat "
+                f"with type {lt.dtype!r}"
+            )
+
+            if needs_i8_conversion(lt.dtype) or (
+                isinstance(lt, ArrowExtensionArray) and lt.dtype.kind in "mM"
+            ):
+                if not isinstance(self.tolerance, datetime.timedelta):
+                    raise MergeError(msg)
+                if self.tolerance < Timedelta(0):
+                    raise MergeError("tolerance must be positive")
+
+            elif is_integer_dtype(lt.dtype):
+                if not is_integer(self.tolerance):
+                    raise MergeError(msg)
+                if self.tolerance < 0:
+                    raise MergeError("tolerance must be positive")
+
+            elif is_float_dtype(lt.dtype):
+                if not is_number(self.tolerance):
+                    raise MergeError(msg)
+                # error: Unsupported operand types for > ("int" and "Number")
+                if self.tolerance < 0:  # type: ignore[operator]
+                    raise MergeError("tolerance must be positive")
+
+            else:
+                raise MergeError("key must be integer, timestamp or float")
+
+    def _convert_values_for_libjoin(
+        self, values: AnyArrayLike, side: str
+    ) -> np.ndarray:
+        # we require sortedness and non-null values in the join keys
+        if not Index(values).is_monotonic_increasing:
+            if isna(values).any():
+                raise ValueError(f"Merge keys contain null values on {side} side")
+            raise ValueError(f"{side} keys must be sorted")
+
+        if isinstance(values, ArrowExtensionArray):
+            values = values._maybe_convert_datelike_array()
+
+        if needs_i8_conversion(values.dtype):
+            values = values.view("i8")
+
+        elif isinstance(values, BaseMaskedArray):
+            # we've verified above that no nulls exist
+            values = values._data
+        elif isinstance(values, ExtensionArray):
+            values = values.to_numpy()
+
+        # error: Incompatible return value type (got "Union[ExtensionArray,
+        # Any, ndarray[Any, Any], ndarray[Any, dtype[Any]], Index, Series]",
+        # expected "ndarray[Any, Any]")
+        return values  # type: ignore[return-value]
+
+    def _get_join_indexers(self) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+        """return the join indexers"""
+
+        # values to compare
+        left_values = (
+            self.left.index._values if self.left_index else self.left_join_keys[-1]
+        )
+        right_values = (
+            self.right.index._values if self.right_index else self.right_join_keys[-1]
+        )
+
+        # _maybe_require_matching_dtypes already checked for dtype matching
+        assert left_values.dtype == right_values.dtype
+
+        tolerance = self.tolerance
+        if tolerance is not None:
+            # TODO: can we reuse a tolerance-conversion function from
+            #  e.g. TimedeltaIndex?
+            if needs_i8_conversion(left_values.dtype) or (
+                isinstance(left_values, ArrowExtensionArray)
+                and left_values.dtype.kind in "mM"
+            ):
+                tolerance = Timedelta(tolerance)
+                # TODO: we have no test cases with PeriodDtype here; probably
+                #  need to adjust tolerance for that case.
+                if left_values.dtype.kind in "mM":
+                    # Make sure the i8 representation for tolerance
+                    #  matches that for left_values/right_values.
+                    if isinstance(left_values, ArrowExtensionArray):
+                        unit = left_values.dtype.pyarrow_dtype.unit
+                    else:
+                        unit = ensure_wrapped_if_datetimelike(left_values).unit
+                    tolerance = tolerance.as_unit(unit)
+
+                tolerance = tolerance._value
+
+        # initial type conversion as needed
+        left_values = self._convert_values_for_libjoin(left_values, "left")
+        right_values = self._convert_values_for_libjoin(right_values, "right")
+
+        # a "by" parameter requires special handling
+        if self.left_by is not None:
+            # remove 'on' parameter from values if one existed
+            if self.left_index and self.right_index:
+                left_join_keys = self.left_join_keys
+                right_join_keys = self.right_join_keys
+            else:
+                left_join_keys = self.left_join_keys[0:-1]
+                right_join_keys = self.right_join_keys[0:-1]
+
+            mapped = [
+                _factorize_keys(
+                    left_join_keys[n],
+                    right_join_keys[n],
+                    sort=False,
+                )
+                for n in range(len(left_join_keys))
+            ]
+
+            if len(left_join_keys) == 1:
+                left_by_values = mapped[0][0]
+                right_by_values = mapped[0][1]
+            else:
+                arrs = [np.concatenate(m[:2]) for m in mapped]
+                shape = tuple(m[2] for m in mapped)
+                group_index = get_group_index(
+                    arrs, shape=shape, sort=False, xnull=False
+                )
+                left_len = len(left_join_keys[0])
+                left_by_values = group_index[:left_len]
+                right_by_values = group_index[left_len:]
+
+            left_by_values = ensure_int64(left_by_values)
+            right_by_values = ensure_int64(right_by_values)
+
+            # choose appropriate function by type
+            func = _asof_by_function(self.direction)
+            return func(
+                left_values,
+                right_values,
+                left_by_values,
+                right_by_values,
+                self.allow_exact_matches,
+                tolerance,
+            )
+        else:
+            # choose appropriate function by type
+            func = _asof_by_function(self.direction)
+            return func(
+                left_values,
+                right_values,
+                None,
+                None,
+                self.allow_exact_matches,
+                tolerance,
+                False,
+            )
+
+
+def _get_multiindex_indexer(
+    join_keys: list[ArrayLike], index: MultiIndex, sort: bool
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+    # left & right join labels and num. of levels at each location
+    mapped = (
+        _factorize_keys(index.levels[n]._values, join_keys[n], sort=sort)
+        for n in range(index.nlevels)
+    )
+    zipped = zip(*mapped)
+    rcodes, lcodes, shape = (list(x) for x in zipped)
+    if sort:
+        rcodes = list(map(np.take, rcodes, index.codes))
+    else:
+        i8copy = lambda a: a.astype("i8", subok=False)
+        rcodes = list(map(i8copy, index.codes))
+
+    # fix right labels if there were any nulls
+    for i, join_key in enumerate(join_keys):
+        mask = index.codes[i] == -1
+        if mask.any():
+            # check if there already was any nulls at this location
+            # if there was, it is factorized to `shape[i] - 1`
+            a = join_key[lcodes[i] == shape[i] - 1]
+            if a.size == 0 or not a[0] != a[0]:
+                shape[i] += 1
+
+            rcodes[i][mask] = shape[i] - 1
+
+    # get flat i8 join keys
+    lkey, rkey = _get_join_keys(lcodes, rcodes, tuple(shape), sort)
+    return lkey, rkey
+
+
+def _get_empty_indexer() -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+    """Return empty join indexers."""
+    return (
+        np.array([], dtype=np.intp),
+        np.array([], dtype=np.intp),
+    )
+
+
+def _get_no_sort_one_missing_indexer(
+    n: int, left_missing: bool
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+    """
+    Return join indexers where all of one side is selected without sorting
+    and none of the other side is selected.
+
+    Parameters
+    ----------
+    n : int
+        Length of indexers to create.
+    left_missing : bool
+        If True, the left indexer will contain only -1's.
+        If False, the right indexer will contain only -1's.
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+        Left indexer
+    np.ndarray[np.intp]
+        Right indexer
+    """
+    idx = np.arange(n, dtype=np.intp)
+    idx_missing = np.full(shape=n, fill_value=-1, dtype=np.intp)
+    if left_missing:
+        return idx_missing, idx
+    return idx, idx_missing
+
+
+def _left_join_on_index(
+    left_ax: Index, right_ax: Index, join_keys: list[ArrayLike], sort: bool = False
+) -> tuple[Index, npt.NDArray[np.intp] | None, npt.NDArray[np.intp]]:
+    if isinstance(right_ax, MultiIndex):
+        lkey, rkey = _get_multiindex_indexer(join_keys, right_ax, sort=sort)
+    else:
+        # error: Incompatible types in assignment (expression has type
+        # "Union[Union[ExtensionArray, ndarray[Any, Any]], Index, Series]",
+        # variable has type "ndarray[Any, dtype[signedinteger[Any]]]")
+        lkey = join_keys[0]  # type: ignore[assignment]
+        # error: Incompatible types in assignment (expression has type "Index",
+        # variable has type "ndarray[Any, dtype[signedinteger[Any]]]")
+        rkey = right_ax._values  # type: ignore[assignment]
+
+    left_key, right_key, count = _factorize_keys(lkey, rkey, sort=sort)
+    left_indexer, right_indexer = libjoin.left_outer_join(
+        left_key, right_key, count, sort=sort
+    )
+
+    if sort or len(left_ax) != len(left_indexer):
+        # if asked to sort or there are 1-to-many matches
+        join_index = left_ax.take(left_indexer)
+        return join_index, left_indexer, right_indexer
+
+    # left frame preserves order & length of its index
+    return left_ax, None, right_indexer
+
+
+def _factorize_keys(
+    lk: ArrayLike,
+    rk: ArrayLike,
+    sort: bool = True,
+    how: str | None = None,
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp], int]:
+    """
+    Encode left and right keys as enumerated types.
+
+    This is used to get the join indexers to be used when merging DataFrames.
+
+    Parameters
+    ----------
+    lk : ndarray, ExtensionArray
+        Left key.
+    rk : ndarray, ExtensionArray
+        Right key.
+    sort : bool, defaults to True
+        If True, the encoding is done such that the unique elements in the
+        keys are sorted.
+    how: str, optional
+        Used to determine if we can use hash-join. If not given, then just factorize
+        keys.
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+        Left (resp. right if called with `key='right'`) labels, as enumerated type.
+    np.ndarray[np.intp]
+        Right (resp. left if called with `key='right'`) labels, as enumerated type.
+    int
+        Number of unique elements in union of left and right labels. -1 if we used
+        a hash-join.
+
+    See Also
+    --------
+    merge : Merge DataFrame or named Series objects
+        with a database-style join.
+    algorithms.factorize : Encode the object as an enumerated type
+        or categorical variable.
+
+    Examples
+    --------
+    >>> lk = np.array(["a", "c", "b"])
+    >>> rk = np.array(["a", "c"])
+
+    Here, the unique values are `'a', 'b', 'c'`. With the default
+    `sort=True`, the encoding will be `{0: 'a', 1: 'b', 2: 'c'}`:
+
+    >>> pd.core.reshape.merge._factorize_keys(lk, rk)
+    (array([0, 2, 1]), array([0, 2]), 3)
+
+    With the `sort=False`, the encoding will correspond to the order
+    in which the unique elements first appear: `{0: 'a', 1: 'c', 2: 'b'}`:
+
+    >>> pd.core.reshape.merge._factorize_keys(lk, rk, sort=False)
+    (array([0, 1, 2]), array([0, 1]), 3)
+    """
+    # TODO: if either is a RangeIndex, we can likely factorize more efficiently?
+
+    if (
+        isinstance(lk.dtype, DatetimeTZDtype) and isinstance(rk.dtype, DatetimeTZDtype)
+    ) or (lib.is_np_dtype(lk.dtype, "M") and lib.is_np_dtype(rk.dtype, "M")):
+        # Extract the ndarray (UTC-localized) values
+        # Note: we dont need the dtypes to match, as these can still be compared
+        lk, rk = cast("DatetimeArray", lk)._ensure_matching_resos(rk)
+        lk = cast("DatetimeArray", lk)._ndarray
+        rk = cast("DatetimeArray", rk)._ndarray
+
+    elif (
+        isinstance(lk.dtype, CategoricalDtype)
+        and isinstance(rk.dtype, CategoricalDtype)
+        and lk.dtype == rk.dtype
+    ):
+        assert isinstance(lk, Categorical)
+        assert isinstance(rk, Categorical)
+        # Cast rk to encoding so we can compare codes with lk
+
+        rk = lk._encode_with_my_categories(rk)
+
+        lk = ensure_int64(lk.codes)
+        rk = ensure_int64(rk.codes)
+
+    elif isinstance(lk, ExtensionArray) and lk.dtype == rk.dtype:
+        if (isinstance(lk.dtype, ArrowDtype) and is_string_dtype(lk.dtype)) or (
+            isinstance(lk.dtype, StringDtype) and lk.dtype.storage == "pyarrow"
+        ):
+            import pyarrow as pa
+            import pyarrow.compute as pc
+
+            len_lk = len(lk)
+            lk = lk._pa_array  # type: ignore[attr-defined]
+            rk = rk._pa_array  # type: ignore[union-attr]
+            dc = (
+                pa.chunked_array(lk.chunks + rk.chunks)  # type: ignore[union-attr]
+                .combine_chunks()
+                .dictionary_encode()
+            )
+
+            llab, rlab, count = (
+                pc.fill_null(dc.indices[slice(len_lk)], -1)
+                .to_numpy()
+                .astype(np.intp, copy=False),
+                pc.fill_null(dc.indices[slice(len_lk, None)], -1)
+                .to_numpy()
+                .astype(np.intp, copy=False),
+                len(dc.dictionary),
+            )
+
+            if sort:
+                uniques = dc.dictionary.to_numpy(zero_copy_only=False)
+                llab, rlab = _sort_labels(uniques, llab, rlab)
+
+            if dc.null_count > 0:
+                lmask = llab == -1
+                lany = lmask.any()
+                rmask = rlab == -1
+                rany = rmask.any()
+                if lany:
+                    np.putmask(llab, lmask, count)
+                if rany:
+                    np.putmask(rlab, rmask, count)
+                count += 1
+            return llab, rlab, count
+
+        if not isinstance(lk, BaseMaskedArray) and not (
+            # exclude arrow dtypes that would get cast to object
+            isinstance(lk.dtype, ArrowDtype)
+            and (
+                is_numeric_dtype(lk.dtype.numpy_dtype)
+                or (is_string_dtype(lk.dtype) and not sort)
+            )
+        ):
+            lk, _ = lk._values_for_factorize()
+
+            # error: Item "ndarray" of "Union[Any, ndarray]" has no attribute
+            # "_values_for_factorize"
+            rk, _ = rk._values_for_factorize()  # type: ignore[union-attr]
+
+    if needs_i8_conversion(lk.dtype) and lk.dtype == rk.dtype:
+        # GH#23917 TODO: Needs tests for non-matching dtypes
+        # GH#23917 TODO: needs tests for case where lk is integer-dtype
+        #  and rk is datetime-dtype
+        lk = np.asarray(lk, dtype=np.int64)
+        rk = np.asarray(rk, dtype=np.int64)
+
+    klass, lk, rk = _convert_arrays_and_get_rizer_klass(lk, rk)
+
+    rizer = klass(
+        max(len(lk), len(rk)),
+        uses_mask=isinstance(rk, (BaseMaskedArray, ArrowExtensionArray)),
+    )
+
+    if isinstance(lk, BaseMaskedArray):
+        assert isinstance(rk, BaseMaskedArray)
+        lk_data, lk_mask = lk._data, lk._mask
+        rk_data, rk_mask = rk._data, rk._mask
+    elif isinstance(lk, ArrowExtensionArray):
+        assert isinstance(rk, ArrowExtensionArray)
+        # we can only get here with numeric dtypes
+        # TODO: Remove when we have a Factorizer for Arrow
+        lk_data = lk.to_numpy(na_value=1, dtype=lk.dtype.numpy_dtype)
+        rk_data = rk.to_numpy(na_value=1, dtype=lk.dtype.numpy_dtype)
+        lk_mask, rk_mask = lk.isna(), rk.isna()
+    else:
+        # Argument 1 to "factorize" of "ObjectFactorizer" has incompatible type
+        # "Union[ndarray[Any, dtype[signedinteger[_64Bit]]],
+        # ndarray[Any, dtype[object_]]]"; expected "ndarray[Any, dtype[object_]]"
+        lk_data, rk_data = lk, rk  # type: ignore[assignment]
+        lk_mask, rk_mask = None, None
+
+    hash_join_available = how == "inner" and not sort and lk.dtype.kind in "iufb"
+    if hash_join_available:
+        rlab = rizer.factorize(rk_data, mask=rk_mask)
+        if rizer.get_count() == len(rlab):
+            ridx, lidx = rizer.hash_inner_join(lk_data, lk_mask)
+            return lidx, ridx, -1
+        else:
+            llab = rizer.factorize(lk_data, mask=lk_mask)
+    else:
+        llab = rizer.factorize(lk_data, mask=lk_mask)
+        rlab = rizer.factorize(rk_data, mask=rk_mask)
+
+    assert llab.dtype == np.dtype(np.intp), llab.dtype
+    assert rlab.dtype == np.dtype(np.intp), rlab.dtype
+
+    count = rizer.get_count()
+
+    if sort:
+        uniques = rizer.uniques.to_array()
+        llab, rlab = _sort_labels(uniques, llab, rlab)
+
+    # NA group
+    lmask = llab == -1
+    lany = lmask.any()
+    rmask = rlab == -1
+    rany = rmask.any()
+
+    if lany or rany:
+        if lany:
+            np.putmask(llab, lmask, count)
+        if rany:
+            np.putmask(rlab, rmask, count)
+        count += 1
+
+    return llab, rlab, count
+
+
+def _convert_arrays_and_get_rizer_klass(
+    lk: ArrayLike, rk: ArrayLike
+) -> tuple[type[libhashtable.Factorizer], ArrayLike, ArrayLike]:
+    klass: type[libhashtable.Factorizer]
+    if is_numeric_dtype(lk.dtype):
+        if lk.dtype != rk.dtype:
+            dtype = find_common_type([lk.dtype, rk.dtype])
+            if isinstance(dtype, ExtensionDtype):
+                cls = dtype.construct_array_type()
+                if not isinstance(lk, ExtensionArray):
+                    lk = cls._from_sequence(lk, dtype=dtype, copy=False)
+                else:
+                    lk = lk.astype(dtype, copy=False)
+
+                if not isinstance(rk, ExtensionArray):
+                    rk = cls._from_sequence(rk, dtype=dtype, copy=False)
+                else:
+                    rk = rk.astype(dtype, copy=False)
+            else:
+                lk = lk.astype(dtype, copy=False)
+                rk = rk.astype(dtype, copy=False)
+        if isinstance(lk, BaseMaskedArray):
+            #  Invalid index type "type" for "Dict[Type[object], Type[Factorizer]]";
+            #  expected type "Type[object]"
+            klass = _factorizers[lk.dtype.type]  # type: ignore[index]
+        elif isinstance(lk.dtype, ArrowDtype):
+            klass = _factorizers[lk.dtype.numpy_dtype.type]
+        else:
+            klass = _factorizers[lk.dtype.type]
+
+    else:
+        klass = libhashtable.ObjectFactorizer
+        lk = ensure_object(lk)
+        rk = ensure_object(rk)
+    return klass, lk, rk
+
+
+def _sort_labels(
+    uniques: np.ndarray, left: npt.NDArray[np.intp], right: npt.NDArray[np.intp]
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
+    llength = len(left)
+    labels = np.concatenate([left, right])
+
+    _, new_labels = algos.safe_sort(uniques, labels, use_na_sentinel=True)
+    new_left, new_right = new_labels[:llength], new_labels[llength:]
+
+    return new_left, new_right
+
+
+def _get_join_keys(
+    llab: list[npt.NDArray[np.int64 | np.intp]],
+    rlab: list[npt.NDArray[np.int64 | np.intp]],
+    shape: Shape,
+    sort: bool,
+) -> tuple[npt.NDArray[np.int64], npt.NDArray[np.int64]]:
+    # how many levels can be done without overflow
+    nlev = next(
+        lev
+        for lev in range(len(shape), 0, -1)
+        if not is_int64_overflow_possible(shape[:lev])
+    )
+
+    # get keys for the first `nlev` levels
+    stride = np.prod(shape[1:nlev], dtype="i8")
+    lkey = stride * llab[0].astype("i8", subok=False, copy=False)
+    rkey = stride * rlab[0].astype("i8", subok=False, copy=False)
+
+    for i in range(1, nlev):
+        with np.errstate(divide="ignore"):
+            stride //= shape[i]
+        lkey += llab[i] * stride
+        rkey += rlab[i] * stride
+
+    if nlev == len(shape):  # all done!
+        return lkey, rkey
+
+    # densify current keys to avoid overflow
+    lkey, rkey, count = _factorize_keys(lkey, rkey, sort=sort)
+
+    llab = [lkey] + llab[nlev:]
+    rlab = [rkey] + rlab[nlev:]
+    shape = (count,) + shape[nlev:]
+
+    return _get_join_keys(llab, rlab, shape, sort)
+
+
+def _should_fill(lname, rname) -> bool:
+    if not isinstance(lname, str) or not isinstance(rname, str):
+        return True
+    return lname == rname
+
+
+def _any(x) -> bool:
+    return x is not None and com.any_not_none(*x)
+
+
+def _validate_operand(obj: DataFrame | Series) -> DataFrame:
+    if isinstance(obj, ABCDataFrame):
+        return obj
+    elif isinstance(obj, ABCSeries):
+        if obj.name is None:
+            raise ValueError("Cannot merge a Series without a name")
+        return obj.to_frame()
+    else:
+        raise TypeError(
+            f"Can only merge Series or DataFrame objects, a {type(obj)} was passed"
+        )
+
+
+def _items_overlap_with_suffix(
+    left: Index, right: Index, suffixes: Suffixes
+) -> tuple[Index, Index]:
+    """
+    Suffixes type validation.
+
+    If two indices overlap, add suffixes to overlapping entries.
+
+    If corresponding suffix is empty, the entry is simply converted to string.
+
+    """
+    if not is_list_like(suffixes, allow_sets=False) or isinstance(suffixes, dict):
+        raise TypeError(
+            f"Passing 'suffixes' as a {type(suffixes)}, is not supported. "
+            "Provide 'suffixes' as a tuple instead."
+        )
+
+    to_rename = left.intersection(right)
+    if len(to_rename) == 0:
+        return left, right
+
+    lsuffix, rsuffix = suffixes
+
+    if not lsuffix and not rsuffix:
+        raise ValueError(f"columns overlap but no suffix specified: {to_rename}")
+
+    def renamer(x, suffix: str | None):
+        """
+        Rename the left and right indices.
+
+        If there is overlap, and suffix is not None, add
+        suffix, otherwise, leave it as-is.
+
+        Parameters
+        ----------
+        x : original column name
+        suffix : str or None
+
+        Returns
+        -------
+        x : renamed column name
+        """
+        if x in to_rename and suffix is not None:
+            return f"{x}{suffix}"
+        return x
+
+    lrenamer = partial(renamer, suffix=lsuffix)
+    rrenamer = partial(renamer, suffix=rsuffix)
+
+    llabels = left._transform_index(lrenamer)
+    rlabels = right._transform_index(rrenamer)
+
+    dups = []
+    if not llabels.is_unique:
+        # Only warn when duplicates are caused because of suffixes, already duplicated
+        # columns in origin should not warn
+        dups = llabels[(llabels.duplicated()) & (~left.duplicated())].tolist()
+    if not rlabels.is_unique:
+        dups.extend(rlabels[(rlabels.duplicated()) & (~right.duplicated())].tolist())
+    if dups:
+        raise MergeError(
+            f"Passing 'suffixes' which cause duplicate columns {set(dups)} is "
+            f"not allowed.",
+        )
+
+    return llabels, rlabels
diff --git a/pandas/pandas/core/reshape/pivot.py b/pandas/pandas/core/reshape/pivot.py
new file mode 100644
index 0000000000000..cfc6f91557781
--- /dev/null
+++ b/pandas/pandas/core/reshape/pivot.py
@@ -0,0 +1,1257 @@
+from __future__ import annotations
+
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.cast import maybe_downcast_to_dtype
+from pandas.core.dtypes.common import (
+    is_list_like,
+    is_nested_list_like,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+import pandas.core.common as com
+from pandas.core.groupby import Grouper
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    get_objs_combined_axis,
+)
+from pandas.core.reshape.concat import concat
+from pandas.core.series import Series
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+    )
+
+    from pandas._typing import (
+        AggFuncType,
+        AggFuncTypeBase,
+        AggFuncTypeDict,
+        IndexLabel,
+        SequenceNotStr,
+    )
+
+    from pandas import DataFrame
+
+
+def pivot_table(
+    data: DataFrame,
+    values=None,
+    index=None,
+    columns=None,
+    aggfunc: AggFuncType = "mean",
+    fill_value=None,
+    margins: bool = False,
+    dropna: bool = True,
+    margins_name: Hashable = "All",
+    observed: bool = True,
+    sort: bool = True,
+    **kwargs,
+) -> DataFrame:
+    """
+    Create a spreadsheet-style pivot table as a DataFrame.
+
+    The levels in the pivot table will be stored in MultiIndex objects
+    (hierarchical indexes) on the index and columns of the result DataFrame.
+
+    Parameters
+    ----------
+    data : DataFrame
+        Input pandas DataFrame object.
+    values : list-like or scalar, optional
+        Column or columns to aggregate.
+    index : column, Grouper, array, or list of the previous
+        Keys to group by on the pivot table index. If a list is passed,
+        it can contain any of the other types (except list). If an array is
+        passed, it must be the same length as the data and will be used in
+        the same manner as column values.
+    columns : column, Grouper, array, or list of the previous
+        Keys to group by on the pivot table column. If a list is passed,
+        it can contain any of the other types (except list). If an array is
+        passed, it must be the same length as the data and will be used in
+        the same manner as column values.
+    aggfunc : function, list of functions, dict, default "mean"
+        If a list of functions is passed, the resulting pivot table will have
+        hierarchical columns whose top level are the function names
+        (inferred from the function objects themselves).
+        If a dict is passed, the key is column to aggregate and the value is
+        function or list of functions. If ``margins=True``, aggfunc will be
+        used to calculate the partial aggregates.
+    fill_value : scalar, default None
+        Value to replace missing values with (in the resulting pivot table,
+        after aggregation).
+    margins : bool, default False
+        If ``margins=True``, special ``All`` columns and rows
+        will be added with partial group aggregates across the categories
+        on the rows and columns.
+    dropna : bool, default True
+        Do not include columns whose entries are all NaN. If True,
+        rows with a NaN value in any column will be omitted before
+        computing margins.
+    margins_name : str, default 'All'
+        Name of the row / column that will contain the totals
+        when margins is True.
+    observed : bool, default False
+        This only applies if any of the groupers are Categoricals.
+        If True: only show observed values for categorical groupers.
+        If False: show all values for categorical groupers.
+
+        .. versionchanged:: 3.0.0
+
+            The default value is now ``True``.
+
+    sort : bool, default True
+        Specifies if the result should be sorted.
+
+        .. versionadded:: 1.3.0
+
+    **kwargs : dict
+        Optional keyword arguments to pass to ``aggfunc``.
+
+        .. versionadded:: 3.0.0
+
+    Returns
+    -------
+    DataFrame
+        An Excel style pivot table.
+
+    See Also
+    --------
+    DataFrame.pivot : Pivot without aggregation that can handle
+        non-numeric data.
+    DataFrame.melt: Unpivot a DataFrame from wide to long format,
+        optionally leaving identifiers set.
+    wide_to_long : Wide panel to long format. Less flexible but more
+        user-friendly than melt.
+
+    Notes
+    -----
+    Reference :ref:`the user guide <reshaping.pivot>` for more examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+    ...         "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+    ...         "C": [
+    ...             "small",
+    ...             "large",
+    ...             "large",
+    ...             "small",
+    ...             "small",
+    ...             "large",
+    ...             "small",
+    ...             "small",
+    ...             "large",
+    ...         ],
+    ...         "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+    ...         "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+    ...     }
+    ... )
+    >>> df
+         A    B      C  D  E
+    0  foo  one  small  1  2
+    1  foo  one  large  2  4
+    2  foo  one  large  2  5
+    3  foo  two  small  3  5
+    4  foo  two  small  3  6
+    5  bar  one  large  4  6
+    6  bar  one  small  5  8
+    7  bar  two  small  6  9
+    8  bar  two  large  7  9
+
+    This first example aggregates values by taking the sum.
+
+    >>> table = pd.pivot_table(
+    ...     df, values="D", index=["A", "B"], columns=["C"], aggfunc="sum"
+    ... )
+    >>> table
+    C        large  small
+    A   B
+    bar one    4.0    5.0
+        two    7.0    6.0
+    foo one    4.0    1.0
+        two    NaN    6.0
+
+    We can also fill missing values using the `fill_value` parameter.
+
+    >>> table = pd.pivot_table(
+    ...     df, values="D", index=["A", "B"], columns=["C"], aggfunc="sum", fill_value=0
+    ... )
+    >>> table
+    C        large  small
+    A   B
+    bar one      4      5
+        two      7      6
+    foo one      4      1
+        two      0      6
+
+    The next example aggregates by taking the mean across multiple columns.
+
+    >>> table = pd.pivot_table(
+    ...     df, values=["D", "E"], index=["A", "C"], aggfunc={"D": "mean", "E": "mean"}
+    ... )
+    >>> table
+                    D         E
+    A   C
+    bar large  5.500000  7.500000
+        small  5.500000  8.500000
+    foo large  2.000000  4.500000
+        small  2.333333  4.333333
+
+    We can also calculate multiple types of aggregations for any given
+    value column.
+
+    >>> table = pd.pivot_table(
+    ...     df,
+    ...     values=["D", "E"],
+    ...     index=["A", "C"],
+    ...     aggfunc={"D": "mean", "E": ["min", "max", "mean"]},
+    ... )
+    >>> table
+                      D   E
+                   mean max      mean  min
+    A   C
+    bar large  5.500000   9  7.500000    6
+        small  5.500000   9  8.500000    8
+    foo large  2.000000   5  4.500000    4
+        small  2.333333   6  4.333333    2
+    """
+    index = _convert_by(index)
+    columns = _convert_by(columns)
+
+    if isinstance(aggfunc, list):
+        pieces: list[DataFrame] = []
+        keys = []
+        for func in aggfunc:
+            _table = __internal_pivot_table(
+                data,
+                values=values,
+                index=index,
+                columns=columns,
+                fill_value=fill_value,
+                aggfunc=func,
+                margins=margins,
+                dropna=dropna,
+                margins_name=margins_name,
+                observed=observed,
+                sort=sort,
+                kwargs=kwargs,
+            )
+            pieces.append(_table)
+            keys.append(getattr(func, "__name__", func))
+
+        table = concat(pieces, keys=keys, axis=1)
+        return table.__finalize__(data, method="pivot_table")
+
+    table = __internal_pivot_table(
+        data,
+        values,
+        index,
+        columns,
+        aggfunc,
+        fill_value,
+        margins,
+        dropna,
+        margins_name,
+        observed,
+        sort,
+        kwargs,
+    )
+    return table.__finalize__(data, method="pivot_table")
+
+
+def __internal_pivot_table(
+    data: DataFrame,
+    values,
+    index,
+    columns,
+    aggfunc: AggFuncTypeBase | AggFuncTypeDict,
+    fill_value,
+    margins: bool,
+    dropna: bool,
+    margins_name: Hashable,
+    observed: bool,
+    sort: bool,
+    kwargs,
+) -> DataFrame:
+    """
+    Helper of :func:`pandas.pivot_table` for any non-list ``aggfunc``.
+    """
+    keys = index + columns
+
+    values_passed = values is not None
+    if values_passed:
+        if is_list_like(values):
+            values_multi = True
+            values = list(values)
+        else:
+            values_multi = False
+            values = [values]
+
+        # GH14938 Make sure value labels are in data
+        for i in values:
+            if i not in data:
+                raise KeyError(i)
+
+        to_filter = []
+        for x in keys + values:
+            if isinstance(x, Grouper):
+                x = x.key
+            try:
+                if x in data:
+                    to_filter.append(x)
+            except TypeError:
+                pass
+        if len(to_filter) < len(data.columns):
+            data = data[to_filter]
+
+    else:
+        values = data.columns
+        for key in keys:
+            try:
+                values = values.drop(key)
+            except (TypeError, ValueError, KeyError):
+                pass
+        values = list(values)
+
+    grouped = data.groupby(keys, observed=observed, sort=sort, dropna=dropna)
+    agged = grouped.agg(aggfunc, **kwargs)
+
+    if dropna and isinstance(agged, ABCDataFrame) and len(agged.columns):
+        agged = agged.dropna(how="all")
+
+    table = agged
+
+    # GH17038, this check should only happen if index is defined (not None)
+    if table.index.nlevels > 1 and index:
+        # Related GH #17123
+        # If index_names are integers, determine whether the integers refer
+        # to the level position or name.
+        index_names = agged.index.names[: len(index)]
+        to_unstack = []
+        for i in range(len(index), len(keys)):
+            name = agged.index.names[i]
+            if name is None or name in index_names:
+                to_unstack.append(i)
+            else:
+                to_unstack.append(name)
+        table = agged.unstack(to_unstack, fill_value=fill_value)
+
+    if not dropna:
+        if isinstance(table.index, MultiIndex):
+            m = MultiIndex.from_product(table.index.levels, names=table.index.names)
+            table = table.reindex(m, axis=0, fill_value=fill_value)
+
+        if isinstance(table.columns, MultiIndex):
+            m = MultiIndex.from_product(table.columns.levels, names=table.columns.names)
+            table = table.reindex(m, axis=1, fill_value=fill_value)
+
+    if sort is True and isinstance(table, ABCDataFrame):
+        table = table.sort_index(axis=1)
+
+    if fill_value is not None:
+        table = table.fillna(fill_value)
+        if aggfunc is len and not observed and lib.is_integer(fill_value):
+            # TODO: can we avoid this?  this used to be handled by
+            #  downcast="infer" in fillna
+            table = table.astype(np.int64)
+
+    if margins:
+        if dropna:
+            data = data[data.notna().all(axis=1)]
+        table = _add_margins(
+            table,
+            data,
+            values,
+            rows=index,
+            cols=columns,
+            aggfunc=aggfunc,
+            kwargs=kwargs,
+            observed=dropna,
+            margins_name=margins_name,
+            fill_value=fill_value,
+        )
+
+    # discard the top level
+    if values_passed and not values_multi and table.columns.nlevels > 1:
+        table.columns = table.columns.droplevel(0)
+    if len(index) == 0 and len(columns) > 0:
+        table = table.T
+
+    # GH 15193 Make sure empty columns are removed if dropna=True
+    if isinstance(table, ABCDataFrame) and dropna:
+        table = table.dropna(how="all", axis=1)
+
+    return table
+
+
+def _add_margins(
+    table: DataFrame | Series,
+    data: DataFrame,
+    values,
+    rows,
+    cols,
+    aggfunc,
+    kwargs,
+    observed: bool,
+    margins_name: Hashable = "All",
+    fill_value=None,
+):
+    if not isinstance(margins_name, str):
+        raise ValueError("margins_name argument must be a string")
+
+    msg = f'Conflicting name "{margins_name}" in margins'
+    for level in table.index.names:
+        if margins_name in table.index.get_level_values(level):
+            raise ValueError(msg)
+
+    grand_margin = _compute_grand_margin(data, values, aggfunc, kwargs, margins_name)
+
+    if table.ndim == 2:
+        # i.e. DataFrame
+        for level in table.columns.names[1:]:
+            if margins_name in table.columns.get_level_values(level):
+                raise ValueError(msg)
+
+    key: str | tuple[str, ...]
+    if len(rows) > 1:
+        key = (margins_name,) + ("",) * (len(rows) - 1)
+    else:
+        key = margins_name
+
+    if not values and isinstance(table, ABCSeries):
+        # If there are no values and the table is a series, then there is only
+        # one column in the data. Compute grand margin and return it.
+        return table._append(table._constructor({key: grand_margin[margins_name]}))
+
+    elif values:
+        marginal_result_set = _generate_marginal_results(
+            table,
+            data,
+            values,
+            rows,
+            cols,
+            aggfunc,
+            kwargs,
+            observed,
+            margins_name,
+        )
+        if not isinstance(marginal_result_set, tuple):
+            return marginal_result_set
+        result, margin_keys, row_margin = marginal_result_set
+    else:
+        # no values, and table is a DataFrame
+        assert isinstance(table, ABCDataFrame)
+        marginal_result_set = _generate_marginal_results_without_values(
+            table, data, rows, cols, aggfunc, kwargs, observed, margins_name
+        )
+        if not isinstance(marginal_result_set, tuple):
+            return marginal_result_set
+        result, margin_keys, row_margin = marginal_result_set
+
+    row_margin = row_margin.reindex(result.columns, fill_value=fill_value)
+    # populate grand margin
+    for k in margin_keys:
+        if isinstance(k, str):
+            row_margin[k] = grand_margin[k]
+        else:
+            row_margin[k] = grand_margin[k[0]]
+
+    from pandas import DataFrame
+
+    margin_dummy = DataFrame(row_margin, columns=Index([key])).T
+
+    row_names = result.index.names
+    # check the result column and leave floats
+
+    for dtype in set(result.dtypes):
+        if isinstance(dtype, ExtensionDtype):
+            # Can hold NA already
+            continue
+
+        cols = result.select_dtypes([dtype]).columns
+        margin_dummy[cols] = margin_dummy[cols].apply(
+            maybe_downcast_to_dtype, args=(dtype,)
+        )
+    result = result._append(margin_dummy)
+    result.index.names = row_names
+
+    return result
+
+
+def _compute_grand_margin(
+    data: DataFrame, values, aggfunc, kwargs, margins_name: Hashable = "All"
+):
+    if values:
+        grand_margin = {}
+        for k, v in data[values].items():
+            try:
+                if isinstance(aggfunc, str):
+                    grand_margin[k] = getattr(v, aggfunc)(**kwargs)
+                elif isinstance(aggfunc, dict):
+                    if isinstance(aggfunc[k], str):
+                        grand_margin[k] = getattr(v, aggfunc[k])(**kwargs)
+                    else:
+                        grand_margin[k] = aggfunc[k](v, **kwargs)
+                else:
+                    grand_margin[k] = aggfunc(v, **kwargs)
+            except TypeError:
+                pass
+        return grand_margin
+    else:
+        return {margins_name: aggfunc(data.index, **kwargs)}
+
+
+def _generate_marginal_results(
+    table,
+    data: DataFrame,
+    values,
+    rows,
+    cols,
+    aggfunc,
+    kwargs,
+    observed: bool,
+    margins_name: Hashable = "All",
+):
+    margin_keys: list | Index
+    if len(cols) > 0:
+        # need to "interleave" the margins
+        table_pieces = []
+        margin_keys = []
+
+        def _all_key(key):
+            return (key, margins_name) + ("",) * (len(cols) - 1)
+
+        if len(rows) > 0:
+            margin = (
+                data[rows + values]
+                .groupby(rows, observed=observed)
+                .agg(aggfunc, **kwargs)
+            )
+            cat_axis = 1
+
+            for key, piece in table.T.groupby(level=0, observed=observed):
+                piece = piece.T
+                all_key = _all_key(key)
+
+                piece[all_key] = margin[key]
+
+                table_pieces.append(piece)
+                margin_keys.append(all_key)
+        else:
+            margin = (
+                data[cols[:1] + values]
+                .groupby(cols[:1], observed=observed)
+                .agg(aggfunc, **kwargs)
+                .T
+            )
+
+            cat_axis = 0
+            for key, piece in table.groupby(level=0, observed=observed):
+                if len(cols) > 1:
+                    all_key = _all_key(key)
+                else:
+                    all_key = margins_name
+                table_pieces.append(piece)
+                transformed_piece = margin[key].to_frame().T
+                if isinstance(piece.index, MultiIndex):
+                    # We are adding an empty level
+                    transformed_piece.index = MultiIndex.from_tuples(
+                        [all_key],
+                        names=piece.index.names
+                        + [
+                            None,
+                        ],
+                    )
+                else:
+                    transformed_piece.index = Index([all_key], name=piece.index.name)
+
+                # append piece for margin into table_piece
+                table_pieces.append(transformed_piece)
+                margin_keys.append(all_key)
+
+        if not table_pieces:
+            # GH 49240
+            return table
+        else:
+            result = concat(table_pieces, axis=cat_axis)
+
+        if len(rows) == 0:
+            return result
+    else:
+        result = table
+        margin_keys = table.columns
+
+    if len(cols) > 0:
+        row_margin = (
+            data[cols + values].groupby(cols, observed=observed).agg(aggfunc, **kwargs)
+        )
+        row_margin = row_margin.stack()
+
+        # GH#26568. Use names instead of indices in case of numeric names
+        new_order_indices = itertools.chain([len(cols)], range(len(cols)))
+        new_order_names = [row_margin.index.names[i] for i in new_order_indices]
+        row_margin.index = row_margin.index.reorder_levels(new_order_names)
+    else:
+        row_margin = data._constructor_sliced(np.nan, index=result.columns)
+
+    return result, margin_keys, row_margin
+
+
+def _generate_marginal_results_without_values(
+    table: DataFrame,
+    data: DataFrame,
+    rows,
+    cols,
+    aggfunc,
+    kwargs,
+    observed: bool,
+    margins_name: Hashable = "All",
+):
+    margin_keys: list | Index
+    if len(cols) > 0:
+        # need to "interleave" the margins
+        margin_keys = []
+
+        def _all_key():
+            if len(cols) == 1:
+                return margins_name
+            return (margins_name,) + ("",) * (len(cols) - 1)
+
+        if len(rows) > 0:
+            margin = data.groupby(rows, observed=observed)[rows].apply(
+                aggfunc, **kwargs
+            )
+            all_key = _all_key()
+            table[all_key] = margin
+            result = table
+            margin_keys.append(all_key)
+
+        else:
+            margin = data.groupby(level=0, observed=observed).apply(aggfunc, **kwargs)
+            all_key = _all_key()
+            table[all_key] = margin
+            result = table
+            margin_keys.append(all_key)
+            return result
+    else:
+        result = table
+        margin_keys = table.columns
+
+    if len(cols):
+        row_margin = data.groupby(cols, observed=observed)[cols].apply(
+            aggfunc, **kwargs
+        )
+    else:
+        row_margin = Series(np.nan, index=result.columns)
+
+    return result, margin_keys, row_margin
+
+
+def _convert_by(by):
+    if by is None:
+        by = []
+    elif (
+        is_scalar(by)
+        or isinstance(by, (np.ndarray, Index, ABCSeries, Grouper))
+        or callable(by)
+    ):
+        by = [by]
+    else:
+        by = list(by)
+    return by
+
+
+def pivot(
+    data: DataFrame,
+    *,
+    columns: IndexLabel,
+    index: IndexLabel | lib.NoDefault = lib.no_default,
+    values: IndexLabel | lib.NoDefault = lib.no_default,
+) -> DataFrame:
+    """
+    Return reshaped DataFrame organized by given index / column values.
+
+    Reshape data (produce a "pivot" table) based on column values. Uses
+    unique values from specified `index` / `columns` to form axes of the
+    resulting DataFrame. This function does not support data
+    aggregation, multiple values will result in a MultiIndex in the
+    columns. See the :ref:`User Guide <reshaping>` for more on reshaping.
+
+    Parameters
+    ----------
+    data : DataFrame
+        Input pandas DataFrame object.
+    columns : str or object or a list of str
+        Column to use to make new frame's columns.
+    index : str or object or a list of str, optional
+        Column to use to make new frame's index. If not given, uses existing index.
+    values : str, object or a list of the previous, optional
+        Column(s) to use for populating new frame's values. If not
+        specified, all remaining columns will be used and the result will
+        have hierarchically indexed columns.
+
+    Returns
+    -------
+    DataFrame
+        Returns reshaped DataFrame.
+
+    Raises
+    ------
+    ValueError:
+        When there are any `index`, `columns` combinations with multiple
+        values. `DataFrame.pivot_table` when you need to aggregate.
+
+    See Also
+    --------
+    DataFrame.pivot_table : Generalization of pivot that can handle
+        duplicate values for one index/column pair.
+    DataFrame.unstack : Pivot based on the index values instead of a
+        column.
+    wide_to_long : Wide panel to long format. Less flexible but more
+        user-friendly than melt.
+
+    Notes
+    -----
+    For finer-tuned control, see hierarchical indexing documentation along
+    with the related stack/unstack methods.
+
+    Reference :ref:`the user guide <reshaping.pivot>` for more examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "foo": ["one", "one", "one", "two", "two", "two"],
+    ...         "bar": ["A", "B", "C", "A", "B", "C"],
+    ...         "baz": [1, 2, 3, 4, 5, 6],
+    ...         "zoo": ["x", "y", "z", "q", "w", "t"],
+    ...     }
+    ... )
+    >>> df
+        foo   bar  baz  zoo
+    0   one   A    1    x
+    1   one   B    2    y
+    2   one   C    3    z
+    3   two   A    4    q
+    4   two   B    5    w
+    5   two   C    6    t
+
+    >>> df.pivot(index="foo", columns="bar", values="baz")
+    bar  A   B   C
+    foo
+    one  1   2   3
+    two  4   5   6
+
+    >>> df.pivot(index="foo", columns="bar")["baz"]
+    bar  A   B   C
+    foo
+    one  1   2   3
+    two  4   5   6
+
+    >>> df.pivot(index="foo", columns="bar", values=["baz", "zoo"])
+          baz       zoo
+    bar   A  B  C   A  B  C
+    foo
+    one   1  2  3   x  y  z
+    two   4  5  6   q  w  t
+
+    You could also assign a list of column names or a list of index names.
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "lev1": [1, 1, 1, 2, 2, 2],
+    ...         "lev2": [1, 1, 2, 1, 1, 2],
+    ...         "lev3": [1, 2, 1, 2, 1, 2],
+    ...         "lev4": [1, 2, 3, 4, 5, 6],
+    ...         "values": [0, 1, 2, 3, 4, 5],
+    ...     }
+    ... )
+    >>> df
+        lev1 lev2 lev3 lev4 values
+    0   1    1    1    1    0
+    1   1    1    2    2    1
+    2   1    2    1    3    2
+    3   2    1    2    4    3
+    4   2    1    1    5    4
+    5   2    2    2    6    5
+
+    >>> df.pivot(index="lev1", columns=["lev2", "lev3"], values="values")
+    lev2    1         2
+    lev3    1    2    1    2
+    lev1
+    1     0.0  1.0  2.0  NaN
+    2     4.0  3.0  NaN  5.0
+
+    >>> df.pivot(index=["lev1", "lev2"], columns=["lev3"], values="values")
+          lev3    1    2
+    lev1  lev2
+       1     1  0.0  1.0
+             2  2.0  NaN
+       2     1  4.0  3.0
+             2  NaN  5.0
+
+    A ValueError is raised if there are any duplicates.
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "foo": ["one", "one", "two", "two"],
+    ...         "bar": ["A", "A", "B", "C"],
+    ...         "baz": [1, 2, 3, 4],
+    ...     }
+    ... )
+    >>> df
+       foo bar  baz
+    0  one   A    1
+    1  one   A    2
+    2  two   B    3
+    3  two   C    4
+
+    Notice that the first two rows are the same for our `index`
+    and `columns` arguments.
+
+    >>> df.pivot(index="foo", columns="bar", values="baz")
+    Traceback (most recent call last):
+       ...
+    ValueError: Index contains duplicate entries, cannot reshape
+    """
+    columns_listlike = com.convert_to_list_like(columns)
+
+    # If columns is None we will create a MultiIndex level with None as name
+    # which might cause duplicated names because None is the default for
+    # level names
+    if any(name is None for name in data.index.names):
+        data = data.copy(deep=False)
+        data.index.names = [
+            name if name is not None else lib.no_default for name in data.index.names
+        ]
+
+    indexed: DataFrame | Series
+    if values is lib.no_default:
+        if index is not lib.no_default:
+            cols = com.convert_to_list_like(index)
+        else:
+            cols = []
+
+        append = index is lib.no_default
+        # error: Unsupported operand types for + ("List[Any]" and "ExtensionArray")
+        # error: Unsupported left operand type for + ("ExtensionArray")
+        indexed = data.set_index(
+            cols + columns_listlike,  # type: ignore[operator]
+            append=append,
+        )
+    else:
+        index_list: list[Index] | list[Series]
+        if index is lib.no_default:
+            if isinstance(data.index, MultiIndex):
+                # GH 23955
+                index_list = [
+                    data.index.get_level_values(i) for i in range(data.index.nlevels)
+                ]
+            else:
+                index_list = [
+                    data._constructor_sliced(data.index, name=data.index.name)
+                ]
+        else:
+            index_list = [data[idx] for idx in com.convert_to_list_like(index)]
+
+        data_columns = [data[col] for col in columns_listlike]
+        index_list.extend(data_columns)
+        multiindex = MultiIndex.from_arrays(index_list)
+
+        if is_list_like(values) and not isinstance(values, tuple):
+            # Exclude tuple because it is seen as a single column name
+            indexed = data._constructor(
+                data[values]._values,
+                index=multiindex,
+                columns=cast("SequenceNotStr", values),
+            )
+        else:
+            indexed = data._constructor_sliced(data[values]._values, index=multiindex)
+    # error: Argument 1 to "unstack" of "DataFrame" has incompatible type "Union
+    # [List[Any], ExtensionArray, ndarray[Any, Any], Index, Series]"; expected
+    # "Hashable"
+    # unstack with a MultiIndex returns a DataFrame
+    result = cast("DataFrame", indexed.unstack(columns_listlike))  # type: ignore[arg-type]
+    result.index.names = [
+        name if name is not lib.no_default else None for name in result.index.names
+    ]
+
+    return result
+
+
+def crosstab(
+    index,
+    columns,
+    values=None,
+    rownames=None,
+    colnames=None,
+    aggfunc=None,
+    margins: bool = False,
+    margins_name: Hashable = "All",
+    dropna: bool = True,
+    normalize: bool | Literal[0, 1, "all", "index", "columns"] = False,
+) -> DataFrame:
+    """
+    Compute a simple cross tabulation of two (or more) factors.
+
+    By default, computes a frequency table of the factors unless an
+    array of values and an aggregation function are passed.
+
+    Parameters
+    ----------
+    index : array-like, Series, or list of arrays/Series
+        Values to group by in the rows.
+    columns : array-like, Series, or list of arrays/Series
+        Values to group by in the columns.
+    values : array-like, optional
+        Array of values to aggregate according to the factors.
+        Requires `aggfunc` be specified.
+    rownames : sequence, default None
+        If passed, must match number of row arrays passed.
+    colnames : sequence, default None
+        If passed, must match number of column arrays passed.
+    aggfunc : function, optional
+        If specified, requires `values` be specified as well.
+    margins : bool, default False
+        Add row/column margins (subtotals).
+    margins_name : str, default 'All'
+        Name of the row/column that will contain the totals
+        when margins is True.
+    dropna : bool, default True
+        Do not include columns whose entries are all NaN.
+    normalize : bool, {'all', 'index', 'columns'}, or {0,1}, default False
+        Normalize by dividing all values by the sum of values.
+
+        - If passed 'all' or `True`, will normalize over all values.
+        - If passed 'index' will normalize over each row.
+        - If passed 'columns' will normalize over each column.
+        - If margins is `True`, will also normalize margin values.
+
+    Returns
+    -------
+    DataFrame
+        Cross tabulation of the data.
+
+    See Also
+    --------
+    DataFrame.pivot : Reshape data based on column values.
+    pivot_table : Create a pivot table as a DataFrame.
+
+    Notes
+    -----
+    Any Series passed will have their name attributes used unless row or column
+    names for the cross-tabulation are specified.
+
+    Any input passed containing Categorical data will have **all** of its
+    categories included in the cross-tabulation, even if the actual data does
+    not contain any instances of a particular category.
+
+    In the event that there aren't overlapping indexes an empty DataFrame will
+    be returned.
+
+    Reference :ref:`the user guide <reshaping.crosstabulations>` for more examples.
+
+    Examples
+    --------
+    >>> a = np.array(
+    ...     [
+    ...         "foo",
+    ...         "foo",
+    ...         "foo",
+    ...         "foo",
+    ...         "bar",
+    ...         "bar",
+    ...         "bar",
+    ...         "bar",
+    ...         "foo",
+    ...         "foo",
+    ...         "foo",
+    ...     ],
+    ...     dtype=object,
+    ... )
+    >>> b = np.array(
+    ...     [
+    ...         "one",
+    ...         "one",
+    ...         "one",
+    ...         "two",
+    ...         "one",
+    ...         "one",
+    ...         "one",
+    ...         "two",
+    ...         "two",
+    ...         "two",
+    ...         "one",
+    ...     ],
+    ...     dtype=object,
+    ... )
+    >>> c = np.array(
+    ...     [
+    ...         "dull",
+    ...         "dull",
+    ...         "shiny",
+    ...         "dull",
+    ...         "dull",
+    ...         "shiny",
+    ...         "shiny",
+    ...         "dull",
+    ...         "shiny",
+    ...         "shiny",
+    ...         "shiny",
+    ...     ],
+    ...     dtype=object,
+    ... )
+    >>> pd.crosstab(a, [b, c], rownames=["a"], colnames=["b", "c"])
+    b   one        two
+    c   dull shiny dull shiny
+    a
+    bar    1     2    1     0
+    foo    2     2    1     2
+
+    Here 'c' and 'f' are not represented in the data and will not be
+    shown in the output because dropna is True by default. Set
+    dropna=False to preserve categories with no data.
+
+    >>> foo = pd.Categorical(["a", "b"], categories=["a", "b", "c"])
+    >>> bar = pd.Categorical(["d", "e"], categories=["d", "e", "f"])
+    >>> pd.crosstab(foo, bar)
+    col_0  d  e
+    row_0
+    a      1  0
+    b      0  1
+    >>> pd.crosstab(foo, bar, dropna=False)
+    col_0  d  e  f
+    row_0
+    a      1  0  0
+    b      0  1  0
+    c      0  0  0
+    """
+    if values is None and aggfunc is not None:
+        raise ValueError("aggfunc cannot be used without values.")
+
+    if values is not None and aggfunc is None:
+        raise ValueError("values cannot be used without an aggfunc.")
+
+    if not is_nested_list_like(index):
+        index = [index]
+    if not is_nested_list_like(columns):
+        columns = [columns]
+
+    common_idx = None
+    pass_objs = [x for x in index + columns if isinstance(x, (ABCSeries, ABCDataFrame))]
+    if pass_objs:
+        common_idx = get_objs_combined_axis(pass_objs, intersect=True, sort=False)
+
+    rownames = _get_names(index, rownames, prefix="row")
+    colnames = _get_names(columns, colnames, prefix="col")
+
+    # duplicate names mapped to unique names for pivot op
+    (
+        rownames_mapper,
+        unique_rownames,
+        colnames_mapper,
+        unique_colnames,
+    ) = _build_names_mapper(rownames, colnames)
+
+    from pandas import DataFrame
+
+    data = {
+        **dict(zip(unique_rownames, index)),
+        **dict(zip(unique_colnames, columns)),
+    }
+    df = DataFrame(data, index=common_idx)
+
+    if values is None:
+        df["__dummy__"] = 0
+        kwargs = {"aggfunc": len, "fill_value": 0}
+    else:
+        df["__dummy__"] = values
+        kwargs = {"aggfunc": aggfunc}
+
+    # error: Argument 7 to "pivot_table" of "DataFrame" has incompatible type
+    # "**Dict[str, object]"; expected "Union[...]"
+    table = df.pivot_table(
+        "__dummy__",
+        index=unique_rownames,
+        columns=unique_colnames,
+        margins=margins,
+        margins_name=margins_name,
+        dropna=dropna,
+        observed=dropna,
+        **kwargs,  # type: ignore[arg-type]
+    )
+
+    # Post-process
+    if normalize is not False:
+        table = _normalize(
+            table, normalize=normalize, margins=margins, margins_name=margins_name
+        )
+
+    table = table.rename_axis(index=rownames_mapper, axis=0)
+    table = table.rename_axis(columns=colnames_mapper, axis=1)
+
+    return table
+
+
+def _normalize(
+    table: DataFrame, normalize, margins: bool, margins_name: Hashable = "All"
+) -> DataFrame:
+    if not isinstance(normalize, (bool, str)):
+        axis_subs = {0: "index", 1: "columns"}
+        try:
+            normalize = axis_subs[normalize]
+        except KeyError as err:
+            raise ValueError("Not a valid normalize argument") from err
+
+    if margins is False:
+        # Actual Normalizations
+        normalizers: dict[bool | str, Callable] = {
+            "all": lambda x: x / x.sum(axis=1).sum(axis=0),
+            "columns": lambda x: x / x.sum(),
+            "index": lambda x: x.div(x.sum(axis=1), axis=0),
+        }
+
+        normalizers[True] = normalizers["all"]
+
+        try:
+            f = normalizers[normalize]
+        except KeyError as err:
+            raise ValueError("Not a valid normalize argument") from err
+
+        table = f(table)
+        table = table.fillna(0)
+
+    elif margins is True:
+        # keep index and column of pivoted table
+        table_index = table.index
+        table_columns = table.columns
+        last_ind_or_col = table.iloc[-1, :].name
+
+        # check if margin name is not in (for MI cases) and not equal to last
+        # index/column and save the column and index margin
+        if (margins_name not in last_ind_or_col) & (margins_name != last_ind_or_col):
+            raise ValueError(f"{margins_name} not in pivoted DataFrame")
+        column_margin = table.iloc[:-1, -1]
+        index_margin = table.iloc[-1, :-1]
+
+        # keep the core table
+        table = table.iloc[:-1, :-1]
+
+        # Normalize core
+        table = _normalize(table, normalize=normalize, margins=False)
+
+        # Fix Margins
+        if normalize == "columns":
+            column_margin = column_margin / column_margin.sum()
+            table = concat([table, column_margin], axis=1)
+            table = table.fillna(0)
+            table.columns = table_columns
+
+        elif normalize == "index":
+            index_margin = index_margin / index_margin.sum()
+            table = table._append(index_margin, ignore_index=True)
+            table = table.fillna(0)
+            table.index = table_index
+
+        elif normalize == "all" or normalize is True:
+            column_margin = column_margin / column_margin.sum()
+            index_margin = index_margin / index_margin.sum()
+            index_margin.loc[margins_name] = 1
+            table = concat([table, column_margin], axis=1)
+            table = table._append(index_margin, ignore_index=True)
+
+            table = table.fillna(0)
+            table.index = table_index
+            table.columns = table_columns
+
+        else:
+            raise ValueError("Not a valid normalize argument")
+
+    else:
+        raise ValueError("Not a valid margins argument")
+
+    return table
+
+
+def _get_names(arrs, names, prefix: str = "row") -> list:
+    if names is None:
+        names = []
+        for i, arr in enumerate(arrs):
+            if isinstance(arr, ABCSeries) and arr.name is not None:
+                names.append(arr.name)
+            else:
+                names.append(f"{prefix}_{i}")
+    else:
+        if len(names) != len(arrs):
+            raise AssertionError("arrays and names must have the same length")
+        if not isinstance(names, list):
+            names = list(names)
+
+    return names
+
+
+def _build_names_mapper(
+    rownames: list[str], colnames: list[str]
+) -> tuple[dict[str, str], list[str], dict[str, str], list[str]]:
+    """
+    Given the names of a DataFrame's rows and columns, returns a set of unique row
+    and column names and mappers that convert to original names.
+
+    A row or column name is replaced if it is duplicate among the rows of the inputs,
+    among the columns of the inputs or between the rows and the columns.
+
+    Parameters
+    ----------
+    rownames: list[str]
+    colnames: list[str]
+
+    Returns
+    -------
+    Tuple(Dict[str, str], List[str], Dict[str, str], List[str])
+
+    rownames_mapper: dict[str, str]
+        a dictionary with new row names as keys and original rownames as values
+    unique_rownames: list[str]
+        a list of rownames with duplicate names replaced by dummy names
+    colnames_mapper: dict[str, str]
+        a dictionary with new column names as keys and original column names as values
+    unique_colnames: list[str]
+        a list of column names with duplicate names replaced by dummy names
+
+    """
+    dup_names = set(rownames) | set(colnames)
+
+    rownames_mapper = {
+        f"row_{i}": name for i, name in enumerate(rownames) if name in dup_names
+    }
+    unique_rownames = [
+        f"row_{i}" if name in dup_names else name for i, name in enumerate(rownames)
+    ]
+
+    colnames_mapper = {
+        f"col_{i}": name for i, name in enumerate(colnames) if name in dup_names
+    }
+    unique_colnames = [
+        f"col_{i}" if name in dup_names else name for i, name in enumerate(colnames)
+    ]
+
+    return rownames_mapper, unique_rownames, colnames_mapper, unique_colnames
diff --git a/pandas/pandas/core/reshape/reshape.py b/pandas/pandas/core/reshape/reshape.py
new file mode 100644
index 0000000000000..c60fe71a7ff28
--- /dev/null
+++ b/pandas/pandas/core/reshape/reshape.py
@@ -0,0 +1,1070 @@
+from __future__ import annotations
+
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config.config import get_option
+
+import pandas._libs.reshape as libreshape
+from pandas.errors import PerformanceWarning
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.cast import (
+    find_common_type,
+    maybe_promote,
+)
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    is_1d_only_ea_dtype,
+    is_integer,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.missing import notna
+
+import pandas.core.algorithms as algos
+from pandas.core.algorithms import (
+    factorize,
+    unique,
+)
+from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
+from pandas.core.arrays.categorical import factorize_from_iterable
+from pandas.core.construction import ensure_wrapped_if_datetimelike
+from pandas.core.frame import DataFrame
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    default_index,
+)
+from pandas.core.reshape.concat import concat
+from pandas.core.series import Series
+from pandas.core.sorting import (
+    compress_group_index,
+    decons_obs_group_ids,
+    get_compressed_ids,
+    get_group_index,
+    get_group_index_sorter,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Level,
+        npt,
+    )
+
+    from pandas.core.arrays import ExtensionArray
+    from pandas.core.indexes.frozen import FrozenList
+
+
+class _Unstacker:
+    """
+    Helper class to unstack data / pivot with multi-level index
+
+    Parameters
+    ----------
+    index : MultiIndex
+    level : int or str, default last level
+        Level to "unstack". Accepts a name for the level.
+    fill_value : scalar, optional
+        Default value to fill in missing values if subgroups do not have the
+        same set of labels. By default, missing values will be replaced with
+        the default fill value for that data type, NaN for float, NaT for
+        datetimelike, etc. For integer types, by default data will converted to
+        float and missing values will be set to NaN.
+    constructor : object
+        Pandas ``DataFrame`` or subclass used to create unstacked
+        response.  If None, DataFrame will be used.
+
+    Examples
+    --------
+    >>> index = pd.MultiIndex.from_tuples(
+    ...     [("one", "a"), ("one", "b"), ("two", "a"), ("two", "b")]
+    ... )
+    >>> s = pd.Series(np.arange(1, 5, dtype=np.int64), index=index)
+    >>> s
+    one  a    1
+         b    2
+    two  a    3
+         b    4
+    dtype: int64
+
+    >>> s.unstack(level=-1)
+         a  b
+    one  1  2
+    two  3  4
+
+    >>> s.unstack(level=0)
+       one  two
+    a    1    3
+    b    2    4
+
+    Returns
+    -------
+    unstacked : DataFrame
+    """
+
+    def __init__(
+        self, index: MultiIndex, level: Level, constructor, sort: bool = True
+    ) -> None:
+        self.constructor = constructor
+        self.sort = sort
+
+        self.index = index.remove_unused_levels()
+
+        self.level = self.index._get_level_number(level)
+
+        # when index includes `nan`, need to lift levels/strides by 1
+        self.lift = 1 if -1 in self.index.codes[self.level] else 0
+
+        # Note: the "pop" below alters these in-place.
+        self.new_index_levels = list(self.index.levels)
+        self.new_index_names = list(self.index.names)
+
+        self.removed_name = self.new_index_names.pop(self.level)
+        self.removed_level = self.new_index_levels.pop(self.level)
+        self.removed_level_full = index.levels[self.level]
+        if not self.sort:
+            unique_codes = unique(self.index.codes[self.level])
+            self.removed_level = self.removed_level.take(unique_codes)
+            self.removed_level_full = self.removed_level_full.take(unique_codes)
+
+        if get_option("performance_warnings"):
+            # Bug fix GH 20601
+            # If the data frame is too big, the number of unique index combination
+            # will cause int32 overflow on windows environments.
+            # We want to check and raise an warning before this happens
+            num_rows = max(index_level.size for index_level in self.new_index_levels)
+            num_columns = self.removed_level.size
+
+            # GH20601: This forces an overflow if the number of cells is too high.
+            # GH 26314: Previous ValueError raised was too restrictive for many users.
+            num_cells = num_rows * num_columns
+            if num_cells > np.iinfo(np.int32).max:
+                warnings.warn(
+                    f"The following operation may generate {num_cells} cells "
+                    f"in the resulting pandas object.",
+                    PerformanceWarning,
+                    stacklevel=find_stack_level(),
+                )
+
+        self._make_selectors()
+
+    @cache_readonly
+    def _indexer_and_to_sort(
+        self,
+    ) -> tuple[
+        npt.NDArray[np.intp],
+        list[np.ndarray],  # each has _some_ signed integer dtype
+    ]:
+        v = self.level
+
+        codes = list(self.index.codes)
+        if not self.sort:
+            # Create new codes considering that labels are already sorted
+            codes = [factorize(code)[0] for code in codes]
+        levs = list(self.index.levels)
+        to_sort = codes[:v] + codes[v + 1 :] + [codes[v]]
+        sizes = tuple(len(x) for x in levs[:v] + levs[v + 1 :] + [levs[v]])
+
+        comp_index, obs_ids = get_compressed_ids(to_sort, sizes)
+        ngroups = len(obs_ids)
+
+        indexer = get_group_index_sorter(comp_index, ngroups)
+        return indexer, to_sort
+
+    @cache_readonly
+    def sorted_labels(self) -> list[np.ndarray]:
+        indexer, to_sort = self._indexer_and_to_sort
+        if self.sort:
+            return [line.take(indexer) for line in to_sort]
+        return to_sort
+
+    def _make_sorted_values(self, values: np.ndarray) -> np.ndarray:
+        indexer, _ = self._indexer_and_to_sort
+        sorted_values = algos.take_nd(values, indexer, axis=0)
+        return sorted_values
+
+    def _make_selectors(self) -> None:
+        new_levels = self.new_index_levels
+
+        # make the mask
+        remaining_labels = self.sorted_labels[:-1]
+        level_sizes = tuple(len(x) for x in new_levels)
+
+        comp_index, obs_ids = get_compressed_ids(remaining_labels, level_sizes)
+        ngroups = len(obs_ids)
+
+        comp_index = ensure_platform_int(comp_index)
+        stride = self.index.levshape[self.level] + self.lift
+        self.full_shape = ngroups, stride
+
+        selector = self.sorted_labels[-1] + stride * comp_index + self.lift
+        mask = np.zeros(np.prod(self.full_shape), dtype=bool)
+        mask.put(selector, True)
+
+        if mask.sum() < len(self.index):
+            raise ValueError("Index contains duplicate entries, cannot reshape")
+
+        self.group_index = comp_index
+        self.mask = mask
+        if self.sort:
+            self.compressor = comp_index.searchsorted(np.arange(ngroups))
+        else:
+            self.compressor = np.sort(np.unique(comp_index, return_index=True)[1])
+
+    @cache_readonly
+    def mask_all(self) -> bool:
+        return bool(self.mask.all())
+
+    @cache_readonly
+    def arange_result(self) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.bool_]]:
+        # We cache this for reuse in ExtensionBlock._unstack
+        dummy_arr = np.arange(len(self.index), dtype=np.intp)
+        new_values, mask = self.get_new_values(dummy_arr, fill_value=-1)
+        return new_values, mask.any(0)
+        # TODO: in all tests we have mask.any(0).all(); can we rely on that?
+
+    def get_result(self, obj, value_columns, fill_value) -> DataFrame:
+        values = obj._values
+        if values.ndim == 1:
+            values = values[:, np.newaxis]
+
+        if value_columns is None and values.shape[1] != 1:  # pragma: no cover
+            raise ValueError("must pass column labels for multi-column data")
+
+        new_values, _ = self.get_new_values(values, fill_value)
+        columns = self.get_new_columns(value_columns)
+        index = self.new_index
+
+        result = self.constructor(
+            new_values, index=index, columns=columns, dtype=new_values.dtype, copy=False
+        )
+        if isinstance(values, np.ndarray):
+            base, new_base = values.base, new_values.base
+        elif isinstance(values, NDArrayBackedExtensionArray):
+            base, new_base = values._ndarray.base, new_values._ndarray.base
+        else:
+            base, new_base = 1, 2  # type: ignore[assignment]
+        if base is new_base:
+            # We can only get here if one of the dimensions is size 1
+            result._mgr.add_references(obj._mgr)
+        return result
+
+    def get_new_values(self, values, fill_value=None):
+        if values.ndim == 1:
+            values = values[:, np.newaxis]
+
+        sorted_values = self._make_sorted_values(values)
+
+        # place the values
+        length, width = self.full_shape
+        stride = values.shape[1]
+        result_width = width * stride
+        result_shape = (length, result_width)
+        mask = self.mask
+        mask_all = self.mask_all
+
+        # we can simply reshape if we don't have a mask
+        if mask_all and len(values):
+            # TODO: Under what circumstances can we rely on sorted_values
+            #  matching values?  When that holds, we can slice instead
+            #  of take (in particular for EAs)
+            new_values = (
+                sorted_values.reshape(length, width, stride)
+                .swapaxes(1, 2)
+                .reshape(result_shape)
+            )
+            new_mask = np.ones(result_shape, dtype=bool)
+            return new_values, new_mask
+
+        dtype = values.dtype
+
+        if isinstance(dtype, ExtensionDtype):
+            # GH#41875
+            # We are assuming that fill_value can be held by this dtype,
+            #  unlike the non-EA case that promotes.
+            cls = dtype.construct_array_type()
+            new_values = cls._empty(result_shape, dtype=dtype)
+            if not mask_all:
+                new_values[:] = fill_value
+        else:
+            if not mask_all:
+                dtype, fill_value = maybe_promote(dtype, fill_value)
+            new_values = np.empty(result_shape, dtype=dtype)
+            if not mask_all:
+                new_values.fill(fill_value)
+
+        name = dtype.name
+        new_mask = np.zeros(result_shape, dtype=bool)
+
+        # we need to convert to a basic dtype
+        # and possibly coerce an input to our output dtype
+        # e.g. ints -> floats
+        if needs_i8_conversion(values.dtype):
+            sorted_values = sorted_values.view("i8")
+            new_values = new_values.view("i8")
+        else:
+            sorted_values = sorted_values.astype(name, copy=False)
+
+        # fill in our values & mask
+        libreshape.unstack(
+            sorted_values,
+            mask.view("u1"),
+            stride,
+            length,
+            width,
+            new_values,
+            new_mask.view("u1"),
+        )
+
+        # reconstruct dtype if needed
+        if needs_i8_conversion(values.dtype):
+            # view as datetime64 so we can wrap in DatetimeArray and use
+            #  DTA's view method
+            new_values = new_values.view("M8[ns]")
+            new_values = ensure_wrapped_if_datetimelike(new_values)
+            new_values = new_values.view(values.dtype)
+
+        return new_values, new_mask
+
+    def get_new_columns(self, value_columns: Index | None):
+        if value_columns is None:
+            if self.lift == 0:
+                return self.removed_level._rename(name=self.removed_name)
+
+            lev = self.removed_level.insert(0, item=self.removed_level._na_value)
+            return lev.rename(self.removed_name)
+
+        stride = len(self.removed_level) + self.lift
+        width = len(value_columns)
+        propagator = np.repeat(np.arange(width), stride)
+
+        new_levels: FrozenList | list[Index]
+
+        if isinstance(value_columns, MultiIndex):
+            # error: Cannot determine type of "__add__"  [has-type]
+            new_levels = value_columns.levels + (  # type: ignore[has-type]
+                self.removed_level_full,
+            )
+            new_names = value_columns.names + (self.removed_name,)
+
+            new_codes = [lab.take(propagator) for lab in value_columns.codes]
+        else:
+            new_levels = [
+                value_columns,
+                self.removed_level_full,
+            ]
+            new_names = [value_columns.name, self.removed_name]
+            new_codes = [propagator]
+
+        repeater = self._repeater
+
+        # The entire level is then just a repetition of the single chunk:
+        new_codes.append(np.tile(repeater, width))
+        return MultiIndex(
+            levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+        )
+
+    @cache_readonly
+    def _repeater(self) -> np.ndarray:
+        # The two indices differ only if the unstacked level had unused items:
+        if len(self.removed_level_full) != len(self.removed_level):
+            # In this case, we remap the new codes to the original level:
+            repeater = self.removed_level_full.get_indexer(self.removed_level)
+            if self.lift:
+                repeater = np.insert(repeater, 0, -1)
+        else:
+            # Otherwise, we just use each level item exactly once:
+            stride = len(self.removed_level) + self.lift
+            repeater = np.arange(stride) - self.lift
+
+        return repeater
+
+    @cache_readonly
+    def new_index(self) -> MultiIndex | Index:
+        # Does not depend on values or value_columns
+        if self.sort:
+            labels = self.sorted_labels[:-1]
+        else:
+            v = self.level
+            codes = list(self.index.codes)
+            labels = codes[:v] + codes[v + 1 :]
+        result_codes = [lab.take(self.compressor) for lab in labels]
+
+        # construct the new index
+        if len(self.new_index_levels) == 1:
+            level, level_codes = self.new_index_levels[0], result_codes[0]
+            if (level_codes == -1).any():
+                level = level.insert(len(level), level._na_value)
+            return level.take(level_codes).rename(self.new_index_names[0])
+
+        return MultiIndex(
+            levels=self.new_index_levels,
+            codes=result_codes,
+            names=self.new_index_names,
+            verify_integrity=False,
+        )
+
+
+def _unstack_multiple(
+    data: Series | DataFrame, clocs, fill_value=None, sort: bool = True
+):
+    if len(clocs) == 0:
+        return data
+
+    # NOTE: This doesn't deal with hierarchical columns yet
+
+    index = data.index
+    index = cast(MultiIndex, index)  # caller is responsible for checking
+
+    # GH 19966 Make sure if MultiIndexed index has tuple name, they will be
+    # recognised as a whole
+    if clocs in index.names:
+        clocs = [clocs]
+    clocs = [index._get_level_number(i) for i in clocs]
+
+    rlocs = [i for i in range(index.nlevels) if i not in clocs]
+
+    clevels = [index.levels[i] for i in clocs]
+    ccodes = [index.codes[i] for i in clocs]
+    cnames = [index.names[i] for i in clocs]
+    rlevels = [index.levels[i] for i in rlocs]
+    rcodes = [index.codes[i] for i in rlocs]
+    rnames = [index.names[i] for i in rlocs]
+
+    shape = tuple(len(x) for x in clevels)
+    group_index = get_group_index(ccodes, shape, sort=False, xnull=False)
+
+    comp_ids, obs_ids = compress_group_index(group_index, sort=False)
+    recons_codes = decons_obs_group_ids(comp_ids, obs_ids, shape, ccodes, xnull=False)
+
+    if not rlocs:
+        # Everything is in clocs, so the dummy df has a regular index
+        dummy_index = Index(obs_ids, name="__placeholder__")
+    else:
+        dummy_index = MultiIndex(
+            levels=rlevels + [obs_ids],
+            codes=rcodes + [comp_ids],
+            names=rnames + ["__placeholder__"],
+            verify_integrity=False,
+        )
+
+    if isinstance(data, Series):
+        dummy = data.copy(deep=False)
+        dummy.index = dummy_index
+
+        unstacked = dummy.unstack("__placeholder__", fill_value=fill_value, sort=sort)
+        new_levels = clevels
+        new_names = cnames
+        new_codes = recons_codes
+    else:
+        if isinstance(data.columns, MultiIndex):
+            result = data
+            while clocs:
+                val = clocs.pop(0)
+                # error: Incompatible types in assignment (expression has type
+                # "DataFrame | Series", variable has type "DataFrame")
+                result = result.unstack(  # type: ignore[assignment]
+                    val, fill_value=fill_value, sort=sort
+                )
+                clocs = [v if v < val else v - 1 for v in clocs]
+
+            return result
+
+        # GH#42579 deep=False to avoid consolidating
+        dummy_df = data.copy(deep=False)
+        dummy_df.index = dummy_index
+
+        # error: Incompatible types in assignment (expression has type "DataFrame |
+        # Series", variable has type "DataFrame")
+        unstacked = dummy_df.unstack(  # type: ignore[assignment]
+            "__placeholder__", fill_value=fill_value, sort=sort
+        )
+        if isinstance(unstacked, Series):
+            unstcols = unstacked.index
+        else:
+            unstcols = unstacked.columns
+        assert isinstance(unstcols, MultiIndex)  # for mypy
+        new_levels = [unstcols.levels[0]] + clevels
+        new_names = [data.columns.name] + cnames
+
+        new_codes = [unstcols.codes[0]]
+        new_codes.extend(rec.take(unstcols.codes[-1]) for rec in recons_codes)
+
+    new_columns = MultiIndex(
+        levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+    )
+
+    if isinstance(unstacked, Series):
+        unstacked.index = new_columns
+    else:
+        unstacked.columns = new_columns
+
+    return unstacked
+
+
+@overload
+def unstack(obj: Series, level, fill_value=..., sort: bool = ...) -> DataFrame: ...
+
+
+@overload
+def unstack(
+    obj: Series | DataFrame, level, fill_value=..., sort: bool = ...
+) -> Series | DataFrame: ...
+
+
+def unstack(
+    obj: Series | DataFrame, level, fill_value=None, sort: bool = True
+) -> Series | DataFrame:
+    if isinstance(level, (tuple, list)):
+        if len(level) != 1:
+            # _unstack_multiple only handles MultiIndexes,
+            # and isn't needed for a single level
+            return _unstack_multiple(obj, level, fill_value=fill_value, sort=sort)
+        else:
+            level = level[0]
+
+    if not is_integer(level) and not level == "__placeholder__":
+        # check if level is valid in case of regular index
+        obj.index._get_level_number(level)
+
+    if isinstance(obj, DataFrame):
+        if isinstance(obj.index, MultiIndex):
+            return _unstack_frame(obj, level, fill_value=fill_value, sort=sort)
+        else:
+            return obj.T.stack()
+    elif not isinstance(obj.index, MultiIndex):
+        # GH 36113
+        # Give nicer error messages when unstack a Series whose
+        # Index is not a MultiIndex.
+        raise ValueError(
+            f"index must be a MultiIndex to unstack, {type(obj.index)} was passed"
+        )
+    else:
+        if is_1d_only_ea_dtype(obj.dtype):
+            return _unstack_extension_series(obj, level, fill_value, sort=sort)
+        unstacker = _Unstacker(
+            obj.index, level=level, constructor=obj._constructor_expanddim, sort=sort
+        )
+        return unstacker.get_result(obj, value_columns=None, fill_value=fill_value)
+
+
+def _unstack_frame(
+    obj: DataFrame, level, fill_value=None, sort: bool = True
+) -> DataFrame:
+    assert isinstance(obj.index, MultiIndex)  # checked by caller
+    unstacker = _Unstacker(
+        obj.index, level=level, constructor=obj._constructor, sort=sort
+    )
+
+    if not obj._can_fast_transpose:
+        mgr = obj._mgr.unstack(unstacker, fill_value=fill_value)
+        return obj._constructor_from_mgr(mgr, axes=mgr.axes)
+    else:
+        return unstacker.get_result(
+            obj, value_columns=obj.columns, fill_value=fill_value
+        )
+
+
+def _unstack_extension_series(
+    series: Series, level, fill_value, sort: bool
+) -> DataFrame:
+    """
+    Unstack an ExtensionArray-backed Series.
+
+    The ExtensionDtype is preserved.
+
+    Parameters
+    ----------
+    series : Series
+        A Series with an ExtensionArray for values
+    level : Any
+        The level name or number.
+    fill_value : Any
+        The user-level (not physical storage) fill value to use for
+        missing values introduced by the reshape. Passed to
+        ``series.values.take``.
+    sort : bool
+        Whether to sort the resulting MuliIndex levels
+
+    Returns
+    -------
+    DataFrame
+        Each column of the DataFrame will have the same dtype as
+        the input Series.
+    """
+    # Defer to the logic in ExtensionBlock._unstack
+    df = series.to_frame()
+    result = df.unstack(level=level, fill_value=fill_value, sort=sort)
+
+    # equiv: result.droplevel(level=0, axis=1)
+    #  but this avoids an extra copy
+    result.columns = result.columns._drop_level_numbers([0])
+    # error: Incompatible return value type (got "DataFrame | Series", expected
+    # "DataFrame")
+    return result  # type: ignore[return-value]
+
+
+def stack(
+    frame: DataFrame, level=-1, dropna: bool = True, sort: bool = True
+) -> Series | DataFrame:
+    """
+    Convert DataFrame to Series with multi-level Index. Columns become the
+    second level of the resulting hierarchical index
+
+    Returns
+    -------
+    stacked : Series or DataFrame
+    """
+
+    def stack_factorize(index):
+        if index.is_unique:
+            return index, np.arange(len(index))
+        codes, categories = factorize_from_iterable(index)
+        return categories, codes
+
+    N, K = frame.shape
+
+    # Will also convert negative level numbers and check if out of bounds.
+    level_num = frame.columns._get_level_number(level)
+
+    if isinstance(frame.columns, MultiIndex):
+        return _stack_multi_columns(
+            frame, level_num=level_num, dropna=dropna, sort=sort
+        )
+    elif isinstance(frame.index, MultiIndex):
+        new_levels = list(frame.index.levels)
+        new_codes = [lab.repeat(K) for lab in frame.index.codes]
+
+        clev, clab = stack_factorize(frame.columns)
+        new_levels.append(clev)
+        new_codes.append(np.tile(clab, N).ravel())
+
+        new_names = list(frame.index.names)
+        new_names.append(frame.columns.name)
+        new_index = MultiIndex(
+            levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+        )
+    else:
+        levels, (ilab, clab) = zip(*map(stack_factorize, (frame.index, frame.columns)))
+        codes = ilab.repeat(K), np.tile(clab, N).ravel()
+        new_index = MultiIndex(
+            levels=levels,
+            codes=codes,
+            names=[frame.index.name, frame.columns.name],
+            verify_integrity=False,
+        )
+
+    new_values: ArrayLike
+    if not frame.empty and frame._is_homogeneous_type:
+        # For homogeneous EAs, frame._values will coerce to object. So
+        # we concatenate instead.
+        dtypes = list(frame.dtypes._values)
+        dtype = dtypes[0]
+
+        if isinstance(dtype, ExtensionDtype):
+            arr = dtype.construct_array_type()
+            new_values = arr._concat_same_type(
+                [col._values for _, col in frame.items()]
+            )
+            new_values = _reorder_for_extension_array_stack(new_values, N, K)
+        else:
+            # homogeneous, non-EA
+            new_values = frame._values.ravel()
+
+    else:
+        # non-homogeneous
+        new_values = frame._values.ravel()
+
+    if dropna:
+        mask = notna(new_values)
+        new_values = new_values[mask]
+        new_index = new_index[mask]
+
+    return frame._constructor_sliced(new_values, index=new_index)
+
+
+def stack_multiple(frame: DataFrame, level, dropna: bool = True, sort: bool = True):
+    # If all passed levels match up to column names, no
+    # ambiguity about what to do
+    if all(lev in frame.columns.names for lev in level):
+        result = frame
+        for lev in level:
+            # error: Incompatible types in assignment (expression has type
+            # "Series | DataFrame", variable has type "DataFrame")
+            result = stack(result, lev, dropna=dropna, sort=sort)  # type: ignore[assignment]
+
+    # Otherwise, level numbers may change as each successive level is stacked
+    elif all(isinstance(lev, int) for lev in level):
+        # As each stack is done, the level numbers decrease, so we need
+        #  to account for that when level is a sequence of ints
+        result = frame
+        # _get_level_number() checks level numbers are in range and converts
+        # negative numbers to positive
+        level = [frame.columns._get_level_number(lev) for lev in level]
+
+        while level:
+            lev = level.pop(0)
+            # error: Incompatible types in assignment (expression has type
+            # "Series | DataFrame", variable has type "DataFrame")
+            result = stack(result, lev, dropna=dropna, sort=sort)  # type: ignore[assignment]
+            # Decrement all level numbers greater than current, as these
+            # have now shifted down by one
+            level = [v if v <= lev else v - 1 for v in level]
+
+    else:
+        raise ValueError(
+            "level should contain all level names or all level "
+            "numbers, not a mixture of the two."
+        )
+
+    return result
+
+
+def _stack_multi_column_index(columns: MultiIndex) -> MultiIndex | Index:
+    """Creates a MultiIndex from the first N-1 levels of this MultiIndex."""
+    if len(columns.levels) <= 2:
+        return columns.levels[0]._rename(name=columns.names[0])
+
+    levs = (
+        [lev[c] if c >= 0 else None for c in codes]
+        for lev, codes in zip(columns.levels[:-1], columns.codes[:-1])
+    )
+
+    # Remove duplicate tuples in the MultiIndex.
+    tuples = zip(*levs)
+    unique_tuples = (key for key, _ in itertools.groupby(tuples))
+    new_levs = zip(*unique_tuples)
+
+    # The dtype of each level must be explicitly set to avoid inferring the wrong type.
+    # See GH-36991.
+    return MultiIndex.from_arrays(
+        [
+            # Not all indices can accept None values.
+            Index(new_lev, dtype=lev.dtype) if None not in new_lev else new_lev
+            for new_lev, lev in zip(new_levs, columns.levels)
+        ],
+        names=columns.names[:-1],
+    )
+
+
+def _stack_multi_columns(
+    frame: DataFrame, level_num: int = -1, dropna: bool = True, sort: bool = True
+) -> DataFrame:
+    def _convert_level_number(level_num: int, columns: Index):
+        """
+        Logic for converting the level number to something we can safely pass
+        to swaplevel.
+
+        If `level_num` matches a column name return the name from
+        position `level_num`, otherwise return `level_num`.
+        """
+        if level_num in columns.names:
+            return columns.names[level_num]
+
+        return level_num
+
+    this = frame.copy(deep=False)
+    mi_cols = this.columns  # cast(MultiIndex, this.columns)
+    assert isinstance(mi_cols, MultiIndex)  # caller is responsible
+
+    # this makes life much simpler
+    if level_num != mi_cols.nlevels - 1:
+        # roll levels to put selected level at end
+        roll_columns = mi_cols
+        for i in range(level_num, mi_cols.nlevels - 1):
+            # Need to check if the ints conflict with level names
+            lev1 = _convert_level_number(i, roll_columns)
+            lev2 = _convert_level_number(i + 1, roll_columns)
+            roll_columns = roll_columns.swaplevel(lev1, lev2)
+        this.columns = mi_cols = roll_columns
+
+    if not mi_cols._is_lexsorted() and sort:
+        # Workaround the edge case where 0 is one of the column names,
+        # which interferes with trying to sort based on the first
+        # level
+        level_to_sort = _convert_level_number(0, mi_cols)
+        this = this.sort_index(level=level_to_sort, axis=1)
+        mi_cols = this.columns
+
+    mi_cols = cast(MultiIndex, mi_cols)
+    new_columns = _stack_multi_column_index(mi_cols)
+
+    # time to ravel the values
+    new_data = {}
+    level_vals = mi_cols.levels[-1]
+    level_codes = unique(mi_cols.codes[-1])
+    if sort:
+        level_codes = np.sort(level_codes)
+    level_vals_nan = level_vals.insert(len(level_vals), None)
+
+    level_vals_used = np.take(level_vals_nan, level_codes)
+    levsize = len(level_codes)
+    drop_cols = []
+    for key in new_columns:
+        try:
+            loc = this.columns.get_loc(key)
+        except KeyError:
+            drop_cols.append(key)
+            continue
+
+        # can make more efficient?
+        # we almost always return a slice
+        # but if unsorted can get a boolean
+        # indexer
+        if not isinstance(loc, slice):
+            slice_len = len(loc)
+        else:
+            slice_len = loc.stop - loc.start
+
+        if slice_len != levsize:
+            chunk = this.loc[:, this.columns[loc]]
+            chunk.columns = level_vals_nan.take(chunk.columns.codes[-1])
+            value_slice = chunk.reindex(columns=level_vals_used).values
+        else:
+            subset = this.iloc[:, loc]
+            dtype = find_common_type(subset.dtypes.tolist())
+            if isinstance(dtype, ExtensionDtype):
+                # TODO(EA2D): won't need special case, can go through .values
+                #  paths below (might change to ._values)
+                value_slice = dtype.construct_array_type()._concat_same_type(
+                    [x._values.astype(dtype, copy=False) for _, x in subset.items()]
+                )
+                N, K = subset.shape
+                idx = np.arange(N * K).reshape(K, N).T.reshape(-1)
+                value_slice = value_slice.take(idx)
+            else:
+                value_slice = subset.values
+
+        if value_slice.ndim > 1:
+            # i.e. not extension
+            value_slice = value_slice.ravel()
+
+        new_data[key] = value_slice
+
+    if len(drop_cols) > 0:
+        new_columns = new_columns.difference(drop_cols)
+
+    N = len(this)
+
+    if isinstance(this.index, MultiIndex):
+        new_levels = list(this.index.levels)
+        new_names = list(this.index.names)
+        new_codes = [lab.repeat(levsize) for lab in this.index.codes]
+    else:
+        old_codes, old_levels = factorize_from_iterable(this.index)
+        new_levels = [old_levels]
+        new_codes = [old_codes.repeat(levsize)]
+        new_names = [this.index.name]  # something better?
+
+    new_levels.append(level_vals)
+    new_codes.append(np.tile(level_codes, N))
+    new_names.append(frame.columns.names[level_num])
+
+    new_index = MultiIndex(
+        levels=new_levels, codes=new_codes, names=new_names, verify_integrity=False
+    )
+
+    result = frame._constructor(new_data, index=new_index, columns=new_columns)
+
+    if frame.columns.nlevels > 1:
+        desired_columns = frame.columns._drop_level_numbers([level_num]).unique()
+        if not result.columns.equals(desired_columns):
+            result = result[desired_columns]
+
+    # more efficient way to go about this? can do the whole masking biz but
+    # will only save a small amount of time...
+    if dropna:
+        result = result.dropna(axis=0, how="all")
+
+    return result
+
+
+def _reorder_for_extension_array_stack(
+    arr: ExtensionArray, n_rows: int, n_columns: int
+) -> ExtensionArray:
+    """
+    Re-orders the values when stacking multiple extension-arrays.
+
+    The indirect stacking method used for EAs requires a followup
+    take to get the order correct.
+
+    Parameters
+    ----------
+    arr : ExtensionArray
+    n_rows, n_columns : int
+        The number of rows and columns in the original DataFrame.
+
+    Returns
+    -------
+    taken : ExtensionArray
+        The original `arr` with elements re-ordered appropriately
+
+    Examples
+    --------
+    >>> arr = np.array(["a", "b", "c", "d", "e", "f"])
+    >>> _reorder_for_extension_array_stack(arr, 2, 3)
+    array(['a', 'c', 'e', 'b', 'd', 'f'], dtype='<U1')
+
+    >>> _reorder_for_extension_array_stack(arr, 3, 2)
+    array(['a', 'd', 'b', 'e', 'c', 'f'], dtype='<U1')
+    """
+    # final take to get the order correct.
+    # idx is an indexer like
+    # [c0r0, c1r0, c2r0, ...,
+    #  c0r1, c1r1, c2r1, ...]
+    idx = np.arange(n_rows * n_columns).reshape(n_columns, n_rows).T.reshape(-1)
+    return arr.take(idx)
+
+
+def stack_v3(frame: DataFrame, level: list[int]) -> Series | DataFrame:
+    if frame.columns.nunique() != len(frame.columns):
+        raise ValueError("Columns with duplicate values are not supported in stack")
+    if not len(level):
+        return frame
+    set_levels = set(level)
+    stack_cols = frame.columns._drop_level_numbers(
+        [k for k in range(frame.columns.nlevels - 1, -1, -1) if k not in set_levels]
+    )
+
+    result = stack_reshape(frame, level, set_levels, stack_cols)
+
+    # Construct the correct MultiIndex by combining the frame's index and
+    # stacked columns.
+    ratio = 0 if frame.empty else len(result) // len(frame)
+
+    index_levels: list | FrozenList
+    if isinstance(frame.index, MultiIndex):
+        index_levels = frame.index.levels
+        index_codes = list(np.tile(frame.index.codes, (1, ratio)))
+    else:
+        codes, uniques = factorize(frame.index, use_na_sentinel=False)
+        index_levels = [uniques]
+        index_codes = list(np.tile(codes, (1, ratio)))
+
+    if len(level) > 1:
+        # Arrange columns in the order we want to take them, e.g. level=[2, 0, 1]
+        sorter = np.argsort(level)
+        assert isinstance(stack_cols, MultiIndex)
+        ordered_stack_cols = stack_cols._reorder_ilevels(sorter)
+    else:
+        ordered_stack_cols = stack_cols
+    ordered_stack_cols_unique = ordered_stack_cols.unique()
+    if isinstance(ordered_stack_cols, MultiIndex):
+        column_levels = ordered_stack_cols.levels
+        column_codes = ordered_stack_cols.drop_duplicates().codes
+    else:
+        column_levels = [ordered_stack_cols_unique]
+        column_codes = [factorize(ordered_stack_cols_unique, use_na_sentinel=False)[0]]
+
+    # error: Incompatible types in assignment (expression has type "list[ndarray[Any,
+    # dtype[Any]]]", variable has type "FrozenList")
+    column_codes = [np.repeat(codes, len(frame)) for codes in column_codes]  # type: ignore[assignment]
+    result.index = MultiIndex(
+        levels=index_levels + column_levels,
+        codes=index_codes + column_codes,
+        names=frame.index.names + list(ordered_stack_cols.names),
+        verify_integrity=False,
+    )
+
+    # sort result, but faster than calling sort_index since we know the order we need
+    len_df = len(frame)
+    n_uniques = len(ordered_stack_cols_unique)
+    indexer = np.arange(n_uniques)
+    idxs = np.tile(len_df * indexer, len_df) + np.repeat(np.arange(len_df), n_uniques)
+    result = result.take(idxs)
+
+    # Reshape/rename if needed and dropna
+    if result.ndim == 2 and frame.columns.nlevels == len(level):
+        if len(result.columns) == 0:
+            result = Series(index=result.index)
+        else:
+            result = result.iloc[:, 0]
+    if result.ndim == 1:
+        result.name = None
+
+    return result
+
+
+def stack_reshape(
+    frame: DataFrame, level: list[int], set_levels: set[int], stack_cols: Index
+) -> Series | DataFrame:
+    """Reshape the data of a frame for stack.
+
+    This function takes care of most of the work that stack needs to do. Caller
+    will sort the result once the appropriate index is set.
+
+    Parameters
+    ----------
+    frame: DataFrame
+        DataFrame that is to be stacked.
+    level: list of ints.
+        Levels of the columns to stack.
+    set_levels: set of ints.
+        Same as level, but as a set.
+    stack_cols: Index.
+        Columns of the result when the DataFrame is stacked.
+
+    Returns
+    -------
+    The data of behind the stacked DataFrame.
+    """
+    # If we need to drop `level` from columns, it needs to be in descending order
+    drop_levnums = sorted(level, reverse=True)
+
+    # Grab data for each unique index to be stacked
+    buf = []
+    for idx in stack_cols.unique():
+        if len(frame.columns) == 1:
+            data = frame.copy(deep=False)
+        else:
+            if not isinstance(frame.columns, MultiIndex) and not isinstance(idx, tuple):
+                # GH#57750 - if the frame is an Index with tuples, .loc below will fail
+                column_indexer = idx
+            else:
+                # Take the data from frame corresponding to this idx value
+                if len(level) == 1:
+                    idx = (idx,)
+                gen = iter(idx)
+                column_indexer = tuple(
+                    next(gen) if k in set_levels else slice(None)
+                    for k in range(frame.columns.nlevels)
+                )
+            data = frame.loc[:, column_indexer]
+
+        if len(level) < frame.columns.nlevels:
+            data.columns = data.columns._drop_level_numbers(drop_levnums)
+        elif stack_cols.nlevels == 1:
+            if data.ndim == 1:
+                data.name = 0
+            else:
+                data.columns = default_index(len(data.columns))
+        buf.append(data)
+
+    if len(buf) > 0 and not frame.empty:
+        result = concat(buf, ignore_index=True)
+    else:
+        # input is empty
+        if len(level) < frame.columns.nlevels:
+            # concat column order may be different from dropping the levels
+            new_columns = frame.columns._drop_level_numbers(drop_levnums).unique()
+        else:
+            new_columns = [0]
+        result = DataFrame(columns=new_columns, dtype=frame._values.dtype)
+
+    if len(level) < frame.columns.nlevels:
+        # concat column order may be different from dropping the levels
+        desired_columns = frame.columns._drop_level_numbers(drop_levnums).unique()
+        if not result.columns.equals(desired_columns):
+            result = result[desired_columns]
+
+    return result
diff --git a/pandas/pandas/core/reshape/tile.py b/pandas/pandas/core/reshape/tile.py
new file mode 100644
index 0000000000000..034b861a83f43
--- /dev/null
+++ b/pandas/pandas/core/reshape/tile.py
@@ -0,0 +1,672 @@
+"""
+Quantilization functions and related stuff
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    Timedelta,
+    Timestamp,
+    lib,
+)
+
+from pandas.core.dtypes.common import (
+    ensure_platform_int,
+    is_bool_dtype,
+    is_integer,
+    is_list_like,
+    is_numeric_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.generic import ABCSeries
+from pandas.core.dtypes.missing import isna
+
+from pandas import (
+    Categorical,
+    Index,
+    IntervalIndex,
+)
+import pandas.core.algorithms as algos
+from pandas.core.arrays.datetimelike import dtype_to_unit
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        DtypeObj,
+        IntervalLeftRight,
+    )
+
+
+def cut(
+    x,
+    bins,
+    right: bool = True,
+    labels=None,
+    retbins: bool = False,
+    precision: int = 3,
+    include_lowest: bool = False,
+    duplicates: str = "raise",
+    ordered: bool = True,
+):
+    """
+    Bin values into discrete intervals.
+
+    Use `cut` when you need to segment and sort data values into bins. This
+    function is also useful for going from a continuous variable to a
+    categorical variable. For example, `cut` could convert ages to groups of
+    age ranges. Supports binning into an equal number of bins, or a
+    pre-specified array of bins.
+
+    Parameters
+    ----------
+    x : 1d ndarray or Series
+        The input array to be binned. Must be 1-dimensional.
+    bins : int, sequence of scalars, or IntervalIndex
+        The criteria to bin by.
+
+        * int : Defines the number of equal-width bins in the range of `x`. The
+          range of `x` is extended by .1% on each side to include the minimum
+          and maximum values of `x`.
+        * sequence of scalars : Defines the bin edges allowing for non-uniform
+          width. No extension of the range of `x` is done.
+        * IntervalIndex : Defines the exact bins to be used. Note that
+          IntervalIndex for `bins` must be non-overlapping.
+
+    right : bool, default True
+        Indicates whether `bins` includes the rightmost edge or not. If
+        ``right == True`` (the default), then the `bins` ``[1, 2, 3, 4]``
+        indicate (1,2], (2,3], (3,4]. This argument is ignored when
+        `bins` is an IntervalIndex.
+    labels : array or False, default None
+        Specifies the labels for the returned bins. Must be the same length as
+        the resulting bins. If False, returns only integer indicators of the
+        bins. This affects the type of the output container (see below).
+        This argument is ignored when `bins` is an IntervalIndex. If True,
+        raises an error. When `ordered=False`, labels must be provided.
+    retbins : bool, default False
+        Whether to return the bins or not. Useful when bins is provided
+        as a scalar.
+    precision : int, default 3
+        The precision at which to store and display the bins labels.
+    include_lowest : bool, default False
+        Whether the first interval should be left-inclusive or not.
+    duplicates : {default 'raise', 'drop'}, optional
+        If bin edges are not unique, raise ValueError or drop non-uniques.
+    ordered : bool, default True
+        Whether the labels are ordered or not. Applies to returned types
+        Categorical and Series (with Categorical dtype). If True,
+        the resulting categorical will be ordered. If False, the resulting
+        categorical will be unordered (labels must be provided).
+
+    Returns
+    -------
+    out : Categorical, Series, or ndarray
+        An array-like object representing the respective bin for each value
+        of `x`. The type depends on the value of `labels`.
+
+        * None (default) : returns a Series for Series `x` or a
+          Categorical for all other inputs. The values stored within
+          are Interval dtype.
+
+        * sequence of scalars : returns a Series for Series `x` or a
+          Categorical for all other inputs. The values stored within
+          are whatever the type in the sequence is.
+
+        * False : returns a 1d ndarray or Series of integers.
+
+    bins : numpy.ndarray or IntervalIndex.
+        The computed or specified bins. Only returned when `retbins=True`.
+        For scalar or sequence `bins`, this is an ndarray with the computed
+        bins. If set `duplicates=drop`, `bins` will drop non-unique bin. For
+        an IntervalIndex `bins`, this is equal to `bins`.
+
+    See Also
+    --------
+    qcut : Discretize variable into equal-sized buckets based on rank
+        or based on sample quantiles.
+    Categorical : Array type for storing data that come from a
+        fixed set of values.
+    Series : One-dimensional array with axis labels (including time series).
+    IntervalIndex : Immutable Index implementing an ordered, sliceable set.
+    numpy.histogram_bin_edges: Function to calculate only the edges of the bins
+        used by the histogram function.
+
+    Notes
+    -----
+    Any NA values will be NA in the result. Out of bounds values will be NA in
+    the resulting Series or Categorical object.
+
+    ``numpy.histogram_bin_edges`` can be used along with cut to calculate bins according
+    to some predefined methods.
+
+    Reference :ref:`the user guide <reshaping.tile.cut>` for more examples.
+
+    Examples
+    --------
+    Discretize into three equal-sized bins.
+
+    >>> pd.cut(np.array([1, 7, 5, 4, 6, 3]), 3)
+    ... # doctest: +ELLIPSIS
+    [(0.994, 3.0], (5.0, 7.0], (3.0, 5.0], (3.0, 5.0], (5.0, 7.0], ...
+    Categories (3, interval[float64, right]): [(0.994, 3.0] < (3.0, 5.0] ...
+
+    >>> pd.cut(np.array([1, 7, 5, 4, 6, 3]), 3, retbins=True)
+    ... # doctest: +ELLIPSIS
+    ([(0.994, 3.0], (5.0, 7.0], (3.0, 5.0], (3.0, 5.0], (5.0, 7.0], ...
+    Categories (3, interval[float64, right]): [(0.994, 3.0] < (3.0, 5.0] ...
+    array([0.994, 3.   , 5.   , 7.   ]))
+
+    Discovers the same bins, but assign them specific labels. Notice that
+    the returned Categorical's categories are `labels` and is ordered.
+
+    >>> pd.cut(np.array([1, 7, 5, 4, 6, 3]), 3, labels=["bad", "medium", "good"])
+    ['bad', 'good', 'medium', 'medium', 'good', 'bad']
+    Categories (3, object): ['bad' < 'medium' < 'good']
+
+    ``ordered=False`` will result in unordered categories when labels are passed.
+    This parameter can be used to allow non-unique labels:
+
+    >>> pd.cut(np.array([1, 7, 5, 4, 6, 3]), 3, labels=["B", "A", "B"], ordered=False)
+    ['B', 'B', 'A', 'A', 'B', 'B']
+    Categories (2, object): ['A', 'B']
+
+    ``labels=False`` implies you just want the bins back.
+
+    >>> pd.cut([0, 1, 1, 2], bins=4, labels=False)
+    array([0, 1, 1, 3])
+
+    Passing a Series as an input returns a Series with categorical dtype:
+
+    >>> s = pd.Series(np.array([2, 4, 6, 8, 10]), index=["a", "b", "c", "d", "e"])
+    >>> pd.cut(s, 3)
+    ... # doctest: +ELLIPSIS
+    a    (1.992, 4.667]
+    b    (1.992, 4.667]
+    c    (4.667, 7.333]
+    d     (7.333, 10.0]
+    e     (7.333, 10.0]
+    dtype: category
+    Categories (3, interval[float64, right]): [(1.992, 4.667] < (4.667, ...
+
+    Passing a Series as an input returns a Series with mapping value.
+    It is used to map numerically to intervals based on bins.
+
+    >>> s = pd.Series(np.array([2, 4, 6, 8, 10]), index=["a", "b", "c", "d", "e"])
+    >>> pd.cut(s, [0, 2, 4, 6, 8, 10], labels=False, retbins=True, right=False)
+    ... # doctest: +ELLIPSIS
+    (a    1.0
+     b    2.0
+     c    3.0
+     d    4.0
+     e    NaN
+     dtype: float64,
+     array([ 0,  2,  4,  6,  8, 10]))
+
+    Use `drop` optional when bins is not unique
+
+    >>> pd.cut(
+    ...     s,
+    ...     [0, 2, 4, 6, 10, 10],
+    ...     labels=False,
+    ...     retbins=True,
+    ...     right=False,
+    ...     duplicates="drop",
+    ... )
+    ... # doctest: +ELLIPSIS
+    (a    1.0
+     b    2.0
+     c    3.0
+     d    3.0
+     e    NaN
+     dtype: float64,
+     array([ 0,  2,  4,  6, 10]))
+
+    Passing an IntervalIndex for `bins` results in those categories exactly.
+    Notice that values not covered by the IntervalIndex are set to NaN. 0
+    is to the left of the first bin (which is closed on the right), and 1.5
+    falls between two bins.
+
+    >>> bins = pd.IntervalIndex.from_tuples([(0, 1), (2, 3), (4, 5)])
+    >>> pd.cut([0, 0.5, 1.5, 2.5, 4.5], bins)
+    [NaN, (0.0, 1.0], NaN, (2.0, 3.0], (4.0, 5.0]]
+    Categories (3, interval[int64, right]): [(0, 1] < (2, 3] < (4, 5]]
+
+    Using np.histogram_bin_edges with cut
+
+    >>> pd.cut(
+    ...     np.array([1, 7, 5, 4]),
+    ...     bins=np.histogram_bin_edges(np.array([1, 7, 5, 4]), bins="auto"),
+    ... )
+    ... # doctest: +ELLIPSIS
+    [NaN, (5.0, 7.0], (3.0, 5.0], (3.0, 5.0]]
+    Categories (3, interval[float64, right]): [(1.0, 3.0] < (3.0, 5.0] < (5.0, 7.0]]
+    """
+    # NOTE: this binning code is changed a bit from histogram for var(x) == 0
+
+    original = x
+    x_idx = _preprocess_for_cut(x)
+    x_idx, _ = _coerce_to_type(x_idx)
+
+    if not np.iterable(bins):
+        bins = _nbins_to_bins(x_idx, bins, right)
+
+    elif isinstance(bins, IntervalIndex):
+        if bins.is_overlapping:
+            raise ValueError("Overlapping IntervalIndex is not accepted.")
+
+    else:
+        bins = Index(bins)
+        if not bins.is_monotonic_increasing:
+            raise ValueError("bins must increase monotonically.")
+
+    fac, bins = _bins_to_cuts(
+        x_idx,
+        bins,
+        right=right,
+        labels=labels,
+        precision=precision,
+        include_lowest=include_lowest,
+        duplicates=duplicates,
+        ordered=ordered,
+    )
+
+    return _postprocess_for_cut(fac, bins, retbins, original)
+
+
+def qcut(
+    x,
+    q,
+    labels=None,
+    retbins: bool = False,
+    precision: int = 3,
+    duplicates: str = "raise",
+):
+    """
+    Quantile-based discretization function.
+
+    Discretize variable into equal-sized buckets based on rank or based
+    on sample quantiles. For example 1000 values for 10 quantiles would
+    produce a Categorical object indicating quantile membership for each data point.
+
+    Parameters
+    ----------
+    x : 1d ndarray or Series
+        Input Numpy array or pandas Series object to be discretized.
+    q : int or list-like of float
+        Number of quantiles. 10 for deciles, 4 for quartiles, etc. Alternately
+        array of quantiles, e.g. [0, .25, .5, .75, 1.] for quartiles.
+    labels : array or False, default None
+        Used as labels for the resulting bins. Must be of the same length as
+        the resulting bins. If False, return only integer indicators of the
+        bins. If True, raises an error.
+    retbins : bool, optional
+        Whether to return the (bins, labels) or not. Can be useful if bins
+        is given as a scalar.
+    precision : int, optional
+        The precision at which to store and display the bins labels.
+    duplicates : {default 'raise', 'drop'}, optional
+        If bin edges are not unique, raise ValueError or drop non-uniques.
+
+    Returns
+    -------
+    out : Categorical or Series or array of integers if labels is False
+        The return type (Categorical or Series) depends on the input: a Series
+        of type category if input is a Series else Categorical. Bins are
+        represented as categories when categorical data is returned.
+    bins : ndarray of floats
+        Returned only if `retbins` is True.
+
+    See Also
+    --------
+    cut : Bin values into discrete intervals.
+    Series.quantile : Return value at the given quantile.
+
+    Notes
+    -----
+    Out of bounds values will be NA in the resulting Categorical object
+
+    Examples
+    --------
+    >>> pd.qcut(range(5), 4)
+    ... # doctest: +ELLIPSIS
+    [(-0.001, 1.0], (-0.001, 1.0], (1.0, 2.0], (2.0, 3.0], (3.0, 4.0]]
+    Categories (4, interval[float64, right]): [(-0.001, 1.0] < (1.0, 2.0] ...
+
+    >>> pd.qcut(range(5), 3, labels=["good", "medium", "bad"])
+    ... # doctest: +SKIP
+    [good, good, medium, bad, bad]
+    Categories (3, object): [good < medium < bad]
+
+    >>> pd.qcut(range(5), 4, labels=False)
+    array([0, 0, 1, 2, 3])
+    """
+    original = x
+    x_idx = _preprocess_for_cut(x)
+    x_idx, _ = _coerce_to_type(x_idx)
+
+    if is_integer(q):
+        quantiles = np.linspace(0, 1, q + 1)
+        # Round up rather than to nearest if not representable in base 2
+        np.putmask(
+            quantiles,
+            q * quantiles != np.arange(q + 1),
+            np.nextafter(quantiles, 1),
+        )
+    else:
+        quantiles = q
+
+    bins = x_idx.to_series().dropna().quantile(quantiles)
+
+    fac, bins = _bins_to_cuts(
+        x_idx,
+        Index(bins),
+        labels=labels,
+        precision=precision,
+        include_lowest=True,
+        duplicates=duplicates,
+    )
+
+    return _postprocess_for_cut(fac, bins, retbins, original)
+
+
+def _nbins_to_bins(x_idx: Index, nbins: int, right: bool) -> Index:
+    """
+    If a user passed an integer N for bins, convert this to a sequence of N
+    equal(ish)-sized bins.
+    """
+    if is_scalar(nbins) and nbins < 1:
+        raise ValueError("`bins` should be a positive integer.")
+
+    if x_idx.size == 0:
+        raise ValueError("Cannot cut empty array")
+
+    rng = (x_idx.min(), x_idx.max())
+    mn, mx = rng
+
+    if is_numeric_dtype(x_idx.dtype) and (np.isinf(mn) or np.isinf(mx)):
+        # GH#24314
+        raise ValueError(
+            "cannot specify integer `bins` when input data contains infinity"
+        )
+
+    if mn == mx:  # adjust end points before binning
+        if _is_dt_or_td(x_idx.dtype):
+            # using seconds=1 is pretty arbitrary here
+            # error: Argument 1 to "dtype_to_unit" has incompatible type
+            # "dtype[Any] | ExtensionDtype"; expected "DatetimeTZDtype | dtype[Any]"
+            unit = dtype_to_unit(x_idx.dtype)  # type: ignore[arg-type]
+            td = Timedelta(seconds=1).as_unit(unit)
+            # Use DatetimeArray/TimedeltaArray method instead of linspace
+            # error: Item "ExtensionArray" of "ExtensionArray | ndarray[Any, Any]"
+            # has no attribute "_generate_range"
+            bins = x_idx._values._generate_range(  # type: ignore[union-attr]
+                start=mn - td, end=mx + td, periods=nbins + 1, freq=None, unit=unit
+            )
+        else:
+            mn -= 0.001 * abs(mn) if mn != 0 else 0.001
+            mx += 0.001 * abs(mx) if mx != 0 else 0.001
+
+            bins = np.linspace(mn, mx, nbins + 1, endpoint=True)
+    else:  # adjust end points after binning
+        if _is_dt_or_td(x_idx.dtype):
+            # Use DatetimeArray/TimedeltaArray method instead of linspace
+
+            # error: Argument 1 to "dtype_to_unit" has incompatible type
+            # "dtype[Any] | ExtensionDtype"; expected "DatetimeTZDtype | dtype[Any]"
+            unit = dtype_to_unit(x_idx.dtype)  # type: ignore[arg-type]
+            # error: Item "ExtensionArray" of "ExtensionArray | ndarray[Any, Any]"
+            # has no attribute "_generate_range"
+            bins = x_idx._values._generate_range(  # type: ignore[union-attr]
+                start=mn, end=mx, periods=nbins + 1, freq=None, unit=unit
+            )
+        else:
+            bins = np.linspace(mn, mx, nbins + 1, endpoint=True)
+        adj = (mx - mn) * 0.001  # 0.1% of the range
+        if right:
+            bins[0] -= adj
+        else:
+            bins[-1] += adj
+
+    return Index(bins)
+
+
+def _bins_to_cuts(
+    x_idx: Index,
+    bins: Index,
+    right: bool = True,
+    labels=None,
+    precision: int = 3,
+    include_lowest: bool = False,
+    duplicates: str = "raise",
+    ordered: bool = True,
+):
+    if not ordered and labels is None:
+        raise ValueError("'labels' must be provided if 'ordered = False'")
+
+    if duplicates not in ["raise", "drop"]:
+        raise ValueError(
+            "invalid value for 'duplicates' parameter, valid options are: raise, drop"
+        )
+
+    result: Categorical | np.ndarray
+
+    if isinstance(bins, IntervalIndex):
+        # we have a fast-path here
+        ids = bins.get_indexer(x_idx)
+        cat_dtype = CategoricalDtype(bins, ordered=True)
+        result = Categorical.from_codes(ids, dtype=cat_dtype, validate=False)
+        return result, bins
+
+    unique_bins = algos.unique(bins)
+    if len(unique_bins) < len(bins) and len(bins) != 2:
+        if duplicates == "raise":
+            raise ValueError(
+                f"Bin edges must be unique: {bins!r}.\n"
+                f"You can drop duplicate edges by setting the 'duplicates' kwarg"
+            )
+        bins = unique_bins
+
+    side: Literal["left", "right"] = "left" if right else "right"
+
+    try:
+        ids = bins.searchsorted(x_idx, side=side)
+    except TypeError as err:
+        # e.g. test_datetime_nan_error if bins are DatetimeArray and x_idx
+        #  is integers
+        if x_idx.dtype.kind == "m":
+            raise ValueError("bins must be of timedelta64 dtype") from err
+        elif x_idx.dtype.kind == bins.dtype.kind == "M":
+            raise ValueError(
+                "Cannot use timezone-naive bins with timezone-aware values, "
+                "or vice-versa"
+            ) from err
+        elif x_idx.dtype.kind == "M":
+            raise ValueError("bins must be of datetime64 dtype") from err
+        else:
+            raise
+    ids = ensure_platform_int(ids)
+
+    if include_lowest:
+        ids[x_idx == bins[0]] = 1
+
+    na_mask = isna(x_idx) | (ids == len(bins)) | (ids == 0)
+    has_nas = na_mask.any()
+
+    if labels is not False:
+        if not (labels is None or is_list_like(labels)):
+            raise ValueError(
+                "Bin labels must either be False, None or passed in as a "
+                "list-like argument"
+            )
+
+        if labels is None:
+            labels = _format_labels(
+                bins, precision, right=right, include_lowest=include_lowest
+            )
+        elif ordered and len(set(labels)) != len(labels):
+            raise ValueError(
+                "labels must be unique if ordered=True; pass ordered=False "
+                "for duplicate labels"
+            )
+        else:
+            if len(labels) != len(bins) - 1:
+                raise ValueError(
+                    "Bin labels must be one fewer than the number of bin edges"
+                )
+
+        if not isinstance(getattr(labels, "dtype", None), CategoricalDtype):
+            labels = Categorical(
+                labels,
+                categories=labels if len(set(labels)) == len(labels) else None,
+                ordered=ordered,
+            )
+        # TODO: handle mismatch between categorical label order and pandas.cut order.
+        np.putmask(ids, na_mask, 0)
+        result = algos.take_nd(labels, ids - 1)
+
+    else:
+        result = ids - 1
+        if has_nas:
+            result = result.astype(np.float64)
+            np.putmask(result, na_mask, np.nan)
+
+    return result, bins
+
+
+def _coerce_to_type(x: Index) -> tuple[Index, DtypeObj | None]:
+    """
+    if the passed data is of datetime/timedelta, bool or nullable int type,
+    this method converts it to numeric so that cut or qcut method can
+    handle it
+    """
+    dtype: DtypeObj | None = None
+
+    if _is_dt_or_td(x.dtype):
+        dtype = x.dtype
+    elif is_bool_dtype(x.dtype):
+        # GH 20303
+        x = x.astype(np.int64)
+    # To support cut and qcut for IntegerArray we convert to float dtype.
+    # Will properly support in the future.
+    # https://github.com/pandas-dev/pandas/pull/31290
+    # https://github.com/pandas-dev/pandas/issues/31389
+    elif isinstance(x.dtype, ExtensionDtype) and is_numeric_dtype(x.dtype):
+        x_arr = x.to_numpy(dtype=np.float64, na_value=np.nan)
+        x = Index(x_arr)
+
+    return Index(x), dtype
+
+
+def _is_dt_or_td(dtype: DtypeObj) -> bool:
+    # Note: the dtype here comes from an Index.dtype, so we know that that any
+    #  dt64/td64 dtype is of a supported unit.
+    return isinstance(dtype, DatetimeTZDtype) or lib.is_np_dtype(dtype, "mM")
+
+
+def _format_labels(
+    bins: Index,
+    precision: int,
+    right: bool = True,
+    include_lowest: bool = False,
+) -> IntervalIndex:
+    """based on the dtype, return our labels"""
+    closed: IntervalLeftRight = "right" if right else "left"
+
+    formatter: Callable[[Any], Timestamp] | Callable[[Any], Timedelta]
+
+    if _is_dt_or_td(bins.dtype):
+        # error: Argument 1 to "dtype_to_unit" has incompatible type
+        # "dtype[Any] | ExtensionDtype"; expected "DatetimeTZDtype | dtype[Any]"
+        unit = dtype_to_unit(bins.dtype)  # type: ignore[arg-type]
+        formatter = lambda x: x
+        adjust = lambda x: x - Timedelta(1, unit=unit).as_unit(unit)
+    else:
+        precision = _infer_precision(precision, bins)
+        formatter = lambda x: _round_frac(x, precision)
+        adjust = lambda x: x - 10 ** (-precision)
+
+    breaks = [formatter(b) for b in bins]
+    if right and include_lowest:
+        # adjust lhs of first interval by precision to account for being right closed
+        breaks[0] = adjust(breaks[0])
+
+    if _is_dt_or_td(bins.dtype):
+        # error: "Index" has no attribute "as_unit"
+        breaks = type(bins)(breaks).as_unit(unit)  # type: ignore[attr-defined]
+
+    return IntervalIndex.from_breaks(breaks, closed=closed)
+
+
+def _preprocess_for_cut(x) -> Index:
+    """
+    handles preprocessing for cut where we convert passed
+    input to array, strip the index information and store it
+    separately
+    """
+    # Check that the passed array is a Pandas or Numpy object
+    # We don't want to strip away a Pandas data-type here (e.g. datetimetz)
+    ndim = getattr(x, "ndim", None)
+    if ndim is None:
+        x = np.asarray(x)
+    if x.ndim != 1:
+        raise ValueError("Input array must be 1 dimensional")
+
+    return Index(x)
+
+
+def _postprocess_for_cut(fac, bins, retbins: bool, original):
+    """
+    handles post processing for the cut method where
+    we combine the index information if the originally passed
+    datatype was a series
+    """
+    if isinstance(original, ABCSeries):
+        fac = original._constructor(fac, index=original.index, name=original.name)
+
+    if not retbins:
+        return fac
+
+    if isinstance(bins, Index) and is_numeric_dtype(bins.dtype):
+        bins = bins._values
+
+    return fac, bins
+
+
+def _round_frac(x, precision: int):
+    """
+    Round the fractional part of the given number
+    """
+    if not np.isfinite(x) or x == 0:
+        return x
+    else:
+        frac, whole = np.modf(x)
+        if whole == 0:
+            digits = -int(np.floor(np.log10(abs(frac)))) - 1 + precision
+        else:
+            digits = precision
+        return np.around(x, digits)
+
+
+def _infer_precision(base_precision: int, bins: Index) -> int:
+    """
+    Infer an appropriate precision for _round_frac
+    """
+    for precision in range(base_precision, 20):
+        levels = np.asarray([_round_frac(b, precision) for b in bins])
+        if algos.unique(levels).size == bins.size:
+            return precision
+    return base_precision  # default
diff --git a/pandas/pandas/core/roperator.py b/pandas/pandas/core/roperator.py
new file mode 100644
index 0000000000000..9ea4bea41cdea
--- /dev/null
+++ b/pandas/pandas/core/roperator.py
@@ -0,0 +1,63 @@
+"""
+Reversed Operations not available in the stdlib operator module.
+Defining these instead of using lambdas allows us to reference them by name.
+"""
+
+from __future__ import annotations
+
+import operator
+
+
+def radd(left, right):
+    return right + left
+
+
+def rsub(left, right):
+    return right - left
+
+
+def rmul(left, right):
+    return right * left
+
+
+def rdiv(left, right):
+    return right / left
+
+
+def rtruediv(left, right):
+    return right / left
+
+
+def rfloordiv(left, right):
+    return right // left
+
+
+def rmod(left, right):
+    # check if right is a string as % is the string
+    # formatting operation; this is a TypeError
+    # otherwise perform the op
+    if isinstance(right, str):
+        typ = type(left).__name__
+        raise TypeError(f"{typ} cannot perform the operation mod")
+
+    return right % left
+
+
+def rdivmod(left, right):
+    return divmod(right, left)
+
+
+def rpow(left, right):
+    return right**left
+
+
+def rand_(left, right):
+    return operator.and_(right, left)
+
+
+def ror_(left, right):
+    return operator.or_(right, left)
+
+
+def rxor(left, right):
+    return operator.xor(right, left)
diff --git a/pandas/pandas/core/sample.py b/pandas/pandas/core/sample.py
new file mode 100644
index 0000000000000..5b1c4b6a331f5
--- /dev/null
+++ b/pandas/pandas/core/sample.py
@@ -0,0 +1,155 @@
+"""
+Module containing utilities for NDFrame.sample() and .GroupBy.sample()
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import AxisInt
+
+    from pandas.core.generic import NDFrame
+
+
+def preprocess_weights(obj: NDFrame, weights, axis: AxisInt) -> np.ndarray:
+    """
+    Process and validate the `weights` argument to `NDFrame.sample` and
+    `.GroupBy.sample`.
+
+    Returns `weights` as an ndarray[np.float64], validated except for normalizing
+    weights (because that must be done groupwise in groupby sampling).
+    """
+    # If a series, align with frame
+    if isinstance(weights, ABCSeries):
+        weights = weights.reindex(obj.axes[axis])
+
+    # Strings acceptable if a dataframe and axis = 0
+    if isinstance(weights, str):
+        if isinstance(obj, ABCDataFrame):
+            if axis == 0:
+                try:
+                    weights = obj[weights]
+                except KeyError as err:
+                    raise KeyError(
+                        "String passed to weights not a valid column"
+                    ) from err
+            else:
+                raise ValueError(
+                    "Strings can only be passed to "
+                    "weights when sampling from rows on "
+                    "a DataFrame"
+                )
+        else:
+            raise ValueError(
+                "Strings cannot be passed as weights when sampling from a Series."
+            )
+
+    if isinstance(obj, ABCSeries):
+        func = obj._constructor
+    else:
+        func = obj._constructor_sliced
+
+    weights = func(weights, dtype="float64")._values
+
+    if len(weights) != obj.shape[axis]:
+        raise ValueError("Weights and axis to be sampled must be of same length")
+
+    if lib.has_infs(weights):
+        raise ValueError("weight vector may not include `inf` values")
+
+    if (weights < 0).any():
+        raise ValueError("weight vector many not include negative values")
+
+    missing = np.isnan(weights)
+    if missing.any():
+        # Don't modify weights in place
+        weights = weights.copy()
+        weights[missing] = 0
+    return weights
+
+
+def process_sampling_size(
+    n: int | None, frac: float | None, replace: bool
+) -> int | None:
+    """
+    Process and validate the `n` and `frac` arguments to `NDFrame.sample` and
+    `.GroupBy.sample`.
+
+    Returns None if `frac` should be used (variable sampling sizes), otherwise returns
+    the constant sampling size.
+    """
+    # If no frac or n, default to n=1.
+    if n is None and frac is None:
+        n = 1
+    elif n is not None and frac is not None:
+        raise ValueError("Please enter a value for `frac` OR `n`, not both")
+    elif n is not None:
+        if n < 0:
+            raise ValueError(
+                "A negative number of rows requested. Please provide `n` >= 0."
+            )
+        if n % 1 != 0:
+            raise ValueError("Only integers accepted as `n` values")
+    else:
+        assert frac is not None  # for mypy
+        if frac > 1 and not replace:
+            raise ValueError(
+                "Replace has to be set to `True` when "
+                "upsampling the population `frac` > 1."
+            )
+        if frac < 0:
+            raise ValueError(
+                "A negative number of rows requested. Please provide `frac` >= 0."
+            )
+
+    return n
+
+
+def sample(
+    obj_len: int,
+    size: int,
+    replace: bool,
+    weights: np.ndarray | None,
+    random_state: np.random.RandomState | np.random.Generator,
+) -> np.ndarray:
+    """
+    Randomly sample `size` indices in `np.arange(obj_len)`
+
+    Parameters
+    ----------
+    obj_len : int
+        The length of the indices being considered
+    size : int
+        The number of values to choose
+    replace : bool
+        Allow or disallow sampling of the same row more than once.
+    weights : np.ndarray[np.float64] or None
+        If None, equal probability weighting, otherwise weights according
+        to the vector normalized
+    random_state: np.random.RandomState or np.random.Generator
+        State used for the random sampling
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+    """
+    if weights is not None:
+        weight_sum = weights.sum()
+        if weight_sum != 0:
+            weights = weights / weight_sum
+        else:
+            raise ValueError("Invalid weights: weights sum to zero")
+
+    return random_state.choice(obj_len, size=size, replace=replace, p=weights).astype(
+        np.intp, copy=False
+    )
diff --git a/pandas/pandas/core/series.py b/pandas/pandas/core/series.py
new file mode 100644
index 0000000000000..da46f8ede3409
--- /dev/null
+++ b/pandas/pandas/core/series.py
@@ -0,0 +1,7255 @@
+"""
+Data structure for 1-dimensional cross-sectional and time series data
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    Mapping,
+    Sequence,
+)
+import functools
+import operator
+import sys
+from textwrap import dedent
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    properties,
+    reshape,
+)
+from pandas._libs.lib import is_range_indexer
+from pandas.compat import PYPY
+from pandas.compat._constants import REF_COUNT
+from pandas.compat._optional import import_optional_dependency
+from pandas.compat.numpy import function as nv
+from pandas.errors import (
+    ChainedAssignmentError,
+    InvalidIndexError,
+)
+from pandas.errors.cow import (
+    _chained_assignment_method_msg,
+    _chained_assignment_msg,
+)
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    deprecate_nonkeyword_arguments,
+    doc,
+    set_module,
+)
+from pandas.util._validators import (
+    validate_ascending,
+    validate_bool_kwarg,
+    validate_percentile,
+)
+
+from pandas.core.dtypes.astype import astype_is_view
+from pandas.core.dtypes.cast import (
+    LossySetitemError,
+    construct_1d_arraylike_from_scalar,
+    find_common_type,
+    infer_dtype_from,
+    maybe_box_native,
+    maybe_cast_pointwise_result,
+)
+from pandas.core.dtypes.common import (
+    is_dict_like,
+    is_float,
+    is_integer,
+    is_iterator,
+    is_list_like,
+    is_object_dtype,
+    is_scalar,
+    pandas_dtype,
+    validate_all_hashable,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    ExtensionDtype,
+    SparseDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.inference import is_hashable
+from pandas.core.dtypes.missing import (
+    isna,
+    na_value_for_dtype,
+    notna,
+    remove_na_arraylike,
+)
+
+from pandas.core import (
+    algorithms,
+    base,
+    common as com,
+    nanops,
+    ops,
+    roperator,
+)
+from pandas.core.accessor import Accessor
+from pandas.core.apply import SeriesApply
+from pandas.core.arrays import ExtensionArray
+from pandas.core.arrays.arrow import (
+    ListAccessor,
+    StructAccessor,
+)
+from pandas.core.arrays.categorical import CategoricalAccessor
+from pandas.core.arrays.sparse import SparseAccessor
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.construction import (
+    array as pd_array,
+    extract_array,
+    sanitize_array,
+)
+from pandas.core.generic import (
+    NDFrame,
+    make_doc,
+)
+from pandas.core.indexers import (
+    disallow_ndim_indexing,
+    unpack_1tuple,
+)
+from pandas.core.indexes.accessors import CombinedDatetimelikeProperties
+from pandas.core.indexes.api import (
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    default_index,
+    ensure_index,
+    maybe_sequence_to_range,
+)
+import pandas.core.indexes.base as ibase
+from pandas.core.indexes.multi import maybe_droplevels
+from pandas.core.indexing import (
+    check_bool_indexer,
+    check_dict_or_set_indexers,
+)
+from pandas.core.internals import SingleBlockManager
+from pandas.core.methods import selectn
+from pandas.core.shared_docs import _shared_docs
+from pandas.core.sorting import (
+    ensure_key_mapped,
+    nargsort,
+)
+from pandas.core.strings.accessor import StringMethods
+from pandas.core.tools.datetimes import to_datetime
+
+import pandas.io.formats.format as fmt
+from pandas.io.formats.info import (
+    INFO_DOCSTRING,
+    SeriesInfo,
+    series_sub_kwargs,
+)
+import pandas.plotting
+
+if TYPE_CHECKING:
+    from pandas._libs.internals import BlockValuesRefs
+    from pandas._typing import (
+        AggFuncType,
+        AnyAll,
+        AnyArrayLike,
+        ArrayLike,
+        Axis,
+        AxisInt,
+        CorrelationMethod,
+        DropKeep,
+        Dtype,
+        DtypeObj,
+        FilePath,
+        Frequency,
+        IgnoreRaise,
+        IndexKeyFunc,
+        IndexLabel,
+        Level,
+        ListLike,
+        MutableMappingT,
+        NaPosition,
+        NumpySorter,
+        NumpyValueArrayLike,
+        QuantileInterpolation,
+        ReindexMethod,
+        Renamer,
+        Scalar,
+        Self,
+        SortKind,
+        StorageOptions,
+        Suffixes,
+        ValueKeyFunc,
+        WriteBuffer,
+        npt,
+    )
+
+    from pandas.core.frame import DataFrame
+    from pandas.core.groupby.generic import SeriesGroupBy
+
+__all__ = ["Series"]
+
+_shared_doc_kwargs = {
+    "axes": "index",
+    "klass": "Series",
+    "axes_single_arg": "{0 or 'index'}",
+    "axis": """axis : {0 or 'index'}
+        Unused. Parameter needed for compatibility with DataFrame.""",
+    "inplace": """inplace : bool, default False
+        If True, performs operation inplace and returns None.""",
+    "unique": "np.ndarray",
+    "duplicated": "Series",
+    "optional_by": "",
+    "optional_reindex": """
+index : array-like, optional
+    New labels for the index. Preferably an Index object to avoid
+    duplicating data.
+axis : int or str, optional
+    Unused.""",
+}
+
+# ----------------------------------------------------------------------
+# Series class
+
+
+# error: Cannot override final attribute "ndim" (previously declared in base
+# class "NDFrame")
+# error: Cannot override final attribute "size" (previously declared in base
+# class "NDFrame")
+# definition in base class "NDFrame"
+@set_module("pandas")
+class Series(base.IndexOpsMixin, NDFrame):  # type: ignore[misc]
+    """
+    One-dimensional ndarray with axis labels (including time series).
+
+    Labels need not be unique but must be a hashable type. The object
+    supports both integer- and label-based indexing and provides a host of
+    methods for performing operations involving the index. Statistical
+    methods from ndarray have been overridden to automatically exclude
+    missing data (currently represented as NaN).
+
+    Operations between Series (+, -, /, \\*, \\*\\*) align values based on their
+    associated index values-- they need not be the same length. The result
+    index will be the sorted union of the two indexes.
+
+    Parameters
+    ----------
+    data : array-like, Iterable, dict, or scalar value
+        Contains data stored in Series. If data is a dict, argument order is
+        maintained. Unordered sets are not supported.
+    index : array-like or Index (1d)
+        Values must be hashable and have the same length as `data`.
+        Non-unique index values are allowed. Will default to
+        RangeIndex (0, 1, 2, ..., n) if not provided. If data is dict-like
+        and index is None, then the keys in the data are used as the index. If the
+        index is not None, the resulting Series is reindexed with the index values.
+    dtype : str, numpy.dtype, or ExtensionDtype, optional
+        Data type for the output Series. If not specified, this will be
+        inferred from `data`.
+        See the :ref:`user guide <basics.dtypes>` for more usages.
+        If ``data`` is Series then is ignored.
+    name : Hashable, default None
+        The name to give to the Series.
+    copy : bool, default False
+        Copy input data. Only affects Series or 1d ndarray input. See examples.
+
+    See Also
+    --------
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+    Index : Immutable sequence used for indexing and alignment.
+
+    Notes
+    -----
+    Please reference the :ref:`User Guide <basics.series>` for more information.
+
+    Examples
+    --------
+    Constructing Series from a dictionary with an Index specified
+
+    >>> d = {"a": 1, "b": 2, "c": 3}
+    >>> ser = pd.Series(data=d, index=["a", "b", "c"])
+    >>> ser
+    a   1
+    b   2
+    c   3
+    dtype: int64
+
+    The keys of the dictionary match with the Index values, hence the Index
+    values have no effect.
+
+    >>> d = {"a": 1, "b": 2, "c": 3}
+    >>> ser = pd.Series(data=d, index=["x", "y", "z"])
+    >>> ser
+    x   NaN
+    y   NaN
+    z   NaN
+    dtype: float64
+
+    Note that the Index is first built with the keys from the dictionary.
+    After this the Series is reindexed with the given Index values, hence we
+    get all NaN as a result.
+
+    Constructing Series from a list with `copy=False`.
+
+    >>> r = [1, 2]
+    >>> ser = pd.Series(r, copy=False)
+    >>> ser.iloc[0] = 999
+    >>> r
+    [1, 2]
+    >>> ser
+    0    999
+    1      2
+    dtype: int64
+
+    Due to input data type the Series has a `copy` of
+    the original data even though `copy=False`, so
+    the data is unchanged.
+
+    Constructing Series from a 1d ndarray with `copy=False`.
+
+    >>> r = np.array([1, 2])
+    >>> ser = pd.Series(r, copy=False)
+    >>> ser.iloc[0] = 999
+    >>> r
+    array([999,   2])
+    >>> ser
+    0    999
+    1      2
+    dtype: int64
+
+    Due to input data type the Series has a `view` on
+    the original data, so
+    the data is changed as well.
+    """
+
+    _typ = "series"
+    _HANDLED_TYPES = (Index, ExtensionArray, np.ndarray)
+
+    _name: Hashable
+    _metadata: list[str] = ["_name"]
+    _internal_names_set = {"index", "name"} | NDFrame._internal_names_set
+    _accessors = {"dt", "cat", "str", "sparse"}
+    _hidden_attrs = (
+        base.IndexOpsMixin._hidden_attrs | NDFrame._hidden_attrs | frozenset([])
+    )
+
+    # similar to __array_priority__, positions Series after DataFrame
+    #  but before Index and ExtensionArray.  Should NOT be overridden by subclasses.
+    __pandas_priority__ = 3000
+
+    # Override cache_readonly bc Series is mutable
+    # error: Incompatible types in assignment (expression has type "property",
+    # base class "IndexOpsMixin" defined the type as "Callable[[IndexOpsMixin], bool]")
+    hasnans = property(  # type: ignore[assignment]
+        # error: "Callable[[IndexOpsMixin], bool]" has no attribute "fget"
+        base.IndexOpsMixin.hasnans.fget,  # type: ignore[attr-defined]
+        doc=base.IndexOpsMixin.hasnans.__doc__,
+    )
+    _mgr: SingleBlockManager
+
+    # ----------------------------------------------------------------------
+    # Constructors
+
+    def __init__(
+        self,
+        data=None,
+        index=None,
+        dtype: Dtype | None = None,
+        name=None,
+        copy: bool | None = None,
+    ) -> None:
+        allow_mgr = False
+        if (
+            isinstance(data, SingleBlockManager)
+            and index is None
+            and dtype is None
+            and (copy is False or copy is None)
+        ):
+            if not allow_mgr:
+                # GH#52419
+                warnings.warn(
+                    f"Passing a {type(data).__name__} to {type(self).__name__} "
+                    "is deprecated and will raise in a future version. "
+                    "Use public APIs instead.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+            data = data.copy(deep=False)
+            # GH#33357 called with just the SingleBlockManager
+            NDFrame.__init__(self, data)
+            self.name = name
+            return
+
+        if isinstance(data, (ExtensionArray, np.ndarray)):
+            if copy is not False:
+                if dtype is None or astype_is_view(data.dtype, pandas_dtype(dtype)):
+                    data = data.copy()
+        if copy is None:
+            copy = False
+
+        if isinstance(data, SingleBlockManager) and not copy:
+            data = data.copy(deep=False)
+
+            if not allow_mgr:
+                warnings.warn(
+                    f"Passing a {type(data).__name__} to {type(self).__name__} "
+                    "is deprecated and will raise in a future version. "
+                    "Use public APIs instead.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+
+        name = ibase.maybe_extract_name(name, data, type(self))
+
+        if index is not None:
+            index = ensure_index(index)
+
+        if dtype is not None:
+            dtype = self._validate_dtype(dtype)
+
+        if data is None:
+            index = index if index is not None else default_index(0)
+            if len(index) or dtype is not None:
+                data = na_value_for_dtype(pandas_dtype(dtype), compat=False)
+            else:
+                data = []
+
+        if isinstance(data, MultiIndex):
+            raise NotImplementedError(
+                "initializing a Series from a MultiIndex is not supported"
+            )
+
+        refs = None
+        if isinstance(data, Index):
+            if dtype is not None:
+                data = data.astype(dtype)
+
+            refs = data._references
+            copy = False
+
+        elif isinstance(data, np.ndarray):
+            if len(data.dtype):
+                # GH#13296 we are dealing with a compound dtype, which
+                #  should be treated as 2D
+                raise ValueError(
+                    "Cannot construct a Series from an ndarray with "
+                    "compound dtype.  Use DataFrame instead."
+                )
+        elif isinstance(data, Series):
+            if index is None:
+                index = data.index
+                data = data._mgr.copy(deep=False)
+            else:
+                data = data.reindex(index)
+                copy = False
+                data = data._mgr
+        elif isinstance(data, Mapping):
+            data, index = self._init_dict(data, index, dtype)
+            dtype = None
+            copy = False
+        elif isinstance(data, SingleBlockManager):
+            if index is None:
+                index = data.index
+            elif not data.index.equals(index) or copy:
+                # GH#19275 SingleBlockManager input should only be called
+                # internally
+                raise AssertionError(
+                    "Cannot pass both SingleBlockManager "
+                    "`data` argument and a different "
+                    "`index` argument. `copy` must be False."
+                )
+
+            if not allow_mgr:
+                warnings.warn(
+                    f"Passing a {type(data).__name__} to {type(self).__name__} "
+                    "is deprecated and will raise in a future version. "
+                    "Use public APIs instead.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+                allow_mgr = True
+
+        elif isinstance(data, ExtensionArray):
+            pass
+        else:
+            data = com.maybe_iterable_to_list(data)
+            if is_list_like(data) and not len(data) and dtype is None:
+                # GH 29405: Pre-2.0, this defaulted to float.
+                dtype = np.dtype(object)
+
+        if index is None:
+            if not is_list_like(data):
+                data = [data]
+            index = default_index(len(data))
+        elif is_list_like(data):
+            com.require_length_match(data, index)
+
+        # create/copy the manager
+        if isinstance(data, SingleBlockManager):
+            if dtype is not None:
+                data = data.astype(dtype=dtype)
+            elif copy:
+                data = data.copy()
+        else:
+            data = sanitize_array(data, index, dtype, copy)
+            data = SingleBlockManager.from_array(data, index, refs=refs)
+
+        NDFrame.__init__(self, data)
+        self.name = name
+        self._set_axis(0, index)
+
+    def _init_dict(
+        self, data: Mapping, index: Index | None = None, dtype: DtypeObj | None = None
+    ):
+        """
+        Derive the "_mgr" and "index" attributes of a new Series from a
+        dictionary input.
+
+        Parameters
+        ----------
+        data : dict or dict-like
+            Data used to populate the new Series.
+        index : Index or None, default None
+            Index for the new Series: if None, use dict keys.
+        dtype : np.dtype, ExtensionDtype, or None, default None
+            The dtype for the new Series: if None, infer from data.
+
+        Returns
+        -------
+        _data : BlockManager for the new Series
+        index : index for the new Series
+        """
+        # Looking for NaN in dict doesn't work ({np.nan : 1}[float('nan')]
+        # raises KeyError), so we iterate the entire dict, and align
+        if data:
+            # GH:34717, issue was using zip to extract key and values from data.
+            # using generators in effects the performance.
+            # Below is the new way of extracting the keys and values
+
+            keys = maybe_sequence_to_range(tuple(data.keys()))
+            values = list(data.values())  # Generating list of values- faster way
+        elif index is not None:
+            # fastpath for Series(data=None). Just use broadcasting a scalar
+            # instead of reindexing.
+            if len(index) or dtype is not None:
+                values = na_value_for_dtype(pandas_dtype(dtype), compat=False)
+            else:
+                values = []
+            keys = index
+        else:
+            keys, values = default_index(0), []
+
+        # Input is now list-like, so rely on "standard" construction:
+        s = Series(values, index=keys, dtype=dtype)
+
+        # Now we just make sure the order is respected, if any
+        if data and index is not None:
+            s = s.reindex(index)
+        return s._mgr, s.index
+
+    # ----------------------------------------------------------------------
+
+    def __arrow_c_stream__(self, requested_schema=None):
+        """
+        Export the pandas Series as an Arrow C stream PyCapsule.
+
+        This relies on pyarrow to convert the pandas Series to the Arrow
+        format (and follows the default behavior of ``pyarrow.Array.from_pandas``
+        in its handling of the index, i.e. to ignore it).
+        This conversion is not necessarily zero-copy.
+
+        Parameters
+        ----------
+        requested_schema : PyCapsule, default None
+            The schema to which the dataframe should be casted, passed as a
+            PyCapsule containing a C ArrowSchema representation of the
+            requested schema.
+
+        Returns
+        -------
+        PyCapsule
+        """
+        pa = import_optional_dependency("pyarrow", min_version="16.0.0")
+        type = (
+            pa.DataType._import_from_c_capsule(requested_schema)
+            if requested_schema is not None
+            else None
+        )
+        ca = pa.array(self, type=type)
+        if not isinstance(ca, pa.ChunkedArray):
+            ca = pa.chunked_array([ca])
+        return ca.__arrow_c_stream__()
+
+    # ----------------------------------------------------------------------
+
+    @property
+    def _constructor(self) -> type[Series]:
+        return Series
+
+    def _constructor_from_mgr(self, mgr, axes):
+        ser = Series._from_mgr(mgr, axes=axes)
+        ser._name = None  # caller is responsible for setting real name
+
+        if type(self) is Series:
+            # This would also work `if self._constructor is Series`, but
+            #  this check is slightly faster, benefiting the most-common case.
+            return ser
+
+        # We assume that the subclass __init__ knows how to handle a
+        #  pd.Series object.
+        return self._constructor(ser)
+
+    @property
+    def _constructor_expanddim(self) -> Callable[..., DataFrame]:
+        """
+        Used when a manipulation result has one higher dimension as the
+        original, such as Series.to_frame()
+        """
+        from pandas.core.frame import DataFrame
+
+        return DataFrame
+
+    def _constructor_expanddim_from_mgr(self, mgr, axes):
+        from pandas.core.frame import DataFrame
+
+        df = DataFrame._from_mgr(mgr, axes=mgr.axes)
+
+        if type(self) is Series:
+            # This would also work `if self._constructor_expanddim is DataFrame`,
+            #  but this check is slightly faster, benefiting the most-common case.
+            return df
+
+        # We assume that the subclass __init__ knows how to handle a
+        #  pd.DataFrame object.
+        return self._constructor_expanddim(df)
+
+    # types
+    @property
+    def _can_hold_na(self) -> bool:
+        return self._mgr._can_hold_na
+
+    # ndarray compatibility
+    @property
+    def dtype(self) -> DtypeObj:
+        """
+        Return the dtype object of the underlying data.
+
+        See Also
+        --------
+        Series.dtypes : Return the dtype object of the underlying data.
+        Series.astype : Cast a pandas object to a specified dtype dtype.
+        Series.convert_dtypes : Convert columns to the best possible dtypes using dtypes
+            supporting pd.NA.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.dtype
+        dtype('int64')
+        """
+        return self._mgr.dtype
+
+    @property
+    def dtypes(self) -> DtypeObj:
+        """
+        Return the dtype object of the underlying data.
+
+        See Also
+        --------
+        DataFrame.dtypes :  Return the dtypes in the DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.dtypes
+        dtype('int64')
+        """
+        # DataFrame compatibility
+        return self.dtype
+
+    @property
+    def name(self) -> Hashable:
+        """
+        Return the name of the Series.
+
+        The name of a Series becomes its index or column name if it is used
+        to form a DataFrame. It is also used whenever displaying the Series
+        using the interpreter.
+
+        Returns
+        -------
+        label (hashable object)
+            The name of the Series, also the column name if part of a DataFrame.
+
+        See Also
+        --------
+        Series.rename : Sets the Series name when given a scalar input.
+        Index.name : Corresponding Index property.
+
+        Examples
+        --------
+        The Series name can be set initially when calling the constructor.
+
+        >>> s = pd.Series([1, 2, 3], dtype=np.int64, name="Numbers")
+        >>> s
+        0    1
+        1    2
+        2    3
+        Name: Numbers, dtype: int64
+        >>> s.name = "Integers"
+        >>> s
+        0    1
+        1    2
+        2    3
+        Name: Integers, dtype: int64
+
+        The name of a Series within a DataFrame is its column name.
+
+        >>> df = pd.DataFrame(
+        ...     [[1, 2], [3, 4], [5, 6]], columns=["Odd Numbers", "Even Numbers"]
+        ... )
+        >>> df
+           Odd Numbers  Even Numbers
+        0            1             2
+        1            3             4
+        2            5             6
+        >>> df["Even Numbers"].name
+        'Even Numbers'
+        """
+        return self._name
+
+    @name.setter
+    def name(self, value: Hashable) -> None:
+        validate_all_hashable(value, error_name=f"{type(self).__name__}.name")
+        object.__setattr__(self, "_name", value)
+
+    @property
+    def values(self):
+        """
+        Return Series as ndarray or ndarray-like depending on the dtype.
+
+        .. warning::
+
+           We recommend using :attr:`Series.array` or
+           :meth:`Series.to_numpy`, depending on whether you need
+           a reference to the underlying data or a NumPy array.
+
+        Returns
+        -------
+        numpy.ndarray or ndarray-like
+
+        See Also
+        --------
+        Series.array : Reference to the underlying data.
+        Series.to_numpy : A NumPy array representing the underlying data.
+
+        Examples
+        --------
+        >>> pd.Series([1, 2, 3]).values
+        array([1, 2, 3])
+
+        >>> pd.Series(list("aabc")).values
+        array(['a', 'a', 'b', 'c'], dtype=object)
+
+        >>> pd.Series(list("aabc")).astype("category").values
+        ['a', 'a', 'b', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+
+        Timezone aware datetime data is converted to UTC:
+
+        >>> pd.Series(pd.date_range("20130101", periods=3, tz="US/Eastern")).values
+        array(['2013-01-01T05:00:00.000000000',
+               '2013-01-02T05:00:00.000000000',
+               '2013-01-03T05:00:00.000000000'], dtype='datetime64[ns]')
+        """
+        return self._mgr.external_values()
+
+    @property
+    def _values(self):
+        """
+        Return the internal repr of this data (defined by Block.interval_values).
+        This are the values as stored in the Block (ndarray or ExtensionArray
+        depending on the Block class), with datetime64[ns] and timedelta64[ns]
+        wrapped in ExtensionArrays to match Index._values behavior.
+
+        Differs from the public ``.values`` for certain data types, because of
+        historical backwards compatibility of the public attribute (e.g. period
+        returns object ndarray and datetimetz a datetime64[ns] ndarray for
+        ``.values`` while it returns an ExtensionArray for ``._values`` in those
+        cases).
+
+        Differs from ``.array`` in that this still returns the numpy array if
+        the Block is backed by a numpy array (except for datetime64 and
+        timedelta64 dtypes), while ``.array`` ensures to always return an
+        ExtensionArray.
+
+        Overview:
+
+        dtype       | values        | _values       | array                 |
+        ----------- | ------------- | ------------- | --------------------- |
+        Numeric     | ndarray       | ndarray       | NumpyExtensionArray   |
+        Category    | Categorical   | Categorical   | Categorical           |
+        dt64[ns]    | ndarray[M8ns] | DatetimeArray | DatetimeArray         |
+        dt64[ns tz] | ndarray[M8ns] | DatetimeArray | DatetimeArray         |
+        td64[ns]    | ndarray[m8ns] | TimedeltaArray| TimedeltaArray        |
+        Period      | ndarray[obj]  | PeriodArray   | PeriodArray           |
+        Nullable    | EA            | EA            | EA                    |
+
+        """
+        return self._mgr.internal_values()
+
+    @property
+    def _references(self) -> BlockValuesRefs:
+        return self._mgr._block.refs
+
+    @Appender(base.IndexOpsMixin.array.__doc__)  # type: ignore[prop-decorator]
+    @property
+    def array(self) -> ExtensionArray:
+        return self._mgr.array_values()
+
+    def __len__(self) -> int:
+        """
+        Return the length of the Series.
+        """
+        return len(self._mgr)
+
+    # ----------------------------------------------------------------------
+    # NDArray Compat
+    def __array__(
+        self, dtype: npt.DTypeLike | None = None, copy: bool | None = None
+    ) -> np.ndarray:
+        """
+        Return the values as a NumPy array.
+
+        Users should not call this directly. Rather, it is invoked by
+        :func:`numpy.array` and :func:`numpy.asarray`.
+
+        Parameters
+        ----------
+        dtype : str or numpy.dtype, optional
+            The dtype to use for the resulting NumPy array. By default,
+            the dtype is inferred from the data.
+
+        copy : bool or None, optional
+            See :func:`numpy.asarray`.
+
+        Returns
+        -------
+        numpy.ndarray
+            The values in the series converted to a :class:`numpy.ndarray`
+            with the specified `dtype`.
+
+        See Also
+        --------
+        array : Create a new array from data.
+        Series.array : Zero-copy view to the array backing the Series.
+        Series.to_numpy : Series method for similar behavior.
+
+        Examples
+        --------
+        >>> ser = pd.Series([1, 2, 3])
+        >>> np.asarray(ser)
+        array([1, 2, 3])
+
+        For timezone-aware data, the timezones may be retained with
+        ``dtype='object'``
+
+        >>> tzser = pd.Series(pd.date_range("2000", periods=2, tz="CET"))
+        >>> np.asarray(tzser, dtype="object")
+        array([Timestamp('2000-01-01 00:00:00+0100', tz='CET'),
+               Timestamp('2000-01-02 00:00:00+0100', tz='CET')],
+              dtype=object)
+
+        Or the values may be localized to UTC and the tzinfo discarded with
+        ``dtype='datetime64[ns]'``
+
+        >>> np.asarray(tzser, dtype="datetime64[ns]")  # doctest: +ELLIPSIS
+        array(['1999-12-31T23:00:00.000000000', ...],
+              dtype='datetime64[ns]')
+        """
+        values = self._values
+        if copy is None:
+            # Note: branch avoids `copy=None` for NumPy 1.x support
+            arr = np.asarray(values, dtype=dtype)
+        else:
+            arr = np.array(values, dtype=dtype, copy=copy)
+
+        if copy is True:
+            return arr
+        if copy is False or astype_is_view(values.dtype, arr.dtype):
+            arr = arr.view()
+            arr.flags.writeable = False
+        return arr
+
+    # ----------------------------------------------------------------------
+
+    # indexers
+    @property
+    def axes(self) -> list[Index]:
+        """
+        Return a list of the row axis labels.
+        """
+        return [self.index]
+
+    # ----------------------------------------------------------------------
+    # Indexing Methods
+
+    def _ixs(self, i: int, axis: AxisInt = 0) -> Any:
+        """
+        Return the i-th value or values in the Series by location.
+
+        Parameters
+        ----------
+        i : int
+
+        Returns
+        -------
+        scalar
+        """
+        return self._values[i]
+
+    def _slice(self, slobj: slice, axis: AxisInt = 0) -> Series:
+        # axis kwarg is retained for compat with NDFrame method
+        #  _slice is *always* positional
+        mgr = self._mgr.get_slice(slobj, axis=axis)
+        out = self._constructor_from_mgr(mgr, axes=mgr.axes)
+        out._name = self._name
+        return out.__finalize__(self)
+
+    def __getitem__(self, key):
+        check_dict_or_set_indexers(key)
+        key = com.apply_if_callable(key, self)
+
+        if key is Ellipsis:
+            return self.copy(deep=False)
+
+        key_is_scalar = is_scalar(key)
+        if isinstance(key, (list, tuple)):
+            key = unpack_1tuple(key)
+
+        elif key_is_scalar:
+            # Note: GH#50617 in 3.0 we changed int key to always be treated as
+            #  a label, matching DataFrame behavior.
+            return self._get_value(key)
+
+        # Convert generator to list before going through hashable part
+        # (We will iterate through the generator there to check for slices)
+        if is_iterator(key):
+            key = list(key)
+
+        if is_hashable(key) and not isinstance(key, slice):
+            # Otherwise index.get_value will raise InvalidIndexError
+            try:
+                # For labels that don't resolve as scalars like tuples and frozensets
+                result = self._get_value(key)
+
+                return result
+
+            except (KeyError, TypeError, InvalidIndexError):
+                # InvalidIndexError for e.g. generator
+                #  see test_series_getitem_corner_generator
+                if isinstance(key, tuple) and isinstance(self.index, MultiIndex):
+                    # We still have the corner case where a tuple is a key
+                    # in the first level of our MultiIndex
+                    return self._get_values_tuple(key)
+
+        if isinstance(key, slice):
+            # Do slice check before somewhat-costly is_bool_indexer
+            return self._getitem_slice(key)
+
+        if com.is_bool_indexer(key):
+            key = check_bool_indexer(self.index, key)
+            key = np.asarray(key, dtype=bool)
+            return self._get_rows_with_mask(key)
+
+        return self._get_with(key)
+
+    def _get_with(self, key):
+        # other: fancy integer or otherwise
+        if isinstance(key, ABCDataFrame):
+            raise TypeError(
+                "Indexing a Series with DataFrame is not "
+                "supported, use the appropriate DataFrame column"
+            )
+        elif isinstance(key, tuple):
+            return self._get_values_tuple(key)
+
+        return self.loc[key]
+
+    def _get_values_tuple(self, key: tuple):
+        # mpl hackaround
+        if com.any_none(*key):
+            # mpl compat if we look up e.g. ser[:, np.newaxis];
+            #  see tests.series.timeseries.test_mpl_compat_hack
+            # the asarray is needed to avoid returning a 2D DatetimeArray
+            result = np.asarray(self._values[key])
+            disallow_ndim_indexing(result)
+            return result
+
+        if not isinstance(self.index, MultiIndex):
+            raise KeyError("key of type tuple not found and not a MultiIndex")
+
+        # If key is contained, would have returned by now
+        indexer, new_index = self.index.get_loc_level(key)
+        new_ser = self._constructor(self._values[indexer], index=new_index, copy=False)
+        if isinstance(indexer, slice):
+            new_ser._mgr.add_references(self._mgr)
+        return new_ser.__finalize__(self)
+
+    def _get_rows_with_mask(self, indexer: npt.NDArray[np.bool_]) -> Series:
+        new_mgr = self._mgr.get_rows_with_mask(indexer)
+        return self._constructor_from_mgr(new_mgr, axes=new_mgr.axes).__finalize__(self)
+
+    def _get_value(self, label, takeable: bool = False):
+        """
+        Quickly retrieve single value at passed index label.
+
+        Parameters
+        ----------
+        label : object
+        takeable : interpret the index as indexers, default False
+
+        Returns
+        -------
+        scalar value
+        """
+        if takeable:
+            return self._values[label]
+
+        # Similar to Index.get_value, but we do not fall back to positional
+        loc = self.index.get_loc(label)
+
+        if is_integer(loc):
+            return self._values[loc]
+
+        if isinstance(self.index, MultiIndex):
+            mi = self.index
+            new_values = self._values[loc]
+            if len(new_values) == 1 and mi.nlevels == 1:
+                # If more than one level left, we can not return a scalar
+                return new_values[0]
+
+            new_index = mi[loc]
+            new_index = maybe_droplevels(new_index, label)
+            new_ser = self._constructor(
+                new_values, index=new_index, name=self.name, copy=False
+            )
+            if isinstance(loc, slice):
+                new_ser._mgr.add_references(self._mgr)
+            return new_ser.__finalize__(self)
+
+        else:
+            return self.iloc[loc]
+
+    def __setitem__(self, key, value) -> None:
+        if not PYPY:
+            if sys.getrefcount(self) <= 3:
+                warnings.warn(
+                    _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
+                )
+
+        check_dict_or_set_indexers(key)
+        key = com.apply_if_callable(key, self)
+
+        if key is Ellipsis:
+            key = slice(None)
+
+        if isinstance(key, slice):
+            indexer = self.index._convert_slice_indexer(key, kind="getitem")
+            return self._set_values(indexer, value)
+
+        try:
+            self._set_with_engine(key, value)
+        except KeyError:
+            # We have a scalar (or for MultiIndex or object-dtype, scalar-like)
+            #  key that is not present in self.index.
+            # GH#12862 adding a new key to the Series
+            self.loc[key] = value
+
+        except (TypeError, ValueError, LossySetitemError):
+            # The key was OK, but we cannot set the value losslessly
+            indexer = self.index.get_loc(key)
+            self._set_values(indexer, value)
+
+        except InvalidIndexError as err:
+            if isinstance(key, tuple) and not isinstance(self.index, MultiIndex):
+                # cases with MultiIndex don't get here bc they raise KeyError
+                # e.g. test_basic_getitem_setitem_corner
+                raise KeyError(
+                    "key of type tuple not found and not a MultiIndex"
+                ) from err
+
+            if com.is_bool_indexer(key):
+                key = check_bool_indexer(self.index, key)
+                key = np.asarray(key, dtype=bool)
+
+                if (
+                    is_list_like(value)
+                    and len(value) != len(self)
+                    and not isinstance(value, Series)
+                    and not is_object_dtype(self.dtype)
+                ):
+                    # Series will be reindexed to have matching length inside
+                    #  _where call below
+                    # GH#44265
+                    indexer = key.nonzero()[0]
+                    self._set_values(indexer, value)
+                    return
+
+                # otherwise with listlike other we interpret series[mask] = other
+                #  as series[mask] = other[mask]
+                try:
+                    self._where(~key, value, inplace=True)
+                except InvalidIndexError:
+                    # test_where_dups
+                    self.iloc[key] = value
+                return
+
+            else:
+                self._set_with(key, value)
+
+    def _set_with_engine(self, key, value) -> None:
+        loc = self.index.get_loc(key)
+
+        # this is equivalent to self._values[key] = value
+        self._mgr.setitem_inplace(loc, value)
+
+    def _set_with(self, key, value) -> None:
+        # We got here via exception-handling off of InvalidIndexError, so
+        #  key should always be listlike at this point.
+        assert not isinstance(key, tuple)
+
+        if is_iterator(key):
+            # Without this, the call to infer_dtype will consume the generator
+            key = list(key)
+
+        self._set_labels(key, value)
+
+    def _set_labels(self, key, value) -> None:
+        key = com.asarray_tuplesafe(key)
+        indexer: np.ndarray = self.index.get_indexer(key)
+        mask = indexer == -1
+        if mask.any():
+            raise KeyError(f"{key[mask]} not in index")
+        self._set_values(indexer, value)
+
+    def _set_values(self, key, value) -> None:
+        if isinstance(key, (Index, Series)):
+            key = key._values
+
+        self._mgr = self._mgr.setitem(indexer=key, value=value)
+
+    def _set_value(self, label, value, takeable: bool = False) -> None:
+        """
+        Quickly set single value at passed label.
+
+        If label is not contained, a new object is created with the label
+        placed at the end of the result index.
+
+        Parameters
+        ----------
+        label : object
+            Partial indexing with MultiIndex not allowed.
+        value : object
+            Scalar value.
+        takeable : interpret the index as indexers, default False
+        """
+        if not takeable:
+            try:
+                loc = self.index.get_loc(label)
+            except KeyError:
+                # set using a non-recursive method
+                self.loc[label] = value
+                return
+        else:
+            loc = label
+
+        self._set_values(loc, value)
+
+    # ----------------------------------------------------------------------
+    # Unsorted
+
+    def repeat(self, repeats: int | Sequence[int], axis: None = None) -> Series:
+        """
+        Repeat elements of a Series.
+
+        Returns a new Series where each element of the current Series
+        is repeated consecutively a given number of times.
+
+        Parameters
+        ----------
+        repeats : int or array of ints
+            The number of repetitions for each element. This should be a
+            non-negative integer. Repeating 0 times will return an empty
+            Series.
+        axis : None
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            Newly created Series with repeated elements.
+
+        See Also
+        --------
+        Index.repeat : Equivalent function for Index.
+        numpy.repeat : Similar method for :class:`numpy.ndarray`.
+
+        Examples
+        --------
+        >>> s = pd.Series(["a", "b", "c"])
+        >>> s
+        0    a
+        1    b
+        2    c
+        dtype: object
+        >>> s.repeat(2)
+        0    a
+        0    a
+        1    b
+        1    b
+        2    c
+        2    c
+        dtype: object
+        >>> s.repeat([1, 2, 3])
+        0    a
+        1    b
+        1    b
+        2    c
+        2    c
+        2    c
+        dtype: object
+        """
+        nv.validate_repeat((), {"axis": axis})
+        new_index = self.index.repeat(repeats)
+        new_values = self._values.repeat(repeats)
+        return self._constructor(new_values, index=new_index, copy=False).__finalize__(
+            self, method="repeat"
+        )
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: Literal[False] = ...,
+        name: Level = ...,
+        inplace: Literal[False] = ...,
+        allow_duplicates: bool = ...,
+    ) -> DataFrame: ...
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: Literal[True],
+        name: Level = ...,
+        inplace: Literal[False] = ...,
+        allow_duplicates: bool = ...,
+    ) -> Series: ...
+
+    @overload
+    def reset_index(
+        self,
+        level: IndexLabel = ...,
+        *,
+        drop: bool = ...,
+        name: Level = ...,
+        inplace: Literal[True],
+        allow_duplicates: bool = ...,
+    ) -> None: ...
+
+    def reset_index(
+        self,
+        level: IndexLabel | None = None,
+        *,
+        drop: bool = False,
+        name: Level = lib.no_default,
+        inplace: bool = False,
+        allow_duplicates: bool = False,
+    ) -> DataFrame | Series | None:
+        """
+        Generate a new DataFrame or Series with the index reset.
+
+        This is useful when the index needs to be treated as a column, or
+        when the index is meaningless and needs to be reset to the default
+        before another operation.
+
+        Parameters
+        ----------
+        level : int, str, tuple, or list, default optional
+            For a Series with a MultiIndex, only remove the specified levels
+            from the index. Removes all levels by default.
+        drop : bool, default False
+            Just reset the index, without inserting it as a column in
+            the new DataFrame.
+        name : object, optional
+            The name to use for the column containing the original Series
+            values. Uses ``self.name`` by default. This argument is ignored
+            when `drop` is True.
+        inplace : bool, default False
+            Modify the Series in place (do not create a new object).
+        allow_duplicates : bool, default False
+            Allow duplicate column labels to be created.
+
+            .. versionadded:: 1.5.0
+
+        Returns
+        -------
+        Series or DataFrame or None
+            When `drop` is False (the default), a DataFrame is returned.
+            The newly created columns will come first in the DataFrame,
+            followed by the original Series values.
+            When `drop` is True, a `Series` is returned.
+            In either case, if ``inplace=True``, no value is returned.
+
+        See Also
+        --------
+        DataFrame.reset_index: Analogous function for DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     name="foo",
+        ...     index=pd.Index(["a", "b", "c", "d"], name="idx"),
+        ... )
+
+        Generate a DataFrame with default index.
+
+        >>> s.reset_index()
+          idx  foo
+        0   a    1
+        1   b    2
+        2   c    3
+        3   d    4
+
+        To specify the name of the new column use `name`.
+
+        >>> s.reset_index(name="values")
+          idx  values
+        0   a       1
+        1   b       2
+        2   c       3
+        3   d       4
+
+        To generate a new Series with the default set `drop` to True.
+
+        >>> s.reset_index(drop=True)
+        0    1
+        1    2
+        2    3
+        3    4
+        Name: foo, dtype: int64
+
+        The `level` parameter is interesting for Series with a multi-level
+        index.
+
+        >>> arrays = [
+        ...     np.array(["bar", "bar", "baz", "baz"]),
+        ...     np.array(["one", "two", "one", "two"]),
+        ... ]
+        >>> s2 = pd.Series(
+        ...     range(4),
+        ...     name="foo",
+        ...     index=pd.MultiIndex.from_arrays(arrays, names=["a", "b"]),
+        ... )
+
+        To remove a specific level from the Index, use `level`.
+
+        >>> s2.reset_index(level="a")
+               a  foo
+        b
+        one  bar    0
+        two  bar    1
+        one  baz    2
+        two  baz    3
+
+        If `level` is not set, all levels are removed from the Index.
+
+        >>> s2.reset_index()
+             a    b  foo
+        0  bar  one    0
+        1  bar  two    1
+        2  baz  one    2
+        3  baz  two    3
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        if drop:
+            new_index = default_index(len(self))
+            if level is not None:
+                level_list: Sequence[Hashable]
+                if not isinstance(level, (tuple, list)):
+                    level_list = [level]
+                else:
+                    level_list = level
+                level_list = [self.index._get_level_number(lev) for lev in level_list]
+                if len(level_list) < self.index.nlevels:
+                    new_index = self.index.droplevel(level_list)
+
+            if inplace:
+                self.index = new_index
+            else:
+                new_ser = self.copy(deep=False)
+                new_ser.index = new_index
+                return new_ser.__finalize__(self, method="reset_index")
+        elif inplace:
+            raise TypeError(
+                "Cannot reset_index inplace on a Series to create a DataFrame"
+            )
+        else:
+            if name is lib.no_default:
+                # For backwards compatibility, keep columns as [0] instead of
+                #  [None] when self.name is None
+                if self.name is None:
+                    name = 0
+                else:
+                    name = self.name
+
+            df = self.to_frame(name)
+            return df.reset_index(
+                level=level, drop=drop, allow_duplicates=allow_duplicates
+            )
+        return None
+
+    # ----------------------------------------------------------------------
+    # Rendering Methods
+
+    def __repr__(self) -> str:
+        """
+        Return a string representation for a particular Series.
+        """
+        repr_params = fmt.get_series_repr_params()
+        return self.to_string(**repr_params)
+
+    @overload
+    def to_string(
+        self,
+        buf: None = ...,
+        *,
+        na_rep: str = ...,
+        float_format: str | None = ...,
+        header: bool = ...,
+        index: bool = ...,
+        length: bool = ...,
+        dtype=...,
+        name=...,
+        max_rows: int | None = ...,
+        min_rows: int | None = ...,
+    ) -> str: ...
+
+    @overload
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        na_rep: str = ...,
+        float_format: str | None = ...,
+        header: bool = ...,
+        index: bool = ...,
+        length: bool = ...,
+        dtype=...,
+        name=...,
+        max_rows: int | None = ...,
+        min_rows: int | None = ...,
+    ) -> None: ...
+
+    @deprecate_nonkeyword_arguments(
+        version="4.0", allowed_args=["self", "buf"], name="to_string"
+    )
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        na_rep: str = "NaN",
+        float_format: str | None = None,
+        header: bool = True,
+        index: bool = True,
+        length: bool = False,
+        dtype: bool = False,
+        name: bool = False,
+        max_rows: int | None = None,
+        min_rows: int | None = None,
+    ) -> str | None:
+        """
+        Render a string representation of the Series.
+
+        Parameters
+        ----------
+        buf : StringIO-like, optional
+            Buffer to write to.
+        na_rep : str, optional
+            String representation of NaN to use, default 'NaN'.
+        float_format : one-parameter function, optional
+            Formatter function to apply to columns' elements if they are
+            floats, default None.
+        header : bool, default True
+            Add the Series header (index name).
+        index : bool, optional
+            Add index (row) labels, default True.
+        length : bool, default False
+            Add the Series length.
+        dtype : bool, default False
+            Add the Series dtype.
+        name : bool, default False
+            Add the Series name if not None.
+        max_rows : int, optional
+            Maximum number of rows to show before truncating. If None, show
+            all.
+        min_rows : int, optional
+            The number of rows to display in a truncated repr (when number
+            of rows is above `max_rows`).
+
+        Returns
+        -------
+        str or None
+            String representation of Series if ``buf=None``, otherwise None.
+
+        See Also
+        --------
+        Series.to_dict : Convert Series to dict object.
+        Series.to_frame : Convert Series to DataFrame object.
+        Series.to_markdown : Print Series in Markdown-friendly format.
+        Series.to_timestamp : Cast to DatetimeIndex of Timestamps.
+
+        Examples
+        --------
+        >>> ser = pd.Series([1, 2, 3]).to_string()
+        >>> ser
+        '0    1\\n1    2\\n2    3'
+        """
+        formatter = fmt.SeriesFormatter(
+            self,
+            name=name,
+            length=length,
+            header=header,
+            index=index,
+            dtype=dtype,
+            na_rep=na_rep,
+            float_format=float_format,
+            min_rows=min_rows,
+            max_rows=max_rows,
+        )
+        result = formatter.to_string()
+
+        # catch contract violations
+        if not isinstance(result, str):
+            raise AssertionError(
+                "result must be of type str, type "
+                f"of result is {type(result).__name__!r}"
+            )
+
+        if buf is None:
+            return result
+        else:
+            if hasattr(buf, "write"):
+                buf.write(result)
+            else:
+                with open(buf, "w", encoding="utf-8") as f:
+                    f.write(result)
+        return None
+
+    @overload
+    def to_markdown(
+        self,
+        buf: None = ...,
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> str: ...
+
+    @overload
+    def to_markdown(
+        self,
+        buf: IO[str],
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> None: ...
+
+    @overload
+    def to_markdown(
+        self,
+        buf: IO[str] | None,
+        *,
+        mode: str = ...,
+        index: bool = ...,
+        storage_options: StorageOptions | None = ...,
+        **kwargs,
+    ) -> str | None: ...
+
+    @doc(
+        klass=_shared_doc_kwargs["klass"],
+        storage_options=_shared_docs["storage_options"],
+        examples=dedent(
+            """Examples
+            --------
+            >>> s = pd.Series(["elk", "pig", "dog", "quetzal"], name="animal")
+            >>> print(s.to_markdown())
+            |    | animal   |
+            |---:|:---------|
+            |  0 | elk      |
+            |  1 | pig      |
+            |  2 | dog      |
+            |  3 | quetzal  |
+
+            Output markdown with a tabulate option.
+
+            >>> print(s.to_markdown(tablefmt="grid"))
+            +----+----------+
+            |    | animal   |
+            +====+==========+
+            |  0 | elk      |
+            +----+----------+
+            |  1 | pig      |
+            +----+----------+
+            |  2 | dog      |
+            +----+----------+
+            |  3 | quetzal  |
+            +----+----------+"""
+        ),
+    )
+    @deprecate_nonkeyword_arguments(
+        version="4.0", allowed_args=["self", "buf"], name="to_markdown"
+    )
+    def to_markdown(
+        self,
+        buf: IO[str] | None = None,
+        mode: str = "wt",
+        index: bool = True,
+        storage_options: StorageOptions | None = None,
+        **kwargs,
+    ) -> str | None:
+        """
+        Print {klass} in Markdown-friendly format.
+
+        Parameters
+        ----------
+        buf : str, Path or StringIO-like, optional, default None
+            Buffer to write to. If None, the output is returned as a string.
+        mode : str, optional
+            Mode in which file is opened, "wt" by default.
+        index : bool, optional, default True
+            Add index (row) labels.
+
+        {storage_options}
+
+        **kwargs
+            These parameters will be passed to `tabulate \
+                <https://pypi.org/project/tabulate>`_.
+
+        Returns
+        -------
+        str
+            {klass} in Markdown-friendly format.
+
+        See Also
+        --------
+        Series.to_frame : Rrite a text representation of object to the system clipboard.
+        Series.to_latex : Render Series to LaTeX-formatted table.
+
+        Notes
+        -----
+        Requires the `tabulate <https://pypi.org/project/tabulate>`_ package.
+
+        {examples}
+        """
+        return self.to_frame().to_markdown(
+            buf, mode=mode, index=index, storage_options=storage_options, **kwargs
+        )
+
+    # ----------------------------------------------------------------------
+
+    def items(self) -> Iterable[tuple[Hashable, Any]]:
+        """
+        Lazily iterate over (index, value) tuples.
+
+        This method returns an iterable tuple (index, value). This is
+        convenient if you want to create a lazy iterator.
+
+        Returns
+        -------
+        iterable
+            Iterable of tuples containing the (index, value) pairs from a
+            Series.
+
+        See Also
+        --------
+        DataFrame.items : Iterate over (column name, Series) pairs.
+        DataFrame.iterrows : Iterate over DataFrame rows as (index, Series) pairs.
+
+        Examples
+        --------
+        >>> s = pd.Series(["A", "B", "C"])
+        >>> for index, value in s.items():
+        ...     print(f"Index : {index}, Value : {value}")
+        Index : 0, Value : A
+        Index : 1, Value : B
+        Index : 2, Value : C
+        """
+        return zip(iter(self.index), iter(self))
+
+    # ----------------------------------------------------------------------
+    # Misc public methods
+
+    def keys(self) -> Index:
+        """
+        Return alias for index.
+
+        Returns
+        -------
+        Index
+            Index of the Series.
+
+        See Also
+        --------
+        Series.index : The index (axis labels) of the Series.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3], index=[0, 1, 2])
+        >>> s.keys()
+        Index([0, 1, 2], dtype='int64')
+        """
+        return self.index
+
+    @overload
+    def to_dict(
+        self, *, into: type[MutableMappingT] | MutableMappingT
+    ) -> MutableMappingT: ...
+
+    @overload
+    def to_dict(self, *, into: type[dict] = ...) -> dict: ...
+
+    # error: Incompatible default for argument "into" (default has type "type[
+    # dict[Any, Any]]", argument has type "type[MutableMappingT] | MutableMappingT")
+    def to_dict(
+        self,
+        *,
+        into: type[MutableMappingT] | MutableMappingT = dict,  # type: ignore[assignment]
+    ) -> MutableMappingT:
+        """
+        Convert Series to {label -> value} dict or dict-like object.
+
+        Parameters
+        ----------
+        into : class, default dict
+            The collections.abc.MutableMapping subclass to use as the return
+            object. Can be the actual class or an empty instance of the mapping
+            type you want.  If you want a collections.defaultdict, you must
+            pass it initialized.
+
+        Returns
+        -------
+        collections.abc.MutableMapping
+            Key-value representation of Series.
+
+        See Also
+        --------
+        Series.to_list: Converts Series to a list of the values.
+        Series.to_numpy: Converts Series to NumPy ndarray.
+        Series.array: ExtensionArray of the data backing this Series.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s.to_dict()
+        {0: 1, 1: 2, 2: 3, 3: 4}
+        >>> from collections import OrderedDict, defaultdict
+        >>> s.to_dict(into=OrderedDict)
+        OrderedDict([(0, 1), (1, 2), (2, 3), (3, 4)])
+        >>> dd = defaultdict(list)
+        >>> s.to_dict(into=dd)
+        defaultdict(<class 'list'>, {0: 1, 1: 2, 2: 3, 3: 4})
+        """
+        # GH16122
+        into_c = com.standardize_mapping(into)
+
+        if is_object_dtype(self.dtype) or isinstance(self.dtype, ExtensionDtype):
+            return into_c((k, maybe_box_native(v)) for k, v in self.items())
+        else:
+            # Not an object dtype => all types will be the same so let the default
+            # indexer return native python type
+            return into_c(self.items())
+
+    def to_frame(self, name: Hashable = lib.no_default) -> DataFrame:
+        """
+        Convert Series to DataFrame.
+
+        Parameters
+        ----------
+        name : object, optional
+            The passed name should substitute for the series name (if it has
+            one).
+
+        Returns
+        -------
+        DataFrame
+            DataFrame representation of Series.
+
+        See Also
+        --------
+        Series.to_dict : Convert Series to dict object.
+
+        Examples
+        --------
+        >>> s = pd.Series(["a", "b", "c"], name="vals")
+        >>> s.to_frame()
+          vals
+        0    a
+        1    b
+        2    c
+        """
+        columns: Index
+        if name is lib.no_default:
+            name = self.name
+            if name is None:
+                # default to [0], same as we would get with DataFrame(self)
+                columns = default_index(1)
+            else:
+                columns = Index([name])
+        else:
+            columns = Index([name])
+
+        mgr = self._mgr.to_2d_mgr(columns)
+        df = self._constructor_expanddim_from_mgr(mgr, axes=mgr.axes)
+        return df.__finalize__(self, method="to_frame")
+
+    def _set_name(
+        self, name, inplace: bool = False, deep: bool | None = None
+    ) -> Series:
+        """
+        Set the Series name.
+
+        Parameters
+        ----------
+        name : str
+        inplace : bool
+            Whether to modify `self` directly or return a copy.
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        ser = self if inplace else self.copy(deep=False)
+        ser.name = name
+        return ser
+
+    @Appender(
+        dedent(
+            """
+        Examples
+        --------
+        >>> ser = pd.Series([390., 350., 30., 20.],
+        ...                 index=['Falcon', 'Falcon', 'Parrot', 'Parrot'],
+        ...                 name="Max Speed")
+        >>> ser
+        Falcon    390.0
+        Falcon    350.0
+        Parrot     30.0
+        Parrot     20.0
+        Name: Max Speed, dtype: float64
+
+        We can pass a list of values to group the Series data by custom labels:
+
+        >>> ser.groupby(["a", "b", "a", "b"]).mean()
+        a    210.0
+        b    185.0
+        Name: Max Speed, dtype: float64
+
+        Grouping by numeric labels yields similar results:
+
+        >>> ser.groupby([0, 1, 0, 1]).mean()
+        0    210.0
+        1    185.0
+        Name: Max Speed, dtype: float64
+
+        We can group by a level of the index:
+
+        >>> ser.groupby(level=0).mean()
+        Falcon    370.0
+        Parrot     25.0
+        Name: Max Speed, dtype: float64
+
+        We can group by a condition applied to the Series values:
+
+        >>> ser.groupby(ser > 100).mean()
+        Max Speed
+        False     25.0
+        True     370.0
+        Name: Max Speed, dtype: float64
+
+        **Grouping by Indexes**
+
+        We can groupby different levels of a hierarchical index
+        using the `level` parameter:
+
+        >>> arrays = [['Falcon', 'Falcon', 'Parrot', 'Parrot'],
+        ...           ['Captive', 'Wild', 'Captive', 'Wild']]
+        >>> index = pd.MultiIndex.from_arrays(arrays, names=('Animal', 'Type'))
+        >>> ser = pd.Series([390., 350., 30., 20.], index=index, name="Max Speed")
+        >>> ser
+        Animal  Type
+        Falcon  Captive    390.0
+                Wild       350.0
+        Parrot  Captive     30.0
+                Wild        20.0
+        Name: Max Speed, dtype: float64
+
+        >>> ser.groupby(level=0).mean()
+        Animal
+        Falcon    370.0
+        Parrot     25.0
+        Name: Max Speed, dtype: float64
+
+        We can also group by the 'Type' level of the hierarchical index
+        to get the mean speed for each type:
+
+        >>> ser.groupby(level="Type").mean()
+        Type
+        Captive    210.0
+        Wild       185.0
+        Name: Max Speed, dtype: float64
+
+        We can also choose to include `NA` in group keys or not by defining
+        `dropna` parameter, the default setting is `True`.
+
+        >>> ser = pd.Series([1, 2, 3, 3], index=["a", 'a', 'b', np.nan])
+        >>> ser.groupby(level=0).sum()
+        a    3
+        b    3
+        dtype: int64
+
+        To include `NA` values in the group keys, set `dropna=False`:
+
+        >>> ser.groupby(level=0, dropna=False).sum()
+        a    3
+        b    3
+        NaN  3
+        dtype: int64
+
+        We can also group by a custom list with NaN values to handle
+        missing group labels:
+
+        >>> arrays = ['Falcon', 'Falcon', 'Parrot', 'Parrot']
+        >>> ser = pd.Series([390., 350., 30., 20.], index=arrays, name="Max Speed")
+        >>> ser.groupby(["a", "b", "a", np.nan]).mean()
+        a    210.0
+        b    350.0
+        Name: Max Speed, dtype: float64
+
+        >>> ser.groupby(["a", "b", "a", np.nan], dropna=False).mean()
+        a    210.0
+        b    350.0
+        NaN   20.0
+        Name: Max Speed, dtype: float64
+        """
+        )
+    )
+    @Appender(_shared_docs["groupby"] % _shared_doc_kwargs)
+    def groupby(
+        self,
+        by=None,
+        level: IndexLabel | None = None,
+        as_index: bool = True,
+        sort: bool = True,
+        group_keys: bool = True,
+        observed: bool = False,
+        dropna: bool = True,
+    ) -> SeriesGroupBy:
+        from pandas.core.groupby.generic import SeriesGroupBy
+
+        if level is None and by is None:
+            raise TypeError("You have to supply one of 'by' and 'level'")
+        if not as_index:
+            raise TypeError("as_index=False only valid with DataFrame")
+
+        return SeriesGroupBy(
+            obj=self,
+            keys=by,
+            level=level,
+            as_index=as_index,
+            sort=sort,
+            group_keys=group_keys,
+            observed=observed,
+            dropna=dropna,
+        )
+
+    # ----------------------------------------------------------------------
+    # Statistics, overridden ndarray methods
+
+    # TODO: integrate bottleneck
+    def count(self) -> int:
+        """
+        Return number of non-NA/null observations in the Series.
+
+        Returns
+        -------
+        int
+            Number of non-null values in the Series.
+
+        See Also
+        --------
+        DataFrame.count : Count non-NA cells for each column or row.
+
+        Examples
+        --------
+        >>> s = pd.Series([0.0, 1.0, np.nan])
+        >>> s.count()
+        2
+        """
+        return notna(self._values).sum().astype("int64")
+
+    def mode(self, dropna: bool = True) -> Series:
+        """
+        Return the mode(s) of the Series.
+
+        The mode is the value that appears most often. There can be multiple modes.
+
+        Always returns Series even if only one value is returned.
+
+        Parameters
+        ----------
+        dropna : bool, default True
+            Don't consider counts of NaN/NaT.
+
+        Returns
+        -------
+        Series
+            Modes of the Series in sorted order.
+
+        See Also
+        --------
+        numpy.mode : Equivalent numpy function for computing median.
+        Series.sum : Sum of the values.
+        Series.median : Median of the values.
+        Series.std : Standard deviation of the values.
+        Series.var : Variance of the values.
+        Series.min : Minimum value.
+        Series.max : Maximum value.
+
+        Examples
+        --------
+        >>> s = pd.Series([2, 4, 2, 2, 4, None])
+        >>> s.mode()
+        0    2.0
+        dtype: float64
+
+        More than one mode:
+
+        >>> s = pd.Series([2, 4, 8, 2, 4, None])
+        >>> s.mode()
+        0    2.0
+        1    4.0
+        dtype: float64
+
+        With and without considering null value:
+
+        >>> s = pd.Series([2, 4, None, None, 4, None])
+        >>> s.mode(dropna=False)
+        0   NaN
+        dtype: float64
+        >>> s = pd.Series([2, 4, None, None, 4, None])
+        >>> s.mode()
+        0    4.0
+        dtype: float64
+        """
+        # TODO: Add option for bins like value_counts()
+        values = self._values
+        if isinstance(values, np.ndarray):
+            res_values = algorithms.mode(values, dropna=dropna)
+        else:
+            res_values = values._mode(dropna=dropna)
+
+        # Ensure index is type stable (should always use int index)
+        return self._constructor(
+            res_values,
+            index=range(len(res_values)),
+            name=self.name,
+            copy=False,
+            dtype=self.dtype,
+        ).__finalize__(self, method="mode")
+
+    def unique(self) -> ArrayLike:
+        """
+        Return unique values of Series object.
+
+        Uniques are returned in order of appearance. Hash table-based unique,
+        therefore does NOT sort.
+
+        Returns
+        -------
+        ndarray or ExtensionArray
+            The unique values returned as a NumPy array. See Notes.
+
+        See Also
+        --------
+        Series.drop_duplicates : Return Series with duplicate values removed.
+        unique : Top-level unique method for any 1-d array-like object.
+        Index.unique : Return Index with unique values from an Index object.
+
+        Notes
+        -----
+        Returns the unique values as a NumPy array. In case of an
+        extension-array backed Series, a new
+        :class:`~api.extensions.ExtensionArray` of that type with just
+        the unique values is returned. This includes
+
+            * Categorical
+            * Period
+            * Datetime with Timezone
+            * Datetime without Timezone
+            * Timedelta
+            * Interval
+            * Sparse
+            * IntegerNA
+
+        See Examples section.
+
+        Examples
+        --------
+        >>> pd.Series([2, 1, 3, 3], name="A").unique()
+        array([2, 1, 3])
+
+        >>> pd.Series([pd.Timestamp("2016-01-01") for _ in range(3)]).unique()
+        <DatetimeArray>
+        ['2016-01-01 00:00:00']
+        Length: 1, dtype: datetime64[s]
+
+        >>> pd.Series(
+        ...     [pd.Timestamp("2016-01-01", tz="US/Eastern") for _ in range(3)]
+        ... ).unique()
+        <DatetimeArray>
+        ['2016-01-01 00:00:00-05:00']
+        Length: 1, dtype: datetime64[s, US/Eastern]
+
+        An Categorical will return categories in the order of
+        appearance and with the same dtype.
+
+        >>> pd.Series(pd.Categorical(list("baabc"))).unique()
+        ['b', 'a', 'c']
+        Categories (3, object): ['a', 'b', 'c']
+        >>> pd.Series(
+        ...     pd.Categorical(list("baabc"), categories=list("abc"), ordered=True)
+        ... ).unique()
+        ['b', 'a', 'c']
+        Categories (3, object): ['a' < 'b' < 'c']
+        """
+        return super().unique()
+
+    @overload
+    def drop_duplicates(
+        self,
+        *,
+        keep: DropKeep = ...,
+        inplace: Literal[False] = ...,
+        ignore_index: bool = ...,
+    ) -> Series: ...
+
+    @overload
+    def drop_duplicates(
+        self, *, keep: DropKeep = ..., inplace: Literal[True], ignore_index: bool = ...
+    ) -> None: ...
+
+    @overload
+    def drop_duplicates(
+        self, *, keep: DropKeep = ..., inplace: bool = ..., ignore_index: bool = ...
+    ) -> Series | None: ...
+
+    def drop_duplicates(
+        self,
+        *,
+        keep: DropKeep = "first",
+        inplace: bool = False,
+        ignore_index: bool = False,
+    ) -> Series | None:
+        """
+        Return Series with duplicate values removed.
+
+        Parameters
+        ----------
+        keep : {'first', 'last', ``False``}, default 'first'
+            Method to handle dropping duplicates:
+
+            - 'first' : Drop duplicates except for the first occurrence.
+            - 'last' : Drop duplicates except for the last occurrence.
+            - ``False`` : Drop all duplicates.
+
+        inplace : bool, default ``False``
+            If ``True``, performs operation inplace and returns None.
+
+        ignore_index : bool, default ``False``
+            If ``True``, the resulting axis will be labeled 0, 1, …, n - 1.
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        Series or None
+            Series with duplicates dropped or None if ``inplace=True``.
+
+        See Also
+        --------
+        Index.drop_duplicates : Equivalent method on Index.
+        DataFrame.drop_duplicates : Equivalent method on DataFrame.
+        Series.duplicated : Related method on Series, indicating duplicate
+            Series values.
+        Series.unique : Return unique values as an array.
+
+        Examples
+        --------
+        Generate a Series with duplicated entries.
+
+        >>> s = pd.Series(
+        ...     ["llama", "cow", "llama", "beetle", "llama", "hippo"], name="animal"
+        ... )
+        >>> s
+        0     llama
+        1       cow
+        2     llama
+        3    beetle
+        4     llama
+        5     hippo
+        Name: animal, dtype: object
+
+        With the 'keep' parameter, the selection behavior of duplicated values
+        can be changed. The value 'first' keeps the first occurrence for each
+        set of duplicated entries. The default value of keep is 'first'.
+
+        >>> s.drop_duplicates()
+        0     llama
+        1       cow
+        3    beetle
+        5     hippo
+        Name: animal, dtype: object
+
+        The value 'last' for parameter 'keep' keeps the last occurrence for
+        each set of duplicated entries.
+
+        >>> s.drop_duplicates(keep="last")
+        1       cow
+        3    beetle
+        4     llama
+        5     hippo
+        Name: animal, dtype: object
+
+        The value ``False`` for parameter 'keep' discards all sets of
+        duplicated entries.
+
+        >>> s.drop_duplicates(keep=False)
+        1       cow
+        3    beetle
+        5     hippo
+        Name: animal, dtype: object
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        result = super().drop_duplicates(keep=keep)
+
+        if ignore_index:
+            result.index = default_index(len(result))
+
+        if inplace:
+            self._update_inplace(result)
+            return None
+        else:
+            return result
+
+    def duplicated(self, keep: DropKeep = "first") -> Series:
+        """
+        Indicate duplicate Series values.
+
+        Duplicated values are indicated as ``True`` values in the resulting
+        Series. Either all duplicates, all except the first or all except the
+        last occurrence of duplicates can be indicated.
+
+        Parameters
+        ----------
+        keep : {'first', 'last', False}, default 'first'
+            Method to handle dropping duplicates:
+
+            - 'first' : Mark duplicates as ``True`` except for the first
+              occurrence.
+            - 'last' : Mark duplicates as ``True`` except for the last
+              occurrence.
+            - ``False`` : Mark all duplicates as ``True``.
+
+        Returns
+        -------
+        Series[bool]
+            Series indicating whether each value has occurred in the
+            preceding values.
+
+        See Also
+        --------
+        Index.duplicated : Equivalent method on pandas.Index.
+        DataFrame.duplicated : Equivalent method on pandas.DataFrame.
+        Series.drop_duplicates : Remove duplicate values from Series.
+
+        Examples
+        --------
+        By default, for each set of duplicated values, the first occurrence is
+        set on False and all others on True:
+
+        >>> animals = pd.Series(["llama", "cow", "llama", "beetle", "llama"])
+        >>> animals.duplicated()
+        0    False
+        1    False
+        2     True
+        3    False
+        4     True
+        dtype: bool
+
+        which is equivalent to
+
+        >>> animals.duplicated(keep="first")
+        0    False
+        1    False
+        2     True
+        3    False
+        4     True
+        dtype: bool
+
+        By using 'last', the last occurrence of each set of duplicated values
+        is set on False and all others on True:
+
+        >>> animals.duplicated(keep="last")
+        0     True
+        1    False
+        2     True
+        3    False
+        4    False
+        dtype: bool
+
+        By setting keep on ``False``, all duplicates are True:
+
+        >>> animals.duplicated(keep=False)
+        0     True
+        1    False
+        2     True
+        3    False
+        4     True
+        dtype: bool
+        """
+        res = self._duplicated(keep=keep)
+        result = self._constructor(res, index=self.index, copy=False)
+        return result.__finalize__(self, method="duplicated")
+
+    def idxmin(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Hashable:
+        """
+        Return the row label of the minimum value.
+
+        If multiple values equal the minimum, the first row label with that
+        value is returned.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        skipna : bool, default True
+            Exclude NA/null values. If the entire Series is NA, or if ``skipna=False``
+            and there is an NA value, this method will raise a ``ValueError``.
+        *args, **kwargs
+            Additional arguments and keywords have no effect but might be
+            accepted for compatibility with NumPy.
+
+        Returns
+        -------
+        Index
+            Label of the minimum value.
+
+        Raises
+        ------
+        ValueError
+            If the Series is empty.
+
+        See Also
+        --------
+        numpy.argmin : Return indices of the minimum values
+            along the given axis.
+        DataFrame.idxmin : Return index of first occurrence of minimum
+            over requested axis.
+        Series.idxmax : Return index *label* of the first occurrence
+            of maximum of values.
+
+        Notes
+        -----
+        This method is the Series version of ``ndarray.argmin``. This method
+        returns the label of the minimum, while ``ndarray.argmin`` returns
+        the position. To get the position, use ``series.values.argmin()``.
+
+        Examples
+        --------
+        >>> s = pd.Series(data=[1, None, 4, 1], index=["A", "B", "C", "D"])
+        >>> s
+        A    1.0
+        B    NaN
+        C    4.0
+        D    1.0
+        dtype: float64
+
+        >>> s.idxmin()
+        'A'
+        """
+        axis = self._get_axis_number(axis)
+        iloc = self.argmin(axis, skipna, *args, **kwargs)
+        return self.index[iloc]
+
+    def idxmax(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Hashable:
+        """
+        Return the row label of the maximum value.
+
+        If multiple values equal the maximum, the first row label with that
+        value is returned.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        skipna : bool, default True
+            Exclude NA/null values. If the entire Series is NA, or if ``skipna=False``
+            and there is an NA value, this method will raise a ``ValueError``.
+        *args, **kwargs
+            Additional arguments and keywords have no effect but might be
+            accepted for compatibility with NumPy.
+
+        Returns
+        -------
+        Index
+            Label of the maximum value.
+
+        Raises
+        ------
+        ValueError
+            If the Series is empty.
+
+        See Also
+        --------
+        numpy.argmax : Return indices of the maximum values
+            along the given axis.
+        DataFrame.idxmax : Return index of first occurrence of maximum
+            over requested axis.
+        Series.idxmin : Return index *label* of the first occurrence
+            of minimum of values.
+
+        Notes
+        -----
+        This method is the Series version of ``ndarray.argmax``. This method
+        returns the label of the maximum, while ``ndarray.argmax`` returns
+        the position. To get the position, use ``series.values.argmax()``.
+
+        Examples
+        --------
+        >>> s = pd.Series(data=[1, None, 4, 3, 4], index=["A", "B", "C", "D", "E"])
+        >>> s
+        A    1.0
+        B    NaN
+        C    4.0
+        D    3.0
+        E    4.0
+        dtype: float64
+
+        >>> s.idxmax()
+        'C'
+        """
+        axis = self._get_axis_number(axis)
+        iloc = self.argmax(axis, skipna, *args, **kwargs)
+        return self.index[iloc]
+
+    def round(self, decimals: int = 0, *args, **kwargs) -> Series:
+        """
+        Round each value in a Series to the given number of decimals.
+
+        Parameters
+        ----------
+        decimals : int, default 0
+            Number of decimal places to round to. If decimals is negative,
+            it specifies the number of positions to the left of the decimal point.
+        *args, **kwargs
+            Additional arguments and keywords have no effect but might be
+            accepted for compatibility with NumPy.
+
+        Returns
+        -------
+        Series
+            Rounded values of the Series.
+
+        See Also
+        --------
+        numpy.around : Round values of an np.array.
+        DataFrame.round : Round values of a DataFrame.
+        Series.dt.round : Round values of data to the specified freq.
+
+        Notes
+        -----
+        For values exactly halfway between rounded decimal values, pandas rounds
+        to the nearest even value (e.g. -0.5 and 0.5 round to 0.0, 1.5 and 2.5
+        round to 2.0, etc.).
+
+        Examples
+        --------
+        >>> s = pd.Series([-0.5, 0.1, 2.5, 1.3, 2.7])
+        >>> s.round()
+        0   -0.0
+        1    0.0
+        2    2.0
+        3    1.0
+        4    3.0
+        dtype: float64
+        """
+        nv.validate_round(args, kwargs)
+        new_mgr = self._mgr.round(decimals=decimals)
+        return self._constructor_from_mgr(new_mgr, axes=new_mgr.axes).__finalize__(
+            self, method="round"
+        )
+
+    @overload
+    def quantile(
+        self, q: float = ..., interpolation: QuantileInterpolation = ...
+    ) -> float: ...
+
+    @overload
+    def quantile(
+        self,
+        q: Sequence[float] | AnyArrayLike,
+        interpolation: QuantileInterpolation = ...,
+    ) -> Series: ...
+
+    @overload
+    def quantile(
+        self,
+        q: float | Sequence[float] | AnyArrayLike = ...,
+        interpolation: QuantileInterpolation = ...,
+    ) -> float | Series: ...
+
+    def quantile(
+        self,
+        q: float | Sequence[float] | AnyArrayLike = 0.5,
+        interpolation: QuantileInterpolation = "linear",
+    ) -> float | Series:
+        """
+        Return value at the given quantile.
+
+        Parameters
+        ----------
+        q : float or array-like, default 0.5 (50% quantile)
+            The quantile(s) to compute, which can lie in range: 0 <= q <= 1.
+        interpolation : {'linear', 'lower', 'higher', 'midpoint', 'nearest'}
+            This optional parameter specifies the interpolation method to use,
+            when the desired quantile lies between two data points `i` and `j`:
+
+                * linear: `i + (j - i) * (x-i)/(j-i)`, where `(x-i)/(j-i)` is
+                  the fractional part of the index surrounded by `i > j`.
+                * lower: `i`.
+                * higher: `j`.
+                * nearest: `i` or `j` whichever is nearest.
+                * midpoint: (`i` + `j`) / 2.
+
+        Returns
+        -------
+        float or Series
+            If ``q`` is an array, a Series will be returned where the
+            index is ``q`` and the values are the quantiles, otherwise
+            a float will be returned.
+
+        See Also
+        --------
+        core.window.Rolling.quantile : Calculate the rolling quantile.
+        numpy.percentile : Returns the q-th percentile(s) of the array elements.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s.quantile(0.5)
+        2.5
+        >>> s.quantile([0.25, 0.5, 0.75])
+        0.25    1.75
+        0.50    2.50
+        0.75    3.25
+        dtype: float64
+        """
+        validate_percentile(q)
+
+        # We dispatch to DataFrame so that core.internals only has to worry
+        #  about 2D cases.
+        df = self.to_frame()
+
+        result = df.quantile(q=q, interpolation=interpolation, numeric_only=False)
+        if result.ndim == 2:
+            result = result.iloc[:, 0]
+
+        if is_list_like(q):
+            result.name = self.name
+            idx = Index(q, dtype=np.float64)
+            return self._constructor(result, index=idx, name=self.name)
+        else:
+            # scalar
+            return result.iloc[0]
+
+    def corr(
+        self,
+        other: Series,
+        method: CorrelationMethod = "pearson",
+        min_periods: int | None = None,
+    ) -> float:
+        """
+        Compute correlation with `other` Series, excluding missing values.
+
+        The two `Series` objects are not required to be the same length and will be
+        aligned internally before the correlation function is applied.
+
+        Parameters
+        ----------
+        other : Series
+            Series with which to compute the correlation.
+        method : {'pearson', 'kendall', 'spearman'} or callable
+            Method used to compute correlation:
+
+            - pearson : Standard correlation coefficient
+            - kendall : Kendall Tau correlation coefficient
+            - spearman : Spearman rank correlation
+            - callable: Callable with input two 1d ndarrays and returning a float.
+
+            .. warning::
+                Note that the returned matrix from corr will have 1 along the
+                diagonals and will be symmetric regardless of the callable's
+                behavior.
+        min_periods : int, optional
+            Minimum number of observations needed to have a valid result.
+
+        Returns
+        -------
+        float
+            Correlation with other.
+
+        See Also
+        --------
+        DataFrame.corr : Compute pairwise correlation between columns.
+        DataFrame.corrwith : Compute pairwise correlation with another
+            DataFrame or Series.
+
+        Notes
+        -----
+        Pearson, Kendall and Spearman correlation are currently computed using pairwise complete observations.
+
+        * `Pearson correlation coefficient <https://en.wikipedia.org/wiki/Pearson_correlation_coefficient>`_
+        * `Kendall rank correlation coefficient <https://en.wikipedia.org/wiki/Kendall_rank_correlation_coefficient>`_
+        * `Spearman's rank correlation coefficient <https://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient>`_
+
+        Automatic data alignment: as with all pandas operations, automatic data alignment is performed for this method.
+        ``corr()`` automatically considers values with matching indices.
+
+        Examples
+        --------
+        >>> def histogram_intersection(a, b):
+        ...     v = np.minimum(a, b).sum().round(decimals=1)
+        ...     return v
+        >>> s1 = pd.Series([0.2, 0.0, 0.6, 0.2])
+        >>> s2 = pd.Series([0.3, 0.6, 0.0, 0.1])
+        >>> s1.corr(s2, method=histogram_intersection)
+        0.3
+
+        Pandas auto-aligns the values with matching indices
+
+        >>> s1 = pd.Series([1, 2, 3], index=[0, 1, 2])
+        >>> s2 = pd.Series([1, 2, 3], index=[2, 1, 0])
+        >>> s1.corr(s2)
+        -1.0
+
+        If the input is a constant array, the correlation is not defined in this case,
+        and ``np.nan`` is returned.
+
+        >>> s1 = pd.Series([0.45, 0.45])
+        >>> s1.corr(s1)
+        nan
+        """  # noqa: E501
+        this, other = self.align(other, join="inner")
+        if len(this) == 0:
+            return np.nan
+
+        this_values = this.to_numpy(dtype=float, na_value=np.nan, copy=False)
+        other_values = other.to_numpy(dtype=float, na_value=np.nan, copy=False)
+
+        if method in ["pearson", "spearman", "kendall"] or callable(method):
+            return nanops.nancorr(
+                this_values, other_values, method=method, min_periods=min_periods
+            )
+
+        raise ValueError(
+            "method must be either 'pearson', "
+            "'spearman', 'kendall', or a callable, "
+            f"'{method}' was supplied"
+        )
+
+    def cov(
+        self,
+        other: Series,
+        min_periods: int | None = None,
+        ddof: int | None = 1,
+    ) -> float:
+        """
+        Compute covariance with Series, excluding missing values.
+
+        The two `Series` objects are not required to be the same length and
+        will be aligned internally before the covariance is calculated.
+
+        Parameters
+        ----------
+        other : Series
+            Series with which to compute the covariance.
+        min_periods : int, optional
+            Minimum number of observations needed to have a valid result.
+        ddof : int, default 1
+            Delta degrees of freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+
+        Returns
+        -------
+        float
+            Covariance between Series and other normalized by N-1
+            (unbiased estimator).
+
+        See Also
+        --------
+        DataFrame.cov : Compute pairwise covariance of columns.
+
+        Examples
+        --------
+        >>> s1 = pd.Series([0.90010907, 0.13484424, 0.62036035])
+        >>> s2 = pd.Series([0.12528585, 0.26962463, 0.51111198])
+        >>> s1.cov(s2)
+        -0.01685762652715874
+        """
+        this, other = self.align(other, join="inner")
+        if len(this) == 0:
+            return np.nan
+        this_values = this.to_numpy(dtype=float, na_value=np.nan, copy=False)
+        other_values = other.to_numpy(dtype=float, na_value=np.nan, copy=False)
+        return nanops.nancov(
+            this_values, other_values, min_periods=min_periods, ddof=ddof
+        )
+
+    @doc(
+        klass="Series",
+        extra_params="",
+        other_klass="DataFrame",
+        examples=dedent(
+            """
+        Difference with previous row
+
+        >>> s = pd.Series([1, 1, 2, 3, 5, 8])
+        >>> s.diff()
+        0    NaN
+        1    0.0
+        2    1.0
+        3    1.0
+        4    2.0
+        5    3.0
+        dtype: float64
+
+        Difference with 3rd previous row
+
+        >>> s.diff(periods=3)
+        0    NaN
+        1    NaN
+        2    NaN
+        3    2.0
+        4    4.0
+        5    6.0
+        dtype: float64
+
+        Difference with following row
+
+        >>> s.diff(periods=-1)
+        0    0.0
+        1   -1.0
+        2   -1.0
+        3   -2.0
+        4   -3.0
+        5    NaN
+        dtype: float64
+
+        Overflow in input dtype
+
+        >>> s = pd.Series([1, 0], dtype=np.uint8)
+        >>> s.diff()
+        0      NaN
+        1    255.0
+        dtype: float64"""
+        ),
+    )
+    def diff(self, periods: int = 1) -> Series:
+        """
+        First discrete difference of element.
+
+        Calculates the difference of a {klass} element compared with another
+        element in the {klass} (default is element in previous row).
+
+        Parameters
+        ----------
+        periods : int, default 1
+            Periods to shift for calculating difference, accepts negative
+            values.
+        {extra_params}
+        Returns
+        -------
+        {klass}
+            First differences of the Series.
+
+        See Also
+        --------
+        {klass}.pct_change: Percent change over given number of periods.
+        {klass}.shift: Shift index by desired number of periods with an
+            optional time freq.
+        {other_klass}.diff: First discrete difference of object.
+
+        Notes
+        -----
+        For boolean dtypes, this uses :meth:`operator.xor` rather than
+        :meth:`operator.sub`.
+        The result is calculated according to current dtype in {klass},
+        however dtype of the result is always float64.
+
+        Examples
+        --------
+        {examples}
+        """
+        if not lib.is_integer(periods):
+            if not (is_float(periods) and periods.is_integer()):
+                raise ValueError("periods must be an integer")
+        result = algorithms.diff(self._values, periods)
+        return self._constructor(result, index=self.index, copy=False).__finalize__(
+            self, method="diff"
+        )
+
+    def autocorr(self, lag: int = 1) -> float:
+        """
+        Compute the lag-N autocorrelation.
+
+        This method computes the Pearson correlation between
+        the Series and its shifted self.
+
+        Parameters
+        ----------
+        lag : int, default 1
+            Number of lags to apply before performing autocorrelation.
+
+        Returns
+        -------
+        float
+            The Pearson correlation between self and self.shift(lag).
+
+        See Also
+        --------
+        Series.corr : Compute the correlation between two Series.
+        Series.shift : Shift index by desired number of periods.
+        DataFrame.corr : Compute pairwise correlation of columns.
+        DataFrame.corrwith : Compute pairwise correlation between rows or
+            columns of two DataFrame objects.
+
+        Notes
+        -----
+        If the Pearson correlation is not well defined return 'NaN'.
+
+        Examples
+        --------
+        >>> s = pd.Series([0.25, 0.5, 0.2, -0.05])
+        >>> s.autocorr()  # doctest: +ELLIPSIS
+        0.10355...
+        >>> s.autocorr(lag=2)  # doctest: +ELLIPSIS
+        -0.99999...
+
+        If the Pearson correlation is not well defined, then 'NaN' is returned.
+
+        >>> s = pd.Series([1, 0, 0, 0])
+        >>> s.autocorr()
+        nan
+        """
+        return self.corr(cast(Series, self.shift(lag)))
+
+    def dot(self, other: AnyArrayLike | DataFrame) -> Series | np.ndarray:
+        """
+        Compute the dot product between the Series and the columns of other.
+
+        This method computes the dot product between the Series and another
+        one, or the Series and each columns of a DataFrame, or the Series and
+        each columns of an array.
+
+        It can also be called using `self @ other`.
+
+        Parameters
+        ----------
+        other : Series, DataFrame or array-like
+            The other object to compute the dot product with its columns.
+
+        Returns
+        -------
+        scalar, Series or numpy.ndarray
+            Return the dot product of the Series and other if other is a
+            Series, the Series of the dot product of Series and each rows of
+            other if other is a DataFrame or a numpy.ndarray between the Series
+            and each columns of the numpy array.
+
+        See Also
+        --------
+        DataFrame.dot: Compute the matrix product with the DataFrame.
+        Series.mul: Multiplication of series and other, element-wise.
+
+        Notes
+        -----
+        The Series and other has to share the same index if other is a Series
+        or a DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series([0, 1, 2, 3])
+        >>> other = pd.Series([-1, 2, -3, 4])
+        >>> s.dot(other)
+        8
+        >>> s @ other
+        8
+        >>> df = pd.DataFrame([[0, 1], [-2, 3], [4, -5], [6, 7]])
+        >>> s.dot(df)
+        0    24
+        1    14
+        dtype: int64
+        >>> arr = np.array([[0, 1], [-2, 3], [4, -5], [6, 7]])
+        >>> s.dot(arr)
+        array([24, 14])
+        """
+        if isinstance(other, (Series, ABCDataFrame)):
+            common = self.index.union(other.index)
+            if len(common) > len(self.index) or len(common) > len(other.index):
+                raise ValueError("matrices are not aligned")
+
+            left = self.reindex(index=common)
+            right = other.reindex(index=common)
+            lvals = left.values
+            rvals = right.values
+        else:
+            lvals = self.values
+            rvals = np.asarray(other)
+            if lvals.shape[0] != rvals.shape[0]:
+                raise Exception(
+                    f"Dot product shape mismatch, {lvals.shape} vs {rvals.shape}"
+                )
+
+        if isinstance(other, ABCDataFrame):
+            return self._constructor(
+                np.dot(lvals, rvals), index=other.columns, copy=False
+            ).__finalize__(self, method="dot")
+        elif isinstance(other, Series):
+            return np.dot(lvals, rvals)
+        elif isinstance(rvals, np.ndarray):
+            return np.dot(lvals, rvals)
+        else:  # pragma: no cover
+            raise TypeError(f"unsupported type: {type(other)}")
+
+    def __matmul__(self, other):
+        """
+        Matrix multiplication using binary `@` operator.
+        """
+        return self.dot(other)
+
+    def __rmatmul__(self, other):
+        """
+        Matrix multiplication using binary `@` operator.
+        """
+        return self.dot(np.transpose(other))
+
+    @doc(base.IndexOpsMixin.searchsorted, klass="Series")
+    # Signature of "searchsorted" incompatible with supertype "IndexOpsMixin"
+    def searchsorted(  # type: ignore[override]
+        self,
+        value: NumpyValueArrayLike | ExtensionArray,
+        side: Literal["left", "right"] = "left",
+        sorter: NumpySorter | None = None,
+    ) -> npt.NDArray[np.intp] | np.intp:
+        return base.IndexOpsMixin.searchsorted(self, value, side=side, sorter=sorter)
+
+    # -------------------------------------------------------------------
+    # Combination
+
+    def _append(
+        self, to_append, ignore_index: bool = False, verify_integrity: bool = False
+    ):
+        from pandas.core.reshape.concat import concat
+
+        if isinstance(to_append, (list, tuple)):
+            to_concat = [self]
+            to_concat.extend(to_append)
+        else:
+            to_concat = [self, to_append]
+        if any(isinstance(x, (ABCDataFrame,)) for x in to_concat[1:]):
+            msg = "to_append should be a Series or list/tuple of Series, got DataFrame"
+            raise TypeError(msg)
+        return concat(
+            to_concat, ignore_index=ignore_index, verify_integrity=verify_integrity
+        )
+
+    @doc(
+        _shared_docs["compare"],
+        dedent(
+            """
+        Returns
+        -------
+        Series or DataFrame
+            If axis is 0 or 'index' the result will be a Series.
+            The resulting index will be a MultiIndex with 'self' and 'other'
+            stacked alternately at the inner level.
+
+            If axis is 1 or 'columns' the result will be a DataFrame.
+            It will have two columns namely 'self' and 'other'.
+
+        See Also
+        --------
+        DataFrame.compare : Compare with another DataFrame and show differences.
+
+        Notes
+        -----
+        Matching NaNs will not appear as a difference.
+
+        Examples
+        --------
+        >>> s1 = pd.Series(["a", "b", "c", "d", "e"])
+        >>> s2 = pd.Series(["a", "a", "c", "b", "e"])
+
+        Align the differences on columns
+
+        >>> s1.compare(s2)
+          self other
+        1    b     a
+        3    d     b
+
+        Stack the differences on indices
+
+        >>> s1.compare(s2, align_axis=0)
+        1  self     b
+           other    a
+        3  self     d
+           other    b
+        dtype: object
+
+        Keep all original rows
+
+        >>> s1.compare(s2, keep_shape=True)
+          self other
+        0  NaN   NaN
+        1    b     a
+        2  NaN   NaN
+        3    d     b
+        4  NaN   NaN
+
+        Keep all original rows and also all original values
+
+        >>> s1.compare(s2, keep_shape=True, keep_equal=True)
+          self other
+        0    a     a
+        1    b     a
+        2    c     c
+        3    d     b
+        4    e     e
+        """
+        ),
+        klass=_shared_doc_kwargs["klass"],
+    )
+    def compare(
+        self,
+        other: Series,
+        align_axis: Axis = 1,
+        keep_shape: bool = False,
+        keep_equal: bool = False,
+        result_names: Suffixes = ("self", "other"),
+    ) -> DataFrame | Series:
+        return super().compare(
+            other=other,
+            align_axis=align_axis,
+            keep_shape=keep_shape,
+            keep_equal=keep_equal,
+            result_names=result_names,
+        )
+
+    def combine(
+        self,
+        other: Series | Hashable,
+        func: Callable[[Hashable, Hashable], Hashable],
+        fill_value: Hashable | None = None,
+    ) -> Series:
+        """
+        Combine the Series with a Series or scalar according to `func`.
+
+        Combine the Series and `other` using `func` to perform elementwise
+        selection for combined Series.
+        `fill_value` is assumed when value is missing at some index
+        from one of the two objects being combined.
+
+        Parameters
+        ----------
+        other : Series or scalar
+            The value(s) to be combined with the `Series`.
+        func : function
+            Function that takes two scalars as inputs and returns an element.
+        fill_value : scalar, optional
+            The value to assume when an index is missing from
+            one Series or the other. The default specifies to use the
+            appropriate NaN value for the underlying dtype of the Series.
+
+        Returns
+        -------
+        Series
+            The result of combining the Series with the other object.
+
+        See Also
+        --------
+        Series.combine_first : Combine Series values, choosing the calling
+            Series' values first.
+
+        Examples
+        --------
+        Consider 2 Datasets ``s1`` and ``s2`` containing
+        highest clocked speeds of different birds.
+
+        >>> s1 = pd.Series({"falcon": 330.0, "eagle": 160.0})
+        >>> s1
+        falcon    330.0
+        eagle     160.0
+        dtype: float64
+        >>> s2 = pd.Series({"falcon": 345.0, "eagle": 200.0, "duck": 30.0})
+        >>> s2
+        falcon    345.0
+        eagle     200.0
+        duck       30.0
+        dtype: float64
+
+        Now, to combine the two datasets and view the highest speeds
+        of the birds across the two datasets
+
+        >>> s1.combine(s2, max)
+        duck        NaN
+        eagle     200.0
+        falcon    345.0
+        dtype: float64
+
+        In the previous example, the resulting value for duck is missing,
+        because the maximum of a NaN and a float is a NaN.
+        So, in the example, we set ``fill_value=0``,
+        so the maximum value returned will be the value from some dataset.
+
+        >>> s1.combine(s2, max, fill_value=0)
+        duck       30.0
+        eagle     200.0
+        falcon    345.0
+        dtype: float64
+        """
+        if fill_value is None:
+            fill_value = na_value_for_dtype(self.dtype, compat=False)
+
+        if isinstance(other, Series):
+            # If other is a Series, result is based on union of Series,
+            # so do this element by element
+            new_index = self.index.union(other.index)
+            new_name = ops.get_op_result_name(self, other)
+            new_values = np.empty(len(new_index), dtype=object)
+            with np.errstate(all="ignore"):
+                for i, idx in enumerate(new_index):
+                    lv = self.get(idx, fill_value)
+                    rv = other.get(idx, fill_value)
+                    new_values[i] = func(lv, rv)
+        else:
+            # Assume that other is a scalar, so apply the function for
+            # each element in the Series
+            new_index = self.index
+            new_values = np.empty(len(new_index), dtype=object)
+            with np.errstate(all="ignore"):
+                new_values[:] = [func(lv, other) for lv in self._values]
+            new_name = self.name
+
+        # try_float=False is to match agg_series
+        npvalues = lib.maybe_convert_objects(new_values, try_float=False)
+        # same_dtype here is a kludge to avoid casting e.g. [True, False] to
+        #  ["True", "False"]
+        same_dtype = isinstance(self.dtype, (StringDtype, CategoricalDtype))
+        res_values = maybe_cast_pointwise_result(
+            npvalues, self.dtype, same_dtype=same_dtype
+        )
+        return self._constructor(res_values, index=new_index, name=new_name, copy=False)
+
+    def combine_first(self, other) -> Series:
+        """
+        Update null elements with value in the same location in 'other'.
+
+        Combine two Series objects by filling null values in one Series with
+        non-null values from the other Series. Result index will be the union
+        of the two indexes.
+
+        Parameters
+        ----------
+        other : Series
+            The value(s) to be used for filling null values.
+
+        Returns
+        -------
+        Series
+            The result of combining the provided Series with the other object.
+
+        See Also
+        --------
+        Series.combine : Perform element-wise operation on two Series
+            using a given function.
+
+        Examples
+        --------
+        >>> s1 = pd.Series([1, np.nan])
+        >>> s2 = pd.Series([3, 4, 5])
+        >>> s1.combine_first(s2)
+        0    1.0
+        1    4.0
+        2    5.0
+        dtype: float64
+
+        Null values still persist if the location of that null value
+        does not exist in `other`
+
+        >>> s1 = pd.Series({"falcon": np.nan, "eagle": 160.0})
+        >>> s2 = pd.Series({"eagle": 200.0, "duck": 30.0})
+        >>> s1.combine_first(s2)
+        duck       30.0
+        eagle     160.0
+        falcon      NaN
+        dtype: float64
+        """
+        from pandas.core.reshape.concat import concat
+
+        if self.dtype == other.dtype:
+            if self.index.equals(other.index):
+                return self.mask(self.isna(), other)
+            elif self._can_hold_na and not isinstance(self.dtype, SparseDtype):
+                this, other = self.align(other, join="outer")
+                return this.mask(this.isna(), other)
+
+        new_index = self.index.union(other.index)
+
+        this = self
+        # identify the index subset to keep for each series
+        keep_other = other.index.difference(this.index[notna(this)])
+        keep_this = this.index.difference(keep_other)
+
+        this = this.reindex(keep_this)
+        other = other.reindex(keep_other)
+
+        if this.dtype.kind == "M" and other.dtype.kind != "M":
+            # TODO: try to match resos?
+            other = to_datetime(other)
+        combined = concat([this, other])
+        combined = combined.reindex(new_index)
+        return combined.__finalize__(self, method="combine_first")
+
+    def update(self, other: Series | Sequence | Mapping) -> None:
+        """
+        Modify Series in place using values from passed Series.
+
+        Uses non-NA values from passed Series to make updates. Aligns
+        on index.
+
+        Parameters
+        ----------
+        other : Series, or object coercible into Series
+            Other Series that provides values to update the current Series.
+
+        See Also
+        --------
+        Series.combine : Perform element-wise operation on two Series
+            using a given function.
+        Series.transform: Modify a Series using a function.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.update(pd.Series([4, 5, 6]))
+        >>> s
+        0    4
+        1    5
+        2    6
+        dtype: int64
+
+        >>> s = pd.Series(["a", "b", "c"])
+        >>> s.update(pd.Series(["d", "e"], index=[0, 2]))
+        >>> s
+        0    d
+        1    b
+        2    e
+        dtype: object
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.update(pd.Series([4, 5, 6, 7, 8]))
+        >>> s
+        0    4
+        1    5
+        2    6
+        dtype: int64
+
+        If ``other`` contains NaNs the corresponding values are not updated
+        in the original Series.
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.update(pd.Series([4, np.nan, 6]))
+        >>> s
+        0    4
+        1    2
+        2    6
+        dtype: int64
+
+        ``other`` can also be a non-Series object type
+        that is coercible into a Series
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.update([4, np.nan, 6])
+        >>> s
+        0    4
+        1    2
+        2    6
+        dtype: int64
+
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.update({1: 9})
+        >>> s
+        0    1
+        1    9
+        2    3
+        dtype: int64
+        """
+        if not PYPY:
+            if sys.getrefcount(self) <= REF_COUNT:
+                warnings.warn(
+                    _chained_assignment_method_msg,
+                    ChainedAssignmentError,
+                    stacklevel=2,
+                )
+
+        if not isinstance(other, Series):
+            other = Series(other)
+
+        other = other.reindex_like(self)
+        mask = notna(other)
+
+        self._mgr = self._mgr.putmask(mask=mask, new=other)
+
+    # ----------------------------------------------------------------------
+    # Reindexing, sorting
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> Series: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> None: ...
+
+    @overload
+    def sort_values(
+        self,
+        *,
+        axis: Axis = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: bool = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        ignore_index: bool = ...,
+        key: ValueKeyFunc = ...,
+    ) -> Series | None: ...
+
+    def sort_values(
+        self,
+        *,
+        axis: Axis = 0,
+        ascending: bool | Sequence[bool] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: NaPosition = "last",
+        ignore_index: bool = False,
+        key: ValueKeyFunc | None = None,
+    ) -> Series | None:
+        """
+        Sort by the values.
+
+        Sort a Series in ascending or descending order by some
+        criterion.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        ascending : bool or list of bools, default True
+            If True, sort values in ascending order, otherwise descending.
+        inplace : bool, default False
+            If True, perform operation in-place.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+            Choice of sorting algorithm. See also :func:`numpy.sort` for more
+            information. 'mergesort' and 'stable' are the only stable  algorithms.
+        na_position : {'first' or 'last'}, default 'last'
+            Argument 'first' puts NaNs at the beginning, 'last' puts NaNs at
+            the end.
+        ignore_index : bool, default False
+            If True, the resulting axis will be labeled 0, 1, …, n - 1.
+        key : callable, optional
+            If not None, apply the key function to the series values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect a
+            ``Series`` and return an array-like.
+
+        Returns
+        -------
+        Series or None
+            Series ordered by values or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.sort_index : Sort by the Series indices.
+        DataFrame.sort_values : Sort DataFrame by the values along either axis.
+        DataFrame.sort_index : Sort DataFrame by indices.
+
+        Examples
+        --------
+        >>> s = pd.Series([np.nan, 1, 3, 10, 5])
+        >>> s
+        0     NaN
+        1     1.0
+        2     3.0
+        3     10.0
+        4     5.0
+        dtype: float64
+
+        Sort values ascending order (default behavior)
+
+        >>> s.sort_values(ascending=True)
+        1     1.0
+        2     3.0
+        4     5.0
+        3    10.0
+        0     NaN
+        dtype: float64
+
+        Sort values descending order
+
+        >>> s.sort_values(ascending=False)
+        3    10.0
+        4     5.0
+        2     3.0
+        1     1.0
+        0     NaN
+        dtype: float64
+
+        Sort values putting NAs first
+
+        >>> s.sort_values(na_position="first")
+        0     NaN
+        1     1.0
+        2     3.0
+        4     5.0
+        3    10.0
+        dtype: float64
+
+        Sort a series of strings
+
+        >>> s = pd.Series(["z", "b", "d", "a", "c"])
+        >>> s
+        0    z
+        1    b
+        2    d
+        3    a
+        4    c
+        dtype: object
+
+        >>> s.sort_values()
+        3    a
+        1    b
+        4    c
+        2    d
+        0    z
+        dtype: object
+
+        Sort using a key function. Your `key` function will be
+        given the ``Series`` of values and should return an array-like.
+
+        >>> s = pd.Series(["a", "B", "c", "D", "e"])
+        >>> s.sort_values()
+        1    B
+        3    D
+        0    a
+        2    c
+        4    e
+        dtype: object
+        >>> s.sort_values(key=lambda x: x.str.lower())
+        0    a
+        1    B
+        2    c
+        3    D
+        4    e
+        dtype: object
+
+        NumPy ufuncs work well here. For example, we can
+        sort by the ``sin`` of the value
+
+        >>> s = pd.Series([-4, -2, 0, 2, 4])
+        >>> s.sort_values(key=np.sin)
+        1   -2
+        4    4
+        2    0
+        0   -4
+        3    2
+        dtype: int64
+
+        More complicated user-defined functions can be used,
+        as long as they expect a Series and return an array-like
+
+        >>> s.sort_values(key=lambda x: (np.tan(x.cumsum())))
+        0   -4
+        3    2
+        4    4
+        1   -2
+        2    0
+        dtype: int64
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        # Validate the axis parameter
+        self._get_axis_number(axis)
+
+        if is_list_like(ascending):
+            ascending = cast(Sequence[bool], ascending)
+            if len(ascending) != 1:
+                raise ValueError(
+                    f"Length of ascending ({len(ascending)}) must be 1 for Series"
+                )
+            ascending = ascending[0]
+
+        ascending = validate_ascending(ascending)
+
+        if na_position not in ["first", "last"]:
+            raise ValueError(f"invalid na_position: {na_position}")
+
+        # GH 35922. Make sorting stable by leveraging nargsort
+        if key:
+            values_to_sort = cast(Series, ensure_key_mapped(self, key))._values
+        else:
+            values_to_sort = self._values
+        sorted_index = nargsort(values_to_sort, kind, bool(ascending), na_position)
+
+        if is_range_indexer(sorted_index, len(sorted_index)):
+            if inplace:
+                return self._update_inplace(self)
+            return self.copy(deep=False)
+
+        result = self._constructor(
+            self._values[sorted_index], index=self.index[sorted_index], copy=False
+        )
+
+        if ignore_index:
+            result.index = default_index(len(sorted_index))
+
+        if not inplace:
+            return result.__finalize__(self, method="sort_values")
+        self._update_inplace(result)
+        return None
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[True],
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> None: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: Literal[False] = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> Series: ...
+
+    @overload
+    def sort_index(
+        self,
+        *,
+        axis: Axis = ...,
+        level: IndexLabel = ...,
+        ascending: bool | Sequence[bool] = ...,
+        inplace: bool = ...,
+        kind: SortKind = ...,
+        na_position: NaPosition = ...,
+        sort_remaining: bool = ...,
+        ignore_index: bool = ...,
+        key: IndexKeyFunc = ...,
+    ) -> Series | None: ...
+
+    def sort_index(
+        self,
+        *,
+        axis: Axis = 0,
+        level: IndexLabel | None = None,
+        ascending: bool | Sequence[bool] = True,
+        inplace: bool = False,
+        kind: SortKind = "quicksort",
+        na_position: NaPosition = "last",
+        sort_remaining: bool = True,
+        ignore_index: bool = False,
+        key: IndexKeyFunc | None = None,
+    ) -> Series | None:
+        """
+        Sort Series by index labels.
+
+        Returns a new Series sorted by label if `inplace` argument is
+        ``False``, otherwise updates the original series and returns None.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        level : int, optional
+            If not None, sort on values in specified index level(s).
+        ascending : bool or list-like of bools, default True
+            Sort ascending vs. descending. When the index is a MultiIndex the
+            sort direction can be controlled for each level individually.
+        inplace : bool, default False
+            If True, perform operation in-place.
+        kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+            Choice of sorting algorithm. See also :func:`numpy.sort` for more
+            information. 'mergesort' and 'stable' are the only stable algorithms. For
+            DataFrames, this option is only applied when sorting on a single
+            column or label.
+        na_position : {'first', 'last'}, default 'last'
+            If 'first' puts NaNs at the beginning, 'last' puts NaNs at the end.
+            Not implemented for MultiIndex.
+        sort_remaining : bool, default True
+            If True and sorting by level and index is multilevel, sort by other
+            levels too (in order) after sorting by specified level.
+        ignore_index : bool, default False
+            If True, the resulting axis will be labeled 0, 1, …, n - 1.
+        key : callable, optional
+            If not None, apply the key function to the index values
+            before sorting. This is similar to the `key` argument in the
+            builtin :meth:`sorted` function, with the notable difference that
+            this `key` function should be *vectorized*. It should expect an
+            ``Index`` and return an ``Index`` of the same shape.
+
+        Returns
+        -------
+        Series or None
+            The original Series sorted by the labels or None if ``inplace=True``.
+
+        See Also
+        --------
+        DataFrame.sort_index: Sort DataFrame by the index.
+        DataFrame.sort_values: Sort DataFrame by the value.
+        Series.sort_values : Sort Series by the value.
+
+        Examples
+        --------
+        >>> s = pd.Series(["a", "b", "c", "d"], index=[3, 2, 1, 4])
+        >>> s.sort_index()
+        1    c
+        2    b
+        3    a
+        4    d
+        dtype: object
+
+        Sort Descending
+
+        >>> s.sort_index(ascending=False)
+        4    d
+        3    a
+        2    b
+        1    c
+        dtype: object
+
+        By default NaNs are put at the end, but use `na_position` to place
+        them at the beginning
+
+        >>> s = pd.Series(["a", "b", "c", "d"], index=[3, 2, 1, np.nan])
+        >>> s.sort_index(na_position="first")
+        NaN     d
+         1.0    c
+         2.0    b
+         3.0    a
+        dtype: object
+
+        Specify index level to sort
+
+        >>> arrays = [
+        ...     np.array(["qux", "qux", "foo", "foo", "baz", "baz", "bar", "bar"]),
+        ...     np.array(["two", "one", "two", "one", "two", "one", "two", "one"]),
+        ... ]
+        >>> s = pd.Series([1, 2, 3, 4, 5, 6, 7, 8], index=arrays)
+        >>> s.sort_index(level=1)
+        bar  one    8
+        baz  one    6
+        foo  one    4
+        qux  one    2
+        bar  two    7
+        baz  two    5
+        foo  two    3
+        qux  two    1
+        dtype: int64
+
+        Does not sort by remaining levels when sorting by levels
+
+        >>> s.sort_index(level=1, sort_remaining=False)
+        qux  one    2
+        foo  one    4
+        baz  one    6
+        bar  one    8
+        qux  two    1
+        foo  two    3
+        baz  two    5
+        bar  two    7
+        dtype: int64
+
+        Apply a key function before sorting
+
+        >>> s = pd.Series([1, 2, 3, 4], index=["A", "b", "C", "d"])
+        >>> s.sort_index(key=lambda x: x.str.lower())
+        A    1
+        b    2
+        C    3
+        d    4
+        dtype: int64
+        """
+
+        return super().sort_index(
+            axis=axis,
+            level=level,
+            ascending=ascending,
+            inplace=inplace,
+            kind=kind,
+            na_position=na_position,
+            sort_remaining=sort_remaining,
+            ignore_index=ignore_index,
+            key=key,
+        )
+
+    def argsort(
+        self,
+        axis: Axis = 0,
+        kind: SortKind = "quicksort",
+        order: None = None,
+        stable: None = None,
+    ) -> Series:
+        """
+        Return the integer indices that would sort the Series values.
+
+        Override ndarray.argsort. Argsorts the value, omitting NA/null values,
+        and places the result in the same locations as the non-NA values.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        kind : {'mergesort', 'quicksort', 'heapsort', 'stable'}, default 'quicksort'
+            Choice of sorting algorithm. See :func:`numpy.sort` for more
+            information. 'mergesort' and 'stable' are the only stable algorithms.
+        order : None
+            Has no effect but is accepted for compatibility with numpy.
+        stable : None
+            Has no effect but is accepted for compatibility with numpy.
+
+        Returns
+        -------
+        Series[np.intp]
+            Positions of values within the sort order with -1 indicating
+            nan values.
+
+        See Also
+        --------
+        numpy.ndarray.argsort : Returns the indices that would sort this array.
+
+        Examples
+        --------
+        >>> s = pd.Series([3, 2, 1])
+        >>> s.argsort()
+        0    2
+        1    1
+        2    0
+        dtype: int64
+        """
+        if axis != -1:
+            # GH#54257 We allow -1 here so that np.argsort(series) works
+            self._get_axis_number(axis)
+
+        result = self.array.argsort(kind=kind)
+
+        res = self._constructor(
+            result, index=self.index, name=self.name, dtype=np.intp, copy=False
+        )
+        return res.__finalize__(self, method="argsort")
+
+    def nlargest(
+        self, n: int = 5, keep: Literal["first", "last", "all"] = "first"
+    ) -> Series:
+        """
+        Return the largest `n` elements.
+
+        Parameters
+        ----------
+        n : int, default 5
+            Return this many descending sorted values.
+        keep : {'first', 'last', 'all'}, default 'first'
+            When there are duplicate values that cannot all fit in a
+            Series of `n` elements:
+
+            - ``first`` : return the first `n` occurrences in order
+              of appearance.
+            - ``last`` : return the last `n` occurrences in reverse
+              order of appearance.
+            - ``all`` : keep all occurrences. This can result in a Series of
+              size larger than `n`.
+
+        Returns
+        -------
+        Series
+            The `n` largest values in the Series, sorted in decreasing order.
+
+        See Also
+        --------
+        Series.nsmallest: Get the `n` smallest elements.
+        Series.sort_values: Sort Series by values.
+        Series.head: Return the first `n` rows.
+
+        Notes
+        -----
+        Faster than ``.sort_values(ascending=False).head(n)`` for small `n`
+        relative to the size of the ``Series`` object.
+
+        Examples
+        --------
+        >>> countries_population = {
+        ...     "Italy": 59000000,
+        ...     "France": 65000000,
+        ...     "Malta": 434000,
+        ...     "Maldives": 434000,
+        ...     "Brunei": 434000,
+        ...     "Iceland": 337000,
+        ...     "Nauru": 11300,
+        ...     "Tuvalu": 11300,
+        ...     "Anguilla": 11300,
+        ...     "Montserrat": 5200,
+        ... }
+        >>> s = pd.Series(countries_population)
+        >>> s
+        Italy       59000000
+        France      65000000
+        Malta         434000
+        Maldives      434000
+        Brunei        434000
+        Iceland       337000
+        Nauru          11300
+        Tuvalu         11300
+        Anguilla       11300
+        Montserrat      5200
+        dtype: int64
+
+        The `n` largest elements where ``n=5`` by default.
+
+        >>> s.nlargest()
+        France      65000000
+        Italy       59000000
+        Malta         434000
+        Maldives      434000
+        Brunei        434000
+        dtype: int64
+
+        The `n` largest elements where ``n=3``. Default `keep` value is 'first'
+        so Malta will be kept.
+
+        >>> s.nlargest(3)
+        France    65000000
+        Italy     59000000
+        Malta       434000
+        dtype: int64
+
+        The `n` largest elements where ``n=3`` and keeping the last duplicates.
+        Brunei will be kept since it is the last with value 434000 based on
+        the index order.
+
+        >>> s.nlargest(3, keep="last")
+        France      65000000
+        Italy       59000000
+        Brunei        434000
+        dtype: int64
+
+        The `n` largest elements where ``n=3`` with all duplicates kept. Note
+        that the returned Series has five elements due to the three duplicates.
+
+        >>> s.nlargest(3, keep="all")
+        France      65000000
+        Italy       59000000
+        Malta         434000
+        Maldives      434000
+        Brunei        434000
+        dtype: int64
+        """
+        return selectn.SelectNSeries(self, n=n, keep=keep).nlargest()
+
+    def nsmallest(
+        self, n: int = 5, keep: Literal["first", "last", "all"] = "first"
+    ) -> Series:
+        """
+        Return the smallest `n` elements.
+
+        Parameters
+        ----------
+        n : int, default 5
+            Return this many ascending sorted values.
+        keep : {'first', 'last', 'all'}, default 'first'
+            When there are duplicate values that cannot all fit in a
+            Series of `n` elements:
+
+            - ``first`` : return the first `n` occurrences in order
+              of appearance.
+            - ``last`` : return the last `n` occurrences in reverse
+              order of appearance.
+            - ``all`` : keep all occurrences. This can result in a Series of
+              size larger than `n`.
+
+        Returns
+        -------
+        Series
+            The `n` smallest values in the Series, sorted in increasing order.
+
+        See Also
+        --------
+        Series.nlargest: Get the `n` largest elements.
+        Series.sort_values: Sort Series by values.
+        Series.head: Return the first `n` rows.
+
+        Notes
+        -----
+        Faster than ``.sort_values().head(n)`` for small `n` relative to
+        the size of the ``Series`` object.
+
+        Examples
+        --------
+        >>> countries_population = {
+        ...     "Italy": 59000000,
+        ...     "France": 65000000,
+        ...     "Brunei": 434000,
+        ...     "Malta": 434000,
+        ...     "Maldives": 434000,
+        ...     "Iceland": 337000,
+        ...     "Nauru": 11300,
+        ...     "Tuvalu": 11300,
+        ...     "Anguilla": 11300,
+        ...     "Montserrat": 5200,
+        ... }
+        >>> s = pd.Series(countries_population)
+        >>> s
+        Italy       59000000
+        France      65000000
+        Brunei        434000
+        Malta         434000
+        Maldives      434000
+        Iceland       337000
+        Nauru          11300
+        Tuvalu         11300
+        Anguilla       11300
+        Montserrat      5200
+        dtype: int64
+
+        The `n` smallest elements where ``n=5`` by default.
+
+        >>> s.nsmallest()
+        Montserrat    5200
+        Nauru        11300
+        Tuvalu       11300
+        Anguilla     11300
+        Iceland     337000
+        dtype: int64
+
+        The `n` smallest elements where ``n=3``. Default `keep` value is
+        'first' so Nauru and Tuvalu will be kept.
+
+        >>> s.nsmallest(3)
+        Montserrat   5200
+        Nauru       11300
+        Tuvalu      11300
+        dtype: int64
+
+        The `n` smallest elements where ``n=3`` and keeping the last
+        duplicates. Anguilla and Tuvalu will be kept since they are the last
+        with value 11300 based on the index order.
+
+        >>> s.nsmallest(3, keep="last")
+        Montserrat   5200
+        Anguilla    11300
+        Tuvalu      11300
+        dtype: int64
+
+        The `n` smallest elements where ``n=3`` with all duplicates kept. Note
+        that the returned Series has four elements due to the three duplicates.
+
+        >>> s.nsmallest(3, keep="all")
+        Montserrat   5200
+        Nauru       11300
+        Tuvalu      11300
+        Anguilla    11300
+        dtype: int64
+        """
+        return selectn.SelectNSeries(self, n=n, keep=keep).nsmallest()
+
+    def swaplevel(
+        self, i: Level = -2, j: Level = -1, copy: bool | lib.NoDefault = lib.no_default
+    ) -> Series:
+        """
+        Swap levels i and j in a :class:`MultiIndex`.
+
+        Default is to swap the two innermost levels of the index.
+
+        Parameters
+        ----------
+        i, j : int or str
+            Levels of the indices to be swapped. Can pass level name as string.
+        copy : bool, default True
+                    Whether to copy underlying data.
+
+                    .. note::
+                        The `copy` keyword will change behavior in pandas 3.0.
+                        `Copy-on-Write
+                        <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                        will be enabled by default, which means that all methods with a
+                        `copy` keyword will use a lazy copy mechanism to defer the copy
+                        and ignore the `copy` keyword. The `copy` keyword will be
+                        removed in a future version of pandas.
+
+                        You can already get the future behavior and improvements through
+                        enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+        Returns
+        -------
+        Series
+            Series with levels swapped in MultiIndex.
+
+        See Also
+        --------
+        DataFrame.swaplevel : Swap levels i and j in a :class:`DataFrame`.
+        Series.reorder_levels : Rearrange index levels using input order.
+        MultiIndex.swaplevel : Swap levels i and j in a :class:`MultiIndex`.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     ["A", "B", "A", "C"],
+        ...     index=[
+        ...         ["Final exam", "Final exam", "Coursework", "Coursework"],
+        ...         ["History", "Geography", "History", "Geography"],
+        ...         ["January", "February", "March", "April"],
+        ...     ],
+        ... )
+        >>> s
+        Final exam  History     January      A
+                    Geography   February     B
+        Coursework  History     March        A
+                    Geography   April        C
+        dtype: object
+
+        In the following example, we will swap the levels of the indices.
+        Here, we will swap the levels column-wise, but levels can be swapped row-wise
+        in a similar manner. Note that column-wise is the default behavior.
+        By not supplying any arguments for i and j, we swap the last and second to
+        last indices.
+
+        >>> s.swaplevel()
+        Final exam  January     History         A
+                    February    Geography       B
+        Coursework  March       History         A
+                    April       Geography       C
+        dtype: object
+
+        By supplying one argument, we can choose which index to swap the last
+        index with. We can for example swap the first index with the last one as
+        follows.
+
+        >>> s.swaplevel(0)
+        January     History     Final exam      A
+        February    Geography   Final exam      B
+        March       History     Coursework      A
+        April       Geography   Coursework      C
+        dtype: object
+
+        We can also define explicitly which indices we want to swap by supplying values
+        for both i and j. Here, we for example swap the first and second indices.
+
+        >>> s.swaplevel(0, 1)
+        History     Final exam  January         A
+        Geography   Final exam  February        B
+        History     Coursework  March           A
+        Geography   Coursework  April           C
+        dtype: object
+        """
+        self._check_copy_deprecation(copy)
+        assert isinstance(self.index, MultiIndex)
+        result = self.copy(deep=False)
+        result.index = self.index.swaplevel(i, j)
+        return result
+
+    def reorder_levels(self, order: Sequence[Level]) -> Series:
+        """
+        Rearrange index levels using input order.
+
+        May not drop or duplicate levels.
+
+        Parameters
+        ----------
+        order : list of int representing new level order
+            Reference level by number or key.
+
+        Returns
+        -------
+        Series
+            Type of caller with index as MultiIndex (new object).
+
+        See Also
+        --------
+        DataFrame.reorder_levels : Rearrange index or column levels using
+            input ``order``.
+
+        Examples
+        --------
+        >>> arrays = [
+        ...     np.array(["dog", "dog", "cat", "cat", "bird", "bird"]),
+        ...     np.array(["white", "black", "white", "black", "white", "black"]),
+        ... ]
+        >>> s = pd.Series([1, 2, 3, 3, 5, 2], index=arrays)
+        >>> s
+        dog   white    1
+              black    2
+        cat   white    3
+              black    3
+        bird  white    5
+              black    2
+        dtype: int64
+        >>> s.reorder_levels([1, 0])
+        white  dog     1
+        black  dog     2
+        white  cat     3
+        black  cat     3
+        white  bird    5
+        black  bird    2
+        dtype: int64
+        """
+        if not isinstance(self.index, MultiIndex):  # pragma: no cover
+            raise Exception("Can only reorder levels on a hierarchical axis.")
+
+        result = self.copy(deep=False)
+        assert isinstance(result.index, MultiIndex)
+        result.index = result.index.reorder_levels(order)
+        return result
+
+    def explode(self, ignore_index: bool = False) -> Series:
+        """
+        Transform each element of a list-like to a row.
+
+        Parameters
+        ----------
+        ignore_index : bool, default False
+            If True, the resulting index will be labeled 0, 1, …, n - 1.
+
+        Returns
+        -------
+        Series
+            Exploded lists to rows; index will be duplicated for these rows.
+
+        See Also
+        --------
+        Series.str.split : Split string values on specified separator.
+        Series.unstack : Unstack, a.k.a. pivot, Series with MultiIndex
+            to produce DataFrame.
+        DataFrame.melt : Unpivot a DataFrame from wide format to long format.
+        DataFrame.explode : Explode a DataFrame from list-like
+            columns to long format.
+
+        Notes
+        -----
+        This routine will explode list-likes including lists, tuples, sets,
+        Series, and np.ndarray. The result dtype of the subset rows will
+        be object. Scalars will be returned unchanged, and empty list-likes will
+        result in a np.nan for that row. In addition, the ordering of elements in
+        the output will be non-deterministic when exploding sets.
+
+        Reference :ref:`the user guide <reshaping.explode>` for more examples.
+
+        Examples
+        --------
+        >>> s = pd.Series([[1, 2, 3], "foo", [], [3, 4]])
+        >>> s
+        0    [1, 2, 3]
+        1          foo
+        2           []
+        3       [3, 4]
+        dtype: object
+
+        >>> s.explode()
+        0      1
+        0      2
+        0      3
+        1    foo
+        2    NaN
+        3      3
+        3      4
+        dtype: object
+        """
+        if isinstance(self.dtype, ExtensionDtype):
+            values, counts = self._values._explode()
+        elif len(self) and is_object_dtype(self.dtype):
+            values, counts = reshape.explode(np.asarray(self._values))
+        else:
+            result = self.copy()
+            return result.reset_index(drop=True) if ignore_index else result
+
+        if ignore_index:
+            index: Index = default_index(len(values))
+        else:
+            index = self.index.repeat(counts)
+
+        return self._constructor(values, index=index, name=self.name, copy=False)
+
+    def unstack(
+        self,
+        level: IndexLabel = -1,
+        fill_value: Hashable | None = None,
+        sort: bool = True,
+    ) -> DataFrame:
+        """
+        Unstack, also known as pivot, Series with MultiIndex to produce DataFrame.
+
+        Parameters
+        ----------
+        level : int, str, or list of these, default last level
+            Level(s) to unstack, can pass level name.
+        fill_value : scalar value, default None
+            Value to use when replacing NaN values.
+        sort : bool, default True
+            Sort the level(s) in the resulting MultiIndex columns.
+
+        Returns
+        -------
+        DataFrame
+            Unstacked Series.
+
+        See Also
+        --------
+        DataFrame.unstack : Pivot the MultiIndex of a DataFrame.
+
+        Notes
+        -----
+        Reference :ref:`the user guide <reshaping.stacking>` for more examples.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     [1, 2, 3, 4],
+        ...     index=pd.MultiIndex.from_product([["one", "two"], ["a", "b"]]),
+        ... )
+        >>> s
+        one  a    1
+             b    2
+        two  a    3
+             b    4
+        dtype: int64
+
+        >>> s.unstack(level=-1)
+             a  b
+        one  1  2
+        two  3  4
+
+        >>> s.unstack(level=0)
+           one  two
+        a    1    3
+        b    2    4
+        """
+        from pandas.core.reshape.reshape import unstack
+
+        return unstack(self, level, fill_value, sort)
+
+    # ----------------------------------------------------------------------
+    # function application
+
+    def map(
+        self,
+        arg: Callable | Mapping | Series,
+        na_action: Literal["ignore"] | None = None,
+        **kwargs,
+    ) -> Series:
+        """
+        Map values of Series according to an input mapping or function.
+
+        Used for substituting each value in a Series with another value,
+        that may be derived from a function, a ``dict`` or
+        a :class:`Series`.
+
+        Parameters
+        ----------
+        arg : function, collections.abc.Mapping subclass or Series
+            Mapping correspondence.
+        na_action : {None, 'ignore'}, default None
+            If 'ignore', propagate NaN values, without passing them to the
+            mapping correspondence.
+        **kwargs
+            Additional keyword arguments to pass as keywords arguments to
+            `arg`.
+
+            .. versionadded:: 3.0.0
+
+        Returns
+        -------
+        Series
+            Same index as caller.
+
+        See Also
+        --------
+        Series.apply : For applying more complex functions on a Series.
+        Series.replace: Replace values given in `to_replace` with `value`.
+        DataFrame.apply : Apply a function row-/column-wise.
+        DataFrame.map : Apply a function elementwise on a whole DataFrame.
+
+        Notes
+        -----
+        When ``arg`` is a dictionary, values in Series that are not in the
+        dictionary (as keys) are converted to ``NaN``. However, if the
+        dictionary is a ``dict`` subclass that defines ``__missing__`` (i.e.
+        provides a method for default values), then this default is used
+        rather than ``NaN``.
+
+        Examples
+        --------
+        >>> s = pd.Series(["cat", "dog", np.nan, "rabbit"])
+        >>> s
+        0      cat
+        1      dog
+        2      NaN
+        3   rabbit
+        dtype: object
+
+        ``map`` accepts a ``dict`` or a ``Series``. Values that are not found
+        in the ``dict`` are converted to ``NaN``, unless the dict has a default
+        value (e.g. ``defaultdict``):
+
+        >>> s.map({"cat": "kitten", "dog": "puppy"})
+        0   kitten
+        1    puppy
+        2      NaN
+        3      NaN
+        dtype: object
+
+        It also accepts a function:
+
+        >>> s.map("I am a {}".format)
+        0       I am a cat
+        1       I am a dog
+        2       I am a nan
+        3    I am a rabbit
+        dtype: object
+
+        To avoid applying the function to missing values (and keep them as
+        ``NaN``) ``na_action='ignore'`` can be used:
+
+        >>> s.map("I am a {}".format, na_action="ignore")
+        0     I am a cat
+        1     I am a dog
+        2            NaN
+        3  I am a rabbit
+        dtype: object
+        """
+        if callable(arg):
+            arg = functools.partial(arg, **kwargs)
+        new_values = self._map_values(arg, na_action=na_action)
+        return self._constructor(new_values, index=self.index, copy=False).__finalize__(
+            self, method="map"
+        )
+
+    def _gotitem(self, key, ndim, subset=None) -> Self:
+        """
+        Sub-classes to define. Return a sliced object.
+
+        Parameters
+        ----------
+        key : string / list of selections
+        ndim : {1, 2}
+            Requested ndim of result.
+        subset : object, default None
+            Subset to act on.
+        """
+        return self
+
+    _agg_see_also_doc = dedent(
+        """
+    See Also
+    --------
+    Series.apply : Invoke function on a Series.
+    Series.transform : Transform function producing a Series with like indexes.
+    """
+    )
+
+    _agg_examples_doc = dedent(
+        """
+    Examples
+    --------
+    >>> s = pd.Series([1, 2, 3, 4])
+    >>> s
+    0    1
+    1    2
+    2    3
+    3    4
+    dtype: int64
+
+    >>> s.agg('min')
+    1
+
+    >>> s.agg(['min', 'max'])
+    min   1
+    max   4
+    dtype: int64
+    """
+    )
+
+    @doc(
+        _shared_docs["aggregate"],
+        klass=_shared_doc_kwargs["klass"],
+        axis=_shared_doc_kwargs["axis"],
+        see_also=_agg_see_also_doc,
+        examples=_agg_examples_doc,
+    )
+    def aggregate(self, func=None, axis: Axis = 0, *args, **kwargs):
+        # Validate the axis parameter
+        self._get_axis_number(axis)
+
+        # if func is None, will switch to user-provided "named aggregation" kwargs
+        if func is None:
+            func = dict(kwargs.items())
+
+        op = SeriesApply(self, func, args=args, kwargs=kwargs)
+        result = op.agg()
+        return result
+
+    agg = aggregate
+
+    @doc(
+        _shared_docs["transform"],
+        klass=_shared_doc_kwargs["klass"],
+        axis=_shared_doc_kwargs["axis"],
+    )
+    def transform(
+        self, func: AggFuncType, axis: Axis = 0, *args, **kwargs
+    ) -> DataFrame | Series:
+        # Validate axis argument
+        self._get_axis_number(axis)
+        ser = self.copy(deep=False)
+        result = SeriesApply(ser, func=func, args=args, kwargs=kwargs).transform()
+        return result
+
+    def apply(
+        self,
+        func: AggFuncType,
+        args: tuple[Any, ...] = (),
+        *,
+        by_row: Literal[False, "compat"] = "compat",
+        **kwargs,
+    ) -> DataFrame | Series:
+        """
+        Invoke function on values of Series.
+
+        Can be ufunc (a NumPy function that applies to the entire Series)
+        or a Python function that only works on single values.
+
+        Parameters
+        ----------
+        func : function
+            Python function or NumPy ufunc to apply.
+        args : tuple
+            Positional arguments passed to func after the series value.
+        by_row : False or "compat", default "compat"
+            If ``"compat"`` and func is a callable, func will be passed each element of
+            the Series, like ``Series.map``. If func is a list or dict of
+            callables, will first try to translate each func into pandas methods. If
+            that doesn't work, will try call to apply again with ``by_row="compat"``
+            and if that fails, will call apply again with ``by_row=False``
+            (backward compatible).
+            If False, the func will be passed the whole Series at once.
+
+            ``by_row`` has no effect when ``func`` is a string.
+
+            .. versionadded:: 2.1.0
+        **kwargs
+            Additional keyword arguments passed to func.
+
+        Returns
+        -------
+        Series or DataFrame
+            If func returns a Series object the result will be a DataFrame.
+
+        See Also
+        --------
+        Series.map: For element-wise operations.
+        Series.agg: Only perform aggregating type operations.
+        Series.transform: Only perform transforming type operations.
+
+        Notes
+        -----
+        Functions that mutate the passed object can produce unexpected
+        behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+        for more details.
+
+        Examples
+        --------
+        Create a series with typical summer temperatures for each city.
+
+        >>> s = pd.Series([20, 21, 12], index=["London", "New York", "Helsinki"])
+        >>> s
+        London      20
+        New York    21
+        Helsinki    12
+        dtype: int64
+
+        Square the values by defining a function and passing it as an
+        argument to ``apply()``.
+
+        >>> def square(x):
+        ...     return x**2
+        >>> s.apply(square)
+        London      400
+        New York    441
+        Helsinki    144
+        dtype: int64
+
+        Square the values by passing an anonymous function as an
+        argument to ``apply()``.
+
+        >>> s.apply(lambda x: x**2)
+        London      400
+        New York    441
+        Helsinki    144
+        dtype: int64
+
+        Define a custom function that needs additional positional
+        arguments and pass these additional arguments using the
+        ``args`` keyword.
+
+        >>> def subtract_custom_value(x, custom_value):
+        ...     return x - custom_value
+
+        >>> s.apply(subtract_custom_value, args=(5,))
+        London      15
+        New York    16
+        Helsinki     7
+        dtype: int64
+
+        Define a custom function that takes keyword arguments
+        and pass these arguments to ``apply``.
+
+        >>> def add_custom_values(x, **kwargs):
+        ...     for month in kwargs:
+        ...         x += kwargs[month]
+        ...     return x
+
+        >>> s.apply(add_custom_values, june=30, july=20, august=25)
+        London      95
+        New York    96
+        Helsinki    87
+        dtype: int64
+
+        Use a function from the Numpy library.
+
+        >>> s.apply(np.log)
+        London      2.995732
+        New York    3.044522
+        Helsinki    2.484907
+        dtype: float64
+        """
+        return SeriesApply(
+            self,
+            func,
+            by_row=by_row,
+            args=args,
+            kwargs=kwargs,
+        ).apply()
+
+    def _reindex_indexer(
+        self,
+        new_index: Index | None,
+        indexer: npt.NDArray[np.intp] | None,
+    ) -> Series:
+        # Note: new_index is None iff indexer is None
+        # if not None, indexer is np.intp
+        if indexer is None and (
+            new_index is None or new_index.names == self.index.names
+        ):
+            return self.copy(deep=False)
+
+        new_values = algorithms.take_nd(
+            self._values, indexer, allow_fill=True, fill_value=None
+        )
+        return self._constructor(new_values, index=new_index, copy=False)
+
+    def _needs_reindex_multi(self, axes, method, level) -> bool:
+        """
+        Check if we do need a multi reindex; this is for compat with
+        higher dims.
+        """
+        return False
+
+    @overload
+    def rename(
+        self,
+        index: Renamer | Hashable | None = ...,
+        *,
+        axis: Axis | None = ...,
+        copy: bool | lib.NoDefault = ...,
+        inplace: Literal[True],
+        level: Level | None = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Series | None: ...
+
+    @overload
+    def rename(
+        self,
+        index: Renamer | Hashable | None = ...,
+        *,
+        axis: Axis | None = ...,
+        copy: bool | lib.NoDefault = ...,
+        inplace: Literal[False] = ...,
+        level: Level | None = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Series: ...
+
+    def rename(
+        self,
+        index: Renamer | Hashable | None = None,
+        *,
+        axis: Axis | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = False,
+        level: Level | None = None,
+        errors: IgnoreRaise = "ignore",
+    ) -> Series | None:
+        """
+        Alter Series index labels or name.
+
+        Function / dict values must be unique (1-to-1). Labels not contained in
+        a dict / Series will be left as-is. Extra labels listed don't throw an
+        error.
+
+        Alternatively, change ``Series.name`` with a scalar value.
+
+        See the :ref:`user guide <basics.rename>` for more.
+
+        Parameters
+        ----------
+        index : scalar, hashable sequence, dict-like or function optional
+            Functions or dict-like are transformations to apply to
+            the index.
+            Scalar or hashable sequence-like will alter the ``Series.name``
+            attribute.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        copy : bool, default False
+            Also copy underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+        inplace : bool, default False
+            Whether to return a new Series. If True the value of copy is ignored.
+        level : int or level name, default None
+            In case of MultiIndex, only rename labels in the specified level.
+        errors : {'ignore', 'raise'}, default 'ignore'
+            If 'raise', raise `KeyError` when a `dict-like mapper` or
+            `index` contains labels that are not present in the index being transformed.
+            If 'ignore', existing keys will be renamed and extra keys will be ignored.
+
+        Returns
+        -------
+        Series
+            A shallow copy with index labels or name altered, or the same object
+            if ``inplace=True`` and index is not a dict or callable else None.
+
+        See Also
+        --------
+        DataFrame.rename : Corresponding DataFrame method.
+        Series.rename_axis : Set the name of the axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s
+        0    1
+        1    2
+        2    3
+        dtype: int64
+        >>> s.rename("my_name")  # scalar, changes Series.name
+        0    1
+        1    2
+        2    3
+        Name: my_name, dtype: int64
+        >>> s.rename(lambda x: x**2)  # function, changes labels
+        0    1
+        1    2
+        4    3
+        dtype: int64
+        >>> s.rename({1: 3, 2: 5})  # mapping, changes labels
+        0    1
+        3    2
+        5    3
+        dtype: int64
+        """
+        self._check_copy_deprecation(copy)
+        if axis is not None:
+            # Make sure we raise if an invalid 'axis' is passed.
+            axis = self._get_axis_number(axis)
+
+        if callable(index) or is_dict_like(index):
+            # error: Argument 1 to "_rename" of "NDFrame" has incompatible
+            # type "Union[Union[Mapping[Any, Hashable], Callable[[Any],
+            # Hashable]], Hashable, None]"; expected "Union[Mapping[Any,
+            # Hashable], Callable[[Any], Hashable], None]"
+            return super()._rename(
+                index,  # type: ignore[arg-type]
+                inplace=inplace,
+                level=level,
+                errors=errors,
+            )
+        else:
+            return self._set_name(index, inplace=inplace)
+
+    @Appender(
+        """
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s
+        0    1
+        1    2
+        2    3
+        dtype: int64
+
+        >>> s.set_axis(['a', 'b', 'c'], axis=0)
+        a    1
+        b    2
+        c    3
+        dtype: int64
+    """
+    )
+    @Substitution(
+        klass=_shared_doc_kwargs["klass"],
+        axes_single_arg=_shared_doc_kwargs["axes_single_arg"],
+        extended_summary_sub="",
+        axis_description_sub="",
+        see_also_sub="",
+    )
+    @Appender(NDFrame.set_axis.__doc__)
+    def set_axis(
+        self,
+        labels,
+        *,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Series:
+        return super().set_axis(labels, axis=axis, copy=copy)
+
+    # error: Cannot determine type of 'reindex'
+    @doc(
+        NDFrame.reindex,  # type: ignore[has-type]
+        klass=_shared_doc_kwargs["klass"],
+        optional_reindex=_shared_doc_kwargs["optional_reindex"],
+    )
+    def reindex(  # type: ignore[override]
+        self,
+        index=None,
+        *,
+        axis: Axis | None = None,
+        method: ReindexMethod | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+        level: Level | None = None,
+        fill_value: Scalar | None = None,
+        limit: int | None = None,
+        tolerance=None,
+    ) -> Series:
+        return super().reindex(
+            index=index,
+            method=method,
+            level=level,
+            fill_value=fill_value,
+            limit=limit,
+            tolerance=tolerance,
+            copy=copy,
+        )
+
+    @overload  # type: ignore[override]
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = ...,
+        inplace: Literal[True],
+    ) -> None: ...
+
+    @overload
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = ...,
+        inplace: Literal[False] = ...,
+    ) -> Self: ...
+
+    @overload
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = ...,
+        *,
+        index=...,
+        axis: Axis = ...,
+        copy: bool | lib.NoDefault = ...,
+        inplace: bool = ...,
+    ) -> Self | None: ...
+
+    def rename_axis(
+        self,
+        mapper: IndexLabel | lib.NoDefault = lib.no_default,
+        *,
+        index=lib.no_default,
+        axis: Axis = 0,
+        copy: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = False,
+    ) -> Self | None:
+        """
+        Set the name of the axis for the index.
+
+        Parameters
+        ----------
+        mapper : scalar, list-like, optional
+            Value to set the axis name attribute.
+
+            Use either ``mapper`` and ``axis`` to
+            specify the axis to target with ``mapper``, or ``index``.
+
+        index : scalar, list-like, dict-like or function, optional
+            A scalar, list-like, dict-like or functions transformations to
+            apply to that axis' values.
+        axis : {0 or 'index'}, default 0
+            The axis to rename. For `Series` this parameter is unused and defaults to 0.
+        copy : bool, default False
+            Also copy underlying data.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+        inplace : bool, default False
+            Modifies the object directly, instead of creating a new Series
+            or DataFrame.
+
+        Returns
+        -------
+        Series, or None
+            The same type as the caller or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.rename : Alter Series index labels or name.
+        DataFrame.rename : Alter DataFrame index labels or name.
+        Index.rename : Set new names on index.
+
+        Examples
+        --------
+
+        >>> s = pd.Series(["dog", "cat", "monkey"])
+        >>> s
+        0       dog
+        1       cat
+        2    monkey
+        dtype: object
+        >>> s.rename_axis("animal")
+        animal
+        0    dog
+        1    cat
+        2    monkey
+        dtype: object
+        """
+        return super().rename_axis(
+            mapper=mapper,
+            index=index,
+            axis=axis,
+            inplace=inplace,
+            copy=copy,
+        )
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: Literal[True],
+        errors: IgnoreRaise = ...,
+    ) -> None: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: Literal[False] = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Series: ...
+
+    @overload
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = ...,
+        *,
+        axis: Axis = ...,
+        index: IndexLabel | ListLike = ...,
+        columns: IndexLabel | ListLike = ...,
+        level: Level | None = ...,
+        inplace: bool = ...,
+        errors: IgnoreRaise = ...,
+    ) -> Series | None: ...
+
+    def drop(
+        self,
+        labels: IndexLabel | ListLike = None,
+        *,
+        axis: Axis = 0,
+        index: IndexLabel | ListLike = None,
+        columns: IndexLabel | ListLike = None,
+        level: Level | None = None,
+        inplace: bool = False,
+        errors: IgnoreRaise = "raise",
+    ) -> Series | None:
+        """
+        Return Series with specified index labels removed.
+
+        Remove elements of a Series based on specifying the index labels.
+        When using a multi-index, labels on different levels can be removed
+        by specifying the level.
+
+        Parameters
+        ----------
+        labels : single label or list-like
+            Index labels to drop.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        index : single label or list-like
+            Redundant for application on Series, but 'index' can be used instead
+            of 'labels'.
+        columns : single label or list-like
+            No change is made to the Series; use 'index' or 'labels' instead.
+        level : int or level name, optional
+            For MultiIndex, level for which the labels will be removed.
+        inplace : bool, default False
+            If True, do operation inplace and return None.
+        errors : {'ignore', 'raise'}, default 'raise'
+            If 'ignore', suppress error and only existing labels are dropped.
+
+        Returns
+        -------
+        Series or None
+            Series with specified index labels removed or None if ``inplace=True``.
+
+        Raises
+        ------
+        KeyError
+            If none of the labels are found in the index.
+
+        See Also
+        --------
+        Series.reindex : Return only specified index labels of Series.
+        Series.dropna : Return series without null values.
+        Series.drop_duplicates : Return Series with duplicate values removed.
+        DataFrame.drop : Drop specified labels from rows or columns.
+
+        Examples
+        --------
+        >>> s = pd.Series(data=np.arange(3), index=["A", "B", "C"])
+        >>> s
+        A  0
+        B  1
+        C  2
+        dtype: int64
+
+        Drop labels B and C
+
+        >>> s.drop(labels=["B", "C"])
+        A  0
+        dtype: int64
+
+        Drop 2nd level label in MultiIndex Series
+
+        >>> midx = pd.MultiIndex(
+        ...     levels=[["llama", "cow", "falcon"], ["speed", "weight", "length"]],
+        ...     codes=[[0, 0, 0, 1, 1, 1, 2, 2, 2], [0, 1, 2, 0, 1, 2, 0, 1, 2]],
+        ... )
+        >>> s = pd.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3], index=midx)
+        >>> s
+        llama   speed      45.0
+                weight    200.0
+                length      1.2
+        cow     speed      30.0
+                weight    250.0
+                length      1.5
+        falcon  speed     320.0
+                weight      1.0
+                length      0.3
+        dtype: float64
+
+        >>> s.drop(labels="weight", level=1)
+        llama   speed      45.0
+                length      1.2
+        cow     speed      30.0
+                length      1.5
+        falcon  speed     320.0
+                length      0.3
+        dtype: float64
+        """
+        return super().drop(
+            labels=labels,
+            axis=axis,
+            index=index,
+            columns=columns,
+            level=level,
+            inplace=inplace,
+            errors=errors,
+        )
+
+    def pop(self, item: Hashable) -> Any:
+        """
+        Return item and drops from series. Raise KeyError if not found.
+
+        Parameters
+        ----------
+        item : label
+            Index of the element that needs to be removed.
+
+        Returns
+        -------
+        scalar
+            Value that is popped from series.
+
+        See Also
+        --------
+        Series.drop: Drop specified values from Series.
+        Series.drop_duplicates: Return Series with duplicate values removed.
+
+        Examples
+        --------
+        >>> ser = pd.Series([1, 2, 3])
+
+        >>> ser.pop(0)
+        1
+
+        >>> ser
+        1    2
+        2    3
+        dtype: int64
+        """
+        return super().pop(item=item)
+
+    @doc(INFO_DOCSTRING, **series_sub_kwargs)
+    def info(
+        self,
+        verbose: bool | None = None,
+        buf: IO[str] | None = None,
+        max_cols: int | None = None,
+        memory_usage: bool | str | None = None,
+        show_counts: bool = True,
+    ) -> None:
+        return SeriesInfo(self, memory_usage).render(
+            buf=buf,
+            max_cols=max_cols,
+            verbose=verbose,
+            show_counts=show_counts,
+        )
+
+    def memory_usage(self, index: bool = True, deep: bool = False) -> int:
+        """
+        Return the memory usage of the Series.
+
+        The memory usage can optionally include the contribution of
+        the index and of elements of `object` dtype.
+
+        Parameters
+        ----------
+        index : bool, default True
+            Specifies whether to include the memory usage of the Series index.
+        deep : bool, default False
+            If True, introspect the data deeply by interrogating
+            `object` dtypes for system-level memory consumption, and include
+            it in the returned value.
+
+        Returns
+        -------
+        int
+            Bytes of memory consumed.
+
+        See Also
+        --------
+        numpy.ndarray.nbytes : Total bytes consumed by the elements of the
+            array.
+        DataFrame.memory_usage : Bytes consumed by a DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series(range(3))
+        >>> s.memory_usage()
+        152
+
+        Not including the index gives the size of the rest of the data, which
+        is necessarily smaller:
+
+        >>> s.memory_usage(index=False)
+        24
+
+        The memory footprint of `object` values is ignored by default:
+
+        >>> s = pd.Series(["a", "b"])
+        >>> s.values
+        array(['a', 'b'], dtype=object)
+        >>> s.memory_usage()
+        144
+        >>> s.memory_usage(deep=True)
+        244
+        """
+        v = self._memory_usage(deep=deep)
+        if index:
+            v += self.index.memory_usage(deep=deep)
+        return v
+
+    def isin(self, values) -> Series:
+        """
+        Whether elements in Series are contained in `values`.
+
+        Return a boolean Series showing whether each element in the Series
+        matches an element in the passed sequence of `values` exactly.
+
+        Parameters
+        ----------
+        values : set or list-like
+            The sequence of values to test. Passing in a single string will
+            raise a ``TypeError``. Instead, turn a single string into a
+            list of one element.
+
+        Returns
+        -------
+        Series
+            Series of booleans indicating if each element is in values.
+
+        Raises
+        ------
+        TypeError
+          * If `values` is a string
+
+        See Also
+        --------
+        DataFrame.isin : Equivalent method on DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     ["llama", "cow", "llama", "beetle", "llama", "hippo"], name="animal"
+        ... )
+        >>> s.isin(["cow", "llama"])
+        0     True
+        1     True
+        2     True
+        3    False
+        4     True
+        5    False
+        Name: animal, dtype: bool
+
+        To invert the boolean values, use the ``~`` operator:
+
+        >>> ~s.isin(["cow", "llama"])
+        0    False
+        1    False
+        2    False
+        3     True
+        4    False
+        5     True
+        Name: animal, dtype: bool
+
+        Passing a single string as ``s.isin('llama')`` will raise an error. Use
+        a list of one element instead:
+
+        >>> s.isin(["llama"])
+        0     True
+        1    False
+        2     True
+        3    False
+        4     True
+        5    False
+        Name: animal, dtype: bool
+
+        Strings and integers are distinct and are therefore not comparable:
+
+        >>> pd.Series([1]).isin(["1"])
+        0    False
+        dtype: bool
+        >>> pd.Series([1.1]).isin(["1.1"])
+        0    False
+        dtype: bool
+        """
+        result = algorithms.isin(self._values, values)
+        return self._constructor(result, index=self.index, copy=False).__finalize__(
+            self, method="isin"
+        )
+
+    def between(
+        self,
+        left,
+        right,
+        inclusive: Literal["both", "neither", "left", "right"] = "both",
+    ) -> Series:
+        """
+        Return boolean Series equivalent to left <= series <= right.
+
+        This function returns a boolean vector containing `True` wherever the
+        corresponding Series element is between the boundary values `left` and
+        `right`. NA values are treated as `False`.
+
+        Parameters
+        ----------
+        left : scalar or list-like
+            Left boundary.
+        right : scalar or list-like
+            Right boundary.
+        inclusive : {"both", "neither", "left", "right"}
+            Include boundaries. Whether to set each bound as closed or open.
+
+            .. versionchanged:: 1.3.0
+
+        Returns
+        -------
+        Series
+            Series representing whether each element is between left and
+            right (inclusive).
+
+        See Also
+        --------
+        Series.gt : Greater than of series and other.
+        Series.lt : Less than of series and other.
+
+        Notes
+        -----
+        This function is equivalent to ``(left <= ser) & (ser <= right)``
+
+        Examples
+        --------
+        >>> s = pd.Series([2, 0, 4, 8, np.nan])
+
+        Boundary values are included by default:
+
+        >>> s.between(1, 4)
+        0     True
+        1    False
+        2     True
+        3    False
+        4    False
+        dtype: bool
+
+        With `inclusive` set to ``"neither"`` boundary values are excluded:
+
+        >>> s.between(1, 4, inclusive="neither")
+        0     True
+        1    False
+        2    False
+        3    False
+        4    False
+        dtype: bool
+
+        `left` and `right` can be any scalar value:
+
+        >>> s = pd.Series(["Alice", "Bob", "Carol", "Eve"])
+        >>> s.between("Anna", "Daniel")
+        0    False
+        1     True
+        2     True
+        3    False
+        dtype: bool
+        """
+        if inclusive == "both":
+            lmask = self >= left
+            rmask = self <= right
+        elif inclusive == "left":
+            lmask = self >= left
+            rmask = self < right
+        elif inclusive == "right":
+            lmask = self > left
+            rmask = self <= right
+        elif inclusive == "neither":
+            lmask = self > left
+            rmask = self < right
+        else:
+            raise ValueError(
+                "Inclusive has to be either string of 'both',"
+                "'left', 'right', or 'neither'."
+            )
+
+        return lmask & rmask
+
+    def case_when(
+        self,
+        caselist: list[
+            tuple[
+                ArrayLike | Callable[[Series], Series | np.ndarray | Sequence[bool]],
+                ArrayLike | Scalar | Callable[[Series], Series | np.ndarray],
+            ],
+        ],
+    ) -> Series:
+        """
+        Replace values where the conditions are True.
+
+        .. versionadded:: 2.2.0
+
+        Parameters
+        ----------
+        caselist : A list of tuples of conditions and expected replacements
+            Takes the form:  ``(condition0, replacement0)``,
+            ``(condition1, replacement1)``, ... .
+            ``condition`` should be a 1-D boolean array-like object
+            or a callable. If ``condition`` is a callable,
+            it is computed on the Series
+            and should return a boolean Series or array.
+            The callable must not change the input Series
+            (though pandas doesn`t check it). ``replacement`` should be a
+            1-D array-like object, a scalar or a callable.
+            If ``replacement`` is a callable, it is computed on the Series
+            and should return a scalar or Series. The callable
+            must not change the input Series
+            (though pandas doesn`t check it).
+
+        Returns
+        -------
+        Series
+            A new Series with values replaced based on the provided conditions.
+
+        See Also
+        --------
+        Series.mask : Replace values where the condition is True.
+
+        Examples
+        --------
+        >>> c = pd.Series([6, 7, 8, 9], name="c")
+        >>> a = pd.Series([0, 0, 1, 2])
+        >>> b = pd.Series([0, 3, 4, 5])
+
+        >>> c.case_when(
+        ...     caselist=[
+        ...         (a.gt(0), a),  # condition, replacement
+        ...         (b.gt(0), b),
+        ...     ]
+        ... )
+        0    6
+        1    3
+        2    1
+        3    2
+        Name: c, dtype: int64
+        """
+        if not isinstance(caselist, list):
+            raise TypeError(
+                f"The caselist argument should be a list; instead got {type(caselist)}"
+            )
+
+        if not caselist:
+            raise ValueError(
+                "provide at least one boolean condition, "
+                "with a corresponding replacement."
+            )
+
+        for num, entry in enumerate(caselist):
+            if not isinstance(entry, tuple):
+                raise TypeError(
+                    f"Argument {num} must be a tuple; instead got {type(entry)}."
+                )
+            if len(entry) != 2:
+                raise ValueError(
+                    f"Argument {num} must have length 2; "
+                    "a condition and replacement; "
+                    f"instead got length {len(entry)}."
+                )
+        caselist = [
+            (
+                com.apply_if_callable(condition, self),
+                com.apply_if_callable(replacement, self),
+            )
+            for condition, replacement in caselist
+        ]
+        default = self.copy(deep=False)
+        conditions, replacements = zip(*caselist)
+        common_dtypes = [infer_dtype_from(arg)[0] for arg in [*replacements, default]]
+        if len(set(common_dtypes)) > 1:
+            common_dtype = find_common_type(common_dtypes)
+            updated_replacements = []
+            for condition, replacement in zip(conditions, replacements):
+                if is_scalar(replacement):
+                    replacement = construct_1d_arraylike_from_scalar(
+                        value=replacement, length=len(condition), dtype=common_dtype
+                    )
+                elif isinstance(replacement, ABCSeries):
+                    replacement = replacement.astype(common_dtype)
+                else:
+                    replacement = pd_array(replacement, dtype=common_dtype)
+                updated_replacements.append(replacement)
+            replacements = updated_replacements
+            default = default.astype(common_dtype)
+
+        counter = range(len(conditions) - 1, -1, -1)
+        for position, condition, replacement in zip(
+            counter, reversed(conditions), reversed(replacements)
+        ):
+            try:
+                default = default.mask(
+                    condition, other=replacement, axis=0, inplace=False, level=None
+                )
+            except Exception as error:
+                raise ValueError(
+                    f"Failed to apply condition{position} and replacement{position}."
+                ) from error
+        return default
+
+    # error: Cannot determine type of 'isna'
+    @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
+    def isna(self) -> Series:
+        return NDFrame.isna(self)
+
+    # error: Cannot determine type of 'isna'
+    @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
+    def isnull(self) -> Series:
+        """
+        Series.isnull is an alias for Series.isna.
+        """
+        return super().isnull()
+
+    # error: Cannot determine type of 'notna'
+    @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
+    def notna(self) -> Series:
+        return super().notna()
+
+    # error: Cannot determine type of 'notna'
+    @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
+    def notnull(self) -> Series:
+        """
+        Series.notnull is an alias for Series.notna.
+        """
+        return super().notnull()
+
+    @overload
+    def dropna(
+        self,
+        *,
+        axis: Axis = ...,
+        inplace: Literal[False] = ...,
+        how: AnyAll | None = ...,
+        ignore_index: bool = ...,
+    ) -> Series: ...
+
+    @overload
+    def dropna(
+        self,
+        *,
+        axis: Axis = ...,
+        inplace: Literal[True],
+        how: AnyAll | None = ...,
+        ignore_index: bool = ...,
+    ) -> None: ...
+
+    def dropna(
+        self,
+        *,
+        axis: Axis = 0,
+        inplace: bool = False,
+        how: AnyAll | None = None,
+        ignore_index: bool = False,
+    ) -> Series | None:
+        """
+        Return a new Series with missing values removed.
+
+        See the :ref:`User Guide <missing_data>` for more on which values are
+        considered missing, and how to work with missing data.
+
+        Parameters
+        ----------
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+        inplace : bool, default False
+            If True, do operation inplace and return None.
+        how : str, optional
+            Not in use. Kept for compatibility.
+        ignore_index : bool, default ``False``
+            If ``True``, the resulting axis will be labeled 0, 1, …, n - 1.
+
+            .. versionadded:: 2.0.0
+
+        Returns
+        -------
+        Series or None
+            Series with NA entries dropped from it or None if ``inplace=True``.
+
+        See Also
+        --------
+        Series.isna: Indicate missing values.
+        Series.notna : Indicate existing (non-missing) values.
+        Series.fillna : Replace missing values.
+        DataFrame.dropna : Drop rows or columns which contain NA values.
+        Index.dropna : Drop missing indices.
+
+        Examples
+        --------
+        >>> ser = pd.Series([1.0, 2.0, np.nan])
+        >>> ser
+        0    1.0
+        1    2.0
+        2    NaN
+        dtype: float64
+
+        Drop NA values from a Series.
+
+        >>> ser.dropna()
+        0    1.0
+        1    2.0
+        dtype: float64
+
+        Empty strings are not considered NA values. ``None`` is considered an
+        NA value.
+
+        >>> ser = pd.Series([np.nan, 2, pd.NaT, "", None, "I stay"])
+        >>> ser
+        0       NaN
+        1         2
+        2       NaT
+        3
+        4      None
+        5    I stay
+        dtype: object
+        >>> ser.dropna()
+        1         2
+        3
+        5    I stay
+        dtype: object
+        """
+        inplace = validate_bool_kwarg(inplace, "inplace")
+        ignore_index = validate_bool_kwarg(ignore_index, "ignore_index")
+        # Validate the axis parameter
+        self._get_axis_number(axis or 0)
+
+        if self._can_hold_na:
+            result = remove_na_arraylike(self)
+        else:
+            if not inplace:
+                result = self.copy(deep=False)
+            else:
+                result = self
+
+        if ignore_index:
+            result.index = default_index(len(result))
+
+        if inplace:
+            return self._update_inplace(result)
+        else:
+            return result
+
+    # ----------------------------------------------------------------------
+    # Time series-oriented methods
+
+    def to_timestamp(
+        self,
+        freq: Frequency | None = None,
+        how: Literal["s", "e", "start", "end"] = "start",
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Series:
+        """
+        Cast to DatetimeIndex of Timestamps, at *beginning* of period.
+
+        This can be changed to the *end* of the period, by specifying `how="e"`.
+
+        Parameters
+        ----------
+        freq : str, default frequency of PeriodIndex
+            Desired frequency.
+        how : {'s', 'e', 'start', 'end'}
+            Convention for converting period to timestamp; start of period
+            vs. end.
+        copy : bool, default False
+            Whether or not to return a copy.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        Series with DatetimeIndex
+            Series with the PeriodIndex cast to DatetimeIndex.
+
+        See Also
+        --------
+        Series.to_period: Inverse method to cast DatetimeIndex to PeriodIndex.
+        DataFrame.to_timestamp: Equivalent method for DataFrame.
+
+        Examples
+        --------
+        >>> idx = pd.PeriodIndex(["2023", "2024", "2025"], freq="Y")
+        >>> s1 = pd.Series([1, 2, 3], index=idx)
+        >>> s1
+        2023    1
+        2024    2
+        2025    3
+        Freq: Y-DEC, dtype: int64
+
+        The resulting frequency of the Timestamps is `YearBegin`
+
+        >>> s1 = s1.to_timestamp()
+        >>> s1
+        2023-01-01    1
+        2024-01-01    2
+        2025-01-01    3
+        Freq: YS-JAN, dtype: int64
+
+        Using `freq` which is the offset that the Timestamps will have
+
+        >>> s2 = pd.Series([1, 2, 3], index=idx)
+        >>> s2 = s2.to_timestamp(freq="M")
+        >>> s2
+        2023-01-31    1
+        2024-01-31    2
+        2025-01-31    3
+        Freq: YE-JAN, dtype: int64
+        """
+        self._check_copy_deprecation(copy)
+        if not isinstance(self.index, PeriodIndex):
+            raise TypeError(f"unsupported Type {type(self.index).__name__}")
+
+        new_obj = self.copy(deep=False)
+        new_index = self.index.to_timestamp(freq=freq, how=how)
+        setattr(new_obj, "index", new_index)
+        return new_obj
+
+    def to_period(
+        self,
+        freq: str | None = None,
+        copy: bool | lib.NoDefault = lib.no_default,
+    ) -> Series:
+        """
+        Convert Series from DatetimeIndex to PeriodIndex.
+
+        Parameters
+        ----------
+        freq : str, default None
+            Frequency associated with the PeriodIndex.
+        copy : bool, default False
+            Whether or not to return a copy.
+
+            .. note::
+                The `copy` keyword will change behavior in pandas 3.0.
+                `Copy-on-Write
+                <https://pandas.pydata.org/docs/dev/user_guide/copy_on_write.html>`__
+                will be enabled by default, which means that all methods with a
+                `copy` keyword will use a lazy copy mechanism to defer the copy and
+                ignore the `copy` keyword. The `copy` keyword will be removed in a
+                future version of pandas.
+
+                You can already get the future behavior and improvements through
+                enabling copy on write ``pd.options.mode.copy_on_write = True``
+
+            .. deprecated:: 3.0.0
+
+        Returns
+        -------
+        Series
+            Series with index converted to PeriodIndex.
+
+        See Also
+        --------
+        DataFrame.to_period: Equivalent method for DataFrame.
+        Series.dt.to_period: Convert DateTime column values.
+
+        Examples
+        --------
+        >>> idx = pd.DatetimeIndex(["2023", "2024", "2025"])
+        >>> s = pd.Series([1, 2, 3], index=idx)
+        >>> s = s.to_period()
+        >>> s
+        2023    1
+        2024    2
+        2025    3
+        Freq: Y-DEC, dtype: int64
+
+        Viewing the index
+
+        >>> s.index
+        PeriodIndex(['2023', '2024', '2025'], dtype='period[Y-DEC]')
+        """
+        self._check_copy_deprecation(copy)
+        if not isinstance(self.index, DatetimeIndex):
+            raise TypeError(f"unsupported Type {type(self.index).__name__}")
+
+        new_obj = self.copy(deep=False)
+        new_index = self.index.to_period(freq=freq)
+        setattr(new_obj, "index", new_index)
+        return new_obj
+
+    # ----------------------------------------------------------------------
+    # Add index
+    _AXIS_ORDERS: list[Literal["index", "columns"]] = ["index"]
+    _AXIS_LEN = len(_AXIS_ORDERS)
+    _info_axis_number: Literal[0] = 0
+    _info_axis_name: Literal["index"] = "index"
+
+    index = properties.AxisProperty(
+        axis=0,
+        doc="""
+        The index (axis labels) of the Series.
+
+        The index of a Series is used to label and identify each element of the
+        underlying data. The index can be thought of as an immutable ordered set
+        (technically a multi-set, as it may contain duplicate labels), and is
+        used to index and align data in pandas.
+
+        Returns
+        -------
+        Index
+            The index labels of the Series.
+
+        See Also
+        --------
+        Series.reindex : Conform Series to new index.
+        Index : The base pandas index type.
+
+        Notes
+        -----
+        For more information on pandas indexing, see the `indexing user guide
+        <https://pandas.pydata.org/docs/user_guide/indexing.html>`__.
+
+        Examples
+        --------
+        To create a Series with a custom index and view the index labels:
+
+        >>> cities = ['Kolkata', 'Chicago', 'Toronto', 'Lisbon']
+        >>> populations = [14.85, 2.71, 2.93, 0.51]
+        >>> city_series = pd.Series(populations, index=cities)
+        >>> city_series.index
+        Index(['Kolkata', 'Chicago', 'Toronto', 'Lisbon'], dtype='object')
+
+        To change the index labels of an existing Series:
+
+        >>> city_series.index = ['KOL', 'CHI', 'TOR', 'LIS']
+        >>> city_series.index
+        Index(['KOL', 'CHI', 'TOR', 'LIS'], dtype='object')
+        """,
+    )
+
+    # ----------------------------------------------------------------------
+    # Accessor Methods
+    # ----------------------------------------------------------------------
+    str = Accessor("str", StringMethods)
+    dt = Accessor("dt", CombinedDatetimelikeProperties)
+    cat = Accessor("cat", CategoricalAccessor)
+    plot = Accessor("plot", pandas.plotting.PlotAccessor)
+    sparse = Accessor("sparse", SparseAccessor)
+    struct = Accessor("struct", StructAccessor)
+    list = Accessor("list", ListAccessor)
+
+    # ----------------------------------------------------------------------
+    # Add plotting methods to Series
+    hist = pandas.plotting.hist_series
+
+    # ----------------------------------------------------------------------
+    # Template-Based Arithmetic/Comparison Methods
+
+    def _cmp_method(self, other, op):
+        res_name = ops.get_op_result_name(self, other)
+
+        if isinstance(other, Series) and not self._indexed_same(other):
+            raise ValueError("Can only compare identically-labeled Series objects")
+
+        lvalues = self._values
+        rvalues = extract_array(other, extract_numpy=True, extract_range=True)
+
+        res_values = ops.comparison_op(lvalues, rvalues, op)
+
+        return self._construct_result(res_values, name=res_name)
+
+    def _logical_method(self, other, op):
+        res_name = ops.get_op_result_name(self, other)
+        self, other = self._align_for_op(other, align_asobject=True)
+
+        lvalues = self._values
+        rvalues = extract_array(other, extract_numpy=True, extract_range=True)
+
+        res_values = ops.logical_op(lvalues, rvalues, op)
+        return self._construct_result(res_values, name=res_name)
+
+    def _arith_method(self, other, op):
+        self, other = self._align_for_op(other)
+        return base.IndexOpsMixin._arith_method(self, other, op)
+
+    def _align_for_op(self, right, align_asobject: bool = False):
+        """align lhs and rhs Series"""
+        # TODO: Different from DataFrame._align_for_op, list, tuple and ndarray
+        # are not coerced here
+        # because Series has inconsistencies described in GH#13637
+        left = self
+
+        if isinstance(right, Series):
+            # avoid repeated alignment
+            if not left.index.equals(right.index):
+                if align_asobject:
+                    if left.dtype not in (object, np.bool_) or right.dtype not in (
+                        object,
+                        np.bool_,
+                    ):
+                        pass
+                        # GH#52538 no longer cast in these cases
+                    else:
+                        # to keep original value's dtype for bool ops
+                        left = left.astype(object)
+                        right = right.astype(object)
+
+                left, right = left.align(right)
+
+        return left, right
+
+    def _binop(self, other: Series, func, level=None, fill_value=None) -> Series:
+        """
+        Perform generic binary operation with optional fill value.
+
+        Parameters
+        ----------
+        other : Series
+        func : binary operator
+        fill_value : float or object
+            Value to substitute for NA/null values. If both Series are NA in a
+            location, the result will be NA regardless of the passed fill value.
+        level : int or level name, default None
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+
+        Returns
+        -------
+        Series
+        """
+        this = self
+
+        if not self.index.equals(other.index):
+            this, other = self.align(other, level=level, join="outer")
+
+        this_vals, other_vals = ops.fill_binop(this._values, other._values, fill_value)
+
+        with np.errstate(all="ignore"):
+            result = func(this_vals, other_vals)
+
+        name = ops.get_op_result_name(self, other)
+        out = this._construct_result(result, name)
+        return cast(Series, out)
+
+    def _construct_result(
+        self, result: ArrayLike | tuple[ArrayLike, ArrayLike], name: Hashable
+    ) -> Series | tuple[Series, Series]:
+        """
+        Construct an appropriately-labelled Series from the result of an op.
+
+        Parameters
+        ----------
+        result : ndarray or ExtensionArray
+        name : Label
+
+        Returns
+        -------
+        Series
+            In the case of __divmod__ or __rdivmod__, a 2-tuple of Series.
+        """
+        if isinstance(result, tuple):
+            # produced by divmod or rdivmod
+
+            res1 = self._construct_result(result[0], name=name)
+            res2 = self._construct_result(result[1], name=name)
+
+            # GH#33427 assertions to keep mypy happy
+            assert isinstance(res1, Series)
+            assert isinstance(res2, Series)
+            return (res1, res2)
+
+        # TODO: result should always be ArrayLike, but this fails for some
+        #  JSONArray tests
+        dtype = getattr(result, "dtype", None)
+        out = self._constructor(result, index=self.index, dtype=dtype, copy=False)
+        out = out.__finalize__(self)
+
+        # Set the result's name after __finalize__ is called because __finalize__
+        #  would set it back to self.name
+        out.name = name
+        return out
+
+    def _flex_method(self, other, op, *, level=None, fill_value=None, axis: Axis = 0):
+        if axis is not None:
+            self._get_axis_number(axis)
+
+        res_name = ops.get_op_result_name(self, other)
+
+        if isinstance(other, Series):
+            return self._binop(other, op, level=level, fill_value=fill_value)
+        elif isinstance(other, (np.ndarray, list, tuple)):
+            if len(other) != len(self):
+                raise ValueError("Lengths must be equal")
+            other = self._constructor(other, self.index, copy=False)
+            result = self._binop(other, op, level=level, fill_value=fill_value)
+            result._name = res_name
+            return result
+        else:
+            if fill_value is not None:
+                if isna(other):
+                    return op(self, fill_value)
+                self = self.fillna(fill_value)
+
+            return op(self, other)
+
+    def eq(
+        self,
+        other,
+        level: Level | None = None,
+        fill_value: float | None = None,
+        axis: Axis = 0,
+    ) -> Series:
+        """
+        Return Equal to of series and other, element-wise (binary operator `eq`).
+
+        Equivalent to ``series == other``, but with support to substitute a fill_value
+        for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            The second operand in this operation.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.ge : Return elementwise Greater than or equal to of series and other.
+        Series.le : Return elementwise Less than or equal to of series and other.
+        Series.gt : Return elementwise Greater than of series and other.
+        Series.lt : Return elementwise Less than of series and other.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        dtype: float64
+        >>> b = pd.Series([1, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+        >>> b
+        a    1.0
+        b    NaN
+        d    1.0
+        e    NaN
+        dtype: float64
+        >>> a.eq(b, fill_value=0)
+        a     True
+        b    False
+        c    False
+        d    False
+        e    False
+        dtype: bool
+        """
+        return self._flex_method(
+            other, operator.eq, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("ne", "series"))
+    def ne(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.ne, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def le(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        """
+        Return Less than or equal to of series and other, \
+        element-wise (binary operator `le`).
+
+        Equivalent to ``series <= other``, but with support to substitute a
+        fill_value for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            The second operand in this operation.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.ge : Return elementwise Greater than or equal to of series and other.
+        Series.lt : Return elementwise Less than of series and other.
+        Series.gt : Return elementwise Greater than of series and other.
+        Series.eq : Return elementwise equal to of series and other.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan, 1], index=['a', 'b', 'c', 'd', 'e'])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        e    1.0
+        dtype: float64
+        >>> b = pd.Series([0, 1, 2, np.nan, 1], index=['a', 'b', 'c', 'd', 'f'])
+        >>> b
+        a    0.0
+        b    1.0
+        c    2.0
+        d    NaN
+        f    1.0
+        dtype: float64
+        >>> a.le(b, fill_value=0)
+        a    False
+        b     True
+        c     True
+        d    False
+        e    False
+        f     True
+        dtype: bool
+        """
+        return self._flex_method(
+            other, operator.le, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("lt", "series"))
+    def lt(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.lt, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def ge(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        """
+        Return Greater than or equal to of series and other, \
+        element-wise (binary operator `ge`).
+
+        Equivalent to ``series >= other``, but with support to substitute a
+        fill_value for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            The second operand in this operation.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.gt : Greater than comparison, element-wise.
+        Series.le : Less than or equal to comparison, element-wise.
+        Series.lt : Less than comparison, element-wise.
+        Series.eq : Equal to comparison, element-wise.
+        Series.ne : Not equal to comparison, element-wise.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan, 1], index=["a", "b", "c", "d", "e"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        e    1.0
+        dtype: float64
+        >>> b = pd.Series([0, 1, 2, np.nan, 1], index=["a", "b", "c", "d", "f"])
+        >>> b
+        a    0.0
+        b    1.0
+        c    2.0
+        d    NaN
+        f    1.0
+        dtype: float64
+        >>> a.ge(b, fill_value=0)
+        a     True
+        b     True
+        c    False
+        d    False
+        e     True
+        f    False
+        dtype: bool
+        """
+        return self._flex_method(
+            other, operator.ge, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("gt", "series"))
+    def gt(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.gt, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def add(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        """
+        Return Addition of series and other, element-wise (binary operator `add`).
+
+        Equivalent to ``series + other``, but with support to substitute a fill_value
+        for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            With which to compute the addition.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.radd : Reverse of the Addition operator, see
+            `Python documentation
+            <https://docs.python.org/3/reference/datamodel.html#emulating-numeric-types>`_
+            for more details.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        dtype: float64
+        >>> b = pd.Series([1, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+        >>> b
+        a    1.0
+        b    NaN
+        d    1.0
+        e    NaN
+        dtype: float64
+        >>> a.add(b, fill_value=0)
+        a    2.0
+        b    1.0
+        c    1.0
+        d    1.0
+        e    NaN
+        dtype: float64
+        """
+        return self._flex_method(
+            other, operator.add, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("radd", "series"))
+    def radd(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.radd, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("sub", "series"))
+    def sub(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.sub, level=level, fill_value=fill_value, axis=axis
+        )
+
+    subtract = sub
+
+    @Appender(ops.make_flex_doc("rsub", "series"))
+    def rsub(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rsub, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def mul(
+        self,
+        other,
+        level: Level | None = None,
+        fill_value: float | None = None,
+        axis: Axis = 0,
+    ) -> Series:
+        """
+        Return Multiplication of series and other, element-wise (binary operator `mul`).
+
+        Equivalent to ``series * other``, but with support to substitute
+        a fill_value for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            With which to compute the multiplication.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.rmul : Reverse of the Multiplication operator, see
+            `Python documentation
+            <https://docs.python.org/3/reference/datamodel.html#emulating-numeric-types>`_
+            for more details.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        dtype: float64
+        >>> b = pd.Series([1, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+        >>> b
+        a    1.0
+        b    NaN
+        d    1.0
+        e    NaN
+        dtype: float64
+        >>> a.multiply(b, fill_value=0)
+        a    1.0
+        b    0.0
+        c    0.0
+        d    0.0
+        e    NaN
+        dtype: float64
+        >>> a.mul(5, fill_value=0)
+        a    5.0
+        b    5.0
+        c    5.0
+        d    0.0
+        dtype: float64
+        """
+        return self._flex_method(
+            other, operator.mul, level=level, fill_value=fill_value, axis=axis
+        )
+
+    multiply = mul
+
+    @Appender(ops.make_flex_doc("rmul", "series"))
+    def rmul(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rmul, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def truediv(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        """
+        Return Floating division of series and other, \
+        element-wise (binary operator `truediv`).
+
+        Equivalent to ``series / other``, but with support to substitute a
+        fill_value for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            Series with which to compute division.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.rtruediv : Reverse of the Floating division operator, see
+            `Python documentation
+            <https://docs.python.org/3/reference/datamodel.html#emulating-numeric-types>`_
+            for more details.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        dtype: float64
+        >>> b = pd.Series([1, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+        >>> b
+        a    1.0
+        b    NaN
+        d    1.0
+        e    NaN
+        dtype: float64
+        >>> a.divide(b, fill_value=0)
+        a    1.0
+        b    inf
+        c    inf
+        d    0.0
+        e    NaN
+        dtype: float64
+        """
+        return self._flex_method(
+            other, operator.truediv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    div = truediv
+    divide = truediv
+
+    @Appender(ops.make_flex_doc("rtruediv", "series"))
+    def rtruediv(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rtruediv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    rdiv = rtruediv
+
+    @Appender(ops.make_flex_doc("floordiv", "series"))
+    def floordiv(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.floordiv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rfloordiv", "series"))
+    def rfloordiv(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rfloordiv, level=level, fill_value=fill_value, axis=axis
+        )
+
+    def mod(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        """
+        Return Modulo of series and other, element-wise (binary operator `mod`).
+
+        Equivalent to ``series % other``, but with support to substitute a
+        fill_value for missing data in either one of the inputs.
+
+        Parameters
+        ----------
+        other : Series or scalar value
+            Series with which to compute modulo.
+        level : int or name
+            Broadcast across a level, matching Index values on the
+            passed MultiIndex level.
+        fill_value : None or float value, default None (NaN)
+            Fill existing missing (NaN) values, and any new element needed for
+            successful Series alignment, with this value before computation.
+            If data in both corresponding Series locations is missing
+            the result of filling (at that location) will be missing.
+        axis : {0 or 'index'}
+            Unused. Parameter needed for compatibility with DataFrame.
+
+        Returns
+        -------
+        Series
+            The result of the operation.
+
+        See Also
+        --------
+        Series.rmod : Reverse of the Modulo operator, see
+            `Python documentation
+            <https://docs.python.org/3/reference/datamodel.html#emulating-numeric-types>`_
+            for more details.
+
+        Examples
+        --------
+        >>> a = pd.Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        >>> a
+        a    1.0
+        b    1.0
+        c    1.0
+        d    NaN
+        dtype: float64
+        >>> b = pd.Series([1, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+        >>> b
+        a    1.0
+        b    NaN
+        d    1.0
+        e    NaN
+        dtype: float64
+        >>> a.mod(b, fill_value=0)
+        a    0.0
+        b    NaN
+        c    NaN
+        d    0.0
+        e    NaN
+        dtype: float64
+        """
+        return self._flex_method(
+            other, operator.mod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rmod", "series"))
+    def rmod(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rmod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("pow", "series"))
+    def pow(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, operator.pow, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rpow", "series"))
+    def rpow(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rpow, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("divmod", "series"))
+    def divmod(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, divmod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    @Appender(ops.make_flex_doc("rdivmod", "series"))
+    def rdivmod(self, other, level=None, fill_value=None, axis: Axis = 0) -> Series:
+        return self._flex_method(
+            other, roperator.rdivmod, level=level, fill_value=fill_value, axis=axis
+        )
+
+    # ----------------------------------------------------------------------
+    # Reductions
+
+    def _reduce(
+        self,
+        op,
+        # error: Variable "pandas.core.series.Series.str" is not valid as a type
+        name: str,  # type: ignore[valid-type]
+        *,
+        axis: Axis = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        filter_type=None,
+        **kwds,
+    ):
+        """
+        Perform a reduction operation.
+
+        If we have an ndarray as a value, then simply perform the operation,
+        otherwise delegate to the object.
+        """
+        delegate = self._values
+
+        if axis is not None:
+            self._get_axis_number(axis)
+
+        if isinstance(delegate, ExtensionArray):
+            # dispatch to ExtensionArray interface
+            return delegate._reduce(name, skipna=skipna, **kwds)
+
+        else:
+            # dispatch to numpy arrays
+            if numeric_only and self.dtype.kind not in "iufcb":
+                # i.e. not is_numeric_dtype(self.dtype)
+                kwd_name = "numeric_only"
+                if name in ["any", "all"]:
+                    kwd_name = "bool_only"
+                # GH#47500 - change to TypeError to match other methods
+                raise TypeError(
+                    f"Series.{name} does not allow {kwd_name}={numeric_only} "
+                    "with non-numeric dtypes."
+                )
+            return op(delegate, skipna=skipna, **kwds)
+
+    @Appender(make_doc("any", ndim=1))
+    # error: Signature of "any" incompatible with supertype "NDFrame"
+    def any(  # type: ignore[override]
+        self,
+        *,
+        axis: Axis = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> bool:
+        nv.validate_logical_func((), kwargs, fname="any")
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+        return self._reduce(
+            nanops.nanany,
+            name="any",
+            axis=axis,
+            numeric_only=bool_only,
+            skipna=skipna,
+            filter_type="bool",
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="all")
+    @Appender(make_doc("all", ndim=1))
+    def all(
+        self,
+        axis: Axis = 0,
+        bool_only: bool = False,
+        skipna: bool = True,
+        **kwargs,
+    ) -> bool:
+        nv.validate_logical_func((), kwargs, fname="all")
+        validate_bool_kwarg(skipna, "skipna", none_allowed=False)
+        return self._reduce(
+            nanops.nanall,
+            name="all",
+            axis=axis,
+            numeric_only=bool_only,
+            skipna=skipna,
+            filter_type="bool",
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="min")
+    def min(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        """
+        Return the minimum of the values over the requested axis.
+
+        If you want the *index* of the minimum, use ``idxmin``.
+        This is the equivalent of the ``numpy.ndarray`` method ``argmin``.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            The minimum of the values in the Series.
+
+        See Also
+        --------
+        numpy.min : Equivalent numpy function for arrays.
+        Series.min : Return the minimum.
+        Series.max : Return the maximum.
+        Series.idxmin : Return the index of the minimum.
+        Series.idxmax : Return the index of the maximum.
+        DataFrame.min : Return the minimum over the requested axis.
+        DataFrame.max : Return the maximum over the requested axis.
+        DataFrame.idxmin : Return the index of the minimum over the requested axis.
+        DataFrame.idxmax : Return the index of the maximum over the requested axis.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_arrays(
+        ...     [["warm", "warm", "cold", "cold"], ["dog", "falcon", "fish", "spider"]],
+        ...     names=["blooded", "animal"],
+        ... )
+        >>> s = pd.Series([4, 2, 0, 8], name="legs", index=idx)
+        >>> s
+        blooded  animal
+        warm     dog       4
+                 falcon    2
+        cold     fish      0
+                 spider    8
+        Name: legs, dtype: int64
+
+        >>> s.min()
+        0
+        """
+        return NDFrame.min(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="max")
+    def max(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        """
+        Return the maximum of the values over the requested axis.
+
+        If you want the *index* of the maximum, use ``idxmax``.
+        This is the equivalent of the ``numpy.ndarray`` method ``argmax``.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            The maximum of the values in the Series.
+
+        See Also
+        --------
+        numpy.max : Equivalent numpy function for arrays.
+        Series.min : Return the minimum.
+        Series.max : Return the maximum.
+        Series.idxmin : Return the index of the minimum.
+        Series.idxmax : Return the index of the maximum.
+        DataFrame.min : Return the minimum over the requested axis.
+        DataFrame.max : Return the maximum over the requested axis.
+        DataFrame.idxmin : Return the index of the minimum over the requested axis.
+        DataFrame.idxmax : Return the index of the maximum over the requested axis.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_arrays(
+        ...     [["warm", "warm", "cold", "cold"], ["dog", "falcon", "fish", "spider"]],
+        ...     names=["blooded", "animal"],
+        ... )
+        >>> s = pd.Series([4, 2, 0, 8], name="legs", index=idx)
+        >>> s
+        blooded  animal
+        warm     dog       4
+                 falcon    2
+        cold     fish      0
+                 spider    8
+        Name: legs, dtype: int64
+
+        >>> s.max()
+        8
+        """
+        return NDFrame.max(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="sum")
+    def sum(
+        self,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ):
+        """
+        Return the sum of the values over the requested axis.
+
+        This is equivalent to the method ``numpy.sum``.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.sum with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+
+        min_count : int, default 0
+            The required number of valid values to perform the operation. If fewer than
+            ``min_count`` non-NA values are present the result will be NA.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            Sum of the values for the requested axis.
+
+        See Also
+        --------
+        numpy.sum : Equivalent numpy function for computing sum.
+        Series.mean : Mean of the values.
+        Series.median : Median of the values.
+        Series.std : Standard deviation of the values.
+        Series.var : Variance of the values.
+        Series.min : Minimum value.
+        Series.max : Maximum value.
+
+        Examples
+        --------
+        >>> idx = pd.MultiIndex.from_arrays(
+        ...     [["warm", "warm", "cold", "cold"], ["dog", "falcon", "fish", "spider"]],
+        ...     names=["blooded", "animal"],
+        ... )
+        >>> s = pd.Series([4, 2, 0, 8], name="legs", index=idx)
+        >>> s
+        blooded  animal
+        warm     dog       4
+                 falcon    2
+        cold     fish      0
+                 spider    8
+        Name: legs, dtype: int64
+
+        >>> s.sum()
+        14
+
+        By default, the sum of an empty or all-NA Series is ``0``.
+
+        >>> pd.Series([], dtype="float64").sum()  # min_count=0 is the default
+        0.0
+
+        This can be controlled with the ``min_count`` parameter. For example, if
+        you'd like the sum of an empty series to be NaN, pass ``min_count=1``.
+
+        >>> pd.Series([], dtype="float64").sum(min_count=1)
+        nan
+
+        Thanks to the ``skipna`` parameter, ``min_count`` handles all-NA and
+        empty series identically.
+
+        >>> pd.Series([np.nan]).sum()
+        0.0
+
+        >>> pd.Series([np.nan]).sum(min_count=1)
+        nan
+        """
+        return NDFrame.sum(
+            self,
+            axis=axis,
+            skipna=skipna,
+            numeric_only=numeric_only,
+            min_count=min_count,
+            **kwargs,
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="prod")
+    @doc(make_doc("prod", ndim=1))
+    def prod(
+        self,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        min_count: int = 0,
+        **kwargs,
+    ):
+        return NDFrame.prod(
+            self,
+            axis=axis,
+            skipna=skipna,
+            numeric_only=numeric_only,
+            min_count=min_count,
+            **kwargs,
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="mean")
+    def mean(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Any:
+        """
+        Return the mean of the values over the requested axis.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            Mean of the values for the requested axis.
+
+        See Also
+        --------
+        numpy.median : Equivalent numpy function for computing median.
+        Series.sum : Sum of the values.
+        Series.median : Median of the values.
+        Series.std : Standard deviation of the values.
+        Series.var : Variance of the values.
+        Series.min : Minimum value.
+        Series.max : Maximum value.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.mean()
+        2.0
+        """
+        return NDFrame.mean(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="median")
+    def median(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ) -> Any:
+        """
+        Return the median of the values over the requested axis.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            Median of the values for the requested axis.
+
+        See Also
+        --------
+        numpy.median : Equivalent numpy function for computing median.
+        Series.sum : Sum of the values.
+        Series.median : Median of the values.
+        Series.std : Standard deviation of the values.
+        Series.var : Variance of the values.
+        Series.min : Minimum value.
+        Series.max : Maximum value.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 3])
+        >>> s.median()
+        2.0
+
+        With a DataFrame
+
+        >>> df = pd.DataFrame({"a": [1, 2], "b": [2, 3]}, index=["tiger", "zebra"])
+        >>> df
+               a   b
+        tiger  1   2
+        zebra  2   3
+        >>> df.median()
+        a   1.5
+        b   2.5
+        dtype: float64
+
+        Using axis=1
+
+        >>> df.median(axis=1)
+        tiger   1.5
+        zebra   2.5
+        dtype: float64
+
+        In this case, `numeric_only` should be set to `True`
+        to avoid getting an error.
+
+        >>> df = pd.DataFrame({"a": [1, 2], "b": ["T", "Z"]}, index=["tiger", "zebra"])
+        >>> df.median(numeric_only=True)
+        a   1.5
+        dtype: float64
+        """
+        return NDFrame.median(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="sem")
+    @doc(make_doc("sem", ndim=1))
+    def sem(
+        self,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        return NDFrame.sem(
+            self,
+            axis=axis,
+            skipna=skipna,
+            ddof=ddof,
+            numeric_only=numeric_only,
+            **kwargs,
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="var")
+    def var(
+        self,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        """
+        Return unbiased variance over requested axis.
+
+        Normalized by N-1 by default. This can be changed using the ddof argument.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            For `Series` this parameter is unused and defaults to 0.
+
+            .. warning::
+
+                The behavior of DataFrame.var with ``axis=None`` is deprecated,
+                in a future version this will reduce over both axes and return a scalar
+                To retain the old behavior, pass axis=0 (or do not pass axis).
+
+        skipna : bool, default True
+            Exclude NA/null values. If an entire row/column is NA, the result
+            will be NA.
+        ddof : int, default 1
+            Delta Degrees of Freedom. The divisor used in calculations is N - ddof,
+            where N represents the number of elements.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns. Not implemented for Series.
+        **kwargs :
+            Additional keywords passed.
+
+        Returns
+        -------
+        scalar or Series (if level specified)
+            Unbiased variance over requested axis.
+
+        See Also
+        --------
+        numpy.var : Equivalent function in NumPy.
+        Series.std : Returns the standard deviation of the Series.
+        DataFrame.var : Returns the variance of the DataFrame.
+        DataFrame.std : Return standard deviation of the values over
+            the requested axis.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "person_id": [0, 1, 2, 3],
+        ...         "age": [21, 25, 62, 43],
+        ...         "height": [1.61, 1.87, 1.49, 2.01],
+        ...     }
+        ... ).set_index("person_id")
+        >>> df
+                   age  height
+        person_id
+        0           21    1.61
+        1           25    1.87
+        2           62    1.49
+        3           43    2.01
+
+        >>> df.var()
+        age       352.916667
+        height      0.056367
+        dtype: float64
+
+        Alternatively, ``ddof=0`` can be set to normalize by N instead of N-1:
+
+        >>> df.var(ddof=0)
+        age       264.687500
+        height      0.042275
+        dtype: float64
+        """
+        return NDFrame.var(
+            self,
+            axis=axis,
+            skipna=skipna,
+            ddof=ddof,
+            numeric_only=numeric_only,
+            **kwargs,
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="std")
+    @doc(make_doc("std", ndim=1))
+    def std(
+        self,
+        axis: Axis | None = None,
+        skipna: bool = True,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        return NDFrame.std(
+            self,
+            axis=axis,
+            skipna=skipna,
+            ddof=ddof,
+            numeric_only=numeric_only,
+            **kwargs,
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="skew")
+    @doc(make_doc("skew", ndim=1))
+    def skew(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        return NDFrame.skew(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    @deprecate_nonkeyword_arguments(version="4.0", allowed_args=["self"], name="kurt")
+    def kurt(
+        self,
+        axis: Axis | None = 0,
+        skipna: bool = True,
+        numeric_only: bool = False,
+        **kwargs,
+    ):
+        """
+        Return unbiased kurtosis over requested axis.
+
+        Kurtosis obtained using Fisher's definition of
+        kurtosis (kurtosis of normal == 0.0). Normalized by N-1.
+
+        Parameters
+        ----------
+        axis : {index (0)}
+            Axis for the function to be applied on.
+            For `Series` this parameter is unused and defaults to 0.
+
+            For DataFrames, specifying ``axis=None`` will apply the aggregation
+            across both axes.
+
+            .. versionadded:: 2.0.0
+
+        skipna : bool, default True
+            Exclude NA/null values when computing the result.
+        numeric_only : bool, default False
+            Include only float, int, boolean columns.
+
+        **kwargs
+            Additional keyword arguments to be passed to the function.
+
+        Returns
+        -------
+        scalar
+            Unbiased kurtosis.
+
+        See Also
+        --------
+        Series.skew : Return unbiased skew over requested axis.
+        Series.var : Return unbiased variance over requested axis.
+        Series.std : Return unbiased standard deviation over requested axis.
+
+        Examples
+        --------
+        >>> s = pd.Series([1, 2, 2, 3], index=["cat", "dog", "dog", "mouse"])
+        >>> s
+        cat    1
+        dog    2
+        dog    2
+        mouse  3
+        dtype: int64
+        >>> s.kurt()
+        1.5
+        """
+        return NDFrame.kurt(
+            self, axis=axis, skipna=skipna, numeric_only=numeric_only, **kwargs
+        )
+
+    kurtosis = kurt
+    product = prod
+
+    @doc(make_doc("cummin", ndim=1))
+    def cummin(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return NDFrame.cummin(self, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cummax", ndim=1))
+    def cummax(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return NDFrame.cummax(self, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cumsum", ndim=1))
+    def cumsum(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return NDFrame.cumsum(self, axis, skipna, *args, **kwargs)
+
+    @doc(make_doc("cumprod", 1))
+    def cumprod(self, axis: Axis = 0, skipna: bool = True, *args, **kwargs) -> Self:
+        return NDFrame.cumprod(self, axis, skipna, *args, **kwargs)
diff --git a/pandas/pandas/core/shared_docs.py b/pandas/pandas/core/shared_docs.py
new file mode 100644
index 0000000000000..81fa508ae6d23
--- /dev/null
+++ b/pandas/pandas/core/shared_docs.py
@@ -0,0 +1,651 @@
+from __future__ import annotations
+
+_shared_docs: dict[str, str] = {}
+
+_shared_docs["aggregate"] = """
+Aggregate using one or more operations over the specified axis.
+
+Parameters
+----------
+func : function, str, list or dict
+    Function to use for aggregating the data. If a function, must either
+    work when passed a {klass} or when passed to {klass}.apply.
+
+    Accepted combinations are:
+
+    - function
+    - string function name
+    - list of functions and/or function names, e.g. ``[np.sum, 'mean']``
+    - dict of axis labels -> functions, function names or list of such.
+{axis}
+*args
+    Positional arguments to pass to `func`.
+**kwargs
+    Keyword arguments to pass to `func`.
+
+Returns
+-------
+scalar, Series or DataFrame
+
+    The return can be:
+
+    * scalar : when Series.agg is called with single function
+    * Series : when DataFrame.agg is called with a single function
+    * DataFrame : when DataFrame.agg is called with several functions
+{see_also}
+Notes
+-----
+The aggregation operations are always performed over an axis, either the
+index (default) or the column axis. This behavior is different from
+`numpy` aggregation functions (`mean`, `median`, `prod`, `sum`, `std`,
+`var`), where the default is to compute the aggregation of the flattened
+array, e.g., ``numpy.mean(arr_2d)`` as opposed to
+``numpy.mean(arr_2d, axis=0)``.
+
+`agg` is an alias for `aggregate`. Use the alias.
+
+Functions that mutate the passed object can produce unexpected
+behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+for more details.
+
+A passed user-defined-function will be passed a Series for evaluation.
+
+If ``func`` defines an index relabeling, ``axis`` must be ``0`` or ``index``.
+{examples}"""
+
+_shared_docs["compare"] = """
+Compare to another {klass} and show the differences.
+
+Parameters
+----------
+other : {klass}
+    Object to compare with.
+
+align_axis : {{0 or 'index', 1 or 'columns'}}, default 1
+    Determine which axis to align the comparison on.
+
+    * 0, or 'index' : Resulting differences are stacked vertically
+      with rows drawn alternately from self and other.
+    * 1, or 'columns' : Resulting differences are aligned horizontally
+      with columns drawn alternately from self and other.
+
+keep_shape : bool, default False
+    If true, all rows and columns are kept.
+    Otherwise, only the ones with different values are kept.
+
+keep_equal : bool, default False
+    If true, the result keeps values that are equal.
+    Otherwise, equal values are shown as NaNs.
+
+result_names : tuple, default ('self', 'other')
+    Set the dataframes names in the comparison.
+
+    .. versionadded:: 1.5.0
+"""
+
+_shared_docs["groupby"] = """
+Group %(klass)s using a mapper or by a Series of columns.
+
+A groupby operation involves some combination of splitting the
+object, applying a function, and combining the results. This can be
+used to group large amounts of data and compute operations on these
+groups.
+
+Parameters
+----------
+by : mapping, function, label, pd.Grouper or list of such
+    Used to determine the groups for the groupby.
+    If ``by`` is a function, it's called on each value of the object's
+    index. If a dict or Series is passed, the Series or dict VALUES
+    will be used to determine the groups (the Series' values are first
+    aligned; see ``.align()`` method). If a list or ndarray of length
+    equal to the selected axis is passed (see the `groupby user guide
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html#splitting-an-object-into-groups>`_),
+    the values are used as-is to determine the groups. A label or list
+    of labels may be passed to group by the columns in ``self``.
+    Notice that a tuple is interpreted as a (single) key.
+level : int, level name, or sequence of such, default None
+    If the axis is a MultiIndex (hierarchical), group by a particular
+    level or levels. Do not specify both ``by`` and ``level``.
+as_index : bool, default True
+    Return object with group labels as the
+    index. Only relevant for DataFrame input. as_index=False is
+    effectively "SQL-style" grouped output. This argument has no effect
+    on filtrations (see the `filtrations in the user guide
+    <https://pandas.pydata.org/docs/dev/user_guide/groupby.html#filtration>`_),
+    such as ``head()``, ``tail()``, ``nth()`` and in transformations
+    (see the `transformations in the user guide
+    <https://pandas.pydata.org/docs/dev/user_guide/groupby.html#transformation>`_).
+sort : bool, default True
+    Sort group keys. Get better performance by turning this off.
+    Note this does not influence the order of observations within each
+    group. Groupby preserves the order of rows within each group. If False,
+    the groups will appear in the same order as they did in the original DataFrame.
+    This argument has no effect on filtrations (see the `filtrations in the user guide
+    <https://pandas.pydata.org/docs/dev/user_guide/groupby.html#filtration>`_),
+    such as ``head()``, ``tail()``, ``nth()`` and in transformations
+    (see the `transformations in the user guide
+    <https://pandas.pydata.org/docs/dev/user_guide/groupby.html#transformation>`_).
+
+    .. versionchanged:: 2.0.0
+
+        Specifying ``sort=False`` with an ordered categorical grouper will no
+        longer sort the values.
+
+group_keys : bool, default True
+    When calling apply and the ``by`` argument produces a like-indexed
+    (i.e. :ref:`a transform <groupby.transform>`) result, add group keys to
+    index to identify pieces. By default group keys are not included
+    when the result's index (and column) labels match the inputs, and
+    are included otherwise.
+
+    .. versionchanged:: 1.5.0
+
+       Warns that ``group_keys`` will no longer be ignored when the
+       result from ``apply`` is a like-indexed Series or DataFrame.
+       Specify ``group_keys`` explicitly to include the group keys or
+       not.
+
+    .. versionchanged:: 2.0.0
+
+       ``group_keys`` now defaults to ``True``.
+
+observed : bool, default True
+    This only applies if any of the groupers are Categoricals.
+    If True: only show observed values for categorical groupers.
+    If False: show all values for categorical groupers.
+
+    .. versionchanged:: 3.0.0
+
+        The default value is now ``True``.
+
+dropna : bool, default True
+    If True, and if group keys contain NA values, NA values together
+    with row/column will be dropped.
+    If False, NA values will also be treated as the key in groups.
+
+Returns
+-------
+pandas.api.typing.%(klass)sGroupBy
+    Returns a groupby object that contains information about the groups.
+
+See Also
+--------
+resample : Convenience method for frequency conversion and resampling
+    of time series.
+
+Notes
+-----
+See the `user guide
+<https://pandas.pydata.org/pandas-docs/stable/groupby.html>`__ for more
+detailed usage and examples, including splitting an object into groups,
+iterating through groups, selecting a group, aggregation, and more.
+
+The implementation of groupby is hash-based, meaning in particular that
+objects that compare as equal will be considered to be in the same group.
+An exception to this is that pandas has special handling of NA values:
+any NA values will be collapsed to a single group, regardless of how
+they compare. See the user guide linked above for more details.
+"""
+
+_shared_docs["transform"] = """
+Call ``func`` on self producing a {klass} with the same axis shape as self.
+
+Parameters
+----------
+func : function, str, list-like or dict-like
+    Function to use for transforming the data. If a function, must either
+    work when passed a {klass} or when passed to {klass}.apply. If func
+    is both list-like and dict-like, dict-like behavior takes precedence.
+
+    Accepted combinations are:
+
+    - function
+    - string function name
+    - list-like of functions and/or function names, e.g. ``[np.exp, 'sqrt']``
+    - dict-like of axis labels -> functions, function names or list-like of such.
+{axis}
+*args
+    Positional arguments to pass to `func`.
+**kwargs
+    Keyword arguments to pass to `func`.
+
+Returns
+-------
+{klass}
+    A {klass} that must have the same length as self.
+
+Raises
+------
+ValueError : If the returned {klass} has a different length than self.
+
+See Also
+--------
+{klass}.agg : Only perform aggregating type operations.
+{klass}.apply : Invoke function on a {klass}.
+
+Notes
+-----
+Functions that mutate the passed object can produce unexpected
+behavior or errors and are not supported. See :ref:`gotchas.udf-mutation`
+for more details.
+
+Examples
+--------
+>>> df = pd.DataFrame({{'A': range(3), 'B': range(1, 4)}})
+>>> df
+   A  B
+0  0  1
+1  1  2
+2  2  3
+>>> df.transform(lambda x: x + 1)
+   A  B
+0  1  2
+1  2  3
+2  3  4
+
+Even though the resulting {klass} must have the same length as the
+input {klass}, it is possible to provide several input functions:
+
+>>> s = pd.Series(range(3))
+>>> s
+0    0
+1    1
+2    2
+dtype: int64
+>>> s.transform([np.sqrt, np.exp])
+       sqrt        exp
+0  0.000000   1.000000
+1  1.000000   2.718282
+2  1.414214   7.389056
+
+You can call transform on a GroupBy object:
+
+>>> df = pd.DataFrame({{
+...     "Date": [
+...         "2015-05-08", "2015-05-07", "2015-05-06", "2015-05-05",
+...         "2015-05-08", "2015-05-07", "2015-05-06", "2015-05-05"],
+...     "Data": [5, 8, 6, 1, 50, 100, 60, 120],
+... }})
+>>> df
+         Date  Data
+0  2015-05-08     5
+1  2015-05-07     8
+2  2015-05-06     6
+3  2015-05-05     1
+4  2015-05-08    50
+5  2015-05-07   100
+6  2015-05-06    60
+7  2015-05-05   120
+>>> df.groupby('Date')['Data'].transform('sum')
+0     55
+1    108
+2     66
+3    121
+4     55
+5    108
+6     66
+7    121
+Name: Data, dtype: int64
+
+>>> df = pd.DataFrame({{
+...     "c": [1, 1, 1, 2, 2, 2, 2],
+...     "type": ["m", "n", "o", "m", "m", "n", "n"]
+... }})
+>>> df
+   c type
+0  1    m
+1  1    n
+2  1    o
+3  2    m
+4  2    m
+5  2    n
+6  2    n
+>>> df['size'] = df.groupby('c')['type'].transform(len)
+>>> df
+   c type size
+0  1    m    3
+1  1    n    3
+2  1    o    3
+3  2    m    4
+4  2    m    4
+5  2    n    4
+6  2    n    4
+"""
+
+_shared_docs["storage_options"] = """storage_options : dict, optional
+    Extra options that make sense for a particular storage connection, e.g.
+    host, port, username, password, etc. For HTTP(S) URLs the key-value pairs
+    are forwarded to ``urllib.request.Request`` as header options. For other
+    URLs (e.g. starting with "s3://", and "gcs://") the key-value pairs are
+    forwarded to ``fsspec.open``. Please see ``fsspec`` and ``urllib`` for more
+    details, and for more examples on storage options refer `here
+    <https://pandas.pydata.org/docs/user_guide/io.html?
+    highlight=storage_options#reading-writing-remote-files>`_."""
+
+_shared_docs["compression_options"] = """compression : str or dict, default 'infer'
+    For on-the-fly compression of the output data. If 'infer' and '%s' is
+    path-like, then detect compression from the following extensions: '.gz',
+    '.bz2', '.zip', '.xz', '.zst', '.tar', '.tar.gz', '.tar.xz' or '.tar.bz2'
+    (otherwise no compression).
+    Set to ``None`` for no compression.
+    Can also be a dict with key ``'method'`` set
+    to one of {``'zip'``, ``'gzip'``, ``'bz2'``, ``'zstd'``, ``'xz'``, ``'tar'``} and
+    other key-value pairs are forwarded to
+    ``zipfile.ZipFile``, ``gzip.GzipFile``,
+    ``bz2.BZ2File``, ``zstandard.ZstdCompressor``, ``lzma.LZMAFile`` or
+    ``tarfile.TarFile``, respectively.
+    As an example, the following could be passed for faster compression and to create
+    a reproducible gzip archive:
+    ``compression={'method': 'gzip', 'compresslevel': 1, 'mtime': 1}``.
+
+    .. versionadded:: 1.5.0
+        Added support for `.tar` files."""
+
+_shared_docs["decompression_options"] = """compression : str or dict, default 'infer'
+    For on-the-fly decompression of on-disk data. If 'infer' and '%s' is
+    path-like, then detect compression from the following extensions: '.gz',
+    '.bz2', '.zip', '.xz', '.zst', '.tar', '.tar.gz', '.tar.xz' or '.tar.bz2'
+    (otherwise no compression).
+    If using 'zip' or 'tar', the ZIP file must contain only one data file to be read in.
+    Set to ``None`` for no decompression.
+    Can also be a dict with key ``'method'`` set
+    to one of {``'zip'``, ``'gzip'``, ``'bz2'``, ``'zstd'``, ``'xz'``, ``'tar'``} and
+    other key-value pairs are forwarded to
+    ``zipfile.ZipFile``, ``gzip.GzipFile``,
+    ``bz2.BZ2File``, ``zstandard.ZstdDecompressor``, ``lzma.LZMAFile`` or
+    ``tarfile.TarFile``, respectively.
+    As an example, the following could be passed for Zstandard decompression using a
+    custom compression dictionary:
+    ``compression={'method': 'zstd', 'dict_data': my_compression_dict}``.
+
+    .. versionadded:: 1.5.0
+        Added support for `.tar` files."""
+
+_shared_docs["replace"] = """
+    Replace values given in `to_replace` with `value`.
+
+    Values of the {klass} are replaced with other values dynamically.
+    This differs from updating with ``.loc`` or ``.iloc``, which require
+    you to specify a location to update with some value.
+
+    Parameters
+    ----------
+    to_replace : str, regex, list, dict, Series, int, float, or None
+        How to find the values that will be replaced.
+
+        * numeric, str or regex:
+
+            - numeric: numeric values equal to `to_replace` will be
+              replaced with `value`
+            - str: string exactly matching `to_replace` will be replaced
+              with `value`
+            - regex: regexs matching `to_replace` will be replaced with
+              `value`
+
+        * list of str, regex, or numeric:
+
+            - First, if `to_replace` and `value` are both lists, they
+              **must** be the same length.
+            - Second, if ``regex=True`` then all of the strings in **both**
+              lists will be interpreted as regexs otherwise they will match
+              directly. This doesn't matter much for `value` since there
+              are only a few possible substitution regexes you can use.
+            - str, regex and numeric rules apply as above.
+
+        * dict:
+
+            - Dicts can be used to specify different replacement values
+              for different existing values. For example,
+              ``{{'a': 'b', 'y': 'z'}}`` replaces the value 'a' with 'b' and
+              'y' with 'z'. To use a dict in this way, the optional `value`
+              parameter should not be given.
+            - For a DataFrame a dict can specify that different values
+              should be replaced in different columns. For example,
+              ``{{'a': 1, 'b': 'z'}}`` looks for the value 1 in column 'a'
+              and the value 'z' in column 'b' and replaces these values
+              with whatever is specified in `value`. The `value` parameter
+              should not be ``None`` in this case. You can treat this as a
+              special case of passing two lists except that you are
+              specifying the column to search in.
+            - For a DataFrame nested dictionaries, e.g.,
+              ``{{'a': {{'b': np.nan}}}}``, are read as follows: look in column
+              'a' for the value 'b' and replace it with NaN. The optional `value`
+              parameter should not be specified to use a nested dict in this
+              way. You can nest regular expressions as well. Note that
+              column names (the top-level dictionary keys in a nested
+              dictionary) **cannot** be regular expressions.
+
+        * None:
+
+            - This means that the `regex` argument must be a string,
+              compiled regular expression, or list, dict, ndarray or
+              Series of such elements. If `value` is also ``None`` then
+              this **must** be a nested dictionary or Series.
+
+        See the examples section for examples of each of these.
+    value : scalar, dict, list, str, regex, default None
+        Value to replace any values matching `to_replace` with.
+        For a DataFrame a dict of values can be used to specify which
+        value to use for each column (columns not in the dict will not be
+        filled). Regular expressions, strings and lists or dicts of such
+        objects are also allowed.
+    {inplace}
+    regex : bool or same types as `to_replace`, default False
+        Whether to interpret `to_replace` and/or `value` as regular
+        expressions. Alternatively, this could be a regular expression or a
+        list, dict, or array of regular expressions in which case
+        `to_replace` must be ``None``.
+
+    Returns
+    -------
+    {klass}
+        Object after replacement.
+
+    Raises
+    ------
+    AssertionError
+        * If `regex` is not a ``bool`` and `to_replace` is not
+          ``None``.
+
+    TypeError
+        * If `to_replace` is not a scalar, array-like, ``dict``, or ``None``
+        * If `to_replace` is a ``dict`` and `value` is not a ``list``,
+          ``dict``, ``ndarray``, or ``Series``
+        * If `to_replace` is ``None`` and `regex` is not compilable
+          into a regular expression or is a list, dict, ndarray, or
+          Series.
+        * When replacing multiple ``bool`` or ``datetime64`` objects and
+          the arguments to `to_replace` does not match the type of the
+          value being replaced
+
+    ValueError
+        * If a ``list`` or an ``ndarray`` is passed to `to_replace` and
+          `value` but they are not the same length.
+
+    See Also
+    --------
+    Series.fillna : Fill NA values.
+    DataFrame.fillna : Fill NA values.
+    Series.where : Replace values based on boolean condition.
+    DataFrame.where : Replace values based on boolean condition.
+    DataFrame.map: Apply a function to a Dataframe elementwise.
+    Series.map: Map values of Series according to an input mapping or function.
+    Series.str.replace : Simple string replacement.
+
+    Notes
+    -----
+    * Regex substitution is performed under the hood with ``re.sub``. The
+      rules for substitution for ``re.sub`` are the same.
+    * Regular expressions will only substitute on strings, meaning you
+      cannot provide, for example, a regular expression matching floating
+      point numbers and expect the columns in your frame that have a
+      numeric dtype to be matched. However, if those floating point
+      numbers *are* strings, then you can do this.
+    * This method has *a lot* of options. You are encouraged to experiment
+      and play with this method to gain intuition about how it works.
+    * When dict is used as the `to_replace` value, it is like
+      key(s) in the dict are the to_replace part and
+      value(s) in the dict are the value parameter.
+
+    Examples
+    --------
+
+    **Scalar `to_replace` and `value`**
+
+    >>> s = pd.Series([1, 2, 3, 4, 5])
+    >>> s.replace(1, 5)
+    0    5
+    1    2
+    2    3
+    3    4
+    4    5
+    dtype: int64
+
+    >>> df = pd.DataFrame({{'A': [0, 1, 2, 3, 4],
+    ...                    'B': [5, 6, 7, 8, 9],
+    ...                    'C': ['a', 'b', 'c', 'd', 'e']}})
+    >>> df.replace(0, 5)
+        A  B  C
+    0  5  5  a
+    1  1  6  b
+    2  2  7  c
+    3  3  8  d
+    4  4  9  e
+
+    **List-like `to_replace`**
+
+    >>> df.replace([0, 1, 2, 3], 4)
+        A  B  C
+    0  4  5  a
+    1  4  6  b
+    2  4  7  c
+    3  4  8  d
+    4  4  9  e
+
+    >>> df.replace([0, 1, 2, 3], [4, 3, 2, 1])
+        A  B  C
+    0  4  5  a
+    1  3  6  b
+    2  2  7  c
+    3  1  8  d
+    4  4  9  e
+
+    **dict-like `to_replace`**
+
+    >>> df.replace({{0: 10, 1: 100}})
+            A  B  C
+    0   10  5  a
+    1  100  6  b
+    2    2  7  c
+    3    3  8  d
+    4    4  9  e
+
+    >>> df.replace({{'A': 0, 'B': 5}}, 100)
+            A    B  C
+    0  100  100  a
+    1    1    6  b
+    2    2    7  c
+    3    3    8  d
+    4    4    9  e
+
+    >>> df.replace({{'A': {{0: 100, 4: 400}}}})
+            A  B  C
+    0  100  5  a
+    1    1  6  b
+    2    2  7  c
+    3    3  8  d
+    4  400  9  e
+
+    **Regular expression `to_replace`**
+
+    >>> df = pd.DataFrame({{'A': ['bat', 'foo', 'bait'],
+    ...                    'B': ['abc', 'bar', 'xyz']}})
+    >>> df.replace(to_replace=r'^ba.$', value='new', regex=True)
+            A    B
+    0   new  abc
+    1   foo  new
+    2  bait  xyz
+
+    >>> df.replace({{'A': r'^ba.$'}}, {{'A': 'new'}}, regex=True)
+            A    B
+    0   new  abc
+    1   foo  bar
+    2  bait  xyz
+
+    >>> df.replace(regex=r'^ba.$', value='new')
+            A    B
+    0   new  abc
+    1   foo  new
+    2  bait  xyz
+
+    >>> df.replace(regex={{r'^ba.$': 'new', 'foo': 'xyz'}})
+            A    B
+    0   new  abc
+    1   xyz  new
+    2  bait  xyz
+
+    >>> df.replace(regex=[r'^ba.$', 'foo'], value='new')
+            A    B
+    0   new  abc
+    1   new  new
+    2  bait  xyz
+
+    Compare the behavior of ``s.replace({{'a': None}})`` and
+    ``s.replace('a', None)`` to understand the peculiarities
+    of the `to_replace` parameter:
+
+    >>> s = pd.Series([10, 'a', 'a', 'b', 'a'])
+
+    When one uses a dict as the `to_replace` value, it is like the
+    value(s) in the dict are equal to the `value` parameter.
+    ``s.replace({{'a': None}})`` is equivalent to
+    ``s.replace(to_replace={{'a': None}}, value=None)``:
+
+    >>> s.replace({{'a': None}})
+    0      10
+    1    None
+    2    None
+    3       b
+    4    None
+    dtype: object
+
+    If ``None`` is explicitly passed for ``value``, it will be respected:
+
+    >>> s.replace('a', None)
+    0      10
+    1    None
+    2    None
+    3       b
+    4    None
+    dtype: object
+
+        .. versionchanged:: 1.4.0
+            Previously the explicit ``None`` was silently ignored.
+
+    When ``regex=True``, ``value`` is not ``None`` and `to_replace` is a string,
+    the replacement will be applied in all columns of the DataFrame.
+
+    >>> df = pd.DataFrame({{'A': [0, 1, 2, 3, 4],
+    ...                    'B': ['a', 'b', 'c', 'd', 'e'],
+    ...                    'C': ['f', 'g', 'h', 'i', 'j']}})
+
+    >>> df.replace(to_replace='^[a-g]', value='e', regex=True)
+        A  B  C
+    0  0  e  e
+    1  1  e  e
+    2  2  e  h
+    3  3  e  i
+    4  4  e  j
+
+    If ``value`` is not ``None`` and `to_replace` is a dictionary, the dictionary
+    keys will be the DataFrame columns that the replacement will be applied.
+
+    >>> df.replace(to_replace={{'B': '^[a-c]', 'C': '^[h-j]'}}, value='e', regex=True)
+        A  B  C
+    0  0  e  f
+    1  1  e  g
+    2  2  e  e
+    3  3  d  e
+    4  4  e  e
+"""
diff --git a/pandas/pandas/core/sorting.py b/pandas/pandas/core/sorting.py
new file mode 100644
index 0000000000000..0d8f42694ccb4
--- /dev/null
+++ b/pandas/pandas/core/sorting.py
@@ -0,0 +1,730 @@
+"""miscellaneous sorting / groupby utilities"""
+
+from __future__ import annotations
+
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    algos,
+    hashtable,
+    lib,
+)
+from pandas._libs.hashtable import unique_label_indices
+
+from pandas.core.dtypes.common import (
+    ensure_int64,
+    ensure_platform_int,
+)
+from pandas.core.dtypes.generic import (
+    ABCMultiIndex,
+    ABCRangeIndex,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.construction import extract_array
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        AxisInt,
+        IndexKeyFunc,
+        Level,
+        NaPosition,
+        Shape,
+        SortKind,
+        npt,
+    )
+
+    from pandas import (
+        MultiIndex,
+        Series,
+    )
+    from pandas.core.arrays import ExtensionArray
+    from pandas.core.indexes.base import Index
+
+
+def get_indexer_indexer(
+    target: Index,
+    level: Level | list[Level] | None,
+    ascending: list[bool] | bool,
+    kind: SortKind,
+    na_position: NaPosition,
+    sort_remaining: bool,
+    key: IndexKeyFunc,
+) -> npt.NDArray[np.intp] | None:
+    """
+    Helper method that return the indexer according to input parameters for
+    the sort_index method of DataFrame and Series.
+
+    Parameters
+    ----------
+    target : Index
+    level : int or level name or list of ints or list of level names
+    ascending : bool or list of bools, default True
+    kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}
+    na_position : {'first', 'last'}
+    sort_remaining : bool
+    key : callable, optional
+
+    Returns
+    -------
+    Optional[ndarray[intp]]
+        The indexer for the new index.
+    """
+
+    # error: Incompatible types in assignment (expression has type
+    # "Union[ExtensionArray, ndarray[Any, Any], Index, Series]", variable has
+    # type "Index")
+    target = ensure_key_mapped(target, key, levels=level)  # type: ignore[assignment]
+    target = target._sort_levels_monotonic()
+
+    if level is not None:
+        _, indexer = target.sortlevel(
+            level,
+            ascending=ascending,
+            sort_remaining=sort_remaining,
+            na_position=na_position,
+        )
+    elif (np.all(ascending) and target.is_monotonic_increasing) or (
+        not np.any(ascending) and target.is_monotonic_decreasing
+    ):
+        # Check monotonic-ness before sort an index (GH 11080)
+        return None
+    elif isinstance(target, ABCMultiIndex):
+        codes = [lev.codes for lev in target._get_codes_for_sorting()]
+        indexer = lexsort_indexer(
+            codes, orders=ascending, na_position=na_position, codes_given=True
+        )
+    else:
+        # ascending can only be a Sequence for MultiIndex
+        indexer = nargsort(
+            target,
+            kind=kind,
+            ascending=cast(bool, ascending),
+            na_position=na_position,
+        )
+    return indexer
+
+
+def get_group_index(
+    labels, shape: Shape, sort: bool, xnull: bool
+) -> npt.NDArray[np.int64]:
+    """
+    For the particular label_list, gets the offsets into the hypothetical list
+    representing the totally ordered cartesian product of all possible label
+    combinations, *as long as* this space fits within int64 bounds;
+    otherwise, though group indices identify unique combinations of
+    labels, they cannot be deconstructed.
+    - If `sort`, rank of returned ids preserve lexical ranks of labels.
+      i.e. returned id's can be used to do lexical sort on labels;
+    - If `xnull` nulls (-1 labels) are passed through.
+
+    Parameters
+    ----------
+    labels : sequence of arrays
+        Integers identifying levels at each location
+    shape : tuple[int, ...]
+        Number of unique levels at each location
+    sort : bool
+        If the ranks of returned ids should match lexical ranks of labels
+    xnull : bool
+        If true nulls are excluded. i.e. -1 values in the labels are
+        passed through.
+
+    Returns
+    -------
+    An array of type int64 where two elements are equal if their corresponding
+    labels are equal at all location.
+
+    Notes
+    -----
+    The length of `labels` and `shape` must be identical.
+    """
+
+    def _int64_cut_off(shape) -> int:
+        acc = 1
+        for i, mul in enumerate(shape):
+            acc *= int(mul)
+            if not acc < lib.i8max:
+                return i
+        return len(shape)
+
+    def maybe_lift(lab, size: int) -> tuple[np.ndarray, int]:
+        # promote nan values (assigned -1 label in lab array)
+        # so that all output values are non-negative
+        return (lab + 1, size + 1) if (lab == -1).any() else (lab, size)
+
+    labels = [ensure_int64(x) for x in labels]
+    lshape = list(shape)
+    if not xnull:
+        for i, (lab, size) in enumerate(zip(labels, shape)):
+            labels[i], lshape[i] = maybe_lift(lab, size)
+
+    # Iteratively process all the labels in chunks sized so less
+    # than lib.i8max unique int ids will be required for each chunk
+    while True:
+        # how many levels can be done without overflow:
+        nlev = _int64_cut_off(lshape)
+
+        # compute flat ids for the first `nlev` levels
+        stride = np.prod(lshape[1:nlev], dtype="i8")
+        out = stride * labels[0].astype("i8", subok=False, copy=False)
+
+        for i in range(1, nlev):
+            if lshape[i] == 0:
+                stride = np.int64(0)
+            else:
+                stride //= lshape[i]
+            out += labels[i] * stride
+
+        if xnull:  # exclude nulls
+            mask = labels[0] == -1
+            for lab in labels[1:nlev]:
+                mask |= lab == -1
+            out[mask] = -1
+
+        if nlev == len(lshape):  # all levels done!
+            break
+
+        # compress what has been done so far in order to avoid overflow
+        # to retain lexical ranks, obs_ids should be sorted
+        comp_ids, obs_ids = compress_group_index(out, sort=sort)
+
+        labels = [comp_ids] + labels[nlev:]
+        lshape = [len(obs_ids)] + lshape[nlev:]
+
+    return out
+
+
+def get_compressed_ids(
+    labels, sizes: Shape
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.int64]]:
+    """
+    Group_index is offsets into cartesian product of all possible labels. This
+    space can be huge, so this function compresses it, by computing offsets
+    (comp_ids) into the list of unique labels (obs_group_ids).
+
+    Parameters
+    ----------
+    labels : list of label arrays
+    sizes : tuple[int] of size of the levels
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+        comp_ids
+    np.ndarray[np.int64]
+        obs_group_ids
+    """
+    ids = get_group_index(labels, sizes, sort=True, xnull=False)
+    return compress_group_index(ids, sort=True)
+
+
+def is_int64_overflow_possible(shape: Shape) -> bool:
+    the_prod = 1
+    for x in shape:
+        the_prod *= int(x)
+
+    return the_prod >= lib.i8max
+
+
+def _decons_group_index(
+    comp_labels: npt.NDArray[np.intp], shape: Shape
+) -> list[npt.NDArray[np.intp]]:
+    # reconstruct labels
+    if is_int64_overflow_possible(shape):
+        # at some point group indices are factorized,
+        # and may not be deconstructed here! wrong path!
+        raise ValueError("cannot deconstruct factorized group indices!")
+
+    label_list = []
+    factor = 1
+    y = np.array(0)
+    x = comp_labels
+    for i in reversed(range(len(shape))):
+        labels = (x - y) % (factor * shape[i]) // factor
+        np.putmask(labels, comp_labels < 0, -1)
+        label_list.append(labels)
+        y = labels * factor
+        factor *= shape[i]
+    return label_list[::-1]
+
+
+def decons_obs_group_ids(
+    comp_ids: npt.NDArray[np.intp],
+    obs_ids: npt.NDArray[np.intp],
+    shape: Shape,
+    labels: Sequence[npt.NDArray[np.signedinteger]],
+    xnull: bool,
+) -> list[npt.NDArray[np.intp]]:
+    """
+    Reconstruct labels from observed group ids.
+
+    Parameters
+    ----------
+    comp_ids : np.ndarray[np.intp]
+    obs_ids: np.ndarray[np.intp]
+    shape : tuple[int]
+    labels : Sequence[np.ndarray[np.signedinteger]]
+    xnull : bool
+        If nulls are excluded; i.e. -1 labels are passed through.
+    """
+    if not xnull:
+        lift = np.fromiter(((a == -1).any() for a in labels), dtype=np.intp)
+        arr_shape = np.asarray(shape, dtype=np.intp) + lift
+        shape = tuple(arr_shape)
+
+    if not is_int64_overflow_possible(shape):
+        # obs ids are deconstructable! take the fast route!
+        out = _decons_group_index(obs_ids, shape)
+        return out if xnull or not lift.any() else [x - y for x, y in zip(out, lift)]
+
+    indexer = unique_label_indices(comp_ids)
+    return [lab[indexer].astype(np.intp, subok=False, copy=True) for lab in labels]
+
+
+def lexsort_indexer(
+    keys: Sequence[ArrayLike | Index | Series],
+    orders=None,
+    na_position: str = "last",
+    key: Callable | None = None,
+    codes_given: bool = False,
+) -> npt.NDArray[np.intp]:
+    """
+    Performs lexical sorting on a set of keys
+
+    Parameters
+    ----------
+    keys : Sequence[ArrayLike | Index | Series]
+        Sequence of arrays to be sorted by the indexer
+        Sequence[Series] is only if key is not None.
+    orders : bool or list of booleans, optional
+        Determines the sorting order for each element in keys. If a list,
+        it must be the same length as keys. This determines whether the
+        corresponding element in keys should be sorted in ascending
+        (True) or descending (False) order. if bool, applied to all
+        elements as above. if None, defaults to True.
+    na_position : {'first', 'last'}, default 'last'
+        Determines placement of NA elements in the sorted list ("last" or "first")
+    key : Callable, optional
+        Callable key function applied to every element in keys before sorting
+    codes_given: bool, False
+        Avoid categorical materialization if codes are already provided.
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+    """
+    from pandas.core.arrays import Categorical
+
+    if na_position not in ["last", "first"]:
+        raise ValueError(f"invalid na_position: {na_position}")
+
+    if isinstance(orders, bool):
+        orders = itertools.repeat(orders, len(keys))
+    elif orders is None:
+        orders = itertools.repeat(True, len(keys))
+    else:
+        orders = reversed(orders)
+
+    labels = []
+
+    for k, order in zip(reversed(keys), orders):
+        k = ensure_key_mapped(k, key)
+        if codes_given:
+            codes = cast(np.ndarray, k)
+            n = codes.max() + 1 if len(codes) else 0
+        else:
+            cat = Categorical(k, ordered=True)
+            codes = cat.codes
+            n = len(cat.categories)
+
+        mask = codes == -1
+
+        if na_position == "last" and mask.any():
+            codes = np.where(mask, n, codes)
+
+        # not order means descending
+        if not order:
+            codes = np.where(mask, codes, n - codes - 1)
+
+        labels.append(codes)
+
+    return np.lexsort(labels)
+
+
+def nargsort(
+    items: ArrayLike | Index | Series,
+    kind: SortKind = "quicksort",
+    ascending: bool = True,
+    na_position: str = "last",
+    key: Callable | None = None,
+    mask: npt.NDArray[np.bool_] | None = None,
+) -> npt.NDArray[np.intp]:
+    """
+    Intended to be a drop-in replacement for np.argsort which handles NaNs.
+
+    Adds ascending, na_position, and key parameters.
+
+    (GH #6399, #5231, #27237)
+
+    Parameters
+    ----------
+    items : np.ndarray, ExtensionArray, Index, or Series
+    kind : {'quicksort', 'mergesort', 'heapsort', 'stable'}, default 'quicksort'
+    ascending : bool, default True
+    na_position : {'first', 'last'}, default 'last'
+    key : Optional[Callable], default None
+    mask : Optional[np.ndarray[bool]], default None
+        Passed when called by ExtensionArray.argsort.
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+    """
+
+    if key is not None:
+        # see TestDataFrameSortKey, TestRangeIndex::test_sort_values_key
+        items = ensure_key_mapped(items, key)
+        return nargsort(
+            items,
+            kind=kind,
+            ascending=ascending,
+            na_position=na_position,
+            key=None,
+            mask=mask,
+        )
+
+    if isinstance(items, ABCRangeIndex):
+        return items.argsort(ascending=ascending)
+    elif not isinstance(items, ABCMultiIndex):
+        items = extract_array(items)
+    else:
+        raise TypeError(
+            "nargsort does not support MultiIndex. Use index.sort_values instead."
+        )
+
+    if mask is None:
+        mask = np.asarray(isna(items))
+
+    if not isinstance(items, np.ndarray):
+        # i.e. ExtensionArray
+        return items.argsort(
+            ascending=ascending,
+            kind=kind,
+            na_position=na_position,
+        )
+
+    idx = np.arange(len(items))
+    non_nans = items[~mask]
+    non_nan_idx = idx[~mask]
+
+    nan_idx = np.nonzero(mask)[0]
+    if not ascending:
+        non_nans = non_nans[::-1]
+        non_nan_idx = non_nan_idx[::-1]
+    indexer = non_nan_idx[non_nans.argsort(kind=kind)]
+    if not ascending:
+        indexer = indexer[::-1]
+    # Finally, place the NaNs at the end or the beginning according to
+    # na_position
+    if na_position == "last":
+        indexer = np.concatenate([indexer, nan_idx])
+    elif na_position == "first":
+        indexer = np.concatenate([nan_idx, indexer])
+    else:
+        raise ValueError(f"invalid na_position: {na_position}")
+    return ensure_platform_int(indexer)
+
+
+def nargminmax(values: ExtensionArray, method: str, axis: AxisInt = 0):
+    """
+    Implementation of np.argmin/argmax but for ExtensionArray and which
+    handles missing values.
+
+    Parameters
+    ----------
+    values : ExtensionArray
+    method : {"argmax", "argmin"}
+    axis : int, default 0
+
+    Returns
+    -------
+    int
+    """
+    assert method in {"argmax", "argmin"}
+    func = np.argmax if method == "argmax" else np.argmin
+
+    mask = np.asarray(isna(values))
+    arr_values = values._values_for_argsort()
+
+    if arr_values.ndim > 1:
+        if mask.any():
+            if axis == 1:
+                zipped = zip(arr_values, mask)
+            else:
+                zipped = zip(arr_values.T, mask.T)
+            return np.array([_nanargminmax(v, m, func) for v, m in zipped])
+        return func(arr_values, axis=axis)
+
+    return _nanargminmax(arr_values, mask, func)
+
+
+def _nanargminmax(values: np.ndarray, mask: npt.NDArray[np.bool_], func) -> int:
+    """
+    See nanargminmax.__doc__.
+    """
+    idx = np.arange(values.shape[0])
+    non_nans = values[~mask]
+    non_nan_idx = idx[~mask]
+
+    return non_nan_idx[func(non_nans)]
+
+
+def _ensure_key_mapped_multiindex(
+    index: MultiIndex, key: Callable, level=None
+) -> MultiIndex:
+    """
+    Returns a new MultiIndex in which key has been applied
+    to all levels specified in level (or all levels if level
+    is None). Used for key sorting for MultiIndex.
+
+    Parameters
+    ----------
+    index : MultiIndex
+        Index to which to apply the key function on the
+        specified levels.
+    key : Callable
+        Function that takes an Index and returns an Index of
+        the same shape. This key is applied to each level
+        separately. The name of the level can be used to
+        distinguish different levels for application.
+    level : list-like, int or str, default None
+        Level or list of levels to apply the key function to.
+        If None, key function is applied to all levels. Other
+        levels are left unchanged.
+
+    Returns
+    -------
+    labels : MultiIndex
+        Resulting MultiIndex with modified levels.
+    """
+
+    if level is not None:
+        if isinstance(level, (str, int)):
+            level_iter = [level]
+        else:
+            level_iter = level
+
+        sort_levels: range | set = {index._get_level_number(lev) for lev in level_iter}
+    else:
+        sort_levels = range(index.nlevels)
+
+    mapped = [
+        ensure_key_mapped(index._get_level_values(level), key)
+        if level in sort_levels
+        else index._get_level_values(level)
+        for level in range(index.nlevels)
+    ]
+
+    return type(index).from_arrays(mapped)
+
+
+def ensure_key_mapped(
+    values: ArrayLike | Index | Series, key: Callable | None, levels=None
+) -> ArrayLike | Index | Series:
+    """
+    Applies a callable key function to the values function and checks
+    that the resulting value has the same shape. Can be called on Index
+    subclasses, Series, DataFrames, or ndarrays.
+
+    Parameters
+    ----------
+    values : Series, DataFrame, Index subclass, or ndarray
+    key : Optional[Callable], key to be called on the values array
+    levels : Optional[List], if values is a MultiIndex, list of levels to
+    apply the key to.
+    """
+    from pandas.core.indexes.api import Index
+
+    if not key:
+        return values
+
+    if isinstance(values, ABCMultiIndex):
+        return _ensure_key_mapped_multiindex(values, key, level=levels)
+
+    result = key(values.copy())
+    if len(result) != len(values):
+        raise ValueError(
+            "User-provided `key` function must not change the shape of the array."
+        )
+
+    try:
+        if isinstance(
+            values, Index
+        ):  # convert to a new Index subclass, not necessarily the same
+            result = Index(result, tupleize_cols=False)
+        else:
+            # try to revert to original type otherwise
+            type_of_values = type(values)
+            #  error: Too many arguments for "ExtensionArray"
+            result = type_of_values(result)  # type: ignore[call-arg]
+    except TypeError as err:
+        raise TypeError(
+            f"User-provided `key` function returned an invalid type {type(result)} \
+            which could not be converted to {type(values)}."
+        ) from err
+
+    return result
+
+
+def get_indexer_dict(
+    label_list: list[np.ndarray], keys: list[Index]
+) -> dict[Hashable, npt.NDArray[np.intp]]:
+    """
+    Returns
+    -------
+    dict:
+        Labels mapped to indexers.
+    """
+    shape = tuple(len(x) for x in keys)
+
+    group_index = get_group_index(label_list, shape, sort=True, xnull=True)
+    if np.all(group_index == -1):
+        # Short-circuit, lib.indices_fast will return the same
+        return {}
+    ngroups = (
+        ((group_index.size and group_index.max()) + 1)
+        if is_int64_overflow_possible(shape)
+        else np.prod(shape, dtype="i8")
+    )
+
+    sorter = get_group_index_sorter(group_index, ngroups)
+
+    sorted_labels = [lab.take(sorter) for lab in label_list]
+    group_index = group_index.take(sorter)
+
+    return lib.indices_fast(sorter, group_index, keys, sorted_labels)
+
+
+# ----------------------------------------------------------------------
+# sorting levels...cleverly?
+
+
+def get_group_index_sorter(
+    group_index: npt.NDArray[np.intp], ngroups: int | None = None
+) -> npt.NDArray[np.intp]:
+    """
+    algos.groupsort_indexer implements `counting sort` and it is at least
+    O(ngroups), where
+        ngroups = prod(shape)
+        shape = map(len, keys)
+    that is, linear in the number of combinations (cartesian product) of unique
+    values of groupby keys. This can be huge when doing multi-key groupby.
+    np.argsort(kind='mergesort') is O(count x log(count)) where count is the
+    length of the data-frame;
+    Both algorithms are `stable` sort and that is necessary for correctness of
+    groupby operations. e.g. consider:
+        df.groupby(key)[col].transform('first')
+
+    Parameters
+    ----------
+    group_index : np.ndarray[np.intp]
+        signed integer dtype
+    ngroups : int or None, default None
+
+    Returns
+    -------
+    np.ndarray[np.intp]
+    """
+    if ngroups is None:
+        ngroups = 1 + group_index.max()
+    count = len(group_index)
+    alpha = 0.0  # taking complexities literally; there may be
+    beta = 1.0  # some room for fine-tuning these parameters
+    do_groupsort = count > 0 and ((alpha + beta * ngroups) < (count * np.log(count)))
+    if do_groupsort:
+        sorter, _ = algos.groupsort_indexer(
+            ensure_platform_int(group_index),
+            ngroups,
+        )
+        # sorter _should_ already be intp, but mypy is not yet able to verify
+    else:
+        sorter = group_index.argsort(kind="mergesort")
+    return ensure_platform_int(sorter)
+
+
+def compress_group_index(
+    group_index: npt.NDArray[np.int64], sort: bool = True
+) -> tuple[npt.NDArray[np.int64], npt.NDArray[np.int64]]:
+    """
+    Group_index is offsets into cartesian product of all possible labels. This
+    space can be huge, so this function compresses it, by computing offsets
+    (comp_ids) into the list of unique labels (obs_group_ids).
+    """
+    if len(group_index) and np.all(group_index[1:] >= group_index[:-1]):
+        # GH 53806: fast path for sorted group_index
+        unique_mask = np.concatenate(
+            [group_index[:1] > -1, group_index[1:] != group_index[:-1]]
+        )
+        comp_ids = unique_mask.cumsum()
+        comp_ids -= 1
+        obs_group_ids = group_index[unique_mask]
+    else:
+        size_hint = len(group_index)
+        table = hashtable.Int64HashTable(size_hint)
+
+        group_index = ensure_int64(group_index)
+
+        # note, group labels come out ascending (ie, 1,2,3 etc)
+        comp_ids, obs_group_ids = table.get_labels_groupby(group_index)
+
+        if sort and len(obs_group_ids) > 0:
+            obs_group_ids, comp_ids = _reorder_by_uniques(obs_group_ids, comp_ids)
+
+    return ensure_int64(comp_ids), ensure_int64(obs_group_ids)
+
+
+def _reorder_by_uniques(
+    uniques: npt.NDArray[np.int64], labels: npt.NDArray[np.intp]
+) -> tuple[npt.NDArray[np.int64], npt.NDArray[np.intp]]:
+    """
+    Parameters
+    ----------
+    uniques : np.ndarray[np.int64]
+    labels : np.ndarray[np.intp]
+
+    Returns
+    -------
+    np.ndarray[np.int64]
+    np.ndarray[np.intp]
+    """
+    # sorter is index where elements ought to go
+    sorter = uniques.argsort()
+
+    # reverse_indexer is where elements came from
+    reverse_indexer = np.empty(len(sorter), dtype=np.intp)
+    reverse_indexer.put(sorter, np.arange(len(sorter)))
+
+    mask = labels < 0
+
+    # move labels to right locations (ie, unsort ascending labels)
+    labels = reverse_indexer.take(labels)
+    np.putmask(labels, mask, -1)
+
+    # sort observed ids
+    uniques = uniques.take(sorter)
+
+    return uniques, labels
diff --git a/pandas/pandas/core/sparse/__init__.py b/pandas/pandas/core/sparse/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/sparse/api.py b/pandas/pandas/core/sparse/api.py
new file mode 100644
index 0000000000000..6650a5c4e90a0
--- /dev/null
+++ b/pandas/pandas/core/sparse/api.py
@@ -0,0 +1,5 @@
+from pandas.core.dtypes.dtypes import SparseDtype
+
+from pandas.core.arrays.sparse import SparseArray
+
+__all__ = ["SparseArray", "SparseDtype"]
diff --git a/pandas/pandas/core/strings/__init__.py b/pandas/pandas/core/strings/__init__.py
new file mode 100644
index 0000000000000..d4ce75f768c5d
--- /dev/null
+++ b/pandas/pandas/core/strings/__init__.py
@@ -0,0 +1,28 @@
+"""
+Implementation of pandas.Series.str and its interface.
+
+* strings.accessor.StringMethods : Accessor for Series.str
+* strings.base.BaseStringArrayMethods: Mixin ABC for EAs to implement str methods
+
+Most methods on the StringMethods accessor follow the pattern:
+
+    1. extract the array from the series (or index)
+    2. Call that array's implementation of the string method
+    3. Wrap the result (in a Series, index, or DataFrame)
+
+Pandas extension arrays implementing string methods should inherit from
+pandas.core.strings.base.BaseStringArrayMethods. This is an ABC defining
+the various string methods. To avoid namespace clashes and pollution,
+these are prefixed with `_str_`. So ``Series.str.upper()`` calls
+``Series.array._str_upper()``. The interface isn't currently public
+to other string extension arrays.
+"""
+# Pandas current implementation is in ObjectStringArrayMixin. This is designed
+# to work on object-dtype ndarrays.
+#
+# BaseStringArrayMethods
+#  - ObjectStringArrayMixin
+#     - StringArray
+#     - NumpyExtensionArray
+#     - Categorical
+#     - ArrowStringArray
diff --git a/pandas/pandas/core/strings/accessor.py b/pandas/pandas/core/strings/accessor.py
new file mode 100644
index 0000000000000..81f7441846589
--- /dev/null
+++ b/pandas/pandas/core/strings/accessor.py
@@ -0,0 +1,3996 @@
+from __future__ import annotations
+
+import codecs
+from functools import wraps
+import re
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs import lib
+from pandas._typing import (
+    AlignJoin,
+    DtypeObj,
+    F,
+    Scalar,
+    npt,
+)
+from pandas.util._decorators import Appender
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_bool_dtype,
+    is_extension_array_dtype,
+    is_integer,
+    is_list_like,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_re,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.arrays import ExtensionArray
+from pandas.core.base import NoNewAttributesMixin
+from pandas.core.construction import extract_array
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterator,
+    )
+
+    from pandas._typing import NpDtype
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+_shared_docs: dict[str, str] = {}
+_cpython_optimized_encoders = (
+    "utf-8",
+    "utf8",
+    "latin-1",
+    "latin1",
+    "iso-8859-1",
+    "mbcs",
+    "ascii",
+)
+_cpython_optimized_decoders = _cpython_optimized_encoders + ("utf-16", "utf-32")
+
+
+def forbid_nonstring_types(
+    forbidden: list[str] | None, name: str | None = None
+) -> Callable[[F], F]:
+    """
+    Decorator to forbid specific types for a method of StringMethods.
+
+    For calling `.str.{method}` on a Series or Index, it is necessary to first
+    initialize the :class:`StringMethods` object, and then call the method.
+    However, different methods allow different input types, and so this can not
+    be checked during :meth:`StringMethods.__init__`, but must be done on a
+    per-method basis. This decorator exists to facilitate this process, and
+    make it explicit which (inferred) types are disallowed by the method.
+
+    :meth:`StringMethods.__init__` allows the *union* of types its different
+    methods allow (after skipping NaNs; see :meth:`StringMethods._validate`),
+    namely: ['string', 'empty', 'bytes', 'mixed', 'mixed-integer'].
+
+    The default string types ['string', 'empty'] are allowed for all methods.
+    For the additional types ['bytes', 'mixed', 'mixed-integer'], each method
+    then needs to forbid the types it is not intended for.
+
+    Parameters
+    ----------
+    forbidden : list-of-str or None
+        List of forbidden non-string types, may be one or more of
+        `['bytes', 'mixed', 'mixed-integer']`.
+    name : str, default None
+        Name of the method to use in the error message. By default, this is
+        None, in which case the name from the method being wrapped will be
+        copied. However, for working with further wrappers (like _pat_wrapper
+        and _noarg_wrapper), it is necessary to specify the name.
+
+    Returns
+    -------
+    func : wrapper
+        The method to which the decorator is applied, with an added check that
+        enforces the inferred type to not be in the list of forbidden types.
+
+    Raises
+    ------
+    TypeError
+        If the inferred type of the underlying data is in `forbidden`.
+    """
+    # deal with None
+    forbidden = [] if forbidden is None else forbidden
+
+    allowed_types = {"string", "empty", "bytes", "mixed", "mixed-integer"} - set(
+        forbidden
+    )
+
+    def _forbid_nonstring_types(func: F) -> F:
+        func_name = func.__name__ if name is None else name
+
+        @wraps(func)
+        def wrapper(self, *args, **kwargs):
+            if self._inferred_dtype not in allowed_types:
+                msg = (
+                    f"Cannot use .str.{func_name} with values of "
+                    f"inferred dtype '{self._inferred_dtype}'."
+                )
+                raise TypeError(msg)
+            return func(self, *args, **kwargs)
+
+        wrapper.__name__ = func_name
+        return cast(F, wrapper)
+
+    return _forbid_nonstring_types
+
+
+def _map_and_wrap(name: str | None, docstring: str | None):
+    @forbid_nonstring_types(["bytes"], name=name)
+    def wrapper(self):
+        result = getattr(self._data.array, f"_str_{name}")()
+        return self._wrap_result(
+            result, returns_string=name not in ("isnumeric", "isdecimal")
+        )
+
+    wrapper.__doc__ = docstring
+    return wrapper
+
+
+class StringMethods(NoNewAttributesMixin):
+    """
+    Vectorized string functions for Series and Index.
+
+    NAs stay NA unless handled otherwise by a particular method.
+    Patterned after Python's string methods, with some inspiration from
+    R's stringr package.
+
+    Parameters
+    ----------
+    data : Series or Index
+        The content of the Series or Index.
+
+    See Also
+    --------
+    Series.str : Vectorized string functions for Series.
+    Index.str : Vectorized string functions for Index.
+
+    Examples
+    --------
+    >>> s = pd.Series(["A_Str_Series"])
+    >>> s
+    0    A_Str_Series
+    dtype: object
+
+    >>> s.str.split("_")
+    0    [A, Str, Series]
+    dtype: object
+
+    >>> s.str.replace("_", "")
+    0    AStrSeries
+    dtype: object
+    """
+
+    # Note: see the docstring in pandas.core.strings.__init__
+    # for an explanation of the implementation.
+    # TODO: Dispatch all the methods
+    # Currently the following are not dispatched to the array
+    # * cat
+    # * extractall
+
+    def __init__(self, data) -> None:
+        from pandas.core.arrays.string_ import StringDtype
+
+        self._inferred_dtype = self._validate(data)
+        self._is_categorical = isinstance(data.dtype, CategoricalDtype)
+        self._is_string = isinstance(data.dtype, StringDtype)
+        self._data = data
+
+        self._index = self._name = None
+        if isinstance(data, ABCSeries):
+            self._index = data.index
+            self._name = data.name
+
+        # ._values.categories works for both Series/Index
+        self._parent = data._values.categories if self._is_categorical else data
+        # save orig to blow up categoricals to the right type
+        self._orig = data
+        self._freeze()
+
+    @staticmethod
+    def _validate(data):
+        """
+        Auxiliary function for StringMethods, infers and checks dtype of data.
+
+        This is a "first line of defence" at the creation of the StringMethods-
+        object, and just checks that the dtype is in the
+        *union* of the allowed types over all string methods below; this
+        restriction is then refined on a per-method basis using the decorator
+        @forbid_nonstring_types (more info in the corresponding docstring).
+
+        This really should exclude all series/index with any non-string values,
+        but that isn't practical for performance reasons until we have a str
+        dtype (GH 9343 / 13877)
+
+        Parameters
+        ----------
+        data : The content of the Series
+
+        Returns
+        -------
+        dtype : inferred dtype of data
+        """
+        if isinstance(data, ABCMultiIndex):
+            raise AttributeError(
+                "Can only use .str accessor with Index, not MultiIndex"
+            )
+
+        # see _libs/lib.pyx for list of inferred types
+        allowed_types = ["string", "empty", "bytes", "mixed", "mixed-integer"]
+
+        data = extract_array(data)
+
+        values = getattr(data, "categories", data)  # categorical / normal
+
+        inferred_dtype = lib.infer_dtype(values, skipna=True)
+
+        if inferred_dtype not in allowed_types:
+            raise AttributeError(
+                f"Can only use .str accessor with string values, not {inferred_dtype}"
+            )
+        return inferred_dtype
+
+    def __getitem__(self, key):
+        result = self._data.array._str_getitem(key)
+        return self._wrap_result(result)
+
+    def __iter__(self) -> Iterator:
+        raise TypeError(f"'{type(self).__name__}' object is not iterable")
+
+    def _wrap_result(
+        self,
+        result,
+        name=None,
+        expand: bool | None = None,
+        fill_value=np.nan,
+        returns_string: bool = True,
+        dtype=None,
+    ):
+        from pandas import (
+            Index,
+            MultiIndex,
+        )
+
+        if not hasattr(result, "ndim") or not hasattr(result, "dtype"):
+            if isinstance(result, ABCDataFrame):
+                result = result.__finalize__(self._orig, name="str")
+            return result
+        assert result.ndim < 3
+
+        # We can be wrapping a string / object / categorical result, in which
+        # case we'll want to return the same dtype as the input.
+        # Or we can be wrapping a numeric output, in which case we don't want
+        # to return a StringArray.
+        # Ideally the array method returns the right array type.
+        if expand is None:
+            # infer from ndim if expand is not specified
+            expand = result.ndim != 1
+        elif expand is True and not isinstance(self._orig, ABCIndex):
+            # required when expand=True is explicitly specified
+            # not needed when inferred
+            if isinstance(result.dtype, ArrowDtype):
+                import pyarrow as pa
+
+                from pandas.compat import pa_version_under11p0
+
+                from pandas.core.arrays.arrow.array import ArrowExtensionArray
+
+                value_lengths = pa.compute.list_value_length(result._pa_array)
+                max_len = pa.compute.max(value_lengths).as_py()
+                min_len = pa.compute.min(value_lengths).as_py()
+                if result._hasna:
+                    # ArrowExtensionArray.fillna doesn't work for list scalars
+                    result = ArrowExtensionArray(
+                        result._pa_array.fill_null([None] * max_len)
+                    )
+                if min_len < max_len:
+                    # append nulls to each scalar list element up to max_len
+                    if not pa_version_under11p0:
+                        result = ArrowExtensionArray(
+                            pa.compute.list_slice(
+                                result._pa_array,
+                                start=0,
+                                stop=max_len,
+                                return_fixed_size_list=True,
+                            )
+                        )
+                    else:
+                        all_null = np.full(max_len, fill_value=None, dtype=object)
+                        values = result.to_numpy()
+                        new_values = []
+                        for row in values:
+                            if len(row) < max_len:
+                                nulls = all_null[: max_len - len(row)]
+                                row = np.append(row, nulls)
+                            new_values.append(row)
+                        pa_type = result._pa_array.type
+                        result = ArrowExtensionArray(pa.array(new_values, type=pa_type))
+                if name is None:
+                    name = range(max_len)
+                result = (
+                    pa.compute.list_flatten(result._pa_array)
+                    .to_numpy()
+                    .reshape(len(result), max_len)
+                )
+                result = {
+                    label: ArrowExtensionArray(pa.array(res))
+                    for label, res in zip(name, result.T)
+                }
+            elif is_object_dtype(result):
+
+                def cons_row(x):
+                    if is_list_like(x):
+                        return x
+                    else:
+                        return [x]
+
+                result = [cons_row(x) for x in result]
+                if result and not self._is_string:
+                    # propagate nan values to match longest sequence (GH 18450)
+                    max_len = max(len(x) for x in result)
+                    result = [
+                        x * max_len if len(x) == 0 or x[0] is np.nan else x
+                        for x in result
+                    ]
+
+        if not isinstance(expand, bool):
+            raise ValueError("expand must be True or False")
+
+        if expand is False:
+            # if expand is False, result should have the same name
+            # as the original otherwise specified
+            if name is None:
+                name = getattr(result, "name", None)
+            if name is None:
+                # do not use logical or, _orig may be a DataFrame
+                # which has "name" column
+                name = self._orig.name
+
+        # Wait until we are sure result is a Series or Index before
+        # checking attributes (GH 12180)
+        if isinstance(self._orig, ABCIndex):
+            # if result is a boolean np.array, return the np.array
+            # instead of wrapping it into a boolean Index (GH 8875)
+            if is_bool_dtype(result):
+                return result
+
+            if expand:
+                result = list(result)
+                out: Index = MultiIndex.from_tuples(result, names=name)
+                if out.nlevels == 1:
+                    # We had all tuples of length-one, which are
+                    # better represented as a regular Index.
+                    out = out.get_level_values(0)
+                return out
+            else:
+                return Index(result, name=name, dtype=dtype)
+        else:
+            index = self._orig.index
+            # This is a mess.
+            _dtype: DtypeObj | str | None = dtype
+            vdtype = getattr(result, "dtype", None)
+            if _dtype is not None:
+                pass
+            elif self._is_string:
+                if is_bool_dtype(vdtype):
+                    _dtype = result.dtype
+                elif returns_string:
+                    _dtype = self._orig.dtype
+                else:
+                    _dtype = vdtype
+            elif vdtype is not None:
+                _dtype = vdtype
+
+            if expand:
+                cons = self._orig._constructor_expanddim
+                result = cons(result, columns=name, index=index, dtype=_dtype)
+            else:
+                # Must be a Series
+                cons = self._orig._constructor
+                result = cons(result, name=name, index=index, dtype=_dtype)
+            result = result.__finalize__(self._orig, method="str")
+            if name is not None and result.ndim == 1:
+                # __finalize__ might copy over the original name, but we may
+                # want the new name (e.g. str.extract).
+                result.name = name
+            return result
+
+    def _get_series_list(self, others):
+        """
+        Auxiliary function for :meth:`str.cat`. Turn potentially mixed input
+        into a list of Series (elements without an index must match the length
+        of the calling Series/Index).
+
+        Parameters
+        ----------
+        others : Series, DataFrame, np.ndarray, list-like or list-like of
+            Objects that are either Series, Index or np.ndarray (1-dim).
+
+        Returns
+        -------
+        list of Series
+            Others transformed into list of Series.
+        """
+        from pandas import (
+            DataFrame,
+            Series,
+        )
+
+        # self._orig is either Series or Index
+        idx = self._orig if isinstance(self._orig, ABCIndex) else self._orig.index
+
+        # Generally speaking, all objects without an index inherit the index
+        # `idx` of the calling Series/Index - i.e. must have matching length.
+        # Objects with an index (i.e. Series/Index/DataFrame) keep their own.
+        if isinstance(others, ABCSeries):
+            return [others]
+        elif isinstance(others, ABCIndex):
+            return [Series(others, index=idx, dtype=others.dtype)]
+        elif isinstance(others, ABCDataFrame):
+            return [others[x] for x in others]
+        elif isinstance(others, np.ndarray) and others.ndim == 2:
+            others = DataFrame(others, index=idx)
+            return [others[x] for x in others]
+        elif is_list_like(others, allow_sets=False):
+            try:
+                others = list(others)  # ensure iterators do not get read twice etc
+            except TypeError:
+                # e.g. ser.str, raise below
+                pass
+            else:
+                # in case of list-like `others`, all elements must be
+                # either Series/Index/np.ndarray (1-dim)...
+                if all(
+                    isinstance(x, (ABCSeries, ABCIndex, ExtensionArray))
+                    or (isinstance(x, np.ndarray) and x.ndim == 1)
+                    for x in others
+                ):
+                    los: list[Series] = []
+                    while others:  # iterate through list and append each element
+                        los = los + self._get_series_list(others.pop(0))
+                    return los
+                # ... or just strings
+                elif all(not is_list_like(x) for x in others):
+                    return [Series(others, index=idx)]
+        raise TypeError(
+            "others must be Series, Index, DataFrame, np.ndarray "
+            "or list-like (either containing only strings or "
+            "containing only objects of type Series/Index/"
+            "np.ndarray[1-dim])"
+        )
+
+    @forbid_nonstring_types(["bytes", "mixed", "mixed-integer"])
+    def cat(
+        self,
+        others=None,
+        sep: str | None = None,
+        na_rep=None,
+        join: AlignJoin = "left",
+    ) -> str | Series | Index:
+        """
+        Concatenate strings in the Series/Index with given separator.
+
+        If `others` is specified, this function concatenates the Series/Index
+        and elements of `others` element-wise.
+        If `others` is not passed, then all values in the Series/Index are
+        concatenated into a single string with a given `sep`.
+
+        Parameters
+        ----------
+        others : Series, Index, DataFrame, np.ndarray or list-like
+            Series, Index, DataFrame, np.ndarray (one- or two-dimensional) and
+            other list-likes of strings must have the same length as the
+            calling Series/Index, with the exception of indexed objects (i.e.
+            Series/Index/DataFrame) if `join` is not None.
+
+            If others is a list-like that contains a combination of Series,
+            Index or np.ndarray (1-dim), then all elements will be unpacked and
+            must satisfy the above criteria individually.
+
+            If others is None, the method returns the concatenation of all
+            strings in the calling Series/Index.
+        sep : str, default ''
+            The separator between the different elements/columns. By default
+            the empty string `''` is used.
+        na_rep : str or None, default None
+            Representation that is inserted for all missing values:
+
+            - If `na_rep` is None, and `others` is None, missing values in the
+              Series/Index are omitted from the result.
+            - If `na_rep` is None, and `others` is not None, a row containing a
+              missing value in any of the columns (before concatenation) will
+              have a missing value in the result.
+        join : {'left', 'right', 'outer', 'inner'}, default 'left'
+            Determines the join-style between the calling Series/Index and any
+            Series/Index/DataFrame in `others` (objects without an index need
+            to match the length of the calling Series/Index). To disable
+            alignment, use `.values` on any Series/Index/DataFrame in `others`.
+
+        Returns
+        -------
+        str, Series or Index
+            If `others` is None, `str` is returned, otherwise a `Series/Index`
+            (same type as caller) of objects is returned.
+
+        See Also
+        --------
+        split : Split each string in the Series/Index.
+        join : Join lists contained as elements in the Series/Index.
+
+        Examples
+        --------
+        When not passing `others`, all values are concatenated into a single
+        string:
+
+        >>> s = pd.Series(["a", "b", np.nan, "d"])
+        >>> s.str.cat(sep=" ")
+        'a b d'
+
+        By default, NA values in the Series are ignored. Using `na_rep`, they
+        can be given a representation:
+
+        >>> s.str.cat(sep=" ", na_rep="?")
+        'a b ? d'
+
+        If `others` is specified, corresponding values are concatenated with
+        the separator. Result will be a Series of strings.
+
+        >>> s.str.cat(["A", "B", "C", "D"], sep=",")
+        0    a,A
+        1    b,B
+        2    NaN
+        3    d,D
+        dtype: object
+
+        Missing values will remain missing in the result, but can again be
+        represented using `na_rep`
+
+        >>> s.str.cat(["A", "B", "C", "D"], sep=",", na_rep="-")
+        0    a,A
+        1    b,B
+        2    -,C
+        3    d,D
+        dtype: object
+
+        If `sep` is not specified, the values are concatenated without
+        separation.
+
+        >>> s.str.cat(["A", "B", "C", "D"], na_rep="-")
+        0    aA
+        1    bB
+        2    -C
+        3    dD
+        dtype: object
+
+        Series with different indexes can be aligned before concatenation. The
+        `join`-keyword works as in other methods.
+
+        >>> t = pd.Series(["d", "a", "e", "c"], index=[3, 0, 4, 2])
+        >>> s.str.cat(t, join="left", na_rep="-")
+        0    aa
+        1    b-
+        2    -c
+        3    dd
+        dtype: object
+        >>>
+        >>> s.str.cat(t, join="outer", na_rep="-")
+        0    aa
+        1    b-
+        2    -c
+        3    dd
+        4    -e
+        dtype: object
+        >>>
+        >>> s.str.cat(t, join="inner", na_rep="-")
+        0    aa
+        2    -c
+        3    dd
+        dtype: object
+        >>>
+        >>> s.str.cat(t, join="right", na_rep="-")
+        3    dd
+        0    aa
+        4    -e
+        2    -c
+        dtype: object
+
+        For more examples, see :ref:`here <text.concatenate>`.
+        """
+        # TODO: dispatch
+        from pandas import (
+            Index,
+            Series,
+            concat,
+        )
+
+        if isinstance(others, str):
+            raise ValueError("Did you mean to supply a `sep` keyword?")
+        if sep is None:
+            sep = ""
+
+        if isinstance(self._orig, ABCIndex):
+            data = Series(self._orig, index=self._orig, dtype=self._orig.dtype)
+        else:  # Series
+            data = self._orig
+
+        # concatenate Series/Index with itself if no "others"
+        if others is None:
+            # error: Incompatible types in assignment (expression has type
+            # "ndarray", variable has type "Series")
+            data = ensure_object(data)  # type: ignore[assignment]
+            na_mask = isna(data)
+            if na_rep is None and na_mask.any():
+                return sep.join(data[~na_mask])
+            elif na_rep is not None and na_mask.any():
+                return sep.join(np.where(na_mask, na_rep, data))
+            else:
+                return sep.join(data)
+
+        try:
+            # turn anything in "others" into lists of Series
+            others = self._get_series_list(others)
+        except ValueError as err:  # do not catch TypeError raised by _get_series_list
+            raise ValueError(
+                "If `others` contains arrays or lists (or other "
+                "list-likes without an index), these must all be "
+                "of the same length as the calling Series/Index."
+            ) from err
+
+        # align if required
+        if any(not data.index.equals(x.index) for x in others):
+            # Need to add keys for uniqueness in case of duplicate columns
+            others = concat(
+                others,
+                axis=1,
+                join=(join if join == "inner" else "outer"),
+                keys=range(len(others)),
+                sort=False,
+            )
+            data, others = data.align(others, join=join)
+            others = [others[x] for x in others]  # again list of Series
+
+        all_cols = [ensure_object(x) for x in [data] + others]
+        na_masks = np.array([isna(x) for x in all_cols])
+        union_mask = np.logical_or.reduce(na_masks, axis=0)
+
+        if na_rep is None and union_mask.any():
+            # no na_rep means NaNs for all rows where any column has a NaN
+            # only necessary if there are actually any NaNs
+            result = np.empty(len(data), dtype=object)
+            np.putmask(result, union_mask, np.nan)
+
+            not_masked = ~union_mask
+            result[not_masked] = cat_safe([x[not_masked] for x in all_cols], sep)
+        elif na_rep is not None and union_mask.any():
+            # fill NaNs with na_rep in case there are actually any NaNs
+            all_cols = [
+                np.where(nm, na_rep, col) for nm, col in zip(na_masks, all_cols)
+            ]
+            result = cat_safe(all_cols, sep)
+        else:
+            # no NaNs - can just concatenate
+            result = cat_safe(all_cols, sep)
+
+        out: Index | Series
+        if isinstance(self._orig.dtype, CategoricalDtype):
+            # We need to infer the new categories.
+            dtype = self._orig.dtype.categories.dtype
+        else:
+            dtype = self._orig.dtype
+        if isinstance(self._orig, ABCIndex):
+            # add dtype for case that result is all-NA
+            if isna(result).all():
+                dtype = object  # type: ignore[assignment]
+
+            out = Index(result, dtype=dtype, name=self._orig.name)
+        else:  # Series
+            res_ser = Series(
+                result, dtype=dtype, index=data.index, name=self._orig.name, copy=False
+            )
+            out = res_ser.__finalize__(self._orig, method="str_cat")
+        return out
+
+    _shared_docs["str_split"] = r"""
+    Split strings around given separator/delimiter.
+
+    Splits the string in the Series/Index from the %(side)s,
+    at the specified delimiter string.
+
+    Parameters
+    ----------
+    pat : str%(pat_regex)s, optional
+        %(pat_description)s.
+        If not specified, split on whitespace.
+    n : int, default -1 (all)
+        Limit number of splits in output.
+        ``None``, 0 and -1 will be interpreted as return all splits.
+    expand : bool, default False
+        Expand the split strings into separate columns.
+
+        - If ``True``, return DataFrame/MultiIndex expanding dimensionality.
+        - If ``False``, return Series/Index, containing lists of strings.
+    %(regex_argument)s
+    Returns
+    -------
+    Series, Index, DataFrame or MultiIndex
+        Type matches caller unless ``expand=True`` (see Notes).
+    %(raises_split)s
+    See Also
+    --------
+    Series.str.split : Split strings around given separator/delimiter.
+    Series.str.rsplit : Splits string around given separator/delimiter,
+        starting from the right.
+    Series.str.join : Join lists contained as elements in the Series/Index
+        with passed delimiter.
+    str.split : Standard library version for split.
+    str.rsplit : Standard library version for rsplit.
+
+    Notes
+    -----
+    The handling of the `n` keyword depends on the number of found splits:
+
+    - If found splits > `n`,  make first `n` splits only
+    - If found splits <= `n`, make all splits
+    - If for a certain row the number of found splits < `n`,
+      append `None` for padding up to `n` if ``expand=True``
+
+    If using ``expand=True``, Series and Index callers return DataFrame and
+    MultiIndex objects, respectively.
+    %(regex_pat_note)s
+    Examples
+    --------
+    >>> s = pd.Series(
+    ...     [
+    ...         "this is a regular sentence",
+    ...         "https://docs.python.org/3/tutorial/index.html",
+    ...         np.nan
+    ...     ]
+    ... )
+    >>> s
+    0                       this is a regular sentence
+    1    https://docs.python.org/3/tutorial/index.html
+    2                                              NaN
+    dtype: object
+
+    In the default setting, the string is split by whitespace.
+
+    >>> s.str.split()
+    0                   [this, is, a, regular, sentence]
+    1    [https://docs.python.org/3/tutorial/index.html]
+    2                                                NaN
+    dtype: object
+
+    Without the `n` parameter, the outputs of `rsplit` and `split`
+    are identical.
+
+    >>> s.str.rsplit()
+    0                   [this, is, a, regular, sentence]
+    1    [https://docs.python.org/3/tutorial/index.html]
+    2                                                NaN
+    dtype: object
+
+    The `n` parameter can be used to limit the number of splits on the
+    delimiter. The outputs of `split` and `rsplit` are different.
+
+    >>> s.str.split(n=2)
+    0                     [this, is, a regular sentence]
+    1    [https://docs.python.org/3/tutorial/index.html]
+    2                                                NaN
+    dtype: object
+
+    >>> s.str.rsplit(n=2)
+    0                     [this is a, regular, sentence]
+    1    [https://docs.python.org/3/tutorial/index.html]
+    2                                                NaN
+    dtype: object
+
+    The `pat` parameter can be used to split by other characters.
+
+    >>> s.str.split(pat="/")
+    0                         [this is a regular sentence]
+    1    [https:, , docs.python.org, 3, tutorial, index...
+    2                                                  NaN
+    dtype: object
+
+    When using ``expand=True``, the split elements will expand out into
+    separate columns. If NaN is present, it is propagated throughout
+    the columns during the split.
+
+    >>> s.str.split(expand=True)
+                                                   0     1     2        3         4
+    0                                           this    is     a  regular  sentence
+    1  https://docs.python.org/3/tutorial/index.html  None  None     None      None
+    2                                            NaN   NaN   NaN      NaN       NaN
+
+    For slightly more complex use cases like splitting the html document name
+    from a url, a combination of parameter settings can be used.
+
+    >>> s.str.rsplit("/", n=1, expand=True)
+                                        0           1
+    0          this is a regular sentence        None
+    1  https://docs.python.org/3/tutorial  index.html
+    2                                 NaN         NaN
+    %(regex_examples)s"""
+
+    @Appender(
+        _shared_docs["str_split"]
+        % {
+            "side": "beginning",
+            "pat_regex": " or compiled regex",
+            "pat_description": "String or regular expression to split on",
+            "regex_argument": """
+    regex : bool, default None
+        Determines if the passed-in pattern is a regular expression:
+
+        - If ``True``, assumes the passed-in pattern is a regular expression
+        - If ``False``, treats the pattern as a literal string.
+        - If ``None`` and `pat` length is 1, treats `pat` as a literal string.
+        - If ``None`` and `pat` length is not 1, treats `pat` as a regular expression.
+        - Cannot be set to False if `pat` is a compiled regex
+
+        .. versionadded:: 1.4.0
+         """,
+            "raises_split": """
+                      Raises
+                      ------
+                      ValueError
+                          * if `regex` is False and `pat` is a compiled regex
+                      """,
+            "regex_pat_note": """
+    Use of `regex =False` with a `pat` as a compiled regex will raise an error.
+            """,
+            "method": "split",
+            "regex_examples": r"""
+    Remember to escape special characters when explicitly using regular expressions.
+
+    >>> s = pd.Series(["foo and bar plus baz"])
+    >>> s.str.split(r"and|plus", expand=True)
+        0   1   2
+    0 foo bar baz
+
+    Regular expressions can be used to handle urls or file names.
+    When `pat` is a string and ``regex=None`` (the default), the given `pat` is compiled
+    as a regex only if ``len(pat) != 1``.
+
+    >>> s = pd.Series(['foojpgbar.jpg'])
+    >>> s.str.split(r".", expand=True)
+               0    1
+    0  foojpgbar  jpg
+
+    >>> s.str.split(r"\.jpg", expand=True)
+               0 1
+    0  foojpgbar
+
+    When ``regex=True``, `pat` is interpreted as a regex
+
+    >>> s.str.split(r"\.jpg", regex=True, expand=True)
+               0 1
+    0  foojpgbar
+
+    A compiled regex can be passed as `pat`
+
+    >>> import re
+    >>> s.str.split(re.compile(r"\.jpg"), expand=True)
+               0 1
+    0  foojpgbar
+
+    When ``regex=False``, `pat` is interpreted as the string itself
+
+    >>> s.str.split(r"\.jpg", regex=False, expand=True)
+                   0
+    0  foojpgbar.jpg
+    """,
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def split(
+        self,
+        pat: str | re.Pattern | None = None,
+        *,
+        n=-1,
+        expand: bool = False,
+        regex: bool | None = None,
+    ):
+        if regex is False and is_re(pat):
+            raise ValueError(
+                "Cannot use a compiled regex as replacement pattern with regex=False"
+            )
+        if is_re(pat):
+            regex = True
+        result = self._data.array._str_split(pat, n, expand, regex)
+        if self._data.dtype == "category":
+            dtype = self._data.dtype.categories.dtype
+        else:
+            dtype = object if self._data.dtype == object else None
+        return self._wrap_result(
+            result, expand=expand, returns_string=expand, dtype=dtype
+        )
+
+    @Appender(
+        _shared_docs["str_split"]
+        % {
+            "side": "end",
+            "pat_regex": "",
+            "pat_description": "String to split on",
+            "regex_argument": "",
+            "raises_split": "",
+            "regex_pat_note": "",
+            "method": "rsplit",
+            "regex_examples": "",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def rsplit(self, pat=None, *, n=-1, expand: bool = False):
+        result = self._data.array._str_rsplit(pat, n=n)
+        dtype = object if self._data.dtype == object else None
+        return self._wrap_result(
+            result, expand=expand, returns_string=expand, dtype=dtype
+        )
+
+    _shared_docs["str_partition"] = """
+    Split the string at the %(side)s occurrence of `sep`.
+
+    This method splits the string at the %(side)s occurrence of `sep`,
+    and returns 3 elements containing the part before the separator,
+    the separator itself, and the part after the separator.
+    If the separator is not found, return %(return)s.
+
+    Parameters
+    ----------
+    sep : str, default whitespace
+        String to split on.
+    expand : bool, default True
+        If True, return DataFrame/MultiIndex expanding dimensionality.
+        If False, return Series/Index.
+
+    Returns
+    -------
+    DataFrame/MultiIndex or Series/Index of objects
+        Returns appropriate type based on `expand` parameter with strings
+        split based on the `sep` parameter.
+
+    See Also
+    --------
+    %(also)s
+    Series.str.split : Split strings around given separators.
+    str.partition : Standard library version.
+
+    Examples
+    --------
+
+    >>> s = pd.Series(['Linda van der Berg', 'George Pitt-Rivers'])
+    >>> s
+    0    Linda van der Berg
+    1    George Pitt-Rivers
+    dtype: object
+
+    >>> s.str.partition()
+            0  1             2
+    0   Linda     van der Berg
+    1  George      Pitt-Rivers
+
+    To partition by the last space instead of the first one:
+
+    >>> s.str.rpartition()
+                   0  1            2
+    0  Linda van der            Berg
+    1         George     Pitt-Rivers
+
+    To partition by something different than a space:
+
+    >>> s.str.partition('-')
+                        0  1       2
+    0  Linda van der Berg
+    1         George Pitt  -  Rivers
+
+    To return a Series containing tuples instead of a DataFrame:
+
+    >>> s.str.partition('-', expand=False)
+    0    (Linda van der Berg, , )
+    1    (George Pitt, -, Rivers)
+    dtype: object
+
+    Also available on indices:
+
+    >>> idx = pd.Index(['X 123', 'Y 999'])
+    >>> idx
+    Index(['X 123', 'Y 999'], dtype='object')
+
+    Which will create a MultiIndex:
+
+    >>> idx.str.partition()
+    MultiIndex([('X', ' ', '123'),
+                ('Y', ' ', '999')],
+               )
+
+    Or an index with tuples with ``expand=False``:
+
+    >>> idx.str.partition(expand=False)
+    Index([('X', ' ', '123'), ('Y', ' ', '999')], dtype='object')
+    """
+
+    @Appender(
+        _shared_docs["str_partition"]
+        % {
+            "side": "first",
+            "return": "3 elements containing the string itself, followed by two "
+            "empty strings",
+            "also": "rpartition : Split the string at the last occurrence of `sep`.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def partition(self, sep: str = " ", expand: bool = True):
+        result = self._data.array._str_partition(sep, expand)
+        if self._data.dtype == "category":
+            dtype = self._data.dtype.categories.dtype
+        else:
+            dtype = object if self._data.dtype == object else None
+        return self._wrap_result(
+            result, expand=expand, returns_string=expand, dtype=dtype
+        )
+
+    @Appender(
+        _shared_docs["str_partition"]
+        % {
+            "side": "last",
+            "return": "3 elements containing two empty strings, followed by the "
+            "string itself",
+            "also": "partition : Split the string at the first occurrence of `sep`.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def rpartition(self, sep: str = " ", expand: bool = True):
+        result = self._data.array._str_rpartition(sep, expand)
+        if self._data.dtype == "category":
+            dtype = self._data.dtype.categories.dtype
+        else:
+            dtype = object if self._data.dtype == object else None
+        return self._wrap_result(
+            result, expand=expand, returns_string=expand, dtype=dtype
+        )
+
+    def get(self, i):
+        """
+        Extract element from each component at specified position or with specified key.
+
+        Extract element from lists, tuples, dict, or strings in each element in the
+        Series/Index.
+
+        Parameters
+        ----------
+        i : int or hashable dict label
+            Position or key of element to extract.
+
+        Returns
+        -------
+        Series or Index
+            Series or Index where each value is the extracted element from
+            the corresponding input component.
+
+        See Also
+        --------
+        Series.str.extract : Extract capture groups in the regex as columns
+            in a DataFrame.
+
+        Examples
+        --------
+        >>> s = pd.Series(
+        ...     [
+        ...         "String",
+        ...         (1, 2, 3),
+        ...         ["a", "b", "c"],
+        ...         123,
+        ...         -456,
+        ...         {1: "Hello", "2": "World"},
+        ...     ]
+        ... )
+        >>> s
+        0                        String
+        1                     (1, 2, 3)
+        2                     [a, b, c]
+        3                           123
+        4                          -456
+        5    {1: 'Hello', '2': 'World'}
+        dtype: object
+
+        >>> s.str.get(1)
+        0        t
+        1        2
+        2        b
+        3      NaN
+        4      NaN
+        5    Hello
+        dtype: object
+
+        >>> s.str.get(-1)
+        0      g
+        1      3
+        2      c
+        3    NaN
+        4    NaN
+        5    None
+        dtype: object
+
+        Return element with given key
+
+        >>> s = pd.Series(
+        ...     [
+        ...         {"name": "Hello", "value": "World"},
+        ...         {"name": "Goodbye", "value": "Planet"},
+        ...     ]
+        ... )
+        >>> s.str.get("name")
+        0      Hello
+        1    Goodbye
+        dtype: object
+        """
+        result = self._data.array._str_get(i)
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def join(self, sep: str):
+        """
+        Join lists contained as elements in the Series/Index with passed delimiter.
+
+        If the elements of a Series are lists themselves, join the content of these
+        lists using the delimiter passed to the function.
+        This function is an equivalent to :meth:`str.join`.
+
+        Parameters
+        ----------
+        sep : str
+            Delimiter to use between list entries.
+
+        Returns
+        -------
+        Series/Index: object
+            The list entries concatenated by intervening occurrences of the
+            delimiter.
+
+        Raises
+        ------
+        AttributeError
+            If the supplied Series contains neither strings nor lists.
+
+        See Also
+        --------
+        str.join : Standard library version of this method.
+        Series.str.split : Split strings around given separator/delimiter.
+
+        Notes
+        -----
+        If any of the list items is not a string object, the result of the join
+        will be `NaN`.
+
+        Examples
+        --------
+        Example with a list that contains non-string elements.
+
+        >>> s = pd.Series(
+        ...     [
+        ...         ["lion", "elephant", "zebra"],
+        ...         [1.1, 2.2, 3.3],
+        ...         ["cat", np.nan, "dog"],
+        ...         ["cow", 4.5, "goat"],
+        ...         ["duck", ["swan", "fish"], "guppy"],
+        ...     ]
+        ... )
+        >>> s
+        0        [lion, elephant, zebra]
+        1                [1.1, 2.2, 3.3]
+        2                [cat, nan, dog]
+        3               [cow, 4.5, goat]
+        4    [duck, [swan, fish], guppy]
+        dtype: object
+
+        Join all lists using a '-'. The lists containing object(s) of types other
+        than str will produce a NaN.
+
+        >>> s.str.join("-")
+        0    lion-elephant-zebra
+        1                    NaN
+        2                    NaN
+        3                    NaN
+        4                    NaN
+        dtype: object
+        """
+        result = self._data.array._str_join(sep)
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def contains(
+        self,
+        pat,
+        case: bool = True,
+        flags: int = 0,
+        na=lib.no_default,
+        regex: bool = True,
+    ):
+        r"""
+        Test if pattern or regex is contained within a string of a Series or Index.
+
+        Return boolean Series or Index based on whether a given pattern or regex is
+        contained within a string of a Series or Index.
+
+        Parameters
+        ----------
+        pat : str
+            Character sequence or regular expression.
+        case : bool, default True
+            If True, case sensitive.
+        flags : int, default 0 (no flags)
+            Flags to pass through to the re module, e.g. re.IGNORECASE.
+        na : scalar, optional
+            Fill value for missing values. The default depends on dtype of the
+            array. For object-dtype, ``numpy.nan`` is used. For the nullable
+            ``StringDtype``, ``pandas.NA`` is used. For the ``"str"`` dtype,
+            ``False`` is used.
+        regex : bool, default True
+            If True, assumes the pat is a regular expression.
+
+            If False, treats the pat as a literal string.
+
+        Returns
+        -------
+        Series or Index of boolean values
+            A Series or Index of boolean values indicating whether the
+            given pattern is contained within the string of each element
+            of the Series or Index.
+
+        See Also
+        --------
+        match : Analogous, but stricter, relying on re.match instead of re.search.
+        Series.str.startswith : Test if the start of each string element matches a
+            pattern.
+        Series.str.endswith : Same as startswith, but tests the end of string.
+
+        Examples
+        --------
+        Returning a Series of booleans using only a literal pattern.
+
+        >>> s1 = pd.Series(["Mouse", "dog", "house and parrot", "23", np.nan])
+        >>> s1.str.contains("og", regex=False)
+        0    False
+        1     True
+        2    False
+        3    False
+        4      NaN
+        dtype: object
+
+        Returning an Index of booleans using only a literal pattern.
+
+        >>> ind = pd.Index(["Mouse", "dog", "house and parrot", "23.0", np.nan])
+        >>> ind.str.contains("23", regex=False)
+        Index([False, False, False, True, nan], dtype='object')
+
+        Specifying case sensitivity using `case`.
+
+        >>> s1.str.contains("oG", case=True, regex=True)
+        0    False
+        1    False
+        2    False
+        3    False
+        4      NaN
+        dtype: object
+
+        Specifying `na` to be `False` instead of `NaN` replaces NaN values
+        with `False`. If Series or Index does not contain NaN values
+        the resultant dtype will be `bool`, otherwise, an `object` dtype.
+
+        >>> s1.str.contains("og", na=False, regex=True)
+        0    False
+        1     True
+        2    False
+        3    False
+        4    False
+        dtype: bool
+
+        Returning 'house' or 'dog' when either expression occurs in a string.
+
+        >>> s1.str.contains("house|dog", regex=True)
+        0    False
+        1     True
+        2     True
+        3    False
+        4      NaN
+        dtype: object
+
+        Ignoring case sensitivity using `flags` with regex.
+
+        >>> import re
+        >>> s1.str.contains("PARROT", flags=re.IGNORECASE, regex=True)
+        0    False
+        1    False
+        2     True
+        3    False
+        4      NaN
+        dtype: object
+
+        Returning any digit using regular expression.
+
+        >>> s1.str.contains("\\d", regex=True)
+        0    False
+        1    False
+        2    False
+        3     True
+        4      NaN
+        dtype: object
+
+        Ensure `pat` is a not a literal pattern when `regex` is set to True.
+        Note in the following example one might expect only `s2[1]` and `s2[3]` to
+        return `True`. However, '.0' as a regex matches any character
+        followed by a 0.
+
+        >>> s2 = pd.Series(["40", "40.0", "41", "41.0", "35"])
+        >>> s2.str.contains(".0", regex=True)
+        0     True
+        1     True
+        2    False
+        3     True
+        4    False
+        dtype: bool
+        """
+        if regex and re.compile(pat).groups:
+            warnings.warn(
+                "This pattern is interpreted as a regular expression, and has "
+                "match groups. To actually get the groups, use str.extract.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        result = self._data.array._str_contains(pat, case, flags, na, regex)
+        return self._wrap_result(result, fill_value=na, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def match(self, pat: str, case: bool = True, flags: int = 0, na=lib.no_default):
+        """
+        Determine if each string starts with a match of a regular expression.
+
+        Determines whether each string in the Series or Index starts with a
+        match to a specified regular expression. This function is especially
+        useful for validating prefixes, such as ensuring that codes, tags, or
+        identifiers begin with a specific pattern.
+
+        Parameters
+        ----------
+        pat : str
+            Character sequence.
+        case : bool, default True
+            If True, case sensitive.
+        flags : int, default 0 (no flags)
+            Regex module flags, e.g. re.IGNORECASE.
+        na : scalar, optional
+            Fill value for missing values. The default depends on dtype of the
+            array. For object-dtype, ``numpy.nan`` is used. For the nullable
+            ``StringDtype``, ``pandas.NA`` is used. For the ``"str"`` dtype,
+            ``False`` is used.
+
+        Returns
+        -------
+        Series/Index/array of boolean values
+            A Series, Index, or array of boolean values indicating whether the start
+            of each string matches the pattern. The result will be of the same type
+            as the input.
+
+        See Also
+        --------
+        fullmatch : Stricter matching that requires the entire string to match.
+        contains : Analogous, but less strict, relying on re.search instead of
+            re.match.
+        extract : Extract matched groups.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["horse", "eagle", "donkey"])
+        >>> ser.str.match("e")
+        0   False
+        1   True
+        2   False
+        dtype: bool
+        """
+        result = self._data.array._str_match(pat, case=case, flags=flags, na=na)
+        return self._wrap_result(result, fill_value=na, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def fullmatch(self, pat, case: bool = True, flags: int = 0, na=lib.no_default):
+        """
+        Determine if each string entirely matches a regular expression.
+
+        Checks if each string in the Series or Index fully matches the
+        specified regular expression pattern. This function is useful when the
+        requirement is for an entire string to conform to a pattern, such as
+        validating formats like phone numbers or email addresses.
+
+        Parameters
+        ----------
+        pat : str
+            Character sequence or regular expression.
+        case : bool, default True
+            If True, case sensitive.
+        flags : int, default 0 (no flags)
+            Regex module flags, e.g. re.IGNORECASE.
+        na : scalar, optional
+            Fill value for missing values. The default depends on dtype of the
+            array. For object-dtype, ``numpy.nan`` is used. For the nullable
+            ``StringDtype``, ``pandas.NA`` is used. For the ``"str"`` dtype,
+            ``False`` is used.
+
+        Returns
+        -------
+        Series/Index/array of boolean values
+            The function returns a Series, Index, or array of boolean values,
+            where True indicates that the entire string matches the regular
+            expression pattern and False indicates that it does not.
+
+        See Also
+        --------
+        match : Similar, but also returns `True` when only a *prefix* of the string
+            matches the regular expression.
+        extract : Extract matched groups.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["cat", "duck", "dove"])
+        >>> ser.str.fullmatch(r"d.+")
+        0   False
+        1    True
+        2    True
+        dtype: bool
+        """
+        result = self._data.array._str_fullmatch(pat, case=case, flags=flags, na=na)
+        return self._wrap_result(result, fill_value=na, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def replace(
+        self,
+        pat: str | re.Pattern | dict,
+        repl: str | Callable | None = None,
+        n: int = -1,
+        case: bool | None = None,
+        flags: int = 0,
+        regex: bool = False,
+    ):
+        r"""
+        Replace each occurrence of pattern/regex in the Series/Index.
+
+        Equivalent to :meth:`str.replace` or :func:`re.sub`, depending on
+        the regex value.
+
+        Parameters
+        ----------
+        pat : str, compiled regex, or a dict
+            String can be a character sequence or regular expression.
+            Dictionary contains <key : value> pairs of strings to be replaced
+            along with the updated value.
+        repl : str or callable
+            Replacement string or a callable. The callable is passed the regex
+            match object and must return a replacement string to be used.
+            Must have a value of None if `pat` is a dict
+            See :func:`re.sub`.
+        n : int, default -1 (all)
+            Number of replacements to make from start.
+        case : bool, default None
+            Determines if replace is case sensitive:
+
+            - If True, case sensitive (the default if `pat` is a string)
+            - Set to False for case insensitive
+            - Cannot be set if `pat` is a compiled regex.
+
+        flags : int, default 0 (no flags)
+            Regex module flags, e.g. re.IGNORECASE. Cannot be set if `pat` is a compiled
+            regex.
+        regex : bool, default False
+            Determines if the passed-in pattern is a regular expression:
+
+            - If True, assumes the passed-in pattern is a regular expression.
+            - If False, treats the pattern as a literal string
+            - Cannot be set to False if `pat` is a compiled regex or `repl` is
+              a callable.
+
+        Returns
+        -------
+        Series or Index of object
+            A copy of the object with all matching occurrences of `pat` replaced by
+            `repl`.
+
+        Raises
+        ------
+        ValueError
+            * if `regex` is False and `repl` is a callable or `pat` is a compiled
+              regex
+            * if `pat` is a compiled regex and `case` or `flags` is set
+            * if `pat` is a dictionary and `repl` is not None.
+
+        See Also
+        --------
+        Series.str.replace : Method to replace occurrences of a substring with another
+            substring.
+        Series.str.extract : Extract substrings using a regular expression.
+        Series.str.findall : Find all occurrences of a pattern or regex in each string.
+        Series.str.split : Split each string by a specified delimiter or pattern.
+
+        Notes
+        -----
+        When `pat` is a compiled regex, all flags should be included in the
+        compiled regex. Use of `case`, `flags`, or `regex=False` with a compiled
+        regex will raise an error.
+
+        Examples
+        --------
+        When `pat` is a dictionary, every key in `pat` is replaced
+        with its corresponding value:
+
+        >>> pd.Series(["A", "B", np.nan]).str.replace(pat={"A": "a", "B": "b"})
+        0    a
+        1    b
+        2    NaN
+        dtype: object
+
+        When `pat` is a string and `regex` is True, the given `pat`
+        is compiled as a regex. When `repl` is a string, it replaces matching
+        regex patterns as with :meth:`re.sub`. NaN value(s) in the Series are
+        left as is:
+
+        >>> pd.Series(["foo", "fuz", np.nan]).str.replace("f.", "ba", regex=True)
+        0    bao
+        1    baz
+        2    NaN
+        dtype: object
+
+        When `pat` is a string and `regex` is False, every `pat` is replaced with
+        `repl` as with :meth:`str.replace`:
+
+        >>> pd.Series(["f.o", "fuz", np.nan]).str.replace("f.", "ba", regex=False)
+        0    bao
+        1    fuz
+        2    NaN
+        dtype: object
+
+        When `repl` is a callable, it is called on every `pat` using
+        :func:`re.sub`. The callable should expect one positional argument
+        (a regex object) and return a string.
+
+        To get the idea:
+
+        >>> pd.Series(["foo", "fuz", np.nan]).str.replace("f", repr, regex=True)
+        0    <re.Match object; span=(0, 1), match='f'>oo
+        1    <re.Match object; span=(0, 1), match='f'>uz
+        2                                            NaN
+        dtype: object
+
+        Reverse every lowercase alphabetic word:
+
+        >>> repl = lambda m: m.group(0)[::-1]
+        >>> ser = pd.Series(["foo 123", "bar baz", np.nan])
+        >>> ser.str.replace(r"[a-z]+", repl, regex=True)
+        0    oof 123
+        1    rab zab
+        2        NaN
+        dtype: object
+
+        Using regex groups (extract second group and swap case):
+
+        >>> pat = r"(?P<one>\w+) (?P<two>\w+) (?P<three>\w+)"
+        >>> repl = lambda m: m.group("two").swapcase()
+        >>> ser = pd.Series(["One Two Three", "Foo Bar Baz"])
+        >>> ser.str.replace(pat, repl, regex=True)
+        0    tWO
+        1    bAR
+        dtype: object
+
+        Using a compiled regex with flags
+
+        >>> import re
+        >>> regex_pat = re.compile(r"FUZ", flags=re.IGNORECASE)
+        >>> pd.Series(["foo", "fuz", np.nan]).str.replace(regex_pat, "bar", regex=True)
+        0    foo
+        1    bar
+        2    NaN
+        dtype: object
+        """
+        if isinstance(pat, dict) and repl is not None:
+            raise ValueError("repl cannot be used when pat is a dictionary")
+
+        # Check whether repl is valid (GH 13438, GH 15055)
+        if not isinstance(pat, dict) and not (isinstance(repl, str) or callable(repl)):
+            raise TypeError("repl must be a string or callable")
+
+        is_compiled_re = is_re(pat)
+        if regex or regex is None:
+            if is_compiled_re and (case is not None or flags != 0):
+                raise ValueError(
+                    "case and flags cannot be set when pat is a compiled regex"
+                )
+
+        elif is_compiled_re:
+            raise ValueError(
+                "Cannot use a compiled regex as replacement pattern with regex=False"
+            )
+        elif callable(repl):
+            raise ValueError("Cannot use a callable replacement when regex=False")
+
+        if case is None:
+            case = True
+
+        res_output = self._data
+        if not isinstance(pat, dict):
+            pat = {pat: repl}
+
+        for key, value in pat.items():
+            result = res_output.array._str_replace(
+                key, value, n=n, case=case, flags=flags, regex=regex
+            )
+            res_output = self._wrap_result(result)
+
+        return res_output
+
+    @forbid_nonstring_types(["bytes"])
+    def repeat(self, repeats):
+        """
+        Duplicate each string in the Series or Index.
+
+        Duplicates each string in the Series or Index, either by applying the
+        same repeat count to all elements or by using different repeat values
+        for each element.
+
+        Parameters
+        ----------
+        repeats : int or sequence of int
+            Same value for all (int) or different value per (sequence).
+
+        Returns
+        -------
+        Series or pandas.Index
+            Series or Index of repeated string objects specified by
+            input parameter repeats.
+
+        See Also
+        --------
+        Series.str.lower : Convert all characters in each string to lowercase.
+        Series.str.upper : Convert all characters in each string to uppercase.
+        Series.str.title : Convert each string to title case (capitalizing the first
+            letter of each word).
+        Series.str.strip : Remove leading and trailing whitespace from each string.
+        Series.str.replace : Replace occurrences of a substring with another substring
+            in each string.
+        Series.str.ljust : Left-justify each string in the Series/Index by padding with
+            a specified character.
+        Series.str.rjust : Right-justify each string in the Series/Index by padding with
+            a specified character.
+
+        Examples
+        --------
+        >>> s = pd.Series(["a", "b", "c"])
+        >>> s
+        0    a
+        1    b
+        2    c
+        dtype: object
+
+        Single int repeats string in Series
+
+        >>> s.str.repeat(repeats=2)
+        0    aa
+        1    bb
+        2    cc
+        dtype: object
+
+        Sequence of int repeats corresponding string in Series
+
+        >>> s.str.repeat(repeats=[1, 2, 3])
+        0      a
+        1     bb
+        2    ccc
+        dtype: object
+        """
+        result = self._data.array._str_repeat(repeats)
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def pad(
+        self,
+        width: int,
+        side: Literal["left", "right", "both"] = "left",
+        fillchar: str = " ",
+    ):
+        """
+        Pad strings in the Series/Index up to width.
+
+        This function pads strings in a Series or Index to a specified width,
+        filling the extra space with a character of your choice. It provides
+        flexibility in positioning the padding, allowing it to be added to the
+        left, right, or both sides. This is useful for formatting strings to
+        align text or ensure consistent string lengths in data processing.
+
+        Parameters
+        ----------
+        width : int
+            Minimum width of resulting string; additional characters will be filled
+            with character defined in `fillchar`.
+        side : {'left', 'right', 'both'}, default 'left'
+            Side from which to fill resulting string.
+        fillchar : str, default ' '
+            Additional character for filling, default is whitespace.
+
+        Returns
+        -------
+        Series or Index of object
+            Returns Series or Index with minimum number of char in object.
+
+        See Also
+        --------
+        Series.str.rjust : Fills the left side of strings with an arbitrary
+            character. Equivalent to ``Series.str.pad(side='left')``.
+        Series.str.ljust : Fills the right side of strings with an arbitrary
+            character. Equivalent to ``Series.str.pad(side='right')``.
+        Series.str.center : Fills both sides of strings with an arbitrary
+            character. Equivalent to ``Series.str.pad(side='both')``.
+        Series.str.zfill : Pad strings in the Series/Index by prepending '0'
+            character. Equivalent to ``Series.str.pad(side='left', fillchar='0')``.
+
+        Examples
+        --------
+        >>> s = pd.Series(["caribou", "tiger"])
+        >>> s
+        0    caribou
+        1      tiger
+        dtype: object
+
+        >>> s.str.pad(width=10)
+        0       caribou
+        1         tiger
+        dtype: object
+
+        >>> s.str.pad(width=10, side="right", fillchar="-")
+        0    caribou---
+        1    tiger-----
+        dtype: object
+
+        >>> s.str.pad(width=10, side="both", fillchar="-")
+        0    -caribou--
+        1    --tiger---
+        dtype: object
+        """
+        if not isinstance(fillchar, str):
+            msg = f"fillchar must be a character, not {type(fillchar).__name__}"
+            raise TypeError(msg)
+
+        if len(fillchar) != 1:
+            raise TypeError("fillchar must be a character, not str")
+
+        if not is_integer(width):
+            msg = f"width must be of integer type, not {type(width).__name__}"
+            raise TypeError(msg)
+
+        result = self._data.array._str_pad(width, side=side, fillchar=fillchar)
+        return self._wrap_result(result)
+
+    _shared_docs["str_pad"] = """
+    Pad %(side)s side of strings in the Series/Index.
+
+    Equivalent to :meth:`str.%(method)s`.
+
+    Parameters
+    ----------
+    width : int
+        Minimum width of resulting string; additional characters will be filled
+        with ``fillchar``.
+    fillchar : str
+        Additional character for filling, default is whitespace.
+
+    Returns
+    -------
+    Series/Index of objects.
+        A Series or Index where the strings are modified by :meth:`str.%(method)s`.
+
+    See Also
+    --------
+    Series.str.rjust : Fills the left side of strings with an arbitrary
+        character.
+    Series.str.ljust : Fills the right side of strings with an arbitrary
+        character.
+    Series.str.center : Fills both sides of strings with an arbitrary
+        character.
+    Series.str.zfill : Pad strings in the Series/Index by prepending '0'
+        character.
+
+    Examples
+    --------
+    For Series.str.center:
+
+    >>> ser = pd.Series(['dog', 'bird', 'mouse'])
+    >>> ser.str.center(8, fillchar='.')
+    0   ..dog...
+    1   ..bird..
+    2   .mouse..
+    dtype: object
+
+    For Series.str.ljust:
+
+    >>> ser = pd.Series(['dog', 'bird', 'mouse'])
+    >>> ser.str.ljust(8, fillchar='.')
+    0   dog.....
+    1   bird....
+    2   mouse...
+    dtype: object
+
+    For Series.str.rjust:
+
+    >>> ser = pd.Series(['dog', 'bird', 'mouse'])
+    >>> ser.str.rjust(8, fillchar='.')
+    0   .....dog
+    1   ....bird
+    2   ...mouse
+    dtype: object
+    """
+
+    @Appender(_shared_docs["str_pad"] % {"side": "left and right", "method": "center"})
+    @forbid_nonstring_types(["bytes"])
+    def center(self, width: int, fillchar: str = " "):
+        return self.pad(width, side="both", fillchar=fillchar)
+
+    @Appender(_shared_docs["str_pad"] % {"side": "right", "method": "ljust"})
+    @forbid_nonstring_types(["bytes"])
+    def ljust(self, width: int, fillchar: str = " "):
+        return self.pad(width, side="right", fillchar=fillchar)
+
+    @Appender(_shared_docs["str_pad"] % {"side": "left", "method": "rjust"})
+    @forbid_nonstring_types(["bytes"])
+    def rjust(self, width: int, fillchar: str = " "):
+        return self.pad(width, side="left", fillchar=fillchar)
+
+    @forbid_nonstring_types(["bytes"])
+    def zfill(self, width: int):
+        """
+        Pad strings in the Series/Index by prepending '0' characters.
+
+        Strings in the Series/Index are padded with '0' characters on the
+        left of the string to reach a total string length  `width`. Strings
+        in the Series/Index with length greater or equal to `width` are
+        unchanged.
+
+        Parameters
+        ----------
+        width : int
+            Minimum length of resulting string; strings with length less
+            than `width` be prepended with '0' characters.
+
+        Returns
+        -------
+        Series/Index of objects.
+            A Series or Index where the strings are prepended with '0' characters.
+
+        See Also
+        --------
+        Series.str.rjust : Fills the left side of strings with an arbitrary
+            character.
+        Series.str.ljust : Fills the right side of strings with an arbitrary
+            character.
+        Series.str.pad : Fills the specified sides of strings with an arbitrary
+            character.
+        Series.str.center : Fills both sides of strings with an arbitrary
+            character.
+
+        Notes
+        -----
+        Differs from :meth:`str.zfill` which has special handling
+        for '+'/'-' in the string.
+
+        Examples
+        --------
+        >>> s = pd.Series(["-1", "1", "1000", 10, np.nan])
+        >>> s
+        0      -1
+        1       1
+        2    1000
+        3      10
+        4     NaN
+        dtype: object
+
+        Note that ``10`` and ``NaN`` are not strings, therefore they are
+        converted to ``NaN``. The minus sign in ``'-1'`` is treated as a
+        special character and the zero is added to the right of it
+        (:meth:`str.zfill` would have moved it to the left). ``1000``
+        remains unchanged as it is longer than `width`.
+
+        >>> s.str.zfill(3)
+        0     -01
+        1     001
+        2    1000
+        3     NaN
+        4     NaN
+        dtype: object
+        """
+        if not is_integer(width):
+            msg = f"width must be of integer type, not {type(width).__name__}"
+            raise TypeError(msg)
+        f = lambda x: x.zfill(width)
+        result = self._data.array._str_map(f)
+        return self._wrap_result(result)
+
+    def slice(self, start=None, stop=None, step=None):
+        """
+        Slice substrings from each element in the Series or Index.
+
+        Slicing substrings from strings in a Series or Index helps extract
+        specific portions of data, making it easier to analyze or manipulate
+        text. This is useful for tasks like parsing structured text fields or
+        isolating parts of strings with a consistent format.
+
+        Parameters
+        ----------
+        start : int, optional
+            Start position for slice operation.
+        stop : int, optional
+            Stop position for slice operation.
+        step : int, optional
+            Step size for slice operation.
+
+        Returns
+        -------
+        Series or Index of object
+            Series or Index from sliced substring from original string object.
+
+        See Also
+        --------
+        Series.str.slice_replace : Replace a slice with a string.
+        Series.str.get : Return element at position.
+            Equivalent to `Series.str.slice(start=i, stop=i+1)` with `i`
+            being the position.
+
+        Examples
+        --------
+        >>> s = pd.Series(["koala", "dog", "chameleon"])
+        >>> s
+        0        koala
+        1          dog
+        2    chameleon
+        dtype: object
+
+        >>> s.str.slice(start=1)
+        0        oala
+        1          og
+        2    hameleon
+        dtype: object
+
+        >>> s.str.slice(start=-1)
+        0           a
+        1           g
+        2           n
+        dtype: object
+
+        >>> s.str.slice(stop=2)
+        0    ko
+        1    do
+        2    ch
+        dtype: object
+
+        >>> s.str.slice(step=2)
+        0      kaa
+        1       dg
+        2    caeen
+        dtype: object
+
+        >>> s.str.slice(start=0, stop=5, step=3)
+        0    kl
+        1     d
+        2    cm
+        dtype: object
+
+        Equivalent behaviour to:
+
+        >>> s.str[0:5:3]
+        0    kl
+        1     d
+        2    cm
+        dtype: object
+        """
+        result = self._data.array._str_slice(start, stop, step)
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def slice_replace(self, start=None, stop=None, repl=None):
+        """
+        Replace a positional slice of a string with another value.
+
+        This function allows replacing specific parts of a string in a Series
+        or Index by specifying start and stop positions. It is useful for
+        modifying substrings in a controlled way, such as updating sections of
+        text based on their positions or patterns.
+
+        Parameters
+        ----------
+        start : int, optional
+            Left index position to use for the slice. If not specified (None),
+            the slice is unbounded on the left, i.e. slice from the start
+            of the string.
+        stop : int, optional
+            Right index position to use for the slice. If not specified (None),
+            the slice is unbounded on the right, i.e. slice until the
+            end of the string.
+        repl : str, optional
+            String for replacement. If not specified (None), the sliced region
+            is replaced with an empty string.
+
+        Returns
+        -------
+        Series or Index
+            Same type as the original object.
+
+        See Also
+        --------
+        Series.str.slice : Just slicing without replacement.
+
+        Examples
+        --------
+        >>> s = pd.Series(["a", "ab", "abc", "abdc", "abcde"])
+        >>> s
+        0        a
+        1       ab
+        2      abc
+        3     abdc
+        4    abcde
+        dtype: object
+
+        Specify just `start`, meaning replace `start` until the end of the
+        string with `repl`.
+
+        >>> s.str.slice_replace(1, repl="X")
+        0    aX
+        1    aX
+        2    aX
+        3    aX
+        4    aX
+        dtype: object
+
+        Specify just `stop`, meaning the start of the string to `stop` is replaced
+        with `repl`, and the rest of the string is included.
+
+        >>> s.str.slice_replace(stop=2, repl="X")
+        0       X
+        1       X
+        2      Xc
+        3     Xdc
+        4    Xcde
+        dtype: object
+
+        Specify `start` and `stop`, meaning the slice from `start` to `stop` is
+        replaced with `repl`. Everything before or after `start` and `stop` is
+        included as is.
+
+        >>> s.str.slice_replace(start=1, stop=3, repl="X")
+        0      aX
+        1      aX
+        2      aX
+        3     aXc
+        4    aXde
+        dtype: object
+        """
+        result = self._data.array._str_slice_replace(start, stop, repl)
+        return self._wrap_result(result)
+
+    def decode(
+        self, encoding, errors: str = "strict", dtype: str | DtypeObj | None = None
+    ):
+        """
+        Decode character string in the Series/Index using indicated encoding.
+
+        Equivalent to :meth:`str.decode` in python2 and :meth:`bytes.decode` in
+        python3.
+
+        Parameters
+        ----------
+        encoding : str
+            Specifies the encoding to be used.
+        errors : str, optional
+            Specifies the error handling scheme.
+            Possible values are those supported by :meth:`bytes.decode`.
+        dtype : str or dtype, optional
+            The dtype of the result. When not ``None``, must be either a string or
+            object dtype. When ``None``, the dtype of the result is determined by
+            ``pd.options.future.infer_string``.
+
+            .. versionadded:: 2.3.0
+
+        Returns
+        -------
+        Series or Index
+            A Series or Index with decoded strings.
+
+        See Also
+        --------
+        Series.str.encode : Encodes strings into bytes in a Series/Index.
+
+        Examples
+        --------
+        For Series:
+
+        >>> ser = pd.Series([b"cow", b"123", b"()"])
+        >>> ser.str.decode("ascii")
+        0   cow
+        1   123
+        2   ()
+        dtype: object
+        """
+        if dtype is not None and not is_string_dtype(dtype):
+            raise ValueError(f"dtype must be string or object, got {dtype=}")
+        if dtype is None and get_option("future.infer_string"):
+            dtype = "str"
+        # TODO: Add a similar _bytes interface.
+        if encoding in _cpython_optimized_decoders:
+            # CPython optimized implementation
+            f = lambda x: x.decode(encoding, errors)
+        else:
+            decoder = codecs.getdecoder(encoding)
+            f = lambda x: decoder(x, errors)[0]
+        arr = self._data.array
+        result = arr._str_map(f)
+        return self._wrap_result(result, dtype=dtype)
+
+    @forbid_nonstring_types(["bytes"])
+    def encode(self, encoding, errors: str = "strict"):
+        """
+        Encode character string in the Series/Index using indicated encoding.
+
+        Equivalent to :meth:`str.encode`.
+
+        Parameters
+        ----------
+        encoding : str
+            Specifies the encoding to be used.
+        errors : str, optional
+            Specifies the error handling scheme.
+            Possible values are those supported by :meth:`str.encode`.
+
+        Returns
+        -------
+        Series/Index of objects
+            A Series or Index with strings encoded into bytes.
+
+        See Also
+        --------
+        Series.str.decode : Decodes bytes into strings in a Series/Index.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["cow", "123", "()"])
+        >>> ser.str.encode(encoding="ascii")
+        0     b'cow'
+        1     b'123'
+        2      b'()'
+        dtype: object
+        """
+        result = self._data.array._str_encode(encoding, errors)
+        return self._wrap_result(result, returns_string=False)
+
+    _shared_docs["str_strip"] = r"""
+    Remove %(position)s characters.
+
+    Strip whitespaces (including newlines) or a set of specified characters
+    from each string in the Series/Index from %(side)s.
+    Replaces any non-strings in Series with NaNs.
+    Equivalent to :meth:`str.%(method)s`.
+
+    Parameters
+    ----------
+    to_strip : str or None, default None
+        Specifying the set of characters to be removed.
+        All combinations of this set of characters will be stripped.
+        If None then whitespaces are removed.
+
+    Returns
+    -------
+    Series or Index of object
+        Series or Index with the strings being stripped from the %(side)s.
+
+    See Also
+    --------
+    Series.str.strip : Remove leading and trailing characters in Series/Index.
+    Series.str.lstrip : Remove leading characters in Series/Index.
+    Series.str.rstrip : Remove trailing characters in Series/Index.
+
+    Examples
+    --------
+    >>> s = pd.Series(['1. Ant.  ', '2. Bee!\n', '3. Cat?\t', np.nan, 10, True])
+    >>> s
+    0    1. Ant.
+    1    2. Bee!\n
+    2    3. Cat?\t
+    3          NaN
+    4           10
+    5         True
+    dtype: object
+
+    >>> s.str.strip()
+    0    1. Ant.
+    1    2. Bee!
+    2    3. Cat?
+    3        NaN
+    4        NaN
+    5        NaN
+    dtype: object
+
+    >>> s.str.lstrip('123.')
+    0    Ant.
+    1    Bee!\n
+    2    Cat?\t
+    3       NaN
+    4       NaN
+    5       NaN
+    dtype: object
+
+    >>> s.str.rstrip('.!? \n\t')
+    0    1. Ant
+    1    2. Bee
+    2    3. Cat
+    3       NaN
+    4       NaN
+    5       NaN
+    dtype: object
+
+    >>> s.str.strip('123.!? \n\t')
+    0    Ant
+    1    Bee
+    2    Cat
+    3    NaN
+    4    NaN
+    5    NaN
+    dtype: object
+    """
+
+    @Appender(
+        _shared_docs["str_strip"]
+        % {
+            "side": "left and right sides",
+            "method": "strip",
+            "position": "leading and trailing",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def strip(self, to_strip=None):
+        result = self._data.array._str_strip(to_strip)
+        return self._wrap_result(result)
+
+    @Appender(
+        _shared_docs["str_strip"]
+        % {"side": "left side", "method": "lstrip", "position": "leading"}
+    )
+    @forbid_nonstring_types(["bytes"])
+    def lstrip(self, to_strip=None):
+        result = self._data.array._str_lstrip(to_strip)
+        return self._wrap_result(result)
+
+    @Appender(
+        _shared_docs["str_strip"]
+        % {"side": "right side", "method": "rstrip", "position": "trailing"}
+    )
+    @forbid_nonstring_types(["bytes"])
+    def rstrip(self, to_strip=None):
+        result = self._data.array._str_rstrip(to_strip)
+        return self._wrap_result(result)
+
+    _shared_docs["str_removefix"] = r"""
+    Remove a %(side)s from an object series.
+
+    If the %(side)s is not present, the original string will be returned.
+
+    Parameters
+    ----------
+    %(side)s : str
+        Remove the %(side)s of the string.
+
+    Returns
+    -------
+    Series/Index: object
+        The Series or Index with given %(side)s removed.
+
+    See Also
+    --------
+    Series.str.remove%(other_side)s : Remove a %(other_side)s from an object series.
+
+    Examples
+    --------
+    >>> s = pd.Series(["str_foo", "str_bar", "no_prefix"])
+    >>> s
+    0    str_foo
+    1    str_bar
+    2    no_prefix
+    dtype: object
+    >>> s.str.removeprefix("str_")
+    0    foo
+    1    bar
+    2    no_prefix
+    dtype: object
+
+    >>> s = pd.Series(["foo_str", "bar_str", "no_suffix"])
+    >>> s
+    0    foo_str
+    1    bar_str
+    2    no_suffix
+    dtype: object
+    >>> s.str.removesuffix("_str")
+    0    foo
+    1    bar
+    2    no_suffix
+    dtype: object
+    """
+
+    @Appender(
+        _shared_docs["str_removefix"] % {"side": "prefix", "other_side": "suffix"}
+    )
+    @forbid_nonstring_types(["bytes"])
+    def removeprefix(self, prefix: str):
+        result = self._data.array._str_removeprefix(prefix)
+        return self._wrap_result(result)
+
+    @Appender(
+        _shared_docs["str_removefix"] % {"side": "suffix", "other_side": "prefix"}
+    )
+    @forbid_nonstring_types(["bytes"])
+    def removesuffix(self, suffix: str):
+        result = self._data.array._str_removesuffix(suffix)
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def wrap(
+        self,
+        width: int,
+        expand_tabs: bool = True,
+        tabsize: int = 8,
+        replace_whitespace: bool = True,
+        drop_whitespace: bool = True,
+        initial_indent: str = "",
+        subsequent_indent: str = "",
+        fix_sentence_endings: bool = False,
+        break_long_words: bool = True,
+        break_on_hyphens: bool = True,
+        max_lines: int | None = None,
+        placeholder: str = " [...]",
+    ):
+        r"""
+        Wrap strings in Series/Index at specified line width.
+
+        This method has the same keyword parameters and defaults as
+            :class:`textwrap.TextWrapper`.
+
+        Parameters
+        ----------
+        width : int, optional
+            Maximum line width.
+        expand_tabs : bool, optional
+            If True, tab characters will be expanded to spaces (default: True).
+        tabsize : int, optional
+            If expand_tabs is true, then all tab characters in text will be
+            expanded to zero or more spaces, depending on the current column
+            and the given tab size (default: 8).
+        replace_whitespace : bool, optional
+            If True, each whitespace character (as defined by string.whitespace)
+            remaining after tab expansion will be replaced by a single space
+            (default: True).
+        drop_whitespace : bool, optional
+            If True, whitespace that, after wrapping, happens to end up at the
+            beginning or end of a line is dropped (default: True).
+        initial_indent : str, optional
+            String that will be prepended to the first line of wrapped output.
+            Counts towards the length of the first line. The empty string is
+            not indented (default: '').
+        subsequent_indent : str, optional
+            String that will be prepended to all lines of wrapped output except
+            the first. Counts towards the length of each line except the first
+            (default: '').
+        fix_sentence_endings : bool, optional
+            If true, TextWrapper attempts to detect sentence endings and ensure
+            that sentences are always separated by exactly two spaces. This is
+            generally desired for text in a monospaced font. However, the sentence
+            detection algorithm is imperfect: it assumes that a sentence ending
+            consists of a lowercase letter followed by one of '.', '!', or '?',
+            possibly followed by one of '"' or "'", followed by a space. One
+            problem with this algorithm is that it is unable to detect the
+            difference between “Dr.” in `[...] Dr. Frankenstein's monster [...]`
+            and “Spot.” in `[...] See Spot. See Spot run [...]`
+            Since the sentence detection algorithm relies on string.lowercase
+            for the definition of “lowercase letter”, and a convention of using
+            two spaces after a period to separate sentences on the same line,
+            it is specific to English-language texts (default: False).
+        break_long_words : bool, optional
+            If True, then words longer than width will be broken in order to ensure
+            that no lines are longer than width. If it is false, long words will
+            not be broken, and some lines may be longer than width (default: True).
+        break_on_hyphens : bool, optional
+            If True, wrapping will occur preferably on whitespace and right after
+            hyphens in compound words, as it is customary in English. If false,
+            only whitespaces will be considered as potentially good places for line
+            breaks, but you need to set break_long_words to false if you want truly
+            insecable words (default: True).
+        max_lines : int, optional
+            If not None, then the output will contain at most max_lines lines, with
+            placeholder appearing at the end of the output (default: None).
+        placeholder : str, optional
+            String that will appear at the end of the output text if it has been
+            truncated (default: ' [...]').
+
+        Returns
+        -------
+        Series or Index
+            A Series or Index where the strings are wrapped at the specified line width.
+
+        See Also
+        --------
+        Series.str.strip : Remove leading and trailing characters in Series/Index.
+        Series.str.lstrip : Remove leading characters in Series/Index.
+        Series.str.rstrip : Remove trailing characters in Series/Index.
+
+        Notes
+        -----
+        Internally, this method uses a :class:`textwrap.TextWrapper` instance with
+        default settings. To achieve behavior matching R's stringr library str_wrap
+        function, use the arguments:
+
+        - expand_tabs = False
+        - replace_whitespace = True
+        - drop_whitespace = True
+        - break_long_words = False
+        - break_on_hyphens = False
+
+        Examples
+        --------
+        >>> s = pd.Series(["line to be wrapped", "another line to be wrapped"])
+        >>> s.str.wrap(12)
+        0             line to be\nwrapped
+        1    another line\nto be\nwrapped
+        dtype: object
+        """
+        result = self._data.array._str_wrap(
+            width=width,
+            expand_tabs=expand_tabs,
+            tabsize=tabsize,
+            replace_whitespace=replace_whitespace,
+            drop_whitespace=drop_whitespace,
+            initial_indent=initial_indent,
+            subsequent_indent=subsequent_indent,
+            fix_sentence_endings=fix_sentence_endings,
+            break_long_words=break_long_words,
+            break_on_hyphens=break_on_hyphens,
+            max_lines=max_lines,
+            placeholder=placeholder,
+        )
+        return self._wrap_result(result)
+
+    @forbid_nonstring_types(["bytes"])
+    def get_dummies(
+        self,
+        sep: str = "|",
+        dtype: NpDtype | None = None,
+    ):
+        """
+        Return DataFrame of dummy/indicator variables for Series.
+
+        Each string in Series is split by sep and returned as a DataFrame
+        of dummy/indicator variables.
+
+        Parameters
+        ----------
+        sep : str, default "|"
+            String to split on.
+        dtype : dtype, default np.int64
+            Data type for new columns. Only a single dtype is allowed.
+
+        Returns
+        -------
+        DataFrame
+            Dummy variables corresponding to values of the Series.
+
+        See Also
+        --------
+        get_dummies : Convert categorical variable into dummy/indicator
+            variables.
+
+        Examples
+        --------
+        >>> pd.Series(["a|b", "a", "a|c"]).str.get_dummies()
+           a  b  c
+        0  1  1  0
+        1  1  0  0
+        2  1  0  1
+
+        >>> pd.Series(["a|b", np.nan, "a|c"]).str.get_dummies()
+           a  b  c
+        0  1  1  0
+        1  0  0  0
+        2  1  0  1
+
+        >>> pd.Series(["a|b", np.nan, "a|c"]).str.get_dummies(dtype=bool)
+                a      b      c
+        0   True   True    False
+        1   False  False   False
+        2   True   False   True
+        """
+        from pandas.core.frame import DataFrame
+
+        if dtype is not None and not (is_numeric_dtype(dtype) or is_bool_dtype(dtype)):
+            raise ValueError("Only numeric or boolean dtypes are supported for 'dtype'")
+        # we need to cast to Series of strings as only that has all
+        # methods available for making the dummies...
+        result, name = self._data.array._str_get_dummies(sep, dtype)
+        if is_extension_array_dtype(dtype):
+            return self._wrap_result(
+                DataFrame(result, columns=name, dtype=dtype),
+                name=name,
+                returns_string=False,
+            )
+        return self._wrap_result(
+            result,
+            name=name,
+            expand=True,
+            returns_string=False,
+        )
+
+    @forbid_nonstring_types(["bytes"])
+    def translate(self, table):
+        """
+        Map all characters in the string through the given mapping table.
+
+        This method is equivalent to the standard :meth:`str.translate`
+        method for strings. It maps each character in the string to a new
+        character according to the translation table provided. Unmapped
+        characters are left unchanged, while characters mapped to None
+        are removed.
+
+        Parameters
+        ----------
+        table : dict
+            Table is a mapping of Unicode ordinals to Unicode ordinals, strings, or
+            None. Unmapped characters are left untouched.
+            Characters mapped to None are deleted. :meth:`str.maketrans` is a
+            helper function for making translation tables.
+
+        Returns
+        -------
+        Series or Index
+            A new Series or Index with translated strings.
+
+        See Also
+        --------
+        Series.str.replace : Replace occurrences of pattern/regex in the
+            Series with some other string.
+        Index.str.replace : Replace occurrences of pattern/regex in the
+            Index with some other string.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["El niño", "Françoise"])
+        >>> mytable = str.maketrans({"ñ": "n", "ç": "c"})
+        >>> ser.str.translate(mytable)
+        0   El nino
+        1   Francoise
+        dtype: object
+        """
+        result = self._data.array._str_translate(table)
+        dtype = object if self._data.dtype == "object" else None
+        return self._wrap_result(result, dtype=dtype)
+
+    @forbid_nonstring_types(["bytes"])
+    def count(self, pat, flags: int = 0):
+        r"""
+        Count occurrences of pattern in each string of the Series/Index.
+
+        This function is used to count the number of times a particular regex
+        pattern is repeated in each of the string elements of the
+        :class:`~pandas.Series`.
+
+        Parameters
+        ----------
+        pat : str
+            Valid regular expression.
+        flags : int, default 0, meaning no flags
+            Flags for the `re` module. For a complete list, `see here
+            <https://docs.python.org/3/howto/regex.html#compilation-flags>`_.
+
+        Returns
+        -------
+        Series or Index
+            Same type as the calling object containing the integer counts.
+
+        See Also
+        --------
+        re : Standard library module for regular expressions.
+        str.count : Standard library version, without regular expression support.
+
+        Notes
+        -----
+        Some characters need to be escaped when passing in `pat`.
+        eg. ``'$'`` has a special meaning in regex and must be escaped when
+        finding this literal character.
+
+        Examples
+        --------
+        >>> s = pd.Series(["A", "B", "Aaba", "Baca", np.nan, "CABA", "cat"])
+        >>> s.str.count("a")
+        0    0.0
+        1    0.0
+        2    2.0
+        3    2.0
+        4    NaN
+        5    0.0
+        6    1.0
+        dtype: float64
+
+        Escape ``'$'`` to find the literal dollar sign.
+
+        >>> s = pd.Series(["$", "B", "Aab$", "$$ca", "C$B$", "cat"])
+        >>> s.str.count("\\$")
+        0    1
+        1    0
+        2    1
+        3    2
+        4    2
+        5    0
+        dtype: int64
+
+        This is also available on Index
+
+        >>> pd.Index(["A", "A", "Aaba", "cat"]).str.count("a")
+        Index([0, 0, 2, 1], dtype='int64')
+        """
+        result = self._data.array._str_count(pat, flags)
+        return self._wrap_result(result, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def startswith(
+        self, pat: str | tuple[str, ...], na: Scalar | lib.NoDefault = lib.no_default
+    ) -> Series | Index:
+        """
+        Test if the start of each string element matches a pattern.
+
+        Equivalent to :meth:`str.startswith`.
+
+        Parameters
+        ----------
+        pat : str or tuple[str, ...]
+            Character sequence or tuple of strings. Regular expressions are not
+            accepted.
+        na : scalar, optional
+            Object shown if element tested is not a string. The default depends
+            on dtype of the array. For object-dtype, ``numpy.nan`` is used.
+            For the nullable ``StringDtype``, ``pandas.NA`` is used.
+            For the ``"str"`` dtype, ``False`` is used.
+
+        Returns
+        -------
+        Series or Index of bool
+            A Series of booleans indicating whether the given pattern matches
+            the start of each string element.
+
+        See Also
+        --------
+        str.startswith : Python standard library string method.
+        Series.str.endswith : Same as startswith, but tests the end of string.
+        Series.str.contains : Tests if string element contains a pattern.
+
+        Examples
+        --------
+        >>> s = pd.Series(["bat", "Bear", "cat", np.nan])
+        >>> s
+        0     bat
+        1    Bear
+        2     cat
+        3     NaN
+        dtype: object
+
+        >>> s.str.startswith("b")
+        0     True
+        1    False
+        2    False
+        3      NaN
+        dtype: object
+
+        >>> s.str.startswith(("b", "B"))
+        0     True
+        1     True
+        2    False
+        3      NaN
+        dtype: object
+
+        Specifying `na` to be `False` instead of `NaN`.
+
+        >>> s.str.startswith("b", na=False)
+        0     True
+        1    False
+        2    False
+        3    False
+        dtype: bool
+        """
+        if not isinstance(pat, (str, tuple)):
+            msg = f"expected a string or tuple, not {type(pat).__name__}"
+            raise TypeError(msg)
+        result = self._data.array._str_startswith(pat, na=na)
+        return self._wrap_result(result, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def endswith(
+        self, pat: str | tuple[str, ...], na: Scalar | lib.NoDefault = lib.no_default
+    ) -> Series | Index:
+        """
+        Test if the end of each string element matches a pattern.
+
+        Equivalent to :meth:`str.endswith`.
+
+        Parameters
+        ----------
+        pat : str or tuple[str, ...]
+            Character sequence or tuple of strings. Regular expressions are not
+            accepted.
+        na : scalar, optional
+            Object shown if element tested is not a string. The default depends
+            on dtype of the array. For object-dtype, ``numpy.nan`` is used.
+            For the nullable ``StringDtype``, ``pandas.NA`` is used.
+            For the ``"str"`` dtype, ``False`` is used.
+
+        Returns
+        -------
+        Series or Index of bool
+            A Series of booleans indicating whether the given pattern matches
+            the end of each string element.
+
+        See Also
+        --------
+        str.endswith : Python standard library string method.
+        Series.str.startswith : Same as endswith, but tests the start of string.
+        Series.str.contains : Tests if string element contains a pattern.
+
+        Examples
+        --------
+        >>> s = pd.Series(["bat", "bear", "caT", np.nan])
+        >>> s
+        0     bat
+        1    bear
+        2     caT
+        3     NaN
+        dtype: object
+
+        >>> s.str.endswith("t")
+        0     True
+        1    False
+        2    False
+        3      NaN
+        dtype: object
+
+        >>> s.str.endswith(("t", "T"))
+        0     True
+        1    False
+        2     True
+        3      NaN
+        dtype: object
+
+        Specifying `na` to be `False` instead of `NaN`.
+
+        >>> s.str.endswith("t", na=False)
+        0     True
+        1    False
+        2    False
+        3    False
+        dtype: bool
+        """
+        if not isinstance(pat, (str, tuple)):
+            msg = f"expected a string or tuple, not {type(pat).__name__}"
+            raise TypeError(msg)
+        result = self._data.array._str_endswith(pat, na=na)
+        return self._wrap_result(result, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def findall(self, pat, flags: int = 0):
+        """
+        Find all occurrences of pattern or regular expression in the Series/Index.
+
+        Equivalent to applying :func:`re.findall` to all the elements in the
+        Series/Index.
+
+        Parameters
+        ----------
+        pat : str
+            Pattern or regular expression.
+        flags : int, default 0
+            Flags from ``re`` module, e.g. `re.IGNORECASE` (default is 0, which
+            means no flags).
+
+        Returns
+        -------
+        Series/Index of lists of strings
+            All non-overlapping matches of pattern or regular expression in each
+            string of this Series/Index.
+
+        See Also
+        --------
+        count : Count occurrences of pattern or regular expression in each string
+            of the Series/Index.
+        extractall : For each string in the Series, extract groups from all matches
+            of regular expression and return a DataFrame with one row for each
+            match and one column for each group.
+        re.findall : The equivalent ``re`` function to all non-overlapping matches
+            of pattern or regular expression in string, as a list of strings.
+
+        Examples
+        --------
+        >>> s = pd.Series(["Lion", "Monkey", "Rabbit"])
+
+        The search for the pattern 'Monkey' returns one match:
+
+        >>> s.str.findall("Monkey")
+        0          []
+        1    [Monkey]
+        2          []
+        dtype: object
+
+        On the other hand, the search for the pattern 'MONKEY' doesn't return any
+        match:
+
+        >>> s.str.findall("MONKEY")
+        0    []
+        1    []
+        2    []
+        dtype: object
+
+        Flags can be added to the pattern or regular expression. For instance,
+        to find the pattern 'MONKEY' ignoring the case:
+
+        >>> import re
+        >>> s.str.findall("MONKEY", flags=re.IGNORECASE)
+        0          []
+        1    [Monkey]
+        2          []
+        dtype: object
+
+        When the pattern matches more than one string in the Series, all matches
+        are returned:
+
+        >>> s.str.findall("on")
+        0    [on]
+        1    [on]
+        2      []
+        dtype: object
+
+        Regular expressions are supported too. For instance, the search for all the
+        strings ending with the word 'on' is shown next:
+
+        >>> s.str.findall("on$")
+        0    [on]
+        1      []
+        2      []
+        dtype: object
+
+        If the pattern is found more than once in the same string, then a list of
+        multiple strings is returned:
+
+        >>> s.str.findall("b")
+        0        []
+        1        []
+        2    [b, b]
+        dtype: object
+        """
+        result = self._data.array._str_findall(pat, flags)
+        return self._wrap_result(result, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def extract(
+        self, pat: str, flags: int = 0, expand: bool = True
+    ) -> DataFrame | Series | Index:
+        r"""
+        Extract capture groups in the regex `pat` as columns in a DataFrame.
+
+        For each subject string in the Series, extract groups from the
+        first match of regular expression `pat`.
+
+        Parameters
+        ----------
+        pat : str
+            Regular expression pattern with capturing groups.
+        flags : int, default 0 (no flags)
+            Flags from the ``re`` module, e.g. ``re.IGNORECASE``, that
+            modify regular expression matching for things like case,
+            spaces, etc. For more details, see :mod:`re`.
+        expand : bool, default True
+            If True, return DataFrame with one column per capture group.
+            If False, return a Series/Index if there is one capture group
+            or DataFrame if there are multiple capture groups.
+
+        Returns
+        -------
+        DataFrame or Series or Index
+            A DataFrame with one row for each subject string, and one
+            column for each group. Any capture group names in regular
+            expression pat will be used for column names; otherwise
+            capture group numbers will be used. The dtype of each result
+            column is always object, even when no match is found. If
+            ``expand=False`` and pat has only one capture group, then
+            return a Series (if subject is a Series) or Index (if subject
+            is an Index).
+
+        See Also
+        --------
+        extractall : Returns all matches (not just the first match).
+
+        Examples
+        --------
+        A pattern with two groups will return a DataFrame with two columns.
+        Non-matches will be NaN.
+
+        >>> s = pd.Series(["a1", "b2", "c3"])
+        >>> s.str.extract(r"([ab])(\d)")
+            0    1
+        0    a    1
+        1    b    2
+        2  NaN  NaN
+
+        A pattern may contain optional groups.
+
+        >>> s.str.extract(r"([ab])?(\d)")
+            0  1
+        0    a  1
+        1    b  2
+        2  NaN  3
+
+        Named groups will become column names in the result.
+
+        >>> s.str.extract(r"(?P<letter>[ab])(?P<digit>\d)")
+        letter digit
+        0      a     1
+        1      b     2
+        2    NaN   NaN
+
+        A pattern with one group will return a DataFrame with one column
+        if expand=True.
+
+        >>> s.str.extract(r"[ab](\d)", expand=True)
+            0
+        0    1
+        1    2
+        2  NaN
+
+        A pattern with one group will return a Series if expand=False.
+
+        >>> s.str.extract(r"[ab](\d)", expand=False)
+        0      1
+        1      2
+        2    NaN
+        dtype: object
+        """
+        from pandas import DataFrame
+
+        if not isinstance(expand, bool):
+            raise ValueError("expand must be True or False")
+
+        regex = re.compile(pat, flags=flags)
+        if regex.groups == 0:
+            raise ValueError("pattern contains no capture groups")
+
+        if not expand and regex.groups > 1 and isinstance(self._data, ABCIndex):
+            raise ValueError("only one regex group is supported with Index")
+
+        obj = self._data
+        result_dtype = _result_dtype(obj)
+
+        returns_df = regex.groups > 1 or expand
+
+        if returns_df:
+            name = None
+            columns = _get_group_names(regex)
+
+            if obj.array.size == 0:
+                result = DataFrame(columns=columns, dtype=result_dtype)
+
+            else:
+                result_list = self._data.array._str_extract(
+                    pat, flags=flags, expand=returns_df
+                )
+
+                result_index: Index | None
+                if isinstance(obj, ABCSeries):
+                    result_index = obj.index
+                else:
+                    result_index = None
+
+                result = DataFrame(
+                    result_list, columns=columns, index=result_index, dtype=result_dtype
+                )
+
+        else:
+            name = _get_single_group_name(regex)
+            result = self._data.array._str_extract(pat, flags=flags, expand=returns_df)
+        return self._wrap_result(result, name=name, dtype=result_dtype)
+
+    @forbid_nonstring_types(["bytes"])
+    def extractall(self, pat, flags: int = 0) -> DataFrame:
+        r"""
+        Extract capture groups in the regex `pat` as columns in DataFrame.
+
+        For each subject string in the Series, extract groups from all
+        matches of regular expression pat. When each subject string in the
+        Series has exactly one match, extractall(pat).xs(0, level='match')
+        is the same as extract(pat).
+
+        Parameters
+        ----------
+        pat : str
+            Regular expression pattern with capturing groups.
+        flags : int, default 0 (no flags)
+            A ``re`` module flag, for example ``re.IGNORECASE``. These allow
+            to modify regular expression matching for things like case, spaces,
+            etc. Multiple flags can be combined with the bitwise OR operator,
+            for example ``re.IGNORECASE | re.MULTILINE``.
+
+        Returns
+        -------
+        DataFrame
+            A ``DataFrame`` with one row for each match, and one column for each
+            group. Its rows have a ``MultiIndex`` with first levels that come from
+            the subject ``Series``. The last level is named 'match' and indexes the
+            matches in each item of the ``Series``. Any capture group names in
+            regular expression pat will be used for column names; otherwise capture
+            group numbers will be used.
+
+        See Also
+        --------
+        extract : Returns first match only (not all matches).
+
+        Examples
+        --------
+        A pattern with one group will return a DataFrame with one column.
+        Indices with no matches will not appear in the result.
+
+        >>> s = pd.Series(["a1a2", "b1", "c1"], index=["A", "B", "C"])
+        >>> s.str.extractall(r"[ab](\d)")
+                0
+        match
+        A 0      1
+          1      2
+        B 0      1
+
+        Capture group names are used for column names of the result.
+
+        >>> s.str.extractall(r"[ab](?P<digit>\d)")
+                digit
+        match
+        A 0         1
+          1         2
+        B 0         1
+
+        A pattern with two groups will return a DataFrame with two columns.
+
+        >>> s.str.extractall(r"(?P<letter>[ab])(?P<digit>\d)")
+                letter digit
+        match
+        A 0          a     1
+          1          a     2
+        B 0          b     1
+
+        Optional groups that do not match are NaN in the result.
+
+        >>> s.str.extractall(r"(?P<letter>[ab])?(?P<digit>\d)")
+                letter digit
+        match
+        A 0          a     1
+          1          a     2
+        B 0          b     1
+        C 0        NaN     1
+        """
+        # TODO: dispatch
+        return str_extractall(self._orig, pat, flags)
+
+    _shared_docs["find"] = """
+    Return %(side)s indexes in each strings in the Series/Index.
+
+    Each of returned indexes corresponds to the position where the
+    substring is fully contained between [start:end]. Return -1 on
+    failure. Equivalent to standard :meth:`str.%(method)s`.
+
+    Parameters
+    ----------
+    sub : str
+        Substring being searched.
+    start : int
+        Left edge index.
+    end : int
+        Right edge index.
+
+    Returns
+    -------
+    Series or Index of int.
+        A Series (if the input is a Series) or an Index (if the input is an
+        Index) of the %(side)s indexes corresponding to the positions where the
+        substring is found in each string of the input.
+
+    See Also
+    --------
+    %(also)s
+
+    Examples
+    --------
+    For Series.str.find:
+
+    >>> ser = pd.Series(["_cow_", "duck_", "do_v_e"])
+    >>> ser.str.find("_")
+    0   0
+    1   4
+    2   2
+    dtype: int64
+
+    For Series.str.rfind:
+
+    >>> ser = pd.Series(["_cow_", "duck_", "do_v_e"])
+    >>> ser.str.rfind("_")
+    0   4
+    1   4
+    2   4
+    dtype: int64
+    """
+
+    @Appender(
+        _shared_docs["find"]
+        % {
+            "side": "lowest",
+            "method": "find",
+            "also": "rfind : Return highest indexes in each strings.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def find(self, sub, start: int = 0, end=None):
+        if not isinstance(sub, str):
+            msg = f"expected a string object, not {type(sub).__name__}"
+            raise TypeError(msg)
+
+        result = self._data.array._str_find(sub, start, end)
+        return self._wrap_result(result, returns_string=False)
+
+    @Appender(
+        _shared_docs["find"]
+        % {
+            "side": "highest",
+            "method": "rfind",
+            "also": "find : Return lowest indexes in each strings.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def rfind(self, sub, start: int = 0, end=None):
+        if not isinstance(sub, str):
+            msg = f"expected a string object, not {type(sub).__name__}"
+            raise TypeError(msg)
+
+        result = self._data.array._str_rfind(sub, start=start, end=end)
+        return self._wrap_result(result, returns_string=False)
+
+    @forbid_nonstring_types(["bytes"])
+    def normalize(self, form):
+        """
+        Return the Unicode normal form for the strings in the Series/Index.
+
+        For more information on the forms, see the
+        :func:`unicodedata.normalize`.
+
+        Parameters
+        ----------
+        form : {'NFC', 'NFKC', 'NFD', 'NFKD'}
+            Unicode form.
+
+        Returns
+        -------
+        Series/Index of objects
+            A Series or Index of strings in the same Unicode form specified by `form`.
+            The returned object retains the same type as the input (Series or Index),
+            and contains the normalized strings.
+
+        See Also
+        --------
+        Series.str.upper : Convert all characters in each string to uppercase.
+        Series.str.lower : Convert all characters in each string to lowercase.
+        Series.str.title : Convert each string to title case (capitalizing the
+            first letter of each word).
+        Series.str.strip : Remove leading and trailing whitespace from each string.
+        Series.str.replace : Replace occurrences of a substring with another substring
+            in each string.
+
+        Examples
+        --------
+        >>> ser = pd.Series(["ñ"])
+        >>> ser.str.normalize("NFC") == ser.str.normalize("NFD")
+        0   False
+        dtype: bool
+        """
+        result = self._data.array._str_normalize(form)
+        return self._wrap_result(result)
+
+    _shared_docs["index"] = """
+    Return %(side)s indexes in each string in Series/Index.
+
+    Each of the returned indexes corresponds to the position where the
+    substring is fully contained between [start:end]. This is the same
+    as ``str.%(similar)s`` except instead of returning -1, it raises a
+    ValueError when the substring is not found. Equivalent to standard
+    ``str.%(method)s``.
+
+    Parameters
+    ----------
+    sub : str
+        Substring being searched.
+    start : int
+        Left edge index.
+    end : int
+        Right edge index.
+
+    Returns
+    -------
+    Series or Index of object
+        Returns a Series or an Index of the %(side)s indexes
+        in each string of the input.
+
+    See Also
+    --------
+    %(also)s
+
+    Examples
+    --------
+    For Series.str.index:
+
+    >>> ser = pd.Series(["horse", "eagle", "donkey"])
+    >>> ser.str.index("e")
+    0   4
+    1   0
+    2   4
+    dtype: int64
+
+    For Series.str.rindex:
+
+    >>> ser = pd.Series(["Deer", "eagle", "Sheep"])
+    >>> ser.str.rindex("e")
+    0   2
+    1   4
+    2   3
+    dtype: int64
+    """
+
+    @Appender(
+        _shared_docs["index"]
+        % {
+            "side": "lowest",
+            "similar": "find",
+            "method": "index",
+            "also": "rindex : Return highest indexes in each strings.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def index(self, sub, start: int = 0, end=None):
+        if not isinstance(sub, str):
+            msg = f"expected a string object, not {type(sub).__name__}"
+            raise TypeError(msg)
+
+        result = self._data.array._str_index(sub, start=start, end=end)
+        return self._wrap_result(result, returns_string=False)
+
+    @Appender(
+        _shared_docs["index"]
+        % {
+            "side": "highest",
+            "similar": "rfind",
+            "method": "rindex",
+            "also": "index : Return lowest indexes in each strings.",
+        }
+    )
+    @forbid_nonstring_types(["bytes"])
+    def rindex(self, sub, start: int = 0, end=None):
+        if not isinstance(sub, str):
+            msg = f"expected a string object, not {type(sub).__name__}"
+            raise TypeError(msg)
+
+        result = self._data.array._str_rindex(sub, start=start, end=end)
+        return self._wrap_result(result, returns_string=False)
+
+    def len(self):
+        """
+        Compute the length of each element in the Series/Index.
+
+        The element may be a sequence (such as a string, tuple or list) or a collection
+        (such as a dictionary).
+
+        Returns
+        -------
+        Series or Index of int
+            A Series or Index of integer values indicating the length of each
+            element in the Series or Index.
+
+        See Also
+        --------
+        str.len : Python built-in function returning the length of an object.
+        Series.size : Returns the length of the Series.
+
+        Examples
+        --------
+        Returns the length (number of characters) in a string. Returns the
+        number of entries for dictionaries, lists or tuples.
+
+        >>> s = pd.Series(
+        ...     ["dog", "", 5, {"foo": "bar"}, [2, 3, 5, 7], ("one", "two", "three")]
+        ... )
+        >>> s
+        0                  dog
+        1
+        2                    5
+        3       {'foo': 'bar'}
+        4         [2, 3, 5, 7]
+        5    (one, two, three)
+        dtype: object
+        >>> s.str.len()
+        0    3.0
+        1    0.0
+        2    NaN
+        3    1.0
+        4    4.0
+        5    3.0
+        dtype: float64
+        """
+        result = self._data.array._str_len()
+        return self._wrap_result(result, returns_string=False)
+
+    _shared_docs["casemethods"] = """
+    Convert strings in the Series/Index to %(type)s.
+    %(version)s
+    Equivalent to :meth:`str.%(method)s`.
+
+    Returns
+    -------
+    Series or Index of objects
+        A Series or Index where the strings are modified by :meth:`str.%(method)s`.
+
+    See Also
+    --------
+    Series.str.lower : Converts all characters to lowercase.
+    Series.str.upper : Converts all characters to uppercase.
+    Series.str.title : Converts first character of each word to uppercase and
+        remaining to lowercase.
+    Series.str.capitalize : Converts first character to uppercase and
+        remaining to lowercase.
+    Series.str.swapcase : Converts uppercase to lowercase and lowercase to
+        uppercase.
+    Series.str.casefold: Removes all case distinctions in the string.
+
+    Examples
+    --------
+    >>> s = pd.Series(['lower', 'CAPITALS', 'this is a sentence', 'SwApCaSe'])
+    >>> s
+    0                 lower
+    1              CAPITALS
+    2    this is a sentence
+    3              SwApCaSe
+    dtype: object
+
+    >>> s.str.lower()
+    0                 lower
+    1              capitals
+    2    this is a sentence
+    3              swapcase
+    dtype: object
+
+    >>> s.str.upper()
+    0                 LOWER
+    1              CAPITALS
+    2    THIS IS A SENTENCE
+    3              SWAPCASE
+    dtype: object
+
+    >>> s.str.title()
+    0                 Lower
+    1              Capitals
+    2    This Is A Sentence
+    3              Swapcase
+    dtype: object
+
+    >>> s.str.capitalize()
+    0                 Lower
+    1              Capitals
+    2    This is a sentence
+    3              Swapcase
+    dtype: object
+
+    >>> s.str.swapcase()
+    0                 LOWER
+    1              capitals
+    2    THIS IS A SENTENCE
+    3              sWaPcAsE
+    dtype: object
+    """
+    # Types:
+    #   cases:
+    #       upper, lower, title, capitalize, swapcase, casefold
+    #   boolean:
+    #     isalpha, isnumeric isalnum isdigit isdecimal isspace islower
+    #     isupper istitle isascii
+    # _doc_args holds dict of strings to use in substituting casemethod docs
+    _doc_args: dict[str, dict[str, str]] = {}
+    _doc_args["lower"] = {"type": "lowercase", "method": "lower", "version": ""}
+    _doc_args["upper"] = {"type": "uppercase", "method": "upper", "version": ""}
+    _doc_args["title"] = {"type": "titlecase", "method": "title", "version": ""}
+    _doc_args["capitalize"] = {
+        "type": "be capitalized",
+        "method": "capitalize",
+        "version": "",
+    }
+    _doc_args["swapcase"] = {
+        "type": "be swapcased",
+        "method": "swapcase",
+        "version": "",
+    }
+    _doc_args["casefold"] = {
+        "type": "be casefolded",
+        "method": "casefold",
+        "version": "",
+    }
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["lower"])
+    @forbid_nonstring_types(["bytes"])
+    def lower(self):
+        result = self._data.array._str_lower()
+        return self._wrap_result(result)
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["upper"])
+    @forbid_nonstring_types(["bytes"])
+    def upper(self):
+        result = self._data.array._str_upper()
+        return self._wrap_result(result)
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["title"])
+    @forbid_nonstring_types(["bytes"])
+    def title(self):
+        result = self._data.array._str_title()
+        return self._wrap_result(result)
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["capitalize"])
+    @forbid_nonstring_types(["bytes"])
+    def capitalize(self):
+        result = self._data.array._str_capitalize()
+        return self._wrap_result(result)
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["swapcase"])
+    @forbid_nonstring_types(["bytes"])
+    def swapcase(self):
+        result = self._data.array._str_swapcase()
+        return self._wrap_result(result)
+
+    @Appender(_shared_docs["casemethods"] % _doc_args["casefold"])
+    @forbid_nonstring_types(["bytes"])
+    def casefold(self):
+        result = self._data.array._str_casefold()
+        return self._wrap_result(result)
+
+    _shared_docs["ismethods"] = """
+    Check whether all characters in each string are %(type)s.
+
+    This is equivalent to running the Python string method
+    :meth:`str.%(method)s` for each element of the Series/Index. If a string
+    has zero characters, ``False`` is returned for that check.
+
+    Returns
+    -------
+    Series or Index of bool
+        Series or Index of boolean values with the same length as the original
+        Series/Index.
+    """
+    _shared_docs["isalpha"] = """
+    See Also
+    --------
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+
+    >>> s1 = pd.Series(['one', 'one1', '1', ''])
+    >>> s1.str.isalpha()
+    0     True
+    1    False
+    2    False
+    3    False
+    dtype: bool
+    """
+    _shared_docs["isnumeric"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+    The ``s.str.isnumeric`` method is the same as ``s3.str.isdigit`` but
+    also includes other characters that can represent quantities such as
+    unicode fractions.
+
+    >>> s1 = pd.Series(['one', 'one1', '1', '', '³', '⅕'])
+    >>> s1.str.isnumeric()
+    0    False
+    1    False
+    2     True
+    3    False
+    4     True
+    5     True
+    dtype: bool
+
+    For a string to be considered numeric, all its characters must have a Unicode
+    numeric property matching :py:meth:`str.is_numeric`. As a consequence,
+    the following cases are **not** recognized as numeric:
+
+    - **Decimal numbers** (e.g., "1.1"): due to period ``"."``
+    - **Negative numbers** (e.g., "-5"):  due to minus sign ``"-"``
+    - **Scientific notation** (e.g., "1e3"): due to characters like ``"e"``
+
+    >>> s2 = pd.Series(["1.1", "-5", "1e3"])
+    >>> s2.str.isnumeric()
+    0    False
+    1    False
+    2    False
+    dtype: bool
+    """
+    _shared_docs["isalnum"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+    >>> s1 = pd.Series(['one', 'one1', '1', ''])
+    >>> s1.str.isalnum()
+    0     True
+    1     True
+    2     True
+    3    False
+    dtype: bool
+
+    Note that checks against characters mixed with any additional punctuation
+    or whitespace will evaluate to false for an alphanumeric check.
+
+    >>> s2 = pd.Series(['A B', '1.5', '3,000'])
+    >>> s2.str.isalnum()
+    0    False
+    1    False
+    2    False
+    dtype: bool
+    """
+    _shared_docs["isdecimal"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+    The ``s3.str.isdecimal`` method checks for characters used to form
+    numbers in base 10.
+
+    >>> s3 = pd.Series(['23', '³', '⅕', ''])
+    >>> s3.str.isdecimal()
+    0     True
+    1    False
+    2    False
+    3    False
+    dtype: bool
+    """
+    _shared_docs["isdigit"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+    Similar to ``str.isdecimal`` but also includes special digits, like
+    superscripted and subscripted digits in unicode.
+
+    >>> s3 = pd.Series(['23', '³', '⅕', ''])
+    >>> s3.str.isdigit()
+    0     True
+    1     True
+    2    False
+    3    False
+    dtype: bool
+    """
+
+    _shared_docs["isspace"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+
+    >>> s4 = pd.Series([' ', '\\t\\r\\n ', ''])
+    >>> s4.str.isspace()
+    0     True
+    1     True
+    2    False
+    dtype: bool
+    """
+    _shared_docs["islower"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+
+    >>> s5 = pd.Series(['leopard', 'Golden Eagle', 'SNAKE', ''])
+    >>> s5.str.islower()
+    0     True
+    1    False
+    2    False
+    3    False
+    dtype: bool
+    """
+
+    _shared_docs["isupper"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.istitle : Check whether all characters are titlecase.
+
+    Examples
+    --------
+
+    >>> s5 = pd.Series(['leopard', 'Golden Eagle', 'SNAKE', ''])
+    >>> s5.str.isupper()
+    0    False
+    1    False
+    2     True
+    3    False
+    dtype: bool
+    """
+    _shared_docs["istitle"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.isascii : Check whether all characters are ascii.
+    Series.str.isupper : Check whether all characters are uppercase.
+
+    Examples
+    --------
+    The ``s5.str.istitle`` method checks for whether all words are in title
+    case (whether only the first letter of each word is capitalized). Words are
+    assumed to be as any sequence of non-numeric characters separated by
+    whitespace characters.
+
+    >>> s5 = pd.Series(['leopard', 'Golden Eagle', 'SNAKE', ''])
+    >>> s5.str.istitle()
+    0    False
+    1     True
+    2    False
+    3    False
+    dtype: bool
+    """
+    _shared_docs["isascii"] = """
+    See Also
+    --------
+    Series.str.isalpha : Check whether all characters are alphabetic.
+    Series.str.isnumeric : Check whether all characters are numeric.
+    Series.str.isalnum : Check whether all characters are alphanumeric.
+    Series.str.isdigit : Check whether all characters are digits.
+    Series.str.isdecimal : Check whether all characters are decimal.
+    Series.str.isspace : Check whether all characters are whitespace.
+    Series.str.islower : Check whether all characters are lowercase.
+    Series.str.istitle : Check whether all characters are titlecase.
+    Series.str.isupper : Check whether all characters are uppercase.
+
+    Examples
+    ------------
+    The ``s5.str.isascii`` method checks for whether all characters are ascii
+    characters, which includes digits 0-9, capital and lowercase letters A-Z,
+    and some other special characters.
+
+    >>> s5 = pd.Series(['ö', 'see123', 'hello world', ''])
+    >>> s5.str.isascii()
+    0    False
+    1     True
+    2     True
+    3     True
+    dtype: bool
+    """
+
+    _doc_args["isalnum"] = {"type": "alphanumeric", "method": "isalnum"}
+    _doc_args["isalpha"] = {"type": "alphabetic", "method": "isalpha"}
+    _doc_args["isdigit"] = {"type": "digits", "method": "isdigit"}
+    _doc_args["isspace"] = {"type": "whitespace", "method": "isspace"}
+    _doc_args["islower"] = {"type": "lowercase", "method": "islower"}
+    _doc_args["isascii"] = {"type": "ascii", "method": "isascii"}
+    _doc_args["isupper"] = {"type": "uppercase", "method": "isupper"}
+    _doc_args["istitle"] = {"type": "titlecase", "method": "istitle"}
+    _doc_args["isnumeric"] = {"type": "numeric", "method": "isnumeric"}
+    _doc_args["isdecimal"] = {"type": "decimal", "method": "isdecimal"}
+    # force _noarg_wrapper return type with dtype=np.dtype(bool) (GH 29624)
+
+    isalnum = _map_and_wrap(
+        "isalnum",
+        docstring=_shared_docs["ismethods"] % _doc_args["isalnum"]
+        + _shared_docs["isalnum"],
+    )
+    isalpha = _map_and_wrap(
+        "isalpha",
+        docstring=_shared_docs["ismethods"] % _doc_args["isalpha"]
+        + _shared_docs["isalpha"],
+    )
+    isdigit = _map_and_wrap(
+        "isdigit",
+        docstring=_shared_docs["ismethods"] % _doc_args["isdigit"]
+        + _shared_docs["isdigit"],
+    )
+    isspace = _map_and_wrap(
+        "isspace",
+        docstring=_shared_docs["ismethods"] % _doc_args["isspace"]
+        + _shared_docs["isspace"],
+    )
+    islower = _map_and_wrap(
+        "islower",
+        docstring=_shared_docs["ismethods"] % _doc_args["islower"]
+        + _shared_docs["islower"],
+    )
+    isascii = _map_and_wrap(
+        "isascii",
+        docstring=_shared_docs["ismethods"] % _doc_args["isascii"]
+        + _shared_docs["isascii"],
+    )
+    isupper = _map_and_wrap(
+        "isupper",
+        docstring=_shared_docs["ismethods"] % _doc_args["isupper"]
+        + _shared_docs["isupper"],
+    )
+    istitle = _map_and_wrap(
+        "istitle",
+        docstring=_shared_docs["ismethods"] % _doc_args["istitle"]
+        + _shared_docs["istitle"],
+    )
+    isnumeric = _map_and_wrap(
+        "isnumeric",
+        docstring=_shared_docs["ismethods"] % _doc_args["isnumeric"]
+        + _shared_docs["isnumeric"],
+    )
+    isdecimal = _map_and_wrap(
+        "isdecimal",
+        docstring=_shared_docs["ismethods"] % _doc_args["isdecimal"]
+        + _shared_docs["isdecimal"],
+    )
+
+
+def cat_safe(list_of_columns: list[npt.NDArray[np.object_]], sep: str):
+    """
+    Auxiliary function for :meth:`str.cat`.
+
+    Same signature as cat_core, but handles TypeErrors in concatenation, which
+    happen if the arrays in list_of columns have the wrong dtypes or content.
+
+    Parameters
+    ----------
+    list_of_columns : list of numpy arrays
+        List of arrays to be concatenated with sep;
+        these arrays may not contain NaNs!
+    sep : string
+        The separator string for concatenating the columns.
+
+    Returns
+    -------
+    nd.array
+        The concatenation of list_of_columns with sep.
+    """
+    try:
+        result = cat_core(list_of_columns, sep)
+    except TypeError:
+        # if there are any non-string values (wrong dtype or hidden behind
+        # object dtype), np.sum will fail; catch and return with better message
+        for column in list_of_columns:
+            dtype = lib.infer_dtype(column, skipna=True)
+            if dtype not in ["string", "empty"]:
+                raise TypeError(
+                    "Concatenation requires list-likes containing only "
+                    "strings (or missing values). Offending values found in "
+                    f"column {dtype}"
+                ) from None
+    return result
+
+
+def cat_core(list_of_columns: list, sep: str):
+    """
+    Auxiliary function for :meth:`str.cat`
+
+    Parameters
+    ----------
+    list_of_columns : list of numpy arrays
+        List of arrays to be concatenated with sep;
+        these arrays may not contain NaNs!
+    sep : string
+        The separator string for concatenating the columns.
+
+    Returns
+    -------
+    nd.array
+        The concatenation of list_of_columns with sep.
+    """
+    if sep == "":
+        # no need to interleave sep if it is empty
+        arr_of_cols = np.asarray(list_of_columns, dtype=object)
+        return np.sum(arr_of_cols, axis=0)
+    list_with_sep = [sep] * (2 * len(list_of_columns) - 1)
+    list_with_sep[::2] = list_of_columns
+    arr_with_sep = np.asarray(list_with_sep, dtype=object)
+    return np.sum(arr_with_sep, axis=0)
+
+
+def _result_dtype(arr):
+    # workaround #27953
+    # ideally we just pass `dtype=arr.dtype` unconditionally, but this fails
+    # when the list of values is empty.
+    from pandas.core.arrays.string_ import StringDtype
+
+    if isinstance(arr.dtype, (ArrowDtype, StringDtype)):
+        return arr.dtype
+    return object
+
+
+def _get_single_group_name(regex: re.Pattern) -> Hashable:
+    if regex.groupindex:
+        return next(iter(regex.groupindex))
+    else:
+        return None
+
+
+def _get_group_names(regex: re.Pattern) -> list[Hashable] | range:
+    """
+    Get named groups from compiled regex.
+
+    Unnamed groups are numbered.
+
+    Parameters
+    ----------
+    regex : compiled regex
+
+    Returns
+    -------
+    list of column labels
+    """
+    rng = range(regex.groups)
+    names = {v: k for k, v in regex.groupindex.items()}
+    if not names:
+        return rng
+    result: list[Hashable] = [names.get(1 + i, i) for i in rng]
+    arr = np.array(result)
+    if arr.dtype.kind == "i" and lib.is_range_indexer(arr, len(arr)):
+        return rng
+    return result
+
+
+def str_extractall(arr, pat, flags: int = 0) -> DataFrame:
+    regex = re.compile(pat, flags=flags)
+    # the regex must contain capture groups.
+    if regex.groups == 0:
+        raise ValueError("pattern contains no capture groups")
+
+    if isinstance(arr, ABCIndex):
+        arr = arr.to_series().reset_index(drop=True).astype(arr.dtype)
+
+    columns = _get_group_names(regex)
+    match_list = []
+    index_list = []
+    is_mi = arr.index.nlevels > 1
+
+    for subject_key, subject in arr.items():
+        if isinstance(subject, str):
+            if not is_mi:
+                subject_key = (subject_key,)
+
+            for match_i, match_tuple in enumerate(regex.findall(subject)):
+                if isinstance(match_tuple, str):
+                    match_tuple = (match_tuple,)
+                na_tuple = [np.nan if group == "" else group for group in match_tuple]
+                match_list.append(na_tuple)
+                result_key = tuple(subject_key + (match_i,))
+                index_list.append(result_key)
+
+    from pandas import MultiIndex
+
+    index = MultiIndex.from_tuples(index_list, names=arr.index.names + ["match"])
+    dtype = _result_dtype(arr)
+
+    result = arr._constructor_expanddim(
+        match_list, index=index, columns=columns, dtype=dtype
+    )
+    return result
diff --git a/pandas/pandas/core/strings/base.py b/pandas/pandas/core/strings/base.py
new file mode 100644
index 0000000000000..78c4f3acbe1aa
--- /dev/null
+++ b/pandas/pandas/core/strings/base.py
@@ -0,0 +1,274 @@
+from __future__ import annotations
+
+import abc
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+from pandas._libs import lib
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+    import re
+
+    from pandas._typing import (
+        NpDtype,
+        Scalar,
+        Self,
+    )
+
+
+class BaseStringArrayMethods(abc.ABC):
+    """
+    Base class for extension arrays implementing string methods.
+
+    This is where our ExtensionArrays can override the implementation of
+    Series.str.<method>. We don't expect this to work with
+    3rd-party extension arrays.
+
+    * User calls Series.str.<method>
+    * pandas extracts the extension array from the Series
+    * pandas calls ``extension_array._str_<method>(*args, **kwargs)``
+    * pandas wraps the result, to return to the user.
+
+    See :ref:`Series.str` for the docstring of each method.
+    """
+
+    def _str_getitem(self, key):
+        if isinstance(key, slice):
+            return self._str_slice(start=key.start, stop=key.stop, step=key.step)
+        else:
+            return self._str_get(key)
+
+    @abc.abstractmethod
+    def _str_count(self, pat, flags: int = 0):
+        pass
+
+    @abc.abstractmethod
+    def _str_pad(
+        self,
+        width: int,
+        side: Literal["left", "right", "both"] = "left",
+        fillchar: str = " ",
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_contains(
+        self, pat, case: bool = True, flags: int = 0, na=None, regex: bool = True
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_startswith(self, pat, na=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_endswith(self, pat, na=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_replace(
+        self,
+        pat: str | re.Pattern,
+        repl: str | Callable,
+        n: int = -1,
+        case: bool = True,
+        flags: int = 0,
+        regex: bool = True,
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_repeat(self, repeats: int | Sequence[int]):
+        pass
+
+    @abc.abstractmethod
+    def _str_match(
+        self,
+        pat: str,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_fullmatch(
+        self,
+        pat: str | re.Pattern,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_encode(self, encoding, errors: str = "strict"):
+        pass
+
+    @abc.abstractmethod
+    def _str_find(self, sub, start: int = 0, end=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_rfind(self, sub, start: int = 0, end=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_findall(self, pat, flags: int = 0):
+        pass
+
+    @abc.abstractmethod
+    def _str_get(self, i):
+        pass
+
+    @abc.abstractmethod
+    def _str_index(self, sub, start: int = 0, end=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_rindex(self, sub, start: int = 0, end=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_join(self, sep: str):
+        pass
+
+    @abc.abstractmethod
+    def _str_partition(self, sep: str, expand):
+        pass
+
+    @abc.abstractmethod
+    def _str_rpartition(self, sep: str, expand):
+        pass
+
+    @abc.abstractmethod
+    def _str_len(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_slice(self, start=None, stop=None, step=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_slice_replace(self, start=None, stop=None, repl=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_translate(self, table):
+        pass
+
+    @abc.abstractmethod
+    def _str_wrap(self, width: int, **kwargs):
+        pass
+
+    @abc.abstractmethod
+    def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
+        pass
+
+    @abc.abstractmethod
+    def _str_isalnum(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isalpha(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isascii(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isdecimal(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isdigit(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_islower(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isnumeric(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isspace(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_istitle(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_isupper(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_capitalize(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_casefold(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_title(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_swapcase(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_lower(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_upper(self):
+        pass
+
+    @abc.abstractmethod
+    def _str_normalize(self, form):
+        pass
+
+    @abc.abstractmethod
+    def _str_strip(self, to_strip=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_lstrip(self, to_strip=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_rstrip(self, to_strip=None):
+        pass
+
+    @abc.abstractmethod
+    def _str_removeprefix(self, prefix: str) -> Self:
+        pass
+
+    @abc.abstractmethod
+    def _str_removesuffix(self, suffix: str) -> Self:
+        pass
+
+    @abc.abstractmethod
+    def _str_split(
+        self, pat=None, n=-1, expand: bool = False, regex: bool | None = None
+    ):
+        pass
+
+    @abc.abstractmethod
+    def _str_rsplit(self, pat=None, n=-1):
+        pass
+
+    @abc.abstractmethod
+    def _str_extract(self, pat: str, flags: int = 0, expand: bool = True):
+        pass
diff --git a/pandas/pandas/core/strings/object_array.py b/pandas/pandas/core/strings/object_array.py
new file mode 100644
index 0000000000000..0adb7b51cf2b7
--- /dev/null
+++ b/pandas/pandas/core/strings/object_array.py
@@ -0,0 +1,539 @@
+from __future__ import annotations
+
+import functools
+import re
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+import unicodedata
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+import pandas._libs.missing as libmissing
+import pandas._libs.ops as libops
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.strings.base import BaseStringArrayMethods
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        NpDtype,
+        Scalar,
+    )
+
+
+class ObjectStringArrayMixin(BaseStringArrayMethods):
+    """
+    String Methods operating on object-dtype ndarrays.
+    """
+
+    def __len__(self) -> int:
+        # For typing, _str_map relies on the object being sized.
+        raise NotImplementedError
+
+    def _str_map(
+        self,
+        f,
+        na_value=lib.no_default,
+        dtype: NpDtype | None = None,
+        convert: bool = True,
+    ):
+        """
+        Map a callable over valid elements of the array.
+
+        Parameters
+        ----------
+        f : Callable
+            A function to call on each non-NA element.
+        na_value : Scalar, optional
+            The value to set for NA values. Might also be used for the
+            fill value if the callable `f` raises an exception.
+            This defaults to ``self.dtype.na_value`` which is ``np.nan``
+            for object-dtype and Categorical and ``pd.NA`` for StringArray.
+        dtype : Dtype, optional
+            The dtype of the result array.
+        convert : bool, default True
+            Whether to call `maybe_convert_objects` on the resulting ndarray
+        """
+        if dtype is None:
+            dtype = np.dtype("object")
+        if na_value is lib.no_default:
+            na_value = self.dtype.na_value  # type: ignore[attr-defined]
+
+        if not len(self):
+            return np.array([], dtype=dtype)
+
+        arr = np.asarray(self, dtype=object)
+        mask = isna(arr)
+        map_convert = convert and not np.all(mask)
+        try:
+            result = lib.map_infer_mask(
+                arr, f, mask.view(np.uint8), convert=map_convert
+            )
+        except (TypeError, AttributeError) as err:
+            # Reraise the exception if callable `f` got wrong number of args.
+            # The user may want to be warned by this, instead of getting NaN
+            p_err = (
+                r"((takes)|(missing)) (?(2)from \d+ to )?\d+ "
+                r"(?(3)required )positional arguments?"
+            )
+
+            if len(err.args) >= 1 and re.search(p_err, err.args[0]):
+                # FIXME: this should be totally avoidable
+                raise err
+
+            def g(x):
+                # This type of fallback behavior can be removed once
+                # we remove object-dtype .str accessor.
+                try:
+                    return f(x)
+                except (TypeError, AttributeError):
+                    return na_value
+
+            return self._str_map(g, na_value=na_value, dtype=dtype)
+        if not isinstance(result, np.ndarray):
+            return result
+        if na_value is not np.nan:
+            np.putmask(result, mask, na_value)
+            if convert and result.dtype == object:
+                result = lib.maybe_convert_objects(result)
+        return result
+
+    def _str_count(self, pat, flags: int = 0):
+        regex = re.compile(pat, flags=flags)
+        f = lambda x: len(regex.findall(x))
+        return self._str_map(f, dtype="int64")
+
+    def _str_pad(
+        self,
+        width: int,
+        side: Literal["left", "right", "both"] = "left",
+        fillchar: str = " ",
+    ):
+        if side == "left":
+            f = lambda x: x.rjust(width, fillchar)
+        elif side == "right":
+            f = lambda x: x.ljust(width, fillchar)
+        elif side == "both":
+            f = lambda x: x.center(width, fillchar)
+        else:  # pragma: no cover
+            raise ValueError("Invalid side")
+        return self._str_map(f)
+
+    def _str_contains(
+        self,
+        pat,
+        case: bool = True,
+        flags: int = 0,
+        na=lib.no_default,
+        regex: bool = True,
+    ):
+        if regex:
+            if not case:
+                flags |= re.IGNORECASE
+
+            pat = re.compile(pat, flags=flags)
+
+            f = lambda x: pat.search(x) is not None
+        else:
+            if case:
+                f = lambda x: pat in x
+            else:
+                upper_pat = pat.upper()
+                f = lambda x: upper_pat in x.upper()
+        if na is not lib.no_default and not isna(na) and not isinstance(na, bool):
+            # GH#59561
+            warnings.warn(
+                "Allowing a non-bool 'na' in obj.str.contains is deprecated "
+                "and will raise in a future version.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        return self._str_map(f, na, dtype=np.dtype("bool"))
+
+    def _str_startswith(self, pat, na=lib.no_default):
+        f = lambda x: x.startswith(pat)
+        if na is not lib.no_default and not isna(na) and not isinstance(na, bool):
+            # GH#59561
+            warnings.warn(
+                "Allowing a non-bool 'na' in obj.str.startswith is deprecated "
+                "and will raise in a future version.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        return self._str_map(f, na_value=na, dtype=np.dtype(bool))
+
+    def _str_endswith(self, pat, na=lib.no_default):
+        f = lambda x: x.endswith(pat)
+        if na is not lib.no_default and not isna(na) and not isinstance(na, bool):
+            # GH#59561
+            warnings.warn(
+                "Allowing a non-bool 'na' in obj.str.endswith is deprecated "
+                "and will raise in a future version.",
+                FutureWarning,
+                stacklevel=find_stack_level(),
+            )
+        return self._str_map(f, na_value=na, dtype=np.dtype(bool))
+
+    def _str_replace(
+        self,
+        pat: str | re.Pattern,
+        repl: str | Callable,
+        n: int = -1,
+        case: bool = True,
+        flags: int = 0,
+        regex: bool = True,
+    ):
+        if case is False:
+            # add case flag, if provided
+            flags |= re.IGNORECASE
+
+        if regex or flags or callable(repl):
+            if not isinstance(pat, re.Pattern):
+                if regex is False:
+                    pat = re.escape(pat)
+                pat = re.compile(pat, flags=flags)
+
+            n = n if n >= 0 else 0
+            f = lambda x: pat.sub(repl=repl, string=x, count=n)
+        else:
+            f = lambda x: x.replace(pat, repl, n)
+
+        return self._str_map(f, dtype=str)
+
+    def _str_repeat(self, repeats: int | Sequence[int]):
+        if lib.is_integer(repeats):
+            rint = cast(int, repeats)
+
+            def scalar_rep(x):
+                try:
+                    return bytes.__mul__(x, rint)
+                except TypeError:
+                    return str.__mul__(x, rint)
+
+            return self._str_map(scalar_rep, dtype=str)
+        else:
+            from pandas.core.arrays.string_ import BaseStringArray
+
+            def rep(x, r):
+                if x is libmissing.NA:
+                    return x
+                try:
+                    return bytes.__mul__(x, r)
+                except TypeError:
+                    return str.__mul__(x, r)
+
+            result = libops.vec_binop(
+                np.asarray(self),
+                np.asarray(repeats, dtype=object),
+                rep,
+            )
+            if not isinstance(self, BaseStringArray):
+                return result
+            # Not going through map, so we have to do this here.
+            return type(self)._from_sequence(result, dtype=self.dtype)
+
+    def _str_match(
+        self,
+        pat: str,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        if not case:
+            flags |= re.IGNORECASE
+
+        regex = re.compile(pat, flags=flags)
+
+        f = lambda x: regex.match(x) is not None
+        return self._str_map(f, na_value=na, dtype=np.dtype(bool))
+
+    def _str_fullmatch(
+        self,
+        pat: str | re.Pattern,
+        case: bool = True,
+        flags: int = 0,
+        na: Scalar | lib.NoDefault = lib.no_default,
+    ):
+        if not case:
+            flags |= re.IGNORECASE
+
+        regex = re.compile(pat, flags=flags)
+
+        f = lambda x: regex.fullmatch(x) is not None
+        return self._str_map(f, na_value=na, dtype=np.dtype(bool))
+
+    def _str_encode(self, encoding, errors: str = "strict"):
+        f = lambda x: x.encode(encoding, errors=errors)
+        return self._str_map(f, dtype=object)
+
+    def _str_find(self, sub, start: int = 0, end=None):
+        return self._str_find_(sub, start, end, side="left")
+
+    def _str_rfind(self, sub, start: int = 0, end=None):
+        return self._str_find_(sub, start, end, side="right")
+
+    def _str_find_(self, sub, start, end, side):
+        if side == "left":
+            method = "find"
+        elif side == "right":
+            method = "rfind"
+        else:  # pragma: no cover
+            raise ValueError("Invalid side")
+
+        if end is None:
+            f = lambda x: getattr(x, method)(sub, start)
+        else:
+            f = lambda x: getattr(x, method)(sub, start, end)
+        return self._str_map(f, dtype="int64")
+
+    def _str_findall(self, pat, flags: int = 0):
+        regex = re.compile(pat, flags=flags)
+        return self._str_map(regex.findall, dtype="object")
+
+    def _str_get(self, i):
+        def f(x):
+            if isinstance(x, dict):
+                return x.get(i)
+            elif len(x) > i >= -len(x):
+                return x[i]
+            return self.dtype.na_value  # type: ignore[attr-defined]
+
+        return self._str_map(f)
+
+    def _str_index(self, sub, start: int = 0, end=None):
+        if end:
+            f = lambda x: x.index(sub, start, end)
+        else:
+            f = lambda x: x.index(sub, start, end)
+        return self._str_map(f, dtype="int64")
+
+    def _str_rindex(self, sub, start: int = 0, end=None):
+        if end:
+            f = lambda x: x.rindex(sub, start, end)
+        else:
+            f = lambda x: x.rindex(sub, start, end)
+        return self._str_map(f, dtype="int64")
+
+    def _str_join(self, sep: str):
+        return self._str_map(sep.join)
+
+    def _str_partition(self, sep: str, expand):
+        result = self._str_map(lambda x: x.partition(sep), dtype="object")
+        return result
+
+    def _str_rpartition(self, sep: str, expand):
+        return self._str_map(lambda x: x.rpartition(sep), dtype="object")
+
+    def _str_len(self):
+        return self._str_map(len, dtype="int64")
+
+    def _str_slice(self, start=None, stop=None, step=None):
+        obj = slice(start, stop, step)
+        return self._str_map(lambda x: x[obj])
+
+    def _str_slice_replace(self, start=None, stop=None, repl=None):
+        if repl is None:
+            repl = ""
+
+        def f(x):
+            if x[start:stop] == "":
+                local_stop = start
+            else:
+                local_stop = stop
+            y = ""
+            if start is not None:
+                y += x[:start]
+            y += repl
+            if stop is not None:
+                y += x[local_stop:]
+            return y
+
+        return self._str_map(f)
+
+    def _str_split(
+        self,
+        pat: str | re.Pattern | None = None,
+        n=-1,
+        expand: bool = False,
+        regex: bool | None = None,
+    ):
+        if pat is None:
+            if n is None or n == 0:
+                n = -1
+            f = lambda x: x.split(pat, n)
+        else:
+            new_pat: str | re.Pattern
+            if regex is True or isinstance(pat, re.Pattern):
+                new_pat = re.compile(pat)
+            elif regex is False:
+                new_pat = pat
+            # regex is None so link to old behavior #43563
+            else:
+                if len(pat) == 1:
+                    new_pat = pat
+                else:
+                    new_pat = re.compile(pat)
+
+            if isinstance(new_pat, re.Pattern):
+                if n is None or n == -1:
+                    n = 0
+                f = lambda x: new_pat.split(x, maxsplit=n)
+            else:
+                if n is None or n == 0:
+                    n = -1
+                f = lambda x: x.split(pat, n)
+        return self._str_map(f, dtype=object)
+
+    def _str_rsplit(self, pat=None, n=-1):
+        if n is None or n == 0:
+            n = -1
+        f = lambda x: x.rsplit(pat, n)
+        return self._str_map(f, dtype="object")
+
+    def _str_translate(self, table):
+        return self._str_map(lambda x: x.translate(table))
+
+    def _str_wrap(self, width: int, **kwargs):
+        kwargs["width"] = width
+        tw = textwrap.TextWrapper(**kwargs)
+        return self._str_map(lambda s: "\n".join(tw.wrap(s)))
+
+    def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
+        from pandas import Series
+
+        if dtype is None:
+            dtype = np.int64
+        arr = Series(self).fillna("")
+        try:
+            arr = sep + arr + sep
+        except (TypeError, NotImplementedError):
+            arr = sep + arr.astype(str) + sep
+
+        tags: set[str] = set()
+        for ts in Series(arr, copy=False).str.split(sep):
+            tags.update(ts)
+        tags2 = sorted(tags - {""})
+
+        _dtype = pandas_dtype(dtype)
+        dummies_dtype: NpDtype
+        if isinstance(_dtype, np.dtype):
+            dummies_dtype = _dtype
+        else:
+            dummies_dtype = np.bool_
+        dummies = np.empty((len(arr), len(tags2)), dtype=dummies_dtype, order="F")
+
+        def _isin(test_elements: str, element: str) -> bool:
+            return element in test_elements
+
+        for i, t in enumerate(tags2):
+            pat = sep + t + sep
+            dummies[:, i] = lib.map_infer(
+                arr.to_numpy(), functools.partial(_isin, element=pat)
+            )
+        return dummies, tags2
+
+    def _str_upper(self):
+        return self._str_map(lambda x: x.upper())
+
+    def _str_isalnum(self):
+        return self._str_map(str.isalnum, dtype="bool")
+
+    def _str_isalpha(self):
+        return self._str_map(str.isalpha, dtype="bool")
+
+    def _str_isascii(self):
+        return self._str_map(str.isascii, dtype="bool")
+
+    def _str_isdecimal(self):
+        return self._str_map(str.isdecimal, dtype="bool")
+
+    def _str_isdigit(self):
+        return self._str_map(str.isdigit, dtype="bool")
+
+    def _str_islower(self):
+        return self._str_map(str.islower, dtype="bool")
+
+    def _str_isnumeric(self):
+        return self._str_map(str.isnumeric, dtype="bool")
+
+    def _str_isspace(self):
+        return self._str_map(str.isspace, dtype="bool")
+
+    def _str_istitle(self):
+        return self._str_map(str.istitle, dtype="bool")
+
+    def _str_isupper(self):
+        return self._str_map(str.isupper, dtype="bool")
+
+    def _str_capitalize(self):
+        return self._str_map(str.capitalize)
+
+    def _str_casefold(self):
+        return self._str_map(str.casefold)
+
+    def _str_title(self):
+        return self._str_map(str.title)
+
+    def _str_swapcase(self):
+        return self._str_map(str.swapcase)
+
+    def _str_lower(self):
+        return self._str_map(str.lower)
+
+    def _str_normalize(self, form):
+        f = lambda x: unicodedata.normalize(form, x)
+        return self._str_map(f)
+
+    def _str_strip(self, to_strip=None):
+        return self._str_map(lambda x: x.strip(to_strip))
+
+    def _str_lstrip(self, to_strip=None):
+        return self._str_map(lambda x: x.lstrip(to_strip))
+
+    def _str_rstrip(self, to_strip=None):
+        return self._str_map(lambda x: x.rstrip(to_strip))
+
+    def _str_removeprefix(self, prefix: str):
+        return self._str_map(lambda x: x.removeprefix(prefix))
+
+    def _str_removesuffix(self, suffix: str):
+        return self._str_map(lambda x: x.removesuffix(suffix))
+
+    def _str_extract(self, pat: str, flags: int = 0, expand: bool = True):
+        regex = re.compile(pat, flags=flags)
+        na_value = self.dtype.na_value  # type: ignore[attr-defined]
+
+        if not expand:
+
+            def g(x):
+                m = regex.search(x)
+                return m.groups()[0] if m else na_value
+
+            return self._str_map(g, convert=False)
+
+        empty_row = [na_value] * regex.groups
+
+        def f(x):
+            if not isinstance(x, str):
+                return empty_row
+            m = regex.search(x)
+            if m:
+                return [na_value if item is None else item for item in m.groups()]
+            else:
+                return empty_row
+
+        return [f(val) for val in np.asarray(self)]
diff --git a/pandas/pandas/core/tools/__init__.py b/pandas/pandas/core/tools/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/tools/datetimes.py b/pandas/pandas/core/tools/datetimes.py
new file mode 100644
index 0000000000000..0a10001a3113f
--- /dev/null
+++ b/pandas/pandas/core/tools/datetimes.py
@@ -0,0 +1,1193 @@
+from __future__ import annotations
+
+from collections import abc
+from datetime import date
+from functools import partial
+from itertools import islice
+from typing import (
+    TYPE_CHECKING,
+    TypedDict,
+    Union,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    tslib,
+)
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    Timedelta,
+    Timestamp,
+    astype_overflowsafe,
+    is_supported_dtype,
+    timezones as libtimezones,
+)
+from pandas._libs.tslibs.conversion import cast_from_unit_vectorized
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas._libs.tslibs.parsing import (
+    DateParseError,
+    guess_datetime_format,
+)
+from pandas._libs.tslibs.strptime import array_strptime
+from pandas._typing import (
+    AnyArrayLike,
+    ArrayLike,
+    DateTimeErrorChoices,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_float,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_numeric_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    DatetimeTZDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+from pandas.arrays import (
+    DatetimeArray,
+    IntegerArray,
+    NumpyExtensionArray,
+)
+from pandas.core.algorithms import unique
+from pandas.core.arrays import ArrowExtensionArray
+from pandas.core.arrays.base import ExtensionArray
+from pandas.core.arrays.datetimes import (
+    maybe_convert_dtype,
+    objects_to_datetime64,
+    tz_to_dtype,
+)
+from pandas.core.construction import extract_array
+from pandas.core.indexes.base import Index
+from pandas.core.indexes.datetimes import DatetimeIndex
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+    )
+
+    from pandas._libs.tslibs.nattype import NaTType
+    from pandas._libs.tslibs.timedeltas import UnitChoices
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+# ---------------------------------------------------------------------
+# types used in annotations
+
+ArrayConvertible = Union[list, tuple, AnyArrayLike]
+Scalar = Union[float, str]
+DatetimeScalar = Union[Scalar, date, np.datetime64]
+
+DatetimeScalarOrArrayConvertible = Union[DatetimeScalar, ArrayConvertible]
+
+DatetimeDictArg = Union[list[Scalar], tuple[Scalar, ...], AnyArrayLike]
+
+
+class YearMonthDayDict(TypedDict, total=True):
+    year: DatetimeDictArg
+    month: DatetimeDictArg
+    day: DatetimeDictArg
+
+
+class FulldatetimeDict(YearMonthDayDict, total=False):
+    hour: DatetimeDictArg
+    hours: DatetimeDictArg
+    minute: DatetimeDictArg
+    minutes: DatetimeDictArg
+    second: DatetimeDictArg
+    seconds: DatetimeDictArg
+    ms: DatetimeDictArg
+    us: DatetimeDictArg
+    ns: DatetimeDictArg
+
+
+DictConvertible = Union[FulldatetimeDict, "DataFrame"]
+start_caching_at = 50
+
+
+# ---------------------------------------------------------------------
+
+
+def _guess_datetime_format_for_array(arr, dayfirst: bool | None = False) -> str | None:
+    # Try to guess the format based on the first non-NaN element, return None if can't
+    if (first_non_null := tslib.first_non_null(arr)) != -1:
+        if type(first_non_nan_element := arr[first_non_null]) is str:
+            # GH#32264 np.str_ object
+            guessed_format = guess_datetime_format(
+                first_non_nan_element, dayfirst=dayfirst
+            )
+            if guessed_format is not None:
+                return guessed_format
+            # If there are multiple non-null elements, warn about
+            # how parsing might not be consistent
+            if tslib.first_non_null(arr[first_non_null + 1 :]) != -1:
+                warnings.warn(
+                    "Could not infer format, so each element will be parsed "
+                    "individually, falling back to `dateutil`. To ensure parsing is "
+                    "consistent and as-expected, please specify a format.",
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+    return None
+
+
+def should_cache(
+    arg: ArrayConvertible, unique_share: float = 0.7, check_count: int | None = None
+) -> bool:
+    """
+    Decides whether to do caching.
+
+    If the percent of unique elements among `check_count` elements less
+    than `unique_share * 100` then we can do caching.
+
+    Parameters
+    ----------
+    arg: listlike, tuple, 1-d array, Series
+    unique_share: float, default=0.7, optional
+        0 < unique_share < 1
+    check_count: int, optional
+        0 <= check_count <= len(arg)
+
+    Returns
+    -------
+    do_caching: bool
+
+    Notes
+    -----
+    By default for a sequence of less than 50 items in size, we don't do
+    caching; for the number of elements less than 5000, we take ten percent of
+    all elements to check for a uniqueness share; if the sequence size is more
+    than 5000, then we check only the first 500 elements.
+    All constants were chosen empirically by.
+    """
+    do_caching = True
+
+    # default realization
+    if check_count is None:
+        # in this case, the gain from caching is negligible
+        if len(arg) <= start_caching_at:
+            return False
+
+        if len(arg) <= 5000:
+            check_count = len(arg) // 10
+        else:
+            check_count = 500
+    else:
+        assert 0 <= check_count <= len(arg), (
+            "check_count must be in next bounds: [0; len(arg)]"
+        )
+        if check_count == 0:
+            return False
+
+    assert 0 < unique_share < 1, "unique_share must be in next bounds: (0; 1)"
+
+    try:
+        # We can't cache if the items are not hashable.
+        unique_elements = set(islice(arg, check_count))
+    except TypeError:
+        return False
+    if len(unique_elements) > check_count * unique_share:
+        do_caching = False
+    return do_caching
+
+
+def _maybe_cache(
+    arg: ArrayConvertible,
+    format: str | None,
+    cache: bool,
+    convert_listlike: Callable,
+) -> Series:
+    """
+    Create a cache of unique dates from an array of dates
+
+    Parameters
+    ----------
+    arg : listlike, tuple, 1-d array, Series
+    format : string
+        Strftime format to parse time
+    cache : bool
+        True attempts to create a cache of converted values
+    convert_listlike : function
+        Conversion function to apply on dates
+
+    Returns
+    -------
+    cache_array : Series
+        Cache of converted, unique dates. Can be empty
+    """
+    from pandas import Series
+
+    cache_array = Series(dtype=object)
+
+    if cache:
+        # Perform a quicker unique check
+        if not should_cache(arg):
+            return cache_array
+
+        if not isinstance(arg, (np.ndarray, ExtensionArray, Index, ABCSeries)):
+            arg = np.array(arg)
+
+        unique_dates = unique(arg)
+        if len(unique_dates) < len(arg):
+            cache_dates = convert_listlike(unique_dates, format)
+            # GH#45319
+            try:
+                cache_array = Series(cache_dates, index=unique_dates, copy=False)
+            except OutOfBoundsDatetime:
+                return cache_array
+            # GH#39882 and GH#35888 in case of None and NaT we get duplicates
+            if not cache_array.index.is_unique:
+                cache_array = cache_array[~cache_array.index.duplicated()]
+    return cache_array
+
+
+def _box_as_indexlike(
+    dt_array: ArrayLike, utc: bool = False, name: Hashable | None = None
+) -> Index:
+    """
+    Properly boxes the ndarray of datetimes to DatetimeIndex
+    if it is possible or to generic Index instead
+
+    Parameters
+    ----------
+    dt_array: 1-d array
+        Array of datetimes to be wrapped in an Index.
+    utc : bool
+        Whether to convert/localize timestamps to UTC.
+    name : string, default None
+        Name for a resulting index
+
+    Returns
+    -------
+    result : datetime of converted dates
+        - DatetimeIndex if convertible to sole datetime64 type
+        - general Index otherwise
+    """
+
+    if lib.is_np_dtype(dt_array.dtype, "M"):
+        tz = "utc" if utc else None
+        return DatetimeIndex(dt_array, tz=tz, name=name)
+    return Index(dt_array, name=name, dtype=dt_array.dtype)
+
+
+def _convert_and_box_cache(
+    arg: DatetimeScalarOrArrayConvertible,
+    cache_array: Series,
+    name: Hashable | None = None,
+) -> Index:
+    """
+    Convert array of dates with a cache and wrap the result in an Index.
+
+    Parameters
+    ----------
+    arg : integer, float, string, datetime, list, tuple, 1-d array, Series
+    cache_array : Series
+        Cache of converted, unique dates
+    name : string, default None
+        Name for a DatetimeIndex
+
+    Returns
+    -------
+    result : Index-like of converted dates
+    """
+    from pandas import Series
+
+    result = Series(arg, dtype=cache_array.index.dtype).map(cache_array)
+    return _box_as_indexlike(result._values, utc=False, name=name)
+
+
+def _convert_listlike_datetimes(
+    arg,
+    format: str | None,
+    name: Hashable | None = None,
+    utc: bool = False,
+    unit: str | None = None,
+    errors: DateTimeErrorChoices = "raise",
+    dayfirst: bool | None = None,
+    yearfirst: bool | None = None,
+    exact: bool = True,
+):
+    """
+    Helper function for to_datetime. Performs the conversions of 1D listlike
+    of dates
+
+    Parameters
+    ----------
+    arg : list, tuple, ndarray, Series, Index
+        date to be parsed
+    name : object
+        None or string for the Index name
+    utc : bool
+        Whether to convert/localize timestamps to UTC.
+    unit : str
+        None or string of the frequency of the passed data
+    errors : str
+        error handing behaviors from to_datetime, 'raise', 'coerce'
+    dayfirst : bool
+        dayfirst parsing behavior from to_datetime
+    yearfirst : bool
+        yearfirst parsing behavior from to_datetime
+    exact : bool, default True
+        exact format matching behavior from to_datetime
+
+    Returns
+    -------
+    Index-like of parsed dates
+    """
+    if isinstance(arg, (list, tuple)):
+        arg = np.array(arg, dtype="O")
+    elif isinstance(arg, NumpyExtensionArray):
+        arg = np.array(arg)
+
+    arg_dtype = getattr(arg, "dtype", None)
+    # these are shortcutable
+    tz = "utc" if utc else None
+    if isinstance(arg_dtype, DatetimeTZDtype):
+        if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
+            return DatetimeIndex(arg, tz=tz, name=name)
+        if utc:
+            arg = arg.tz_convert(None).tz_localize("utc")
+        return arg
+
+    elif isinstance(arg_dtype, ArrowDtype) and arg_dtype.type is Timestamp:
+        # TODO: Combine with above if DTI/DTA supports Arrow timestamps
+        if utc:
+            # pyarrow uses UTC, not lowercase utc
+            if isinstance(arg, Index):
+                arg_array = cast(ArrowExtensionArray, arg.array)
+                if arg_dtype.pyarrow_dtype.tz is not None:
+                    arg_array = arg_array._dt_tz_convert("UTC")
+                else:
+                    arg_array = arg_array._dt_tz_localize("UTC")
+                arg = Index(arg_array)
+            else:
+                # ArrowExtensionArray
+                if arg_dtype.pyarrow_dtype.tz is not None:
+                    arg = arg._dt_tz_convert("UTC")
+                else:
+                    arg = arg._dt_tz_localize("UTC")
+        return arg
+
+    elif lib.is_np_dtype(arg_dtype, "M"):
+        if not is_supported_dtype(arg_dtype):
+            # We go to closest supported reso, i.e. "s"
+            arg = astype_overflowsafe(
+                np.asarray(arg),
+                np.dtype("M8[s]"),
+                is_coerce=errors == "coerce",
+            )
+
+        if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
+            return DatetimeIndex(arg, tz=tz, name=name)
+        elif utc:
+            # DatetimeArray, DatetimeIndex
+            return arg.tz_localize("utc")
+
+        return arg
+
+    elif unit is not None:
+        if format is not None:
+            raise ValueError("cannot specify both format and unit")
+        return _to_datetime_with_unit(arg, unit, name, utc, errors)
+    elif getattr(arg, "ndim", 1) > 1:
+        raise TypeError(
+            "arg must be a string, datetime, list, tuple, 1-d array, or Series"
+        )
+
+    # warn if passing timedelta64, raise for PeriodDtype
+    # NB: this must come after unit transformation
+    try:
+        arg, _ = maybe_convert_dtype(arg, copy=False, tz=libtimezones.maybe_get_tz(tz))
+    except TypeError:
+        if errors == "coerce":
+            npvalues = np.full(len(arg), np.datetime64("NaT", "ns"))
+            return DatetimeIndex(npvalues, name=name)
+        raise
+
+    arg = ensure_object(arg)
+
+    if format is None:
+        format = _guess_datetime_format_for_array(arg, dayfirst=dayfirst)
+
+    # `format` could be inferred, or user didn't ask for mixed-format parsing.
+    if format is not None and format != "mixed":
+        return _array_strptime_with_fallback(arg, name, utc, format, exact, errors)
+
+    result, tz_parsed = objects_to_datetime64(
+        arg,
+        dayfirst=dayfirst,
+        yearfirst=yearfirst,
+        utc=utc,
+        errors=errors,
+        allow_object=True,
+    )
+
+    if tz_parsed is not None:
+        # We can take a shortcut since the datetime64 numpy array
+        # is in UTC
+        out_unit = np.datetime_data(result.dtype)[0]
+        dtype = tz_to_dtype(tz_parsed, out_unit)
+        dt64_values = result.view(f"M8[{dtype.unit}]")
+        dta = DatetimeArray._simple_new(dt64_values, dtype=dtype)
+        return DatetimeIndex._simple_new(dta, name=name)
+
+    return _box_as_indexlike(result, utc=utc, name=name)
+
+
+def _array_strptime_with_fallback(
+    arg,
+    name,
+    utc: bool,
+    fmt: str,
+    exact: bool,
+    errors: str,
+) -> Index:
+    """
+    Call array_strptime, with fallback behavior depending on 'errors'.
+    """
+    result, tz_out = array_strptime(arg, fmt, exact=exact, errors=errors, utc=utc)
+    if tz_out is not None:
+        unit = np.datetime_data(result.dtype)[0]
+        dtype = DatetimeTZDtype(tz=tz_out, unit=unit)
+        dta = DatetimeArray._simple_new(result, dtype=dtype)
+        if utc:
+            dta = dta.tz_convert("UTC")
+        return Index(dta, name=name)
+    elif result.dtype != object and utc:
+        unit = np.datetime_data(result.dtype)[0]
+        res = Index(result, dtype=f"M8[{unit}, UTC]", name=name)
+        return res
+    return Index(result, dtype=result.dtype, name=name)
+
+
+def _to_datetime_with_unit(arg, unit, name, utc: bool, errors: str) -> Index:
+    """
+    to_datetime specalized to the case where a 'unit' is passed.
+    """
+    arg = extract_array(arg, extract_numpy=True)
+
+    # GH#30050 pass an ndarray to tslib.array_to_datetime
+    # because it expects an ndarray argument
+    if isinstance(arg, IntegerArray):
+        arr = arg.astype(f"datetime64[{unit}]")
+        tz_parsed = None
+    else:
+        arg = np.asarray(arg)
+
+        if arg.dtype.kind in "iu":
+            # Note we can't do "f" here because that could induce unwanted
+            #  rounding GH#14156, GH#20445
+            arr = arg.astype(f"datetime64[{unit}]", copy=False)
+            try:
+                arr = astype_overflowsafe(arr, np.dtype("M8[ns]"), copy=False)
+            except OutOfBoundsDatetime:
+                if errors == "raise":
+                    raise
+                arg = arg.astype(object)
+                return _to_datetime_with_unit(arg, unit, name, utc, errors)
+            tz_parsed = None
+
+        elif arg.dtype.kind == "f":
+            with np.errstate(over="raise"):
+                try:
+                    arr = cast_from_unit_vectorized(arg, unit=unit)
+                except OutOfBoundsDatetime as err:
+                    if errors != "raise":
+                        return _to_datetime_with_unit(
+                            arg.astype(object), unit, name, utc, errors
+                        )
+                    raise OutOfBoundsDatetime(
+                        f"cannot convert input with unit '{unit}'"
+                    ) from err
+
+            arr = arr.view("M8[ns]")
+            tz_parsed = None
+        else:
+            arg = arg.astype(object, copy=False)
+            arr, tz_parsed = tslib.array_to_datetime(
+                arg,
+                utc=utc,
+                errors=errors,
+                unit_for_numerics=unit,
+                creso=NpyDatetimeUnit.NPY_FR_ns.value,
+            )
+
+    result = DatetimeIndex(arr, name=name)
+    if not isinstance(result, DatetimeIndex):
+        return result
+
+    # GH#23758: We may still need to localize the result with tz
+    # GH#25546: Apply tz_parsed first (from arg), then tz (from caller)
+    # result will be naive but in UTC
+    result = result.tz_localize("UTC").tz_convert(tz_parsed)
+
+    if utc:
+        if result.tz is None:
+            result = result.tz_localize("utc")
+        else:
+            result = result.tz_convert("utc")
+    return result
+
+
+def _adjust_to_origin(arg, origin, unit):
+    """
+    Helper function for to_datetime.
+    Adjust input argument to the specified origin
+
+    Parameters
+    ----------
+    arg : list, tuple, ndarray, Series, Index
+        date to be adjusted
+    origin : 'julian' or Timestamp
+        origin offset for the arg
+    unit : str
+        passed unit from to_datetime, must be 'D'
+
+    Returns
+    -------
+    ndarray or scalar of adjusted date(s)
+    """
+    if origin == "julian":
+        original = arg
+        j0 = Timestamp(0).to_julian_date()
+        if unit != "D":
+            raise ValueError("unit must be 'D' for origin='julian'")
+        try:
+            arg = arg - j0
+        except TypeError as err:
+            raise ValueError(
+                "incompatible 'arg' type for given 'origin'='julian'"
+            ) from err
+
+        # preemptively check this for a nice range
+        j_max = Timestamp.max.to_julian_date() - j0
+        j_min = Timestamp.min.to_julian_date() - j0
+        if np.any(arg > j_max) or np.any(arg < j_min):
+            raise OutOfBoundsDatetime(
+                f"{original} is Out of Bounds for origin='julian'"
+            )
+    else:
+        # arg must be numeric
+        if not (
+            (is_integer(arg) or is_float(arg)) or is_numeric_dtype(np.asarray(arg))
+        ):
+            raise ValueError(
+                f"'{arg}' is not compatible with origin='{origin}'; "
+                "it must be numeric with a unit specified"
+            )
+
+        # we are going to offset back to unix / epoch time
+        try:
+            offset = Timestamp(origin, unit=unit)
+        except OutOfBoundsDatetime as err:
+            raise OutOfBoundsDatetime(f"origin {origin} is Out of Bounds") from err
+        except ValueError as err:
+            raise ValueError(
+                f"origin {origin} cannot be converted to a Timestamp"
+            ) from err
+
+        if offset.tz is not None:
+            raise ValueError(f"origin offset {offset} must be tz-naive")
+        td_offset = offset - Timestamp(0)
+
+        # convert the offset to the unit of the arg
+        # this should be lossless in terms of precision
+        ioffset = td_offset // Timedelta(1, unit=unit)
+
+        # scalars & ndarray-like can handle the addition
+        if is_list_like(arg) and not isinstance(arg, (ABCSeries, Index, np.ndarray)):
+            arg = np.asarray(arg)
+        arg = arg + ioffset
+    return arg
+
+
+@overload
+def to_datetime(
+    arg: DatetimeScalar,
+    errors: DateTimeErrorChoices = ...,
+    dayfirst: bool = ...,
+    yearfirst: bool = ...,
+    utc: bool = ...,
+    format: str | None = ...,
+    exact: bool = ...,
+    unit: str | None = ...,
+    origin=...,
+    cache: bool = ...,
+) -> Timestamp: ...
+
+
+@overload
+def to_datetime(
+    arg: Series | DictConvertible,
+    errors: DateTimeErrorChoices = ...,
+    dayfirst: bool = ...,
+    yearfirst: bool = ...,
+    utc: bool = ...,
+    format: str | None = ...,
+    exact: bool = ...,
+    unit: str | None = ...,
+    origin=...,
+    cache: bool = ...,
+) -> Series: ...
+
+
+@overload
+def to_datetime(
+    arg: list | tuple | Index | ArrayLike,
+    errors: DateTimeErrorChoices = ...,
+    dayfirst: bool = ...,
+    yearfirst: bool = ...,
+    utc: bool = ...,
+    format: str | None = ...,
+    exact: bool = ...,
+    unit: str | None = ...,
+    origin=...,
+    cache: bool = ...,
+) -> DatetimeIndex: ...
+
+
+def to_datetime(
+    arg: DatetimeScalarOrArrayConvertible | DictConvertible,
+    errors: DateTimeErrorChoices = "raise",
+    dayfirst: bool = False,
+    yearfirst: bool = False,
+    utc: bool = False,
+    format: str | None = None,
+    exact: bool | lib.NoDefault = lib.no_default,
+    unit: str | None = None,
+    origin: str = "unix",
+    cache: bool = True,
+) -> DatetimeIndex | Series | DatetimeScalar | NaTType | None:
+    """
+    Convert argument to datetime.
+
+    This function converts a scalar, array-like, :class:`Series` or
+    :class:`DataFrame`/dict-like to a pandas datetime object.
+
+    Parameters
+    ----------
+    arg : int, float, str, datetime, list, tuple, 1-d array, Series, DataFrame/dict-like
+        The object to convert to a datetime. If a :class:`DataFrame` is provided, the
+        method expects minimally the following columns: :const:`"year"`,
+        :const:`"month"`, :const:`"day"`. The column "year"
+        must be specified in 4-digit format.
+    errors : {'raise', 'coerce'}, default 'raise'
+        - If :const:`'raise'`, then invalid parsing will raise an exception.
+        - If :const:`'coerce'`, then invalid parsing will be set as :const:`NaT`.
+    dayfirst : bool, default False
+        Specify a date parse order if `arg` is str or is list-like.
+        If :const:`True`, parses dates with the day first, e.g. :const:`"10/11/12"`
+        is parsed as :const:`2012-11-10`.
+
+        .. warning::
+
+            ``dayfirst=True`` is not strict, but will prefer to parse
+            with day first.
+
+    yearfirst : bool, default False
+        Specify a date parse order if `arg` is str or is list-like.
+
+        - If :const:`True` parses dates with the year first, e.g.
+          :const:`"10/11/12"` is parsed as :const:`2010-11-12`.
+        - If both `dayfirst` and `yearfirst` are :const:`True`, `yearfirst` is
+          preceded (same as :mod:`dateutil`).
+
+        .. warning::
+
+            ``yearfirst=True`` is not strict, but will prefer to parse
+            with year first.
+
+    utc : bool, default False
+        Control timezone-related parsing, localization and conversion.
+
+        - If :const:`True`, the function *always* returns a timezone-aware
+          UTC-localized :class:`Timestamp`, :class:`Series` or
+          :class:`DatetimeIndex`. To do this, timezone-naive inputs are
+          *localized* as UTC, while timezone-aware inputs are *converted* to UTC.
+
+        - If :const:`False` (default), inputs will not be coerced to UTC.
+          Timezone-naive inputs will remain naive, while timezone-aware ones
+          will keep their time offsets. Limitations exist for mixed
+          offsets (typically, daylight savings), see :ref:`Examples
+          <to_datetime_tz_examples>` section for details.
+
+        See also: pandas general documentation about `timezone conversion and
+        localization
+        <https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html
+        #time-zone-handling>`_.
+
+    format : str, default None
+        The strftime to parse time, e.g. :const:`"%d/%m/%Y"`. See
+        `strftime documentation
+        <https://docs.python.org/3/library/datetime.html
+        #strftime-and-strptime-behavior>`_ for more information on choices, though
+        note that :const:`"%f"` will parse all the way up to nanoseconds.
+        You can also pass:
+
+        - "ISO8601", to parse any `ISO8601 <https://en.wikipedia.org/wiki/ISO_8601>`_
+          time string (not necessarily in exactly the same format);
+        - "mixed", to infer the format for each element individually. This is risky,
+          and you should probably use it along with `dayfirst`.
+
+        .. note::
+
+            If a :class:`DataFrame` is passed, then `format` has no effect.
+
+    exact : bool, default True
+        Control how `format` is used:
+
+        - If :const:`True`, require an exact `format` match.
+        - If :const:`False`, allow the `format` to match anywhere in the target
+          string.
+
+        Cannot be used alongside ``format='ISO8601'`` or ``format='mixed'``.
+    unit : str, default 'ns'
+        The unit of the arg (D,s,ms,us,ns) denote the unit, which is an
+        integer or float number. This will be based off the origin.
+        Example, with ``unit='ms'`` and ``origin='unix'``, this would calculate
+        the number of milliseconds to the unix epoch start.
+    origin : scalar, default 'unix'
+        Define the reference date. The numeric values would be parsed as number
+        of units (defined by `unit`) since this reference date.
+
+        - If :const:`'unix'` (or POSIX) time; origin is set to 1970-01-01.
+        - If :const:`'julian'`, unit must be :const:`'D'`, and origin is set to
+          beginning of Julian Calendar. Julian day number :const:`0` is assigned
+          to the day starting at noon on January 1, 4713 BC.
+        - If Timestamp convertible (Timestamp, dt.datetime, np.datetimt64 or date
+          string), origin is set to Timestamp identified by origin.
+        - If a float or integer, origin is the difference
+          (in units determined by the ``unit`` argument) relative to 1970-01-01.
+    cache : bool, default True
+        If :const:`True`, use a cache of unique, converted dates to apply the
+        datetime conversion. May produce significant speed-up when parsing
+        duplicate date strings, especially ones with timezone offsets. The cache
+        is only used when there are at least 50 values. The presence of
+        out-of-bounds values will render the cache unusable and may slow down
+        parsing.
+
+    Returns
+    -------
+    datetime
+        If parsing succeeded.
+        Return type depends on input (types in parenthesis correspond to
+        fallback in case of unsuccessful timezone or out-of-range timestamp
+        parsing):
+
+        - scalar: :class:`Timestamp` (or :class:`datetime.datetime`)
+        - array-like: :class:`DatetimeIndex` (or :class:`Series` with
+          :class:`object` dtype containing :class:`datetime.datetime`)
+        - Series: :class:`Series` of :class:`datetime64` dtype (or
+          :class:`Series` of :class:`object` dtype containing
+          :class:`datetime.datetime`)
+        - DataFrame: :class:`Series` of :class:`datetime64` dtype (or
+          :class:`Series` of :class:`object` dtype containing
+          :class:`datetime.datetime`)
+
+    Raises
+    ------
+    ParserError
+        When parsing a date from string fails.
+    ValueError
+        When another datetime conversion error happens. For example when one
+        of 'year', 'month', day' columns is missing in a :class:`DataFrame`, or
+        when a Timezone-aware :class:`datetime.datetime` is found in an array-like
+        of mixed time offsets, and ``utc=False``, or when parsing datetimes
+        with mixed time zones unless ``utc=True``. If parsing datetimes with mixed
+        time zones, please specify ``utc=True``.
+
+    See Also
+    --------
+    DataFrame.astype : Cast argument to a specified dtype.
+    to_timedelta : Convert argument to timedelta.
+    convert_dtypes : Convert dtypes.
+
+    Notes
+    -----
+
+    Many input types are supported, and lead to different output types:
+
+    - **scalars** can be int, float, str, datetime object (from stdlib :mod:`datetime`
+      module or :mod:`numpy`). They are converted to :class:`Timestamp` when
+      possible, otherwise they are converted to :class:`datetime.datetime`.
+      None/NaN/null scalars are converted to :const:`NaT`.
+
+    - **array-like** can contain int, float, str, datetime objects. They are
+      converted to :class:`DatetimeIndex` when possible, otherwise they are
+      converted to :class:`Index` with :class:`object` dtype, containing
+      :class:`datetime.datetime`. None/NaN/null entries are converted to
+      :const:`NaT` in both cases.
+
+    - **Series** are converted to :class:`Series` with :class:`datetime64`
+      dtype when possible, otherwise they are converted to :class:`Series` with
+      :class:`object` dtype, containing :class:`datetime.datetime`. None/NaN/null
+      entries are converted to :const:`NaT` in both cases.
+
+    - **DataFrame/dict-like** are converted to :class:`Series` with
+      :class:`datetime64` dtype. For each row a datetime is created from assembling
+      the various dataframe columns. Column keys can be common abbreviations
+      like ['year', 'month', 'day', 'minute', 'second', 'ms', 'us', 'ns']) or
+      plurals of the same.
+
+    The following causes are responsible for :class:`datetime.datetime` objects
+    being returned (possibly inside an :class:`Index` or a :class:`Series` with
+    :class:`object` dtype) instead of a proper pandas designated type
+    (:class:`Timestamp`, :class:`DatetimeIndex` or :class:`Series`
+    with :class:`datetime64` dtype):
+
+    - when any input element is before :const:`Timestamp.min` or after
+      :const:`Timestamp.max`, see `timestamp limitations
+      <https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html
+      #timeseries-timestamp-limits>`_.
+
+    - when ``utc=False`` (default) and the input is an array-like or
+      :class:`Series` containing mixed naive/aware datetime, or aware with mixed
+      time offsets. Note that this happens in the (quite frequent) situation when
+      the timezone has a daylight savings policy. In that case you may wish to
+      use ``utc=True``.
+
+    Examples
+    --------
+
+    **Handling various input formats**
+
+    Assembling a datetime from multiple columns of a :class:`DataFrame`. The keys
+    can be common abbreviations like ['year', 'month', 'day', 'minute', 'second',
+    'ms', 'us', 'ns']) or plurals of the same
+
+    >>> df = pd.DataFrame({"year": [2015, 2016], "month": [2, 3], "day": [4, 5]})
+    >>> pd.to_datetime(df)
+    0   2015-02-04
+    1   2016-03-05
+    dtype: datetime64[s]
+
+    Using a unix epoch time
+
+    >>> pd.to_datetime(1490195805, unit="s")
+    Timestamp('2017-03-22 15:16:45')
+    >>> pd.to_datetime(1490195805433502912, unit="ns")
+    Timestamp('2017-03-22 15:16:45.433502912')
+
+    .. warning:: For float arg, precision rounding might happen. To prevent
+        unexpected behavior use a fixed-width exact type.
+
+    Using a non-unix epoch origin
+
+    >>> pd.to_datetime([1, 2, 3], unit="D", origin=pd.Timestamp("1960-01-01"))
+    DatetimeIndex(['1960-01-02', '1960-01-03', '1960-01-04'],
+                  dtype='datetime64[ns]', freq=None)
+
+    **Differences with strptime behavior**
+
+    :const:`"%f"` will parse all the way up to nanoseconds.
+
+    >>> pd.to_datetime("2018-10-26 12:00:00.0000000011", format="%Y-%m-%d %H:%M:%S.%f")
+    Timestamp('2018-10-26 12:00:00.000000001')
+
+    **Non-convertible date/times**
+
+    Passing ``errors='coerce'`` will force an out-of-bounds date to :const:`NaT`,
+    in addition to forcing non-dates (or non-parseable dates) to :const:`NaT`.
+
+    >>> pd.to_datetime("invalid for Ymd", format="%Y%m%d", errors="coerce")
+    NaT
+
+    .. _to_datetime_tz_examples:
+
+    **Timezones and time offsets**
+
+    The default behaviour (``utc=False``) is as follows:
+
+    - Timezone-naive inputs are converted to timezone-naive :class:`DatetimeIndex`:
+
+    >>> pd.to_datetime(["2018-10-26 12:00:00", "2018-10-26 13:00:15"])
+    DatetimeIndex(['2018-10-26 12:00:00', '2018-10-26 13:00:15'],
+                  dtype='datetime64[s]', freq=None)
+
+    - Timezone-aware inputs *with constant time offset* are converted to
+      timezone-aware :class:`DatetimeIndex`:
+
+    >>> pd.to_datetime(["2018-10-26 12:00 -0500", "2018-10-26 13:00 -0500"])
+    DatetimeIndex(['2018-10-26 12:00:00-05:00', '2018-10-26 13:00:00-05:00'],
+                  dtype='datetime64[s, UTC-05:00]', freq=None)
+
+    - However, timezone-aware inputs *with mixed time offsets* (for example
+      issued from a timezone with daylight savings, such as Europe/Paris)
+      are **not successfully converted** to a :class:`DatetimeIndex`.
+      Parsing datetimes with mixed time zones will raise a ValueError unless
+      ``utc=True``:
+
+    >>> pd.to_datetime(
+    ...     ["2020-10-25 02:00 +0200", "2020-10-25 04:00 +0100"]
+    ... )  # doctest: +SKIP
+    ValueError: Mixed timezones detected. Pass utc=True in to_datetime
+    or tz='UTC' in DatetimeIndex to convert to a common timezone.
+
+    - To create a :class:`Series` with mixed offsets and ``object`` dtype, please use
+      :meth:`Series.apply` and :func:`datetime.datetime.strptime`:
+
+    >>> import datetime as dt
+    >>> ser = pd.Series(["2020-10-25 02:00 +0200", "2020-10-25 04:00 +0100"])
+    >>> ser.apply(lambda x: dt.datetime.strptime(x, "%Y-%m-%d %H:%M %z"))
+    0    2020-10-25 02:00:00+02:00
+    1    2020-10-25 04:00:00+01:00
+    dtype: object
+
+    - A mix of timezone-aware and timezone-naive inputs will also raise a ValueError
+      unless ``utc=True``:
+
+    >>> from datetime import datetime
+    >>> pd.to_datetime(
+    ...     ["2020-01-01 01:00:00-01:00", datetime(2020, 1, 1, 3, 0)]
+    ... )  # doctest: +SKIP
+    ValueError: Mixed timezones detected. Pass utc=True in to_datetime
+    or tz='UTC' in DatetimeIndex to convert to a common timezone.
+
+    |
+
+    Setting ``utc=True`` solves most of the above issues:
+
+    - Timezone-naive inputs are *localized* as UTC
+
+    >>> pd.to_datetime(["2018-10-26 12:00", "2018-10-26 13:00"], utc=True)
+    DatetimeIndex(['2018-10-26 12:00:00+00:00', '2018-10-26 13:00:00+00:00'],
+                  dtype='datetime64[s, UTC]', freq=None)
+
+    - Timezone-aware inputs are *converted* to UTC (the output represents the
+      exact same datetime, but viewed from the UTC time offset `+00:00`).
+
+    >>> pd.to_datetime(["2018-10-26 12:00 -0530", "2018-10-26 12:00 -0500"], utc=True)
+    DatetimeIndex(['2018-10-26 17:30:00+00:00', '2018-10-26 17:00:00+00:00'],
+                  dtype='datetime64[s, UTC]', freq=None)
+
+    - Inputs can contain both string or datetime, the above
+      rules still apply
+
+    >>> pd.to_datetime(["2018-10-26 12:00", datetime(2020, 1, 1, 18)], utc=True)
+    DatetimeIndex(['2018-10-26 12:00:00+00:00', '2020-01-01 18:00:00+00:00'],
+                  dtype='datetime64[us, UTC]', freq=None)
+    """
+    if exact is not lib.no_default and format in {"mixed", "ISO8601"}:
+        raise ValueError("Cannot use 'exact' when 'format' is 'mixed' or 'ISO8601'")
+    if arg is None:
+        return None
+
+    if origin != "unix":
+        arg = _adjust_to_origin(arg, origin, unit)
+
+    convert_listlike = partial(
+        _convert_listlike_datetimes,
+        utc=utc,
+        unit=unit,
+        dayfirst=dayfirst,
+        yearfirst=yearfirst,
+        errors=errors,
+        exact=exact,  # type: ignore[arg-type]
+    )
+    result: Timestamp | NaTType | Series | Index
+
+    if isinstance(arg, Timestamp):
+        result = arg
+        if utc:
+            if arg.tz is not None:
+                result = arg.tz_convert("utc")
+            else:
+                result = arg.tz_localize("utc")
+    elif isinstance(arg, ABCSeries):
+        cache_array = _maybe_cache(arg, format, cache, convert_listlike)
+        if not cache_array.empty:
+            result = arg.map(cache_array)
+        else:
+            values = convert_listlike(arg._values, format)
+            result = arg._constructor(values, index=arg.index, name=arg.name)
+    elif isinstance(arg, (ABCDataFrame, abc.MutableMapping)):
+        result = _assemble_from_unit_mappings(arg, errors, utc)
+    elif isinstance(arg, Index):
+        cache_array = _maybe_cache(arg, format, cache, convert_listlike)
+        if not cache_array.empty:
+            result = _convert_and_box_cache(arg, cache_array, name=arg.name)
+        else:
+            result = convert_listlike(arg, format, name=arg.name)
+    elif is_list_like(arg):
+        try:
+            # error: Argument 1 to "_maybe_cache" has incompatible type
+            # "Union[float, str, datetime, List[Any], Tuple[Any, ...], ExtensionArray,
+            # ndarray[Any, Any], Series]"; expected "Union[List[Any], Tuple[Any, ...],
+            # Union[Union[ExtensionArray, ndarray[Any, Any]], Index, Series], Series]"
+            argc = cast(
+                Union[list, tuple, ExtensionArray, np.ndarray, "Series", Index], arg
+            )
+            cache_array = _maybe_cache(argc, format, cache, convert_listlike)
+        except OutOfBoundsDatetime:
+            # caching attempts to create a DatetimeIndex, which may raise
+            # an OOB. If that's the desired behavior, then just reraise...
+            if errors == "raise":
+                raise
+            # ... otherwise, continue without the cache.
+            from pandas import Series
+
+            cache_array = Series([], dtype=object)  # just an empty array
+        if not cache_array.empty:
+            result = _convert_and_box_cache(argc, cache_array)
+        else:
+            result = convert_listlike(argc, format)
+    else:
+        result = convert_listlike(np.array([arg]), format)[0]
+        if isinstance(arg, bool) and isinstance(result, np.bool_):
+            result = bool(result)  # TODO: avoid this kludge.
+
+    #  error: Incompatible return value type (got "Union[Timestamp, NaTType,
+    # Series, Index]", expected "Union[DatetimeIndex, Series, float, str,
+    # NaTType, None]")
+    return result  # type: ignore[return-value]
+
+
+# mappings for assembling units
+_unit_map = {
+    "year": "year",
+    "years": "year",
+    "month": "month",
+    "months": "month",
+    "day": "day",
+    "days": "day",
+    "hour": "h",
+    "hours": "h",
+    "minute": "m",
+    "minutes": "m",
+    "second": "s",
+    "seconds": "s",
+    "ms": "ms",
+    "millisecond": "ms",
+    "milliseconds": "ms",
+    "us": "us",
+    "microsecond": "us",
+    "microseconds": "us",
+    "ns": "ns",
+    "nanosecond": "ns",
+    "nanoseconds": "ns",
+}
+
+
+def _assemble_from_unit_mappings(
+    arg, errors: DateTimeErrorChoices, utc: bool
+) -> Series:
+    """
+    assemble the unit specified fields from the arg (DataFrame)
+    Return a Series for actual parsing
+
+    Parameters
+    ----------
+    arg : DataFrame
+    errors : {'raise', 'coerce'}, default 'raise'
+
+        - If :const:`'raise'`, then invalid parsing will raise an exception
+        - If :const:`'coerce'`, then invalid parsing will be set as :const:`NaT`
+    utc : bool
+        Whether to convert/localize timestamps to UTC.
+
+    Returns
+    -------
+    Series
+    """
+    from pandas import (
+        DataFrame,
+        to_numeric,
+        to_timedelta,
+    )
+
+    arg = DataFrame(arg)
+    if not arg.columns.is_unique:
+        raise ValueError("cannot assemble with duplicate keys")
+
+    # replace passed unit with _unit_map
+    def f(value):
+        if value in _unit_map:
+            return _unit_map[value]
+
+        # m is case significant
+        if value.lower() in _unit_map:
+            return _unit_map[value.lower()]
+
+        return value
+
+    unit = {k: f(k) for k in arg.keys()}
+    unit_rev = {v: k for k, v in unit.items()}
+
+    # we require at least Ymd
+    required = ["year", "month", "day"]
+    req = set(required) - set(unit_rev.keys())
+    if len(req):
+        _required = ",".join(sorted(req))
+        raise ValueError(
+            "to assemble mappings requires at least that "
+            f"[year, month, day] be specified: [{_required}] is missing"
+        )
+
+    # keys we don't recognize
+    excess = set(unit_rev.keys()) - set(_unit_map.values())
+    if len(excess):
+        _excess = ",".join(sorted(excess))
+        raise ValueError(
+            f"extra keys have been passed to the datetime assemblage: [{_excess}]"
+        )
+
+    def coerce(values):
+        # we allow coercion to if errors allows
+        values = to_numeric(values, errors=errors)
+
+        # prevent prevision issues in case of float32 # GH#60506
+        if is_float_dtype(values.dtype):
+            values = values.astype("float64")
+
+        # prevent overflow in case of int8 or int16
+        if is_integer_dtype(values.dtype):
+            values = values.astype("int64")
+        return values
+
+    values = (
+        coerce(arg[unit_rev["year"]]) * 10000
+        + coerce(arg[unit_rev["month"]]) * 100
+        + coerce(arg[unit_rev["day"]])
+    )
+    try:
+        values = to_datetime(values, format="%Y%m%d", errors=errors, utc=utc)
+    except (TypeError, ValueError) as err:
+        raise ValueError(f"cannot assemble the datetimes: {err}") from err
+
+    units: list[UnitChoices] = ["h", "m", "s", "ms", "us", "ns"]
+    for u in units:
+        value = unit_rev.get(u)
+        if value is not None and value in arg:
+            try:
+                values += to_timedelta(coerce(arg[value]), unit=u, errors=errors)
+            except (TypeError, ValueError) as err:
+                raise ValueError(
+                    f"cannot assemble the datetimes [{value}]: {err}"
+                ) from err
+    return values
+
+
+__all__ = [
+    "DateParseError",
+    "should_cache",
+    "to_datetime",
+]
diff --git a/pandas/pandas/core/tools/numeric.py b/pandas/pandas/core/tools/numeric.py
new file mode 100644
index 0000000000000..bc45343d6e2d3
--- /dev/null
+++ b/pandas/pandas/core/tools/numeric.py
@@ -0,0 +1,316 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
+from pandas._libs.tslibs import (
+    Timedelta,
+    Timestamp,
+)
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.cast import maybe_downcast_numeric
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_bool_dtype,
+    is_decimal,
+    is_integer_dtype,
+    is_number,
+    is_numeric_dtype,
+    is_scalar,
+    is_string_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import ArrowDtype
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+
+from pandas.core.arrays import BaseMaskedArray
+from pandas.core.arrays.string_ import StringDtype
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        DateTimeErrorChoices,
+        DtypeBackend,
+        npt,
+    )
+
+
+def to_numeric(
+    arg,
+    errors: DateTimeErrorChoices = "raise",
+    downcast: Literal["integer", "signed", "unsigned", "float"] | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+):
+    """
+    Convert argument to a numeric type.
+
+    The default return dtype is `float64` or `int64`
+    depending on the data supplied. Use the `downcast` parameter
+    to obtain other dtypes.
+
+    Please note that precision loss may occur if really large numbers
+    are passed in. Due to the internal limitations of `ndarray`, if
+    numbers smaller than `-9223372036854775808` (np.iinfo(np.int64).min)
+    or larger than `18446744073709551615` (np.iinfo(np.uint64).max) are
+    passed in, it is very likely they will be converted to float so that
+    they can be stored in an `ndarray`. These warnings apply similarly to
+    `Series` since it internally leverages `ndarray`.
+
+    Parameters
+    ----------
+    arg : scalar, list, tuple, 1-d array, or Series
+        Argument to be converted.
+
+    errors : {'raise', 'coerce'}, default 'raise'
+        - If 'raise', then invalid parsing will raise an exception.
+        - If 'coerce', then invalid parsing will be set as NaN.
+
+    downcast : str, default None
+        Can be 'integer', 'signed', 'unsigned', or 'float'.
+        If not None, and if the data has been successfully cast to a
+        numerical dtype (or if the data was numeric to begin with),
+        downcast that resulting data to the smallest numerical dtype
+        possible according to the following rules:
+
+        - 'integer' or 'signed': smallest signed int dtype (min.: np.int8)
+        - 'unsigned': smallest unsigned int dtype (min.: np.uint8)
+        - 'float': smallest float dtype (min.: np.float32)
+
+        As this behaviour is separate from the core conversion to
+        numeric values, any errors raised during the downcasting
+        will be surfaced regardless of the value of the 'errors' input.
+
+        In addition, downcasting will only occur if the size
+        of the resulting data's dtype is strictly larger than
+        the dtype it is to be cast to, so if none of the dtypes
+        checked satisfy that specification, no downcasting will be
+        performed on the data.
+
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed object
+        * ``"pyarrow"``: returns with pyarrow-backed nullable object
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    ret
+        Numeric if parsing succeeded.
+        Return type depends on input.  Series if Series, otherwise ndarray.
+
+    See Also
+    --------
+    DataFrame.astype : Cast argument to a specified dtype.
+    to_datetime : Convert argument to datetime.
+    to_timedelta : Convert argument to timedelta.
+    numpy.ndarray.astype : Cast a numpy array to a specified type.
+    DataFrame.convert_dtypes : Convert dtypes.
+
+    Examples
+    --------
+    Take separate series and convert to numeric, coercing when told to
+
+    >>> s = pd.Series(["1.0", "2", -3])
+    >>> pd.to_numeric(s)
+    0    1.0
+    1    2.0
+    2   -3.0
+    dtype: float64
+    >>> pd.to_numeric(s, downcast="float")
+    0    1.0
+    1    2.0
+    2   -3.0
+    dtype: float32
+    >>> pd.to_numeric(s, downcast="signed")
+    0    1
+    1    2
+    2   -3
+    dtype: int8
+    >>> s = pd.Series(["apple", "1.0", "2", -3])
+    >>> pd.to_numeric(s, errors="coerce")
+    0    NaN
+    1    1.0
+    2    2.0
+    3   -3.0
+    dtype: float64
+
+    Downcasting of nullable integer and floating dtypes is supported:
+
+    >>> s = pd.Series([1, 2, 3], dtype="Int64")
+    >>> pd.to_numeric(s, downcast="integer")
+    0    1
+    1    2
+    2    3
+    dtype: Int8
+    >>> s = pd.Series([1.0, 2.1, 3.0], dtype="Float64")
+    >>> pd.to_numeric(s, downcast="float")
+    0    1.0
+    1    2.1
+    2    3.0
+    dtype: Float32
+    """
+    if downcast not in (None, "integer", "signed", "unsigned", "float"):
+        raise ValueError("invalid downcasting method provided")
+
+    if errors not in ("raise", "coerce"):
+        raise ValueError("invalid error value specified")
+
+    check_dtype_backend(dtype_backend)
+
+    is_series = False
+    is_index = False
+    is_scalars = False
+
+    if isinstance(arg, ABCSeries):
+        is_series = True
+        values = arg.values
+    elif isinstance(arg, ABCIndex):
+        is_index = True
+        if needs_i8_conversion(arg.dtype):
+            values = arg.view("i8")
+        else:
+            values = arg.values
+    elif isinstance(arg, (list, tuple)):
+        values = np.array(arg, dtype="O")
+    elif is_scalar(arg):
+        if is_decimal(arg):
+            return float(arg)
+        if is_number(arg):
+            return arg
+        if isinstance(arg, (Timedelta, Timestamp)):
+            return arg._value
+        is_scalars = True
+        values = np.array([arg], dtype="O")
+    elif getattr(arg, "ndim", 1) > 1:
+        raise TypeError("arg must be a list, tuple, 1-d array, or Series")
+    else:
+        values = arg
+
+    # GH33013: for IntegerArray & FloatingArray extract non-null values for casting
+    # save mask to reconstruct the full array after casting
+    mask: npt.NDArray[np.bool_] | None = None
+    if isinstance(values, BaseMaskedArray):
+        mask = values._mask
+        values = values._data[~mask]
+
+    values_dtype = getattr(values, "dtype", None)
+    if isinstance(values_dtype, ArrowDtype):
+        mask = values.isna()
+        values = values.dropna().to_numpy()
+    new_mask: np.ndarray | None = None
+    if is_numeric_dtype(values_dtype):
+        pass
+    elif lib.is_np_dtype(values_dtype, "mM"):
+        values = values.view(np.int64)
+    else:
+        values = ensure_object(values)
+        coerce_numeric = errors != "raise"
+        values, new_mask = lib.maybe_convert_numeric(  # type: ignore[call-overload]
+            values,
+            set(),
+            coerce_numeric=coerce_numeric,
+            convert_to_masked_nullable=dtype_backend is not lib.no_default
+            or (
+                isinstance(values_dtype, StringDtype)
+                and values_dtype.na_value is libmissing.NA
+            ),
+        )
+
+    if new_mask is not None:
+        # Remove unnecessary values, is expected later anyway and enables
+        # downcasting
+        values = values[~new_mask]
+    elif (dtype_backend is not lib.no_default and new_mask is None) or (
+        isinstance(values_dtype, StringDtype) and values_dtype.na_value is libmissing.NA
+    ):
+        new_mask = np.zeros(values.shape, dtype=np.bool_)
+
+    # attempt downcast only if the data has been successfully converted
+    # to a numerical dtype and if a downcast method has been specified
+    if downcast is not None and is_numeric_dtype(values.dtype):
+        typecodes: str | None = None
+
+        if downcast in ("integer", "signed"):
+            typecodes = np.typecodes["Integer"]
+        elif downcast == "unsigned" and (not len(values) or np.min(values) >= 0):
+            typecodes = np.typecodes["UnsignedInteger"]
+        elif downcast == "float":
+            typecodes = np.typecodes["Float"]
+
+            # pandas support goes only to np.float32,
+            # as float dtypes smaller than that are
+            # extremely rare and not well supported
+            float_32_char = np.dtype(np.float32).char
+            float_32_ind = typecodes.index(float_32_char)
+            typecodes = typecodes[float_32_ind:]
+
+        if typecodes is not None:
+            # from smallest to largest
+            for typecode in typecodes:
+                dtype = np.dtype(typecode)
+                if dtype.itemsize <= values.dtype.itemsize:
+                    values = maybe_downcast_numeric(values, dtype)
+
+                    # successful conversion
+                    if values.dtype == dtype:
+                        break
+
+    # GH33013: for IntegerArray, BooleanArray & FloatingArray need to reconstruct
+    # masked array
+    if (mask is not None or new_mask is not None) and not is_string_dtype(values.dtype):
+        if mask is None or (new_mask is not None and new_mask.shape == mask.shape):
+            # GH 52588
+            mask = new_mask
+        else:
+            mask = mask.copy()
+        assert isinstance(mask, np.ndarray)
+        data = np.zeros(mask.shape, dtype=values.dtype)
+        data[~mask] = values
+
+        from pandas.core.arrays import (
+            ArrowExtensionArray,
+            BooleanArray,
+            FloatingArray,
+            IntegerArray,
+        )
+
+        klass: type[IntegerArray | BooleanArray | FloatingArray]
+        if is_integer_dtype(data.dtype):
+            klass = IntegerArray
+        elif is_bool_dtype(data.dtype):
+            klass = BooleanArray
+        else:
+            klass = FloatingArray
+        values = klass(data, mask)
+
+        if dtype_backend == "pyarrow" or isinstance(values_dtype, ArrowDtype):
+            values = ArrowExtensionArray(values.__arrow_array__())
+
+    if is_series:
+        return arg._constructor(values, index=arg.index, name=arg.name)
+    elif is_index:
+        # because we want to coerce to numeric if possible,
+        # do not use _shallow_copy
+        from pandas import Index
+
+        return Index(values, name=arg.name)
+    elif is_scalars:
+        return values[0]
+    else:
+        return values
diff --git a/pandas/pandas/core/tools/timedeltas.py b/pandas/pandas/core/tools/timedeltas.py
new file mode 100644
index 0000000000000..8d82a5c213910
--- /dev/null
+++ b/pandas/pandas/core/tools/timedeltas.py
@@ -0,0 +1,247 @@
+"""
+timedelta support tools
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    NaT,
+    NaTType,
+)
+from pandas._libs.tslibs.timedeltas import (
+    Timedelta,
+    disallow_ambiguous_unit,
+    parse_timedelta_unit,
+)
+
+from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.dtypes import ArrowDtype
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+
+from pandas.core.arrays.timedeltas import sequence_to_td64ns
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+    from datetime import timedelta
+
+    from pandas._libs.tslibs.timedeltas import UnitChoices
+    from pandas._typing import (
+        ArrayLike,
+        DateTimeErrorChoices,
+    )
+
+    from pandas import (
+        Index,
+        Series,
+        TimedeltaIndex,
+    )
+
+
+@overload
+def to_timedelta(
+    arg: str | float | timedelta,
+    unit: UnitChoices | None = ...,
+    errors: DateTimeErrorChoices = ...,
+) -> Timedelta: ...
+
+
+@overload
+def to_timedelta(
+    arg: Series,
+    unit: UnitChoices | None = ...,
+    errors: DateTimeErrorChoices = ...,
+) -> Series: ...
+
+
+@overload
+def to_timedelta(
+    arg: list | tuple | range | ArrayLike | Index,
+    unit: UnitChoices | None = ...,
+    errors: DateTimeErrorChoices = ...,
+) -> TimedeltaIndex: ...
+
+
+def to_timedelta(
+    arg: str
+    | int
+    | float
+    | timedelta
+    | list
+    | tuple
+    | range
+    | ArrayLike
+    | Index
+    | Series,
+    unit: UnitChoices | None = None,
+    errors: DateTimeErrorChoices = "raise",
+) -> Timedelta | TimedeltaIndex | Series | NaTType | Any:
+    """
+    Convert argument to timedelta.
+
+    Timedeltas are absolute differences in times, expressed in difference
+    units (e.g. days, hours, minutes, seconds). This method converts
+    an argument from a recognized timedelta format / value into
+    a Timedelta type.
+
+    Parameters
+    ----------
+    arg : str, timedelta, list-like or Series
+        The data to be converted to timedelta.
+
+        .. versionchanged:: 2.0
+            Strings with units 'M', 'Y' and 'y' do not represent
+            unambiguous timedelta values and will raise an exception.
+
+    unit : str, optional
+        Denotes the unit of the arg for numeric `arg`. Defaults to ``"ns"``.
+
+        Possible values:
+
+        * 'W'
+        * 'D' / 'days' / 'day'
+        * 'hours' / 'hour' / 'hr' / 'h' / 'H'
+        * 'm' / 'minute' / 'min' / 'minutes'
+        * 's' / 'seconds' / 'sec' / 'second' / 'S'
+        * 'ms' / 'milliseconds' / 'millisecond' / 'milli' / 'millis'
+        * 'us' / 'microseconds' / 'microsecond' / 'micro' / 'micros'
+        * 'ns' / 'nanoseconds' / 'nano' / 'nanos' / 'nanosecond'
+
+        Must not be specified when `arg` contains strings and ``errors="raise"``.
+
+        .. deprecated:: 2.2.0
+            Units 'H'and 'S' are deprecated and will be removed
+            in a future version. Please use 'h' and 's'.
+
+    errors : {'raise', 'coerce'}, default 'raise'
+        - If 'raise', then invalid parsing will raise an exception.
+        - If 'coerce', then invalid parsing will be set as NaT.
+
+    Returns
+    -------
+    timedelta
+        If parsing succeeded.
+        Return type depends on input:
+
+        - list-like: TimedeltaIndex of timedelta64 dtype
+        - Series: Series of timedelta64 dtype
+        - scalar: Timedelta
+
+    See Also
+    --------
+    DataFrame.astype : Cast argument to a specified dtype.
+    to_datetime : Convert argument to datetime.
+    convert_dtypes : Convert dtypes.
+
+    Notes
+    -----
+    If the precision is higher than nanoseconds, the precision of the duration is
+    truncated to nanoseconds for string inputs.
+
+    Examples
+    --------
+    Parsing a single string to a Timedelta:
+
+    >>> pd.to_timedelta("1 days 06:05:01.00003")
+    Timedelta('1 days 06:05:01.000030')
+    >>> pd.to_timedelta("15.5us")
+    Timedelta('0 days 00:00:00.000015500')
+
+    Parsing a list or array of strings:
+
+    >>> pd.to_timedelta(["1 days 06:05:01.00003", "15.5us", "nan"])
+    TimedeltaIndex(['1 days 06:05:01.000030', '0 days 00:00:00.000015500', NaT],
+                   dtype='timedelta64[ns]', freq=None)
+
+    Converting numbers by specifying the `unit` keyword argument:
+
+    >>> pd.to_timedelta(np.arange(5), unit="s")
+    TimedeltaIndex(['0 days 00:00:00', '0 days 00:00:01', '0 days 00:00:02',
+                    '0 days 00:00:03', '0 days 00:00:04'],
+                   dtype='timedelta64[ns]', freq=None)
+    >>> pd.to_timedelta(np.arange(5), unit="D")
+    TimedeltaIndex(['0 days', '1 days', '2 days', '3 days', '4 days'],
+                   dtype='timedelta64[ns]', freq=None)
+    """
+    if unit is not None:
+        unit = parse_timedelta_unit(unit)
+        disallow_ambiguous_unit(unit)
+
+    if errors not in ("raise", "coerce"):
+        raise ValueError("errors must be one of 'raise', or 'coerce'.")
+
+    if arg is None:
+        return arg
+    elif isinstance(arg, ABCSeries):
+        values = _convert_listlike(arg._values, unit=unit, errors=errors)
+        return arg._constructor(values, index=arg.index, name=arg.name)
+    elif isinstance(arg, ABCIndex):
+        return _convert_listlike(arg, unit=unit, errors=errors, name=arg.name)
+    elif isinstance(arg, np.ndarray) and arg.ndim == 0:
+        # extract array scalar and process below
+        # error: Incompatible types in assignment (expression has type "object",
+        # variable has type "Union[str, int, float, timedelta, List[Any],
+        # Tuple[Any, ...], Union[Union[ExtensionArray, ndarray[Any, Any]], Index,
+        # Series]]")  [assignment]
+        arg = lib.item_from_zerodim(arg)  # type: ignore[assignment]
+    elif is_list_like(arg) and getattr(arg, "ndim", 1) == 1:
+        return _convert_listlike(arg, unit=unit, errors=errors)
+    elif getattr(arg, "ndim", 1) > 1:
+        raise TypeError(
+            "arg must be a string, timedelta, list, tuple, 1-d array, or Series"
+        )
+
+    if isinstance(arg, str) and unit is not None:
+        raise ValueError("unit must not be specified if the input is/contains a str")
+
+    # ...so it must be a scalar value. Return scalar.
+    return _coerce_scalar_to_timedelta_type(arg, unit=unit, errors=errors)
+
+
+def _coerce_scalar_to_timedelta_type(
+    r, unit: UnitChoices | None = "ns", errors: DateTimeErrorChoices = "raise"
+) -> Timedelta | NaTType:
+    """Convert string 'r' to a timedelta object."""
+    result: Timedelta | NaTType
+
+    try:
+        result = Timedelta(r, unit)
+    except ValueError:
+        if errors == "raise":
+            raise
+        # coerce
+        result = NaT
+
+    return result
+
+
+def _convert_listlike(
+    arg,
+    unit: UnitChoices | None = None,
+    errors: DateTimeErrorChoices = "raise",
+    name: Hashable | None = None,
+):
+    """Convert a list of objects to a timedelta index object."""
+    arg_dtype = getattr(arg, "dtype", None)
+    if isinstance(arg, (list, tuple)) or arg_dtype is None:
+        arg = np.array(arg, dtype=object)
+    elif isinstance(arg_dtype, ArrowDtype) and arg_dtype.kind == "m":
+        return arg
+
+    td64arr = sequence_to_td64ns(arg, unit=unit, errors=errors, copy=False)[0]
+
+    from pandas import TimedeltaIndex
+
+    value = TimedeltaIndex(td64arr, name=name)
+    return value
diff --git a/pandas/pandas/core/tools/times.py b/pandas/pandas/core/tools/times.py
new file mode 100644
index 0000000000000..f6a610ebc36ab
--- /dev/null
+++ b/pandas/pandas/core/tools/times.py
@@ -0,0 +1,153 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    time,
+)
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs.lib import is_list_like
+
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import notna
+
+if TYPE_CHECKING:
+    from pandas._typing import DateTimeErrorChoices
+
+
+def to_time(
+    arg,
+    format: str | None = None,
+    infer_time_format: bool = False,
+    errors: DateTimeErrorChoices = "raise",
+):
+    """
+    Parse time strings to time objects using fixed strptime formats ("%H:%M",
+    "%H%M", "%I:%M%p", "%I%M%p", "%H:%M:%S", "%H%M%S", "%I:%M:%S%p",
+    "%I%M%S%p")
+
+    Use infer_time_format if all the strings are in the same format to speed
+    up conversion.
+
+    Parameters
+    ----------
+    arg : string in time format, datetime.time, list, tuple, 1-d array,  Series
+    format : str, default None
+        Format used to convert arg into a time object.  If None, fixed formats
+        are used.
+    infer_time_format: bool, default False
+        Infer the time format based on the first non-NaN element.  If all
+        strings are in the same format, this will speed up conversion.
+    errors : {'raise', 'coerce'}, default 'raise'
+        - If 'raise', then invalid parsing will raise an exception
+        - If 'coerce', then invalid parsing will be set as None
+
+    Returns
+    -------
+    datetime.time
+    """
+    if errors not in ("raise", "coerce"):
+        raise ValueError("errors must be one of 'raise', or 'coerce'.")
+
+    def _convert_listlike(arg, format):
+        if isinstance(arg, (list, tuple)):
+            arg = np.array(arg, dtype="O")
+
+        elif getattr(arg, "ndim", 1) > 1:
+            raise TypeError(
+                "arg must be a string, datetime, list, tuple, 1-d array, or Series"
+            )
+
+        arg = np.asarray(arg, dtype="O")
+
+        if infer_time_format and format is None:
+            format = _guess_time_format_for_array(arg)
+
+        times: list[time | None] = []
+        if format is not None:
+            for element in arg:
+                try:
+                    times.append(datetime.strptime(element, format).time())
+                except (ValueError, TypeError) as err:
+                    if errors == "raise":
+                        msg = (
+                            f"Cannot convert {element} to a time with given "
+                            f"format {format}"
+                        )
+                        raise ValueError(msg) from err
+                    times.append(None)
+        else:
+            formats = _time_formats[:]
+            format_found = False
+            for element in arg:
+                time_object = None
+                try:
+                    time_object = time.fromisoformat(element)
+                except (ValueError, TypeError):
+                    for time_format in formats:
+                        try:
+                            time_object = datetime.strptime(element, time_format).time()
+                            if not format_found:
+                                # Put the found format in front
+                                fmt = formats.pop(formats.index(time_format))
+                                formats.insert(0, fmt)
+                                format_found = True
+                            break
+                        except (ValueError, TypeError):
+                            continue
+
+                if time_object is not None:
+                    times.append(time_object)
+                elif errors == "raise":
+                    raise ValueError(f"Cannot convert arg {arg} to a time")
+                else:
+                    times.append(None)
+
+        return times
+
+    if arg is None:
+        return arg
+    elif isinstance(arg, time):
+        return arg
+    elif isinstance(arg, ABCSeries):
+        values = _convert_listlike(arg._values, format)
+        return arg._constructor(values, index=arg.index, name=arg.name)
+    elif isinstance(arg, ABCIndex):
+        return _convert_listlike(arg, format)
+    elif is_list_like(arg):
+        return _convert_listlike(arg, format)
+
+    return _convert_listlike(np.array([arg]), format)[0]
+
+
+# Fixed time formats for time parsing
+_time_formats = [
+    "%H:%M",
+    "%H%M",
+    "%I:%M%p",
+    "%I%M%p",
+    "%H:%M:%S",
+    "%H%M%S",
+    "%I:%M:%S%p",
+    "%I%M%S%p",
+]
+
+
+def _guess_time_format_for_array(arr):
+    # Try to guess the format based on the first non-NaN element
+    non_nan_elements = notna(arr).nonzero()[0]
+    if len(non_nan_elements):
+        element = arr[non_nan_elements[0]]
+        for time_format in _time_formats:
+            try:
+                datetime.strptime(element, time_format)
+                return time_format
+            except ValueError:
+                pass
+
+    return None
diff --git a/pandas/pandas/core/util/__init__.py b/pandas/pandas/core/util/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/core/util/hashing.py b/pandas/pandas/core/util/hashing.py
new file mode 100644
index 0000000000000..e120e69dc27cf
--- /dev/null
+++ b/pandas/pandas/core/util/hashing.py
@@ -0,0 +1,347 @@
+"""
+data hash pandas / numpy objects
+"""
+
+from __future__ import annotations
+
+import itertools
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs.hashing import hash_object_array
+
+from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.dtypes import CategoricalDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCExtensionArray,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterable,
+        Iterator,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        MultiIndex,
+        Series,
+    )
+
+
+# 16 byte long hashing key
+_default_hash_key = "0123456789123456"
+
+
+def combine_hash_arrays(
+    arrays: Iterator[np.ndarray], num_items: int
+) -> npt.NDArray[np.uint64]:
+    """
+    Parameters
+    ----------
+    arrays : Iterator[np.ndarray]
+    num_items : int
+
+    Returns
+    -------
+    np.ndarray[uint64]
+
+    Should be the same as CPython's tupleobject.c
+    """
+    try:
+        first = next(arrays)
+    except StopIteration:
+        return np.array([], dtype=np.uint64)
+
+    arrays = itertools.chain([first], arrays)
+
+    mult = np.uint64(1000003)
+    out = np.zeros_like(first) + np.uint64(0x345678)
+    last_i = 0
+    for i, a in enumerate(arrays):
+        inverse_i = num_items - i
+        out ^= a
+        out *= mult
+        mult += np.uint64(82520 + inverse_i + inverse_i)
+        last_i = i
+    assert last_i + 1 == num_items, "Fed in wrong num_items"
+    out += np.uint64(97531)
+    return out
+
+
+def hash_pandas_object(
+    obj: Index | DataFrame | Series,
+    index: bool = True,
+    encoding: str = "utf8",
+    hash_key: str | None = _default_hash_key,
+    categorize: bool = True,
+) -> Series:
+    """
+    Return a data hash of the Index/Series/DataFrame.
+
+    Parameters
+    ----------
+    obj : Index, Series, or DataFrame
+    index : bool, default True
+        Include the index in the hash (if Series/DataFrame).
+    encoding : str, default 'utf8'
+        Encoding for data & key when strings.
+    hash_key : str, default _default_hash_key
+        Hash_key for string key to encode.
+    categorize : bool, default True
+        Whether to first categorize object arrays before hashing. This is more
+        efficient when the array contains duplicate values.
+
+    Returns
+    -------
+    Series of uint64
+        Same length as the object.
+
+    Examples
+    --------
+    >>> pd.util.hash_pandas_object(pd.Series([1, 2, 3]))
+    0    14639053686158035780
+    1     3869563279212530728
+    2      393322362522515241
+    dtype: uint64
+    """
+    from pandas import Series
+
+    if hash_key is None:
+        hash_key = _default_hash_key
+
+    if isinstance(obj, ABCMultiIndex):
+        return Series(hash_tuples(obj, encoding, hash_key), dtype="uint64", copy=False)
+
+    elif isinstance(obj, ABCIndex):
+        h = hash_array(obj._values, encoding, hash_key, categorize).astype(
+            "uint64", copy=False
+        )
+        ser = Series(h, index=obj, dtype="uint64", copy=False)
+
+    elif isinstance(obj, ABCSeries):
+        h = hash_array(obj._values, encoding, hash_key, categorize).astype(
+            "uint64", copy=False
+        )
+        if index:
+            index_iter = (
+                hash_pandas_object(
+                    obj.index,
+                    index=False,
+                    encoding=encoding,
+                    hash_key=hash_key,
+                    categorize=categorize,
+                )._values
+                for _ in [None]
+            )
+            arrays = itertools.chain([h], index_iter)
+            h = combine_hash_arrays(arrays, 2)
+
+        ser = Series(h, index=obj.index, dtype="uint64", copy=False)
+
+    elif isinstance(obj, ABCDataFrame):
+        hashes = (
+            hash_array(series._values, encoding, hash_key, categorize)
+            for _, series in obj.items()
+        )
+        num_items = len(obj.columns)
+        if index:
+            index_hash_generator = (
+                hash_pandas_object(
+                    obj.index,
+                    index=False,
+                    encoding=encoding,
+                    hash_key=hash_key,
+                    categorize=categorize,
+                )._values
+                for _ in [None]
+            )
+            num_items += 1
+
+            # keep `hashes` specifically a generator to keep mypy happy
+            _hashes = itertools.chain(hashes, index_hash_generator)
+            hashes = (x for x in _hashes)
+        h = combine_hash_arrays(hashes, num_items)
+
+        ser = Series(h, index=obj.index, dtype="uint64", copy=False)
+    else:
+        raise TypeError(f"Unexpected type for hashing {type(obj)}")
+
+    return ser
+
+
+def hash_tuples(
+    vals: MultiIndex | Iterable[tuple[Hashable, ...]],
+    encoding: str = "utf8",
+    hash_key: str = _default_hash_key,
+) -> npt.NDArray[np.uint64]:
+    """
+    Hash an MultiIndex / listlike-of-tuples efficiently.
+
+    Parameters
+    ----------
+    vals : MultiIndex or listlike-of-tuples
+    encoding : str, default 'utf8'
+    hash_key : str, default _default_hash_key
+
+    Returns
+    -------
+    ndarray[np.uint64] of hashed values
+    """
+    if not is_list_like(vals):
+        raise TypeError("must be convertible to a list-of-tuples")
+
+    from pandas import (
+        Categorical,
+        MultiIndex,
+    )
+
+    if not isinstance(vals, ABCMultiIndex):
+        mi = MultiIndex.from_tuples(vals)
+    else:
+        mi = vals
+
+    # create a list-of-Categoricals
+    cat_vals = [
+        Categorical._simple_new(
+            mi.codes[level],
+            CategoricalDtype(categories=mi.levels[level], ordered=False),
+        )
+        for level in range(mi.nlevels)
+    ]
+
+    # hash the list-of-ndarrays
+    hashes = (
+        cat._hash_pandas_object(encoding=encoding, hash_key=hash_key, categorize=False)
+        for cat in cat_vals
+    )
+    h = combine_hash_arrays(hashes, len(cat_vals))
+
+    return h
+
+
+def hash_array(
+    vals: ArrayLike,
+    encoding: str = "utf8",
+    hash_key: str = _default_hash_key,
+    categorize: bool = True,
+) -> npt.NDArray[np.uint64]:
+    """
+    Given a 1d array, return an array of deterministic integers.
+
+    Parameters
+    ----------
+    vals : ndarray or ExtensionArray
+        The input array to hash.
+    encoding : str, default 'utf8'
+        Encoding for data & key when strings.
+    hash_key : str, default _default_hash_key
+        Hash_key for string key to encode.
+    categorize : bool, default True
+        Whether to first categorize object arrays before hashing. This is more
+        efficient when the array contains duplicate values.
+
+    Returns
+    -------
+    ndarray[np.uint64, ndim=1]
+        Hashed values, same length as the vals.
+
+    See Also
+    --------
+    util.hash_pandas_object : Return a data hash of the Index/Series/DataFrame.
+    util.hash_tuples : Hash an MultiIndex / listlike-of-tuples efficiently.
+
+    Examples
+    --------
+    >>> pd.util.hash_array(np.array([1, 2, 3]))
+    array([ 6238072747940578789, 15839785061582574730,  2185194620014831856],
+      dtype=uint64)
+    """
+    if not hasattr(vals, "dtype"):
+        raise TypeError("must pass a ndarray-like")
+
+    if isinstance(vals, ABCExtensionArray):
+        return vals._hash_pandas_object(
+            encoding=encoding, hash_key=hash_key, categorize=categorize
+        )
+
+    if not isinstance(vals, np.ndarray):
+        # GH#42003
+        raise TypeError(
+            "hash_array requires np.ndarray or ExtensionArray, not "
+            f"{type(vals).__name__}. Use hash_pandas_object instead."
+        )
+
+    return _hash_ndarray(vals, encoding, hash_key, categorize)
+
+
+def _hash_ndarray(
+    vals: np.ndarray,
+    encoding: str = "utf8",
+    hash_key: str = _default_hash_key,
+    categorize: bool = True,
+) -> npt.NDArray[np.uint64]:
+    """
+    See hash_array.__doc__.
+    """
+    dtype = vals.dtype
+
+    # _hash_ndarray only takes 64-bit values, so handle 128-bit by parts
+    if np.issubdtype(dtype, np.complex128):
+        hash_real = _hash_ndarray(vals.real, encoding, hash_key, categorize)
+        hash_imag = _hash_ndarray(vals.imag, encoding, hash_key, categorize)
+        return hash_real + 23 * hash_imag
+
+    # First, turn whatever array this is into unsigned 64-bit ints, if we can
+    # manage it.
+    if dtype == bool:
+        vals = vals.astype("u8")
+    elif issubclass(dtype.type, (np.datetime64, np.timedelta64)):
+        vals = vals.view("i8").astype("u8", copy=False)
+    elif issubclass(dtype.type, np.number) and dtype.itemsize <= 8:
+        vals = vals.view(f"u{vals.dtype.itemsize}").astype("u8")
+    else:
+        # With repeated values, its MUCH faster to categorize object dtypes,
+        # then hash and rename categories. We allow skipping the categorization
+        # when the values are known/likely to be unique.
+        if categorize:
+            from pandas import (
+                Categorical,
+                Index,
+                factorize,
+            )
+
+            codes, categories = factorize(vals, sort=False)
+            dtype = CategoricalDtype(categories=Index(categories), ordered=False)
+            cat = Categorical._simple_new(codes, dtype)
+            return cat._hash_pandas_object(
+                encoding=encoding, hash_key=hash_key, categorize=False
+            )
+
+        try:
+            vals = hash_object_array(vals, hash_key, encoding)
+        except TypeError:
+            # we have mixed types
+            vals = hash_object_array(
+                vals.astype(str).astype(object), hash_key, encoding
+            )
+
+    # Then, redistribute these 64-bit ints within the space of 64-bit ints
+    vals ^= vals >> 30
+    vals *= np.uint64(0xBF58476D1CE4E5B9)
+    vals ^= vals >> 27
+    vals *= np.uint64(0x94D049BB133111EB)
+    vals ^= vals >> 31
+    return vals
diff --git a/pandas/pandas/core/util/numba_.py b/pandas/pandas/core/util/numba_.py
new file mode 100644
index 0000000000000..d3f00c08e0e2c
--- /dev/null
+++ b/pandas/pandas/core/util/numba_.py
@@ -0,0 +1,150 @@
+"""Common utilities for Numba operations"""
+
+from __future__ import annotations
+
+import inspect
+import types
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import NumbaUtilError
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+GLOBAL_USE_NUMBA: bool = False
+
+
+def maybe_use_numba(engine: str | None) -> bool:
+    """Signal whether to use numba routines."""
+    return engine == "numba" or (engine is None and GLOBAL_USE_NUMBA)
+
+
+def set_use_numba(enable: bool = False) -> None:
+    global GLOBAL_USE_NUMBA
+    if enable:
+        import_optional_dependency("numba")
+    GLOBAL_USE_NUMBA = enable
+
+
+def get_jit_arguments(engine_kwargs: dict[str, bool] | None = None) -> dict[str, bool]:
+    """
+    Return arguments to pass to numba.JIT, falling back on pandas default JIT settings.
+
+    Parameters
+    ----------
+    engine_kwargs : dict, default None
+        user passed keyword arguments for numba.JIT
+
+    Returns
+    -------
+    dict[str, bool]
+        nopython, nogil, parallel
+
+    Raises
+    ------
+    NumbaUtilError
+    """
+    if engine_kwargs is None:
+        engine_kwargs = {}
+
+    nopython = engine_kwargs.get("nopython", True)
+    nogil = engine_kwargs.get("nogil", False)
+    parallel = engine_kwargs.get("parallel", False)
+    return {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+
+
+def jit_user_function(func: Callable) -> Callable:
+    """
+    If user function is not jitted already, mark the user's function
+    as jitable.
+
+    Parameters
+    ----------
+    func : function
+        user defined function
+
+    Returns
+    -------
+    function
+        Numba JITed function, or function marked as JITable by numba
+    """
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    if numba.extending.is_jitted(func):
+        # Don't jit a user passed jitted function
+        numba_func = func
+    elif getattr(np, func.__name__, False) is func or isinstance(
+        func, types.BuiltinFunctionType
+    ):
+        # Not necessary to jit builtins or np functions
+        # This will mess up register_jitable
+        numba_func = func
+    else:
+        numba_func = numba.extending.register_jitable(func)
+
+    return numba_func
+
+
+_sentinel = object()
+
+
+def prepare_function_arguments(
+    func: Callable, args: tuple, kwargs: dict, *, num_required_args: int
+) -> tuple[tuple, dict]:
+    """
+    Prepare arguments for jitted function. As numba functions do not support kwargs,
+    we try to move kwargs into args if possible.
+
+    Parameters
+    ----------
+    func : function
+        User defined function
+    args : tuple
+        User input positional arguments
+    kwargs : dict
+        User input keyword arguments
+    num_required_args : int
+        The number of leading positional arguments we will pass to udf.
+        These are not supplied by the user.
+        e.g. for groupby we require "values", "index" as the first two arguments:
+        `numba_func(group, group_index, *args)`, in this case num_required_args=2.
+        See :func:`pandas.core.groupby.numba_.generate_numba_agg_func`
+
+    Returns
+    -------
+    tuple[tuple, dict]
+        args, kwargs
+
+    """
+    if not kwargs:
+        return args, kwargs
+
+    # the udf should have this pattern: def udf(arg1, arg2, ..., *args, **kwargs):...
+    signature = inspect.signature(func)
+    arguments = signature.bind(*[_sentinel] * num_required_args, *args, **kwargs)
+    arguments.apply_defaults()
+    # Ref: https://peps.python.org/pep-0362/
+    # Arguments which could be passed as part of either *args or **kwargs
+    # will be included only in the BoundArguments.args attribute.
+    args = arguments.args
+    kwargs = arguments.kwargs
+
+    if kwargs:
+        # Note: in case numba supports keyword-only arguments in
+        # a future version, we should remove this check. But this
+        # seems unlikely to happen soon.
+
+        raise NumbaUtilError(
+            "numba does not support keyword-only arguments"
+            "https://github.com/numba/numba/issues/2916, "
+            "https://github.com/numba/numba/issues/6846"
+        )
+
+    args = args[num_required_args:]
+    return args, kwargs
diff --git a/pandas/pandas/core/window/__init__.py b/pandas/pandas/core/window/__init__.py
new file mode 100644
index 0000000000000..857e12e5467a6
--- /dev/null
+++ b/pandas/pandas/core/window/__init__.py
@@ -0,0 +1,23 @@
+from pandas.core.window.ewm import (
+    ExponentialMovingWindow,
+    ExponentialMovingWindowGroupby,
+)
+from pandas.core.window.expanding import (
+    Expanding,
+    ExpandingGroupby,
+)
+from pandas.core.window.rolling import (
+    Rolling,
+    RollingGroupby,
+    Window,
+)
+
+__all__ = [
+    "Expanding",
+    "ExpandingGroupby",
+    "ExponentialMovingWindow",
+    "ExponentialMovingWindowGroupby",
+    "Rolling",
+    "RollingGroupby",
+    "Window",
+]
diff --git a/pandas/pandas/core/window/common.py b/pandas/pandas/core/window/common.py
new file mode 100644
index 0000000000000..004a3555f0212
--- /dev/null
+++ b/pandas/pandas/core/window/common.py
@@ -0,0 +1,170 @@
+"""Common utility functions for rolling operations"""
+
+from __future__ import annotations
+
+from collections import defaultdict
+from typing import cast
+
+import numpy as np
+
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+from pandas.core.indexes.api import MultiIndex
+
+
+def flex_binary_moment(arg1, arg2, f, pairwise: bool = False):
+    if isinstance(arg1, ABCSeries) and isinstance(arg2, ABCSeries):
+        X, Y = prep_binary(arg1, arg2)
+        return f(X, Y)
+
+    elif isinstance(arg1, ABCDataFrame):
+        from pandas import DataFrame
+
+        def dataframe_from_int_dict(data, frame_template) -> DataFrame:
+            result = DataFrame(data, index=frame_template.index)
+            if len(result.columns) > 0:
+                result.columns = frame_template.columns[result.columns]
+            else:
+                result.columns = frame_template.columns.copy()
+            return result
+
+        results = {}
+        if isinstance(arg2, ABCDataFrame):
+            if pairwise is False:
+                if arg1 is arg2:
+                    # special case in order to handle duplicate column names
+                    for i in range(len(arg1.columns)):
+                        results[i] = f(arg1.iloc[:, i], arg2.iloc[:, i])
+                    return dataframe_from_int_dict(results, arg1)
+                else:
+                    if not arg1.columns.is_unique:
+                        raise ValueError("'arg1' columns are not unique")
+                    if not arg2.columns.is_unique:
+                        raise ValueError("'arg2' columns are not unique")
+                    X, Y = arg1.align(arg2, join="outer")
+                    X, Y = prep_binary(X, Y)
+                    res_columns = arg1.columns.union(arg2.columns)
+                    for col in res_columns:
+                        if col in X and col in Y:
+                            results[col] = f(X[col], Y[col])
+                    return DataFrame(results, index=X.index, columns=res_columns)
+            elif pairwise is True:
+                results = defaultdict(dict)
+                for i in range(len(arg1.columns)):
+                    for j in range(len(arg2.columns)):
+                        if j < i and arg2 is arg1:
+                            # Symmetric case
+                            results[i][j] = results[j][i]
+                        else:
+                            results[i][j] = f(
+                                *prep_binary(arg1.iloc[:, i], arg2.iloc[:, j])
+                            )
+
+                from pandas import concat
+
+                result_index = arg1.index.union(arg2.index)
+                if len(result_index):
+                    # construct result frame
+                    result = concat(
+                        [
+                            concat(
+                                [results[i][j] for j in range(len(arg2.columns))],
+                                ignore_index=True,
+                            )
+                            for i in range(len(arg1.columns))
+                        ],
+                        ignore_index=True,
+                        axis=1,
+                    )
+                    result.columns = arg1.columns
+
+                    # set the index and reorder
+                    if arg2.columns.nlevels > 1:
+                        # mypy needs to know columns is a MultiIndex, Index doesn't
+                        # have levels attribute
+                        arg2.columns = cast(MultiIndex, arg2.columns)
+                        # GH 21157: Equivalent to MultiIndex.from_product(
+                        #  [result_index], <unique combinations of arg2.columns.levels>,
+                        # )
+                        # A normal MultiIndex.from_product will produce too many
+                        # combinations.
+                        result_level = np.tile(
+                            result_index, len(result) // len(result_index)
+                        )
+                        arg2_levels = (
+                            np.repeat(
+                                arg2.columns.get_level_values(i),
+                                len(result) // len(arg2.columns),
+                            )
+                            for i in range(arg2.columns.nlevels)
+                        )
+                        result_names = list(arg2.columns.names) + [result_index.name]
+                        result.index = MultiIndex.from_arrays(
+                            [*arg2_levels, result_level], names=result_names
+                        )
+                        # GH 34440
+                        num_levels = len(result.index.levels)
+                        new_order = [num_levels - 1] + list(range(num_levels - 1))
+                        result = result.reorder_levels(new_order).sort_index()
+                    else:
+                        result.index = MultiIndex.from_product(
+                            [range(len(arg2.columns)), range(len(result_index))]
+                        )
+                        result = result.swaplevel(1, 0).sort_index()
+                        result.index = MultiIndex.from_product(
+                            [result_index] + [arg2.columns]
+                        )
+                else:
+                    # empty result
+                    result = DataFrame(
+                        index=MultiIndex(
+                            levels=[arg1.index, arg2.columns], codes=[[], []]
+                        ),
+                        columns=arg2.columns,
+                        dtype="float64",
+                    )
+
+                # reset our index names to arg1 names
+                # reset our column names to arg2 names
+                # careful not to mutate the original names
+                result.columns = result.columns.set_names(arg1.columns.names)
+                result.index = result.index.set_names(
+                    result_index.names + arg2.columns.names
+                )
+
+                return result
+        else:
+            results = {
+                i: f(*prep_binary(arg1.iloc[:, i], arg2))
+                for i in range(len(arg1.columns))
+            }
+            return dataframe_from_int_dict(results, arg1)
+
+    else:
+        return flex_binary_moment(arg2, arg1, f)
+
+
+def zsqrt(x):
+    with np.errstate(all="ignore"):
+        result = np.sqrt(x)
+        mask = x < 0
+
+    if isinstance(x, ABCDataFrame):
+        if mask._values.any():
+            result[mask] = 0
+    else:
+        if mask.any():
+            result[mask] = 0
+
+    return result
+
+
+def prep_binary(arg1, arg2):
+    # mask out values, this also makes a common index...
+    X = arg1 + 0 * arg2
+    Y = arg2 + 0 * arg1
+
+    return X, Y
diff --git a/pandas/pandas/core/window/doc.py b/pandas/pandas/core/window/doc.py
new file mode 100644
index 0000000000000..6dbc52a99e70c
--- /dev/null
+++ b/pandas/pandas/core/window/doc.py
@@ -0,0 +1,174 @@
+"""Any shareable docstring components for rolling/expanding/ewm"""
+
+from __future__ import annotations
+
+from textwrap import dedent
+
+from pandas.core.shared_docs import _shared_docs
+
+_shared_docs = dict(**_shared_docs)
+
+
+def create_section_header(header: str) -> str:
+    """Create numpydoc section header"""
+    return f"{header}\n{'-' * len(header)}\n"
+
+
+template_header = "\nCalculate the {window_method} {aggregation_description}.\n\n"
+
+template_returns = dedent(
+    """
+    Series or DataFrame
+        Return type is the same as the original object with ``np.float64`` dtype.\n
+    """
+).replace("\n", "", 1)
+
+template_see_also = dedent(
+    """
+    Series.{window_method} : Calling {window_method} with Series data.
+    DataFrame.{window_method} : Calling {window_method} with DataFrames.
+    Series.{agg_method} : Aggregating {agg_method} for Series.
+    DataFrame.{agg_method} : Aggregating {agg_method} for DataFrame.\n
+    """
+).replace("\n", "", 1)
+
+kwargs_numeric_only = dedent(
+    """
+    numeric_only : bool, default False
+        Include only float, int, boolean columns.
+
+        .. versionadded:: 1.5.0\n
+    """
+).replace("\n", "", 1)
+
+kwargs_scipy = dedent(
+    """
+    **kwargs
+        Keyword arguments to configure the ``SciPy`` weighted window type.\n
+    """
+).replace("\n", "", 1)
+
+window_apply_parameters = dedent(
+    """
+    func : function
+        Must produce a single value from an ndarray input if ``raw=True``
+        or a single value from a Series if ``raw=False``. Can also accept a
+        Numba JIT function with ``engine='numba'`` specified.
+
+    raw : bool, default False
+        * ``False`` : passes each row or column as a Series to the
+          function.
+        * ``True`` : the passed function will receive ndarray
+          objects instead.
+          If you are just applying a NumPy reduction function this will
+          achieve much better performance.
+
+    engine : str, default None
+        * ``'cython'`` : Runs rolling apply through C-extensions from cython.
+        * ``'numba'`` : Runs rolling apply through JIT compiled code from numba.
+          Only available when ``raw`` is set to ``True``.
+        * ``None`` : Defaults to ``'cython'`` or globally setting ``compute.use_numba``
+
+    engine_kwargs : dict, default None
+        * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+        * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+          and ``parallel`` dictionary keys. The values must either be ``True`` or
+          ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+          ``{{'nopython': True, 'nogil': False, 'parallel': False}}`` and will be
+          applied to both the ``func`` and the ``apply`` rolling aggregation.
+
+    args : tuple, default None
+        Positional arguments to be passed into func.
+
+    kwargs : dict, default None
+        Keyword arguments to be passed into func.\n
+    """
+).replace("\n", "", 1)
+
+template_pipe = """
+Apply a ``func`` with arguments to this %(klass)s object and return its result.
+
+Use `.pipe` when you want to improve readability by chaining together
+functions that expect Series, DataFrames, GroupBy, Rolling, Expanding or Resampler
+objects.
+Instead of writing
+
+>>> h = lambda x, arg2, arg3: x + 1 - arg2 * arg3
+>>> g = lambda x, arg1: x * 5 / arg1
+>>> f = lambda x: x ** 4
+>>> df = pd.DataFrame({'A': [1, 2, 3, 4]}, index=pd.date_range('2012-08-02', periods=4))
+>>> h(g(f(df.rolling('2D')), arg1=1), arg2=2, arg3=3)  # doctest: +SKIP
+
+You can write
+
+>>> (df.rolling('2D')
+...    .pipe(f)
+...    .pipe(g, arg1=1)
+...    .pipe(h, arg2=2, arg3=3))  # doctest: +SKIP
+
+which is much more readable.
+
+Parameters
+----------
+func : callable or tuple of (callable, str)
+    Function to apply to this %(klass)s object or, alternatively,
+    a `(callable, data_keyword)` tuple where `data_keyword` is a
+    string indicating the keyword of `callable` that expects the
+    %(klass)s object.
+*args : iterable, optional
+       Positional arguments passed into `func`.
+**kwargs : dict, optional
+         A dictionary of keyword arguments passed into `func`.
+
+Returns
+-------
+%(klass)s
+    The original object with the function `func` applied.
+
+See Also
+--------
+Series.pipe : Apply a function with arguments to a series.
+DataFrame.pipe: Apply a function with arguments to a dataframe.
+apply : Apply function to each group instead of to the
+    full %(klass)s object.
+
+Notes
+-----
+See more `here
+<https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html#piping-function-calls>`_
+
+Examples
+--------
+%(examples)s
+"""
+
+numba_notes = (
+    "See :ref:`window.numba_engine` and :ref:`enhancingperf.numba` for "
+    "extended documentation and performance considerations for the Numba engine.\n\n"
+)
+
+
+def window_agg_numba_parameters(version: str = "1.3") -> str:
+    return (
+        dedent(
+            """
+    engine : str, default None
+        * ``'cython'`` : Runs the operation through C-extensions from cython.
+        * ``'numba'`` : Runs the operation through JIT compiled code from numba.
+        * ``None`` : Defaults to ``'cython'`` or globally setting ``compute.use_numba``
+
+          .. versionadded:: {version}.0
+
+    engine_kwargs : dict, default None
+        * For ``'cython'`` engine, there are no accepted ``engine_kwargs``
+        * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+          and ``parallel`` dictionary keys. The values must either be ``True`` or
+          ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+          ``{{'nopython': True, 'nogil': False, 'parallel': False}}``
+
+          .. versionadded:: {version}.0\n
+    """
+        )
+        .replace("\n", "", 1)
+        .replace("{version}", version)
+    )
diff --git a/pandas/pandas/core/window/ewm.py b/pandas/pandas/core/window/ewm.py
new file mode 100644
index 0000000000000..73e4de6ea6208
--- /dev/null
+++ b/pandas/pandas/core/window/ewm.py
@@ -0,0 +1,1096 @@
+from __future__ import annotations
+
+import datetime
+from functools import partial
+from textwrap import dedent
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs.tslibs import Timedelta
+import pandas._libs.window.aggregations as window_aggregations
+from pandas.util._decorators import doc
+
+from pandas.core.dtypes.common import (
+    is_datetime64_dtype,
+    is_numeric_dtype,
+)
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.generic import ABCSeries
+from pandas.core.dtypes.missing import isna
+
+from pandas.core import common
+from pandas.core.arrays.datetimelike import dtype_to_unit
+from pandas.core.indexers.objects import (
+    BaseIndexer,
+    ExponentialMovingWindowIndexer,
+    GroupbyIndexer,
+)
+from pandas.core.util.numba_ import (
+    get_jit_arguments,
+    maybe_use_numba,
+)
+from pandas.core.window.common import zsqrt
+from pandas.core.window.doc import (
+    _shared_docs,
+    create_section_header,
+    kwargs_numeric_only,
+    numba_notes,
+    template_header,
+    template_returns,
+    template_see_also,
+    window_agg_numba_parameters,
+)
+from pandas.core.window.numba_ import (
+    generate_numba_ewm_func,
+    generate_numba_ewm_table_func,
+)
+from pandas.core.window.online import (
+    EWMMeanState,
+    generate_online_numba_ewma_func,
+)
+from pandas.core.window.rolling import (
+    BaseWindow,
+    BaseWindowGroupby,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        TimedeltaConvertibleTypes,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+    from pandas.core.generic import NDFrame
+
+
+def get_center_of_mass(
+    comass: float | None,
+    span: float | None,
+    halflife: float | None,
+    alpha: float | None,
+) -> float:
+    valid_count = common.count_not_none(comass, span, halflife, alpha)
+    if valid_count > 1:
+        raise ValueError("comass, span, halflife, and alpha are mutually exclusive")
+
+    # Convert to center of mass; domain checks ensure 0 < alpha <= 1
+    if comass is not None:
+        if comass < 0:
+            raise ValueError("comass must satisfy: comass >= 0")
+    elif span is not None:
+        if span < 1:
+            raise ValueError("span must satisfy: span >= 1")
+        comass = (span - 1) / 2
+    elif halflife is not None:
+        if halflife <= 0:
+            raise ValueError("halflife must satisfy: halflife > 0")
+        decay = 1 - np.exp(np.log(0.5) / halflife)
+        comass = 1 / decay - 1
+    elif alpha is not None:
+        if alpha <= 0 or alpha > 1:
+            raise ValueError("alpha must satisfy: 0 < alpha <= 1")
+        comass = (1 - alpha) / alpha
+    else:
+        raise ValueError("Must pass one of comass, span, halflife, or alpha")
+
+    return float(comass)
+
+
+def _calculate_deltas(
+    times: np.ndarray | NDFrame,
+    halflife: float | TimedeltaConvertibleTypes | None,
+) -> npt.NDArray[np.float64]:
+    """
+    Return the diff of the times divided by the half-life. These values are used in
+    the calculation of the ewm mean.
+
+    Parameters
+    ----------
+    times : np.ndarray, Series
+        Times corresponding to the observations. Must be monotonically increasing
+        and ``datetime64[ns]`` dtype.
+    halflife : float, str, timedelta, optional
+        Half-life specifying the decay
+
+    Returns
+    -------
+    np.ndarray
+        Diff of the times divided by the half-life
+    """
+    unit = dtype_to_unit(times.dtype)
+    if isinstance(times, ABCSeries):
+        times = times._values
+    _times = np.asarray(times.view(np.int64), dtype=np.float64)
+    _halflife = float(Timedelta(halflife).as_unit(unit)._value)
+    return np.diff(_times) / _halflife
+
+
+class ExponentialMovingWindow(BaseWindow):
+    r"""
+    Provide exponentially weighted (EW) calculations.
+
+    Exactly one of ``com``, ``span``, ``halflife``, or ``alpha`` must be
+    provided if ``times`` is not provided. If ``times`` is provided and ``adjust=True``,
+    ``halflife`` and one of ``com``, ``span`` or ``alpha`` may be provided.
+    If ``times`` is provided and ``adjust=False``, ``halflife`` must be the only
+    provided decay-specification parameter.
+
+    Parameters
+    ----------
+    com : float, optional
+        Specify decay in terms of center of mass
+
+        :math:`\alpha = 1 / (1 + com)`, for :math:`com \geq 0`.
+
+    span : float, optional
+        Specify decay in terms of span
+
+        :math:`\alpha = 2 / (span + 1)`, for :math:`span \geq 1`.
+
+    halflife : float, str, timedelta, optional
+        Specify decay in terms of half-life
+
+        :math:`\alpha = 1 - \exp\left(-\ln(2) / halflife\right)`, for
+        :math:`halflife > 0`.
+
+        If ``times`` is specified, a timedelta convertible unit over which an
+        observation decays to half its value. Only applicable to ``mean()``,
+        and halflife value will not apply to the other functions.
+
+    alpha : float, optional
+        Specify smoothing factor :math:`\alpha` directly
+
+        :math:`0 < \alpha \leq 1`.
+
+    min_periods : int, default 0
+        Minimum number of observations in window required to have a value;
+        otherwise, result is ``np.nan``.
+
+    adjust : bool, default True
+        Divide by decaying adjustment factor in beginning periods to account
+        for imbalance in relative weightings (viewing EWMA as a moving average).
+
+        - When ``adjust=True`` (default), the EW function is calculated using weights
+          :math:`w_i = (1 - \alpha)^i`. For example, the EW moving average of the series
+          [:math:`x_0, x_1, ..., x_t`] would be:
+
+        .. math::
+            y_t = \frac{x_t + (1 - \alpha)x_{t-1} + (1 - \alpha)^2 x_{t-2} + ... + (1 -
+            \alpha)^t x_0}{1 + (1 - \alpha) + (1 - \alpha)^2 + ... + (1 - \alpha)^t}
+
+        - When ``adjust=False``, the exponentially weighted function is calculated
+          recursively:
+
+        .. math::
+            \begin{split}
+                y_0 &= x_0\\
+                y_t &= (1 - \alpha) y_{t-1} + \alpha x_t,
+            \end{split}
+    ignore_na : bool, default False
+        Ignore missing values when calculating weights.
+
+        - When ``ignore_na=False`` (default), weights are based on absolute positions.
+          For example, the weights of :math:`x_0` and :math:`x_2` used in calculating
+          the final weighted average of [:math:`x_0`, None, :math:`x_2`] are
+          :math:`(1-\alpha)^2` and :math:`1` if ``adjust=True``, and
+          :math:`(1-\alpha)^2` and :math:`\alpha` if ``adjust=False``.
+
+        - When ``ignore_na=True``, weights are based
+          on relative positions. For example, the weights of :math:`x_0` and :math:`x_2`
+          used in calculating the final weighted average of
+          [:math:`x_0`, None, :math:`x_2`] are :math:`1-\alpha` and :math:`1` if
+          ``adjust=True``, and :math:`1-\alpha` and :math:`\alpha` if ``adjust=False``.
+
+    times : np.ndarray, Series, default None
+
+        Only applicable to ``mean()``.
+
+        Times corresponding to the observations. Must be monotonically increasing and
+        ``datetime64[ns]`` dtype.
+
+        If 1-D array like, a sequence with the same shape as the observations.
+
+    method : str {'single', 'table'}, default 'single'
+        .. versionadded:: 1.4.0
+
+        Execute the rolling operation per single column or row (``'single'``)
+        or over the entire object (``'table'``).
+
+        This argument is only implemented when specifying ``engine='numba'``
+        in the method call.
+
+        Only applicable to ``mean()``
+
+    Returns
+    -------
+    pandas.api.typing.ExponentialMovingWindow
+        An instance of ExponentialMovingWindow for further exponentially weighted (EW)
+        calculations, e.g. using the ``mean`` method.
+
+    See Also
+    --------
+    rolling : Provides rolling window calculations.
+    expanding : Provides expanding transformations.
+
+    Notes
+    -----
+    See :ref:`Windowing Operations <window.exponentially_weighted>`
+    for further usage details and examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({'B': [0, 1, 2, np.nan, 4]})
+    >>> df
+         B
+    0  0.0
+    1  1.0
+    2  2.0
+    3  NaN
+    4  4.0
+
+    >>> df.ewm(com=0.5).mean()
+              B
+    0  0.000000
+    1  0.750000
+    2  1.615385
+    3  1.615385
+    4  3.670213
+    >>> df.ewm(alpha=2 / 3).mean()
+              B
+    0  0.000000
+    1  0.750000
+    2  1.615385
+    3  1.615385
+    4  3.670213
+
+    **adjust**
+
+    >>> df.ewm(com=0.5, adjust=True).mean()
+              B
+    0  0.000000
+    1  0.750000
+    2  1.615385
+    3  1.615385
+    4  3.670213
+    >>> df.ewm(com=0.5, adjust=False).mean()
+              B
+    0  0.000000
+    1  0.666667
+    2  1.555556
+    3  1.555556
+    4  3.650794
+
+    **ignore_na**
+
+    >>> df.ewm(com=0.5, ignore_na=True).mean()
+              B
+    0  0.000000
+    1  0.750000
+    2  1.615385
+    3  1.615385
+    4  3.225000
+    >>> df.ewm(com=0.5, ignore_na=False).mean()
+              B
+    0  0.000000
+    1  0.750000
+    2  1.615385
+    3  1.615385
+    4  3.670213
+
+    **times**
+
+    Exponentially weighted mean with weights calculated with a timedelta ``halflife``
+    relative to ``times``.
+
+    >>> times = ['2020-01-01', '2020-01-03', '2020-01-10', '2020-01-15', '2020-01-17']
+    >>> df.ewm(halflife='4 days', times=pd.DatetimeIndex(times)).mean()
+              B
+    0  0.000000
+    1  0.585786
+    2  1.523889
+    3  1.523889
+    4  3.233686
+    """
+
+    _attributes = [
+        "com",
+        "span",
+        "halflife",
+        "alpha",
+        "min_periods",
+        "adjust",
+        "ignore_na",
+        "times",
+        "method",
+    ]
+
+    def __init__(
+        self,
+        obj: NDFrame,
+        com: float | None = None,
+        span: float | None = None,
+        halflife: float | TimedeltaConvertibleTypes | None = None,
+        alpha: float | None = None,
+        min_periods: int | None = 0,
+        adjust: bool = True,
+        ignore_na: bool = False,
+        times: np.ndarray | NDFrame | None = None,
+        method: str = "single",
+        *,
+        selection=None,
+    ) -> None:
+        super().__init__(
+            obj=obj,
+            min_periods=1 if min_periods is None else max(int(min_periods), 1),
+            on=None,
+            center=False,
+            closed=None,
+            method=method,
+            selection=selection,
+        )
+        self.com = com
+        self.span = span
+        self.halflife = halflife
+        self.alpha = alpha
+        self.adjust = adjust
+        self.ignore_na = ignore_na
+        self.times = times
+        if self.times is not None:
+            times_dtype = getattr(self.times, "dtype", None)
+            if not (
+                is_datetime64_dtype(times_dtype)
+                or isinstance(times_dtype, DatetimeTZDtype)
+            ):
+                raise ValueError("times must be datetime64 dtype.")
+            if len(self.times) != len(obj):
+                raise ValueError("times must be the same length as the object.")
+            if not isinstance(self.halflife, (str, datetime.timedelta, np.timedelta64)):
+                raise ValueError("halflife must be a timedelta convertible object")
+            if isna(self.times).any():
+                raise ValueError("Cannot convert NaT values to integer")
+            self._deltas = _calculate_deltas(self.times, self.halflife)
+            # Halflife is no longer applicable when calculating COM
+            # But allow COM to still be calculated if the user passes other decay args
+            if common.count_not_none(self.com, self.span, self.alpha) > 0:
+                if not self.adjust:
+                    raise NotImplementedError(
+                        "None of com, span, or alpha can be specified if "
+                        "times is provided and adjust=False"
+                    )
+                self._com = get_center_of_mass(self.com, self.span, None, self.alpha)
+            else:
+                self._com = 1.0
+        else:
+            if self.halflife is not None and isinstance(
+                self.halflife, (str, datetime.timedelta, np.timedelta64)
+            ):
+                raise ValueError(
+                    "halflife can only be a timedelta convertible argument if "
+                    "times is not None."
+                )
+            # Without times, points are equally spaced
+            self._deltas = np.ones(max(self.obj.shape[0] - 1, 0), dtype=np.float64)
+            self._com = get_center_of_mass(
+                # error: Argument 3 to "get_center_of_mass" has incompatible type
+                # "Union[float, Any, None, timedelta64, signedinteger[_64Bit]]";
+                # expected "Optional[float]"
+                self.com,
+                self.span,
+                self.halflife,  # type: ignore[arg-type]
+                self.alpha,
+            )
+
+    def _check_window_bounds(
+        self, start: np.ndarray, end: np.ndarray, num_vals: int
+    ) -> None:
+        # emw algorithms are iterative with each point
+        # ExponentialMovingWindowIndexer "bounds" are the entire window
+        pass
+
+    def _get_window_indexer(self) -> BaseIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+        """
+        return ExponentialMovingWindowIndexer()
+
+    def online(
+        self, engine: str = "numba", engine_kwargs=None
+    ) -> OnlineExponentialMovingWindow:
+        """
+        Return an ``OnlineExponentialMovingWindow`` object to calculate
+        exponentially moving window aggregations in an online method.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        engine: str, default ``'numba'``
+            Execution engine to calculate online aggregations.
+            Applies to all supported aggregation methods.
+
+        engine_kwargs : dict, default None
+            Applies to all supported aggregation methods.
+
+            * For ``'numba'`` engine, the engine can accept ``nopython``, ``nogil``
+              and ``parallel`` dictionary keys. The values must either be ``True`` or
+              ``False``. The default ``engine_kwargs`` for the ``'numba'`` engine is
+              ``{{'nopython': True, 'nogil': False, 'parallel': False}}`` and will be
+              applied to the function
+
+        Returns
+        -------
+        OnlineExponentialMovingWindow
+        """
+        return OnlineExponentialMovingWindow(
+            obj=self.obj,
+            com=self.com,
+            span=self.span,
+            halflife=self.halflife,
+            alpha=self.alpha,
+            min_periods=self.min_periods,
+            adjust=self.adjust,
+            ignore_na=self.ignore_na,
+            times=self.times,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+            selection=self._selection,
+        )
+
+    @doc(
+        _shared_docs["aggregate"],
+        see_also=dedent(
+            """
+        See Also
+        --------
+        pandas.DataFrame.rolling.aggregate
+        """
+        ),
+        examples=dedent(
+            """
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        >>> df
+           A  B  C
+        0  1  4  7
+        1  2  5  8
+        2  3  6  9
+
+        >>> df.ewm(alpha=0.5).mean()
+                  A         B         C
+        0  1.000000  4.000000  7.000000
+        1  1.666667  4.666667  7.666667
+        2  2.428571  5.428571  8.428571
+        """
+        ),
+        klass="Series/Dataframe",
+        axis="",
+    )
+    def aggregate(self, func=None, *args, **kwargs):
+        return super().aggregate(func, *args, **kwargs)
+
+    agg = aggregate
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4])
+        >>> ser.ewm(alpha=.2).mean()
+        0    1.000000
+        1    1.555556
+        2    2.147541
+        3    2.775068
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) mean",
+        agg_method="mean",
+    )
+    def mean(
+        self,
+        numeric_only: bool = False,
+        engine=None,
+        engine_kwargs=None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "single":
+                func = generate_numba_ewm_func
+            else:
+                func = generate_numba_ewm_table_func
+            ewm_func = func(
+                **get_jit_arguments(engine_kwargs),
+                com=self._com,
+                adjust=self.adjust,
+                ignore_na=self.ignore_na,
+                deltas=tuple(self._deltas),
+                normalize=True,
+            )
+            return self._apply(ewm_func, name="mean")
+        elif engine in ("cython", None):
+            if engine_kwargs is not None:
+                raise ValueError("cython engine does not accept engine_kwargs")
+
+            deltas = None if self.times is None else self._deltas
+            window_func = partial(
+                window_aggregations.ewm,
+                com=self._com,
+                adjust=self.adjust,
+                ignore_na=self.ignore_na,
+                deltas=deltas,
+                normalize=True,
+            )
+            return self._apply(window_func, name="mean", numeric_only=numeric_only)
+        else:
+            raise ValueError("engine must be either 'numba' or 'cython'")
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4])
+        >>> ser.ewm(alpha=.2).sum()
+        0    1.000
+        1    2.800
+        2    5.240
+        3    8.192
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) sum",
+        agg_method="sum",
+    )
+    def sum(
+        self,
+        numeric_only: bool = False,
+        engine=None,
+        engine_kwargs=None,
+    ):
+        if not self.adjust:
+            raise NotImplementedError("sum is not implemented with adjust=False")
+        if self.times is not None:
+            raise NotImplementedError("sum is not implemented with times")
+        if maybe_use_numba(engine):
+            if self.method == "single":
+                func = generate_numba_ewm_func
+            else:
+                func = generate_numba_ewm_table_func
+            ewm_func = func(
+                **get_jit_arguments(engine_kwargs),
+                com=self._com,
+                adjust=self.adjust,
+                ignore_na=self.ignore_na,
+                deltas=tuple(self._deltas),
+                normalize=False,
+            )
+            return self._apply(ewm_func, name="sum")
+        elif engine in ("cython", None):
+            if engine_kwargs is not None:
+                raise ValueError("cython engine does not accept engine_kwargs")
+
+            deltas = None if self.times is None else self._deltas
+            window_func = partial(
+                window_aggregations.ewm,
+                com=self._com,
+                adjust=self.adjust,
+                ignore_na=self.ignore_na,
+                deltas=deltas,
+                normalize=False,
+            )
+            return self._apply(window_func, name="sum", numeric_only=numeric_only)
+        else:
+            raise ValueError("engine must be either 'numba' or 'cython'")
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """\
+        bias : bool, default False
+            Use a standard estimation bias correction.
+        """
+        ),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4])
+        >>> ser.ewm(alpha=.2).std()
+        0         NaN
+        1    0.707107
+        2    0.995893
+        3    1.277320
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) standard deviation",
+        agg_method="std",
+    )
+    def std(self, bias: bool = False, numeric_only: bool = False):
+        if (
+            numeric_only
+            and self._selected_obj.ndim == 1
+            and not is_numeric_dtype(self._selected_obj.dtype)
+        ):
+            # Raise directly so error message says std instead of var
+            raise NotImplementedError(
+                f"{type(self).__name__}.std does not implement numeric_only"
+            )
+        if self.times is not None:
+            raise NotImplementedError("std is not implemented with times")
+        return zsqrt(self.var(bias=bias, numeric_only=numeric_only))
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """\
+        bias : bool, default False
+            Use a standard estimation bias correction.
+        """
+        ),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4])
+        >>> ser.ewm(alpha=.2).var()
+        0         NaN
+        1    0.500000
+        2    0.991803
+        3    1.631547
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) variance",
+        agg_method="var",
+    )
+    def var(self, bias: bool = False, numeric_only: bool = False):
+        if self.times is not None:
+            raise NotImplementedError("var is not implemented with times")
+        window_func = window_aggregations.ewmcov
+        wfunc = partial(
+            window_func,
+            com=self._com,
+            adjust=self.adjust,
+            ignore_na=self.ignore_na,
+            bias=bias,
+        )
+
+        def var_func(values, begin, end, min_periods):
+            return wfunc(values, begin, end, min_periods, values)
+
+        return self._apply(var_func, name="var", numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """\
+        other : Series or DataFrame , optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndex DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        bias : bool, default False
+            Use a standard estimation bias correction.
+        """
+        ),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser1 = pd.Series([1, 2, 3, 4])
+        >>> ser2 = pd.Series([10, 11, 13, 16])
+        >>> ser1.ewm(alpha=.2).cov(ser2)
+        0         NaN
+        1    0.500000
+        2    1.524590
+        3    3.408836
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) sample covariance",
+        agg_method="cov",
+    )
+    def cov(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        bias: bool = False,
+        numeric_only: bool = False,
+    ):
+        if self.times is not None:
+            raise NotImplementedError("cov is not implemented with times")
+
+        from pandas import Series
+
+        self._validate_numeric_only("cov", numeric_only)
+
+        def cov_func(x, y):
+            x_array = self._prep_values(x)
+            y_array = self._prep_values(y)
+            window_indexer = self._get_window_indexer()
+            min_periods = (
+                self.min_periods
+                if self.min_periods is not None
+                else window_indexer.window_size
+            )
+            start, end = window_indexer.get_window_bounds(
+                num_values=len(x_array),
+                min_periods=min_periods,
+                center=self.center,
+                closed=self.closed,
+                step=self.step,
+            )
+            result = window_aggregations.ewmcov(
+                x_array,
+                start,
+                end,
+                # error: Argument 4 to "ewmcov" has incompatible type
+                # "Optional[int]"; expected "int"
+                self.min_periods,  # type: ignore[arg-type]
+                y_array,
+                self._com,
+                self.adjust,
+                self.ignore_na,
+                bias,
+            )
+            return Series(result, index=x.index, name=x.name, copy=False)
+
+        return self._apply_pairwise(
+            self._selected_obj, other, pairwise, cov_func, numeric_only
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """\
+        other : Series or DataFrame, optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndex DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        """
+        ),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser1 = pd.Series([1, 2, 3, 4])
+        >>> ser2 = pd.Series([10, 11, 13, 16])
+        >>> ser1.ewm(alpha=.2).corr(ser2)
+        0         NaN
+        1    1.000000
+        2    0.982821
+        3    0.977802
+        dtype: float64
+        """
+        ),
+        window_method="ewm",
+        aggregation_description="(exponential weighted moment) sample correlation",
+        agg_method="corr",
+    )
+    def corr(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        numeric_only: bool = False,
+    ):
+        if self.times is not None:
+            raise NotImplementedError("corr is not implemented with times")
+
+        from pandas import Series
+
+        self._validate_numeric_only("corr", numeric_only)
+
+        def cov_func(x, y):
+            x_array = self._prep_values(x)
+            y_array = self._prep_values(y)
+            window_indexer = self._get_window_indexer()
+            min_periods = (
+                self.min_periods
+                if self.min_periods is not None
+                else window_indexer.window_size
+            )
+            start, end = window_indexer.get_window_bounds(
+                num_values=len(x_array),
+                min_periods=min_periods,
+                center=self.center,
+                closed=self.closed,
+                step=self.step,
+            )
+
+            def _cov(X, Y):
+                return window_aggregations.ewmcov(
+                    X,
+                    start,
+                    end,
+                    min_periods,
+                    Y,
+                    self._com,
+                    self.adjust,
+                    self.ignore_na,
+                    True,
+                )
+
+            with np.errstate(all="ignore"):
+                cov = _cov(x_array, y_array)
+                x_var = _cov(x_array, x_array)
+                y_var = _cov(y_array, y_array)
+                result = cov / zsqrt(x_var * y_var)
+            return Series(result, index=x.index, name=x.name, copy=False)
+
+        return self._apply_pairwise(
+            self._selected_obj, other, pairwise, cov_func, numeric_only
+        )
+
+
+class ExponentialMovingWindowGroupby(BaseWindowGroupby, ExponentialMovingWindow):
+    """
+    Provide an exponential moving window groupby implementation.
+    """
+
+    _attributes = ExponentialMovingWindow._attributes + BaseWindowGroupby._attributes
+
+    def __init__(self, obj, *args, _grouper=None, **kwargs) -> None:
+        super().__init__(obj, *args, _grouper=_grouper, **kwargs)
+
+        if not obj.empty and self.times is not None:
+            # sort the times and recalculate the deltas according to the groups
+            groupby_order = np.concatenate(list(self._grouper.indices.values()))
+            self._deltas = _calculate_deltas(
+                self.times.take(groupby_order),
+                self.halflife,
+            )
+
+    def _get_window_indexer(self) -> GroupbyIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+
+        Returns
+        -------
+        GroupbyIndexer
+        """
+        window_indexer = GroupbyIndexer(
+            groupby_indices=self._grouper.indices,
+            window_indexer=ExponentialMovingWindowIndexer,
+        )
+        return window_indexer
+
+
+class OnlineExponentialMovingWindow(ExponentialMovingWindow):
+    def __init__(
+        self,
+        obj: NDFrame,
+        com: float | None = None,
+        span: float | None = None,
+        halflife: float | TimedeltaConvertibleTypes | None = None,
+        alpha: float | None = None,
+        min_periods: int | None = 0,
+        adjust: bool = True,
+        ignore_na: bool = False,
+        times: np.ndarray | NDFrame | None = None,
+        engine: str = "numba",
+        engine_kwargs: dict[str, bool] | None = None,
+        *,
+        selection=None,
+    ) -> None:
+        if times is not None:
+            raise NotImplementedError(
+                "times is not implemented with online operations."
+            )
+        super().__init__(
+            obj=obj,
+            com=com,
+            span=span,
+            halflife=halflife,
+            alpha=alpha,
+            min_periods=min_periods,
+            adjust=adjust,
+            ignore_na=ignore_na,
+            times=times,
+            selection=selection,
+        )
+        self._mean = EWMMeanState(self._com, self.adjust, self.ignore_na, obj.shape)
+        if maybe_use_numba(engine):
+            self.engine = engine
+            self.engine_kwargs = engine_kwargs
+        else:
+            raise ValueError("'numba' is the only supported engine")
+
+    def reset(self) -> None:
+        """
+        Reset the state captured by `update` calls.
+        """
+        self._mean.reset()
+
+    def aggregate(self, func=None, *args, **kwargs):
+        raise NotImplementedError("aggregate is not implemented.")
+
+    def std(self, bias: bool = False, *args, **kwargs):
+        raise NotImplementedError("std is not implemented.")
+
+    def corr(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        numeric_only: bool = False,
+    ):
+        raise NotImplementedError("corr is not implemented.")
+
+    def cov(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        bias: bool = False,
+        numeric_only: bool = False,
+    ):
+        raise NotImplementedError("cov is not implemented.")
+
+    def var(self, bias: bool = False, numeric_only: bool = False):
+        raise NotImplementedError("var is not implemented.")
+
+    def mean(self, *args, update=None, update_times=None, **kwargs):
+        """
+        Calculate an online exponentially weighted mean.
+
+        Parameters
+        ----------
+        update: DataFrame or Series, default None
+            New values to continue calculating the
+            exponentially weighted mean from the last values and weights.
+            Values should be float64 dtype.
+
+            ``update`` needs to be ``None`` the first time the
+            exponentially weighted mean is calculated.
+
+        update_times: Series or 1-D np.ndarray, default None
+            New times to continue calculating the
+            exponentially weighted mean from the last values and weights.
+            If ``None``, values are assumed to be evenly spaced
+            in time.
+            This feature is currently unsupported.
+
+        Returns
+        -------
+        DataFrame or Series
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"a": range(5), "b": range(5, 10)})
+        >>> online_ewm = df.head(2).ewm(0.5).online()
+        >>> online_ewm.mean()
+              a     b
+        0  0.00  5.00
+        1  0.75  5.75
+        >>> online_ewm.mean(update=df.tail(3))
+                  a         b
+        2  1.615385  6.615385
+        3  2.550000  7.550000
+        4  3.520661  8.520661
+        >>> online_ewm.reset()
+        >>> online_ewm.mean()
+              a     b
+        0  0.00  5.00
+        1  0.75  5.75
+        """
+        result_kwargs = {}
+        is_frame = self._selected_obj.ndim == 2
+        if update_times is not None:
+            raise NotImplementedError("update_times is not implemented.")
+        update_deltas = np.ones(
+            max(self._selected_obj.shape[-1] - 1, 0), dtype=np.float64
+        )
+        if update is not None:
+            if self._mean.last_ewm is None:
+                raise ValueError(
+                    "Must call mean with update=None first before passing update"
+                )
+            result_from = 1
+            result_kwargs["index"] = update.index
+            if is_frame:
+                last_value = self._mean.last_ewm[np.newaxis, :]
+                result_kwargs["columns"] = update.columns
+            else:
+                last_value = self._mean.last_ewm
+                result_kwargs["name"] = update.name
+            np_array = np.concatenate((last_value, update.to_numpy()))
+        else:
+            result_from = 0
+            result_kwargs["index"] = self._selected_obj.index
+            if is_frame:
+                result_kwargs["columns"] = self._selected_obj.columns
+            else:
+                result_kwargs["name"] = self._selected_obj.name
+            np_array = self._selected_obj.astype(np.float64).to_numpy()
+        ewma_func = generate_online_numba_ewma_func(
+            **get_jit_arguments(self.engine_kwargs)
+        )
+        result = self._mean.run_ewm(
+            np_array if is_frame else np_array[:, np.newaxis],
+            update_deltas,
+            self.min_periods,
+            ewma_func,
+        )
+        if not is_frame:
+            result = result.squeeze()
+        result = result[result_from:]
+        result = self._selected_obj._constructor(result, **result_kwargs)
+        return result
diff --git a/pandas/pandas/core/window/expanding.py b/pandas/pandas/core/window/expanding.py
new file mode 100644
index 0000000000000..81c89e1ef5428
--- /dev/null
+++ b/pandas/pandas/core/window/expanding.py
@@ -0,0 +1,1092 @@
+from __future__ import annotations
+
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    final,
+    overload,
+)
+
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    doc,
+)
+
+from pandas.core.indexers.objects import (
+    BaseIndexer,
+    ExpandingIndexer,
+    GroupbyIndexer,
+)
+from pandas.core.window.doc import (
+    _shared_docs,
+    create_section_header,
+    kwargs_numeric_only,
+    numba_notes,
+    template_header,
+    template_pipe,
+    template_returns,
+    template_see_also,
+    window_agg_numba_parameters,
+    window_apply_parameters,
+)
+from pandas.core.window.rolling import (
+    BaseWindowGroupby,
+    RollingAndExpandingMixin,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import (
+        Concatenate,
+        P,
+        QuantileInterpolation,
+        Self,
+        T,
+        WindowingRankType,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+    from pandas.core.generic import NDFrame
+
+
+class Expanding(RollingAndExpandingMixin):
+    """
+    Provide expanding window calculations.
+
+    An expanding window yields the value of an aggregation statistic with all the data
+    available up to that point in time.
+
+    Parameters
+    ----------
+    min_periods : int, default 1
+        Minimum number of observations in window required to have a value;
+        otherwise, result is ``np.nan``.
+
+    method : str {'single', 'table'}, default 'single'
+        Execute the rolling operation per single column or row (``'single'``)
+        or over the entire object (``'table'``).
+
+        This argument is only implemented when specifying ``engine='numba'``
+        in the method call.
+
+        .. versionadded:: 1.3.0
+
+    Returns
+    -------
+    pandas.api.typing.Expanding
+        An instance of Expanding for further expanding window calculations,
+        e.g. using the ``sum`` method.
+
+    See Also
+    --------
+    rolling : Provides rolling window calculations.
+    ewm : Provides exponential weighted functions.
+
+    Notes
+    -----
+    See :ref:`Windowing Operations <window.expanding>` for further usage details
+    and examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    >>> df
+         B
+    0  0.0
+    1  1.0
+    2  2.0
+    3  NaN
+    4  4.0
+
+    **min_periods**
+
+    Expanding sum with 1 vs 3 observations needed to calculate a value.
+
+    >>> df.expanding(1).sum()
+         B
+    0  0.0
+    1  1.0
+    2  3.0
+    3  3.0
+    4  7.0
+    >>> df.expanding(3).sum()
+         B
+    0  NaN
+    1  NaN
+    2  3.0
+    3  3.0
+    4  7.0
+    """
+
+    _attributes: list[str] = ["min_periods", "method"]
+
+    def __init__(
+        self,
+        obj: NDFrame,
+        min_periods: int = 1,
+        method: str = "single",
+        selection=None,
+    ) -> None:
+        super().__init__(
+            obj=obj,
+            min_periods=min_periods,
+            method=method,
+            selection=selection,
+        )
+
+    def _get_window_indexer(self) -> BaseIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+        """
+        return ExpandingIndexer()
+
+    @doc(
+        _shared_docs["aggregate"],
+        see_also=dedent(
+            """
+        See Also
+        --------
+        DataFrame.aggregate : Similar DataFrame method.
+        Series.aggregate : Similar Series method.
+        """
+        ),
+        examples=dedent(
+            """
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        >>> df
+           A  B  C
+        0  1  4  7
+        1  2  5  8
+        2  3  6  9
+
+        >>> df.ewm(alpha=0.5).mean()
+                  A         B         C
+        0  1.000000  4.000000  7.000000
+        1  1.666667  4.666667  7.666667
+        2  2.428571  5.428571  8.428571
+        """
+        ),
+        klass="Series/Dataframe",
+        axis="",
+    )
+    def aggregate(self, func=None, *args, **kwargs):
+        return super().aggregate(func, *args, **kwargs)
+
+    agg = aggregate
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().count()
+        a    1.0
+        b    2.0
+        c    3.0
+        d    4.0
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="count of non NaN observations",
+        agg_method="count",
+    )
+    def count(self, numeric_only: bool = False):
+        return super().count(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        window_apply_parameters,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().apply(lambda s: s.max() - 2 * s.min())
+        a   -1.0
+        b    0.0
+        c    1.0
+        d    2.0
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="custom aggregation function",
+        agg_method="apply",
+    )
+    def apply(
+        self,
+        func: Callable[..., Any],
+        raw: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        args: tuple[Any, ...] | None = None,
+        kwargs: dict[str, Any] | None = None,
+    ):
+        return super().apply(
+            func,
+            raw=raw,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    @final
+    @Substitution(
+        klass="Expanding",
+        examples="""
+    >>> df = pd.DataFrame({'A': [1, 2, 3, 4]},
+    ...                   index=pd.date_range('2012-08-02', periods=4))
+    >>> df
+                A
+    2012-08-02  1
+    2012-08-03  2
+    2012-08-04  3
+    2012-08-05  4
+
+    To get the difference between each expanding window's maximum and minimum
+    value in one pass, you can do
+
+    >>> df.expanding().pipe(lambda x: x.max() - x.min())
+                  A
+    2012-08-02  0.0
+    2012-08-03  1.0
+    2012-08-04  2.0
+    2012-08-05  3.0""",
+    )
+    @Appender(template_pipe)
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        return super().pipe(func, *args, **kwargs)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().sum()
+        a     1.0
+        b     3.0
+        c     6.0
+        d    10.0
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="sum",
+        agg_method="sum",
+    )
+    def sum(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().sum(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([3, 2, 1, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().max()
+        a    3.0
+        b    3.0
+        c    3.0
+        d    4.0
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="maximum",
+        agg_method="max",
+    )
+    def max(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().max(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([2, 3, 4, 1], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().min()
+        a    2.0
+        b    2.0
+        c    2.0
+        d    1.0
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="minimum",
+        agg_method="min",
+    )
+    def min(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().min(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().mean()
+        a    1.0
+        b    1.5
+        c    2.0
+        d    2.5
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="mean",
+        agg_method="mean",
+    )
+    def mean(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().mean(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser.expanding().median()
+        a    1.0
+        b    1.5
+        c    2.0
+        d    2.5
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="median",
+        agg_method="median",
+    )
+    def median(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().median(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.\n
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        window_agg_numba_parameters("1.4"),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "numpy.std : Equivalent method for NumPy array.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        The default ``ddof`` of 1 used in :meth:`Series.std` is different
+        than the default ``ddof`` of 0 in :func:`numpy.std`.
+
+        A minimum of one period is required for the rolling calculation.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([5, 5, 6, 7, 5, 5, 5])
+
+        >>> s.expanding(3).std()
+        0         NaN
+        1         NaN
+        2    0.577350
+        3    0.957427
+        4    0.894427
+        5    0.836660
+        6    0.786796
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="standard deviation",
+        agg_method="std",
+    )
+    def std(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().std(
+            ddof=ddof,
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.\n
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        window_agg_numba_parameters("1.4"),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "numpy.var : Equivalent method for NumPy array.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        The default ``ddof`` of 1 used in :meth:`Series.var` is different
+        than the default ``ddof`` of 0 in :func:`numpy.var`.
+
+        A minimum of one period is required for the rolling calculation.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([5, 5, 6, 7, 5, 5, 5])
+
+        >>> s.expanding(3).var()
+        0         NaN
+        1         NaN
+        2    0.333333
+        3    0.916667
+        4    0.800000
+        5    0.700000
+        6    0.619048
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="variance",
+        agg_method="var",
+    )
+    def var(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().var(
+            ddof=ddof,
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.\n
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        "A minimum of one period is required for the calculation.\n\n",
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([0, 1, 2, 3])
+
+        >>> s.expanding().sem()
+        0         NaN
+        1    0.707107
+        2    0.707107
+        3    0.745356
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="standard error of mean",
+        agg_method="sem",
+    )
+    def sem(self, ddof: int = 1, numeric_only: bool = False):
+        return super().sem(ddof=ddof, numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "scipy.stats.skew : Third moment of a probability density.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        "A minimum of three periods is required for the rolling calculation.\n\n",
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([-1, 0, 2, -1, 2], index=['a', 'b', 'c', 'd', 'e'])
+        >>> ser.expanding().skew()
+        a         NaN
+        b         NaN
+        c    0.935220
+        d    1.414214
+        e    0.315356
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="unbiased skewness",
+        agg_method="skew",
+    )
+    def skew(self, numeric_only: bool = False):
+        return super().skew(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "scipy.stats.kurtosis : Reference SciPy method.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        "A minimum of four periods is required for the calculation.\n\n",
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show a rolling calculation with a window size of
+        four matching the equivalent function call using `scipy.stats`.
+
+        >>> arr = [1, 2, 3, 4, 999]
+        >>> import scipy.stats
+        >>> print(f"{{scipy.stats.kurtosis(arr[:-1], bias=False):.6f}}")
+        -1.200000
+        >>> print(f"{{scipy.stats.kurtosis(arr, bias=False):.6f}}")
+        4.999874
+        >>> s = pd.Series(arr)
+        >>> s.expanding(4).kurt()
+        0         NaN
+        1         NaN
+        2         NaN
+        3   -1.200000
+        4    4.999874
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="Fisher's definition of kurtosis without bias",
+        agg_method="kurt",
+    )
+    def kurt(self, numeric_only: bool = False):
+        return super().kurt(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        GroupBy.first : Similar method for GroupBy objects.
+        Expanding.last : Method to get the last element in each window.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show an expanding calculation with a window size of
+        three.
+
+        >>> s = pd.Series(range(5))
+        >>> s.expanding(3).first()
+        0         NaN
+        1         NaN
+        2         0.0
+        3         0.0
+        4         0.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="First (left-most) element of the window",
+        agg_method="first",
+    )
+    def first(self, numeric_only: bool = False):
+        return super().first(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        GroupBy.last : Similar method for GroupBy objects.
+        Expanding.first : Method to get the first element in each window.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show an expanding calculation with a window size of
+        three.
+
+        >>> s = pd.Series(range(5))
+        >>> s.expanding(3).last()
+        0         NaN
+        1         NaN
+        2         2.0
+        3         3.0
+        4         4.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="Last (right-most) element of the window",
+        agg_method="last",
+    )
+    def last(self, numeric_only: bool = False):
+        return super().last(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        q : float
+            Quantile to compute. 0 <= quantile <= 1.
+
+            .. deprecated:: 2.1.0
+                This was renamed from 'quantile' to 'q' in version 2.1.0.
+        interpolation : {{'linear', 'lower', 'higher', 'midpoint', 'nearest'}}
+            This optional parameter specifies the interpolation method to use,
+            when the desired quantile lies between two data points `i` and `j`:
+
+                * linear: `i + (j - i) * fraction`, where `fraction` is the
+                  fractional part of the index surrounded by `i` and `j`.
+                * lower: `i`.
+                * higher: `j`.
+                * nearest: `i` or `j` whichever is nearest.
+                * midpoint: (`i` + `j`) / 2.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4, 5, 6], index=['a', 'b', 'c', 'd', 'e', 'f'])
+        >>> ser.expanding(min_periods=4).quantile(.25)
+        a     NaN
+        b     NaN
+        c     NaN
+        d    1.75
+        e    2.00
+        f    2.25
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="quantile",
+        agg_method="quantile",
+    )
+    def quantile(
+        self,
+        q: float,
+        interpolation: QuantileInterpolation = "linear",
+        numeric_only: bool = False,
+    ):
+        return super().quantile(
+            q=q,
+            interpolation=interpolation,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        ".. versionadded:: 1.4.0 \n\n",
+        create_section_header("Parameters"),
+        dedent(
+            """
+        method : {{'average', 'min', 'max'}}, default 'average'
+            How to rank the group of records that have the same value (i.e. ties):
+
+            * average: average rank of the group
+            * min: lowest rank in the group
+            * max: highest rank in the group
+
+        ascending : bool, default True
+            Whether or not the elements should be ranked in ascending order.
+        pct : bool, default False
+            Whether or not to display the returned rankings in percentile
+            form.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([1, 4, 2, 3, 5, 3])
+        >>> s.expanding().rank()
+        0    1.0
+        1    2.0
+        2    2.0
+        3    3.0
+        4    5.0
+        5    3.5
+        dtype: float64
+
+        >>> s.expanding().rank(method="max")
+        0    1.0
+        1    2.0
+        2    2.0
+        3    3.0
+        4    5.0
+        5    4.0
+        dtype: float64
+
+        >>> s.expanding().rank(method="min")
+        0    1.0
+        1    2.0
+        2    2.0
+        3    3.0
+        4    5.0
+        5    3.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="expanding",
+        aggregation_description="rank",
+        agg_method="rank",
+    )
+    def rank(
+        self,
+        method: WindowingRankType = "average",
+        ascending: bool = True,
+        pct: bool = False,
+        numeric_only: bool = False,
+    ):
+        return super().rank(
+            method=method,
+            ascending=ascending,
+            pct=pct,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        other : Series or DataFrame, optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndexed DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser1 = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser2 = pd.Series([10, 11, 13, 16], index=['a', 'b', 'c', 'd'])
+        >>> ser1.expanding().cov(ser2)
+        a         NaN
+        b    0.500000
+        c    1.500000
+        d    3.333333
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="sample covariance",
+        agg_method="cov",
+    )
+    def cov(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        return super().cov(
+            other=other,
+            pairwise=pairwise,
+            ddof=ddof,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        other : Series or DataFrame, optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndexed DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.\n
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        cov : Similar method to calculate covariance.
+        numpy.corrcoef : NumPy Pearson's correlation calculation.
+        """
+        ).replace("\n", "", 1),
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        This function uses Pearson's definition of correlation
+        (https://en.wikipedia.org/wiki/Pearson_correlation_coefficient).
+
+        When `other` is not specified, the output will be self correlation (e.g.
+        all 1's), except for :class:`~pandas.DataFrame` inputs with `pairwise`
+        set to `True`.
+
+        Function will return ``NaN`` for correlations of equal valued sequences;
+        this is the result of a 0/0 division error.
+
+        When `pairwise` is set to `False`, only matching columns between `self` and
+        `other` will be used.
+
+        When `pairwise` is set to `True`, the output will be a MultiIndex DataFrame
+        with the original index on the first level, and the `other` DataFrame
+        columns on the second level.
+
+        In the case of missing elements, only complete pairwise observations
+        will be used.\n
+        """
+        ),
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser1 = pd.Series([1, 2, 3, 4], index=['a', 'b', 'c', 'd'])
+        >>> ser2 = pd.Series([10, 11, 13, 16], index=['a', 'b', 'c', 'd'])
+        >>> ser1.expanding().corr(ser2)
+        a         NaN
+        b    1.000000
+        c    0.981981
+        d    0.975900
+        dtype: float64
+        """
+        ),
+        window_method="expanding",
+        aggregation_description="correlation",
+        agg_method="corr",
+    )
+    def corr(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        return super().corr(
+            other=other,
+            pairwise=pairwise,
+            ddof=ddof,
+            numeric_only=numeric_only,
+        )
+
+
+class ExpandingGroupby(BaseWindowGroupby, Expanding):
+    """
+    Provide a expanding groupby implementation.
+    """
+
+    _attributes = Expanding._attributes + BaseWindowGroupby._attributes
+
+    def _get_window_indexer(self) -> GroupbyIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+
+        Returns
+        -------
+        GroupbyIndexer
+        """
+        window_indexer = GroupbyIndexer(
+            groupby_indices=self._grouper.indices,
+            window_indexer=ExpandingIndexer,
+        )
+        return window_indexer
diff --git a/pandas/pandas/core/window/numba_.py b/pandas/pandas/core/window/numba_.py
new file mode 100644
index 0000000000000..171d3bc1d1c35
--- /dev/null
+++ b/pandas/pandas/core/window/numba_.py
@@ -0,0 +1,357 @@
+from __future__ import annotations
+
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.util.numba_ import jit_user_function
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pandas._typing import Scalar
+
+
+@functools.cache
+def generate_numba_apply_func(
+    func: Callable[..., Scalar],
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+):
+    """
+    Generate a numba jitted apply function specified by values from engine_kwargs.
+
+    1. jit the user's function
+    2. Return a rolling apply function with the jitted function inline
+
+    Configurations specified in engine_kwargs apply to both the user's
+    function _AND_ the rolling apply function.
+
+    Parameters
+    ----------
+    func : function
+        function to be applied to each window and will be JITed
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+    numba_func = jit_user_function(func)
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def roll_apply(
+        values: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        minimum_periods: int,
+        *args: Any,
+    ) -> np.ndarray:
+        result = np.empty(len(begin))
+        for i in numba.prange(len(result)):
+            start = begin[i]
+            stop = end[i]
+            window = values[start:stop]
+            count_nan = np.sum(np.isnan(window))
+            if len(window) - count_nan >= minimum_periods:
+                result[i] = numba_func(window, *args)
+            else:
+                result[i] = np.nan
+        return result
+
+    return roll_apply
+
+
+@functools.cache
+def generate_numba_ewm_func(
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+    com: float,
+    adjust: bool,
+    ignore_na: bool,
+    deltas: tuple,
+    normalize: bool,
+):
+    """
+    Generate a numba jitted ewm mean or sum function specified by values
+    from engine_kwargs.
+
+    Parameters
+    ----------
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+    com : float
+    adjust : bool
+    ignore_na : bool
+    deltas : tuple
+    normalize : bool
+
+    Returns
+    -------
+    Numba function
+    """
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def ewm(
+        values: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        minimum_periods: int,
+    ) -> np.ndarray:
+        result = np.empty(len(values))
+        alpha = 1.0 / (1.0 + com)
+        old_wt_factor = 1.0 - alpha
+        new_wt = 1.0 if adjust else alpha
+
+        for i in numba.prange(len(begin)):
+            start = begin[i]
+            stop = end[i]
+            window = values[start:stop]
+            sub_result = np.empty(len(window))
+
+            weighted = window[0]
+            nobs = int(not np.isnan(weighted))
+            sub_result[0] = weighted if nobs >= minimum_periods else np.nan
+            old_wt = 1.0
+
+            for j in range(1, len(window)):
+                cur = window[j]
+                is_observation = not np.isnan(cur)
+                nobs += is_observation
+                if not np.isnan(weighted):
+                    if is_observation or not ignore_na:
+                        if normalize:
+                            # note that len(deltas) = len(vals) - 1 and deltas[i]
+                            # is to be used in conjunction with vals[i+1]
+                            old_wt *= old_wt_factor ** deltas[start + j - 1]
+                            if not adjust and com == 1:
+                                # update in case of irregular-interval time series
+                                new_wt = 1.0 - old_wt
+                        else:
+                            weighted = old_wt_factor * weighted
+                        if is_observation:
+                            if normalize:
+                                # avoid numerical errors on constant series
+                                if weighted != cur:
+                                    weighted = old_wt * weighted + new_wt * cur
+                                    if normalize:
+                                        weighted = weighted / (old_wt + new_wt)
+                                if adjust:
+                                    old_wt += new_wt
+                                else:
+                                    old_wt = 1.0
+                            else:
+                                weighted += cur
+                elif is_observation:
+                    weighted = cur
+
+                sub_result[j] = weighted if nobs >= minimum_periods else np.nan
+
+            result[start:stop] = sub_result
+
+        return result
+
+    return ewm
+
+
+@functools.cache
+def generate_numba_table_func(
+    func: Callable[..., np.ndarray],
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+):
+    """
+    Generate a numba jitted function to apply window calculations table-wise.
+
+    Func will be passed a M window size x N number of columns array, and
+    must return a 1 x N number of columns array.
+
+    1. jit the user's function
+    2. Return a rolling apply function with the jitted function inline
+
+    Parameters
+    ----------
+    func : function
+        function to be applied to each window and will be JITed
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+    numba_func = jit_user_function(func)
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def roll_table(
+        values: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        minimum_periods: int,
+        *args: Any,
+    ):
+        result = np.empty((len(begin), values.shape[1]))
+        min_periods_mask = np.empty(result.shape)
+        for i in numba.prange(len(result)):
+            start = begin[i]
+            stop = end[i]
+            window = values[start:stop]
+            count_nan = np.sum(np.isnan(window), axis=0)
+            nan_mask = len(window) - count_nan >= minimum_periods
+            if nan_mask.any():
+                result[i, :] = numba_func(window, *args)
+            min_periods_mask[i, :] = nan_mask
+        result = np.where(min_periods_mask, result, np.nan)
+        return result
+
+    return roll_table
+
+
+# This function will no longer be needed once numba supports
+# axis for all np.nan* agg functions
+# https://github.com/numba/numba/issues/1269
+@functools.cache
+def generate_manual_numpy_nan_agg_with_axis(nan_func):
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=True, nogil=True, parallel=True)
+    def nan_agg_with_axis(table):
+        result = np.empty(table.shape[1])
+        for i in numba.prange(table.shape[1]):
+            partition = table[:, i]
+            result[i] = nan_func(partition)
+        return result
+
+    return nan_agg_with_axis
+
+
+@functools.cache
+def generate_numba_ewm_table_func(
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+    com: float,
+    adjust: bool,
+    ignore_na: bool,
+    deltas: tuple,
+    normalize: bool,
+):
+    """
+    Generate a numba jitted ewm mean or sum function applied table wise specified
+    by values from engine_kwargs.
+
+    Parameters
+    ----------
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+    com : float
+    adjust : bool
+    ignore_na : bool
+    deltas : tuple
+    normalize: bool
+
+    Returns
+    -------
+    Numba function
+    """
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def ewm_table(
+        values: np.ndarray,
+        begin: np.ndarray,
+        end: np.ndarray,
+        minimum_periods: int,
+    ) -> np.ndarray:
+        alpha = 1.0 / (1.0 + com)
+        old_wt_factor = 1.0 - alpha
+        new_wt = 1.0 if adjust else alpha
+        old_wt = np.ones(values.shape[1])
+
+        result = np.empty(values.shape)
+        weighted = values[0].copy()
+        nobs = (~np.isnan(weighted)).astype(np.int64)
+        result[0] = np.where(nobs >= minimum_periods, weighted, np.nan)
+        for i in range(1, len(values)):
+            cur = values[i]
+            is_observations = ~np.isnan(cur)
+            nobs += is_observations.astype(np.int64)
+            for j in numba.prange(len(cur)):
+                if not np.isnan(weighted[j]):
+                    if is_observations[j] or not ignore_na:
+                        if normalize:
+                            # note that len(deltas) = len(vals) - 1 and deltas[i]
+                            # is to be used in conjunction with vals[i+1]
+                            old_wt[j] *= old_wt_factor ** deltas[i - 1]
+                            if not adjust and com == 1:
+                                # update in case of irregular-interval time series
+                                new_wt = 1.0 - old_wt[j]
+                        else:
+                            weighted[j] = old_wt_factor * weighted[j]
+                        if is_observations[j]:
+                            if normalize:
+                                # avoid numerical errors on constant series
+                                if weighted[j] != cur[j]:
+                                    weighted[j] = (
+                                        old_wt[j] * weighted[j] + new_wt * cur[j]
+                                    )
+                                    if normalize:
+                                        weighted[j] = weighted[j] / (old_wt[j] + new_wt)
+                                if adjust:
+                                    old_wt[j] += new_wt
+                                else:
+                                    old_wt[j] = 1.0
+                            else:
+                                weighted[j] += cur[j]
+                elif is_observations[j]:
+                    weighted[j] = cur[j]
+
+            result[i] = np.where(nobs >= minimum_periods, weighted, np.nan)
+
+        return result
+
+    return ewm_table
diff --git a/pandas/pandas/core/window/online.py b/pandas/pandas/core/window/online.py
new file mode 100644
index 0000000000000..72236bf5ccea2
--- /dev/null
+++ b/pandas/pandas/core/window/online.py
@@ -0,0 +1,117 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+
+
+def generate_online_numba_ewma_func(
+    nopython: bool,
+    nogil: bool,
+    parallel: bool,
+):
+    """
+    Generate a numba jitted groupby ewma function specified by values
+    from engine_kwargs.
+
+    Parameters
+    ----------
+    nopython : bool
+        nopython to be passed into numba.jit
+    nogil : bool
+        nogil to be passed into numba.jit
+    parallel : bool
+        parallel to be passed into numba.jit
+
+    Returns
+    -------
+    Numba function
+    """
+    if TYPE_CHECKING:
+        import numba
+    else:
+        numba = import_optional_dependency("numba")
+
+    @numba.jit(nopython=nopython, nogil=nogil, parallel=parallel)
+    def online_ewma(
+        values: np.ndarray,
+        deltas: np.ndarray,
+        minimum_periods: int,
+        old_wt_factor: float,
+        new_wt: float,
+        old_wt: np.ndarray,
+        adjust: bool,
+        ignore_na: bool,
+    ):
+        """
+        Compute online exponentially weighted mean per column over 2D values.
+
+        Takes the first observation as is, then computes the subsequent
+        exponentially weighted mean accounting minimum periods.
+        """
+        result = np.empty(values.shape)
+        weighted_avg = values[0].copy()
+        nobs = (~np.isnan(weighted_avg)).astype(np.int64)
+        result[0] = np.where(nobs >= minimum_periods, weighted_avg, np.nan)
+
+        for i in range(1, len(values)):
+            cur = values[i]
+            is_observations = ~np.isnan(cur)
+            nobs += is_observations.astype(np.int64)
+            for j in numba.prange(len(cur)):
+                if not np.isnan(weighted_avg[j]):
+                    if is_observations[j] or not ignore_na:
+                        # note that len(deltas) = len(vals) - 1 and deltas[i] is to be
+                        # used in conjunction with vals[i+1]
+                        old_wt[j] *= old_wt_factor ** deltas[j - 1]
+                        if is_observations[j]:
+                            # avoid numerical errors on constant series
+                            if weighted_avg[j] != cur[j]:
+                                weighted_avg[j] = (
+                                    (old_wt[j] * weighted_avg[j]) + (new_wt * cur[j])
+                                ) / (old_wt[j] + new_wt)
+                            if adjust:
+                                old_wt[j] += new_wt
+                            else:
+                                old_wt[j] = 1.0
+                elif is_observations[j]:
+                    weighted_avg[j] = cur[j]
+
+            result[i] = np.where(nobs >= minimum_periods, weighted_avg, np.nan)
+
+        return result, old_wt
+
+    return online_ewma
+
+
+class EWMMeanState:
+    def __init__(self, com, adjust, ignore_na, shape) -> None:
+        alpha = 1.0 / (1.0 + com)
+        self.shape = shape
+        self.adjust = adjust
+        self.ignore_na = ignore_na
+        self.new_wt = 1.0 if adjust else alpha
+        self.old_wt_factor = 1.0 - alpha
+        self.old_wt = np.ones(self.shape[-1])
+        self.last_ewm = None
+
+    def run_ewm(self, weighted_avg, deltas, min_periods, ewm_func):
+        result, old_wt = ewm_func(
+            weighted_avg,
+            deltas,
+            min_periods,
+            self.old_wt_factor,
+            self.new_wt,
+            self.old_wt,
+            self.adjust,
+            self.ignore_na,
+        )
+        self.old_wt = old_wt
+        self.last_ewm = result[-1]
+        return result
+
+    def reset(self) -> None:
+        self.old_wt = np.ones(self.shape[-1])
+        self.last_ewm = None
diff --git a/pandas/pandas/core/window/rolling.py b/pandas/pandas/core/window/rolling.py
new file mode 100644
index 0000000000000..69fce8cf2137e
--- /dev/null
+++ b/pandas/pandas/core/window/rolling.py
@@ -0,0 +1,3109 @@
+"""
+Provide a generic structure to support window functions,
+similar to how we have a Groupby object.
+"""
+
+from __future__ import annotations
+
+import copy
+from datetime import timedelta
+from functools import partial
+import inspect
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    final,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.tslibs import (
+    BaseOffset,
+    Timedelta,
+    to_offset,
+)
+import pandas._libs.window.aggregations as window_aggregations
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import DataError
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+    doc,
+)
+
+from pandas.core.dtypes.common import (
+    ensure_float64,
+    is_bool,
+    is_integer,
+    is_numeric_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import ArrowDtype
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import notna
+
+from pandas.core._numba import executor
+from pandas.core.algorithms import factorize
+from pandas.core.apply import (
+    ResamplerWindowApply,
+    reconstruct_func,
+)
+from pandas.core.arrays import ExtensionArray
+from pandas.core.base import SelectionMixin
+import pandas.core.common as com
+from pandas.core.indexers.objects import (
+    BaseIndexer,
+    FixedWindowIndexer,
+    GroupbyIndexer,
+    VariableWindowIndexer,
+)
+from pandas.core.indexes.api import (
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    TimedeltaIndex,
+)
+from pandas.core.reshape.concat import concat
+from pandas.core.util.numba_ import (
+    get_jit_arguments,
+    maybe_use_numba,
+    prepare_function_arguments,
+)
+from pandas.core.window.common import (
+    flex_binary_moment,
+    zsqrt,
+)
+from pandas.core.window.doc import (
+    _shared_docs,
+    create_section_header,
+    kwargs_numeric_only,
+    kwargs_scipy,
+    numba_notes,
+    template_header,
+    template_pipe,
+    template_returns,
+    template_see_also,
+    window_agg_numba_parameters,
+    window_apply_parameters,
+)
+from pandas.core.window.numba_ import (
+    generate_manual_numpy_nan_agg_with_axis,
+    generate_numba_apply_func,
+    generate_numba_table_func,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+    from collections.abc import (
+        Hashable,
+        Iterator,
+        Sized,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        Concatenate,
+        NDFrameT,
+        QuantileInterpolation,
+        P,
+        Self,
+        T,
+        WindowingRankType,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+    from pandas.core.generic import NDFrame
+    from pandas.core.groupby.ops import BaseGrouper
+
+from pandas.core.arrays.datetimelike import dtype_to_unit
+
+
+class BaseWindow(SelectionMixin):
+    """Provides utilities for performing windowing operations."""
+
+    _attributes: list[str] = []
+    exclusions: frozenset[Hashable] = frozenset()
+    _on: Index
+
+    def __init__(
+        self,
+        obj: NDFrame,
+        window=None,
+        min_periods: int | None = None,
+        center: bool | None = False,
+        win_type: str | None = None,
+        on: str | Index | None = None,
+        closed: str | None = None,
+        step: int | None = None,
+        method: str = "single",
+        *,
+        selection=None,
+    ) -> None:
+        self.obj = obj
+        self.on = on
+        self.closed = closed
+        self.step = step
+        self.window = window
+        self.min_periods = min_periods
+        self.center = center
+        self.win_type = win_type
+        self.method = method
+        self._win_freq_i8: int | None = None
+        if self.on is None:
+            self._on = self.obj.index
+        elif isinstance(self.on, Index):
+            self._on = self.on
+        elif isinstance(self.obj, ABCDataFrame) and self.on in self.obj.columns:
+            self._on = Index(self.obj[self.on])
+        else:
+            raise ValueError(
+                f"invalid on specified as {self.on}, "
+                "must be a column (of DataFrame), an Index or None"
+            )
+
+        self._selection = selection
+        self._validate()
+
+    def _validate(self) -> None:
+        if self.center is not None and not is_bool(self.center):
+            raise ValueError("center must be a boolean")
+        if self.min_periods is not None:
+            if not is_integer(self.min_periods):
+                raise ValueError("min_periods must be an integer")
+            if self.min_periods < 0:
+                raise ValueError("min_periods must be >= 0")
+            if is_integer(self.window) and self.min_periods > self.window:
+                raise ValueError(
+                    f"min_periods {self.min_periods} must be <= window {self.window}"
+                )
+        if self.closed is not None and self.closed not in [
+            "right",
+            "both",
+            "left",
+            "neither",
+        ]:
+            raise ValueError("closed must be 'right', 'left', 'both' or 'neither'")
+        if not isinstance(self.obj, (ABCSeries, ABCDataFrame)):
+            raise TypeError(f"invalid type: {type(self)}")
+        if isinstance(self.window, BaseIndexer):
+            # Validate that the passed BaseIndexer subclass has
+            # a get_window_bounds with the correct signature.
+            get_window_bounds_signature = inspect.signature(
+                self.window.get_window_bounds
+            ).parameters.keys()
+            expected_signature = inspect.signature(
+                BaseIndexer().get_window_bounds
+            ).parameters.keys()
+            if get_window_bounds_signature != expected_signature:
+                raise ValueError(
+                    f"{type(self.window).__name__} does not implement "
+                    f"the correct signature for get_window_bounds"
+                )
+        if self.method not in ["table", "single"]:
+            raise ValueError("method must be 'table' or 'single")
+        if self.step is not None:
+            if not is_integer(self.step):
+                raise ValueError("step must be an integer")
+            if self.step < 0:
+                raise ValueError("step must be >= 0")
+
+    def _check_window_bounds(
+        self, start: np.ndarray, end: np.ndarray, num_vals: int
+    ) -> None:
+        if len(start) != len(end):
+            raise ValueError(
+                f"start ({len(start)}) and end ({len(end)}) bounds must be the "
+                f"same length"
+            )
+        if len(start) != (num_vals + (self.step or 1) - 1) // (self.step or 1):
+            raise ValueError(
+                f"start and end bounds ({len(start)}) must be the same length "
+                f"as the object ({num_vals}) divided by the step ({self.step}) "
+                f"if given and rounded up"
+            )
+
+    def _slice_axis_for_step(self, index: Index, result: Sized | None = None) -> Index:
+        """
+        Slices the index for a given result and the preset step.
+        """
+        return (
+            index
+            if result is None or len(result) == len(index)
+            else index[:: self.step]
+        )
+
+    def _validate_numeric_only(self, name: str, numeric_only: bool) -> None:
+        """
+        Validate numeric_only argument, raising if invalid for the input.
+
+        Parameters
+        ----------
+        name : str
+            Name of the operator (kernel).
+        numeric_only : bool
+            Value passed by user.
+        """
+        if (
+            self._selected_obj.ndim == 1
+            and numeric_only
+            and not is_numeric_dtype(self._selected_obj.dtype)
+        ):
+            raise NotImplementedError(
+                f"{type(self).__name__}.{name} does not implement numeric_only"
+            )
+
+    def _make_numeric_only(self, obj: NDFrameT) -> NDFrameT:
+        """Subset DataFrame to numeric columns.
+
+        Parameters
+        ----------
+        obj : DataFrame
+
+        Returns
+        -------
+        obj subset to numeric-only columns.
+        """
+        result = obj.select_dtypes(include=["number"], exclude=["timedelta"])
+        return result
+
+    def _create_data(self, obj: NDFrameT, numeric_only: bool = False) -> NDFrameT:
+        """
+        Split data into blocks & return conformed data.
+        """
+        # filter out the on from the object
+        if self.on is not None and not isinstance(self.on, Index) and obj.ndim == 2:
+            obj = obj.reindex(columns=obj.columns.difference([self.on], sort=False))
+        if obj.ndim > 1 and numeric_only:
+            obj = self._make_numeric_only(obj)
+        return obj
+
+    def _gotitem(self, key, ndim, subset=None):
+        """
+        Sub-classes to define. Return a sliced object.
+
+        Parameters
+        ----------
+        key : str / list of selections
+        ndim : {1, 2}
+            requested ndim of result
+        subset : object, default None
+            subset to act on
+        """
+        # create a new object to prevent aliasing
+        if subset is None:
+            subset = self.obj
+
+        # we need to make a shallow copy of ourselves
+        # with the same groupby
+        kwargs = {attr: getattr(self, attr) for attr in self._attributes}
+
+        selection = self._infer_selection(key, subset)
+        new_win = type(self)(subset, selection=selection, **kwargs)
+        return new_win
+
+    def __getattr__(self, attr: str):
+        if attr in self._internal_names_set:
+            return object.__getattribute__(self, attr)
+        if attr in self.obj:
+            return self[attr]
+
+        raise AttributeError(
+            f"'{type(self).__name__}' object has no attribute '{attr}'"
+        )
+
+    def _dir_additions(self):
+        return self.obj._dir_additions()
+
+    def __repr__(self) -> str:
+        """
+        Provide a nice str repr of our rolling object.
+        """
+        attrs_list = (
+            f"{attr_name}={getattr(self, attr_name)}"
+            for attr_name in self._attributes
+            if getattr(self, attr_name, None) is not None and attr_name[0] != "_"
+        )
+        attrs = ",".join(attrs_list)
+        return f"{type(self).__name__} [{attrs}]"
+
+    def __iter__(self) -> Iterator:
+        obj = self._selected_obj.set_axis(self._on)
+        obj = self._create_data(obj)
+        indexer = self._get_window_indexer()
+
+        start, end = indexer.get_window_bounds(
+            num_values=len(obj),
+            min_periods=self.min_periods,
+            center=self.center,
+            closed=self.closed,
+            step=self.step,
+        )
+        self._check_window_bounds(start, end, len(obj))
+
+        for s, e in zip(start, end):
+            result = obj.iloc[slice(s, e)]
+            yield result
+
+    def _prep_values(self, values: ArrayLike) -> np.ndarray:
+        """Convert input to numpy arrays for Cython routines"""
+        if needs_i8_conversion(values.dtype):
+            raise NotImplementedError(
+                f"ops for {type(self).__name__} for this "
+                f"dtype {values.dtype} are not implemented"
+            )
+        # GH #12373 : rolling functions error on float32 data
+        # make sure the data is coerced to float64
+        try:
+            if isinstance(values, ExtensionArray):
+                values = values.to_numpy(np.float64, na_value=np.nan)
+            else:
+                values = ensure_float64(values)
+        except (ValueError, TypeError) as err:
+            raise TypeError(f"cannot handle this type -> {values.dtype}") from err
+
+        # Convert inf to nan for C funcs
+        inf = np.isinf(values)
+        if inf.any():
+            values = np.where(inf, np.nan, values)
+
+        return values
+
+    def _insert_on_column(self, result: DataFrame, obj: DataFrame) -> None:
+        # if we have an 'on' column we want to put it back into
+        # the results in the same location
+        from pandas import Series
+
+        if self.on is not None and not self._on.equals(obj.index):
+            name = self._on.name
+            extra_col = Series(self._on, index=self.obj.index, name=name, copy=False)
+            if name in result.columns:
+                # TODO: sure we want to overwrite results?
+                result[name] = extra_col
+            elif name in result.index.names:
+                pass
+            elif name in self._selected_obj.columns:
+                # insert in the same location as we had in _selected_obj
+                old_cols = self._selected_obj.columns
+                new_cols = result.columns
+                old_loc = old_cols.get_loc(name)
+                overlap = new_cols.intersection(old_cols[:old_loc])
+                new_loc = len(overlap)
+                result.insert(new_loc, name, extra_col)
+            else:
+                # insert at the end
+                result[name] = extra_col
+
+    @property
+    def _index_array(self) -> npt.NDArray[np.int64] | None:
+        # TODO: why do we get here with e.g. MultiIndex?
+        if isinstance(self._on, (PeriodIndex, DatetimeIndex, TimedeltaIndex)):
+            return self._on.asi8
+        elif isinstance(self._on.dtype, ArrowDtype) and self._on.dtype.kind in "mM":
+            return self._on.to_numpy(dtype=np.int64)
+        return None
+
+    def _resolve_output(self, out: DataFrame, obj: DataFrame) -> DataFrame:
+        """Validate and finalize result."""
+        if out.shape[1] == 0 and obj.shape[1] > 0:
+            raise DataError("No numeric types to aggregate")
+        if out.shape[1] == 0:
+            return obj.astype("float64")
+
+        self._insert_on_column(out, obj)
+        return out
+
+    def _get_window_indexer(self) -> BaseIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+        """
+        if isinstance(self.window, BaseIndexer):
+            return self.window
+        if self._win_freq_i8 is not None:
+            return VariableWindowIndexer(
+                index_array=self._index_array,
+                window_size=self._win_freq_i8,
+                center=self.center,
+            )
+        return FixedWindowIndexer(window_size=self.window)
+
+    def _apply_series(
+        self, homogeneous_func: Callable[..., ArrayLike], name: str | None = None
+    ) -> Series:
+        """
+        Series version of _apply_columnwise
+        """
+        obj = self._create_data(self._selected_obj)
+
+        if name == "count":
+            # GH 12541: Special case for count where we support date-like types
+            obj = notna(obj).astype(int)
+        try:
+            values = self._prep_values(obj._values)
+        except (TypeError, NotImplementedError) as err:
+            raise DataError("No numeric types to aggregate") from err
+
+        result = homogeneous_func(values)
+        index = self._slice_axis_for_step(obj.index, result)
+        return obj._constructor(result, index=index, name=obj.name)
+
+    def _apply_columnwise(
+        self,
+        homogeneous_func: Callable[..., ArrayLike],
+        name: str,
+        numeric_only: bool = False,
+    ) -> DataFrame | Series:
+        """
+        Apply the given function to the DataFrame broken down into homogeneous
+        sub-frames.
+        """
+        self._validate_numeric_only(name, numeric_only)
+        if self._selected_obj.ndim == 1:
+            return self._apply_series(homogeneous_func, name)
+
+        obj = self._create_data(self._selected_obj, numeric_only)
+        if name == "count":
+            # GH 12541: Special case for count where we support date-like types
+            obj = notna(obj).astype(int)
+            obj._mgr = obj._mgr.consolidate()
+
+        taker = []
+        res_values = []
+        for i, arr in enumerate(obj._iter_column_arrays()):
+            # GH#42736 operate column-wise instead of block-wise
+            # As of 2.0, hfunc will raise for nuisance columns
+            try:
+                arr = self._prep_values(arr)
+            except (TypeError, NotImplementedError) as err:
+                raise DataError(
+                    f"Cannot aggregate non-numeric type: {arr.dtype}"
+                ) from err
+            res = homogeneous_func(arr)
+            res_values.append(res)
+            taker.append(i)
+
+        index = self._slice_axis_for_step(
+            obj.index, res_values[0] if len(res_values) > 0 else None
+        )
+        df = type(obj)._from_arrays(
+            res_values,
+            index=index,
+            columns=obj.columns.take(taker),
+            verify_integrity=False,
+        )
+
+        return self._resolve_output(df, obj)
+
+    def _apply_tablewise(
+        self,
+        homogeneous_func: Callable[..., ArrayLike],
+        name: str | None = None,
+        numeric_only: bool = False,
+    ) -> DataFrame | Series:
+        """
+        Apply the given function to the DataFrame across the entire object
+        """
+        if self._selected_obj.ndim == 1:
+            raise ValueError("method='table' not applicable for Series objects.")
+        obj = self._create_data(self._selected_obj, numeric_only)
+        values = self._prep_values(obj.to_numpy())
+        result = homogeneous_func(values)
+        index = self._slice_axis_for_step(obj.index, result)
+        columns = (
+            obj.columns
+            if result.shape[1] == len(obj.columns)
+            else obj.columns[:: self.step]
+        )
+        out = obj._constructor(result, index=index, columns=columns)
+
+        return self._resolve_output(out, obj)
+
+    def _apply_pairwise(
+        self,
+        target: DataFrame | Series,
+        other: DataFrame | Series | None,
+        pairwise: bool | None,
+        func: Callable[[DataFrame | Series, DataFrame | Series], DataFrame | Series],
+        numeric_only: bool,
+    ) -> DataFrame | Series:
+        """
+        Apply the given pairwise function given 2 pandas objects (DataFrame/Series)
+        """
+        target = self._create_data(target, numeric_only)
+        if other is None:
+            other = target
+            # only default unset
+            pairwise = True if pairwise is None else pairwise
+        elif not isinstance(other, (ABCDataFrame, ABCSeries)):
+            raise ValueError("other must be a DataFrame or Series")
+        elif other.ndim == 2 and numeric_only:
+            other = self._make_numeric_only(other)
+
+        return flex_binary_moment(target, other, func, pairwise=bool(pairwise))
+
+    def _apply(
+        self,
+        func: Callable[..., Any],
+        name: str,
+        numeric_only: bool = False,
+        numba_args: tuple[Any, ...] = (),
+        **kwargs,
+    ):
+        """
+        Rolling statistical measure using supplied function.
+
+        Designed to be used with passed-in Cython array-based functions.
+
+        Parameters
+        ----------
+        func : callable function to apply
+        name : str,
+        numba_args : tuple
+            args to be passed when func is a numba func
+        **kwargs
+            additional arguments for rolling function and window function
+
+        Returns
+        -------
+        y : type of input
+        """
+        window_indexer = self._get_window_indexer()
+        min_periods = (
+            self.min_periods
+            if self.min_periods is not None
+            else window_indexer.window_size
+        )
+
+        def homogeneous_func(values: np.ndarray):
+            # calculation function
+
+            if values.size == 0:
+                return values.copy()
+
+            def calc(x):
+                start, end = window_indexer.get_window_bounds(
+                    num_values=len(x),
+                    min_periods=min_periods,
+                    center=self.center,
+                    closed=self.closed,
+                    step=self.step,
+                )
+                self._check_window_bounds(start, end, len(x))
+
+                return func(x, start, end, min_periods, *numba_args)
+
+            with np.errstate(all="ignore"):
+                result = calc(values)
+
+            return result
+
+        if self.method == "single":
+            return self._apply_columnwise(homogeneous_func, name, numeric_only)
+        else:
+            return self._apply_tablewise(homogeneous_func, name, numeric_only)
+
+    def _numba_apply(
+        self,
+        func: Callable[..., Any],
+        engine_kwargs: dict[str, bool] | None = None,
+        **func_kwargs,
+    ):
+        window_indexer = self._get_window_indexer()
+        min_periods = (
+            self.min_periods
+            if self.min_periods is not None
+            else window_indexer.window_size
+        )
+        obj = self._create_data(self._selected_obj)
+        values = self._prep_values(obj.to_numpy())
+        if values.ndim == 1:
+            values = values.reshape(-1, 1)
+        start, end = window_indexer.get_window_bounds(
+            num_values=len(values),
+            min_periods=min_periods,
+            center=self.center,
+            closed=self.closed,
+            step=self.step,
+        )
+        self._check_window_bounds(start, end, len(values))
+        # For now, map everything to float to match the Cython impl
+        # even though it is wrong
+        # TODO: Could preserve correct dtypes in future
+        # xref #53214
+        dtype_mapping = executor.float_dtype_mapping
+        aggregator = executor.generate_shared_aggregator(
+            func,
+            dtype_mapping,
+            is_grouped_kernel=False,
+            **get_jit_arguments(engine_kwargs),
+        )
+        result = aggregator(
+            values.T, start=start, end=end, min_periods=min_periods, **func_kwargs
+        ).T
+        index = self._slice_axis_for_step(obj.index, result)
+        if obj.ndim == 1:
+            result = result.squeeze()
+            out = obj._constructor(result, index=index, name=obj.name)
+            return out
+        else:
+            columns = self._slice_axis_for_step(obj.columns, result.T)
+            out = obj._constructor(result, index=index, columns=columns)
+            return self._resolve_output(out, obj)
+
+    def aggregate(self, func=None, *args, **kwargs):
+        relabeling, func, columns, order = reconstruct_func(func, **kwargs)
+        result = ResamplerWindowApply(self, func, args=args, kwargs=kwargs).agg()
+        if isinstance(result, ABCDataFrame) and relabeling:
+            result = result.iloc[:, order]
+            result.columns = columns  # type: ignore[union-attr]
+        if result is None:
+            return self.apply(func, raw=False, args=args, kwargs=kwargs)
+        return result
+
+    agg = aggregate
+
+
+class BaseWindowGroupby(BaseWindow):
+    """
+    Provide the groupby windowing facilities.
+    """
+
+    _grouper: BaseGrouper
+    _as_index: bool
+    _attributes: list[str] = ["_grouper"]
+
+    def __init__(
+        self,
+        obj: DataFrame | Series,
+        *args,
+        _grouper: BaseGrouper,
+        _as_index: bool = True,
+        **kwargs,
+    ) -> None:
+        from pandas.core.groupby.ops import BaseGrouper
+
+        if not isinstance(_grouper, BaseGrouper):
+            raise ValueError("Must pass a BaseGrouper object.")
+        self._grouper = _grouper
+        self._as_index = _as_index
+        # GH 32262: It's convention to keep the grouping column in
+        # groupby.<agg_func>, but unexpected to users in
+        # groupby.rolling.<agg_func>
+        obj = obj.drop(columns=self._grouper.names, errors="ignore")
+        # GH 15354
+        if kwargs.get("step") is not None:
+            raise NotImplementedError("step not implemented for groupby")
+        super().__init__(obj, *args, **kwargs)
+
+    def _apply(
+        self,
+        func: Callable[..., Any],
+        name: str,
+        numeric_only: bool = False,
+        numba_args: tuple[Any, ...] = (),
+        **kwargs,
+    ) -> DataFrame | Series:
+        result = super()._apply(
+            func,
+            name,
+            numeric_only,
+            numba_args,
+            **kwargs,
+        )
+        # Reconstruct the resulting MultiIndex
+        # 1st set of levels = group by labels
+        # 2nd set of levels = original DataFrame/Series index
+        grouped_object_index = self.obj.index
+        grouped_index_name = [*grouped_object_index.names]
+        groupby_keys = copy.copy(self._grouper.names)
+        result_index_names = groupby_keys + grouped_index_name
+
+        drop_columns = [
+            key
+            for key in self._grouper.names
+            if key not in self.obj.index.names or key is None
+        ]
+
+        if len(drop_columns) != len(groupby_keys):
+            # Our result will have still kept the column in the result
+            result = result.drop(columns=drop_columns, errors="ignore")
+
+        codes = self._grouper.codes
+        levels = copy.copy(self._grouper.levels)
+
+        group_indices = self._grouper.indices.values()
+        if group_indices:
+            indexer = np.concatenate(list(group_indices))
+        else:
+            indexer = np.array([], dtype=np.intp)
+        codes = [c.take(indexer) for c in codes]
+
+        # if the index of the original dataframe needs to be preserved, append
+        # this index (but reordered) to the codes/levels from the groupby
+        if grouped_object_index is not None:
+            idx = grouped_object_index.take(indexer)
+            if not isinstance(idx, MultiIndex):
+                idx = MultiIndex.from_arrays([idx])
+            codes.extend(list(idx.codes))
+            levels.extend(list(idx.levels))
+
+        result_index = MultiIndex(
+            levels, codes, names=result_index_names, verify_integrity=False
+        )
+
+        result.index = result_index
+        if not self._as_index:
+            result = result.reset_index(level=list(range(len(groupby_keys))))
+        return result
+
+    def _apply_pairwise(
+        self,
+        target: DataFrame | Series,
+        other: DataFrame | Series | None,
+        pairwise: bool | None,
+        func: Callable[[DataFrame | Series, DataFrame | Series], DataFrame | Series],
+        numeric_only: bool,
+    ) -> DataFrame | Series:
+        """
+        Apply the given pairwise function given 2 pandas objects (DataFrame/Series)
+        """
+        # Manually drop the grouping column first
+        target = target.drop(columns=self._grouper.names, errors="ignore")
+        result = super()._apply_pairwise(target, other, pairwise, func, numeric_only)
+        # 1) Determine the levels + codes of the groupby levels
+        if other is not None and not all(
+            len(group) == len(other) for group in self._grouper.indices.values()
+        ):
+            # GH 42915
+            # len(other) != len(any group), so must reindex (expand) the result
+            # from flex_binary_moment to a "transform"-like result
+            # per groupby combination
+            old_result_len = len(result)
+            result = concat(
+                [
+                    result.take(gb_indices).reindex(result.index)
+                    for gb_indices in self._grouper.indices.values()
+                ]
+            )
+
+            gb_pairs = (
+                com.maybe_make_list(pair) for pair in self._grouper.indices.keys()
+            )
+            groupby_codes = []
+            groupby_levels = []
+            # e.g. [[1, 2], [4, 5]] as [[1, 4], [2, 5]]
+            for gb_level_pair in map(list, zip(*gb_pairs)):
+                labels = np.repeat(np.array(gb_level_pair), old_result_len)
+                codes, levels = factorize(labels)
+                groupby_codes.append(codes)
+                groupby_levels.append(levels)
+        else:
+            # pairwise=True or len(other) == len(each group), so repeat
+            # the groupby labels by the number of columns in the original object
+            groupby_codes = self._grouper.codes
+            # error: Incompatible types in assignment (expression has type
+            # "List[Index]", variable has type "List[Union[ndarray, Index]]")
+            groupby_levels = self._grouper.levels  # type: ignore[assignment]
+
+            group_indices = self._grouper.indices.values()
+            if group_indices:
+                indexer = np.concatenate(list(group_indices))
+            else:
+                indexer = np.array([], dtype=np.intp)
+
+            if target.ndim == 1:
+                repeat_by = 1
+            else:
+                repeat_by = len(target.columns)
+            groupby_codes = [
+                np.repeat(c.take(indexer), repeat_by) for c in groupby_codes
+            ]
+        # 2) Determine the levels + codes of the result from super()._apply_pairwise
+        if isinstance(result.index, MultiIndex):
+            result_codes = list(result.index.codes)
+            result_levels = list(result.index.levels)
+            result_names = list(result.index.names)
+        else:
+            idx_codes, idx_levels = factorize(result.index)
+            result_codes = [idx_codes]
+            result_levels = [idx_levels]
+            result_names = [result.index.name]
+
+            # 3) Create the resulting index by combining 1) + 2)
+        result_codes = groupby_codes + result_codes
+        result_levels = groupby_levels + result_levels
+        result_names = self._grouper.names + result_names
+
+        result_index = MultiIndex(
+            result_levels, result_codes, names=result_names, verify_integrity=False
+        )
+        result.index = result_index
+        return result
+
+    def _create_data(self, obj: NDFrameT, numeric_only: bool = False) -> NDFrameT:
+        """
+        Split data into blocks & return conformed data.
+        """
+        # Ensure the object we're rolling over is monotonically sorted relative
+        # to the groups
+        # GH 36197
+        if not obj.empty:
+            groupby_order = np.concatenate(list(self._grouper.indices.values())).astype(
+                np.int64
+            )
+            obj = obj.take(groupby_order)
+        return super()._create_data(obj, numeric_only)
+
+    def _gotitem(self, key, ndim, subset=None):
+        # we are setting the index on the actual object
+        # here so our index is carried through to the selected obj
+        # when we do the splitting for the groupby
+        if self.on is not None:
+            # GH 43355
+            subset = self.obj.set_index(self._on)
+        return super()._gotitem(key, ndim, subset=subset)
+
+
+class Window(BaseWindow):
+    """
+    Provide rolling window calculations.
+
+    Parameters
+    ----------
+    window : int, timedelta, str, offset, or BaseIndexer subclass
+        Interval of the moving window.
+
+        If an integer, the delta between the start and end of each window.
+        The number of points in the window depends on the ``closed`` argument.
+
+        If a timedelta, str, or offset, the time period of each window. Each
+        window will be a variable sized based on the observations included in
+        the time-period. This is only valid for datetimelike indexes.
+        To learn more about the offsets & frequency strings, please see
+        :ref:`this link<timeseries.offset_aliases>`.
+
+        If a BaseIndexer subclass, the window boundaries
+        based on the defined ``get_window_bounds`` method. Additional rolling
+        keyword arguments, namely ``min_periods``, ``center``, ``closed`` and
+        ``step`` will be passed to ``get_window_bounds``.
+
+    min_periods : int, default None
+        Minimum number of observations in window required to have a value;
+        otherwise, result is ``np.nan``.
+
+        For a window that is specified by an offset, ``min_periods`` will default to 1.
+
+        For a window that is specified by an integer, ``min_periods`` will default
+        to the size of the window.
+
+    center : bool, default False
+        If False, set the window labels as the right edge of the window index.
+
+        If True, set the window labels as the center of the window index.
+
+    win_type : str, default None
+        If ``None``, all points are evenly weighted.
+
+        If a string, it must be a valid `scipy.signal window function
+        <https://docs.scipy.org/doc/scipy/reference/signal.windows.html#module-scipy.signal.windows>`__.
+
+        Certain Scipy window types require additional parameters to be passed
+        in the aggregation function. The additional parameters must match
+        the keywords specified in the Scipy window type method signature.
+
+    on : str, optional
+        For a DataFrame, a column label or Index level on which
+        to calculate the rolling window, rather than the DataFrame's index.
+
+        Provided integer column is ignored and excluded from result since
+        an integer index is not used to calculate the rolling window.
+
+    closed : str, default None
+        Determines the inclusivity of points in the window
+
+        If ``'right'``, uses the window (first, last] meaning the last point
+        is included in the calculations.
+
+        If ``'left'``, uses the window [first, last) meaning the first point
+        is included in the calculations.
+
+        If ``'both'``, uses the window [first, last] meaning all points in
+        the window are included in the calculations.
+
+        If ``'neither'``, uses the window (first, last) meaning the first
+        and last points in the window are excluded from calculations.
+
+        () and [] are referencing open and closed set
+        notation respetively.
+
+        Default ``None`` (``'right'``).
+
+    step : int, default None
+        Evaluate the window at every ``step`` result, equivalent to slicing as
+        ``[::step]``. ``window`` must be an integer. Using a step argument other
+        than None or 1 will produce a result with a different shape than the input.
+
+        .. versionadded:: 1.5.0
+
+    method : str {'single', 'table'}, default 'single'
+
+        .. versionadded:: 1.3.0
+
+        Execute the rolling operation per single column or row (``'single'``)
+        or over the entire object (``'table'``).
+
+        This argument is only implemented when specifying ``engine='numba'``
+        in the method call.
+
+    Returns
+    -------
+    pandas.api.typing.Window or pandas.api.typing.Rolling
+        An instance of Window is returned if ``win_type`` is passed. Otherwise,
+        an instance of Rolling is returned.
+
+    See Also
+    --------
+    expanding : Provides expanding transformations.
+    ewm : Provides exponential weighted functions.
+
+    Notes
+    -----
+    See :ref:`Windowing Operations <window.generic>` for further usage details
+    and examples.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    >>> df
+         B
+    0  0.0
+    1  1.0
+    2  2.0
+    3  NaN
+    4  4.0
+
+    **window**
+
+    Rolling sum with a window length of 2 observations.
+
+    >>> df.rolling(2).sum()
+         B
+    0  NaN
+    1  1.0
+    2  3.0
+    3  NaN
+    4  NaN
+
+    Rolling sum with a window span of 2 seconds.
+
+    >>> df_time = pd.DataFrame(
+    ...     {"B": [0, 1, 2, np.nan, 4]},
+    ...     index=[
+    ...         pd.Timestamp("20130101 09:00:00"),
+    ...         pd.Timestamp("20130101 09:00:02"),
+    ...         pd.Timestamp("20130101 09:00:03"),
+    ...         pd.Timestamp("20130101 09:00:05"),
+    ...         pd.Timestamp("20130101 09:00:06"),
+    ...     ],
+    ... )
+
+    >>> df_time
+                           B
+    2013-01-01 09:00:00  0.0
+    2013-01-01 09:00:02  1.0
+    2013-01-01 09:00:03  2.0
+    2013-01-01 09:00:05  NaN
+    2013-01-01 09:00:06  4.0
+
+    >>> df_time.rolling("2s").sum()
+                           B
+    2013-01-01 09:00:00  0.0
+    2013-01-01 09:00:02  1.0
+    2013-01-01 09:00:03  3.0
+    2013-01-01 09:00:05  NaN
+    2013-01-01 09:00:06  4.0
+
+    Rolling sum with forward looking windows with 2 observations.
+
+    >>> indexer = pd.api.indexers.FixedForwardWindowIndexer(window_size=2)
+    >>> df.rolling(window=indexer, min_periods=1).sum()
+         B
+    0  1.0
+    1  3.0
+    2  2.0
+    3  4.0
+    4  4.0
+
+    **min_periods**
+
+    Rolling sum with a window length of 2 observations, but only needs a minimum of 1
+    observation to calculate a value.
+
+    >>> df.rolling(2, min_periods=1).sum()
+         B
+    0  0.0
+    1  1.0
+    2  3.0
+    3  2.0
+    4  4.0
+
+    **center**
+
+    Rolling sum with the result assigned to the center of the window index.
+
+    >>> df.rolling(3, min_periods=1, center=True).sum()
+         B
+    0  1.0
+    1  3.0
+    2  3.0
+    3  6.0
+    4  4.0
+
+    >>> df.rolling(3, min_periods=1, center=False).sum()
+         B
+    0  0.0
+    1  1.0
+    2  3.0
+    3  3.0
+    4  6.0
+
+    **step**
+
+    Rolling sum with a window length of 2 observations, minimum of 1 observation to
+    calculate a value, and a step of 2.
+
+    >>> df.rolling(2, min_periods=1, step=2).sum()
+         B
+    0  0.0
+    2  3.0
+    4  4.0
+
+    **win_type**
+
+    Rolling sum with a window length of 2, using the Scipy ``'gaussian'``
+    window type. ``std`` is required in the aggregation function.
+
+    >>> df.rolling(2, win_type="gaussian").sum(std=3)
+              B
+    0       NaN
+    1  0.986207
+    2  2.958621
+    3       NaN
+    4       NaN
+
+    **on**
+
+    Rolling sum with a window length of 2 days.
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "A": [
+    ...             pd.to_datetime("2020-01-01"),
+    ...             pd.to_datetime("2020-01-01"),
+    ...             pd.to_datetime("2020-01-02"),
+    ...         ],
+    ...         "B": [1, 2, 3],
+    ...     },
+    ...     index=pd.date_range("2020", periods=3),
+    ... )
+
+    >>> df
+                        A  B
+    2020-01-01 2020-01-01  1
+    2020-01-02 2020-01-01  2
+    2020-01-03 2020-01-02  3
+
+    >>> df.rolling("2D", on="A").sum()
+                        A    B
+    2020-01-01 2020-01-01  1.0
+    2020-01-02 2020-01-01  3.0
+    2020-01-03 2020-01-02  6.0
+    """
+
+    _attributes = [
+        "window",
+        "min_periods",
+        "center",
+        "win_type",
+        "on",
+        "closed",
+        "step",
+        "method",
+    ]
+
+    def _validate(self) -> None:
+        super()._validate()
+
+        if not isinstance(self.win_type, str):
+            raise ValueError(f"Invalid win_type {self.win_type}")
+        signal = import_optional_dependency(
+            "scipy.signal.windows", extra="Scipy is required to generate window weight."
+        )
+        self._scipy_weight_generator = getattr(signal, self.win_type, None)
+        if self._scipy_weight_generator is None:
+            raise ValueError(f"Invalid win_type {self.win_type}")
+
+        if isinstance(self.window, BaseIndexer):
+            raise NotImplementedError(
+                "BaseIndexer subclasses not implemented with win_types."
+            )
+        if not is_integer(self.window) or self.window < 0:
+            raise ValueError("window must be an integer 0 or greater")
+
+        if self.method != "single":
+            raise NotImplementedError("'single' is the only supported method type.")
+
+    def _center_window(self, result: np.ndarray, offset: int) -> np.ndarray:
+        """
+        Center the result in the window for weighted rolling aggregations.
+        """
+        if offset > 0:
+            lead_indexer = [slice(offset, None)]
+            result = np.copy(result[tuple(lead_indexer)])
+        return result
+
+    def _apply(
+        self,
+        func: Callable[[np.ndarray, int, int], np.ndarray],
+        name: str,
+        numeric_only: bool = False,
+        numba_args: tuple[Any, ...] = (),
+        **kwargs,
+    ):
+        """
+        Rolling with weights statistical measure using supplied function.
+
+        Designed to be used with passed-in Cython array-based functions.
+
+        Parameters
+        ----------
+        func : callable function to apply
+        name : str,
+        numeric_only : bool, default False
+            Whether to only operate on bool, int, and float columns
+        numba_args : tuple
+            unused
+        **kwargs
+            additional arguments for scipy windows if necessary
+
+        Returns
+        -------
+        y : type of input
+        """
+        # "None" not callable  [misc]
+        window = self._scipy_weight_generator(  # type: ignore[misc]
+            self.window, **kwargs
+        )
+        offset = (len(window) - 1) // 2 if self.center else 0
+
+        def homogeneous_func(values: np.ndarray):
+            # calculation function
+
+            if values.size == 0:
+                return values.copy()
+
+            def calc(x):
+                additional_nans = np.full(offset, np.nan)
+                x = np.concatenate((x, additional_nans))
+                return func(
+                    x,
+                    window,
+                    self.min_periods if self.min_periods is not None else len(window),
+                )
+
+            with np.errstate(all="ignore"):
+                # Our weighted aggregations return memoryviews
+                result = np.asarray(calc(values))
+
+            if self.center:
+                result = self._center_window(result, offset)
+
+            return result
+
+        return self._apply_columnwise(homogeneous_func, name, numeric_only)[
+            :: self.step
+        ]
+
+    @doc(
+        _shared_docs["aggregate"],
+        see_also=dedent(
+            """
+        See Also
+        --------
+        DataFrame.aggregate : Similar DataFrame method.
+        Series.aggregate : Similar Series method.
+        """
+        ),
+        examples=dedent(
+            """
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        >>> df
+           A  B  C
+        0  1  4  7
+        1  2  5  8
+        2  3  6  9
+
+        >>> df.rolling(2, win_type="boxcar").agg("mean")
+             A    B    C
+        0  NaN  NaN  NaN
+        1  1.5  4.5  7.5
+        2  2.5  5.5  8.5
+        """
+        ),
+        klass="Series/DataFrame",
+        axis="",
+    )
+    def aggregate(self, func=None, *args, **kwargs):
+        result = ResamplerWindowApply(self, func, args=args, kwargs=kwargs).agg()
+        if result is None:
+            # these must apply directly
+            result = func(self)
+
+        return result
+
+    agg = aggregate
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        kwargs_scipy,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([0, 1, 5, 2, 8])
+
+        To get an instance of :class:`~pandas.core.window.rolling.Window` we need
+        to pass the parameter `win_type`.
+
+        >>> type(ser.rolling(2, win_type='gaussian'))
+        <class 'pandas.core.window.rolling.Window'>
+
+        In order to use the `SciPy` Gaussian window we need to provide the parameters
+        `M` and `std`. The parameter `M` corresponds to 2 in our example.
+        We pass the second parameter `std` as a parameter of the following method
+        (`sum` in this case):
+
+        >>> ser.rolling(2, win_type='gaussian').sum(std=3)
+        0         NaN
+        1    0.986207
+        2    5.917243
+        3    6.903450
+        4    9.862071
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="weighted window sum",
+        agg_method="sum",
+    )
+    def sum(self, numeric_only: bool = False, **kwargs):
+        window_func = window_aggregations.roll_weighted_sum
+        # error: Argument 1 to "_apply" of "Window" has incompatible type
+        # "Callable[[ndarray, ndarray, int], ndarray]"; expected
+        # "Callable[[ndarray, int, int], ndarray]"
+        return self._apply(
+            window_func,  # type: ignore[arg-type]
+            name="sum",
+            numeric_only=numeric_only,
+            **kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        kwargs_scipy,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([0, 1, 5, 2, 8])
+
+        To get an instance of :class:`~pandas.core.window.rolling.Window` we need
+        to pass the parameter `win_type`.
+
+        >>> type(ser.rolling(2, win_type='gaussian'))
+        <class 'pandas.core.window.rolling.Window'>
+
+        In order to use the `SciPy` Gaussian window we need to provide the parameters
+        `M` and `std`. The parameter `M` corresponds to 2 in our example.
+        We pass the second parameter `std` as a parameter of the following method:
+
+        >>> ser.rolling(2, win_type='gaussian').mean(std=3)
+        0    NaN
+        1    0.5
+        2    3.0
+        3    3.5
+        4    5.0
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="weighted window mean",
+        agg_method="mean",
+    )
+    def mean(self, numeric_only: bool = False, **kwargs):
+        window_func = window_aggregations.roll_weighted_mean
+        # error: Argument 1 to "_apply" of "Window" has incompatible type
+        # "Callable[[ndarray, ndarray, int], ndarray]"; expected
+        # "Callable[[ndarray, int, int], ndarray]"
+        return self._apply(
+            window_func,  # type: ignore[arg-type]
+            name="mean",
+            numeric_only=numeric_only,
+            **kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        kwargs_scipy,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([0, 1, 5, 2, 8])
+
+        To get an instance of :class:`~pandas.core.window.rolling.Window` we need
+        to pass the parameter `win_type`.
+
+        >>> type(ser.rolling(2, win_type='gaussian'))
+        <class 'pandas.core.window.rolling.Window'>
+
+        In order to use the `SciPy` Gaussian window we need to provide the parameters
+        `M` and `std`. The parameter `M` corresponds to 2 in our example.
+        We pass the second parameter `std` as a parameter of the following method:
+
+        >>> ser.rolling(2, win_type='gaussian').var(std=3)
+        0     NaN
+        1     0.5
+        2     8.0
+        3     4.5
+        4    18.0
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="weighted window variance",
+        agg_method="var",
+    )
+    def var(self, ddof: int = 1, numeric_only: bool = False, **kwargs):
+        window_func = partial(window_aggregations.roll_weighted_var, ddof=ddof)
+        kwargs.pop("name", None)
+        return self._apply(window_func, name="var", numeric_only=numeric_only, **kwargs)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        kwargs_scipy,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([0, 1, 5, 2, 8])
+
+        To get an instance of :class:`~pandas.core.window.rolling.Window` we need
+        to pass the parameter `win_type`.
+
+        >>> type(ser.rolling(2, win_type='gaussian'))
+        <class 'pandas.core.window.rolling.Window'>
+
+        In order to use the `SciPy` Gaussian window we need to provide the parameters
+        `M` and `std`. The parameter `M` corresponds to 2 in our example.
+        We pass the second parameter `std` as a parameter of the following method:
+
+        >>> ser.rolling(2, win_type='gaussian').std(std=3)
+        0         NaN
+        1    0.707107
+        2    2.828427
+        3    2.121320
+        4    4.242641
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="weighted window standard deviation",
+        agg_method="std",
+    )
+    def std(self, ddof: int = 1, numeric_only: bool = False, **kwargs):
+        return zsqrt(
+            self.var(ddof=ddof, name="std", numeric_only=numeric_only, **kwargs)
+        )
+
+
+class RollingAndExpandingMixin(BaseWindow):
+    def count(self, numeric_only: bool = False):
+        window_func = window_aggregations.roll_sum
+        return self._apply(window_func, name="count", numeric_only=numeric_only)
+
+    def apply(
+        self,
+        func: Callable[..., Any],
+        raw: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        args: tuple[Any, ...] | None = None,
+        kwargs: dict[str, Any] | None = None,
+    ):
+        if args is None:
+            args = ()
+        if kwargs is None:
+            kwargs = {}
+
+        if not is_bool(raw):
+            raise ValueError("raw parameter must be `True` or `False`")
+
+        numba_args: tuple[Any, ...] = ()
+        if maybe_use_numba(engine):
+            if raw is False:
+                raise ValueError("raw must be `True` when using the numba engine")
+            numba_args, kwargs = prepare_function_arguments(
+                func, args, kwargs, num_required_args=1
+            )
+            if self.method == "single":
+                apply_func = generate_numba_apply_func(
+                    func, **get_jit_arguments(engine_kwargs)
+                )
+            else:
+                apply_func = generate_numba_table_func(
+                    func, **get_jit_arguments(engine_kwargs)
+                )
+        elif engine in ("cython", None):
+            if engine_kwargs is not None:
+                raise ValueError("cython engine does not accept engine_kwargs")
+            apply_func = self._generate_cython_apply_func(args, kwargs, raw, func)
+        else:
+            raise ValueError("engine must be either 'numba' or 'cython'")
+
+        return self._apply(
+            apply_func,
+            name="apply",
+            numba_args=numba_args,
+        )
+
+    def _generate_cython_apply_func(
+        self,
+        args: tuple[Any, ...],
+        kwargs: dict[str, Any],
+        raw: bool | np.bool_,
+        function: Callable[..., Any],
+    ) -> Callable[[np.ndarray, np.ndarray, np.ndarray, int], np.ndarray]:
+        from pandas import Series
+
+        window_func = partial(
+            window_aggregations.roll_apply,
+            args=args,
+            kwargs=kwargs,
+            raw=bool(raw),
+            function=function,
+        )
+
+        def apply_func(values, begin, end, min_periods, raw=raw):
+            if not raw:
+                # GH 45912
+                values = Series(values, index=self._on, copy=False)
+            return window_func(values, begin, end, min_periods)
+
+        return apply_func
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        return com.pipe(self, func, *args, **kwargs)
+
+    def sum(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                func = generate_manual_numpy_nan_agg_with_axis(np.nansum)
+                return self.apply(
+                    func,
+                    raw=True,
+                    engine=engine,
+                    engine_kwargs=engine_kwargs,
+                )
+            else:
+                from pandas.core._numba.kernels import sliding_sum
+
+                return self._numba_apply(sliding_sum, engine_kwargs)
+        window_func = window_aggregations.roll_sum
+        return self._apply(window_func, name="sum", numeric_only=numeric_only)
+
+    def max(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                func = generate_manual_numpy_nan_agg_with_axis(np.nanmax)
+                return self.apply(
+                    func,
+                    raw=True,
+                    engine=engine,
+                    engine_kwargs=engine_kwargs,
+                )
+            else:
+                from pandas.core._numba.kernels import sliding_min_max
+
+                return self._numba_apply(sliding_min_max, engine_kwargs, is_max=True)
+        window_func = window_aggregations.roll_max
+        return self._apply(window_func, name="max", numeric_only=numeric_only)
+
+    def min(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                func = generate_manual_numpy_nan_agg_with_axis(np.nanmin)
+                return self.apply(
+                    func,
+                    raw=True,
+                    engine=engine,
+                    engine_kwargs=engine_kwargs,
+                )
+            else:
+                from pandas.core._numba.kernels import sliding_min_max
+
+                return self._numba_apply(sliding_min_max, engine_kwargs, is_max=False)
+        window_func = window_aggregations.roll_min
+        return self._apply(window_func, name="min", numeric_only=numeric_only)
+
+    def mean(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                func = generate_manual_numpy_nan_agg_with_axis(np.nanmean)
+                return self.apply(
+                    func,
+                    raw=True,
+                    engine=engine,
+                    engine_kwargs=engine_kwargs,
+                )
+            else:
+                from pandas.core._numba.kernels import sliding_mean
+
+                return self._numba_apply(sliding_mean, engine_kwargs)
+        window_func = window_aggregations.roll_mean
+        return self._apply(window_func, name="mean", numeric_only=numeric_only)
+
+    def median(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                func = generate_manual_numpy_nan_agg_with_axis(np.nanmedian)
+            else:
+                func = np.nanmedian
+
+            return self.apply(
+                func,
+                raw=True,
+                engine=engine,
+                engine_kwargs=engine_kwargs,
+            )
+        window_func = window_aggregations.roll_median_c
+        return self._apply(window_func, name="median", numeric_only=numeric_only)
+
+    def std(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                raise NotImplementedError("std not supported with method='table'")
+            from pandas.core._numba.kernels import sliding_var
+
+            return zsqrt(self._numba_apply(sliding_var, engine_kwargs, ddof=ddof))
+        window_func = window_aggregations.roll_var
+
+        def zsqrt_func(values, begin, end, min_periods):
+            return zsqrt(window_func(values, begin, end, min_periods, ddof=ddof))
+
+        return self._apply(
+            zsqrt_func,
+            name="std",
+            numeric_only=numeric_only,
+        )
+
+    def var(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        if maybe_use_numba(engine):
+            if self.method == "table":
+                raise NotImplementedError("var not supported with method='table'")
+            from pandas.core._numba.kernels import sliding_var
+
+            return self._numba_apply(sliding_var, engine_kwargs, ddof=ddof)
+        window_func = partial(window_aggregations.roll_var, ddof=ddof)
+        return self._apply(
+            window_func,
+            name="var",
+            numeric_only=numeric_only,
+        )
+
+    def skew(self, numeric_only: bool = False):
+        window_func = window_aggregations.roll_skew
+        return self._apply(
+            window_func,
+            name="skew",
+            numeric_only=numeric_only,
+        )
+
+    def sem(self, ddof: int = 1, numeric_only: bool = False):
+        # Raise here so error message says sem instead of std
+        self._validate_numeric_only("sem", numeric_only)
+        return self.std(numeric_only=numeric_only) / (
+            self.count(numeric_only=numeric_only) - ddof
+        ).pow(0.5)
+
+    def kurt(self, numeric_only: bool = False):
+        window_func = window_aggregations.roll_kurt
+        return self._apply(
+            window_func,
+            name="kurt",
+            numeric_only=numeric_only,
+        )
+
+    def first(self, numeric_only: bool = False):
+        window_func = window_aggregations.roll_first
+        return self._apply(
+            window_func,
+            name="first",
+            numeric_only=numeric_only,
+        )
+
+    def last(self, numeric_only: bool = False):
+        window_func = window_aggregations.roll_last
+        return self._apply(
+            window_func,
+            name="last",
+            numeric_only=numeric_only,
+        )
+
+    def quantile(
+        self,
+        q: float,
+        interpolation: QuantileInterpolation = "linear",
+        numeric_only: bool = False,
+    ):
+        if q == 1.0:
+            window_func = window_aggregations.roll_max
+        elif q == 0.0:
+            window_func = window_aggregations.roll_min
+        else:
+            window_func = partial(
+                window_aggregations.roll_quantile,
+                quantile=q,
+                interpolation=interpolation,
+            )
+
+        return self._apply(window_func, name="quantile", numeric_only=numeric_only)
+
+    def rank(
+        self,
+        method: WindowingRankType = "average",
+        ascending: bool = True,
+        pct: bool = False,
+        numeric_only: bool = False,
+    ):
+        window_func = partial(
+            window_aggregations.roll_rank,
+            method=method,
+            ascending=ascending,
+            percentile=pct,
+        )
+
+        return self._apply(window_func, name="rank", numeric_only=numeric_only)
+
+    def cov(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        if self.step is not None:
+            raise NotImplementedError("step not implemented for cov")
+        self._validate_numeric_only("cov", numeric_only)
+
+        from pandas import Series
+
+        def cov_func(x, y):
+            x_array = self._prep_values(x)
+            y_array = self._prep_values(y)
+            window_indexer = self._get_window_indexer()
+            min_periods = (
+                self.min_periods
+                if self.min_periods is not None
+                else window_indexer.window_size
+            )
+            start, end = window_indexer.get_window_bounds(
+                num_values=len(x_array),
+                min_periods=min_periods,
+                center=self.center,
+                closed=self.closed,
+                step=self.step,
+            )
+            self._check_window_bounds(start, end, len(x_array))
+
+            with np.errstate(all="ignore"):
+                mean_x_y = window_aggregations.roll_mean(
+                    x_array * y_array, start, end, min_periods
+                )
+                mean_x = window_aggregations.roll_mean(x_array, start, end, min_periods)
+                mean_y = window_aggregations.roll_mean(y_array, start, end, min_periods)
+                count_x_y = window_aggregations.roll_sum(
+                    notna(x_array + y_array).astype(np.float64), start, end, 0
+                )
+                result = (mean_x_y - mean_x * mean_y) * (count_x_y / (count_x_y - ddof))
+            return Series(result, index=x.index, name=x.name, copy=False)
+
+        return self._apply_pairwise(
+            self._selected_obj, other, pairwise, cov_func, numeric_only
+        )
+
+    def corr(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        if self.step is not None:
+            raise NotImplementedError("step not implemented for corr")
+        self._validate_numeric_only("corr", numeric_only)
+
+        from pandas import Series
+
+        def corr_func(x, y):
+            x_array = self._prep_values(x)
+            y_array = self._prep_values(y)
+            window_indexer = self._get_window_indexer()
+            min_periods = (
+                self.min_periods
+                if self.min_periods is not None
+                else window_indexer.window_size
+            )
+            start, end = window_indexer.get_window_bounds(
+                num_values=len(x_array),
+                min_periods=min_periods,
+                center=self.center,
+                closed=self.closed,
+                step=self.step,
+            )
+            self._check_window_bounds(start, end, len(x_array))
+
+            with np.errstate(all="ignore"):
+                mean_x_y = window_aggregations.roll_mean(
+                    x_array * y_array, start, end, min_periods
+                )
+                mean_x = window_aggregations.roll_mean(x_array, start, end, min_periods)
+                mean_y = window_aggregations.roll_mean(y_array, start, end, min_periods)
+                count_x_y = window_aggregations.roll_sum(
+                    notna(x_array + y_array).astype(np.float64), start, end, 0
+                )
+                x_var = window_aggregations.roll_var(
+                    x_array, start, end, min_periods, ddof
+                )
+                y_var = window_aggregations.roll_var(
+                    y_array, start, end, min_periods, ddof
+                )
+                numerator = (mean_x_y - mean_x * mean_y) * (
+                    count_x_y / (count_x_y - ddof)
+                )
+                denominator = (x_var * y_var) ** 0.5
+                result = numerator / denominator
+            return Series(result, index=x.index, name=x.name, copy=False)
+
+        return self._apply_pairwise(
+            self._selected_obj, other, pairwise, corr_func, numeric_only
+        )
+
+
+class Rolling(RollingAndExpandingMixin):
+    _attributes: list[str] = [
+        "window",
+        "min_periods",
+        "center",
+        "win_type",
+        "on",
+        "closed",
+        "step",
+        "method",
+    ]
+
+    def _validate(self) -> None:
+        super()._validate()
+
+        # we allow rolling on a datetimelike index
+        if (
+            self.obj.empty
+            or isinstance(self._on, (DatetimeIndex, TimedeltaIndex, PeriodIndex))
+            or (isinstance(self._on.dtype, ArrowDtype) and self._on.dtype.kind in "mM")
+        ) and isinstance(self.window, (str, BaseOffset, timedelta)):
+            self._validate_datetimelike_monotonic()
+
+            # this will raise ValueError on non-fixed freqs
+            try:
+                freq = to_offset(self.window)
+            except (TypeError, ValueError) as err:
+                raise ValueError(
+                    f"passed window {self.window} is not "
+                    "compatible with a datetimelike index"
+                ) from err
+            if isinstance(self._on, PeriodIndex):
+                # error: Incompatible types in assignment (expression has type
+                # "float", variable has type "Optional[int]")
+                self._win_freq_i8 = freq.nanos / (  # type: ignore[assignment]
+                    self._on.freq.nanos / self._on.freq.n
+                )
+            else:
+                try:
+                    unit = dtype_to_unit(self._on.dtype)  # type: ignore[arg-type]
+                except TypeError:
+                    # if not a datetime dtype, eg for empty dataframes
+                    unit = "ns"
+                self._win_freq_i8 = Timedelta(freq.nanos).as_unit(unit)._value
+
+            # min_periods must be an integer
+            if self.min_periods is None:
+                self.min_periods = 1
+
+            if self.step is not None:
+                raise NotImplementedError(
+                    "step is not supported with frequency windows"
+                )
+
+        elif isinstance(self.window, BaseIndexer):
+            # Passed BaseIndexer subclass should handle all other rolling kwargs
+            pass
+        elif not is_integer(self.window) or self.window < 0:
+            raise ValueError("window must be an integer 0 or greater")
+
+    def _validate_datetimelike_monotonic(self) -> None:
+        """
+        Validate self._on is monotonic (increasing or decreasing) and has
+        no NaT values for frequency windows.
+        """
+        if self._on.hasnans:
+            self._raise_monotonic_error("values must not have NaT")
+        if not (self._on.is_monotonic_increasing or self._on.is_monotonic_decreasing):
+            self._raise_monotonic_error("values must be monotonic")
+
+    def _raise_monotonic_error(self, msg: str):
+        on = self.on
+        if on is None:
+            on = "index"
+        raise ValueError(f"{on} {msg}")
+
+    @doc(
+        _shared_docs["aggregate"],
+        see_also=dedent(
+            """
+        See Also
+        --------
+        Series.rolling : Calling object with Series data.
+        DataFrame.rolling : Calling object with DataFrame data.
+        """
+        ),
+        examples=dedent(
+            """
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        >>> df
+           A  B  C
+        0  1  4  7
+        1  2  5  8
+        2  3  6  9
+
+        >>> df.rolling(2).sum()
+             A     B     C
+        0  NaN   NaN   NaN
+        1  3.0   9.0  15.0
+        2  5.0  11.0  17.0
+
+        >>> df.rolling(2).agg({"A": "sum", "B": "min"})
+             A    B
+        0  NaN  NaN
+        1  3.0  4.0
+        2  5.0  5.0
+        """
+        ),
+        klass="Series/Dataframe",
+        axis="",
+    )
+    def aggregate(self, func=None, *args, **kwargs):
+        return super().aggregate(func, *args, **kwargs)
+
+    agg = aggregate
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([2, 3, np.nan, 10])
+        >>> s.rolling(2).count()
+        0    NaN
+        1    2.0
+        2    1.0
+        3    1.0
+        dtype: float64
+        >>> s.rolling(3).count()
+        0    NaN
+        1    NaN
+        2    2.0
+        3    2.0
+        dtype: float64
+        >>> s.rolling(4).count()
+        0    NaN
+        1    NaN
+        2    NaN
+        3    3.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="count of non NaN observations",
+        agg_method="count",
+    )
+    def count(self, numeric_only: bool = False):
+        return super().count(numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        window_apply_parameters,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 6, 5, 4])
+        >>> ser.rolling(2).apply(lambda s: s.sum() - s.min())
+        0    NaN
+        1    6.0
+        2    6.0
+        3    5.0
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="custom aggregation function",
+        agg_method="apply",
+    )
+    def apply(
+        self,
+        func: Callable[..., Any],
+        raw: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        args: tuple[Any, ...] | None = None,
+        kwargs: dict[str, Any] | None = None,
+    ):
+        return super().apply(
+            func,
+            raw=raw,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+            args=args,
+            kwargs=kwargs,
+        )
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    @final
+    @Substitution(
+        klass="Rolling",
+        examples="""
+    >>> df = pd.DataFrame({'A': [1, 2, 3, 4]},
+    ...                   index=pd.date_range('2012-08-02', periods=4))
+    >>> df
+                A
+    2012-08-02  1
+    2012-08-03  2
+    2012-08-04  3
+    2012-08-05  4
+
+    To get the difference between each rolling 2-day window's maximum and minimum
+    value in one pass, you can do
+
+    >>> df.rolling('2D').pipe(lambda x: x.max() - x.min())
+                  A
+    2012-08-02  0.0
+    2012-08-03  1.0
+    2012-08-04  1.0
+    2012-08-05  1.0""",
+    )
+    @Appender(template_pipe)
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        return super().pipe(func, *args, **kwargs)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([1, 2, 3, 4, 5])
+        >>> s
+        0    1
+        1    2
+        2    3
+        3    4
+        4    5
+        dtype: int64
+
+        >>> s.rolling(3).sum()
+        0     NaN
+        1     NaN
+        2     6.0
+        3     9.0
+        4    12.0
+        dtype: float64
+
+        >>> s.rolling(3, center=True).sum()
+        0     NaN
+        1     6.0
+        2     9.0
+        3    12.0
+        4     NaN
+        dtype: float64
+
+        For DataFrame, each sum is computed column-wise.
+
+        >>> df = pd.DataFrame({{"A": s, "B": s ** 2}})
+        >>> df
+           A   B
+        0  1   1
+        1  2   4
+        2  3   9
+        3  4  16
+        4  5  25
+
+        >>> df.rolling(3).sum()
+              A     B
+        0   NaN   NaN
+        1   NaN   NaN
+        2   6.0  14.0
+        3   9.0  29.0
+        4  12.0  50.0
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="sum",
+        agg_method="sum",
+    )
+    def sum(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().sum(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        dedent(
+            """
+            *args : iterable, optional
+                Positional arguments passed into ``func``.\n
+            """
+        ).replace("\n", "", 1),
+        window_agg_numba_parameters(),
+        dedent(
+            """
+            **kwargs : mapping, optional
+                A dictionary of keyword arguments passed into ``func``.\n
+            """
+        ).replace("\n", "", 1),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 2, 3, 4])
+        >>> ser.rolling(2).max()
+        0    NaN
+        1    2.0
+        2    3.0
+        3    4.0
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="maximum",
+        agg_method="max",
+    )
+    def max(
+        self,
+        numeric_only: bool = False,
+        *args,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+        **kwargs,
+    ):
+        return super().max(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """
+        Performing a rolling minimum with a window size of 3.
+
+        >>> s = pd.Series([4, 3, 5, 2, 6])
+        >>> s.rolling(3).min()
+        0    NaN
+        1    NaN
+        2    3.0
+        3    2.0
+        4    2.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="minimum",
+        agg_method="min",
+    )
+    def min(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().min(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """
+        The below examples will show rolling mean calculations with window sizes of
+        two and three, respectively.
+
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s.rolling(2).mean()
+        0    NaN
+        1    1.5
+        2    2.5
+        3    3.5
+        dtype: float64
+
+        >>> s.rolling(3).mean()
+        0    NaN
+        1    NaN
+        2    2.0
+        3    3.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="mean",
+        agg_method="mean",
+    )
+    def mean(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().mean(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        window_agg_numba_parameters(),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        numba_notes,
+        create_section_header("Examples"),
+        dedent(
+            """
+        Compute the rolling median of a series with a window size of 3.
+
+        >>> s = pd.Series([0, 1, 2, 3, 4])
+        >>> s.rolling(3).median()
+        0    NaN
+        1    NaN
+        2    1.0
+        3    2.0
+        4    3.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="median",
+        agg_method="median",
+    )
+    def median(
+        self,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().median(
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        window_agg_numba_parameters("1.4"),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "numpy.std : Equivalent method for NumPy array.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        The default ``ddof`` of 1 used in :meth:`Series.std` is different
+        than the default ``ddof`` of 0 in :func:`numpy.std`.
+
+        A minimum of one period is required for the rolling calculation.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([5, 5, 6, 7, 5, 5, 5])
+        >>> s.rolling(3).std()
+        0         NaN
+        1         NaN
+        2    0.577350
+        3    1.000000
+        4    1.000000
+        5    1.154701
+        6    0.000000
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="standard deviation",
+        agg_method="std",
+    )
+    def std(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().std(
+            ddof=ddof,
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        window_agg_numba_parameters("1.4"),
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "numpy.var : Equivalent method for NumPy array.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        The default ``ddof`` of 1 used in :meth:`Series.var` is different
+        than the default ``ddof`` of 0 in :func:`numpy.var`.
+
+        A minimum of one period is required for the rolling calculation.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([5, 5, 6, 7, 5, 5, 5])
+        >>> s.rolling(3).var()
+        0         NaN
+        1         NaN
+        2    0.333333
+        3    1.000000
+        4    1.000000
+        5    1.333333
+        6    0.000000
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="variance",
+        agg_method="var",
+    )
+    def var(
+        self,
+        ddof: int = 1,
+        numeric_only: bool = False,
+        engine: Literal["cython", "numba"] | None = None,
+        engine_kwargs: dict[str, bool] | None = None,
+    ):
+        return super().var(
+            ddof=ddof,
+            numeric_only=numeric_only,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "scipy.stats.skew : Third moment of a probability density.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        A minimum of three periods is required for the rolling calculation.\n
+        """
+        ),
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser = pd.Series([1, 5, 2, 7, 15, 6])
+        >>> ser.rolling(3).skew().round(6)
+        0         NaN
+        1         NaN
+        2    1.293343
+        3   -0.585583
+        4    0.670284
+        5    1.652317
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="unbiased skewness",
+        agg_method="skew",
+    )
+    def skew(self, numeric_only: bool = False):
+        return super().skew(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Notes"),
+        "A minimum of one period is required for the calculation.\n\n",
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([0, 1, 2, 3])
+        >>> s.rolling(2, min_periods=1).sem()
+        0         NaN
+        1    0.707107
+        2    0.707107
+        3    0.707107
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="standard error of mean",
+        agg_method="sem",
+    )
+    def sem(self, ddof: int = 1, numeric_only: bool = False):
+        # Raise here so error message says sem instead of std
+        self._validate_numeric_only("sem", numeric_only)
+        return self.std(numeric_only=numeric_only) / (
+            self.count(numeric_only) - ddof
+        ).pow(0.5)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        "scipy.stats.kurtosis : Reference SciPy method.\n",
+        template_see_also,
+        create_section_header("Notes"),
+        "A minimum of four periods is required for the calculation.\n\n",
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show a rolling calculation with a window size of
+        four matching the equivalent function call using `scipy.stats`.
+
+        >>> arr = [1, 2, 3, 4, 999]
+        >>> import scipy.stats
+        >>> print(f"{{scipy.stats.kurtosis(arr[:-1], bias=False):.6f}}")
+        -1.200000
+        >>> print(f"{{scipy.stats.kurtosis(arr[1:], bias=False):.6f}}")
+        3.999946
+        >>> s = pd.Series(arr)
+        >>> s.rolling(4).kurt()
+        0         NaN
+        1         NaN
+        2         NaN
+        3   -1.200000
+        4    3.999946
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="Fisher's definition of kurtosis without bias",
+        agg_method="kurt",
+    )
+    def kurt(self, numeric_only: bool = False):
+        return super().kurt(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        GroupBy.first : Similar method for GroupBy objects.
+        Rolling.last : Method to get the last element in each window.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show a rolling calculation with a window size of
+        three.
+
+        >>> s = pd.Series(range(5))
+        >>> s.rolling(3).first()
+        0         NaN
+        1         NaN
+        2         0.0
+        3         1.0
+        4         2.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="First (left-most) element of the window",
+        agg_method="first",
+    )
+    def first(self, numeric_only: bool = False):
+        return super().first(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        GroupBy.last : Similar method for GroupBy objects.
+        Rolling.first : Method to get the first element in each window.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        The example below will show a rolling calculation with a window size of
+        three.
+
+        >>> s = pd.Series(range(5))
+        >>> s.rolling(3).last()
+        0         NaN
+        1         NaN
+        2         2.0
+        3         3.0
+        4         4.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="Last (right-most) element of the window",
+        agg_method="last",
+    )
+    def last(self, numeric_only: bool = False):
+        return super().last(numeric_only=numeric_only)
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        q : float
+            Quantile to compute. 0 <= quantile <= 1.
+
+            .. deprecated:: 2.1.0
+                This was renamed from 'quantile' to 'q' in version 2.1.0.
+        interpolation : {{'linear', 'lower', 'higher', 'midpoint', 'nearest'}}
+            This optional parameter specifies the interpolation method to use,
+            when the desired quantile lies between two data points `i` and `j`:
+
+                * linear: `i + (j - i) * fraction`, where `fraction` is the
+                  fractional part of the index surrounded by `i` and `j`.
+                * lower: `i`.
+                * higher: `j`.
+                * nearest: `i` or `j` whichever is nearest.
+                * midpoint: (`i` + `j`) / 2.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([1, 2, 3, 4])
+        >>> s.rolling(2).quantile(.4, interpolation='lower')
+        0    NaN
+        1    1.0
+        2    2.0
+        3    3.0
+        dtype: float64
+
+        >>> s.rolling(2).quantile(.4, interpolation='midpoint')
+        0    NaN
+        1    1.5
+        2    2.5
+        3    3.5
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="quantile",
+        agg_method="quantile",
+    )
+    def quantile(
+        self,
+        q: float,
+        interpolation: QuantileInterpolation = "linear",
+        numeric_only: bool = False,
+    ):
+        return super().quantile(
+            q=q,
+            interpolation=interpolation,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        ".. versionadded:: 1.4.0 \n\n",
+        create_section_header("Parameters"),
+        dedent(
+            """
+        method : {{'average', 'min', 'max'}}, default 'average'
+            How to rank the group of records that have the same value (i.e. ties):
+
+            * average: average rank of the group
+            * min: lowest rank in the group
+            * max: highest rank in the group
+
+        ascending : bool, default True
+            Whether or not the elements should be ranked in ascending order.
+        pct : bool, default False
+            Whether or not to display the returned rankings in percentile
+            form.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """
+        >>> s = pd.Series([1, 4, 2, 3, 5, 3])
+        >>> s.rolling(3).rank()
+        0    NaN
+        1    NaN
+        2    2.0
+        3    2.0
+        4    3.0
+        5    1.5
+        dtype: float64
+
+        >>> s.rolling(3).rank(method="max")
+        0    NaN
+        1    NaN
+        2    2.0
+        3    2.0
+        4    3.0
+        5    2.0
+        dtype: float64
+
+        >>> s.rolling(3).rank(method="min")
+        0    NaN
+        1    NaN
+        2    2.0
+        3    2.0
+        4    3.0
+        5    1.0
+        dtype: float64
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="rank",
+        agg_method="rank",
+    )
+    def rank(
+        self,
+        method: WindowingRankType = "average",
+        ascending: bool = True,
+        pct: bool = False,
+        numeric_only: bool = False,
+    ):
+        return super().rank(
+            method=method,
+            ascending=ascending,
+            pct=pct,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        other : Series or DataFrame, optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndexed DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        template_see_also,
+        create_section_header("Examples"),
+        dedent(
+            """\
+        >>> ser1 = pd.Series([1, 2, 3, 4])
+        >>> ser2 = pd.Series([1, 4, 5, 8])
+        >>> ser1.rolling(2).cov(ser2)
+        0    NaN
+        1    1.5
+        2    0.5
+        3    1.5
+        dtype: float64
+        """
+        ),
+        window_method="rolling",
+        aggregation_description="sample covariance",
+        agg_method="cov",
+    )
+    def cov(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        return super().cov(
+            other=other,
+            pairwise=pairwise,
+            ddof=ddof,
+            numeric_only=numeric_only,
+        )
+
+    @doc(
+        template_header,
+        create_section_header("Parameters"),
+        dedent(
+            """
+        other : Series or DataFrame, optional
+            If not supplied then will default to self and produce pairwise
+            output.
+        pairwise : bool, default None
+            If False then only matching columns between self and other will be
+            used and the output will be a DataFrame.
+            If True then all pairwise combinations will be calculated and the
+            output will be a MultiIndexed DataFrame in the case of DataFrame
+            inputs. In the case of missing elements, only complete pairwise
+            observations will be used.
+        ddof : int, default 1
+            Delta Degrees of Freedom.  The divisor used in calculations
+            is ``N - ddof``, where ``N`` represents the number of elements.
+        """
+        ).replace("\n", "", 1),
+        kwargs_numeric_only,
+        create_section_header("Returns"),
+        template_returns,
+        create_section_header("See Also"),
+        dedent(
+            """
+        cov : Similar method to calculate covariance.
+        numpy.corrcoef : NumPy Pearson's correlation calculation.
+        """
+        ).replace("\n", "", 1),
+        template_see_also,
+        create_section_header("Notes"),
+        dedent(
+            """
+        This function uses Pearson's definition of correlation
+        (https://en.wikipedia.org/wiki/Pearson_correlation_coefficient).
+
+        When `other` is not specified, the output will be self correlation (e.g.
+        all 1's), except for :class:`~pandas.DataFrame` inputs with `pairwise`
+        set to `True`.
+
+        Function will return ``NaN`` for correlations of equal valued sequences;
+        this is the result of a 0/0 division error.
+
+        When `pairwise` is set to `False`, only matching columns between `self` and
+        `other` will be used.
+
+        When `pairwise` is set to `True`, the output will be a MultiIndex DataFrame
+        with the original index on the first level, and the `other` DataFrame
+        columns on the second level.
+
+        In the case of missing elements, only complete pairwise observations
+        will be used.\n
+        """
+        ).replace("\n", "", 1),
+        create_section_header("Examples"),
+        dedent(
+            """
+        The below example shows a rolling calculation with a window size of
+        four matching the equivalent function call using :meth:`numpy.corrcoef`.
+
+        >>> v1 = [3, 3, 3, 5, 8]
+        >>> v2 = [3, 4, 4, 4, 8]
+        >>> np.corrcoef(v1[:-1], v2[:-1])
+        array([[1.        , 0.33333333],
+               [0.33333333, 1.        ]])
+        >>> np.corrcoef(v1[1:], v2[1:])
+        array([[1.       , 0.9169493],
+               [0.9169493, 1.       ]])
+        >>> s1 = pd.Series(v1)
+        >>> s2 = pd.Series(v2)
+        >>> s1.rolling(4).corr(s2)
+        0         NaN
+        1         NaN
+        2         NaN
+        3    0.333333
+        4    0.916949
+        dtype: float64
+
+        The below example shows a similar rolling calculation on a
+        DataFrame using the pairwise option.
+
+        >>> matrix = np.array([[51., 35.],
+        ...                    [49., 30.],
+        ...                    [47., 32.],
+        ...                    [46., 31.],
+        ...                    [50., 36.]])
+        >>> np.corrcoef(matrix[:-1, 0], matrix[:-1, 1])
+        array([[1.       , 0.6263001],
+               [0.6263001, 1.       ]])
+        >>> np.corrcoef(matrix[1:, 0], matrix[1:, 1])
+        array([[1.        , 0.55536811],
+               [0.55536811, 1.        ]])
+        >>> df = pd.DataFrame(matrix, columns=['X', 'Y'])
+        >>> df
+              X     Y
+        0  51.0  35.0
+        1  49.0  30.0
+        2  47.0  32.0
+        3  46.0  31.0
+        4  50.0  36.0
+        >>> df.rolling(4).corr(pairwise=True)
+                    X         Y
+        0 X       NaN       NaN
+          Y       NaN       NaN
+        1 X       NaN       NaN
+          Y       NaN       NaN
+        2 X       NaN       NaN
+          Y       NaN       NaN
+        3 X  1.000000  0.626300
+          Y  0.626300  1.000000
+        4 X  1.000000  0.555368
+          Y  0.555368  1.000000
+        """
+        ).replace("\n", "", 1),
+        window_method="rolling",
+        aggregation_description="correlation",
+        agg_method="corr",
+    )
+    def corr(
+        self,
+        other: DataFrame | Series | None = None,
+        pairwise: bool | None = None,
+        ddof: int = 1,
+        numeric_only: bool = False,
+    ):
+        return super().corr(
+            other=other,
+            pairwise=pairwise,
+            ddof=ddof,
+            numeric_only=numeric_only,
+        )
+
+
+Rolling.__doc__ = Window.__doc__
+
+
+class RollingGroupby(BaseWindowGroupby, Rolling):
+    """
+    Provide a rolling groupby implementation.
+    """
+
+    _attributes = Rolling._attributes + BaseWindowGroupby._attributes
+
+    def _get_window_indexer(self) -> GroupbyIndexer:
+        """
+        Return an indexer class that will compute the window start and end bounds
+
+        Returns
+        -------
+        GroupbyIndexer
+        """
+        rolling_indexer: type[BaseIndexer]
+        indexer_kwargs: dict[str, Any] | None = None
+        index_array = self._index_array
+        if isinstance(self.window, BaseIndexer):
+            rolling_indexer = type(self.window)
+            indexer_kwargs = self.window.__dict__.copy()
+            assert isinstance(indexer_kwargs, dict)  # for mypy
+            # We'll be using the index of each group later
+            indexer_kwargs.pop("index_array", None)
+            window = self.window
+        elif self._win_freq_i8 is not None:
+            rolling_indexer = VariableWindowIndexer
+            # error: Incompatible types in assignment (expression has type
+            # "int", variable has type "BaseIndexer")
+            window = self._win_freq_i8  # type: ignore[assignment]
+        else:
+            rolling_indexer = FixedWindowIndexer
+            window = self.window
+        window_indexer = GroupbyIndexer(
+            index_array=index_array,
+            window_size=window,
+            groupby_indices=self._grouper.indices,
+            window_indexer=rolling_indexer,
+            indexer_kwargs=indexer_kwargs,
+        )
+        return window_indexer
+
+    def _validate_datetimelike_monotonic(self) -> None:
+        """
+        Validate that each group in self._on is monotonic
+        """
+        # GH 46061
+        if self._on.hasnans:
+            self._raise_monotonic_error("values must not have NaT")
+        for group_indices in self._grouper.indices.values():
+            group_on = self._on.take(group_indices)
+            if not (
+                group_on.is_monotonic_increasing or group_on.is_monotonic_decreasing
+            ):
+                on = "index" if self.on is None else self.on
+                raise ValueError(
+                    f"Each group within {on} must be monotonic. "
+                    f"Sort the values in {on} first."
+                )
diff --git a/pandas/pandas/errors/__init__.py b/pandas/pandas/errors/__init__.py
new file mode 100644
index 0000000000000..2b5bc450e41d6
--- /dev/null
+++ b/pandas/pandas/errors/__init__.py
@@ -0,0 +1,947 @@
+"""
+Expose public exceptions & warnings
+"""
+
+from __future__ import annotations
+
+import ctypes
+
+from pandas._config.config import OptionError
+
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+)
+
+from pandas.util.version import InvalidVersion
+
+
+class IntCastingNaNError(ValueError):
+    """
+    Exception raised when converting (``astype``) an array with NaN to an integer type.
+
+    This error occurs when attempting to cast a data structure containing non-finite
+    values (such as NaN or infinity) to an integer data type. Integer types do not
+    support non-finite values, so such conversions are explicitly disallowed to
+    prevent silent data corruption or unexpected behavior.
+
+    See Also
+    --------
+    DataFrame.astype : Method to cast a pandas DataFrame object to a specified dtype.
+    Series.astype : Method to cast a pandas Series object to a specified dtype.
+
+    Examples
+    --------
+    >>> pd.DataFrame(np.array([[1, np.nan], [2, 3]]), dtype="i8")
+    Traceback (most recent call last):
+    IntCastingNaNError: Cannot convert non-finite values (NA or inf) to integer
+    """
+
+
+class NullFrequencyError(ValueError):
+    """
+    Exception raised when a ``freq`` cannot be null.
+
+    Particularly ``DatetimeIndex.shift``, ``TimedeltaIndex.shift``,
+    ``PeriodIndex.shift``.
+
+    See Also
+    --------
+    Index.shift : Shift values of Index.
+    Series.shift : Shift values of Series.
+
+    Examples
+    --------
+    >>> df = pd.DatetimeIndex(["2011-01-01 10:00", "2011-01-01"], freq=None)
+    >>> df.shift(2)
+    Traceback (most recent call last):
+    NullFrequencyError: Cannot shift with no freq
+    """
+
+
+class PerformanceWarning(Warning):
+    """
+    Warning raised when there is a possible performance impact.
+
+    See Also
+    --------
+    DataFrame.set_index : Set the DataFrame index using existing columns.
+    DataFrame.loc : Access a group of rows and columns by label(s) \
+    or a boolean array.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {"jim": [0, 0, 1, 1], "joe": ["x", "x", "z", "y"], "jolie": [1, 2, 3, 4]}
+    ... )
+    >>> df = df.set_index(["jim", "joe"])
+    >>> df
+              jolie
+    jim  joe
+    0    x    1
+         x    2
+    1    z    3
+         y    4
+    >>> df.loc[(1, "z")]  # doctest: +SKIP
+    # PerformanceWarning: indexing past lexsort depth may impact performance.
+    df.loc[(1, 'z')]
+              jolie
+    jim  joe
+    1    z        3
+    """
+
+
+class UnsupportedFunctionCall(ValueError):
+    """
+    Exception raised when attempting to call a unsupported numpy function.
+
+    For example, ``np.cumsum(groupby_object)``.
+
+    See Also
+    --------
+    DataFrame.groupby : Group DataFrame using a mapper or by a Series of columns.
+    Series.groupby : Group Series using a mapper or by a Series of columns.
+    core.groupby.GroupBy.cumsum : Compute cumulative sum for each group.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {"A": [0, 0, 1, 1], "B": ["x", "x", "z", "y"], "C": [1, 2, 3, 4]}
+    ... )
+    >>> np.cumsum(df.groupby(["A"]))
+    Traceback (most recent call last):
+    UnsupportedFunctionCall: numpy operations are not valid with groupby.
+    Use .groupby(...).cumsum() instead
+    """
+
+
+class UnsortedIndexError(KeyError):
+    """
+    Error raised when slicing a MultiIndex which has not been lexsorted.
+
+    Subclass of `KeyError`.
+
+    See Also
+    --------
+    DataFrame.sort_index : Sort a DataFrame by its index.
+    DataFrame.set_index : Set the DataFrame index using existing columns.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "cat": [0, 0, 1, 1],
+    ...         "color": ["white", "white", "brown", "black"],
+    ...         "lives": [4, 4, 3, 7],
+    ...     },
+    ... )
+    >>> df = df.set_index(["cat", "color"])
+    >>> df
+                lives
+    cat  color
+    0    white    4
+         white    4
+    1    brown    3
+         black    7
+    >>> df.loc[(0, "black") : (1, "white")]
+    Traceback (most recent call last):
+    UnsortedIndexError: 'Key length (2) was greater
+    than MultiIndex lexsort depth (1)'
+    """
+
+
+class ParserError(ValueError):
+    """
+    Exception that is raised by an error encountered in parsing file contents.
+
+    This is a generic error raised for errors encountered when functions like
+    `read_csv` or `read_html` are parsing contents of a file.
+
+    See Also
+    --------
+    read_csv : Read CSV (comma-separated) file into a DataFrame.
+    read_html : Read HTML table into a DataFrame.
+
+    Examples
+    --------
+    >>> data = '''a,b,c
+    ... cat,foo,bar
+    ... dog,foo,"baz'''
+    >>> from io import StringIO
+    >>> pd.read_csv(StringIO(data), skipfooter=1, engine="python")
+    Traceback (most recent call last):
+    ParserError: ',' expected after '"'. Error could possibly be due
+    to parsing errors in the skipped footer rows
+    """
+
+
+class DtypeWarning(Warning):
+    """
+    Warning raised when reading different dtypes in a column from a file.
+
+    Raised for a dtype incompatibility. This can happen whenever `read_csv`
+    or `read_table` encounter non-uniform dtypes in a column(s) of a given
+    CSV file.
+
+    See Also
+    --------
+    read_csv : Read CSV (comma-separated) file into a DataFrame.
+    read_table : Read general delimited file into a DataFrame.
+
+    Notes
+    -----
+    This warning is issued when dealing with larger files because the dtype
+    checking happens per chunk read.
+
+    Despite the warning, the CSV file is read with mixed types in a single
+    column which will be an object type. See the examples below to better
+    understand this issue.
+
+    Examples
+    --------
+    This example creates and reads a large CSV file with a column that contains
+    `int` and `str`.
+
+    >>> df = pd.DataFrame(
+    ...     {
+    ...         "a": (["1"] * 100000 + ["X"] * 100000 + ["1"] * 100000),
+    ...         "b": ["b"] * 300000,
+    ...     }
+    ... )  # doctest: +SKIP
+    >>> df.to_csv("test.csv", index=False)  # doctest: +SKIP
+    >>> df2 = pd.read_csv("test.csv")  # doctest: +SKIP
+    ... # DtypeWarning: Columns (0: a) have mixed types
+
+    Important to notice that ``df2`` will contain both `str` and `int` for the
+    same input, '1'.
+
+    >>> df2.iloc[262140, 0]  # doctest: +SKIP
+    '1'
+    >>> type(df2.iloc[262140, 0])  # doctest: +SKIP
+    <class 'str'>
+    >>> df2.iloc[262150, 0]  # doctest: +SKIP
+    1
+    >>> type(df2.iloc[262150, 0])  # doctest: +SKIP
+    <class 'int'>
+
+    One way to solve this issue is using the `dtype` parameter in the
+    `read_csv` and `read_table` functions to explicit the conversion:
+
+    >>> df2 = pd.read_csv("test.csv", sep=",", dtype={"a": str})  # doctest: +SKIP
+
+    No warning was issued.
+    """
+
+
+class EmptyDataError(ValueError):
+    """
+    Exception raised in ``pd.read_csv`` when empty data or header is encountered.
+
+    This error is typically encountered when attempting to read an empty file or
+    an invalid file where no data or headers are present.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (CSV) file into DataFrame.
+    errors.ParserError : Exception that is raised by an error encountered in parsing
+        file contents.
+    errors.DtypeWarning : Warning raised when reading different dtypes in a column
+        from a file.
+
+    Examples
+    --------
+    >>> from io import StringIO
+    >>> empty = StringIO()
+    >>> pd.read_csv(empty)
+    Traceback (most recent call last):
+    EmptyDataError: No columns to parse from file
+    """
+
+
+class ParserWarning(Warning):
+    """
+    Warning raised when reading a file that doesn't use the default 'c' parser.
+
+    Raised by `pd.read_csv` and `pd.read_table` when it is necessary to change
+    parsers, generally from the default 'c' parser to 'python'.
+
+    It happens due to a lack of support or functionality for parsing a
+    particular attribute of a CSV file with the requested engine.
+
+    Currently, 'c' unsupported options include the following parameters:
+
+    1. `sep` other than a single character (e.g. regex separators)
+    2. `skipfooter` higher than 0
+
+    The warning can be avoided by adding `engine='python'` as a parameter in
+    `pd.read_csv` and `pd.read_table` methods.
+
+    See Also
+    --------
+    pd.read_csv : Read CSV (comma-separated) file into DataFrame.
+    pd.read_table : Read general delimited file into DataFrame.
+
+    Examples
+    --------
+    Using a `sep` in `pd.read_csv` other than a single character:
+
+    >>> import io
+    >>> csv = '''a;b;c
+    ...           1;1,8
+    ...           1;2,1'''
+    >>> df = pd.read_csv(io.StringIO(csv), sep="[;,]")  # doctest: +SKIP
+    ... # ParserWarning: Falling back to the 'python' engine...
+
+    Adding `engine='python'` to `pd.read_csv` removes the Warning:
+
+    >>> df = pd.read_csv(io.StringIO(csv), sep="[;,]", engine="python")
+    """
+
+
+class MergeError(ValueError):
+    """
+    Exception raised when merging data.
+
+    Subclass of ``ValueError``.
+
+    See Also
+    --------
+    DataFrame.join : For joining DataFrames on their indexes.
+    merge : For merging two DataFrames on a common set of keys.
+
+    Examples
+    --------
+    >>> left = pd.DataFrame(
+    ...     {"a": ["a", "b", "b", "d"], "b": ["cat", "dog", "weasel", "horse"]},
+    ...     index=range(4),
+    ... )
+    >>> right = pd.DataFrame(
+    ...     {"a": ["a", "b", "c", "d"], "c": ["meow", "bark", "chirp", "nay"]},
+    ...     index=range(4),
+    ... ).set_index("a")
+    >>> left.join(
+    ...     right,
+    ...     on="a",
+    ...     validate="one_to_one",
+    ... )
+    Traceback (most recent call last):
+    MergeError: Merge keys are not unique in left dataset; not a one-to-one merge
+    """
+
+
+class AbstractMethodError(NotImplementedError):
+    """
+    Raise this error instead of NotImplementedError for abstract methods.
+
+    The `AbstractMethodError` is designed for use in classes that follow an abstract
+    base class pattern. By raising this error in the method, it ensures that a subclass
+    must implement the method to provide specific functionality. This is useful in a
+    framework or library where certain methods must be implemented by the user to
+    ensure correct behavior.
+
+    Parameters
+    ----------
+    class_instance : object
+        The instance of the class where the abstract method is being called.
+    methodtype : str, default "method"
+        A string indicating the type of method that is abstract.
+        Must be one of {"method", "classmethod", "staticmethod", "property"}.
+
+    See Also
+    --------
+    api.extensions.ExtensionArray
+        An example of a pandas extension mechanism that requires implementing
+        specific abstract methods.
+    NotImplementedError
+        A built-in exception that can also be used for abstract methods but lacks
+        the specificity of `AbstractMethodError` in indicating the need for subclass
+        implementation.
+
+    Examples
+    --------
+    >>> class Foo:
+    ...     @classmethod
+    ...     def classmethod(cls):
+    ...         raise pd.errors.AbstractMethodError(cls, methodtype="classmethod")
+    ...
+    ...     def method(self):
+    ...         raise pd.errors.AbstractMethodError(self)
+    >>> test = Foo.classmethod()
+    Traceback (most recent call last):
+    AbstractMethodError: This classmethod must be defined in the concrete class Foo
+
+    >>> test2 = Foo().method()
+    Traceback (most recent call last):
+    AbstractMethodError: This classmethod must be defined in the concrete class Foo
+    """
+
+    def __init__(self, class_instance, methodtype: str = "method") -> None:
+        types = {"method", "classmethod", "staticmethod", "property"}
+        if methodtype not in types:
+            raise ValueError(
+                f"methodtype must be one of {methodtype}, got {types} instead."
+            )
+        self.methodtype = methodtype
+        self.class_instance = class_instance
+
+    def __str__(self) -> str:
+        if self.methodtype == "classmethod":
+            name = self.class_instance.__name__
+        else:
+            name = type(self.class_instance).__name__
+        return f"This {self.methodtype} must be defined in the concrete class {name}"
+
+
+class NumbaUtilError(Exception):
+    """
+    Error raised for unsupported Numba engine routines.
+
+    See Also
+    --------
+    DataFrame.groupby : Group DataFrame using a mapper or by a Series of columns.
+    Series.groupby : Group Series using a mapper or by a Series of columns.
+    DataFrame.agg : Aggregate using one or more operations over the specified axis.
+    Series.agg : Aggregate using one or more operations over the specified axis.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     {"key": ["a", "a", "b", "b"], "data": [1, 2, 3, 4]}, columns=["key", "data"]
+    ... )
+    >>> def incorrect_function(x):
+    ...     return sum(x) * 2.7
+    >>> df.groupby("key").agg(incorrect_function, engine="numba")
+    Traceback (most recent call last):
+    NumbaUtilError: The first 2 arguments to incorrect_function
+    must be ['values', 'index']
+    """
+
+
+class DuplicateLabelError(ValueError):
+    """
+    Error raised when an operation would introduce duplicate labels.
+
+    This error is typically encountered when performing operations on objects
+    with `allows_duplicate_labels=False` and the operation would result in
+    duplicate labels in the index. Duplicate labels can lead to ambiguities
+    in indexing and reduce data integrity.
+
+    See Also
+    --------
+    Series.set_flags : Return a new ``Series`` object with updated flags.
+    DataFrame.set_flags : Return a new ``DataFrame`` object with updated flags.
+    Series.reindex : Conform ``Series`` object to new index with optional filling logic.
+    DataFrame.reindex : Conform ``DataFrame`` object to new index with optional filling
+        logic.
+
+    Examples
+    --------
+    >>> s = pd.Series([0, 1, 2], index=["a", "b", "c"]).set_flags(
+    ...     allows_duplicate_labels=False
+    ... )
+    >>> s.reindex(["a", "a", "b"])
+    Traceback (most recent call last):
+       ...
+    DuplicateLabelError: Index has duplicates.
+          positions
+    label
+    a        [0, 1]
+    """
+
+
+class InvalidIndexError(Exception):
+    """
+    Exception raised when attempting to use an invalid index key.
+
+    This exception is triggered when a user attempts to access or manipulate
+    data in a pandas DataFrame or Series using an index key that is not valid
+    for the given object. This may occur in cases such as using a malformed
+    slice, a mismatched key for a ``MultiIndex``, or attempting to access an index
+    element that does not exist.
+
+    See Also
+    --------
+    MultiIndex : A multi-level, or hierarchical, index object for pandas objects.
+
+    Examples
+    --------
+    >>> idx = pd.MultiIndex.from_product([["x", "y"], [0, 1]])
+    >>> df = pd.DataFrame([[1, 1, 2, 2], [3, 3, 4, 4]], columns=idx)
+    >>> df
+        x       y
+        0   1   0   1
+    0   1   1   2   2
+    1   3   3   4   4
+    >>> df[:, 0]
+    Traceback (most recent call last):
+    InvalidIndexError: (slice(None, None, None), 0)
+    """
+
+
+class DataError(Exception):
+    """
+    Exception raised when performing an operation on non-numerical data.
+
+    For example, calling ``ohlc`` on a non-numerical column or a function
+    on a rolling window.
+
+    See Also
+    --------
+    Series.rolling : Provide rolling window calculations on Series object.
+    DataFrame.rolling : Provide rolling window calculations on DataFrame object.
+
+    Examples
+    --------
+    >>> ser = pd.Series(["a", "b", "c"])
+    >>> ser.rolling(2).sum()
+    Traceback (most recent call last):
+    DataError: No numeric types to aggregate
+    """
+
+
+class SpecificationError(Exception):
+    """
+    Exception raised by ``agg`` when the functions are ill-specified.
+
+    The exception raised in two scenarios.
+
+    The first way is calling ``agg`` on a
+    Dataframe or Series using a nested renamer (dict-of-dict).
+
+    The second way is calling ``agg`` on a Dataframe with duplicated functions
+    names without assigning column name.
+
+    See Also
+    --------
+    DataFrame.agg : Aggregate using one or more operations over the specified axis.
+    Series.agg : Aggregate using one or more operations over the specified axis.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"A": [1, 1, 1, 2, 2], "B": range(5), "C": range(5)})
+    >>> df.groupby("A").B.agg({"foo": "count"})  # doctest: +SKIP
+    ... # SpecificationError: nested renamer is not supported
+
+    >>> df.groupby("A").agg({"B": {"foo": ["sum", "max"]}})  # doctest: +SKIP
+    ... # SpecificationError: nested renamer is not supported
+
+    >>> df.groupby("A").agg(["min", "min"])  # doctest: +SKIP
+    ... # SpecificationError: nested renamer is not supported
+    """
+
+
+class ChainedAssignmentError(Warning):
+    """
+    Warning raised when trying to set using chained assignment.
+
+    When the ``mode.copy_on_write`` option is enabled, chained assignment can
+    never work. In such a situation, we are always setting into a temporary
+    object that is the result of an indexing operation (getitem), which under
+    Copy-on-Write always behaves as a copy. Thus, assigning through a chain
+    can never update the original Series or DataFrame.
+
+    For more information on Copy-on-Write,
+    see :ref:`the user guide<copy_on_write>`.
+
+    See Also
+    --------
+    options.mode.copy_on_write : Global setting for enabling or disabling
+        Copy-on-Write behavior.
+
+    Examples
+    --------
+    >>> pd.options.mode.copy_on_write = True
+    >>> df = pd.DataFrame({"A": [1, 1, 1, 2, 2]}, columns=["A"])
+    >>> df["A"][0:3] = 10  # doctest: +SKIP
+    ... # ChainedAssignmentError: ...
+    >>> pd.options.mode.copy_on_write = False
+    """
+
+
+class NumExprClobberingError(NameError):
+    """
+    Exception raised when trying to use a built-in numexpr name as a variable name.
+
+    ``eval`` or ``query`` will throw the error if the engine is set
+    to 'numexpr'. 'numexpr' is the default engine value for these methods if the
+    numexpr package is installed.
+
+    See Also
+    --------
+    eval : Evaluate a Python expression as a string using various backends.
+    DataFrame.query : Query the columns of a DataFrame with a boolean expression.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"abs": [1, 1, 1]})
+    >>> df.query("abs > 2")  # doctest: +SKIP
+    ... # NumExprClobberingError: Variables in expression "(abs) > (2)" overlap...
+    >>> sin, a = 1, 2
+    >>> pd.eval("sin + a", engine="numexpr")  # doctest: +SKIP
+    ... # NumExprClobberingError: Variables in expression "(sin) + (a)" overlap...
+    """
+
+
+class UndefinedVariableError(NameError):
+    """
+    Exception raised by ``query`` or ``eval`` when using an undefined variable name.
+
+    It will also specify whether the undefined variable is local or not.
+
+    Parameters
+    ----------
+    name : str
+        The name of the undefined variable.
+    is_local : bool or None, optional
+        Indicates whether the undefined variable is considered a local variable.
+        If ``True``, the error message specifies it as a local variable.
+        If ``False`` or ``None``, the variable is treated as a non-local name.
+
+    See Also
+    --------
+    DataFrame.query : Query the columns of a DataFrame with a boolean expression.
+    DataFrame.eval : Evaluate a string describing operations on DataFrame columns.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"A": [1, 1, 1]})
+    >>> df.query("A > x")  # doctest: +SKIP
+    ... # UndefinedVariableError: name 'x' is not defined
+    >>> df.query("A > @y")  # doctest: +SKIP
+    ... # UndefinedVariableError: local variable 'y' is not defined
+    >>> pd.eval("x + 1")  # doctest: +SKIP
+    ... # UndefinedVariableError: name 'x' is not defined
+    """
+
+    def __init__(self, name: str, is_local: bool | None = None) -> None:
+        base_msg = f"{name!r} is not defined"
+        if is_local:
+            msg = f"local variable {base_msg}"
+        else:
+            msg = f"name {base_msg}"
+        super().__init__(msg)
+
+
+class IndexingError(Exception):
+    """
+    Exception is raised when trying to index and there is a mismatch in dimensions.
+
+    Raised by properties like :attr:`.pandas.DataFrame.iloc` when
+    an indexer is out of bounds or :attr:`.pandas.DataFrame.loc` when its index is
+    unalignable to the frame index.
+
+    See Also
+    --------
+    DataFrame.iloc : Purely integer-location based indexing for \
+    selection by position.
+    DataFrame.loc : Access a group of rows and columns by label(s) \
+    or a boolean array.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"A": [1, 1, 1]})
+    >>> df.loc[..., ..., "A"]  # doctest: +SKIP
+    ... # IndexingError: indexer may only contain one '...' entry
+    >>> df = pd.DataFrame({"A": [1, 1, 1]})
+    >>> df.loc[1, ..., ...]  # doctest: +SKIP
+    ... # IndexingError: Too many indexers
+    >>> df[pd.Series([True], dtype=bool)]  # doctest: +SKIP
+    ... # IndexingError: Unalignable boolean Series provided as indexer...
+    >>> s = pd.Series(range(2), index=pd.MultiIndex.from_product([["a", "b"], ["c"]]))
+    >>> s.loc["a", "c", "d"]  # doctest: +SKIP
+    ... # IndexingError: Too many indexers
+    """
+
+
+class PyperclipException(RuntimeError):
+    """
+    Exception raised when clipboard functionality is unsupported.
+
+    Raised by ``to_clipboard()`` and ``read_clipboard()``.
+    """
+
+
+class PyperclipWindowsException(PyperclipException):
+    """
+    Exception raised when clipboard functionality is unsupported by Windows.
+
+    Access to the clipboard handle would be denied due to some other
+    window process is accessing it.
+    """
+
+    def __init__(self, message: str) -> None:
+        # attr only exists on Windows, so typing fails on other platforms
+        message += f" ({ctypes.WinError()})"  # type: ignore[attr-defined]
+        super().__init__(message)
+
+
+class CSSWarning(UserWarning):
+    """
+    Warning is raised when converting css styling fails.
+
+    This can be due to the styling not having an equivalent value or because the
+    styling isn't properly formatted.
+
+    See Also
+    --------
+    DataFrame.style : Returns a Styler object for applying CSS-like styles.
+    io.formats.style.Styler : Helps style a DataFrame or Series according to the
+        data with HTML and CSS.
+    io.formats.style.Styler.to_excel : Export styled DataFrame to Excel.
+    io.formats.style.Styler.to_html : Export styled DataFrame to HTML.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"A": [1, 1, 1]})
+    >>> df.style.map(lambda x: "background-color: blueGreenRed;").to_excel(
+    ...     "styled.xlsx"
+    ... )  # doctest: +SKIP
+    CSSWarning: Unhandled color format: 'blueGreenRed'
+    >>> df.style.map(lambda x: "border: 1px solid red red;").to_excel(
+    ...     "styled.xlsx"
+    ... )  # doctest: +SKIP
+    CSSWarning: Unhandled color format: 'blueGreenRed'
+    """
+
+
+class PossibleDataLossError(Exception):
+    """
+    Exception raised when trying to open a HDFStore file when already opened.
+
+    This error is triggered when there is a potential risk of data loss due to
+    conflicting operations on an HDFStore file. It serves to prevent unintended
+    overwrites or data corruption by enforcing exclusive access to the file.
+
+    See Also
+    --------
+    HDFStore : Dict-like IO interface for storing pandas objects in PyTables.
+    HDFStore.open : Open an HDFStore file in the specified mode.
+
+    Examples
+    --------
+    >>> store = pd.HDFStore("my-store", "a")  # doctest: +SKIP
+    >>> store.open("w")  # doctest: +SKIP
+    """
+
+
+class ClosedFileError(Exception):
+    """
+    Exception is raised when trying to perform an operation on a closed HDFStore file.
+
+    ``ClosedFileError`` is specific to operations on ``HDFStore`` objects. Once an
+    HDFStore is closed, its resources are no longer available, and any further attempt
+    to access data or perform file operations will raise this exception.
+
+    See Also
+    --------
+    HDFStore.close : Closes the PyTables file handle.
+    HDFStore.open : Opens the file in the specified mode.
+    HDFStore.is_open : Returns a boolean indicating whether the file is open.
+
+    Examples
+    --------
+    >>> store = pd.HDFStore("my-store", "a")  # doctest: +SKIP
+    >>> store.close()  # doctest: +SKIP
+    >>> store.keys()  # doctest: +SKIP
+    ... # ClosedFileError: my-store file is not open!
+    """
+
+
+class IncompatibilityWarning(Warning):
+    """
+    Warning raised when trying to use where criteria on an incompatible HDF5 file.
+    """
+
+
+class AttributeConflictWarning(Warning):
+    """
+    Warning raised when index attributes conflict when using HDFStore.
+
+    Occurs when attempting to append an index with a different
+    name than the existing index on an HDFStore or attempting to append an index with a
+    different frequency than the existing index on an HDFStore.
+
+    See Also
+    --------
+    HDFStore : Dict-like IO interface for storing pandas objects in PyTables.
+    DataFrame.to_hdf : Write the contained data to an HDF5 file using HDFStore.
+    read_hdf : Read from an HDF5 file into a DataFrame.
+
+    Examples
+    --------
+    >>> idx1 = pd.Index(["a", "b"], name="name1")
+    >>> df1 = pd.DataFrame([[1, 2], [3, 4]], index=idx1)
+    >>> df1.to_hdf("file", "data", "w", append=True)  # doctest: +SKIP
+    >>> idx2 = pd.Index(["c", "d"], name="name2")
+    >>> df2 = pd.DataFrame([[5, 6], [7, 8]], index=idx2)
+    >>> df2.to_hdf("file", "data", "a", append=True)  # doctest: +SKIP
+    AttributeConflictWarning: the [index_name] attribute of the existing index is
+    [name1] which conflicts with the new [name2]...
+    """
+
+
+class DatabaseError(OSError):
+    """
+    Error is raised when executing SQL with bad syntax or SQL that throws an error.
+
+    Raised by :func:`.pandas.read_sql` when a bad SQL statement is passed in.
+
+    See Also
+    --------
+    read_sql : Read SQL query or database table into a DataFrame.
+
+    Examples
+    --------
+    >>> from sqlite3 import connect
+    >>> conn = connect(":memory:")
+    >>> pd.read_sql("select * test", conn)  # doctest: +SKIP
+    """
+
+
+class PossiblePrecisionLoss(Warning):
+    """
+    Warning raised by to_stata on a column with a value outside or equal to int64.
+
+    When the column value is outside or equal to the int64 value the column is
+    converted to a float64 dtype.
+
+    See Also
+    --------
+    DataFrame.to_stata : Export DataFrame object to Stata dta format.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"s": pd.Series([1, 2**53], dtype=np.int64)})
+    >>> df.to_stata("test")  # doctest: +SKIP
+    """
+
+
+class ValueLabelTypeMismatch(Warning):
+    """
+    Warning raised by to_stata on a category column that contains non-string values.
+
+    When exporting data to Stata format using the `to_stata` method, category columns
+    must have string values as labels. If a category column contains non-string values
+    (e.g., integers, floats, or other types), this warning is raised to indicate that
+    the Stata file may not correctly represent the data.
+
+    See Also
+    --------
+    DataFrame.to_stata : Export DataFrame object to Stata dta format.
+    Series.cat : Accessor for categorical properties of the Series values.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"categories": pd.Series(["a", 2], dtype="category")})
+    >>> df.to_stata("test")  # doctest: +SKIP
+    """
+
+
+class InvalidColumnName(Warning):
+    """
+    Warning raised by to_stata the column contains a non-valid stata name.
+
+    Because the column name is an invalid Stata variable, the name needs to be
+    converted.
+
+    See Also
+    --------
+    DataFrame.to_stata : Export DataFrame object to Stata dta format.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({"0categories": pd.Series([2, 2])})
+    >>> df.to_stata("test")  # doctest: +SKIP
+    """
+
+
+class CategoricalConversionWarning(Warning):
+    """
+    Warning is raised when reading a partial labeled Stata file using a iterator.
+
+    This warning helps ensure data integrity and alerts users to potential issues
+    during the incremental reading of Stata files with labeled data, allowing for
+    additional checks and adjustments as necessary.
+
+    See Also
+    --------
+    read_stata : Read a Stata file into a DataFrame.
+    Categorical : Represents a categorical variable in pandas.
+
+    Examples
+    --------
+    >>> from pandas.io.stata import StataReader
+    >>> with StataReader("dta_file", chunksize=2) as reader:  # doctest: +SKIP
+    ...     for i, block in enumerate(reader):
+    ...         print(i, block)
+    ... # CategoricalConversionWarning: One or more series with value labels...
+    """
+
+
+class LossySetitemError(Exception):
+    """
+    Raised when trying to do a __setitem__ on an np.ndarray that is not lossless.
+
+    Notes
+    -----
+    This is an internal error.
+    """
+
+
+class NoBufferPresent(Exception):
+    """
+    Exception is raised in _get_data_buffer to signal that there is no requested buffer.
+    """
+
+
+class InvalidComparison(Exception):
+    """
+    Exception is raised by _validate_comparison_value to indicate an invalid comparison.
+
+    Notes
+    -----
+    This is an internal error.
+    """
+
+
+__all__ = [
+    "AbstractMethodError",
+    "AttributeConflictWarning",
+    "CSSWarning",
+    "CategoricalConversionWarning",
+    "ChainedAssignmentError",
+    "ClosedFileError",
+    "DataError",
+    "DatabaseError",
+    "DtypeWarning",
+    "DuplicateLabelError",
+    "EmptyDataError",
+    "IncompatibilityWarning",
+    "IndexingError",
+    "IntCastingNaNError",
+    "InvalidColumnName",
+    "InvalidComparison",
+    "InvalidIndexError",
+    "InvalidVersion",
+    "LossySetitemError",
+    "MergeError",
+    "NoBufferPresent",
+    "NullFrequencyError",
+    "NumExprClobberingError",
+    "NumbaUtilError",
+    "OptionError",
+    "OutOfBoundsDatetime",
+    "OutOfBoundsTimedelta",
+    "ParserError",
+    "ParserWarning",
+    "PerformanceWarning",
+    "PossibleDataLossError",
+    "PossiblePrecisionLoss",
+    "PyperclipException",
+    "PyperclipWindowsException",
+    "SpecificationError",
+    "UndefinedVariableError",
+    "UnsortedIndexError",
+    "UnsupportedFunctionCall",
+    "ValueLabelTypeMismatch",
+]
diff --git a/pandas/pandas/errors/cow.py b/pandas/pandas/errors/cow.py
new file mode 100644
index 0000000000000..4815a0f5328d7
--- /dev/null
+++ b/pandas/pandas/errors/cow.py
@@ -0,0 +1,24 @@
+_chained_assignment_msg = (
+    "A value is trying to be set on a copy of a DataFrame or Series "
+    "through chained assignment.\n"
+    "When using the Copy-on-Write mode, such chained assignment never works "
+    "to update the original DataFrame or Series, because the intermediate "
+    "object on which we are setting values always behaves as a copy.\n\n"
+    "Try using '.loc[row_indexer, col_indexer] = value' instead, to perform "
+    "the assignment in a single step.\n\n"
+    "See the caveats in the documentation: "
+    "https://pandas.pydata.org/pandas-docs/stable/user_guide/"
+    "copy_on_write.html"
+)
+
+
+_chained_assignment_method_msg = (
+    "A value is trying to be set on a copy of a DataFrame or Series "
+    "through chained assignment using an inplace method.\n"
+    "When using the Copy-on-Write mode, such inplace method never works "
+    "to update the original DataFrame or Series, because the intermediate "
+    "object on which we are setting values always behaves as a copy.\n\n"
+    "For example, when doing 'df[col].method(value, inplace=True)', try "
+    "using 'df.method({col: value}, inplace=True)' instead, to perform "
+    "the operation inplace on the original object.\n\n"
+)
diff --git a/pandas/pandas/io/__init__.py b/pandas/pandas/io/__init__.py
new file mode 100644
index 0000000000000..1c7e531debb14
--- /dev/null
+++ b/pandas/pandas/io/__init__.py
@@ -0,0 +1,13 @@
+# ruff: noqa: TC004
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    # import modules that have public classes/functions
+    from pandas.io import (
+        formats,
+        json,
+        stata,
+    )
+
+    # mark only those modules as public
+    __all__ = ["formats", "json", "stata"]
diff --git a/pandas/pandas/io/_util.py b/pandas/pandas/io/_util.py
new file mode 100644
index 0000000000000..6827fbe9c998e
--- /dev/null
+++ b/pandas/pandas/io/_util.py
@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+from pandas.compat import (
+    pa_version_under18p0,
+    pa_version_under19p0,
+)
+from pandas.compat._optional import import_optional_dependency
+
+import pandas as pd
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    import pyarrow
+
+    from pandas._typing import DtypeBackend
+
+
+def _arrow_dtype_mapping() -> dict:
+    pa = import_optional_dependency("pyarrow")
+    return {
+        pa.int8(): pd.Int8Dtype(),
+        pa.int16(): pd.Int16Dtype(),
+        pa.int32(): pd.Int32Dtype(),
+        pa.int64(): pd.Int64Dtype(),
+        pa.uint8(): pd.UInt8Dtype(),
+        pa.uint16(): pd.UInt16Dtype(),
+        pa.uint32(): pd.UInt32Dtype(),
+        pa.uint64(): pd.UInt64Dtype(),
+        pa.bool_(): pd.BooleanDtype(),
+        pa.string(): pd.StringDtype(),
+        pa.float32(): pd.Float32Dtype(),
+        pa.float64(): pd.Float64Dtype(),
+        pa.string(): pd.StringDtype(),
+        pa.large_string(): pd.StringDtype(),
+    }
+
+
+def _arrow_string_types_mapper() -> Callable:
+    pa = import_optional_dependency("pyarrow")
+
+    mapping = {
+        pa.string(): pd.StringDtype(na_value=np.nan),
+        pa.large_string(): pd.StringDtype(na_value=np.nan),
+    }
+    if not pa_version_under18p0:
+        mapping[pa.string_view()] = pd.StringDtype(na_value=np.nan)
+
+    return mapping.get
+
+
+def arrow_table_to_pandas(
+    table: pyarrow.Table,
+    dtype_backend: DtypeBackend | Literal["numpy"] | lib.NoDefault = lib.no_default,
+    null_to_int64: bool = False,
+    to_pandas_kwargs: dict | None = None,
+) -> pd.DataFrame:
+    pa = import_optional_dependency("pyarrow")
+
+    to_pandas_kwargs = {} if to_pandas_kwargs is None else to_pandas_kwargs
+
+    types_mapper: type[pd.ArrowDtype] | None | Callable
+    if dtype_backend == "numpy_nullable":
+        mapping = _arrow_dtype_mapping()
+        if null_to_int64:
+            # Modify the default mapping to also map null to Int64
+            # (to match other engines - only for CSV parser)
+            mapping[pa.null()] = pd.Int64Dtype()
+        types_mapper = mapping.get
+    elif dtype_backend == "pyarrow":
+        types_mapper = pd.ArrowDtype
+    elif using_string_dtype():
+        if pa_version_under19p0:
+            types_mapper = _arrow_string_types_mapper()
+        else:
+            types_mapper = None
+    elif dtype_backend is lib.no_default or dtype_backend == "numpy":
+        types_mapper = None
+    else:
+        raise NotImplementedError
+
+    df = table.to_pandas(types_mapper=types_mapper, **to_pandas_kwargs)
+    return df
diff --git a/pandas/pandas/io/api.py b/pandas/pandas/io/api.py
new file mode 100644
index 0000000000000..d4982399a604b
--- /dev/null
+++ b/pandas/pandas/io/api.py
@@ -0,0 +1,63 @@
+"""
+Data IO api
+"""
+
+from pandas.io.clipboards import read_clipboard
+from pandas.io.excel import (
+    ExcelFile,
+    ExcelWriter,
+    read_excel,
+)
+from pandas.io.feather_format import read_feather
+from pandas.io.html import read_html
+from pandas.io.json import read_json
+from pandas.io.orc import read_orc
+from pandas.io.parquet import read_parquet
+from pandas.io.parsers import (
+    read_csv,
+    read_fwf,
+    read_table,
+)
+from pandas.io.pickle import (
+    read_pickle,
+    to_pickle,
+)
+from pandas.io.pytables import (
+    HDFStore,
+    read_hdf,
+)
+from pandas.io.sas import read_sas
+from pandas.io.spss import read_spss
+from pandas.io.sql import (
+    read_sql,
+    read_sql_query,
+    read_sql_table,
+)
+from pandas.io.stata import read_stata
+from pandas.io.xml import read_xml
+
+__all__ = [
+    "ExcelFile",
+    "ExcelWriter",
+    "HDFStore",
+    "read_clipboard",
+    "read_csv",
+    "read_excel",
+    "read_feather",
+    "read_fwf",
+    "read_hdf",
+    "read_html",
+    "read_json",
+    "read_orc",
+    "read_parquet",
+    "read_pickle",
+    "read_sas",
+    "read_spss",
+    "read_sql",
+    "read_sql_query",
+    "read_sql_table",
+    "read_stata",
+    "read_table",
+    "read_xml",
+    "to_pickle",
+]
diff --git a/pandas/pandas/io/clipboard/__init__.py b/pandas/pandas/io/clipboard/__init__.py
new file mode 100644
index 0000000000000..6491849925e86
--- /dev/null
+++ b/pandas/pandas/io/clipboard/__init__.py
@@ -0,0 +1,747 @@
+"""
+Pyperclip
+
+A cross-platform clipboard module for Python,
+with copy & paste functions for plain text.
+By Al Sweigart al@inventwithpython.com
+Licence at LICENSES/PYPERCLIP_LICENSE
+
+Usage:
+  import pyperclip
+  pyperclip.copy('The text to be copied to the clipboard.')
+  spam = pyperclip.paste()
+
+  if not pyperclip.is_available():
+    print("Copy functionality unavailable!")
+
+On Windows, no additional modules are needed.
+On Mac, the pyobjc module is used, falling back to the pbcopy and pbpaste cli
+    commands. (These commands should come with OS X.).
+On Linux, install xclip, xsel, or wl-clipboard (for "wayland" sessions) via
+package manager.
+For example, in Debian:
+    sudo apt-get install xclip
+    sudo apt-get install xsel
+    sudo apt-get install wl-clipboard
+
+Otherwise on Linux, you will need the PyQt5 modules installed.
+
+This module does not work with PyGObject yet.
+
+Cygwin is currently not supported.
+
+Security Note: This module runs programs with these names:
+    - pbcopy
+    - pbpaste
+    - xclip
+    - xsel
+    - wl-copy/wl-paste
+    - klipper
+    - qdbus
+A malicious user could rename or add programs with these names, tricking
+Pyperclip into running them with whatever permissions the Python process has.
+
+"""
+
+__version__ = "1.8.2"
+
+
+import contextlib
+import ctypes
+from ctypes import (
+    c_size_t,
+    c_wchar,
+    c_wchar_p,
+    get_errno,
+    sizeof,
+)
+import os
+import platform
+from shutil import which as _executable_exists
+import subprocess
+import time
+import warnings
+
+from pandas.errors import (
+    PyperclipException,
+    PyperclipWindowsException,
+)
+from pandas.util._exceptions import find_stack_level
+
+# `import PyQt4` sys.exit()s if DISPLAY is not in the environment.
+# Thus, we need to detect the presence of $DISPLAY manually
+# and not load PyQt4 if it is absent.
+HAS_DISPLAY = os.getenv("DISPLAY")
+
+EXCEPT_MSG = """
+    Pyperclip could not find a copy/paste mechanism for your system.
+    For more information, please visit
+    https://pyperclip.readthedocs.io/en/latest/index.html#not-implemented-error
+    """
+
+ENCODING = "utf-8"
+
+
+class PyperclipTimeoutException(PyperclipException):
+    pass
+
+
+def _stringifyText(text) -> str:
+    acceptedTypes = (str, int, float, bool)
+    if not isinstance(text, acceptedTypes):
+        raise PyperclipException(
+            f"only str, int, float, and bool values "
+            f"can be copied to the clipboard, not {type(text).__name__}"
+        )
+    return str(text)
+
+
+def init_osx_pbcopy_clipboard():
+    def copy_osx_pbcopy(text):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        with subprocess.Popen(
+            ["pbcopy", "w"], stdin=subprocess.PIPE, close_fds=True
+        ) as p:
+            p.communicate(input=text.encode(ENCODING))
+
+    def paste_osx_pbcopy():
+        with subprocess.Popen(
+            ["pbpaste", "r"], stdout=subprocess.PIPE, close_fds=True
+        ) as p:
+            stdout = p.communicate()[0]
+        return stdout.decode(ENCODING)
+
+    return copy_osx_pbcopy, paste_osx_pbcopy
+
+
+def init_osx_pyobjc_clipboard():
+    def copy_osx_pyobjc(text):
+        """Copy string argument to clipboard"""
+        text = _stringifyText(text)  # Converts non-str values to str.
+        newStr = Foundation.NSString.stringWithString_(text).nsstring()
+        newData = newStr.dataUsingEncoding_(Foundation.NSUTF8StringEncoding)
+        board = AppKit.NSPasteboard.generalPasteboard()
+        board.declareTypes_owner_([AppKit.NSStringPboardType], None)
+        board.setData_forType_(newData, AppKit.NSStringPboardType)
+
+    def paste_osx_pyobjc():
+        """Returns contents of clipboard"""
+        board = AppKit.NSPasteboard.generalPasteboard()
+        content = board.stringForType_(AppKit.NSStringPboardType)
+        return content
+
+    return copy_osx_pyobjc, paste_osx_pyobjc
+
+
+def init_qt_clipboard():
+    global QApplication
+    # $DISPLAY should exist
+
+    # Try to import from qtpy, but if that fails try PyQt5 then PyQt4
+    try:
+        from qtpy.QtWidgets import QApplication
+    except ImportError:
+        try:
+            from PyQt5.QtWidgets import QApplication
+        except ImportError:
+            from PyQt4.QtGui import QApplication
+
+    app = QApplication.instance()
+    if app is None:
+        app = QApplication([])
+
+    def copy_qt(text):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        cb = app.clipboard()
+        cb.setText(text)
+
+    def paste_qt() -> str:
+        cb = app.clipboard()
+        return str(cb.text())
+
+    return copy_qt, paste_qt
+
+
+def init_xclip_clipboard():
+    DEFAULT_SELECTION = "c"
+    PRIMARY_SELECTION = "p"
+
+    def copy_xclip(text, primary=False):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        selection = DEFAULT_SELECTION
+        if primary:
+            selection = PRIMARY_SELECTION
+        with subprocess.Popen(
+            ["xclip", "-selection", selection], stdin=subprocess.PIPE, close_fds=True
+        ) as p:
+            p.communicate(input=text.encode(ENCODING))
+
+    def paste_xclip(primary=False):
+        selection = DEFAULT_SELECTION
+        if primary:
+            selection = PRIMARY_SELECTION
+        with subprocess.Popen(
+            ["xclip", "-selection", selection, "-o"],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            close_fds=True,
+        ) as p:
+            stdout = p.communicate()[0]
+        # Intentionally ignore extraneous output on stderr when clipboard is empty
+        return stdout.decode(ENCODING)
+
+    return copy_xclip, paste_xclip
+
+
+def init_xsel_clipboard():
+    DEFAULT_SELECTION = "-b"
+    PRIMARY_SELECTION = "-p"
+
+    def copy_xsel(text, primary=False):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        selection_flag = DEFAULT_SELECTION
+        if primary:
+            selection_flag = PRIMARY_SELECTION
+        with subprocess.Popen(
+            ["xsel", selection_flag, "-i"], stdin=subprocess.PIPE, close_fds=True
+        ) as p:
+            p.communicate(input=text.encode(ENCODING))
+
+    def paste_xsel(primary=False):
+        selection_flag = DEFAULT_SELECTION
+        if primary:
+            selection_flag = PRIMARY_SELECTION
+        with subprocess.Popen(
+            ["xsel", selection_flag, "-o"], stdout=subprocess.PIPE, close_fds=True
+        ) as p:
+            stdout = p.communicate()[0]
+        return stdout.decode(ENCODING)
+
+    return copy_xsel, paste_xsel
+
+
+def init_wl_clipboard():
+    PRIMARY_SELECTION = "-p"
+
+    def copy_wl(text, primary=False):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        args = ["wl-copy"]
+        if primary:
+            args.append(PRIMARY_SELECTION)
+        if not text:
+            args.append("--clear")
+            subprocess.check_call(args, close_fds=True)
+        else:
+            p = subprocess.Popen(args, stdin=subprocess.PIPE, close_fds=True)
+            p.communicate(input=text.encode(ENCODING))
+
+    def paste_wl(primary=False):
+        args = ["wl-paste", "-n"]
+        if primary:
+            args.append(PRIMARY_SELECTION)
+        p = subprocess.Popen(args, stdout=subprocess.PIPE, close_fds=True)
+        stdout, _stderr = p.communicate()
+        return stdout.decode(ENCODING)
+
+    return copy_wl, paste_wl
+
+
+def init_klipper_clipboard():
+    def copy_klipper(text):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        with subprocess.Popen(
+            [
+                "qdbus",
+                "org.kde.klipper",
+                "/klipper",
+                "setClipboardContents",
+                text.encode(ENCODING),
+            ],
+            stdin=subprocess.PIPE,
+            close_fds=True,
+        ) as p:
+            p.communicate(input=None)
+
+    def paste_klipper():
+        with subprocess.Popen(
+            ["qdbus", "org.kde.klipper", "/klipper", "getClipboardContents"],
+            stdout=subprocess.PIPE,
+            close_fds=True,
+        ) as p:
+            stdout = p.communicate()[0]
+
+        # Workaround for https://bugs.kde.org/show_bug.cgi?id=342874
+        # TODO: https://github.com/asweigart/pyperclip/issues/43
+        clipboardContents = stdout.decode(ENCODING)
+        # even if blank, Klipper will append a newline at the end
+        assert len(clipboardContents) > 0
+        # make sure that newline is there
+        assert clipboardContents.endswith("\n")
+        if clipboardContents.endswith("\n"):
+            clipboardContents = clipboardContents[:-1]
+        return clipboardContents
+
+    return copy_klipper, paste_klipper
+
+
+def init_dev_clipboard_clipboard():
+    def copy_dev_clipboard(text):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        if text == "":
+            warnings.warn(
+                "Pyperclip cannot copy a blank string to the clipboard on Cygwin. "
+                "This is effectively a no-op.",
+                stacklevel=find_stack_level(),
+            )
+        if "\r" in text:
+            warnings.warn(
+                "Pyperclip cannot handle \\r characters on Cygwin.",
+                stacklevel=find_stack_level(),
+            )
+
+        with open("/dev/clipboard", "w", encoding="utf-8") as fd:
+            fd.write(text)
+
+    def paste_dev_clipboard() -> str:
+        with open("/dev/clipboard", encoding="utf-8") as fd:
+            content = fd.read()
+        return content
+
+    return copy_dev_clipboard, paste_dev_clipboard
+
+
+def init_no_clipboard():
+    class ClipboardUnavailable:
+        def __call__(self, *args, **kwargs):
+            raise PyperclipException(EXCEPT_MSG)
+
+        def __bool__(self) -> bool:
+            return False
+
+    return ClipboardUnavailable(), ClipboardUnavailable()
+
+
+# Windows-related clipboard functions:
+class CheckedCall:
+    def __init__(self, f) -> None:
+        super().__setattr__("f", f)
+
+    def __call__(self, *args):
+        ret = self.f(*args)
+        if not ret and get_errno():
+            raise PyperclipWindowsException("Error calling " + self.f.__name__)
+        return ret
+
+    def __setattr__(self, key, value):
+        setattr(self.f, key, value)
+
+
+def init_windows_clipboard():
+    global HGLOBAL, LPVOID, DWORD, LPCSTR, INT
+    global HWND, HINSTANCE, HMENU, BOOL, UINT, HANDLE
+    from ctypes.wintypes import (
+        BOOL,
+        DWORD,
+        HANDLE,
+        HGLOBAL,
+        HINSTANCE,
+        HMENU,
+        HWND,
+        INT,
+        LPCSTR,
+        LPVOID,
+        UINT,
+    )
+
+    windll = ctypes.windll
+    msvcrt = ctypes.CDLL("msvcrt")
+
+    safeCreateWindowExA = CheckedCall(windll.user32.CreateWindowExA)
+    safeCreateWindowExA.argtypes = [
+        DWORD,
+        LPCSTR,
+        LPCSTR,
+        DWORD,
+        INT,
+        INT,
+        INT,
+        INT,
+        HWND,
+        HMENU,
+        HINSTANCE,
+        LPVOID,
+    ]
+    safeCreateWindowExA.restype = HWND
+
+    safeDestroyWindow = CheckedCall(windll.user32.DestroyWindow)
+    safeDestroyWindow.argtypes = [HWND]
+    safeDestroyWindow.restype = BOOL
+
+    OpenClipboard = windll.user32.OpenClipboard
+    OpenClipboard.argtypes = [HWND]
+    OpenClipboard.restype = BOOL
+
+    safeCloseClipboard = CheckedCall(windll.user32.CloseClipboard)
+    safeCloseClipboard.argtypes = []
+    safeCloseClipboard.restype = BOOL
+
+    safeEmptyClipboard = CheckedCall(windll.user32.EmptyClipboard)
+    safeEmptyClipboard.argtypes = []
+    safeEmptyClipboard.restype = BOOL
+
+    safeGetClipboardData = CheckedCall(windll.user32.GetClipboardData)
+    safeGetClipboardData.argtypes = [UINT]
+    safeGetClipboardData.restype = HANDLE
+
+    safeSetClipboardData = CheckedCall(windll.user32.SetClipboardData)
+    safeSetClipboardData.argtypes = [UINT, HANDLE]
+    safeSetClipboardData.restype = HANDLE
+
+    safeGlobalAlloc = CheckedCall(windll.kernel32.GlobalAlloc)
+    safeGlobalAlloc.argtypes = [UINT, c_size_t]
+    safeGlobalAlloc.restype = HGLOBAL
+
+    safeGlobalLock = CheckedCall(windll.kernel32.GlobalLock)
+    safeGlobalLock.argtypes = [HGLOBAL]
+    safeGlobalLock.restype = LPVOID
+
+    safeGlobalUnlock = CheckedCall(windll.kernel32.GlobalUnlock)
+    safeGlobalUnlock.argtypes = [HGLOBAL]
+    safeGlobalUnlock.restype = BOOL
+
+    wcslen = CheckedCall(msvcrt.wcslen)
+    wcslen.argtypes = [c_wchar_p]
+    wcslen.restype = UINT
+
+    GMEM_MOVEABLE = 0x0002
+    CF_UNICODETEXT = 13
+
+    @contextlib.contextmanager
+    def window():
+        """
+        Context that provides a valid Windows hwnd.
+        """
+        # we really just need the hwnd, so setting "STATIC"
+        # as predefined lpClass is just fine.
+        hwnd = safeCreateWindowExA(
+            0, b"STATIC", None, 0, 0, 0, 0, 0, None, None, None, None
+        )
+        try:
+            yield hwnd
+        finally:
+            safeDestroyWindow(hwnd)
+
+    @contextlib.contextmanager
+    def clipboard(hwnd):
+        """
+        Context manager that opens the clipboard and prevents
+        other applications from modifying the clipboard content.
+        """
+        # We may not get the clipboard handle immediately because
+        # some other application is accessing it (?)
+        # We try for at least 500ms to get the clipboard.
+        t = time.time() + 0.5
+        success = False
+        while time.time() < t:
+            success = OpenClipboard(hwnd)
+            if success:
+                break
+            time.sleep(0.01)
+        if not success:
+            raise PyperclipWindowsException("Error calling OpenClipboard")
+
+        try:
+            yield
+        finally:
+            safeCloseClipboard()
+
+    def copy_windows(text):
+        # This function is heavily based on
+        # http://msdn.com/ms649016#_win32_Copying_Information_to_the_Clipboard
+
+        text = _stringifyText(text)  # Converts non-str values to str.
+
+        with window() as hwnd:
+            # http://msdn.com/ms649048
+            # If an application calls OpenClipboard with hwnd set to NULL,
+            # EmptyClipboard sets the clipboard owner to NULL;
+            # this causes SetClipboardData to fail.
+            # => We need a valid hwnd to copy something.
+            with clipboard(hwnd):
+                safeEmptyClipboard()
+
+                if text:
+                    # http://msdn.com/ms649051
+                    # If the hMem parameter identifies a memory object,
+                    # the object must have been allocated using the
+                    # function with the GMEM_MOVEABLE flag.
+                    count = wcslen(text) + 1
+                    handle = safeGlobalAlloc(GMEM_MOVEABLE, count * sizeof(c_wchar))
+                    locked_handle = safeGlobalLock(handle)
+
+                    ctypes.memmove(
+                        c_wchar_p(locked_handle),
+                        c_wchar_p(text),
+                        count * sizeof(c_wchar),
+                    )
+
+                    safeGlobalUnlock(handle)
+                    safeSetClipboardData(CF_UNICODETEXT, handle)
+
+    def paste_windows():
+        with clipboard(None):
+            handle = safeGetClipboardData(CF_UNICODETEXT)
+            if not handle:
+                # GetClipboardData may return NULL with errno == NO_ERROR
+                # if the clipboard is empty.
+                # (Also, it may return a handle to an empty buffer,
+                # but technically that's not empty)
+                return ""
+            return c_wchar_p(handle).value
+
+    return copy_windows, paste_windows
+
+
+def init_wsl_clipboard():
+    def copy_wsl(text):
+        text = _stringifyText(text)  # Converts non-str values to str.
+        with subprocess.Popen(["clip.exe"], stdin=subprocess.PIPE, close_fds=True) as p:
+            p.communicate(input=text.encode(ENCODING))
+
+    def paste_wsl():
+        with subprocess.Popen(
+            ["powershell.exe", "-command", "Get-Clipboard"],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            close_fds=True,
+        ) as p:
+            stdout = p.communicate()[0]
+        # WSL appends "\r\n" to the contents.
+        return stdout[:-2].decode(ENCODING)
+
+    return copy_wsl, paste_wsl
+
+
+# Automatic detection of clipboard mechanisms
+# and importing is done in determine_clipboard():
+def determine_clipboard():
+    """
+    Determine the OS/platform and set the copy() and paste() functions
+    accordingly.
+    """
+    global Foundation, AppKit, qtpy, PyQt4, PyQt5
+
+    # Setup for the CYGWIN platform:
+    if (
+        "cygwin" in platform.system().lower()
+    ):  # Cygwin has a variety of values returned by platform.system(),
+        # such as 'CYGWIN_NT-6.1'
+        # FIXME(pyperclip#55): pyperclip currently does not support Cygwin,
+        # see https://github.com/asweigart/pyperclip/issues/55
+        if os.path.exists("/dev/clipboard"):
+            warnings.warn(
+                "Pyperclip's support for Cygwin is not perfect, "
+                "see https://github.com/asweigart/pyperclip/issues/55",
+                stacklevel=find_stack_level(),
+            )
+            return init_dev_clipboard_clipboard()
+
+    # Setup for the WINDOWS platform:
+    elif os.name == "nt" or platform.system() == "Windows":
+        return init_windows_clipboard()
+
+    if platform.system() == "Linux":
+        if _executable_exists("wslconfig.exe"):
+            return init_wsl_clipboard()
+
+    # Setup for the macOS platform:
+    if os.name == "mac" or platform.system() == "Darwin":
+        try:
+            import AppKit
+            import Foundation  # check if pyobjc is installed
+        except ImportError:
+            return init_osx_pbcopy_clipboard()
+        else:
+            return init_osx_pyobjc_clipboard()
+
+    # Setup for the LINUX platform:
+    if HAS_DISPLAY:
+        if os.environ.get("WAYLAND_DISPLAY") and _executable_exists("wl-copy"):
+            return init_wl_clipboard()
+        if _executable_exists("xsel"):
+            return init_xsel_clipboard()
+        if _executable_exists("xclip"):
+            return init_xclip_clipboard()
+        if _executable_exists("klipper") and _executable_exists("qdbus"):
+            return init_klipper_clipboard()
+
+        try:
+            # qtpy is a small abstraction layer that lets you write applications
+            # using a single api call to either PyQt or PySide.
+            # https://pypi.python.org/project/QtPy
+            import qtpy  # check if qtpy is installed
+        except ImportError:
+            # If qtpy isn't installed, fall back on importing PyQt4.
+            try:
+                import PyQt5  # check if PyQt5 is installed
+            except ImportError:
+                try:
+                    import PyQt4  # check if PyQt4 is installed
+                except ImportError:
+                    pass  # We want to fail fast for all non-ImportError exceptions.
+                else:
+                    return init_qt_clipboard()
+            else:
+                return init_qt_clipboard()
+        else:
+            return init_qt_clipboard()
+
+    return init_no_clipboard()
+
+
+def set_clipboard(clipboard):
+    """
+    Explicitly sets the clipboard mechanism. The "clipboard mechanism" is how
+    the copy() and paste() functions interact with the operating system to
+    implement the copy/paste feature. The clipboard parameter must be one of:
+        - pbcopy
+        - pyobjc (default on macOS)
+        - qt
+        - xclip
+        - xsel
+        - klipper
+        - windows (default on Windows)
+        - no (this is what is set when no clipboard mechanism can be found)
+    """
+    global copy, paste
+
+    clipboard_types = {
+        "pbcopy": init_osx_pbcopy_clipboard,
+        "pyobjc": init_osx_pyobjc_clipboard,
+        "qt": init_qt_clipboard,  # TODO - split this into 'qtpy', 'pyqt4', and 'pyqt5'
+        "xclip": init_xclip_clipboard,
+        "xsel": init_xsel_clipboard,
+        "wl-clipboard": init_wl_clipboard,
+        "klipper": init_klipper_clipboard,
+        "windows": init_windows_clipboard,
+        "no": init_no_clipboard,
+    }
+
+    if clipboard not in clipboard_types:
+        allowed_clipboard_types = [repr(_) for _ in clipboard_types]
+        raise ValueError(
+            f"Argument must be one of {', '.join(allowed_clipboard_types)}"
+        )
+
+    # Sets pyperclip's copy() and paste() functions:
+    copy, paste = clipboard_types[clipboard]()
+
+
+def lazy_load_stub_copy(text):
+    """
+    A stub function for copy(), which will load the real copy() function when
+    called so that the real copy() function is used for later calls.
+
+    This allows users to import pyperclip without having determine_clipboard()
+    automatically run, which will automatically select a clipboard mechanism.
+    This could be a problem if it selects, say, the memory-heavy PyQt4 module
+    but the user was just going to immediately call set_clipboard() to use a
+    different clipboard mechanism.
+
+    The lazy loading this stub function implements gives the user a chance to
+    call set_clipboard() to pick another clipboard mechanism. Or, if the user
+    simply calls copy() or paste() without calling set_clipboard() first,
+    will fall back on whatever clipboard mechanism that determine_clipboard()
+    automatically chooses.
+    """
+    global copy, paste
+    copy, paste = determine_clipboard()
+    return copy(text)
+
+
+def lazy_load_stub_paste():
+    """
+    A stub function for paste(), which will load the real paste() function when
+    called so that the real paste() function is used for later calls.
+
+    This allows users to import pyperclip without having determine_clipboard()
+    automatically run, which will automatically select a clipboard mechanism.
+    This could be a problem if it selects, say, the memory-heavy PyQt4 module
+    but the user was just going to immediately call set_clipboard() to use a
+    different clipboard mechanism.
+
+    The lazy loading this stub function implements gives the user a chance to
+    call set_clipboard() to pick another clipboard mechanism. Or, if the user
+    simply calls copy() or paste() without calling set_clipboard() first,
+    will fall back on whatever clipboard mechanism that determine_clipboard()
+    automatically chooses.
+    """
+    global copy, paste
+    copy, paste = determine_clipboard()
+    return paste()
+
+
+def is_available() -> bool:
+    return copy != lazy_load_stub_copy and paste != lazy_load_stub_paste
+
+
+# Initially, copy() and paste() are set to lazy loading wrappers which will
+# set `copy` and `paste` to real functions the first time they're used, unless
+# set_clipboard() or determine_clipboard() is called first.
+copy, paste = lazy_load_stub_copy, lazy_load_stub_paste
+
+
+def waitForPaste(timeout=None):
+    """This function call blocks until a non-empty text string exists on the
+    clipboard. It returns this text.
+
+    This function raises PyperclipTimeoutException if timeout was set to
+    a number of seconds that has elapsed without non-empty text being put on
+    the clipboard."""
+    startTime = time.time()
+    while True:
+        clipboardText = paste()
+        if clipboardText != "":
+            return clipboardText
+        time.sleep(0.01)
+
+        if timeout is not None and time.time() > startTime + timeout:
+            raise PyperclipTimeoutException(
+                "waitForPaste() timed out after " + str(timeout) + " seconds."
+            )
+
+
+def waitForNewPaste(timeout=None):
+    """This function call blocks until a new text string exists on the
+    clipboard that is different from the text that was there when the function
+    was first called. It returns this text.
+
+    This function raises PyperclipTimeoutException if timeout was set to
+    a number of seconds that has elapsed without non-empty text being put on
+    the clipboard."""
+    startTime = time.time()
+    originalText = paste()
+    while True:
+        currentText = paste()
+        if currentText != originalText:
+            return currentText
+        time.sleep(0.01)
+
+        if timeout is not None and time.time() > startTime + timeout:
+            raise PyperclipTimeoutException(
+                "waitForNewPaste() timed out after " + str(timeout) + " seconds."
+            )
+
+
+__all__ = [
+    "copy",
+    "paste",
+    "waitForPaste",
+    "waitForNewPaste",
+    "set_clipboard",
+    "determine_clipboard",
+]
+
+# pandas aliases
+clipboard_get = paste
+clipboard_set = copy
diff --git a/pandas/pandas/io/clipboards.py b/pandas/pandas/io/clipboards.py
new file mode 100644
index 0000000000000..2ed241f0b9bca
--- /dev/null
+++ b/pandas/pandas/io/clipboards.py
@@ -0,0 +1,198 @@
+"""io on the clipboard"""
+
+from __future__ import annotations
+
+from io import StringIO
+from typing import TYPE_CHECKING
+import warnings
+
+from pandas._libs import lib
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.generic import ABCDataFrame
+
+from pandas import (
+    get_option,
+    option_context,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import DtypeBackend
+
+
+def read_clipboard(
+    sep: str = r"\s+",
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    **kwargs,
+):  # pragma: no cover
+    r"""
+    Read text from clipboard and pass to :func:`~pandas.read_csv`.
+
+    Parses clipboard contents similar to how CSV files are parsed
+    using :func:`~pandas.read_csv`.
+
+    Parameters
+    ----------
+    sep : str, default '\\s+'
+        A string or regex delimiter. The default of ``'\\s+'`` denotes
+        one or more whitespace characters.
+
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    **kwargs
+        See :func:`~pandas.read_csv` for the full argument list.
+
+    Returns
+    -------
+    DataFrame
+        A parsed :class:`~pandas.DataFrame` object.
+
+    See Also
+    --------
+    DataFrame.to_clipboard : Copy object to the system clipboard.
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+    read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+    >>> df.to_clipboard()  # doctest: +SKIP
+    >>> pd.read_clipboard()  # doctest: +SKIP
+         A  B  C
+    0    1  2  3
+    1    4  5  6
+    """
+    encoding = kwargs.pop("encoding", "utf-8")
+
+    # only utf-8 is valid for passed value because that's what clipboard
+    # supports
+    if encoding is not None and encoding.lower().replace("-", "") != "utf8":
+        raise NotImplementedError("reading from clipboard only supports utf-8 encoding")
+
+    check_dtype_backend(dtype_backend)
+
+    from pandas.io.clipboard import clipboard_get
+    from pandas.io.parsers import read_csv
+
+    text = clipboard_get()
+
+    # Try to decode (if needed, as "text" might already be a string here).
+    try:
+        text = text.decode(kwargs.get("encoding") or get_option("display.encoding"))
+    except AttributeError:
+        pass
+
+    # Excel copies into clipboard with \t separation
+    # inspect no more then the 10 first lines, if they
+    # all contain an equal number (>0) of tabs, infer
+    # that this came from excel and set 'sep' accordingly
+    lines = text[:10000].split("\n")[:-1][:10]
+
+    # Need to remove leading white space, since read_csv
+    # accepts:
+    #    a  b
+    # 0  1  2
+    # 1  3  4
+
+    counts = {x.lstrip(" ").count("\t") for x in lines}
+    if len(lines) > 1 and len(counts) == 1 and counts.pop() != 0:
+        sep = "\t"
+        # check the number of leading tabs in the first line
+        # to account for index columns
+        index_length = len(lines[0]) - len(lines[0].lstrip(" \t"))
+        if index_length != 0:
+            kwargs.setdefault("index_col", list(range(index_length)))
+
+    elif not isinstance(sep, str):
+        raise ValueError(f"{sep=} must be a string")
+
+    # Regex separator currently only works with python engine.
+    # Default to python if separator is multi-character (regex)
+    if len(sep) > 1 and kwargs.get("engine") is None:
+        kwargs["engine"] = "python"
+    elif len(sep) > 1 and kwargs.get("engine") == "c":
+        warnings.warn(
+            "read_clipboard with regex separator does not work properly with c engine.",
+            stacklevel=find_stack_level(),
+        )
+
+    return read_csv(StringIO(text), sep=sep, dtype_backend=dtype_backend, **kwargs)
+
+
+def to_clipboard(
+    obj, excel: bool | None = True, sep: str | None = None, **kwargs
+) -> None:  # pragma: no cover
+    """
+    Attempt to write text representation of object to the system clipboard
+    The clipboard can be then pasted into Excel for example.
+
+    Parameters
+    ----------
+    obj : the object to write to the clipboard
+    excel : bool, defaults to True
+            if True, use the provided separator, writing in a csv
+            format for allowing easy pasting into excel.
+            if False, write a string representation of the object
+            to the clipboard
+    sep : optional, defaults to tab
+    other keywords are passed to to_csv
+
+    Notes
+    -----
+    Requirements for your platform
+      - Linux: xclip, or xsel (with PyQt4 modules)
+      - Windows:
+      - OS X:
+    """
+    encoding = kwargs.pop("encoding", "utf-8")
+
+    # testing if an invalid encoding is passed to clipboard
+    if encoding is not None and encoding.lower().replace("-", "") != "utf8":
+        raise ValueError("clipboard only supports utf-8 encoding")
+
+    from pandas.io.clipboard import clipboard_set
+
+    if excel is None:
+        excel = True
+
+    if excel:
+        try:
+            if sep is None:
+                sep = "\t"
+            buf = StringIO()
+
+            # clipboard_set (pyperclip) expects unicode
+            obj.to_csv(buf, sep=sep, encoding="utf-8", **kwargs)
+            text = buf.getvalue()
+
+            clipboard_set(text)
+            return
+        except TypeError:
+            warnings.warn(
+                "to_clipboard in excel mode requires a single character separator.",
+                stacklevel=find_stack_level(),
+            )
+    elif sep is not None:
+        warnings.warn(
+            "to_clipboard with excel=False ignores the sep argument.",
+            stacklevel=find_stack_level(),
+        )
+
+    if isinstance(obj, ABCDataFrame):
+        # str(df) has various unhelpful defaults, like truncation
+        with option_context("display.max_colwidth", None):
+            objstr = obj.to_string(**kwargs)
+    else:
+        objstr = str(obj)
+    clipboard_set(objstr)
diff --git a/pandas/pandas/io/common.py b/pandas/pandas/io/common.py
new file mode 100644
index 0000000000000..e0076eb486976
--- /dev/null
+++ b/pandas/pandas/io/common.py
@@ -0,0 +1,1284 @@
+"""Common IO api utilities"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+import codecs
+from collections import defaultdict
+from collections.abc import (
+    Hashable,
+    Mapping,
+    Sequence,
+)
+import dataclasses
+import functools
+import gzip
+from io import (
+    BufferedIOBase,
+    BytesIO,
+    RawIOBase,
+    StringIO,
+    TextIOBase,
+    TextIOWrapper,
+)
+import mmap
+import os
+from pathlib import Path
+import re
+import tarfile
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+    AnyStr,
+    DefaultDict,
+    Generic,
+    Literal,
+    TypeVar,
+    cast,
+    overload,
+)
+from urllib.parse import (
+    urljoin,
+    urlparse as parse_url,
+    uses_netloc,
+    uses_params,
+    uses_relative,
+)
+import warnings
+import zipfile
+
+from pandas._typing import (
+    BaseBuffer,
+    ReadCsvBuffer,
+)
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_file_like,
+    is_integer,
+    is_list_like,
+)
+from pandas.core.dtypes.generic import ABCMultiIndex
+
+from pandas.core.shared_docs import _shared_docs
+
+_VALID_URLS = set(uses_relative + uses_netloc + uses_params)
+_VALID_URLS.discard("")
+_RFC_3986_PATTERN = re.compile(r"^[A-Za-z][A-Za-z0-9+\-+.]*://")
+
+BaseBufferT = TypeVar("BaseBufferT", bound=BaseBuffer)
+
+
+if TYPE_CHECKING:
+    from types import TracebackType
+
+    from pandas._typing import (
+        CompressionDict,
+        CompressionOptions,
+        FilePath,
+        ReadBuffer,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+    from pandas import MultiIndex
+
+
+@dataclasses.dataclass
+class IOArgs:
+    """
+    Return value of io/common.py:_get_filepath_or_buffer.
+    """
+
+    filepath_or_buffer: str | BaseBuffer
+    encoding: str
+    mode: str
+    compression: CompressionDict
+    should_close: bool = False
+
+
+@dataclasses.dataclass
+class IOHandles(Generic[AnyStr]):
+    """
+    Return value of io/common.py:get_handle
+
+    Can be used as a context manager.
+
+    This is used to easily close created buffers and to handle corner cases when
+    TextIOWrapper is inserted.
+
+    handle: The file handle to be used.
+    created_handles: All file handles that are created by get_handle
+    is_wrapped: Whether a TextIOWrapper needs to be detached.
+    """
+
+    # handle might not implement the IO-interface
+    handle: IO[AnyStr]
+    compression: CompressionDict
+    created_handles: list[IO[bytes] | IO[str]] = dataclasses.field(default_factory=list)
+    is_wrapped: bool = False
+
+    def close(self) -> None:
+        """
+        Close all created buffers.
+
+        Note: If a TextIOWrapper was inserted, it is flushed and detached to
+        avoid closing the potentially user-created buffer.
+        """
+        if self.is_wrapped:
+            assert isinstance(self.handle, TextIOWrapper)
+            self.handle.flush()
+            self.handle.detach()
+            self.created_handles.remove(self.handle)
+        for handle in self.created_handles:
+            handle.close()
+        self.created_handles = []
+        self.is_wrapped = False
+
+    def __enter__(self) -> IOHandles[AnyStr]:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+
+def is_url(url: object) -> bool:
+    """
+    Check to see if a URL has a valid protocol.
+
+    Parameters
+    ----------
+    url : str or unicode
+
+    Returns
+    -------
+    isurl : bool
+        If `url` has a valid protocol return True otherwise False.
+    """
+    if not isinstance(url, str):
+        return False
+    return parse_url(url).scheme in _VALID_URLS
+
+
+@overload
+def _expand_user(filepath_or_buffer: str) -> str: ...
+
+
+@overload
+def _expand_user(filepath_or_buffer: BaseBufferT) -> BaseBufferT: ...
+
+
+def _expand_user(filepath_or_buffer: str | BaseBufferT) -> str | BaseBufferT:
+    """
+    Return the argument with an initial component of ~ or ~user
+    replaced by that user's home directory.
+
+    Parameters
+    ----------
+    filepath_or_buffer : object to be converted if possible
+
+    Returns
+    -------
+    expanded_filepath_or_buffer : an expanded filepath or the
+                                  input if not expandable
+    """
+    if isinstance(filepath_or_buffer, str):
+        return os.path.expanduser(filepath_or_buffer)
+    return filepath_or_buffer
+
+
+def validate_header_arg(header: object) -> None:
+    if header is None:
+        return
+    if is_integer(header):
+        header = cast(int, header)
+        if header < 0:
+            # GH 27779
+            raise ValueError(
+                "Passing negative integer to header is invalid. "
+                "For no header, use header=None instead"
+            )
+        return
+    if is_list_like(header, allow_sets=False):
+        header = cast(Sequence, header)
+        if not all(map(is_integer, header)):
+            raise ValueError("header must be integer or list of integers")
+        if any(i < 0 for i in header):
+            raise ValueError("cannot specify multi-index header with negative integers")
+        return
+    if is_bool(header):
+        raise TypeError(
+            "Passing a bool to header is invalid. Use header=None for no header or "
+            "header=int or list-like of ints to specify "
+            "the row(s) making up the column names"
+        )
+    # GH 16338
+    raise ValueError("header must be integer or list of integers")
+
+
+@overload
+def stringify_path(
+    filepath_or_buffer: FilePath, convert_file_like: bool = ...
+) -> str: ...
+
+
+@overload
+def stringify_path(
+    filepath_or_buffer: BaseBufferT, convert_file_like: bool = ...
+) -> BaseBufferT: ...
+
+
+def stringify_path(
+    filepath_or_buffer: FilePath | BaseBufferT,
+    convert_file_like: bool = False,
+) -> str | BaseBufferT:
+    """
+    Attempt to convert a path-like object to a string.
+
+    Parameters
+    ----------
+    filepath_or_buffer : object to be converted
+
+    Returns
+    -------
+    str_filepath_or_buffer : maybe a string version of the object
+
+    Notes
+    -----
+    Objects supporting the fspath protocol are coerced
+    according to its __fspath__ method.
+
+    Any other object is passed through unchanged, which includes bytes,
+    strings, buffers, or anything else that's not even path-like.
+    """
+    if not convert_file_like and is_file_like(filepath_or_buffer):
+        # GH 38125: some fsspec objects implement os.PathLike but have already opened a
+        # file. This prevents opening the file a second time. infer_compression calls
+        # this function with convert_file_like=True to infer the compression.
+        return cast(BaseBufferT, filepath_or_buffer)
+
+    if isinstance(filepath_or_buffer, os.PathLike):
+        filepath_or_buffer = filepath_or_buffer.__fspath__()
+    return _expand_user(filepath_or_buffer)
+
+
+def urlopen(*args: Any, **kwargs: Any) -> Any:
+    """
+    Lazy-import wrapper for stdlib urlopen, as that imports a big chunk of
+    the stdlib.
+    """
+    import urllib.request
+
+    return urllib.request.urlopen(*args, **kwargs)  # noqa: TID251
+
+
+def is_fsspec_url(url: FilePath | BaseBuffer) -> bool:
+    """
+    Returns true if the given URL looks like
+    something fsspec can handle
+    """
+    return (
+        isinstance(url, str)
+        and bool(_RFC_3986_PATTERN.match(url))
+        and not url.startswith(("http://", "https://"))
+    )
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    compression_options=_shared_docs["compression_options"] % "filepath_or_buffer",
+)
+def _get_filepath_or_buffer(
+    filepath_or_buffer: FilePath | BaseBuffer,
+    encoding: str = "utf-8",
+    compression: CompressionOptions | None = None,
+    mode: str = "r",
+    storage_options: StorageOptions | None = None,
+) -> IOArgs:
+    """
+    If the filepath_or_buffer is a url, translate and return the buffer.
+    Otherwise passthrough.
+
+    Parameters
+    ----------
+    filepath_or_buffer : a url, filepath (str or pathlib.Path),
+                         or buffer
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    encoding : the encoding to use to decode bytes, default is 'utf-8'
+    mode : str, optional
+
+    {storage_options}
+
+
+    Returns the dataclass IOArgs.
+    """
+    filepath_or_buffer = stringify_path(filepath_or_buffer)
+
+    # handle compression dict
+    compression_method, compression = get_compression_method(compression)
+    compression_method = infer_compression(filepath_or_buffer, compression_method)
+
+    # GH21227 internal compression is not used for non-binary handles.
+    if compression_method and hasattr(filepath_or_buffer, "write") and "b" not in mode:
+        warnings.warn(
+            "compression has no effect when passing a non-binary object as input.",
+            RuntimeWarning,
+            stacklevel=find_stack_level(),
+        )
+        compression_method = None
+
+    compression = dict(compression, method=compression_method)
+
+    # bz2 and xz do not write the byte order mark for utf-16 and utf-32
+    # print a warning when writing such files
+    if (
+        "w" in mode
+        and compression_method in ["bz2", "xz"]
+        and encoding in ["utf-16", "utf-32"]
+    ):
+        warnings.warn(
+            f"{compression} will not write the byte order mark for {encoding}",
+            UnicodeWarning,
+            stacklevel=find_stack_level(),
+        )
+
+    if "a" in mode and compression_method in ["zip", "tar"]:
+        # GH56778
+        warnings.warn(
+            "zip and tar do not support mode 'a' properly. "
+            "This combination will result in multiple files with same name "
+            "being added to the archive.",
+            RuntimeWarning,
+            stacklevel=find_stack_level(),
+        )
+
+    # Use binary mode when converting path-like objects to file-like objects (fsspec)
+    # except when text mode is explicitly requested. The original mode is returned if
+    # fsspec is not used.
+    fsspec_mode = mode
+    if "t" not in fsspec_mode and "b" not in fsspec_mode:
+        fsspec_mode += "b"
+
+    if isinstance(filepath_or_buffer, str) and is_url(filepath_or_buffer):
+        # TODO: fsspec can also handle HTTP via requests, but leaving this
+        # unchanged. using fsspec appears to break the ability to infer if the
+        # server responded with gzipped data
+        storage_options = storage_options or {}
+
+        # waiting until now for importing to match intended lazy logic of
+        # urlopen function defined elsewhere in this module
+        import urllib.request
+
+        # assuming storage_options is to be interpreted as headers
+        req_info = urllib.request.Request(filepath_or_buffer, headers=storage_options)
+        with urlopen(req_info) as req:
+            content_encoding = req.headers.get("Content-Encoding", None)
+            if content_encoding == "gzip":
+                # Override compression based on Content-Encoding header
+                compression = {"method": "gzip"}
+            reader = BytesIO(req.read())
+        return IOArgs(
+            filepath_or_buffer=reader,
+            encoding=encoding,
+            compression=compression,
+            should_close=True,
+            mode=fsspec_mode,
+        )
+
+    if is_fsspec_url(filepath_or_buffer):
+        assert isinstance(
+            filepath_or_buffer, str
+        )  # just to appease mypy for this branch
+        # two special-case s3-like protocols; these have special meaning in Hadoop,
+        # but are equivalent to just "s3" from fsspec's point of view
+        # cc #11071
+        if filepath_or_buffer.startswith("s3a://"):
+            filepath_or_buffer = filepath_or_buffer.replace("s3a://", "s3://")
+        if filepath_or_buffer.startswith("s3n://"):
+            filepath_or_buffer = filepath_or_buffer.replace("s3n://", "s3://")
+        fsspec = import_optional_dependency("fsspec")
+
+        # If botocore is installed we fallback to reading with anon=True
+        # to allow reads from public buckets
+        err_types_to_retry_with_anon: list[Any] = []
+        try:
+            import_optional_dependency("botocore")
+            from botocore.exceptions import (
+                ClientError,
+                NoCredentialsError,
+            )
+
+            err_types_to_retry_with_anon = [
+                ClientError,
+                NoCredentialsError,
+                PermissionError,
+            ]
+        except ImportError:
+            pass
+
+        try:
+            file_obj = fsspec.open(
+                filepath_or_buffer, mode=fsspec_mode, **(storage_options or {})
+            ).open()
+        # GH 34626 Reads from Public Buckets without Credentials needs anon=True
+        except tuple(err_types_to_retry_with_anon):
+            if storage_options is None:
+                storage_options = {"anon": True}
+            else:
+                # don't mutate user input.
+                storage_options = dict(storage_options)
+                storage_options["anon"] = True
+            file_obj = fsspec.open(
+                filepath_or_buffer, mode=fsspec_mode, **(storage_options or {})
+            ).open()
+
+        return IOArgs(
+            filepath_or_buffer=file_obj,
+            encoding=encoding,
+            compression=compression,
+            should_close=True,
+            mode=fsspec_mode,
+        )
+    elif storage_options:
+        raise ValueError(
+            "storage_options passed with file object or non-fsspec file path"
+        )
+
+    if isinstance(filepath_or_buffer, (str, bytes, mmap.mmap)):
+        return IOArgs(
+            filepath_or_buffer=_expand_user(filepath_or_buffer),
+            encoding=encoding,
+            compression=compression,
+            should_close=False,
+            mode=mode,
+        )
+
+    # is_file_like requires (read | write) & __iter__ but __iter__ is only
+    # needed for read_csv(engine=python)
+    if not (
+        hasattr(filepath_or_buffer, "read") or hasattr(filepath_or_buffer, "write")
+    ):
+        msg = f"Invalid file path or buffer object type: {type(filepath_or_buffer)}"
+        raise ValueError(msg)
+
+    return IOArgs(
+        filepath_or_buffer=filepath_or_buffer,
+        encoding=encoding,
+        compression=compression,
+        should_close=False,
+        mode=mode,
+    )
+
+
+def file_path_to_url(path: str) -> str:
+    """
+    converts an absolute native path to a FILE URL.
+
+    Parameters
+    ----------
+    path : a path in native format
+
+    Returns
+    -------
+    a valid FILE URL
+    """
+    # lazify expensive import (~30ms)
+    from urllib.request import pathname2url
+
+    return urljoin("file:", pathname2url(path))
+
+
+extension_to_compression = {
+    ".tar": "tar",
+    ".tar.gz": "tar",
+    ".tar.bz2": "tar",
+    ".tar.xz": "tar",
+    ".gz": "gzip",
+    ".bz2": "bz2",
+    ".zip": "zip",
+    ".xz": "xz",
+    ".zst": "zstd",
+}
+_supported_compressions = set(extension_to_compression.values())
+
+
+def get_compression_method(
+    compression: CompressionOptions,
+) -> tuple[str | None, CompressionDict]:
+    """
+    Simplifies a compression argument to a compression method string and
+    a mapping containing additional arguments.
+
+    Parameters
+    ----------
+    compression : str or mapping
+        If string, specifies the compression method. If mapping, value at key
+        'method' specifies compression method.
+
+    Returns
+    -------
+    tuple of ({compression method}, Optional[str]
+              {compression arguments}, Dict[str, Any])
+
+    Raises
+    ------
+    ValueError on mapping missing 'method' key
+    """
+    compression_method: str | None
+    if isinstance(compression, Mapping):
+        compression_args = dict(compression)
+        try:
+            compression_method = compression_args.pop("method")
+        except KeyError as err:
+            raise ValueError("If mapping, compression must have key 'method'") from err
+    else:
+        compression_args = {}
+        compression_method = compression
+    return compression_method, compression_args
+
+
+@doc(compression_options=_shared_docs["compression_options"] % "filepath_or_buffer")
+def infer_compression(
+    filepath_or_buffer: FilePath | BaseBuffer, compression: str | None
+) -> str | None:
+    """
+    Get the compression method for filepath_or_buffer. If compression='infer',
+    the inferred compression method is returned. Otherwise, the input
+    compression method is returned unchanged, unless it's invalid, in which
+    case an error is raised.
+
+    Parameters
+    ----------
+    filepath_or_buffer : str or file handle
+        File path or object.
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    Returns
+    -------
+    string or None
+
+    Raises
+    ------
+    ValueError on invalid compression specified.
+    """
+    if compression is None:
+        return None
+
+    # Infer compression
+    if compression == "infer":
+        # Convert all path types (e.g. pathlib.Path) to strings
+        if isinstance(filepath_or_buffer, str) and "::" in filepath_or_buffer:
+            # chained URLs contain ::
+            filepath_or_buffer = filepath_or_buffer.split("::")[0]
+        filepath_or_buffer = stringify_path(filepath_or_buffer, convert_file_like=True)
+        if not isinstance(filepath_or_buffer, str):
+            # Cannot infer compression of a buffer, assume no compression
+            return None
+
+        # Infer compression from the filename/URL extension
+        for extension, compression in extension_to_compression.items():
+            if filepath_or_buffer.lower().endswith(extension):
+                return compression
+        return None
+
+    # Compression has been specified. Check that it's valid
+    if compression in _supported_compressions:
+        return compression
+
+    valid = ["infer", None] + sorted(_supported_compressions)
+    msg = (
+        f"Unrecognized compression type: {compression}\n"
+        f"Valid compression types are {valid}"
+    )
+    raise ValueError(msg)
+
+
+def check_parent_directory(path: Path | str) -> None:
+    """
+    Check if parent directory of a file exists, raise OSError if it does not
+
+    Parameters
+    ----------
+    path: Path or str
+        Path to check parent directory of
+    """
+    parent = Path(path).parent
+    if not parent.is_dir():
+        raise OSError(rf"Cannot save file into a non-existent directory: '{parent}'")
+
+
+@overload
+def get_handle(
+    path_or_buf: FilePath | BaseBuffer,
+    mode: str,
+    *,
+    encoding: str | None = ...,
+    compression: CompressionOptions = ...,
+    memory_map: bool = ...,
+    is_text: Literal[False],
+    errors: str | None = ...,
+    storage_options: StorageOptions = ...,
+) -> IOHandles[bytes]: ...
+
+
+@overload
+def get_handle(
+    path_or_buf: FilePath | BaseBuffer,
+    mode: str,
+    *,
+    encoding: str | None = ...,
+    compression: CompressionOptions = ...,
+    memory_map: bool = ...,
+    is_text: Literal[True] = ...,
+    errors: str | None = ...,
+    storage_options: StorageOptions = ...,
+) -> IOHandles[str]: ...
+
+
+@overload
+def get_handle(
+    path_or_buf: FilePath | BaseBuffer,
+    mode: str,
+    *,
+    encoding: str | None = ...,
+    compression: CompressionOptions = ...,
+    memory_map: bool = ...,
+    is_text: bool = ...,
+    errors: str | None = ...,
+    storage_options: StorageOptions = ...,
+) -> IOHandles[str] | IOHandles[bytes]: ...
+
+
+@doc(compression_options=_shared_docs["compression_options"] % "path_or_buf")
+def get_handle(
+    path_or_buf: FilePath | BaseBuffer,
+    mode: str,
+    *,
+    encoding: str | None = None,
+    compression: CompressionOptions | None = None,
+    memory_map: bool = False,
+    is_text: bool = True,
+    errors: str | None = None,
+    storage_options: StorageOptions | None = None,
+) -> IOHandles[str] | IOHandles[bytes]:
+    """
+    Get file handle for given path/buffer and mode.
+
+    Parameters
+    ----------
+    path_or_buf : str or file handle
+        File path or object.
+    mode : str
+        Mode to open path_or_buf with.
+    encoding : str or None
+        Encoding to use.
+    {compression_options}
+
+           May be a dict with key 'method' as compression mode
+           and other keys as compression options if compression
+           mode is 'zip'.
+
+           Passing compression options as keys in dict is
+           supported for compression modes 'gzip', 'bz2', 'zstd' and 'zip'.
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    memory_map : bool, default False
+        See parsers._parser_params for more information. Only used by read_csv.
+    is_text : bool, default True
+        Whether the type of the content passed to the file/buffer is string or
+        bytes. This is not the same as `"b" not in mode`. If a string content is
+        passed to a binary file/buffer, a wrapper is inserted.
+    errors : str, default 'strict'
+        Specifies how encoding and decoding errors are to be handled.
+        See the errors argument for :func:`open` for a full list
+        of options.
+    storage_options: StorageOptions = None
+        Passed to _get_filepath_or_buffer
+
+    Returns the dataclass IOHandles
+    """
+    # Windows does not default to utf-8. Set to utf-8 for a consistent behavior
+    encoding = encoding or "utf-8"
+
+    errors = errors or "strict"
+
+    # read_csv does not know whether the buffer is opened in binary/text mode
+    if _is_binary_mode(path_or_buf, mode) and "b" not in mode:
+        mode += "b"
+
+    # validate encoding and errors
+    codecs.lookup(encoding)
+    if isinstance(errors, str):
+        codecs.lookup_error(errors)
+
+    # open URLs
+    ioargs = _get_filepath_or_buffer(
+        path_or_buf,
+        encoding=encoding,
+        compression=compression,
+        mode=mode,
+        storage_options=storage_options,
+    )
+
+    handle = ioargs.filepath_or_buffer
+    handles: list[BaseBuffer]
+
+    # memory mapping needs to be the first step
+    # only used for read_csv
+    handle, memory_map, handles = _maybe_memory_map(handle, memory_map)
+
+    is_path = isinstance(handle, str)
+    compression_args = dict(ioargs.compression)
+    compression = compression_args.pop("method")
+
+    # Only for write methods
+    if "r" not in mode and is_path:
+        check_parent_directory(str(handle))
+
+    if compression:
+        if compression != "zstd":
+            # compression libraries do not like an explicit text-mode
+            ioargs.mode = ioargs.mode.replace("t", "")
+        elif compression == "zstd" and "b" not in ioargs.mode:
+            # python-zstandard defaults to text mode, but we always expect
+            # compression libraries to use binary mode.
+            ioargs.mode += "b"
+
+        # GZ Compression
+        if compression == "gzip":
+            if isinstance(handle, str):
+                # error: Incompatible types in assignment (expression has type
+                # "GzipFile", variable has type "Union[str, BaseBuffer]")
+                handle = gzip.GzipFile(  # type: ignore[assignment]
+                    filename=handle,
+                    mode=ioargs.mode,
+                    **compression_args,
+                )
+            else:
+                handle = gzip.GzipFile(
+                    # No overload variant of "GzipFile" matches argument types
+                    # "Union[str, BaseBuffer]", "str", "Dict[str, Any]"
+                    fileobj=handle,  # type: ignore[call-overload]
+                    mode=ioargs.mode,
+                    **compression_args,
+                )
+
+        # BZ Compression
+        elif compression == "bz2":
+            import bz2
+
+            # Overload of "BZ2File" to handle pickle protocol 5
+            # "Union[str, BaseBuffer]", "str", "Dict[str, Any]"
+            handle = bz2.BZ2File(  # type: ignore[call-overload]
+                handle,
+                mode=ioargs.mode,
+                **compression_args,
+            )
+
+        # ZIP Compression
+        elif compression == "zip":
+            # error: Argument 1 to "_BytesZipFile" has incompatible type
+            # "Union[str, BaseBuffer]"; expected "Union[Union[str, PathLike[str]],
+            # ReadBuffer[bytes], WriteBuffer[bytes]]"
+            handle = _BytesZipFile(
+                handle,  # type: ignore[arg-type]
+                ioargs.mode,
+                **compression_args,
+            )
+            if handle.buffer.mode == "r":
+                handles.append(handle)
+                zip_names = handle.buffer.namelist()
+                if len(zip_names) == 1:
+                    handle = handle.buffer.open(zip_names.pop())
+                elif not zip_names:
+                    raise ValueError(f"Zero files found in ZIP file {path_or_buf}")
+                else:
+                    raise ValueError(
+                        "Multiple files found in ZIP file. "
+                        f"Only one file per ZIP: {zip_names}"
+                    )
+
+        # TAR Encoding
+        elif compression == "tar":
+            compression_args.setdefault("mode", ioargs.mode)
+            if isinstance(handle, str):
+                handle = _BytesTarFile(name=handle, **compression_args)
+            else:
+                # error: Argument "fileobj" to "_BytesTarFile" has incompatible
+                # type "BaseBuffer"; expected "Union[ReadBuffer[bytes],
+                # WriteBuffer[bytes], None]"
+                handle = _BytesTarFile(
+                    fileobj=handle,  # type: ignore[arg-type]
+                    **compression_args,
+                )
+            assert isinstance(handle, _BytesTarFile)
+            if "r" in handle.buffer.mode:
+                handles.append(handle)
+                files = handle.buffer.getnames()
+                if len(files) == 1:
+                    file = handle.buffer.extractfile(files[0])
+                    assert file is not None
+                    handle = file
+                elif not files:
+                    raise ValueError(f"Zero files found in TAR archive {path_or_buf}")
+                else:
+                    raise ValueError(
+                        "Multiple files found in TAR archive. "
+                        f"Only one file per TAR archive: {files}"
+                    )
+
+        # XZ Compression
+        elif compression == "xz":
+            # error: Argument 1 to "LZMAFile" has incompatible type "Union[str,
+            # BaseBuffer]"; expected "Optional[Union[Union[str, bytes, PathLike[str],
+            # PathLike[bytes]], IO[bytes]], None]"
+            import lzma
+
+            handle = lzma.LZMAFile(
+                handle,  # type: ignore[arg-type]
+                ioargs.mode,
+                **compression_args,
+            )
+
+        # Zstd Compression
+        elif compression == "zstd":
+            zstd = import_optional_dependency("zstandard")
+            if "r" in ioargs.mode:
+                open_args = {"dctx": zstd.ZstdDecompressor(**compression_args)}
+            else:
+                open_args = {"cctx": zstd.ZstdCompressor(**compression_args)}
+            handle = zstd.open(
+                handle,
+                mode=ioargs.mode,
+                **open_args,
+            )
+
+        # Unrecognized Compression
+        else:
+            msg = f"Unrecognized compression type: {compression}"
+            raise ValueError(msg)
+
+        assert not isinstance(handle, str)
+        handles.append(handle)
+
+    elif isinstance(handle, str):
+        # Check whether the filename is to be opened in binary mode.
+        # Binary mode does not support 'encoding' and 'newline'.
+        if ioargs.encoding and "b" not in ioargs.mode:
+            # Encoding
+            handle = open(
+                handle,
+                ioargs.mode,
+                encoding=ioargs.encoding,
+                errors=errors,
+                newline="",
+            )
+        else:
+            # Binary mode
+            handle = open(handle, ioargs.mode)
+        handles.append(handle)
+
+    # Convert BytesIO or file objects passed with an encoding
+    is_wrapped = False
+    if not is_text and ioargs.mode == "rb" and isinstance(handle, TextIOBase):
+        # not added to handles as it does not open/buffer resources
+        handle = _BytesIOWrapper(
+            handle,
+            encoding=ioargs.encoding,
+        )
+    elif is_text and (
+        compression or memory_map or _is_binary_mode(handle, ioargs.mode)
+    ):
+        if (
+            not hasattr(handle, "readable")
+            or not hasattr(handle, "writable")
+            or not hasattr(handle, "seekable")
+        ):
+            handle = _IOWrapper(handle)
+        # error: Value of type variable "_BufferT_co" of "TextIOWrapper" cannot
+        # be "_IOWrapper | BaseBuffer" [type-var]
+        handle = TextIOWrapper(
+            handle,  # type: ignore[type-var]
+            encoding=ioargs.encoding,
+            errors=errors,
+            newline="",
+        )
+        handles.append(handle)
+        # only marked as wrapped when the caller provided a handle
+        is_wrapped = not (
+            isinstance(ioargs.filepath_or_buffer, str) or ioargs.should_close
+        )
+
+    if "r" in ioargs.mode and not hasattr(handle, "read"):
+        raise TypeError(
+            "Expected file path name or file-like object, "
+            f"got {type(ioargs.filepath_or_buffer)} type"
+        )
+
+    handles.reverse()  # close the most recently added buffer first
+    if ioargs.should_close:
+        assert not isinstance(ioargs.filepath_or_buffer, str)
+        handles.append(ioargs.filepath_or_buffer)
+
+    return IOHandles(
+        # error: Argument "handle" to "IOHandles" has incompatible type
+        # "Union[TextIOWrapper, GzipFile, BaseBuffer, typing.IO[bytes],
+        # typing.IO[Any]]"; expected "pandas._typing.IO[Any]"
+        handle=handle,  # type: ignore[arg-type]
+        # error: Argument "created_handles" to "IOHandles" has incompatible type
+        # "List[BaseBuffer]"; expected "List[Union[IO[bytes], IO[str]]]"
+        created_handles=handles,  # type: ignore[arg-type]
+        is_wrapped=is_wrapped,
+        compression=ioargs.compression,
+    )
+
+
+# error: Definition of "__enter__" in base class "IOBase" is incompatible
+# with definition in base class "BinaryIO"
+class _BufferedWriter(BytesIO, ABC):  # type: ignore[misc]
+    """
+    Some objects do not support multiple .write() calls (TarFile and ZipFile).
+    This wrapper writes to the underlying buffer on close.
+    """
+
+    buffer = BytesIO()
+
+    @abstractmethod
+    def write_to_buffer(self) -> None: ...
+
+    def close(self) -> None:
+        if self.closed:
+            # already closed
+            return
+        if self.getbuffer().nbytes:
+            # write to buffer
+            self.seek(0)
+            with self.buffer:
+                self.write_to_buffer()
+        else:
+            self.buffer.close()
+        super().close()
+
+
+class _BytesTarFile(_BufferedWriter):
+    def __init__(
+        self,
+        name: str | None = None,
+        mode: Literal["r", "a", "w", "x"] = "r",
+        fileobj: ReadBuffer[bytes] | WriteBuffer[bytes] | None = None,
+        archive_name: str | None = None,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__()
+        self.archive_name = archive_name
+        self.name = name
+        # error: Incompatible types in assignment (expression has type "TarFile",
+        # base class "_BufferedWriter" defined the type as "BytesIO")
+        self.buffer: tarfile.TarFile = tarfile.TarFile.open(  # type: ignore[assignment]
+            name=name,
+            mode=self.extend_mode(mode),
+            fileobj=fileobj,
+            **kwargs,
+        )
+
+    def extend_mode(self, mode: str) -> str:
+        mode = mode.replace("b", "")
+        if mode != "w":
+            return mode
+        if self.name is not None:
+            suffix = Path(self.name).suffix
+            if suffix in (".gz", ".xz", ".bz2"):
+                mode = f"{mode}:{suffix[1:]}"
+        return mode
+
+    def infer_filename(self) -> str | None:
+        """
+        If an explicit archive_name is not given, we still want the file inside the zip
+        file not to be named something.tar, because that causes confusion (GH39465).
+        """
+        if self.name is None:
+            return None
+
+        filename = Path(self.name)
+        if filename.suffix == ".tar":
+            return filename.with_suffix("").name
+        elif filename.suffix in (".tar.gz", ".tar.bz2", ".tar.xz"):
+            return filename.with_suffix("").with_suffix("").name
+        return filename.name
+
+    def write_to_buffer(self) -> None:
+        # TarFile needs a non-empty string
+        archive_name = self.archive_name or self.infer_filename() or "tar"
+        tarinfo = tarfile.TarInfo(name=archive_name)
+        tarinfo.size = len(self.getvalue())
+        self.buffer.addfile(tarinfo, self)
+
+
+class _BytesZipFile(_BufferedWriter):
+    def __init__(
+        self,
+        file: FilePath | ReadBuffer[bytes] | WriteBuffer[bytes],
+        mode: str,
+        archive_name: str | None = None,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__()
+        mode = mode.replace("b", "")
+        self.archive_name = archive_name
+
+        kwargs.setdefault("compression", zipfile.ZIP_DEFLATED)
+        # error: Incompatible types in assignment (expression has type "ZipFile",
+        # base class "_BufferedWriter" defined the type as "BytesIO")
+        self.buffer: zipfile.ZipFile = zipfile.ZipFile(  # type: ignore[assignment]
+            file, mode, **kwargs
+        )
+
+    def infer_filename(self) -> str | None:
+        """
+        If an explicit archive_name is not given, we still want the file inside the zip
+        file not to be named something.zip, because that causes confusion (GH39465).
+        """
+        if isinstance(self.buffer.filename, (os.PathLike, str)):
+            filename = Path(self.buffer.filename)
+            if filename.suffix == ".zip":
+                return filename.with_suffix("").name
+            return filename.name
+        return None
+
+    def write_to_buffer(self) -> None:
+        # ZipFile needs a non-empty string
+        archive_name = self.archive_name or self.infer_filename() or "zip"
+        self.buffer.writestr(archive_name, self.getvalue())
+
+
+class _IOWrapper:
+    # TextIOWrapper is overly strict: it request that the buffer has seekable, readable,
+    # and writable. If we have a read-only buffer, we shouldn't need writable and vice
+    # versa. Some buffers, are seek/read/writ-able but they do not have the "-able"
+    # methods, e.g., tempfile.SpooledTemporaryFile.
+    # If a buffer does not have the above "-able" methods, we simple assume they are
+    # seek/read/writ-able.
+    def __init__(self, buffer: BaseBuffer) -> None:
+        self.buffer = buffer
+
+    def __getattr__(self, name: str) -> Any:
+        return getattr(self.buffer, name)
+
+    def readable(self) -> bool:
+        if hasattr(self.buffer, "readable"):
+            return self.buffer.readable()
+        return True
+
+    def seekable(self) -> bool:
+        if hasattr(self.buffer, "seekable"):
+            return self.buffer.seekable()
+        return True
+
+    def writable(self) -> bool:
+        if hasattr(self.buffer, "writable"):
+            return self.buffer.writable()
+        return True
+
+
+class _BytesIOWrapper:
+    # Wrapper that wraps a StringIO buffer and reads bytes from it
+    # Created for compat with pyarrow read_csv
+    def __init__(self, buffer: StringIO | TextIOBase, encoding: str = "utf-8") -> None:
+        self.buffer = buffer
+        self.encoding = encoding
+        # Because a character can be represented by more than 1 byte,
+        # it is possible that reading will produce more bytes than n
+        # We store the extra bytes in this overflow variable, and append the
+        # overflow to the front of the bytestring the next time reading is performed
+        self.overflow = b""
+
+    def __getattr__(self, attr: str) -> Any:
+        return getattr(self.buffer, attr)
+
+    def read(self, n: int | None = -1) -> bytes:
+        assert self.buffer is not None
+        bytestring = self.buffer.read(n).encode(self.encoding)
+        # When n=-1/n greater than remaining bytes: Read entire file/rest of file
+        combined_bytestring = self.overflow + bytestring
+        if n is None or n < 0 or n >= len(combined_bytestring):
+            self.overflow = b""
+            return combined_bytestring
+        else:
+            to_return = combined_bytestring[:n]
+            self.overflow = combined_bytestring[n:]
+            return to_return
+
+
+def _maybe_memory_map(
+    handle: str | BaseBuffer, memory_map: bool
+) -> tuple[str | BaseBuffer, bool, list[BaseBuffer]]:
+    """Try to memory map file/buffer."""
+    handles: list[BaseBuffer] = []
+    memory_map &= hasattr(handle, "fileno") or isinstance(handle, str)
+    if not memory_map:
+        return handle, memory_map, handles
+
+    # mmap used by only read_csv
+    handle = cast(ReadCsvBuffer, handle)
+
+    # need to open the file first
+    if isinstance(handle, str):
+        handle = open(handle, "rb")
+        handles.append(handle)
+
+    try:
+        # open mmap and adds *-able
+        # error: Argument 1 to "_IOWrapper" has incompatible type "mmap";
+        # expected "BaseBuffer"
+        wrapped = _IOWrapper(
+            mmap.mmap(
+                handle.fileno(),
+                0,
+                access=mmap.ACCESS_READ,  # type: ignore[arg-type]
+            )
+        )
+    finally:
+        for handle in reversed(handles):
+            # error: "BaseBuffer" has no attribute "close"
+            handle.close()  # type: ignore[attr-defined]
+
+    return wrapped, memory_map, [wrapped]
+
+
+def file_exists(filepath_or_buffer: FilePath | BaseBuffer) -> bool:
+    """Test whether file exists."""
+    exists = False
+    filepath_or_buffer = stringify_path(filepath_or_buffer)
+    if not isinstance(filepath_or_buffer, str):
+        return exists
+    try:
+        exists = os.path.exists(filepath_or_buffer)
+        # gh-5874: if the filepath is too long will raise here
+    except (TypeError, ValueError):
+        pass
+    return exists
+
+
+def _is_binary_mode(handle: FilePath | BaseBuffer, mode: str) -> bool:
+    """Whether the handle is opened in binary mode"""
+    # specified by user
+    if "t" in mode or "b" in mode:
+        return "b" in mode
+
+    # exceptions
+    text_classes = (
+        # classes that expect string but have 'b' in mode
+        codecs.StreamWriter,
+        codecs.StreamReader,
+        codecs.StreamReaderWriter,
+    )
+    if issubclass(type(handle), text_classes):
+        return False
+
+    return isinstance(handle, _get_binary_io_classes()) or "b" in getattr(
+        handle, "mode", mode
+    )
+
+
+@functools.lru_cache
+def _get_binary_io_classes() -> tuple[type, ...]:
+    """IO classes that that expect bytes"""
+    binary_classes: tuple[type, ...] = (BufferedIOBase, RawIOBase)
+
+    # python-zstandard doesn't use any of the builtin base classes; instead we
+    # have to use the `zstd.ZstdDecompressionReader` class for isinstance checks.
+    # Unfortunately `zstd.ZstdDecompressionReader` isn't exposed by python-zstandard
+    # so we have to get it from a `zstd.ZstdDecompressor` instance.
+    # See also https://github.com/indygreg/python-zstandard/pull/165.
+    zstd = import_optional_dependency("zstandard", errors="ignore")
+    if zstd is not None:
+        with zstd.ZstdDecompressor().stream_reader(b"") as reader:
+            binary_classes += (type(reader),)
+
+    return binary_classes
+
+
+def is_potential_multi_index(
+    columns: Sequence[Hashable] | MultiIndex,
+    index_col: bool | Sequence[int] | None = None,
+) -> bool:
+    """
+    Check whether or not the `columns` parameter
+    could be converted into a MultiIndex.
+
+    Parameters
+    ----------
+    columns : array-like
+        Object which may or may not be convertible into a MultiIndex
+    index_col : None, bool or list, optional
+        Column or columns to use as the (possibly hierarchical) index
+
+    Returns
+    -------
+    bool : Whether or not columns could become a MultiIndex
+    """
+    if index_col is None or isinstance(index_col, bool):
+        index_columns = set()
+    else:
+        index_columns = set(index_col)
+
+    return bool(
+        len(columns)
+        and not isinstance(columns, ABCMultiIndex)
+        and all(isinstance(c, tuple) for c in columns if c not in index_columns)
+    )
+
+
+def dedup_names(
+    names: Sequence[Hashable], is_potential_multiindex: bool
+) -> Sequence[Hashable]:
+    """
+    Rename column names if duplicates exist.
+
+    Currently the renaming is done by appending a period and an autonumeric,
+    but a custom pattern may be supported in the future.
+
+    Examples
+    --------
+    >>> dedup_names(["x", "y", "x", "x"], is_potential_multiindex=False)
+    ['x', 'y', 'x.1', 'x.2']
+    """
+    names = list(names)  # so we can index
+    counts: DefaultDict[Hashable, int] = defaultdict(int)
+
+    for i, col in enumerate(names):
+        cur_count = counts[col]
+
+        while cur_count > 0:
+            counts[col] = cur_count + 1
+
+            if is_potential_multiindex:
+                # for mypy
+                assert isinstance(col, tuple)
+                col = col[:-1] + (f"{col[-1]}.{cur_count}",)
+            else:
+                col = f"{col}.{cur_count}"
+            cur_count = counts[col]
+
+        names[i] = col
+        counts[col] = cur_count + 1
+
+    return names
diff --git a/pandas/pandas/io/excel/__init__.py b/pandas/pandas/io/excel/__init__.py
new file mode 100644
index 0000000000000..f13d7afa63d84
--- /dev/null
+++ b/pandas/pandas/io/excel/__init__.py
@@ -0,0 +1,19 @@
+from pandas.io.excel._base import (
+    ExcelFile,
+    ExcelWriter,
+    read_excel,
+)
+from pandas.io.excel._odswriter import ODSWriter as _ODSWriter
+from pandas.io.excel._openpyxl import OpenpyxlWriter as _OpenpyxlWriter
+from pandas.io.excel._util import register_writer
+from pandas.io.excel._xlsxwriter import XlsxWriter as _XlsxWriter
+
+__all__ = ["ExcelFile", "ExcelWriter", "read_excel"]
+
+
+register_writer(_OpenpyxlWriter)
+
+register_writer(_XlsxWriter)
+
+
+register_writer(_ODSWriter)
diff --git a/pandas/pandas/io/excel/_base.py b/pandas/pandas/io/excel/_base.py
new file mode 100644
index 0000000000000..460af65a60bf6
--- /dev/null
+++ b/pandas/pandas/io/excel/_base.py
@@ -0,0 +1,1860 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    Mapping,
+    Sequence,
+)
+import datetime
+from decimal import Decimal
+from functools import partial
+import os
+from textwrap import fill
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+    Generic,
+    Literal,
+    TypeVar,
+    Union,
+    cast,
+    overload,
+)
+import warnings
+import zipfile
+
+from pandas._config import config
+
+from pandas._libs import lib
+from pandas._libs.parsers import STR_NA_VALUES
+from pandas.compat._optional import (
+    get_version,
+    import_optional_dependency,
+)
+from pandas.errors import EmptyDataError
+from pandas.util._decorators import (
+    Appender,
+    doc,
+)
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_decimal,
+    is_file_like,
+    is_float,
+    is_integer,
+    is_list_like,
+)
+
+from pandas.core.frame import DataFrame
+from pandas.core.shared_docs import _shared_docs
+from pandas.util.version import Version
+
+from pandas.io.common import (
+    IOHandles,
+    get_handle,
+    stringify_path,
+    validate_header_arg,
+)
+from pandas.io.excel._util import (
+    fill_mi_header,
+    get_default_engine,
+    get_writer,
+    maybe_convert_usecols,
+    pop_header_name,
+)
+from pandas.io.parsers import TextParser
+from pandas.io.parsers.readers import validate_integer
+
+if TYPE_CHECKING:
+    from types import TracebackType
+
+    from pandas._typing import (
+        DtypeArg,
+        DtypeBackend,
+        ExcelWriterIfSheetExists,
+        FilePath,
+        HashableT,
+        IntStrT,
+        ReadBuffer,
+        Self,
+        SequenceNotStr,
+        StorageOptions,
+        WriteExcelBuffer,
+    )
+_read_excel_doc = (
+    """
+Read an Excel file into a ``pandas`` ``DataFrame``.
+
+Supports `xls`, `xlsx`, `xlsm`, `xlsb`, `odf`, `ods` and `odt` file extensions
+read from a local filesystem or URL. Supports an option to read
+a single sheet or a list of sheets.
+
+Parameters
+----------
+io : str, ExcelFile, xlrd.Book, path object, or file-like object
+    Any valid string path is acceptable. The string could be a URL. Valid
+    URL schemes include http, ftp, s3, and file. For file URLs, a host is
+    expected. A local file could be: ``file://localhost/path/to/table.xlsx``.
+
+    If you want to pass in a path object, pandas accepts any ``os.PathLike``.
+
+    By file-like object, we refer to objects with a ``read()`` method,
+    such as a file handle (e.g. via builtin ``open`` function)
+    or ``StringIO``.
+
+    .. deprecated:: 2.1.0
+        Passing byte strings is deprecated. To read from a
+        byte string, wrap it in a ``BytesIO`` object.
+sheet_name : str, int, list, or None, default 0
+    Strings are used for sheet names. Integers are used in zero-indexed
+    sheet positions (chart sheets do not count as a sheet position).
+    Lists of strings/integers are used to request multiple sheets.
+    When ``None``, will return a dictionary containing DataFrames for each sheet.
+
+    Available cases:
+
+    * Defaults to ``0``: 1st sheet as a `DataFrame`
+    * ``1``: 2nd sheet as a `DataFrame`
+    * ``"Sheet1"``: Load sheet with name "Sheet1"
+    * ``[0, 1, "Sheet5"]``: Load first, second and sheet named "Sheet5"
+      as a dict of `DataFrame`
+    * ``None``: Returns a dictionary containing DataFrames for each sheet..
+
+header : int, list of int, default 0
+    Row (0-indexed) to use for the column labels of the parsed
+    DataFrame. If a list of integers is passed those row positions will
+    be combined into a ``MultiIndex``. Use None if there is no header.
+names : array-like, default None
+    List of column names to use. If file contains no header row,
+    then you should explicitly pass header=None.
+index_col : int, str, list of int, default None
+    Column (0-indexed) to use as the row labels of the DataFrame.
+    Pass None if there is no such column.  If a list is passed,
+    those columns will be combined into a ``MultiIndex``.  If a
+    subset of data is selected with ``usecols``, index_col
+    is based on the subset.
+
+    Missing values will be forward filled to allow roundtripping with
+    ``to_excel`` for ``merged_cells=True``. To avoid forward filling the
+    missing values use ``set_index`` after reading the data instead of
+    ``index_col``.
+usecols : str, list-like, or callable, default None
+    * If None, then parse all columns.
+    * If str, then indicates comma separated list of Excel column letters
+      and column ranges (e.g. "A:E" or "A,C,E:F"). Ranges are inclusive of
+      both sides.
+    * If list of int, then indicates list of column numbers to be parsed
+      (0-indexed).
+    * If list of string, then indicates list of column names to be parsed.
+    * If callable, then evaluate each column name against it and parse the
+      column if the callable returns ``True``.
+
+    Returns a subset of the columns according to behavior above.
+dtype : Type name or dict of column -> type, default None
+    Data type for data or columns. E.g. {{'a': np.float64, 'b': np.int32}}
+    Use ``object`` to preserve data as stored in Excel and not interpret dtype,
+    which will necessarily result in ``object`` dtype.
+    If converters are specified, they will be applied INSTEAD
+    of dtype conversion.
+    If you use ``None``, it will infer the dtype of each column based on the data.
+engine : {{'openpyxl', 'calamine', 'odf', 'pyxlsb', 'xlrd'}}, default None
+    If io is not a buffer or path, this must be set to identify io.
+    Engine compatibility :
+
+    - ``openpyxl`` supports newer Excel file formats.
+    - ``calamine`` supports Excel (.xls, .xlsx, .xlsm, .xlsb)
+      and OpenDocument (.ods) file formats.
+    - ``odf`` supports OpenDocument file formats (.odf, .ods, .odt).
+    - ``pyxlsb`` supports Binary Excel files.
+    - ``xlrd`` supports old-style Excel files (.xls).
+
+    When ``engine=None``, the following logic will be used to determine the engine:
+
+    - If ``path_or_buffer`` is an OpenDocument format (.odf, .ods, .odt),
+      then `odf <https://pypi.org/project/odfpy/>`_ will be used.
+    - Otherwise if ``path_or_buffer`` is an xls format, ``xlrd`` will be used.
+    - Otherwise if ``path_or_buffer`` is in xlsb format, ``pyxlsb`` will be used.
+    - Otherwise ``openpyxl`` will be used.
+converters : dict, default None
+    Dict of functions for converting values in certain columns. Keys can
+    either be integers or column labels, values are functions that take one
+    input argument, the Excel cell content, and return the transformed
+    content.
+true_values : list, default None
+    Values to consider as True.
+false_values : list, default None
+    Values to consider as False.
+skiprows : list-like, int, or callable, optional
+    Line numbers to skip (0-indexed) or number of lines to skip (int) at the
+    start of the file. If callable, the callable function will be evaluated
+    against the row indices, returning True if the row should be skipped and
+    False otherwise. An example of a valid callable argument would be ``lambda
+    x: x in [0, 2]``.
+nrows : int, default None
+    Number of rows to parse.
+na_values : scalar, str, list-like, or dict, default None
+    Additional strings to recognize as NA/NaN. If dict passed, specific
+    per-column NA values. By default the following values are interpreted
+    as NaN: '"""
+    + fill("', '".join(sorted(STR_NA_VALUES)), 70, subsequent_indent="    ")
+    + """'.
+keep_default_na : bool, default True
+    Whether or not to include the default NaN values when parsing the data.
+    Depending on whether ``na_values`` is passed in, the behavior is as follows:
+
+    * If ``keep_default_na`` is True, and ``na_values`` are specified,
+      ``na_values`` is appended to the default NaN values used for parsing.
+    * If ``keep_default_na`` is True, and ``na_values`` are not specified, only
+      the default NaN values are used for parsing.
+    * If ``keep_default_na`` is False, and ``na_values`` are specified, only
+      the NaN values specified ``na_values`` are used for parsing.
+    * If ``keep_default_na`` is False, and ``na_values`` are not specified, no
+      strings will be parsed as NaN.
+
+    Note that if `na_filter` is passed in as False, the ``keep_default_na`` and
+    ``na_values`` parameters will be ignored.
+na_filter : bool, default True
+    Detect missing value markers (empty strings and the value of na_values). In
+    data without any NAs, passing ``na_filter=False`` can improve the
+    performance of reading a large file.
+verbose : bool, default False
+    Indicate number of NA values placed in non-numeric columns.
+parse_dates : bool, list-like, or dict, default False
+    The behavior is as follows:
+
+    * ``bool``. If True -> try parsing the index.
+    * ``list`` of int or names. e.g. If [1, 2, 3] -> try parsing columns 1, 2, 3
+      each as a separate date column.
+    * ``list`` of lists. e.g.  If [[1, 3]] -> combine columns 1 and 3 and parse as
+      a single date column.
+    * ``dict``, e.g. {{'foo' : [1, 3]}} -> parse columns 1, 3 as date and call
+      result 'foo'
+
+    If a column or index contains an unparsable date, the entire column or
+    index will be returned unaltered as an object data type. If you don`t want to
+    parse some cells as date just change their type in Excel to "Text".
+    For non-standard datetime parsing, use ``pd.to_datetime`` after ``pd.read_excel``.
+
+    Note: A fast-path exists for iso8601-formatted dates.
+date_format : str or dict of column -> format, default ``None``
+   If used in conjunction with ``parse_dates``, will parse dates according to this
+   format. For anything more complex,
+   please read in as ``object`` and then apply :func:`to_datetime` as-needed.
+
+   .. versionadded:: 2.0.0
+thousands : str, default None
+    Thousands separator for parsing string columns to numeric.  Note that
+    this parameter is only necessary for columns stored as TEXT in Excel,
+    any numeric columns will automatically be parsed, regardless of display
+    format.
+decimal : str, default '.'
+    Character to recognize as decimal point for parsing string columns to numeric.
+    Note that this parameter is only necessary for columns stored as TEXT in Excel,
+    any numeric columns will automatically be parsed, regardless of display
+    format.(e.g. use ',' for European data).
+
+    .. versionadded:: 1.4.0
+
+comment : str, default None
+    Comments out remainder of line. Pass a character or characters to this
+    argument to indicate comments in the input file. Any data between the
+    comment string and the end of the current line is ignored.
+skipfooter : int, default 0
+    Rows at the end to skip (0-indexed).
+{storage_options}
+
+dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+    Back-end data type applied to the resultant :class:`DataFrame`
+    (still experimental). If not specified, the default behavior
+    is to not use nullable data types. If specified, the behavior
+    is as follows:
+
+    * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+    * ``"pyarrow"``: returns pyarrow-backed nullable
+      :class:`ArrowDtype` :class:`DataFrame`
+
+    .. versionadded:: 2.0
+
+engine_kwargs : dict, optional
+    Arbitrary keyword arguments passed to excel engine.
+
+Returns
+-------
+DataFrame or dict of DataFrames
+    DataFrame from the passed in Excel file. See notes in sheet_name
+    argument for more information on when a dict of DataFrames is returned.
+
+See Also
+--------
+DataFrame.to_excel : Write DataFrame to an Excel file.
+DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file.
+read_csv : Read a comma-separated values (csv) file into DataFrame.
+read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+Notes
+-----
+For specific information on the methods used for each Excel engine, refer to the pandas
+:ref:`user guide <io.excel_reader>`
+
+Examples
+--------
+The file can be read using the file name as string or an open file object:
+
+>>> pd.read_excel('tmp.xlsx', index_col=0)  # doctest: +SKIP
+       Name  Value
+0   string1      1
+1   string2      2
+2  #Comment      3
+
+>>> pd.read_excel(open('tmp.xlsx', 'rb'),
+...               sheet_name='Sheet3')  # doctest: +SKIP
+   Unnamed: 0      Name  Value
+0           0   string1      1
+1           1   string2      2
+2           2  #Comment      3
+
+Index and header can be specified via the `index_col` and `header` arguments
+
+>>> pd.read_excel('tmp.xlsx', index_col=None, header=None)  # doctest: +SKIP
+     0         1      2
+0  NaN      Name  Value
+1  0.0   string1      1
+2  1.0   string2      2
+3  2.0  #Comment      3
+
+Column types are inferred but can be explicitly specified
+
+>>> pd.read_excel('tmp.xlsx', index_col=0,
+...               dtype={{'Name': str, 'Value': float}})  # doctest: +SKIP
+       Name  Value
+0   string1    1.0
+1   string2    2.0
+2  #Comment    3.0
+
+True, False, and NA values, and thousands separators have defaults,
+but can be explicitly specified, too. Supply the values you would like
+as strings or lists of strings!
+
+>>> pd.read_excel('tmp.xlsx', index_col=0,
+...               na_values=['string1', 'string2'])  # doctest: +SKIP
+       Name  Value
+0       NaN      1
+1       NaN      2
+2  #Comment      3
+
+Comment lines in the excel input file can be skipped using the
+``comment`` kwarg.
+
+>>> pd.read_excel('tmp.xlsx', index_col=0, comment='#')  # doctest: +SKIP
+      Name  Value
+0  string1    1.0
+1  string2    2.0
+2     None    NaN
+"""
+)
+
+
+@overload
+def read_excel(
+    io,
+    # sheet name is str or int -> DataFrame
+    sheet_name: str | int = ...,
+    *,
+    header: int | Sequence[int] | None = ...,
+    names: SequenceNotStr[Hashable] | range | None = ...,
+    index_col: int | str | Sequence[int] | None = ...,
+    usecols: int
+    | str
+    | Sequence[int]
+    | Sequence[str]
+    | Callable[[HashableT], bool]
+    | None = ...,
+    dtype: DtypeArg | None = ...,
+    engine: Literal["xlrd", "openpyxl", "odf", "pyxlsb", "calamine"] | None = ...,
+    converters: dict[str, Callable] | dict[int, Callable] | None = ...,
+    true_values: Iterable[Hashable] | None = ...,
+    false_values: Iterable[Hashable] | None = ...,
+    skiprows: Sequence[int] | int | Callable[[int], object] | None = ...,
+    nrows: int | None = ...,
+    na_values=...,
+    keep_default_na: bool = ...,
+    na_filter: bool = ...,
+    verbose: bool = ...,
+    parse_dates: list | dict | bool = ...,
+    date_format: dict[Hashable, str] | str | None = ...,
+    thousands: str | None = ...,
+    decimal: str = ...,
+    comment: str | None = ...,
+    skipfooter: int = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> DataFrame: ...
+
+
+@overload
+def read_excel(
+    io,
+    # sheet name is list or None -> dict[IntStrT, DataFrame]
+    sheet_name: list[IntStrT] | None,
+    *,
+    header: int | Sequence[int] | None = ...,
+    names: SequenceNotStr[Hashable] | range | None = ...,
+    index_col: int | str | Sequence[int] | None = ...,
+    usecols: int
+    | str
+    | Sequence[int]
+    | Sequence[str]
+    | Callable[[HashableT], bool]
+    | None = ...,
+    dtype: DtypeArg | None = ...,
+    engine: Literal["xlrd", "openpyxl", "odf", "pyxlsb", "calamine"] | None = ...,
+    converters: dict[str, Callable] | dict[int, Callable] | None = ...,
+    true_values: Iterable[Hashable] | None = ...,
+    false_values: Iterable[Hashable] | None = ...,
+    skiprows: Sequence[int] | int | Callable[[int], object] | None = ...,
+    nrows: int | None = ...,
+    na_values=...,
+    keep_default_na: bool = ...,
+    na_filter: bool = ...,
+    verbose: bool = ...,
+    parse_dates: list | dict | bool = ...,
+    date_format: dict[Hashable, str] | str | None = ...,
+    thousands: str | None = ...,
+    decimal: str = ...,
+    comment: str | None = ...,
+    skipfooter: int = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> dict[IntStrT, DataFrame]: ...
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+@Appender(_read_excel_doc)
+def read_excel(
+    io,
+    sheet_name: str | int | list[IntStrT] | None = 0,
+    *,
+    header: int | Sequence[int] | None = 0,
+    names: SequenceNotStr[Hashable] | range | None = None,
+    index_col: int | str | Sequence[int] | None = None,
+    usecols: int
+    | str
+    | Sequence[int]
+    | Sequence[str]
+    | Callable[[HashableT], bool]
+    | None = None,
+    dtype: DtypeArg | None = None,
+    engine: Literal["xlrd", "openpyxl", "odf", "pyxlsb", "calamine"] | None = None,
+    converters: dict[str, Callable] | dict[int, Callable] | None = None,
+    true_values: Iterable[Hashable] | None = None,
+    false_values: Iterable[Hashable] | None = None,
+    skiprows: Sequence[int] | int | Callable[[int], object] | None = None,
+    nrows: int | None = None,
+    na_values=None,
+    keep_default_na: bool = True,
+    na_filter: bool = True,
+    verbose: bool = False,
+    parse_dates: list | dict | bool = False,
+    date_format: dict[Hashable, str] | str | None = None,
+    thousands: str | None = None,
+    decimal: str = ".",
+    comment: str | None = None,
+    skipfooter: int = 0,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    engine_kwargs: dict | None = None,
+) -> DataFrame | dict[IntStrT, DataFrame]:
+    check_dtype_backend(dtype_backend)
+    should_close = False
+    if engine_kwargs is None:
+        engine_kwargs = {}
+
+    if not isinstance(io, ExcelFile):
+        should_close = True
+        io = ExcelFile(
+            io,
+            storage_options=storage_options,
+            engine=engine,
+            engine_kwargs=engine_kwargs,
+        )
+    elif engine and engine != io.engine:
+        raise ValueError(
+            "Engine should not be specified when passing "
+            "an ExcelFile - ExcelFile already has the engine set"
+        )
+
+    try:
+        data = io.parse(
+            sheet_name=sheet_name,
+            header=header,
+            names=names,
+            index_col=index_col,
+            usecols=usecols,
+            dtype=dtype,
+            converters=converters,
+            true_values=true_values,
+            false_values=false_values,
+            skiprows=skiprows,
+            nrows=nrows,
+            na_values=na_values,
+            keep_default_na=keep_default_na,
+            na_filter=na_filter,
+            verbose=verbose,
+            parse_dates=parse_dates,
+            date_format=date_format,
+            thousands=thousands,
+            decimal=decimal,
+            comment=comment,
+            skipfooter=skipfooter,
+            dtype_backend=dtype_backend,
+        )
+    finally:
+        # make sure to close opened file handles
+        if should_close:
+            io.close()
+    return data
+
+
+_WorkbookT = TypeVar("_WorkbookT")
+
+
+class BaseExcelReader(Generic[_WorkbookT]):
+    book: _WorkbookT
+
+    def __init__(
+        self,
+        filepath_or_buffer,
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        if engine_kwargs is None:
+            engine_kwargs = {}
+
+        self.handles = IOHandles(
+            handle=filepath_or_buffer, compression={"method": None}
+        )
+        if not isinstance(filepath_or_buffer, (ExcelFile, self._workbook_class)):
+            self.handles = get_handle(
+                filepath_or_buffer, "rb", storage_options=storage_options, is_text=False
+            )
+
+        if isinstance(self.handles.handle, self._workbook_class):
+            self.book = self.handles.handle
+        elif hasattr(self.handles.handle, "read"):
+            # N.B. xlrd.Book has a read attribute too
+            self.handles.handle.seek(0)
+            try:
+                self.book = self.load_workbook(self.handles.handle, engine_kwargs)
+            except Exception:
+                self.close()
+                raise
+        else:
+            raise ValueError(
+                "Must explicitly set engine if not passing in buffer or path for io."
+            )
+
+    @property
+    def _workbook_class(self) -> type[_WorkbookT]:
+        raise NotImplementedError
+
+    def load_workbook(self, filepath_or_buffer, engine_kwargs) -> _WorkbookT:
+        raise NotImplementedError
+
+    def close(self) -> None:
+        if hasattr(self, "book"):
+            if hasattr(self.book, "close"):
+                # pyxlsb: opens a TemporaryFile
+                # openpyxl: https://stackoverflow.com/questions/31416842/
+                #     openpyxl-does-not-close-excel-workbook-in-read-only-mode
+                self.book.close()
+            elif hasattr(self.book, "release_resources"):
+                # xlrd
+                # https://github.com/python-excel/xlrd/blob/2.0.1/xlrd/book.py#L548
+                self.book.release_resources()
+        self.handles.close()
+
+    @property
+    def sheet_names(self) -> list[str]:
+        raise NotImplementedError
+
+    def get_sheet_by_name(self, name: str):
+        raise NotImplementedError
+
+    def get_sheet_by_index(self, index: int):
+        raise NotImplementedError
+
+    def get_sheet_data(self, sheet, rows: int | None = None):
+        raise NotImplementedError
+
+    def raise_if_bad_sheet_by_index(self, index: int) -> None:
+        n_sheets = len(self.sheet_names)
+        if index >= n_sheets:
+            raise ValueError(
+                f"Worksheet index {index} is invalid, {n_sheets} worksheets found"
+            )
+
+    def raise_if_bad_sheet_by_name(self, name: str) -> None:
+        if name not in self.sheet_names:
+            raise ValueError(f"Worksheet named '{name}' not found")
+
+    def _check_skiprows_func(
+        self,
+        skiprows: Callable,
+        rows_to_use: int,
+    ) -> int:
+        """
+        Determine how many file rows are required to obtain `nrows` data
+        rows when `skiprows` is a function.
+
+        Parameters
+        ----------
+        skiprows : function
+            The function passed to read_excel by the user.
+        rows_to_use : int
+            The number of rows that will be needed for the header and
+            the data.
+
+        Returns
+        -------
+        int
+        """
+        i = 0
+        rows_used_so_far = 0
+        while rows_used_so_far < rows_to_use:
+            if not skiprows(i):
+                rows_used_so_far += 1
+            i += 1
+        return i
+
+    def _calc_rows(
+        self,
+        header: int | Sequence[int] | None,
+        index_col: int | Sequence[int] | None,
+        skiprows: Sequence[int] | int | Callable[[int], object] | None,
+        nrows: int | None,
+    ) -> int | None:
+        """
+        If nrows specified, find the number of rows needed from the
+        file, otherwise return None.
+
+
+        Parameters
+        ----------
+        header : int, list of int, or None
+            See read_excel docstring.
+        index_col : int, str, list of int, or None
+            See read_excel docstring.
+        skiprows : list-like, int, callable, or None
+            See read_excel docstring.
+        nrows : int or None
+            See read_excel docstring.
+
+        Returns
+        -------
+        int or None
+        """
+        if nrows is None:
+            return None
+        if header is None:
+            header_rows = 1
+        elif is_integer(header):
+            header = cast(int, header)
+            header_rows = 1 + header
+        else:
+            header = cast(Sequence, header)
+            header_rows = 1 + header[-1]
+        # If there is a MultiIndex header and an index then there is also
+        # a row containing just the index name(s)
+        if is_list_like(header) and index_col is not None:
+            header = cast(Sequence, header)
+            if len(header) > 1:
+                header_rows += 1
+        if skiprows is None:
+            return header_rows + nrows
+        if is_integer(skiprows):
+            skiprows = cast(int, skiprows)
+            return header_rows + nrows + skiprows
+        if is_list_like(skiprows):
+
+            def f(skiprows: Sequence, x: int) -> bool:
+                return x in skiprows
+
+            skiprows = cast(Sequence, skiprows)
+            return self._check_skiprows_func(partial(f, skiprows), header_rows + nrows)
+        if callable(skiprows):
+            return self._check_skiprows_func(
+                skiprows,
+                header_rows + nrows,
+            )
+        # else unexpected skiprows type: read_excel will not optimize
+        # the number of rows read from file
+        return None
+
+    def parse(
+        self,
+        sheet_name: str | int | list[int] | list[str] | None = 0,
+        header: int | Sequence[int] | None = 0,
+        names: SequenceNotStr[Hashable] | range | None = None,
+        index_col: int | Sequence[int] | None = None,
+        usecols=None,
+        dtype: DtypeArg | None = None,
+        true_values: Iterable[Hashable] | None = None,
+        false_values: Iterable[Hashable] | None = None,
+        skiprows: Sequence[int] | int | Callable[[int], object] | None = None,
+        nrows: int | None = None,
+        na_values=None,
+        verbose: bool = False,
+        parse_dates: list | dict | bool = False,
+        date_format: dict[Hashable, str] | str | None = None,
+        thousands: str | None = None,
+        decimal: str = ".",
+        comment: str | None = None,
+        skipfooter: int = 0,
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+        **kwds,
+    ):
+        validate_header_arg(header)
+        validate_integer("nrows", nrows)
+
+        ret_dict = False
+
+        # Keep sheetname to maintain backwards compatibility.
+        sheets: list[int] | list[str]
+        if isinstance(sheet_name, list):
+            sheets = sheet_name
+            ret_dict = True
+        elif sheet_name is None:
+            sheets = self.sheet_names
+            ret_dict = True
+        elif isinstance(sheet_name, str):
+            sheets = [sheet_name]
+        else:
+            sheets = [sheet_name]
+
+        # handle same-type duplicates.
+        sheets = cast(Union[list[int], list[str]], list(dict.fromkeys(sheets).keys()))
+
+        output = {}
+
+        last_sheetname = None
+        for asheetname in sheets:
+            last_sheetname = asheetname
+            if verbose:
+                print(f"Reading sheet {asheetname}")
+
+            if isinstance(asheetname, str):
+                sheet = self.get_sheet_by_name(asheetname)
+            else:  # assume an integer if not a string
+                sheet = self.get_sheet_by_index(asheetname)
+
+            file_rows_needed = self._calc_rows(header, index_col, skiprows, nrows)
+            data = self.get_sheet_data(sheet, file_rows_needed)
+            if hasattr(sheet, "close"):
+                # pyxlsb opens two TemporaryFiles
+                sheet.close()
+            usecols = maybe_convert_usecols(usecols)
+
+            if not data:
+                output[asheetname] = DataFrame()
+                continue
+
+            output = self._parse_sheet(
+                data=data,
+                output=output,
+                asheetname=asheetname,
+                header=header,
+                names=names,
+                index_col=index_col,
+                usecols=usecols,
+                dtype=dtype,
+                skiprows=skiprows,
+                nrows=nrows,
+                true_values=true_values,
+                false_values=false_values,
+                na_values=na_values,
+                parse_dates=parse_dates,
+                date_format=date_format,
+                thousands=thousands,
+                decimal=decimal,
+                comment=comment,
+                skipfooter=skipfooter,
+                dtype_backend=dtype_backend,
+                **kwds,
+            )
+
+        if last_sheetname is None:
+            raise ValueError("Sheet name is an empty list")
+
+        if ret_dict:
+            return output
+        else:
+            return output[last_sheetname]
+
+    def _parse_sheet(
+        self,
+        data: list,
+        output: dict,
+        asheetname: str | int | None = None,
+        header: int | Sequence[int] | None = 0,
+        names: SequenceNotStr[Hashable] | range | None = None,
+        index_col: int | Sequence[int] | None = None,
+        usecols=None,
+        dtype: DtypeArg | None = None,
+        skiprows: Sequence[int] | int | Callable[[int], object] | None = None,
+        nrows: int | None = None,
+        true_values: Iterable[Hashable] | None = None,
+        false_values: Iterable[Hashable] | None = None,
+        na_values=None,
+        parse_dates: list | dict | bool = False,
+        date_format: dict[Hashable, str] | str | None = None,
+        thousands: str | None = None,
+        decimal: str = ".",
+        comment: str | None = None,
+        skipfooter: int = 0,
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+        **kwds,
+    ):
+        is_list_header = False
+        is_len_one_list_header = False
+        if is_list_like(header):
+            assert isinstance(header, Sequence)
+            is_list_header = True
+            if len(header) == 1:
+                is_len_one_list_header = True
+
+        if is_len_one_list_header:
+            header = cast(Sequence[int], header)[0]
+
+        # forward fill and pull out names for MultiIndex column
+        header_names = None
+        if header is not None and is_list_like(header):
+            assert isinstance(header, Sequence)
+
+            header_names = []
+            control_row = [True] * len(data[0])
+
+            for row in header:
+                if is_integer(skiprows):
+                    assert isinstance(skiprows, int)
+                    row += skiprows
+
+                if row > len(data) - 1:
+                    raise ValueError(
+                        f"header index {row} exceeds maximum index "
+                        f"{len(data) - 1} of data.",
+                    )
+
+                data[row], control_row = fill_mi_header(data[row], control_row)
+
+                if index_col is not None:
+                    header_name, _ = pop_header_name(data[row], index_col)
+                    header_names.append(header_name)
+
+        # If there is a MultiIndex header and an index then there is also
+        # a row containing just the index name(s)
+        has_index_names = False
+        if is_list_header and not is_len_one_list_header and index_col is not None:
+            index_col_set: set[int]
+            if isinstance(index_col, int):
+                index_col_set = {index_col}
+            else:
+                assert isinstance(index_col, Sequence)
+                index_col_set = set(index_col)
+
+            # We have to handle mi without names. If any of the entries in the data
+            # columns are not empty, this is a regular row
+            assert isinstance(header, Sequence)
+            if len(header) < len(data):
+                potential_index_names = data[len(header)]
+                has_index_names = all(
+                    x == "" or x is None
+                    for i, x in enumerate(potential_index_names)
+                    if not control_row[i] and i not in index_col_set
+                )
+
+        if is_list_like(index_col):
+            # Forward fill values for MultiIndex index.
+            if header is None:
+                offset = 0
+            elif isinstance(header, int):
+                offset = 1 + header
+            else:
+                offset = 1 + max(header)
+
+            # GH34673: if MultiIndex names present and not defined in the header,
+            # offset needs to be incremented so that forward filling starts
+            # from the first MI value instead of the name
+            if has_index_names:
+                offset += 1
+
+            # Check if we have an empty dataset
+            # before trying to collect data.
+            if offset < len(data):
+                assert isinstance(index_col, Sequence)
+
+                for col in index_col:
+                    last = data[offset][col]
+
+                    for row in range(offset + 1, len(data)):
+                        if data[row][col] == "" or data[row][col] is None:
+                            data[row][col] = last
+                        else:
+                            last = data[row][col]
+
+        # GH 12292 : error when read one empty column from excel file
+        try:
+            parser = TextParser(
+                data,
+                names=names,
+                header=header,
+                index_col=index_col,
+                has_index_names=has_index_names,
+                dtype=dtype,
+                true_values=true_values,
+                false_values=false_values,
+                skiprows=skiprows,
+                nrows=nrows,
+                na_values=na_values,
+                skip_blank_lines=False,  # GH 39808
+                parse_dates=parse_dates,
+                date_format=date_format,
+                thousands=thousands,
+                decimal=decimal,
+                comment=comment,
+                skipfooter=skipfooter,
+                usecols=usecols,
+                dtype_backend=dtype_backend,
+                **kwds,
+            )
+
+            output[asheetname] = parser.read(nrows=nrows)
+
+            if header_names:
+                output[asheetname].columns = output[asheetname].columns.set_names(
+                    header_names
+                )
+
+        except EmptyDataError:
+            # No Data, return an empty DataFrame
+            output[asheetname] = DataFrame()
+
+        except Exception as err:
+            err.args = (f"{err.args[0]} (sheet: {asheetname})", *err.args[1:])
+            raise err
+
+        return output
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+class ExcelWriter(Generic[_WorkbookT]):
+    """
+    Class for writing DataFrame objects into excel sheets.
+
+    Default is to use:
+
+    * `xlsxwriter <https://pypi.org/project/XlsxWriter/>`__ for xlsx files if xlsxwriter
+      is installed otherwise `openpyxl <https://pypi.org/project/openpyxl/>`__
+    * `odf <https://pypi.org/project/odfpy/>`__ for ods files
+
+    See :meth:`DataFrame.to_excel` for typical usage.
+
+    The writer should be used as a context manager. Otherwise, call `close()` to save
+    and close any opened file handles.
+
+    Parameters
+    ----------
+    path : str or typing.BinaryIO
+        Path to xls or xlsx or ods file.
+    engine : str (optional)
+        Engine to use for writing. If None, defaults to
+        ``io.excel.<extension>.writer``.  NOTE: can only be passed as a keyword
+        argument.
+    date_format : str, default None
+        Format string for dates written into Excel files (e.g. 'YYYY-MM-DD').
+    datetime_format : str, default None
+        Format string for datetime objects written into Excel files.
+        (e.g. 'YYYY-MM-DD HH:MM:SS').
+    mode : {{'w', 'a'}}, default 'w'
+        File mode to use (write or append). Append does not work with fsspec URLs.
+    {storage_options}
+
+    if_sheet_exists : {{'error', 'new', 'replace', 'overlay'}}, default 'error'
+        How to behave when trying to write to a sheet that already
+        exists (append mode only).
+
+        * error: raise a ValueError.
+        * new: Create a new sheet, with a name determined by the engine.
+        * replace: Delete the contents of the sheet before writing to it.
+        * overlay: Write contents to the existing sheet without first removing,
+          but possibly over top of, the existing contents.
+
+        .. versionadded:: 1.3.0
+
+        .. versionchanged:: 1.4.0
+
+           Added ``overlay`` option
+
+    engine_kwargs : dict, optional
+        Keyword arguments to be passed into the engine. These will be passed to
+        the following functions of the respective engines:
+
+        * xlsxwriter: ``xlsxwriter.Workbook(file, **engine_kwargs)``
+        * openpyxl (write mode): ``openpyxl.Workbook(**engine_kwargs)``
+        * openpyxl (append mode): ``openpyxl.load_workbook(file, **engine_kwargs)``
+        * odf: ``odf.opendocument.OpenDocumentSpreadsheet(**engine_kwargs)``
+
+        .. versionadded:: 1.3.0
+
+    See Also
+    --------
+    read_excel : Read an Excel sheet values (xlsx) file into DataFrame.
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+    read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+    Notes
+    -----
+    For compatibility with CSV writers, ExcelWriter serializes lists
+    and dicts to strings before writing.
+
+    Examples
+    --------
+    Default usage:
+
+    >>> df = pd.DataFrame([["ABC", "XYZ"]], columns=["Foo", "Bar"])  # doctest: +SKIP
+    >>> with pd.ExcelWriter("path_to_file.xlsx") as writer:
+    ...     df.to_excel(writer)  # doctest: +SKIP
+
+    To write to separate sheets in a single file:
+
+    >>> df1 = pd.DataFrame([["AAA", "BBB"]], columns=["Spam", "Egg"])  # doctest: +SKIP
+    >>> df2 = pd.DataFrame([["ABC", "XYZ"]], columns=["Foo", "Bar"])  # doctest: +SKIP
+    >>> with pd.ExcelWriter("path_to_file.xlsx") as writer:
+    ...     df1.to_excel(writer, sheet_name="Sheet1")  # doctest: +SKIP
+    ...     df2.to_excel(writer, sheet_name="Sheet2")  # doctest: +SKIP
+
+    You can set the date format or datetime format:
+
+    >>> from datetime import date, datetime  # doctest: +SKIP
+    >>> df = pd.DataFrame(
+    ...     [
+    ...         [date(2014, 1, 31), date(1999, 9, 24)],
+    ...         [datetime(1998, 5, 26, 23, 33, 4), datetime(2014, 2, 28, 13, 5, 13)],
+    ...     ],
+    ...     index=["Date", "Datetime"],
+    ...     columns=["X", "Y"],
+    ... )  # doctest: +SKIP
+    >>> with pd.ExcelWriter(
+    ...     "path_to_file.xlsx",
+    ...     date_format="YYYY-MM-DD",
+    ...     datetime_format="YYYY-MM-DD HH:MM:SS",
+    ... ) as writer:
+    ...     df.to_excel(writer)  # doctest: +SKIP
+
+    You can also append to an existing Excel file:
+
+    >>> with pd.ExcelWriter("path_to_file.xlsx", mode="a", engine="openpyxl") as writer:
+    ...     df.to_excel(writer, sheet_name="Sheet3")  # doctest: +SKIP
+
+    Here, the `if_sheet_exists` parameter can be set to replace a sheet if it
+    already exists:
+
+    >>> with pd.ExcelWriter(
+    ...     "path_to_file.xlsx",
+    ...     mode="a",
+    ...     engine="openpyxl",
+    ...     if_sheet_exists="replace",
+    ... ) as writer:
+    ...     df.to_excel(writer, sheet_name="Sheet1")  # doctest: +SKIP
+
+    You can also write multiple DataFrames to a single sheet. Note that the
+    ``if_sheet_exists`` parameter needs to be set to ``overlay``:
+
+    >>> with pd.ExcelWriter(
+    ...     "path_to_file.xlsx",
+    ...     mode="a",
+    ...     engine="openpyxl",
+    ...     if_sheet_exists="overlay",
+    ... ) as writer:
+    ...     df1.to_excel(writer, sheet_name="Sheet1")
+    ...     df2.to_excel(writer, sheet_name="Sheet1", startcol=3)  # doctest: +SKIP
+
+    You can store Excel file in RAM:
+
+    >>> import io
+    >>> df = pd.DataFrame([["ABC", "XYZ"]], columns=["Foo", "Bar"])
+    >>> buffer = io.BytesIO()
+    >>> with pd.ExcelWriter(buffer) as writer:
+    ...     df.to_excel(writer)
+
+    You can pack Excel file into zip archive:
+
+    >>> import zipfile  # doctest: +SKIP
+    >>> df = pd.DataFrame([["ABC", "XYZ"]], columns=["Foo", "Bar"])  # doctest: +SKIP
+    >>> with zipfile.ZipFile("path_to_file.zip", "w") as zf:
+    ...     with zf.open("filename.xlsx", "w") as buffer:
+    ...         with pd.ExcelWriter(buffer) as writer:
+    ...             df.to_excel(writer)  # doctest: +SKIP
+
+    You can specify additional arguments to the underlying engine:
+
+    >>> with pd.ExcelWriter(
+    ...     "path_to_file.xlsx",
+    ...     engine="xlsxwriter",
+    ...     engine_kwargs={{"options": {{"nan_inf_to_errors": True}}}},
+    ... ) as writer:
+    ...     df.to_excel(writer)  # doctest: +SKIP
+
+    In append mode, ``engine_kwargs`` are passed through to
+    openpyxl's ``load_workbook``:
+
+    >>> with pd.ExcelWriter(
+    ...     "path_to_file.xlsx",
+    ...     engine="openpyxl",
+    ...     mode="a",
+    ...     engine_kwargs={{"keep_vba": True}},
+    ... ) as writer:
+    ...     df.to_excel(writer, sheet_name="Sheet2")  # doctest: +SKIP
+    """
+
+    # Defining an ExcelWriter implementation (see abstract methods for more...)
+
+    # - Mandatory
+    #   - ``write_cells(self, cells, sheet_name=None, startrow=0, startcol=0)``
+    #     --> called to write additional DataFrames to disk
+    #   - ``_supported_extensions`` (tuple of supported extensions), used to
+    #      check that engine supports the given extension.
+    #   - ``_engine`` - string that gives the engine name. Necessary to
+    #     instantiate class directly and bypass ``ExcelWriterMeta`` engine
+    #     lookup.
+    #   - ``save(self)`` --> called to save file to disk
+    # - Mostly mandatory (i.e. should at least exist)
+    #   - book, cur_sheet, path
+
+    # - Optional:
+    #   - ``__init__(self, path, engine=None, **kwargs)`` --> always called
+    #     with path as first argument.
+
+    # You also need to register the class with ``register_writer()``.
+    # Technically, ExcelWriter implementations don't need to subclass
+    # ExcelWriter.
+
+    _engine: str
+    _supported_extensions: tuple[str, ...]
+
+    def __new__(
+        cls,
+        path: FilePath | WriteExcelBuffer | ExcelWriter,
+        engine: str | None = None,
+        date_format: str | None = None,
+        datetime_format: str | None = None,
+        mode: str = "w",
+        storage_options: StorageOptions | None = None,
+        if_sheet_exists: ExcelWriterIfSheetExists | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> Self:
+        # only switch class if generic(ExcelWriter)
+        if cls is ExcelWriter:
+            if engine is None or (isinstance(engine, str) and engine == "auto"):
+                if isinstance(path, str):
+                    ext = os.path.splitext(path)[-1][1:]
+                else:
+                    ext = "xlsx"
+
+                try:
+                    engine = config.get_option(f"io.excel.{ext}.writer")
+                    if engine == "auto":
+                        engine = get_default_engine(ext, mode="writer")
+                except KeyError as err:
+                    raise ValueError(f"No engine for filetype: '{ext}'") from err
+
+            # for mypy
+            assert engine is not None
+            #  error: Incompatible types in assignment (expression has type
+            #  "type[ExcelWriter[Any]]", variable has type "type[Self]")
+            cls = get_writer(engine)  # type: ignore[assignment]
+
+        return object.__new__(cls)
+
+    # declare external properties you can count on
+    _path = None
+
+    @property
+    def supported_extensions(self) -> tuple[str, ...]:
+        """Extensions that writer engine supports."""
+        return self._supported_extensions
+
+    @property
+    def engine(self) -> str:
+        """Name of engine."""
+        return self._engine
+
+    @property
+    def sheets(self) -> dict[str, Any]:
+        """Mapping of sheet names to sheet objects."""
+        raise NotImplementedError
+
+    @property
+    def book(self) -> _WorkbookT:
+        """
+        Book instance. Class type will depend on the engine used.
+
+        This attribute can be used to access engine-specific features.
+        """
+        raise NotImplementedError
+
+    def _write_cells(
+        self,
+        cells,
+        sheet_name: str | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        freeze_panes: tuple[int, int] | None = None,
+    ) -> None:
+        """
+        Write given formatted cells into Excel an excel sheet
+
+        Parameters
+        ----------
+        cells : generator
+            cell of formatted data to save to Excel sheet
+        sheet_name : str, default None
+            Name of Excel sheet, if None, then use self.cur_sheet
+        startrow : upper left cell row to dump data frame
+        startcol : upper left cell column to dump data frame
+        freeze_panes: int tuple of length 2
+            contains the bottom-most row and right-most column to freeze
+        """
+        raise NotImplementedError
+
+    def _save(self) -> None:
+        """
+        Save workbook to disk.
+        """
+        raise NotImplementedError
+
+    def __init__(
+        self,
+        path: FilePath | WriteExcelBuffer | ExcelWriter,
+        engine: str | None = None,
+        date_format: str | None = None,
+        datetime_format: str | None = None,
+        mode: str = "w",
+        storage_options: StorageOptions | None = None,
+        if_sheet_exists: ExcelWriterIfSheetExists | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+    ) -> None:
+        # validate that this engine can handle the extension
+        if isinstance(path, str):
+            ext = os.path.splitext(path)[-1]
+            self.check_extension(ext)
+
+        # use mode to open the file
+        if "b" not in mode:
+            mode += "b"
+        # use "a" for the user to append data to excel but internally use "r+" to let
+        # the excel backend first read the existing file and then write any data to it
+        mode = mode.replace("a", "r+")
+
+        if if_sheet_exists not in (None, "error", "new", "replace", "overlay"):
+            raise ValueError(
+                f"'{if_sheet_exists}' is not valid for if_sheet_exists. "
+                "Valid options are 'error', 'new', 'replace' and 'overlay'."
+            )
+        if if_sheet_exists and "r+" not in mode:
+            raise ValueError("if_sheet_exists is only valid in append mode (mode='a')")
+        if if_sheet_exists is None:
+            if_sheet_exists = "error"
+        self._if_sheet_exists = if_sheet_exists
+
+        # cast ExcelWriter to avoid adding 'if self._handles is not None'
+        self._handles = IOHandles(
+            cast(IO[bytes], path), compression={"compression": None}
+        )
+        if not isinstance(path, ExcelWriter):
+            self._handles = get_handle(
+                path, mode, storage_options=storage_options, is_text=False
+            )
+        self._cur_sheet = None
+
+        if date_format is None:
+            self._date_format = "YYYY-MM-DD"
+        else:
+            self._date_format = date_format
+        if datetime_format is None:
+            self._datetime_format = "YYYY-MM-DD HH:MM:SS"
+        else:
+            self._datetime_format = datetime_format
+
+        self._mode = mode
+
+    @property
+    def date_format(self) -> str:
+        """
+        Format string for dates written into Excel files (e.g. 'YYYY-MM-DD').
+        """
+        return self._date_format
+
+    @property
+    def datetime_format(self) -> str:
+        """
+        Format string for dates written into Excel files (e.g. 'YYYY-MM-DD').
+        """
+        return self._datetime_format
+
+    @property
+    def if_sheet_exists(self) -> str:
+        """
+        How to behave when writing to a sheet that already exists in append mode.
+        """
+        return self._if_sheet_exists
+
+    def __fspath__(self) -> str:
+        return getattr(self._handles.handle, "name", "")
+
+    def _get_sheet_name(self, sheet_name: str | None) -> str:
+        if sheet_name is None:
+            sheet_name = self._cur_sheet
+        if sheet_name is None:  # pragma: no cover
+            raise ValueError("Must pass explicit sheet_name or set _cur_sheet property")
+        return sheet_name
+
+    def _value_with_fmt(
+        self, val
+    ) -> tuple[
+        int | float | bool | str | datetime.datetime | datetime.date, str | None
+    ]:
+        """
+        Convert numpy types to Python types for the Excel writers.
+
+        Parameters
+        ----------
+        val : object
+            Value to be written into cells
+
+        Returns
+        -------
+        Tuple with the first element being the converted value and the second
+            being an optional format
+        """
+        fmt = None
+
+        if is_integer(val):
+            val = int(val)
+        elif is_float(val):
+            val = float(val)
+        elif is_bool(val):
+            val = bool(val)
+        elif is_decimal(val):
+            val = Decimal(val)
+        elif isinstance(val, datetime.datetime):
+            fmt = self._datetime_format
+        elif isinstance(val, datetime.date):
+            fmt = self._date_format
+        elif isinstance(val, datetime.timedelta):
+            val = val.total_seconds() / 86400
+            fmt = "0"
+        else:
+            val = str(val)
+            # GH#56954
+            # Excel's limitation on cell contents is 32767 characters
+            # xref https://support.microsoft.com/en-au/office/excel-specifications-and-limits-1672b34d-7043-467e-8e27-269d656771c3
+            if len(val) > 32767:
+                warnings.warn(
+                    f"Cell contents too long ({len(val)}), "
+                    "truncated to 32767 characters",
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+        return val, fmt
+
+    @classmethod
+    def check_extension(cls, ext: str) -> Literal[True]:
+        """
+        checks that path's extension against the Writer's supported
+        extensions.  If it isn't supported, raises UnsupportedFiletypeError.
+        """
+        if ext.startswith("."):
+            ext = ext[1:]
+        if not any(ext in extension for extension in cls._supported_extensions):
+            raise ValueError(f"Invalid extension for engine '{cls.engine}': '{ext}'")
+        return True
+
+    # Allow use as a contextmanager
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+    def close(self) -> None:
+        """synonym for save, to make it more file-like"""
+        self._save()
+        self._handles.close()
+
+
+XLS_SIGNATURES = (
+    b"\x09\x00\x04\x00\x07\x00\x10\x00",  # BIFF2
+    b"\x09\x02\x06\x00\x00\x00\x10\x00",  # BIFF3
+    b"\x09\x04\x06\x00\x00\x00\x10\x00",  # BIFF4
+    b"\xd0\xcf\x11\xe0\xa1\xb1\x1a\xe1",  # Compound File Binary
+)
+ZIP_SIGNATURE = b"PK\x03\x04"
+PEEK_SIZE = max(map(len, XLS_SIGNATURES + (ZIP_SIGNATURE,)))
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def inspect_excel_format(
+    content_or_path: FilePath | ReadBuffer[bytes],
+    storage_options: StorageOptions | None = None,
+) -> str | None:
+    """
+    Inspect the path or content of an excel file and get its format.
+
+    Adopted from xlrd: https://github.com/python-excel/xlrd.
+
+    Parameters
+    ----------
+    content_or_path : str or file-like object
+        Path to file or content of file to inspect. May be a URL.
+    {storage_options}
+
+    Returns
+    -------
+    str or None
+        Format of file if it can be determined.
+
+    Raises
+    ------
+    ValueError
+        If resulting stream is empty.
+    BadZipFile
+        If resulting stream does not have an XLS signature and is not a valid zipfile.
+    """
+    with get_handle(
+        content_or_path, "rb", storage_options=storage_options, is_text=False
+    ) as handle:
+        stream = handle.handle
+        stream.seek(0)
+        buf = stream.read(PEEK_SIZE)
+        if buf is None:
+            raise ValueError("stream is empty")
+        assert isinstance(buf, bytes)
+        peek = buf
+        stream.seek(0)
+
+        if any(peek.startswith(sig) for sig in XLS_SIGNATURES):
+            return "xls"
+        elif not peek.startswith(ZIP_SIGNATURE):
+            return None
+
+        with zipfile.ZipFile(stream) as zf:
+            # Workaround for some third party files that use forward slashes and
+            # lower case names.
+            component_names = {
+                name.replace("\\", "/").lower() for name in zf.namelist()
+            }
+
+        if "xl/workbook.xml" in component_names:
+            return "xlsx"
+        if "xl/workbook.bin" in component_names:
+            return "xlsb"
+        if "content.xml" in component_names:
+            return "ods"
+        return "zip"
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+class ExcelFile:
+    """
+    Class for parsing tabular Excel sheets into DataFrame objects.
+
+    See read_excel for more documentation.
+
+    Parameters
+    ----------
+    path_or_buffer : str, bytes, pathlib.Path,
+        A file-like object, xlrd workbook or openpyxl workbook.
+        If a string or path object, expected to be a path to a
+        .xls, .xlsx, .xlsb, .xlsm, .odf, .ods, or .odt file.
+    engine : str, default None
+        If io is not a buffer or path, this must be set to identify io.
+        Supported engines: ``xlrd``, ``openpyxl``, ``odf``, ``pyxlsb``, ``calamine``
+        Engine compatibility :
+
+        - ``xlrd`` supports old-style Excel files (.xls).
+        - ``openpyxl`` supports newer Excel file formats.
+        - ``odf`` supports OpenDocument file formats (.odf, .ods, .odt).
+        - ``pyxlsb`` supports Binary Excel files.
+        - ``calamine`` supports Excel (.xls, .xlsx, .xlsm, .xlsb)
+          and OpenDocument (.ods) file formats.
+
+        .. versionchanged:: 1.2.0
+
+           The engine `xlrd <https://xlrd.readthedocs.io/en/latest/>`_
+           now only supports old-style ``.xls`` files.
+           When ``engine=None``, the following logic will be
+           used to determine the engine:
+
+           - If ``path_or_buffer`` is an OpenDocument format (.odf, .ods, .odt),
+             then `odf <https://pypi.org/project/odfpy/>`_ will be used.
+           - Otherwise if ``path_or_buffer`` is an xls format,
+             ``xlrd`` will be used.
+           - Otherwise if ``path_or_buffer`` is in xlsb format,
+             `pyxlsb <https://pypi.org/project/pyxlsb/>`_ will be used.
+
+        .. versionadded:: 1.3.0
+
+           - Otherwise if `openpyxl <https://pypi.org/project/openpyxl/>`_ is installed,
+             then ``openpyxl`` will be used.
+           - Otherwise if ``xlrd >= 2.0`` is installed, a ``ValueError`` will be raised.
+
+        .. warning::
+
+           Please do not report issues when using ``xlrd`` to read ``.xlsx`` files.
+           This is not supported, switch to using ``openpyxl`` instead.
+    {storage_options}
+    engine_kwargs : dict, optional
+        Arbitrary keyword arguments passed to excel engine.
+
+    See Also
+    --------
+    DataFrame.to_excel : Write DataFrame to an Excel file.
+    DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file.
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+    read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+    Examples
+    --------
+    >>> file = pd.ExcelFile("myfile.xlsx")  # doctest: +SKIP
+    >>> with pd.ExcelFile("myfile.xls") as xls:  # doctest: +SKIP
+    ...     df1 = pd.read_excel(xls, "Sheet1")  # doctest: +SKIP
+    """
+
+    from pandas.io.excel._calamine import CalamineReader
+    from pandas.io.excel._odfreader import ODFReader
+    from pandas.io.excel._openpyxl import OpenpyxlReader
+    from pandas.io.excel._pyxlsb import PyxlsbReader
+    from pandas.io.excel._xlrd import XlrdReader
+
+    _engines: Mapping[str, Any] = {
+        "xlrd": XlrdReader,
+        "openpyxl": OpenpyxlReader,
+        "odf": ODFReader,
+        "pyxlsb": PyxlsbReader,
+        "calamine": CalamineReader,
+    }
+
+    def __init__(
+        self,
+        path_or_buffer,
+        engine: str | None = None,
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        if engine_kwargs is None:
+            engine_kwargs = {}
+
+        if engine is not None and engine not in self._engines:
+            raise ValueError(f"Unknown engine: {engine}")
+
+        # Always a string
+        self._io = stringify_path(path_or_buffer)
+
+        if engine is None:
+            # Only determine ext if it is needed
+            ext: str | None = None
+
+            if not isinstance(
+                path_or_buffer, (str, os.PathLike, ExcelFile)
+            ) and not is_file_like(path_or_buffer):
+                # GH#56692 - avoid importing xlrd if possible
+                if import_optional_dependency("xlrd", errors="ignore") is None:
+                    xlrd_version = None
+                else:
+                    import xlrd
+
+                    xlrd_version = Version(get_version(xlrd))
+
+                if xlrd_version is not None and isinstance(path_or_buffer, xlrd.Book):
+                    ext = "xls"
+
+            if ext is None:
+                ext = inspect_excel_format(
+                    content_or_path=path_or_buffer, storage_options=storage_options
+                )
+                if ext is None:
+                    raise ValueError(
+                        "Excel file format cannot be determined, you must specify "
+                        "an engine manually."
+                    )
+
+            engine = config.get_option(f"io.excel.{ext}.reader")
+            if engine == "auto":
+                engine = get_default_engine(ext, mode="reader")
+
+        assert engine is not None
+        self.engine = engine
+        self.storage_options = storage_options
+
+        self._reader = self._engines[engine](
+            self._io,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    def __fspath__(self):
+        return self._io
+
+    def parse(
+        self,
+        sheet_name: str | int | list[int] | list[str] | None = 0,
+        header: int | Sequence[int] | None = 0,
+        names: SequenceNotStr[Hashable] | range | None = None,
+        index_col: int | Sequence[int] | None = None,
+        usecols=None,
+        converters=None,
+        true_values: Iterable[Hashable] | None = None,
+        false_values: Iterable[Hashable] | None = None,
+        skiprows: Sequence[int] | int | Callable[[int], object] | None = None,
+        nrows: int | None = None,
+        na_values=None,
+        parse_dates: list | dict | bool = False,
+        date_format: str | dict[Hashable, str] | None = None,
+        thousands: str | None = None,
+        comment: str | None = None,
+        skipfooter: int = 0,
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+        **kwds,
+    ) -> DataFrame | dict[str, DataFrame] | dict[int, DataFrame]:
+        """
+        Parse specified sheet(s) into a DataFrame.
+
+        Equivalent to read_excel(ExcelFile, ...)  See the read_excel
+        docstring for more info on accepted parameters.
+
+        Parameters
+        ----------
+        sheet_name : str, int, list, or None, default 0
+            Strings are used for sheet names. Integers are used in zero-indexed
+            sheet positions (chart sheets do not count as a sheet position).
+            Lists of strings/integers are used to request multiple sheets.
+            When ``None``, will return a dictionary containing DataFrames for
+            each sheet.
+        header : int, list of int, default 0
+            Row (0-indexed) to use for the column labels of the parsed
+            DataFrame. If a list of integers is passed those row positions will
+            be combined into a ``MultiIndex``. Use None if there is no header.
+        names : array-like, default None
+            List of column names to use. If file contains no header row,
+            then you should explicitly pass header=None.
+        index_col : int, str, list of int, default None
+            Column (0-indexed) to use as the row labels of the DataFrame.
+            Pass None if there is no such column.  If a list is passed,
+            those columns will be combined into a ``MultiIndex``.  If a
+            subset of data is selected with ``usecols``, index_col
+            is based on the subset.
+
+            Missing values will be forward filled to allow roundtripping with
+            ``to_excel`` for ``merged_cells=True``. To avoid forward filling the
+            missing values use ``set_index`` after reading the data instead of
+            ``index_col``.
+        usecols : str, list-like, or callable, default None
+            * If None, then parse all columns.
+            * If str, then indicates comma separated list of Excel column letters
+              and column ranges (e.g. "A:E" or "A,C,E:F"). Ranges are inclusive of
+              both sides.
+            * If list of int, then indicates list of column numbers to be parsed
+              (0-indexed).
+            * If list of string, then indicates list of column names to be parsed.
+            * If callable, then evaluate each column name against it and parse the
+              column if the callable returns ``True``.
+
+            Returns a subset of the columns according to behavior above.
+        converters : dict, default None
+            Dict of functions for converting values in certain columns. Keys can
+            either be integers or column labels, values are functions that take one
+            input argument, the Excel cell content, and return the transformed
+            content.
+        true_values : list, default None
+            Values to consider as True.
+        false_values : list, default None
+            Values to consider as False.
+        skiprows : list-like, int, or callable, optional
+            Line numbers to skip (0-indexed) or number of lines to skip (int) at the
+            start of the file. If callable, the callable function will be evaluated
+            against the row indices, returning True if the row should be skipped and
+            False otherwise. An example of a valid callable argument would be ``lambda
+            x: x in [0, 2]``.
+        nrows : int, default None
+            Number of rows to parse.
+        na_values : scalar, str, list-like, or dict, default None
+            Additional strings to recognize as NA/NaN. If dict passed, specific
+            per-column NA values.
+        parse_dates : bool, list-like, or dict, default False
+            The behavior is as follows:
+
+            * ``bool``. If True -> try parsing the index.
+            * ``list`` of int or names. e.g. If [1, 2, 3] -> try parsing columns 1, 2, 3
+              each as a separate date column.
+            * ``list`` of lists. e.g.  If [[1, 3]] -> combine columns 1 and 3 and
+              parse as a single date column.
+            * ``dict``, e.g. {{'foo' : [1, 3]}} -> parse columns 1, 3 as date and call
+              result 'foo'
+
+            If a column or index contains an unparsable date, the entire column or
+            index will be returned unaltered as an object data type. If you
+            don`t want to parse some cells as date just change their type
+            in Excel to "Text".For non-standard datetime parsing, use
+            ``pd.to_datetime`` after ``pd.read_excel``.
+
+            Note: A fast-path exists for iso8601-formatted dates.
+        date_format : str or dict of column -> format, default ``None``
+           If used in conjunction with ``parse_dates``, will parse dates
+           according to this format. For anything more complex,
+           please read in as ``object`` and then apply :func:`to_datetime` as-needed.
+        thousands : str, default None
+            Thousands separator for parsing string columns to numeric.  Note that
+            this parameter is only necessary for columns stored as TEXT in Excel,
+            any numeric columns will automatically be parsed, regardless of display
+            format.
+        comment : str, default None
+            Comments out remainder of line. Pass a character or characters to this
+            argument to indicate comments in the input file. Any data between the
+            comment string and the end of the current line is ignored.
+        skipfooter : int, default 0
+            Rows at the end to skip (0-indexed).
+        dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+            Back-end data type applied to the resultant :class:`DataFrame`
+            (still experimental). If not specified, the default behavior
+            is to not use nullable data types. If specified, the behavior
+            is as follows:
+
+            * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+            * ``"pyarrow"``: returns pyarrow-backed nullable
+              :class:`ArrowDtype` :class:`DataFrame`
+
+            .. versionadded:: 2.0
+        **kwds : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+
+        Returns
+        -------
+        DataFrame or dict of DataFrames
+            DataFrame from the passed in Excel file.
+
+        See Also
+        --------
+        read_excel : Read an Excel sheet values (xlsx) file into DataFrame.
+        read_csv : Read a comma-separated values (csv) file into DataFrame.
+        read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+        >>> df.to_excel("myfile.xlsx")  # doctest: +SKIP
+        >>> file = pd.ExcelFile("myfile.xlsx")  # doctest: +SKIP
+        >>> file.parse()  # doctest: +SKIP
+        """
+        return self._reader.parse(
+            sheet_name=sheet_name,
+            header=header,
+            names=names,
+            index_col=index_col,
+            usecols=usecols,
+            converters=converters,
+            true_values=true_values,
+            false_values=false_values,
+            skiprows=skiprows,
+            nrows=nrows,
+            na_values=na_values,
+            parse_dates=parse_dates,
+            date_format=date_format,
+            thousands=thousands,
+            comment=comment,
+            skipfooter=skipfooter,
+            dtype_backend=dtype_backend,
+            **kwds,
+        )
+
+    @property
+    def book(self):
+        """
+        Gets the Excel workbook.
+
+        Workbook is the top-level container for all document information.
+
+        Returns
+        -------
+        Excel Workbook
+            The workbook object of the type defined by the engine being used.
+
+        See Also
+        --------
+        read_excel : Read an Excel file into a pandas DataFrame.
+
+        Examples
+        --------
+        >>> file = pd.ExcelFile("myfile.xlsx")  # doctest: +SKIP
+        >>> file.book  # doctest: +SKIP
+        <openpyxl.workbook.workbook.Workbook object at 0x11eb5ad70>
+        >>> file.book.path  # doctest: +SKIP
+        '/xl/workbook.xml'
+        >>> file.book.active  # doctest: +SKIP
+        <openpyxl.worksheet._read_only.ReadOnlyWorksheet object at 0x11eb5b370>
+        >>> file.book.sheetnames  # doctest: +SKIP
+        ['Sheet1', 'Sheet2']
+        """
+        return self._reader.book
+
+    @property
+    def sheet_names(self):
+        """
+        Names of the sheets in the document.
+
+        This is particularly useful for loading a specific sheet into a DataFrame when
+        you do not know the sheet names beforehand.
+
+        Returns
+        -------
+        list of str
+            List of sheet names in the document.
+
+        See Also
+        --------
+        ExcelFile.parse : Parse a sheet into a DataFrame.
+        read_excel : Read an Excel file into a pandas DataFrame. If you know the sheet
+            names, it may be easier to specify them directly to read_excel.
+
+        Examples
+        --------
+        >>> file = pd.ExcelFile("myfile.xlsx")  # doctest: +SKIP
+        >>> file.sheet_names  # doctest: +SKIP
+        ["Sheet1", "Sheet2"]
+        """
+        return self._reader.sheet_names
+
+    def close(self) -> None:
+        """close io if necessary"""
+        self._reader.close()
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
diff --git a/pandas/pandas/io/excel/_calamine.py b/pandas/pandas/io/excel/_calamine.py
new file mode 100644
index 0000000000000..b8994e679d4b1
--- /dev/null
+++ b/pandas/pandas/io/excel/_calamine.py
@@ -0,0 +1,124 @@
+from __future__ import annotations
+
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Union,
+)
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+
+import pandas as pd
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.excel._base import BaseExcelReader
+
+if TYPE_CHECKING:
+    from python_calamine import (
+        CalamineSheet,
+        CalamineWorkbook,
+    )
+
+    from pandas._typing import (
+        FilePath,
+        NaTType,
+        ReadBuffer,
+        Scalar,
+        StorageOptions,
+    )
+
+_CellValue = Union[int, float, str, bool, time, date, datetime, timedelta]
+
+
+class CalamineReader(BaseExcelReader["CalamineWorkbook"]):
+    @doc(storage_options=_shared_docs["storage_options"])
+    def __init__(
+        self,
+        filepath_or_buffer: FilePath | ReadBuffer[bytes],
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        Reader using calamine engine (xlsx/xls/xlsb/ods).
+
+        Parameters
+        ----------
+        filepath_or_buffer : str, path to be parsed or
+            an open readable stream.
+        {storage_options}
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        import_optional_dependency("python_calamine")
+        super().__init__(
+            filepath_or_buffer,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @property
+    def _workbook_class(self) -> type[CalamineWorkbook]:
+        from python_calamine import CalamineWorkbook
+
+        return CalamineWorkbook
+
+    def load_workbook(
+        self, filepath_or_buffer: FilePath | ReadBuffer[bytes], engine_kwargs: Any
+    ) -> CalamineWorkbook:
+        from python_calamine import load_workbook
+
+        return load_workbook(
+            filepath_or_buffer,
+            **engine_kwargs,
+        )
+
+    @property
+    def sheet_names(self) -> list[str]:
+        from python_calamine import SheetTypeEnum
+
+        return [
+            sheet.name
+            for sheet in self.book.sheets_metadata
+            if sheet.typ == SheetTypeEnum.WorkSheet
+        ]
+
+    def get_sheet_by_name(self, name: str) -> CalamineSheet:
+        self.raise_if_bad_sheet_by_name(name)
+        return self.book.get_sheet_by_name(name)
+
+    def get_sheet_by_index(self, index: int) -> CalamineSheet:
+        self.raise_if_bad_sheet_by_index(index)
+        return self.book.get_sheet_by_index(index)
+
+    def get_sheet_data(
+        self, sheet: CalamineSheet, file_rows_needed: int | None = None
+    ) -> list[list[Scalar | NaTType | time]]:
+        def _convert_cell(value: _CellValue) -> Scalar | NaTType | time:
+            if isinstance(value, float):
+                val = int(value)
+                if val == value:
+                    return val
+                else:
+                    return value
+            elif isinstance(value, date):
+                return pd.Timestamp(value)
+            elif isinstance(value, timedelta):
+                return pd.Timedelta(value)
+            elif isinstance(value, time):
+                return value
+
+            return value
+
+        rows: list[list[_CellValue]] = sheet.to_python(
+            skip_empty_area=False, nrows=file_rows_needed
+        )
+        data = [[_convert_cell(cell) for cell in row] for row in rows]
+
+        return data
diff --git a/pandas/pandas/io/excel/_odfreader.py b/pandas/pandas/io/excel/_odfreader.py
new file mode 100644
index 0000000000000..f79417d11080d
--- /dev/null
+++ b/pandas/pandas/io/excel/_odfreader.py
@@ -0,0 +1,249 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    cast,
+)
+
+import numpy as np
+
+from pandas._typing import (
+    FilePath,
+    ReadBuffer,
+    Scalar,
+    StorageOptions,
+)
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+
+import pandas as pd
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.excel._base import BaseExcelReader
+
+if TYPE_CHECKING:
+    from odf.opendocument import OpenDocument
+
+    from pandas._libs.tslibs.nattype import NaTType
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+class ODFReader(BaseExcelReader["OpenDocument"]):
+    def __init__(
+        self,
+        filepath_or_buffer: FilePath | ReadBuffer[bytes],
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        Read tables out of OpenDocument formatted files.
+
+        Parameters
+        ----------
+        filepath_or_buffer : str, path to be parsed or
+            an open readable stream.
+        {storage_options}
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        import_optional_dependency("odf")
+        super().__init__(
+            filepath_or_buffer,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @property
+    def _workbook_class(self) -> type[OpenDocument]:
+        from odf.opendocument import OpenDocument
+
+        return OpenDocument
+
+    def load_workbook(
+        self, filepath_or_buffer: FilePath | ReadBuffer[bytes], engine_kwargs
+    ) -> OpenDocument:
+        from odf.opendocument import load
+
+        return load(filepath_or_buffer, **engine_kwargs)
+
+    @property
+    def empty_value(self) -> str:
+        """Property for compat with other readers."""
+        return ""
+
+    @property
+    def sheet_names(self) -> list[str]:
+        """Return a list of sheet names present in the document"""
+        from odf.table import Table
+
+        tables = self.book.getElementsByType(Table)
+        return [t.getAttribute("name") for t in tables]
+
+    def get_sheet_by_index(self, index: int):
+        from odf.table import Table
+
+        self.raise_if_bad_sheet_by_index(index)
+        tables = self.book.getElementsByType(Table)
+        return tables[index]
+
+    def get_sheet_by_name(self, name: str):
+        from odf.table import Table
+
+        self.raise_if_bad_sheet_by_name(name)
+        tables = self.book.getElementsByType(Table)
+
+        for table in tables:
+            if table.getAttribute("name") == name:
+                return table
+
+        self.close()
+        raise ValueError(f"sheet {name} not found")
+
+    def get_sheet_data(
+        self, sheet, file_rows_needed: int | None = None
+    ) -> list[list[Scalar | NaTType]]:
+        """
+        Parse an ODF Table into a list of lists
+        """
+        from odf.table import (
+            CoveredTableCell,
+            TableCell,
+            TableRow,
+        )
+
+        covered_cell_name = CoveredTableCell().qname
+        table_cell_name = TableCell().qname
+        cell_names = {covered_cell_name, table_cell_name}
+
+        sheet_rows = sheet.getElementsByType(TableRow)
+        empty_rows = 0
+        max_row_len = 0
+
+        table: list[list[Scalar | NaTType]] = []
+
+        for sheet_row in sheet_rows:
+            empty_cells = 0
+            table_row: list[Scalar | NaTType] = []
+
+            for sheet_cell in sheet_row.childNodes:
+                if hasattr(sheet_cell, "qname") and sheet_cell.qname in cell_names:
+                    if sheet_cell.qname == table_cell_name:
+                        value = self._get_cell_value(sheet_cell)
+                    else:
+                        value = self.empty_value
+
+                    column_repeat = self._get_column_repeat(sheet_cell)
+
+                    # Queue up empty values, writing only if content succeeds them
+                    if value == self.empty_value:
+                        empty_cells += column_repeat
+                    else:
+                        table_row.extend([self.empty_value] * empty_cells)
+                        empty_cells = 0
+                        table_row.extend([value] * column_repeat)
+
+            if max_row_len < len(table_row):
+                max_row_len = len(table_row)
+
+            row_repeat = self._get_row_repeat(sheet_row)
+            if len(table_row) == 0:
+                empty_rows += row_repeat
+            else:
+                # add blank rows to our table
+                table.extend([[self.empty_value]] * empty_rows)
+                empty_rows = 0
+                table.extend(table_row for _ in range(row_repeat))
+            if file_rows_needed is not None and len(table) >= file_rows_needed:
+                break
+
+        # Make our table square
+        for row in table:
+            if len(row) < max_row_len:
+                row.extend([self.empty_value] * (max_row_len - len(row)))
+
+        return table
+
+    def _get_row_repeat(self, row) -> int:
+        """
+        Return number of times this row was repeated
+        Repeating an empty row appeared to be a common way
+        of representing sparse rows in the table.
+        """
+        from odf.namespaces import TABLENS
+
+        return int(row.attributes.get((TABLENS, "number-rows-repeated"), 1))
+
+    def _get_column_repeat(self, cell) -> int:
+        from odf.namespaces import TABLENS
+
+        return int(cell.attributes.get((TABLENS, "number-columns-repeated"), 1))
+
+    def _get_cell_value(self, cell) -> Scalar | NaTType:
+        from odf.namespaces import OFFICENS
+
+        if str(cell) == "#N/A":
+            return np.nan
+
+        cell_type = cell.attributes.get((OFFICENS, "value-type"))
+        if cell_type == "boolean":
+            if str(cell) == "TRUE":
+                return True
+            return False
+        if cell_type is None:
+            return self.empty_value
+        elif cell_type == "float":
+            # GH5394
+            cell_value = float(cell.attributes.get((OFFICENS, "value")))
+            val = int(cell_value)
+            if val == cell_value:
+                return val
+            return cell_value
+        elif cell_type == "percentage":
+            cell_value = cell.attributes.get((OFFICENS, "value"))
+            return float(cell_value)
+        elif cell_type == "string":
+            return self._get_cell_string_value(cell)
+        elif cell_type == "currency":
+            cell_value = cell.attributes.get((OFFICENS, "value"))
+            return float(cell_value)
+        elif cell_type == "date":
+            cell_value = cell.attributes.get((OFFICENS, "date-value"))
+            return pd.Timestamp(cell_value)
+        elif cell_type == "time":
+            stamp = pd.Timestamp(str(cell))
+            # cast needed here because Scalar doesn't include datetime.time
+            return cast(Scalar, stamp.time())
+        else:
+            self.close()
+            raise ValueError(f"Unrecognized type {cell_type}")
+
+    def _get_cell_string_value(self, cell) -> str:
+        """
+        Find and decode OpenDocument text:s tags that represent
+        a run length encoded sequence of space characters.
+        """
+        from odf.element import Element
+        from odf.namespaces import TEXTNS
+        from odf.office import Annotation
+        from odf.text import S
+
+        office_annotation = Annotation().qname
+        text_s = S().qname
+
+        value = []
+
+        for fragment in cell.childNodes:
+            if isinstance(fragment, Element):
+                if fragment.qname == text_s:
+                    spaces = int(fragment.attributes.get((TEXTNS, "c"), 1))
+                    value.append(" " * spaces)
+                elif fragment.qname == office_annotation:
+                    continue
+                else:
+                    # recursive impl needed in case of nested fragments
+                    # with multiple spaces
+                    # https://github.com/pandas-dev/pandas/pull/36175#discussion_r484639704
+                    value.append(self._get_cell_string_value(fragment))
+            else:
+                value.append(str(fragment).strip("\n"))
+        return "".join(value)
diff --git a/pandas/pandas/io/excel/_odswriter.py b/pandas/pandas/io/excel/_odswriter.py
new file mode 100644
index 0000000000000..ba4919c9298ed
--- /dev/null
+++ b/pandas/pandas/io/excel/_odswriter.py
@@ -0,0 +1,357 @@
+from __future__ import annotations
+
+from collections import defaultdict
+import datetime
+import json
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    DefaultDict,
+    cast,
+    overload,
+)
+
+from pandas.io.excel._base import ExcelWriter
+from pandas.io.excel._util import (
+    combine_kwargs,
+    validate_freeze_panes,
+)
+
+if TYPE_CHECKING:
+    from odf.opendocument import OpenDocumentSpreadsheet
+
+    from pandas._typing import (
+        ExcelWriterIfSheetExists,
+        FilePath,
+        StorageOptions,
+        WriteExcelBuffer,
+    )
+
+    from pandas.io.formats.excel import ExcelCell
+
+
+class ODSWriter(ExcelWriter):
+    _engine = "odf"
+    _supported_extensions = (".ods",)
+
+    def __init__(  # pyright: ignore[reportInconsistentConstructor]
+        self,
+        path: FilePath | WriteExcelBuffer | ExcelWriter,
+        engine: str | None = None,
+        date_format: str | None = None,
+        datetime_format: str | None = None,
+        mode: str = "w",
+        storage_options: StorageOptions | None = None,
+        if_sheet_exists: ExcelWriterIfSheetExists | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+        **kwargs: Any,
+    ) -> None:
+        from odf.opendocument import OpenDocumentSpreadsheet
+
+        if mode == "a":
+            raise ValueError("Append mode is not supported with odf!")
+
+        engine_kwargs = combine_kwargs(engine_kwargs, kwargs)
+        self._book = OpenDocumentSpreadsheet(**engine_kwargs)
+
+        super().__init__(
+            path,
+            mode=mode,
+            storage_options=storage_options,
+            if_sheet_exists=if_sheet_exists,
+            engine_kwargs=engine_kwargs,
+        )
+
+        self._style_dict: dict[str, str] = {}
+
+    @property
+    def book(self) -> OpenDocumentSpreadsheet:
+        """
+        Book instance of class odf.opendocument.OpenDocumentSpreadsheet.
+
+        This attribute can be used to access engine-specific features.
+        """
+        return self._book
+
+    @property
+    def sheets(self) -> dict[str, Any]:
+        """Mapping of sheet names to sheet objects."""
+        from odf.table import Table
+
+        result = {
+            sheet.getAttribute("name"): sheet
+            for sheet in self.book.getElementsByType(Table)
+        }
+        return result
+
+    def _save(self) -> None:
+        """
+        Save workbook to disk.
+        """
+        for sheet in self.sheets.values():
+            self.book.spreadsheet.addElement(sheet)
+        self.book.save(self._handles.handle)
+
+    def _write_cells(
+        self,
+        cells: list[ExcelCell],
+        sheet_name: str | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        freeze_panes: tuple[int, int] | None = None,
+    ) -> None:
+        """
+        Write the frame cells using odf
+        """
+        from odf.table import (
+            Table,
+            TableCell,
+            TableRow,
+        )
+        from odf.text import P
+
+        sheet_name = self._get_sheet_name(sheet_name)
+        assert sheet_name is not None
+
+        if sheet_name in self.sheets:
+            wks = self.sheets[sheet_name]
+        else:
+            wks = Table(name=sheet_name)
+            self.book.spreadsheet.addElement(wks)
+
+        if validate_freeze_panes(freeze_panes):
+            freeze_panes = cast(tuple[int, int], freeze_panes)
+            self._create_freeze_panes(sheet_name, freeze_panes)
+
+        for _ in range(startrow):
+            wks.addElement(TableRow())
+
+        rows: DefaultDict = defaultdict(TableRow)
+        col_count: DefaultDict = defaultdict(int)
+
+        for cell in sorted(cells, key=lambda cell: (cell.row, cell.col)):
+            # only add empty cells if the row is still empty
+            if not col_count[cell.row]:
+                for _ in range(startcol):
+                    rows[cell.row].addElement(TableCell())
+
+            # fill with empty cells if needed
+            for _ in range(cell.col - col_count[cell.row]):
+                rows[cell.row].addElement(TableCell())
+                col_count[cell.row] += 1
+
+            pvalue, tc = self._make_table_cell(cell)
+            rows[cell.row].addElement(tc)
+            col_count[cell.row] += 1
+            p = P(text=pvalue)
+            tc.addElement(p)
+
+        # add all rows to the sheet
+        if len(rows) > 0:
+            for row_nr in range(max(rows.keys()) + 1):
+                wks.addElement(rows[row_nr])
+
+    def _make_table_cell_attributes(self, cell: ExcelCell) -> dict[str, int | str]:
+        """Convert cell attributes to OpenDocument attributes
+
+        Parameters
+        ----------
+        cell : ExcelCell
+            Spreadsheet cell data
+
+        Returns
+        -------
+        attributes : Dict[str, Union[int, str]]
+            Dictionary with attributes and attribute values
+        """
+        attributes: dict[str, int | str] = {}
+        style_name = self._process_style(cell.style)
+        if style_name is not None:
+            attributes["stylename"] = style_name
+        if cell.mergestart is not None and cell.mergeend is not None:
+            attributes["numberrowsspanned"] = max(1, cell.mergestart)
+            attributes["numbercolumnsspanned"] = cell.mergeend
+        return attributes
+
+    def _make_table_cell(self, cell: ExcelCell) -> tuple[object, Any]:
+        """Convert cell data to an OpenDocument spreadsheet cell
+
+        Parameters
+        ----------
+        cell : ExcelCell
+            Spreadsheet cell data
+
+        Returns
+        -------
+        pvalue, cell : Tuple[str, TableCell]
+            Display value, Cell value
+        """
+        from odf.table import TableCell
+
+        attributes = self._make_table_cell_attributes(cell)
+        val, fmt = self._value_with_fmt(cell.val)
+        pvalue = value = val
+        if isinstance(val, bool):
+            value = str(val).lower()
+            pvalue = str(val).upper()
+            return (
+                pvalue,
+                TableCell(
+                    valuetype="boolean",
+                    booleanvalue=value,
+                    attributes=attributes,
+                ),
+            )
+        elif isinstance(val, datetime.datetime):
+            # Fast formatting
+            value = val.isoformat()
+            # Slow but locale-dependent
+            pvalue = val.strftime("%c")
+            return (
+                pvalue,
+                TableCell(valuetype="date", datevalue=value, attributes=attributes),
+            )
+        elif isinstance(val, datetime.date):
+            # Fast formatting
+            value = f"{val.year}-{val.month:02d}-{val.day:02d}"
+            # Slow but locale-dependent
+            pvalue = val.strftime("%x")
+            return (
+                pvalue,
+                TableCell(valuetype="date", datevalue=value, attributes=attributes),
+            )
+        elif isinstance(val, str):
+            return (
+                pvalue,
+                TableCell(
+                    valuetype="string",
+                    stringvalue=value,
+                    attributes=attributes,
+                ),
+            )
+        else:
+            return (
+                pvalue,
+                TableCell(
+                    valuetype="float",
+                    value=value,
+                    attributes=attributes,
+                ),
+            )
+
+    @overload
+    def _process_style(self, style: dict[str, Any]) -> str: ...
+
+    @overload
+    def _process_style(self, style: None) -> None: ...
+
+    def _process_style(self, style: dict[str, Any] | None) -> str | None:
+        """Convert a style dictionary to a OpenDocument style sheet
+
+        Parameters
+        ----------
+        style : Dict
+            Style dictionary
+
+        Returns
+        -------
+        style_key : str
+            Unique style key for later reference in sheet
+        """
+        from odf.style import (
+            ParagraphProperties,
+            Style,
+            TableCellProperties,
+            TextProperties,
+        )
+
+        if style is None:
+            return None
+        style_key = json.dumps(style)
+        if style_key in self._style_dict:
+            return self._style_dict[style_key]
+        name = f"pd{len(self._style_dict) + 1}"
+        self._style_dict[style_key] = name
+        odf_style = Style(name=name, family="table-cell")
+        if "font" in style:
+            font = style["font"]
+            if font.get("bold", False):
+                odf_style.addElement(TextProperties(fontweight="bold"))
+        if "borders" in style:
+            borders = style["borders"]
+            for side, thickness in borders.items():
+                thickness_translation = {"thin": "0.75pt solid #000000"}
+                odf_style.addElement(
+                    TableCellProperties(
+                        attributes={f"border{side}": thickness_translation[thickness]}
+                    )
+                )
+        if "alignment" in style:
+            alignment = style["alignment"]
+            horizontal = alignment.get("horizontal")
+            if horizontal:
+                odf_style.addElement(ParagraphProperties(textalign=horizontal))
+            vertical = alignment.get("vertical")
+            if vertical:
+                odf_style.addElement(TableCellProperties(verticalalign=vertical))
+        self.book.styles.addElement(odf_style)
+        return name
+
+    def _create_freeze_panes(
+        self, sheet_name: str, freeze_panes: tuple[int, int]
+    ) -> None:
+        """
+        Create freeze panes in the sheet.
+
+        Parameters
+        ----------
+        sheet_name : str
+            Name of the spreadsheet
+        freeze_panes : tuple of (int, int)
+            Freeze pane location x and y
+        """
+        from odf.config import (
+            ConfigItem,
+            ConfigItemMapEntry,
+            ConfigItemMapIndexed,
+            ConfigItemMapNamed,
+            ConfigItemSet,
+        )
+
+        config_item_set = ConfigItemSet(name="ooo:view-settings")
+        self.book.settings.addElement(config_item_set)
+
+        config_item_map_indexed = ConfigItemMapIndexed(name="Views")
+        config_item_set.addElement(config_item_map_indexed)
+
+        config_item_map_entry = ConfigItemMapEntry()
+        config_item_map_indexed.addElement(config_item_map_entry)
+
+        config_item_map_named = ConfigItemMapNamed(name="Tables")
+        config_item_map_entry.addElement(config_item_map_named)
+
+        config_item_map_entry = ConfigItemMapEntry(name=sheet_name)
+        config_item_map_named.addElement(config_item_map_entry)
+
+        config_item_map_entry.addElement(
+            ConfigItem(name="HorizontalSplitMode", type="short", text="2")
+        )
+        config_item_map_entry.addElement(
+            ConfigItem(name="VerticalSplitMode", type="short", text="2")
+        )
+        config_item_map_entry.addElement(
+            ConfigItem(
+                name="HorizontalSplitPosition", type="int", text=str(freeze_panes[0])
+            )
+        )
+        config_item_map_entry.addElement(
+            ConfigItem(
+                name="VerticalSplitPosition", type="int", text=str(freeze_panes[1])
+            )
+        )
+        config_item_map_entry.addElement(
+            ConfigItem(name="PositionRight", type="int", text=str(freeze_panes[0]))
+        )
+        config_item_map_entry.addElement(
+            ConfigItem(name="PositionBottom", type="int", text=str(freeze_panes[1]))
+        )
diff --git a/pandas/pandas/io/excel/_openpyxl.py b/pandas/pandas/io/excel/_openpyxl.py
new file mode 100644
index 0000000000000..3055c68a93cbc
--- /dev/null
+++ b/pandas/pandas/io/excel/_openpyxl.py
@@ -0,0 +1,640 @@
+from __future__ import annotations
+
+import mmap
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.excel._base import (
+    BaseExcelReader,
+    ExcelWriter,
+)
+from pandas.io.excel._util import (
+    combine_kwargs,
+    validate_freeze_panes,
+)
+
+if TYPE_CHECKING:
+    from openpyxl import Workbook
+    from openpyxl.descriptors.serialisable import Serialisable
+    from openpyxl.styles import Fill
+
+    from pandas._typing import (
+        ExcelWriterIfSheetExists,
+        FilePath,
+        ReadBuffer,
+        Scalar,
+        StorageOptions,
+        WriteExcelBuffer,
+    )
+
+
+class OpenpyxlWriter(ExcelWriter):
+    _engine = "openpyxl"
+    _supported_extensions = (".xlsx", ".xlsm")
+
+    def __init__(  # pyright: ignore[reportInconsistentConstructor]
+        self,
+        path: FilePath | WriteExcelBuffer | ExcelWriter,
+        engine: str | None = None,
+        date_format: str | None = None,
+        datetime_format: str | None = None,
+        mode: str = "w",
+        storage_options: StorageOptions | None = None,
+        if_sheet_exists: ExcelWriterIfSheetExists | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+        **kwargs,
+    ) -> None:
+        # Use the openpyxl module as the Excel writer.
+        from openpyxl.workbook import Workbook
+
+        engine_kwargs = combine_kwargs(engine_kwargs, kwargs)
+
+        super().__init__(
+            path,
+            mode=mode,
+            storage_options=storage_options,
+            if_sheet_exists=if_sheet_exists,
+            engine_kwargs=engine_kwargs,
+        )
+
+        # ExcelWriter replaced "a" by "r+" to allow us to first read the excel file from
+        # the file and later write to it
+        if "r+" in self._mode:  # Load from existing workbook
+            from openpyxl import load_workbook
+
+            try:
+                self._book = load_workbook(self._handles.handle, **engine_kwargs)
+            except TypeError:
+                self._handles.handle.close()
+                raise
+            self._handles.handle.seek(0)
+        else:
+            # Create workbook object with default optimized_write=True.
+            try:
+                self._book = Workbook(**engine_kwargs)
+            except TypeError:
+                self._handles.handle.close()
+                raise
+
+            if self.book.worksheets:
+                self.book.remove(self.book.worksheets[0])
+
+    @property
+    def book(self) -> Workbook:
+        """
+        Book instance of class openpyxl.workbook.Workbook.
+
+        This attribute can be used to access engine-specific features.
+        """
+        return self._book
+
+    @property
+    def sheets(self) -> dict[str, Any]:
+        """Mapping of sheet names to sheet objects."""
+        result = {name: self.book[name] for name in self.book.sheetnames}
+        return result
+
+    def _save(self) -> None:
+        """
+        Save workbook to disk.
+        """
+        self.book.save(self._handles.handle)
+        if "r+" in self._mode and not isinstance(self._handles.handle, mmap.mmap):
+            # truncate file to the written content
+            self._handles.handle.truncate()
+
+    @classmethod
+    def _convert_to_style_kwargs(cls, style_dict: dict) -> dict[str, Serialisable]:
+        """
+        Convert a style_dict to a set of kwargs suitable for initializing
+        or updating-on-copy an openpyxl v2 style object.
+
+        Parameters
+        ----------
+        style_dict : dict
+            A dict with zero or more of the following keys (or their synonyms).
+                'font'
+                'fill'
+                'border' ('borders')
+                'alignment'
+                'number_format'
+                'protection'
+
+        Returns
+        -------
+        style_kwargs : dict
+            A dict with the same, normalized keys as ``style_dict`` but each
+            value has been replaced with a native openpyxl style object of the
+            appropriate class.
+        """
+        _style_key_map = {"borders": "border"}
+
+        style_kwargs: dict[str, Serialisable] = {}
+        for k, v in style_dict.items():
+            k = _style_key_map.get(k, k)
+            _conv_to_x = getattr(cls, f"_convert_to_{k}", lambda x: None)
+            new_v = _conv_to_x(v)
+            if new_v:
+                style_kwargs[k] = new_v
+
+        return style_kwargs
+
+    @classmethod
+    def _convert_to_color(cls, color_spec):
+        """
+        Convert ``color_spec`` to an openpyxl v2 Color object.
+
+        Parameters
+        ----------
+        color_spec : str, dict
+            A 32-bit ARGB hex string, or a dict with zero or more of the
+            following keys.
+                'rgb'
+                'indexed'
+                'auto'
+                'theme'
+                'tint'
+                'index'
+                'type'
+
+        Returns
+        -------
+        color : openpyxl.styles.Color
+        """
+        from openpyxl.styles import Color
+
+        if isinstance(color_spec, str):
+            return Color(color_spec)
+        else:
+            return Color(**color_spec)
+
+    @classmethod
+    def _convert_to_font(cls, font_dict):
+        """
+        Convert ``font_dict`` to an openpyxl v2 Font object.
+
+        Parameters
+        ----------
+        font_dict : dict
+            A dict with zero or more of the following keys (or their synonyms).
+                'name'
+                'size' ('sz')
+                'bold' ('b')
+                'italic' ('i')
+                'underline' ('u')
+                'strikethrough' ('strike')
+                'color'
+                'vertAlign' ('vertalign')
+                'charset'
+                'scheme'
+                'family'
+                'outline'
+                'shadow'
+                'condense'
+
+        Returns
+        -------
+        font : openpyxl.styles.Font
+        """
+        from openpyxl.styles import Font
+
+        _font_key_map = {
+            "sz": "size",
+            "b": "bold",
+            "i": "italic",
+            "u": "underline",
+            "strike": "strikethrough",
+            "vertalign": "vertAlign",
+        }
+
+        font_kwargs = {}
+        for k, v in font_dict.items():
+            k = _font_key_map.get(k, k)
+            if k == "color":
+                v = cls._convert_to_color(v)
+            font_kwargs[k] = v
+
+        return Font(**font_kwargs)
+
+    @classmethod
+    def _convert_to_stop(cls, stop_seq):
+        """
+        Convert ``stop_seq`` to a list of openpyxl v2 Color objects,
+        suitable for initializing the ``GradientFill`` ``stop`` parameter.
+
+        Parameters
+        ----------
+        stop_seq : iterable
+            An iterable that yields objects suitable for consumption by
+            ``_convert_to_color``.
+
+        Returns
+        -------
+        stop : list of openpyxl.styles.Color
+        """
+        return map(cls._convert_to_color, stop_seq)
+
+    @classmethod
+    def _convert_to_fill(cls, fill_dict: dict[str, Any]) -> Fill:
+        """
+        Convert ``fill_dict`` to an openpyxl v2 Fill object.
+
+        Parameters
+        ----------
+        fill_dict : dict
+            A dict with one or more of the following keys (or their synonyms),
+                'fill_type' ('patternType', 'patterntype')
+                'start_color' ('fgColor', 'fgcolor')
+                'end_color' ('bgColor', 'bgcolor')
+            or one or more of the following keys (or their synonyms).
+                'type' ('fill_type')
+                'degree'
+                'left'
+                'right'
+                'top'
+                'bottom'
+                'stop'
+
+        Returns
+        -------
+        fill : openpyxl.styles.Fill
+        """
+        from openpyxl.styles import (
+            GradientFill,
+            PatternFill,
+        )
+
+        _pattern_fill_key_map = {
+            "patternType": "fill_type",
+            "patterntype": "fill_type",
+            "fgColor": "start_color",
+            "fgcolor": "start_color",
+            "bgColor": "end_color",
+            "bgcolor": "end_color",
+        }
+
+        _gradient_fill_key_map = {"fill_type": "type"}
+
+        pfill_kwargs = {}
+        gfill_kwargs = {}
+        for k, v in fill_dict.items():
+            pk = _pattern_fill_key_map.get(k)
+            gk = _gradient_fill_key_map.get(k)
+            if pk in ["start_color", "end_color"]:
+                v = cls._convert_to_color(v)
+            if gk == "stop":
+                v = cls._convert_to_stop(v)
+            if pk:
+                pfill_kwargs[pk] = v
+            elif gk:
+                gfill_kwargs[gk] = v
+            else:
+                pfill_kwargs[k] = v
+                gfill_kwargs[k] = v
+
+        try:
+            return PatternFill(**pfill_kwargs)
+        except TypeError:
+            return GradientFill(**gfill_kwargs)
+
+    @classmethod
+    def _convert_to_side(cls, side_spec):
+        """
+        Convert ``side_spec`` to an openpyxl v2 Side object.
+
+        Parameters
+        ----------
+        side_spec : str, dict
+            A string specifying the border style, or a dict with zero or more
+            of the following keys (or their synonyms).
+                'style' ('border_style')
+                'color'
+
+        Returns
+        -------
+        side : openpyxl.styles.Side
+        """
+        from openpyxl.styles import Side
+
+        _side_key_map = {"border_style": "style"}
+
+        if isinstance(side_spec, str):
+            return Side(style=side_spec)
+
+        side_kwargs = {}
+        for k, v in side_spec.items():
+            k = _side_key_map.get(k, k)
+            if k == "color":
+                v = cls._convert_to_color(v)
+            side_kwargs[k] = v
+
+        return Side(**side_kwargs)
+
+    @classmethod
+    def _convert_to_border(cls, border_dict):
+        """
+        Convert ``border_dict`` to an openpyxl v2 Border object.
+
+        Parameters
+        ----------
+        border_dict : dict
+            A dict with zero or more of the following keys (or their synonyms).
+                'left'
+                'right'
+                'top'
+                'bottom'
+                'diagonal'
+                'diagonal_direction'
+                'vertical'
+                'horizontal'
+                'diagonalUp' ('diagonalup')
+                'diagonalDown' ('diagonaldown')
+                'outline'
+
+        Returns
+        -------
+        border : openpyxl.styles.Border
+        """
+        from openpyxl.styles import Border
+
+        _border_key_map = {"diagonalup": "diagonalUp", "diagonaldown": "diagonalDown"}
+
+        border_kwargs = {}
+        for k, v in border_dict.items():
+            k = _border_key_map.get(k, k)
+            if k == "color":
+                v = cls._convert_to_color(v)
+            if k in ["left", "right", "top", "bottom", "diagonal"]:
+                v = cls._convert_to_side(v)
+            border_kwargs[k] = v
+
+        return Border(**border_kwargs)
+
+    @classmethod
+    def _convert_to_alignment(cls, alignment_dict):
+        """
+        Convert ``alignment_dict`` to an openpyxl v2 Alignment object.
+
+        Parameters
+        ----------
+        alignment_dict : dict
+            A dict with zero or more of the following keys (or their synonyms).
+                'horizontal'
+                'vertical'
+                'text_rotation'
+                'wrap_text'
+                'shrink_to_fit'
+                'indent'
+        Returns
+        -------
+        alignment : openpyxl.styles.Alignment
+        """
+        from openpyxl.styles import Alignment
+
+        return Alignment(**alignment_dict)
+
+    @classmethod
+    def _convert_to_number_format(cls, number_format_dict):
+        """
+        Convert ``number_format_dict`` to an openpyxl v2.1.0 number format
+        initializer.
+
+        Parameters
+        ----------
+        number_format_dict : dict
+            A dict with zero or more of the following keys.
+                'format_code' : str
+
+        Returns
+        -------
+        number_format : str
+        """
+        return number_format_dict["format_code"]
+
+    @classmethod
+    def _convert_to_protection(cls, protection_dict):
+        """
+        Convert ``protection_dict`` to an openpyxl v2 Protection object.
+
+        Parameters
+        ----------
+        protection_dict : dict
+            A dict with zero or more of the following keys.
+                'locked'
+                'hidden'
+
+        Returns
+        -------
+        """
+        from openpyxl.styles import Protection
+
+        return Protection(**protection_dict)
+
+    def _write_cells(
+        self,
+        cells,
+        sheet_name: str | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        freeze_panes: tuple[int, int] | None = None,
+    ) -> None:
+        # Write the frame cells using openpyxl.
+        sheet_name = self._get_sheet_name(sheet_name)
+
+        _style_cache: dict[str, dict[str, Serialisable]] = {}
+
+        if sheet_name in self.sheets and self._if_sheet_exists != "new":
+            if "r+" in self._mode:
+                if self._if_sheet_exists == "replace":
+                    old_wks = self.sheets[sheet_name]
+                    target_index = self.book.index(old_wks)
+                    del self.book[sheet_name]
+                    wks = self.book.create_sheet(sheet_name, target_index)
+                elif self._if_sheet_exists == "error":
+                    raise ValueError(
+                        f"Sheet '{sheet_name}' already exists and "
+                        f"if_sheet_exists is set to 'error'."
+                    )
+                elif self._if_sheet_exists == "overlay":
+                    wks = self.sheets[sheet_name]
+                else:
+                    raise ValueError(
+                        f"'{self._if_sheet_exists}' is not valid for if_sheet_exists. "
+                        "Valid options are 'error', 'new', 'replace' and 'overlay'."
+                    )
+            else:
+                wks = self.sheets[sheet_name]
+        else:
+            wks = self.book.create_sheet()
+            wks.title = sheet_name
+
+        if validate_freeze_panes(freeze_panes):
+            freeze_panes = cast(tuple[int, int], freeze_panes)
+            wks.freeze_panes = wks.cell(
+                row=freeze_panes[0] + 1, column=freeze_panes[1] + 1
+            )
+
+        for cell in cells:
+            xcell = wks.cell(
+                row=startrow + cell.row + 1, column=startcol + cell.col + 1
+            )
+            xcell.value, fmt = self._value_with_fmt(cell.val)
+            if fmt:
+                xcell.number_format = fmt
+
+            style_kwargs: dict[str, Serialisable] | None = {}
+            if cell.style:
+                key = str(cell.style)
+                style_kwargs = _style_cache.get(key)
+                if style_kwargs is None:
+                    style_kwargs = self._convert_to_style_kwargs(cell.style)
+                    _style_cache[key] = style_kwargs
+
+            if style_kwargs:
+                for k, v in style_kwargs.items():
+                    setattr(xcell, k, v)
+
+            if cell.mergestart is not None and cell.mergeend is not None:
+                wks.merge_cells(
+                    start_row=startrow + cell.row + 1,
+                    start_column=startcol + cell.col + 1,
+                    end_column=startcol + cell.mergeend + 1,
+                    end_row=startrow + cell.mergestart + 1,
+                )
+
+                # When cells are merged only the top-left cell is preserved
+                # The behaviour of the other cells in a merged range is
+                # undefined
+                if style_kwargs:
+                    first_row = startrow + cell.row + 1
+                    last_row = startrow + cell.mergestart + 1
+                    first_col = startcol + cell.col + 1
+                    last_col = startcol + cell.mergeend + 1
+
+                    for row in range(first_row, last_row + 1):
+                        for col in range(first_col, last_col + 1):
+                            if row == first_row and col == first_col:
+                                # Ignore first cell. It is already handled.
+                                continue
+                            xcell = wks.cell(column=col, row=row)
+                            for k, v in style_kwargs.items():
+                                setattr(xcell, k, v)
+
+
+class OpenpyxlReader(BaseExcelReader["Workbook"]):
+    @doc(storage_options=_shared_docs["storage_options"])
+    def __init__(
+        self,
+        filepath_or_buffer: FilePath | ReadBuffer[bytes],
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        Reader using openpyxl engine.
+
+        Parameters
+        ----------
+        filepath_or_buffer : str, path object or Workbook
+            Object to be parsed.
+        {storage_options}
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        import_optional_dependency("openpyxl")
+        super().__init__(
+            filepath_or_buffer,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @property
+    def _workbook_class(self) -> type[Workbook]:
+        from openpyxl import Workbook
+
+        return Workbook
+
+    def load_workbook(
+        self, filepath_or_buffer: FilePath | ReadBuffer[bytes], engine_kwargs
+    ) -> Workbook:
+        from openpyxl import load_workbook
+
+        default_kwargs = {"read_only": True, "data_only": True, "keep_links": False}
+
+        return load_workbook(
+            filepath_or_buffer,
+            **(default_kwargs | engine_kwargs),
+        )
+
+    @property
+    def sheet_names(self) -> list[str]:
+        return [sheet.title for sheet in self.book.worksheets]
+
+    def get_sheet_by_name(self, name: str):
+        self.raise_if_bad_sheet_by_name(name)
+        return self.book[name]
+
+    def get_sheet_by_index(self, index: int):
+        self.raise_if_bad_sheet_by_index(index)
+        return self.book.worksheets[index]
+
+    def _convert_cell(self, cell) -> Scalar:
+        from openpyxl.cell.cell import (
+            TYPE_ERROR,
+            TYPE_NUMERIC,
+        )
+
+        if cell.value is None:
+            return ""  # compat with xlrd
+        elif cell.data_type == TYPE_ERROR:
+            return np.nan
+        elif cell.data_type == TYPE_NUMERIC:
+            val = int(cell.value)
+            if val == cell.value:
+                return val
+            return float(cell.value)
+
+        return cell.value
+
+    def get_sheet_data(
+        self, sheet, file_rows_needed: int | None = None
+    ) -> list[list[Scalar]]:
+        if self.book.read_only:
+            sheet.reset_dimensions()
+
+        data: list[list[Scalar]] = []
+        last_row_with_data = -1
+        for row_number, row in enumerate(sheet.rows):
+            converted_row = [self._convert_cell(cell) for cell in row]
+            while converted_row and converted_row[-1] == "":
+                # trim trailing empty elements
+                converted_row.pop()
+            if converted_row:
+                last_row_with_data = row_number
+            data.append(converted_row)
+            if file_rows_needed is not None and len(data) >= file_rows_needed:
+                break
+
+        # Trim trailing empty rows
+        data = data[: last_row_with_data + 1]
+
+        if len(data) > 0:
+            # extend rows to max width
+            max_width = max(len(data_row) for data_row in data)
+            if min(len(data_row) for data_row in data) < max_width:
+                empty_cell: list[Scalar] = [""]
+                data = [
+                    data_row + (max_width - len(data_row)) * empty_cell
+                    for data_row in data
+                ]
+
+        return data
diff --git a/pandas/pandas/io/excel/_pyxlsb.py b/pandas/pandas/io/excel/_pyxlsb.py
new file mode 100644
index 0000000000000..a6e42616c2043
--- /dev/null
+++ b/pandas/pandas/io/excel/_pyxlsb.py
@@ -0,0 +1,127 @@
+# pyright: reportMissingImports=false
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.excel._base import BaseExcelReader
+
+if TYPE_CHECKING:
+    from pyxlsb import Workbook
+
+    from pandas._typing import (
+        FilePath,
+        ReadBuffer,
+        Scalar,
+        StorageOptions,
+    )
+
+
+class PyxlsbReader(BaseExcelReader["Workbook"]):
+    @doc(storage_options=_shared_docs["storage_options"])
+    def __init__(
+        self,
+        filepath_or_buffer: FilePath | ReadBuffer[bytes],
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        Reader using pyxlsb engine.
+
+        Parameters
+        ----------
+        filepath_or_buffer : str, path object, or Workbook
+            Object to be parsed.
+        {storage_options}
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        import_optional_dependency("pyxlsb")
+        # This will call load_workbook on the filepath or buffer
+        # And set the result to the book-attribute
+        super().__init__(
+            filepath_or_buffer,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @property
+    def _workbook_class(self) -> type[Workbook]:
+        from pyxlsb import Workbook
+
+        return Workbook
+
+    def load_workbook(
+        self, filepath_or_buffer: FilePath | ReadBuffer[bytes], engine_kwargs
+    ) -> Workbook:
+        from pyxlsb import open_workbook
+
+        # TODO: hack in buffer capability
+        # This might need some modifications to the Pyxlsb library
+        # Actual work for opening it is in xlsbpackage.py, line 20-ish
+
+        return open_workbook(filepath_or_buffer, **engine_kwargs)
+
+    @property
+    def sheet_names(self) -> list[str]:
+        return self.book.sheets
+
+    def get_sheet_by_name(self, name: str):
+        self.raise_if_bad_sheet_by_name(name)
+        return self.book.get_sheet(name)
+
+    def get_sheet_by_index(self, index: int):
+        self.raise_if_bad_sheet_by_index(index)
+        # pyxlsb sheets are indexed from 1 onwards
+        # There's a fix for this in the source, but the pypi package doesn't have it
+        return self.book.get_sheet(index + 1)
+
+    def _convert_cell(self, cell) -> Scalar:
+        # TODO: there is no way to distinguish between floats and datetimes in pyxlsb
+        # This means that there is no way to read datetime types from an xlsb file yet
+        if cell.v is None:
+            return ""  # Prevents non-named columns from not showing up as Unnamed: i
+        if isinstance(cell.v, float):
+            val = int(cell.v)
+            if val == cell.v:
+                return val
+            else:
+                return float(cell.v)
+
+        return cell.v
+
+    def get_sheet_data(
+        self,
+        sheet,
+        file_rows_needed: int | None = None,
+    ) -> list[list[Scalar]]:
+        data: list[list[Scalar]] = []
+        previous_row_number = -1
+        # When sparse=True the rows can have different lengths and empty rows are
+        # not returned. The cells are namedtuples of row, col, value (r, c, v).
+        for row in sheet.rows(sparse=True):
+            row_number = row[0].r
+            converted_row = [self._convert_cell(cell) for cell in row]
+            while converted_row and converted_row[-1] == "":
+                # trim trailing empty elements
+                converted_row.pop()
+            if converted_row:
+                data.extend([[]] * (row_number - previous_row_number - 1))
+                data.append(converted_row)
+                previous_row_number = row_number
+            if file_rows_needed is not None and len(data) >= file_rows_needed:
+                break
+        if data:
+            # extend rows to max_width
+            max_width = max(len(data_row) for data_row in data)
+            if min(len(data_row) for data_row in data) < max_width:
+                empty_cell: list[Scalar] = [""]
+                data = [
+                    data_row + (max_width - len(data_row)) * empty_cell
+                    for data_row in data
+                ]
+        return data
diff --git a/pandas/pandas/io/excel/_util.py b/pandas/pandas/io/excel/_util.py
new file mode 100644
index 0000000000000..e7c5d518abaee
--- /dev/null
+++ b/pandas/pandas/io/excel/_util.py
@@ -0,0 +1,328 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    MutableMapping,
+    Sequence,
+)
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    TypeVar,
+    overload,
+)
+
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.dtypes.common import (
+    is_integer,
+    is_list_like,
+)
+
+if TYPE_CHECKING:
+    from pandas.io.excel._base import ExcelWriter
+
+    ExcelWriter_t = type[ExcelWriter]
+    usecols_func = TypeVar("usecols_func", bound=Callable[[Hashable], object])
+
+_writers: MutableMapping[str, ExcelWriter_t] = {}
+
+
+def register_writer(klass: ExcelWriter_t) -> None:
+    """
+    Add engine to the excel writer registry.io.excel.
+
+    You must use this method to integrate with ``to_excel``.
+
+    Parameters
+    ----------
+    klass : ExcelWriter
+    """
+    if not callable(klass):
+        raise ValueError("Can only register callables as engines")
+    engine_name = klass._engine
+    _writers[engine_name] = klass
+
+
+def get_default_engine(ext: str, mode: Literal["reader", "writer"] = "reader") -> str:
+    """
+    Return the default reader/writer for the given extension.
+
+    Parameters
+    ----------
+    ext : str
+        The excel file extension for which to get the default engine.
+    mode : str {'reader', 'writer'}
+        Whether to get the default engine for reading or writing.
+        Either 'reader' or 'writer'
+
+    Returns
+    -------
+    str
+        The default engine for the extension.
+    """
+    _default_readers = {
+        "xlsx": "openpyxl",
+        "xlsm": "openpyxl",
+        "xlsb": "pyxlsb",
+        "xls": "xlrd",
+        "ods": "odf",
+    }
+    _default_writers = {
+        "xlsx": "openpyxl",
+        "xlsm": "openpyxl",
+        "xlsb": "pyxlsb",
+        "ods": "odf",
+    }
+    assert mode in ["reader", "writer"]
+    if mode == "writer":
+        # Prefer xlsxwriter over openpyxl if installed
+        xlsxwriter = import_optional_dependency("xlsxwriter", errors="warn")
+        if xlsxwriter:
+            _default_writers["xlsx"] = "xlsxwriter"
+        return _default_writers[ext]
+    else:
+        return _default_readers[ext]
+
+
+def get_writer(engine_name: str) -> ExcelWriter_t:
+    try:
+        return _writers[engine_name]
+    except KeyError as err:
+        raise ValueError(f"No Excel writer '{engine_name}'") from err
+
+
+def _excel2num(x: str) -> int:
+    """
+    Convert Excel column name like 'AB' to 0-based column index.
+
+    Parameters
+    ----------
+    x : str
+        The Excel column name to convert to a 0-based column index.
+
+    Returns
+    -------
+    num : int
+        The column index corresponding to the name.
+
+    Raises
+    ------
+    ValueError
+        Part of the Excel column name was invalid.
+    """
+    index = 0
+
+    for c in x.upper().strip():
+        cp = ord(c)
+
+        if cp < ord("A") or cp > ord("Z"):
+            raise ValueError(f"Invalid column name: {x}")
+
+        index = index * 26 + cp - ord("A") + 1
+
+    return index - 1
+
+
+def _range2cols(areas: str) -> list[int]:
+    """
+    Convert comma separated list of column names and ranges to indices.
+
+    Parameters
+    ----------
+    areas : str
+        A string containing a sequence of column ranges (or areas).
+
+    Returns
+    -------
+    cols : list
+        A list of 0-based column indices.
+
+    Examples
+    --------
+    >>> _range2cols("A:E")
+    [0, 1, 2, 3, 4]
+    >>> _range2cols("A,C,Z:AB")
+    [0, 2, 25, 26, 27]
+    """
+    cols: list[int] = []
+
+    for rng in areas.split(","):
+        if ":" in rng:
+            rngs = rng.split(":")
+            cols.extend(range(_excel2num(rngs[0]), _excel2num(rngs[1]) + 1))
+        else:
+            cols.append(_excel2num(rng))
+
+    return cols
+
+
+@overload
+def maybe_convert_usecols(usecols: str | list[int]) -> list[int]: ...
+
+
+@overload
+def maybe_convert_usecols(usecols: list[str]) -> list[str]: ...
+
+
+@overload
+def maybe_convert_usecols(usecols: usecols_func) -> usecols_func: ...
+
+
+@overload
+def maybe_convert_usecols(usecols: None) -> None: ...
+
+
+def maybe_convert_usecols(
+    usecols: str | list[int] | list[str] | usecols_func | None,
+) -> None | list[int] | list[str] | usecols_func:
+    """
+    Convert `usecols` into a compatible format for parsing in `parsers.py`.
+
+    Parameters
+    ----------
+    usecols : object
+        The use-columns object to potentially convert.
+
+    Returns
+    -------
+    converted : object
+        The compatible format of `usecols`.
+    """
+    if usecols is None:
+        return usecols
+
+    if is_integer(usecols):
+        raise ValueError(
+            "Passing an integer for `usecols` is no longer supported.  "
+            "Please pass in a list of int from 0 to `usecols` inclusive instead."
+        )
+
+    if isinstance(usecols, str):
+        return _range2cols(usecols)
+
+    return usecols
+
+
+@overload
+def validate_freeze_panes(freeze_panes: tuple[int, int]) -> Literal[True]: ...
+
+
+@overload
+def validate_freeze_panes(freeze_panes: None) -> Literal[False]: ...
+
+
+def validate_freeze_panes(freeze_panes: tuple[int, int] | None) -> bool:
+    if freeze_panes is not None:
+        if len(freeze_panes) == 2 and all(
+            isinstance(item, int) for item in freeze_panes
+        ):
+            return True
+
+        raise ValueError(
+            "freeze_panes must be of form (row, column) "
+            "where row and column are integers"
+        )
+
+    # freeze_panes wasn't specified, return False so it won't be applied
+    # to output sheet
+    return False
+
+
+def fill_mi_header(
+    row: list[Hashable], control_row: list[bool]
+) -> tuple[list[Hashable], list[bool]]:
+    """
+    Forward fill blank entries in row but only inside the same parent index.
+
+    Used for creating headers in Multiindex.
+
+    Parameters
+    ----------
+    row : list
+        List of items in a single row.
+    control_row : list of bool
+        Helps to determine if particular column is in same parent index as the
+        previous value. Used to stop propagation of empty cells between
+        different indexes.
+
+    Returns
+    -------
+    Returns changed row and control_row
+    """
+    last = row[0]
+    for i in range(1, len(row)):
+        if not control_row[i]:
+            last = row[i]
+
+        if row[i] == "" or row[i] is None:
+            row[i] = last
+        else:
+            control_row[i] = False
+            last = row[i]
+
+    return row, control_row
+
+
+def pop_header_name(
+    row: list[Hashable], index_col: int | Sequence[int]
+) -> tuple[Hashable | None, list[Hashable]]:
+    """
+    Pop the header name for MultiIndex parsing.
+
+    Parameters
+    ----------
+    row : list
+        The data row to parse for the header name.
+    index_col : int, list
+        The index columns for our data. Assumed to be non-null.
+
+    Returns
+    -------
+    header_name : str
+        The extracted header name.
+    trimmed_row : list
+        The original data row with the header name removed.
+    """
+    # Pop out header name and fill w/blank.
+    if is_list_like(index_col):
+        assert isinstance(index_col, Iterable)
+        i = max(index_col)
+    else:
+        assert not isinstance(index_col, Iterable)
+        i = index_col
+
+    header_name = row[i]
+    header_name = None if header_name == "" else header_name
+
+    return header_name, row[:i] + [""] + row[i + 1 :]
+
+
+def combine_kwargs(engine_kwargs: dict[str, Any] | None, kwargs: dict) -> dict:
+    """
+    Used to combine two sources of kwargs for the backend engine.
+
+    Use of kwargs is deprecated, this function is solely for use in 1.3 and should
+    be removed in 1.4/2.0. Also _base.ExcelWriter.__new__ ensures either engine_kwargs
+    or kwargs must be None or empty respectively.
+
+    Parameters
+    ----------
+    engine_kwargs: dict
+        kwargs to be passed through to the engine.
+    kwargs: dict
+        kwargs to be psased through to the engine (deprecated)
+
+    Returns
+    -------
+    engine_kwargs combined with kwargs
+    """
+    if engine_kwargs is None:
+        result = {}
+    else:
+        result = engine_kwargs.copy()
+    result.update(kwargs)
+    return result
diff --git a/pandas/pandas/io/excel/_xlrd.py b/pandas/pandas/io/excel/_xlrd.py
new file mode 100644
index 0000000000000..5d39a840336eb
--- /dev/null
+++ b/pandas/pandas/io/excel/_xlrd.py
@@ -0,0 +1,140 @@
+from __future__ import annotations
+
+from datetime import time
+import math
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.excel._base import BaseExcelReader
+
+if TYPE_CHECKING:
+    from xlrd import Book
+
+    from pandas._typing import (
+        Scalar,
+        StorageOptions,
+    )
+
+
+class XlrdReader(BaseExcelReader["Book"]):
+    @doc(storage_options=_shared_docs["storage_options"])
+    def __init__(
+        self,
+        filepath_or_buffer,
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        Reader using xlrd engine.
+
+        Parameters
+        ----------
+        filepath_or_buffer : str, path object or Workbook
+            Object to be parsed.
+        {storage_options}
+        engine_kwargs : dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        err_msg = "Install xlrd >= 2.0.1 for xls Excel support"
+        import_optional_dependency("xlrd", extra=err_msg)
+        super().__init__(
+            filepath_or_buffer,
+            storage_options=storage_options,
+            engine_kwargs=engine_kwargs,
+        )
+
+    @property
+    def _workbook_class(self) -> type[Book]:
+        from xlrd import Book
+
+        return Book
+
+    def load_workbook(self, filepath_or_buffer, engine_kwargs) -> Book:
+        from xlrd import open_workbook
+
+        if hasattr(filepath_or_buffer, "read"):
+            data = filepath_or_buffer.read()
+            return open_workbook(file_contents=data, **engine_kwargs)
+        else:
+            return open_workbook(filepath_or_buffer, **engine_kwargs)
+
+    @property
+    def sheet_names(self):
+        return self.book.sheet_names()
+
+    def get_sheet_by_name(self, name):
+        self.raise_if_bad_sheet_by_name(name)
+        return self.book.sheet_by_name(name)
+
+    def get_sheet_by_index(self, index):
+        self.raise_if_bad_sheet_by_index(index)
+        return self.book.sheet_by_index(index)
+
+    def get_sheet_data(
+        self, sheet, file_rows_needed: int | None = None
+    ) -> list[list[Scalar]]:
+        from xlrd import (
+            XL_CELL_BOOLEAN,
+            XL_CELL_DATE,
+            XL_CELL_ERROR,
+            XL_CELL_NUMBER,
+            xldate,
+        )
+
+        epoch1904 = self.book.datemode
+
+        def _parse_cell(cell_contents, cell_typ):
+            """
+            converts the contents of the cell into a pandas appropriate object
+            """
+            if cell_typ == XL_CELL_DATE:
+                # Use the newer xlrd datetime handling.
+                try:
+                    cell_contents = xldate.xldate_as_datetime(cell_contents, epoch1904)
+                except OverflowError:
+                    return cell_contents
+
+                # Excel doesn't distinguish between dates and time,
+                # so we treat dates on the epoch as times only.
+                # Also, Excel supports 1900 and 1904 epochs.
+                year = (cell_contents.timetuple())[0:3]
+                if (not epoch1904 and year == (1899, 12, 31)) or (
+                    epoch1904 and year == (1904, 1, 1)
+                ):
+                    cell_contents = time(
+                        cell_contents.hour,
+                        cell_contents.minute,
+                        cell_contents.second,
+                        cell_contents.microsecond,
+                    )
+
+            elif cell_typ == XL_CELL_ERROR:
+                cell_contents = np.nan
+            elif cell_typ == XL_CELL_BOOLEAN:
+                cell_contents = bool(cell_contents)
+            elif cell_typ == XL_CELL_NUMBER:
+                # GH5394 - Excel 'numbers' are always floats
+                # it's a minimal perf hit and less surprising
+                if math.isfinite(cell_contents):
+                    # GH54564 - don't attempt to convert NaN/Inf
+                    val = int(cell_contents)
+                    if val == cell_contents:
+                        cell_contents = val
+            return cell_contents
+
+        nrows = sheet.nrows
+        if file_rows_needed is not None:
+            nrows = min(nrows, file_rows_needed)
+        return [
+            [
+                _parse_cell(value, typ)
+                for value, typ in zip(sheet.row_values(i), sheet.row_types(i))
+            ]
+            for i in range(nrows)
+        ]
diff --git a/pandas/pandas/io/excel/_xlsxwriter.py b/pandas/pandas/io/excel/_xlsxwriter.py
new file mode 100644
index 0000000000000..4a7b8eee2bfce
--- /dev/null
+++ b/pandas/pandas/io/excel/_xlsxwriter.py
@@ -0,0 +1,284 @@
+from __future__ import annotations
+
+import json
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas.io.excel._base import ExcelWriter
+from pandas.io.excel._util import (
+    combine_kwargs,
+    validate_freeze_panes,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ExcelWriterIfSheetExists,
+        FilePath,
+        StorageOptions,
+        WriteExcelBuffer,
+    )
+
+
+class _XlsxStyler:
+    # Map from openpyxl-oriented styles to flatter xlsxwriter representation
+    # Ordering necessary for both determinism and because some are keyed by
+    # prefixes of others.
+    STYLE_MAPPING: dict[str, list[tuple[tuple[str, ...], str]]] = {
+        "font": [
+            (("name",), "font_name"),
+            (("sz",), "font_size"),
+            (("size",), "font_size"),
+            (("color", "rgb"), "font_color"),
+            (("color",), "font_color"),
+            (("b",), "bold"),
+            (("bold",), "bold"),
+            (("i",), "italic"),
+            (("italic",), "italic"),
+            (("u",), "underline"),
+            (("underline",), "underline"),
+            (("strike",), "font_strikeout"),
+            (("vertAlign",), "font_script"),
+            (("vertalign",), "font_script"),
+        ],
+        "number_format": [(("format_code",), "num_format"), ((), "num_format")],
+        "protection": [(("locked",), "locked"), (("hidden",), "hidden")],
+        "alignment": [
+            (("horizontal",), "align"),
+            (("vertical",), "valign"),
+            (("text_rotation",), "rotation"),
+            (("wrap_text",), "text_wrap"),
+            (("indent",), "indent"),
+            (("shrink_to_fit",), "shrink"),
+        ],
+        "fill": [
+            (("patternType",), "pattern"),
+            (("patterntype",), "pattern"),
+            (("fill_type",), "pattern"),
+            (("start_color", "rgb"), "fg_color"),
+            (("fgColor", "rgb"), "fg_color"),
+            (("fgcolor", "rgb"), "fg_color"),
+            (("start_color",), "fg_color"),
+            (("fgColor",), "fg_color"),
+            (("fgcolor",), "fg_color"),
+            (("end_color", "rgb"), "bg_color"),
+            (("bgColor", "rgb"), "bg_color"),
+            (("bgcolor", "rgb"), "bg_color"),
+            (("end_color",), "bg_color"),
+            (("bgColor",), "bg_color"),
+            (("bgcolor",), "bg_color"),
+        ],
+        "border": [
+            (("color", "rgb"), "border_color"),
+            (("color",), "border_color"),
+            (("style",), "border"),
+            (("top", "color", "rgb"), "top_color"),
+            (("top", "color"), "top_color"),
+            (("top", "style"), "top"),
+            (("top",), "top"),
+            (("right", "color", "rgb"), "right_color"),
+            (("right", "color"), "right_color"),
+            (("right", "style"), "right"),
+            (("right",), "right"),
+            (("bottom", "color", "rgb"), "bottom_color"),
+            (("bottom", "color"), "bottom_color"),
+            (("bottom", "style"), "bottom"),
+            (("bottom",), "bottom"),
+            (("left", "color", "rgb"), "left_color"),
+            (("left", "color"), "left_color"),
+            (("left", "style"), "left"),
+            (("left",), "left"),
+        ],
+    }
+
+    @classmethod
+    def convert(cls, style_dict, num_format_str=None) -> dict[str, Any]:
+        """
+        converts a style_dict to an xlsxwriter format dict
+
+        Parameters
+        ----------
+        style_dict : style dictionary to convert
+        num_format_str : optional number format string
+        """
+        # Create a XlsxWriter format object.
+        props = {}
+
+        if num_format_str is not None:
+            props["num_format"] = num_format_str
+
+        if style_dict is None:
+            return props
+
+        if "borders" in style_dict:
+            style_dict = style_dict.copy()
+            style_dict["border"] = style_dict.pop("borders")
+
+        for style_group_key, style_group in style_dict.items():
+            for src, dst in cls.STYLE_MAPPING.get(style_group_key, []):
+                # src is a sequence of keys into a nested dict
+                # dst is a flat key
+                if dst in props:
+                    continue
+                v = style_group
+                for k in src:
+                    try:
+                        v = v[k]
+                    except (KeyError, TypeError):
+                        break
+                else:
+                    props[dst] = v
+
+        if isinstance(props.get("pattern"), str):
+            # TODO: support other fill patterns
+            props["pattern"] = 0 if props["pattern"] == "none" else 1
+
+        for k in ["border", "top", "right", "bottom", "left"]:
+            if isinstance(props.get(k), str):
+                try:
+                    props[k] = [
+                        "none",
+                        "thin",
+                        "medium",
+                        "dashed",
+                        "dotted",
+                        "thick",
+                        "double",
+                        "hair",
+                        "mediumDashed",
+                        "dashDot",
+                        "mediumDashDot",
+                        "dashDotDot",
+                        "mediumDashDotDot",
+                        "slantDashDot",
+                    ].index(props[k])
+                except ValueError:
+                    props[k] = 2
+
+        if isinstance(props.get("font_script"), str):
+            props["font_script"] = ["baseline", "superscript", "subscript"].index(
+                props["font_script"]
+            )
+
+        if isinstance(props.get("underline"), str):
+            props["underline"] = {
+                "none": 0,
+                "single": 1,
+                "double": 2,
+                "singleAccounting": 33,
+                "doubleAccounting": 34,
+            }[props["underline"]]
+
+        # GH 30107 - xlsxwriter uses different name
+        if props.get("valign") == "center":
+            props["valign"] = "vcenter"
+
+        return props
+
+
+class XlsxWriter(ExcelWriter):
+    _engine = "xlsxwriter"
+    _supported_extensions = (".xlsx",)
+
+    def __init__(  # pyright: ignore[reportInconsistentConstructor]
+        self,
+        path: FilePath | WriteExcelBuffer | ExcelWriter,
+        engine: str | None = None,
+        date_format: str | None = None,
+        datetime_format: str | None = None,
+        mode: str = "w",
+        storage_options: StorageOptions | None = None,
+        if_sheet_exists: ExcelWriterIfSheetExists | None = None,
+        engine_kwargs: dict[str, Any] | None = None,
+        **kwargs,
+    ) -> None:
+        # Use the xlsxwriter module as the Excel writer.
+        from xlsxwriter import Workbook
+
+        engine_kwargs = combine_kwargs(engine_kwargs, kwargs)
+
+        if mode == "a":
+            raise ValueError("Append mode is not supported with xlsxwriter!")
+
+        super().__init__(
+            path,
+            engine=engine,
+            date_format=date_format,
+            datetime_format=datetime_format,
+            mode=mode,
+            storage_options=storage_options,
+            if_sheet_exists=if_sheet_exists,
+            engine_kwargs=engine_kwargs,
+        )
+
+        try:
+            self._book = Workbook(self._handles.handle, **engine_kwargs)
+        except TypeError:
+            self._handles.handle.close()
+            raise
+
+    @property
+    def book(self):
+        """
+        Book instance of class xlsxwriter.Workbook.
+
+        This attribute can be used to access engine-specific features.
+        """
+        return self._book
+
+    @property
+    def sheets(self) -> dict[str, Any]:
+        result = self.book.sheetnames
+        return result
+
+    def _save(self) -> None:
+        """
+        Save workbook to disk.
+        """
+        self.book.close()
+
+    def _write_cells(
+        self,
+        cells,
+        sheet_name: str | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        freeze_panes: tuple[int, int] | None = None,
+    ) -> None:
+        # Write the frame cells using xlsxwriter.
+        sheet_name = self._get_sheet_name(sheet_name)
+
+        wks = self.book.get_worksheet_by_name(sheet_name)
+        if wks is None:
+            wks = self.book.add_worksheet(sheet_name)
+
+        style_dict = {"null": None}
+
+        if validate_freeze_panes(freeze_panes):
+            wks.freeze_panes(*(freeze_panes))
+
+        for cell in cells:
+            val, fmt = self._value_with_fmt(cell.val)
+
+            stylekey = json.dumps(cell.style)
+            if fmt:
+                stylekey += fmt
+
+            if stylekey in style_dict:
+                style = style_dict[stylekey]
+            else:
+                style = self.book.add_format(_XlsxStyler.convert(cell.style, fmt))
+                style_dict[stylekey] = style
+
+            if cell.mergestart is not None and cell.mergeend is not None:
+                wks.merge_range(
+                    startrow + cell.row,
+                    startcol + cell.col,
+                    startrow + cell.mergestart,
+                    startcol + cell.mergeend,
+                    val,
+                    style,
+                )
+            else:
+                wks.write(startrow + cell.row, startcol + cell.col, val, style)
diff --git a/pandas/pandas/io/feather_format.py b/pandas/pandas/io/feather_format.py
new file mode 100644
index 0000000000000..565c53f0f3fc5
--- /dev/null
+++ b/pandas/pandas/io/feather_format.py
@@ -0,0 +1,157 @@
+"""feather-format compat"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import warnings
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import doc
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.api import DataFrame
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io._util import arrow_table_to_pandas
+from pandas.io.common import get_handle
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        DtypeBackend,
+        FilePath,
+        ReadBuffer,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def to_feather(
+    df: DataFrame,
+    path: FilePath | WriteBuffer[bytes],
+    storage_options: StorageOptions | None = None,
+    **kwargs: Any,
+) -> None:
+    """
+    Write a DataFrame to the binary Feather format.
+
+    Parameters
+    ----------
+    df : DataFrame
+    path : str, path object, or file-like object
+    {storage_options}
+    **kwargs :
+        Additional keywords passed to `pyarrow.feather.write_feather`.
+
+    """
+    import_optional_dependency("pyarrow")
+    from pyarrow import feather
+
+    if not isinstance(df, DataFrame):
+        raise ValueError("feather only support IO with DataFrames")
+
+    with get_handle(
+        path, "wb", storage_options=storage_options, is_text=False
+    ) as handles:
+        feather.write_feather(df, handles.handle, **kwargs)
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def read_feather(
+    path: FilePath | ReadBuffer[bytes],
+    columns: Sequence[Hashable] | None = None,
+    use_threads: bool = True,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame:
+    """
+    Load a feather-format object from the file path.
+
+    Feather is particularly useful for scenarios that require efficient
+    serialization and deserialization of tabular data. It supports
+    schema preservation, making it a reliable choice for use cases
+    such as sharing data between Python and R, or persisting intermediate
+    results during data processing pipelines. This method provides additional
+    flexibility with options for selective column reading, thread parallelism,
+    and choosing the backend for data types.
+
+    Parameters
+    ----------
+    path : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``read()`` function. The string could be a URL.
+        Valid URL schemes include http, ftp, s3, and file. For file URLs, a host is
+        expected. A local file could be: ``file://localhost/path/to/table.feather``.
+    columns : sequence, default None
+        If not provided, all columns are read.
+    use_threads : bool, default True
+        Whether to parallelize reading using multiple threads.
+    {storage_options}
+
+    dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`.
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    type of object stored in file
+        DataFrame object stored in the file.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (csv) file into a pandas DataFrame.
+    read_excel : Read an Excel file into a pandas DataFrame.
+    read_spss : Read an SPSS file into a pandas DataFrame.
+    read_orc : Load an ORC object into a pandas DataFrame.
+    read_sas : Read SAS file into a pandas DataFrame.
+
+    Examples
+    --------
+    >>> df = pd.read_feather("path/to/file.feather")  # doctest: +SKIP
+    """
+    import_optional_dependency("pyarrow")
+    from pyarrow import feather
+
+    # import utils to register the pyarrow extension types
+    import pandas.core.arrays.arrow.extension_types  # pyright: ignore[reportUnusedImport] # noqa: F401
+
+    check_dtype_backend(dtype_backend)
+
+    with get_handle(
+        path, "rb", storage_options=storage_options, is_text=False
+    ) as handles:
+        if dtype_backend is lib.no_default and not using_string_dtype():
+            with warnings.catch_warnings():
+                warnings.filterwarnings(
+                    "ignore",
+                    "make_block is deprecated",
+                    DeprecationWarning,
+                )
+
+                return feather.read_feather(
+                    handles.handle, columns=columns, use_threads=bool(use_threads)
+                )
+
+        pa_table = feather.read_table(
+            handles.handle, columns=columns, use_threads=bool(use_threads)
+        )
+        return arrow_table_to_pandas(pa_table, dtype_backend=dtype_backend)
diff --git a/pandas/pandas/io/formats/__init__.py b/pandas/pandas/io/formats/__init__.py
new file mode 100644
index 0000000000000..895669c342f97
--- /dev/null
+++ b/pandas/pandas/io/formats/__init__.py
@@ -0,0 +1,9 @@
+# ruff: noqa: TC004
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    # import modules that have public classes/functions
+    from pandas.io.formats import style
+
+    # and mark only those modules as public
+    __all__ = ["style"]
diff --git a/pandas/pandas/io/formats/_color_data.py b/pandas/pandas/io/formats/_color_data.py
new file mode 100644
index 0000000000000..2e7cb7f29646e
--- /dev/null
+++ b/pandas/pandas/io/formats/_color_data.py
@@ -0,0 +1,157 @@
+# GH37967: Enable the use of CSS named colors, as defined in
+# matplotlib.colors.CSS4_COLORS, when exporting to Excel.
+# This data has been copied here, instead of being imported from matplotlib,
+# not to have ``to_excel`` methods require matplotlib.
+# source: matplotlib._color_data (3.3.3)
+from __future__ import annotations
+
+CSS4_COLORS = {
+    "aliceblue": "F0F8FF",
+    "antiquewhite": "FAEBD7",
+    "aqua": "00FFFF",
+    "aquamarine": "7FFFD4",
+    "azure": "F0FFFF",
+    "beige": "F5F5DC",
+    "bisque": "FFE4C4",
+    "black": "000000",
+    "blanchedalmond": "FFEBCD",
+    "blue": "0000FF",
+    "blueviolet": "8A2BE2",
+    "brown": "A52A2A",
+    "burlywood": "DEB887",
+    "cadetblue": "5F9EA0",
+    "chartreuse": "7FFF00",
+    "chocolate": "D2691E",
+    "coral": "FF7F50",
+    "cornflowerblue": "6495ED",
+    "cornsilk": "FFF8DC",
+    "crimson": "DC143C",
+    "cyan": "00FFFF",
+    "darkblue": "00008B",
+    "darkcyan": "008B8B",
+    "darkgoldenrod": "B8860B",
+    "darkgray": "A9A9A9",
+    "darkgreen": "006400",
+    "darkgrey": "A9A9A9",
+    "darkkhaki": "BDB76B",
+    "darkmagenta": "8B008B",
+    "darkolivegreen": "556B2F",
+    "darkorange": "FF8C00",
+    "darkorchid": "9932CC",
+    "darkred": "8B0000",
+    "darksalmon": "E9967A",
+    "darkseagreen": "8FBC8F",
+    "darkslateblue": "483D8B",
+    "darkslategray": "2F4F4F",
+    "darkslategrey": "2F4F4F",
+    "darkturquoise": "00CED1",
+    "darkviolet": "9400D3",
+    "deeppink": "FF1493",
+    "deepskyblue": "00BFFF",
+    "dimgray": "696969",
+    "dimgrey": "696969",
+    "dodgerblue": "1E90FF",
+    "firebrick": "B22222",
+    "floralwhite": "FFFAF0",
+    "forestgreen": "228B22",
+    "fuchsia": "FF00FF",
+    "gainsboro": "DCDCDC",
+    "ghostwhite": "F8F8FF",
+    "gold": "FFD700",
+    "goldenrod": "DAA520",
+    "gray": "808080",
+    "green": "008000",
+    "greenyellow": "ADFF2F",
+    "grey": "808080",
+    "honeydew": "F0FFF0",
+    "hotpink": "FF69B4",
+    "indianred": "CD5C5C",
+    "indigo": "4B0082",
+    "ivory": "FFFFF0",
+    "khaki": "F0E68C",
+    "lavender": "E6E6FA",
+    "lavenderblush": "FFF0F5",
+    "lawngreen": "7CFC00",
+    "lemonchiffon": "FFFACD",
+    "lightblue": "ADD8E6",
+    "lightcoral": "F08080",
+    "lightcyan": "E0FFFF",
+    "lightgoldenrodyellow": "FAFAD2",
+    "lightgray": "D3D3D3",
+    "lightgreen": "90EE90",
+    "lightgrey": "D3D3D3",
+    "lightpink": "FFB6C1",
+    "lightsalmon": "FFA07A",
+    "lightseagreen": "20B2AA",
+    "lightskyblue": "87CEFA",
+    "lightslategray": "778899",
+    "lightslategrey": "778899",
+    "lightsteelblue": "B0C4DE",
+    "lightyellow": "FFFFE0",
+    "lime": "00FF00",
+    "limegreen": "32CD32",
+    "linen": "FAF0E6",
+    "magenta": "FF00FF",
+    "maroon": "800000",
+    "mediumaquamarine": "66CDAA",
+    "mediumblue": "0000CD",
+    "mediumorchid": "BA55D3",
+    "mediumpurple": "9370DB",
+    "mediumseagreen": "3CB371",
+    "mediumslateblue": "7B68EE",
+    "mediumspringgreen": "00FA9A",
+    "mediumturquoise": "48D1CC",
+    "mediumvioletred": "C71585",
+    "midnightblue": "191970",
+    "mintcream": "F5FFFA",
+    "mistyrose": "FFE4E1",
+    "moccasin": "FFE4B5",
+    "navajowhite": "FFDEAD",
+    "navy": "000080",
+    "oldlace": "FDF5E6",
+    "olive": "808000",
+    "olivedrab": "6B8E23",
+    "orange": "FFA500",
+    "orangered": "FF4500",
+    "orchid": "DA70D6",
+    "palegoldenrod": "EEE8AA",
+    "palegreen": "98FB98",
+    "paleturquoise": "AFEEEE",
+    "palevioletred": "DB7093",
+    "papayawhip": "FFEFD5",
+    "peachpuff": "FFDAB9",
+    "peru": "CD853F",
+    "pink": "FFC0CB",
+    "plum": "DDA0DD",
+    "powderblue": "B0E0E6",
+    "purple": "800080",
+    "rebeccapurple": "663399",
+    "red": "FF0000",
+    "rosybrown": "BC8F8F",
+    "royalblue": "4169E1",
+    "saddlebrown": "8B4513",
+    "salmon": "FA8072",
+    "sandybrown": "F4A460",
+    "seagreen": "2E8B57",
+    "seashell": "FFF5EE",
+    "sienna": "A0522D",
+    "silver": "C0C0C0",
+    "skyblue": "87CEEB",
+    "slateblue": "6A5ACD",
+    "slategray": "708090",
+    "slategrey": "708090",
+    "snow": "FFFAFA",
+    "springgreen": "00FF7F",
+    "steelblue": "4682B4",
+    "tan": "D2B48C",
+    "teal": "008080",
+    "thistle": "D8BFD8",
+    "tomato": "FF6347",
+    "turquoise": "40E0D0",
+    "violet": "EE82EE",
+    "wheat": "F5DEB3",
+    "white": "FFFFFF",
+    "whitesmoke": "F5F5F5",
+    "yellow": "FFFF00",
+    "yellowgreen": "9ACD32",
+}
diff --git a/pandas/pandas/io/formats/console.py b/pandas/pandas/io/formats/console.py
new file mode 100644
index 0000000000000..d76593b41a996
--- /dev/null
+++ b/pandas/pandas/io/formats/console.py
@@ -0,0 +1,95 @@
+"""
+Internal module for console introspection
+"""
+
+from __future__ import annotations
+
+from shutil import get_terminal_size
+
+
+def get_console_size() -> tuple[int | None, int | None]:
+    """
+    Return console size as tuple = (width, height).
+
+    Returns (None,None) in non-interactive session.
+    """
+    from pandas import get_option
+
+    display_width = get_option("display.width")
+    display_height = get_option("display.max_rows")
+
+    # Consider
+    # interactive shell terminal, can detect term size
+    # interactive non-shell terminal (ipnb/ipqtconsole), cannot detect term
+    # size non-interactive script, should disregard term size
+
+    # in addition
+    # width,height have default values, but setting to 'None' signals
+    # should use Auto-Detection, But only in interactive shell-terminal.
+    # Simple. yeah.
+
+    if in_interactive_session():
+        if in_ipython_frontend():
+            # sane defaults for interactive non-shell terminal
+            # match default for width,height in config_init
+            from pandas._config.config import get_default_val
+
+            terminal_width = get_default_val("display.width")
+            terminal_height = get_default_val("display.max_rows")
+        else:
+            # pure terminal
+            terminal_width, terminal_height = get_terminal_size()
+    else:
+        terminal_width, terminal_height = None, None
+
+    # Note if the User sets width/Height to None (auto-detection)
+    # and we're in a script (non-inter), this will return (None,None)
+    # caller needs to deal.
+    return display_width or terminal_width, display_height or terminal_height
+
+
+# ----------------------------------------------------------------------
+# Detect our environment
+
+
+def in_interactive_session() -> bool:
+    """
+    Check if we're running in an interactive shell.
+
+    Returns
+    -------
+    bool
+        True if running under python/ipython interactive shell.
+    """
+    from pandas import get_option
+
+    def check_main() -> bool:
+        try:
+            import __main__ as main
+        except ModuleNotFoundError:
+            return get_option("mode.sim_interactive")
+        return not hasattr(main, "__file__") or get_option("mode.sim_interactive")
+
+    try:
+        # error: Name '__IPYTHON__' is not defined
+        return __IPYTHON__ or check_main()  # type: ignore[name-defined]
+    except NameError:
+        return check_main()
+
+
+def in_ipython_frontend() -> bool:
+    """
+    Check if we're inside an IPython zmq frontend.
+
+    Returns
+    -------
+    bool
+    """
+    try:
+        # error: Name 'get_ipython' is not defined
+        ip = get_ipython()  # type: ignore[name-defined]
+        return "zmq" in str(type(ip)).lower()
+    except NameError:
+        pass
+
+    return False
diff --git a/pandas/pandas/io/formats/css.py b/pandas/pandas/io/formats/css.py
new file mode 100644
index 0000000000000..10c970887e03b
--- /dev/null
+++ b/pandas/pandas/io/formats/css.py
@@ -0,0 +1,425 @@
+"""
+Utilities for interpreting CSS from Stylers for formatting non-HTML outputs.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import TYPE_CHECKING
+import warnings
+
+from pandas.errors import CSSWarning
+from pandas.util._exceptions import find_stack_level
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Generator,
+        Iterable,
+        Iterator,
+    )
+
+
+def _side_expander(prop_fmt: str) -> Callable:
+    """
+    Wrapper to expand shorthand property into top, right, bottom, left properties
+
+    Parameters
+    ----------
+    side : str
+        The border side to expand into properties
+
+    Returns
+    -------
+        function: Return to call when a 'border(-{side}): {value}' string is encountered
+    """
+
+    def expand(self: CSSResolver, prop: str, value: str) -> Generator[tuple[str, str]]:
+        """
+        Expand shorthand property into side-specific property (top, right, bottom, left)
+
+        Parameters
+        ----------
+            prop (str): CSS property name
+            value (str): String token for property
+
+        Yields
+        ------
+            Tuple (str, str): Expanded property, value
+        """
+        tokens = value.split()
+        try:
+            mapping = self.SIDE_SHORTHANDS[len(tokens)]
+        except KeyError:
+            warnings.warn(
+                f'Could not expand "{prop}: {value}"',
+                CSSWarning,
+                stacklevel=find_stack_level(),
+            )
+            return
+        for key, idx in zip(self.SIDES, mapping):
+            yield prop_fmt.format(key), tokens[idx]
+
+    return expand
+
+
+def _border_expander(side: str = "") -> Callable:
+    """
+    Wrapper to expand 'border' property into border color, style, and width properties
+
+    Parameters
+    ----------
+    side : str
+        The border side to expand into properties
+
+    Returns
+    -------
+        function: Return to call when a 'border(-{side}): {value}' string is encountered
+    """
+    if side != "":
+        side = f"-{side}"
+
+    def expand(self: CSSResolver, prop: str, value: str) -> Generator[tuple[str, str]]:
+        """
+        Expand border into color, style, and width tuples
+
+        Parameters
+        ----------
+            prop : str
+                CSS property name passed to styler
+            value : str
+                Value passed to styler for property
+
+        Yields
+        ------
+            Tuple (str, str): Expanded property, value
+        """
+        tokens = value.split()
+        if len(tokens) == 0 or len(tokens) > 3:
+            warnings.warn(
+                f'Too many tokens provided to "{prop}" (expected 1-3)',
+                CSSWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        # TODO: Can we use current color as initial value to comply with CSS standards?
+        border_declarations = {
+            f"border{side}-color": "black",
+            f"border{side}-style": "none",
+            f"border{side}-width": "medium",
+        }
+        for token in tokens:
+            if token.lower() in self.BORDER_STYLES:
+                border_declarations[f"border{side}-style"] = token
+            elif any(ratio in token.lower() for ratio in self.BORDER_WIDTH_RATIOS):
+                border_declarations[f"border{side}-width"] = token
+            else:
+                border_declarations[f"border{side}-color"] = token
+            # TODO: Warn user if item entered more than once (e.g. "border: red green")
+
+        # Per CSS, "border" will reset previous "border-*" definitions
+        yield from self.atomize(border_declarations.items())
+
+    return expand
+
+
+class CSSResolver:
+    """
+    A callable for parsing and resolving CSS to atomic properties.
+    """
+
+    UNIT_RATIOS = {
+        "pt": ("pt", 1),
+        "em": ("em", 1),
+        "rem": ("pt", 12),
+        "ex": ("em", 0.5),
+        # 'ch':
+        "px": ("pt", 0.75),
+        "pc": ("pt", 12),
+        "in": ("pt", 72),
+        "cm": ("in", 1 / 2.54),
+        "mm": ("in", 1 / 25.4),
+        "q": ("mm", 0.25),
+        "!!default": ("em", 0),
+    }
+
+    FONT_SIZE_RATIOS = UNIT_RATIOS.copy()
+    FONT_SIZE_RATIOS.update(
+        {
+            "%": ("em", 0.01),
+            "xx-small": ("rem", 0.5),
+            "x-small": ("rem", 0.625),
+            "small": ("rem", 0.8),
+            "medium": ("rem", 1),
+            "large": ("rem", 1.125),
+            "x-large": ("rem", 1.5),
+            "xx-large": ("rem", 2),
+            "smaller": ("em", 1 / 1.2),
+            "larger": ("em", 1.2),
+            "!!default": ("em", 1),
+        }
+    )
+
+    MARGIN_RATIOS = UNIT_RATIOS.copy()
+    MARGIN_RATIOS.update({"none": ("pt", 0)})
+
+    BORDER_WIDTH_RATIOS = UNIT_RATIOS.copy()
+    BORDER_WIDTH_RATIOS.update(
+        {
+            "none": ("pt", 0),
+            "thick": ("px", 4),
+            "medium": ("px", 2),
+            "thin": ("px", 1),
+            # Default: medium only if solid
+        }
+    )
+
+    BORDER_STYLES = [
+        "none",
+        "hidden",
+        "dotted",
+        "dashed",
+        "solid",
+        "double",
+        "groove",
+        "ridge",
+        "inset",
+        "outset",
+        "mediumdashdot",
+        "dashdotdot",
+        "hair",
+        "mediumdashdotdot",
+        "dashdot",
+        "slantdashdot",
+        "mediumdashed",
+    ]
+
+    SIDE_SHORTHANDS = {
+        1: [0, 0, 0, 0],
+        2: [0, 1, 0, 1],
+        3: [0, 1, 2, 1],
+        4: [0, 1, 2, 3],
+    }
+
+    SIDES = ("top", "right", "bottom", "left")
+
+    CSS_EXPANSIONS = {
+        **{
+            (f"border-{prop}" if prop else "border"): _border_expander(prop)
+            for prop in ["", "top", "right", "bottom", "left"]
+        },
+        **{
+            f"border-{prop}": _side_expander(f"border-{{:s}}-{prop}")
+            for prop in ["color", "style", "width"]
+        },
+        "margin": _side_expander("margin-{:s}"),
+        "padding": _side_expander("padding-{:s}"),
+    }
+
+    def __call__(
+        self,
+        declarations: str | Iterable[tuple[str, str]],
+        inherited: dict[str, str] | None = None,
+    ) -> dict[str, str]:
+        """
+        The given declarations to atomic properties.
+
+        Parameters
+        ----------
+        declarations_str : str | Iterable[tuple[str, str]]
+            A CSS string or set of CSS declaration tuples
+            e.g. "font-weight: bold; background: blue" or
+            {("font-weight", "bold"), ("background", "blue")}
+        inherited : dict, optional
+            Atomic properties indicating the inherited style context in which
+            declarations_str is to be resolved. ``inherited`` should already
+            be resolved, i.e. valid output of this method.
+
+        Returns
+        -------
+        dict
+            Atomic CSS 2.2 properties.
+
+        Examples
+        --------
+        >>> resolve = CSSResolver()
+        >>> inherited = {"font-family": "serif", "font-weight": "bold"}
+        >>> out = resolve(
+        ...     '''
+        ...               border-color: BLUE RED;
+        ...               font-size: 1em;
+        ...               font-size: 2em;
+        ...               font-weight: normal;
+        ...               font-weight: inherit;
+        ...               ''',
+        ...     inherited,
+        ... )
+        >>> sorted(out.items())  # doctest: +NORMALIZE_WHITESPACE
+        [('border-bottom-color', 'blue'),
+         ('border-left-color', 'red'),
+         ('border-right-color', 'red'),
+         ('border-top-color', 'blue'),
+         ('font-family', 'serif'),
+         ('font-size', '24pt'),
+         ('font-weight', 'bold')]
+        """
+        if isinstance(declarations, str):
+            declarations = self.parse(declarations)
+        props = dict(self.atomize(declarations))
+        if inherited is None:
+            inherited = {}
+
+        props = self._update_initial(props, inherited)
+        props = self._update_font_size(props, inherited)
+        return self._update_other_units(props)
+
+    def _update_initial(
+        self,
+        props: dict[str, str],
+        inherited: dict[str, str],
+    ) -> dict[str, str]:
+        # 1. resolve inherited, initial
+        for prop, val in inherited.items():
+            if prop not in props:
+                props[prop] = val
+
+        new_props = props.copy()
+        for prop, val in props.items():
+            if val == "inherit":
+                val = inherited.get(prop, "initial")
+
+            if val in ("initial", None):
+                # we do not define a complete initial stylesheet
+                del new_props[prop]
+            else:
+                new_props[prop] = val
+        return new_props
+
+    def _update_font_size(
+        self,
+        props: dict[str, str],
+        inherited: dict[str, str],
+    ) -> dict[str, str]:
+        # 2. resolve relative font size
+        if props.get("font-size"):
+            props["font-size"] = self.size_to_pt(
+                props["font-size"],
+                self._get_font_size(inherited),
+                conversions=self.FONT_SIZE_RATIOS,
+            )
+        return props
+
+    def _get_font_size(self, props: dict[str, str]) -> float | None:
+        if props.get("font-size"):
+            font_size_string = props["font-size"]
+            return self._get_float_font_size_from_pt(font_size_string)
+        return None
+
+    def _get_float_font_size_from_pt(self, font_size_string: str) -> float:
+        assert font_size_string.endswith("pt")
+        return float(font_size_string.rstrip("pt"))
+
+    def _update_other_units(self, props: dict[str, str]) -> dict[str, str]:
+        font_size = self._get_font_size(props)
+        # 3. TODO: resolve other font-relative units
+        for side in self.SIDES:
+            prop = f"border-{side}-width"
+            if prop in props:
+                props[prop] = self.size_to_pt(
+                    props[prop],
+                    em_pt=font_size,
+                    conversions=self.BORDER_WIDTH_RATIOS,
+                )
+
+            for prop in [f"margin-{side}", f"padding-{side}"]:
+                if prop in props:
+                    # TODO: support %
+                    props[prop] = self.size_to_pt(
+                        props[prop],
+                        em_pt=font_size,
+                        conversions=self.MARGIN_RATIOS,
+                    )
+        return props
+
+    def size_to_pt(
+        self, in_val: str, em_pt: float | None = None, conversions: dict = UNIT_RATIOS
+    ) -> str:
+        def _error() -> str:
+            warnings.warn(
+                f"Unhandled size: {in_val!r}",
+                CSSWarning,
+                stacklevel=find_stack_level(),
+            )
+            return self.size_to_pt("1!!default", conversions=conversions)
+
+        match = re.match(r"^(\S*?)([a-zA-Z%!].*)", in_val)
+        if match is None:
+            return _error()
+
+        val, unit = match.groups()
+        if val == "":
+            # hack for 'large' etc.
+            val = 1
+        else:
+            try:
+                val = float(val)
+            except ValueError:
+                return _error()
+
+        while unit != "pt":
+            if unit == "em":
+                if em_pt is None:
+                    unit = "rem"
+                else:
+                    val *= em_pt
+                    unit = "pt"
+                continue
+
+            try:
+                unit, mul = conversions[unit]
+            except KeyError:
+                return _error()
+            val *= mul
+
+        val = round(val, 5)
+        if int(val) == val:
+            size_fmt = f"{int(val):d}pt"
+        else:
+            size_fmt = f"{val:f}pt"
+        return size_fmt
+
+    def atomize(self, declarations: Iterable) -> Generator[tuple[str, str]]:
+        for prop, value in declarations:
+            prop = prop.lower()
+            value = value.lower()
+            if prop in self.CSS_EXPANSIONS:
+                expand = self.CSS_EXPANSIONS[prop]
+                yield from expand(self, prop, value)
+            else:
+                yield prop, value
+
+    def parse(self, declarations_str: str) -> Iterator[tuple[str, str]]:
+        """
+        Generates (prop, value) pairs from declarations.
+
+        In a future version may generate parsed tokens from tinycss/tinycss2
+
+        Parameters
+        ----------
+        declarations_str : str
+        """
+        for decl in declarations_str.split(";"):
+            if not decl.strip():
+                continue
+            prop, sep, val = decl.partition(":")
+            prop = prop.strip().lower()
+            # TODO: don't lowercase case sensitive parts of values (strings)
+            val = val.strip().lower()
+            if sep:
+                yield prop, val
+            else:
+                warnings.warn(
+                    f"Ill-formatted attribute: expected a colon in {decl!r}",
+                    CSSWarning,
+                    stacklevel=find_stack_level(),
+                )
diff --git a/pandas/pandas/io/formats/csvs.py b/pandas/pandas/io/formats/csvs.py
new file mode 100644
index 0000000000000..75bcb51ef4be2
--- /dev/null
+++ b/pandas/pandas/io/formats/csvs.py
@@ -0,0 +1,334 @@
+"""
+Module for formatting output data into CSV files.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Hashable,
+    Iterable,
+    Iterator,
+    Sequence,
+)
+import csv as csvlib
+import os
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+
+import numpy as np
+
+from pandas._libs import writers as libwriters
+from pandas._typing import SequenceNotStr
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.generic import (
+    ABCDatetimeIndex,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCPeriodIndex,
+)
+from pandas.core.dtypes.missing import notna
+
+from pandas.core.indexes.api import Index
+
+from pandas.io.common import get_handle
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        FloatFormatType,
+        IndexLabel,
+        StorageOptions,
+        WriteBuffer,
+        npt,
+    )
+
+    from pandas.io.formats.format import DataFrameFormatter
+
+
+_DEFAULT_CHUNKSIZE_CELLS = 100_000
+
+
+class CSVFormatter:
+    cols: npt.NDArray[np.object_]
+
+    def __init__(
+        self,
+        formatter: DataFrameFormatter,
+        path_or_buf: FilePath | WriteBuffer[str] | WriteBuffer[bytes] = "",
+        sep: str = ",",
+        cols: Sequence[Hashable] | None = None,
+        index_label: IndexLabel | None = None,
+        mode: str = "w",
+        encoding: str | None = None,
+        errors: str = "strict",
+        compression: CompressionOptions = "infer",
+        quoting: int | None = None,
+        lineterminator: str | None = "\n",
+        chunksize: int | None = None,
+        quotechar: str | None = '"',
+        date_format: str | None = None,
+        doublequote: bool = True,
+        escapechar: str | None = None,
+        storage_options: StorageOptions | None = None,
+    ) -> None:
+        self.fmt = formatter
+
+        self.obj = self.fmt.frame
+
+        self.filepath_or_buffer = path_or_buf
+        self.encoding = encoding
+        self.compression: CompressionOptions = compression
+        self.mode = mode
+        self.storage_options = storage_options
+
+        self.sep = sep
+        self.index_label = self._initialize_index_label(index_label)
+        self.errors = errors
+        self.quoting = quoting or csvlib.QUOTE_MINIMAL
+        self.quotechar = self._initialize_quotechar(quotechar)
+        self.doublequote = doublequote
+        self.escapechar = escapechar
+        self.lineterminator = lineterminator or os.linesep
+        self.date_format = date_format
+        self.cols = self._initialize_columns(cols)
+        self.chunksize = self._initialize_chunksize(chunksize)
+
+    @property
+    def na_rep(self) -> str:
+        return self.fmt.na_rep
+
+    @property
+    def float_format(self) -> FloatFormatType | None:
+        return self.fmt.float_format
+
+    @property
+    def decimal(self) -> str:
+        return self.fmt.decimal
+
+    @property
+    def header(self) -> bool | SequenceNotStr[str]:
+        return self.fmt.header
+
+    @property
+    def index(self) -> bool:
+        return self.fmt.index
+
+    def _initialize_index_label(self, index_label: IndexLabel | None) -> IndexLabel:
+        if index_label is not False:
+            if index_label is None:
+                return self._get_index_label_from_obj()
+            elif not isinstance(index_label, (list, tuple, np.ndarray, ABCIndex)):
+                # given a string for a DF with Index
+                return [index_label]
+        return index_label
+
+    def _get_index_label_from_obj(self) -> Sequence[Hashable]:
+        if isinstance(self.obj.index, ABCMultiIndex):
+            return self._get_index_label_multiindex()
+        else:
+            return self._get_index_label_flat()
+
+    def _get_index_label_multiindex(self) -> Sequence[Hashable]:
+        return [name or "" for name in self.obj.index.names]
+
+    def _get_index_label_flat(self) -> Sequence[Hashable]:
+        index_label = self.obj.index.name
+        return [""] if index_label is None else [index_label]
+
+    def _initialize_quotechar(self, quotechar: str | None) -> str | None:
+        if self.quoting != csvlib.QUOTE_NONE:
+            # prevents crash in _csv
+            return quotechar
+        return None
+
+    @property
+    def has_mi_columns(self) -> bool:
+        return bool(isinstance(self.obj.columns, ABCMultiIndex))
+
+    def _initialize_columns(
+        self, cols: Iterable[Hashable] | None
+    ) -> npt.NDArray[np.object_]:
+        # validate mi options
+        if self.has_mi_columns:
+            if cols is not None:
+                msg = "cannot specify cols with a MultiIndex on the columns"
+                raise TypeError(msg)
+
+        if cols is not None:
+            if isinstance(cols, ABCIndex):
+                cols = cols._get_values_for_csv(**self._number_format)
+            else:
+                cols = list(cols)
+            self.obj = self.obj.loc[:, cols]
+
+        # update columns to include possible multiplicity of dupes
+        # and make sure cols is just a list of labels
+        new_cols = self.obj.columns
+        return new_cols._get_values_for_csv(**self._number_format)
+
+    def _initialize_chunksize(self, chunksize: int | None) -> int:
+        if chunksize is None:
+            return (_DEFAULT_CHUNKSIZE_CELLS // (len(self.cols) or 1)) or 1
+        return int(chunksize)
+
+    @property
+    def _number_format(self) -> dict[str, Any]:
+        """Dictionary used for storing number formatting settings."""
+        return {
+            "na_rep": self.na_rep,
+            "float_format": self.float_format,
+            "date_format": self.date_format,
+            "quoting": self.quoting,
+            "decimal": self.decimal,
+        }
+
+    @cache_readonly
+    def data_index(self) -> Index:
+        data_index = self.obj.index
+        if (
+            isinstance(data_index, (ABCDatetimeIndex, ABCPeriodIndex))
+            and self.date_format is not None
+        ):
+            data_index = Index(
+                [x.strftime(self.date_format) if notna(x) else "" for x in data_index]
+            )
+        elif isinstance(data_index, ABCMultiIndex):
+            data_index = data_index.remove_unused_levels()
+        return data_index
+
+    @property
+    def nlevels(self) -> int:
+        if self.index:
+            return getattr(self.data_index, "nlevels", 1)
+        else:
+            return 0
+
+    @property
+    def _has_aliases(self) -> bool:
+        return isinstance(self.header, (tuple, list, np.ndarray, ABCIndex))
+
+    @property
+    def _need_to_save_header(self) -> bool:
+        return bool(self._has_aliases or self.header)
+
+    @property
+    def write_cols(self) -> SequenceNotStr[Hashable]:
+        if self._has_aliases:
+            assert not isinstance(self.header, bool)
+            if len(self.header) != len(self.cols):
+                raise ValueError(
+                    f"Writing {len(self.cols)} cols but got {len(self.header)} aliases"
+                )
+            return self.header
+        else:
+            # self.cols is an ndarray derived from Index._get_values_for_csv,
+            #  so its entries are strings, i.e. hashable
+            return cast(SequenceNotStr[Hashable], self.cols)
+
+    @property
+    def encoded_labels(self) -> list[Hashable]:
+        encoded_labels: list[Hashable] = []
+
+        if self.index and self.index_label:
+            assert isinstance(self.index_label, Sequence)
+            encoded_labels = list(self.index_label)
+
+        if not self.has_mi_columns or self._has_aliases:
+            encoded_labels += list(self.write_cols)
+
+        return encoded_labels
+
+    def save(self) -> None:
+        """
+        Create the writer & save.
+        """
+        # apply compression and byte/text conversion
+        with get_handle(
+            self.filepath_or_buffer,
+            self.mode,
+            encoding=self.encoding,
+            errors=self.errors,
+            compression=self.compression,
+            storage_options=self.storage_options,
+        ) as handles:
+            # Note: self.encoding is irrelevant here
+            self.writer = csvlib.writer(
+                handles.handle,
+                lineterminator=self.lineterminator,
+                delimiter=self.sep,
+                quoting=self.quoting,
+                doublequote=self.doublequote,
+                escapechar=self.escapechar,
+                quotechar=self.quotechar,
+            )
+
+            self._save()
+
+    def _save(self) -> None:
+        if self._need_to_save_header:
+            self._save_header()
+        self._save_body()
+
+    def _save_header(self) -> None:
+        if not self.has_mi_columns or self._has_aliases:
+            self.writer.writerow(self.encoded_labels)
+        else:
+            for row in self._generate_multiindex_header_rows():
+                self.writer.writerow(row)
+
+    def _generate_multiindex_header_rows(self) -> Iterator[list[Hashable]]:
+        columns = self.obj.columns
+        for i in range(columns.nlevels):
+            # we need at least 1 index column to write our col names
+            col_line = []
+            if self.index:
+                # name is the first column
+                col_line.append(columns.names[i])
+
+                if isinstance(self.index_label, list) and len(self.index_label) > 1:
+                    col_line.extend([""] * (len(self.index_label) - 1))
+
+            col_line.extend(columns._get_level_values(i))
+            yield col_line
+
+        # Write out the index line if it's not empty.
+        # Otherwise, we will print out an extraneous
+        # blank line between the mi and the data rows.
+        if self.encoded_labels and set(self.encoded_labels) != {""}:
+            yield self.encoded_labels + [""] * len(columns)
+
+    def _save_body(self) -> None:
+        nrows = len(self.data_index)
+        chunks = (nrows // self.chunksize) + 1
+        for i in range(chunks):
+            start_i = i * self.chunksize
+            end_i = min(start_i + self.chunksize, nrows)
+            if start_i >= end_i:
+                break
+            self._save_chunk(start_i, end_i)
+
+    def _save_chunk(self, start_i: int, end_i: int) -> None:
+        # create the data for a chunk
+        slicer = slice(start_i, end_i)
+        df = self.obj.iloc[slicer]
+
+        res = df._get_values_for_csv(**self._number_format)
+        data = list(res._iter_column_arrays())
+
+        ix = (
+            self.data_index[slicer]._get_values_for_csv(**self._number_format)
+            if self.nlevels != 0
+            else np.empty(end_i - start_i)
+        )
+        libwriters.write_csv_rows(
+            data,
+            ix,
+            self.nlevels,
+            self.cols,
+            self.writer,
+        )
diff --git a/pandas/pandas/io/formats/excel.py b/pandas/pandas/io/formats/excel.py
new file mode 100644
index 0000000000000..5fde6577e9f95
--- /dev/null
+++ b/pandas/pandas/io/formats/excel.py
@@ -0,0 +1,947 @@
+"""
+Utilities for conversion to writer-agnostic Excel representation.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Hashable,
+    Iterable,
+    Mapping,
+    Sequence,
+)
+import functools
+import itertools
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs.lib import is_list_like
+from pandas.util._decorators import doc
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes import missing
+from pandas.core.dtypes.common import (
+    is_float,
+    is_scalar,
+)
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    PeriodIndex,
+)
+import pandas.core.common as com
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.formats._color_data import CSS4_COLORS
+from pandas.io.formats.css import (
+    CSSResolver,
+    CSSWarning,
+)
+from pandas.io.formats.format import get_level_lengths
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ExcelWriterMergeCells,
+        FilePath,
+        IndexLabel,
+        StorageOptions,
+        WriteExcelBuffer,
+    )
+
+    from pandas import ExcelWriter
+
+
+class ExcelCell:
+    __fields__ = ("row", "col", "val", "style", "mergestart", "mergeend")
+    __slots__ = __fields__
+
+    def __init__(
+        self,
+        row: int,
+        col: int,
+        val,
+        style=None,
+        mergestart: int | None = None,
+        mergeend: int | None = None,
+    ) -> None:
+        self.row = row
+        self.col = col
+        self.val = val
+        self.style = style
+        self.mergestart = mergestart
+        self.mergeend = mergeend
+
+
+class CssExcelCell(ExcelCell):
+    def __init__(
+        self,
+        row: int,
+        col: int,
+        val,
+        style: dict | None,
+        css_styles: dict[tuple[int, int], list[tuple[str, Any]]] | None,
+        css_row: int,
+        css_col: int,
+        css_converter: Callable | None,
+        **kwargs,
+    ) -> None:
+        if css_styles and css_converter:
+            # Use dict to get only one (case-insensitive) declaration per property
+            declaration_dict = {
+                prop.lower(): val for prop, val in css_styles[css_row, css_col]
+            }
+            # Convert to frozenset for order-invariant caching
+            unique_declarations = frozenset(declaration_dict.items())
+            style = css_converter(unique_declarations)
+
+        super().__init__(row=row, col=col, val=val, style=style, **kwargs)
+
+
+class CSSToExcelConverter:
+    """
+    A callable for converting CSS declarations to ExcelWriter styles
+
+    Supports parts of CSS 2.2, with minimal CSS 3.0 support (e.g. text-shadow),
+    focusing on font styling, backgrounds, borders and alignment.
+
+    Operates by first computing CSS styles in a fairly generic
+    way (see :meth:`compute_css`) then determining Excel style
+    properties from CSS properties (see :meth:`build_xlstyle`).
+
+    Parameters
+    ----------
+    inherited : str, optional
+        CSS declarations understood to be the containing scope for the
+        CSS processed by :meth:`__call__`.
+    """
+
+    NAMED_COLORS = CSS4_COLORS
+
+    VERTICAL_MAP = {
+        "top": "top",
+        "text-top": "top",
+        "middle": "center",
+        "baseline": "bottom",
+        "bottom": "bottom",
+        "text-bottom": "bottom",
+        # OpenXML also has 'justify', 'distributed'
+    }
+
+    BOLD_MAP = {
+        "bold": True,
+        "bolder": True,
+        "600": True,
+        "700": True,
+        "800": True,
+        "900": True,
+        "normal": False,
+        "lighter": False,
+        "100": False,
+        "200": False,
+        "300": False,
+        "400": False,
+        "500": False,
+    }
+
+    ITALIC_MAP = {
+        "normal": False,
+        "italic": True,
+        "oblique": True,
+    }
+
+    FAMILY_MAP = {
+        "serif": 1,  # roman
+        "sans-serif": 2,  # swiss
+        "cursive": 4,  # script
+        "fantasy": 5,  # decorative
+    }
+
+    BORDER_STYLE_MAP = {
+        style.lower(): style
+        for style in [
+            "dashed",
+            "mediumDashDot",
+            "dashDotDot",
+            "hair",
+            "dotted",
+            "mediumDashDotDot",
+            "double",
+            "dashDot",
+            "slantDashDot",
+            "mediumDashed",
+        ]
+    }
+
+    # NB: Most of the methods here could be classmethods, as only __init__
+    #     and __call__ make use of instance attributes.  We leave them as
+    #     instancemethods so that users can easily experiment with extensions
+    #     without monkey-patching.
+    inherited: dict[str, str] | None
+
+    def __init__(self, inherited: str | None = None) -> None:
+        if inherited is not None:
+            self.inherited = self.compute_css(inherited)
+        else:
+            self.inherited = None
+        # We should avoid cache on the __call__ method.
+        # Otherwise once the method __call__ has been called
+        # garbage collection no longer deletes the instance.
+        self._call_cached = functools.cache(self._call_uncached)
+
+    compute_css = CSSResolver()
+
+    def __call__(
+        self, declarations: str | frozenset[tuple[str, str]]
+    ) -> dict[str, dict[str, str]]:
+        """
+        Convert CSS declarations to ExcelWriter style.
+
+        Parameters
+        ----------
+        declarations : str | frozenset[tuple[str, str]]
+            CSS string or set of CSS declaration tuples.
+            e.g. "font-weight: bold; background: blue" or
+            {("font-weight", "bold"), ("background", "blue")}
+
+        Returns
+        -------
+        xlstyle : dict
+            A style as interpreted by ExcelWriter when found in
+            ExcelCell.style.
+        """
+        return self._call_cached(declarations)
+
+    def _call_uncached(
+        self, declarations: str | frozenset[tuple[str, str]]
+    ) -> dict[str, dict[str, str]]:
+        properties = self.compute_css(declarations, self.inherited)
+        return self.build_xlstyle(properties)
+
+    def build_xlstyle(self, props: Mapping[str, str]) -> dict[str, dict[str, str]]:
+        out = {
+            "alignment": self.build_alignment(props),
+            "border": self.build_border(props),
+            "fill": self.build_fill(props),
+            "font": self.build_font(props),
+            "number_format": self.build_number_format(props),
+        }
+
+        # TODO: handle cell width and height: needs support in pandas.io.excel
+
+        def remove_none(d: dict[str, str | None]) -> None:
+            """Remove key where value is None, through nested dicts"""
+            for k, v in list(d.items()):
+                if v is None:
+                    del d[k]
+                elif isinstance(v, dict):
+                    remove_none(v)
+                    if not v:
+                        del d[k]
+
+        remove_none(out)
+        return out
+
+    def build_alignment(self, props: Mapping[str, str]) -> dict[str, bool | str | None]:
+        # TODO: text-indent, padding-left -> alignment.indent
+        return {
+            "horizontal": props.get("text-align"),
+            "vertical": self._get_vertical_alignment(props),
+            "wrap_text": self._get_is_wrap_text(props),
+        }
+
+    def _get_vertical_alignment(self, props: Mapping[str, str]) -> str | None:
+        vertical_align = props.get("vertical-align")
+        if vertical_align:
+            return self.VERTICAL_MAP.get(vertical_align)
+        return None
+
+    def _get_is_wrap_text(self, props: Mapping[str, str]) -> bool | None:
+        if props.get("white-space") is None:
+            return None
+        return bool(props["white-space"] not in ("nowrap", "pre", "pre-line"))
+
+    def build_border(
+        self, props: Mapping[str, str]
+    ) -> dict[str, dict[str, str | None]]:
+        return {
+            side: {
+                "style": self._border_style(
+                    props.get(f"border-{side}-style"),
+                    props.get(f"border-{side}-width"),
+                    self.color_to_excel(props.get(f"border-{side}-color")),
+                ),
+                "color": self.color_to_excel(props.get(f"border-{side}-color")),
+            }
+            for side in ["top", "right", "bottom", "left"]
+        }
+
+    def _border_style(
+        self, style: str | None, width: str | None, color: str | None
+    ) -> str | None:
+        # convert styles and widths to openxml, one of:
+        #       'dashDot'
+        #       'dashDotDot'
+        #       'dashed'
+        #       'dotted'
+        #       'double'
+        #       'hair'
+        #       'medium'
+        #       'mediumDashDot'
+        #       'mediumDashDotDot'
+        #       'mediumDashed'
+        #       'slantDashDot'
+        #       'thick'
+        #       'thin'
+        if width is None and style is None and color is None:
+            # Return None will remove "border" from style dictionary
+            return None
+
+        if width is None and style is None:
+            # Return "none" will keep "border" in style dictionary
+            return "none"
+
+        if style in ("none", "hidden"):
+            return "none"
+
+        width_name = self._get_width_name(width)
+        if width_name is None:
+            return "none"
+
+        if style in (None, "groove", "ridge", "inset", "outset", "solid"):
+            # not handled
+            return width_name
+
+        if style == "double":
+            return "double"
+        if style == "dotted":
+            if width_name in ("hair", "thin"):
+                return "dotted"
+            return "mediumDashDotDot"
+        if style == "dashed":
+            if width_name in ("hair", "thin"):
+                return "dashed"
+            return "mediumDashed"
+        elif style in self.BORDER_STYLE_MAP:
+            # Excel-specific styles
+            return self.BORDER_STYLE_MAP[style]
+        else:
+            warnings.warn(
+                f"Unhandled border style format: {style!r}",
+                CSSWarning,
+                stacklevel=find_stack_level(),
+            )
+            return "none"
+
+    def _get_width_name(self, width_input: str | None) -> str | None:
+        width = self._width_to_float(width_input)
+        if width < 1e-5:
+            return None
+        elif width < 1.3:
+            return "thin"
+        elif width < 2.8:
+            return "medium"
+        return "thick"
+
+    def _width_to_float(self, width: str | None) -> float:
+        if width is None:
+            width = "2pt"
+        return self._pt_to_float(width)
+
+    def _pt_to_float(self, pt_string: str) -> float:
+        assert pt_string.endswith("pt")
+        return float(pt_string.rstrip("pt"))
+
+    def build_fill(self, props: Mapping[str, str]):
+        # TODO: perhaps allow for special properties
+        #       -excel-pattern-bgcolor and -excel-pattern-type
+        fill_color = props.get("background-color")
+        if fill_color not in (None, "transparent", "none"):
+            return {"fgColor": self.color_to_excel(fill_color), "patternType": "solid"}
+
+    def build_number_format(self, props: Mapping[str, str]) -> dict[str, str | None]:
+        fc = props.get("number-format")
+        fc = fc.replace("§", ";") if isinstance(fc, str) else fc
+        return {"format_code": fc}
+
+    def build_font(
+        self, props: Mapping[str, str]
+    ) -> dict[str, bool | float | str | None]:
+        font_names = self._get_font_names(props)
+        decoration = self._get_decoration(props)
+        return {
+            "name": font_names[0] if font_names else None,
+            "family": self._select_font_family(font_names),
+            "size": self._get_font_size(props),
+            "bold": self._get_is_bold(props),
+            "italic": self._get_is_italic(props),
+            "underline": ("single" if "underline" in decoration else None),
+            "strike": ("line-through" in decoration) or None,
+            "color": self.color_to_excel(props.get("color")),
+            # shadow if nonzero digit before shadow color
+            "shadow": self._get_shadow(props),
+        }
+
+    def _get_is_bold(self, props: Mapping[str, str]) -> bool | None:
+        weight = props.get("font-weight")
+        if weight:
+            return self.BOLD_MAP.get(weight)
+        return None
+
+    def _get_is_italic(self, props: Mapping[str, str]) -> bool | None:
+        font_style = props.get("font-style")
+        if font_style:
+            return self.ITALIC_MAP.get(font_style)
+        return None
+
+    def _get_decoration(self, props: Mapping[str, str]) -> Sequence[str]:
+        decoration = props.get("text-decoration")
+        if decoration is not None:
+            return decoration.split()
+        else:
+            return ()
+
+    def _get_underline(self, decoration: Sequence[str]) -> str | None:
+        if "underline" in decoration:
+            return "single"
+        return None
+
+    def _get_shadow(self, props: Mapping[str, str]) -> bool | None:
+        if "text-shadow" in props:
+            return bool(re.search("^[^#(]*[1-9]", props["text-shadow"]))
+        return None
+
+    def _get_font_names(self, props: Mapping[str, str]) -> Sequence[str]:
+        font_names_tmp = re.findall(
+            r"""(?x)
+            (
+            "(?:[^"]|\\")+"
+            |
+            '(?:[^']|\\')+'
+            |
+            [^'",]+
+            )(?=,|\s*$)
+        """,
+            props.get("font-family", ""),
+        )
+
+        font_names = []
+        for name in font_names_tmp:
+            if name[:1] == '"':
+                name = name[1:-1].replace('\\"', '"')
+            elif name[:1] == "'":
+                name = name[1:-1].replace("\\'", "'")
+            else:
+                name = name.strip()
+            if name:
+                font_names.append(name)
+        return font_names
+
+    def _get_font_size(self, props: Mapping[str, str]) -> float | None:
+        size = props.get("font-size")
+        if size is None:
+            return size
+        return self._pt_to_float(size)
+
+    def _select_font_family(self, font_names: Sequence[str]) -> int | None:
+        family = None
+        for name in font_names:
+            family = self.FAMILY_MAP.get(name)
+            if family:
+                break
+
+        return family
+
+    def color_to_excel(self, val: str | None) -> str | None:
+        if val is None:
+            return None
+
+        if self._is_hex_color(val):
+            return self._convert_hex_to_excel(val)
+
+        try:
+            return self.NAMED_COLORS[val]
+        except KeyError:
+            warnings.warn(
+                f"Unhandled color format: {val!r}",
+                CSSWarning,
+                stacklevel=find_stack_level(),
+            )
+        return None
+
+    def _is_hex_color(self, color_string: str) -> bool:
+        return bool(color_string.startswith("#"))
+
+    def _convert_hex_to_excel(self, color_string: str) -> str:
+        code = color_string.lstrip("#")
+        if self._is_shorthand_color(color_string):
+            return (code[0] * 2 + code[1] * 2 + code[2] * 2).upper()
+        else:
+            return code.upper()
+
+    def _is_shorthand_color(self, color_string: str) -> bool:
+        """Check if color code is shorthand.
+
+        #FFF is a shorthand as opposed to full #FFFFFF.
+        """
+        code = color_string.lstrip("#")
+        if len(code) == 3:
+            return True
+        elif len(code) == 6:
+            return False
+        else:
+            raise ValueError(f"Unexpected color {color_string}")
+
+
+class ExcelFormatter:
+    """
+    Class for formatting a DataFrame to a list of ExcelCells,
+
+    Parameters
+    ----------
+    df : DataFrame or Styler
+    na_rep: na representation
+    float_format : str, default None
+        Format string for floating point numbers
+    cols : sequence, optional
+        Columns to write
+    header : bool or sequence of str, default True
+        Write out column names. If a list of string is given it is
+        assumed to be aliases for the column names
+    index : bool, default True
+        output row names (index)
+    index_label : str or sequence, default None
+        Column label for index column(s) if desired. If None is given, and
+        `header` and `index` are True, then the index names are used. A
+        sequence should be given if the DataFrame uses MultiIndex.
+    merge_cells : bool or 'columns', default False
+        Format MultiIndex column headers and Hierarchical Rows as merged cells
+        if True. Merge MultiIndex column headers only if 'columns'.
+        .. versionchanged:: 3.0.0
+            Added the 'columns' option.
+    inf_rep : str, default `'inf'`
+        representation for np.inf values (which aren't representable in Excel)
+        A `'-'` sign will be added in front of -inf.
+    style_converter : callable, optional
+        This translates Styler styles (CSS) into ExcelWriter styles.
+        Defaults to ``CSSToExcelConverter()``.
+        It should have signature css_declarations string -> excel style.
+        This is only called for body cells.
+    """
+
+    max_rows = 2**20
+    max_cols = 2**14
+
+    def __init__(
+        self,
+        df,
+        na_rep: str = "",
+        float_format: str | None = None,
+        cols: Sequence[Hashable] | None = None,
+        header: Sequence[Hashable] | bool = True,
+        index: bool = True,
+        index_label: IndexLabel | None = None,
+        merge_cells: ExcelWriterMergeCells = False,
+        inf_rep: str = "inf",
+        style_converter: Callable | None = None,
+    ) -> None:
+        self.rowcounter = 0
+        self.na_rep = na_rep
+        if not isinstance(df, DataFrame):
+            self.styler = df
+            self.styler._compute()  # calculate applied styles
+            df = df.data
+            if style_converter is None:
+                style_converter = CSSToExcelConverter()
+            self.style_converter: Callable | None = style_converter
+        else:
+            self.styler = None
+            self.style_converter = None
+        self.df = df
+        if cols is not None:
+            # all missing, raise
+            if not len(Index(cols).intersection(df.columns)):
+                raise KeyError("passes columns are not ALL present dataframe")
+
+            if len(Index(cols).intersection(df.columns)) != len(set(cols)):
+                # Deprecated in GH#17295, enforced in 1.0.0
+                raise KeyError("Not all names specified in 'columns' are found")
+
+            self.df = df.reindex(columns=cols)
+
+        self.columns = self.df.columns
+        self.float_format = float_format
+        self.index = index
+        self.index_label = index_label
+        self.header = header
+
+        if not isinstance(merge_cells, bool) and merge_cells != "columns":
+            raise ValueError(f"Unexpected value for {merge_cells=}.")
+        self.merge_cells = merge_cells
+        self.inf_rep = inf_rep
+
+    def _format_value(self, val):
+        if is_scalar(val) and missing.isna(val):
+            val = self.na_rep
+        elif is_float(val):
+            if missing.isposinf_scalar(val):
+                val = self.inf_rep
+            elif missing.isneginf_scalar(val):
+                val = f"-{self.inf_rep}"
+            elif self.float_format is not None:
+                val = float(self.float_format % val)
+        if getattr(val, "tzinfo", None) is not None:
+            raise ValueError(
+                "Excel does not support datetimes with "
+                "timezones. Please ensure that datetimes "
+                "are timezone unaware before writing to Excel."
+            )
+        return val
+
+    def _format_header_mi(self) -> Iterable[ExcelCell]:
+        if self.columns.nlevels > 1:
+            if not self.index:
+                raise NotImplementedError(
+                    "Writing to Excel with MultiIndex columns and no "
+                    "index ('index'=False) is not yet implemented."
+                )
+
+        if not (self._has_aliases or self.header):
+            return
+
+        columns = self.columns
+        merge_columns = self.merge_cells in {True, "columns"}
+        level_strs = columns._format_multi(sparsify=merge_columns, include_names=False)
+        level_lengths = get_level_lengths(level_strs)
+        coloffset = 0
+        lnum = 0
+
+        if self.index and isinstance(self.df.index, MultiIndex):
+            coloffset = self.df.index.nlevels - 1
+
+        for lnum, name in enumerate(columns.names):
+            yield ExcelCell(
+                row=lnum,
+                col=coloffset,
+                val=name,
+                style=None,
+            )
+
+        for lnum, (spans, levels, level_codes) in enumerate(
+            zip(level_lengths, columns.levels, columns.codes)
+        ):
+            values = levels.take(level_codes)
+            for i, span_val in spans.items():
+                mergestart, mergeend = None, None
+                if merge_columns and span_val > 1:
+                    mergestart, mergeend = lnum, coloffset + i + span_val
+                yield CssExcelCell(
+                    row=lnum,
+                    col=coloffset + i + 1,
+                    val=values[i],
+                    style=None,
+                    css_styles=getattr(self.styler, "ctx_columns", None),
+                    css_row=lnum,
+                    css_col=i,
+                    css_converter=self.style_converter,
+                    mergestart=mergestart,
+                    mergeend=mergeend,
+                )
+        self.rowcounter = lnum
+
+    def _format_header_regular(self) -> Iterable[ExcelCell]:
+        if self._has_aliases or self.header:
+            coloffset = 0
+
+            if self.index:
+                coloffset = 1
+                if isinstance(self.df.index, MultiIndex):
+                    coloffset = len(self.df.index.names)
+
+            colnames = self.columns
+            if self._has_aliases:
+                self.header = cast(Sequence, self.header)
+                if len(self.header) != len(self.columns):
+                    raise ValueError(
+                        f"Writing {len(self.columns)} cols "
+                        f"but got {len(self.header)} aliases"
+                    )
+                colnames = self.header
+
+            for colindex, colname in enumerate(colnames):
+                yield CssExcelCell(
+                    row=self.rowcounter,
+                    col=colindex + coloffset,
+                    val=colname,
+                    style=None,
+                    css_styles=getattr(self.styler, "ctx_columns", None),
+                    css_row=0,
+                    css_col=colindex,
+                    css_converter=self.style_converter,
+                )
+
+    def _format_header(self) -> Iterable[ExcelCell]:
+        gen: Iterable[ExcelCell]
+
+        if isinstance(self.columns, MultiIndex):
+            gen = self._format_header_mi()
+        else:
+            gen = self._format_header_regular()
+
+        gen2: Iterable[ExcelCell] = ()
+
+        if self.df.index.names:
+            row = [x if x is not None else "" for x in self.df.index.names] + [
+                ""
+            ] * len(self.columns)
+            if functools.reduce(lambda x, y: x and y, (x != "" for x in row)):
+                gen2 = (
+                    ExcelCell(self.rowcounter, colindex, val, None)
+                    for colindex, val in enumerate(row)
+                )
+                self.rowcounter += 1
+        return itertools.chain(gen, gen2)
+
+    def _format_body(self) -> Iterable[ExcelCell]:
+        if isinstance(self.df.index, MultiIndex):
+            return self._format_hierarchical_rows()
+        else:
+            return self._format_regular_rows()
+
+    def _format_regular_rows(self) -> Iterable[ExcelCell]:
+        if self._has_aliases or self.header:
+            self.rowcounter += 1
+
+        # output index and index_label?
+        if self.index:
+            # check aliases
+            # if list only take first as this is not a MultiIndex
+            if self.index_label and isinstance(
+                self.index_label, (list, tuple, np.ndarray, Index)
+            ):
+                index_label = self.index_label[0]
+            # if string good to go
+            elif self.index_label and isinstance(self.index_label, str):
+                index_label = self.index_label
+            else:
+                index_label = self.df.index.names[0]
+
+            if isinstance(self.columns, MultiIndex):
+                self.rowcounter += 1
+
+            if index_label and self.header is not False:
+                yield ExcelCell(self.rowcounter - 1, 0, index_label, None)
+
+            # write index_values
+            index_values = self.df.index
+            if isinstance(self.df.index, PeriodIndex):
+                index_values = self.df.index.to_timestamp()
+
+            for idx, idxval in enumerate(index_values):
+                yield CssExcelCell(
+                    row=self.rowcounter + idx,
+                    col=0,
+                    val=idxval,
+                    style=None,
+                    css_styles=getattr(self.styler, "ctx_index", None),
+                    css_row=idx,
+                    css_col=0,
+                    css_converter=self.style_converter,
+                )
+            coloffset = 1
+        else:
+            coloffset = 0
+
+        yield from self._generate_body(coloffset)
+
+    def _format_hierarchical_rows(self) -> Iterable[ExcelCell]:
+        if self._has_aliases or self.header:
+            self.rowcounter += 1
+
+        gcolidx = 0
+
+        if self.index:
+            index_labels = self.df.index.names
+            # check for aliases
+            if self.index_label and isinstance(
+                self.index_label, (list, tuple, np.ndarray, Index)
+            ):
+                index_labels = self.index_label
+
+            # MultiIndex columns require an extra row
+            # with index names (blank if None) for
+            # unambiguous round-trip, Issue #11328
+            if isinstance(self.columns, MultiIndex):
+                self.rowcounter += 1
+
+            # if index labels are not empty go ahead and dump
+            if com.any_not_none(*index_labels) and self.header is not False:
+                for cidx, name in enumerate(index_labels):
+                    yield ExcelCell(self.rowcounter - 1, cidx, name, None)
+
+            if self.merge_cells and self.merge_cells != "columns":
+                # Format hierarchical rows as merged cells.
+                level_strs = self.df.index._format_multi(
+                    sparsify=True, include_names=False
+                )
+                level_lengths = get_level_lengths(level_strs)
+
+                for spans, levels, level_codes in zip(
+                    level_lengths, self.df.index.levels, self.df.index.codes
+                ):
+                    values = levels.take(
+                        level_codes,
+                        allow_fill=levels._can_hold_na,
+                        fill_value=levels._na_value,
+                    )
+                    # GH#60099
+                    if isinstance(values[0], Period):
+                        values = values.to_timestamp()
+
+                    for i, span_val in spans.items():
+                        mergestart, mergeend = None, None
+                        if span_val > 1:
+                            mergestart = self.rowcounter + i + span_val - 1
+                            mergeend = gcolidx
+                        yield CssExcelCell(
+                            row=self.rowcounter + i,
+                            col=gcolidx,
+                            val=values[i],
+                            style=None,
+                            css_styles=getattr(self.styler, "ctx_index", None),
+                            css_row=i,
+                            css_col=gcolidx,
+                            css_converter=self.style_converter,
+                            mergestart=mergestart,
+                            mergeend=mergeend,
+                        )
+                    gcolidx += 1
+
+            else:
+                # Format hierarchical rows with non-merged values.
+                for indexcolvals in zip(*self.df.index):
+                    for idx, indexcolval in enumerate(indexcolvals):
+                        # GH#60099
+                        if isinstance(indexcolval, Period):
+                            indexcolval = indexcolval.to_timestamp()
+
+                        yield CssExcelCell(
+                            row=self.rowcounter + idx,
+                            col=gcolidx,
+                            val=indexcolval,
+                            style=None,
+                            css_styles=getattr(self.styler, "ctx_index", None),
+                            css_row=idx,
+                            css_col=gcolidx,
+                            css_converter=self.style_converter,
+                        )
+                    gcolidx += 1
+
+        yield from self._generate_body(gcolidx)
+
+    @property
+    def _has_aliases(self) -> bool:
+        """Whether the aliases for column names are present."""
+        return is_list_like(self.header)
+
+    def _generate_body(self, coloffset: int) -> Iterable[ExcelCell]:
+        # Write the body of the frame data series by series.
+        for colidx in range(len(self.columns)):
+            series = self.df.iloc[:, colidx]
+            for i, val in enumerate(series):
+                yield CssExcelCell(
+                    row=self.rowcounter + i,
+                    col=colidx + coloffset,
+                    val=val,
+                    style=None,
+                    css_styles=getattr(self.styler, "ctx", None),
+                    css_row=i,
+                    css_col=colidx,
+                    css_converter=self.style_converter,
+                )
+
+    def get_formatted_cells(self) -> Iterable[ExcelCell]:
+        for cell in itertools.chain(self._format_header(), self._format_body()):
+            cell.val = self._format_value(cell.val)
+            yield cell
+
+    @doc(storage_options=_shared_docs["storage_options"])
+    def write(
+        self,
+        writer: FilePath | WriteExcelBuffer | ExcelWriter,
+        sheet_name: str = "Sheet1",
+        startrow: int = 0,
+        startcol: int = 0,
+        freeze_panes: tuple[int, int] | None = None,
+        engine: str | None = None,
+        storage_options: StorageOptions | None = None,
+        engine_kwargs: dict | None = None,
+    ) -> None:
+        """
+        writer : path-like, file-like, or ExcelWriter object
+            File path or existing ExcelWriter
+        sheet_name : str, default 'Sheet1'
+            Name of sheet which will contain DataFrame
+        startrow :
+            upper left cell row to dump data frame
+        startcol :
+            upper left cell column to dump data frame
+        freeze_panes : tuple of integer (length 2), default None
+            Specifies the one-based bottommost row and rightmost column that
+            is to be frozen
+        engine : string, default None
+            write engine to use if writer is a path - you can also set this
+            via the options ``io.excel.xlsx.writer``,
+            or ``io.excel.xlsm.writer``.
+
+        {storage_options}
+
+        engine_kwargs: dict, optional
+            Arbitrary keyword arguments passed to excel engine.
+        """
+        from pandas.io.excel import ExcelWriter
+
+        num_rows, num_cols = self.df.shape
+        if num_rows > self.max_rows or num_cols > self.max_cols:
+            raise ValueError(
+                f"This sheet is too large! Your sheet size is: {num_rows}, {num_cols} "
+                f"Max sheet size is: {self.max_rows}, {self.max_cols}"
+            )
+
+        if engine_kwargs is None:
+            engine_kwargs = {}
+
+        formatted_cells = self.get_formatted_cells()
+        if isinstance(writer, ExcelWriter):
+            need_save = False
+        else:
+            writer = ExcelWriter(
+                writer,
+                engine=engine,
+                storage_options=storage_options,
+                engine_kwargs=engine_kwargs,
+            )
+            need_save = True
+
+        try:
+            writer._write_cells(
+                formatted_cells,
+                sheet_name,
+                startrow=startrow,
+                startcol=startcol,
+                freeze_panes=freeze_panes,
+            )
+        finally:
+            # make sure to close opened file handles
+            if need_save:
+                writer.close()
diff --git a/pandas/pandas/io/formats/format.py b/pandas/pandas/io/formats/format.py
new file mode 100644
index 0000000000000..b7fbc4e5e22b7
--- /dev/null
+++ b/pandas/pandas/io/formats/format.py
@@ -0,0 +1,2056 @@
+"""
+Internal module for formatting output data in csv, html, xml,
+and latex files. This module also applies to display formatting.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Generator,
+    Hashable,
+    Mapping,
+    Sequence,
+)
+from contextlib import contextmanager
+from csv import QUOTE_NONE
+from decimal import Decimal
+from functools import partial
+from io import StringIO
+import math
+import re
+from shutil import get_terminal_size
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Final,
+    cast,
+)
+
+import numpy as np
+
+from pandas._config.config import (
+    get_option,
+    set_option,
+)
+
+from pandas._libs import lib
+from pandas._libs.missing import NA
+from pandas._libs.tslibs import (
+    NaT,
+    Timedelta,
+    Timestamp,
+)
+from pandas._libs.tslibs.nattype import NaTType
+
+from pandas.core.dtypes.common import (
+    is_complex_dtype,
+    is_float,
+    is_integer,
+    is_list_like,
+    is_numeric_dtype,
+    is_scalar,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    notna,
+)
+
+from pandas.core.arrays import (
+    Categorical,
+    DatetimeArray,
+    ExtensionArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.base import PandasObject
+import pandas.core.common as com
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    ensure_index,
+)
+from pandas.core.indexes.datetimes import DatetimeIndex
+from pandas.core.indexes.timedeltas import TimedeltaIndex
+
+from pandas.io.common import (
+    check_parent_directory,
+    stringify_path,
+)
+from pandas.io.formats import printing
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        ArrayLike,
+        Axes,
+        ColspaceArgType,
+        ColspaceType,
+        CompressionOptions,
+        FilePath,
+        FloatFormatType,
+        FormattersType,
+        IndexLabel,
+        SequenceNotStr,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+common_docstring: Final = """
+        Parameters
+        ----------
+        buf : str, Path or StringIO-like, optional, default None
+            Buffer to write to. If None, the output is returned as a string.
+        columns : array-like, optional, default None
+            The subset of columns to write. Writes all columns by default.
+        col_space : %(col_space_type)s, optional
+            %(col_space)s.
+        header : %(header_type)s, optional
+            %(header)s.
+        index : bool, optional, default True
+            Whether to print index (row) labels.
+        na_rep : str, optional, default 'NaN'
+            String representation of ``NaN`` to use.
+        formatters : list, tuple or dict of one-param. functions, optional
+            Formatter functions to apply to columns' elements by position or
+            name.
+            The result of each function must be a unicode string.
+            List/tuple must be of length equal to the number of columns.
+        float_format : one-parameter function, optional, default None
+            Formatter function to apply to columns' elements if they are
+            floats. This function must return a unicode string and will be
+            applied only to the non-``NaN`` elements, with ``NaN`` being
+            handled by ``na_rep``.
+        sparsify : bool, optional, default True
+            Set to False for a DataFrame with a hierarchical index to print
+            every multiindex key at each row.
+        index_names : bool, optional, default True
+            Prints the names of the indexes.
+        justify : str, default None
+            How to justify the column labels. If None uses the option from
+            the print configuration (controlled by set_option), 'right' out
+            of the box. Valid values are
+
+            * left
+            * right
+            * center
+            * justify
+            * justify-all
+            * start
+            * end
+            * inherit
+            * match-parent
+            * initial
+            * unset.
+        max_rows : int, optional
+            Maximum number of rows to display in the console.
+        max_cols : int, optional
+            Maximum number of columns to display in the console.
+        show_dimensions : bool, default False
+            Display DataFrame dimensions (number of rows by number of columns).
+        decimal : str, default '.'
+            Character recognized as decimal separator, e.g. ',' in Europe.
+    """
+
+VALID_JUSTIFY_PARAMETERS = (
+    "left",
+    "right",
+    "center",
+    "justify",
+    "justify-all",
+    "start",
+    "end",
+    "inherit",
+    "match-parent",
+    "initial",
+    "unset",
+)
+
+return_docstring: Final = """
+        Returns
+        -------
+        str or None
+            If buf is None, returns the result as a string. Otherwise returns
+            None.
+    """
+
+
+class SeriesFormatter:
+    """
+    Implement the main logic of Series.to_string, which underlies
+    Series.__repr__.
+    """
+
+    def __init__(
+        self,
+        series: Series,
+        *,
+        length: bool | str = True,
+        header: bool = True,
+        index: bool = True,
+        na_rep: str = "NaN",
+        name: bool = False,
+        float_format: str | None = None,
+        dtype: bool = True,
+        max_rows: int | None = None,
+        min_rows: int | None = None,
+    ) -> None:
+        self.series = series
+        self.buf = StringIO()
+        self.name = name
+        self.na_rep = na_rep
+        self.header = header
+        self.length = length
+        self.index = index
+        self.max_rows = max_rows
+        self.min_rows = min_rows
+
+        if float_format is None:
+            float_format = get_option("display.float_format")
+        self.float_format = float_format
+        self.dtype = dtype
+        self.adj = printing.get_adjustment()
+
+        self._chk_truncate()
+
+    def _chk_truncate(self) -> None:
+        self.tr_row_num: int | None
+
+        min_rows = self.min_rows
+        max_rows = self.max_rows
+        # truncation determined by max_rows, actual truncated number of rows
+        # used below by min_rows
+        is_truncated_vertically = max_rows and (len(self.series) > max_rows)
+        series = self.series
+        if is_truncated_vertically:
+            max_rows = cast(int, max_rows)
+            if min_rows:
+                # if min_rows is set (not None or 0), set max_rows to minimum
+                # of both
+                max_rows = min(min_rows, max_rows)
+            if max_rows == 1:
+                row_num = max_rows
+                series = series.iloc[:max_rows]
+            else:
+                row_num = max_rows // 2
+                _len = len(series)
+                _slice = np.hstack(
+                    [np.arange(row_num), np.arange(_len - row_num, _len)]
+                )
+                series = series.iloc[_slice]
+            self.tr_row_num = row_num
+        else:
+            self.tr_row_num = None
+        self.tr_series = series
+        self.is_truncated_vertically = is_truncated_vertically
+
+    def _get_footer(self) -> str:
+        name = self.series.name
+        footer = ""
+
+        index = self.series.index
+        if (
+            isinstance(index, (DatetimeIndex, PeriodIndex, TimedeltaIndex))
+            and index.freq is not None
+        ):
+            footer += f"Freq: {index.freqstr}"
+
+        if self.name is not False and name is not None:
+            if footer:
+                footer += ", "
+
+            series_name = printing.pprint_thing(name, escape_chars=("\t", "\r", "\n"))
+            footer += f"Name: {series_name}"
+
+        if self.length is True or (
+            self.length == "truncate" and self.is_truncated_vertically
+        ):
+            if footer:
+                footer += ", "
+            footer += f"Length: {len(self.series)}"
+
+        if self.dtype is not False and self.dtype is not None:
+            dtype_name = getattr(self.tr_series.dtype, "name", None)
+            if dtype_name:
+                if footer:
+                    footer += ", "
+                footer += f"dtype: {printing.pprint_thing(dtype_name)}"
+
+        # level infos are added to the end and in a new line, like it is done
+        # for Categoricals
+        if isinstance(self.tr_series.dtype, CategoricalDtype):
+            level_info = self.tr_series._values._get_repr_footer()
+            if footer:
+                footer += "\n"
+            footer += level_info
+
+        return str(footer)
+
+    def _get_formatted_values(self) -> list[str]:
+        return format_array(
+            self.tr_series._values,
+            None,
+            float_format=self.float_format,
+            na_rep=self.na_rep,
+            leading_space=self.index,
+        )
+
+    def to_string(self) -> str:
+        series = self.tr_series
+        footer = self._get_footer()
+
+        if len(series) == 0:
+            return f"{type(self.series).__name__}([], {footer})"
+
+        index = series.index
+        have_header = _has_names(index)
+        if isinstance(index, MultiIndex):
+            fmt_index = index._format_multi(include_names=True, sparsify=None)
+            adj = printing.get_adjustment()
+            fmt_index = adj.adjoin(2, *fmt_index).split("\n")
+        else:
+            fmt_index = index._format_flat(include_name=True)
+        fmt_values = self._get_formatted_values()
+
+        if self.is_truncated_vertically:
+            n_header_rows = 0
+            row_num = self.tr_row_num
+            row_num = cast(int, row_num)
+            width = self.adj.len(fmt_values[row_num - 1])
+            if width > 3:
+                dot_str = "..."
+            else:
+                dot_str = ".."
+            # Series uses mode=center because it has single value columns
+            # DataFrame uses mode=left
+            dot_str = self.adj.justify([dot_str], width, mode="center")[0]
+            fmt_values.insert(row_num + n_header_rows, dot_str)
+            fmt_index.insert(row_num + 1, "")
+
+        if self.index:
+            result = self.adj.adjoin(3, *[fmt_index[1:], fmt_values])
+        else:
+            result = self.adj.adjoin(3, fmt_values)
+
+        if self.header and have_header:
+            result = fmt_index[0] + "\n" + result
+
+        if footer:
+            result += "\n" + footer
+
+        return str("".join(result))
+
+
+def get_dataframe_repr_params() -> dict[str, Any]:
+    """Get the parameters used to repr(dataFrame) calls using DataFrame.to_string.
+
+    Supplying these parameters to DataFrame.to_string is equivalent to calling
+    ``repr(DataFrame)``. This is useful if you want to adjust the repr output.
+
+    .. versionadded:: 1.4.0
+
+    Example
+    -------
+    >>> import pandas as pd
+    >>>
+    >>> df = pd.DataFrame([[1, 2], [3, 4]])
+    >>> repr_params = pd.io.formats.format.get_dataframe_repr_params()
+    >>> repr(df) == df.to_string(**repr_params)
+    True
+    """
+    from pandas.io.formats import console
+
+    if get_option("display.expand_frame_repr"):
+        line_width, _ = console.get_console_size()
+    else:
+        line_width = None
+    return {
+        "max_rows": get_option("display.max_rows"),
+        "min_rows": get_option("display.min_rows"),
+        "max_cols": get_option("display.max_columns"),
+        "max_colwidth": get_option("display.max_colwidth"),
+        "show_dimensions": get_option("display.show_dimensions"),
+        "line_width": line_width,
+    }
+
+
+def get_series_repr_params() -> dict[str, Any]:
+    """Get the parameters used to repr(Series) calls using Series.to_string.
+
+    Supplying these parameters to Series.to_string is equivalent to calling
+    ``repr(series)``. This is useful if you want to adjust the series repr output.
+
+    .. versionadded:: 1.4.0
+
+    Example
+    -------
+    >>> import pandas as pd
+    >>>
+    >>> ser = pd.Series([1, 2, 3, 4])
+    >>> repr_params = pd.io.formats.format.get_series_repr_params()
+    >>> repr(ser) == ser.to_string(**repr_params)
+    True
+    """
+    width, height = get_terminal_size()
+    max_rows_opt = get_option("display.max_rows")
+    max_rows = height if max_rows_opt == 0 else max_rows_opt
+    min_rows = height if max_rows_opt == 0 else get_option("display.min_rows")
+
+    return {
+        "name": True,
+        "dtype": True,
+        "min_rows": min_rows,
+        "max_rows": max_rows,
+        "length": get_option("display.show_dimensions"),
+    }
+
+
+class DataFrameFormatter:
+    """
+    Class for processing dataframe formatting options and data.
+
+    Used by DataFrame.to_string, which backs DataFrame.__repr__.
+    """
+
+    __doc__ = __doc__ if __doc__ else ""
+    __doc__ += common_docstring + return_docstring
+
+    def __init__(
+        self,
+        frame: DataFrame,
+        columns: Axes | None = None,
+        col_space: ColspaceArgType | None = None,
+        header: bool | SequenceNotStr[str] = True,
+        index: bool = True,
+        na_rep: str = "NaN",
+        formatters: FormattersType | None = None,
+        justify: str | None = None,
+        float_format: FloatFormatType | None = None,
+        sparsify: bool | None = None,
+        index_names: bool = True,
+        max_rows: int | None = None,
+        min_rows: int | None = None,
+        max_cols: int | None = None,
+        show_dimensions: bool | str = False,
+        decimal: str = ".",
+        bold_rows: bool = False,
+        escape: bool = True,
+    ) -> None:
+        self.frame = frame
+        self.columns = self._initialize_columns(columns)
+        self.col_space = self._initialize_colspace(col_space)
+        self.header = header
+        self.index = index
+        self.na_rep = na_rep
+        self.formatters = self._initialize_formatters(formatters)
+        self.justify = self._initialize_justify(justify)
+        self.float_format = float_format
+        self.sparsify = self._initialize_sparsify(sparsify)
+        self.show_index_names = index_names
+        self.decimal = decimal
+        self.bold_rows = bold_rows
+        self.escape = escape
+        self.max_rows = max_rows
+        self.min_rows = min_rows
+        self.max_cols = max_cols
+        self.show_dimensions = show_dimensions
+
+        self.max_cols_fitted = self._calc_max_cols_fitted()
+        self.max_rows_fitted = self._calc_max_rows_fitted()
+
+        self.tr_frame = self.frame
+        self.truncate()
+        self.adj = printing.get_adjustment()
+
+    def get_strcols(self) -> list[list[str]]:
+        """
+        Render a DataFrame to a list of columns (as lists of strings).
+        """
+        strcols = self._get_strcols_without_index()
+
+        if self.index:
+            str_index = self._get_formatted_index(self.tr_frame)
+            strcols.insert(0, str_index)
+
+        return strcols
+
+    @property
+    def should_show_dimensions(self) -> bool:
+        return self.show_dimensions is True or (
+            self.show_dimensions == "truncate" and self.is_truncated
+        )
+
+    @property
+    def is_truncated(self) -> bool:
+        return bool(self.is_truncated_horizontally or self.is_truncated_vertically)
+
+    @property
+    def is_truncated_horizontally(self) -> bool:
+        return bool(self.max_cols_fitted and (len(self.columns) > self.max_cols_fitted))
+
+    @property
+    def is_truncated_vertically(self) -> bool:
+        return bool(self.max_rows_fitted and (len(self.frame) > self.max_rows_fitted))
+
+    @property
+    def dimensions_info(self) -> str:
+        return f"\n\n[{len(self.frame)} rows x {len(self.frame.columns)} columns]"
+
+    @property
+    def has_index_names(self) -> bool:
+        return _has_names(self.frame.index)
+
+    @property
+    def has_column_names(self) -> bool:
+        return _has_names(self.frame.columns)
+
+    @property
+    def show_row_idx_names(self) -> bool:
+        return all((self.has_index_names, self.index, self.show_index_names))
+
+    @property
+    def show_col_idx_names(self) -> bool:
+        return all((self.has_column_names, self.show_index_names, self.header))
+
+    @property
+    def max_rows_displayed(self) -> int:
+        return min(self.max_rows or len(self.frame), len(self.frame))
+
+    def _initialize_sparsify(self, sparsify: bool | None) -> bool:
+        if sparsify is None:
+            return get_option("display.multi_sparse")
+        return sparsify
+
+    def _initialize_formatters(
+        self, formatters: FormattersType | None
+    ) -> FormattersType:
+        if formatters is None:
+            return {}
+        elif len(self.frame.columns) == len(formatters) or isinstance(formatters, dict):
+            return formatters
+        else:
+            raise ValueError(
+                f"Formatters length({len(formatters)}) should match "
+                f"DataFrame number of columns({len(self.frame.columns)})"
+            )
+
+    def _initialize_justify(self, justify: str | None) -> str:
+        if justify is None:
+            return get_option("display.colheader_justify")
+        else:
+            return justify
+
+    def _initialize_columns(self, columns: Axes | None) -> Index:
+        if columns is not None:
+            cols = ensure_index(columns)
+            self.frame = self.frame[cols]
+            return cols
+        else:
+            return self.frame.columns
+
+    def _initialize_colspace(self, col_space: ColspaceArgType | None) -> ColspaceType:
+        result: ColspaceType
+
+        if col_space is None:
+            result = {}
+        elif isinstance(col_space, (int, str)):
+            result = {"": col_space}
+            result.update({column: col_space for column in self.frame.columns})
+        elif isinstance(col_space, Mapping):
+            for column in col_space.keys():
+                if column not in self.frame.columns and column != "":
+                    raise ValueError(
+                        f"Col_space is defined for an unknown column: {column}"
+                    )
+            result = col_space
+        else:
+            if len(self.frame.columns) != len(col_space):
+                raise ValueError(
+                    f"Col_space length({len(col_space)}) should match "
+                    f"DataFrame number of columns({len(self.frame.columns)})"
+                )
+            result = dict(zip(self.frame.columns, col_space))
+        return result
+
+    def _calc_max_cols_fitted(self) -> int | None:
+        """Number of columns fitting the screen."""
+        if not self._is_in_terminal():
+            return self.max_cols
+
+        width, _ = get_terminal_size()
+        if self._is_screen_narrow(width):
+            return width
+        else:
+            return self.max_cols
+
+    def _calc_max_rows_fitted(self) -> int | None:
+        """Number of rows with data fitting the screen."""
+        max_rows: int | None
+
+        if self._is_in_terminal():
+            _, height = get_terminal_size()
+            if self.max_rows == 0:
+                # rows available to fill with actual data
+                return height - self._get_number_of_auxiliary_rows()
+
+            if self._is_screen_short(height):
+                max_rows = height
+            else:
+                max_rows = self.max_rows
+        else:
+            max_rows = self.max_rows
+
+        return self._adjust_max_rows(max_rows)
+
+    def _adjust_max_rows(self, max_rows: int | None) -> int | None:
+        """Adjust max_rows using display logic.
+
+        See description here:
+        https://pandas.pydata.org/docs/dev/user_guide/options.html#frequently-used-options
+
+        GH #37359
+        """
+        if max_rows:
+            if (len(self.frame) > max_rows) and self.min_rows:
+                # if truncated, set max_rows showed to min_rows
+                max_rows = min(self.min_rows, max_rows)
+        return max_rows
+
+    def _is_in_terminal(self) -> bool:
+        """Check if the output is to be shown in terminal."""
+        return bool(self.max_cols == 0 or self.max_rows == 0)
+
+    def _is_screen_narrow(self, max_width) -> bool:
+        return bool(self.max_cols == 0 and len(self.frame.columns) > max_width)
+
+    def _is_screen_short(self, max_height) -> bool:
+        return bool(self.max_rows == 0 and len(self.frame) > max_height)
+
+    def _get_number_of_auxiliary_rows(self) -> int:
+        """Get number of rows occupied by prompt, dots and dimension info."""
+        dot_row = 1
+        prompt_row = 1
+        num_rows = dot_row + prompt_row
+
+        if self.show_dimensions:
+            num_rows += len(self.dimensions_info.splitlines())
+
+        if self.header:
+            num_rows += 1
+
+        return num_rows
+
+    def truncate(self) -> None:
+        """
+        Check whether the frame should be truncated. If so, slice the frame up.
+        """
+        if self.is_truncated_horizontally:
+            self._truncate_horizontally()
+
+        if self.is_truncated_vertically:
+            self._truncate_vertically()
+
+    def _truncate_horizontally(self) -> None:
+        """Remove columns, which are not to be displayed and adjust formatters.
+
+        Attributes affected:
+            - tr_frame
+            - formatters
+            - tr_col_num
+        """
+        assert self.max_cols_fitted is not None
+        col_num = self.max_cols_fitted // 2
+        if col_num >= 1:
+            _len = len(self.tr_frame.columns)
+            _slice = np.hstack([np.arange(col_num), np.arange(_len - col_num, _len)])
+            self.tr_frame = self.tr_frame.iloc[:, _slice]
+
+            # truncate formatter
+            if isinstance(self.formatters, (list, tuple)):
+                self.formatters = [
+                    *self.formatters[:col_num],
+                    *self.formatters[-col_num:],
+                ]
+        else:
+            col_num = cast(int, self.max_cols)
+            self.tr_frame = self.tr_frame.iloc[:, :col_num]
+        self.tr_col_num: int = col_num
+
+    def _truncate_vertically(self) -> None:
+        """Remove rows, which are not to be displayed.
+
+        Attributes affected:
+            - tr_frame
+            - tr_row_num
+        """
+        assert self.max_rows_fitted is not None
+        row_num = self.max_rows_fitted // 2
+        if row_num >= 1:
+            _len = len(self.tr_frame)
+            _slice = np.hstack([np.arange(row_num), np.arange(_len - row_num, _len)])
+            self.tr_frame = self.tr_frame.iloc[_slice]
+        else:
+            row_num = cast(int, self.max_rows)
+            self.tr_frame = self.tr_frame.iloc[:row_num, :]
+        self.tr_row_num = row_num
+
+    def _get_strcols_without_index(self) -> list[list[str]]:
+        strcols: list[list[str]] = []
+
+        if not is_list_like(self.header) and not self.header:
+            for i, c in enumerate(self.tr_frame):
+                fmt_values = self.format_col(i)
+                fmt_values = _make_fixed_width(
+                    strings=fmt_values,
+                    justify=self.justify,
+                    minimum=int(self.col_space.get(c, 0)),
+                    adj=self.adj,
+                )
+                strcols.append(fmt_values)
+            return strcols
+
+        if is_list_like(self.header):
+            # cast here since can't be bool if is_list_like
+            self.header = cast(list[str], self.header)
+            if len(self.header) != len(self.columns):
+                raise ValueError(
+                    f"Writing {len(self.columns)} cols "
+                    f"but got {len(self.header)} aliases"
+                )
+            str_columns = [[label] for label in self.header]
+        else:
+            str_columns = self._get_formatted_column_labels(self.tr_frame)
+
+        if self.show_row_idx_names:
+            for x in str_columns:
+                x.append("")
+
+        for i, c in enumerate(self.tr_frame):
+            cheader = str_columns[i]
+            header_colwidth = max(
+                int(self.col_space.get(c, 0)), *(self.adj.len(x) for x in cheader)
+            )
+            fmt_values = self.format_col(i)
+            fmt_values = _make_fixed_width(
+                fmt_values, self.justify, minimum=header_colwidth, adj=self.adj
+            )
+
+            max_len = max(*(self.adj.len(x) for x in fmt_values), header_colwidth)
+            cheader = self.adj.justify(cheader, max_len, mode=self.justify)
+            strcols.append(cheader + fmt_values)
+
+        return strcols
+
+    def format_col(self, i: int) -> list[str]:
+        frame = self.tr_frame
+        formatter = self._get_formatter(i)
+        return format_array(
+            frame.iloc[:, i]._values,
+            formatter,
+            float_format=self.float_format,
+            na_rep=self.na_rep,
+            space=self.col_space.get(frame.columns[i]),
+            decimal=self.decimal,
+            leading_space=self.index,
+        )
+
+    def _get_formatter(self, i: str | int) -> Callable | None:
+        if isinstance(self.formatters, (list, tuple)):
+            if is_integer(i):
+                i = cast(int, i)
+                return self.formatters[i]
+            else:
+                return None
+        else:
+            if is_integer(i) and i not in self.columns:
+                i = self.columns[i]
+            return self.formatters.get(i, None)
+
+    def _get_formatted_column_labels(self, frame: DataFrame) -> list[list[str]]:
+        from pandas.core.indexes.multi import sparsify_labels
+
+        columns = frame.columns
+
+        if isinstance(columns, MultiIndex):
+            fmt_columns = columns._format_multi(sparsify=False, include_names=False)
+            if self.sparsify and len(fmt_columns):
+                fmt_columns = sparsify_labels(fmt_columns)
+
+            str_columns = [list(x) for x in zip(*fmt_columns)]
+        else:
+            fmt_columns = columns._format_flat(include_name=False)
+            str_columns = [
+                [
+                    " " + x
+                    if not self._get_formatter(i) and is_numeric_dtype(dtype)
+                    else x
+                ]
+                for i, (x, dtype) in enumerate(zip(fmt_columns, self.frame.dtypes))
+            ]
+        return str_columns
+
+    def _get_formatted_index(self, frame: DataFrame) -> list[str]:
+        # Note: this is only used by to_string() and to_latex(), not by
+        # to_html(). so safe to cast col_space here.
+        col_space = {k: cast(int, v) for k, v in self.col_space.items()}
+        index = frame.index
+        columns = frame.columns
+        fmt = self._get_formatter("__index__")
+
+        if isinstance(index, MultiIndex):
+            fmt_index = index._format_multi(
+                sparsify=self.sparsify,
+                include_names=self.show_row_idx_names,
+                formatter=fmt,
+            )
+        else:
+            fmt_index = [
+                index._format_flat(include_name=self.show_row_idx_names, formatter=fmt)
+            ]
+
+        fmt_index = [
+            tuple(
+                _make_fixed_width(
+                    list(x), justify="left", minimum=col_space.get("", 0), adj=self.adj
+                )
+            )
+            for x in fmt_index
+        ]
+
+        adjoined = self.adj.adjoin(1, *fmt_index).split("\n")
+
+        # empty space for columns
+        if self.show_col_idx_names:
+            col_header = [str(x) for x in self._get_column_name_list()]
+        else:
+            col_header = [""] * columns.nlevels
+
+        if self.header:
+            return col_header + adjoined
+        else:
+            return adjoined
+
+    def _get_column_name_list(self) -> list[Hashable]:
+        names: list[Hashable] = []
+        columns = self.frame.columns
+        if isinstance(columns, MultiIndex):
+            names.extend("" if name is None else name for name in columns.names)
+        else:
+            names.append("" if columns.name is None else columns.name)
+        return names
+
+
+class DataFrameRenderer:
+    """Class for creating dataframe output in multiple formats.
+
+    Called in pandas.core.generic.NDFrame:
+        - to_csv
+        - to_latex
+
+    Called in pandas.DataFrame:
+        - to_html
+        - to_string
+
+    Parameters
+    ----------
+    fmt : DataFrameFormatter
+        Formatter with the formatting options.
+    """
+
+    def __init__(self, fmt: DataFrameFormatter) -> None:
+        self.fmt = fmt
+
+    def to_html(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        encoding: str | None = None,
+        classes: str | list | tuple | None = None,
+        notebook: bool = False,
+        border: int | bool | None = None,
+        table_id: str | None = None,
+        render_links: bool = False,
+    ) -> str | None:
+        """
+        Render a DataFrame to a html table.
+
+        Parameters
+        ----------
+        buf : str, path object, file-like object, or None, default None
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a string ``write()`` function. If None, the result is
+            returned as a string.
+        encoding : str, default “utf-8”
+            Set character encoding.
+        classes : str or list-like
+            classes to include in the `class` attribute of the opening
+            ``<table>`` tag, in addition to the default "dataframe".
+        notebook : {True, False}, optional, default False
+            Whether the generated HTML is for IPython Notebook.
+        border : int or bool
+            When an integer value is provided, it sets the border attribute in
+            the opening tag, specifying the thickness of the border.
+            If ``False`` or ``0`` is passed, the border attribute will not
+            be present in the ``<table>`` tag.
+            The default value for this parameter is governed by
+            ``pd.options.display.html.border``.
+        table_id : str, optional
+            A css id is included in the opening `<table>` tag if specified.
+        render_links : bool, default False
+            Convert URLs to HTML links.
+        """
+        from pandas.io.formats.html import (
+            HTMLFormatter,
+            NotebookFormatter,
+        )
+
+        Klass = NotebookFormatter if notebook else HTMLFormatter
+
+        html_formatter = Klass(
+            self.fmt,
+            classes=classes,
+            border=border,
+            table_id=table_id,
+            render_links=render_links,
+        )
+        string = html_formatter.to_string()
+        return save_to_buffer(string, buf=buf, encoding=encoding)
+
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        encoding: str | None = None,
+        line_width: int | None = None,
+    ) -> str | None:
+        """
+        Render a DataFrame to a console-friendly tabular output.
+
+        Parameters
+        ----------
+        buf : str, path object, file-like object, or None, default None
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a string ``write()`` function. If None, the result is
+            returned as a string.
+        encoding: str, default “utf-8”
+            Set character encoding.
+        line_width : int, optional
+            Width to wrap a line in characters.
+        """
+        from pandas.io.formats.string import StringFormatter
+
+        string_formatter = StringFormatter(self.fmt, line_width=line_width)
+        string = string_formatter.to_string()
+        return save_to_buffer(string, buf=buf, encoding=encoding)
+
+    def to_csv(
+        self,
+        path_or_buf: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = None,
+        encoding: str | None = None,
+        sep: str = ",",
+        columns: Sequence[Hashable] | None = None,
+        index_label: IndexLabel | None = None,
+        mode: str = "w",
+        compression: CompressionOptions = "infer",
+        quoting: int | None = None,
+        quotechar: str = '"',
+        lineterminator: str | None = None,
+        chunksize: int | None = None,
+        date_format: str | None = None,
+        doublequote: bool = True,
+        escapechar: str | None = None,
+        errors: str = "strict",
+        storage_options: StorageOptions | None = None,
+    ) -> str | None:
+        """
+        Render dataframe as comma-separated file.
+        """
+        from pandas.io.formats.csvs import CSVFormatter
+
+        if path_or_buf is None:
+            created_buffer = True
+            path_or_buf = StringIO()
+        else:
+            created_buffer = False
+
+        csv_formatter = CSVFormatter(
+            path_or_buf=path_or_buf,
+            lineterminator=lineterminator,
+            sep=sep,
+            encoding=encoding,
+            errors=errors,
+            compression=compression,
+            quoting=quoting,
+            cols=columns,
+            index_label=index_label,
+            mode=mode,
+            chunksize=chunksize,
+            quotechar=quotechar,
+            date_format=date_format,
+            doublequote=doublequote,
+            escapechar=escapechar,
+            storage_options=storage_options,
+            formatter=self.fmt,
+        )
+        csv_formatter.save()
+
+        if created_buffer:
+            assert isinstance(path_or_buf, StringIO)
+            content = path_or_buf.getvalue()
+            path_or_buf.close()
+            return content
+
+        return None
+
+
+def save_to_buffer(
+    string: str,
+    buf: FilePath | WriteBuffer[str] | None = None,
+    encoding: str | None = None,
+) -> str | None:
+    """
+    Perform serialization. Write to buf or return as string if buf is None.
+    """
+    with _get_buffer(buf, encoding=encoding) as fd:
+        fd.write(string)
+        if buf is None:
+            # error: "WriteBuffer[str]" has no attribute "getvalue"
+            return fd.getvalue()  # type: ignore[attr-defined]
+        return None
+
+
+@contextmanager
+def _get_buffer(
+    buf: FilePath | WriteBuffer[str] | None, encoding: str | None = None
+) -> Generator[WriteBuffer[str]] | Generator[StringIO]:
+    """
+    Context manager to open, yield and close buffer for filenames or Path-like
+    objects, otherwise yield buf unchanged.
+    """
+    if buf is not None:
+        buf = stringify_path(buf)
+    else:
+        buf = StringIO()
+
+    if encoding is None:
+        encoding = "utf-8"
+    elif not isinstance(buf, str):
+        raise ValueError("buf is not a file name and encoding is specified.")
+
+    if hasattr(buf, "write"):
+        # Incompatible types in "yield" (actual type "Union[str, WriteBuffer[str],
+        # StringIO]", expected type "Union[WriteBuffer[str], StringIO]")
+        yield buf  # type: ignore[misc]
+    elif isinstance(buf, str):
+        check_parent_directory(str(buf))
+        with open(buf, "w", encoding=encoding, newline="") as f:
+            # GH#30034 open instead of codecs.open prevents a file leak
+            #  if we have an invalid encoding argument.
+            # newline="" is needed to roundtrip correctly on
+            #  windows test_to_latex_filename
+            yield f
+    else:
+        raise TypeError("buf is not a file name and it has no write method")
+
+
+# ----------------------------------------------------------------------
+# Array formatters
+
+
+def format_array(
+    values: ArrayLike,
+    formatter: Callable | None,
+    float_format: FloatFormatType | None = None,
+    na_rep: str = "NaN",
+    digits: int | None = None,
+    space: str | int | None = None,
+    justify: str = "right",
+    decimal: str = ".",
+    leading_space: bool | None = True,
+    quoting: int | None = None,
+    fallback_formatter: Callable | None = None,
+) -> list[str]:
+    """
+    Format an array for printing.
+
+    Parameters
+    ----------
+    values : np.ndarray or ExtensionArray
+    formatter
+    float_format
+    na_rep
+    digits
+    space
+    justify
+    decimal
+    leading_space : bool, optional, default True
+        Whether the array should be formatted with a leading space.
+        When an array as a column of a Series or DataFrame, we do want
+        the leading space to pad between columns.
+
+        When formatting an Index subclass
+        (e.g. IntervalIndex._get_values_for_csv), we don't want the
+        leading space since it should be left-aligned.
+    fallback_formatter
+
+    Returns
+    -------
+    List[str]
+    """
+    fmt_klass: type[_GenericArrayFormatter]
+    if lib.is_np_dtype(values.dtype, "M"):
+        fmt_klass = _Datetime64Formatter
+        values = cast(DatetimeArray, values)
+    elif isinstance(values.dtype, DatetimeTZDtype):
+        fmt_klass = _Datetime64TZFormatter
+        values = cast(DatetimeArray, values)
+    elif lib.is_np_dtype(values.dtype, "m"):
+        fmt_klass = _Timedelta64Formatter
+        values = cast(TimedeltaArray, values)
+    elif isinstance(values.dtype, ExtensionDtype):
+        fmt_klass = _ExtensionArrayFormatter
+    elif lib.is_np_dtype(values.dtype, "fc"):
+        fmt_klass = FloatArrayFormatter
+    elif lib.is_np_dtype(values.dtype, "iu"):
+        fmt_klass = _IntArrayFormatter
+    else:
+        fmt_klass = _GenericArrayFormatter
+
+    if space is None:
+        space = 12
+
+    if float_format is None:
+        float_format = get_option("display.float_format")
+
+    if digits is None:
+        digits = get_option("display.precision")
+
+    fmt_obj = fmt_klass(
+        values,
+        digits=digits,
+        na_rep=na_rep,
+        float_format=float_format,
+        formatter=formatter,
+        space=space,
+        justify=justify,
+        decimal=decimal,
+        leading_space=leading_space,
+        quoting=quoting,
+        fallback_formatter=fallback_formatter,
+    )
+
+    return fmt_obj.get_result()
+
+
+class _GenericArrayFormatter:
+    def __init__(
+        self,
+        values: ArrayLike,
+        digits: int = 7,
+        formatter: Callable | None = None,
+        na_rep: str = "NaN",
+        space: str | int = 12,
+        float_format: FloatFormatType | None = None,
+        justify: str = "right",
+        decimal: str = ".",
+        quoting: int | None = None,
+        fixed_width: bool = True,
+        leading_space: bool | None = True,
+        fallback_formatter: Callable | None = None,
+    ) -> None:
+        self.values = values
+        self.digits = digits
+        self.na_rep = na_rep
+        self.space = space
+        self.formatter = formatter
+        self.float_format = float_format
+        self.justify = justify
+        self.decimal = decimal
+        self.quoting = quoting
+        self.fixed_width = fixed_width
+        self.leading_space = leading_space
+        self.fallback_formatter = fallback_formatter
+
+    def get_result(self) -> list[str]:
+        fmt_values = self._format_strings()
+        return _make_fixed_width(fmt_values, self.justify)
+
+    def _format_strings(self) -> list[str]:
+        if self.float_format is None:
+            float_format = get_option("display.float_format")
+            if float_format is None:
+                precision = get_option("display.precision")
+                float_format = lambda x: _trim_zeros_single_float(
+                    f"{x: .{precision:d}f}"
+                )
+        else:
+            float_format = self.float_format
+
+        if self.formatter is not None:
+            formatter = self.formatter
+        elif self.fallback_formatter is not None:
+            formatter = self.fallback_formatter
+        else:
+            quote_strings = self.quoting is not None and self.quoting != QUOTE_NONE
+            formatter = partial(
+                printing.pprint_thing,
+                escape_chars=("\t", "\r", "\n"),
+                quote_strings=quote_strings,
+            )
+
+        def _format(x):
+            if self.na_rep is not None and is_scalar(x) and isna(x):
+                if x is None:
+                    return "None"
+                elif x is NA:
+                    return str(NA)
+                elif x is NaT or isinstance(x, (np.datetime64, np.timedelta64)):
+                    return "NaT"
+                return self.na_rep
+            elif isinstance(x, PandasObject):
+                return str(x)
+            elif isinstance(x, StringDtype):
+                return repr(x)
+            else:
+                # object dtype
+                return str(formatter(x))
+
+        vals = self.values
+        if not isinstance(vals, np.ndarray):
+            raise TypeError(
+                "ExtensionArray formatting should use _ExtensionArrayFormatter"
+            )
+        inferred = lib.map_infer(vals, is_float)
+        is_float_type = (
+            inferred
+            # vals may have 2 or more dimensions
+            & np.all(notna(vals), axis=tuple(range(1, len(vals.shape))))
+        )
+        leading_space = self.leading_space
+        if leading_space is None:
+            leading_space = is_float_type.any()
+
+        fmt_values = []
+        for i, v in enumerate(vals):
+            if (not is_float_type[i] or self.formatter is not None) and leading_space:
+                fmt_values.append(f" {_format(v)}")
+            elif is_float_type[i]:
+                fmt_values.append(float_format(v))
+            else:
+                if leading_space is False:
+                    # False specifically, so that the default is
+                    # to include a space if we get here.
+                    tpl = "{v}"
+                else:
+                    tpl = " {v}"
+                fmt_values.append(tpl.format(v=_format(v)))
+
+        return fmt_values
+
+
+class FloatArrayFormatter(_GenericArrayFormatter):
+    def __init__(self, *args, **kwargs) -> None:
+        super().__init__(*args, **kwargs)
+
+        # float_format is expected to be a string
+        # formatter should be used to pass a function
+        if self.float_format is not None and self.formatter is None:
+            # GH21625, GH22270
+            self.fixed_width = False
+            if callable(self.float_format):
+                self.formatter = self.float_format
+                self.float_format = None
+
+    def _value_formatter(
+        self,
+        float_format: FloatFormatType | None = None,
+        threshold: float | None = None,
+    ) -> Callable:
+        """Returns a function to be applied on each value to format it"""
+        # the float_format parameter supersedes self.float_format
+        if float_format is None:
+            float_format = self.float_format
+
+        # we are going to compose different functions, to first convert to
+        # a string, then replace the decimal symbol, and finally chop according
+        # to the threshold
+
+        # when there is no float_format, we use str instead of '%g'
+        # because str(0.0) = '0.0' while '%g' % 0.0 = '0'
+        if float_format:
+
+            def base_formatter(v):
+                assert float_format is not None  # for mypy
+                # error: "str" not callable
+                # error: Unexpected keyword argument "value" for "__call__" of
+                # "EngFormatter"
+                return (
+                    float_format(value=v)  # type: ignore[operator,call-arg]
+                    if notna(v)
+                    else self.na_rep
+                )
+
+        else:
+
+            def base_formatter(v):
+                return str(v) if notna(v) else self.na_rep
+
+        if self.decimal != ".":
+
+            def decimal_formatter(v):
+                return base_formatter(v).replace(".", self.decimal, 1)
+
+        else:
+            decimal_formatter = base_formatter
+
+        if threshold is None:
+            return decimal_formatter
+
+        def formatter(value):
+            if notna(value):
+                if abs(value) > threshold:
+                    return decimal_formatter(value)
+                else:
+                    return decimal_formatter(0.0)
+            else:
+                return self.na_rep
+
+        return formatter
+
+    def get_result_as_array(self) -> np.ndarray:
+        """
+        Returns the float values converted into strings using
+        the parameters given at initialisation, as a numpy array
+        """
+
+        def format_with_na_rep(
+            values: ArrayLike, formatter: Callable, na_rep: str
+        ) -> np.ndarray:
+            mask = isna(values)
+            formatted = np.array(
+                [
+                    formatter(val) if not m else na_rep
+                    for val, m in zip(values.ravel(), mask.ravel())
+                ]
+            ).reshape(values.shape)
+            return formatted
+
+        def format_complex_with_na_rep(
+            values: ArrayLike, formatter: Callable, na_rep: str
+        ) -> np.ndarray:
+            real_values = np.real(values).ravel()  # type: ignore[arg-type]
+            imag_values = np.imag(values).ravel()  # type: ignore[arg-type]
+            real_mask, imag_mask = isna(real_values), isna(imag_values)
+            formatted_lst = []
+            for val, real_val, imag_val, re_isna, im_isna in zip(
+                values.ravel(),
+                real_values,
+                imag_values,
+                real_mask,
+                imag_mask,
+            ):
+                if not re_isna and not im_isna:
+                    formatted_lst.append(formatter(val))
+                elif not re_isna:  # xxx+nanj
+                    formatted_lst.append(f"{formatter(real_val)}+{na_rep}j")
+                elif not im_isna:  # nan[+/-]xxxj
+                    # The imaginary part may either start with a "-" or a space
+                    imag_formatted = formatter(imag_val).strip()
+                    if imag_formatted.startswith("-"):
+                        formatted_lst.append(f"{na_rep}{imag_formatted}j")
+                    else:
+                        formatted_lst.append(f"{na_rep}+{imag_formatted}j")
+                else:  # nan+nanj
+                    formatted_lst.append(f"{na_rep}+{na_rep}j")
+            return np.array(formatted_lst).reshape(values.shape)
+
+        if self.formatter is not None:
+            return format_with_na_rep(self.values, self.formatter, self.na_rep)
+
+        if self.fixed_width:
+            threshold = get_option("display.chop_threshold")
+        else:
+            threshold = None
+
+        # if we have a fixed_width, we'll need to try different float_format
+        def format_values_with(float_format):
+            formatter = self._value_formatter(float_format, threshold)
+
+            # default formatter leaves a space to the left when formatting
+            # floats, must be consistent for left-justifying NaNs (GH #25061)
+            na_rep = " " + self.na_rep if self.justify == "left" else self.na_rep
+
+            # different formatting strategies for complex and non-complex data
+            # need to distinguish complex and float NaNs (GH #53762)
+            values = self.values
+            is_complex = is_complex_dtype(values)
+
+            # separate the wheat from the chaff
+            if is_complex:
+                values = format_complex_with_na_rep(values, formatter, na_rep)
+            else:
+                values = format_with_na_rep(values, formatter, na_rep)
+
+            if self.fixed_width:
+                if is_complex:
+                    result = _trim_zeros_complex(values, self.decimal)
+                else:
+                    result = _trim_zeros_float(values, self.decimal)
+                return np.asarray(result, dtype="object")
+
+            return values
+
+        # There is a special default string when we are fixed-width
+        # The default is otherwise to use str instead of a formatting string
+        float_format: FloatFormatType | None
+        if self.float_format is None:
+            if self.fixed_width:
+                if self.leading_space is True:
+                    fmt_str = "{value: .{digits:d}f}"
+                else:
+                    fmt_str = "{value:.{digits:d}f}"
+                float_format = partial(fmt_str.format, digits=self.digits)
+            else:
+                float_format = self.float_format
+        else:
+            float_format = lambda value: self.float_format % value
+
+        formatted_values = format_values_with(float_format)
+
+        if not self.fixed_width:
+            return formatted_values
+
+        # we need do convert to engineering format if some values are too small
+        # and would appear as 0, or if some values are too big and take too
+        # much space
+
+        if len(formatted_values) > 0:
+            maxlen = max(len(x) for x in formatted_values)
+            too_long = maxlen > self.digits + 6
+        else:
+            too_long = False
+
+        abs_vals = np.abs(self.values)
+        # this is pretty arbitrary for now
+        # large values: more that 8 characters including decimal symbol
+        # and first digit, hence > 1e6
+        has_large_values = (abs_vals > 1e6).any()
+        has_small_values = ((abs_vals < 10 ** (-self.digits)) & (abs_vals > 0)).any()
+
+        if has_small_values or (too_long and has_large_values):
+            if self.leading_space is True:
+                fmt_str = "{value: .{digits:d}e}"
+            else:
+                fmt_str = "{value:.{digits:d}e}"
+            float_format = partial(fmt_str.format, digits=self.digits)
+            formatted_values = format_values_with(float_format)
+
+        return formatted_values
+
+    def _format_strings(self) -> list[str]:
+        return list(self.get_result_as_array())
+
+
+class _IntArrayFormatter(_GenericArrayFormatter):
+    def _format_strings(self) -> list[str]:
+        if self.leading_space is False:
+            formatter_str = lambda x: f"{x:d}".format(x=x)
+        else:
+            formatter_str = lambda x: f"{x: d}".format(x=x)
+        formatter = self.formatter or formatter_str
+        fmt_values = [formatter(x) for x in self.values]
+        return fmt_values
+
+
+class _Datetime64Formatter(_GenericArrayFormatter):
+    values: DatetimeArray
+
+    def __init__(
+        self,
+        values: DatetimeArray,
+        nat_rep: str = "NaT",
+        date_format: None = None,
+        **kwargs,
+    ) -> None:
+        super().__init__(values, **kwargs)
+        self.nat_rep = nat_rep
+        self.date_format = date_format
+
+    def _format_strings(self) -> list[str]:
+        """we by definition have DO NOT have a TZ"""
+        values = self.values
+
+        if self.formatter is not None:
+            return [self.formatter(x) for x in values]
+
+        fmt_values = values._format_native_types(
+            na_rep=self.nat_rep, date_format=self.date_format
+        )
+        return fmt_values.tolist()
+
+
+class _ExtensionArrayFormatter(_GenericArrayFormatter):
+    values: ExtensionArray
+
+    def _format_strings(self) -> list[str]:
+        values = self.values
+
+        formatter = self.formatter
+        fallback_formatter = None
+        if formatter is None:
+            fallback_formatter = values._formatter(boxed=True)
+
+        if isinstance(values, Categorical):
+            # Categorical is special for now, so that we can preserve tzinfo
+            array = values._internal_get_values()
+        else:
+            array = np.asarray(values, dtype=object)
+
+        fmt_values = format_array(
+            array,
+            formatter,
+            float_format=self.float_format,
+            na_rep=self.na_rep,
+            digits=self.digits,
+            space=self.space,
+            justify=self.justify,
+            decimal=self.decimal,
+            leading_space=self.leading_space,
+            quoting=self.quoting,
+            fallback_formatter=fallback_formatter,
+        )
+        return fmt_values
+
+
+def format_percentiles(
+    percentiles: np.ndarray | Sequence[float],
+) -> list[str]:
+    """
+    Outputs rounded and formatted percentiles.
+
+    Parameters
+    ----------
+    percentiles : list-like, containing floats from interval [0,1]
+
+    Returns
+    -------
+    formatted : list of strings
+
+    Notes
+    -----
+    Rounding precision is chosen so that: (1) if any two elements of
+    ``percentiles`` differ, they remain different after rounding
+    (2) no entry is *rounded* to 0% or 100%.
+    Any non-integer is always rounded to at least 1 decimal place.
+
+    Examples
+    --------
+    Keeps all entries different after rounding:
+
+    >>> format_percentiles([0.01999, 0.02001, 0.5, 0.666666, 0.9999])
+    ['1.999%', '2.001%', '50%', '66.667%', '99.99%']
+
+    No element is rounded to 0% or 100% (unless already equal to it).
+    Duplicates are allowed:
+
+    >>> format_percentiles([0, 0.5, 0.02001, 0.5, 0.666666, 0.9999])
+    ['0%', '50%', '2.0%', '50%', '66.67%', '99.99%']
+    """
+    percentiles = np.asarray(percentiles)
+
+    # It checks for np.nan as well
+    if (
+        not is_numeric_dtype(percentiles)
+        or not np.all(percentiles >= 0)
+        or not np.all(percentiles <= 1)
+    ):
+        raise ValueError("percentiles should all be in the interval [0,1]")
+
+    percentiles = 100 * percentiles
+    prec = get_precision(percentiles)
+    percentiles_round_type = percentiles.round(prec).astype(int)
+
+    int_idx = np.isclose(percentiles_round_type, percentiles)
+
+    if np.all(int_idx):
+        out = percentiles_round_type.astype(str)
+        return [i + "%" for i in out]
+
+    unique_pcts = np.unique(percentiles)
+    prec = get_precision(unique_pcts)
+    out = np.empty_like(percentiles, dtype=object)
+    out[int_idx] = percentiles[int_idx].round().astype(int).astype(str)
+
+    out[~int_idx] = percentiles[~int_idx].round(prec).astype(str)
+    return [i + "%" for i in out]
+
+
+def get_precision(array: np.ndarray | Sequence[float]) -> int:
+    to_begin = array[0] if array[0] > 0 else None
+    to_end = 100 - array[-1] if array[-1] < 100 else None
+    diff = np.ediff1d(array, to_begin=to_begin, to_end=to_end)
+    diff = abs(diff)
+    prec = -np.floor(np.log10(np.min(diff))).astype(int)
+    prec = max(1, prec)
+    return prec
+
+
+def _format_datetime64(x: NaTType | Timestamp, nat_rep: str = "NaT") -> str:
+    if x is NaT:
+        return nat_rep
+
+    # Timestamp.__str__ falls back to datetime.datetime.__str__ = isoformat(sep=' ')
+    # so it already uses string formatting rather than strftime (faster).
+    return str(x)
+
+
+def _format_datetime64_dateonly(
+    x: NaTType | Timestamp,
+    nat_rep: str = "NaT",
+    date_format: str | None = None,
+) -> str:
+    if isinstance(x, NaTType):
+        return nat_rep
+
+    if date_format:
+        return x.strftime(date_format)
+    else:
+        # Timestamp._date_repr relies on string formatting (faster than strftime)
+        return x._date_repr
+
+
+def get_format_datetime64(
+    is_dates_only: bool, nat_rep: str = "NaT", date_format: str | None = None
+) -> Callable:
+    """Return a formatter callable taking a datetime64 as input and providing
+    a string as output"""
+
+    if is_dates_only:
+        return lambda x: _format_datetime64_dateonly(
+            x, nat_rep=nat_rep, date_format=date_format
+        )
+    else:
+        return lambda x: _format_datetime64(x, nat_rep=nat_rep)
+
+
+class _Datetime64TZFormatter(_Datetime64Formatter):
+    values: DatetimeArray
+
+    def _format_strings(self) -> list[str]:
+        """we by definition have a TZ"""
+        ido = self.values._is_dates_only
+        values = self.values.astype(object)
+        formatter = self.formatter or get_format_datetime64(
+            ido, date_format=self.date_format
+        )
+        fmt_values = [formatter(x) for x in values]
+
+        return fmt_values
+
+
+class _Timedelta64Formatter(_GenericArrayFormatter):
+    values: TimedeltaArray
+
+    def __init__(
+        self,
+        values: TimedeltaArray,
+        nat_rep: str = "NaT",
+        **kwargs,
+    ) -> None:
+        # TODO: nat_rep is never passed, na_rep is.
+        super().__init__(values, **kwargs)
+        self.nat_rep = nat_rep
+
+    def _format_strings(self) -> list[str]:
+        formatter = self.formatter or get_format_timedelta64(
+            self.values, nat_rep=self.nat_rep, box=False
+        )
+        return [formatter(x) for x in self.values]
+
+
+def get_format_timedelta64(
+    values: TimedeltaArray,
+    nat_rep: str | float = "NaT",
+    box: bool = False,
+) -> Callable:
+    """
+    Return a formatter function for a range of timedeltas.
+    These will all have the same format argument
+
+    If box, then show the return in quotes
+    """
+    even_days = values._is_dates_only
+
+    if even_days:
+        format = None
+    else:
+        format = "long"
+
+    def _formatter(x):
+        if x is None or (is_scalar(x) and isna(x)):
+            return nat_rep
+
+        if not isinstance(x, Timedelta):
+            x = Timedelta(x)
+
+        # Timedelta._repr_base uses string formatting (faster than strftime)
+        result = x._repr_base(format=format)
+        if box:
+            result = f"'{result}'"
+        return result
+
+    return _formatter
+
+
+def _make_fixed_width(
+    strings: list[str],
+    justify: str = "right",
+    minimum: int | None = None,
+    adj: printing._TextAdjustment | None = None,
+) -> list[str]:
+    if len(strings) == 0 or justify == "all":
+        return strings
+
+    if adj is None:
+        adjustment = printing.get_adjustment()
+    else:
+        adjustment = adj
+
+    max_len = max(adjustment.len(x) for x in strings)
+
+    if minimum is not None:
+        max_len = max(minimum, max_len)
+
+    conf_max = get_option("display.max_colwidth")
+    if conf_max is not None and max_len > conf_max:
+        max_len = conf_max
+
+    def just(x: str) -> str:
+        if conf_max is not None:
+            if (conf_max > 3) & (adjustment.len(x) > max_len):
+                x = x[: max_len - 3] + "..."
+        return x
+
+    strings = [just(x) for x in strings]
+    result = adjustment.justify(strings, max_len, mode=justify)
+    return result
+
+
+def _trim_zeros_complex(str_complexes: ArrayLike, decimal: str = ".") -> list[str]:
+    """
+    Separates the real and imaginary parts from the complex number, and
+    executes the _trim_zeros_float method on each of those.
+    """
+    real_part, imag_part = [], []
+    for x in str_complexes:
+        # Complex numbers are represented as "(-)xxx(+/-)xxxj"
+        # The split will give [{"", "-"}, "xxx", "+/-", "xxx", "j", ""]
+        # Therefore, the imaginary part is the 4th and 3rd last elements,
+        # and the real part is everything before the imaginary part
+        trimmed = re.split(r"(?<!e)([j+-])", x)
+        real_part.append("".join(trimmed[:-4]))
+        imag_part.append("".join(trimmed[-4:-2]))
+
+    # We want to align the lengths of the real and imaginary parts of each complex
+    # number, as well as the lengths the real (resp. complex) parts of all numbers
+    # in the array
+    n = len(str_complexes)
+    padded_parts = _trim_zeros_float(real_part + imag_part, decimal)
+    if len(padded_parts) == 0:
+        return []
+    padded_length = max(len(part) for part in padded_parts) - 1
+    padded = [
+        real_pt  # real part, possibly NaN
+        + imag_pt[0]  # +/-
+        + f"{imag_pt[1:]:>{padded_length}}"  # complex part (no sign), possibly nan
+        + "j"
+        for real_pt, imag_pt in zip(padded_parts[:n], padded_parts[n:])
+    ]
+    return padded
+
+
+def _trim_zeros_single_float(str_float: str) -> str:
+    """
+    Trims trailing zeros after a decimal point,
+    leaving just one if necessary.
+    """
+    str_float = str_float.rstrip("0")
+    if str_float.endswith("."):
+        str_float += "0"
+
+    return str_float
+
+
+def _trim_zeros_float(
+    str_floats: ArrayLike | list[str], decimal: str = "."
+) -> list[str]:
+    """
+    Trims the maximum number of trailing zeros equally from
+    all numbers containing decimals, leaving just one if
+    necessary.
+    """
+    trimmed = str_floats
+    number_regex = re.compile(rf"^\s*[\+-]?[0-9]+\{decimal}[0-9]*$")
+
+    def is_number_with_decimal(x) -> bool:
+        return re.match(number_regex, x) is not None
+
+    def should_trim(values: ArrayLike | list[str]) -> bool:
+        """
+        Determine if an array of strings should be trimmed.
+
+        Returns True if all numbers containing decimals (defined by the
+        above regular expression) within the array end in a zero, otherwise
+        returns False.
+        """
+        numbers = [x for x in values if is_number_with_decimal(x)]
+        return len(numbers) > 0 and all(x.endswith("0") for x in numbers)
+
+    while should_trim(trimmed):
+        trimmed = [x[:-1] if is_number_with_decimal(x) else x for x in trimmed]
+
+    # leave one 0 after the decimal points if need be.
+    result = [
+        x + "0" if is_number_with_decimal(x) and x.endswith(decimal) else x
+        for x in trimmed
+    ]
+    return result
+
+
+def _has_names(index: Index) -> bool:
+    if isinstance(index, MultiIndex):
+        return com.any_not_none(*index.names)
+    else:
+        return index.name is not None
+
+
+class EngFormatter:
+    """
+    Formats float values according to engineering format.
+
+    Based on matplotlib.ticker.EngFormatter
+    """
+
+    # The SI engineering prefixes
+    ENG_PREFIXES = {
+        -24: "y",
+        -21: "z",
+        -18: "a",
+        -15: "f",
+        -12: "p",
+        -9: "n",
+        -6: "u",
+        -3: "m",
+        0: "",
+        3: "k",
+        6: "M",
+        9: "G",
+        12: "T",
+        15: "P",
+        18: "E",
+        21: "Z",
+        24: "Y",
+    }
+
+    def __init__(
+        self, accuracy: int | None = None, use_eng_prefix: bool = False
+    ) -> None:
+        self.accuracy = accuracy
+        self.use_eng_prefix = use_eng_prefix
+
+    def __call__(self, num: float) -> str:
+        """
+        Formats a number in engineering notation, appending a letter
+        representing the power of 1000 of the original number. Some examples:
+        >>> format_eng = EngFormatter(accuracy=0, use_eng_prefix=True)
+        >>> format_eng(0)
+        ' 0'
+        >>> format_eng = EngFormatter(accuracy=1, use_eng_prefix=True)
+        >>> format_eng(1_000_000)
+        ' 1.0M'
+        >>> format_eng = EngFormatter(accuracy=2, use_eng_prefix=False)
+        >>> format_eng("-1e-6")
+        '-1.00E-06'
+
+        @param num: the value to represent
+        @type num: either a numeric value or a string that can be converted to
+                   a numeric value (as per decimal.Decimal constructor)
+
+        @return: engineering formatted string
+        """
+        dnum = Decimal(str(num))
+
+        if Decimal.is_nan(dnum):
+            return "NaN"
+
+        if Decimal.is_infinite(dnum):
+            return "inf"
+
+        sign = 1
+
+        if dnum < 0:  # pragma: no cover
+            sign = -1
+            dnum = -dnum
+
+        if dnum != 0:
+            pow10 = Decimal(int(math.floor(dnum.log10() / 3) * 3))
+        else:
+            pow10 = Decimal(0)
+
+        pow10 = pow10.min(max(self.ENG_PREFIXES.keys()))
+        pow10 = pow10.max(min(self.ENG_PREFIXES.keys()))
+        int_pow10 = int(pow10)
+
+        if self.use_eng_prefix:
+            prefix = self.ENG_PREFIXES[int_pow10]
+        elif int_pow10 < 0:
+            prefix = f"E-{-int_pow10:02d}"
+        else:
+            prefix = f"E+{int_pow10:02d}"
+
+        mant = sign * dnum / (10**pow10)
+
+        if self.accuracy is None:  # pragma: no cover
+            format_str = "{mant: g}{prefix}"
+        else:
+            format_str = f"{{mant: .{self.accuracy:d}f}}{{prefix}}"
+
+        formatted = format_str.format(mant=mant, prefix=prefix)
+
+        return formatted
+
+
+def set_eng_float_format(accuracy: int = 3, use_eng_prefix: bool = False) -> None:
+    """
+    Format float representation in DataFrame with SI notation.
+
+    Sets the floating-point display format for ``DataFrame`` objects using engineering
+    notation (SI units), allowing easier readability of values across wide ranges.
+
+    Parameters
+    ----------
+    accuracy : int, default 3
+        Number of decimal digits after the floating point.
+    use_eng_prefix : bool, default False
+        Whether to represent a value with SI prefixes.
+
+    Returns
+    -------
+    None
+        This method does not return a value. it updates the global display format
+        for floats in DataFrames.
+
+    See Also
+    --------
+    set_option : Set the value of the specified option or options.
+    reset_option : Reset one or more options to their default value.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame([1e-9, 1e-3, 1, 1e3, 1e6])
+    >>> df
+                  0
+    0  1.000000e-09
+    1  1.000000e-03
+    2  1.000000e+00
+    3  1.000000e+03
+    4  1.000000e+06
+
+    >>> pd.set_eng_float_format(accuracy=1)
+    >>> df
+             0
+    0  1.0E-09
+    1  1.0E-03
+    2  1.0E+00
+    3  1.0E+03
+    4  1.0E+06
+
+    >>> pd.set_eng_float_format(use_eng_prefix=True)
+    >>> df
+            0
+    0  1.000n
+    1  1.000m
+    2   1.000
+    3  1.000k
+    4  1.000M
+
+    >>> pd.set_eng_float_format(accuracy=1, use_eng_prefix=True)
+    >>> df
+          0
+    0  1.0n
+    1  1.0m
+    2   1.0
+    3  1.0k
+    4  1.0M
+
+    >>> pd.set_option("display.float_format", None)  # unset option
+    """
+    set_option("display.float_format", EngFormatter(accuracy, use_eng_prefix))
+
+
+def get_level_lengths(
+    levels: Any, sentinel: bool | object | str = ""
+) -> list[dict[int, int]]:
+    """
+    For each index in each level the function returns lengths of indexes.
+
+    Parameters
+    ----------
+    levels : list of lists
+        List of values on for level.
+    sentinel : string, optional
+        Value which states that no new index starts on there.
+
+    Returns
+    -------
+    Returns list of maps. For each level returns map of indexes (key is index
+    in row and value is length of index).
+    """
+    if len(levels) == 0:
+        return []
+
+    control = [True] * len(levels[0])
+
+    result = []
+    for level in levels:
+        last_index = 0
+
+        lengths = {}
+        for i, key in enumerate(level):
+            if control[i] and key == sentinel:
+                pass
+            else:
+                control[i] = False
+                lengths[last_index] = i - last_index
+                last_index = i
+
+        lengths[last_index] = len(level) - last_index
+
+        result.append(lengths)
+
+    return result
+
+
+def buffer_put_lines(buf: WriteBuffer[str], lines: list[str]) -> None:
+    """
+    Appends lines to a buffer.
+
+    Parameters
+    ----------
+    buf
+        The buffer to write to
+    lines
+        The lines to append.
+    """
+    if any(isinstance(x, str) for x in lines):
+        lines = [str(x) for x in lines]
+    buf.write("\n".join(lines))
diff --git a/pandas/pandas/io/formats/html.py b/pandas/pandas/io/formats/html.py
new file mode 100644
index 0000000000000..c4884ef4ce4a9
--- /dev/null
+++ b/pandas/pandas/io/formats/html.py
@@ -0,0 +1,650 @@
+"""
+Module for formatting output data in HTML.
+"""
+
+from __future__ import annotations
+
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Final,
+    cast,
+)
+
+from pandas._config import get_option
+
+from pandas._libs import lib
+
+from pandas import (
+    MultiIndex,
+    option_context,
+)
+
+from pandas.io.common import is_url
+from pandas.io.formats.format import (
+    DataFrameFormatter,
+    get_level_lengths,
+)
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterable,
+        Mapping,
+    )
+
+
+class HTMLFormatter:
+    """
+    Internal class for formatting output data in html.
+    This class is intended for shared functionality between
+    DataFrame.to_html() and DataFrame._repr_html_().
+    Any logic in common with other output formatting methods
+    should ideally be inherited from classes in format.py
+    and this class responsible for only producing html markup.
+    """
+
+    indent_delta: Final = 2
+
+    def __init__(
+        self,
+        formatter: DataFrameFormatter,
+        classes: str | list[str] | tuple[str, ...] | None = None,
+        border: int | bool | None = None,
+        table_id: str | None = None,
+        render_links: bool = False,
+    ) -> None:
+        self.fmt = formatter
+        self.classes = classes
+
+        self.frame = self.fmt.frame
+        self.columns = self.fmt.tr_frame.columns
+        self.elements: list[str] = []
+        self.bold_rows = self.fmt.bold_rows
+        self.escape = self.fmt.escape
+        self.show_dimensions = self.fmt.show_dimensions
+        if border is None or border is True:
+            border = cast(int, get_option("display.html.border"))
+        elif not border:
+            border = None
+
+        self.border = border
+        self.table_id = table_id
+        self.render_links = render_links
+
+        self.col_space = {}
+        is_multi_index = isinstance(self.columns, MultiIndex)
+        for column, value in self.fmt.col_space.items():
+            col_space_value = f"{value}px" if isinstance(value, int) else value
+            self.col_space[column] = col_space_value
+            # GH 53885: Handling case where column is index
+            # Flatten the data in the multi index and add in the map
+            if is_multi_index and isinstance(column, tuple):
+                for column_index in column:
+                    self.col_space[str(column_index)] = col_space_value
+
+    def to_string(self) -> str:
+        lines = self.render()
+        if any(isinstance(x, str) for x in lines):
+            lines = [str(x) for x in lines]
+        return "\n".join(lines)
+
+    def render(self) -> list[str]:
+        self._write_table()
+
+        if self.should_show_dimensions:
+            by = chr(215)  # ×  # noqa: RUF003
+            self.write(
+                f"<p>{len(self.frame)} rows {by} {len(self.frame.columns)} columns</p>"
+            )
+
+        return self.elements
+
+    @property
+    def should_show_dimensions(self) -> bool:
+        return self.fmt.should_show_dimensions
+
+    @property
+    def show_row_idx_names(self) -> bool:
+        return self.fmt.show_row_idx_names
+
+    @property
+    def show_col_idx_names(self) -> bool:
+        return self.fmt.show_col_idx_names
+
+    @property
+    def row_levels(self) -> int:
+        if self.fmt.index:
+            # showing (row) index
+            return self.frame.index.nlevels
+        elif self.show_col_idx_names:
+            # see gh-22579
+            # Column misalignment also occurs for
+            # a standard index when the columns index is named.
+            # If the row index is not displayed a column of
+            # blank cells need to be included before the DataFrame values.
+            return 1
+        # not showing (row) index
+        return 0
+
+    def _get_columns_formatted_values(self) -> Iterable:
+        return self.columns
+
+    @property
+    def is_truncated(self) -> bool:
+        return self.fmt.is_truncated
+
+    @property
+    def ncols(self) -> int:
+        return len(self.fmt.tr_frame.columns)
+
+    def write(self, s: Any, indent: int = 0) -> None:
+        rs = pprint_thing(s)
+        self.elements.append(" " * indent + rs)
+
+    def write_th(
+        self, s: Any, header: bool = False, indent: int = 0, tags: str | None = None
+    ) -> None:
+        """
+        Method for writing a formatted <th> cell.
+
+        If col_space is set on the formatter then that is used for
+        the value of min-width.
+
+        Parameters
+        ----------
+        s : object
+            The data to be written inside the cell.
+        header : bool, default False
+            Set to True if the <th> is for use inside <thead>.  This will
+            cause min-width to be set if there is one.
+        indent : int, default 0
+            The indentation level of the cell.
+        tags : str, default None
+            Tags to include in the cell.
+
+        Returns
+        -------
+        A written <th> cell.
+        """
+        col_space = self.col_space.get(s, None)
+
+        if header and col_space is not None:
+            tags = tags or ""
+            tags += f'style="min-width: {col_space};"'
+
+        self._write_cell(s, kind="th", indent=indent, tags=tags)
+
+    def write_td(self, s: Any, indent: int = 0, tags: str | None = None) -> None:
+        self._write_cell(s, kind="td", indent=indent, tags=tags)
+
+    def _write_cell(
+        self, s: Any, kind: str = "td", indent: int = 0, tags: str | None = None
+    ) -> None:
+        if tags is not None:
+            start_tag = f"<{kind} {tags}>"
+        else:
+            start_tag = f"<{kind}>"
+
+        if self.escape:
+            # escape & first to prevent double escaping of &
+            esc = {"&": r"&amp;", "<": r"&lt;", ">": r"&gt;"}
+        else:
+            esc = {}
+
+        rs = pprint_thing(s, escape_chars=esc).strip()
+        # replace spaces betweens strings with non-breaking spaces
+        rs = rs.replace("  ", "&nbsp;&nbsp;")
+
+        if self.render_links and is_url(rs):
+            rs_unescaped = pprint_thing(s, escape_chars={}).strip()
+            start_tag += f'<a href="{rs_unescaped}" target="_blank">'
+            end_a = "</a>"
+        else:
+            end_a = ""
+
+        self.write(f"{start_tag}{rs}{end_a}</{kind}>", indent)
+
+    def write_tr(
+        self,
+        line: Iterable,
+        indent: int = 0,
+        indent_delta: int = 0,
+        header: bool = False,
+        align: str | None = None,
+        tags: dict[int, str] | None = None,
+        nindex_levels: int = 0,
+    ) -> None:
+        if tags is None:
+            tags = {}
+
+        if align is None:
+            self.write("<tr>", indent)
+        else:
+            self.write(f'<tr style="text-align: {align};">', indent)
+        indent += indent_delta
+
+        for i, s in enumerate(line):
+            val_tag = tags.get(i, None)
+            if header or (self.bold_rows and i < nindex_levels):
+                self.write_th(s, indent=indent, header=header, tags=val_tag)
+            else:
+                self.write_td(s, indent, tags=val_tag)
+
+        indent -= indent_delta
+        self.write("</tr>", indent)
+
+    def _write_table(self, indent: int = 0) -> None:
+        _classes = ["dataframe"]  # Default class.
+        use_mathjax = get_option("display.html.use_mathjax")
+        if not use_mathjax:
+            _classes.append("tex2jax_ignore")
+            _classes.append("mathjax_ignore")
+        if self.classes is not None:
+            if isinstance(self.classes, str):
+                self.classes = self.classes.split()
+            if not isinstance(self.classes, (list, tuple)):
+                raise TypeError(
+                    "classes must be a string, list, "
+                    f"or tuple, not {type(self.classes)}"
+                )
+            _classes.extend(self.classes)
+
+        if self.table_id is None:
+            id_section = ""
+        else:
+            id_section = f' id="{self.table_id}"'
+
+        if self.border is None:
+            border_attr = ""
+        else:
+            border_attr = f' border="{self.border}"'
+
+        self.write(
+            f'<table{border_attr} class="{" ".join(_classes)}"{id_section}>',
+            indent,
+        )
+
+        if self.fmt.header or self.show_row_idx_names:
+            self._write_header(indent + self.indent_delta)
+
+        self._write_body(indent + self.indent_delta)
+
+        self.write("</table>", indent)
+
+    def _write_col_header(self, indent: int) -> None:
+        row: list[Hashable]
+        is_truncated_horizontally = self.fmt.is_truncated_horizontally
+        if isinstance(self.columns, MultiIndex):
+            template = 'colspan="{span:d}" halign="left"'
+
+            sentinel: lib.NoDefault | bool
+            if self.fmt.sparsify:
+                # GH3547
+                sentinel = lib.no_default
+            else:
+                sentinel = False
+            levels = self.columns._format_multi(sparsify=sentinel, include_names=False)
+            level_lengths = get_level_lengths(levels, sentinel)
+            inner_lvl = len(level_lengths) - 1
+            for lnum, (records, values) in enumerate(zip(level_lengths, levels)):
+                if is_truncated_horizontally:
+                    # modify the header lines
+                    ins_col = self.fmt.tr_col_num
+                    if self.fmt.sparsify:
+                        recs_new = {}
+                        # Increment tags after ... col.
+                        for tag, span in list(records.items()):
+                            if tag >= ins_col:
+                                recs_new[tag + 1] = span
+                            elif tag + span > ins_col:
+                                recs_new[tag] = span + 1
+                                if lnum == inner_lvl:
+                                    values = (
+                                        values[:ins_col] + ("...",) + values[ins_col:]
+                                    )
+                                else:
+                                    # sparse col headers do not receive a ...
+                                    values = (
+                                        values[:ins_col]
+                                        + (values[ins_col - 1],)
+                                        + values[ins_col:]
+                                    )
+                            else:
+                                recs_new[tag] = span
+                            # if ins_col lies between tags, all col headers
+                            # get ...
+                            if tag + span == ins_col:
+                                recs_new[ins_col] = 1
+                                values = values[:ins_col] + ("...",) + values[ins_col:]
+                        records = recs_new
+                        inner_lvl = len(level_lengths) - 1
+                        if lnum == inner_lvl:
+                            records[ins_col] = 1
+                    else:
+                        recs_new = {}
+                        for tag, span in list(records.items()):
+                            if tag >= ins_col:
+                                recs_new[tag + 1] = span
+                            else:
+                                recs_new[tag] = span
+                        recs_new[ins_col] = 1
+                        records = recs_new
+                        values = values[:ins_col] + ["..."] + values[ins_col:]
+
+                # see gh-22579
+                # Column Offset Bug with to_html(index=False) with
+                # MultiIndex Columns and Index.
+                # Initially fill row with blank cells before column names.
+                # TODO: Refactor to remove code duplication with code
+                # block below for standard columns index.
+                row = [""] * (self.row_levels - 1)
+                if self.fmt.index or self.show_col_idx_names:
+                    # see gh-22747
+                    # If to_html(index_names=False) do not show columns
+                    # index names.
+                    # TODO: Refactor to use _get_column_name_list from
+                    # DataFrameFormatter class and create a
+                    # _get_formatted_column_labels function for code
+                    # parity with DataFrameFormatter class.
+                    if self.fmt.show_index_names:
+                        name = self.columns.names[lnum]
+                        row.append(pprint_thing(name or ""))
+                    else:
+                        row.append("")
+
+                tags = {}
+                j = len(row)
+                for i, v in enumerate(values):
+                    if i in records:
+                        if records[i] > 1:
+                            tags[j] = template.format(span=records[i])
+                    else:
+                        continue
+                    j += 1
+                    row.append(v)
+                self.write_tr(row, indent, self.indent_delta, tags=tags, header=True)
+        else:
+            # see gh-22579
+            # Column misalignment also occurs for
+            # a standard index when the columns index is named.
+            # Initially fill row with blank cells before column names.
+            # TODO: Refactor to remove code duplication with code block
+            # above for columns MultiIndex.
+            row = [""] * (self.row_levels - 1)
+            if self.fmt.index or self.show_col_idx_names:
+                # see gh-22747
+                # If to_html(index_names=False) do not show columns
+                # index names.
+                # TODO: Refactor to use _get_column_name_list from
+                # DataFrameFormatter class.
+                if self.fmt.show_index_names:
+                    row.append(self.columns.name or "")
+                else:
+                    row.append("")
+            row.extend(self._get_columns_formatted_values())
+            align = self.fmt.justify
+
+            if is_truncated_horizontally:
+                ins_col = self.row_levels + self.fmt.tr_col_num
+                row.insert(ins_col, "...")
+
+            self.write_tr(row, indent, self.indent_delta, header=True, align=align)
+
+    def _write_row_header(self, indent: int) -> None:
+        is_truncated_horizontally = self.fmt.is_truncated_horizontally
+        row = [x if x is not None else "" for x in self.frame.index.names] + [""] * (
+            self.ncols + (1 if is_truncated_horizontally else 0)
+        )
+        self.write_tr(row, indent, self.indent_delta, header=True)
+
+    def _write_header(self, indent: int) -> None:
+        self.write("<thead>", indent)
+
+        if self.fmt.header:
+            self._write_col_header(indent + self.indent_delta)
+
+        if self.show_row_idx_names:
+            self._write_row_header(indent + self.indent_delta)
+
+        self.write("</thead>", indent)
+
+    def _get_formatted_values(self) -> dict[int, list[str]]:
+        with option_context("display.max_colwidth", None):
+            fmt_values = {i: self.fmt.format_col(i) for i in range(self.ncols)}
+        return fmt_values
+
+    def _write_body(self, indent: int) -> None:
+        self.write("<tbody>", indent)
+        fmt_values = self._get_formatted_values()
+
+        # write values
+        if self.fmt.index and isinstance(self.frame.index, MultiIndex):
+            self._write_hierarchical_rows(fmt_values, indent + self.indent_delta)
+        else:
+            self._write_regular_rows(fmt_values, indent + self.indent_delta)
+
+        self.write("</tbody>", indent)
+
+    def _write_regular_rows(
+        self, fmt_values: Mapping[int, list[str]], indent: int
+    ) -> None:
+        is_truncated_horizontally = self.fmt.is_truncated_horizontally
+        is_truncated_vertically = self.fmt.is_truncated_vertically
+
+        nrows = len(self.fmt.tr_frame)
+
+        if self.fmt.index:
+            fmt = self.fmt._get_formatter("__index__")
+            if fmt is not None:
+                index_values = self.fmt.tr_frame.index.map(fmt)
+            else:
+                # only reached with non-Multi index
+                index_values = self.fmt.tr_frame.index._format_flat(include_name=False)
+
+        row: list[str] = []
+        for i in range(nrows):
+            if is_truncated_vertically and i == (self.fmt.tr_row_num):
+                str_sep_row = ["..."] * len(row)
+                self.write_tr(
+                    str_sep_row,
+                    indent,
+                    self.indent_delta,
+                    tags=None,
+                    nindex_levels=self.row_levels,
+                )
+
+            row = []
+            if self.fmt.index:
+                row.append(index_values[i])
+            # see gh-22579
+            # Column misalignment also occurs for
+            # a standard index when the columns index is named.
+            # Add blank cell before data cells.
+            elif self.show_col_idx_names:
+                row.append("")
+            row.extend(fmt_values[j][i] for j in range(self.ncols))
+
+            if is_truncated_horizontally:
+                dot_col_ix = self.fmt.tr_col_num + self.row_levels
+                row.insert(dot_col_ix, "...")
+            self.write_tr(
+                row, indent, self.indent_delta, tags=None, nindex_levels=self.row_levels
+            )
+
+    def _write_hierarchical_rows(
+        self, fmt_values: Mapping[int, list[str]], indent: int
+    ) -> None:
+        template = 'rowspan="{span}" valign="top"'
+
+        is_truncated_horizontally = self.fmt.is_truncated_horizontally
+        is_truncated_vertically = self.fmt.is_truncated_vertically
+        frame = self.fmt.tr_frame
+        nrows = len(frame)
+
+        assert isinstance(frame.index, MultiIndex)
+        idx_values = frame.index._format_multi(sparsify=False, include_names=False)
+        idx_values = list(zip(*idx_values))
+
+        if self.fmt.sparsify:
+            # GH3547
+            sentinel = lib.no_default
+            levels = frame.index._format_multi(sparsify=sentinel, include_names=False)
+
+            level_lengths = get_level_lengths(levels, sentinel)
+            inner_lvl = len(level_lengths) - 1
+            if is_truncated_vertically:
+                # Insert ... row and adjust idx_values and
+                # level_lengths to take this into account.
+                ins_row = self.fmt.tr_row_num
+                inserted = False
+                for lnum, records in enumerate(level_lengths):
+                    rec_new = {}
+                    for tag, span in list(records.items()):
+                        if tag >= ins_row:
+                            rec_new[tag + 1] = span
+                        elif tag + span > ins_row:
+                            rec_new[tag] = span + 1
+
+                            # GH 14882 - Make sure insertion done once
+                            if not inserted:
+                                dot_row = list(idx_values[ins_row - 1])
+                                dot_row[-1] = "..."
+                                idx_values.insert(ins_row, tuple(dot_row))
+                                inserted = True
+                            else:
+                                dot_row = list(idx_values[ins_row])
+                                dot_row[inner_lvl - lnum] = "..."
+                                idx_values[ins_row] = tuple(dot_row)
+                        else:
+                            rec_new[tag] = span
+                        # If ins_row lies between tags, all cols idx cols
+                        # receive ...
+                        if tag + span == ins_row:
+                            rec_new[ins_row] = 1
+                            if lnum == 0:
+                                idx_values.insert(
+                                    ins_row, tuple(["..."] * len(level_lengths))
+                                )
+
+                            # GH 14882 - Place ... in correct level
+                            elif inserted:
+                                dot_row = list(idx_values[ins_row])
+                                dot_row[inner_lvl - lnum] = "..."
+                                idx_values[ins_row] = tuple(dot_row)
+                    level_lengths[lnum] = rec_new
+
+                level_lengths[inner_lvl][ins_row] = 1
+                for ix_col in fmt_values:
+                    fmt_values[ix_col].insert(ins_row, "...")
+                nrows += 1
+
+            for i in range(nrows):
+                row = []
+                tags = {}
+
+                sparse_offset = 0
+                j = 0
+                for records, v in zip(level_lengths, idx_values[i]):
+                    if i in records:
+                        if records[i] > 1:
+                            tags[j] = template.format(span=records[i])
+                    else:
+                        sparse_offset += 1
+                        continue
+
+                    j += 1
+                    row.append(v)
+
+                row.extend(fmt_values[j][i] for j in range(self.ncols))
+                if is_truncated_horizontally:
+                    row.insert(
+                        self.row_levels - sparse_offset + self.fmt.tr_col_num, "..."
+                    )
+                self.write_tr(
+                    row,
+                    indent,
+                    self.indent_delta,
+                    tags=tags,
+                    nindex_levels=len(levels) - sparse_offset,
+                )
+        else:
+            row = []
+            for i in range(len(frame)):
+                if is_truncated_vertically and i == (self.fmt.tr_row_num):
+                    str_sep_row = ["..."] * len(row)
+                    self.write_tr(
+                        str_sep_row,
+                        indent,
+                        self.indent_delta,
+                        tags=None,
+                        nindex_levels=self.row_levels,
+                    )
+
+                idx_values = list(
+                    zip(*frame.index._format_multi(sparsify=False, include_names=False))
+                )
+                row = []
+                row.extend(idx_values[i])
+                row.extend(fmt_values[j][i] for j in range(self.ncols))
+                if is_truncated_horizontally:
+                    row.insert(self.row_levels + self.fmt.tr_col_num, "...")
+                self.write_tr(
+                    row,
+                    indent,
+                    self.indent_delta,
+                    tags=None,
+                    nindex_levels=frame.index.nlevels,
+                )
+
+
+class NotebookFormatter(HTMLFormatter):
+    """
+    Internal class for formatting output data in html for display in Jupyter
+    Notebooks. This class is intended for functionality specific to
+    DataFrame._repr_html_() and DataFrame.to_html(notebook=True)
+    """
+
+    def _get_formatted_values(self) -> dict[int, list[str]]:
+        return {i: self.fmt.format_col(i) for i in range(self.ncols)}
+
+    def _get_columns_formatted_values(self) -> list[str]:
+        # only reached with non-Multi Index
+        return self.columns._format_flat(include_name=False)
+
+    def write_style(self) -> None:
+        # We use the "scoped" attribute here so that the desired
+        # style properties for the data frame are not then applied
+        # throughout the entire notebook.
+        template_first = """\
+            <style scoped>"""
+        template_last = """\
+            </style>"""
+        template_select = """\
+                .dataframe %s {
+                    %s: %s;
+                }"""
+        element_props = [
+            ("tbody tr th:only-of-type", "vertical-align", "middle"),
+            ("tbody tr th", "vertical-align", "top"),
+        ]
+        if isinstance(self.columns, MultiIndex):
+            element_props.append(("thead tr th", "text-align", "left"))
+            if self.show_row_idx_names:
+                element_props.append(
+                    ("thead tr:last-of-type th", "text-align", "right")
+                )
+        else:
+            element_props.append(("thead th", "text-align", "right"))
+        template_mid = "\n\n".join(template_select % t for t in element_props)
+        template = dedent(f"{template_first}\n{template_mid}\n{template_last}")
+        self.write(template)
+
+    def render(self) -> list[str]:
+        self.write("<div>")
+        self.write_style()
+        super().render()
+        self.write("</div>")
+        return self.elements
diff --git a/pandas/pandas/io/formats/info.py b/pandas/pandas/io/formats/info.py
new file mode 100644
index 0000000000000..c9a6e94a0c7c1
--- /dev/null
+++ b/pandas/pandas/io/formats/info.py
@@ -0,0 +1,1106 @@
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+import sys
+from textwrap import dedent
+from typing import TYPE_CHECKING
+
+from pandas._config import get_option
+
+from pandas.io.formats import format as fmt
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Iterable,
+        Iterator,
+        Mapping,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        Dtype,
+        WriteBuffer,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+
+frame_max_cols_sub = dedent(
+    """\
+    max_cols : int, optional
+        When to switch from the verbose to the truncated output. If the
+        DataFrame has more than `max_cols` columns, the truncated output
+        is used. By default, the setting in
+        ``pandas.options.display.max_info_columns`` is used."""
+)
+
+
+show_counts_sub = dedent(
+    """\
+    show_counts : bool, optional
+        Whether to show the non-null counts. By default, this is shown
+        only if the DataFrame is smaller than
+        ``pandas.options.display.max_info_rows`` and
+        ``pandas.options.display.max_info_columns``. A value of True always
+        shows the counts, and False never shows the counts."""
+)
+
+
+frame_examples_sub = dedent(
+    """\
+    >>> int_values = [1, 2, 3, 4, 5]
+    >>> text_values = ['alpha', 'beta', 'gamma', 'delta', 'epsilon']
+    >>> float_values = [0.0, 0.25, 0.5, 0.75, 1.0]
+    >>> df = pd.DataFrame({"int_col": int_values, "text_col": text_values,
+    ...                   "float_col": float_values})
+    >>> df
+        int_col text_col  float_col
+    0        1    alpha       0.00
+    1        2     beta       0.25
+    2        3    gamma       0.50
+    3        4    delta       0.75
+    4        5  epsilon       1.00
+
+    Prints information of all columns:
+
+    >>> df.info(verbose=True)
+    <class 'pandas.DataFrame'>
+    RangeIndex: 5 entries, 0 to 4
+    Data columns (total 3 columns):
+     #   Column     Non-Null Count  Dtype
+    ---  ------     --------------  -----
+     0   int_col    5 non-null      int64
+     1   text_col   5 non-null      object
+     2   float_col  5 non-null      float64
+    dtypes: float64(1), int64(1), object(1)
+    memory usage: 248.0+ bytes
+
+    Prints a summary of columns count and its dtypes but not per column
+    information:
+
+    >>> df.info(verbose=False)
+    <class 'pandas.DataFrame'>
+    RangeIndex: 5 entries, 0 to 4
+    Columns: 3 entries, int_col to float_col
+    dtypes: float64(1), int64(1), object(1)
+    memory usage: 248.0+ bytes
+
+    Pipe output of DataFrame.info to buffer instead of sys.stdout, get
+    buffer content and writes to a text file:
+
+    >>> import io
+    >>> buffer = io.StringIO()
+    >>> df.info(buf=buffer)
+    >>> s = buffer.getvalue()
+    >>> with open("df_info.txt", "w",
+    ...           encoding="utf-8") as f:  # doctest: +SKIP
+    ...     f.write(s)
+    260
+
+    The `memory_usage` parameter allows deep introspection mode, specially
+    useful for big DataFrames and fine-tune memory optimization:
+
+    >>> random_strings_array = np.random.choice(['a', 'b', 'c'], 10 ** 6)
+    >>> df = pd.DataFrame({
+    ...     'column_1': np.random.choice(['a', 'b', 'c'], 10 ** 6),
+    ...     'column_2': np.random.choice(['a', 'b', 'c'], 10 ** 6),
+    ...     'column_3': np.random.choice(['a', 'b', 'c'], 10 ** 6)
+    ... })
+    >>> df.info()
+    <class 'pandas.DataFrame'>
+    RangeIndex: 1000000 entries, 0 to 999999
+    Data columns (total 3 columns):
+     #   Column    Non-Null Count    Dtype
+    ---  ------    --------------    -----
+     0   column_1  1000000 non-null  object
+     1   column_2  1000000 non-null  object
+     2   column_3  1000000 non-null  object
+    dtypes: object(3)
+    memory usage: 22.9+ MB
+
+    >>> df.info(memory_usage='deep')
+    <class 'pandas.DataFrame'>
+    RangeIndex: 1000000 entries, 0 to 999999
+    Data columns (total 3 columns):
+     #   Column    Non-Null Count    Dtype
+    ---  ------    --------------    -----
+     0   column_1  1000000 non-null  object
+     1   column_2  1000000 non-null  object
+     2   column_3  1000000 non-null  object
+    dtypes: object(3)
+    memory usage: 165.9 MB"""
+)
+
+
+frame_see_also_sub = dedent(
+    """\
+    DataFrame.describe: Generate descriptive statistics of DataFrame
+        columns.
+    DataFrame.memory_usage: Memory usage of DataFrame columns."""
+)
+
+
+frame_sub_kwargs = {
+    "klass": "DataFrame",
+    "type_sub": " and columns",
+    "max_cols_sub": frame_max_cols_sub,
+    "show_counts_sub": show_counts_sub,
+    "examples_sub": frame_examples_sub,
+    "see_also_sub": frame_see_also_sub,
+    "version_added_sub": "",
+}
+
+
+series_examples_sub = dedent(
+    """\
+    >>> int_values = [1, 2, 3, 4, 5]
+    >>> text_values = ['alpha', 'beta', 'gamma', 'delta', 'epsilon']
+    >>> s = pd.Series(text_values, index=int_values)
+    >>> s.info()
+    <class 'pandas.Series'>
+    Index: 5 entries, 1 to 5
+    Series name: None
+    Non-Null Count  Dtype
+    --------------  -----
+    5 non-null      object
+    dtypes: object(1)
+    memory usage: 80.0+ bytes
+
+    Prints a summary excluding information about its values:
+
+    >>> s.info(verbose=False)
+    <class 'pandas.Series'>
+    Index: 5 entries, 1 to 5
+    dtypes: object(1)
+    memory usage: 80.0+ bytes
+
+    Pipe output of Series.info to buffer instead of sys.stdout, get
+    buffer content and writes to a text file:
+
+    >>> import io
+    >>> buffer = io.StringIO()
+    >>> s.info(buf=buffer)
+    >>> s = buffer.getvalue()
+    >>> with open("df_info.txt", "w",
+    ...           encoding="utf-8") as f:  # doctest: +SKIP
+    ...     f.write(s)
+    260
+
+    The `memory_usage` parameter allows deep introspection mode, specially
+    useful for big Series and fine-tune memory optimization:
+
+    >>> random_strings_array = np.random.choice(['a', 'b', 'c'], 10 ** 6)
+    >>> s = pd.Series(np.random.choice(['a', 'b', 'c'], 10 ** 6))
+    >>> s.info()
+    <class 'pandas.Series'>
+    RangeIndex: 1000000 entries, 0 to 999999
+    Series name: None
+    Non-Null Count    Dtype
+    --------------    -----
+    1000000 non-null  object
+    dtypes: object(1)
+    memory usage: 7.6+ MB
+
+    >>> s.info(memory_usage='deep')
+    <class 'pandas.Series'>
+    RangeIndex: 1000000 entries, 0 to 999999
+    Series name: None
+    Non-Null Count    Dtype
+    --------------    -----
+    1000000 non-null  object
+    dtypes: object(1)
+    memory usage: 55.3 MB"""
+)
+
+
+series_see_also_sub = dedent(
+    """\
+    Series.describe: Generate descriptive statistics of Series.
+    Series.memory_usage: Memory usage of Series."""
+)
+series_max_cols_sub = dedent(
+    """\
+    max_cols : int, optional
+        Unused, exists only for compatibility with DataFrame.info."""
+)
+
+
+series_sub_kwargs = {
+    "klass": "Series",
+    "type_sub": "",
+    "max_cols_sub": series_max_cols_sub,
+    "show_counts_sub": show_counts_sub,
+    "examples_sub": series_examples_sub,
+    "see_also_sub": series_see_also_sub,
+    "version_added_sub": "\n.. versionadded:: 1.4.0\n",
+}
+
+
+INFO_DOCSTRING = dedent(
+    """
+    Print a concise summary of a {klass}.
+
+    This method prints information about a {klass} including
+    the index dtype{type_sub}, non-null values and memory usage.
+    {version_added_sub}\
+
+    Parameters
+    ----------
+    verbose : bool, optional
+        Whether to print the full summary. By default, the setting in
+        ``pandas.options.display.max_info_columns`` is followed.
+    buf : writable buffer, defaults to sys.stdout
+        Where to send the output. By default, the output is printed to
+        sys.stdout. Pass a writable buffer if you need to further process
+        the output.
+    {max_cols_sub}
+    memory_usage : bool, str, optional
+        Specifies whether total memory usage of the {klass}
+        elements (including the index) should be displayed. By default,
+        this follows the ``pandas.options.display.memory_usage`` setting.
+
+        True always show memory usage. False never shows memory usage.
+        A value of 'deep' is equivalent to "True with deep introspection".
+        Memory usage is shown in human-readable units (base-2
+        representation). Without deep introspection a memory estimation is
+        made based in column dtype and number of rows assuming values
+        consume the same memory amount for corresponding dtypes. With deep
+        memory introspection, a real memory usage calculation is performed
+        at the cost of computational resources. See the
+        :ref:`Frequently Asked Questions <df-memory-usage>` for more
+        details.
+    {show_counts_sub}
+
+    Returns
+    -------
+    None
+        This method prints a summary of a {klass} and returns None.
+
+    See Also
+    --------
+    {see_also_sub}
+
+    Examples
+    --------
+    {examples_sub}
+    """
+)
+
+
+def _put_str(s: str | Dtype, space: int) -> str:
+    """
+    Make string of specified length, padding to the right if necessary.
+
+    Parameters
+    ----------
+    s : Union[str, Dtype]
+        String to be formatted.
+    space : int
+        Length to force string to be of.
+
+    Returns
+    -------
+    str
+        String coerced to given length.
+
+    Examples
+    --------
+    >>> pd.io.formats.info._put_str("panda", 6)
+    'panda '
+    >>> pd.io.formats.info._put_str("panda", 4)
+    'pand'
+    """
+    return str(s)[:space].ljust(space)
+
+
+def _sizeof_fmt(num: float, size_qualifier: str) -> str:
+    """
+    Return size in human readable format.
+
+    Parameters
+    ----------
+    num : int
+        Size in bytes.
+    size_qualifier : str
+        Either empty, or '+' (if lower bound).
+
+    Returns
+    -------
+    str
+        Size in human readable format.
+
+    Examples
+    --------
+    >>> _sizeof_fmt(23028, "")
+    '22.5 KB'
+
+    >>> _sizeof_fmt(23028, "+")
+    '22.5+ KB'
+    """
+    for x in ["bytes", "KB", "MB", "GB", "TB"]:
+        if num < 1024.0:
+            return f"{num:3.1f}{size_qualifier} {x}"
+        num /= 1024.0
+    return f"{num:3.1f}{size_qualifier} PB"
+
+
+def _initialize_memory_usage(
+    memory_usage: bool | str | None = None,
+) -> bool | str:
+    """Get memory usage based on inputs and display options."""
+    if memory_usage is None:
+        memory_usage = get_option("display.memory_usage")
+    return memory_usage
+
+
+class _BaseInfo(ABC):
+    """
+    Base class for DataFrameInfo and SeriesInfo.
+
+    Parameters
+    ----------
+    data : DataFrame or Series
+        Either dataframe or series.
+    memory_usage : bool or str, optional
+        If "deep", introspect the data deeply by interrogating object dtypes
+        for system-level memory consumption, and include it in the returned
+        values.
+    """
+
+    data: DataFrame | Series
+    memory_usage: bool | str
+
+    @property
+    @abstractmethod
+    def dtypes(self) -> Iterable[Dtype]:
+        """
+        Dtypes.
+
+        Returns
+        -------
+        dtypes : sequence
+            Dtype of each of the DataFrame's columns (or one series column).
+        """
+
+    @property
+    @abstractmethod
+    def dtype_counts(self) -> Mapping[str, int]:
+        """Mapping dtype - number of counts."""
+
+    @property
+    @abstractmethod
+    def non_null_counts(self) -> list[int] | Series:
+        """Sequence of non-null counts for all columns or column (if series)."""
+
+    @property
+    @abstractmethod
+    def memory_usage_bytes(self) -> int:
+        """
+        Memory usage in bytes.
+
+        Returns
+        -------
+        memory_usage_bytes : int
+            Object's total memory usage in bytes.
+        """
+
+    @property
+    def memory_usage_string(self) -> str:
+        """Memory usage in a form of human readable string."""
+        return f"{_sizeof_fmt(self.memory_usage_bytes, self.size_qualifier)}\n"
+
+    @property
+    def size_qualifier(self) -> str:
+        size_qualifier = ""
+        if self.memory_usage:
+            if self.memory_usage != "deep":
+                # size_qualifier is just a best effort; not guaranteed to catch
+                # all cases (e.g., it misses categorical data even with object
+                # categories)
+                if (
+                    "object" in self.dtype_counts
+                    or self.data.index._is_memory_usage_qualified
+                ):
+                    size_qualifier = "+"
+        return size_qualifier
+
+    @abstractmethod
+    def render(
+        self,
+        *,
+        buf: WriteBuffer[str] | None,
+        max_cols: int | None,
+        verbose: bool | None,
+        show_counts: bool | None,
+    ) -> None:
+        pass
+
+
+class DataFrameInfo(_BaseInfo):
+    """
+    Class storing dataframe-specific info.
+    """
+
+    def __init__(
+        self,
+        data: DataFrame,
+        memory_usage: bool | str | None = None,
+    ) -> None:
+        self.data: DataFrame = data
+        self.memory_usage = _initialize_memory_usage(memory_usage)
+
+    @property
+    def dtype_counts(self) -> Mapping[str, int]:
+        return _get_dataframe_dtype_counts(self.data)
+
+    @property
+    def dtypes(self) -> Iterable[Dtype]:
+        """
+        Dtypes.
+
+        Returns
+        -------
+        dtypes
+            Dtype of each of the DataFrame's columns.
+        """
+        return self.data.dtypes
+
+    @property
+    def ids(self) -> Index:
+        """
+        Column names.
+
+        Returns
+        -------
+        ids : Index
+            DataFrame's column names.
+        """
+        return self.data.columns
+
+    @property
+    def col_count(self) -> int:
+        """Number of columns to be summarized."""
+        return len(self.ids)
+
+    @property
+    def non_null_counts(self) -> Series:
+        """Sequence of non-null counts for all columns or column (if series)."""
+        return self.data.count()
+
+    @property
+    def memory_usage_bytes(self) -> int:
+        deep = self.memory_usage == "deep"
+        return self.data.memory_usage(index=True, deep=deep).sum()
+
+    def render(
+        self,
+        *,
+        buf: WriteBuffer[str] | None,
+        max_cols: int | None,
+        verbose: bool | None,
+        show_counts: bool | None,
+    ) -> None:
+        printer = _DataFrameInfoPrinter(
+            info=self,
+            max_cols=max_cols,
+            verbose=verbose,
+            show_counts=show_counts,
+        )
+        printer.to_buffer(buf)
+
+
+class SeriesInfo(_BaseInfo):
+    """
+    Class storing series-specific info.
+    """
+
+    def __init__(
+        self,
+        data: Series,
+        memory_usage: bool | str | None = None,
+    ) -> None:
+        self.data: Series = data
+        self.memory_usage = _initialize_memory_usage(memory_usage)
+
+    def render(
+        self,
+        *,
+        buf: WriteBuffer[str] | None = None,
+        max_cols: int | None = None,
+        verbose: bool | None = None,
+        show_counts: bool | None = None,
+    ) -> None:
+        if max_cols is not None:
+            raise ValueError(
+                "Argument `max_cols` can only be passed "
+                "in DataFrame.info, not Series.info"
+            )
+        printer = _SeriesInfoPrinter(
+            info=self,
+            verbose=verbose,
+            show_counts=show_counts,
+        )
+        printer.to_buffer(buf)
+
+    @property
+    def non_null_counts(self) -> list[int]:
+        return [self.data.count()]
+
+    @property
+    def dtypes(self) -> Iterable[Dtype]:
+        return [self.data.dtypes]
+
+    @property
+    def dtype_counts(self) -> Mapping[str, int]:
+        from pandas.core.frame import DataFrame
+
+        return _get_dataframe_dtype_counts(DataFrame(self.data))
+
+    @property
+    def memory_usage_bytes(self) -> int:
+        """Memory usage in bytes.
+
+        Returns
+        -------
+        memory_usage_bytes : int
+            Object's total memory usage in bytes.
+        """
+        deep = self.memory_usage == "deep"
+        return self.data.memory_usage(index=True, deep=deep)
+
+
+class _InfoPrinterAbstract:
+    """
+    Class for printing dataframe or series info.
+    """
+
+    def to_buffer(self, buf: WriteBuffer[str] | None = None) -> None:
+        """Save dataframe info into buffer."""
+        table_builder = self._create_table_builder()
+        lines = table_builder.get_lines()
+        if buf is None:  # pragma: no cover
+            buf = sys.stdout
+        fmt.buffer_put_lines(buf, lines)
+
+    @abstractmethod
+    def _create_table_builder(self) -> _TableBuilderAbstract:
+        """Create instance of table builder."""
+
+
+class _DataFrameInfoPrinter(_InfoPrinterAbstract):
+    """
+    Class for printing dataframe info.
+
+    Parameters
+    ----------
+    info : DataFrameInfo
+        Instance of DataFrameInfo.
+    max_cols : int, optional
+        When to switch from the verbose to the truncated output.
+    verbose : bool, optional
+        Whether to print the full summary.
+    show_counts : bool, optional
+        Whether to show the non-null counts.
+    """
+
+    def __init__(
+        self,
+        info: DataFrameInfo,
+        max_cols: int | None = None,
+        verbose: bool | None = None,
+        show_counts: bool | None = None,
+    ) -> None:
+        self.info = info
+        self.data = info.data
+        self.verbose = verbose
+        self.max_cols = self._initialize_max_cols(max_cols)
+        self.show_counts = self._initialize_show_counts(show_counts)
+
+    @property
+    def max_rows(self) -> int:
+        """Maximum info rows to be displayed."""
+        return get_option("display.max_info_rows")
+
+    @property
+    def exceeds_info_cols(self) -> bool:
+        """Check if number of columns to be summarized does not exceed maximum."""
+        return bool(self.col_count > self.max_cols)
+
+    @property
+    def exceeds_info_rows(self) -> bool:
+        """Check if number of rows to be summarized does not exceed maximum."""
+        return bool(len(self.data) > self.max_rows)
+
+    @property
+    def col_count(self) -> int:
+        """Number of columns to be summarized."""
+        return self.info.col_count
+
+    def _initialize_max_cols(self, max_cols: int | None) -> int:
+        if max_cols is None:
+            return get_option("display.max_info_columns")
+        return max_cols
+
+    def _initialize_show_counts(self, show_counts: bool | None) -> bool:
+        if show_counts is None:
+            return bool(not self.exceeds_info_cols and not self.exceeds_info_rows)
+        else:
+            return show_counts
+
+    def _create_table_builder(self) -> _DataFrameTableBuilder:
+        """
+        Create instance of table builder based on verbosity and display settings.
+        """
+        if self.verbose:
+            return _DataFrameTableBuilderVerbose(
+                info=self.info,
+                with_counts=self.show_counts,
+            )
+        elif self.verbose is False:  # specifically set to False, not necessarily None
+            return _DataFrameTableBuilderNonVerbose(info=self.info)
+        elif self.exceeds_info_cols:
+            return _DataFrameTableBuilderNonVerbose(info=self.info)
+        else:
+            return _DataFrameTableBuilderVerbose(
+                info=self.info,
+                with_counts=self.show_counts,
+            )
+
+
+class _SeriesInfoPrinter(_InfoPrinterAbstract):
+    """Class for printing series info.
+
+    Parameters
+    ----------
+    info : SeriesInfo
+        Instance of SeriesInfo.
+    verbose : bool, optional
+        Whether to print the full summary.
+    show_counts : bool, optional
+        Whether to show the non-null counts.
+    """
+
+    def __init__(
+        self,
+        info: SeriesInfo,
+        verbose: bool | None = None,
+        show_counts: bool | None = None,
+    ) -> None:
+        self.info = info
+        self.data = info.data
+        self.verbose = verbose
+        self.show_counts = self._initialize_show_counts(show_counts)
+
+    def _create_table_builder(self) -> _SeriesTableBuilder:
+        """
+        Create instance of table builder based on verbosity.
+        """
+        if self.verbose or self.verbose is None:
+            return _SeriesTableBuilderVerbose(
+                info=self.info,
+                with_counts=self.show_counts,
+            )
+        else:
+            return _SeriesTableBuilderNonVerbose(info=self.info)
+
+    def _initialize_show_counts(self, show_counts: bool | None) -> bool:
+        if show_counts is None:
+            return True
+        else:
+            return show_counts
+
+
+class _TableBuilderAbstract(ABC):
+    """
+    Abstract builder for info table.
+    """
+
+    _lines: list[str]
+    info: _BaseInfo
+
+    @abstractmethod
+    def get_lines(self) -> list[str]:
+        """Product in a form of list of lines (strings)."""
+
+    @property
+    def data(self) -> DataFrame | Series:
+        return self.info.data
+
+    @property
+    def dtypes(self) -> Iterable[Dtype]:
+        """Dtypes of each of the DataFrame's columns."""
+        return self.info.dtypes
+
+    @property
+    def dtype_counts(self) -> Mapping[str, int]:
+        """Mapping dtype - number of counts."""
+        return self.info.dtype_counts
+
+    @property
+    def display_memory_usage(self) -> bool:
+        """Whether to display memory usage."""
+        return bool(self.info.memory_usage)
+
+    @property
+    def memory_usage_string(self) -> str:
+        """Memory usage string with proper size qualifier."""
+        return self.info.memory_usage_string
+
+    @property
+    def non_null_counts(self) -> list[int] | Series:
+        return self.info.non_null_counts
+
+    def add_object_type_line(self) -> None:
+        """Add line with string representation of dataframe to the table."""
+        self._lines.append(str(type(self.data)))
+
+    def add_index_range_line(self) -> None:
+        """Add line with range of indices to the table."""
+        self._lines.append(self.data.index._summary())
+
+    def add_dtypes_line(self) -> None:
+        """Add summary line with dtypes present in dataframe."""
+        collected_dtypes = [
+            f"{key}({val:d})" for key, val in sorted(self.dtype_counts.items())
+        ]
+        self._lines.append(f"dtypes: {', '.join(collected_dtypes)}")
+
+
+class _DataFrameTableBuilder(_TableBuilderAbstract):
+    """
+    Abstract builder for dataframe info table.
+
+    Parameters
+    ----------
+    info : DataFrameInfo.
+        Instance of DataFrameInfo.
+    """
+
+    def __init__(self, *, info: DataFrameInfo) -> None:
+        self.info: DataFrameInfo = info
+
+    def get_lines(self) -> list[str]:
+        self._lines = []
+        if self.col_count == 0:
+            self._fill_empty_info()
+        else:
+            self._fill_non_empty_info()
+        return self._lines
+
+    def _fill_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to empty dataframe."""
+        self.add_object_type_line()
+        self.add_index_range_line()
+        self._lines.append(f"Empty {type(self.data).__name__}\n")
+
+    @abstractmethod
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty dataframe."""
+
+    @property
+    def data(self) -> DataFrame:
+        """DataFrame."""
+        return self.info.data
+
+    @property
+    def ids(self) -> Index:
+        """Dataframe columns."""
+        return self.info.ids
+
+    @property
+    def col_count(self) -> int:
+        """Number of dataframe columns to be summarized."""
+        return self.info.col_count
+
+    def add_memory_usage_line(self) -> None:
+        """Add line containing memory usage."""
+        self._lines.append(f"memory usage: {self.memory_usage_string}")
+
+
+class _DataFrameTableBuilderNonVerbose(_DataFrameTableBuilder):
+    """
+    Dataframe info table builder for non-verbose output.
+    """
+
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty dataframe."""
+        self.add_object_type_line()
+        self.add_index_range_line()
+        self.add_columns_summary_line()
+        self.add_dtypes_line()
+        if self.display_memory_usage:
+            self.add_memory_usage_line()
+
+    def add_columns_summary_line(self) -> None:
+        self._lines.append(self.ids._summary(name="Columns"))
+
+
+class _TableBuilderVerboseMixin(_TableBuilderAbstract):
+    """
+    Mixin for verbose info output.
+    """
+
+    SPACING: str = " " * 2
+    strrows: Sequence[Sequence[str]]
+    gross_column_widths: Sequence[int]
+    with_counts: bool
+
+    @property
+    @abstractmethod
+    def headers(self) -> Sequence[str]:
+        """Headers names of the columns in verbose table."""
+
+    @property
+    def header_column_widths(self) -> Sequence[int]:
+        """Widths of header columns (only titles)."""
+        return [len(col) for col in self.headers]
+
+    def _get_gross_column_widths(self) -> Sequence[int]:
+        """Get widths of columns containing both headers and actual content."""
+        body_column_widths = self._get_body_column_widths()
+        return [
+            max(*widths)
+            for widths in zip(self.header_column_widths, body_column_widths)
+        ]
+
+    def _get_body_column_widths(self) -> Sequence[int]:
+        """Get widths of table content columns."""
+        strcols: Sequence[Sequence[str]] = list(zip(*self.strrows))
+        return [max(len(x) for x in col) for col in strcols]
+
+    def _gen_rows(self) -> Iterator[Sequence[str]]:
+        """
+        Generator function yielding rows content.
+
+        Each element represents a row comprising a sequence of strings.
+        """
+        if self.with_counts:
+            return self._gen_rows_with_counts()
+        else:
+            return self._gen_rows_without_counts()
+
+    @abstractmethod
+    def _gen_rows_with_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data with counts."""
+
+    @abstractmethod
+    def _gen_rows_without_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data without counts."""
+
+    def add_header_line(self) -> None:
+        header_line = self.SPACING.join(
+            [
+                _put_str(header, col_width)
+                for header, col_width in zip(self.headers, self.gross_column_widths)
+            ]
+        )
+        self._lines.append(header_line)
+
+    def add_separator_line(self) -> None:
+        separator_line = self.SPACING.join(
+            [
+                _put_str("-" * header_colwidth, gross_colwidth)
+                for header_colwidth, gross_colwidth in zip(
+                    self.header_column_widths, self.gross_column_widths
+                )
+            ]
+        )
+        self._lines.append(separator_line)
+
+    def add_body_lines(self) -> None:
+        for row in self.strrows:
+            body_line = self.SPACING.join(
+                [
+                    _put_str(col, gross_colwidth)
+                    for col, gross_colwidth in zip(row, self.gross_column_widths)
+                ]
+            )
+            self._lines.append(body_line)
+
+    def _gen_non_null_counts(self) -> Iterator[str]:
+        """Iterator with string representation of non-null counts."""
+        for count in self.non_null_counts:
+            yield f"{count} non-null"
+
+    def _gen_dtypes(self) -> Iterator[str]:
+        """Iterator with string representation of column dtypes."""
+        for dtype in self.dtypes:
+            yield pprint_thing(dtype)
+
+
+class _DataFrameTableBuilderVerbose(_DataFrameTableBuilder, _TableBuilderVerboseMixin):
+    """
+    Dataframe info table builder for verbose output.
+    """
+
+    def __init__(
+        self,
+        *,
+        info: DataFrameInfo,
+        with_counts: bool,
+    ) -> None:
+        self.info = info
+        self.with_counts = with_counts
+        self.strrows: Sequence[Sequence[str]] = list(self._gen_rows())
+        self.gross_column_widths: Sequence[int] = self._get_gross_column_widths()
+
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty dataframe."""
+        self.add_object_type_line()
+        self.add_index_range_line()
+        self.add_columns_summary_line()
+        self.add_header_line()
+        self.add_separator_line()
+        self.add_body_lines()
+        self.add_dtypes_line()
+        if self.display_memory_usage:
+            self.add_memory_usage_line()
+
+    @property
+    def headers(self) -> Sequence[str]:
+        """Headers names of the columns in verbose table."""
+        if self.with_counts:
+            return [" # ", "Column", "Non-Null Count", "Dtype"]
+        return [" # ", "Column", "Dtype"]
+
+    def add_columns_summary_line(self) -> None:
+        self._lines.append(f"Data columns (total {self.col_count} columns):")
+
+    def _gen_rows_without_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data without counts."""
+        yield from zip(
+            self._gen_line_numbers(),
+            self._gen_columns(),
+            self._gen_dtypes(),
+        )
+
+    def _gen_rows_with_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data with counts."""
+        yield from zip(
+            self._gen_line_numbers(),
+            self._gen_columns(),
+            self._gen_non_null_counts(),
+            self._gen_dtypes(),
+        )
+
+    def _gen_line_numbers(self) -> Iterator[str]:
+        """Iterator with string representation of column numbers."""
+        for i, _ in enumerate(self.ids):
+            yield f" {i}"
+
+    def _gen_columns(self) -> Iterator[str]:
+        """Iterator with string representation of column names."""
+        for col in self.ids:
+            yield pprint_thing(col)
+
+
+class _SeriesTableBuilder(_TableBuilderAbstract):
+    """
+    Abstract builder for series info table.
+
+    Parameters
+    ----------
+    info : SeriesInfo.
+        Instance of SeriesInfo.
+    """
+
+    def __init__(self, *, info: SeriesInfo) -> None:
+        self.info: SeriesInfo = info
+
+    def get_lines(self) -> list[str]:
+        self._lines = []
+        self._fill_non_empty_info()
+        return self._lines
+
+    @property
+    def data(self) -> Series:
+        """Series."""
+        return self.info.data
+
+    def add_memory_usage_line(self) -> None:
+        """Add line containing memory usage."""
+        self._lines.append(f"memory usage: {self.memory_usage_string}")
+
+    @abstractmethod
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty series."""
+
+
+class _SeriesTableBuilderNonVerbose(_SeriesTableBuilder):
+    """
+    Series info table builder for non-verbose output.
+    """
+
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty series."""
+        self.add_object_type_line()
+        self.add_index_range_line()
+        self.add_dtypes_line()
+        if self.display_memory_usage:
+            self.add_memory_usage_line()
+
+
+class _SeriesTableBuilderVerbose(_SeriesTableBuilder, _TableBuilderVerboseMixin):
+    """
+    Series info table builder for verbose output.
+    """
+
+    def __init__(
+        self,
+        *,
+        info: SeriesInfo,
+        with_counts: bool,
+    ) -> None:
+        self.info = info
+        self.with_counts = with_counts
+        self.strrows: Sequence[Sequence[str]] = list(self._gen_rows())
+        self.gross_column_widths: Sequence[int] = self._get_gross_column_widths()
+
+    def _fill_non_empty_info(self) -> None:
+        """Add lines to the info table, pertaining to non-empty series."""
+        self.add_object_type_line()
+        self.add_index_range_line()
+        self.add_series_name_line()
+        self.add_header_line()
+        self.add_separator_line()
+        self.add_body_lines()
+        self.add_dtypes_line()
+        if self.display_memory_usage:
+            self.add_memory_usage_line()
+
+    def add_series_name_line(self) -> None:
+        self._lines.append(f"Series name: {self.data.name}")
+
+    @property
+    def headers(self) -> Sequence[str]:
+        """Headers names of the columns in verbose table."""
+        if self.with_counts:
+            return ["Non-Null Count", "Dtype"]
+        return ["Dtype"]
+
+    def _gen_rows_without_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data without counts."""
+        yield from self._gen_dtypes()
+
+    def _gen_rows_with_counts(self) -> Iterator[Sequence[str]]:
+        """Iterator with string representation of body data with counts."""
+        yield from zip(
+            self._gen_non_null_counts(),
+            self._gen_dtypes(),
+        )
+
+
+def _get_dataframe_dtype_counts(df: DataFrame) -> Mapping[str, int]:
+    """
+    Create mapping between datatypes and their number of occurrences.
+    """
+    # groupby dtype.name to collect e.g. Categorical columns
+    return df.dtypes.value_counts().groupby(lambda x: x.name).sum()
diff --git a/pandas/pandas/io/formats/printing.py b/pandas/pandas/io/formats/printing.py
new file mode 100644
index 0000000000000..ab27321ffe83c
--- /dev/null
+++ b/pandas/pandas/io/formats/printing.py
@@ -0,0 +1,585 @@
+"""
+Printing tools.
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Callable,
+    Iterable,
+    Mapping,
+    Sequence,
+)
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    TypeVar,
+    Union,
+)
+from unicodedata import east_asian_width
+
+from pandas._config import get_option
+
+from pandas.core.dtypes.inference import is_sequence
+
+from pandas.io.formats.console import get_console_size
+
+if TYPE_CHECKING:
+    from pandas._typing import ListLike
+EscapeChars = Union[Mapping[str, str], Iterable[str]]
+_KT = TypeVar("_KT")
+_VT = TypeVar("_VT")
+
+
+def adjoin(space: int, *lists: list[str], **kwargs: Any) -> str:
+    """
+    Glues together two sets of strings using the amount of space requested.
+    The idea is to prettify.
+
+    ----------
+    space : int
+        number of spaces for padding
+    lists : str
+        list of str which being joined
+    strlen : callable
+        function used to calculate the length of each str. Needed for unicode
+        handling.
+    justfunc : callable
+        function used to justify str. Needed for unicode handling.
+    """
+    strlen = kwargs.pop("strlen", len)
+    justfunc = kwargs.pop("justfunc", _adj_justify)
+
+    newLists = []
+    lengths = [max(map(strlen, x)) + space for x in lists[:-1]]
+    # not the last one
+    lengths.append(max(map(len, lists[-1])))
+    maxLen = max(map(len, lists))
+    for i, lst in enumerate(lists):
+        nl = justfunc(lst, lengths[i], mode="left")
+        nl = ([" " * lengths[i]] * (maxLen - len(lst))) + nl
+        newLists.append(nl)
+    toJoin = zip(*newLists)
+    return "\n".join("".join(lines) for lines in toJoin)
+
+
+def _adj_justify(texts: Iterable[str], max_len: int, mode: str = "right") -> list[str]:
+    """
+    Perform ljust, center, rjust against string or list-like
+    """
+    if mode == "left":
+        return [x.ljust(max_len) for x in texts]
+    elif mode == "center":
+        return [x.center(max_len) for x in texts]
+    else:
+        return [x.rjust(max_len) for x in texts]
+
+
+# Unicode consolidation
+# ---------------------
+#
+# pprinting utility functions for generating Unicode text or
+# bytes(3.x)/str(2.x) representations of objects.
+# Try to use these as much as possible rather than rolling your own.
+#
+# When to use
+# -----------
+#
+# 1) If you're writing code internal to pandas (no I/O directly involved),
+#    use pprint_thing().
+#
+#    It will always return unicode text which can handled by other
+#    parts of the package without breakage.
+#
+# 2) if you need to write something out to file, use
+#    pprint_thing_encoded(encoding).
+#
+#    If no encoding is specified, it defaults to utf-8. Since encoding pure
+#    ascii with utf-8 is a no-op you can safely use the default utf-8 if you're
+#    working with straight ascii.
+
+
+def _pprint_seq(
+    seq: ListLike, _nest_lvl: int = 0, max_seq_items: int | None = None, **kwds: Any
+) -> str:
+    """
+    internal. pprinter for iterables. you should probably use pprint_thing()
+    rather than calling this directly.
+
+    bounds length of printed sequence, depending on options
+    """
+    if isinstance(seq, set):
+        fmt = "{{{body}}}"
+    elif isinstance(seq, frozenset):
+        fmt = "frozenset({{{body}}})"
+    else:
+        fmt = "[{body}]" if hasattr(seq, "__setitem__") else "({body})"
+
+    if max_seq_items is False:
+        max_items = None
+    else:
+        max_items = max_seq_items or get_option("max_seq_items") or len(seq)
+
+    s = iter(seq)
+    # handle sets, no slicing
+    r = []
+    max_items_reached = False
+    for i, item in enumerate(s):
+        if (max_items is not None) and (i >= max_items):
+            max_items_reached = True
+            break
+        r.append(pprint_thing(item, _nest_lvl + 1, max_seq_items=max_seq_items, **kwds))
+    body = ", ".join(r)
+
+    if max_items_reached:
+        body += ", ..."
+    elif isinstance(seq, tuple) and len(seq) == 1:
+        body += ","
+
+    return fmt.format(body=body)
+
+
+def _pprint_dict(
+    seq: Mapping, _nest_lvl: int = 0, max_seq_items: int | None = None, **kwds: Any
+) -> str:
+    """
+    internal. pprinter for iterables. you should probably use pprint_thing()
+    rather than calling this directly.
+    """
+    fmt = "{{{things}}}"
+    pairs = []
+
+    pfmt = "{key}: {val}"
+
+    if max_seq_items is False:
+        nitems = len(seq)
+    else:
+        nitems = max_seq_items or get_option("max_seq_items") or len(seq)
+
+    for k, v in list(seq.items())[:nitems]:
+        pairs.append(
+            pfmt.format(
+                key=pprint_thing(k, _nest_lvl + 1, max_seq_items=max_seq_items, **kwds),
+                val=pprint_thing(v, _nest_lvl + 1, max_seq_items=max_seq_items, **kwds),
+            )
+        )
+
+    if nitems < len(seq):
+        return fmt.format(things=", ".join(pairs) + ", ...")
+    else:
+        return fmt.format(things=", ".join(pairs))
+
+
+def pprint_thing(
+    thing: object,
+    _nest_lvl: int = 0,
+    escape_chars: EscapeChars | None = None,
+    default_escapes: bool = False,
+    quote_strings: bool = False,
+    max_seq_items: int | None = None,
+) -> str:
+    """
+    This function is the sanctioned way of converting objects
+    to a string representation and properly handles nested sequences.
+
+    Parameters
+    ----------
+    thing : anything to be formatted
+    _nest_lvl : internal use only. pprint_thing() is mutually-recursive
+        with pprint_sequence, this argument is used to keep track of the
+        current nesting level, and limit it.
+    escape_chars : list[str] or Mapping[str, str], optional
+        Characters to escape. If a Mapping is passed the values are the
+        replacements
+    default_escapes : bool, default False
+        Whether the input escape characters replaces or adds to the defaults
+    max_seq_items : int or None, default None
+        Pass through to other pretty printers to limit sequence printing
+
+    Returns
+    -------
+    str
+    """
+
+    def as_escaped_string(
+        thing: Any, escape_chars: EscapeChars | None = escape_chars
+    ) -> str:
+        translate = {"\t": r"\t", "\n": r"\n", "\r": r"\r", "'": r"\'"}
+        if isinstance(escape_chars, Mapping):
+            if default_escapes:
+                translate.update(escape_chars)
+            else:
+                translate = escape_chars  # type: ignore[assignment]
+            escape_chars = list(escape_chars.keys())
+        else:
+            escape_chars = escape_chars or ()
+
+        result = str(thing)
+        for c in escape_chars:
+            result = result.replace(c, translate[c])
+        return result
+
+    if hasattr(thing, "__next__"):
+        return str(thing)
+    elif isinstance(thing, Mapping) and _nest_lvl < get_option(
+        "display.pprint_nest_depth"
+    ):
+        result = _pprint_dict(
+            thing, _nest_lvl, quote_strings=True, max_seq_items=max_seq_items
+        )
+    elif is_sequence(thing) and _nest_lvl < get_option("display.pprint_nest_depth"):
+        result = _pprint_seq(
+            # error: Argument 1 to "_pprint_seq" has incompatible type "object";
+            # expected "ExtensionArray | ndarray[Any, Any] | Index | Series |
+            # SequenceNotStr[Any] | range"
+            thing,  # type: ignore[arg-type]
+            _nest_lvl,
+            escape_chars=escape_chars,
+            quote_strings=quote_strings,
+            max_seq_items=max_seq_items,
+        )
+    elif isinstance(thing, str) and quote_strings:
+        result = f"'{as_escaped_string(thing)}'"
+    else:
+        result = as_escaped_string(thing)
+
+    return result
+
+
+def pprint_thing_encoded(
+    object: object, encoding: str = "utf-8", errors: str = "replace"
+) -> bytes:
+    value = pprint_thing(object)  # get unicode representation of object
+    return value.encode(encoding, errors)
+
+
+def enable_data_resource_formatter(enable: bool) -> None:
+    if "IPython" not in sys.modules:
+        # definitely not in IPython
+        return
+    from IPython import get_ipython
+
+    # error: Call to untyped function "get_ipython" in typed context
+    ip = get_ipython()  # type: ignore[no-untyped-call]
+    if ip is None:
+        # still not in IPython
+        return
+
+    formatters = ip.display_formatter.formatters
+    mimetype = "application/vnd.dataresource+json"
+
+    if enable:
+        if mimetype not in formatters:
+            # define tableschema formatter
+            from IPython.core.formatters import BaseFormatter
+            from traitlets import ObjectName
+
+            class TableSchemaFormatter(BaseFormatter):
+                print_method = ObjectName("_repr_data_resource_")
+                _return_type = (dict,)
+
+            # register it:
+            formatters[mimetype] = TableSchemaFormatter()
+        # enable it if it's been disabled:
+        formatters[mimetype].enabled = True
+    # unregister tableschema mime-type
+    elif mimetype in formatters:
+        formatters[mimetype].enabled = False
+
+
+def default_pprint(thing: Any, max_seq_items: int | None = None) -> str:
+    return pprint_thing(
+        thing,
+        escape_chars=("\t", "\r", "\n"),
+        quote_strings=True,
+        max_seq_items=max_seq_items,
+    )
+
+
+def format_object_summary(
+    obj: ListLike,
+    formatter: Callable,
+    is_justify: bool = True,
+    name: str | None = None,
+    indent_for_name: bool = True,
+    line_break_each_value: bool = False,
+) -> str:
+    """
+    Return the formatted obj as a unicode string
+
+    Parameters
+    ----------
+    obj : object
+        must be iterable and support __getitem__
+    formatter : callable
+        string formatter for an element
+    is_justify : bool
+        should justify the display
+    name : name, optional
+        defaults to the class name of the obj
+    indent_for_name : bool, default True
+        Whether subsequent lines should be indented to
+        align with the name.
+    line_break_each_value : bool, default False
+        If True, inserts a line break for each value of ``obj``.
+        If False, only break lines when the a line of values gets wider
+        than the display width.
+
+    Returns
+    -------
+    summary string
+    """
+    display_width, _ = get_console_size()
+    if display_width is None:
+        display_width = get_option("display.width") or 80
+    if name is None:
+        name = type(obj).__name__
+
+    if indent_for_name:
+        name_len = len(name)
+        space1 = f"\n{(' ' * (name_len + 1))}"
+        space2 = f"\n{(' ' * (name_len + 2))}"
+    else:
+        space1 = "\n"
+        space2 = "\n "  # space for the opening '['
+
+    n = len(obj)
+    if line_break_each_value:
+        # If we want to vertically align on each value of obj, we need to
+        # separate values by a line break and indent the values
+        sep = ",\n " + " " * len(name)
+    else:
+        sep = ","
+    max_seq_items = get_option("display.max_seq_items") or n
+
+    # are we a truncated display
+    is_truncated = n > max_seq_items
+
+    # adj can optionally handle unicode eastern asian width
+    adj = get_adjustment()
+
+    def _extend_line(
+        s: str, line: str, value: str, display_width: int, next_line_prefix: str
+    ) -> tuple[str, str]:
+        if adj.len(line.rstrip()) + adj.len(value.rstrip()) >= display_width:
+            s += line.rstrip()
+            line = next_line_prefix
+        line += value
+        return s, line
+
+    def best_len(values: list[str]) -> int:
+        if values:
+            return max(adj.len(x) for x in values)
+        else:
+            return 0
+
+    close = ", "
+
+    if n == 0:
+        summary = f"[]{close}"
+    elif n == 1 and not line_break_each_value:
+        first = formatter(obj[0])
+        summary = f"[{first}]{close}"
+    elif n == 2 and not line_break_each_value:
+        first = formatter(obj[0])
+        last = formatter(obj[-1])
+        summary = f"[{first}, {last}]{close}"
+    else:
+        if max_seq_items == 1:
+            # If max_seq_items=1 show only last element
+            head = []
+            tail = [formatter(x) for x in obj[-1:]]
+        elif n > max_seq_items:
+            n = min(max_seq_items // 2, 10)
+            head = [formatter(x) for x in obj[:n]]
+            tail = [formatter(x) for x in obj[-n:]]
+        else:
+            head = []
+            tail = [formatter(x) for x in obj]
+
+        # adjust all values to max length if needed
+        if is_justify:
+            if line_break_each_value:
+                # Justify each string in the values of head and tail, so the
+                # strings will right align when head and tail are stacked
+                # vertically.
+                head, tail = _justify(head, tail)
+            elif is_truncated or not (
+                len(", ".join(head)) < display_width
+                and len(", ".join(tail)) < display_width
+            ):
+                # Each string in head and tail should align with each other
+                max_length = max(best_len(head), best_len(tail))
+                head = [x.rjust(max_length) for x in head]
+                tail = [x.rjust(max_length) for x in tail]
+            # If we are not truncated and we are only a single
+            # line, then don't justify
+
+        if line_break_each_value:
+            # Now head and tail are of type List[Tuple[str]]. Below we
+            # convert them into List[str], so there will be one string per
+            # value. Also truncate items horizontally if wider than
+            # max_space
+            max_space = display_width - len(space2)
+            value = tail[0]
+            max_items = 1
+            for num_items in reversed(range(1, len(value) + 1)):
+                pprinted_seq = _pprint_seq(value, max_seq_items=num_items)
+                if len(pprinted_seq) < max_space:
+                    max_items = num_items
+                    break
+            head = [_pprint_seq(x, max_seq_items=max_items) for x in head]
+            tail = [_pprint_seq(x, max_seq_items=max_items) for x in tail]
+
+        summary = ""
+        line = space2
+
+        for head_value in head:
+            word = head_value + sep + " "
+            summary, line = _extend_line(summary, line, word, display_width, space2)
+
+        if is_truncated:
+            # remove trailing space of last line
+            summary += line.rstrip() + space2 + "..."
+            line = space2
+
+        for tail_item in tail[:-1]:
+            word = tail_item + sep + " "
+            summary, line = _extend_line(summary, line, word, display_width, space2)
+
+        # last value: no sep added + 1 space of width used for trailing ','
+        summary, line = _extend_line(summary, line, tail[-1], display_width - 2, space2)
+        summary += line
+
+        # right now close is either '' or ', '
+        # Now we want to include the ']', but not the maybe space.
+        close = "]" + close.rstrip(" ")
+        summary += close
+
+        if len(summary) > (display_width) or line_break_each_value:
+            summary += space1
+        else:  # one row
+            summary += " "
+
+        # remove initial space
+        summary = "[" + summary[len(space2) :]
+
+    return summary
+
+
+def _justify(
+    head: list[Sequence[str]], tail: list[Sequence[str]]
+) -> tuple[list[tuple[str, ...]], list[tuple[str, ...]]]:
+    """
+    Justify items in head and tail, so they are right-aligned when stacked.
+
+    Parameters
+    ----------
+    head : list-like of list-likes of strings
+    tail : list-like of list-likes of strings
+
+    Returns
+    -------
+    tuple of list of tuples of strings
+        Same as head and tail, but items are right aligned when stacked
+        vertically.
+
+    Examples
+    --------
+    >>> _justify([["a", "b"]], [["abc", "abcd"]])
+    ([('  a', '   b')], [('abc', 'abcd')])
+    """
+    combined = head + tail
+
+    # For each position for the sequences in ``combined``,
+    # find the length of the largest string.
+    max_length = [0] * len(combined[0])
+    for inner_seq in combined:
+        length = [len(item) for item in inner_seq]
+        max_length = [max(x, y) for x, y in zip(max_length, length)]
+
+    # justify each item in each list-like in head and tail using max_length
+    head_tuples = [
+        tuple(x.rjust(max_len) for x, max_len in zip(seq, max_length)) for seq in head
+    ]
+    tail_tuples = [
+        tuple(x.rjust(max_len) for x, max_len in zip(seq, max_length)) for seq in tail
+    ]
+    return head_tuples, tail_tuples
+
+
+class PrettyDict(dict[_KT, _VT]):
+    """Dict extension to support abbreviated __repr__"""
+
+    def __repr__(self) -> str:
+        return pprint_thing(self)
+
+
+class _TextAdjustment:
+    def __init__(self) -> None:
+        self.encoding = get_option("display.encoding")
+
+    def len(self, text: str) -> int:
+        return len(text)
+
+    def justify(self, texts: Any, max_len: int, mode: str = "right") -> list[str]:
+        """
+        Perform ljust, center, rjust against string or list-like
+        """
+        if mode == "left":
+            return [x.ljust(max_len) for x in texts]
+        elif mode == "center":
+            return [x.center(max_len) for x in texts]
+        else:
+            return [x.rjust(max_len) for x in texts]
+
+    def adjoin(self, space: int, *lists: Any, **kwargs: Any) -> str:
+        return adjoin(space, *lists, strlen=self.len, justfunc=self.justify, **kwargs)
+
+
+class _EastAsianTextAdjustment(_TextAdjustment):
+    def __init__(self) -> None:
+        super().__init__()
+        if get_option("display.unicode.ambiguous_as_wide"):
+            self.ambiguous_width = 2
+        else:
+            self.ambiguous_width = 1
+
+        # Definition of East Asian Width
+        # https://unicode.org/reports/tr11/
+        # Ambiguous width can be changed by option
+        self._EAW_MAP = {"Na": 1, "N": 1, "W": 2, "F": 2, "H": 1}
+
+    def len(self, text: str) -> int:
+        """
+        Calculate display width considering unicode East Asian Width
+        """
+        if not isinstance(text, str):
+            return len(text)
+
+        return sum(
+            self._EAW_MAP.get(east_asian_width(c), self.ambiguous_width) for c in text
+        )
+
+    def justify(
+        self, texts: Iterable[str], max_len: int, mode: str = "right"
+    ) -> list[str]:
+        # re-calculate padding space per str considering East Asian Width
+        def _get_pad(t: str) -> int:
+            return max_len - self.len(t) + len(t)
+
+        if mode == "left":
+            return [x.ljust(_get_pad(x)) for x in texts]
+        elif mode == "center":
+            return [x.center(_get_pad(x)) for x in texts]
+        else:
+            return [x.rjust(_get_pad(x)) for x in texts]
+
+
+def get_adjustment() -> _TextAdjustment:
+    use_east_asian_width = get_option("display.unicode.east_asian_width")
+    if use_east_asian_width:
+        return _EastAsianTextAdjustment()
+    else:
+        return _TextAdjustment()
diff --git a/pandas/pandas/io/formats/string.py b/pandas/pandas/io/formats/string.py
new file mode 100644
index 0000000000000..ca41726de08cf
--- /dev/null
+++ b/pandas/pandas/io/formats/string.py
@@ -0,0 +1,207 @@
+"""
+Module for formatting output data in console (to string).
+"""
+
+from __future__ import annotations
+
+from shutil import get_terminal_size
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.io.formats.printing import pprint_thing
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+    from pandas.io.formats.format import DataFrameFormatter
+
+
+class StringFormatter:
+    """Formatter for string representation of a dataframe."""
+
+    def __init__(self, fmt: DataFrameFormatter, line_width: int | None = None) -> None:
+        self.fmt = fmt
+        self.adj = fmt.adj
+        self.frame = fmt.frame
+        self.line_width = line_width
+
+    def to_string(self) -> str:
+        text = self._get_string_representation()
+        if self.fmt.should_show_dimensions:
+            text = f"{text}{self.fmt.dimensions_info}"
+        return text
+
+    def _get_strcols(self) -> list[list[str]]:
+        strcols = self.fmt.get_strcols()
+        if self.fmt.is_truncated:
+            strcols = self._insert_dot_separators(strcols)
+        return strcols
+
+    def _get_string_representation(self) -> str:
+        if self.fmt.frame.empty:
+            return self._empty_info_line
+
+        strcols = self._get_strcols()
+
+        if self.line_width is None:
+            # no need to wrap around just print the whole frame
+            return self.adj.adjoin(1, *strcols)
+
+        if self._need_to_wrap_around:
+            return self._join_multiline(strcols)
+
+        return self._fit_strcols_to_terminal_width(strcols)
+
+    @property
+    def _empty_info_line(self) -> str:
+        return (
+            f"Empty {type(self.frame).__name__}\n"
+            f"Columns: {pprint_thing(self.frame.columns)}\n"
+            f"Index: {pprint_thing(self.frame.index)}"
+        )
+
+    @property
+    def _need_to_wrap_around(self) -> bool:
+        return bool(self.fmt.max_cols is None or self.fmt.max_cols > 0)
+
+    def _insert_dot_separators(self, strcols: list[list[str]]) -> list[list[str]]:
+        str_index = self.fmt._get_formatted_index(self.fmt.tr_frame)
+        index_length = len(str_index)
+
+        if self.fmt.is_truncated_horizontally:
+            strcols = self._insert_dot_separator_horizontal(strcols, index_length)
+
+        if self.fmt.is_truncated_vertically:
+            strcols = self._insert_dot_separator_vertical(strcols, index_length)
+
+        return strcols
+
+    @property
+    def _adjusted_tr_col_num(self) -> int:
+        return self.fmt.tr_col_num + 1 if self.fmt.index else self.fmt.tr_col_num
+
+    def _insert_dot_separator_horizontal(
+        self, strcols: list[list[str]], index_length: int
+    ) -> list[list[str]]:
+        strcols.insert(self._adjusted_tr_col_num, [" ..."] * index_length)
+        return strcols
+
+    def _insert_dot_separator_vertical(
+        self, strcols: list[list[str]], index_length: int
+    ) -> list[list[str]]:
+        n_header_rows = index_length - len(self.fmt.tr_frame)
+        row_num = self.fmt.tr_row_num
+        for ix, col in enumerate(strcols):
+            cwidth = self.adj.len(col[row_num])
+
+            if self.fmt.is_truncated_horizontally:
+                is_dot_col = ix == self._adjusted_tr_col_num
+            else:
+                is_dot_col = False
+
+            if cwidth > 3 or is_dot_col:
+                dots = "..."
+            else:
+                dots = ".."
+
+            if ix == 0 and self.fmt.index:
+                dot_mode = "left"
+            elif is_dot_col:
+                cwidth = 4
+                dot_mode = "right"
+            else:
+                dot_mode = "right"
+
+            dot_str = self.adj.justify([dots], cwidth, mode=dot_mode)[0]
+            col.insert(row_num + n_header_rows, dot_str)
+        return strcols
+
+    def _join_multiline(self, strcols_input: Iterable[list[str]]) -> str:
+        lwidth = self.line_width
+        adjoin_width = 1
+        strcols = list(strcols_input)
+
+        if self.fmt.index:
+            idx = strcols.pop(0)
+            lwidth -= np.array([self.adj.len(x) for x in idx]).max() + adjoin_width
+
+        col_widths = [
+            np.array([self.adj.len(x) for x in col]).max() if len(col) > 0 else 0
+            for col in strcols
+        ]
+
+        assert lwidth is not None
+        col_bins = _binify(col_widths, lwidth)
+        nbins = len(col_bins)
+
+        str_lst = []
+        start = 0
+        for i, end in enumerate(col_bins):
+            row = strcols[start:end]
+            if self.fmt.index:
+                row.insert(0, idx)
+            if nbins > 1:
+                nrows = len(row[-1])
+                if end <= len(strcols) and i < nbins - 1:
+                    row.append([" \\"] + ["  "] * (nrows - 1))
+                else:
+                    row.append([" "] * nrows)
+            str_lst.append(self.adj.adjoin(adjoin_width, *row))
+            start = end
+        return "\n\n".join(str_lst)
+
+    def _fit_strcols_to_terminal_width(self, strcols: list[list[str]]) -> str:
+        from pandas import Series
+
+        lines = self.adj.adjoin(1, *strcols).split("\n")
+        max_len = Series(lines).str.len().max()
+        # plus truncate dot col
+        width, _ = get_terminal_size()
+        dif = max_len - width
+        # '+ 1' to avoid too wide repr (GH PR #17023)
+        adj_dif = dif + 1
+        col_lens = Series([Series(ele).str.len().max() for ele in strcols])
+        n_cols = len(col_lens)
+        counter = 0
+        while adj_dif > 0 and n_cols > 1:
+            counter += 1
+            mid = round(n_cols / 2)
+            mid_ix = col_lens.index[mid]
+            col_len = col_lens[mid_ix]
+            # adjoin adds one
+            adj_dif -= col_len + 1
+            col_lens = col_lens.drop(mid_ix)
+            n_cols = len(col_lens)
+
+        # subtract index column
+        max_cols_fitted = n_cols - self.fmt.index
+        # GH-21180. Ensure that we print at least two.
+        max_cols_fitted = max(max_cols_fitted, 2)
+        self.fmt.max_cols_fitted = max_cols_fitted
+
+        # Call again _truncate to cut frame appropriately
+        # and then generate string representation
+        self.fmt.truncate()
+        strcols = self._get_strcols()
+        return self.adj.adjoin(1, *strcols)
+
+
+def _binify(cols: list[int], line_width: int) -> list[int]:
+    adjoin_width = 1
+    bins = []
+    curr_width = 0
+    i_last_column = len(cols) - 1
+    for i, w in enumerate(cols):
+        w_adjoined = w + adjoin_width
+        curr_width += w_adjoined
+        if i_last_column == i:
+            wrap = curr_width + 1 > line_width and i > 0
+        else:
+            wrap = curr_width + 2 > line_width and i > 0
+        if wrap:
+            bins.append(i)
+            curr_width = w_adjoined
+
+    bins.append(len(cols))
+    return bins
diff --git a/pandas/pandas/io/formats/style.py b/pandas/pandas/io/formats/style.py
new file mode 100644
index 0000000000000..0f734a81795c4
--- /dev/null
+++ b/pandas/pandas/io/formats/style.py
@@ -0,0 +1,4370 @@
+"""
+Module for applying conditional formatting to DataFrames and Series.
+"""
+
+from __future__ import annotations
+
+import copy
+from functools import partial
+import operator
+import textwrap
+from typing import (
+    TYPE_CHECKING,
+    overload,
+)
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._decorators import (
+    Substitution,
+    doc,
+)
+
+import pandas as pd
+from pandas import (
+    IndexSlice,
+    RangeIndex,
+)
+import pandas.core.common as com
+from pandas.core.frame import (
+    DataFrame,
+    Series,
+)
+from pandas.core.generic import NDFrame
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.formats.format import save_to_buffer
+
+jinja2 = import_optional_dependency("jinja2", extra="DataFrame.style requires jinja2.")
+
+from pandas.io.formats.style_render import (
+    CSSProperties,
+    CSSStyles,
+    ExtFormatter,
+    StylerRenderer,
+    Subset,
+    Tooltips,
+    format_table_styles,
+    maybe_convert_css_to_tuples,
+    non_reducing_slice,
+    refactor_levels,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Sequence,
+    )
+
+    from matplotlib.colors import Colormap
+
+    from pandas._typing import (
+        Any,
+        Axis,
+        AxisInt,
+        Concatenate,
+        ExcelWriterMergeCells,
+        FilePath,
+        IndexLabel,
+        IntervalClosedType,
+        Level,
+        P,
+        QuantileInterpolation,
+        Scalar,
+        Self,
+        StorageOptions,
+        T,
+        WriteBuffer,
+        WriteExcelBuffer,
+    )
+
+    from pandas import ExcelWriter
+
+
+####
+# Shared Doc Strings
+
+subset_args = """subset : label, array-like, IndexSlice, optional
+            A valid 2d input to `DataFrame.loc[<subset>]`, or, in the case of a 1d input
+            or single key, to `DataFrame.loc[:, <subset>]` where the columns are
+            prioritised, to limit ``data`` to *before* applying the function."""
+
+properties_args = """props : str, default None
+           CSS properties to use for highlighting. If ``props`` is given, ``color``
+           is not used."""
+
+coloring_args = """color : str, default '{default}'
+           Background color to use for highlighting."""
+
+buffering_args = """buf : str, path object, file-like object, optional
+         String, path object (implementing ``os.PathLike[str]``), or file-like
+         object implementing a string ``write()`` function. If ``None``, the result is
+         returned as a string."""
+
+encoding_args = """encoding : str, optional
+              Character encoding setting for file output (and meta tags if available).
+              Defaults to ``pandas.options.styler.render.encoding`` value of "utf-8"."""
+
+#
+###
+
+
+class Styler(StylerRenderer):
+    r"""
+    Helps style a DataFrame or Series according to the data with HTML and CSS.
+
+    This class provides methods for styling and formatting a Pandas DataFrame or Series.
+    The styled output can be rendered as HTML or LaTeX, and it supports CSS-based
+    styling, allowing users to control colors, font styles, and other visual aspects of
+    tabular data. It is particularly useful for presenting DataFrame objects in a
+    Jupyter Notebook environment or when exporting styled tables for reports and
+
+    Parameters
+    ----------
+    data : Series or DataFrame
+        Data to be styled - either a Series or DataFrame.
+    precision : int, optional
+        Precision to round floats to. If not given defaults to
+        ``pandas.options.styler.format.precision``.
+
+        .. versionchanged:: 1.4.0
+    table_styles : list-like, default None
+        List of {selector: (attr, value)} dicts; see Notes.
+    uuid : str, default None
+        A unique identifier to avoid CSS collisions; generated automatically.
+    caption : str, tuple, default None
+        String caption to attach to the table. Tuple only used for LaTeX dual captions.
+    table_attributes : str, default None
+        Items that show up in the opening ``<table>`` tag
+        in addition to automatic (by default) id.
+    cell_ids : bool, default True
+        If True, each cell will have an ``id`` attribute in their HTML tag.
+        The ``id`` takes the form ``T_<uuid>_row<num_row>_col<num_col>``
+        where ``<uuid>`` is the unique identifier, ``<num_row>`` is the row
+        number and ``<num_col>`` is the column number.
+    na_rep : str, optional
+        Representation for missing values.
+        If ``na_rep`` is None, no special formatting is applied, and falls back to
+        ``pandas.options.styler.format.na_rep``.
+
+    uuid_len : int, default 5
+        If ``uuid`` is not specified, the length of the ``uuid`` to randomly generate
+        expressed in hex characters, in range [0, 32].
+    decimal : str, optional
+        Character used as decimal separator for floats, complex and integers. If not
+        given uses ``pandas.options.styler.format.decimal``.
+
+        .. versionadded:: 1.3.0
+
+    thousands : str, optional, default None
+        Character used as thousands separator for floats, complex and integers. If not
+        given uses ``pandas.options.styler.format.thousands``.
+
+        .. versionadded:: 1.3.0
+
+    escape : str, optional
+        Use 'html' to replace the characters ``&``, ``<``, ``>``, ``'``, and ``"``
+        in cell display string with HTML-safe sequences.
+
+        Use 'latex' to replace the characters ``&``, ``%``, ``$``, ``#``, ``_``,
+        ``{``, ``}``, ``~``, ``^``, and ``\`` in the cell display string with
+        LaTeX-safe sequences. Use 'latex-math' to replace the characters
+        the same way as in 'latex' mode, except for math substrings,
+        which either are surrounded by two characters ``$`` or start with
+        the character ``\(`` and end with ``\)``.
+        If not given uses ``pandas.options.styler.format.escape``.
+
+        .. versionadded:: 1.3.0
+    formatter : str, callable, dict, optional
+        Object to define how values are displayed. See ``Styler.format``. If not given
+        uses ``pandas.options.styler.format.formatter``.
+
+        .. versionadded:: 1.4.0
+
+    Attributes
+    ----------
+    index : data.index Index
+    columns : data.columns Index
+    env : Jinja2 jinja2.Environment
+    template_html : Jinja2 Template
+    template_html_table : Jinja2 Template
+    template_html_style : Jinja2 Template
+    template_latex : Jinja2 Template
+    loader : Jinja2 Loader
+
+    See Also
+    --------
+    DataFrame.style : Return a Styler object containing methods for building
+        a styled HTML representation for the DataFrame.
+
+    Notes
+    -----
+    .. warning::
+
+       ``Styler`` is primarily intended for use on safe input that you control.
+       When using ``Styler`` on untrusted, user-provided input to serve HTML,
+       you should set ``escape="html"`` to prevent security vulnerabilities.
+       See the Jinja2 documentation on escaping HTML for more.
+
+    Most styling will be done by passing style functions into
+    ``Styler.apply`` or ``Styler.map``. Style functions should
+    return values with strings containing CSS ``'attr: value'`` that will
+    be applied to the indicated cells.
+
+    If using in the Jupyter notebook, Styler has defined a ``_repr_html_``
+    to automatically render itself. Otherwise call Styler.to_html to get
+    the generated HTML.
+
+    CSS classes are attached to the generated HTML
+
+    * Index and Column names include ``index_name`` and ``level<k>``
+      where `k` is its level in a MultiIndex
+    * Index label cells include
+
+      * ``row_heading``
+      * ``row<n>`` where `n` is the numeric position of the row
+      * ``level<k>`` where `k` is the level in a MultiIndex
+
+    * Column label cells include
+
+      * ``col_heading``
+      * ``col<n>`` where `n` is the numeric position of the column
+      * ``level<k>`` where `k` is the level in a MultiIndex
+
+    * Blank cells include ``blank``
+    * Data cells include ``data``
+    * Trimmed cells include ``col_trim`` or ``row_trim``.
+
+    Any, or all, or these classes can be renamed by using the ``css_class_names``
+    argument in ``Styler.set_table_styles``, giving a value such as
+    *{"row": "MY_ROW_CLASS", "col_trim": "", "row_trim": ""}*.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame(
+    ...     [[1.0, 2.0, 3.0], [4, 5, 6]], index=["a", "b"], columns=["A", "B", "C"]
+    ... )
+    >>> pd.io.formats.style.Styler(
+    ...     df, precision=2, caption="My table"
+    ... )  # doctest: +SKIP
+
+    Please see:
+    `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+    """
+
+    def __init__(
+        self,
+        data: DataFrame | Series,
+        precision: int | None = None,
+        table_styles: CSSStyles | None = None,
+        uuid: str | None = None,
+        caption: str | tuple | list | None = None,
+        table_attributes: str | None = None,
+        cell_ids: bool = True,
+        na_rep: str | None = None,
+        uuid_len: int = 5,
+        decimal: str | None = None,
+        thousands: str | None = None,
+        escape: str | None = None,
+        formatter: ExtFormatter | None = None,
+    ) -> None:
+        super().__init__(
+            data=data,
+            uuid=uuid,
+            uuid_len=uuid_len,
+            table_styles=table_styles,
+            table_attributes=table_attributes,
+            caption=caption,
+            cell_ids=cell_ids,
+            precision=precision,
+        )
+
+        # validate ordered args
+        thousands = thousands or get_option("styler.format.thousands")
+        decimal = decimal or get_option("styler.format.decimal")
+        na_rep = na_rep or get_option("styler.format.na_rep")
+        escape = escape or get_option("styler.format.escape")
+        formatter = formatter or get_option("styler.format.formatter")
+        # precision is handled by superclass as default for performance
+
+        self.format(
+            formatter=formatter,
+            precision=precision,
+            na_rep=na_rep,
+            escape=escape,
+            decimal=decimal,
+            thousands=thousands,
+        )
+
+    def concat(self, other: Styler) -> Styler:
+        """
+        Append another Styler to combine the output into a single table.
+
+        .. versionadded:: 1.5.0
+
+        Parameters
+        ----------
+        other : Styler
+            The other Styler object which has already been styled and formatted. The
+            data for this Styler must have the same columns as the original, and the
+            number of index levels must also be the same to render correctly.
+
+        Returns
+        -------
+        Styler
+            Instance of class with specified Styler appended.
+
+        See Also
+        --------
+        Styler.clear : Reset the ``Styler``, removing any previously applied styles.
+        Styler.export : Export the styles applied to the current Styler.
+
+        Notes
+        -----
+        The purpose of this method is to extend existing styled dataframes with other
+        metrics that may be useful but may not conform to the original's structure.
+        For example adding a sub total row, or displaying metrics such as means,
+        variance or counts.
+
+        Styles that are applied using the ``apply``, ``map``, ``apply_index``
+        and ``map_index``, and formatting applied with ``format`` and
+        ``format_index`` will be preserved.
+
+        .. warning::
+            Only the output methods ``to_html``, ``to_string`` and ``to_latex``
+            currently work with concatenated Stylers.
+
+            Other output methods, including ``to_excel``, **do not** work with
+            concatenated Stylers.
+
+        The following should be noted:
+
+          - ``table_styles``, ``table_attributes``, ``caption`` and ``uuid`` are all
+            inherited from the original Styler and not ``other``.
+          - hidden columns and hidden index levels will be inherited from the
+            original Styler
+          - ``css`` will be inherited from the original Styler, and the value of
+            keys ``data``, ``row_heading`` and ``row`` will be prepended with
+            ``foot0_``. If more concats are chained, their styles will be prepended
+            with ``foot1_``, ''foot_2'', etc., and if a concatenated style have
+            another concatenated style, the second style will be prepended with
+            ``foot{parent}_foot{child}_``.
+
+        A common use case is to concatenate user defined functions with
+        ``DataFrame.agg`` or with described statistics via ``DataFrame.describe``.
+        See examples.
+
+        Examples
+        --------
+        A common use case is adding totals rows, or otherwise, via methods calculated
+        in ``DataFrame.agg``.
+
+        >>> df = pd.DataFrame(
+        ...     [[4, 6], [1, 9], [3, 4], [5, 5], [9, 6]],
+        ...     columns=["Mike", "Jim"],
+        ...     index=["Mon", "Tue", "Wed", "Thurs", "Fri"],
+        ... )
+        >>> styler = df.style.concat(df.agg(["sum"]).style)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/footer_simple.png
+
+        Since the concatenated object is a Styler the existing functionality can be
+        used to conditionally format it as well as the original.
+
+        >>> descriptors = df.agg(["sum", "mean", lambda s: s.dtype])
+        >>> descriptors.index = ["Total", "Average", "dtype"]
+        >>> other = (
+        ...     descriptors.style.highlight_max(
+        ...         axis=1, subset=(["Total", "Average"], slice(None))
+        ...     )
+        ...     .format(subset=("Average", slice(None)), precision=2, decimal=",")
+        ...     .map(lambda v: "font-weight: bold;")
+        ... )
+        >>> styler = df.style.highlight_max(color="salmon").set_table_styles(
+        ...     [{"selector": ".foot_row0", "props": "border-top: 1px solid black;"}]
+        ... )
+        >>> styler.concat(other)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/footer_extended.png
+
+        When ``other`` has fewer index levels than the original Styler it is possible
+        to extend the index in ``other``, with placeholder levels.
+
+        >>> df = pd.DataFrame(
+        ...     [[1], [2]], index=pd.MultiIndex.from_product([[0], [1, 2]])
+        ... )
+        >>> descriptors = df.agg(["sum"])
+        >>> descriptors.index = pd.MultiIndex.from_product([[""], descriptors.index])
+        >>> df.style.concat(descriptors.style)  # doctest: +SKIP
+        """
+        if not isinstance(other, Styler):
+            raise TypeError("`other` must be of type `Styler`")
+        if not self.data.columns.equals(other.data.columns):
+            raise ValueError("`other.data` must have same columns as `Styler.data`")
+        if not self.data.index.nlevels == other.data.index.nlevels:
+            raise ValueError(
+                "number of index levels must be same in `other` "
+                "as in `Styler`. See documentation for suggestions."
+            )
+        self.concatenated.append(other)
+        return self
+
+    def _repr_html_(self) -> str | None:
+        """
+        Hooks into Jupyter notebook rich display system, which calls _repr_html_ by
+        default if an object is returned at the end of a cell.
+        """
+        if get_option("styler.render.repr") == "html":
+            return self.to_html()
+        return None
+
+    def _repr_latex_(self) -> str | None:
+        if get_option("styler.render.repr") == "latex":
+            return self.to_latex()
+        return None
+
+    def set_tooltips(
+        self,
+        ttips: DataFrame,
+        props: CSSProperties | None = None,
+        css_class: str | None = None,
+        as_title_attribute: bool = False,
+    ) -> Styler:
+        """
+        Set the DataFrame of strings on ``Styler`` generating ``:hover`` tooltips.
+
+        These string based tooltips are only applicable to ``<td>`` HTML elements,
+        and cannot be used for column or index headers.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        ttips : DataFrame
+            DataFrame containing strings that will be translated to tooltips, mapped
+            by identical column and index values that must exist on the underlying
+            Styler data. None, NaN values, and empty strings will be ignored and
+            not affect the rendered HTML.
+        props : list-like or str, optional
+            List of (attr, value) tuples or a valid CSS string. If ``None`` adopts
+            the internal default values described in notes.
+        css_class : str, optional
+            Name of the tooltip class used in CSS, should conform to HTML standards.
+            Only useful if integrating tooltips with external CSS. If ``None`` uses the
+            internal default value 'pd-t'.
+        as_title_attribute : bool, default False
+            Add the tooltip text as title attribute to resultant <td> element. If True
+            then props and css_class arguments are ignored.
+
+        Returns
+        -------
+        Styler
+            Instance of class with DataFrame set for strings on ``Styler``
+                generating ``:hover`` tooltips.
+
+        See Also
+        --------
+        Styler.set_table_attributes : Set the table attributes added to the
+            ``<table>`` HTML element.
+        Styler.set_table_styles : Set the table styles included within the
+            ``<style>`` HTML element.
+
+        Notes
+        -----
+        Tooltips are created by adding `<span class="pd-t"></span>` to each data cell
+        and then manipulating the table level CSS to attach pseudo hover and pseudo
+        after selectors to produce the required the results.
+
+        The default properties for the tooltip CSS class are:
+
+        - visibility: hidden
+        - position: absolute
+        - z-index: 1
+        - background-color: black
+        - color: white
+        - transform: translate(-20px, -20px)
+
+        The property 'visibility: hidden;' is a key prerequisite to the hover
+        functionality, and should always be included in any manual properties
+        specification, using the ``props`` argument.
+
+        Tooltips are not designed to be efficient, and can add large amounts of
+        additional HTML for larger tables, since they also require that ``cell_ids``
+        is forced to `True`.
+
+        If multiline tooltips are required, or if styling is not required and/or
+        space is of concern, then utilizing as_title_attribute as True will store
+        the tooltip on the <td> title attribute. This will cause no CSS
+        to be generated nor will the <span> elements. Storing tooltips through
+        the title attribute will mean that tooltip styling effects do not apply.
+
+        Examples
+        --------
+        Basic application
+
+        >>> df = pd.DataFrame(data=[[0, 1], [2, 3]])
+        >>> ttips = pd.DataFrame(
+        ...     data=[["Min", ""], [np.nan, "Max"]], columns=df.columns, index=df.index
+        ... )
+        >>> s = df.style.set_tooltips(ttips).to_html()
+
+        Optionally controlling the tooltip visual display
+
+        >>> df.style.set_tooltips(
+        ...     ttips,
+        ...     css_class="tt-add",
+        ...     props=[
+        ...         ("visibility", "hidden"),
+        ...         ("position", "absolute"),
+        ...         ("z-index", 1),
+        ...     ],
+        ... )  # doctest: +SKIP
+        >>> df.style.set_tooltips(
+        ...     ttips,
+        ...     css_class="tt-add",
+        ...     props="visibility:hidden; position:absolute; z-index:1;",
+        ... )
+        ... # doctest: +SKIP
+
+        Multiline tooltips with smaller size footprint
+
+        >>> df.style.set_tooltips(ttips, as_title_attribute=True)  # doctest: +SKIP
+        """
+        if not self.cell_ids:
+            # tooltips not optimised for individual cell check. requires reasonable
+            # redesign and more extensive code for a feature that might be rarely used.
+            raise NotImplementedError(
+                "Tooltips can only render with 'cell_ids' is True."
+            )
+        if not ttips.index.is_unique or not ttips.columns.is_unique:
+            raise KeyError(
+                "Tooltips render only if `ttips` has unique index and columns."
+            )
+        if self.tooltips is None:  # create a default instance if necessary
+            self.tooltips = Tooltips()
+        self.tooltips.tt_data = ttips
+        if not as_title_attribute:
+            if props:
+                self.tooltips.class_properties = props
+            if css_class:
+                self.tooltips.class_name = css_class
+        else:
+            self.tooltips.as_title_attribute = as_title_attribute
+
+        return self
+
+    @doc(
+        NDFrame.to_excel,
+        klass="Styler",
+        storage_options=_shared_docs["storage_options"],
+        storage_options_versionadded="1.5.0",
+        encoding_parameter=textwrap.dedent(
+            """\
+        encoding : str or None, default None
+            Unused parameter, present for compatibility.
+        """
+        ),
+        verbose_parameter=textwrap.dedent(
+            """\
+        verbose : str, default True
+            Optional unused parameter, present for compatibility.
+        """
+        ),
+        extra_parameters="",
+    )
+    def to_excel(
+        self,
+        excel_writer: FilePath | WriteExcelBuffer | ExcelWriter,
+        sheet_name: str = "Sheet1",
+        na_rep: str = "",
+        float_format: str | None = None,
+        columns: Sequence[Hashable] | None = None,
+        header: Sequence[Hashable] | bool = True,
+        index: bool = True,
+        index_label: IndexLabel | None = None,
+        startrow: int = 0,
+        startcol: int = 0,
+        engine: str | None = None,
+        merge_cells: ExcelWriterMergeCells = True,
+        encoding: str | None = None,
+        inf_rep: str = "inf",
+        verbose: bool = True,
+        freeze_panes: tuple[int, int] | None = None,
+        storage_options: StorageOptions | None = None,
+    ) -> None:
+        from pandas.io.formats.excel import ExcelFormatter
+
+        formatter = ExcelFormatter(
+            self,
+            na_rep=na_rep,
+            cols=columns,
+            header=header,
+            float_format=float_format,
+            index=index,
+            index_label=index_label,
+            merge_cells=merge_cells,
+            inf_rep=inf_rep,
+        )
+        formatter.write(
+            excel_writer,
+            sheet_name=sheet_name,
+            startrow=startrow,
+            startcol=startcol,
+            freeze_panes=freeze_panes,
+            engine=engine,
+            storage_options=storage_options,
+        )
+
+    @overload
+    def to_latex(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        column_format: str | None = ...,
+        position: str | None = ...,
+        position_float: str | None = ...,
+        hrules: bool | None = ...,
+        clines: str | None = ...,
+        label: str | None = ...,
+        caption: str | tuple | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        multirow_align: str | None = ...,
+        multicol_align: str | None = ...,
+        siunitx: bool = ...,
+        environment: str | None = ...,
+        encoding: str | None = ...,
+        convert_css: bool = ...,
+    ) -> None: ...
+
+    @overload
+    def to_latex(
+        self,
+        buf: None = ...,
+        *,
+        column_format: str | None = ...,
+        position: str | None = ...,
+        position_float: str | None = ...,
+        hrules: bool | None = ...,
+        clines: str | None = ...,
+        label: str | None = ...,
+        caption: str | tuple | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        multirow_align: str | None = ...,
+        multicol_align: str | None = ...,
+        siunitx: bool = ...,
+        environment: str | None = ...,
+        encoding: str | None = ...,
+        convert_css: bool = ...,
+    ) -> str: ...
+
+    def to_latex(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        column_format: str | None = None,
+        position: str | None = None,
+        position_float: str | None = None,
+        hrules: bool | None = None,
+        clines: str | None = None,
+        label: str | None = None,
+        caption: str | tuple | None = None,
+        sparse_index: bool | None = None,
+        sparse_columns: bool | None = None,
+        multirow_align: str | None = None,
+        multicol_align: str | None = None,
+        siunitx: bool = False,
+        environment: str | None = None,
+        encoding: str | None = None,
+        convert_css: bool = False,
+    ) -> str | None:
+        r"""
+        Write Styler to a file, buffer or string in LaTeX format.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        buf : str, path object, file-like object, or None, default None
+            String, path object (implementing ``os.PathLike[str]``), or file-like
+            object implementing a string ``write()`` function. If None, the result is
+            returned as a string.
+        column_format : str, optional
+            The LaTeX column specification placed in location:
+
+            \\begin{tabular}{<column_format>}
+
+            Defaults to 'l' for index and
+            non-numeric data columns, and, for numeric data columns,
+            to 'r' by default, or 'S' if ``siunitx`` is ``True``.
+        position : str, optional
+            The LaTeX positional argument (e.g. 'h!') for tables, placed in location:
+
+            ``\\begin{table}[<position>]``.
+        position_float : {"centering", "raggedleft", "raggedright"}, optional
+            The LaTeX float command placed in location:
+
+            \\begin{table}[<position>]
+
+            \\<position_float>
+
+            Cannot be used if ``environment`` is "longtable".
+        hrules : bool
+            Set to `True` to add \\toprule, \\midrule and \\bottomrule from the
+            {booktabs} LaTeX package.
+            Defaults to ``pandas.options.styler.latex.hrules``, which is `False`.
+
+            .. versionchanged:: 1.4.0
+        clines : str, optional
+            Use to control adding \\cline commands for the index labels separation.
+            Possible values are:
+
+              - `None`: no cline commands are added (default).
+              - `"all;data"`: a cline is added for every index value extending the
+                width of the table, including data entries.
+              - `"all;index"`: as above with lines extending only the width of the
+                index entries.
+              - `"skip-last;data"`: a cline is added for each index value except the
+                last level (which is never sparsified), extending the widtn of the
+                table.
+              - `"skip-last;index"`: as above with lines extending only the width of the
+                index entries.
+
+            .. versionadded:: 1.4.0
+        label : str, optional
+            The LaTeX label included as: \\label{<label>}.
+            This is used with \\ref{<label>} in the main .tex file.
+        caption : str, tuple, optional
+            If string, the LaTeX table caption included as: \\caption{<caption>}.
+            If tuple, i.e ("full caption", "short caption"), the caption included
+            as: \\caption[<caption[1]>]{<caption[0]>}.
+        sparse_index : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each row.
+            Defaults to ``pandas.options.styler.sparse.index``, which is `True`.
+        sparse_columns : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each
+            column. Defaults to ``pandas.options.styler.sparse.columns``, which
+            is `True`.
+        multirow_align : {"c", "t", "b", "naive"}, optional
+            If sparsifying hierarchical MultiIndexes whether to align text centrally,
+            at the top or bottom using the multirow package. If not given defaults to
+            ``pandas.options.styler.latex.multirow_align``, which is `"c"`.
+            If "naive" is given renders without multirow.
+
+            .. versionchanged:: 1.4.0
+        multicol_align : {"r", "c", "l", "naive-l", "naive-r"}, optional
+            If sparsifying hierarchical MultiIndex columns whether to align text at
+            the left, centrally, or at the right. If not given defaults to
+            ``pandas.options.styler.latex.multicol_align``, which is "r".
+            If a naive option is given renders without multicol.
+            Pipe decorators can also be added to non-naive values to draw vertical
+            rules, e.g. "\|r" will draw a rule on the left side of right aligned merged
+            cells.
+
+            .. versionchanged:: 1.4.0
+        siunitx : bool, default False
+            Set to ``True`` to structure LaTeX compatible with the {siunitx} package.
+        environment : str, optional
+            If given, the environment that will replace 'table' in ``\\begin{table}``.
+            If 'longtable' is specified then a more suitable template is
+            rendered. If not given defaults to
+            ``pandas.options.styler.latex.environment``, which is `None`.
+
+            .. versionadded:: 1.4.0
+        encoding : str, optional
+            Character encoding setting. Defaults
+            to ``pandas.options.styler.render.encoding``, which is "utf-8".
+        convert_css : bool, default False
+            Convert simple cell-styles from CSS to LaTeX format. Any CSS not found in
+            conversion table is dropped. A style can be forced by adding option
+            `--latex`. See notes.
+
+        Returns
+        -------
+        str or None
+            If `buf` is None, returns the result as a string. Otherwise returns `None`.
+
+        See Also
+        --------
+        Styler.format: Format the text display value of cells.
+
+        Notes
+        -----
+        **Latex Packages**
+
+        For the following features we recommend the following LaTeX inclusions:
+
+        ===================== ==========================================================
+        Feature               Inclusion
+        ===================== ==========================================================
+        sparse columns        none: included within default {tabular} environment
+        sparse rows           \\usepackage{multirow}
+        hrules                \\usepackage{booktabs}
+        colors                \\usepackage[table]{xcolor}
+        siunitx               \\usepackage{siunitx}
+        bold (with siunitx)   | \\usepackage{etoolbox}
+                              | \\robustify\\bfseries
+                              | \\sisetup{detect-all = true}  *(within {document})*
+        italic (with siunitx) | \\usepackage{etoolbox}
+                              | \\robustify\\itshape
+                              | \\sisetup{detect-all = true}  *(within {document})*
+        environment           \\usepackage{longtable} if arg is "longtable"
+                              | or any other relevant environment package
+        hyperlinks            \\usepackage{hyperref}
+        ===================== ==========================================================
+
+        **Cell Styles**
+
+        LaTeX styling can only be rendered if the accompanying styling functions have
+        been constructed with appropriate LaTeX commands. All styling
+        functionality is built around the concept of a CSS ``(<attribute>, <value>)``
+        pair (see `Table Visualization <../../user_guide/style.ipynb>`_), and this
+        should be replaced by a LaTeX
+        ``(<command>, <options>)`` approach. Each cell will be styled individually
+        using nested LaTeX commands with their accompanied options.
+
+        For example the following code will highlight and bold a cell in HTML-CSS:
+
+        >>> df = pd.DataFrame([[1, 2], [3, 4]])
+        >>> s = df.style.highlight_max(axis=None,
+        ...                            props='background-color:red; font-weight:bold;')
+        >>> s.to_html()  # doctest: +SKIP
+
+        The equivalent using LaTeX only commands is the following:
+
+        >>> s = df.style.highlight_max(axis=None,
+        ...                            props='cellcolor:{red}; bfseries: ;')
+        >>> s.to_latex()  # doctest: +SKIP
+
+        Internally these structured LaTeX ``(<command>, <options>)`` pairs
+        are translated to the
+        ``display_value`` with the default structure:
+        ``\<command><options> <display_value>``.
+        Where there are multiple commands the latter is nested recursively, so that
+        the above example highlighted cell is rendered as
+        ``\cellcolor{red} \bfseries 4``.
+
+        Occasionally this format does not suit the applied command, or
+        combination of LaTeX packages that is in use, so additional flags can be
+        added to the ``<options>``, within the tuple, to result in different
+        positions of required braces (the **default** being the same as ``--nowrap``):
+
+        =================================== ============================================
+        Tuple Format                           Output Structure
+        =================================== ============================================
+        (<command>,<options>)               \\<command><options> <display_value>
+        (<command>,<options> ``--nowrap``)  \\<command><options> <display_value>
+        (<command>,<options> ``--rwrap``)   \\<command><options>{<display_value>}
+        (<command>,<options> ``--wrap``)    {\\<command><options> <display_value>}
+        (<command>,<options> ``--lwrap``)   {\\<command><options>} <display_value>
+        (<command>,<options> ``--dwrap``)   {\\<command><options>}{<display_value>}
+        =================================== ============================================
+
+        For example the `textbf` command for font-weight
+        should always be used with `--rwrap` so ``('textbf', '--rwrap')`` will render a
+        working cell, wrapped with braces, as ``\textbf{<display_value>}``.
+
+        A more comprehensive example is as follows:
+
+        >>> df = pd.DataFrame([[1, 2.2, "dogs"], [3, 4.4, "cats"], [2, 6.6, "cows"]],
+        ...                   index=["ix1", "ix2", "ix3"],
+        ...                   columns=["Integers", "Floats", "Strings"])
+        >>> s = df.style.highlight_max(
+        ...     props='cellcolor:[HTML]{FFFF00}; color:{red};'
+        ...           'textit:--rwrap; textbf:--rwrap;'
+        ... )
+        >>> s.to_latex()  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/latex_1.png
+
+        **Table Styles**
+
+        Internally Styler uses its ``table_styles`` object to parse the
+        ``column_format``, ``position``, ``position_float``, and ``label``
+        input arguments. These arguments are added to table styles in the format:
+
+        .. code-block:: python
+
+            set_table_styles([
+                {"selector": "column_format", "props": f":{column_format};"},
+                {"selector": "position", "props": f":{position};"},
+                {"selector": "position_float", "props": f":{position_float};"},
+                {"selector": "label", "props": f":{{{label.replace(':','§')}}};"}
+            ], overwrite=False)
+
+        Exception is made for the ``hrules`` argument which, in fact, controls all three
+        commands: ``toprule``, ``bottomrule`` and ``midrule`` simultaneously. Instead of
+        setting ``hrules`` to ``True``, it is also possible to set each
+        individual rule definition, by manually setting the ``table_styles``,
+        for example below we set a regular ``toprule``, set an ``hline`` for
+        ``bottomrule`` and exclude the ``midrule``:
+
+        .. code-block:: python
+
+            set_table_styles([
+                {'selector': 'toprule', 'props': ':toprule;'},
+                {'selector': 'bottomrule', 'props': ':hline;'},
+            ], overwrite=False)
+
+        If other ``commands`` are added to table styles they will be detected, and
+        positioned immediately above the '\\begin{tabular}' command. For example to
+        add odd and even row coloring, from the {colortbl} package, in format
+        ``\rowcolors{1}{pink}{red}``, use:
+
+        .. code-block:: python
+
+            set_table_styles([
+                {'selector': 'rowcolors', 'props': ':{1}{pink}{red};'}
+            ], overwrite=False)
+
+        A more comprehensive example using these arguments is as follows:
+
+        >>> df.columns = pd.MultiIndex.from_tuples([
+        ...     ("Numeric", "Integers"),
+        ...     ("Numeric", "Floats"),
+        ...     ("Non-Numeric", "Strings")
+        ... ])
+        >>> df.index = pd.MultiIndex.from_tuples([
+        ...     ("L0", "ix1"), ("L0", "ix2"), ("L1", "ix3")
+        ... ])
+        >>> s = df.style.highlight_max(
+        ...     props='cellcolor:[HTML]{FFFF00}; color:{red}; itshape:; bfseries:;'
+        ... )
+        >>> s.to_latex(
+        ...     column_format="rrrrr", position="h", position_float="centering",
+        ...     hrules=True, label="table:5", caption="Styled LaTeX Table",
+        ...     multirow_align="t", multicol_align="r"
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/latex_2.png
+
+        **Formatting**
+
+        To format values :meth:`Styler.format` should be used prior to calling
+        `Styler.to_latex`, as well as other methods such as :meth:`Styler.hide`
+        for example:
+
+        >>> s.clear()
+        >>> s.table_styles = []
+        >>> s.caption = None
+        >>> s.format({
+        ...     ("Numeric", "Integers"): '\\${}',
+        ...     ("Numeric", "Floats"): '{:.3f}',
+        ...     ("Non-Numeric", "Strings"): str.upper
+        ... })  # doctest: +SKIP
+                        Numeric      Non-Numeric
+                  Integers   Floats    Strings
+        L0    ix1       $1   2.200      DOGS
+              ix2       $3   4.400      CATS
+        L1    ix3       $2   6.600      COWS
+
+        >>> s.to_latex()  # doctest: +SKIP
+        \begin{tabular}{llrrl}
+        {} & {} & \multicolumn{2}{r}{Numeric} & {Non-Numeric} \\
+        {} & {} & {Integers} & {Floats} & {Strings} \\
+        \multirow[c]{2}{*}{L0} & ix1 & \\$1 & 2.200 & DOGS \\
+         & ix2 & \$3 & 4.400 & CATS \\
+        L1 & ix3 & \$2 & 6.600 & COWS \\
+        \end{tabular}
+
+        **CSS Conversion**
+
+        This method can convert a Styler constructured with HTML-CSS to LaTeX using
+        the following limited conversions.
+
+        ================== ==================== ============= ==========================
+        CSS Attribute      CSS value            LaTeX Command LaTeX Options
+        ================== ==================== ============= ==========================
+        font-weight        | bold               | bfseries
+                           | bolder             | bfseries
+        font-style         | italic             | itshape
+                           | oblique            | slshape
+        background-color   | red                cellcolor     | {red}--lwrap
+                           | #fe01ea                          | [HTML]{FE01EA}--lwrap
+                           | #f0e                             | [HTML]{FF00EE}--lwrap
+                           | rgb(128,255,0)                   | [rgb]{0.5,1,0}--lwrap
+                           | rgba(128,0,0,0.5)                | [rgb]{0.5,0,0}--lwrap
+                           | rgb(25%,255,50%)                 | [rgb]{0.25,1,0.5}--lwrap
+        color              | red                color         | {red}
+                           | #fe01ea                          | [HTML]{FE01EA}
+                           | #f0e                             | [HTML]{FF00EE}
+                           | rgb(128,255,0)                   | [rgb]{0.5,1,0}
+                           | rgba(128,0,0,0.5)                | [rgb]{0.5,0,0}
+                           | rgb(25%,255,50%)                 | [rgb]{0.25,1,0.5}
+        ================== ==================== ============= ==========================
+
+        It is also possible to add user-defined LaTeX only styles to a HTML-CSS Styler
+        using the ``--latex`` flag, and to add LaTeX parsing options that the
+        converter will detect within a CSS-comment.
+
+        >>> df = pd.DataFrame([[1]])
+        >>> df.style.set_properties(
+        ...     **{"font-weight": "bold /* --dwrap */", "Huge": "--latex--rwrap"}
+        ... ).to_latex(convert_css=True)  # doctest: +SKIP
+        \begin{tabular}{lr}
+        {} & {0} \\
+        0 & {\bfseries}{\Huge{1}} \\
+        \end{tabular}
+
+        Examples
+        --------
+        Below we give a complete step by step example adding some advanced features
+        and noting some common gotchas.
+
+        First we create the DataFrame and Styler as usual, including MultiIndex rows
+        and columns, which allow for more advanced formatting options:
+
+        >>> cidx = pd.MultiIndex.from_arrays([
+        ...     ["Equity", "Equity", "Equity", "Equity",
+        ...      "Stats", "Stats", "Stats", "Stats", "Rating"],
+        ...     ["Energy", "Energy", "Consumer", "Consumer", "", "", "", "", ""],
+        ...     ["BP", "Shell", "H&M", "Unilever",
+        ...      "Std Dev", "Variance", "52w High", "52w Low", ""]
+        ... ])
+        >>> iidx = pd.MultiIndex.from_arrays([
+        ...     ["Equity", "Equity", "Equity", "Equity"],
+        ...     ["Energy", "Energy", "Consumer", "Consumer"],
+        ...     ["BP", "Shell", "H&M", "Unilever"]
+        ... ])
+        >>> styler = pd.DataFrame([
+        ...     [1, 0.8, 0.66, 0.72, 32.1678, 32.1678**2, 335.12, 240.89, "Buy"],
+        ...     [0.8, 1.0, 0.69, 0.79, 1.876, 1.876**2, 14.12, 19.78, "Hold"],
+        ...     [0.66, 0.69, 1.0, 0.86, 7, 7**2, 210.9, 140.6, "Buy"],
+        ...     [0.72, 0.79, 0.86, 1.0, 213.76, 213.76**2, 2807, 3678, "Sell"],
+        ... ], columns=cidx, index=iidx).style
+
+        Second we will format the display and, since our table is quite wide, will
+        hide the repeated level-0 of the index:
+
+        >>> (styler.format(subset="Equity", precision=2)
+        ...     .format(subset="Stats", precision=1, thousands=",")
+        ...     .format(subset="Rating", formatter=str.upper)
+        ...     .format_index(escape="latex", axis=1)
+        ...     .format_index(escape="latex", axis=0)
+        ...     .hide(level=0, axis=0))  # doctest: +SKIP
+
+        Note that one of the string entries of the index and column headers is "H&M".
+        Without applying the `escape="latex"` option to the `format_index` method the
+        resultant LaTeX will fail to render, and the error returned is quite
+        difficult to debug. Using the appropriate escape the "&" is converted to "\\&".
+
+        Thirdly we will apply some (CSS-HTML) styles to our object. We will use a
+        builtin method and also define our own method to highlight the stock
+        recommendation:
+
+        >>> def rating_color(v):
+        ...     if v == "Buy":
+        ...         color = "#33ff85"
+        ...     elif v == "Sell":
+        ...         color = "#ff5933"
+        ...     else:
+        ...         color = "#ffdd33"
+        ...     return f"color: {color}; font-weight: bold;"
+        >>> (styler.background_gradient(cmap="inferno", subset="Equity", vmin=0, vmax=1)
+        ...     .map(rating_color, subset="Rating"))  # doctest: +SKIP
+
+        All the above styles will work with HTML (see below) and LaTeX upon conversion:
+
+        .. figure:: ../../_static/style/latex_stocks_html.png
+
+        However, we finally want to add one LaTeX only style
+        (from the {graphicx} package), that is not easy to convert from CSS and
+        pandas does not support it. Notice the `--latex` flag used here,
+        as well as `--rwrap` to ensure this is formatted correctly and
+        not ignored upon conversion.
+
+        >>> styler.map_index(
+        ...     lambda v: "rotatebox:{45}--rwrap--latex;", level=2, axis=1
+        ... )  # doctest: +SKIP
+
+        Finally we render our LaTeX adding in other options as required:
+
+        >>> styler.to_latex(
+        ...     caption="Selected stock correlation and simple statistics.",
+        ...     clines="skip-last;data",
+        ...     convert_css=True,
+        ...     position_float="centering",
+        ...     multicol_align="|c|",
+        ...     hrules=True,
+        ... )  # doctest: +SKIP
+        \begin{table}
+        \centering
+        \caption{Selected stock correlation and simple statistics.}
+        \begin{tabular}{llrrrrrrrrl}
+        \toprule
+         &  & \multicolumn{4}{|c|}{Equity} & \multicolumn{4}{|c|}{Stats} & Rating \\
+         &  & \multicolumn{2}{|c|}{Energy} & \multicolumn{2}{|c|}{Consumer} &
+        \multicolumn{4}{|c|}{} &  \\
+         &  & \rotatebox{45}{BP} & \rotatebox{45}{Shell} & \rotatebox{45}{H\&M} &
+        \rotatebox{45}{Unilever} & \rotatebox{45}{Std Dev} & \rotatebox{45}{Variance} &
+        \rotatebox{45}{52w High} & \rotatebox{45}{52w Low} & \rotatebox{45}{} \\
+        \midrule
+        \multirow[c]{2}{*}{Energy} & BP & {\cellcolor[HTML]{FCFFA4}}
+        \color[HTML]{000000} 1.00 & {\cellcolor[HTML]{FCA50A}} \color[HTML]{000000}
+        0.80 & {\cellcolor[HTML]{EB6628}} \color[HTML]{F1F1F1} 0.66 &
+        {\cellcolor[HTML]{F68013}} \color[HTML]{F1F1F1} 0.72 & 32.2 & 1,034.8 & 335.1
+        & 240.9 & \color[HTML]{33FF85} \bfseries BUY \\
+         & Shell & {\cellcolor[HTML]{FCA50A}} \color[HTML]{000000} 0.80 &
+        {\cellcolor[HTML]{FCFFA4}} \color[HTML]{000000} 1.00 &
+        {\cellcolor[HTML]{F1731D}} \color[HTML]{F1F1F1} 0.69 &
+        {\cellcolor[HTML]{FCA108}} \color[HTML]{000000} 0.79 & 1.9 & 3.5 & 14.1 &
+        19.8 & \color[HTML]{FFDD33} \bfseries HOLD \\
+        \cline{1-11}
+        \multirow[c]{2}{*}{Consumer} & H\&M & {\cellcolor[HTML]{EB6628}}
+        \color[HTML]{F1F1F1} 0.66 & {\cellcolor[HTML]{F1731D}} \color[HTML]{F1F1F1}
+        0.69 & {\cellcolor[HTML]{FCFFA4}} \color[HTML]{000000} 1.00 &
+        {\cellcolor[HTML]{FAC42A}} \color[HTML]{000000} 0.86 & 7.0 & 49.0 & 210.9 &
+        140.6 & \color[HTML]{33FF85} \bfseries BUY \\
+         & Unilever & {\cellcolor[HTML]{F68013}} \color[HTML]{F1F1F1} 0.72 &
+        {\cellcolor[HTML]{FCA108}} \color[HTML]{000000} 0.79 &
+        {\cellcolor[HTML]{FAC42A}} \color[HTML]{000000} 0.86 &
+        {\cellcolor[HTML]{FCFFA4}} \color[HTML]{000000} 1.00 & 213.8 & 45,693.3 &
+        2,807.0 & 3,678.0 & \color[HTML]{FF5933} \bfseries SELL \\
+        \cline{1-11}
+        \bottomrule
+        \end{tabular}
+        \end{table}
+
+        .. figure:: ../../_static/style/latex_stocks.png
+        """
+        obj = self._copy(deepcopy=True)  # manipulate table_styles on obj, not self
+
+        table_selectors = (
+            [style["selector"] for style in self.table_styles]
+            if self.table_styles is not None
+            else []
+        )
+
+        if column_format is not None:
+            # add more recent setting to table_styles
+            obj.set_table_styles(
+                [{"selector": "column_format", "props": f":{column_format}"}],
+                overwrite=False,
+            )
+        elif "column_format" in table_selectors:
+            pass  # adopt what has been previously set in table_styles
+        else:
+            # create a default: set float, complex, int cols to 'r' ('S'), index to 'l'
+            _original_columns = self.data.columns
+            self.data.columns = RangeIndex(stop=len(self.data.columns))
+            numeric_cols = self.data._get_numeric_data().columns.to_list()
+            self.data.columns = _original_columns
+            column_format = ""
+            for level in range(self.index.nlevels):
+                column_format += "" if self.hide_index_[level] else "l"
+            for ci, _ in enumerate(self.data.columns):
+                if ci not in self.hidden_columns:
+                    column_format += (
+                        ("r" if not siunitx else "S") if ci in numeric_cols else "l"
+                    )
+            obj.set_table_styles(
+                [{"selector": "column_format", "props": f":{column_format}"}],
+                overwrite=False,
+            )
+
+        if position:
+            obj.set_table_styles(
+                [{"selector": "position", "props": f":{position}"}],
+                overwrite=False,
+            )
+
+        if position_float:
+            if environment == "longtable":
+                raise ValueError(
+                    "`position_float` cannot be used in 'longtable' `environment`"
+                )
+            if position_float not in ["raggedright", "raggedleft", "centering"]:
+                raise ValueError(
+                    f"`position_float` should be one of "
+                    f"'raggedright', 'raggedleft', 'centering', "
+                    f"got: '{position_float}'"
+                )
+            obj.set_table_styles(
+                [{"selector": "position_float", "props": f":{position_float}"}],
+                overwrite=False,
+            )
+
+        hrules = get_option("styler.latex.hrules") if hrules is None else hrules
+        if hrules:
+            obj.set_table_styles(
+                [
+                    {"selector": "toprule", "props": ":toprule"},
+                    {"selector": "midrule", "props": ":midrule"},
+                    {"selector": "bottomrule", "props": ":bottomrule"},
+                ],
+                overwrite=False,
+            )
+
+        if label:
+            obj.set_table_styles(
+                [{"selector": "label", "props": f":{{{label.replace(':', '§')}}}"}],
+                overwrite=False,
+            )
+
+        if caption:
+            obj.set_caption(caption)
+
+        if sparse_index is None:
+            sparse_index = get_option("styler.sparse.index")
+        if sparse_columns is None:
+            sparse_columns = get_option("styler.sparse.columns")
+        environment = environment or get_option("styler.latex.environment")
+        multicol_align = multicol_align or get_option("styler.latex.multicol_align")
+        multirow_align = multirow_align or get_option("styler.latex.multirow_align")
+        latex = obj._render_latex(
+            sparse_index=sparse_index,
+            sparse_columns=sparse_columns,
+            multirow_align=multirow_align,
+            multicol_align=multicol_align,
+            environment=environment,
+            convert_css=convert_css,
+            siunitx=siunitx,
+            clines=clines,
+        )
+
+        encoding = (
+            (encoding or get_option("styler.render.encoding"))
+            if isinstance(buf, str)  # i.e. a filepath
+            else encoding
+        )
+        return save_to_buffer(latex, buf=buf, encoding=encoding)
+
+    @overload
+    def to_typst(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        encoding: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+    ) -> None: ...
+
+    @overload
+    def to_typst(
+        self,
+        buf: None = ...,
+        *,
+        encoding: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+    ) -> str: ...
+
+    @Substitution(buf=buffering_args, encoding=encoding_args)
+    def to_typst(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        encoding: str | None = None,
+        sparse_index: bool | None = None,
+        sparse_columns: bool | None = None,
+        max_rows: int | None = None,
+        max_columns: int | None = None,
+    ) -> str | None:
+        """
+        Write Styler to a file, buffer or string in Typst format.
+
+        .. versionadded:: 3.0.0
+
+        Parameters
+        ----------
+        %(buf)s
+        %(encoding)s
+        sparse_index : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each row.
+            Defaults to ``pandas.options.styler.sparse.index`` value.
+        sparse_columns : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each
+            column. Defaults to ``pandas.options.styler.sparse.columns`` value.
+        max_rows : int, optional
+            The maximum number of rows that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_rows``, which is None.
+        max_columns : int, optional
+            The maximum number of columns that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_columns``, which is None.
+
+            Rows and columns may be reduced if the number of total elements is
+            large. This value is set to ``pandas.options.styler.render.max_elements``,
+            which is 262144 (18 bit browser rendering).
+
+        Returns
+        -------
+        str or None
+            If `buf` is None, returns the result as a string. Otherwise returns `None`.
+
+        See Also
+        --------
+        DataFrame.to_typst : Write a DataFrame to a file,
+            buffer or string in Typst format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.style.to_typst()  # doctest: +SKIP
+
+        .. code-block:: typst
+
+            #table(
+              columns: 3,
+              [], [A], [B],
+
+              [0], [1], [3],
+              [1], [2], [4],
+            )
+        """
+        obj = self._copy(deepcopy=True)
+
+        if sparse_index is None:
+            sparse_index = get_option("styler.sparse.index")
+        if sparse_columns is None:
+            sparse_columns = get_option("styler.sparse.columns")
+
+        text = obj._render_typst(
+            sparse_columns=sparse_columns,
+            sparse_index=sparse_index,
+            max_rows=max_rows,
+            max_cols=max_columns,
+        )
+        return save_to_buffer(
+            text, buf=buf, encoding=(encoding if buf is not None else None)
+        )
+
+    @overload
+    def to_html(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        table_uuid: str | None = ...,
+        table_attributes: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        bold_headers: bool = ...,
+        caption: str | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+        encoding: str | None = ...,
+        doctype_html: bool = ...,
+        exclude_styles: bool = ...,
+        **kwargs,
+    ) -> None: ...
+
+    @overload
+    def to_html(
+        self,
+        buf: None = ...,
+        *,
+        table_uuid: str | None = ...,
+        table_attributes: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        bold_headers: bool = ...,
+        caption: str | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+        encoding: str | None = ...,
+        doctype_html: bool = ...,
+        exclude_styles: bool = ...,
+        **kwargs,
+    ) -> str: ...
+
+    @Substitution(buf=buffering_args, encoding=encoding_args)
+    def to_html(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        table_uuid: str | None = None,
+        table_attributes: str | None = None,
+        sparse_index: bool | None = None,
+        sparse_columns: bool | None = None,
+        bold_headers: bool = False,
+        caption: str | None = None,
+        max_rows: int | None = None,
+        max_columns: int | None = None,
+        encoding: str | None = None,
+        doctype_html: bool = False,
+        exclude_styles: bool = False,
+        **kwargs,
+    ) -> str | None:
+        """
+        Write Styler to a file, buffer or string in HTML-CSS format.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        %(buf)s
+        table_uuid : str, optional
+            Id attribute assigned to the <table> HTML element in the format:
+
+            ``<table id="T_<table_uuid>" ..>``
+
+            If not given uses Styler's initially assigned value.
+        table_attributes : str, optional
+            Attributes to assign within the `<table>` HTML element in the format:
+
+            ``<table .. <table_attributes> >``
+
+            If not given defaults to Styler's preexisting value.
+        sparse_index : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each row.
+            Defaults to ``pandas.options.styler.sparse.index`` value.
+
+            .. versionadded:: 1.4.0
+        sparse_columns : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each
+            column. Defaults to ``pandas.options.styler.sparse.columns`` value.
+
+            .. versionadded:: 1.4.0
+        bold_headers : bool, optional
+            Adds "font-weight: bold;" as a CSS property to table style header cells.
+
+            .. versionadded:: 1.4.0
+        caption : str, optional
+            Set, or overwrite, the caption on Styler before rendering.
+
+            .. versionadded:: 1.4.0
+        max_rows : int, optional
+            The maximum number of rows that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_rows/max_columns``.
+
+            .. versionadded:: 1.4.0
+        max_columns : int, optional
+            The maximum number of columns that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_columns``, which is None.
+
+            Rows and columns may be reduced if the number of total elements is
+            large. This value is set to ``pandas.options.styler.render.max_elements``,
+            which is 262144 (18 bit browser rendering).
+
+            .. versionadded:: 1.4.0
+        %(encoding)s
+        doctype_html : bool, default False
+            Whether to output a fully structured HTML file including all
+            HTML elements, or just the core ``<style>`` and ``<table>`` elements.
+        exclude_styles : bool, default False
+            Whether to include the ``<style>`` element and all associated element
+            ``class`` and ``id`` identifiers, or solely the ``<table>`` element without
+            styling identifiers.
+        **kwargs
+            Any additional keyword arguments are passed through to the jinja2
+            ``self.template.render`` process. This is useful when you need to provide
+            additional variables for a custom template.
+
+        Returns
+        -------
+        str or None
+            If `buf` is None, returns the result as a string. Otherwise returns `None`.
+
+        See Also
+        --------
+        DataFrame.to_html: Write a DataFrame to a file, buffer or string in HTML format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> print(df.style.to_html())  # doctest: +SKIP
+        <style type="text/css">
+        </style>
+        <table id="T_1e78e">
+          <thead>
+            <tr>
+              <th class="blank level0" >&nbsp;</th>
+              <th id="T_1e78e_level0_col0" class="col_heading level0 col0" >A</th>
+              <th id="T_1e78e_level0_col1" class="col_heading level0 col1" >B</th>
+            </tr>
+        ...
+        """
+        obj = self._copy(deepcopy=True)  # manipulate table_styles on obj, not self
+
+        if table_uuid:
+            obj.set_uuid(table_uuid)
+
+        if table_attributes:
+            obj.set_table_attributes(table_attributes)
+
+        if sparse_index is None:
+            sparse_index = get_option("styler.sparse.index")
+        if sparse_columns is None:
+            sparse_columns = get_option("styler.sparse.columns")
+
+        if bold_headers:
+            obj.set_table_styles(
+                [{"selector": "th", "props": "font-weight: bold;"}], overwrite=False
+            )
+
+        if caption is not None:
+            obj.set_caption(caption)
+
+        # Build HTML string..
+        html = obj._render_html(
+            sparse_index=sparse_index,
+            sparse_columns=sparse_columns,
+            max_rows=max_rows,
+            max_cols=max_columns,
+            exclude_styles=exclude_styles,
+            encoding=encoding or get_option("styler.render.encoding"),
+            doctype_html=doctype_html,
+            **kwargs,
+        )
+
+        return save_to_buffer(
+            html, buf=buf, encoding=(encoding if buf is not None else None)
+        )
+
+    @overload
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str],
+        *,
+        encoding: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+        delimiter: str = ...,
+    ) -> None: ...
+
+    @overload
+    def to_string(
+        self,
+        buf: None = ...,
+        *,
+        encoding: str | None = ...,
+        sparse_index: bool | None = ...,
+        sparse_columns: bool | None = ...,
+        max_rows: int | None = ...,
+        max_columns: int | None = ...,
+        delimiter: str = ...,
+    ) -> str: ...
+
+    @Substitution(buf=buffering_args, encoding=encoding_args)
+    def to_string(
+        self,
+        buf: FilePath | WriteBuffer[str] | None = None,
+        *,
+        encoding: str | None = None,
+        sparse_index: bool | None = None,
+        sparse_columns: bool | None = None,
+        max_rows: int | None = None,
+        max_columns: int | None = None,
+        delimiter: str = " ",
+    ) -> str | None:
+        """
+        Write Styler to a file, buffer or string in text format.
+
+        .. versionadded:: 1.5.0
+
+        Parameters
+        ----------
+        %(buf)s
+        %(encoding)s
+        sparse_index : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each row.
+            Defaults to ``pandas.options.styler.sparse.index`` value.
+        sparse_columns : bool, optional
+            Whether to sparsify the display of a hierarchical index. Setting to False
+            will display each explicit level element in a hierarchical key for each
+            column. Defaults to ``pandas.options.styler.sparse.columns`` value.
+        max_rows : int, optional
+            The maximum number of rows that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_rows``, which is None.
+        max_columns : int, optional
+            The maximum number of columns that will be rendered. Defaults to
+            ``pandas.options.styler.render.max_columns``, which is None.
+
+            Rows and columns may be reduced if the number of total elements is
+            large. This value is set to ``pandas.options.styler.render.max_elements``,
+            which is 262144 (18 bit browser rendering).
+        delimiter : str, default single space
+            The separator between data elements.
+
+        Returns
+        -------
+        str or None
+            If `buf` is None, returns the result as a string. Otherwise returns `None`.
+
+        See Also
+        --------
+        DataFrame.to_string : Render a DataFrame to a console-friendly tabular output.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.style.to_string()
+        ' A B\\n0 1 3\\n1 2 4\\n'
+        """
+        obj = self._copy(deepcopy=True)
+
+        if sparse_index is None:
+            sparse_index = get_option("styler.sparse.index")
+        if sparse_columns is None:
+            sparse_columns = get_option("styler.sparse.columns")
+
+        text = obj._render_string(
+            sparse_columns=sparse_columns,
+            sparse_index=sparse_index,
+            max_rows=max_rows,
+            max_cols=max_columns,
+            delimiter=delimiter,
+        )
+        return save_to_buffer(
+            text, buf=buf, encoding=(encoding if buf is not None else None)
+        )
+
+    def set_td_classes(self, classes: DataFrame) -> Styler:
+        """
+        Set the ``class`` attribute of ``<td>`` HTML elements.
+
+        Parameters
+        ----------
+        classes : DataFrame
+            DataFrame containing strings that will be translated to CSS classes,
+            mapped by identical column and index key values that must exist on the
+            underlying Styler data. None, NaN values, and empty strings will
+            be ignored and not affect the rendered HTML.
+
+        Returns
+        -------
+        Styler
+            Instance of class with ``class`` attribute set for ``<td>``
+                HTML elements.
+
+        See Also
+        --------
+        Styler.set_table_styles: Set the table styles included within the ``<style>``
+            HTML element.
+        Styler.set_table_attributes: Set the table attributes added to the ``<table>``
+            HTML element.
+
+        Notes
+        -----
+        Can be used in combination with ``Styler.set_table_styles`` to define an
+        internal CSS solution without reference to external CSS files.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(data=[[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+        >>> classes = pd.DataFrame(
+        ...     [["min-val red", "", "blue"], ["red", None, "blue max-val"]],
+        ...     index=df.index,
+        ...     columns=df.columns,
+        ... )
+        >>> df.style.set_td_classes(classes)  # doctest: +SKIP
+
+        Using `MultiIndex` columns and a `classes` `DataFrame` as a subset of the
+        underlying,
+
+        >>> df = pd.DataFrame(
+        ...     [[1, 2], [3, 4]],
+        ...     index=["a", "b"],
+        ...     columns=[["level0", "level0"], ["level1a", "level1b"]],
+        ... )
+        >>> classes = pd.DataFrame(
+        ...     ["min-val"], index=["a"], columns=[["level0"], ["level1a"]]
+        ... )
+        >>> df.style.set_td_classes(classes)  # doctest: +SKIP
+
+        Form of the output with new additional css classes,
+
+        >>> from pandas.io.formats.style import Styler
+        >>> df = pd.DataFrame([[1]])
+        >>> css = pd.DataFrame([["other-class"]])
+        >>> s = Styler(df, uuid="_", cell_ids=False).set_td_classes(css)
+        >>> s.hide(axis=0).to_html()  # doctest: +SKIP
+        '<style type="text/css"></style>'
+        '<table id="T__">'
+        '  <thead>'
+        '    <tr><th class="col_heading level0 col0" >0</th></tr>'
+        '  </thead>'
+        '  <tbody>'
+        '    <tr><td class="data row0 col0 other-class" >1</td></tr>'
+        '  </tbody>'
+        '</table>'
+        """
+        if not classes.index.is_unique or not classes.columns.is_unique:
+            raise KeyError(
+                "Classes render only if `classes` has unique index and columns."
+            )
+        classes = classes.reindex_like(self.data)
+
+        for r, row_tup in enumerate(classes.itertuples()):
+            for c, value in enumerate(row_tup[1:]):
+                if not (pd.isna(value) or value == ""):
+                    self.cell_context[(r, c)] = str(value)
+
+        return self
+
+    def _update_ctx(self, attrs: DataFrame) -> None:
+        """
+        Update the state of the ``Styler`` for data cells.
+
+        Collects a mapping of {index_label: [('<property>', '<value>'), ..]}.
+
+        Parameters
+        ----------
+        attrs : DataFrame
+            should contain strings of '<property>: <value>;<prop2>: <val2>'
+            Whitespace shouldn't matter and the final trailing ';' shouldn't
+            matter.
+        """
+        if not self.index.is_unique or not self.columns.is_unique:
+            raise KeyError(
+                "`Styler.apply` and `.map` are not compatible "
+                "with non-unique index or columns."
+            )
+
+        for cn in attrs.columns:
+            j = self.columns.get_loc(cn)
+            ser = attrs[cn]
+            for rn, c in ser.items():
+                if not c or pd.isna(c):
+                    continue
+                css_list = maybe_convert_css_to_tuples(c)
+                i = self.index.get_loc(rn)
+                self.ctx[(i, j)].extend(css_list)
+
+    def _update_ctx_header(self, attrs: DataFrame, axis: AxisInt) -> None:
+        """
+        Update the state of the ``Styler`` for header cells.
+
+        Collects a mapping of {index_label: [('<property>', '<value>'), ..]}.
+
+        Parameters
+        ----------
+        attrs : Series
+            Should contain strings of '<property>: <value>;<prop2>: <val2>', and an
+            integer index.
+            Whitespace shouldn't matter and the final trailing ';' shouldn't
+            matter.
+        axis : int
+            Identifies whether the ctx object being updated is the index or columns
+        """
+        for j in attrs.columns:
+            ser = attrs[j]
+            for i, c in ser.items():
+                if not c or pd.isna(c):
+                    continue
+                css_list = maybe_convert_css_to_tuples(c)
+                if axis == 0:
+                    self.ctx_index[(i, j)].extend(css_list)
+                else:
+                    self.ctx_columns[(j, i)].extend(css_list)
+
+    def _copy(self, deepcopy: bool = False) -> Styler:
+        """
+        Copies a Styler, allowing for deepcopy or shallow copy
+
+        Copying a Styler aims to recreate a new Styler object which contains the same
+        data and styles as the original.
+
+        Data dependent attributes [copied and NOT exported]:
+          - formatting (._display_funcs)
+          - hidden index values or column values (.hidden_rows, .hidden_columns)
+          - tooltips
+          - cell_context (cell css classes)
+          - ctx (cell css styles)
+          - caption
+          - concatenated stylers
+
+        Non-data dependent attributes [copied and exported]:
+          - css
+          - hidden index state and hidden columns state (.hide_index_, .hide_columns_)
+          - table_attributes
+          - table_styles
+          - applied styles (_todo)
+
+        """
+        # GH 40675, 52728
+        styler = type(self)(
+            self.data,  # populates attributes 'data', 'columns', 'index' as shallow
+        )
+        shallow = [  # simple string or boolean immutables
+            "hide_index_",
+            "hide_columns_",
+            "hide_column_names",
+            "hide_index_names",
+            "table_attributes",
+            "cell_ids",
+            "caption",
+            "uuid",
+            "uuid_len",
+            "template_latex",  # also copy templates if these have been customised
+            "template_html_style",
+            "template_html_table",
+            "template_html",
+        ]
+        deep = [  # nested lists or dicts
+            "css",
+            "concatenated",
+            "_display_funcs",
+            "_display_funcs_index",
+            "_display_funcs_columns",
+            "_display_funcs_index_names",
+            "_display_funcs_column_names",
+            "hidden_rows",
+            "hidden_columns",
+            "ctx",
+            "ctx_index",
+            "ctx_columns",
+            "cell_context",
+            "_todo",
+            "table_styles",
+            "tooltips",
+        ]
+
+        for attr in shallow:
+            setattr(styler, attr, getattr(self, attr))
+
+        for attr in deep:
+            val = getattr(self, attr)
+            setattr(styler, attr, copy.deepcopy(val) if deepcopy else val)
+
+        return styler
+
+    def __copy__(self) -> Styler:
+        return self._copy(deepcopy=False)
+
+    def __deepcopy__(self, memo) -> Styler:
+        return self._copy(deepcopy=True)
+
+    def clear(self) -> None:
+        """
+        Reset the ``Styler``, removing any previously applied styles.
+
+        Returns None.
+
+        See Also
+        --------
+        Styler.apply : Apply a CSS-styling function column-wise, row-wise,
+            or table-wise.
+        Styler.export : Export the styles applied to the current Styler.
+        Styler.map : Apply a CSS-styling function elementwise.
+        Styler.use : Set the styles on the current Styler.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, np.nan]})
+
+        After any added style:
+
+        >>> df.style.highlight_null(color="yellow")  # doctest: +SKIP
+
+        Remove it with:
+
+        >>> df.style.clear()  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        # create default GH 40675
+        clean_copy = Styler(self.data, uuid=self.uuid)
+        clean_attrs = [a for a in clean_copy.__dict__ if not callable(a)]
+        self_attrs = [a for a in self.__dict__ if not callable(a)]  # maybe more attrs
+        for attr in clean_attrs:
+            setattr(self, attr, getattr(clean_copy, attr))
+        for attr in set(self_attrs).difference(clean_attrs):
+            delattr(self, attr)
+
+    def _apply(
+        self,
+        func: Callable,
+        axis: Axis | None = 0,
+        subset: Subset | None = None,
+        **kwargs,
+    ) -> Styler:
+        subset = slice(None) if subset is None else subset
+        subset = non_reducing_slice(subset)
+        data = self.data.loc[subset]
+        if data.empty:
+            result = DataFrame()
+        elif axis is None:
+            result = func(data, **kwargs)
+            if not isinstance(result, DataFrame):
+                if not isinstance(result, np.ndarray):
+                    raise TypeError(
+                        f"Function {func!r} must return a DataFrame or ndarray "
+                        f"when passed to `Styler.apply` with axis=None"
+                    )
+                if data.shape != result.shape:
+                    raise ValueError(
+                        f"Function {func!r} returned ndarray with wrong shape.\n"
+                        f"Result has shape: {result.shape}\n"
+                        f"Expected shape: {data.shape}"
+                    )
+                result = DataFrame(result, index=data.index, columns=data.columns)
+        else:
+            axis = self.data._get_axis_number(axis)
+            if axis == 0:
+                result = data.apply(func, axis=0, **kwargs)
+            else:
+                result = data.T.apply(func, axis=0, **kwargs).T  # see GH 42005
+
+        if isinstance(result, Series):
+            raise ValueError(
+                f"Function {func!r} resulted in the apply method collapsing to a "
+                f"Series.\nUsually, this is the result of the function returning a "
+                f"single value, instead of list-like."
+            )
+        msg = (
+            f"Function {func!r} created invalid {{0}} labels.\nUsually, this is "
+            f"the result of the function returning a "
+            f"{'Series' if axis is not None else 'DataFrame'} which contains invalid "
+            f"labels, or returning an incorrectly shaped, list-like object which "
+            f"cannot be mapped to labels, possibly due to applying the function along "
+            f"the wrong axis.\n"
+            f"Result {{0}} has shape: {{1}}\n"
+            f"Expected {{0}} shape:   {{2}}"
+        )
+        if not all(result.index.isin(data.index)):
+            raise ValueError(msg.format("index", result.index.shape, data.index.shape))
+        if not all(result.columns.isin(data.columns)):
+            raise ValueError(
+                msg.format("columns", result.columns.shape, data.columns.shape)
+            )
+        self._update_ctx(result)
+        return self
+
+    @Substitution(subset=subset_args)
+    def apply(
+        self,
+        func: Callable,
+        axis: Axis | None = 0,
+        subset: Subset | None = None,
+        **kwargs,
+    ) -> Styler:
+        """
+        Apply a CSS-styling function column-wise, row-wise, or table-wise.
+
+        Updates the HTML representation with the result.
+
+        Parameters
+        ----------
+        func : function
+            ``func`` should take a Series if ``axis`` in [0,1] and return a list-like
+            object of same length, or a Series, not necessarily of same length, with
+            valid index labels considering ``subset``.
+            ``func`` should take a DataFrame if ``axis`` is ``None`` and return either
+            an ndarray with the same shape or a DataFrame, not necessarily of the same
+            shape, with valid index and columns labels considering ``subset``.
+
+            .. versionchanged:: 1.3.0
+
+            .. versionchanged:: 1.4.0
+
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``), or to the entire DataFrame at once
+            with ``axis=None``.
+        %(subset)s
+        **kwargs : dict
+            Pass along to ``func``.
+
+        Returns
+        -------
+        Styler
+            Instance of class with CSS applied to its HTML representation.
+
+        See Also
+        --------
+        Styler.map_index: Apply a CSS-styling function to headers elementwise.
+        Styler.apply_index: Apply a CSS-styling function to headers level-wise.
+        Styler.map: Apply a CSS-styling function elementwise.
+
+        Notes
+        -----
+        The elements of the output of ``func`` should be CSS styles as strings, in the
+        format 'attribute: value; attribute2: value2; ...' or,
+        if nothing is to be applied to that element, an empty string or ``None``.
+
+        This is similar to ``DataFrame.apply``, except that ``axis=None``
+        applies the function to the entire DataFrame at once,
+        rather than column-wise or row-wise.
+
+        Examples
+        --------
+        >>> def highlight_max(x, color):
+        ...     return np.where(x == np.nanmax(x.to_numpy()), f"color: {color};", None)
+        >>> df = pd.DataFrame(np.random.randn(5, 2), columns=["A", "B"])
+        >>> df.style.apply(highlight_max, color="red")  # doctest: +SKIP
+        >>> df.style.apply(highlight_max, color="blue", axis=1)  # doctest: +SKIP
+        >>> df.style.apply(highlight_max, color="green", axis=None)  # doctest: +SKIP
+
+        Using ``subset`` to restrict application to a single column or multiple columns
+
+        >>> df.style.apply(highlight_max, color="red", subset="A")
+        ... # doctest: +SKIP
+        >>> df.style.apply(highlight_max, color="red", subset=["A", "B"])
+        ... # doctest: +SKIP
+
+        Using a 2d input to ``subset`` to select rows in addition to columns
+
+        >>> df.style.apply(highlight_max, color="red", subset=([0, 1, 2], slice(None)))
+        ... # doctest: +SKIP
+        >>> df.style.apply(highlight_max, color="red", subset=(slice(0, 5, 2), "A"))
+        ... # doctest: +SKIP
+
+        Using a function which returns a Series / DataFrame of unequal length but
+        containing valid index labels
+
+        >>> df = pd.DataFrame([[1, 2], [3, 4], [4, 6]], index=["A1", "A2", "Total"])
+        >>> total_style = pd.Series("font-weight: bold;", index=["Total"])
+        >>> df.style.apply(lambda s: total_style)  # doctest: +SKIP
+
+        See `Table Visualization <../../user_guide/style.ipynb>`_ user guide for
+        more details.
+        """
+        self._todo.append(
+            (lambda instance: instance._apply, (func, axis, subset), kwargs)
+        )
+        return self
+
+    def _apply_index(
+        self,
+        func: Callable,
+        axis: Axis = 0,
+        level: Level | list[Level] | None = None,
+        method: str = "apply",
+        **kwargs,
+    ) -> Styler:
+        axis = self.data._get_axis_number(axis)
+        obj = self.index if axis == 0 else self.columns
+
+        levels_ = refactor_levels(level, obj)
+        data = DataFrame(obj.to_list()).loc[:, levels_]
+
+        if method == "apply":
+            result = data.apply(func, axis=0, **kwargs)
+        elif method == "map":
+            result = data.map(func, **kwargs)
+
+        self._update_ctx_header(result, axis)
+        return self
+
+    @doc(
+        this="apply",
+        wise="level-wise",
+        alt="map",
+        altwise="elementwise",
+        func="take a Series and return a string array of the same length",
+        input_note="the index as a Series, if an Index, or a level of a MultiIndex",
+        output_note="an identically sized array of CSS styles as strings",
+        var="label",
+        ret='np.where(label == "B", "background-color: yellow;", "")',
+        ret2='["background-color: yellow;" if "x" in v else "" for v in label]',
+    )
+    def apply_index(
+        self,
+        func: Callable,
+        axis: AxisInt | str = 0,
+        level: Level | list[Level] | None = None,
+        **kwargs,
+    ) -> Styler:
+        """
+        Apply a CSS-styling function to the index or column headers, {wise}.
+
+        Updates the HTML representation with the result.
+
+        .. versionadded:: 1.4.0
+
+        .. versionadded:: 2.1.0
+           Styler.applymap_index was deprecated and renamed to Styler.map_index.
+
+        Parameters
+        ----------
+        func : function
+            ``func`` should {func}.
+        axis : {{0, 1, "index", "columns"}}
+            The headers over which to apply the function.
+        level : int, str, list, optional
+            If index is MultiIndex the level(s) over which to apply the function.
+        **kwargs : dict
+            Pass along to ``func``.
+
+        Returns
+        -------
+        Styler
+            Instance of class with CSS applied to its HTML representation.
+
+        See Also
+        --------
+        Styler.{alt}_index: Apply a CSS-styling function to headers {altwise}.
+        Styler.apply: Apply a CSS-styling function column-wise, row-wise, or table-wise.
+        Styler.map: Apply a CSS-styling function elementwise.
+
+        Notes
+        -----
+        Each input to ``func`` will be {input_note}. The output of ``func`` should be
+        {output_note}, in the format 'attribute: value; attribute2: value2; ...'
+        or, if nothing is to be applied to that element, an empty string or ``None``.
+
+        Examples
+        --------
+        Basic usage to conditionally highlight values in the index.
+
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], index=["A", "B"])
+        >>> def color_b({var}):
+        ...     return {ret}
+        >>> df.style.{this}_index(color_b)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/appmaphead1.png
+
+        Selectively applying to specific levels of MultiIndex columns.
+
+        >>> midx = pd.MultiIndex.from_product([["ix", "jy"], [0, 1], ["x3", "z4"]])
+        >>> df = pd.DataFrame([np.arange(8)], columns=midx)
+        >>> def highlight_x({var}):
+        ...     return {ret2}
+        >>> df.style.{this}_index(
+        ...     highlight_x, axis="columns", level=[0, 2])  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/appmaphead2.png
+        """
+        self._todo.append(
+            (
+                lambda instance: instance._apply_index,
+                (func, axis, level, "apply"),
+                kwargs,
+            )
+        )
+        return self
+
+    @doc(
+        apply_index,
+        this="map",
+        wise="elementwise",
+        alt="apply",
+        altwise="level-wise",
+        func="take a scalar and return a string",
+        input_note="an index value, if an Index, or a level value of a MultiIndex",
+        output_note="CSS styles as a string",
+        var="label",
+        ret='"background-color: yellow;" if label == "B" else None',
+        ret2='"background-color: yellow;" if "x" in label else None',
+    )
+    def map_index(
+        self,
+        func: Callable,
+        axis: AxisInt | str = 0,
+        level: Level | list[Level] | None = None,
+        **kwargs,
+    ) -> Styler:
+        self._todo.append(
+            (
+                lambda instance: instance._apply_index,
+                (func, axis, level, "map"),
+                kwargs,
+            )
+        )
+        return self
+
+    def _map(self, func: Callable, subset: Subset | None = None, **kwargs) -> Styler:
+        func = partial(func, **kwargs)  # map doesn't take kwargs?
+        if subset is None:
+            subset = IndexSlice[:]
+        subset = non_reducing_slice(subset)
+        result = self.data.loc[subset].map(func)
+        self._update_ctx(result)
+        return self
+
+    @Substitution(subset=subset_args)
+    def map(self, func: Callable, subset: Subset | None = None, **kwargs) -> Styler:
+        """
+        Apply a CSS-styling function elementwise.
+
+        Updates the HTML representation with the result.
+
+        Parameters
+        ----------
+        func : function
+            ``func`` should take a scalar and return a string.
+        %(subset)s
+        **kwargs : dict
+            Pass along to ``func``.
+
+        Returns
+        -------
+        Styler
+            Instance of class with CSS-styling function applied elementwise.
+
+        See Also
+        --------
+        Styler.map_index: Apply a CSS-styling function to headers elementwise.
+        Styler.apply_index: Apply a CSS-styling function to headers level-wise.
+        Styler.apply: Apply a CSS-styling function column-wise, row-wise, or table-wise.
+
+        Notes
+        -----
+        The elements of the output of ``func`` should be CSS styles as strings, in the
+        format 'attribute: value; attribute2: value2; ...' or,
+        if nothing is to be applied to that element, an empty string or ``None``.
+
+        Examples
+        --------
+        >>> def color_negative(v, color):
+        ...     return f"color: {color};" if v < 0 else None
+        >>> df = pd.DataFrame(np.random.randn(5, 2), columns=["A", "B"])
+        >>> df.style.map(color_negative, color="red")  # doctest: +SKIP
+
+        Using ``subset`` to restrict application to a single column or multiple columns
+
+        >>> df.style.map(color_negative, color="red", subset="A")
+        ... # doctest: +SKIP
+        >>> df.style.map(color_negative, color="red", subset=["A", "B"])
+        ... # doctest: +SKIP
+
+        Using a 2d input to ``subset`` to select rows in addition to columns
+
+        >>> df.style.map(
+        ...     color_negative, color="red", subset=([0, 1, 2], slice(None))
+        ... )  # doctest: +SKIP
+        >>> df.style.map(color_negative, color="red", subset=(slice(0, 5, 2), "A"))
+        ... # doctest: +SKIP
+
+        See `Table Visualization <../../user_guide/style.ipynb>`_ user guide for
+        more details.
+        """
+        self._todo.append((lambda instance: instance._map, (func, subset), kwargs))
+        return self
+
+    def set_table_attributes(self, attributes: str) -> Styler:
+        """
+        Set the table attributes added to the ``<table>`` HTML element.
+
+        These are items in addition to automatic (by default) ``id`` attribute.
+
+        Parameters
+        ----------
+        attributes : str
+            Table attributes to be added to the ``<table>`` HTML element.
+
+        Returns
+        -------
+        Styler
+            Instance of class with specified table attributes set.
+
+        See Also
+        --------
+        Styler.set_table_styles: Set the table styles included within the ``<style>``
+            HTML element.
+        Styler.set_td_classes: Set the DataFrame of strings added to the ``class``
+            attribute of ``<td>`` HTML elements.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(np.random.randn(10, 4))
+        >>> df.style.set_table_attributes('class="pure-table"')  # doctest: +SKIP
+        # ... <table class="pure-table"> ...
+        """
+        self.table_attributes = attributes
+        return self
+
+    def export(self) -> dict[str, Any]:
+        """
+        Export the styles applied to the current Styler.
+
+        Can be applied to a second Styler with ``Styler.use``.
+
+        Returns
+        -------
+        dict
+            Contains data-independent (exportable) styles applied to current Styler.
+
+        See Also
+        --------
+        Styler.use: Set the styles on the current Styler.
+        Styler.copy: Create a copy of the current Styler.
+
+        Notes
+        -----
+        This method is designed to copy non-data dependent attributes of
+        one Styler to another. It differs from ``Styler.copy`` where data and
+        data dependent attributes are also copied.
+
+        The following items are exported since they are not generally data dependent:
+
+          - Styling functions added by the ``apply`` and ``map``
+          - Whether axes and names are hidden from the display, if unambiguous.
+          - Table attributes
+          - Table styles
+
+        The following attributes are considered data dependent and therefore not
+        exported:
+
+          - Caption
+          - UUID
+          - Tooltips
+          - Any hidden rows or columns identified by Index labels
+          - Any formatting applied using ``Styler.format``
+          - Any CSS classes added using ``Styler.set_td_classes``
+
+        Examples
+        --------
+
+        >>> styler = pd.DataFrame([[1, 2], [3, 4]]).style
+        >>> styler2 = pd.DataFrame([[9, 9, 9]]).style
+        >>> styler.hide(axis=0).highlight_max(axis=1)  # doctest: +SKIP
+        >>> export = styler.export()
+        >>> styler2.use(export)  # doctest: +SKIP
+        """
+        return {
+            "apply": copy.copy(self._todo),
+            "table_attributes": self.table_attributes,
+            "table_styles": copy.copy(self.table_styles),
+            "hide_index": all(self.hide_index_),
+            "hide_columns": all(self.hide_columns_),
+            "hide_index_names": self.hide_index_names,
+            "hide_column_names": self.hide_column_names,
+            "css": copy.copy(self.css),
+        }
+
+    def use(self, styles: dict[str, Any]) -> Styler:
+        """
+        Set the styles on the current Styler.
+
+        Possibly uses styles from ``Styler.export``.
+
+        Parameters
+        ----------
+        styles : dict(str, Any)
+            List of attributes to add to Styler. Dict keys should contain only:
+              - "apply": list of styler functions, typically added with ``apply`` or
+                ``map``.
+              - "table_attributes": HTML attributes, typically added with
+                ``set_table_attributes``.
+              - "table_styles": CSS selectors and properties, typically added with
+                ``set_table_styles``.
+              - "hide_index":  whether the index is hidden, typically added with
+                ``hide_index``, or a boolean list for hidden levels.
+              - "hide_columns": whether column headers are hidden, typically added with
+                ``hide_columns``, or a boolean list for hidden levels.
+              - "hide_index_names": whether index names are hidden.
+              - "hide_column_names": whether column header names are hidden.
+              - "css": the css class names used.
+
+        Returns
+        -------
+        Styler
+            Instance of class with defined styler attributes added.
+
+        See Also
+        --------
+        Styler.export : Export the non data dependent attributes to the current Styler.
+
+        Examples
+        --------
+
+        >>> styler = pd.DataFrame([[1, 2], [3, 4]]).style
+        >>> styler2 = pd.DataFrame([[9, 9, 9]]).style
+        >>> styler.hide(axis=0).highlight_max(axis=1)  # doctest: +SKIP
+        >>> export = styler.export()
+        >>> styler2.use(export)  # doctest: +SKIP
+        """
+        self._todo.extend(styles.get("apply", []))
+        table_attributes: str = self.table_attributes or ""
+        obj_table_atts: str = (
+            ""
+            if styles.get("table_attributes") is None
+            else str(styles.get("table_attributes"))
+        )
+        self.set_table_attributes((table_attributes + " " + obj_table_atts).strip())
+        if styles.get("table_styles"):
+            self.set_table_styles(styles.get("table_styles"), overwrite=False)
+
+        for obj in ["index", "columns"]:
+            hide_obj = styles.get("hide_" + obj)
+            if hide_obj is not None:
+                if isinstance(hide_obj, bool):
+                    n = getattr(self, obj).nlevels
+                    setattr(self, "hide_" + obj + "_", [hide_obj] * n)
+                else:
+                    setattr(self, "hide_" + obj + "_", hide_obj)
+
+        self.hide_index_names = styles.get("hide_index_names", False)
+        self.hide_column_names = styles.get("hide_column_names", False)
+        if styles.get("css"):
+            self.css = styles.get("css")  # type: ignore[assignment]
+        return self
+
+    def set_uuid(self, uuid: str) -> Styler:
+        """
+        Set the uuid applied to ``id`` attributes of HTML elements.
+
+        Parameters
+        ----------
+        uuid : str
+            The uuid to be applied to ``id`` attributes of HTML elements.
+
+        Returns
+        -------
+        Styler
+            Instance of class with specified uuid for `id` attributes set.
+
+        See Also
+        --------
+        Styler.set_caption : Set the text added to a ``<caption>`` HTML element.
+        Styler.set_td_classes : Set the ``class`` attribute of ``<td>`` HTML elements.
+        Styler.set_tooltips : Set the DataFrame of strings on ``Styler`` generating
+            ``:hover`` tooltips.
+
+        Notes
+        -----
+        Almost all HTML elements within the table, and including the ``<table>`` element
+        are assigned ``id`` attributes. The format is ``T_uuid_<extra>`` where
+        ``<extra>`` is typically a more specific identifier, such as ``row1_col2``.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], index=["A", "B"], columns=["c1", "c2"])
+
+        You can get the `id` attributes with the following:
+
+        >>> print((df).style.to_html())  # doctest: +SKIP
+
+        To add a title to column `c1`, its `id` is T_20a7d_level0_col0:
+
+        >>> df.style.set_uuid("T_20a7d_level0_col0").set_caption("Test")
+        ... # doctest: +SKIP
+
+        Please see:
+        `Table visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        self.uuid = uuid
+        return self
+
+    def set_caption(self, caption: str | tuple | list) -> Styler:
+        """
+        Set the text added to a ``<caption>`` HTML element.
+
+        Parameters
+        ----------
+        caption : str, tuple, list
+            For HTML output either the string input is used or the first element of the
+            tuple. For LaTeX the string input provides a caption and the additional
+            tuple input allows for full captions and short captions, in that order.
+
+        Returns
+        -------
+        Styler
+            Instance of class with text set for ``<caption>`` HTML element.
+
+        See Also
+        --------
+        Styler.set_td_classes : Set the ``class`` attribute of ``<td>`` HTML elements.
+        Styler.set_tooltips : Set the DataFrame of strings on ``Styler`` generating
+            ``:hover`` tooltips.
+        Styler.set_uuid : Set the uuid applied to ``id`` attributes of HTML elements.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.style.set_caption("test")  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        msg = "`caption` must be either a string or 2-tuple of strings."
+        if isinstance(caption, (list, tuple)):
+            if (
+                len(caption) != 2
+                or not isinstance(caption[0], str)
+                or not isinstance(caption[1], str)
+            ):
+                raise ValueError(msg)
+        elif not isinstance(caption, str):
+            raise ValueError(msg)
+        self.caption = caption
+        return self
+
+    def set_sticky(
+        self,
+        axis: Axis = 0,
+        pixel_size: int | None = None,
+        levels: Level | list[Level] | None = None,
+    ) -> Styler:
+        """
+        Add CSS to permanently display the index or column headers in a scrolling frame.
+
+        Parameters
+        ----------
+        axis : {0 or 'index', 1 or 'columns'}, default 0
+            Whether to make the index or column headers sticky.
+        pixel_size : int, optional
+            Required to configure the width of index cells or the height of column
+            header cells when sticking a MultiIndex (or with a named Index).
+            Defaults to 75 and 25 respectively.
+        levels : int, str, list, optional
+            If ``axis`` is a MultiIndex the specific levels to stick. If ``None`` will
+            stick all levels.
+
+        Returns
+        -------
+        Styler
+            Instance of class with CSS set for permanently displaying headers
+                in scrolling frame.
+
+        See Also
+        --------
+        Styler.set_properties : Set defined CSS-properties to each ``<td>``
+            HTML element for the given subset.
+
+        Notes
+        -----
+        This method uses the CSS 'position: sticky;' property to display. It is
+        designed to work with visible axes, therefore both:
+
+          - `styler.set_sticky(axis="index").hide(axis="index")`
+          - `styler.set_sticky(axis="columns").hide(axis="columns")`
+
+        may produce strange behaviour due to CSS controls with missing elements.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+        >>> df.style.set_sticky(axis="index")  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        axis = self.data._get_axis_number(axis)
+        obj = self.data.index if axis == 0 else self.data.columns
+        pixel_size = (75 if axis == 0 else 25) if not pixel_size else pixel_size
+
+        props = "position:sticky; background-color:inherit;"
+        if not isinstance(obj, pd.MultiIndex):
+            # handling MultiIndexes requires different CSS
+
+            if axis == 1:
+                # stick the first <tr> of <head> and, if index names, the second <tr>
+                # if self._hide_columns then no <thead><tr> here will exist: no conflict
+                styles: CSSStyles = [
+                    {
+                        "selector": "thead tr:nth-child(1) th",
+                        "props": props + "top:0px; z-index:2;",
+                    }
+                ]
+                if self.index.names[0] is not None:
+                    styles[0]["props"] = (
+                        props + f"top:0px; z-index:2; height:{pixel_size}px;"
+                    )
+                    styles.append(
+                        {
+                            "selector": "thead tr:nth-child(2) th",
+                            "props": props
+                            + f"top:{pixel_size}px; z-index:2; height:{pixel_size}px; ",
+                        }
+                    )
+            else:
+                # stick the first <th> of each <tr> in both <thead> and <tbody>
+                # if self._hide_index then no <th> will exist in <tbody>: no conflict
+                # but <th> will exist in <thead>: conflict with initial element
+                styles = [
+                    {
+                        "selector": "thead tr th:nth-child(1)",
+                        "props": props + "left:0px; z-index:3 !important;",
+                    },
+                    {
+                        "selector": "tbody tr th:nth-child(1)",
+                        "props": props + "left:0px; z-index:1;",
+                    },
+                ]
+
+        else:
+            # handle the MultiIndex case
+            range_idx = list(range(obj.nlevels))
+            levels_: list[int] = refactor_levels(levels, obj) if levels else range_idx
+            levels_ = sorted(levels_)
+
+            if axis == 1:
+                styles = []
+                for i, level in enumerate(levels_):
+                    styles.append(
+                        {
+                            "selector": f"thead tr:nth-child({level + 1}) th",
+                            "props": props
+                            + (
+                                f"top:{i * pixel_size}px; height:{pixel_size}px; "
+                                "z-index:2;"
+                            ),
+                        }
+                    )
+                if not all(name is None for name in self.index.names):
+                    styles.append(
+                        {
+                            "selector": f"thead tr:nth-child({obj.nlevels + 1}) th",
+                            "props": props
+                            + (
+                                f"top:{(len(levels_)) * pixel_size}px; "
+                                f"height:{pixel_size}px; z-index:2;"
+                            ),
+                        }
+                    )
+
+            else:
+                styles = []
+                for i, level in enumerate(levels_):
+                    props_ = props + (
+                        f"left:{i * pixel_size}px; "
+                        f"min-width:{pixel_size}px; "
+                        f"max-width:{pixel_size}px; "
+                    )
+                    styles.extend(
+                        [
+                            {
+                                "selector": f"thead tr th:nth-child({level + 1})",
+                                "props": props_ + "z-index:3 !important;",
+                            },
+                            {
+                                "selector": f"tbody tr th.level{level}",
+                                "props": props_ + "z-index:1;",
+                            },
+                        ]
+                    )
+
+        return self.set_table_styles(styles, overwrite=False)
+
+    def set_table_styles(
+        self,
+        table_styles: dict[Any, CSSStyles] | CSSStyles | None = None,
+        axis: AxisInt = 0,
+        overwrite: bool = True,
+        css_class_names: dict[str, str] | None = None,
+    ) -> Styler:
+        """
+        Set the table styles included within the ``<style>`` HTML element.
+
+        This function can be used to style the entire table, columns, rows or
+        specific HTML selectors.
+
+        Parameters
+        ----------
+        table_styles : list or dict
+            If supplying a list, each individual table_style should be a
+            dictionary with ``selector`` and ``props`` keys. ``selector``
+            should be a CSS selector that the style will be applied to
+            (automatically prefixed by the table's UUID) and ``props``
+            should be a list of tuples with ``(attribute, value)``.
+            If supplying a dict, the dict keys should correspond to
+            column names or index values, depending upon the specified
+            `axis` argument. These will be mapped to row or col CSS
+            selectors. MultiIndex values as dict keys should be
+            in their respective tuple form. The dict values should be
+            a list as specified in the form with CSS selectors and
+            props that will be applied to the specified row or column.
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``). Only used if `table_styles` is
+            dict.
+        overwrite : bool, default True
+            Styles are replaced if `True`, or extended if `False`. CSS
+            rules are preserved so most recent styles set will dominate
+            if selectors intersect.
+        css_class_names : dict, optional
+            A dict of strings used to replace the default CSS classes described below.
+
+            .. versionadded:: 1.4.0
+
+        Returns
+        -------
+        Styler
+            Instance of class with specified table styles set.
+
+        See Also
+        --------
+        Styler.set_td_classes: Set the DataFrame of strings added to the ``class``
+            attribute of ``<td>`` HTML elements.
+        Styler.set_table_attributes: Set the table attributes added to the ``<table>``
+            HTML element.
+
+        Notes
+        -----
+        The default CSS classes dict, whose values can be replaced is as follows:
+
+        .. code-block:: python
+
+            css_class_names = {
+                "row_heading": "row_heading",
+                "col_heading": "col_heading",
+                "index_name": "index_name",
+                "col": "col",
+                "row": "row",
+                "col_trim": "col_trim",
+                "row_trim": "row_trim",
+                "level": "level",
+                "data": "data",
+                "blank": "blank",
+                "foot": "foot",
+            }
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(np.random.randn(10, 4), columns=["A", "B", "C", "D"])
+        >>> df.style.set_table_styles(
+        ...     [{"selector": "tr:hover", "props": [("background-color", "yellow")]}]
+        ... )  # doctest: +SKIP
+
+        Or with CSS strings
+
+        >>> df.style.set_table_styles(
+        ...     [
+        ...         {
+        ...             "selector": "tr:hover",
+        ...             "props": "background-color: yellow; font-size: 1em;",
+        ...         }
+        ...     ]
+        ... )  # doctest: +SKIP
+
+        Adding column styling by name
+
+        >>> df.style.set_table_styles(
+        ...     {
+        ...         "A": [{"selector": "", "props": [("color", "red")]}],
+        ...         "B": [{"selector": "td", "props": "color: blue;"}],
+        ...     },
+        ...     overwrite=False,
+        ... )  # doctest: +SKIP
+
+        Adding row styling
+
+        >>> df.style.set_table_styles(
+        ...     {0: [{"selector": "td:hover", "props": [("font-size", "25px")]}]},
+        ...     axis=1,
+        ...     overwrite=False,
+        ... )  # doctest: +SKIP
+
+        See `Table Visualization <../../user_guide/style.ipynb>`_ user guide for
+        more details.
+        """
+        if css_class_names is not None:
+            self.css = {**self.css, **css_class_names}
+
+        if table_styles is None:
+            return self
+        elif isinstance(table_styles, dict):
+            axis = self.data._get_axis_number(axis)
+            obj = self.data.index if axis == 1 else self.data.columns
+            idf = f".{self.css['row']}" if axis == 1 else f".{self.css['col']}"
+
+            table_styles = [
+                {
+                    "selector": str(s["selector"]) + idf + str(idx),
+                    "props": maybe_convert_css_to_tuples(s["props"]),
+                }
+                for key, styles in table_styles.items()
+                for idx in obj.get_indexer_for([key])
+                for s in format_table_styles(styles)
+            ]
+        else:
+            table_styles = [
+                {
+                    "selector": s["selector"],
+                    "props": maybe_convert_css_to_tuples(s["props"]),
+                }
+                for s in table_styles
+            ]
+
+        if not overwrite and self.table_styles is not None:
+            self.table_styles.extend(table_styles)
+        else:
+            self.table_styles = table_styles
+        return self
+
+    def hide(
+        self,
+        subset: Subset | None = None,
+        axis: Axis = 0,
+        level: Level | list[Level] | None = None,
+        names: bool = False,
+    ) -> Styler:
+        """
+        Hide the entire index / column headers, or specific rows / columns from display.
+
+        .. versionadded:: 1.4.0
+
+        Parameters
+        ----------
+        subset : label, array-like, IndexSlice, optional
+            A valid 1d input or single key along the axis within
+            `DataFrame.loc[<subset>, :]` or `DataFrame.loc[:, <subset>]` depending
+            upon ``axis``, to limit ``data`` to select hidden rows / columns.
+        axis : {"index", 0, "columns", 1}
+            Apply to the index or columns.
+        level : int, str, list
+            The level(s) to hide in a MultiIndex if hiding the entire index / column
+            headers. Cannot be used simultaneously with ``subset``.
+        names : bool
+            Whether to hide the level name(s) of the index / columns headers in the case
+            it (or at least one the levels) remains visible.
+
+        Returns
+        -------
+        Styler
+            Instance of class with specified headers/rows/columns hidden from display.
+
+        See Also
+        --------
+        Styler.apply : Apply a CSS-styling function column-wise, row-wise,
+            or table-wise.
+        Styler.map : Apply a CSS-styling function elementwise.
+
+        Notes
+        -----
+        .. warning::
+           This method only works with the output methods ``to_html``, ``to_string``
+           and ``to_latex``.
+
+           Other output methods, including ``to_excel``, ignore this hiding method
+           and will display all data.
+
+        This method has multiple functionality depending upon the combination
+        of the ``subset``, ``level`` and ``names`` arguments (see examples). The
+        ``axis`` argument is used only to control whether the method is applied to row
+        or column headers:
+
+        .. list-table:: Argument combinations
+           :widths: 10 20 10 60
+           :header-rows: 1
+
+           * - ``subset``
+             - ``level``
+             - ``names``
+             - Effect
+           * - None
+             - None
+             - False
+             - The axis-Index is hidden entirely.
+           * - None
+             - None
+             - True
+             - Only the axis-Index names are hidden.
+           * - None
+             - Int, Str, List
+             - False
+             - Specified axis-MultiIndex levels are hidden entirely.
+           * - None
+             - Int, Str, List
+             - True
+             - Specified axis-MultiIndex levels are hidden entirely and the names of
+               remaining axis-MultiIndex levels.
+           * - Subset
+             - None
+             - False
+             - The specified data rows/columns are hidden, but the axis-Index itself,
+               and names, remain unchanged.
+           * - Subset
+             - None
+             - True
+             - The specified data rows/columns and axis-Index names are hidden, but
+               the axis-Index itself remains unchanged.
+           * - Subset
+             - Int, Str, List
+             - Boolean
+             - ValueError: cannot supply ``subset`` and ``level`` simultaneously.
+
+        Note this method only hides the identified elements so can be chained to hide
+        multiple elements in sequence.
+
+        Examples
+        --------
+        Simple application hiding specific rows:
+
+        >>> df = pd.DataFrame([[1, 2], [3, 4], [5, 6]], index=["a", "b", "c"])
+        >>> df.style.hide(["a", "b"])  # doctest: +SKIP
+             0    1
+        c    5    6
+
+        Hide the index and retain the data values:
+
+        >>> midx = pd.MultiIndex.from_product([["x", "y"], ["a", "b", "c"]])
+        >>> df = pd.DataFrame(np.random.randn(6, 6), index=midx, columns=midx)
+        >>> df.style.format("{:.1f}").hide()  # doctest: +SKIP
+                         x                    y
+           a      b      c      a      b      c
+         0.1    0.0    0.4    1.3    0.6   -1.4
+         0.7    1.0    1.3    1.5   -0.0   -0.2
+         1.4   -0.8    1.6   -0.2   -0.4   -0.3
+         0.4    1.0   -0.2   -0.8   -1.2    1.1
+        -0.6    1.2    1.8    1.9    0.3    0.3
+         0.8    0.5   -0.3    1.2    2.2   -0.8
+
+        Hide specific rows in a MultiIndex but retain the index:
+
+        >>> df.style.format("{:.1f}").hide(subset=(slice(None), ["a", "c"]))
+        ... # doctest: +SKIP
+                                 x                    y
+                   a      b      c      a      b      c
+        x   b    0.7    1.0    1.3    1.5   -0.0   -0.2
+        y   b   -0.6    1.2    1.8    1.9    0.3    0.3
+
+        Hide specific rows and the index through chaining:
+
+        >>> df.style.format("{:.1f}").hide(subset=(slice(None), ["a", "c"])).hide()
+        ... # doctest: +SKIP
+                         x                    y
+           a      b      c      a      b      c
+         0.7    1.0    1.3    1.5   -0.0   -0.2
+        -0.6    1.2    1.8    1.9    0.3    0.3
+
+        Hide a specific level:
+
+        >>> df.style.format("{:,.1f}").hide(level=1)  # doctest: +SKIP
+                             x                    y
+               a      b      c      a      b      c
+        x    0.1    0.0    0.4    1.3    0.6   -1.4
+             0.7    1.0    1.3    1.5   -0.0   -0.2
+             1.4   -0.8    1.6   -0.2   -0.4   -0.3
+        y    0.4    1.0   -0.2   -0.8   -1.2    1.1
+            -0.6    1.2    1.8    1.9    0.3    0.3
+             0.8    0.5   -0.3    1.2    2.2   -0.8
+
+        Hiding just the index level names:
+
+        >>> df.index.names = ["lev0", "lev1"]
+        >>> df.style.format("{:,.1f}").hide(names=True)  # doctest: +SKIP
+                                 x                    y
+                   a      b      c      a      b      c
+        x   a    0.1    0.0    0.4    1.3    0.6   -1.4
+            b    0.7    1.0    1.3    1.5   -0.0   -0.2
+            c    1.4   -0.8    1.6   -0.2   -0.4   -0.3
+        y   a    0.4    1.0   -0.2   -0.8   -1.2    1.1
+            b   -0.6    1.2    1.8    1.9    0.3    0.3
+            c    0.8    0.5   -0.3    1.2    2.2   -0.8
+
+        Examples all produce equivalently transposed effects with ``axis="columns"``.
+        """
+        axis = self.data._get_axis_number(axis)
+        if axis == 0:
+            obj, objs, alt = "index", "index", "rows"
+        else:
+            obj, objs, alt = "column", "columns", "columns"
+
+        if level is not None and subset is not None:
+            raise ValueError("`subset` and `level` cannot be passed simultaneously")
+
+        if subset is None:
+            if level is None and names:
+                # this combination implies user shows the index and hides just names
+                setattr(self, f"hide_{obj}_names", True)
+                return self
+
+            levels_ = refactor_levels(level, getattr(self, objs))
+            setattr(
+                self,
+                f"hide_{objs}_",
+                [lev in levels_ for lev in range(getattr(self, objs).nlevels)],
+            )
+        else:
+            if axis == 0:
+                subset_ = IndexSlice[subset, :]  # new var so mypy reads not Optional
+            else:
+                subset_ = IndexSlice[:, subset]  # new var so mypy reads not Optional
+            subset = non_reducing_slice(subset_)
+            hide = self.data.loc[subset]
+            h_els = getattr(self, objs).get_indexer_for(getattr(hide, objs))
+            setattr(self, f"hidden_{alt}", h_els)
+
+        if names:
+            setattr(self, f"hide_{obj}_names", True)
+        return self
+
+    # -----------------------------------------------------------------------
+    # A collection of "builtin" styles
+    # -----------------------------------------------------------------------
+
+    def _get_numeric_subset_default(self):
+        # Returns a boolean mask indicating where `self.data` has numerical columns.
+        # Choosing a mask as opposed to the column names also works for
+        # boolean column labels (GH47838).
+        return self.data.columns.isin(self.data.select_dtypes(include=np.number))
+
+    @doc(
+        name="background",
+        alt="text",
+        image_prefix="bg",
+        text_threshold="""text_color_threshold : float or int\n
+            Luminance threshold for determining text color in [0, 1]. Facilitates text\n
+            visibility across varying background colors. All text is dark if 0, and\n
+            light if 1, defaults to 0.408.""",
+    )
+    @Substitution(subset=subset_args)
+    def background_gradient(
+        self,
+        cmap: str | Colormap = "PuBu",
+        low: float = 0,
+        high: float = 0,
+        axis: Axis | None = 0,
+        subset: Subset | None = None,
+        text_color_threshold: float = 0.408,
+        vmin: float | None = None,
+        vmax: float | None = None,
+        gmap: Sequence | None = None,
+    ) -> Styler:
+        """
+        Color the {name} in a gradient style.
+
+        The {name} color is determined according
+        to the data in each column, row or frame, or by a given
+        gradient map. Requires matplotlib.
+
+        Parameters
+        ----------
+        cmap : str or colormap
+            Matplotlib colormap.
+        low : float
+            Compress the color range at the low end. This is a multiple of the data
+            range to extend below the minimum; good values usually in [0, 1],
+            defaults to 0.
+        high : float
+            Compress the color range at the high end. This is a multiple of the data
+            range to extend above the maximum; good values usually in [0, 1],
+            defaults to 0.
+        axis : {{0, 1, "index", "columns", None}}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``), or to the entire DataFrame at once
+            with ``axis=None``.
+        %(subset)s
+        {text_threshold}
+        vmin : float, optional
+            Minimum data value that corresponds to colormap minimum value.
+            If not specified the minimum value of the data (or gmap) will be used.
+        vmax : float, optional
+            Maximum data value that corresponds to colormap maximum value.
+            If not specified the maximum value of the data (or gmap) will be used.
+        gmap : array-like, optional
+            Gradient map for determining the {name} colors. If not supplied
+            will use the underlying data from rows, columns or frame. If given as an
+            ndarray or list-like must be an identical shape to the underlying data
+            considering ``axis`` and ``subset``. If given as DataFrame or Series must
+            have same index and column labels considering ``axis`` and ``subset``.
+            If supplied, ``vmin`` and ``vmax`` should be given relative to this
+            gradient map.
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Styler
+            Instance of class with {name} colored in gradient style.
+
+        See Also
+        --------
+        Styler.{alt}_gradient: Color the {alt} in a gradient style.
+
+        Notes
+        -----
+        When using ``low`` and ``high`` the range
+        of the gradient, given by the data if ``gmap`` is not given or by ``gmap``,
+        is extended at the low end effectively by
+        `map.min - low * map.range` and at the high end by
+        `map.max + high * map.range` before the colors are normalized and determined.
+
+        If combining with ``vmin`` and ``vmax`` the `map.min`, `map.max` and
+        `map.range` are replaced by values according to the values derived from
+        ``vmin`` and ``vmax``.
+
+        This method will preselect numeric columns and ignore non-numeric columns
+        unless a ``gmap`` is supplied in which case no preselection occurs.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     columns=["City", "Temp (c)", "Rain (mm)", "Wind (m/s)"],
+        ...     data=[
+        ...         ["Stockholm", 21.6, 5.0, 3.2],
+        ...         ["Oslo", 22.4, 13.3, 3.1],
+        ...         ["Copenhagen", 24.5, 0.0, 6.7],
+        ...     ],
+        ... )
+
+        Shading the values column-wise, with ``axis=0``, preselecting numeric columns
+
+        >>> df.style.{name}_gradient(axis=0)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_ax0.png
+
+        Shading all values collectively using ``axis=None``
+
+        >>> df.style.{name}_gradient(axis=None)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_axNone.png
+
+        Compress the color map from the both ``low`` and ``high`` ends
+
+        >>> df.style.{name}_gradient(axis=None, low=0.75, high=1.0)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_axNone_lowhigh.png
+
+        Manually setting ``vmin`` and ``vmax`` gradient thresholds
+
+        >>> df.style.{name}_gradient(axis=None, vmin=6.7, vmax=21.6)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_axNone_vminvmax.png
+
+        Setting a ``gmap`` and applying to all columns with another ``cmap``
+
+        >>> df.style.{name}_gradient(axis=0, gmap=df['Temp (c)'], cmap='YlOrRd')
+        ... # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_gmap.png
+
+        Setting the gradient map for a dataframe (i.e. ``axis=None``), we need to
+        explicitly state ``subset`` to match the ``gmap`` shape
+
+        >>> gmap = np.array([[1, 2, 3], [2, 3, 4], [3, 4, 5]])
+        >>> df.style.{name}_gradient(axis=None, gmap=gmap,
+        ...     cmap='YlOrRd', subset=['Temp (c)', 'Rain (mm)', 'Wind (m/s)']
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/{image_prefix}_axNone_gmap.png
+        """
+        if subset is None and gmap is None:
+            subset = self._get_numeric_subset_default()
+
+        self.apply(
+            _background_gradient,
+            cmap=cmap,
+            subset=subset,
+            axis=axis,
+            low=low,
+            high=high,
+            text_color_threshold=text_color_threshold,
+            vmin=vmin,
+            vmax=vmax,
+            gmap=gmap,
+        )
+        return self
+
+    @doc(
+        background_gradient,
+        name="text",
+        alt="background",
+        image_prefix="tg",
+        text_threshold="",
+    )
+    def text_gradient(
+        self,
+        cmap: str | Colormap = "PuBu",
+        low: float = 0,
+        high: float = 0,
+        axis: Axis | None = 0,
+        subset: Subset | None = None,
+        vmin: float | None = None,
+        vmax: float | None = None,
+        gmap: Sequence | None = None,
+    ) -> Styler:
+        if subset is None and gmap is None:
+            subset = self._get_numeric_subset_default()
+
+        return self.apply(
+            _background_gradient,
+            cmap=cmap,
+            subset=subset,
+            axis=axis,
+            low=low,
+            high=high,
+            vmin=vmin,
+            vmax=vmax,
+            gmap=gmap,
+            text_only=True,
+        )
+
+    @Substitution(subset=subset_args)
+    def set_properties(self, subset: Subset | None = None, **kwargs) -> Styler:
+        """
+        Set defined CSS-properties to each ``<td>`` HTML element for the given subset.
+
+        Parameters
+        ----------
+        %(subset)s
+        **kwargs : dict
+            A dictionary of property, value pairs to be set for each cell.
+
+        Returns
+        -------
+        Styler
+            Instance of class with CSS-properties set for each ``<td>`` HTML element
+                in the given subset
+
+        See Also
+        --------
+        Styler.set_sticky : Add CSS to permanently display the index or column
+            headers in a scrolling frame.
+
+        Notes
+        -----
+        This is a convenience methods which wraps the :meth:`Styler.map` calling a
+        function returning the CSS-properties independently of the data.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(np.random.randn(10, 4))
+        >>> df.style.set_properties(color="white", align="right")  # doctest: +SKIP
+        >>> df.style.set_properties(**{"background-color": "yellow"})  # doctest: +SKIP
+
+        See `Table Visualization <../../user_guide/style.ipynb>`_ user guide for
+        more details.
+        """
+        values = "".join([f"{p}: {v};" for p, v in kwargs.items()])
+        return self.map(lambda x: values, subset=subset)
+
+    @Substitution(subset=subset_args)
+    def bar(
+        self,
+        subset: Subset | None = None,
+        axis: Axis | None = 0,
+        *,
+        color: str | list | tuple | None = None,
+        cmap: Any | None = None,
+        width: float = 100,
+        height: float = 100,
+        align: str | float | Callable = "mid",
+        vmin: float | None = None,
+        vmax: float | None = None,
+        props: str = "width: 10em;",
+    ) -> Styler:
+        """
+        Draw bar chart in the cell backgrounds.
+
+        .. versionchanged:: 1.4.0
+
+        Parameters
+        ----------
+        %(subset)s
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``), or to the entire DataFrame at once
+            with ``axis=None``.
+        color : str or 2-tuple/list
+            If a str is passed, the color is the same for both
+            negative and positive numbers. If 2-tuple/list is used, the
+            first element is the color_negative and the second is the
+            color_positive (eg: ['#d65f5f', '#5fba7d']).
+        cmap : str, matplotlib.cm.ColorMap
+            A string name of a matplotlib Colormap, or a Colormap object. Cannot be
+            used together with ``color``.
+
+            .. versionadded:: 1.4.0
+        width : float, default 100
+            The percentage of the cell, measured from the left, in which to draw the
+            bars, in [0, 100].
+        height : float, default 100
+            The percentage height of the bar in the cell, centrally aligned, in [0,100].
+
+            .. versionadded:: 1.4.0
+        align : str, int, float, callable, default 'mid'
+            How to align the bars within the cells relative to a width adjusted center.
+            If string must be one of:
+
+            - 'left' : bars are drawn rightwards from the minimum data value.
+            - 'right' : bars are drawn leftwards from the maximum data value.
+            - 'zero' : a value of zero is located at the center of the cell.
+            - 'mid' : a value of (max-min)/2 is located at the center of the cell,
+              or if all values are negative (positive) the zero is
+              aligned at the right (left) of the cell.
+            - 'mean' : the mean value of the data is located at the center of the cell.
+
+            If a float or integer is given this will indicate the center of the cell.
+
+            If a callable should take a 1d or 2d array and return a scalar.
+
+            .. versionchanged:: 1.4.0
+
+        vmin : float, optional
+            Minimum bar value, defining the left hand limit
+            of the bar drawing range, lower values are clipped to `vmin`.
+            When None (default): the minimum value of the data will be used.
+        vmax : float, optional
+            Maximum bar value, defining the right hand limit
+            of the bar drawing range, higher values are clipped to `vmax`.
+            When None (default): the maximum value of the data will be used.
+        props : str, optional
+            The base CSS of the cell that is extended to add the bar chart. Defaults to
+            `"width: 10em;"`.
+
+            .. versionadded:: 1.4.0
+
+        Returns
+        -------
+        Styler
+            Contains list-like attribute with bar chart data as formatted CSS.
+
+        See Also
+        --------
+        PlotAccessor.bar : Vertical bar plot.
+        PlotAccessor.line : Plot Series or DataFrame as lines.
+        PlotAccessor.pie : Generate a pie plot.
+
+        Notes
+        -----
+        This section of the user guide:
+        `Table Visualization <../../user_guide/style.ipynb>`_ gives
+        a number of examples for different settings and color coordination.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [3, 4, 5, 6]})
+        >>> df.style.bar(subset=["A"], color="gray")  # doctest: +SKIP
+        """
+        if color is None and cmap is None:
+            color = "#d65f5f"
+        elif color is not None and cmap is not None:
+            raise ValueError("`color` and `cmap` cannot both be given")
+        elif color is not None:
+            if (isinstance(color, (list, tuple)) and len(color) > 2) or not isinstance(
+                color, (str, list, tuple)
+            ):
+                raise ValueError(
+                    "`color` must be string or list or tuple of 2 strings,"
+                    "(eg: color=['#d65f5f', '#5fba7d'])"
+                )
+
+        if not 0 <= width <= 100:
+            raise ValueError(f"`width` must be a value in [0, 100], got {width}")
+        if not 0 <= height <= 100:
+            raise ValueError(f"`height` must be a value in [0, 100], got {height}")
+
+        if subset is None:
+            subset = self._get_numeric_subset_default()
+
+        self.apply(
+            _bar,
+            subset=subset,
+            axis=axis,
+            align=align,
+            colors=color,
+            cmap=cmap,
+            width=width / 100,
+            height=height / 100,
+            vmin=vmin,
+            vmax=vmax,
+            base_css=props,
+        )
+
+        return self
+
+    @Substitution(
+        subset=subset_args,
+        props=properties_args,
+        color=coloring_args.format(default="red"),
+    )
+    def highlight_null(
+        self,
+        color: str = "red",
+        subset: Subset | None = None,
+        props: str | None = None,
+    ) -> Styler:
+        """
+        Highlight missing values with a style.
+
+        Parameters
+        ----------
+        %(color)s
+
+            .. versionadded:: 1.5.0
+
+        %(subset)s
+
+        %(props)s
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Styler
+            Instance of class where null values are highlighted with given style.
+
+        See Also
+        --------
+        Styler.highlight_max: Highlight the maximum with a style.
+        Styler.highlight_min: Highlight the minimum with a style.
+        Styler.highlight_between: Highlight a defined range with a style.
+        Styler.highlight_quantile: Highlight values defined by a quantile with a style.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [1, 2], "B": [3, np.nan]})
+        >>> df.style.highlight_null(color="yellow")  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+
+        def f(data: DataFrame, props: str) -> np.ndarray:
+            return np.where(pd.isna(data).to_numpy(), props, "")
+
+        if props is None:
+            props = f"background-color: {color};"
+        return self.apply(f, axis=None, subset=subset, props=props)
+
+    @Substitution(
+        subset=subset_args,
+        color=coloring_args.format(default="yellow"),
+        props=properties_args,
+    )
+    def highlight_max(
+        self,
+        subset: Subset | None = None,
+        color: str = "yellow",
+        axis: Axis | None = 0,
+        props: str | None = None,
+    ) -> Styler:
+        """
+        Highlight the maximum with a style.
+
+        Parameters
+        ----------
+        %(subset)s
+        %(color)s
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``), or to the entire DataFrame at once
+            with ``axis=None``.
+        %(props)s
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Styler
+            Instance of class where max value is highlighted in given style.
+
+        See Also
+        --------
+        Styler.highlight_null: Highlight missing values with a style.
+        Styler.highlight_min: Highlight the minimum with a style.
+        Styler.highlight_between: Highlight a defined range with a style.
+        Styler.highlight_quantile: Highlight values defined by a quantile with a style.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [2, 1], "B": [3, 4]})
+        >>> df.style.highlight_max(color="yellow")  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+
+        if props is None:
+            props = f"background-color: {color};"
+        return self.apply(
+            partial(_highlight_value, op="max"),
+            axis=axis,
+            subset=subset,
+            props=props,
+        )
+
+    @Substitution(
+        subset=subset_args,
+        color=coloring_args.format(default="yellow"),
+        props=properties_args,
+    )
+    def highlight_min(
+        self,
+        subset: Subset | None = None,
+        color: str = "yellow",
+        axis: Axis | None = 0,
+        props: str | None = None,
+    ) -> Styler:
+        """
+        Highlight the minimum with a style.
+
+        Parameters
+        ----------
+        %(subset)s
+        %(color)s
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Apply to each column (``axis=0`` or ``'index'``), to each row
+            (``axis=1`` or ``'columns'``), or to the entire DataFrame at once
+            with ``axis=None``.
+        %(props)s
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        Styler
+            Instance of class where min value is highlighted in given style.
+
+        See Also
+        --------
+        Styler.highlight_null: Highlight missing values with a style.
+        Styler.highlight_max: Highlight the maximum with a style.
+        Styler.highlight_between: Highlight a defined range with a style.
+        Styler.highlight_quantile: Highlight values defined by a quantile with a style.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"A": [2, 1], "B": [3, 4]})
+        >>> df.style.highlight_min(color="yellow")  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+
+        if props is None:
+            props = f"background-color: {color};"
+        return self.apply(
+            partial(_highlight_value, op="min"),
+            axis=axis,
+            subset=subset,
+            props=props,
+        )
+
+    @Substitution(
+        subset=subset_args,
+        color=coloring_args.format(default="yellow"),
+        props=properties_args,
+    )
+    def highlight_between(
+        self,
+        subset: Subset | None = None,
+        color: str = "yellow",
+        axis: Axis | None = 0,
+        left: Scalar | Sequence | None = None,
+        right: Scalar | Sequence | None = None,
+        inclusive: IntervalClosedType = "both",
+        props: str | None = None,
+    ) -> Styler:
+        """
+        Highlight a defined range with a style.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        %(subset)s
+        %(color)s
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            If ``left`` or ``right`` given as sequence, axis along which to apply those
+            boundaries. See examples.
+        left : scalar or datetime-like, or sequence or array-like, default None
+            Left bound for defining the range.
+        right : scalar or datetime-like, or sequence or array-like, default None
+            Right bound for defining the range.
+        inclusive : {'both', 'neither', 'left', 'right'}
+            Identify whether bounds are closed or open.
+        %(props)s
+
+        Returns
+        -------
+        Styler
+            Instance of class with range highlighted in given style.
+
+        See Also
+        --------
+        Styler.highlight_null: Highlight missing values with a style.
+        Styler.highlight_max: Highlight the maximum with a style.
+        Styler.highlight_min: Highlight the minimum with a style.
+        Styler.highlight_quantile: Highlight values defined by a quantile with a style.
+
+        Notes
+        -----
+        If ``left`` is ``None`` only the right bound is applied.
+        If ``right`` is ``None`` only the left bound is applied. If both are ``None``
+        all values are highlighted.
+
+        ``axis`` is only needed if ``left`` or ``right`` are provided as a sequence or
+        an array-like object for aligning the shapes. If ``left`` and ``right`` are
+        both scalars then all ``axis`` inputs will give the same result.
+
+        This function only works with compatible ``dtypes``. For example a datetime-like
+        region can only use equivalent datetime-like ``left`` and ``right`` arguments.
+        Use ``subset`` to control regions which have multiple ``dtypes``.
+
+        Examples
+        --------
+        Basic usage
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "One": [1.2, 1.6, 1.5],
+        ...         "Two": [2.9, 2.1, 2.5],
+        ...         "Three": [3.1, 3.2, 3.8],
+        ...     }
+        ... )
+        >>> df.style.highlight_between(left=2.1, right=2.9)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hbetw_basic.png
+
+        Using a range input sequence along an ``axis``, in this case setting a ``left``
+        and ``right`` for each column individually
+
+        >>> df.style.highlight_between(
+        ...     left=[1.4, 2.4, 3.4], right=[1.6, 2.6, 3.6], axis=1, color="#fffd75"
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hbetw_seq.png
+
+        Using ``axis=None`` and providing the ``left`` argument as an array that
+        matches the input DataFrame, with a constant ``right``
+
+        >>> df.style.highlight_between(
+        ...     left=[[2, 2, 3], [2, 2, 3], [3, 3, 3]],
+        ...     right=3.5,
+        ...     axis=None,
+        ...     color="#fffd75",
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hbetw_axNone.png
+
+        Using ``props`` instead of default background coloring
+
+        >>> df.style.highlight_between(
+        ...     left=1.5, right=3.5, props="font-weight:bold;color:#e83e8c"
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hbetw_props.png
+        """
+        if props is None:
+            props = f"background-color: {color};"
+        return self.apply(
+            _highlight_between,
+            axis=axis,
+            subset=subset,
+            props=props,
+            left=left,
+            right=right,
+            inclusive=inclusive,
+        )
+
+    @Substitution(
+        subset=subset_args,
+        color=coloring_args.format(default="yellow"),
+        props=properties_args,
+    )
+    def highlight_quantile(
+        self,
+        subset: Subset | None = None,
+        color: str = "yellow",
+        axis: Axis | None = 0,
+        q_left: float = 0.0,
+        q_right: float = 1.0,
+        interpolation: QuantileInterpolation = "linear",
+        inclusive: IntervalClosedType = "both",
+        props: str | None = None,
+    ) -> Styler:
+        """
+        Highlight values defined by a quantile with a style.
+
+        .. versionadded:: 1.3.0
+
+        Parameters
+        ----------
+        %(subset)s
+        %(color)s
+        axis : {0 or 'index', 1 or 'columns', None}, default 0
+            Axis along which to determine and highlight quantiles. If ``None`` quantiles
+            are measured over the entire DataFrame. See examples.
+        q_left : float, default 0
+            Left bound, in [0, q_right), for the target quantile range.
+        q_right : float, default 1
+            Right bound, in (q_left, 1], for the target quantile range.
+        interpolation : {'linear', 'lower', 'higher', 'midpoint', 'nearest'}
+            Argument passed to ``Series.quantile`` or ``DataFrame.quantile`` for
+            quantile estimation.
+        inclusive : {'both', 'neither', 'left', 'right'}
+            Identify whether quantile bounds are closed or open.
+        %(props)s
+
+        Returns
+        -------
+        Styler
+            Instance of class where values in quantile highlighted with given style.
+
+        See Also
+        --------
+        Styler.highlight_null: Highlight missing values with a style.
+        Styler.highlight_max: Highlight the maximum with a style.
+        Styler.highlight_min: Highlight the minimum with a style.
+        Styler.highlight_between: Highlight a defined range with a style.
+
+        Notes
+        -----
+        This function does not work with ``str`` dtypes.
+
+        Examples
+        --------
+        Using ``axis=None`` and apply a quantile to all collective data
+
+        >>> df = pd.DataFrame(np.arange(10).reshape(2, 5) + 1)
+        >>> df.style.highlight_quantile(axis=None, q_left=0.8, color="#fffd75")
+        ... # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hq_axNone.png
+
+        Or highlight quantiles row-wise or column-wise, in this case by row-wise
+
+        >>> df.style.highlight_quantile(axis=1, q_left=0.8, color="#fffd75")
+        ... # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hq_ax1.png
+
+        Use ``props`` instead of default background coloring
+
+        >>> df.style.highlight_quantile(
+        ...     axis=None,
+        ...     q_left=0.2,
+        ...     q_right=0.8,
+        ...     props="font-weight:bold;color:#e83e8c",
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/hq_props.png
+        """
+        subset_ = slice(None) if subset is None else subset
+        subset_ = non_reducing_slice(subset_)
+        data = self.data.loc[subset_]
+
+        # after quantile is found along axis, e.g. along rows,
+        # applying the calculated quantile to alternate axis, e.g. to each column
+        quantiles = [q_left, q_right]
+        if axis is None:
+            q = Series(data.to_numpy().ravel()).quantile(
+                q=quantiles, interpolation=interpolation
+            )
+            axis_apply: int | None = None
+        else:
+            axis = self.data._get_axis_number(axis)
+            q = data.quantile(
+                axis=axis, numeric_only=False, q=quantiles, interpolation=interpolation
+            )
+            axis_apply = 1 - axis
+
+        if props is None:
+            props = f"background-color: {color};"
+        return self.apply(
+            _highlight_between,
+            axis=axis_apply,
+            subset=subset,
+            props=props,
+            left=q.iloc[0],
+            right=q.iloc[1],
+            inclusive=inclusive,
+        )
+
+    @classmethod
+    def from_custom_template(
+        cls,
+        searchpath: Sequence[str],
+        html_table: str | None = None,
+        html_style: str | None = None,
+    ) -> type[Styler]:
+        """
+        Factory function for creating a subclass of ``Styler``.
+
+        Uses custom templates and Jinja environment.
+
+        .. versionchanged:: 1.3.0
+
+        Parameters
+        ----------
+        searchpath : str or list
+            Path or paths of directories containing the templates.
+        html_table : str
+            Name of your custom template to replace the html_table template.
+
+            .. versionadded:: 1.3.0
+
+        html_style : str
+            Name of your custom template to replace the html_style template.
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        MyStyler : subclass of Styler
+            Has the correct ``env``,``template_html``, ``template_html_table`` and
+            ``template_html_style`` class attributes set.
+
+        See Also
+        --------
+        Styler.export : Export the styles applied to the current Styler.
+        Styler.use : Set the styles on the current Styler.
+
+        Examples
+        --------
+        >>> from pandas.io.formats.style import Styler
+        >>> EasyStyler = Styler.from_custom_template(
+        ...     "path/to/template",
+        ...     "template.tpl",
+        ... )  # doctest: +SKIP
+        >>> df = pd.DataFrame({"A": [1, 2]})
+        >>> EasyStyler(df)  # doctest: +SKIP
+
+        Please see:
+        `Table Visualization <../../user_guide/style.ipynb>`_ for more examples.
+        """
+        loader = jinja2.ChoiceLoader([jinja2.FileSystemLoader(searchpath), cls.loader])
+
+        # mypy doesn't like dynamically-defined classes
+        # error: Variable "cls" is not valid as a type
+        # error: Invalid base class "cls"
+        class MyStyler(cls):  # type: ignore[valid-type,misc]
+            env = jinja2.Environment(loader=loader)
+            if html_table:
+                template_html_table = env.get_template(html_table)
+            if html_style:
+                template_html_style = env.get_template(html_style)
+
+        return MyStyler
+
+    @overload
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T],
+        *args: P.args,
+        **kwargs: P.kwargs,
+    ) -> T: ...
+
+    @overload
+    def pipe(
+        self,
+        func: tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T: ...
+
+    def pipe(
+        self,
+        func: Callable[Concatenate[Self, P], T] | tuple[Callable[..., T], str],
+        *args: Any,
+        **kwargs: Any,
+    ) -> T:
+        """
+        Apply ``func(self, *args, **kwargs)``, and return the result.
+
+        Parameters
+        ----------
+        func : function
+            Function to apply to the Styler.  Alternatively, a
+            ``(callable, keyword)`` tuple where ``keyword`` is a string
+            indicating the keyword of ``callable`` that expects the Styler.
+        *args : optional
+            Arguments passed to `func`.
+        **kwargs : optional
+            A dictionary of keyword arguments passed into ``func``.
+
+        Returns
+        -------
+        object :
+            The value returned by ``func``.
+
+        See Also
+        --------
+        DataFrame.pipe : Analogous method for DataFrame.
+        Styler.apply : Apply a CSS-styling function column-wise, row-wise, or
+            table-wise.
+
+        Notes
+        -----
+        Like :meth:`DataFrame.pipe`, this method can simplify the
+        application of several user-defined functions to a styler.  Instead
+        of writing:
+
+        .. code-block:: python
+
+            f(g(df.style.format(precision=3), arg1=a), arg2=b, arg3=c)
+
+        users can write:
+
+        .. code-block:: python
+
+            (df.style.format(precision=3).pipe(g, arg1=a).pipe(f, arg2=b, arg3=c))
+
+        In particular, this allows users to define functions that take a
+        styler object, along with other parameters, and return the styler after
+        making styling changes (such as calling :meth:`Styler.apply` or
+        :meth:`Styler.set_properties`).
+
+        Examples
+        --------
+
+        **Common Use**
+
+        A common usage pattern is to pre-define styling operations which
+        can be easily applied to a generic styler in a single ``pipe`` call.
+
+        >>> def some_highlights(styler, min_color="red", max_color="blue"):
+        ...     styler.highlight_min(color=min_color, axis=None)
+        ...     styler.highlight_max(color=max_color, axis=None)
+        ...     styler.highlight_null()
+        ...     return styler
+        >>> df = pd.DataFrame([[1, 2, 3, pd.NA], [pd.NA, 4, 5, 6]], dtype="Int64")
+        >>> df.style.pipe(some_highlights, min_color="green")  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/df_pipe_hl.png
+
+        Since the method returns a ``Styler`` object it can be chained with other
+        methods as if applying the underlying highlighters directly.
+
+        >>> (
+        ...     df.style.format("{:.1f}")
+        ...     .pipe(some_highlights, min_color="green")
+        ...     .highlight_between(left=2, right=5)
+        ... )  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/df_pipe_hl2.png
+
+        **Advanced Use**
+
+        Sometimes it may be necessary to pre-define styling functions, but in the case
+        where those functions rely on the styler, data or context. Since
+        ``Styler.use`` and ``Styler.export`` are designed to be non-data dependent,
+        they cannot be used for this purpose. Additionally the ``Styler.apply``
+        and ``Styler.format`` type methods are not context aware, so a solution
+        is to use ``pipe`` to dynamically wrap this functionality.
+
+        Suppose we want to code a generic styling function that highlights the final
+        level of a MultiIndex. The number of levels in the Index is dynamic so we
+        need the ``Styler`` context to define the level.
+
+        >>> def highlight_last_level(styler):
+        ...     return styler.apply_index(
+        ...         lambda v: "background-color: pink; color: yellow",
+        ...         axis="columns",
+        ...         level=styler.columns.nlevels - 1,
+        ...     )  # doctest: +SKIP
+        >>> df.columns = pd.MultiIndex.from_product([["A", "B"], ["X", "Y"]])
+        >>> df.style.pipe(highlight_last_level)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/df_pipe_applymap.png
+
+        Additionally suppose we want to highlight a column header if there is any
+        missing data in that column.
+        In this case we need the data object itself to determine the effect on the
+        column headers.
+
+        >>> def highlight_header_missing(styler, level):
+        ...     def dynamic_highlight(s):
+        ...         return np.where(
+        ...             styler.data.isna().any(), "background-color: red;", ""
+        ...         )
+        ...
+        ...     return styler.apply_index(dynamic_highlight, axis=1, level=level)
+        >>> df.style.pipe(highlight_header_missing, level=1)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/df_pipe_applydata.png
+        """
+        return com.pipe(self, func, *args, **kwargs)
+
+
+def _validate_apply_axis_arg(
+    arg: NDFrame | Sequence | np.ndarray,
+    arg_name: str,
+    dtype: Any | None,
+    data: NDFrame,
+) -> np.ndarray:
+    """
+    For the apply-type methods, ``axis=None`` creates ``data`` as DataFrame, and for
+    ``axis=[1,0]`` it creates a Series. Where ``arg`` is expected as an element
+    of some operator with ``data`` we must make sure that the two are compatible shapes,
+    or raise.
+
+    Parameters
+    ----------
+    arg : sequence, Series or DataFrame
+        the user input arg
+    arg_name : string
+        name of the arg for use in error messages
+    dtype : numpy dtype, optional
+        forced numpy dtype if given
+    data : Series or DataFrame
+        underling subset of Styler data on which operations are performed
+
+    Returns
+    -------
+    ndarray
+    """
+    dtype = {"dtype": dtype} if dtype else {}
+    # raise if input is wrong for axis:
+    if isinstance(arg, Series) and isinstance(data, DataFrame):
+        raise ValueError(
+            f"'{arg_name}' is a Series but underlying data for operations "
+            f"is a DataFrame since 'axis=None'"
+        )
+    if isinstance(arg, DataFrame) and isinstance(data, Series):
+        raise ValueError(
+            f"'{arg_name}' is a DataFrame but underlying data for "
+            f"operations is a Series with 'axis in [0,1]'"
+        )
+    if isinstance(arg, (Series, DataFrame)):  # align indx / cols to data
+        arg = arg.reindex_like(data).to_numpy(**dtype)
+    else:
+        arg = np.asarray(arg, **dtype)
+        assert isinstance(arg, np.ndarray)  # mypy requirement
+        if arg.shape != data.shape:  # check valid input
+            raise ValueError(
+                f"supplied '{arg_name}' is not correct shape for data over "
+                f"selected 'axis': got {arg.shape}, "
+                f"expected {data.shape}"
+            )
+    return arg
+
+
+def _background_gradient(
+    data,
+    cmap: str | Colormap = "PuBu",
+    low: float = 0,
+    high: float = 0,
+    text_color_threshold: float = 0.408,
+    vmin: float | None = None,
+    vmax: float | None = None,
+    gmap: Sequence | np.ndarray | DataFrame | Series | None = None,
+    text_only: bool = False,
+) -> list[str] | DataFrame:
+    """
+    Color background in a range according to the data or a gradient map
+    """
+    if gmap is None:  # the data is used the gmap
+        gmap = data.to_numpy(dtype=float, na_value=np.nan)
+    else:  # else validate gmap against the underlying data
+        gmap = _validate_apply_axis_arg(gmap, "gmap", float, data)
+
+    smin = np.nanmin(gmap) if vmin is None else vmin
+    smax = np.nanmax(gmap) if vmax is None else vmax
+    rng = smax - smin
+    _matplotlib = import_optional_dependency(
+        "matplotlib", extra="Styler.background_gradient requires matplotlib."
+    )
+    # extend lower / upper bounds, compresses color range
+    norm = _matplotlib.colors.Normalize(smin - (rng * low), smax + (rng * high))
+
+    if cmap is None:
+        rgbas = _matplotlib.colormaps[_matplotlib.rcParams["image.cmap"]](norm(gmap))
+    else:
+        rgbas = _matplotlib.colormaps.get_cmap(cmap)(norm(gmap))
+
+    def relative_luminance(rgba) -> float:
+        """
+        Calculate relative luminance of a color.
+
+        The calculation adheres to the W3C standards
+        (https://www.w3.org/WAI/GL/wiki/Relative_luminance)
+
+        Parameters
+        ----------
+        color : rgb or rgba tuple
+
+        Returns
+        -------
+        float
+            The relative luminance as a value from 0 to 1
+        """
+        r, g, b = (
+            x / 12.92 if x <= 0.04045 else ((x + 0.055) / 1.055) ** 2.4
+            for x in rgba[:3]
+        )
+        return 0.2126 * r + 0.7152 * g + 0.0722 * b
+
+    def css(rgba, text_only) -> str:
+        if not text_only:
+            dark = relative_luminance(rgba) < text_color_threshold
+            text_color = "#f1f1f1" if dark else "#000000"
+            return (
+                f"background-color: {_matplotlib.colors.rgb2hex(rgba)};"
+                f"color: {text_color};"
+            )
+        else:
+            return f"color: {_matplotlib.colors.rgb2hex(rgba)};"
+
+    if data.ndim == 1:
+        return [css(rgba, text_only) for rgba in rgbas]
+    else:
+        return DataFrame(
+            [[css(rgba, text_only) for rgba in row] for row in rgbas],
+            index=data.index,
+            columns=data.columns,
+        )
+
+
+def _highlight_between(
+    data: NDFrame,
+    props: str,
+    left: Scalar | Sequence | np.ndarray | NDFrame | None = None,
+    right: Scalar | Sequence | np.ndarray | NDFrame | None = None,
+    inclusive: bool | str = True,
+) -> np.ndarray:
+    """
+    Return an array of css props based on condition of data values within given range.
+    """
+    if np.iterable(left) and not isinstance(left, str):
+        left = _validate_apply_axis_arg(left, "left", None, data)
+
+    if np.iterable(right) and not isinstance(right, str):
+        right = _validate_apply_axis_arg(right, "right", None, data)
+
+    # get ops with correct boundary attribution
+    if inclusive == "both":
+        ops = (operator.ge, operator.le)
+    elif inclusive == "neither":
+        ops = (operator.gt, operator.lt)
+    elif inclusive == "left":
+        ops = (operator.ge, operator.lt)
+    elif inclusive == "right":
+        ops = (operator.gt, operator.le)
+    else:
+        raise ValueError(
+            f"'inclusive' values can be 'both', 'left', 'right', or 'neither' "
+            f"got {inclusive}"
+        )
+
+    g_left = (
+        # error: Argument 2 to "ge" has incompatible type "Union[str, float,
+        # Period, Timedelta, Interval[Any], datetime64, timedelta64, datetime,
+        # Sequence[Any], ndarray[Any, Any], NDFrame]"; expected "Union
+        # [SupportsDunderLE, SupportsDunderGE, SupportsDunderGT, SupportsDunderLT]"
+        ops[0](data, left)  # type: ignore[arg-type]
+        if left is not None
+        else np.full(data.shape, True, dtype=bool)
+    )
+    if isinstance(g_left, (DataFrame, Series)):
+        g_left = g_left.where(pd.notna(g_left), False)
+    l_right = (
+        # error: Argument 2 to "le" has incompatible type "Union[str, float,
+        # Period, Timedelta, Interval[Any], datetime64, timedelta64, datetime,
+        # Sequence[Any], ndarray[Any, Any], NDFrame]"; expected "Union
+        # [SupportsDunderLE, SupportsDunderGE, SupportsDunderGT, SupportsDunderLT]"
+        ops[1](data, right)  # type: ignore[arg-type]
+        if right is not None
+        else np.full(data.shape, True, dtype=bool)
+    )
+    if isinstance(l_right, (DataFrame, Series)):
+        l_right = l_right.where(pd.notna(l_right), False)
+    return np.where(g_left & l_right, props, "")
+
+
+def _highlight_value(data: DataFrame | Series, op: str, props: str) -> np.ndarray:
+    """
+    Return an array of css strings based on the condition of values matching an op.
+    """
+    value = getattr(data, op)(skipna=True)
+    if isinstance(data, DataFrame):  # min/max must be done twice to return scalar
+        value = getattr(value, op)(skipna=True)
+    cond = data == value
+    cond = cond.where(pd.notna(cond), False)
+    return np.where(cond, props, "")
+
+
+def _bar(
+    data: NDFrame,
+    align: str | float | Callable,
+    colors: str | list | tuple,
+    cmap: Any,
+    width: float,
+    height: float,
+    vmin: float | None,
+    vmax: float | None,
+    base_css: str,
+):
+    """
+    Draw bar chart in data cells using HTML CSS linear gradient.
+
+    Parameters
+    ----------
+    data : Series or DataFrame
+        Underling subset of Styler data on which operations are performed.
+    align : str in {"left", "right", "mid", "zero", "mean"}, int, float, callable
+        Method for how bars are structured or scalar value of centre point.
+    colors : list-like of str
+        Two listed colors as string in valid CSS.
+    width : float in [0,1]
+        The percentage of the cell, measured from left, where drawn bars will reside.
+    height : float in [0,1]
+        The percentage of the cell's height where drawn bars will reside, centrally
+        aligned.
+    vmin : float, optional
+        Overwrite the minimum value of the window.
+    vmax : float, optional
+        Overwrite the maximum value of the window.
+    base_css : str
+        Additional CSS that is included in the cell before bars are drawn.
+    """
+
+    def css_bar(start: float, end: float, color: str) -> str:
+        """
+        Generate CSS code to draw a bar from start to end in a table cell.
+
+        Uses linear-gradient.
+
+        Parameters
+        ----------
+        start : float
+            Relative positional start of bar coloring in [0,1]
+        end : float
+            Relative positional end of the bar coloring in [0,1]
+        color : str
+            CSS valid color to apply.
+
+        Returns
+        -------
+        str : The CSS applicable to the cell.
+
+        Notes
+        -----
+        Uses ``base_css`` from outer scope.
+        """
+        cell_css = base_css
+        if end > start:
+            cell_css += "background: linear-gradient(90deg,"
+            if start > 0:
+                cell_css += (
+                    f" transparent {start * 100:.1f}%, {color} {start * 100:.1f}%,"
+                )
+            cell_css += f" {color} {end * 100:.1f}%, transparent {end * 100:.1f}%)"
+        return cell_css
+
+    def css_calc(x, left: float, right: float, align: str, color: str | list | tuple):
+        """
+        Return the correct CSS for bar placement based on calculated values.
+
+        Parameters
+        ----------
+        x : float
+            Value which determines the bar placement.
+        left : float
+            Value marking the left side of calculation, usually minimum of data.
+        right : float
+            Value marking the right side of the calculation, usually maximum of data
+            (left < right).
+        align : {"left", "right", "zero", "mid"}
+            How the bars will be positioned.
+            "left", "right", "zero" can be used with any values for ``left``, ``right``.
+            "mid" can only be used where ``left <= 0`` and ``right >= 0``.
+            "zero" is used to specify a center when all values ``x``, ``left``,
+            ``right`` are translated, e.g. by say a mean or median.
+
+        Returns
+        -------
+        str : Resultant CSS with linear gradient.
+
+        Notes
+        -----
+        Uses ``colors``, ``width`` and ``height`` from outer scope.
+        """
+        if pd.isna(x):
+            return base_css
+
+        if isinstance(color, (list, tuple)):
+            color = color[0] if x < 0 else color[1]
+        assert isinstance(color, str)  # mypy redefinition
+
+        x = left if x < left else x
+        x = right if x > right else x  # trim data if outside of the window
+
+        start: float = 0
+        end: float = 1
+
+        if align == "left":
+            # all proportions are measured from the left side between left and right
+            end = (x - left) / (right - left)
+
+        elif align == "right":
+            # all proportions are measured from the right side between left and right
+            start = (x - left) / (right - left)
+
+        else:
+            z_frac: float = 0.5  # location of zero based on the left-right range
+            if align == "zero":
+                # all proportions are measured from the center at zero
+                limit: float = max(abs(left), abs(right))
+                left, right = -limit, limit
+            elif align == "mid":
+                # bars drawn from zero either leftwards or rightwards with center at mid
+                mid: float = (left + right) / 2
+                z_frac = (
+                    -mid / (right - left) + 0.5 if mid < 0 else -left / (right - left)
+                )
+
+            if x < 0:
+                start, end = (x - left) / (right - left), z_frac
+            else:
+                start, end = z_frac, (x - left) / (right - left)
+
+        ret = css_bar(start * width, end * width, color)
+        if height < 1 and "background: linear-gradient(" in ret:
+            return (
+                ret + f" no-repeat center; background-size: 100% {height * 100:.1f}%;"
+            )
+        else:
+            return ret
+
+    values = data.to_numpy()
+    # A tricky way to address the issue where np.nanmin/np.nanmax fail to handle pd.NA.
+    left = np.nanmin(data.min(skipna=True)) if vmin is None else vmin
+    right = np.nanmax(data.max(skipna=True)) if vmax is None else vmax
+    z: float = 0  # adjustment to translate data
+
+    if align == "mid":
+        if left >= 0:  # "mid" is documented to act as "left" if all values positive
+            align, left = "left", 0 if vmin is None else vmin
+        elif right <= 0:  # "mid" is documented to act as "right" if all values negative
+            align, right = "right", 0 if vmax is None else vmax
+    elif align == "mean":
+        z, align = np.nanmean(values), "zero"
+    elif callable(align):
+        z, align = align(values), "zero"
+    elif isinstance(align, (float, int)):
+        z, align = float(align), "zero"
+    elif align not in ("left", "right", "zero"):
+        raise ValueError(
+            "`align` should be in {'left', 'right', 'mid', 'mean', 'zero'} or be a "
+            "value defining the center line or a callable that returns a float"
+        )
+
+    rgbas = None
+    if cmap is not None:
+        # use the matplotlib colormap input
+        _matplotlib = import_optional_dependency(
+            "matplotlib", extra="Styler.bar requires matplotlib."
+        )
+        cmap = (
+            _matplotlib.colormaps[cmap]
+            if isinstance(cmap, str)
+            else cmap  # assumed to be a Colormap instance as documented
+        )
+        norm = _matplotlib.colors.Normalize(left, right)
+        rgbas = cmap(norm(values))
+        if data.ndim == 1:
+            rgbas = [_matplotlib.colors.rgb2hex(rgba) for rgba in rgbas]
+        else:
+            rgbas = [
+                [_matplotlib.colors.rgb2hex(rgba) for rgba in row] for row in rgbas
+            ]
+
+    assert isinstance(align, str)  # mypy: should now be in [left, right, mid, zero]
+    if data.ndim == 1:
+        return [
+            css_calc(
+                x - z, left - z, right - z, align, colors if rgbas is None else rgbas[i]
+            )
+            for i, x in enumerate(values)
+        ]
+    else:
+        return np.array(
+            [
+                [
+                    css_calc(
+                        x - z,
+                        left - z,
+                        right - z,
+                        align,
+                        colors if rgbas is None else rgbas[i][j],
+                    )
+                    for j, x in enumerate(row)
+                ]
+                for i, row in enumerate(values)
+            ]
+        )
diff --git a/pandas/pandas/io/formats/style_render.py b/pandas/pandas/io/formats/style_render.py
new file mode 100644
index 0000000000000..482ed316c7ce4
--- /dev/null
+++ b/pandas/pandas/io/formats/style_render.py
@@ -0,0 +1,2697 @@
+from __future__ import annotations
+
+from collections import defaultdict
+from collections.abc import (
+    Callable,
+    Sequence,
+)
+from functools import partial
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    DefaultDict,
+    Optional,
+    TypedDict,
+    Union,
+)
+from uuid import uuid4
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+
+from pandas.core.dtypes.common import (
+    is_complex,
+    is_float,
+    is_integer,
+)
+from pandas.core.dtypes.generic import ABCSeries
+
+from pandas import (
+    DataFrame,
+    Index,
+    IndexSlice,
+    MultiIndex,
+    Series,
+    isna,
+)
+from pandas.api.types import is_list_like
+import pandas.core.common as com
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        Axis,
+        Level,
+    )
+jinja2 = import_optional_dependency("jinja2", extra="DataFrame.style requires jinja2.")
+from markupsafe import escape as escape_html  # markupsafe is jinja2 dependency
+
+BaseFormatter = Union[str, Callable]
+ExtFormatter = Union[BaseFormatter, dict[Any, Optional[BaseFormatter]]]
+CSSPair = tuple[str, Union[str, float]]
+CSSList = list[CSSPair]
+CSSProperties = Union[str, CSSList]
+
+
+class CSSDict(TypedDict):
+    selector: str
+    props: CSSProperties
+
+
+CSSStyles = list[CSSDict]
+Subset = Union[slice, Sequence, Index]
+
+
+class StylerRenderer:
+    """
+    Base class to process rendering a Styler with a specified jinja2 template.
+    """
+
+    loader = jinja2.PackageLoader("pandas", "io/formats/templates")
+    env = jinja2.Environment(loader=loader, trim_blocks=True)
+    template_html = env.get_template("html.tpl")
+    template_html_table = env.get_template("html_table.tpl")
+    template_html_style = env.get_template("html_style.tpl")
+    template_latex = env.get_template("latex.tpl")
+    template_typst = env.get_template("typst.tpl")
+    template_string = env.get_template("string.tpl")
+
+    def __init__(
+        self,
+        data: DataFrame | Series,
+        uuid: str | None = None,
+        uuid_len: int = 5,
+        table_styles: CSSStyles | None = None,
+        table_attributes: str | None = None,
+        caption: str | tuple | list | None = None,
+        cell_ids: bool = True,
+        precision: int | None = None,
+    ) -> None:
+        # validate ordered args
+        if isinstance(data, Series):
+            data = data.to_frame()
+        if not isinstance(data, DataFrame):
+            raise TypeError("``data`` must be a Series or DataFrame")
+        self.data: DataFrame = data
+        self.index: Index = data.index
+        self.columns: Index = data.columns
+        if not isinstance(uuid_len, int) or uuid_len < 0:
+            raise TypeError("``uuid_len`` must be an integer in range [0, 32].")
+        self.uuid = uuid or uuid4().hex[: min(32, uuid_len)]
+        self.uuid_len = len(self.uuid)
+        self.table_styles = table_styles
+        self.table_attributes = table_attributes
+        self.caption = caption
+        self.cell_ids = cell_ids
+        self.css = {
+            "row_heading": "row_heading",
+            "col_heading": "col_heading",
+            "index_name": "index_name",
+            "col": "col",
+            "row": "row",
+            "col_trim": "col_trim",
+            "row_trim": "row_trim",
+            "level": "level",
+            "data": "data",
+            "blank": "blank",
+            "foot": "foot",
+        }
+        self.concatenated: list[StylerRenderer] = []
+        # add rendering variables
+        self.hide_index_names: bool = False
+        self.hide_column_names: bool = False
+        self.hide_index_: list = [False] * self.index.nlevels
+        self.hide_columns_: list = [False] * self.columns.nlevels
+        self.hidden_rows: Sequence[int] = []  # sequence for specific hidden rows/cols
+        self.hidden_columns: Sequence[int] = []
+        self.ctx: DefaultDict[tuple[int, int], CSSList] = defaultdict(list)
+        self.ctx_index: DefaultDict[tuple[int, int], CSSList] = defaultdict(list)
+        self.ctx_columns: DefaultDict[tuple[int, int], CSSList] = defaultdict(list)
+        self.cell_context: DefaultDict[tuple[int, int], str] = defaultdict(str)
+        self._todo: list[tuple[Callable, tuple, dict]] = []
+        self.tooltips: Tooltips | None = None
+        precision = (
+            get_option("styler.format.precision") if precision is None else precision
+        )
+        self._display_funcs: DefaultDict[  # maps (row, col) -> format func
+            tuple[int, int], Callable[[Any], str]
+        ] = defaultdict(lambda: partial(_default_formatter, precision=precision))
+        self._display_funcs_index: DefaultDict[  # maps (row, level) -> format func
+            tuple[int, int], Callable[[Any], str]
+        ] = defaultdict(lambda: partial(_default_formatter, precision=precision))
+        self._display_funcs_index_names: DefaultDict[  # maps index level -> format func
+            int, Callable[[Any], str]
+        ] = defaultdict(lambda: partial(_default_formatter, precision=precision))
+        self._display_funcs_columns: DefaultDict[  # maps (level, col) -> format func
+            tuple[int, int], Callable[[Any], str]
+        ] = defaultdict(lambda: partial(_default_formatter, precision=precision))
+        self._display_funcs_column_names: DefaultDict[  # maps col level -> format func
+            int, Callable[[Any], str]
+        ] = defaultdict(lambda: partial(_default_formatter, precision=precision))
+
+    def _render(
+        self,
+        sparse_index: bool,
+        sparse_columns: bool,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        blank: str = "",
+    ):
+        """
+        Computes and applies styles and then generates the general render dicts.
+
+        Also extends the `ctx` and `ctx_index` attributes with those of concatenated
+        stylers for use within `_translate_latex`
+        """
+        self._compute()
+        dxs = []
+        ctx_len = len(self.index)
+        for i, concatenated in enumerate(self.concatenated):
+            concatenated.hide_index_ = self.hide_index_
+            concatenated.hidden_columns = self.hidden_columns
+            foot = f"{self.css['foot']}{i}"
+            concatenated.css = {
+                **self.css,
+                "data": f"{foot}_data",
+                "row_heading": f"{foot}_row_heading",
+                "row": f"{foot}_row",
+                "foot": f"{foot}_foot",
+            }
+            dx = concatenated._render(
+                sparse_index, sparse_columns, max_rows, max_cols, blank
+            )
+            dxs.append(dx)
+
+            for (r, c), v in concatenated.ctx.items():
+                self.ctx[(r + ctx_len, c)] = v
+            for (r, c), v in concatenated.ctx_index.items():
+                self.ctx_index[(r + ctx_len, c)] = v
+
+            ctx_len += len(concatenated.index)
+
+        d = self._translate(
+            sparse_index, sparse_columns, max_rows, max_cols, blank, dxs
+        )
+        return d
+
+    def _render_html(
+        self,
+        sparse_index: bool,
+        sparse_columns: bool,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        **kwargs,
+    ) -> str:
+        """
+        Renders the ``Styler`` including all applied styles to HTML.
+        Generates a dict with necessary kwargs passed to jinja2 template.
+        """
+        d = self._render(sparse_index, sparse_columns, max_rows, max_cols, "&nbsp;")
+        d.update(kwargs)
+        return self.template_html.render(
+            **d,
+            html_table_tpl=self.template_html_table,
+            html_style_tpl=self.template_html_style,
+        )
+
+    def _render_latex(
+        self, sparse_index: bool, sparse_columns: bool, clines: str | None, **kwargs
+    ) -> str:
+        """
+        Render a Styler in latex format
+        """
+        d = self._render(sparse_index, sparse_columns, None, None)
+        self._translate_latex(d, clines=clines)
+        self.template_latex.globals["parse_wrap"] = _parse_latex_table_wrapping
+        self.template_latex.globals["parse_table"] = _parse_latex_table_styles
+        self.template_latex.globals["parse_cell"] = _parse_latex_cell_styles
+        self.template_latex.globals["parse_header"] = _parse_latex_header_span
+        d.update(kwargs)
+        return self.template_latex.render(**d)
+
+    def _render_typst(
+        self,
+        sparse_index: bool,
+        sparse_columns: bool,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        **kwargs,
+    ) -> str:
+        """
+        Render a Styler in typst format
+        """
+        d = self._render(sparse_index, sparse_columns, max_rows, max_cols)
+        d.update(kwargs)
+        return self.template_typst.render(**d)
+
+    def _render_string(
+        self,
+        sparse_index: bool,
+        sparse_columns: bool,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        **kwargs,
+    ) -> str:
+        """
+        Render a Styler in string format
+        """
+        d = self._render(sparse_index, sparse_columns, max_rows, max_cols)
+        d.update(kwargs)
+        return self.template_string.render(**d)
+
+    def _compute(self):
+        """
+        Execute the style functions built up in `self._todo`.
+
+        Relies on the conventions that all style functions go through
+        .apply or .map. The append styles to apply as tuples of
+
+        (application method, *args, **kwargs)
+        """
+        self.ctx.clear()
+        self.ctx_index.clear()
+        self.ctx_columns.clear()
+        r = self
+        for func, args, kwargs in self._todo:
+            r = func(self)(*args, **kwargs)
+        return r
+
+    def _translate(
+        self,
+        sparse_index: bool,
+        sparse_cols: bool,
+        max_rows: int | None = None,
+        max_cols: int | None = None,
+        blank: str = "&nbsp;",
+        dxs: list[dict] | None = None,
+    ):
+        """
+        Process Styler data and settings into a dict for template rendering.
+
+        Convert data and settings from ``Styler`` attributes such as ``self.data``,
+        ``self.tooltips`` including applying any methods in ``self._todo``.
+
+        Parameters
+        ----------
+        sparse_index : bool
+            Whether to sparsify the index or print all hierarchical index elements.
+            Upstream defaults are typically to `pandas.options.styler.sparse.index`.
+        sparse_cols : bool
+            Whether to sparsify the columns or print all hierarchical column elements.
+            Upstream defaults are typically to `pandas.options.styler.sparse.columns`.
+        max_rows, max_cols : int, optional
+            Specific max rows and cols. max_elements always take precedence in render.
+        blank : str
+            Entry to top-left blank cells.
+        dxs : list[dict]
+            The render dicts of the concatenated Stylers.
+
+        Returns
+        -------
+        d : dict
+            The following structure: {uuid, table_styles, caption, head, body,
+            cellstyle, table_attributes}
+        """
+        if dxs is None:
+            dxs = []
+        self.css["blank_value"] = blank
+
+        # construct render dict
+        d = {
+            "uuid": self.uuid,
+            "table_styles": format_table_styles(self.table_styles or []),
+            "caption": self.caption,
+        }
+
+        max_elements = get_option("styler.render.max_elements")
+        max_rows = max_rows if max_rows else get_option("styler.render.max_rows")
+        max_cols = max_cols if max_cols else get_option("styler.render.max_columns")
+        max_rows, max_cols = _get_trimming_maximums(
+            len(self.data.index),
+            len(self.data.columns),
+            max_elements,
+            max_rows,
+            max_cols,
+        )
+
+        self.cellstyle_map_columns: DefaultDict[tuple[CSSPair, ...], list[str]] = (
+            defaultdict(list)
+        )
+        head = self._translate_header(sparse_cols, max_cols)
+        d.update({"head": head})
+
+        # for sparsifying a MultiIndex and for use with latex clines
+        idx_lengths = _get_level_lengths(
+            self.index, sparse_index, max_rows, self.hidden_rows
+        )
+        d.update({"index_lengths": idx_lengths})
+
+        self.cellstyle_map: DefaultDict[tuple[CSSPair, ...], list[str]] = defaultdict(
+            list
+        )
+        self.cellstyle_map_index: DefaultDict[tuple[CSSPair, ...], list[str]] = (
+            defaultdict(list)
+        )
+        body: list = self._translate_body(idx_lengths, max_rows, max_cols)
+        d.update({"body": body})
+
+        ctx_maps = {
+            "cellstyle": "cellstyle_map",
+            "cellstyle_index": "cellstyle_map_index",
+            "cellstyle_columns": "cellstyle_map_columns",
+        }  # add the cell_ids styles map to the render dictionary in right format
+        for k, attr in ctx_maps.items():
+            map = [
+                {"props": list(props), "selectors": selectors}
+                for props, selectors in getattr(self, attr).items()
+            ]
+            d.update({k: map})
+
+        for dx in dxs:  # self.concatenated is not empty
+            d["body"].extend(dx["body"])  # type: ignore[union-attr]
+            d["cellstyle"].extend(dx["cellstyle"])  # type: ignore[union-attr]
+            d["cellstyle_index"].extend(  # type: ignore[union-attr]
+                dx["cellstyle_index"]
+            )
+
+        table_attr = self.table_attributes
+        if not get_option("styler.html.mathjax"):
+            table_attr = table_attr or ""
+            if 'class="' in table_attr:
+                table_attr = table_attr.replace(
+                    'class="', 'class="tex2jax_ignore mathjax_ignore '
+                )
+            else:
+                table_attr += ' class="tex2jax_ignore mathjax_ignore"'
+        d.update({"table_attributes": table_attr})
+
+        if self.tooltips:
+            d = self.tooltips._translate(self, d)
+
+        return d
+
+    def _translate_header(self, sparsify_cols: bool, max_cols: int):
+        """
+        Build each <tr> within table <head> as a list
+
+        Using the structure:
+             +----------------------------+---------------+---------------------------+
+             |  index_blanks ...          | column_name_0 |  column_headers (level_0) |
+          1) |       ..                   |       ..      |             ..            |
+             |  index_blanks ...          | column_name_n |  column_headers (level_n) |
+             +----------------------------+---------------+---------------------------+
+          2) |  index_names (level_0 to level_n) ...      | column_blanks ...         |
+             +----------------------------+---------------+---------------------------+
+
+        Parameters
+        ----------
+        sparsify_cols : bool
+            Whether column_headers section will add colspan attributes (>1) to elements.
+        max_cols : int
+            Maximum number of columns to render. If exceeded will contain `...` filler.
+
+        Returns
+        -------
+        head : list
+            The associated HTML elements needed for template rendering.
+        """
+        # for sparsifying a MultiIndex
+        col_lengths = _get_level_lengths(
+            self.columns, sparsify_cols, max_cols, self.hidden_columns
+        )
+
+        clabels = self.data.columns.tolist()
+        if self.data.columns.nlevels == 1:
+            clabels = [[x] for x in clabels]
+        clabels = list(zip(*clabels))
+
+        head = []
+        # 1) column headers
+        for r, hide in enumerate(self.hide_columns_):
+            if hide or not clabels:
+                continue
+
+            header_row = self._generate_col_header_row(
+                (r, clabels), max_cols, col_lengths
+            )
+            head.append(header_row)
+
+        # 2) index names
+        if (
+            self.data.index.names
+            and com.any_not_none(*self.data.index.names)
+            and not all(self.hide_index_)
+            and not self.hide_index_names
+        ):
+            index_names_row = self._generate_index_names_row(
+                clabels, max_cols, col_lengths
+            )
+            head.append(index_names_row)
+
+        return head
+
+    def _generate_col_header_row(
+        self, iter: Sequence, max_cols: int, col_lengths: dict
+    ):
+        """
+        Generate the row containing column headers:
+
+         +----------------------------+---------------+---------------------------+
+         |  index_blanks ...          | column_name_i |  column_headers (level_i) |
+         +----------------------------+---------------+---------------------------+
+
+        Parameters
+        ----------
+        iter : tuple
+            Looping variables from outer scope
+        max_cols : int
+            Permissible number of columns
+        col_lengths :
+            c
+
+        Returns
+        -------
+        list of elements
+        """
+
+        r, clabels = iter
+
+        # number of index blanks is governed by number of hidden index levels
+        index_blanks = [
+            _element("th", self.css["blank"], self.css["blank_value"], True)
+        ] * (self.index.nlevels - sum(self.hide_index_) - 1)
+
+        name = self.data.columns.names[r]
+
+        is_display = name is not None and not self.hide_column_names
+        value = name if is_display else self.css["blank_value"]
+        display_value = (
+            self._display_funcs_column_names[r](value) if is_display else None
+        )
+        column_name = [
+            _element(
+                "th",
+                (
+                    f"{self.css['blank']} {self.css['level']}{r}"
+                    if name is None
+                    else f"{self.css['index_name']} {self.css['level']}{r}"
+                ),
+                value,
+                not all(self.hide_index_),
+                display_value=display_value,
+            )
+        ]
+
+        column_headers: list = []
+        visible_col_count: int = 0
+        for c, value in enumerate(clabels[r]):
+            header_element_visible = _is_visible(c, r, col_lengths)
+            if header_element_visible:
+                visible_col_count += col_lengths.get((r, c), 0)
+            if self._check_trim(
+                visible_col_count,
+                max_cols,
+                column_headers,
+                "th",
+                f"{self.css['col_heading']} {self.css['level']}{r} "
+                f"{self.css['col_trim']}",
+            ):
+                break
+
+            header_element = _element(
+                "th",
+                (
+                    f"{self.css['col_heading']} {self.css['level']}{r} "
+                    f"{self.css['col']}{c}"
+                ),
+                value,
+                header_element_visible,
+                display_value=self._display_funcs_columns[(r, c)](value),
+                attributes=(
+                    f'colspan="{col_lengths.get((r, c), 0)}"'
+                    if col_lengths.get((r, c), 0) > 1
+                    else ""
+                ),
+            )
+
+            if self.cell_ids:
+                header_element["id"] = f"{self.css['level']}{r}_{self.css['col']}{c}"
+            if (
+                header_element_visible
+                and (r, c) in self.ctx_columns
+                and self.ctx_columns[r, c]
+            ):
+                header_element["id"] = f"{self.css['level']}{r}_{self.css['col']}{c}"
+                self.cellstyle_map_columns[tuple(self.ctx_columns[r, c])].append(
+                    f"{self.css['level']}{r}_{self.css['col']}{c}"
+                )
+
+            column_headers.append(header_element)
+
+        return index_blanks + column_name + column_headers
+
+    def _generate_index_names_row(
+        self, iter: Sequence, max_cols: int, col_lengths: dict
+    ):
+        """
+        Generate the row containing index names
+
+         +----------------------------+---------------+---------------------------+
+         |  index_names (level_0 to level_n) ...      | column_blanks ...         |
+         +----------------------------+---------------+---------------------------+
+
+        Parameters
+        ----------
+        iter : tuple
+            Looping variables from outer scope
+        max_cols : int
+            Permissible number of columns
+
+        Returns
+        -------
+        list of elements
+        """
+
+        clabels = iter
+
+        index_names = [
+            _element(
+                "th",
+                f"{self.css['index_name']} {self.css['level']}{c}",
+                self.css["blank_value"] if name is None else name,
+                not self.hide_index_[c],
+                display_value=(
+                    None if name is None else self._display_funcs_index_names[c](name)
+                ),
+            )
+            for c, name in enumerate(self.data.index.names)
+        ]
+
+        column_blanks: list = []
+        visible_col_count: int = 0
+        if clabels:
+            last_level = self.columns.nlevels - 1  # use last level since never sparsed
+            for c, value in enumerate(clabels[last_level]):
+                header_element_visible = _is_visible(c, last_level, col_lengths)
+                if header_element_visible:
+                    visible_col_count += 1
+                if self._check_trim(
+                    visible_col_count,
+                    max_cols,
+                    column_blanks,
+                    "th",
+                    f"{self.css['blank']} {self.css['col']}{c} {self.css['col_trim']}",
+                    self.css["blank_value"],
+                ):
+                    break
+
+                column_blanks.append(
+                    _element(
+                        "th",
+                        f"{self.css['blank']} {self.css['col']}{c}",
+                        self.css["blank_value"],
+                        c not in self.hidden_columns,
+                    )
+                )
+
+        return index_names + column_blanks
+
+    def _translate_body(self, idx_lengths: dict, max_rows: int, max_cols: int):
+        """
+        Build each <tr> within table <body> as a list
+
+        Use the following structure:
+          +--------------------------------------------+---------------------------+
+          |  index_header_0    ...    index_header_n   |  data_by_column   ...     |
+          +--------------------------------------------+---------------------------+
+
+        Also add elements to the cellstyle_map for more efficient grouped elements in
+        <style></style> block
+
+        Parameters
+        ----------
+        sparsify_index : bool
+            Whether index_headers section will add rowspan attributes (>1) to elements.
+
+        Returns
+        -------
+        body : list
+            The associated HTML elements needed for template rendering.
+        """
+        rlabels = self.data.index.tolist()
+        if not isinstance(self.data.index, MultiIndex):
+            rlabels = [[x] for x in rlabels]
+
+        body: list = []
+        visible_row_count: int = 0
+        for r, row_tup in [
+            z for z in enumerate(self.data.itertuples()) if z[0] not in self.hidden_rows
+        ]:
+            visible_row_count += 1
+            if self._check_trim(
+                visible_row_count,
+                max_rows,
+                body,
+                "row",
+            ):
+                break
+
+            body_row = self._generate_body_row(
+                (r, row_tup, rlabels), max_cols, idx_lengths
+            )
+            body.append(body_row)
+        return body
+
+    def _check_trim(
+        self,
+        count: int,
+        max: int,
+        obj: list,
+        element: str,
+        css: str | None = None,
+        value: str = "...",
+    ) -> bool:
+        """
+        Indicates whether to break render loops and append a trimming indicator
+
+        Parameters
+        ----------
+        count : int
+            The loop count of previous visible items.
+        max : int
+            The allowable rendered items in the loop.
+        obj : list
+            The current render collection of the rendered items.
+        element : str
+            The type of element to append in the case a trimming indicator is needed.
+        css : str, optional
+            The css to add to the trimming indicator element.
+        value : str, optional
+            The value of the elements display if necessary.
+
+        Returns
+        -------
+        result : bool
+            Whether a trimming element was required and appended.
+        """
+        if count > max:
+            if element == "row":
+                obj.append(self._generate_trimmed_row(max))
+            else:
+                obj.append(_element(element, css, value, True, attributes=""))
+            return True
+        return False
+
+    def _generate_trimmed_row(self, max_cols: int) -> list:
+        """
+        When a render has too many rows we generate a trimming row containing "..."
+
+        Parameters
+        ----------
+        max_cols : int
+            Number of permissible columns
+
+        Returns
+        -------
+        list of elements
+        """
+        index_headers = [
+            _element(
+                "th",
+                (
+                    f"{self.css['row_heading']} {self.css['level']}{c} "
+                    f"{self.css['row_trim']}"
+                ),
+                "...",
+                not self.hide_index_[c],
+                attributes="",
+            )
+            for c in range(self.data.index.nlevels)
+        ]
+
+        data: list = []
+        visible_col_count: int = 0
+        for c, _ in enumerate(self.columns):
+            data_element_visible = c not in self.hidden_columns
+            if data_element_visible:
+                visible_col_count += 1
+            if self._check_trim(
+                visible_col_count,
+                max_cols,
+                data,
+                "td",
+                f"{self.css['data']} {self.css['row_trim']} {self.css['col_trim']}",
+            ):
+                break
+
+            data.append(
+                _element(
+                    "td",
+                    f"{self.css['data']} {self.css['col']}{c} {self.css['row_trim']}",
+                    "...",
+                    data_element_visible,
+                    attributes="",
+                )
+            )
+
+        return index_headers + data
+
+    def _generate_body_row(
+        self,
+        iter: tuple,
+        max_cols: int,
+        idx_lengths: dict,
+    ):
+        """
+        Generate a regular row for the body section of appropriate format.
+
+          +--------------------------------------------+---------------------------+
+          |  index_header_0    ...    index_header_n   |  data_by_column   ...     |
+          +--------------------------------------------+---------------------------+
+
+        Parameters
+        ----------
+        iter : tuple
+            Iterable from outer scope: row number, row data tuple, row index labels.
+        max_cols : int
+            Number of permissible columns.
+        idx_lengths : dict
+            A map of the sparsification structure of the index
+
+        Returns
+        -------
+            list of elements
+        """
+        r, row_tup, rlabels = iter
+
+        index_headers = []
+        for c, value in enumerate(rlabels[r]):
+            header_element_visible = (
+                _is_visible(r, c, idx_lengths) and not self.hide_index_[c]
+            )
+            header_element = _element(
+                "th",
+                (
+                    f"{self.css['row_heading']} {self.css['level']}{c} "
+                    f"{self.css['row']}{r}"
+                ),
+                value,
+                header_element_visible,
+                display_value=self._display_funcs_index[(r, c)](value),
+                attributes=(
+                    f'rowspan="{idx_lengths.get((c, r), 0)}"'
+                    if idx_lengths.get((c, r), 0) > 1
+                    else ""
+                ),
+            )
+
+            if self.cell_ids:
+                header_element["id"] = (
+                    f"{self.css['level']}{c}_{self.css['row']}{r}"  # id is given
+                )
+            if (
+                header_element_visible
+                and (r, c) in self.ctx_index
+                and self.ctx_index[r, c]
+            ):
+                # always add id if a style is specified
+                header_element["id"] = f"{self.css['level']}{c}_{self.css['row']}{r}"
+                self.cellstyle_map_index[tuple(self.ctx_index[r, c])].append(
+                    f"{self.css['level']}{c}_{self.css['row']}{r}"
+                )
+
+            index_headers.append(header_element)
+
+        data: list = []
+        visible_col_count: int = 0
+        for c, value in enumerate(row_tup[1:]):
+            data_element_visible = (
+                c not in self.hidden_columns and r not in self.hidden_rows
+            )
+            if data_element_visible:
+                visible_col_count += 1
+            if self._check_trim(
+                visible_col_count,
+                max_cols,
+                data,
+                "td",
+                f"{self.css['data']} {self.css['row']}{r} {self.css['col_trim']}",
+            ):
+                break
+
+            # add custom classes from cell context
+            cls = ""
+            if (r, c) in self.cell_context:
+                cls = " " + self.cell_context[r, c]
+
+            data_element = _element(
+                "td",
+                (f"{self.css['data']} {self.css['row']}{r} {self.css['col']}{c}{cls}"),
+                value,
+                data_element_visible,
+                attributes="",
+                display_value=self._display_funcs[(r, c)](value),
+            )
+
+            if self.cell_ids:
+                data_element["id"] = f"{self.css['row']}{r}_{self.css['col']}{c}"
+            if data_element_visible and (r, c) in self.ctx and self.ctx[r, c]:
+                # always add id if needed due to specified style
+                data_element["id"] = f"{self.css['row']}{r}_{self.css['col']}{c}"
+                self.cellstyle_map[tuple(self.ctx[r, c])].append(
+                    f"{self.css['row']}{r}_{self.css['col']}{c}"
+                )
+
+            data.append(data_element)
+
+        return index_headers + data
+
+    def _translate_latex(self, d: dict, clines: str | None) -> None:
+        r"""
+        Post-process the default render dict for the LaTeX template format.
+
+        Processing items included are:
+          - Remove hidden columns from the non-headers part of the body.
+          - Place cellstyles directly in td cells rather than use cellstyle_map.
+          - Remove hidden indexes or reinsert missing th elements if part of multiindex
+            or multirow sparsification (so that \multirow and \multicol work correctly).
+        """
+        index_levels = self.index.nlevels
+        # GH 52218
+        visible_index_level_n = max(1, index_levels - sum(self.hide_index_))
+        d["head"] = [
+            [
+                {**col, "cellstyle": self.ctx_columns[r, c - visible_index_level_n]}
+                for c, col in enumerate(row)
+                if col["is_visible"]
+            ]
+            for r, row in enumerate(d["head"])
+        ]
+
+        def _concatenated_visible_rows(obj, n, row_indices):
+            """
+            Extract all visible row indices recursively from concatenated stylers.
+            """
+            row_indices.extend(
+                [r + n for r in range(len(obj.index)) if r not in obj.hidden_rows]
+            )
+            n += len(obj.index)
+            for concatenated in obj.concatenated:
+                n = _concatenated_visible_rows(concatenated, n, row_indices)
+            return n
+
+        def concatenated_visible_rows(obj):
+            row_indices: list[int] = []
+            _concatenated_visible_rows(obj, 0, row_indices)
+            # TODO try to consolidate the concat visible rows
+            # methods to a single function / recursion for simplicity
+            return row_indices
+
+        body = []
+        for r, row in zip(concatenated_visible_rows(self), d["body"]):
+            # note: cannot enumerate d["body"] because rows were dropped if hidden
+            # during _translate_body so must zip to acquire the true r-index associated
+            # with the ctx obj which contains the cell styles.
+            if all(self.hide_index_):
+                row_body_headers = []
+            else:
+                row_body_headers = [
+                    {
+                        **col,
+                        "display_value": (
+                            col["display_value"] if col["is_visible"] else ""
+                        ),
+                        "cellstyle": self.ctx_index[r, c],
+                    }
+                    for c, col in enumerate(row[:index_levels])
+                    if (col["type"] == "th" and not self.hide_index_[c])
+                ]
+
+            row_body_cells = [
+                {**col, "cellstyle": self.ctx[r, c]}
+                for c, col in enumerate(row[index_levels:])
+                if (col["is_visible"] and col["type"] == "td")
+            ]
+
+            body.append(row_body_headers + row_body_cells)
+        d["body"] = body
+
+        # clines are determined from info on index_lengths and hidden_rows and input
+        # to a dict defining which row clines should be added in the template.
+        if clines not in [
+            None,
+            "all;data",
+            "all;index",
+            "skip-last;data",
+            "skip-last;index",
+        ]:
+            raise ValueError(
+                f"`clines` value of {clines} is invalid. Should either be None or one "
+                f"of 'all;data', 'all;index', 'skip-last;data', 'skip-last;index'."
+            )
+        if clines is not None:
+            data_len = len(row_body_cells) if "data" in clines and d["body"] else 0
+
+            d["clines"] = defaultdict(list)
+            visible_row_indexes: list[int] = [
+                r for r in range(len(self.data.index)) if r not in self.hidden_rows
+            ]
+            visible_index_levels: list[int] = [
+                i for i in range(index_levels) if not self.hide_index_[i]
+            ]
+            for rn, r in enumerate(visible_row_indexes):
+                for lvln, lvl in enumerate(visible_index_levels):
+                    if lvl == index_levels - 1 and "skip-last" in clines:
+                        continue
+                    idx_len = d["index_lengths"].get((lvl, r), None)
+                    if idx_len is not None:  # i.e. not a sparsified entry
+                        d["clines"][rn + idx_len].append(
+                            f"\\cline{{{lvln + 1}-{len(visible_index_levels) + data_len}}}"  # noqa: E501
+                        )
+
+    def format(
+        self,
+        formatter: ExtFormatter | None = None,
+        subset: Subset | None = None,
+        na_rep: str | None = None,
+        precision: int | None = None,
+        decimal: str = ".",
+        thousands: str | None = None,
+        escape: str | None = None,
+        hyperlinks: str | None = None,
+    ) -> StylerRenderer:
+        r"""
+        Format the text display value of cells.
+
+        Parameters
+        ----------
+        formatter : str, callable, dict or None
+            Object to define how values are displayed. See notes.
+        subset : label, array-like, IndexSlice, optional
+            A valid 2d input to `DataFrame.loc[<subset>]`, or, in the case of a 1d input
+            or single key, to `DataFrame.loc[:, <subset>]` where the columns are
+            prioritised, to limit ``data`` to *before* applying the function.
+        na_rep : str, optional
+            Representation for missing values.
+            If ``na_rep`` is None, no special formatting is applied.
+        precision : int, optional
+            Floating point precision to use for display purposes, if not determined by
+            the specified ``formatter``.
+
+            .. versionadded:: 1.3.0
+
+        decimal : str, default "."
+            Character used as decimal separator for floats, complex and integers.
+
+            .. versionadded:: 1.3.0
+
+        thousands : str, optional, default None
+            Character used as thousands separator for floats, complex and integers.
+
+            .. versionadded:: 1.3.0
+
+        escape : str, optional
+            Use 'html' to replace the characters ``&``, ``<``, ``>``, ``'``, and ``"``
+            in cell display string with HTML-safe sequences.
+            Use 'latex' to replace the characters ``&``, ``%``, ``$``, ``#``, ``_``,
+            ``{``, ``}``, ``~``, ``^``, and ``\`` in the cell display string with
+            LaTeX-safe sequences.
+            Use 'latex-math' to replace the characters the same way as in 'latex' mode,
+            except for math substrings, which either are surrounded
+            by two characters ``$`` or start with the character ``\(`` and
+            end with ``\)``. Escaping is done before ``formatter``.
+
+            .. versionadded:: 1.3.0
+
+        hyperlinks : {"html", "latex"}, optional
+            Convert string patterns containing https://, http://, ftp:// or www. to
+            HTML <a> tags as clickable URL hyperlinks if "html", or LaTeX \href
+            commands if "latex".
+
+            .. versionadded:: 1.4.0
+
+        Returns
+        -------
+        Styler
+            Returns itself for chaining.
+
+        See Also
+        --------
+        Styler.format_index: Format the text display value of index labels.
+
+        Notes
+        -----
+        This method assigns a formatting function, ``formatter``, to each cell in the
+        DataFrame. If ``formatter`` is ``None``, then the default formatter is used.
+        If a callable then that function should take a data value as input and return
+        a displayable representation, such as a string. If ``formatter`` is
+        given as a string this is assumed to be a valid Python format specification
+        and is wrapped to a callable as ``string.format(x)``. If a ``dict`` is given,
+        keys should correspond to column names, and values should be string or
+        callable, as above.
+
+        The default formatter currently expresses floats and complex numbers with the
+        pandas display precision unless using the ``precision`` argument here. The
+        default formatter does not adjust the representation of missing values unless
+        the ``na_rep`` argument is used.
+
+        The ``subset`` argument defines which region to apply the formatting function
+        to. If the ``formatter`` argument is given in dict form but does not include
+        all columns within the subset then these columns will have the default formatter
+        applied. Any columns in the formatter dict excluded from the subset will
+        be ignored.
+
+        When using a ``formatter`` string the dtypes must be compatible, otherwise a
+        `ValueError` will be raised.
+
+        When instantiating a Styler, default formatting can be applied by setting the
+        ``pandas.options``:
+
+          - ``styler.format.formatter``: default None.
+          - ``styler.format.na_rep``: default None.
+          - ``styler.format.precision``: default 6.
+          - ``styler.format.decimal``: default ".".
+          - ``styler.format.thousands``: default None.
+          - ``styler.format.escape``: default None.
+
+        .. warning::
+           `Styler.format` is ignored when using the output format `Styler.to_excel`,
+           since Excel and Python have inherently different formatting structures.
+           However, it is possible to use the `number-format` pseudo CSS attribute
+           to force Excel permissible formatting. See examples.
+
+        Examples
+        --------
+        Using ``na_rep`` and ``precision`` with the default ``formatter``
+
+        >>> df = pd.DataFrame([[np.nan, 1.0, 'A'], [2.0, np.nan, 3.0]])
+        >>> df.style.format(na_rep='MISS', precision=3)  # doctest: +SKIP
+                0       1       2
+        0    MISS   1.000       A
+        1   2.000    MISS   3.000
+
+        Using a ``formatter`` specification on consistent column dtypes
+
+        >>> df.style.format('{:.2f}', na_rep='MISS', subset=[0, 1])  # doctest: +SKIP
+                0      1          2
+        0    MISS   1.00          A
+        1    2.00   MISS   3.000000
+
+        Using the default ``formatter`` for unspecified columns
+
+        >>> df.style.format({0: '{:.2f}', 1: '£ {:.1f}'},
+        ...                 na_rep='MISS', precision=1)  # doctest: +SKIP
+                 0      1     2
+        0    MISS   £ 1.0     A
+        1    2.00    MISS   3.0
+
+        Multiple ``na_rep`` or ``precision`` specifications under the default
+        ``formatter``.
+
+        >>> (df.style.format(na_rep='MISS', precision=1, subset=[0]).format(
+        ...     na_rep='PASS', precision=2, subset=[1, 2]))  # doctest: +SKIP
+                0      1      2
+        0    MISS   1.00      A
+        1     2.0   PASS   3.00
+
+        Using a callable ``formatter`` function.
+
+        >>> func = lambda s: 'STRING' if isinstance(s, str) else 'FLOAT'
+        >>> df.style.format({0: '{:.1f}', 2: func},
+        ...                 precision=4, na_rep='MISS')  # doctest: +SKIP
+                0        1        2
+        0    MISS   1.0000   STRING
+        1     2.0     MISS    FLOAT
+
+        Using a ``formatter`` with HTML ``escape`` and ``na_rep``.
+
+        >>> df = pd.DataFrame([['<div></div>', '"A&B"', None]])
+        >>> s = df.style.format(
+        ...     '<a href="a.com/{0}">{0}</a>', escape="html", na_rep="NA")
+        >>> s.to_html()  # doctest: +SKIP
+        ...
+        <td .. ><a href="a.com/&lt;div&gt;&lt;/div&gt;">&lt;div&gt;&lt;/div&gt;</a></td>
+        <td .. ><a href="a.com/&#34;A&amp;B&#34;">&#34;A&amp;B&#34;</a></td>
+        <td .. >NA</td>
+        ...
+
+        Using a ``formatter`` with ``escape`` in 'latex' mode.
+
+        >>> df = pd.DataFrame([["123"], ["~ ^"], ["$%#"]])
+        >>> df.style.format("\\textbf{{{}}}",
+        ...                 escape="latex").to_latex()  # doctest: +SKIP
+        \begin{tabular}{ll}
+         & 0 \\
+        0 & \textbf{123} \\
+        1 & \textbf{\textasciitilde \space \textasciicircum } \\
+        2 & \textbf{\$\%\#} \\
+        \end{tabular}
+
+        Applying ``escape`` in 'latex-math' mode. In the example below
+        we enter math mode using the character ``$``.
+
+        >>> df = pd.DataFrame([
+        ...     [r"$\sum_{i=1}^{10} a_i$ a~b $\alpha = \frac{\beta}{\zeta^2}$"],
+        ...     [r"%#^ $ \$x^2 $"]])
+        >>> df.style.format(escape="latex-math").to_latex()  # doctest: +SKIP
+        \begin{tabular}{ll}
+         & 0 \\
+        0 & $\sum_{i=1}^{10} a_i$ a\textasciitilde b $\alpha = \frac{\beta}{\zeta^2}$ \\
+        1 & \%\#\textasciicircum \space $ \$x^2 $ \\
+        \end{tabular}
+
+        We can use the character ``\(`` to enter math mode and the character ``\)``
+        to close math mode.
+
+        >>> df = pd.DataFrame([
+        ...     [r"\(\sum_{i=1}^{10} a_i\) a~b \(\alpha = \frac{\beta}{\zeta^2}\)"],
+        ...     [r"%#^ \( \$x^2 \)"]])
+        >>> df.style.format(escape="latex-math").to_latex()  # doctest: +SKIP
+        \begin{tabular}{ll}
+         & 0 \\
+        0 & \(\sum_{i=1}^{10} a_i\) a\textasciitilde b \(\alpha
+        = \frac{\beta}{\zeta^2}\) \\
+        1 & \%\#\textasciicircum \space \( \$x^2 \) \\
+        \end{tabular}
+
+        If we have in one DataFrame cell a combination of both shorthands
+        for math formulas, the shorthand with the sign ``$`` will be applied.
+
+        >>> df = pd.DataFrame([
+        ...     [r"\( x^2 \)  $x^2$"],
+        ...     [r"$\frac{\beta}{\zeta}$ \(\frac{\beta}{\zeta}\)"]])
+        >>> df.style.format(escape="latex-math").to_latex()  # doctest: +SKIP
+        \begin{tabular}{ll}
+         & 0 \\
+        0 & \textbackslash ( x\textasciicircum 2 \textbackslash )  $x^2$ \\
+        1 & $\frac{\beta}{\zeta}$ \textbackslash (\textbackslash
+        frac\{\textbackslash beta\}\{\textbackslash zeta\}\textbackslash ) \\
+        \end{tabular}
+
+        Pandas defines a `number-format` pseudo CSS attribute instead of the `.format`
+        method to create `to_excel` permissible formatting. Note that semi-colons are
+        CSS protected characters but used as separators in Excel's format string.
+        Replace semi-colons with the section separator character (ASCII-245) when
+        defining the formatting here.
+
+        >>> df = pd.DataFrame({"A": [1, 0, -1]})
+        >>> pseudo_css = "number-format: 0§[Red](0)§-§@;"
+        >>> filename = "formatted_file.xlsx"
+        >>> df.style.map(lambda v: pseudo_css).to_excel(filename)  # doctest: +SKIP
+
+        .. figure:: ../../_static/style/format_excel_css.png
+        """
+        if all(
+            (
+                formatter is None,
+                subset is None,
+                precision is None,
+                decimal == ".",
+                thousands is None,
+                na_rep is None,
+                escape is None,
+                hyperlinks is None,
+            )
+        ):
+            self._display_funcs.clear()
+            return self  # clear the formatter / revert to default and avoid looping
+
+        subset = slice(None) if subset is None else subset
+        subset = non_reducing_slice(subset)
+        data = self.data.loc[subset]
+
+        if not isinstance(formatter, dict):
+            formatter = {col: formatter for col in data.columns}
+
+        cis = self.columns.get_indexer_for(data.columns)
+        ris = self.index.get_indexer_for(data.index)
+        for ci in cis:
+            format_func = _maybe_wrap_formatter(
+                formatter.get(self.columns[ci]),
+                na_rep=na_rep,
+                precision=precision,
+                decimal=decimal,
+                thousands=thousands,
+                escape=escape,
+                hyperlinks=hyperlinks,
+            )
+            for ri in ris:
+                self._display_funcs[(ri, ci)] = format_func
+
+        return self
+
+    def format_index(
+        self,
+        formatter: ExtFormatter | None = None,
+        axis: Axis = 0,
+        level: Level | list[Level] | None = None,
+        na_rep: str | None = None,
+        precision: int | None = None,
+        decimal: str = ".",
+        thousands: str | None = None,
+        escape: str | None = None,
+        hyperlinks: str | None = None,
+    ) -> StylerRenderer:
+        r"""
+        Format the text display value of index labels or column headers.
+
+        .. versionadded:: 1.4.0
+
+        Parameters
+        ----------
+        formatter : str, callable, dict or None
+            Object to define how values are displayed. See notes.
+        axis : {0, "index", 1, "columns"}
+            Whether to apply the formatter to the index or column headers.
+        level : int, str, list
+            The level(s) over which to apply the generic formatter.
+        na_rep : str, optional
+            Representation for missing values.
+            If ``na_rep`` is None, no special formatting is applied.
+        precision : int, optional
+            Floating point precision to use for display purposes, if not determined by
+            the specified ``formatter``.
+        decimal : str, default "."
+            Character used as decimal separator for floats, complex and integers.
+        thousands : str, optional, default None
+            Character used as thousands separator for floats, complex and integers.
+        escape : str, optional
+            Use 'html' to replace the characters ``&``, ``<``, ``>``, ``'``, and ``"``
+            in cell display string with HTML-safe sequences.
+            Use 'latex' to replace the characters ``&``, ``%``, ``$``, ``#``, ``_``,
+            ``{``, ``}``, ``~``, ``^``, and ``\`` in the cell display string with
+            LaTeX-safe sequences.
+            Escaping is done before ``formatter``.
+        hyperlinks : {"html", "latex"}, optional
+            Convert string patterns containing https://, http://, ftp:// or www. to
+            HTML <a> tags as clickable URL hyperlinks if "html", or LaTeX \href
+            commands if "latex".
+
+        Returns
+        -------
+        Styler
+            Returns itself for chaining.
+
+        See Also
+        --------
+        Styler.format: Format the text display value of data cells.
+
+        Notes
+        -----
+        This method assigns a formatting function, ``formatter``, to each level label
+        in the DataFrame's index or column headers. If ``formatter`` is ``None``,
+        then the default formatter is used.
+        If a callable then that function should take a label value as input and return
+        a displayable representation, such as a string. If ``formatter`` is
+        given as a string this is assumed to be a valid Python format specification
+        and is wrapped to a callable as ``string.format(x)``. If a ``dict`` is given,
+        keys should correspond to MultiIndex level numbers or names, and values should
+        be string or callable, as above.
+
+        The default formatter currently expresses floats and complex numbers with the
+        pandas display precision unless using the ``precision`` argument here. The
+        default formatter does not adjust the representation of missing values unless
+        the ``na_rep`` argument is used.
+
+        The ``level`` argument defines which levels of a MultiIndex to apply the
+        method to. If the ``formatter`` argument is given in dict form but does
+        not include all levels within the level argument then these unspecified levels
+        will have the default formatter applied. Any levels in the formatter dict
+        specifically excluded from the level argument will be ignored.
+
+        When using a ``formatter`` string the dtypes must be compatible, otherwise a
+        `ValueError` will be raised.
+
+        .. warning::
+           `Styler.format_index` is ignored when using the output format
+           `Styler.to_excel`, since Excel and Python have inherently different
+           formatting structures.
+           However, it is possible to use the `number-format` pseudo CSS attribute
+           to force Excel permissible formatting. See documentation for `Styler.format`.
+
+        Examples
+        --------
+        Using ``na_rep`` and ``precision`` with the default ``formatter``
+
+        >>> df = pd.DataFrame([[1, 2, 3]], columns=[2.0, np.nan, 4.0])
+        >>> df.style.format_index(axis=1, na_rep='MISS', precision=3)  # doctest: +SKIP
+            2.000    MISS   4.000
+        0       1       2       3
+
+        Using a ``formatter`` specification on consistent dtypes in a level
+
+        >>> df.style.format_index('{:.2f}', axis=1, na_rep='MISS')  # doctest: +SKIP
+             2.00   MISS    4.00
+        0       1      2       3
+
+        Using the default ``formatter`` for unspecified levels
+
+        >>> df = pd.DataFrame([[1, 2, 3]],
+        ...                   columns=pd.MultiIndex.from_arrays(
+        ...                   [["a", "a", "b"], [2, np.nan, 4]]))
+        >>> df.style.format_index({0: lambda v: v.upper()}, axis=1, precision=1)
+        ... # doctest: +SKIP
+                       A       B
+              2.0    nan     4.0
+        0       1      2       3
+
+        Using a callable ``formatter`` function.
+
+        >>> func = lambda s: 'STRING' if isinstance(s, str) else 'FLOAT'
+        >>> df.style.format_index(func, axis=1, na_rep='MISS')
+        ... # doctest: +SKIP
+                  STRING  STRING
+            FLOAT   MISS   FLOAT
+        0       1      2       3
+
+        Using a ``formatter`` with HTML ``escape`` and ``na_rep``.
+
+        >>> df = pd.DataFrame([[1, 2, 3]], columns=['"A"', 'A&B', None])
+        >>> s = df.style.format_index('$ {0}', axis=1, escape="html", na_rep="NA")
+        ... # doctest: +SKIP
+        <th .. >$ &#34;A&#34;</th>
+        <th .. >$ A&amp;B</th>
+        <th .. >NA</td>
+        ...
+
+        Using a ``formatter`` with LaTeX ``escape``.
+
+        >>> df = pd.DataFrame([[1, 2, 3]], columns=["123", "~", "$%#"])
+        >>> df.style.format_index("\\textbf{{{}}}", escape="latex", axis=1).to_latex()
+        ... # doctest: +SKIP
+        \begin{tabular}{lrrr}
+        {} & {\textbf{123}} & {\textbf{\textasciitilde }} & {\textbf{\$\%\#}} \\
+        0 & 1 & 2 & 3 \\
+        \end{tabular}
+        """
+        axis = self.data._get_axis_number(axis)
+        if axis == 0:
+            display_funcs_, obj = self._display_funcs_index, self.index
+        else:
+            display_funcs_, obj = self._display_funcs_columns, self.columns
+        levels_ = refactor_levels(level, obj)
+
+        if all(
+            (
+                formatter is None,
+                level is None,
+                precision is None,
+                decimal == ".",
+                thousands is None,
+                na_rep is None,
+                escape is None,
+                hyperlinks is None,
+            )
+        ):
+            display_funcs_.clear()
+            return self  # clear the formatter / revert to default and avoid looping
+
+        if not isinstance(formatter, dict):
+            formatter = {level: formatter for level in levels_}
+        else:
+            formatter = {
+                obj._get_level_number(level): formatter_
+                for level, formatter_ in formatter.items()
+            }
+
+        for lvl in levels_:
+            format_func = _maybe_wrap_formatter(
+                formatter.get(lvl),
+                na_rep=na_rep,
+                precision=precision,
+                decimal=decimal,
+                thousands=thousands,
+                escape=escape,
+                hyperlinks=hyperlinks,
+            )
+
+            for idx in [(i, lvl) if axis == 0 else (lvl, i) for i in range(len(obj))]:
+                display_funcs_[idx] = format_func
+
+        return self
+
+    def relabel_index(
+        self,
+        labels: Sequence | Index,
+        axis: Axis = 0,
+        level: Level | list[Level] | None = None,
+    ) -> StylerRenderer:
+        r"""
+        Relabel the index, or column header, keys to display a set of specified values.
+
+        .. versionadded:: 1.5.0
+
+        Parameters
+        ----------
+        labels : list-like or Index
+            New labels to display. Must have same length as the underlying values not
+            hidden.
+        axis : {"index", 0, "columns", 1}
+            Apply to the index or columns.
+        level : int, str, list, optional
+            The level(s) over which to apply the new labels. If `None` will apply
+            to all levels of an Index or MultiIndex which are not hidden.
+
+        Returns
+        -------
+        Styler
+            Returns itself for chaining.
+
+        See Also
+        --------
+        Styler.format_index: Format the text display value of index or column headers.
+        Styler.hide: Hide the index, column headers, or specified data from display.
+
+        Notes
+        -----
+        As part of Styler, this method allows the display of an index to be
+        completely user-specified without affecting the underlying DataFrame data,
+        index, or column headers. This means that the flexibility of indexing is
+        maintained whilst the final display is customisable.
+
+        Since Styler is designed to be progressively constructed with method chaining,
+        this method is adapted to react to the **currently specified hidden elements**.
+        This is useful because it means one does not have to specify all the new
+        labels if the majority of an index, or column headers, have already been hidden.
+        The following produce equivalent display (note the length of ``labels`` in
+        each case).
+
+        .. code-block:: python
+
+            # relabel first, then hide
+            df = pd.DataFrame({"col": ["a", "b", "c"]})
+            df.style.relabel_index(["A", "B", "C"]).hide([0, 1])
+            # hide first, then relabel
+            df = pd.DataFrame({"col": ["a", "b", "c"]})
+            df.style.hide([0, 1]).relabel_index(["C"])
+
+        This method should be used, rather than :meth:`Styler.format_index`, in one of
+        the following cases (see examples):
+
+          - A specified set of labels are required which are not a function of the
+            underlying index keys.
+          - The function of the underlying index keys requires a counter variable,
+            such as those available upon enumeration.
+
+        Examples
+        --------
+        Basic use
+
+        >>> df = pd.DataFrame({"col": ["a", "b", "c"]})
+        >>> df.style.relabel_index(["A", "B", "C"])  # doctest: +SKIP
+             col
+        A      a
+        B      b
+        C      c
+
+        Chaining with pre-hidden elements
+
+        >>> df.style.hide([0, 1]).relabel_index(["C"])  # doctest: +SKIP
+             col
+        C      c
+
+        Using a MultiIndex
+
+        >>> midx = pd.MultiIndex.from_product([[0, 1], [0, 1], [0, 1]])
+        >>> df = pd.DataFrame({"col": list(range(8))}, index=midx)
+        >>> styler = df.style  # doctest: +SKIP
+                  col
+        0  0  0     0
+              1     1
+           1  0     2
+              1     3
+        1  0  0     4
+              1     5
+           1  0     6
+              1     7
+        >>> styler.hide(
+        ...     (midx.get_level_values(0) == 0) | (midx.get_level_values(1) == 0)
+        ... )
+        ... # doctest: +SKIP
+        >>> styler.hide(level=[0, 1])  # doctest: +SKIP
+        >>> styler.relabel_index(["binary6", "binary7"])  # doctest: +SKIP
+                  col
+        binary6     6
+        binary7     7
+
+        We can also achieve the above by indexing first and then re-labeling
+
+        >>> styler = df.loc[[(1, 1, 0), (1, 1, 1)]].style
+        >>> styler.hide(level=[0, 1]).relabel_index(["binary6", "binary7"])
+        ... # doctest: +SKIP
+                  col
+        binary6     6
+        binary7     7
+
+        Defining a formatting function which uses an enumeration counter. Also note
+        that the value of the index key is passed in the case of string labels so it
+        can also be inserted into the label, using curly brackets (or double curly
+        brackets if the string if pre-formatted),
+
+        >>> df = pd.DataFrame({"samples": np.random.rand(10)})
+        >>> styler = df.loc[np.random.randint(0, 10, 3)].style
+        >>> styler.relabel_index([f"sample{i + 1} ({{}})" for i in range(3)])
+        ... # doctest: +SKIP
+                         samples
+        sample1 (5)     0.315811
+        sample2 (0)     0.495941
+        sample3 (2)     0.067946
+        """
+        axis = self.data._get_axis_number(axis)
+        if axis == 0:
+            display_funcs_, obj = self._display_funcs_index, self.index
+            hidden_labels, hidden_lvls = self.hidden_rows, self.hide_index_
+        else:
+            display_funcs_, obj = self._display_funcs_columns, self.columns
+            hidden_labels, hidden_lvls = self.hidden_columns, self.hide_columns_
+        visible_len = len(obj) - len(set(hidden_labels))
+        if len(labels) != visible_len:
+            raise ValueError(
+                "``labels`` must be of length equal to the number of "
+                f"visible labels along ``axis`` ({visible_len})."
+            )
+
+        if level is None:
+            level = [i for i in range(obj.nlevels) if not hidden_lvls[i]]
+        levels_ = refactor_levels(level, obj)
+
+        def alias_(x, value):
+            if isinstance(value, str):
+                return value.format(x)
+            return value
+
+        for ai, i in enumerate([i for i in range(len(obj)) if i not in hidden_labels]):
+            if len(levels_) == 1:
+                idx = (i, levels_[0]) if axis == 0 else (levels_[0], i)
+                display_funcs_[idx] = partial(alias_, value=labels[ai])
+            else:
+                for aj, lvl in enumerate(levels_):
+                    idx = (i, lvl) if axis == 0 else (lvl, i)
+                    display_funcs_[idx] = partial(alias_, value=labels[ai][aj])
+
+        return self
+
+    def format_index_names(
+        self,
+        formatter: ExtFormatter | None = None,
+        axis: Axis = 0,
+        level: Level | list[Level] | None = None,
+        na_rep: str | None = None,
+        precision: int | None = None,
+        decimal: str = ".",
+        thousands: str | None = None,
+        escape: str | None = None,
+        hyperlinks: str | None = None,
+    ) -> StylerRenderer:
+        r"""
+        Format the text display value of index names or column names.
+
+        .. versionadded:: 3.0
+
+        Parameters
+        ----------
+        formatter : str, callable, dict or None
+            Object to define how values are displayed. See notes.
+        axis : {0, "index", 1, "columns"}
+            Whether to apply the formatter to the index or column headers.
+        level : int, str, list
+            The level(s) over which to apply the generic formatter.
+        na_rep : str, optional
+            Representation for missing values.
+            If ``na_rep`` is None, no special formatting is applied.
+        precision : int, optional
+            Floating point precision to use for display purposes, if not determined by
+            the specified ``formatter``.
+        decimal : str, default "."
+            Character used as decimal separator for floats, complex and integers.
+        thousands : str, optional, default None
+            Character used as thousands separator for floats, complex and integers.
+        escape : str, optional
+            Use 'html' to replace the characters ``&``, ``<``, ``>``, ``'``, and ``"``
+            in cell display string with HTML-safe sequences.
+            Use 'latex' to replace the characters ``&``, ``%``, ``$``, ``#``, ``_``,
+            ``{``, ``}``, ``~``, ``^``, and ``\`` in the cell display string with
+            LaTeX-safe sequences.
+            Escaping is done before ``formatter``.
+        hyperlinks : {"html", "latex"}, optional
+            Convert string patterns containing https://, http://, ftp:// or www. to
+            HTML <a> tags as clickable URL hyperlinks if "html", or LaTeX \href
+            commands if "latex".
+
+        Returns
+        -------
+        Styler
+            Returns itself for chaining.
+
+        Raises
+        ------
+        ValueError
+            If the `formatter` is a string and the dtypes are incompatible.
+
+        See Also
+        --------
+        Styler.format_index: Format the text display value of index labels
+            or column headers.
+
+        Notes
+        -----
+        This method has a similar signature to :meth:`Styler.format_index`. Since
+        `names` are generally label based, and often not numeric, the typical features
+        expected to be more frequently used here are ``escape`` and ``hyperlinks``.
+
+        .. warning::
+            `Styler.format_index_names` is ignored when using the output format
+            `Styler.to_excel`, since Excel and Python have inherently different
+            formatting structures.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     [[1, 2], [3, 4]],
+        ...     index=pd.Index(["a", "b"], name="idx"),
+        ... )
+        >>> df  # doctest: +SKIP
+             0  1
+        idx
+        a    1  2
+        b    3  4
+        >>> df.style.format_index_names(lambda x: x.upper(), axis=0)  # doctest: +SKIP
+             0  1
+        IDX
+        a    1  2
+        b    3  4
+        """
+        axis = self.data._get_axis_number(axis)
+        if axis == 0:
+            display_funcs_, obj = self._display_funcs_index_names, self.index
+        else:
+            display_funcs_, obj = self._display_funcs_column_names, self.columns
+        levels_ = refactor_levels(level, obj)
+
+        if all(
+            (
+                formatter is None,
+                level is None,
+                precision is None,
+                decimal == ".",
+                thousands is None,
+                na_rep is None,
+                escape is None,
+                hyperlinks is None,
+            )
+        ):
+            display_funcs_.clear()
+            return self  # clear the formatter / revert to default and avoid looping
+
+        if not isinstance(formatter, dict):
+            formatter = {level: formatter for level in levels_}
+        else:
+            formatter = {
+                obj._get_level_number(level): formatter_
+                for level, formatter_ in formatter.items()
+            }
+
+        for lvl in levels_:
+            format_func = _maybe_wrap_formatter(
+                formatter.get(lvl),
+                na_rep=na_rep,
+                precision=precision,
+                decimal=decimal,
+                thousands=thousands,
+                escape=escape,
+                hyperlinks=hyperlinks,
+            )
+            display_funcs_[lvl] = format_func
+
+        return self
+
+
+def _element(
+    html_element: str,
+    html_class: str | None,
+    value: Any,
+    is_visible: bool,
+    **kwargs,
+) -> dict:
+    """
+    Template to return container with information for a <td></td> or <th></th> element.
+    """
+    if "display_value" not in kwargs or kwargs["display_value"] is None:
+        kwargs["display_value"] = value
+    return {
+        "type": html_element,
+        "value": value,
+        "class": html_class,
+        "is_visible": is_visible,
+        **kwargs,
+    }
+
+
+def _get_trimming_maximums(
+    rn,
+    cn,
+    max_elements,
+    max_rows=None,
+    max_cols=None,
+    scaling_factor: float = 0.8,
+) -> tuple[int, int]:
+    """
+    Recursively reduce the number of rows and columns to satisfy max elements.
+
+    Parameters
+    ----------
+    rn, cn : int
+        The number of input rows / columns
+    max_elements : int
+        The number of allowable elements
+    max_rows, max_cols : int, optional
+        Directly specify an initial maximum rows or columns before compression.
+    scaling_factor : float
+        Factor at which to reduce the number of rows / columns to fit.
+
+    Returns
+    -------
+    rn, cn : tuple
+        New rn and cn values that satisfy the max_elements constraint
+    """
+
+    def scale_down(rn, cn):
+        if cn >= rn:
+            return rn, int(cn * scaling_factor)
+        else:
+            return int(rn * scaling_factor), cn
+
+    if max_rows:
+        rn = max_rows if rn > max_rows else rn
+    if max_cols:
+        cn = max_cols if cn > max_cols else cn
+
+    while rn * cn > max_elements:
+        rn, cn = scale_down(rn, cn)
+
+    return rn, cn
+
+
+def _get_level_lengths(
+    index: Index,
+    sparsify: bool,
+    max_index: int,
+    hidden_elements: Sequence[int] | None = None,
+):
+    """
+    Given an index, find the level length for each element.
+
+    Parameters
+    ----------
+    index : Index
+        Index or columns to determine lengths of each element
+    sparsify : bool
+        Whether to hide or show each distinct element in a MultiIndex
+    max_index : int
+        The maximum number of elements to analyse along the index due to trimming
+    hidden_elements : sequence of int
+        Index positions of elements hidden from display in the index affecting
+        length
+
+    Returns
+    -------
+    Dict :
+        Result is a dictionary of (level, initial_position): span
+    """
+    if isinstance(index, MultiIndex):
+        levels = index._format_multi(sparsify=lib.no_default, include_names=False)
+    else:
+        levels = index._format_flat(include_name=False)
+
+    if hidden_elements is None:
+        hidden_elements = []
+
+    lengths = {}
+    if not isinstance(index, MultiIndex):
+        for i, value in enumerate(levels):
+            if i not in hidden_elements:
+                lengths[(0, i)] = 1
+        return lengths
+
+    for i, lvl in enumerate(levels):
+        visible_row_count = 0  # used to break loop due to display trimming
+        for j, row in enumerate(lvl):
+            if visible_row_count > max_index:
+                break
+            if not sparsify:
+                # then lengths will always equal 1 since no aggregation.
+                if j not in hidden_elements:
+                    lengths[(i, j)] = 1
+                    visible_row_count += 1
+            elif (row is not lib.no_default) and (j not in hidden_elements):
+                # this element has not been sparsified so must be the start of section
+                last_label = j
+                lengths[(i, last_label)] = 1
+                visible_row_count += 1
+            elif row is not lib.no_default:
+                # even if the above is hidden, keep track of it in case length > 1 and
+                # later elements are visible
+                last_label = j
+                lengths[(i, last_label)] = 0
+            elif j not in hidden_elements:
+                # then element must be part of sparsified section and is visible
+                visible_row_count += 1
+                if visible_row_count > max_index:
+                    break  # do not add a length since the render trim limit reached
+                if lengths[(i, last_label)] == 0:
+                    # if previous iteration was first-of-section but hidden then offset
+                    last_label = j
+                    lengths[(i, last_label)] = 1
+                else:
+                    # else add to previous iteration
+                    lengths[(i, last_label)] += 1
+
+    non_zero_lengths = {
+        element: length for element, length in lengths.items() if length >= 1
+    }
+
+    return non_zero_lengths
+
+
+def _is_visible(idx_row, idx_col, lengths) -> bool:
+    """
+    Index -> {(idx_row, idx_col): bool}).
+    """
+    return (idx_col, idx_row) in lengths
+
+
+def format_table_styles(styles: CSSStyles) -> CSSStyles:
+    """
+    looks for multiple CSS selectors and separates them:
+    [{'selector': 'td, th', 'props': 'a:v;'}]
+        ---> [{'selector': 'td', 'props': 'a:v;'},
+              {'selector': 'th', 'props': 'a:v;'}]
+    """
+    return [
+        {"selector": selector, "props": css_dict["props"]}
+        for css_dict in styles
+        for selector in css_dict["selector"].split(",")
+    ]
+
+
+def _default_formatter(x: Any, precision: int, thousands: bool = False) -> Any:
+    """
+    Format the display of a value
+
+    Parameters
+    ----------
+    x : Any
+        Input variable to be formatted
+    precision : Int
+        Floating point precision used if ``x`` is float or complex.
+    thousands : bool, default False
+        Whether to group digits with thousands separated with ",".
+
+    Returns
+    -------
+    value : Any
+        Matches input type, or string if input is float or complex or int with sep.
+    """
+    if is_float(x) or is_complex(x):
+        return f"{x:,.{precision}f}" if thousands else f"{x:.{precision}f}"
+    elif is_integer(x):
+        return f"{x:,}" if thousands else str(x)
+    return x
+
+
+def _wrap_decimal_thousands(
+    formatter: Callable, decimal: str, thousands: str | None
+) -> Callable:
+    """
+    Takes a string formatting function and wraps logic to deal with thousands and
+    decimal parameters, in the case that they are non-standard and that the input
+    is a (float, complex, int).
+    """
+
+    def wrapper(x):
+        if is_float(x) or is_integer(x) or is_complex(x):
+            if decimal != "." and thousands is not None and thousands != ",":
+                return (
+                    formatter(x)
+                    .replace(",", "§_§-")  # rare string to avoid "," <-> "." clash.
+                    .replace(".", decimal)
+                    .replace("§_§-", thousands)
+                )
+            elif decimal != "." and (thousands is None or thousands == ","):
+                return formatter(x).replace(".", decimal)
+            elif decimal == "." and thousands is not None and thousands != ",":
+                return formatter(x).replace(",", thousands)
+        return formatter(x)
+
+    return wrapper
+
+
+def _str_escape(x, escape):
+    """if escaping: only use on str, else return input"""
+    if isinstance(x, str):
+        if escape == "html":
+            return escape_html(x)
+        elif escape == "latex":
+            return _escape_latex(x)
+        elif escape == "latex-math":
+            return _escape_latex_math(x)
+        else:
+            raise ValueError(
+                f"`escape` only permitted in {{'html', 'latex', 'latex-math'}}, \
+got {escape}"
+            )
+    return x
+
+
+def _render_href(x, format):
+    """uses regex to detect a common URL pattern and converts to href tag in format."""
+    if isinstance(x, str):
+        if format == "html":
+            href = '<a href="{0}" target="_blank">{0}</a>'
+        elif format == "latex":
+            href = r"\href{{{0}}}{{{0}}}"
+        else:
+            raise ValueError("``hyperlinks`` format can only be 'html' or 'latex'")
+        pat = r"((http|ftp)s?:\/\/|www.)[\w/\-?=%.:@]+\.[\w/\-&?=%.,':;~!@#$*()\[\]]+"
+        return re.sub(pat, lambda m: href.format(m.group(0)), x)
+    return x
+
+
+def _maybe_wrap_formatter(
+    formatter: BaseFormatter | None = None,
+    na_rep: str | None = None,
+    precision: int | None = None,
+    decimal: str = ".",
+    thousands: str | None = None,
+    escape: str | None = None,
+    hyperlinks: str | None = None,
+) -> Callable:
+    """
+    Allows formatters to be expressed as str, callable or None, where None returns
+    a default formatting function. wraps with na_rep, and precision where they are
+    available.
+    """
+    # Get initial func from input string, input callable, or from default factory
+    if isinstance(formatter, str):
+        func_0 = lambda x: formatter.format(x)
+    elif callable(formatter):
+        func_0 = formatter
+    elif formatter is None:
+        precision = (
+            get_option("styler.format.precision") if precision is None else precision
+        )
+        func_0 = partial(
+            _default_formatter, precision=precision, thousands=(thousands is not None)
+        )
+    else:
+        raise TypeError(f"'formatter' expected str or callable, got {type(formatter)}")
+
+    # Replace chars if escaping
+    if escape is not None:
+        func_1 = lambda x: func_0(_str_escape(x, escape=escape))
+    else:
+        func_1 = func_0
+
+    # Replace decimals and thousands if non-standard inputs detected
+    if decimal != "." or (thousands is not None and thousands != ","):
+        func_2 = _wrap_decimal_thousands(func_1, decimal=decimal, thousands=thousands)
+    else:
+        func_2 = func_1
+
+    # Render links
+    if hyperlinks is not None:
+        func_3 = lambda x: func_2(_render_href(x, format=hyperlinks))
+    else:
+        func_3 = func_2
+
+    # Replace missing values if na_rep
+    if na_rep is None:
+        return func_3
+    else:
+        return lambda x: na_rep if (isna(x) is True) else func_3(x)
+
+
+def non_reducing_slice(slice_: Subset):
+    """
+    Ensure that a slice doesn't reduce to a Series or Scalar.
+
+    Any user-passed `subset` should have this called on it
+    to make sure we're always working with DataFrames.
+    """
+    # default to column slice, like DataFrame
+    # ['A', 'B'] -> IndexSlices[:, ['A', 'B']]
+    kinds = (ABCSeries, np.ndarray, Index, list, str)
+    if isinstance(slice_, kinds):
+        slice_ = IndexSlice[:, slice_]
+
+    def pred(part) -> bool:
+        """
+        Returns
+        -------
+        bool
+            True if slice does *not* reduce,
+            False if `part` is a tuple.
+        """
+        # true when slice does *not* reduce, False when part is a tuple,
+        # i.e. MultiIndex slice
+        if isinstance(part, tuple):
+            # GH#39421 check for sub-slice:
+            return any((isinstance(s, slice) or is_list_like(s)) for s in part)
+        else:
+            return isinstance(part, slice) or is_list_like(part)
+
+    if not is_list_like(slice_):
+        if not isinstance(slice_, slice):
+            # a 1-d slice, like df.loc[1]
+            slice_ = [[slice_]]
+        else:
+            # slice(a, b, c)
+            slice_ = [slice_]  # to tuplize later
+    else:
+        # error: Item "slice" of "Union[slice, Sequence[Any]]" has no attribute
+        # "__iter__" (not iterable) -> is specifically list_like in conditional
+        slice_ = [p if pred(p) else [p] for p in slice_]  # type: ignore[union-attr]
+    return tuple(slice_)
+
+
+def maybe_convert_css_to_tuples(style: CSSProperties) -> CSSList:
+    """
+    Convert css-string to sequence of tuples format if needed.
+    'color:red; border:1px solid black;' -> [('color', 'red'),
+                                             ('border','1px solid red')]
+    """
+    if isinstance(style, str):
+        if style and ":" not in style:
+            raise ValueError(
+                "Styles supplied as string must follow CSS rule formats, "
+                f"for example 'attr: val;'. '{style}' was given."
+            )
+        s = style.split(";")
+        return [
+            (x.split(":")[0].strip(), ":".join(x.split(":")[1:]).strip())
+            for x in s
+            if x.strip() != ""
+        ]
+
+    return style
+
+
+def refactor_levels(
+    level: Level | list[Level] | None,
+    obj: Index,
+) -> list[int]:
+    """
+    Returns a consistent levels arg for use in ``hide_index`` or ``hide_columns``.
+
+    Parameters
+    ----------
+    level : int, str, list
+        Original ``level`` arg supplied to above methods.
+    obj:
+        Either ``self.index`` or ``self.columns``
+
+    Returns
+    -------
+    list : refactored arg with a list of levels to hide
+    """
+    if level is None:
+        levels_: list[int] = list(range(obj.nlevels))
+    elif isinstance(level, int):
+        levels_ = [level]
+    elif isinstance(level, str):
+        levels_ = [obj._get_level_number(level)]
+    elif isinstance(level, list):
+        levels_ = [
+            obj._get_level_number(lev) if not isinstance(lev, int) else lev
+            for lev in level
+        ]
+    else:
+        raise ValueError("`level` must be of type `int`, `str` or list of such")
+    return levels_
+
+
+class Tooltips:
+    """
+    An extension to ``Styler`` that allows for and manipulates tooltips on hover
+    of ``<td>`` cells in the HTML result.
+
+    Parameters
+    ----------
+    css_name: str, default "pd-t"
+        Name of the CSS class that controls visualisation of tooltips.
+    css_props: list-like, default; see Notes
+        List of (attr, value) tuples defining properties of the CSS class.
+    tooltips: DataFrame, default empty
+        DataFrame of strings aligned with underlying Styler data for tooltip
+        display.
+    as_title_attribute: bool, default False
+        Flag to use title attribute based tooltips (True) or <span> based
+        tooltips (False).
+        Add the tooltip text as title attribute to resultant <td> element. If
+        True, no CSS is generated and styling effects do not apply.
+
+    Notes
+    -----
+    The default properties for the tooltip CSS class are:
+
+        - visibility: hidden
+        - position: absolute
+        - z-index: 1
+        - background-color: black
+        - color: white
+        - transform: translate(-20px, -20px)
+
+    Hidden visibility is a key prerequisite to the hover functionality, and should
+    always be included in any manual properties specification.
+    """
+
+    def __init__(
+        self,
+        css_props: CSSProperties = [  # noqa: B006
+            ("visibility", "hidden"),
+            ("position", "absolute"),
+            ("z-index", 1),
+            ("background-color", "black"),
+            ("color", "white"),
+            ("transform", "translate(-20px, -20px)"),
+        ],
+        css_name: str = "pd-t",
+        tooltips: DataFrame = DataFrame(),
+        as_title_attribute: bool = False,
+    ) -> None:
+        self.class_name = css_name
+        self.class_properties = css_props
+        self.tt_data = tooltips
+        self.table_styles: CSSStyles = []
+        self.as_title_attribute = as_title_attribute
+
+    @property
+    def _class_styles(self):
+        """
+        Combine the ``_Tooltips`` CSS class name and CSS properties to the format
+        required to extend the underlying ``Styler`` `table_styles` to allow
+        tooltips to render in HTML.
+
+        Returns
+        -------
+        styles : List
+        """
+        return [
+            {
+                "selector": f".{self.class_name}",
+                "props": maybe_convert_css_to_tuples(self.class_properties),
+            }
+        ]
+
+    def _pseudo_css(
+        self, uuid: str, name: str, row: int, col: int, text: str
+    ) -> list[CSSDict]:
+        """
+        For every table data-cell that has a valid tooltip (not None, NaN or
+        empty string) must create two pseudo CSS entries for the specific
+        <td> element id which are added to overall table styles:
+        an on hover visibility change and a content change
+        dependent upon the user's chosen display string.
+
+        For example:
+            [{"selector": "T__row1_col1:hover .pd-t",
+             "props": [("visibility", "visible")]},
+            {"selector": "T__row1_col1 .pd-t::after",
+             "props": [("content", "Some Valid Text String")]}]
+
+        Parameters
+        ----------
+        uuid: str
+            The uuid of the Styler instance
+        name: str
+            The css-name of the class used for styling tooltips
+        row : int
+            The row index of the specified tooltip string data
+        col : int
+            The col index of the specified tooltip string data
+        text : str
+            The textual content of the tooltip to be displayed in HTML.
+
+        Returns
+        -------
+        pseudo_css : List
+        """
+        selector_id = "#T_" + uuid + "_row" + str(row) + "_col" + str(col)
+        return [
+            {
+                "selector": selector_id + f":hover .{name}",
+                "props": [("visibility", "visible")],
+            },
+            {
+                "selector": selector_id + f" .{name}::after",
+                "props": [("content", f'"{text}"')],
+            },
+        ]
+
+    def _translate(self, styler: StylerRenderer, d: dict):
+        """
+        Mutate the render dictionary to allow for tooltips:
+
+        - Add ``<span>`` HTML element to each data cells ``display_value``. Ignores
+          headers.
+        - Add table level CSS styles to control pseudo classes.
+
+        Parameters
+        ----------
+        styler_data : DataFrame
+            Underlying ``Styler`` DataFrame used for reindexing.
+        uuid : str
+            The underlying ``Styler`` uuid for CSS id.
+        d : dict
+            The dictionary prior to final render
+
+        Returns
+        -------
+        render_dict : Dict
+        """
+        self.tt_data = self.tt_data.reindex_like(styler.data)
+        if self.tt_data.empty:
+            return d
+
+        mask = (self.tt_data.isna()) | (self.tt_data.eq(""))  # empty string = no ttip
+        # this conditional adds tooltips via pseudo css and <span> elements.
+        if not self.as_title_attribute:
+            name = self.class_name
+            self.table_styles = [
+                style
+                for sublist in [
+                    self._pseudo_css(
+                        styler.uuid, name, i, j, str(self.tt_data.iloc[i, j])
+                    )
+                    for i in range(len(self.tt_data.index))
+                    for j in range(len(self.tt_data.columns))
+                    if not (
+                        mask.iloc[i, j]
+                        or i in styler.hidden_rows
+                        or j in styler.hidden_columns
+                    )
+                ]
+                for style in sublist
+            ]
+
+            # add span class to every cell since there is at least 1 non-empty tooltip
+            if self.table_styles:
+                for row in d["body"]:
+                    for item in row:
+                        if item["type"] == "td":
+                            item["display_value"] = (
+                                str(item["display_value"])
+                                + f'<span class="{self.class_name}"></span>'
+                            )
+                d["table_styles"].extend(self._class_styles)
+                d["table_styles"].extend(self.table_styles)
+        # this conditional adds tooltips as extra "title" attribute on a <td> element
+        else:
+            index_offset = self.tt_data.index.nlevels
+            body = d["body"]
+            for i in range(len(self.tt_data.index)):
+                for j in range(len(self.tt_data.columns)):
+                    if (
+                        not mask.iloc[i, j]
+                        or i in styler.hidden_rows
+                        or j in styler.hidden_columns
+                    ):
+                        row = body[i]
+                        item = row[j + index_offset]
+                        value = self.tt_data.iloc[i, j]
+                        item["attributes"] += f' title="{value}"'
+        return d
+
+
+def _parse_latex_table_wrapping(table_styles: CSSStyles, caption: str | None) -> bool:
+    """
+    Indicate whether LaTeX {tabular} should be wrapped with a {table} environment.
+
+    Parses the `table_styles` and detects any selectors which must be included outside
+    of {tabular}, i.e. indicating that wrapping must occur, and therefore return True,
+    or if a caption exists and requires similar.
+    """
+    IGNORED_WRAPPERS = ["toprule", "midrule", "bottomrule", "column_format"]
+    # ignored selectors are included with {tabular} so do not need wrapping
+    return (
+        table_styles is not None
+        and any(d["selector"] not in IGNORED_WRAPPERS for d in table_styles)
+    ) or caption is not None
+
+
+def _parse_latex_table_styles(table_styles: CSSStyles, selector: str) -> str | None:
+    """
+    Return the first 'props' 'value' from ``tables_styles`` identified by ``selector``.
+
+    Examples
+    --------
+    >>> table_styles = [
+    ...     {"selector": "foo", "props": [("attr", "value")]},
+    ...     {"selector": "bar", "props": [("attr", "overwritten")]},
+    ...     {"selector": "bar", "props": [("a1", "baz"), ("a2", "ignore")]},
+    ... ]
+    >>> _parse_latex_table_styles(table_styles, selector="bar")
+    'baz'
+
+    Notes
+    -----
+    The replacement of "§" with ":" is to avoid the CSS problem where ":" has structural
+    significance and cannot be used in LaTeX labels, but is often required by them.
+    """
+    for style in table_styles[::-1]:  # in reverse for most recently applied style
+        if style["selector"] == selector:
+            return str(style["props"][0][1]).replace("§", ":")
+    return None
+
+
+def _parse_latex_cell_styles(
+    latex_styles: CSSList, display_value: str, convert_css: bool = False
+) -> str:
+    r"""
+    Mutate the ``display_value`` string including LaTeX commands from ``latex_styles``.
+
+    This method builds a recursive latex chain of commands based on the
+    CSSList input, nested around ``display_value``.
+
+    If a CSS style is given as ('<command>', '<options>') this is translated to
+    '\<command><options>{display_value}', and this value is treated as the
+    display value for the next iteration.
+
+    The most recent style forms the inner component, for example for styles:
+    `[('c1', 'o1'), ('c2', 'o2')]` this returns: `\c1o1{\c2o2{display_value}}`
+
+    Sometimes latex commands have to be wrapped with curly braces in different ways:
+    We create some parsing flags to identify the different behaviours:
+
+     - `--rwrap`        : `\<command><options>{<display_value>}`
+     - `--wrap`         : `{\<command><options> <display_value>}`
+     - `--nowrap`       : `\<command><options> <display_value>`
+     - `--lwrap`        : `{\<command><options>} <display_value>`
+     - `--dwrap`        : `{\<command><options>}{<display_value>}`
+
+    For example for styles:
+    `[('c1', 'o1--wrap'), ('c2', 'o2')]` this returns: `{\c1o1 \c2o2{display_value}}
+    """
+    if convert_css:
+        latex_styles = _parse_latex_css_conversion(latex_styles)
+    for command, options in latex_styles[::-1]:  # in reverse for most recent style
+        formatter = {
+            "--wrap": f"{{\\{command}--to_parse {display_value}}}",
+            "--nowrap": f"\\{command}--to_parse {display_value}",
+            "--lwrap": f"{{\\{command}--to_parse}} {display_value}",
+            "--rwrap": f"\\{command}--to_parse{{{display_value}}}",
+            "--dwrap": f"{{\\{command}--to_parse}}{{{display_value}}}",
+        }
+        display_value = f"\\{command}{options} {display_value}"
+        for arg in ["--nowrap", "--wrap", "--lwrap", "--rwrap", "--dwrap"]:
+            if arg in str(options):
+                display_value = formatter[arg].replace(
+                    "--to_parse", _parse_latex_options_strip(value=options, arg=arg)
+                )
+                break  # only ever one purposeful entry
+    return display_value
+
+
+def _parse_latex_header_span(
+    cell: dict[str, Any],
+    multirow_align: str,
+    multicol_align: str,
+    wrap: bool = False,
+    convert_css: bool = False,
+) -> str:
+    r"""
+    Refactor the cell `display_value` if a 'colspan' or 'rowspan' attribute is present.
+
+    'rowspan' and 'colspan' do not occur simultaneously. If they are detected then
+    the `display_value` is altered to a LaTeX `multirow` or `multicol` command
+    respectively, with the appropriate cell-span.
+
+    ``wrap`` is used to enclose the `display_value` in braces which is needed for
+    column headers using an siunitx package.
+
+    Requires the package {multirow}, whereas multicol support is usually built in
+    to the {tabular} environment.
+
+    Examples
+    --------
+    >>> cell = {"cellstyle": "", "display_value": "text", "attributes": 'colspan="3"'}
+    >>> _parse_latex_header_span(cell, "t", "c")
+    '\\multicolumn{3}{c}{text}'
+    """
+    display_val = _parse_latex_cell_styles(
+        cell["cellstyle"], cell["display_value"], convert_css
+    )
+    if "attributes" in cell:
+        attrs = cell["attributes"]
+        if 'colspan="' in attrs:
+            colspan = attrs[attrs.find('colspan="') + 9 :]  # len('colspan="') = 9
+            colspan = int(colspan[: colspan.find('"')])
+            if "naive-l" == multicol_align:
+                out = f"{{{display_val}}}" if wrap else f"{display_val}"
+                blanks = " & {}" if wrap else " &"
+                return out + blanks * (colspan - 1)
+            elif "naive-r" == multicol_align:
+                out = f"{{{display_val}}}" if wrap else f"{display_val}"
+                blanks = "{} & " if wrap else "& "
+                return blanks * (colspan - 1) + out
+            return f"\\multicolumn{{{colspan}}}{{{multicol_align}}}{{{display_val}}}"
+        elif 'rowspan="' in attrs:
+            if multirow_align == "naive":
+                return display_val
+            rowspan = attrs[attrs.find('rowspan="') + 9 :]
+            rowspan = int(rowspan[: rowspan.find('"')])
+            return f"\\multirow[{multirow_align}]{{{rowspan}}}{{*}}{{{display_val}}}"
+    if wrap:
+        return f"{{{display_val}}}"
+    else:
+        return display_val
+
+
+def _parse_latex_options_strip(value: str | float, arg: str) -> str:
+    """
+    Strip a css_value which may have latex wrapping arguments, css comment identifiers,
+    and whitespaces, to a valid string for latex options parsing.
+
+    For example: 'red /* --wrap */  ' --> 'red'
+    """
+    return str(value).replace(arg, "").replace("/*", "").replace("*/", "").strip()
+
+
+def _parse_latex_css_conversion(styles: CSSList) -> CSSList:
+    """
+    Convert CSS (attribute,value) pairs to equivalent LaTeX (command,options) pairs.
+
+    Ignore conversion if tagged with `--latex` option, skipped if no conversion found.
+    """
+
+    def font_weight(value, arg) -> tuple[str, str] | None:
+        if value in ("bold", "bolder"):
+            return "bfseries", f"{arg}"
+        return None
+
+    def font_style(value, arg) -> tuple[str, str] | None:
+        if value == "italic":
+            return "itshape", f"{arg}"
+        if value == "oblique":
+            return "slshape", f"{arg}"
+        return None
+
+    def color(value, user_arg, command, comm_arg):
+        """
+        CSS colors have 5 formats to process:
+
+         - 6 digit hex code: "#ff23ee"     --> [HTML]{FF23EE}
+         - 3 digit hex code: "#f0e"        --> [HTML]{FF00EE}
+         - rgba: rgba(128, 255, 0, 0.5)    --> [rgb]{0.502, 1.000, 0.000}
+         - rgb: rgb(128, 255, 0,)          --> [rbg]{0.502, 1.000, 0.000}
+         - string: red                     --> {red}
+
+        Additionally rgb or rgba can be expressed in % which is also parsed.
+        """
+        arg = user_arg if user_arg != "" else comm_arg
+
+        if value[0] == "#" and len(value) == 7:  # color is hex code
+            return command, f"[HTML]{{{value[1:].upper()}}}{arg}"
+        if value[0] == "#" and len(value) == 4:  # color is short hex code
+            val = f"{value[1].upper() * 2}{value[2].upper() * 2}{value[3].upper() * 2}"
+            return command, f"[HTML]{{{val}}}{arg}"
+        elif value[:3] == "rgb":  # color is rgb or rgba
+            r = re.findall("(?<=\\()[0-9\\s%]+(?=,)", value)[0].strip()
+            r = float(r[:-1]) / 100 if "%" in r else int(r) / 255
+            g = re.findall("(?<=,)[0-9\\s%]+(?=,)", value)[0].strip()
+            g = float(g[:-1]) / 100 if "%" in g else int(g) / 255
+            if value[3] == "a":  # color is rgba
+                b = re.findall("(?<=,)[0-9\\s%]+(?=,)", value)[1].strip()
+            else:  # color is rgb
+                b = re.findall("(?<=,)[0-9\\s%]+(?=\\))", value)[0].strip()
+            b = float(b[:-1]) / 100 if "%" in b else int(b) / 255
+            return command, f"[rgb]{{{r:.3f}, {g:.3f}, {b:.3f}}}{arg}"
+        else:
+            return command, f"{{{value}}}{arg}"  # color is likely string-named
+
+    CONVERTED_ATTRIBUTES: dict[str, Callable] = {
+        "font-weight": font_weight,
+        "background-color": partial(color, command="cellcolor", comm_arg="--lwrap"),
+        "color": partial(color, command="color", comm_arg=""),
+        "font-style": font_style,
+    }
+
+    latex_styles: CSSList = []
+    for attribute, value in styles:
+        if isinstance(value, str) and "--latex" in value:
+            # return the style without conversion but drop '--latex'
+            latex_styles.append((attribute, value.replace("--latex", "")))
+        if attribute in CONVERTED_ATTRIBUTES:
+            arg = ""
+            for x in ["--wrap", "--nowrap", "--lwrap", "--dwrap", "--rwrap"]:
+                if x in str(value):
+                    arg, value = x, _parse_latex_options_strip(value, x)
+                    break
+            latex_style = CONVERTED_ATTRIBUTES[attribute](value, arg)
+            if latex_style is not None:
+                latex_styles.extend([latex_style])
+    return latex_styles
+
+
+def _escape_latex(s: str) -> str:
+    r"""
+    Replace the characters ``&``, ``%``, ``$``, ``#``, ``_``, ``{``, ``}``,
+    ``~``, ``^``, and ``\`` in the string with LaTeX-safe sequences.
+
+    Use this if you need to display text that might contain such characters in LaTeX.
+
+    Parameters
+    ----------
+    s : str
+        Input to be escaped
+
+    Return
+    ------
+    str :
+        Escaped string
+    """
+    return (
+        s.replace("\\", "ab2§=§8yz")  # rare string for final conversion: avoid \\ clash
+        .replace("ab2§=§8yz ", "ab2§=§8yz\\space ")  # since \backslash gobbles spaces
+        .replace("&", "\\&")
+        .replace("%", "\\%")
+        .replace("$", "\\$")
+        .replace("#", "\\#")
+        .replace("_", "\\_")
+        .replace("{", "\\{")
+        .replace("}", "\\}")
+        .replace("~ ", "~\\space ")  # since \textasciitilde gobbles spaces
+        .replace("~", "\\textasciitilde ")
+        .replace("^ ", "^\\space ")  # since \textasciicircum gobbles spaces
+        .replace("^", "\\textasciicircum ")
+        .replace("ab2§=§8yz", "\\textbackslash ")
+    )
+
+
+def _math_mode_with_dollar(s: str) -> str:
+    r"""
+    All characters in LaTeX math mode are preserved.
+
+    The substrings in LaTeX math mode, which start with
+    the character ``$`` and end with ``$``, are preserved
+    without escaping. Otherwise regular LaTeX escaping applies.
+
+    Parameters
+    ----------
+    s : str
+        Input to be escaped
+
+    Return
+    ------
+    str :
+        Escaped string
+    """
+    s = s.replace(r"\$", r"rt8§=§7wz")
+    pattern = re.compile(r"\$.*?\$")
+    pos = 0
+    ps = pattern.search(s, pos)
+    res = []
+    while ps:
+        res.append(_escape_latex(s[pos : ps.span()[0]]))
+        res.append(ps.group())
+        pos = ps.span()[1]
+        ps = pattern.search(s, pos)
+
+    res.append(_escape_latex(s[pos : len(s)]))
+    return "".join(res).replace(r"rt8§=§7wz", r"\$")
+
+
+def _math_mode_with_parentheses(s: str) -> str:
+    r"""
+    All characters in LaTeX math mode are preserved.
+
+    The substrings in LaTeX math mode, which start with
+    the character ``\(`` and end with ``\)``, are preserved
+    without escaping. Otherwise regular LaTeX escaping applies.
+
+    Parameters
+    ----------
+    s : str
+        Input to be escaped
+
+    Return
+    ------
+    str :
+        Escaped string
+    """
+    s = s.replace(r"\(", r"LEFT§=§6yzLEFT").replace(r"\)", r"RIGHTab5§=§RIGHT")
+    res = []
+    for item in re.split(r"LEFT§=§6yz|ab5§=§RIGHT", s):
+        if item.startswith("LEFT") and item.endswith("RIGHT"):
+            res.append(item.replace("LEFT", r"\(").replace("RIGHT", r"\)"))
+        elif "LEFT" in item and "RIGHT" in item:
+            res.append(
+                _escape_latex(item).replace("LEFT", r"\(").replace("RIGHT", r"\)")
+            )
+        else:
+            res.append(
+                _escape_latex(item)
+                .replace("LEFT", r"\textbackslash (")
+                .replace("RIGHT", r"\textbackslash )")
+            )
+    return "".join(res)
+
+
+def _escape_latex_math(s: str) -> str:
+    r"""
+    All characters in LaTeX math mode are preserved.
+
+    The substrings in LaTeX math mode, which either are surrounded
+    by two characters ``$`` or start with the character ``\(`` and end with ``\)``,
+    are preserved without escaping. Otherwise regular LaTeX escaping applies.
+
+    Parameters
+    ----------
+    s : str
+        Input to be escaped
+
+    Return
+    ------
+    str :
+        Escaped string
+    """
+    s = s.replace(r"\$", r"rt8§=§7wz")
+    ps_d = re.compile(r"\$.*?\$").search(s, 0)
+    ps_p = re.compile(r"\(.*?\)").search(s, 0)
+    mode = []
+    if ps_d:
+        mode.append(ps_d.span()[0])
+    if ps_p:
+        mode.append(ps_p.span()[0])
+    if len(mode) == 0:
+        return _escape_latex(s.replace(r"rt8§=§7wz", r"\$"))
+    if s[mode[0]] == r"$":
+        return _math_mode_with_dollar(s.replace(r"rt8§=§7wz", r"\$"))
+    if s[mode[0] - 1 : mode[0] + 1] == r"\(":
+        return _math_mode_with_parentheses(s.replace(r"rt8§=§7wz", r"\$"))
+    else:
+        return _escape_latex(s.replace(r"rt8§=§7wz", r"\$"))
diff --git a/pandas/pandas/io/formats/templates/html.tpl b/pandas/pandas/io/formats/templates/html.tpl
new file mode 100644
index 0000000000000..8c63be3ad788a
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/html.tpl
@@ -0,0 +1,16 @@
+{# Update the html_style/table_structure.html documentation too #}
+{% if doctype_html %}
+<!DOCTYPE html>
+<html>
+<head>
+<meta charset="{{encoding}}">
+{% if not exclude_styles %}{% include html_style_tpl %}{% endif %}
+</head>
+<body>
+{% include html_table_tpl %}
+</body>
+</html>
+{% elif not doctype_html %}
+{% if not exclude_styles %}{% include html_style_tpl %}{% endif %}
+{% include html_table_tpl %}
+{% endif %}
diff --git a/pandas/pandas/io/formats/templates/html_style.tpl b/pandas/pandas/io/formats/templates/html_style.tpl
new file mode 100644
index 0000000000000..5c3fcd97f51bb
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/html_style.tpl
@@ -0,0 +1,26 @@
+{%- block before_style -%}{%- endblock before_style -%}
+{% block style %}
+<style type="text/css">
+{% block table_styles %}
+{% for s in table_styles %}
+#T_{{uuid}} {{s.selector}} {
+{% for p,val in s.props %}
+  {{p}}: {{val}};
+{% endfor %}
+}
+{% endfor %}
+{% endblock table_styles %}
+{% block before_cellstyle %}{% endblock before_cellstyle %}
+{% block cellstyle %}
+{% for cs in [cellstyle, cellstyle_index, cellstyle_columns] %}
+{% for s in cs %}
+{% for selector in s.selectors %}{% if not loop.first %}, {% endif %}#T_{{uuid}}_{{selector}}{% endfor %} {
+{% for p,val in s.props %}
+  {{p}}: {{val}};
+{% endfor %}
+}
+{% endfor %}
+{% endfor %}
+{% endblock cellstyle %}
+</style>
+{% endblock style %}
diff --git a/pandas/pandas/io/formats/templates/html_table.tpl b/pandas/pandas/io/formats/templates/html_table.tpl
new file mode 100644
index 0000000000000..17118d2bb21cc
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/html_table.tpl
@@ -0,0 +1,63 @@
+{% block before_table %}{% endblock before_table %}
+{% block table %}
+{% if exclude_styles %}
+<table>
+{% else %}
+<table id="T_{{uuid}}"{% if table_attributes %} {{table_attributes}}{% endif %}>
+{% endif %}
+{% block caption %}
+{% if caption and caption is string %}
+  <caption>{{caption}}</caption>
+{% elif caption and caption is sequence %}
+  <caption>{{caption[0]}}</caption>
+{% endif %}
+{% endblock caption %}
+{% block thead %}
+  <thead>
+{% block before_head_rows %}{% endblock %}
+{% for r in head %}
+{% block head_tr scoped %}
+    <tr>
+{% if exclude_styles %}
+{% for c in r %}
+{% if c.is_visible != False %}
+      <{{c.type}} {{c.attributes}}>{{c.display_value}}</{{c.type}}>
+{% endif %}
+{% endfor %}
+{% else %}
+{% for c in r %}
+{% if c.is_visible != False %}
+      <{{c.type}} {%- if c.id is defined %} id="T_{{uuid}}_{{c.id}}" {%- endif %} class="{{c.class}}" {{c.attributes}}>{{c.display_value}}</{{c.type}}>
+{% endif %}
+{% endfor %}
+{% endif %}
+    </tr>
+{% endblock head_tr %}
+{% endfor %}
+{% block after_head_rows %}{% endblock %}
+  </thead>
+{% endblock thead %}
+{% block tbody %}
+  <tbody>
+{% block before_rows %}{% endblock before_rows %}
+{% for r in body %}
+{% block tr scoped %}
+    <tr>
+{% if exclude_styles %}
+{% for c in r %}{% if c.is_visible != False %}
+      <{{c.type}} {{c.attributes}}>{{c.display_value}}</{{c.type}}>
+{% endif %}{% endfor %}
+{% else %}
+{% for c in r %}{% if c.is_visible != False %}
+      <{{c.type}} {%- if c.id is defined %} id="T_{{uuid}}_{{c.id}}" {%- endif %} class="{{c.class}}" {{c.attributes}}>{{c.display_value}}</{{c.type}}>
+{% endif %}{% endfor %}
+{% endif %}
+    </tr>
+{% endblock tr %}
+{% endfor %}
+{% block after_rows %}{% endblock after_rows %}
+  </tbody>
+{% endblock tbody %}
+</table>
+{% endblock table %}
+{% block after_table %}{% endblock after_table %}
diff --git a/pandas/pandas/io/formats/templates/latex.tpl b/pandas/pandas/io/formats/templates/latex.tpl
new file mode 100644
index 0000000000000..ae341bbc29823
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/latex.tpl
@@ -0,0 +1,5 @@
+{% if environment == "longtable" %}
+{% include "latex_longtable.tpl" %}
+{% else %}
+{% include "latex_table.tpl" %}
+{% endif %}
diff --git a/pandas/pandas/io/formats/templates/latex_longtable.tpl b/pandas/pandas/io/formats/templates/latex_longtable.tpl
new file mode 100644
index 0000000000000..b97843eeb918d
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/latex_longtable.tpl
@@ -0,0 +1,82 @@
+\begin{longtable}
+{%- set position = parse_table(table_styles, 'position') %}
+{%- if position is not none %}
+[{{position}}]
+{%- endif %}
+{%- set column_format = parse_table(table_styles, 'column_format') %}
+{% raw %}{{% endraw %}{{column_format}}{% raw %}}{% endraw %}
+
+{% for style in table_styles %}
+{% if style['selector'] not in ['position', 'position_float', 'caption', 'toprule', 'midrule', 'bottomrule', 'column_format', 'label'] %}
+\{{style['selector']}}{{parse_table(table_styles, style['selector'])}}
+{% endif %}
+{% endfor %}
+{% if caption and caption is string %}
+\caption{% raw %}{{% endraw %}{{caption}}{% raw %}}{% endraw %}
+{%- set label = parse_table(table_styles, 'label') %}
+{%- if label is not none %}
+ \label{{label}}
+{%- endif %} \\
+{% elif caption and caption is sequence %}
+\caption[{{caption[1]}}]{% raw %}{{% endraw %}{{caption[0]}}{% raw %}}{% endraw %}
+{%- set label = parse_table(table_styles, 'label') %}
+{%- if label is not none %}
+ \label{{label}}
+{%- endif %} \\
+{% else %}
+{%- set label = parse_table(table_styles, 'label') %}
+{%- if label is not none %}
+\label{{label}} \\
+{% endif %}
+{% endif %}
+{% set toprule = parse_table(table_styles, 'toprule') %}
+{% if toprule is not none %}
+\{{toprule}}
+{% endif %}
+{% for row in head %}
+{% for c in row %}{%- if not loop.first %} & {% endif %}{{parse_header(c, multirow_align, multicol_align, siunitx)}}{% endfor %} \\
+{% endfor %}
+{% set midrule = parse_table(table_styles, 'midrule') %}
+{% if midrule is not none %}
+\{{midrule}}
+{% endif %}
+\endfirsthead
+{% if caption and caption is string %}
+\caption[]{% raw %}{{% endraw %}{{caption}}{% raw %}}{% endraw %} \\
+{% elif caption and caption is sequence %}
+\caption[]{% raw %}{{% endraw %}{{caption[0]}}{% raw %}}{% endraw %} \\
+{% endif %}
+{% if toprule is not none %}
+\{{toprule}}
+{% endif %}
+{% for row in head %}
+{% for c in row %}{%- if not loop.first %} & {% endif %}{{parse_header(c, multirow_align, multicol_align, siunitx)}}{% endfor %} \\
+{% endfor %}
+{% if midrule is not none %}
+\{{midrule}}
+{% endif %}
+\endhead
+{% if midrule is not none %}
+\{{midrule}}
+{% endif %}
+\multicolumn{% raw %}{{% endraw %}{{body[0]|length}}{% raw %}}{% endraw %}{r}{Continued on next page} \\
+{% if midrule is not none %}
+\{{midrule}}
+{% endif %}
+\endfoot
+{% set bottomrule = parse_table(table_styles, 'bottomrule') %}
+{% if bottomrule is not none %}
+\{{bottomrule}}
+{% endif %}
+\endlastfoot
+{% for row in body %}
+{% for c in row %}{% if not loop.first %} & {% endif %}
+  {%- if c.type == 'th' %}{{parse_header(c, multirow_align, multicol_align)}}{% else %}{{parse_cell(c.cellstyle, c.display_value, convert_css)}}{% endif %}
+{%- endfor %} \\
+{% if clines and clines[loop.index] | length > 0 %}
+  {%- for cline in clines[loop.index] %}{% if not loop.first %} {% endif %}{{ cline }}{% endfor %}
+
+{% endif %}
+{% endfor %}
+\end{longtable}
+{% raw %}{% endraw %}
diff --git a/pandas/pandas/io/formats/templates/latex_table.tpl b/pandas/pandas/io/formats/templates/latex_table.tpl
new file mode 100644
index 0000000000000..7858cb4c94553
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/latex_table.tpl
@@ -0,0 +1,57 @@
+{% if environment or parse_wrap(table_styles, caption) %}
+\begin{% raw %}{{% endraw %}{{environment if environment else "table"}}{% raw %}}{% endraw %}
+{%- set position = parse_table(table_styles, 'position') %}
+{%- if position is not none %}
+[{{position}}]
+{%- endif %}
+
+{% set position_float = parse_table(table_styles, 'position_float') %}
+{% if position_float is not none%}
+\{{position_float}}
+{% endif %}
+{% if caption and caption is string %}
+\caption{% raw %}{{% endraw %}{{caption}}{% raw %}}{% endraw %}
+
+{% elif caption and caption is sequence %}
+\caption[{{caption[1]}}]{% raw %}{{% endraw %}{{caption[0]}}{% raw %}}{% endraw %}
+
+{% endif %}
+{% for style in table_styles %}
+{% if style['selector'] not in ['position', 'position_float', 'caption', 'toprule', 'midrule', 'bottomrule', 'column_format'] %}
+\{{style['selector']}}{{parse_table(table_styles, style['selector'])}}
+{% endif %}
+{% endfor %}
+{% endif %}
+\begin{tabular}
+{%- set column_format = parse_table(table_styles, 'column_format') %}
+{% raw %}{{% endraw %}{{column_format}}{% raw %}}{% endraw %}
+
+{% set toprule = parse_table(table_styles, 'toprule') %}
+{% if toprule is not none %}
+\{{toprule}}
+{% endif %}
+{% for row in head %}
+{% for c in row %}{%- if not loop.first %} & {% endif %}{{parse_header(c, multirow_align, multicol_align, siunitx, convert_css)}}{% endfor %} \\
+{% endfor %}
+{% set midrule = parse_table(table_styles, 'midrule') %}
+{% if midrule is not none %}
+\{{midrule}}
+{% endif %}
+{% for row in body %}
+{% for c in row %}{% if not loop.first %} & {% endif %}
+  {%- if c.type == 'th' %}{{parse_header(c, multirow_align, multicol_align, False, convert_css)}}{% else %}{{parse_cell(c.cellstyle, c.display_value, convert_css)}}{% endif %}
+{%- endfor %} \\
+{% if clines and clines[loop.index] | length > 0 %}
+  {%- for cline in clines[loop.index] %}{% if not loop.first %} {% endif %}{{ cline }}{% endfor %}
+
+{% endif %}
+{% endfor %}
+{% set bottomrule = parse_table(table_styles, 'bottomrule') %}
+{% if bottomrule is not none %}
+\{{bottomrule}}
+{% endif %}
+\end{tabular}
+{% if environment or parse_wrap(table_styles, caption) %}
+\end{% raw %}{{% endraw %}{{environment if environment else "table"}}{% raw %}}{% endraw %}
+
+{% endif %}
diff --git a/pandas/pandas/io/formats/templates/string.tpl b/pandas/pandas/io/formats/templates/string.tpl
new file mode 100644
index 0000000000000..06aeb2b4e413c
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/string.tpl
@@ -0,0 +1,12 @@
+{% for r in head %}
+{% for c in r %}{% if c["is_visible"] %}
+{{ c["display_value"] }}{% if not loop.last %}{{ delimiter }}{% endif %}
+{% endif %}{% endfor %}
+
+{% endfor %}
+{% for r in body %}
+{% for c in r %}{% if c["is_visible"] %}
+{{ c["display_value"] }}{% if not loop.last %}{{ delimiter }}{% endif %}
+{% endif %}{% endfor %}
+
+{% endfor %}
diff --git a/pandas/pandas/io/formats/templates/typst.tpl b/pandas/pandas/io/formats/templates/typst.tpl
new file mode 100644
index 0000000000000..66de8f31b405e
--- /dev/null
+++ b/pandas/pandas/io/formats/templates/typst.tpl
@@ -0,0 +1,12 @@
+#table(
+  columns: {{ head[0] | length }},
+{% for r in head %}
+  {% for c in r %}[{% if c["is_visible"] %}{{ c["display_value"] }}{% endif %}],{% if not loop.last %} {% endif%}{% endfor %}
+
+{% endfor %}
+
+{% for r in body %}
+  {% for c in r %}[{% if c["is_visible"] %}{{ c["display_value"] }}{% endif %}],{% if not loop.last %} {% endif%}{% endfor %}
+
+{% endfor %}
+)
diff --git a/pandas/pandas/io/formats/xml.py b/pandas/pandas/io/formats/xml.py
new file mode 100644
index 0000000000000..febf43b9a1018
--- /dev/null
+++ b/pandas/pandas/io/formats/xml.py
@@ -0,0 +1,555 @@
+"""
+:mod:`pandas.io.formats.xml` is a module for formatting data in XML.
+"""
+
+from __future__ import annotations
+
+import codecs
+import io
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    final,
+)
+
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import (
+    cache_readonly,
+    doc,
+)
+
+from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.missing import isna
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import get_handle
+from pandas.io.xml import get_data_from_filepath
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        ReadBuffer,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+    from pandas import DataFrame
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    compression_options=_shared_docs["compression_options"] % "path_or_buffer",
+)
+class _BaseXMLFormatter:
+    """
+    Subclass for formatting data in XML.
+
+    Parameters
+    ----------
+    path_or_buffer : str or file-like
+        This can be either a string of raw XML, a valid URL,
+        file or file-like object.
+
+    index : bool
+        Whether to include index in xml document.
+
+    row_name : str
+        Name for root of xml document. Default is 'data'.
+
+    root_name : str
+        Name for row elements of xml document. Default is 'row'.
+
+    na_rep : str
+        Missing data representation.
+
+    attrs_cols : list
+        List of columns to write as attributes in row element.
+
+    elem_cols : list
+        List of columns to write as children in row element.
+
+    namespaces : dict
+        The namespaces to define in XML document as dicts with key
+        being namespace and value the URI.
+
+    prefix : str
+        The prefix for each element in XML document including root.
+
+    encoding : str
+        Encoding of xml object or document.
+
+    xml_declaration : bool
+        Whether to include xml declaration at top line item in xml.
+
+    pretty_print : bool
+        Whether to write xml document with line breaks and indentation.
+
+    stylesheet : str or file-like
+        A URL, file, file-like object, or a raw string containing XSLT.
+
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    {storage_options}
+
+    See also
+    --------
+    pandas.io.formats.xml.EtreeXMLFormatter
+    pandas.io.formats.xml.LxmlXMLFormatter
+
+    """
+
+    def __init__(
+        self,
+        frame: DataFrame,
+        path_or_buffer: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = None,
+        index: bool = True,
+        root_name: str | None = "data",
+        row_name: str | None = "row",
+        na_rep: str | None = None,
+        attr_cols: list[str] | None = None,
+        elem_cols: list[str] | None = None,
+        namespaces: dict[str | None, str] | None = None,
+        prefix: str | None = None,
+        encoding: str = "utf-8",
+        xml_declaration: bool | None = True,
+        pretty_print: bool | None = True,
+        stylesheet: FilePath | ReadBuffer[str] | ReadBuffer[bytes] | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+    ) -> None:
+        self.frame = frame
+        self.path_or_buffer = path_or_buffer
+        self.index = index
+        self.root_name = root_name
+        self.row_name = row_name
+        self.na_rep = na_rep
+        self.attr_cols = attr_cols
+        self.elem_cols = elem_cols
+        self.namespaces = namespaces
+        self.prefix = prefix
+        self.encoding = encoding
+        self.xml_declaration = xml_declaration
+        self.pretty_print = pretty_print
+        self.stylesheet = stylesheet
+        self.compression: CompressionOptions = compression
+        self.storage_options = storage_options
+
+        self.orig_cols = self.frame.columns.tolist()
+        self.frame_dicts = self._process_dataframe()
+
+        self._validate_columns()
+        self._validate_encoding()
+        self.prefix_uri = self._get_prefix_uri()
+        self._handle_indexes()
+
+    def _build_tree(self) -> bytes:
+        """
+        Build tree from  data.
+
+        This method initializes the root and builds attributes and elements
+        with optional namespaces.
+        """
+        raise AbstractMethodError(self)
+
+    @final
+    def _validate_columns(self) -> None:
+        """
+        Validate elems_cols and attrs_cols.
+
+        This method will check if columns is list-like.
+
+        Raises
+        ------
+        ValueError
+            * If value is not a list and less then length of nodes.
+        """
+        if self.attr_cols and not is_list_like(self.attr_cols):
+            raise TypeError(
+                f"{type(self.attr_cols).__name__} is not a valid type for attr_cols"
+            )
+
+        if self.elem_cols and not is_list_like(self.elem_cols):
+            raise TypeError(
+                f"{type(self.elem_cols).__name__} is not a valid type for elem_cols"
+            )
+
+    @final
+    def _validate_encoding(self) -> None:
+        """
+        Validate encoding.
+
+        This method will check if encoding is among listed under codecs.
+
+        Raises
+        ------
+        LookupError
+            * If encoding is not available in codecs.
+        """
+
+        codecs.lookup(self.encoding)
+
+    @final
+    def _process_dataframe(self) -> dict[int | str, dict[str, Any]]:
+        """
+        Adjust Data Frame to fit xml output.
+
+        This method will adjust underlying data frame for xml output,
+        including optionally replacing missing values and including indexes.
+        """
+
+        df = self.frame
+
+        if self.index:
+            df = df.reset_index()
+
+        if self.na_rep is not None:
+            df = df.fillna(self.na_rep)
+
+        return df.to_dict(orient="index")
+
+    @final
+    def _handle_indexes(self) -> None:
+        """
+        Handle indexes.
+
+        This method will add indexes into attr_cols or elem_cols.
+        """
+
+        if not self.index:
+            return
+
+        first_key = next(iter(self.frame_dicts))
+        indexes: list[str] = [
+            x for x in self.frame_dicts[first_key].keys() if x not in self.orig_cols
+        ]
+
+        if self.attr_cols:
+            self.attr_cols = indexes + self.attr_cols
+
+        if self.elem_cols:
+            self.elem_cols = indexes + self.elem_cols
+
+    def _get_prefix_uri(self) -> str:
+        """
+        Get uri of namespace prefix.
+
+        This method retrieves corresponding URI to prefix in namespaces.
+
+        Raises
+        ------
+        KeyError
+            *If prefix is not included in namespace dict.
+        """
+
+        raise AbstractMethodError(self)
+
+    @final
+    def _other_namespaces(self) -> dict:
+        """
+        Define other namespaces.
+
+        This method will build dictionary of namespaces attributes
+        for root element, conditionally with optional namespaces and
+        prefix.
+        """
+
+        nmsp_dict: dict[str, str] = {}
+        if self.namespaces:
+            nmsp_dict = {
+                f"xmlns{p if p == '' else f':{p}'}": n
+                for p, n in self.namespaces.items()
+                if n != self.prefix_uri[1:-1]
+            }
+
+        return nmsp_dict
+
+    @final
+    def _build_attribs(self, d: dict[str, Any], elem_row: Any) -> Any:
+        """
+        Create attributes of row.
+
+        This method adds attributes using attr_cols to row element and
+        works with tuples for multindex or hierarchical columns.
+        """
+
+        if not self.attr_cols:
+            return elem_row
+
+        for col in self.attr_cols:
+            attr_name = self._get_flat_col_name(col)
+            try:
+                if not isna(d[col]):
+                    elem_row.attrib[attr_name] = str(d[col])
+            except KeyError as err:
+                raise KeyError(f"no valid column, {col}") from err
+        return elem_row
+
+    @final
+    def _get_flat_col_name(self, col: str | tuple) -> str:
+        flat_col = col
+        if isinstance(col, tuple):
+            flat_col = (
+                "".join([str(c) for c in col]).strip()
+                if "" in col
+                else "_".join([str(c) for c in col]).strip()
+            )
+        return f"{self.prefix_uri}{flat_col}"
+
+    @cache_readonly
+    def _sub_element_cls(self):
+        raise AbstractMethodError(self)
+
+    @final
+    def _build_elems(self, d: dict[str, Any], elem_row: Any) -> None:
+        """
+        Create child elements of row.
+
+        This method adds child elements using elem_cols to row element and
+        works with tuples for multindex or hierarchical columns.
+        """
+        sub_element_cls = self._sub_element_cls
+
+        if not self.elem_cols:
+            return
+
+        for col in self.elem_cols:
+            elem_name = self._get_flat_col_name(col)
+            try:
+                val = None if isna(d[col]) or d[col] == "" else str(d[col])
+                sub_element_cls(elem_row, elem_name).text = val
+            except KeyError as err:
+                raise KeyError(f"no valid column, {col}") from err
+
+    @final
+    def write_output(self) -> str | None:
+        xml_doc = self._build_tree()
+
+        if self.path_or_buffer is not None:
+            with get_handle(
+                self.path_or_buffer,
+                "wb",
+                compression=self.compression,
+                storage_options=self.storage_options,
+                is_text=False,
+            ) as handles:
+                handles.handle.write(xml_doc)
+            return None
+
+        else:
+            return xml_doc.decode(self.encoding).rstrip()
+
+
+class EtreeXMLFormatter(_BaseXMLFormatter):
+    """
+    Class for formatting data in xml using Python standard library
+    modules: `xml.etree.ElementTree` and `xml.dom.minidom`.
+    """
+
+    def _build_tree(self) -> bytes:
+        from xml.etree.ElementTree import (
+            Element,
+            SubElement,
+            tostring,
+        )
+
+        self.root = Element(
+            f"{self.prefix_uri}{self.root_name}", attrib=self._other_namespaces()
+        )
+
+        for d in self.frame_dicts.values():
+            elem_row = SubElement(self.root, f"{self.prefix_uri}{self.row_name}")
+
+            if not self.attr_cols and not self.elem_cols:
+                self.elem_cols = list(d.keys())
+                self._build_elems(d, elem_row)
+
+            else:
+                elem_row = self._build_attribs(d, elem_row)
+                self._build_elems(d, elem_row)
+
+        self.out_xml = tostring(
+            self.root,
+            method="xml",
+            encoding=self.encoding,
+            xml_declaration=self.xml_declaration,
+        )
+
+        if self.pretty_print:
+            self.out_xml = self._prettify_tree()
+
+        if self.stylesheet is not None:
+            raise ValueError(
+                "To use stylesheet, you need lxml installed and selected as parser."
+            )
+
+        return self.out_xml
+
+    def _get_prefix_uri(self) -> str:
+        from xml.etree.ElementTree import register_namespace
+
+        uri = ""
+        if self.namespaces:
+            for p, n in self.namespaces.items():
+                if isinstance(p, str) and isinstance(n, str):
+                    register_namespace(p, n)
+            if self.prefix:
+                try:
+                    uri = f"{{{self.namespaces[self.prefix]}}}"
+                except KeyError as err:
+                    raise KeyError(
+                        f"{self.prefix} is not included in namespaces"
+                    ) from err
+            elif "" in self.namespaces:
+                uri = f"{{{self.namespaces['']}}}"
+            else:
+                uri = ""
+
+        return uri
+
+    @cache_readonly
+    def _sub_element_cls(self):
+        from xml.etree.ElementTree import SubElement
+
+        return SubElement
+
+    def _prettify_tree(self) -> bytes:
+        """
+        Output tree for pretty print format.
+
+        This method will pretty print xml with line breaks and indentation.
+        """
+
+        from xml.dom.minidom import parseString
+
+        dom = parseString(self.out_xml)
+
+        return dom.toprettyxml(indent="  ", encoding=self.encoding)
+
+
+class LxmlXMLFormatter(_BaseXMLFormatter):
+    """
+    Class for formatting data in xml using Python standard library
+    modules: `xml.etree.ElementTree` and `xml.dom.minidom`.
+    """
+
+    def __init__(self, *args, **kwargs) -> None:
+        super().__init__(*args, **kwargs)
+
+        self._convert_empty_str_key()
+
+    def _build_tree(self) -> bytes:
+        """
+        Build tree from  data.
+
+        This method initializes the root and builds attributes and elements
+        with optional namespaces.
+        """
+        from lxml.etree import (
+            Element,
+            SubElement,
+            tostring,
+        )
+
+        self.root = Element(f"{self.prefix_uri}{self.root_name}", nsmap=self.namespaces)
+
+        for d in self.frame_dicts.values():
+            elem_row = SubElement(self.root, f"{self.prefix_uri}{self.row_name}")
+
+            if not self.attr_cols and not self.elem_cols:
+                self.elem_cols = list(d.keys())
+                self._build_elems(d, elem_row)
+
+            else:
+                elem_row = self._build_attribs(d, elem_row)
+                self._build_elems(d, elem_row)
+
+        self.out_xml = tostring(
+            self.root,
+            pretty_print=self.pretty_print,
+            method="xml",
+            encoding=self.encoding,
+            xml_declaration=self.xml_declaration,
+        )
+
+        if self.stylesheet is not None:
+            self.out_xml = self._transform_doc()
+
+        return self.out_xml
+
+    def _convert_empty_str_key(self) -> None:
+        """
+        Replace zero-length string in `namespaces`.
+
+        This method will replace '' with None to align to `lxml`
+        requirement that empty string prefixes are not allowed.
+        """
+
+        if self.namespaces and "" in self.namespaces.keys():
+            self.namespaces[None] = self.namespaces.pop("", "default")
+
+    def _get_prefix_uri(self) -> str:
+        uri = ""
+        if self.namespaces:
+            if self.prefix:
+                try:
+                    uri = f"{{{self.namespaces[self.prefix]}}}"
+                except KeyError as err:
+                    raise KeyError(
+                        f"{self.prefix} is not included in namespaces"
+                    ) from err
+            elif "" in self.namespaces:
+                uri = f"{{{self.namespaces['']}}}"
+            else:
+                uri = ""
+
+        return uri
+
+    @cache_readonly
+    def _sub_element_cls(self):
+        from lxml.etree import SubElement
+
+        return SubElement
+
+    def _transform_doc(self) -> bytes:
+        """
+        Parse stylesheet from file or buffer and run it.
+
+        This method will parse stylesheet object into tree for parsing
+        conditionally by its specific object type, then transforms
+        original tree with XSLT script.
+        """
+        from lxml.etree import (
+            XSLT,
+            XMLParser,
+            fromstring,
+            parse,
+        )
+
+        style_doc = self.stylesheet
+        assert style_doc is not None  # is ensured by caller
+
+        handle_data = get_data_from_filepath(
+            filepath_or_buffer=style_doc,
+            encoding=self.encoding,
+            compression=self.compression,
+            storage_options=self.storage_options,
+        )
+
+        with handle_data as xml_data:
+            curr_parser = XMLParser(encoding=self.encoding)
+
+            if isinstance(xml_data, io.StringIO):
+                xsl_doc = fromstring(
+                    xml_data.getvalue().encode(self.encoding), parser=curr_parser
+                )
+            else:
+                xsl_doc = parse(xml_data, parser=curr_parser)
+
+        transformer = XSLT(xsl_doc)
+        new_doc = transformer(self.root)
+
+        return bytes(new_doc)
diff --git a/pandas/pandas/io/html.py b/pandas/pandas/io/html.py
new file mode 100644
index 0000000000000..183af3a03221b
--- /dev/null
+++ b/pandas/pandas/io/html.py
@@ -0,0 +1,1248 @@
+"""
+:mod:`pandas.io.html` is a module containing functionality for dealing with
+HTML IO.
+
+"""
+
+from __future__ import annotations
+
+from collections import abc
+import errno
+import numbers
+import os
+import re
+from re import Pattern
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    cast,
+)
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    AbstractMethodError,
+    EmptyDataError,
+)
+from pandas.util._decorators import doc
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import is_list_like
+
+from pandas import isna
+from pandas.core.indexes.base import Index
+from pandas.core.indexes.multi import MultiIndex
+from pandas.core.series import Series
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import (
+    get_handle,
+    is_url,
+    stringify_path,
+    validate_header_arg,
+)
+from pandas.io.formats.printing import pprint_thing
+from pandas.io.parsers import TextParser
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Iterable,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        BaseBuffer,
+        DtypeBackend,
+        FilePath,
+        HTMLFlavors,
+        ReadBuffer,
+        StorageOptions,
+    )
+
+    from pandas import DataFrame
+
+#############
+# READ HTML #
+#############
+_RE_WHITESPACE = re.compile(r"[\r\n]+|\s{2,}")
+
+
+def _remove_whitespace(s: str, regex: Pattern = _RE_WHITESPACE) -> str:
+    """
+    Replace extra whitespace inside of a string with a single space.
+
+    Parameters
+    ----------
+    s : str or unicode
+        The string from which to remove extra whitespace.
+    regex : re.Pattern
+        The regular expression to use to remove extra whitespace.
+
+    Returns
+    -------
+    subd : str or unicode
+        `s` with all extra whitespace replaced with a single space.
+    """
+    return regex.sub(" ", s.strip())
+
+
+def _get_skiprows(skiprows: int | Sequence[int] | slice | None) -> int | Sequence[int]:
+    """
+    Get an iterator given an integer, slice or container.
+
+    Parameters
+    ----------
+    skiprows : int, slice, container
+        The iterator to use to skip rows; can also be a slice.
+
+    Raises
+    ------
+    TypeError
+        * If `skiprows` is not a slice, integer, or Container
+
+    Returns
+    -------
+    it : iterable
+        A proper iterator to use to skip rows of a DataFrame.
+    """
+    if isinstance(skiprows, slice):
+        start, step = skiprows.start or 0, skiprows.step or 1
+        return list(range(start, skiprows.stop, step))
+    elif isinstance(skiprows, numbers.Integral) or is_list_like(skiprows):
+        return cast("int | Sequence[int]", skiprows)
+    elif skiprows is None:
+        return 0
+    raise TypeError(f"{type(skiprows).__name__} is not a valid type for skipping rows")
+
+
+def _read(
+    obj: FilePath | BaseBuffer,
+    encoding: str | None,
+    storage_options: StorageOptions | None,
+) -> str | bytes:
+    """
+    Try to read from a url, file or string.
+
+    Parameters
+    ----------
+    obj : str, unicode, path object, or file-like object
+
+    Returns
+    -------
+    raw_text : str
+    """
+    try:
+        with get_handle(
+            obj, "r", encoding=encoding, storage_options=storage_options
+        ) as handles:
+            return handles.handle.read()
+    except OSError as err:
+        if not is_url(obj):
+            raise FileNotFoundError(
+                f"[Errno {errno.ENOENT}] {os.strerror(errno.ENOENT)}: {obj}"
+            ) from err
+        raise
+
+
+class _HtmlFrameParser:
+    """
+    Base class for parsers that parse HTML into DataFrames.
+
+    Parameters
+    ----------
+    io : str or file-like
+        This can be either a string path, a valid URL using the HTTP,
+        FTP, or FILE protocols or a file-like object.
+
+    match : str or regex
+        The text to match in the document.
+
+    attrs : dict
+        List of HTML <table> element attributes to match.
+
+    encoding : str
+        Encoding to be used by parser
+
+    displayed_only : bool
+        Whether or not items with "display:none" should be ignored
+
+    extract_links : {None, "all", "header", "body", "footer"}
+        Table elements in the specified section(s) with <a> tags will have their
+        href extracted.
+
+        .. versionadded:: 1.5.0
+
+    Attributes
+    ----------
+    io : str or file-like
+        raw HTML, URL, or file-like object
+
+    match : regex
+        The text to match in the raw HTML
+
+    attrs : dict-like
+        A dictionary of valid table attributes to use to search for table
+        elements.
+
+    encoding : str
+        Encoding to be used by parser
+
+    displayed_only : bool
+        Whether or not items with "display:none" should be ignored
+
+    extract_links : {None, "all", "header", "body", "footer"}
+        Table elements in the specified section(s) with <a> tags will have their
+        href extracted.
+
+        .. versionadded:: 1.5.0
+
+    Notes
+    -----
+    To subclass this class effectively you must override the following methods:
+        * :func:`_build_doc`
+        * :func:`_attr_getter`
+        * :func:`_href_getter`
+        * :func:`_text_getter`
+        * :func:`_parse_td`
+        * :func:`_parse_thead_tr`
+        * :func:`_parse_tbody_tr`
+        * :func:`_parse_tfoot_tr`
+        * :func:`_parse_tables`
+        * :func:`_equals_tag`
+    See each method's respective documentation for details on their
+    functionality.
+    """
+
+    def __init__(
+        self,
+        io: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+        match: str | Pattern,
+        attrs: dict[str, str] | None,
+        encoding: str,
+        displayed_only: bool,
+        extract_links: Literal[None, "header", "footer", "body", "all"],
+        storage_options: StorageOptions = None,
+    ) -> None:
+        self.io = io
+        self.match = match
+        self.attrs = attrs
+        self.encoding = encoding
+        self.displayed_only = displayed_only
+        self.extract_links = extract_links
+        self.storage_options = storage_options
+
+    def parse_tables(self):
+        """
+        Parse and return all tables from the DOM.
+
+        Returns
+        -------
+        list of parsed (header, body, footer) tuples from tables.
+        """
+        tables = self._parse_tables(self._build_doc(), self.match, self.attrs)
+        return (self._parse_thead_tbody_tfoot(table) for table in tables)
+
+    def _attr_getter(self, obj, attr):
+        """
+        Return the attribute value of an individual DOM node.
+
+        Parameters
+        ----------
+        obj : node-like
+            A DOM node.
+
+        attr : str or unicode
+            The attribute, such as "colspan"
+
+        Returns
+        -------
+        str or unicode
+            The attribute value.
+        """
+        # Both lxml and BeautifulSoup have the same implementation:
+        return obj.get(attr)
+
+    def _href_getter(self, obj) -> str | None:
+        """
+        Return a href if the DOM node contains a child <a> or None.
+
+        Parameters
+        ----------
+        obj : node-like
+            A DOM node.
+
+        Returns
+        -------
+        href : str or unicode
+            The href from the <a> child of the DOM node.
+        """
+        raise AbstractMethodError(self)
+
+    def _text_getter(self, obj):
+        """
+        Return the text of an individual DOM node.
+
+        Parameters
+        ----------
+        obj : node-like
+            A DOM node.
+
+        Returns
+        -------
+        text : str or unicode
+            The text from an individual DOM node.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_td(self, obj):
+        """
+        Return the td elements from a row element.
+
+        Parameters
+        ----------
+        obj : node-like
+            A DOM <tr> node.
+
+        Returns
+        -------
+        list of node-like
+            These are the elements of each row, i.e., the columns.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_thead_tr(self, table):
+        """
+        Return the list of thead row elements from the parsed table element.
+
+        Parameters
+        ----------
+        table : a table element that contains zero or more thead elements.
+
+        Returns
+        -------
+        list of node-like
+            These are the <tr> row elements of a table.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_tbody_tr(self, table):
+        """
+        Return the list of tbody row elements from the parsed table element.
+
+        HTML5 table bodies consist of either 0 or more <tbody> elements (which
+        only contain <tr> elements) or 0 or more <tr> elements. This method
+        checks for both structures.
+
+        Parameters
+        ----------
+        table : a table element that contains row elements.
+
+        Returns
+        -------
+        list of node-like
+            These are the <tr> row elements of a table.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_tfoot_tr(self, table):
+        """
+        Return the list of tfoot row elements from the parsed table element.
+
+        Parameters
+        ----------
+        table : a table element that contains row elements.
+
+        Returns
+        -------
+        list of node-like
+            These are the <tr> row elements of a table.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_tables(self, document, match, attrs):
+        """
+        Return all tables from the parsed DOM.
+
+        Parameters
+        ----------
+        document : the DOM from which to parse the table element.
+
+        match : str or regular expression
+            The text to search for in the DOM tree.
+
+        attrs : dict
+            A dictionary of table attributes that can be used to disambiguate
+            multiple tables on a page.
+
+        Raises
+        ------
+        ValueError : `match` does not match any text in the document.
+
+        Returns
+        -------
+        list of node-like
+            HTML <table> elements to be parsed into raw data.
+        """
+        raise AbstractMethodError(self)
+
+    def _equals_tag(self, obj, tag) -> bool:
+        """
+        Return whether an individual DOM node matches a tag
+
+        Parameters
+        ----------
+        obj : node-like
+            A DOM node.
+
+        tag : str
+            Tag name to be checked for equality.
+
+        Returns
+        -------
+        boolean
+            Whether `obj`'s tag name is `tag`
+        """
+        raise AbstractMethodError(self)
+
+    def _build_doc(self):
+        """
+        Return a tree-like object that can be used to iterate over the DOM.
+
+        Returns
+        -------
+        node-like
+            The DOM from which to parse the table element.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_thead_tbody_tfoot(self, table_html):
+        """
+        Given a table, return parsed header, body, and foot.
+
+        Parameters
+        ----------
+        table_html : node-like
+
+        Returns
+        -------
+        tuple of (header, body, footer), each a list of list-of-text rows.
+
+        Notes
+        -----
+        Header and body are lists-of-lists. Top level list is a list of
+        rows. Each row is a list of str text.
+
+        Logic: Use <thead>, <tbody>, <tfoot> elements to identify
+               header, body, and footer, otherwise:
+               - Put all rows into body
+               - Move rows from top of body to header only if
+                 all elements inside row are <th>
+               - Move rows from bottom of body to footer only if
+                 all elements inside row are <th>
+        """
+        header_rows = self._parse_thead_tr(table_html)
+        body_rows = self._parse_tbody_tr(table_html)
+        footer_rows = self._parse_tfoot_tr(table_html)
+
+        def row_is_all_th(row):
+            return all(self._equals_tag(t, "th") for t in self._parse_td(row))
+
+        if not header_rows:
+            # The table has no <thead>. Move the top all-<th> rows from
+            # body_rows to header_rows. (This is a common case because many
+            # tables in the wild have no <thead> or <tfoot>
+            while body_rows and row_is_all_th(body_rows[0]):
+                header_rows.append(body_rows.pop(0))
+
+        header, rem = self._expand_colspan_rowspan(header_rows, section="header")
+        body, rem = self._expand_colspan_rowspan(
+            body_rows,
+            section="body",
+            remainder=rem,
+            overflow=len(footer_rows) > 0,
+        )
+        footer, _ = self._expand_colspan_rowspan(
+            footer_rows, section="footer", remainder=rem, overflow=False
+        )
+
+        return header, body, footer
+
+    def _expand_colspan_rowspan(
+        self,
+        rows,
+        section: Literal["header", "footer", "body"],
+        remainder: list[tuple[int, str | tuple, int]] | None = None,
+        overflow: bool = True,
+    ) -> tuple[list[list], list[tuple[int, str | tuple, int]]]:
+        """
+        Given a list of <tr>s, return a list of text rows.
+
+        Parameters
+        ----------
+        rows : list of node-like
+            List of <tr>s
+        section : the section that the rows belong to (header, body or footer).
+        remainder: list[tuple[int, str | tuple, int]] | None
+            Any remainder from the expansion of previous section
+        overflow: bool
+            If true, return any partial rows as 'remainder'. If not, use up any
+            partial rows. True by default.
+
+        Returns
+        -------
+        list of list
+            Each returned row is a list of str text, or tuple (text, link)
+            if extract_links is not None.
+        remainder
+            Remaining partial rows if any. If overflow is False, an empty list
+            is returned.
+
+        Notes
+        -----
+        Any cell with ``rowspan`` or ``colspan`` will have its contents copied
+        to subsequent cells.
+        """
+        all_texts = []  # list of rows, each a list of str
+        text: str | tuple
+        remainder = remainder if remainder is not None else []
+
+        for tr in rows:
+            texts = []  # the output for this row
+            next_remainder = []
+
+            index = 0
+            tds = self._parse_td(tr)
+            for td in tds:
+                # Append texts from previous rows with rowspan>1 that come
+                # before this <td>
+                while remainder and remainder[0][0] <= index:
+                    prev_i, prev_text, prev_rowspan = remainder.pop(0)
+                    texts.append(prev_text)
+                    if prev_rowspan > 1:
+                        next_remainder.append((prev_i, prev_text, prev_rowspan - 1))
+                    index += 1
+
+                # Append the text from this <td>, colspan times
+                text = _remove_whitespace(self._text_getter(td))
+                if self.extract_links in ("all", section):
+                    href = self._href_getter(td)
+                    text = (text, href)
+                rowspan = int(self._attr_getter(td, "rowspan") or 1)
+                colspan = int(self._attr_getter(td, "colspan") or 1)
+
+                for _ in range(colspan):
+                    texts.append(text)
+                    if rowspan > 1:
+                        next_remainder.append((index, text, rowspan - 1))
+                    index += 1
+
+            # Append texts from previous rows at the final position
+            for prev_i, prev_text, prev_rowspan in remainder:
+                texts.append(prev_text)
+                if prev_rowspan > 1:
+                    next_remainder.append((prev_i, prev_text, prev_rowspan - 1))
+
+            all_texts.append(texts)
+            remainder = next_remainder
+
+        if not overflow:
+            # Append rows that only appear because the previous row had non-1
+            # rowspan
+            while remainder:
+                next_remainder = []
+                texts = []
+                for prev_i, prev_text, prev_rowspan in remainder:
+                    texts.append(prev_text)
+                    if prev_rowspan > 1:
+                        next_remainder.append((prev_i, prev_text, prev_rowspan - 1))
+                all_texts.append(texts)
+                remainder = next_remainder
+
+        return all_texts, remainder
+
+    def _handle_hidden_tables(self, tbl_list, attr_name: str):
+        """
+        Return list of tables, potentially removing hidden elements
+
+        Parameters
+        ----------
+        tbl_list : list of node-like
+            Type of list elements will vary depending upon parser used
+        attr_name : str
+            Name of the accessor for retrieving HTML attributes
+
+        Returns
+        -------
+        list of node-like
+            Return type matches `tbl_list`
+        """
+        if not self.displayed_only:
+            return tbl_list
+
+        return [
+            x
+            for x in tbl_list
+            if "display:none"
+            not in getattr(x, attr_name).get("style", "").replace(" ", "")
+        ]
+
+
+class _BeautifulSoupHtml5LibFrameParser(_HtmlFrameParser):
+    """
+    HTML to DataFrame parser that uses BeautifulSoup under the hood.
+
+    See Also
+    --------
+    pandas.io.html._HtmlFrameParser
+    pandas.io.html._LxmlFrameParser
+
+    Notes
+    -----
+    Documentation strings for this class are in the base class
+    :class:`pandas.io.html._HtmlFrameParser`.
+    """
+
+    def _parse_tables(self, document, match, attrs):
+        element_name = "table"
+        tables = document.find_all(element_name, attrs=attrs)
+        if not tables:
+            raise ValueError("No tables found")
+
+        result = []
+        unique_tables = set()
+        tables = self._handle_hidden_tables(tables, "attrs")
+
+        for table in tables:
+            if self.displayed_only:
+                for elem in table.find_all("style"):
+                    elem.decompose()
+
+                for elem in table.find_all(style=re.compile(r"display:\s*none")):
+                    elem.decompose()
+
+            if table not in unique_tables and table.find(string=match) is not None:
+                result.append(table)
+            unique_tables.add(table)
+        if not result:
+            raise ValueError(f"No tables found matching pattern {match.pattern!r}")
+        return result
+
+    def _href_getter(self, obj) -> str | None:
+        a = obj.find("a", href=True)
+        return None if not a else a["href"]
+
+    def _text_getter(self, obj):
+        return obj.text
+
+    def _equals_tag(self, obj, tag) -> bool:
+        return obj.name == tag
+
+    def _parse_td(self, row):
+        return row.find_all(("td", "th"), recursive=False)
+
+    def _parse_thead_tr(self, table):
+        return table.select("thead tr")
+
+    def _parse_tbody_tr(self, table):
+        from_tbody = table.select("tbody tr")
+        from_root = table.find_all("tr", recursive=False)
+        # HTML spec: at most one of these lists has content
+        return from_tbody + from_root
+
+    def _parse_tfoot_tr(self, table):
+        return table.select("tfoot tr")
+
+    def _setup_build_doc(self):
+        raw_text = _read(self.io, self.encoding, self.storage_options)
+        if not raw_text:
+            raise ValueError(f"No text parsed from document: {self.io}")
+        return raw_text
+
+    def _build_doc(self):
+        from bs4 import BeautifulSoup
+
+        bdoc = self._setup_build_doc()
+        if isinstance(bdoc, bytes) and self.encoding is not None:
+            udoc = bdoc.decode(self.encoding)
+            from_encoding = None
+        else:
+            udoc = bdoc
+            from_encoding = self.encoding
+
+        soup = BeautifulSoup(udoc, features="html5lib", from_encoding=from_encoding)
+
+        for br in soup.find_all("br"):
+            br.replace_with("\n" + br.text)
+
+        return soup
+
+
+def _build_xpath_expr(attrs) -> str:
+    """
+    Build an xpath expression to simulate bs4's ability to pass in kwargs to
+    search for attributes when using the lxml parser.
+
+    Parameters
+    ----------
+    attrs : dict
+        A dict of HTML attributes. These are NOT checked for validity.
+
+    Returns
+    -------
+    expr : unicode
+        An XPath expression that checks for the given HTML attributes.
+    """
+    # give class attribute as class_ because class is a python keyword
+    if "class_" in attrs:
+        attrs["class"] = attrs.pop("class_")
+
+    s = " and ".join([f"@{k}={v!r}" for k, v in attrs.items()])
+    return f"[{s}]"
+
+
+_re_namespace = {"re": "http://exslt.org/regular-expressions"}
+
+
+class _LxmlFrameParser(_HtmlFrameParser):
+    """
+    HTML to DataFrame parser that uses lxml under the hood.
+
+    Warning
+    -------
+    This parser can only handle HTTP, FTP, and FILE urls.
+
+    See Also
+    --------
+    _HtmlFrameParser
+    _BeautifulSoupLxmlFrameParser
+
+    Notes
+    -----
+    Documentation strings for this class are in the base class
+    :class:`_HtmlFrameParser`.
+    """
+
+    def _href_getter(self, obj) -> str | None:
+        href = obj.xpath(".//a/@href")
+        return None if not href else href[0]
+
+    def _text_getter(self, obj):
+        return obj.text_content()
+
+    def _parse_td(self, row):
+        # Look for direct children only: the "row" element here may be a
+        # <thead> or <tfoot> (see _parse_thead_tr).
+        return row.xpath("./td|./th")
+
+    def _parse_tables(self, document, match, kwargs):
+        pattern = match.pattern
+
+        # 1. check all descendants for the given pattern and only search tables
+        # GH 49929
+        xpath_expr = f"//table[.//text()[re:test(., {pattern!r})]]"
+
+        # if any table attributes were given build an xpath expression to
+        # search for them
+        if kwargs:
+            xpath_expr += _build_xpath_expr(kwargs)
+
+        tables = document.xpath(xpath_expr, namespaces=_re_namespace)
+
+        tables = self._handle_hidden_tables(tables, "attrib")
+        if self.displayed_only:
+            for table in tables:
+                # lxml utilizes XPATH 1.0 which does not have regex
+                # support. As a result, we find all elements with a style
+                # attribute and iterate them to check for display:none
+                for elem in table.xpath(".//style"):
+                    elem.drop_tree()
+                for elem in table.xpath(".//*[@style]"):
+                    if "display:none" in elem.attrib.get("style", "").replace(" ", ""):
+                        elem.drop_tree()
+        if not tables:
+            raise ValueError(f"No tables found matching regex {pattern!r}")
+        return tables
+
+    def _equals_tag(self, obj, tag) -> bool:
+        return obj.tag == tag
+
+    def _build_doc(self):
+        """
+        Raises
+        ------
+        ValueError
+            * If a URL that lxml cannot parse is passed.
+
+        Exception
+            * Any other ``Exception`` thrown. For example, trying to parse a
+              URL that is syntactically correct on a machine with no internet
+              connection will fail.
+
+        See Also
+        --------
+        pandas.io.html._HtmlFrameParser._build_doc
+        """
+        from lxml.etree import XMLSyntaxError
+        from lxml.html import (
+            HTMLParser,
+            parse,
+        )
+
+        parser = HTMLParser(recover=True, encoding=self.encoding)
+
+        if is_url(self.io):
+            with get_handle(self.io, "r", storage_options=self.storage_options) as f:
+                r = parse(f.handle, parser=parser)
+        else:
+            # try to parse the input in the simplest way
+            try:
+                r = parse(self.io, parser=parser)
+            except OSError as err:
+                raise FileNotFoundError(
+                    f"[Errno {errno.ENOENT}] {os.strerror(errno.ENOENT)}: {self.io}"
+                ) from err
+        try:
+            r = r.getroot()
+        except AttributeError:
+            pass
+        else:
+            if not hasattr(r, "text_content"):
+                raise XMLSyntaxError("no text parsed from document", 0, 0, 0)
+
+        for br in r.xpath("*//br"):
+            br.tail = "\n" + (br.tail or "")
+
+        return r
+
+    def _parse_thead_tr(self, table):
+        rows = []
+
+        for thead in table.xpath(".//thead"):
+            rows.extend(thead.xpath("./tr"))
+
+            # HACK: lxml does not clean up the clearly-erroneous
+            # <thead><th>foo</th><th>bar</th></thead>. (Missing <tr>). Add
+            # the <thead> and _pretend_ it's a <tr>; _parse_td() will find its
+            # children as though it's a <tr>.
+            #
+            # Better solution would be to use html5lib.
+            elements_at_root = thead.xpath("./td|./th")
+            if elements_at_root:
+                rows.append(thead)
+
+        return rows
+
+    def _parse_tbody_tr(self, table):
+        from_tbody = table.xpath(".//tbody//tr")
+        from_root = table.xpath("./tr")
+        # HTML spec: at most one of these lists has content
+        return from_tbody + from_root
+
+    def _parse_tfoot_tr(self, table):
+        return table.xpath(".//tfoot//tr")
+
+
+def _expand_elements(body) -> None:
+    data = [len(elem) for elem in body]
+    lens = Series(data)
+    lens_max = lens.max()
+    not_max = lens[lens != lens_max]
+
+    empty = [""]
+    for ind, length in not_max.items():
+        body[ind] += empty * (lens_max - length)
+
+
+def _data_to_frame(**kwargs):
+    head, body, foot = kwargs.pop("data")
+    header = kwargs.pop("header")
+    kwargs["skiprows"] = _get_skiprows(kwargs["skiprows"])
+    if head:
+        body = head + body
+
+        # Infer header when there is a <thead> or top <th>-only rows
+        if header is None:
+            if len(head) == 1:
+                header = 0
+            else:
+                # ignore all-empty-text rows
+                header = [i for i, row in enumerate(head) if any(text for text in row)]
+
+    if foot:
+        body += foot
+
+    # fill out elements of body that are "ragged"
+    _expand_elements(body)
+    with TextParser(body, header=header, **kwargs) as tp:
+        return tp.read()
+
+
+_valid_parsers = {
+    "lxml": _LxmlFrameParser,
+    None: _LxmlFrameParser,
+    "html5lib": _BeautifulSoupHtml5LibFrameParser,
+    "bs4": _BeautifulSoupHtml5LibFrameParser,
+}
+
+
+def _parser_dispatch(flavor: HTMLFlavors | None) -> type[_HtmlFrameParser]:
+    """
+    Choose the parser based on the input flavor.
+
+    Parameters
+    ----------
+    flavor : {{"lxml", "html5lib", "bs4"}} or None
+        The type of parser to use. This must be a valid backend.
+
+    Returns
+    -------
+    cls : _HtmlFrameParser subclass
+        The parser class based on the requested input flavor.
+
+    Raises
+    ------
+    ValueError
+        * If `flavor` is not a valid backend.
+    ImportError
+        * If you do not have the requested `flavor`
+    """
+    valid_parsers = list(_valid_parsers.keys())
+    if flavor not in valid_parsers:
+        raise ValueError(
+            f"{flavor!r} is not a valid flavor, valid flavors are {valid_parsers}"
+        )
+
+    if flavor in ("bs4", "html5lib"):
+        import_optional_dependency("html5lib")
+        import_optional_dependency("bs4")
+    else:
+        import_optional_dependency("lxml.etree")
+    return _valid_parsers[flavor]
+
+
+def _print_as_set(s) -> str:
+    arg = ", ".join([pprint_thing(el) for el in s])
+    return f"{{{arg}}}"
+
+
+def _validate_flavor(flavor):
+    if flavor is None:
+        flavor = "lxml", "bs4"
+    elif isinstance(flavor, str):
+        flavor = (flavor,)
+    elif isinstance(flavor, abc.Iterable):
+        if not all(isinstance(flav, str) for flav in flavor):
+            raise TypeError(
+                f"Object of type {type(flavor).__name__!r} "
+                f"is not an iterable of strings"
+            )
+    else:
+        msg = repr(flavor) if isinstance(flavor, str) else str(flavor)
+        msg += " is not a valid flavor"
+        raise ValueError(msg)
+
+    flavor = tuple(flavor)
+    valid_flavors = set(_valid_parsers)
+    flavor_set = set(flavor)
+
+    if not flavor_set & valid_flavors:
+        raise ValueError(
+            f"{_print_as_set(flavor_set)} is not a valid set of flavors, valid "
+            f"flavors are {_print_as_set(valid_flavors)}"
+        )
+    return flavor
+
+
+def _parse(
+    flavor,
+    io,
+    match,
+    attrs,
+    encoding,
+    displayed_only,
+    extract_links,
+    storage_options,
+    **kwargs,
+):
+    flavor = _validate_flavor(flavor)
+    compiled_match = re.compile(match)  # you can pass a compiled regex here
+
+    retained = None
+    for flav in flavor:
+        parser = _parser_dispatch(flav)
+        p = parser(
+            io,
+            compiled_match,
+            attrs,
+            encoding,
+            displayed_only,
+            extract_links,
+            storage_options,
+        )
+
+        try:
+            tables = p.parse_tables()
+        except ValueError as caught:
+            # if `io` is an io-like object, check if it's seekable
+            # and try to rewind it before trying the next parser
+            if hasattr(io, "seekable") and io.seekable():
+                io.seek(0)
+            elif hasattr(io, "seekable") and not io.seekable():
+                # if we couldn't rewind it, let the user know
+                raise ValueError(
+                    f"The flavor {flav} failed to parse your input. "
+                    "Since you passed a non-rewindable file "
+                    "object, we can't rewind it to try "
+                    "another parser. Try read_html() with a different flavor."
+                ) from caught
+
+            retained = caught
+        else:
+            break
+    else:
+        assert retained is not None  # for mypy
+        raise retained
+
+    ret = []
+    for table in tables:
+        try:
+            df = _data_to_frame(data=table, **kwargs)
+            # Cast MultiIndex header to an Index of tuples when extracting header
+            # links and replace nan with None (therefore can't use mi.to_flat_index()).
+            # This maintains consistency of selection (e.g. df.columns.str[1])
+            if extract_links in ("all", "header") and isinstance(
+                df.columns, MultiIndex
+            ):
+                df.columns = Index(
+                    ((col[0], None if isna(col[1]) else col[1]) for col in df.columns),
+                    tupleize_cols=False,
+                )
+
+            ret.append(df)
+        except EmptyDataError:  # empty table
+            continue
+    return ret
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def read_html(
+    io: FilePath | ReadBuffer[str],
+    *,
+    match: str | Pattern = ".+",
+    flavor: HTMLFlavors | Sequence[HTMLFlavors] | None = None,
+    header: int | Sequence[int] | None = None,
+    index_col: int | Sequence[int] | None = None,
+    skiprows: int | Sequence[int] | slice | None = None,
+    attrs: dict[str, str] | None = None,
+    parse_dates: bool = False,
+    thousands: str | None = ",",
+    encoding: str | None = None,
+    decimal: str = ".",
+    converters: dict | None = None,
+    na_values: Iterable[object] | None = None,
+    keep_default_na: bool = True,
+    displayed_only: bool = True,
+    extract_links: Literal[None, "header", "footer", "body", "all"] = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    storage_options: StorageOptions = None,
+) -> list[DataFrame]:
+    r"""
+    Read HTML tables into a ``list`` of ``DataFrame`` objects.
+
+    Parameters
+    ----------
+    io : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a string ``read()`` function.
+        The string can represent a URL. Note that
+        lxml only accepts the http, ftp and file url protocols. If you have a
+        URL that starts with ``'https'`` you might try removing the ``'s'``.
+
+        .. deprecated:: 2.1.0
+            Passing html literal strings is deprecated.
+            Wrap literal string/bytes input in ``io.StringIO``/``io.BytesIO`` instead.
+
+    match : str or compiled regular expression, optional
+        The set of tables containing text matching this regex or string will be
+        returned. Unless the HTML is extremely simple you will probably need to
+        pass a non-empty string here. Defaults to '.+' (match any non-empty
+        string). The default value will return all tables contained on a page.
+        This value is converted to a regular expression so that there is
+        consistent behavior between Beautiful Soup and lxml.
+
+    flavor : {{"lxml", "html5lib", "bs4"}} or list-like, optional
+        The parsing engine (or list of parsing engines) to use. 'bs4' and
+        'html5lib' are synonymous with each other, they are both there for
+        backwards compatibility. The default of ``None`` tries to use ``lxml``
+        to parse and if that fails it falls back on ``bs4`` + ``html5lib``.
+
+    header : int or list-like, optional
+        The row (or list of rows for a :class:`~pandas.MultiIndex`) to use to
+        make the columns headers.
+
+    index_col : int or list-like, optional
+        The column (or list of columns) to use to create the index.
+
+    skiprows : int, list-like or slice, optional
+        Number of rows to skip after parsing the column integer. 0-based. If a
+        sequence of integers or a slice is given, will skip the rows indexed by
+        that sequence.  Note that a single element sequence means 'skip the nth
+        row' whereas an integer means 'skip n rows'.
+
+    attrs : dict, optional
+        This is a dictionary of attributes that you can pass to use to identify
+        the table in the HTML. These are not checked for validity before being
+        passed to lxml or Beautiful Soup. However, these attributes must be
+        valid HTML table attributes to work correctly. For example, ::
+
+            attrs = {{"id": "table"}}
+
+        is a valid attribute dictionary because the 'id' HTML tag attribute is
+        a valid HTML attribute for *any* HTML tag as per `this document
+        <https://html.spec.whatwg.org/multipage/dom.html#global-attributes>`__. ::
+
+            attrs = {{"asdf": "table"}}
+
+        is *not* a valid attribute dictionary because 'asdf' is not a valid
+        HTML attribute even if it is a valid XML attribute.  Valid HTML 4.01
+        table attributes can be found `here
+        <http://www.w3.org/TR/REC-html40/struct/tables.html#h-11.2>`__. A
+        working draft of the HTML 5 spec can be found `here
+        <https://html.spec.whatwg.org/multipage/tables.html>`__. It contains the
+        latest information on table attributes for the modern web.
+
+    parse_dates : bool, optional
+        See :func:`~read_csv` for more details.
+
+    thousands : str, optional
+        Separator to use to parse thousands. Defaults to ``','``.
+
+    encoding : str, optional
+        The encoding used to decode the web page. Defaults to ``None``.``None``
+        preserves the previous encoding behavior, which depends on the
+        underlying parser library (e.g., the parser library will try to use
+        the encoding provided by the document).
+
+    decimal : str, default '.'
+        Character to recognize as decimal point (e.g. use ',' for European
+        data).
+
+    converters : dict, default None
+        Dict of functions for converting values in certain columns. Keys can
+        either be integers or column labels, values are functions that take one
+        input argument, the cell (not column) content, and return the
+        transformed content.
+
+    na_values : iterable, default None
+        Custom NA values.
+
+    keep_default_na : bool, default True
+        If na_values are specified and keep_default_na is False the default NaN
+        values are overridden, otherwise they're appended to.
+
+    displayed_only : bool, default True
+        Whether elements with "display: none" should be parsed.
+
+    extract_links : {{None, "all", "header", "body", "footer"}}
+        Table elements in the specified section(s) with <a> tags will have their
+        href extracted.
+
+        .. versionadded:: 1.5.0
+
+    dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    {storage_options}
+
+        .. versionadded:: 2.1.0
+
+    Returns
+    -------
+    dfs
+        A list of DataFrames.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+
+    Notes
+    -----
+    Before using this function you should read the :ref:`gotchas about the
+    HTML parsing libraries <io.html.gotchas>`.
+
+    Expect to do some cleanup after you call this function. For example, you
+    might need to manually assign column names if the column names are
+    converted to NaN when you pass the `header=0` argument. We try to assume as
+    little as possible about the structure of the table and push the
+    idiosyncrasies of the HTML contained in the table to the user.
+
+    This function searches for ``<table>`` elements and only for ``<tr>``
+    and ``<th>`` rows and ``<td>`` elements within each ``<tr>`` or ``<th>``
+    element in the table. ``<td>`` stands for "table data". This function
+    attempts to properly handle ``colspan`` and ``rowspan`` attributes.
+    If the function has a ``<thead>`` argument, it is used to construct
+    the header, otherwise the function attempts to find the header within
+    the body (by putting rows with only ``<th>`` elements into the header).
+
+    Similar to :func:`~read_csv` the `header` argument is applied
+    **after** `skiprows` is applied.
+
+    This function will *always* return a list of :class:`DataFrame` *or*
+    it will fail, i.e., it will *not* return an empty list, save for some
+    rare cases.
+    It might return an empty list in case of inputs with single row and
+    ``<td>`` containing only whitespaces.
+
+    Examples
+    --------
+    See the :ref:`read_html documentation in the IO section of the docs
+    <io.read_html>` for some examples of reading in HTML tables.
+    """
+    # Type check here. We don't want to parse only to fail because of an
+    # invalid value of an integer skiprows.
+    if isinstance(skiprows, numbers.Integral) and skiprows < 0:
+        raise ValueError(
+            "cannot skip rows starting from the end of the "
+            "data (you passed a negative value)"
+        )
+    if extract_links not in [None, "header", "footer", "body", "all"]:
+        raise ValueError(
+            "`extract_links` must be one of "
+            '{None, "header", "footer", "body", "all"}, got '
+            f'"{extract_links}"'
+        )
+
+    validate_header_arg(header)
+    check_dtype_backend(dtype_backend)
+
+    io = stringify_path(io)
+
+    return _parse(
+        flavor=flavor,
+        io=io,
+        match=match,
+        header=header,
+        index_col=index_col,
+        skiprows=skiprows,
+        parse_dates=parse_dates,
+        thousands=thousands,
+        attrs=attrs,
+        encoding=encoding,
+        decimal=decimal,
+        converters=converters,
+        na_values=na_values,
+        keep_default_na=keep_default_na,
+        displayed_only=displayed_only,
+        extract_links=extract_links,
+        dtype_backend=dtype_backend,
+        storage_options=storage_options,
+    )
diff --git a/pandas/pandas/io/json/__init__.py b/pandas/pandas/io/json/__init__.py
new file mode 100644
index 0000000000000..39f78e26d6041
--- /dev/null
+++ b/pandas/pandas/io/json/__init__.py
@@ -0,0 +1,15 @@
+from pandas.io.json._json import (
+    read_json,
+    to_json,
+    ujson_dumps,
+    ujson_loads,
+)
+from pandas.io.json._table_schema import build_table_schema
+
+__all__ = [
+    "build_table_schema",
+    "read_json",
+    "to_json",
+    "ujson_dumps",
+    "ujson_loads",
+]
diff --git a/pandas/pandas/io/json/_json.py b/pandas/pandas/io/json/_json.py
new file mode 100644
index 0000000000000..e032e26d771d7
--- /dev/null
+++ b/pandas/pandas/io/json/_json.py
@@ -0,0 +1,1383 @@
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from collections import abc
+from itertools import islice
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Generic,
+    Literal,
+    TypeVar,
+    final,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.json import (
+    ujson_dumps,
+    ujson_loads,
+)
+from pandas._libs.tslibs import iNaT
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import doc
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import (
+    ensure_str,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    isna,
+    notna,
+    to_datetime,
+)
+from pandas.core.reshape.concat import concat
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io._util import arrow_table_to_pandas
+from pandas.io.common import (
+    IOHandles,
+    dedup_names,
+    get_handle,
+    is_potential_multi_index,
+    stringify_path,
+)
+from pandas.io.json._normalize import convert_to_line_delimits
+from pandas.io.json._table_schema import (
+    build_table_schema,
+    parse_table_schema,
+    set_default_names,
+)
+from pandas.io.parsers.readers import validate_integer
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Mapping,
+    )
+    from types import TracebackType
+
+    from pandas._typing import (
+        CompressionOptions,
+        DtypeArg,
+        DtypeBackend,
+        FilePath,
+        IndexLabel,
+        JSONEngine,
+        JSONSerializable,
+        ReadBuffer,
+        Self,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+    from pandas.core.generic import NDFrame
+
+FrameSeriesStrT = TypeVar("FrameSeriesStrT", bound=Literal["frame", "series"])
+
+
+# interface to/from
+@overload
+def to_json(
+    path_or_buf: FilePath | WriteBuffer[str] | WriteBuffer[bytes],
+    obj: NDFrame,
+    orient: str | None = ...,
+    date_format: str = ...,
+    double_precision: int = ...,
+    force_ascii: bool = ...,
+    date_unit: str = ...,
+    default_handler: Callable[[Any], JSONSerializable] | None = ...,
+    lines: bool = ...,
+    compression: CompressionOptions = ...,
+    index: bool | None = ...,
+    indent: int = ...,
+    storage_options: StorageOptions = ...,
+    mode: Literal["a", "w"] = ...,
+) -> None: ...
+
+
+@overload
+def to_json(
+    path_or_buf: None,
+    obj: NDFrame,
+    orient: str | None = ...,
+    date_format: str = ...,
+    double_precision: int = ...,
+    force_ascii: bool = ...,
+    date_unit: str = ...,
+    default_handler: Callable[[Any], JSONSerializable] | None = ...,
+    lines: bool = ...,
+    compression: CompressionOptions = ...,
+    index: bool | None = ...,
+    indent: int = ...,
+    storage_options: StorageOptions = ...,
+    mode: Literal["a", "w"] = ...,
+) -> str: ...
+
+
+def to_json(
+    path_or_buf: FilePath | WriteBuffer[str] | WriteBuffer[bytes] | None,
+    obj: NDFrame,
+    orient: str | None = None,
+    date_format: str = "epoch",
+    double_precision: int = 10,
+    force_ascii: bool = True,
+    date_unit: str = "ms",
+    default_handler: Callable[[Any], JSONSerializable] | None = None,
+    lines: bool = False,
+    compression: CompressionOptions = "infer",
+    index: bool | None = None,
+    indent: int = 0,
+    storage_options: StorageOptions | None = None,
+    mode: Literal["a", "w"] = "w",
+) -> str | None:
+    if orient in ["records", "values"] and index is True:
+        raise ValueError(
+            "'index=True' is only valid when 'orient' is 'split', 'table', "
+            "'index', or 'columns'."
+        )
+    elif orient in ["index", "columns"] and index is False:
+        raise ValueError(
+            "'index=False' is only valid when 'orient' is 'split', 'table', "
+            "'records', or 'values'."
+        )
+    elif index is None:
+        # will be ignored for orient='records' and 'values'
+        index = True
+
+    if lines and orient != "records":
+        raise ValueError("'lines' keyword only valid when 'orient' is records")
+
+    if mode not in ["a", "w"]:
+        msg = (
+            f"mode={mode} is not a valid option."
+            "Only 'w' and 'a' are currently supported."
+        )
+        raise ValueError(msg)
+
+    if mode == "a" and (not lines or orient != "records"):
+        msg = (
+            "mode='a' (append) is only supported when "
+            "lines is True and orient is 'records'"
+        )
+        raise ValueError(msg)
+
+    if orient == "table" and isinstance(obj, Series):
+        obj = obj.to_frame(name=obj.name or "values")
+
+    writer: type[Writer]
+    if orient == "table" and isinstance(obj, DataFrame):
+        writer = JSONTableWriter
+    elif isinstance(obj, Series):
+        writer = SeriesWriter
+    elif isinstance(obj, DataFrame):
+        writer = FrameWriter
+    else:
+        raise NotImplementedError("'obj' should be a Series or a DataFrame")
+
+    s = writer(
+        obj,
+        orient=orient,
+        date_format=date_format,
+        double_precision=double_precision,
+        ensure_ascii=force_ascii,
+        date_unit=date_unit,
+        default_handler=default_handler,
+        index=index,
+        indent=indent,
+    ).write()
+
+    if lines:
+        s = convert_to_line_delimits(s)
+
+    if path_or_buf is not None:
+        # apply compression and byte/text conversion
+        with get_handle(
+            path_or_buf, mode, compression=compression, storage_options=storage_options
+        ) as handles:
+            handles.handle.write(s)
+    else:
+        return s
+    return None
+
+
+class Writer(ABC):
+    _default_orient: str
+
+    def __init__(
+        self,
+        obj: NDFrame,
+        orient: str | None,
+        date_format: str,
+        double_precision: int,
+        ensure_ascii: bool,
+        date_unit: str,
+        index: bool,
+        default_handler: Callable[[Any], JSONSerializable] | None = None,
+        indent: int = 0,
+    ) -> None:
+        self.obj = obj
+
+        if orient is None:
+            orient = self._default_orient
+
+        self.orient = orient
+        self.date_format = date_format
+        self.double_precision = double_precision
+        self.ensure_ascii = ensure_ascii
+        self.date_unit = date_unit
+        self.default_handler = default_handler
+        self.index = index
+        self.indent = indent
+        self._format_axes()
+
+    def _format_axes(self) -> None:
+        raise AbstractMethodError(self)
+
+    def write(self) -> str:
+        iso_dates = self.date_format == "iso"
+        return ujson_dumps(
+            self.obj_to_write,
+            orient=self.orient,
+            double_precision=self.double_precision,
+            ensure_ascii=self.ensure_ascii,
+            date_unit=self.date_unit,
+            iso_dates=iso_dates,
+            default_handler=self.default_handler,
+            indent=self.indent,
+        )
+
+    @property
+    @abstractmethod
+    def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
+        """Object to write in JSON format."""
+
+
+class SeriesWriter(Writer):
+    _default_orient = "index"
+
+    @property
+    def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
+        if not self.index and self.orient == "split":
+            return {"name": self.obj.name, "data": self.obj.values}
+        else:
+            return self.obj
+
+    def _format_axes(self) -> None:
+        if not self.obj.index.is_unique and self.orient == "index":
+            raise ValueError(f"Series index must be unique for orient='{self.orient}'")
+
+
+class FrameWriter(Writer):
+    _default_orient = "columns"
+
+    @property
+    def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
+        if not self.index and self.orient == "split":
+            obj_to_write = self.obj.to_dict(orient="split")
+            del obj_to_write["index"]
+        else:
+            obj_to_write = self.obj
+        return obj_to_write
+
+    def _format_axes(self) -> None:
+        """
+        Try to format axes if they are datelike.
+        """
+        if not self.obj.index.is_unique and self.orient in ("index", "columns"):
+            raise ValueError(
+                f"DataFrame index must be unique for orient='{self.orient}'."
+            )
+        if not self.obj.columns.is_unique and self.orient in (
+            "index",
+            "columns",
+            "records",
+        ):
+            raise ValueError(
+                f"DataFrame columns must be unique for orient='{self.orient}'."
+            )
+
+
+class JSONTableWriter(FrameWriter):
+    _default_orient = "records"
+
+    def __init__(
+        self,
+        obj,
+        orient: str | None,
+        date_format: str,
+        double_precision: int,
+        ensure_ascii: bool,
+        date_unit: str,
+        index: bool,
+        default_handler: Callable[[Any], JSONSerializable] | None = None,
+        indent: int = 0,
+    ) -> None:
+        """
+        Adds a `schema` attribute with the Table Schema, resets
+        the index (can't do in caller, because the schema inference needs
+        to know what the index is, forces orient to records, and forces
+        date_format to 'iso'.
+        """
+        super().__init__(
+            obj,
+            orient,
+            date_format,
+            double_precision,
+            ensure_ascii,
+            date_unit,
+            index,
+            default_handler=default_handler,
+            indent=indent,
+        )
+
+        if date_format != "iso":
+            msg = (
+                "Trying to write with `orient='table'` and "
+                f"`date_format='{date_format}'`. Table Schema requires dates "
+                "to be formatted with `date_format='iso'`"
+            )
+            raise ValueError(msg)
+
+        self.schema = build_table_schema(obj, index=self.index)
+        if self.index:
+            obj = set_default_names(obj)
+
+        # NotImplemented on a column MultiIndex
+        if obj.ndim == 2 and isinstance(obj.columns, MultiIndex):
+            raise NotImplementedError(
+                "orient='table' is not supported for MultiIndex columns"
+            )
+
+        # TODO: Do this timedelta properly in objToJSON.c See GH #15137
+        if ((obj.ndim == 1) and (obj.name in set(obj.index.names))) or len(
+            obj.columns.intersection(obj.index.names)
+        ):
+            msg = "Overlapping names between the index and columns"
+            raise ValueError(msg)
+
+        timedeltas = obj.select_dtypes(include=["timedelta"]).columns
+        copied = False
+        if len(timedeltas):
+            obj = obj.copy()
+            copied = True
+            obj[timedeltas] = obj[timedeltas].map(lambda x: x.isoformat())
+
+        # exclude index from obj if index=False
+        if not self.index:
+            self.obj = obj.reset_index(drop=True)
+        else:
+            # Convert PeriodIndex to datetimes before serializing
+            if isinstance(obj.index.dtype, PeriodDtype):
+                if not copied:
+                    obj = obj.copy(deep=False)
+                obj.index = obj.index.to_timestamp()
+            self.obj = obj.reset_index(drop=False)
+        self.date_format = "iso"
+        self.orient = "records"
+        self.index = index
+
+    @property
+    def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
+        return {"schema": self.schema, "data": self.obj}
+
+
+@overload
+def read_json(
+    path_or_buf: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+    *,
+    orient: str | None = ...,
+    typ: Literal["frame"] = ...,
+    dtype: DtypeArg | None = ...,
+    convert_axes: bool | None = ...,
+    convert_dates: bool | list[str] = ...,
+    keep_default_dates: bool = ...,
+    precise_float: bool = ...,
+    date_unit: str | None = ...,
+    encoding: str | None = ...,
+    encoding_errors: str | None = ...,
+    lines: bool = ...,
+    chunksize: int,
+    compression: CompressionOptions = ...,
+    nrows: int | None = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    engine: JSONEngine = ...,
+) -> JsonReader[Literal["frame"]]: ...
+
+
+@overload
+def read_json(
+    path_or_buf: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+    *,
+    orient: str | None = ...,
+    typ: Literal["series"],
+    dtype: DtypeArg | None = ...,
+    convert_axes: bool | None = ...,
+    convert_dates: bool | list[str] = ...,
+    keep_default_dates: bool = ...,
+    precise_float: bool = ...,
+    date_unit: str | None = ...,
+    encoding: str | None = ...,
+    encoding_errors: str | None = ...,
+    lines: bool = ...,
+    chunksize: int,
+    compression: CompressionOptions = ...,
+    nrows: int | None = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    engine: JSONEngine = ...,
+) -> JsonReader[Literal["series"]]: ...
+
+
+@overload
+def read_json(
+    path_or_buf: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+    *,
+    orient: str | None = ...,
+    typ: Literal["series"],
+    dtype: DtypeArg | None = ...,
+    convert_axes: bool | None = ...,
+    convert_dates: bool | list[str] = ...,
+    keep_default_dates: bool = ...,
+    precise_float: bool = ...,
+    date_unit: str | None = ...,
+    encoding: str | None = ...,
+    encoding_errors: str | None = ...,
+    lines: bool = ...,
+    chunksize: None = ...,
+    compression: CompressionOptions = ...,
+    nrows: int | None = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    engine: JSONEngine = ...,
+) -> Series: ...
+
+
+@overload
+def read_json(
+    path_or_buf: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+    *,
+    orient: str | None = ...,
+    typ: Literal["frame"] = ...,
+    dtype: DtypeArg | None = ...,
+    convert_axes: bool | None = ...,
+    convert_dates: bool | list[str] = ...,
+    keep_default_dates: bool = ...,
+    precise_float: bool = ...,
+    date_unit: str | None = ...,
+    encoding: str | None = ...,
+    encoding_errors: str | None = ...,
+    lines: bool = ...,
+    chunksize: None = ...,
+    compression: CompressionOptions = ...,
+    nrows: int | None = ...,
+    storage_options: StorageOptions = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    engine: JSONEngine = ...,
+) -> DataFrame: ...
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    decompression_options=_shared_docs["decompression_options"] % "path_or_buf",
+)
+def read_json(
+    path_or_buf: FilePath | ReadBuffer[str] | ReadBuffer[bytes],
+    *,
+    orient: str | None = None,
+    typ: Literal["frame", "series"] = "frame",
+    dtype: DtypeArg | None = None,
+    convert_axes: bool | None = None,
+    convert_dates: bool | list[str] = True,
+    keep_default_dates: bool = True,
+    precise_float: bool = False,
+    date_unit: str | None = None,
+    encoding: str | None = None,
+    encoding_errors: str | None = "strict",
+    lines: bool = False,
+    chunksize: int | None = None,
+    compression: CompressionOptions = "infer",
+    nrows: int | None = None,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    engine: JSONEngine = "ujson",
+) -> DataFrame | Series | JsonReader:
+    """
+    Convert a JSON string to pandas object.
+
+    This method reads JSON files or JSON-like data and converts them into pandas
+    objects. It supports a variety of input formats, including line-delimited JSON,
+    compressed files, and various data representations (table, records, index-based,
+    etc.). When `chunksize` is specified, an iterator is returned instead of loading
+    the entire data into memory.
+
+    Parameters
+    ----------
+    path_or_buf : a str path, path object or file-like object
+        Any valid string path is acceptable. The string could be a URL. Valid
+        URL schemes include http, ftp, s3, and file. For file URLs, a host is
+        expected. A local file could be:
+        ``file://localhost/path/to/table.json``.
+
+        If you want to pass in a path object, pandas accepts any
+        ``os.PathLike``.
+
+        By file-like object, we refer to objects with a ``read()`` method,
+        such as a file handle (e.g. via builtin ``open`` function)
+        or ``StringIO``.
+
+        .. deprecated:: 2.1.0
+            Passing json literal strings is deprecated.
+
+    orient : str, optional
+        Indication of expected JSON string format.
+        Compatible JSON strings can be produced by ``to_json()`` with a
+        corresponding orient value.
+        The set of possible orients is:
+
+        - ``'split'`` : dict like
+          ``{{index -> [index], columns -> [columns], data -> [values]}}``
+        - ``'records'`` : list like
+          ``[{{column -> value}}, ... , {{column -> value}}]``
+        - ``'index'`` : dict like ``{{index -> {{column -> value}}}}``
+        - ``'columns'`` : dict like ``{{column -> {{index -> value}}}}``
+        - ``'values'`` : just the values array
+        - ``'table'`` : dict like ``{{'schema': {{schema}}, 'data': {{data}}}}``
+
+        The allowed and default values depend on the value
+        of the `typ` parameter.
+
+        * when ``typ == 'series'``,
+
+          - allowed orients are ``{{'split','records','index'}}``
+          - default is ``'index'``
+          - The Series index must be unique for orient ``'index'``.
+
+        * when ``typ == 'frame'``,
+
+          - allowed orients are ``{{'split','records','index',
+            'columns','values', 'table'}}``
+          - default is ``'columns'``
+          - The DataFrame index must be unique for orients ``'index'`` and
+            ``'columns'``.
+          - The DataFrame columns must be unique for orients ``'index'``,
+            ``'columns'``, and ``'records'``.
+
+    typ : {{'frame', 'series'}}, default 'frame'
+        The type of object to recover.
+
+    dtype : bool or dict, default None
+        If True, infer dtypes; if a dict of column to dtype, then use those;
+        if False, then don't infer dtypes at all, applies only to the data.
+
+        For all ``orient`` values except ``'table'``, default is True.
+
+    convert_axes : bool, default None
+        Try to convert the axes to the proper dtypes.
+
+        For all ``orient`` values except ``'table'``, default is True.
+
+    convert_dates : bool or list of str, default True
+        If True then default datelike columns may be converted (depending on
+        keep_default_dates).
+        If False, no dates will be converted.
+        If a list of column names, then those columns will be converted and
+        default datelike columns may also be converted (depending on
+        keep_default_dates).
+
+    keep_default_dates : bool, default True
+        If parsing dates (convert_dates is not False), then try to parse the
+        default datelike columns.
+        A column label is datelike if
+
+        * it ends with ``'_at'``,
+
+        * it ends with ``'_time'``,
+
+        * it begins with ``'timestamp'``,
+
+        * it is ``'modified'``, or
+
+        * it is ``'date'``.
+
+    precise_float : bool, default False
+        Set to enable usage of higher precision (strtod) function when
+        decoding string to double values. Default (False) is to use fast but
+        less precise builtin functionality.
+
+    date_unit : str, default None
+        The timestamp unit to detect if converting dates. The default behaviour
+        is to try and detect the correct precision, but if this is not desired
+        then pass one of 's', 'ms', 'us' or 'ns' to force parsing only seconds,
+        milliseconds, microseconds or nanoseconds respectively.
+
+    encoding : str, default is 'utf-8'
+        The encoding to use to decode py3 bytes.
+
+    encoding_errors : str, optional, default "strict"
+        How encoding errors are treated. `List of possible values
+        <https://docs.python.org/3/library/codecs.html#error-handlers>`_ .
+
+        .. versionadded:: 1.3.0
+
+    lines : bool, default False
+        Read the file as a json object per line.
+
+    chunksize : int, optional
+        Return JsonReader object for iteration.
+        See the `line-delimited json docs
+        <https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#line-delimited-json>`_
+        for more information on ``chunksize``.
+        This can only be passed if `lines=True`.
+        If this is None, the file will be read into memory all at once.
+    {decompression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    nrows : int, optional
+        The number of lines from the line-delimited jsonfile that has to be read.
+        This can only be passed if `lines=True`.
+        If this is None, all the rows will be returned.
+
+    {storage_options}
+
+    dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    engine : {{"ujson", "pyarrow"}}, default "ujson"
+        Parser engine to use. The ``"pyarrow"`` engine is only available when
+        ``lines=True``.
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    Series, DataFrame, or pandas.api.typing.JsonReader
+        A JsonReader is returned when ``chunksize`` is not ``0`` or ``None``.
+        Otherwise, the type returned depends on the value of ``typ``.
+
+    See Also
+    --------
+    DataFrame.to_json : Convert a DataFrame to a JSON string.
+    Series.to_json : Convert a Series to a JSON string.
+    json_normalize : Normalize semi-structured JSON data into a flat table.
+
+    Notes
+    -----
+    Specific to ``orient='table'``, if a :class:`DataFrame` with a literal
+    :class:`Index` name of `index` gets written with :func:`to_json`, the
+    subsequent read operation will incorrectly set the :class:`Index` name to
+    ``None``. This is because `index` is also used by :func:`DataFrame.to_json`
+    to denote a missing :class:`Index` name, and the subsequent
+    :func:`read_json` operation cannot distinguish between the two. The same
+    limitation is encountered with a :class:`MultiIndex` and any names
+    beginning with ``'level_'``.
+
+    Examples
+    --------
+    >>> from io import StringIO
+    >>> df = pd.DataFrame([['a', 'b'], ['c', 'd']],
+    ...                   index=['row 1', 'row 2'],
+    ...                   columns=['col 1', 'col 2'])
+
+    Encoding/decoding a Dataframe using ``'split'`` formatted JSON:
+
+    >>> df.to_json(orient='split')
+        '\
+{{\
+"columns":["col 1","col 2"],\
+"index":["row 1","row 2"],\
+"data":[["a","b"],["c","d"]]\
+}}\
+'
+    >>> pd.read_json(StringIO(_), orient='split')  # noqa: F821
+          col 1 col 2
+    row 1     a     b
+    row 2     c     d
+
+    Encoding/decoding a Dataframe using ``'index'`` formatted JSON:
+
+    >>> df.to_json(orient='index')
+    '{{"row 1":{{"col 1":"a","col 2":"b"}},"row 2":{{"col 1":"c","col 2":"d"}}}}'
+
+    >>> pd.read_json(StringIO(_), orient='index')  # noqa: F821
+          col 1 col 2
+    row 1     a     b
+    row 2     c     d
+
+    Encoding/decoding a Dataframe using ``'records'`` formatted JSON.
+    Note that index labels are not preserved with this encoding.
+
+    >>> df.to_json(orient='records')
+    '[{{"col 1":"a","col 2":"b"}},{{"col 1":"c","col 2":"d"}}]'
+    >>> pd.read_json(StringIO(_), orient='records')  # noqa: F821
+      col 1 col 2
+    0     a     b
+    1     c     d
+
+    Encoding with Table Schema
+
+    >>> df.to_json(orient='table')
+        '\
+{{"schema":{{"fields":[\
+{{"name":"index","type":"string"}},\
+{{"name":"col 1","type":"string"}},\
+{{"name":"col 2","type":"string"}}],\
+"primaryKey":["index"],\
+"pandas_version":"1.4.0"}},\
+"data":[\
+{{"index":"row 1","col 1":"a","col 2":"b"}},\
+{{"index":"row 2","col 1":"c","col 2":"d"}}]\
+}}\
+'
+
+    The following example uses ``dtype_backend="numpy_nullable"``
+
+    >>> data = '''{{"index": {{"0": 0, "1": 1}},
+    ...        "a": {{"0": 1, "1": null}},
+    ...        "b": {{"0": 2.5, "1": 4.5}},
+    ...        "c": {{"0": true, "1": false}},
+    ...        "d": {{"0": "a", "1": "b"}},
+    ...        "e": {{"0": 1577.2, "1": 1577.1}}}}'''
+    >>> pd.read_json(StringIO(data), dtype_backend="numpy_nullable")
+       index     a    b      c  d       e
+    0      0     1  2.5   True  a  1577.2
+    1      1  <NA>  4.5  False  b  1577.1
+    """
+    if orient == "table" and dtype:
+        raise ValueError("cannot pass both dtype and orient='table'")
+    if orient == "table" and convert_axes:
+        raise ValueError("cannot pass both convert_axes and orient='table'")
+
+    check_dtype_backend(dtype_backend)
+
+    if dtype is None and orient != "table":
+        # error: Incompatible types in assignment (expression has type "bool", variable
+        # has type "Union[ExtensionDtype, str, dtype[Any], Type[str], Type[float],
+        # Type[int], Type[complex], Type[bool], Type[object], Dict[Hashable,
+        # Union[ExtensionDtype, Union[str, dtype[Any]], Type[str], Type[float],
+        # Type[int], Type[complex], Type[bool], Type[object]]], None]")
+        dtype = True  # type: ignore[assignment]
+    if convert_axes is None and orient != "table":
+        convert_axes = True
+
+    json_reader = JsonReader(
+        path_or_buf,
+        orient=orient,
+        typ=typ,
+        dtype=dtype,
+        convert_axes=convert_axes,
+        convert_dates=convert_dates,
+        keep_default_dates=keep_default_dates,
+        precise_float=precise_float,
+        date_unit=date_unit,
+        encoding=encoding,
+        lines=lines,
+        chunksize=chunksize,
+        compression=compression,
+        nrows=nrows,
+        storage_options=storage_options,
+        encoding_errors=encoding_errors,
+        dtype_backend=dtype_backend,
+        engine=engine,
+    )
+
+    if chunksize:
+        return json_reader
+    else:
+        return json_reader.read()
+
+
+class JsonReader(abc.Iterator, Generic[FrameSeriesStrT]):
+    """
+    JsonReader provides an interface for reading in a JSON file.
+
+    If initialized with ``lines=True`` and ``chunksize``, can be iterated over
+    ``chunksize`` lines at a time. Otherwise, calling ``read`` reads in the
+    whole document.
+    """
+
+    def __init__(
+        self,
+        filepath_or_buffer,
+        orient,
+        typ: FrameSeriesStrT,
+        dtype,
+        convert_axes: bool | None,
+        convert_dates,
+        keep_default_dates: bool,
+        precise_float: bool,
+        date_unit,
+        encoding,
+        lines: bool,
+        chunksize: int | None,
+        compression: CompressionOptions,
+        nrows: int | None,
+        storage_options: StorageOptions | None = None,
+        encoding_errors: str | None = "strict",
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+        engine: JSONEngine = "ujson",
+    ) -> None:
+        self.orient = orient
+        self.typ = typ
+        self.dtype = dtype
+        self.convert_axes = convert_axes
+        self.convert_dates = convert_dates
+        self.keep_default_dates = keep_default_dates
+        self.precise_float = precise_float
+        self.date_unit = date_unit
+        self.encoding = encoding
+        self.engine = engine
+        self.compression = compression
+        self.storage_options = storage_options
+        self.lines = lines
+        self.chunksize = chunksize
+        self.nrows_seen = 0
+        self.nrows = nrows
+        self.encoding_errors = encoding_errors
+        self.handles: IOHandles[str] | None = None
+        self.dtype_backend = dtype_backend
+
+        if self.engine not in {"pyarrow", "ujson"}:
+            raise ValueError(
+                f"The engine type {self.engine} is currently not supported."
+            )
+        if self.chunksize is not None:
+            self.chunksize = validate_integer("chunksize", self.chunksize, 1)
+            if not self.lines:
+                raise ValueError("chunksize can only be passed if lines=True")
+            if self.engine == "pyarrow":
+                raise ValueError(
+                    "currently pyarrow engine doesn't support chunksize parameter"
+                )
+        if self.nrows is not None:
+            self.nrows = validate_integer("nrows", self.nrows, 0)
+            if not self.lines:
+                raise ValueError("nrows can only be passed if lines=True")
+        if self.engine == "pyarrow":
+            if not self.lines:
+                raise ValueError(
+                    "currently pyarrow engine only supports "
+                    "the line-delimited JSON format"
+                )
+            self.data = filepath_or_buffer
+        elif self.engine == "ujson":
+            data = self._get_data_from_filepath(filepath_or_buffer)
+            # If self.chunksize, we prepare the data for the `__next__` method.
+            # Otherwise, we read it into memory for the `read` method.
+            if not (self.chunksize or self.nrows):
+                with self:
+                    self.data = data.read()
+            else:
+                self.data = data
+
+    def _get_data_from_filepath(self, filepath_or_buffer):
+        """
+        The function read_json accepts three input types:
+            1. filepath (string-like)
+            2. file-like object (e.g. open file object, StringIO)
+        """
+        filepath_or_buffer = stringify_path(filepath_or_buffer)
+        try:
+            self.handles = get_handle(
+                filepath_or_buffer,
+                "r",
+                encoding=self.encoding,
+                compression=self.compression,
+                storage_options=self.storage_options,
+                errors=self.encoding_errors,
+            )
+        except OSError as err:
+            raise FileNotFoundError(
+                f"File {filepath_or_buffer} does not exist"
+            ) from err
+        filepath_or_buffer = self.handles.handle
+        return filepath_or_buffer
+
+    def _combine_lines(self, lines) -> str:
+        """
+        Combines a list of JSON objects into one JSON object.
+        """
+        return (
+            f"[{','.join([line for line in (line.strip() for line in lines) if line])}]"
+        )
+
+    @overload
+    def read(self: JsonReader[Literal["frame"]]) -> DataFrame: ...
+
+    @overload
+    def read(self: JsonReader[Literal["series"]]) -> Series: ...
+
+    @overload
+    def read(self: JsonReader[Literal["frame", "series"]]) -> DataFrame | Series: ...
+
+    def read(self) -> DataFrame | Series:
+        """
+        Read the whole JSON input into a pandas object.
+        """
+        obj: DataFrame | Series
+        with self:
+            if self.engine == "pyarrow":
+                pyarrow_json = import_optional_dependency("pyarrow.json")
+                pa_table = pyarrow_json.read_json(self.data)
+                return arrow_table_to_pandas(pa_table, dtype_backend=self.dtype_backend)
+            elif self.engine == "ujson":
+                if self.lines:
+                    if self.chunksize:
+                        obj = concat(self)
+                    elif self.nrows:
+                        lines = list(islice(self.data, self.nrows))
+                        lines_json = self._combine_lines(lines)
+                        obj = self._get_object_parser(lines_json)
+                    else:
+                        data = ensure_str(self.data)
+                        data_lines = data.split("\n")
+                        obj = self._get_object_parser(self._combine_lines(data_lines))
+                else:
+                    obj = self._get_object_parser(self.data)
+                if self.dtype_backend is not lib.no_default:
+                    return obj.convert_dtypes(
+                        infer_objects=False, dtype_backend=self.dtype_backend
+                    )
+                else:
+                    return obj
+
+    def _get_object_parser(self, json: str) -> DataFrame | Series:
+        """
+        Parses a json document into a pandas object.
+        """
+        typ = self.typ
+        dtype = self.dtype
+        kwargs = {
+            "orient": self.orient,
+            "dtype": self.dtype,
+            "convert_axes": self.convert_axes,
+            "convert_dates": self.convert_dates,
+            "keep_default_dates": self.keep_default_dates,
+            "precise_float": self.precise_float,
+            "date_unit": self.date_unit,
+            "dtype_backend": self.dtype_backend,
+        }
+        if typ == "frame":
+            return FrameParser(json, **kwargs).parse()
+        elif typ == "series":
+            if not isinstance(dtype, bool):
+                kwargs["dtype"] = dtype
+            return SeriesParser(json, **kwargs).parse()
+        else:
+            raise ValueError(f"{typ=} must be 'frame' or 'series'.")
+
+    def close(self) -> None:
+        """
+        If we opened a stream earlier, in _get_data_from_filepath, we should
+        close it.
+
+        If an open stream or file was passed, we leave it open.
+        """
+        if self.handles is not None:
+            self.handles.close()
+
+    def __iter__(self) -> Self:
+        return self
+
+    @overload
+    def __next__(self: JsonReader[Literal["frame"]]) -> DataFrame: ...
+
+    @overload
+    def __next__(self: JsonReader[Literal["series"]]) -> Series: ...
+
+    @overload
+    def __next__(
+        self: JsonReader[Literal["frame", "series"]],
+    ) -> DataFrame | Series: ...
+
+    def __next__(self) -> DataFrame | Series:
+        if self.nrows and self.nrows_seen >= self.nrows:
+            self.close()
+            raise StopIteration
+
+        lines = list(islice(self.data, self.chunksize))
+        if not lines:
+            self.close()
+            raise StopIteration
+
+        try:
+            lines_json = self._combine_lines(lines)
+            obj = self._get_object_parser(lines_json)
+
+            # Make sure that the returned objects have the right index.
+            obj.index = range(self.nrows_seen, self.nrows_seen + len(obj))
+            self.nrows_seen += len(obj)
+        except Exception as ex:
+            self.close()
+            raise ex
+
+        if self.dtype_backend is not lib.no_default:
+            return obj.convert_dtypes(
+                infer_objects=False, dtype_backend=self.dtype_backend
+            )
+        else:
+            return obj
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+
+class Parser:
+    _split_keys: tuple[str, ...]
+    _default_orient: str
+
+    _STAMP_UNITS = ("s", "ms", "us", "ns")
+    _MIN_STAMPS = {
+        "s": 31536000,
+        "ms": 31536000000,
+        "us": 31536000000000,
+        "ns": 31536000000000000,
+    }
+    json: str
+
+    def __init__(
+        self,
+        json: str,
+        orient,
+        dtype: DtypeArg | None = None,
+        convert_axes: bool = True,
+        convert_dates: bool | list[str] = True,
+        keep_default_dates: bool = False,
+        precise_float: bool = False,
+        date_unit=None,
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    ) -> None:
+        self.json = json
+
+        if orient is None:
+            orient = self._default_orient
+
+        self.orient = orient
+
+        self.dtype = dtype
+
+        if date_unit is not None:
+            date_unit = date_unit.lower()
+            if date_unit not in self._STAMP_UNITS:
+                raise ValueError(f"date_unit must be one of {self._STAMP_UNITS}")
+            self.min_stamp = self._MIN_STAMPS[date_unit]
+        else:
+            self.min_stamp = self._MIN_STAMPS["s"]
+
+        self.precise_float = precise_float
+        self.convert_axes = convert_axes
+        self.convert_dates = convert_dates
+        self.date_unit = date_unit
+        self.keep_default_dates = keep_default_dates
+        self.dtype_backend = dtype_backend
+
+    @final
+    def check_keys_split(self, decoded: dict) -> None:
+        """
+        Checks that dict has only the appropriate keys for orient='split'.
+        """
+        bad_keys = set(decoded.keys()).difference(set(self._split_keys))
+        if bad_keys:
+            bad_keys_joined = ", ".join(bad_keys)
+            raise ValueError(f"JSON data had unexpected key(s): {bad_keys_joined}")
+
+    @final
+    def parse(self) -> DataFrame | Series:
+        obj = self._parse()
+
+        if self.convert_axes:
+            obj = self._convert_axes(obj)
+        obj = self._try_convert_types(obj)
+        return obj
+
+    def _parse(self) -> DataFrame | Series:
+        raise AbstractMethodError(self)
+
+    @final
+    def _convert_axes(self, obj: DataFrame | Series) -> DataFrame | Series:
+        """
+        Try to convert axes.
+        """
+        for axis_name in obj._AXIS_ORDERS:
+            ax = obj._get_axis(axis_name)
+            ser = Series(ax, dtype=ax.dtype, copy=False)
+            new_ser, result = self._try_convert_data(
+                name=axis_name,
+                data=ser,
+                use_dtypes=False,
+                convert_dates=True,
+                is_axis=True,
+            )
+            if result:
+                new_axis = Index(new_ser, dtype=new_ser.dtype, copy=False)
+                setattr(obj, axis_name, new_axis)
+        return obj
+
+    def _try_convert_types(self, obj):
+        raise AbstractMethodError(self)
+
+    @final
+    def _try_convert_data(
+        self,
+        name: Hashable,
+        data: Series,
+        use_dtypes: bool = True,
+        convert_dates: bool | list[str] = True,
+        is_axis: bool = False,
+    ) -> tuple[Series, bool]:
+        """
+        Try to parse a Series into a column by inferring dtype.
+        """
+        org_data = data
+        # don't try to coerce, unless a force conversion
+        if use_dtypes:
+            if not self.dtype:
+                if all(notna(data)):
+                    return data, False
+
+                filled = data.fillna(np.nan)
+
+                return filled, True
+
+            elif self.dtype is True:
+                pass
+            elif not _should_convert_dates(
+                convert_dates, self.keep_default_dates, name
+            ):
+                # convert_dates takes precedence over columns listed in dtypes
+                dtype = (
+                    self.dtype.get(name) if isinstance(self.dtype, dict) else self.dtype
+                )
+                if dtype is not None:
+                    try:
+                        return data.astype(dtype), True
+                    except (TypeError, ValueError):
+                        return data, False
+
+        if convert_dates:
+            new_data = self._try_convert_to_date(data)
+            if new_data is not data:
+                return new_data, True
+
+        converted = False
+        if self.dtype_backend is not lib.no_default and not is_axis:
+            # Fall through for conversion later on
+            return data, True
+        elif is_string_dtype(data.dtype):
+            # try float
+            try:
+                data = data.astype("float64")
+                converted = True
+            except (TypeError, ValueError):
+                pass
+
+        if data.dtype.kind == "f" and data.dtype != "float64":
+            # coerce floats to 64
+            try:
+                data = data.astype("float64")
+                converted = True
+            except (TypeError, ValueError):
+                pass
+
+        # don't coerce 0-len data
+        if len(data) and data.dtype in ("float", "object"):
+            # coerce ints if we can
+            try:
+                new_data = org_data.astype("int64")
+                if (new_data == data).all():
+                    data = new_data
+                    converted = True
+            except (TypeError, ValueError, OverflowError):
+                pass
+
+        if data.dtype == "int" and data.dtype != "int64":
+            # coerce ints to 64
+            try:
+                data = data.astype("int64")
+                converted = True
+            except (TypeError, ValueError):
+                pass
+
+        # if we have an index, we want to preserve dtypes
+        if name == "index" and len(data):
+            if self.orient == "split":
+                return data, False
+
+        return data, converted
+
+    @final
+    def _try_convert_to_date(self, data: Series) -> Series:
+        """
+        Try to parse a ndarray like into a date column.
+
+        Try to coerce object in epoch/iso formats and integer/float in epoch
+        formats.
+        """
+        # no conversion on empty
+        if not len(data):
+            return data
+
+        new_data = data
+
+        if new_data.dtype == "string":
+            new_data = new_data.astype(object)
+
+        if new_data.dtype == "object":
+            try:
+                new_data = data.astype("int64")
+            except OverflowError:
+                return data
+            except (TypeError, ValueError):
+                pass
+
+        # ignore numbers that are out of range
+        if issubclass(new_data.dtype.type, np.number):
+            in_range = (
+                isna(new_data._values)
+                | (new_data > self.min_stamp)
+                | (new_data._values == iNaT)
+            )
+            if not in_range.all():
+                return data
+
+        date_units = (self.date_unit,) if self.date_unit else self._STAMP_UNITS
+        for date_unit in date_units:
+            try:
+                return to_datetime(new_data, errors="raise", unit=date_unit)
+            except (ValueError, OverflowError, TypeError):
+                continue
+        return data
+
+
+class SeriesParser(Parser):
+    _default_orient = "index"
+    _split_keys = ("name", "index", "data")
+
+    def _parse(self) -> Series:
+        data = ujson_loads(self.json, precise_float=self.precise_float)
+
+        if self.orient == "split":
+            decoded = {str(k): v for k, v in data.items()}
+            self.check_keys_split(decoded)
+            return Series(**decoded)
+        else:
+            return Series(data)
+
+    def _try_convert_types(self, obj: Series) -> Series:
+        obj, _ = self._try_convert_data("data", obj, convert_dates=self.convert_dates)
+        return obj
+
+
+class FrameParser(Parser):
+    _default_orient = "columns"
+    _split_keys = ("columns", "index", "data")
+
+    def _parse(self) -> DataFrame:
+        json = self.json
+        orient = self.orient
+
+        if orient == "split":
+            decoded = {
+                str(k): v
+                for k, v in ujson_loads(json, precise_float=self.precise_float).items()
+            }
+            self.check_keys_split(decoded)
+            orig_names = [
+                (tuple(col) if isinstance(col, list) else col)
+                for col in decoded["columns"]
+            ]
+            decoded["columns"] = dedup_names(
+                orig_names,
+                is_potential_multi_index(orig_names, None),
+            )
+            return DataFrame(dtype=None, **decoded)
+        elif orient == "index":
+            return DataFrame.from_dict(
+                ujson_loads(json, precise_float=self.precise_float),
+                dtype=None,
+                orient="index",
+            )
+        elif orient == "table":
+            return parse_table_schema(json, precise_float=self.precise_float)
+        else:
+            # includes orient == "columns"
+            return DataFrame(
+                ujson_loads(json, precise_float=self.precise_float), dtype=None
+            )
+
+    def _try_convert_types(self, obj: DataFrame) -> DataFrame:
+        arrays = []
+        for col_label, series in obj.items():
+            result, _ = self._try_convert_data(
+                col_label,
+                series,
+                convert_dates=_should_convert_dates(
+                    self.convert_dates,
+                    keep_default_dates=self.keep_default_dates,
+                    col=col_label,
+                ),
+            )
+            arrays.append(result.array)
+        return DataFrame._from_arrays(
+            arrays, obj.columns, obj.index, verify_integrity=False
+        )
+
+
+def _should_convert_dates(
+    convert_dates: bool | list[str],
+    keep_default_dates: bool,
+    col: Hashable,
+) -> bool:
+    """
+    Return bool whether a DataFrame column should be cast to datetime.
+    """
+    if convert_dates is False:
+        # convert_dates=True means follow keep_default_dates
+        return False
+    elif not isinstance(convert_dates, bool) and col in set(convert_dates):
+        return True
+    elif not keep_default_dates:
+        return False
+    elif not isinstance(col, str):
+        return False
+    col_lower = col.lower()
+    if (
+        col_lower.endswith(("_at", "_time"))
+        or col_lower in {"modified", "date", "datetime"}
+        or col_lower.startswith("timestamp")
+    ):
+        return True
+    return False
diff --git a/pandas/pandas/io/json/_normalize.py b/pandas/pandas/io/json/_normalize.py
new file mode 100644
index 0000000000000..45c8876dbe3e5
--- /dev/null
+++ b/pandas/pandas/io/json/_normalize.py
@@ -0,0 +1,604 @@
+# ---------------------------------------------------------------------
+# JSON normalization routines
+from __future__ import annotations
+
+from collections import (
+    abc,
+    defaultdict,
+)
+import copy
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    DefaultDict,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs.writers import convert_json_to_lines
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+    from pandas._typing import (
+        IgnoreRaise,
+        Scalar,
+    )
+
+
+def convert_to_line_delimits(s: str) -> str:
+    """
+    Helper function that converts JSON lists to line delimited JSON.
+    """
+    # Determine we have a JSON list to turn to lines otherwise just return the
+    # json object, only lists can
+    if not s[0] == "[" and s[-1] == "]":
+        return s
+    s = s[1:-1]
+
+    return convert_json_to_lines(s)
+
+
+@overload
+def nested_to_record(
+    ds: dict,
+    prefix: str = ...,
+    sep: str = ...,
+    level: int = ...,
+    max_level: int | None = ...,
+) -> dict[str, Any]: ...
+
+
+@overload
+def nested_to_record(
+    ds: list[dict],
+    prefix: str = ...,
+    sep: str = ...,
+    level: int = ...,
+    max_level: int | None = ...,
+) -> list[dict[str, Any]]: ...
+
+
+def nested_to_record(
+    ds: dict | list[dict],
+    prefix: str = "",
+    sep: str = ".",
+    level: int = 0,
+    max_level: int | None = None,
+) -> dict[str, Any] | list[dict[str, Any]]:
+    """
+    A simplified json_normalize
+
+    Converts a nested dict into a flat dict ("record"), unlike json_normalize,
+    it does not attempt to extract a subset of the data.
+
+    Parameters
+    ----------
+    ds : dict or list of dicts
+    prefix: the prefix, optional, default: ""
+    sep : str, default '.'
+        Nested records will generate names separated by sep,
+        e.g., for sep='.', { 'foo' : { 'bar' : 0 } } -> foo.bar
+    level: int, optional, default: 0
+        The number of levels in the json string.
+
+    max_level: int, optional, default: None
+        The max depth to normalize.
+
+    Returns
+    -------
+    d - dict or list of dicts, matching `ds`
+
+    Examples
+    --------
+    >>> nested_to_record(
+    ...     dict(flat1=1, dict1=dict(c=1, d=2), nested=dict(e=dict(c=1, d=2), d=2))
+    ... )
+    {\
+'flat1': 1, \
+'dict1.c': 1, \
+'dict1.d': 2, \
+'nested.e.c': 1, \
+'nested.e.d': 2, \
+'nested.d': 2\
+}
+    """
+    singleton = False
+    if isinstance(ds, dict):
+        ds = [ds]
+        singleton = True
+    new_ds = []
+    for d in ds:
+        new_d = copy.deepcopy(d)
+        for k, v in d.items():
+            # each key gets renamed with prefix
+            if not isinstance(k, str):
+                k = str(k)
+            if level == 0:
+                newkey = k
+            else:
+                newkey = prefix + sep + k
+
+            # flatten if type is dict and
+            # current dict level  < maximum level provided and
+            # only dicts gets recurse-flattened
+            # only at level>1 do we rename the rest of the keys
+            if not isinstance(v, dict) or (
+                max_level is not None and level >= max_level
+            ):
+                if level != 0:  # so we skip copying for top level, common case
+                    v = new_d.pop(k)
+                    new_d[newkey] = v
+                continue
+
+            v = new_d.pop(k)
+            new_d.update(nested_to_record(v, newkey, sep, level + 1, max_level))
+        new_ds.append(new_d)
+
+    if singleton:
+        return new_ds[0]
+    return new_ds
+
+
+def _normalise_json(
+    data: Any,
+    key_string: str,
+    normalized_dict: dict[str, Any],
+    separator: str,
+) -> dict[str, Any]:
+    """
+    Main recursive function
+    Designed for the most basic use case of pd.json_normalize(data)
+    intended as a performance improvement, see #15621
+
+    Parameters
+    ----------
+    data : Any
+        Type dependent on types contained within nested Json
+    key_string : str
+        New key (with separator(s) in) for data
+    normalized_dict : dict
+        The new normalized/flattened Json dict
+    separator : str, default '.'
+        Nested records will generate names separated by sep,
+        e.g., for sep='.', { 'foo' : { 'bar' : 0 } } -> foo.bar
+    """
+    if isinstance(data, dict):
+        for key, value in data.items():
+            new_key = f"{key_string}{separator}{key}"
+
+            if not key_string:
+                new_key = new_key.removeprefix(separator)
+
+            _normalise_json(
+                data=value,
+                key_string=new_key,
+                normalized_dict=normalized_dict,
+                separator=separator,
+            )
+    else:
+        normalized_dict[key_string] = data
+    return normalized_dict
+
+
+def _normalise_json_ordered(data: dict[str, Any], separator: str) -> dict[str, Any]:
+    """
+    Order the top level keys and then recursively go to depth
+
+    Parameters
+    ----------
+    data : dict or list of dicts
+    separator : str, default '.'
+        Nested records will generate names separated by sep,
+        e.g., for sep='.', { 'foo' : { 'bar' : 0 } } -> foo.bar
+
+    Returns
+    -------
+    dict or list of dicts, matching `normalised_json_object`
+    """
+    top_dict_ = {k: v for k, v in data.items() if not isinstance(v, dict)}
+    nested_dict_ = _normalise_json(
+        data={k: v for k, v in data.items() if isinstance(v, dict)},
+        key_string="",
+        normalized_dict={},
+        separator=separator,
+    )
+    return {**top_dict_, **nested_dict_}
+
+
+def _simple_json_normalize(
+    ds: dict | list[dict],
+    sep: str = ".",
+) -> dict | list[dict] | Any:
+    """
+    A optimized basic json_normalize
+
+    Converts a nested dict into a flat dict ("record"), unlike
+    json_normalize and nested_to_record it doesn't do anything clever.
+    But for the most basic use cases it enhances performance.
+    E.g. pd.json_normalize(data)
+
+    Parameters
+    ----------
+    ds : dict or list of dicts
+    sep : str, default '.'
+        Nested records will generate names separated by sep,
+        e.g., for sep='.', { 'foo' : { 'bar' : 0 } } -> foo.bar
+
+    Returns
+    -------
+    frame : DataFrame
+    d - dict or list of dicts, matching `normalised_json_object`
+
+    Examples
+    --------
+    >>> _simple_json_normalize(
+    ...     {
+    ...         "flat1": 1,
+    ...         "dict1": {"c": 1, "d": 2},
+    ...         "nested": {"e": {"c": 1, "d": 2}, "d": 2},
+    ...     }
+    ... )
+    {\
+'flat1': 1, \
+'dict1.c': 1, \
+'dict1.d': 2, \
+'nested.e.c': 1, \
+'nested.e.d': 2, \
+'nested.d': 2\
+}
+
+    """
+    normalised_json_object = {}
+    # expect a dictionary, as most jsons are. However, lists are perfectly valid
+    if isinstance(ds, dict):
+        normalised_json_object = _normalise_json_ordered(data=ds, separator=sep)
+    elif isinstance(ds, list):
+        normalised_json_list = [_simple_json_normalize(row, sep=sep) for row in ds]
+        return normalised_json_list
+    return normalised_json_object
+
+
+def json_normalize(
+    data: dict | list[dict] | Series,
+    record_path: str | list | None = None,
+    meta: str | list[str | list[str]] | None = None,
+    meta_prefix: str | None = None,
+    record_prefix: str | None = None,
+    errors: IgnoreRaise = "raise",
+    sep: str = ".",
+    max_level: int | None = None,
+) -> DataFrame:
+    """
+    Normalize semi-structured JSON data into a flat table.
+
+    This method is designed to transform semi-structured JSON data, such as nested
+    dictionaries or lists, into a flat table. This is particularly useful when
+    handling JSON-like data structures that contain deeply nested fields.
+
+    Parameters
+    ----------
+    data : dict, list of dicts, or Series of dicts
+        Unserialized JSON objects.
+    record_path : str or list of str, default None
+        Path in each object to list of records. If not passed, data will be
+        assumed to be an array of records.
+    meta : list of paths (str or list of str), default None
+        Fields to use as metadata for each record in resulting table.
+    meta_prefix : str, default None
+        If True, prefix records with dotted path, e.g. foo.bar.field if
+        meta is ['foo', 'bar'].
+    record_prefix : str, default None
+        If True, prefix records with dotted path, e.g. foo.bar.field if
+        path to records is ['foo', 'bar'].
+    errors : {'raise', 'ignore'}, default 'raise'
+        Configures error handling.
+
+        * 'ignore' : will ignore KeyError if keys listed in meta are not
+          always present.
+        * 'raise' : will raise KeyError if keys listed in meta are not
+          always present.
+    sep : str, default '.'
+        Nested records will generate names separated by sep.
+        e.g., for sep='.', {'foo': {'bar': 0}} -> foo.bar.
+    max_level : int, default None
+        Max number of levels(depth of dict) to normalize.
+        if None, normalizes all levels.
+
+    Returns
+    -------
+    DataFrame
+        The normalized data, represented as a pandas DataFrame.
+
+    See Also
+    --------
+    DataFrame : Two-dimensional, size-mutable, potentially heterogeneous tabular data.
+    Series : One-dimensional ndarray with axis labels (including time series).
+
+    Examples
+    --------
+    >>> data = [
+    ...     {"id": 1, "name": {"first": "Coleen", "last": "Volk"}},
+    ...     {"name": {"given": "Mark", "family": "Regner"}},
+    ...     {"id": 2, "name": "Faye Raker"},
+    ... ]
+    >>> pd.json_normalize(data)
+        id name.first name.last name.given name.family        name
+    0  1.0     Coleen      Volk        NaN         NaN         NaN
+    1  NaN        NaN       NaN       Mark      Regner         NaN
+    2  2.0        NaN       NaN        NaN         NaN  Faye Raker
+
+    >>> data = [
+    ...     {
+    ...         "id": 1,
+    ...         "name": "Cole Volk",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ...     {"name": "Mark Reg", "fitness": {"height": 130, "weight": 60}},
+    ...     {
+    ...         "id": 2,
+    ...         "name": "Faye Raker",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ... ]
+    >>> pd.json_normalize(data, max_level=0)
+        id        name                        fitness
+    0  1.0   Cole Volk  {'height': 130, 'weight': 60}
+    1  NaN    Mark Reg  {'height': 130, 'weight': 60}
+    2  2.0  Faye Raker  {'height': 130, 'weight': 60}
+
+    Normalizes nested data up to level 1.
+
+    >>> data = [
+    ...     {
+    ...         "id": 1,
+    ...         "name": "Cole Volk",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ...     {"name": "Mark Reg", "fitness": {"height": 130, "weight": 60}},
+    ...     {
+    ...         "id": 2,
+    ...         "name": "Faye Raker",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ... ]
+    >>> pd.json_normalize(data, max_level=1)
+        id        name  fitness.height  fitness.weight
+    0  1.0   Cole Volk             130              60
+    1  NaN    Mark Reg             130              60
+    2  2.0  Faye Raker             130              60
+
+    >>> data = [
+    ...     {
+    ...         "id": 1,
+    ...         "name": "Cole Volk",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ...     {"name": "Mark Reg", "fitness": {"height": 130, "weight": 60}},
+    ...     {
+    ...         "id": 2,
+    ...         "name": "Faye Raker",
+    ...         "fitness": {"height": 130, "weight": 60},
+    ...     },
+    ... ]
+    >>> series = pd.Series(data, index=pd.Index(["a", "b", "c"]))
+    >>> pd.json_normalize(series)
+        id        name  fitness.height  fitness.weight
+    a  1.0   Cole Volk             130              60
+    b  NaN    Mark Reg             130              60
+    c  2.0  Faye Raker             130              60
+
+    >>> data = [
+    ...     {
+    ...         "state": "Florida",
+    ...         "shortname": "FL",
+    ...         "info": {"governor": "Rick Scott"},
+    ...         "counties": [
+    ...             {"name": "Dade", "population": 12345},
+    ...             {"name": "Broward", "population": 40000},
+    ...             {"name": "Palm Beach", "population": 60000},
+    ...         ],
+    ...     },
+    ...     {
+    ...         "state": "Ohio",
+    ...         "shortname": "OH",
+    ...         "info": {"governor": "John Kasich"},
+    ...         "counties": [
+    ...             {"name": "Summit", "population": 1234},
+    ...             {"name": "Cuyahoga", "population": 1337},
+    ...         ],
+    ...     },
+    ... ]
+    >>> result = pd.json_normalize(
+    ...     data, "counties", ["state", "shortname", ["info", "governor"]]
+    ... )
+    >>> result
+             name  population    state shortname info.governor
+    0        Dade       12345   Florida    FL    Rick Scott
+    1     Broward       40000   Florida    FL    Rick Scott
+    2  Palm Beach       60000   Florida    FL    Rick Scott
+    3      Summit        1234   Ohio       OH    John Kasich
+    4    Cuyahoga        1337   Ohio       OH    John Kasich
+
+    >>> data = {"A": [1, 2]}
+    >>> pd.json_normalize(data, "A", record_prefix="Prefix.")
+        Prefix.0
+    0          1
+    1          2
+
+    Returns normalized data with columns prefixed with the given string.
+    """
+
+    def _pull_field(
+        js: dict[str, Any], spec: list | str, extract_record: bool = False
+    ) -> Scalar | Iterable:
+        """Internal function to pull field"""
+        result = js
+        try:
+            if isinstance(spec, list):
+                for field in spec:
+                    if result is None:
+                        raise KeyError(field)
+                    result = result[field]
+            else:
+                result = result[spec]
+        except KeyError as e:
+            if extract_record:
+                raise KeyError(
+                    f"Key {e} not found. If specifying a record_path, all elements of "
+                    f"data should have the path."
+                ) from e
+            if errors == "ignore":
+                return np.nan
+            else:
+                raise KeyError(
+                    f"Key {e} not found. To replace missing values of {e} with "
+                    f"np.nan, pass in errors='ignore'"
+                ) from e
+
+        return result
+
+    def _pull_records(js: dict[str, Any], spec: list | str) -> list:
+        """
+        Internal function to pull field for records, and similar to
+        _pull_field, but require to return list. And will raise error
+        if has non iterable value.
+        """
+        result = _pull_field(js, spec, extract_record=True)
+
+        # GH 31507 GH 30145, GH 26284 if result is not list, raise TypeError if not
+        # null, otherwise return an empty list
+        if not isinstance(result, list):
+            if pd.isnull(result):
+                result = []
+            else:
+                raise TypeError(
+                    f"Path must contain list or null, "
+                    f"but got {type(result).__name__} at {spec!r}"
+                )
+        return result
+
+    if isinstance(data, Series):
+        index = data.index
+    else:
+        index = None
+
+    if isinstance(data, list) and not data:
+        return DataFrame()
+    elif isinstance(data, dict):
+        # A bit of a hackjob
+        data = [data]
+    elif isinstance(data, abc.Iterable) and not isinstance(data, str):
+        # GH35923 Fix pd.json_normalize to not skip the first element of a
+        # generator input
+        data = list(data)
+    else:
+        raise NotImplementedError
+
+    # check to see if a simple recursive function is possible to
+    # improve performance (see #15621) but only for cases such
+    # as pd.Dataframe(data) or pd.Dataframe(data, sep)
+    if (
+        record_path is None
+        and meta is None
+        and meta_prefix is None
+        and record_prefix is None
+        and max_level is None
+    ):
+        return DataFrame(_simple_json_normalize(data, sep=sep), index=index)
+
+    if record_path is None:
+        if any([isinstance(x, dict) for x in y.values()] for y in data):
+            # naive normalization, this is idempotent for flat records
+            # and potentially will inflate the data considerably for
+            # deeply nested structures:
+            #  {VeryLong: { b: 1,c:2}} -> {VeryLong.b:1 ,VeryLong.c:@}
+            #
+            # TODO: handle record value which are lists, at least error
+            #       reasonably
+            data = nested_to_record(data, sep=sep, max_level=max_level)
+        return DataFrame(data, index=index)
+    elif not isinstance(record_path, list):
+        record_path = [record_path]
+
+    if meta is None:
+        meta = []
+    elif not isinstance(meta, list):
+        meta = [meta]
+
+    _meta = [m if isinstance(m, list) else [m] for m in meta]
+
+    # Disastrously inefficient for now
+    records: list = []
+    lengths = []
+
+    meta_vals: DefaultDict = defaultdict(list)
+    meta_keys = [sep.join(val) for val in _meta]
+
+    def _recursive_extract(data, path, seen_meta, level: int = 0) -> None:
+        if isinstance(data, dict):
+            data = [data]
+        if len(path) > 1:
+            for obj in data:
+                for val, key in zip(_meta, meta_keys):
+                    if level + 1 == len(val):
+                        seen_meta[key] = _pull_field(obj, val[-1])
+
+                _recursive_extract(obj[path[0]], path[1:], seen_meta, level=level + 1)
+        else:
+            for obj in data:
+                recs = _pull_records(obj, path[0])
+                recs = [
+                    nested_to_record(r, sep=sep, max_level=max_level)
+                    if isinstance(r, dict)
+                    else r
+                    for r in recs
+                ]
+
+                # For repeating the metadata later
+                lengths.append(len(recs))
+                for val, key in zip(_meta, meta_keys):
+                    if level + 1 > len(val):
+                        meta_val = seen_meta[key]
+                    else:
+                        meta_val = _pull_field(obj, val[level:])
+                    meta_vals[key].append(meta_val)
+                records.extend(recs)
+
+    _recursive_extract(data, record_path, {}, level=0)
+
+    result = DataFrame(records)
+
+    if record_prefix is not None:
+        result = result.rename(columns=lambda x: f"{record_prefix}{x}")
+
+    # Data types, a problem
+    for k, v in meta_vals.items():
+        if meta_prefix is not None:
+            k = meta_prefix + k
+
+        if k in result:
+            raise ValueError(
+                f"Conflicting metadata name {k}, need distinguishing prefix "
+            )
+        # GH 37782
+
+        values = np.array(v, dtype=object)
+
+        if values.ndim > 1:
+            # GH 37782
+            values = np.empty((len(v),), dtype=object)
+            for i, val in enumerate(v):
+                values[i] = val
+
+        result[k] = values.repeat(lengths)
+    if index is not None:
+        result.index = index.repeat(lengths)
+    return result
diff --git a/pandas/pandas/io/json/_table_schema.py b/pandas/pandas/io/json/_table_schema.py
new file mode 100644
index 0000000000000..7879be18b52c9
--- /dev/null
+++ b/pandas/pandas/io/json/_table_schema.py
@@ -0,0 +1,401 @@
+"""
+Table Schema builders
+
+https://specs.frictionlessdata.io/table-schema/
+"""
+
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+from pandas._libs import lib
+from pandas._libs.json import ujson_loads
+from pandas._libs.tslibs import timezones
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import _registry as registry
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_integer_dtype,
+    is_numeric_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+
+from pandas import DataFrame
+import pandas.core.common as com
+
+from pandas.tseries.frequencies import to_offset
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        DtypeObj,
+        JSONSerializable,
+    )
+
+    from pandas import Series
+    from pandas.core.indexes.multi import MultiIndex
+
+
+TABLE_SCHEMA_VERSION = "1.4.0"
+
+
+def as_json_table_type(x: DtypeObj) -> str:
+    """
+    Convert a NumPy / pandas type to its corresponding json_table.
+
+    Parameters
+    ----------
+    x : np.dtype or ExtensionDtype
+
+    Returns
+    -------
+    str
+        the Table Schema data types
+
+    Notes
+    -----
+    This table shows the relationship between NumPy / pandas dtypes,
+    and Table Schema dtypes.
+
+    ==============  =================
+    Pandas type     Table Schema type
+    ==============  =================
+    int64           integer
+    float64         number
+    bool            boolean
+    datetime64[ns]  datetime
+    timedelta64[ns] duration
+    object          str
+    categorical     any
+    =============== =================
+    """
+    if is_integer_dtype(x):
+        return "integer"
+    elif is_bool_dtype(x):
+        return "boolean"
+    elif is_numeric_dtype(x):
+        return "number"
+    elif lib.is_np_dtype(x, "M") or isinstance(x, (DatetimeTZDtype, PeriodDtype)):
+        return "datetime"
+    elif lib.is_np_dtype(x, "m"):
+        return "duration"
+    elif isinstance(x, ExtensionDtype):
+        return "any"
+    elif is_string_dtype(x):
+        return "string"
+    else:
+        return "any"
+
+
+def set_default_names(data):
+    """Sets index names to 'index' for regular, or 'level_x' for Multi"""
+    if com.all_not_none(*data.index.names):
+        nms = data.index.names
+        if len(nms) == 1 and data.index.name == "index":
+            warnings.warn(
+                "Index name of 'index' is not round-trippable.",
+                stacklevel=find_stack_level(),
+            )
+        elif len(nms) > 1 and any(x.startswith("level_") for x in nms):
+            warnings.warn(
+                "Index names beginning with 'level_' are not round-trippable.",
+                stacklevel=find_stack_level(),
+            )
+        return data
+
+    data = data.copy(deep=False)
+    if data.index.nlevels > 1:
+        data.index.names = com.fill_missing_names(data.index.names)
+    else:
+        data.index.name = data.index.name or "index"
+    return data
+
+
+def convert_pandas_type_to_json_field(arr) -> dict[str, JSONSerializable]:
+    dtype = arr.dtype
+    name: JSONSerializable
+    if arr.name is None:
+        name = "values"
+    else:
+        name = arr.name
+    field: dict[str, JSONSerializable] = {
+        "name": name,
+        "type": as_json_table_type(dtype),
+    }
+
+    if isinstance(dtype, CategoricalDtype):
+        cats = dtype.categories
+        ordered = dtype.ordered
+
+        field["constraints"] = {"enum": list(cats)}
+        field["ordered"] = ordered
+    elif isinstance(dtype, PeriodDtype):
+        field["freq"] = dtype.freq.freqstr
+    elif isinstance(dtype, DatetimeTZDtype):
+        if timezones.is_utc(dtype.tz):
+            field["tz"] = "UTC"
+        else:
+            zone = timezones.get_timezone(dtype.tz)
+            if isinstance(zone, str):
+                field["tz"] = zone
+    elif isinstance(dtype, ExtensionDtype):
+        field["extDtype"] = dtype.name
+    return field
+
+
+def convert_json_field_to_pandas_type(field) -> str | CategoricalDtype:
+    """
+    Converts a JSON field descriptor into its corresponding NumPy / pandas type
+
+    Parameters
+    ----------
+    field
+        A JSON field descriptor
+
+    Returns
+    -------
+    dtype
+
+    Raises
+    ------
+    ValueError
+        If the type of the provided field is unknown or currently unsupported
+
+    Examples
+    --------
+    >>> convert_json_field_to_pandas_type({"name": "an_int", "type": "integer"})
+    'int64'
+
+    >>> convert_json_field_to_pandas_type(
+    ...     {
+    ...         "name": "a_categorical",
+    ...         "type": "any",
+    ...         "constraints": {"enum": ["a", "b", "c"]},
+    ...         "ordered": True,
+    ...     }
+    ... )
+    CategoricalDtype(categories=['a', 'b', 'c'], ordered=True, categories_dtype=object)
+
+    >>> convert_json_field_to_pandas_type({"name": "a_datetime", "type": "datetime"})
+    'datetime64[ns]'
+
+    >>> convert_json_field_to_pandas_type(
+    ...     {"name": "a_datetime_with_tz", "type": "datetime", "tz": "US/Central"}
+    ... )
+    'datetime64[ns, US/Central]'
+    """
+    typ = field["type"]
+    if typ == "string":
+        return "object"
+    elif typ == "integer":
+        return field.get("extDtype", "int64")
+    elif typ == "number":
+        return field.get("extDtype", "float64")
+    elif typ == "boolean":
+        return field.get("extDtype", "bool")
+    elif typ == "duration":
+        return "timedelta64"
+    elif typ == "datetime":
+        if field.get("tz"):
+            return f"datetime64[ns, {field['tz']}]"
+        elif field.get("freq"):
+            # GH#9586 rename frequency M to ME for offsets
+            offset = to_offset(field["freq"])
+            freq = PeriodDtype(offset)._freqstr
+            # GH#47747 using datetime over period to minimize the change surface
+            return f"period[{freq}]"
+        else:
+            return "datetime64[ns]"
+    elif typ == "any":
+        if "constraints" in field and "ordered" in field:
+            return CategoricalDtype(
+                categories=field["constraints"]["enum"], ordered=field["ordered"]
+            )
+        elif "extDtype" in field:
+            return registry.find(field["extDtype"])
+        else:
+            return "object"
+
+    raise ValueError(f"Unsupported or invalid field type: {typ}")
+
+
+def build_table_schema(
+    data: DataFrame | Series,
+    index: bool = True,
+    primary_key: bool | None = None,
+    version: bool = True,
+) -> dict[str, JSONSerializable]:
+    """
+    Create a Table schema from ``data``.
+
+    This method is a utility to generate a JSON-serializable schema
+    representation of a pandas Series or DataFrame, compatible with the
+    Table Schema specification. It enables structured data to be shared
+    and validated in various applications, ensuring consistency and
+    interoperability.
+
+    Parameters
+    ----------
+    data : Series or DataFrame
+        The input data for which the table schema is to be created.
+    index : bool, default True
+        Whether to include ``data.index`` in the schema.
+    primary_key : bool or None, default True
+        Column names to designate as the primary key.
+        The default `None` will set `'primaryKey'` to the index
+        level or levels if the index is unique.
+    version : bool, default True
+        Whether to include a field `pandas_version` with the version
+        of pandas that last revised the table schema. This version
+        can be different from the installed pandas version.
+
+    Returns
+    -------
+    dict
+        A dictionary representing the Table schema.
+
+    See Also
+    --------
+    DataFrame.to_json : Convert the object to a JSON string.
+    read_json : Convert a JSON string to pandas object.
+
+    Notes
+    -----
+    See `Table Schema
+    <https://pandas.pydata.org/docs/user_guide/io.html#table-schema>`__ for
+    conversion types.
+    Timedeltas as converted to ISO8601 duration format with
+    9 decimal places after the seconds field for nanosecond precision.
+
+    Categoricals are converted to the `any` dtype, and use the `enum` field
+    constraint to list the allowed values. The `ordered` attribute is included
+    in an `ordered` field.
+
+    Examples
+    --------
+    >>> from pandas.io.json._table_schema import build_table_schema
+    >>> df = pd.DataFrame(
+    ...     {'A': [1, 2, 3],
+    ...      'B': ['a', 'b', 'c'],
+    ...      'C': pd.date_range('2016-01-01', freq='D', periods=3),
+    ...      }, index=pd.Index(range(3), name='idx'))
+    >>> build_table_schema(df)
+    {'fields': \
+[{'name': 'idx', 'type': 'integer'}, \
+{'name': 'A', 'type': 'integer'}, \
+{'name': 'B', 'type': 'string'}, \
+{'name': 'C', 'type': 'datetime'}], \
+'primaryKey': ['idx'], \
+'pandas_version': '1.4.0'}
+    """
+    if index is True:
+        data = set_default_names(data)
+
+    schema: dict[str, Any] = {}
+    fields = []
+
+    if index:
+        if data.index.nlevels > 1:
+            data.index = cast("MultiIndex", data.index)
+            for level, name in zip(data.index.levels, data.index.names):
+                new_field = convert_pandas_type_to_json_field(level)
+                new_field["name"] = name
+                fields.append(new_field)
+        else:
+            fields.append(convert_pandas_type_to_json_field(data.index))
+
+    if data.ndim > 1:
+        for column, s in data.items():
+            fields.append(convert_pandas_type_to_json_field(s))
+    else:
+        fields.append(convert_pandas_type_to_json_field(data))
+
+    schema["fields"] = fields
+    if index and data.index.is_unique and primary_key is None:
+        if data.index.nlevels == 1:
+            schema["primaryKey"] = [data.index.name]
+        else:
+            schema["primaryKey"] = data.index.names
+    elif primary_key is not None:
+        schema["primaryKey"] = primary_key
+
+    if version:
+        schema["pandas_version"] = TABLE_SCHEMA_VERSION
+    return schema
+
+
+def parse_table_schema(json, precise_float: bool) -> DataFrame:
+    """
+    Builds a DataFrame from a given schema
+
+    Parameters
+    ----------
+    json :
+        A JSON table schema
+    precise_float : bool
+        Flag controlling precision when decoding string to double values, as
+        dictated by ``read_json``
+
+    Returns
+    -------
+    df : DataFrame
+
+    Raises
+    ------
+    NotImplementedError
+        If the JSON table schema contains either timezone or timedelta data
+
+    Notes
+    -----
+        Because :func:`DataFrame.to_json` uses the string 'index' to denote a
+        name-less :class:`Index`, this function sets the name of the returned
+        :class:`DataFrame` to ``None`` when said string is encountered with a
+        normal :class:`Index`. For a :class:`MultiIndex`, the same limitation
+        applies to any strings beginning with 'level_'. Therefore, an
+        :class:`Index` name of 'index'  and :class:`MultiIndex` names starting
+        with 'level_' are not supported.
+
+    See Also
+    --------
+    build_table_schema : Inverse function.
+    pandas.read_json
+    """
+    table = ujson_loads(json, precise_float=precise_float)
+    col_order = [field["name"] for field in table["schema"]["fields"]]
+    df = DataFrame(table["data"], columns=col_order)[col_order]
+
+    dtypes = {
+        field["name"]: convert_json_field_to_pandas_type(field)
+        for field in table["schema"]["fields"]
+    }
+
+    # No ISO constructor for Timedelta as of yet, so need to raise
+    if "timedelta64" in dtypes.values():
+        raise NotImplementedError(
+            'table="orient" can not yet read ISO-formatted Timedelta data'
+        )
+
+    df = df.astype(dtypes)
+
+    if "primaryKey" in table["schema"]:
+        df = df.set_index(table["schema"]["primaryKey"])
+        if len(df.index.names) == 1:
+            if df.index.name == "index":
+                df.index.name = None
+        else:
+            df.index.names = [
+                None if x.startswith("level_") else x for x in df.index.names
+            ]
+
+    return df
diff --git a/pandas/pandas/io/orc.py b/pandas/pandas/io/orc.py
new file mode 100644
index 0000000000000..1a2d564d5b44d
--- /dev/null
+++ b/pandas/pandas/io/orc.py
@@ -0,0 +1,244 @@
+"""orc compat"""
+
+from __future__ import annotations
+
+import io
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+)
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.indexes.api import default_index
+
+from pandas.io._util import arrow_table_to_pandas
+from pandas.io.common import (
+    get_handle,
+    is_fsspec_url,
+)
+
+if TYPE_CHECKING:
+    import fsspec
+    import pyarrow.fs
+
+    from pandas._typing import (
+        DtypeBackend,
+        FilePath,
+        ReadBuffer,
+        WriteBuffer,
+    )
+
+    from pandas.core.frame import DataFrame
+
+
+def read_orc(
+    path: FilePath | ReadBuffer[bytes],
+    columns: list[str] | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    filesystem: pyarrow.fs.FileSystem | fsspec.spec.AbstractFileSystem | None = None,
+    **kwargs: Any,
+) -> DataFrame:
+    """
+    Load an ORC object from the file path, returning a DataFrame.
+
+    This method reads an ORC (Optimized Row Columnar) file into a pandas
+    DataFrame using the `pyarrow.orc` library. ORC is a columnar storage format
+    that provides efficient compression and fast retrieval for analytical workloads.
+    It allows reading specific columns, handling different filesystem
+    types (such as local storage, cloud storage via fsspec, or pyarrow filesystem),
+    and supports different data type backends, including `numpy_nullable` and `pyarrow`.
+
+    Parameters
+    ----------
+    path : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``read()`` function. The string could be a URL.
+        Valid URL schemes include http, ftp, s3, and file. For file URLs, a host is
+        expected. A local file could be:
+        ``file://localhost/path/to/table.orc``.
+    columns : list, default None
+        If not None, only these columns will be read from the file.
+        Output always follows the ordering of the file and not the columns list.
+        This mirrors the original behaviour of
+        :external+pyarrow:py:meth:`pyarrow.orc.ORCFile.read`.
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    filesystem : fsspec or pyarrow filesystem, default None
+        Filesystem object to use when reading the orc file.
+
+        .. versionadded:: 2.1.0
+
+    **kwargs
+        Any additional kwargs are passed to pyarrow.
+
+    Returns
+    -------
+    DataFrame
+        DataFrame based on the ORC file.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (csv) file into a pandas DataFrame.
+    read_excel : Read an Excel file into a pandas DataFrame.
+    read_spss : Read an SPSS file into a pandas DataFrame.
+    read_sas : Load a SAS file into a pandas DataFrame.
+    read_feather : Load a feather-format object into a pandas DataFrame.
+
+    Notes
+    -----
+    Before using this function you should read the :ref:`user guide about ORC <io.orc>`
+    and :ref:`install optional dependencies <install.warn_orc>`.
+
+    If ``path`` is a URI scheme pointing to a local or remote file (e.g. "s3://"),
+    a ``pyarrow.fs`` filesystem will be attempted to read the file. You can also pass a
+    pyarrow or fsspec filesystem object into the filesystem keyword to override this
+    behavior.
+
+    Examples
+    --------
+    >>> result = pd.read_orc("example_pa.orc")  # doctest: +SKIP
+    """
+    # we require a newer version of pyarrow than we support for orc
+
+    orc = import_optional_dependency("pyarrow.orc")
+
+    check_dtype_backend(dtype_backend)
+
+    with get_handle(path, "rb", is_text=False) as handles:
+        source = handles.handle
+        if is_fsspec_url(path) and filesystem is None:
+            pa = import_optional_dependency("pyarrow")
+            pa_fs = import_optional_dependency("pyarrow.fs")
+            try:
+                filesystem, source = pa_fs.FileSystem.from_uri(path)
+            except (TypeError, pa.ArrowInvalid):
+                pass
+
+        pa_table = orc.read_table(
+            source=source, columns=columns, filesystem=filesystem, **kwargs
+        )
+    return arrow_table_to_pandas(pa_table, dtype_backend=dtype_backend)
+
+
+def to_orc(
+    df: DataFrame,
+    path: FilePath | WriteBuffer[bytes] | None = None,
+    *,
+    engine: Literal["pyarrow"] = "pyarrow",
+    index: bool | None = None,
+    engine_kwargs: dict[str, Any] | None = None,
+) -> bytes | None:
+    """
+    Write a DataFrame to the ORC format.
+
+    .. versionadded:: 1.5.0
+
+    Parameters
+    ----------
+    df : DataFrame
+        The dataframe to be written to ORC. Raises NotImplementedError
+        if dtype of one or more columns is category, unsigned integers,
+        intervals, periods or sparse.
+    path : str, file-like object or None, default None
+        If a string, it will be used as Root Directory path
+        when writing a partitioned dataset. By file-like object,
+        we refer to objects with a write() method, such as a file handle
+        (e.g. via builtin open function). If path is None,
+        a bytes object is returned.
+    engine : str, default 'pyarrow'
+        ORC library to use.
+    index : bool, optional
+        If ``True``, include the dataframe's index(es) in the file output. If
+        ``False``, they will not be written to the file.
+        If ``None``, similar to ``infer`` the dataframe's index(es)
+        will be saved. However, instead of being saved as values,
+        the RangeIndex will be stored as a range in the metadata so it
+        doesn't require much space and is faster. Other indexes will
+        be included as columns in the file output.
+    engine_kwargs : dict[str, Any] or None, default None
+        Additional keyword arguments passed to :func:`pyarrow.orc.write_table`.
+
+    Returns
+    -------
+    bytes if no path argument is provided else None
+
+    Raises
+    ------
+    NotImplementedError
+        Dtype of one or more columns is category, unsigned integers, interval,
+        period or sparse.
+    ValueError
+        engine is not pyarrow.
+
+    Notes
+    -----
+    * Before using this function you should read the
+      :ref:`user guide about ORC <io.orc>` and
+      :ref:`install optional dependencies <install.warn_orc>`.
+    * This function requires `pyarrow <https://arrow.apache.org/docs/python/>`_
+      library.
+    * For supported dtypes please refer to `supported ORC features in Arrow
+      <https://arrow.apache.org/docs/cpp/orc.html#data-types>`__.
+    * Currently timezones in datetime columns are not preserved when a
+      dataframe is converted into ORC files.
+    """
+    if index is None:
+        index = df.index.names[0] is not None
+    if engine_kwargs is None:
+        engine_kwargs = {}
+
+    # validate index
+    # --------------
+
+    # validate that we have only a default index
+    # raise on anything else as we don't serialize the index
+
+    if not df.index.equals(default_index(len(df))):
+        raise ValueError(
+            "orc does not support serializing a non-default index for the index; "
+            "you can .reset_index() to make the index into column(s)"
+        )
+
+    if df.index.name is not None:
+        raise ValueError("orc does not serialize index meta-data on a default index")
+
+    if engine != "pyarrow":
+        raise ValueError("engine must be 'pyarrow'")
+    pyarrow = import_optional_dependency(engine, min_version="10.0.1")
+    pa = import_optional_dependency("pyarrow")
+    orc = import_optional_dependency("pyarrow.orc")
+
+    was_none = path is None
+    if was_none:
+        path = io.BytesIO()
+    assert path is not None  # For mypy
+    with get_handle(path, "wb", is_text=False) as handles:
+        try:
+            orc.write_table(
+                pyarrow.Table.from_pandas(df, preserve_index=index),
+                handles.handle,
+                **engine_kwargs,
+            )
+        except (TypeError, pa.ArrowNotImplementedError) as e:
+            raise NotImplementedError(
+                "The dtype of one or more columns is not supported yet."
+            ) from e
+
+    if was_none:
+        assert isinstance(path, io.BytesIO)  # For mypy
+        return path.getvalue()
+    return None
diff --git a/pandas/pandas/io/parquet.py b/pandas/pandas/io/parquet.py
new file mode 100644
index 0000000000000..6a5a83088e986
--- /dev/null
+++ b/pandas/pandas/io/parquet.py
@@ -0,0 +1,658 @@
+"""parquet compat"""
+
+from __future__ import annotations
+
+import io
+import json
+import os
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+)
+from warnings import (
+    catch_warnings,
+    filterwarnings,
+)
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import doc
+from pandas.util._validators import check_dtype_backend
+
+from pandas import (
+    DataFrame,
+    get_option,
+)
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io._util import arrow_table_to_pandas
+from pandas.io.common import (
+    IOHandles,
+    get_handle,
+    is_fsspec_url,
+    is_url,
+    stringify_path,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        DtypeBackend,
+        FilePath,
+        ReadBuffer,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+
+def get_engine(engine: str) -> BaseImpl:
+    """return our implementation"""
+    if engine == "auto":
+        engine = get_option("io.parquet.engine")
+
+    if engine == "auto":
+        # try engines in this order
+        engine_classes = [PyArrowImpl, FastParquetImpl]
+
+        error_msgs = ""
+        for engine_class in engine_classes:
+            try:
+                return engine_class()
+            except ImportError as err:
+                error_msgs += "\n - " + str(err)
+
+        raise ImportError(
+            "Unable to find a usable engine; "
+            "tried using: 'pyarrow', 'fastparquet'.\n"
+            "A suitable version of "
+            "pyarrow or fastparquet is required for parquet "
+            "support.\n"
+            "Trying to import the above resulted in these errors:"
+            f"{error_msgs}"
+        )
+
+    if engine == "pyarrow":
+        return PyArrowImpl()
+    elif engine == "fastparquet":
+        return FastParquetImpl()
+
+    raise ValueError("engine must be one of 'pyarrow', 'fastparquet'")
+
+
+def _get_path_or_handle(
+    path: FilePath | ReadBuffer[bytes] | WriteBuffer[bytes],
+    fs: Any,
+    storage_options: StorageOptions | None = None,
+    mode: str = "rb",
+    is_dir: bool = False,
+) -> tuple[
+    FilePath | ReadBuffer[bytes] | WriteBuffer[bytes], IOHandles[bytes] | None, Any
+]:
+    """File handling for PyArrow."""
+    path_or_handle = stringify_path(path)
+    if fs is not None:
+        pa_fs = import_optional_dependency("pyarrow.fs", errors="ignore")
+        fsspec = import_optional_dependency("fsspec", errors="ignore")
+        if pa_fs is not None and isinstance(fs, pa_fs.FileSystem):
+            if storage_options:
+                raise NotImplementedError(
+                    "storage_options not supported with a pyarrow FileSystem."
+                )
+        elif fsspec is not None and isinstance(fs, fsspec.spec.AbstractFileSystem):
+            pass
+        else:
+            raise ValueError(
+                f"filesystem must be a pyarrow or fsspec FileSystem, "
+                f"not a {type(fs).__name__}"
+            )
+    if is_fsspec_url(path_or_handle) and fs is None:
+        if storage_options is None:
+            pa = import_optional_dependency("pyarrow")
+            pa_fs = import_optional_dependency("pyarrow.fs")
+
+            try:
+                fs, path_or_handle = pa_fs.FileSystem.from_uri(path)
+            except (TypeError, pa.ArrowInvalid):
+                pass
+        if fs is None:
+            fsspec = import_optional_dependency("fsspec")
+            fs, path_or_handle = fsspec.core.url_to_fs(
+                path_or_handle, **(storage_options or {})
+            )
+    elif storage_options and (not is_url(path_or_handle) or mode != "rb"):
+        # can't write to a remote url
+        # without making use of fsspec at the moment
+        raise ValueError("storage_options passed with buffer, or non-supported URL")
+
+    handles = None
+    if (
+        not fs
+        and not is_dir
+        and isinstance(path_or_handle, str)
+        and not os.path.isdir(path_or_handle)
+    ):
+        # use get_handle only when we are very certain that it is not a directory
+        # fsspec resources can also point to directories
+        # this branch is used for example when reading from non-fsspec URLs
+        handles = get_handle(
+            path_or_handle, mode, is_text=False, storage_options=storage_options
+        )
+        fs = None
+        path_or_handle = handles.handle
+    return path_or_handle, handles, fs
+
+
+class BaseImpl:
+    @staticmethod
+    def validate_dataframe(df: DataFrame) -> None:
+        if not isinstance(df, DataFrame):
+            raise ValueError("to_parquet only supports IO with DataFrames")
+
+    def write(self, df: DataFrame, path, compression, **kwargs) -> None:
+        raise AbstractMethodError(self)
+
+    def read(self, path, columns=None, **kwargs) -> DataFrame:
+        raise AbstractMethodError(self)
+
+
+class PyArrowImpl(BaseImpl):
+    def __init__(self) -> None:
+        import_optional_dependency(
+            "pyarrow", extra="pyarrow is required for parquet support."
+        )
+        import pyarrow.parquet
+
+        # import utils to register the pyarrow extension types
+        import pandas.core.arrays.arrow.extension_types  # pyright: ignore[reportUnusedImport] # noqa: F401
+
+        self.api = pyarrow
+
+    def write(
+        self,
+        df: DataFrame,
+        path: FilePath | WriteBuffer[bytes],
+        compression: str | None = "snappy",
+        index: bool | None = None,
+        storage_options: StorageOptions | None = None,
+        partition_cols: list[str] | None = None,
+        filesystem=None,
+        **kwargs,
+    ) -> None:
+        self.validate_dataframe(df)
+
+        from_pandas_kwargs: dict[str, Any] = {"schema": kwargs.pop("schema", None)}
+        if index is not None:
+            from_pandas_kwargs["preserve_index"] = index
+
+        table = self.api.Table.from_pandas(df, **from_pandas_kwargs)
+
+        if df.attrs:
+            df_metadata = {"PANDAS_ATTRS": json.dumps(df.attrs)}
+            existing_metadata = table.schema.metadata
+            merged_metadata = {**existing_metadata, **df_metadata}
+            table = table.replace_schema_metadata(merged_metadata)
+
+        path_or_handle, handles, filesystem = _get_path_or_handle(
+            path,
+            filesystem,
+            storage_options=storage_options,
+            mode="wb",
+            is_dir=partition_cols is not None,
+        )
+        if (
+            isinstance(path_or_handle, io.BufferedWriter)
+            and hasattr(path_or_handle, "name")
+            and isinstance(path_or_handle.name, (str, bytes))
+        ):
+            if isinstance(path_or_handle.name, bytes):
+                path_or_handle = path_or_handle.name.decode()
+            else:
+                path_or_handle = path_or_handle.name
+
+        try:
+            if partition_cols is not None:
+                # writes to multiple files under the given path
+                self.api.parquet.write_to_dataset(
+                    table,
+                    path_or_handle,
+                    compression=compression,
+                    partition_cols=partition_cols,
+                    filesystem=filesystem,
+                    **kwargs,
+                )
+            else:
+                # write to single output file
+                self.api.parquet.write_table(
+                    table,
+                    path_or_handle,
+                    compression=compression,
+                    filesystem=filesystem,
+                    **kwargs,
+                )
+        finally:
+            if handles is not None:
+                handles.close()
+
+    def read(
+        self,
+        path,
+        columns=None,
+        filters=None,
+        dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+        storage_options: StorageOptions | None = None,
+        filesystem=None,
+        to_pandas_kwargs: dict[str, Any] | None = None,
+        **kwargs,
+    ) -> DataFrame:
+        kwargs["use_pandas_metadata"] = True
+
+        path_or_handle, handles, filesystem = _get_path_or_handle(
+            path,
+            filesystem,
+            storage_options=storage_options,
+            mode="rb",
+        )
+        try:
+            pa_table = self.api.parquet.read_table(
+                path_or_handle,
+                columns=columns,
+                filesystem=filesystem,
+                filters=filters,
+                **kwargs,
+            )
+            with catch_warnings():
+                filterwarnings(
+                    "ignore",
+                    "make_block is deprecated",
+                    DeprecationWarning,
+                )
+                result = arrow_table_to_pandas(
+                    pa_table,
+                    dtype_backend=dtype_backend,
+                    to_pandas_kwargs=to_pandas_kwargs,
+                )
+
+            if pa_table.schema.metadata:
+                if b"PANDAS_ATTRS" in pa_table.schema.metadata:
+                    df_metadata = pa_table.schema.metadata[b"PANDAS_ATTRS"]
+                    result.attrs = json.loads(df_metadata)
+            return result
+        finally:
+            if handles is not None:
+                handles.close()
+
+
+class FastParquetImpl(BaseImpl):
+    def __init__(self) -> None:
+        # since pandas is a dependency of fastparquet
+        # we need to import on first use
+        fastparquet = import_optional_dependency(
+            "fastparquet", extra="fastparquet is required for parquet support."
+        )
+        self.api = fastparquet
+
+    def write(
+        self,
+        df: DataFrame,
+        path,
+        compression: Literal["snappy", "gzip", "brotli"] | None = "snappy",
+        index=None,
+        partition_cols=None,
+        storage_options: StorageOptions | None = None,
+        filesystem=None,
+        **kwargs,
+    ) -> None:
+        self.validate_dataframe(df)
+
+        if "partition_on" in kwargs and partition_cols is not None:
+            raise ValueError(
+                "Cannot use both partition_on and "
+                "partition_cols. Use partition_cols for partitioning data"
+            )
+        if "partition_on" in kwargs:
+            partition_cols = kwargs.pop("partition_on")
+
+        if partition_cols is not None:
+            kwargs["file_scheme"] = "hive"
+
+        if filesystem is not None:
+            raise NotImplementedError(
+                "filesystem is not implemented for the fastparquet engine."
+            )
+
+        # cannot use get_handle as write() does not accept file buffers
+        path = stringify_path(path)
+        if is_fsspec_url(path):
+            fsspec = import_optional_dependency("fsspec")
+
+            # if filesystem is provided by fsspec, file must be opened in 'wb' mode.
+            kwargs["open_with"] = lambda path, _: fsspec.open(
+                path, "wb", **(storage_options or {})
+            ).open()
+        elif storage_options:
+            raise ValueError(
+                "storage_options passed with file object or non-fsspec file path"
+            )
+
+        with catch_warnings(record=True):
+            self.api.write(
+                path,
+                df,
+                compression=compression,
+                write_index=index,
+                partition_on=partition_cols,
+                **kwargs,
+            )
+
+    def read(
+        self,
+        path,
+        columns=None,
+        filters=None,
+        storage_options: StorageOptions | None = None,
+        filesystem=None,
+        to_pandas_kwargs: dict | None = None,
+        **kwargs,
+    ) -> DataFrame:
+        parquet_kwargs: dict[str, Any] = {}
+        dtype_backend = kwargs.pop("dtype_backend", lib.no_default)
+        # We are disabling nullable dtypes for fastparquet pending discussion
+        parquet_kwargs["pandas_nulls"] = False
+        if dtype_backend is not lib.no_default:
+            raise ValueError(
+                "The 'dtype_backend' argument is not supported for the "
+                "fastparquet engine"
+            )
+        if filesystem is not None:
+            raise NotImplementedError(
+                "filesystem is not implemented for the fastparquet engine."
+            )
+        if to_pandas_kwargs is not None:
+            raise NotImplementedError(
+                "to_pandas_kwargs is not implemented for the fastparquet engine."
+            )
+        path = stringify_path(path)
+        handles = None
+        if is_fsspec_url(path):
+            fsspec = import_optional_dependency("fsspec")
+
+            parquet_kwargs["fs"] = fsspec.open(path, "rb", **(storage_options or {})).fs
+        elif isinstance(path, str) and not os.path.isdir(path):
+            # use get_handle only when we are very certain that it is not a directory
+            # fsspec resources can also point to directories
+            # this branch is used for example when reading from non-fsspec URLs
+            handles = get_handle(
+                path, "rb", is_text=False, storage_options=storage_options
+            )
+            path = handles.handle
+
+        try:
+            parquet_file = self.api.ParquetFile(path, **parquet_kwargs)
+            with catch_warnings():
+                filterwarnings(
+                    "ignore",
+                    "make_block is deprecated",
+                    DeprecationWarning,
+                )
+                return parquet_file.to_pandas(
+                    columns=columns, filters=filters, **kwargs
+                )
+        finally:
+            if handles is not None:
+                handles.close()
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def to_parquet(
+    df: DataFrame,
+    path: FilePath | WriteBuffer[bytes] | None = None,
+    engine: str = "auto",
+    compression: str | None = "snappy",
+    index: bool | None = None,
+    storage_options: StorageOptions | None = None,
+    partition_cols: list[str] | None = None,
+    filesystem: Any = None,
+    **kwargs,
+) -> bytes | None:
+    """
+    Write a DataFrame to the parquet format.
+
+    Parameters
+    ----------
+    df : DataFrame
+    path : str, path object, file-like object, or None, default None
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``write()`` function. If None, the result is
+        returned as bytes. If a string, it will be used as Root Directory path
+        when writing a partitioned dataset. The engine fastparquet does not
+        accept file-like objects.
+    engine : {{'auto', 'pyarrow', 'fastparquet'}}, default 'auto'
+        Parquet library to use. If 'auto', then the option
+        ``io.parquet.engine`` is used. The default ``io.parquet.engine``
+        behavior is to try 'pyarrow', falling back to 'fastparquet' if
+        'pyarrow' is unavailable.
+
+        When using the ``'pyarrow'`` engine and no storage options are provided
+        and a filesystem is implemented by both ``pyarrow.fs`` and ``fsspec``
+        (e.g. "s3://"), then the ``pyarrow.fs`` filesystem is attempted first.
+        Use the filesystem keyword with an instantiated fsspec filesystem
+        if you wish to use its implementation.
+    compression : {{'snappy', 'gzip', 'brotli', 'lz4', 'zstd', None}},
+        default 'snappy'. Name of the compression to use. Use ``None``
+        for no compression.
+    index : bool, default None
+        If ``True``, include the dataframe's index(es) in the file output. If
+        ``False``, they will not be written to the file.
+        If ``None``, similar to ``True`` the dataframe's index(es)
+        will be saved. However, instead of being saved as values,
+        the RangeIndex will be stored as a range in the metadata so it
+        doesn't require much space and is faster. Other indexes will
+        be included as columns in the file output.
+    partition_cols : str or list, optional, default None
+        Column names by which to partition the dataset.
+        Columns are partitioned in the order they are given.
+        Must be None if path is not a string.
+    {storage_options}
+
+    filesystem : fsspec or pyarrow filesystem, default None
+        Filesystem object to use when reading the parquet file. Only implemented
+        for ``engine="pyarrow"``.
+
+        .. versionadded:: 2.1.0
+
+    kwargs
+        Additional keyword arguments passed to the engine.
+
+    Returns
+    -------
+    bytes if no path argument is provided else None
+    """
+    if isinstance(partition_cols, str):
+        partition_cols = [partition_cols]
+    impl = get_engine(engine)
+
+    path_or_buf: FilePath | WriteBuffer[bytes] = io.BytesIO() if path is None else path
+
+    impl.write(
+        df,
+        path_or_buf,
+        compression=compression,
+        index=index,
+        partition_cols=partition_cols,
+        storage_options=storage_options,
+        filesystem=filesystem,
+        **kwargs,
+    )
+
+    if path is None:
+        assert isinstance(path_or_buf, io.BytesIO)
+        return path_or_buf.getvalue()
+    else:
+        return None
+
+
+@doc(storage_options=_shared_docs["storage_options"])
+def read_parquet(
+    path: FilePath | ReadBuffer[bytes],
+    engine: str = "auto",
+    columns: list[str] | None = None,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    filesystem: Any = None,
+    filters: list[tuple] | list[list[tuple]] | None = None,
+    to_pandas_kwargs: dict | None = None,
+    **kwargs,
+) -> DataFrame:
+    """
+    Load a parquet object from the file path, returning a DataFrame.
+
+    The function automatically handles reading the data from a parquet file
+    and creates a DataFrame with the appropriate structure.
+
+    Parameters
+    ----------
+    path : str, path object or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``read()`` function.
+        The string could be a URL. Valid URL schemes include http, ftp, s3,
+        gs, and file. For file URLs, a host is expected. A local file could be:
+        ``file://localhost/path/to/table.parquet``.
+        A file URL can also be a path to a directory that contains multiple
+        partitioned parquet files. Both pyarrow and fastparquet support
+        paths to directories as well as file URLs. A directory path could be:
+        ``file://localhost/path/to/tables`` or ``s3://bucket/partition_dir``.
+    engine : {{'auto', 'pyarrow', 'fastparquet'}}, default 'auto'
+        Parquet library to use. If 'auto', then the option
+        ``io.parquet.engine`` is used. The default ``io.parquet.engine``
+        behavior is to try 'pyarrow', falling back to 'fastparquet' if
+        'pyarrow' is unavailable.
+
+        When using the ``'pyarrow'`` engine and no storage options are provided
+        and a filesystem is implemented by both ``pyarrow.fs`` and ``fsspec``
+        (e.g. "s3://"), then the ``pyarrow.fs`` filesystem is attempted first.
+        Use the filesystem keyword with an instantiated fsspec filesystem
+        if you wish to use its implementation.
+    columns : list, default=None
+        If not None, only these columns will be read from the file.
+    {storage_options}
+
+        .. versionadded:: 1.3.0
+
+    dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    filesystem : fsspec or pyarrow filesystem, default None
+        Filesystem object to use when reading the parquet file. Only implemented
+        for ``engine="pyarrow"``.
+
+        .. versionadded:: 2.1.0
+
+    filters : List[Tuple] or List[List[Tuple]], default None
+        To filter out data.
+        Filter syntax: [[(column, op, val), ...],...]
+        where op is [==, =, >, >=, <, <=, !=, in, not in]
+        The innermost tuples are transposed into a set of filters applied
+        through an `AND` operation.
+        The outer list combines these sets of filters through an `OR`
+        operation.
+        A single list of tuples can also be used, meaning that no `OR`
+        operation between set of filters is to be conducted.
+
+        Using this argument will NOT result in row-wise filtering of the final
+        partitions unless ``engine="pyarrow"`` is also specified.  For
+        other engines, filtering is only performed at the partition level, that is,
+        to prevent the loading of some row-groups and/or files.
+
+        .. versionadded:: 2.1.0
+
+    to_pandas_kwargs : dict | None, default None
+        Keyword arguments to pass through to :func:`pyarrow.Table.to_pandas`
+        when ``engine="pyarrow"``.
+
+        .. versionadded:: 3.0.0
+
+    **kwargs
+        Any additional kwargs are passed to the engine.
+
+    Returns
+    -------
+    DataFrame
+        DataFrame based on parquet file.
+
+    See Also
+    --------
+    DataFrame.to_parquet : Create a parquet object that serializes a DataFrame.
+
+    Examples
+    --------
+    >>> original_df = pd.DataFrame({{"foo": range(5), "bar": range(5, 10)}})
+    >>> original_df
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    >>> df_parquet_bytes = original_df.to_parquet()
+    >>> from io import BytesIO
+    >>> restored_df = pd.read_parquet(BytesIO(df_parquet_bytes))
+    >>> restored_df
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    >>> restored_df.equals(original_df)
+    True
+    >>> restored_bar = pd.read_parquet(BytesIO(df_parquet_bytes), columns=["bar"])
+    >>> restored_bar
+        bar
+    0    5
+    1    6
+    2    7
+    3    8
+    4    9
+    >>> restored_bar.equals(original_df[["bar"]])
+    True
+
+    The function uses `kwargs` that are passed directly to the engine.
+    In the following example, we use the `filters` argument of the pyarrow
+    engine to filter the rows of the DataFrame.
+
+    Since `pyarrow` is the default engine, we can omit the `engine` argument.
+    Note that the `filters` argument is implemented by the `pyarrow` engine,
+    which can benefit from multithreading and also potentially be more
+    economical in terms of memory.
+
+    >>> sel = [("foo", ">", 2)]
+    >>> restored_part = pd.read_parquet(BytesIO(df_parquet_bytes), filters=sel)
+    >>> restored_part
+        foo  bar
+    0    3    8
+    1    4    9
+    """
+
+    impl = get_engine(engine)
+    check_dtype_backend(dtype_backend)
+
+    return impl.read(
+        path,
+        columns=columns,
+        filters=filters,
+        storage_options=storage_options,
+        dtype_backend=dtype_backend,
+        filesystem=filesystem,
+        to_pandas_kwargs=to_pandas_kwargs,
+        **kwargs,
+    )
diff --git a/pandas/pandas/io/parsers/__init__.py b/pandas/pandas/io/parsers/__init__.py
new file mode 100644
index 0000000000000..ff11968db15f0
--- /dev/null
+++ b/pandas/pandas/io/parsers/__init__.py
@@ -0,0 +1,9 @@
+from pandas.io.parsers.readers import (
+    TextFileReader,
+    TextParser,
+    read_csv,
+    read_fwf,
+    read_table,
+)
+
+__all__ = ["TextFileReader", "TextParser", "read_csv", "read_fwf", "read_table"]
diff --git a/pandas/pandas/io/parsers/arrow_parser_wrapper.py b/pandas/pandas/io/parsers/arrow_parser_wrapper.py
new file mode 100644
index 0000000000000..8cadde1ad6537
--- /dev/null
+++ b/pandas/pandas/io/parsers/arrow_parser_wrapper.py
@@ -0,0 +1,295 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+import warnings
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    ParserError,
+    ParserWarning,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.inference import is_integer
+
+from pandas.io._util import arrow_table_to_pandas
+from pandas.io.parsers.base_parser import ParserBase
+
+if TYPE_CHECKING:
+    from pandas._typing import ReadBuffer
+
+    from pandas import DataFrame
+
+
+class ArrowParserWrapper(ParserBase):
+    """
+    Wrapper for the pyarrow engine for read_csv()
+    """
+
+    def __init__(self, src: ReadBuffer[bytes], **kwds) -> None:
+        super().__init__(kwds)
+        self.kwds = kwds
+        self.src = src
+
+        self._parse_kwds()
+
+    def _parse_kwds(self) -> None:
+        """
+        Validates keywords before passing to pyarrow.
+        """
+        encoding: str | None = self.kwds.get("encoding")
+        self.encoding = "utf-8" if encoding is None else encoding
+
+        na_values = self.kwds["na_values"]
+        if isinstance(na_values, dict):
+            raise ValueError(
+                "The pyarrow engine doesn't support passing a dict for na_values"
+            )
+        self.na_values = list(self.kwds["na_values"])
+
+    def _get_pyarrow_options(self) -> None:
+        """
+        Rename some arguments to pass to pyarrow
+        """
+        mapping = {
+            "usecols": "include_columns",
+            "na_values": "null_values",
+            "escapechar": "escape_char",
+            "skip_blank_lines": "ignore_empty_lines",
+            "decimal": "decimal_point",
+            "quotechar": "quote_char",
+        }
+        for pandas_name, pyarrow_name in mapping.items():
+            if pandas_name in self.kwds and self.kwds.get(pandas_name) is not None:
+                self.kwds[pyarrow_name] = self.kwds.pop(pandas_name)
+
+        # Date format handling
+        # If we get a string, we need to convert it into a list for pyarrow
+        # If we get a dict, we want to parse those separately
+        date_format = self.date_format
+        if isinstance(date_format, str):
+            date_format = [date_format]
+        else:
+            # In case of dict, we don't want to propagate through, so
+            # just set to pyarrow default of None
+
+            # Ideally, in future we disable pyarrow dtype inference (read in as string)
+            # to prevent misreads.
+            date_format = None
+        self.kwds["timestamp_parsers"] = date_format
+
+        self.parse_options = {
+            option_name: option_value
+            for option_name, option_value in self.kwds.items()
+            if option_value is not None
+            and option_name
+            in ("delimiter", "quote_char", "escape_char", "ignore_empty_lines")
+        }
+
+        on_bad_lines = self.kwds.get("on_bad_lines")
+        if on_bad_lines is not None:
+            if callable(on_bad_lines):
+                self.parse_options["invalid_row_handler"] = on_bad_lines
+            elif on_bad_lines == ParserBase.BadLineHandleMethod.ERROR:
+                self.parse_options["invalid_row_handler"] = (
+                    None  # PyArrow raises an exception by default
+                )
+            elif on_bad_lines == ParserBase.BadLineHandleMethod.WARN:
+
+                def handle_warning(invalid_row) -> str:
+                    warnings.warn(
+                        f"Expected {invalid_row.expected_columns} columns, but found "
+                        f"{invalid_row.actual_columns}: {invalid_row.text}",
+                        ParserWarning,
+                        stacklevel=find_stack_level(),
+                    )
+                    return "skip"
+
+                self.parse_options["invalid_row_handler"] = handle_warning
+            elif on_bad_lines == ParserBase.BadLineHandleMethod.SKIP:
+                self.parse_options["invalid_row_handler"] = lambda _: "skip"
+
+        self.convert_options = {
+            option_name: option_value
+            for option_name, option_value in self.kwds.items()
+            if option_value is not None
+            and option_name
+            in (
+                "include_columns",
+                "null_values",
+                "true_values",
+                "false_values",
+                "decimal_point",
+                "timestamp_parsers",
+            )
+        }
+        self.convert_options["strings_can_be_null"] = "" in self.kwds["null_values"]
+        # autogenerated column names are prefixed with 'f' in pyarrow.csv
+        if self.header is None and "include_columns" in self.convert_options:
+            self.convert_options["include_columns"] = [
+                f"f{n}" for n in self.convert_options["include_columns"]
+            ]
+
+        self.read_options = {
+            "autogenerate_column_names": self.header is None,
+            "skip_rows": self.header
+            if self.header is not None
+            else self.kwds["skiprows"],
+            "encoding": self.encoding,
+        }
+
+    def _finalize_pandas_output(self, frame: DataFrame) -> DataFrame:
+        """
+        Processes data read in based on kwargs.
+
+        Parameters
+        ----------
+        frame: DataFrame
+            The DataFrame to process.
+
+        Returns
+        -------
+        DataFrame
+            The processed DataFrame.
+        """
+        num_cols = len(frame.columns)
+        multi_index_named = True
+        if self.header is None:
+            if self.names is None:
+                if self.header is None:
+                    self.names = range(num_cols)
+            if len(self.names) != num_cols:
+                # usecols is passed through to pyarrow, we only handle index col here
+                # The only way self.names is not the same length as number of cols is
+                # if we have int index_col. We should just pad the names(they will get
+                # removed anyways) to expected length then.
+                columns_prefix = [str(x) for x in range(num_cols - len(self.names))]
+                self.names = columns_prefix + self.names
+                multi_index_named = False
+            frame.columns = self.names
+
+        frame = self._do_date_conversions(frame.columns, frame)
+        if self.index_col is not None:
+            index_to_set = self.index_col.copy()
+            for i, item in enumerate(self.index_col):
+                if is_integer(item):
+                    index_to_set[i] = frame.columns[item]
+                # String case
+                elif item not in frame.columns:
+                    raise ValueError(f"Index {item} invalid")
+
+                # Process dtype for index_col and drop from dtypes
+                if self.dtype is not None:
+                    key, new_dtype = (
+                        (item, self.dtype.get(item))
+                        if self.dtype.get(item) is not None
+                        else (frame.columns[item], self.dtype.get(frame.columns[item]))
+                    )
+                    if new_dtype is not None:
+                        frame[key] = frame[key].astype(new_dtype)
+                        del self.dtype[key]
+
+            frame.set_index(index_to_set, drop=True, inplace=True)
+            # Clear names if headerless and no name given
+            if self.header is None and not multi_index_named:
+                frame.index.names = [None] * len(frame.index.names)
+
+        if self.dtype is not None:
+            # Ignore non-existent columns from dtype mapping
+            # like other parsers do
+            if isinstance(self.dtype, dict):
+                self.dtype = {
+                    k: pandas_dtype(v)
+                    for k, v in self.dtype.items()
+                    if k in frame.columns
+                }
+            else:
+                self.dtype = pandas_dtype(self.dtype)
+            try:
+                frame = frame.astype(self.dtype)
+            except TypeError as err:
+                # GH#44901 reraise to keep api consistent
+                raise ValueError(str(err)) from err
+        return frame
+
+    def _validate_usecols(self, usecols) -> None:
+        if lib.is_list_like(usecols) and not all(isinstance(x, str) for x in usecols):
+            raise ValueError(
+                "The pyarrow engine does not allow 'usecols' to be integer "
+                "column positions. Pass a list of string column names instead."
+            )
+        elif callable(usecols):
+            raise ValueError(
+                "The pyarrow engine does not allow 'usecols' to be a callable."
+            )
+
+    def read(self) -> DataFrame:
+        """
+        Reads the contents of a CSV file into a DataFrame and
+        processes it according to the kwargs passed in the
+        constructor.
+
+        Returns
+        -------
+        DataFrame
+            The DataFrame created from the CSV file.
+        """
+        pa = import_optional_dependency("pyarrow")
+        pyarrow_csv = import_optional_dependency("pyarrow.csv")
+        self._get_pyarrow_options()
+
+        try:
+            convert_options = pyarrow_csv.ConvertOptions(**self.convert_options)
+        except TypeError as err:
+            include = self.convert_options.get("include_columns", None)
+            if include is not None:
+                self._validate_usecols(include)
+
+            nulls = self.convert_options.get("null_values", set())
+            if not lib.is_list_like(nulls) or not all(
+                isinstance(x, str) for x in nulls
+            ):
+                raise TypeError(
+                    "The 'pyarrow' engine requires all na_values to be strings"
+                ) from err
+
+            raise
+
+        try:
+            table = pyarrow_csv.read_csv(
+                self.src,
+                read_options=pyarrow_csv.ReadOptions(**self.read_options),
+                parse_options=pyarrow_csv.ParseOptions(**self.parse_options),
+                convert_options=convert_options,
+            )
+        except pa.ArrowInvalid as e:
+            raise ParserError(e) from e
+
+        dtype_backend = self.kwds["dtype_backend"]
+
+        # Convert all pa.null() cols -> float64 (non nullable)
+        # else Int64 (nullable case, see below)
+        if dtype_backend is lib.no_default:
+            new_schema = table.schema
+            new_type = pa.float64()
+            for i, arrow_type in enumerate(table.schema.types):
+                if pa.types.is_null(arrow_type):
+                    new_schema = new_schema.set(
+                        i, new_schema.field(i).with_type(new_type)
+                    )
+
+            table = table.cast(new_schema)
+
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore",
+                "make_block is deprecated",
+                DeprecationWarning,
+            )
+            frame = arrow_table_to_pandas(
+                table, dtype_backend=dtype_backend, null_to_int64=True
+            )
+
+        return self._finalize_pandas_output(frame)
diff --git a/pandas/pandas/io/parsers/base_parser.py b/pandas/pandas/io/parsers/base_parser.py
new file mode 100644
index 0000000000000..c283f600eb971
--- /dev/null
+++ b/pandas/pandas/io/parsers/base_parser.py
@@ -0,0 +1,991 @@
+from __future__ import annotations
+
+from collections import defaultdict
+from copy import copy
+import csv
+from enum import Enum
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+    final,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    parsers,
+)
+import pandas._libs.ops as libops
+from pandas._libs.parsers import STR_NA_VALUES
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    ParserError,
+    ParserWarning,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_dict_like,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_list_like,
+    is_object_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.missing import isna
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    StringDtype,
+)
+from pandas.core import algorithms
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    BaseMaskedArray,
+    BooleanArray,
+    FloatingArray,
+    IntegerArray,
+)
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    default_index,
+    ensure_index_from_sequences,
+)
+from pandas.core.series import Series
+from pandas.core.tools import datetimes as tools
+
+from pandas.io.common import is_potential_multi_index
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Iterable,
+        Mapping,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        DtypeArg,
+        Hashable,
+        HashableT,
+        Scalar,
+        SequenceT,
+    )
+
+
+class ParserBase:
+    class BadLineHandleMethod(Enum):
+        ERROR = 0
+        WARN = 1
+        SKIP = 2
+
+    _implicit_index: bool
+    _first_chunk: bool
+    keep_default_na: bool
+    dayfirst: bool
+    cache_dates: bool
+    usecols_dtype: str | None
+
+    def __init__(self, kwds) -> None:
+        self._implicit_index = False
+
+        self.names = kwds.get("names")
+        self.orig_names: Sequence[Hashable] | None = None
+
+        self.index_col = kwds.get("index_col", None)
+        self.unnamed_cols: set = set()
+        self.index_names: Sequence[Hashable] | None = None
+        self.col_names: Sequence[Hashable] | None = None
+
+        parse_dates = kwds.pop("parse_dates", False)
+        if parse_dates is None or lib.is_bool(parse_dates):
+            parse_dates = bool(parse_dates)
+        elif not isinstance(parse_dates, list):
+            raise TypeError(
+                "Only booleans and lists are accepted for the 'parse_dates' parameter"
+            )
+        self.parse_dates: bool | list = parse_dates
+        self.date_parser = kwds.pop("date_parser", lib.no_default)
+        self.date_format = kwds.pop("date_format", None)
+        self.dayfirst = kwds.pop("dayfirst", False)
+
+        self.na_values = kwds.get("na_values")
+        self.na_fvalues = kwds.get("na_fvalues")
+        self.na_filter = kwds.get("na_filter", False)
+        self.keep_default_na = kwds.get("keep_default_na", True)
+
+        self.dtype = copy(kwds.get("dtype", None))
+        self.converters = kwds.get("converters")
+        self.dtype_backend = kwds.get("dtype_backend")
+
+        self.true_values = kwds.get("true_values")
+        self.false_values = kwds.get("false_values")
+        self.cache_dates = kwds.pop("cache_dates", True)
+
+        # validate header options for mi
+        self.header = kwds.get("header")
+        if is_list_like(self.header, allow_sets=False):
+            if kwds.get("usecols"):
+                raise ValueError(
+                    "cannot specify usecols when specifying a multi-index header"
+                )
+            if kwds.get("names"):
+                raise ValueError(
+                    "cannot specify names when specifying a multi-index header"
+                )
+
+            # validate index_col that only contains integers
+            if self.index_col is not None:
+                # In this case we can pin down index_col as list[int]
+                if is_integer(self.index_col):
+                    self.index_col = [self.index_col]
+                elif not (
+                    is_list_like(self.index_col, allow_sets=False)
+                    and all(map(is_integer, self.index_col))
+                ):
+                    raise ValueError(
+                        "index_col must only contain integers of column positions "
+                        "when specifying a multi-index header"
+                    )
+                else:
+                    self.index_col = list(self.index_col)
+
+        self._first_chunk = True
+
+        self.usecols, self.usecols_dtype = _validate_usecols_arg(kwds["usecols"])
+
+        # Fallback to error to pass a sketchy test(test_override_set_noconvert_columns)
+        # Normally, this arg would get pre-processed earlier on
+        self.on_bad_lines = kwds.get("on_bad_lines", self.BadLineHandleMethod.ERROR)
+
+    def close(self) -> None:
+        pass
+
+    @final
+    def _should_parse_dates(self, i: int) -> bool:
+        if isinstance(self.parse_dates, bool):
+            return self.parse_dates
+        else:
+            if self.index_names is not None:
+                name = self.index_names[i]
+            else:
+                name = None
+            j = i if self.index_col is None else self.index_col[i]
+
+            return (j in self.parse_dates) or (
+                name is not None and name in self.parse_dates
+            )
+
+    @final
+    def _extract_multi_indexer_columns(
+        self,
+        header,
+        index_names: Sequence[Hashable] | None,
+        passed_names: bool = False,
+    ) -> tuple[
+        Sequence[Hashable], Sequence[Hashable] | None, Sequence[Hashable] | None, bool
+    ]:
+        """
+        Extract and return the names, index_names, col_names if the column
+        names are a MultiIndex.
+
+        Parameters
+        ----------
+        header: list of lists
+            The header rows
+        index_names: list, optional
+            The names of the future index
+        passed_names: bool, default False
+            A flag specifying if names where passed
+
+        """
+        if len(header) < 2:
+            return header[0], index_names, None, passed_names
+
+        # the names are the tuples of the header that are not the index cols
+        # 0 is the name of the index, assuming index_col is a list of column
+        # numbers
+        ic = self.index_col
+        if ic is None:
+            ic = []
+
+        if not isinstance(ic, (list, tuple, np.ndarray)):
+            ic = [ic]
+        sic = set(ic)
+
+        # clean the index_names
+        index_names = header.pop(-1)
+        index_names, _, _ = self._clean_index_names(index_names, self.index_col)
+
+        # extract the columns
+        field_count = len(header[0])
+
+        # check if header lengths are equal
+        if not all(len(header_iter) == field_count for header_iter in header[1:]):
+            raise ParserError("Header rows must have an equal number of columns.")
+
+        def extract(r):
+            return tuple(r[i] for i in range(field_count) if i not in sic)
+
+        columns = list(zip(*(extract(r) for r in header)))
+        names = columns.copy()
+        for single_ic in sorted(ic):
+            names.insert(single_ic, single_ic)
+
+        # Clean the column names (if we have an index_col).
+        if len(ic):
+            col_names = [
+                r[ic[0]]
+                if ((r[ic[0]] is not None) and r[ic[0]] not in self.unnamed_cols)
+                else None
+                for r in header
+            ]
+        else:
+            col_names = [None] * len(header)
+
+        passed_names = True
+
+        return names, index_names, col_names, passed_names
+
+    @final
+    def _maybe_make_multi_index_columns(
+        self,
+        columns: SequenceT,
+        col_names: Sequence[Hashable] | None = None,
+    ) -> SequenceT | MultiIndex:
+        # possibly create a column mi here
+        if is_potential_multi_index(columns):
+            columns_mi = cast("Sequence[tuple[Hashable, ...]]", columns)
+            return MultiIndex.from_tuples(columns_mi, names=col_names)
+        return columns
+
+    @final
+    def _make_index(
+        self, alldata, columns, indexnamerow: list[Scalar] | None = None
+    ) -> tuple[Index | None, Sequence[Hashable] | MultiIndex]:
+        index: Index | None
+        if isinstance(self.index_col, list) and len(self.index_col):
+            to_remove = []
+            indexes = []
+            for idx in self.index_col:
+                if isinstance(idx, str):
+                    raise ValueError(f"Index {idx} invalid")
+                to_remove.append(idx)
+                indexes.append(alldata[idx])
+            # remove index items from content and columns, don't pop in
+            # loop
+            for i in sorted(to_remove, reverse=True):
+                alldata.pop(i)
+                if not self._implicit_index:
+                    columns.pop(i)
+            index = self._agg_index(indexes)
+
+            # add names for the index
+            if indexnamerow:
+                coffset = len(indexnamerow) - len(columns)
+                index = index.set_names(indexnamerow[:coffset])
+        else:
+            index = None
+
+        # maybe create a mi on the columns
+        columns = self._maybe_make_multi_index_columns(columns, self.col_names)
+
+        return index, columns
+
+    @final
+    def _clean_mapping(self, mapping):
+        """converts col numbers to names"""
+        if not isinstance(mapping, dict):
+            return mapping
+        clean = {}
+        # for mypy
+        assert self.orig_names is not None
+
+        for col, v in mapping.items():
+            if isinstance(col, int) and col not in self.orig_names:
+                col = self.orig_names[col]
+            clean[col] = v
+        if isinstance(mapping, defaultdict):
+            remaining_cols = set(self.orig_names) - set(clean.keys())
+            clean.update({col: mapping[col] for col in remaining_cols})
+        return clean
+
+    @final
+    def _agg_index(self, index) -> Index:
+        arrays = []
+        converters = self._clean_mapping(self.converters)
+        clean_dtypes = self._clean_mapping(self.dtype)
+
+        if self.index_names is not None:
+            names: Iterable = self.index_names
+        else:
+            names = itertools.cycle([None])
+        for i, (arr, name) in enumerate(zip(index, names)):
+            if self._should_parse_dates(i):
+                arr = date_converter(
+                    arr,
+                    col=self.index_names[i] if self.index_names is not None else None,
+                    dayfirst=self.dayfirst,
+                    cache_dates=self.cache_dates,
+                    date_format=self.date_format,
+                )
+
+            if self.na_filter:
+                col_na_values = self.na_values
+                col_na_fvalues = self.na_fvalues
+            else:
+                col_na_values = set()
+                col_na_fvalues = set()
+
+            if isinstance(self.na_values, dict):
+                assert self.index_names is not None
+                col_name = self.index_names[i]
+                if col_name is not None:
+                    col_na_values, col_na_fvalues = get_na_values(
+                        col_name, self.na_values, self.na_fvalues, self.keep_default_na
+                    )
+                else:
+                    col_na_values, col_na_fvalues = set(), set()
+
+            cast_type = None
+            index_converter = False
+            if self.index_names is not None:
+                if isinstance(clean_dtypes, dict):
+                    cast_type = clean_dtypes.get(self.index_names[i], None)
+
+                if isinstance(converters, dict):
+                    index_converter = converters.get(self.index_names[i]) is not None
+
+            try_num_bool = not (
+                (cast_type and is_string_dtype(cast_type)) or index_converter
+            )
+
+            arr, _ = self._infer_types(
+                arr, col_na_values | col_na_fvalues, cast_type is None, try_num_bool
+            )
+            if cast_type is not None:
+                # Don't perform RangeIndex inference
+                idx = Index(arr, name=name, dtype=cast_type)
+            else:
+                idx = ensure_index_from_sequences([arr], [name])
+            arrays.append(idx)
+
+        if len(arrays) == 1:
+            return arrays[0]
+        else:
+            return MultiIndex.from_arrays(arrays)
+
+    @final
+    def _set_noconvert_dtype_columns(
+        self, col_indices: list[int], names: Sequence[Hashable]
+    ) -> set[int]:
+        """
+        Set the columns that should not undergo dtype conversions.
+
+        Currently, any column that is involved with date parsing will not
+        undergo such conversions. If usecols is specified, the positions of the columns
+        not to cast is relative to the usecols not to all columns.
+
+        Parameters
+        ----------
+        col_indices: The indices specifying order and positions of the columns
+        names: The column names which order is corresponding with the order
+               of col_indices
+
+        Returns
+        -------
+        A set of integers containing the positions of the columns not to convert.
+        """
+        usecols: list[int] | list[str] | None
+        noconvert_columns = set()
+        if self.usecols_dtype == "integer":
+            # A set of integers will be converted to a list in
+            # the correct order every single time.
+            usecols = sorted(self.usecols)
+        elif callable(self.usecols) or self.usecols_dtype not in ("empty", None):
+            # The names attribute should have the correct columns
+            # in the proper order for indexing with parse_dates.
+            usecols = col_indices
+        else:
+            # Usecols is empty.
+            usecols = None
+
+        def _set(x) -> int:
+            if usecols is not None and is_integer(x):
+                x = usecols[x]
+
+            if not is_integer(x):
+                x = col_indices[names.index(x)]
+
+            return x
+
+        if isinstance(self.parse_dates, list):
+            validate_parse_dates_presence(self.parse_dates, names)
+            for val in self.parse_dates:
+                noconvert_columns.add(_set(val))
+
+        elif self.parse_dates:
+            if isinstance(self.index_col, list):
+                for k in self.index_col:
+                    noconvert_columns.add(_set(k))
+            elif self.index_col is not None:
+                noconvert_columns.add(_set(self.index_col))
+
+        return noconvert_columns
+
+    @final
+    def _infer_types(
+        self, values, na_values, no_dtype_specified, try_num_bool: bool = True
+    ) -> tuple[ArrayLike, int]:
+        """
+        Infer types of values, possibly casting
+
+        Parameters
+        ----------
+        values : ndarray
+        na_values : set
+        no_dtype_specified: Specifies if we want to cast explicitly
+        try_num_bool : bool, default try
+           try to cast values to numeric (first preference) or boolean
+
+        Returns
+        -------
+        converted : ndarray or ExtensionArray
+        na_count : int
+        """
+        na_count = 0
+        if issubclass(values.dtype.type, (np.number, np.bool_)):
+            # If our array has numeric dtype, we don't have to check for strings in isin
+            na_values = np.array([val for val in na_values if not isinstance(val, str)])
+            mask = algorithms.isin(values, na_values)
+            na_count = mask.astype("uint8", copy=False).sum()
+            if na_count > 0:
+                if is_integer_dtype(values):
+                    values = values.astype(np.float64)
+                np.putmask(values, mask, np.nan)
+            return values, na_count
+
+        dtype_backend = self.dtype_backend
+        non_default_dtype_backend = (
+            no_dtype_specified and dtype_backend is not lib.no_default
+        )
+        result: ArrayLike
+
+        if try_num_bool and is_object_dtype(values.dtype):
+            # exclude e.g DatetimeIndex here
+            try:
+                result, result_mask = lib.maybe_convert_numeric(
+                    values,
+                    na_values,
+                    False,
+                    convert_to_masked_nullable=non_default_dtype_backend,  # type: ignore[arg-type]
+                )
+            except (ValueError, TypeError):
+                # e.g. encountering datetime string gets ValueError
+                #  TypeError can be raised in floatify
+                na_count = parsers.sanitize_objects(values, na_values)
+                result = values
+            else:
+                if non_default_dtype_backend:
+                    if result_mask is None:
+                        result_mask = np.zeros(result.shape, dtype=np.bool_)
+
+                    if result_mask.all():
+                        result = IntegerArray(
+                            np.ones(result_mask.shape, dtype=np.int64), result_mask
+                        )
+                    elif is_integer_dtype(result):
+                        result = IntegerArray(result, result_mask)
+                    elif is_bool_dtype(result):
+                        result = BooleanArray(result, result_mask)
+                    elif is_float_dtype(result):
+                        result = FloatingArray(result, result_mask)
+
+                    na_count = result_mask.sum()
+                else:
+                    na_count = isna(result).sum()
+        else:
+            result = values
+            if values.dtype == np.object_:
+                na_count = parsers.sanitize_objects(values, na_values)
+
+        if result.dtype == np.object_ and try_num_bool:
+            result, bool_mask = libops.maybe_convert_bool(
+                np.asarray(values),
+                true_values=self.true_values,
+                false_values=self.false_values,
+                convert_to_masked_nullable=non_default_dtype_backend,  # type: ignore[arg-type]
+            )
+            if result.dtype == np.bool_ and non_default_dtype_backend:
+                if bool_mask is None:
+                    bool_mask = np.zeros(result.shape, dtype=np.bool_)
+                result = BooleanArray(result, bool_mask)
+            elif result.dtype == np.object_ and non_default_dtype_backend:
+                # read_excel sends array of datetime objects
+                if not lib.is_datetime_array(result, skipna=True):
+                    dtype = StringDtype()
+                    cls = dtype.construct_array_type()
+                    result = cls._from_sequence(values, dtype=dtype)
+
+        if dtype_backend == "pyarrow":
+            pa = import_optional_dependency("pyarrow")
+            if isinstance(result, np.ndarray):
+                result = ArrowExtensionArray(pa.array(result, from_pandas=True))
+            elif isinstance(result, BaseMaskedArray):
+                if result._mask.all():
+                    # We want an arrow null array here
+                    result = ArrowExtensionArray(pa.array([None] * len(result)))
+                else:
+                    result = ArrowExtensionArray(
+                        pa.array(result._data, mask=result._mask)
+                    )
+            else:
+                result = ArrowExtensionArray(
+                    pa.array(result.to_numpy(), from_pandas=True)
+                )
+
+        return result, na_count
+
+    @overload
+    def _do_date_conversions(
+        self,
+        names: Index,
+        data: DataFrame,
+    ) -> DataFrame: ...
+
+    @overload
+    def _do_date_conversions(
+        self,
+        names: Sequence[Hashable],
+        data: Mapping[Hashable, ArrayLike],
+    ) -> Mapping[Hashable, ArrayLike]: ...
+
+    @final
+    def _do_date_conversions(
+        self,
+        names: Sequence[Hashable] | Index,
+        data: Mapping[Hashable, ArrayLike] | DataFrame,
+    ) -> Mapping[Hashable, ArrayLike] | DataFrame:
+        if not isinstance(self.parse_dates, list):
+            return data
+        for colspec in self.parse_dates:
+            if isinstance(colspec, int) and colspec not in data:
+                colspec = names[colspec]
+            if (isinstance(self.index_col, list) and colspec in self.index_col) or (
+                isinstance(self.index_names, list) and colspec in self.index_names
+            ):
+                continue
+            result = date_converter(
+                data[colspec],
+                col=colspec,
+                dayfirst=self.dayfirst,
+                cache_dates=self.cache_dates,
+                date_format=self.date_format,
+            )
+            # error: Unsupported target for indexed assignment
+            # ("Mapping[Hashable, ExtensionArray | ndarray[Any, Any]] | DataFrame")
+            data[colspec] = result  # type: ignore[index]
+
+        return data
+
+    @final
+    def _check_data_length(
+        self,
+        columns: Sequence[Hashable],
+        data: Sequence[ArrayLike],
+    ) -> None:
+        """Checks if length of data is equal to length of column names.
+
+        One set of trailing commas is allowed. self.index_col not False
+        results in a ParserError previously when lengths do not match.
+
+        Parameters
+        ----------
+        columns: list of column names
+        data: list of array-likes containing the data column-wise.
+        """
+        if not self.index_col and len(columns) != len(data) and columns:
+            empty_str = is_object_dtype(data[-1]) and data[-1] == ""
+            # error: No overload variant of "__ror__" of "ndarray" matches
+            # argument type "ExtensionArray"
+            empty_str_or_na = empty_str | isna(data[-1])  # type: ignore[operator]
+            if len(columns) == len(data) - 1 and np.all(empty_str_or_na):
+                return
+            warnings.warn(
+                "Length of header or names does not match length of data. This leads "
+                "to a loss of data with index_col=False.",
+                ParserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+    @final
+    def _validate_usecols_names(self, usecols: SequenceT, names: Sequence) -> SequenceT:
+        """
+        Validates that all usecols are present in a given
+        list of names. If not, raise a ValueError that
+        shows what usecols are missing.
+
+        Parameters
+        ----------
+        usecols : iterable of usecols
+            The columns to validate are present in names.
+        names : iterable of names
+            The column names to check against.
+
+        Returns
+        -------
+        usecols : iterable of usecols
+            The `usecols` parameter if the validation succeeds.
+
+        Raises
+        ------
+        ValueError : Columns were missing. Error message will list them.
+        """
+        missing = [c for c in usecols if c not in names]
+        if len(missing) > 0:
+            raise ValueError(
+                f"Usecols do not match columns, columns expected but not found: "
+                f"{missing}"
+            )
+
+        return usecols
+
+    @final
+    def _clean_index_names(self, columns, index_col) -> tuple[list | None, list, list]:
+        if not is_index_col(index_col):
+            return None, columns, index_col
+
+        columns = list(columns)
+
+        # In case of no rows and multiindex columns we have to set index_names to
+        # list of Nones GH#38292
+        if not columns:
+            return [None] * len(index_col), columns, index_col
+
+        cp_cols = list(columns)
+        index_names: list[str | int | None] = []
+
+        # don't mutate
+        index_col = list(index_col)
+
+        for i, c in enumerate(index_col):
+            if isinstance(c, str):
+                index_names.append(c)
+                for j, name in enumerate(cp_cols):
+                    if name == c:
+                        index_col[i] = j
+                        columns.remove(name)
+                        break
+            else:
+                name = cp_cols[c]
+                columns.remove(name)
+                index_names.append(name)
+
+        # Only clean index names that were placeholders.
+        for i, name in enumerate(index_names):
+            if isinstance(name, str) and name in self.unnamed_cols:
+                index_names[i] = None
+
+        return index_names, columns, index_col
+
+    @final
+    def _get_empty_meta(
+        self, columns: Sequence[HashableT], dtype: DtypeArg | None = None
+    ) -> tuple[Index, list[HashableT], dict[HashableT, Series]]:
+        columns = list(columns)
+
+        index_col = self.index_col
+        index_names = self.index_names
+
+        # Convert `dtype` to a defaultdict of some kind.
+        # This will enable us to write `dtype[col_name]`
+        # without worrying about KeyError issues later on.
+        dtype_dict: defaultdict[Hashable, Any]
+        if not is_dict_like(dtype):
+            # if dtype == None, default will be object.
+            dtype_dict = defaultdict(lambda: dtype)
+        else:
+            dtype = cast(dict, dtype)
+            dtype_dict = defaultdict(
+                lambda: None,
+                {columns[k] if is_integer(k) else k: v for k, v in dtype.items()},
+            )
+
+        # Even though we have no data, the "index" of the empty DataFrame
+        # could for example still be an empty MultiIndex. Thus, we need to
+        # check whether we have any index columns specified, via either:
+        #
+        # 1) index_col (column indices)
+        # 2) index_names (column names)
+        #
+        # Both must be non-null to ensure a successful construction. Otherwise,
+        # we have to create a generic empty Index.
+        index: Index
+        if (index_col is None or index_col is False) or index_names is None:
+            index = default_index(0)
+        else:
+            # TODO: We could return default_index(0) if dtype_dict[name] is None
+            data = [
+                Index([], name=name, dtype=dtype_dict[name]) for name in index_names
+            ]
+            if len(data) == 1:
+                index = data[0]
+            else:
+                index = MultiIndex.from_arrays(data)
+            index_col.sort()
+
+            for i, n in enumerate(index_col):
+                columns.pop(n - i)
+
+        col_dict = {
+            col_name: Series([], dtype=dtype_dict[col_name]) for col_name in columns
+        }
+
+        return index, columns, col_dict
+
+
+def date_converter(
+    date_col,
+    col: Hashable,
+    dayfirst: bool = False,
+    cache_dates: bool = True,
+    date_format: dict[Hashable, str] | str | None = None,
+):
+    if date_col.dtype.kind in "Mm":
+        return date_col
+
+    date_fmt = date_format.get(col) if isinstance(date_format, dict) else date_format
+
+    str_objs = lib.ensure_string_array(np.asarray(date_col))
+    try:
+        result = tools.to_datetime(
+            str_objs,
+            format=date_fmt,
+            utc=False,
+            dayfirst=dayfirst,
+            cache=cache_dates,
+        )
+    except (ValueError, TypeError):
+        # test_usecols_with_parse_dates4
+        # test_multi_index_parse_dates
+        return str_objs
+
+    if isinstance(result, DatetimeIndex):
+        arr = result.to_numpy()
+        arr.flags.writeable = True
+        return arr
+    return result._values
+
+
+parser_defaults = {
+    "delimiter": None,
+    "escapechar": None,
+    "quotechar": '"',
+    "quoting": csv.QUOTE_MINIMAL,
+    "doublequote": True,
+    "skipinitialspace": False,
+    "lineterminator": None,
+    "header": "infer",
+    "index_col": None,
+    "names": None,
+    "skiprows": None,
+    "skipfooter": 0,
+    "nrows": None,
+    "na_values": None,
+    "keep_default_na": True,
+    "true_values": None,
+    "false_values": None,
+    "converters": None,
+    "dtype": None,
+    "cache_dates": True,
+    "thousands": None,
+    "comment": None,
+    "decimal": ".",
+    # 'engine': 'c',
+    "parse_dates": False,
+    "dayfirst": False,
+    "date_format": None,
+    "usecols": None,
+    # 'iterator': False,
+    "chunksize": None,
+    "encoding": None,
+    "compression": None,
+    "skip_blank_lines": True,
+    "encoding_errors": "strict",
+    "on_bad_lines": ParserBase.BadLineHandleMethod.ERROR,
+    "dtype_backend": lib.no_default,
+}
+
+
+def get_na_values(col, na_values, na_fvalues, keep_default_na: bool):
+    """
+    Get the NaN values for a given column.
+
+    Parameters
+    ----------
+    col : str
+        The name of the column.
+    na_values : array-like, dict
+        The object listing the NaN values as strings.
+    na_fvalues : array-like, dict
+        The object listing the NaN values as floats.
+    keep_default_na : bool
+        If `na_values` is a dict, and the column is not mapped in the
+        dictionary, whether to return the default NaN values or the empty set.
+
+    Returns
+    -------
+    nan_tuple : A length-two tuple composed of
+
+        1) na_values : the string NaN values for that column.
+        2) na_fvalues : the float NaN values for that column.
+    """
+    if isinstance(na_values, dict):
+        if col in na_values:
+            return na_values[col], na_fvalues[col]
+        else:
+            if keep_default_na:
+                return STR_NA_VALUES, set()
+
+            return set(), set()
+    else:
+        return na_values, na_fvalues
+
+
+def is_index_col(col) -> bool:
+    return col is not None and col is not False
+
+
+def validate_parse_dates_presence(
+    parse_dates: bool | list, columns: Sequence[Hashable]
+) -> set:
+    """
+    Check if parse_dates are in columns.
+
+    If user has provided names for parse_dates, check if those columns
+    are available.
+
+    Parameters
+    ----------
+    columns : list
+        List of names of the dataframe.
+
+    Returns
+    -------
+    The names of the columns which will get parsed later if a list
+    is given as specification.
+
+    Raises
+    ------
+    ValueError
+        If column to parse_date is not in dataframe.
+
+    """
+    if not isinstance(parse_dates, list):
+        return set()
+
+    missing = set()
+    unique_cols = set()
+    for col in parse_dates:
+        if isinstance(col, str):
+            if col not in columns:
+                missing.add(col)
+            else:
+                unique_cols.add(col)
+        elif col in columns:
+            unique_cols.add(col)
+        else:
+            unique_cols.add(columns[col])
+    if missing:
+        missing_cols = ", ".join(sorted(missing))
+        raise ValueError(f"Missing column provided to 'parse_dates': '{missing_cols}'")
+    return unique_cols
+
+
+def _validate_usecols_arg(usecols):
+    """
+    Validate the 'usecols' parameter.
+
+    Checks whether or not the 'usecols' parameter contains all integers
+    (column selection by index), strings (column by name) or is a callable.
+    Raises a ValueError if that is not the case.
+
+    Parameters
+    ----------
+    usecols : list-like, callable, or None
+        List of columns to use when parsing or a callable that can be used
+        to filter a list of table columns.
+
+    Returns
+    -------
+    usecols_tuple : tuple
+        A tuple of (verified_usecols, usecols_dtype).
+
+        'verified_usecols' is either a set if an array-like is passed in or
+        'usecols' if a callable or None is passed in.
+
+        'usecols_dtype` is the inferred dtype of 'usecols' if an array-like
+        is passed in or None if a callable or None is passed in.
+    """
+    msg = (
+        "'usecols' must either be list-like of all strings, all unicode, "
+        "all integers or a callable."
+    )
+    if usecols is not None:
+        if callable(usecols):
+            return usecols, None
+
+        if not is_list_like(usecols):
+            # see gh-20529
+            #
+            # Ensure it is iterable container but not string.
+            raise ValueError(msg)
+
+        usecols_dtype = lib.infer_dtype(usecols, skipna=False)
+
+        if usecols_dtype not in ("empty", "integer", "string"):
+            raise ValueError(msg)
+
+        usecols = set(usecols)
+
+        return usecols, usecols_dtype
+    return usecols, None
+
+
+@overload
+def evaluate_callable_usecols(
+    usecols: Callable[[Hashable], object],
+    names: Iterable[Hashable],
+) -> set[int]: ...
+
+
+@overload
+def evaluate_callable_usecols(
+    usecols: SequenceT, names: Iterable[Hashable]
+) -> SequenceT: ...
+
+
+def evaluate_callable_usecols(
+    usecols: Callable[[Hashable], object] | SequenceT,
+    names: Iterable[Hashable],
+) -> SequenceT | set[int]:
+    """
+    Check whether or not the 'usecols' parameter
+    is a callable.  If so, enumerates the 'names'
+    parameter and returns a set of indices for
+    each entry in 'names' that evaluates to True.
+    If not a callable, returns 'usecols'.
+    """
+    if callable(usecols):
+        return {i for i, name in enumerate(names) if usecols(name)}
+    return usecols
diff --git a/pandas/pandas/io/parsers/c_parser_wrapper.py b/pandas/pandas/io/parsers/c_parser_wrapper.py
new file mode 100644
index 0000000000000..818c9f5ff6b80
--- /dev/null
+++ b/pandas/pandas/io/parsers/c_parser_wrapper.py
@@ -0,0 +1,413 @@
+from __future__ import annotations
+
+from collections import defaultdict
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas._libs import (
+    lib,
+    parsers,
+)
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import DtypeWarning
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.concat import (
+    concat_compat,
+    union_categoricals,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+from pandas.core.indexes.api import ensure_index_from_sequences
+
+from pandas.io.common import (
+    dedup_names,
+    is_potential_multi_index,
+)
+from pandas.io.parsers.base_parser import (
+    ParserBase,
+    ParserError,
+    date_converter,
+    evaluate_callable_usecols,
+    is_index_col,
+    validate_parse_dates_presence,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Mapping,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        AnyArrayLike,
+        ArrayLike,
+        DtypeArg,
+        DtypeObj,
+        ReadCsvBuffer,
+        SequenceT,
+    )
+
+    from pandas import (
+        Index,
+        MultiIndex,
+    )
+
+
+class CParserWrapper(ParserBase):
+    low_memory: bool
+    _reader: parsers.TextReader
+
+    def __init__(self, src: ReadCsvBuffer[str], **kwds) -> None:
+        super().__init__(kwds)
+        self.kwds = kwds
+        kwds = kwds.copy()
+
+        self.low_memory = kwds.pop("low_memory", False)
+
+        # #2442
+        # error: Cannot determine type of 'index_col'
+        kwds["allow_leading_cols"] = (
+            self.index_col is not False  # type: ignore[has-type]
+        )
+
+        # GH20529, validate usecol arg before TextReader
+        kwds["usecols"] = self.usecols
+
+        # Have to pass int, would break tests using TextReader directly otherwise :(
+        kwds["on_bad_lines"] = self.on_bad_lines.value
+
+        for key in (
+            "storage_options",
+            "encoding",
+            "memory_map",
+            "compression",
+        ):
+            kwds.pop(key, None)
+
+        kwds["dtype"] = ensure_dtype_objs(kwds.get("dtype", None))
+        if "dtype_backend" not in kwds or kwds["dtype_backend"] is lib.no_default:
+            kwds["dtype_backend"] = "numpy"
+        if kwds["dtype_backend"] == "pyarrow":
+            # Fail here loudly instead of in cython after reading
+            import_optional_dependency("pyarrow")
+        self._reader = parsers.TextReader(src, **kwds)
+
+        self.unnamed_cols = self._reader.unnamed_cols
+
+        # error: Cannot determine type of 'names'
+        passed_names = self.names is None  # type: ignore[has-type]
+
+        if self._reader.header is None:
+            self.names = None
+        else:
+            # error: Cannot determine type of 'names'
+            # error: Cannot determine type of 'index_names'
+            (
+                self.names,  # type: ignore[has-type]
+                self.index_names,
+                self.col_names,
+                passed_names,
+            ) = self._extract_multi_indexer_columns(
+                self._reader.header,
+                self.index_names,  # type: ignore[has-type]
+                passed_names,
+            )
+
+        # error: Cannot determine type of 'names'
+        if self.names is None:  # type: ignore[has-type]
+            self.names = list(range(self._reader.table_width))
+
+        # gh-9755
+        #
+        # need to set orig_names here first
+        # so that proper indexing can be done
+        # with _set_noconvert_columns
+        #
+        # once names has been filtered, we will
+        # then set orig_names again to names
+        # error: Cannot determine type of 'names'
+        self.orig_names = self.names[:]  # type: ignore[has-type]
+
+        if self.usecols:
+            usecols = evaluate_callable_usecols(self.usecols, self.orig_names)
+
+            # GH 14671
+            # assert for mypy, orig_names is List or None, None would error in issubset
+            assert self.orig_names is not None
+            if self.usecols_dtype == "string" and not set(usecols).issubset(
+                self.orig_names
+            ):
+                self._validate_usecols_names(usecols, self.orig_names)
+
+            # error: Cannot determine type of 'names'
+            if len(self.names) > len(usecols):  # type: ignore[has-type]
+                # error: Cannot determine type of 'names'
+                self.names = [  # type: ignore[has-type]
+                    n
+                    # error: Cannot determine type of 'names'
+                    for i, n in enumerate(self.names)  # type: ignore[has-type]
+                    if (i in usecols or n in usecols)
+                ]
+
+            # error: Cannot determine type of 'names'
+            if len(self.names) < len(usecols):  # type: ignore[has-type]
+                # error: Cannot determine type of 'names'
+                self._validate_usecols_names(
+                    usecols,
+                    self.names,  # type: ignore[has-type]
+                )
+
+        # error: Cannot determine type of 'names'
+        validate_parse_dates_presence(self.parse_dates, self.names)  # type: ignore[has-type]
+        self._set_noconvert_columns()
+
+        # error: Cannot determine type of 'names'
+        self.orig_names = self.names  # type: ignore[has-type]
+
+        # error: Cannot determine type of 'index_col'
+        if self._reader.leading_cols == 0 and is_index_col(
+            self.index_col  # type: ignore[has-type]
+        ):
+            (
+                index_names,
+                # error: Cannot determine type of 'names'
+                self.names,  # type: ignore[has-type]
+                self.index_col,
+            ) = self._clean_index_names(
+                # error: Cannot determine type of 'names'
+                self.names,  # type: ignore[has-type]
+                # error: Cannot determine type of 'index_col'
+                self.index_col,  # type: ignore[has-type]
+            )
+
+            if self.index_names is None:
+                self.index_names = index_names
+
+        if self._reader.header is None and not passed_names:
+            assert self.index_names is not None
+            self.index_names = [None] * len(self.index_names)
+
+        self._implicit_index = self._reader.leading_cols > 0
+
+    def close(self) -> None:
+        # close handles opened by C parser
+        try:
+            self._reader.close()
+        except ValueError:
+            pass
+
+    def _set_noconvert_columns(self) -> None:
+        """
+        Set the columns that should not undergo dtype conversions.
+
+        Currently, any column that is involved with date parsing will not
+        undergo such conversions.
+        """
+        assert self.orig_names is not None
+        # error: Cannot determine type of 'names'
+
+        # much faster than using orig_names.index(x) xref GH#44106
+        names_dict = {x: i for i, x in enumerate(self.orig_names)}
+        col_indices = [names_dict[x] for x in self.names]  # type: ignore[has-type]
+        # error: Cannot determine type of 'names'
+        noconvert_columns = self._set_noconvert_dtype_columns(
+            col_indices,
+            self.names,  # type: ignore[has-type]
+        )
+        for col in noconvert_columns:
+            self._reader.set_noconvert(col)
+
+    def read(
+        self,
+        nrows: int | None = None,
+    ) -> tuple[
+        Index | MultiIndex | None,
+        Sequence[Hashable] | MultiIndex,
+        Mapping[Hashable, AnyArrayLike],
+    ]:
+        index: Index | MultiIndex | None
+        column_names: Sequence[Hashable] | MultiIndex
+        try:
+            if self.low_memory:
+                chunks = self._reader.read_low_memory(nrows)
+                # destructive to chunks
+                data = _concatenate_chunks(chunks, self.names)  # type: ignore[has-type]
+
+            else:
+                data = self._reader.read(nrows)
+        except StopIteration:
+            if self._first_chunk:
+                self._first_chunk = False
+                names = dedup_names(
+                    self.orig_names,
+                    is_potential_multi_index(self.orig_names, self.index_col),
+                )
+                index, columns, col_dict = self._get_empty_meta(
+                    names,
+                    dtype=self.dtype,
+                )
+                # error: Incompatible types in assignment (expression has type
+                # "list[Hashable] | MultiIndex", variable has type "list[Hashable]")
+                columns = self._maybe_make_multi_index_columns(  # type: ignore[assignment]
+                    columns, self.col_names
+                )
+
+                columns = _filter_usecols(self.usecols, columns)
+
+                col_dict = {k: v for k, v in col_dict.items() if k in columns}
+
+                return index, columns, col_dict
+
+            else:
+                self.close()
+                raise
+
+        # Done with first read, next time raise StopIteration
+        self._first_chunk = False
+
+        # error: Cannot determine type of 'names'
+        names = self.names  # type: ignore[has-type]
+
+        if self._reader.leading_cols:
+            # implicit index, no index names
+            arrays = []
+
+            if self.index_col and self._reader.leading_cols != len(self.index_col):
+                raise ParserError(
+                    "Could not construct index. Requested to use "
+                    f"{len(self.index_col)} number of columns, but "
+                    f"{self._reader.leading_cols} left to parse."
+                )
+
+            for i in range(self._reader.leading_cols):
+                if self.index_col is None:
+                    values = data.pop(i)
+                else:
+                    values = data.pop(self.index_col[i])
+
+                if self._should_parse_dates(i):
+                    values = date_converter(
+                        values,
+                        col=self.index_names[i]
+                        if self.index_names is not None
+                        else None,
+                        dayfirst=self.dayfirst,
+                        cache_dates=self.cache_dates,
+                        date_format=self.date_format,
+                    )
+                arrays.append(values)
+
+            index = ensure_index_from_sequences(arrays)
+
+            names = _filter_usecols(self.usecols, names)
+
+            names = dedup_names(names, is_potential_multi_index(names, self.index_col))
+
+            # rename dict keys
+            data_tups = sorted(data.items())
+            data = {k: v for k, (i, v) in zip(names, data_tups)}
+
+            date_data = self._do_date_conversions(names, data)
+
+            # maybe create a mi on the columns
+            column_names = self._maybe_make_multi_index_columns(names, self.col_names)
+
+        else:
+            # rename dict keys
+            data_tups = sorted(data.items())
+
+            # ugh, mutation
+
+            # assert for mypy, orig_names is List or None, None would error in list(...)
+            assert self.orig_names is not None
+            names = list(self.orig_names)
+            names = dedup_names(names, is_potential_multi_index(names, self.index_col))
+
+            names = _filter_usecols(self.usecols, names)
+
+            # columns as list
+            alldata = [x[1] for x in data_tups]
+            if self.usecols is None:
+                self._check_data_length(names, alldata)
+
+            data = {k: v for k, (i, v) in zip(names, data_tups)}
+
+            date_data = self._do_date_conversions(names, data)
+            index, column_names = self._make_index(alldata, names)
+
+        return index, column_names, date_data
+
+
+def _filter_usecols(usecols, names: SequenceT) -> SequenceT | list[Hashable]:
+    # hackish
+    usecols = evaluate_callable_usecols(usecols, names)
+    if usecols is not None and len(names) != len(usecols):
+        return [name for i, name in enumerate(names) if i in usecols or name in usecols]
+    return names
+
+
+def _concatenate_chunks(
+    chunks: list[dict[int, ArrayLike]], column_names: list[str]
+) -> dict:
+    """
+    Concatenate chunks of data read with low_memory=True.
+
+    The tricky part is handling Categoricals, where different chunks
+    may have different inferred categories.
+    """
+    names = list(chunks[0].keys())
+    warning_columns = []
+
+    result: dict = {}
+    for name in names:
+        arrs = [chunk.pop(name) for chunk in chunks]
+        # Check each arr for consistent types.
+        dtypes = {a.dtype for a in arrs}
+        non_cat_dtypes = {x for x in dtypes if not isinstance(x, CategoricalDtype)}
+
+        dtype = dtypes.pop()
+        if isinstance(dtype, CategoricalDtype):
+            result[name] = union_categoricals(arrs, sort_categories=False)
+        else:
+            result[name] = concat_compat(arrs)
+            if len(non_cat_dtypes) > 1 and result[name].dtype == np.dtype(object):
+                warning_columns.append(column_names[name])
+
+    if warning_columns:
+        warning_names = ", ".join(
+            [f"{index}: {name}" for index, name in enumerate(warning_columns)]
+        )
+        warning_message = " ".join(
+            [
+                f"Columns ({warning_names}) have mixed types. "
+                f"Specify dtype option on import or set low_memory=False."
+            ]
+        )
+        warnings.warn(warning_message, DtypeWarning, stacklevel=find_stack_level())
+    return result
+
+
+def ensure_dtype_objs(
+    dtype: DtypeArg | dict[Hashable, DtypeArg] | None,
+) -> DtypeObj | dict[Hashable, DtypeObj] | None:
+    """
+    Ensure we have either None, a dtype object, or a dictionary mapping to
+    dtype objects.
+    """
+    if isinstance(dtype, defaultdict):
+        # "None" not callable  [misc]
+        default_dtype = pandas_dtype(dtype.default_factory())  # type: ignore[misc]
+        dtype_converted: defaultdict = defaultdict(lambda: default_dtype)
+        for key in dtype.keys():
+            dtype_converted[key] = pandas_dtype(dtype[key])
+        return dtype_converted
+    elif isinstance(dtype, dict):
+        return {k: pandas_dtype(dtype[k]) for k in dtype}
+    elif dtype is not None:
+        return pandas_dtype(dtype)
+    return dtype
diff --git a/pandas/pandas/io/parsers/python_parser.py b/pandas/pandas/io/parsers/python_parser.py
new file mode 100644
index 0000000000000..e7b5c7f06a79a
--- /dev/null
+++ b/pandas/pandas/io/parsers/python_parser.py
@@ -0,0 +1,1552 @@
+from __future__ import annotations
+
+from collections import (
+    abc,
+    defaultdict,
+)
+import csv
+from io import StringIO
+import re
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+    DefaultDict,
+    Literal,
+    cast,
+    final,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+    ParserWarning,
+)
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.astype import astype_array
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_extension_array_dtype,
+    is_integer,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_string_dtype,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.inference import is_dict_like
+
+from pandas.core import algorithms
+from pandas.core.arrays import (
+    Categorical,
+    ExtensionArray,
+)
+from pandas.core.arrays.boolean import BooleanDtype
+from pandas.core.indexes.api import Index
+
+from pandas.io.common import (
+    dedup_names,
+    is_potential_multi_index,
+)
+from pandas.io.parsers.base_parser import (
+    ParserBase,
+    evaluate_callable_usecols,
+    get_na_values,
+    parser_defaults,
+    validate_parse_dates_presence,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Hashable,
+        Iterator,
+        Mapping,
+        Sequence,
+    )
+
+    from pandas._typing import (
+        ArrayLike,
+        DtypeObj,
+        ReadCsvBuffer,
+        Scalar,
+        T,
+    )
+
+    from pandas import (
+        MultiIndex,
+        Series,
+    )
+
+# BOM character (byte order mark)
+# This exists at the beginning of a file to indicate endianness
+# of a file (stream). Unfortunately, this marker screws up parsing,
+# so we need to remove it if we see it.
+_BOM = "\ufeff"
+
+
+class PythonParser(ParserBase):
+    _no_thousands_columns: set[int]
+
+    def __init__(self, f: ReadCsvBuffer[str] | list, **kwds) -> None:
+        """
+        Workhorse function for processing nested list into DataFrame
+        """
+        super().__init__(kwds)
+
+        self.data: Iterator[list[str]] | list[list[Scalar]] = []
+        self.buf: list = []
+        self.pos = 0
+        self.line_pos = 0
+
+        self.skiprows = kwds["skiprows"]
+
+        if callable(self.skiprows):
+            self.skipfunc = self.skiprows
+        else:
+            self.skipfunc = lambda x: x in self.skiprows
+
+        self.skipfooter = _validate_skipfooter_arg(kwds["skipfooter"])
+        self.delimiter = kwds["delimiter"]
+
+        self.quotechar = kwds["quotechar"]
+        if isinstance(self.quotechar, str):
+            self.quotechar = str(self.quotechar)
+
+        self.escapechar = kwds["escapechar"]
+        self.doublequote = kwds["doublequote"]
+        self.skipinitialspace = kwds["skipinitialspace"]
+        self.lineterminator = kwds["lineterminator"]
+        self.quoting = kwds["quoting"]
+        self.skip_blank_lines = kwds["skip_blank_lines"]
+
+        # Passed from read_excel
+        self.has_index_names = kwds.get("has_index_names", False)
+
+        self.thousands = kwds["thousands"]
+        self.decimal = kwds["decimal"]
+
+        self.comment = kwds["comment"]
+
+        # Set self.data to something that can read lines.
+        if isinstance(f, list):
+            # read_excel: f is a nested list, can contain non-str
+            self.data = f
+        else:
+            assert hasattr(f, "readline")
+            # yields list of str
+            self.data = self._make_reader(f)
+
+        # Get columns in two steps: infer from data, then
+        # infer column indices from self.usecols if it is specified.
+        self._col_indices: list[int] | None = None
+        columns: list[list[Scalar | None]]
+        (
+            columns,
+            self.num_original_columns,
+            self.unnamed_cols,
+        ) = self._infer_columns()
+
+        # Now self.columns has the set of columns that we will process.
+        # The original set is stored in self.original_columns.
+        # error: Cannot determine type of 'index_names'
+        (
+            self.columns,
+            self.index_names,
+            self.col_names,
+            _,
+        ) = self._extract_multi_indexer_columns(
+            columns,
+            self.index_names,  # type: ignore[has-type]
+        )
+
+        # get popped off for index
+        self.orig_names: list[Hashable] = list(self.columns)
+
+        index_names, self.orig_names, self.columns = self._get_index_name()
+        if self.index_names is None:
+            self.index_names = index_names
+
+        if self._col_indices is None:
+            self._col_indices = list(range(len(self.columns)))
+
+        self._no_thousands_columns = self._set_no_thousand_columns()
+
+        if len(self.decimal) != 1:
+            raise ValueError("Only length-1 decimal markers supported")
+
+    @cache_readonly
+    def num(self) -> re.Pattern:
+        decimal = re.escape(self.decimal)
+        if self.thousands is None:
+            regex = rf"^[\-\+]?[0-9]*({decimal}[0-9]*)?([0-9]?(E|e)\-?[0-9]+)?$"
+        else:
+            thousands = re.escape(self.thousands)
+            regex = (
+                rf"^[\-\+]?([0-9]+{thousands}|[0-9])*({decimal}[0-9]*)?"
+                rf"([0-9]?(E|e)\-?[0-9]+)?$"
+            )
+        return re.compile(regex)
+
+    def _make_reader(self, f: IO[str] | ReadCsvBuffer[str]) -> Iterator[list[str]]:
+        sep = self.delimiter
+
+        if sep is None or len(sep) == 1:
+            if self.lineterminator:
+                raise ValueError(
+                    "Custom line terminators not supported in python parser (yet)"
+                )
+
+            class MyDialect(csv.Dialect):
+                delimiter = self.delimiter
+                quotechar = self.quotechar
+                escapechar = self.escapechar
+                doublequote = self.doublequote
+                skipinitialspace = self.skipinitialspace
+                quoting = self.quoting
+                lineterminator = "\n"
+
+            dia = MyDialect
+
+            if sep is not None:
+                dia.delimiter = sep
+            else:
+                # attempt to sniff the delimiter from the first valid line,
+                # i.e. no comment line and not in skiprows
+                line = f.readline()
+                lines = self._check_comments([[line]])[0]
+                while self.skipfunc(self.pos) or not lines:
+                    self.pos += 1
+                    line = f.readline()
+                    lines = self._check_comments([[line]])[0]
+                lines_str = cast(list[str], lines)
+
+                # since `line` was a string, lines will be a list containing
+                # only a single string
+                line = lines_str[0]
+
+                self.pos += 1
+                self.line_pos += 1
+                sniffed = csv.Sniffer().sniff(line)
+                dia.delimiter = sniffed.delimiter
+
+                # Note: encoding is irrelevant here
+                line_rdr = csv.reader(StringIO(line), dialect=dia)
+                self.buf.extend(list(line_rdr))
+
+            # Note: encoding is irrelevant here
+            reader = csv.reader(f, dialect=dia, strict=True)
+
+        else:
+
+            def _read():
+                line = f.readline()
+                pat = re.compile(sep)
+
+                yield pat.split(line.strip())
+
+                for line in f:
+                    yield pat.split(line.strip())
+
+            reader = _read()
+
+        return reader
+
+    def read(
+        self, rows: int | None = None
+    ) -> tuple[
+        Index | None,
+        Sequence[Hashable] | MultiIndex,
+        Mapping[Hashable, ArrayLike | Series],
+    ]:
+        try:
+            content = self._get_lines(rows)
+        except StopIteration:
+            if self._first_chunk:
+                content = []
+            else:
+                self.close()
+                raise
+
+        # done with first read, next time raise StopIteration
+        self._first_chunk = False
+
+        index: Index | None
+        columns: Sequence[Hashable] = list(self.orig_names)
+        if not len(content):  # pragma: no cover
+            # DataFrame with the right metadata, even though it's length 0
+            # error: Cannot determine type of 'index_col'
+            names = dedup_names(
+                self.orig_names,
+                is_potential_multi_index(
+                    self.orig_names,
+                    self.index_col,  # type: ignore[has-type]
+                ),
+            )
+            index, columns, col_dict = self._get_empty_meta(
+                names,
+                self.dtype,
+            )
+            conv_columns = self._maybe_make_multi_index_columns(columns, self.col_names)
+            return index, conv_columns, col_dict
+
+        # handle new style for names in index
+        indexnamerow = None
+        if self.has_index_names and sum(
+            int(v == "" or v is None) for v in content[0]
+        ) == len(columns):
+            indexnamerow = content[0]
+            content = content[1:]
+
+        alldata = self._rows_to_cols(content)
+        data, columns = self._exclude_implicit_index(alldata)
+
+        conv_data = self._convert_data(data)
+        conv_data = self._do_date_conversions(columns, conv_data)
+
+        index, result_columns = self._make_index(alldata, columns, indexnamerow)
+
+        return index, result_columns, conv_data
+
+    def _exclude_implicit_index(
+        self,
+        alldata: list[np.ndarray],
+    ) -> tuple[Mapping[Hashable, np.ndarray], Sequence[Hashable]]:
+        # error: Cannot determine type of 'index_col'
+        names = dedup_names(
+            self.orig_names,
+            is_potential_multi_index(
+                self.orig_names,
+                self.index_col,  # type: ignore[has-type]
+            ),
+        )
+
+        offset = 0
+        if self._implicit_index:
+            # error: Cannot determine type of 'index_col'
+            offset = len(self.index_col)  # type: ignore[has-type]
+
+        len_alldata = len(alldata)
+        self._check_data_length(names, alldata)
+
+        return {
+            name: alldata[i + offset] for i, name in enumerate(names) if i < len_alldata
+        }, names
+
+    # legacy
+    def get_chunk(
+        self, size: int | None = None
+    ) -> tuple[
+        Index | None,
+        Sequence[Hashable] | MultiIndex,
+        Mapping[Hashable, ArrayLike | Series],
+    ]:
+        if size is None:
+            # error: "PythonParser" has no attribute "chunksize"
+            size = self.chunksize  # type: ignore[attr-defined]
+        return self.read(rows=size)
+
+    def _convert_data(
+        self,
+        data: Mapping[Hashable, np.ndarray],
+    ) -> Mapping[Hashable, ArrayLike]:
+        # apply converters
+        clean_conv = self._clean_mapping(self.converters)
+        clean_dtypes = self._clean_mapping(self.dtype)
+
+        # Apply NA values.
+        clean_na_values = {}
+        clean_na_fvalues = {}
+
+        if isinstance(self.na_values, dict):
+            for col in self.na_values:
+                if col is not None:
+                    na_value = self.na_values[col]
+                    na_fvalue = self.na_fvalues[col]
+
+                    if isinstance(col, int) and col not in self.orig_names:
+                        col = self.orig_names[col]
+
+                    clean_na_values[col] = na_value
+                    clean_na_fvalues[col] = na_fvalue
+        else:
+            clean_na_values = self.na_values
+            clean_na_fvalues = self.na_fvalues
+
+        return self._convert_to_ndarrays(
+            data,
+            clean_na_values,
+            clean_na_fvalues,
+            clean_conv,
+            clean_dtypes,
+        )
+
+    @final
+    def _convert_to_ndarrays(
+        self,
+        dct: Mapping,
+        na_values,
+        na_fvalues,
+        converters=None,
+        dtypes=None,
+    ) -> dict[Any, np.ndarray]:
+        result = {}
+        parse_date_cols = validate_parse_dates_presence(self.parse_dates, self.columns)
+        for c, values in dct.items():
+            conv_f = None if converters is None else converters.get(c, None)
+            if isinstance(dtypes, dict):
+                cast_type = dtypes.get(c, None)
+            else:
+                # single dtype or None
+                cast_type = dtypes
+
+            if self.na_filter:
+                col_na_values, col_na_fvalues = get_na_values(
+                    c, na_values, na_fvalues, self.keep_default_na
+                )
+            else:
+                col_na_values, col_na_fvalues = set(), set()
+
+            if c in parse_date_cols:
+                # GH#26203 Do not convert columns which get converted to dates
+                # but replace nans to ensure to_datetime works
+                mask = algorithms.isin(values, set(col_na_values) | col_na_fvalues)  # pyright: ignore[reportArgumentType]
+                np.putmask(values, mask, np.nan)
+                result[c] = values
+                continue
+
+            if conv_f is not None:
+                # conv_f applied to data before inference
+                if cast_type is not None:
+                    warnings.warn(
+                        (
+                            "Both a converter and dtype were specified "
+                            f"for column {c} - only the converter will be used."
+                        ),
+                        ParserWarning,
+                        stacklevel=find_stack_level(),
+                    )
+
+                try:
+                    values = lib.map_infer(values, conv_f)
+                except ValueError:
+                    mask = algorithms.isin(values, list(na_values)).view(np.uint8)
+                    values = lib.map_infer_mask(values, conv_f, mask)
+
+                cvals, na_count = self._infer_types(
+                    values,
+                    set(col_na_values) | col_na_fvalues,
+                    cast_type is None,
+                    try_num_bool=False,
+                )
+            else:
+                is_ea = is_extension_array_dtype(cast_type)
+                is_str_or_ea_dtype = is_ea or is_string_dtype(cast_type)
+                # skip inference if specified dtype is object
+                # or casting to an EA
+                try_num_bool = not (cast_type and is_str_or_ea_dtype)
+
+                # general type inference and conversion
+                cvals, na_count = self._infer_types(
+                    values,
+                    set(col_na_values) | col_na_fvalues,
+                    cast_type is None,
+                    try_num_bool,
+                )
+
+                # type specified in dtype param or cast_type is an EA
+                if cast_type is not None:
+                    cast_type = pandas_dtype(cast_type)
+                if cast_type and (cvals.dtype != cast_type or is_ea):
+                    if not is_ea and na_count > 0:
+                        if is_bool_dtype(cast_type):
+                            raise ValueError(f"Bool column has NA values in column {c}")
+                    cvals = self._cast_types(cvals, cast_type, c)
+
+            result[c] = cvals
+        return result
+
+    @final
+    def _cast_types(self, values: ArrayLike, cast_type: DtypeObj, column) -> ArrayLike:
+        """
+        Cast values to specified type
+
+        Parameters
+        ----------
+        values : ndarray or ExtensionArray
+        cast_type : np.dtype or ExtensionDtype
+           dtype to cast values to
+        column : string
+            column name - used only for error reporting
+
+        Returns
+        -------
+        converted : ndarray or ExtensionArray
+        """
+        if isinstance(cast_type, CategoricalDtype):
+            known_cats = cast_type.categories is not None
+
+            if not is_object_dtype(values.dtype) and not known_cats:
+                # TODO: this is for consistency with
+                # c-parser which parses all categories
+                # as strings
+                values = lib.ensure_string_array(
+                    values, skipna=False, convert_na_value=False
+                )
+
+            cats = Index(values).unique().dropna()
+            values = Categorical._from_inferred_categories(
+                cats, cats.get_indexer(values), cast_type, true_values=self.true_values
+            )
+
+        # use the EA's implementation of casting
+        elif isinstance(cast_type, ExtensionDtype):
+            array_type = cast_type.construct_array_type()
+            try:
+                if isinstance(cast_type, BooleanDtype):
+                    # error: Unexpected keyword argument "true_values" for
+                    # "_from_sequence_of_strings" of "ExtensionArray"
+                    values_str = [str(val) for val in values]
+                    return array_type._from_sequence_of_strings(  # type: ignore[call-arg]
+                        values_str,
+                        dtype=cast_type,
+                        true_values=self.true_values,  # pyright: ignore[reportCallIssue]
+                        false_values=self.false_values,  # pyright: ignore[reportCallIssue]
+                        none_values=self.na_values,  # pyright: ignore[reportCallIssue]
+                    )
+                else:
+                    return array_type._from_sequence_of_strings(values, dtype=cast_type)
+            except NotImplementedError as err:
+                raise NotImplementedError(
+                    f"Extension Array: {array_type} must implement "
+                    "_from_sequence_of_strings in order to be used in parser methods"
+                ) from err
+
+        elif isinstance(values, ExtensionArray):
+            values = values.astype(cast_type, copy=False)
+        elif issubclass(cast_type.type, str):
+            # TODO: why skipna=True here and False above? some tests depend
+            #  on it here, but nothing fails if we change it above
+            #  (as no tests get there as of 2022-12-06)
+            values = lib.ensure_string_array(
+                values, skipna=True, convert_na_value=False
+            )
+        else:
+            try:
+                values = astype_array(values, cast_type, copy=True)
+            except ValueError as err:
+                raise ValueError(
+                    f"Unable to convert column {column} to type {cast_type}"
+                ) from err
+        return values
+
+    @cache_readonly
+    def _have_mi_columns(self) -> bool:
+        if self.header is None:
+            return False
+
+        header = self.header
+        if isinstance(header, (list, tuple, np.ndarray)):
+            return len(header) > 1
+        else:
+            return False
+
+    def _infer_columns(
+        self,
+    ) -> tuple[list[list[Scalar | None]], int, set[Scalar | None]]:
+        names = self.names
+        num_original_columns = 0
+        clear_buffer = True
+        unnamed_cols: set[Scalar | None] = set()
+
+        if self.header is not None:
+            header = self.header
+            have_mi_columns = self._have_mi_columns
+
+            if isinstance(header, (list, tuple, np.ndarray)):
+                # we have a mi columns, so read an extra line
+                if have_mi_columns:
+                    header = list(header) + [header[-1] + 1]
+            else:
+                header = [header]
+
+            columns: list[list[Scalar | None]] = []
+            for level, hr in enumerate(header):
+                try:
+                    line = self._buffered_line()
+
+                    while self.line_pos <= hr:
+                        line = self._next_line()
+
+                except StopIteration as err:
+                    if 0 < self.line_pos <= hr and (
+                        not have_mi_columns or hr != header[-1]
+                    ):
+                        # If no rows we want to raise a different message and if
+                        # we have mi columns, the last line is not part of the header
+                        joi = list(map(str, header[:-1] if have_mi_columns else header))
+                        msg = f"[{','.join(joi)}], len of {len(joi)}, "
+                        raise ValueError(
+                            f"Passed header={msg}but only {self.line_pos} lines in file"
+                        ) from err
+
+                    # We have an empty file, so check
+                    # if columns are provided. That will
+                    # serve as the 'line' for parsing
+                    if have_mi_columns and hr > 0:
+                        if clear_buffer:
+                            self.buf.clear()
+                        columns.append([None] * len(columns[-1]))
+                        return columns, num_original_columns, unnamed_cols
+
+                    if not self.names:
+                        raise EmptyDataError("No columns to parse from file") from err
+
+                    line = self.names[:]
+
+                this_columns: list[Scalar | None] = []
+                this_unnamed_cols = []
+
+                for i, c in enumerate(line):
+                    if c == "":
+                        if have_mi_columns:
+                            col_name = f"Unnamed: {i}_level_{level}"
+                        else:
+                            col_name = f"Unnamed: {i}"
+
+                        this_unnamed_cols.append(i)
+                        this_columns.append(col_name)
+                    else:
+                        this_columns.append(c)
+
+                if not have_mi_columns:
+                    counts: DefaultDict = defaultdict(int)
+                    # Ensure that regular columns are used before unnamed ones
+                    # to keep given names and mangle unnamed columns
+                    col_loop_order = [
+                        i
+                        for i in range(len(this_columns))
+                        if i not in this_unnamed_cols
+                    ] + this_unnamed_cols
+
+                    # TODO: Use pandas.io.common.dedup_names instead (see #50371)
+                    for i in col_loop_order:
+                        col = this_columns[i]
+                        old_col = col
+                        cur_count = counts[col]
+
+                        if cur_count > 0:
+                            while cur_count > 0:
+                                counts[old_col] = cur_count + 1
+                                col = f"{old_col}.{cur_count}"
+                                if col in this_columns:
+                                    cur_count += 1
+                                else:
+                                    cur_count = counts[col]
+
+                            if (
+                                self.dtype is not None
+                                and is_dict_like(self.dtype)
+                                and self.dtype.get(old_col) is not None
+                                and self.dtype.get(col) is None
+                            ):
+                                self.dtype.update({col: self.dtype.get(old_col)})
+                        this_columns[i] = col
+                        counts[col] = cur_count + 1
+                elif have_mi_columns:
+                    # if we have grabbed an extra line, but its not in our
+                    # format so save in the buffer, and create an blank extra
+                    # line for the rest of the parsing code
+                    if hr == header[-1]:
+                        lc = len(this_columns)
+                        # error: Cannot determine type of 'index_col'
+                        sic = self.index_col  # type: ignore[has-type]
+                        ic = len(sic) if sic is not None else 0
+                        unnamed_count = len(this_unnamed_cols)
+
+                        # if wrong number of blanks or no index, not our format
+                        if (lc != unnamed_count and lc - ic > unnamed_count) or ic == 0:
+                            clear_buffer = False
+                            this_columns = [None] * lc
+                            self.buf = [self.buf[-1]]
+
+                columns.append(this_columns)
+                unnamed_cols.update({this_columns[i] for i in this_unnamed_cols})
+
+                if len(columns) == 1:
+                    num_original_columns = len(this_columns)
+
+            if clear_buffer:
+                self.buf.clear()
+
+            first_line: list[Scalar] | None
+            if names is not None:
+                # Read first row after header to check if data are longer
+                try:
+                    first_line = self._next_line()
+                except StopIteration:
+                    first_line = None
+
+                len_first_data_row = 0 if first_line is None else len(first_line)
+
+                if len(names) > len(columns[0]) and len(names) > len_first_data_row:
+                    raise ValueError(
+                        "Number of passed names did not match "
+                        "number of header fields in the file"
+                    )
+                if len(columns) > 1:
+                    raise TypeError("Cannot pass names with multi-index columns")
+
+                if self.usecols is not None:
+                    # Set _use_cols. We don't store columns because they are
+                    # overwritten.
+                    self._handle_usecols(columns, names, num_original_columns)
+                else:
+                    num_original_columns = len(names)
+                if self._col_indices is not None and len(names) != len(
+                    self._col_indices
+                ):
+                    columns = [[names[i] for i in sorted(self._col_indices)]]
+                else:
+                    columns = [names]
+            else:
+                columns = self._handle_usecols(
+                    columns, columns[0], num_original_columns
+                )
+        else:
+            ncols = len(self._header_line)
+            num_original_columns = ncols
+
+            if not names:
+                columns = [list(range(ncols))]
+                columns = self._handle_usecols(columns, columns[0], ncols)
+            elif self.usecols is None or len(names) >= ncols:
+                columns = self._handle_usecols([names], names, ncols)
+                num_original_columns = len(names)
+            elif not callable(self.usecols) and len(names) != len(self.usecols):
+                raise ValueError(
+                    "Number of passed names did not match number of "
+                    "header fields in the file"
+                )
+            else:
+                # Ignore output but set used columns.
+                columns = [names]
+                self._handle_usecols(columns, columns[0], ncols)
+
+        return columns, num_original_columns, unnamed_cols
+
+    @cache_readonly
+    def _header_line(self):
+        # Store line for reuse in _get_index_name
+        if self.header is not None:
+            return None
+
+        try:
+            line = self._buffered_line()
+        except StopIteration as err:
+            if not self.names:
+                raise EmptyDataError("No columns to parse from file") from err
+
+            line = self.names[:]
+        return line
+
+    def _handle_usecols(
+        self,
+        columns: list[list[Scalar | None]],
+        usecols_key: list[Scalar | None],
+        num_original_columns: int,
+    ) -> list[list[Scalar | None]]:
+        """
+        Sets self._col_indices
+
+        usecols_key is used if there are string usecols.
+        """
+        col_indices: set[int] | list[int]
+        if self.usecols is not None:
+            if callable(self.usecols):
+                col_indices = evaluate_callable_usecols(self.usecols, usecols_key)
+            elif any(isinstance(u, str) for u in self.usecols):
+                if len(columns) > 1:
+                    raise ValueError(
+                        "If using multiple headers, usecols must be integers."
+                    )
+                col_indices = []
+
+                for col in self.usecols:
+                    if isinstance(col, str):
+                        try:
+                            col_indices.append(usecols_key.index(col))
+                        except ValueError:
+                            self._validate_usecols_names(self.usecols, usecols_key)
+                    else:
+                        col_indices.append(col)
+            else:
+                missing_usecols = [
+                    col for col in self.usecols if col >= num_original_columns
+                ]
+                if missing_usecols:
+                    raise ParserError(
+                        "Defining usecols with out-of-bounds indices is not allowed. "
+                        f"{missing_usecols} are out-of-bounds.",
+                    )
+                col_indices = self.usecols
+
+            columns = [
+                [n for i, n in enumerate(column) if i in col_indices]
+                for column in columns
+            ]
+            self._col_indices = sorted(col_indices)
+        return columns
+
+    def _buffered_line(self) -> list[Scalar]:
+        """
+        Return a line from buffer, filling buffer if required.
+        """
+        if len(self.buf) > 0:
+            return self.buf[0]
+        else:
+            return self._next_line()
+
+    def _check_for_bom(self, first_row: list[Scalar]) -> list[Scalar]:
+        """
+        Checks whether the file begins with the BOM character.
+        If it does, remove it. In addition, if there is quoting
+        in the field subsequent to the BOM, remove it as well
+        because it technically takes place at the beginning of
+        the name, not the middle of it.
+        """
+        # first_row will be a list, so we need to check
+        # that that list is not empty before proceeding.
+        if not first_row:
+            return first_row
+
+        # The first element of this row is the one that could have the
+        # BOM that we want to remove. Check that the first element is a
+        # string before proceeding.
+        if not isinstance(first_row[0], str):
+            return first_row
+
+        # Check that the string is not empty, as that would
+        # obviously not have a BOM at the start of it.
+        if not first_row[0]:
+            return first_row
+
+        # Since the string is non-empty, check that it does
+        # in fact begin with a BOM.
+        first_elt = first_row[0][0]
+        if first_elt != _BOM:
+            return first_row
+
+        first_row_bom = first_row[0]
+        new_row: str
+
+        if len(first_row_bom) > 1 and first_row_bom[1] == self.quotechar:
+            start = 2
+            quote = first_row_bom[1]
+            end = first_row_bom[2:].index(quote) + 2
+
+            # Extract the data between the quotation marks
+            new_row = first_row_bom[start:end]
+
+            # Extract any remaining data after the second
+            # quotation mark.
+            if len(first_row_bom) > end + 1:
+                new_row += first_row_bom[end + 1 :]
+
+        else:
+            # No quotation so just remove BOM from first element
+            new_row = first_row_bom[1:]
+
+        new_row_list: list[Scalar] = [new_row]
+        return new_row_list + first_row[1:]
+
+    def _is_line_empty(self, line: Sequence[Scalar]) -> bool:
+        """
+        Check if a line is empty or not.
+
+        Parameters
+        ----------
+        line : str, array-like
+            The line of data to check.
+
+        Returns
+        -------
+        boolean : Whether or not the line is empty.
+        """
+        return not line or all(not x for x in line)
+
+    def _next_line(self) -> list[Scalar]:
+        if isinstance(self.data, list):
+            while self.skipfunc(self.pos):
+                if self.pos >= len(self.data):
+                    break
+                self.pos += 1
+
+            while True:
+                try:
+                    line = self._check_comments([self.data[self.pos]])[0]
+                    self.pos += 1
+                    # either uncommented or blank to begin with
+                    if not self.skip_blank_lines and (
+                        self._is_line_empty(self.data[self.pos - 1]) or line
+                    ):
+                        break
+                    if self.skip_blank_lines:
+                        ret = self._remove_empty_lines([line])
+                        if ret:
+                            line = ret[0]
+                            break
+                except IndexError as err:
+                    raise StopIteration from err
+        else:
+            while self.skipfunc(self.pos):
+                self.pos += 1
+                next(self.data)
+
+            while True:
+                orig_line = self._next_iter_line(row_num=self.pos + 1)
+                self.pos += 1
+
+                if orig_line is not None:
+                    line = self._check_comments([orig_line])[0]
+
+                    if self.skip_blank_lines:
+                        ret = self._remove_empty_lines([line])
+
+                        if ret:
+                            line = ret[0]
+                            break
+                    elif self._is_line_empty(orig_line) or line:
+                        break
+
+        # This was the first line of the file,
+        # which could contain the BOM at the
+        # beginning of it.
+        if self.pos == 1:
+            line = self._check_for_bom(line)
+
+        self.line_pos += 1
+        self.buf.append(line)
+        return line
+
+    def _alert_malformed(self, msg: str, row_num: int) -> None:
+        """
+        Alert a user about a malformed row, depending on value of
+        `self.on_bad_lines` enum.
+
+        If `self.on_bad_lines` is ERROR, the alert will be `ParserError`.
+        If `self.on_bad_lines` is WARN, the alert will be printed out.
+
+        Parameters
+        ----------
+        msg: str
+            The error message to display.
+        row_num: int
+            The row number where the parsing error occurred.
+            Because this row number is displayed, we 1-index,
+            even though we 0-index internally.
+        """
+        if self.on_bad_lines == self.BadLineHandleMethod.ERROR:
+            raise ParserError(msg)
+        if self.on_bad_lines == self.BadLineHandleMethod.WARN:
+            warnings.warn(
+                f"Skipping line {row_num}: {msg}\n",
+                ParserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+    def _next_iter_line(self, row_num: int) -> list[Scalar] | None:
+        """
+        Wrapper around iterating through `self.data` (CSV source).
+
+        When a CSV error is raised, we check for specific
+        error messages that allow us to customize the
+        error message displayed to the user.
+
+        Parameters
+        ----------
+        row_num: int
+            The row number of the line being parsed.
+        """
+        try:
+            assert not isinstance(self.data, list)
+            line = next(self.data)
+            # lie about list[str] vs list[Scalar] to minimize ignores
+            return line  # type: ignore[return-value]
+        except csv.Error as e:
+            if self.on_bad_lines in (
+                self.BadLineHandleMethod.ERROR,
+                self.BadLineHandleMethod.WARN,
+            ):
+                msg = str(e)
+
+                if "NULL byte" in msg or "line contains NUL" in msg:
+                    msg = (
+                        "NULL byte detected. This byte "
+                        "cannot be processed in Python's "
+                        "native csv library at the moment, "
+                        "so please pass in engine='c' instead"
+                    )
+
+                if self.skipfooter > 0:
+                    reason = (
+                        "Error could possibly be due to "
+                        "parsing errors in the skipped footer rows "
+                        "(the skipfooter keyword is only applied "
+                        "after Python's csv library has parsed "
+                        "all rows)."
+                    )
+                    msg += ". " + reason
+
+                self._alert_malformed(msg, row_num)
+            return None
+
+    def _check_comments(self, lines: list[list[Scalar]]) -> list[list[Scalar]]:
+        if self.comment is None:
+            return lines
+        ret = []
+        for line in lines:
+            rl = []
+            for x in line:
+                if (
+                    not isinstance(x, str)
+                    or self.comment not in x
+                    or x in self.na_values
+                ):
+                    rl.append(x)
+                else:
+                    x = x[: x.find(self.comment)]
+                    if len(x) > 0:
+                        rl.append(x)
+                    break
+            ret.append(rl)
+        return ret
+
+    def _remove_empty_lines(self, lines: list[list[T]]) -> list[list[T]]:
+        """
+        Iterate through the lines and remove any that are
+        either empty or contain only one whitespace value
+
+        Parameters
+        ----------
+        lines : list of list of Scalars
+            The array of lines that we are to filter.
+
+        Returns
+        -------
+        filtered_lines : list of list of Scalars
+            The same array of lines with the "empty" ones removed.
+        """
+        # Remove empty lines and lines with only one whitespace value
+        ret = [
+            line
+            for line in lines
+            if (
+                len(line) > 1
+                or (
+                    len(line) == 1 and (not isinstance(line[0], str) or line[0].strip())
+                )
+            )
+        ]
+        return ret
+
+    def _check_thousands(self, lines: list[list[Scalar]]) -> list[list[Scalar]]:
+        if self.thousands is None:
+            return lines
+
+        return self._search_replace_num_columns(
+            lines=lines, search=self.thousands, replace=""
+        )
+
+    def _search_replace_num_columns(
+        self, lines: list[list[Scalar]], search: str, replace: str
+    ) -> list[list[Scalar]]:
+        ret = []
+        for line in lines:
+            rl = []
+            for i, x in enumerate(line):
+                if (
+                    not isinstance(x, str)
+                    or search not in x
+                    or i in self._no_thousands_columns
+                    or not self.num.search(x.strip())
+                ):
+                    rl.append(x)
+                else:
+                    rl.append(x.replace(search, replace))
+            ret.append(rl)
+        return ret
+
+    def _check_decimal(self, lines: list[list[Scalar]]) -> list[list[Scalar]]:
+        if self.decimal == parser_defaults["decimal"]:
+            return lines
+
+        return self._search_replace_num_columns(
+            lines=lines, search=self.decimal, replace="."
+        )
+
+    def _get_index_name(
+        self,
+    ) -> tuple[Sequence[Hashable] | None, list[Hashable], list[Hashable]]:
+        """
+        Try several cases to get lines:
+
+        0) There are headers on row 0 and row 1 and their
+        total summed lengths equals the length of the next line.
+        Treat row 0 as columns and row 1 as indices
+        1) Look for implicit index: there are more columns
+        on row 1 than row 0. If this is true, assume that row
+        1 lists index columns and row 0 lists normal columns.
+        2) Get index from the columns if it was listed.
+        """
+        columns: Sequence[Hashable] = self.orig_names
+        orig_names = list(columns)
+        columns = list(columns)
+
+        line: list[Scalar] | None
+        if self._header_line is not None:
+            line = self._header_line
+        else:
+            try:
+                line = self._next_line()
+            except StopIteration:
+                line = None
+
+        next_line: list[Scalar] | None
+        try:
+            next_line = self._next_line()
+        except StopIteration:
+            next_line = None
+
+        # implicitly index_col=0 b/c 1 fewer column names
+        implicit_first_cols = 0
+        if line is not None:
+            # leave it 0, #2442
+            # Case 1
+            # error: Cannot determine type of 'index_col'
+            index_col = self.index_col  # type: ignore[has-type]
+            if index_col is not False:
+                implicit_first_cols = len(line) - self.num_original_columns
+
+            # Case 0
+            if (
+                next_line is not None
+                and self.header is not None
+                and index_col is not False
+            ):
+                if len(next_line) == len(line) + self.num_original_columns:
+                    # column and index names on diff rows
+                    self.index_col = list(range(len(line)))
+                    self.buf = self.buf[1:]
+
+                    for c in reversed(line):
+                        columns.insert(0, c)
+
+                    # Update list of original names to include all indices.
+                    orig_names = list(columns)
+                    self.num_original_columns = len(columns)
+                    return line, orig_names, columns
+
+        if implicit_first_cols > 0:
+            # Case 1
+            self._implicit_index = True
+            if self.index_col is None:
+                self.index_col = list(range(implicit_first_cols))
+
+            index_name = None
+
+        else:
+            # Case 2
+            (index_name, _, self.index_col) = self._clean_index_names(
+                columns, self.index_col
+            )
+
+        return index_name, orig_names, columns
+
+    def _rows_to_cols(self, content: list[list[Scalar]]) -> list[np.ndarray]:
+        col_len = self.num_original_columns
+
+        if self._implicit_index:
+            col_len += len(self.index_col)
+
+        max_len = max(len(row) for row in content)
+
+        # Check that there are no rows with too many
+        # elements in their row (rows with too few
+        # elements are padded with NaN).
+        # error: Non-overlapping identity check (left operand type: "List[int]",
+        # right operand type: "Literal[False]")
+        if (
+            max_len > col_len
+            and self.index_col is not False  # type: ignore[comparison-overlap]
+            and self.usecols is None
+        ):
+            footers = self.skipfooter if self.skipfooter else 0
+            bad_lines = []
+
+            iter_content = enumerate(content)
+            content_len = len(content)
+            content = []
+
+            for i, _content in iter_content:
+                actual_len = len(_content)
+
+                if actual_len > col_len:
+                    if callable(self.on_bad_lines):
+                        new_l = self.on_bad_lines(_content)
+                        if new_l is not None:
+                            content.append(new_l)  # pyright: ignore[reportArgumentType]
+                    elif self.on_bad_lines in (
+                        self.BadLineHandleMethod.ERROR,
+                        self.BadLineHandleMethod.WARN,
+                    ):
+                        row_num = self.pos - (content_len - i + footers)
+                        bad_lines.append((row_num, actual_len))
+
+                        if self.on_bad_lines == self.BadLineHandleMethod.ERROR:
+                            break
+                else:
+                    content.append(_content)
+
+            for row_num, actual_len in bad_lines:
+                msg = (
+                    f"Expected {col_len} fields in line {row_num + 1}, saw {actual_len}"
+                )
+                if (
+                    self.delimiter
+                    and len(self.delimiter) > 1
+                    and self.quoting != csv.QUOTE_NONE
+                ):
+                    # see gh-13374
+                    reason = (
+                        "Error could possibly be due to quotes being "
+                        "ignored when a multi-char delimiter is used."
+                    )
+                    msg += ". " + reason
+
+                self._alert_malformed(msg, row_num + 1)
+
+        # see gh-13320
+        zipped_content = list(lib.to_object_array(content, min_width=col_len).T)
+
+        if self.usecols:
+            assert self._col_indices is not None
+            col_indices = self._col_indices
+
+            if self._implicit_index:
+                zipped_content = [
+                    a
+                    for i, a in enumerate(zipped_content)
+                    if (
+                        i < len(self.index_col)
+                        or i - len(self.index_col) in col_indices
+                    )
+                ]
+            else:
+                zipped_content = [
+                    a for i, a in enumerate(zipped_content) if i in col_indices
+                ]
+        return zipped_content
+
+    def _get_lines(self, rows: int | None = None) -> list[list[Scalar]]:
+        lines = self.buf
+        new_rows = None
+
+        # already fetched some number
+        if rows is not None:
+            # we already have the lines in the buffer
+            if len(self.buf) >= rows:
+                new_rows, self.buf = self.buf[:rows], self.buf[rows:]
+
+            # need some lines
+            else:
+                rows -= len(self.buf)
+
+        if new_rows is None:
+            if isinstance(self.data, list):
+                if self.pos > len(self.data):
+                    raise StopIteration
+                if rows is None:
+                    new_rows = self.data[self.pos :]
+                    new_pos = len(self.data)
+                else:
+                    new_rows = self.data[self.pos : self.pos + rows]
+                    new_pos = self.pos + rows
+
+                new_rows = self._remove_skipped_rows(new_rows)
+                lines.extend(new_rows)
+                self.pos = new_pos
+
+            else:
+                new_rows = []
+                try:
+                    if rows is not None:
+                        row_index = 0
+                        row_ct = 0
+                        offset = self.pos if self.pos is not None else 0
+                        while row_ct < rows:
+                            new_row = next(self.data)
+                            if not self.skipfunc(offset + row_index):
+                                row_ct += 1
+                            row_index += 1
+                            new_rows.append(new_row)
+
+                        len_new_rows = len(new_rows)
+                        new_rows = self._remove_skipped_rows(new_rows)
+                        lines.extend(new_rows)
+                    else:
+                        rows = 0
+
+                        while True:
+                            next_row = self._next_iter_line(row_num=self.pos + rows + 1)
+                            rows += 1
+
+                            if next_row is not None:
+                                new_rows.append(next_row)
+                        len_new_rows = len(new_rows)
+
+                except StopIteration:
+                    len_new_rows = len(new_rows)
+                    new_rows = self._remove_skipped_rows(new_rows)
+                    lines.extend(new_rows)
+                    if len(lines) == 0:
+                        raise
+                self.pos += len_new_rows
+
+            self.buf = []
+        else:
+            lines = new_rows
+
+        if self.skipfooter:
+            lines = lines[: -self.skipfooter]
+
+        lines = self._check_comments(lines)
+        if self.skip_blank_lines:
+            lines = self._remove_empty_lines(lines)
+        lines = self._check_thousands(lines)
+        return self._check_decimal(lines)
+
+    def _remove_skipped_rows(self, new_rows: list[list[Scalar]]) -> list[list[Scalar]]:
+        if self.skiprows:
+            return [
+                row for i, row in enumerate(new_rows) if not self.skipfunc(i + self.pos)
+            ]
+        return new_rows
+
+    def _set_no_thousand_columns(self) -> set[int]:
+        no_thousands_columns: set[int] = set()
+        if self.columns and self.parse_dates:
+            assert self._col_indices is not None
+            no_thousands_columns = self._set_noconvert_dtype_columns(
+                self._col_indices, self.columns
+            )
+        if self.columns and self.dtype:
+            assert self._col_indices is not None
+            for i, col in zip(self._col_indices, self.columns):
+                if not isinstance(self.dtype, dict) and not is_numeric_dtype(
+                    self.dtype
+                ):
+                    no_thousands_columns.add(i)
+                if (
+                    isinstance(self.dtype, dict)
+                    and col in self.dtype
+                    and (
+                        not is_numeric_dtype(self.dtype[col])
+                        or is_bool_dtype(self.dtype[col])
+                    )
+                ):
+                    no_thousands_columns.add(i)
+        return no_thousands_columns
+
+
+class FixedWidthReader(abc.Iterator):
+    """
+    A reader of fixed-width lines.
+    """
+
+    def __init__(
+        self,
+        f: IO[str] | ReadCsvBuffer[str],
+        colspecs: list[tuple[int, int]] | Literal["infer"],
+        delimiter: str | None,
+        comment: str | None,
+        skiprows: set[int] | None = None,
+        infer_nrows: int = 100,
+    ) -> None:
+        self.f = f
+        self.buffer: Iterator | None = None
+        self.delimiter = "\r\n" + delimiter if delimiter else "\n\r\t "
+        self.comment = comment
+        if colspecs == "infer":
+            self.colspecs = self.detect_colspecs(
+                infer_nrows=infer_nrows, skiprows=skiprows
+            )
+        else:
+            self.colspecs = colspecs
+
+        if not isinstance(self.colspecs, (tuple, list)):
+            raise TypeError(
+                "column specifications must be a list or tuple, "
+                f"input was a {type(colspecs).__name__}"
+            )
+
+        for colspec in self.colspecs:
+            if not (
+                isinstance(colspec, (tuple, list))
+                and len(colspec) == 2
+                and isinstance(colspec[0], (int, np.integer, type(None)))
+                and isinstance(colspec[1], (int, np.integer, type(None)))
+            ):
+                raise TypeError(
+                    "Each column specification must be "
+                    "2 element tuple or list of integers"
+                )
+
+    def get_rows(self, infer_nrows: int, skiprows: set[int] | None = None) -> list[str]:
+        """
+        Read rows from self.f, skipping as specified.
+
+        We distinguish buffer_rows (the first <= infer_nrows
+        lines) from the rows returned to detect_colspecs
+        because it's simpler to leave the other locations
+        with skiprows logic alone than to modify them to
+        deal with the fact we skipped some rows here as
+        well.
+
+        Parameters
+        ----------
+        infer_nrows : int
+            Number of rows to read from self.f, not counting
+            rows that are skipped.
+        skiprows: set, optional
+            Indices of rows to skip.
+
+        Returns
+        -------
+        detect_rows : list of str
+            A list containing the rows to read.
+
+        """
+        if skiprows is None:
+            skiprows = set()
+        buffer_rows = []
+        detect_rows = []
+        for i, row in enumerate(self.f):
+            if i not in skiprows:
+                detect_rows.append(row)
+            buffer_rows.append(row)
+            if len(detect_rows) >= infer_nrows:
+                break
+        self.buffer = iter(buffer_rows)
+        return detect_rows
+
+    def detect_colspecs(
+        self, infer_nrows: int = 100, skiprows: set[int] | None = None
+    ) -> list[tuple[int, int]]:
+        # Regex escape the delimiters
+        delimiters = "".join([rf"\{x}" for x in self.delimiter])
+        pattern = re.compile(f"([^{delimiters}]+)")
+        rows = self.get_rows(infer_nrows, skiprows)
+        if not rows:
+            raise EmptyDataError("No rows from which to infer column width")
+        max_len = max(map(len, rows))
+        mask = np.zeros(max_len + 1, dtype=int)
+        if self.comment is not None:
+            rows = [row.partition(self.comment)[0] for row in rows]
+        for row in rows:
+            for m in pattern.finditer(row):
+                mask[m.start() : m.end()] = 1
+        shifted = np.roll(mask, 1)
+        shifted[0] = 0
+        edges = np.where((mask ^ shifted) == 1)[0]
+        edge_pairs = list(zip(edges[::2], edges[1::2]))
+        return edge_pairs
+
+    def __next__(self) -> list[str]:
+        # Argument 1 to "next" has incompatible type "Union[IO[str],
+        # ReadCsvBuffer[str]]"; expected "SupportsNext[str]"
+        if self.buffer is not None:
+            try:
+                line = next(self.buffer)
+            except StopIteration:
+                self.buffer = None
+                line = next(self.f)  # type: ignore[arg-type]
+        else:
+            line = next(self.f)  # type: ignore[arg-type]
+        # Note: 'colspecs' is a sequence of half-open intervals.
+        return [line[from_:to].strip(self.delimiter) for (from_, to) in self.colspecs]
+
+
+class FixedWidthFieldParser(PythonParser):
+    """
+    Specialization that Converts fixed-width fields into DataFrames.
+    See PythonParser for details.
+    """
+
+    def __init__(self, f: ReadCsvBuffer[str], **kwds) -> None:
+        # Support iterators, convert to a list.
+        self.colspecs = kwds.pop("colspecs")
+        self.infer_nrows = kwds.pop("infer_nrows")
+        PythonParser.__init__(self, f, **kwds)
+
+    def _make_reader(self, f: IO[str] | ReadCsvBuffer[str]) -> FixedWidthReader:
+        return FixedWidthReader(
+            f,
+            self.colspecs,
+            self.delimiter,
+            self.comment,
+            self.skiprows,
+            self.infer_nrows,
+        )
+
+    def _remove_empty_lines(self, lines: list[list[T]]) -> list[list[T]]:
+        """
+        Returns the list of lines without the empty ones. With fixed-width
+        fields, empty lines become arrays of empty strings.
+
+        See PythonParser._remove_empty_lines.
+        """
+        return [
+            line
+            for line in lines
+            if any(not isinstance(e, str) or e.strip() for e in line)
+        ]
+
+
+def _validate_skipfooter_arg(skipfooter: int) -> int:
+    """
+    Validate the 'skipfooter' parameter.
+
+    Checks whether 'skipfooter' is a non-negative integer.
+    Raises a ValueError if that is not the case.
+
+    Parameters
+    ----------
+    skipfooter : non-negative integer
+        The number of rows to skip at the end of the file.
+
+    Returns
+    -------
+    validated_skipfooter : non-negative integer
+        The original input if the validation succeeds.
+
+    Raises
+    ------
+    ValueError : 'skipfooter' was not a non-negative integer.
+    """
+    if not is_integer(skipfooter):
+        raise ValueError("skipfooter must be an integer")
+
+    if skipfooter < 0:
+        raise ValueError("skipfooter cannot be negative")
+
+    # Incompatible return value type (got "Union[int, integer[Any]]", expected "int")
+    return skipfooter  # type: ignore[return-value]
diff --git a/pandas/pandas/io/parsers/readers.py b/pandas/pandas/io/parsers/readers.py
new file mode 100644
index 0000000000000..67193f930b4dc
--- /dev/null
+++ b/pandas/pandas/io/parsers/readers.py
@@ -0,0 +1,1938 @@
+"""
+Module contains tools for processing files into DataFrames or other objects
+
+GH#48849 provides a convenient way of deprecating keyword arguments
+"""
+
+from __future__ import annotations
+
+from collections import (
+    abc,
+    defaultdict,
+)
+import csv
+import sys
+from textwrap import fill
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    Any,
+    Generic,
+    Literal,
+    TypedDict,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.parsers import STR_NA_VALUES
+from pandas.errors import (
+    AbstractMethodError,
+    ParserWarning,
+)
+from pandas.util._decorators import (
+    Appender,
+    set_module,
+)
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import (
+    is_file_like,
+    is_float,
+    is_integer,
+    is_list_like,
+    pandas_dtype,
+)
+
+from pandas import Series
+from pandas.core.frame import DataFrame
+from pandas.core.indexes.api import RangeIndex
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import (
+    IOHandles,
+    get_handle,
+    stringify_path,
+    validate_header_arg,
+)
+from pandas.io.parsers.arrow_parser_wrapper import ArrowParserWrapper
+from pandas.io.parsers.base_parser import (
+    ParserBase,
+    is_index_col,
+    parser_defaults,
+)
+from pandas.io.parsers.c_parser_wrapper import CParserWrapper
+from pandas.io.parsers.python_parser import (
+    FixedWidthFieldParser,
+    PythonParser,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterable,
+        Mapping,
+        Sequence,
+    )
+    from types import TracebackType
+
+    from pandas._typing import (
+        CompressionOptions,
+        CSVEngine,
+        DtypeArg,
+        DtypeBackend,
+        FilePath,
+        HashableT,
+        IndexLabel,
+        ReadCsvBuffer,
+        Self,
+        StorageOptions,
+        Unpack,
+        UsecolsArgType,
+    )
+
+    class _read_shared(TypedDict, Generic[HashableT], total=False):
+        # annotations shared between read_csv/fwf/table's overloads
+        # NOTE: Keep in sync with the annotations of the implementation
+        sep: str | None | lib.NoDefault
+        delimiter: str | None | lib.NoDefault
+        header: int | Sequence[int] | None | Literal["infer"]
+        names: Sequence[Hashable] | None | lib.NoDefault
+        index_col: IndexLabel | Literal[False] | None
+        usecols: UsecolsArgType
+        dtype: DtypeArg | None
+        engine: CSVEngine | None
+        converters: Mapping[HashableT, Callable] | None
+        true_values: list | None
+        false_values: list | None
+        skipinitialspace: bool
+        skiprows: list[int] | int | Callable[[Hashable], bool] | None
+        skipfooter: int
+        nrows: int | None
+        na_values: (
+            Hashable | Iterable[Hashable] | Mapping[Hashable, Iterable[Hashable]] | None
+        )
+        keep_default_na: bool
+        na_filter: bool
+        skip_blank_lines: bool
+        parse_dates: bool | Sequence[Hashable] | None
+        date_format: str | dict[Hashable, str] | None
+        dayfirst: bool
+        cache_dates: bool
+        compression: CompressionOptions
+        thousands: str | None
+        decimal: str
+        lineterminator: str | None
+        quotechar: str
+        quoting: int
+        doublequote: bool
+        escapechar: str | None
+        comment: str | None
+        encoding: str | None
+        encoding_errors: str | None
+        dialect: str | csv.Dialect | None
+        on_bad_lines: str
+        low_memory: bool
+        memory_map: bool
+        float_precision: Literal["high", "legacy", "round_trip"] | None
+        storage_options: StorageOptions | None
+        dtype_backend: DtypeBackend | lib.NoDefault
+else:
+    _read_shared = dict
+
+
+_doc_read_csv_and_table = r"""
+{summary}
+
+Also supports optionally iterating or breaking of the file
+into chunks.
+
+Additional help can be found in the online docs for
+`IO Tools <https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html>`_.
+
+Parameters
+----------
+filepath_or_buffer : str, path object or file-like object
+    Any valid string path is acceptable. The string could be a URL. Valid
+    URL schemes include http, ftp, s3, gs, and file. For file URLs, a host is
+    expected. A local file could be: file://localhost/path/to/table.csv.
+
+    If you want to pass in a path object, pandas accepts any ``os.PathLike``.
+
+    By file-like object, we refer to objects with a ``read()`` method, such as
+    a file handle (e.g. via builtin ``open`` function) or ``StringIO``.
+sep : str, default {_default_sep}
+    Character or regex pattern to treat as the delimiter. If ``sep=None``, the
+    C engine cannot automatically detect
+    the separator, but the Python parsing engine can, meaning the latter will
+    be used and automatically detect the separator from only the first valid
+    row of the file by Python's builtin sniffer tool, ``csv.Sniffer``.
+    In addition, separators longer than 1 character and different from
+    ``'\s+'`` will be interpreted as regular expressions and will also force
+    the use of the Python parsing engine. Note that regex delimiters are prone
+    to ignoring quoted data. Regex example: ``'\r\t'``.
+delimiter : str, optional
+    Alias for ``sep``.
+header : int, Sequence of int, 'infer' or None, default 'infer'
+    Row number(s) containing column labels and marking the start of the
+    data (zero-indexed). Default behavior is to infer the column names: if no ``names``
+    are passed the behavior is identical to ``header=0`` and column
+    names are inferred from the first line of the file, if column
+    names are passed explicitly to ``names`` then the behavior is identical to
+    ``header=None``. Explicitly pass ``header=0`` to be able to
+    replace existing names. The header can be a list of integers that
+    specify row locations for a :class:`~pandas.MultiIndex` on the columns
+    e.g. ``[0, 1, 3]``. Intervening rows that are not specified will be
+    skipped (e.g. 2 in this example is skipped). Note that this
+    parameter ignores commented lines and empty lines if
+    ``skip_blank_lines=True``, so ``header=0`` denotes the first line of
+    data rather than the first line of the file.
+
+    When inferred from the file contents, headers are kept distinct from
+    each other by renaming duplicate names with a numeric suffix of the form
+    ``".{{count}}"`` starting from 1, e.g. ``"foo"`` and ``"foo.1"``.
+    Empty headers are named ``"Unnamed: {{i}}"`` or ``"Unnamed: {{i}}_level_{{level}}"``
+    in the case of MultiIndex columns.
+names : Sequence of Hashable, optional
+    Sequence of column labels to apply. If the file contains a header row,
+    then you should explicitly pass ``header=0`` to override the column names.
+    Duplicates in this list are not allowed.
+index_col : Hashable, Sequence of Hashable or False, optional
+  Column(s) to use as row label(s), denoted either by column labels or column
+  indices.  If a sequence of labels or indices is given, :class:`~pandas.MultiIndex`
+  will be formed for the row labels.
+
+  Note: ``index_col=False`` can be used to force pandas to *not* use the first
+  column as the index, e.g., when you have a malformed file with delimiters at
+  the end of each line.
+usecols : Sequence of Hashable or Callable, optional
+    Subset of columns to select, denoted either by column labels or column indices.
+    If list-like, all elements must either
+    be positional (i.e. integer indices into the document columns) or strings
+    that correspond to column names provided either by the user in ``names`` or
+    inferred from the document header row(s). If ``names`` are given, the document
+    header row(s) are not taken into account. For example, a valid list-like
+    ``usecols`` parameter would be ``[0, 1, 2]`` or ``['foo', 'bar', 'baz']``.
+    Element order is ignored, so ``usecols=[0, 1]`` is the same as ``[1, 0]``.
+    To instantiate a :class:`~pandas.DataFrame` from ``data`` with element order
+    preserved use ``pd.read_csv(data, usecols=['foo', 'bar'])[['foo', 'bar']]``
+    for columns in ``['foo', 'bar']`` order or
+    ``pd.read_csv(data, usecols=['foo', 'bar'])[['bar', 'foo']]``
+    for ``['bar', 'foo']`` order.
+
+    If callable, the callable function will be evaluated against the column
+    names, returning names where the callable function evaluates to ``True``. An
+    example of a valid callable argument would be ``lambda x: x.upper() in
+    ['AAA', 'BBB', 'DDD']``. Using this parameter results in much faster
+    parsing time and lower memory usage.
+dtype : dtype or dict of {{Hashable : dtype}}, optional
+    Data type(s) to apply to either the whole dataset or individual columns.
+    E.g., ``{{'a': np.float64, 'b': np.int32, 'c': 'Int64'}}``
+    Use ``str`` or ``object`` together with suitable ``na_values`` settings
+    to preserve and not interpret ``dtype``.
+    If ``converters`` are specified, they will be applied INSTEAD
+    of ``dtype`` conversion.
+
+    .. versionadded:: 1.5.0
+
+        Support for ``defaultdict`` was added. Specify a ``defaultdict`` as input where
+        the default determines the ``dtype`` of the columns which are not explicitly
+        listed.
+engine : {{'c', 'python', 'pyarrow'}}, optional
+    Parser engine to use. The C and pyarrow engines are faster, while the python engine
+    is currently more feature-complete. Multithreading is currently only supported by
+    the pyarrow engine.
+
+    .. versionadded:: 1.4.0
+
+        The 'pyarrow' engine was added as an *experimental* engine, and some features
+        are unsupported, or may not work correctly, with this engine.
+converters : dict of {{Hashable : Callable}}, optional
+    Functions for converting values in specified columns. Keys can either
+    be column labels or column indices.
+true_values : list, optional
+    Values to consider as ``True`` in addition to case-insensitive variants of 'True'.
+false_values : list, optional
+    Values to consider as ``False`` in addition to case-insensitive variants of 'False'.
+skipinitialspace : bool, default False
+    Skip spaces after delimiter.
+skiprows : int, list of int or Callable, optional
+    Line numbers to skip (0-indexed) or number of lines to skip (``int``)
+    at the start of the file.
+
+    If callable, the callable function will be evaluated against the row
+    indices, returning ``True`` if the row should be skipped and ``False`` otherwise.
+    An example of a valid callable argument would be ``lambda x: x in [0, 2]``.
+skipfooter : int, default 0
+    Number of lines at bottom of file to skip (Unsupported with ``engine='c'``).
+nrows : int, optional
+    Number of rows of file to read. Useful for reading pieces of large files.
+    Refers to the number of data rows in the returned DataFrame, excluding:
+
+    * The header row containing column names.
+    * Rows before the header row, if ``header=1`` or larger.
+
+    Example usage:
+
+    * To read the first 999,999 (non-header) rows:
+      ``read_csv(..., nrows=999999)``
+
+    * To read rows 1,000,000 through 1,999,999:
+      ``read_csv(..., skiprows=1000000, nrows=999999)``
+na_values : Hashable, Iterable of Hashable or dict of {{Hashable : Iterable}}, optional
+    Additional strings to recognize as ``NA``/``NaN``. If ``dict`` passed, specific
+    per-column ``NA`` values.  By default the following values are interpreted as
+    ``NaN``: "{na_values_str}".
+keep_default_na : bool, default True
+    Whether or not to include the default ``NaN`` values when parsing the data.
+    Depending on whether ``na_values`` is passed in, the behavior is as follows:
+
+    * If ``keep_default_na`` is ``True``, and ``na_values`` are specified, ``na_values``
+      is appended to the default ``NaN`` values used for parsing.
+    * If ``keep_default_na`` is ``True``, and ``na_values`` are not specified, only
+      the default ``NaN`` values are used for parsing.
+    * If ``keep_default_na`` is ``False``, and ``na_values`` are specified, only
+      the ``NaN`` values specified ``na_values`` are used for parsing.
+    * If ``keep_default_na`` is ``False``, and ``na_values`` are not specified, no
+      strings will be parsed as ``NaN``.
+
+    Note that if ``na_filter`` is passed in as ``False``, the ``keep_default_na`` and
+    ``na_values`` parameters will be ignored.
+na_filter : bool, default True
+    Detect missing value markers (empty strings and the value of ``na_values``). In
+    data without any ``NA`` values, passing ``na_filter=False`` can improve the
+    performance of reading a large file.
+skip_blank_lines : bool, default True
+    If ``True``, skip over blank lines rather than interpreting as ``NaN`` values.
+parse_dates : bool, None, list of Hashable, default None
+    The behavior is as follows:
+
+    * ``bool``. If ``True`` -> try parsing the index.
+    * ``None``. Behaves like ``True`` if ``date_format`` is specified.
+    * ``list`` of ``int`` or names. e.g. If ``[1, 2, 3]`` -> try parsing columns 1, 2, 3
+      each as a separate date column.
+
+    If a column or index cannot be represented as an array of ``datetime``,
+    say because of an unparsable value or a mixture of timezones, the column
+    or index will be returned unaltered as an ``object`` data type. For
+    non-standard ``datetime`` parsing, use :func:`~pandas.to_datetime` after
+    :func:`~pandas.read_csv`.
+
+    Note: A fast-path exists for iso8601-formatted dates.
+date_format : str or dict of column -> format, optional
+    Format to use for parsing dates and/or times when used in conjunction with ``parse_dates``.
+    The strftime to parse time, e.g. :const:`"%d/%m/%Y"`. See
+    `strftime documentation
+    <https://docs.python.org/3/library/datetime.html
+    #strftime-and-strptime-behavior>`_ for more information on choices, though
+    note that :const:`"%f"` will parse all the way up to nanoseconds.
+    You can also pass:
+
+    - "ISO8601", to parse any `ISO8601 <https://en.wikipedia.org/wiki/ISO_8601>`_
+      time string (not necessarily in exactly the same format);
+    - "mixed", to infer the format for each element individually. This is risky,
+      and you should probably use it along with `dayfirst`.
+
+    .. versionadded:: 2.0.0
+dayfirst : bool, default False
+    DD/MM format dates, international and European format.
+cache_dates : bool, default True
+    If ``True``, use a cache of unique, converted dates to apply the ``datetime``
+    conversion. May produce significant speed-up when parsing duplicate
+    date strings, especially ones with timezone offsets.
+
+iterator : bool, default False
+    Return ``TextFileReader`` object for iteration or getting chunks with
+    ``get_chunk()``.
+chunksize : int, optional
+    Number of lines to read from the file per chunk. Passing a value will cause the
+    function to return a ``TextFileReader`` object for iteration.
+    See the `IO Tools docs
+    <https://pandas.pydata.org/pandas-docs/stable/io.html#io-chunking>`_
+    for more information on ``iterator`` and ``chunksize``.
+
+{decompression_options}
+
+    .. versionchanged:: 1.4.0 Zstandard support.
+
+thousands : str (length 1), optional
+    Character acting as the thousands separator in numerical values.
+decimal : str (length 1), default '.'
+    Character to recognize as decimal point (e.g., use ',' for European data).
+lineterminator : str (length 1), optional
+    Character used to denote a line break. Only valid with C parser.
+quotechar : str (length 1), optional
+    Character used to denote the start and end of a quoted item. Quoted
+    items can include the ``delimiter`` and it will be ignored.
+quoting : {{0 or csv.QUOTE_MINIMAL, 1 or csv.QUOTE_ALL, 2 or csv.QUOTE_NONNUMERIC, 3 or csv.QUOTE_NONE}}, default csv.QUOTE_MINIMAL
+    Control field quoting behavior per ``csv.QUOTE_*`` constants. Default is
+    ``csv.QUOTE_MINIMAL`` (i.e., 0) which implies that only fields containing special
+    characters are quoted (e.g., characters defined in ``quotechar``, ``delimiter``,
+    or ``lineterminator``.
+doublequote : bool, default True
+   When ``quotechar`` is specified and ``quoting`` is not ``QUOTE_NONE``, indicate
+   whether or not to interpret two consecutive ``quotechar`` elements INSIDE a
+   field as a single ``quotechar`` element.
+escapechar : str (length 1), optional
+    Character used to escape other characters.
+comment : str (length 1), optional
+    Character indicating that the remainder of line should not be parsed.
+    If found at the beginning
+    of a line, the line will be ignored altogether. This parameter must be a
+    single character. Like empty lines (as long as ``skip_blank_lines=True``),
+    fully commented lines are ignored by the parameter ``header`` but not by
+    ``skiprows``. For example, if ``comment='#'``, parsing
+    ``#empty\\na,b,c\\n1,2,3`` with ``header=0`` will result in ``'a,b,c'`` being
+    treated as the header.
+encoding : str, optional, default 'utf-8'
+    Encoding to use for UTF when reading/writing (ex. ``'utf-8'``). `List of Python
+    standard encodings
+    <https://docs.python.org/3/library/codecs.html#standard-encodings>`_ .
+
+encoding_errors : str, optional, default 'strict'
+    How encoding errors are treated. `List of possible values
+    <https://docs.python.org/3/library/codecs.html#error-handlers>`_ .
+
+    .. versionadded:: 1.3.0
+
+dialect : str or csv.Dialect, optional
+    If provided, this parameter will override values (default or not) for the
+    following parameters: ``delimiter``, ``doublequote``, ``escapechar``,
+    ``skipinitialspace``, ``quotechar``, and ``quoting``. If it is necessary to
+    override values, a ``ParserWarning`` will be issued. See ``csv.Dialect``
+    documentation for more details.
+on_bad_lines : {{'error', 'warn', 'skip'}} or Callable, default 'error'
+    Specifies what to do upon encountering a bad line (a line with too many fields).
+    Allowed values are:
+
+    - ``'error'``, raise an Exception when a bad line is encountered.
+    - ``'warn'``, raise a warning when a bad line is encountered and skip that line.
+    - ``'skip'``, skip bad lines without raising or warning when they are encountered.
+    - Callable, function that will process a single bad line.
+        - With ``engine='python'``, function with signature
+          ``(bad_line: list[str]) -> list[str] | None``.
+          ``bad_line`` is a list of strings split by the ``sep``.
+          If the function returns ``None``, the bad line will be ignored.
+          If the function returns a new ``list`` of strings with more elements than
+          expected, a ``ParserWarning`` will be emitted while dropping extra elements.
+        - With ``engine='pyarrow'``, function with signature
+          as described in pyarrow documentation: `invalid_row_handler
+          <https://arrow.apache.org/docs/python/generated/pyarrow.csv.ParseOptions.html
+          #pyarrow.csv.ParseOptions.invalid_row_handler>`_.
+
+    .. versionadded:: 1.3.0
+
+    .. versionadded:: 1.4.0
+
+        Callable
+
+    .. versionchanged:: 2.2.0
+
+        Callable for ``engine='pyarrow'``
+
+low_memory : bool, default True
+    Internally process the file in chunks, resulting in lower memory use
+    while parsing, but possibly mixed type inference.  To ensure no mixed
+    types either set ``False``, or specify the type with the ``dtype`` parameter.
+    Note that the entire file is read into a single :class:`~pandas.DataFrame`
+    regardless, use the ``chunksize`` or ``iterator`` parameter to return the data in
+    chunks. (Only valid with C parser).
+memory_map : bool, default False
+    If a filepath is provided for ``filepath_or_buffer``, map the file object
+    directly onto memory and access the data directly from there. Using this
+    option can improve performance because there is no longer any I/O overhead.
+float_precision : {{'high', 'legacy', 'round_trip'}}, optional
+    Specifies which converter the C engine should use for floating-point
+    values. The options are ``None`` or ``'high'`` for the ordinary converter,
+    ``'legacy'`` for the original lower precision pandas converter, and
+    ``'round_trip'`` for the round-trip converter.
+
+{storage_options}
+
+dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+    Back-end data type applied to the resultant :class:`DataFrame`
+    (still experimental). If not specified, the default behavior
+    is to not use nullable data types. If specified, the behavior
+    is as follows:
+
+    * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+    * ``"pyarrow"``: returns pyarrow-backed nullable :class:`ArrowDtype` :class:`DataFrame`
+
+    .. versionadded:: 2.0
+
+Returns
+-------
+DataFrame or TextFileReader
+    A comma-separated values (csv) file is returned as two-dimensional
+    data structure with labeled axes.
+
+See Also
+--------
+DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file.
+{see_also_func_name} : {see_also_func_summary}
+read_fwf : Read a table of fixed-width formatted lines into DataFrame.
+
+Examples
+--------
+>>> pd.{func_name}('data.csv')  # doctest: +SKIP
+   Name  Value
+0   foo      1
+1   bar      2
+2  #baz      3
+
+Index and header can be specified via the `index_col` and `header` arguments.
+
+>>> pd.{func_name}('data.csv', header=None)  # doctest: +SKIP
+      0      1
+0  Name  Value
+1   foo      1
+2   bar      2
+3  #baz      3
+
+>>> pd.{func_name}('data.csv', index_col='Value')  # doctest: +SKIP
+       Name
+Value
+1       foo
+2       bar
+3      #baz
+
+Column types are inferred but can be explicitly specified using the dtype argument.
+
+>>> pd.{func_name}('data.csv', dtype={{'Value': float}})  # doctest: +SKIP
+   Name  Value
+0   foo    1.0
+1   bar    2.0
+2  #baz    3.0
+
+True, False, and NA values, and thousands separators have defaults,
+but can be explicitly specified, too. Supply the values you would like
+as strings or lists of strings!
+
+>>> pd.{func_name}('data.csv', na_values=['foo', 'bar'])  # doctest: +SKIP
+   Name  Value
+0   NaN      1
+1   NaN      2
+2  #baz      3
+
+Comment lines in the input file can be skipped using the `comment` argument.
+
+>>> pd.{func_name}('data.csv', comment='#')  # doctest: +SKIP
+  Name  Value
+0  foo      1
+1  bar      2
+
+By default, columns with dates will be read as ``object`` rather than  ``datetime``.
+
+>>> df = pd.{func_name}('tmp.csv')  # doctest: +SKIP
+
+>>> df  # doctest: +SKIP
+   col 1       col 2            col 3
+0     10  10/04/2018  Sun 15 Jan 2023
+1     20  15/04/2018  Fri 12 May 2023
+
+>>> df.dtypes  # doctest: +SKIP
+col 1     int64
+col 2    object
+col 3    object
+dtype: object
+
+Specific columns can be parsed as dates by using the `parse_dates` and
+`date_format` arguments.
+
+>>> df = pd.{func_name}(
+...     'tmp.csv',
+...     parse_dates=[1, 2],
+...     date_format={{'col 2': '%d/%m/%Y', 'col 3': '%a %d %b %Y'}},
+... )  # doctest: +SKIP
+
+>>> df.dtypes  # doctest: +SKIP
+col 1             int64
+col 2    datetime64[ns]
+col 3    datetime64[ns]
+dtype: object
+"""  # noqa: E501
+
+
+class _C_Parser_Defaults(TypedDict):
+    na_filter: Literal[True]
+    low_memory: Literal[True]
+    memory_map: Literal[False]
+    float_precision: None
+
+
+_c_parser_defaults: _C_Parser_Defaults = {
+    "na_filter": True,
+    "low_memory": True,
+    "memory_map": False,
+    "float_precision": None,
+}
+
+
+class _Fwf_Defaults(TypedDict):
+    colspecs: Literal["infer"]
+    infer_nrows: Literal[100]
+    widths: None
+
+
+_fwf_defaults: _Fwf_Defaults = {"colspecs": "infer", "infer_nrows": 100, "widths": None}
+_c_unsupported = {"skipfooter"}
+_python_unsupported = {"low_memory", "float_precision"}
+_pyarrow_unsupported = {
+    "skipfooter",
+    "float_precision",
+    "chunksize",
+    "comment",
+    "nrows",
+    "thousands",
+    "memory_map",
+    "dialect",
+    "quoting",
+    "lineterminator",
+    "converters",
+    "iterator",
+    "dayfirst",
+    "skipinitialspace",
+    "low_memory",
+}
+
+
+@overload
+def validate_integer(name: str, val: None, min_val: int = ...) -> None: ...
+
+
+@overload
+def validate_integer(name: str, val: float, min_val: int = ...) -> int: ...
+
+
+@overload
+def validate_integer(name: str, val: int | None, min_val: int = ...) -> int | None: ...
+
+
+def validate_integer(
+    name: str, val: int | float | None, min_val: int = 0
+) -> int | None:
+    """
+    Checks whether the 'name' parameter for parsing is either
+    an integer OR float that can SAFELY be cast to an integer
+    without losing accuracy. Raises a ValueError if that is
+    not the case.
+
+    Parameters
+    ----------
+    name : str
+        Parameter name (used for error reporting)
+    val : int or float
+        The value to check
+    min_val : int
+        Minimum allowed value (val < min_val will result in a ValueError)
+    """
+    if val is None:
+        return val
+
+    msg = f"'{name:s}' must be an integer >={min_val:d}"
+    if is_float(val):
+        if int(val) != val:
+            raise ValueError(msg)
+        val = int(val)
+    elif not (is_integer(val) and val >= min_val):
+        raise ValueError(msg)
+
+    return int(val)
+
+
+def _validate_names(names: Sequence[Hashable] | None) -> None:
+    """
+    Raise ValueError if the `names` parameter contains duplicates or has an
+    invalid data type.
+
+    Parameters
+    ----------
+    names : array-like or None
+        An array containing a list of the names used for the output DataFrame.
+
+    Raises
+    ------
+    ValueError
+        If names are not unique or are not ordered (e.g. set).
+    """
+    if names is not None:
+        if len(names) != len(set(names)):
+            raise ValueError("Duplicate names are not allowed.")
+        if not (
+            is_list_like(names, allow_sets=False) or isinstance(names, abc.KeysView)
+        ):
+            raise ValueError("Names should be an ordered collection.")
+
+
+def _read(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str], kwds
+) -> DataFrame | TextFileReader:
+    """Generic reader of line files."""
+    # if we pass a date_format and parse_dates=False, we should not parse the
+    # dates GH#44366
+    if kwds.get("parse_dates", None) is None:
+        if kwds.get("date_format", None) is None:
+            kwds["parse_dates"] = False
+        else:
+            kwds["parse_dates"] = True
+
+    # Extract some of the arguments (pass chunksize on).
+    iterator = kwds.get("iterator", False)
+    chunksize = kwds.get("chunksize", None)
+
+    # Check type of encoding_errors
+    errors = kwds.get("encoding_errors", "strict")
+    if not isinstance(errors, str):
+        raise ValueError(
+            f"encoding_errors must be a string, got {type(errors).__name__}"
+        )
+
+    if kwds.get("engine") == "pyarrow":
+        if iterator:
+            raise ValueError(
+                "The 'iterator' option is not supported with the 'pyarrow' engine"
+            )
+
+        if chunksize is not None:
+            raise ValueError(
+                "The 'chunksize' option is not supported with the 'pyarrow' engine"
+            )
+    else:
+        chunksize = validate_integer("chunksize", chunksize, 1)
+
+    nrows = kwds.get("nrows", None)
+
+    # Check for duplicates in names.
+    _validate_names(kwds.get("names", None))
+
+    # Create the parser.
+    parser = TextFileReader(filepath_or_buffer, **kwds)
+
+    if chunksize or iterator:
+        return parser
+
+    with parser:
+        return parser.read(nrows)
+
+
+@overload
+def read_csv(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: Literal[True],
+    chunksize: int | None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_csv(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: bool = ...,
+    chunksize: int,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_csv(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: Literal[False] = ...,
+    chunksize: None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame: ...
+
+
+@overload
+def read_csv(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: bool = ...,
+    chunksize: int | None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame | TextFileReader: ...
+
+
+@Appender(
+    _doc_read_csv_and_table.format(
+        func_name="read_csv",
+        summary="Read a comma-separated values (csv) file into DataFrame.",
+        see_also_func_name="read_table",
+        see_also_func_summary="Read general delimited file into DataFrame.",
+        na_values_str=fill(
+            '", "'.join(sorted(STR_NA_VALUES)), 70, subsequent_indent="    "
+        ),
+        _default_sep="','",
+        storage_options=_shared_docs["storage_options"],
+        decompression_options=_shared_docs["decompression_options"]
+        % "filepath_or_buffer",
+    )
+)
+@set_module("pandas")
+def read_csv(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    sep: str | None | lib.NoDefault = lib.no_default,
+    delimiter: str | None | lib.NoDefault = None,
+    # Column and Index Locations and Names
+    header: int | Sequence[int] | None | Literal["infer"] = "infer",
+    names: Sequence[Hashable] | None | lib.NoDefault = lib.no_default,
+    index_col: IndexLabel | Literal[False] | None = None,
+    usecols: UsecolsArgType = None,
+    # General Parsing Configuration
+    dtype: DtypeArg | None = None,
+    engine: CSVEngine | None = None,
+    converters: Mapping[HashableT, Callable] | None = None,
+    true_values: list | None = None,
+    false_values: list | None = None,
+    skipinitialspace: bool = False,
+    skiprows: list[int] | int | Callable[[Hashable], bool] | None = None,
+    skipfooter: int = 0,
+    nrows: int | None = None,
+    # NA and Missing Data Handling
+    na_values: Hashable
+    | Iterable[Hashable]
+    | Mapping[Hashable, Iterable[Hashable]]
+    | None = None,
+    keep_default_na: bool = True,
+    na_filter: bool = True,
+    skip_blank_lines: bool = True,
+    # Datetime Handling
+    parse_dates: bool | Sequence[Hashable] | None = None,
+    date_format: str | dict[Hashable, str] | None = None,
+    dayfirst: bool = False,
+    cache_dates: bool = True,
+    # Iteration
+    iterator: bool = False,
+    chunksize: int | None = None,
+    # Quoting, Compression, and File Format
+    compression: CompressionOptions = "infer",
+    thousands: str | None = None,
+    decimal: str = ".",
+    lineterminator: str | None = None,
+    quotechar: str = '"',
+    quoting: int = csv.QUOTE_MINIMAL,
+    doublequote: bool = True,
+    escapechar: str | None = None,
+    comment: str | None = None,
+    encoding: str | None = None,
+    encoding_errors: str | None = "strict",
+    dialect: str | csv.Dialect | None = None,
+    # Error Handling
+    on_bad_lines: str = "error",
+    # Internal
+    low_memory: bool = _c_parser_defaults["low_memory"],
+    memory_map: bool = False,
+    float_precision: Literal["high", "legacy", "round_trip"] | None = None,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame | TextFileReader:
+    # locals() should never be modified
+    kwds = locals().copy()
+    del kwds["filepath_or_buffer"]
+    del kwds["sep"]
+
+    kwds_defaults = _refine_defaults_read(
+        dialect,
+        delimiter,
+        engine,
+        sep,
+        on_bad_lines,
+        names,
+        defaults={"delimiter": ","},
+        dtype_backend=dtype_backend,
+    )
+    kwds.update(kwds_defaults)
+
+    return _read(filepath_or_buffer, kwds)
+
+
+@overload
+def read_table(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: Literal[True],
+    chunksize: int | None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_table(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: bool = ...,
+    chunksize: int,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_table(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: Literal[False] = ...,
+    chunksize: None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame: ...
+
+
+@overload
+def read_table(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    iterator: bool = ...,
+    chunksize: int | None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame | TextFileReader: ...
+
+
+@Appender(
+    _doc_read_csv_and_table.format(
+        func_name="read_table",
+        summary="Read general delimited file into DataFrame.",
+        see_also_func_name="read_csv",
+        see_also_func_summary=(
+            "Read a comma-separated values (csv) file into DataFrame."
+        ),
+        na_values_str=fill(
+            '", "'.join(sorted(STR_NA_VALUES)), 70, subsequent_indent="    "
+        ),
+        _default_sep=r"'\\t' (tab-stop)",
+        storage_options=_shared_docs["storage_options"],
+        decompression_options=_shared_docs["decompression_options"]
+        % "filepath_or_buffer",
+    )
+)
+@set_module("pandas")
+def read_table(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    sep: str | None | lib.NoDefault = lib.no_default,
+    delimiter: str | None | lib.NoDefault = None,
+    # Column and Index Locations and Names
+    header: int | Sequence[int] | None | Literal["infer"] = "infer",
+    names: Sequence[Hashable] | None | lib.NoDefault = lib.no_default,
+    index_col: IndexLabel | Literal[False] | None = None,
+    usecols: UsecolsArgType = None,
+    # General Parsing Configuration
+    dtype: DtypeArg | None = None,
+    engine: CSVEngine | None = None,
+    converters: Mapping[HashableT, Callable] | None = None,
+    true_values: list | None = None,
+    false_values: list | None = None,
+    skipinitialspace: bool = False,
+    skiprows: list[int] | int | Callable[[Hashable], bool] | None = None,
+    skipfooter: int = 0,
+    nrows: int | None = None,
+    # NA and Missing Data Handling
+    na_values: Hashable
+    | Iterable[Hashable]
+    | Mapping[Hashable, Iterable[Hashable]]
+    | None = None,
+    keep_default_na: bool = True,
+    na_filter: bool = True,
+    skip_blank_lines: bool = True,
+    # Datetime Handling
+    parse_dates: bool | Sequence[Hashable] | None = None,
+    date_format: str | dict[Hashable, str] | None = None,
+    dayfirst: bool = False,
+    cache_dates: bool = True,
+    # Iteration
+    iterator: bool = False,
+    chunksize: int | None = None,
+    # Quoting, Compression, and File Format
+    compression: CompressionOptions = "infer",
+    thousands: str | None = None,
+    decimal: str = ".",
+    lineterminator: str | None = None,
+    quotechar: str = '"',
+    quoting: int = csv.QUOTE_MINIMAL,
+    doublequote: bool = True,
+    escapechar: str | None = None,
+    comment: str | None = None,
+    encoding: str | None = None,
+    encoding_errors: str | None = "strict",
+    dialect: str | csv.Dialect | None = None,
+    # Error Handling
+    on_bad_lines: str = "error",
+    # Internal
+    low_memory: bool = _c_parser_defaults["low_memory"],
+    memory_map: bool = False,
+    float_precision: Literal["high", "legacy", "round_trip"] | None = None,
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame | TextFileReader:
+    # locals() should never be modified
+    kwds = locals().copy()
+    del kwds["filepath_or_buffer"]
+    del kwds["sep"]
+
+    kwds_defaults = _refine_defaults_read(
+        dialect,
+        delimiter,
+        engine,
+        sep,
+        on_bad_lines,
+        names,
+        defaults={"delimiter": "\t"},
+        dtype_backend=dtype_backend,
+    )
+    kwds.update(kwds_defaults)
+
+    return _read(filepath_or_buffer, kwds)
+
+
+@overload
+def read_fwf(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    colspecs: Sequence[tuple[int, int]] | str | None = ...,
+    widths: Sequence[int] | None = ...,
+    infer_nrows: int = ...,
+    iterator: Literal[True],
+    chunksize: int | None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_fwf(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    colspecs: Sequence[tuple[int, int]] | str | None = ...,
+    widths: Sequence[int] | None = ...,
+    infer_nrows: int = ...,
+    iterator: bool = ...,
+    chunksize: int,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> TextFileReader: ...
+
+
+@overload
+def read_fwf(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    colspecs: Sequence[tuple[int, int]] | str | None = ...,
+    widths: Sequence[int] | None = ...,
+    infer_nrows: int = ...,
+    iterator: Literal[False] = ...,
+    chunksize: None = ...,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame: ...
+
+
+@set_module("pandas")
+def read_fwf(
+    filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
+    *,
+    colspecs: Sequence[tuple[int, int]] | str | None = "infer",
+    widths: Sequence[int] | None = None,
+    infer_nrows: int = 100,
+    iterator: bool = False,
+    chunksize: int | None = None,
+    **kwds: Unpack[_read_shared[HashableT]],
+) -> DataFrame | TextFileReader:
+    r"""
+    Read a table of fixed-width formatted lines into DataFrame.
+
+    Also supports optionally iterating or breaking of the file
+    into chunks.
+
+    Additional help can be found in the `online docs for IO Tools
+    <https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html>`_.
+
+    Parameters
+    ----------
+    filepath_or_buffer : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a text ``read()`` function.The string could be a URL.
+        Valid URL schemes include http, ftp, s3, and file. For file URLs, a host is
+        expected. A local file could be:
+        ``file://localhost/path/to/table.csv``.
+    colspecs : list of tuple (int, int) or 'infer'. optional
+        A list of tuples giving the extents of the fixed-width
+        fields of each line as half-open intervals (i.e.,  [from, to] ).
+        String value 'infer' can be used to instruct the parser to try
+        detecting the column specifications from the first 100 rows of
+        the data which are not being skipped via skiprows (default='infer').
+    widths : list of int, optional
+        A list of field widths which can be used instead of 'colspecs' if
+        the intervals are contiguous.
+    infer_nrows : int, default 100
+        The number of rows to consider when letting the parser determine the
+        `colspecs`.
+    iterator : bool, default False
+        Return ``TextFileReader`` object for iteration or getting chunks with
+        ``get_chunk()``.
+    chunksize : int, optional
+        Number of lines to read from the file per chunk.
+    **kwds : optional
+        Optional keyword arguments can be passed to ``TextFileReader``.
+
+    Returns
+    -------
+    DataFrame or TextFileReader
+        A comma-separated values (csv) file is returned as two-dimensional
+        data structure with labeled axes.
+
+    See Also
+    --------
+    DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file.
+    read_csv : Read a comma-separated values (csv) file into DataFrame.
+
+    Examples
+    --------
+    >>> pd.read_fwf("data.csv")  # doctest: +SKIP
+    """
+    # Check input arguments.
+    if colspecs is None and widths is None:
+        raise ValueError("Must specify either colspecs or widths")
+    if colspecs not in (None, "infer") and widths is not None:
+        raise ValueError("You must specify only one of 'widths' and 'colspecs'")
+
+    # Compute 'colspecs' from 'widths', if specified.
+    if widths is not None:
+        colspecs, col = [], 0
+        for w in widths:
+            colspecs.append((col, col + w))
+            col += w
+
+    # for mypy
+    assert colspecs is not None
+
+    # GH#40830
+    # Ensure length of `colspecs` matches length of `names`
+    names = kwds.get("names")
+    if names is not None and names is not lib.no_default:
+        if len(names) != len(colspecs) and colspecs != "infer":
+            # need to check len(index_col) as it might contain
+            # unnamed indices, in which case it's name is not required
+            len_index = 0
+            if kwds.get("index_col") is not None:
+                index_col: Any = kwds.get("index_col")
+                if index_col is not False:
+                    if not is_list_like(index_col):
+                        len_index = 1
+                    else:
+                        # for mypy: handled in the if-branch
+                        assert index_col is not lib.no_default
+
+                        len_index = len(index_col)
+            if kwds.get("usecols") is None and len(names) + len_index != len(colspecs):
+                # If usecols is used colspec may be longer than names
+                raise ValueError("Length of colspecs must match length of names")
+
+    check_dtype_backend(kwds.setdefault("dtype_backend", lib.no_default))
+    return _read(
+        filepath_or_buffer,
+        kwds
+        | {
+            "colspecs": colspecs,
+            "infer_nrows": infer_nrows,
+            "engine": "python-fwf",
+            "iterator": iterator,
+            "chunksize": chunksize,
+        },
+    )
+
+
+class TextFileReader(abc.Iterator):
+    """
+
+    Passed dialect overrides any of the related parser options
+
+    """
+
+    def __init__(
+        self,
+        f: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str] | list,
+        engine: CSVEngine | None = None,
+        **kwds,
+    ) -> None:
+        if engine is not None:
+            engine_specified = True
+        else:
+            engine = "python"
+            engine_specified = False
+        self.engine = engine
+        self._engine_specified = kwds.get("engine_specified", engine_specified)
+
+        _validate_skipfooter(kwds)
+
+        dialect = _extract_dialect(kwds)
+        if dialect is not None:
+            if engine == "pyarrow":
+                raise ValueError(
+                    "The 'dialect' option is not supported with the 'pyarrow' engine"
+                )
+            kwds = _merge_with_dialect_properties(dialect, kwds)
+
+        if kwds.get("header", "infer") == "infer":
+            kwds["header"] = 0 if kwds.get("names") is None else None
+
+        self.orig_options = kwds
+
+        # miscellanea
+        self._currow = 0
+
+        options = self._get_options_with_defaults(engine)
+        options["storage_options"] = kwds.get("storage_options", None)
+
+        self.chunksize = options.pop("chunksize", None)
+        self.nrows = options.pop("nrows", None)
+
+        self._check_file_or_buffer(f, engine)
+        self.options, self.engine = self._clean_options(options, engine)
+
+        if "has_index_names" in kwds:
+            self.options["has_index_names"] = kwds["has_index_names"]
+
+        self.handles: IOHandles | None = None
+        self._engine = self._make_engine(f, self.engine)
+
+    def close(self) -> None:
+        if self.handles is not None:
+            self.handles.close()
+        self._engine.close()
+
+    def _get_options_with_defaults(self, engine: CSVEngine) -> dict[str, Any]:
+        kwds = self.orig_options
+
+        options = {}
+        default: object | None
+
+        for argname, default in parser_defaults.items():
+            value = kwds.get(argname, default)
+
+            # see gh-12935
+            if (
+                engine == "pyarrow"
+                and argname in _pyarrow_unsupported
+                and value != default
+                and value != getattr(value, "value", default)
+            ):
+                raise ValueError(
+                    f"The {argname!r} option is not supported with the 'pyarrow' engine"
+                )
+            options[argname] = value
+
+        for argname, default in _c_parser_defaults.items():
+            if argname in kwds:
+                value = kwds[argname]
+
+                if engine != "c" and value != default:
+                    # TODO: Refactor this logic, its pretty convoluted
+                    if "python" in engine and argname not in _python_unsupported:
+                        pass
+                    elif "pyarrow" in engine and argname not in _pyarrow_unsupported:
+                        pass
+                    else:
+                        raise ValueError(
+                            f"The {argname!r} option is not supported with the "
+                            f"{engine!r} engine"
+                        )
+            else:
+                value = default
+            options[argname] = value
+
+        if engine == "python-fwf":
+            for argname, default in _fwf_defaults.items():
+                options[argname] = kwds.get(argname, default)
+
+        return options
+
+    def _check_file_or_buffer(self, f, engine: CSVEngine) -> None:
+        # see gh-16530
+        if is_file_like(f) and engine != "c" and not hasattr(f, "__iter__"):
+            # The C engine doesn't need the file-like to have the "__iter__"
+            # attribute. However, the Python engine needs "__iter__(...)"
+            # when iterating through such an object, meaning it
+            # needs to have that attribute
+            raise ValueError(
+                "The 'python' engine cannot iterate through this file buffer."
+            )
+        if hasattr(f, "encoding"):
+            file_encoding = f.encoding
+            orig_reader_enc = self.orig_options.get("encoding", None)
+            any_none = file_encoding is None or orig_reader_enc is None
+            if file_encoding != orig_reader_enc and not any_none:
+                file_path = getattr(f, "name", None)
+                raise ValueError(
+                    f"The specified reader encoding {orig_reader_enc} is different "
+                    f"from the encoding {file_encoding} of file {file_path}."
+                )
+
+    def _clean_options(
+        self, options: dict[str, Any], engine: CSVEngine
+    ) -> tuple[dict[str, Any], CSVEngine]:
+        result = options.copy()
+
+        fallback_reason = None
+
+        # C engine not supported yet
+        if engine == "c":
+            if options["skipfooter"] > 0:
+                fallback_reason = "the 'c' engine does not support skipfooter"
+                engine = "python"
+
+        sep = options["delimiter"]
+
+        if sep is not None and len(sep) > 1:
+            if engine == "c" and sep == r"\s+":
+                # delim_whitespace passed on to pandas._libs.parsers.TextReader
+                result["delim_whitespace"] = True
+                del result["delimiter"]
+            elif engine not in ("python", "python-fwf"):
+                # wait until regex engine integrated
+                fallback_reason = (
+                    f"the '{engine}' engine does not support "
+                    "regex separators (separators > 1 char and "
+                    r"different from '\s+' are interpreted as regex)"
+                )
+                engine = "python"
+        elif sep is not None:
+            encodeable = True
+            encoding = sys.getfilesystemencoding() or "utf-8"
+            try:
+                if len(sep.encode(encoding)) > 1:
+                    encodeable = False
+            except UnicodeDecodeError:
+                encodeable = False
+            if not encodeable and engine not in ("python", "python-fwf"):
+                fallback_reason = (
+                    f"the separator encoded in {encoding} "
+                    f"is > 1 char long, and the '{engine}' engine "
+                    "does not support such separators"
+                )
+                engine = "python"
+
+        quotechar = options["quotechar"]
+        if quotechar is not None and isinstance(quotechar, (str, bytes)):
+            if (
+                len(quotechar) == 1
+                and ord(quotechar) > 127
+                and engine not in ("python", "python-fwf")
+            ):
+                fallback_reason = (
+                    "ord(quotechar) > 127, meaning the "
+                    "quotechar is larger than one byte, "
+                    f"and the '{engine}' engine does not support such quotechars"
+                )
+                engine = "python"
+
+        if fallback_reason and self._engine_specified:
+            raise ValueError(fallback_reason)
+
+        if engine == "c":
+            for arg in _c_unsupported:
+                del result[arg]
+
+        if "python" in engine:
+            for arg in _python_unsupported:
+                if fallback_reason and result[arg] != _c_parser_defaults.get(arg):
+                    raise ValueError(
+                        "Falling back to the 'python' engine because "
+                        f"{fallback_reason}, but this causes {arg!r} to be "
+                        "ignored as it is not supported by the 'python' engine."
+                    )
+                del result[arg]
+
+        if fallback_reason:
+            warnings.warn(
+                (
+                    "Falling back to the 'python' engine because "
+                    f"{fallback_reason}; you can avoid this warning by specifying "
+                    "engine='python'."
+                ),
+                ParserWarning,
+                stacklevel=find_stack_level(),
+            )
+
+        index_col = options["index_col"]
+        names = options["names"]
+        converters = options["converters"]
+        na_values = options["na_values"]
+        skiprows = options["skiprows"]
+
+        validate_header_arg(options["header"])
+
+        if index_col is True:
+            raise ValueError("The value of index_col couldn't be 'True'")
+        if is_index_col(index_col):
+            if not isinstance(index_col, (list, tuple, np.ndarray)):
+                index_col = [index_col]
+        result["index_col"] = index_col
+
+        names = list(names) if names is not None else names
+
+        # type conversion-related
+        if converters is not None:
+            if not isinstance(converters, dict):
+                raise TypeError(
+                    "Type converters must be a dict or subclass, "
+                    f"input was a {type(converters).__name__}"
+                )
+        else:
+            converters = {}
+
+        # Converting values to NA
+        keep_default_na = options["keep_default_na"]
+        floatify = engine != "pyarrow"
+        na_values, na_fvalues = _clean_na_values(
+            na_values, keep_default_na, floatify=floatify
+        )
+
+        # handle skiprows; this is internally handled by the
+        # c-engine, so only need for python and pyarrow parsers
+        if engine == "pyarrow":
+            if not is_integer(skiprows) and skiprows is not None:
+                # pyarrow expects skiprows to be passed as an integer
+                raise ValueError(
+                    "skiprows argument must be an integer when using engine='pyarrow'"
+                )
+        else:
+            if is_integer(skiprows):
+                skiprows = range(skiprows)
+            if skiprows is None:
+                skiprows = set()
+            elif not callable(skiprows):
+                skiprows = set(skiprows)
+
+        # put stuff back
+        result["names"] = names
+        result["converters"] = converters
+        result["na_values"] = na_values
+        result["na_fvalues"] = na_fvalues
+        result["skiprows"] = skiprows
+
+        return result, engine
+
+    def __next__(self) -> DataFrame:
+        try:
+            return self.get_chunk()
+        except StopIteration:
+            self.close()
+            raise
+
+    def _make_engine(
+        self,
+        f: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str] | list | IO,
+        engine: CSVEngine = "c",
+    ) -> ParserBase:
+        mapping: dict[str, type[ParserBase]] = {
+            "c": CParserWrapper,
+            "python": PythonParser,
+            "pyarrow": ArrowParserWrapper,
+            "python-fwf": FixedWidthFieldParser,
+        }
+
+        if engine not in mapping:
+            raise ValueError(
+                f"Unknown engine: {engine} (valid options are {mapping.keys()})"
+            )
+        if not isinstance(f, list):
+            # open file here
+            is_text = True
+            mode = "r"
+            if engine == "pyarrow":
+                is_text = False
+                mode = "rb"
+            elif (
+                engine == "c"
+                and self.options.get("encoding", "utf-8") == "utf-8"
+                and isinstance(stringify_path(f), str)
+            ):
+                # c engine can decode utf-8 bytes, adding TextIOWrapper makes
+                # the c-engine especially for memory_map=True far slower
+                is_text = False
+                if "b" not in mode:
+                    mode += "b"
+            self.handles = get_handle(
+                f,
+                mode,
+                encoding=self.options.get("encoding", None),
+                compression=self.options.get("compression", None),
+                memory_map=self.options.get("memory_map", False),
+                is_text=is_text,
+                errors=self.options.get("encoding_errors", "strict"),
+                storage_options=self.options.get("storage_options", None),
+            )
+            assert self.handles is not None
+            f = self.handles.handle
+
+        elif engine != "python":
+            msg = f"Invalid file path or buffer object type: {type(f)}"
+            raise ValueError(msg)
+
+        try:
+            return mapping[engine](f, **self.options)
+        except Exception:
+            if self.handles is not None:
+                self.handles.close()
+            raise
+
+    def _failover_to_python(self) -> None:
+        raise AbstractMethodError(self)
+
+    def read(self, nrows: int | None = None) -> DataFrame:
+        if self.engine == "pyarrow":
+            try:
+                # error: "ParserBase" has no attribute "read"
+                df = self._engine.read()  # type: ignore[attr-defined]
+            except Exception:
+                self.close()
+                raise
+        else:
+            nrows = validate_integer("nrows", nrows)
+            try:
+                # error: "ParserBase" has no attribute "read"
+                (
+                    index,
+                    columns,
+                    col_dict,
+                ) = self._engine.read(  # type: ignore[attr-defined]
+                    nrows
+                )
+            except Exception:
+                self.close()
+                raise
+
+            if index is None:
+                if col_dict:
+                    # Any column is actually fine:
+                    new_rows = len(next(iter(col_dict.values())))
+                    index = RangeIndex(self._currow, self._currow + new_rows)
+                else:
+                    new_rows = 0
+            else:
+                new_rows = len(index)
+
+            if hasattr(self, "orig_options"):
+                dtype_arg = self.orig_options.get("dtype", None)
+            else:
+                dtype_arg = None
+
+            if isinstance(dtype_arg, dict):
+                dtype = defaultdict(lambda: None)  # type: ignore[var-annotated]
+                dtype.update(dtype_arg)
+            elif dtype_arg is not None and pandas_dtype(dtype_arg) in (
+                np.str_,
+                np.object_,
+            ):
+                dtype = defaultdict(lambda: dtype_arg)
+            else:
+                dtype = None
+
+            if dtype is not None:
+                new_col_dict = {}
+                for k, v in col_dict.items():
+                    d = (
+                        dtype[k]
+                        if pandas_dtype(dtype[k]) in (np.str_, np.object_)
+                        else None
+                    )
+                    new_col_dict[k] = Series(v, index=index, dtype=d, copy=False)
+            else:
+                new_col_dict = col_dict
+
+            df = DataFrame(
+                new_col_dict,
+                columns=columns,
+                index=index,
+                copy=False,
+            )
+
+            self._currow += new_rows
+        return df
+
+    def get_chunk(self, size: int | None = None) -> DataFrame:
+        if size is None:
+            size = self.chunksize
+        if self.nrows is not None:
+            if self._currow >= self.nrows:
+                raise StopIteration
+            if size is None:
+                size = self.nrows - self._currow
+            else:
+                size = min(size, self.nrows - self._currow)
+        return self.read(nrows=size)
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+
+def TextParser(*args, **kwds) -> TextFileReader:
+    """
+    Converts lists of lists/tuples into DataFrames with proper type inference
+    and optional (e.g. string to datetime) conversion. Also enables iterating
+    lazily over chunks of large files
+
+    Parameters
+    ----------
+    data : file-like object or list
+    delimiter : separator character to use
+    dialect : str or csv.Dialect instance, optional
+        Ignored if delimiter is longer than 1 character
+    names : sequence, default
+    header : int, default 0
+        Row to use to parse column labels. Defaults to the first row. Prior
+        rows will be discarded
+    index_col : int or list, optional
+        Column or columns to use as the (possibly hierarchical) index
+    has_index_names: bool, default False
+        True if the cols defined in index_col have an index name and are
+        not in the header.
+    na_values : scalar, str, list-like, or dict, optional
+        Additional strings to recognize as NA/NaN.
+    keep_default_na : bool, default True
+    thousands : str, optional
+        Thousands separator
+    comment : str, optional
+        Comment out remainder of line
+    parse_dates : bool, default False
+    date_format : str or dict of column -> format, default ``None``
+
+        .. versionadded:: 2.0.0
+    skiprows : list of integers
+        Row numbers to skip
+    skipfooter : int
+        Number of line at bottom of file to skip
+    converters : dict, optional
+        Dict of functions for converting values in certain columns. Keys can
+        either be integers or column labels, values are functions that take one
+        input argument, the cell (not column) content, and return the
+        transformed content.
+    encoding : str, optional
+        Encoding to use for UTF when reading/writing (ex. 'utf-8')
+    float_precision : str, optional
+        Specifies which converter the C engine should use for floating-point
+        values. The options are `None` or `high` for the ordinary converter,
+        `legacy` for the original lower precision pandas converter, and
+        `round_trip` for the round-trip converter.
+    """
+    kwds["engine"] = "python"
+    return TextFileReader(*args, **kwds)
+
+
+def _clean_na_values(na_values, keep_default_na: bool = True, floatify: bool = True):
+    na_fvalues: set | dict
+    if na_values is None:
+        if keep_default_na:
+            na_values = STR_NA_VALUES
+        else:
+            na_values = set()
+        na_fvalues = set()
+    elif isinstance(na_values, dict):
+        old_na_values = na_values.copy()
+        na_values = {}  # Prevent aliasing.
+
+        # Convert the values in the na_values dictionary
+        # into array-likes for further use. This is also
+        # where we append the default NaN values, provided
+        # that `keep_default_na=True`.
+        for k, v in old_na_values.items():
+            if not is_list_like(v):
+                v = [v]
+
+            if keep_default_na:
+                v = set(v) | STR_NA_VALUES
+
+            na_values[k] = _stringify_na_values(v, floatify)
+        na_fvalues = {k: _floatify_na_values(v) for k, v in na_values.items()}
+    else:
+        if not is_list_like(na_values):
+            na_values = [na_values]
+        na_values = _stringify_na_values(na_values, floatify)
+        if keep_default_na:
+            na_values = na_values | STR_NA_VALUES
+
+        na_fvalues = _floatify_na_values(na_values)
+
+    return na_values, na_fvalues
+
+
+def _floatify_na_values(na_values):
+    # create float versions of the na_values
+    result = set()
+    for v in na_values:
+        try:
+            v = float(v)
+            if not np.isnan(v):
+                result.add(v)
+        except (TypeError, ValueError, OverflowError):
+            pass
+    return result
+
+
+def _stringify_na_values(na_values, floatify: bool) -> set[str | float]:
+    """return a stringified and numeric for these values"""
+    result: list[str | float] = []
+    for x in na_values:
+        result.append(str(x))
+        result.append(x)
+        try:
+            v = float(x)
+
+            # we are like 999 here
+            if v == int(v):
+                v = int(v)
+                result.append(f"{v}.0")
+                result.append(str(v))
+
+            if floatify:
+                result.append(v)
+        except (TypeError, ValueError, OverflowError):
+            pass
+        if floatify:
+            try:
+                result.append(int(x))
+            except (TypeError, ValueError, OverflowError):
+                pass
+    return set(result)
+
+
+def _refine_defaults_read(
+    dialect: str | csv.Dialect | None,
+    delimiter: str | None | lib.NoDefault,
+    engine: CSVEngine | None,
+    sep: str | None | lib.NoDefault,
+    on_bad_lines: str | Callable,
+    names: Sequence[Hashable] | None | lib.NoDefault,
+    defaults: dict[str, Any],
+    dtype_backend: DtypeBackend | lib.NoDefault,
+):
+    """Validate/refine default values of input parameters of read_csv, read_table.
+
+    Parameters
+    ----------
+    dialect : str or csv.Dialect
+        If provided, this parameter will override values (default or not) for the
+        following parameters: `delimiter`, `doublequote`, `escapechar`,
+        `skipinitialspace`, `quotechar`, and `quoting`. If it is necessary to
+        override values, a ParserWarning will be issued. See csv.Dialect
+        documentation for more details.
+    delimiter : str or object
+        Alias for sep.
+    engine : {{'c', 'python'}}
+        Parser engine to use. The C engine is faster while the python engine is
+        currently more feature-complete.
+    sep : str or object
+        A delimiter provided by the user (str) or a sentinel value, i.e.
+        pandas._libs.lib.no_default.
+    on_bad_lines : str, callable
+        An option for handling bad lines or a sentinel value(None).
+    names : array-like, optional
+        List of column names to use. If the file contains a header row,
+        then you should explicitly pass ``header=0`` to override the column names.
+        Duplicates in this list are not allowed.
+    defaults: dict
+        Default values of input parameters.
+
+    Returns
+    -------
+    kwds : dict
+        Input parameters with correct values.
+    """
+    # fix types for sep, delimiter to Union(str, Any)
+    delim_default = defaults["delimiter"]
+    kwds: dict[str, Any] = {}
+    # gh-23761
+    #
+    # When a dialect is passed, it overrides any of the overlapping
+    # parameters passed in directly. We don't want to warn if the
+    # default parameters were passed in (since it probably means
+    # that the user didn't pass them in explicitly in the first place).
+    #
+    # "delimiter" is the annoying corner case because we alias it to
+    # "sep" before doing comparison to the dialect values later on.
+    # Thus, we need a flag to indicate that we need to "override"
+    # the comparison to dialect values by checking if default values
+    # for BOTH "delimiter" and "sep" were provided.
+    if dialect is not None:
+        kwds["sep_override"] = delimiter is None and (
+            sep is lib.no_default or sep == delim_default
+        )
+
+    if delimiter and (sep is not lib.no_default):
+        raise ValueError("Specified a sep and a delimiter; you can only specify one.")
+
+    kwds["names"] = None if names is lib.no_default else names
+
+    # Alias sep -> delimiter.
+    if delimiter is None:
+        delimiter = sep
+
+    if delimiter == "\n":
+        raise ValueError(
+            r"Specified \n as separator or delimiter. This forces the python engine "
+            "which does not accept a line terminator. Hence it is not allowed to use "
+            "the line terminator as separator.",
+        )
+
+    if delimiter is lib.no_default:
+        # assign default separator value
+        kwds["delimiter"] = delim_default
+    else:
+        kwds["delimiter"] = delimiter
+
+    if engine is not None:
+        kwds["engine_specified"] = True
+    else:
+        kwds["engine"] = "c"
+        kwds["engine_specified"] = False
+
+    if on_bad_lines == "error":
+        kwds["on_bad_lines"] = ParserBase.BadLineHandleMethod.ERROR
+    elif on_bad_lines == "warn":
+        kwds["on_bad_lines"] = ParserBase.BadLineHandleMethod.WARN
+    elif on_bad_lines == "skip":
+        kwds["on_bad_lines"] = ParserBase.BadLineHandleMethod.SKIP
+    elif callable(on_bad_lines):
+        if engine not in ["python", "pyarrow"]:
+            raise ValueError(
+                "on_bad_line can only be a callable function "
+                "if engine='python' or 'pyarrow'"
+            )
+        kwds["on_bad_lines"] = on_bad_lines
+    else:
+        raise ValueError(f"Argument {on_bad_lines} is invalid for on_bad_lines")
+
+    check_dtype_backend(dtype_backend)
+
+    kwds["dtype_backend"] = dtype_backend
+
+    return kwds
+
+
+def _extract_dialect(kwds: dict[str, Any]) -> csv.Dialect | None:
+    """
+    Extract concrete csv dialect instance.
+
+    Returns
+    -------
+    csv.Dialect or None
+    """
+    if kwds.get("dialect") is None:
+        return None
+
+    dialect = kwds["dialect"]
+    if dialect in csv.list_dialects():
+        dialect = csv.get_dialect(dialect)
+
+    _validate_dialect(dialect)
+
+    return dialect
+
+
+MANDATORY_DIALECT_ATTRS = (
+    "delimiter",
+    "doublequote",
+    "escapechar",
+    "skipinitialspace",
+    "quotechar",
+    "quoting",
+)
+
+
+def _validate_dialect(dialect: csv.Dialect) -> None:
+    """
+    Validate csv dialect instance.
+
+    Raises
+    ------
+    ValueError
+        If incorrect dialect is provided.
+    """
+    for param in MANDATORY_DIALECT_ATTRS:
+        if not hasattr(dialect, param):
+            raise ValueError(f"Invalid dialect {dialect} provided")
+
+
+def _merge_with_dialect_properties(
+    dialect: csv.Dialect,
+    defaults: dict[str, Any],
+) -> dict[str, Any]:
+    """
+    Merge default kwargs in TextFileReader with dialect parameters.
+
+    Parameters
+    ----------
+    dialect : csv.Dialect
+        Concrete csv dialect. See csv.Dialect documentation for more details.
+    defaults : dict
+        Keyword arguments passed to TextFileReader.
+
+    Returns
+    -------
+    kwds : dict
+        Updated keyword arguments, merged with dialect parameters.
+    """
+    kwds = defaults.copy()
+
+    for param in MANDATORY_DIALECT_ATTRS:
+        dialect_val = getattr(dialect, param)
+
+        parser_default = parser_defaults[param]
+        provided = kwds.get(param, parser_default)
+
+        # Messages for conflicting values between the dialect
+        # instance and the actual parameters provided.
+        conflict_msgs = []
+
+        # Don't warn if the default parameter was passed in,
+        # even if it conflicts with the dialect (gh-23761).
+        if provided not in (parser_default, dialect_val):
+            msg = (
+                f"Conflicting values for '{param}': '{provided}' was "
+                f"provided, but the dialect specifies '{dialect_val}'. "
+                "Using the dialect-specified value."
+            )
+
+            # Annoying corner case for not warning about
+            # conflicts between dialect and delimiter parameter.
+            # Refer to the outer "_read_" function for more info.
+            if not (param == "delimiter" and kwds.pop("sep_override", False)):
+                conflict_msgs.append(msg)
+
+        if conflict_msgs:
+            warnings.warn(
+                "\n\n".join(conflict_msgs), ParserWarning, stacklevel=find_stack_level()
+            )
+        kwds[param] = dialect_val
+    return kwds
+
+
+def _validate_skipfooter(kwds: dict[str, Any]) -> None:
+    """
+    Check whether skipfooter is compatible with other kwargs in TextFileReader.
+
+    Parameters
+    ----------
+    kwds : dict
+        Keyword arguments passed to TextFileReader.
+
+    Raises
+    ------
+    ValueError
+        If skipfooter is not compatible with other parameters.
+    """
+    if kwds.get("skipfooter"):
+        if kwds.get("iterator") or kwds.get("chunksize"):
+            raise ValueError("'skipfooter' not supported for iteration")
+        if kwds.get("nrows"):
+            raise ValueError("'skipfooter' not supported with 'nrows'")
diff --git a/pandas/pandas/io/pickle.py b/pandas/pandas/io/pickle.py
new file mode 100644
index 0000000000000..f0441f583bea2
--- /dev/null
+++ b/pandas/pandas/io/pickle.py
@@ -0,0 +1,209 @@
+"""pickle compat"""
+
+from __future__ import annotations
+
+import pickle
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import warnings
+
+from pandas.compat import pickle_compat
+from pandas.util._decorators import doc
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import get_handle
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        ReadPickleBuffer,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    compression_options=_shared_docs["compression_options"] % "filepath_or_buffer",
+)
+def to_pickle(
+    obj: Any,
+    filepath_or_buffer: FilePath | WriteBuffer[bytes],
+    compression: CompressionOptions = "infer",
+    protocol: int = pickle.HIGHEST_PROTOCOL,
+    storage_options: StorageOptions | None = None,
+) -> None:
+    """
+    Pickle (serialize) object to file.
+
+    Parameters
+    ----------
+    obj : any object
+        Any python object.
+    filepath_or_buffer : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``write()`` function.
+        Also accepts URL. URL has to be of S3 or GCS.
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    protocol : int
+        Int which indicates which protocol should be used by the pickler,
+        default HIGHEST_PROTOCOL (see [1], paragraph 12.1.2). The possible
+        values for this parameter depend on the version of Python. For Python
+        2.x, possible values are 0, 1, 2. For Python>=3.0, 3 is a valid value.
+        For Python >= 3.4, 4 is a valid value. A negative value for the
+        protocol parameter is equivalent to setting its value to
+        HIGHEST_PROTOCOL.
+
+    {storage_options}
+
+        .. [1] https://docs.python.org/3/library/pickle.html
+
+    See Also
+    --------
+    read_pickle : Load pickled pandas object (or any object) from file.
+    DataFrame.to_hdf : Write DataFrame to an HDF5 file.
+    DataFrame.to_sql : Write DataFrame to a SQL database.
+    DataFrame.to_parquet : Write a DataFrame to the binary parquet format.
+
+    Examples
+    --------
+    >>> original_df = pd.DataFrame(
+    ...     {{"foo": range(5), "bar": range(5, 10)}}
+    ... )  # doctest: +SKIP
+    >>> original_df  # doctest: +SKIP
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    >>> pd.to_pickle(original_df, "./dummy.pkl")  # doctest: +SKIP
+
+    >>> unpickled_df = pd.read_pickle("./dummy.pkl")  # doctest: +SKIP
+    >>> unpickled_df  # doctest: +SKIP
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    """
+    if protocol < 0:
+        protocol = pickle.HIGHEST_PROTOCOL
+
+    with get_handle(
+        filepath_or_buffer,
+        "wb",
+        compression=compression,
+        is_text=False,
+        storage_options=storage_options,
+    ) as handles:
+        # letting pickle write directly to the buffer is more memory-efficient
+        pickle.dump(obj, handles.handle, protocol=protocol)
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    decompression_options=_shared_docs["decompression_options"] % "filepath_or_buffer",
+)
+def read_pickle(
+    filepath_or_buffer: FilePath | ReadPickleBuffer,
+    compression: CompressionOptions = "infer",
+    storage_options: StorageOptions | None = None,
+) -> DataFrame | Series:
+    """
+    Load pickled pandas object (or any object) from file and return unpickled object.
+
+    .. warning::
+
+       Loading pickled data received from untrusted sources can be
+       unsafe. See `here <https://docs.python.org/3/library/pickle.html>`__.
+
+    Parameters
+    ----------
+    filepath_or_buffer : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``readlines()`` function.
+        Also accepts URL. URL is not limited to S3 and GCS.
+
+    {decompression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    {storage_options}
+
+    Returns
+    -------
+    object
+        The unpickled pandas object (or any object) that was stored in file.
+
+    See Also
+    --------
+    DataFrame.to_pickle : Pickle (serialize) DataFrame object to file.
+    Series.to_pickle : Pickle (serialize) Series object to file.
+    read_hdf : Read HDF5 file into a DataFrame.
+    read_sql : Read SQL query or database table into a DataFrame.
+    read_parquet : Load a parquet object, returning a DataFrame.
+
+    Notes
+    -----
+    read_pickle is only guaranteed to be backwards compatible to pandas 1.0
+    provided the object was serialized with to_pickle.
+
+    Examples
+    --------
+    >>> original_df = pd.DataFrame(
+    ...     {{"foo": range(5), "bar": range(5, 10)}}
+    ... )  # doctest: +SKIP
+    >>> original_df  # doctest: +SKIP
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    >>> pd.to_pickle(original_df, "./dummy.pkl")  # doctest: +SKIP
+
+    >>> unpickled_df = pd.read_pickle("./dummy.pkl")  # doctest: +SKIP
+    >>> unpickled_df  # doctest: +SKIP
+       foo  bar
+    0    0    5
+    1    1    6
+    2    2    7
+    3    3    8
+    4    4    9
+    """
+    # TypeError for Cython complaints about object.__new__ vs Tick.__new__
+    excs_to_catch = (AttributeError, ImportError, ModuleNotFoundError, TypeError)
+    with get_handle(
+        filepath_or_buffer,
+        "rb",
+        compression=compression,
+        is_text=False,
+        storage_options=storage_options,
+    ) as handles:
+        # 1) try standard library Pickle
+        # 2) try pickle_compat (older pandas version) to handle subclass changes
+        try:
+            with warnings.catch_warnings(record=True):
+                # We want to silence any warnings about, e.g. moved modules.
+                warnings.simplefilter("ignore", Warning)
+                return pickle.load(handles.handle)
+        except excs_to_catch:
+            # e.g.
+            #  "No module named 'pandas.core.sparse.series'"
+            #  "Can't get attribute '_nat_unpickle' on <module 'pandas._libs.tslib"
+            handles.handle.seek(0)
+            return pickle_compat.Unpickler(handles.handle).load()
diff --git a/pandas/pandas/io/pytables.py b/pandas/pandas/io/pytables.py
new file mode 100644
index 0000000000000..a689cfbcb1418
--- /dev/null
+++ b/pandas/pandas/io/pytables.py
@@ -0,0 +1,5510 @@
+"""
+High level interface to PyTables for reading and writing pandas data structures
+to disk
+"""
+
+from __future__ import annotations
+
+from contextlib import suppress
+import copy
+from datetime import (
+    date,
+    tzinfo,
+)
+import itertools
+import os
+import re
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Final,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import (
+    config,
+    get_option,
+    using_string_dtype,
+)
+
+from pandas._libs import (
+    lib,
+    writers as libwriters,
+)
+from pandas._libs.lib import is_string_array
+from pandas._libs.tslibs import timezones
+from pandas.compat._optional import import_optional_dependency
+from pandas.compat.pickle_compat import patch_pickle
+from pandas.errors import (
+    AttributeConflictWarning,
+    ClosedFileError,
+    IncompatibilityWarning,
+    PerformanceWarning,
+    PossibleDataLossError,
+)
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_bool_dtype,
+    is_complex_dtype,
+    is_list_like,
+    is_string_dtype,
+    needs_i8_conversion,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import array_equivalent
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    StringDtype,
+    TimedeltaIndex,
+    concat,
+    isna,
+)
+from pandas.core.arrays import (
+    Categorical,
+    DatetimeArray,
+    PeriodArray,
+)
+from pandas.core.arrays.datetimes import tz_to_dtype
+from pandas.core.arrays.string_ import BaseStringArray
+import pandas.core.common as com
+from pandas.core.computation.pytables import (
+    PyTablesExpr,
+    maybe_expression,
+)
+from pandas.core.construction import (
+    array as pd_array,
+    extract_array,
+)
+from pandas.core.indexes.api import ensure_index
+
+from pandas.io.common import stringify_path
+from pandas.io.formats.printing import (
+    adjoin,
+    pprint_thing,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Iterator,
+        Sequence,
+    )
+    from types import TracebackType
+
+    from tables import (
+        Col,
+        File,
+        Node,
+    )
+
+    from pandas._typing import (
+        AnyArrayLike,
+        ArrayLike,
+        AxisInt,
+        DtypeArg,
+        FilePath,
+        Self,
+        Shape,
+        npt,
+    )
+
+    from pandas.core.internals import Block
+
+# versioning attribute
+_version = "0.15.2"
+
+# encoding
+_default_encoding = "UTF-8"
+
+
+def _ensure_encoding(encoding: str | None) -> str:
+    # set the encoding if we need
+    if encoding is None:
+        encoding = _default_encoding
+
+    return encoding
+
+
+def _ensure_str(name):
+    """
+    Ensure that an index / column name is a str (python 3); otherwise they
+    may be np.string dtype. Non-string dtypes are passed through unchanged.
+
+    https://github.com/pandas-dev/pandas/issues/13492
+    """
+    if isinstance(name, str):
+        name = str(name)
+    return name
+
+
+Term = PyTablesExpr
+
+
+def _ensure_term(where, scope_level: int):
+    """
+    Ensure that the where is a Term or a list of Term.
+
+    This makes sure that we are capturing the scope of variables that are
+    passed create the terms here with a frame_level=2 (we are 2 levels down)
+    """
+    # only consider list/tuple here as an ndarray is automatically a coordinate
+    # list
+    level = scope_level + 1
+    if isinstance(where, (list, tuple)):
+        where = [
+            Term(term, scope_level=level + 1) if maybe_expression(term) else term
+            for term in where
+            if term is not None
+        ]
+    elif maybe_expression(where):
+        where = Term(where, scope_level=level)
+    return where if where is None or len(where) else None
+
+
+incompatibility_doc: Final = """
+where criteria is being ignored as this version [%s] is too old (or
+not-defined), read the file in and write it out to a new file to upgrade (with
+the copy_to method)
+"""
+
+attribute_conflict_doc: Final = """
+the [%s] attribute of the existing index is [%s] which conflicts with the new
+[%s], resetting the attribute to None
+"""
+
+performance_doc: Final = """
+your performance may suffer as PyTables will pickle object types that it cannot
+map directly to c-types [inferred_type->%s,key->%s] [items->%s]
+"""
+
+# formats
+_FORMAT_MAP = {"f": "fixed", "fixed": "fixed", "t": "table", "table": "table"}
+
+# axes map
+_AXES_MAP = {DataFrame: [0]}
+
+# register our configuration options
+dropna_doc: Final = """
+: boolean
+    drop ALL nan rows when appending to a table
+"""
+format_doc: Final = """
+: format
+    default format writing format, if None, then
+    put will default to 'fixed' and append will default to 'table'
+"""
+
+with config.config_prefix("io.hdf"):
+    config.register_option("dropna_table", False, dropna_doc, validator=config.is_bool)
+    config.register_option(
+        "default_format",
+        None,
+        format_doc,
+        validator=config.is_one_of_factory(["fixed", "table", None]),
+    )
+
+# oh the troubles to reduce import time
+_table_mod = None
+_table_file_open_policy_is_strict = False
+
+
+def _tables():
+    global _table_mod
+    global _table_file_open_policy_is_strict
+    if _table_mod is None:
+        import tables
+
+        _table_mod = tables
+
+        # set the file open policy
+        # return the file open policy; this changes as of pytables 3.1
+        # depending on the HDF5 version
+        with suppress(AttributeError):
+            _table_file_open_policy_is_strict = (
+                tables.file._FILE_OPEN_POLICY == "strict"
+            )
+
+    return _table_mod
+
+
+# interface to/from ###
+
+
+def to_hdf(
+    path_or_buf: FilePath | HDFStore,
+    key: str,
+    value: DataFrame | Series,
+    mode: str = "a",
+    complevel: int | None = None,
+    complib: str | None = None,
+    append: bool = False,
+    format: str | None = None,
+    index: bool = True,
+    min_itemsize: int | dict[str, int] | None = None,
+    nan_rep=None,
+    dropna: bool | None = None,
+    data_columns: Literal[True] | list[str] | None = None,
+    errors: str = "strict",
+    encoding: str = "UTF-8",
+) -> None:
+    """store this object, close it if we opened it"""
+    if append:
+        f = lambda store: store.append(
+            key,
+            value,
+            format=format,
+            index=index,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            dropna=dropna,
+            data_columns=data_columns,
+            errors=errors,
+            encoding=encoding,
+        )
+    else:
+        # NB: dropna is not passed to `put`
+        f = lambda store: store.put(
+            key,
+            value,
+            format=format,
+            index=index,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            data_columns=data_columns,
+            errors=errors,
+            encoding=encoding,
+            dropna=dropna,
+        )
+
+    if isinstance(path_or_buf, HDFStore):
+        f(path_or_buf)
+    else:
+        path_or_buf = stringify_path(path_or_buf)
+        with HDFStore(
+            path_or_buf, mode=mode, complevel=complevel, complib=complib
+        ) as store:
+            f(store)
+
+
+def read_hdf(
+    path_or_buf: FilePath | HDFStore,
+    key=None,
+    mode: str = "r",
+    errors: str = "strict",
+    where: str | list | None = None,
+    start: int | None = None,
+    stop: int | None = None,
+    columns: list[str] | None = None,
+    iterator: bool = False,
+    chunksize: int | None = None,
+    **kwargs,
+):
+    """
+    Read from the store, close it if we opened it.
+
+    Retrieve pandas object stored in file, optionally based on where
+    criteria.
+
+    .. warning::
+
+       Pandas uses PyTables for reading and writing HDF5 files, which allows
+       serializing object-dtype data with pickle when using the "fixed" format.
+       Loading pickled data received from untrusted sources can be unsafe.
+
+       See: https://docs.python.org/3/library/pickle.html for more.
+
+    Parameters
+    ----------
+    path_or_buf : str, path object, pandas.HDFStore
+        Any valid string path is acceptable. Only supports the local file system,
+        remote URLs and file-like objects are not supported.
+
+        If you want to pass in a path object, pandas accepts any
+        ``os.PathLike``.
+
+        Alternatively, pandas accepts an open :class:`pandas.HDFStore` object.
+
+    key : object, optional
+        The group identifier in the store. Can be omitted if the HDF file
+        contains a single pandas object.
+    mode : {'r', 'r+', 'a'}, default 'r'
+        Mode to use when opening the file. Ignored if path_or_buf is a
+        :class:`pandas.HDFStore`. Default is 'r'.
+    errors : str, default 'strict'
+        Specifies how encoding and decoding errors are to be handled.
+        See the errors argument for :func:`open` for a full list
+        of options.
+    where : list, optional
+        A list of Term (or convertible) objects.
+    start : int, optional
+        Row number to start selection.
+    stop : int, optional
+        Row number to stop selection.
+    columns : list, optional
+        A list of columns names to return.
+    iterator : bool, optional
+        Return an iterator object.
+    chunksize : int, optional
+        Number of rows to include in an iteration when using an iterator.
+    **kwargs
+        Additional keyword arguments passed to HDFStore.
+
+    Returns
+    -------
+    object
+        The selected object. Return type depends on the object stored.
+
+    See Also
+    --------
+    DataFrame.to_hdf : Write a HDF file from a DataFrame.
+    HDFStore : Low-level access to HDF files.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame([[1, 1.0, "a"]], columns=["x", "y", "z"])  # doctest: +SKIP
+    >>> df.to_hdf("./store.h5", "data")  # doctest: +SKIP
+    >>> reread = pd.read_hdf("./store.h5")  # doctest: +SKIP
+    """
+    if mode not in ["r", "r+", "a"]:
+        raise ValueError(
+            f"mode {mode} is not allowed while performing a read. "
+            f"Allowed modes are r, r+ and a."
+        )
+    # grab the scope
+    if where is not None:
+        where = _ensure_term(where, scope_level=1)
+
+    if isinstance(path_or_buf, HDFStore):
+        if not path_or_buf.is_open:
+            raise OSError("The HDFStore must be open for reading.")
+
+        store = path_or_buf
+        auto_close = False
+    else:
+        path_or_buf = stringify_path(path_or_buf)
+        if not isinstance(path_or_buf, str):
+            raise NotImplementedError(
+                "Support for generic buffers has not been implemented."
+            )
+        try:
+            exists = os.path.exists(path_or_buf)
+
+        # if filepath is too long
+        except (TypeError, ValueError):
+            exists = False
+
+        if not exists:
+            raise FileNotFoundError(f"File {path_or_buf} does not exist")
+
+        store = HDFStore(path_or_buf, mode=mode, errors=errors, **kwargs)
+        # can't auto open/close if we are using an iterator
+        # so delegate to the iterator
+        auto_close = True
+
+    try:
+        if key is None:
+            groups = store.groups()
+            if len(groups) == 0:
+                raise ValueError(
+                    "Dataset(s) incompatible with Pandas data types, "
+                    "not table, or no datasets found in HDF5 file."
+                )
+            candidate_only_group = groups[0]
+
+            # For the HDF file to have only one dataset, all other groups
+            # should then be metadata groups for that candidate group. (This
+            # assumes that the groups() method enumerates parent groups
+            # before their children.)
+            for group_to_check in groups[1:]:
+                if not _is_metadata_of(group_to_check, candidate_only_group):
+                    raise ValueError(
+                        "key must be provided when HDF5 "
+                        "file contains multiple datasets."
+                    )
+            key = candidate_only_group._v_pathname
+        return store.select(
+            key,
+            where=where,
+            start=start,
+            stop=stop,
+            columns=columns,
+            iterator=iterator,
+            chunksize=chunksize,
+            auto_close=auto_close,
+        )
+    except (ValueError, TypeError, LookupError):
+        if not isinstance(path_or_buf, HDFStore):
+            # if there is an error, close the store if we opened it.
+            with suppress(AttributeError):
+                store.close()
+
+        raise
+
+
+def _is_metadata_of(group: Node, parent_group: Node) -> bool:
+    """Check if a given group is a metadata group for a given parent_group."""
+    if group._v_depth <= parent_group._v_depth:
+        return False
+
+    current = group
+    while current._v_depth > 1:
+        parent = current._v_parent
+        if parent == parent_group and current._v_name == "meta":
+            return True
+        current = current._v_parent
+    return False
+
+
+class HDFStore:
+    """
+    Dict-like IO interface for storing pandas objects in PyTables.
+
+    Either Fixed or Table format.
+
+    .. warning::
+
+       Pandas uses PyTables for reading and writing HDF5 files, which allows
+       serializing object-dtype data with pickle when using the "fixed" format.
+       Loading pickled data received from untrusted sources can be unsafe.
+
+       See: https://docs.python.org/3/library/pickle.html for more.
+
+    Parameters
+    ----------
+    path : str
+        File path to HDF5 file.
+    mode : {'a', 'w', 'r', 'r+'}, default 'a'
+
+        ``'r'``
+            Read-only; no data can be modified.
+        ``'w'``
+            Write; a new file is created (an existing file with the same
+            name would be deleted).
+        ``'a'``
+            Append; an existing file is opened for reading and writing,
+            and if the file does not exist it is created.
+        ``'r+'``
+            It is similar to ``'a'``, but the file must already exist.
+    complevel : int, 0-9, default None
+        Specifies a compression level for data.
+        A value of 0 or None disables compression.
+    complib : {'zlib', 'lzo', 'bzip2', 'blosc'}, default 'zlib'
+        Specifies the compression library to be used.
+        These additional compressors for Blosc are supported
+        (default if no compressor specified: 'blosc:blosclz'):
+        {'blosc:blosclz', 'blosc:lz4', 'blosc:lz4hc', 'blosc:snappy',
+         'blosc:zlib', 'blosc:zstd'}.
+        Specifying a compression library which is not available issues
+        a ValueError.
+    fletcher32 : bool, default False
+        If applying compression use the fletcher32 checksum.
+    **kwargs
+        These parameters will be passed to the PyTables open_file method.
+
+    Examples
+    --------
+    >>> bar = pd.DataFrame(np.random.randn(10, 4))
+    >>> store = pd.HDFStore("test.h5")
+    >>> store["foo"] = bar  # write to HDF5
+    >>> bar = store["foo"]  # retrieve
+    >>> store.close()
+
+    **Create or load HDF5 file in-memory**
+
+    When passing the `driver` option to the PyTables open_file method through
+    **kwargs, the HDF5 file is loaded or created in-memory and will only be
+    written when closed:
+
+    >>> bar = pd.DataFrame(np.random.randn(10, 4))
+    >>> store = pd.HDFStore("test.h5", driver="H5FD_CORE")
+    >>> store["foo"] = bar
+    >>> store.close()  # only now, data is written to disk
+    """
+
+    _handle: File | None
+    _mode: str
+
+    def __init__(
+        self,
+        path,
+        mode: str = "a",
+        complevel: int | None = None,
+        complib=None,
+        fletcher32: bool = False,
+        **kwargs,
+    ) -> None:
+        if "format" in kwargs:
+            raise ValueError("format is not a defined argument for HDFStore")
+
+        tables = import_optional_dependency("tables")
+
+        if complib is not None and complib not in tables.filters.all_complibs:
+            raise ValueError(
+                f"complib only supports {tables.filters.all_complibs} compression."
+            )
+
+        if complib is None and complevel is not None:
+            complib = tables.filters.default_complib
+
+        self._path = stringify_path(path)
+        if mode is None:
+            mode = "a"
+        self._mode = mode
+        self._handle = None
+        self._complevel = complevel if complevel else 0
+        self._complib = complib
+        self._fletcher32 = fletcher32
+        self._filters = None
+        self.open(mode=mode, **kwargs)
+
+    def __fspath__(self) -> str:
+        return self._path
+
+    @property
+    def root(self):
+        """return the root node"""
+        self._check_if_open()
+        assert self._handle is not None  # for mypy
+        return self._handle.root
+
+    @property
+    def filename(self) -> str:
+        return self._path
+
+    def __getitem__(self, key: str):
+        return self.get(key)
+
+    def __setitem__(self, key: str, value) -> None:
+        self.put(key, value)
+
+    def __delitem__(self, key: str) -> int | None:
+        return self.remove(key)
+
+    def __getattr__(self, name: str):
+        """allow attribute access to get stores"""
+        try:
+            return self.get(name)
+        except (KeyError, ClosedFileError):
+            pass
+        raise AttributeError(
+            f"'{type(self).__name__}' object has no attribute '{name}'"
+        )
+
+    def __contains__(self, key: str) -> bool:
+        """
+        check for existence of this key
+        can match the exact pathname or the pathnm w/o the leading '/'
+        """
+        node = self.get_node(key)
+        if node is not None:
+            name = node._v_pathname
+            if key in (name, name[1:]):
+                return True
+        return False
+
+    def __len__(self) -> int:
+        return len(self.groups())
+
+    def __repr__(self) -> str:
+        pstr = pprint_thing(self._path)
+        return f"{type(self)}\nFile path: {pstr}\n"
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+    def keys(self, include: str = "pandas") -> list[str]:
+        """
+        Return a list of keys corresponding to objects stored in HDFStore.
+
+        Parameters
+        ----------
+
+        include : str, default 'pandas'
+                When kind equals 'pandas' return pandas objects.
+                When kind equals 'native' return native HDF5 Table objects.
+
+        Returns
+        -------
+        list
+            List of ABSOLUTE path-names (e.g. have the leading '/').
+
+        Raises
+        ------
+        raises ValueError if kind has an illegal value
+
+        See Also
+        --------
+        HDFStore.info : Prints detailed information on the store.
+        HDFStore.get_node : Returns the node with the key.
+        HDFStore.get_storer : Returns the storer object for a key.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df)  # doctest: +SKIP
+        >>> store.get("data")  # doctest: +SKIP
+        >>> print(store.keys())  # doctest: +SKIP
+        ['/data1', '/data2']
+        >>> store.close()  # doctest: +SKIP
+        """
+        if include == "pandas":
+            return [n._v_pathname for n in self.groups()]
+
+        elif include == "native":
+            assert self._handle is not None  # mypy
+            return [
+                n._v_pathname for n in self._handle.walk_nodes("/", classname="Table")
+            ]
+        raise ValueError(
+            f"`include` should be either 'pandas' or 'native' but is '{include}'"
+        )
+
+    def __iter__(self) -> Iterator[str]:
+        return iter(self.keys())
+
+    def items(self) -> Iterator[tuple[str, list]]:
+        """
+        iterate on key->group
+        """
+        for g in self.groups():
+            yield g._v_pathname, g
+
+    def open(self, mode: str = "a", **kwargs) -> None:
+        """
+        Open the file in the specified mode
+
+        Parameters
+        ----------
+        mode : {'a', 'w', 'r', 'r+'}, default 'a'
+            See HDFStore docstring or tables.open_file for info about modes
+        **kwargs
+            These parameters will be passed to the PyTables open_file method.
+        """
+        tables = _tables()
+
+        if self._mode != mode:
+            # if we are changing a write mode to read, ok
+            if self._mode in ["a", "w"] and mode in ["r", "r+"]:
+                pass
+            elif mode in ["w"]:
+                # this would truncate, raise here
+                if self.is_open:
+                    raise PossibleDataLossError(
+                        f"Re-opening the file [{self._path}] with mode [{self._mode}] "
+                        "will delete the current file!"
+                    )
+
+            self._mode = mode
+
+        # close and reopen the handle
+        if self.is_open:
+            self.close()
+
+        if self._complevel and self._complevel > 0:
+            self._filters = _tables().Filters(
+                self._complevel, self._complib, fletcher32=self._fletcher32
+            )
+
+        if _table_file_open_policy_is_strict and self.is_open:
+            msg = (
+                "Cannot open HDF5 file, which is already opened, "
+                "even in read-only mode."
+            )
+            raise ValueError(msg)
+
+        self._handle = tables.open_file(self._path, self._mode, **kwargs)
+
+    def close(self) -> None:
+        """
+        Close the PyTables file handle
+        """
+        if self._handle is not None:
+            self._handle.close()
+        self._handle = None
+
+    @property
+    def is_open(self) -> bool:
+        """
+        return a boolean indicating whether the file is open
+        """
+        if self._handle is None:
+            return False
+        return bool(self._handle.isopen)
+
+    def flush(self, fsync: bool = False) -> None:
+        """
+        Force all buffered modifications to be written to disk.
+
+        Parameters
+        ----------
+        fsync : bool (default False)
+          call ``os.fsync()`` on the file handle to force writing to disk.
+
+        Notes
+        -----
+        Without ``fsync=True``, flushing may not guarantee that the OS writes
+        to disk. With fsync, the operation will block until the OS claims the
+        file has been written; however, other caching layers may still
+        interfere.
+        """
+        if self._handle is not None:
+            self._handle.flush()
+            if fsync:
+                with suppress(OSError):
+                    os.fsync(self._handle.fileno())
+
+    def get(self, key: str):
+        """
+        Retrieve pandas object stored in file.
+
+        Parameters
+        ----------
+        key : str
+            Object to retrieve from file. Raises KeyError if not found.
+
+        Returns
+        -------
+        object
+            Same type as object stored in file.
+
+        See Also
+        --------
+        HDFStore.get_node : Returns the node with the key.
+        HDFStore.get_storer : Returns the storer object for a key.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df)  # doctest: +SKIP
+        >>> store.get("data")  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+        """
+        with patch_pickle():
+            # GH#31167 Without this patch, pickle doesn't know how to unpickle
+            #  old DateOffset objects now that they are cdef classes.
+            group = self.get_node(key)
+            if group is None:
+                raise KeyError(f"No object named {key} in the file")
+            return self._read_group(group)
+
+    def select(
+        self,
+        key: str,
+        where=None,
+        start=None,
+        stop=None,
+        columns=None,
+        iterator: bool = False,
+        chunksize: int | None = None,
+        auto_close: bool = False,
+    ):
+        """
+        Retrieve pandas object stored in file, optionally based on where criteria.
+
+        .. warning::
+
+           Pandas uses PyTables for reading and writing HDF5 files, which allows
+           serializing object-dtype data with pickle when using the "fixed" format.
+           Loading pickled data received from untrusted sources can be unsafe.
+
+           See: https://docs.python.org/3/library/pickle.html for more.
+
+        Parameters
+        ----------
+        key : str
+            Object being retrieved from file.
+        where : list or None
+            List of Term (or convertible) objects, optional.
+        start : int or None
+            Row number to start selection.
+        stop : int, default None
+            Row number to stop selection.
+        columns : list or None
+            A list of columns that if not None, will limit the return columns.
+        iterator : bool or False
+            Returns an iterator.
+        chunksize : int or None
+            Number or rows to include in iteration, return an iterator.
+        auto_close : bool or False
+            Should automatically close the store when finished.
+
+        Returns
+        -------
+        object
+            Retrieved object from file.
+
+        See Also
+        --------
+        HDFStore.select_as_coordinates : Returns the selection as an index.
+        HDFStore.select_column : Returns a single column from the table.
+        HDFStore.select_as_multiple : Retrieves pandas objects from multiple tables.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df)  # doctest: +SKIP
+        >>> store.get("data")  # doctest: +SKIP
+        >>> print(store.keys())  # doctest: +SKIP
+        ['/data1', '/data2']
+        >>> store.select("/data1")  # doctest: +SKIP
+           A  B
+        0  1  2
+        1  3  4
+        >>> store.select("/data1", where="columns == A")  # doctest: +SKIP
+           A
+        0  1
+        1  3
+        >>> store.close()  # doctest: +SKIP
+        """
+        group = self.get_node(key)
+        if group is None:
+            raise KeyError(f"No object named {key} in the file")
+
+        # create the storer and axes
+        where = _ensure_term(where, scope_level=1)
+        s = self._create_storer(group)
+        s.infer_axes()
+
+        # function to call on iteration
+        def func(_start, _stop, _where):
+            return s.read(start=_start, stop=_stop, where=_where, columns=columns)
+
+        # create the iterator
+        it = TableIterator(
+            self,
+            s,
+            func,
+            where=where,
+            nrows=s.nrows,
+            start=start,
+            stop=stop,
+            iterator=iterator,
+            chunksize=chunksize,
+            auto_close=auto_close,
+        )
+
+        return it.get_result()
+
+    def select_as_coordinates(
+        self,
+        key: str,
+        where=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ):
+        """
+        return the selection as an Index
+
+        .. warning::
+
+           Pandas uses PyTables for reading and writing HDF5 files, which allows
+           serializing object-dtype data with pickle when using the "fixed" format.
+           Loading pickled data received from untrusted sources can be unsafe.
+
+           See: https://docs.python.org/3/library/pickle.html for more.
+
+
+        Parameters
+        ----------
+        key : str
+        where : list of Term (or convertible) objects, optional
+        start : integer (defaults to None), row number to start selection
+        stop  : integer (defaults to None), row number to stop selection
+        """
+        where = _ensure_term(where, scope_level=1)
+        tbl = self.get_storer(key)
+        if not isinstance(tbl, Table):
+            raise TypeError("can only read_coordinates with a table")
+        return tbl.read_coordinates(where=where, start=start, stop=stop)
+
+    def select_column(
+        self,
+        key: str,
+        column: str,
+        start: int | None = None,
+        stop: int | None = None,
+    ):
+        """
+        return a single column from the table. This is generally only useful to
+        select an indexable
+
+        .. warning::
+
+           Pandas uses PyTables for reading and writing HDF5 files, which allows
+           serializing object-dtype data with pickle when using the "fixed" format.
+           Loading pickled data received from untrusted sources can be unsafe.
+
+           See: https://docs.python.org/3/library/pickle.html for more.
+
+        Parameters
+        ----------
+        key : str
+        column : str
+            The column of interest.
+        start : int or None, default None
+        stop : int or None, default None
+
+        Raises
+        ------
+        raises KeyError if the column is not found (or key is not a valid
+            store)
+        raises ValueError if the column can not be extracted individually (it
+            is part of a data block)
+
+        """
+        tbl = self.get_storer(key)
+        if not isinstance(tbl, Table):
+            raise TypeError("can only read_column with a table")
+        return tbl.read_column(column=column, start=start, stop=stop)
+
+    def select_as_multiple(
+        self,
+        keys,
+        where=None,
+        selector=None,
+        columns=None,
+        start=None,
+        stop=None,
+        iterator: bool = False,
+        chunksize: int | None = None,
+        auto_close: bool = False,
+    ):
+        """
+        Retrieve pandas objects from multiple tables.
+
+        .. warning::
+
+           Pandas uses PyTables for reading and writing HDF5 files, which allows
+           serializing object-dtype data with pickle when using the "fixed" format.
+           Loading pickled data received from untrusted sources can be unsafe.
+
+           See: https://docs.python.org/3/library/pickle.html for more.
+
+        Parameters
+        ----------
+        keys : a list of the tables
+        selector : the table to apply the where criteria (defaults to keys[0]
+            if not supplied)
+        columns : the columns I want back
+        start : integer (defaults to None), row number to start selection
+        stop  : integer (defaults to None), row number to stop selection
+        iterator : bool, return an iterator, default False
+        chunksize : nrows to include in iteration, return an iterator
+        auto_close : bool, default False
+            Should automatically close the store when finished.
+
+        Raises
+        ------
+        raises KeyError if keys or selector is not found or keys is empty
+        raises TypeError if keys is not a list or tuple
+        raises ValueError if the tables are not ALL THE SAME DIMENSIONS
+        """
+        # default to single select
+        where = _ensure_term(where, scope_level=1)
+        if isinstance(keys, (list, tuple)) and len(keys) == 1:
+            keys = keys[0]
+        if isinstance(keys, str):
+            return self.select(
+                key=keys,
+                where=where,
+                columns=columns,
+                start=start,
+                stop=stop,
+                iterator=iterator,
+                chunksize=chunksize,
+                auto_close=auto_close,
+            )
+
+        if not isinstance(keys, (list, tuple)):
+            raise TypeError("keys must be a list/tuple")
+
+        if not len(keys):
+            raise ValueError("keys must have a non-zero length")
+
+        if selector is None:
+            selector = keys[0]
+
+        # collect the tables
+        tbls = [self.get_storer(k) for k in keys]
+        s = self.get_storer(selector)
+
+        # validate rows
+        nrows = None
+        for t, k in itertools.chain([(s, selector)], zip(tbls, keys)):
+            if t is None:
+                raise KeyError(f"Invalid table [{k}]")
+            if not t.is_table:
+                raise TypeError(
+                    f"object [{t.pathname}] is not a table, and cannot be used in all "
+                    "select as multiple"
+                )
+
+            if nrows is None:
+                nrows = t.nrows
+            elif t.nrows != nrows:
+                raise ValueError("all tables must have exactly the same nrows!")
+
+        # The isinstance checks here are redundant with the check above,
+        #  but necessary for mypy; see GH#29757
+        _tbls = [x for x in tbls if isinstance(x, Table)]
+
+        # axis is the concentration axes
+        axis = {t.non_index_axes[0][0] for t in _tbls}.pop()
+
+        def func(_start, _stop, _where):
+            # retrieve the objs, _where is always passed as a set of
+            # coordinates here
+            objs = [
+                t.read(where=_where, columns=columns, start=_start, stop=_stop)
+                for t in tbls
+            ]
+
+            # concat and return
+            return concat(objs, axis=axis, verify_integrity=False)._consolidate()
+
+        # create the iterator
+        it = TableIterator(
+            self,
+            s,
+            func,
+            where=where,
+            nrows=nrows,
+            start=start,
+            stop=stop,
+            iterator=iterator,
+            chunksize=chunksize,
+            auto_close=auto_close,
+        )
+
+        return it.get_result(coordinates=True)
+
+    def put(
+        self,
+        key: str,
+        value: DataFrame | Series,
+        format=None,
+        index: bool = True,
+        append: bool = False,
+        complib=None,
+        complevel: int | None = None,
+        min_itemsize: int | dict[str, int] | None = None,
+        nan_rep=None,
+        data_columns: Literal[True] | list[str] | None = None,
+        encoding=None,
+        errors: str = "strict",
+        track_times: bool = True,
+        dropna: bool = False,
+    ) -> None:
+        """
+        Store object in HDFStore.
+
+        This method writes a pandas DataFrame or Series into an HDF5 file using
+        either the fixed or table format. The `table` format allows additional
+        operations like incremental appends and queries but may have performance
+        trade-offs. The `fixed` format provides faster read/write operations but
+        does not support appends or queries.
+
+        Parameters
+        ----------
+        key : str
+            Key of object to store in file.
+        value : {Series, DataFrame}
+            Value of object to store in file.
+        format : 'fixed(f)|table(t)', default is 'fixed'
+            Format to use when storing object in HDFStore. Value can be one of:
+
+            ``'fixed'``
+                Fixed format.  Fast writing/reading. Not-appendable, nor searchable.
+            ``'table'``
+                Table format.  Write as a PyTables Table structure which may perform
+                worse but allow more flexible operations like searching / selecting
+                subsets of the data.
+        index : bool, default True
+            Write DataFrame index as a column.
+        append : bool, default False
+            This will force Table format, append the input data to the existing.
+        complib : default None
+            This parameter is currently not accepted.
+        complevel : int, 0-9, default None
+            Specifies a compression level for data.
+            A value of 0 or None disables compression.
+        min_itemsize : int, dict, or None
+            Dict of columns that specify minimum str sizes.
+        nan_rep : str
+            Str to use as str nan representation.
+        data_columns : list of columns or True, default None
+            List of columns to create as data columns, or True to use all columns.
+            See `here
+            <https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#query-via-data-columns>`__.
+        encoding : str, default None
+            Provide an encoding for strings.
+        errors : str, default 'strict'
+            The error handling scheme to use for encoding errors.
+            The default is 'strict' meaning that encoding errors raise a
+            UnicodeEncodeError.  Other possible values are 'ignore', 'replace' and
+            'xmlcharrefreplace' as well as any other name registered with
+            codecs.register_error that can handle UnicodeEncodeErrors.
+        track_times : bool, default True
+            Parameter is propagated to 'create_table' method of 'PyTables'.
+            If set to False it enables to have the same h5 files (same hashes)
+            independent on creation time.
+        dropna : bool, default False, optional
+            Remove missing values.
+
+        See Also
+        --------
+        HDFStore.info : Prints detailed information on the store.
+        HDFStore.get_storer : Returns the storer object for a key.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df)  # doctest: +SKIP
+        """
+        if format is None:
+            format = get_option("io.hdf.default_format") or "fixed"
+        format = self._validate_format(format)
+        self._write_to_group(
+            key,
+            value,
+            format=format,
+            index=index,
+            append=append,
+            complib=complib,
+            complevel=complevel,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            data_columns=data_columns,
+            encoding=encoding,
+            errors=errors,
+            track_times=track_times,
+            dropna=dropna,
+        )
+
+    def remove(self, key: str, where=None, start=None, stop=None) -> int | None:
+        """
+        Remove pandas object partially by specifying the where condition
+
+        Parameters
+        ----------
+        key : str
+            Node to remove or delete rows from
+        where : list of Term (or convertible) objects, optional
+        start : integer (defaults to None), row number to start selection
+        stop  : integer (defaults to None), row number to stop selection
+
+        Returns
+        -------
+        number of rows removed (or None if not a Table)
+
+        Raises
+        ------
+        raises KeyError if key is not a valid store
+
+        """
+        where = _ensure_term(where, scope_level=1)
+        try:
+            s = self.get_storer(key)
+        except KeyError:
+            # the key is not a valid store, re-raising KeyError
+            raise
+        except AssertionError:
+            # surface any assertion errors for e.g. debugging
+            raise
+        except Exception as err:
+            # In tests we get here with ClosedFileError, TypeError, and
+            #  _table_mod.NoSuchNodeError.  TODO: Catch only these?
+
+            if where is not None:
+                raise ValueError(
+                    "trying to remove a node with a non-None where clause!"
+                ) from err
+
+            # we are actually trying to remove a node (with children)
+            node = self.get_node(key)
+            if node is not None:
+                node._f_remove(recursive=True)
+                return None
+
+        # remove the node
+        if com.all_none(where, start, stop):
+            s.group._f_remove(recursive=True)
+            return None
+
+        # delete from the table
+        if not s.is_table:
+            raise ValueError("can only remove with where on objects written as tables")
+        return s.delete(where=where, start=start, stop=stop)
+
+    def append(
+        self,
+        key: str,
+        value: DataFrame | Series,
+        format=None,
+        axes=None,
+        index: bool | list[str] = True,
+        append: bool = True,
+        complib=None,
+        complevel: int | None = None,
+        columns=None,
+        min_itemsize: int | dict[str, int] | None = None,
+        nan_rep=None,
+        chunksize: int | None = None,
+        expectedrows=None,
+        dropna: bool | None = None,
+        data_columns: Literal[True] | list[str] | None = None,
+        encoding=None,
+        errors: str = "strict",
+    ) -> None:
+        """
+        Append to Table in file.
+
+        Node must already exist and be Table format.
+
+        Parameters
+        ----------
+        key : str
+            Key of object to append.
+        value : {Series, DataFrame}
+            Value of object to append.
+        format : 'table' is the default
+            Format to use when storing object in HDFStore.  Value can be one of:
+
+            ``'table'``
+                Table format. Write as a PyTables Table structure which may perform
+                worse but allow more flexible operations like searching / selecting
+                subsets of the data.
+        axes : default None
+            This parameter is currently not accepted.
+        index : bool, default True
+            Write DataFrame index as a column.
+        append : bool, default True
+            Append the input data to the existing.
+        complib : default None
+            This parameter is currently not accepted.
+        complevel : int, 0-9, default None
+            Specifies a compression level for data.
+            A value of 0 or None disables compression.
+        columns : default None
+            This parameter is currently not accepted, try data_columns.
+        min_itemsize : int, dict, or None
+            Dict of columns that specify minimum str sizes.
+        nan_rep : str
+            Str to use as str nan representation.
+        chunksize : int or None
+            Size to chunk the writing.
+        expectedrows : int
+            Expected TOTAL row size of this table.
+        dropna : bool, default False, optional
+            Do not write an ALL nan row to the store settable
+            by the option 'io.hdf.dropna_table'.
+        data_columns : list of columns, or True, default None
+            List of columns to create as indexed data columns for on-disk
+            queries, or True to use all columns. By default only the axes
+            of the object are indexed. See `here
+            <https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html#query-via-data-columns>`__.
+        encoding : default None
+            Provide an encoding for str.
+        errors : str, default 'strict'
+            The error handling scheme to use for encoding errors.
+            The default is 'strict' meaning that encoding errors raise a
+            UnicodeEncodeError.  Other possible values are 'ignore', 'replace' and
+            'xmlcharrefreplace' as well as any other name registered with
+            codecs.register_error that can handle UnicodeEncodeErrors.
+
+        See Also
+        --------
+        HDFStore.append_to_multiple : Append to multiple tables.
+
+        Notes
+        -----
+        Does *not* check if data being appended overlaps with existing
+        data in the table, so be careful
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df1, format="table")  # doctest: +SKIP
+        >>> df2 = pd.DataFrame([[5, 6], [7, 8]], columns=["A", "B"])
+        >>> store.append("data", df2)  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+           A  B
+        0  1  2
+        1  3  4
+        0  5  6
+        1  7  8
+        """
+        if columns is not None:
+            raise TypeError(
+                "columns is not a supported keyword in append, try data_columns"
+            )
+
+        if dropna is None:
+            dropna = get_option("io.hdf.dropna_table")
+        if format is None:
+            format = get_option("io.hdf.default_format") or "table"
+        format = self._validate_format(format)
+        self._write_to_group(
+            key,
+            value,
+            format=format,
+            axes=axes,
+            index=index,
+            append=append,
+            complib=complib,
+            complevel=complevel,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            chunksize=chunksize,
+            expectedrows=expectedrows,
+            dropna=dropna,
+            data_columns=data_columns,
+            encoding=encoding,
+            errors=errors,
+        )
+
+    def append_to_multiple(
+        self,
+        d: dict,
+        value,
+        selector,
+        data_columns=None,
+        axes=None,
+        dropna: bool = False,
+        **kwargs,
+    ) -> None:
+        """
+        Append to multiple tables
+
+        Parameters
+        ----------
+        d : a dict of table_name to table_columns, None is acceptable as the
+            values of one node (this will get all the remaining columns)
+        value : a pandas object
+        selector : a string that designates the indexable table; all of its
+            columns will be designed as data_columns, unless data_columns is
+            passed, in which case these are used
+        data_columns : list of columns to create as data columns, or True to
+            use all columns
+        dropna : if evaluates to True, drop rows from all tables if any single
+                 row in each table has all NaN. Default False.
+
+        Notes
+        -----
+        axes parameter is currently not accepted
+
+        """
+        if axes is not None:
+            raise TypeError(
+                "axes is currently not accepted as a parameter to append_to_multiple; "
+                "you can create the tables independently instead"
+            )
+
+        if not isinstance(d, dict):
+            raise ValueError(
+                "append_to_multiple must have a dictionary specified as the "
+                "way to split the value"
+            )
+
+        if selector not in d:
+            raise ValueError(
+                "append_to_multiple requires a selector that is in passed dict"
+            )
+
+        # figure out the splitting axis (the non_index_axis)
+        axis = next(iter(set(range(value.ndim)) - set(_AXES_MAP[type(value)])))
+
+        # figure out how to split the value
+        remain_key = None
+        remain_values: list = []
+        for k, v in d.items():
+            if v is None:
+                if remain_key is not None:
+                    raise ValueError(
+                        "append_to_multiple can only have one value in d that is None"
+                    )
+                remain_key = k
+            else:
+                remain_values.extend(v)
+        if remain_key is not None:
+            ordered = value.axes[axis]
+            ordd = ordered.difference(Index(remain_values))
+            ordd = sorted(ordered.get_indexer(ordd))
+            d[remain_key] = ordered.take(ordd)
+
+        # data_columns
+        if data_columns is None:
+            data_columns = d[selector]
+
+        # ensure rows are synchronized across the tables
+        if dropna:
+            idxs = (value[cols].dropna(how="all").index for cols in d.values())
+            valid_index = next(idxs)
+            for index in idxs:
+                valid_index = valid_index.intersection(index)
+            value = value.loc[valid_index]
+
+        min_itemsize = kwargs.pop("min_itemsize", None)
+
+        # append
+        for k, v in d.items():
+            dc = data_columns if k == selector else None
+
+            # compute the val
+            val = value.reindex(v, axis=axis)
+
+            filtered = (
+                {key: value for (key, value) in min_itemsize.items() if key in v}
+                if min_itemsize is not None
+                else None
+            )
+            self.append(k, val, data_columns=dc, min_itemsize=filtered, **kwargs)
+
+    def create_table_index(
+        self,
+        key: str,
+        columns=None,
+        optlevel: int | None = None,
+        kind: str | None = None,
+    ) -> None:
+        """
+        Create a pytables index on the table.
+
+        Parameters
+        ----------
+        key : str
+        columns : None, bool, or listlike[str]
+            Indicate which columns to create an index on.
+
+            * False : Do not create any indexes.
+            * True : Create indexes on all columns.
+            * None : Create indexes on all columns.
+            * listlike : Create indexes on the given columns.
+
+        optlevel : int or None, default None
+            Optimization level, if None, pytables defaults to 6.
+        kind : str or None, default None
+            Kind of index, if None, pytables defaults to "medium".
+
+        Raises
+        ------
+        TypeError: raises if the node is not a table
+        """
+        # version requirements
+        _tables()
+        s = self.get_storer(key)
+        if s is None:
+            return
+
+        if not isinstance(s, Table):
+            raise TypeError("cannot create table index on a Fixed format store")
+        s.create_index(columns=columns, optlevel=optlevel, kind=kind)
+
+    def groups(self) -> list:
+        """
+        Return a list of all the top-level nodes.
+
+        Each node returned is not a pandas storage object.
+
+        Returns
+        -------
+        list
+            List of objects.
+
+        See Also
+        --------
+        HDFStore.get_node : Returns the node with the key.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df)  # doctest: +SKIP
+        >>> print(store.groups())  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+        [/data (Group) ''
+          children := ['axis0' (Array), 'axis1' (Array), 'block0_values' (Array),
+          'block0_items' (Array)]]
+        """
+        _tables()
+        self._check_if_open()
+        assert self._handle is not None  # for mypy
+        assert _table_mod is not None  # for mypy
+        return [
+            g
+            for g in self._handle.walk_groups()
+            if (
+                not isinstance(g, _table_mod.link.Link)
+                and (
+                    getattr(g._v_attrs, "pandas_type", None)
+                    or getattr(g, "table", None)
+                    or (isinstance(g, _table_mod.table.Table) and g._v_name != "table")
+                )
+            )
+        ]
+
+    def walk(self, where: str = "/") -> Iterator[tuple[str, list[str], list[str]]]:
+        """
+        Walk the pytables group hierarchy for pandas objects.
+
+        This generator will yield the group path, subgroups and pandas object
+        names for each group.
+
+        Any non-pandas PyTables objects that are not a group will be ignored.
+
+        The `where` group itself is listed first (preorder), then each of its
+        child groups (following an alphanumerical order) is also traversed,
+        following the same procedure.
+
+        Parameters
+        ----------
+        where : str, default "/"
+            Group where to start walking.
+
+        Yields
+        ------
+        path : str
+            Full path to a group (without trailing '/').
+        groups : list
+            Names (strings) of the groups contained in `path`.
+        leaves : list
+            Names (strings) of the pandas objects contained in `path`.
+
+        See Also
+        --------
+        HDFStore.info : Prints detailed information on the store.
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data", df1, format="table")  # doctest: +SKIP
+        >>> df2 = pd.DataFrame([[5, 6], [7, 8]], columns=["A", "B"])
+        >>> store.append("data", df2)  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+        >>> for group in store.walk():  # doctest: +SKIP
+        ...     print(group)  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+        """
+        _tables()
+        self._check_if_open()
+        assert self._handle is not None  # for mypy
+        assert _table_mod is not None  # for mypy
+
+        for g in self._handle.walk_groups(where):
+            if getattr(g._v_attrs, "pandas_type", None) is not None:
+                continue
+
+            groups = []
+            leaves = []
+            for child in g._v_children.values():
+                pandas_type = getattr(child._v_attrs, "pandas_type", None)
+                if pandas_type is None:
+                    if isinstance(child, _table_mod.group.Group):
+                        groups.append(child._v_name)
+                else:
+                    leaves.append(child._v_name)
+
+            yield (g._v_pathname.rstrip("/"), groups, leaves)
+
+    def get_node(self, key: str) -> Node | None:
+        """return the node with the key or None if it does not exist"""
+        self._check_if_open()
+        if not key.startswith("/"):
+            key = "/" + key
+
+        assert self._handle is not None
+        assert _table_mod is not None  # for mypy
+        try:
+            node = self._handle.get_node(self.root, key)
+        except _table_mod.exceptions.NoSuchNodeError:
+            return None
+
+        assert isinstance(node, _table_mod.Node), type(node)
+        return node
+
+    def get_storer(self, key: str) -> GenericFixed | Table:
+        """return the storer object for a key, raise if not in the file"""
+        group = self.get_node(key)
+        if group is None:
+            raise KeyError(f"No object named {key} in the file")
+
+        s = self._create_storer(group)
+        s.infer_axes()
+        return s
+
+    def copy(
+        self,
+        file,
+        mode: str = "w",
+        propindexes: bool = True,
+        keys=None,
+        complib=None,
+        complevel: int | None = None,
+        fletcher32: bool = False,
+        overwrite: bool = True,
+    ) -> HDFStore:
+        """
+        Copy the existing store to a new file, updating in place.
+
+        Parameters
+        ----------
+        propindexes : bool, default True
+            Restore indexes in copied file.
+        keys : list, optional
+            List of keys to include in the copy (defaults to all).
+        overwrite : bool, default True
+            Whether to overwrite (remove and replace) existing nodes in the new store.
+        mode, complib, complevel, fletcher32 same as in HDFStore.__init__
+
+        Returns
+        -------
+        open file handle of the new store
+        """
+        new_store = HDFStore(
+            file, mode=mode, complib=complib, complevel=complevel, fletcher32=fletcher32
+        )
+        if keys is None:
+            keys = list(self.keys())
+        if not isinstance(keys, (tuple, list)):
+            keys = [keys]
+        for k in keys:
+            s = self.get_storer(k)
+            if s is not None:
+                if k in new_store:
+                    if overwrite:
+                        new_store.remove(k)
+
+                data = self.select(k)
+                if isinstance(s, Table):
+                    index: bool | list[str] = False
+                    if propindexes:
+                        index = [a.name for a in s.axes if a.is_indexed]
+                    new_store.append(
+                        k,
+                        data,
+                        index=index,
+                        data_columns=getattr(s, "data_columns", None),
+                        encoding=s.encoding,
+                    )
+                else:
+                    new_store.put(k, data, encoding=s.encoding)
+
+        return new_store
+
+    def info(self) -> str:
+        """
+        Print detailed information on the store.
+
+        Returns
+        -------
+        str
+            A String containing the python pandas class name, filepath to the HDF5
+            file and all the object keys along with their respective dataframe shapes.
+
+        See Also
+        --------
+        HDFStore.get_storer : Returns the storer object for a key.
+
+        Examples
+        --------
+        >>> df1 = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        >>> df2 = pd.DataFrame([[5, 6], [7, 8]], columns=["C", "D"])
+        >>> store = pd.HDFStore("store.h5", "w")  # doctest: +SKIP
+        >>> store.put("data1", df1)  # doctest: +SKIP
+        >>> store.put("data2", df2)  # doctest: +SKIP
+        >>> print(store.info())  # doctest: +SKIP
+        >>> store.close()  # doctest: +SKIP
+        <class 'pandas.io.pytables.HDFStore'>
+        File path: store.h5
+        /data1            frame        (shape->[2,2])
+        /data2            frame        (shape->[2,2])
+        """
+        path = pprint_thing(self._path)
+        output = f"{type(self)}\nFile path: {path}\n"
+
+        if self.is_open:
+            lkeys = sorted(self.keys())
+            if len(lkeys):
+                keys = []
+                values = []
+
+                for k in lkeys:
+                    try:
+                        s = self.get_storer(k)
+                        if s is not None:
+                            keys.append(pprint_thing(s.pathname or k))
+                            values.append(pprint_thing(s or "invalid_HDFStore node"))
+                    except AssertionError:
+                        # surface any assertion errors for e.g. debugging
+                        raise
+                    except Exception as detail:
+                        keys.append(k)
+                        dstr = pprint_thing(detail)
+                        values.append(f"[invalid_HDFStore node: {dstr}]")
+
+                output += adjoin(12, keys, values)
+            else:
+                output += "Empty"
+        else:
+            output += "File is CLOSED"
+
+        return output
+
+    # ------------------------------------------------------------------------
+    # private methods
+
+    def _check_if_open(self) -> None:
+        if not self.is_open:
+            raise ClosedFileError(f"{self._path} file is not open!")
+
+    def _validate_format(self, format: str) -> str:
+        """validate / deprecate formats"""
+        # validate
+        try:
+            format = _FORMAT_MAP[format.lower()]
+        except KeyError as err:
+            raise TypeError(f"invalid HDFStore format specified [{format}]") from err
+
+        return format
+
+    def _create_storer(
+        self,
+        group,
+        format=None,
+        value: DataFrame | Series | None = None,
+        encoding: str = "UTF-8",
+        errors: str = "strict",
+    ) -> GenericFixed | Table:
+        """return a suitable class to operate"""
+        cls: type[GenericFixed | Table]
+
+        if value is not None and not isinstance(value, (Series, DataFrame)):
+            raise TypeError("value must be None, Series, or DataFrame")
+
+        pt = getattr(group._v_attrs, "pandas_type", None)
+        tt = getattr(group._v_attrs, "table_type", None)
+
+        # infer the pt from the passed value
+        if pt is None:
+            if value is None:
+                _tables()
+                assert _table_mod is not None  # for mypy
+                if getattr(group, "table", None) or isinstance(
+                    group, _table_mod.table.Table
+                ):
+                    pt = "frame_table"
+                    tt = "generic_table"
+                else:
+                    raise TypeError(
+                        "cannot create a storer if the object is not existing "
+                        "nor a value are passed"
+                    )
+            else:
+                if isinstance(value, Series):
+                    pt = "series"
+                else:
+                    pt = "frame"
+
+                # we are actually a table
+                if format == "table":
+                    pt += "_table"
+
+        # a storer node
+        if "table" not in pt:
+            _STORER_MAP = {"series": SeriesFixed, "frame": FrameFixed}
+            try:
+                cls = _STORER_MAP[pt]
+            except KeyError as err:
+                raise TypeError(
+                    f"cannot properly create the storer for: [_STORER_MAP] [group->"
+                    f"{group},value->{type(value)},format->{format}"
+                ) from err
+            return cls(self, group, encoding=encoding, errors=errors)
+
+        # existing node (and must be a table)
+        if tt is None:
+            # if we are a writer, determine the tt
+            if value is not None:
+                if pt == "series_table":
+                    index = getattr(value, "index", None)
+                    if index is not None:
+                        if index.nlevels == 1:
+                            tt = "appendable_series"
+                        elif index.nlevels > 1:
+                            tt = "appendable_multiseries"
+                elif pt == "frame_table":
+                    index = getattr(value, "index", None)
+                    if index is not None:
+                        if index.nlevels == 1:
+                            tt = "appendable_frame"
+                        elif index.nlevels > 1:
+                            tt = "appendable_multiframe"
+
+        _TABLE_MAP = {
+            "generic_table": GenericTable,
+            "appendable_series": AppendableSeriesTable,
+            "appendable_multiseries": AppendableMultiSeriesTable,
+            "appendable_frame": AppendableFrameTable,
+            "appendable_multiframe": AppendableMultiFrameTable,
+            "worm": WORMTable,
+        }
+        try:
+            cls = _TABLE_MAP[tt]  # type: ignore[index]
+        except KeyError as err:
+            raise TypeError(
+                f"cannot properly create the storer for: [_TABLE_MAP] [group->"
+                f"{group},value->{type(value)},format->{format}"
+            ) from err
+
+        return cls(self, group, encoding=encoding, errors=errors)
+
+    def _write_to_group(
+        self,
+        key: str,
+        value: DataFrame | Series,
+        format,
+        axes=None,
+        index: bool | list[str] = True,
+        append: bool = False,
+        complib=None,
+        complevel: int | None = None,
+        fletcher32=None,
+        min_itemsize: int | dict[str, int] | None = None,
+        chunksize: int | None = None,
+        expectedrows=None,
+        dropna: bool = False,
+        nan_rep=None,
+        data_columns=None,
+        encoding=None,
+        errors: str = "strict",
+        track_times: bool = True,
+    ) -> None:
+        # we don't want to store a table node at all if our object is 0-len
+        # as there are not dtypes
+        if getattr(value, "empty", None) and (format == "table" or append):
+            return
+
+        group = self._identify_group(key, append)
+
+        s = self._create_storer(group, format, value, encoding=encoding, errors=errors)
+        if append:
+            # raise if we are trying to append to a Fixed format,
+            #       or a table that exists (and we are putting)
+            if not s.is_table or (s.is_table and format == "fixed" and s.is_exists):
+                raise ValueError("Can only append to Tables")
+            if not s.is_exists:
+                s.set_object_info()
+        else:
+            s.set_object_info()
+
+        if not s.is_table and complib:
+            raise ValueError("Compression not supported on Fixed format stores")
+
+        # write the object
+        s.write(
+            obj=value,
+            axes=axes,
+            append=append,
+            complib=complib,
+            complevel=complevel,
+            fletcher32=fletcher32,
+            min_itemsize=min_itemsize,
+            chunksize=chunksize,
+            expectedrows=expectedrows,
+            dropna=dropna,
+            nan_rep=nan_rep,
+            data_columns=data_columns,
+            track_times=track_times,
+        )
+
+        if isinstance(s, Table) and index:
+            s.create_index(columns=index)
+
+    def _read_group(self, group: Node):
+        s = self._create_storer(group)
+        s.infer_axes()
+        return s.read()
+
+    def _identify_group(self, key: str, append: bool) -> Node:
+        """Identify HDF5 group based on key, delete/create group if needed."""
+        group = self.get_node(key)
+
+        # we make this assertion for mypy; the get_node call will already
+        # have raised if this is incorrect
+        assert self._handle is not None
+
+        # remove the node if we are not appending
+        if group is not None and not append:
+            self._handle.remove_node(group, recursive=True)
+            group = None
+
+        if group is None:
+            group = self._create_nodes_and_group(key)
+
+        return group
+
+    def _create_nodes_and_group(self, key: str) -> Node:
+        """Create nodes from key and return group name."""
+        # assertion for mypy
+        assert self._handle is not None
+
+        paths = key.split("/")
+        # recursively create the groups
+        path = "/"
+        for p in paths:
+            if not len(p):
+                continue
+            new_path = path
+            if not path.endswith("/"):
+                new_path += "/"
+            new_path += p
+            group = self.get_node(new_path)
+            if group is None:
+                group = self._handle.create_group(path, p)
+            path = new_path
+        return group
+
+
+class TableIterator:
+    """
+    Define the iteration interface on a table
+
+    Parameters
+    ----------
+    store : HDFStore
+    s     : the referred storer
+    func  : the function to execute the query
+    where : the where of the query
+    nrows : the rows to iterate on
+    start : the passed start value (default is None)
+    stop  : the passed stop value (default is None)
+    iterator : bool, default False
+        Whether to use the default iterator.
+    chunksize : the passed chunking value (default is 100000)
+    auto_close : bool, default False
+        Whether to automatically close the store at the end of iteration.
+    """
+
+    chunksize: int | None
+    store: HDFStore
+    s: GenericFixed | Table
+
+    def __init__(
+        self,
+        store: HDFStore,
+        s: GenericFixed | Table,
+        func,
+        where,
+        nrows,
+        start=None,
+        stop=None,
+        iterator: bool = False,
+        chunksize: int | None = None,
+        auto_close: bool = False,
+    ) -> None:
+        self.store = store
+        self.s = s
+        self.func = func
+        self.where = where
+
+        # set start/stop if they are not set if we are a table
+        if self.s.is_table:
+            if nrows is None:
+                nrows = 0
+            if start is None:
+                start = 0
+            if stop is None:
+                stop = nrows
+            stop = min(nrows, stop)
+
+        self.nrows = nrows
+        self.start = start
+        self.stop = stop
+
+        self.coordinates = None
+        if iterator or chunksize is not None:
+            if chunksize is None:
+                chunksize = 100000
+            self.chunksize = int(chunksize)
+        else:
+            self.chunksize = None
+
+        self.auto_close = auto_close
+
+    def __iter__(self) -> Iterator:
+        # iterate
+        current = self.start
+        if self.coordinates is None:
+            raise ValueError("Cannot iterate until get_result is called.")
+        while current < self.stop:
+            stop = min(current + self.chunksize, self.stop)
+            value = self.func(None, None, self.coordinates[current:stop])
+            current = stop
+            if value is None or not len(value):
+                continue
+
+            yield value
+
+        self.close()
+
+    def close(self) -> None:
+        if self.auto_close:
+            self.store.close()
+
+    def get_result(self, coordinates: bool = False):
+        #  return the actual iterator
+        if self.chunksize is not None:
+            if not isinstance(self.s, Table):
+                raise TypeError("can only use an iterator or chunksize on a table")
+
+            self.coordinates = self.s.read_coordinates(where=self.where)
+
+            return self
+
+        # if specified read via coordinates (necessary for multiple selections
+        if coordinates:
+            if not isinstance(self.s, Table):
+                raise TypeError("can only read_coordinates on a table")
+            where = self.s.read_coordinates(
+                where=self.where, start=self.start, stop=self.stop
+            )
+        else:
+            where = self.where
+
+        # directly return the result
+        results = self.func(self.start, self.stop, where)
+        self.close()
+        return results
+
+
+class IndexCol:
+    """
+    an index column description class
+
+    Parameters
+    ----------
+    axis   : axis which I reference
+    values : the ndarray like converted values
+    kind   : a string description of this type
+    typ    : the pytables type
+    pos    : the position in the pytables
+
+    """
+
+    is_an_indexable: bool = True
+    is_data_indexable: bool = True
+    _info_fields = ["freq", "tz", "index_name"]
+
+    def __init__(
+        self,
+        name: str,
+        values=None,
+        kind=None,
+        typ=None,
+        cname: str | None = None,
+        axis=None,
+        pos=None,
+        freq=None,
+        tz=None,
+        index_name=None,
+        ordered=None,
+        table=None,
+        meta=None,
+        metadata=None,
+    ) -> None:
+        if not isinstance(name, str):
+            raise ValueError("`name` must be a str.")
+
+        self.values = values
+        self.kind = kind
+        self.typ = typ
+        self.name = name
+        self.cname = cname or name
+        self.axis = axis
+        self.pos = pos
+        self.freq = freq
+        self.tz = tz
+        self.index_name = index_name
+        self.ordered = ordered
+        self.table = table
+        self.meta = meta
+        self.metadata = metadata
+
+        if pos is not None:
+            self.set_pos(pos)
+
+        # These are ensured as long as the passed arguments match the
+        #  constructor annotations.
+        assert isinstance(self.name, str)
+        assert isinstance(self.cname, str)
+
+    @property
+    def itemsize(self) -> int:
+        # Assumes self.typ has already been initialized
+        return self.typ.itemsize
+
+    @property
+    def kind_attr(self) -> str:
+        return f"{self.name}_kind"
+
+    def set_pos(self, pos: int) -> None:
+        """set the position of this column in the Table"""
+        self.pos = pos
+        if pos is not None and self.typ is not None:
+            self.typ._v_pos = pos
+
+    def __repr__(self) -> str:
+        temp = tuple(
+            map(pprint_thing, (self.name, self.cname, self.axis, self.pos, self.kind))
+        )
+        return ",".join(
+            [
+                f"{key}->{value}"
+                for key, value in zip(["name", "cname", "axis", "pos", "kind"], temp)
+            ]
+        )
+
+    def __eq__(self, other: object) -> bool:
+        """compare 2 col items"""
+        return all(
+            getattr(self, a, None) == getattr(other, a, None)
+            for a in ["name", "cname", "axis", "pos"]
+        )
+
+    def __ne__(self, other) -> bool:
+        return not self.__eq__(other)
+
+    @property
+    def is_indexed(self) -> bool:
+        """return whether I am an indexed column"""
+        if not hasattr(self.table, "cols"):
+            # e.g. if infer hasn't been called yet, self.table will be None.
+            return False
+        return getattr(self.table.cols, self.cname).is_indexed
+
+    def convert(
+        self, values: np.ndarray, nan_rep, encoding: str, errors: str
+    ) -> tuple[np.ndarray, np.ndarray] | tuple[Index, Index]:
+        """
+        Convert the data from this selection to the appropriate pandas type.
+        """
+        assert isinstance(values, np.ndarray), type(values)
+
+        # values is a recarray
+        if values.dtype.fields is not None:
+            # Copy, otherwise values will be a view
+            # preventing the original recarry from being free'ed
+            values = values[self.cname].copy()
+
+        val_kind = self.kind
+        values = _maybe_convert(values, val_kind, encoding, errors)
+        kwargs = {}
+        kwargs["name"] = self.index_name
+
+        if self.freq is not None:
+            kwargs["freq"] = self.freq
+
+        factory: type[Index | DatetimeIndex] = Index
+        if lib.is_np_dtype(values.dtype, "M") or isinstance(
+            values.dtype, DatetimeTZDtype
+        ):
+            factory = DatetimeIndex
+        elif values.dtype == "i8" and "freq" in kwargs:
+            # PeriodIndex data is stored as i8
+            # error: Incompatible types in assignment (expression has type
+            # "Callable[[Any, KwArg(Any)], PeriodIndex]", variable has type
+            # "Union[Type[Index], Type[DatetimeIndex]]")
+            factory = lambda x, **kwds: PeriodIndex.from_ordinals(  # type: ignore[assignment]
+                x, freq=kwds.get("freq", None)
+            )._rename(kwds["name"])
+
+        # making an Index instance could throw a number of different errors
+        try:
+            new_pd_index = factory(values, **kwargs)
+        except ValueError:
+            # if the output freq is different that what we recorded,
+            # it should be None (see also 'doc example part 2')
+            if "freq" in kwargs:
+                kwargs["freq"] = None
+            new_pd_index = factory(values, **kwargs)
+
+        final_pd_index: Index
+        if self.tz is not None and isinstance(new_pd_index, DatetimeIndex):
+            final_pd_index = new_pd_index.tz_localize("UTC").tz_convert(self.tz)
+        else:
+            final_pd_index = new_pd_index
+        return final_pd_index, final_pd_index
+
+    def take_data(self):
+        """return the values"""
+        return self.values
+
+    @property
+    def attrs(self):
+        return self.table._v_attrs
+
+    @property
+    def description(self):
+        return self.table.description
+
+    @property
+    def col(self):
+        """return my current col description"""
+        return getattr(self.description, self.cname, None)
+
+    @property
+    def cvalues(self):
+        """return my cython values"""
+        return self.values
+
+    def __iter__(self) -> Iterator:
+        return iter(self.values)
+
+    def maybe_set_size(self, min_itemsize=None) -> None:
+        """
+        maybe set a string col itemsize:
+            min_itemsize can be an integer or a dict with this columns name
+            with an integer size
+        """
+        if self.kind == "string":
+            if isinstance(min_itemsize, dict):
+                min_itemsize = min_itemsize.get(self.name)
+
+            if min_itemsize is not None and self.typ.itemsize < min_itemsize:
+                self.typ = _tables().StringCol(itemsize=min_itemsize, pos=self.pos)
+
+    def validate_names(self) -> None:
+        pass
+
+    def validate_and_set(self, handler: AppendableTable, append: bool) -> None:
+        self.table = handler.table
+        self.validate_col()
+        self.validate_attr(append)
+        self.validate_metadata(handler)
+        self.write_metadata(handler)
+        self.set_attr()
+
+    def validate_col(self, itemsize=None):
+        """validate this column: return the compared against itemsize"""
+        # validate this column for string truncation (or reset to the max size)
+        if self.kind == "string":
+            c = self.col
+            if c is not None:
+                if itemsize is None:
+                    itemsize = self.itemsize
+                if c.itemsize < itemsize:
+                    raise ValueError(
+                        f"Trying to store a string with len [{itemsize}] in "
+                        f"[{self.cname}] column but\nthis column has a limit of "
+                        f"[{c.itemsize}]!\nConsider using min_itemsize to "
+                        "preset the sizes on these columns"
+                    )
+                return c.itemsize
+
+        return None
+
+    def validate_attr(self, append: bool) -> None:
+        # check for backwards incompatibility
+        if append:
+            existing_kind = getattr(self.attrs, self.kind_attr, None)
+            if existing_kind is not None and existing_kind != self.kind:
+                raise TypeError(
+                    f"incompatible kind in col [{existing_kind} - {self.kind}]"
+                )
+
+    def update_info(self, info) -> None:
+        """
+        set/update the info for this indexable with the key/value
+        if there is a conflict raise/warn as needed
+        """
+        for key in self._info_fields:
+            value = getattr(self, key, None)
+            idx = info.setdefault(self.name, {})
+
+            existing_value = idx.get(key)
+            if key in idx and value is not None and existing_value != value:
+                # frequency/name just warn
+                if key in ["freq", "index_name"]:
+                    ws = attribute_conflict_doc % (key, existing_value, value)
+                    warnings.warn(
+                        ws, AttributeConflictWarning, stacklevel=find_stack_level()
+                    )
+
+                    # reset
+                    idx[key] = None
+                    setattr(self, key, None)
+
+                else:
+                    raise ValueError(
+                        f"invalid info for [{self.name}] for [{key}], "
+                        f"existing_value [{existing_value}] conflicts with "
+                        f"new value [{value}]"
+                    )
+            elif value is not None or existing_value is not None:
+                idx[key] = value
+
+    def set_info(self, info) -> None:
+        """set my state from the passed info"""
+        idx = info.get(self.name)
+        if idx is not None:
+            self.__dict__.update(idx)
+
+    def set_attr(self) -> None:
+        """set the kind for this column"""
+        setattr(self.attrs, self.kind_attr, self.kind)
+
+    def validate_metadata(self, handler: AppendableTable) -> None:
+        """validate that kind=category does not change the categories"""
+        if self.meta == "category":
+            new_metadata = self.metadata
+            cur_metadata = handler.read_metadata(self.cname)
+            if (
+                new_metadata is not None
+                and cur_metadata is not None
+                and not array_equivalent(
+                    new_metadata, cur_metadata, strict_nan=True, dtype_equal=True
+                )
+            ):
+                raise ValueError(
+                    "cannot append a categorical with "
+                    "different categories to the existing"
+                )
+
+    def write_metadata(self, handler: AppendableTable) -> None:
+        """set the meta data"""
+        if self.metadata is not None:
+            handler.write_metadata(self.cname, self.metadata)
+
+
+class GenericIndexCol(IndexCol):
+    """an index which is not represented in the data of the table"""
+
+    @property
+    def is_indexed(self) -> bool:
+        return False
+
+    def convert(
+        self, values: np.ndarray, nan_rep, encoding: str, errors: str
+    ) -> tuple[Index, Index]:
+        """
+        Convert the data from this selection to the appropriate pandas type.
+
+        Parameters
+        ----------
+        values : np.ndarray
+        nan_rep : str
+        encoding : str
+        errors : str
+        """
+        assert isinstance(values, np.ndarray), type(values)
+
+        index = RangeIndex(len(values))
+        return index, index
+
+    def set_attr(self) -> None:
+        pass
+
+
+class DataCol(IndexCol):
+    """
+    a data holding column, by definition this is not indexable
+
+    Parameters
+    ----------
+    data   : the actual data
+    cname  : the column name in the table to hold the data (typically
+                values)
+    meta   : a string description of the metadata
+    metadata : the actual metadata
+    """
+
+    is_an_indexable = False
+    is_data_indexable = False
+    _info_fields = ["tz", "ordered"]
+
+    def __init__(
+        self,
+        name: str,
+        values=None,
+        kind=None,
+        typ=None,
+        cname: str | None = None,
+        pos=None,
+        tz=None,
+        ordered=None,
+        table=None,
+        meta=None,
+        metadata=None,
+        dtype: DtypeArg | None = None,
+        data=None,
+    ) -> None:
+        super().__init__(
+            name=name,
+            values=values,
+            kind=kind,
+            typ=typ,
+            pos=pos,
+            cname=cname,
+            tz=tz,
+            ordered=ordered,
+            table=table,
+            meta=meta,
+            metadata=metadata,
+        )
+        self.dtype = dtype
+        self.data = data
+
+    @property
+    def dtype_attr(self) -> str:
+        return f"{self.name}_dtype"
+
+    @property
+    def meta_attr(self) -> str:
+        return f"{self.name}_meta"
+
+    def __repr__(self) -> str:
+        temp = tuple(
+            map(
+                pprint_thing, (self.name, self.cname, self.dtype, self.kind, self.shape)
+            )
+        )
+        return ",".join(
+            [
+                f"{key}->{value}"
+                for key, value in zip(["name", "cname", "dtype", "kind", "shape"], temp)
+            ]
+        )
+
+    def __eq__(self, other: object) -> bool:
+        """compare 2 col items"""
+        return all(
+            getattr(self, a, None) == getattr(other, a, None)
+            for a in ["name", "cname", "dtype", "pos"]
+        )
+
+    def set_data(self, data: ArrayLike) -> None:
+        assert data is not None
+        assert self.dtype is None
+
+        data, dtype_name = _get_data_and_dtype_name(data)
+
+        self.data = data
+        self.dtype = dtype_name
+        self.kind = _dtype_to_kind(dtype_name)
+
+    def take_data(self):
+        """return the data"""
+        return self.data
+
+    @classmethod
+    def _get_atom(cls, values: ArrayLike) -> Col:
+        """
+        Get an appropriately typed and shaped pytables.Col object for values.
+        """
+        dtype = values.dtype
+        # error: Item "ExtensionDtype" of "Union[ExtensionDtype, dtype[Any]]" has no
+        # attribute "itemsize"
+        itemsize = dtype.itemsize  # type: ignore[union-attr]
+
+        shape = values.shape
+        if values.ndim == 1:
+            # EA, use block shape pretending it is 2D
+            # TODO(EA2D): not necessary with 2D EAs
+            shape = (1, values.size)
+
+        if isinstance(values, Categorical):
+            codes = values.codes
+            atom = cls.get_atom_data(shape, kind=codes.dtype.name)
+        elif lib.is_np_dtype(dtype, "M") or isinstance(dtype, DatetimeTZDtype):
+            atom = cls.get_atom_datetime64(shape)
+        elif lib.is_np_dtype(dtype, "m"):
+            atom = cls.get_atom_timedelta64(shape)
+        elif is_complex_dtype(dtype):
+            atom = _tables().ComplexCol(itemsize=itemsize, shape=shape[0])
+        elif is_string_dtype(dtype):
+            atom = cls.get_atom_string(shape, itemsize)
+        else:
+            atom = cls.get_atom_data(shape, kind=dtype.name)
+
+        return atom
+
+    @classmethod
+    def get_atom_string(cls, shape, itemsize):
+        return _tables().StringCol(itemsize=itemsize, shape=shape[0])
+
+    @classmethod
+    def get_atom_coltype(cls, kind: str) -> type[Col]:
+        """return the PyTables column class for this column"""
+        if kind.startswith("uint"):
+            k4 = kind[4:]
+            col_name = f"UInt{k4}Col"
+        elif kind.startswith("period"):
+            # we store as integer
+            col_name = "Int64Col"
+        else:
+            kcap = kind.capitalize()
+            col_name = f"{kcap}Col"
+
+        return getattr(_tables(), col_name)
+
+    @classmethod
+    def get_atom_data(cls, shape, kind: str) -> Col:
+        return cls.get_atom_coltype(kind=kind)(shape=shape[0])
+
+    @classmethod
+    def get_atom_datetime64(cls, shape):
+        return _tables().Int64Col(shape=shape[0])
+
+    @classmethod
+    def get_atom_timedelta64(cls, shape):
+        return _tables().Int64Col(shape=shape[0])
+
+    @property
+    def shape(self):
+        return getattr(self.data, "shape", None)
+
+    @property
+    def cvalues(self):
+        """return my cython values"""
+        return self.data
+
+    def validate_attr(self, append) -> None:
+        """validate that we have the same order as the existing & same dtype"""
+        if append:
+            existing_fields = getattr(self.attrs, self.kind_attr, None)
+            if existing_fields is not None and existing_fields != list(self.values):
+                raise ValueError("appended items do not match existing items in table!")
+
+            existing_dtype = getattr(self.attrs, self.dtype_attr, None)
+            if existing_dtype is not None and existing_dtype != self.dtype:
+                raise ValueError(
+                    "appended items dtype do not match existing items dtype in table!"
+                )
+
+    def convert(self, values: np.ndarray, nan_rep, encoding: str, errors: str):
+        """
+        Convert the data from this selection to the appropriate pandas type.
+
+        Parameters
+        ----------
+        values : np.ndarray
+        nan_rep :
+        encoding : str
+        errors : str
+
+        Returns
+        -------
+        index : listlike to become an Index
+        data : ndarraylike to become a column
+        """
+        assert isinstance(values, np.ndarray), type(values)
+
+        # values is a recarray
+        if values.dtype.fields is not None:
+            values = values[self.cname]
+
+        assert self.typ is not None
+        if self.dtype is None:
+            # Note: in tests we never have timedelta64 or datetime64,
+            #  so the _get_data_and_dtype_name may be unnecessary
+            converted, dtype_name = _get_data_and_dtype_name(values)
+            kind = _dtype_to_kind(dtype_name)
+        else:
+            converted = values
+            dtype_name = self.dtype
+            kind = self.kind
+
+        assert isinstance(converted, np.ndarray)  # for mypy
+
+        # use the meta if needed
+        meta = self.meta
+        metadata = self.metadata
+        ordered = self.ordered
+        tz = self.tz
+
+        assert dtype_name is not None
+        # convert to the correct dtype
+        dtype = dtype_name
+
+        # reverse converts
+        if dtype.startswith("datetime64"):
+            # recreate with tz if indicated
+            converted = _set_tz(converted, tz, dtype)
+
+        elif dtype == "timedelta64":
+            converted = np.asarray(converted, dtype="m8[ns]")
+        elif dtype == "date":
+            try:
+                converted = np.asarray(
+                    [date.fromordinal(v) for v in converted], dtype=object
+                )
+            except ValueError:
+                converted = np.asarray(
+                    [date.fromtimestamp(v) for v in converted], dtype=object
+                )
+
+        elif meta == "category":
+            # we have a categorical
+            categories = metadata
+            codes = converted.ravel()
+
+            # if we have stored a NaN in the categories
+            # then strip it; in theory we could have BOTH
+            # -1s in the codes and nulls :<
+            if categories is None:
+                # Handle case of NaN-only categorical columns in which case
+                # the categories are an empty array; when this is stored,
+                # pytables cannot write a zero-len array, so on readback
+                # the categories would be None and `read_hdf()` would fail.
+                categories = Index([], dtype=np.float64)
+            else:
+                mask = isna(categories)
+                if mask.any():
+                    categories = categories[~mask]
+                    codes[codes != -1] -= mask.astype(int).cumsum()._values
+
+            converted = Categorical.from_codes(
+                codes, categories=categories, ordered=ordered, validate=False
+            )
+
+        else:
+            try:
+                converted = converted.astype(dtype, copy=False)
+            except TypeError:
+                converted = converted.astype("O", copy=False)
+
+        # convert nans / decode
+        if kind == "string":
+            converted = _unconvert_string_array(
+                converted, nan_rep=nan_rep, encoding=encoding, errors=errors
+            )
+
+        return self.values, converted
+
+    def set_attr(self) -> None:
+        """set the data for this column"""
+        setattr(self.attrs, self.kind_attr, self.values)
+        setattr(self.attrs, self.meta_attr, self.meta)
+        assert self.dtype is not None
+        setattr(self.attrs, self.dtype_attr, self.dtype)
+
+
+class DataIndexableCol(DataCol):
+    """represent a data column that can be indexed"""
+
+    is_data_indexable = True
+
+    def validate_names(self) -> None:
+        if not is_string_dtype(Index(self.values).dtype):
+            # TODO: should the message here be more specifically non-str?
+            raise ValueError("cannot have non-object label DataIndexableCol")
+
+    @classmethod
+    def get_atom_string(cls, shape, itemsize):
+        return _tables().StringCol(itemsize=itemsize)
+
+    @classmethod
+    def get_atom_data(cls, shape, kind: str) -> Col:
+        return cls.get_atom_coltype(kind=kind)()
+
+    @classmethod
+    def get_atom_datetime64(cls, shape):
+        return _tables().Int64Col()
+
+    @classmethod
+    def get_atom_timedelta64(cls, shape):
+        return _tables().Int64Col()
+
+
+class GenericDataIndexableCol(DataIndexableCol):
+    """represent a generic pytables data column"""
+
+
+class Fixed:
+    """
+    represent an object in my store
+    facilitate read/write of various types of objects
+    this is an abstract base class
+
+    Parameters
+    ----------
+    parent : HDFStore
+    group : Node
+        The group node where the table resides.
+    """
+
+    pandas_kind: str
+    format_type: str = "fixed"  # GH#30962 needed by dask
+    obj_type: type[DataFrame | Series]
+    ndim: int
+    parent: HDFStore
+    is_table: bool = False
+
+    def __init__(
+        self,
+        parent: HDFStore,
+        group: Node,
+        encoding: str | None = "UTF-8",
+        errors: str = "strict",
+    ) -> None:
+        assert isinstance(parent, HDFStore), type(parent)
+        assert _table_mod is not None  # needed for mypy
+        assert isinstance(group, _table_mod.Node), type(group)
+        self.parent = parent
+        self.group = group
+        self.encoding = _ensure_encoding(encoding)
+        self.errors = errors
+
+    @property
+    def is_old_version(self) -> bool:
+        return self.version[0] <= 0 and self.version[1] <= 10 and self.version[2] < 1
+
+    @property
+    def version(self) -> tuple[int, int, int]:
+        """compute and set our version"""
+        version = getattr(self.group._v_attrs, "pandas_version", None)
+        if isinstance(version, str):
+            version_tup = tuple(int(x) for x in version.split("."))
+            if len(version_tup) == 2:
+                version_tup = version_tup + (0,)
+            assert len(version_tup) == 3  # needed for mypy
+            return version_tup
+        else:
+            return (0, 0, 0)
+
+    @property
+    def pandas_type(self):
+        return getattr(self.group._v_attrs, "pandas_type", None)
+
+    def __repr__(self) -> str:
+        """return a pretty representation of myself"""
+        self.infer_axes()
+        s = self.shape
+        if s is not None:
+            if isinstance(s, (list, tuple)):
+                jshape = ",".join([pprint_thing(x) for x in s])
+                s = f"[{jshape}]"
+            return f"{self.pandas_type:12.12} (shape->{s})"
+        return self.pandas_type
+
+    def set_object_info(self) -> None:
+        """set my pandas type & version"""
+        self.attrs.pandas_type = str(self.pandas_kind)
+        self.attrs.pandas_version = str(_version)
+
+    def copy(self) -> Fixed:
+        new_self = copy.copy(self)
+        return new_self
+
+    @property
+    def shape(self):
+        return self.nrows
+
+    @property
+    def pathname(self):
+        return self.group._v_pathname
+
+    @property
+    def _handle(self):
+        return self.parent._handle
+
+    @property
+    def _filters(self):
+        return self.parent._filters
+
+    @property
+    def _complevel(self) -> int:
+        return self.parent._complevel
+
+    @property
+    def _fletcher32(self) -> bool:
+        return self.parent._fletcher32
+
+    @property
+    def attrs(self):
+        return self.group._v_attrs
+
+    def set_attrs(self) -> None:
+        """set our object attributes"""
+
+    def get_attrs(self) -> None:
+        """get our object attributes"""
+
+    @property
+    def storable(self):
+        """return my storable"""
+        return self.group
+
+    @property
+    def is_exists(self) -> bool:
+        return False
+
+    @property
+    def nrows(self):
+        return getattr(self.storable, "nrows", None)
+
+    def validate(self, other) -> Literal[True] | None:
+        """validate against an existing storable"""
+        if other is None:
+            return None
+        return True
+
+    def validate_version(self, where=None) -> None:
+        """are we trying to operate on an old version?"""
+
+    def infer_axes(self) -> bool:
+        """
+        infer the axes of my storer
+        return a boolean indicating if we have a valid storer or not
+        """
+        s = self.storable
+        if s is None:
+            return False
+        self.get_attrs()
+        return True
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> Series | DataFrame:
+        raise NotImplementedError(
+            "cannot read on an abstract storer: subclasses should implement"
+        )
+
+    def write(self, obj, **kwargs) -> None:
+        raise NotImplementedError(
+            "cannot write on an abstract storer: subclasses should implement"
+        )
+
+    def delete(
+        self, where=None, start: int | None = None, stop: int | None = None
+    ) -> int | None:
+        """
+        support fully deleting the node in its entirety (only) - where
+        specification must be None
+        """
+        if com.all_none(where, start, stop):
+            self._handle.remove_node(self.group, recursive=True)
+            return None
+
+        raise TypeError("cannot delete on an abstract storer")
+
+
+class GenericFixed(Fixed):
+    """a generified fixed version"""
+
+    _index_type_map = {DatetimeIndex: "datetime", PeriodIndex: "period"}
+    _reverse_index_map = {v: k for k, v in _index_type_map.items()}
+    attributes: list[str] = []
+
+    # indexer helpers
+    def _class_to_alias(self, cls) -> str:
+        return self._index_type_map.get(cls, "")
+
+    def _alias_to_class(self, alias):
+        if isinstance(alias, type):  # pragma: no cover
+            # compat: for a short period of time master stored types
+            return alias
+        return self._reverse_index_map.get(alias, Index)
+
+    def _get_index_factory(self, attrs):
+        index_class = self._alias_to_class(getattr(attrs, "index_class", ""))
+
+        factory: Callable
+
+        if index_class == DatetimeIndex:
+
+            def f(values, freq=None, tz=None):
+                # data are already in UTC, localize and convert if tz present
+                dta = DatetimeArray._simple_new(
+                    values.values, dtype=values.dtype, freq=freq
+                )
+                result = DatetimeIndex._simple_new(dta, name=None)
+                if tz is not None:
+                    result = result.tz_localize("UTC").tz_convert(tz)
+                return result
+
+            factory = f
+        elif index_class == PeriodIndex:
+
+            def f(values, freq=None, tz=None):
+                dtype = PeriodDtype(freq)
+                parr = PeriodArray._simple_new(values, dtype=dtype)
+                return PeriodIndex._simple_new(parr, name=None)
+
+            factory = f
+        else:
+            factory = index_class
+
+        kwargs = {}
+        if "freq" in attrs:
+            kwargs["freq"] = attrs["freq"]
+            if index_class is Index:
+                # DTI/PI would be gotten by _alias_to_class
+                factory = TimedeltaIndex
+
+        if "tz" in attrs:
+            kwargs["tz"] = attrs["tz"]
+            assert index_class is DatetimeIndex  # just checking
+
+        return factory, kwargs
+
+    def validate_read(self, columns, where) -> None:
+        """
+        raise if any keywords are passed which are not-None
+        """
+        if columns is not None:
+            raise TypeError(
+                "cannot pass a column specification when reading "
+                "a Fixed format store. this store must be selected in its entirety"
+            )
+        if where is not None:
+            raise TypeError(
+                "cannot pass a where specification when reading "
+                "from a Fixed format store. this store must be selected in its entirety"
+            )
+
+    @property
+    def is_exists(self) -> bool:
+        return True
+
+    def set_attrs(self) -> None:
+        """set our object attributes"""
+        self.attrs.encoding = self.encoding
+        self.attrs.errors = self.errors
+
+    def get_attrs(self) -> None:
+        """retrieve our attributes"""
+        self.encoding = _ensure_encoding(getattr(self.attrs, "encoding", None))
+        self.errors = getattr(self.attrs, "errors", "strict")
+        for n in self.attributes:
+            setattr(self, n, getattr(self.attrs, n, None))
+
+    def write(self, obj, **kwargs) -> None:
+        self.set_attrs()
+
+    def read_array(self, key: str, start: int | None = None, stop: int | None = None):
+        """read an array for the specified node (off of group"""
+        import tables
+
+        node = getattr(self.group, key)
+        attrs = node._v_attrs
+
+        transposed = getattr(attrs, "transposed", False)
+
+        if isinstance(node, tables.VLArray):
+            ret = node[0][start:stop]
+            dtype = getattr(attrs, "value_type", None)
+            if dtype is not None:
+                ret = pd_array(ret, dtype=dtype)
+        else:
+            dtype = getattr(attrs, "value_type", None)
+            shape = getattr(attrs, "shape", None)
+
+            if shape is not None:
+                # length 0 axis
+                ret = np.empty(shape, dtype=dtype)
+            else:
+                ret = node[start:stop]
+
+            if dtype and dtype.startswith("datetime64"):
+                # reconstruct a timezone if indicated
+                tz = getattr(attrs, "tz", None)
+                ret = _set_tz(ret, tz, dtype)
+
+            elif dtype == "timedelta64":
+                ret = np.asarray(ret, dtype="m8[ns]")
+
+        if transposed:
+            return ret.T
+        else:
+            return ret
+
+    def read_index(
+        self, key: str, start: int | None = None, stop: int | None = None
+    ) -> Index:
+        variety = getattr(self.attrs, f"{key}_variety")
+
+        if variety == "multi":
+            return self.read_multi_index(key, start=start, stop=stop)
+        elif variety == "regular":
+            node = getattr(self.group, key)
+            index = self.read_index_node(node, start=start, stop=stop)
+            return index
+        else:  # pragma: no cover
+            raise TypeError(f"unrecognized index variety: {variety}")
+
+    def write_index(self, key: str, index: Index) -> None:
+        if isinstance(index, MultiIndex):
+            setattr(self.attrs, f"{key}_variety", "multi")
+            self.write_multi_index(key, index)
+        else:
+            setattr(self.attrs, f"{key}_variety", "regular")
+            converted = _convert_index("index", index, self.encoding, self.errors)
+
+            self.write_array(key, converted.values)
+
+            node = getattr(self.group, key)
+            node._v_attrs.kind = converted.kind
+            node._v_attrs.name = index.name
+
+            if isinstance(index, (DatetimeIndex, PeriodIndex)):
+                node._v_attrs.index_class = self._class_to_alias(type(index))
+
+            if isinstance(index, (DatetimeIndex, PeriodIndex, TimedeltaIndex)):
+                node._v_attrs.freq = index.freq
+
+            if isinstance(index, DatetimeIndex) and index.tz is not None:
+                node._v_attrs.tz = _get_tz(index.tz)
+
+    def write_multi_index(self, key: str, index: MultiIndex) -> None:
+        setattr(self.attrs, f"{key}_nlevels", index.nlevels)
+
+        for i, (lev, level_codes, name) in enumerate(
+            zip(index.levels, index.codes, index.names)
+        ):
+            # write the level
+            if isinstance(lev.dtype, ExtensionDtype):
+                raise NotImplementedError(
+                    "Saving a MultiIndex with an extension dtype is not supported."
+                )
+            level_key = f"{key}_level{i}"
+            conv_level = _convert_index(level_key, lev, self.encoding, self.errors)
+            self.write_array(level_key, conv_level.values)
+            node = getattr(self.group, level_key)
+            node._v_attrs.kind = conv_level.kind
+            node._v_attrs.name = name
+
+            # write the name
+            setattr(node._v_attrs, f"{key}_name{name}", name)
+
+            # write the labels
+            label_key = f"{key}_label{i}"
+            self.write_array(label_key, level_codes)
+
+    def read_multi_index(
+        self, key: str, start: int | None = None, stop: int | None = None
+    ) -> MultiIndex:
+        nlevels = getattr(self.attrs, f"{key}_nlevels")
+
+        levels = []
+        codes = []
+        names: list[Hashable] = []
+        for i in range(nlevels):
+            level_key = f"{key}_level{i}"
+            node = getattr(self.group, level_key)
+            lev = self.read_index_node(node, start=start, stop=stop)
+            levels.append(lev)
+            names.append(lev.name)
+
+            label_key = f"{key}_label{i}"
+            level_codes = self.read_array(label_key, start=start, stop=stop)
+            codes.append(level_codes)
+
+        return MultiIndex(
+            levels=levels, codes=codes, names=names, verify_integrity=True
+        )
+
+    def read_index_node(
+        self, node: Node, start: int | None = None, stop: int | None = None
+    ) -> Index:
+        data = node[start:stop]
+        # If the index was an empty array write_array_empty() will
+        # have written a sentinel. Here we replace it with the original.
+        if "shape" in node._v_attrs and np.prod(node._v_attrs.shape) == 0:
+            data = np.empty(node._v_attrs.shape, dtype=node._v_attrs.value_type)
+        kind = node._v_attrs.kind
+        name = None
+
+        if "name" in node._v_attrs:
+            name = _ensure_str(node._v_attrs.name)
+
+        attrs = node._v_attrs
+        factory, kwargs = self._get_index_factory(attrs)
+
+        if kind in ("date", "object"):
+            index = factory(
+                _unconvert_index(
+                    data, kind, encoding=self.encoding, errors=self.errors
+                ),
+                dtype=object,
+                **kwargs,
+            )
+        else:
+            index = factory(
+                _unconvert_index(
+                    data, kind, encoding=self.encoding, errors=self.errors
+                ),
+                **kwargs,
+            )
+
+        index.name = name
+
+        return index
+
+    def write_array_empty(self, key: str, value: ArrayLike) -> None:
+        """write a 0-len array"""
+        # ugly hack for length 0 axes
+        arr = np.empty((1,) * value.ndim)
+        self._handle.create_array(self.group, key, arr)
+        node = getattr(self.group, key)
+        node._v_attrs.value_type = str(value.dtype)
+        node._v_attrs.shape = value.shape
+
+    def write_array(
+        self, key: str, obj: AnyArrayLike, items: Index | None = None
+    ) -> None:
+        # TODO: we only have a few tests that get here, the only EA
+        #  that gets passed is DatetimeArray, and we never have
+        #  both self._filters and EA
+
+        value = extract_array(obj, extract_numpy=True)
+
+        if key in self.group:
+            self._handle.remove_node(self.group, key)
+
+        # Transform needed to interface with pytables row/col notation
+        empty_array = value.size == 0
+        transposed = False
+
+        if isinstance(value.dtype, CategoricalDtype):
+            raise NotImplementedError(
+                "Cannot store a category dtype in a HDF5 dataset that uses format="
+                '"fixed". Use format="table".'
+            )
+        if not empty_array:
+            if hasattr(value, "T"):
+                # ExtensionArrays (1d) may not have transpose.
+                value = value.T
+                transposed = True
+
+        atom = None
+        if self._filters is not None:
+            with suppress(ValueError):
+                # get the atom for this datatype
+                atom = _tables().Atom.from_dtype(value.dtype)
+
+        if atom is not None:
+            # We only get here if self._filters is non-None and
+            #  the Atom.from_dtype call succeeded
+
+            # create an empty chunked array and fill it from value
+            if not empty_array:
+                ca = self._handle.create_carray(
+                    self.group, key, atom, value.shape, filters=self._filters
+                )
+                ca[:] = value
+
+            else:
+                self.write_array_empty(key, value)
+
+        elif value.dtype.type == np.object_:
+            # infer the type, warn if we have a non-string type here (for
+            # performance)
+            inferred_type = lib.infer_dtype(value, skipna=False)
+            if empty_array:
+                pass
+            elif inferred_type == "string":
+                pass
+            elif get_option("performance_warnings"):
+                ws = performance_doc % (inferred_type, key, items)
+                warnings.warn(ws, PerformanceWarning, stacklevel=find_stack_level())
+
+            vlarr = self._handle.create_vlarray(self.group, key, _tables().ObjectAtom())
+            vlarr.append(value)
+
+        elif lib.is_np_dtype(value.dtype, "M"):
+            self._handle.create_array(self.group, key, value.view("i8"))
+            getattr(self.group, key)._v_attrs.value_type = str(value.dtype)
+        elif isinstance(value.dtype, DatetimeTZDtype):
+            # store as UTC
+            # with a zone
+
+            # error: Item "ExtensionArray" of "Union[Any, ExtensionArray]" has no
+            # attribute "asi8"
+            self._handle.create_array(
+                self.group,
+                key,
+                value.asi8,  # type: ignore[union-attr]
+            )
+
+            node = getattr(self.group, key)
+            # error: Item "ExtensionArray" of "Union[Any, ExtensionArray]" has no
+            # attribute "tz"
+            node._v_attrs.tz = _get_tz(value.tz)  # type: ignore[union-attr]
+            node._v_attrs.value_type = f"datetime64[{value.dtype.unit}]"
+        elif lib.is_np_dtype(value.dtype, "m"):
+            self._handle.create_array(self.group, key, value.view("i8"))
+            getattr(self.group, key)._v_attrs.value_type = "timedelta64"
+        elif isinstance(value, BaseStringArray):
+            vlarr = self._handle.create_vlarray(self.group, key, _tables().ObjectAtom())
+            vlarr.append(value.to_numpy())
+            node = getattr(self.group, key)
+            node._v_attrs.value_type = str(value.dtype)
+        elif empty_array:
+            self.write_array_empty(key, value)
+        else:
+            self._handle.create_array(self.group, key, value)
+
+        getattr(self.group, key)._v_attrs.transposed = transposed
+
+
+class SeriesFixed(GenericFixed):
+    pandas_kind = "series"
+    attributes = ["name"]
+
+    name: Hashable
+
+    @property
+    def shape(self) -> tuple[int] | None:
+        try:
+            return (len(self.group.values),)
+        except (TypeError, AttributeError):
+            return None
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> Series:
+        self.validate_read(columns, where)
+        index = self.read_index("index", start=start, stop=stop)
+        values = self.read_array("values", start=start, stop=stop)
+        result = Series(values, index=index, name=self.name, copy=False)
+        if (
+            using_string_dtype()
+            and isinstance(values, np.ndarray)
+            and is_string_array(values, skipna=True)
+        ):
+            result = result.astype(StringDtype(na_value=np.nan))
+        return result
+
+    def write(self, obj, **kwargs) -> None:
+        super().write(obj, **kwargs)
+        self.write_index("index", obj.index)
+        self.write_array("values", obj)
+        self.attrs.name = obj.name
+
+
+class BlockManagerFixed(GenericFixed):
+    attributes = ["ndim", "nblocks"]
+
+    nblocks: int
+
+    @property
+    def shape(self) -> Shape | None:
+        try:
+            ndim = self.ndim
+
+            # items
+            items = 0
+            for i in range(self.nblocks):
+                node = getattr(self.group, f"block{i}_items")
+                shape = getattr(node, "shape", None)
+                if shape is not None:
+                    items += shape[0]
+
+            # data shape
+            node = self.group.block0_values
+            shape = getattr(node, "shape", None)
+            if shape is not None:
+                shape = list(shape[0 : (ndim - 1)])
+            else:
+                shape = []
+
+            shape.append(items)
+
+            return shape
+        except AttributeError:
+            return None
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> DataFrame:
+        # start, stop applied to rows, so 0th axis only
+        self.validate_read(columns, where)
+        select_axis = self.obj_type()._get_block_manager_axis(0)
+
+        axes = []
+        for i in range(self.ndim):
+            _start, _stop = (start, stop) if i == select_axis else (None, None)
+            ax = self.read_index(f"axis{i}", start=_start, stop=_stop)
+            axes.append(ax)
+
+        items = axes[0]
+        dfs = []
+
+        for i in range(self.nblocks):
+            blk_items = self.read_index(f"block{i}_items")
+            values = self.read_array(f"block{i}_values", start=_start, stop=_stop)
+
+            columns = items[items.get_indexer(blk_items)]
+            df = DataFrame(values.T, columns=columns, index=axes[1], copy=False)
+            if (
+                using_string_dtype()
+                and isinstance(values, np.ndarray)
+                and is_string_array(values, skipna=True)
+            ):
+                df = df.astype(StringDtype(na_value=np.nan))
+            dfs.append(df)
+
+        if len(dfs) > 0:
+            out = concat(dfs, axis=1).copy()
+            return out.reindex(columns=items)
+
+        return DataFrame(columns=axes[0], index=axes[1])
+
+    def write(self, obj, **kwargs) -> None:
+        super().write(obj, **kwargs)
+
+        data = obj._mgr
+        if not data.is_consolidated():
+            data = data.consolidate()
+
+        self.attrs.ndim = data.ndim
+        for i, ax in enumerate(data.axes):
+            if i == 0 and (not ax.is_unique):
+                raise ValueError("Columns index has to be unique for fixed format")
+            self.write_index(f"axis{i}", ax)
+
+        # Supporting mixed-type DataFrame objects...nontrivial
+        self.attrs.nblocks = len(data.blocks)
+        for i, blk in enumerate(data.blocks):
+            # I have no idea why, but writing values before items fixed #2299
+            blk_items = data.items.take(blk.mgr_locs)
+            self.write_array(f"block{i}_values", blk.values, items=blk_items)
+            self.write_index(f"block{i}_items", blk_items)
+
+
+class FrameFixed(BlockManagerFixed):
+    pandas_kind = "frame"
+    obj_type = DataFrame
+
+
+class Table(Fixed):
+    """
+    represent a table:
+        facilitate read/write of various types of tables
+
+    Attrs in Table Node
+    -------------------
+    These are attributes that are store in the main table node, they are
+    necessary to recreate these tables when read back in.
+
+    index_axes    : a list of tuples of the (original indexing axis and
+        index column)
+    non_index_axes: a list of tuples of the (original index axis and
+        columns on a non-indexing axis)
+    values_axes   : a list of the columns which comprise the data of this
+        table
+    data_columns  : a list of the columns that we are allowing indexing
+        (these become single columns in values_axes)
+    nan_rep       : the string to use for nan representations for string
+        objects
+    levels        : the names of levels
+    metadata      : the names of the metadata columns
+    """
+
+    pandas_kind = "wide_table"
+    format_type: str = "table"  # GH#30962 needed by dask
+    table_type: str
+    levels: int | list[Hashable] = 1
+    is_table = True
+
+    metadata: list
+
+    def __init__(
+        self,
+        parent: HDFStore,
+        group: Node,
+        encoding: str | None = None,
+        errors: str = "strict",
+        index_axes: list[IndexCol] | None = None,
+        non_index_axes: list[tuple[AxisInt, Any]] | None = None,
+        values_axes: list[DataCol] | None = None,
+        data_columns: list | None = None,
+        info: dict | None = None,
+        nan_rep=None,
+    ) -> None:
+        super().__init__(parent, group, encoding=encoding, errors=errors)
+        self.index_axes = index_axes or []
+        self.non_index_axes = non_index_axes or []
+        self.values_axes = values_axes or []
+        self.data_columns = data_columns or []
+        self.info = info or {}
+        self.nan_rep = nan_rep
+
+    @property
+    def table_type_short(self) -> str:
+        return self.table_type.split("_")[0]
+
+    def __repr__(self) -> str:
+        """return a pretty representation of myself"""
+        self.infer_axes()
+        jdc = ",".join(self.data_columns) if len(self.data_columns) else ""
+        dc = f",dc->[{jdc}]"
+
+        ver = ""
+        if self.is_old_version:
+            jver = ".".join([str(x) for x in self.version])
+            ver = f"[{jver}]"
+
+        jindex_axes = ",".join([a.name for a in self.index_axes])
+        return (
+            f"{self.pandas_type:12.12}{ver} "
+            f"(typ->{self.table_type_short},nrows->{self.nrows},"
+            f"ncols->{self.ncols},indexers->[{jindex_axes}]{dc})"
+        )
+
+    def __getitem__(self, c: str):
+        """return the axis for c"""
+        for a in self.axes:
+            if c == a.name:
+                return a
+        return None
+
+    def validate(self, other) -> None:
+        """validate against an existing table"""
+        if other is None:
+            return
+
+        if other.table_type != self.table_type:
+            raise TypeError(
+                "incompatible table_type with existing "
+                f"[{other.table_type} - {self.table_type}]"
+            )
+
+        for c in ["index_axes", "non_index_axes", "values_axes"]:
+            sv = getattr(self, c, None)
+            ov = getattr(other, c, None)
+            if sv != ov:
+                # show the error for the specific axes
+                # Argument 1 to "enumerate" has incompatible type
+                # "Optional[Any]"; expected "Iterable[Any]"  [arg-type]
+                for i, sax in enumerate(sv):  # type: ignore[arg-type]
+                    # Value of type "Optional[Any]" is not indexable  [index]
+                    oax = ov[i]  # type: ignore[index]
+                    if sax != oax:
+                        if c == "values_axes" and sax.kind != oax.kind:
+                            raise ValueError(
+                                f"Cannot serialize the column [{oax.values[0]}] "
+                                f"because its data contents are not [{sax.kind}] "
+                                f"but [{oax.kind}] object dtype"
+                            )
+                        raise ValueError(
+                            f"invalid combination of [{c}] on appending data "
+                            f"[{sax}] vs current table [{oax}]"
+                        )
+
+                # should never get here
+                raise Exception(
+                    f"invalid combination of [{c}] on appending data [{sv}] vs "
+                    f"current table [{ov}]"
+                )
+
+    @property
+    def is_multi_index(self) -> bool:
+        """the levels attribute is 1 or a list in the case of a multi-index"""
+        return isinstance(self.levels, list)
+
+    def validate_multiindex(
+        self, obj: DataFrame | Series
+    ) -> tuple[DataFrame, list[Hashable]]:
+        """
+        validate that we can store the multi-index; reset and return the
+        new object
+        """
+        levels = com.fill_missing_names(obj.index.names)
+        try:
+            reset_obj = obj.reset_index()
+        except ValueError as err:
+            raise ValueError(
+                "duplicate names/columns in the multi-index when storing as a table"
+            ) from err
+        assert isinstance(reset_obj, DataFrame)  # for mypy
+        return reset_obj, levels
+
+    @property
+    def nrows_expected(self) -> int:
+        """based on our axes, compute the expected nrows"""
+        return np.prod([i.cvalues.shape[0] for i in self.index_axes])
+
+    @property
+    def is_exists(self) -> bool:
+        """has this table been created"""
+        return "table" in self.group
+
+    @property
+    def storable(self):
+        return getattr(self.group, "table", None)
+
+    @property
+    def table(self):
+        """return the table group (this is my storable)"""
+        return self.storable
+
+    @property
+    def dtype(self):
+        return self.table.dtype
+
+    @property
+    def description(self):
+        return self.table.description
+
+    @property
+    def axes(self) -> itertools.chain[IndexCol]:
+        return itertools.chain(self.index_axes, self.values_axes)
+
+    @property
+    def ncols(self) -> int:
+        """the number of total columns in the values axes"""
+        return sum(len(a.values) for a in self.values_axes)
+
+    @property
+    def is_transposed(self) -> bool:
+        return False
+
+    @property
+    def data_orientation(self) -> tuple[int, ...]:
+        """return a tuple of my permutated axes, non_indexable at the front"""
+        return tuple(
+            itertools.chain(
+                [int(a[0]) for a in self.non_index_axes],
+                [int(a.axis) for a in self.index_axes],
+            )
+        )
+
+    def queryables(self) -> dict[str, Any]:
+        """return a dict of the kinds allowable columns for this object"""
+        # mypy doesn't recognize DataFrame._AXIS_NAMES, so we re-write it here
+        axis_names = {0: "index", 1: "columns"}
+
+        # compute the values_axes queryables
+        d1 = [(a.cname, a) for a in self.index_axes]
+        d2 = [(axis_names[axis], None) for axis, values in self.non_index_axes]
+        d3 = [
+            (v.cname, v) for v in self.values_axes if v.name in set(self.data_columns)
+        ]
+
+        return dict(d1 + d2 + d3)
+
+    def index_cols(self) -> list[tuple[Any, Any]]:
+        """return a list of my index cols"""
+        # Note: each `i.cname` below is assured to be a str.
+        return [(i.axis, i.cname) for i in self.index_axes]
+
+    def values_cols(self) -> list[str]:
+        """return a list of my values cols"""
+        return [i.cname for i in self.values_axes]
+
+    def _get_metadata_path(self, key: str) -> str:
+        """return the metadata pathname for this key"""
+        group = self.group._v_pathname
+        return f"{group}/meta/{key}/meta"
+
+    def write_metadata(self, key: str, values: np.ndarray) -> None:
+        """
+        Write out a metadata array to the key as a fixed-format Series.
+
+        Parameters
+        ----------
+        key : str
+        values : ndarray
+        """
+        self.parent.put(
+            self._get_metadata_path(key),
+            Series(values, copy=False),
+            format="table",
+            encoding=self.encoding,
+            errors=self.errors,
+            nan_rep=self.nan_rep,
+        )
+
+    def read_metadata(self, key: str):
+        """return the meta data array for this key"""
+        if getattr(getattr(self.group, "meta", None), key, None) is not None:
+            return self.parent.select(self._get_metadata_path(key))
+        return None
+
+    def set_attrs(self) -> None:
+        """set our table type & indexables"""
+        self.attrs.table_type = str(self.table_type)
+        self.attrs.index_cols = self.index_cols()
+        self.attrs.values_cols = self.values_cols()
+        self.attrs.non_index_axes = self.non_index_axes
+        self.attrs.data_columns = self.data_columns
+        self.attrs.nan_rep = self.nan_rep
+        self.attrs.encoding = self.encoding
+        self.attrs.errors = self.errors
+        self.attrs.levels = self.levels
+        self.attrs.info = self.info
+
+    def get_attrs(self) -> None:
+        """retrieve our attributes"""
+        self.non_index_axes = getattr(self.attrs, "non_index_axes", None) or []
+        self.data_columns = getattr(self.attrs, "data_columns", None) or []
+        self.info = getattr(self.attrs, "info", None) or {}
+        self.nan_rep = getattr(self.attrs, "nan_rep", None)
+        self.encoding = _ensure_encoding(getattr(self.attrs, "encoding", None))
+        self.errors = getattr(self.attrs, "errors", "strict")
+        self.levels: list[Hashable] = getattr(self.attrs, "levels", None) or []
+        self.index_axes = [a for a in self.indexables if a.is_an_indexable]
+        self.values_axes = [a for a in self.indexables if not a.is_an_indexable]
+
+    def validate_version(self, where=None) -> None:
+        """are we trying to operate on an old version?"""
+        if where is not None:
+            if self.is_old_version:
+                ws = incompatibility_doc % ".".join([str(x) for x in self.version])
+                warnings.warn(
+                    ws,
+                    IncompatibilityWarning,
+                    stacklevel=find_stack_level(),
+                )
+
+    def validate_min_itemsize(self, min_itemsize) -> None:
+        """
+        validate the min_itemsize doesn't contain items that are not in the
+        axes this needs data_columns to be defined
+        """
+        if min_itemsize is None:
+            return
+        if not isinstance(min_itemsize, dict):
+            return
+
+        q = self.queryables()
+        for k in min_itemsize:
+            # ok, apply generally
+            if k == "values":
+                continue
+            if k not in q:
+                raise ValueError(
+                    f"min_itemsize has the key [{k}] which is not an axis or "
+                    "data_column"
+                )
+
+    @cache_readonly
+    def indexables(self):
+        """create/cache the indexables if they don't exist"""
+        _indexables = []
+
+        desc = self.description
+        table_attrs = self.table.attrs
+
+        # Note: each of the `name` kwargs below are str, ensured
+        #  by the definition in index_cols.
+        # index columns
+        for i, (axis, name) in enumerate(self.attrs.index_cols):
+            atom = getattr(desc, name)
+            md = self.read_metadata(name)
+            meta = "category" if md is not None else None
+
+            kind_attr = f"{name}_kind"
+            kind = getattr(table_attrs, kind_attr, None)
+
+            index_col = IndexCol(
+                name=name,
+                axis=axis,
+                pos=i,
+                kind=kind,
+                typ=atom,
+                table=self.table,
+                meta=meta,
+                metadata=md,
+            )
+            _indexables.append(index_col)
+
+        # values columns
+        dc = set(self.data_columns)
+        base_pos = len(_indexables)
+
+        def f(i, c: str) -> DataCol:
+            assert isinstance(c, str)
+            klass = DataCol
+            if c in dc:
+                klass = DataIndexableCol
+
+            atom = getattr(desc, c)
+            adj_name = _maybe_adjust_name(c, self.version)
+
+            # TODO: why kind_attr here?
+            values = getattr(table_attrs, f"{adj_name}_kind", None)
+            dtype = getattr(table_attrs, f"{adj_name}_dtype", None)
+            # Argument 1 to "_dtype_to_kind" has incompatible type
+            # "Optional[Any]"; expected "str"  [arg-type]
+            kind = _dtype_to_kind(dtype)  # type: ignore[arg-type]
+
+            md = self.read_metadata(c)
+            # TODO: figure out why these two versions of `meta` dont always match.
+            #  meta = "category" if md is not None else None
+            meta = getattr(table_attrs, f"{adj_name}_meta", None)
+
+            obj = klass(
+                name=adj_name,
+                cname=c,
+                values=values,
+                kind=kind,
+                pos=base_pos + i,
+                typ=atom,
+                table=self.table,
+                meta=meta,
+                metadata=md,
+                dtype=dtype,
+            )
+            return obj
+
+        # Note: the definition of `values_cols` ensures that each
+        #  `c` below is a str.
+        _indexables.extend([f(i, c) for i, c in enumerate(self.attrs.values_cols)])
+
+        return _indexables
+
+    def create_index(
+        self, columns=None, optlevel=None, kind: str | None = None
+    ) -> None:
+        """
+        Create a pytables index on the specified columns.
+
+        Parameters
+        ----------
+        columns : None, bool, or listlike[str]
+            Indicate which columns to create an index on.
+
+            * False : Do not create any indexes.
+            * True : Create indexes on all columns.
+            * None : Create indexes on all columns.
+            * listlike : Create indexes on the given columns.
+
+        optlevel : int or None, default None
+            Optimization level, if None, pytables defaults to 6.
+        kind : str or None, default None
+            Kind of index, if None, pytables defaults to "medium".
+
+        Raises
+        ------
+        TypeError if trying to create an index on a complex-type column.
+
+        Notes
+        -----
+        Cannot index Time64Col or ComplexCol.
+        Pytables must be >= 3.0.
+        """
+        if not self.infer_axes():
+            return
+        if columns is False:
+            return
+
+        # index all indexables and data_columns
+        if columns is None or columns is True:
+            columns = [a.cname for a in self.axes if a.is_data_indexable]
+        if not isinstance(columns, (tuple, list)):
+            columns = [columns]
+
+        kw = {}
+        if optlevel is not None:
+            kw["optlevel"] = optlevel
+        if kind is not None:
+            kw["kind"] = kind
+
+        table = self.table
+        for c in columns:
+            v = getattr(table.cols, c, None)
+            if v is not None:
+                # remove the index if the kind/optlevel have changed
+                if v.is_indexed:
+                    index = v.index
+                    cur_optlevel = index.optlevel
+                    cur_kind = index.kind
+
+                    if kind is not None and cur_kind != kind:
+                        v.remove_index()
+                    else:
+                        kw["kind"] = cur_kind
+
+                    if optlevel is not None and cur_optlevel != optlevel:
+                        v.remove_index()
+                    else:
+                        kw["optlevel"] = cur_optlevel
+
+                # create the index
+                if not v.is_indexed:
+                    if v.type.startswith("complex"):
+                        raise TypeError(
+                            "Columns containing complex values can be stored but "
+                            "cannot be indexed when using table format. Either use "
+                            "fixed format, set index=False, or do not include "
+                            "the columns containing complex values to "
+                            "data_columns when initializing the table."
+                        )
+                    v.create_index(**kw)
+            elif c in self.non_index_axes[0][1]:
+                # GH 28156
+                raise AttributeError(
+                    f"column {c} is not a data_column.\n"
+                    f"In order to read column {c} you must reload the dataframe \n"
+                    f"into HDFStore and include {c} with the data_columns argument."
+                )
+
+    def _read_axes(
+        self, where, start: int | None = None, stop: int | None = None
+    ) -> list[tuple[np.ndarray, np.ndarray] | tuple[Index, Index]]:
+        """
+        Create the axes sniffed from the table.
+
+        Parameters
+        ----------
+        where : ???
+        start : int or None, default None
+        stop : int or None, default None
+
+        Returns
+        -------
+        List[Tuple[index_values, column_values]]
+        """
+        # create the selection
+        selection = Selection(self, where=where, start=start, stop=stop)
+        values = selection.select()
+
+        results = []
+        # convert the data
+        for a in self.axes:
+            a.set_info(self.info)
+            res = a.convert(
+                values,
+                nan_rep=self.nan_rep,
+                encoding=self.encoding,
+                errors=self.errors,
+            )
+            results.append(res)
+
+        return results
+
+    @classmethod
+    def get_object(cls, obj, transposed: bool):
+        """return the data for this obj"""
+        return obj
+
+    def validate_data_columns(self, data_columns, min_itemsize, non_index_axes) -> list:
+        """
+        take the input data_columns and min_itemize and create a data
+        columns spec
+        """
+        if not len(non_index_axes):
+            return []
+
+        axis, axis_labels = non_index_axes[0]
+        info = self.info.get(axis, {})
+        if info.get("type") == "MultiIndex" and data_columns:
+            raise ValueError(
+                f"cannot use a multi-index on axis [{axis}] with "
+                f"data_columns {data_columns}"
+            )
+
+        # evaluate the passed data_columns, True == use all columns
+        # take only valid axis labels
+        if data_columns is True:
+            data_columns = list(axis_labels)
+        elif data_columns is None:
+            data_columns = []
+
+        # if min_itemsize is a dict, add the keys (exclude 'values')
+        if isinstance(min_itemsize, dict):
+            existing_data_columns = set(data_columns)
+            data_columns = list(data_columns)  # ensure we do not modify
+            data_columns.extend(
+                [
+                    k
+                    for k in min_itemsize.keys()
+                    if k != "values" and k not in existing_data_columns
+                ]
+            )
+
+        # return valid columns in the order of our axis
+        return [c for c in data_columns if c in axis_labels]
+
+    def _create_axes(
+        self,
+        axes,
+        obj: DataFrame,
+        validate: bool = True,
+        nan_rep=None,
+        data_columns=None,
+        min_itemsize=None,
+    ):
+        """
+        Create and return the axes.
+
+        Parameters
+        ----------
+        axes: list or None
+            The names or numbers of the axes to create.
+        obj : DataFrame
+            The object to create axes on.
+        validate: bool, default True
+            Whether to validate the obj against an existing object already written.
+        nan_rep :
+            A value to use for string column nan_rep.
+        data_columns : List[str], True, or None, default None
+            Specify the columns that we want to create to allow indexing on.
+
+            * True : Use all available columns.
+            * None : Use no columns.
+            * List[str] : Use the specified columns.
+
+        min_itemsize: Dict[str, int] or None, default None
+            The min itemsize for a column in bytes.
+        """
+        if not isinstance(obj, DataFrame):
+            group = self.group._v_name
+            raise TypeError(
+                f"cannot properly create the storer for: [group->{group},"
+                f"value->{type(obj)}]"
+            )
+
+        # set the default axes if needed
+        if axes is None:
+            axes = [0]
+
+        # map axes to numbers
+        axes = [obj._get_axis_number(a) for a in axes]
+
+        # do we have an existing table (if so, use its axes & data_columns)
+        if self.infer_axes():
+            table_exists = True
+            axes = [a.axis for a in self.index_axes]
+            data_columns = list(self.data_columns)
+            nan_rep = self.nan_rep
+            # TODO: do we always have validate=True here?
+        else:
+            table_exists = False
+
+        new_info = self.info
+
+        assert self.ndim == 2  # with next check, we must have len(axes) == 1
+        # currently support on ndim-1 axes
+        if len(axes) != self.ndim - 1:
+            raise ValueError(
+                "currently only support ndim-1 indexers in an AppendableTable"
+            )
+
+        # create according to the new data
+        new_non_index_axes: list = []
+
+        # nan_representation
+        if nan_rep is None:
+            nan_rep = "nan"
+
+        # We construct the non-index-axis first, since that alters new_info
+        idx = next(x for x in [0, 1] if x not in axes)
+
+        a = obj.axes[idx]
+        # we might be able to change the axes on the appending data if necessary
+        append_axis = list(a)
+        if table_exists:
+            indexer = len(new_non_index_axes)  # i.e. 0
+            exist_axis = self.non_index_axes[indexer][1]
+            if not array_equivalent(
+                np.array(append_axis),
+                np.array(exist_axis),
+                strict_nan=True,
+                dtype_equal=True,
+            ):
+                # ahah! -> reindex
+                if array_equivalent(
+                    np.array(sorted(append_axis)),
+                    np.array(sorted(exist_axis)),
+                    strict_nan=True,
+                    dtype_equal=True,
+                ):
+                    append_axis = exist_axis
+
+        # the non_index_axes info
+        info = new_info.setdefault(idx, {})
+        info["names"] = list(a.names)
+        info["type"] = type(a).__name__
+
+        new_non_index_axes.append((idx, append_axis))
+
+        # Now we can construct our new index axis
+        idx = axes[0]
+        a = obj.axes[idx]
+        axis_name = obj._get_axis_name(idx)
+        new_index = _convert_index(axis_name, a, self.encoding, self.errors)
+        new_index.axis = idx
+
+        # Because we are always 2D, there is only one new_index, so
+        #  we know it will have pos=0
+        new_index.set_pos(0)
+        new_index.update_info(new_info)
+        new_index.maybe_set_size(min_itemsize)  # check for column conflicts
+
+        new_index_axes = [new_index]
+        j = len(new_index_axes)  # i.e. 1
+        assert j == 1
+
+        # reindex by our non_index_axes & compute data_columns
+        assert len(new_non_index_axes) == 1
+        for a in new_non_index_axes:
+            obj = _reindex_axis(obj, a[0], a[1])
+
+        transposed = new_index.axis == 1
+
+        # figure out data_columns and get out blocks
+        data_columns = self.validate_data_columns(
+            data_columns, min_itemsize, new_non_index_axes
+        )
+
+        frame = self.get_object(obj, transposed)._consolidate()
+
+        blocks, blk_items = self._get_blocks_and_items(
+            frame, table_exists, new_non_index_axes, self.values_axes, data_columns
+        )
+
+        # add my values
+        vaxes = []
+        for i, (blk, b_items) in enumerate(zip(blocks, blk_items)):
+            # shape of the data column are the indexable axes
+            klass = DataCol
+            name = None
+
+            # we have a data_column
+            if data_columns and len(b_items) == 1 and b_items[0] in data_columns:
+                klass = DataIndexableCol
+                name = b_items[0]
+                if not (name is None or isinstance(name, str)):
+                    # TODO: should the message here be more specifically non-str?
+                    raise ValueError("cannot have non-object label DataIndexableCol")
+
+            # make sure that we match up the existing columns
+            # if we have an existing table
+            existing_col: DataCol | None
+
+            if table_exists and validate:
+                try:
+                    existing_col = self.values_axes[i]
+                except (IndexError, KeyError) as err:
+                    raise ValueError(
+                        f"Incompatible appended table [{blocks}]"
+                        f"with existing table [{self.values_axes}]"
+                    ) from err
+            else:
+                existing_col = None
+
+            new_name = name or f"values_block_{i}"
+            data_converted = _maybe_convert_for_string_atom(
+                new_name,
+                blk.values,
+                existing_col=existing_col,
+                min_itemsize=min_itemsize,
+                nan_rep=nan_rep,
+                encoding=self.encoding,
+                errors=self.errors,
+                columns=b_items,
+            )
+            adj_name = _maybe_adjust_name(new_name, self.version)
+
+            typ = klass._get_atom(data_converted)
+            kind = _dtype_to_kind(data_converted.dtype.name)
+            tz = None
+            if getattr(data_converted, "tz", None) is not None:
+                tz = _get_tz(data_converted.tz)
+
+            meta = metadata = ordered = None
+            if isinstance(data_converted.dtype, CategoricalDtype):
+                ordered = data_converted.ordered
+                meta = "category"
+                metadata = np.asarray(data_converted.categories).ravel()
+            elif isinstance(blk.dtype, StringDtype):
+                meta = str(blk.dtype)
+
+            data, dtype_name = _get_data_and_dtype_name(data_converted)
+
+            col = klass(
+                name=adj_name,
+                cname=new_name,
+                values=list(b_items),
+                typ=typ,
+                pos=j,
+                kind=kind,
+                tz=tz,
+                ordered=ordered,
+                meta=meta,
+                metadata=metadata,
+                dtype=dtype_name,
+                data=data,
+            )
+            col.update_info(new_info)
+
+            vaxes.append(col)
+
+            j += 1
+
+        dcs = [col.name for col in vaxes if col.is_data_indexable]
+
+        new_table = type(self)(
+            parent=self.parent,
+            group=self.group,
+            encoding=self.encoding,
+            errors=self.errors,
+            index_axes=new_index_axes,
+            non_index_axes=new_non_index_axes,
+            values_axes=vaxes,
+            data_columns=dcs,
+            info=new_info,
+            nan_rep=nan_rep,
+        )
+        if hasattr(self, "levels"):
+            # TODO: get this into constructor, only for appropriate subclass
+            new_table.levels = self.levels
+
+        new_table.validate_min_itemsize(min_itemsize)
+
+        if validate and table_exists:
+            new_table.validate(self)
+
+        return new_table
+
+    @staticmethod
+    def _get_blocks_and_items(
+        frame: DataFrame,
+        table_exists: bool,
+        new_non_index_axes,
+        values_axes,
+        data_columns,
+    ):
+        # Helper to clarify non-state-altering parts of _create_axes
+        def get_blk_items(mgr):
+            return [mgr.items.take(blk.mgr_locs) for blk in mgr.blocks]
+
+        mgr = frame._mgr
+        blocks: list[Block] = list(mgr.blocks)
+        blk_items: list[Index] = get_blk_items(mgr)
+
+        if len(data_columns):
+            # TODO: prove that we only get here with axis == 1?
+            #  It is the case in all extant tests, but NOT the case
+            #  outside this `if len(data_columns)` check.
+
+            axis, axis_labels = new_non_index_axes[0]
+            new_labels = Index(axis_labels).difference(Index(data_columns))
+            mgr = frame.reindex(new_labels, axis=axis)._mgr
+
+            blocks = list(mgr.blocks)
+            blk_items = get_blk_items(mgr)
+            for c in data_columns:
+                # This reindex would raise ValueError if we had a duplicate
+                #  index, so we can infer that (as long as axis==1) we
+                #  get a single column back, so a single block.
+                mgr = frame.reindex([c], axis=axis)._mgr
+                blocks.extend(mgr.blocks)
+                blk_items.extend(get_blk_items(mgr))
+
+        # reorder the blocks in the same order as the existing table if we can
+        if table_exists:
+            by_items = {
+                tuple(b_items.tolist()): (b, b_items)
+                for b, b_items in zip(blocks, blk_items)
+            }
+            new_blocks: list[Block] = []
+            new_blk_items = []
+            for ea in values_axes:
+                items = tuple(ea.values)
+                try:
+                    b, b_items = by_items.pop(items)
+                    new_blocks.append(b)
+                    new_blk_items.append(b_items)
+                except (IndexError, KeyError) as err:
+                    jitems = ",".join([pprint_thing(item) for item in items])
+                    raise ValueError(
+                        f"cannot match existing table structure for [{jitems}] "
+                        "on appending data"
+                    ) from err
+            blocks = new_blocks
+            blk_items = new_blk_items
+
+        return blocks, blk_items
+
+    def process_axes(self, obj, selection: Selection, columns=None) -> DataFrame:
+        """process axes filters"""
+        # make a copy to avoid side effects
+        if columns is not None:
+            columns = list(columns)
+
+        # make sure to include levels if we have them
+        if columns is not None and self.is_multi_index:
+            assert isinstance(self.levels, list)  # assured by is_multi_index
+            for n in self.levels:
+                if n not in columns:
+                    columns.insert(0, n)
+
+        # reorder by any non_index_axes & limit to the select columns
+        for axis, labels in self.non_index_axes:
+            obj = _reindex_axis(obj, axis, labels, columns)
+
+            def process_filter(field, filt, op):
+                for axis_name in obj._AXIS_ORDERS:
+                    axis_number = obj._get_axis_number(axis_name)
+                    axis_values = obj._get_axis(axis_name)
+                    assert axis_number is not None
+
+                    # see if the field is the name of an axis
+                    if field == axis_name:
+                        # if we have a multi-index, then need to include
+                        # the levels
+                        if self.is_multi_index:
+                            filt = filt.union(Index(self.levels))
+
+                        takers = op(axis_values, filt)
+                        return obj.loc(axis=axis_number)[takers]
+
+                    # this might be the name of a file IN an axis
+                    elif field in axis_values:
+                        # we need to filter on this dimension
+                        values = ensure_index(getattr(obj, field).values)
+                        filt = ensure_index(filt)
+
+                        # hack until we support reversed dim flags
+                        if isinstance(obj, DataFrame):
+                            axis_number = 1 - axis_number
+
+                        takers = op(values, filt)
+                        return obj.loc(axis=axis_number)[takers]
+
+                raise ValueError(f"cannot find the field [{field}] for filtering!")
+
+        # apply the selection filters (but keep in the same order)
+        if selection.filter is not None:
+            for field, op, filt in selection.filter.format():
+                obj = process_filter(field, filt, op)
+
+        return obj
+
+    def create_description(
+        self,
+        complib,
+        complevel: int | None,
+        fletcher32: bool,
+        expectedrows: int | None,
+    ) -> dict[str, Any]:
+        """create the description of the table from the axes & values"""
+        # provided expected rows if its passed
+        if expectedrows is None:
+            expectedrows = max(self.nrows_expected, 10000)
+
+        d = {"name": "table", "expectedrows": expectedrows}
+
+        # description from the axes & values
+        d["description"] = {a.cname: a.typ for a in self.axes}
+
+        if complib:
+            if complevel is None:
+                complevel = self._complevel or 9
+            filters = _tables().Filters(
+                complevel=complevel,
+                complib=complib,
+                fletcher32=fletcher32 or self._fletcher32,
+            )
+            d["filters"] = filters
+        elif self._filters is not None:
+            d["filters"] = self._filters
+
+        return d
+
+    def read_coordinates(
+        self, where=None, start: int | None = None, stop: int | None = None
+    ):
+        """
+        select coordinates (row numbers) from a table; return the
+        coordinates object
+        """
+        # validate the version
+        self.validate_version(where)
+
+        # infer the data kind
+        if not self.infer_axes():
+            return False
+
+        # create the selection
+        selection = Selection(self, where=where, start=start, stop=stop)
+        coords = selection.select_coords()
+        if selection.filter is not None:
+            for field, op, filt in selection.filter.format():
+                data = self.read_column(
+                    field, start=coords.min(), stop=coords.max() + 1
+                )
+                coords = coords[op(data.iloc[coords - coords.min()], filt).values]
+
+        return Index(coords)
+
+    def read_column(
+        self,
+        column: str,
+        where=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ):
+        """
+        return a single column from the table, generally only indexables
+        are interesting
+        """
+        # validate the version
+        self.validate_version()
+
+        # infer the data kind
+        if not self.infer_axes():
+            return False
+
+        if where is not None:
+            raise TypeError("read_column does not currently accept a where clause")
+
+        # find the axes
+        for a in self.axes:
+            if column == a.name:
+                if not a.is_data_indexable:
+                    raise ValueError(
+                        f"column [{column}] can not be extracted individually; "
+                        "it is not data indexable"
+                    )
+
+                # column must be an indexable or a data column
+                c = getattr(self.table.cols, column)
+                a.set_info(self.info)
+                col_values = a.convert(
+                    c[start:stop],
+                    nan_rep=self.nan_rep,
+                    encoding=self.encoding,
+                    errors=self.errors,
+                )
+                cvs = col_values[1]
+                dtype = getattr(self.table.attrs, f"{column}_meta", None)
+                return Series(cvs, name=column, copy=False, dtype=dtype)
+
+        raise KeyError(f"column [{column}] not found in the table")
+
+
+class WORMTable(Table):
+    """
+    a write-once read-many table: this format DOES NOT ALLOW appending to a
+    table. writing is a one-time operation the data are stored in a format
+    that allows for searching the data on disk
+    """
+
+    table_type = "worm"
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ):
+        """
+        read the indices and the indexing array, calculate offset rows and return
+        """
+        raise NotImplementedError("WORMTable needs to implement read")
+
+    def write(self, obj, **kwargs) -> None:
+        """
+        write in a format that we can search later on (but cannot append
+        to): write out the indices and the values using _write_array
+        (e.g. a CArray) create an indexing table so that we can search
+        """
+        raise NotImplementedError("WORMTable needs to implement write")
+
+
+class AppendableTable(Table):
+    """support the new appendable table formats"""
+
+    table_type = "appendable"
+
+    # error: Signature of "write" incompatible with supertype "Fixed"
+    def write(  # type: ignore[override]
+        self,
+        obj,
+        axes=None,
+        append: bool = False,
+        complib=None,
+        complevel=None,
+        fletcher32=None,
+        min_itemsize=None,
+        chunksize: int | None = None,
+        expectedrows=None,
+        dropna: bool = False,
+        nan_rep=None,
+        data_columns=None,
+        track_times: bool = True,
+    ) -> None:
+        if not append and self.is_exists:
+            self._handle.remove_node(self.group, "table")
+
+        # create the axes
+        table = self._create_axes(
+            axes=axes,
+            obj=obj,
+            validate=append,
+            min_itemsize=min_itemsize,
+            nan_rep=nan_rep,
+            data_columns=data_columns,
+        )
+
+        for a in table.axes:
+            a.validate_names()
+
+        if not table.is_exists:
+            # create the table
+            options = table.create_description(
+                complib=complib,
+                complevel=complevel,
+                fletcher32=fletcher32,
+                expectedrows=expectedrows,
+            )
+
+            # set the table attributes
+            table.set_attrs()
+
+            options["track_times"] = track_times
+
+            # create the table
+            table._handle.create_table(table.group, **options)
+
+        # update my info
+        table.attrs.info = table.info
+
+        # validate the axes and set the kinds
+        for a in table.axes:
+            a.validate_and_set(table, append)
+
+        # add the rows
+        table.write_data(chunksize, dropna=dropna)
+
+    def write_data(self, chunksize: int | None, dropna: bool = False) -> None:
+        """
+        we form the data into a 2-d including indexes,values,mask write chunk-by-chunk
+        """
+        names = self.dtype.names
+        nrows = self.nrows_expected
+
+        # if dropna==True, then drop ALL nan rows
+        masks = []
+        if dropna:
+            for a in self.values_axes:
+                # figure the mask: only do if we can successfully process this
+                # column, otherwise ignore the mask
+                mask = isna(a.data).all(axis=0)
+                if isinstance(mask, np.ndarray):
+                    masks.append(mask.astype("u1", copy=False))
+
+        # consolidate masks
+        if len(masks):
+            mask = masks[0]
+            for m in masks[1:]:
+                mask = mask & m
+            mask = mask.ravel()
+        else:
+            mask = None
+
+        # broadcast the indexes if needed
+        indexes = [a.cvalues for a in self.index_axes]
+        nindexes = len(indexes)
+        assert nindexes == 1, nindexes  # ensures we dont need to broadcast
+
+        # transpose the values so first dimension is last
+        # reshape the values if needed
+        values = [a.take_data() for a in self.values_axes]
+        values = [v.transpose(np.roll(np.arange(v.ndim), v.ndim - 1)) for v in values]
+        bvalues = []
+        for i, v in enumerate(values):
+            new_shape = (nrows,) + self.dtype[names[nindexes + i]].shape
+            bvalues.append(v.reshape(new_shape))
+
+        # write the chunks
+        if chunksize is None:
+            chunksize = 100000
+
+        rows = np.empty(min(chunksize, nrows), dtype=self.dtype)
+        chunks = nrows // chunksize + 1
+        for i in range(chunks):
+            start_i = i * chunksize
+            end_i = min((i + 1) * chunksize, nrows)
+            if start_i >= end_i:
+                break
+
+            self.write_data_chunk(
+                rows,
+                indexes=[a[start_i:end_i] for a in indexes],
+                mask=mask[start_i:end_i] if mask is not None else None,
+                values=[v[start_i:end_i] for v in bvalues],
+            )
+
+    def write_data_chunk(
+        self,
+        rows: np.ndarray,
+        indexes: list[np.ndarray],
+        mask: npt.NDArray[np.bool_] | None,
+        values: list[np.ndarray],
+    ) -> None:
+        """
+        Parameters
+        ----------
+        rows : an empty memory space where we are putting the chunk
+        indexes : an array of the indexes
+        mask : an array of the masks
+        values : an array of the values
+        """
+        # 0 len
+        for v in values:
+            if not np.prod(v.shape):
+                return
+
+        nrows = indexes[0].shape[0]
+        if nrows != len(rows):
+            rows = np.empty(nrows, dtype=self.dtype)
+        names = self.dtype.names
+        nindexes = len(indexes)
+
+        # indexes
+        for i, idx in enumerate(indexes):
+            rows[names[i]] = idx
+
+        # values
+        for i, v in enumerate(values):
+            rows[names[i + nindexes]] = v
+
+        # mask
+        if mask is not None:
+            m = ~mask.ravel().astype(bool, copy=False)
+            if not m.all():
+                rows = rows[m]
+
+        if len(rows):
+            self.table.append(rows)
+            self.table.flush()
+
+    def delete(
+        self, where=None, start: int | None = None, stop: int | None = None
+    ) -> int | None:
+        # delete all rows (and return the nrows)
+        if where is None or not len(where):
+            if start is None and stop is None:
+                nrows = self.nrows
+                self._handle.remove_node(self.group, recursive=True)
+            else:
+                # pytables<3.0 would remove a single row with stop=None
+                if stop is None:
+                    stop = self.nrows
+                nrows = self.table.remove_rows(start=start, stop=stop)
+                self.table.flush()
+            return nrows
+
+        # infer the data kind
+        if not self.infer_axes():
+            return None
+
+        # create the selection
+        table = self.table
+        selection = Selection(self, where, start=start, stop=stop)
+        values = selection.select_coords()
+
+        # delete the rows in reverse order
+        sorted_series = Series(values, copy=False).sort_values()
+        ln = len(sorted_series)
+
+        if ln:
+            # construct groups of consecutive rows
+            diff = sorted_series.diff()
+            groups = list(diff[diff > 1].index)
+
+            # 1 group
+            if not len(groups):
+                groups = [0]
+
+            # final element
+            if groups[-1] != ln:
+                groups.append(ln)
+
+            # initial element
+            if groups[0] != 0:
+                groups.insert(0, 0)
+
+            # we must remove in reverse order!
+            pg = groups.pop()
+            for g in reversed(groups):
+                rows = sorted_series.take(range(g, pg))
+                table.remove_rows(
+                    start=rows[rows.index[0]], stop=rows[rows.index[-1]] + 1
+                )
+                pg = g
+
+            self.table.flush()
+
+        # return the number of rows removed
+        return ln
+
+
+class AppendableFrameTable(AppendableTable):
+    """support the new appendable table formats"""
+
+    pandas_kind = "frame_table"
+    table_type = "appendable_frame"
+    ndim = 2
+    obj_type: type[DataFrame | Series] = DataFrame
+
+    @property
+    def is_transposed(self) -> bool:
+        return self.index_axes[0].axis == 1
+
+    @classmethod
+    def get_object(cls, obj, transposed: bool):
+        """these are written transposed"""
+        if transposed:
+            obj = obj.T
+        return obj
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ):
+        # validate the version
+        self.validate_version(where)
+
+        # infer the data kind
+        if not self.infer_axes():
+            return None
+
+        result = self._read_axes(where=where, start=start, stop=stop)
+
+        info = (
+            self.info.get(self.non_index_axes[0][0], {})
+            if len(self.non_index_axes)
+            else {}
+        )
+
+        inds = [i for i, ax in enumerate(self.axes) if ax is self.index_axes[0]]
+        assert len(inds) == 1
+        ind = inds[0]
+
+        index = result[ind][0]
+
+        frames = []
+        for i, a in enumerate(self.axes):
+            if a not in self.values_axes:
+                continue
+            index_vals, cvalues = result[i]
+
+            # we could have a multi-index constructor here
+            # ensure_index doesn't recognized our list-of-tuples here
+            if info.get("type") != "MultiIndex":
+                cols = Index(index_vals)
+            else:
+                cols = MultiIndex.from_tuples(index_vals)
+
+            names = info.get("names")
+            if names is not None:
+                cols.set_names(names, inplace=True)
+
+            if self.is_transposed:
+                values = cvalues
+                index_ = cols
+                cols_ = Index(index, name=getattr(index, "name", None))
+            else:
+                values = cvalues.T
+                index_ = Index(index, name=getattr(index, "name", None))
+                cols_ = cols
+
+            # if we have a DataIndexableCol, its shape will only be 1 dim
+            if values.ndim == 1 and isinstance(values, np.ndarray):
+                values = values.reshape((1, values.shape[0]))
+
+            if isinstance(values, (np.ndarray, DatetimeArray)):
+                df = DataFrame(values.T, columns=cols_, index=index_, copy=False)
+            elif isinstance(values, Index):
+                df = DataFrame(values, columns=cols_, index=index_)
+            else:
+                # Categorical
+                df = DataFrame._from_arrays([values], columns=cols_, index=index_)
+            if not (using_string_dtype() and values.dtype.kind == "O"):
+                assert (df.dtypes == values.dtype).all(), (df.dtypes, values.dtype)
+
+            # If str / string dtype is stored in meta, use that.
+            converted = False
+            for column in cols_:
+                dtype = getattr(self.table.attrs, f"{column}_meta", None)
+                if dtype in ["str", "string"]:
+                    df[column] = df[column].astype(dtype)
+                    converted = True
+            # Otherwise try inference.
+            if (
+                not converted
+                and using_string_dtype()
+                and isinstance(values, np.ndarray)
+                and is_string_array(
+                    values,
+                    skipna=True,
+                )
+            ):
+                df = df.astype(StringDtype(na_value=np.nan))
+            frames.append(df)
+
+        if len(frames) == 1:
+            df = frames[0]
+        else:
+            df = concat(frames, axis=1)
+
+        selection = Selection(self, where=where, start=start, stop=stop)
+        # apply the selection filters & axis orderings
+        df = self.process_axes(df, selection=selection, columns=columns)
+        return df
+
+
+class AppendableSeriesTable(AppendableFrameTable):
+    """support the new appendable table formats"""
+
+    pandas_kind = "series_table"
+    table_type = "appendable_series"
+    ndim = 2
+    obj_type = Series
+
+    @property
+    def is_transposed(self) -> bool:
+        return False
+
+    @classmethod
+    def get_object(cls, obj, transposed: bool):
+        return obj
+
+    # error: Signature of "write" incompatible with supertype "Fixed"
+    def write(self, obj, data_columns=None, **kwargs) -> None:  # type: ignore[override]
+        """we are going to write this as a frame table"""
+        if not isinstance(obj, DataFrame):
+            name = obj.name or "values"
+            obj = obj.to_frame(name)
+        super().write(obj=obj, data_columns=obj.columns.tolist(), **kwargs)
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> Series:
+        is_multi_index = self.is_multi_index
+        if columns is not None and is_multi_index:
+            assert isinstance(self.levels, list)  # needed for mypy
+            for n in self.levels:
+                if n not in columns:
+                    columns.insert(0, n)
+        s = super().read(where=where, columns=columns, start=start, stop=stop)
+        if is_multi_index:
+            s.set_index(self.levels, inplace=True)
+
+        s = s.iloc[:, 0]
+
+        # remove the default name
+        if s.name == "values":
+            s.name = None
+        return s
+
+
+class AppendableMultiSeriesTable(AppendableSeriesTable):
+    """support the new appendable table formats"""
+
+    pandas_kind = "series_table"
+    table_type = "appendable_multiseries"
+
+    #  error: Signature of "write" incompatible with supertype "Fixed"
+    def write(self, obj, **kwargs) -> None:  # type: ignore[override]
+        """we are going to write this as a frame table"""
+        name = obj.name or "values"
+        newobj, self.levels = self.validate_multiindex(obj)
+        assert isinstance(self.levels, list)  # for mypy
+        cols = list(self.levels)
+        cols.append(name)
+        newobj.columns = Index(cols)
+        super().write(obj=newobj, **kwargs)
+
+
+class GenericTable(AppendableFrameTable):
+    """a table that read/writes the generic pytables table format"""
+
+    pandas_kind = "frame_table"
+    table_type = "generic_table"
+    ndim = 2
+    obj_type = DataFrame
+    levels: list[Hashable]
+
+    @property
+    def pandas_type(self) -> str:
+        return self.pandas_kind
+
+    @property
+    def storable(self):
+        return getattr(self.group, "table", None) or self.group
+
+    def get_attrs(self) -> None:
+        """retrieve our attributes"""
+        self.non_index_axes = []
+        self.nan_rep = None
+        self.levels = []
+
+        self.index_axes = [a for a in self.indexables if a.is_an_indexable]
+        self.values_axes = [a for a in self.indexables if not a.is_an_indexable]
+        self.data_columns = [a.name for a in self.values_axes]
+
+    @cache_readonly
+    def indexables(self):
+        """create the indexables from the table description"""
+        d = self.description
+
+        # TODO: can we get a typ for this?  AFAICT it is the only place
+        #  where we aren't passing one
+        # the index columns is just a simple index
+        md = self.read_metadata("index")
+        meta = "category" if md is not None else None
+        index_col = GenericIndexCol(
+            name="index", axis=0, table=self.table, meta=meta, metadata=md
+        )
+
+        _indexables: list[GenericIndexCol | GenericDataIndexableCol] = [index_col]
+
+        for i, n in enumerate(d._v_names):
+            assert isinstance(n, str)
+
+            atom = getattr(d, n)
+            md = self.read_metadata(n)
+            meta = "category" if md is not None else None
+            dc = GenericDataIndexableCol(
+                name=n,
+                pos=i,
+                values=[n],
+                typ=atom,
+                table=self.table,
+                meta=meta,
+                metadata=md,
+            )
+            _indexables.append(dc)
+
+        return _indexables
+
+    # error: Signature of "write" incompatible with supertype "AppendableTable"
+    def write(self, **kwargs) -> None:  # type: ignore[override]
+        raise NotImplementedError("cannot write on an generic table")
+
+
+class AppendableMultiFrameTable(AppendableFrameTable):
+    """a frame with a multi-index"""
+
+    table_type = "appendable_multiframe"
+    obj_type = DataFrame
+    ndim = 2
+    _re_levels = re.compile(r"^level_\d+$")
+
+    @property
+    def table_type_short(self) -> str:
+        return "appendable_multi"
+
+    # error: Signature of "write" incompatible with supertype "Fixed"
+    def write(self, obj, data_columns=None, **kwargs) -> None:  # type: ignore[override]
+        if data_columns is None:
+            data_columns = []
+        elif data_columns is True:
+            data_columns = obj.columns.tolist()
+        obj, self.levels = self.validate_multiindex(obj)
+        assert isinstance(self.levels, list)  # for mypy
+        for n in self.levels:
+            if n not in data_columns:
+                data_columns.insert(0, n)
+        super().write(obj=obj, data_columns=data_columns, **kwargs)
+
+    def read(
+        self,
+        where=None,
+        columns=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> DataFrame:
+        df = super().read(where=where, columns=columns, start=start, stop=stop)
+        df = df.set_index(self.levels)
+
+        # remove names for 'level_%d'
+        df.index = df.index.set_names(
+            [None if self._re_levels.search(name) else name for name in df.index.names]
+        )
+
+        return df
+
+
+def _reindex_axis(
+    obj: DataFrame, axis: AxisInt, labels: Index, other=None
+) -> DataFrame:
+    ax = obj._get_axis(axis)
+    labels = ensure_index(labels)
+
+    # try not to reindex even if other is provided
+    # if it equals our current index
+    if other is not None:
+        other = ensure_index(other)
+    if (other is None or labels.equals(other)) and labels.equals(ax):
+        return obj
+
+    labels = ensure_index(labels.unique())
+    if other is not None:
+        labels = ensure_index(other.unique()).intersection(labels, sort=False)
+    if not labels.equals(ax):
+        slicer: list[slice | Index] = [slice(None, None)] * obj.ndim
+        slicer[axis] = labels
+        obj = obj.loc[tuple(slicer)]
+    return obj
+
+
+# tz to/from coercion
+
+
+def _get_tz(tz: tzinfo) -> str | tzinfo:
+    """for a tz-aware type, return an encoded zone"""
+    zone = timezones.get_timezone(tz)
+    return zone
+
+
+def _set_tz(
+    values: npt.NDArray[np.int64], tz: str | tzinfo | None, datetime64_dtype: str
+) -> DatetimeArray:
+    """
+    Coerce the values to a DatetimeArray with appropriate tz.
+
+    Parameters
+    ----------
+    values : ndarray[int64]
+    tz : str, tzinfo, or None
+    datetime64_dtype : str, e.g. "datetime64[ns]", "datetime64[25s]"
+    """
+    assert values.dtype == "i8", values.dtype
+    # Argument "tz" to "tz_to_dtype" has incompatible type "str | tzinfo | None";
+    # expected "tzinfo"
+    unit, _ = np.datetime_data(datetime64_dtype)  # parsing dtype: unit, count
+    dtype = tz_to_dtype(tz=tz, unit=unit)  # type: ignore[arg-type]
+    dta = DatetimeArray._from_sequence(values, dtype=dtype)
+    return dta
+
+
+def _convert_index(name: str, index: Index, encoding: str, errors: str) -> IndexCol:
+    assert isinstance(name, str)
+
+    index_name = index.name
+    # error: Argument 1 to "_get_data_and_dtype_name" has incompatible type "Index";
+    # expected "Union[ExtensionArray, ndarray]"
+    converted, dtype_name = _get_data_and_dtype_name(index)  # type: ignore[arg-type]
+    kind = _dtype_to_kind(dtype_name)
+    atom = DataIndexableCol._get_atom(converted)
+
+    if (
+        lib.is_np_dtype(index.dtype, "iu")
+        or needs_i8_conversion(index.dtype)
+        or is_bool_dtype(index.dtype)
+    ):
+        # Includes Index, RangeIndex, DatetimeIndex, TimedeltaIndex, PeriodIndex,
+        #  in which case "kind" is "integer", "integer", "datetime64",
+        #  "timedelta64", and "integer", respectively.
+        return IndexCol(
+            name,
+            values=converted,
+            kind=kind,
+            typ=atom,
+            freq=getattr(index, "freq", None),
+            tz=getattr(index, "tz", None),
+            index_name=index_name,
+        )
+
+    if isinstance(index, MultiIndex):
+        raise TypeError("MultiIndex not supported here!")
+
+    inferred_type = lib.infer_dtype(index, skipna=False)
+    # we won't get inferred_type of "datetime64" or "timedelta64" as these
+    #  would go through the DatetimeIndex/TimedeltaIndex paths above
+
+    values = np.asarray(index)
+
+    if inferred_type == "date":
+        converted = np.asarray([v.toordinal() for v in values], dtype=np.int32)
+        return IndexCol(
+            name, converted, "date", _tables().Time32Col(), index_name=index_name
+        )
+    elif inferred_type == "string":
+        converted = _convert_string_array(values, encoding, errors)
+        itemsize = converted.dtype.itemsize
+        return IndexCol(
+            name,
+            converted,
+            "string",
+            _tables().StringCol(itemsize),
+            index_name=index_name,
+        )
+
+    elif inferred_type in ["integer", "floating"]:
+        return IndexCol(
+            name, values=converted, kind=kind, typ=atom, index_name=index_name
+        )
+    else:
+        assert isinstance(converted, np.ndarray) and converted.dtype == object
+        assert kind == "object", kind
+        atom = _tables().ObjectAtom()
+        return IndexCol(name, converted, kind, atom, index_name=index_name)
+
+
+def _unconvert_index(data, kind: str, encoding: str, errors: str) -> np.ndarray | Index:
+    index: Index | np.ndarray
+
+    if kind.startswith("datetime64"):
+        if kind == "datetime64":
+            # created before we stored resolution information
+            index = DatetimeIndex(data)
+        else:
+            index = DatetimeIndex(data.view(kind))
+    elif kind == "timedelta64":
+        index = TimedeltaIndex(data)
+    elif kind == "date":
+        try:
+            index = np.asarray([date.fromordinal(v) for v in data], dtype=object)
+        except ValueError:
+            index = np.asarray([date.fromtimestamp(v) for v in data], dtype=object)
+    elif kind in ("integer", "float", "bool"):
+        index = np.asarray(data)
+    elif kind in ("string"):
+        index = _unconvert_string_array(
+            data, nan_rep=None, encoding=encoding, errors=errors
+        )
+    elif kind == "object":
+        index = np.asarray(data[0])
+    else:  # pragma: no cover
+        raise ValueError(f"unrecognized index type {kind}")
+    return index
+
+
+def _maybe_convert_for_string_atom(
+    name: str,
+    bvalues: ArrayLike,
+    existing_col,
+    min_itemsize,
+    nan_rep,
+    encoding,
+    errors,
+    columns: list[str],
+):
+    if isinstance(bvalues.dtype, StringDtype):
+        # "ndarray[Any, Any]" has no attribute "to_numpy"
+        bvalues = bvalues.to_numpy()  # type: ignore[union-attr]
+    if bvalues.dtype != object:
+        return bvalues
+
+    bvalues = cast(np.ndarray, bvalues)
+
+    dtype_name = bvalues.dtype.name
+    inferred_type = lib.infer_dtype(bvalues, skipna=False)
+
+    if inferred_type == "date":
+        raise TypeError("[date] is not implemented as a table column")
+    if inferred_type == "datetime":
+        # after GH#8260
+        # this only would be hit for a multi-timezone dtype which is an error
+        raise TypeError(
+            "too many timezones in this block, create separate data columns"
+        )
+
+    if not (inferred_type == "string" or dtype_name == "object"):
+        return bvalues
+
+    mask = isna(bvalues)
+    data = bvalues.copy()
+    data[mask] = nan_rep
+
+    if existing_col and mask.any() and len(nan_rep) > existing_col.itemsize:
+        raise ValueError("NaN representation is too large for existing column size")
+
+    # see if we have a valid string type
+    inferred_type = lib.infer_dtype(data, skipna=False)
+    if inferred_type != "string":
+        # we cannot serialize this data, so report an exception on a column
+        # by column basis
+
+        # expected behaviour:
+        # search block for a non-string object column by column
+        for i in range(data.shape[0]):
+            col = data[i]
+            inferred_type = lib.infer_dtype(col, skipna=False)
+            if inferred_type != "string":
+                error_column_label = columns[i] if len(columns) > i else f"No.{i}"
+                raise TypeError(
+                    f"Cannot serialize the column [{error_column_label}]\n"
+                    f"because its data contents are not [string] but "
+                    f"[{inferred_type}] object dtype"
+                )
+
+    # itemsize is the maximum length of a string (along any dimension)
+
+    data_converted = _convert_string_array(data, encoding, errors).reshape(data.shape)
+    itemsize = data_converted.itemsize
+
+    # specified min_itemsize?
+    if isinstance(min_itemsize, dict):
+        min_itemsize = int(min_itemsize.get(name) or min_itemsize.get("values") or 0)
+    itemsize = max(min_itemsize or 0, itemsize)
+
+    # check for column in the values conflicts
+    if existing_col is not None:
+        eci = existing_col.validate_col(itemsize)
+        if eci is not None and eci > itemsize:
+            itemsize = eci
+
+    data_converted = data_converted.astype(f"|S{itemsize}", copy=False)
+    return data_converted
+
+
+def _convert_string_array(data: np.ndarray, encoding: str, errors: str) -> np.ndarray:
+    """
+    Take a string-like that is object dtype and coerce to a fixed size string type.
+
+    Parameters
+    ----------
+    data : np.ndarray[object]
+    encoding : str
+    errors : str
+        Handler for encoding errors.
+
+    Returns
+    -------
+    np.ndarray[fixed-length-string]
+    """
+    # encode if needed
+    if len(data):
+        data = (
+            Series(data.ravel(), copy=False)
+            .str.encode(encoding, errors)
+            ._values.reshape(data.shape)
+        )
+
+    # create the sized dtype
+    ensured = ensure_object(data.ravel())
+    itemsize = max(1, libwriters.max_len_string_array(ensured))
+
+    data = np.asarray(data, dtype=f"S{itemsize}")
+    return data
+
+
+def _unconvert_string_array(
+    data: np.ndarray, nan_rep, encoding: str, errors: str
+) -> np.ndarray:
+    """
+    Inverse of _convert_string_array.
+
+    Parameters
+    ----------
+    data : np.ndarray[fixed-length-string]
+    nan_rep : the storage repr of NaN
+    encoding : str
+    errors : str
+        Handler for encoding errors.
+
+    Returns
+    -------
+    np.ndarray[object]
+        Decoded data.
+    """
+    shape = data.shape
+    data = np.asarray(data.ravel(), dtype=object)
+
+    if len(data):
+        itemsize = libwriters.max_len_string_array(ensure_object(data))
+        dtype = f"U{itemsize}"
+
+        if isinstance(data[0], bytes):
+            ser = Series(data, copy=False).str.decode(encoding, errors=errors)
+            data = ser.to_numpy()
+            data.flags.writeable = True
+        else:
+            data = data.astype(dtype, copy=False).astype(object, copy=False)
+
+    if nan_rep is None:
+        nan_rep = "nan"
+
+    libwriters.string_array_replace_from_nan_rep(data, nan_rep)
+    return data.reshape(shape)
+
+
+def _maybe_convert(values: np.ndarray, val_kind: str, encoding: str, errors: str):
+    assert isinstance(val_kind, str), type(val_kind)
+    if _need_convert(val_kind):
+        conv = _get_converter(val_kind, encoding, errors)
+        values = conv(values)
+    return values
+
+
+def _get_converter(kind: str, encoding: str, errors: str):
+    if kind == "datetime64":
+        return lambda x: np.asarray(x, dtype="M8[ns]")
+    elif "datetime64" in kind:
+        return lambda x: np.asarray(x, dtype=kind)
+    elif kind == "string":
+        return lambda x: _unconvert_string_array(
+            x, nan_rep=None, encoding=encoding, errors=errors
+        )
+    else:  # pragma: no cover
+        raise ValueError(f"invalid kind {kind}")
+
+
+def _need_convert(kind: str) -> bool:
+    if kind in ("datetime64", "string") or "datetime64" in kind:
+        return True
+    return False
+
+
+def _maybe_adjust_name(name: str, version: Sequence[int]) -> str:
+    """
+    Prior to 0.10.1, we named values blocks like: values_block_0 an the
+    name values_0, adjust the given name if necessary.
+
+    Parameters
+    ----------
+    name : str
+    version : Tuple[int, int, int]
+
+    Returns
+    -------
+    str
+    """
+    if isinstance(version, str) or len(version) < 3:
+        raise ValueError("Version is incorrect, expected sequence of 3 integers.")
+
+    if version[0] == 0 and version[1] <= 10 and version[2] == 0:
+        m = re.search(r"values_block_(\d+)", name)
+        if m:
+            grp = m.groups()[0]
+            name = f"values_{grp}"
+    return name
+
+
+def _dtype_to_kind(dtype_str: str) -> str:
+    """
+    Find the "kind" string describing the given dtype name.
+    """
+    if dtype_str.startswith(("string", "bytes")):
+        kind = "string"
+    elif dtype_str.startswith("float"):
+        kind = "float"
+    elif dtype_str.startswith("complex"):
+        kind = "complex"
+    elif dtype_str.startswith(("int", "uint")):
+        kind = "integer"
+    elif dtype_str.startswith("datetime64"):
+        kind = dtype_str
+    elif dtype_str.startswith("timedelta"):
+        kind = "timedelta64"
+    elif dtype_str.startswith("bool"):
+        kind = "bool"
+    elif dtype_str.startswith("category"):
+        kind = "category"
+    elif dtype_str.startswith("period"):
+        # We store the `freq` attr so we can restore from integers
+        kind = "integer"
+    elif dtype_str == "object":
+        kind = "object"
+    elif dtype_str == "str":
+        kind = "str"
+    else:
+        raise ValueError(f"cannot interpret dtype of [{dtype_str}]")
+
+    return kind
+
+
+def _get_data_and_dtype_name(data: ArrayLike):
+    """
+    Convert the passed data into a storable form and a dtype string.
+    """
+    if isinstance(data, Categorical):
+        data = data.codes
+
+    if isinstance(data.dtype, DatetimeTZDtype):
+        # For datetime64tz we need to drop the TZ in tests TODO: why?
+        dtype_name = f"datetime64[{data.dtype.unit}]"
+    else:
+        dtype_name = data.dtype.name
+
+    if data.dtype.kind in "mM":
+        data = np.asarray(data.view("i8"))
+        # TODO: we used to reshape for the dt64tz case, but no longer
+        #  doing that doesn't seem to break anything.  why?
+
+    elif isinstance(data, PeriodIndex):
+        data = data.asi8
+
+    data = np.asarray(data)
+    return data, dtype_name
+
+
+class Selection:
+    """
+    Carries out a selection operation on a tables.Table object.
+
+    Parameters
+    ----------
+    table : a Table object
+    where : list of Terms (or convertible to)
+    start, stop: indices to start and/or stop selection
+
+    """
+
+    def __init__(
+        self,
+        table: Table,
+        where=None,
+        start: int | None = None,
+        stop: int | None = None,
+    ) -> None:
+        self.table = table
+        self.where = where
+        self.start = start
+        self.stop = stop
+        self.condition = None
+        self.filter = None
+        self.terms = None
+        self.coordinates = None
+
+        if is_list_like(where):
+            # see if we have a passed coordinate like
+            with suppress(ValueError):
+                inferred = lib.infer_dtype(where, skipna=False)
+                if inferred in ("integer", "boolean"):
+                    where = np.asarray(where)
+                    if where.dtype == np.bool_:
+                        start, stop = self.start, self.stop
+                        if start is None:
+                            start = 0
+                        if stop is None:
+                            stop = self.table.nrows
+                        self.coordinates = np.arange(start, stop)[where]
+                    elif issubclass(where.dtype.type, np.integer):
+                        if (self.start is not None and (where < self.start).any()) or (
+                            self.stop is not None and (where >= self.stop).any()
+                        ):
+                            raise ValueError(
+                                "where must have index locations >= start and < stop"
+                            )
+                        self.coordinates = where
+
+        if self.coordinates is None:
+            self.terms = self.generate(where)
+
+            # create the numexpr & the filter
+            if self.terms is not None:
+                self.condition, self.filter = self.terms.evaluate()
+
+    @overload
+    def generate(self, where: dict | list | tuple | str) -> PyTablesExpr: ...
+
+    @overload
+    def generate(self, where: None) -> None: ...
+
+    def generate(self, where: dict | list | tuple | str | None) -> PyTablesExpr | None:
+        """where can be a : dict,list,tuple,string"""
+        if where is None:
+            return None
+
+        q = self.table.queryables()
+        try:
+            return PyTablesExpr(where, queryables=q, encoding=self.table.encoding)
+        except NameError as err:
+            # raise a nice message, suggesting that the user should use
+            # data_columns
+            qkeys = ",".join(q.keys())
+            msg = dedent(
+                f"""\
+                The passed where expression: {where}
+                            contains an invalid variable reference
+                            all of the variable references must be a reference to
+                            an axis (e.g. 'index' or 'columns'), or a data_column
+                            The currently defined references are: {qkeys}
+                """
+            )
+            raise ValueError(msg) from err
+
+    def select(self):
+        """
+        generate the selection
+        """
+        if self.condition is not None:
+            return self.table.table.read_where(
+                self.condition.format(), start=self.start, stop=self.stop
+            )
+        elif self.coordinates is not None:
+            return self.table.table.read_coordinates(self.coordinates)
+        return self.table.table.read(start=self.start, stop=self.stop)
+
+    def select_coords(self):
+        """
+        generate the selection
+        """
+        start, stop = self.start, self.stop
+        nrows = self.table.nrows
+        if start is None:
+            start = 0
+        elif start < 0:
+            start += nrows
+        if stop is None:
+            stop = nrows
+        elif stop < 0:
+            stop += nrows
+
+        if self.condition is not None:
+            return self.table.table.get_where_list(
+                self.condition.format(), start=start, stop=stop, sort=True
+            )
+        elif self.coordinates is not None:
+            return self.coordinates
+
+        return np.arange(start, stop)
diff --git a/pandas/pandas/io/sas/__init__.py b/pandas/pandas/io/sas/__init__.py
new file mode 100644
index 0000000000000..317730745b6e3
--- /dev/null
+++ b/pandas/pandas/io/sas/__init__.py
@@ -0,0 +1,3 @@
+from pandas.io.sas.sasreader import read_sas
+
+__all__ = ["read_sas"]
diff --git a/pandas/pandas/io/sas/sas7bdat.py b/pandas/pandas/io/sas/sas7bdat.py
new file mode 100644
index 0000000000000..792af5ff713a3
--- /dev/null
+++ b/pandas/pandas/io/sas/sas7bdat.py
@@ -0,0 +1,739 @@
+"""
+Read SAS7BDAT files
+
+Based on code written by Jared Hobbs:
+  https://bitbucket.org/jaredhobbs/sas7bdat
+
+See also:
+  https://github.com/BioStatMatt/sas7bdat
+
+Partial documentation of the file format:
+  https://cran.r-project.org/package=sas7bdat/vignettes/sas7bdat.pdf
+
+Reference for binary data compression:
+  http://collaboration.cmc.ec.gc.ca/science/rpn/biblio/ddj/Website/articles/CUJ/1992/9210/ross/ross.htm
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+import sys
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._config import get_option
+
+from pandas._libs.byteswap import (
+    read_double_with_byteswap,
+    read_float_with_byteswap,
+    read_uint16_with_byteswap,
+    read_uint32_with_byteswap,
+    read_uint64_with_byteswap,
+)
+from pandas._libs.sas import (
+    Parser,
+    get_subheader_index,
+)
+from pandas._libs.tslibs.conversion import cast_from_unit_vectorized
+from pandas.errors import EmptyDataError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Timestamp,
+)
+
+from pandas.io.common import get_handle
+import pandas.io.sas.sas_constants as const
+from pandas.io.sas.sasreader import SASReader
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        ReadBuffer,
+    )
+
+
+_unix_origin = Timestamp("1970-01-01")
+_sas_origin = Timestamp("1960-01-01")
+
+
+def _convert_datetimes(sas_datetimes: pd.Series, unit: str) -> pd.Series:
+    """
+    Convert to Timestamp if possible, otherwise to datetime.datetime.
+    SAS float64 lacks precision for more than ms resolution so the fit
+    to datetime.datetime is ok.
+
+    Parameters
+    ----------
+    sas_datetimes : {Series, Sequence[float]}
+       Dates or datetimes in SAS
+    unit : {'d', 's'}
+       "d" if the floats represent dates, "s" for datetimes
+
+    Returns
+    -------
+    Series
+       Series of datetime64 dtype or datetime.datetime.
+    """
+    td = (_sas_origin - _unix_origin).as_unit("s")
+    if unit == "s":
+        millis = cast_from_unit_vectorized(
+            sas_datetimes._values, unit="s", out_unit="ms"
+        )
+        dt64ms = millis.view("M8[ms]") + td
+        return pd.Series(dt64ms, index=sas_datetimes.index, copy=False)
+    else:
+        vals = np.array(sas_datetimes, dtype="M8[D]") + td
+        return pd.Series(vals, dtype="M8[s]", index=sas_datetimes.index, copy=False)
+
+
+class _Column:
+    col_id: int
+    name: str | bytes
+    label: str | bytes
+    format: str | bytes
+    ctype: bytes
+    length: int
+
+    def __init__(
+        self,
+        col_id: int,
+        # These can be bytes when convert_header_text is False
+        name: str | bytes,
+        label: str | bytes,
+        format: str | bytes,
+        ctype: bytes,
+        length: int,
+    ) -> None:
+        self.col_id = col_id
+        self.name = name
+        self.label = label
+        self.format = format
+        self.ctype = ctype
+        self.length = length
+
+
+# SAS7BDAT represents a SAS data file in SAS7BDAT format.
+class SAS7BDATReader(SASReader):
+    """
+    Read SAS files in SAS7BDAT format.
+
+    Parameters
+    ----------
+    path_or_buf : path name or buffer
+        Name of SAS file or file-like object pointing to SAS file
+        contents.
+    index : column identifier, defaults to None
+        Column to use as index.
+    convert_dates : bool, defaults to True
+        Attempt to convert dates to Pandas datetime values.  Note that
+        some rarely used SAS date formats may be unsupported.
+    blank_missing : bool, defaults to True
+        Convert empty strings to missing values (SAS uses blanks to
+        indicate missing character variables).
+    chunksize : int, defaults to None
+        Return SAS7BDATReader object for iterations, returns chunks
+        with given number of lines.
+    encoding : str, 'infer', defaults to None
+        String encoding acc. to Python standard encodings,
+        encoding='infer' tries to detect the encoding from the file header,
+        encoding=None will leave the data in binary format.
+    convert_text : bool, defaults to True
+        If False, text variables are left as raw bytes.
+    convert_header_text : bool, defaults to True
+        If False, header text, including column names, are left as raw
+        bytes.
+    """
+
+    _int_length: int
+    _cached_page: bytes | None
+
+    def __init__(
+        self,
+        path_or_buf: FilePath | ReadBuffer[bytes],
+        index=None,
+        convert_dates: bool = True,
+        blank_missing: bool = True,
+        chunksize: int | None = None,
+        encoding: str | None = None,
+        convert_text: bool = True,
+        convert_header_text: bool = True,
+        compression: CompressionOptions = "infer",
+    ) -> None:
+        self.index = index
+        self.convert_dates = convert_dates
+        self.blank_missing = blank_missing
+        self.chunksize = chunksize
+        self.encoding = encoding
+        self.convert_text = convert_text
+        self.convert_header_text = convert_header_text
+
+        self.default_encoding = "latin-1"
+        self.compression = b""
+        self.column_names_raw: list[bytes] = []
+        self.column_names: list[str | bytes] = []
+        self.column_formats: list[str | bytes] = []
+        self.columns: list[_Column] = []
+
+        self._current_page_data_subheader_pointers: list[tuple[int, int]] = []
+        self._cached_page = None
+        self._column_data_lengths: list[int] = []
+        self._column_data_offsets: list[int] = []
+        self._column_types: list[bytes] = []
+
+        self._current_row_in_file_index = 0
+        self._current_row_on_page_index = 0
+        self._current_row_in_file_index = 0
+
+        self.handles = get_handle(
+            path_or_buf, "rb", is_text=False, compression=compression
+        )
+
+        self._path_or_buf = self.handles.handle
+
+        # Same order as const.SASIndex
+        self._subheader_processors = [
+            self._process_rowsize_subheader,
+            self._process_columnsize_subheader,
+            self._process_subheader_counts,
+            self._process_columntext_subheader,
+            self._process_columnname_subheader,
+            self._process_columnattributes_subheader,
+            self._process_format_subheader,
+            self._process_columnlist_subheader,
+            None,  # Data
+        ]
+
+        try:
+            self._get_properties()
+            self._parse_metadata()
+        except Exception:
+            self.close()
+            raise
+
+    def column_data_lengths(self) -> np.ndarray:
+        """Return a numpy int64 array of the column data lengths"""
+        return np.asarray(self._column_data_lengths, dtype=np.int64)
+
+    def column_data_offsets(self) -> np.ndarray:
+        """Return a numpy int64 array of the column offsets"""
+        return np.asarray(self._column_data_offsets, dtype=np.int64)
+
+    def column_types(self) -> np.ndarray:
+        """
+        Returns a numpy character array of the column types:
+           s (string) or d (double)
+        """
+        return np.asarray(self._column_types, dtype=np.dtype("S1"))
+
+    def close(self) -> None:
+        self.handles.close()
+
+    def _get_properties(self) -> None:
+        # Check magic number
+        self._path_or_buf.seek(0)
+        self._cached_page = self._path_or_buf.read(288)
+        if self._cached_page[0 : len(const.magic)] != const.magic:
+            raise ValueError("magic number mismatch (not a SAS file?)")
+
+        # Get alignment information
+        buf = self._read_bytes(const.align_1_offset, const.align_1_length)
+        if buf == const.u64_byte_checker_value:
+            self.U64 = True
+            self._int_length = 8
+            self._page_bit_offset = const.page_bit_offset_x64
+            self._subheader_pointer_length = const.subheader_pointer_length_x64
+        else:
+            self.U64 = False
+            self._page_bit_offset = const.page_bit_offset_x86
+            self._subheader_pointer_length = const.subheader_pointer_length_x86
+            self._int_length = 4
+        buf = self._read_bytes(const.align_2_offset, const.align_2_length)
+        if buf == const.align_1_checker_value:
+            align1 = const.align_2_value
+        else:
+            align1 = 0
+
+        # Get endianness information
+        buf = self._read_bytes(const.endianness_offset, const.endianness_length)
+        if buf == b"\x01":
+            self.byte_order = "<"
+            self.need_byteswap = sys.byteorder == "big"
+        else:
+            self.byte_order = ">"
+            self.need_byteswap = sys.byteorder == "little"
+
+        # Get encoding information
+        buf = self._read_bytes(const.encoding_offset, const.encoding_length)[0]
+        if buf in const.encoding_names:
+            self.inferred_encoding = const.encoding_names[buf]
+            if self.encoding == "infer":
+                self.encoding = self.inferred_encoding
+        else:
+            self.inferred_encoding = f"unknown (code={buf})"
+
+        # Timestamp is epoch 01/01/1960
+        epoch = datetime(1960, 1, 1)
+        x = self._read_float(
+            const.date_created_offset + align1, const.date_created_length
+        )
+        self.date_created = epoch + pd.to_timedelta(x, unit="s")
+        x = self._read_float(
+            const.date_modified_offset + align1, const.date_modified_length
+        )
+        self.date_modified = epoch + pd.to_timedelta(x, unit="s")
+
+        self.header_length = self._read_uint(
+            const.header_size_offset + align1, const.header_size_length
+        )
+
+        # Read the rest of the header into cached_page.
+        buf = self._path_or_buf.read(self.header_length - 288)
+        self._cached_page += buf
+        # error: Argument 1 to "len" has incompatible type "Optional[bytes]";
+        #  expected "Sized"
+        if len(self._cached_page) != self.header_length:  # type: ignore[arg-type]
+            raise ValueError("The SAS7BDAT file appears to be truncated.")
+
+        self._page_length = self._read_uint(
+            const.page_size_offset + align1, const.page_size_length
+        )
+
+    def __next__(self) -> DataFrame:
+        da = self.read(nrows=self.chunksize or 1)
+        if da.empty:
+            self.close()
+            raise StopIteration
+        return da
+
+    # Read a single float of the given width (4 or 8).
+    def _read_float(self, offset: int, width: int) -> float:
+        assert self._cached_page is not None
+        if width == 4:
+            return read_float_with_byteswap(
+                self._cached_page, offset, self.need_byteswap
+            )
+        elif width == 8:
+            return read_double_with_byteswap(
+                self._cached_page, offset, self.need_byteswap
+            )
+        else:
+            self.close()
+            raise ValueError("invalid float width")
+
+    # Read a single unsigned integer of the given width (1, 2, 4 or 8).
+    def _read_uint(self, offset: int, width: int) -> int:
+        assert self._cached_page is not None
+        if width == 1:
+            return self._read_bytes(offset, 1)[0]
+        elif width == 2:
+            return read_uint16_with_byteswap(
+                self._cached_page, offset, self.need_byteswap
+            )
+        elif width == 4:
+            return read_uint32_with_byteswap(
+                self._cached_page, offset, self.need_byteswap
+            )
+        elif width == 8:
+            return read_uint64_with_byteswap(
+                self._cached_page, offset, self.need_byteswap
+            )
+        else:
+            self.close()
+            raise ValueError("invalid int width")
+
+    def _read_bytes(self, offset: int, length: int):
+        assert self._cached_page is not None
+        if offset + length > len(self._cached_page):
+            self.close()
+            raise ValueError("The cached page is too small.")
+        return self._cached_page[offset : offset + length]
+
+    def _parse_metadata(self) -> None:
+        done = False
+        while not done:
+            self._cached_page = self._path_or_buf.read(self._page_length)
+            if len(self._cached_page) <= 0:
+                break
+            if len(self._cached_page) != self._page_length:
+                raise ValueError("Failed to read a meta data page from the SAS file.")
+            done = self._process_page_meta()
+
+    def _process_page_meta(self) -> bool:
+        self._read_page_header()
+        pt = const.page_meta_types + [const.page_amd_type, const.page_mix_type]
+        if self._current_page_type in pt:
+            self._process_page_metadata()
+        is_data_page = self._current_page_type == const.page_data_type
+        is_mix_page = self._current_page_type == const.page_mix_type
+        return bool(
+            is_data_page
+            or is_mix_page
+            or self._current_page_data_subheader_pointers != []
+        )
+
+    def _read_page_header(self) -> None:
+        bit_offset = self._page_bit_offset
+        tx = const.page_type_offset + bit_offset
+        self._current_page_type = (
+            self._read_uint(tx, const.page_type_length) & const.page_type_mask2
+        )
+        tx = const.block_count_offset + bit_offset
+        self._current_page_block_count = self._read_uint(tx, const.block_count_length)
+        tx = const.subheader_count_offset + bit_offset
+        self._current_page_subheaders_count = self._read_uint(
+            tx, const.subheader_count_length
+        )
+
+    def _process_page_metadata(self) -> None:
+        bit_offset = self._page_bit_offset
+
+        for i in range(self._current_page_subheaders_count):
+            offset = const.subheader_pointers_offset + bit_offset
+            total_offset = offset + self._subheader_pointer_length * i
+
+            subheader_offset = self._read_uint(total_offset, self._int_length)
+            total_offset += self._int_length
+
+            subheader_length = self._read_uint(total_offset, self._int_length)
+            total_offset += self._int_length
+
+            subheader_compression = self._read_uint(total_offset, 1)
+            total_offset += 1
+
+            subheader_type = self._read_uint(total_offset, 1)
+
+            if (
+                subheader_length == 0
+                or subheader_compression == const.truncated_subheader_id
+            ):
+                continue
+
+            subheader_signature = self._read_bytes(subheader_offset, self._int_length)
+            subheader_index = get_subheader_index(subheader_signature)
+            subheader_processor = self._subheader_processors[subheader_index]
+
+            if subheader_processor is None:
+                f1 = subheader_compression in (const.compressed_subheader_id, 0)
+                f2 = subheader_type == const.compressed_subheader_type
+                if self.compression and f1 and f2:
+                    self._current_page_data_subheader_pointers.append(
+                        (subheader_offset, subheader_length)
+                    )
+                else:
+                    self.close()
+                    raise ValueError(
+                        f"Unknown subheader signature {subheader_signature}"
+                    )
+            else:
+                subheader_processor(subheader_offset, subheader_length)
+
+    def _process_rowsize_subheader(self, offset: int, length: int) -> None:
+        int_len = self._int_length
+        lcs_offset = offset
+        lcp_offset = offset
+        if self.U64:
+            lcs_offset += 682
+            lcp_offset += 706
+        else:
+            lcs_offset += 354
+            lcp_offset += 378
+
+        self.row_length = self._read_uint(
+            offset + const.row_length_offset_multiplier * int_len,
+            int_len,
+        )
+        self.row_count = self._read_uint(
+            offset + const.row_count_offset_multiplier * int_len,
+            int_len,
+        )
+        self.col_count_p1 = self._read_uint(
+            offset + const.col_count_p1_multiplier * int_len, int_len
+        )
+        self.col_count_p2 = self._read_uint(
+            offset + const.col_count_p2_multiplier * int_len, int_len
+        )
+        mx = const.row_count_on_mix_page_offset_multiplier * int_len
+        self._mix_page_row_count = self._read_uint(offset + mx, int_len)
+        self._lcs = self._read_uint(lcs_offset, 2)
+        self._lcp = self._read_uint(lcp_offset, 2)
+
+    def _process_columnsize_subheader(self, offset: int, length: int) -> None:
+        int_len = self._int_length
+        offset += int_len
+        self.column_count = self._read_uint(offset, int_len)
+        if self.col_count_p1 + self.col_count_p2 != self.column_count:
+            print(
+                f"Warning: column count mismatch ({self.col_count_p1} + "
+                f"{self.col_count_p2} != {self.column_count})\n"
+            )
+
+    # Unknown purpose
+    def _process_subheader_counts(self, offset: int, length: int) -> None:
+        pass
+
+    def _process_columntext_subheader(self, offset: int, length: int) -> None:
+        offset += self._int_length
+        text_block_size = self._read_uint(offset, const.text_block_size_length)
+
+        buf = self._read_bytes(offset, text_block_size)
+        cname_raw = buf[0:text_block_size].rstrip(b"\x00 ")
+        self.column_names_raw.append(cname_raw)
+
+        if len(self.column_names_raw) == 1:
+            compression_literal = b""
+            for cl in const.compression_literals:
+                if cl in cname_raw:
+                    compression_literal = cl
+            self.compression = compression_literal
+            offset -= self._int_length
+
+            offset1 = offset + 16
+            if self.U64:
+                offset1 += 4
+
+            buf = self._read_bytes(offset1, self._lcp)
+            compression_literal = buf.rstrip(b"\x00")
+            if compression_literal == b"":
+                self._lcs = 0
+                offset1 = offset + 32
+                if self.U64:
+                    offset1 += 4
+                buf = self._read_bytes(offset1, self._lcp)
+                self.creator_proc = buf[0 : self._lcp]
+            elif compression_literal == const.rle_compression:
+                offset1 = offset + 40
+                if self.U64:
+                    offset1 += 4
+                buf = self._read_bytes(offset1, self._lcp)
+                self.creator_proc = buf[0 : self._lcp]
+            elif self._lcs > 0:
+                self._lcp = 0
+                offset1 = offset + 16
+                if self.U64:
+                    offset1 += 4
+                buf = self._read_bytes(offset1, self._lcs)
+                self.creator_proc = buf[0 : self._lcp]
+            if hasattr(self, "creator_proc"):
+                self.creator_proc = self._convert_header_text(self.creator_proc)  # pyright: ignore[reportArgumentType]
+
+    def _process_columnname_subheader(self, offset: int, length: int) -> None:
+        int_len = self._int_length
+        offset += int_len
+        column_name_pointers_count = (length - 2 * int_len - 12) // 8
+        for i in range(column_name_pointers_count):
+            text_subheader = (
+                offset
+                + const.column_name_pointer_length * (i + 1)
+                + const.column_name_text_subheader_offset
+            )
+            col_name_offset = (
+                offset
+                + const.column_name_pointer_length * (i + 1)
+                + const.column_name_offset_offset
+            )
+            col_name_length = (
+                offset
+                + const.column_name_pointer_length * (i + 1)
+                + const.column_name_length_offset
+            )
+
+            idx = self._read_uint(
+                text_subheader, const.column_name_text_subheader_length
+            )
+            col_offset = self._read_uint(
+                col_name_offset, const.column_name_offset_length
+            )
+            col_len = self._read_uint(col_name_length, const.column_name_length_length)
+
+            name_raw = self.column_names_raw[idx]
+            cname = name_raw[col_offset : col_offset + col_len]
+            self.column_names.append(self._convert_header_text(cname))
+
+    def _process_columnattributes_subheader(self, offset: int, length: int) -> None:
+        int_len = self._int_length
+        column_attributes_vectors_count = (length - 2 * int_len - 12) // (int_len + 8)
+        for i in range(column_attributes_vectors_count):
+            col_data_offset = (
+                offset + int_len + const.column_data_offset_offset + i * (int_len + 8)
+            )
+            col_data_len = (
+                offset
+                + 2 * int_len
+                + const.column_data_length_offset
+                + i * (int_len + 8)
+            )
+            col_types = (
+                offset + 2 * int_len + const.column_type_offset + i * (int_len + 8)
+            )
+
+            x = self._read_uint(col_data_offset, int_len)
+            self._column_data_offsets.append(x)
+
+            x = self._read_uint(col_data_len, const.column_data_length_length)
+            self._column_data_lengths.append(x)
+
+            x = self._read_uint(col_types, const.column_type_length)
+            self._column_types.append(b"d" if x == 1 else b"s")
+
+    def _process_columnlist_subheader(self, offset: int, length: int) -> None:
+        # unknown purpose
+        pass
+
+    def _process_format_subheader(self, offset: int, length: int) -> None:
+        int_len = self._int_length
+        text_subheader_format = (
+            offset + const.column_format_text_subheader_index_offset + 3 * int_len
+        )
+        col_format_offset = offset + const.column_format_offset_offset + 3 * int_len
+        col_format_len = offset + const.column_format_length_offset + 3 * int_len
+        text_subheader_label = (
+            offset + const.column_label_text_subheader_index_offset + 3 * int_len
+        )
+        col_label_offset = offset + const.column_label_offset_offset + 3 * int_len
+        col_label_len = offset + const.column_label_length_offset + 3 * int_len
+
+        x = self._read_uint(
+            text_subheader_format, const.column_format_text_subheader_index_length
+        )
+        format_idx = min(x, len(self.column_names_raw) - 1)
+
+        format_start = self._read_uint(
+            col_format_offset, const.column_format_offset_length
+        )
+        format_len = self._read_uint(col_format_len, const.column_format_length_length)
+
+        label_idx = self._read_uint(
+            text_subheader_label, const.column_label_text_subheader_index_length
+        )
+        label_idx = min(label_idx, len(self.column_names_raw) - 1)
+
+        label_start = self._read_uint(
+            col_label_offset, const.column_label_offset_length
+        )
+        label_len = self._read_uint(col_label_len, const.column_label_length_length)
+
+        label_names = self.column_names_raw[label_idx]
+        column_label = self._convert_header_text(
+            label_names[label_start : label_start + label_len]
+        )
+        format_names = self.column_names_raw[format_idx]
+        column_format = self._convert_header_text(
+            format_names[format_start : format_start + format_len]
+        )
+        current_column_number = len(self.columns)
+
+        col = _Column(
+            current_column_number,
+            self.column_names[current_column_number],
+            column_label,
+            column_format,
+            self._column_types[current_column_number],
+            self._column_data_lengths[current_column_number],
+        )
+
+        self.column_formats.append(column_format)
+        self.columns.append(col)
+
+    def read(self, nrows: int | None = None) -> DataFrame:
+        if (nrows is None) and (self.chunksize is not None):
+            nrows = self.chunksize
+        elif nrows is None:
+            nrows = self.row_count
+
+        if len(self._column_types) == 0:
+            self.close()
+            raise EmptyDataError("No columns to parse from file")
+
+        if nrows > 0 and self._current_row_in_file_index >= self.row_count:
+            return DataFrame()
+
+        nrows = min(nrows, self.row_count - self._current_row_in_file_index)
+
+        nd = self._column_types.count(b"d")
+        ns = self._column_types.count(b"s")
+
+        self._string_chunk = np.empty((ns, nrows), dtype=object)
+        self._byte_chunk = np.zeros((nd, 8 * nrows), dtype=np.uint8)
+
+        self._current_row_in_chunk_index = 0
+        p = Parser(self)
+        p.read(nrows)
+
+        rslt = self._chunk_to_dataframe()
+        if self.index is not None:
+            rslt = rslt.set_index(self.index)
+
+        return rslt
+
+    def _read_next_page(self):
+        self._current_page_data_subheader_pointers = []
+        self._cached_page = self._path_or_buf.read(self._page_length)
+        if len(self._cached_page) <= 0:
+            return True
+        elif len(self._cached_page) != self._page_length:
+            self.close()
+            msg = (
+                "failed to read complete page from file (read "
+                f"{len(self._cached_page):d} of {self._page_length:d} bytes)"
+            )
+            raise ValueError(msg)
+
+        self._read_page_header()
+        if self._current_page_type in const.page_meta_types:
+            self._process_page_metadata()
+
+        if self._current_page_type not in const.page_meta_types + [
+            const.page_data_type,
+            const.page_mix_type,
+        ]:
+            return self._read_next_page()
+
+        return False
+
+    def _chunk_to_dataframe(self) -> DataFrame:
+        n = self._current_row_in_chunk_index
+        m = self._current_row_in_file_index
+        ix = range(m - n, m)
+        rslt = {}
+
+        js, jb = 0, 0
+        infer_string = get_option("future.infer_string")
+        for j in range(self.column_count):
+            name = self.column_names[j]
+
+            if self._column_types[j] == b"d":
+                col_arr = self._byte_chunk[jb, :].view(dtype=self.byte_order + "d")
+                rslt[name] = pd.Series(col_arr, dtype=np.float64, index=ix, copy=False)
+                if self.convert_dates:
+                    if self.column_formats[j] in const.sas_date_formats:
+                        rslt[name] = _convert_datetimes(rslt[name], "d")
+                    elif self.column_formats[j] in const.sas_datetime_formats:
+                        rslt[name] = _convert_datetimes(rslt[name], "s")
+                jb += 1
+            elif self._column_types[j] == b"s":
+                rslt[name] = pd.Series(self._string_chunk[js, :], index=ix, copy=False)
+                if self.convert_text and (self.encoding is not None):
+                    rslt[name] = self._decode_string(rslt[name].str)
+                    if infer_string:
+                        rslt[name] = rslt[name].astype("str")
+
+                js += 1
+            else:
+                self.close()
+                raise ValueError(f"unknown column type {self._column_types[j]!r}")
+
+        df = DataFrame(rslt, columns=self.column_names, index=ix, copy=False)
+        return df
+
+    def _decode_string(self, b):
+        return b.decode(self.encoding or self.default_encoding)
+
+    def _convert_header_text(self, b: bytes) -> str | bytes:
+        if self.convert_header_text:
+            return self._decode_string(b)
+        else:
+            return b
diff --git a/pandas/pandas/io/sas/sas_constants.py b/pandas/pandas/io/sas/sas_constants.py
new file mode 100644
index 0000000000000..8da7becd76d3b
--- /dev/null
+++ b/pandas/pandas/io/sas/sas_constants.py
@@ -0,0 +1,310 @@
+from __future__ import annotations
+
+from typing import Final
+
+magic: Final = (
+    b"\x00\x00\x00\x00\x00\x00\x00\x00"
+    b"\x00\x00\x00\x00\xc2\xea\x81\x60"
+    b"\xb3\x14\x11\xcf\xbd\x92\x08\x00"
+    b"\x09\xc7\x31\x8c\x18\x1f\x10\x11"
+)
+
+align_1_checker_value: Final = b"3"
+align_1_offset: Final = 32
+align_1_length: Final = 1
+align_1_value: Final = 4
+u64_byte_checker_value: Final = b"3"
+align_2_offset: Final = 35
+align_2_length: Final = 1
+align_2_value: Final = 4
+endianness_offset: Final = 37
+endianness_length: Final = 1
+platform_offset: Final = 39
+platform_length: Final = 1
+encoding_offset: Final = 70
+encoding_length: Final = 1
+dataset_offset: Final = 92
+dataset_length: Final = 64
+file_type_offset: Final = 156
+file_type_length: Final = 8
+date_created_offset: Final = 164
+date_created_length: Final = 8
+date_modified_offset: Final = 172
+date_modified_length: Final = 8
+header_size_offset: Final = 196
+header_size_length: Final = 4
+page_size_offset: Final = 200
+page_size_length: Final = 4
+page_count_offset: Final = 204
+page_count_length: Final = 4
+sas_release_offset: Final = 216
+sas_release_length: Final = 8
+sas_server_type_offset: Final = 224
+sas_server_type_length: Final = 16
+os_version_number_offset: Final = 240
+os_version_number_length: Final = 16
+os_maker_offset: Final = 256
+os_maker_length: Final = 16
+os_name_offset: Final = 272
+os_name_length: Final = 16
+page_bit_offset_x86: Final = 16
+page_bit_offset_x64: Final = 32
+subheader_pointer_length_x86: Final = 12
+subheader_pointer_length_x64: Final = 24
+page_type_offset: Final = 0
+page_type_length: Final = 2
+block_count_offset: Final = 2
+block_count_length: Final = 2
+subheader_count_offset: Final = 4
+subheader_count_length: Final = 2
+page_type_mask: Final = 0x0F00
+# Keep "page_comp_type" bits
+page_type_mask2: Final = 0xF000 | page_type_mask
+page_meta_type: Final = 0x0000
+page_data_type: Final = 0x0100
+page_mix_type: Final = 0x0200
+page_amd_type: Final = 0x0400
+page_meta2_type: Final = 0x4000
+page_comp_type: Final = 0x9000
+page_meta_types: Final = [page_meta_type, page_meta2_type]
+subheader_pointers_offset: Final = 8
+truncated_subheader_id: Final = 1
+compressed_subheader_id: Final = 4
+compressed_subheader_type: Final = 1
+text_block_size_length: Final = 2
+row_length_offset_multiplier: Final = 5
+row_count_offset_multiplier: Final = 6
+col_count_p1_multiplier: Final = 9
+col_count_p2_multiplier: Final = 10
+row_count_on_mix_page_offset_multiplier: Final = 15
+column_name_pointer_length: Final = 8
+column_name_text_subheader_offset: Final = 0
+column_name_text_subheader_length: Final = 2
+column_name_offset_offset: Final = 2
+column_name_offset_length: Final = 2
+column_name_length_offset: Final = 4
+column_name_length_length: Final = 2
+column_data_offset_offset: Final = 8
+column_data_length_offset: Final = 8
+column_data_length_length: Final = 4
+column_type_offset: Final = 14
+column_type_length: Final = 1
+column_format_text_subheader_index_offset: Final = 22
+column_format_text_subheader_index_length: Final = 2
+column_format_offset_offset: Final = 24
+column_format_offset_length: Final = 2
+column_format_length_offset: Final = 26
+column_format_length_length: Final = 2
+column_label_text_subheader_index_offset: Final = 28
+column_label_text_subheader_index_length: Final = 2
+column_label_offset_offset: Final = 30
+column_label_offset_length: Final = 2
+column_label_length_offset: Final = 32
+column_label_length_length: Final = 2
+rle_compression: Final = b"SASYZCRL"
+rdc_compression: Final = b"SASYZCR2"
+
+compression_literals: Final = [rle_compression, rdc_compression]
+
+# Incomplete list of encodings, using SAS nomenclature:
+# https://support.sas.com/documentation/onlinedoc/dfdmstudio/2.6/dmpdmsug/Content/dfU_Encodings_SAS.html
+# corresponding to the Python documentation of standard encodings
+# https://docs.python.org/3/library/codecs.html#standard-encodings
+encoding_names: Final = {
+    20: "utf-8",
+    29: "latin1",
+    30: "latin2",
+    31: "latin3",
+    32: "latin4",
+    33: "cyrillic",
+    34: "arabic",
+    35: "greek",
+    36: "hebrew",
+    37: "latin5",
+    38: "latin6",
+    39: "cp874",
+    40: "latin9",
+    41: "cp437",
+    42: "cp850",
+    43: "cp852",
+    44: "cp857",
+    45: "cp858",
+    46: "cp862",
+    47: "cp864",
+    48: "cp865",
+    49: "cp866",
+    50: "cp869",
+    51: "cp874",
+    # 52: "",  # not found
+    # 53: "",  # not found
+    # 54: "",  # not found
+    55: "cp720",
+    56: "cp737",
+    57: "cp775",
+    58: "cp860",
+    59: "cp863",
+    60: "cp1250",
+    61: "cp1251",
+    62: "cp1252",
+    63: "cp1253",
+    64: "cp1254",
+    65: "cp1255",
+    66: "cp1256",
+    67: "cp1257",
+    68: "cp1258",
+    118: "cp950",
+    # 119: "",  # not found
+    123: "big5",
+    125: "gb2312",
+    126: "cp936",
+    134: "euc_jp",
+    136: "cp932",
+    138: "shift_jis",
+    140: "euc-kr",
+    141: "cp949",
+    227: "latin8",
+    # 228: "", # not found
+    # 229: ""  # not found
+}
+
+
+class SASIndex:
+    row_size_index: Final = 0
+    column_size_index: Final = 1
+    subheader_counts_index: Final = 2
+    column_text_index: Final = 3
+    column_name_index: Final = 4
+    column_attributes_index: Final = 5
+    format_and_label_index: Final = 6
+    column_list_index: Final = 7
+    data_subheader_index: Final = 8
+
+
+subheader_signature_to_index: Final = {
+    b"\xf7\xf7\xf7\xf7": SASIndex.row_size_index,
+    b"\x00\x00\x00\x00\xf7\xf7\xf7\xf7": SASIndex.row_size_index,
+    b"\xf7\xf7\xf7\xf7\x00\x00\x00\x00": SASIndex.row_size_index,
+    b"\xf7\xf7\xf7\xf7\xff\xff\xfb\xfe": SASIndex.row_size_index,
+    b"\xf6\xf6\xf6\xf6": SASIndex.column_size_index,
+    b"\x00\x00\x00\x00\xf6\xf6\xf6\xf6": SASIndex.column_size_index,
+    b"\xf6\xf6\xf6\xf6\x00\x00\x00\x00": SASIndex.column_size_index,
+    b"\xf6\xf6\xf6\xf6\xff\xff\xfb\xfe": SASIndex.column_size_index,
+    b"\x00\xfc\xff\xff": SASIndex.subheader_counts_index,
+    b"\xff\xff\xfc\x00": SASIndex.subheader_counts_index,
+    b"\x00\xfc\xff\xff\xff\xff\xff\xff": SASIndex.subheader_counts_index,
+    b"\xff\xff\xff\xff\xff\xff\xfc\x00": SASIndex.subheader_counts_index,
+    b"\xfd\xff\xff\xff": SASIndex.column_text_index,
+    b"\xff\xff\xff\xfd": SASIndex.column_text_index,
+    b"\xfd\xff\xff\xff\xff\xff\xff\xff": SASIndex.column_text_index,
+    b"\xff\xff\xff\xff\xff\xff\xff\xfd": SASIndex.column_text_index,
+    b"\xff\xff\xff\xff": SASIndex.column_name_index,
+    b"\xff\xff\xff\xff\xff\xff\xff\xff": SASIndex.column_name_index,
+    b"\xfc\xff\xff\xff": SASIndex.column_attributes_index,
+    b"\xff\xff\xff\xfc": SASIndex.column_attributes_index,
+    b"\xfc\xff\xff\xff\xff\xff\xff\xff": SASIndex.column_attributes_index,
+    b"\xff\xff\xff\xff\xff\xff\xff\xfc": SASIndex.column_attributes_index,
+    b"\xfe\xfb\xff\xff": SASIndex.format_and_label_index,
+    b"\xff\xff\xfb\xfe": SASIndex.format_and_label_index,
+    b"\xfe\xfb\xff\xff\xff\xff\xff\xff": SASIndex.format_and_label_index,
+    b"\xff\xff\xff\xff\xff\xff\xfb\xfe": SASIndex.format_and_label_index,
+    b"\xfe\xff\xff\xff": SASIndex.column_list_index,
+    b"\xff\xff\xff\xfe": SASIndex.column_list_index,
+    b"\xfe\xff\xff\xff\xff\xff\xff\xff": SASIndex.column_list_index,
+    b"\xff\xff\xff\xff\xff\xff\xff\xfe": SASIndex.column_list_index,
+}
+
+
+# List of frequently used SAS date and datetime formats
+# http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_intervals_sect009.htm
+# https://github.com/epam/parso/blob/master/src/main/java/com/epam/parso/impl/SasFileConstants.java
+sas_date_formats: Final = (
+    "DATE",
+    "DAY",
+    "DDMMYY",
+    "DOWNAME",
+    "JULDAY",
+    "JULIAN",
+    "MMDDYY",
+    "MMYY",
+    "MMYYC",
+    "MMYYD",
+    "MMYYP",
+    "MMYYS",
+    "MMYYN",
+    "MONNAME",
+    "MONTH",
+    "MONYY",
+    "QTR",
+    "QTRR",
+    "NENGO",
+    "WEEKDATE",
+    "WEEKDATX",
+    "WEEKDAY",
+    "WEEKV",
+    "WORDDATE",
+    "WORDDATX",
+    "YEAR",
+    "YYMM",
+    "YYMMC",
+    "YYMMD",
+    "YYMMP",
+    "YYMMS",
+    "YYMMN",
+    "YYMON",
+    "YYMMDD",
+    "YYQ",
+    "YYQC",
+    "YYQD",
+    "YYQP",
+    "YYQS",
+    "YYQN",
+    "YYQR",
+    "YYQRC",
+    "YYQRD",
+    "YYQRP",
+    "YYQRS",
+    "YYQRN",
+    "YYMMDDP",
+    "YYMMDDC",
+    "E8601DA",
+    "YYMMDDN",
+    "MMDDYYC",
+    "MMDDYYS",
+    "MMDDYYD",
+    "YYMMDDS",
+    "B8601DA",
+    "DDMMYYN",
+    "YYMMDDD",
+    "DDMMYYB",
+    "DDMMYYP",
+    "MMDDYYP",
+    "YYMMDDB",
+    "MMDDYYN",
+    "DDMMYYC",
+    "DDMMYYD",
+    "DDMMYYS",
+    "MINGUO",
+)
+
+sas_datetime_formats: Final = (
+    "DATETIME",
+    "DTWKDATX",
+    "B8601DN",
+    "B8601DT",
+    "B8601DX",
+    "B8601DZ",
+    "B8601LX",
+    "E8601DN",
+    "E8601DT",
+    "E8601DX",
+    "E8601DZ",
+    "E8601LX",
+    "DATEAMPM",
+    "DTDATE",
+    "DTMONYY",
+    "DTMONYY",
+    "DTWKDATX",
+    "DTYEAR",
+    "TOD",
+    "MDYAMPM",
+)
diff --git a/pandas/pandas/io/sas/sas_xport.py b/pandas/pandas/io/sas/sas_xport.py
new file mode 100644
index 0000000000000..a9c45e720fd56
--- /dev/null
+++ b/pandas/pandas/io/sas/sas_xport.py
@@ -0,0 +1,505 @@
+"""
+Read a SAS XPort format file into a Pandas DataFrame.
+
+Based on code from Jack Cushman (github.com/jcushman/xport).
+
+The file format is defined here:
+
+https://support.sas.com/content/dam/SAS/support/en/technical-papers/record-layout-of-a-sas-version-5-or-6-data-set-in-sas-transport-xport-format.pdf
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+import struct
+from typing import TYPE_CHECKING
+import warnings
+
+import numpy as np
+
+from pandas.util._decorators import Appender
+from pandas.util._exceptions import find_stack_level
+
+import pandas as pd
+
+from pandas.io.common import get_handle
+from pandas.io.sas.sasreader import SASReader
+
+if TYPE_CHECKING:
+    from pandas._typing import (
+        CompressionOptions,
+        DatetimeNaTType,
+        FilePath,
+        ReadBuffer,
+    )
+_correct_line1 = (
+    "HEADER RECORD*******LIBRARY HEADER RECORD!!!!!!!000000000000000000000000000000  "
+)
+_correct_header1 = (
+    "HEADER RECORD*******MEMBER  HEADER RECORD!!!!!!!000000000000000001600000000"
+)
+_correct_header2 = (
+    "HEADER RECORD*******DSCRPTR HEADER RECORD!!!!!!!000000000000000000000000000000  "
+)
+_correct_obs_header = (
+    "HEADER RECORD*******OBS     HEADER RECORD!!!!!!!000000000000000000000000000000  "
+)
+_fieldkeys = [
+    "ntype",
+    "nhfun",
+    "field_length",
+    "nvar0",
+    "name",
+    "label",
+    "nform",
+    "nfl",
+    "num_decimals",
+    "nfj",
+    "nfill",
+    "niform",
+    "nifl",
+    "nifd",
+    "npos",
+    "_",
+]
+
+
+_base_params_doc = """\
+Parameters
+----------
+filepath_or_buffer : str or file-like object
+    Path to SAS file or object implementing binary read method."""
+
+_params2_doc = """\
+index : identifier of index column
+    Identifier of column that should be used as index of the DataFrame.
+encoding : str
+    Encoding for text data.
+chunksize : int
+    Read file `chunksize` lines at a time, returns iterator."""
+
+_format_params_doc = """\
+format : str
+    File format, only `xport` is currently supported."""
+
+_iterator_doc = """\
+iterator : bool, default False
+    Return XportReader object for reading file incrementally."""
+
+
+_read_sas_doc = f"""Read a SAS file into a DataFrame.
+
+{_base_params_doc}
+{_format_params_doc}
+{_params2_doc}
+{_iterator_doc}
+
+Returns
+-------
+DataFrame or XportReader
+
+Examples
+--------
+Read a SAS Xport file:
+
+>>> df = pd.read_sas('filename.XPT')
+
+Read a Xport file in 10,000 line chunks:
+
+>>> itr = pd.read_sas('filename.XPT', chunksize=10000)
+>>> for chunk in itr:
+>>>     do_something(chunk)
+
+"""
+
+_xport_reader_doc = f"""\
+Class for reading SAS Xport files.
+
+{_base_params_doc}
+{_params2_doc}
+
+Attributes
+----------
+member_info : list
+    Contains information about the file
+fields : list
+    Contains information about the variables in the file
+"""
+
+_read_method_doc = """\
+Read observations from SAS Xport file, returning as data frame.
+
+Parameters
+----------
+nrows : int
+    Number of rows to read from data file; if None, read whole
+    file.
+
+Returns
+-------
+A DataFrame.
+"""
+
+
+def _parse_date(datestr: str) -> DatetimeNaTType:
+    """Given a date in xport format, return Python date."""
+    try:
+        # e.g. "16FEB11:10:07:55"
+        return datetime.strptime(datestr, "%d%b%y:%H:%M:%S")
+    except ValueError:
+        return pd.NaT
+
+
+def _split_line(s: str, parts):
+    """
+    Parameters
+    ----------
+    s: str
+        Fixed-length string to split
+    parts: list of (name, length) pairs
+        Used to break up string, name '_' will be filtered from output.
+
+    Returns
+    -------
+    Dict of name:contents of string at given location.
+    """
+    out = {}
+    start = 0
+    for name, length in parts:
+        out[name] = s[start : start + length].strip()
+        start += length
+    del out["_"]
+    return out
+
+
+def _handle_truncated_float_vec(vec, nbytes):
+    # This feature is not well documented, but some SAS XPORT files
+    # have 2-7 byte "truncated" floats.  To read these truncated
+    # floats, pad them with zeros on the right to make 8 byte floats.
+    #
+    # References:
+    # https://github.com/jcushman/xport/pull/3
+    # The R "foreign" library
+
+    if nbytes != 8:
+        vec1 = np.zeros(len(vec), np.dtype("S8"))
+        dtype = np.dtype(f"S{nbytes},S{8 - nbytes}")
+        vec2 = vec1.view(dtype=dtype)
+        vec2["f0"] = vec
+        return vec2
+
+    return vec
+
+
+def _parse_float_vec(vec):
+    """
+    Parse a vector of float values representing IBM 8 byte floats into
+    native 8 byte floats.
+    """
+    dtype = np.dtype(">u4,>u4")
+    vec1 = vec.view(dtype=dtype)
+    xport1 = vec1["f0"]
+    xport2 = vec1["f1"]
+
+    # Start by setting first half of ieee number to first half of IBM
+    # number sans exponent
+    ieee1 = xport1 & 0x00FFFFFF
+
+    # The fraction bit to the left of the binary point in the ieee
+    # format was set and the number was shifted 0, 1, 2, or 3
+    # places. This will tell us how to adjust the ibm exponent to be a
+    # power of 2 ieee exponent and how to shift the fraction bits to
+    # restore the correct magnitude.
+    shift = np.zeros(len(vec), dtype=np.uint8)
+    shift[np.where(xport1 & 0x00200000)] = 1
+    shift[np.where(xport1 & 0x00400000)] = 2
+    shift[np.where(xport1 & 0x00800000)] = 3
+
+    # shift the ieee number down the correct number of places then
+    # set the second half of the ieee number to be the second half
+    # of the ibm number shifted appropriately, ored with the bits
+    # from the first half that would have been shifted in if we
+    # could shift a double. All we are worried about are the low
+    # order 3 bits of the first half since we're only shifting by
+    # 1, 2, or 3.
+    ieee1 >>= shift
+    ieee2 = (xport2 >> shift) | ((xport1 & 0x00000007) << (29 + (3 - shift)))
+
+    # clear the 1 bit to the left of the binary point
+    ieee1 &= 0xFFEFFFFF
+
+    # set the exponent of the ieee number to be the actual exponent
+    # plus the shift count + 1023. Or this into the first half of the
+    # ieee number. The ibm exponent is excess 64 but is adjusted by 65
+    # since during conversion to ibm format the exponent is
+    # incremented by 1 and the fraction bits left 4 positions to the
+    # right of the radix point.  (had to add >> 24 because C treats &
+    # 0x7f as 0x7f000000 and Python doesn't)
+    ieee1 |= ((((((xport1 >> 24) & 0x7F) - 65) << 2) + shift + 1023) << 20) | (
+        xport1 & 0x80000000
+    )
+
+    ieee = np.empty((len(ieee1),), dtype=">u4,>u4")
+    ieee["f0"] = ieee1
+    ieee["f1"] = ieee2
+    ieee = ieee.view(dtype=">f8")
+    ieee = ieee.astype("f8")
+
+    return ieee
+
+
+class XportReader(SASReader):
+    __doc__ = _xport_reader_doc
+
+    def __init__(
+        self,
+        filepath_or_buffer: FilePath | ReadBuffer[bytes],
+        index=None,
+        encoding: str | None = "ISO-8859-1",
+        chunksize: int | None = None,
+        compression: CompressionOptions = "infer",
+    ) -> None:
+        self._encoding = encoding
+        self._lines_read = 0
+        self._index = index
+        self._chunksize = chunksize
+
+        self.handles = get_handle(
+            filepath_or_buffer,
+            "rb",
+            encoding=encoding,
+            is_text=False,
+            compression=compression,
+        )
+        self.filepath_or_buffer = self.handles.handle
+
+        try:
+            self._read_header()
+        except Exception:
+            self.close()
+            raise
+
+    def close(self) -> None:
+        self.handles.close()
+
+    def _get_row(self):
+        return self.filepath_or_buffer.read(80).decode()
+
+    def _read_header(self) -> None:
+        self.filepath_or_buffer.seek(0)
+
+        # read file header
+        line1 = self._get_row()
+        if line1 != _correct_line1:
+            if "**COMPRESSED**" in line1:
+                # this was created with the PROC CPORT method and can't be read
+                # https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/movefile/p1bm6aqp3fw4uin1hucwh718f6kp.htm
+                raise ValueError(
+                    "Header record indicates a CPORT file, which is not readable."
+                )
+            raise ValueError("Header record is not an XPORT file.")
+
+        line2 = self._get_row()
+        fif = [["prefix", 24], ["version", 8], ["OS", 8], ["_", 24], ["created", 16]]
+        file_info = _split_line(line2, fif)
+        if file_info["prefix"] != "SAS     SAS     SASLIB":
+            raise ValueError("Header record has invalid prefix.")
+        file_info["created"] = _parse_date(file_info["created"])
+        self.file_info = file_info
+
+        line3 = self._get_row()
+        file_info["modified"] = _parse_date(line3[:16])
+
+        # read member header
+        header1 = self._get_row()
+        header2 = self._get_row()
+        headflag1 = header1.startswith(_correct_header1)
+        headflag2 = header2 == _correct_header2
+        if not (headflag1 and headflag2):
+            raise ValueError("Member header not found")
+        # usually 140, could be 135
+        fieldnamelength = int(header1[-5:-2])
+
+        # member info
+        mem = [
+            ["prefix", 8],
+            ["set_name", 8],
+            ["sasdata", 8],
+            ["version", 8],
+            ["OS", 8],
+            ["_", 24],
+            ["created", 16],
+        ]
+        member_info = _split_line(self._get_row(), mem)
+        mem = [["modified", 16], ["_", 16], ["label", 40], ["type", 8]]
+        member_info.update(_split_line(self._get_row(), mem))
+        member_info["modified"] = _parse_date(member_info["modified"])
+        member_info["created"] = _parse_date(member_info["created"])
+        self.member_info = member_info
+
+        # read field names
+        types = {1: "numeric", 2: "char"}
+        fieldcount = int(self._get_row()[54:58])
+        datalength = fieldnamelength * fieldcount
+        # round up to nearest 80
+        if datalength % 80:
+            datalength += 80 - datalength % 80
+        fielddata = self.filepath_or_buffer.read(datalength)
+        fields = []
+        obs_length = 0
+        while len(fielddata) >= fieldnamelength:
+            # pull data for one field
+            fieldbytes, fielddata = (
+                fielddata[:fieldnamelength],
+                fielddata[fieldnamelength:],
+            )
+
+            # rest at end gets ignored, so if field is short, pad out
+            # to match struct pattern below
+            fieldbytes = fieldbytes.ljust(140)
+
+            fieldstruct = struct.unpack(">hhhh8s40s8shhh2s8shhl52s", fieldbytes)
+            field = dict(zip(_fieldkeys, fieldstruct))
+            del field["_"]
+            field["ntype"] = types[field["ntype"]]
+            fl = field["field_length"]
+            if field["ntype"] == "numeric" and ((fl < 2) or (fl > 8)):
+                msg = f"Floating field width {fl} is not between 2 and 8."
+                raise TypeError(msg)
+
+            for k, v in field.items():
+                try:
+                    field[k] = v.strip()
+                except AttributeError:
+                    pass
+
+            obs_length += field["field_length"]
+            fields += [field]
+
+        header = self._get_row()
+        if not header == _correct_obs_header:
+            raise ValueError("Observation header not found.")
+
+        self.fields = fields
+        self.record_length = obs_length
+        self.record_start = self.filepath_or_buffer.tell()
+
+        self.nobs = self._record_count()
+        self.columns = [x["name"].decode() for x in self.fields]
+
+        # Setup the dtype.
+        dtypel = [
+            ("s" + str(i), "S" + str(field["field_length"]))
+            for i, field in enumerate(self.fields)
+        ]
+        dtype = np.dtype(dtypel)
+        self._dtype = dtype
+
+    def __next__(self) -> pd.DataFrame:
+        return self.read(nrows=self._chunksize or 1)
+
+    def _record_count(self) -> int:
+        """
+        Get number of records in file.
+
+        This is maybe suboptimal because we have to seek to the end of
+        the file.
+
+        Side effect: returns file position to record_start.
+        """
+        self.filepath_or_buffer.seek(0, 2)
+        total_records_length = self.filepath_or_buffer.tell() - self.record_start
+
+        if total_records_length % 80 != 0:
+            warnings.warn(
+                "xport file may be corrupted.",
+                stacklevel=find_stack_level(),
+            )
+
+        if self.record_length > 80:
+            self.filepath_or_buffer.seek(self.record_start)
+            return total_records_length // self.record_length
+
+        self.filepath_or_buffer.seek(-80, 2)
+        last_card_bytes = self.filepath_or_buffer.read(80)
+        last_card = np.frombuffer(last_card_bytes, dtype=np.uint64)
+
+        # 8 byte blank
+        ix = np.flatnonzero(last_card == 2314885530818453536)
+
+        if len(ix) == 0:
+            tail_pad = 0
+        else:
+            tail_pad = 8 * len(ix)
+
+        self.filepath_or_buffer.seek(self.record_start)
+
+        return (total_records_length - tail_pad) // self.record_length
+
+    def get_chunk(self, size: int | None = None) -> pd.DataFrame:
+        """
+        Reads lines from Xport file and returns as dataframe
+
+        Parameters
+        ----------
+        size : int, defaults to None
+            Number of lines to read.  If None, reads whole file.
+
+        Returns
+        -------
+        DataFrame
+        """
+        if size is None:
+            size = self._chunksize
+        return self.read(nrows=size)
+
+    def _missing_double(self, vec):
+        v = vec.view(dtype="u1,u1,u2,u4")
+        miss = (v["f1"] == 0) & (v["f2"] == 0) & (v["f3"] == 0)
+        miss1 = (
+            ((v["f0"] >= 0x41) & (v["f0"] <= 0x5A))
+            | (v["f0"] == 0x5F)
+            | (v["f0"] == 0x2E)
+        )
+        miss &= miss1
+        return miss
+
+    @Appender(_read_method_doc)
+    def read(self, nrows: int | None = None) -> pd.DataFrame:
+        if nrows is None:
+            nrows = self.nobs
+
+        read_lines = min(nrows, self.nobs - self._lines_read)
+        read_len = read_lines * self.record_length
+        if read_len <= 0:
+            self.close()
+            raise StopIteration
+        raw = self.filepath_or_buffer.read(read_len)
+        data = np.frombuffer(raw, dtype=self._dtype, count=read_lines)
+
+        df_data = {}
+        for j, x in enumerate(self.columns):
+            vec = data["s" + str(j)]
+            ntype = self.fields[j]["ntype"]
+            if ntype == "numeric":
+                vec = _handle_truncated_float_vec(vec, self.fields[j]["field_length"])
+                miss = self._missing_double(vec)
+                v = _parse_float_vec(vec)
+                v[miss] = np.nan
+            elif self.fields[j]["ntype"] == "char":
+                v = [y.rstrip() for y in vec]
+
+                if self._encoding is not None:
+                    v = [y.decode(self._encoding) for y in v]
+
+            df_data.update({x: v})
+        df = pd.DataFrame(df_data)
+
+        if self._index is None:
+            df.index = pd.Index(range(self._lines_read, self._lines_read + read_lines))
+        else:
+            df = df.set_index(self._index)
+
+        self._lines_read += read_lines
+
+        return df
diff --git a/pandas/pandas/io/sas/sasreader.py b/pandas/pandas/io/sas/sasreader.py
new file mode 100644
index 0000000000000..6daf4a24781bd
--- /dev/null
+++ b/pandas/pandas/io/sas/sasreader.py
@@ -0,0 +1,185 @@
+"""
+Read SAS sas7bdat or xport files.
+"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from collections.abc import Iterator
+from typing import (
+    TYPE_CHECKING,
+    overload,
+)
+
+from pandas.util._decorators import doc
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import stringify_path
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+    from types import TracebackType
+
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        ReadBuffer,
+        Self,
+    )
+
+    from pandas import DataFrame
+
+
+class SASReader(Iterator["DataFrame"], ABC):
+    """
+    Abstract class for XportReader and SAS7BDATReader.
+    """
+
+    @abstractmethod
+    def read(self, nrows: int | None = None) -> DataFrame: ...
+
+    @abstractmethod
+    def close(self) -> None: ...
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        self.close()
+
+
+@overload
+def read_sas(
+    filepath_or_buffer: FilePath | ReadBuffer[bytes],
+    *,
+    format: str | None = ...,
+    index: Hashable | None = ...,
+    encoding: str | None = ...,
+    chunksize: int = ...,
+    iterator: bool = ...,
+    compression: CompressionOptions = ...,
+) -> SASReader: ...
+
+
+@overload
+def read_sas(
+    filepath_or_buffer: FilePath | ReadBuffer[bytes],
+    *,
+    format: str | None = ...,
+    index: Hashable | None = ...,
+    encoding: str | None = ...,
+    chunksize: None = ...,
+    iterator: bool = ...,
+    compression: CompressionOptions = ...,
+) -> DataFrame | SASReader: ...
+
+
+@doc(decompression_options=_shared_docs["decompression_options"] % "filepath_or_buffer")
+def read_sas(
+    filepath_or_buffer: FilePath | ReadBuffer[bytes],
+    *,
+    format: str | None = None,
+    index: Hashable | None = None,
+    encoding: str | None = None,
+    chunksize: int | None = None,
+    iterator: bool = False,
+    compression: CompressionOptions = "infer",
+) -> DataFrame | SASReader:
+    """
+    Read SAS files stored as either XPORT or SAS7BDAT format files.
+
+    Parameters
+    ----------
+    filepath_or_buffer : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a binary ``read()`` function. The string could be a URL.
+        Valid URL schemes include http, ftp, s3, and file. For file URLs, a host is
+        expected. A local file could be:
+        ``file://localhost/path/to/table.sas7bdat``.
+    format : str {{'xport', 'sas7bdat'}} or None
+        If None, file format is inferred from file extension. If 'xport' or
+        'sas7bdat', uses the corresponding format.
+    index : identifier of index column, defaults to None
+        Identifier of column that should be used as index of the DataFrame.
+    encoding : str, default is None
+        Encoding for text data.  If None, text data are stored as raw bytes.
+    chunksize : int
+        Read file `chunksize` lines at a time, returns iterator.
+    iterator : bool, defaults to False
+        If True, returns an iterator for reading the file incrementally.
+    {decompression_options}
+
+    Returns
+    -------
+    DataFrame, SAS7BDATReader, or XportReader
+        DataFrame if iterator=False and chunksize=None, else SAS7BDATReader
+        or XportReader, file format is inferred from file extension.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (csv) file into a pandas DataFrame.
+    read_excel : Read an Excel file into a pandas DataFrame.
+    read_spss : Read an SPSS file into a pandas DataFrame.
+    read_orc : Load an ORC object into a pandas DataFrame.
+    read_feather : Load a feather-format object into a pandas DataFrame.
+
+    Examples
+    --------
+    >>> df = pd.read_sas("sas_data.sas7bdat")  # doctest: +SKIP
+    """
+    if format is None:
+        buffer_error_msg = (
+            "If this is a buffer object rather "
+            "than a string name, you must specify a format string"
+        )
+        filepath_or_buffer = stringify_path(filepath_or_buffer)
+        if not isinstance(filepath_or_buffer, str):
+            raise ValueError(buffer_error_msg)
+        fname = filepath_or_buffer.lower()
+        if ".xpt" in fname:
+            format = "xport"
+        elif ".sas7bdat" in fname:
+            format = "sas7bdat"
+        else:
+            raise ValueError(
+                f"unable to infer format of SAS file from filename: {fname!r}"
+            )
+
+    reader: SASReader
+    if format.lower() == "xport":
+        from pandas.io.sas.sas_xport import XportReader
+
+        reader = XportReader(
+            filepath_or_buffer,
+            index=index,
+            encoding=encoding,
+            chunksize=chunksize,
+            compression=compression,
+        )
+    elif format.lower() == "sas7bdat":
+        from pandas.io.sas.sas7bdat import SAS7BDATReader
+
+        reader = SAS7BDATReader(
+            filepath_or_buffer,
+            index=index,
+            encoding=encoding,
+            chunksize=chunksize,
+            compression=compression,
+        )
+    else:
+        raise ValueError("unknown SAS format")
+
+    if iterator or chunksize:
+        return reader
+
+    with reader:
+        return reader.read()
diff --git a/pandas/pandas/io/spss.py b/pandas/pandas/io/spss.py
new file mode 100644
index 0000000000000..dfada10c719c9
--- /dev/null
+++ b/pandas/pandas/io/spss.py
@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.inference import is_list_like
+
+from pandas.io.common import stringify_path
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+    from pathlib import Path
+
+    from pandas._typing import DtypeBackend
+
+    from pandas import DataFrame
+
+
+def read_spss(
+    path: str | Path,
+    usecols: Sequence[str] | None = None,
+    convert_categoricals: bool = True,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    **kwargs: Any,
+) -> DataFrame:
+    """
+    Load an SPSS file from the file path, returning a DataFrame.
+
+    Parameters
+    ----------
+    path : str or Path
+        File path.
+    usecols : list-like, optional
+        Return a subset of the columns. If None, return all columns.
+    convert_categoricals : bool, default is True
+        Convert categorical columns into pd.Categorical.
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed
+          nullable :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+    **kwargs
+        Additional keyword arguments that can be passed to :func:`pyreadstat.read_sav`.
+
+        .. versionadded:: 3.0
+
+    Returns
+    -------
+    DataFrame
+        DataFrame based on the SPSS file.
+
+    See Also
+    --------
+    read_csv : Read a comma-separated values (csv) file into a pandas DataFrame.
+    read_excel : Read an Excel file into a pandas DataFrame.
+    read_sas : Read an SAS file into a pandas DataFrame.
+    read_orc : Load an ORC object into a pandas DataFrame.
+    read_feather : Load a feather-format object into a pandas DataFrame.
+
+    Examples
+    --------
+    >>> df = pd.read_spss("spss_data.sav")  # doctest: +SKIP
+    """
+    pyreadstat = import_optional_dependency("pyreadstat")
+    check_dtype_backend(dtype_backend)
+
+    if usecols is not None:
+        if not is_list_like(usecols):
+            raise TypeError("usecols must be list-like.")
+        usecols = list(usecols)  # pyreadstat requires a list
+
+    df, metadata = pyreadstat.read_sav(
+        stringify_path(path),
+        usecols=usecols,
+        apply_value_formats=convert_categoricals,
+        **kwargs,
+    )
+    df.attrs = metadata.__dict__
+    if dtype_backend is not lib.no_default:
+        df = df.convert_dtypes(dtype_backend=dtype_backend)
+    return df
diff --git a/pandas/pandas/io/sql.py b/pandas/pandas/io/sql.py
new file mode 100644
index 0000000000000..0e0f07c0f8ff3
--- /dev/null
+++ b/pandas/pandas/io/sql.py
@@ -0,0 +1,2960 @@
+"""
+Collection of query wrappers / abstractions to both facilitate data
+retrieval and to reduce dependency on DB-specific API.
+"""
+
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from contextlib import (
+    ExitStack,
+    contextmanager,
+)
+from datetime import (
+    date,
+    datetime,
+    time,
+)
+from functools import partial
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    overload,
+)
+import warnings
+
+import numpy as np
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    AbstractMethodError,
+    DatabaseError,
+)
+from pandas.util._exceptions import find_stack_level
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import (
+    is_dict_like,
+    is_list_like,
+    is_object_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.missing import isna
+
+from pandas import get_option
+from pandas.core.api import (
+    DataFrame,
+    Series,
+)
+from pandas.core.arrays import ArrowExtensionArray
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.base import PandasObject
+import pandas.core.common as com
+from pandas.core.common import maybe_make_list
+from pandas.core.internals.construction import convert_object_array
+from pandas.core.tools.datetimes import to_datetime
+
+from pandas.io._util import arrow_table_to_pandas
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Generator,
+        Iterator,
+        Mapping,
+    )
+
+    from sqlalchemy import Table
+    from sqlalchemy.sql.expression import (
+        Delete,
+        Select,
+        TextClause,
+    )
+
+    from pandas._typing import (
+        DtypeArg,
+        DtypeBackend,
+        IndexLabel,
+        Self,
+    )
+
+    from pandas import Index
+
+# -----------------------------------------------------------------------------
+# -- Helper functions
+
+
+def _process_parse_dates_argument(parse_dates):
+    """Process parse_dates argument for read_sql functions"""
+    # handle non-list entries for parse_dates gracefully
+    if parse_dates is True or parse_dates is None or parse_dates is False:
+        parse_dates = []
+
+    elif not hasattr(parse_dates, "__iter__"):
+        parse_dates = [parse_dates]
+    return parse_dates
+
+
+def _handle_date_column(
+    col, utc: bool = False, format: str | dict[str, Any] | None = None
+):
+    if isinstance(format, dict):
+        # GH35185 Allow custom error values in parse_dates argument of
+        # read_sql like functions.
+        # Format can take on custom to_datetime argument values such as
+        # {"errors": "coerce"} or {"dayfirst": True}
+        return to_datetime(col, **format)
+    else:
+        # Allow passing of formatting string for integers
+        # GH17855
+        if format is None and (
+            issubclass(col.dtype.type, np.floating)
+            or issubclass(col.dtype.type, np.integer)
+        ):
+            format = "s"
+        if format in ["D", "d", "h", "m", "s", "ms", "us", "ns"]:
+            return to_datetime(col, errors="coerce", unit=format, utc=utc)
+        elif isinstance(col.dtype, DatetimeTZDtype):
+            # coerce to UTC timezone
+            # GH11216
+            return to_datetime(col, utc=True)
+        else:
+            return to_datetime(col, errors="coerce", format=format, utc=utc)
+
+
+def _parse_date_columns(data_frame: DataFrame, parse_dates) -> DataFrame:
+    """
+    Force non-datetime columns to be read as such.
+    Supports both string formatted and integer timestamp columns.
+    """
+    parse_dates = _process_parse_dates_argument(parse_dates)
+
+    # we want to coerce datetime64_tz dtypes for now to UTC
+    # we could in theory do a 'nice' conversion from a FixedOffset tz
+    # GH11216
+    for i, (col_name, df_col) in enumerate(data_frame.items()):
+        if isinstance(df_col.dtype, DatetimeTZDtype) or col_name in parse_dates:
+            try:
+                fmt = parse_dates[col_name]
+            except (KeyError, TypeError):
+                fmt = None
+            data_frame.isetitem(i, _handle_date_column(df_col, format=fmt))
+
+    return data_frame
+
+
+def _convert_arrays_to_dataframe(
+    data,
+    columns,
+    coerce_float: bool = True,
+    dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+) -> DataFrame:
+    content = lib.to_object_array_tuples(data)
+    idx_len = content.shape[0]
+    arrays = convert_object_array(
+        list(content.T),
+        dtype=None,
+        coerce_float=coerce_float,
+        dtype_backend=dtype_backend,
+    )
+    if dtype_backend == "pyarrow":
+        pa = import_optional_dependency("pyarrow")
+
+        result_arrays = []
+        for arr in arrays:
+            pa_array = pa.array(arr, from_pandas=True)
+            if arr.dtype == "string":
+                # TODO: Arrow still infers strings arrays as regular strings instead
+                # of large_string, which is what we preserver everywhere else for
+                # dtype_backend="pyarrow". We may want to reconsider this
+                pa_array = pa_array.cast(pa.string())
+            result_arrays.append(ArrowExtensionArray(pa_array))
+        arrays = result_arrays  # type: ignore[assignment]
+    if arrays:
+        return DataFrame._from_arrays(
+            arrays, columns=columns, index=range(idx_len), verify_integrity=False
+        )
+    else:
+        return DataFrame(columns=columns)
+
+
+def _wrap_result(
+    data,
+    columns,
+    index_col=None,
+    coerce_float: bool = True,
+    parse_dates=None,
+    dtype: DtypeArg | None = None,
+    dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+) -> DataFrame:
+    """Wrap result set of a SQLAlchemy query in a DataFrame."""
+    frame = _convert_arrays_to_dataframe(data, columns, coerce_float, dtype_backend)
+
+    if dtype:
+        frame = frame.astype(dtype)
+
+    frame = _parse_date_columns(frame, parse_dates)
+
+    if index_col is not None:
+        frame = frame.set_index(index_col)
+
+    return frame
+
+
+def _wrap_result_adbc(
+    df: DataFrame,
+    *,
+    index_col=None,
+    parse_dates=None,
+    dtype: DtypeArg | None = None,
+    dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+) -> DataFrame:
+    """Wrap result set of a SQLAlchemy query in a DataFrame."""
+    if dtype:
+        df = df.astype(dtype)
+
+    df = _parse_date_columns(df, parse_dates)
+
+    if index_col is not None:
+        df = df.set_index(index_col)
+
+    return df
+
+
+# -----------------------------------------------------------------------------
+# -- Read and write to DataFrames
+
+
+@overload
+def read_sql_table(  # pyright: ignore[reportOverlappingOverload]
+    table_name: str,
+    con,
+    schema=...,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = ...,
+    columns: list[str] | None = ...,
+    chunksize: None = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> DataFrame: ...
+
+
+@overload
+def read_sql_table(
+    table_name: str,
+    con,
+    schema=...,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = ...,
+    columns: list[str] | None = ...,
+    chunksize: int = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> Iterator[DataFrame]: ...
+
+
+def read_sql_table(
+    table_name: str,
+    con,
+    schema: str | None = None,
+    index_col: str | list[str] | None = None,
+    coerce_float: bool = True,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = None,
+    columns: list[str] | None = None,
+    chunksize: int | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame | Iterator[DataFrame]:
+    """
+    Read SQL database table into a DataFrame.
+
+    Given a table name and a SQLAlchemy connectable, returns a DataFrame.
+    This function does not support DBAPI connections.
+
+    Parameters
+    ----------
+    table_name : str
+        Name of SQL table in database.
+    con : SQLAlchemy connectable or str
+        A database URI could be provided as str.
+        SQLite DBAPI connection mode not supported.
+    schema : str, default None
+        Name of SQL schema in database to query (if database flavor
+        supports this). Uses default schema if None (default).
+    index_col : str or list of str, optional, default: None
+        Column(s) to set as index(MultiIndex).
+    coerce_float : bool, default True
+        Attempts to convert values of non-string, non-numeric objects (like
+        decimal.Decimal) to floating point. Can result in loss of Precision.
+    parse_dates : list or dict, default None
+        - List of column names to parse as dates.
+        - Dict of ``{column_name: format string}`` where format string is
+          strftime compatible in case of parsing string times or is one of
+          (D, s, ns, ms, us) in case of parsing integer timestamps.
+        - Dict of ``{column_name: arg dict}``, where the arg dict corresponds
+          to the keyword arguments of :func:`pandas.to_datetime`
+          Especially useful with databases without native Datetime support,
+          such as SQLite.
+    columns : list, default None
+        List of column names to select from SQL table.
+    chunksize : int, default None
+        If specified, returns an iterator where `chunksize` is the number of
+        rows to include in each chunk.
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    DataFrame or Iterator[DataFrame]
+        A SQL table is returned as two-dimensional data structure with labeled
+        axes.
+
+    See Also
+    --------
+    read_sql_query : Read SQL query into a DataFrame.
+    read_sql : Read SQL query or database table into a DataFrame.
+
+    Notes
+    -----
+    Any datetime values with time zone information will be converted to UTC.
+
+    Examples
+    --------
+    >>> pd.read_sql_table("table_name", "postgres:///db_name")  # doctest:+SKIP
+    """
+
+    check_dtype_backend(dtype_backend)
+    if dtype_backend is lib.no_default:
+        dtype_backend = "numpy"  # type: ignore[assignment]
+    assert dtype_backend is not lib.no_default
+
+    with pandasSQL_builder(con, schema=schema, need_transaction=True) as pandas_sql:
+        if not pandas_sql.has_table(table_name):
+            raise ValueError(f"Table {table_name} not found")
+
+        table = pandas_sql.read_table(
+            table_name,
+            index_col=index_col,
+            coerce_float=coerce_float,
+            parse_dates=parse_dates,
+            columns=columns,
+            chunksize=chunksize,
+            dtype_backend=dtype_backend,
+        )
+
+    if table is not None:
+        return table
+    else:
+        raise ValueError(f"Table {table_name} not found", con)
+
+
+@overload
+def read_sql_query(  # pyright: ignore[reportOverlappingOverload]
+    sql,
+    con,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    params: list[Any] | Mapping[str, Any] | None = ...,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = ...,
+    chunksize: None = ...,
+    dtype: DtypeArg | None = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> DataFrame: ...
+
+
+@overload
+def read_sql_query(
+    sql,
+    con,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    params: list[Any] | Mapping[str, Any] | None = ...,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = ...,
+    chunksize: int = ...,
+    dtype: DtypeArg | None = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+) -> Iterator[DataFrame]: ...
+
+
+def read_sql_query(
+    sql,
+    con,
+    index_col: str | list[str] | None = None,
+    coerce_float: bool = True,
+    params: list[Any] | Mapping[str, Any] | None = None,
+    parse_dates: list[str] | dict[str, str] | dict[str, dict[str, Any]] | None = None,
+    chunksize: int | None = None,
+    dtype: DtypeArg | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame | Iterator[DataFrame]:
+    """
+    Read SQL query into a DataFrame.
+
+    Returns a DataFrame corresponding to the result set of the query
+    string. Optionally provide an `index_col` parameter to use one of the
+    columns as the index, otherwise default integer index will be used.
+
+    Parameters
+    ----------
+    sql : str SQL query or SQLAlchemy Selectable (select or text object)
+        SQL query to be executed.
+    con : SQLAlchemy connectable, str, or sqlite3 connection
+        Using SQLAlchemy makes it possible to use any DB supported by that
+        library. If a DBAPI2 object, only sqlite3 is supported.
+    index_col : str or list of str, optional, default: None
+        Column(s) to set as index(MultiIndex).
+    coerce_float : bool, default True
+        Attempts to convert values of non-string, non-numeric objects (like
+        decimal.Decimal) to floating point. Useful for SQL result sets.
+    params : list, tuple or mapping, optional, default: None
+        List of parameters to pass to execute method.  The syntax used
+        to pass parameters is database driver dependent. Check your
+        database driver documentation for which of the five syntax styles,
+        described in PEP 249's paramstyle, is supported.
+        Eg. for psycopg2, uses %(name)s so use params={'name' : 'value'}.
+    parse_dates : list or dict, default: None
+        - List of column names to parse as dates.
+        - Dict of ``{column_name: format string}`` where format string is
+          strftime compatible in case of parsing string times, or is one of
+          (D, s, ns, ms, us) in case of parsing integer timestamps.
+        - Dict of ``{column_name: arg dict}``, where the arg dict corresponds
+          to the keyword arguments of :func:`pandas.to_datetime`
+          Especially useful with databases without native Datetime support,
+          such as SQLite.
+    chunksize : int, default None
+        If specified, return an iterator where `chunksize` is the number of
+        rows to include in each chunk.
+    dtype : Type name or dict of columns
+        Data type for data or columns. E.g. np.float64 or
+        {'a': np.float64, 'b': np.int32, 'c': 'Int64'}.
+
+        .. versionadded:: 1.3.0
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    DataFrame or Iterator[DataFrame]
+        Returns a DataFrame object that contains the result set of the
+        executed SQL query, in relation to the specified database connection.
+
+    See Also
+    --------
+    read_sql_table : Read SQL database table into a DataFrame.
+    read_sql : Read SQL query or database table into a DataFrame.
+
+    Notes
+    -----
+    Any datetime values with time zone information parsed via the `parse_dates`
+    parameter will be converted to UTC.
+
+    Examples
+    --------
+    >>> from sqlalchemy import create_engine  # doctest: +SKIP
+    >>> engine = create_engine("sqlite:///database.db")  # doctest: +SKIP
+    >>> sql_query = "SELECT int_column FROM test_data"  # doctest: +SKIP
+    >>> with engine.connect() as conn, conn.begin():  # doctest: +SKIP
+    ...     data = pd.read_sql_query(sql_query, conn)  # doctest: +SKIP
+    """
+
+    check_dtype_backend(dtype_backend)
+    if dtype_backend is lib.no_default:
+        dtype_backend = "numpy"  # type: ignore[assignment]
+    assert dtype_backend is not lib.no_default
+
+    with pandasSQL_builder(con) as pandas_sql:
+        return pandas_sql.read_query(
+            sql,
+            index_col=index_col,
+            params=params,
+            coerce_float=coerce_float,
+            parse_dates=parse_dates,
+            chunksize=chunksize,
+            dtype=dtype,
+            dtype_backend=dtype_backend,
+        )
+
+
+@overload
+def read_sql(  # pyright: ignore[reportOverlappingOverload]
+    sql,
+    con,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    params=...,
+    parse_dates=...,
+    columns: list[str] = ...,
+    chunksize: None = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    dtype: DtypeArg | None = None,
+) -> DataFrame: ...
+
+
+@overload
+def read_sql(
+    sql,
+    con,
+    index_col: str | list[str] | None = ...,
+    coerce_float=...,
+    params=...,
+    parse_dates=...,
+    columns: list[str] = ...,
+    chunksize: int = ...,
+    dtype_backend: DtypeBackend | lib.NoDefault = ...,
+    dtype: DtypeArg | None = None,
+) -> Iterator[DataFrame]: ...
+
+
+def read_sql(
+    sql,
+    con,
+    index_col: str | list[str] | None = None,
+    coerce_float: bool = True,
+    params=None,
+    parse_dates=None,
+    columns: list[str] | None = None,
+    chunksize: int | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    dtype: DtypeArg | None = None,
+) -> DataFrame | Iterator[DataFrame]:
+    """
+    Read SQL query or database table into a DataFrame.
+
+    This function is a convenience wrapper around ``read_sql_table`` and
+    ``read_sql_query`` (for backward compatibility). It will delegate
+    to the specific function depending on the provided input. A SQL query
+    will be routed to ``read_sql_query``, while a database table name will
+    be routed to ``read_sql_table``. Note that the delegated function might
+    have more specific notes about their functionality not listed here.
+
+    Parameters
+    ----------
+    sql : str or SQLAlchemy Selectable (select or text object)
+        SQL query to be executed or a table name.
+    con : ADBC Connection, SQLAlchemy connectable, str, or sqlite3 connection
+        ADBC provides high performance I/O with native type support, where available.
+        Using SQLAlchemy makes it possible to use any DB supported by that
+        library. If a DBAPI2 object, only sqlite3 is supported. The user is responsible
+        for engine disposal and connection closure for the ADBC connection and
+        SQLAlchemy connectable; str connections are closed automatically. See
+        `here <https://docs.sqlalchemy.org/en/20/core/connections.html>`_.
+    index_col : str or list of str, optional, default: None
+        Column(s) to set as index(MultiIndex).
+    coerce_float : bool, default True
+        Attempts to convert values of non-string, non-numeric objects (like
+        decimal.Decimal) to floating point, useful for SQL result sets.
+    params : list, tuple or dict, optional, default: None
+        List of parameters to pass to execute method.  The syntax used
+        to pass parameters is database driver dependent. Check your
+        database driver documentation for which of the five syntax styles,
+        described in PEP 249's paramstyle, is supported.
+        Eg. for psycopg2, uses %(name)s so use params={'name' : 'value'}.
+    parse_dates : list or dict, default: None
+        - List of column names to parse as dates.
+        - Dict of ``{column_name: format string}`` where format string is
+          strftime compatible in case of parsing string times, or is one of
+          (D, s, ns, ms, us) in case of parsing integer timestamps.
+        - Dict of ``{column_name: arg dict}``, where the arg dict corresponds
+          to the keyword arguments of :func:`pandas.to_datetime`
+          Especially useful with databases without native Datetime support,
+          such as SQLite.
+    columns : list, default: None
+        List of column names to select from SQL table (only used when reading
+        a table).
+    chunksize : int, default None
+        If specified, return an iterator where `chunksize` is the
+        number of rows to include in each chunk.
+    dtype_backend : {'numpy_nullable', 'pyarrow'}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+    dtype : Type name or dict of columns
+        Data type for data or columns. E.g. np.float64 or
+        {'a': np.float64, 'b': np.int32, 'c': 'Int64'}.
+        The argument is ignored if a table is passed instead of a query.
+
+        .. versionadded:: 2.0.0
+
+    Returns
+    -------
+    DataFrame or Iterator[DataFrame]
+        Returns a DataFrame object that contains the result set of the
+        executed SQL query or an SQL Table based on the provided input,
+        in relation to the specified database connection.
+
+    See Also
+    --------
+    read_sql_table : Read SQL database table into a DataFrame.
+    read_sql_query : Read SQL query into a DataFrame.
+
+    Notes
+    -----
+    ``pandas`` does not attempt to sanitize SQL statements;
+    instead it simply forwards the statement you are executing
+    to the underlying driver, which may or may not sanitize from there.
+    Please refer to the underlying driver documentation for any details.
+    Generally, be wary when accepting statements from arbitrary sources.
+
+    Examples
+    --------
+    Read data from SQL via either a SQL query or a SQL tablename.
+    When using a SQLite database only SQL queries are accepted,
+    providing only the SQL tablename will result in an error.
+
+    >>> from sqlite3 import connect
+    >>> conn = connect(":memory:")
+    >>> df = pd.DataFrame(
+    ...     data=[[0, "10/11/12"], [1, "12/11/10"]],
+    ...     columns=["int_column", "date_column"],
+    ... )
+    >>> df.to_sql(name="test_data", con=conn)
+    2
+
+    >>> pd.read_sql("SELECT int_column, date_column FROM test_data", conn)
+       int_column date_column
+    0           0    10/11/12
+    1           1    12/11/10
+
+    >>> pd.read_sql("test_data", "postgres:///db_name")  # doctest:+SKIP
+
+    For parameterized query, using ``params`` is recommended over string interpolation.
+
+    >>> from sqlalchemy import text
+    >>> sql = text(
+    ...     "SELECT int_column, date_column FROM test_data WHERE int_column=:int_val"
+    ... )
+    >>> pd.read_sql(sql, conn, params={"int_val": 1})  # doctest:+SKIP
+       int_column date_column
+    0           1    12/11/10
+
+    Apply date parsing to columns through the ``parse_dates`` argument
+    The ``parse_dates`` argument calls ``pd.to_datetime`` on the provided columns.
+    Custom argument values for applying ``pd.to_datetime`` on a column are specified
+    via a dictionary format:
+
+    >>> pd.read_sql(
+    ...     "SELECT int_column, date_column FROM test_data",
+    ...     conn,
+    ...     parse_dates={"date_column": {"format": "%d/%m/%y"}},
+    ... )
+       int_column date_column
+    0           0  2012-11-10
+    1           1  2010-11-12
+
+    .. versionadded:: 2.2.0
+
+       pandas now supports reading via ADBC drivers
+
+    >>> from adbc_driver_postgresql import dbapi  # doctest:+SKIP
+    >>> with dbapi.connect("postgres:///db_name") as conn:  # doctest:+SKIP
+    ...     pd.read_sql("SELECT int_column FROM test_data", conn)
+       int_column
+    0           0
+    1           1
+    """
+
+    check_dtype_backend(dtype_backend)
+    if dtype_backend is lib.no_default:
+        dtype_backend = "numpy"  # type: ignore[assignment]
+    assert dtype_backend is not lib.no_default
+
+    with pandasSQL_builder(con) as pandas_sql:
+        if isinstance(pandas_sql, SQLiteDatabase):
+            return pandas_sql.read_query(
+                sql,
+                index_col=index_col,
+                params=params,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                chunksize=chunksize,
+                dtype_backend=dtype_backend,
+                dtype=dtype,
+            )
+
+        try:
+            _is_table_name = pandas_sql.has_table(sql)
+        except Exception:
+            # using generic exception to catch errors from sql drivers (GH24988)
+            _is_table_name = False
+
+        if _is_table_name:
+            return pandas_sql.read_table(
+                sql,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                columns=columns,
+                chunksize=chunksize,
+                dtype_backend=dtype_backend,
+            )
+        else:
+            return pandas_sql.read_query(
+                sql,
+                index_col=index_col,
+                params=params,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                chunksize=chunksize,
+                dtype_backend=dtype_backend,
+                dtype=dtype,
+            )
+
+
+def to_sql(
+    frame,
+    name: str,
+    con,
+    schema: str | None = None,
+    if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+    index: bool = True,
+    index_label: IndexLabel | None = None,
+    chunksize: int | None = None,
+    dtype: DtypeArg | None = None,
+    method: Literal["multi"] | Callable | None = None,
+    engine: str = "auto",
+    **engine_kwargs,
+) -> int | None:
+    """
+    Write records stored in a DataFrame to a SQL database.
+
+    .. warning::
+        The pandas library does not attempt to sanitize inputs provided via a to_sql call.
+        Please refer to the documentation for the underlying database driver to see if it
+        will properly prevent injection, or alternatively be advised of a security risk when
+        executing arbitrary commands in a to_sql call.
+
+    Parameters
+    ----------
+    frame : DataFrame, Series
+    name : str
+        Name of SQL table.
+    con : ADBC Connection, SQLAlchemy connectable, str, or sqlite3 connection
+        or sqlite3 DBAPI2 connection
+        ADBC provides high performance I/O with native type support, where available.
+        Using SQLAlchemy makes it possible to use any DB supported by that
+        library.
+        If a DBAPI2 object, only sqlite3 is supported.
+    schema : str, optional
+        Name of SQL schema in database to write to (if database flavor
+        supports this). If None, use default schema (default).
+    if_exists : {'fail', 'replace', 'append', 'delete_rows'}, default 'fail'
+        - fail: If table exists, do nothing.
+        - replace: If table exists, drop it, recreate it, and insert data.
+        - append: If table exists, insert data. Create if does not exist.
+        - delete_rows: If a table exists, delete all records and insert data.
+    index : bool, default True
+        Write DataFrame index as a column.
+    index_label : str or sequence, optional
+        Column label for index column(s). If None is given (default) and
+        `index` is True, then the index names are used.
+        A sequence should be given if the DataFrame uses MultiIndex.
+    chunksize : int, optional
+        Specify the number of rows in each batch to be written at a time.
+        By default, all rows will be written at once.
+    dtype : dict or scalar, optional
+        Specifying the datatype for columns. If a dictionary is used, the
+        keys should be the column names and the values should be the
+        SQLAlchemy types or strings for the sqlite3 fallback mode. If a
+        scalar is provided, it will be applied to all columns.
+    method : {None, 'multi', callable}, optional
+        Controls the SQL insertion clause used:
+
+        - None : Uses standard SQL ``INSERT`` clause (one per row).
+        - ``'multi'``: Pass multiple values in a single ``INSERT`` clause.
+        - callable with signature ``(pd_table, conn, keys, data_iter) -> int | None``.
+
+        Details and a sample callable implementation can be found in the
+        section :ref:`insert method <io.sql.method>`.
+    engine : {'auto', 'sqlalchemy'}, default 'auto'
+        SQL engine library to use. If 'auto', then the option
+        ``io.sql.engine`` is used. The default ``io.sql.engine``
+        behavior is 'sqlalchemy'
+
+        .. versionadded:: 1.3.0
+
+    **engine_kwargs
+        Any additional kwargs are passed to the engine.
+
+    Returns
+    -------
+    None or int
+        Number of rows affected by to_sql. None is returned if the callable
+        passed into ``method`` does not return an integer number of rows.
+
+        .. versionadded:: 1.4.0
+
+    Notes
+    -----
+    The returned rows affected is the sum of the ``rowcount`` attribute of ``sqlite3.Cursor``
+    or SQLAlchemy connectable. If using ADBC the returned rows are the result
+    of ``Cursor.adbc_ingest``. The returned value may not reflect the exact number of written
+    rows as stipulated in the
+    `sqlite3 <https://docs.python.org/3/library/sqlite3.html#sqlite3.Cursor.rowcount>`__ or
+    `SQLAlchemy <https://docs.sqlalchemy.org/en/14/core/connections.html#sqlalchemy.engine.BaseCursorResult.rowcount>`__
+    """  # noqa: E501
+    if if_exists not in ("fail", "replace", "append", "delete_rows"):
+        raise ValueError(f"'{if_exists}' is not valid for if_exists")
+
+    if isinstance(frame, Series):
+        frame = frame.to_frame()
+    elif not isinstance(frame, DataFrame):
+        raise NotImplementedError(
+            "'frame' argument should be either a Series or a DataFrame"
+        )
+
+    with pandasSQL_builder(con, schema=schema, need_transaction=True) as pandas_sql:
+        return pandas_sql.to_sql(
+            frame,
+            name,
+            if_exists=if_exists,
+            index=index,
+            index_label=index_label,
+            schema=schema,
+            chunksize=chunksize,
+            dtype=dtype,
+            method=method,
+            engine=engine,
+            **engine_kwargs,
+        )
+
+
+def has_table(table_name: str, con, schema: str | None = None) -> bool:
+    """
+    Check if DataBase has named table.
+
+    Parameters
+    ----------
+    table_name: string
+        Name of SQL table.
+    con: ADBC Connection, SQLAlchemy connectable, str, or sqlite3 connection
+        ADBC provides high performance I/O with native type support, where available.
+        Using SQLAlchemy makes it possible to use any DB supported by that
+        library.
+        If a DBAPI2 object, only sqlite3 is supported.
+    schema : string, default None
+        Name of SQL schema in database to write to (if database flavor supports
+        this). If None, use default schema (default).
+
+    Returns
+    -------
+    boolean
+    """
+    with pandasSQL_builder(con, schema=schema) as pandas_sql:
+        return pandas_sql.has_table(table_name)
+
+
+table_exists = has_table
+
+
+def pandasSQL_builder(
+    con,
+    schema: str | None = None,
+    need_transaction: bool = False,
+) -> PandasSQL:
+    """
+    Convenience function to return the correct PandasSQL subclass based on the
+    provided parameters.  Also creates a sqlalchemy connection and transaction
+    if necessary.
+    """
+    import sqlite3
+
+    if isinstance(con, sqlite3.Connection) or con is None:
+        return SQLiteDatabase(con)
+
+    sqlalchemy = import_optional_dependency("sqlalchemy", errors="ignore")
+
+    if isinstance(con, str) and sqlalchemy is None:
+        raise ImportError("Using URI string without sqlalchemy installed.")
+
+    if sqlalchemy is not None and isinstance(con, (str, sqlalchemy.engine.Connectable)):
+        return SQLDatabase(con, schema, need_transaction)
+
+    adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+    if adbc and isinstance(con, adbc.Connection):
+        return ADBCDatabase(con)
+
+    warnings.warn(
+        "pandas only supports SQLAlchemy connectable (engine/connection) or "
+        "database string URI or sqlite3 DBAPI2 connection. Other DBAPI2 "
+        "objects are not tested. Please consider using SQLAlchemy.",
+        UserWarning,
+        stacklevel=find_stack_level(),
+    )
+    return SQLiteDatabase(con)
+
+
+class SQLTable(PandasObject):
+    """
+    For mapping Pandas tables to SQL tables.
+    Uses fact that table is reflected by SQLAlchemy to
+    do better type conversions.
+    Also holds various flags needed to avoid having to
+    pass them between functions all the time.
+    """
+
+    # TODO: support for multiIndex
+
+    def __init__(
+        self,
+        name: str,
+        pandas_sql_engine,
+        frame=None,
+        index: bool | str | list[str] | None = True,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        prefix: str = "pandas",
+        index_label=None,
+        schema=None,
+        keys=None,
+        dtype: DtypeArg | None = None,
+    ) -> None:
+        self.name = name
+        self.pd_sql = pandas_sql_engine
+        self.prefix = prefix
+        self.frame = frame
+        self.index = self._index_name(index, index_label)
+        self.schema = schema
+        self.if_exists = if_exists
+        self.keys = keys
+        self.dtype = dtype
+
+        if frame is not None:
+            # We want to initialize based on a dataframe
+            self.table = self._create_table_setup()
+        else:
+            # no data provided, read-only mode
+            self.table = self.pd_sql.get_table(self.name, self.schema)
+
+        if self.table is None:
+            raise ValueError(f"Could not init table '{name}'")
+
+        if not len(self.name):
+            raise ValueError("Empty table name specified")
+
+    def exists(self):
+        return self.pd_sql.has_table(self.name, self.schema)
+
+    def sql_schema(self) -> str:
+        from sqlalchemy.schema import CreateTable
+
+        return str(CreateTable(self.table).compile(self.pd_sql.con))
+
+    def _execute_create(self) -> None:
+        # Inserting table into database, add to MetaData object
+        self.table = self.table.to_metadata(self.pd_sql.meta)
+        with self.pd_sql.run_transaction():
+            self.table.create(bind=self.pd_sql.con)
+
+    def create(self) -> None:
+        if self.exists():
+            if self.if_exists == "fail":
+                raise ValueError(f"Table '{self.name}' already exists.")
+            elif self.if_exists == "replace":
+                self.pd_sql.drop_table(self.name, self.schema)
+                self._execute_create()
+            elif self.if_exists == "append":
+                pass
+            elif self.if_exists == "delete_rows":
+                self.pd_sql.delete_rows(self.name, self.schema)
+            else:
+                raise ValueError(f"'{self.if_exists}' is not valid for if_exists")
+        else:
+            self._execute_create()
+
+    def _execute_insert(self, conn, keys: list[str], data_iter) -> int:
+        """
+        Execute SQL statement inserting data
+
+        Parameters
+        ----------
+        conn : sqlalchemy.engine.Engine or sqlalchemy.engine.Connection
+        keys : list of str
+           Column names
+        data_iter : generator of list
+           Each item contains a list of values to be inserted
+        """
+        data = [dict(zip(keys, row)) for row in data_iter]
+        result = self.pd_sql.execute(self.table.insert(), data)
+        return result.rowcount
+
+    def _execute_insert_multi(self, conn, keys: list[str], data_iter) -> int:
+        """
+        Alternative to _execute_insert for DBs support multi-value INSERT.
+
+        Note: multi-value insert is usually faster for analytics DBs
+        and tables containing a few columns
+        but performance degrades quickly with increase of columns.
+
+        """
+
+        from sqlalchemy import insert
+
+        data = [dict(zip(keys, row)) for row in data_iter]
+        stmt = insert(self.table).values(data)
+        result = self.pd_sql.execute(stmt)
+        return result.rowcount
+
+    def insert_data(self) -> tuple[list[str], list[np.ndarray]]:
+        if self.index is not None:
+            temp = self.frame.copy(deep=False)
+            temp.index.names = self.index
+            try:
+                temp.reset_index(inplace=True)
+            except ValueError as err:
+                raise ValueError(f"duplicate name in index/columns: {err}") from err
+        else:
+            temp = self.frame
+
+        column_names = list(map(str, temp.columns))
+        ncols = len(column_names)
+        # this just pre-allocates the list: None's will be replaced with ndarrays
+        # error: List item 0 has incompatible type "None"; expected "ndarray"
+        data_list: list[np.ndarray] = [None] * ncols  # type: ignore[list-item]
+
+        for i, (_, ser) in enumerate(temp.items()):
+            if ser.dtype.kind == "M":
+                if isinstance(ser._values, ArrowExtensionArray):
+                    import pyarrow as pa
+
+                    if pa.types.is_date(ser.dtype.pyarrow_dtype):
+                        # GH#53854 to_pydatetime not supported for pyarrow date dtypes
+                        d = ser._values.to_numpy(dtype=object)
+                    else:
+                        d = ser.dt.to_pydatetime()._values
+                else:
+                    d = ser._values.to_pydatetime()
+            elif ser.dtype.kind == "m":
+                vals = ser._values
+                if isinstance(vals, ArrowExtensionArray):
+                    vals = vals.to_numpy(dtype=np.dtype("m8[ns]"))
+                # store as integers, see GH#6921, GH#7076
+                d = vals.view("i8").astype(object)
+            else:
+                d = ser._values.astype(object)
+
+            assert isinstance(d, np.ndarray), type(d)
+
+            if ser._can_hold_na:
+                # Note: this will miss timedeltas since they are converted to int
+                mask = isna(d)
+                d[mask] = None
+
+            data_list[i] = d
+
+        return column_names, data_list
+
+    def insert(
+        self,
+        chunksize: int | None = None,
+        method: Literal["multi"] | Callable | None = None,
+    ) -> int | None:
+        # set insert method
+        if method is None:
+            exec_insert = self._execute_insert
+        elif method == "multi":
+            exec_insert = self._execute_insert_multi
+        elif callable(method):
+            exec_insert = partial(method, self)
+        else:
+            raise ValueError(f"Invalid parameter `method`: {method}")
+
+        keys, data_list = self.insert_data()
+
+        nrows = len(self.frame)
+
+        if nrows == 0:
+            return 0
+
+        if chunksize is None:
+            chunksize = nrows
+        elif chunksize == 0:
+            raise ValueError("chunksize argument should be non-zero")
+
+        chunks = (nrows // chunksize) + 1
+        total_inserted = None
+        with self.pd_sql.run_transaction() as conn:
+            for i in range(chunks):
+                start_i = i * chunksize
+                end_i = min((i + 1) * chunksize, nrows)
+                if start_i >= end_i:
+                    break
+
+                chunk_iter = zip(*(arr[start_i:end_i] for arr in data_list))
+                num_inserted = exec_insert(conn, keys, chunk_iter)
+                # GH 46891
+                if num_inserted is not None:
+                    if total_inserted is None:
+                        total_inserted = num_inserted
+                    else:
+                        total_inserted += num_inserted
+        return total_inserted
+
+    def _query_iterator(
+        self,
+        result,
+        exit_stack: ExitStack,
+        chunksize: int | None,
+        columns,
+        coerce_float: bool = True,
+        parse_dates=None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> Generator[DataFrame]:
+        """Return generator through chunked result set."""
+        has_read_data = False
+        with exit_stack:
+            while True:
+                data = result.fetchmany(chunksize)
+                if not data:
+                    if not has_read_data:
+                        yield DataFrame.from_records(
+                            [], columns=columns, coerce_float=coerce_float
+                        )
+                    break
+
+                has_read_data = True
+                self.frame = _convert_arrays_to_dataframe(
+                    data, columns, coerce_float, dtype_backend
+                )
+
+                self._harmonize_columns(
+                    parse_dates=parse_dates, dtype_backend=dtype_backend
+                )
+
+                if self.index is not None:
+                    self.frame.set_index(self.index, inplace=True)
+
+                yield self.frame
+
+    def read(
+        self,
+        exit_stack: ExitStack,
+        coerce_float: bool = True,
+        parse_dates=None,
+        columns=None,
+        chunksize: int | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        from sqlalchemy import select
+
+        if columns is not None and len(columns) > 0:
+            cols = [self.table.c[n] for n in columns]
+            if self.index is not None:
+                for idx in self.index[::-1]:
+                    cols.insert(0, self.table.c[idx])
+            sql_select = select(*cols)
+        else:
+            sql_select = select(self.table)
+        result = self.pd_sql.execute(sql_select)
+        column_names = result.keys()
+
+        if chunksize is not None:
+            return self._query_iterator(
+                result,
+                exit_stack,
+                chunksize,
+                column_names,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype_backend=dtype_backend,
+            )
+        else:
+            data = result.fetchall()
+            self.frame = _convert_arrays_to_dataframe(
+                data, column_names, coerce_float, dtype_backend
+            )
+
+            self._harmonize_columns(
+                parse_dates=parse_dates, dtype_backend=dtype_backend
+            )
+
+            if self.index is not None:
+                self.frame.set_index(self.index, inplace=True)
+
+            return self.frame
+
+    def _index_name(self, index, index_label):
+        # for writing: index=True to include index in sql table
+        if index is True:
+            nlevels = self.frame.index.nlevels
+            # if index_label is specified, set this as index name(s)
+            if index_label is not None:
+                if not isinstance(index_label, list):
+                    index_label = [index_label]
+                if len(index_label) != nlevels:
+                    raise ValueError(
+                        "Length of 'index_label' should match number of "
+                        f"levels, which is {nlevels}"
+                    )
+                return index_label
+            # return the used column labels for the index columns
+            if (
+                nlevels == 1
+                and "index" not in self.frame.columns
+                and self.frame.index.name is None
+            ):
+                return ["index"]
+            else:
+                return com.fill_missing_names(self.frame.index.names)
+
+        # for reading: index=(list of) string to specify column to set as index
+        elif isinstance(index, str):
+            return [index]
+        elif isinstance(index, list):
+            return index
+        else:
+            return None
+
+    def _get_column_names_and_types(self, dtype_mapper):
+        column_names_and_types = []
+        if self.index is not None:
+            for i, idx_label in enumerate(self.index):
+                idx_type = dtype_mapper(self.frame.index._get_level_values(i))
+                column_names_and_types.append((str(idx_label), idx_type, True))
+
+        column_names_and_types += [
+            (str(self.frame.columns[i]), dtype_mapper(self.frame.iloc[:, i]), False)
+            for i in range(len(self.frame.columns))
+        ]
+
+        return column_names_and_types
+
+    def _create_table_setup(self):
+        from sqlalchemy import (
+            Column,
+            PrimaryKeyConstraint,
+            Table,
+        )
+        from sqlalchemy.schema import MetaData
+
+        column_names_and_types = self._get_column_names_and_types(self._sqlalchemy_type)
+
+        columns: list[Any] = [
+            Column(name, typ, index=is_index)
+            for name, typ, is_index in column_names_and_types
+        ]
+
+        if self.keys is not None:
+            if not is_list_like(self.keys):
+                keys = [self.keys]
+            else:
+                keys = self.keys
+            pkc = PrimaryKeyConstraint(*keys, name=self.name + "_pk")
+            columns.append(pkc)
+
+        schema = self.schema or self.pd_sql.meta.schema
+
+        # At this point, attach to new metadata, only attach to self.meta
+        # once table is created.
+        meta = MetaData()
+        return Table(self.name, meta, *columns, schema=schema)
+
+    def _harmonize_columns(
+        self,
+        parse_dates=None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> None:
+        """
+        Make the DataFrame's column types align with the SQL table
+        column types.
+        Need to work around limited NA value support. Floats are always
+        fine, ints must always be floats if there are Null values.
+        Booleans are hard because converting bool column with None replaces
+        all Nones with false. Therefore only convert bool if there are no
+        NA values.
+        Datetimes should already be converted to np.datetime64 if supported,
+        but here we also force conversion if required.
+        """
+        parse_dates = _process_parse_dates_argument(parse_dates)
+
+        for sql_col in self.table.columns:
+            col_name = sql_col.name
+            try:
+                df_col = self.frame[col_name]
+
+                # Handle date parsing upfront; don't try to convert columns
+                # twice
+                if col_name in parse_dates:
+                    try:
+                        fmt = parse_dates[col_name]
+                    except TypeError:
+                        fmt = None
+                    self.frame[col_name] = _handle_date_column(df_col, format=fmt)
+                    continue
+
+                # the type the dataframe column should have
+                col_type = self._get_dtype(sql_col.type)
+
+                if (
+                    col_type is datetime
+                    or col_type is date
+                    or col_type is DatetimeTZDtype
+                ):
+                    # Convert tz-aware Datetime SQL columns to UTC
+                    utc = col_type is DatetimeTZDtype
+                    self.frame[col_name] = _handle_date_column(df_col, utc=utc)
+                elif dtype_backend == "numpy" and col_type is float:
+                    # floats support NA, can always convert!
+                    self.frame[col_name] = df_col.astype(col_type)
+                elif (
+                    using_string_dtype()
+                    and is_string_dtype(col_type)
+                    and is_object_dtype(self.frame[col_name])
+                ):
+                    self.frame[col_name] = df_col.astype(col_type)
+                elif dtype_backend == "numpy" and len(df_col) == df_col.count():
+                    # No NA values, can convert ints and bools
+                    if col_type is np.dtype("int64") or col_type is bool:
+                        self.frame[col_name] = df_col.astype(col_type)
+            except KeyError:
+                pass  # this column not in results
+
+    def _sqlalchemy_type(self, col: Index | Series):
+        dtype: DtypeArg = self.dtype or {}
+        if is_dict_like(dtype):
+            dtype = cast(dict, dtype)
+            if col.name in dtype:
+                return dtype[col.name]
+
+        # Infer type of column, while ignoring missing values.
+        # Needed for inserting typed data containing NULLs, GH 8778.
+        col_type = lib.infer_dtype(col, skipna=True)
+
+        from sqlalchemy.types import (
+            TIMESTAMP,
+            BigInteger,
+            Boolean,
+            Date,
+            DateTime,
+            Float,
+            Integer,
+            SmallInteger,
+            Text,
+            Time,
+        )
+
+        if col_type in ("datetime64", "datetime"):
+            # GH 9086: TIMESTAMP is the suggested type if the column contains
+            # timezone information
+            try:
+                # error: Item "Index" of "Union[Index, Series]" has no attribute "dt"
+                if col.dt.tz is not None:  # type: ignore[union-attr]
+                    return TIMESTAMP(timezone=True)
+            except AttributeError:
+                # The column is actually a DatetimeIndex
+                # GH 26761 or an Index with date-like data e.g. 9999-01-01
+                if getattr(col, "tz", None) is not None:
+                    return TIMESTAMP(timezone=True)
+            return DateTime
+        if col_type == "timedelta64":
+            warnings.warn(
+                "the 'timedelta' type is not supported, and will be "
+                "written as integer values (ns frequency) to the database.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+            return BigInteger
+        elif col_type == "floating":
+            if col.dtype == "float32":
+                return Float(precision=23)
+            else:
+                return Float(precision=53)
+        elif col_type == "integer":
+            # GH35076 Map pandas integer to optimal SQLAlchemy integer type
+            if col.dtype.name.lower() in ("int8", "uint8", "int16"):
+                return SmallInteger
+            elif col.dtype.name.lower() in ("uint16", "int32"):
+                return Integer
+            elif col.dtype.name.lower() == "uint64":
+                raise ValueError("Unsigned 64 bit integer datatype is not supported")
+            else:
+                return BigInteger
+        elif col_type == "boolean":
+            return Boolean
+        elif col_type == "date":
+            return Date
+        elif col_type == "time":
+            return Time
+        elif col_type == "complex":
+            raise ValueError("Complex datatypes not supported")
+
+        return Text
+
+    def _get_dtype(self, sqltype):
+        from sqlalchemy.types import (
+            TIMESTAMP,
+            Boolean,
+            Date,
+            DateTime,
+            Float,
+            Integer,
+            String,
+        )
+
+        if isinstance(sqltype, Float):
+            return float
+        elif isinstance(sqltype, Integer):
+            # TODO: Refine integer size.
+            return np.dtype("int64")
+        elif isinstance(sqltype, TIMESTAMP):
+            # we have a timezone capable type
+            if not sqltype.timezone:
+                return datetime
+            return DatetimeTZDtype
+        elif isinstance(sqltype, DateTime):
+            # Caution: np.datetime64 is also a subclass of np.number.
+            return datetime
+        elif isinstance(sqltype, Date):
+            return date
+        elif isinstance(sqltype, Boolean):
+            return bool
+        elif isinstance(sqltype, String):
+            if using_string_dtype():
+                return StringDtype(na_value=np.nan)
+
+        return object
+
+
+class PandasSQL(PandasObject, ABC):
+    """
+    Subclasses Should define read_query and to_sql.
+    """
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(self, *args) -> None:
+        pass
+
+    def read_table(
+        self,
+        table_name: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        columns=None,
+        schema: str | None = None,
+        chunksize: int | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        raise NotImplementedError
+
+    @abstractmethod
+    def read_query(
+        self,
+        sql: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        params=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        pass
+
+    @abstractmethod
+    def to_sql(
+        self,
+        frame,
+        name: str,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        index: bool = True,
+        index_label=None,
+        schema=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        method: Literal["multi"] | Callable | None = None,
+        engine: str = "auto",
+        **engine_kwargs,
+    ) -> int | None:
+        pass
+
+    @abstractmethod
+    def execute(self, sql: str | Select | TextClause, params=None):
+        pass
+
+    @abstractmethod
+    def has_table(self, name: str, schema: str | None = None) -> bool:
+        pass
+
+    @abstractmethod
+    def _create_sql_schema(
+        self,
+        frame: DataFrame,
+        table_name: str,
+        keys: list[str] | None = None,
+        dtype: DtypeArg | None = None,
+        schema: str | None = None,
+    ) -> str:
+        pass
+
+
+class BaseEngine:
+    def insert_records(
+        self,
+        table: SQLTable,
+        con,
+        frame,
+        name: str,
+        index: bool | str | list[str] | None = True,
+        schema=None,
+        chunksize: int | None = None,
+        method=None,
+        **engine_kwargs,
+    ) -> int | None:
+        """
+        Inserts data into already-prepared table
+        """
+        raise AbstractMethodError(self)
+
+
+class SQLAlchemyEngine(BaseEngine):
+    def __init__(self) -> None:
+        import_optional_dependency(
+            "sqlalchemy", extra="sqlalchemy is required for SQL support."
+        )
+
+    def insert_records(
+        self,
+        table: SQLTable,
+        con,
+        frame,
+        name: str,
+        index: bool | str | list[str] | None = True,
+        schema=None,
+        chunksize: int | None = None,
+        method=None,
+        **engine_kwargs,
+    ) -> int | None:
+        from sqlalchemy import exc
+
+        try:
+            return table.insert(chunksize=chunksize, method=method)
+        except exc.StatementError as err:
+            # GH34431
+            # https://stackoverflow.com/a/67358288/6067848
+            msg = r"""(\(1054, "Unknown column 'inf(e0)?' in 'field list'"\))(?#
+            )|inf can not be used with MySQL"""
+            err_text = str(err.orig)
+            if re.search(msg, err_text):
+                raise ValueError("inf cannot be used with MySQL") from err
+            raise err
+
+
+def get_engine(engine: str) -> BaseEngine:
+    """return our implementation"""
+    if engine == "auto":
+        engine = get_option("io.sql.engine")
+
+    if engine == "auto":
+        # try engines in this order
+        engine_classes = [SQLAlchemyEngine]
+
+        error_msgs = ""
+        for engine_class in engine_classes:
+            try:
+                return engine_class()
+            except ImportError as err:
+                error_msgs += "\n - " + str(err)
+
+        raise ImportError(
+            "Unable to find a usable engine; "
+            "tried using: 'sqlalchemy'.\n"
+            "A suitable version of "
+            "sqlalchemy is required for sql I/O "
+            "support.\n"
+            "Trying to import the above resulted in these errors:"
+            f"{error_msgs}"
+        )
+
+    if engine == "sqlalchemy":
+        return SQLAlchemyEngine()
+
+    raise ValueError("engine must be one of 'auto', 'sqlalchemy'")
+
+
+class SQLDatabase(PandasSQL):
+    """
+    This class enables conversion between DataFrame and SQL databases
+    using SQLAlchemy to handle DataBase abstraction.
+
+    Parameters
+    ----------
+    con : SQLAlchemy Connectable or URI string.
+        Connectable to connect with the database. Using SQLAlchemy makes it
+        possible to use any DB supported by that library.
+    schema : string, default None
+        Name of SQL schema in database to write to (if database flavor
+        supports this). If None, use default schema (default).
+    need_transaction : bool, default False
+        If True, SQLDatabase will create a transaction.
+
+    """
+
+    def __init__(
+        self, con, schema: str | None = None, need_transaction: bool = False
+    ) -> None:
+        from sqlalchemy import create_engine
+        from sqlalchemy.engine import Engine
+        from sqlalchemy.schema import MetaData
+
+        # self.exit_stack cleans up the Engine and Connection and commits the
+        # transaction if any of those objects was created below.
+        # Cleanup happens either in self.__exit__ or at the end of the iterator
+        # returned by read_sql when chunksize is not None.
+        self.exit_stack = ExitStack()
+        if isinstance(con, str):
+            con = create_engine(con)
+            self.exit_stack.callback(con.dispose)
+        if isinstance(con, Engine):
+            con = self.exit_stack.enter_context(con.connect())
+        if need_transaction and not con.in_transaction():
+            self.exit_stack.enter_context(con.begin())
+        self.con = con
+        self.meta = MetaData(schema=schema)
+        self.returns_generator = False
+
+    def __exit__(self, *args) -> None:
+        if not self.returns_generator:
+            self.exit_stack.close()
+
+    @contextmanager
+    def run_transaction(self):
+        if not self.con.in_transaction():
+            with self.con.begin():
+                yield self.con
+        else:
+            yield self.con
+
+    def execute(self, sql: str | Select | TextClause | Delete, params=None):
+        """Simple passthrough to SQLAlchemy connectable"""
+        from sqlalchemy.exc import SQLAlchemyError
+
+        args = [] if params is None else [params]
+        if isinstance(sql, str):
+            execute_function = self.con.exec_driver_sql
+        else:
+            execute_function = self.con.execute
+
+        try:
+            return execute_function(sql, *args)
+        except SQLAlchemyError as exc:
+            raise DatabaseError(f"Execution failed on sql '{sql}': {exc}") from exc
+
+    def read_table(
+        self,
+        table_name: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        columns=None,
+        schema: str | None = None,
+        chunksize: int | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        """
+        Read SQL database table into a DataFrame.
+
+        Parameters
+        ----------
+        table_name : str
+            Name of SQL table in database.
+        index_col : string, optional, default: None
+            Column to set as index.
+        coerce_float : bool, default True
+            Attempts to convert values of non-string, non-numeric objects
+            (like decimal.Decimal) to floating point. This can result in
+            loss of precision.
+        parse_dates : list or dict, default: None
+            - List of column names to parse as dates.
+            - Dict of ``{column_name: format string}`` where format string is
+              strftime compatible in case of parsing string times, or is one of
+              (D, s, ns, ms, us) in case of parsing integer timestamps.
+            - Dict of ``{column_name: arg}``, where the arg corresponds
+              to the keyword arguments of :func:`pandas.to_datetime`.
+              Especially useful with databases without native Datetime support,
+              such as SQLite.
+        columns : list, default: None
+            List of column names to select from SQL table.
+        schema : string, default None
+            Name of SQL schema in database to query (if database flavor
+            supports this).  If specified, this overwrites the default
+            schema of the SQL database object.
+        chunksize : int, default None
+            If specified, return an iterator where `chunksize` is the number
+            of rows to include in each chunk.
+        dtype_backend : {'numpy_nullable', 'pyarrow'}
+            Back-end data type applied to the resultant :class:`DataFrame`
+            (still experimental). If not specified, the default behavior
+            is to not use nullable data types. If specified, the behavior
+            is as follows:
+
+            * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+            * ``"pyarrow"``: returns pyarrow-backed nullable
+              :class:`ArrowDtype` :class:`DataFrame`
+
+            .. versionadded:: 2.0
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        pandas.read_sql_table
+        SQLDatabase.read_query
+
+        """
+        self.meta.reflect(bind=self.con, only=[table_name], views=True)
+        table = SQLTable(table_name, self, index=index_col, schema=schema)
+        if chunksize is not None:
+            self.returns_generator = True
+        return table.read(
+            self.exit_stack,
+            coerce_float=coerce_float,
+            parse_dates=parse_dates,
+            columns=columns,
+            chunksize=chunksize,
+            dtype_backend=dtype_backend,
+        )
+
+    @staticmethod
+    def _query_iterator(
+        result,
+        exit_stack: ExitStack,
+        chunksize: int,
+        columns,
+        index_col=None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> Generator[DataFrame]:
+        """Return generator through chunked result set"""
+        has_read_data = False
+        with exit_stack:
+            while True:
+                data = result.fetchmany(chunksize)
+                if not data:
+                    if not has_read_data:
+                        yield _wrap_result(
+                            [],
+                            columns,
+                            index_col=index_col,
+                            coerce_float=coerce_float,
+                            parse_dates=parse_dates,
+                            dtype=dtype,
+                            dtype_backend=dtype_backend,
+                        )
+                    break
+
+                has_read_data = True
+                yield _wrap_result(
+                    data,
+                    columns,
+                    index_col=index_col,
+                    coerce_float=coerce_float,
+                    parse_dates=parse_dates,
+                    dtype=dtype,
+                    dtype_backend=dtype_backend,
+                )
+
+    def read_query(
+        self,
+        sql: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        params=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        """
+        Read SQL query into a DataFrame.
+
+        Parameters
+        ----------
+        sql : str
+            SQL query to be executed.
+        index_col : string, optional, default: None
+            Column name to use as index for the returned DataFrame object.
+        coerce_float : bool, default True
+            Attempt to convert values of non-string, non-numeric objects (like
+            decimal.Decimal) to floating point, useful for SQL result sets.
+        params : list, tuple or dict, optional, default: None
+            List of parameters to pass to execute method.  The syntax used
+            to pass parameters is database driver dependent. Check your
+            database driver documentation for which of the five syntax styles,
+            described in PEP 249's paramstyle, is supported.
+            Eg. for psycopg2, uses %(name)s so use params={'name' : 'value'}
+        parse_dates : list or dict, default: None
+            - List of column names to parse as dates.
+            - Dict of ``{column_name: format string}`` where format string is
+              strftime compatible in case of parsing string times, or is one of
+              (D, s, ns, ms, us) in case of parsing integer timestamps.
+            - Dict of ``{column_name: arg dict}``, where the arg dict
+              corresponds to the keyword arguments of
+              :func:`pandas.to_datetime` Especially useful with databases
+              without native Datetime support, such as SQLite.
+        chunksize : int, default None
+            If specified, return an iterator where `chunksize` is the number
+            of rows to include in each chunk.
+        dtype : Type name or dict of columns
+            Data type for data or columns. E.g. np.float64 or
+            {'a': np.float64, 'b': np.int32, 'c': 'Int64'}
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        read_sql_table : Read SQL database table into a DataFrame.
+        read_sql
+
+        """
+        result = self.execute(sql, params)
+        columns = result.keys()
+
+        if chunksize is not None:
+            self.returns_generator = True
+            return self._query_iterator(
+                result,
+                self.exit_stack,
+                chunksize,
+                columns,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype=dtype,
+                dtype_backend=dtype_backend,
+            )
+        else:
+            data = result.fetchall()
+            frame = _wrap_result(
+                data,
+                columns,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype=dtype,
+                dtype_backend=dtype_backend,
+            )
+            return frame
+
+    read_sql = read_query
+
+    def prep_table(
+        self,
+        frame,
+        name: str,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        index: bool | str | list[str] | None = True,
+        index_label=None,
+        schema=None,
+        dtype: DtypeArg | None = None,
+    ) -> SQLTable:
+        """
+        Prepares table in the database for data insertion. Creates it if needed, etc.
+        """
+        if dtype:
+            if not is_dict_like(dtype):
+                # error: Value expression in dictionary comprehension has incompatible
+                # type "Union[ExtensionDtype, str, dtype[Any], Type[object],
+                # Dict[Hashable, Union[ExtensionDtype, Union[str, dtype[Any]],
+                # Type[str], Type[float], Type[int], Type[complex], Type[bool],
+                # Type[object]]]]"; expected type "Union[ExtensionDtype, str,
+                # dtype[Any], Type[object]]"
+                dtype = {col_name: dtype for col_name in frame}  # type: ignore[misc]
+            else:
+                dtype = cast(dict, dtype)
+
+            from sqlalchemy.types import TypeEngine
+
+            for col, my_type in dtype.items():
+                if isinstance(my_type, type) and issubclass(my_type, TypeEngine):
+                    pass
+                elif isinstance(my_type, TypeEngine):
+                    pass
+                else:
+                    raise ValueError(f"The type of {col} is not a SQLAlchemy type")
+
+        table = SQLTable(
+            name,
+            self,
+            frame=frame,
+            index=index,
+            if_exists=if_exists,
+            index_label=index_label,
+            schema=schema,
+            dtype=dtype,
+        )
+        table.create()
+        return table
+
+    def check_case_sensitive(
+        self,
+        name: str,
+        schema: str | None,
+    ) -> None:
+        """
+        Checks table name for issues with case-sensitivity.
+        Method is called after data is inserted.
+        """
+        if not name.isdigit() and not name.islower():
+            # check for potentially case sensitivity issues (GH7815)
+            # Only check when name is not a number and name is not lower case
+            from sqlalchemy import inspect as sqlalchemy_inspect
+
+            insp = sqlalchemy_inspect(self.con)
+            table_names = insp.get_table_names(schema=schema or self.meta.schema)
+            if name not in table_names:
+                msg = (
+                    f"The provided table name '{name}' is not found exactly as "
+                    "such in the database after writing the table, possibly "
+                    "due to case sensitivity issues. Consider using lower "
+                    "case table names."
+                )
+                warnings.warn(
+                    msg,
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+
+    def to_sql(
+        self,
+        frame,
+        name: str,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        index: bool = True,
+        index_label=None,
+        schema: str | None = None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        method: Literal["multi"] | Callable | None = None,
+        engine: str = "auto",
+        **engine_kwargs,
+    ) -> int | None:
+        """
+        Write records stored in a DataFrame to a SQL database.
+
+        Parameters
+        ----------
+        frame : DataFrame
+        name : string
+            Name of SQL table.
+        if_exists : {'fail', 'replace', 'append', 'delete_rows'}, default 'fail'
+            - fail: If table exists, do nothing.
+            - replace: If table exists, drop it, recreate it, and insert data.
+            - append: If table exists, insert data. Create if does not exist.
+            - delete_rows: If a table exists, delete all records and insert data.
+        index : boolean, default True
+            Write DataFrame index as a column.
+        index_label : string or sequence, default None
+            Column label for index column(s). If None is given (default) and
+            `index` is True, then the index names are used.
+            A sequence should be given if the DataFrame uses MultiIndex.
+        schema : string, default None
+            Name of SQL schema in database to write to (if database flavor
+            supports this). If specified, this overwrites the default
+            schema of the SQLDatabase object.
+        chunksize : int, default None
+            If not None, then rows will be written in batches of this size at a
+            time.  If None, all rows will be written at once.
+        dtype : single type or dict of column name to SQL type, default None
+            Optional specifying the datatype for columns. The SQL type should
+            be a SQLAlchemy type. If all columns are of the same type, one
+            single value can be used.
+        method : {None', 'multi', callable}, default None
+            Controls the SQL insertion clause used:
+
+            * None : Uses standard SQL ``INSERT`` clause (one per row).
+            * 'multi': Pass multiple values in a single ``INSERT`` clause.
+            * callable with signature ``(pd_table, conn, keys, data_iter)``.
+
+            Details and a sample callable implementation can be found in the
+            section :ref:`insert method <io.sql.method>`.
+        engine : {'auto', 'sqlalchemy'}, default 'auto'
+            SQL engine library to use. If 'auto', then the option
+            ``io.sql.engine`` is used. The default ``io.sql.engine``
+            behavior is 'sqlalchemy'
+
+            .. versionadded:: 1.3.0
+
+        **engine_kwargs
+            Any additional kwargs are passed to the engine.
+        """
+        sql_engine = get_engine(engine)
+
+        table = self.prep_table(
+            frame=frame,
+            name=name,
+            if_exists=if_exists,
+            index=index,
+            index_label=index_label,
+            schema=schema,
+            dtype=dtype,
+        )
+
+        total_inserted = sql_engine.insert_records(
+            table=table,
+            con=self.con,
+            frame=frame,
+            name=name,
+            index=index,
+            schema=schema,
+            chunksize=chunksize,
+            method=method,
+            **engine_kwargs,
+        )
+
+        self.check_case_sensitive(name=name, schema=schema)
+        return total_inserted
+
+    @property
+    def tables(self):
+        return self.meta.tables
+
+    def has_table(self, name: str, schema: str | None = None) -> bool:
+        from sqlalchemy import inspect as sqlalchemy_inspect
+
+        insp = sqlalchemy_inspect(self.con)
+        return insp.has_table(name, schema or self.meta.schema)
+
+    def get_table(self, table_name: str, schema: str | None = None) -> Table:
+        from sqlalchemy import (
+            Numeric,
+            Table,
+        )
+
+        schema = schema or self.meta.schema
+        tbl = Table(table_name, self.meta, autoload_with=self.con, schema=schema)
+        for column in tbl.columns:
+            if isinstance(column.type, Numeric):
+                column.type.asdecimal = False
+        return tbl
+
+    def drop_table(self, table_name: str, schema: str | None = None) -> None:
+        schema = schema or self.meta.schema
+        if self.has_table(table_name, schema):
+            self.meta.reflect(
+                bind=self.con, only=[table_name], schema=schema, views=True
+            )
+            with self.run_transaction():
+                self.get_table(table_name, schema).drop(bind=self.con)
+            self.meta.clear()
+
+    def delete_rows(self, table_name: str, schema: str | None = None) -> None:
+        schema = schema or self.meta.schema
+        if self.has_table(table_name, schema):
+            self.meta.reflect(
+                bind=self.con, only=[table_name], schema=schema, views=True
+            )
+            table = self.get_table(table_name, schema)
+            self.execute(table.delete()).close()
+            self.meta.clear()
+
+    def _create_sql_schema(
+        self,
+        frame: DataFrame,
+        table_name: str,
+        keys: list[str] | None = None,
+        dtype: DtypeArg | None = None,
+        schema: str | None = None,
+    ) -> str:
+        table = SQLTable(
+            table_name,
+            self,
+            frame=frame,
+            index=False,
+            keys=keys,
+            dtype=dtype,
+            schema=schema,
+        )
+        return str(table.sql_schema())
+
+
+# ---- SQL without SQLAlchemy ---
+
+
+class ADBCDatabase(PandasSQL):
+    """
+    This class enables conversion between DataFrame and SQL databases
+    using ADBC to handle DataBase abstraction.
+
+    Parameters
+    ----------
+    con : adbc_driver_manager.dbapi.Connection
+    """
+
+    def __init__(self, con) -> None:
+        self.con = con
+
+    @contextmanager
+    def run_transaction(self):
+        with self.con.cursor() as cur:
+            try:
+                yield cur
+            except Exception:
+                self.con.rollback()
+                raise
+            self.con.commit()
+
+    def execute(self, sql: str | Select | TextClause, params=None):
+        from adbc_driver_manager import Error
+
+        if not isinstance(sql, str):
+            raise TypeError("Query must be a string unless using sqlalchemy.")
+        args = [] if params is None else [params]
+        cur = self.con.cursor()
+        try:
+            cur.execute(sql, *args)
+            return cur
+        except Error as exc:
+            try:
+                self.con.rollback()
+            except Error as inner_exc:  # pragma: no cover
+                ex = DatabaseError(
+                    f"Execution failed on sql: {sql}\n{exc}\nunable to rollback"
+                )
+                raise ex from inner_exc
+
+            ex = DatabaseError(f"Execution failed on sql '{sql}': {exc}")
+            raise ex from exc
+
+    def read_table(
+        self,
+        table_name: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        columns=None,
+        schema: str | None = None,
+        chunksize: int | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        """
+        Read SQL database table into a DataFrame.
+
+        Parameters
+        ----------
+        table_name : str
+            Name of SQL table in database.
+        coerce_float : bool, default True
+            Raises NotImplementedError
+        parse_dates : list or dict, default: None
+            - List of column names to parse as dates.
+            - Dict of ``{column_name: format string}`` where format string is
+              strftime compatible in case of parsing string times, or is one of
+              (D, s, ns, ms, us) in case of parsing integer timestamps.
+            - Dict of ``{column_name: arg}``, where the arg corresponds
+              to the keyword arguments of :func:`pandas.to_datetime`.
+              Especially useful with databases without native Datetime support,
+              such as SQLite.
+        columns : list, default: None
+            List of column names to select from SQL table.
+        schema : string, default None
+            Name of SQL schema in database to query (if database flavor
+            supports this).  If specified, this overwrites the default
+            schema of the SQL database object.
+        chunksize : int, default None
+            Raises NotImplementedError
+        dtype_backend : {'numpy_nullable', 'pyarrow'}
+            Back-end data type applied to the resultant :class:`DataFrame`
+            (still experimental). If not specified, the default behavior
+            is to not use nullable data types. If specified, the behavior
+            is as follows:
+
+            * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+            * ``"pyarrow"``: returns pyarrow-backed nullable
+              :class:`ArrowDtype` :class:`DataFrame`
+
+            .. versionadded:: 2.0
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        pandas.read_sql_table
+        SQLDatabase.read_query
+
+        """
+        if coerce_float is not True:
+            raise NotImplementedError(
+                "'coerce_float' is not implemented for ADBC drivers"
+            )
+        if chunksize:
+            raise NotImplementedError("'chunksize' is not implemented for ADBC drivers")
+
+        if columns:
+            if index_col:
+                index_select = maybe_make_list(index_col)
+            else:
+                index_select = []
+            to_select = index_select + columns
+            select_list = ", ".join(f'"{x}"' for x in to_select)
+        else:
+            select_list = "*"
+        if schema:
+            stmt = f"SELECT {select_list} FROM {schema}.{table_name}"
+        else:
+            stmt = f"SELECT {select_list} FROM {table_name}"
+
+        with self.execute(stmt) as cur:
+            pa_table = cur.fetch_arrow_table()
+            df = arrow_table_to_pandas(pa_table, dtype_backend=dtype_backend)
+
+        return _wrap_result_adbc(
+            df,
+            index_col=index_col,
+            parse_dates=parse_dates,
+        )
+
+    def read_query(
+        self,
+        sql: str,
+        index_col: str | list[str] | None = None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        params=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        """
+        Read SQL query into a DataFrame.
+
+        Parameters
+        ----------
+        sql : str
+            SQL query to be executed.
+        index_col : string, optional, default: None
+            Column name to use as index for the returned DataFrame object.
+        coerce_float : bool, default True
+            Raises NotImplementedError
+        params : list, tuple or dict, optional, default: None
+            Raises NotImplementedError
+        parse_dates : list or dict, default: None
+            - List of column names to parse as dates.
+            - Dict of ``{column_name: format string}`` where format string is
+              strftime compatible in case of parsing string times, or is one of
+              (D, s, ns, ms, us) in case of parsing integer timestamps.
+            - Dict of ``{column_name: arg dict}``, where the arg dict
+              corresponds to the keyword arguments of
+              :func:`pandas.to_datetime` Especially useful with databases
+              without native Datetime support, such as SQLite.
+        chunksize : int, default None
+            Raises NotImplementedError
+        dtype : Type name or dict of columns
+            Data type for data or columns. E.g. np.float64 or
+            {'a': np.float64, 'b': np.int32, 'c': 'Int64'}
+
+            .. versionadded:: 1.3.0
+
+        Returns
+        -------
+        DataFrame
+
+        See Also
+        --------
+        read_sql_table : Read SQL database table into a DataFrame.
+        read_sql
+
+        """
+        if coerce_float is not True:
+            raise NotImplementedError(
+                "'coerce_float' is not implemented for ADBC drivers"
+            )
+        if params:
+            raise NotImplementedError("'params' is not implemented for ADBC drivers")
+        if chunksize:
+            raise NotImplementedError("'chunksize' is not implemented for ADBC drivers")
+
+        with self.execute(sql) as cur:
+            pa_table = cur.fetch_arrow_table()
+            df = arrow_table_to_pandas(pa_table, dtype_backend=dtype_backend)
+
+        return _wrap_result_adbc(
+            df,
+            index_col=index_col,
+            parse_dates=parse_dates,
+            dtype=dtype,
+        )
+
+    read_sql = read_query
+
+    def to_sql(
+        self,
+        frame,
+        name: str,
+        if_exists: Literal["fail", "replace", "append", "delete_rows"] = "fail",
+        index: bool = True,
+        index_label=None,
+        schema: str | None = None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        method: Literal["multi"] | Callable | None = None,
+        engine: str = "auto",
+        **engine_kwargs,
+    ) -> int | None:
+        """
+        Write records stored in a DataFrame to a SQL database.
+
+        Parameters
+        ----------
+        frame : DataFrame
+        name : string
+            Name of SQL table.
+        if_exists : {'fail', 'replace', 'append'}, default 'fail'
+            - fail: If table exists, do nothing.
+            - replace: If table exists, drop it, recreate it, and insert data.
+            - append: If table exists, insert data. Create if does not exist.
+            - delete_rows: If a table exists, delete all records and insert data.
+        index : boolean, default True
+            Write DataFrame index as a column.
+        index_label : string or sequence, default None
+            Raises NotImplementedError
+        schema : string, default None
+            Name of SQL schema in database to write to (if database flavor
+            supports this). If specified, this overwrites the default
+            schema of the SQLDatabase object.
+        chunksize : int, default None
+            Raises NotImplementedError
+        dtype : single type or dict of column name to SQL type, default None
+            Raises NotImplementedError
+        method : {None', 'multi', callable}, default None
+            Raises NotImplementedError
+        engine : {'auto', 'sqlalchemy'}, default 'auto'
+            Raises NotImplementedError if not set to 'auto'
+        """
+        pa = import_optional_dependency("pyarrow")
+        from adbc_driver_manager import Error
+
+        if index_label:
+            raise NotImplementedError(
+                "'index_label' is not implemented for ADBC drivers"
+            )
+        if chunksize:
+            raise NotImplementedError("'chunksize' is not implemented for ADBC drivers")
+        if dtype:
+            raise NotImplementedError("'dtype' is not implemented for ADBC drivers")
+        if method:
+            raise NotImplementedError("'method' is not implemented for ADBC drivers")
+        if engine != "auto":
+            raise NotImplementedError(
+                "engine != 'auto' not implemented for ADBC drivers"
+            )
+
+        if schema:
+            table_name = f"{schema}.{name}"
+        else:
+            table_name = name
+
+        # pandas if_exists="append" will still create the
+        # table if it does not exist; ADBC is more explicit with append/create
+        # as applicable modes, so the semantics get blurred across
+        # the libraries
+        mode = "create"
+        if self.has_table(name, schema):
+            if if_exists == "fail":
+                raise ValueError(f"Table '{table_name}' already exists.")
+            elif if_exists == "replace":
+                sql_statement = f"DROP TABLE {table_name}"
+                self.execute(sql_statement).close()
+            elif if_exists == "append":
+                mode = "append"
+            elif if_exists == "delete_rows":
+                mode = "append"
+                self.delete_rows(name, schema)
+
+        try:
+            tbl = pa.Table.from_pandas(frame, preserve_index=index)
+        except pa.ArrowNotImplementedError as exc:
+            raise ValueError("datatypes not supported") from exc
+
+        with self.con.cursor() as cur:
+            try:
+                total_inserted = cur.adbc_ingest(
+                    table_name=name, data=tbl, mode=mode, db_schema_name=schema
+                )
+            except Error as exc:
+                raise DatabaseError(
+                    f"Failed to insert records on table={name} with {mode=}"
+                ) from exc
+
+        self.con.commit()
+        return total_inserted
+
+    def has_table(self, name: str, schema: str | None = None) -> bool:
+        meta = self.con.adbc_get_objects(
+            db_schema_filter=schema, table_name_filter=name
+        ).read_all()
+
+        for catalog_schema in meta["catalog_db_schemas"].to_pylist():
+            if not catalog_schema:
+                continue
+            for schema_record in catalog_schema:
+                if not schema_record:
+                    continue
+
+                for table_record in schema_record["db_schema_tables"]:
+                    if table_record["table_name"] == name:
+                        return True
+
+        return False
+
+    def delete_rows(self, name: str, schema: str | None = None) -> None:
+        table_name = f"{schema}.{name}" if schema else name
+        if self.has_table(name, schema):
+            self.execute(f"DELETE FROM {table_name}").close()
+
+    def _create_sql_schema(
+        self,
+        frame: DataFrame,
+        table_name: str,
+        keys: list[str] | None = None,
+        dtype: DtypeArg | None = None,
+        schema: str | None = None,
+    ) -> str:
+        raise NotImplementedError("not implemented for adbc")
+
+
+# sqlite-specific sql strings and handler class
+# dictionary used for readability purposes
+_SQL_TYPES = {
+    "string": "TEXT",
+    "floating": "REAL",
+    "integer": "INTEGER",
+    "datetime": "TIMESTAMP",
+    "date": "DATE",
+    "time": "TIME",
+    "boolean": "INTEGER",
+}
+
+
+def _get_unicode_name(name: object) -> str:
+    try:
+        uname = str(name).encode("utf-8", "strict").decode("utf-8")
+    except UnicodeError as err:
+        raise ValueError(f"Cannot convert identifier to UTF-8: '{name}'") from err
+    return uname
+
+
+def _get_valid_sqlite_name(name: object) -> str:
+    # See https://stackoverflow.com/questions/6514274/how-do-you-escape-strings\
+    # -for-sqlite-table-column-names-in-python
+    # Ensure the string can be encoded as UTF-8.
+    # Ensure the string does not include any NUL characters.
+    # Replace all " with "".
+    # Wrap the entire thing in double quotes.
+
+    uname = _get_unicode_name(name)
+    if not len(uname):
+        raise ValueError("Empty table or column name specified")
+
+    nul_index = uname.find("\x00")
+    if nul_index >= 0:
+        raise ValueError("SQLite identifier cannot contain NULs")
+    return '"' + uname.replace('"', '""') + '"'
+
+
+class SQLiteTable(SQLTable):
+    """
+    Patch the SQLTable for fallback support.
+    Instead of a table variable just use the Create Table statement.
+    """
+
+    def __init__(self, *args, **kwargs) -> None:
+        super().__init__(*args, **kwargs)
+
+        self._register_date_adapters()
+
+    def _register_date_adapters(self) -> None:
+        # GH 8341
+        # register an adapter callable for datetime.time object
+        import sqlite3
+
+        # this will transform time(12,34,56,789) into '12:34:56.000789'
+        # (this is what sqlalchemy does)
+        def _adapt_time(t) -> str:
+            # This is faster than strftime
+            return f"{t.hour:02d}:{t.minute:02d}:{t.second:02d}.{t.microsecond:06d}"
+
+        # Also register adapters for date/datetime and co
+        # xref https://docs.python.org/3.12/library/sqlite3.html#adapter-and-converter-recipes
+        # Python 3.12+ doesn't auto-register adapters for us anymore
+
+        adapt_date_iso = lambda val: val.isoformat()
+        adapt_datetime_iso = lambda val: val.isoformat(" ")
+
+        sqlite3.register_adapter(time, _adapt_time)
+
+        sqlite3.register_adapter(date, adapt_date_iso)
+        sqlite3.register_adapter(datetime, adapt_datetime_iso)
+
+        convert_date = lambda val: date.fromisoformat(val.decode())
+        convert_timestamp = lambda val: datetime.fromisoformat(val.decode())
+
+        sqlite3.register_converter("date", convert_date)
+        sqlite3.register_converter("timestamp", convert_timestamp)
+
+    def sql_schema(self) -> str:
+        return str(";\n".join(self.table))
+
+    def _execute_create(self) -> None:
+        with self.pd_sql.run_transaction() as cur:
+            for stmt in self.table:
+                cur.execute(stmt)
+
+    def insert_statement(self, *, num_rows: int) -> str:
+        names = list(map(str, self.frame.columns))
+        wld = "?"  # wildcard char
+        escape = _get_valid_sqlite_name
+
+        if self.index is not None:
+            for idx in self.index[::-1]:
+                names.insert(0, idx)
+
+        bracketed_names = [escape(column) for column in names]
+        col_names = ",".join(bracketed_names)
+
+        row_wildcards = ",".join([wld] * len(names))
+        wildcards = ",".join([f"({row_wildcards})" for _ in range(num_rows)])
+        insert_statement = (
+            f"INSERT INTO {escape(self.name)} ({col_names}) VALUES {wildcards}"
+        )
+        return insert_statement
+
+    def _execute_insert(self, conn, keys, data_iter) -> int:
+        from sqlite3 import Error
+
+        data_list = list(data_iter)
+        try:
+            conn.executemany(self.insert_statement(num_rows=1), data_list)
+        except Error as exc:
+            raise DatabaseError("Execution failed") from exc
+        return conn.rowcount
+
+    def _execute_insert_multi(self, conn, keys, data_iter) -> int:
+        data_list = list(data_iter)
+        flattened_data = [x for row in data_list for x in row]
+        conn.execute(self.insert_statement(num_rows=len(data_list)), flattened_data)
+        return conn.rowcount
+
+    def _create_table_setup(self):
+        """
+        Return a list of SQL statements that creates a table reflecting the
+        structure of a DataFrame.  The first entry will be a CREATE TABLE
+        statement while the rest will be CREATE INDEX statements.
+        """
+        column_names_and_types = self._get_column_names_and_types(self._sql_type_name)
+        escape = _get_valid_sqlite_name
+
+        create_tbl_stmts = [
+            escape(cname) + " " + ctype for cname, ctype, _ in column_names_and_types
+        ]
+
+        if self.keys is not None and len(self.keys):
+            if not is_list_like(self.keys):
+                keys = [self.keys]
+            else:
+                keys = self.keys
+            cnames_br = ", ".join([escape(c) for c in keys])
+            create_tbl_stmts.append(
+                f"CONSTRAINT {self.name}_pk PRIMARY KEY ({cnames_br})"
+            )
+        if self.schema:
+            schema_name = self.schema + "."
+        else:
+            schema_name = ""
+        create_stmts = [
+            "CREATE TABLE "
+            + schema_name
+            + escape(self.name)
+            + " (\n"
+            + ",\n  ".join(create_tbl_stmts)
+            + "\n)"
+        ]
+
+        ix_cols = [cname for cname, _, is_index in column_names_and_types if is_index]
+        if len(ix_cols):
+            cnames = "_".join(ix_cols)
+            cnames_br = ",".join([escape(c) for c in ix_cols])
+            create_stmts.append(
+                "CREATE INDEX "
+                + escape("ix_" + self.name + "_" + cnames)
+                + "ON "
+                + escape(self.name)
+                + " ("
+                + cnames_br
+                + ")"
+            )
+
+        return create_stmts
+
+    def _sql_type_name(self, col):
+        dtype: DtypeArg = self.dtype or {}
+        if is_dict_like(dtype):
+            dtype = cast(dict, dtype)
+            if col.name in dtype:
+                return dtype[col.name]
+
+        # Infer type of column, while ignoring missing values.
+        # Needed for inserting typed data containing NULLs, GH 8778.
+        col_type = lib.infer_dtype(col, skipna=True)
+
+        if col_type == "timedelta64":
+            warnings.warn(
+                "the 'timedelta' type is not supported, and will be "
+                "written as integer values (ns frequency) to the database.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+            col_type = "integer"
+
+        elif col_type == "datetime64":
+            col_type = "datetime"
+
+        elif col_type == "empty":
+            col_type = "string"
+
+        elif col_type == "complex":
+            raise ValueError("Complex datatypes not supported")
+
+        if col_type not in _SQL_TYPES:
+            col_type = "string"
+
+        return _SQL_TYPES[col_type]
+
+
+class SQLiteDatabase(PandasSQL):
+    """
+    Version of SQLDatabase to support SQLite connections (fallback without
+    SQLAlchemy). This should only be used internally.
+
+    Parameters
+    ----------
+    con : sqlite connection object
+
+    """
+
+    def __init__(self, con) -> None:
+        self.con = con
+
+    @contextmanager
+    def run_transaction(self):
+        cur = self.con.cursor()
+        try:
+            yield cur
+            self.con.commit()
+        except Exception:
+            self.con.rollback()
+            raise
+        finally:
+            cur.close()
+
+    def execute(self, sql: str | Select | TextClause, params=None):
+        from sqlite3 import Error
+
+        if not isinstance(sql, str):
+            raise TypeError("Query must be a string unless using sqlalchemy.")
+        args = [] if params is None else [params]
+        cur = self.con.cursor()
+        try:
+            cur.execute(sql, *args)
+            return cur
+        except Error as exc:
+            try:
+                self.con.rollback()
+            except Error as inner_exc:  # pragma: no cover
+                ex = DatabaseError(
+                    f"Execution failed on sql: {sql}\n{exc}\nunable to rollback"
+                )
+                raise ex from inner_exc
+
+            ex = DatabaseError(f"Execution failed on sql '{sql}': {exc}")
+            raise ex from exc
+
+    @staticmethod
+    def _query_iterator(
+        cursor,
+        chunksize: int,
+        columns,
+        index_col=None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> Generator[DataFrame]:
+        """Return generator through chunked result set"""
+        has_read_data = False
+        while True:
+            data = cursor.fetchmany(chunksize)
+            if type(data) == tuple:
+                data = list(data)
+            if not data:
+                cursor.close()
+                if not has_read_data:
+                    result = DataFrame.from_records(
+                        [], columns=columns, coerce_float=coerce_float
+                    )
+                    if dtype:
+                        result = result.astype(dtype)
+                    yield result
+                break
+
+            has_read_data = True
+            yield _wrap_result(
+                data,
+                columns,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype=dtype,
+                dtype_backend=dtype_backend,
+            )
+
+    def read_query(
+        self,
+        sql,
+        index_col=None,
+        coerce_float: bool = True,
+        parse_dates=None,
+        params=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        dtype_backend: DtypeBackend | Literal["numpy"] = "numpy",
+    ) -> DataFrame | Iterator[DataFrame]:
+        cursor = self.execute(sql, params)
+        columns = [col_desc[0] for col_desc in cursor.description]
+
+        if chunksize is not None:
+            return self._query_iterator(
+                cursor,
+                chunksize,
+                columns,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype=dtype,
+                dtype_backend=dtype_backend,
+            )
+        else:
+            data = self._fetchall_as_list(cursor)
+            cursor.close()
+
+            frame = _wrap_result(
+                data,
+                columns,
+                index_col=index_col,
+                coerce_float=coerce_float,
+                parse_dates=parse_dates,
+                dtype=dtype,
+                dtype_backend=dtype_backend,
+            )
+            return frame
+
+    def _fetchall_as_list(self, cur):
+        result = cur.fetchall()
+        if not isinstance(result, list):
+            result = list(result)
+        return result
+
+    def to_sql(
+        self,
+        frame,
+        name: str,
+        if_exists: str = "fail",
+        index: bool = True,
+        index_label=None,
+        schema=None,
+        chunksize: int | None = None,
+        dtype: DtypeArg | None = None,
+        method: Literal["multi"] | Callable | None = None,
+        engine: str = "auto",
+        **engine_kwargs,
+    ) -> int | None:
+        """
+        Write records stored in a DataFrame to a SQL database.
+
+        Parameters
+        ----------
+        frame: DataFrame
+        name: string
+            Name of SQL table.
+        if_exists: {'fail', 'replace', 'append', 'delete_rows'}, default 'fail'
+            fail: If table exists, do nothing.
+            replace: If table exists, drop it, recreate it, and insert data.
+            append: If table exists, insert data. Create if it does not exist.
+            delete_rows: If a table exists, delete all records and insert data.
+        index : bool, default True
+            Write DataFrame index as a column
+        index_label : string or sequence, default None
+            Column label for index column(s). If None is given (default) and
+            `index` is True, then the index names are used.
+            A sequence should be given if the DataFrame uses MultiIndex.
+        schema : string, default None
+            Ignored parameter included for compatibility with SQLAlchemy
+            version of ``to_sql``.
+        chunksize : int, default None
+            If not None, then rows will be written in batches of this
+            size at a time. If None, all rows will be written at once.
+        dtype : single type or dict of column name to SQL type, default None
+            Optional specifying the datatype for columns. The SQL type should
+            be a string. If all columns are of the same type, one single value
+            can be used.
+        method : {None, 'multi', callable}, default None
+            Controls the SQL insertion clause used:
+
+            * None : Uses standard SQL ``INSERT`` clause (one per row).
+            * 'multi': Pass multiple values in a single ``INSERT`` clause.
+            * callable with signature ``(pd_table, conn, keys, data_iter)``.
+
+            Details and a sample callable implementation can be found in the
+            section :ref:`insert method <io.sql.method>`.
+        """
+        if dtype:
+            if not is_dict_like(dtype):
+                # error: Value expression in dictionary comprehension has incompatible
+                # type "Union[ExtensionDtype, str, dtype[Any], Type[object],
+                # Dict[Hashable, Union[ExtensionDtype, Union[str, dtype[Any]],
+                # Type[str], Type[float], Type[int], Type[complex], Type[bool],
+                # Type[object]]]]"; expected type "Union[ExtensionDtype, str,
+                # dtype[Any], Type[object]]"
+                dtype = {col_name: dtype for col_name in frame}  # type: ignore[misc]
+            else:
+                dtype = cast(dict, dtype)
+
+            for col, my_type in dtype.items():
+                if not isinstance(my_type, str):
+                    raise ValueError(f"{col} ({my_type}) not a string")
+
+        table = SQLiteTable(
+            name,
+            self,
+            frame=frame,
+            index=index,
+            if_exists=if_exists,
+            index_label=index_label,
+            dtype=dtype,
+        )
+        table.create()
+        return table.insert(chunksize, method)
+
+    def has_table(self, name: str, schema: str | None = None) -> bool:
+        wld = "?"
+        query = f"""
+        SELECT
+            name
+        FROM
+            sqlite_master
+        WHERE
+            type IN ('table', 'view')
+            AND name={wld};
+        """
+
+        return len(self.execute(query, [name]).fetchall()) > 0
+
+    def get_table(self, table_name: str, schema: str | None = None) -> None:
+        return None  # not supported in fallback mode
+
+    def drop_table(self, name: str, schema: str | None = None) -> None:
+        drop_sql = f"DROP TABLE {_get_valid_sqlite_name(name)}"
+        self.execute(drop_sql).close()
+
+    def delete_rows(self, name: str, schema: str | None = None) -> None:
+        delete_sql = f"DELETE FROM {_get_valid_sqlite_name(name)}"
+        if self.has_table(name, schema):
+            self.execute(delete_sql).close()
+
+    def _create_sql_schema(
+        self,
+        frame,
+        table_name: str,
+        keys=None,
+        dtype: DtypeArg | None = None,
+        schema: str | None = None,
+    ) -> str:
+        table = SQLiteTable(
+            table_name,
+            self,
+            frame=frame,
+            index=False,
+            keys=keys,
+            dtype=dtype,
+            schema=schema,
+        )
+        return str(table.sql_schema())
+
+
+def get_schema(
+    frame,
+    name: str,
+    keys=None,
+    con=None,
+    dtype: DtypeArg | None = None,
+    schema: str | None = None,
+) -> str:
+    """
+    Get the SQL db table schema for the given frame.
+
+    Parameters
+    ----------
+    frame : DataFrame
+    name : str
+        name of SQL table
+    keys : string or sequence, default: None
+        columns to use a primary key
+    con: ADBC Connection, SQLAlchemy connectable, sqlite3 connection, default: None
+        ADBC provides high performance I/O with native type support, where available.
+        Using SQLAlchemy makes it possible to use any DB supported by that
+        library
+        If a DBAPI2 object, only sqlite3 is supported.
+    dtype : dict of column name to SQL type, default None
+        Optional specifying the datatype for columns. The SQL type should
+        be a SQLAlchemy type, or a string for sqlite3 fallback connection.
+    schema: str, default: None
+        Optional specifying the schema to be used in creating the table.
+    """
+    with pandasSQL_builder(con=con) as pandas_sql:
+        return pandas_sql._create_sql_schema(
+            frame, name, keys=keys, dtype=dtype, schema=schema
+        )
diff --git a/pandas/pandas/io/stata.py b/pandas/pandas/io/stata.py
new file mode 100644
index 0000000000000..34d95fb59a21c
--- /dev/null
+++ b/pandas/pandas/io/stata.py
@@ -0,0 +1,3841 @@
+"""
+Module contains tools for processing Stata files into DataFrames
+
+The StataReader below was originally written by Joe Presbrey as part of PyDTA.
+It has been extended and improved by Skipper Seabold from the Statsmodels
+project who also developed the StataWriter and was finally added to pandas in
+a once again improved version.
+
+You can find more information on http://presbrey.mit.edu/PyDTA and
+https://www.statsmodels.org/devel/
+"""
+
+from __future__ import annotations
+
+from collections import abc
+from datetime import (
+    datetime,
+    timedelta,
+)
+from io import BytesIO
+import os
+import struct
+import sys
+from typing import (
+    IO,
+    TYPE_CHECKING,
+    AnyStr,
+    Final,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.lib import infer_dtype
+from pandas._libs.writers import max_len_string_array
+from pandas.errors import (
+    CategoricalConversionWarning,
+    InvalidColumnName,
+    PossiblePrecisionLoss,
+    ValueLabelTypeMismatch,
+)
+from pandas.util._decorators import (
+    Appender,
+    doc,
+)
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.common import (
+    ensure_object,
+    is_numeric_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+from pandas import (
+    Categorical,
+    DatetimeIndex,
+    NaT,
+    Timestamp,
+    isna,
+    to_datetime,
+)
+from pandas.core.frame import DataFrame
+from pandas.core.indexes.base import Index
+from pandas.core.indexes.range import RangeIndex
+from pandas.core.series import Series
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import get_handle
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Sequence,
+    )
+    from types import TracebackType
+    from typing import Literal
+
+    from pandas._typing import (
+        CompressionOptions,
+        FilePath,
+        ReadBuffer,
+        Self,
+        StorageOptions,
+        WriteBuffer,
+    )
+
+_version_error = (
+    "Version of given Stata file is {version}. pandas supports importing "
+    "versions 102, 103, 104, 105, 108, 110 (Stata 7), 111 (Stata 7SE),  "
+    "113 (Stata 8/9), 114 (Stata 10/11), 115 (Stata 12), 117 (Stata 13), "
+    "118 (Stata 14/15/16), and 119 (Stata 15/16, over 32,767 variables)."
+)
+
+_statafile_processing_params1 = """\
+convert_dates : bool, default True
+    Convert date variables to DataFrame time values.
+convert_categoricals : bool, default True
+    Read value labels and convert columns to Categorical/Factor variables."""
+
+_statafile_processing_params2 = """\
+index_col : str, optional
+    Column to set as index.
+convert_missing : bool, default False
+    Flag indicating whether to convert missing values to their Stata
+    representations.  If False, missing values are replaced with nan.
+    If True, columns containing missing values are returned with
+    object data types and missing values are represented by
+    StataMissingValue objects.
+preserve_dtypes : bool, default True
+    Preserve Stata datatypes. If False, numeric data are upcast to pandas
+    default types for foreign data (float64 or int64).
+columns : list or None
+    Columns to retain.  Columns will be returned in the given order.  None
+    returns all columns.
+order_categoricals : bool, default True
+    Flag indicating whether converted categorical data are ordered."""
+
+_chunksize_params = """\
+chunksize : int, default None
+    Return StataReader object for iterations, returns chunks with
+    given number of lines."""
+
+_iterator_params = """\
+iterator : bool, default False
+    Return StataReader object."""
+
+_reader_notes = """\
+Notes
+-----
+Categorical variables read through an iterator may not have the same
+categories and dtype. This occurs when  a variable stored in a DTA
+file is associated to an incomplete set of value labels that only
+label a strict subset of the values."""
+
+_read_stata_doc = f"""
+Read Stata file into DataFrame.
+
+Parameters
+----------
+filepath_or_buffer : str, path object or file-like object
+    Any valid string path is acceptable. The string could be a URL. Valid
+    URL schemes include http, ftp, s3, and file. For file URLs, a host is
+    expected. A local file could be: ``file://localhost/path/to/table.dta``.
+
+    If you want to pass in a path object, pandas accepts any ``os.PathLike``.
+
+    By file-like object, we refer to objects with a ``read()`` method,
+    such as a file handle (e.g. via builtin ``open`` function)
+    or ``StringIO``.
+{_statafile_processing_params1}
+{_statafile_processing_params2}
+{_chunksize_params}
+{_iterator_params}
+{_shared_docs["decompression_options"] % "filepath_or_buffer"}
+{_shared_docs["storage_options"]}
+
+Returns
+-------
+DataFrame, pandas.api.typing.StataReader
+    If iterator or chunksize, returns StataReader, else DataFrame.
+
+See Also
+--------
+io.stata.StataReader : Low-level reader for Stata data files.
+DataFrame.to_stata: Export Stata data files.
+
+{_reader_notes}
+
+Examples
+--------
+
+Creating a dummy stata for this example
+
+>>> df = pd.DataFrame({{'animal': ['falcon', 'parrot', 'falcon', 'parrot'],
+...                   'speed': [350, 18, 361, 15]}})  # doctest: +SKIP
+>>> df.to_stata('animals.dta')  # doctest: +SKIP
+
+Read a Stata dta file:
+
+>>> df = pd.read_stata('animals.dta')  # doctest: +SKIP
+
+Read a Stata dta file in 10,000 line chunks:
+
+>>> values = np.random.randint(0, 10, size=(20_000, 1), dtype="uint8")  # doctest: +SKIP
+>>> df = pd.DataFrame(values, columns=["i"])  # doctest: +SKIP
+>>> df.to_stata('filename.dta')  # doctest: +SKIP
+
+>>> with pd.read_stata('filename.dta', chunksize=10000) as itr:  # doctest: +SKIP
+>>>     for chunk in itr:
+...         # Operate on a single chunk, e.g., chunk.mean()
+...         pass  # doctest: +SKIP
+"""
+
+_read_method_doc = f"""\
+Reads observations from Stata file, converting them into a dataframe
+
+Parameters
+----------
+nrows : int
+    Number of lines to read from data file, if None read whole file.
+{_statafile_processing_params1}
+{_statafile_processing_params2}
+
+Returns
+-------
+DataFrame
+"""
+
+_stata_reader_doc = f"""\
+Class for reading Stata dta files.
+
+Parameters
+----------
+path_or_buf : path (string), buffer or path object
+    string, pathlib.Path or object
+    implementing a binary read() functions.
+{_statafile_processing_params1}
+{_statafile_processing_params2}
+{_chunksize_params}
+{_shared_docs["decompression_options"]}
+{_shared_docs["storage_options"]}
+
+{_reader_notes}
+"""
+
+
+_date_formats = ["%tc", "%tC", "%td", "%d", "%tw", "%tm", "%tq", "%th", "%ty"]
+
+
+stata_epoch: Final = datetime(1960, 1, 1)
+unix_epoch: Final = datetime(1970, 1, 1)
+
+
+def _stata_elapsed_date_to_datetime_vec(dates: Series, fmt: str) -> Series:
+    """
+    Convert from SIF to datetime. https://www.stata.com/help.cgi?datetime
+
+    Parameters
+    ----------
+    dates : Series
+        The Stata Internal Format date to convert to datetime according to fmt
+    fmt : str
+        The format to convert to. Can be, tc, td, tw, tm, tq, th, ty
+        Returns
+
+    Returns
+    -------
+    converted : Series
+        The converted dates
+
+    Examples
+    --------
+    >>> dates = pd.Series([52])
+    >>> _stata_elapsed_date_to_datetime_vec(dates, "%tw")
+    0   1961-01-01
+    dtype: datetime64[s]
+
+    Notes
+    -----
+    datetime/c - tc
+        milliseconds since 01jan1960 00:00:00.000, assuming 86,400 s/day
+    datetime/C - tC - NOT IMPLEMENTED
+        milliseconds since 01jan1960 00:00:00.000, adjusted for leap seconds
+    date - td
+        days since 01jan1960 (01jan1960 = 0)
+    weekly date - tw
+        weeks since 1960w1
+        This assumes 52 weeks in a year, then adds 7 * remainder of the weeks.
+        The datetime value is the start of the week in terms of days in the
+        year, not ISO calendar weeks.
+    monthly date - tm
+        months since 1960m1
+    quarterly date - tq
+        quarters since 1960q1
+    half-yearly date - th
+        half-years since 1960h1 yearly
+    date - ty
+        years since 0000
+    """
+
+    if fmt.startswith(("%tc", "tc")):
+        # Delta ms relative to base
+        td = np.timedelta64(stata_epoch - unix_epoch, "ms")
+        res = np.array(dates._values, dtype="M8[ms]") + td
+        return Series(res, index=dates.index)
+
+    elif fmt.startswith(("%td", "td", "%d", "d")):
+        # Delta days relative to base
+        td = np.timedelta64(stata_epoch - unix_epoch, "D")
+        res = np.array(dates._values, dtype="M8[D]") + td
+        return Series(res, index=dates.index)
+
+    elif fmt.startswith(("%tm", "tm")):
+        # Delta months relative to base
+        ordinals = dates + (stata_epoch.year - unix_epoch.year) * 12
+        res = np.array(ordinals, dtype="M8[M]").astype("M8[s]")
+        return Series(res, index=dates.index)
+
+    elif fmt.startswith(("%tq", "tq")):
+        # Delta quarters relative to base
+        ordinals = dates + (stata_epoch.year - unix_epoch.year) * 4
+        res = np.array(ordinals, dtype="M8[3M]").astype("M8[s]")
+        return Series(res, index=dates.index)
+
+    elif fmt.startswith(("%th", "th")):
+        # Delta half-years relative to base
+        ordinals = dates + (stata_epoch.year - unix_epoch.year) * 2
+        res = np.array(ordinals, dtype="M8[6M]").astype("M8[s]")
+        return Series(res, index=dates.index)
+
+    elif fmt.startswith(("%ty", "ty")):
+        # Years -- not delta
+        ordinals = dates - 1970
+        res = np.array(ordinals, dtype="M8[Y]").astype("M8[s]")
+        return Series(res, index=dates.index)
+
+    bad_locs = np.isnan(dates)
+    has_bad_values = False
+    if bad_locs.any():
+        has_bad_values = True
+        dates._values[bad_locs] = 1.0  # Replace with NaT
+    dates = dates.astype(np.int64)
+
+    if fmt.startswith(("%tC", "tC")):
+        warnings.warn(
+            "Encountered %tC format. Leaving in Stata Internal Format.",
+            stacklevel=find_stack_level(),
+        )
+        conv_dates = Series(dates, dtype=object)
+        if has_bad_values:
+            conv_dates[bad_locs] = NaT
+        return conv_dates
+    # does not count leap days - 7 days is a week.
+    # 52nd week may have more than 7 days
+    elif fmt.startswith(("%tw", "tw")):
+        year = stata_epoch.year + dates // 52
+        days = (dates % 52) * 7
+        per_y = (year - 1970).array.view("Period[Y]")
+        per_d = per_y.asfreq("D", how="S")
+        per_d_shifted = per_d + days._values
+        per_s = per_d_shifted.asfreq("s", how="S")
+        conv_dates_arr = per_s.view("M8[s]")
+        conv_dates = Series(conv_dates_arr, index=dates.index)
+
+    else:
+        raise ValueError(f"Date fmt {fmt} not understood")
+
+    if has_bad_values:  # Restore NaT for bad values
+        conv_dates[bad_locs] = NaT
+
+    return conv_dates
+
+
+def _datetime_to_stata_elapsed_vec(dates: Series, fmt: str) -> Series:
+    """
+    Convert from datetime to SIF. https://www.stata.com/help.cgi?datetime
+
+    Parameters
+    ----------
+    dates : Series
+        Series or array containing datetime or datetime64[ns] to
+        convert to the Stata Internal Format given by fmt
+    fmt : str
+        The format to convert to. Can be, tc, td, tw, tm, tq, th, ty
+    """
+    index = dates.index
+    NS_PER_DAY = 24 * 3600 * 1000 * 1000 * 1000
+    US_PER_DAY = NS_PER_DAY / 1000
+    MS_PER_DAY = NS_PER_DAY / 1_000_000
+
+    def parse_dates_safe(
+        dates: Series, delta: bool = False, year: bool = False, days: bool = False
+    ) -> DataFrame:
+        d = {}
+        if lib.is_np_dtype(dates.dtype, "M"):
+            if delta:
+                time_delta = dates.dt.as_unit("ms") - Timestamp(stata_epoch).as_unit(
+                    "ms"
+                )
+                d["delta"] = time_delta._values.view(np.int64)
+            if days or year:
+                date_index = DatetimeIndex(dates)
+                d["year"] = date_index._data.year
+                d["month"] = date_index._data.month
+            if days:
+                year_start = np.asarray(dates).astype("M8[Y]").astype(dates.dtype)
+                diff = dates - year_start
+                d["days"] = np.asarray(diff).astype("m8[D]").view("int64")
+
+        elif infer_dtype(dates, skipna=False) == "datetime":
+            if delta:
+                delta = dates._values - stata_epoch
+
+                def f(x: timedelta) -> float:
+                    return US_PER_DAY * x.days + 1000000 * x.seconds + x.microseconds
+
+                v = np.vectorize(f)
+                d["delta"] = v(delta)
+            if year:
+                year_month = dates.apply(lambda x: 100 * x.year + x.month)
+                d["year"] = year_month._values // 100
+                d["month"] = year_month._values - d["year"] * 100
+            if days:
+
+                def g(x: datetime) -> int:
+                    return (x - datetime(x.year, 1, 1)).days
+
+                v = np.vectorize(g)
+                d["days"] = v(dates)
+        else:
+            raise ValueError(
+                "Columns containing dates must contain either "
+                "datetime64, datetime or null values."
+            )
+
+        return DataFrame(d, index=index)
+
+    bad_loc = isna(dates)
+    index = dates.index
+    if bad_loc.any():
+        if lib.is_np_dtype(dates.dtype, "M"):
+            dates._values[bad_loc] = to_datetime(stata_epoch)
+        else:
+            dates._values[bad_loc] = stata_epoch
+
+    if fmt in ["%tc", "tc"]:
+        d = parse_dates_safe(dates, delta=True)
+        conv_dates = d.delta
+    elif fmt in ["%tC", "tC"]:
+        warnings.warn(
+            "Stata Internal Format tC not supported.",
+            stacklevel=find_stack_level(),
+        )
+        conv_dates = dates
+    elif fmt in ["%td", "td"]:
+        d = parse_dates_safe(dates, delta=True)
+        conv_dates = d.delta // MS_PER_DAY
+    elif fmt in ["%tw", "tw"]:
+        d = parse_dates_safe(dates, year=True, days=True)
+        conv_dates = 52 * (d.year - stata_epoch.year) + d.days // 7
+    elif fmt in ["%tm", "tm"]:
+        d = parse_dates_safe(dates, year=True)
+        conv_dates = 12 * (d.year - stata_epoch.year) + d.month - 1
+    elif fmt in ["%tq", "tq"]:
+        d = parse_dates_safe(dates, year=True)
+        conv_dates = 4 * (d.year - stata_epoch.year) + (d.month - 1) // 3
+    elif fmt in ["%th", "th"]:
+        d = parse_dates_safe(dates, year=True)
+        conv_dates = 2 * (d.year - stata_epoch.year) + (d.month > 6).astype(int)
+    elif fmt in ["%ty", "ty"]:
+        d = parse_dates_safe(dates, year=True)
+        conv_dates = d.year
+    else:
+        raise ValueError(f"Format {fmt} is not a known Stata date format")
+
+    conv_dates = Series(conv_dates, dtype=np.float64, copy=False)
+    missing_value = struct.unpack("<d", b"\x00\x00\x00\x00\x00\x00\xe0\x7f")[0]
+    conv_dates[bad_loc] = missing_value
+
+    return Series(conv_dates, index=index, copy=False)
+
+
+excessive_string_length_error: Final = """
+Fixed width strings in Stata .dta files are limited to 244 (or fewer)
+characters.  Column '{0}' does not satisfy this restriction. Use the
+'version=117' parameter to write the newer (Stata 13 and later) format.
+"""
+
+
+precision_loss_doc: Final = """
+Column converted from {0} to {1}, and some data are outside of the lossless
+conversion range. This may result in a loss of precision in the saved data.
+"""
+
+
+value_label_mismatch_doc: Final = """
+Stata value labels (pandas categories) must be strings. Column {0} contains
+non-string labels which will be converted to strings.  Please check that the
+Stata data file created has not lost information due to duplicate labels.
+"""
+
+
+invalid_name_doc: Final = """
+Not all pandas column names were valid Stata variable names.
+The following replacements have been made:
+
+    {0}
+
+If this is not what you expect, please make sure you have Stata-compliant
+column names in your DataFrame (strings only, max 32 characters, only
+alphanumerics and underscores, no Stata reserved words)
+"""
+
+
+categorical_conversion_warning: Final = """
+One or more series with value labels are not fully labeled. Reading this
+dataset with an iterator results in categorical variable with different
+categories. This occurs since it is not possible to know all possible values
+until the entire dataset has been read. To avoid this warning, you can either
+read dataset without an iterator, or manually convert categorical data by
+``convert_categoricals`` to False and then accessing the variable labels
+through the value_labels method of the reader.
+"""
+
+
+def _cast_to_stata_types(data: DataFrame) -> DataFrame:
+    """
+    Checks the dtypes of the columns of a pandas DataFrame for
+    compatibility with the data types and ranges supported by Stata, and
+    converts if necessary.
+
+    Parameters
+    ----------
+    data : DataFrame
+        The DataFrame to check and convert
+
+    Notes
+    -----
+    Numeric columns in Stata must be one of int8, int16, int32, float32 or
+    float64, with some additional value restrictions.  int8 and int16 columns
+    are checked for violations of the value restrictions and upcast if needed.
+    int64 data is not usable in Stata, and so it is downcast to int32 whenever
+    the value are in the int32 range, and sidecast to float64 when larger than
+    this range.  If the int64 values are outside of the range of those
+    perfectly representable as float64 values, a warning is raised.
+
+    bool columns are cast to int8.  uint columns are converted to int of the
+    same size if there is no loss in precision, otherwise are upcast to a
+    larger type.  uint64 is currently not supported since it is concerted to
+    object in a DataFrame.
+    """
+    ws = ""
+    # original, if small, if large
+    conversion_data: tuple[
+        tuple[type, type, type],
+        tuple[type, type, type],
+        tuple[type, type, type],
+        tuple[type, type, type],
+        tuple[type, type, type],
+    ] = (
+        (np.bool_, np.int8, np.int8),
+        (np.uint8, np.int8, np.int16),
+        (np.uint16, np.int16, np.int32),
+        (np.uint32, np.int32, np.int64),
+        (np.uint64, np.int64, np.float64),
+    )
+
+    float32_max = struct.unpack("<f", b"\xff\xff\xff\x7e")[0]
+    float64_max = struct.unpack("<d", b"\xff\xff\xff\xff\xff\xff\xdf\x7f")[0]
+
+    for col in data:
+        # Cast from unsupported types to supported types
+        is_nullable_int = (
+            isinstance(data[col].dtype, ExtensionDtype)
+            and data[col].dtype.kind in "iub"
+        )
+        # We need to find orig_missing before altering data below
+        orig_missing = data[col].isna()
+        if is_nullable_int:
+            fv = 0 if data[col].dtype.kind in "iu" else False
+            # Replace with NumPy-compatible column
+            data[col] = data[col].fillna(fv).astype(data[col].dtype.numpy_dtype)
+        elif isinstance(data[col].dtype, ExtensionDtype):
+            if getattr(data[col].dtype, "numpy_dtype", None) is not None:
+                data[col] = data[col].astype(data[col].dtype.numpy_dtype)
+            elif is_string_dtype(data[col].dtype):
+                # TODO could avoid converting string dtype to object here,
+                # but handle string dtype in _encode_strings
+                data[col] = data[col].astype("object")
+                # generate_table checks for None values
+                data.loc[data[col].isna(), col] = None
+
+        dtype = data[col].dtype
+        empty_df = data.shape[0] == 0
+        for c_data in conversion_data:
+            if dtype == c_data[0]:
+                if empty_df or data[col].max() <= np.iinfo(c_data[1]).max:
+                    dtype = c_data[1]
+                else:
+                    dtype = c_data[2]
+                if c_data[2] == np.int64:  # Warn if necessary
+                    if data[col].max() >= 2**53:
+                        ws = precision_loss_doc.format("uint64", "float64")
+
+                data[col] = data[col].astype(dtype)
+
+        # Check values and upcast if necessary
+
+        if dtype == np.int8 and not empty_df:
+            if data[col].max() > 100 or data[col].min() < -127:
+                data[col] = data[col].astype(np.int16)
+        elif dtype == np.int16 and not empty_df:
+            if data[col].max() > 32740 or data[col].min() < -32767:
+                data[col] = data[col].astype(np.int32)
+        elif dtype == np.int64:
+            if empty_df or (
+                data[col].max() <= 2147483620 and data[col].min() >= -2147483647
+            ):
+                data[col] = data[col].astype(np.int32)
+            else:
+                data[col] = data[col].astype(np.float64)
+                if data[col].max() >= 2**53 or data[col].min() <= -(2**53):
+                    ws = precision_loss_doc.format("int64", "float64")
+        elif dtype in (np.float32, np.float64):
+            if np.isinf(data[col]).any():
+                raise ValueError(
+                    f"Column {col} contains infinity or -infinity"
+                    "which is outside the range supported by Stata."
+                )
+            value = data[col].max()
+            if dtype == np.float32 and value > float32_max:
+                data[col] = data[col].astype(np.float64)
+            elif dtype == np.float64:
+                if value > float64_max:
+                    raise ValueError(
+                        f"Column {col} has a maximum value ({value}) outside the range "
+                        f"supported by Stata ({float64_max})"
+                    )
+        if is_nullable_int:
+            if orig_missing.any():
+                # Replace missing by Stata sentinel value
+                sentinel = StataMissingValue.BASE_MISSING_VALUES[data[col].dtype.name]
+                data.loc[orig_missing, col] = sentinel
+    if ws:
+        warnings.warn(
+            ws,
+            PossiblePrecisionLoss,
+            stacklevel=find_stack_level(),
+        )
+
+    return data
+
+
+class StataValueLabel:
+    """
+    Parse a categorical column and prepare formatted output
+
+    Parameters
+    ----------
+    catarray : Series
+        Categorical Series to encode
+    encoding : {"latin-1", "utf-8"}
+        Encoding to use for value labels.
+    """
+
+    def __init__(
+        self, catarray: Series, encoding: Literal["latin-1", "utf-8"] = "latin-1"
+    ) -> None:
+        if encoding not in ("latin-1", "utf-8"):
+            raise ValueError("Only latin-1 and utf-8 are supported.")
+        self.labname = catarray.name
+        self._encoding = encoding
+        categories = catarray.cat.categories
+        self.value_labels = enumerate(categories)
+
+        self._prepare_value_labels()
+
+    def _prepare_value_labels(self) -> None:
+        """Encode value labels."""
+
+        self.text_len = 0
+        self.txt: list[bytes] = []
+        self.n = 0
+        # Offsets (length of categories), converted to int32
+        self.off = np.array([], dtype=np.int32)
+        # Values, converted to int32
+        self.val = np.array([], dtype=np.int32)
+        self.len = 0
+
+        # Compute lengths and setup lists of offsets and labels
+        offsets: list[int] = []
+        values: list[float] = []
+        for vl in self.value_labels:
+            category: str | bytes = vl[1]
+            if not isinstance(category, str):
+                category = str(category)
+                warnings.warn(
+                    value_label_mismatch_doc.format(self.labname),
+                    ValueLabelTypeMismatch,
+                    stacklevel=find_stack_level(),
+                )
+            category = category.encode(self._encoding)
+            offsets.append(self.text_len)
+            self.text_len += len(category) + 1  # +1 for the padding
+            values.append(vl[0])
+            self.txt.append(category)
+            self.n += 1
+
+        # Ensure int32
+        self.off = np.array(offsets, dtype=np.int32)
+        self.val = np.array(values, dtype=np.int32)
+
+        # Total length
+        self.len = 4 + 4 + 4 * self.n + 4 * self.n + self.text_len
+
+    def generate_value_label(self, byteorder: str) -> bytes:
+        """
+        Generate the binary representation of the value labels.
+
+        Parameters
+        ----------
+        byteorder : str
+            Byte order of the output
+
+        Returns
+        -------
+        value_label : bytes
+            Bytes containing the formatted value label
+        """
+        encoding = self._encoding
+        bio = BytesIO()
+        null_byte = b"\x00"
+
+        # len
+        bio.write(struct.pack(byteorder + "i", self.len))
+
+        # labname
+        labname = str(self.labname)[:32].encode(encoding)
+        lab_len = 32 if encoding not in ("utf-8", "utf8") else 128
+        labname = _pad_bytes(labname, lab_len + 1)
+        bio.write(labname)
+
+        # padding - 3 bytes
+        for i in range(3):
+            bio.write(struct.pack("c", null_byte))
+
+        # value_label_table
+        # n - int32
+        bio.write(struct.pack(byteorder + "i", self.n))
+
+        # textlen  - int32
+        bio.write(struct.pack(byteorder + "i", self.text_len))
+
+        # off - int32 array (n elements)
+        for offset in self.off:
+            bio.write(struct.pack(byteorder + "i", offset))
+
+        # val - int32 array (n elements)
+        for value in self.val:
+            bio.write(struct.pack(byteorder + "i", value))
+
+        # txt - Text labels, null terminated
+        for text in self.txt:
+            bio.write(text + null_byte)
+
+        return bio.getvalue()
+
+
+class StataNonCatValueLabel(StataValueLabel):
+    """
+    Prepare formatted version of value labels
+
+    Parameters
+    ----------
+    labname : str
+        Value label name
+    value_labels: Dictionary
+        Mapping of values to labels
+    encoding : {"latin-1", "utf-8"}
+        Encoding to use for value labels.
+    """
+
+    def __init__(
+        self,
+        labname: str,
+        value_labels: dict[float, str],
+        encoding: Literal["latin-1", "utf-8"] = "latin-1",
+    ) -> None:
+        if encoding not in ("latin-1", "utf-8"):
+            raise ValueError("Only latin-1 and utf-8 are supported.")
+
+        self.labname = labname
+        self._encoding = encoding
+        self.value_labels = sorted(  # type: ignore[assignment]
+            value_labels.items(), key=lambda x: x[0]
+        )
+        self._prepare_value_labels()
+
+
+class StataMissingValue:
+    """
+    An observation's missing value.
+
+    Parameters
+    ----------
+    value : {int, float}
+        The Stata missing value code
+
+    Notes
+    -----
+    More information: <https://www.stata.com/help.cgi?missing>
+
+    Integer missing values make the code '.', '.a', ..., '.z' to the ranges
+    101 ... 127 (for int8), 32741 ... 32767  (for int16) and 2147483621 ...
+    2147483647 (for int32).  Missing values for floating point data types are
+    more complex but the pattern is simple to discern from the following table.
+
+    np.float32 missing values (float in Stata)
+    0000007f    .
+    0008007f    .a
+    0010007f    .b
+    ...
+    00c0007f    .x
+    00c8007f    .y
+    00d0007f    .z
+
+    np.float64 missing values (double in Stata)
+    000000000000e07f    .
+    000000000001e07f    .a
+    000000000002e07f    .b
+    ...
+    000000000018e07f    .x
+    000000000019e07f    .y
+    00000000001ae07f    .z
+    """
+
+    # Construct a dictionary of missing values
+    MISSING_VALUES: dict[float, str] = {}
+    bases: Final = (101, 32741, 2147483621)
+    for b in bases:
+        # Conversion to long to avoid hash issues on 32 bit platforms #8968
+        MISSING_VALUES[b] = "."
+        for i in range(1, 27):
+            MISSING_VALUES[i + b] = "." + chr(96 + i)
+
+    float32_base: bytes = b"\x00\x00\x00\x7f"
+    increment_32: int = struct.unpack("<i", b"\x00\x08\x00\x00")[0]
+    for i in range(27):
+        key = struct.unpack("<f", float32_base)[0]
+        MISSING_VALUES[key] = "."
+        if i > 0:
+            MISSING_VALUES[key] += chr(96 + i)
+        int_value = struct.unpack("<i", struct.pack("<f", key))[0] + increment_32
+        float32_base = struct.pack("<i", int_value)
+
+    float64_base: bytes = b"\x00\x00\x00\x00\x00\x00\xe0\x7f"
+    increment_64 = struct.unpack("q", b"\x00\x00\x00\x00\x00\x01\x00\x00")[0]
+    for i in range(27):
+        key = struct.unpack("<d", float64_base)[0]
+        MISSING_VALUES[key] = "."
+        if i > 0:
+            MISSING_VALUES[key] += chr(96 + i)
+        int_value = struct.unpack("q", struct.pack("<d", key))[0] + increment_64
+        float64_base = struct.pack("q", int_value)
+
+    BASE_MISSING_VALUES: Final = {
+        "int8": 101,
+        "int16": 32741,
+        "int32": 2147483621,
+        "float32": struct.unpack("<f", float32_base)[0],
+        "float64": struct.unpack("<d", float64_base)[0],
+    }
+
+    def __init__(self, value: float) -> None:
+        self._value = value
+        # Conversion to int to avoid hash issues on 32 bit platforms #8968
+        value = int(value) if value < 2147483648 else float(value)
+        self._str = self.MISSING_VALUES[value]
+
+    @property
+    def string(self) -> str:
+        """
+        The Stata representation of the missing value: '.', '.a'..'.z'
+
+        Returns
+        -------
+        str
+            The representation of the missing value.
+        """
+        return self._str
+
+    @property
+    def value(self) -> float:
+        """
+        The binary representation of the missing value.
+
+        Returns
+        -------
+        {int, float}
+            The binary representation of the missing value.
+        """
+        return self._value
+
+    def __str__(self) -> str:
+        return self.string
+
+    def __repr__(self) -> str:
+        return f"{type(self)}({self})"
+
+    def __eq__(self, other: object) -> bool:
+        return (
+            isinstance(other, type(self))
+            and self.string == other.string
+            and self.value == other.value
+        )
+
+    @classmethod
+    def get_base_missing_value(cls, dtype: np.dtype) -> float:
+        if dtype.type is np.int8:
+            value = cls.BASE_MISSING_VALUES["int8"]
+        elif dtype.type is np.int16:
+            value = cls.BASE_MISSING_VALUES["int16"]
+        elif dtype.type is np.int32:
+            value = cls.BASE_MISSING_VALUES["int32"]
+        elif dtype.type is np.float32:
+            value = cls.BASE_MISSING_VALUES["float32"]
+        elif dtype.type is np.float64:
+            value = cls.BASE_MISSING_VALUES["float64"]
+        else:
+            raise ValueError("Unsupported dtype")
+        return value
+
+
+class StataParser:
+    def __init__(self) -> None:
+        # type          code.
+        # --------------------
+        # str1        1 = 0x01
+        # str2        2 = 0x02
+        # ...
+        # str244    244 = 0xf4
+        # byte      251 = 0xfb  (sic)
+        # int       252 = 0xfc
+        # long      253 = 0xfd
+        # float     254 = 0xfe
+        # double    255 = 0xff
+        # --------------------
+        # NOTE: the byte type seems to be reserved for categorical variables
+        # with a label, but the underlying variable is -127 to 100
+        # we're going to drop the label and cast to int
+        self.DTYPE_MAP = dict(
+            [(i, np.dtype(f"S{i}")) for i in range(1, 245)]
+            + [
+                (251, np.dtype(np.int8)),
+                (252, np.dtype(np.int16)),
+                (253, np.dtype(np.int32)),
+                (254, np.dtype(np.float32)),
+                (255, np.dtype(np.float64)),
+            ]
+        )
+        self.DTYPE_MAP_XML: dict[int, np.dtype] = {
+            32768: np.dtype(np.uint8),  # Keys to GSO
+            65526: np.dtype(np.float64),
+            65527: np.dtype(np.float32),
+            65528: np.dtype(np.int32),
+            65529: np.dtype(np.int16),
+            65530: np.dtype(np.int8),
+        }
+        self.TYPE_MAP = list(tuple(range(251)) + tuple("bhlfd"))
+        self.TYPE_MAP_XML = {
+            # Not really a Q, unclear how to handle byteswap
+            32768: "Q",
+            65526: "d",
+            65527: "f",
+            65528: "l",
+            65529: "h",
+            65530: "b",
+        }
+        # NOTE: technically, some of these are wrong. there are more numbers
+        # that can be represented. it's the 27 ABOVE and BELOW the max listed
+        # numeric data type in [U] 12.2.2 of the 11.2 manual
+        float32_min = b"\xff\xff\xff\xfe"
+        float32_max = b"\xff\xff\xff\x7e"
+        float64_min = b"\xff\xff\xff\xff\xff\xff\xef\xff"
+        float64_max = b"\xff\xff\xff\xff\xff\xff\xdf\x7f"
+        self.VALID_RANGE = {
+            "b": (-127, 100),
+            "h": (-32767, 32740),
+            "l": (-2147483647, 2147483620),
+            "f": (
+                np.float32(struct.unpack("<f", float32_min)[0]),
+                np.float32(struct.unpack("<f", float32_max)[0]),
+            ),
+            "d": (
+                np.float64(struct.unpack("<d", float64_min)[0]),
+                np.float64(struct.unpack("<d", float64_max)[0]),
+            ),
+        }
+        self.OLD_VALID_RANGE = {
+            "b": (-128, 126),
+            "h": (-32768, 32766),
+            "l": (-2147483648, 2147483646),
+            "f": (
+                np.float32(struct.unpack("<f", float32_min)[0]),
+                np.float32(struct.unpack("<f", float32_max)[0]),
+            ),
+            "d": (
+                np.float64(struct.unpack("<d", float64_min)[0]),
+                np.float64(struct.unpack("<d", float64_max)[0]),
+            ),
+        }
+
+        self.OLD_TYPE_MAPPING = {
+            98: 251,  # byte
+            105: 252,  # int
+            108: 253,  # long
+            102: 254,  # float
+            100: 255,  # double
+        }
+
+        # These missing values are the generic '.' in Stata, and are used
+        # to replace nans
+        self.MISSING_VALUES: dict[str, int | np.float32 | np.float64] = {
+            "b": 101,
+            "h": 32741,
+            "l": 2147483621,
+            "f": np.float32(struct.unpack("<f", b"\x00\x00\x00\x7f")[0]),
+            "d": np.float64(
+                struct.unpack("<d", b"\x00\x00\x00\x00\x00\x00\xe0\x7f")[0]
+            ),
+        }
+        self.NUMPY_TYPE_MAP = {
+            "b": "i1",
+            "h": "i2",
+            "l": "i4",
+            "f": "f4",
+            "d": "f8",
+            "Q": "u8",
+        }
+
+        # Reserved words cannot be used as variable names
+        self.RESERVED_WORDS = {
+            "aggregate",
+            "array",
+            "boolean",
+            "break",
+            "byte",
+            "case",
+            "catch",
+            "class",
+            "colvector",
+            "complex",
+            "const",
+            "continue",
+            "default",
+            "delegate",
+            "delete",
+            "do",
+            "double",
+            "else",
+            "eltypedef",
+            "end",
+            "enum",
+            "explicit",
+            "export",
+            "external",
+            "float",
+            "for",
+            "friend",
+            "function",
+            "global",
+            "goto",
+            "if",
+            "inline",
+            "int",
+            "local",
+            "long",
+            "NULL",
+            "pragma",
+            "protected",
+            "quad",
+            "rowvector",
+            "short",
+            "typedef",
+            "typename",
+            "virtual",
+            "_all",
+            "_N",
+            "_skip",
+            "_b",
+            "_pi",
+            "str#",
+            "in",
+            "_pred",
+            "strL",
+            "_coef",
+            "_rc",
+            "using",
+            "_cons",
+            "_se",
+            "with",
+            "_n",
+        }
+
+
+class StataReader(StataParser, abc.Iterator):
+    __doc__ = _stata_reader_doc
+
+    _path_or_buf: IO[bytes]  # Will be assigned by `_open_file`.
+
+    def __init__(
+        self,
+        path_or_buf: FilePath | ReadBuffer[bytes],
+        convert_dates: bool = True,
+        convert_categoricals: bool = True,
+        index_col: str | None = None,
+        convert_missing: bool = False,
+        preserve_dtypes: bool = True,
+        columns: Sequence[str] | None = None,
+        order_categoricals: bool = True,
+        chunksize: int | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+    ) -> None:
+        super().__init__()
+
+        # Arguments to the reader (can be temporarily overridden in
+        # calls to read).
+        self._convert_dates = convert_dates
+        self._convert_categoricals = convert_categoricals
+        self._index_col = index_col
+        self._convert_missing = convert_missing
+        self._preserve_dtypes = preserve_dtypes
+        self._columns = columns
+        self._order_categoricals = order_categoricals
+        self._original_path_or_buf = path_or_buf
+        self._compression = compression
+        self._storage_options = storage_options
+        self._encoding = ""
+        self._chunksize = chunksize
+        self._using_iterator = False
+        self._entered = False
+        if self._chunksize is None:
+            self._chunksize = 1
+        elif not isinstance(chunksize, int) or chunksize <= 0:
+            raise ValueError("chunksize must be a positive integer when set.")
+
+        # State variables for the file
+        self._close_file: Callable[[], None] | None = None
+        self._column_selector_set = False
+        self._value_label_dict: dict[str, dict[int, str]] = {}
+        self._value_labels_read = False
+        self._dtype: np.dtype | None = None
+        self._lines_read = 0
+
+        self._native_byteorder = _set_endianness(sys.byteorder)
+
+    def _ensure_open(self) -> None:
+        """
+        Ensure the file has been opened and its header data read.
+        """
+        if not hasattr(self, "_path_or_buf"):
+            self._open_file()
+
+    def _open_file(self) -> None:
+        """
+        Open the file (with compression options, etc.), and read header information.
+        """
+        if not self._entered:
+            warnings.warn(
+                "StataReader is being used without using a context manager. "
+                "Using StataReader as a context manager is the only supported method.",
+                ResourceWarning,
+                stacklevel=find_stack_level(),
+            )
+        handles = get_handle(
+            self._original_path_or_buf,
+            "rb",
+            storage_options=self._storage_options,
+            is_text=False,
+            compression=self._compression,
+        )
+        if hasattr(handles.handle, "seekable") and handles.handle.seekable():
+            # If the handle is directly seekable, use it without an extra copy.
+            self._path_or_buf = handles.handle
+            self._close_file = handles.close
+        else:
+            # Copy to memory, and ensure no encoding.
+            with handles:
+                self._path_or_buf = BytesIO(handles.handle.read())
+            self._close_file = self._path_or_buf.close
+
+        self._read_header()
+        self._setup_dtype()
+
+    def __enter__(self) -> Self:
+        """enter context manager"""
+        self._entered = True
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        if self._close_file:
+            self._close_file()
+
+    def _set_encoding(self) -> None:
+        """
+        Set string encoding which depends on file version
+        """
+        if self._format_version < 118:
+            self._encoding = "latin-1"
+        else:
+            self._encoding = "utf-8"
+
+    def _read_int8(self) -> int:
+        return struct.unpack("b", self._path_or_buf.read(1))[0]
+
+    def _read_uint8(self) -> int:
+        return struct.unpack("B", self._path_or_buf.read(1))[0]
+
+    def _read_uint16(self) -> int:
+        return struct.unpack(f"{self._byteorder}H", self._path_or_buf.read(2))[0]
+
+    def _read_uint32(self) -> int:
+        return struct.unpack(f"{self._byteorder}I", self._path_or_buf.read(4))[0]
+
+    def _read_uint64(self) -> int:
+        return struct.unpack(f"{self._byteorder}Q", self._path_or_buf.read(8))[0]
+
+    def _read_int16(self) -> int:
+        return struct.unpack(f"{self._byteorder}h", self._path_or_buf.read(2))[0]
+
+    def _read_int32(self) -> int:
+        return struct.unpack(f"{self._byteorder}i", self._path_or_buf.read(4))[0]
+
+    def _read_int64(self) -> int:
+        return struct.unpack(f"{self._byteorder}q", self._path_or_buf.read(8))[0]
+
+    def _read_char8(self) -> bytes:
+        return struct.unpack("c", self._path_or_buf.read(1))[0]
+
+    def _read_int16_count(self, count: int) -> tuple[int, ...]:
+        return struct.unpack(
+            f"{self._byteorder}{'h' * count}",
+            self._path_or_buf.read(2 * count),
+        )
+
+    def _read_header(self) -> None:
+        first_char = self._read_char8()
+        if first_char == b"<":
+            self._read_new_header()
+        else:
+            self._read_old_header(first_char)
+
+    def _read_new_header(self) -> None:
+        # The first part of the header is common to 117 - 119.
+        self._path_or_buf.read(27)  # stata_dta><header><release>
+        self._format_version = int(self._path_or_buf.read(3))
+        if self._format_version not in [117, 118, 119]:
+            raise ValueError(_version_error.format(version=self._format_version))
+        self._set_encoding()
+        self._path_or_buf.read(21)  # </release><byteorder>
+        self._byteorder = ">" if self._path_or_buf.read(3) == b"MSF" else "<"
+        self._path_or_buf.read(15)  # </byteorder><K>
+        self._nvar = (
+            self._read_uint16() if self._format_version <= 118 else self._read_uint32()
+        )
+        self._path_or_buf.read(7)  # </K><N>
+
+        self._nobs = self._get_nobs()
+        self._path_or_buf.read(11)  # </N><label>
+        self._data_label = self._get_data_label()
+        self._path_or_buf.read(19)  # </label><timestamp>
+        self._time_stamp = self._get_time_stamp()
+        self._path_or_buf.read(26)  # </timestamp></header><map>
+        self._path_or_buf.read(8)  # 0x0000000000000000
+        self._path_or_buf.read(8)  # position of <map>
+
+        self._seek_vartypes = self._read_int64() + 16
+        self._seek_varnames = self._read_int64() + 10
+        self._seek_sortlist = self._read_int64() + 10
+        self._seek_formats = self._read_int64() + 9
+        self._seek_value_label_names = self._read_int64() + 19
+
+        # Requires version-specific treatment
+        self._seek_variable_labels = self._get_seek_variable_labels()
+
+        self._path_or_buf.read(8)  # <characteristics>
+        self._data_location = self._read_int64() + 6
+        self._seek_strls = self._read_int64() + 7
+        self._seek_value_labels = self._read_int64() + 14
+
+        self._typlist, self._dtyplist = self._get_dtypes(self._seek_vartypes)
+
+        self._path_or_buf.seek(self._seek_varnames)
+        self._varlist = self._get_varlist()
+
+        self._path_or_buf.seek(self._seek_sortlist)
+        self._srtlist = self._read_int16_count(self._nvar + 1)[:-1]
+
+        self._path_or_buf.seek(self._seek_formats)
+        self._fmtlist = self._get_fmtlist()
+
+        self._path_or_buf.seek(self._seek_value_label_names)
+        self._lbllist = self._get_lbllist()
+
+        self._path_or_buf.seek(self._seek_variable_labels)
+        self._variable_labels = self._get_variable_labels()
+
+    # Get data type information, works for versions 117-119.
+    def _get_dtypes(
+        self, seek_vartypes: int
+    ) -> tuple[list[int | str], list[str | np.dtype]]:
+        self._path_or_buf.seek(seek_vartypes)
+        typlist = []
+        dtyplist = []
+        for _ in range(self._nvar):
+            typ = self._read_uint16()
+            if typ <= 2045:
+                typlist.append(typ)
+                dtyplist.append(str(typ))
+            else:
+                try:
+                    typlist.append(self.TYPE_MAP_XML[typ])  # type: ignore[arg-type]
+                    dtyplist.append(self.DTYPE_MAP_XML[typ])  # type: ignore[arg-type]
+                except KeyError as err:
+                    raise ValueError(f"cannot convert stata types [{typ}]") from err
+
+        return typlist, dtyplist  # type: ignore[return-value]
+
+    def _get_varlist(self) -> list[str]:
+        # 33 in order formats, 129 in formats 118 and 119
+        b = 33 if self._format_version < 118 else 129
+        return [self._decode(self._path_or_buf.read(b)) for _ in range(self._nvar)]
+
+    # Returns the format list
+    def _get_fmtlist(self) -> list[str]:
+        if self._format_version >= 118:
+            b = 57
+        elif self._format_version > 113:
+            b = 49
+        elif self._format_version > 104:
+            b = 12
+        else:
+            b = 7
+
+        return [self._decode(self._path_or_buf.read(b)) for _ in range(self._nvar)]
+
+    # Returns the label list
+    def _get_lbllist(self) -> list[str]:
+        if self._format_version >= 118:
+            b = 129
+        elif self._format_version > 108:
+            b = 33
+        else:
+            b = 9
+        return [self._decode(self._path_or_buf.read(b)) for _ in range(self._nvar)]
+
+    def _get_variable_labels(self) -> list[str]:
+        if self._format_version >= 118:
+            vlblist = [
+                self._decode(self._path_or_buf.read(321)) for _ in range(self._nvar)
+            ]
+        elif self._format_version > 105:
+            vlblist = [
+                self._decode(self._path_or_buf.read(81)) for _ in range(self._nvar)
+            ]
+        else:
+            vlblist = [
+                self._decode(self._path_or_buf.read(32)) for _ in range(self._nvar)
+            ]
+        return vlblist
+
+    def _get_nobs(self) -> int:
+        if self._format_version >= 118:
+            return self._read_uint64()
+        elif self._format_version >= 103:
+            return self._read_uint32()
+        else:
+            return self._read_uint16()
+
+    def _get_data_label(self) -> str:
+        if self._format_version >= 118:
+            strlen = self._read_uint16()
+            return self._decode(self._path_or_buf.read(strlen))
+        elif self._format_version == 117:
+            strlen = self._read_int8()
+            return self._decode(self._path_or_buf.read(strlen))
+        elif self._format_version > 105:
+            return self._decode(self._path_or_buf.read(81))
+        else:
+            return self._decode(self._path_or_buf.read(32))
+
+    def _get_time_stamp(self) -> str:
+        if self._format_version >= 118:
+            strlen = self._read_int8()
+            return self._path_or_buf.read(strlen).decode("utf-8")
+        elif self._format_version == 117:
+            strlen = self._read_int8()
+            return self._decode(self._path_or_buf.read(strlen))
+        elif self._format_version > 104:
+            return self._decode(self._path_or_buf.read(18))
+        else:
+            raise ValueError
+
+    def _get_seek_variable_labels(self) -> int:
+        if self._format_version == 117:
+            self._path_or_buf.read(8)  # <variable_labels>, throw away
+            # Stata 117 data files do not follow the described format.  This is
+            # a work around that uses the previous label, 33 bytes for each
+            # variable, 20 for the closing tag and 17 for the opening tag
+            return self._seek_value_label_names + (33 * self._nvar) + 20 + 17
+        elif self._format_version >= 118:
+            return self._read_int64() + 17
+        else:
+            raise ValueError
+
+    def _read_old_header(self, first_char: bytes) -> None:
+        self._format_version = int(first_char[0])
+        if self._format_version not in [
+            102,
+            103,
+            104,
+            105,
+            108,
+            110,
+            111,
+            113,
+            114,
+            115,
+        ]:
+            raise ValueError(_version_error.format(version=self._format_version))
+        self._set_encoding()
+        # Note 102 format will have a zero in this header position, so support
+        # relies on little-endian being set whenever this value isn't one,
+        # even though for later releases strictly speaking the value should
+        # be either one or two to be valid
+        self._byteorder = ">" if self._read_int8() == 0x1 else "<"
+        self._filetype = self._read_int8()
+        self._path_or_buf.read(1)  # unused
+
+        self._nvar = self._read_uint16()
+        self._nobs = self._get_nobs()
+
+        self._data_label = self._get_data_label()
+
+        if self._format_version >= 105:
+            self._time_stamp = self._get_time_stamp()
+
+        # descriptors
+        if self._format_version >= 111:
+            typlist = [int(c) for c in self._path_or_buf.read(self._nvar)]
+        else:
+            buf = self._path_or_buf.read(self._nvar)
+            typlistb = np.frombuffer(buf, dtype=np.uint8)
+            typlist = []
+            for tp in typlistb:
+                if tp in self.OLD_TYPE_MAPPING:
+                    typlist.append(self.OLD_TYPE_MAPPING[tp])
+                else:
+                    typlist.append(tp - 127)  # bytes
+
+        try:
+            self._typlist = [self.TYPE_MAP[typ] for typ in typlist]
+        except ValueError as err:
+            invalid_types = ",".join([str(x) for x in typlist])
+            raise ValueError(f"cannot convert stata types [{invalid_types}]") from err
+        try:
+            self._dtyplist = [self.DTYPE_MAP[typ] for typ in typlist]
+        except ValueError as err:
+            invalid_dtypes = ",".join([str(x) for x in typlist])
+            raise ValueError(f"cannot convert stata dtypes [{invalid_dtypes}]") from err
+
+        if self._format_version > 108:
+            self._varlist = [
+                self._decode(self._path_or_buf.read(33)) for _ in range(self._nvar)
+            ]
+        else:
+            self._varlist = [
+                self._decode(self._path_or_buf.read(9)) for _ in range(self._nvar)
+            ]
+        self._srtlist = self._read_int16_count(self._nvar + 1)[:-1]
+
+        self._fmtlist = self._get_fmtlist()
+
+        self._lbllist = self._get_lbllist()
+
+        self._variable_labels = self._get_variable_labels()
+
+        # ignore expansion fields (Format 105 and later)
+        # When reading, read five bytes; the last four bytes now tell you
+        # the size of the next read, which you discard.  You then continue
+        # like this until you read 5 bytes of zeros.
+
+        if self._format_version > 104:
+            while True:
+                data_type = self._read_int8()
+                if self._format_version > 108:
+                    data_len = self._read_int32()
+                else:
+                    data_len = self._read_int16()
+                if data_type == 0:
+                    break
+                self._path_or_buf.read(data_len)
+
+        # necessary data to continue parsing
+        self._data_location = self._path_or_buf.tell()
+
+    def _setup_dtype(self) -> np.dtype:
+        """Map between numpy and state dtypes"""
+        if self._dtype is not None:
+            return self._dtype
+
+        dtypes = []  # Convert struct data types to numpy data type
+        for i, typ in enumerate(self._typlist):
+            if typ in self.NUMPY_TYPE_MAP:
+                typ = cast(str, typ)  # only strs in NUMPY_TYPE_MAP
+                dtypes.append((f"s{i}", f"{self._byteorder}{self.NUMPY_TYPE_MAP[typ]}"))
+            else:
+                dtypes.append((f"s{i}", f"S{typ}"))
+        self._dtype = np.dtype(dtypes)
+
+        return self._dtype
+
+    def _decode(self, s: bytes) -> str:
+        # have bytes not strings, so must decode
+        s = s.partition(b"\0")[0]
+        try:
+            return s.decode(self._encoding)
+        except UnicodeDecodeError:
+            # GH 25960, fallback to handle incorrect format produced when 117
+            # files are converted to 118 files in Stata
+            encoding = self._encoding
+            msg = f"""
+One or more strings in the dta file could not be decoded using {encoding}, and
+so the fallback encoding of latin-1 is being used.  This can happen when a file
+has been incorrectly encoded by Stata or some other software. You should verify
+the string values returned are correct."""
+            warnings.warn(
+                msg,
+                UnicodeWarning,
+                stacklevel=find_stack_level(),
+            )
+            return s.decode("latin-1")
+
+    def _read_new_value_labels(self) -> None:
+        """Reads value labels with variable length strings (108 and later format)"""
+        if self._format_version >= 117:
+            self._path_or_buf.seek(self._seek_value_labels)
+        else:
+            assert self._dtype is not None
+            offset = self._nobs * self._dtype.itemsize
+            self._path_or_buf.seek(self._data_location + offset)
+
+        while True:
+            if self._format_version >= 117:
+                if self._path_or_buf.read(5) == b"</val":  # <lbl>
+                    break  # end of value label table
+
+            slength = self._path_or_buf.read(4)
+            if not slength:
+                break  # end of value label table (format < 117), or end-of-file
+            if self._format_version == 108:
+                labname = self._decode(self._path_or_buf.read(9))
+            elif self._format_version <= 117:
+                labname = self._decode(self._path_or_buf.read(33))
+            else:
+                labname = self._decode(self._path_or_buf.read(129))
+            self._path_or_buf.read(3)  # padding
+
+            n = self._read_uint32()
+            txtlen = self._read_uint32()
+            off = np.frombuffer(
+                self._path_or_buf.read(4 * n), dtype=f"{self._byteorder}i4", count=n
+            )
+            val = np.frombuffer(
+                self._path_or_buf.read(4 * n), dtype=f"{self._byteorder}i4", count=n
+            )
+            ii = np.argsort(off)
+            off = off[ii]
+            val = val[ii]
+            txt = self._path_or_buf.read(txtlen)
+            self._value_label_dict[labname] = {}
+            for i in range(n):
+                end = off[i + 1] if i < n - 1 else txtlen
+                self._value_label_dict[labname][val[i]] = self._decode(
+                    txt[off[i] : end]
+                )
+
+            if self._format_version >= 117:
+                self._path_or_buf.read(6)  # </lbl>
+
+    def _read_old_value_labels(self) -> None:
+        """Reads value labels with fixed-length strings (105 and earlier format)"""
+        assert self._dtype is not None
+        offset = self._nobs * self._dtype.itemsize
+        self._path_or_buf.seek(self._data_location + offset)
+
+        while True:
+            if not self._path_or_buf.read(2):
+                # end-of-file may have been reached, if so stop here
+                break
+
+            # otherwise back up and read again, taking byteorder into account
+            self._path_or_buf.seek(-2, os.SEEK_CUR)
+            n = self._read_uint16()
+            labname = self._decode(self._path_or_buf.read(9))
+            self._path_or_buf.read(1)  # padding
+            codes = np.frombuffer(
+                self._path_or_buf.read(2 * n), dtype=f"{self._byteorder}i2", count=n
+            )
+            self._value_label_dict[labname] = {}
+            for i in range(n):
+                self._value_label_dict[labname][codes[i]] = self._decode(
+                    self._path_or_buf.read(8)
+                )
+
+    def _read_value_labels(self) -> None:
+        self._ensure_open()
+        if self._value_labels_read:
+            # Don't read twice
+            return
+
+        if self._format_version >= 108:
+            self._read_new_value_labels()
+        else:
+            self._read_old_value_labels()
+        self._value_labels_read = True
+
+    def _read_strls(self) -> None:
+        self._path_or_buf.seek(self._seek_strls)
+        # Wrap v_o in a string to allow uint64 values as keys on 32bit OS
+        self.GSO = {"0": ""}
+        while True:
+            if self._path_or_buf.read(3) != b"GSO":
+                break
+
+            if self._format_version == 117:
+                v_o = self._read_uint64()
+            else:
+                buf = self._path_or_buf.read(12)
+                # Only tested on little endian machine.
+                v_size = 2 if self._format_version == 118 else 3
+                if self._byteorder == "<":
+                    buf = buf[0:v_size] + buf[4 : (12 - v_size)]
+                else:
+                    buf = buf[4 - v_size : 4] + buf[(4 + v_size) :]
+                v_o = struct.unpack(f"{self._byteorder}Q", buf)[0]
+            typ = self._read_uint8()
+            length = self._read_uint32()
+            va = self._path_or_buf.read(length)
+            if typ == 130:
+                decoded_va = va[0:-1].decode(self._encoding)
+            else:
+                # Stata says typ 129 can be binary, so use str
+                decoded_va = str(va)
+                # Wrap v_o in a string to allow uint64 values as keys on 32bit OS
+            self.GSO[str(v_o)] = decoded_va
+
+    def __next__(self) -> DataFrame:
+        self._using_iterator = True
+        return self.read(nrows=self._chunksize)
+
+    def get_chunk(self, size: int | None = None) -> DataFrame:
+        """
+        Reads lines from Stata file and returns as dataframe
+
+        Parameters
+        ----------
+        size : int, defaults to None
+            Number of lines to read.  If None, reads whole file.
+
+        Returns
+        -------
+        DataFrame
+        """
+        if size is None:
+            size = self._chunksize
+        return self.read(nrows=size)
+
+    @Appender(_read_method_doc)
+    def read(
+        self,
+        nrows: int | None = None,
+        convert_dates: bool | None = None,
+        convert_categoricals: bool | None = None,
+        index_col: str | None = None,
+        convert_missing: bool | None = None,
+        preserve_dtypes: bool | None = None,
+        columns: Sequence[str] | None = None,
+        order_categoricals: bool | None = None,
+    ) -> DataFrame:
+        self._ensure_open()
+
+        # Handle options
+        if convert_dates is None:
+            convert_dates = self._convert_dates
+        if convert_categoricals is None:
+            convert_categoricals = self._convert_categoricals
+        if convert_missing is None:
+            convert_missing = self._convert_missing
+        if preserve_dtypes is None:
+            preserve_dtypes = self._preserve_dtypes
+        if columns is None:
+            columns = self._columns
+        if order_categoricals is None:
+            order_categoricals = self._order_categoricals
+        if index_col is None:
+            index_col = self._index_col
+        if nrows is None:
+            nrows = self._nobs
+
+        # Handle empty file or chunk.  If reading incrementally raise
+        # StopIteration.  If reading the whole thing return an empty
+        # data frame.
+        if (self._nobs == 0) and nrows == 0:
+            data = DataFrame(columns=self._varlist)
+            # Apply dtypes correctly
+            for i, col in enumerate(data.columns):
+                dt = self._dtyplist[i]
+                if isinstance(dt, np.dtype):
+                    if dt.char != "S":
+                        data[col] = data[col].astype(dt)
+            if columns is not None:
+                data = self._do_select_columns(data, columns)
+            return data
+
+        if (self._format_version >= 117) and (not self._value_labels_read):
+            self._read_strls()
+
+        # Read data
+        assert self._dtype is not None
+        dtype = self._dtype
+        max_read_len = (self._nobs - self._lines_read) * dtype.itemsize
+        read_len = nrows * dtype.itemsize
+        read_len = min(read_len, max_read_len)
+        if read_len <= 0:
+            # Iterator has finished, should never be here unless
+            # we are reading the file incrementally
+            if convert_categoricals:
+                self._read_value_labels()
+            raise StopIteration
+        offset = self._lines_read * dtype.itemsize
+        self._path_or_buf.seek(self._data_location + offset)
+        read_lines = min(nrows, self._nobs - self._lines_read)
+        raw_data = np.frombuffer(
+            self._path_or_buf.read(read_len), dtype=dtype, count=read_lines
+        )
+
+        self._lines_read += read_lines
+
+        # if necessary, swap the byte order to native here
+        if self._byteorder != self._native_byteorder:
+            raw_data = raw_data.byteswap().view(raw_data.dtype.newbyteorder())
+
+        if convert_categoricals:
+            self._read_value_labels()
+
+        if len(raw_data) == 0:
+            data = DataFrame(columns=self._varlist)
+        else:
+            data = DataFrame.from_records(raw_data)
+            data.columns = Index(self._varlist)
+
+        # If index is not specified, use actual row number rather than
+        # restarting at 0 for each chunk.
+        if index_col is None:
+            data.index = RangeIndex(
+                self._lines_read - read_lines, self._lines_read
+            )  # set attr instead of set_index to avoid copy
+
+        if columns is not None:
+            data = self._do_select_columns(data, columns)
+
+        # Decode strings
+        for col, typ in zip(data, self._typlist):
+            if isinstance(typ, int):
+                data[col] = data[col].apply(self._decode)
+
+        data = self._insert_strls(data)
+
+        # Convert columns (if needed) to match input type
+        valid_dtypes = [i for i, dtyp in enumerate(self._dtyplist) if dtyp is not None]
+        object_type = np.dtype(object)
+        for idx in valid_dtypes:
+            dtype = data.iloc[:, idx].dtype
+            if dtype not in (object_type, self._dtyplist[idx]):
+                data.isetitem(idx, data.iloc[:, idx].astype(dtype))
+
+        data = self._do_convert_missing(data, convert_missing)
+
+        if convert_dates:
+            for i, fmt in enumerate(self._fmtlist):
+                if any(fmt.startswith(date_fmt) for date_fmt in _date_formats):
+                    data.isetitem(
+                        i, _stata_elapsed_date_to_datetime_vec(data.iloc[:, i], fmt)
+                    )
+
+        if convert_categoricals:
+            data = self._do_convert_categoricals(
+                data, self._value_label_dict, self._lbllist, order_categoricals
+            )
+
+        if not preserve_dtypes:
+            retyped_data = []
+            convert = False
+            for col in data:
+                dtype = data[col].dtype
+                if dtype in (np.dtype(np.float16), np.dtype(np.float32)):
+                    dtype = np.dtype(np.float64)
+                    convert = True
+                elif dtype in (
+                    np.dtype(np.int8),
+                    np.dtype(np.int16),
+                    np.dtype(np.int32),
+                ):
+                    dtype = np.dtype(np.int64)
+                    convert = True
+                retyped_data.append((col, data[col].astype(dtype)))
+            if convert:
+                data = DataFrame.from_dict(dict(retyped_data))
+
+        if index_col is not None:
+            data = data.set_index(data.pop(index_col))
+
+        return data
+
+    def _do_convert_missing(self, data: DataFrame, convert_missing: bool) -> DataFrame:
+        # missing code for double was different in version 105 and prior
+        old_missingdouble = float.fromhex("0x1.0p333")
+
+        # Check for missing values, and replace if found
+        replacements = {}
+        for i in range(len(data.columns)):
+            fmt = self._typlist[i]
+            # recode instances of the old missing code to the currently used value
+            if self._format_version <= 105 and fmt == "d":
+                data.iloc[:, i] = data.iloc[:, i].replace(
+                    old_missingdouble, self.MISSING_VALUES["d"]
+                )
+
+            if self._format_version <= 111:
+                if fmt not in self.OLD_VALID_RANGE:
+                    continue
+
+                fmt = cast(str, fmt)  # only strs in OLD_VALID_RANGE
+                nmin, nmax = self.OLD_VALID_RANGE[fmt]
+            else:
+                if fmt not in self.VALID_RANGE:
+                    continue
+
+                fmt = cast(str, fmt)  # only strs in VALID_RANGE
+                nmin, nmax = self.VALID_RANGE[fmt]
+            series = data.iloc[:, i]
+
+            # appreciably faster to do this with ndarray instead of Series
+            svals = series._values
+            missing = (svals < nmin) | (svals > nmax)
+
+            if not missing.any():
+                continue
+
+            if convert_missing:  # Replacement follows Stata notation
+                missing_loc = np.nonzero(np.asarray(missing))[0]
+                umissing, umissing_loc = np.unique(series[missing], return_inverse=True)
+                replacement = Series(series, dtype=object)
+                for j, um in enumerate(umissing):
+                    if self._format_version <= 111:
+                        missing_value = StataMissingValue(
+                            float(self.MISSING_VALUES[fmt])
+                        )
+                    else:
+                        missing_value = StataMissingValue(um)
+
+                    loc = missing_loc[umissing_loc == j]
+                    replacement.iloc[loc] = missing_value
+            else:  # All replacements are identical
+                dtype = series.dtype
+                if dtype not in (np.float32, np.float64):
+                    dtype = np.float64
+                replacement = Series(series, dtype=dtype)
+                # Note: operating on ._values is much faster than directly
+                # TODO: can we fix that?
+                replacement._values[missing] = np.nan
+            replacements[i] = replacement
+        if replacements:
+            for idx, value in replacements.items():
+                data.isetitem(idx, value)
+        return data
+
+    def _insert_strls(self, data: DataFrame) -> DataFrame:
+        if not hasattr(self, "GSO") or len(self.GSO) == 0:
+            return data
+        for i, typ in enumerate(self._typlist):
+            if typ != "Q":
+                continue
+            # Wrap v_o in a string to allow uint64 values as keys on 32bit OS
+            data.isetitem(i, [self.GSO[str(k)] for k in data.iloc[:, i]])
+        return data
+
+    def _do_select_columns(self, data: DataFrame, columns: Sequence[str]) -> DataFrame:
+        if not self._column_selector_set:
+            column_set = set(columns)
+            if len(column_set) != len(columns):
+                raise ValueError("columns contains duplicate entries")
+            unmatched = column_set.difference(data.columns)
+            if unmatched:
+                joined = ", ".join(list(unmatched))
+                raise ValueError(
+                    "The following columns were not "
+                    f"found in the Stata data set: {joined}"
+                )
+            # Copy information for retained columns for later processing
+            dtyplist = []
+            typlist = []
+            fmtlist = []
+            lbllist = []
+            for col in columns:
+                i = data.columns.get_loc(col)  # type: ignore[no-untyped-call]
+                dtyplist.append(self._dtyplist[i])
+                typlist.append(self._typlist[i])
+                fmtlist.append(self._fmtlist[i])
+                lbllist.append(self._lbllist[i])
+
+            self._dtyplist = dtyplist
+            self._typlist = typlist
+            self._fmtlist = fmtlist
+            self._lbllist = lbllist
+            self._column_selector_set = True
+
+        return data[columns]
+
+    def _do_convert_categoricals(
+        self,
+        data: DataFrame,
+        value_label_dict: dict[str, dict[int, str]],
+        lbllist: Sequence[str],
+        order_categoricals: bool,
+    ) -> DataFrame:
+        """
+        Converts categorical columns to Categorical type.
+        """
+        if not value_label_dict:
+            return data
+        cat_converted_data = []
+        for col, label in zip(data, lbllist):
+            if label in value_label_dict:
+                # Explicit call with ordered=True
+                vl = value_label_dict[label]
+                keys = np.array(list(vl.keys()))
+                column = data[col]
+                key_matches = column.isin(keys)
+                if self._using_iterator and key_matches.all():
+                    initial_categories: np.ndarray | None = keys
+                    # If all categories are in the keys and we are iterating,
+                    # use the same keys for all chunks. If some are missing
+                    # value labels, then we will fall back to the categories
+                    # varying across chunks.
+                else:
+                    if self._using_iterator:
+                        # warn is using an iterator
+                        warnings.warn(
+                            categorical_conversion_warning,
+                            CategoricalConversionWarning,
+                            stacklevel=find_stack_level(),
+                        )
+                    initial_categories = None
+                cat_data = Categorical(
+                    column, categories=initial_categories, ordered=order_categoricals
+                )
+                if initial_categories is None:
+                    # If None here, then we need to match the cats in the Categorical
+                    categories = []
+                    for category in cat_data.categories:
+                        if category in vl:
+                            categories.append(vl[category])
+                        else:
+                            categories.append(category)
+                else:
+                    # If all cats are matched, we can use the values
+                    categories = list(vl.values())
+                try:
+                    # Try to catch duplicate categories
+                    # TODO: if we get a non-copying rename_categories, use that
+                    cat_data = cat_data.rename_categories(categories)
+                except ValueError as err:
+                    vc = Series(categories, copy=False).value_counts()
+                    repeated_cats = list(vc.index[vc > 1])
+                    repeats = "-" * 80 + "\n" + "\n".join(repeated_cats)
+                    # GH 25772
+                    msg = f"""
+Value labels for column {col} are not unique. These cannot be converted to
+pandas categoricals.
+
+Either read the file with `convert_categoricals` set to False or use the
+low level interface in `StataReader` to separately read the values and the
+value_labels.
+
+The repeated labels are:
+{repeats}
+"""
+                    raise ValueError(msg) from err
+                # TODO: is the next line needed above in the data(...) method?
+                cat_series = Series(cat_data, index=data.index, copy=False)
+                cat_converted_data.append((col, cat_series))
+            else:
+                cat_converted_data.append((col, data[col]))
+        data = DataFrame(dict(cat_converted_data), copy=False)
+        return data
+
+    @property
+    def data_label(self) -> str:
+        """
+        Return data label of Stata file.
+
+        The data label is a descriptive string associated with the dataset
+        stored in the Stata file. This property provides access to that
+        label, if one is present.
+
+        See Also
+        --------
+        io.stata.StataReader.variable_labels : Return a dict associating each variable
+            name with corresponding label.
+        DataFrame.to_stata : Export DataFrame object to Stata dta format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([(1,)], columns=["variable"])
+        >>> time_stamp = pd.Timestamp(2000, 2, 29, 14, 21)
+        >>> data_label = "This is a data file."
+        >>> path = "/My_path/filename.dta"
+        >>> df.to_stata(
+        ...     path,
+        ...     time_stamp=time_stamp,  # doctest: +SKIP
+        ...     data_label=data_label,  # doctest: +SKIP
+        ...     version=None,
+        ... )  # doctest: +SKIP
+        >>> with pd.io.stata.StataReader(path) as reader:  # doctest: +SKIP
+        ...     print(reader.data_label)  # doctest: +SKIP
+        This is a data file.
+        """
+        self._ensure_open()
+        return self._data_label
+
+    @property
+    def time_stamp(self) -> str:
+        """
+        Return time stamp of Stata file.
+        """
+        self._ensure_open()
+        return self._time_stamp
+
+    def variable_labels(self) -> dict[str, str]:
+        """
+        Return a dict associating each variable name with corresponding label.
+
+        This method retrieves variable labels from a Stata file. Variable labels are
+        mappings between variable names and their corresponding descriptive labels
+        in a Stata dataset.
+
+        Returns
+        -------
+        dict
+            A python dictionary.
+
+        See Also
+        --------
+        read_stata : Read Stata file into DataFrame.
+        DataFrame.to_stata : Export DataFrame object to Stata dta format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["col_1", "col_2"])
+        >>> time_stamp = pd.Timestamp(2000, 2, 29, 14, 21)
+        >>> path = "/My_path/filename.dta"
+        >>> variable_labels = {"col_1": "This is an example"}
+        >>> df.to_stata(
+        ...     path,
+        ...     time_stamp=time_stamp,  # doctest: +SKIP
+        ...     variable_labels=variable_labels,
+        ...     version=None,
+        ... )  # doctest: +SKIP
+        >>> with pd.io.stata.StataReader(path) as reader:  # doctest: +SKIP
+        ...     print(reader.variable_labels())  # doctest: +SKIP
+        {'index': '', 'col_1': 'This is an example', 'col_2': ''}
+        >>> pd.read_stata(path)  # doctest: +SKIP
+            index col_1 col_2
+        0       0    1    2
+        1       1    3    4
+        """
+        self._ensure_open()
+        return dict(zip(self._varlist, self._variable_labels))
+
+    def value_labels(self) -> dict[str, dict[int, str]]:
+        """
+        Return a nested dict associating each variable name to its value and label.
+
+        This method retrieves the value labels from a Stata file. Value labels are
+        mappings between the coded values and their corresponding descriptive labels
+        in a Stata dataset.
+
+        Returns
+        -------
+        dict
+            A python dictionary.
+
+        See Also
+        --------
+        read_stata : Read Stata file into DataFrame.
+        DataFrame.to_stata : Export DataFrame object to Stata dta format.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame([[1, 2], [3, 4]], columns=["col_1", "col_2"])
+        >>> time_stamp = pd.Timestamp(2000, 2, 29, 14, 21)
+        >>> path = "/My_path/filename.dta"
+        >>> value_labels = {"col_1": {3: "x"}}
+        >>> df.to_stata(
+        ...     path,
+        ...     time_stamp=time_stamp,  # doctest: +SKIP
+        ...     value_labels=value_labels,
+        ...     version=None,
+        ... )  # doctest: +SKIP
+        >>> with pd.io.stata.StataReader(path) as reader:  # doctest: +SKIP
+        ...     print(reader.value_labels())  # doctest: +SKIP
+        {'col_1': {3: 'x'}}
+        >>> pd.read_stata(path)  # doctest: +SKIP
+            index col_1 col_2
+        0       0    1    2
+        1       1    x    4
+        """
+        if not self._value_labels_read:
+            self._read_value_labels()
+
+        return self._value_label_dict
+
+
+@Appender(_read_stata_doc)
+def read_stata(
+    filepath_or_buffer: FilePath | ReadBuffer[bytes],
+    *,
+    convert_dates: bool = True,
+    convert_categoricals: bool = True,
+    index_col: str | None = None,
+    convert_missing: bool = False,
+    preserve_dtypes: bool = True,
+    columns: Sequence[str] | None = None,
+    order_categoricals: bool = True,
+    chunksize: int | None = None,
+    iterator: bool = False,
+    compression: CompressionOptions = "infer",
+    storage_options: StorageOptions | None = None,
+) -> DataFrame | StataReader:
+    reader = StataReader(
+        filepath_or_buffer,
+        convert_dates=convert_dates,
+        convert_categoricals=convert_categoricals,
+        index_col=index_col,
+        convert_missing=convert_missing,
+        preserve_dtypes=preserve_dtypes,
+        columns=columns,
+        order_categoricals=order_categoricals,
+        chunksize=chunksize,
+        storage_options=storage_options,
+        compression=compression,
+    )
+
+    if iterator or chunksize:
+        return reader
+
+    with reader:
+        return reader.read()
+
+
+def _set_endianness(endianness: str) -> str:
+    if endianness.lower() in ["<", "little"]:
+        return "<"
+    elif endianness.lower() in [">", "big"]:
+        return ">"
+    else:  # pragma : no cover
+        raise ValueError(f"Endianness {endianness} not understood")
+
+
+def _pad_bytes(name: AnyStr, length: int) -> AnyStr:
+    """
+    Take a char string and pads it with null bytes until it's length chars.
+    """
+    if isinstance(name, bytes):
+        return name + b"\x00" * (length - len(name))
+    return name + "\x00" * (length - len(name))
+
+
+def _convert_datetime_to_stata_type(fmt: str) -> np.dtype:
+    """
+    Convert from one of the stata date formats to a type in TYPE_MAP.
+    """
+    if fmt in [
+        "tc",
+        "%tc",
+        "td",
+        "%td",
+        "tw",
+        "%tw",
+        "tm",
+        "%tm",
+        "tq",
+        "%tq",
+        "th",
+        "%th",
+        "ty",
+        "%ty",
+    ]:
+        return np.dtype(np.float64)  # Stata expects doubles for SIFs
+    else:
+        raise NotImplementedError(f"Format {fmt} not implemented")
+
+
+def _maybe_convert_to_int_keys(convert_dates: dict, varlist: list[Hashable]) -> dict:
+    new_dict = {}
+    for key, value in convert_dates.items():
+        if not convert_dates[key].startswith("%"):  # make sure proper fmts
+            convert_dates[key] = "%" + value
+        if key in varlist:
+            new_dict[varlist.index(key)] = convert_dates[key]
+        else:
+            if not isinstance(key, int):
+                raise ValueError("convert_dates key must be a column or an integer")
+            new_dict[key] = convert_dates[key]
+    return new_dict
+
+
+def _dtype_to_stata_type(dtype: np.dtype, column: Series) -> int:
+    """
+    Convert dtype types to stata types. Returns the byte of the given ordinal.
+    See TYPE_MAP and comments for an explanation. This is also explained in
+    the dta spec.
+    1 - 244 are strings of this length
+                         Pandas    Stata
+    251 - for int8      byte
+    252 - for int16     int
+    253 - for int32     long
+    254 - for float32   float
+    255 - for double    double
+
+    If there are dates to convert, then dtype will already have the correct
+    type inserted.
+    """
+    # TODO: expand to handle datetime to integer conversion
+    if dtype.type is np.object_:  # try to coerce it to the biggest string
+        # not memory efficient, what else could we
+        # do?
+        itemsize = max_len_string_array(ensure_object(column._values))
+        return max(itemsize, 1)
+    elif dtype.type is np.float64:
+        return 255
+    elif dtype.type is np.float32:
+        return 254
+    elif dtype.type is np.int32:
+        return 253
+    elif dtype.type is np.int16:
+        return 252
+    elif dtype.type is np.int8:
+        return 251
+    else:  # pragma : no cover
+        raise NotImplementedError(f"Data type {dtype} not supported.")
+
+
+def _dtype_to_default_stata_fmt(
+    dtype: np.dtype, column: Series, dta_version: int = 114, force_strl: bool = False
+) -> str:
+    """
+    Map numpy dtype to stata's default format for this type. Not terribly
+    important since users can change this in Stata. Semantics are
+
+    object  -> "%DDs" where DD is the length of the string.  If not a string,
+                raise ValueError
+    float64 -> "%10.0g"
+    float32 -> "%9.0g"
+    int64   -> "%9.0g"
+    int32   -> "%12.0g"
+    int16   -> "%8.0g"
+    int8    -> "%8.0g"
+    strl    -> "%9s"
+    """
+    # TODO: Refactor to combine type with format
+    # TODO: expand this to handle a default datetime format?
+    if dta_version < 117:
+        max_str_len = 244
+    else:
+        max_str_len = 2045
+        if force_strl:
+            return "%9s"
+    if dtype.type is np.object_:
+        itemsize = max_len_string_array(ensure_object(column._values))
+        if itemsize > max_str_len:
+            if dta_version >= 117:
+                return "%9s"
+            else:
+                raise ValueError(excessive_string_length_error.format(column.name))
+        return "%" + str(max(itemsize, 1)) + "s"
+    elif dtype == np.float64:
+        return "%10.0g"
+    elif dtype == np.float32:
+        return "%9.0g"
+    elif dtype == np.int32:
+        return "%12.0g"
+    elif dtype in (np.int8, np.int16):
+        return "%8.0g"
+    else:  # pragma : no cover
+        raise NotImplementedError(f"Data type {dtype} not supported.")
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    compression_options=_shared_docs["compression_options"] % "fname",
+)
+class StataWriter(StataParser):
+    """
+    A class for writing Stata binary dta files
+
+    Parameters
+    ----------
+    fname : path (string), buffer or path object
+        string, pathlib.Path or
+        object implementing a binary write() functions. If using a buffer
+        then the buffer will not be automatically closed after the file
+        is written.
+    data : DataFrame
+        Input to save
+    convert_dates : dict
+        Dictionary mapping columns containing datetime types to stata internal
+        format to use when writing the dates. Options are 'tc', 'td', 'tm',
+        'tw', 'th', 'tq', 'ty'. Column can be either an integer or a name.
+        Datetime columns that do not have a conversion type specified will be
+        converted to 'tc'. Raises NotImplementedError if a datetime column has
+        timezone information
+    write_index : bool
+        Write the index to Stata dataset.
+    byteorder : str
+        Can be ">", "<", "little", or "big". default is `sys.byteorder`
+    time_stamp : datetime
+        A datetime to use as file creation date.  Default is the current time
+    data_label : str
+        A label for the data set.  Must be 80 characters or smaller.
+    variable_labels : dict
+        Dictionary containing columns as keys and variable labels as values.
+        Each label must be 80 characters or smaller.
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    {storage_options}
+
+    value_labels : dict of dicts
+        Dictionary containing columns as keys and dictionaries of column value
+        to labels as values. The combined length of all labels for a single
+        variable must be 32,000 characters or smaller.
+
+        .. versionadded:: 1.4.0
+
+    Returns
+    -------
+    writer : StataWriter instance
+        The StataWriter instance has a write_file method, which will
+        write the file to the given `fname`.
+
+    Raises
+    ------
+    NotImplementedError
+        * If datetimes contain timezone information
+    ValueError
+        * Columns listed in convert_dates are neither datetime64[ns]
+          or datetime
+        * Column dtype is not representable in Stata
+        * Column listed in convert_dates is not in DataFrame
+        * Categorical label contains more than 32,000 characters
+
+    Examples
+    --------
+    >>> data = pd.DataFrame([[1.0, 1]], columns=["a", "b"])
+    >>> writer = StataWriter("./data_file.dta", data)
+    >>> writer.write_file()
+
+    Directly write a zip file
+    >>> compression = {{"method": "zip", "archive_name": "data_file.dta"}}
+    >>> writer = StataWriter("./data_file.zip", data, compression=compression)
+    >>> writer.write_file()
+
+    Save a DataFrame with dates
+    >>> from datetime import datetime
+    >>> data = pd.DataFrame([[datetime(2000, 1, 1)]], columns=["date"])
+    >>> writer = StataWriter("./date_data_file.dta", data, {{"date": "tw"}})
+    >>> writer.write_file()
+    """
+
+    _max_string_length = 244
+    _encoding: Literal["latin-1", "utf-8"] = "latin-1"
+
+    def __init__(
+        self,
+        fname: FilePath | WriteBuffer[bytes],
+        data: DataFrame,
+        convert_dates: dict[Hashable, str] | None = None,
+        write_index: bool = True,
+        byteorder: str | None = None,
+        time_stamp: datetime | None = None,
+        data_label: str | None = None,
+        variable_labels: dict[Hashable, str] | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+        *,
+        value_labels: dict[Hashable, dict[float, str]] | None = None,
+    ) -> None:
+        super().__init__()
+        self.data = data
+        self._convert_dates = {} if convert_dates is None else convert_dates
+        self._write_index = write_index
+        self._time_stamp = time_stamp
+        self._data_label = data_label
+        self._variable_labels = variable_labels
+        self._non_cat_value_labels = value_labels
+        self._value_labels: list[StataValueLabel] = []
+        self._has_value_labels = np.array([], dtype=bool)
+        self._compression = compression
+        self._output_file: IO[bytes] | None = None
+        self._converted_names: dict[Hashable, str] = {}
+        # attach nobs, nvars, data, varlist, typlist
+        self._prepare_pandas(data)
+        self.storage_options = storage_options
+
+        if byteorder is None:
+            byteorder = sys.byteorder
+        self._byteorder = _set_endianness(byteorder)
+        self._fname = fname
+        self.type_converters = {253: np.int32, 252: np.int16, 251: np.int8}
+
+    def _write(self, to_write: str) -> None:
+        """
+        Helper to call encode before writing to file for Python 3 compat.
+        """
+        self.handles.handle.write(to_write.encode(self._encoding))
+
+    def _write_bytes(self, value: bytes) -> None:
+        """
+        Helper to assert file is open before writing.
+        """
+        self.handles.handle.write(value)
+
+    def _prepare_non_cat_value_labels(
+        self, data: DataFrame
+    ) -> list[StataNonCatValueLabel]:
+        """
+        Check for value labels provided for non-categorical columns. Value
+        labels
+        """
+        non_cat_value_labels: list[StataNonCatValueLabel] = []
+        if self._non_cat_value_labels is None:
+            return non_cat_value_labels
+
+        for labname, labels in self._non_cat_value_labels.items():
+            if labname in self._converted_names:
+                colname = self._converted_names[labname]
+            elif labname in data.columns:
+                colname = str(labname)
+            else:
+                raise KeyError(
+                    f"Can't create value labels for {labname}, it wasn't "
+                    "found in the dataset."
+                )
+
+            if not is_numeric_dtype(data[colname].dtype):
+                # Labels should not be passed explicitly for categorical
+                # columns that will be converted to int
+                raise ValueError(
+                    f"Can't create value labels for {labname}, value labels "
+                    "can only be applied to numeric columns."
+                )
+            svl = StataNonCatValueLabel(colname, labels, self._encoding)
+            non_cat_value_labels.append(svl)
+        return non_cat_value_labels
+
+    def _prepare_categoricals(self, data: DataFrame) -> DataFrame:
+        """
+        Check for categorical columns, retain categorical information for
+        Stata file and convert categorical data to int
+        """
+        is_cat = [isinstance(dtype, CategoricalDtype) for dtype in data.dtypes]
+        if not any(is_cat):
+            return data
+
+        self._has_value_labels |= np.array(is_cat)
+
+        get_base_missing_value = StataMissingValue.get_base_missing_value
+        data_formatted = []
+        for col, col_is_cat in zip(data, is_cat):
+            if col_is_cat:
+                svl = StataValueLabel(data[col], encoding=self._encoding)
+                self._value_labels.append(svl)
+                dtype = data[col].cat.codes.dtype
+                if dtype == np.int64:
+                    raise ValueError(
+                        "It is not possible to export "
+                        "int64-based categorical data to Stata."
+                    )
+                values = data[col].cat.codes._values.copy()
+
+                # Upcast if needed so that correct missing values can be set
+                if values.max() >= get_base_missing_value(dtype):
+                    if dtype == np.int8:
+                        dtype = np.dtype(np.int16)
+                    elif dtype == np.int16:
+                        dtype = np.dtype(np.int32)
+                    else:
+                        dtype = np.dtype(np.float64)
+                    values = np.array(values, dtype=dtype)
+
+                # Replace missing values with Stata missing value for type
+                values[values == -1] = get_base_missing_value(dtype)
+                data_formatted.append((col, values))
+            else:
+                data_formatted.append((col, data[col]))
+        return DataFrame.from_dict(dict(data_formatted))
+
+    def _replace_nans(self, data: DataFrame) -> DataFrame:
+        # return data
+        """
+        Checks floating point data columns for nans, and replaces these with
+        the generic Stata for missing value (.)
+        """
+        for c in data:
+            dtype = data[c].dtype
+            if dtype in (np.float32, np.float64):
+                if dtype == np.float32:
+                    replacement = self.MISSING_VALUES["f"]
+                else:
+                    replacement = self.MISSING_VALUES["d"]
+                data[c] = data[c].fillna(replacement)
+
+        return data
+
+    def _update_strl_names(self) -> None:
+        """No-op, forward compatibility"""
+
+    def _validate_variable_name(self, name: str) -> str:
+        """
+        Validate variable names for Stata export.
+
+        Parameters
+        ----------
+        name : str
+            Variable name
+
+        Returns
+        -------
+        str
+            The validated name with invalid characters replaced with
+            underscores.
+
+        Notes
+        -----
+        Stata 114 and 117 support ascii characters in a-z, A-Z, 0-9
+        and _.
+        """
+        for c in name:
+            if (
+                (c < "A" or c > "Z")
+                and (c < "a" or c > "z")
+                and (c < "0" or c > "9")
+                and c != "_"
+            ):
+                name = name.replace(c, "_")
+        return name
+
+    def _check_column_names(self, data: DataFrame) -> DataFrame:
+        """
+        Checks column names to ensure that they are valid Stata column names.
+        This includes checks for:
+            * Non-string names
+            * Stata keywords
+            * Variables that start with numbers
+            * Variables with names that are too long
+
+        When an illegal variable name is detected, it is converted, and if
+        dates are exported, the variable name is propagated to the date
+        conversion dictionary
+        """
+        converted_names: dict[Hashable, str] = {}
+        columns = list(data.columns)
+        original_columns = columns[:]
+
+        duplicate_var_id = 0
+        for j, name in enumerate(columns):
+            orig_name = name
+            if not isinstance(name, str):
+                name = str(name)
+
+            name = self._validate_variable_name(name)
+
+            # Variable name must not be a reserved word
+            if name in self.RESERVED_WORDS:
+                name = "_" + name
+
+            # Variable name may not start with a number
+            if "0" <= name[0] <= "9":
+                name = "_" + name
+
+            name = name[: min(len(name), 32)]
+
+            if not name == orig_name:
+                # check for duplicates
+                while columns.count(name) > 0:
+                    # prepend ascending number to avoid duplicates
+                    name = "_" + str(duplicate_var_id) + name
+                    name = name[: min(len(name), 32)]
+                    duplicate_var_id += 1
+                converted_names[orig_name] = name
+
+            columns[j] = name
+
+        data.columns = Index(columns)
+
+        # Check date conversion, and fix key if needed
+        if self._convert_dates:
+            for c, o in zip(columns, original_columns):
+                if c != o:
+                    self._convert_dates[c] = self._convert_dates[o]
+                    del self._convert_dates[o]
+
+        if converted_names:
+            conversion_warning = []
+            for orig_name, name in converted_names.items():
+                msg = f"{orig_name}   ->   {name}"
+                conversion_warning.append(msg)
+
+            ws = invalid_name_doc.format("\n    ".join(conversion_warning))
+            warnings.warn(
+                ws,
+                InvalidColumnName,
+                stacklevel=find_stack_level(),
+            )
+
+        self._converted_names = converted_names
+        self._update_strl_names()
+
+        return data
+
+    def _set_formats_and_types(self, dtypes: Series) -> None:
+        self.fmtlist: list[str] = []
+        self.typlist: list[int] = []
+        for col, dtype in dtypes.items():
+            self.fmtlist.append(_dtype_to_default_stata_fmt(dtype, self.data[col]))
+            self.typlist.append(_dtype_to_stata_type(dtype, self.data[col]))
+
+    def _prepare_pandas(self, data: DataFrame) -> None:
+        # NOTE: we might need a different API / class for pandas objects so
+        # we can set different semantics - handle this with a PR to pandas.io
+
+        data = data.copy()
+
+        if self._write_index:
+            temp = data.reset_index()
+            if isinstance(temp, DataFrame):
+                data = temp
+
+        # Ensure column names are strings
+        data = self._check_column_names(data)
+
+        # Check columns for compatibility with stata, upcast if necessary
+        # Raise if outside the supported range
+        data = _cast_to_stata_types(data)
+
+        # Replace NaNs with Stata missing values
+        data = self._replace_nans(data)
+
+        # Set all columns to initially unlabelled
+        self._has_value_labels = np.repeat(False, data.shape[1])
+
+        # Create value labels for non-categorical data
+        non_cat_value_labels = self._prepare_non_cat_value_labels(data)
+
+        non_cat_columns = [svl.labname for svl in non_cat_value_labels]
+        has_non_cat_val_labels = data.columns.isin(non_cat_columns)
+        self._has_value_labels |= has_non_cat_val_labels
+        self._value_labels.extend(non_cat_value_labels)
+
+        # Convert categoricals to int data, and strip labels
+        data = self._prepare_categoricals(data)
+
+        self.nobs, self.nvar = data.shape
+        self.data = data
+        self.varlist = data.columns.tolist()
+
+        dtypes = data.dtypes
+
+        # Ensure all date columns are converted
+        for col in data:
+            if col in self._convert_dates:
+                continue
+            if lib.is_np_dtype(data[col].dtype, "M"):
+                self._convert_dates[col] = "tc"
+
+        self._convert_dates = _maybe_convert_to_int_keys(
+            self._convert_dates, self.varlist
+        )
+        for key in self._convert_dates:
+            new_type = _convert_datetime_to_stata_type(self._convert_dates[key])
+            dtypes.iloc[key] = np.dtype(new_type)
+
+        # Verify object arrays are strings and encode to bytes
+        self._encode_strings()
+
+        self._set_formats_and_types(dtypes)
+
+        # set the given format for the datetime cols
+        if self._convert_dates is not None:
+            for key in self._convert_dates:
+                if isinstance(key, int):
+                    self.fmtlist[key] = self._convert_dates[key]
+
+    def _encode_strings(self) -> None:
+        """
+        Encode strings in dta-specific encoding
+
+        Do not encode columns marked for date conversion or for strL
+        conversion. The strL converter independently handles conversion and
+        also accepts empty string arrays.
+        """
+        convert_dates = self._convert_dates
+        # _convert_strl is not available in dta 114
+        convert_strl = getattr(self, "_convert_strl", [])
+        for i, col in enumerate(self.data):
+            # Skip columns marked for date conversion or strl conversion
+            if i in convert_dates or col in convert_strl:
+                continue
+            column = self.data[col]
+            dtype = column.dtype
+            # TODO could also handle string dtype here specifically
+            if dtype.type is np.object_:
+                inferred_dtype = infer_dtype(column, skipna=True)
+                if not ((inferred_dtype == "string") or len(column) == 0):
+                    col = column.name
+                    raise ValueError(
+                        f"""\
+Column `{col}` cannot be exported.\n\nOnly string-like object arrays
+containing all strings or a mix of strings and None can be exported.
+Object arrays containing only null values are prohibited. Other object
+types cannot be exported and must first be converted to one of the
+supported types."""
+                    )
+                encoded = self.data[col].str.encode(self._encoding)
+                # If larger than _max_string_length do nothing
+                if (
+                    max_len_string_array(ensure_object(encoded._values))
+                    <= self._max_string_length
+                ):
+                    self.data[col] = encoded
+
+    def write_file(self) -> None:
+        """
+        Export DataFrame object to Stata dta format.
+
+        This method writes the contents of a pandas DataFrame to a `.dta` file
+        compatible with Stata. It includes features for handling value labels,
+        variable types, and metadata like timestamps and data labels. The output
+        file can then be read and used in Stata or other compatible statistical
+        tools.
+
+        See Also
+        --------
+        read_stata : Read Stata file into DataFrame.
+        DataFrame.to_stata : Export DataFrame object to Stata dta format.
+        io.stata.StataWriter : A class for writing Stata binary dta files.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "fully_labelled": [1, 2, 3, 3, 1],
+        ...         "partially_labelled": [1.0, 2.0, np.nan, 9.0, np.nan],
+        ...         "Y": [7, 7, 9, 8, 10],
+        ...         "Z": pd.Categorical(["j", "k", "l", "k", "j"]),
+        ...     }
+        ... )
+        >>> path = "/My_path/filename.dta"
+        >>> labels = {
+        ...     "fully_labelled": {1: "one", 2: "two", 3: "three"},
+        ...     "partially_labelled": {1.0: "one", 2.0: "two"},
+        ... }
+        >>> writer = pd.io.stata.StataWriter(
+        ...     path, df, value_labels=labels
+        ... )  # doctest: +SKIP
+        >>> writer.write_file()  # doctest: +SKIP
+        >>> df = pd.read_stata(path)  # doctest: +SKIP
+        >>> df  # doctest: +SKIP
+            index fully_labelled  partially_labeled  Y  Z
+        0       0            one                one  7  j
+        1       1            two                two  7  k
+        2       2          three                NaN  9  l
+        3       3          three                9.0  8  k
+        4       4            one                NaN 10  j
+        """
+        with get_handle(
+            self._fname,
+            "wb",
+            compression=self._compression,
+            is_text=False,
+            storage_options=self.storage_options,
+        ) as self.handles:
+            if self.handles.compression["method"] is not None:
+                # ZipFile creates a file (with the same name) for each write call.
+                # Write it first into a buffer and then write the buffer to the ZipFile.
+                self._output_file, self.handles.handle = self.handles.handle, BytesIO()
+                self.handles.created_handles.append(self.handles.handle)
+
+            try:
+                self._write_header(
+                    data_label=self._data_label, time_stamp=self._time_stamp
+                )
+                self._write_map()
+                self._write_variable_types()
+                self._write_varnames()
+                self._write_sortlist()
+                self._write_formats()
+                self._write_value_label_names()
+                self._write_variable_labels()
+                self._write_expansion_fields()
+                self._write_characteristics()
+                records = self._prepare_data()
+                self._write_data(records)
+                self._write_strls()
+                self._write_value_labels()
+                self._write_file_close_tag()
+                self._write_map()
+                self._close()
+            except Exception as exc:
+                self.handles.close()
+                if isinstance(self._fname, (str, os.PathLike)) and os.path.isfile(
+                    self._fname
+                ):
+                    try:
+                        os.unlink(self._fname)
+                    except OSError:
+                        warnings.warn(
+                            f"This save was not successful but {self._fname} could not "
+                            "be deleted. This file is not valid.",
+                            ResourceWarning,
+                            stacklevel=find_stack_level(),
+                        )
+                raise exc
+
+    def _close(self) -> None:
+        """
+        Close the file if it was created by the writer.
+
+        If a buffer or file-like object was passed in, for example a GzipFile,
+        then leave this file open for the caller to close.
+        """
+        # write compression
+        if self._output_file is not None:
+            assert isinstance(self.handles.handle, BytesIO)
+            bio, self.handles.handle = self.handles.handle, self._output_file
+            self.handles.handle.write(bio.getvalue())
+
+    def _write_map(self) -> None:
+        """No-op, future compatibility"""
+
+    def _write_file_close_tag(self) -> None:
+        """No-op, future compatibility"""
+
+    def _write_characteristics(self) -> None:
+        """No-op, future compatibility"""
+
+    def _write_strls(self) -> None:
+        """No-op, future compatibility"""
+
+    def _write_expansion_fields(self) -> None:
+        """Write 5 zeros for expansion fields"""
+        self._write(_pad_bytes("", 5))
+
+    def _write_value_labels(self) -> None:
+        for vl in self._value_labels:
+            self._write_bytes(vl.generate_value_label(self._byteorder))
+
+    def _write_header(
+        self,
+        data_label: str | None = None,
+        time_stamp: datetime | None = None,
+    ) -> None:
+        byteorder = self._byteorder
+        # ds_format - just use 114
+        self._write_bytes(struct.pack("b", 114))
+        # byteorder
+        self._write((byteorder == ">" and "\x01") or "\x02")
+        # filetype
+        self._write("\x01")
+        # unused
+        self._write("\x00")
+        # number of vars, 2 bytes
+        self._write_bytes(struct.pack(byteorder + "h", self.nvar)[:2])
+        # number of obs, 4 bytes
+        self._write_bytes(struct.pack(byteorder + "i", self.nobs)[:4])
+        # data label 81 bytes, char, null terminated
+        if data_label is None:
+            self._write_bytes(self._null_terminate_bytes(_pad_bytes("", 80)))
+        else:
+            self._write_bytes(
+                self._null_terminate_bytes(_pad_bytes(data_label[:80], 80))
+            )
+        # time stamp, 18 bytes, char, null terminated
+        # format dd Mon yyyy hh:mm
+        if time_stamp is None:
+            time_stamp = datetime.now()
+        elif not isinstance(time_stamp, datetime):
+            raise ValueError("time_stamp should be datetime type")
+        # GH #13856
+        # Avoid locale-specific month conversion
+        months = [
+            "Jan",
+            "Feb",
+            "Mar",
+            "Apr",
+            "May",
+            "Jun",
+            "Jul",
+            "Aug",
+            "Sep",
+            "Oct",
+            "Nov",
+            "Dec",
+        ]
+        month_lookup = {i + 1: month for i, month in enumerate(months)}
+        ts = (
+            time_stamp.strftime("%d ")
+            + month_lookup[time_stamp.month]
+            + time_stamp.strftime(" %Y %H:%M")
+        )
+        self._write_bytes(self._null_terminate_bytes(ts))
+
+    def _write_variable_types(self) -> None:
+        for typ in self.typlist:
+            self._write_bytes(struct.pack("B", typ))
+
+    def _write_varnames(self) -> None:
+        # varlist names are checked by _check_column_names
+        # varlist, requires null terminated
+        for name in self.varlist:
+            name = self._null_terminate_str(name)
+            name = _pad_bytes(name[:32], 33)
+            self._write(name)
+
+    def _write_sortlist(self) -> None:
+        # srtlist, 2*(nvar+1), int array, encoded by byteorder
+        srtlist = _pad_bytes("", 2 * (self.nvar + 1))
+        self._write(srtlist)
+
+    def _write_formats(self) -> None:
+        # fmtlist, 49*nvar, char array
+        for fmt in self.fmtlist:
+            self._write(_pad_bytes(fmt, 49))
+
+    def _write_value_label_names(self) -> None:
+        # lbllist, 33*nvar, char array
+        for i in range(self.nvar):
+            # Use variable name when categorical
+            if self._has_value_labels[i]:
+                name = self.varlist[i]
+                name = self._null_terminate_str(name)
+                name = _pad_bytes(name[:32], 33)
+                self._write(name)
+            else:  # Default is empty label
+                self._write(_pad_bytes("", 33))
+
+    def _write_variable_labels(self) -> None:
+        # Missing labels are 80 blank characters plus null termination
+        blank = _pad_bytes("", 81)
+
+        if self._variable_labels is None:
+            for i in range(self.nvar):
+                self._write(blank)
+            return
+
+        for col in self.data:
+            if col in self._variable_labels:
+                label = self._variable_labels[col]
+                if len(label) > 80:
+                    raise ValueError("Variable labels must be 80 characters or fewer")
+                is_latin1 = all(ord(c) < 256 for c in label)
+                if not is_latin1:
+                    raise ValueError(
+                        "Variable labels must contain only characters that "
+                        "can be encoded in Latin-1"
+                    )
+                self._write(_pad_bytes(label, 81))
+            else:
+                self._write(blank)
+
+    def _convert_strls(self, data: DataFrame) -> DataFrame:
+        """No-op, future compatibility"""
+        return data
+
+    def _prepare_data(self) -> np.rec.recarray:
+        data = self.data
+        typlist = self.typlist
+        convert_dates = self._convert_dates
+        # 1. Convert dates
+        if self._convert_dates is not None:
+            for i, col in enumerate(data):
+                if i in convert_dates:
+                    data[col] = _datetime_to_stata_elapsed_vec(
+                        data[col], self.fmtlist[i]
+                    )
+        # 2. Convert strls
+        data = self._convert_strls(data)
+
+        # 3. Convert bad string data to '' and pad to correct length
+        dtypes = {}
+        native_byteorder = self._byteorder == _set_endianness(sys.byteorder)
+        for i, col in enumerate(data):
+            typ = typlist[i]
+            if typ <= self._max_string_length:
+                dc = data[col].fillna("")
+                data[col] = dc.apply(_pad_bytes, args=(typ,))
+                stype = f"S{typ}"
+                dtypes[col] = stype
+                data[col] = data[col].astype(stype)
+            else:
+                dtype = data[col].dtype
+                if not native_byteorder:
+                    dtype = dtype.newbyteorder(self._byteorder)
+                dtypes[col] = dtype
+
+        return data.to_records(index=False, column_dtypes=dtypes)
+
+    def _write_data(self, records: np.rec.recarray) -> None:
+        self._write_bytes(records.tobytes())
+
+    @staticmethod
+    def _null_terminate_str(s: str) -> str:
+        s += "\x00"
+        return s
+
+    def _null_terminate_bytes(self, s: str) -> bytes:
+        return self._null_terminate_str(s).encode(self._encoding)
+
+
+def _dtype_to_stata_type_117(dtype: np.dtype, column: Series, force_strl: bool) -> int:
+    """
+    Converts dtype types to stata types. Returns the byte of the given ordinal.
+    See TYPE_MAP and comments for an explanation. This is also explained in
+    the dta spec.
+    1 - 2045 are strings of this length
+                Pandas    Stata
+    32768 - for object    strL
+    65526 - for int8      byte
+    65527 - for int16     int
+    65528 - for int32     long
+    65529 - for float32   float
+    65530 - for double    double
+
+    If there are dates to convert, then dtype will already have the correct
+    type inserted.
+    """
+    # TODO: expand to handle datetime to integer conversion
+    if force_strl:
+        return 32768
+    if dtype.type is np.object_:  # try to coerce it to the biggest string
+        # not memory efficient, what else could we
+        # do?
+        itemsize = max_len_string_array(ensure_object(column._values))
+        itemsize = max(itemsize, 1)
+        if itemsize <= 2045:
+            return itemsize
+        return 32768
+    elif dtype.type is np.float64:
+        return 65526
+    elif dtype.type is np.float32:
+        return 65527
+    elif dtype.type is np.int32:
+        return 65528
+    elif dtype.type is np.int16:
+        return 65529
+    elif dtype.type is np.int8:
+        return 65530
+    else:  # pragma : no cover
+        raise NotImplementedError(f"Data type {dtype} not supported.")
+
+
+def _pad_bytes_new(name: str | bytes, length: int) -> bytes:
+    """
+    Takes a bytes instance and pads it with null bytes until it's length chars.
+    """
+    if isinstance(name, str):
+        name = bytes(name, "utf-8")
+    return name + b"\x00" * (length - len(name))
+
+
+class StataStrLWriter:
+    """
+    Converter for Stata StrLs
+
+    Stata StrLs map 8 byte values to strings which are stored using a
+    dictionary-like format where strings are keyed to two values.
+
+    Parameters
+    ----------
+    df : DataFrame
+        DataFrame to convert
+    columns : Sequence[str]
+        List of columns names to convert to StrL
+    version : int, optional
+        dta version.  Currently supports 117, 118 and 119
+    byteorder : str, optional
+        Can be ">", "<", "little", or "big". default is `sys.byteorder`
+
+    Notes
+    -----
+    Supports creation of the StrL block of a dta file for dta versions
+    117, 118 and 119.  These differ in how the GSO is stored.  118 and
+    119 store the GSO lookup value as a uint32 and a uint64, while 117
+    uses two uint32s. 118 and 119 also encode all strings as unicode
+    which is required by the format.  117 uses 'latin-1' a fixed width
+    encoding that extends the 7-bit ascii table with an additional 128
+    characters.
+    """
+
+    def __init__(
+        self,
+        df: DataFrame,
+        columns: Sequence[str],
+        version: int = 117,
+        byteorder: str | None = None,
+    ) -> None:
+        if version not in (117, 118, 119):
+            raise ValueError("Only dta versions 117, 118 and 119 supported")
+        self._dta_ver = version
+
+        self.df = df
+        self.columns = columns
+        self._gso_table = {"": (0, 0)}
+        if byteorder is None:
+            byteorder = sys.byteorder
+        self._byteorder = _set_endianness(byteorder)
+        # Flag whether chosen byteorder matches the system on which we're running
+        self._native_byteorder = self._byteorder == _set_endianness(sys.byteorder)
+
+        gso_v_type = "I"  # uint32
+        gso_o_type = "Q"  # uint64
+        self._encoding = "utf-8"
+        if version == 117:
+            o_size = 4
+            gso_o_type = "I"  # 117 used uint32
+            self._encoding = "latin-1"
+        elif version == 118:
+            o_size = 6
+        else:  # version == 119
+            o_size = 5
+        if self._native_byteorder:
+            self._o_offet = 2 ** (8 * (8 - o_size))
+        else:
+            self._o_offet = 2 ** (8 * o_size)
+        self._gso_o_type = gso_o_type
+        self._gso_v_type = gso_v_type
+
+    def _convert_key(self, key: tuple[int, int]) -> int:
+        v, o = key
+        if self._native_byteorder:
+            return v + self._o_offet * o
+        else:
+            # v, o will be swapped when applying byteorder
+            return o + self._o_offet * v
+
+    def generate_table(self) -> tuple[dict[str, tuple[int, int]], DataFrame]:
+        """
+        Generates the GSO lookup table for the DataFrame
+
+        Returns
+        -------
+        gso_table : dict
+            Ordered dictionary using the string found as keys
+            and their lookup position (v,o) as values
+        gso_df : DataFrame
+            DataFrame where strl columns have been converted to
+            (v,o) values
+
+        Notes
+        -----
+        Modifies the DataFrame in-place.
+
+        The DataFrame returned encodes the (v,o) values as uint64s. The
+        encoding depends on the dta version, and can be expressed as
+
+        enc = v + o * 2 ** (o_size * 8)
+
+        so that v is stored in the lower bits and o is in the upper
+        bits. o_size is
+
+          * 117: 4
+          * 118: 6
+          * 119: 5
+        """
+        gso_table = self._gso_table
+        gso_df = self.df
+        columns = list(gso_df.columns)
+        selected = gso_df[self.columns]
+        col_index = [(col, columns.index(col)) for col in self.columns]
+        keys = np.empty(selected.shape, dtype=np.uint64)
+        for o, (idx, row) in enumerate(selected.iterrows()):
+            for j, (col, v) in enumerate(col_index):
+                val = row[col]
+                # Allow columns with mixed str and None (GH 23633)
+                val = "" if val is None else val
+                key = gso_table.get(val, None)
+                if key is None:
+                    # Stata prefers human numbers
+                    key = (v + 1, o + 1)
+                    gso_table[val] = key
+                keys[o, j] = self._convert_key(key)
+        for i, col in enumerate(self.columns):
+            gso_df[col] = keys[:, i]
+
+        return gso_table, gso_df
+
+    def generate_blob(self, gso_table: dict[str, tuple[int, int]]) -> bytes:
+        """
+        Generates the binary blob of GSOs that is written to the dta file.
+
+        Parameters
+        ----------
+        gso_table : dict
+            Ordered dictionary (str, vo)
+
+        Returns
+        -------
+        gso : bytes
+            Binary content of dta file to be placed between strl tags
+
+        Notes
+        -----
+        Output format depends on dta version.  117 uses two uint32s to
+        express v and o while 118+ uses a uint32 for v and a uint64 for o.
+        """
+        # Format information
+        # Length includes null term
+        # 117
+        # GSOvvvvooootllllxxxxxxxxxxxxxxx...x
+        #  3  u4  u4 u1 u4  string + null term
+        #
+        # 118, 119
+        # GSOvvvvooooooootllllxxxxxxxxxxxxxxx...x
+        #  3  u4   u8   u1 u4    string + null term
+
+        bio = BytesIO()
+        gso = bytes("GSO", "ascii")
+        gso_type = struct.pack(self._byteorder + "B", 130)
+        null = struct.pack(self._byteorder + "B", 0)
+        v_type = self._byteorder + self._gso_v_type
+        o_type = self._byteorder + self._gso_o_type
+        len_type = self._byteorder + "I"
+        for strl, vo in gso_table.items():
+            if vo == (0, 0):
+                continue
+            v, o = vo
+
+            # GSO
+            bio.write(gso)
+
+            # vvvv
+            bio.write(struct.pack(v_type, v))
+
+            # oooo / oooooooo
+            bio.write(struct.pack(o_type, o))
+
+            # t
+            bio.write(gso_type)
+
+            # llll
+            utf8_string = bytes(strl, "utf-8")
+            bio.write(struct.pack(len_type, len(utf8_string) + 1))
+
+            # xxx...xxx
+            bio.write(utf8_string)
+            bio.write(null)
+
+        return bio.getvalue()
+
+
+class StataWriter117(StataWriter):
+    """
+    A class for writing Stata binary dta files in Stata 13 format (117)
+
+    Parameters
+    ----------
+    fname : path (string), buffer or path object
+        string, pathlib.Path or
+        object implementing a binary write() functions. If using a buffer
+        then the buffer will not be automatically closed after the file
+        is written.
+    data : DataFrame
+        Input to save
+    convert_dates : dict
+        Dictionary mapping columns containing datetime types to stata internal
+        format to use when writing the dates. Options are 'tc', 'td', 'tm',
+        'tw', 'th', 'tq', 'ty'. Column can be either an integer or a name.
+        Datetime columns that do not have a conversion type specified will be
+        converted to 'tc'. Raises NotImplementedError if a datetime column has
+        timezone information
+    write_index : bool
+        Write the index to Stata dataset.
+    byteorder : str
+        Can be ">", "<", "little", or "big". default is `sys.byteorder`
+    time_stamp : datetime
+        A datetime to use as file creation date.  Default is the current time
+    data_label : str
+        A label for the data set.  Must be 80 characters or smaller.
+    variable_labels : dict
+        Dictionary containing columns as keys and variable labels as values.
+        Each label must be 80 characters or smaller.
+    convert_strl : list
+        List of columns names to convert to Stata StrL format.  Columns with
+        more than 2045 characters are automatically written as StrL.
+        Smaller columns can be converted by including the column name.  Using
+        StrLs can reduce output file size when strings are longer than 8
+        characters, and either frequently repeated or sparse.
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    value_labels : dict of dicts
+        Dictionary containing columns as keys and dictionaries of column value
+        to labels as values. The combined length of all labels for a single
+        variable must be 32,000 characters or smaller.
+
+        .. versionadded:: 1.4.0
+
+    Returns
+    -------
+    writer : StataWriter117 instance
+        The StataWriter117 instance has a write_file method, which will
+        write the file to the given `fname`.
+
+    Raises
+    ------
+    NotImplementedError
+        * If datetimes contain timezone information
+    ValueError
+        * Columns listed in convert_dates are neither datetime64[ns]
+          or datetime
+        * Column dtype is not representable in Stata
+        * Column listed in convert_dates is not in DataFrame
+        * Categorical label contains more than 32,000 characters
+
+    Examples
+    --------
+    >>> data = pd.DataFrame([[1.0, 1, "a"]], columns=["a", "b", "c"])
+    >>> writer = pd.io.stata.StataWriter117("./data_file.dta", data)
+    >>> writer.write_file()
+
+    Directly write a zip file
+    >>> compression = {"method": "zip", "archive_name": "data_file.dta"}
+    >>> writer = pd.io.stata.StataWriter117(
+    ...     "./data_file.zip", data, compression=compression
+    ... )
+    >>> writer.write_file()
+
+    Or with long strings stored in strl format
+    >>> data = pd.DataFrame(
+    ...     [["A relatively long string"], [""], [""]], columns=["strls"]
+    ... )
+    >>> writer = pd.io.stata.StataWriter117(
+    ...     "./data_file_with_long_strings.dta", data, convert_strl=["strls"]
+    ... )
+    >>> writer.write_file()
+    """
+
+    _max_string_length = 2045
+    _dta_version = 117
+
+    def __init__(
+        self,
+        fname: FilePath | WriteBuffer[bytes],
+        data: DataFrame,
+        convert_dates: dict[Hashable, str] | None = None,
+        write_index: bool = True,
+        byteorder: str | None = None,
+        time_stamp: datetime | None = None,
+        data_label: str | None = None,
+        variable_labels: dict[Hashable, str] | None = None,
+        convert_strl: Sequence[Hashable] | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+        *,
+        value_labels: dict[Hashable, dict[float, str]] | None = None,
+    ) -> None:
+        # Copy to new list since convert_strl might be modified later
+        self._convert_strl: list[Hashable] = []
+        if convert_strl is not None:
+            self._convert_strl.extend(convert_strl)
+
+        super().__init__(
+            fname,
+            data,
+            convert_dates,
+            write_index,
+            byteorder=byteorder,
+            time_stamp=time_stamp,
+            data_label=data_label,
+            variable_labels=variable_labels,
+            value_labels=value_labels,
+            compression=compression,
+            storage_options=storage_options,
+        )
+        self._map: dict[str, int] = {}
+        self._strl_blob = b""
+
+    @staticmethod
+    def _tag(val: str | bytes, tag: str) -> bytes:
+        """Surround val with <tag></tag>"""
+        if isinstance(val, str):
+            val = bytes(val, "utf-8")
+        return bytes("<" + tag + ">", "utf-8") + val + bytes("</" + tag + ">", "utf-8")
+
+    def _update_map(self, tag: str) -> None:
+        """Update map location for tag with file position"""
+        assert self.handles.handle is not None
+        self._map[tag] = self.handles.handle.tell()
+
+    def _write_header(
+        self,
+        data_label: str | None = None,
+        time_stamp: datetime | None = None,
+    ) -> None:
+        """Write the file header"""
+        byteorder = self._byteorder
+        self._write_bytes(bytes("<stata_dta>", "utf-8"))
+        bio = BytesIO()
+        # ds_format - 117
+        bio.write(self._tag(bytes(str(self._dta_version), "utf-8"), "release"))
+        # byteorder
+        bio.write(self._tag((byteorder == ">" and "MSF") or "LSF", "byteorder"))
+        # number of vars, 2 bytes in 117 and 118, 4 byte in 119
+        nvar_type = "H" if self._dta_version <= 118 else "I"
+        bio.write(self._tag(struct.pack(byteorder + nvar_type, self.nvar), "K"))
+        # 117 uses 4 bytes, 118 uses 8
+        nobs_size = "I" if self._dta_version == 117 else "Q"
+        bio.write(self._tag(struct.pack(byteorder + nobs_size, self.nobs), "N"))
+        # data label 81 bytes, char, null terminated
+        label = data_label[:80] if data_label is not None else ""
+        encoded_label = label.encode(self._encoding)
+        label_size = "B" if self._dta_version == 117 else "H"
+        label_len = struct.pack(byteorder + label_size, len(encoded_label))
+        encoded_label = label_len + encoded_label
+        bio.write(self._tag(encoded_label, "label"))
+        # time stamp, 18 bytes, char, null terminated
+        # format dd Mon yyyy hh:mm
+        if time_stamp is None:
+            time_stamp = datetime.now()
+        elif not isinstance(time_stamp, datetime):
+            raise ValueError("time_stamp should be datetime type")
+        # Avoid locale-specific month conversion
+        months = [
+            "Jan",
+            "Feb",
+            "Mar",
+            "Apr",
+            "May",
+            "Jun",
+            "Jul",
+            "Aug",
+            "Sep",
+            "Oct",
+            "Nov",
+            "Dec",
+        ]
+        month_lookup = {i + 1: month for i, month in enumerate(months)}
+        ts = (
+            time_stamp.strftime("%d ")
+            + month_lookup[time_stamp.month]
+            + time_stamp.strftime(" %Y %H:%M")
+        )
+        # '\x11' added due to inspection of Stata file
+        stata_ts = b"\x11" + bytes(ts, "utf-8")
+        bio.write(self._tag(stata_ts, "timestamp"))
+        self._write_bytes(self._tag(bio.getvalue(), "header"))
+
+    def _write_map(self) -> None:
+        """
+        Called twice during file write. The first populates the values in
+        the map with 0s.  The second call writes the final map locations when
+        all blocks have been written.
+        """
+        if not self._map:
+            self._map = {
+                "stata_data": 0,
+                "map": self.handles.handle.tell(),
+                "variable_types": 0,
+                "varnames": 0,
+                "sortlist": 0,
+                "formats": 0,
+                "value_label_names": 0,
+                "variable_labels": 0,
+                "characteristics": 0,
+                "data": 0,
+                "strls": 0,
+                "value_labels": 0,
+                "stata_data_close": 0,
+                "end-of-file": 0,
+            }
+        # Move to start of map
+        self.handles.handle.seek(self._map["map"])
+        bio = BytesIO()
+        for val in self._map.values():
+            bio.write(struct.pack(self._byteorder + "Q", val))
+        self._write_bytes(self._tag(bio.getvalue(), "map"))
+
+    def _write_variable_types(self) -> None:
+        self._update_map("variable_types")
+        bio = BytesIO()
+        for typ in self.typlist:
+            bio.write(struct.pack(self._byteorder + "H", typ))
+        self._write_bytes(self._tag(bio.getvalue(), "variable_types"))
+
+    def _write_varnames(self) -> None:
+        self._update_map("varnames")
+        bio = BytesIO()
+        # 118 scales by 4 to accommodate utf-8 data worst case encoding
+        vn_len = 32 if self._dta_version == 117 else 128
+        for name in self.varlist:
+            name = self._null_terminate_str(name)
+            name = _pad_bytes_new(name[:32].encode(self._encoding), vn_len + 1)
+            bio.write(name)
+        self._write_bytes(self._tag(bio.getvalue(), "varnames"))
+
+    def _write_sortlist(self) -> None:
+        self._update_map("sortlist")
+        sort_size = 2 if self._dta_version < 119 else 4
+        self._write_bytes(self._tag(b"\x00" * sort_size * (self.nvar + 1), "sortlist"))
+
+    def _write_formats(self) -> None:
+        self._update_map("formats")
+        bio = BytesIO()
+        fmt_len = 49 if self._dta_version == 117 else 57
+        for fmt in self.fmtlist:
+            bio.write(_pad_bytes_new(fmt.encode(self._encoding), fmt_len))
+        self._write_bytes(self._tag(bio.getvalue(), "formats"))
+
+    def _write_value_label_names(self) -> None:
+        self._update_map("value_label_names")
+        bio = BytesIO()
+        # 118 scales by 4 to accommodate utf-8 data worst case encoding
+        vl_len = 32 if self._dta_version == 117 else 128
+        for i in range(self.nvar):
+            # Use variable name when categorical
+            name = ""  # default name
+            if self._has_value_labels[i]:
+                name = self.varlist[i]
+            name = self._null_terminate_str(name)
+            encoded_name = _pad_bytes_new(name[:32].encode(self._encoding), vl_len + 1)
+            bio.write(encoded_name)
+        self._write_bytes(self._tag(bio.getvalue(), "value_label_names"))
+
+    def _write_variable_labels(self) -> None:
+        # Missing labels are 80 blank characters plus null termination
+        self._update_map("variable_labels")
+        bio = BytesIO()
+        # 118 scales by 4 to accommodate utf-8 data worst case encoding
+        vl_len = 80 if self._dta_version == 117 else 320
+        blank = _pad_bytes_new("", vl_len + 1)
+
+        if self._variable_labels is None:
+            for _ in range(self.nvar):
+                bio.write(blank)
+            self._write_bytes(self._tag(bio.getvalue(), "variable_labels"))
+            return
+
+        for col in self.data:
+            if col in self._variable_labels:
+                label = self._variable_labels[col]
+                if len(label) > 80:
+                    raise ValueError("Variable labels must be 80 characters or fewer")
+                try:
+                    encoded = label.encode(self._encoding)
+                except UnicodeEncodeError as err:
+                    raise ValueError(
+                        "Variable labels must contain only characters that "
+                        f"can be encoded in {self._encoding}"
+                    ) from err
+
+                bio.write(_pad_bytes_new(encoded, vl_len + 1))
+            else:
+                bio.write(blank)
+        self._write_bytes(self._tag(bio.getvalue(), "variable_labels"))
+
+    def _write_characteristics(self) -> None:
+        self._update_map("characteristics")
+        self._write_bytes(self._tag(b"", "characteristics"))
+
+    def _write_data(self, records: np.rec.recarray) -> None:
+        self._update_map("data")
+        self._write_bytes(b"<data>")
+        self._write_bytes(records.tobytes())
+        self._write_bytes(b"</data>")
+
+    def _write_strls(self) -> None:
+        self._update_map("strls")
+        self._write_bytes(self._tag(self._strl_blob, "strls"))
+
+    def _write_expansion_fields(self) -> None:
+        """No-op in dta 117+"""
+
+    def _write_value_labels(self) -> None:
+        self._update_map("value_labels")
+        bio = BytesIO()
+        for vl in self._value_labels:
+            lab = vl.generate_value_label(self._byteorder)
+            lab = self._tag(lab, "lbl")
+            bio.write(lab)
+        self._write_bytes(self._tag(bio.getvalue(), "value_labels"))
+
+    def _write_file_close_tag(self) -> None:
+        self._update_map("stata_data_close")
+        self._write_bytes(bytes("</stata_dta>", "utf-8"))
+        self._update_map("end-of-file")
+
+    def _update_strl_names(self) -> None:
+        """
+        Update column names for conversion to strl if they might have been
+        changed to comply with Stata naming rules
+        """
+        # Update convert_strl if names changed
+        for orig, new in self._converted_names.items():
+            if orig in self._convert_strl:
+                idx = self._convert_strl.index(orig)
+                self._convert_strl[idx] = new
+
+    def _convert_strls(self, data: DataFrame) -> DataFrame:
+        """
+        Convert columns to StrLs if either very large or in the
+        convert_strl variable
+        """
+        convert_cols = [
+            col
+            for i, col in enumerate(data)
+            if self.typlist[i] == 32768 or col in self._convert_strl
+        ]
+
+        if convert_cols:
+            ssw = StataStrLWriter(
+                data, convert_cols, version=self._dta_version, byteorder=self._byteorder
+            )
+            tab, new_data = ssw.generate_table()
+            data = new_data
+            self._strl_blob = ssw.generate_blob(tab)
+        return data
+
+    def _set_formats_and_types(self, dtypes: Series) -> None:
+        self.typlist = []
+        self.fmtlist = []
+        for col, dtype in dtypes.items():
+            force_strl = col in self._convert_strl
+            fmt = _dtype_to_default_stata_fmt(
+                dtype,
+                self.data[col],
+                dta_version=self._dta_version,
+                force_strl=force_strl,
+            )
+            self.fmtlist.append(fmt)
+            self.typlist.append(
+                _dtype_to_stata_type_117(dtype, self.data[col], force_strl)
+            )
+
+
+class StataWriterUTF8(StataWriter117):
+    """
+    Stata binary dta file writing in Stata 15 (118) and 16 (119) formats
+
+    DTA 118 and 119 format files support unicode string data (both fixed
+    and strL) format. Unicode is also supported in value labels, variable
+    labels and the dataset label. Format 119 is automatically used if the
+    file contains more than 32,767 variables.
+
+    Parameters
+    ----------
+    fname : path (string), buffer or path object
+        string, pathlib.Path or
+        object implementing a binary write() functions. If using a buffer
+        then the buffer will not be automatically closed after the file
+        is written.
+    data : DataFrame
+        Input to save
+    convert_dates : dict, default None
+        Dictionary mapping columns containing datetime types to stata internal
+        format to use when writing the dates. Options are 'tc', 'td', 'tm',
+        'tw', 'th', 'tq', 'ty'. Column can be either an integer or a name.
+        Datetime columns that do not have a conversion type specified will be
+        converted to 'tc'. Raises NotImplementedError if a datetime column has
+        timezone information
+    write_index : bool, default True
+        Write the index to Stata dataset.
+    byteorder : str, default None
+        Can be ">", "<", "little", or "big". default is `sys.byteorder`
+    time_stamp : datetime, default None
+        A datetime to use as file creation date.  Default is the current time
+    data_label : str, default None
+        A label for the data set.  Must be 80 characters or smaller.
+    variable_labels : dict, default None
+        Dictionary containing columns as keys and variable labels as values.
+        Each label must be 80 characters or smaller.
+    convert_strl : list, default None
+        List of columns names to convert to Stata StrL format.  Columns with
+        more than 2045 characters are automatically written as StrL.
+        Smaller columns can be converted by including the column name.  Using
+        StrLs can reduce output file size when strings are longer than 8
+        characters, and either frequently repeated or sparse.
+    version : int, default None
+        The dta version to use. By default, uses the size of data to determine
+        the version. 118 is used if data.shape[1] <= 32767, and 119 is used
+        for storing larger DataFrames.
+    {compression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    value_labels : dict of dicts
+        Dictionary containing columns as keys and dictionaries of column value
+        to labels as values. The combined length of all labels for a single
+        variable must be 32,000 characters or smaller.
+
+        .. versionadded:: 1.4.0
+
+    Returns
+    -------
+    StataWriterUTF8
+        The instance has a write_file method, which will write the file to the
+        given `fname`.
+
+    Raises
+    ------
+    NotImplementedError
+        * If datetimes contain timezone information
+    ValueError
+        * Columns listed in convert_dates are neither datetime64[ns]
+          or datetime
+        * Column dtype is not representable in Stata
+        * Column listed in convert_dates is not in DataFrame
+        * Categorical label contains more than 32,000 characters
+
+    Examples
+    --------
+    Using Unicode data and column names
+
+    >>> from pandas.io.stata import StataWriterUTF8
+    >>> data = pd.DataFrame([[1.0, 1, "ᴬ"]], columns=["a", "β", "ĉ"])
+    >>> writer = StataWriterUTF8("./data_file.dta", data)
+    >>> writer.write_file()
+
+    Directly write a zip file
+    >>> compression = {"method": "zip", "archive_name": "data_file.dta"}
+    >>> writer = StataWriterUTF8("./data_file.zip", data, compression=compression)
+    >>> writer.write_file()
+
+    Or with long strings stored in strl format
+
+    >>> data = pd.DataFrame(
+    ...     [["ᴀ relatively long ŝtring"], [""], [""]], columns=["strls"]
+    ... )
+    >>> writer = StataWriterUTF8(
+    ...     "./data_file_with_long_strings.dta", data, convert_strl=["strls"]
+    ... )
+    >>> writer.write_file()
+    """
+
+    _encoding: Literal["utf-8"] = "utf-8"
+
+    def __init__(
+        self,
+        fname: FilePath | WriteBuffer[bytes],
+        data: DataFrame,
+        convert_dates: dict[Hashable, str] | None = None,
+        write_index: bool = True,
+        byteorder: str | None = None,
+        time_stamp: datetime | None = None,
+        data_label: str | None = None,
+        variable_labels: dict[Hashable, str] | None = None,
+        convert_strl: Sequence[Hashable] | None = None,
+        version: int | None = None,
+        compression: CompressionOptions = "infer",
+        storage_options: StorageOptions | None = None,
+        *,
+        value_labels: dict[Hashable, dict[float, str]] | None = None,
+    ) -> None:
+        if version is None:
+            version = 118 if data.shape[1] <= 32767 else 119
+        elif version not in (118, 119):
+            raise ValueError("version must be either 118 or 119.")
+        elif version == 118 and data.shape[1] > 32767:
+            raise ValueError(
+                "You must use version 119 for data sets containing more than"
+                "32,767 variables"
+            )
+
+        super().__init__(
+            fname,
+            data,
+            convert_dates=convert_dates,
+            write_index=write_index,
+            byteorder=byteorder,
+            time_stamp=time_stamp,
+            data_label=data_label,
+            variable_labels=variable_labels,
+            value_labels=value_labels,
+            convert_strl=convert_strl,
+            compression=compression,
+            storage_options=storage_options,
+        )
+        # Override version set in StataWriter117 init
+        self._dta_version = version
+
+    def _validate_variable_name(self, name: str) -> str:
+        """
+        Validate variable names for Stata export.
+
+        Parameters
+        ----------
+        name : str
+            Variable name
+
+        Returns
+        -------
+        str
+            The validated name with invalid characters replaced with
+            underscores.
+
+        Notes
+        -----
+        Stata 118+ support most unicode characters. The only limitation is in
+        the ascii range where the characters supported are a-z, A-Z, 0-9 and _.
+        """
+        # High code points appear to be acceptable
+        for c in name:
+            if (
+                (
+                    ord(c) < 128
+                    and (c < "A" or c > "Z")
+                    and (c < "a" or c > "z")
+                    and (c < "0" or c > "9")
+                    and c != "_"
+                )
+                or 128 <= ord(c) < 192
+                or c in {"×", "÷"}  # noqa: RUF001
+            ):
+                name = name.replace(c, "_")
+
+        return name
diff --git a/pandas/pandas/io/xml.py b/pandas/pandas/io/xml.py
new file mode 100644
index 0000000000000..0fcf27af42fde
--- /dev/null
+++ b/pandas/pandas/io/xml.py
@@ -0,0 +1,1140 @@
+"""
+:mod:``pandas.io.xml`` is a module for reading XML.
+"""
+
+from __future__ import annotations
+
+import io
+from os import PathLike
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas._libs import lib
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    AbstractMethodError,
+    ParserError,
+)
+from pandas.util._decorators import doc
+from pandas.util._validators import check_dtype_backend
+
+from pandas.core.dtypes.common import is_list_like
+
+from pandas.core.shared_docs import _shared_docs
+
+from pandas.io.common import (
+    get_handle,
+    infer_compression,
+    is_fsspec_url,
+    is_url,
+    stringify_path,
+)
+from pandas.io.parsers import TextParser
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Sequence,
+    )
+    from xml.etree.ElementTree import Element
+
+    from lxml import etree
+
+    from pandas._typing import (
+        CompressionOptions,
+        ConvertersArg,
+        DtypeArg,
+        DtypeBackend,
+        FilePath,
+        ParseDatesArg,
+        ReadBuffer,
+        StorageOptions,
+        XMLParsers,
+    )
+
+    from pandas import DataFrame
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    decompression_options=_shared_docs["decompression_options"] % "path_or_buffer",
+)
+class _XMLFrameParser:
+    """
+    Internal subclass to parse XML into DataFrames.
+
+    Parameters
+    ----------
+    path_or_buffer : a valid JSON ``str``, path object or file-like object
+        Any valid string path is acceptable. The string could be a URL. Valid
+        URL schemes include http, ftp, s3, and file.
+
+    xpath : str or regex
+        The ``XPath`` expression to parse required set of nodes for
+        migration to :class:`~pandas.DataFrame`. ``etree`` supports limited ``XPath``.
+
+    namespaces : dict
+        The namespaces defined in XML document (``xmlns:namespace='URI'``)
+        as dicts with key being namespace and value the URI.
+
+    elems_only : bool
+        Parse only the child elements at the specified ``xpath``.
+
+    attrs_only : bool
+        Parse only the attributes at the specified ``xpath``.
+
+    names : list
+        Column names for :class:`~pandas.DataFrame` of parsed XML data.
+
+    dtype : dict
+        Data type for data or columns. E.g. {{'a': np.float64,
+        'b': np.int32, 'c': 'Int64'}}
+
+        .. versionadded:: 1.5.0
+
+    converters : dict, optional
+        Dict of functions for converting values in certain columns. Keys can
+        either be integers or column labels.
+
+        .. versionadded:: 1.5.0
+
+    parse_dates : bool or list of int or names or list of lists or dict
+        Converts either index or select columns to datetimes
+
+        .. versionadded:: 1.5.0
+
+    encoding : str
+        Encoding of xml object or document.
+
+    stylesheet : str or file-like
+        URL, file, file-like object, or a raw string containing XSLT,
+        ``etree`` does not support XSLT but retained for consistency.
+
+    iterparse : dict, optional
+        Dict with row element as key and list of descendant elements
+        and/or attributes as value to be retrieved in iterparsing of
+        XML document.
+
+        .. versionadded:: 1.5.0
+
+    {decompression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    {storage_options}
+
+    See also
+    --------
+    pandas.io.xml._EtreeFrameParser
+    pandas.io.xml._LxmlFrameParser
+
+    Notes
+    -----
+    To subclass this class effectively you must override the following methods:`
+        * :func:`parse_data`
+        * :func:`_parse_nodes`
+        * :func:`_iterparse_nodes`
+        * :func:`_parse_doc`
+        * :func:`_validate_names`
+        * :func:`_validate_path`
+
+
+    See each method's respective documentation for details on their
+    functionality.
+    """
+
+    def __init__(
+        self,
+        path_or_buffer: FilePath | ReadBuffer[bytes] | ReadBuffer[str],
+        xpath: str,
+        namespaces: dict[str, str] | None,
+        elems_only: bool,
+        attrs_only: bool,
+        names: Sequence[str] | None,
+        dtype: DtypeArg | None,
+        converters: ConvertersArg | None,
+        parse_dates: ParseDatesArg | None,
+        encoding: str | None,
+        stylesheet: FilePath | ReadBuffer[bytes] | ReadBuffer[str] | None,
+        iterparse: dict[str, list[str]] | None,
+        compression: CompressionOptions,
+        storage_options: StorageOptions,
+    ) -> None:
+        self.path_or_buffer = path_or_buffer
+        self.xpath = xpath
+        self.namespaces = namespaces
+        self.elems_only = elems_only
+        self.attrs_only = attrs_only
+        self.names = names
+        self.dtype = dtype
+        self.converters = converters
+        self.parse_dates = parse_dates
+        self.encoding = encoding
+        self.stylesheet = stylesheet
+        self.iterparse = iterparse
+        self.compression: CompressionOptions = compression
+        self.storage_options = storage_options
+
+    def parse_data(self) -> list[dict[str, str | None]]:
+        """
+        Parse xml data.
+
+        This method will call the other internal methods to
+        validate ``xpath``, names, parse and return specific nodes.
+        """
+
+        raise AbstractMethodError(self)
+
+    def _parse_nodes(self, elems: list[Any]) -> list[dict[str, str | None]]:
+        """
+        Parse xml nodes.
+
+        This method will parse the children and attributes of elements
+        in ``xpath``, conditionally for only elements, only attributes
+        or both while optionally renaming node names.
+
+        Raises
+        ------
+        ValueError
+            * If only elements and only attributes are specified.
+
+        Notes
+        -----
+        Namespace URIs will be removed from return node values. Also,
+        elements with missing children or attributes compared to siblings
+        will have optional keys filled with None values.
+        """
+
+        dicts: list[dict[str, str | None]]
+
+        if self.elems_only and self.attrs_only:
+            raise ValueError("Either element or attributes can be parsed not both.")
+        if self.elems_only:
+            if self.names:
+                dicts = [
+                    {
+                        **(
+                            {el.tag: el.text}
+                            if el.text and not el.text.isspace()
+                            else {}
+                        ),
+                        **{
+                            nm: ch.text if ch.text else None
+                            for nm, ch in zip(self.names, el.findall("*"))
+                        },
+                    }
+                    for el in elems
+                ]
+            else:
+                dicts = [
+                    {ch.tag: ch.text if ch.text else None for ch in el.findall("*")}
+                    for el in elems
+                ]
+
+        elif self.attrs_only:
+            dicts = [
+                {k: v if v else None for k, v in el.attrib.items()} for el in elems
+            ]
+
+        elif self.names:
+            dicts = [
+                {
+                    **el.attrib,
+                    **({el.tag: el.text} if el.text and not el.text.isspace() else {}),
+                    **{
+                        nm: ch.text if ch.text else None
+                        for nm, ch in zip(self.names, el.findall("*"))
+                    },
+                }
+                for el in elems
+            ]
+
+        else:
+            dicts = [
+                {
+                    **el.attrib,
+                    **({el.tag: el.text} if el.text and not el.text.isspace() else {}),
+                    **{ch.tag: ch.text if ch.text else None for ch in el.findall("*")},
+                }
+                for el in elems
+            ]
+
+        dicts = [
+            {k.split("}")[1] if "}" in k else k: v for k, v in d.items()} for d in dicts
+        ]
+
+        keys = list(dict.fromkeys([k for d in dicts for k in d.keys()]))
+        dicts = [{k: d[k] if k in d.keys() else None for k in keys} for d in dicts]
+
+        if self.names:
+            dicts = [dict(zip(self.names, d.values())) for d in dicts]
+
+        return dicts
+
+    def _iterparse_nodes(self, iterparse: Callable) -> list[dict[str, str | None]]:
+        """
+        Iterparse xml nodes.
+
+        This method will read in local disk, decompressed XML files for elements
+        and underlying descendants using iterparse, a method to iterate through
+        an XML tree without holding entire XML tree in memory.
+
+        Raises
+        ------
+        TypeError
+            * If ``iterparse`` is not a dict or its dict value is not list-like.
+        ParserError
+            * If ``path_or_buffer`` is not a physical file on disk or file-like object.
+            * If no data is returned from selected items in ``iterparse``.
+
+        Notes
+        -----
+        Namespace URIs will be removed from return node values. Also,
+        elements with missing children or attributes in submitted list
+        will have optional keys filled with None values.
+        """
+
+        dicts: list[dict[str, str | None]] = []
+        row: dict[str, str | None] | None = None
+
+        if not isinstance(self.iterparse, dict):
+            raise TypeError(
+                f"{type(self.iterparse).__name__} is not a valid type for iterparse"
+            )
+
+        row_node = next(iter(self.iterparse.keys())) if self.iterparse else ""
+        if not is_list_like(self.iterparse[row_node]):
+            raise TypeError(
+                f"{type(self.iterparse[row_node])} is not a valid type "
+                "for value in iterparse"
+            )
+
+        if (not hasattr(self.path_or_buffer, "read")) and (
+            not isinstance(self.path_or_buffer, (str, PathLike))
+            or is_url(self.path_or_buffer)
+            or is_fsspec_url(self.path_or_buffer)
+            or (
+                isinstance(self.path_or_buffer, str)
+                and self.path_or_buffer.startswith(("<?xml", "<"))
+            )
+            or infer_compression(self.path_or_buffer, "infer") is not None
+        ):
+            raise ParserError(
+                "iterparse is designed for large XML files that are fully extracted on "
+                "local disk and not as compressed files or online sources."
+            )
+
+        iterparse_repeats = len(self.iterparse[row_node]) != len(
+            set(self.iterparse[row_node])
+        )
+
+        for event, elem in iterparse(self.path_or_buffer, events=("start", "end")):
+            curr_elem = elem.tag.split("}")[1] if "}" in elem.tag else elem.tag
+
+            if event == "start":
+                if curr_elem == row_node:
+                    row = {}
+
+            if row is not None:
+                if self.names and iterparse_repeats:
+                    for col, nm in zip(self.iterparse[row_node], self.names):
+                        if curr_elem == col:
+                            elem_val = elem.text if elem.text else None
+                            if elem_val not in row.values() and nm not in row:
+                                row[nm] = elem_val
+
+                        if col in elem.attrib:
+                            if elem.attrib[col] not in row.values() and nm not in row:
+                                row[nm] = elem.attrib[col]
+                else:
+                    for col in self.iterparse[row_node]:
+                        if curr_elem == col:
+                            row[col] = elem.text if elem.text else None
+                        if col in elem.attrib:
+                            row[col] = elem.attrib[col]
+
+            if event == "end":
+                if curr_elem == row_node and row is not None:
+                    dicts.append(row)
+                    row = None
+
+                elem.clear()
+                if hasattr(elem, "getprevious"):
+                    while (
+                        elem.getprevious() is not None and elem.getparent() is not None
+                    ):
+                        del elem.getparent()[0]
+
+        if dicts == []:
+            raise ParserError("No result from selected items in iterparse.")
+
+        keys = list(dict.fromkeys([k for d in dicts for k in d.keys()]))
+        dicts = [{k: d[k] if k in d.keys() else None for k in keys} for d in dicts]
+
+        if self.names:
+            dicts = [dict(zip(self.names, d.values())) for d in dicts]
+
+        return dicts
+
+    def _validate_path(self) -> list[Any]:
+        """
+        Validate ``xpath``.
+
+        This method checks for syntax, evaluation, or empty nodes return.
+
+        Raises
+        ------
+        SyntaxError
+            * If xpah is not supported or issues with namespaces.
+
+        ValueError
+            * If xpah does not return any nodes.
+        """
+
+        raise AbstractMethodError(self)
+
+    def _validate_names(self) -> None:
+        """
+        Validate names.
+
+        This method will check if names is a list-like and aligns
+        with length of parse nodes.
+
+        Raises
+        ------
+        ValueError
+            * If value is not a list and less then length of nodes.
+        """
+        raise AbstractMethodError(self)
+
+    def _parse_doc(
+        self, raw_doc: FilePath | ReadBuffer[bytes] | ReadBuffer[str]
+    ) -> Element | etree._Element:
+        """
+        Build tree from path_or_buffer.
+
+        This method will parse XML object into tree
+        either from string/bytes or file location.
+        """
+        raise AbstractMethodError(self)
+
+
+class _EtreeFrameParser(_XMLFrameParser):
+    """
+    Internal class to parse XML into DataFrames with the Python
+    standard library XML module: `xml.etree.ElementTree`.
+    """
+
+    def parse_data(self) -> list[dict[str, str | None]]:
+        from xml.etree.ElementTree import iterparse
+
+        if self.stylesheet is not None:
+            raise ValueError(
+                "To use stylesheet, you need lxml installed and selected as parser."
+            )
+
+        if self.iterparse is None:
+            self.xml_doc = self._parse_doc(self.path_or_buffer)
+            elems = self._validate_path()
+
+        self._validate_names()
+
+        xml_dicts: list[dict[str, str | None]] = (
+            self._parse_nodes(elems)
+            if self.iterparse is None
+            else self._iterparse_nodes(iterparse)
+        )
+
+        return xml_dicts
+
+    def _validate_path(self) -> list[Any]:
+        """
+        Notes
+        -----
+        ``etree`` supports limited ``XPath``. If user attempts a more complex
+        expression syntax error will raise.
+        """
+
+        msg = (
+            "xpath does not return any nodes or attributes. "
+            "Be sure to specify in `xpath` the parent nodes of "
+            "children and attributes to parse. "
+            "If document uses namespaces denoted with "
+            "xmlns, be sure to define namespaces and "
+            "use them in xpath."
+        )
+        try:
+            elems = self.xml_doc.findall(self.xpath, namespaces=self.namespaces)
+            children = [ch for el in elems for ch in el.findall("*")]
+            attrs = {k: v for el in elems for k, v in el.attrib.items()}
+
+            if elems is None:
+                raise ValueError(msg)
+
+            if elems is not None:
+                if self.elems_only and children == []:
+                    raise ValueError(msg)
+                if self.attrs_only and attrs == {}:
+                    raise ValueError(msg)
+                if children == [] and attrs == {}:
+                    raise ValueError(msg)
+
+        except (KeyError, SyntaxError) as err:
+            raise SyntaxError(
+                "You have used an incorrect or unsupported XPath "
+                "expression for etree library or you used an "
+                "undeclared namespace prefix."
+            ) from err
+
+        return elems
+
+    def _validate_names(self) -> None:
+        children: list[Any]
+
+        if self.names:
+            if self.iterparse:
+                children = self.iterparse[next(iter(self.iterparse))]
+            else:
+                parent = self.xml_doc.find(self.xpath, namespaces=self.namespaces)
+                children = parent.findall("*") if parent is not None else []
+
+            if is_list_like(self.names):
+                if len(self.names) < len(children):
+                    raise ValueError(
+                        "names does not match length of child elements in xpath."
+                    )
+            else:
+                raise TypeError(
+                    f"{type(self.names).__name__} is not a valid type for names"
+                )
+
+    def _parse_doc(
+        self, raw_doc: FilePath | ReadBuffer[bytes] | ReadBuffer[str]
+    ) -> Element:
+        from xml.etree.ElementTree import (
+            XMLParser,
+            parse,
+        )
+
+        handle_data = get_data_from_filepath(
+            filepath_or_buffer=raw_doc,
+            encoding=self.encoding,
+            compression=self.compression,
+            storage_options=self.storage_options,
+        )
+
+        with handle_data as xml_data:
+            curr_parser = XMLParser(encoding=self.encoding)
+            document = parse(xml_data, parser=curr_parser)
+
+        return document.getroot()
+
+
+class _LxmlFrameParser(_XMLFrameParser):
+    """
+    Internal class to parse XML into :class:`~pandas.DataFrame` with third-party
+    full-featured XML library, ``lxml``, that supports
+    ``XPath`` 1.0 and XSLT 1.0.
+    """
+
+    def parse_data(self) -> list[dict[str, str | None]]:
+        """
+        Parse xml data.
+
+        This method will call the other internal methods to
+        validate ``xpath``, names, optionally parse and run XSLT,
+        and parse original or transformed XML and return specific nodes.
+        """
+        from lxml.etree import iterparse
+
+        if self.iterparse is None:
+            self.xml_doc = self._parse_doc(self.path_or_buffer)
+
+            if self.stylesheet:
+                self.xsl_doc = self._parse_doc(self.stylesheet)
+                self.xml_doc = self._transform_doc()
+
+            elems = self._validate_path()
+
+        self._validate_names()
+
+        xml_dicts: list[dict[str, str | None]] = (
+            self._parse_nodes(elems)
+            if self.iterparse is None
+            else self._iterparse_nodes(iterparse)
+        )
+
+        return xml_dicts
+
+    def _validate_path(self) -> list[Any]:
+        msg = (
+            "xpath does not return any nodes or attributes. "
+            "Be sure to specify in `xpath` the parent nodes of "
+            "children and attributes to parse. "
+            "If document uses namespaces denoted with "
+            "xmlns, be sure to define namespaces and "
+            "use them in xpath."
+        )
+
+        elems = self.xml_doc.xpath(self.xpath, namespaces=self.namespaces)
+        children = [ch for el in elems for ch in el.xpath("*")]
+        attrs = {k: v for el in elems for k, v in el.attrib.items()}
+
+        if elems == []:
+            raise ValueError(msg)
+
+        if elems != []:
+            if self.elems_only and children == []:
+                raise ValueError(msg)
+            if self.attrs_only and attrs == {}:
+                raise ValueError(msg)
+            if children == [] and attrs == {}:
+                raise ValueError(msg)
+
+        return elems
+
+    def _validate_names(self) -> None:
+        children: list[Any]
+
+        if self.names:
+            if self.iterparse:
+                children = self.iterparse[next(iter(self.iterparse))]
+            else:
+                children = self.xml_doc.xpath(
+                    self.xpath + "[1]/*", namespaces=self.namespaces
+                )
+
+            if is_list_like(self.names):
+                if len(self.names) < len(children):
+                    raise ValueError(
+                        "names does not match length of child elements in xpath."
+                    )
+            else:
+                raise TypeError(
+                    f"{type(self.names).__name__} is not a valid type for names"
+                )
+
+    def _parse_doc(
+        self, raw_doc: FilePath | ReadBuffer[bytes] | ReadBuffer[str]
+    ) -> etree._Element:
+        from lxml.etree import (
+            XMLParser,
+            fromstring,
+            parse,
+        )
+
+        handle_data = get_data_from_filepath(
+            filepath_or_buffer=raw_doc,
+            encoding=self.encoding,
+            compression=self.compression,
+            storage_options=self.storage_options,
+        )
+
+        with handle_data as xml_data:
+            curr_parser = XMLParser(encoding=self.encoding)
+
+            if isinstance(xml_data, io.StringIO):
+                if self.encoding is None:
+                    raise TypeError(
+                        "Can not pass encoding None when input is StringIO."
+                    )
+
+                document = fromstring(
+                    xml_data.getvalue().encode(self.encoding), parser=curr_parser
+                )
+            else:
+                document = parse(xml_data, parser=curr_parser)
+
+        return document
+
+    def _transform_doc(self) -> etree._XSLTResultTree:
+        """
+        Transform original tree using stylesheet.
+
+        This method will transform original xml using XSLT script into
+        am ideally flatter xml document for easier parsing and migration
+        to Data Frame.
+        """
+        from lxml.etree import XSLT
+
+        transformer = XSLT(self.xsl_doc)
+        new_doc = transformer(self.xml_doc)
+
+        return new_doc
+
+
+def get_data_from_filepath(
+    filepath_or_buffer: FilePath | bytes | ReadBuffer[bytes] | ReadBuffer[str],
+    encoding: str | None,
+    compression: CompressionOptions,
+    storage_options: StorageOptions,
+):
+    """
+    Extract raw XML data.
+
+    The method accepts two input types:
+        1. filepath (string-like)
+        2. file-like object (e.g. open file object, StringIO)
+    """
+    filepath_or_buffer = stringify_path(filepath_or_buffer)  # type: ignore[arg-type]
+    with get_handle(  # pyright: ignore[reportCallIssue]
+        filepath_or_buffer,  # pyright: ignore[reportArgumentType]
+        "r",
+        encoding=encoding,
+        compression=compression,
+        storage_options=storage_options,
+    ) as handle_obj:
+        return (
+            preprocess_data(handle_obj.handle.read())
+            if hasattr(handle_obj.handle, "read")
+            else handle_obj.handle
+        )
+
+
+def preprocess_data(data) -> io.StringIO | io.BytesIO:
+    """
+    Convert extracted raw data.
+
+    This method will return underlying data of extracted XML content.
+    The data either has a `read` attribute (e.g. a file object or a
+    StringIO/BytesIO) or is a string or bytes that is an XML document.
+    """
+
+    if isinstance(data, str):
+        data = io.StringIO(data)
+
+    elif isinstance(data, bytes):
+        data = io.BytesIO(data)
+
+    return data
+
+
+def _data_to_frame(data, **kwargs) -> DataFrame:
+    """
+    Convert parsed data to Data Frame.
+
+    This method will bind xml dictionary data of keys and values
+    into named columns of Data Frame using the built-in TextParser
+    class that build Data Frame and infers specific dtypes.
+    """
+
+    tags = next(iter(data))
+    nodes = [list(d.values()) for d in data]
+
+    try:
+        with TextParser(nodes, names=tags, **kwargs) as tp:
+            return tp.read()
+    except ParserError as err:
+        raise ParserError(
+            "XML document may be too complex for import. "
+            "Try to flatten document and use distinct "
+            "element and attribute names."
+        ) from err
+
+
+def _parse(
+    path_or_buffer: FilePath | ReadBuffer[bytes] | ReadBuffer[str],
+    xpath: str,
+    namespaces: dict[str, str] | None,
+    elems_only: bool,
+    attrs_only: bool,
+    names: Sequence[str] | None,
+    dtype: DtypeArg | None,
+    converters: ConvertersArg | None,
+    parse_dates: ParseDatesArg | None,
+    encoding: str | None,
+    parser: XMLParsers,
+    stylesheet: FilePath | ReadBuffer[bytes] | ReadBuffer[str] | None,
+    iterparse: dict[str, list[str]] | None,
+    compression: CompressionOptions,
+    storage_options: StorageOptions,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+    **kwargs,
+) -> DataFrame:
+    """
+    Call internal parsers.
+
+    This method will conditionally call internal parsers:
+    LxmlFrameParser and/or EtreeParser.
+
+    Raises
+    ------
+    ImportError
+        * If lxml is not installed if selected as parser.
+
+    ValueError
+        * If parser is not lxml or etree.
+    """
+
+    p: _EtreeFrameParser | _LxmlFrameParser
+
+    if parser == "lxml":
+        lxml = import_optional_dependency("lxml.etree", errors="ignore")
+
+        if lxml is not None:
+            p = _LxmlFrameParser(
+                path_or_buffer,
+                xpath,
+                namespaces,
+                elems_only,
+                attrs_only,
+                names,
+                dtype,
+                converters,
+                parse_dates,
+                encoding,
+                stylesheet,
+                iterparse,
+                compression,
+                storage_options,
+            )
+        else:
+            raise ImportError("lxml not found, please install or use the etree parser.")
+
+    elif parser == "etree":
+        p = _EtreeFrameParser(
+            path_or_buffer,
+            xpath,
+            namespaces,
+            elems_only,
+            attrs_only,
+            names,
+            dtype,
+            converters,
+            parse_dates,
+            encoding,
+            stylesheet,
+            iterparse,
+            compression,
+            storage_options,
+        )
+    else:
+        raise ValueError("Values for parser can only be lxml or etree.")
+
+    data_dicts = p.parse_data()
+
+    return _data_to_frame(
+        data=data_dicts,
+        dtype=dtype,
+        converters=converters,
+        parse_dates=parse_dates,
+        dtype_backend=dtype_backend,
+        **kwargs,
+    )
+
+
+@doc(
+    storage_options=_shared_docs["storage_options"],
+    decompression_options=_shared_docs["decompression_options"] % "path_or_buffer",
+)
+def read_xml(
+    path_or_buffer: FilePath | ReadBuffer[bytes] | ReadBuffer[str],
+    *,
+    xpath: str = "./*",
+    namespaces: dict[str, str] | None = None,
+    elems_only: bool = False,
+    attrs_only: bool = False,
+    names: Sequence[str] | None = None,
+    dtype: DtypeArg | None = None,
+    converters: ConvertersArg | None = None,
+    parse_dates: ParseDatesArg | None = None,
+    # encoding can not be None for lxml and StringIO input
+    encoding: str | None = "utf-8",
+    parser: XMLParsers = "lxml",
+    stylesheet: FilePath | ReadBuffer[bytes] | ReadBuffer[str] | None = None,
+    iterparse: dict[str, list[str]] | None = None,
+    compression: CompressionOptions = "infer",
+    storage_options: StorageOptions | None = None,
+    dtype_backend: DtypeBackend | lib.NoDefault = lib.no_default,
+) -> DataFrame:
+    r"""
+    Read XML document into a :class:`~pandas.DataFrame` object.
+
+    .. versionadded:: 1.3.0
+
+    Parameters
+    ----------
+    path_or_buffer : str, path object, or file-like object
+        String, path object (implementing ``os.PathLike[str]``), or file-like
+        object implementing a ``read()`` function. The string can be a path.
+        The string can further be a URL. Valid URL schemes
+        include http, ftp, s3, and file.
+
+        .. deprecated:: 2.1.0
+            Passing xml literal strings is deprecated.
+            Wrap literal xml input in ``io.StringIO`` or ``io.BytesIO`` instead.
+
+    xpath : str, optional, default './\*'
+        The ``XPath`` to parse required set of nodes for migration to
+        :class:`~pandas.DataFrame`.``XPath`` should return a collection of elements
+        and not a single element. Note: The ``etree`` parser supports limited ``XPath``
+        expressions. For more complex ``XPath``, use ``lxml`` which requires
+        installation.
+
+    namespaces : dict, optional
+        The namespaces defined in XML document as dicts with key being
+        namespace prefix and value the URI. There is no need to include all
+        namespaces in XML, only the ones used in ``xpath`` expression.
+        Note: if XML document uses default namespace denoted as
+        `xmlns='<URI>'` without a prefix, you must assign any temporary
+        namespace prefix such as 'doc' to the URI in order to parse
+        underlying nodes and/or attributes.
+
+    elems_only : bool, optional, default False
+        Parse only the child elements at the specified ``xpath``. By default,
+        all child elements and non-empty text nodes are returned.
+
+    attrs_only :  bool, optional, default False
+        Parse only the attributes at the specified ``xpath``.
+        By default, all attributes are returned.
+
+    names :  list-like, optional
+        Column names for DataFrame of parsed XML data. Use this parameter to
+        rename original element names and distinguish same named elements and
+        attributes.
+
+    dtype : Type name or dict of column -> type, optional
+        Data type for data or columns. E.g. {{'a': np.float64, 'b': np.int32,
+        'c': 'Int64'}}
+        Use `str` or `object` together with suitable `na_values` settings
+        to preserve and not interpret dtype.
+        If converters are specified, they will be applied INSTEAD
+        of dtype conversion.
+
+        .. versionadded:: 1.5.0
+
+    converters : dict, optional
+        Dict of functions for converting values in certain columns. Keys can either
+        be integers or column labels.
+
+        .. versionadded:: 1.5.0
+
+    parse_dates : bool or list of int or names or list of lists or dict, default False
+        Identifiers to parse index or columns to datetime. The behavior is as follows:
+
+        * boolean. If True -> try parsing the index.
+        * list of int or names. e.g. If [1, 2, 3] -> try parsing columns 1, 2, 3
+          each as a separate date column.
+        * list of lists. e.g.  If [[1, 3]] -> combine columns 1 and 3 and parse as
+          a single date column.
+        * dict, e.g. {{'foo' : [1, 3]}} -> parse columns 1, 3 as date and call
+          result 'foo'
+
+        .. versionadded:: 1.5.0
+
+    encoding : str, optional, default 'utf-8'
+        Encoding of XML document.
+
+    parser : {{'lxml','etree'}}, default 'lxml'
+        Parser module to use for retrieval of data. Only 'lxml' and
+        'etree' are supported. With 'lxml' more complex ``XPath`` searches
+        and ability to use XSLT stylesheet are supported.
+
+    stylesheet : str, path object or file-like object
+        A URL, file-like object, or a string path containing an XSLT script.
+        This stylesheet should flatten complex, deeply nested XML documents
+        for easier parsing. To use this feature you must have ``lxml`` module
+        installed and specify 'lxml' as ``parser``. The ``xpath`` must
+        reference nodes of transformed XML document generated after XSLT
+        transformation and not the original XML document. Only XSLT 1.0
+        scripts and not later versions is currently supported.
+
+    iterparse : dict, optional
+        The nodes or attributes to retrieve in iterparsing of XML document
+        as a dict with key being the name of repeating element and value being
+        list of elements or attribute names that are descendants of the repeated
+        element. Note: If this option is used, it will replace ``xpath`` parsing
+        and unlike ``xpath``, descendants do not need to relate to each other but can
+        exist any where in document under the repeating element. This memory-
+        efficient method should be used for very large XML files (500MB, 1GB, or 5GB+).
+        For example, ``{{"row_element": ["child_elem", "attr", "grandchild_elem"]}}``.
+
+        .. versionadded:: 1.5.0
+
+    {decompression_options}
+
+        .. versionchanged:: 1.4.0 Zstandard support.
+
+    {storage_options}
+
+    dtype_backend : {{'numpy_nullable', 'pyarrow'}}
+        Back-end data type applied to the resultant :class:`DataFrame`
+        (still experimental). If not specified, the default behavior
+        is to not use nullable data types. If specified, the behavior
+        is as follows:
+
+        * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame`
+        * ``"pyarrow"``: returns pyarrow-backed nullable
+          :class:`ArrowDtype` :class:`DataFrame`
+
+        .. versionadded:: 2.0
+
+    Returns
+    -------
+    df
+        A DataFrame.
+
+    See Also
+    --------
+    read_json : Convert a JSON string to pandas object.
+    read_html : Read HTML tables into a list of DataFrame objects.
+
+    Notes
+    -----
+    This method is best designed to import shallow XML documents in
+    following format which is the ideal fit for the two-dimensions of a
+    ``DataFrame`` (row by column). ::
+
+            <root>
+                <row>
+                  <column1>data</column1>
+                  <column2>data</column2>
+                  <column3>data</column3>
+                  ...
+               </row>
+               <row>
+                  ...
+               </row>
+               ...
+            </root>
+
+    As a file format, XML documents can be designed any way including
+    layout of elements and attributes as long as it conforms to W3C
+    specifications. Therefore, this method is a convenience handler for
+    a specific flatter design and not all possible XML structures.
+
+    However, for more complex XML documents, ``stylesheet`` allows you to
+    temporarily redesign original document with XSLT (a special purpose
+    language) for a flatter version for migration to a DataFrame.
+
+    This function will *always* return a single :class:`DataFrame` or raise
+    exceptions due to issues with XML document, ``xpath``, or other
+    parameters.
+
+    See the :ref:`read_xml documentation in the IO section of the docs
+    <io.read_xml>` for more information in using this method to parse XML
+    files to DataFrames.
+
+    Examples
+    --------
+    >>> from io import StringIO
+    >>> xml = '''<?xml version='1.0' encoding='utf-8'?>
+    ... <data xmlns="http://example.com">
+    ...  <row>
+    ...    <shape>square</shape>
+    ...    <degrees>360</degrees>
+    ...    <sides>4.0</sides>
+    ...  </row>
+    ...  <row>
+    ...    <shape>circle</shape>
+    ...    <degrees>360</degrees>
+    ...    <sides/>
+    ...  </row>
+    ...  <row>
+    ...    <shape>triangle</shape>
+    ...    <degrees>180</degrees>
+    ...    <sides>3.0</sides>
+    ...  </row>
+    ... </data>'''
+
+    >>> df = pd.read_xml(StringIO(xml))
+    >>> df
+          shape  degrees  sides
+    0    square      360    4.0
+    1    circle      360    NaN
+    2  triangle      180    3.0
+
+    >>> xml = '''<?xml version='1.0' encoding='utf-8'?>
+    ... <data>
+    ...   <row shape="square" degrees="360" sides="4.0"/>
+    ...   <row shape="circle" degrees="360"/>
+    ...   <row shape="triangle" degrees="180" sides="3.0"/>
+    ... </data>'''
+
+    >>> df = pd.read_xml(StringIO(xml), xpath=".//row")
+    >>> df
+          shape  degrees  sides
+    0    square      360    4.0
+    1    circle      360    NaN
+    2  triangle      180    3.0
+
+    >>> xml = '''<?xml version='1.0' encoding='utf-8'?>
+    ... <doc:data xmlns:doc="https://example.com">
+    ...   <doc:row>
+    ...     <doc:shape>square</doc:shape>
+    ...     <doc:degrees>360</doc:degrees>
+    ...     <doc:sides>4.0</doc:sides>
+    ...   </doc:row>
+    ...   <doc:row>
+    ...     <doc:shape>circle</doc:shape>
+    ...     <doc:degrees>360</doc:degrees>
+    ...     <doc:sides/>
+    ...   </doc:row>
+    ...   <doc:row>
+    ...     <doc:shape>triangle</doc:shape>
+    ...     <doc:degrees>180</doc:degrees>
+    ...     <doc:sides>3.0</doc:sides>
+    ...   </doc:row>
+    ... </doc:data>'''
+
+    >>> df = pd.read_xml(
+    ...     StringIO(xml),
+    ...     xpath="//doc:row",
+    ...     namespaces={{"doc": "https://example.com"}},
+    ... )
+    >>> df
+          shape  degrees  sides
+    0    square      360    4.0
+    1    circle      360    NaN
+    2  triangle      180    3.0
+
+    >>> xml_data = '''
+    ...         <data>
+    ...            <row>
+    ...               <index>0</index>
+    ...               <a>1</a>
+    ...               <b>2.5</b>
+    ...               <c>True</c>
+    ...               <d>a</d>
+    ...               <e>2019-12-31 00:00:00</e>
+    ...            </row>
+    ...            <row>
+    ...               <index>1</index>
+    ...               <b>4.5</b>
+    ...               <c>False</c>
+    ...               <d>b</d>
+    ...               <e>2019-12-31 00:00:00</e>
+    ...            </row>
+    ...         </data>
+    ...         '''
+
+    >>> df = pd.read_xml(
+    ...     StringIO(xml_data), dtype_backend="numpy_nullable", parse_dates=["e"]
+    ... )
+    >>> df
+       index     a    b      c  d          e
+    0      0     1  2.5   True  a 2019-12-31
+    1      1  <NA>  4.5  False  b 2019-12-31
+    """
+    check_dtype_backend(dtype_backend)
+
+    return _parse(
+        path_or_buffer=path_or_buffer,
+        xpath=xpath,
+        namespaces=namespaces,
+        elems_only=elems_only,
+        attrs_only=attrs_only,
+        names=names,
+        dtype=dtype,
+        converters=converters,
+        parse_dates=parse_dates,
+        encoding=encoding,
+        parser=parser,
+        stylesheet=stylesheet,
+        iterparse=iterparse,
+        compression=compression,
+        storage_options=storage_options,
+        dtype_backend=dtype_backend,
+    )
diff --git a/pandas/pandas/meson.build b/pandas/pandas/meson.build
new file mode 100644
index 0000000000000..840ac257bba09
--- /dev/null
+++ b/pandas/pandas/meson.build
@@ -0,0 +1,53 @@
+incdir_numpy = run_command(
+    py,
+    [
+        '-c',
+        '''
+import os
+import numpy as np
+try:
+    # Check if include directory is inside the pandas dir
+    # e.g. a venv created inside the pandas dir
+    # If so, convert it to a relative path
+    incdir = os.path.relpath(np.get_include())
+except Exception:
+    incdir = np.get_include()
+print(incdir)
+     ''',
+    ],
+    check: true,
+).stdout().strip()
+
+inc_np = include_directories(incdir_numpy)
+inc_pd = include_directories('_libs/include')
+
+fs.copyfile('__init__.py')
+
+subdir('_libs')
+
+subdirs_list = [
+    '_config',
+    '_testing',
+    'api',
+    'arrays',
+    'compat',
+    'core',
+    'errors',
+    'io',
+    'plotting',
+    'tests',
+    'tseries',
+    'util',
+]
+foreach subdir : subdirs_list
+    install_subdir(subdir, install_dir: py.get_install_dir() / 'pandas')
+endforeach
+
+top_level_py_list = [
+    '__init__.py',
+    '_typing.py',
+    '_version.py',
+    'conftest.py',
+    'testing.py',
+]
+py.install_sources(top_level_py_list, subdir: 'pandas')
diff --git a/pandas/pandas/plotting/__init__.py b/pandas/pandas/plotting/__init__.py
new file mode 100644
index 0000000000000..837bfaf82ca27
--- /dev/null
+++ b/pandas/pandas/plotting/__init__.py
@@ -0,0 +1,99 @@
+"""
+Plotting public API.
+
+Authors of third-party plotting backends should implement a module with a
+public ``plot(data, kind, **kwargs)``. The parameter `data` will contain
+the data structure and can be a `Series` or a `DataFrame`. For example,
+for ``df.plot()`` the parameter `data` will contain the DataFrame `df`.
+In some cases, the data structure is transformed before being sent to
+the backend (see PlotAccessor.__call__ in pandas/plotting/_core.py for
+the exact transformations).
+
+The parameter `kind` will be one of:
+
+- line
+- bar
+- barh
+- box
+- hist
+- kde
+- area
+- pie
+- scatter
+- hexbin
+
+See the pandas API reference for documentation on each kind of plot.
+
+Any other keyword argument is currently assumed to be backend specific,
+but some parameters may be unified and added to the signature in the
+future (e.g. `title` which should be useful for any backend).
+
+Currently, all the Matplotlib functions in pandas are accessed through
+the selected backend. For example, `pandas.plotting.boxplot` (equivalent
+to `DataFrame.boxplot`) is also accessed in the selected backend. This
+is expected to change, and the exact API is under discussion. But with
+the current version, backends are expected to implement the next functions:
+
+- plot (describe above, used for `Series.plot` and `DataFrame.plot`)
+- hist_series and hist_frame (for `Series.hist` and `DataFrame.hist`)
+- boxplot (`pandas.plotting.boxplot(df)` equivalent to `DataFrame.boxplot`)
+- boxplot_frame and boxplot_frame_groupby
+- register and deregister (register converters for the tick formats)
+- Plots not called as `Series` and `DataFrame` methods:
+  - table
+  - andrews_curves
+  - autocorrelation_plot
+  - bootstrap_plot
+  - lag_plot
+  - parallel_coordinates
+  - radviz
+  - scatter_matrix
+
+Use the code in pandas/plotting/_matplotib.py and
+https://github.com/pyviz/hvplot as a reference on how to write a backend.
+
+For the discussion about the API see
+https://github.com/pandas-dev/pandas/issues/26747.
+"""
+
+from pandas.plotting._core import (
+    PlotAccessor,
+    boxplot,
+    boxplot_frame,
+    boxplot_frame_groupby,
+    hist_frame,
+    hist_series,
+)
+from pandas.plotting._misc import (
+    andrews_curves,
+    autocorrelation_plot,
+    bootstrap_plot,
+    deregister as deregister_matplotlib_converters,
+    lag_plot,
+    parallel_coordinates,
+    plot_params,
+    radviz,
+    register as register_matplotlib_converters,
+    scatter_matrix,
+    table,
+)
+
+__all__ = [
+    "PlotAccessor",
+    "andrews_curves",
+    "autocorrelation_plot",
+    "bootstrap_plot",
+    "boxplot",
+    "boxplot_frame",
+    "boxplot_frame_groupby",
+    "deregister_matplotlib_converters",
+    "hist_frame",
+    "hist_series",
+    "lag_plot",
+    "parallel_coordinates",
+    "plot_params",
+    "radviz",
+    "register_matplotlib_converters",
+    "scatter_matrix",
+    "table",
+]
diff --git a/pandas/pandas/plotting/_core.py b/pandas/pandas/plotting/_core.py
new file mode 100644
index 0000000000000..9670b5439c87e
--- /dev/null
+++ b/pandas/pandas/plotting/_core.py
@@ -0,0 +1,2027 @@
+from __future__ import annotations
+
+import importlib
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+)
+
+from pandas._config import get_option
+
+from pandas.util._decorators import (
+    Appender,
+    Substitution,
+)
+
+from pandas.core.dtypes.common import (
+    is_integer,
+    is_list_like,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCSeries,
+)
+
+from pandas.core.base import PandasObject
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Hashable,
+        Sequence,
+    )
+    import types
+
+    from matplotlib.axes import Axes
+    import numpy as np
+
+    from pandas._typing import IndexLabel
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+    from pandas.core.groupby.generic import DataFrameGroupBy
+
+
+def holds_integer(column: Index) -> bool:
+    return column.inferred_type in {"integer", "mixed-integer"}
+
+
+def hist_series(
+    self: Series,
+    by=None,
+    ax=None,
+    grid: bool = True,
+    xlabelsize: int | None = None,
+    xrot: float | None = None,
+    ylabelsize: int | None = None,
+    yrot: float | None = None,
+    figsize: tuple[int, int] | None = None,
+    bins: int | Sequence[int] = 10,
+    backend: str | None = None,
+    legend: bool = False,
+    **kwargs,
+):
+    """
+    Draw histogram of the input series using matplotlib.
+
+    Parameters
+    ----------
+    by : object, optional
+        If passed, then used to form histograms for separate groups.
+    ax : matplotlib axis object
+        If not passed, uses gca().
+    grid : bool, default True
+        Whether to show axis grid lines.
+    xlabelsize : int, default None
+        If specified changes the x-axis label size.
+    xrot : float, default None
+        Rotation of x axis labels.
+    ylabelsize : int, default None
+        If specified changes the y-axis label size.
+    yrot : float, default None
+        Rotation of y axis labels.
+    figsize : tuple, default None
+        Figure size in inches by default.
+    bins : int or sequence, default 10
+        Number of histogram bins to be used. If an integer is given, bins + 1
+        bin edges are calculated and returned. If bins is a sequence, gives
+        bin edges, including left edge of first bin and right edge of last
+        bin. In this case, bins is returned unmodified.
+    backend : str, default None
+        Backend to use instead of the backend specified in the option
+        ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+        specify the ``plotting.backend`` for the whole session, set
+        ``pd.options.plotting.backend``.
+    legend : bool, default False
+        Whether to show the legend.
+
+    **kwargs
+        To be passed to the actual plotting function.
+
+    Returns
+    -------
+    matplotlib.axes.Axes
+        A histogram plot.
+
+    See Also
+    --------
+    matplotlib.axes.Axes.hist : Plot a histogram using matplotlib.
+
+    Examples
+    --------
+    For Series:
+
+    .. plot::
+        :context: close-figs
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([1, 2, 2, 4, 6, 6], index=lst)
+        >>> hist = ser.hist()
+
+    For Groupby:
+
+    .. plot::
+        :context: close-figs
+
+        >>> lst = ["a", "a", "a", "b", "b", "b"]
+        >>> ser = pd.Series([1, 2, 2, 4, 6, 6], index=lst)
+        >>> hist = ser.groupby(level=0).hist()
+    """
+    plot_backend = _get_plot_backend(backend)
+    return plot_backend.hist_series(
+        self,
+        by=by,
+        ax=ax,
+        grid=grid,
+        xlabelsize=xlabelsize,
+        xrot=xrot,
+        ylabelsize=ylabelsize,
+        yrot=yrot,
+        figsize=figsize,
+        bins=bins,
+        legend=legend,
+        **kwargs,
+    )
+
+
+def hist_frame(
+    data: DataFrame,
+    column: IndexLabel | None = None,
+    by=None,
+    grid: bool = True,
+    xlabelsize: int | None = None,
+    xrot: float | None = None,
+    ylabelsize: int | None = None,
+    yrot: float | None = None,
+    ax=None,
+    sharex: bool = False,
+    sharey: bool = False,
+    figsize: tuple[int, int] | None = None,
+    layout: tuple[int, int] | None = None,
+    bins: int | Sequence[int] = 10,
+    backend: str | None = None,
+    legend: bool = False,
+    **kwargs,
+):
+    """
+    Make a histogram of the DataFrame's columns.
+
+    A `histogram`_ is a representation of the distribution of data.
+    This function calls :meth:`matplotlib.pyplot.hist`, on each series in
+    the DataFrame, resulting in one histogram per column.
+
+    .. _histogram: https://en.wikipedia.org/wiki/Histogram
+
+    Parameters
+    ----------
+    data : DataFrame
+        The pandas object holding the data.
+    column : str or sequence, optional
+        If passed, will be used to limit data to a subset of columns.
+    by : object, optional
+        If passed, then used to form histograms for separate groups.
+    grid : bool, default True
+        Whether to show axis grid lines.
+    xlabelsize : int, default None
+        If specified changes the x-axis label size.
+    xrot : float, default None
+        Rotation of x axis labels. For example, a value of 90 displays the
+        x labels rotated 90 degrees clockwise.
+    ylabelsize : int, default None
+        If specified changes the y-axis label size.
+    yrot : float, default None
+        Rotation of y axis labels. For example, a value of 90 displays the
+        y labels rotated 90 degrees clockwise.
+    ax : Matplotlib axes object, default None
+        The axes to plot the histogram on.
+    sharex : bool, default True if ax is None else False
+        In case subplots=True, share x axis and set some x axis labels to
+        invisible; defaults to True if ax is None otherwise False if an ax
+        is passed in.
+        Note that passing in both an ax and sharex=True will alter all x axis
+        labels for all subplots in a figure.
+    sharey : bool, default False
+        In case subplots=True, share y axis and set some y axis labels to
+        invisible.
+    figsize : tuple, optional
+        The size in inches of the figure to create. Uses the value in
+        `matplotlib.rcParams` by default.
+    layout : tuple, optional
+        Tuple of (rows, columns) for the layout of the histograms.
+    bins : int or sequence, default 10
+        Number of histogram bins to be used. If an integer is given, bins + 1
+        bin edges are calculated and returned. If bins is a sequence, gives
+        bin edges, including left edge of first bin and right edge of last
+        bin. In this case, bins is returned unmodified.
+
+    backend : str, default None
+        Backend to use instead of the backend specified in the option
+        ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+        specify the ``plotting.backend`` for the whole session, set
+        ``pd.options.plotting.backend``.
+
+    legend : bool, default False
+        Whether to show the legend.
+
+    **kwargs
+        All other plotting keyword arguments to be passed to
+        :meth:`matplotlib.pyplot.hist`.
+
+    Returns
+    -------
+    matplotlib.Axes or numpy.ndarray of them
+        Returns a AxesSubplot object a numpy array of AxesSubplot objects.
+
+    See Also
+    --------
+    matplotlib.pyplot.hist : Plot a histogram using matplotlib.
+
+    Examples
+    --------
+    This example draws a histogram based on the length and width of
+    some animals, displayed in three bins
+
+    .. plot::
+        :context: close-figs
+
+        >>> data = {
+        ...     "length": [1.5, 0.5, 1.2, 0.9, 3],
+        ...     "width": [0.7, 0.2, 0.15, 0.2, 1.1],
+        ... }
+        >>> index = ["pig", "rabbit", "duck", "chicken", "horse"]
+        >>> df = pd.DataFrame(data, index=index)
+        >>> hist = df.hist(bins=3)
+    """
+    plot_backend = _get_plot_backend(backend)
+    return plot_backend.hist_frame(
+        data,
+        column=column,
+        by=by,
+        grid=grid,
+        xlabelsize=xlabelsize,
+        xrot=xrot,
+        ylabelsize=ylabelsize,
+        yrot=yrot,
+        ax=ax,
+        sharex=sharex,
+        sharey=sharey,
+        figsize=figsize,
+        layout=layout,
+        legend=legend,
+        bins=bins,
+        **kwargs,
+    )
+
+
+_boxplot_doc = """
+Make a box plot from DataFrame columns.
+
+Make a box-and-whisker plot from DataFrame columns, optionally grouped
+by some other columns. A box plot is a method for graphically depicting
+groups of numerical data through their quartiles.
+The box extends from the Q1 to Q3 quartile values of the data,
+with a line at the median (Q2). The whiskers extend from the edges
+of box to show the range of the data. By default, they extend no more than
+`1.5 * IQR (IQR = Q3 - Q1)` from the edges of the box, ending at the farthest
+data point within that interval. Outliers are plotted as separate dots.
+
+For further details see
+Wikipedia's entry for `boxplot <https://en.wikipedia.org/wiki/Box_plot>`_.
+
+Parameters
+----------
+%(data)s\
+column : str or list of str, optional
+    Column name or list of names, or vector.
+    Can be any valid input to :meth:`pandas.DataFrame.groupby`.
+by : str or array-like, optional
+    Column in the DataFrame to :meth:`pandas.DataFrame.groupby`.
+    One box-plot will be done per value of columns in `by`.
+ax : object of class matplotlib.axes.Axes, optional
+    The matplotlib axes to be used by boxplot.
+fontsize : float or str
+    Tick label font size in points or as a string (e.g., `large`).
+rot : float, default 0
+    The rotation angle of labels (in degrees)
+    with respect to the screen coordinate system.
+grid : bool, default True
+    Setting this to True will show the grid.
+figsize : A tuple (width, height) in inches
+    The size of the figure to create in matplotlib.
+layout : tuple (rows, columns), optional
+    For example, (3, 5) will display the subplots
+    using 3 rows and 5 columns, starting from the top-left.
+return_type : {'axes', 'dict', 'both'} or None, default 'axes'
+    The kind of object to return. The default is ``axes``.
+
+    * 'axes' returns the matplotlib axes the boxplot is drawn on.
+    * 'dict' returns a dictionary whose values are the matplotlib
+      Lines of the boxplot.
+    * 'both' returns a namedtuple with the axes and dict.
+    * when grouping with ``by``, a Series mapping columns to
+      ``return_type`` is returned.
+
+      If ``return_type`` is `None`, a NumPy array
+      of axes with the same shape as ``layout`` is returned.
+%(backend)s\
+
+**kwargs
+    All other plotting keyword arguments to be passed to
+    :func:`matplotlib.pyplot.boxplot`.
+
+Returns
+-------
+result
+    See Notes.
+
+See Also
+--------
+Series.plot.hist: Make a histogram.
+matplotlib.pyplot.boxplot : Matplotlib equivalent plot.
+
+Notes
+-----
+The return type depends on the `return_type` parameter:
+
+* 'axes' : object of class matplotlib.axes.Axes
+* 'dict' : dict of matplotlib.lines.Line2D objects
+* 'both' : a namedtuple with structure (ax, lines)
+
+For data grouped with ``by``, return a Series of the above or a numpy
+array:
+
+* :class:`~pandas.Series`
+* :class:`~numpy.array` (for ``return_type = None``)
+
+Use ``return_type='dict'`` when you want to tweak the appearance
+of the lines after plotting. In this case a dict containing the Lines
+making up the boxes, caps, fliers, medians, and whiskers is returned.
+
+Examples
+--------
+
+Boxplots can be created for every column in the dataframe
+by ``df.boxplot()`` or indicating the columns to be used:
+
+.. plot::
+    :context: close-figs
+
+    >>> np.random.seed(1234)
+    >>> df = pd.DataFrame(np.random.randn(10, 4),
+    ...                   columns=['Col1', 'Col2', 'Col3', 'Col4'])
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2', 'Col3'])  # doctest: +SKIP
+
+Boxplots of variables distributions grouped by the values of a third
+variable can be created using the option ``by``. For instance:
+
+.. plot::
+    :context: close-figs
+
+    >>> df = pd.DataFrame(np.random.randn(10, 2),
+    ...                   columns=['Col1', 'Col2'])
+    >>> df['X'] = pd.Series(['A', 'A', 'A', 'A', 'A',
+    ...                      'B', 'B', 'B', 'B', 'B'])
+    >>> boxplot = df.boxplot(by='X')
+
+A list of strings (i.e. ``['X', 'Y']``) can be passed to boxplot
+in order to group the data by combination of the variables in the x-axis:
+
+.. plot::
+    :context: close-figs
+
+    >>> df = pd.DataFrame(np.random.randn(10, 3),
+    ...                   columns=['Col1', 'Col2', 'Col3'])
+    >>> df['X'] = pd.Series(['A', 'A', 'A', 'A', 'A',
+    ...                      'B', 'B', 'B', 'B', 'B'])
+    >>> df['Y'] = pd.Series(['A', 'B', 'A', 'B', 'A',
+    ...                      'B', 'A', 'B', 'A', 'B'])
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2'], by=['X', 'Y'])
+
+The layout of boxplot can be adjusted giving a tuple to ``layout``:
+
+.. plot::
+    :context: close-figs
+
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2'], by='X',
+    ...                      layout=(2, 1))
+
+Additional formatting can be done to the boxplot, like suppressing the grid
+(``grid=False``), rotating the labels in the x-axis (i.e. ``rot=45``)
+or changing the fontsize (i.e. ``fontsize=15``):
+
+.. plot::
+    :context: close-figs
+
+    >>> boxplot = df.boxplot(grid=False, rot=45, fontsize=15)  # doctest: +SKIP
+
+The parameter ``return_type`` can be used to select the type of element
+returned by `boxplot`.  When ``return_type='axes'`` is selected,
+the matplotlib axes on which the boxplot is drawn are returned:
+
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2'], return_type='axes')
+    >>> type(boxplot)
+    <class 'matplotlib.axes._axes.Axes'>
+
+When grouping with ``by``, a Series mapping columns to ``return_type``
+is returned:
+
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2'], by='X',
+    ...                      return_type='axes')
+    >>> type(boxplot)
+    <class 'pandas.Series'>
+
+If ``return_type`` is `None`, a NumPy array of axes with the same shape
+as ``layout`` is returned:
+
+    >>> boxplot = df.boxplot(column=['Col1', 'Col2'], by='X',
+    ...                      return_type=None)
+    >>> type(boxplot)
+    <class 'numpy.ndarray'>
+"""
+
+_backend_doc = """\
+backend : str, default None
+    Backend to use instead of the backend specified in the option
+    ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+    specify the ``plotting.backend`` for the whole session, set
+    ``pd.options.plotting.backend``.
+"""
+
+
+_bar_or_line_doc = """
+        Parameters
+        ----------
+        x : label or position, optional
+            Allows plotting of one column versus another. If not specified,
+            the index of the DataFrame is used.
+        y : label or position, optional
+            Allows plotting of one column versus another. If not specified,
+            all numerical columns are used.
+        color : str, array-like, or dict, optional
+            The color for each of the DataFrame's columns. Possible values are:
+
+            - A single color string referred to by name, RGB or RGBA code,
+                for instance 'red' or '#a98d19'.
+
+            - A sequence of color strings referred to by name, RGB or RGBA
+                code, which will be used for each column recursively. For
+                instance ['green','yellow'] each column's %(kind)s will be filled in
+                green or yellow, alternatively. If there is only a single column to
+                be plotted, then only the first color from the color list will be
+                used.
+
+            - A dict of the form {column name : color}, so that each column will be
+                colored accordingly. For example, if your columns are called `a` and
+                `b`, then passing {'a': 'green', 'b': 'red'} will color %(kind)ss for
+                column `a` in green and %(kind)ss for column `b` in red.
+
+        **kwargs
+            Additional keyword arguments are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        matplotlib.axes.Axes or np.ndarray of them
+            An ndarray is returned with one :class:`matplotlib.axes.Axes`
+            per column when ``subplots=True``.
+"""
+
+
+@Substitution(data="data : DataFrame\n    The data to visualize.\n", backend="")
+@Appender(_boxplot_doc)
+def boxplot(
+    data: DataFrame,
+    column: str | list[str] | None = None,
+    by: str | list[str] | None = None,
+    ax: Axes | None = None,
+    fontsize: float | str | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    figsize: tuple[float, float] | None = None,
+    layout: tuple[int, int] | None = None,
+    return_type: str | None = None,
+    **kwargs,
+):
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.boxplot(
+        data,
+        column=column,
+        by=by,
+        ax=ax,
+        fontsize=fontsize,
+        rot=rot,
+        grid=grid,
+        figsize=figsize,
+        layout=layout,
+        return_type=return_type,
+        **kwargs,
+    )
+
+
+@Substitution(data="", backend=_backend_doc)
+@Appender(_boxplot_doc)
+def boxplot_frame(
+    self: DataFrame,
+    column=None,
+    by=None,
+    ax=None,
+    fontsize: int | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    return_type=None,
+    backend=None,
+    **kwargs,
+):
+    plot_backend = _get_plot_backend(backend)
+    return plot_backend.boxplot_frame(
+        self,
+        column=column,
+        by=by,
+        ax=ax,
+        fontsize=fontsize,
+        rot=rot,
+        grid=grid,
+        figsize=figsize,
+        layout=layout,
+        return_type=return_type,
+        **kwargs,
+    )
+
+
+def boxplot_frame_groupby(
+    grouped: DataFrameGroupBy,
+    subplots: bool = True,
+    column=None,
+    fontsize: int | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    ax=None,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    sharex: bool = False,
+    sharey: bool = True,
+    backend=None,
+    **kwargs,
+):
+    """
+    Make box plots from DataFrameGroupBy data.
+
+    Parameters
+    ----------
+    grouped : DataFrameGroupBy
+        The grouped DataFrame object over which to create the box plots.
+    subplots : bool
+        * ``False`` - no subplots will be used
+        * ``True`` - create a subplot for each group.
+    column : column name or list of names, or vector
+        Can be any valid input to groupby.
+    fontsize : float or str
+        Font size for the labels.
+    rot : float
+        Rotation angle of labels (in degrees) on the x-axis.
+    grid : bool
+        Whether to show grid lines on the plot.
+    ax : Matplotlib axis object, default None
+        The axes on which to draw the plots. If None, uses the current axes.
+    figsize : tuple of (float, float)
+        The figure size in inches (width, height).
+    layout : tuple (optional)
+        The layout of the plot: (rows, columns).
+    sharex : bool, default False
+        Whether x-axes will be shared among subplots.
+    sharey : bool, default True
+        Whether y-axes will be shared among subplots.
+    backend : str, default None
+        Backend to use instead of the backend specified in the option
+        ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+        specify the ``plotting.backend`` for the whole session, set
+        ``pd.options.plotting.backend``.
+    **kwargs
+        All other plotting keyword arguments to be passed to
+        matplotlib's boxplot function.
+
+    Returns
+    -------
+    dict or DataFrame.boxplot return value
+        If ``subplots=True``, returns a dictionary of group keys to the boxplot
+        return values. If ``subplots=False``, returns the boxplot return value
+        of a single DataFrame.
+
+    See Also
+    --------
+    DataFrame.boxplot : Create a box plot from a DataFrame.
+    Series.plot : Plot a Series.
+
+    Examples
+    --------
+    You can create boxplots for grouped data and show them as separate subplots:
+
+    .. plot::
+        :context: close-figs
+
+        >>> import itertools
+        >>> tuples = [t for t in itertools.product(range(1000), range(4))]
+        >>> index = pd.MultiIndex.from_tuples(tuples, names=["lvl0", "lvl1"])
+        >>> data = np.random.randn(len(index), 4)
+        >>> df = pd.DataFrame(data, columns=list("ABCD"), index=index)
+        >>> grouped = df.groupby(level="lvl1")
+        >>> grouped.boxplot(rot=45, fontsize=12, figsize=(8, 10))  # doctest: +SKIP
+
+    The ``subplots=False`` option shows the boxplots in a single figure.
+
+    .. plot::
+        :context: close-figs
+
+        >>> grouped.boxplot(subplots=False, rot=45, fontsize=12)  # doctest: +SKIP
+    """
+    plot_backend = _get_plot_backend(backend)
+    return plot_backend.boxplot_frame_groupby(
+        grouped,
+        subplots=subplots,
+        column=column,
+        fontsize=fontsize,
+        rot=rot,
+        grid=grid,
+        ax=ax,
+        figsize=figsize,
+        layout=layout,
+        sharex=sharex,
+        sharey=sharey,
+        **kwargs,
+    )
+
+
+class PlotAccessor(PandasObject):
+    """
+    Make plots of Series or DataFrame.
+
+    Uses the backend specified by the
+    option ``plotting.backend``. By default, matplotlib is used.
+
+    Parameters
+    ----------
+    data : Series or DataFrame
+        The object for which the method is called.
+
+    Attributes
+    ----------
+    x : label or position, default None
+        Only used if data is a DataFrame.
+    y : label, position or list of label, positions, default None
+        Allows plotting of one column versus another. Only used if data is a
+        DataFrame.
+    kind : str
+        The kind of plot to produce:
+
+        - 'line' : line plot (default)
+        - 'bar' : vertical bar plot
+        - 'barh' : horizontal bar plot
+        - 'hist' : histogram
+        - 'box' : boxplot
+        - 'kde' : Kernel Density Estimation plot
+        - 'density' : same as 'kde'
+        - 'area' : area plot
+        - 'pie' : pie plot
+        - 'scatter' : scatter plot (DataFrame only)
+        - 'hexbin' : hexbin plot (DataFrame only)
+    ax : matplotlib axes object, default None
+        An axes of the current figure.
+    subplots : bool or sequence of iterables, default False
+        Whether to group columns into subplots:
+
+        - ``False`` : No subplots will be used
+        - ``True`` : Make separate subplots for each column.
+        - sequence of iterables of column labels: Create a subplot for each
+          group of columns. For example `[('a', 'c'), ('b', 'd')]` will
+          create 2 subplots: one with columns 'a' and 'c', and one
+          with columns 'b' and 'd'. Remaining columns that aren't specified
+          will be plotted in additional subplots (one per column).
+
+          .. versionadded:: 1.5.0
+
+    sharex : bool, default True if ax is None else False
+        In case ``subplots=True``, share x axis and set some x axis labels
+        to invisible; defaults to True if ax is None otherwise False if
+        an ax is passed in; Be aware, that passing in both an ax and
+        ``sharex=True`` will alter all x axis labels for all axis in a figure.
+    sharey : bool, default False
+        In case ``subplots=True``, share y axis and set some y axis labels to invisible.
+    layout : tuple, optional
+        (rows, columns) for the layout of subplots.
+    figsize : a tuple (width, height) in inches
+        Size of a figure object.
+    use_index : bool, default True
+        Use index as ticks for x axis.
+    title : str or list
+        Title to use for the plot. If a string is passed, print the string
+        at the top of the figure. If a list is passed and `subplots` is
+        True, print each item in the list above the corresponding subplot.
+    grid : bool, default None (matlab style default)
+        Axis grid lines.
+    legend : bool or {'reverse'}
+        Place legend on axis subplots.
+    style : list or dict
+        The matplotlib line style per column.
+    logx : bool or 'sym', default False
+        Use log scaling or symlog scaling on x axis.
+
+    logy : bool or 'sym' default False
+        Use log scaling or symlog scaling on y axis.
+
+    loglog : bool or 'sym', default False
+        Use log scaling or symlog scaling on both x and y axes.
+
+    xticks : sequence
+        Values to use for the xticks.
+    yticks : sequence
+        Values to use for the yticks.
+    xlim : 2-tuple/list
+        Set the x limits of the current axes.
+    ylim : 2-tuple/list
+        Set the y limits of the current axes.
+    xlabel : label, optional
+        Name to use for the xlabel on x-axis. Default uses index name as xlabel, or the
+        x-column name for planar plots.
+
+        .. versionchanged:: 2.0.0
+
+            Now applicable to histograms.
+
+    ylabel : label, optional
+        Name to use for the ylabel on y-axis. Default will show no ylabel, or the
+        y-column name for planar plots.
+
+        .. versionchanged:: 2.0.0
+
+            Now applicable to histograms.
+
+    rot : float, default None
+        Rotation for ticks (xticks for vertical, yticks for horizontal
+        plots).
+    fontsize : float, default None
+        Font size for xticks and yticks.
+    colormap : str or matplotlib colormap object, default None
+        Colormap to select colors from. If string, load colormap with that
+        name from matplotlib.
+    colorbar : bool, optional
+        If True, plot colorbar (only relevant for 'scatter' and 'hexbin'
+        plots).
+    position : float
+        Specify relative alignments for bar plot layout.
+        From 0 (left/bottom-end) to 1 (right/top-end). Default is 0.5
+        (center).
+    table : bool, Series or DataFrame, default False
+        If True, draw a table using the data in the DataFrame and the data
+        will be transposed to meet matplotlib's default layout.
+        If a Series or DataFrame is passed, use passed data to draw a
+        table.
+    yerr : DataFrame, Series, array-like, dict and str
+        See :ref:`Plotting with Error Bars <visualization.errorbars>` for
+        detail.
+    xerr : DataFrame, Series, array-like, dict and str
+        Equivalent to yerr.
+    stacked : bool, default False in line and bar plots, and True in area plot
+        If True, create stacked plot.
+    secondary_y : bool or sequence, default False
+        Whether to plot on the secondary y-axis if a list/tuple, which
+        columns to plot on secondary y-axis.
+    mark_right : bool, default True
+        When using a secondary_y axis, automatically mark the column
+        labels with "(right)" in the legend.
+    include_bool : bool, default is False
+        If True, boolean values can be plotted.
+    backend : str, default None
+        Backend to use instead of the backend specified in the option
+        ``plotting.backend``. For instance, 'matplotlib'. Alternatively, to
+        specify the ``plotting.backend`` for the whole session, set
+        ``pd.options.plotting.backend``.
+    **kwargs
+        Options to pass to matplotlib plotting method.
+
+    Returns
+    -------
+    :class:`matplotlib.axes.Axes` or numpy.ndarray of them
+        If the backend is not the default matplotlib one, the return value
+        will be the object returned by the backend.
+
+    See Also
+    --------
+    matplotlib.pyplot.plot : Plot y versus x as lines and/or markers.
+    DataFrame.hist : Make a histogram.
+    DataFrame.boxplot : Make a box plot.
+    DataFrame.plot.scatter : Make a scatter plot with varying marker
+        point size and color.
+    DataFrame.plot.hexbin : Make a hexagonal binning plot of
+        two variables.
+    DataFrame.plot.kde : Make Kernel Density Estimate plot using
+        Gaussian kernels.
+    DataFrame.plot.area : Make a stacked area plot.
+    DataFrame.plot.bar : Make a bar plot.
+    DataFrame.plot.barh : Make a horizontal bar plot.
+
+    Notes
+    -----
+    - See matplotlib documentation online for more on this subject
+    - If `kind` = 'bar' or 'barh', you can specify relative alignments
+      for bar plot layout by `position` keyword.
+      From 0 (left/bottom-end) to 1 (right/top-end). Default is 0.5
+      (center)
+
+    Examples
+    --------
+    For Series:
+
+    .. plot::
+        :context: close-figs
+
+        >>> ser = pd.Series([1, 2, 3, 3])
+        >>> plot = ser.plot(kind="hist", title="My plot")
+
+    For DataFrame:
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "length": [1.5, 0.5, 1.2, 0.9, 3],
+        ...         "width": [0.7, 0.2, 0.15, 0.2, 1.1],
+        ...     },
+        ...     index=["pig", "rabbit", "duck", "chicken", "horse"],
+        ... )
+        >>> plot = df.plot(title="DataFrame Plot")
+
+    For SeriesGroupBy:
+
+    .. plot::
+        :context: close-figs
+
+        >>> lst = [-1, -2, -3, 1, 2, 3]
+        >>> ser = pd.Series([1, 2, 2, 4, 6, 6], index=lst)
+        >>> plot = ser.groupby(lambda x: x > 0).plot(title="SeriesGroupBy Plot")
+
+    For DataFrameGroupBy:
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.DataFrame({"col1": [1, 2, 3, 4], "col2": ["A", "B", "A", "B"]})
+        >>> plot = df.groupby("col2").plot(kind="bar", title="DataFrameGroupBy Plot")
+    """
+
+    _common_kinds = ("line", "bar", "barh", "kde", "density", "area", "hist", "box")
+    _series_kinds = ("pie",)
+    _dataframe_kinds = ("scatter", "hexbin")
+    _kind_aliases = {"density": "kde"}
+    _all_kinds = _common_kinds + _series_kinds + _dataframe_kinds
+
+    def __init__(self, data: Series | DataFrame) -> None:
+        self._parent = data
+
+    @staticmethod
+    def _get_call_args(backend_name: str, data: Series | DataFrame, args, kwargs):
+        """
+        This function makes calls to this accessor `__call__` method compatible
+        with the previous `SeriesPlotMethods.__call__` and
+        `DataFramePlotMethods.__call__`. Those had slightly different
+        signatures, since `DataFramePlotMethods` accepted `x` and `y`
+        parameters.
+        """
+        if isinstance(data, ABCSeries):
+            arg_def = [
+                ("kind", "line"),
+                ("ax", None),
+                ("figsize", None),
+                ("use_index", True),
+                ("title", None),
+                ("grid", None),
+                ("legend", False),
+                ("style", None),
+                ("logx", False),
+                ("logy", False),
+                ("loglog", False),
+                ("xticks", None),
+                ("yticks", None),
+                ("xlim", None),
+                ("ylim", None),
+                ("rot", None),
+                ("fontsize", None),
+                ("colormap", None),
+                ("table", False),
+                ("yerr", None),
+                ("xerr", None),
+                ("label", None),
+                ("secondary_y", False),
+                ("xlabel", None),
+                ("ylabel", None),
+            ]
+        elif isinstance(data, ABCDataFrame):
+            arg_def = [
+                ("x", None),
+                ("y", None),
+                ("kind", "line"),
+                ("ax", None),
+                ("subplots", False),
+                ("sharex", None),
+                ("sharey", False),
+                ("layout", None),
+                ("figsize", None),
+                ("use_index", True),
+                ("title", None),
+                ("grid", None),
+                ("legend", True),
+                ("style", None),
+                ("logx", False),
+                ("logy", False),
+                ("loglog", False),
+                ("xticks", None),
+                ("yticks", None),
+                ("xlim", None),
+                ("ylim", None),
+                ("rot", None),
+                ("fontsize", None),
+                ("colormap", None),
+                ("table", False),
+                ("yerr", None),
+                ("xerr", None),
+                ("secondary_y", False),
+                ("xlabel", None),
+                ("ylabel", None),
+            ]
+        else:
+            raise TypeError(
+                f"Called plot accessor for type {type(data).__name__}, "
+                "expected Series or DataFrame"
+            )
+
+        if args and isinstance(data, ABCSeries):
+            positional_args = str(args)[1:-1]
+            keyword_args = ", ".join(
+                [f"{name}={value!r}" for (name, _), value in zip(arg_def, args)]
+            )
+            msg = (
+                "`Series.plot()` should not be called with positional "
+                "arguments, only keyword arguments. The order of "
+                "positional arguments will change in the future. "
+                f"Use `Series.plot({keyword_args})` instead of "
+                f"`Series.plot({positional_args})`."
+            )
+            raise TypeError(msg)
+
+        pos_args = {name: value for (name, _), value in zip(arg_def, args)}
+        if backend_name == "pandas.plotting._matplotlib":
+            kwargs = dict(arg_def, **pos_args, **kwargs)
+        else:
+            kwargs = dict(pos_args, **kwargs)
+
+        x = kwargs.pop("x", None)
+        y = kwargs.pop("y", None)
+        kind = kwargs.pop("kind", "line")
+        return x, y, kind, kwargs
+
+    def __call__(self, *args, **kwargs):
+        plot_backend = _get_plot_backend(kwargs.pop("backend", None))
+
+        x, y, kind, kwargs = self._get_call_args(
+            plot_backend.__name__, self._parent, args, kwargs
+        )
+
+        kind = self._kind_aliases.get(kind, kind)
+
+        # when using another backend, get out of the way
+        if plot_backend.__name__ != "pandas.plotting._matplotlib":
+            return plot_backend.plot(self._parent, x=x, y=y, kind=kind, **kwargs)
+
+        if kind not in self._all_kinds:
+            raise ValueError(
+                f"{kind} is not a valid plot kind Valid plot kinds: {self._all_kinds}"
+            )
+
+        data = self._parent
+
+        if isinstance(data, ABCSeries):
+            kwargs["reuse_plot"] = True
+
+        if kind in self._dataframe_kinds:
+            if isinstance(data, ABCDataFrame):
+                return plot_backend.plot(data, x=x, y=y, kind=kind, **kwargs)
+            else:
+                raise ValueError(f"plot kind {kind} can only be used for data frames")
+        elif kind in self._series_kinds:
+            if isinstance(data, ABCDataFrame):
+                if y is None and kwargs.get("subplots") is False:
+                    raise ValueError(
+                        f"{kind} requires either y column or 'subplots=True'"
+                    )
+                if y is not None:
+                    if is_integer(y) and not holds_integer(data.columns):
+                        y = data.columns[y]
+                    # converted to series actually. copy to not modify
+                    data = data[y].copy(deep=False)
+                    data.index.name = y
+        elif isinstance(data, ABCDataFrame):
+            data_cols = data.columns
+            if x is not None:
+                if is_integer(x) and not holds_integer(data.columns):
+                    x = data_cols[x]
+                elif not isinstance(data[x], ABCSeries):
+                    raise ValueError("x must be a label or position")
+                data = data.set_index(x)
+            if y is not None:
+                # check if we have y as int or list of ints
+                int_ylist = is_list_like(y) and all(is_integer(c) for c in y)
+                int_y_arg = is_integer(y) or int_ylist
+                if int_y_arg and not holds_integer(data.columns):
+                    y = data_cols[y]
+
+                label_kw = kwargs["label"] if "label" in kwargs else False
+                for kw in ["xerr", "yerr"]:
+                    if kw in kwargs and (
+                        isinstance(kwargs[kw], str) or is_integer(kwargs[kw])
+                    ):
+                        try:
+                            kwargs[kw] = data[kwargs[kw]]
+                        except (IndexError, KeyError, TypeError):
+                            pass
+
+                data = data[y]
+
+                if isinstance(data, ABCSeries):
+                    label_name = label_kw or y
+                    data.name = label_name
+                else:
+                    # error: Argument 1 to "len" has incompatible type "Any | bool";
+                    # expected "Sized"  [arg-type]
+                    match = is_list_like(label_kw) and len(label_kw) == len(y)  # type: ignore[arg-type]
+                    if label_kw and not match:
+                        raise ValueError(
+                            "label should be list-like and same length as y"
+                        )
+                    label_name = label_kw or data.columns
+                    data.columns = label_name
+
+        return plot_backend.plot(data, kind=kind, **kwargs)
+
+    __call__.__doc__ = __doc__
+
+    @Appender(
+        """
+        See Also
+        --------
+        matplotlib.pyplot.plot : Plot y versus x as lines and/or markers.
+
+        Examples
+        --------
+
+        .. plot::
+            :context: close-figs
+
+            >>> s = pd.Series([1, 3, 2])
+            >>> s.plot.line()  # doctest: +SKIP
+
+        .. plot::
+            :context: close-figs
+
+            The following example shows the populations for some animals
+            over the years.
+
+            >>> df = pd.DataFrame({
+            ...     'pig': [20, 18, 489, 675, 1776],
+            ...     'horse': [4, 25, 281, 600, 1900]
+            ... }, index=[1990, 1997, 2003, 2009, 2014])
+            >>> lines = df.plot.line()
+
+        .. plot::
+           :context: close-figs
+
+           An example with subplots, so an array of axes is returned.
+
+           >>> axes = df.plot.line(subplots=True)
+           >>> type(axes)
+           <class 'numpy.ndarray'>
+
+        .. plot::
+           :context: close-figs
+
+           Let's repeat the same example, but specifying colors for
+           each column (in this case, for each animal).
+
+           >>> axes = df.plot.line(
+           ...     subplots=True, color={"pig": "pink", "horse": "#742802"}
+           ... )
+
+        .. plot::
+            :context: close-figs
+
+            The following example shows the relationship between both
+            populations.
+
+            >>> lines = df.plot.line(x='pig', y='horse')
+        """
+    )
+    @Substitution(kind="line")
+    @Appender(_bar_or_line_doc)
+    def line(
+        self,
+        x: Hashable | None = None,
+        y: Hashable | None = None,
+        color: str | Sequence[str] | dict | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Plot Series or DataFrame as lines.
+
+        This function is useful to plot lines using DataFrame's values
+        as coordinates.
+        """
+        if color is not None:
+            kwargs["color"] = color
+        return self(kind="line", x=x, y=y, **kwargs)
+
+    @Appender(
+        """
+        See Also
+        --------
+        DataFrame.plot.barh : Horizontal bar plot.
+        DataFrame.plot : Make plots of a DataFrame.
+        matplotlib.pyplot.bar : Make a bar plot with matplotlib.
+
+        Examples
+        --------
+        Basic plot.
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame({'lab': ['A', 'B', 'C'], 'val': [10, 30, 20]})
+            >>> ax = df.plot.bar(x='lab', y='val', rot=0)
+
+        Plot a whole dataframe to a bar plot. Each column is assigned a
+        distinct color, and each row is nested in a group along the
+        horizontal axis.
+
+        .. plot::
+            :context: close-figs
+
+            >>> speed = [0.1, 17.5, 40, 48, 52, 69, 88]
+            >>> lifespan = [2, 8, 70, 1.5, 25, 12, 28]
+            >>> index = ['snail', 'pig', 'elephant',
+            ...          'rabbit', 'giraffe', 'coyote', 'horse']
+            >>> df = pd.DataFrame({'speed': speed,
+            ...                    'lifespan': lifespan}, index=index)
+            >>> ax = df.plot.bar(rot=0)
+
+        Plot stacked bar charts for the DataFrame
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.bar(stacked=True)
+
+        Instead of nesting, the figure can be split by column with
+        ``subplots=True``. In this case, a :class:`numpy.ndarray` of
+        :class:`matplotlib.axes.Axes` are returned.
+
+        .. plot::
+            :context: close-figs
+
+            >>> axes = df.plot.bar(rot=0, subplots=True)
+            >>> axes[1].legend(loc=2)  # doctest: +SKIP
+
+        If you don't like the default colours, you can specify how you'd
+        like each column to be colored.
+
+        .. plot::
+            :context: close-figs
+
+            >>> axes = df.plot.bar(
+            ...     rot=0, subplots=True, color={"speed": "red", "lifespan": "green"}
+            ... )
+            >>> axes[1].legend(loc=2)  # doctest: +SKIP
+
+        Plot a single column.
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.bar(y='speed', rot=0)
+
+        Plot only selected categories for the DataFrame.
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.bar(x='lifespan', rot=0)
+    """
+    )
+    @Substitution(kind="bar")
+    @Appender(_bar_or_line_doc)
+    def bar(
+        self,
+        x: Hashable | None = None,
+        y: Hashable | None = None,
+        color: str | Sequence[str] | dict | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Vertical bar plot.
+
+        A bar plot is a plot that presents categorical data with
+        rectangular bars with lengths proportional to the values that they
+        represent. A bar plot shows comparisons among discrete categories. One
+        axis of the plot shows the specific categories being compared, and the
+        other axis represents a measured value.
+        """
+        if color is not None:
+            kwargs["color"] = color
+        return self(kind="bar", x=x, y=y, **kwargs)
+
+    @Appender(
+        """
+        See Also
+        --------
+        DataFrame.plot.bar : Vertical bar plot.
+        DataFrame.plot : Make plots of DataFrame using matplotlib.
+        matplotlib.axes.Axes.bar : Plot a vertical bar plot using matplotlib.
+
+        Examples
+        --------
+        Basic example
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame({'lab': ['A', 'B', 'C'], 'val': [10, 30, 20]})
+            >>> ax = df.plot.barh(x='lab', y='val')
+
+        Plot a whole DataFrame to a horizontal bar plot
+
+        .. plot::
+            :context: close-figs
+
+            >>> speed = [0.1, 17.5, 40, 48, 52, 69, 88]
+            >>> lifespan = [2, 8, 70, 1.5, 25, 12, 28]
+            >>> index = ['snail', 'pig', 'elephant',
+            ...          'rabbit', 'giraffe', 'coyote', 'horse']
+            >>> df = pd.DataFrame({'speed': speed,
+            ...                    'lifespan': lifespan}, index=index)
+            >>> ax = df.plot.barh()
+
+        Plot stacked barh charts for the DataFrame
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.barh(stacked=True)
+
+        We can specify colors for each column
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.barh(color={"speed": "red", "lifespan": "green"})
+
+        Plot a column of the DataFrame to a horizontal bar plot
+
+        .. plot::
+            :context: close-figs
+
+            >>> speed = [0.1, 17.5, 40, 48, 52, 69, 88]
+            >>> lifespan = [2, 8, 70, 1.5, 25, 12, 28]
+            >>> index = ['snail', 'pig', 'elephant',
+            ...          'rabbit', 'giraffe', 'coyote', 'horse']
+            >>> df = pd.DataFrame({'speed': speed,
+            ...                    'lifespan': lifespan}, index=index)
+            >>> ax = df.plot.barh(y='speed')
+
+        Plot DataFrame versus the desired column
+
+        .. plot::
+            :context: close-figs
+
+            >>> speed = [0.1, 17.5, 40, 48, 52, 69, 88]
+            >>> lifespan = [2, 8, 70, 1.5, 25, 12, 28]
+            >>> index = ['snail', 'pig', 'elephant',
+            ...          'rabbit', 'giraffe', 'coyote', 'horse']
+            >>> df = pd.DataFrame({'speed': speed,
+            ...                    'lifespan': lifespan}, index=index)
+            >>> ax = df.plot.barh(x='lifespan')
+    """
+    )
+    @Substitution(kind="bar")
+    @Appender(_bar_or_line_doc)
+    def barh(
+        self,
+        x: Hashable | None = None,
+        y: Hashable | None = None,
+        color: str | Sequence[str] | dict | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Make a horizontal bar plot.
+
+        A horizontal bar plot is a plot that presents quantitative data with
+        rectangular bars with lengths proportional to the values that they
+        represent. A bar plot shows comparisons among discrete categories. One
+        axis of the plot shows the specific categories being compared, and the
+        other axis represents a measured value.
+        """
+        if color is not None:
+            kwargs["color"] = color
+        return self(kind="barh", x=x, y=y, **kwargs)
+
+    def box(self, by: IndexLabel | None = None, **kwargs) -> PlotAccessor:
+        r"""
+        Make a box plot of the DataFrame columns.
+
+        A box plot is a method for graphically depicting groups of numerical
+        data through their quartiles.
+        The box extends from the Q1 to Q3 quartile values of the data,
+        with a line at the median (Q2). The whiskers extend from the edges
+        of box to show the range of the data. The position of the whiskers
+        is set by default to 1.5*IQR (IQR = Q3 - Q1) from the edges of the
+        box. Outlier points are those past the end of the whiskers.
+
+        For further details see Wikipedia's
+        entry for `boxplot <https://en.wikipedia.org/wiki/Box_plot>`__.
+
+        A consideration when using this chart is that the box and the whiskers
+        can overlap, which is very common when plotting small sets of data.
+
+        Parameters
+        ----------
+        by : str or sequence
+            Column in the DataFrame to group by.
+
+            .. versionchanged:: 1.4.0
+
+               Previously, `by` is silently ignore and makes no groupings
+
+        **kwargs
+            Additional keywords are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        :class:`matplotlib.axes.Axes` or numpy.ndarray of them
+            The matplotlib axes containing the box plot.
+
+        See Also
+        --------
+        DataFrame.boxplot: Another method to draw a box plot.
+        Series.plot.box: Draw a box plot from a Series object.
+        matplotlib.pyplot.boxplot: Draw a box plot in matplotlib.
+
+        Examples
+        --------
+        Draw a box plot from a DataFrame with four columns of randomly
+        generated data.
+
+        .. plot::
+            :context: close-figs
+
+            >>> data = np.random.randn(25, 4)
+            >>> df = pd.DataFrame(data, columns=list("ABCD"))
+            >>> ax = df.plot.box()
+
+        You can also generate groupings if you specify the `by` parameter (which
+        can take a column name, or a list or tuple of column names):
+
+        .. versionchanged:: 1.4.0
+
+        .. plot::
+            :context: close-figs
+
+            >>> age_list = [8, 10, 12, 14, 72, 74, 76, 78, 20, 25, 30, 35, 60, 85]
+            >>> df = pd.DataFrame({"gender": list("MMMMMMMMFFFFFF"), "age": age_list})
+            >>> ax = df.plot.box(column="age", by="gender", figsize=(10, 8))
+        """
+        return self(kind="box", by=by, **kwargs)
+
+    def hist(
+        self, by: IndexLabel | None = None, bins: int = 10, **kwargs
+    ) -> PlotAccessor:
+        """
+        Draw one histogram of the DataFrame's columns.
+
+        A histogram is a representation of the distribution of data.
+        This function groups the values of all given Series in the DataFrame
+        into bins and draws all bins in one :class:`matplotlib.axes.Axes`.
+        This is useful when the DataFrame's Series are in a similar scale.
+
+        Parameters
+        ----------
+        by : str or sequence, optional
+            Column in the DataFrame to group by.
+
+            .. versionchanged:: 1.4.0
+
+               Previously, `by` is silently ignore and makes no groupings
+
+        bins : int, default 10
+            Number of histogram bins to be used.
+        **kwargs
+            Additional keyword arguments are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        :class:`matplotlib.axes.Axes`
+            Return a histogram plot.
+
+        See Also
+        --------
+        DataFrame.hist : Draw histograms per DataFrame's Series.
+        Series.hist : Draw a histogram with Series' data.
+
+        Examples
+        --------
+        When we roll a die 6000 times, we expect to get each value around 1000
+        times. But when we roll two dice and sum the result, the distribution
+        is going to be quite different. A histogram illustrates those
+        distributions.
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(np.random.randint(1, 7, 6000), columns=["one"])
+            >>> df["two"] = df["one"] + np.random.randint(1, 7, 6000)
+            >>> ax = df.plot.hist(bins=12, alpha=0.5)
+
+        A grouped histogram can be generated by providing the parameter `by` (which
+        can be a column name, or a list of column names):
+
+        .. plot::
+            :context: close-figs
+
+            >>> age_list = [8, 10, 12, 14, 72, 74, 76, 78, 20, 25, 30, 35, 60, 85]
+            >>> df = pd.DataFrame({"gender": list("MMMMMMMMFFFFFF"), "age": age_list})
+            >>> ax = df.plot.hist(column=["age"], by="gender", figsize=(10, 8))
+        """
+        return self(kind="hist", by=by, bins=bins, **kwargs)
+
+    def kde(
+        self,
+        bw_method: Literal["scott", "silverman"] | float | Callable | None = None,
+        ind: np.ndarray | int | None = None,
+        weights: np.ndarray | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Generate Kernel Density Estimate plot using Gaussian kernels.
+
+        In statistics, `kernel density estimation`_ (KDE) is a non-parametric
+        way to estimate the probability density function (PDF) of a random
+        variable. This function uses Gaussian kernels and includes automatic
+        bandwidth determination.
+
+        .. _kernel density estimation:
+            https://en.wikipedia.org/wiki/Kernel_density_estimation
+
+        Parameters
+        ----------
+        bw_method : str, scalar or callable, optional
+            The method used to calculate the estimator bandwidth. This can be
+            'scott', 'silverman', a scalar constant or a callable.
+            If None (default), 'scott' is used.
+            See :class:`scipy.stats.gaussian_kde` for more information.
+        ind : NumPy array or int, optional
+            Evaluation points for the estimated PDF. If None (default),
+            1000 equally spaced points are used. If `ind` is a NumPy array, the
+            KDE is evaluated at the points passed. If `ind` is an integer,
+            `ind` number of equally spaced points are used.
+        weights : NumPy array, optional
+            Weights of datapoints. This must be the same shape as datapoints.
+            If None (default), the samples are assumed to be equally weighted.
+        **kwargs
+            Additional keyword arguments are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        matplotlib.axes.Axes or numpy.ndarray of them
+            The matplotlib axes containing the KDE plot.
+
+        See Also
+        --------
+        scipy.stats.gaussian_kde : Representation of a kernel-density
+            estimate using Gaussian kernels. This is the function used
+            internally to estimate the PDF.
+
+        Examples
+        --------
+        Given a Series of points randomly sampled from an unknown
+        distribution, estimate its PDF using KDE with automatic
+        bandwidth determination and plot the results, evaluating them at
+        1000 equally spaced points (default):
+
+        .. plot::
+            :context: close-figs
+
+            >>> s = pd.Series([1, 2, 2.5, 3, 3.5, 4, 5])
+            >>> ax = s.plot.kde()
+
+        A scalar bandwidth can be specified. Using a small bandwidth value can
+        lead to over-fitting, while using a large bandwidth value may result
+        in under-fitting:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = s.plot.kde(bw_method=0.3)
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = s.plot.kde(bw_method=3)
+
+        Finally, the `ind` parameter determines the evaluation points for the
+        plot of the estimated PDF:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = s.plot.kde(ind=[1, 2, 3, 4, 5])
+
+        For DataFrame, it works in the same way:
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(
+            ...     {
+            ...         "x": [1, 2, 2.5, 3, 3.5, 4, 5],
+            ...         "y": [4, 4, 4.5, 5, 5.5, 6, 6],
+            ...     }
+            ... )
+            >>> ax = df.plot.kde()
+
+        A scalar bandwidth can be specified. Using a small bandwidth value can
+        lead to over-fitting, while using a large bandwidth value may result
+        in under-fitting:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.kde(bw_method=0.3)
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.kde(bw_method=3)
+
+        Finally, the `ind` parameter determines the evaluation points for the
+        plot of the estimated PDF:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.kde(ind=[1, 2, 3, 4, 5, 6])
+        """
+        return self(kind="kde", bw_method=bw_method, ind=ind, weights=weights, **kwargs)
+
+    density = kde
+
+    def area(
+        self,
+        x: Hashable | None = None,
+        y: Hashable | None = None,
+        stacked: bool = True,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Draw a stacked area plot.
+
+        An area plot displays quantitative data visually.
+        This function wraps the matplotlib area function.
+
+        Parameters
+        ----------
+        x : label or position, optional
+            Coordinates for the X axis. By default uses the index.
+        y : label or position, optional
+            Column to plot. By default uses all columns.
+        stacked : bool, default True
+            Area plots are stacked by default. Set to False to create a
+            unstacked plot.
+        **kwargs
+            Additional keyword arguments are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        matplotlib.axes.Axes or numpy.ndarray
+            Area plot, or array of area plots if subplots is True.
+
+        See Also
+        --------
+        DataFrame.plot : Make plots of DataFrame using matplotlib.
+
+        Examples
+        --------
+        Draw an area plot based on basic business metrics:
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(
+            ...     {
+            ...         "sales": [3, 2, 3, 9, 10, 6],
+            ...         "signups": [5, 5, 6, 12, 14, 13],
+            ...         "visits": [20, 42, 28, 62, 81, 50],
+            ...     },
+            ...     index=pd.date_range(
+            ...         start="2018/01/01", end="2018/07/01", freq="ME"
+            ...     ),
+            ... )
+            >>> ax = df.plot.area()
+
+        Area plots are stacked by default. To produce an unstacked plot,
+        pass ``stacked=False``:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.area(stacked=False)
+
+        Draw an area plot for a single column:
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax = df.plot.area(y="sales")
+
+        Draw with a different `x`:
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(
+            ...     {
+            ...         "sales": [3, 2, 3],
+            ...         "visits": [20, 42, 28],
+            ...         "day": [1, 2, 3],
+            ...     }
+            ... )
+            >>> ax = df.plot.area(x="day")
+        """
+        return self(kind="area", x=x, y=y, stacked=stacked, **kwargs)
+
+    def pie(self, y: IndexLabel | None = None, **kwargs) -> PlotAccessor:
+        """
+        Generate a pie plot.
+
+        A pie plot is a proportional representation of the numerical data in a
+        column. This function wraps :meth:`matplotlib.pyplot.pie` for the
+        specified column. If no column reference is passed and
+        ``subplots=True`` a pie plot is drawn for each numerical column
+        independently.
+
+        Parameters
+        ----------
+        y : int or label, optional
+            Label or position of the column to plot.
+            If not provided, ``subplots=True`` argument must be passed.
+        **kwargs
+            Keyword arguments to pass on to :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        matplotlib.axes.Axes or np.ndarray of them
+            A NumPy array is returned when `subplots` is True.
+
+        See Also
+        --------
+        Series.plot.pie : Generate a pie plot for a Series.
+        DataFrame.plot : Make plots of a DataFrame.
+
+        Examples
+        --------
+        In the example below we have a DataFrame with the information about
+        planet's mass and radius. We pass the 'mass' column to the
+        pie function to get a pie plot.
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(
+            ...     {"mass": [0.330, 4.87, 5.97], "radius": [2439.7, 6051.8, 6378.1]},
+            ...     index=["Mercury", "Venus", "Earth"],
+            ... )
+            >>> plot = df.plot.pie(y="mass", figsize=(5, 5))
+
+        .. plot::
+            :context: close-figs
+
+            >>> plot = df.plot.pie(subplots=True, figsize=(11, 6))
+        """
+        if y is not None:
+            kwargs["y"] = y
+        if (
+            isinstance(self._parent, ABCDataFrame)
+            and kwargs.get("y", None) is None
+            and not kwargs.get("subplots", False)
+        ):
+            raise ValueError("pie requires either y column or 'subplots=True'")
+        return self(kind="pie", **kwargs)
+
+    def scatter(
+        self,
+        x: Hashable,
+        y: Hashable,
+        s: Hashable | Sequence[Hashable] | None = None,
+        c: Hashable | Sequence[Hashable] | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Create a scatter plot with varying marker point size and color.
+
+        The coordinates of each point are defined by two dataframe columns and
+        filled circles are used to represent each point. This kind of plot is
+        useful to see complex correlations between two variables. Points could
+        be for instance natural 2D coordinates like longitude and latitude in
+        a map or, in general, any pair of metrics that can be plotted against
+        each other.
+
+        Parameters
+        ----------
+        x : int or str
+            The column name or column position to be used as horizontal
+            coordinates for each point.
+        y : int or str
+            The column name or column position to be used as vertical
+            coordinates for each point.
+        s : str, scalar or array-like, optional
+            The size of each point. Possible values are:
+
+            - A string with the name of the column to be used for marker's size.
+
+            - A single scalar so all points have the same size.
+
+            - A sequence of scalars, which will be used for each point's size
+              recursively. For instance, when passing [2,14] all points size
+              will be either 2 or 14, alternatively.
+
+        c : str, int or array-like, optional
+            The color of each point. Possible values are:
+
+            - A single color string referred to by name, RGB or RGBA code,
+              for instance 'red' or '#a98d19'.
+
+            - A sequence of color strings referred to by name, RGB or RGBA
+              code, which will be used for each point's color recursively. For
+              instance ['green','yellow'] all points will be filled in green or
+              yellow, alternatively.
+
+            - A column name or position whose values will be used to color the
+              marker points according to a colormap.
+
+        **kwargs
+            Keyword arguments to pass on to :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        :class:`matplotlib.axes.Axes` or numpy.ndarray of them
+            The matplotlib axes containing the scatter plot.
+
+        See Also
+        --------
+        matplotlib.pyplot.scatter : Scatter plot using multiple input data
+            formats.
+
+        Examples
+        --------
+        Let's see how to draw a scatter plot using coordinates from the values
+        in a DataFrame's columns.
+
+        .. plot::
+            :context: close-figs
+
+            >>> df = pd.DataFrame(
+            ...     [
+            ...         [5.1, 3.5, 0],
+            ...         [4.9, 3.0, 0],
+            ...         [7.0, 3.2, 1],
+            ...         [6.4, 3.2, 1],
+            ...         [5.9, 3.0, 2],
+            ...     ],
+            ...     columns=["length", "width", "species"],
+            ... )
+            >>> ax1 = df.plot.scatter(x="length", y="width", c="DarkBlue")
+
+        And now with the color determined by a column as well.
+
+        .. plot::
+            :context: close-figs
+
+            >>> ax2 = df.plot.scatter(
+            ...     x="length", y="width", c="species", colormap="viridis"
+            ... )
+        """
+        return self(kind="scatter", x=x, y=y, s=s, c=c, **kwargs)
+
+    def hexbin(
+        self,
+        x: Hashable,
+        y: Hashable,
+        C: Hashable | None = None,
+        reduce_C_function: Callable | None = None,
+        gridsize: int | tuple[int, int] | None = None,
+        **kwargs,
+    ) -> PlotAccessor:
+        """
+        Generate a hexagonal binning plot.
+
+        Generate a hexagonal binning plot of `x` versus `y`. If `C` is `None`
+        (the default), this is a histogram of the number of occurrences
+        of the observations at ``(x[i], y[i])``.
+
+        If `C` is specified, specifies values at given coordinates
+        ``(x[i], y[i])``. These values are accumulated for each hexagonal
+        bin and then reduced according to `reduce_C_function`,
+        having as default the NumPy's mean function (:meth:`numpy.mean`).
+        (If `C` is specified, it must also be a 1-D sequence
+        of the same length as `x` and `y`, or a column label.)
+
+        Parameters
+        ----------
+        x : int or str
+            The column label or position for x points.
+        y : int or str
+            The column label or position for y points.
+        C : int or str, optional
+            The column label or position for the value of `(x, y)` point.
+        reduce_C_function : callable, default `np.mean`
+            Function of one argument that reduces all the values in a bin to
+            a single number (e.g. `np.mean`, `np.max`, `np.sum`, `np.std`).
+        gridsize : int or tuple of (int, int), default 100
+            The number of hexagons in the x-direction.
+            The corresponding number of hexagons in the y-direction is
+            chosen in a way that the hexagons are approximately regular.
+            Alternatively, gridsize can be a tuple with two elements
+            specifying the number of hexagons in the x-direction and the
+            y-direction.
+        **kwargs
+            Additional keyword arguments are documented in
+            :meth:`DataFrame.plot`.
+
+        Returns
+        -------
+        matplotlib.Axes
+            The matplotlib ``Axes`` on which the hexbin is plotted.
+
+        See Also
+        --------
+        DataFrame.plot : Make plots of a DataFrame.
+        matplotlib.pyplot.hexbin : Hexagonal binning plot using matplotlib,
+            the matplotlib function that is used under the hood.
+
+        Examples
+        --------
+        The following examples are generated with random data from
+        a normal distribution.
+
+        .. plot::
+            :context: close-figs
+
+            >>> n = 10000
+            >>> df = pd.DataFrame({"x": np.random.randn(n), "y": np.random.randn(n)})
+            >>> ax = df.plot.hexbin(x="x", y="y", gridsize=20)
+
+        The next example uses `C` and `np.sum` as `reduce_C_function`.
+        Note that `'observations'` values ranges from 1 to 5 but the result
+        plot shows values up to more than 25. This is because of the
+        `reduce_C_function`.
+
+        .. plot::
+            :context: close-figs
+
+            >>> n = 500
+            >>> df = pd.DataFrame(
+            ...     {
+            ...         "coord_x": np.random.uniform(-3, 3, size=n),
+            ...         "coord_y": np.random.uniform(30, 50, size=n),
+            ...         "observations": np.random.randint(1, 5, size=n),
+            ...     }
+            ... )
+            >>> ax = df.plot.hexbin(
+            ...     x="coord_x",
+            ...     y="coord_y",
+            ...     C="observations",
+            ...     reduce_C_function=np.sum,
+            ...     gridsize=10,
+            ...     cmap="viridis",
+            ... )
+        """
+        if reduce_C_function is not None:
+            kwargs["reduce_C_function"] = reduce_C_function
+        if gridsize is not None:
+            kwargs["gridsize"] = gridsize
+
+        return self(kind="hexbin", x=x, y=y, C=C, **kwargs)
+
+
+_backends: dict[str, types.ModuleType] = {}
+
+
+def _load_backend(backend: str) -> types.ModuleType:
+    """
+    Load a pandas plotting backend.
+
+    Parameters
+    ----------
+    backend : str
+        The identifier for the backend. Either an entrypoint item registered
+        with importlib.metadata, "matplotlib", or a module name.
+
+    Returns
+    -------
+    types.ModuleType
+        The imported backend.
+    """
+    from importlib.metadata import entry_points
+
+    if backend == "matplotlib":
+        # Because matplotlib is an optional dependency and first-party backend,
+        # we need to attempt an import here to raise an ImportError if needed.
+        try:
+            module = importlib.import_module("pandas.plotting._matplotlib")
+        except ImportError:
+            raise ImportError(
+                "matplotlib is required for plotting when the "
+                'default backend "matplotlib" is selected.'
+            ) from None
+        return module
+
+    found_backend = False
+
+    eps = entry_points()
+    key = "pandas_plotting_backends"
+    # entry_points lost dict API ~ PY 3.10
+    # https://github.com/python/importlib_metadata/issues/298
+    if hasattr(eps, "select"):
+        entry = eps.select(group=key)
+    else:
+        # Argument 2 to "get" of "dict" has incompatible type "Tuple[]";
+        # expected "EntryPoints"  [arg-type]
+        entry = eps.get(key, ())  # type: ignore[arg-type]
+    for entry_point in entry:
+        found_backend = entry_point.name == backend
+        if found_backend:
+            module = entry_point.load()
+            break
+
+    if not found_backend:
+        # Fall back to unregistered, module name approach.
+        try:
+            module = importlib.import_module(backend)
+            found_backend = True
+        except ImportError:
+            # We re-raise later on.
+            pass
+
+    if found_backend:
+        if hasattr(module, "plot"):
+            # Validate that the interface is implemented when the option is set,
+            # rather than at plot time.
+            return module
+
+    raise ValueError(
+        f"Could not find plotting backend '{backend}'. Ensure that you've "
+        f"installed the package providing the '{backend}' entrypoint, or that "
+        "the package has a top-level `.plot` method."
+    )
+
+
+def _get_plot_backend(backend: str | None = None):
+    """
+    Return the plotting backend to use (e.g. `pandas.plotting._matplotlib`).
+
+    The plotting system of pandas uses matplotlib by default, but the idea here
+    is that it can also work with other third-party backends. This function
+    returns the module which provides a top-level `.plot` method that will
+    actually do the plotting. The backend is specified from a string, which
+    either comes from the keyword argument `backend`, or, if not specified, from
+    the option `pandas.options.plotting.backend`. All the rest of the code in
+    this file uses the backend specified there for the plotting.
+
+    The backend is imported lazily, as matplotlib is a soft dependency, and
+    pandas can be used without it being installed.
+
+    Notes
+    -----
+    Modifies `_backends` with imported backend as a side effect.
+    """
+    backend_str: str = backend or get_option("plotting.backend")
+
+    if backend_str in _backends:
+        return _backends[backend_str]
+
+    module = _load_backend(backend_str)
+    _backends[backend_str] = module
+    return module
diff --git a/pandas/pandas/plotting/_matplotlib/__init__.py b/pandas/pandas/plotting/_matplotlib/__init__.py
new file mode 100644
index 0000000000000..ff28868aa0033
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/__init__.py
@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas.plotting._matplotlib.boxplot import (
+    BoxPlot,
+    boxplot,
+    boxplot_frame,
+    boxplot_frame_groupby,
+)
+from pandas.plotting._matplotlib.converter import (
+    deregister,
+    register,
+)
+from pandas.plotting._matplotlib.core import (
+    AreaPlot,
+    BarhPlot,
+    BarPlot,
+    HexBinPlot,
+    LinePlot,
+    PiePlot,
+    ScatterPlot,
+)
+from pandas.plotting._matplotlib.hist import (
+    HistPlot,
+    KdePlot,
+    hist_frame,
+    hist_series,
+)
+from pandas.plotting._matplotlib.misc import (
+    andrews_curves,
+    autocorrelation_plot,
+    bootstrap_plot,
+    lag_plot,
+    parallel_coordinates,
+    radviz,
+    scatter_matrix,
+)
+from pandas.plotting._matplotlib.tools import table
+
+if TYPE_CHECKING:
+    from pandas.plotting._matplotlib.core import MPLPlot
+
+PLOT_CLASSES: dict[str, type[MPLPlot]] = {
+    "line": LinePlot,
+    "bar": BarPlot,
+    "barh": BarhPlot,
+    "box": BoxPlot,
+    "hist": HistPlot,
+    "kde": KdePlot,
+    "area": AreaPlot,
+    "pie": PiePlot,
+    "scatter": ScatterPlot,
+    "hexbin": HexBinPlot,
+}
+
+
+def plot(data, kind, **kwargs):
+    # Importing pyplot at the top of the file (before the converters are
+    # registered) causes problems in matplotlib 2 (converters seem to not
+    # work)
+    import matplotlib.pyplot as plt
+
+    if kwargs.pop("reuse_plot", False):
+        ax = kwargs.get("ax")
+        if ax is None and len(plt.get_fignums()) > 0:
+            with plt.rc_context():
+                ax = plt.gca()
+            kwargs["ax"] = getattr(ax, "left_ax", ax)
+    plot_obj = PLOT_CLASSES[kind](data, **kwargs)
+    plot_obj.generate()
+    plt.draw_if_interactive()
+    return plot_obj.result
+
+
+__all__ = [
+    "andrews_curves",
+    "autocorrelation_plot",
+    "bootstrap_plot",
+    "boxplot",
+    "boxplot_frame",
+    "boxplot_frame_groupby",
+    "deregister",
+    "hist_frame",
+    "hist_series",
+    "lag_plot",
+    "parallel_coordinates",
+    "plot",
+    "radviz",
+    "register",
+    "scatter_matrix",
+    "table",
+]
diff --git a/pandas/pandas/plotting/_matplotlib/boxplot.py b/pandas/pandas/plotting/_matplotlib/boxplot.py
new file mode 100644
index 0000000000000..af77972da8634
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/boxplot.py
@@ -0,0 +1,560 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Literal,
+    NamedTuple,
+)
+import warnings
+
+import matplotlib as mpl
+import numpy as np
+
+from pandas._libs import lib
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import is_dict_like
+from pandas.core.dtypes.generic import ABCSeries
+from pandas.core.dtypes.missing import remove_na_arraylike
+
+import pandas as pd
+import pandas.core.common as com
+from pandas.util.version import Version
+
+from pandas.io.formats.printing import pprint_thing
+from pandas.plotting._matplotlib.core import (
+    LinePlot,
+    MPLPlot,
+)
+from pandas.plotting._matplotlib.groupby import create_iter_data_given_by
+from pandas.plotting._matplotlib.style import get_standard_colors
+from pandas.plotting._matplotlib.tools import (
+    create_subplots,
+    flatten_axes,
+    maybe_adjust_figure,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Collection
+
+    from matplotlib.axes import Axes
+    from matplotlib.figure import Figure
+    from matplotlib.lines import Line2D
+
+    from pandas._typing import MatplotlibColor
+
+
+def _set_ticklabels(ax: Axes, labels: list[str], is_vertical: bool, **kwargs) -> None:
+    """Set the tick labels of a given axis.
+
+    Due to https://github.com/matplotlib/matplotlib/pull/17266, we need to handle the
+    case of repeated ticks (due to `FixedLocator`) and thus we duplicate the number of
+    labels.
+    """
+    ticks = ax.get_xticks() if is_vertical else ax.get_yticks()
+    if len(ticks) != len(labels):
+        i, remainder = divmod(len(ticks), len(labels))
+        if Version(mpl.__version__) < Version("3.10"):
+            assert remainder == 0, remainder
+        labels *= i
+    if is_vertical:
+        ax.set_xticklabels(labels, **kwargs)
+    else:
+        ax.set_yticklabels(labels, **kwargs)
+
+
+class BoxPlot(LinePlot):
+    @property
+    def _kind(self) -> Literal["box"]:
+        return "box"
+
+    _layout_type = "horizontal"
+
+    _valid_return_types = (None, "axes", "dict", "both")
+
+    class BP(NamedTuple):
+        # namedtuple to hold results
+        ax: Axes
+        lines: dict[str, list[Line2D]]
+
+    def __init__(self, data, return_type: str = "axes", **kwargs) -> None:
+        if return_type not in self._valid_return_types:
+            raise ValueError("return_type must be {None, 'axes', 'dict', 'both'}")
+
+        self.return_type = return_type
+        # Do not call LinePlot.__init__ which may fill nan
+        MPLPlot.__init__(self, data, **kwargs)
+
+        if self.subplots:
+            # Disable label ax sharing. Otherwise, all subplots shows last
+            # column label
+            if self.orientation == "vertical":
+                self.sharex = False
+            else:
+                self.sharey = False
+
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls, ax: Axes, y: np.ndarray, column_num=None, return_type: str = "axes", **kwds
+    ):
+        ys: np.ndarray | list[np.ndarray]
+        if y.ndim == 2:
+            ys = [remove_na_arraylike(v) for v in y]
+            # Boxplot fails with empty arrays, so need to add a NaN
+            #   if any cols are empty
+            # GH 8181
+            ys = [v if v.size > 0 else np.array([np.nan]) for v in ys]
+        else:
+            ys = remove_na_arraylike(y)
+        bp = ax.boxplot(ys, **kwds)
+
+        if return_type == "dict":
+            return bp, bp
+        elif return_type == "both":
+            return cls.BP(ax=ax, lines=bp), bp
+        else:
+            return ax, bp
+
+    def _validate_color_args(self, color, colormap):
+        if color is lib.no_default:
+            return None
+
+        if colormap is not None:
+            warnings.warn(
+                "'color' and 'colormap' cannot be used simultaneously. Using 'color'",
+                stacklevel=find_stack_level(),
+            )
+
+        if isinstance(color, dict):
+            valid_keys = ["boxes", "whiskers", "medians", "caps"]
+            for key in color:
+                if key not in valid_keys:
+                    raise ValueError(
+                        f"color dict contains invalid key '{key}'. "
+                        f"The key must be either {valid_keys}"
+                    )
+        return color
+
+    @cache_readonly
+    def _color_attrs(self):
+        # get standard colors for default
+        # use 2 colors by default, for box/whisker and median
+        # flier colors isn't needed here
+        # because it can be specified by ``sym`` kw
+        return get_standard_colors(num_colors=3, colormap=self.colormap, color=None)
+
+    @cache_readonly
+    def _boxes_c(self):
+        return self._color_attrs[0]
+
+    @cache_readonly
+    def _whiskers_c(self):
+        return self._color_attrs[0]
+
+    @cache_readonly
+    def _medians_c(self):
+        return self._color_attrs[2]
+
+    @cache_readonly
+    def _caps_c(self):
+        return self._color_attrs[0]
+
+    def _get_colors(
+        self,
+        num_colors=None,
+        color_kwds: dict[str, MatplotlibColor]
+        | MatplotlibColor
+        | Collection[MatplotlibColor]
+        | None = "color",
+    ) -> None:
+        pass
+
+    def maybe_color_bp(self, bp) -> None:
+        if isinstance(self.color, dict):
+            boxes = self.color.get("boxes", self._boxes_c)
+            whiskers = self.color.get("whiskers", self._whiskers_c)
+            medians = self.color.get("medians", self._medians_c)
+            caps = self.color.get("caps", self._caps_c)
+        else:
+            # Other types are forwarded to matplotlib
+            # If None, use default colors
+            boxes = self.color or self._boxes_c
+            whiskers = self.color or self._whiskers_c
+            medians = self.color or self._medians_c
+            caps = self.color or self._caps_c
+
+        color_tup = (boxes, whiskers, medians, caps)
+        maybe_color_bp(bp, color_tup=color_tup, **self.kwds)
+
+    def _make_plot(self, fig: Figure) -> None:
+        if self.subplots:
+            self._return_obj = pd.Series(dtype=object)
+
+            # Re-create iterated data if `by` is assigned by users
+            data = (
+                create_iter_data_given_by(self.data, self._kind)
+                if self.by is not None
+                else self.data
+            )
+
+            for i, (label, y) in enumerate(self._iter_data(data=data)):
+                ax = self._get_ax(i)
+                kwds = self.kwds.copy()
+
+                # When by is applied, show title for subplots to know which group it is
+                # just like df.boxplot, and need to apply T on y to provide right input
+                if self.by is not None:
+                    y = y.T
+                    ax.set_title(pprint_thing(label))
+
+                    # When `by` is assigned, the ticklabels will become unique grouped
+                    # values, instead of label which is used as subtitle in this case.
+                    # error: "Index" has no attribute "levels"; maybe "nlevels"?
+                    levels = self.data.columns.levels  # type: ignore[attr-defined]
+                    ticklabels = [pprint_thing(col) for col in levels[0]]
+                else:
+                    ticklabels = [pprint_thing(label)]
+
+                ret, bp = self._plot(
+                    ax, y, column_num=i, return_type=self.return_type, **kwds
+                )
+                self.maybe_color_bp(bp)
+                self._return_obj[label] = ret
+                _set_ticklabels(
+                    ax=ax, labels=ticklabels, is_vertical=self.orientation == "vertical"
+                )
+        else:
+            y = self.data.values.T
+            ax = self._get_ax(0)
+            kwds = self.kwds.copy()
+
+            ret, bp = self._plot(
+                ax, y, column_num=0, return_type=self.return_type, **kwds
+            )
+            self.maybe_color_bp(bp)
+            self._return_obj = ret
+
+            labels = [pprint_thing(left) for left in self.data.columns]
+            if not self.use_index:
+                labels = [pprint_thing(key) for key in range(len(labels))]
+            _set_ticklabels(
+                ax=ax, labels=labels, is_vertical=self.orientation == "vertical"
+            )
+
+    def _make_legend(self) -> None:
+        pass
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        # GH 45465: make sure that the boxplot doesn't ignore xlabel/ylabel
+        if self.xlabel:
+            ax.set_xlabel(pprint_thing(self.xlabel))
+        if self.ylabel:
+            ax.set_ylabel(pprint_thing(self.ylabel))
+
+    @property
+    def orientation(self) -> Literal["horizontal", "vertical"]:
+        if self.kwds.get("vert", True):
+            return "vertical"
+        else:
+            return "horizontal"
+
+    @property
+    def result(self):
+        if self.return_type is None:
+            return super().result
+        else:
+            return self._return_obj
+
+
+def maybe_color_bp(bp, color_tup, **kwds) -> None:
+    # GH#30346, when users specifying those arguments explicitly, our defaults
+    # for these four kwargs should be overridden; if not, use Pandas settings
+    if not kwds.get("boxprops"):
+        mpl.artist.setp(bp["boxes"], color=color_tup[0], alpha=1)
+    if not kwds.get("whiskerprops"):
+        mpl.artist.setp(bp["whiskers"], color=color_tup[1], alpha=1)
+    if not kwds.get("medianprops"):
+        mpl.artist.setp(bp["medians"], color=color_tup[2], alpha=1)
+    if not kwds.get("capprops"):
+        mpl.artist.setp(bp["caps"], color=color_tup[3], alpha=1)
+
+
+def _grouped_plot_by_column(
+    plotf,
+    data,
+    columns=None,
+    by=None,
+    numeric_only: bool = True,
+    grid: bool = False,
+    figsize: tuple[float, float] | None = None,
+    ax=None,
+    layout=None,
+    return_type=None,
+    **kwargs,
+):
+    grouped = data.groupby(by, observed=False)
+    if columns is None:
+        if not isinstance(by, (list, tuple)):
+            by = [by]
+        columns = data._get_numeric_data().columns.difference(by)
+    naxes = len(columns)
+    fig, axes = create_subplots(
+        naxes=naxes,
+        sharex=kwargs.pop("sharex", True),
+        sharey=kwargs.pop("sharey", True),
+        figsize=figsize,
+        ax=ax,
+        layout=layout,
+    )
+
+    # GH 45465: move the "by" label based on "vert"
+    xlabel, ylabel = kwargs.pop("xlabel", None), kwargs.pop("ylabel", None)
+    if kwargs.get("vert", True):
+        xlabel = xlabel or by
+    else:
+        ylabel = ylabel or by
+
+    ax_values = []
+
+    for ax, col in zip(flatten_axes(axes), columns):
+        gp_col = grouped[col]
+        keys, values = zip(*gp_col)
+        re_plotf = plotf(keys, values, ax, xlabel=xlabel, ylabel=ylabel, **kwargs)
+        ax.set_title(col)
+        ax_values.append(re_plotf)
+        ax.grid(grid)
+
+    result = pd.Series(ax_values, index=columns, copy=False)
+
+    # Return axes in multiplot case, maybe revisit later # 985
+    if return_type is None:
+        result = axes
+
+    byline = by[0] if len(by) == 1 else by
+    fig.suptitle(f"Boxplot grouped by {byline}")
+    maybe_adjust_figure(fig, bottom=0.15, top=0.9, left=0.1, right=0.9, wspace=0.2)
+
+    return result
+
+
+def boxplot(
+    data,
+    column=None,
+    by=None,
+    ax=None,
+    fontsize: int | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    return_type=None,
+    **kwds,
+):
+    import matplotlib.pyplot as plt
+
+    # validate return_type:
+    if return_type not in BoxPlot._valid_return_types:
+        raise ValueError("return_type must be {'axes', 'dict', 'both'}")
+
+    if isinstance(data, ABCSeries):
+        data = data.to_frame("x")
+        column = "x"
+
+    def _get_colors():
+        #  num_colors=3 is required as method maybe_color_bp takes the colors
+        #  in positions 0 and 2.
+        #  if colors not provided, use same defaults as DataFrame.plot.box
+        result_list = get_standard_colors(num_colors=3)
+        result = np.take(result_list, [0, 0, 2])
+        result = np.append(result, "k")
+
+        colors = kwds.pop("color", None)
+        if colors:
+            if is_dict_like(colors):
+                # replace colors in result array with user-specified colors
+                # taken from the colors dict parameter
+                # "boxes" value placed in position 0, "whiskers" in 1, etc.
+                valid_keys = ["boxes", "whiskers", "medians", "caps"]
+                key_to_index = dict(zip(valid_keys, range(4)))
+                for key, value in colors.items():
+                    if key in valid_keys:
+                        result[key_to_index[key]] = value
+                    else:
+                        raise ValueError(
+                            f"color dict contains invalid key '{key}'. "
+                            f"The key must be either {valid_keys}"
+                        )
+            else:
+                result.fill(colors)
+
+        return result
+
+    def plot_group(keys, values, ax: Axes, **kwds):
+        # GH 45465: xlabel/ylabel need to be popped out before plotting happens
+        xlabel, ylabel = kwds.pop("xlabel", None), kwds.pop("ylabel", None)
+        if xlabel:
+            ax.set_xlabel(pprint_thing(xlabel))
+        if ylabel:
+            ax.set_ylabel(pprint_thing(ylabel))
+
+        keys = [pprint_thing(x) for x in keys]
+        values = [np.asarray(remove_na_arraylike(v), dtype=object) for v in values]
+        bp = ax.boxplot(values, **kwds)
+        if fontsize is not None:
+            ax.tick_params(axis="both", labelsize=fontsize)
+
+        # GH 45465: x/y are flipped when "vert" changes
+        _set_ticklabels(
+            ax=ax, labels=keys, is_vertical=kwds.get("vert", True), rotation=rot
+        )
+        maybe_color_bp(bp, color_tup=colors, **kwds)
+
+        # Return axes in multiplot case, maybe revisit later # 985
+        if return_type == "dict":
+            return bp
+        elif return_type == "both":
+            return BoxPlot.BP(ax=ax, lines=bp)
+        else:
+            return ax
+
+    colors = _get_colors()
+    if column is None:
+        columns = None
+    elif isinstance(column, (list, tuple)):
+        columns = column
+    else:
+        columns = [column]
+
+    if by is not None:
+        # Prefer array return type for 2-D plots to match the subplot layout
+        # https://github.com/pandas-dev/pandas/pull/12216#issuecomment-241175580
+        result = _grouped_plot_by_column(
+            plot_group,
+            data,
+            columns=columns,
+            by=by,
+            grid=grid,
+            figsize=figsize,
+            ax=ax,
+            layout=layout,
+            return_type=return_type,
+            **kwds,
+        )
+    else:
+        if return_type is None:
+            return_type = "axes"
+        if layout is not None:
+            raise ValueError("The 'layout' keyword is not supported when 'by' is None")
+
+        if ax is None:
+            rc = {"figure.figsize": figsize} if figsize is not None else {}
+            with mpl.rc_context(rc):
+                ax = plt.gca()
+        data = data._get_numeric_data()
+        naxes = len(data.columns)
+        if naxes == 0:
+            raise ValueError(
+                "boxplot method requires numerical columns, nothing to plot."
+            )
+        if columns is None:
+            columns = data.columns
+        else:
+            data = data[columns]
+
+        result = plot_group(columns, data.values.T, ax, **kwds)
+        ax.grid(grid)
+
+    return result
+
+
+def boxplot_frame(
+    self,
+    column=None,
+    by=None,
+    ax=None,
+    fontsize: int | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    return_type=None,
+    **kwds,
+):
+    import matplotlib.pyplot as plt
+
+    ax = boxplot(
+        self,
+        column=column,
+        by=by,
+        ax=ax,
+        fontsize=fontsize,
+        grid=grid,
+        rot=rot,
+        figsize=figsize,
+        layout=layout,
+        return_type=return_type,
+        **kwds,
+    )
+    plt.draw_if_interactive()
+    return ax
+
+
+def boxplot_frame_groupby(
+    grouped,
+    subplots: bool = True,
+    column=None,
+    fontsize: int | None = None,
+    rot: int = 0,
+    grid: bool = True,
+    ax=None,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    sharex: bool = False,
+    sharey: bool = True,
+    **kwds,
+):
+    if subplots is True:
+        naxes = len(grouped)
+        fig, axes = create_subplots(
+            naxes=naxes,
+            squeeze=False,
+            ax=ax,
+            sharex=sharex,
+            sharey=sharey,
+            figsize=figsize,
+            layout=layout,
+        )
+        data = {}
+        for (key, group), ax in zip(grouped, flatten_axes(axes)):
+            d = group.boxplot(
+                ax=ax, column=column, fontsize=fontsize, rot=rot, grid=grid, **kwds
+            )
+            ax.set_title(pprint_thing(key))
+            data[key] = d
+        ret = pd.Series(data)
+        maybe_adjust_figure(fig, bottom=0.15, top=0.9, left=0.1, right=0.9, wspace=0.2)
+    else:
+        keys, frames = zip(*grouped)
+        df = pd.concat(frames, keys=keys, axis=1)
+
+        # GH 16748, DataFrameGroupby fails when subplots=False and `column` argument
+        # is assigned, and in this case, since `df` here becomes MI after groupby,
+        # so we need to couple the keys (grouped values) and column (original df
+        # column) together to search for subset to plot
+        if column is not None:
+            column = com.convert_to_list_like(column)
+            multi_key = pd.MultiIndex.from_product([keys, column])
+            column = list(multi_key.values)
+        ret = df.boxplot(
+            column=column,
+            fontsize=fontsize,
+            rot=rot,
+            grid=grid,
+            ax=ax,
+            figsize=figsize,
+            layout=layout,
+            **kwds,
+        )
+    return ret
diff --git a/pandas/pandas/plotting/_matplotlib/converter.py b/pandas/pandas/plotting/_matplotlib/converter.py
new file mode 100644
index 0000000000000..4c00049075d03
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/converter.py
@@ -0,0 +1,1111 @@
+from __future__ import annotations
+
+import contextlib
+import datetime as pydt
+from datetime import (
+    datetime,
+    tzinfo,
+)
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+import matplotlib as mpl
+import matplotlib.dates as mdates
+import matplotlib.units as munits
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    Timestamp,
+    to_offset,
+)
+from pandas._libs.tslibs.dtypes import (
+    FreqGroup,
+    periods_per_day,
+)
+from pandas._typing import (
+    F,
+    npt,
+)
+
+from pandas.core.dtypes.common import (
+    is_float,
+    is_float_dtype,
+    is_integer,
+    is_integer_dtype,
+    is_nested_list_like,
+)
+
+from pandas import (
+    Index,
+    Series,
+    get_option,
+)
+import pandas.core.common as com
+from pandas.core.indexes.datetimes import date_range
+from pandas.core.indexes.period import (
+    Period,
+    PeriodIndex,
+    period_range,
+)
+import pandas.core.tools.datetimes as tools
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+    from matplotlib.axis import Axis
+
+    from pandas._libs.tslibs.offsets import BaseOffset
+
+
+_mpl_units = {}  # Cache for units overwritten by us
+
+
+def get_pairs():
+    pairs = [
+        (Timestamp, DatetimeConverter),
+        (Period, PeriodConverter),
+        (pydt.datetime, DatetimeConverter),
+        (pydt.date, DatetimeConverter),
+        (pydt.time, TimeConverter),
+        (np.datetime64, DatetimeConverter),
+    ]
+    return pairs
+
+
+def register_pandas_matplotlib_converters(func: F) -> F:
+    """
+    Decorator applying pandas_converters.
+    """
+
+    @functools.wraps(func)
+    def wrapper(*args, **kwargs):
+        with pandas_converters():
+            return func(*args, **kwargs)
+
+    return cast(F, wrapper)
+
+
+@contextlib.contextmanager
+def pandas_converters() -> Generator[None]:
+    """
+    Context manager registering pandas' converters for a plot.
+
+    See Also
+    --------
+    register_pandas_matplotlib_converters : Decorator that applies this.
+    """
+    value = get_option("plotting.matplotlib.register_converters")
+
+    if value:
+        # register for True or "auto"
+        register()
+    try:
+        yield
+    finally:
+        if value == "auto":
+            # only deregister for "auto"
+            deregister()
+
+
+def register() -> None:
+    pairs = get_pairs()
+    for type_, cls in pairs:
+        # Cache previous converter if present
+        if type_ in munits.registry and not isinstance(munits.registry[type_], cls):
+            previous = munits.registry[type_]
+            _mpl_units[type_] = previous
+        # Replace with pandas converter
+        munits.registry[type_] = cls()
+
+
+def deregister() -> None:
+    # Renamed in pandas.plotting.__init__
+    for type_, cls in get_pairs():
+        # We use type to catch our classes directly, no inheritance
+        if type(munits.registry.get(type_)) is cls:
+            munits.registry.pop(type_)
+
+    # restore the old keys
+    for unit, formatter in _mpl_units.items():
+        if type(formatter) not in {DatetimeConverter, PeriodConverter, TimeConverter}:
+            # make it idempotent by excluding ours.
+            munits.registry[unit] = formatter
+
+
+def _to_ordinalf(tm: pydt.time) -> float:
+    tot_sec = tm.hour * 3600 + tm.minute * 60 + tm.second + tm.microsecond / 10**6
+    return tot_sec
+
+
+def time2num(d):
+    if isinstance(d, str):
+        parsed = Timestamp(d)
+        return _to_ordinalf(parsed.time())
+    if isinstance(d, pydt.time):
+        return _to_ordinalf(d)
+    return d
+
+
+class TimeConverter(munits.ConversionInterface):
+    @staticmethod
+    def convert(value, unit, axis):
+        valid_types = (str, pydt.time)
+        if isinstance(value, valid_types) or is_integer(value) or is_float(value):
+            return time2num(value)
+        if isinstance(value, Index):
+            return value.map(time2num)
+        if isinstance(value, (list, tuple, np.ndarray, Index)):
+            return [time2num(x) for x in value]
+        return value
+
+    @staticmethod
+    def axisinfo(unit, axis) -> munits.AxisInfo | None:
+        if unit != "time":
+            return None
+
+        majloc = mpl.ticker.AutoLocator()  # pyright: ignore[reportAttributeAccessIssue]
+        majfmt = TimeFormatter(majloc)
+        return munits.AxisInfo(majloc=majloc, majfmt=majfmt, label="time")
+
+    @staticmethod
+    def default_units(x, axis) -> str:
+        return "time"
+
+
+# time formatter
+class TimeFormatter(mpl.ticker.Formatter):  # pyright: ignore[reportAttributeAccessIssue]
+    def __init__(self, locs) -> None:
+        self.locs = locs
+
+    def __call__(self, x, pos: int | None = 0) -> str:
+        """
+        Return the time of day as a formatted string.
+
+        Parameters
+        ----------
+        x : float
+            The time of day specified as seconds since 00:00 (midnight),
+            with up to microsecond precision.
+        pos
+            Unused
+
+        Returns
+        -------
+        str
+            A string in HH:MM:SS.mmmuuu format. Microseconds,
+            milliseconds and seconds are only displayed if non-zero.
+        """
+        fmt = "%H:%M:%S.%f"
+        s = int(x)
+        msus = round((x - s) * 10**6)
+        ms = msus // 1000
+        us = msus % 1000
+        m, s = divmod(s, 60)
+        h, m = divmod(m, 60)
+        _, h = divmod(h, 24)
+        if us != 0:
+            return pydt.time(h, m, s, msus).strftime(fmt)
+        elif ms != 0:
+            return pydt.time(h, m, s, msus).strftime(fmt)[:-3]
+        elif s != 0:
+            return pydt.time(h, m, s).strftime("%H:%M:%S")
+
+        return pydt.time(h, m).strftime("%H:%M")
+
+
+# Period Conversion
+
+
+class PeriodConverter(mdates.DateConverter):
+    @staticmethod
+    def convert(values, units, axis):
+        if is_nested_list_like(values):
+            values = [PeriodConverter._convert_1d(v, units, axis) for v in values]
+        else:
+            values = PeriodConverter._convert_1d(values, units, axis)
+        return values
+
+    @staticmethod
+    def _convert_1d(values, units, axis):
+        if not hasattr(axis, "freq"):
+            raise TypeError("Axis must have `freq` set to convert to Periods")
+        valid_types = (str, datetime, Period, pydt.date, pydt.time, np.datetime64)
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore", "Period with BDay freq is deprecated", category=FutureWarning
+            )
+            warnings.filterwarnings(
+                "ignore", r"PeriodDtype\[B\] is deprecated", category=FutureWarning
+            )
+            if (
+                isinstance(values, valid_types)
+                or is_integer(values)
+                or is_float(values)
+            ):
+                return get_datevalue(values, axis.freq)
+            elif isinstance(values, PeriodIndex):
+                return values.asfreq(axis.freq).asi8
+            elif isinstance(values, Index):
+                return values.map(lambda x: get_datevalue(x, axis.freq))
+            elif lib.infer_dtype(values, skipna=False) == "period":
+                # https://github.com/pandas-dev/pandas/issues/24304
+                # convert ndarray[period] -> PeriodIndex
+                return PeriodIndex(values, freq=axis.freq).asi8
+            elif isinstance(values, (list, tuple, np.ndarray, Index)):
+                return [get_datevalue(x, axis.freq) for x in values]
+        return values
+
+
+def get_datevalue(date, freq):
+    if isinstance(date, Period):
+        return date.asfreq(freq).ordinal
+    elif isinstance(date, (str, datetime, pydt.date, pydt.time, np.datetime64)):
+        return Period(date, freq).ordinal
+    elif (
+        is_integer(date)
+        or is_float(date)
+        or (isinstance(date, (np.ndarray, Index)) and (date.size == 1))
+    ):
+        return date
+    elif date is None:
+        return None
+    raise ValueError(f"Unrecognizable date '{date}'")
+
+
+# Datetime Conversion
+class DatetimeConverter(mdates.DateConverter):
+    @staticmethod
+    def convert(values, unit, axis):
+        # values might be a 1-d array, or a list-like of arrays.
+        if is_nested_list_like(values):
+            values = [DatetimeConverter._convert_1d(v, unit, axis) for v in values]
+        else:
+            values = DatetimeConverter._convert_1d(values, unit, axis)
+        return values
+
+    @staticmethod
+    def _convert_1d(values, unit, axis):
+        def try_parse(values):
+            try:
+                return mdates.date2num(tools.to_datetime(values))
+            except Exception:
+                return values
+
+        if isinstance(values, (datetime, pydt.date, np.datetime64, pydt.time)):
+            return mdates.date2num(values)
+        elif is_integer(values) or is_float(values):
+            return values
+        elif isinstance(values, str):
+            return try_parse(values)
+        elif isinstance(values, (list, tuple, np.ndarray, Index, Series)):
+            if isinstance(values, Series):
+                # https://github.com/matplotlib/matplotlib/issues/11391
+                # Series was skipped. Convert to DatetimeIndex to get asi8
+                values = Index(values)
+            if isinstance(values, Index):
+                values = values.values
+            if not isinstance(values, np.ndarray):
+                values = com.asarray_tuplesafe(values)
+
+            if is_integer_dtype(values) or is_float_dtype(values):
+                return values
+
+            try:
+                values = tools.to_datetime(values)
+            except Exception:
+                pass
+
+            values = mdates.date2num(values)
+
+        return values
+
+    @staticmethod
+    def axisinfo(unit: tzinfo | None, axis) -> munits.AxisInfo:
+        """
+        Return the :class:`~matplotlib.units.AxisInfo` for *unit*.
+
+        *unit* is a tzinfo instance or None.
+        The *axis* argument is required but not used.
+        """
+        tz = unit
+
+        majloc = PandasAutoDateLocator(tz=tz)
+        majfmt = PandasAutoDateFormatter(majloc, tz=tz)
+        datemin = pydt.date(2000, 1, 1)
+        datemax = pydt.date(2010, 1, 1)
+
+        return munits.AxisInfo(
+            majloc=majloc, majfmt=majfmt, label="", default_limits=(datemin, datemax)
+        )
+
+
+class PandasAutoDateFormatter(mdates.AutoDateFormatter):
+    def __init__(self, locator, tz=None, defaultfmt: str = "%Y-%m-%d") -> None:
+        mdates.AutoDateFormatter.__init__(self, locator, tz, defaultfmt)
+
+
+class PandasAutoDateLocator(mdates.AutoDateLocator):
+    def get_locator(self, dmin, dmax):
+        """Pick the best locator based on a distance."""
+        tot_sec = (dmax - dmin).total_seconds()
+
+        if abs(tot_sec) < self.minticks:
+            self._freq = -1
+            locator = MilliSecondLocator(self.tz)
+            locator.set_axis(self.axis)
+
+            # error: Item "None" of "Axis | _DummyAxis | _AxisWrapper | None"
+            # has no attribute "get_data_interval"
+            locator.axis.set_view_interval(  # type: ignore[union-attr]
+                *self.axis.get_view_interval()  # type: ignore[union-attr]
+            )
+            locator.axis.set_data_interval(  # type: ignore[union-attr]
+                *self.axis.get_data_interval()  # type: ignore[union-attr]
+            )
+            return locator
+
+        return mdates.AutoDateLocator.get_locator(self, dmin, dmax)
+
+    def _get_unit(self):
+        return MilliSecondLocator.get_unit_generic(self._freq)
+
+
+class MilliSecondLocator(mdates.DateLocator):
+    UNIT = 1.0 / (24 * 3600 * 1000)
+
+    def __init__(self, tz) -> None:
+        mdates.DateLocator.__init__(self, tz)
+        self._interval = 1.0
+
+    def _get_unit(self):
+        return self.get_unit_generic(-1)
+
+    @staticmethod
+    def get_unit_generic(freq):
+        unit = mdates.RRuleLocator.get_unit_generic(freq)
+        if unit < 0:
+            return MilliSecondLocator.UNIT
+        return unit
+
+    def __call__(self):
+        # if no data have been set, this will tank with a ValueError
+        try:
+            dmin, dmax = self.viewlim_to_dt()
+        except ValueError:
+            return []
+
+        # We need to cap at the endpoints of valid datetime
+        nmax, nmin = mdates.date2num((dmax, dmin))
+
+        num = (nmax - nmin) * 86400 * 1000
+        max_millis_ticks = 6
+        for interval in [1, 10, 50, 100, 200, 500]:
+            if num <= interval * (max_millis_ticks - 1):
+                self._interval = interval
+                break
+            # We went through the whole loop without breaking, default to 1
+            self._interval = 1000.0
+
+        estimate = (nmax - nmin) / (self._get_unit() * self._get_interval())
+
+        if estimate > self.MAXTICKS * 2:
+            raise RuntimeError(
+                "MillisecondLocator estimated to generate "
+                f"{estimate:d} ticks from {dmin} to {dmax}: exceeds Locator.MAXTICKS"
+                f"* 2 ({self.MAXTICKS * 2:d}) "
+            )
+
+        interval = self._get_interval()
+        freq = f"{interval}ms"
+        tz = self.tz.tzname(None)
+        st = dmin.replace(tzinfo=None)
+        ed = dmax.replace(tzinfo=None)
+        all_dates = date_range(start=st, end=ed, freq=freq, tz=tz).astype(object)
+
+        try:
+            if len(all_dates) > 0:
+                locs = self.raise_if_exceeds(mdates.date2num(all_dates))
+                return locs
+        except Exception:  # pragma: no cover
+            pass
+
+        lims = mdates.date2num([dmin, dmax])
+        return lims
+
+    def _get_interval(self):
+        return self._interval
+
+    def autoscale(self):
+        """
+        Set the view limits to include the data range.
+        """
+        # We need to cap at the endpoints of valid datetime
+        dmin, dmax = self.datalim_to_dt()
+
+        vmin = mdates.date2num(dmin)
+        vmax = mdates.date2num(dmax)
+
+        return self.nonsingular(vmin, vmax)
+
+
+# Fixed frequency dynamic tick locators and formatters
+
+# -------------------------------------------------------------------------
+# --- Locators ---
+# -------------------------------------------------------------------------
+
+
+def _get_default_annual_spacing(nyears) -> tuple[int, int]:
+    """
+    Returns a default spacing between consecutive ticks for annual data.
+    """
+    if nyears < 11:
+        (min_spacing, maj_spacing) = (1, 1)
+    elif nyears < 20:
+        (min_spacing, maj_spacing) = (1, 2)
+    elif nyears < 50:
+        (min_spacing, maj_spacing) = (1, 5)
+    elif nyears < 100:
+        (min_spacing, maj_spacing) = (5, 10)
+    elif nyears < 200:
+        (min_spacing, maj_spacing) = (5, 25)
+    elif nyears < 600:
+        (min_spacing, maj_spacing) = (10, 50)
+    else:
+        factor = nyears // 1000 + 1
+        (min_spacing, maj_spacing) = (factor * 20, factor * 100)
+    return (min_spacing, maj_spacing)
+
+
+def _period_break(dates: PeriodIndex, period: str) -> npt.NDArray[np.intp]:
+    """
+    Returns the indices where the given period changes.
+
+    Parameters
+    ----------
+    dates : PeriodIndex
+        Array of intervals to monitor.
+    period : str
+        Name of the period to monitor.
+    """
+    mask = _period_break_mask(dates, period)
+    return np.nonzero(mask)[0]
+
+
+def _period_break_mask(dates: PeriodIndex, period: str) -> npt.NDArray[np.bool_]:
+    current = getattr(dates, period)
+    previous = getattr(dates - 1 * dates.freq, period)
+    return current != previous
+
+
+def has_level_label(label_flags: npt.NDArray[np.intp], vmin: float) -> bool:
+    """
+    Returns true if the ``label_flags`` indicate there is at least one label
+    for this level.
+
+    if the minimum view limit is not an exact integer, then the first tick
+    label won't be shown, so we must adjust for that.
+    """
+    if label_flags.size == 0 or (
+        label_flags.size == 1 and label_flags[0] == 0 and vmin % 1 > 0.0
+    ):
+        return False
+    else:
+        return True
+
+
+def _get_periods_per_ymd(freq: BaseOffset) -> tuple[int, int, int]:
+    # error: "BaseOffset" has no attribute "_period_dtype_code"
+    dtype_code = freq._period_dtype_code  # type: ignore[attr-defined]
+    freq_group = FreqGroup.from_period_dtype_code(dtype_code)
+
+    ppd = -1  # placeholder for above-day freqs
+
+    if dtype_code >= FreqGroup.FR_HR.value:  # pyright: ignore[reportAttributeAccessIssue]
+        # error: "BaseOffset" has no attribute "_creso"
+        ppd = periods_per_day(freq._creso)  # type: ignore[attr-defined]
+        ppm = 28 * ppd
+        ppy = 365 * ppd
+    elif freq_group == FreqGroup.FR_BUS:
+        ppm = 19
+        ppy = 261
+    elif freq_group == FreqGroup.FR_DAY:
+        ppm = 28
+        ppy = 365
+    elif freq_group == FreqGroup.FR_WK:
+        ppm = 3
+        ppy = 52
+    elif freq_group == FreqGroup.FR_MTH:
+        ppm = 1
+        ppy = 12
+    elif freq_group == FreqGroup.FR_QTR:
+        ppm = -1  # placerholder
+        ppy = 4
+    elif freq_group == FreqGroup.FR_ANN:
+        ppm = -1  # placeholder
+        ppy = 1
+    else:
+        raise NotImplementedError(f"Unsupported frequency: {dtype_code}")
+
+    return ppd, ppm, ppy
+
+
+@functools.cache
+def _daily_finder(vmin: float, vmax: float, freq: BaseOffset) -> np.ndarray:
+    # error: "BaseOffset" has no attribute "_period_dtype_code"
+    dtype_code = freq._period_dtype_code  # type: ignore[attr-defined]
+
+    periodsperday, periodspermonth, periodsperyear = _get_periods_per_ymd(freq)
+
+    # save this for later usage
+    vmin_orig = vmin
+    (vmin, vmax) = (int(vmin), int(vmax))
+    span = vmax - vmin + 1
+
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore", "Period with BDay freq is deprecated", category=FutureWarning
+        )
+        warnings.filterwarnings(
+            "ignore", r"PeriodDtype\[B\] is deprecated", category=FutureWarning
+        )
+        dates_ = period_range(
+            start=Period(ordinal=vmin, freq=freq),
+            end=Period(ordinal=vmax, freq=freq),
+            freq=freq,
+        )
+
+    # Initialize the output
+    info = np.zeros(
+        span, dtype=[("val", np.int64), ("maj", bool), ("min", bool), ("fmt", "|S20")]
+    )
+    info["val"][:] = dates_.asi8
+    info["fmt"][:] = ""
+    info["maj"][[0, -1]] = True
+    # .. and set some shortcuts
+    info_maj = info["maj"]
+    info_min = info["min"]
+    info_fmt = info["fmt"]
+
+    def first_label(label_flags):
+        if (label_flags[0] == 0) and (label_flags.size > 1) and ((vmin_orig % 1) > 0.0):
+            return label_flags[1]
+        else:
+            return label_flags[0]
+
+    # Case 1. Less than a month
+    if span <= periodspermonth:
+        day_start = _period_break(dates_, "day")
+        month_start = _period_break(dates_, "month")
+        year_start = _period_break(dates_, "year")
+
+        def _hour_finder(label_interval: int, force_year_start: bool) -> None:
+            target = dates_.hour
+            mask = _period_break_mask(dates_, "hour")
+            info_maj[day_start] = True
+            info_min[mask & (target % label_interval == 0)] = True
+            info_fmt[mask & (target % label_interval == 0)] = "%H:%M"
+            info_fmt[day_start] = "%H:%M\n%d-%b"
+            info_fmt[year_start] = "%H:%M\n%d-%b\n%Y"
+            if force_year_start and not has_level_label(year_start, vmin_orig):
+                info_fmt[first_label(day_start)] = "%H:%M\n%d-%b\n%Y"
+
+        def _minute_finder(label_interval: int) -> None:
+            target = dates_.minute
+            hour_start = _period_break(dates_, "hour")
+            mask = _period_break_mask(dates_, "minute")
+            info_maj[hour_start] = True
+            info_min[mask & (target % label_interval == 0)] = True
+            info_fmt[mask & (target % label_interval == 0)] = "%H:%M"
+            info_fmt[day_start] = "%H:%M\n%d-%b"
+            info_fmt[year_start] = "%H:%M\n%d-%b\n%Y"
+
+        def _second_finder(label_interval: int) -> None:
+            target = dates_.second
+            minute_start = _period_break(dates_, "minute")
+            mask = _period_break_mask(dates_, "second")
+            info_maj[minute_start] = True
+            info_min[mask & (target % label_interval == 0)] = True
+            info_fmt[mask & (target % label_interval == 0)] = "%H:%M:%S"
+            info_fmt[day_start] = "%H:%M:%S\n%d-%b"
+            info_fmt[year_start] = "%H:%M:%S\n%d-%b\n%Y"
+
+        if span < periodsperday / 12000:
+            _second_finder(1)
+        elif span < periodsperday / 6000:
+            _second_finder(2)
+        elif span < periodsperday / 2400:
+            _second_finder(5)
+        elif span < periodsperday / 1200:
+            _second_finder(10)
+        elif span < periodsperday / 800:
+            _second_finder(15)
+        elif span < periodsperday / 400:
+            _second_finder(30)
+        elif span < periodsperday / 150:
+            _minute_finder(1)
+        elif span < periodsperday / 70:
+            _minute_finder(2)
+        elif span < periodsperday / 24:
+            _minute_finder(5)
+        elif span < periodsperday / 12:
+            _minute_finder(15)
+        elif span < periodsperday / 6:
+            _minute_finder(30)
+        elif span < periodsperday / 2.5:
+            _hour_finder(1, False)
+        elif span < periodsperday / 1.5:
+            _hour_finder(2, False)
+        elif span < periodsperday * 1.25:
+            _hour_finder(3, False)
+        elif span < periodsperday * 2.5:
+            _hour_finder(6, True)
+        elif span < periodsperday * 4:
+            _hour_finder(12, True)
+        else:
+            info_maj[month_start] = True
+            info_min[day_start] = True
+            info_fmt[day_start] = "%d"
+            info_fmt[month_start] = "%d\n%b"
+            info_fmt[year_start] = "%d\n%b\n%Y"
+            if not has_level_label(year_start, vmin_orig):
+                if not has_level_label(month_start, vmin_orig):
+                    info_fmt[first_label(day_start)] = "%d\n%b\n%Y"
+                else:
+                    info_fmt[first_label(month_start)] = "%d\n%b\n%Y"
+
+    # Case 2. Less than three months
+    elif span <= periodsperyear // 4:
+        month_start = _period_break(dates_, "month")
+        info_maj[month_start] = True
+        if dtype_code < FreqGroup.FR_HR.value:  # pyright: ignore[reportAttributeAccessIssue]
+            info["min"] = True
+        else:
+            day_start = _period_break(dates_, "day")
+            info["min"][day_start] = True
+        week_start = _period_break(dates_, "week")
+        year_start = _period_break(dates_, "year")
+        info_fmt[week_start] = "%d"
+        info_fmt[month_start] = "\n\n%b"
+        info_fmt[year_start] = "\n\n%b\n%Y"
+        if not has_level_label(year_start, vmin_orig):
+            if not has_level_label(month_start, vmin_orig):
+                info_fmt[first_label(week_start)] = "\n\n%b\n%Y"
+            else:
+                info_fmt[first_label(month_start)] = "\n\n%b\n%Y"
+    # Case 3. Less than 14 months ...............
+    elif span <= 1.15 * periodsperyear:
+        year_start = _period_break(dates_, "year")
+        month_start = _period_break(dates_, "month")
+        week_start = _period_break(dates_, "week")
+        info_maj[month_start] = True
+        info_min[week_start] = True
+        info_min[year_start] = False
+        info_min[month_start] = False
+        info_fmt[month_start] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+        if not has_level_label(year_start, vmin_orig):
+            info_fmt[first_label(month_start)] = "%b\n%Y"
+    # Case 4. Less than 2.5 years ...............
+    elif span <= 2.5 * periodsperyear:
+        year_start = _period_break(dates_, "year")
+        quarter_start = _period_break(dates_, "quarter")
+        month_start = _period_break(dates_, "month")
+        info_maj[quarter_start] = True
+        info_min[month_start] = True
+        info_fmt[quarter_start] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+    # Case 4. Less than 4 years .................
+    elif span <= 4 * periodsperyear:
+        year_start = _period_break(dates_, "year")
+        month_start = _period_break(dates_, "month")
+        info_maj[year_start] = True
+        info_min[month_start] = True
+        info_min[year_start] = False
+
+        month_break = dates_[month_start].month
+        jan_or_jul = month_start[(month_break == 1) | (month_break == 7)]
+        info_fmt[jan_or_jul] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+    # Case 5. Less than 11 years ................
+    elif span <= 11 * periodsperyear:
+        year_start = _period_break(dates_, "year")
+        quarter_start = _period_break(dates_, "quarter")
+        info_maj[year_start] = True
+        info_min[quarter_start] = True
+        info_min[year_start] = False
+        info_fmt[year_start] = "%Y"
+    # Case 6. More than 12 years ................
+    else:
+        year_start = _period_break(dates_, "year")
+        year_break = dates_[year_start].year
+        nyears = span / periodsperyear
+        (min_anndef, maj_anndef) = _get_default_annual_spacing(nyears)
+        major_idx = year_start[(year_break % maj_anndef == 0)]
+        info_maj[major_idx] = True
+        minor_idx = year_start[(year_break % min_anndef == 0)]
+        info_min[minor_idx] = True
+        info_fmt[major_idx] = "%Y"
+
+    return info
+
+
+@functools.cache
+def _monthly_finder(vmin: float, vmax: float, freq: BaseOffset) -> np.ndarray:
+    _, _, periodsperyear = _get_periods_per_ymd(freq)
+
+    vmin_orig = vmin
+    (vmin, vmax) = (int(vmin), int(vmax))
+    span = vmax - vmin + 1
+
+    # Initialize the output
+    info = np.zeros(
+        span, dtype=[("val", int), ("maj", bool), ("min", bool), ("fmt", "|S8")]
+    )
+    info["val"] = np.arange(vmin, vmax + 1)
+    dates_ = info["val"]
+    info["fmt"] = ""
+    year_start = (dates_ % 12 == 0).nonzero()[0]
+    info_maj = info["maj"]
+    info_fmt = info["fmt"]
+
+    if span <= 1.15 * periodsperyear:
+        info_maj[year_start] = True
+        info["min"] = True
+
+        info_fmt[:] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+
+        if not has_level_label(year_start, vmin_orig):
+            if dates_.size > 1:
+                idx = 1
+            else:
+                idx = 0
+            info_fmt[idx] = "%b\n%Y"
+
+    elif span <= 2.5 * periodsperyear:
+        quarter_start = (dates_ % 3 == 0).nonzero()
+        info_maj[year_start] = True
+        # TODO: Check the following : is it really info['fmt'] ?
+        #  2023-09-15 this is reached in test_finder_monthly
+        info["fmt"][quarter_start] = True
+        info["min"] = True
+
+        info_fmt[quarter_start] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+
+    elif span <= 4 * periodsperyear:
+        info_maj[year_start] = True
+        info["min"] = True
+
+        jan_or_jul = (dates_ % 12 == 0) | (dates_ % 12 == 6)
+        info_fmt[jan_or_jul] = "%b"
+        info_fmt[year_start] = "%b\n%Y"
+
+    elif span <= 11 * periodsperyear:
+        quarter_start = (dates_ % 3 == 0).nonzero()
+        info_maj[year_start] = True
+        info["min"][quarter_start] = True
+
+        info_fmt[year_start] = "%Y"
+
+    else:
+        nyears = span / periodsperyear
+        (min_anndef, maj_anndef) = _get_default_annual_spacing(nyears)
+        years = dates_[year_start] // 12 + 1
+        major_idx = year_start[(years % maj_anndef == 0)]
+        info_maj[major_idx] = True
+        info["min"][year_start[(years % min_anndef == 0)]] = True
+
+        info_fmt[major_idx] = "%Y"
+
+    return info
+
+
+@functools.cache
+def _quarterly_finder(vmin: float, vmax: float, freq: BaseOffset) -> np.ndarray:
+    _, _, periodsperyear = _get_periods_per_ymd(freq)
+    vmin_orig = vmin
+    (vmin, vmax) = (int(vmin), int(vmax))
+    span = vmax - vmin + 1
+
+    info = np.zeros(
+        span, dtype=[("val", int), ("maj", bool), ("min", bool), ("fmt", "|S8")]
+    )
+    info["val"] = np.arange(vmin, vmax + 1)
+    info["fmt"] = ""
+    dates_ = info["val"]
+    info_maj = info["maj"]
+    info_fmt = info["fmt"]
+    year_start = (dates_ % 4 == 0).nonzero()[0]
+
+    if span <= 3.5 * periodsperyear:
+        info_maj[year_start] = True
+        info["min"] = True
+
+        info_fmt[:] = "Q%q"
+        info_fmt[year_start] = "Q%q\n%F"
+        if not has_level_label(year_start, vmin_orig):
+            if dates_.size > 1:
+                idx = 1
+            else:
+                idx = 0
+            info_fmt[idx] = "Q%q\n%F"
+
+    elif span <= 11 * periodsperyear:
+        info_maj[year_start] = True
+        info["min"] = True
+        info_fmt[year_start] = "%F"
+
+    else:
+        # https://github.com/pandas-dev/pandas/pull/47602
+        years = dates_[year_start] // 4 + 1970
+        nyears = span / periodsperyear
+        (min_anndef, maj_anndef) = _get_default_annual_spacing(nyears)
+        major_idx = year_start[(years % maj_anndef == 0)]
+        info_maj[major_idx] = True
+        info["min"][year_start[(years % min_anndef == 0)]] = True
+        info_fmt[major_idx] = "%F"
+
+    return info
+
+
+@functools.cache
+def _annual_finder(vmin: float, vmax: float, freq: BaseOffset) -> np.ndarray:
+    # Note: small difference here vs other finders in adding 1 to vmax
+    (vmin, vmax) = (int(vmin), int(vmax + 1))
+    span = vmax - vmin + 1
+
+    info = np.zeros(
+        span, dtype=[("val", int), ("maj", bool), ("min", bool), ("fmt", "|S8")]
+    )
+    info["val"] = np.arange(vmin, vmax + 1)
+    info["fmt"] = ""
+    dates_ = info["val"]
+
+    (min_anndef, maj_anndef) = _get_default_annual_spacing(span)
+    major_idx = dates_ % maj_anndef == 0
+    minor_idx = dates_ % min_anndef == 0
+    info["maj"][major_idx] = True
+    info["min"][minor_idx] = True
+    info["fmt"][major_idx] = "%Y"
+
+    return info
+
+
+def get_finder(freq: BaseOffset):
+    # error: "BaseOffset" has no attribute "_period_dtype_code"
+    dtype_code = freq._period_dtype_code  # type: ignore[attr-defined]
+    fgroup = FreqGroup.from_period_dtype_code(dtype_code)
+
+    if fgroup == FreqGroup.FR_ANN:
+        return _annual_finder
+    elif fgroup == FreqGroup.FR_QTR:
+        return _quarterly_finder
+    elif fgroup == FreqGroup.FR_MTH:
+        return _monthly_finder
+    elif (dtype_code >= FreqGroup.FR_BUS.value) or fgroup == FreqGroup.FR_WK:  # pyright: ignore[reportAttributeAccessIssue]
+        return _daily_finder
+    else:  # pragma: no cover
+        raise NotImplementedError(f"Unsupported frequency: {dtype_code}")
+
+
+class TimeSeries_DateLocator(mpl.ticker.Locator):  # pyright: ignore[reportAttributeAccessIssue]
+    """
+    Locates the ticks along an axis controlled by a :class:`Series`.
+
+    Parameters
+    ----------
+    freq : BaseOffset
+        Valid frequency specifier.
+    minor_locator : {False, True}, optional
+        Whether the locator is for minor ticks (True) or not.
+    dynamic_mode : {True, False}, optional
+        Whether the locator should work in dynamic mode.
+    base : {int}, optional
+    quarter : {int}, optional
+    month : {int}, optional
+    day : {int}, optional
+    """
+
+    axis: Axis
+
+    def __init__(
+        self,
+        freq: BaseOffset,
+        minor_locator: bool = False,
+        dynamic_mode: bool = True,
+        base: int = 1,
+        quarter: int = 1,
+        month: int = 1,
+        day: int = 1,
+        plot_obj=None,
+    ) -> None:
+        freq = to_offset(freq, is_period=True)
+        self.freq = freq
+        self.base = base
+        (self.quarter, self.month, self.day) = (quarter, month, day)
+        self.isminor = minor_locator
+        self.isdynamic = dynamic_mode
+        self.offset = 0
+        self.plot_obj = plot_obj
+        self.finder = get_finder(freq)
+
+    def _get_default_locs(self, vmin, vmax):
+        """Returns the default locations of ticks."""
+        locator = self.finder(vmin, vmax, self.freq)
+
+        if self.isminor:
+            return np.compress(locator["min"], locator["val"])
+        return np.compress(locator["maj"], locator["val"])
+
+    def __call__(self):
+        """Return the locations of the ticks."""
+        # axis calls Locator.set_axis inside set_m<xxxx>_formatter
+
+        vi = tuple(self.axis.get_view_interval())
+        vmin, vmax = vi
+        if vmax < vmin:
+            vmin, vmax = vmax, vmin
+        if self.isdynamic:
+            locs = self._get_default_locs(vmin, vmax)
+        else:  # pragma: no cover
+            base = self.base
+            (d, m) = divmod(vmin, base)
+            vmin = (d + 1) * base
+            # error: No overload variant of "range" matches argument types "float",
+            # "float", "int"
+            locs = list(range(vmin, vmax + 1, base))  # type: ignore[call-overload]
+        return locs
+
+    def autoscale(self):
+        """
+        Sets the view limits to the nearest multiples of base that contain the
+        data.
+        """
+        # requires matplotlib >= 0.98.0
+        (vmin, vmax) = self.axis.get_data_interval()
+
+        locs = self._get_default_locs(vmin, vmax)
+        (vmin, vmax) = locs[[0, -1]]
+        if vmin == vmax:
+            vmin -= 1
+            vmax += 1
+        return mpl.transforms.nonsingular(vmin, vmax)
+
+
+# -------------------------------------------------------------------------
+# --- Formatter ---
+# -------------------------------------------------------------------------
+
+
+class TimeSeries_DateFormatter(mpl.ticker.Formatter):  # pyright: ignore[reportAttributeAccessIssue]
+    """
+    Formats the ticks along an axis controlled by a :class:`PeriodIndex`.
+
+    Parameters
+    ----------
+    freq : BaseOffset
+        Valid frequency specifier.
+    minor_locator : bool, default False
+        Whether the current formatter should apply to minor ticks (True) or
+        major ticks (False).
+    dynamic_mode : bool, default True
+        Whether the formatter works in dynamic mode or not.
+    """
+
+    axis: Axis
+
+    def __init__(
+        self,
+        freq: BaseOffset,
+        minor_locator: bool = False,
+        dynamic_mode: bool = True,
+        plot_obj=None,
+    ) -> None:
+        freq = to_offset(freq, is_period=True)
+        self.format = None
+        self.freq = freq
+        self.locs: list[Any] = []  # unused, for matplotlib compat
+        self.formatdict: dict[Any, Any] | None = None
+        self.isminor = minor_locator
+        self.isdynamic = dynamic_mode
+        self.offset = 0
+        self.plot_obj = plot_obj
+        self.finder = get_finder(freq)
+
+    def _set_default_format(self, vmin, vmax):
+        """Returns the default ticks spacing."""
+        info = self.finder(vmin, vmax, self.freq)
+
+        if self.isminor:
+            format = np.compress(info["min"] & np.logical_not(info["maj"]), info)
+        else:
+            format = np.compress(info["maj"], info)
+        self.formatdict = {x: f for (x, _, _, f) in format}
+        return self.formatdict
+
+    def set_locs(self, locs) -> None:
+        """Sets the locations of the ticks"""
+        # don't actually use the locs. This is just needed to work with
+        # matplotlib. Force to use vmin, vmax
+
+        self.locs = locs
+
+        (vmin, vmax) = tuple(self.axis.get_view_interval())
+        if vmax < vmin:
+            (vmin, vmax) = (vmax, vmin)
+        self._set_default_format(vmin, vmax)
+
+    def __call__(self, x, pos: int | None = 0) -> str:
+        if self.formatdict is None:
+            return ""
+        else:
+            fmt = self.formatdict.pop(x, "")
+            if isinstance(fmt, np.bytes_):
+                fmt = fmt.decode("utf-8")
+            with warnings.catch_warnings():
+                warnings.filterwarnings(
+                    "ignore",
+                    "Period with BDay freq is deprecated",
+                    category=FutureWarning,
+                )
+                period = Period(ordinal=int(x), freq=self.freq)
+            assert isinstance(period, Period)
+            return period.strftime(fmt)
+
+
+class TimeSeries_TimedeltaFormatter(mpl.ticker.Formatter):  # pyright: ignore[reportAttributeAccessIssue]
+    """
+    Formats the ticks along an axis controlled by a :class:`TimedeltaIndex`.
+    """
+
+    axis: Axis
+
+    @staticmethod
+    def format_timedelta_ticks(x, pos, n_decimals: int) -> str:
+        """
+        Convert seconds to 'D days HH:MM:SS.F'
+        """
+        s, ns = divmod(x, 10**9)  # TODO(non-nano): this looks like it assumes ns
+        m, s = divmod(s, 60)
+        h, m = divmod(m, 60)
+        d, h = divmod(h, 24)
+        decimals = int(ns * 10 ** (n_decimals - 9))
+        s = f"{int(h):02d}:{int(m):02d}:{int(s):02d}"
+        if n_decimals > 0:
+            s += f".{decimals:0{n_decimals}d}"
+        if d != 0:
+            s = f"{int(d):d} days {s}"
+        return s
+
+    def __call__(self, x, pos: int | None = 0) -> str:
+        (vmin, vmax) = tuple(self.axis.get_view_interval())
+        n_decimals = min(int(np.ceil(np.log10(100 * 10**9 / abs(vmax - vmin)))), 9)
+        return self.format_timedelta_ticks(x, pos, n_decimals)
diff --git a/pandas/pandas/plotting/_matplotlib/core.py b/pandas/pandas/plotting/_matplotlib/core.py
new file mode 100644
index 0000000000000..1035150302d2c
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/core.py
@@ -0,0 +1,2153 @@
+from __future__ import annotations
+
+from abc import (
+    ABC,
+    abstractmethod,
+)
+from collections.abc import (
+    Hashable,
+    Iterable,
+    Iterator,
+    Sequence,
+)
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    cast,
+    final,
+)
+import warnings
+
+import matplotlib as mpl
+import numpy as np
+
+from pandas._libs import lib
+from pandas.errors import AbstractMethodError
+from pandas.util._decorators import cache_readonly
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import (
+    is_any_real_numeric_dtype,
+    is_bool,
+    is_float,
+    is_float_dtype,
+    is_hashable,
+    is_integer,
+    is_integer_dtype,
+    is_iterator,
+    is_list_like,
+    is_number,
+    is_numeric_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    ExtensionDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCDatetimeIndex,
+    ABCIndex,
+    ABCMultiIndex,
+    ABCPeriodIndex,
+    ABCSeries,
+)
+from pandas.core.dtypes.missing import isna
+
+import pandas.core.common as com
+from pandas.util.version import Version
+
+from pandas.io.formats.printing import pprint_thing
+from pandas.plotting._matplotlib import tools
+from pandas.plotting._matplotlib.converter import register_pandas_matplotlib_converters
+from pandas.plotting._matplotlib.groupby import reconstruct_data_with_by
+from pandas.plotting._matplotlib.misc import unpack_single_str_list
+from pandas.plotting._matplotlib.style import get_standard_colors
+from pandas.plotting._matplotlib.timeseries import (
+    decorate_axes,
+    format_dateaxis,
+    maybe_convert_index,
+    maybe_resample,
+    use_dynamic_x,
+)
+from pandas.plotting._matplotlib.tools import (
+    create_subplots,
+    flatten_axes,
+    format_date_labels,
+    get_all_lines,
+    get_xlim,
+    handle_shared_axes,
+)
+
+if TYPE_CHECKING:
+    from matplotlib.artist import Artist
+    from matplotlib.axes import Axes
+    from matplotlib.axis import Axis
+    from matplotlib.figure import Figure
+
+    from pandas._typing import (
+        IndexLabel,
+        NDFrameT,
+        PlottingOrientation,
+        npt,
+    )
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+
+def holds_integer(column: Index) -> bool:
+    return column.inferred_type in {"integer", "mixed-integer"}
+
+
+def _color_in_style(style: str) -> bool:
+    """
+    Check if there is a color letter in the style string.
+    """
+    return not set(mpl.colors.BASE_COLORS).isdisjoint(style)
+
+
+class MPLPlot(ABC):
+    """
+    Base class for assembling a pandas plot using matplotlib
+
+    Parameters
+    ----------
+    data :
+
+    """
+
+    @property
+    @abstractmethod
+    def _kind(self) -> str:
+        """Specify kind str. Must be overridden in child class"""
+        raise NotImplementedError
+
+    _layout_type = "vertical"
+    _default_rot = 0
+
+    @property
+    def orientation(self) -> str | None:
+        return None
+
+    data: DataFrame
+
+    def __init__(
+        self,
+        data,
+        kind=None,
+        by: IndexLabel | None = None,
+        subplots: bool | Sequence[Sequence[str]] = False,
+        sharex: bool | None = None,
+        sharey: bool = False,
+        use_index: bool = True,
+        figsize: tuple[float, float] | None = None,
+        grid=None,
+        legend: bool | str = True,
+        rot=None,
+        ax=None,
+        fig=None,
+        title=None,
+        xlim=None,
+        ylim=None,
+        xticks=None,
+        yticks=None,
+        xlabel: Hashable | None = None,
+        ylabel: Hashable | None = None,
+        fontsize: int | None = None,
+        secondary_y: bool | tuple | list | np.ndarray = False,
+        colormap=None,
+        table: bool = False,
+        layout=None,
+        include_bool: bool = False,
+        column: IndexLabel | None = None,
+        *,
+        logx: bool | None | Literal["sym"] = False,
+        logy: bool | None | Literal["sym"] = False,
+        loglog: bool | None | Literal["sym"] = False,
+        mark_right: bool = True,
+        stacked: bool = False,
+        label: Hashable | None = None,
+        style=None,
+        **kwds,
+    ) -> None:
+        # if users assign an empty list or tuple, raise `ValueError`
+        # similar to current `df.box` and `df.hist` APIs.
+        if by in ([], ()):
+            raise ValueError("No group keys passed!")
+        self.by = com.maybe_make_list(by)
+
+        # Assign the rest of columns into self.columns if by is explicitly defined
+        # while column is not, only need `columns` in hist/box plot when it's DF
+        # TODO: Might deprecate `column` argument in future PR (#28373)
+        if isinstance(data, ABCDataFrame):
+            if column:
+                self.columns = com.maybe_make_list(column)
+            elif self.by is None:
+                self.columns = [
+                    col for col in data.columns if is_numeric_dtype(data[col])
+                ]
+            else:
+                self.columns = [
+                    col
+                    for col in data.columns
+                    if col not in self.by and is_numeric_dtype(data[col])
+                ]
+
+        # For `hist` plot, need to get grouped original data before `self.data` is
+        # updated later
+        if self.by is not None and self._kind == "hist":
+            self._grouped = data.groupby(unpack_single_str_list(self.by))
+
+        self.kind = kind
+
+        self.subplots = type(self)._validate_subplots_kwarg(
+            subplots, data, kind=self._kind
+        )
+
+        self.sharex = type(self)._validate_sharex(sharex, ax, by)
+        self.sharey = sharey
+        self.figsize = figsize
+        self.layout = layout
+
+        self.xticks = xticks
+        self.yticks = yticks
+        self.xlim = xlim
+        self.ylim = ylim
+        self.title = title
+        self.use_index = use_index
+        self.xlabel = xlabel
+        self.ylabel = ylabel
+
+        self.fontsize = fontsize
+
+        if rot is not None:
+            self.rot = rot
+            # need to know for format_date_labels since it's rotated to 30 by
+            # default
+            self._rot_set = True
+        else:
+            self._rot_set = False
+            self.rot = self._default_rot
+
+        if grid is None:
+            grid = False if secondary_y else mpl.rcParams["axes.grid"]
+
+        self.grid = grid
+        self.legend = legend
+        self.legend_handles: list[Artist] = []
+        self.legend_labels: list[Hashable] = []
+
+        self.logx = type(self)._validate_log_kwd("logx", logx)
+        self.logy = type(self)._validate_log_kwd("logy", logy)
+        self.loglog = type(self)._validate_log_kwd("loglog", loglog)
+        self.label = label
+        self.style = style
+        self.mark_right = mark_right
+        self.stacked = stacked
+
+        # ax may be an Axes object or (if self.subplots) an ndarray of
+        #  Axes objects
+        self.ax = ax
+        # TODO: deprecate fig keyword as it is ignored, not passed in tests
+        #  as of 2023-11-05
+
+        # parse errorbar input if given
+        xerr = kwds.pop("xerr", None)
+        yerr = kwds.pop("yerr", None)
+        nseries = self._get_nseries(data)
+        xerr, data = type(self)._parse_errorbars("xerr", xerr, data, nseries)
+        yerr, data = type(self)._parse_errorbars("yerr", yerr, data, nseries)
+        self.errors = {"xerr": xerr, "yerr": yerr}
+        self.data = data
+
+        if not isinstance(secondary_y, (bool, tuple, list, np.ndarray, ABCIndex)):
+            secondary_y = [secondary_y]
+        self.secondary_y = secondary_y
+
+        # ugly TypeError if user passes matplotlib's `cmap` name.
+        # Probably better to accept either.
+        if "cmap" in kwds and colormap:
+            raise TypeError("Only specify one of `cmap` and `colormap`.")
+        if "cmap" in kwds:
+            self.colormap = kwds.pop("cmap")
+        else:
+            self.colormap = colormap
+
+        self.table = table
+        self.include_bool = include_bool
+
+        self.kwds = kwds
+
+        color = kwds.pop("color", lib.no_default)
+        self.color = self._validate_color_args(color, self.colormap)
+        assert "color" not in self.kwds
+
+        self.data = self._ensure_frame(self.data)
+
+    @final
+    @staticmethod
+    def _validate_sharex(sharex: bool | None, ax, by) -> bool:
+        if sharex is None:
+            # if by is defined, subplots are used and sharex should be False
+            if ax is None and by is None:
+                sharex = True
+            else:
+                # if we get an axis, the users should do the visibility
+                # setting...
+                sharex = False
+        elif not is_bool(sharex):
+            raise TypeError("sharex must be a bool or None")
+        return bool(sharex)
+
+    @classmethod
+    def _validate_log_kwd(
+        cls,
+        kwd: str,
+        value: bool | None | Literal["sym"],
+    ) -> bool | None | Literal["sym"]:
+        if (
+            value is None
+            or isinstance(value, bool)
+            or (isinstance(value, str) and value == "sym")
+        ):
+            return value
+        raise ValueError(
+            f"keyword '{kwd}' should be bool, None, or 'sym', not '{value}'"
+        )
+
+    @final
+    @staticmethod
+    def _validate_subplots_kwarg(
+        subplots: bool | Sequence[Sequence[str]], data: Series | DataFrame, kind: str
+    ) -> bool | list[tuple[int, ...]]:
+        """
+        Validate the subplots parameter
+
+        - check type and content
+        - check for duplicate columns
+        - check for invalid column names
+        - convert column names into indices
+        - add missing columns in a group of their own
+        See comments in code below for more details.
+
+        Parameters
+        ----------
+        subplots : subplots parameters as passed to PlotAccessor
+
+        Returns
+        -------
+        validated subplots : a bool or a list of tuples of column indices. Columns
+        in the same tuple will be grouped together in the resulting plot.
+        """
+
+        if isinstance(subplots, bool):
+            return subplots
+        elif not isinstance(subplots, Iterable):
+            raise ValueError("subplots should be a bool or an iterable")
+
+        supported_kinds = (
+            "line",
+            "bar",
+            "barh",
+            "hist",
+            "kde",
+            "density",
+            "area",
+            "pie",
+        )
+        if kind not in supported_kinds:
+            raise ValueError(
+                "When subplots is an iterable, kind must be "
+                f"one of {', '.join(supported_kinds)}. Got {kind}."
+            )
+
+        if isinstance(data, ABCSeries):
+            raise NotImplementedError(
+                "An iterable subplots for a Series is not supported."
+            )
+
+        columns = data.columns
+        if isinstance(columns, ABCMultiIndex):
+            raise NotImplementedError(
+                "An iterable subplots for a DataFrame with a MultiIndex column "
+                "is not supported."
+            )
+
+        if columns.nunique() != len(columns):
+            raise NotImplementedError(
+                "An iterable subplots for a DataFrame with non-unique column "
+                "labels is not supported."
+            )
+
+        # subplots is a list of tuples where each tuple is a group of
+        # columns to be grouped together (one ax per group).
+        # we consolidate the subplots list such that:
+        # - the tuples contain indices instead of column names
+        # - the columns that aren't yet in the list are added in a group
+        #   of their own.
+        # For example with columns from a to g, and
+        # subplots = [(a, c), (b, f, e)],
+        # we end up with [(ai, ci), (bi, fi, ei), (di,), (gi,)]
+        # This way, we can handle self.subplots in a homogeneous manner
+        # later.
+        # TODO: also accept indices instead of just names?
+
+        out = []
+        seen_columns: set[Hashable] = set()
+        for group in subplots:
+            if not is_list_like(group):
+                raise ValueError(
+                    "When subplots is an iterable, each entry "
+                    "should be a list/tuple of column names."
+                )
+            idx_locs = columns.get_indexer_for(group)
+            if (idx_locs == -1).any():
+                bad_labels = np.extract(idx_locs == -1, group)
+                raise ValueError(
+                    f"Column label(s) {list(bad_labels)} not found in the DataFrame."
+                )
+            unique_columns = set(group)
+            duplicates = seen_columns.intersection(unique_columns)
+            if duplicates:
+                raise ValueError(
+                    "Each column should be in only one subplot. "
+                    f"Columns {duplicates} were found in multiple subplots."
+                )
+            seen_columns = seen_columns.union(unique_columns)
+            out.append(tuple(idx_locs))
+
+        unseen_columns = columns.difference(seen_columns)
+        for column in unseen_columns:
+            idx_loc = columns.get_loc(column)
+            out.append((idx_loc,))
+        return out
+
+    def _validate_color_args(self, color, colormap):
+        if color is lib.no_default:
+            # It was not provided by the user
+            if "colors" in self.kwds and colormap is not None:
+                warnings.warn(
+                    "'color' and 'colormap' cannot be used simultaneously. "
+                    "Using 'color'",
+                    stacklevel=find_stack_level(),
+                )
+            return None
+        if self.nseries == 1 and color is not None and not is_list_like(color):
+            # support series.plot(color='green')
+            color = [color]
+
+        if isinstance(color, tuple) and self.nseries == 1 and len(color) in (3, 4):
+            # support RGB and RGBA tuples in series plot
+            color = [color]
+
+        if colormap is not None:
+            warnings.warn(
+                "'color' and 'colormap' cannot be used simultaneously. Using 'color'",
+                stacklevel=find_stack_level(),
+            )
+
+        if self.style is not None:
+            if isinstance(self.style, dict):
+                styles = [self.style[col] for col in self.columns if col in self.style]
+            elif is_list_like(self.style):
+                styles = self.style
+            else:
+                styles = [self.style]
+            # need only a single match
+            for s in styles:
+                if _color_in_style(s):
+                    raise ValueError(
+                        "Cannot pass 'style' string with a color symbol and "
+                        "'color' keyword argument. Please use one or the "
+                        "other or pass 'style' without a color symbol"
+                    )
+        return color
+
+    @final
+    @staticmethod
+    def _iter_data(
+        data: DataFrame | dict[Hashable, Series | DataFrame],
+    ) -> Iterator[tuple[Hashable, np.ndarray]]:
+        for col, values in data.items():
+            # This was originally written to use values.values before EAs
+            #  were implemented; adding np.asarray(...) to keep consistent
+            #  typing.
+            yield col, np.asarray(values.values)
+
+    def _get_nseries(self, data: Series | DataFrame) -> int:
+        # When `by` is explicitly assigned, grouped data size will be defined, and
+        # this will determine number of subplots to have, aka `self.nseries`
+        if data.ndim == 1:
+            return 1
+        elif self.by is not None and self._kind == "hist":
+            return len(self._grouped)
+        elif self.by is not None and self._kind == "box":
+            return len(self.columns)
+        else:
+            return data.shape[1]
+
+    @final
+    @property
+    def nseries(self) -> int:
+        return self._get_nseries(self.data)
+
+    @final
+    def generate(self) -> None:
+        self._compute_plot_data()
+        fig = self.fig
+        self._make_plot(fig)
+        self._add_table()
+        self._make_legend()
+        self._adorn_subplots(fig)
+
+        for ax in self.axes:
+            self._post_plot_logic_common(ax)
+            self._post_plot_logic(ax, self.data)
+
+    @final
+    @staticmethod
+    def _has_plotted_object(ax: Axes) -> bool:
+        """check whether ax has data"""
+        return len(ax.lines) != 0 or len(ax.artists) != 0 or len(ax.containers) != 0
+
+    @final
+    def _maybe_right_yaxis(self, ax: Axes, axes_num: int) -> Axes:
+        if not self.on_right(axes_num):
+            # secondary axes may be passed via ax kw
+            return self._get_ax_layer(ax)
+
+        if hasattr(ax, "right_ax"):
+            # if it has right_ax property, ``ax`` must be left axes
+            return ax.right_ax
+        elif hasattr(ax, "left_ax"):
+            # if it has left_ax property, ``ax`` must be right axes
+            return ax
+        else:
+            # otherwise, create twin axes
+            orig_ax, new_ax = ax, ax.twinx()
+            # TODO: use Matplotlib public API when available
+            new_ax._get_lines = orig_ax._get_lines  # type: ignore[attr-defined]
+            # TODO #54485
+            new_ax._get_patches_for_fill = (  # type: ignore[attr-defined]
+                orig_ax._get_patches_for_fill  # type: ignore[attr-defined]
+            )
+            # TODO #54485
+            orig_ax.right_ax, new_ax.left_ax = (  # type: ignore[attr-defined]
+                new_ax,
+                orig_ax,
+            )
+
+            if not self._has_plotted_object(orig_ax):  # no data on left y
+                orig_ax.get_yaxis().set_visible(False)
+
+            if self.logy is True or self.loglog is True:
+                new_ax.set_yscale("log")
+            elif self.logy == "sym" or self.loglog == "sym":
+                new_ax.set_yscale("symlog")
+            return new_ax
+
+    @final
+    @cache_readonly
+    def fig(self) -> Figure:
+        return self._axes_and_fig[1]
+
+    @final
+    @cache_readonly
+    # TODO: can we annotate this as both a Sequence[Axes] and ndarray[object]?
+    def axes(self) -> Sequence[Axes]:
+        return self._axes_and_fig[0]
+
+    @final
+    @cache_readonly
+    def _axes_and_fig(self) -> tuple[Sequence[Axes], Figure]:
+        import matplotlib.pyplot as plt
+
+        if self.subplots:
+            naxes = (
+                self.nseries if isinstance(self.subplots, bool) else len(self.subplots)
+            )
+            fig, axes = create_subplots(
+                naxes=naxes,
+                sharex=self.sharex,
+                sharey=self.sharey,
+                figsize=self.figsize,
+                ax=self.ax,
+                layout=self.layout,
+                layout_type=self._layout_type,
+            )
+        elif self.ax is None:
+            fig = plt.figure(figsize=self.figsize)
+            axes = fig.add_subplot(111)
+        else:
+            fig = self.ax.get_figure()
+            if self.figsize is not None:
+                fig.set_size_inches(self.figsize)
+            axes = self.ax
+
+        axes = np.fromiter(flatten_axes(axes), dtype=object)
+
+        if self.logx is True or self.loglog is True:
+            [a.set_xscale("log") for a in axes]
+        elif self.logx == "sym" or self.loglog == "sym":
+            [a.set_xscale("symlog") for a in axes]
+
+        if self.logy is True or self.loglog is True:
+            [a.set_yscale("log") for a in axes]
+        elif self.logy == "sym" or self.loglog == "sym":
+            [a.set_yscale("symlog") for a in axes]
+
+        axes_seq = cast(Sequence["Axes"], axes)
+        return axes_seq, fig
+
+    @property
+    def result(self):
+        """
+        Return result axes
+        """
+        if self.subplots:
+            if self.layout is not None and not is_list_like(self.ax):
+                # error: "Sequence[Any]" has no attribute "reshape"
+                return self.axes.reshape(*self.layout)  # type: ignore[attr-defined]
+            else:
+                return self.axes
+        else:
+            sec_true = isinstance(self.secondary_y, bool) and self.secondary_y
+            # error: Argument 1 to "len" has incompatible type "Union[bool,
+            # Tuple[Any, ...], List[Any], ndarray[Any, Any]]"; expected "Sized"
+            all_sec = (
+                is_list_like(self.secondary_y) and len(self.secondary_y) == self.nseries  # type: ignore[arg-type]
+            )
+            if sec_true or all_sec:
+                # if all data is plotted on secondary, return right axes
+                return self._get_ax_layer(self.axes[0], primary=False)
+            else:
+                return self.axes[0]
+
+    @final
+    @staticmethod
+    def _convert_to_ndarray(data):
+        # GH31357: categorical columns are processed separately
+        if isinstance(data.dtype, CategoricalDtype):
+            return data
+
+        # GH32073: cast to float if values contain nulled integers
+        if (is_integer_dtype(data.dtype) or is_float_dtype(data.dtype)) and isinstance(
+            data.dtype, ExtensionDtype
+        ):
+            return data.to_numpy(dtype="float", na_value=np.nan)
+
+        # GH25587: cast ExtensionArray of pandas (IntegerArray, etc.) to
+        # np.ndarray before plot.
+        if len(data) > 0:
+            return np.asarray(data)
+
+        return data
+
+    @final
+    def _ensure_frame(self, data) -> DataFrame:
+        if isinstance(data, ABCSeries):
+            label = self.label
+            if label is None and data.name is None:
+                label = ""
+            if label is None:
+                # We'll end up with columns of [0] instead of [None]
+                data = data.to_frame()
+            else:
+                data = data.to_frame(name=label)
+        elif self._kind in ("hist", "box"):
+            cols = self.columns if self.by is None else self.columns + self.by
+            data = data.loc[:, cols]
+        return data
+
+    @final
+    def _compute_plot_data(self) -> None:
+        data = self.data
+
+        # GH15079 reconstruct data if by is defined
+        if self.by is not None:
+            self.subplots = True
+            data = reconstruct_data_with_by(self.data, by=self.by, cols=self.columns)
+
+        # GH16953, infer_objects is needed as fallback, for ``Series``
+        # with ``dtype == object``
+        data = data.infer_objects()
+        include_type = [np.number, "datetime", "datetimetz", "timedelta"]
+
+        # GH23719, allow plotting boolean
+        if self.include_bool is True:
+            include_type.append(np.bool_)
+
+        # GH22799, exclude datetime-like type for boxplot
+        exclude_type = None
+        if self._kind == "box":
+            # TODO: change after solving issue 27881
+            include_type = [np.number]
+            exclude_type = ["timedelta"]
+
+        # GH 18755, include object and category type for scatter plot
+        if self._kind == "scatter":
+            include_type.extend(["object", "category", "string"])
+
+        numeric_data = data.select_dtypes(include=include_type, exclude=exclude_type)
+
+        is_empty = numeric_data.shape[-1] == 0
+        # no non-numeric frames or series allowed
+        if is_empty:
+            raise TypeError("no numeric data to plot")
+
+        self.data = numeric_data.apply(type(self)._convert_to_ndarray)
+
+    def _make_plot(self, fig: Figure) -> None:
+        raise AbstractMethodError(self)
+
+    @final
+    def _add_table(self) -> None:
+        if self.table is False:
+            return
+        elif self.table is True:
+            data = self.data.transpose()
+        else:
+            data = self.table
+        ax = self._get_ax(0)
+        tools.table(ax, data)
+
+    @final
+    def _post_plot_logic_common(self, ax: Axes) -> None:
+        """Common post process for each axes"""
+        if self.orientation == "vertical" or self.orientation is None:
+            type(self)._apply_axis_properties(
+                ax.xaxis, rot=self.rot, fontsize=self.fontsize
+            )
+            type(self)._apply_axis_properties(ax.yaxis, fontsize=self.fontsize)
+
+            if hasattr(ax, "right_ax"):
+                type(self)._apply_axis_properties(
+                    ax.right_ax.yaxis, fontsize=self.fontsize
+                )
+
+        elif self.orientation == "horizontal":
+            type(self)._apply_axis_properties(
+                ax.yaxis, rot=self.rot, fontsize=self.fontsize
+            )
+            type(self)._apply_axis_properties(ax.xaxis, fontsize=self.fontsize)
+
+            if hasattr(ax, "right_ax"):
+                type(self)._apply_axis_properties(
+                    ax.right_ax.yaxis, fontsize=self.fontsize
+                )
+        else:  # pragma no cover
+            raise ValueError
+
+    @abstractmethod
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        """Post process for each axes. Overridden in child classes"""
+
+    @final
+    def _adorn_subplots(self, fig: Figure) -> None:
+        """Common post process unrelated to data"""
+        if len(self.axes) > 0:
+            all_axes = self._get_subplots(fig)
+            nrows, ncols = self._get_axes_layout(fig)
+            handle_shared_axes(
+                axarr=all_axes,
+                nplots=len(all_axes),
+                naxes=nrows * ncols,
+                nrows=nrows,
+                ncols=ncols,
+                sharex=self.sharex,
+                sharey=self.sharey,
+            )
+
+        for ax in self.axes:
+            ax = getattr(ax, "right_ax", ax)
+            if self.yticks is not None:
+                ax.set_yticks(self.yticks)
+
+            if self.xticks is not None:
+                ax.set_xticks(self.xticks)
+
+            if self.ylim is not None:
+                ax.set_ylim(self.ylim)
+
+            if self.xlim is not None:
+                ax.set_xlim(self.xlim)
+
+            # GH9093, currently Pandas does not show ylabel, so if users provide
+            # ylabel will set it as ylabel in the plot.
+            if self.ylabel is not None:
+                ax.set_ylabel(pprint_thing(self.ylabel))
+
+            ax.grid(self.grid)
+
+        if self.title:
+            if self.subplots:
+                if is_list_like(self.title):
+                    if len(self.title) != self.nseries:
+                        raise ValueError(
+                            "The length of `title` must equal the number "
+                            "of columns if using `title` of type `list` "
+                            "and `subplots=True`.\n"
+                            f"length of title = {len(self.title)}\n"
+                            f"number of columns = {self.nseries}"
+                        )
+
+                    for ax, title in zip(self.axes, self.title):
+                        ax.set_title(title)
+                else:
+                    fig.suptitle(self.title)
+            else:
+                if is_list_like(self.title):
+                    msg = (
+                        "Using `title` of type `list` is not supported "
+                        "unless `subplots=True` is passed"
+                    )
+                    raise ValueError(msg)
+                self.axes[0].set_title(self.title)
+
+    @final
+    @staticmethod
+    def _apply_axis_properties(
+        axis: Axis, rot=None, fontsize: int | None = None
+    ) -> None:
+        """
+        Tick creation within matplotlib is reasonably expensive and is
+        internally deferred until accessed as Ticks are created/destroyed
+        multiple times per draw. It's therefore beneficial for us to avoid
+        accessing unless we will act on the Tick.
+        """
+        if rot is not None or fontsize is not None:
+            # rot=0 is a valid setting, hence the explicit None check
+            labels = axis.get_majorticklabels() + axis.get_minorticklabels()
+            for label in labels:
+                if rot is not None:
+                    label.set_rotation(rot)
+                if fontsize is not None:
+                    label.set_fontsize(fontsize)
+
+    @final
+    @property
+    def legend_title(self) -> str | None:
+        if not isinstance(self.data.columns, ABCMultiIndex):
+            name = self.data.columns.name
+            if name is not None:
+                name = pprint_thing(name)
+            return name
+        else:
+            stringified = map(pprint_thing, self.data.columns.names)
+            return ",".join(stringified)
+
+    @final
+    def _mark_right_label(self, label: str, index: int) -> str:
+        """
+        Append ``(right)`` to the label of a line if it's plotted on the right axis.
+
+        Note that ``(right)`` is only appended when ``subplots=False``.
+        """
+        if not self.subplots and self.mark_right and self.on_right(index):
+            label += " (right)"
+        return label
+
+    @final
+    def _append_legend_handles_labels(self, handle: Artist, label: str) -> None:
+        """
+        Append current handle and label to ``legend_handles`` and ``legend_labels``.
+
+        These will be used to make the legend.
+        """
+        self.legend_handles.append(handle)
+        self.legend_labels.append(label)
+
+    def _make_legend(self) -> None:
+        ax, leg = self._get_ax_legend(self.axes[0])
+
+        handles = []
+        labels = []
+        title = ""
+
+        if not self.subplots:
+            if leg is not None:
+                title = leg.get_title().get_text()
+                # Replace leg.legend_handles because it misses marker info
+                if Version(mpl.__version__) < Version("3.7"):
+                    handles = leg.legendHandles
+                else:
+                    handles = leg.legend_handles
+                labels = [x.get_text() for x in leg.get_texts()]
+
+            if self.legend:
+                if self.legend == "reverse":
+                    handles += reversed(self.legend_handles)
+                    labels += reversed(self.legend_labels)
+                else:
+                    handles += self.legend_handles
+                    labels += self.legend_labels
+
+                if self.legend_title is not None:
+                    title = self.legend_title
+
+            if len(handles) > 0:
+                ax.legend(handles, labels, loc="best", title=title)
+
+        elif self.subplots and self.legend:
+            for ax in self.axes:
+                if ax.get_visible():
+                    with warnings.catch_warnings():
+                        warnings.filterwarnings(
+                            "ignore",
+                            "No artists with labels found to put in legend.",
+                            UserWarning,
+                        )
+                        ax.legend(loc="best")
+
+    @final
+    @staticmethod
+    def _get_ax_legend(ax: Axes):
+        """
+        Take in axes and return ax and legend under different scenarios
+        """
+        leg = ax.get_legend()
+
+        other_ax = getattr(ax, "left_ax", None) or getattr(ax, "right_ax", None)
+        other_leg = None
+        if other_ax is not None:
+            other_leg = other_ax.get_legend()
+        if leg is None and other_leg is not None:
+            leg = other_leg
+            ax = other_ax
+        return ax, leg
+
+    _need_to_set_index = False
+
+    @final
+    def _get_xticks(self):
+        index = self.data.index
+        is_datetype = index.inferred_type in ("datetime", "date", "datetime64", "time")
+
+        # TODO: be stricter about x?
+        x: list[int] | np.ndarray
+        if self.use_index:
+            if isinstance(index, ABCPeriodIndex):
+                # test_mixed_freq_irreg_period
+                x = index.to_timestamp()._mpl_repr()
+                # TODO: why do we need to do to_timestamp() here but not other
+                #  places where we call mpl_repr?
+            elif is_any_real_numeric_dtype(index.dtype):
+                # Matplotlib supports numeric values or datetime objects as
+                # xaxis values. Taking LBYL approach here, by the time
+                # matplotlib raises exception when using non numeric/datetime
+                # values for xaxis, several actions are already taken by plt.
+                x = index._mpl_repr()
+            elif isinstance(index, ABCDatetimeIndex) or is_datetype:
+                x = index._mpl_repr()
+            else:
+                self._need_to_set_index = True
+                x = list(range(len(index)))
+        else:
+            x = list(range(len(index)))
+
+        return x
+
+    @classmethod
+    @register_pandas_matplotlib_converters
+    def _plot(
+        cls, ax: Axes, x, y: np.ndarray, style=None, is_errorbar: bool = False, **kwds
+    ):
+        mask = isna(y)
+        if mask.any():
+            y = np.ma.array(y)
+            y = np.ma.masked_where(mask, y)
+
+        if isinstance(x, ABCIndex):
+            x = x._mpl_repr()
+
+        if is_errorbar:
+            if "xerr" in kwds:
+                kwds["xerr"] = np.array(kwds.get("xerr"))
+            if "yerr" in kwds:
+                kwds["yerr"] = np.array(kwds.get("yerr"))
+            return ax.errorbar(x, y, **kwds)
+        else:
+            # prevent style kwarg from going to errorbar, where it is unsupported
+            args = (x, y, style) if style is not None else (x, y)
+            return ax.plot(*args, **kwds)
+
+    def _get_custom_index_name(self):
+        """Specify whether xlabel/ylabel should be used to override index name"""
+        return self.xlabel
+
+    @final
+    def _get_index_name(self) -> str | None:
+        if isinstance(self.data.index, ABCMultiIndex):
+            name = self.data.index.names
+            if com.any_not_none(*name):
+                name = ",".join([pprint_thing(x) for x in name])
+            else:
+                name = None
+        else:
+            name = self.data.index.name
+            if name is not None:
+                name = pprint_thing(name)
+
+        # GH 45145, override the default axis label if one is provided.
+        index_name = self._get_custom_index_name()
+        if index_name is not None:
+            name = pprint_thing(index_name)
+
+        return name
+
+    @final
+    @classmethod
+    def _get_ax_layer(cls, ax, primary: bool = True):
+        """get left (primary) or right (secondary) axes"""
+        if primary:
+            return getattr(ax, "left_ax", ax)
+        else:
+            return getattr(ax, "right_ax", ax)
+
+    @final
+    def _col_idx_to_axis_idx(self, col_idx: int) -> int:
+        """Return the index of the axis where the column at col_idx should be plotted"""
+        if isinstance(self.subplots, list):
+            # Subplots is a list: some columns will be grouped together in the same ax
+            return next(
+                group_idx
+                for (group_idx, group) in enumerate(self.subplots)
+                if col_idx in group
+            )
+        else:
+            # subplots is True: one ax per column
+            return col_idx
+
+    @final
+    def _get_ax(self, i: int) -> Axes:
+        # get the twinx ax if appropriate
+        if self.subplots:
+            i = self._col_idx_to_axis_idx(i)
+            ax = self.axes[i]
+            ax = self._maybe_right_yaxis(ax, i)
+            # error: Unsupported target for indexed assignment ("Sequence[Any]")
+            self.axes[i] = ax  # type: ignore[index]
+        else:
+            ax = self.axes[0]
+            ax = self._maybe_right_yaxis(ax, i)
+
+        ax.get_yaxis().set_visible(True)
+        return ax
+
+    @final
+    def on_right(self, i: int) -> bool:
+        if isinstance(self.secondary_y, bool):
+            return self.secondary_y
+
+        if isinstance(self.secondary_y, (tuple, list, np.ndarray, ABCIndex)):
+            return self.data.columns[i] in self.secondary_y
+
+    @final
+    def _apply_style_colors(
+        self, colors, kwds: dict[str, Any], col_num: int, label: str
+    ):
+        """
+        Manage style and color based on column number and its label.
+        Returns tuple of appropriate style and kwds which "color" may be added.
+        """
+        style = None
+        if self.style is not None:
+            if isinstance(self.style, list):
+                try:
+                    style = self.style[col_num]
+                except IndexError:
+                    pass
+            elif isinstance(self.style, dict):
+                style = self.style.get(label, style)
+            else:
+                style = self.style
+
+        has_color = "color" in kwds or self.colormap is not None
+        nocolor_style = style is None or not _color_in_style(style)
+        if (has_color or self.subplots) and nocolor_style:
+            if isinstance(colors, dict):
+                kwds["color"] = colors[label]
+            else:
+                kwds["color"] = colors[col_num % len(colors)]
+        return style, kwds
+
+    def _get_colors(
+        self,
+        num_colors: int | None = None,
+        color_kwds: str = "color",
+    ):
+        if num_colors is None:
+            num_colors = self.nseries
+        if color_kwds == "color":
+            color = self.color
+        else:
+            color = self.kwds.get(color_kwds)
+        return get_standard_colors(
+            num_colors=num_colors,
+            colormap=self.colormap,
+            color=color,
+        )
+
+    # TODO: tighter typing for first return?
+    @final
+    @staticmethod
+    def _parse_errorbars(
+        label: str, err, data: NDFrameT, nseries: int
+    ) -> tuple[Any, NDFrameT]:
+        """
+        Look for error keyword arguments and return the actual errorbar data
+        or return the error DataFrame/dict
+
+        Error bars can be specified in several ways:
+            Series: the user provides a pandas.Series object of the same
+                    length as the data
+            ndarray: provides a np.ndarray of the same length as the data
+            DataFrame/dict: error values are paired with keys matching the
+                    key in the plotted DataFrame
+            str: the name of the column within the plotted DataFrame
+
+        Asymmetrical error bars are also supported, however raw error values
+        must be provided in this case. For a ``N`` length :class:`Series`, a
+        ``2xN`` array should be provided indicating lower and upper (or left
+        and right) errors. For a ``MxN`` :class:`DataFrame`, asymmetrical errors
+        should be in a ``Mx2xN`` array.
+        """
+        if err is None:
+            return None, data
+
+        def match_labels(data, e):
+            e = e.reindex(data.index)
+            return e
+
+        # key-matched DataFrame
+        if isinstance(err, ABCDataFrame):
+            err = match_labels(data, err)
+        # key-matched dict
+        elif isinstance(err, dict):
+            pass
+
+        # Series of error values
+        elif isinstance(err, ABCSeries):
+            # broadcast error series across data
+            err = match_labels(data, err)
+            err = np.atleast_2d(err)
+            err = np.tile(err, (nseries, 1))
+
+        # errors are a column in the dataframe
+        elif isinstance(err, str):
+            evalues = data[err].values
+            data = data[data.columns.drop(err)]
+            err = np.atleast_2d(evalues)
+            err = np.tile(err, (nseries, 1))
+
+        elif is_list_like(err):
+            if is_iterator(err):
+                err = np.atleast_2d(list(err))
+            else:
+                # raw error values
+                err = np.atleast_2d(err)
+
+            err_shape = err.shape
+
+            # asymmetrical error bars
+            if isinstance(data, ABCSeries) and err_shape[0] == 2:
+                err = np.expand_dims(err, 0)
+                err_shape = err.shape
+                if err_shape[2] != len(data):
+                    raise ValueError(
+                        "Asymmetrical error bars should be provided "
+                        f"with the shape (2, {len(data)})"
+                    )
+            elif isinstance(data, ABCDataFrame) and err.ndim == 3:
+                if (
+                    (err_shape[0] != nseries)
+                    or (err_shape[1] != 2)
+                    or (err_shape[2] != len(data))
+                ):
+                    raise ValueError(
+                        "Asymmetrical error bars should be provided "
+                        f"with the shape ({nseries}, 2, {len(data)})"
+                    )
+
+            # broadcast errors to each data series
+            if len(err) == 1:
+                err = np.tile(err, (nseries, 1))
+
+        elif is_number(err):
+            err = np.tile(
+                [err],
+                (nseries, len(data)),
+            )
+
+        else:
+            msg = f"No valid {label} detected"
+            raise ValueError(msg)
+
+        return err, data
+
+    @final
+    def _get_errorbars(
+        self, label=None, index=None, xerr: bool = True, yerr: bool = True
+    ) -> dict[str, Any]:
+        errors = {}
+
+        for kw, flag in zip(["xerr", "yerr"], [xerr, yerr]):
+            if flag:
+                err = self.errors[kw]
+                # user provided label-matched dataframe of errors
+                if isinstance(err, (ABCDataFrame, dict)):
+                    if label is not None and label in err.keys():
+                        err = err[label]
+                    else:
+                        err = None
+                elif index is not None and err is not None:
+                    err = err[index]
+
+                if err is not None:
+                    errors[kw] = err
+        return errors
+
+    @final
+    def _get_subplots(self, fig: Figure) -> list[Axes]:
+        if Version(mpl.__version__) < Version("3.8"):
+            Klass = mpl.axes.Subplot
+        else:
+            Klass = mpl.axes.Axes
+
+        return [
+            ax
+            for ax in fig.get_axes()
+            if (isinstance(ax, Klass) and ax.get_subplotspec() is not None)
+        ]
+
+    @final
+    def _get_axes_layout(self, fig: Figure) -> tuple[int, int]:
+        axes = self._get_subplots(fig)
+        x_set = set()
+        y_set = set()
+        for ax in axes:
+            # check axes coordinates to estimate layout
+            points = ax.get_position().get_points()
+            x_set.add(points[0][0])
+            y_set.add(points[0][1])
+        return (len(y_set), len(x_set))
+
+
+class PlanePlot(MPLPlot, ABC):
+    """
+    Abstract class for plotting on plane, currently scatter and hexbin.
+    """
+
+    _layout_type = "single"
+
+    def __init__(self, data, x, y, **kwargs) -> None:
+        MPLPlot.__init__(self, data, **kwargs)
+        if x is None or y is None:
+            raise ValueError(self._kind + " requires an x and y column")
+        if is_integer(x) and not holds_integer(self.data.columns):
+            x = self.data.columns[x]
+        if is_integer(y) and not holds_integer(self.data.columns):
+            y = self.data.columns[y]
+
+        self.x = x
+        self.y = y
+
+    @final
+    def _get_nseries(self, data: Series | DataFrame) -> int:
+        return 1
+
+    @final
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        x, y = self.x, self.y
+        xlabel = self.xlabel if self.xlabel is not None else pprint_thing(x)
+        ylabel = self.ylabel if self.ylabel is not None else pprint_thing(y)
+        # error: Argument 1 to "set_xlabel" of "_AxesBase" has incompatible
+        # type "Hashable"; expected "str"
+        ax.set_xlabel(xlabel)  # type: ignore[arg-type]
+        ax.set_ylabel(ylabel)  # type: ignore[arg-type]
+
+    @final
+    def _plot_colorbar(self, ax: Axes, *, fig: Figure, **kwds):
+        # Addresses issues #10611 and #10678:
+        # When plotting scatterplots and hexbinplots in IPython
+        # inline backend the colorbar axis height tends not to
+        # exactly match the parent axis height.
+        # The difference is due to small fractional differences
+        # in floating points with similar representation.
+        # To deal with this, this method forces the colorbar
+        # height to take the height of the parent axes.
+        # For a more detailed description of the issue
+        # see the following link:
+        # https://github.com/ipython/ipython/issues/11215
+
+        # GH33389, if ax is used multiple times, we should always
+        # use the last one which contains the latest information
+        # about the ax
+        img = ax.collections[-1]
+        return fig.colorbar(img, ax=ax, **kwds)
+
+
+class ScatterPlot(PlanePlot):
+    @property
+    def _kind(self) -> Literal["scatter"]:
+        return "scatter"
+
+    def __init__(
+        self,
+        data,
+        x,
+        y,
+        s=None,
+        c=None,
+        *,
+        colorbar: bool | lib.NoDefault = lib.no_default,
+        norm=None,
+        **kwargs,
+    ) -> None:
+        if s is None:
+            # hide the matplotlib default for size, in case we want to change
+            # the handling of this argument later
+            s = 20
+        elif is_hashable(s) and s in data.columns:
+            s = data[s]
+        self.s = s
+
+        self.colorbar = colorbar
+        self.norm = norm
+
+        super().__init__(data, x, y, **kwargs)
+        if is_integer(c) and not holds_integer(self.data.columns):
+            c = self.data.columns[c]
+        self.c = c
+
+    def _make_plot(self, fig: Figure) -> None:
+        x, y, c, data = self.x, self.y, self.c, self.data
+        ax = self.axes[0]
+
+        c_is_column = is_hashable(c) and c in self.data.columns
+
+        color_by_categorical = c_is_column and isinstance(
+            self.data[c].dtype, CategoricalDtype
+        )
+
+        color = self.color
+        c_values = self._get_c_values(color, color_by_categorical, c_is_column)
+        norm, cmap = self._get_norm_and_cmap(c_values, color_by_categorical)
+        cb = self._get_colorbar(c_values, c_is_column)
+
+        if self.legend:
+            label = self.label
+        else:
+            label = None
+
+        # if a list of non color strings is passed in as c, color points
+        # by uniqueness of the strings, such same strings get same color
+        create_colors = not self._are_valid_colors(c_values)
+        if create_colors:
+            color_mapping = self._get_color_mapping(c_values)
+            c_values = [color_mapping[s] for s in c_values]
+
+            # build legend for labeling custom colors
+            ax.legend(
+                handles=[
+                    mpl.patches.Circle((0, 0), facecolor=c, label=s)
+                    for s, c in color_mapping.items()
+                ]
+            )
+
+        scatter = ax.scatter(
+            data[x].values,
+            data[y].values,
+            c=c_values,
+            label=label,
+            cmap=cmap,
+            norm=norm,
+            s=self.s,
+            **self.kwds,
+        )
+
+        if cb:
+            cbar_label = c if c_is_column else ""
+            cbar = self._plot_colorbar(ax, fig=fig, label=cbar_label)
+            if color_by_categorical:
+                n_cats = len(self.data[c].cat.categories)
+                cbar.set_ticks(np.linspace(0.5, n_cats - 0.5, n_cats))
+                cbar.ax.set_yticklabels(self.data[c].cat.categories)
+
+        if label is not None:
+            self._append_legend_handles_labels(
+                # error: Argument 2 to "_append_legend_handles_labels" of
+                # "MPLPlot" has incompatible type "Hashable"; expected "str"
+                scatter,
+                label,  # type: ignore[arg-type]
+            )
+
+        errors_x = self._get_errorbars(label=x, index=0, yerr=False)
+        errors_y = self._get_errorbars(label=y, index=0, xerr=False)
+        if len(errors_x) > 0 or len(errors_y) > 0:
+            err_kwds = dict(errors_x, **errors_y)
+            err_kwds["ecolor"] = scatter.get_facecolor()[0]
+            ax.errorbar(data[x].values, data[y].values, linestyle="none", **err_kwds)
+
+    def _get_c_values(self, color, color_by_categorical: bool, c_is_column: bool):
+        c = self.c
+        if c is not None and color is not None:
+            raise TypeError("Specify exactly one of `c` and `color`")
+        if c is None and color is None:
+            c_values = mpl.rcParams["patch.facecolor"]
+        elif color is not None:
+            c_values = color
+        elif color_by_categorical:
+            c_values = self.data[c].cat.codes
+        elif c_is_column:
+            c_values = self.data[c].values
+        else:
+            c_values = c
+        return c_values
+
+    def _are_valid_colors(self, c_values: Series) -> bool:
+        # check if c_values contains strings and if these strings are valid mpl colors.
+        # no need to check numerics as these (and mpl colors) will be validated for us
+        # in .Axes.scatter._parse_scatter_color_args(...)
+        unique = np.unique(c_values)
+        try:
+            if len(c_values) and all(isinstance(c, str) for c in unique):
+                mpl.colors.to_rgba_array(unique)
+
+            return True
+
+        except (TypeError, ValueError) as _:
+            return False
+
+    def _get_color_mapping(self, c_values: Series) -> dict[str, np.ndarray]:
+        unique = np.unique(c_values)
+        n_colors = len(unique)
+
+        # passing `None` here will default to :rc:`image.cmap`
+        cmap = mpl.colormaps.get_cmap(self.colormap)
+        colors = cmap(np.linspace(0, 1, n_colors))  # RGB tuples
+
+        return dict(zip(unique, colors))
+
+    def _get_norm_and_cmap(self, c_values, color_by_categorical: bool):
+        c = self.c
+        if self.colormap is not None:
+            cmap = mpl.colormaps.get_cmap(self.colormap)
+        # cmap is only used if c_values are integers, otherwise UserWarning.
+        # GH-53908: additionally call isinstance() because is_integer_dtype
+        # returns True for "b" (meaning "blue" and not int8 in this context)
+        elif not isinstance(c_values, str) and is_integer_dtype(c_values):
+            # pandas uses colormap, matplotlib uses cmap.
+            cmap = mpl.colormaps["Greys"]
+        else:
+            cmap = None
+
+        if color_by_categorical and cmap is not None:
+            n_cats = len(self.data[c].cat.categories)
+            cmap = mpl.colors.ListedColormap([cmap(i) for i in range(cmap.N)])
+            bounds = np.linspace(0, n_cats, n_cats + 1)
+            norm = mpl.colors.BoundaryNorm(bounds, cmap.N)
+            # TODO: warn that we are ignoring self.norm if user specified it?
+            #  Doesn't happen in any tests 2023-11-09
+        else:
+            norm = self.norm
+        return norm, cmap
+
+    def _get_colorbar(self, c_values, c_is_column: bool) -> bool:
+        # plot colorbar if
+        # 1. colormap is assigned, and
+        # 2.`c` is a column containing only numeric values
+        plot_colorbar = self.colormap or c_is_column
+        cb = self.colorbar
+        if cb is lib.no_default:
+            return is_numeric_dtype(c_values) and plot_colorbar
+        return cb
+
+
+class HexBinPlot(PlanePlot):
+    @property
+    def _kind(self) -> Literal["hexbin"]:
+        return "hexbin"
+
+    def __init__(self, data, x, y, C=None, *, colorbar: bool = True, **kwargs) -> None:
+        super().__init__(data, x, y, **kwargs)
+        if is_integer(C) and not holds_integer(self.data.columns):
+            C = self.data.columns[C]
+        self.C = C
+
+        self.colorbar = colorbar
+
+        # Scatter plot allows to plot objects data
+        if len(self.data[self.x]._get_numeric_data()) == 0:
+            raise ValueError(self._kind + " requires x column to be numeric")
+        if len(self.data[self.y]._get_numeric_data()) == 0:
+            raise ValueError(self._kind + " requires y column to be numeric")
+
+    def _make_plot(self, fig: Figure) -> None:
+        x, y, data, C = self.x, self.y, self.data, self.C
+        ax = self.axes[0]
+        # pandas uses colormap, matplotlib uses cmap.
+        cmap = self.colormap or "BuGn"
+        cmap = mpl.colormaps.get_cmap(cmap)
+        cb = self.colorbar
+
+        if C is None:
+            c_values = None
+        else:
+            c_values = data[C].values
+
+        ax.hexbin(data[x].values, data[y].values, C=c_values, cmap=cmap, **self.kwds)
+        if cb:
+            self._plot_colorbar(ax, fig=fig)
+
+    def _make_legend(self) -> None:
+        pass
+
+
+class LinePlot(MPLPlot):
+    _default_rot = 0
+
+    @property
+    def orientation(self) -> PlottingOrientation:
+        return "vertical"
+
+    @property
+    def _kind(self) -> Literal["line", "area", "hist", "kde", "box"]:
+        return "line"
+
+    def __init__(self, data, **kwargs) -> None:
+        from pandas.plotting import plot_params
+
+        MPLPlot.__init__(self, data, **kwargs)
+        if self.stacked:
+            self.data = self.data.fillna(value=0)
+        self.x_compat = plot_params["x_compat"]
+        if "x_compat" in self.kwds:
+            self.x_compat = bool(self.kwds.pop("x_compat"))
+
+    @final
+    def _is_ts_plot(self) -> bool:
+        # this is slightly deceptive
+        return not self.x_compat and self.use_index and self._use_dynamic_x()
+
+    @final
+    def _use_dynamic_x(self) -> bool:
+        return use_dynamic_x(self._get_ax(0), self.data)
+
+    def _make_plot(self, fig: Figure) -> None:
+        if self._is_ts_plot():
+            data = maybe_convert_index(self._get_ax(0), self.data)
+
+            x = data.index  # dummy, not used
+            plotf = self._ts_plot
+            it = data.items()
+        else:
+            x = self._get_xticks()
+            # error: Incompatible types in assignment (expression has type
+            # "Callable[[Any, Any, Any, Any, Any, Any, KwArg(Any)], Any]", variable has
+            # type "Callable[[Any, Any, Any, Any, KwArg(Any)], Any]")
+            plotf = self._plot  # type: ignore[assignment]
+            # error: Incompatible types in assignment (expression has type
+            # "Iterator[tuple[Hashable, ndarray[Any, Any]]]", variable has
+            # type "Iterable[tuple[Hashable, Series]]")
+            it = self._iter_data(data=self.data)  # type: ignore[assignment]
+
+        stacking_id = self._get_stacking_id()
+        is_errorbar = com.any_not_none(*self.errors.values())
+
+        colors = self._get_colors()
+        for i, (label, y) in enumerate(it):
+            ax = self._get_ax(i)
+            kwds = self.kwds.copy()
+            if self.color is not None:
+                kwds["color"] = self.color
+            style, kwds = self._apply_style_colors(
+                colors,
+                kwds,
+                i,
+                # error: Argument 4 to "_apply_style_colors" of "MPLPlot" has
+                # incompatible type "Hashable"; expected "str"
+                label,  # type: ignore[arg-type]
+            )
+
+            errors = self._get_errorbars(label=label, index=i)
+            kwds = dict(kwds, **errors)
+
+            label = pprint_thing(label)
+            label = self._mark_right_label(label, index=i)
+            kwds["label"] = label
+
+            newlines = plotf(
+                ax,
+                x,
+                y,
+                style=style,
+                column_num=i,
+                stacking_id=stacking_id,
+                is_errorbar=is_errorbar,
+                **kwds,
+            )
+            self._append_legend_handles_labels(newlines[0], label)
+
+            if self._is_ts_plot():
+                # reset of xlim should be used for ts data
+                # TODO: GH28021, should find a way to change view limit on xaxis
+                lines = get_all_lines(ax)
+                left, right = get_xlim(lines)
+                ax.set_xlim(left, right)
+
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls,
+        ax: Axes,
+        x,
+        y: np.ndarray,
+        style=None,
+        column_num=None,
+        stacking_id=None,
+        **kwds,
+    ):
+        # column_num is used to get the target column from plotf in line and
+        # area plots
+        if column_num == 0:
+            cls._initialize_stacker(ax, stacking_id, len(y))
+        y_values = cls._get_stacked_values(ax, stacking_id, y, kwds["label"])
+        lines = MPLPlot._plot(ax, x, y_values, style=style, **kwds)
+        cls._update_stacker(ax, stacking_id, y)
+        return lines
+
+    @final
+    def _ts_plot(self, ax: Axes, x, data: Series, style=None, **kwds):
+        # accept x to be consistent with normal plot func,
+        # x is not passed to tsplot as it uses data.index as x coordinate
+        # column_num must be in kwds for stacking purpose
+        freq, data = maybe_resample(data, ax, kwds)
+
+        # Set ax with freq info
+        decorate_axes(ax, freq)
+        # digging deeper
+        if hasattr(ax, "left_ax"):
+            decorate_axes(ax.left_ax, freq)
+        if hasattr(ax, "right_ax"):
+            decorate_axes(ax.right_ax, freq)
+        # TODO #54485
+        ax._plot_data.append((data, self._kind, kwds))  # type: ignore[attr-defined]
+
+        lines = self._plot(ax, data.index, np.asarray(data.values), style=style, **kwds)
+        # set date formatter, locators and rescale limits
+        # TODO #54485
+        format_dateaxis(ax, ax.freq, data.index)  # type: ignore[arg-type, attr-defined]
+        return lines
+
+    @final
+    def _get_stacking_id(self) -> int | None:
+        if self.stacked:
+            return id(self.data)
+        else:
+            return None
+
+    @final
+    @classmethod
+    def _initialize_stacker(cls, ax: Axes, stacking_id, n: int) -> None:
+        if stacking_id is None:
+            return
+        if not hasattr(ax, "_stacker_pos_prior"):
+            # TODO #54485
+            ax._stacker_pos_prior = {}  # type: ignore[attr-defined]
+        if not hasattr(ax, "_stacker_neg_prior"):
+            # TODO #54485
+            ax._stacker_neg_prior = {}  # type: ignore[attr-defined]
+        # TODO #54485
+        ax._stacker_pos_prior[stacking_id] = np.zeros(n)  # type: ignore[attr-defined]
+        # TODO #54485
+        ax._stacker_neg_prior[stacking_id] = np.zeros(n)  # type: ignore[attr-defined]
+
+    @final
+    @classmethod
+    def _get_stacked_values(
+        cls, ax: Axes, stacking_id: int | None, values: np.ndarray, label
+    ) -> np.ndarray:
+        if stacking_id is None:
+            return values
+        if not hasattr(ax, "_stacker_pos_prior"):
+            # stacker may not be initialized for subplots
+            cls._initialize_stacker(ax, stacking_id, len(values))
+
+        if (values >= 0).all():
+            # TODO #54485
+            return (
+                ax._stacker_pos_prior[stacking_id]  # type: ignore[attr-defined]
+                + values
+            )
+        elif (values <= 0).all():
+            # TODO #54485
+            return (
+                ax._stacker_neg_prior[stacking_id]  # type: ignore[attr-defined]
+                + values
+            )
+
+        raise ValueError(
+            "When stacked is True, each column must be either "
+            "all positive or all negative. "
+            f"Column '{label}' contains both positive and negative values"
+        )
+
+    @final
+    @classmethod
+    def _update_stacker(cls, ax: Axes, stacking_id: int | None, values) -> None:
+        if stacking_id is None:
+            return
+        if (values >= 0).all():
+            # TODO #54485
+            ax._stacker_pos_prior[stacking_id] += values  # type: ignore[attr-defined]
+        elif (values <= 0).all():
+            # TODO #54485
+            ax._stacker_neg_prior[stacking_id] += values  # type: ignore[attr-defined]
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        def get_label(i):
+            if is_float(i) and i.is_integer():
+                i = int(i)
+            try:
+                return pprint_thing(data.index[i])
+            except Exception:
+                return ""
+
+        if self._need_to_set_index:
+            xticks = ax.get_xticks()
+            xticklabels = [get_label(x) for x in xticks]
+            # error: Argument 1 to "FixedLocator" has incompatible type "ndarray[Any,
+            # Any]"; expected "Sequence[float]"
+            ax.xaxis.set_major_locator(mpl.ticker.FixedLocator(xticks))  # type: ignore[arg-type]
+            ax.set_xticklabels(xticklabels)
+
+        # If the index is an irregular time series, then by default
+        # we rotate the tick labels. The exception is if there are
+        # subplots which don't share their x-axes, in which we case
+        # we don't rotate the ticklabels as by default the subplots
+        # would be too close together.
+        condition = (
+            not self._use_dynamic_x()
+            and (data.index._is_all_dates and self.use_index)
+            and (not self.subplots or (self.subplots and self.sharex))
+        )
+
+        index_name = self._get_index_name()
+
+        if condition:
+            # irregular TS rotated 30 deg. by default
+            # probably a better place to check / set this.
+            if not self._rot_set:
+                self.rot = 30
+            format_date_labels(ax, rot=self.rot)
+
+        if index_name is not None and self.use_index:
+            ax.set_xlabel(index_name)
+
+
+class AreaPlot(LinePlot):
+    @property
+    def _kind(self) -> Literal["area"]:
+        return "area"
+
+    def __init__(self, data, **kwargs) -> None:
+        kwargs.setdefault("stacked", True)
+        data = data.fillna(value=0)
+        LinePlot.__init__(self, data, **kwargs)
+
+        if not self.stacked:
+            # use smaller alpha to distinguish overlap
+            self.kwds.setdefault("alpha", 0.5)
+
+        if self.logy or self.loglog:
+            raise ValueError("Log-y scales are not supported in area plot")
+
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls,
+        ax: Axes,
+        x,
+        y: np.ndarray,
+        style=None,
+        column_num=None,
+        stacking_id=None,
+        is_errorbar: bool = False,
+        **kwds,
+    ):
+        if column_num == 0:
+            cls._initialize_stacker(ax, stacking_id, len(y))
+        y_values = cls._get_stacked_values(ax, stacking_id, y, kwds["label"])
+
+        # need to remove label, because subplots uses mpl legend as it is
+        line_kwds = kwds.copy()
+        line_kwds.pop("label")
+        lines = MPLPlot._plot(ax, x, y_values, style=style, **line_kwds)
+
+        # get data from the line to get coordinates for fill_between
+        xdata, y_values = lines[0].get_data(orig=False)
+
+        # unable to use ``_get_stacked_values`` here to get starting point
+        if stacking_id is None:
+            start = np.zeros(len(y))
+        elif (y >= 0).all():
+            # TODO #54485
+            start = ax._stacker_pos_prior[stacking_id]  # type: ignore[attr-defined]
+        elif (y <= 0).all():
+            # TODO #54485
+            start = ax._stacker_neg_prior[stacking_id]  # type: ignore[attr-defined]
+        else:
+            start = np.zeros(len(y))
+
+        if "color" not in kwds:
+            kwds["color"] = lines[0].get_color()
+
+        rect = ax.fill_between(xdata, start, y_values, **kwds)
+        cls._update_stacker(ax, stacking_id, y)
+
+        # LinePlot expects list of artists
+        res = [rect]
+        return res
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        LinePlot._post_plot_logic(self, ax, data)
+
+        is_shared_y = len(list(ax.get_shared_y_axes())) > 0
+        # do not override the default axis behaviour in case of shared y axes
+        if self.ylim is None and not is_shared_y:
+            if (data >= 0).all().all():
+                ax.set_ylim(0, None)
+            elif (data <= 0).all().all():
+                ax.set_ylim(None, 0)
+
+
+class BarPlot(MPLPlot):
+    @property
+    def _kind(self) -> Literal["bar", "barh"]:
+        return "bar"
+
+    _default_rot = 90
+
+    @property
+    def orientation(self) -> PlottingOrientation:
+        return "vertical"
+
+    def __init__(
+        self,
+        data,
+        *,
+        align="center",
+        bottom=0,
+        left=0,
+        width=0.5,
+        position=0.5,
+        log=False,
+        **kwargs,
+    ) -> None:
+        # we have to treat a series differently than a
+        # 1-column DataFrame w.r.t. color handling
+        self._is_series = isinstance(data, ABCSeries)
+        self.bar_width = width
+        self._align = align
+        self._position = position
+        self.tick_pos = np.arange(len(data))
+
+        if is_list_like(bottom):
+            bottom = np.array(bottom)
+        if is_list_like(left):
+            left = np.array(left)
+        self.bottom = bottom
+        self.left = left
+
+        self.log = log
+
+        MPLPlot.__init__(self, data, **kwargs)
+
+    @cache_readonly
+    def ax_pos(self) -> np.ndarray:
+        return self.tick_pos - self.tickoffset
+
+    @cache_readonly
+    def tickoffset(self):
+        if self.stacked or self.subplots:
+            return self.bar_width * self._position
+        elif self._align == "edge":
+            w = self.bar_width / self.nseries
+            return self.bar_width * (self._position - 0.5) + w * 0.5
+        else:
+            return self.bar_width * self._position
+
+    @cache_readonly
+    def lim_offset(self):
+        if self.stacked or self.subplots:
+            if self._align == "edge":
+                return self.bar_width / 2
+            else:
+                return 0
+        elif self._align == "edge":
+            w = self.bar_width / self.nseries
+            return w * 0.5
+        else:
+            return 0
+
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls,
+        ax: Axes,
+        x,
+        y: np.ndarray,
+        w,
+        start: int | npt.NDArray[np.intp] = 0,
+        log: bool = False,
+        **kwds,
+    ):
+        return ax.bar(x, y, w, bottom=start, log=log, **kwds)
+
+    @property
+    def _start_base(self):
+        return self.bottom
+
+    def _make_plot(self, fig: Figure) -> None:
+        colors = self._get_colors()
+        ncolors = len(colors)
+
+        pos_prior = neg_prior = np.zeros(len(self.data))
+        K = self.nseries
+
+        data = self.data.fillna(0)
+        for i, (label, y) in enumerate(self._iter_data(data=data)):
+            ax = self._get_ax(i)
+            kwds = self.kwds.copy()
+            if self._is_series:
+                kwds["color"] = colors
+            elif isinstance(colors, dict):
+                kwds["color"] = colors[label]
+            else:
+                kwds["color"] = colors[i % ncolors]
+
+            errors = self._get_errorbars(label=label, index=i)
+            kwds = dict(kwds, **errors)
+
+            label = pprint_thing(label)
+            label = self._mark_right_label(label, index=i)
+
+            if (("yerr" in kwds) or ("xerr" in kwds)) and (kwds.get("ecolor") is None):
+                kwds["ecolor"] = mpl.rcParams["xtick.color"]
+
+            start = 0
+            if self.log and (y >= 1).all():
+                start = 1
+            start = start + self._start_base
+
+            kwds["align"] = self._align
+            if self.subplots:
+                w = self.bar_width / 2
+                rect = self._plot(
+                    ax,
+                    self.ax_pos + w,
+                    y,
+                    self.bar_width,
+                    start=start,
+                    label=label,
+                    log=self.log,
+                    **kwds,
+                )
+                ax.set_title(label)
+            elif self.stacked:
+                mask = y >= 0
+                start = np.where(mask, pos_prior, neg_prior) + self._start_base
+                w = self.bar_width / 2
+                rect = self._plot(
+                    ax,
+                    self.ax_pos + w,
+                    y,
+                    self.bar_width,
+                    start=start,
+                    label=label,
+                    log=self.log,
+                    **kwds,
+                )
+                pos_prior = pos_prior + np.where(mask, y, 0)
+                neg_prior = neg_prior + np.where(mask, 0, y)
+            else:
+                w = self.bar_width / K
+                rect = self._plot(
+                    ax,
+                    self.ax_pos + (i + 0.5) * w,
+                    y,
+                    w,
+                    start=start,
+                    label=label,
+                    log=self.log,
+                    **kwds,
+                )
+            self._append_legend_handles_labels(rect, label)
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        if self.use_index:
+            str_index = [pprint_thing(key) for key in data.index]
+        else:
+            str_index = [pprint_thing(key) for key in range(data.shape[0])]
+
+        s_edge = self.ax_pos[0] - 0.25 + self.lim_offset
+        e_edge = self.ax_pos[-1] + 0.25 + self.bar_width + self.lim_offset
+
+        self._decorate_ticks(ax, self._get_index_name(), str_index, s_edge, e_edge)
+
+    def _decorate_ticks(
+        self,
+        ax: Axes,
+        name: str | None,
+        ticklabels: list[str],
+        start_edge: float,
+        end_edge: float,
+    ) -> None:
+        ax.set_xlim((start_edge, end_edge))
+
+        if self.xticks is not None:
+            ax.set_xticks(np.array(self.xticks))
+        else:
+            ax.set_xticks(self.tick_pos)
+            ax.set_xticklabels(ticklabels)
+
+        if name is not None and self.use_index:
+            ax.set_xlabel(name)
+
+
+class BarhPlot(BarPlot):
+    @property
+    def _kind(self) -> Literal["barh"]:
+        return "barh"
+
+    _default_rot = 0
+
+    @property
+    def orientation(self) -> Literal["horizontal"]:
+        return "horizontal"
+
+    @property
+    def _start_base(self):
+        return self.left
+
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls,
+        ax: Axes,
+        x,
+        y: np.ndarray,
+        w,
+        start: int | npt.NDArray[np.intp] = 0,
+        log: bool = False,
+        **kwds,
+    ):
+        return ax.barh(x, y, w, left=start, log=log, **kwds)
+
+    def _get_custom_index_name(self):
+        return self.ylabel
+
+    def _decorate_ticks(
+        self,
+        ax: Axes,
+        name: str | None,
+        ticklabels: list[str],
+        start_edge: float,
+        end_edge: float,
+    ) -> None:
+        # horizontal bars
+        ax.set_ylim((start_edge, end_edge))
+        ax.set_yticks(self.tick_pos)
+        ax.set_yticklabels(ticklabels)
+        if name is not None and self.use_index:
+            ax.set_ylabel(name)
+        # error: Argument 1 to "set_xlabel" of "_AxesBase" has incompatible type
+        # "Hashable | None"; expected "str"
+        ax.set_xlabel(self.xlabel)  # type: ignore[arg-type]
+
+
+class PiePlot(MPLPlot):
+    @property
+    def _kind(self) -> Literal["pie"]:
+        return "pie"
+
+    _layout_type = "horizontal"
+
+    def __init__(self, data: Series | DataFrame, kind=None, **kwargs) -> None:
+        data = data.fillna(value=0)
+        lt_zero = data < 0
+        if isinstance(data, ABCDataFrame) and lt_zero.any().any():
+            raise ValueError(f"{self._kind} plot doesn't allow negative values")
+        elif isinstance(data, ABCSeries) and lt_zero.any():
+            raise ValueError(f"{self._kind} plot doesn't allow negative values")
+        MPLPlot.__init__(self, data, kind=kind, **kwargs)
+
+    @classmethod
+    def _validate_log_kwd(
+        cls,
+        kwd: str,
+        value: bool | None | Literal["sym"],
+    ) -> bool | None | Literal["sym"]:
+        super()._validate_log_kwd(kwd=kwd, value=value)
+        if value is not False:
+            warnings.warn(
+                f"PiePlot ignores the '{kwd}' keyword",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+        return False
+
+    def _validate_color_args(self, color, colormap) -> None:
+        # TODO: warn if color is passed and ignored?
+        return None
+
+    def _make_plot(self, fig: Figure) -> None:
+        colors = self._get_colors(num_colors=len(self.data), color_kwds="colors")
+        self.kwds.setdefault("colors", colors)
+
+        for i, (label, y) in enumerate(self._iter_data(data=self.data)):
+            ax = self._get_ax(i)
+
+            kwds = self.kwds.copy()
+
+            def blank_labeler(label, value):
+                if value == 0:
+                    return ""
+                else:
+                    return label
+
+            idx = [pprint_thing(v) for v in self.data.index]
+            labels = kwds.pop("labels", idx)
+            # labels is used for each wedge's labels
+            # Blank out labels for values of 0 so they don't overlap
+            # with nonzero wedges
+            if labels is not None:
+                blabels = [blank_labeler(left, value) for left, value in zip(labels, y)]
+            else:
+                blabels = None
+            results = ax.pie(y, labels=blabels, **kwds)
+
+            if kwds.get("autopct", None) is not None:
+                # error: Need more than 2 values to unpack (3 expected)
+                patches, texts, autotexts = results  # type: ignore[misc]
+            else:
+                # error: Too many values to unpack (2 expected, 3 provided)
+                patches, texts = results  # type: ignore[misc]
+                autotexts = []
+
+            if self.fontsize is not None:
+                for t in texts + autotexts:
+                    t.set_fontsize(self.fontsize)
+
+            # leglabels is used for legend labels
+            leglabels = labels if labels is not None else idx
+            for _patch, _leglabel in zip(patches, leglabels):
+                self._append_legend_handles_labels(_patch, _leglabel)
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        pass
diff --git a/pandas/pandas/plotting/_matplotlib/groupby.py b/pandas/pandas/plotting/_matplotlib/groupby.py
new file mode 100644
index 0000000000000..783f79710097c
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/groupby.py
@@ -0,0 +1,141 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.missing import remove_na_arraylike
+
+from pandas import (
+    MultiIndex,
+    concat,
+)
+
+from pandas.plotting._matplotlib.misc import unpack_single_str_list
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from pandas._typing import IndexLabel
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+def create_iter_data_given_by(
+    data: DataFrame, kind: str = "hist"
+) -> dict[Hashable, DataFrame | Series]:
+    """
+    Create data for iteration given `by` is assigned or not, and it is only
+    used in both hist and boxplot.
+
+    If `by` is assigned, return a dictionary of DataFrames in which the key of
+    dictionary is the values in groups.
+    If `by` is not assigned, return input as is, and this preserves current
+    status of iter_data.
+
+    Parameters
+    ----------
+    data : reformatted grouped data from `_compute_plot_data` method.
+    kind : str, plot kind. This function is only used for `hist` and `box` plots.
+
+    Returns
+    -------
+    iter_data : DataFrame or Dictionary of DataFrames
+
+    Examples
+    --------
+    If `by` is assigned:
+
+    >>> import numpy as np
+    >>> tuples = [("h1", "a"), ("h1", "b"), ("h2", "a"), ("h2", "b")]
+    >>> mi = pd.MultiIndex.from_tuples(tuples)
+    >>> value = [[1, 3, np.nan, np.nan], [3, 4, np.nan, np.nan], [np.nan, np.nan, 5, 6]]
+    >>> data = pd.DataFrame(value, columns=mi)
+    >>> create_iter_data_given_by(data)
+    {'h1':     h1
+         a    b
+    0  1.0  3.0
+    1  3.0  4.0
+    2  NaN  NaN, 'h2':     h2
+         a    b
+    0  NaN  NaN
+    1  NaN  NaN
+    2  5.0  6.0}
+    """
+
+    # For `hist` plot, before transformation, the values in level 0 are values
+    # in groups and subplot titles, and later used for column subselection and
+    # iteration; For `box` plot, values in level 1 are column names to show,
+    # and are used for iteration and as subplots titles.
+    if kind == "hist":
+        level = 0
+    else:
+        level = 1
+
+    # Select sub-columns based on the value of level of MI, and if `by` is
+    # assigned, data must be a MI DataFrame
+    assert isinstance(data.columns, MultiIndex)
+    return {
+        col: data.loc[:, data.columns.get_level_values(level) == col]
+        for col in data.columns.levels[level]
+    }
+
+
+def reconstruct_data_with_by(
+    data: DataFrame, by: IndexLabel, cols: IndexLabel
+) -> DataFrame:
+    """
+    Internal function to group data, and reassign multiindex column names onto the
+    result in order to let grouped data be used in _compute_plot_data method.
+
+    Parameters
+    ----------
+    data : Original DataFrame to plot
+    by : grouped `by` parameter selected by users
+    cols : columns of data set (excluding columns used in `by`)
+
+    Returns
+    -------
+    Output is the reconstructed DataFrame with MultiIndex columns. The first level
+    of MI is unique values of groups, and second level of MI is the columns
+    selected by users.
+
+    Examples
+    --------
+    >>> d = {"h": ["h1", "h1", "h2"], "a": [1, 3, 5], "b": [3, 4, 6]}
+    >>> df = pd.DataFrame(d)
+    >>> reconstruct_data_with_by(df, by="h", cols=["a", "b"])
+       h1      h2
+       a     b     a     b
+    0  1.0   3.0   NaN   NaN
+    1  3.0   4.0   NaN   NaN
+    2  NaN   NaN   5.0   6.0
+    """
+    by_modified = unpack_single_str_list(by)
+    grouped = data.groupby(by_modified)
+
+    data_list = []
+    for key, group in grouped:
+        # error: List item 1 has incompatible type "Union[Hashable,
+        # Sequence[Hashable]]"; expected "Iterable[Hashable]"
+        columns = MultiIndex.from_product([[key], cols])  # type: ignore[list-item]
+        sub_group = group[cols]
+        sub_group.columns = columns
+        data_list.append(sub_group)
+
+    data = concat(data_list, axis=1)
+    return data
+
+
+def reformat_hist_y_given_by(y: np.ndarray, by: IndexLabel | None) -> np.ndarray:
+    """Internal function to reformat y given `by` is applied or not for hist plot.
+
+    If by is None, input y is 1-d with NaN removed; and if by is not None, groupby
+    will take place and input y is multi-dimensional array.
+    """
+    if by is not None and len(y.shape) > 1:
+        return np.array([remove_na_arraylike(col) for col in y.T]).T
+    return remove_na_arraylike(y)
diff --git a/pandas/pandas/plotting/_matplotlib/hist.py b/pandas/pandas/plotting/_matplotlib/hist.py
new file mode 100644
index 0000000000000..1a423ad49c294
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/hist.py
@@ -0,0 +1,572 @@
+from __future__ import annotations
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Literal,
+    final,
+)
+
+import numpy as np
+
+from pandas.core.dtypes.common import (
+    is_integer,
+    is_list_like,
+)
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+)
+from pandas.core.dtypes.missing import (
+    isna,
+    remove_na_arraylike,
+)
+
+from pandas.io.formats.printing import pprint_thing
+from pandas.plotting._matplotlib.core import (
+    LinePlot,
+    MPLPlot,
+)
+from pandas.plotting._matplotlib.groupby import (
+    create_iter_data_given_by,
+    reformat_hist_y_given_by,
+)
+from pandas.plotting._matplotlib.misc import unpack_single_str_list
+from pandas.plotting._matplotlib.tools import (
+    create_subplots,
+    flatten_axes,
+    maybe_adjust_figure,
+    set_ticks_props,
+)
+
+if TYPE_CHECKING:
+    from matplotlib.axes import Axes
+    from matplotlib.container import BarContainer
+    from matplotlib.figure import Figure
+    from matplotlib.patches import Polygon
+
+    from pandas._typing import PlottingOrientation
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+class HistPlot(LinePlot):
+    @property
+    def _kind(self) -> Literal["hist", "kde"]:
+        return "hist"
+
+    def __init__(
+        self,
+        data,
+        bins: int | np.ndarray | list[np.ndarray] = 10,
+        bottom: int | np.ndarray = 0,
+        *,
+        range=None,
+        weights=None,
+        **kwargs,
+    ) -> None:
+        if is_list_like(bottom):
+            bottom = np.array(bottom)
+        self.bottom = bottom
+
+        self._bin_range = range
+        self.weights = weights
+
+        self.xlabel = kwargs.get("xlabel")
+        self.ylabel = kwargs.get("ylabel")
+        # Do not call LinePlot.__init__ which may fill nan
+        MPLPlot.__init__(self, data, **kwargs)
+
+        self.bins = self._adjust_bins(bins)
+
+    def _adjust_bins(self, bins: int | np.ndarray | list[np.ndarray]):
+        if is_integer(bins):
+            if self.by is not None:
+                by_modified = unpack_single_str_list(self.by)
+                grouped = self.data.groupby(by_modified)[self.columns]
+                bins = [self._calculate_bins(group, bins) for key, group in grouped]
+            else:
+                bins = self._calculate_bins(self.data, bins)
+        return bins
+
+    def _calculate_bins(self, data: Series | DataFrame, bins) -> np.ndarray:
+        """Calculate bins given data"""
+        nd_values = data.infer_objects()._get_numeric_data()
+        values = nd_values.values
+        if nd_values.ndim == 2:
+            values = values.reshape(-1)
+        values = values[~isna(values)]
+
+        return np.histogram_bin_edges(values, bins=bins, range=self._bin_range)
+
+    # error: Signature of "_plot" incompatible with supertype "LinePlot"
+    @classmethod
+    def _plot(  # type: ignore[override]
+        cls,
+        ax: Axes,
+        y: np.ndarray,
+        style=None,
+        bottom: int | np.ndarray = 0,
+        column_num: int = 0,
+        stacking_id=None,
+        *,
+        bins,
+        **kwds,
+        # might return a subset from the possible return types of Axes.hist(...)[2]?
+    ) -> BarContainer | Polygon | list[BarContainer | Polygon]:
+        if column_num == 0:
+            cls._initialize_stacker(ax, stacking_id, len(bins) - 1)
+
+        base = np.zeros(len(bins) - 1)
+        bottom = bottom + cls._get_stacked_values(ax, stacking_id, base, kwds["label"])
+        # ignore style
+        n, bins, patches = ax.hist(y, bins=bins, bottom=bottom, **kwds)
+        cls._update_stacker(ax, stacking_id, n)
+        return patches
+
+    def _make_plot(self, fig: Figure) -> None:
+        colors = self._get_colors()
+        stacking_id = self._get_stacking_id()
+
+        # Re-create iterated data if `by` is assigned by users
+        data = (
+            create_iter_data_given_by(self.data, self._kind)
+            if self.by is not None
+            else self.data
+        )
+        for i, (label, y) in enumerate(self._iter_data(data=data)):
+            ax = self._get_ax(i)
+
+            kwds = self.kwds.copy()
+            if self.color is not None:
+                kwds["color"] = self.color
+
+            label = pprint_thing(label)
+            label = self._mark_right_label(label, index=i)
+            kwds["label"] = label
+
+            style, kwds = self._apply_style_colors(colors, kwds, i, label)
+            if style is not None:
+                kwds["style"] = style
+
+            self._make_plot_keywords(kwds, y)
+
+            # the bins is multi-dimension array now and each plot need only 1-d and
+            # when by is applied, label should be columns that are grouped
+            if self.by is not None:
+                kwds["bins"] = kwds["bins"][i]
+                kwds["label"] = self.columns
+                kwds.pop("color")
+
+            if self.weights is not None:
+                kwds["weights"] = type(self)._get_column_weights(self.weights, i, y)
+
+            y = reformat_hist_y_given_by(y, self.by)
+
+            artists = self._plot(ax, y, column_num=i, stacking_id=stacking_id, **kwds)
+
+            # when by is applied, show title for subplots to know which group it is
+            if self.by is not None:
+                ax.set_title(pprint_thing(label))
+
+            # error: Value of type "Polygon" is not indexable
+            self._append_legend_handles_labels(artists[0], label)  # type: ignore[index,arg-type]
+
+    def _make_plot_keywords(self, kwds: dict[str, Any], y: np.ndarray) -> None:
+        """merge BoxPlot/KdePlot properties to passed kwds"""
+        # y is required for KdePlot
+        kwds["bottom"] = self.bottom
+        kwds["bins"] = self.bins
+
+    @final
+    @staticmethod
+    def _get_column_weights(weights, i: int, y):
+        # We allow weights to be a multi-dimensional array, e.g. a (10, 2) array,
+        # and each sub-array (10,) will be called in each iteration. If users only
+        # provide 1D array, we assume the same weights is used for all iterations
+        if weights is not None:
+            if np.ndim(weights) != 1 and np.shape(weights)[-1] != 1:
+                try:
+                    weights = weights[:, i]
+                except IndexError as err:
+                    raise ValueError(
+                        "weights must have the same shape as data, "
+                        "or be a single column"
+                    ) from err
+            weights = weights[~isna(y)]
+        return weights
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        if self.orientation == "horizontal":
+            # error: Argument 1 to "set_xlabel" of "_AxesBase" has incompatible
+            # type "Hashable"; expected "str"
+            ax.set_xlabel(
+                "Frequency" if self.xlabel is None else self.xlabel  # type: ignore[arg-type]
+            )
+            ax.set_ylabel(self.ylabel)  # type: ignore[arg-type]
+        else:
+            ax.set_xlabel(self.xlabel)  # type: ignore[arg-type]
+            ax.set_ylabel(
+                "Frequency" if self.ylabel is None else self.ylabel  # type: ignore[arg-type]
+            )
+
+    @property
+    def orientation(self) -> PlottingOrientation:
+        if self.kwds.get("orientation", None) == "horizontal":
+            return "horizontal"
+        else:
+            return "vertical"
+
+
+class KdePlot(HistPlot):
+    @property
+    def _kind(self) -> Literal["kde"]:
+        return "kde"
+
+    @property
+    def orientation(self) -> Literal["vertical"]:
+        return "vertical"
+
+    def __init__(
+        self, data, bw_method=None, ind=None, *, weights=None, **kwargs
+    ) -> None:
+        # Do not call LinePlot.__init__ which may fill nan
+        MPLPlot.__init__(self, data, **kwargs)
+        self.bw_method = bw_method
+        self.ind = ind
+        self.weights = weights
+
+    @staticmethod
+    def _get_ind(y: np.ndarray, ind):
+        if ind is None:
+            # np.nanmax() and np.nanmin() ignores the missing values
+            sample_range = np.nanmax(y) - np.nanmin(y)
+            ind = np.linspace(
+                np.nanmin(y) - 0.5 * sample_range,
+                np.nanmax(y) + 0.5 * sample_range,
+                1000,
+            )
+        elif is_integer(ind):
+            sample_range = np.nanmax(y) - np.nanmin(y)
+            ind = np.linspace(
+                np.nanmin(y) - 0.5 * sample_range,
+                np.nanmax(y) + 0.5 * sample_range,
+                ind,
+            )
+        return ind
+
+    @classmethod
+    # error: Signature of "_plot" incompatible with supertype "MPLPlot"
+    def _plot(  #  type: ignore[override]
+        cls,
+        ax: Axes,
+        y: np.ndarray,
+        style=None,
+        bw_method=None,
+        weights=None,
+        ind=None,
+        column_num=None,
+        stacking_id: int | None = None,
+        **kwds,
+    ):
+        from scipy.stats import gaussian_kde
+
+        y = remove_na_arraylike(y)
+        gkde = gaussian_kde(y, bw_method=bw_method, weights=weights)
+
+        y = gkde.evaluate(ind)
+        lines = MPLPlot._plot(ax, ind, y, style=style, **kwds)
+        return lines
+
+    def _make_plot_keywords(self, kwds: dict[str, Any], y: np.ndarray) -> None:
+        kwds["bw_method"] = self.bw_method
+        kwds["ind"] = type(self)._get_ind(y, ind=self.ind)
+
+    def _post_plot_logic(self, ax: Axes, data) -> None:
+        ax.set_ylabel("Density")
+
+
+def _grouped_plot(
+    plotf,
+    data: Series | DataFrame,
+    column=None,
+    by=None,
+    numeric_only: bool = True,
+    figsize: tuple[float, float] | None = None,
+    sharex: bool = True,
+    sharey: bool = True,
+    layout=None,
+    rot: float = 0,
+    ax=None,
+    **kwargs,
+):
+    # error: Non-overlapping equality check (left operand type: "Optional[Tuple[float,
+    # float]]", right operand type: "Literal['default']")
+    if figsize == "default":  # type: ignore[comparison-overlap]
+        # allowed to specify mpl default with 'default'
+        raise ValueError(
+            "figsize='default' is no longer supported. "
+            "Specify figure size by tuple instead"
+        )
+
+    grouped = data.groupby(by)
+    if column is not None:
+        grouped = grouped[column]
+
+    naxes = len(grouped)
+    fig, axes = create_subplots(
+        naxes=naxes, figsize=figsize, sharex=sharex, sharey=sharey, ax=ax, layout=layout
+    )
+
+    for ax, (key, group) in zip(flatten_axes(axes), grouped):
+        if numeric_only and isinstance(group, ABCDataFrame):
+            group = group._get_numeric_data()
+        plotf(group, ax, **kwargs)
+        ax.set_title(pprint_thing(key))
+
+    return fig, axes
+
+
+def _grouped_hist(
+    data: Series | DataFrame,
+    column=None,
+    by=None,
+    ax=None,
+    bins: int = 50,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    sharex: bool = False,
+    sharey: bool = False,
+    rot: float = 90,
+    grid: bool = True,
+    xlabelsize: int | None = None,
+    xrot=None,
+    ylabelsize: int | None = None,
+    yrot=None,
+    legend: bool = False,
+    **kwargs,
+):
+    """
+    Grouped histogram
+
+    Parameters
+    ----------
+    data : Series/DataFrame
+    column : object, optional
+    by : object, optional
+    ax : axes, optional
+    bins : int, default 50
+    figsize : tuple, optional
+    layout : optional
+    sharex : bool, default False
+    sharey : bool, default False
+    rot : float, default 90
+    grid : bool, default True
+    legend: : bool, default False
+    kwargs : dict, keyword arguments passed to matplotlib.Axes.hist
+
+    Returns
+    -------
+    collection of Matplotlib Axes
+    """
+    if legend:
+        assert "label" not in kwargs
+        if data.ndim == 1:
+            kwargs["label"] = data.name
+        elif column is None:
+            kwargs["label"] = data.columns
+        else:
+            kwargs["label"] = column
+
+    def plot_group(group, ax) -> None:
+        ax.hist(group.dropna().values, bins=bins, **kwargs)
+        if legend:
+            ax.legend()
+
+    if xrot is None:
+        xrot = rot
+
+    fig, axes = _grouped_plot(
+        plot_group,
+        data,
+        column=column,
+        by=by,
+        sharex=sharex,
+        sharey=sharey,
+        ax=ax,
+        figsize=figsize,
+        layout=layout,
+        rot=rot,
+    )
+
+    set_ticks_props(
+        axes, xlabelsize=xlabelsize, xrot=xrot, ylabelsize=ylabelsize, yrot=yrot
+    )
+
+    maybe_adjust_figure(
+        fig, bottom=0.15, top=0.9, left=0.1, right=0.9, hspace=0.5, wspace=0.3
+    )
+    return axes
+
+
+def hist_series(
+    self: Series,
+    by=None,
+    ax=None,
+    grid: bool = True,
+    xlabelsize: int | None = None,
+    xrot=None,
+    ylabelsize: int | None = None,
+    yrot=None,
+    figsize: tuple[float, float] | None = None,
+    bins: int = 10,
+    legend: bool = False,
+    **kwds,
+):
+    import matplotlib.pyplot as plt
+
+    if legend and "label" in kwds:
+        raise ValueError("Cannot use both legend and label")
+
+    if by is None:
+        if kwds.get("layout", None) is not None:
+            raise ValueError("The 'layout' keyword is not supported when 'by' is None")
+        # hack until the plotting interface is a bit more unified
+        fig = kwds.pop(
+            "figure", plt.gcf() if plt.get_fignums() else plt.figure(figsize=figsize)
+        )
+        if figsize is not None and tuple(figsize) != tuple(fig.get_size_inches()):
+            fig.set_size_inches(*figsize, forward=True)
+        if ax is None:
+            ax = fig.gca()
+        elif ax.get_figure() != fig:
+            raise AssertionError("passed axis not bound to passed figure")
+        values = self.dropna().values
+        if legend:
+            kwds["label"] = self.name
+        ax.hist(values, bins=bins, **kwds)
+        if legend:
+            ax.legend()
+        ax.grid(grid)
+        axes = np.array([ax])
+
+        set_ticks_props(
+            axes,
+            xlabelsize=xlabelsize,
+            xrot=xrot,
+            ylabelsize=ylabelsize,
+            yrot=yrot,
+        )
+
+    else:
+        if "figure" in kwds:
+            raise ValueError(
+                "Cannot pass 'figure' when using the "
+                "'by' argument, since a new 'Figure' instance will be created"
+            )
+        axes = _grouped_hist(
+            self,
+            by=by,
+            ax=ax,
+            grid=grid,
+            figsize=figsize,
+            bins=bins,
+            xlabelsize=xlabelsize,
+            xrot=xrot,
+            ylabelsize=ylabelsize,
+            yrot=yrot,
+            legend=legend,
+            **kwds,
+        )
+
+    if hasattr(axes, "ndim"):
+        if axes.ndim == 1 and len(axes) == 1:
+            return axes[0]
+    return axes
+
+
+def hist_frame(
+    data: DataFrame,
+    column=None,
+    by=None,
+    grid: bool = True,
+    xlabelsize: int | None = None,
+    xrot=None,
+    ylabelsize: int | None = None,
+    yrot=None,
+    ax=None,
+    sharex: bool = False,
+    sharey: bool = False,
+    figsize: tuple[float, float] | None = None,
+    layout=None,
+    bins: int = 10,
+    legend: bool = False,
+    **kwds,
+):
+    if legend and "label" in kwds:
+        raise ValueError("Cannot use both legend and label")
+    if by is not None:
+        axes = _grouped_hist(
+            data,
+            column=column,
+            by=by,
+            ax=ax,
+            grid=grid,
+            figsize=figsize,
+            sharex=sharex,
+            sharey=sharey,
+            layout=layout,
+            bins=bins,
+            xlabelsize=xlabelsize,
+            xrot=xrot,
+            ylabelsize=ylabelsize,
+            yrot=yrot,
+            legend=legend,
+            **kwds,
+        )
+        return axes
+
+    if column is not None:
+        if not isinstance(column, (list, np.ndarray, ABCIndex)):
+            column = [column]
+        data = data[column]
+    # GH32590
+    data = data.select_dtypes(
+        include=(np.number, "datetime64", "datetimetz"), exclude="timedelta"
+    )
+    naxes = len(data.columns)
+
+    if naxes == 0:
+        raise ValueError(
+            "hist method requires numerical or datetime columns, nothing to plot."
+        )
+
+    fig, axes = create_subplots(
+        naxes=naxes,
+        ax=ax,
+        squeeze=False,
+        sharex=sharex,
+        sharey=sharey,
+        figsize=figsize,
+        layout=layout,
+    )
+    can_set_label = "label" not in kwds
+
+    for ax, col in zip(flatten_axes(axes), data.columns):
+        if legend and can_set_label:
+            kwds["label"] = col
+        ax.hist(data[col].dropna().values, bins=bins, **kwds)
+        ax.set_title(col)
+        ax.grid(grid)
+        if legend:
+            ax.legend()
+
+    set_ticks_props(
+        axes, xlabelsize=xlabelsize, xrot=xrot, ylabelsize=ylabelsize, yrot=yrot
+    )
+    maybe_adjust_figure(fig, wspace=0.3, hspace=0.3)
+
+    return axes
diff --git a/pandas/pandas/plotting/_matplotlib/misc.py b/pandas/pandas/plotting/_matplotlib/misc.py
new file mode 100644
index 0000000000000..4a891ec27e8cb
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/misc.py
@@ -0,0 +1,480 @@
+from __future__ import annotations
+
+import random
+from typing import TYPE_CHECKING
+
+import matplotlib as mpl
+import numpy as np
+
+from pandas.core.dtypes.missing import notna
+
+from pandas.io.formats.printing import pprint_thing
+from pandas.plotting._matplotlib.style import get_standard_colors
+from pandas.plotting._matplotlib.tools import (
+    create_subplots,
+    do_adjust_figure,
+    maybe_adjust_figure,
+    set_ticks_props,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Hashable
+
+    from matplotlib.axes import Axes
+    from matplotlib.figure import Figure
+
+    from pandas import (
+        DataFrame,
+        Index,
+        Series,
+    )
+
+
+def scatter_matrix(
+    frame: DataFrame,
+    alpha: float = 0.5,
+    figsize: tuple[float, float] | None = None,
+    ax=None,
+    grid: bool = False,
+    diagonal: str = "hist",
+    marker: str = ".",
+    density_kwds=None,
+    hist_kwds=None,
+    range_padding: float = 0.05,
+    **kwds,
+):
+    df = frame._get_numeric_data()
+    n = df.columns.size
+    naxes = n * n
+    fig, axes = create_subplots(naxes=naxes, figsize=figsize, ax=ax, squeeze=False)
+
+    # no gaps between subplots
+    maybe_adjust_figure(fig, wspace=0, hspace=0)
+
+    mask = notna(df)
+
+    marker = _get_marker_compat(marker)
+
+    hist_kwds = hist_kwds or {}
+    density_kwds = density_kwds or {}
+
+    # GH 14855
+    kwds.setdefault("edgecolors", "none")
+
+    boundaries_list = []
+    for a in df.columns:
+        values = df[a].values[mask[a].values]
+        rmin_, rmax_ = np.min(values), np.max(values)
+        rdelta_ext = (rmax_ - rmin_) * range_padding / 2
+        boundaries_list.append((rmin_ - rdelta_ext, rmax_ + rdelta_ext))
+
+    for i, a in enumerate(df.columns):
+        for j, b in enumerate(df.columns):
+            ax = axes[i, j]
+
+            if i == j:
+                values = df[a].values[mask[a].values]
+
+                # Deal with the diagonal by drawing a histogram there.
+                if diagonal == "hist":
+                    ax.hist(values, **hist_kwds)
+
+                elif diagonal in ("kde", "density"):
+                    from scipy.stats import gaussian_kde
+
+                    y = values
+                    gkde = gaussian_kde(y)
+                    ind = np.linspace(y.min(), y.max(), 1000)
+                    ax.plot(ind, gkde.evaluate(ind), **density_kwds)
+
+                ax.set_xlim(boundaries_list[i])
+
+            else:
+                common = (mask[a] & mask[b]).values
+
+                ax.scatter(
+                    df[b][common], df[a][common], marker=marker, alpha=alpha, **kwds
+                )
+
+                ax.set_xlim(boundaries_list[j])
+                ax.set_ylim(boundaries_list[i])
+
+            ax.set_xlabel(b)
+            ax.set_ylabel(a)
+
+            if j != 0:
+                ax.yaxis.set_visible(False)
+            if i != n - 1:
+                ax.xaxis.set_visible(False)
+
+    if len(df.columns) > 1:
+        lim1 = boundaries_list[0]
+        locs = axes[0][1].yaxis.get_majorticklocs()
+        locs = locs[(lim1[0] <= locs) & (locs <= lim1[1])]
+        adj = (locs - lim1[0]) / (lim1[1] - lim1[0])
+
+        lim0 = axes[0][0].get_ylim()
+        adj = adj * (lim0[1] - lim0[0]) + lim0[0]
+        axes[0][0].yaxis.set_ticks(adj)
+
+        if np.all(locs == locs.astype(int)):
+            # if all ticks are int
+            locs = locs.astype(int)
+        axes[0][0].yaxis.set_ticklabels(locs)
+
+    set_ticks_props(axes, xlabelsize=8, xrot=90, ylabelsize=8, yrot=0)
+
+    return axes
+
+
+def _get_marker_compat(marker):
+    if marker not in mpl.lines.lineMarkers:
+        return "o"
+    return marker
+
+
+def radviz(
+    frame: DataFrame,
+    class_column,
+    ax: Axes | None = None,
+    color=None,
+    colormap=None,
+    **kwds,
+) -> Axes:
+    import matplotlib.pyplot as plt
+
+    def normalize(series):
+        a = min(series)
+        b = max(series)
+        return (series - a) / (b - a)
+
+    n = len(frame)
+    classes = frame[class_column].drop_duplicates()
+    class_col = frame[class_column]
+    df = frame.drop(class_column, axis=1).apply(normalize)
+
+    if ax is None:
+        ax = plt.gca()
+        ax.set_xlim(-1, 1)
+        ax.set_ylim(-1, 1)
+
+    to_plot: dict[Hashable, list[list]] = {}
+    colors = get_standard_colors(
+        num_colors=len(classes), colormap=colormap, color_type="random", color=color
+    )
+
+    for kls in classes:
+        to_plot[kls] = [[], []]
+
+    m = len(frame.columns) - 1
+    s = np.array(
+        [(np.cos(t), np.sin(t)) for t in [2 * np.pi * (i / m) for i in range(m)]]
+    )
+
+    for i in range(n):
+        row = df.iloc[i].values
+        row_ = np.repeat(np.expand_dims(row, axis=1), 2, axis=1)
+        y = (s * row_).sum(axis=0) / row.sum()
+        kls = class_col.iat[i]
+        to_plot[kls][0].append(y[0])
+        to_plot[kls][1].append(y[1])
+
+    for i, kls in enumerate(classes):
+        ax.scatter(
+            to_plot[kls][0],
+            to_plot[kls][1],
+            color=colors[i],
+            label=pprint_thing(kls),
+            **kwds,
+        )
+    ax.legend()
+
+    ax.add_patch(mpl.patches.Circle((0.0, 0.0), radius=1.0, facecolor="none"))
+
+    for xy, name in zip(s, df.columns):
+        ax.add_patch(mpl.patches.Circle(xy, radius=0.025, facecolor="gray"))
+
+        if xy[0] < 0.0 and xy[1] < 0.0:
+            ax.text(
+                xy[0] - 0.025, xy[1] - 0.025, name, ha="right", va="top", size="small"
+            )
+        elif xy[0] < 0.0 <= xy[1]:
+            ax.text(
+                xy[0] - 0.025,
+                xy[1] + 0.025,
+                name,
+                ha="right",
+                va="bottom",
+                size="small",
+            )
+        elif xy[1] < 0.0 <= xy[0]:
+            ax.text(
+                xy[0] + 0.025, xy[1] - 0.025, name, ha="left", va="top", size="small"
+            )
+        elif xy[0] >= 0.0 and xy[1] >= 0.0:
+            ax.text(
+                xy[0] + 0.025, xy[1] + 0.025, name, ha="left", va="bottom", size="small"
+            )
+
+    ax.axis("equal")
+    return ax
+
+
+def andrews_curves(
+    frame: DataFrame,
+    class_column,
+    ax: Axes | None = None,
+    samples: int = 200,
+    color=None,
+    colormap=None,
+    **kwds,
+) -> Axes:
+    import matplotlib.pyplot as plt
+
+    def function(amplitudes):
+        def f(t):
+            x1 = amplitudes[0]
+            result = x1 / np.sqrt(2.0)
+
+            # Take the rest of the coefficients and resize them
+            # appropriately. Take a copy of amplitudes as otherwise numpy
+            # deletes the element from amplitudes itself.
+            coeffs = np.delete(np.copy(amplitudes), 0)
+            coeffs = np.resize(coeffs, (int((coeffs.size + 1) / 2), 2))
+
+            # Generate the harmonics and arguments for the sin and cos
+            # functions.
+            harmonics = np.arange(0, coeffs.shape[0]) + 1
+            trig_args = np.outer(harmonics, t)
+
+            result += np.sum(
+                coeffs[:, 0, np.newaxis] * np.sin(trig_args)
+                + coeffs[:, 1, np.newaxis] * np.cos(trig_args),
+                axis=0,
+            )
+            return result
+
+        return f
+
+    n = len(frame)
+    class_col = frame[class_column]
+    classes = frame[class_column].drop_duplicates()
+    df = frame.drop(class_column, axis=1)
+    t = np.linspace(-np.pi, np.pi, samples)
+    used_legends: set[str] = set()
+
+    color_values = get_standard_colors(
+        num_colors=len(classes), colormap=colormap, color_type="random", color=color
+    )
+    colors = dict(zip(classes, color_values))
+    if ax is None:
+        ax = plt.gca()
+        ax.set_xlim(-np.pi, np.pi)
+    for i in range(n):
+        row = df.iloc[i].values
+        f = function(row)
+        y = f(t)
+        kls = class_col.iat[i]
+        label = pprint_thing(kls)
+        if label not in used_legends:
+            used_legends.add(label)
+            ax.plot(t, y, color=colors[kls], label=label, **kwds)
+        else:
+            ax.plot(t, y, color=colors[kls], **kwds)
+
+    ax.legend(loc="upper right")
+    ax.grid()
+    return ax
+
+
+def bootstrap_plot(
+    series: Series,
+    fig: Figure | None = None,
+    size: int = 50,
+    samples: int = 500,
+    **kwds,
+) -> Figure:
+    import matplotlib.pyplot as plt
+
+    # TODO: is the failure mentioned below still relevant?
+    # random.sample(ndarray, int) fails on python 3.3, sigh
+    data = list(series.values)
+    samplings = [random.sample(data, size) for _ in range(samples)]
+
+    means = np.array([np.mean(sampling) for sampling in samplings])
+    medians = np.array([np.median(sampling) for sampling in samplings])
+    midranges = np.array(
+        [(min(sampling) + max(sampling)) * 0.5 for sampling in samplings]
+    )
+    if fig is None:
+        fig = plt.figure()
+    x = list(range(samples))
+    axes = []
+    ax1 = fig.add_subplot(2, 3, 1)
+    ax1.set_xlabel("Sample")
+    axes.append(ax1)
+    ax1.plot(x, means, **kwds)
+    ax2 = fig.add_subplot(2, 3, 2)
+    ax2.set_xlabel("Sample")
+    axes.append(ax2)
+    ax2.plot(x, medians, **kwds)
+    ax3 = fig.add_subplot(2, 3, 3)
+    ax3.set_xlabel("Sample")
+    axes.append(ax3)
+    ax3.plot(x, midranges, **kwds)
+    ax4 = fig.add_subplot(2, 3, 4)
+    ax4.set_xlabel("Mean")
+    axes.append(ax4)
+    ax4.hist(means, **kwds)
+    ax5 = fig.add_subplot(2, 3, 5)
+    ax5.set_xlabel("Median")
+    axes.append(ax5)
+    ax5.hist(medians, **kwds)
+    ax6 = fig.add_subplot(2, 3, 6)
+    ax6.set_xlabel("Midrange")
+    axes.append(ax6)
+    ax6.hist(midranges, **kwds)
+    for axis in axes:
+        plt.setp(axis.get_xticklabels(), fontsize=8)
+        plt.setp(axis.get_yticklabels(), fontsize=8)
+    if do_adjust_figure(fig):
+        plt.tight_layout()
+    return fig
+
+
+def parallel_coordinates(
+    frame: DataFrame,
+    class_column,
+    cols=None,
+    ax: Axes | None = None,
+    color=None,
+    use_columns: bool = False,
+    xticks=None,
+    colormap=None,
+    axvlines: bool = True,
+    axvlines_kwds=None,
+    sort_labels: bool = False,
+    **kwds,
+) -> Axes:
+    import matplotlib.pyplot as plt
+
+    if axvlines_kwds is None:
+        axvlines_kwds = {"linewidth": 1, "color": "black"}
+
+    n = len(frame)
+    classes = frame[class_column].drop_duplicates()
+    class_col = frame[class_column]
+
+    if cols is None:
+        df = frame.drop(class_column, axis=1)
+    else:
+        df = frame[cols]
+
+    used_legends: set[str] = set()
+
+    ncols = len(df.columns)
+
+    # determine values to use for xticks
+    x: list[int] | Index
+    if use_columns is True:
+        if not np.all(np.isreal(list(df.columns))):
+            raise ValueError("Columns must be numeric to be used as xticks")
+        x = df.columns
+    elif xticks is not None:
+        if not np.all(np.isreal(xticks)):
+            raise ValueError("xticks specified must be numeric")
+        if len(xticks) != ncols:
+            raise ValueError("Length of xticks must match number of columns")
+        x = xticks
+    else:
+        x = list(range(ncols))
+
+    if ax is None:
+        ax = plt.gca()
+
+    color_values = get_standard_colors(
+        num_colors=len(classes), colormap=colormap, color_type="random", color=color
+    )
+
+    if sort_labels:
+        classes = sorted(classes)
+        color_values = sorted(color_values)
+    colors = dict(zip(classes, color_values))
+
+    for i in range(n):
+        y = df.iloc[i].values
+        kls = class_col.iat[i]
+        label = pprint_thing(kls)
+        if label not in used_legends:
+            used_legends.add(label)
+            ax.plot(x, y, color=colors[kls], label=label, **kwds)
+        else:
+            ax.plot(x, y, color=colors[kls], **kwds)
+
+    if axvlines:
+        for i in x:
+            ax.axvline(i, **axvlines_kwds)
+
+    ax.set_xticks(x)
+    ax.set_xticklabels(df.columns)
+    ax.set_xlim(x[0], x[-1])
+    ax.legend(loc="upper right")
+    ax.grid()
+    return ax
+
+
+def lag_plot(series: Series, lag: int = 1, ax: Axes | None = None, **kwds) -> Axes:
+    # workaround because `c='b'` is hardcoded in matplotlib's scatter method
+    import matplotlib.pyplot as plt
+
+    kwds.setdefault("c", plt.rcParams["patch.facecolor"])
+
+    data = series.values
+    y1 = data[:-lag]
+    y2 = data[lag:]
+    if ax is None:
+        ax = plt.gca()
+    ax.set_xlabel("y(t)")
+    ax.set_ylabel(f"y(t + {lag})")
+    ax.scatter(y1, y2, **kwds)
+    return ax
+
+
+def autocorrelation_plot(series: Series, ax: Axes | None = None, **kwds) -> Axes:
+    import matplotlib.pyplot as plt
+
+    n = len(series)
+    data = np.asarray(series)
+    if ax is None:
+        ax = plt.gca()
+        ax.set_xlim(1, n)
+        ax.set_ylim(-1.0, 1.0)
+    mean = np.mean(data)
+    c0 = np.sum((data - mean) ** 2) / n
+
+    def r(h):
+        return ((data[: n - h] - mean) * (data[h:] - mean)).sum() / n / c0
+
+    x = np.arange(n) + 1
+    y = [r(loc) for loc in x]
+    z95 = 1.959963984540054
+    z99 = 2.5758293035489004
+    ax.axhline(y=z99 / np.sqrt(n), linestyle="--", color="grey")
+    ax.axhline(y=z95 / np.sqrt(n), color="grey")
+    ax.axhline(y=0.0, color="black")
+    ax.axhline(y=-z95 / np.sqrt(n), color="grey")
+    ax.axhline(y=-z99 / np.sqrt(n), linestyle="--", color="grey")
+    ax.set_xlabel("Lag")
+    ax.set_ylabel("Autocorrelation")
+    ax.plot(x, y, **kwds)
+    if "label" in kwds:
+        ax.legend()
+    ax.grid()
+    return ax
+
+
+def unpack_single_str_list(keys):
+    # GH 42795
+    if isinstance(keys, list) and len(keys) == 1:
+        keys = keys[0]
+    return keys
diff --git a/pandas/pandas/plotting/_matplotlib/style.py b/pandas/pandas/plotting/_matplotlib/style.py
new file mode 100644
index 0000000000000..962f9711d9916
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/style.py
@@ -0,0 +1,309 @@
+from __future__ import annotations
+
+from collections.abc import (
+    Collection,
+    Iterator,
+    Sequence,
+)
+import itertools
+from typing import (
+    TYPE_CHECKING,
+    cast,
+    overload,
+)
+import warnings
+
+import matplotlib as mpl
+import matplotlib.colors
+import numpy as np
+
+from pandas._typing import MatplotlibColor as Color
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import is_list_like
+
+import pandas.core.common as com
+
+if TYPE_CHECKING:
+    from matplotlib.colors import Colormap
+
+
+@overload
+def get_standard_colors(
+    num_colors: int,
+    colormap: Colormap | None = ...,
+    color_type: str = ...,
+    *,
+    color: dict[str, Color],
+) -> dict[str, Color]: ...
+
+
+@overload
+def get_standard_colors(
+    num_colors: int,
+    colormap: Colormap | None = ...,
+    color_type: str = ...,
+    *,
+    color: Color | Sequence[Color] | None = ...,
+) -> list[Color]: ...
+
+
+@overload
+def get_standard_colors(
+    num_colors: int,
+    colormap: Colormap | None = ...,
+    color_type: str = ...,
+    *,
+    color: dict[str, Color] | Color | Sequence[Color] | None = ...,
+) -> dict[str, Color] | list[Color]: ...
+
+
+def get_standard_colors(
+    num_colors: int,
+    colormap: Colormap | None = None,
+    color_type: str = "default",
+    *,
+    color: dict[str, Color] | Color | Sequence[Color] | None = None,
+) -> dict[str, Color] | list[Color]:
+    """
+    Get standard colors based on `colormap`, `color_type` or `color` inputs.
+
+    Parameters
+    ----------
+    num_colors : int
+        Minimum number of colors to be returned.
+        Ignored if `color` is a dictionary.
+    colormap : :py:class:`matplotlib.colors.Colormap`, optional
+        Matplotlib colormap.
+        When provided, the resulting colors will be derived from the colormap.
+    color_type : {"default", "random"}, optional
+        Type of colors to derive. Used if provided `color` and `colormap` are None.
+        Ignored if either `color` or `colormap` are not None.
+    color : dict or str or sequence, optional
+        Color(s) to be used for deriving sequence of colors.
+        Can be either be a dictionary, or a single color (single color string,
+        or sequence of floats representing a single color),
+        or a sequence of colors.
+
+    Returns
+    -------
+    dict or list
+        Standard colors. Can either be a mapping if `color` was a dictionary,
+        or a list of colors with a length of `num_colors` or more.
+
+    Warns
+    -----
+    UserWarning
+        If both `colormap` and `color` are provided.
+        Parameter `color` will override.
+    """
+    if isinstance(color, dict):
+        return color
+
+    colors = _derive_colors(
+        color=color,
+        colormap=colormap,
+        color_type=color_type,
+        num_colors=num_colors,
+    )
+
+    return list(_cycle_colors(colors, num_colors=num_colors))
+
+
+def _derive_colors(
+    *,
+    color: Color | Collection[Color] | None,
+    colormap: str | Colormap | None,
+    color_type: str,
+    num_colors: int,
+) -> list[Color]:
+    """
+    Derive colors from either `colormap`, `color_type` or `color` inputs.
+
+    Get a list of colors either from `colormap`, or from `color`,
+    or from `color_type` (if both `colormap` and `color` are None).
+
+    Parameters
+    ----------
+    color : str or sequence, optional
+        Color(s) to be used for deriving sequence of colors.
+        Can be either be a single color (single color string, or sequence of floats
+        representing a single color), or a sequence of colors.
+    colormap : :py:class:`matplotlib.colors.Colormap`, optional
+        Matplotlib colormap.
+        When provided, the resulting colors will be derived from the colormap.
+    color_type : {"default", "random"}, optional
+        Type of colors to derive. Used if provided `color` and `colormap` are None.
+        Ignored if either `color` or `colormap`` are not None.
+    num_colors : int
+        Number of colors to be extracted.
+
+    Returns
+    -------
+    list
+        List of colors extracted.
+
+    Warns
+    -----
+    UserWarning
+        If both `colormap` and `color` are provided.
+        Parameter `color` will override.
+    """
+    if color is None and colormap is not None:
+        return _get_colors_from_colormap(colormap, num_colors=num_colors)
+    elif color is not None:
+        if colormap is not None:
+            warnings.warn(
+                "'color' and 'colormap' cannot be used simultaneously. Using 'color'",
+                stacklevel=find_stack_level(),
+            )
+        return _get_colors_from_color(color)
+    else:
+        return _get_colors_from_color_type(color_type, num_colors=num_colors)
+
+
+def _cycle_colors(colors: list[Color], num_colors: int) -> Iterator[Color]:
+    """Cycle colors until achieving max of `num_colors` or length of `colors`.
+
+    Extra colors will be ignored by matplotlib if there are more colors
+    than needed and nothing needs to be done here.
+    """
+    max_colors = max(num_colors, len(colors))
+    yield from itertools.islice(itertools.cycle(colors), max_colors)
+
+
+def _get_colors_from_colormap(
+    colormap: str | Colormap,
+    num_colors: int,
+) -> list[Color]:
+    """Get colors from colormap."""
+    cmap = _get_cmap_instance(colormap)
+    return [cmap(num) for num in np.linspace(0, 1, num=num_colors)]
+
+
+def _get_cmap_instance(colormap: str | Colormap) -> Colormap:
+    """Get instance of matplotlib colormap."""
+    if isinstance(colormap, str):
+        cmap = colormap
+        colormap = mpl.colormaps[colormap]
+        if colormap is None:
+            raise ValueError(f"Colormap {cmap} is not recognized")
+    return colormap
+
+
+def _get_colors_from_color(
+    color: Color | Collection[Color],
+) -> list[Color]:
+    """Get colors from user input color."""
+    if len(color) == 0:
+        raise ValueError(f"Invalid color argument: {color}")
+
+    if _is_single_color(color):
+        color = cast(Color, color)
+        return [color]
+
+    color = cast(Collection[Color], color)
+    return list(_gen_list_of_colors_from_iterable(color))
+
+
+def _is_single_color(color: Color | Collection[Color]) -> bool:
+    """Check if `color` is a single color, not a sequence of colors.
+
+    Single color is of these kinds:
+        - Named color "red", "C0", "firebrick"
+        - Alias "g"
+        - Sequence of floats, such as (0.1, 0.2, 0.3) or (0.1, 0.2, 0.3, 0.4).
+
+    See Also
+    --------
+    _is_single_string_color
+    """
+    if isinstance(color, str) and _is_single_string_color(color):
+        # GH #36972
+        return True
+
+    if _is_floats_color(color):
+        return True
+
+    return False
+
+
+def _gen_list_of_colors_from_iterable(color: Collection[Color]) -> Iterator[Color]:
+    """
+    Yield colors from string of several letters or from collection of colors.
+    """
+    for x in color:
+        if _is_single_color(x):
+            yield x
+        else:
+            raise ValueError(f"Invalid color {x}")
+
+
+def _is_floats_color(color: Color | Collection[Color]) -> bool:
+    """Check if color comprises a sequence of floats representing color."""
+    return bool(
+        is_list_like(color)
+        and (len(color) == 3 or len(color) == 4)
+        and all(isinstance(x, (int, float)) for x in color)
+    )
+
+
+def _get_colors_from_color_type(color_type: str, num_colors: int) -> list[Color]:
+    """Get colors from user input color type."""
+    if color_type == "default":
+        return _get_default_colors(num_colors)
+    elif color_type == "random":
+        return _get_random_colors(num_colors)
+    else:
+        raise ValueError("color_type must be either 'default' or 'random'")
+
+
+def _get_default_colors(num_colors: int) -> list[Color]:
+    """Get `num_colors` of default colors from matplotlib rc params."""
+    colors = [c["color"] for c in mpl.rcParams["axes.prop_cycle"]]
+    return colors[0:num_colors]
+
+
+def _get_random_colors(num_colors: int) -> list[Color]:
+    """Get `num_colors` of random colors."""
+    return [_random_color(num) for num in range(num_colors)]
+
+
+def _random_color(column: int) -> list[float]:
+    """Get a random color represented as a list of length 3"""
+    # GH17525 use common._random_state to avoid resetting the seed
+    rs = com.random_state(column)
+    return rs.rand(3).tolist()
+
+
+def _is_single_string_color(color: Color) -> bool:
+    """Check if `color` is a single string color.
+
+    Examples of single string colors:
+        - 'r'
+        - 'g'
+        - 'red'
+        - 'green'
+        - 'C3'
+        - 'firebrick'
+
+    Parameters
+    ----------
+    color : Color
+        Color string or sequence of floats.
+
+    Returns
+    -------
+    bool
+        True if `color` looks like a valid color.
+        False otherwise.
+    """
+    conv = matplotlib.colors.ColorConverter()
+    try:
+        # error: Argument 1 to "to_rgba" of "ColorConverter" has incompatible type
+        # "str | Sequence[float]"; expected "tuple[float, float, float] | ..."
+        conv.to_rgba(color)  # type: ignore[arg-type]
+    except ValueError:
+        return False
+    else:
+        return True
diff --git a/pandas/pandas/plotting/_matplotlib/timeseries.py b/pandas/pandas/plotting/_matplotlib/timeseries.py
new file mode 100644
index 0000000000000..d95ccad2da565
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/timeseries.py
@@ -0,0 +1,368 @@
+# TODO: Use the fact that axis can have units to simplify the process
+
+from __future__ import annotations
+
+import functools
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+import numpy as np
+
+from pandas._libs.tslibs import (
+    BaseOffset,
+    Period,
+    to_offset,
+)
+from pandas._libs.tslibs.dtypes import (
+    OFFSET_TO_PERIOD_FREQSTR,
+    FreqGroup,
+)
+
+from pandas.core.dtypes.generic import (
+    ABCDatetimeIndex,
+    ABCPeriodIndex,
+    ABCTimedeltaIndex,
+)
+
+from pandas.io.formats.printing import pprint_thing
+from pandas.plotting._matplotlib.converter import (
+    TimeSeries_DateFormatter,
+    TimeSeries_DateLocator,
+    TimeSeries_TimedeltaFormatter,
+)
+from pandas.tseries.frequencies import (
+    get_period_alias,
+    is_subperiod,
+    is_superperiod,
+)
+
+if TYPE_CHECKING:
+    from datetime import timedelta
+
+    from matplotlib.axes import Axes
+
+    from pandas._typing import NDFrameT
+
+    from pandas import (
+        DataFrame,
+        DatetimeIndex,
+        Index,
+        PeriodIndex,
+        Series,
+    )
+
+# ---------------------------------------------------------------------
+# Plotting functions and monkey patches
+
+
+def maybe_resample(series: Series, ax: Axes, kwargs: dict[str, Any]):
+    # resample against axes freq if necessary
+
+    if "how" in kwargs:
+        raise ValueError(
+            "'how' is not a valid keyword for plotting functions. If plotting "
+            "multiple objects on shared axes, resample manually first."
+        )
+
+    freq, ax_freq = _get_freq(ax, series)
+
+    if freq is None:  # pragma: no cover
+        raise ValueError("Cannot use dynamic axis without frequency info")
+
+    # Convert DatetimeIndex to PeriodIndex
+    if isinstance(series.index, ABCDatetimeIndex):
+        series = series.to_period(freq=freq)
+
+    if ax_freq is not None and freq != ax_freq:
+        if is_superperiod(freq, ax_freq):  # upsample input
+            series = series.copy()
+            # error: "Index" has no attribute "asfreq"
+            series.index = series.index.asfreq(  # type: ignore[attr-defined]
+                ax_freq, how="s"
+            )
+            freq = ax_freq
+        elif _is_sup(freq, ax_freq):  # one is weekly
+            # Resampling with PeriodDtype is deprecated, so we convert to
+            #  DatetimeIndex, resample, then convert back.
+            ser_ts = series.to_timestamp()
+            ser_d = ser_ts.resample("D").last().dropna()
+            ser_freq = ser_d.resample(ax_freq).last().dropna()
+            series = ser_freq.to_period(ax_freq)
+            freq = ax_freq
+        elif is_subperiod(freq, ax_freq) or _is_sub(freq, ax_freq):
+            _upsample_others(ax, freq, kwargs)
+        else:  # pragma: no cover
+            raise ValueError("Incompatible frequency conversion")
+    return freq, series
+
+
+def _is_sub(f1: str, f2: str) -> bool:
+    return (f1.startswith("W") and is_subperiod("D", f2)) or (
+        f2.startswith("W") and is_subperiod(f1, "D")
+    )
+
+
+def _is_sup(f1: str, f2: str) -> bool:
+    return (f1.startswith("W") and is_superperiod("D", f2)) or (
+        f2.startswith("W") and is_superperiod(f1, "D")
+    )
+
+
+def _upsample_others(ax: Axes, freq: BaseOffset, kwargs: dict[str, Any]) -> None:
+    legend = ax.get_legend()
+    lines, labels = _replot_ax(ax, freq)
+    _replot_ax(ax, freq)
+
+    other_ax = None
+    if hasattr(ax, "left_ax"):
+        other_ax = ax.left_ax
+    if hasattr(ax, "right_ax"):
+        other_ax = ax.right_ax
+
+    if other_ax is not None:
+        rlines, rlabels = _replot_ax(other_ax, freq)
+        lines.extend(rlines)
+        labels.extend(rlabels)
+
+    if legend is not None and kwargs.get("legend", True) and len(lines) > 0:
+        title: str | None = legend.get_title().get_text()
+        if title == "None":
+            title = None
+        ax.legend(lines, labels, loc="best", title=title)
+
+
+def _replot_ax(ax: Axes, freq: BaseOffset):
+    data = getattr(ax, "_plot_data", None)
+
+    # clear current axes and data
+    # TODO #54485
+    ax._plot_data = []  # type: ignore[attr-defined]
+    ax.clear()
+
+    decorate_axes(ax, freq)
+
+    lines = []
+    labels = []
+    if data is not None:
+        for series, plotf, kwds in data:
+            series = series.copy()
+            idx = series.index.asfreq(freq, how="S")
+            series.index = idx
+            # TODO #54485
+            ax._plot_data.append((series, plotf, kwds))  # type: ignore[attr-defined]
+
+            # for tsplot
+            if isinstance(plotf, str):
+                from pandas.plotting._matplotlib import PLOT_CLASSES
+
+                plotf = PLOT_CLASSES[plotf]._plot
+
+            lines.append(plotf(ax, series.index._mpl_repr(), series.values, **kwds)[0])
+            labels.append(pprint_thing(series.name))
+
+    return lines, labels
+
+
+def decorate_axes(ax: Axes, freq: BaseOffset) -> None:
+    """Initialize axes for time-series plotting"""
+    if not hasattr(ax, "_plot_data"):
+        # TODO #54485
+        ax._plot_data = []  # type: ignore[attr-defined]
+
+    # TODO #54485
+    ax.freq = freq  # type: ignore[attr-defined]
+    xaxis = ax.get_xaxis()
+    # TODO #54485
+    xaxis.freq = freq  # type: ignore[attr-defined]
+
+
+def _get_ax_freq(ax: Axes):
+    """
+    Get the freq attribute of the ax object if set.
+    Also checks shared axes (eg when using secondary yaxis, sharex=True
+    or twinx)
+    """
+    ax_freq = getattr(ax, "freq", None)
+    if ax_freq is None:
+        # check for left/right ax in case of secondary yaxis
+        if hasattr(ax, "left_ax"):
+            ax_freq = getattr(ax.left_ax, "freq", None)
+        elif hasattr(ax, "right_ax"):
+            ax_freq = getattr(ax.right_ax, "freq", None)
+    if ax_freq is None:
+        # check if a shared ax (sharex/twinx) has already freq set
+        shared_axes = ax.get_shared_x_axes().get_siblings(ax)
+        if len(shared_axes) > 1:
+            for shared_ax in shared_axes:
+                ax_freq = getattr(shared_ax, "freq", None)
+                if ax_freq is not None:
+                    break
+    return ax_freq
+
+
+def _get_period_alias(freq: timedelta | BaseOffset | str) -> str | None:
+    if isinstance(freq, BaseOffset):
+        freqstr = freq.name
+    else:
+        freqstr = to_offset(freq, is_period=True).rule_code
+
+    return get_period_alias(freqstr)
+
+
+def _get_freq(ax: Axes, series: Series):
+    # get frequency from data
+    freq = getattr(series.index, "freq", None)
+    if freq is None:
+        freq = getattr(series.index, "inferred_freq", None)
+        freq = to_offset(freq, is_period=True)
+
+    ax_freq = _get_ax_freq(ax)
+
+    # use axes freq if no data freq
+    if freq is None:
+        freq = ax_freq
+
+    # get the period frequency
+    freq = _get_period_alias(freq)
+    return freq, ax_freq
+
+
+def use_dynamic_x(ax: Axes, data: DataFrame | Series) -> bool:
+    freq = _get_index_freq(data.index)
+    ax_freq = _get_ax_freq(ax)
+
+    if freq is None:  # convert irregular if axes has freq info
+        freq = ax_freq
+    # do not use tsplot if irregular was plotted first
+    elif (ax_freq is None) and (len(ax.get_lines()) > 0):
+        return False
+
+    if freq is None:
+        return False
+
+    freq_str = _get_period_alias(freq)
+
+    if freq_str is None:
+        return False
+
+    # FIXME: hack this for 0.10.1, creating more technical debt...sigh
+    if isinstance(data.index, ABCDatetimeIndex):
+        # error: "BaseOffset" has no attribute "_period_dtype_code"
+        freq_str = OFFSET_TO_PERIOD_FREQSTR.get(freq_str, freq_str)
+        base = to_offset(freq_str, is_period=True)._period_dtype_code  # type: ignore[attr-defined]
+        x = data.index
+        if base <= FreqGroup.FR_DAY.value:
+            return x[:1].is_normalized
+        period = Period(x[0], freq_str)
+        assert isinstance(period, Period)
+        return period.to_timestamp().tz_localize(x.tz) == x[0]
+    return True
+
+
+def _get_index_freq(index: Index) -> BaseOffset | None:
+    freq = getattr(index, "freq", None)
+    if freq is None:
+        freq = getattr(index, "inferred_freq", None)
+        if freq == "B":
+            # error: "Index" has no attribute "dayofweek"
+            weekdays = np.unique(index.dayofweek)  # type: ignore[attr-defined]
+            if (5 in weekdays) or (6 in weekdays):
+                freq = None
+
+    freq = to_offset(freq)
+    return freq
+
+
+def maybe_convert_index(ax: Axes, data: NDFrameT) -> NDFrameT:
+    # tsplot converts automatically, but don't want to convert index
+    # over and over for DataFrames
+    if isinstance(data.index, (ABCDatetimeIndex, ABCPeriodIndex)):
+        freq: str | BaseOffset | None = data.index.freq
+
+        if freq is None:
+            # We only get here for DatetimeIndex
+            data.index = cast("DatetimeIndex", data.index)
+            freq = data.index.inferred_freq
+            freq = to_offset(freq)
+
+        if freq is None:
+            freq = _get_ax_freq(ax)
+
+        if freq is None:
+            raise ValueError("Could not get frequency alias for plotting")
+
+        freq_str = _get_period_alias(freq)
+
+        with warnings.catch_warnings():
+            # suppress Period[B] deprecation warning
+            # TODO: need to find an alternative to this before the deprecation
+            #  is enforced!
+            warnings.filterwarnings(
+                "ignore",
+                r"PeriodDtype\[B\] is deprecated",
+                category=FutureWarning,
+            )
+
+            if isinstance(data.index, ABCDatetimeIndex):
+                data = data.tz_localize(None).to_period(freq=freq_str)
+            elif isinstance(data.index, ABCPeriodIndex):
+                data.index = data.index.asfreq(freq=freq_str, how="start")
+    return data
+
+
+# Patch methods for subplot.
+
+
+def _format_coord(freq, t, y) -> str:
+    time_period = Period(ordinal=int(t), freq=freq)
+    return f"t = {time_period}  y = {y:8f}"
+
+
+def format_dateaxis(
+    subplot, freq: BaseOffset, index: DatetimeIndex | PeriodIndex
+) -> None:
+    """
+    Pretty-formats the date axis (x-axis).
+
+    Major and minor ticks are automatically set for the frequency of the
+    current underlying series.  As the dynamic mode is activated by
+    default, changing the limits of the x axis will intelligently change
+    the positions of the ticks.
+    """
+    import matplotlib.pyplot as plt
+
+    # handle index specific formatting
+    # Note: DatetimeIndex does not use this
+    # interface. DatetimeIndex uses matplotlib.date directly
+    if isinstance(index, ABCPeriodIndex):
+        majlocator = TimeSeries_DateLocator(
+            freq, dynamic_mode=True, minor_locator=False, plot_obj=subplot
+        )
+        minlocator = TimeSeries_DateLocator(
+            freq, dynamic_mode=True, minor_locator=True, plot_obj=subplot
+        )
+        subplot.xaxis.set_major_locator(majlocator)
+        subplot.xaxis.set_minor_locator(minlocator)
+
+        majformatter = TimeSeries_DateFormatter(
+            freq, dynamic_mode=True, minor_locator=False, plot_obj=subplot
+        )
+        minformatter = TimeSeries_DateFormatter(
+            freq, dynamic_mode=True, minor_locator=True, plot_obj=subplot
+        )
+        subplot.xaxis.set_major_formatter(majformatter)
+        subplot.xaxis.set_minor_formatter(minformatter)
+
+        # x and y coord info
+        subplot.format_coord = functools.partial(_format_coord, freq)
+
+    elif isinstance(index, ABCTimedeltaIndex):
+        subplot.xaxis.set_major_formatter(TimeSeries_TimedeltaFormatter())
+    else:
+        raise TypeError("index type not supported")
+
+    plt.draw_if_interactive()
diff --git a/pandas/pandas/plotting/_matplotlib/tools.py b/pandas/pandas/plotting/_matplotlib/tools.py
new file mode 100644
index 0000000000000..8ee75e7fe553e
--- /dev/null
+++ b/pandas/pandas/plotting/_matplotlib/tools.py
@@ -0,0 +1,491 @@
+# being a bit too dynamic
+from __future__ import annotations
+
+from math import ceil
+from typing import TYPE_CHECKING
+import warnings
+
+import matplotlib as mpl
+import numpy as np
+
+from pandas.util._exceptions import find_stack_level
+
+from pandas.core.dtypes.common import is_list_like
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndex,
+    ABCSeries,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Iterable,
+    )
+
+    from matplotlib.axes import Axes
+    from matplotlib.axis import Axis
+    from matplotlib.figure import Figure
+    from matplotlib.lines import Line2D
+    from matplotlib.table import Table
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+def do_adjust_figure(fig: Figure) -> bool:
+    """Whether fig has constrained_layout enabled."""
+    if not hasattr(fig, "get_constrained_layout"):
+        return False
+    return not fig.get_constrained_layout()
+
+
+def maybe_adjust_figure(fig: Figure, *args, **kwargs) -> None:
+    """Call fig.subplots_adjust unless fig has constrained_layout enabled."""
+    if do_adjust_figure(fig):
+        fig.subplots_adjust(*args, **kwargs)
+
+
+def format_date_labels(ax: Axes, rot) -> None:
+    # mini version of autofmt_xdate
+    for label in ax.get_xticklabels():
+        label.set_horizontalalignment("right")
+        label.set_rotation(rot)
+    fig = ax.get_figure()
+    if fig is not None:
+        # should always be a Figure but can technically be None
+        maybe_adjust_figure(fig, bottom=0.2)  # type: ignore[arg-type]
+
+
+def table(
+    ax, data: DataFrame | Series, rowLabels=None, colLabels=None, **kwargs
+) -> Table:
+    if isinstance(data, ABCSeries):
+        data = data.to_frame()
+    elif isinstance(data, ABCDataFrame):
+        pass
+    else:
+        raise ValueError("Input data must be DataFrame or Series")
+
+    if rowLabels is None:
+        rowLabels = data.index
+
+    if colLabels is None:
+        colLabels = data.columns
+
+    cellText = data.values
+
+    # error: Argument "cellText" to "table" has incompatible type "ndarray[Any,
+    # Any]"; expected "Sequence[Sequence[str]] | None"
+    return mpl.table.table(
+        ax,
+        cellText=cellText,  # type: ignore[arg-type]
+        rowLabels=rowLabels,
+        colLabels=colLabels,
+        **kwargs,
+    )
+
+
+def _get_layout(
+    nplots: int,
+    layout: tuple[int, int] | None = None,
+    layout_type: str = "box",
+) -> tuple[int, int]:
+    if layout is not None:
+        if not isinstance(layout, (tuple, list)) or len(layout) != 2:
+            raise ValueError("Layout must be a tuple of (rows, columns)")
+
+        nrows, ncols = layout
+
+        if nrows == -1 and ncols > 0:
+            layout = (ceil(nplots / ncols), ncols)
+        elif ncols == -1 and nrows > 0:
+            layout = (nrows, ceil(nplots / nrows))
+        elif ncols <= 0 and nrows <= 0:
+            msg = "At least one dimension of layout must be positive"
+            raise ValueError(msg)
+
+        nrows, ncols = layout
+        if nrows * ncols < nplots:
+            raise ValueError(
+                f"Layout of {nrows}x{ncols} must be larger than required size {nplots}"
+            )
+
+        return layout
+
+    if layout_type == "single":
+        return (1, 1)
+    elif layout_type == "horizontal":
+        return (1, nplots)
+    elif layout_type == "vertical":
+        return (nplots, 1)
+
+    layouts = {1: (1, 1), 2: (1, 2), 3: (2, 2), 4: (2, 2)}
+    try:
+        return layouts[nplots]
+    except KeyError:
+        k = 1
+        while k**2 < nplots:
+            k += 1
+
+        if (k - 1) * k >= nplots:
+            return k, (k - 1)
+        else:
+            return k, k
+
+
+# copied from matplotlib/pyplot.py and modified for pandas.plotting
+
+
+def create_subplots(
+    naxes: int,
+    sharex: bool = False,
+    sharey: bool = False,
+    squeeze: bool = True,
+    subplot_kw=None,
+    ax=None,
+    layout=None,
+    layout_type: str = "box",
+    **fig_kw,
+):
+    """
+    Create a figure with a set of subplots already made.
+
+    This utility wrapper makes it convenient to create common layouts of
+    subplots, including the enclosing figure object, in a single call.
+
+    Parameters
+    ----------
+    naxes : int
+      Number of required axes. Exceeded axes are set invisible. Default is
+      nrows * ncols.
+
+    sharex : bool
+      If True, the X axis will be shared amongst all subplots.
+
+    sharey : bool
+      If True, the Y axis will be shared amongst all subplots.
+
+    squeeze : bool
+
+      If True, extra dimensions are squeezed out from the returned axis object:
+        - if only one subplot is constructed (nrows=ncols=1), the resulting
+        single Axis object is returned as a scalar.
+        - for Nx1 or 1xN subplots, the returned object is a 1-d numpy object
+        array of Axis objects are returned as numpy 1-d arrays.
+        - for NxM subplots with N>1 and M>1 are returned as a 2d array.
+
+      If False, no squeezing is done: the returned axis object is always
+      a 2-d array containing Axis instances, even if it ends up being 1x1.
+
+    subplot_kw : dict
+      Dict with keywords passed to the add_subplot() call used to create each
+      subplots.
+
+    ax : Matplotlib axis object, optional
+
+    layout : tuple
+      Number of rows and columns of the subplot grid.
+      If not specified, calculated from naxes and layout_type
+
+    layout_type : {'box', 'horizontal', 'vertical'}, default 'box'
+      Specify how to layout the subplot grid.
+
+    fig_kw : Other keyword arguments to be passed to the figure() call.
+        Note that all keywords not recognized above will be
+        automatically included here.
+
+    Returns
+    -------
+    fig, ax : tuple
+      - fig is the Matplotlib Figure object
+      - ax can be either a single axis object or an array of axis objects if
+      more than one subplot was created.  The dimensions of the resulting array
+      can be controlled with the squeeze keyword, see above.
+
+    Examples
+    --------
+    x = np.linspace(0, 2*np.pi, 400)
+    y = np.sin(x**2)
+
+    # Just a figure and one subplot
+    f, ax = plt.subplots()
+    ax.plot(x, y)
+    ax.set_title('Simple plot')
+
+    # Two subplots, unpack the output array immediately
+    f, (ax1, ax2) = plt.subplots(1, 2, sharey=True)
+    ax1.plot(x, y)
+    ax1.set_title('Sharing Y axis')
+    ax2.scatter(x, y)
+
+    # Four polar axes
+    plt.subplots(2, 2, subplot_kw=dict(polar=True))
+    """
+    import matplotlib.pyplot as plt
+
+    if subplot_kw is None:
+        subplot_kw = {}
+
+    if ax is None:
+        fig = plt.figure(**fig_kw)
+    else:
+        if is_list_like(ax):
+            if squeeze:
+                ax = np.fromiter(flatten_axes(ax), dtype=object)
+            if layout is not None:
+                warnings.warn(
+                    "When passing multiple axes, layout keyword is ignored.",
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+            if sharex or sharey:
+                warnings.warn(
+                    "When passing multiple axes, sharex and sharey "
+                    "are ignored. These settings must be specified when creating axes.",
+                    UserWarning,
+                    stacklevel=find_stack_level(),
+                )
+            if ax.size == naxes:
+                fig = ax.flat[0].get_figure()
+                return fig, ax
+            else:
+                raise ValueError(
+                    f"The number of passed axes must be {naxes}, the "
+                    "same as the output plot"
+                )
+
+        fig = ax.get_figure()
+        # if ax is passed and a number of subplots is 1, return ax as it is
+        if naxes == 1:
+            if squeeze:
+                return fig, ax
+            else:
+                return fig, np.fromiter(flatten_axes(ax), dtype=object)
+        else:
+            warnings.warn(
+                "To output multiple subplots, the figure containing "
+                "the passed axes is being cleared.",
+                UserWarning,
+                stacklevel=find_stack_level(),
+            )
+            fig.clear()
+
+    nrows, ncols = _get_layout(naxes, layout=layout, layout_type=layout_type)
+    nplots = nrows * ncols
+
+    # Create empty object array to hold all axes.  It's easiest to make it 1-d
+    # so we can just append subplots upon creation, and then
+    axarr = np.empty(nplots, dtype=object)
+
+    # Create first subplot separately, so we can share it if requested
+    ax0 = fig.add_subplot(nrows, ncols, 1, **subplot_kw)
+
+    if sharex:
+        subplot_kw["sharex"] = ax0
+    if sharey:
+        subplot_kw["sharey"] = ax0
+    axarr[0] = ax0
+
+    # Note off-by-one counting because add_subplot uses the MATLAB 1-based
+    # convention.
+    for i in range(1, nplots):
+        kwds = subplot_kw.copy()
+        # Set sharex and sharey to None for blank/dummy axes, these can
+        # interfere with proper axis limits on the visible axes if
+        # they share axes e.g. issue #7528
+        if i >= naxes:
+            kwds["sharex"] = None
+            kwds["sharey"] = None
+        ax = fig.add_subplot(nrows, ncols, i + 1, **kwds)
+        axarr[i] = ax
+
+    if naxes != nplots:
+        for ax in axarr[naxes:]:
+            ax.set_visible(False)
+
+    handle_shared_axes(axarr, nplots, naxes, nrows, ncols, sharex, sharey)
+
+    if squeeze:
+        # Reshape the array to have the final desired dimension (nrow,ncol),
+        # though discarding unneeded dimensions that equal 1.  If we only have
+        # one subplot, just return it instead of a 1-element array.
+        if nplots == 1:
+            axes = axarr[0]
+        else:
+            axes = axarr.reshape(nrows, ncols).squeeze()
+    else:
+        # returned axis array will be always 2-d, even if nrows=ncols=1
+        axes = axarr.reshape(nrows, ncols)
+
+    return fig, axes
+
+
+def _remove_labels_from_axis(axis: Axis) -> None:
+    for t in axis.get_majorticklabels():
+        t.set_visible(False)
+
+    # set_visible will not be effective if
+    # minor axis has NullLocator and NullFormatter (default)
+    if isinstance(axis.get_minor_locator(), mpl.ticker.NullLocator):
+        axis.set_minor_locator(mpl.ticker.AutoLocator())
+    if isinstance(axis.get_minor_formatter(), mpl.ticker.NullFormatter):
+        axis.set_minor_formatter(mpl.ticker.FormatStrFormatter(""))
+    for t in axis.get_minorticklabels():
+        t.set_visible(False)
+
+    axis.get_label().set_visible(False)
+
+
+def _has_externally_shared_axis(ax1: Axes, compare_axis: str) -> bool:
+    """
+    Return whether an axis is externally shared.
+
+    Parameters
+    ----------
+    ax1 : matplotlib.axes.Axes
+        Axis to query.
+    compare_axis : str
+        `"x"` or `"y"` according to whether the X-axis or Y-axis is being
+        compared.
+
+    Returns
+    -------
+    bool
+        `True` if the axis is externally shared. Otherwise `False`.
+
+    Notes
+    -----
+    If two axes with different positions are sharing an axis, they can be
+    referred to as *externally* sharing the common axis.
+
+    If two axes sharing an axis also have the same position, they can be
+    referred to as *internally* sharing the common axis (a.k.a twinning).
+
+    _handle_shared_axes() is only interested in axes externally sharing an
+    axis, regardless of whether either of the axes is also internally sharing
+    with a third axis.
+    """
+    if compare_axis == "x":
+        axes = ax1.get_shared_x_axes()
+    elif compare_axis == "y":
+        axes = ax1.get_shared_y_axes()
+    else:
+        raise ValueError(
+            "_has_externally_shared_axis() needs 'x' or 'y' as a second parameter"
+        )
+
+    axes_siblings = axes.get_siblings(ax1)
+
+    # Retain ax1 and any of its siblings which aren't in the same position as it
+    ax1_points = ax1.get_position().get_points()
+
+    for ax2 in axes_siblings:
+        if not np.array_equal(ax1_points, ax2.get_position().get_points()):
+            return True
+
+    return False
+
+
+def handle_shared_axes(
+    axarr: Iterable[Axes],
+    nplots: int,
+    naxes: int,
+    nrows: int,
+    ncols: int,
+    sharex: bool,
+    sharey: bool,
+) -> None:
+    if nplots > 1:
+        row_num = lambda x: x.get_subplotspec().rowspan.start
+        col_num = lambda x: x.get_subplotspec().colspan.start
+
+        is_first_col = lambda x: x.get_subplotspec().is_first_col()
+
+        if nrows > 1:
+            try:
+                # first find out the ax layout,
+                # so that we can correctly handle 'gaps"
+                layout = np.zeros((nrows + 1, ncols + 1), dtype=np.bool_)
+                for ax in axarr:
+                    layout[row_num(ax), col_num(ax)] = ax.get_visible()
+
+                for ax in axarr:
+                    # only the last row of subplots should get x labels -> all
+                    # other off layout handles the case that the subplot is
+                    # the last in the column, because below is no subplot/gap.
+                    if not layout[row_num(ax) + 1, col_num(ax)]:
+                        continue
+                    if sharex or _has_externally_shared_axis(ax, "x"):
+                        _remove_labels_from_axis(ax.xaxis)
+
+            except IndexError:
+                # if gridspec is used, ax.rowNum and ax.colNum may different
+                # from layout shape. in this case, use last_row logic
+                is_last_row = lambda x: x.get_subplotspec().is_last_row()
+                for ax in axarr:
+                    if is_last_row(ax):
+                        continue
+                    if sharex or _has_externally_shared_axis(ax, "x"):
+                        _remove_labels_from_axis(ax.xaxis)
+
+        if ncols > 1:
+            for ax in axarr:
+                # only the first column should get y labels -> set all other to
+                # off as we only have labels in the first column and we always
+                # have a subplot there, we can skip the layout test
+                if is_first_col(ax):
+                    continue
+                if sharey or _has_externally_shared_axis(ax, "y"):
+                    _remove_labels_from_axis(ax.yaxis)
+
+
+def flatten_axes(axes: Axes | Iterable[Axes]) -> Generator[Axes]:
+    if not is_list_like(axes):
+        yield axes  # type: ignore[misc]
+    elif isinstance(axes, (np.ndarray, ABCIndex)):
+        yield from np.asarray(axes).reshape(-1)
+    else:
+        yield from axes  # type: ignore[misc]
+
+
+def set_ticks_props(
+    axes: Axes | Iterable[Axes],
+    xlabelsize: int | None = None,
+    xrot=None,
+    ylabelsize: int | None = None,
+    yrot=None,
+):
+    for ax in flatten_axes(axes):
+        if xlabelsize is not None:
+            mpl.artist.setp(ax.get_xticklabels(), fontsize=xlabelsize)  # type: ignore[arg-type]
+        if xrot is not None:
+            mpl.artist.setp(ax.get_xticklabels(), rotation=xrot)  # type: ignore[arg-type]
+        if ylabelsize is not None:
+            mpl.artist.setp(ax.get_yticklabels(), fontsize=ylabelsize)  # type: ignore[arg-type]
+        if yrot is not None:
+            mpl.artist.setp(ax.get_yticklabels(), rotation=yrot)  # type: ignore[arg-type]
+    return axes
+
+
+def get_all_lines(ax: Axes) -> list[Line2D]:
+    lines = ax.get_lines()
+
+    if hasattr(ax, "right_ax"):
+        lines += ax.right_ax.get_lines()
+
+    if hasattr(ax, "left_ax"):
+        lines += ax.left_ax.get_lines()
+
+    return lines
+
+
+def get_xlim(lines: Iterable[Line2D]) -> tuple[float, float]:
+    left, right = np.inf, -np.inf
+    for line in lines:
+        x = line.get_xdata(orig=False)
+        left = min(np.nanmin(x), left)
+        right = max(np.nanmax(x), right)
+    return left, right
diff --git a/pandas/pandas/plotting/_misc.py b/pandas/pandas/plotting/_misc.py
new file mode 100644
index 0000000000000..0e0fb23d924bc
--- /dev/null
+++ b/pandas/pandas/plotting/_misc.py
@@ -0,0 +1,759 @@
+from __future__ import annotations
+
+from contextlib import contextmanager
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+from pandas.plotting._core import _get_plot_backend
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Generator,
+        Mapping,
+    )
+
+    from matplotlib.axes import Axes
+    from matplotlib.colors import Colormap
+    from matplotlib.figure import Figure
+    from matplotlib.table import Table
+    import numpy as np
+
+    from pandas import (
+        DataFrame,
+        Series,
+    )
+
+
+def table(ax: Axes, data: DataFrame | Series, **kwargs) -> Table:
+    """
+    Helper function to convert DataFrame and Series to matplotlib.table.
+
+    This method provides an easy way to visualize tabular data within a Matplotlib
+    figure. It automatically extracts index and column labels from the DataFrame
+    or Series, unless explicitly specified. This function is particularly useful
+    when displaying summary tables alongside other plots or when creating static
+    reports. It utilizes the `matplotlib.pyplot.table` backend and allows
+    customization through various styling options available in Matplotlib.
+
+    Parameters
+    ----------
+    ax : Matplotlib axes object
+        The axes on which to draw the table.
+    data : DataFrame or Series
+        Data for table contents.
+    **kwargs
+        Keyword arguments to be passed to matplotlib.table.table.
+        If `rowLabels` or `colLabels` is not specified, data index or column
+        names will be used.
+
+    Returns
+    -------
+    matplotlib table object
+        The created table as a matplotlib Table object.
+
+    See Also
+    --------
+    DataFrame.plot : Make plots of DataFrame using matplotlib.
+    matplotlib.pyplot.table : Create a table from data in a Matplotlib plot.
+
+    Examples
+    --------
+
+    .. plot::
+            :context: close-figs
+
+            >>> import matplotlib.pyplot as plt
+            >>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+            >>> fig, ax = plt.subplots()
+            >>> ax.axis("off")
+            (0.0, 1.0, 0.0, 1.0)
+            >>> table = pd.plotting.table(
+            ...     ax, df, loc="center", cellLoc="center", colWidths=[0.2, 0.2]
+            ... )
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.table(
+        ax=ax, data=data, rowLabels=None, colLabels=None, **kwargs
+    )
+
+
+def register() -> None:
+    """
+    Register pandas formatters and converters with matplotlib.
+
+    This function modifies the global ``matplotlib.units.registry``
+    dictionary. pandas adds custom converters for
+
+    * pd.Timestamp
+    * pd.Period
+    * np.datetime64
+    * datetime.datetime
+    * datetime.date
+    * datetime.time
+
+    See Also
+    --------
+    deregister_matplotlib_converters : Remove pandas formatters and converters.
+
+    Examples
+    --------
+    .. plot::
+       :context: close-figs
+
+        The following line is done automatically by pandas so
+        the plot can be rendered:
+
+        >>> pd.plotting.register_matplotlib_converters()
+
+        >>> df = pd.DataFrame(
+        ...     {"ts": pd.period_range("2020", periods=2, freq="M"), "y": [1, 2]}
+        ... )
+        >>> plot = df.plot.line(x="ts", y="y")
+
+    Unsetting the register manually an error will be raised:
+
+    >>> pd.set_option(
+    ...     "plotting.matplotlib.register_converters", False
+    ... )  # doctest: +SKIP
+    >>> df.plot.line(x="ts", y="y")  # doctest: +SKIP
+    Traceback (most recent call last):
+    TypeError: float() argument must be a string or a real number, not 'Period'
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    plot_backend.register()
+
+
+def deregister() -> None:
+    """
+    Remove pandas formatters and converters.
+
+    Removes the custom converters added by :func:`register`. This
+    attempts to set the state of the registry back to the state before
+    pandas registered its own units. Converters for pandas' own types like
+    Timestamp and Period are removed completely. Converters for types
+    pandas overwrites, like ``datetime.datetime``, are restored to their
+    original value.
+
+    See Also
+    --------
+    register_matplotlib_converters : Register pandas formatters and converters
+        with matplotlib.
+
+    Examples
+    --------
+    .. plot::
+       :context: close-figs
+
+        The following line is done automatically by pandas so
+        the plot can be rendered:
+
+        >>> pd.plotting.register_matplotlib_converters()
+
+        >>> df = pd.DataFrame(
+        ...     {"ts": pd.period_range("2020", periods=2, freq="M"), "y": [1, 2]}
+        ... )
+        >>> plot = df.plot.line(x="ts", y="y")
+
+    Unsetting the register manually an error will be raised:
+
+    >>> pd.set_option(
+    ...     "plotting.matplotlib.register_converters", False
+    ... )  # doctest: +SKIP
+    >>> df.plot.line(x="ts", y="y")  # doctest: +SKIP
+    Traceback (most recent call last):
+    TypeError: float() argument must be a string or a real number, not 'Period'
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    plot_backend.deregister()
+
+
+def scatter_matrix(
+    frame: DataFrame,
+    alpha: float = 0.5,
+    figsize: tuple[float, float] | None = None,
+    ax: Axes | None = None,
+    grid: bool = False,
+    diagonal: str = "hist",
+    marker: str = ".",
+    density_kwds: Mapping[str, Any] | None = None,
+    hist_kwds: Mapping[str, Any] | None = None,
+    range_padding: float = 0.05,
+    **kwargs,
+) -> np.ndarray:
+    """
+    Draw a matrix of scatter plots.
+
+    Each pair of numeric columns in the DataFrame is plotted against each other,
+    resulting in a matrix of scatter plots. The diagonal plots can display either
+    histograms or Kernel Density Estimation (KDE) plots for each variable.
+
+    Parameters
+    ----------
+    frame : DataFrame
+        The data to be plotted.
+    alpha : float, optional
+        Amount of transparency applied.
+    figsize : (float,float), optional
+        A tuple (width, height) in inches.
+    ax : Matplotlib axis object, optional
+        An existing Matplotlib axis object for the plots. If None, a new axis is
+        created.
+    grid : bool, optional
+        Setting this to True will show the grid.
+    diagonal : {'hist', 'kde'}
+        Pick between 'kde' and 'hist' for either Kernel Density Estimation or
+        Histogram plot in the diagonal.
+    marker : str, optional
+        Matplotlib marker type, default '.'.
+    density_kwds : keywords
+        Keyword arguments to be passed to kernel density estimate plot.
+    hist_kwds : keywords
+        Keyword arguments to be passed to hist function.
+    range_padding : float, default 0.05
+        Relative extension of axis range in x and y with respect to
+        (x_max - x_min) or (y_max - y_min).
+    **kwargs
+        Keyword arguments to be passed to scatter function.
+
+    Returns
+    -------
+    numpy.ndarray
+        A matrix of scatter plots.
+
+    See Also
+    --------
+    plotting.parallel_coordinates : Plots parallel coordinates for multivariate data.
+    plotting.andrews_curves : Generates Andrews curves for visualizing clusters of
+        multivariate data.
+    plotting.radviz : Creates a RadViz visualization.
+    plotting.bootstrap_plot : Visualizes uncertainty in data via bootstrap sampling.
+
+    Examples
+    --------
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.DataFrame(np.random.randn(1000, 4), columns=["A", "B", "C", "D"])
+        >>> pd.plotting.scatter_matrix(df, alpha=0.2)
+        array([[<Axes: xlabel='A', ylabel='A'>, <Axes: xlabel='B', ylabel='A'>,
+                <Axes: xlabel='C', ylabel='A'>, <Axes: xlabel='D', ylabel='A'>],
+               [<Axes: xlabel='A', ylabel='B'>, <Axes: xlabel='B', ylabel='B'>,
+                <Axes: xlabel='C', ylabel='B'>, <Axes: xlabel='D', ylabel='B'>],
+               [<Axes: xlabel='A', ylabel='C'>, <Axes: xlabel='B', ylabel='C'>,
+                <Axes: xlabel='C', ylabel='C'>, <Axes: xlabel='D', ylabel='C'>],
+               [<Axes: xlabel='A', ylabel='D'>, <Axes: xlabel='B', ylabel='D'>,
+                <Axes: xlabel='C', ylabel='D'>, <Axes: xlabel='D', ylabel='D'>]],
+              dtype=object)
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.scatter_matrix(
+        frame=frame,
+        alpha=alpha,
+        figsize=figsize,
+        ax=ax,
+        grid=grid,
+        diagonal=diagonal,
+        marker=marker,
+        density_kwds=density_kwds,
+        hist_kwds=hist_kwds,
+        range_padding=range_padding,
+        **kwargs,
+    )
+
+
+def radviz(
+    frame: DataFrame,
+    class_column: str,
+    ax: Axes | None = None,
+    color: list[str] | tuple[str, ...] | None = None,
+    colormap: Colormap | str | None = None,
+    **kwds,
+) -> Axes:
+    """
+    Plot a multidimensional dataset in 2D.
+
+    Each Series in the DataFrame is represented as a evenly distributed
+    slice on a circle. Each data point is rendered in the circle according to
+    the value on each Series. Highly correlated `Series` in the `DataFrame`
+    are placed closer on the unit circle.
+
+    RadViz allow to project a N-dimensional data set into a 2D space where the
+    influence of each dimension can be interpreted as a balance between the
+    influence of all dimensions.
+
+    More info available at the `original article
+    <https://doi.org/10.1145/331770.331775>`_
+    describing RadViz.
+
+    Parameters
+    ----------
+    frame : `DataFrame`
+        Object holding the data.
+    class_column : str
+        Column name containing the name of the data point category.
+    ax : :class:`matplotlib.axes.Axes`, optional
+        A plot instance to which to add the information.
+    color : list[str] or tuple[str], optional
+        Assign a color to each category. Example: ['blue', 'green'].
+    colormap : str or :class:`matplotlib.colors.Colormap`, default None
+        Colormap to select colors from. If string, load colormap with that
+        name from matplotlib.
+    **kwds
+        Options to pass to matplotlib scatter plotting method.
+
+    Returns
+    -------
+    :class:`matplotlib.axes.Axes`
+        The Axes object from Matplotlib.
+
+    See Also
+    --------
+    plotting.andrews_curves : Plot clustering visualization.
+
+    Examples
+    --------
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.DataFrame(
+        ...     {
+        ...         "SepalLength": [6.5, 7.7, 5.1, 5.8, 7.6, 5.0, 5.4, 4.6, 6.7, 4.6],
+        ...         "SepalWidth": [3.0, 3.8, 3.8, 2.7, 3.0, 2.3, 3.0, 3.2, 3.3, 3.6],
+        ...         "PetalLength": [5.5, 6.7, 1.9, 5.1, 6.6, 3.3, 4.5, 1.4, 5.7, 1.0],
+        ...         "PetalWidth": [1.8, 2.2, 0.4, 1.9, 2.1, 1.0, 1.5, 0.2, 2.1, 0.2],
+        ...         "Category": [
+        ...             "virginica",
+        ...             "virginica",
+        ...             "setosa",
+        ...             "virginica",
+        ...             "virginica",
+        ...             "versicolor",
+        ...             "versicolor",
+        ...             "setosa",
+        ...             "virginica",
+        ...             "setosa",
+        ...         ],
+        ...     }
+        ... )
+        >>> pd.plotting.radviz(df, "Category")  # doctest: +SKIP
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.radviz(
+        frame=frame,
+        class_column=class_column,
+        ax=ax,
+        color=color,
+        colormap=colormap,
+        **kwds,
+    )
+
+
+def andrews_curves(
+    frame: DataFrame,
+    class_column: str,
+    ax: Axes | None = None,
+    samples: int = 200,
+    color: list[str] | tuple[str, ...] | None = None,
+    colormap: Colormap | str | None = None,
+    **kwargs,
+) -> Axes:
+    """
+    Generate a matplotlib plot for visualizing clusters of multivariate data.
+
+    Andrews curves have the functional form:
+
+    .. math::
+        f(t) = \\frac{x_1}{\\sqrt{2}} + x_2 \\sin(t) + x_3 \\cos(t) +
+        x_4 \\sin(2t) + x_5 \\cos(2t) + \\cdots
+
+    Where :math:`x` coefficients correspond to the values of each dimension
+    and :math:`t` is linearly spaced between :math:`-\\pi` and :math:`+\\pi`.
+    Each row of frame then corresponds to a single curve.
+
+    Parameters
+    ----------
+    frame : DataFrame
+        Data to be plotted, preferably normalized to (0.0, 1.0).
+    class_column : label
+        Name of the column containing class names.
+    ax : axes object, default None
+        Axes to use.
+    samples : int
+        Number of points to plot in each curve.
+    color : str, list[str] or tuple[str], optional
+        Colors to use for the different classes. Colors can be strings
+        or 3-element floating point RGB values.
+    colormap : str or matplotlib colormap object, default None
+        Colormap to select colors from. If a string, load colormap with that
+        name from matplotlib.
+    **kwargs
+        Options to pass to matplotlib plotting method.
+
+    Returns
+    -------
+    :class:`matplotlib.axes.Axes`
+        The matplotlib Axes object with the plot.
+
+    See Also
+    --------
+    plotting.parallel_coordinates : Plot parallel coordinates chart.
+    DataFrame.plot : Make plots of Series or DataFrame.
+
+    Examples
+    --------
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.read_csv(
+        ...     "https://raw.githubusercontent.com/pandas-dev/"
+        ...     "pandas/main/pandas/tests/io/data/csv/iris.csv"
+        ... )
+        >>> pd.plotting.andrews_curves(df, "Name")  # doctest: +SKIP
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.andrews_curves(
+        frame=frame,
+        class_column=class_column,
+        ax=ax,
+        samples=samples,
+        color=color,
+        colormap=colormap,
+        **kwargs,
+    )
+
+
+def bootstrap_plot(
+    series: Series,
+    fig: Figure | None = None,
+    size: int = 50,
+    samples: int = 500,
+    **kwds,
+) -> Figure:
+    """
+    Bootstrap plot on mean, median and mid-range statistics.
+
+    The bootstrap plot is used to estimate the uncertainty of a statistic
+    by relying on random sampling with replacement [1]_. This function will
+    generate bootstrapping plots for mean, median and mid-range statistics
+    for the given number of samples of the given size.
+
+    .. [1] "Bootstrapping (statistics)" in \
+    https://en.wikipedia.org/wiki/Bootstrapping_%28statistics%29
+
+    Parameters
+    ----------
+    series : pandas.Series
+        Series from where to get the samplings for the bootstrapping.
+    fig : matplotlib.figure.Figure, default None
+        If given, it will use the `fig` reference for plotting instead of
+        creating a new one with default parameters.
+    size : int, default 50
+        Number of data points to consider during each sampling. It must be
+        less than or equal to the length of the `series`.
+    samples : int, default 500
+        Number of times the bootstrap procedure is performed.
+    **kwds
+        Options to pass to matplotlib plotting method.
+
+    Returns
+    -------
+    matplotlib.figure.Figure
+        Matplotlib figure.
+
+    See Also
+    --------
+    DataFrame.plot : Basic plotting for DataFrame objects.
+    Series.plot : Basic plotting for Series objects.
+
+    Examples
+    --------
+    This example draws a basic bootstrap plot for a Series.
+
+    .. plot::
+        :context: close-figs
+
+        >>> s = pd.Series(np.random.uniform(size=100))
+        >>> pd.plotting.bootstrap_plot(s)  # doctest: +SKIP
+        <Figure size 640x480 with 6 Axes>
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.bootstrap_plot(
+        series=series, fig=fig, size=size, samples=samples, **kwds
+    )
+
+
+def parallel_coordinates(
+    frame: DataFrame,
+    class_column: str,
+    cols: list[str] | None = None,
+    ax: Axes | None = None,
+    color: list[str] | tuple[str, ...] | None = None,
+    use_columns: bool = False,
+    xticks: list | tuple | None = None,
+    colormap: Colormap | str | None = None,
+    axvlines: bool = True,
+    axvlines_kwds: Mapping[str, Any] | None = None,
+    sort_labels: bool = False,
+    **kwargs,
+) -> Axes:
+    """
+    Parallel coordinates plotting.
+
+    Parameters
+    ----------
+    frame : DataFrame
+        The DataFrame to be plotted.
+    class_column : str
+        Column name containing class names.
+    cols : list, optional
+        A list of column names to use.
+    ax : matplotlib.axis, optional
+        Matplotlib axis object.
+    color : list or tuple, optional
+        Colors to use for the different classes.
+    use_columns : bool, optional
+        If true, columns will be used as xticks.
+    xticks : list or tuple, optional
+        A list of values to use for xticks.
+    colormap : str or matplotlib colormap, default None
+        Colormap to use for line colors.
+    axvlines : bool, optional
+        If true, vertical lines will be added at each xtick.
+    axvlines_kwds : keywords, optional
+        Options to be passed to axvline method for vertical lines.
+    sort_labels : bool, default False
+        Sort class_column labels, useful when assigning colors.
+    **kwargs
+        Options to pass to matplotlib plotting method.
+
+    Returns
+    -------
+    matplotlib.axes.Axes
+        The matplotlib axes containing the parallel coordinates plot.
+
+    See Also
+    --------
+    plotting.andrews_curves : Generate a matplotlib plot for visualizing clusters
+        of multivariate data.
+    plotting.radviz : Plot a multidimensional dataset in 2D.
+
+    Examples
+    --------
+
+    .. plot::
+        :context: close-figs
+
+        >>> df = pd.read_csv(
+        ...     "https://raw.githubusercontent.com/pandas-dev/"
+        ...     "pandas/main/pandas/tests/io/data/csv/iris.csv"
+        ... )
+        >>> pd.plotting.parallel_coordinates(
+        ...     df, "Name", color=("#556270", "#4ECDC4", "#C7F464")
+        ... )  # doctest: +SKIP
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.parallel_coordinates(
+        frame=frame,
+        class_column=class_column,
+        cols=cols,
+        ax=ax,
+        color=color,
+        use_columns=use_columns,
+        xticks=xticks,
+        colormap=colormap,
+        axvlines=axvlines,
+        axvlines_kwds=axvlines_kwds,
+        sort_labels=sort_labels,
+        **kwargs,
+    )
+
+
+def lag_plot(series: Series, lag: int = 1, ax: Axes | None = None, **kwds) -> Axes:
+    """
+    Lag plot for time series.
+
+    A lag plot is a scatter plot of a time series against a lag of itself. It helps
+    in visualizing the temporal dependence between observations by plotting the values
+    at time `t` on the x-axis and the values at time `t + lag` on the y-axis.
+
+    Parameters
+    ----------
+    series : Series
+        The time series to visualize.
+    lag : int, default 1
+        Lag length of the scatter plot.
+    ax : Matplotlib axis object, optional
+        The matplotlib axis object to use.
+    **kwds
+        Matplotlib scatter method keyword arguments.
+
+    Returns
+    -------
+    matplotlib.axes.Axes
+        The matplotlib Axes object containing the lag plot.
+
+    See Also
+    --------
+    plotting.autocorrelation_plot : Autocorrelation plot for time series.
+    matplotlib.pyplot.scatter : A scatter plot of y vs. x with varying marker size
+        and/or color in Matplotlib.
+
+    Examples
+    --------
+    Lag plots are most commonly used to look for patterns in time series data.
+
+    Given the following time series
+
+    .. plot::
+        :context: close-figs
+
+        >>> np.random.seed(5)
+        >>> x = np.cumsum(np.random.normal(loc=1, scale=5, size=50))
+        >>> s = pd.Series(x)
+        >>> s.plot()  # doctest: +SKIP
+
+    A lag plot with ``lag=1`` returns
+
+    .. plot::
+        :context: close-figs
+
+        >>> pd.plotting.lag_plot(s, lag=1)
+        <Axes: xlabel='y(t)', ylabel='y(t + 1)'>
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.lag_plot(series=series, lag=lag, ax=ax, **kwds)
+
+
+def autocorrelation_plot(series: Series, ax: Axes | None = None, **kwargs) -> Axes:
+    """
+    Autocorrelation plot for time series.
+
+    Parameters
+    ----------
+    series : Series
+        The time series to visualize.
+    ax : Matplotlib axis object, optional
+        The matplotlib axis object to use.
+    **kwargs
+        Options to pass to matplotlib plotting method.
+
+    Returns
+    -------
+    matplotlib.axes.Axes
+        The matplotlib axes containing the autocorrelation plot.
+
+    See Also
+    --------
+    Series.autocorr : Compute the lag-N autocorrelation for a Series.
+    plotting.lag_plot : Lag plot for time series.
+
+    Examples
+    --------
+    The horizontal lines in the plot correspond to 95% and 99% confidence bands.
+
+    The dashed line is 99% confidence band.
+
+    .. plot::
+        :context: close-figs
+
+        >>> spacing = np.linspace(-9 * np.pi, 9 * np.pi, num=1000)
+        >>> s = pd.Series(0.7 * np.random.rand(1000) + 0.3 * np.sin(spacing))
+        >>> pd.plotting.autocorrelation_plot(s)  # doctest: +SKIP
+    """
+    plot_backend = _get_plot_backend("matplotlib")
+    return plot_backend.autocorrelation_plot(series=series, ax=ax, **kwargs)
+
+
+class _Options(dict):
+    """
+    Stores pandas plotting options.
+
+    Allows for parameter aliasing so you can just use parameter names that are
+    the same as the plot function parameters, but is stored in a canonical
+    format that makes it easy to breakdown into groups later.
+
+    See Also
+    --------
+    plotting.register_matplotlib_converters : Register pandas formatters and
+        converters with matplotlib.
+    plotting.bootstrap_plot : Bootstrap plot on mean, median and mid-range statistics.
+    plotting.autocorrelation_plot : Autocorrelation plot for time series.
+    plotting.lag_plot : Lag plot for time series.
+
+    Examples
+    --------
+
+    .. plot::
+            :context: close-figs
+
+             >>> np.random.seed(42)
+             >>> df = pd.DataFrame(
+             ...     {"A": np.random.randn(10), "B": np.random.randn(10)},
+             ...     index=pd.date_range("1/1/2000", freq="4MS", periods=10),
+             ... )
+             >>> with pd.plotting.plot_params.use("x_compat", True):
+             ...     _ = df["A"].plot(color="r")
+             ...     _ = df["B"].plot(color="g")
+    """
+
+    # alias so the names are same as plotting method parameter names
+    _ALIASES = {"x_compat": "xaxis.compat"}
+    _DEFAULT_KEYS = ["xaxis.compat"]
+
+    def __init__(self) -> None:
+        super().__setitem__("xaxis.compat", False)
+
+    def __getitem__(self, key):
+        key = self._get_canonical_key(key)
+        if key not in self:
+            raise ValueError(f"{key} is not a valid pandas plotting option")
+        return super().__getitem__(key)
+
+    def __setitem__(self, key, value) -> None:
+        key = self._get_canonical_key(key)
+        super().__setitem__(key, value)
+
+    def __delitem__(self, key) -> None:
+        key = self._get_canonical_key(key)
+        if key in self._DEFAULT_KEYS:
+            raise ValueError(f"Cannot remove default parameter {key}")
+        super().__delitem__(key)
+
+    def __contains__(self, key) -> bool:
+        key = self._get_canonical_key(key)
+        return super().__contains__(key)
+
+    def reset(self) -> None:
+        """
+        Reset the option store to its initial state
+
+        Returns
+        -------
+        None
+        """
+        # error: Cannot access "__init__" directly
+        self.__init__()  # type: ignore[misc]
+
+    def _get_canonical_key(self, key: str) -> str:
+        return self._ALIASES.get(key, key)
+
+    @contextmanager
+    def use(self, key, value) -> Generator[_Options]:
+        """
+        Temporarily set a parameter value using the with statement.
+        Aliasing allowed.
+        """
+        old_value = self[key]
+        try:
+            self[key] = value
+            yield self
+        finally:
+            self[key] = old_value
+
+
+plot_params = _Options()
diff --git a/pandas/pandas/testing.py b/pandas/pandas/testing.py
new file mode 100644
index 0000000000000..433b22bf1107e
--- /dev/null
+++ b/pandas/pandas/testing.py
@@ -0,0 +1,17 @@
+"""
+Public testing utility functions.
+"""
+
+from pandas._testing import (
+    assert_extension_array_equal,
+    assert_frame_equal,
+    assert_index_equal,
+    assert_series_equal,
+)
+
+__all__ = [
+    "assert_extension_array_equal",
+    "assert_frame_equal",
+    "assert_index_equal",
+    "assert_series_equal",
+]
diff --git a/pandas/pandas/tests/__init__.py b/pandas/pandas/tests/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/api/__init__.py b/pandas/pandas/tests/api/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/api/test_api.py b/pandas/pandas/tests/api/test_api.py
new file mode 100644
index 0000000000000..4a05259a98087
--- /dev/null
+++ b/pandas/pandas/tests/api/test_api.py
@@ -0,0 +1,436 @@
+from __future__ import annotations
+
+import pytest
+
+import pandas as pd
+from pandas import api
+import pandas._testing as tm
+from pandas.api import (
+    extensions as api_extensions,
+    indexers as api_indexers,
+    interchange as api_interchange,
+    types as api_types,
+    typing as api_typing,
+)
+
+
+class Base:
+    def check(self, namespace, expected, ignored=None):
+        # see which names are in the namespace, minus optional
+        # ignored ones
+        # compare vs the expected
+
+        result = sorted(
+            f for f in dir(namespace) if not f.startswith("__") and f != "annotations"
+        )
+        if ignored is not None:
+            result = sorted(set(result) - set(ignored))
+
+        expected = sorted(expected)
+        tm.assert_almost_equal(result, expected)
+
+
+class TestPDApi(Base):
+    # these are optionally imported based on testing
+    # & need to be ignored
+    ignored = ["tests", "locale", "conftest", "_version_meson"]
+
+    # top-level sub-packages
+    public_lib = [
+        "api",
+        "arrays",
+        "options",
+        "test",
+        "testing",
+        "errors",
+        "plotting",
+        "io",
+        "tseries",
+    ]
+    private_lib = ["compat", "core", "pandas", "util", "_built_with_meson"]
+
+    # misc
+    misc = ["IndexSlice", "NaT", "NA"]
+
+    # top-level classes
+    classes = [
+        "ArrowDtype",
+        "Categorical",
+        "CategoricalIndex",
+        "DataFrame",
+        "DateOffset",
+        "DatetimeIndex",
+        "ExcelFile",
+        "ExcelWriter",
+        "Flags",
+        "Grouper",
+        "HDFStore",
+        "Index",
+        "MultiIndex",
+        "Period",
+        "PeriodIndex",
+        "RangeIndex",
+        "Series",
+        "SparseDtype",
+        "StringDtype",
+        "Timedelta",
+        "TimedeltaIndex",
+        "Timestamp",
+        "Interval",
+        "IntervalIndex",
+        "CategoricalDtype",
+        "PeriodDtype",
+        "IntervalDtype",
+        "DatetimeTZDtype",
+        "BooleanDtype",
+        "Int8Dtype",
+        "Int16Dtype",
+        "Int32Dtype",
+        "Int64Dtype",
+        "UInt8Dtype",
+        "UInt16Dtype",
+        "UInt32Dtype",
+        "UInt64Dtype",
+        "Float32Dtype",
+        "Float64Dtype",
+        "NamedAgg",
+    ]
+
+    # these are already deprecated; awaiting removal
+    deprecated_classes: list[str] = []
+
+    # external modules exposed in pandas namespace
+    modules: list[str] = []
+
+    # top-level functions
+    funcs = [
+        "array",
+        "bdate_range",
+        "concat",
+        "crosstab",
+        "cut",
+        "date_range",
+        "interval_range",
+        "eval",
+        "factorize",
+        "get_dummies",
+        "from_dummies",
+        "infer_freq",
+        "isna",
+        "isnull",
+        "lreshape",
+        "melt",
+        "notna",
+        "notnull",
+        "offsets",
+        "merge",
+        "merge_ordered",
+        "merge_asof",
+        "period_range",
+        "pivot",
+        "pivot_table",
+        "qcut",
+        "show_versions",
+        "timedelta_range",
+        "unique",
+        "wide_to_long",
+    ]
+
+    # top-level option funcs
+    funcs_option = [
+        "reset_option",
+        "describe_option",
+        "get_option",
+        "option_context",
+        "set_option",
+        "set_eng_float_format",
+    ]
+
+    # top-level read_* funcs
+    funcs_read = [
+        "read_clipboard",
+        "read_csv",
+        "read_excel",
+        "read_fwf",
+        "read_hdf",
+        "read_html",
+        "read_xml",
+        "read_json",
+        "read_pickle",
+        "read_sas",
+        "read_sql",
+        "read_sql_query",
+        "read_sql_table",
+        "read_stata",
+        "read_table",
+        "read_feather",
+        "read_parquet",
+        "read_orc",
+        "read_spss",
+    ]
+
+    # top-level json funcs
+    funcs_json = ["json_normalize"]
+
+    # top-level to_* funcs
+    funcs_to = ["to_datetime", "to_numeric", "to_pickle", "to_timedelta"]
+
+    # top-level to deprecate in the future
+    deprecated_funcs_in_future: list[str] = []
+
+    # these are already deprecated; awaiting removal
+    deprecated_funcs: list[str] = []
+
+    # private modules in pandas namespace
+    private_modules = [
+        "_config",
+        "_libs",
+        "_is_numpy_dev",
+        "_pandas_datetime_CAPI",
+        "_pandas_parser_CAPI",
+        "_testing",
+        "_typing",
+    ]
+    if not pd._built_with_meson:
+        private_modules.append("_version")
+
+    def test_api(self):
+        checkthese = (
+            self.public_lib
+            + self.private_lib
+            + self.misc
+            + self.modules
+            + self.classes
+            + self.funcs
+            + self.funcs_option
+            + self.funcs_read
+            + self.funcs_json
+            + self.funcs_to
+            + self.private_modules
+        )
+        self.check(namespace=pd, expected=checkthese, ignored=self.ignored)
+
+    def test_api_all(self):
+        expected = set(
+            self.public_lib
+            + self.misc
+            + self.modules
+            + self.classes
+            + self.funcs
+            + self.funcs_option
+            + self.funcs_read
+            + self.funcs_json
+            + self.funcs_to
+        ) - set(self.deprecated_classes)
+        actual = set(pd.__all__)
+
+        extraneous = actual - expected
+        assert not extraneous
+
+        missing = expected - actual
+        assert not missing
+
+    def test_depr(self):
+        deprecated_list = (
+            self.deprecated_classes
+            + self.deprecated_funcs
+            + self.deprecated_funcs_in_future
+        )
+        for depr in deprecated_list:
+            with tm.assert_produces_warning(FutureWarning):
+                _ = getattr(pd, depr)
+
+
+class TestApi(Base):
+    allowed_api_dirs = [
+        "types",
+        "extensions",
+        "indexers",
+        "interchange",
+        "typing",
+        "internals",
+    ]
+    allowed_typing = [
+        "DataFrameGroupBy",
+        "DatetimeIndexResamplerGroupby",
+        "Expanding",
+        "ExpandingGroupby",
+        "ExponentialMovingWindow",
+        "ExponentialMovingWindowGroupby",
+        "FrozenList",
+        "JsonReader",
+        "NaTType",
+        "NAType",
+        "NoDefault",
+        "PeriodIndexResamplerGroupby",
+        "Resampler",
+        "Rolling",
+        "RollingGroupby",
+        "SeriesGroupBy",
+        "StataReader",
+        "SASReader",
+        "TimedeltaIndexResamplerGroupby",
+        "TimeGrouper",
+        "Window",
+    ]
+    allowed_api_types = [
+        "is_any_real_numeric_dtype",
+        "is_array_like",
+        "is_bool",
+        "is_bool_dtype",
+        "is_categorical_dtype",
+        "is_complex",
+        "is_complex_dtype",
+        "is_datetime64_any_dtype",
+        "is_datetime64_dtype",
+        "is_datetime64_ns_dtype",
+        "is_datetime64tz_dtype",
+        "is_dict_like",
+        "is_dtype_equal",
+        "is_extension_array_dtype",
+        "is_file_like",
+        "is_float",
+        "is_float_dtype",
+        "is_hashable",
+        "is_int64_dtype",
+        "is_integer",
+        "is_integer_dtype",
+        "is_interval_dtype",
+        "is_iterator",
+        "is_list_like",
+        "is_named_tuple",
+        "is_number",
+        "is_numeric_dtype",
+        "is_object_dtype",
+        "is_period_dtype",
+        "is_re",
+        "is_re_compilable",
+        "is_scalar",
+        "is_signed_integer_dtype",
+        "is_sparse",
+        "is_string_dtype",
+        "is_timedelta64_dtype",
+        "is_timedelta64_ns_dtype",
+        "is_unsigned_integer_dtype",
+        "pandas_dtype",
+        "infer_dtype",
+        "union_categoricals",
+        "CategoricalDtype",
+        "DatetimeTZDtype",
+        "IntervalDtype",
+        "PeriodDtype",
+    ]
+    allowed_api_interchange = ["from_dataframe", "DataFrame"]
+    allowed_api_indexers = [
+        "check_array_indexer",
+        "BaseIndexer",
+        "FixedForwardWindowIndexer",
+        "VariableOffsetWindowIndexer",
+    ]
+    allowed_api_extensions = [
+        "no_default",
+        "ExtensionDtype",
+        "register_extension_dtype",
+        "register_dataframe_accessor",
+        "register_index_accessor",
+        "register_series_accessor",
+        "take",
+        "ExtensionArray",
+        "ExtensionScalarOpsMixin",
+    ]
+
+    def test_api(self):
+        self.check(api, self.allowed_api_dirs)
+
+    def test_api_typing(self):
+        self.check(api_typing, self.allowed_typing)
+
+    def test_api_types(self):
+        self.check(api_types, self.allowed_api_types)
+
+    def test_api_interchange(self):
+        self.check(api_interchange, self.allowed_api_interchange)
+
+    def test_api_indexers(self):
+        self.check(api_indexers, self.allowed_api_indexers)
+
+    def test_api_extensions(self):
+        self.check(api_extensions, self.allowed_api_extensions)
+
+
+class TestErrors(Base):
+    def test_errors(self):
+        self.check(pd.errors, pd.errors.__all__, ignored=["ctypes", "cow"])
+
+
+class TestUtil(Base):
+    def test_util(self):
+        self.check(
+            pd.util,
+            ["hash_array", "hash_pandas_object"],
+            ignored=[
+                "_decorators",
+                "_test_decorators",
+                "_exceptions",
+                "_validators",
+                "capitalize_first_letter",
+                "version",
+                "_print_versions",
+                "_tester",
+            ],
+        )
+
+
+class TestTesting(Base):
+    funcs = [
+        "assert_frame_equal",
+        "assert_series_equal",
+        "assert_index_equal",
+        "assert_extension_array_equal",
+    ]
+
+    def test_testing(self):
+        from pandas import testing
+
+        self.check(testing, self.funcs)
+
+    def test_util_in_top_level(self):
+        with pytest.raises(AttributeError, match="foo"):
+            pd.util.foo
+
+
+def test_set_module():
+    assert pd.DataFrame.__module__ == "pandas"
+    assert pd.CategoricalDtype.__module__ == "pandas"
+    assert pd.PeriodDtype.__module__ == "pandas"
+    assert pd.IntervalDtype.__module__ == "pandas"
+    assert pd.SparseDtype.__module__ == "pandas"
+    assert pd.ArrowDtype.__module__ == "pandas"
+    assert pd.StringDtype.__module__ == "pandas"
+    assert pd.Index.__module__ == "pandas"
+    assert pd.CategoricalIndex.__module__ == "pandas"
+    assert pd.DatetimeIndex.__module__ == "pandas"
+    assert pd.IntervalIndex.__module__ == "pandas"
+    assert pd.MultiIndex.__module__ == "pandas"
+    assert pd.PeriodIndex.__module__ == "pandas"
+    assert pd.RangeIndex.__module__ == "pandas"
+    assert pd.TimedeltaIndex.__module__ == "pandas"
+    assert pd.Period.__module__ == "pandas"
+    assert pd.Timestamp.__module__ == "pandas"
+    assert pd.Timedelta.__module__ == "pandas"
+    assert pd.concat.__module__ == "pandas"
+    assert pd.isna.__module__ == "pandas"
+    assert pd.notna.__module__ == "pandas"
+    assert pd.merge.__module__ == "pandas"
+    assert pd.merge_ordered.__module__ == "pandas"
+    assert pd.merge_asof.__module__ == "pandas"
+    assert pd.read_csv.__module__ == "pandas"
+    assert pd.read_table.__module__ == "pandas"
+    assert pd.read_fwf.__module__ == "pandas"
+    assert pd.Series.__module__ == "pandas"
+    assert pd.date_range.__module__ == "pandas"
+    assert pd.bdate_range.__module__ == "pandas"
+    assert pd.timedelta_range.__module__ == "pandas"
+    assert pd.NamedAgg.__module__ == "pandas"
+    assert api.typing.SeriesGroupBy.__module__ == "pandas.api.typing"
+    assert api.typing.DataFrameGroupBy.__module__ == "pandas.api.typing"
diff --git a/pandas/pandas/tests/api/test_types.py b/pandas/pandas/tests/api/test_types.py
new file mode 100644
index 0000000000000..bf39370c49d76
--- /dev/null
+++ b/pandas/pandas/tests/api/test_types.py
@@ -0,0 +1,61 @@
+from __future__ import annotations
+
+import pandas._testing as tm
+from pandas.api import types
+from pandas.tests.api.test_api import Base
+
+
+class TestTypes(Base):
+    allowed = [
+        "is_any_real_numeric_dtype",
+        "is_bool",
+        "is_bool_dtype",
+        "is_categorical_dtype",
+        "is_complex",
+        "is_complex_dtype",
+        "is_datetime64_any_dtype",
+        "is_datetime64_dtype",
+        "is_datetime64_ns_dtype",
+        "is_datetime64tz_dtype",
+        "is_dtype_equal",
+        "is_float",
+        "is_float_dtype",
+        "is_int64_dtype",
+        "is_integer",
+        "is_integer_dtype",
+        "is_number",
+        "is_numeric_dtype",
+        "is_object_dtype",
+        "is_scalar",
+        "is_sparse",
+        "is_string_dtype",
+        "is_signed_integer_dtype",
+        "is_timedelta64_dtype",
+        "is_timedelta64_ns_dtype",
+        "is_unsigned_integer_dtype",
+        "is_period_dtype",
+        "is_interval_dtype",
+        "is_re",
+        "is_re_compilable",
+        "is_dict_like",
+        "is_iterator",
+        "is_file_like",
+        "is_list_like",
+        "is_hashable",
+        "is_array_like",
+        "is_named_tuple",
+        "pandas_dtype",
+        "union_categoricals",
+        "infer_dtype",
+        "is_extension_array_dtype",
+    ]
+    deprecated: list[str] = []
+    dtypes = ["CategoricalDtype", "DatetimeTZDtype", "PeriodDtype", "IntervalDtype"]
+
+    def test_types(self):
+        self.check(types, self.allowed + self.dtypes + self.deprecated)
+
+    def test_deprecated_from_api_types(self):
+        for t in self.deprecated:
+            with tm.assert_produces_warning(FutureWarning):
+                getattr(types, t)(1)
diff --git a/pandas/pandas/tests/apply/__init__.py b/pandas/pandas/tests/apply/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/apply/common.py b/pandas/pandas/tests/apply/common.py
new file mode 100644
index 0000000000000..b4d153df54059
--- /dev/null
+++ b/pandas/pandas/tests/apply/common.py
@@ -0,0 +1,7 @@
+from pandas.core.groupby.base import transformation_kernels
+
+# There is no Series.cumcount or DataFrame.cumcount
+series_transform_kernels = [
+    x for x in sorted(transformation_kernels) if x != "cumcount"
+]
+frame_transform_kernels = [x for x in sorted(transformation_kernels) if x != "cumcount"]
diff --git a/pandas/pandas/tests/apply/test_frame_apply.py b/pandas/pandas/tests/apply/test_frame_apply.py
new file mode 100644
index 0000000000000..b9e407adc3051
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_frame_apply.py
@@ -0,0 +1,1793 @@
+from datetime import datetime
+import warnings
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.frame.common import zip_frames
+from pandas.util.version import Version
+
+
+@pytest.fixture
+def int_frame_const_col():
+    """
+    Fixture for DataFrame of ints which are constant per column
+
+    Columns are ['A', 'B', 'C'], with values (per column): [1, 2, 3]
+    """
+    df = DataFrame(
+        np.tile(np.arange(3, dtype="int64"), 6).reshape(6, -1) + 1,
+        columns=["A", "B", "C"],
+    )
+    return df
+
+
+@pytest.fixture(params=["python", pytest.param("numba", marks=pytest.mark.single_cpu)])
+def engine(request):
+    if request.param == "numba":
+        pytest.importorskip("numba")
+    return request.param
+
+
+def test_apply(float_frame, engine, request):
+    if engine == "numba":
+        mark = pytest.mark.xfail(reason="numba engine not supporting numpy ufunc yet")
+        request.node.add_marker(mark)
+    with np.errstate(all="ignore"):
+        # ufunc
+        result = np.sqrt(float_frame["A"])
+        expected = float_frame.apply(np.sqrt, engine=engine)["A"]
+        tm.assert_series_equal(result, expected)
+
+        # aggregator
+        result = float_frame.apply(np.mean, engine=engine)["A"]
+        expected = np.mean(float_frame["A"])
+        assert result == expected
+
+        d = float_frame.index[0]
+        result = float_frame.apply(np.mean, axis=1, engine=engine)
+        expected = np.mean(float_frame.xs(d))
+        assert result[d] == expected
+        assert result.index is float_frame.index
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+@pytest.mark.parametrize("raw", [True, False])
+@pytest.mark.parametrize("nopython", [True, False])
+def test_apply_args(float_frame, axis, raw, engine, nopython):
+    numba = pytest.importorskip("numba")
+    if (
+        engine == "numba"
+        and Version(numba.__version__) == Version("0.61")
+        and is_platform_arm()
+    ):
+        pytest.skip(f"Segfaults on ARM platforms with numba {numba.__version__}")
+    engine_kwargs = {"nopython": nopython}
+    result = float_frame.apply(
+        lambda x, y: x + y,
+        axis,
+        args=(1,),
+        raw=raw,
+        engine=engine,
+        engine_kwargs=engine_kwargs,
+    )
+    expected = float_frame + 1
+    tm.assert_frame_equal(result, expected)
+
+    # GH:58712
+    result = float_frame.apply(
+        lambda x, a, b: x + a + b,
+        args=(1,),
+        b=2,
+        raw=raw,
+        engine=engine,
+        engine_kwargs=engine_kwargs,
+    )
+    expected = float_frame + 3
+    tm.assert_frame_equal(result, expected)
+
+    if engine == "numba":
+        # py signature binding
+        with pytest.raises(TypeError, match="missing a required argument: 'a'"):
+            float_frame.apply(
+                lambda x, a: x + a,
+                b=2,
+                raw=raw,
+                engine=engine,
+                engine_kwargs=engine_kwargs,
+            )
+
+        # keyword-only arguments are not supported in numba
+        with pytest.raises(
+            pd.errors.NumbaUtilError,
+            match="numba does not support keyword-only arguments",
+        ):
+            float_frame.apply(
+                lambda x, a, *, b: x + a + b,
+                args=(1,),
+                b=2,
+                raw=raw,
+                engine=engine,
+                engine_kwargs=engine_kwargs,
+            )
+
+        with pytest.raises(
+            pd.errors.NumbaUtilError,
+            match="numba does not support keyword-only arguments",
+        ):
+            float_frame.apply(
+                lambda *x, b: x[0] + x[1] + b,
+                args=(1,),
+                b=2,
+                raw=raw,
+                engine=engine,
+                engine_kwargs=engine_kwargs,
+            )
+
+
+def test_apply_categorical_func():
+    # GH 9573
+    df = DataFrame({"c0": ["A", "A", "B", "B"], "c1": ["C", "C", "D", "D"]})
+    result = df.apply(lambda ts: ts.astype("category"))
+
+    assert result.shape == (4, 2)
+    assert isinstance(result["c0"].dtype, CategoricalDtype)
+    assert isinstance(result["c1"].dtype, CategoricalDtype)
+
+
+def test_apply_axis1_with_ea():
+    # GH#36785
+    expected = DataFrame({"A": [Timestamp("2013-01-01", tz="UTC")]})
+    result = expected.apply(lambda x: x, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, dtype",
+    [(1, None), (1, CategoricalDtype([1])), (Timestamp("2013-01-01", tz="UTC"), None)],
+)
+def test_agg_axis1_duplicate_index(data, dtype):
+    # GH 42380
+    expected = DataFrame([[data], [data]], index=["a", "a"], dtype=dtype)
+    result = expected.agg(lambda x: x, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_mixed_datetimelike():
+    # mixed datetimelike
+    # GH 7778
+    expected = DataFrame(
+        {
+            "A": date_range("20130101", periods=3),
+            "B": pd.to_timedelta(np.arange(3), unit="s"),
+        }
+    )
+    result = expected.apply(lambda x: x, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", [np.sqrt, np.mean])
+def test_apply_empty(func, engine):
+    # empty
+    empty_frame = DataFrame()
+
+    result = empty_frame.apply(func, engine=engine)
+    assert result.empty
+
+
+def test_apply_float_frame(float_frame, engine):
+    no_rows = float_frame[:0]
+    result = no_rows.apply(lambda x: x.mean(), engine=engine)
+    expected = Series(np.nan, index=float_frame.columns)
+    tm.assert_series_equal(result, expected)
+
+    no_cols = float_frame.loc[:, []]
+    result = no_cols.apply(lambda x: x.mean(), axis=1, engine=engine)
+    expected = Series(np.nan, index=float_frame.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_empty_except_index(engine):
+    # GH 2476
+    expected = DataFrame(index=["a"])
+    result = expected.apply(lambda x: x["a"], axis=1, engine=engine)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_with_reduce_empty():
+    # reduce with an empty DataFrame
+    empty_frame = DataFrame()
+
+    x = []
+    result = empty_frame.apply(x.append, axis=1, result_type="expand")
+    tm.assert_frame_equal(result, empty_frame)
+    result = empty_frame.apply(x.append, axis=1, result_type="reduce")
+    expected = Series([], dtype=np.float64)
+    tm.assert_series_equal(result, expected)
+
+    empty_with_cols = DataFrame(columns=["a", "b", "c"])
+    result = empty_with_cols.apply(x.append, axis=1, result_type="expand")
+    tm.assert_frame_equal(result, empty_with_cols)
+    result = empty_with_cols.apply(x.append, axis=1, result_type="reduce")
+    expected = Series([], dtype=np.float64)
+    tm.assert_series_equal(result, expected)
+
+    # Ensure that x.append hasn't been called
+    assert x == []
+
+
+@pytest.mark.parametrize("func", ["sum", "prod", "any", "all"])
+def test_apply_funcs_over_empty(func):
+    # GH 28213
+    df = DataFrame(columns=["a", "b", "c"])
+
+    result = df.apply(getattr(np, func))
+    expected = getattr(df, func)()
+    if func in ("sum", "prod"):
+        expected = expected.astype(float)
+    tm.assert_series_equal(result, expected)
+
+
+def test_nunique_empty():
+    # GH 28213
+    df = DataFrame(columns=["a", "b", "c"])
+
+    result = df.nunique()
+    expected = Series(0, index=df.columns)
+    tm.assert_series_equal(result, expected)
+
+    result = df.T.nunique()
+    expected = Series([], dtype=np.float64)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_standard_nonunique():
+    df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=["a", "a", "c"])
+
+    result = df.apply(lambda s: s[0], axis=1)
+    expected = Series([1, 4, 7], ["a", "a", "c"])
+    tm.assert_series_equal(result, expected)
+
+    result = df.T.apply(lambda s: s[0], axis=0)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_broadcast_scalars(float_frame):
+    # scalars
+    result = float_frame.apply(np.mean, result_type="broadcast")
+    expected = DataFrame([float_frame.mean()], index=float_frame.index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_broadcast_scalars_axis1(float_frame):
+    result = float_frame.apply(np.mean, axis=1, result_type="broadcast")
+    m = float_frame.mean(axis=1)
+    expected = DataFrame({c: m for c in float_frame.columns})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_broadcast_lists_columns(float_frame):
+    # lists
+    result = float_frame.apply(
+        lambda x: list(range(len(float_frame.columns))),
+        axis=1,
+        result_type="broadcast",
+    )
+    m = list(range(len(float_frame.columns)))
+    expected = DataFrame(
+        [m] * len(float_frame.index),
+        dtype="float64",
+        index=float_frame.index,
+        columns=float_frame.columns,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_broadcast_lists_index(float_frame):
+    result = float_frame.apply(
+        lambda x: list(range(len(float_frame.index))), result_type="broadcast"
+    )
+    m = list(range(len(float_frame.index)))
+    expected = DataFrame(
+        {c: m for c in float_frame.columns},
+        dtype="float64",
+        index=float_frame.index,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_broadcast_list_lambda_func(int_frame_const_col):
+    # preserve columns
+    df = int_frame_const_col
+    result = df.apply(lambda x: [1, 2, 3], axis=1, result_type="broadcast")
+    tm.assert_frame_equal(result, df)
+
+
+def test_apply_broadcast_series_lambda_func(int_frame_const_col):
+    df = int_frame_const_col
+    result = df.apply(
+        lambda x: Series([1, 2, 3], index=list("abc")),
+        axis=1,
+        result_type="broadcast",
+    )
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+def test_apply_raw_float_frame(float_frame, axis, engine):
+    if engine == "numba":
+        pytest.skip("numba can't handle when UDF returns None.")
+
+    def _assert_raw(x):
+        assert isinstance(x, np.ndarray)
+        assert x.ndim == 1
+
+    float_frame.apply(_assert_raw, axis=axis, engine=engine, raw=True)
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+def test_apply_raw_float_frame_lambda(float_frame, axis, engine):
+    result = float_frame.apply(np.mean, axis=axis, engine=engine, raw=True)
+    expected = float_frame.apply(lambda x: x.values.mean(), axis=axis)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_raw_float_frame_no_reduction(float_frame, engine):
+    # no reduction
+    result = float_frame.apply(lambda x: x * 2, engine=engine, raw=True)
+    expected = float_frame * 2
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+def test_apply_raw_mixed_type_frame(axis, engine):
+    if engine == "numba":
+        pytest.skip("isinstance check doesn't work with numba")
+
+    def _assert_raw(x):
+        assert isinstance(x, np.ndarray)
+        assert x.ndim == 1
+
+    # Mixed dtype (GH-32423)
+    df = DataFrame(
+        {
+            "a": 1.0,
+            "b": 2,
+            "c": "foo",
+            "float32": np.array([1.0] * 10, dtype="float32"),
+            "int32": np.array([1] * 10, dtype="int32"),
+        },
+        index=np.arange(10),
+    )
+    df.apply(_assert_raw, axis=axis, engine=engine, raw=True)
+
+
+def test_apply_axis1(float_frame):
+    d = float_frame.index[0]
+    result = float_frame.apply(np.mean, axis=1)[d]
+    expected = np.mean(float_frame.xs(d))
+    assert result == expected
+
+
+def test_apply_mixed_dtype_corner():
+    df = DataFrame({"A": ["foo"], "B": [1.0]})
+    result = df[:0].apply(np.mean, axis=1)
+    # the result here is actually kind of ambiguous, should it be a Series
+    # or a DataFrame?
+    expected = Series(dtype=np.float64)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_mixed_dtype_corner_indexing():
+    df = DataFrame({"A": ["foo"], "B": [1.0]})
+    result = df.apply(lambda x: x["A"], axis=1)
+    expected = Series(["foo"], index=range(1))
+    tm.assert_series_equal(result, expected)
+
+    result = df.apply(lambda x: x["B"], axis=1)
+    expected = Series([1.0], index=range(1))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore::RuntimeWarning")
+@pytest.mark.parametrize("ax", ["index", "columns"])
+@pytest.mark.parametrize(
+    "func", [lambda x: x, lambda x: x.mean()], ids=["identity", "mean"]
+)
+@pytest.mark.parametrize("raw", [True, False])
+@pytest.mark.parametrize("axis", [0, 1])
+def test_apply_empty_infer_type(ax, func, raw, axis, engine, request):
+    df = DataFrame(**{ax: ["a", "b", "c"]})
+
+    with np.errstate(all="ignore"):
+        test_res = func(np.array([], dtype="f8"))
+        is_reduction = not isinstance(test_res, np.ndarray)
+
+        result = df.apply(func, axis=axis, engine=engine, raw=raw)
+        if is_reduction:
+            agg_axis = df._get_agg_axis(axis)
+            assert isinstance(result, Series)
+            assert result.index is agg_axis
+        else:
+            assert isinstance(result, DataFrame)
+
+
+def test_apply_empty_infer_type_broadcast():
+    no_cols = DataFrame(index=["a", "b", "c"])
+    result = no_cols.apply(lambda x: x.mean(), result_type="broadcast")
+    assert isinstance(result, DataFrame)
+
+
+def test_apply_with_args_kwds_add_some(float_frame):
+    def add_some(x, howmuch=0):
+        return x + howmuch
+
+    result = float_frame.apply(add_some, howmuch=2)
+    expected = float_frame.apply(lambda x: x + 2)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_with_args_kwds_agg_and_add(float_frame):
+    def agg_and_add(x, howmuch=0):
+        return x.mean() + howmuch
+
+    result = float_frame.apply(agg_and_add, howmuch=2)
+    expected = float_frame.apply(lambda x: x.mean() + 2)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_with_args_kwds_subtract_and_divide(float_frame):
+    def subtract_and_divide(x, sub, divide=1):
+        return (x - sub) / divide
+
+    result = float_frame.apply(subtract_and_divide, args=(2,), divide=2)
+    expected = float_frame.apply(lambda x: (x - 2.0) / 2.0)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_yield_list(float_frame):
+    result = float_frame.apply(list)
+    tm.assert_frame_equal(result, float_frame)
+
+
+def test_apply_reduce_Series(float_frame):
+    float_frame.iloc[::2, float_frame.columns.get_loc("A")] = np.nan
+    expected = float_frame.mean(axis=1)
+    result = float_frame.apply(np.mean, axis=1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_reduce_to_dict():
+    # GH 25196 37544
+    data = DataFrame([[1, 2], [3, 4]], columns=["c0", "c1"], index=["i0", "i1"])
+
+    result = data.apply(dict, axis=0)
+    expected = Series([{"i0": 1, "i1": 3}, {"i0": 2, "i1": 4}], index=data.columns)
+    tm.assert_series_equal(result, expected)
+
+    result = data.apply(dict, axis=1)
+    expected = Series([{"c0": 1, "c1": 2}, {"c0": 3, "c1": 4}], index=data.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_differently_indexed():
+    df = DataFrame(np.random.default_rng(2).standard_normal((20, 10)))
+
+    result = df.apply(Series.describe, axis=0)
+    expected = DataFrame({i: v.describe() for i, v in df.items()}, columns=df.columns)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.apply(Series.describe, axis=1)
+    expected = DataFrame({i: v.describe() for i, v in df.T.items()}, columns=df.index).T
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_bug():
+    # GH 6125
+    positions = DataFrame(
+        [
+            [1, "ABC0", 50],
+            [1, "YUM0", 20],
+            [1, "DEF0", 20],
+            [2, "ABC1", 50],
+            [2, "YUM1", 20],
+            [2, "DEF1", 20],
+        ],
+        columns=["a", "market", "position"],
+    )
+
+    def f(r):
+        return r["market"]
+
+    expected = positions.apply(f, axis=1)
+
+    positions = DataFrame(
+        [
+            [datetime(2013, 1, 1), "ABC0", 50],
+            [datetime(2013, 1, 2), "YUM0", 20],
+            [datetime(2013, 1, 3), "DEF0", 20],
+            [datetime(2013, 1, 4), "ABC1", 50],
+            [datetime(2013, 1, 5), "YUM1", 20],
+            [datetime(2013, 1, 6), "DEF1", 20],
+        ],
+        columns=["a", "market", "position"],
+    )
+    result = positions.apply(f, axis=1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_convert_objects():
+    expected = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    result = expected.apply(lambda x: x, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_attach_name(float_frame):
+    result = float_frame.apply(lambda x: x.name)
+    expected = Series(float_frame.columns, index=float_frame.columns)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_attach_name_axis1(float_frame):
+    result = float_frame.apply(lambda x: x.name, axis=1)
+    expected = Series(float_frame.index, index=float_frame.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_attach_name_non_reduction(float_frame):
+    # non-reductions
+    result = float_frame.apply(lambda x: np.repeat(x.name, len(x)))
+    expected = DataFrame(
+        np.tile(float_frame.columns, (len(float_frame.index), 1)),
+        index=float_frame.index,
+        columns=float_frame.columns,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_attach_name_non_reduction_axis1(float_frame):
+    result = float_frame.apply(lambda x: np.repeat(x.name, len(x)), axis=1)
+    expected = Series(
+        np.repeat(t[0], len(float_frame.columns)) for t in float_frame.itertuples()
+    )
+    expected.index = float_frame.index
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_multi_index():
+    index = MultiIndex.from_arrays([["a", "a", "b"], ["c", "d", "d"]])
+    s = DataFrame([[1, 2], [3, 4], [5, 6]], index=index, columns=["col1", "col2"])
+    result = s.apply(lambda x: Series({"min": min(x), "max": max(x)}), 1)
+    expected = DataFrame([[1, 2], [3, 4], [5, 6]], index=index, columns=["min", "max"])
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize(
+    "df, dicts",
+    [
+        [
+            DataFrame([["foo", "bar"], ["spam", "eggs"]]),
+            Series([{0: "foo", 1: "spam"}, {0: "bar", 1: "eggs"}]),
+        ],
+        [DataFrame([[0, 1], [2, 3]]), Series([{0: 0, 1: 2}, {0: 1, 1: 3}])],
+    ],
+)
+def test_apply_dict(df, dicts):
+    # GH 8735
+    fn = lambda x: x.to_dict()
+    reduce_true = df.apply(fn, result_type="reduce")
+    reduce_false = df.apply(fn, result_type="expand")
+    reduce_none = df.apply(fn)
+
+    tm.assert_series_equal(reduce_true, dicts)
+    tm.assert_frame_equal(reduce_false, df)
+    tm.assert_series_equal(reduce_none, dicts)
+
+
+def test_apply_non_numpy_dtype():
+    # GH 12244
+    df = DataFrame({"dt": date_range("2015-01-01", periods=3, tz="Europe/Brussels")})
+    result = df.apply(lambda x: x)
+    tm.assert_frame_equal(result, df)
+
+    result = df.apply(lambda x: x + pd.Timedelta("1day"))
+    expected = DataFrame(
+        {"dt": date_range("2015-01-02", periods=3, tz="Europe/Brussels")}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_non_numpy_dtype_category():
+    df = DataFrame({"dt": ["a", "b", "c", "a"]}, dtype="category")
+    result = df.apply(lambda x: x)
+    tm.assert_frame_equal(result, df)
+
+
+def test_apply_dup_names_multi_agg():
+    # GH 21063
+    df = DataFrame([[0, 1], [2, 3]], columns=["a", "a"])
+    expected = DataFrame([[0, 1]], columns=["a", "a"], index=["min"])
+    result = df.agg(["min"])
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", ["apply", "agg"])
+def test_apply_nested_result_axis_1(op):
+    # GH 13820
+    def apply_list(row):
+        return [2 * row["A"], 2 * row["C"], 2 * row["B"]]
+
+    df = DataFrame(np.zeros((4, 4)), columns=list("ABCD"))
+    result = getattr(df, op)(apply_list, axis=1)
+    expected = Series(
+        [[0.0, 0.0, 0.0], [0.0, 0.0, 0.0], [0.0, 0.0, 0.0], [0.0, 0.0, 0.0]]
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_noreduction_tzaware_object():
+    # https://github.com/pandas-dev/pandas/issues/31505
+    expected = DataFrame(
+        {"foo": [Timestamp("2020", tz="UTC")]}, dtype="datetime64[ns, UTC]"
+    )
+    result = expected.apply(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+    result = expected.apply(lambda x: x.copy())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_function_runs_once():
+    # https://github.com/pandas-dev/pandas/issues/30815
+
+    df = DataFrame({"a": [1, 2, 3]})
+    names = []  # Save row names function is applied to
+
+    def reducing_function(row):
+        names.append(row.name)
+
+    def non_reducing_function(row):
+        names.append(row.name)
+        return row
+
+    for func in [reducing_function, non_reducing_function]:
+        del names[:]
+
+        df.apply(func, axis=1)
+        assert names == list(df.index)
+
+
+def test_apply_raw_function_runs_once(engine):
+    # https://github.com/pandas-dev/pandas/issues/34506
+    if engine == "numba":
+        pytest.skip("appending to list outside of numba func is not supported")
+
+    df = DataFrame({"a": [1, 2, 3]})
+    values = []  # Save row values function is applied to
+
+    def reducing_function(row):
+        values.extend(row)
+
+    def non_reducing_function(row):
+        values.extend(row)
+        return row
+
+    for func in [reducing_function, non_reducing_function]:
+        del values[:]
+
+        df.apply(func, engine=engine, raw=True, axis=1)
+        assert values == list(df.a.to_list())
+
+
+def test_apply_with_byte_string():
+    # GH 34529
+    df = DataFrame(np.array([b"abcd", b"efgh"]), columns=["col"])
+    expected = DataFrame(np.array([b"abcd", b"efgh"]), columns=["col"], dtype=object)
+    # After we make the apply we expect a dataframe just
+    # like the original but with the object datatype
+    result = df.apply(lambda x: x.astype("object"))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", ["asd", 12, None, np.nan])
+def test_apply_category_equalness(val):
+    # Check if categorical comparisons on apply, GH 21239
+    df_values = ["asd", None, 12, "asd", "cde", np.nan]
+    df = DataFrame({"a": df_values}, dtype="category")
+
+    result = df.a.apply(lambda x: x == val)
+    expected = Series(
+        [False if pd.isnull(x) else x == val for x in df_values], name="a"
+    )
+    # False since behavior of NaN for categorical dtype has been changed (GH 59966)
+    tm.assert_series_equal(result, expected)
+
+
+# the user has supplied an opaque UDF where
+# they are transforming the input that requires
+# us to infer the output
+
+
+def test_infer_row_shape():
+    # GH 17437
+    # if row shape is changing, infer it
+    df = DataFrame(np.random.default_rng(2).random((10, 2)))
+    result = df.apply(np.fft.fft, axis=0).shape
+    assert result == (10, 2)
+
+    result = df.apply(np.fft.rfft, axis=0).shape
+    assert result == (6, 2)
+
+
+@pytest.mark.parametrize(
+    "ops, by_row, expected",
+    [
+        ({"a": lambda x: x + 1}, "compat", DataFrame({"a": [2, 3]})),
+        ({"a": lambda x: x + 1}, False, DataFrame({"a": [2, 3]})),
+        ({"a": lambda x: x.sum()}, "compat", Series({"a": 3})),
+        ({"a": lambda x: x.sum()}, False, Series({"a": 3})),
+        (
+            {"a": ["sum", np.sum, lambda x: x.sum()]},
+            "compat",
+            DataFrame({"a": [3, 3, 3]}, index=["sum", "sum", "<lambda>"]),
+        ),
+        (
+            {"a": ["sum", np.sum, lambda x: x.sum()]},
+            False,
+            DataFrame({"a": [3, 3, 3]}, index=["sum", "sum", "<lambda>"]),
+        ),
+        ({"a": lambda x: 1}, "compat", DataFrame({"a": [1, 1]})),
+        ({"a": lambda x: 1}, False, Series({"a": 1})),
+    ],
+)
+def test_dictlike_lambda(ops, by_row, expected):
+    # GH53601
+    df = DataFrame({"a": [1, 2]})
+    result = df.apply(ops, by_row=by_row)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops",
+    [
+        {"a": lambda x: x + 1},
+        {"a": lambda x: x.sum()},
+        {"a": ["sum", np.sum, lambda x: x.sum()]},
+        {"a": lambda x: 1},
+    ],
+)
+def test_dictlike_lambda_raises(ops):
+    # GH53601
+    df = DataFrame({"a": [1, 2]})
+    with pytest.raises(ValueError, match="by_row=True not allowed"):
+        df.apply(ops, by_row=True)
+
+
+def test_with_dictlike_columns():
+    # GH 17602
+    df = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+    result = df.apply(lambda x: {"s": x["a"] + x["b"]}, axis=1)
+    expected = Series([{"s": 3} for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+    df["tm"] = [
+        Timestamp("2017-05-01 00:00:00"),
+        Timestamp("2017-05-02 00:00:00"),
+    ]
+    result = df.apply(lambda x: {"s": x["a"] + x["b"]}, axis=1)
+    tm.assert_series_equal(result, expected)
+
+    # compose a series
+    result = (df["a"] + df["b"]).apply(lambda x: {"s": x})
+    expected = Series([{"s": 3}, {"s": 3}])
+    tm.assert_series_equal(result, expected)
+
+
+def test_with_dictlike_columns_with_datetime():
+    # GH 18775
+    df = DataFrame()
+    df["author"] = ["X", "Y", "Z"]
+    df["publisher"] = ["BBC", "NBC", "N24"]
+    df["date"] = pd.to_datetime(
+        ["17-10-2010 07:15:30", "13-05-2011 08:20:35", "15-01-2013 09:09:09"],
+        dayfirst=True,
+    )
+    result = df.apply(lambda x: {}, axis=1)
+    expected = Series([{}, {}, {}])
+    tm.assert_series_equal(result, expected)
+
+
+def test_with_dictlike_columns_with_infer():
+    # GH 17602
+    df = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+    result = df.apply(lambda x: {"s": x["a"] + x["b"]}, axis=1, result_type="expand")
+    expected = DataFrame({"s": [3, 3]})
+    tm.assert_frame_equal(result, expected)
+
+    df["tm"] = [
+        Timestamp("2017-05-01 00:00:00"),
+        Timestamp("2017-05-02 00:00:00"),
+    ]
+    result = df.apply(lambda x: {"s": x["a"] + x["b"]}, axis=1, result_type="expand")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, by_row, expected",
+    [
+        ([lambda x: x + 1], "compat", DataFrame({("a", "<lambda>"): [2, 3]})),
+        ([lambda x: x + 1], False, DataFrame({("a", "<lambda>"): [2, 3]})),
+        ([lambda x: x.sum()], "compat", DataFrame({"a": [3]}, index=["<lambda>"])),
+        ([lambda x: x.sum()], False, DataFrame({"a": [3]}, index=["<lambda>"])),
+        (
+            ["sum", np.sum, lambda x: x.sum()],
+            "compat",
+            DataFrame({"a": [3, 3, 3]}, index=["sum", "sum", "<lambda>"]),
+        ),
+        (
+            ["sum", np.sum, lambda x: x.sum()],
+            False,
+            DataFrame({"a": [3, 3, 3]}, index=["sum", "sum", "<lambda>"]),
+        ),
+        (
+            [lambda x: x + 1, lambda x: 3],
+            "compat",
+            DataFrame([[2, 3], [3, 3]], columns=[["a", "a"], ["<lambda>", "<lambda>"]]),
+        ),
+        (
+            [lambda x: 2, lambda x: 3],
+            False,
+            DataFrame({"a": [2, 3]}, ["<lambda>", "<lambda>"]),
+        ),
+    ],
+)
+def test_listlike_lambda(ops, by_row, expected):
+    # GH53601
+    df = DataFrame({"a": [1, 2]})
+    result = df.apply(ops, by_row=by_row)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops",
+    [
+        [lambda x: x + 1],
+        [lambda x: x.sum()],
+        ["sum", np.sum, lambda x: x.sum()],
+        [lambda x: x + 1, lambda x: 3],
+    ],
+)
+def test_listlike_lambda_raises(ops):
+    # GH53601
+    df = DataFrame({"a": [1, 2]})
+    with pytest.raises(ValueError, match="by_row=True not allowed"):
+        df.apply(ops, by_row=True)
+
+
+def test_with_listlike_columns():
+    # GH 17348
+    df = DataFrame(
+        {
+            "a": Series(np.random.default_rng(2).standard_normal(4)),
+            "b": ["a", "list", "of", "words"],
+            "ts": date_range("2016-10-01", periods=4, freq="h"),
+        }
+    )
+
+    result = df[["a", "b"]].apply(tuple, axis=1)
+    expected = Series([t[1:] for t in df[["a", "b"]].itertuples()])
+    tm.assert_series_equal(result, expected)
+
+    result = df[["a", "ts"]].apply(tuple, axis=1)
+    expected = Series([t[1:] for t in df[["a", "ts"]].itertuples()])
+    tm.assert_series_equal(result, expected)
+
+
+def test_with_listlike_columns_returning_list():
+    # GH 18919
+    df = DataFrame({"x": Series([["a", "b"], ["q"]]), "y": Series([["z"], ["q", "t"]])})
+    df.index = MultiIndex.from_tuples([("i0", "j0"), ("i1", "j1")])
+
+    result = df.apply(lambda row: [el for el in row["x"] if el in row["y"]], axis=1)
+    expected = Series([[], ["q"]], index=df.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_infer_output_shape_columns():
+    # GH 18573
+
+    df = DataFrame(
+        {
+            "number": [1.0, 2.0],
+            "string": ["foo", "bar"],
+            "datetime": [
+                Timestamp("2017-11-29 03:30:00"),
+                Timestamp("2017-11-29 03:45:00"),
+            ],
+        }
+    )
+    result = df.apply(lambda row: (row.number, row.string), axis=1)
+    expected = Series([(t.number, t.string) for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+
+def test_infer_output_shape_listlike_columns():
+    # GH 16353
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((6, 3)), columns=["A", "B", "C"]
+    )
+
+    result = df.apply(lambda x: [1, 2, 3], axis=1)
+    expected = Series([[1, 2, 3] for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+    result = df.apply(lambda x: [1, 2], axis=1)
+    expected = Series([[1, 2] for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [1, 2])
+def test_infer_output_shape_listlike_columns_np_func(val):
+    # GH 17970
+    df = DataFrame({"a": [1, 2, 3]}, index=list("abc"))
+
+    result = df.apply(lambda row: np.ones(val), axis=1)
+    expected = Series([np.ones(val) for t in df.itertuples()], index=df.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_infer_output_shape_listlike_columns_with_timestamp():
+    # GH 17892
+    df = DataFrame(
+        {
+            "a": [
+                Timestamp("2010-02-01"),
+                Timestamp("2010-02-04"),
+                Timestamp("2010-02-05"),
+                Timestamp("2010-02-06"),
+            ],
+            "b": [9, 5, 4, 3],
+            "c": [5, 3, 4, 2],
+            "d": [1, 2, 3, 4],
+        }
+    )
+
+    def fun(x):
+        return (1, 2)
+
+    result = df.apply(fun, axis=1)
+    expected = Series([(1, 2) for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("lst", [[1, 2, 3], [1, 2]])
+def test_consistent_coerce_for_shapes(lst):
+    # we want column names to NOT be propagated
+    # just because the shape matches the input shape
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((4, 3)), columns=["A", "B", "C"]
+    )
+
+    result = df.apply(lambda x: lst, axis=1)
+    expected = Series([lst for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+
+def test_consistent_names(int_frame_const_col):
+    # if a Series is returned, we should use the resulting index names
+    df = int_frame_const_col
+
+    result = df.apply(
+        lambda x: Series([1, 2, 3], index=["test", "other", "cols"]), axis=1
+    )
+    expected = int_frame_const_col.rename(
+        columns={"A": "test", "B": "other", "C": "cols"}
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.apply(lambda x: Series([1, 2], index=["test", "other"]), axis=1)
+    expected = expected[["test", "other"]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type(int_frame_const_col):
+    # result_type should be consistent no matter which
+    # path we take in the code
+    df = int_frame_const_col
+
+    result = df.apply(lambda x: [1, 2, 3], axis=1, result_type="expand")
+    expected = df.copy()
+    expected.columns = range(3)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type_shorter_list(int_frame_const_col):
+    # result_type should be consistent no matter which
+    # path we take in the code
+    df = int_frame_const_col
+    result = df.apply(lambda x: [1, 2], axis=1, result_type="expand")
+    expected = df[["A", "B"]].copy()
+    expected.columns = range(2)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type_broadcast(int_frame_const_col, request, engine):
+    # result_type should be consistent no matter which
+    # path we take in the code
+    if engine == "numba":
+        mark = pytest.mark.xfail(reason="numba engine doesn't support list return")
+        request.node.add_marker(mark)
+    df = int_frame_const_col
+    # broadcast result
+    result = df.apply(
+        lambda x: [1, 2, 3], axis=1, result_type="broadcast", engine=engine
+    )
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type_broadcast_series_func(int_frame_const_col, engine, request):
+    # result_type should be consistent no matter which
+    # path we take in the code
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="numba Series constructor only support ndarrays not list data"
+        )
+        request.node.add_marker(mark)
+    df = int_frame_const_col
+    columns = ["other", "col", "names"]
+    result = df.apply(
+        lambda x: Series([1, 2, 3], index=columns),
+        axis=1,
+        result_type="broadcast",
+        engine=engine,
+    )
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type_series_result(int_frame_const_col, engine, request):
+    # result_type should be consistent no matter which
+    # path we take in the code
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="numba Series constructor only support ndarrays not list data"
+        )
+        request.node.add_marker(mark)
+    df = int_frame_const_col
+    # series result
+    result = df.apply(lambda x: Series([1, 2, 3], index=x.index), axis=1, engine=engine)
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_result_type_series_result_other_index(int_frame_const_col, engine, request):
+    # result_type should be consistent no matter which
+    # path we take in the code
+
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="no support in numba Series constructor for list of columns"
+        )
+        request.node.add_marker(mark)
+    df = int_frame_const_col
+    # series result with other index
+    columns = ["other", "col", "names"]
+    result = df.apply(lambda x: Series([1, 2, 3], index=columns), axis=1, engine=engine)
+    expected = df.copy()
+    expected.columns = columns
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "box",
+    [lambda x: list(x), lambda x: tuple(x), lambda x: np.array(x, dtype="int64")],
+    ids=["list", "tuple", "array"],
+)
+def test_consistency_for_boxed(box, int_frame_const_col):
+    # passing an array or list should not affect the output shape
+    df = int_frame_const_col
+
+    result = df.apply(lambda x: box([1, 2]), axis=1)
+    expected = Series([box([1, 2]) for t in df.itertuples()])
+    tm.assert_series_equal(result, expected)
+
+    result = df.apply(lambda x: box([1, 2]), axis=1, result_type="expand")
+    expected = int_frame_const_col[["A", "B"]].rename(columns={"A": 0, "B": 1})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_transform(axis, float_frame):
+    other_axis = 1 if axis in {0, "index"} else 0
+
+    with np.errstate(all="ignore"):
+        f_abs = np.abs(float_frame)
+        f_sqrt = np.sqrt(float_frame)
+
+        # ufunc
+        expected = f_sqrt.copy()
+        result = float_frame.apply(np.sqrt, axis=axis)
+        tm.assert_frame_equal(result, expected)
+
+        # list-like
+        result = float_frame.apply([np.sqrt], axis=axis)
+        expected = f_sqrt.copy()
+        if axis in {0, "index"}:
+            expected.columns = MultiIndex.from_product([float_frame.columns, ["sqrt"]])
+        else:
+            expected.index = MultiIndex.from_product([float_frame.index, ["sqrt"]])
+        tm.assert_frame_equal(result, expected)
+
+        # multiple items in list
+        # these are in the order as if we are applying both
+        # functions per series and then concatting
+        result = float_frame.apply([np.abs, np.sqrt], axis=axis)
+        expected = zip_frames([f_abs, f_sqrt], axis=other_axis)
+        if axis in {0, "index"}:
+            expected.columns = MultiIndex.from_product(
+                [float_frame.columns, ["absolute", "sqrt"]]
+            )
+        else:
+            expected.index = MultiIndex.from_product(
+                [float_frame.index, ["absolute", "sqrt"]]
+            )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_demo():
+    # demonstration tests
+    df = DataFrame({"A": range(5), "B": 5})
+
+    result = df.agg(["min", "max"])
+    expected = DataFrame(
+        {"A": [0, 4], "B": [5, 5]}, columns=["A", "B"], index=["min", "max"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_demo_dict_agg():
+    # demonstration tests
+    df = DataFrame({"A": range(5), "B": 5})
+    result = df.agg({"A": ["min", "max"], "B": ["sum", "max"]})
+    expected = DataFrame(
+        {"A": [4.0, 0.0, np.nan], "B": [5.0, np.nan, 25.0]},
+        columns=["A", "B"],
+        index=["max", "min", "sum"],
+    )
+    tm.assert_frame_equal(result.reindex_like(expected), expected)
+
+
+def test_agg_with_name_as_column_name():
+    # GH 36212 - Column name is "name"
+    data = {"name": ["foo", "bar"]}
+    df = DataFrame(data)
+
+    # result's name should be None
+    result = df.agg({"name": "count"})
+    expected = Series({"name": 2})
+    tm.assert_series_equal(result, expected)
+
+    # Check if name is still preserved when aggregating series instead
+    result = df["name"].agg({"name": "count"})
+    expected = Series({"name": 2}, name="name")
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_multiple_mixed():
+    # GH 20909
+    mdf = DataFrame(
+        {
+            "A": [1, 2, 3],
+            "B": [1.0, 2.0, 3.0],
+            "C": ["foo", "bar", "baz"],
+        }
+    )
+    expected = DataFrame(
+        {
+            "A": [1, 6],
+            "B": [1.0, 6.0],
+            "C": ["bar", "foobarbaz"],
+        },
+        index=["min", "sum"],
+    )
+    # sorted index
+    result = mdf.agg(["min", "sum"])
+    tm.assert_frame_equal(result, expected)
+
+    result = mdf[["C", "B", "A"]].agg(["sum", "min"])
+    # GH40420: the result of .agg should have an index that is sorted
+    # according to the arguments provided to agg.
+    expected = expected[["C", "B", "A"]].reindex(["sum", "min"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_multiple_mixed_raises():
+    # GH 20909
+    mdf = DataFrame(
+        {
+            "A": [1, 2, 3],
+            "B": [1.0, 2.0, 3.0],
+            "C": ["foo", "bar", "baz"],
+            "D": date_range("20130101", periods=3),
+        }
+    )
+
+    # sorted index
+    msg = "does not support operation"
+    with pytest.raises(TypeError, match=msg):
+        mdf.agg(["min", "sum"])
+
+    with pytest.raises(TypeError, match=msg):
+        mdf[["D", "C", "B", "A"]].agg(["sum", "min"])
+
+
+def test_agg_reduce(axis, float_frame):
+    other_axis = 1 if axis in {0, "index"} else 0
+    name1, name2 = float_frame.axes[other_axis].unique()[:2].sort_values()
+
+    # all reducers
+    expected = pd.concat(
+        [
+            float_frame.mean(axis=axis),
+            float_frame.max(axis=axis),
+            float_frame.sum(axis=axis),
+        ],
+        axis=1,
+    )
+    expected.columns = ["mean", "max", "sum"]
+    expected = expected.T if axis in {0, "index"} else expected
+
+    result = float_frame.agg(["mean", "max", "sum"], axis=axis)
+    tm.assert_frame_equal(result, expected)
+
+    # dict input with scalars
+    func = {name1: "mean", name2: "sum"}
+    result = float_frame.agg(func, axis=axis)
+    expected = Series(
+        [
+            float_frame.loc(other_axis)[name1].mean(),
+            float_frame.loc(other_axis)[name2].sum(),
+        ],
+        index=[name1, name2],
+    )
+    tm.assert_series_equal(result, expected)
+
+    # dict input with lists
+    func = {name1: ["mean"], name2: ["sum"]}
+    result = float_frame.agg(func, axis=axis)
+    expected = DataFrame(
+        {
+            name1: Series([float_frame.loc(other_axis)[name1].mean()], index=["mean"]),
+            name2: Series([float_frame.loc(other_axis)[name2].sum()], index=["sum"]),
+        }
+    )
+    expected = expected.T if axis in {1, "columns"} else expected
+    tm.assert_frame_equal(result, expected)
+
+    # dict input with lists with multiple
+    func = {name1: ["mean", "sum"], name2: ["sum", "max"]}
+    result = float_frame.agg(func, axis=axis)
+    expected = pd.concat(
+        {
+            name1: Series(
+                [
+                    float_frame.loc(other_axis)[name1].mean(),
+                    float_frame.loc(other_axis)[name1].sum(),
+                ],
+                index=["mean", "sum"],
+            ),
+            name2: Series(
+                [
+                    float_frame.loc(other_axis)[name2].sum(),
+                    float_frame.loc(other_axis)[name2].max(),
+                ],
+                index=["sum", "max"],
+            ),
+        },
+        axis=1,
+    )
+    expected = expected.T if axis in {1, "columns"} else expected
+    tm.assert_frame_equal(result, expected)
+
+
+def test_named_agg_reduce_axis1_raises(float_frame):
+    name1, name2 = float_frame.axes[0].unique()[:2].sort_values()
+    msg = "Named aggregation is not supported when axis=1."
+    for axis in [1, "columns"]:
+        with pytest.raises(NotImplementedError, match=msg):
+            float_frame.agg(row1=(name1, "sum"), row2=(name2, "max"), axis=axis)
+
+
+def test_nuiscance_columns():
+    # GH 15015
+    df = DataFrame(
+        {
+            "A": [1, 2, 3],
+            "B": [1.0, 2.0, 3.0],
+            "C": ["foo", "bar", "baz"],
+            "D": date_range("20130101", periods=3),
+        }
+    )
+
+    result = df.agg("min")
+    expected = Series([1, 1.0, "bar", Timestamp("20130101")], index=df.columns)
+    tm.assert_series_equal(result, expected)
+
+    result = df.agg(["min"])
+    expected = DataFrame(
+        [[1, 1.0, "bar", Timestamp("20130101").as_unit("ns")]],
+        index=["min"],
+        columns=df.columns,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    msg = "does not support operation"
+    with pytest.raises(TypeError, match=msg):
+        df.agg("sum")
+
+    result = df[["A", "B", "C"]].agg("sum")
+    expected = Series([6, 6.0, "foobarbaz"], index=["A", "B", "C"])
+    tm.assert_series_equal(result, expected)
+
+    msg = "does not support operation"
+    with pytest.raises(TypeError, match=msg):
+        df.agg(["sum"])
+
+
+@pytest.mark.parametrize("how", ["agg", "apply"])
+def test_non_callable_aggregates(how):
+    # GH 16405
+    # 'size' is a property of frame/series
+    # validate that this is working
+    # GH 39116 - expand to apply
+    df = DataFrame(
+        {"A": [None, 2, 3], "B": [1.0, np.nan, 3.0], "C": ["foo", None, "bar"]}
+    )
+
+    # Function aggregate
+    result = getattr(df, how)({"A": "count"})
+    expected = Series({"A": 2})
+
+    tm.assert_series_equal(result, expected)
+
+    # Non-function aggregate
+    result = getattr(df, how)({"A": "size"})
+    expected = Series({"A": 3})
+
+    tm.assert_series_equal(result, expected)
+
+    # Mix function and non-function aggs
+    result1 = getattr(df, how)(["count", "size"])
+    result2 = getattr(df, how)(
+        {"A": ["count", "size"], "B": ["count", "size"], "C": ["count", "size"]}
+    )
+    expected = DataFrame(
+        {
+            "A": {"count": 2, "size": 3},
+            "B": {"count": 2, "size": 3},
+            "C": {"count": 2, "size": 3},
+        }
+    )
+
+    tm.assert_frame_equal(result1, result2, check_like=True)
+    tm.assert_frame_equal(result2, expected, check_like=True)
+
+    # Just functional string arg is same as calling df.arg()
+    result = getattr(df, how)("count")
+    expected = df.count()
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["agg", "apply"])
+def test_size_as_str(how, axis):
+    # GH 39934
+    df = DataFrame(
+        {"A": [None, 2, 3], "B": [1.0, np.nan, 3.0], "C": ["foo", None, "bar"]}
+    )
+    # Just a string attribute arg same as calling df.arg
+    # on the columns
+    result = getattr(df, how)("size", axis=axis)
+    if axis in (0, "index"):
+        expected = Series(df.shape[0], index=df.columns)
+    else:
+        expected = Series(df.shape[1], index=df.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_listlike_result():
+    # GH-29587 user defined function returning list-likes
+    df = DataFrame({"A": [2, 2, 3], "B": [1.5, np.nan, 1.5], "C": ["foo", None, "bar"]})
+
+    def func(group_col):
+        return list(group_col.dropna().unique())
+
+    result = df.agg(func)
+    expected = Series([[2, 3], [1.5], ["foo", "bar"]], index=["A", "B", "C"])
+    tm.assert_series_equal(result, expected)
+
+    result = df.agg([func])
+    expected = expected.to_frame("func").T
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+@pytest.mark.parametrize(
+    "args, kwargs",
+    [
+        ((1, 2, 3), {}),
+        ((8, 7, 15), {}),
+        ((1, 2), {}),
+        ((1,), {"b": 2}),
+        ((), {"a": 1, "b": 2}),
+        ((), {"a": 2, "b": 1}),
+        ((), {"a": 1, "b": 2, "c": 3}),
+    ],
+)
+def test_agg_args_kwargs(axis, args, kwargs):
+    def f(x, a, b, c=3):
+        return x.sum() + (a + b) / c
+
+    df = DataFrame([[1, 2], [3, 4]])
+
+    if axis == 0:
+        expected = Series([5.0, 7.0])
+    else:
+        expected = Series([4.0, 8.0])
+
+    result = df.agg(f, axis, *args, **kwargs)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("num_cols", [2, 3, 5])
+def test_frequency_is_original(num_cols, engine, request):
+    # GH 22150
+    if engine == "numba":
+        mark = pytest.mark.xfail(reason="numba engine only supports numeric indices")
+        request.node.add_marker(mark)
+    index = pd.DatetimeIndex(["1950-06-30", "1952-10-24", "1953-05-29"])
+    original = index.copy()
+    df = DataFrame(1, index=index, columns=range(num_cols))
+    df.apply(lambda x: x, engine=engine)
+    assert index.freq == original.freq
+
+
+def test_apply_datetime_tz_issue(engine, request):
+    # GH 29052
+
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="numba engine doesn't support non-numeric indexes"
+        )
+        request.node.add_marker(mark)
+
+    timestamps = [
+        Timestamp("2019-03-15 12:34:31.909000+0000", tz="UTC"),
+        Timestamp("2019-03-15 12:34:34.359000+0000", tz="UTC"),
+        Timestamp("2019-03-15 12:34:34.660000+0000", tz="UTC"),
+    ]
+    df = DataFrame(data=[0, 1, 2], index=timestamps)
+    result = df.apply(lambda x: x.name, axis=1, engine=engine)
+    expected = Series(index=timestamps, data=timestamps)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("df", [DataFrame({"A": ["a", None], "B": ["c", "d"]})])
+@pytest.mark.parametrize("method", ["min", "max", "sum"])
+def test_mixed_column_raises(df, method, using_infer_string):
+    # GH 16832
+    if method == "sum":
+        msg = r'can only concatenate str \(not "int"\) to str|does not support'
+    else:
+        msg = "not supported between instances of 'str' and 'float'"
+    if not using_infer_string:
+        with pytest.raises(TypeError, match=msg):
+            getattr(df, method)()
+    else:
+        getattr(df, method)()
+
+
+@pytest.mark.parametrize("col", [1, 1.0, True, "a", np.nan])
+def test_apply_dtype(col):
+    # GH 31466
+    df = DataFrame([[1.0, col]], columns=["a", "b"])
+    result = df.apply(lambda x: x.dtype)
+    expected = df.dtypes
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_mutating():
+    # GH#35462 case where applied func pins a new BlockManager to a row
+    df = DataFrame({"a": range(10), "b": range(10, 20)})
+    df_orig = df.copy()
+
+    def func(row):
+        mgr = row._mgr
+        row.loc["a"] += 1
+        assert row._mgr is not mgr
+        return row
+
+    expected = df.copy()
+    expected["a"] += 1
+
+    result = df.apply(func, axis=1)
+
+    tm.assert_frame_equal(result, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_apply_empty_list_reduce():
+    # GH#35683 get columns correct
+    df = DataFrame([[1, 2], [3, 4], [5, 6], [7, 8], [9, 10]], columns=["a", "b"])
+
+    result = df.apply(lambda x: [], result_type="reduce")
+    expected = Series({"a": [], "b": []}, dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_no_suffix_index(engine, request):
+    # GH36189
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="numba engine doesn't support list-likes/dict-like callables"
+        )
+        request.node.add_marker(mark)
+    pdf = DataFrame([[4, 9]] * 3, columns=["A", "B"])
+    result = pdf.apply(["sum", lambda x: x.sum(), lambda x: x.sum()], engine=engine)
+    expected = DataFrame(
+        {"A": [12, 12, 12], "B": [27, 27, 27]}, index=["sum", "<lambda>", "<lambda>"]
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_raw_returns_string(engine):
+    # https://github.com/pandas-dev/pandas/issues/35940
+    if engine == "numba":
+        pytest.skip("No object dtype support in numba")
+    df = DataFrame({"A": ["aa", "bbb"]})
+    result = df.apply(lambda x: x[0], engine=engine, axis=1, raw=True)
+    expected = Series(["aa", "bbb"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_aggregation_func_column_order():
+    # GH40420: the result of .agg should have an index that is sorted
+    # according to the arguments provided to agg.
+    df = DataFrame(
+        [
+            (1, 0, 0),
+            (2, 0, 0),
+            (3, 0, 0),
+            (4, 5, 4),
+            (5, 6, 6),
+            (6, 7, 7),
+        ],
+        columns=("att1", "att2", "att3"),
+    )
+
+    def sum_div2(s):
+        return s.sum() / 2
+
+    aggs = ["sum", sum_div2, "count", "min"]
+    result = df.agg(aggs)
+    expected = DataFrame(
+        {
+            "att1": [21.0, 10.5, 6.0, 1.0],
+            "att2": [18.0, 9.0, 6.0, 0.0],
+            "att3": [17.0, 8.5, 6.0, 0.0],
+        },
+        index=["sum", "sum_div2", "count", "min"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_getitem_axis_1(engine, request):
+    # GH 13427
+    if engine == "numba":
+        mark = pytest.mark.xfail(
+            reason="numba engine not supporting duplicate index values"
+        )
+        request.node.add_marker(mark)
+    df = DataFrame({"a": [0, 1, 2], "b": [1, 2, 3]})
+    result = df[["a", "a"]].apply(
+        lambda x: x.iloc[0] + x.iloc[1], axis=1, engine=engine
+    )
+    expected = Series([0, 2, 4])
+    tm.assert_series_equal(result, expected)
+
+
+def test_nuisance_depr_passes_through_warnings():
+    # GH 43740
+    # DataFrame.agg with list-likes may emit warnings for both individual
+    # args and for entire columns, but we only want to emit once. We
+    # catch and suppress the warnings for individual args, but need to make
+    # sure if some other warnings were raised, they get passed through to
+    # the user.
+
+    def expected_warning(x):
+        warnings.warn("Hello, World!")
+        return x.sum()
+
+    df = DataFrame({"a": [1, 2, 3]})
+    with tm.assert_produces_warning(UserWarning, match="Hello, World!"):
+        df.agg([expected_warning])
+
+
+def test_apply_type():
+    # GH 46719
+    df = DataFrame(
+        {"col1": [3, "string", float], "col2": [0.25, datetime(2020, 1, 1), np.nan]},
+        index=["a", "b", "c"],
+    )
+
+    # axis=0
+    result = df.apply(type, axis=0)
+    expected = Series({"col1": Series, "col2": Series})
+    tm.assert_series_equal(result, expected)
+
+    # axis=1
+    result = df.apply(type, axis=1)
+    expected = Series({"a": Series, "b": Series, "c": Series})
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_on_empty_dataframe(engine):
+    # GH 39111
+    df = DataFrame({"a": [1, 2], "b": [3, 0]})
+    result = df.head(0).apply(lambda x: max(x["a"], x["b"]), axis=1, engine=engine)
+    expected = Series([], dtype=np.float64)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_return_list():
+    df = DataFrame({"a": [1, 2], "b": [2, 3]})
+    result = df.apply(lambda x: [x.values])
+    expected = DataFrame({"a": [[1, 2]], "b": [[2, 3]]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "test, constant",
+    [
+        ({"a": [1, 2, 3], "b": [1, 1, 1]}, {"a": [1, 2, 3], "b": [1]}),
+        ({"a": [2, 2, 2], "b": [1, 1, 1]}, {"a": [2], "b": [1]}),
+    ],
+)
+def test_unique_agg_type_is_series(test, constant):
+    # GH#22558
+    df1 = DataFrame(test)
+    expected = Series(data=constant, index=["a", "b"], dtype="object")
+    aggregation = {"a": "unique", "b": "unique"}
+
+    result = df1.agg(aggregation)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_any_apply_keyword_non_zero_axis_regression():
+    # https://github.com/pandas-dev/pandas/issues/48656
+    df = DataFrame({"A": [1, 2, 0], "B": [0, 2, 0], "C": [0, 0, 0]})
+    expected = Series([True, True, False])
+    tm.assert_series_equal(df.any(axis=1), expected)
+
+    result = df.apply("any", axis=1)
+    tm.assert_series_equal(result, expected)
+
+    result = df.apply("any", 1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_mapping_func_deprecated():
+    # GH 53325
+    df = DataFrame({"x": [1, 2, 3]})
+
+    def foo1(x, a=1, c=0):
+        return x + a + c
+
+    def foo2(x, b=2, c=0):
+        return x + b + c
+
+    # single func already takes the vectorized path
+    result = df.agg(foo1, 0, 3, c=4)
+    expected = df + 7
+    tm.assert_frame_equal(result, expected)
+
+    result = df.agg([foo1, foo2], 0, 3, c=4)
+    expected = DataFrame(
+        [[8, 8], [9, 9], [10, 10]], columns=[["x", "x"], ["foo1", "foo2"]]
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # TODO: the result below is wrong, should be fixed (GH53325)
+    result = df.agg({"x": foo1}, 0, 3, c=4)
+    expected = DataFrame([2, 3, 4], columns=["x"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_std():
+    df = DataFrame(np.arange(6).reshape(3, 2), columns=["A", "B"])
+
+    result = df.agg(np.std, ddof=1)
+    expected = Series({"A": 2.0, "B": 2.0}, dtype=float)
+    tm.assert_series_equal(result, expected)
+
+    result = df.agg([np.std], ddof=1)
+    expected = DataFrame({"A": 2.0, "B": 2.0}, index=["std"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_dist_like_and_nonunique_columns():
+    # GH#51099
+    df = DataFrame(
+        {"A": [None, 2, 3], "B": [1.0, np.nan, 3.0], "C": ["foo", None, "bar"]}
+    )
+    df.columns = ["A", "A", "C"]
+
+    result = df.agg({"A": "count"})
+    expected = df["A"].count()
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/apply/test_frame_apply_relabeling.py b/pandas/pandas/tests/apply/test_frame_apply_relabeling.py
new file mode 100644
index 0000000000000..57c109abba304
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_frame_apply_relabeling.py
@@ -0,0 +1,109 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gte1p25
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_agg_relabel():
+    # GH 26513
+    df = pd.DataFrame({"A": [1, 2, 1, 2], "B": [1, 2, 3, 4], "C": [3, 4, 5, 6]})
+
+    # simplest case with one column, one func
+    result = df.agg(foo=("B", "sum"))
+    expected = pd.DataFrame({"B": [10]}, index=pd.Index(["foo"]))
+    tm.assert_frame_equal(result, expected)
+
+    # test on same column with different methods
+    result = df.agg(foo=("B", "sum"), bar=("B", "min"))
+    expected = pd.DataFrame({"B": [10, 1]}, index=pd.Index(["foo", "bar"]))
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_relabel_multi_columns_multi_methods():
+    # GH 26513, test on multiple columns with multiple methods
+    df = pd.DataFrame({"A": [1, 2, 1, 2], "B": [1, 2, 3, 4], "C": [3, 4, 5, 6]})
+    result = df.agg(
+        foo=("A", "sum"),
+        bar=("B", "mean"),
+        cat=("A", "min"),
+        dat=("B", "max"),
+        f=("A", "max"),
+        g=("C", "min"),
+    )
+    expected = pd.DataFrame(
+        {
+            "A": [6.0, np.nan, 1.0, np.nan, 2.0, np.nan],
+            "B": [np.nan, 2.5, np.nan, 4.0, np.nan, np.nan],
+            "C": [np.nan, np.nan, np.nan, np.nan, np.nan, 3.0],
+        },
+        index=pd.Index(["foo", "bar", "cat", "dat", "f", "g"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.xfail(np_version_gte1p25, reason="name of min now equals name of np.min")
+def test_agg_relabel_partial_functions():
+    # GH 26513, test on partial, functools or more complex cases
+    df = pd.DataFrame({"A": [1, 2, 1, 2], "B": [1, 2, 3, 4], "C": [3, 4, 5, 6]})
+    result = df.agg(foo=("A", np.mean), bar=("A", "mean"), cat=("A", min))
+    expected = pd.DataFrame(
+        {"A": [1.5, 1.5, 1.0]}, index=pd.Index(["foo", "bar", "cat"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.agg(
+        foo=("A", min),
+        bar=("A", np.min),
+        cat=("B", max),
+        dat=("C", "min"),
+        f=("B", np.sum),
+        kk=("B", lambda x: min(x)),
+    )
+    expected = pd.DataFrame(
+        {
+            "A": [1.0, 1.0, np.nan, np.nan, np.nan, np.nan],
+            "B": [np.nan, np.nan, 4.0, np.nan, 10.0, 1.0],
+            "C": [np.nan, np.nan, np.nan, 3.0, np.nan, np.nan],
+        },
+        index=pd.Index(["foo", "bar", "cat", "dat", "f", "kk"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_namedtuple():
+    # GH 26513
+    df = pd.DataFrame({"A": [0, 1], "B": [1, 2]})
+    result = df.agg(
+        foo=pd.NamedAgg("B", "sum"),
+        bar=pd.NamedAgg("B", "min"),
+        cat=pd.NamedAgg(column="B", aggfunc="count"),
+        fft=pd.NamedAgg("B", aggfunc="max"),
+    )
+
+    expected = pd.DataFrame(
+        {"B": [3, 1, 2, 2]}, index=pd.Index(["foo", "bar", "cat", "fft"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.agg(
+        foo=pd.NamedAgg("A", "min"),
+        bar=pd.NamedAgg(column="B", aggfunc="max"),
+        cat=pd.NamedAgg(column="A", aggfunc="max"),
+    )
+    expected = pd.DataFrame(
+        {"A": [0.0, np.nan, 1.0], "B": [np.nan, 2.0, np.nan]},
+        index=pd.Index(["foo", "bar", "cat"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reconstruct_func():
+    # GH 28472, test to ensure reconstruct_func isn't moved;
+    # This method is used by other libraries (e.g. dask)
+    result = pd.core.apply.reconstruct_func("min")
+    expected = (False, "min", None, None)
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/apply/test_frame_transform.py b/pandas/pandas/tests/apply/test_frame_transform.py
new file mode 100644
index 0000000000000..558d76ae8fdc4
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_frame_transform.py
@@ -0,0 +1,264 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.apply.common import frame_transform_kernels
+from pandas.tests.frame.common import zip_frames
+
+
+def unpack_obj(obj, klass, axis):
+    """
+    Helper to ensure we have the right type of object for a test parametrized
+    over frame_or_series.
+    """
+    if klass is not DataFrame:
+        obj = obj["A"]
+        if axis != 0:
+            pytest.skip(f"Test is only for DataFrame with axis={axis}")
+    return obj
+
+
+def test_transform_ufunc(axis, float_frame, frame_or_series):
+    # GH 35964
+    obj = unpack_obj(float_frame, frame_or_series, axis)
+
+    with np.errstate(all="ignore"):
+        f_sqrt = np.sqrt(obj)
+
+    # ufunc
+    result = obj.transform(np.sqrt, axis=axis)
+    expected = f_sqrt
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, names",
+    [
+        ([np.sqrt], ["sqrt"]),
+        ([np.abs, np.sqrt], ["absolute", "sqrt"]),
+        (np.array([np.sqrt]), ["sqrt"]),
+        (np.array([np.abs, np.sqrt]), ["absolute", "sqrt"]),
+    ],
+)
+def test_transform_listlike(axis, float_frame, ops, names):
+    # GH 35964
+    other_axis = 1 if axis in {0, "index"} else 0
+    with np.errstate(all="ignore"):
+        expected = zip_frames([op(float_frame) for op in ops], axis=other_axis)
+    if axis in {0, "index"}:
+        expected.columns = MultiIndex.from_product([float_frame.columns, names])
+    else:
+        expected.index = MultiIndex.from_product([float_frame.index, names])
+    result = float_frame.transform(ops, axis=axis)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("ops", [[], np.array([])])
+def test_transform_empty_listlike(float_frame, ops, frame_or_series):
+    obj = unpack_obj(float_frame, frame_or_series, 0)
+
+    with pytest.raises(ValueError, match="No transform functions were provided"):
+        obj.transform(ops)
+
+
+def test_transform_listlike_func_with_args():
+    # GH 50624
+    df = DataFrame({"x": [1, 2, 3]})
+
+    def foo1(x, a=1, c=0):
+        return x + a + c
+
+    def foo2(x, b=2, c=0):
+        return x + b + c
+
+    msg = r"foo1\(\) got an unexpected keyword argument 'b'"
+    with pytest.raises(TypeError, match=msg):
+        df.transform([foo1, foo2], 0, 3, b=3, c=4)
+
+    result = df.transform([foo1, foo2], 0, 3, c=4)
+    expected = DataFrame(
+        [[8, 8], [9, 9], [10, 10]],
+        columns=MultiIndex.from_tuples([("x", "foo1"), ("x", "foo2")]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [dict, Series])
+def test_transform_dictlike(axis, float_frame, box):
+    # GH 35964
+    if axis in (0, "index"):
+        e = float_frame.columns[0]
+        expected = float_frame[[e]].transform(np.abs)
+    else:
+        e = float_frame.index[0]
+        expected = float_frame.iloc[[0]].transform(np.abs)
+    result = float_frame.transform(box({e: np.abs}), axis=axis)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_dictlike_mixed():
+    # GH 40018 - mix of lists and non-lists in values of a dictionary
+    df = DataFrame({"a": [1, 2], "b": [1, 4], "c": [1, 4]})
+    result = df.transform({"b": ["sqrt", "abs"], "c": "sqrt"})
+    expected = DataFrame(
+        [[1.0, 1, 1.0], [2.0, 4, 2.0]],
+        columns=MultiIndex([("b", "c"), ("sqrt", "abs")], [(0, 0, 1), (0, 1, 0)]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops",
+    [
+        {},
+        {"A": []},
+        {"A": [], "B": "cumsum"},
+        {"A": "cumsum", "B": []},
+        {"A": [], "B": ["cumsum"]},
+        {"A": ["cumsum"], "B": []},
+    ],
+)
+def test_transform_empty_dictlike(float_frame, ops, frame_or_series):
+    obj = unpack_obj(float_frame, frame_or_series, 0)
+
+    with pytest.raises(ValueError, match="No transform functions were provided"):
+        obj.transform(ops)
+
+
+@pytest.mark.parametrize("use_apply", [True, False])
+def test_transform_udf(axis, float_frame, use_apply, frame_or_series):
+    # GH 35964
+    obj = unpack_obj(float_frame, frame_or_series, axis)
+
+    # transform uses UDF either via apply or passing the entire DataFrame
+    def func(x):
+        # transform is using apply iff x is not a DataFrame
+        if use_apply == isinstance(x, frame_or_series):
+            # Force transform to fallback
+            raise ValueError
+        return x + 1
+
+    result = obj.transform(func, axis=axis)
+    expected = obj + 1
+    tm.assert_equal(result, expected)
+
+
+wont_fail = ["ffill", "bfill", "fillna", "pad", "backfill", "shift"]
+frame_kernels_raise = [x for x in frame_transform_kernels if x not in wont_fail]
+
+
+@pytest.mark.parametrize("op", [*frame_kernels_raise, lambda x: x + 1])
+def test_transform_bad_dtype(op, frame_or_series, request):
+    # GH 35964
+    if op == "ngroup":
+        request.applymarker(
+            pytest.mark.xfail(raises=ValueError, reason="ngroup not valid for NDFrame")
+        )
+
+    obj = DataFrame({"A": 3 * [object]})  # DataFrame that will fail on most transforms
+    obj = tm.get_obj(obj, frame_or_series)
+    error = TypeError
+    msg = "|".join(
+        [
+            "not supported between instances of 'type' and 'type'",
+            "unsupported operand type",
+        ]
+    )
+
+    with pytest.raises(error, match=msg):
+        obj.transform(op)
+    with pytest.raises(error, match=msg):
+        obj.transform([op])
+    with pytest.raises(error, match=msg):
+        obj.transform({"A": op})
+    with pytest.raises(error, match=msg):
+        obj.transform({"A": [op]})
+
+
+@pytest.mark.parametrize("op", frame_kernels_raise)
+def test_transform_failure_typeerror(request, op):
+    # GH 35964
+
+    if op == "ngroup":
+        request.applymarker(
+            pytest.mark.xfail(raises=ValueError, reason="ngroup not valid for NDFrame")
+        )
+
+    # Using object makes most transform kernels fail
+    df = DataFrame({"A": 3 * [object], "B": [1, 2, 3]})
+    error = TypeError
+    msg = "|".join(
+        [
+            "not supported between instances of 'type' and 'type'",
+            "unsupported operand type",
+        ]
+    )
+
+    with pytest.raises(error, match=msg):
+        df.transform([op])
+
+    with pytest.raises(error, match=msg):
+        df.transform({"A": op, "B": op})
+
+    with pytest.raises(error, match=msg):
+        df.transform({"A": [op], "B": [op]})
+
+    with pytest.raises(error, match=msg):
+        df.transform({"A": [op, "shift"], "B": [op]})
+
+
+def test_transform_failure_valueerror():
+    # GH 40211
+    def op(x):
+        if np.sum(np.sum(x)) < 10:
+            raise ValueError
+        return x
+
+    df = DataFrame({"A": [1, 2, 3], "B": [400, 500, 600]})
+    msg = "Transform function failed"
+
+    with pytest.raises(ValueError, match=msg):
+        df.transform([op])
+
+    with pytest.raises(ValueError, match=msg):
+        df.transform({"A": op, "B": op})
+
+    with pytest.raises(ValueError, match=msg):
+        df.transform({"A": [op], "B": [op]})
+
+    with pytest.raises(ValueError, match=msg):
+        df.transform({"A": [op, "shift"], "B": [op]})
+
+
+@pytest.mark.parametrize("use_apply", [True, False])
+def test_transform_passes_args(use_apply, frame_or_series):
+    # GH 35964
+    # transform uses UDF either via apply or passing the entire DataFrame
+    expected_args = [1, 2]
+    expected_kwargs = {"c": 3}
+
+    def f(x, a, b, c):
+        # transform is using apply iff x is not a DataFrame
+        if use_apply == isinstance(x, frame_or_series):
+            # Force transform to fallback
+            raise ValueError
+        assert [a, b] == expected_args
+        assert c == expected_kwargs["c"]
+        return x
+
+    frame_or_series([1]).transform(f, 0, *expected_args, **expected_kwargs)
+
+
+def test_transform_empty_dataframe():
+    # https://github.com/pandas-dev/pandas/issues/39636
+    df = DataFrame([], columns=["col1", "col2"])
+    result = df.transform(lambda x: x + 10)
+    tm.assert_frame_equal(result, df)
+
+    result = df["col1"].transform(lambda x: x + 10)
+    tm.assert_series_equal(result, df["col1"])
diff --git a/pandas/pandas/tests/apply/test_invalid_arg.py b/pandas/pandas/tests/apply/test_invalid_arg.py
new file mode 100644
index 0000000000000..0503bf9166ec7
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_invalid_arg.py
@@ -0,0 +1,375 @@
+# Tests specifically aimed at detecting bad arguments.
+# This file is organized by reason for exception.
+#     1. always invalid argument values
+#     2. missing column(s)
+#     3. incompatible ops/dtype/args/kwargs
+#     4. invalid result shape/type
+# If your test does not fit into one of these categories, add to this list.
+
+from itertools import chain
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import SpecificationError
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("result_type", ["foo", 1])
+def test_result_type_error(result_type):
+    # allowed result_type
+    df = DataFrame(
+        np.tile(np.arange(3, dtype="int64"), 6).reshape(6, -1) + 1,
+        columns=["A", "B", "C"],
+    )
+
+    msg = (
+        "invalid value for result_type, must be one of "
+        "{None, 'reduce', 'broadcast', 'expand'}"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.apply(lambda x: [1, 2, 3], axis=1, result_type=result_type)
+
+
+def test_apply_invalid_axis_value():
+    df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=["a", "a", "c"])
+    msg = "No axis named 2 for object type DataFrame"
+    with pytest.raises(ValueError, match=msg):
+        df.apply(lambda x: x, 2)
+
+
+def test_agg_raises():
+    # GH 26513
+    df = DataFrame({"A": [0, 1], "B": [1, 2]})
+    msg = "Must provide"
+
+    with pytest.raises(TypeError, match=msg):
+        df.agg()
+
+
+def test_map_with_invalid_na_action_raises():
+    # https://github.com/pandas-dev/pandas/issues/32815
+    s = Series([1, 2, 3])
+    msg = "na_action must either be 'ignore' or None"
+    with pytest.raises(ValueError, match=msg):
+        s.map(lambda x: x, na_action="____")
+
+
+@pytest.mark.parametrize("input_na_action", ["____", True])
+def test_map_arg_is_dict_with_invalid_na_action_raises(input_na_action):
+    # https://github.com/pandas-dev/pandas/issues/46588
+    s = Series([1, 2, 3])
+    msg = f"na_action must either be 'ignore' or None, {input_na_action} was passed"
+    with pytest.raises(ValueError, match=msg):
+        s.map({1: 2}, na_action=input_na_action)
+
+
+@pytest.mark.parametrize("method", ["apply", "agg", "transform"])
+@pytest.mark.parametrize("func", [{"A": {"B": "sum"}}, {"A": {"B": ["sum"]}}])
+def test_nested_renamer(frame_or_series, method, func):
+    # GH 35964
+    obj = frame_or_series({"A": [1]})
+    match = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=match):
+        getattr(obj, method)(func)
+
+
+@pytest.mark.parametrize(
+    "renamer",
+    [{"foo": ["min", "max"]}, {"foo": ["min", "max"], "bar": ["sum", "mean"]}],
+)
+def test_series_nested_renamer(renamer):
+    s = Series(range(6), dtype="int64", name="series")
+    msg = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        s.agg(renamer)
+
+
+def test_apply_dict_depr():
+    tsdf = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 3)),
+        columns=["A", "B", "C"],
+        index=date_range("1/1/2000", periods=10),
+    )
+    msg = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        tsdf.A.agg({"foo": ["sum", "mean"]})
+
+
+@pytest.mark.parametrize("method", ["agg", "transform"])
+def test_dict_nested_renaming_depr(method):
+    df = DataFrame({"A": range(5), "B": 5})
+
+    # nested renaming
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        getattr(df, method)({"A": {"foo": "min"}, "B": {"bar": "max"}})
+
+
+@pytest.mark.parametrize("method", ["apply", "agg", "transform"])
+@pytest.mark.parametrize("func", [{"B": "sum"}, {"B": ["sum"]}])
+def test_missing_column(method, func):
+    # GH 40004
+    obj = DataFrame({"A": [1]})
+    msg = r"Label\(s\) \['B'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        getattr(obj, method)(func)
+
+
+def test_transform_mixed_column_name_dtypes():
+    # GH39025
+    df = DataFrame({"a": ["1"]})
+    msg = r"Label\(s\) \[1, 'b'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        df.transform({"a": int, 1: str, "b": int})
+
+
+@pytest.mark.parametrize(
+    "how, args", [("pct_change", ()), ("nsmallest", (1, ["a", "b"])), ("tail", 1)]
+)
+def test_apply_str_axis_1_raises(how, args):
+    # GH 39211 - some ops don't support axis=1
+    df = DataFrame({"a": [1, 2], "b": [3, 4]})
+    msg = f"Operation {how} does not support axis=1"
+    with pytest.raises(ValueError, match=msg):
+        df.apply(how, axis=1, args=args)
+
+
+def test_transform_axis_1_raises():
+    # GH 35964
+    msg = "No axis named 1 for object type Series"
+    with pytest.raises(ValueError, match=msg):
+        Series([1]).transform("sum", axis=1)
+
+
+def test_apply_modify_traceback():
+    data = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    data.loc[4, "C"] = np.nan
+
+    def transform(row):
+        if row["C"].startswith("shin") and row["A"] == "foo":
+            row["D"] = 7
+        return row
+
+    msg = "'float' object has no attribute 'startswith'"
+    with pytest.raises(AttributeError, match=msg):
+        data.apply(transform, axis=1)
+
+
+@pytest.mark.parametrize(
+    "df, func, expected",
+    tm.get_cython_table_params(
+        DataFrame([["a", "b"], ["b", "a"]]), [["cumprod", TypeError]]
+    ),
+)
+def test_agg_cython_table_raises_frame(df, func, expected, axis, using_infer_string):
+    # GH 21224
+    if using_infer_string:
+        expected = (expected, NotImplementedError)
+
+    msg = (
+        "can't multiply sequence by non-int of type 'str'"
+        "|cannot perform cumprod with type str"  # NotImplementedError python backend
+        "|operation 'cumprod' not supported for dtype 'str'"  # TypeError pyarrow
+    )
+    warn = None if isinstance(func, str) else FutureWarning
+    with pytest.raises(expected, match=msg):
+        with tm.assert_produces_warning(warn, match="using DataFrame.cumprod"):
+            df.agg(func, axis=axis)
+
+
+@pytest.mark.parametrize(
+    "series, func, expected",
+    chain(
+        tm.get_cython_table_params(
+            Series("a b c".split()),
+            [
+                ("mean", TypeError),  # mean raises TypeError
+                ("prod", TypeError),
+                ("std", TypeError),
+                ("var", TypeError),
+                ("median", TypeError),
+                ("cumprod", TypeError),
+            ],
+        )
+    ),
+)
+def test_agg_cython_table_raises_series(series, func, expected, using_infer_string):
+    # GH21224
+    msg = r"[Cc]ould not convert|can't multiply sequence by non-int of type"
+    if func == "median" or func is np.nanmedian or func is np.median:
+        msg = r"Cannot convert \['a' 'b' 'c'\] to numeric"
+
+    if using_infer_string and func == "cumprod":
+        expected = (expected, NotImplementedError)
+
+    msg = (
+        msg + "|does not support|has no kernel|Cannot perform|cannot perform|operation"
+    )
+    warn = None if isinstance(func, str) else FutureWarning
+
+    with pytest.raises(expected, match=msg):
+        # e.g. Series('a b'.split()).cumprod() will raise
+        with tm.assert_produces_warning(warn, match="is currently using Series.*"):
+            series.agg(func)
+
+
+def test_agg_none_to_type():
+    # GH 40543
+    df = DataFrame({"a": [None]})
+    msg = re.escape("int() argument must be a string")
+    with pytest.raises(TypeError, match=msg):
+        df.agg({"a": lambda x: int(x.iloc[0])})
+
+
+def test_transform_none_to_type():
+    # GH#34377
+    df = DataFrame({"a": [None]})
+    msg = "argument must be a"
+    with pytest.raises(TypeError, match=msg):
+        df.transform({"a": lambda x: int(x.iloc[0])})
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x: np.array([1, 2]).reshape(-1, 2),
+        lambda x: [1, 2],
+        lambda x: Series([1, 2]),
+    ],
+)
+def test_apply_broadcast_error(func):
+    df = DataFrame(
+        np.tile(np.arange(3, dtype="int64"), 6).reshape(6, -1) + 1,
+        columns=["A", "B", "C"],
+    )
+
+    # > 1 ndim
+    msg = "too many dims to broadcast|cannot broadcast result"
+    with pytest.raises(ValueError, match=msg):
+        df.apply(func, axis=1, result_type="broadcast")
+
+
+def test_transform_and_agg_err_agg(axis, float_frame):
+    # cannot both transform and agg
+    msg = "cannot combine transform and aggregation operations"
+    with pytest.raises(ValueError, match=msg):
+        with np.errstate(all="ignore"):
+            float_frame.agg(["max", "sqrt"], axis=axis)
+
+
+@pytest.mark.filterwarnings("ignore::FutureWarning")  # GH53325
+@pytest.mark.parametrize(
+    "func, msg",
+    [
+        (["sqrt", "max"], "cannot combine transform and aggregation"),
+        (
+            {"foo": np.sqrt, "bar": "sum"},
+            "cannot perform both aggregation and transformation",
+        ),
+    ],
+)
+def test_transform_and_agg_err_series(string_series, func, msg):
+    # we are trying to transform with an aggregator
+    with pytest.raises(ValueError, match=msg):
+        with np.errstate(all="ignore"):
+            string_series.agg(func)
+
+
+@pytest.mark.parametrize("func", [["max", "min"], ["max", "sqrt"]])
+def test_transform_wont_agg_frame(axis, float_frame, func):
+    # GH 35964
+    # cannot both transform and agg
+    msg = "Function did not transform"
+    with pytest.raises(ValueError, match=msg):
+        float_frame.transform(func, axis=axis)
+
+
+@pytest.mark.parametrize("func", [["min", "max"], ["sqrt", "max"]])
+def test_transform_wont_agg_series(string_series, func):
+    # GH 35964
+    # we are trying to transform with an aggregator
+    msg = "Function did not transform"
+
+    with pytest.raises(ValueError, match=msg):
+        string_series.transform(func)
+
+
+@pytest.mark.parametrize(
+    "op_wrapper", [lambda x: x, lambda x: [x], lambda x: {"A": x}, lambda x: {"A": [x]}]
+)
+def test_transform_reducer_raises(all_reductions, frame_or_series, op_wrapper):
+    # GH 35964
+    op = op_wrapper(all_reductions)
+
+    obj = DataFrame({"A": [1, 2, 3]})
+    obj = tm.get_obj(obj, frame_or_series)
+
+    msg = "Function did not transform"
+    with pytest.raises(ValueError, match=msg):
+        obj.transform(op)
+
+
+def test_transform_missing_labels_raises():
+    # GH 58474
+    df = DataFrame({"foo": [2, 4, 6], "bar": [1, 2, 3]}, index=["A", "B", "C"])
+    msg = r"Label\(s\) \['A', 'B'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        df.transform({"A": lambda x: x + 2, "B": lambda x: x * 2}, axis=0)
+
+    msg = r"Label\(s\) \['bar', 'foo'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        df.transform({"foo": lambda x: x + 2, "bar": lambda x: x * 2}, axis=1)
diff --git a/pandas/pandas/tests/apply/test_numba.py b/pandas/pandas/tests/apply/test_numba.py
new file mode 100644
index 0000000000000..75bc3f5b74b9d
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_numba.py
@@ -0,0 +1,129 @@
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [td.skip_if_no("numba"), pytest.mark.single_cpu, pytest.mark.skipif()]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+@pytest.fixture(params=[0, 1])
+def apply_axis(request):
+    return request.param
+
+
+def test_numba_vs_python_noop(float_frame, apply_axis):
+    func = lambda x: x
+    result = float_frame.apply(func, engine="numba", axis=apply_axis)
+    expected = float_frame.apply(func, engine="python", axis=apply_axis)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_numba_vs_python_string_index():
+    # GH#56189
+    df = DataFrame(
+        1,
+        index=Index(["a", "b"], dtype=pd.StringDtype(na_value=np.nan)),
+        columns=Index(["x", "y"], dtype=pd.StringDtype(na_value=np.nan)),
+    )
+    func = lambda x: x
+    result = df.apply(func, engine="numba", axis=0)
+    expected = df.apply(func, engine="python", axis=0)
+    tm.assert_frame_equal(
+        result, expected, check_column_type=False, check_index_type=False
+    )
+
+
+def test_numba_vs_python_indexing():
+    frame = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": [7.0, 8.0, 9.0]},
+        index=Index(["A", "B", "C"]),
+    )
+    row_func = lambda x: x["c"]
+    result = frame.apply(row_func, engine="numba", axis=1)
+    expected = frame.apply(row_func, engine="python", axis=1)
+    tm.assert_series_equal(result, expected)
+
+    col_func = lambda x: x["A"]
+    result = frame.apply(col_func, engine="numba", axis=0)
+    expected = frame.apply(col_func, engine="python", axis=0)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "reduction",
+    [lambda x: x.mean(), lambda x: x.min(), lambda x: x.max(), lambda x: x.sum()],
+)
+def test_numba_vs_python_reductions(reduction, apply_axis):
+    df = DataFrame(np.ones((4, 4), dtype=np.float64))
+    result = df.apply(reduction, engine="numba", axis=apply_axis)
+    expected = df.apply(reduction, engine="python", axis=apply_axis)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("colnames", [[1, 2, 3], [1.0, 2.0, 3.0]])
+def test_numba_numeric_colnames(colnames):
+    # Check that numeric column names lower properly and can be indexed on
+    df = DataFrame(
+        np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype=np.int64), columns=colnames
+    )
+    first_col = colnames[0]
+    f = lambda x: x[first_col]  # Get the first column
+    result = df.apply(f, engine="numba", axis=1)
+    expected = df.apply(f, engine="python", axis=1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_numba_parallel_unsupported(float_frame):
+    f = lambda x: x
+    with pytest.raises(
+        NotImplementedError,
+        match="Parallel apply is not supported when raw=False and engine='numba'",
+    ):
+        float_frame.apply(f, engine="numba", engine_kwargs={"parallel": True})
+
+
+def test_numba_nonunique_unsupported(apply_axis):
+    f = lambda x: x
+    df = DataFrame({"a": [1, 2]}, index=Index(["a", "a"]))
+    with pytest.raises(
+        NotImplementedError,
+        match="The index/columns must be unique when raw=False and engine='numba'",
+    ):
+        df.apply(f, engine="numba", axis=apply_axis)
+
+
+def test_numba_unsupported_dtypes(apply_axis):
+    pytest.importorskip("pyarrow")
+    f = lambda x: x
+    df = DataFrame({"a": [1, 2], "b": ["a", "b"], "c": [4, 5]})
+    df["c"] = df["c"].astype("double[pyarrow]")
+
+    with pytest.raises(
+        ValueError,
+        match="Column b must have a numeric dtype. Found 'object|str' instead",
+    ):
+        df.apply(f, engine="numba", axis=apply_axis)
+
+    with pytest.raises(
+        ValueError,
+        match="Column c is backed by an extension array, "
+        "which is not supported by the numba engine.",
+    ):
+        df["c"].to_frame().apply(f, engine="numba", axis=apply_axis)
diff --git a/pandas/pandas/tests/apply/test_series_apply.py b/pandas/pandas/tests/apply/test_series_apply.py
new file mode 100644
index 0000000000000..9541b0b7495c7
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_series_apply.py
@@ -0,0 +1,667 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.tests.apply.common import series_transform_kernels
+
+
+@pytest.fixture(params=[False, "compat"])
+def by_row(request):
+    return request.param
+
+
+def test_series_map_box_timedelta(by_row):
+    # GH#11349
+    ser = Series(timedelta_range("1 day 1 s", periods=3, freq="h"))
+
+    def f(x):
+        return x.total_seconds() if by_row else x.dt.total_seconds()
+
+    result = ser.apply(f, by_row=by_row)
+
+    expected = ser.map(lambda x: x.total_seconds())
+    tm.assert_series_equal(result, expected)
+
+    expected = Series([86401.0, 90001.0, 93601.0])
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply(datetime_series, by_row):
+    result = datetime_series.apply(np.sqrt, by_row=by_row)
+    with np.errstate(all="ignore"):
+        expected = np.sqrt(datetime_series)
+    tm.assert_series_equal(result, expected)
+
+    # element-wise apply (ufunc)
+    result = datetime_series.apply(np.exp, by_row=by_row)
+    expected = np.exp(datetime_series)
+    tm.assert_series_equal(result, expected)
+
+    # empty series
+    s = Series(dtype=object, name="foo", index=Index([], name="bar"))
+    rs = s.apply(lambda x: x, by_row=by_row)
+    tm.assert_series_equal(s, rs)
+
+    # check all metadata (GH 9322)
+    assert s is not rs
+    assert s.index is rs.index
+    assert s.dtype == rs.dtype
+    assert s.name == rs.name
+
+    # index but no data
+    s = Series(index=[1, 2, 3], dtype=np.float64)
+    rs = s.apply(lambda x: x, by_row=by_row)
+    tm.assert_series_equal(s, rs)
+
+
+def test_apply_map_same_length_inference_bug():
+    s = Series([1, 2])
+
+    def f(x):
+        return (x, x + 1)
+
+    result = s.apply(f, by_row="compat")
+    expected = s.map(f)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_args():
+    s = Series(["foo,bar"])
+
+    result = s.apply(str.split, args=(",",))
+    assert result[0] == ["foo", "bar"]
+    assert isinstance(result[0], list)
+
+
+@pytest.mark.parametrize(
+    "args, kwargs, increment",
+    [((), {}, 0), ((), {"a": 1}, 1), ((2, 3), {}, 32), ((1,), {"c": 2}, 201)],
+)
+def test_agg_args(args, kwargs, increment):
+    # GH 43357
+    def f(x, a=0, b=0, c=0):
+        return x + a + 10 * b + 100 * c
+
+    s = Series([1, 2])
+    result = s.agg(f, 0, *args, **kwargs)
+    expected = s + increment
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_mapping_func_deprecated():
+    # GH 53325
+    s = Series([1, 2, 3])
+
+    def foo1(x, a=1, c=0):
+        return x + a + c
+
+    def foo2(x, b=2, c=0):
+        return x + b + c
+
+    s.agg(foo1, 0, 3, c=4)
+    s.agg([foo1, foo2], 0, 3, c=4)
+    s.agg({"a": foo1, "b": foo2}, 0, 3, c=4)
+
+
+def test_series_apply_map_box_timestamps(by_row):
+    # GH#2689, GH#2627
+    ser = Series(date_range("1/1/2000", periods=10))
+
+    def func(x):
+        return (x.hour, x.day, x.month)
+
+    if not by_row:
+        msg = "Series' object has no attribute 'hour'"
+        with pytest.raises(AttributeError, match=msg):
+            ser.apply(func, by_row=by_row)
+        return
+
+    result = ser.apply(func, by_row=by_row)
+    expected = ser.map(func)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_box_dt64():
+    # ufunc will not be boxed. Same test cases as the test_map_box
+    vals = [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-01-02")]
+    ser = Series(vals, dtype="M8[ns]")
+    assert ser.dtype == "datetime64[ns]"
+    # boxed value must be Timestamp instance
+    res = ser.apply(lambda x: f"{type(x).__name__}_{x.day}_{x.tz}", by_row="compat")
+    exp = Series(["Timestamp_1_None", "Timestamp_2_None"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_apply_box_dt64tz():
+    vals = [
+        pd.Timestamp("2011-01-01", tz="US/Eastern"),
+        pd.Timestamp("2011-01-02", tz="US/Eastern"),
+    ]
+    ser = Series(vals, dtype="M8[ns, US/Eastern]")
+    assert ser.dtype == "datetime64[ns, US/Eastern]"
+    res = ser.apply(lambda x: f"{type(x).__name__}_{x.day}_{x.tz}", by_row="compat")
+    exp = Series(["Timestamp_1_US/Eastern", "Timestamp_2_US/Eastern"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_apply_box_td64():
+    # timedelta
+    vals = [pd.Timedelta("1 days"), pd.Timedelta("2 days")]
+    ser = Series(vals)
+    assert ser.dtype == "timedelta64[ns]"
+    res = ser.apply(lambda x: f"{type(x).__name__}_{x.days}", by_row="compat")
+    exp = Series(["Timedelta_1", "Timedelta_2"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_apply_box_period():
+    # period
+    vals = [pd.Period("2011-01-01", freq="M"), pd.Period("2011-01-02", freq="M")]
+    ser = Series(vals)
+    assert ser.dtype == "Period[M]"
+    res = ser.apply(lambda x: f"{type(x).__name__}_{x.freqstr}", by_row="compat")
+    exp = Series(["Period_M", "Period_M"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_apply_datetimetz(by_row):
+    values = date_range("2011-01-01", "2011-01-02", freq="h").tz_localize("Asia/Tokyo")
+    s = Series(values, name="XX")
+
+    result = s.apply(lambda x: x + pd.offsets.Day(), by_row=by_row)
+    exp_values = date_range("2011-01-02", "2011-01-03", freq="h").tz_localize(
+        "Asia/Tokyo"
+    )
+    exp = Series(exp_values, name="XX")
+    tm.assert_series_equal(result, exp)
+
+    result = s.apply(lambda x: x.hour if by_row else x.dt.hour, by_row=by_row)
+    exp = Series(list(range(24)) + [0], name="XX", dtype="int64" if by_row else "int32")
+    tm.assert_series_equal(result, exp)
+
+    # not vectorized
+    def f(x):
+        return str(x.tz) if by_row else str(x.dt.tz)
+
+    result = s.apply(f, by_row=by_row)
+    if by_row:
+        exp = Series(["Asia/Tokyo"] * 25, name="XX")
+        tm.assert_series_equal(result, exp)
+    else:
+        assert result == "Asia/Tokyo"
+
+
+def test_apply_categorical(by_row, using_infer_string):
+    values = pd.Categorical(list("ABBABCD"), categories=list("DCBA"), ordered=True)
+    ser = Series(values, name="XX", index=list("abcdefg"))
+
+    if not by_row:
+        msg = "Series' object has no attribute 'lower"
+        with pytest.raises(AttributeError, match=msg):
+            ser.apply(lambda x: x.lower(), by_row=by_row)
+        assert ser.apply(lambda x: "A", by_row=by_row) == "A"
+        return
+
+    result = ser.apply(lambda x: x.lower(), by_row=by_row)
+
+    # should be categorical dtype when the number of categories are
+    # the same
+    values = pd.Categorical(list("abbabcd"), categories=list("dcba"), ordered=True)
+    exp = Series(values, name="XX", index=list("abcdefg"))
+    tm.assert_series_equal(result, exp)
+    tm.assert_categorical_equal(result.values, exp.values)
+
+    result = ser.apply(lambda x: "A")
+    exp = Series(["A"] * 7, name="XX", index=list("abcdefg"))
+    tm.assert_series_equal(result, exp)
+    assert result.dtype == object if not using_infer_string else "str"
+
+
+@pytest.mark.parametrize("series", [["1-1", "1-1", np.nan], ["1-1", "1-2", np.nan]])
+def test_apply_categorical_with_nan_values(series, by_row):
+    # GH 20714 bug fixed in: GH 24275
+    s = Series(series, dtype="category")
+    if not by_row:
+        msg = "'Series' object has no attribute 'split'"
+        with pytest.raises(AttributeError, match=msg):
+            s.apply(lambda x: x.split("-")[0], by_row=by_row)
+        return
+    # NaN for cat dtype fixed in (GH 59966)
+    result = s.apply(lambda x: x.split("-")[0] if pd.notna(x) else False, by_row=by_row)
+    result = result.astype(object)
+    expected = Series(["1", "1", False], dtype="category")
+    expected = expected.astype(object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_empty_integer_series_with_datetime_index(by_row):
+    # GH 21245
+    s = Series([], index=date_range(start="2018-01-01", periods=0), dtype=int)
+    result = s.apply(lambda x: x, by_row=by_row)
+    tm.assert_series_equal(result, s)
+
+
+def test_apply_dataframe_iloc():
+    uintDF = DataFrame(np.uint64([1, 2, 3, 4, 5]), columns=["Numbers"])
+    indexDF = DataFrame([2, 3, 2, 1, 2], columns=["Indices"])
+
+    def retrieve(targetRow, targetDF):
+        val = targetDF["Numbers"].iloc[targetRow]
+        return val
+
+    result = indexDF["Indices"].apply(retrieve, args=(uintDF,))
+    expected = Series([3, 4, 3, 2, 3], name="Indices", dtype="uint64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform(string_series, by_row):
+    # transforming functions
+
+    with np.errstate(all="ignore"):
+        f_sqrt = np.sqrt(string_series)
+        f_abs = np.abs(string_series)
+
+        # ufunc
+        result = string_series.apply(np.sqrt, by_row=by_row)
+        expected = f_sqrt.copy()
+        tm.assert_series_equal(result, expected)
+
+        # list-like
+        result = string_series.apply([np.sqrt], by_row=by_row)
+        expected = f_sqrt.to_frame().copy()
+        expected.columns = ["sqrt"]
+        tm.assert_frame_equal(result, expected)
+
+        result = string_series.apply(["sqrt"], by_row=by_row)
+        tm.assert_frame_equal(result, expected)
+
+        # multiple items in list
+        # these are in the order as if we are applying both functions per
+        # series and then concatting
+        expected = concat([f_sqrt, f_abs], axis=1)
+        expected.columns = ["sqrt", "absolute"]
+        result = string_series.apply([np.sqrt, np.abs], by_row=by_row)
+        tm.assert_frame_equal(result, expected)
+
+        # dict, provide renaming
+        expected = concat([f_sqrt, f_abs], axis=1)
+        expected.columns = ["foo", "bar"]
+        expected = expected.unstack().rename("series")
+
+        result = string_series.apply({"foo": np.sqrt, "bar": np.abs}, by_row=by_row)
+        tm.assert_series_equal(result.reindex_like(expected), expected)
+
+
+@pytest.mark.parametrize("op", series_transform_kernels)
+def test_transform_partial_failure(op, request):
+    # GH 35964
+    if op in ("ffill", "bfill", "shift"):
+        request.applymarker(
+            pytest.mark.xfail(reason=f"{op} is successful on any dtype")
+        )
+
+    # Using object makes most transform kernels fail
+    ser = Series(3 * [object])
+
+    if op in ("fillna", "ngroup"):
+        error = ValueError
+        msg = "Transform function failed"
+    else:
+        error = TypeError
+        msg = "|".join(
+            [
+                "not supported between instances of 'type' and 'type'",
+                "unsupported operand type",
+            ]
+        )
+
+    with pytest.raises(error, match=msg):
+        ser.transform([op, "shift"])
+
+    with pytest.raises(error, match=msg):
+        ser.transform({"A": op, "B": "shift"})
+
+    with pytest.raises(error, match=msg):
+        ser.transform({"A": [op], "B": ["shift"]})
+
+    with pytest.raises(error, match=msg):
+        ser.transform({"A": [op, "shift"], "B": [op]})
+
+
+def test_transform_partial_failure_valueerror():
+    # GH 40211
+    def noop(x):
+        return x
+
+    def raising_op(_):
+        raise ValueError
+
+    ser = Series(3 * [object])
+    msg = "Transform function failed"
+
+    with pytest.raises(ValueError, match=msg):
+        ser.transform([noop, raising_op])
+
+    with pytest.raises(ValueError, match=msg):
+        ser.transform({"A": raising_op, "B": noop})
+
+    with pytest.raises(ValueError, match=msg):
+        ser.transform({"A": [raising_op], "B": [noop]})
+
+    with pytest.raises(ValueError, match=msg):
+        ser.transform({"A": [noop, raising_op], "B": [noop]})
+
+
+def test_demo():
+    # demonstration tests
+    s = Series(range(6), dtype="int64", name="series")
+
+    result = s.agg(["min", "max"])
+    expected = Series([0, 5], index=["min", "max"], name="series")
+    tm.assert_series_equal(result, expected)
+
+    result = s.agg({"foo": "min"})
+    expected = Series([0], index=["foo"], name="series")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", [str, lambda x: str(x)])
+def test_apply_map_evaluate_lambdas_the_same(string_series, func, by_row):
+    # test that we are evaluating row-by-row first if by_row="compat"
+    # else vectorized evaluation
+    result = string_series.apply(func, by_row=by_row)
+
+    if by_row:
+        expected = string_series.map(func)
+        tm.assert_series_equal(result, expected)
+    else:
+        assert result == str(string_series)
+
+
+def test_agg_evaluate_lambdas(string_series):
+    # GH53325
+    result = string_series.agg(lambda x: type(x))
+    assert result is Series
+
+    result = string_series.agg(type)
+    assert result is Series
+
+
+@pytest.mark.parametrize("op_name", ["agg", "apply"])
+def test_with_nested_series(datetime_series, op_name):
+    # GH 2316 & GH52123
+    # .agg with a reducer and a transform, what to do
+    result = getattr(datetime_series, op_name)(
+        lambda x: Series([x, x**2], index=["x", "x^2"])
+    )
+    if op_name == "apply":
+        expected = DataFrame({"x": datetime_series, "x^2": datetime_series**2})
+        tm.assert_frame_equal(result, expected)
+    else:
+        expected = Series([datetime_series, datetime_series**2], index=["x", "x^2"])
+        tm.assert_series_equal(result, expected)
+
+
+def test_replicate_describe(string_series):
+    # this also tests a result set that is all scalars
+    expected = string_series.describe()
+    result = string_series.apply(
+        {
+            "count": "count",
+            "mean": "mean",
+            "std": "std",
+            "min": "min",
+            "25%": lambda x: x.quantile(0.25),
+            "50%": "median",
+            "75%": lambda x: x.quantile(0.75),
+            "max": "max",
+        },
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_reduce(string_series):
+    # reductions with named functions
+    result = string_series.agg(["sum", "mean"])
+    expected = Series(
+        [string_series.sum(), string_series.mean()],
+        ["sum", "mean"],
+        name=string_series.name,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "how, kwds",
+    [("agg", {}), ("apply", {"by_row": "compat"}), ("apply", {"by_row": False})],
+)
+def test_non_callable_aggregates(how, kwds):
+    # test agg using non-callable series attributes
+    # GH 39116 - expand to apply
+    s = Series([1, 2, None])
+
+    # Calling agg w/ just a string arg same as calling s.arg
+    result = getattr(s, how)("size", **kwds)
+    expected = s.size
+    assert result == expected
+
+    # test when mixed w/ callable reducers
+    result = getattr(s, how)(["size", "count", "mean"], **kwds)
+    expected = Series({"size": 3.0, "count": 2.0, "mean": 1.5})
+    tm.assert_series_equal(result, expected)
+
+    result = getattr(s, how)({"size": "size", "count": "count", "mean": "mean"}, **kwds)
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_apply_no_suffix_index(by_row):
+    # GH36189
+    s = Series([4] * 3)
+    result = s.apply(["sum", lambda x: x.sum(), lambda x: x.sum()], by_row=by_row)
+    expected = Series([12, 12, 12], index=["sum", "<lambda>", "<lambda>"])
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dti,exp",
+    [
+        (
+            Series([1, 2], index=pd.DatetimeIndex([0, 31536000000])),
+            DataFrame(np.repeat([[1, 2]], 2, axis=0), dtype="int64"),
+        ),
+        (
+            Series(
+                np.arange(10, dtype=np.float64),
+                index=date_range("2020-01-01", periods=10),
+                name="ts",
+            ),
+            DataFrame(np.repeat([[1, 2]], 10, axis=0), dtype="int64"),
+        ),
+    ],
+)
+@pytest.mark.parametrize("aware", [True, False])
+def test_apply_series_on_date_time_index_aware_series(dti, exp, aware):
+    # GH 25959
+    # Calling apply on a localized time series should not cause an error
+    if aware:
+        index = dti.tz_localize("UTC").index
+    else:
+        index = dti.index
+    result = Series(index).apply(lambda x: Series([1, 2]))
+    tm.assert_frame_equal(result, exp)
+
+
+@pytest.mark.parametrize(
+    "by_row, expected", [("compat", Series(np.ones(10), dtype="int64")), (False, 1)]
+)
+def test_apply_scalar_on_date_time_index_aware_series(by_row, expected):
+    # GH 25959
+    # Calling apply on a localized time series should not cause an error
+    series = Series(
+        np.arange(10, dtype=np.float64),
+        index=date_range("2020-01-01", periods=10, tz="UTC"),
+    )
+    result = Series(series.index).apply(lambda x: 1, by_row=by_row)
+    tm.assert_equal(result, expected)
+
+
+def test_apply_to_timedelta(by_row):
+    list_of_valid_strings = ["00:00:01", "00:00:02"]
+    a = pd.to_timedelta(list_of_valid_strings)
+    b = Series(list_of_valid_strings).apply(pd.to_timedelta, by_row=by_row)
+    tm.assert_series_equal(Series(a), b)
+
+    list_of_strings = ["00:00:01", np.nan, pd.NaT, pd.NaT]
+
+    a = pd.to_timedelta(list_of_strings)
+    ser = Series(list_of_strings)
+    b = ser.apply(pd.to_timedelta, by_row=by_row)
+    tm.assert_series_equal(Series(a), b)
+
+
+@pytest.mark.parametrize(
+    "ops, names",
+    [
+        ([np.sum], ["sum"]),
+        ([np.sum, np.mean], ["sum", "mean"]),
+        (np.array([np.sum]), ["sum"]),
+        (np.array([np.sum, np.mean]), ["sum", "mean"]),
+    ],
+)
+@pytest.mark.parametrize(
+    "how, kwargs",
+    [["agg", {}], ["apply", {"by_row": "compat"}], ["apply", {"by_row": False}]],
+)
+def test_apply_listlike_reducer(string_series, ops, names, how, kwargs):
+    # GH 39140
+    expected = Series({name: op(string_series) for name, op in zip(names, ops)})
+    expected.name = "series"
+    result = getattr(string_series, how)(ops, **kwargs)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops",
+    [
+        {"A": np.sum},
+        {"A": np.sum, "B": np.mean},
+        Series({"A": np.sum}),
+        Series({"A": np.sum, "B": np.mean}),
+    ],
+)
+@pytest.mark.parametrize(
+    "how, kwargs",
+    [["agg", {}], ["apply", {"by_row": "compat"}], ["apply", {"by_row": False}]],
+)
+def test_apply_dictlike_reducer(string_series, ops, how, kwargs, by_row):
+    # GH 39140
+    expected = Series({name: op(string_series) for name, op in ops.items()})
+    expected.name = string_series.name
+    result = getattr(string_series, how)(ops, **kwargs)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, names",
+    [
+        ([np.sqrt], ["sqrt"]),
+        ([np.abs, np.sqrt], ["absolute", "sqrt"]),
+        (np.array([np.sqrt]), ["sqrt"]),
+        (np.array([np.abs, np.sqrt]), ["absolute", "sqrt"]),
+    ],
+)
+def test_apply_listlike_transformer(string_series, ops, names, by_row):
+    # GH 39140
+    with np.errstate(all="ignore"):
+        expected = concat([op(string_series) for op in ops], axis=1)
+        expected.columns = names
+        result = string_series.apply(ops, by_row=by_row)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, expected",
+    [
+        ([lambda x: x], DataFrame({"<lambda>": [1, 2, 3]})),
+        ([lambda x: x.sum()], Series([6], index=["<lambda>"])),
+    ],
+)
+def test_apply_listlike_lambda(ops, expected, by_row):
+    # GH53400
+    ser = Series([1, 2, 3])
+    result = ser.apply(ops, by_row=by_row)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops",
+    [
+        {"A": np.sqrt},
+        {"A": np.sqrt, "B": np.exp},
+        Series({"A": np.sqrt}),
+        Series({"A": np.sqrt, "B": np.exp}),
+    ],
+)
+def test_apply_dictlike_transformer(string_series, ops, by_row):
+    # GH 39140
+    with np.errstate(all="ignore"):
+        expected = concat({name: op(string_series) for name, op in ops.items()})
+        expected.name = string_series.name
+        result = string_series.apply(ops, by_row=by_row)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, expected",
+    [
+        (
+            {"a": lambda x: x},
+            Series([1, 2, 3], index=MultiIndex.from_arrays([["a"] * 3, range(3)])),
+        ),
+        ({"a": lambda x: x.sum()}, Series([6], index=["a"])),
+    ],
+)
+def test_apply_dictlike_lambda(ops, by_row, expected):
+    # GH53400
+    ser = Series([1, 2, 3])
+    result = ser.apply(ops, by_row=by_row)
+    tm.assert_equal(result, expected)
+
+
+def test_apply_retains_column_name(by_row):
+    # GH 16380
+    df = DataFrame({"x": range(3)}, Index(range(3), name="x"))
+    result = df.x.apply(lambda x: Series(range(x + 1), Index(range(x + 1), name="y")))
+    expected = DataFrame(
+        [[0.0, np.nan, np.nan], [0.0, 1.0, np.nan], [0.0, 1.0, 2.0]],
+        columns=Index(range(3), name="y"),
+        index=Index(range(3), name="x"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_type():
+    # GH 46719
+    s = Series([3, "string", float], index=["a", "b", "c"])
+    result = s.apply(type)
+    expected = Series([int, str, type], index=["a", "b", "c"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_apply_unpack_nested_data():
+    # GH#55189
+    ser = Series([[1, 2, 3], [4, 5, 6, 7]])
+    result = ser.apply(lambda x: Series(x))
+    expected = DataFrame({0: [1.0, 4.0], 1: [2.0, 5.0], 2: [3.0, 6.0], 3: [np.nan, 7]})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/apply/test_series_apply_relabeling.py b/pandas/pandas/tests/apply/test_series_apply_relabeling.py
new file mode 100644
index 0000000000000..c0a285e6eb38c
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_series_apply_relabeling.py
@@ -0,0 +1,33 @@
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_relabel_no_duplicated_method():
+    # this is to test there is no duplicated method used in agg
+    df = pd.DataFrame({"A": [1, 2, 1, 2], "B": [1, 2, 3, 4]})
+
+    result = df["A"].agg(foo="sum")
+    expected = df["A"].agg({"foo": "sum"})
+    tm.assert_series_equal(result, expected)
+
+    result = df["B"].agg(foo="min", bar="max")
+    expected = df["B"].agg({"foo": "min", "bar": "max"})
+    tm.assert_series_equal(result, expected)
+
+    result = df["B"].agg(foo=sum, bar=min, cat="max")
+    expected = df["B"].agg({"foo": sum, "bar": min, "cat": "max"})
+    tm.assert_series_equal(result, expected)
+
+
+def test_relabel_duplicated_method():
+    # this is to test with nested renaming, duplicated method can be used
+    # if they are assigned with different new names
+    df = pd.DataFrame({"A": [1, 2, 1, 2], "B": [1, 2, 3, 4]})
+
+    result = df["A"].agg(foo="sum", bar="sum")
+    expected = pd.Series([6, 6], index=["foo", "bar"], name="A")
+    tm.assert_series_equal(result, expected)
+
+    result = df["B"].agg(foo=min, bar="min")
+    expected = pd.Series([1, 1], index=["foo", "bar"], name="B")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/apply/test_series_transform.py b/pandas/pandas/tests/apply/test_series_transform.py
new file mode 100644
index 0000000000000..82592c4711ece
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_series_transform.py
@@ -0,0 +1,84 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "args, kwargs, increment",
+    [((), {}, 0), ((), {"a": 1}, 1), ((2, 3), {}, 32), ((1,), {"c": 2}, 201)],
+)
+def test_agg_args(args, kwargs, increment):
+    # GH 43357
+    def f(x, a=0, b=0, c=0):
+        return x + a + 10 * b + 100 * c
+
+    s = Series([1, 2])
+    result = s.transform(f, 0, *args, **kwargs)
+    expected = s + increment
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ops, names",
+    [
+        ([np.sqrt], ["sqrt"]),
+        ([np.abs, np.sqrt], ["absolute", "sqrt"]),
+        (np.array([np.sqrt]), ["sqrt"]),
+        (np.array([np.abs, np.sqrt]), ["absolute", "sqrt"]),
+    ],
+)
+def test_transform_listlike(string_series, ops, names):
+    # GH 35964
+    with np.errstate(all="ignore"):
+        expected = concat([op(string_series) for op in ops], axis=1)
+        expected.columns = names
+        result = string_series.transform(ops)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_transform_listlike_func_with_args():
+    # GH 50624
+
+    s = Series([1, 2, 3])
+
+    def foo1(x, a=1, c=0):
+        return x + a + c
+
+    def foo2(x, b=2, c=0):
+        return x + b + c
+
+    msg = r"foo1\(\) got an unexpected keyword argument 'b'"
+    with pytest.raises(TypeError, match=msg):
+        s.transform([foo1, foo2], 0, 3, b=3, c=4)
+
+    result = s.transform([foo1, foo2], 0, 3, c=4)
+    expected = DataFrame({"foo1": [8, 9, 10], "foo2": [8, 9, 10]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [dict, Series])
+def test_transform_dictlike(string_series, box):
+    # GH 35964
+    with np.errstate(all="ignore"):
+        expected = concat([np.sqrt(string_series), np.abs(string_series)], axis=1)
+    expected.columns = ["foo", "bar"]
+    result = string_series.transform(box({"foo": np.sqrt, "bar": np.abs}))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_dictlike_mixed():
+    # GH 40018 - mix of lists and non-lists in values of a dictionary
+    df = Series([1, 4])
+    result = df.transform({"b": ["sqrt", "abs"], "c": "sqrt"})
+    expected = DataFrame(
+        [[1.0, 1, 1.0], [2.0, 4, 2.0]],
+        columns=MultiIndex([("b", "c"), ("sqrt", "abs")], [(0, 0, 1), (0, 1, 0)]),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/apply/test_str.py b/pandas/pandas/tests/apply/test_str.py
new file mode 100644
index 0000000000000..e5a9492630b13
--- /dev/null
+++ b/pandas/pandas/tests/apply/test_str.py
@@ -0,0 +1,307 @@
+from itertools import chain
+import operator
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    WASM,
+)
+
+from pandas.core.dtypes.common import is_number
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.apply.common import (
+    frame_transform_kernels,
+    series_transform_kernels,
+)
+
+
+@pytest.mark.parametrize("func", ["sum", "mean", "min", "max", "std"])
+@pytest.mark.parametrize(
+    "kwds",
+    [
+        pytest.param({}, id="no_kwds"),
+        pytest.param({"axis": 1}, id="on_axis"),
+        pytest.param({"numeric_only": True}, id="func_kwds"),
+        pytest.param({"axis": 1, "numeric_only": True}, id="axis_and_func_kwds"),
+    ],
+)
+@pytest.mark.parametrize("how", ["agg", "apply"])
+def test_apply_with_string_funcs(float_frame, func, kwds, how):
+    result = getattr(float_frame, how)(func, **kwds)
+    expected = getattr(float_frame, func)(**kwds)
+    tm.assert_series_equal(result, expected)
+
+
+def test_with_string_args(datetime_series, all_numeric_reductions):
+    result = datetime_series.apply(all_numeric_reductions)
+    expected = getattr(datetime_series, all_numeric_reductions)()
+    assert result == expected
+
+
+@pytest.mark.parametrize("op", ["mean", "median", "std", "var"])
+@pytest.mark.parametrize("how", ["agg", "apply"])
+def test_apply_np_reducer(op, how):
+    # GH 39116
+    float_frame = DataFrame({"a": [1, 2], "b": [3, 4]})
+    result = getattr(float_frame, how)(op)
+    # pandas ddof defaults to 1, numpy to 0
+    kwargs = {"ddof": 1} if op in ("std", "var") else {}
+    expected = Series(
+        getattr(np, op)(float_frame, axis=0, **kwargs), index=float_frame.columns
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.skipif(WASM, reason="No fp exception support in wasm")
+@pytest.mark.parametrize(
+    "op", ["abs", "ceil", "cos", "cumsum", "exp", "log", "sqrt", "square"]
+)
+@pytest.mark.parametrize("how", ["transform", "apply"])
+def test_apply_np_transformer(float_frame, op, how):
+    # GH 39116
+
+    # float_frame will _usually_ have negative values, which will
+    #  trigger the warning here, but let's put one in just to be sure
+    float_frame.iloc[0, 0] = -1.0
+    warn = None
+    if op in ["log", "sqrt"]:
+        warn = RuntimeWarning
+
+    with tm.assert_produces_warning(warn, check_stacklevel=False):
+        # float_frame fixture is defined in conftest.py, so we don't check the
+        # stacklevel as otherwise the test would fail.
+        result = getattr(float_frame, how)(op)
+        expected = getattr(np, op)(float_frame)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "series, func, expected",
+    chain(
+        tm.get_cython_table_params(
+            Series(dtype=np.float64),
+            [
+                ("sum", 0),
+                ("max", np.nan),
+                ("min", np.nan),
+                ("all", True),
+                ("any", False),
+                ("mean", np.nan),
+                ("prod", 1),
+                ("std", np.nan),
+                ("var", np.nan),
+                ("median", np.nan),
+            ],
+        ),
+        tm.get_cython_table_params(
+            Series([np.nan, 1, 2, 3]),
+            [
+                ("sum", 6),
+                ("max", 3),
+                ("min", 1),
+                ("all", True),
+                ("any", True),
+                ("mean", 2),
+                ("prod", 6),
+                ("std", 1),
+                ("var", 1),
+                ("median", 2),
+            ],
+        ),
+        tm.get_cython_table_params(
+            Series("a b c".split()),
+            [
+                ("sum", "abc"),
+                ("max", "c"),
+                ("min", "a"),
+                ("all", True),
+                ("any", True),
+            ],
+        ),
+    ),
+)
+def test_agg_cython_table_series(series, func, expected):
+    # GH21224
+    # test reducing functions in
+    # pandas.core.base.SelectionMixin._cython_table
+    warn = None if isinstance(func, str) else FutureWarning
+    with tm.assert_produces_warning(warn, match="is currently using Series.*"):
+        result = series.agg(func)
+    if is_number(expected):
+        assert np.isclose(result, expected, equal_nan=True)
+    else:
+        assert result == expected
+
+
+@pytest.mark.parametrize(
+    "series, func, expected",
+    chain(
+        tm.get_cython_table_params(
+            Series(dtype=np.float64),
+            [
+                ("cumprod", Series([], dtype=np.float64)),
+                ("cumsum", Series([], dtype=np.float64)),
+            ],
+        ),
+        tm.get_cython_table_params(
+            Series([np.nan, 1, 2, 3]),
+            [
+                ("cumprod", Series([np.nan, 1, 2, 6])),
+                ("cumsum", Series([np.nan, 1, 3, 6])),
+            ],
+        ),
+        tm.get_cython_table_params(
+            Series("a b c".split()), [("cumsum", Series(["a", "ab", "abc"]))]
+        ),
+    ),
+)
+def test_agg_cython_table_transform_series(series, func, expected):
+    # GH21224
+    # test transforming functions in
+    # pandas.core.base.SelectionMixin._cython_table (cumprod, cumsum)
+    warn = None if isinstance(func, str) else FutureWarning
+    with tm.assert_produces_warning(warn, match="is currently using Series.*"):
+        result = series.agg(func)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df, func, expected",
+    chain(
+        tm.get_cython_table_params(
+            DataFrame(),
+            [
+                ("sum", Series(dtype="float64")),
+                ("max", Series(dtype="float64")),
+                ("min", Series(dtype="float64")),
+                ("all", Series(dtype=bool)),
+                ("any", Series(dtype=bool)),
+                ("mean", Series(dtype="float64")),
+                ("prod", Series(dtype="float64")),
+                ("std", Series(dtype="float64")),
+                ("var", Series(dtype="float64")),
+                ("median", Series(dtype="float64")),
+            ],
+        ),
+        tm.get_cython_table_params(
+            DataFrame([[np.nan, 1], [1, 2]]),
+            [
+                ("sum", Series([1.0, 3])),
+                ("max", Series([1.0, 2])),
+                ("min", Series([1.0, 1])),
+                ("all", Series([True, True])),
+                ("any", Series([True, True])),
+                ("mean", Series([1, 1.5])),
+                ("prod", Series([1.0, 2])),
+                ("std", Series([np.nan, 0.707107])),
+                ("var", Series([np.nan, 0.5])),
+                ("median", Series([1, 1.5])),
+            ],
+        ),
+    ),
+)
+def test_agg_cython_table_frame(df, func, expected, axis):
+    # GH 21224
+    # test reducing functions in
+    # pandas.core.base.SelectionMixin._cython_table
+    warn = None if isinstance(func, str) else FutureWarning
+    with tm.assert_produces_warning(warn, match="is currently using DataFrame.*"):
+        # GH#53425
+        result = df.agg(func, axis=axis)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df, func, expected",
+    chain(
+        tm.get_cython_table_params(
+            DataFrame(), [("cumprod", DataFrame()), ("cumsum", DataFrame())]
+        ),
+        tm.get_cython_table_params(
+            DataFrame([[np.nan, 1], [1, 2]]),
+            [
+                ("cumprod", DataFrame([[np.nan, 1], [1, 2]])),
+                ("cumsum", DataFrame([[np.nan, 1], [1, 3]])),
+            ],
+        ),
+    ),
+)
+def test_agg_cython_table_transform_frame(df, func, expected, axis):
+    # GH 21224
+    # test transforming functions in
+    # pandas.core.base.SelectionMixin._cython_table (cumprod, cumsum)
+    if axis in ("columns", 1):
+        # operating blockwise doesn't let us preserve dtypes
+        expected = expected.astype("float64")
+
+    warn = None if isinstance(func, str) else FutureWarning
+    with tm.assert_produces_warning(warn, match="is currently using DataFrame.*"):
+        # GH#53425
+        result = df.agg(func, axis=axis)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", series_transform_kernels)
+def test_transform_groupby_kernel_series(request, string_series, op):
+    # GH 35964
+    if op == "ngroup":
+        request.applymarker(
+            pytest.mark.xfail(raises=ValueError, reason="ngroup not valid for NDFrame")
+        )
+    args = [0.0] if op == "fillna" else []
+    ones = np.ones(string_series.shape[0])
+
+    warn = FutureWarning if op == "fillna" else None
+    msg = "SeriesGroupBy.fillna is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        expected = string_series.groupby(ones).transform(op, *args)
+    result = string_series.transform(op, 0, *args)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", frame_transform_kernels)
+def test_transform_groupby_kernel_frame(request, float_frame, op):
+    if op == "ngroup":
+        request.applymarker(
+            pytest.mark.xfail(raises=ValueError, reason="ngroup not valid for NDFrame")
+        )
+
+    # GH 35964
+
+    args = [0.0] if op == "fillna" else []
+    ones = np.ones(float_frame.shape[0])
+    gb = float_frame.groupby(ones)
+
+    warn = FutureWarning if op == "fillna" else None
+    op_msg = "DataFrameGroupBy.fillna is deprecated"
+    with tm.assert_produces_warning(warn, match=op_msg):
+        expected = gb.transform(op, *args)
+
+    result = float_frame.transform(op, 0, *args)
+    tm.assert_frame_equal(result, expected)
+
+    # same thing, but ensuring we have multiple blocks
+    assert "E" not in float_frame.columns
+    float_frame["E"] = float_frame["A"].copy()
+    assert len(float_frame._mgr.blocks) > 1
+
+    ones = np.ones(float_frame.shape[0])
+    gb2 = float_frame.groupby(ones)
+    expected2 = gb2.transform(op, *args)
+    result2 = float_frame.transform(op, 0, *args)
+    tm.assert_frame_equal(result2, expected2)
+
+
+@pytest.mark.parametrize("method", ["abs", "shift", "pct_change", "cumsum", "rank"])
+def test_transform_method_name(method):
+    # GH 19760
+    df = DataFrame({"A": [-1, 2]})
+    result = df.transform(method)
+    expected = operator.methodcaller(method)(df)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/arithmetic/__init__.py b/pandas/pandas/tests/arithmetic/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arithmetic/common.py b/pandas/pandas/tests/arithmetic/common.py
new file mode 100644
index 0000000000000..0730729e2fd94
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/common.py
@@ -0,0 +1,163 @@
+"""
+Assertion helpers for arithmetic tests.
+"""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    array,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    BooleanArray,
+    NumpyExtensionArray,
+)
+
+
+def assert_cannot_add(left, right, msg="cannot add"):
+    """
+    Helper function to assert that two objects cannot be added.
+
+    Parameters
+    ----------
+    left : object
+        The first operand.
+    right : object
+        The second operand.
+    msg : str, default "cannot add"
+        The error message expected in the TypeError.
+    """
+    with pytest.raises(TypeError, match=msg):
+        left + right
+    with pytest.raises(TypeError, match=msg):
+        right + left
+
+
+def assert_invalid_addsub_type(left, right, msg=None):
+    """
+    Helper function to assert that two objects can
+    neither be added nor subtracted.
+
+    Parameters
+    ----------
+    left : object
+        The first operand.
+    right : object
+        The second operand.
+    msg : str or None, default None
+        The error message expected in the TypeError.
+    """
+    with pytest.raises(TypeError, match=msg):
+        left + right
+    with pytest.raises(TypeError, match=msg):
+        right + left
+    with pytest.raises(TypeError, match=msg):
+        left - right
+    with pytest.raises(TypeError, match=msg):
+        right - left
+
+
+def get_upcast_box(left, right, is_cmp: bool = False):
+    """
+    Get the box to use for 'expected' in an arithmetic or comparison operation.
+
+    Parameters
+    left : Any
+    right : Any
+    is_cmp : bool, default False
+        Whether the operation is a comparison method.
+    """
+
+    if isinstance(left, DataFrame) or isinstance(right, DataFrame):
+        return DataFrame
+    if isinstance(left, Series) or isinstance(right, Series):
+        if is_cmp and isinstance(left, Index):
+            # Index does not defer for comparisons
+            return np.array
+        return Series
+    if isinstance(left, Index) or isinstance(right, Index):
+        if is_cmp:
+            return np.array
+        return Index
+    return tm.to_array
+
+
+def assert_invalid_comparison(left, right, box):
+    """
+    Assert that comparison operations with mismatched types behave correctly.
+
+    Parameters
+    ----------
+    left : np.ndarray, ExtensionArray, Index, or Series
+    right : object
+    box : {pd.DataFrame, pd.Series, pd.Index, pd.array, tm.to_array}
+    """
+    # Not for tznaive-tzaware comparison
+
+    # Note: not quite the same as how we do this for tm.box_expected
+    xbox = box if box not in [Index, array] else np.array
+
+    def xbox2(x):
+        # Eventually we'd like this to be tighter, but for now we'll
+        #  just exclude NumpyExtensionArray[bool]
+        if isinstance(x, NumpyExtensionArray):
+            return x._ndarray
+        if isinstance(x, BooleanArray):
+            # NB: we are assuming no pd.NAs for now
+            return x.astype(bool)
+        return x
+
+    # rev_box: box to use for reversed comparisons
+    rev_box = xbox
+    if isinstance(right, Index) and isinstance(left, Series):
+        rev_box = np.array
+
+    result = xbox2(left == right)
+    expected = xbox(np.zeros(result.shape, dtype=np.bool_))
+
+    tm.assert_equal(result, expected)
+
+    result = xbox2(right == left)
+    tm.assert_equal(result, rev_box(expected))
+
+    result = xbox2(left != right)
+    tm.assert_equal(result, ~expected)
+
+    result = xbox2(right != left)
+    tm.assert_equal(result, rev_box(~expected))
+
+    msg = "|".join(
+        [
+            "Invalid comparison between",
+            "Cannot compare type",
+            "not supported between",
+            "invalid type promotion",
+            (
+                # GH#36706 npdev 1.20.0 2020-09-28
+                r"The DTypes <class 'numpy.dtype\[datetime64\]'> and "
+                r"<class 'numpy.dtype\[int64\]'> do not have a common DType. "
+                "For example they cannot be stored in a single array unless the "
+                "dtype is `object`."
+            ),
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        left < right
+    with pytest.raises(TypeError, match=msg):
+        left <= right
+    with pytest.raises(TypeError, match=msg):
+        left > right
+    with pytest.raises(TypeError, match=msg):
+        left >= right
+    with pytest.raises(TypeError, match=msg):
+        right < left
+    with pytest.raises(TypeError, match=msg):
+        right <= left
+    with pytest.raises(TypeError, match=msg):
+        right > left
+    with pytest.raises(TypeError, match=msg):
+        right >= left
diff --git a/pandas/pandas/tests/arithmetic/conftest.py b/pandas/pandas/tests/arithmetic/conftest.py
new file mode 100644
index 0000000000000..c7703b34a5e38
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/conftest.py
@@ -0,0 +1,139 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Index
+
+
+@pytest.fixture(params=[1, np.array(1, dtype=np.int64)])
+def one(request):
+    """
+    Several variants of integer value 1. The zero-dim integer array
+    behaves like an integer.
+
+    This fixture can be used to check that datetimelike indexes handle
+    addition and subtraction of integers and zero-dimensional arrays
+    of integers.
+
+    Examples
+    --------
+    dti = pd.date_range('2016-01-01', periods=2, freq='h')
+    dti
+    DatetimeIndex(['2016-01-01 00:00:00', '2016-01-01 01:00:00'],
+    dtype='datetime64[ns]', freq='h')
+    dti + one
+    DatetimeIndex(['2016-01-01 01:00:00', '2016-01-01 02:00:00'],
+    dtype='datetime64[ns]', freq='h')
+    """
+    return request.param
+
+
+zeros = [
+    box_cls([0] * 5, dtype=dtype)
+    for box_cls in [Index, np.array, pd.array]
+    for dtype in [np.int64, np.uint64, np.float64]
+]
+zeros.extend([box_cls([-0.0] * 5, dtype=np.float64) for box_cls in [Index, np.array]])
+zeros.extend([np.array(0, dtype=dtype) for dtype in [np.int64, np.uint64, np.float64]])
+zeros.extend([np.array(-0.0, dtype=np.float64)])
+zeros.extend([0, 0.0, -0.0])
+
+
+@pytest.fixture(params=zeros)
+def zero(request):
+    """
+    Several types of scalar zeros and length 5 vectors of zeros.
+
+    This fixture can be used to check that numeric-dtype indexes handle
+    division by any zero numeric-dtype.
+
+    Uses vector of length 5 for broadcasting with `numeric_idx` fixture,
+    which creates numeric-dtype vectors also of length 5.
+
+    Examples
+    --------
+    arr = RangeIndex(5)
+    arr / zeros
+    Index([nan, inf, inf, inf, inf], dtype='float64')
+    """
+    return request.param
+
+
+# ------------------------------------------------------------------
+# Scalar Fixtures
+
+
+@pytest.fixture(
+    params=[
+        pd.Timedelta("10m7s").to_pytimedelta(),
+        pd.Timedelta("10m7s"),
+        pd.Timedelta("10m7s").to_timedelta64(),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def scalar_td(request):
+    """
+    Several variants of Timedelta scalars representing 10 minutes and 7 seconds.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        pd.offsets.Day(3),
+        pd.offsets.Hour(72),
+        pd.Timedelta(days=3).to_pytimedelta(),
+        pd.Timedelta("72:00:00"),
+        np.timedelta64(3, "D"),
+        np.timedelta64(72, "h"),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def three_days(request):
+    """
+    Several timedelta-like and DateOffset objects that each represent
+    a 3-day timedelta
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        pd.offsets.Hour(2),
+        pd.offsets.Minute(120),
+        pd.Timedelta(hours=2).to_pytimedelta(),
+        pd.Timedelta(seconds=2 * 3600),
+        np.timedelta64(2, "h"),
+        np.timedelta64(120, "m"),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def two_hours(request):
+    """
+    Several timedelta-like and DateOffset objects that each represent
+    a 2-hour timedelta
+    """
+    return request.param
+
+
+_common_mismatch = [
+    pd.offsets.YearBegin(2),
+    pd.offsets.MonthBegin(1),
+    pd.offsets.Minute(),
+]
+
+
+@pytest.fixture(
+    params=[
+        np.timedelta64(4, "h"),
+        pd.Timedelta(hours=23).to_pytimedelta(),
+        pd.Timedelta("23:00:00"),
+    ]
+    + _common_mismatch
+)
+def not_daily(request):
+    """
+    Several timedelta-like and DateOffset instances that are _not_
+    compatible with Daily frequencies.
+    """
+    return request.param
diff --git a/pandas/pandas/tests/arithmetic/test_array_ops.py b/pandas/pandas/tests/arithmetic/test_array_ops.py
new file mode 100644
index 0000000000000..2c347d965bbf7
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_array_ops.py
@@ -0,0 +1,39 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas._testing as tm
+from pandas.core.ops.array_ops import (
+    comparison_op,
+    na_logical_op,
+)
+
+
+def test_na_logical_op_2d():
+    left = np.arange(8).reshape(4, 2)
+    right = left.astype(object)
+    right[0, 0] = np.nan
+
+    # Check that we fall back to the vec_binop branch
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        operator.or_(left, right)
+
+    result = na_logical_op(left, right, operator.or_)
+    expected = right
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_object_comparison_2d():
+    left = np.arange(9).reshape(3, 3).astype(object)
+    right = left.T
+
+    result = comparison_op(left, right, operator.eq)
+    expected = np.eye(3).astype(bool)
+    tm.assert_numpy_array_equal(result, expected)
+
+    # Ensure that cython doesn't raise on non-writeable arg, which
+    #  we can get from np.broadcast_to
+    right.flags.writeable = False
+    result = comparison_op(left, right, operator.ne)
+    tm.assert_numpy_array_equal(result, ~expected)
diff --git a/pandas/pandas/tests/arithmetic/test_categorical.py b/pandas/pandas/tests/arithmetic/test_categorical.py
new file mode 100644
index 0000000000000..d6f3a13ce6705
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_categorical.py
@@ -0,0 +1,25 @@
+import numpy as np
+
+from pandas import (
+    Categorical,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalComparisons:
+    def test_categorical_nan_equality(self):
+        cat = Series(Categorical(["a", "b", "c", np.nan]))
+        expected = Series([True, True, True, False])
+        result = cat == cat
+        tm.assert_series_equal(result, expected)
+
+    def test_categorical_tuple_equality(self):
+        # GH 18050
+        ser = Series([(0, 0), (0, 1), (0, 0), (1, 0), (1, 1)])
+        expected = Series([True, False, True, False, False])
+        result = ser == (0, 0)
+        tm.assert_series_equal(result, expected)
+
+        result = ser.astype("category") == (0, 0)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arithmetic/test_datetime64.py b/pandas/pandas/tests/arithmetic/test_datetime64.py
new file mode 100644
index 0000000000000..26dfcf088e74b
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_datetime64.py
@@ -0,0 +1,2482 @@
+# Arithmetic tests for DataFrame/Series/Index/Array classes that should
+# behave identically.
+# Specifically for datetime64 and datetime64tz dtypes
+from datetime import (
+    datetime,
+    time,
+    timedelta,
+    timezone,
+)
+from itertools import (
+    product,
+    starmap,
+)
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.conversion import localize_pydatetime
+from pandas._libs.tslibs.offsets import shift_months
+
+import pandas as pd
+from pandas import (
+    DateOffset,
+    DatetimeIndex,
+    NaT,
+    Period,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core import roperator
+from pandas.tests.arithmetic.common import (
+    assert_cannot_add,
+    assert_invalid_addsub_type,
+    assert_invalid_comparison,
+    get_upcast_box,
+)
+
+# ------------------------------------------------------------------
+# Comparisons
+
+
+class TestDatetime64ArrayLikeComparisons:
+    # Comparison tests for datetime64 vectors fully parametrized over
+    #  DataFrame/Series/DatetimeIndex/DatetimeArray.  Ideally all comparison
+    #  tests will eventually end up here.
+
+    def test_compare_zerodim(self, tz_naive_fixture, box_with_array):
+        # Test comparison with zero-dimensional array is unboxed
+        tz = tz_naive_fixture
+        box = box_with_array
+        dti = date_range("20130101", periods=3, tz=tz)
+
+        other = np.array(dti.to_numpy()[0])
+
+        dtarr = tm.box_expected(dti, box)
+        xbox = get_upcast_box(dtarr, other, True)
+        result = dtarr <= other
+        expected = np.array([True, False, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            "foo",
+            -1,
+            99,
+            4.0,
+            object(),
+            timedelta(days=2),
+            # GH#19800, GH#19301 datetime.date comparison raises to
+            #  match DatetimeIndex/Timestamp.  This also matches the behavior
+            #  of stdlib datetime.datetime
+            datetime(2001, 1, 1).date(),
+            # GH#19301 None and NaN are *not* cast to NaT for comparisons
+            None,
+            np.nan,
+        ],
+    )
+    def test_dt64arr_cmp_scalar_invalid(self, other, tz_naive_fixture, box_with_array):
+        # GH#22074, GH#15966
+        tz = tz_naive_fixture
+
+        rng = date_range("1/1/2000", periods=10, tz=tz)
+        dtarr = tm.box_expected(rng, box_with_array)
+        assert_invalid_comparison(dtarr, other, box_with_array)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            # GH#4968 invalid date/int comparisons
+            list(range(10)),
+            np.arange(10),
+            np.arange(10).astype(np.float32),
+            np.arange(10).astype(object),
+            pd.timedelta_range("1ns", periods=10).array,
+            np.array(pd.timedelta_range("1ns", periods=10)),
+            list(pd.timedelta_range("1ns", periods=10)),
+            pd.timedelta_range("1 Day", periods=10).astype(object),
+            pd.period_range("1971-01-01", freq="D", periods=10).array,
+            pd.period_range("1971-01-01", freq="D", periods=10).astype(object),
+        ],
+    )
+    def test_dt64arr_cmp_arraylike_invalid(
+        self, other, tz_naive_fixture, box_with_array
+    ):
+        tz = tz_naive_fixture
+
+        dta = date_range("1970-01-01", freq="ns", periods=10, tz=tz)._data
+        obj = tm.box_expected(dta, box_with_array)
+        assert_invalid_comparison(obj, other, box_with_array)
+
+    def test_dt64arr_cmp_mixed_invalid(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+
+        dta = date_range("1970-01-01", freq="h", periods=5, tz=tz)._data
+
+        other = np.array([0, 1, 2, dta[3], Timedelta(days=1)])
+        result = dta == other
+        expected = np.array([False, False, False, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dta != other
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        msg = "Invalid comparison between|Cannot compare type|not supported between"
+        with pytest.raises(TypeError, match=msg):
+            dta < other
+        with pytest.raises(TypeError, match=msg):
+            dta > other
+        with pytest.raises(TypeError, match=msg):
+            dta <= other
+        with pytest.raises(TypeError, match=msg):
+            dta >= other
+
+    def test_dt64arr_nat_comparison(self, tz_naive_fixture, box_with_array):
+        # GH#22242, GH#22163 DataFrame considered NaT == ts incorrectly
+        tz = tz_naive_fixture
+        box = box_with_array
+
+        ts = Timestamp("2021-01-01", tz=tz)
+        ser = Series([ts, NaT])
+
+        obj = tm.box_expected(ser, box)
+        xbox = get_upcast_box(obj, ts, True)
+
+        expected = Series([True, False], dtype=np.bool_)
+        expected = tm.box_expected(expected, xbox)
+
+        result = obj == ts
+        tm.assert_equal(result, expected)
+
+
+class TestDatetime64SeriesComparison:
+    # TODO: moved from tests.series.test_operators; needs cleanup
+
+    @pytest.mark.parametrize(
+        "pair",
+        [
+            (
+                [Timestamp("2011-01-01"), NaT, Timestamp("2011-01-03")],
+                [NaT, NaT, Timestamp("2011-01-03")],
+            ),
+            (
+                [Timedelta("1 days"), NaT, Timedelta("3 days")],
+                [NaT, NaT, Timedelta("3 days")],
+            ),
+            (
+                [Period("2011-01", freq="M"), NaT, Period("2011-03", freq="M")],
+                [NaT, NaT, Period("2011-03", freq="M")],
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("reverse", [True, False])
+    @pytest.mark.parametrize("dtype", [None, object])
+    @pytest.mark.parametrize(
+        "op, expected",
+        [
+            (operator.eq, [False, False, True]),
+            (operator.ne, [True, True, False]),
+            (operator.lt, [False, False, False]),
+            (operator.gt, [False, False, False]),
+            (operator.ge, [False, False, True]),
+            (operator.le, [False, False, True]),
+        ],
+    )
+    def test_nat_comparisons(
+        self,
+        dtype,
+        index_or_series,
+        reverse,
+        pair,
+        op,
+        expected,
+    ):
+        box = index_or_series
+        lhs, rhs = pair
+        if reverse:
+            # add lhs / rhs switched data
+            lhs, rhs = rhs, lhs
+
+        left = Series(lhs, dtype=dtype)
+        right = box(rhs, dtype=dtype)
+
+        result = op(left, right)
+
+        tm.assert_series_equal(result, Series(expected))
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            [Timestamp("2011-01-01"), NaT, Timestamp("2011-01-03")],
+            [Timedelta("1 days"), NaT, Timedelta("3 days")],
+            [Period("2011-01", freq="M"), NaT, Period("2011-03", freq="M")],
+        ],
+    )
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_nat_comparisons_scalar(self, dtype, data, box_with_array):
+        box = box_with_array
+
+        left = Series(data, dtype=dtype)
+        left = tm.box_expected(left, box)
+        xbox = get_upcast_box(left, NaT, True)
+
+        expected = [False, False, False]
+        expected = tm.box_expected(expected, xbox)
+        if box is pd.array and dtype is object:
+            expected = pd.array(expected, dtype="bool")
+
+        tm.assert_equal(left == NaT, expected)
+        tm.assert_equal(NaT == left, expected)
+
+        expected = [True, True, True]
+        expected = tm.box_expected(expected, xbox)
+        if box is pd.array and dtype is object:
+            expected = pd.array(expected, dtype="bool")
+        tm.assert_equal(left != NaT, expected)
+        tm.assert_equal(NaT != left, expected)
+
+        expected = [False, False, False]
+        expected = tm.box_expected(expected, xbox)
+        if box is pd.array and dtype is object:
+            expected = pd.array(expected, dtype="bool")
+        tm.assert_equal(left < NaT, expected)
+        tm.assert_equal(NaT > left, expected)
+        tm.assert_equal(left <= NaT, expected)
+        tm.assert_equal(NaT >= left, expected)
+
+        tm.assert_equal(left > NaT, expected)
+        tm.assert_equal(NaT < left, expected)
+        tm.assert_equal(left >= NaT, expected)
+        tm.assert_equal(NaT <= left, expected)
+
+    @pytest.mark.parametrize("val", [datetime(2000, 1, 4), datetime(2000, 1, 5)])
+    def test_series_comparison_scalars(self, val):
+        series = Series(date_range("1/1/2000", periods=10))
+
+        result = series > val
+        expected = Series([x > val for x in series])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "left,right", [("lt", "gt"), ("le", "ge"), ("eq", "eq"), ("ne", "ne")]
+    )
+    def test_timestamp_compare_series(self, left, right):
+        # see gh-4982
+        # Make sure we can compare Timestamps on the right AND left hand side.
+        ser = Series(date_range("20010101", periods=10), name="dates")
+        s_nat = ser.copy(deep=True)
+
+        ser[0] = Timestamp("nat")
+        ser[3] = Timestamp("nat")
+
+        left_f = getattr(operator, left)
+        right_f = getattr(operator, right)
+
+        # No NaT
+        expected = left_f(ser, Timestamp("20010109"))
+        result = right_f(Timestamp("20010109"), ser)
+        tm.assert_series_equal(result, expected)
+
+        # NaT
+        expected = left_f(ser, Timestamp("nat"))
+        result = right_f(Timestamp("nat"), ser)
+        tm.assert_series_equal(result, expected)
+
+        # Compare to Timestamp with series containing NaT
+        expected = left_f(s_nat, Timestamp("20010109"))
+        result = right_f(Timestamp("20010109"), s_nat)
+        tm.assert_series_equal(result, expected)
+
+        # Compare to NaT with series containing NaT
+        expected = left_f(s_nat, NaT)
+        result = right_f(NaT, s_nat)
+        tm.assert_series_equal(result, expected)
+
+    def test_dt64arr_timestamp_equality(self, box_with_array):
+        # GH#11034
+        box = box_with_array
+
+        ser = Series([Timestamp("2000-01-29 01:59:00"), Timestamp("2000-01-30"), NaT])
+        ser = tm.box_expected(ser, box)
+        xbox = get_upcast_box(ser, ser, True)
+
+        result = ser != ser
+        expected = tm.box_expected([False, False, True], xbox)
+        tm.assert_equal(result, expected)
+
+        if box is pd.DataFrame:
+            # alignment for frame vs series comparisons deprecated
+            #  in GH#46795 enforced 2.0
+            with pytest.raises(ValueError, match="not aligned"):
+                ser != ser[0]
+
+        else:
+            result = ser != ser[0]
+            expected = tm.box_expected([False, True, True], xbox)
+            tm.assert_equal(result, expected)
+
+        if box is pd.DataFrame:
+            # alignment for frame vs series comparisons deprecated
+            #  in GH#46795 enforced 2.0
+            with pytest.raises(ValueError, match="not aligned"):
+                ser != ser[2]
+        else:
+            result = ser != ser[2]
+            expected = tm.box_expected([True, True, True], xbox)
+            tm.assert_equal(result, expected)
+
+        result = ser == ser
+        expected = tm.box_expected([True, True, False], xbox)
+        tm.assert_equal(result, expected)
+
+        if box is pd.DataFrame:
+            # alignment for frame vs series comparisons deprecated
+            #  in GH#46795 enforced 2.0
+            with pytest.raises(ValueError, match="not aligned"):
+                ser == ser[0]
+        else:
+            result = ser == ser[0]
+            expected = tm.box_expected([True, False, False], xbox)
+            tm.assert_equal(result, expected)
+
+        if box is pd.DataFrame:
+            # alignment for frame vs series comparisons deprecated
+            #  in GH#46795 enforced 2.0
+            with pytest.raises(ValueError, match="not aligned"):
+                ser == ser[2]
+        else:
+            result = ser == ser[2]
+            expected = tm.box_expected([False, False, False], xbox)
+            tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "datetimelike",
+        [
+            Timestamp("20130101"),
+            datetime(2013, 1, 1),
+            np.datetime64("2013-01-01T00:00", "ns"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "op,expected",
+        [
+            (operator.lt, [True, False, False, False]),
+            (operator.le, [True, True, False, False]),
+            (operator.eq, [False, True, False, False]),
+            (operator.gt, [False, False, False, True]),
+        ],
+    )
+    def test_dt64_compare_datetime_scalar(self, datetimelike, op, expected):
+        # GH#17965, test for ability to compare datetime64[ns] columns
+        #  to datetimelike
+        ser = Series(
+            [
+                Timestamp("20120101"),
+                Timestamp("20130101"),
+                np.nan,
+                Timestamp("20130103"),
+            ],
+            name="A",
+        )
+        result = op(ser, datetimelike)
+        expected = Series(expected, name="A")
+        tm.assert_series_equal(result, expected)
+
+    def test_ts_series_numpy_maximum(self):
+        # GH#50864, test numpy.maximum does not fail
+        # given a TimeStamp and Series(with dtype datetime64) comparison
+        ts = Timestamp("2024-07-01")
+        ts_series = Series(
+            ["2024-06-01", "2024-07-01", "2024-08-01"],
+            dtype="datetime64[us]",
+        )
+
+        expected = Series(
+            ["2024-07-01", "2024-07-01", "2024-08-01"],
+            dtype="datetime64[us]",
+        )
+
+        tm.assert_series_equal(expected, np.maximum(ts, ts_series))
+
+
+class TestDatetimeIndexComparisons:
+    # TODO: moved from tests.indexes.test_base; parametrize and de-duplicate
+    def test_comparators(self, comparison_op):
+        index = date_range("2020-01-01", periods=10)
+        element = index[len(index) // 2]
+        element = Timestamp(element).to_datetime64()
+
+        arr = np.array(index)
+        arr_result = comparison_op(arr, element)
+        index_result = comparison_op(index, element)
+
+        assert isinstance(index_result, np.ndarray)
+        tm.assert_numpy_array_equal(arr_result, index_result)
+
+    @pytest.mark.parametrize(
+        "other",
+        [datetime(2016, 1, 1), Timestamp("2016-01-01"), np.datetime64("2016-01-01")],
+    )
+    def test_dti_cmp_datetimelike(self, other, tz_naive_fixture):
+        tz = tz_naive_fixture
+        dti = date_range("2016-01-01", periods=2, tz=tz)
+        if tz is not None:
+            if isinstance(other, np.datetime64):
+                pytest.skip(f"{type(other).__name__} is not tz aware")
+            other = localize_pydatetime(other, dti.tzinfo)
+
+        result = dti == other
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti > other
+        expected = np.array([False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti >= other
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti < other
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti <= other
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_dti_cmp_nat(self, dtype, box_with_array):
+        left = DatetimeIndex([Timestamp("2011-01-01"), NaT, Timestamp("2011-01-03")])
+        right = DatetimeIndex([NaT, NaT, Timestamp("2011-01-03")])
+
+        left = tm.box_expected(left, box_with_array)
+        right = tm.box_expected(right, box_with_array)
+        xbox = get_upcast_box(left, right, True)
+
+        lhs, rhs = left, right
+        if dtype is object:
+            lhs, rhs = left.astype(object), right.astype(object)
+
+        result = rhs == lhs
+        expected = np.array([False, False, True])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(result, expected)
+
+        result = lhs != rhs
+        expected = np.array([True, True, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(result, expected)
+
+        expected = np.array([False, False, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(lhs == NaT, expected)
+        tm.assert_equal(NaT == rhs, expected)
+
+        expected = np.array([True, True, True])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(lhs != NaT, expected)
+        tm.assert_equal(NaT != lhs, expected)
+
+        expected = np.array([False, False, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(lhs < NaT, expected)
+        tm.assert_equal(NaT > lhs, expected)
+
+    def test_dti_cmp_nat_behaves_like_float_cmp_nan(self):
+        fidx1 = pd.Index([1.0, np.nan, 3.0, np.nan, 5.0, 7.0])
+        fidx2 = pd.Index([2.0, 3.0, np.nan, np.nan, 6.0, 7.0])
+
+        didx1 = DatetimeIndex(
+            ["2014-01-01", NaT, "2014-03-01", NaT, "2014-05-01", "2014-07-01"]
+        )
+        didx2 = DatetimeIndex(
+            ["2014-02-01", "2014-03-01", NaT, NaT, "2014-06-01", "2014-07-01"]
+        )
+        darr = np.array(
+            [
+                np.datetime64("2014-02-01 00:00"),
+                np.datetime64("2014-03-01 00:00"),
+                np.datetime64("nat"),
+                np.datetime64("nat"),
+                np.datetime64("2014-06-01 00:00"),
+                np.datetime64("2014-07-01 00:00"),
+            ]
+        )
+
+        cases = [(fidx1, fidx2), (didx1, didx2), (didx1, darr)]
+
+        # Check pd.NaT is handles as the same as np.nan
+        with tm.assert_produces_warning(None):
+            for idx1, idx2 in cases:
+                result = idx1 < idx2
+                expected = np.array([True, False, False, False, True, False])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx2 > idx1
+                expected = np.array([True, False, False, False, True, False])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 <= idx2
+                expected = np.array([True, False, False, False, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx2 >= idx1
+                expected = np.array([True, False, False, False, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 == idx2
+                expected = np.array([False, False, False, False, False, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 != idx2
+                expected = np.array([True, True, True, True, True, False])
+                tm.assert_numpy_array_equal(result, expected)
+
+        with tm.assert_produces_warning(None):
+            for idx1, val in [(fidx1, np.nan), (didx1, NaT)]:
+                result = idx1 < val
+                expected = np.array([False, False, False, False, False, False])
+                tm.assert_numpy_array_equal(result, expected)
+                result = idx1 > val
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 <= val
+                tm.assert_numpy_array_equal(result, expected)
+                result = idx1 >= val
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 == val
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 != val
+                expected = np.array([True, True, True, True, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+        # Check pd.NaT is handles as the same as np.nan
+        with tm.assert_produces_warning(None):
+            for idx1, val in [(fidx1, 3), (didx1, datetime(2014, 3, 1))]:
+                result = idx1 < val
+                expected = np.array([True, False, False, False, False, False])
+                tm.assert_numpy_array_equal(result, expected)
+                result = idx1 > val
+                expected = np.array([False, False, False, False, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 <= val
+                expected = np.array([True, False, True, False, False, False])
+                tm.assert_numpy_array_equal(result, expected)
+                result = idx1 >= val
+                expected = np.array([False, False, True, False, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 == val
+                expected = np.array([False, False, True, False, False, False])
+                tm.assert_numpy_array_equal(result, expected)
+
+                result = idx1 != val
+                expected = np.array([True, True, False, True, True, True])
+                tm.assert_numpy_array_equal(result, expected)
+
+    def test_comparison_tzawareness_compat(self, comparison_op, box_with_array):
+        # GH#18162
+        op = comparison_op
+        box = box_with_array
+
+        dr = date_range("2016-01-01", periods=6)
+        dz = dr.tz_localize("US/Pacific")
+
+        dr = tm.box_expected(dr, box)
+        dz = tm.box_expected(dz, box)
+
+        if box is pd.DataFrame:
+            tolist = lambda x: x.astype(object).values.tolist()[0]
+        else:
+            tolist = list
+
+        if op not in [operator.eq, operator.ne]:
+            msg = (
+                r"Invalid comparison between dtype=datetime64\[ns.*\] "
+                "and (Timestamp|DatetimeArray|list|ndarray)"
+            )
+            with pytest.raises(TypeError, match=msg):
+                op(dr, dz)
+
+            with pytest.raises(TypeError, match=msg):
+                op(dr, tolist(dz))
+            with pytest.raises(TypeError, match=msg):
+                op(dr, np.array(tolist(dz), dtype=object))
+            with pytest.raises(TypeError, match=msg):
+                op(dz, dr)
+
+            with pytest.raises(TypeError, match=msg):
+                op(dz, tolist(dr))
+            with pytest.raises(TypeError, match=msg):
+                op(dz, np.array(tolist(dr), dtype=object))
+
+        # The aware==aware and naive==naive comparisons should *not* raise
+        assert np.all(dr == dr)
+        assert np.all(dr == tolist(dr))
+        assert np.all(tolist(dr) == dr)
+        assert np.all(np.array(tolist(dr), dtype=object) == dr)
+        assert np.all(dr == np.array(tolist(dr), dtype=object))
+
+        assert np.all(dz == dz)
+        assert np.all(dz == tolist(dz))
+        assert np.all(tolist(dz) == dz)
+        assert np.all(np.array(tolist(dz), dtype=object) == dz)
+        assert np.all(dz == np.array(tolist(dz), dtype=object))
+
+    def test_comparison_tzawareness_compat_scalars(self, comparison_op, box_with_array):
+        # GH#18162
+        op = comparison_op
+
+        dr = date_range("2016-01-01", periods=6)
+        dz = dr.tz_localize("US/Pacific")
+
+        dr = tm.box_expected(dr, box_with_array)
+        dz = tm.box_expected(dz, box_with_array)
+
+        # Check comparisons against scalar Timestamps
+        ts = Timestamp("2000-03-14 01:59")
+        ts_tz = Timestamp("2000-03-14 01:59", tz="Europe/Amsterdam")
+
+        assert np.all(dr > ts)
+        msg = r"Invalid comparison between dtype=datetime64\[ns.*\] and Timestamp"
+        if op not in [operator.eq, operator.ne]:
+            with pytest.raises(TypeError, match=msg):
+                op(dr, ts_tz)
+
+        assert np.all(dz > ts_tz)
+        if op not in [operator.eq, operator.ne]:
+            with pytest.raises(TypeError, match=msg):
+                op(dz, ts)
+
+        if op not in [operator.eq, operator.ne]:
+            # GH#12601: Check comparison against Timestamps and DatetimeIndex
+            with pytest.raises(TypeError, match=msg):
+                op(ts, dz)
+
+    @pytest.mark.parametrize(
+        "other",
+        [datetime(2016, 1, 1), Timestamp("2016-01-01"), np.datetime64("2016-01-01")],
+    )
+    # Bug in NumPy? https://github.com/numpy/numpy/issues/13841
+    # Raising in __eq__ will fallback to NumPy, which warns, fails,
+    # then re-raises the original exception. So we just need to ignore.
+    @pytest.mark.filterwarnings("ignore:elementwise comp:DeprecationWarning")
+    def test_scalar_comparison_tzawareness(
+        self, comparison_op, other, tz_aware_fixture, box_with_array
+    ):
+        op = comparison_op
+        tz = tz_aware_fixture
+        dti = date_range("2016-01-01", periods=2, tz=tz)
+
+        dtarr = tm.box_expected(dti, box_with_array)
+        xbox = get_upcast_box(dtarr, other, True)
+        if op in [operator.eq, operator.ne]:
+            exbool = op is operator.ne
+            expected = np.array([exbool, exbool], dtype=bool)
+            expected = tm.box_expected(expected, xbox)
+
+            result = op(dtarr, other)
+            tm.assert_equal(result, expected)
+
+            result = op(other, dtarr)
+            tm.assert_equal(result, expected)
+        else:
+            msg = (
+                r"Invalid comparison between dtype=datetime64\[ns, .*\] "
+                f"and {type(other).__name__}"
+            )
+            with pytest.raises(TypeError, match=msg):
+                op(dtarr, other)
+            with pytest.raises(TypeError, match=msg):
+                op(other, dtarr)
+
+    def test_nat_comparison_tzawareness(self, comparison_op):
+        # GH#19276
+        # tzaware DatetimeIndex should not raise when compared to NaT
+        op = comparison_op
+
+        dti = DatetimeIndex(
+            ["2014-01-01", NaT, "2014-03-01", NaT, "2014-05-01", "2014-07-01"]
+        )
+        expected = np.array([op == operator.ne] * len(dti))
+        result = op(dti, NaT)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = op(dti.tz_localize("US/Pacific"), NaT)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_dti_cmp_str(self, tz_naive_fixture):
+        # GH#22074
+        # regardless of tz, we expect these comparisons are valid
+        tz = tz_naive_fixture
+        rng = date_range("1/1/2000", periods=10, tz=tz)
+        other = "1/1/2000"
+
+        result = rng == other
+        expected = np.array([True] + [False] * 9)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng != other
+        expected = np.array([False] + [True] * 9)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng < other
+        expected = np.array([False] * 10)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng <= other
+        expected = np.array([True] + [False] * 9)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng > other
+        expected = np.array([False] + [True] * 9)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng >= other
+        expected = np.array([True] * 10)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_dti_cmp_list(self):
+        rng = date_range("1/1/2000", periods=10)
+
+        result = rng == list(rng)
+        expected = rng == rng
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            pd.timedelta_range("1D", periods=10),
+            pd.timedelta_range("1D", periods=10).to_series(),
+            pd.timedelta_range("1D", periods=10).asi8.view("m8[ns]"),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_dti_cmp_tdi_tzawareness(self, other):
+        # GH#22074
+        # reversion test that we _don't_ call _assert_tzawareness_compat
+        # when comparing against TimedeltaIndex
+        dti = date_range("2000-01-01", periods=10, tz="Asia/Tokyo")
+
+        result = dti == other
+        expected = np.array([False] * 10)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti != other
+        expected = np.array([True] * 10)
+        tm.assert_numpy_array_equal(result, expected)
+        msg = "Invalid comparison between"
+        with pytest.raises(TypeError, match=msg):
+            dti < other
+        with pytest.raises(TypeError, match=msg):
+            dti <= other
+        with pytest.raises(TypeError, match=msg):
+            dti > other
+        with pytest.raises(TypeError, match=msg):
+            dti >= other
+
+    def test_dti_cmp_object_dtype(self):
+        # GH#22074
+        dti = date_range("2000-01-01", periods=10, tz="Asia/Tokyo")
+
+        other = dti.astype("O")
+
+        result = dti == other
+        expected = np.array([True] * 10)
+        tm.assert_numpy_array_equal(result, expected)
+
+        other = dti.tz_localize(None)
+        result = dti != other
+        tm.assert_numpy_array_equal(result, expected)
+
+        other = np.array(list(dti[:5]) + [Timedelta(days=1)] * 5)
+        result = dti == other
+        expected = np.array([True] * 5 + [False] * 5)
+        tm.assert_numpy_array_equal(result, expected)
+        msg = ">=' not supported between instances of 'Timestamp' and 'Timedelta'"
+        with pytest.raises(TypeError, match=msg):
+            dti >= other
+
+
+# ------------------------------------------------------------------
+# Arithmetic
+
+
+class TestDatetime64Arithmetic:
+    # This class is intended for "finished" tests that are fully parametrized
+    #  over DataFrame/Series/Index/DatetimeArray
+
+    # -------------------------------------------------------------
+    # Addition/Subtraction of timedelta-like
+
+    @pytest.mark.arm_slow
+    def test_dt64arr_add_timedeltalike_scalar(
+        self, tz_naive_fixture, two_hours, box_with_array
+    ):
+        # GH#22005, GH#22163 check DataFrame doesn't raise TypeError
+        tz = tz_naive_fixture
+
+        rng = date_range("2000-01-01", "2000-02-01", tz=tz)
+        expected = date_range("2000-01-01 02:00", "2000-02-01 02:00", tz=tz)
+
+        rng = tm.box_expected(rng, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = rng + two_hours
+        tm.assert_equal(result, expected)
+
+        result = two_hours + rng
+        tm.assert_equal(result, expected)
+
+        rng += two_hours
+        tm.assert_equal(rng, expected)
+
+    def test_dt64arr_sub_timedeltalike_scalar(
+        self, tz_naive_fixture, two_hours, box_with_array
+    ):
+        tz = tz_naive_fixture
+
+        rng = date_range("2000-01-01", "2000-02-01", tz=tz)
+        expected = date_range("1999-12-31 22:00", "2000-01-31 22:00", tz=tz)
+
+        rng = tm.box_expected(rng, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = rng - two_hours
+        tm.assert_equal(result, expected)
+
+        rng -= two_hours
+        tm.assert_equal(rng, expected)
+
+    def test_dt64_array_sub_dt_with_different_timezone(self, box_with_array):
+        t1 = date_range("20130101", periods=3).tz_localize("US/Eastern")
+        t1 = tm.box_expected(t1, box_with_array)
+        t2 = Timestamp("20130101").tz_localize("CET")
+        tnaive = Timestamp(20130101)
+
+        result = t1 - t2
+        expected = TimedeltaIndex(
+            ["0 days 06:00:00", "1 days 06:00:00", "2 days 06:00:00"]
+        )
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = t2 - t1
+        expected = TimedeltaIndex(
+            ["-1 days +18:00:00", "-2 days +18:00:00", "-3 days +18:00:00"]
+        )
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        msg = "Cannot subtract tz-naive and tz-aware datetime-like objects"
+        with pytest.raises(TypeError, match=msg):
+            t1 - tnaive
+
+        with pytest.raises(TypeError, match=msg):
+            tnaive - t1
+
+    def test_dt64_array_sub_dt64_array_with_different_timezone(self, box_with_array):
+        t1 = date_range("20130101", periods=3).tz_localize("US/Eastern")
+        t1 = tm.box_expected(t1, box_with_array)
+        t2 = date_range("20130101", periods=3).tz_localize("CET")
+        t2 = tm.box_expected(t2, box_with_array)
+        tnaive = date_range("20130101", periods=3)
+
+        result = t1 - t2
+        expected = TimedeltaIndex(
+            ["0 days 06:00:00", "0 days 06:00:00", "0 days 06:00:00"]
+        )
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = t2 - t1
+        expected = TimedeltaIndex(
+            ["-1 days +18:00:00", "-1 days +18:00:00", "-1 days +18:00:00"]
+        )
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        msg = "Cannot subtract tz-naive and tz-aware datetime-like objects"
+        with pytest.raises(TypeError, match=msg):
+            t1 - tnaive
+
+        with pytest.raises(TypeError, match=msg):
+            tnaive - t1
+
+    def test_dt64arr_add_sub_td64_nat(self, box_with_array, tz_naive_fixture):
+        # GH#23320 special handling for timedelta64("NaT")
+        tz = tz_naive_fixture
+
+        dti = date_range("1994-04-01", periods=9, tz=tz, freq="QS")
+        other = np.timedelta64("NaT")
+        expected = DatetimeIndex(["NaT"] * 9, tz=tz).as_unit("ns")
+
+        obj = tm.box_expected(dti, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = obj + other
+        tm.assert_equal(result, expected)
+        result = other + obj
+        tm.assert_equal(result, expected)
+        result = obj - other
+        tm.assert_equal(result, expected)
+        msg = "cannot subtract"
+        with pytest.raises(TypeError, match=msg):
+            other - obj
+
+    def test_dt64arr_add_sub_td64ndarray(self, tz_naive_fixture, box_with_array):
+        tz = tz_naive_fixture
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        tdi = TimedeltaIndex(["-1 Day", "-1 Day", "-1 Day"])
+        tdarr = tdi.values
+
+        expected = date_range("2015-12-31", "2016-01-02", periods=3, tz=tz)
+
+        dtarr = tm.box_expected(dti, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = dtarr + tdarr
+        tm.assert_equal(result, expected)
+        result = tdarr + dtarr
+        tm.assert_equal(result, expected)
+
+        expected = date_range("2016-01-02", "2016-01-04", periods=3, tz=tz)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = dtarr - tdarr
+        tm.assert_equal(result, expected)
+        msg = "cannot subtract|(bad|unsupported) operand type for unary"
+        with pytest.raises(TypeError, match=msg):
+            tdarr - dtarr
+
+    # -----------------------------------------------------------------
+    # Subtraction of datetime-like scalars
+
+    @pytest.mark.parametrize(
+        "ts",
+        [
+            Timestamp("2013-01-01"),
+            Timestamp("2013-01-01").to_pydatetime(),
+            Timestamp("2013-01-01").to_datetime64(),
+            # GH#7996, GH#22163 ensure non-nano datetime64 is converted to nano
+            #  for DataFrame operation
+            np.datetime64("2013-01-01", "D"),
+        ],
+    )
+    def test_dt64arr_sub_dtscalar(self, box_with_array, ts):
+        # GH#8554, GH#22163 DataFrame op should _not_ return dt64 dtype
+        idx = date_range("2013-01-01", periods=3)._with_freq(None)
+        idx = tm.box_expected(idx, box_with_array)
+
+        expected = TimedeltaIndex(["0 Days", "1 Day", "2 Days"])
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = idx - ts
+        tm.assert_equal(result, expected)
+
+        result = ts - idx
+        tm.assert_equal(result, -expected)
+        tm.assert_equal(result, -expected)
+
+    def test_dt64arr_sub_timestamp_tzaware(self, box_with_array):
+        ser = date_range("2014-03-17", periods=2, freq="D", tz="US/Eastern")
+        ser = ser._with_freq(None)
+        ts = ser[0]
+
+        ser = tm.box_expected(ser, box_with_array)
+
+        delta_series = Series([np.timedelta64(0, "D"), np.timedelta64(1, "D")])
+        expected = tm.box_expected(delta_series, box_with_array)
+
+        tm.assert_equal(ser - ts, expected)
+        tm.assert_equal(ts - ser, -expected)
+
+    def test_dt64arr_sub_NaT(self, box_with_array, unit):
+        # GH#18808
+        dti = DatetimeIndex([NaT, Timestamp("19900315")]).as_unit(unit)
+        ser = tm.box_expected(dti, box_with_array)
+
+        result = ser - NaT
+        expected = Series([NaT, NaT], dtype=f"timedelta64[{unit}]")
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        dti_tz = dti.tz_localize("Asia/Tokyo")
+        ser_tz = tm.box_expected(dti_tz, box_with_array)
+
+        result = ser_tz - NaT
+        expected = Series([NaT, NaT], dtype=f"timedelta64[{unit}]")
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+    # -------------------------------------------------------------
+    # Subtraction of datetime-like array-like
+
+    def test_dt64arr_sub_dt64object_array(
+        self, performance_warning, box_with_array, tz_naive_fixture
+    ):
+        dti = date_range("2016-01-01", periods=3, tz=tz_naive_fixture)
+        expected = dti - dti
+
+        obj = tm.box_expected(dti, box_with_array)
+        expected = tm.box_expected(expected, box_with_array).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            result = obj - obj.astype(object)
+        tm.assert_equal(result, expected)
+
+    def test_dt64arr_naive_sub_dt64ndarray(self, box_with_array):
+        dti = date_range("2016-01-01", periods=3, tz=None)
+        dt64vals = dti.values
+
+        dtarr = tm.box_expected(dti, box_with_array)
+
+        expected = dtarr - dtarr
+        result = dtarr - dt64vals
+        tm.assert_equal(result, expected)
+        result = dt64vals - dtarr
+        tm.assert_equal(result, expected)
+
+    def test_dt64arr_aware_sub_dt64ndarray_raises(
+        self, tz_aware_fixture, box_with_array
+    ):
+        tz = tz_aware_fixture
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        dt64vals = dti.values
+
+        dtarr = tm.box_expected(dti, box_with_array)
+        msg = "Cannot subtract tz-naive and tz-aware datetime"
+        with pytest.raises(TypeError, match=msg):
+            dtarr - dt64vals
+        with pytest.raises(TypeError, match=msg):
+            dt64vals - dtarr
+
+    # -------------------------------------------------------------
+    # Addition of datetime-like others (invalid)
+
+    def test_dt64arr_add_dtlike_raises(self, tz_naive_fixture, box_with_array):
+        # GH#22163 ensure DataFrame doesn't cast Timestamp to i8
+        # GH#9631
+        tz = tz_naive_fixture
+
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        if tz is None:
+            dti2 = dti.tz_localize("US/Eastern")
+        else:
+            dti2 = dti.tz_localize(None)
+        dtarr = tm.box_expected(dti, box_with_array)
+
+        assert_cannot_add(dtarr, dti.values)
+        assert_cannot_add(dtarr, dti)
+        assert_cannot_add(dtarr, dtarr)
+        assert_cannot_add(dtarr, dti[0])
+        assert_cannot_add(dtarr, dti[0].to_pydatetime())
+        assert_cannot_add(dtarr, dti[0].to_datetime64())
+        assert_cannot_add(dtarr, dti2[0])
+        assert_cannot_add(dtarr, dti2[0].to_pydatetime())
+        assert_cannot_add(dtarr, np.datetime64("2011-01-01", "D"))
+
+    # -------------------------------------------------------------
+    # Other Invalid Addition/Subtraction
+
+    # Note: freq here includes both Tick and non-Tick offsets; this is
+    #  relevant because historically integer-addition was allowed if we had
+    #  a freq.
+    @pytest.mark.parametrize("freq", ["h", "D", "W", "2ME", "MS", "QE", "B", None])
+    @pytest.mark.parametrize("dtype", [None, "uint8"])
+    def test_dt64arr_addsub_intlike(
+        self, dtype, index_or_series_or_array, freq, tz_naive_fixture
+    ):
+        # GH#19959, GH#19123, GH#19012
+        # GH#55860 use index_or_series_or_array instead of box_with_array
+        #  bc DataFrame alignment makes it inapplicable
+        tz = tz_naive_fixture
+
+        if freq is None:
+            dti = DatetimeIndex(["NaT", "2017-04-05 06:07:08"], tz=tz)
+        else:
+            dti = date_range("2016-01-01", periods=2, freq=freq, tz=tz)
+
+        obj = index_or_series_or_array(dti)
+        other = np.array([4, -1])
+        if dtype is not None:
+            other = other.astype(dtype)
+
+        msg = "|".join(
+            [
+                "Addition/subtraction of integers",
+                "cannot subtract DatetimeArray from",
+                # IntegerArray
+                "can only perform ops with numeric values",
+                "unsupported operand type.*Categorical",
+                r"unsupported operand type\(s\) for -: 'int' and 'Timestamp'",
+            ]
+        )
+        assert_invalid_addsub_type(obj, 1, msg)
+        assert_invalid_addsub_type(obj, np.int64(2), msg)
+        assert_invalid_addsub_type(obj, np.array(3, dtype=np.int64), msg)
+        assert_invalid_addsub_type(obj, other, msg)
+        assert_invalid_addsub_type(obj, np.array(other), msg)
+        assert_invalid_addsub_type(obj, pd.array(other), msg)
+        assert_invalid_addsub_type(obj, pd.Categorical(other), msg)
+        assert_invalid_addsub_type(obj, pd.Index(other), msg)
+        assert_invalid_addsub_type(obj, Series(other), msg)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            3.14,
+            np.array([2.0, 3.0]),
+            # GH#13078 datetime +/- Period is invalid
+            Period("2011-01-01", freq="D"),
+            # https://github.com/pandas-dev/pandas/issues/10329
+            time(1, 2, 3),
+        ],
+    )
+    @pytest.mark.parametrize("dti_freq", [None, "D"])
+    def test_dt64arr_add_sub_invalid(self, dti_freq, other, box_with_array):
+        dti = DatetimeIndex(["2011-01-01", "2011-01-02"], freq=dti_freq)
+        dtarr = tm.box_expected(dti, box_with_array)
+        msg = "|".join(
+            [
+                "unsupported operand type",
+                "cannot (add|subtract)",
+                "cannot use operands with types",
+                "ufunc '?(add|subtract)'? cannot use operands with types",
+                "Concatenation operation is not implemented for NumPy arrays",
+            ]
+        )
+        assert_invalid_addsub_type(dtarr, other, msg)
+
+    @pytest.mark.parametrize("pi_freq", ["D", "W", "Q", "h"])
+    @pytest.mark.parametrize("dti_freq", [None, "D"])
+    def test_dt64arr_add_sub_parr(
+        self, dti_freq, pi_freq, box_with_array, box_with_array2
+    ):
+        # GH#20049 subtracting PeriodIndex should raise TypeError
+        dti = DatetimeIndex(["2011-01-01", "2011-01-02"], freq=dti_freq)
+        pi = dti.to_period(pi_freq)
+
+        dtarr = tm.box_expected(dti, box_with_array)
+        parr = tm.box_expected(pi, box_with_array2)
+        msg = "|".join(
+            [
+                "cannot (add|subtract)",
+                "unsupported operand",
+                "descriptor.*requires",
+                "ufunc.*cannot use operands",
+            ]
+        )
+        assert_invalid_addsub_type(dtarr, parr, msg)
+
+    @pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+    def test_dt64arr_addsub_time_objects_raises(self, box_with_array, tz_naive_fixture):
+        # https://github.com/pandas-dev/pandas/issues/10329
+
+        tz = tz_naive_fixture
+
+        obj1 = date_range("2012-01-01", periods=3, tz=tz)
+        obj2 = [time(i, i, i) for i in range(3)]
+
+        obj1 = tm.box_expected(obj1, box_with_array)
+        obj2 = tm.box_expected(obj2, box_with_array)
+
+        msg = "|".join(
+            [
+                "unsupported operand",
+                "cannot subtract DatetimeArray from ndarray",
+            ]
+        )
+        # pandas.errors.PerformanceWarning: Non-vectorized DateOffset being
+        # applied to Series or DatetimeIndex
+        # we aren't testing that here, so ignore.
+        assert_invalid_addsub_type(obj1, obj2, msg=msg)
+
+    # -------------------------------------------------------------
+    # Other invalid operations
+
+    @pytest.mark.parametrize(
+        "dt64_series",
+        [
+            Series([Timestamp("19900315"), Timestamp("19900315")]),
+            Series([NaT, Timestamp("19900315")]),
+            Series([NaT, NaT], dtype="datetime64[ns]"),
+        ],
+    )
+    @pytest.mark.parametrize("one", [1, 1.0, np.array(1)])
+    def test_dt64_mul_div_numeric_invalid(self, one, dt64_series, box_with_array):
+        obj = tm.box_expected(dt64_series, box_with_array)
+
+        msg = "cannot perform .* with this index type"
+
+        # multiplication
+        with pytest.raises(TypeError, match=msg):
+            obj * one
+        with pytest.raises(TypeError, match=msg):
+            one * obj
+
+        # division
+        with pytest.raises(TypeError, match=msg):
+            obj / one
+        with pytest.raises(TypeError, match=msg):
+            one / obj
+
+
+class TestDatetime64DateOffsetArithmetic:
+    # -------------------------------------------------------------
+    # Tick DateOffsets
+
+    # TODO: parametrize over timezone?
+    def test_dt64arr_series_add_tick_DateOffset(self, box_with_array, unit):
+        # GH#4532
+        # operate with pd.offsets
+        ser = Series(
+            [Timestamp("20130101 9:01"), Timestamp("20130101 9:02")]
+        ).dt.as_unit(unit)
+        expected = Series(
+            [Timestamp("20130101 9:01:05"), Timestamp("20130101 9:02:05")]
+        ).dt.as_unit(unit)
+
+        ser = tm.box_expected(ser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = ser + pd.offsets.Second(5)
+        tm.assert_equal(result, expected)
+
+        result2 = pd.offsets.Second(5) + ser
+        tm.assert_equal(result2, expected)
+
+    def test_dt64arr_series_sub_tick_DateOffset(self, box_with_array):
+        # GH#4532
+        # operate with pd.offsets
+        ser = Series([Timestamp("20130101 9:01"), Timestamp("20130101 9:02")])
+        expected = Series(
+            [Timestamp("20130101 9:00:55"), Timestamp("20130101 9:01:55")]
+        )
+
+        ser = tm.box_expected(ser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = ser - pd.offsets.Second(5)
+        tm.assert_equal(result, expected)
+
+        result2 = -pd.offsets.Second(5) + ser
+        tm.assert_equal(result2, expected)
+        msg = "(bad|unsupported) operand type for unary"
+        with pytest.raises(TypeError, match=msg):
+            pd.offsets.Second(5) - ser
+
+    @pytest.mark.parametrize(
+        "cls_name", ["Day", "Hour", "Minute", "Second", "Milli", "Micro", "Nano"]
+    )
+    def test_dt64arr_add_sub_tick_DateOffset_smoke(self, cls_name, box_with_array):
+        # GH#4532
+        # smoke tests for valid DateOffsets
+        ser = Series([Timestamp("20130101 9:01"), Timestamp("20130101 9:02")])
+        ser = tm.box_expected(ser, box_with_array)
+
+        offset_cls = getattr(pd.offsets, cls_name)
+        ser + offset_cls(5)
+        offset_cls(5) + ser
+        ser - offset_cls(5)
+
+    def test_dti_add_tick_tzaware(self, tz_aware_fixture, box_with_array):
+        # GH#21610, GH#22163 ensure DataFrame doesn't return object-dtype
+        tz = tz_aware_fixture
+        if tz == "US/Pacific":
+            dates = date_range("2012-11-01", periods=3, tz=tz)
+            offset = dates + pd.offsets.Hour(5)
+            assert dates[0] + pd.offsets.Hour(5) == offset[0]
+
+        dates = date_range("2010-11-01 00:00", periods=3, tz=tz, freq="h")
+        expected = DatetimeIndex(
+            ["2010-11-01 05:00", "2010-11-01 06:00", "2010-11-01 07:00"],
+            freq="h",
+            tz=tz,
+        ).as_unit("ns")
+
+        dates = tm.box_expected(dates, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        for scalar in [pd.offsets.Hour(5), np.timedelta64(5, "h"), timedelta(hours=5)]:
+            offset = dates + scalar
+            tm.assert_equal(offset, expected)
+            offset = scalar + dates
+            tm.assert_equal(offset, expected)
+
+            roundtrip = offset - scalar
+            tm.assert_equal(roundtrip, dates)
+
+            msg = "|".join(
+                ["bad operand type for unary -", "cannot subtract DatetimeArray"]
+            )
+            with pytest.raises(TypeError, match=msg):
+                scalar - dates
+
+    # -------------------------------------------------------------
+    # RelativeDelta DateOffsets
+
+    def test_dt64arr_add_sub_relativedelta_offsets(self, box_with_array, unit):
+        # GH#10699
+        vec = DatetimeIndex(
+            [
+                Timestamp("2000-01-05 00:15:00"),
+                Timestamp("2000-01-31 00:23:00"),
+                Timestamp("2000-01-01"),
+                Timestamp("2000-03-31"),
+                Timestamp("2000-02-29"),
+                Timestamp("2000-12-31"),
+                Timestamp("2000-05-15"),
+                Timestamp("2001-06-15"),
+            ]
+        ).as_unit(unit)
+        vec = tm.box_expected(vec, box_with_array)
+        vec_items = vec.iloc[0] if box_with_array is pd.DataFrame else vec
+
+        # DateOffset relativedelta fastpath
+        relative_kwargs = [
+            ("years", 2),
+            ("months", 5),
+            ("days", 3),
+            ("hours", 5),
+            ("minutes", 10),
+            ("seconds", 2),
+            ("microseconds", 5),
+        ]
+        for i, (offset_unit, value) in enumerate(relative_kwargs):
+            off = DateOffset(**{offset_unit: value})
+
+            exp_unit = unit
+            if offset_unit == "microseconds" and unit != "ns":
+                exp_unit = "us"
+
+            # TODO(GH#55564): as_unit will be unnecessary
+            expected = DatetimeIndex([x + off for x in vec_items]).as_unit(exp_unit)
+            expected = tm.box_expected(expected, box_with_array)
+            tm.assert_equal(expected, vec + off)
+
+            expected = DatetimeIndex([x - off for x in vec_items]).as_unit(exp_unit)
+            expected = tm.box_expected(expected, box_with_array)
+            tm.assert_equal(expected, vec - off)
+
+            off = DateOffset(**dict(relative_kwargs[: i + 1]))
+
+            expected = DatetimeIndex([x + off for x in vec_items]).as_unit(exp_unit)
+            expected = tm.box_expected(expected, box_with_array)
+            tm.assert_equal(expected, vec + off)
+
+            expected = DatetimeIndex([x - off for x in vec_items]).as_unit(exp_unit)
+            expected = tm.box_expected(expected, box_with_array)
+            tm.assert_equal(expected, vec - off)
+            msg = "(bad|unsupported) operand type for unary"
+            with pytest.raises(TypeError, match=msg):
+                off - vec
+
+    # -------------------------------------------------------------
+    # Non-Tick, Non-RelativeDelta DateOffsets
+
+    # TODO: redundant with test_dt64arr_add_sub_DateOffset?  that includes
+    #  tz-aware cases which this does not
+    @pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+    @pytest.mark.parametrize(
+        "cls_and_kwargs",
+        [
+            "YearBegin",
+            ("YearBegin", {"month": 5}),
+            "YearEnd",
+            ("YearEnd", {"month": 5}),
+            "MonthBegin",
+            "MonthEnd",
+            "SemiMonthEnd",
+            "SemiMonthBegin",
+            "Week",
+            ("Week", {"weekday": 3}),
+            ("Week", {"weekday": 6}),
+            "BusinessDay",
+            "BDay",
+            "QuarterEnd",
+            "QuarterBegin",
+            "CustomBusinessDay",
+            "CDay",
+            "CBMonthEnd",
+            "CBMonthBegin",
+            "BMonthBegin",
+            "BMonthEnd",
+            "BusinessHour",
+            "BYearBegin",
+            "BYearEnd",
+            "BQuarterBegin",
+            ("LastWeekOfMonth", {"weekday": 2}),
+            (
+                "FY5253Quarter",
+                {
+                    "qtr_with_extra_week": 1,
+                    "startingMonth": 1,
+                    "weekday": 2,
+                    "variation": "nearest",
+                },
+            ),
+            ("FY5253", {"weekday": 0, "startingMonth": 2, "variation": "nearest"}),
+            ("WeekOfMonth", {"weekday": 2, "week": 2}),
+            "Easter",
+            ("DateOffset", {"day": 4}),
+            ("DateOffset", {"month": 5}),
+        ],
+    )
+    @pytest.mark.parametrize("normalize", [True, False])
+    @pytest.mark.parametrize("n", [0, 5])
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_dt64arr_add_sub_DateOffsets(
+        self, box_with_array, n, normalize, cls_and_kwargs, unit, tz
+    ):
+        # GH#10699
+        # assert vectorized operation matches pointwise operations
+
+        if isinstance(cls_and_kwargs, tuple):
+            # If cls_name param is a tuple, then 2nd entry is kwargs for
+            # the offset constructor
+            cls_name, kwargs = cls_and_kwargs
+        else:
+            cls_name = cls_and_kwargs
+            kwargs = {}
+
+        if n == 0 and cls_name in [
+            "WeekOfMonth",
+            "LastWeekOfMonth",
+            "FY5253Quarter",
+            "FY5253",
+        ]:
+            # passing n = 0 is invalid for these offset classes
+            return
+
+        vec = (
+            DatetimeIndex(
+                [
+                    Timestamp("2000-01-05 00:15:00"),
+                    Timestamp("2000-01-31 00:23:00"),
+                    Timestamp("2000-01-01"),
+                    Timestamp("2000-03-31"),
+                    Timestamp("2000-02-29"),
+                    Timestamp("2000-12-31"),
+                    Timestamp("2000-05-15"),
+                    Timestamp("2001-06-15"),
+                ]
+            )
+            .as_unit(unit)
+            .tz_localize(tz)
+        )
+        vec = tm.box_expected(vec, box_with_array)
+        vec_items = vec.iloc[0] if box_with_array is pd.DataFrame else vec
+
+        offset_cls = getattr(pd.offsets, cls_name)
+        offset = offset_cls(n, normalize=normalize, **kwargs)
+
+        # TODO(GH#55564): as_unit will be unnecessary
+        expected = DatetimeIndex([x + offset for x in vec_items]).as_unit(unit)
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(expected, vec + offset)
+        tm.assert_equal(expected, offset + vec)
+
+        expected = DatetimeIndex([x - offset for x in vec_items]).as_unit(unit)
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(expected, vec - offset)
+
+        expected = DatetimeIndex([offset + x for x in vec_items]).as_unit(unit)
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(expected, offset + vec)
+        msg = "(bad|unsupported) operand type for unary"
+        with pytest.raises(TypeError, match=msg):
+            offset - vec
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            [pd.offsets.MonthEnd(), pd.offsets.Day(n=2)],
+            [pd.offsets.DateOffset(years=1), pd.offsets.MonthEnd()],
+            # matching offsets
+            [pd.offsets.DateOffset(years=1), pd.offsets.DateOffset(years=1)],
+        ],
+    )
+    @pytest.mark.parametrize("op", [operator.add, roperator.radd, operator.sub])
+    def test_dt64arr_add_sub_offset_array(
+        self, performance_warning, tz_naive_fixture, box_with_array, op, other
+    ):
+        # GH#18849
+        # GH#10699 array of offsets
+
+        tz = tz_naive_fixture
+        dti = date_range("2017-01-01", periods=2, tz=tz)
+        dtarr = tm.box_expected(dti, box_with_array)
+        other = np.array(other)
+        expected = DatetimeIndex([op(dti[n], other[n]) for n in range(len(dti))])
+        expected = tm.box_expected(expected, box_with_array).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = op(dtarr, other)
+        tm.assert_equal(res, expected)
+
+        # Same thing but boxing other
+        other = tm.box_expected(other, box_with_array)
+        if box_with_array is pd.array and op is roperator.radd:
+            # We expect a NumpyExtensionArray, not ndarray[object] here
+            expected = pd.array(expected, dtype=object)
+        with tm.assert_produces_warning(performance_warning):
+            res = op(dtarr, other)
+        tm.assert_equal(res, expected)
+
+    @pytest.mark.parametrize(
+        "op, offset, exp, exp_freq",
+        [
+            (
+                "__add__",
+                DateOffset(months=3, days=10),
+                [
+                    Timestamp("2014-04-11"),
+                    Timestamp("2015-04-11"),
+                    Timestamp("2016-04-11"),
+                    Timestamp("2017-04-11"),
+                ],
+                None,
+            ),
+            (
+                "__add__",
+                DateOffset(months=3),
+                [
+                    Timestamp("2014-04-01"),
+                    Timestamp("2015-04-01"),
+                    Timestamp("2016-04-01"),
+                    Timestamp("2017-04-01"),
+                ],
+                "YS-APR",
+            ),
+            (
+                "__sub__",
+                DateOffset(months=3, days=10),
+                [
+                    Timestamp("2013-09-21"),
+                    Timestamp("2014-09-21"),
+                    Timestamp("2015-09-21"),
+                    Timestamp("2016-09-21"),
+                ],
+                None,
+            ),
+            (
+                "__sub__",
+                DateOffset(months=3),
+                [
+                    Timestamp("2013-10-01"),
+                    Timestamp("2014-10-01"),
+                    Timestamp("2015-10-01"),
+                    Timestamp("2016-10-01"),
+                ],
+                "YS-OCT",
+            ),
+        ],
+    )
+    def test_dti_add_sub_nonzero_mth_offset(
+        self, op, offset, exp, exp_freq, tz_aware_fixture, box_with_array
+    ):
+        # GH 26258
+        tz = tz_aware_fixture
+        date = date_range(start="01 Jan 2014", end="01 Jan 2017", freq="YS", tz=tz)
+        date = tm.box_expected(date, box_with_array, False)
+        mth = getattr(date, op)
+        result = mth(offset)
+
+        expected = DatetimeIndex(exp, tz=tz).as_unit("ns")
+        expected = tm.box_expected(expected, box_with_array, False)
+        tm.assert_equal(result, expected)
+
+    def test_dt64arr_series_add_DateOffset_with_milli(self):
+        # GH 57529
+        dti = DatetimeIndex(
+            [
+                "2000-01-01 00:00:00.012345678",
+                "2000-01-31 00:00:00.012345678",
+                "2000-02-29 00:00:00.012345678",
+            ],
+            dtype="datetime64[ns]",
+        )
+        result = dti + DateOffset(milliseconds=4)
+        expected = DatetimeIndex(
+            [
+                "2000-01-01 00:00:00.016345678",
+                "2000-01-31 00:00:00.016345678",
+                "2000-02-29 00:00:00.016345678",
+            ],
+            dtype="datetime64[ns]",
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = dti + DateOffset(days=1, milliseconds=4)
+        expected = DatetimeIndex(
+            [
+                "2000-01-02 00:00:00.016345678",
+                "2000-02-01 00:00:00.016345678",
+                "2000-03-01 00:00:00.016345678",
+            ],
+            dtype="datetime64[ns]",
+        )
+        tm.assert_index_equal(result, expected)
+
+
+class TestDatetime64OverflowHandling:
+    # TODO: box + de-duplicate
+
+    def test_dt64_overflow_masking(self, box_with_array):
+        # GH#25317
+        left = Series([Timestamp("1969-12-31")], dtype="M8[ns]")
+        right = Series([NaT])
+
+        left = tm.box_expected(left, box_with_array)
+        right = tm.box_expected(right, box_with_array)
+
+        expected = TimedeltaIndex([NaT], dtype="m8[ns]")
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = left - right
+        tm.assert_equal(result, expected)
+
+    def test_dt64_series_arith_overflow(self):
+        # GH#12534, fixed by GH#19024
+        dt = Timestamp("1700-01-31")
+        td = Timedelta("20000 Days")
+        dti = date_range("1949-09-30", freq="100YE", periods=4)
+        ser = Series(dti)
+        msg = "Overflow in int64 addition"
+        with pytest.raises(OverflowError, match=msg):
+            ser - dt
+        with pytest.raises(OverflowError, match=msg):
+            dt - ser
+        with pytest.raises(OverflowError, match=msg):
+            ser + td
+        with pytest.raises(OverflowError, match=msg):
+            td + ser
+
+        ser.iloc[-1] = NaT
+        expected = Series(
+            ["2004-10-03", "2104-10-04", "2204-10-04", "NaT"], dtype="datetime64[ns]"
+        )
+        res = ser + td
+        tm.assert_series_equal(res, expected)
+        res = td + ser
+        tm.assert_series_equal(res, expected)
+
+        ser.iloc[1:] = NaT
+        expected = Series(["91279 Days", "NaT", "NaT", "NaT"], dtype="timedelta64[ns]")
+        res = ser - dt
+        tm.assert_series_equal(res, expected)
+        res = dt - ser
+        tm.assert_series_equal(res, -expected)
+
+    def test_datetimeindex_sub_timestamp_overflow(self):
+        dtimax = pd.to_datetime(["2021-12-28 17:19", Timestamp.max]).as_unit("ns")
+        dtimin = pd.to_datetime(["2021-12-28 17:19", Timestamp.min]).as_unit("ns")
+
+        tsneg = Timestamp("1950-01-01").as_unit("ns")
+        ts_neg_variants = [
+            tsneg,
+            tsneg.to_pydatetime(),
+            tsneg.to_datetime64().astype("datetime64[ns]"),
+            tsneg.to_datetime64().astype("datetime64[D]"),
+        ]
+
+        tspos = Timestamp("1980-01-01").as_unit("ns")
+        ts_pos_variants = [
+            tspos,
+            tspos.to_pydatetime(),
+            tspos.to_datetime64().astype("datetime64[ns]"),
+            tspos.to_datetime64().astype("datetime64[D]"),
+        ]
+        msg = "Overflow in int64 addition"
+        for variant in ts_neg_variants:
+            with pytest.raises(OverflowError, match=msg):
+                dtimax - variant
+
+        expected = Timestamp.max._value - tspos._value
+        for variant in ts_pos_variants:
+            res = dtimax - variant
+            assert res[1]._value == expected
+
+        expected = Timestamp.min._value - tsneg._value
+        for variant in ts_neg_variants:
+            res = dtimin - variant
+            assert res[1]._value == expected
+
+        for variant in ts_pos_variants:
+            with pytest.raises(OverflowError, match=msg):
+                dtimin - variant
+
+    def test_datetimeindex_sub_datetimeindex_overflow(self):
+        # GH#22492, GH#22508
+        dtimax = pd.to_datetime(["2021-12-28 17:19", Timestamp.max]).as_unit("ns")
+        dtimin = pd.to_datetime(["2021-12-28 17:19", Timestamp.min]).as_unit("ns")
+
+        ts_neg = pd.to_datetime(["1950-01-01", "1950-01-01"]).as_unit("ns")
+        ts_pos = pd.to_datetime(["1980-01-01", "1980-01-01"]).as_unit("ns")
+
+        # General tests
+        expected = Timestamp.max._value - ts_pos[1]._value
+        result = dtimax - ts_pos
+        assert result[1]._value == expected
+
+        expected = Timestamp.min._value - ts_neg[1]._value
+        result = dtimin - ts_neg
+        assert result[1]._value == expected
+        msg = "Overflow in int64 addition"
+        with pytest.raises(OverflowError, match=msg):
+            dtimax - ts_neg
+
+        with pytest.raises(OverflowError, match=msg):
+            dtimin - ts_pos
+
+        # Edge cases
+        tmin = pd.to_datetime([Timestamp.min])
+        t1 = tmin + Timedelta.max + Timedelta("1us")
+        with pytest.raises(OverflowError, match=msg):
+            t1 - tmin
+
+        tmax = pd.to_datetime([Timestamp.max])
+        t2 = tmax + Timedelta.min - Timedelta("1us")
+        with pytest.raises(OverflowError, match=msg):
+            tmax - t2
+
+
+class TestTimestampSeriesArithmetic:
+    def test_empty_series_add_sub(self, box_with_array):
+        # GH#13844
+        a = Series(dtype="M8[ns]")
+        b = Series(dtype="m8[ns]")
+        a = box_with_array(a)
+        b = box_with_array(b)
+        tm.assert_equal(a, a + b)
+        tm.assert_equal(a, a - b)
+        tm.assert_equal(a, b + a)
+        msg = "cannot subtract"
+        with pytest.raises(TypeError, match=msg):
+            b - a
+
+    def test_operators_datetimelike(self):
+        # ## timedelta64 ###
+        td1 = Series([timedelta(minutes=5, seconds=3)] * 3)
+        td1.iloc[2] = np.nan
+
+        # ## datetime64 ###
+        dt1 = Series(
+            [
+                Timestamp("20111230"),
+                Timestamp("20120101"),
+                Timestamp("20120103"),
+            ]
+        )
+        dt1.iloc[2] = np.nan
+        dt2 = Series(
+            [
+                Timestamp("20111231"),
+                Timestamp("20120102"),
+                Timestamp("20120104"),
+            ]
+        )
+        dt1 - dt2
+        dt2 - dt1
+
+        # datetime64 with timetimedelta
+        dt1 + td1
+        td1 + dt1
+        dt1 - td1
+
+        # timetimedelta with datetime64
+        td1 + dt1
+        dt1 + td1
+
+    def test_dt64ser_sub_datetime_dtype(self, unit):
+        ts = Timestamp(datetime(1993, 1, 7, 13, 30, 00))
+        dt = datetime(1993, 6, 22, 13, 30)
+        ser = Series([ts], dtype=f"M8[{unit}]")
+        result = ser - dt
+
+        # the expected unit is the max of `unit` and the unit imputed to `dt`,
+        #  which is "us"
+        exp_unit = tm.get_finest_unit(unit, "us")
+        assert result.dtype == f"timedelta64[{exp_unit}]"
+
+    # -------------------------------------------------------------
+    # TODO: This next block of tests came from tests.series.test_operators,
+    # needs to be de-duplicated and parametrized over `box` classes
+
+    @pytest.mark.parametrize(
+        "left, right, op_fail",
+        [
+            [
+                [Timestamp("20111230"), Timestamp("20120101"), NaT],
+                [Timestamp("20111231"), Timestamp("20120102"), Timestamp("20120104")],
+                ["__sub__", "__rsub__"],
+            ],
+            [
+                [Timestamp("20111230"), Timestamp("20120101"), NaT],
+                [timedelta(minutes=5, seconds=3), timedelta(minutes=5, seconds=3), NaT],
+                ["__add__", "__radd__", "__sub__"],
+            ],
+            [
+                [
+                    Timestamp("20111230", tz="US/Eastern"),
+                    Timestamp("20111230", tz="US/Eastern"),
+                    NaT,
+                ],
+                [timedelta(minutes=5, seconds=3), NaT, timedelta(minutes=5, seconds=3)],
+                ["__add__", "__radd__", "__sub__"],
+            ],
+        ],
+    )
+    def test_operators_datetimelike_invalid(
+        self, left, right, op_fail, all_arithmetic_operators
+    ):
+        # these are all TypeError ops
+        op_str = all_arithmetic_operators
+        arg1 = Series(left)
+        arg2 = Series(right)
+        # check that we are getting a TypeError
+        # with 'operate' (from core/ops.py) for the ops that are not
+        # defined
+        op = getattr(arg1, op_str, None)
+        # Previously, _validate_for_numeric_binop in core/indexes/base.py
+        # did this for us.
+        if op_str not in op_fail:
+            with pytest.raises(
+                TypeError, match="operate|[cC]annot|unsupported operand"
+            ):
+                op(arg2)
+        else:
+            # Smoke test
+            op(arg2)
+
+    def test_sub_single_tz(self, unit):
+        # GH#12290
+        s1 = Series([Timestamp("2016-02-10", tz="America/Sao_Paulo")]).dt.as_unit(unit)
+        s2 = Series([Timestamp("2016-02-08", tz="America/Sao_Paulo")]).dt.as_unit(unit)
+        result = s1 - s2
+        expected = Series([Timedelta("2days")]).dt.as_unit(unit)
+        tm.assert_series_equal(result, expected)
+        result = s2 - s1
+        expected = Series([Timedelta("-2days")]).dt.as_unit(unit)
+        tm.assert_series_equal(result, expected)
+
+    def test_dt64tz_series_sub_dtitz(self):
+        # GH#19071 subtracting tzaware DatetimeIndex from tzaware Series
+        # (with same tz) raises, fixed by #19024
+        dti = date_range("1999-09-30", periods=10, tz="US/Pacific")
+        ser = Series(dti)
+        expected = Series(TimedeltaIndex(["0days"] * 10))
+
+        res = dti - ser
+        tm.assert_series_equal(res, expected)
+        res = ser - dti
+        tm.assert_series_equal(res, expected)
+
+    def test_sub_datetime_compat(self, unit):
+        # see GH#14088
+        ser = Series([datetime(2016, 8, 23, 12, tzinfo=timezone.utc), NaT]).dt.as_unit(
+            unit
+        )
+        dt = datetime(2016, 8, 22, 12, tzinfo=timezone.utc)
+        # The datetime object has "us" so we upcast lower units
+        exp_unit = tm.get_finest_unit(unit, "us")
+        exp = Series([Timedelta("1 days"), NaT]).dt.as_unit(exp_unit)
+        result = ser - dt
+        tm.assert_series_equal(result, exp)
+        result2 = ser - Timestamp(dt)
+        tm.assert_series_equal(result2, exp)
+
+    def test_dt64_series_add_mixed_tick_DateOffset(self):
+        # GH#4532
+        # operate with pd.offsets
+        s = Series([Timestamp("20130101 9:01"), Timestamp("20130101 9:02")])
+
+        result = s + pd.offsets.Milli(5)
+        result2 = pd.offsets.Milli(5) + s
+        expected = Series(
+            [Timestamp("20130101 9:01:00.005"), Timestamp("20130101 9:02:00.005")]
+        )
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
+
+        result = s + pd.offsets.Minute(5) + pd.offsets.Milli(5)
+        expected = Series(
+            [Timestamp("20130101 9:06:00.005"), Timestamp("20130101 9:07:00.005")]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_datetime64_ops_nat(self, unit):
+        # GH#11349
+        datetime_series = Series([NaT, Timestamp("19900315")]).dt.as_unit(unit)
+        nat_series_dtype_timestamp = Series([NaT, NaT], dtype=f"datetime64[{unit}]")
+        single_nat_dtype_datetime = Series([NaT], dtype=f"datetime64[{unit}]")
+
+        # subtraction
+        tm.assert_series_equal(-NaT + datetime_series, nat_series_dtype_timestamp)
+        msg = "bad operand type for unary -: 'DatetimeArray'"
+        with pytest.raises(TypeError, match=msg):
+            -single_nat_dtype_datetime + datetime_series
+
+        tm.assert_series_equal(
+            -NaT + nat_series_dtype_timestamp, nat_series_dtype_timestamp
+        )
+        with pytest.raises(TypeError, match=msg):
+            -single_nat_dtype_datetime + nat_series_dtype_timestamp
+
+        # addition
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp + NaT, nat_series_dtype_timestamp
+        )
+        tm.assert_series_equal(
+            NaT + nat_series_dtype_timestamp, nat_series_dtype_timestamp
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp + NaT, nat_series_dtype_timestamp
+        )
+        tm.assert_series_equal(
+            NaT + nat_series_dtype_timestamp, nat_series_dtype_timestamp
+        )
+
+    # -------------------------------------------------------------
+    # Timezone-Centric Tests
+
+    def test_operators_datetimelike_with_timezones(self):
+        tz = "US/Eastern"
+        dt1 = Series(date_range("2000-01-01 09:00:00", periods=5, tz=tz), name="foo")
+        dt2 = dt1.copy()
+        dt2.iloc[2] = np.nan
+
+        td1 = Series(pd.timedelta_range("1 days 1 min", periods=5, freq="h"))
+        td2 = td1.copy()
+        td2.iloc[1] = np.nan
+        assert td2._values.freq is None
+
+        result = dt1 + td1[0]
+        exp = (dt1.dt.tz_localize(None) + td1[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = dt2 + td2[0]
+        exp = (dt2.dt.tz_localize(None) + td2[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        # odd numpy behavior with scalar timedeltas
+        result = td1[0] + dt1
+        exp = (dt1.dt.tz_localize(None) + td1[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = td2[0] + dt2
+        exp = (dt2.dt.tz_localize(None) + td2[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = dt1 - td1[0]
+        exp = (dt1.dt.tz_localize(None) - td1[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+        msg = "(bad|unsupported) operand type for unary"
+        with pytest.raises(TypeError, match=msg):
+            td1[0] - dt1
+
+        result = dt2 - td2[0]
+        exp = (dt2.dt.tz_localize(None) - td2[0]).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+        with pytest.raises(TypeError, match=msg):
+            td2[0] - dt2
+
+        result = dt1 + td1
+        exp = (dt1.dt.tz_localize(None) + td1).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = dt2 + td2
+        exp = (dt2.dt.tz_localize(None) + td2).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = dt1 - td1
+        exp = (dt1.dt.tz_localize(None) - td1).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+
+        result = dt2 - td2
+        exp = (dt2.dt.tz_localize(None) - td2).dt.tz_localize(tz)
+        tm.assert_series_equal(result, exp)
+        msg = "cannot (add|subtract)"
+        with pytest.raises(TypeError, match=msg):
+            td1 - dt1
+        with pytest.raises(TypeError, match=msg):
+            td2 - dt2
+
+
+class TestDatetimeIndexArithmetic:
+    # -------------------------------------------------------------
+    # Binary operations DatetimeIndex and TimedeltaIndex/array
+
+    def test_dti_add_tdi(self, tz_naive_fixture):
+        # GH#17558
+        tz = tz_naive_fixture
+        dti = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10)
+        tdi = pd.timedelta_range("0 days", periods=10)
+        expected = date_range("2017-01-01", periods=10, tz=tz)
+        expected = expected._with_freq(None)
+
+        # add with TimedeltaIndex
+        result = dti + tdi
+        tm.assert_index_equal(result, expected)
+
+        result = tdi + dti
+        tm.assert_index_equal(result, expected)
+
+        # add with timedelta64 array
+        result = dti + tdi.values
+        tm.assert_index_equal(result, expected)
+
+        result = tdi.values + dti
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_iadd_tdi(self, tz_naive_fixture):
+        # GH#17558
+        tz = tz_naive_fixture
+        dti = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10)
+        tdi = pd.timedelta_range("0 days", periods=10)
+        expected = date_range("2017-01-01", periods=10, tz=tz)
+        expected = expected._with_freq(None)
+
+        # iadd with TimedeltaIndex
+        result = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10)
+        result += tdi
+        tm.assert_index_equal(result, expected)
+
+        result = pd.timedelta_range("0 days", periods=10)
+        result += dti
+        tm.assert_index_equal(result, expected)
+
+        # iadd with timedelta64 array
+        result = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10)
+        result += tdi.values
+        tm.assert_index_equal(result, expected)
+
+        result = pd.timedelta_range("0 days", periods=10)
+        result += dti
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_sub_tdi(self, tz_naive_fixture):
+        # GH#17558
+        tz = tz_naive_fixture
+        dti = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10)
+        tdi = pd.timedelta_range("0 days", periods=10)
+        expected = date_range("2017-01-01", periods=10, tz=tz, freq="-1D")
+        expected = expected._with_freq(None)
+
+        # sub with TimedeltaIndex
+        result = dti - tdi
+        tm.assert_index_equal(result, expected)
+
+        msg = "cannot subtract .*TimedeltaArray"
+        with pytest.raises(TypeError, match=msg):
+            tdi - dti
+
+        # sub with timedelta64 array
+        result = dti - tdi.values
+        tm.assert_index_equal(result, expected)
+
+        msg = "cannot subtract a datelike from a TimedeltaArray"
+        with pytest.raises(TypeError, match=msg):
+            tdi.values - dti
+
+    def test_dti_isub_tdi(self, tz_naive_fixture, unit):
+        # GH#17558
+        tz = tz_naive_fixture
+        dti = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10).as_unit(unit)
+        tdi = pd.timedelta_range("0 days", periods=10, unit=unit)
+        expected = date_range("2017-01-01", periods=10, tz=tz, freq="-1D", unit=unit)
+        expected = expected._with_freq(None)
+
+        # isub with TimedeltaIndex
+        result = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10).as_unit(unit)
+        result -= tdi
+        tm.assert_index_equal(result, expected)
+
+        # DTA.__isub__ GH#43904
+        dta = dti._data.copy()
+        dta -= tdi
+        tm.assert_datetime_array_equal(dta, expected._data)
+
+        out = dti._data.copy()
+        np.subtract(out, tdi, out=out)
+        tm.assert_datetime_array_equal(out, expected._data)
+
+        msg = "cannot subtract a datelike from a TimedeltaArray"
+        with pytest.raises(TypeError, match=msg):
+            tdi -= dti
+
+        # isub with timedelta64 array
+        result = DatetimeIndex([Timestamp("2017-01-01", tz=tz)] * 10).as_unit(unit)
+        result -= tdi.values
+        tm.assert_index_equal(result, expected)
+
+        with pytest.raises(TypeError, match=msg):
+            tdi.values -= dti
+
+        with pytest.raises(TypeError, match=msg):
+            tdi._values -= dti
+
+    # -------------------------------------------------------------
+    # Binary Operations DatetimeIndex and datetime-like
+    # TODO: A couple other tests belong in this section.  Move them in
+    # A PR where there isn't already a giant diff.
+
+    # -------------------------------------------------------------
+
+    def test_dta_add_sub_index(self, tz_naive_fixture):
+        # Check that DatetimeArray defers to Index classes
+        dti = date_range("20130101", periods=3, tz=tz_naive_fixture)
+        dta = dti.array
+        result = dta - dti
+        expected = dti - dti
+        tm.assert_index_equal(result, expected)
+
+        tdi = result
+        result = dta + tdi
+        expected = dti + tdi
+        tm.assert_index_equal(result, expected)
+
+        result = dta - tdi
+        expected = dti - tdi
+        tm.assert_index_equal(result, expected)
+
+    def test_sub_dti_dti(self, unit):
+        # previously performed setop (deprecated in 0.16.0), now changed to
+        # return subtraction -> TimeDeltaIndex (GH ...)
+
+        dti = date_range("20130101", periods=3, unit=unit)
+        dti_tz = date_range("20130101", periods=3, unit=unit).tz_localize("US/Eastern")
+        expected = TimedeltaIndex([0, 0, 0]).as_unit(unit)
+
+        result = dti - dti
+        tm.assert_index_equal(result, expected)
+
+        result = dti_tz - dti_tz
+        tm.assert_index_equal(result, expected)
+        msg = "Cannot subtract tz-naive and tz-aware datetime-like objects"
+        with pytest.raises(TypeError, match=msg):
+            dti_tz - dti
+
+        with pytest.raises(TypeError, match=msg):
+            dti - dti_tz
+
+        # isub
+        dti -= dti
+        tm.assert_index_equal(dti, expected)
+
+        # different length raises ValueError
+        dti1 = date_range("20130101", periods=3, unit=unit)
+        dti2 = date_range("20130101", periods=4, unit=unit)
+        msg = "cannot add indices of unequal length"
+        with pytest.raises(ValueError, match=msg):
+            dti1 - dti2
+
+        # NaN propagation
+        dti1 = DatetimeIndex(["2012-01-01", np.nan, "2012-01-03"]).as_unit(unit)
+        dti2 = DatetimeIndex(["2012-01-02", "2012-01-03", np.nan]).as_unit(unit)
+        expected = TimedeltaIndex(["1 days", np.nan, np.nan]).as_unit(unit)
+        result = dti2 - dti1
+        tm.assert_index_equal(result, expected)
+
+    # -------------------------------------------------------------------
+    # TODO: Most of this block is moved from series or frame tests, needs
+    # cleanup, box-parametrization, and de-duplication
+
+    @pytest.mark.parametrize("op", [operator.add, operator.sub])
+    def test_timedelta64_equal_timedelta_supported_ops(self, op, box_with_array):
+        ser = Series(
+            [
+                Timestamp("20130301"),
+                Timestamp("20130228 23:00:00"),
+                Timestamp("20130228 22:00:00"),
+                Timestamp("20130228 21:00:00"),
+            ]
+        )
+        obj = box_with_array(ser)
+
+        intervals = ["D", "h", "m", "s", "us"]
+
+        def timedelta64(*args):
+            # see casting notes in NumPy gh-12927
+            return np.sum(list(starmap(np.timedelta64, zip(args, intervals))))
+
+        for d, h, m, s, us in product(*([range(2)] * 5)):
+            nptd = timedelta64(d, h, m, s, us)
+            pytd = timedelta(days=d, hours=h, minutes=m, seconds=s, microseconds=us)
+            lhs = op(obj, nptd)
+            rhs = op(obj, pytd)
+
+            tm.assert_equal(lhs, rhs)
+
+    def test_ops_nat_mixed_datetime64_timedelta64(self):
+        # GH#11349
+        timedelta_series = Series([NaT, Timedelta("1s")])
+        datetime_series = Series([NaT, Timestamp("19900315")])
+        nat_series_dtype_timedelta = Series([NaT, NaT], dtype="timedelta64[ns]")
+        nat_series_dtype_timestamp = Series([NaT, NaT], dtype="datetime64[ns]")
+        single_nat_dtype_datetime = Series([NaT], dtype="datetime64[ns]")
+        single_nat_dtype_timedelta = Series([NaT], dtype="timedelta64[ns]")
+
+        # subtraction
+        tm.assert_series_equal(
+            datetime_series - single_nat_dtype_datetime, nat_series_dtype_timedelta
+        )
+
+        tm.assert_series_equal(
+            datetime_series - single_nat_dtype_timedelta, nat_series_dtype_timestamp
+        )
+        tm.assert_series_equal(
+            -single_nat_dtype_timedelta + datetime_series, nat_series_dtype_timestamp
+        )
+
+        # without a Series wrapping the NaT, it is ambiguous
+        # whether it is a datetime64 or timedelta64
+        # defaults to interpreting it as timedelta64
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp - single_nat_dtype_datetime,
+            nat_series_dtype_timedelta,
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp - single_nat_dtype_timedelta,
+            nat_series_dtype_timestamp,
+        )
+        tm.assert_series_equal(
+            -single_nat_dtype_timedelta + nat_series_dtype_timestamp,
+            nat_series_dtype_timestamp,
+        )
+        msg = "cannot subtract a datelike"
+        with pytest.raises(TypeError, match=msg):
+            timedelta_series - single_nat_dtype_datetime
+
+        # addition
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp + single_nat_dtype_timedelta,
+            nat_series_dtype_timestamp,
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_timedelta + nat_series_dtype_timestamp,
+            nat_series_dtype_timestamp,
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timestamp + single_nat_dtype_timedelta,
+            nat_series_dtype_timestamp,
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_timedelta + nat_series_dtype_timestamp,
+            nat_series_dtype_timestamp,
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta + single_nat_dtype_datetime,
+            nat_series_dtype_timestamp,
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_datetime + nat_series_dtype_timedelta,
+            nat_series_dtype_timestamp,
+        )
+
+    def test_ufunc_coercions(self, unit):
+        idx = date_range("2011-01-01", periods=3, freq="2D", name="x", unit=unit)
+
+        delta = np.timedelta64(1, "D")
+        exp = date_range("2011-01-02", periods=3, freq="2D", name="x", unit=unit)
+        for result in [idx + delta, np.add(idx, delta)]:
+            assert isinstance(result, DatetimeIndex)
+            tm.assert_index_equal(result, exp)
+            assert result.freq == "2D"
+
+        exp = date_range("2010-12-31", periods=3, freq="2D", name="x", unit=unit)
+
+        for result in [idx - delta, np.subtract(idx, delta)]:
+            assert isinstance(result, DatetimeIndex)
+            tm.assert_index_equal(result, exp)
+            assert result.freq == "2D"
+
+        # When adding/subtracting an ndarray (which has no .freq), the result
+        #  does not infer freq
+        idx = idx._with_freq(None)
+        delta = np.array(
+            [np.timedelta64(1, "D"), np.timedelta64(2, "D"), np.timedelta64(3, "D")]
+        )
+        exp = DatetimeIndex(
+            ["2011-01-02", "2011-01-05", "2011-01-08"], name="x"
+        ).as_unit(unit)
+
+        for result in [idx + delta, np.add(idx, delta)]:
+            tm.assert_index_equal(result, exp)
+            assert result.freq == exp.freq
+
+        exp = DatetimeIndex(
+            ["2010-12-31", "2011-01-01", "2011-01-02"], name="x"
+        ).as_unit(unit)
+        for result in [idx - delta, np.subtract(idx, delta)]:
+            assert isinstance(result, DatetimeIndex)
+            tm.assert_index_equal(result, exp)
+            assert result.freq == exp.freq
+
+    def test_dti_add_series(self, tz_naive_fixture, names):
+        # GH#13905
+        tz = tz_naive_fixture
+        index = DatetimeIndex(
+            ["2016-06-28 05:30", "2016-06-28 05:31"], tz=tz, name=names[0]
+        ).as_unit("ns")
+        ser = Series([Timedelta(seconds=5)] * 2, index=index, name=names[1])
+        expected = Series(index + Timedelta(seconds=5), index=index, name=names[2])
+
+        # passing name arg isn't enough when names[2] is None
+        expected.name = names[2]
+        assert expected.dtype == index.dtype
+        result = ser + index
+        tm.assert_series_equal(result, expected)
+        result2 = index + ser
+        tm.assert_series_equal(result2, expected)
+
+        expected = index + Timedelta(seconds=5)
+        result3 = ser.values + index
+        tm.assert_index_equal(result3, expected)
+        result4 = index + ser.values
+        tm.assert_index_equal(result4, expected)
+
+    @pytest.mark.parametrize("op", [operator.add, roperator.radd, operator.sub])
+    def test_dti_addsub_offset_arraylike(
+        self, performance_warning, tz_naive_fixture, names, op, index_or_series
+    ):
+        # GH#18849, GH#19744
+        other_box = index_or_series
+
+        tz = tz_naive_fixture
+        dti = date_range("2017-01-01", periods=2, tz=tz, name=names[0])
+        other = other_box([pd.offsets.MonthEnd(), pd.offsets.Day(n=2)], name=names[1])
+
+        xbox = get_upcast_box(dti, other)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = op(dti, other)
+
+        expected = DatetimeIndex(
+            [op(dti[n], other[n]) for n in range(len(dti))], name=names[2], freq="infer"
+        )
+        expected = tm.box_expected(expected, xbox).astype(object)
+        tm.assert_equal(res, expected)
+
+    @pytest.mark.parametrize("other_box", [pd.Index, np.array])
+    def test_dti_addsub_object_arraylike(
+        self, performance_warning, tz_naive_fixture, box_with_array, other_box
+    ):
+        tz = tz_naive_fixture
+
+        dti = date_range("2017-01-01", periods=2, tz=tz)
+        dtarr = tm.box_expected(dti, box_with_array)
+        other = other_box([pd.offsets.MonthEnd(), Timedelta(days=4)])
+        xbox = get_upcast_box(dtarr, other)
+
+        expected = DatetimeIndex(["2017-01-31", "2017-01-06"], tz=tz_naive_fixture)
+        expected = tm.box_expected(expected, xbox).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            result = dtarr + other
+        tm.assert_equal(result, expected)
+
+        expected = DatetimeIndex(["2016-12-31", "2016-12-29"], tz=tz_naive_fixture)
+        expected = tm.box_expected(expected, xbox).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            result = dtarr - other
+        tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("years", [-1, 0, 1])
+@pytest.mark.parametrize("months", [-2, 0, 2])
+def test_shift_months(years, months, unit):
+    dti = DatetimeIndex(
+        [
+            Timestamp("2000-01-05 00:15:00"),
+            Timestamp("2000-01-31 00:23:00"),
+            Timestamp("2000-01-01"),
+            Timestamp("2000-02-29"),
+            Timestamp("2000-12-31"),
+        ]
+    ).as_unit(unit)
+    shifted = shift_months(dti.asi8, years * 12 + months, reso=dti._data._creso)
+    shifted_dt64 = shifted.view(f"M8[{dti.unit}]")
+    actual = DatetimeIndex(shifted_dt64)
+
+    raw = [x + pd.offsets.DateOffset(years=years, months=months) for x in dti]
+    expected = DatetimeIndex(raw).as_unit(dti.unit)
+    tm.assert_index_equal(actual, expected)
+
+
+def test_dt64arr_addsub_object_dtype_2d(performance_warning):
+    # block-wise DataFrame operations will require operating on 2D
+    #  DatetimeArray/TimedeltaArray, so check that specifically.
+    dti = date_range("1994-02-13", freq="2W", periods=4)
+    dta = dti._data.reshape((4, 1))
+
+    other = np.array([[pd.offsets.Day(n)] for n in range(4)])
+    assert other.shape == dta.shape
+
+    with tm.assert_produces_warning(performance_warning):
+        result = dta + other
+    with tm.assert_produces_warning(performance_warning):
+        expected = (dta[:, 0] + other[:, 0]).reshape(-1, 1)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+    with tm.assert_produces_warning(performance_warning):
+        # Case where we expect to get a TimedeltaArray back
+        result2 = dta - dta.astype(object)
+
+    assert result2.shape == (4, 1)
+    assert all(td._value == 0 for td in result2.ravel())
+
+
+def test_non_nano_dt64_addsub_np_nat_scalars():
+    # GH 52295
+    ser = Series([1233242342344, 232432434324, 332434242344], dtype="datetime64[ms]")
+    result = ser - np.datetime64("nat", "ms")
+    expected = Series([NaT] * 3, dtype="timedelta64[ms]")
+    tm.assert_series_equal(result, expected)
+
+    result = ser + np.timedelta64("nat", "ms")
+    expected = Series([NaT] * 3, dtype="datetime64[ms]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_non_nano_dt64_addsub_np_nat_scalars_unitless():
+    # GH 52295
+    # TODO: Can we default to the ser unit?
+    ser = Series([1233242342344, 232432434324, 332434242344], dtype="datetime64[ms]")
+    result = ser - np.datetime64("nat")
+    expected = Series([NaT] * 3, dtype="timedelta64[ns]")
+    tm.assert_series_equal(result, expected)
+
+    result = ser + np.timedelta64("nat")
+    expected = Series([NaT] * 3, dtype="datetime64[ns]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_non_nano_dt64_addsub_np_nat_scalars_unsupported_unit():
+    # GH 52295
+    ser = Series([12332, 23243, 33243], dtype="datetime64[s]")
+    result = ser - np.datetime64("nat", "D")
+    expected = Series([NaT] * 3, dtype="timedelta64[s]")
+    tm.assert_series_equal(result, expected)
+
+    result = ser + np.timedelta64("nat", "D")
+    expected = Series([NaT] * 3, dtype="datetime64[s]")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arithmetic/test_interval.py b/pandas/pandas/tests/arithmetic/test_interval.py
new file mode 100644
index 0000000000000..0e316cf419cb0
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_interval.py
@@ -0,0 +1,306 @@
+import operator
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_list_like
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    Index,
+    Interval,
+    IntervalIndex,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    BooleanArray,
+    IntervalArray,
+)
+from pandas.tests.arithmetic.common import get_upcast_box
+
+
+@pytest.fixture(
+    params=[
+        (Index([0, 2, 4, 4]), Index([1, 3, 5, 8])),
+        (Index([0.0, 1.0, 2.0, np.nan]), Index([1.0, 2.0, 3.0, np.nan])),
+        (
+            timedelta_range("0 days", periods=3).insert(3, pd.NaT),
+            timedelta_range("1 day", periods=3).insert(3, pd.NaT),
+        ),
+        (
+            date_range("20170101", periods=3).insert(3, pd.NaT),
+            date_range("20170102", periods=3).insert(3, pd.NaT),
+        ),
+        (
+            date_range("20170101", periods=3, tz="US/Eastern").insert(3, pd.NaT),
+            date_range("20170102", periods=3, tz="US/Eastern").insert(3, pd.NaT),
+        ),
+    ],
+    ids=lambda x: str(x[0].dtype),
+)
+def left_right_dtypes(request):
+    """
+    Fixture for building an IntervalArray from various dtypes
+    """
+    return request.param
+
+
+@pytest.fixture
+def interval_array(left_right_dtypes):
+    """
+    Fixture to generate an IntervalArray of various dtypes containing NA if possible
+    """
+    left, right = left_right_dtypes
+    return IntervalArray.from_arrays(left, right)
+
+
+def create_categorical_intervals(left, right, closed="right"):
+    return Categorical(IntervalIndex.from_arrays(left, right, closed))
+
+
+def create_series_intervals(left, right, closed="right"):
+    return Series(IntervalArray.from_arrays(left, right, closed))
+
+
+def create_series_categorical_intervals(left, right, closed="right"):
+    return Series(Categorical(IntervalIndex.from_arrays(left, right, closed)))
+
+
+class TestComparison:
+    @pytest.fixture(params=[operator.eq, operator.ne])
+    def op(self, request):
+        return request.param
+
+    @pytest.fixture(
+        params=[
+            IntervalArray.from_arrays,
+            IntervalIndex.from_arrays,
+            create_categorical_intervals,
+            create_series_intervals,
+            create_series_categorical_intervals,
+        ],
+        ids=[
+            "IntervalArray",
+            "IntervalIndex",
+            "Categorical[Interval]",
+            "Series[Interval]",
+            "Series[Categorical[Interval]]",
+        ],
+    )
+    def interval_constructor(self, request):
+        """
+        Fixture for all pandas native interval constructors.
+        To be used as the LHS of IntervalArray comparisons.
+        """
+        return request.param
+
+    def elementwise_comparison(self, op, interval_array, other):
+        """
+        Helper that performs elementwise comparisons between `array` and `other`
+        """
+        other = other if is_list_like(other) else [other] * len(interval_array)
+        expected = np.array([op(x, y) for x, y in zip(interval_array, other)])
+        if isinstance(other, Series):
+            return Series(expected, index=other.index)
+        return expected
+
+    def test_compare_scalar_interval(self, op, interval_array):
+        # matches first interval
+        other = interval_array[0]
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # matches on a single endpoint but not both
+        other = Interval(interval_array.left[0], interval_array.right[1])
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_compare_scalar_interval_mixed_closed(self, op, closed, other_closed):
+        interval_array = IntervalArray.from_arrays(range(2), range(1, 3), closed=closed)
+        other = Interval(0, 1, closed=other_closed)
+
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_compare_scalar_na(self, op, interval_array, nulls_fixture, box_with_array):
+        box = box_with_array
+        obj = tm.box_expected(interval_array, box)
+        result = op(obj, nulls_fixture)
+
+        if nulls_fixture is pd.NA:
+            # GH#31882
+            exp = np.ones(interval_array.shape, dtype=bool)
+            expected = BooleanArray(exp, exp)
+        else:
+            expected = self.elementwise_comparison(op, interval_array, nulls_fixture)
+
+        if not (box is Index and nulls_fixture is pd.NA):
+            # don't cast expected from BooleanArray to ndarray[object]
+            xbox = get_upcast_box(obj, nulls_fixture, True)
+            expected = tm.box_expected(expected, xbox)
+
+        tm.assert_equal(result, expected)
+
+        rev = op(nulls_fixture, obj)
+        tm.assert_equal(rev, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            0,
+            1.0,
+            True,
+            "foo",
+            Timestamp("2017-01-01"),
+            Timestamp("2017-01-01", tz="US/Eastern"),
+            Timedelta("0 days"),
+            Period("2017-01-01", "D"),
+        ],
+    )
+    def test_compare_scalar_other(self, op, interval_array, other):
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_compare_list_like_interval(self, op, interval_array, interval_constructor):
+        # same endpoints
+        other = interval_constructor(interval_array.left, interval_array.right)
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_equal(result, expected)
+
+        # different endpoints
+        other = interval_constructor(
+            interval_array.left[::-1], interval_array.right[::-1]
+        )
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_equal(result, expected)
+
+        # all nan endpoints
+        other = interval_constructor([np.nan] * 4, [np.nan] * 4)
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_equal(result, expected)
+
+    def test_compare_list_like_interval_mixed_closed(
+        self, op, interval_constructor, closed, other_closed
+    ):
+        interval_array = IntervalArray.from_arrays(range(2), range(1, 3), closed=closed)
+        other = interval_constructor(range(2), range(1, 3), closed=other_closed)
+
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            (
+                Interval(0, 1),
+                Interval(Timedelta("1 day"), Timedelta("2 days")),
+                Interval(4, 5, "both"),
+                Interval(10, 20, "neither"),
+            ),
+            (0, 1.5, Timestamp("20170103"), np.nan),
+            (
+                Timestamp("20170102", tz="US/Eastern"),
+                Timedelta("2 days"),
+                "baz",
+                pd.NaT,
+            ),
+        ],
+    )
+    def test_compare_list_like_object(self, op, interval_array, other):
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_compare_list_like_nan(self, op, interval_array, nulls_fixture):
+        other = [nulls_fixture] * 4
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            np.arange(4, dtype="int64"),
+            np.arange(4, dtype="float64"),
+            date_range("2017-01-01", periods=4),
+            date_range("2017-01-01", periods=4, tz="US/Eastern"),
+            timedelta_range("0 days", periods=4),
+            period_range("2017-01-01", periods=4, freq="D"),
+            Categorical(list("abab")),
+            Categorical(date_range("2017-01-01", periods=4)),
+            pd.array(list("abcd")),
+            pd.array(["foo", 3.14, None, object()], dtype=object),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def test_compare_list_like_other(self, op, interval_array, other):
+        result = op(interval_array, other)
+        expected = self.elementwise_comparison(op, interval_array, other)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("length", [1, 3, 5])
+    @pytest.mark.parametrize("other_constructor", [IntervalArray, list])
+    def test_compare_length_mismatch_errors(self, op, other_constructor, length):
+        interval_array = IntervalArray.from_arrays(range(4), range(1, 5))
+        other = other_constructor([Interval(0, 1)] * length)
+        with pytest.raises(ValueError, match="Lengths must match to compare"):
+            op(interval_array, other)
+
+    @pytest.mark.parametrize(
+        "constructor, expected_type, assert_func",
+        [
+            (IntervalIndex, np.array, tm.assert_numpy_array_equal),
+            (Series, Series, tm.assert_series_equal),
+        ],
+    )
+    def test_index_series_compat(self, op, constructor, expected_type, assert_func):
+        # IntervalIndex/Series that rely on IntervalArray for comparisons
+        breaks = range(4)
+        index = constructor(IntervalIndex.from_breaks(breaks))
+
+        # scalar comparisons
+        other = index[0]
+        result = op(index, other)
+        expected = expected_type(self.elementwise_comparison(op, index, other))
+        assert_func(result, expected)
+
+        other = breaks[0]
+        result = op(index, other)
+        expected = expected_type(self.elementwise_comparison(op, index, other))
+        assert_func(result, expected)
+
+        # list-like comparisons
+        other = IntervalArray.from_breaks(breaks)
+        result = op(index, other)
+        expected = expected_type(self.elementwise_comparison(op, index, other))
+        assert_func(result, expected)
+
+        other = [index[0], breaks[0], "foo"]
+        result = op(index, other)
+        expected = expected_type(self.elementwise_comparison(op, index, other))
+        assert_func(result, expected)
+
+    @pytest.mark.parametrize("scalars", ["a", False, 1, 1.0, None])
+    def test_comparison_operations(self, scalars):
+        # GH #28981
+        expected = Series([False, False])
+        s = Series([Interval(0, 1), Interval(1, 2)], dtype="interval")
+        result = s == scalars
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arithmetic/test_numeric.py b/pandas/pandas/tests/arithmetic/test_numeric.py
new file mode 100644
index 0000000000000..d205569270705
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_numeric.py
@@ -0,0 +1,1545 @@
+# Arithmetic tests for DataFrame/Series/Index/Array classes that should
+# behave identically.
+# Specifically for numeric dtypes
+from __future__ import annotations
+
+from collections import abc
+from datetime import timedelta
+from decimal import Decimal
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    RangeIndex,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    array,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core import ops
+from pandas.core.computation import expressions as expr
+from pandas.tests.arithmetic.common import (
+    assert_invalid_addsub_type,
+    assert_invalid_comparison,
+)
+
+
+@pytest.fixture(autouse=True, params=[0, 1000000], ids=["numexpr", "python"])
+def switch_numexpr_min_elements(request, monkeypatch):
+    with monkeypatch.context() as m:
+        m.setattr(expr, "_MIN_ELEMENTS", request.param)
+        yield request.param
+
+
+@pytest.fixture(
+    params=[
+        # TODO: add more  dtypes here
+        Index(np.arange(5, dtype="float64")),
+        Index(np.arange(5, dtype="int64")),
+        Index(np.arange(5, dtype="uint64")),
+        RangeIndex(5),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def numeric_idx(request):
+    """
+    Several types of numeric-dtypes Index objects
+    """
+    return request.param
+
+
+def adjust_negative_zero(zero, expected):
+    """
+    Helper to adjust the expected result if we are dividing by -0.0
+    as opposed to 0.0
+    """
+    if np.signbit(np.array(zero)).any():
+        # All entries in the `zero` fixture should be either
+        #  all-negative or no-negative.
+        assert np.signbit(np.array(zero)).all()
+
+        expected *= -1
+
+    return expected
+
+
+def compare_op(series, other, op):
+    left = np.abs(series) if op in (ops.rpow, operator.pow) else series
+    right = np.abs(other) if op in (ops.rpow, operator.pow) else other
+
+    cython_or_numpy = op(left, right)
+    python = left.combine(right, op)
+    if isinstance(other, Series) and not other.index.equals(series.index):
+        python.index = python.index._with_freq(None)
+    tm.assert_series_equal(cython_or_numpy, python)
+
+
+# TODO: remove this kludge once mypy stops giving false positives here
+# List comprehension has incompatible type List[PandasObject]; expected List[RangeIndex]
+#  See GH#29725
+_ldtypes = ["i1", "i2", "i4", "i8", "u1", "u2", "u4", "u8", "f2", "f4", "f8"]
+lefts: list[Index | Series] = [RangeIndex(10, 40, 10)]
+lefts.extend([Series([10, 20, 30], dtype=dtype) for dtype in _ldtypes])
+lefts.extend([Index([10, 20, 30], dtype=dtype) for dtype in _ldtypes if dtype != "f2"])
+
+# ------------------------------------------------------------------
+# Comparisons
+
+
+class TestNumericComparisons:
+    def test_operator_series_comparison_zerorank(self):
+        # GH#13006
+        result = np.float64(0) > Series([1, 2, 3])
+        expected = 0.0 > Series([1, 2, 3])
+        tm.assert_series_equal(result, expected)
+        result = Series([1, 2, 3]) < np.float64(0)
+        expected = Series([1, 2, 3]) < 0.0
+        tm.assert_series_equal(result, expected)
+        result = np.array([0, 1, 2])[0] > Series([0, 1, 2])
+        expected = 0.0 > Series([1, 2, 3])
+        tm.assert_series_equal(result, expected)
+
+    def test_df_numeric_cmp_dt64_raises(self, box_with_array, fixed_now_ts):
+        # GH#8932, GH#22163
+        ts = fixed_now_ts
+        obj = np.array(range(5))
+        obj = tm.box_expected(obj, box_with_array)
+
+        assert_invalid_comparison(obj, ts, box_with_array)
+
+    def test_compare_invalid(self):
+        # GH#8058
+        # ops testing
+        a = Series(np.random.default_rng(2).standard_normal(5), name=0)
+        b = Series(np.random.default_rng(2).standard_normal(5))
+        b.name = pd.Timestamp("2000-01-01")
+        tm.assert_series_equal(a / b, 1 / (b / a))
+
+    def test_numeric_cmp_string_numexpr_path(self, box_with_array, monkeypatch):
+        # GH#36377, GH#35700
+        box = box_with_array
+        xbox = box if box is not Index else np.ndarray
+
+        obj = Series(np.random.default_rng(2).standard_normal(51))
+        obj = tm.box_expected(obj, box, transpose=False)
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 50)
+            result = obj == "a"
+
+        expected = Series(np.zeros(51, dtype=bool))
+        expected = tm.box_expected(expected, xbox, transpose=False)
+        tm.assert_equal(result, expected)
+
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 50)
+            result = obj != "a"
+        tm.assert_equal(result, ~expected)
+
+        msg = "Invalid comparison between dtype=float64 and str"
+        with pytest.raises(TypeError, match=msg):
+            obj < "a"
+
+
+# ------------------------------------------------------------------
+# Numeric dtypes Arithmetic with Datetime/Timedelta Scalar
+
+
+class TestNumericArraylikeArithmeticWithDatetimeLike:
+    @pytest.mark.parametrize("box_cls", [np.array, Index, Series])
+    @pytest.mark.parametrize(
+        "left", lefts, ids=lambda x: type(x).__name__ + str(x.dtype)
+    )
+    def test_mul_td64arr(self, left, box_cls):
+        # GH#22390
+        right = np.array([1, 2, 3], dtype="m8[s]")
+        right = box_cls(right)
+
+        expected = TimedeltaIndex(["10s", "40s", "90s"], dtype=right.dtype)
+
+        if isinstance(left, Series) or box_cls is Series:
+            expected = Series(expected)
+        assert expected.dtype == right.dtype
+
+        result = left * right
+        tm.assert_equal(result, expected)
+
+        result = right * left
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("box_cls", [np.array, Index, Series])
+    @pytest.mark.parametrize(
+        "left", lefts, ids=lambda x: type(x).__name__ + str(x.dtype)
+    )
+    def test_div_td64arr(self, left, box_cls):
+        # GH#22390
+        right = np.array([10, 40, 90], dtype="m8[s]")
+        right = box_cls(right)
+
+        expected = TimedeltaIndex(["1s", "2s", "3s"], dtype=right.dtype)
+        if isinstance(left, Series) or box_cls is Series:
+            expected = Series(expected)
+        assert expected.dtype == right.dtype
+
+        result = right / left
+        tm.assert_equal(result, expected)
+
+        result = right // left
+        tm.assert_equal(result, expected)
+
+        # (true_) needed for min-versions build 2022-12-26
+        msg = "ufunc '(true_)?divide' cannot use operands with types"
+        with pytest.raises(TypeError, match=msg):
+            left / right
+
+        msg = "ufunc 'floor_divide' cannot use operands with types"
+        with pytest.raises(TypeError, match=msg):
+            left // right
+
+    # TODO: also test Tick objects;
+    #  see test_numeric_arr_rdiv_tdscalar for note on these failing
+    @pytest.mark.parametrize(
+        "scalar_td",
+        [
+            Timedelta(days=1),
+            Timedelta(days=1).to_timedelta64(),
+            Timedelta(days=1).to_pytimedelta(),
+            Timedelta(days=1).to_timedelta64().astype("timedelta64[s]"),
+            Timedelta(days=1).to_timedelta64().astype("timedelta64[ms]"),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_numeric_arr_mul_tdscalar(self, scalar_td, numeric_idx, box_with_array):
+        # GH#19333
+        box = box_with_array
+        index = numeric_idx
+        expected = TimedeltaIndex([Timedelta(days=n) for n in range(len(index))])
+        if isinstance(scalar_td, np.timedelta64):
+            dtype = scalar_td.dtype
+            expected = expected.astype(dtype)
+        elif type(scalar_td) is timedelta:
+            expected = expected.astype("m8[us]")
+
+        index = tm.box_expected(index, box)
+        expected = tm.box_expected(expected, box)
+
+        result = index * scalar_td
+        tm.assert_equal(result, expected)
+
+        commute = scalar_td * index
+        tm.assert_equal(commute, expected)
+
+    @pytest.mark.parametrize(
+        "scalar_td",
+        [
+            Timedelta(days=1),
+            Timedelta(days=1).to_timedelta64(),
+            Timedelta(days=1).to_pytimedelta(),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    @pytest.mark.parametrize("dtype", [np.int64, np.float64])
+    def test_numeric_arr_mul_tdscalar_numexpr_path(
+        self, dtype, scalar_td, box_with_array
+    ):
+        # GH#44772 for the float64 case
+        box = box_with_array
+
+        arr_i8 = np.arange(2 * 10**4).astype(np.int64, copy=False)
+        arr = arr_i8.astype(dtype, copy=False)
+        obj = tm.box_expected(arr, box, transpose=False)
+
+        expected = arr_i8.view("timedelta64[D]").astype("timedelta64[ns]")
+        if type(scalar_td) is timedelta:
+            expected = expected.astype("timedelta64[us]")
+
+        expected = tm.box_expected(expected, box, transpose=False)
+
+        result = obj * scalar_td
+        tm.assert_equal(result, expected)
+
+        result = scalar_td * obj
+        tm.assert_equal(result, expected)
+
+    def test_numeric_arr_rdiv_tdscalar(self, three_days, numeric_idx, box_with_array):
+        box = box_with_array
+
+        index = numeric_idx[1:3]
+
+        expected = TimedeltaIndex(["3 Days", "36 Hours"])
+        if isinstance(three_days, np.timedelta64):
+            dtype = three_days.dtype
+            if dtype < np.dtype("m8[s]"):
+                # i.e. resolution is lower -> use lowest supported resolution
+                dtype = np.dtype("m8[s]")
+            expected = expected.astype(dtype)
+        elif type(three_days) is timedelta:
+            expected = expected.astype("m8[us]")
+        elif isinstance(
+            three_days,
+            (pd.offsets.Day, pd.offsets.Hour, pd.offsets.Minute, pd.offsets.Second),
+        ):
+            # closest reso is Second
+            expected = expected.astype("m8[s]")
+
+        index = tm.box_expected(index, box)
+        expected = tm.box_expected(expected, box)
+
+        result = three_days / index
+        tm.assert_equal(result, expected)
+
+        msg = "cannot use operands with types dtype"
+        with pytest.raises(TypeError, match=msg):
+            index / three_days
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            Timedelta(hours=31),
+            Timedelta(hours=31).to_pytimedelta(),
+            Timedelta(hours=31).to_timedelta64(),
+            Timedelta(hours=31).to_timedelta64().astype("m8[h]"),
+            np.timedelta64("NaT"),
+            np.timedelta64("NaT", "D"),
+            pd.offsets.Minute(3),
+            pd.offsets.Second(0),
+            # GH#28080 numeric+datetimelike should raise; Timestamp used
+            #  to raise NullFrequencyError but that behavior was removed in 1.0
+            pd.Timestamp("2021-01-01", tz="Asia/Tokyo"),
+            pd.Timestamp("2021-01-01"),
+            pd.Timestamp("2021-01-01").to_pydatetime(),
+            pd.Timestamp("2021-01-01", tz="UTC").to_pydatetime(),
+            pd.Timestamp("2021-01-01").to_datetime64(),
+            np.datetime64("NaT", "ns"),
+            pd.NaT,
+        ],
+        ids=repr,
+    )
+    def test_add_sub_datetimedeltalike_invalid(
+        self, numeric_idx, other, box_with_array
+    ):
+        box = box_with_array
+
+        left = tm.box_expected(numeric_idx, box)
+        msg = "|".join(
+            [
+                "unsupported operand type",
+                "Addition/subtraction of integers and integer-arrays",
+                "Instead of adding/subtracting",
+                "cannot use operands with types dtype",
+                "Concatenation operation is not implemented for NumPy arrays",
+                "Cannot (add|subtract) NaT (to|from) ndarray",
+                # pd.array vs np.datetime64 case
+                r"operand type\(s\) all returned NotImplemented from __array_ufunc__",
+                "can only perform ops with numeric values",
+                "cannot subtract DatetimeArray from ndarray",
+                # pd.Timedelta(1) + Index([0, 1, 2])
+                "Cannot add or subtract Timedelta from integers",
+            ]
+        )
+        assert_invalid_addsub_type(left, other, msg)
+
+
+# ------------------------------------------------------------------
+# Arithmetic
+
+
+class TestDivisionByZero:
+    def test_div_zero(self, zero, numeric_idx):
+        idx = numeric_idx
+
+        expected = Index([np.nan, np.inf, np.inf, np.inf, np.inf], dtype=np.float64)
+        # We only adjust for Index, because Series does not yet apply
+        #  the adjustment correctly.
+        expected2 = adjust_negative_zero(zero, expected)
+
+        result = idx / zero
+        tm.assert_index_equal(result, expected2)
+        ser_compat = Series(idx).astype("i8") / np.array(zero).astype("i8")
+        tm.assert_series_equal(ser_compat, Series(expected))
+
+    def test_floordiv_zero(self, zero, numeric_idx):
+        idx = numeric_idx
+
+        expected = Index([np.nan, np.inf, np.inf, np.inf, np.inf], dtype=np.float64)
+        # We only adjust for Index, because Series does not yet apply
+        #  the adjustment correctly.
+        expected2 = adjust_negative_zero(zero, expected)
+
+        result = idx // zero
+        tm.assert_index_equal(result, expected2)
+        ser_compat = Series(idx).astype("i8") // np.array(zero).astype("i8")
+        tm.assert_series_equal(ser_compat, Series(expected))
+
+    def test_mod_zero(self, zero, numeric_idx):
+        idx = numeric_idx
+
+        expected = Index([np.nan, np.nan, np.nan, np.nan, np.nan], dtype=np.float64)
+        result = idx % zero
+        tm.assert_index_equal(result, expected)
+        ser_compat = Series(idx).astype("i8") % np.array(zero).astype("i8")
+        tm.assert_series_equal(ser_compat, Series(result))
+
+    def test_divmod_zero(self, zero, numeric_idx):
+        idx = numeric_idx
+
+        exleft = Index([np.nan, np.inf, np.inf, np.inf, np.inf], dtype=np.float64)
+        exright = Index([np.nan, np.nan, np.nan, np.nan, np.nan], dtype=np.float64)
+        exleft = adjust_negative_zero(zero, exleft)
+
+        result = divmod(idx, zero)
+        tm.assert_index_equal(result[0], exleft)
+        tm.assert_index_equal(result[1], exright)
+
+    @pytest.mark.parametrize("op", [operator.truediv, operator.floordiv])
+    def test_div_negative_zero(self, zero, numeric_idx, op):
+        # Check that -1 / -0.0 returns np.inf, not -np.inf
+        if numeric_idx.dtype == np.uint64:
+            pytest.skip(f"Div by negative 0 not relevant for {numeric_idx.dtype}")
+        idx = numeric_idx - 3
+
+        expected = Index([-np.inf, -np.inf, -np.inf, np.nan, np.inf], dtype=np.float64)
+        expected = adjust_negative_zero(zero, expected)
+
+        result = op(idx, zero)
+        tm.assert_index_equal(result, expected)
+
+    # ------------------------------------------------------------------
+
+    @pytest.mark.parametrize("dtype1", [np.int64, np.float64, np.uint64])
+    def test_ser_div_ser(
+        self,
+        switch_numexpr_min_elements,
+        dtype1,
+        any_real_numpy_dtype,
+    ):
+        # no longer do integer div for any ops, but deal with the 0's
+        dtype2 = any_real_numpy_dtype
+
+        first = Series([3, 4, 5, 8], name="first").astype(dtype1)
+        second = Series([0, 0, 0, 3], name="second").astype(dtype2)
+
+        with np.errstate(all="ignore"):
+            expected = Series(
+                first.values.astype(np.float64) / second.values,
+                dtype="float64",
+                name=None,
+            )
+        expected.iloc[0:3] = np.inf
+        if first.dtype == "int64" and second.dtype == "float32":
+            # when using numexpr, the casting rules are slightly different
+            # and int64/float32 combo results in float32 instead of float64
+            if expr.USE_NUMEXPR and switch_numexpr_min_elements == 0:
+                expected = expected.astype("float32")
+
+        result = first / second
+        tm.assert_series_equal(result, expected)
+        assert not result.equals(second / first)
+
+    @pytest.mark.parametrize("dtype1", [np.int64, np.float64, np.uint64])
+    def test_ser_divmod_zero(self, dtype1, any_real_numpy_dtype):
+        # GH#26987
+        dtype2 = any_real_numpy_dtype
+        left = Series([1, 1]).astype(dtype1)
+        right = Series([0, 2]).astype(dtype2)
+
+        # GH#27321 pandas convention is to set 1 // 0 to np.inf, as opposed
+        #  to numpy which sets to np.nan; patch `expected[0]` below
+        expected = left // right, left % right
+        expected = list(expected)
+        expected[0] = expected[0].astype(np.float64)
+        expected[0][0] = np.inf
+        result = divmod(left, right)
+
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+        # rdivmod case
+        result = divmod(left.values, right)
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+    def test_ser_divmod_inf(self):
+        left = Series([np.inf, 1.0])
+        right = Series([np.inf, 2.0])
+
+        expected = left // right, left % right
+        result = divmod(left, right)
+
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+        # rdivmod case
+        result = divmod(left.values, right)
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+    def test_rdiv_zero_compat(self):
+        # GH#8674
+        zero_array = np.array([0] * 5)
+        data = np.random.default_rng(2).standard_normal(5)
+        expected = Series([0.0] * 5)
+
+        result = zero_array / Series(data)
+        tm.assert_series_equal(result, expected)
+
+        result = Series(zero_array) / data
+        tm.assert_series_equal(result, expected)
+
+        result = Series(zero_array) / Series(data)
+        tm.assert_series_equal(result, expected)
+
+    def test_div_zero_inf_signs(self):
+        # GH#9144, inf signing
+        ser = Series([-1, 0, 1], name="first")
+        expected = Series([-np.inf, np.nan, np.inf], name="first")
+
+        result = ser / 0
+        tm.assert_series_equal(result, expected)
+
+    def test_rdiv_zero(self):
+        # GH#9144
+        ser = Series([-1, 0, 1], name="first")
+        expected = Series([0.0, np.nan, 0.0], name="first")
+
+        result = 0 / ser
+        tm.assert_series_equal(result, expected)
+
+    def test_floordiv_div(self):
+        # GH#9144
+        ser = Series([-1, 0, 1], name="first")
+
+        result = ser // 0
+        expected = Series([-np.inf, np.nan, np.inf], name="first")
+        tm.assert_series_equal(result, expected)
+
+    def test_df_div_zero_df(self):
+        # integer div, but deal with the 0's (GH#9144)
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+        result = df / df
+
+        first = Series([1.0, 1.0, 1.0, 1.0])
+        second = Series([np.nan, np.nan, np.nan, 1])
+        expected = pd.DataFrame({"first": first, "second": second})
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_div_zero_array(self):
+        # integer div, but deal with the 0's (GH#9144)
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+
+        first = Series([1.0, 1.0, 1.0, 1.0])
+        second = Series([np.nan, np.nan, np.nan, 1])
+        expected = pd.DataFrame({"first": first, "second": second})
+
+        with np.errstate(all="ignore"):
+            arr = df.values.astype("float") / df.values
+        result = pd.DataFrame(arr, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_div_zero_int(self):
+        # integer div, but deal with the 0's (GH#9144)
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+
+        result = df / 0
+        expected = pd.DataFrame(np.inf, index=df.index, columns=df.columns)
+        expected.iloc[0:3, 1] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        # numpy has a slightly different (wrong) treatment
+        with np.errstate(all="ignore"):
+            arr = df.values.astype("float64") / 0
+        result2 = pd.DataFrame(arr, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_df_div_zero_series_does_not_commute(self):
+        # integer div, but deal with the 0's (GH#9144)
+        df = pd.DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+        ser = df[0]
+        res = ser / df
+        res2 = df / ser
+        assert not res.fillna(0).equals(res2.fillna(0))
+
+    # ------------------------------------------------------------------
+    # Mod By Zero
+
+    def test_df_mod_zero_df(self):
+        # GH#3590, modulo as ints
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+        # this is technically wrong, as the integer portion is coerced to float
+        first = Series([0, 0, 0, 0])
+        first = first.astype("float64")
+        second = Series([np.nan, np.nan, np.nan, 0])
+        expected = pd.DataFrame({"first": first, "second": second})
+        result = df % df
+        tm.assert_frame_equal(result, expected)
+
+        # GH#38939 If we dont pass copy=False, df is consolidated and
+        #  result["first"] is float64 instead of int64
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]}, copy=False)
+        first = Series([0, 0, 0, 0], dtype="int64")
+        second = Series([np.nan, np.nan, np.nan, 0])
+        expected = pd.DataFrame({"first": first, "second": second})
+        result = df % df
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_mod_zero_array(self):
+        # GH#3590, modulo as ints
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+
+        # this is technically wrong, as the integer portion is coerced to float
+        # ###
+        first = Series([0, 0, 0, 0], dtype="float64")
+        second = Series([np.nan, np.nan, np.nan, 0])
+        expected = pd.DataFrame({"first": first, "second": second})
+
+        # numpy has a slightly different (wrong) treatment
+        with np.errstate(all="ignore"):
+            arr = df.values % df.values
+        result2 = pd.DataFrame(arr, index=df.index, columns=df.columns, dtype="float64")
+        result2.iloc[0:3, 1] = np.nan
+        tm.assert_frame_equal(result2, expected)
+
+    def test_df_mod_zero_int(self):
+        # GH#3590, modulo as ints
+        df = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+
+        result = df % 0
+        expected = pd.DataFrame(np.nan, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result, expected)
+
+        # numpy has a slightly different (wrong) treatment
+        with np.errstate(all="ignore"):
+            arr = df.values.astype("float64") % 0
+        result2 = pd.DataFrame(arr, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_df_mod_zero_series_does_not_commute(self):
+        # GH#3590, modulo as ints
+        # not commutative with series
+        df = pd.DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+        ser = df[0]
+        res = ser % df
+        res2 = df % ser
+        assert not res.fillna(0).equals(res2.fillna(0))
+
+
+class TestMultiplicationDivision:
+    # __mul__, __rmul__, __div__, __rdiv__, __floordiv__, __rfloordiv__
+    # for non-timestamp/timedelta/period dtypes
+
+    def test_divide_decimal(self, box_with_array):
+        # resolves issue GH#9787
+        box = box_with_array
+        ser = Series([Decimal(10)])
+        expected = Series([Decimal(5)])
+
+        ser = tm.box_expected(ser, box)
+        expected = tm.box_expected(expected, box)
+
+        result = ser / Decimal(2)
+
+        tm.assert_equal(result, expected)
+
+        result = ser // Decimal(2)
+        tm.assert_equal(result, expected)
+
+    def test_div_equiv_binop(self):
+        # Test Series.div as well as Series.__div__
+        # float/integer issue
+        # GH#7785
+        first = Series([1, 0], name="first")
+        second = Series([-0.01, -0.02], name="second")
+        expected = Series([-0.01, -np.inf])
+
+        result = second.div(first)
+        tm.assert_series_equal(result, expected, check_names=False)
+
+        result = second / first
+        tm.assert_series_equal(result, expected)
+
+    def test_div_int(self, numeric_idx):
+        idx = numeric_idx
+        result = idx / 1
+        expected = idx.astype("float64")
+        tm.assert_index_equal(result, expected)
+
+        result = idx / 2
+        expected = Index(idx.values / 2)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("op", [operator.mul, ops.rmul, operator.floordiv])
+    def test_mul_int_identity(self, op, numeric_idx, box_with_array):
+        idx = numeric_idx
+        idx = tm.box_expected(idx, box_with_array)
+
+        result = op(idx, 1)
+        tm.assert_equal(result, idx)
+
+    def test_mul_int_array(self, numeric_idx):
+        idx = numeric_idx
+        didx = idx * idx
+
+        result = idx * np.array(5, dtype="int64")
+        tm.assert_index_equal(result, idx * 5)
+
+        arr_dtype = "uint64" if idx.dtype == np.uint64 else "int64"
+        result = idx * np.arange(5, dtype=arr_dtype)
+        tm.assert_index_equal(result, didx)
+
+    def test_mul_int_series(self, numeric_idx):
+        idx = numeric_idx
+        didx = idx * idx
+
+        arr_dtype = "uint64" if idx.dtype == np.uint64 else "int64"
+        result = idx * Series(np.arange(5, dtype=arr_dtype))
+        tm.assert_series_equal(result, Series(didx))
+
+    def test_mul_float_series(self, numeric_idx):
+        idx = numeric_idx
+        rng5 = np.arange(5, dtype="float64")
+
+        result = idx * Series(rng5 + 0.1)
+        expected = Series(rng5 * (rng5 + 0.1))
+        tm.assert_series_equal(result, expected)
+
+    def test_mul_index(self, numeric_idx):
+        idx = numeric_idx
+
+        result = idx * idx
+        tm.assert_index_equal(result, idx**2)
+
+    def test_mul_datelike_raises(self, numeric_idx):
+        idx = numeric_idx
+        msg = "cannot perform __rmul__ with this index type"
+        with pytest.raises(TypeError, match=msg):
+            idx * date_range("20130101", periods=5)
+
+    def test_mul_size_mismatch_raises(self, numeric_idx):
+        idx = numeric_idx
+        msg = "operands could not be broadcast together"
+        with pytest.raises(ValueError, match=msg):
+            idx * idx[0:3]
+        with pytest.raises(ValueError, match=msg):
+            idx * np.array([1, 2])
+
+    @pytest.mark.parametrize("op", [operator.pow, ops.rpow])
+    def test_pow_float(self, op, numeric_idx, box_with_array):
+        # test power calculations both ways, GH#14973
+        box = box_with_array
+        idx = numeric_idx
+        expected = Index(op(idx.values, 2.0))
+
+        idx = tm.box_expected(idx, box)
+        expected = tm.box_expected(expected, box)
+
+        result = op(idx, 2.0)
+        tm.assert_equal(result, expected)
+
+    def test_modulo(self, numeric_idx, box_with_array):
+        # GH#9244
+        box = box_with_array
+        idx = numeric_idx
+        expected = Index(idx.values % 2)
+
+        idx = tm.box_expected(idx, box)
+        expected = tm.box_expected(expected, box)
+
+        result = idx % 2
+        tm.assert_equal(result, expected)
+
+    def test_divmod_scalar(self, numeric_idx):
+        idx = numeric_idx
+
+        result = divmod(idx, 2)
+        with np.errstate(all="ignore"):
+            div, mod = divmod(idx.values, 2)
+
+        expected = Index(div), Index(mod)
+        for r, e in zip(result, expected):
+            tm.assert_index_equal(r, e)
+
+    def test_divmod_ndarray(self, numeric_idx):
+        idx = numeric_idx
+        other = np.ones(idx.values.shape, dtype=idx.values.dtype) * 2
+
+        result = divmod(idx, other)
+        with np.errstate(all="ignore"):
+            div, mod = divmod(idx.values, other)
+
+        expected = Index(div), Index(mod)
+        for r, e in zip(result, expected):
+            tm.assert_index_equal(r, e)
+
+    def test_divmod_series(self, numeric_idx):
+        idx = numeric_idx
+        other = np.ones(idx.values.shape, dtype=idx.values.dtype) * 2
+
+        result = divmod(idx, Series(other))
+        with np.errstate(all="ignore"):
+            div, mod = divmod(idx.values, other)
+
+        expected = Series(div), Series(mod)
+        for r, e in zip(result, expected):
+            tm.assert_series_equal(r, e)
+
+    @pytest.mark.parametrize("other", [np.nan, 7, -23, 2.718, -3.14, np.inf])
+    def test_ops_np_scalar(self, other):
+        vals = np.random.default_rng(2).standard_normal((5, 3))
+        f = lambda x: pd.DataFrame(
+            x, index=list("ABCDE"), columns=["jim", "joe", "jolie"]
+        )
+
+        df = f(vals)
+
+        tm.assert_frame_equal(df / np.array(other), f(vals / other))
+        tm.assert_frame_equal(np.array(other) * df, f(vals * other))
+        tm.assert_frame_equal(df + np.array(other), f(vals + other))
+        tm.assert_frame_equal(np.array(other) - df, f(other - vals))
+
+    # TODO: This came from series.test.test_operators, needs cleanup
+    def test_operators_frame(self):
+        # rpow does not work with DataFrame
+        ts = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        ts.name = "ts"
+
+        df = pd.DataFrame({"A": ts})
+
+        tm.assert_series_equal(ts + ts, ts + df["A"], check_names=False)
+        tm.assert_series_equal(ts**ts, ts ** df["A"], check_names=False)
+        tm.assert_series_equal(ts < ts, ts < df["A"], check_names=False)
+        tm.assert_series_equal(ts / ts, ts / df["A"], check_names=False)
+
+    # TODO: this came from tests.series.test_analytics, needs cleanup and
+    #  de-duplication with test_modulo above
+    def test_modulo2(self):
+        with np.errstate(all="ignore"):
+            # GH#3590, modulo as ints
+            p = pd.DataFrame({"first": [3, 4, 5, 8], "second": [0, 0, 0, 3]})
+            result = p["first"] % p["second"]
+            expected = Series(p["first"].values % p["second"].values, dtype="float64")
+            expected.iloc[0:3] = np.nan
+            tm.assert_series_equal(result, expected)
+
+            result = p["first"] % 0
+            expected = Series(np.nan, index=p.index, name="first")
+            tm.assert_series_equal(result, expected)
+
+            p = p.astype("float64")
+            result = p["first"] % p["second"]
+            expected = Series(p["first"].values % p["second"].values)
+            tm.assert_series_equal(result, expected)
+
+            p = p.astype("float64")
+            result = p["first"] % p["second"]
+            result2 = p["second"] % p["first"]
+            assert not result.equals(result2)
+
+    def test_modulo_zero_int(self):
+        # GH#9144
+        with np.errstate(all="ignore"):
+            s = Series([0, 1])
+
+            result = s % 0
+            expected = Series([np.nan, np.nan])
+            tm.assert_series_equal(result, expected)
+
+            result = 0 % s
+            expected = Series([np.nan, 0.0])
+            tm.assert_series_equal(result, expected)
+
+
+class TestAdditionSubtraction:
+    # __add__, __sub__, __radd__, __rsub__, __iadd__, __isub__
+    # for non-timestamp/timedelta/period dtypes
+
+    @pytest.mark.parametrize(
+        "first, second, expected",
+        [
+            (
+                Series([1, 2, 3], index=list("ABC"), name="x"),
+                Series([2, 2, 2], index=list("ABD"), name="x"),
+                Series([3.0, 4.0, np.nan, np.nan], index=list("ABCD"), name="x"),
+            ),
+            (
+                Series([1, 2, 3], index=list("ABC"), name="x"),
+                Series([2, 2, 2, 2], index=list("ABCD"), name="x"),
+                Series([3, 4, 5, np.nan], index=list("ABCD"), name="x"),
+            ),
+        ],
+    )
+    def test_add_series(self, first, second, expected):
+        # GH#1134
+        tm.assert_series_equal(first + second, expected)
+        tm.assert_series_equal(second + first, expected)
+
+    @pytest.mark.parametrize(
+        "first, second, expected",
+        [
+            (
+                pd.DataFrame({"x": [1, 2, 3]}, index=list("ABC")),
+                pd.DataFrame({"x": [2, 2, 2]}, index=list("ABD")),
+                pd.DataFrame({"x": [3.0, 4.0, np.nan, np.nan]}, index=list("ABCD")),
+            ),
+            (
+                pd.DataFrame({"x": [1, 2, 3]}, index=list("ABC")),
+                pd.DataFrame({"x": [2, 2, 2, 2]}, index=list("ABCD")),
+                pd.DataFrame({"x": [3, 4, 5, np.nan]}, index=list("ABCD")),
+            ),
+        ],
+    )
+    def test_add_frames(self, first, second, expected):
+        # GH#1134
+        tm.assert_frame_equal(first + second, expected)
+        tm.assert_frame_equal(second + first, expected)
+
+    # TODO: This came from series.test.test_operators, needs cleanup
+    def test_series_frame_radd_bug(self, fixed_now_ts):
+        # GH#353
+        vals = Series([str(i) for i in range(5)])
+        result = "foo_" + vals
+        expected = vals.map(lambda x: "foo_" + x)
+        tm.assert_series_equal(result, expected)
+
+        frame = pd.DataFrame({"vals": vals})
+        result = "foo_" + frame
+        expected = pd.DataFrame({"vals": vals.map(lambda x: "foo_" + x)})
+        tm.assert_frame_equal(result, expected)
+
+        ts = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+
+        # really raise this time
+        fix_now = fixed_now_ts.to_pydatetime()
+        msg = "|".join(
+            [
+                "unsupported operand type",
+                # wrong error message, see https://github.com/numpy/numpy/issues/18832
+                "Concatenation operation",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            fix_now + ts
+
+        with pytest.raises(TypeError, match=msg):
+            ts + fix_now
+
+    # TODO: This came from series.test.test_operators, needs cleanup
+    def test_datetime64_with_index(self):
+        # arithmetic integer ops with an index
+        ser = Series(np.random.default_rng(2).standard_normal(5))
+        expected = ser - ser.index.to_series()
+        result = ser - ser.index
+        tm.assert_series_equal(result, expected)
+
+        # GH#4629
+        # arithmetic datetime64 ops with an index
+        ser = Series(
+            date_range("20130101", periods=5),
+            index=date_range("20130101", periods=5),
+        )
+        expected = ser - ser.index.to_series()
+        result = ser - ser.index
+        tm.assert_series_equal(result, expected)
+
+        msg = "cannot subtract PeriodArray from DatetimeArray"
+        with pytest.raises(TypeError, match=msg):
+            # GH#18850
+            result = ser - ser.index.to_period()
+
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)),
+            index=date_range("20130101", periods=5),
+        )
+        df["date"] = pd.Timestamp("20130102")
+        df["expected"] = df["date"] - df.index.to_series()
+        df["result"] = df["date"] - df.index
+        tm.assert_series_equal(df["result"], df["expected"], check_names=False)
+
+    # TODO: taken from tests.frame.test_operators, needs cleanup
+    def test_frame_operators(self, float_frame):
+        frame = float_frame
+
+        garbage = np.random.default_rng(2).random(4)
+        colSeries = Series(garbage, index=np.array(frame.columns))
+
+        idSum = frame + frame
+        seriesSum = frame + colSeries
+
+        for col, series in idSum.items():
+            for idx, val in series.items():
+                origVal = frame[col][idx] * 2
+                if not np.isnan(val):
+                    assert val == origVal
+                else:
+                    assert np.isnan(origVal)
+
+        for col, series in seriesSum.items():
+            for idx, val in series.items():
+                origVal = frame[col][idx] + colSeries[col]
+                if not np.isnan(val):
+                    assert val == origVal
+                else:
+                    assert np.isnan(origVal)
+
+    def test_frame_operators_col_align(self, float_frame):
+        frame2 = pd.DataFrame(float_frame, columns=["D", "C", "B", "A"])
+        added = frame2 + frame2
+        expected = frame2 * 2
+        tm.assert_frame_equal(added, expected)
+
+    def test_frame_operators_none_to_nan(self):
+        df = pd.DataFrame({"a": ["a", None, "b"]})
+        tm.assert_frame_equal(df + df, pd.DataFrame({"a": ["aa", np.nan, "bb"]}))
+
+    @pytest.mark.parametrize("dtype", ("float", "int64"))
+    def test_frame_operators_empty_like(self, dtype):
+        # Test for issue #10181
+        frames = [
+            pd.DataFrame(dtype=dtype),
+            pd.DataFrame(columns=["A"], dtype=dtype),
+            pd.DataFrame(index=[0], dtype=dtype),
+        ]
+        for df in frames:
+            assert (df + df).equals(df)
+            tm.assert_frame_equal(df + df, df)
+
+    @pytest.mark.parametrize(
+        "func",
+        [lambda x: x * 2, lambda x: x[::2], lambda x: 5],
+        ids=["multiply", "slice", "constant"],
+    )
+    def test_series_operators_arithmetic(self, all_arithmetic_functions, func):
+        op = all_arithmetic_functions
+        series = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        other = func(series)
+        compare_op(series, other, op)
+
+    @pytest.mark.parametrize(
+        "func", [lambda x: x + 1, lambda x: 5], ids=["add", "constant"]
+    )
+    def test_series_operators_compare(self, comparison_op, func):
+        op = comparison_op
+        series = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        other = func(series)
+        compare_op(series, other, op)
+
+    @pytest.mark.parametrize(
+        "func",
+        [lambda x: x * 2, lambda x: x[::2], lambda x: 5],
+        ids=["multiply", "slice", "constant"],
+    )
+    def test_divmod(self, func):
+        series = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        other = func(series)
+        results = divmod(series, other)
+        if isinstance(other, abc.Iterable) and len(series) != len(other):
+            # if the lengths don't match, this is the test where we use
+            # `tser[::2]`. Pad every other value in `other_np` with nan.
+            other_np = []
+            for n in other:
+                other_np.append(n)
+                other_np.append(np.nan)
+        else:
+            other_np = other
+        other_np = np.asarray(other_np)
+        with np.errstate(all="ignore"):
+            expecteds = divmod(series.values, np.asarray(other_np))
+
+        for result, expected in zip(results, expecteds):
+            # check the values, name, and index separately
+            tm.assert_almost_equal(np.asarray(result), expected)
+
+            assert result.name == series.name
+            tm.assert_index_equal(result.index, series.index._with_freq(None))
+
+    def test_series_divmod_zero(self):
+        # Check that divmod uses pandas convention for division by zero,
+        #  which does not match numpy.
+        # pandas convention has
+        #  1/0 == np.inf
+        #  -1/0 == -np.inf
+        #  1/-0.0 == -np.inf
+        #  -1/-0.0 == np.inf
+        tser = Series(
+            np.arange(1, 11, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        other = tser * 0
+
+        result = divmod(tser, other)
+        exp1 = Series([np.inf] * len(tser), index=tser.index, name="ts")
+        exp2 = Series([np.nan] * len(tser), index=tser.index, name="ts")
+        tm.assert_series_equal(result[0], exp1)
+        tm.assert_series_equal(result[1], exp2)
+
+
+class TestUFuncCompat:
+    # TODO: add more dtypes
+    @pytest.mark.parametrize("holder", [Index, RangeIndex, Series])
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_ufunc_compat(self, holder, dtype):
+        box = Series if holder is Series else Index
+
+        if holder is RangeIndex:
+            if dtype != np.int64:
+                pytest.skip(f"dtype {dtype} not relevant for RangeIndex")
+            idx = RangeIndex(0, 5, name="foo")
+        else:
+            idx = holder(np.arange(5, dtype=dtype), name="foo")
+        result = np.sin(idx)
+        expected = box(np.sin(np.arange(5, dtype=dtype)), name="foo")
+        tm.assert_equal(result, expected)
+
+    # TODO: add more dtypes
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_ufunc_coercions(self, index_or_series, dtype):
+        idx = index_or_series([1, 2, 3, 4, 5], dtype=dtype, name="x")
+        box = index_or_series
+
+        result = np.sqrt(idx)
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index(np.sqrt(np.array([1, 2, 3, 4, 5], dtype=np.float64)), name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+        result = np.divide(idx, 2.0)
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index([0.5, 1.0, 1.5, 2.0, 2.5], dtype=np.float64, name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+        # _evaluate_numeric_binop
+        result = idx + 2.0
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index([3.0, 4.0, 5.0, 6.0, 7.0], dtype=np.float64, name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+        result = idx - 2.0
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index([-1.0, 0.0, 1.0, 2.0, 3.0], dtype=np.float64, name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+        result = idx * 1.0
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index([1.0, 2.0, 3.0, 4.0, 5.0], dtype=np.float64, name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+        result = idx / 2.0
+        assert result.dtype == "f8" and isinstance(result, box)
+        exp = Index([0.5, 1.0, 1.5, 2.0, 2.5], dtype=np.float64, name="x")
+        exp = tm.box_expected(exp, box)
+        tm.assert_equal(result, exp)
+
+    # TODO: add more dtypes
+    @pytest.mark.parametrize("holder", [Index, Series])
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_ufunc_multiple_return_values(self, holder, dtype):
+        obj = holder([1, 2, 3], dtype=dtype, name="x")
+        box = Series if holder is Series else Index
+
+        result = np.modf(obj)
+        assert isinstance(result, tuple)
+        exp1 = Index([0.0, 0.0, 0.0], dtype=np.float64, name="x")
+        exp2 = Index([1.0, 2.0, 3.0], dtype=np.float64, name="x")
+        tm.assert_equal(result[0], tm.box_expected(exp1, box))
+        tm.assert_equal(result[1], tm.box_expected(exp2, box))
+
+    def test_ufunc_at(self):
+        s = Series([0, 1, 2], index=[1, 2, 3], name="x")
+        np.add.at(s, [0, 2], 10)
+        expected = Series([10, 1, 12], index=[1, 2, 3], name="x")
+        tm.assert_series_equal(s, expected)
+
+
+class TestObjectDtypeEquivalence:
+    # Tests that arithmetic operations match operations executed elementwise
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_numarr_with_dtype_add_nan(self, dtype, box_with_array):
+        box = box_with_array
+        ser = Series([1, 2, 3], dtype=dtype)
+        expected = Series([np.nan, np.nan, np.nan], dtype=dtype)
+
+        ser = tm.box_expected(ser, box)
+        expected = tm.box_expected(expected, box)
+
+        result = np.nan + ser
+        tm.assert_equal(result, expected)
+
+        result = ser + np.nan
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_numarr_with_dtype_add_int(self, dtype, box_with_array):
+        box = box_with_array
+        ser = Series([1, 2, 3], dtype=dtype)
+        expected = Series([2, 3, 4], dtype=dtype)
+
+        ser = tm.box_expected(ser, box)
+        expected = tm.box_expected(expected, box)
+
+        result = 1 + ser
+        tm.assert_equal(result, expected)
+
+        result = ser + 1
+        tm.assert_equal(result, expected)
+
+    # TODO: moved from tests.series.test_operators; needs cleanup
+    @pytest.mark.parametrize(
+        "op",
+        [operator.add, operator.sub, operator.mul, operator.truediv, operator.floordiv],
+    )
+    def test_operators_reverse_object(self, op):
+        # GH#56
+        arr = Series(
+            np.random.default_rng(2).standard_normal(10),
+            index=np.arange(10),
+            dtype=object,
+        )
+
+        result = op(1.0, arr)
+        expected = op(1.0, arr.astype(float))
+        tm.assert_series_equal(result.astype(float), expected)
+
+
+class TestNumericArithmeticUnsorted:
+    # Tests in this class have been moved from type-specific test modules
+    #  but not yet sorted, parametrized, and de-duplicated
+    @pytest.mark.parametrize(
+        "op",
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.floordiv,
+            operator.truediv,
+        ],
+    )
+    @pytest.mark.parametrize(
+        "idx1",
+        [
+            RangeIndex(0, 10, 1),
+            RangeIndex(0, 20, 2),
+            RangeIndex(-10, 10, 2),
+            RangeIndex(5, -5, -1),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "idx2",
+        [
+            RangeIndex(0, 10, 1),
+            RangeIndex(0, 20, 2),
+            RangeIndex(-10, 10, 2),
+            RangeIndex(5, -5, -1),
+        ],
+    )
+    def test_binops_index(self, op, idx1, idx2):
+        idx1 = idx1._rename("foo")
+        idx2 = idx2._rename("bar")
+        result = op(idx1, idx2)
+        expected = op(Index(idx1.to_numpy()), Index(idx2.to_numpy()))
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+    @pytest.mark.parametrize(
+        "op",
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.floordiv,
+            operator.truediv,
+        ],
+    )
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            RangeIndex(0, 10, 1),
+            RangeIndex(0, 20, 2),
+            RangeIndex(-10, 10, 2),
+            RangeIndex(5, -5, -1),
+        ],
+    )
+    @pytest.mark.parametrize("scalar", [-1, 1, 2])
+    def test_binops_index_scalar(self, op, idx, scalar):
+        result = op(idx, scalar)
+        expected = op(Index(idx.to_numpy()), scalar)
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+    @pytest.mark.parametrize("idx1", [RangeIndex(0, 10, 1), RangeIndex(0, 20, 2)])
+    @pytest.mark.parametrize("idx2", [RangeIndex(0, 10, 1), RangeIndex(0, 20, 2)])
+    def test_binops_index_pow(self, idx1, idx2):
+        # numpy does not allow powers of negative integers so test separately
+        # https://github.com/numpy/numpy/pull/8127
+        idx1 = idx1._rename("foo")
+        idx2 = idx2._rename("bar")
+        result = pow(idx1, idx2)
+        expected = pow(Index(idx1.to_numpy()), Index(idx2.to_numpy()))
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+    @pytest.mark.parametrize("idx", [RangeIndex(0, 10, 1), RangeIndex(0, 20, 2)])
+    @pytest.mark.parametrize("scalar", [1, 2])
+    def test_binops_index_scalar_pow(self, idx, scalar):
+        # numpy does not allow powers of negative integers so test separately
+        # https://github.com/numpy/numpy/pull/8127
+        result = pow(idx, scalar)
+        expected = pow(Index(idx.to_numpy()), scalar)
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+    # TODO: divmod?
+    @pytest.mark.parametrize(
+        "op",
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.floordiv,
+            operator.truediv,
+            operator.pow,
+            operator.mod,
+        ],
+    )
+    def test_arithmetic_with_frame_or_series(self, op):
+        # check that we return NotImplemented when operating with Series
+        # or DataFrame
+        index = RangeIndex(5)
+        other = Series(np.random.default_rng(2).standard_normal(5))
+
+        expected = op(Series(index), other)
+        result = op(index, other)
+        tm.assert_series_equal(result, expected)
+
+        other = pd.DataFrame(np.random.default_rng(2).standard_normal((2, 5)))
+        expected = op(pd.DataFrame([index, index]), other)
+        result = op(index, other)
+        tm.assert_frame_equal(result, expected)
+
+    def test_numeric_compat2(self):
+        # validate that we are handling the RangeIndex overrides to numeric ops
+        # and returning RangeIndex where possible
+
+        idx = RangeIndex(0, 10, 2)
+
+        result = idx * 2
+        expected = RangeIndex(0, 20, 4)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx + 2
+        expected = RangeIndex(2, 12, 2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx - 2
+        expected = RangeIndex(-2, 8, 2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx / 2
+        expected = RangeIndex(0, 5, 1).astype("float64")
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx / 4
+        expected = RangeIndex(0, 10, 2) / 4
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx // 1
+        expected = idx
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # __mul__
+        result = idx * idx
+        expected = Index(idx.values * idx.values)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # __pow__
+        idx = RangeIndex(0, 1000, 2)
+        result = idx**2
+        expected = Index(idx._values) ** 2
+        tm.assert_index_equal(Index(result.values), expected, exact=True)
+
+    @pytest.mark.parametrize(
+        "idx, div, expected",
+        [
+            # TODO: add more dtypes
+            (RangeIndex(0, 1000, 2), 2, RangeIndex(0, 500, 1)),
+            (RangeIndex(-99, -201, -3), -3, RangeIndex(33, 67, 1)),
+            (
+                RangeIndex(0, 1000, 1),
+                2,
+                Index(RangeIndex(0, 1000, 1)._values) // 2,
+            ),
+            (
+                RangeIndex(0, 100, 1),
+                2.0,
+                Index(RangeIndex(0, 100, 1)._values) // 2.0,
+            ),
+            (RangeIndex(0), 50, RangeIndex(0)),
+            (RangeIndex(2, 4, 2), 3, RangeIndex(0, 1, 1)),
+            (RangeIndex(-5, -10, -6), 4, RangeIndex(-2, -1, 1)),
+            (RangeIndex(-100, -200, 3), 2, RangeIndex(0)),
+        ],
+    )
+    def test_numeric_compat2_floordiv(self, idx, div, expected):
+        # __floordiv__
+        tm.assert_index_equal(idx // div, expected, exact=True)
+
+    @pytest.mark.parametrize("dtype", [np.int64, np.float64])
+    @pytest.mark.parametrize("delta", [1, 0, -1])
+    def test_addsub_arithmetic(self, dtype, delta):
+        # GH#8142
+        delta = dtype(delta)
+        index = Index([10, 11, 12], dtype=dtype)
+        result = index + delta
+        expected = Index(index.values + delta, dtype=dtype)
+        tm.assert_index_equal(result, expected)
+
+        # this subtraction used to fail
+        result = index - delta
+        expected = Index(index.values - delta, dtype=dtype)
+        tm.assert_index_equal(result, expected)
+
+        tm.assert_index_equal(index + index, 2 * index)
+        tm.assert_index_equal(index - index, 0 * index)
+        assert not (index - index).empty
+
+    def test_pow_nan_with_zero(self, box_with_array):
+        left = Index([np.nan, np.nan, np.nan])
+        right = Index([0, 0, 0])
+        expected = Index([1.0, 1.0, 1.0])
+
+        left = tm.box_expected(left, box_with_array)
+        right = tm.box_expected(right, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = left**right
+        tm.assert_equal(result, expected)
+
+
+def test_fill_value_inf_masking():
+    # GH #27464 make sure we mask 0/1 with Inf and not NaN
+    df = pd.DataFrame({"A": [0, 1, 2], "B": [1.1, None, 1.1]})
+
+    other = pd.DataFrame({"A": [1.1, 1.2, 1.3]}, index=[0, 2, 3])
+
+    result = df.rfloordiv(other, fill_value=1)
+
+    expected = pd.DataFrame(
+        {"A": [np.inf, 1.0, 0.0, 1.0], "B": [0.0, np.nan, 0.0, np.nan]}
+    )
+    tm.assert_frame_equal(result, expected, check_index_type=False)
+
+
+def test_dataframe_div_silenced():
+    # GH#26793
+    pdf1 = pd.DataFrame(
+        {
+            "A": np.arange(10),
+            "B": [np.nan, 1, 2, 3, 4] * 2,
+            "C": [np.nan] * 10,
+            "D": np.arange(10),
+        },
+        index=list("abcdefghij"),
+        columns=list("ABCD"),
+    )
+    pdf2 = pd.DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        index=list("abcdefghjk"),
+        columns=list("ABCX"),
+    )
+    with tm.assert_produces_warning(None):
+        pdf1.div(pdf2, fill_value=0)
+
+
+@pytest.mark.parametrize(
+    "data, expected_data",
+    [([0, 1, 2], [0, 2, 4])],
+)
+@pytest.mark.parametrize("box_pandas_1d_array", [Index, Series, tm.to_array])
+@pytest.mark.parametrize("box_1d_array", [Index, Series, tm.to_array, np.array, list])
+def test_integer_array_add_list_like(
+    box_pandas_1d_array, box_1d_array, data, expected_data
+):
+    # GH22606 Verify operators with IntegerArray and list-likes
+    arr = array(data, dtype="Int64")
+    container = box_pandas_1d_array(arr)
+    left = container + box_1d_array(data)
+    right = box_1d_array(data) + container
+
+    if Series in [box_1d_array, box_pandas_1d_array]:
+        cls = Series
+    elif Index in [box_1d_array, box_pandas_1d_array]:
+        cls = Index
+    else:
+        cls = array
+
+    expected = cls(expected_data, dtype="Int64")
+
+    tm.assert_equal(left, expected)
+    tm.assert_equal(right, expected)
+
+
+def test_sub_multiindex_swapped_levels():
+    # GH 9952
+    df = pd.DataFrame(
+        {"a": np.random.default_rng(2).standard_normal(6)},
+        index=pd.MultiIndex.from_product(
+            [["a", "b"], [0, 1, 2]], names=["levA", "levB"]
+        ),
+    )
+    df2 = df.copy()
+    df2.index = df2.index.swaplevel(0, 1)
+    result = df - df2
+    expected = pd.DataFrame([0.0] * 6, columns=["a"], index=df.index)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("power", [1, 2, 5])
+@pytest.mark.parametrize("string_size", [0, 1, 2, 5])
+def test_empty_str_comparison(power, string_size):
+    # GH 37348
+    a = np.array(range(10**power))
+    right = pd.DataFrame(a, dtype=np.int64)
+    left = " " * string_size
+
+    result = right == left
+    expected = pd.DataFrame(np.zeros(right.shape, dtype=bool))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_series_add_sub_with_UInt64():
+    # GH 22023
+    series1 = Series([1, 2, 3])
+    series2 = Series([2, 1, 3], dtype="UInt64")
+
+    result = series1 + series2
+    expected = Series([3, 3, 6], dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+    result = series1 - series2
+    expected = Series([-1, 1, 0], dtype="Float64")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arithmetic/test_object.py b/pandas/pandas/tests/arithmetic/test_object.py
new file mode 100644
index 0000000000000..bc0f78d3aa01a
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_object.py
@@ -0,0 +1,410 @@
+# Arithmetic tests for DataFrame/Series/Index/Array classes that should
+# behave identically.
+# Specifically for object dtype
+import datetime
+from decimal import Decimal
+import operator
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Series,
+    Timestamp,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.core import ops
+
+# ------------------------------------------------------------------
+# Comparisons
+
+
+class TestObjectComparisons:
+    def test_comparison_object_numeric_nas(self, comparison_op):
+        ser = Series(np.random.default_rng(2).standard_normal(10), dtype=object)
+        shifted = ser.shift(2)
+
+        func = comparison_op
+
+        result = func(ser, shifted)
+        expected = func(ser.astype(float), shifted.astype(float))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    def test_object_comparisons(self, infer_string):
+        with option_context("future.infer_string", infer_string):
+            ser = Series(["a", "b", np.nan, "c", "a"])
+
+            result = ser == "a"
+            expected = Series([True, False, False, False, True])
+            tm.assert_series_equal(result, expected)
+
+            result = ser < "a"
+            expected = Series([False, False, False, False, False])
+            tm.assert_series_equal(result, expected)
+
+            result = ser != "a"
+            expected = -(ser == "a")
+            tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_more_na_comparisons(self, dtype):
+        left = Series(["a", np.nan, "c"], dtype=dtype)
+        right = Series(["a", np.nan, "d"], dtype=dtype)
+
+        result = left == right
+        expected = Series([True, False, False])
+        tm.assert_series_equal(result, expected)
+
+        result = left != right
+        expected = Series([False, True, True])
+        tm.assert_series_equal(result, expected)
+
+        result = left == np.nan
+        expected = Series([False, False, False])
+        tm.assert_series_equal(result, expected)
+
+        result = left != np.nan
+        expected = Series([True, True, True])
+        tm.assert_series_equal(result, expected)
+
+
+# ------------------------------------------------------------------
+# Arithmetic
+
+
+class TestArithmetic:
+    def test_add_period_to_array_of_offset(self):
+        # GH#50162
+        per = pd.Period("2012-1-1", freq="D")
+        pi = pd.period_range("2012-1-1", periods=10, freq="D")
+        idx = per - pi
+
+        expected = pd.Index([x + per for x in idx], dtype=object)
+        result = idx + per
+        tm.assert_index_equal(result, expected)
+
+        result = per + idx
+        tm.assert_index_equal(result, expected)
+
+    # TODO: parametrize
+    def test_pow_ops_object(self):
+        # GH#22922
+        # pow is weird with masking & 1, so testing here
+        a = Series([1, np.nan, 1, np.nan], dtype=object)
+        b = Series([1, np.nan, np.nan, 1], dtype=object)
+        result = a**b
+        expected = Series(a.values**b.values, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        result = b**a
+        expected = Series(b.values**a.values, dtype=object)
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    @pytest.mark.parametrize("other", ["category", "Int64"])
+    def test_add_extension_scalar(self, other, box_with_array, op):
+        # GH#22378
+        # Check that scalars satisfying is_extension_array_dtype(obj)
+        # do not incorrectly try to dispatch to an ExtensionArray operation
+
+        arr = Series(["a", "b", "c"])
+        expected = Series([op(x, other) for x in arr])
+
+        arr = tm.box_expected(arr, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = op(arr, other)
+        tm.assert_equal(result, expected)
+
+    def test_objarr_add_str(self, box_with_array):
+        ser = Series(["x", np.nan, "x"])
+        expected = Series(["xa", np.nan, "xa"])
+
+        ser = tm.box_expected(ser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = ser + "a"
+        tm.assert_equal(result, expected)
+
+    def test_objarr_radd_str(self, box_with_array):
+        ser = Series(["x", np.nan, "x"])
+        expected = Series(["ax", np.nan, "ax"])
+
+        ser = tm.box_expected(ser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = "a" + ser
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            [1, 2, 3],
+            [1.1, 2.2, 3.3],
+            [Timestamp("2011-01-01"), Timestamp("2011-01-02"), pd.NaT],
+            ["x", "y", 1],
+        ],
+    )
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_objarr_radd_str_invalid(self, dtype, data, box_with_array):
+        ser = Series(data, dtype=dtype)
+
+        ser = tm.box_expected(ser, box_with_array)
+        msg = "|".join(
+            [
+                "can only concatenate str",
+                "did not contain a loop with signature matching types",
+                "unsupported operand type",
+                "must be str",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            "foo_" + ser
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd, operator.sub, ops.rsub])
+    def test_objarr_add_invalid(self, op, box_with_array):
+        # invalid ops
+        box = box_with_array
+
+        obj_ser = Series(list("abc"), dtype=object, name="objects")
+
+        obj_ser = tm.box_expected(obj_ser, box)
+        msg = "|".join(
+            [
+                "can only concatenate str",
+                "unsupported operand type",
+                "must be str",
+                "has no kernel",
+            ]
+        )
+        with pytest.raises(Exception, match=msg):
+            op(obj_ser, 1)
+        with pytest.raises(Exception, match=msg):
+            op(obj_ser, np.array(1, dtype=np.int64))
+
+    # TODO: Moved from tests.series.test_operators; needs cleanup
+    def test_operators_na_handling(self):
+        ser = Series(["foo", "bar", "baz", np.nan])
+        result = "prefix_" + ser
+        expected = Series(["prefix_foo", "prefix_bar", "prefix_baz", np.nan])
+        tm.assert_series_equal(result, expected)
+
+        result = ser + "_suffix"
+        expected = Series(["foo_suffix", "bar_suffix", "baz_suffix", np.nan])
+        tm.assert_series_equal(result, expected)
+
+    # TODO: parametrize over box
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_series_with_dtype_radd_timedelta(self, dtype):
+        # note this test is _not_ aimed at timedelta64-dtyped Series
+        # as of 2.0 we retain object dtype when ser.dtype == object
+        ser = Series(
+            [pd.Timedelta("1 days"), pd.Timedelta("2 days"), pd.Timedelta("3 days")],
+            dtype=dtype,
+        )
+        expected = Series(
+            [pd.Timedelta("4 days"), pd.Timedelta("5 days"), pd.Timedelta("6 days")],
+            dtype=dtype,
+        )
+
+        result = pd.Timedelta("3 days") + ser
+        tm.assert_series_equal(result, expected)
+
+        result = ser + pd.Timedelta("3 days")
+        tm.assert_series_equal(result, expected)
+
+    # TODO: cleanup & parametrize over box
+    def test_mixed_timezone_series_ops_object(self):
+        # GH#13043
+        ser = Series(
+            [
+                Timestamp("2015-01-01", tz="US/Eastern"),
+                Timestamp("2015-01-01", tz="Asia/Tokyo"),
+            ],
+            name="xxx",
+        )
+        assert ser.dtype == object
+
+        exp = Series(
+            [
+                Timestamp("2015-01-02", tz="US/Eastern"),
+                Timestamp("2015-01-02", tz="Asia/Tokyo"),
+            ],
+            name="xxx",
+        )
+        tm.assert_series_equal(ser + pd.Timedelta("1 days"), exp)
+        tm.assert_series_equal(pd.Timedelta("1 days") + ser, exp)
+
+        # object series & object series
+        ser2 = Series(
+            [
+                Timestamp("2015-01-03", tz="US/Eastern"),
+                Timestamp("2015-01-05", tz="Asia/Tokyo"),
+            ],
+            name="xxx",
+        )
+        assert ser2.dtype == object
+        exp = Series(
+            [pd.Timedelta("2 days"), pd.Timedelta("4 days")], name="xxx", dtype=object
+        )
+        tm.assert_series_equal(ser2 - ser, exp)
+        tm.assert_series_equal(ser - ser2, -exp)
+
+        ser = Series(
+            [pd.Timedelta("01:00:00"), pd.Timedelta("02:00:00")],
+            name="xxx",
+            dtype=object,
+        )
+        assert ser.dtype == object
+
+        exp = Series(
+            [pd.Timedelta("01:30:00"), pd.Timedelta("02:30:00")],
+            name="xxx",
+            dtype=object,
+        )
+        tm.assert_series_equal(ser + pd.Timedelta("00:30:00"), exp)
+        tm.assert_series_equal(pd.Timedelta("00:30:00") + ser, exp)
+
+    # TODO: cleanup & parametrize over box
+    def test_iadd_preserves_name(self):
+        # GH#17067, GH#19723 __iadd__ and __isub__ should preserve index name
+        ser = Series([1, 2, 3])
+        ser.index.name = "foo"
+
+        ser.index += 1
+        assert ser.index.name == "foo"
+
+        ser.index -= 1
+        assert ser.index.name == "foo"
+
+    def test_add_string(self):
+        # from bug report
+        index = pd.Index(["a", "b", "c"])
+        index2 = index + "foo"
+
+        assert "a" not in index2
+        assert "afoo" in index2
+
+    def test_iadd_string(self):
+        index = pd.Index(["a", "b", "c"])
+        # doesn't fail test unless there is a check before `+=`
+        assert "a" in index
+
+        index += "_x"
+        assert "a_x" in index
+
+    def test_add(self):
+        index = pd.Index([str(i) for i in range(10)])
+        expected = pd.Index(index.values * 2)
+        tm.assert_index_equal(index + index, expected)
+        tm.assert_index_equal(index + index.tolist(), expected)
+        tm.assert_index_equal(index.tolist() + index, expected)
+
+        # test add and radd
+        index = pd.Index(list("abc"))
+        expected = pd.Index(["a1", "b1", "c1"])
+        tm.assert_index_equal(index + "1", expected)
+        expected = pd.Index(["1a", "1b", "1c"])
+        tm.assert_index_equal("1" + index, expected)
+
+    def test_sub_fail(self):
+        index = pd.Index([str(i) for i in range(10)])
+
+        msg = "unsupported operand type|Cannot broadcast|sub' not supported"
+        with pytest.raises(TypeError, match=msg):
+            index - "a"
+        with pytest.raises(TypeError, match=msg):
+            index - index
+        with pytest.raises(TypeError, match=msg):
+            index - index.tolist()
+        with pytest.raises(TypeError, match=msg):
+            index.tolist() - index
+
+    def test_sub_object(self):
+        # GH#19369
+        index = pd.Index([Decimal(1), Decimal(2)])
+        expected = pd.Index([Decimal(0), Decimal(1)])
+
+        result = index - Decimal(1)
+        tm.assert_index_equal(result, expected)
+
+        result = index - pd.Index([Decimal(1), Decimal(1)])
+        tm.assert_index_equal(result, expected)
+
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            index - "foo"
+
+        with pytest.raises(TypeError, match=msg):
+            index - np.array([2, "foo"], dtype=object)
+
+    def test_rsub_object(self, fixed_now_ts):
+        # GH#19369
+        index = pd.Index([Decimal(1), Decimal(2)])
+        expected = pd.Index([Decimal(1), Decimal(0)])
+
+        result = Decimal(2) - index
+        tm.assert_index_equal(result, expected)
+
+        result = np.array([Decimal(2), Decimal(2)]) - index
+        tm.assert_index_equal(result, expected)
+
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            "foo" - index
+
+        with pytest.raises(TypeError, match=msg):
+            np.array([True, fixed_now_ts]) - index
+
+
+class MyIndex(pd.Index):
+    # Simple index subclass that tracks ops calls.
+
+    _calls: int
+
+    @classmethod
+    def _simple_new(cls, values, name=None, dtype=None):
+        result = object.__new__(cls)
+        result._data = values
+        result._name = name
+        result._calls = 0
+        result._reset_identity()
+
+        return result
+
+    def __add__(self, other):
+        self._calls += 1
+        return self._simple_new(self._data)
+
+    def __radd__(self, other):
+        return self.__add__(other)
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        [datetime.timedelta(1), datetime.timedelta(2)],
+        [datetime.datetime(2000, 1, 1), datetime.datetime(2000, 1, 2)],
+        [pd.Period("2000"), pd.Period("2001")],
+        ["a", "b"],
+    ],
+    ids=["timedelta", "datetime", "period", "object"],
+)
+def test_index_ops_defer_to_unknown_subclasses(other):
+    # https://github.com/pandas-dev/pandas/issues/31109
+    values = np.array(
+        [datetime.date(2000, 1, 1), datetime.date(2000, 1, 2)], dtype=object
+    )
+    a = MyIndex._simple_new(values)
+    other = pd.Index(other)
+    result = other + a
+    assert isinstance(result, MyIndex)
+    assert a._calls == 1
diff --git a/pandas/pandas/tests/arithmetic/test_period.py b/pandas/pandas/tests/arithmetic/test_period.py
new file mode 100644
index 0000000000000..67762e0b89c73
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_period.py
@@ -0,0 +1,1679 @@
+# Arithmetic tests for DataFrame/Series/Index/Array classes that should
+# behave identically.
+# Specifically for Period dtype
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    IncompatibleFrequency,
+    Period,
+    Timestamp,
+    to_offset,
+)
+
+import pandas as pd
+from pandas import (
+    PeriodIndex,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core import ops
+from pandas.core.arrays import TimedeltaArray
+from pandas.tests.arithmetic.common import (
+    assert_invalid_addsub_type,
+    assert_invalid_comparison,
+    get_upcast_box,
+)
+
+_common_mismatch = [
+    pd.offsets.YearBegin(2),
+    pd.offsets.MonthBegin(1),
+    pd.offsets.Minute(),
+]
+
+
+@pytest.fixture(
+    params=[
+        Timedelta(minutes=30).to_pytimedelta(),
+        np.timedelta64(30, "s"),
+        Timedelta(seconds=30),
+    ]
+    + _common_mismatch
+)
+def not_hourly(request):
+    """
+    Several timedelta-like and DateOffset instances that are _not_
+    compatible with Hourly frequencies.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        np.timedelta64(365, "D"),
+        Timedelta(days=365).to_pytimedelta(),
+        Timedelta(days=365),
+    ]
+    + _common_mismatch
+)
+def mismatched_freq(request):
+    """
+    Several timedelta-like and DateOffset instances that are _not_
+    compatible with Monthly or Annual frequencies.
+    """
+    return request.param
+
+
+# ------------------------------------------------------------------
+# Comparisons
+
+
+class TestPeriodArrayLikeComparisons:
+    # Comparison tests for PeriodDtype vectors fully parametrized over
+    #  DataFrame/Series/PeriodIndex/PeriodArray.  Ideally all comparison
+    #  tests will eventually end up here.
+
+    @pytest.mark.parametrize("other", ["2017", Period("2017", freq="D")])
+    def test_eq_scalar(self, other, box_with_array):
+        idx = PeriodIndex(["2017", "2017", "2018"], freq="D")
+        idx = tm.box_expected(idx, box_with_array)
+        xbox = get_upcast_box(idx, other, True)
+
+        expected = np.array([True, True, False])
+        expected = tm.box_expected(expected, xbox)
+
+        result = idx == other
+
+        tm.assert_equal(result, expected)
+
+    def test_compare_zerodim(self, box_with_array):
+        # GH#26689 make sure we unbox zero-dimensional arrays
+
+        pi = period_range("2000", periods=4)
+        other = np.array(pi.to_numpy()[0])
+
+        pi = tm.box_expected(pi, box_with_array)
+        xbox = get_upcast_box(pi, other, True)
+
+        result = pi <= other
+        expected = np.array([True, False, False, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            "foo",
+            Timestamp("2021-01-01"),
+            Timedelta(days=4),
+            9,
+            9.5,
+            2000,  # specifically don't consider 2000 to match Period("2000", "D")
+            False,
+            None,
+        ],
+    )
+    def test_compare_invalid_scalar(self, box_with_array, scalar):
+        # GH#28980
+        # comparison with scalar that cannot be interpreted as a Period
+        pi = period_range("2000", periods=4)
+        parr = tm.box_expected(pi, box_with_array)
+        assert_invalid_comparison(parr, scalar, box_with_array)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            pd.date_range("2000", periods=4).array,
+            pd.timedelta_range("1D", periods=4).array,
+            np.arange(4),
+            np.arange(4).astype(np.float64),
+            list(range(4)),
+            # match Period semantics by not treating integers as Periods
+            [2000, 2001, 2002, 2003],
+            np.arange(2000, 2004),
+            np.arange(2000, 2004).astype(object),
+            pd.Index([2000, 2001, 2002, 2003]),
+        ],
+    )
+    def test_compare_invalid_listlike(self, box_with_array, other):
+        pi = period_range("2000", periods=4)
+        parr = tm.box_expected(pi, box_with_array)
+        assert_invalid_comparison(parr, other, box_with_array)
+
+    @pytest.mark.parametrize("other_box", [list, np.array, lambda x: x.astype(object)])
+    def test_compare_object_dtype(self, box_with_array, other_box):
+        pi = period_range("2000", periods=5)
+        parr = tm.box_expected(pi, box_with_array)
+
+        other = other_box(pi)
+        xbox = get_upcast_box(parr, other, True)
+
+        expected = np.array([True, True, True, True, True])
+        expected = tm.box_expected(expected, xbox)
+
+        result = parr == other
+        tm.assert_equal(result, expected)
+        result = parr <= other
+        tm.assert_equal(result, expected)
+        result = parr >= other
+        tm.assert_equal(result, expected)
+
+        result = parr != other
+        tm.assert_equal(result, ~expected)
+        result = parr < other
+        tm.assert_equal(result, ~expected)
+        result = parr > other
+        tm.assert_equal(result, ~expected)
+
+        other = other_box(pi[::-1])
+
+        expected = np.array([False, False, True, False, False])
+        expected = tm.box_expected(expected, xbox)
+        result = parr == other
+        tm.assert_equal(result, expected)
+
+        expected = np.array([True, True, True, False, False])
+        expected = tm.box_expected(expected, xbox)
+        result = parr <= other
+        tm.assert_equal(result, expected)
+
+        expected = np.array([False, False, True, True, True])
+        expected = tm.box_expected(expected, xbox)
+        result = parr >= other
+        tm.assert_equal(result, expected)
+
+        expected = np.array([True, True, False, True, True])
+        expected = tm.box_expected(expected, xbox)
+        result = parr != other
+        tm.assert_equal(result, expected)
+
+        expected = np.array([True, True, False, False, False])
+        expected = tm.box_expected(expected, xbox)
+        result = parr < other
+        tm.assert_equal(result, expected)
+
+        expected = np.array([False, False, False, True, True])
+        expected = tm.box_expected(expected, xbox)
+        result = parr > other
+        tm.assert_equal(result, expected)
+
+
+class TestPeriodIndexComparisons:
+    # TODO: parameterize over boxes
+
+    def test_pi_cmp_period(self):
+        idx = period_range("2007-01", periods=20, freq="M")
+        per = idx[10]
+
+        result = idx < per
+        exp = idx.values < idx.values[10]
+        tm.assert_numpy_array_equal(result, exp)
+
+        # Tests Period.__richcmp__ against ndarray[object, ndim=2]
+        result = idx.values.reshape(10, 2) < per
+        tm.assert_numpy_array_equal(result, exp.reshape(10, 2))
+
+        # Tests Period.__richcmp__ against ndarray[object, ndim=0]
+        result = idx < np.array(per)
+        tm.assert_numpy_array_equal(result, exp)
+
+    # TODO: moved from test_datetime64; de-duplicate with version below
+    def test_parr_cmp_period_scalar2(self, box_with_array):
+        pi = period_range("2000-01-01", periods=10, freq="D")
+
+        val = pi[3]
+        expected = [x > val for x in pi]
+
+        ser = tm.box_expected(pi, box_with_array)
+        xbox = get_upcast_box(ser, val, True)
+
+        expected = tm.box_expected(expected, xbox)
+        result = ser > val
+        tm.assert_equal(result, expected)
+
+        val = pi[5]
+        result = ser > val
+        expected = [x > val for x in pi]
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_parr_cmp_period_scalar(self, freq, box_with_array):
+        # GH#13200
+        base = PeriodIndex(["2011-01", "2011-02", "2011-03", "2011-04"], freq=freq)
+        base = tm.box_expected(base, box_with_array)
+        per = Period("2011-02", freq=freq)
+        xbox = get_upcast_box(base, per, True)
+
+        exp = np.array([False, True, False, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base == per, exp)
+        tm.assert_equal(per == base, exp)
+
+        exp = np.array([True, False, True, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base != per, exp)
+        tm.assert_equal(per != base, exp)
+
+        exp = np.array([False, False, True, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base > per, exp)
+        tm.assert_equal(per < base, exp)
+
+        exp = np.array([True, False, False, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base < per, exp)
+        tm.assert_equal(per > base, exp)
+
+        exp = np.array([False, True, True, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base >= per, exp)
+        tm.assert_equal(per <= base, exp)
+
+        exp = np.array([True, True, False, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base <= per, exp)
+        tm.assert_equal(per >= base, exp)
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_parr_cmp_pi(self, freq, box_with_array):
+        # GH#13200
+        base = PeriodIndex(["2011-01", "2011-02", "2011-03", "2011-04"], freq=freq)
+        base = tm.box_expected(base, box_with_array)
+
+        # TODO: could also box idx?
+        idx = PeriodIndex(["2011-02", "2011-01", "2011-03", "2011-05"], freq=freq)
+
+        xbox = get_upcast_box(base, idx, True)
+
+        exp = np.array([False, False, True, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base == idx, exp)
+
+        exp = np.array([True, True, False, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base != idx, exp)
+
+        exp = np.array([False, True, False, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base > idx, exp)
+
+        exp = np.array([True, False, False, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base < idx, exp)
+
+        exp = np.array([False, True, True, False])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base >= idx, exp)
+
+        exp = np.array([True, False, True, True])
+        exp = tm.box_expected(exp, xbox)
+        tm.assert_equal(base <= idx, exp)
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_parr_cmp_pi_mismatched_freq(self, freq, box_with_array):
+        # GH#13200
+        # different base freq
+        base = PeriodIndex(["2011-01", "2011-02", "2011-03", "2011-04"], freq=freq)
+        base = tm.box_expected(base, box_with_array)
+
+        msg = rf"Invalid comparison between dtype=period\[{freq}\] and Period"
+        with pytest.raises(TypeError, match=msg):
+            base <= Period("2011", freq="Y")
+
+        with pytest.raises(TypeError, match=msg):
+            Period("2011", freq="Y") >= base
+
+        # TODO: Could parametrize over boxes for idx?
+        idx = PeriodIndex(["2011", "2012", "2013", "2014"], freq="Y")
+        rev_msg = r"Invalid comparison between dtype=period\[Y-DEC\] and PeriodArray"
+        idx_msg = rev_msg if box_with_array in [tm.to_array, pd.array] else msg
+        with pytest.raises(TypeError, match=idx_msg):
+            base <= idx
+
+        # Different frequency
+        msg = rf"Invalid comparison between dtype=period\[{freq}\] and Period"
+        with pytest.raises(TypeError, match=msg):
+            base <= Period("2011", freq="4M")
+
+        with pytest.raises(TypeError, match=msg):
+            Period("2011", freq="4M") >= base
+
+        idx = PeriodIndex(["2011", "2012", "2013", "2014"], freq="4M")
+        rev_msg = r"Invalid comparison between dtype=period\[4M\] and PeriodArray"
+        idx_msg = rev_msg if box_with_array in [tm.to_array, pd.array] else msg
+        with pytest.raises(TypeError, match=idx_msg):
+            base <= idx
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_pi_cmp_nat(self, freq):
+        idx1 = PeriodIndex(["2011-01", "2011-02", "NaT", "2011-05"], freq=freq)
+        per = idx1[1]
+
+        result = idx1 > per
+        exp = np.array([False, False, False, True])
+        tm.assert_numpy_array_equal(result, exp)
+        result = per < idx1
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 == pd.NaT
+        exp = np.array([False, False, False, False])
+        tm.assert_numpy_array_equal(result, exp)
+        result = pd.NaT == idx1
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 != pd.NaT
+        exp = np.array([True, True, True, True])
+        tm.assert_numpy_array_equal(result, exp)
+        result = pd.NaT != idx1
+        tm.assert_numpy_array_equal(result, exp)
+
+        idx2 = PeriodIndex(["2011-02", "2011-01", "2011-04", "NaT"], freq=freq)
+        result = idx1 < idx2
+        exp = np.array([True, False, False, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 == idx2
+        exp = np.array([False, False, False, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 != idx2
+        exp = np.array([True, True, True, True])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 == idx1
+        exp = np.array([True, True, False, True])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = idx1 != idx1
+        exp = np.array([False, False, True, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_pi_cmp_nat_mismatched_freq_raises(self, freq):
+        idx1 = PeriodIndex(["2011-01", "2011-02", "NaT", "2011-05"], freq=freq)
+
+        diff = PeriodIndex(["2011-02", "2011-01", "2011-04", "NaT"], freq="4M")
+        msg = rf"Invalid comparison between dtype=period\[{freq}\] and PeriodArray"
+        with pytest.raises(TypeError, match=msg):
+            idx1 > diff
+
+        result = idx1 == diff
+        expected = np.array([False, False, False, False], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    # TODO: De-duplicate with test_pi_cmp_nat
+    @pytest.mark.parametrize("dtype", [object, None])
+    def test_comp_nat(self, dtype):
+        left = PeriodIndex([Period("2011-01-01"), pd.NaT, Period("2011-01-03")])
+        right = PeriodIndex([pd.NaT, pd.NaT, Period("2011-01-03")])
+
+        if dtype is not None:
+            left = left.astype(dtype)
+            right = right.astype(dtype)
+
+        result = left == right
+        expected = np.array([False, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = left != right
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(left == pd.NaT, expected)
+        tm.assert_numpy_array_equal(pd.NaT == right, expected)
+
+        expected = np.array([True, True, True])
+        tm.assert_numpy_array_equal(left != pd.NaT, expected)
+        tm.assert_numpy_array_equal(pd.NaT != left, expected)
+
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(left < pd.NaT, expected)
+        tm.assert_numpy_array_equal(pd.NaT > left, expected)
+
+
+class TestPeriodSeriesComparisons:
+    def test_cmp_series_period_series_mixed_freq(self):
+        # GH#13200
+        base = Series(
+            [
+                Period("2011", freq="Y"),
+                Period("2011-02", freq="M"),
+                Period("2013", freq="Y"),
+                Period("2011-04", freq="M"),
+            ]
+        )
+
+        ser = Series(
+            [
+                Period("2012", freq="Y"),
+                Period("2011-01", freq="M"),
+                Period("2013", freq="Y"),
+                Period("2011-05", freq="M"),
+            ]
+        )
+
+        exp = Series([False, False, True, False])
+        tm.assert_series_equal(base == ser, exp)
+
+        exp = Series([True, True, False, True])
+        tm.assert_series_equal(base != ser, exp)
+
+        exp = Series([False, True, False, False])
+        tm.assert_series_equal(base > ser, exp)
+
+        exp = Series([True, False, False, True])
+        tm.assert_series_equal(base < ser, exp)
+
+        exp = Series([False, True, True, False])
+        tm.assert_series_equal(base >= ser, exp)
+
+        exp = Series([True, False, True, True])
+        tm.assert_series_equal(base <= ser, exp)
+
+
+class TestPeriodIndexSeriesComparisonConsistency:
+    """Test PeriodIndex and Period Series Ops consistency"""
+
+    # TODO: needs parametrization+de-duplication
+
+    def _check(self, values, func, expected):
+        # Test PeriodIndex and Period Series Ops consistency
+
+        idx = PeriodIndex(values)
+        result = func(idx)
+
+        # check that we don't pass an unwanted type to tm.assert_equal
+        assert isinstance(expected, (pd.Index, np.ndarray))
+        tm.assert_equal(result, expected)
+
+        s = Series(values)
+        result = func(s)
+
+        exp = Series(expected, name=values.name)
+        tm.assert_series_equal(result, exp)
+
+    def test_pi_comp_period(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+        per = idx[2]
+
+        f = lambda x: x == per
+        exp = np.array([False, False, True, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: per == x
+        self._check(idx, f, exp)
+
+        f = lambda x: x != per
+        exp = np.array([True, True, False, True], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: per != x
+        self._check(idx, f, exp)
+
+        f = lambda x: per >= x
+        exp = np.array([True, True, True, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+        f = lambda x: x > per
+        exp = np.array([False, False, False, True], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+        f = lambda x: per >= x
+        exp = np.array([True, True, True, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+    def test_pi_comp_period_nat(self):
+        idx = PeriodIndex(
+            ["2011-01", "NaT", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+        per = idx[2]
+
+        f = lambda x: x == per
+        exp = np.array([False, False, True, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: per == x
+        self._check(idx, f, exp)
+
+        f = lambda x: x == pd.NaT
+        exp = np.array([False, False, False, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: pd.NaT == x
+        self._check(idx, f, exp)
+
+        f = lambda x: x != per
+        exp = np.array([True, True, False, True], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: per != x
+        self._check(idx, f, exp)
+
+        f = lambda x: x != pd.NaT
+        exp = np.array([True, True, True, True], dtype=np.bool_)
+        self._check(idx, f, exp)
+        f = lambda x: pd.NaT != x
+        self._check(idx, f, exp)
+
+        f = lambda x: per >= x
+        exp = np.array([True, False, True, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+        f = lambda x: x < per
+        exp = np.array([True, False, False, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+        f = lambda x: x > pd.NaT
+        exp = np.array([False, False, False, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+        f = lambda x: pd.NaT >= x
+        exp = np.array([False, False, False, False], dtype=np.bool_)
+        self._check(idx, f, exp)
+
+
+# ------------------------------------------------------------------
+# Arithmetic
+
+
+class TestPeriodFrameArithmetic:
+    def test_ops_frame_period(self):
+        # GH#13043
+        df = pd.DataFrame(
+            {
+                "A": [Period("2015-01", freq="M"), Period("2015-02", freq="M")],
+                "B": [Period("2014-01", freq="M"), Period("2014-02", freq="M")],
+            }
+        )
+        assert df["A"].dtype == "Period[M]"
+        assert df["B"].dtype == "Period[M]"
+
+        p = Period("2015-03", freq="M")
+        off = p.freq
+        # dtype will be object because of original dtype
+        exp = pd.DataFrame(
+            {
+                "A": np.array([2 * off, 1 * off], dtype=object),
+                "B": np.array([14 * off, 13 * off], dtype=object),
+            }
+        )
+        tm.assert_frame_equal(p - df, exp)
+        tm.assert_frame_equal(df - p, -1 * exp)
+
+        df2 = pd.DataFrame(
+            {
+                "A": [Period("2015-05", freq="M"), Period("2015-06", freq="M")],
+                "B": [Period("2015-05", freq="M"), Period("2015-06", freq="M")],
+            }
+        )
+        assert df2["A"].dtype == "Period[M]"
+        assert df2["B"].dtype == "Period[M]"
+
+        exp = pd.DataFrame(
+            {
+                "A": np.array([4 * off, 4 * off], dtype=object),
+                "B": np.array([16 * off, 16 * off], dtype=object),
+            }
+        )
+        tm.assert_frame_equal(df2 - df, exp)
+        tm.assert_frame_equal(df - df2, -1 * exp)
+
+
+class TestPeriodIndexArithmetic:
+    # ---------------------------------------------------------------
+    # __add__/__sub__ with PeriodIndex
+    # PeriodIndex + other is defined for integers and timedelta-like others
+    # PeriodIndex - other is defined for integers, timedelta-like others,
+    #   and PeriodIndex (with matching freq)
+
+    def test_parr_add_iadd_parr_raises(self, box_with_array):
+        rng = period_range("1/1/2000", freq="D", periods=5)
+        other = period_range("1/6/2000", freq="D", periods=5)
+        # TODO: parametrize over boxes for other?
+
+        rng = tm.box_expected(rng, box_with_array)
+        # An earlier implementation of PeriodIndex addition performed
+        # a set operation (union).  This has since been changed to
+        # raise a TypeError. See GH#14164 and GH#13077 for historical
+        # reference.
+        msg = r"unsupported operand type\(s\) for \+: .* and .*"
+        with pytest.raises(TypeError, match=msg):
+            rng + other
+
+        with pytest.raises(TypeError, match=msg):
+            rng += other
+
+    def test_pi_sub_isub_pi(self):
+        # GH#20049
+        # For historical reference see GH#14164, GH#13077.
+        # PeriodIndex subtraction originally performed set difference,
+        # then changed to raise TypeError before being implemented in GH#20049
+        rng = period_range("1/1/2000", freq="D", periods=5)
+        other = period_range("1/6/2000", freq="D", periods=5)
+
+        off = rng.freq
+        expected = pd.Index([-5 * off] * 5)
+        result = rng - other
+        tm.assert_index_equal(result, expected)
+
+        rng -= other
+        tm.assert_index_equal(rng, expected)
+
+    def test_pi_sub_pi_with_nat(self):
+        rng = period_range("1/1/2000", freq="D", periods=5)
+        other = rng[1:].insert(0, pd.NaT)
+        assert other[1:].equals(rng[1:])
+
+        result = rng - other
+        off = rng.freq
+        expected = pd.Index([pd.NaT, 0 * off, 0 * off, 0 * off, 0 * off])
+        tm.assert_index_equal(result, expected)
+
+    def test_parr_sub_pi_mismatched_freq(self, box_with_array, box_with_array2):
+        rng = period_range("1/1/2000", freq="D", periods=5)
+        other = period_range("1/6/2000", freq="h", periods=5)
+
+        rng = tm.box_expected(rng, box_with_array)
+        other = tm.box_expected(other, box_with_array2)
+        msg = r"Input has different freq=[hD] from PeriodArray\(freq=[Dh]\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng - other
+
+    @pytest.mark.parametrize("n", [1, 2, 3, 4])
+    def test_sub_n_gt_1_ticks(self, tick_classes, n):
+        # GH 23878
+        p1_d = "19910905"
+        p2_d = "19920406"
+        p1 = PeriodIndex([p1_d], freq=tick_classes(n))
+        p2 = PeriodIndex([p2_d], freq=tick_classes(n))
+
+        expected = PeriodIndex([p2_d], freq=p2.freq.base) - PeriodIndex(
+            [p1_d], freq=p1.freq.base
+        )
+
+        tm.assert_index_equal((p2 - p1), expected)
+
+    @pytest.mark.parametrize("n", [1, 2, 3, 4])
+    @pytest.mark.parametrize(
+        "offset, kwd_name",
+        [
+            (pd.offsets.YearEnd, "month"),
+            (pd.offsets.QuarterEnd, "startingMonth"),
+            (pd.offsets.MonthEnd, None),
+            (pd.offsets.Week, "weekday"),
+        ],
+    )
+    def test_sub_n_gt_1_offsets(self, offset, kwd_name, n):
+        # GH 23878
+        kwds = {kwd_name: 3} if kwd_name is not None else {}
+        p1_d = "19910905"
+        p2_d = "19920406"
+        freq = offset(n, normalize=False, **kwds)
+        p1 = PeriodIndex([p1_d], freq=freq)
+        p2 = PeriodIndex([p2_d], freq=freq)
+
+        result = p2 - p1
+        expected = PeriodIndex([p2_d], freq=freq.base) - PeriodIndex(
+            [p1_d], freq=freq.base
+        )
+
+        tm.assert_index_equal(result, expected)
+
+    # -------------------------------------------------------------
+    # Invalid Operations
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            # datetime scalars
+            Timestamp("2016-01-01"),
+            Timestamp("2016-01-01").to_pydatetime(),
+            Timestamp("2016-01-01").to_datetime64(),
+            # datetime-like arrays
+            pd.date_range("2016-01-01", periods=3, freq="h"),
+            pd.date_range("2016-01-01", periods=3, tz="Europe/Brussels"),
+            pd.date_range("2016-01-01", periods=3, freq="s")._data,
+            pd.date_range("2016-01-01", periods=3, tz="Asia/Tokyo")._data,
+            # Miscellaneous invalid types
+            3.14,
+            np.array([2.0, 3.0, 4.0]),
+        ],
+    )
+    def test_parr_add_sub_invalid(self, other, box_with_array):
+        # GH#23215
+        rng = period_range("1/1/2000", freq="D", periods=3)
+        rng = tm.box_expected(rng, box_with_array)
+
+        msg = "|".join(
+            [
+                r"(:?cannot add PeriodArray and .*)",
+                r"(:?cannot subtract .* from (:?a\s)?.*)",
+                r"(:?unsupported operand type\(s\) for \+: .* and .*)",
+                r"unsupported operand type\(s\) for [+-]: .* and .*",
+            ]
+        )
+        assert_invalid_addsub_type(rng, other, msg)
+        with pytest.raises(TypeError, match=msg):
+            rng + other
+        with pytest.raises(TypeError, match=msg):
+            other + rng
+        with pytest.raises(TypeError, match=msg):
+            rng - other
+        with pytest.raises(TypeError, match=msg):
+            other - rng
+
+    # -----------------------------------------------------------------
+    # __add__/__sub__ with ndarray[datetime64] and ndarray[timedelta64]
+
+    def test_pi_add_sub_td64_array_non_tick_raises(self):
+        rng = period_range("1/1/2000", freq="Q", periods=3)
+        tdi = TimedeltaIndex(["-1 Day", "-1 Day", "-1 Day"])
+        tdarr = tdi.values
+
+        msg = r"Cannot add or subtract timedelta64\[ns\] dtype from period\[Q-DEC\]"
+        with pytest.raises(TypeError, match=msg):
+            rng + tdarr
+        with pytest.raises(TypeError, match=msg):
+            tdarr + rng
+
+        with pytest.raises(TypeError, match=msg):
+            rng - tdarr
+        msg = r"cannot subtract PeriodArray from TimedeltaArray"
+        with pytest.raises(TypeError, match=msg):
+            tdarr - rng
+
+    def test_pi_add_sub_td64_array_tick(self):
+        # PeriodIndex + Timedelta-like is allowed only with
+        #   tick-like frequencies
+        rng = period_range("1/1/2000", freq="90D", periods=3)
+        tdi = TimedeltaIndex(["-1 Day", "-1 Day", "-1 Day"])
+        tdarr = tdi.values
+
+        expected = period_range("12/31/1999", freq="90D", periods=3)
+        result = rng + tdi
+        tm.assert_index_equal(result, expected)
+        result = rng + tdarr
+        tm.assert_index_equal(result, expected)
+        result = tdi + rng
+        tm.assert_index_equal(result, expected)
+        result = tdarr + rng
+        tm.assert_index_equal(result, expected)
+
+        expected = period_range("1/2/2000", freq="90D", periods=3)
+
+        result = rng - tdi
+        tm.assert_index_equal(result, expected)
+        result = rng - tdarr
+        tm.assert_index_equal(result, expected)
+
+        msg = r"cannot subtract .* from .*"
+        with pytest.raises(TypeError, match=msg):
+            tdarr - rng
+
+        with pytest.raises(TypeError, match=msg):
+            tdi - rng
+
+    @pytest.mark.parametrize("pi_freq", ["D", "W", "Q", "h"])
+    @pytest.mark.parametrize("tdi_freq", [None, "h"])
+    def test_parr_sub_td64array(self, box_with_array, tdi_freq, pi_freq):
+        box = box_with_array
+        xbox = box if box not in [pd.array, tm.to_array] else pd.Index
+
+        tdi = TimedeltaIndex(["1 hours", "2 hours"], freq=tdi_freq)
+        dti = Timestamp("2018-03-07 17:16:40") + tdi
+        pi = dti.to_period(pi_freq)
+
+        # TODO: parametrize over box for pi?
+        td64obj = tm.box_expected(tdi, box)
+
+        if pi_freq == "h":
+            result = pi - td64obj
+            expected = (pi.to_timestamp("s") - tdi).to_period(pi_freq)
+            expected = tm.box_expected(expected, xbox)
+            tm.assert_equal(result, expected)
+
+            # Subtract from scalar
+            result = pi[0] - td64obj
+            expected = (pi[0].to_timestamp("s") - tdi).to_period(pi_freq)
+            expected = tm.box_expected(expected, box)
+            tm.assert_equal(result, expected)
+
+        elif pi_freq == "D":
+            # Tick, but non-compatible
+            msg = (
+                "Cannot add/subtract timedelta-like from PeriodArray that is "
+                "not an integer multiple of the PeriodArray's freq."
+            )
+            with pytest.raises(IncompatibleFrequency, match=msg):
+                pi - td64obj
+
+            with pytest.raises(IncompatibleFrequency, match=msg):
+                pi[0] - td64obj
+
+        else:
+            # With non-Tick freq, we could not add timedelta64 array regardless
+            #  of what its resolution is
+            msg = "Cannot add or subtract timedelta64"
+            with pytest.raises(TypeError, match=msg):
+                pi - td64obj
+            with pytest.raises(TypeError, match=msg):
+                pi[0] - td64obj
+
+    # -----------------------------------------------------------------
+    # operations with array/Index of DateOffset objects
+
+    @pytest.mark.parametrize("box", [np.array, pd.Index])
+    def test_pi_add_offset_array(self, performance_warning, box):
+        # GH#18849
+        pi = PeriodIndex([Period("2015Q1"), Period("2016Q2")])
+        offs = box(
+            [
+                pd.offsets.QuarterEnd(n=1, startingMonth=12),
+                pd.offsets.QuarterEnd(n=-2, startingMonth=12),
+            ]
+        )
+        expected = PeriodIndex([Period("2015Q2"), Period("2015Q4")]).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = pi + offs
+        tm.assert_index_equal(res, expected)
+
+        with tm.assert_produces_warning(performance_warning):
+            res2 = offs + pi
+        tm.assert_index_equal(res2, expected)
+
+        unanchored = np.array([pd.offsets.Hour(n=1), pd.offsets.Minute(n=-2)])
+        # addition/subtraction ops with incompatible offsets should issue
+        # a PerformanceWarning and _then_ raise a TypeError.
+        msg = r"Input cannot be converted to Period\(freq=Q-DEC\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                pi + unanchored
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                unanchored + pi
+
+    @pytest.mark.parametrize("box", [np.array, pd.Index])
+    def test_pi_sub_offset_array(self, performance_warning, box):
+        # GH#18824
+        pi = PeriodIndex([Period("2015Q1"), Period("2016Q2")])
+        other = box(
+            [
+                pd.offsets.QuarterEnd(n=1, startingMonth=12),
+                pd.offsets.QuarterEnd(n=-2, startingMonth=12),
+            ]
+        )
+
+        expected = PeriodIndex([pi[n] - other[n] for n in range(len(pi))])
+        expected = expected.astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = pi - other
+        tm.assert_index_equal(res, expected)
+
+        anchored = box([pd.offsets.MonthEnd(), pd.offsets.Day(n=2)])
+
+        # addition/subtraction ops with anchored offsets should issue
+        # a PerformanceWarning and _then_ raise a TypeError.
+        msg = r"Input has different freq=-1M from Period\(freq=Q-DEC\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                pi - anchored
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                anchored - pi
+
+    def test_pi_add_iadd_int(self, one):
+        # Variants of `one` for #19012
+        rng = period_range("2000-01-01 09:00", freq="h", periods=10)
+        result = rng + one
+        expected = period_range("2000-01-01 10:00", freq="h", periods=10)
+        tm.assert_index_equal(result, expected)
+        rng += one
+        tm.assert_index_equal(rng, expected)
+
+    def test_pi_sub_isub_int(self, one):
+        """
+        PeriodIndex.__sub__ and __isub__ with several representations of
+        the integer 1, e.g. int, np.int64, np.uint8, ...
+        """
+        rng = period_range("2000-01-01 09:00", freq="h", periods=10)
+        result = rng - one
+        expected = period_range("2000-01-01 08:00", freq="h", periods=10)
+        tm.assert_index_equal(result, expected)
+        rng -= one
+        tm.assert_index_equal(rng, expected)
+
+    @pytest.mark.parametrize("five", [5, np.array(5, dtype=np.int64)])
+    def test_pi_sub_intlike(self, five):
+        rng = period_range("2007-01", periods=50)
+
+        result = rng - five
+        exp = rng + (-five)
+        tm.assert_index_equal(result, exp)
+
+    def test_pi_add_sub_int_array_freqn_gt1(self):
+        # GH#47209 test adding array of ints when freq.n > 1 matches
+        #  scalar behavior
+        pi = period_range("2016-01-01", periods=10, freq="2D")
+        arr = np.arange(10)
+        result = pi + arr
+        expected = pd.Index([x + y for x, y in zip(pi, arr)])
+        tm.assert_index_equal(result, expected)
+
+        result = pi - arr
+        expected = pd.Index([x - y for x, y in zip(pi, arr)])
+        tm.assert_index_equal(result, expected)
+
+    def test_pi_sub_isub_offset(self):
+        # offset
+        # DateOffset
+        rng = period_range("2014", "2024", freq="Y")
+        result = rng - pd.offsets.YearEnd(5)
+        expected = period_range("2009", "2019", freq="Y")
+        tm.assert_index_equal(result, expected)
+        rng -= pd.offsets.YearEnd(5)
+        tm.assert_index_equal(rng, expected)
+
+        rng = period_range("2014-01", "2016-12", freq="M")
+        result = rng - pd.offsets.MonthEnd(5)
+        expected = period_range("2013-08", "2016-07", freq="M")
+        tm.assert_index_equal(result, expected)
+
+        rng -= pd.offsets.MonthEnd(5)
+        tm.assert_index_equal(rng, expected)
+
+    @pytest.mark.parametrize("transpose", [True, False])
+    def test_pi_add_offset_n_gt1(self, box_with_array, transpose):
+        # GH#23215
+        # add offset to PeriodIndex with freq.n > 1
+
+        per = Period("2016-01", freq="2M")
+        pi = PeriodIndex([per])
+
+        expected = PeriodIndex(["2016-03"], freq="2M")
+
+        pi = tm.box_expected(pi, box_with_array, transpose=transpose)
+        expected = tm.box_expected(expected, box_with_array, transpose=transpose)
+
+        result = pi + per.freq
+        tm.assert_equal(result, expected)
+
+        result = per.freq + pi
+        tm.assert_equal(result, expected)
+
+    def test_pi_add_offset_n_gt1_not_divisible(self, box_with_array):
+        # GH#23215
+        # PeriodIndex with freq.n > 1 add offset with offset.n % freq.n != 0
+        pi = PeriodIndex(["2016-01"], freq="2M")
+        expected = PeriodIndex(["2016-04"], freq="2M")
+
+        pi = tm.box_expected(pi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = pi + to_offset("3ME")
+        tm.assert_equal(result, expected)
+
+        result = to_offset("3ME") + pi
+        tm.assert_equal(result, expected)
+
+    # ---------------------------------------------------------------
+    # __add__/__sub__ with integer arrays
+
+    @pytest.mark.parametrize("int_holder", [np.array, pd.Index])
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_pi_add_intarray(self, int_holder, op):
+        # GH#19959
+        pi = PeriodIndex([Period("2015Q1"), Period("NaT")])
+        other = int_holder([4, -1])
+
+        result = op(pi, other)
+        expected = PeriodIndex([Period("2016Q1"), Period("NaT")])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("int_holder", [np.array, pd.Index])
+    def test_pi_sub_intarray(self, int_holder):
+        # GH#19959
+        pi = PeriodIndex([Period("2015Q1"), Period("NaT")])
+        other = int_holder([4, -1])
+
+        result = pi - other
+        expected = PeriodIndex([Period("2014Q1"), Period("NaT")])
+        tm.assert_index_equal(result, expected)
+
+        msg = r"bad operand type for unary -: 'PeriodArray'"
+        with pytest.raises(TypeError, match=msg):
+            other - pi
+
+    # ---------------------------------------------------------------
+    # Timedelta-like (timedelta, timedelta64, Timedelta, Tick)
+    # TODO: Some of these are misnomers because of non-Tick DateOffsets
+
+    def test_parr_add_timedeltalike_minute_gt1(self, three_days, box_with_array):
+        # GH#23031 adding a time-delta-like offset to a PeriodArray that has
+        # minute frequency with n != 1.  A more general case is tested below
+        # in test_pi_add_timedeltalike_tick_gt1, but here we write out the
+        # expected result more explicitly.
+        other = three_days
+        rng = period_range("2014-05-01", periods=3, freq="2D")
+        rng = tm.box_expected(rng, box_with_array)
+
+        expected = PeriodIndex(["2014-05-04", "2014-05-06", "2014-05-08"], freq="2D")
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = rng + other
+        tm.assert_equal(result, expected)
+
+        result = other + rng
+        tm.assert_equal(result, expected)
+
+        # subtraction
+        expected = PeriodIndex(["2014-04-28", "2014-04-30", "2014-05-02"], freq="2D")
+        expected = tm.box_expected(expected, box_with_array)
+        result = rng - other
+        tm.assert_equal(result, expected)
+
+        msg = "|".join(
+            [
+                r"bad operand type for unary -: 'PeriodArray'",
+                r"cannot subtract PeriodArray from timedelta64\[[hD]\]",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            other - rng
+
+    @pytest.mark.parametrize("freqstr", ["5ns", "5us", "5ms", "5s", "5min", "5h", "5D"])
+    def test_parr_add_timedeltalike_tick_gt1(self, three_days, freqstr, box_with_array):
+        # GH#23031 adding a time-delta-like offset to a PeriodArray that has
+        # tick-like frequency with n != 1
+        other = three_days
+        rng = period_range("2014-05-01", periods=6, freq=freqstr)
+        first = rng[0]
+        rng = tm.box_expected(rng, box_with_array)
+
+        expected = period_range(first + other, periods=6, freq=freqstr)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = rng + other
+        tm.assert_equal(result, expected)
+
+        result = other + rng
+        tm.assert_equal(result, expected)
+
+        # subtraction
+        expected = period_range(first - other, periods=6, freq=freqstr)
+        expected = tm.box_expected(expected, box_with_array)
+        result = rng - other
+        tm.assert_equal(result, expected)
+        msg = "|".join(
+            [
+                r"bad operand type for unary -: 'PeriodArray'",
+                r"cannot subtract PeriodArray from timedelta64\[[hD]\]",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            other - rng
+
+    def test_pi_add_iadd_timedeltalike_daily(self, three_days):
+        # Tick
+        other = three_days
+        rng = period_range("2014-05-01", "2014-05-15", freq="D")
+        expected = period_range("2014-05-04", "2014-05-18", freq="D")
+
+        result = rng + other
+        tm.assert_index_equal(result, expected)
+
+        rng += other
+        tm.assert_index_equal(rng, expected)
+
+    def test_pi_sub_isub_timedeltalike_daily(self, three_days):
+        # Tick-like 3 Days
+        other = three_days
+        rng = period_range("2014-05-01", "2014-05-15", freq="D")
+        expected = period_range("2014-04-28", "2014-05-12", freq="D")
+
+        result = rng - other
+        tm.assert_index_equal(result, expected)
+
+        rng -= other
+        tm.assert_index_equal(rng, expected)
+
+    def test_parr_add_sub_timedeltalike_freq_mismatch_daily(
+        self, not_daily, box_with_array
+    ):
+        other = not_daily
+        rng = period_range("2014-05-01", "2014-05-15", freq="D")
+        rng = tm.box_expected(rng, box_with_array)
+
+        msg = "|".join(
+            [
+                # non-timedelta-like DateOffset
+                "Input has different freq(=.+)? from Period.*?\\(freq=D\\)",
+                # timedelta/td64/Timedelta but not a multiple of 24H
+                "Cannot add/subtract timedelta-like from PeriodArray that is "
+                "not an integer multiple of the PeriodArray's freq.",
+            ]
+        )
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng + other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng += other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng - other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng -= other
+
+    def test_pi_add_iadd_timedeltalike_hourly(self, two_hours):
+        other = two_hours
+        rng = period_range("2014-01-01 10:00", "2014-01-05 10:00", freq="h")
+        expected = period_range("2014-01-01 12:00", "2014-01-05 12:00", freq="h")
+
+        result = rng + other
+        tm.assert_index_equal(result, expected)
+
+        rng += other
+        tm.assert_index_equal(rng, expected)
+
+    def test_parr_add_timedeltalike_mismatched_freq_hourly(
+        self, not_hourly, box_with_array
+    ):
+        other = not_hourly
+        rng = period_range("2014-01-01 10:00", "2014-01-05 10:00", freq="h")
+        rng = tm.box_expected(rng, box_with_array)
+        msg = "|".join(
+            [
+                # non-timedelta-like DateOffset
+                "Input has different freq(=.+)? from Period.*?\\(freq=h\\)",
+                # timedelta/td64/Timedelta but not a multiple of 24H
+                "Cannot add/subtract timedelta-like from PeriodArray that is "
+                "not an integer multiple of the PeriodArray's freq.",
+            ]
+        )
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng + other
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng += other
+
+    def test_pi_sub_isub_timedeltalike_hourly(self, two_hours):
+        other = two_hours
+        rng = period_range("2014-01-01 10:00", "2014-01-05 10:00", freq="h")
+        expected = period_range("2014-01-01 08:00", "2014-01-05 08:00", freq="h")
+
+        result = rng - other
+        tm.assert_index_equal(result, expected)
+
+        rng -= other
+        tm.assert_index_equal(rng, expected)
+
+    def test_add_iadd_timedeltalike_annual(self):
+        # offset
+        # DateOffset
+        rng = period_range("2014", "2024", freq="Y")
+        result = rng + pd.offsets.YearEnd(5)
+        expected = period_range("2019", "2029", freq="Y")
+        tm.assert_index_equal(result, expected)
+        rng += pd.offsets.YearEnd(5)
+        tm.assert_index_equal(rng, expected)
+
+    def test_pi_add_sub_timedeltalike_freq_mismatch_annual(self, mismatched_freq):
+        other = mismatched_freq
+        rng = period_range("2014", "2024", freq="Y")
+        msg = "Input has different freq(=.+)? from Period.*?\\(freq=Y-DEC\\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng + other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng += other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng - other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng -= other
+
+    def test_pi_add_iadd_timedeltalike_M(self):
+        rng = period_range("2014-01", "2016-12", freq="M")
+        expected = period_range("2014-06", "2017-05", freq="M")
+
+        result = rng + pd.offsets.MonthEnd(5)
+        tm.assert_index_equal(result, expected)
+
+        rng += pd.offsets.MonthEnd(5)
+        tm.assert_index_equal(rng, expected)
+
+    def test_pi_add_sub_timedeltalike_freq_mismatch_monthly(self, mismatched_freq):
+        other = mismatched_freq
+        rng = period_range("2014-01", "2016-12", freq="M")
+        msg = "Input has different freq(=.+)? from Period.*?\\(freq=M\\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng + other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng += other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng - other
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            rng -= other
+
+    @pytest.mark.parametrize("transpose", [True, False])
+    def test_parr_add_sub_td64_nat(self, box_with_array, transpose):
+        # GH#23320 special handling for timedelta64("NaT")
+        pi = period_range("1994-04-01", periods=9, freq="19D")
+        other = np.timedelta64("NaT")
+        expected = PeriodIndex(["NaT"] * 9, freq="19D")
+
+        obj = tm.box_expected(pi, box_with_array, transpose=transpose)
+        expected = tm.box_expected(expected, box_with_array, transpose=transpose)
+
+        result = obj + other
+        tm.assert_equal(result, expected)
+        result = other + obj
+        tm.assert_equal(result, expected)
+        result = obj - other
+        tm.assert_equal(result, expected)
+        msg = r"cannot subtract .* from .*"
+        with pytest.raises(TypeError, match=msg):
+            other - obj
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            np.array(["NaT"] * 9, dtype="m8[ns]"),
+            TimedeltaArray._from_sequence(["NaT"] * 9, dtype="m8[ns]"),
+        ],
+    )
+    def test_parr_add_sub_tdt64_nat_array(self, box_with_array, other):
+        pi = period_range("1994-04-01", periods=9, freq="19D")
+        expected = PeriodIndex(["NaT"] * 9, freq="19D")
+
+        obj = tm.box_expected(pi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = obj + other
+        tm.assert_equal(result, expected)
+        result = other + obj
+        tm.assert_equal(result, expected)
+        result = obj - other
+        tm.assert_equal(result, expected)
+        msg = r"cannot subtract .* from .*"
+        with pytest.raises(TypeError, match=msg):
+            other - obj
+
+        # some but not *all* NaT
+        other = other.copy()
+        other[0] = np.timedelta64(0, "ns")
+        expected = PeriodIndex([pi[0]] + ["NaT"] * 8, freq="19D")
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = obj + other
+        tm.assert_equal(result, expected)
+        result = other + obj
+        tm.assert_equal(result, expected)
+        result = obj - other
+        tm.assert_equal(result, expected)
+        with pytest.raises(TypeError, match=msg):
+            other - obj
+
+    # ---------------------------------------------------------------
+    # Unsorted
+
+    def test_parr_add_sub_index(self):
+        # Check that PeriodArray defers to Index on arithmetic ops
+        pi = period_range("2000-12-31", periods=3)
+        parr = pi.array
+
+        result = parr - pi
+        expected = pi - pi
+        tm.assert_index_equal(result, expected)
+
+    def test_parr_add_sub_object_array(self, performance_warning):
+        pi = period_range("2000-12-31", periods=3, freq="D")
+        parr = pi.array
+
+        other = np.array([Timedelta(days=1), pd.offsets.Day(2), 3])
+
+        with tm.assert_produces_warning(performance_warning):
+            result = parr + other
+
+        expected = PeriodIndex(
+            ["2001-01-01", "2001-01-03", "2001-01-05"], freq="D"
+        )._data.astype(object)
+        tm.assert_equal(result, expected)
+
+        with tm.assert_produces_warning(performance_warning):
+            result = parr - other
+
+        expected = PeriodIndex(["2000-12-30"] * 3, freq="D")._data.astype(object)
+        tm.assert_equal(result, expected)
+
+    def test_period_add_timestamp_raises(self, box_with_array):
+        # GH#17983
+        ts = Timestamp("2017")
+        per = Period("2017", freq="M")
+
+        arr = pd.Index([per], dtype="Period[M]")
+        arr = tm.box_expected(arr, box_with_array)
+
+        msg = "cannot add PeriodArray and Timestamp"
+        with pytest.raises(TypeError, match=msg):
+            arr + ts
+        with pytest.raises(TypeError, match=msg):
+            ts + arr
+        if box_with_array is pd.DataFrame:
+            # TODO: before implementing resolution-inference we got the same
+            #  message with DataFrame and non-DataFrame.  Why did that change?
+            msg = "cannot add PeriodArray and Timestamp"
+        else:
+            msg = "cannot add PeriodArray and DatetimeArray"
+        with pytest.raises(TypeError, match=msg):
+            arr + Series([ts])
+        with pytest.raises(TypeError, match=msg):
+            Series([ts]) + arr
+        with pytest.raises(TypeError, match=msg):
+            arr + pd.Index([ts])
+        with pytest.raises(TypeError, match=msg):
+            pd.Index([ts]) + arr
+
+        if box_with_array is pd.DataFrame:
+            msg = "cannot add PeriodArray and DatetimeArray"
+        else:
+            msg = r"unsupported operand type\(s\) for \+: 'Period' and 'DatetimeArray"
+        with pytest.raises(TypeError, match=msg):
+            arr + pd.DataFrame([ts])
+        if box_with_array is pd.DataFrame:
+            msg = "cannot add PeriodArray and DatetimeArray"
+        else:
+            msg = r"unsupported operand type\(s\) for \+: 'DatetimeArray' and 'Period'"
+        with pytest.raises(TypeError, match=msg):
+            pd.DataFrame([ts]) + arr
+
+
+class TestPeriodSeriesArithmetic:
+    def test_parr_add_timedeltalike_scalar(self, three_days, box_with_array):
+        # GH#13043
+        ser = Series(
+            [Period("2015-01-01", freq="D"), Period("2015-01-02", freq="D")],
+            name="xxx",
+        )
+        assert ser.dtype == "Period[D]"
+
+        expected = Series(
+            [Period("2015-01-04", freq="D"), Period("2015-01-05", freq="D")],
+            name="xxx",
+        )
+
+        obj = tm.box_expected(ser, box_with_array)
+        if box_with_array is pd.DataFrame:
+            assert (obj.dtypes == "Period[D]").all()
+
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = obj + three_days
+        tm.assert_equal(result, expected)
+
+        result = three_days + obj
+        tm.assert_equal(result, expected)
+
+    def test_ops_series_period(self):
+        # GH#13043
+        ser = Series(
+            [Period("2015-01-01", freq="D"), Period("2015-01-02", freq="D")],
+            name="xxx",
+        )
+        assert ser.dtype == "Period[D]"
+
+        per = Period("2015-01-10", freq="D")
+        off = per.freq
+        # dtype will be object because of original dtype
+        expected = Series([9 * off, 8 * off], name="xxx", dtype=object)
+        tm.assert_series_equal(per - ser, expected)
+        tm.assert_series_equal(ser - per, -1 * expected)
+
+        s2 = Series(
+            [Period("2015-01-05", freq="D"), Period("2015-01-04", freq="D")],
+            name="xxx",
+        )
+        assert s2.dtype == "Period[D]"
+
+        expected = Series([4 * off, 2 * off], name="xxx", dtype=object)
+        tm.assert_series_equal(s2 - ser, expected)
+        tm.assert_series_equal(ser - s2, -1 * expected)
+
+
+class TestPeriodIndexSeriesMethods:
+    """Test PeriodIndex and Period Series Ops consistency"""
+
+    def _check(self, values, func, expected):
+        idx = PeriodIndex(values)
+        result = func(idx)
+        tm.assert_equal(result, expected)
+
+        ser = Series(values)
+        result = func(ser)
+
+        exp = Series(expected, name=values.name)
+        tm.assert_series_equal(result, exp)
+
+    def test_pi_ops(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+
+        expected = PeriodIndex(
+            ["2011-03", "2011-04", "2011-05", "2011-06"], freq="M", name="idx"
+        )
+
+        self._check(idx, lambda x: x + 2, expected)
+        self._check(idx, lambda x: 2 + x, expected)
+
+        self._check(idx + 2, lambda x: x - 2, idx)
+
+        result = idx - Period("2011-01", freq="M")
+        off = idx.freq
+        exp = pd.Index([0 * off, 1 * off, 2 * off, 3 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        result = Period("2011-01", freq="M") - idx
+        exp = pd.Index([0 * off, -1 * off, -2 * off, -3 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+    @pytest.mark.parametrize("ng", ["str", 1.5])
+    @pytest.mark.parametrize(
+        "func",
+        [
+            lambda obj, ng: obj + ng,
+            lambda obj, ng: ng + obj,
+            lambda obj, ng: obj - ng,
+            lambda obj, ng: ng - obj,
+            lambda obj, ng: np.add(obj, ng),
+            lambda obj, ng: np.add(ng, obj),
+            lambda obj, ng: np.subtract(obj, ng),
+            lambda obj, ng: np.subtract(ng, obj),
+        ],
+    )
+    def test_parr_ops_errors(self, ng, func, box_with_array):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+        obj = tm.box_expected(idx, box_with_array)
+        msg = "|".join(
+            [
+                r"unsupported operand type\(s\)",
+                "can only concatenate",
+                r"must be str",
+                "object to str implicitly",
+            ]
+        )
+
+        with pytest.raises(TypeError, match=msg):
+            func(obj, ng)
+
+    def test_pi_ops_nat(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="M", name="idx"
+        )
+        expected = PeriodIndex(
+            ["2011-03", "2011-04", "NaT", "2011-06"], freq="M", name="idx"
+        )
+
+        self._check(idx, lambda x: x + 2, expected)
+        self._check(idx, lambda x: 2 + x, expected)
+        self._check(idx, lambda x: np.add(x, 2), expected)
+
+        self._check(idx + 2, lambda x: x - 2, idx)
+        self._check(idx + 2, lambda x: np.subtract(x, 2), idx)
+
+        # freq with mult
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="2M", name="idx"
+        )
+        expected = PeriodIndex(
+            ["2011-07", "2011-08", "NaT", "2011-10"], freq="2M", name="idx"
+        )
+
+        self._check(idx, lambda x: x + 3, expected)
+        self._check(idx, lambda x: 3 + x, expected)
+        self._check(idx, lambda x: np.add(x, 3), expected)
+
+        self._check(idx + 3, lambda x: x - 3, idx)
+        self._check(idx + 3, lambda x: np.subtract(x, 3), idx)
+
+    def test_pi_ops_array_int(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="M", name="idx"
+        )
+        f = lambda x: x + np.array([1, 2, 3, 4])
+        exp = PeriodIndex(
+            ["2011-02", "2011-04", "NaT", "2011-08"], freq="M", name="idx"
+        )
+        self._check(idx, f, exp)
+
+        f = lambda x: np.add(x, np.array([4, -1, 1, 2]))
+        exp = PeriodIndex(
+            ["2011-05", "2011-01", "NaT", "2011-06"], freq="M", name="idx"
+        )
+        self._check(idx, f, exp)
+
+        f = lambda x: x - np.array([1, 2, 3, 4])
+        exp = PeriodIndex(
+            ["2010-12", "2010-12", "NaT", "2010-12"], freq="M", name="idx"
+        )
+        self._check(idx, f, exp)
+
+        f = lambda x: np.subtract(x, np.array([3, 2, 3, -2]))
+        exp = PeriodIndex(
+            ["2010-10", "2010-12", "NaT", "2011-06"], freq="M", name="idx"
+        )
+        self._check(idx, f, exp)
+
+    def test_pi_ops_offset(self):
+        idx = PeriodIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01", "2011-04-01"],
+            freq="D",
+            name="idx",
+        )
+        f = lambda x: x + pd.offsets.Day()
+        exp = PeriodIndex(
+            ["2011-01-02", "2011-02-02", "2011-03-02", "2011-04-02"],
+            freq="D",
+            name="idx",
+        )
+        self._check(idx, f, exp)
+
+        f = lambda x: x + pd.offsets.Day(2)
+        exp = PeriodIndex(
+            ["2011-01-03", "2011-02-03", "2011-03-03", "2011-04-03"],
+            freq="D",
+            name="idx",
+        )
+        self._check(idx, f, exp)
+
+        f = lambda x: x - pd.offsets.Day(2)
+        exp = PeriodIndex(
+            ["2010-12-30", "2011-01-30", "2011-02-27", "2011-03-30"],
+            freq="D",
+            name="idx",
+        )
+        self._check(idx, f, exp)
+
+    def test_pi_offset_errors(self):
+        idx = PeriodIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01", "2011-04-01"],
+            freq="D",
+            name="idx",
+        )
+        ser = Series(idx)
+
+        msg = (
+            "Cannot add/subtract timedelta-like from PeriodArray that is not "
+            "an integer multiple of the PeriodArray's freq"
+        )
+        for obj in [idx, ser]:
+            with pytest.raises(IncompatibleFrequency, match=msg):
+                obj + pd.offsets.Hour(2)
+
+            with pytest.raises(IncompatibleFrequency, match=msg):
+                pd.offsets.Hour(2) + obj
+
+            with pytest.raises(IncompatibleFrequency, match=msg):
+                obj - pd.offsets.Hour(2)
+
+    def test_pi_sub_period(self):
+        # GH#13071
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+
+        result = idx - Period("2012-01", freq="M")
+        off = idx.freq
+        exp = pd.Index([-12 * off, -11 * off, -10 * off, -9 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        result = np.subtract(idx, Period("2012-01", freq="M"))
+        tm.assert_index_equal(result, exp)
+
+        result = Period("2012-01", freq="M") - idx
+        exp = pd.Index([12 * off, 11 * off, 10 * off, 9 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        result = np.subtract(Period("2012-01", freq="M"), idx)
+        tm.assert_index_equal(result, exp)
+
+        exp = TimedeltaIndex([np.nan, np.nan, np.nan, np.nan], name="idx")
+        result = idx - Period("NaT", freq="M")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == exp.freq
+
+        result = Period("NaT", freq="M") - idx
+        tm.assert_index_equal(result, exp)
+        assert result.freq == exp.freq
+
+    def test_pi_sub_pdnat(self):
+        # GH#13071, GH#19389
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="M", name="idx"
+        )
+        exp = TimedeltaIndex([pd.NaT] * 4, name="idx")
+        tm.assert_index_equal(pd.NaT - idx, exp)
+        tm.assert_index_equal(idx - pd.NaT, exp)
+
+    def test_pi_sub_period_nat(self):
+        # GH#13071
+        idx = PeriodIndex(
+            ["2011-01", "NaT", "2011-03", "2011-04"], freq="M", name="idx"
+        )
+
+        result = idx - Period("2012-01", freq="M")
+        off = idx.freq
+        exp = pd.Index([-12 * off, pd.NaT, -10 * off, -9 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        result = Period("2012-01", freq="M") - idx
+        exp = pd.Index([12 * off, pd.NaT, 10 * off, 9 * off], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        exp = TimedeltaIndex([np.nan, np.nan, np.nan, np.nan], name="idx")
+        tm.assert_index_equal(idx - Period("NaT", freq="M"), exp)
+        tm.assert_index_equal(Period("NaT", freq="M") - idx, exp)
diff --git a/pandas/pandas/tests/arithmetic/test_timedelta64.py b/pandas/pandas/tests/arithmetic/test_timedelta64.py
new file mode 100644
index 0000000000000..87e085fb22878
--- /dev/null
+++ b/pandas/pandas/tests/arithmetic/test_timedelta64.py
@@ -0,0 +1,2176 @@
+# Arithmetic tests for DataFrame/Series/Index/Array classes that should
+# behave identically.
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+from pandas.errors import OutOfBoundsDatetime
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    NaT,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    offsets,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import NumpyExtensionArray
+from pandas.tests.arithmetic.common import (
+    assert_invalid_addsub_type,
+    assert_invalid_comparison,
+    get_upcast_box,
+)
+
+
+def assert_dtype(obj, expected_dtype):
+    """
+    Helper to check the dtype for a Series, Index, or single-column DataFrame.
+    """
+    dtype = tm.get_dtype(obj)
+
+    assert dtype == expected_dtype
+
+
+def get_expected_name(box, names):
+    if box is DataFrame:
+        # Since we are operating with a DataFrame and a non-DataFrame,
+        # the non-DataFrame is cast to Series and its name ignored.
+        exname = names[0]
+    elif box in [tm.to_array, pd.array]:
+        exname = names[1]
+    else:
+        exname = names[2]
+    return exname
+
+
+# ------------------------------------------------------------------
+# Timedelta64[ns] dtype Comparisons
+
+
+class TestTimedelta64ArrayLikeComparisons:
+    # Comparison tests for timedelta64[ns] vectors fully parametrized over
+    #  DataFrame/Series/TimedeltaIndex/TimedeltaArray.  Ideally all comparison
+    #  tests will eventually end up here.
+
+    def test_compare_timedelta64_zerodim(self, box_with_array):
+        # GH#26689 should unbox when comparing with zerodim array
+        box = box_with_array
+        xbox = box_with_array if box_with_array not in [Index, pd.array] else np.ndarray
+
+        tdi = timedelta_range("2h", periods=4)
+        other = np.array(tdi.to_numpy()[0])
+
+        tdi = tm.box_expected(tdi, box)
+        res = tdi <= other
+        expected = np.array([True, False, False, False])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(res, expected)
+
+    @pytest.mark.parametrize(
+        "td_scalar",
+        [
+            timedelta(days=1),
+            Timedelta(days=1),
+            Timedelta(days=1).to_timedelta64(),
+            offsets.Hour(24),
+        ],
+    )
+    def test_compare_timedeltalike_scalar(self, box_with_array, td_scalar):
+        # regression test for GH#5963
+        box = box_with_array
+        xbox = box if box not in [Index, pd.array] else np.ndarray
+
+        ser = Series([timedelta(days=1), timedelta(days=2)])
+        ser = tm.box_expected(ser, box)
+        actual = ser > td_scalar
+        expected = Series([False, True])
+        expected = tm.box_expected(expected, xbox)
+        tm.assert_equal(actual, expected)
+
+    @pytest.mark.parametrize(
+        "invalid",
+        [
+            345600000000000,
+            "a",
+            Timestamp("2021-01-01"),
+            Timestamp("2021-01-01").now("UTC"),
+            Timestamp("2021-01-01").now().to_datetime64(),
+            Timestamp("2021-01-01").now().to_pydatetime(),
+            Timestamp("2021-01-01").date(),
+            np.array(4),  # zero-dim mismatched dtype
+        ],
+    )
+    def test_td64_comparisons_invalid(self, box_with_array, invalid):
+        # GH#13624 for str
+        box = box_with_array
+
+        rng = timedelta_range("1 days", periods=10)
+        obj = tm.box_expected(rng, box)
+
+        assert_invalid_comparison(obj, invalid, box)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            list(range(10)),
+            np.arange(10),
+            np.arange(10).astype(np.float32),
+            np.arange(10).astype(object),
+            pd.date_range("1970-01-01", periods=10, tz="UTC").array,
+            np.array(pd.date_range("1970-01-01", periods=10)),
+            list(pd.date_range("1970-01-01", periods=10)),
+            pd.date_range("1970-01-01", periods=10).astype(object),
+            pd.period_range("1971-01-01", freq="D", periods=10).array,
+            pd.period_range("1971-01-01", freq="D", periods=10).astype(object),
+        ],
+    )
+    def test_td64arr_cmp_arraylike_invalid(self, other, box_with_array):
+        # We don't parametrize this over box_with_array because listlike
+        #  other plays poorly with assert_invalid_comparison reversed checks
+
+        rng = timedelta_range("1 days", periods=10)._data
+        rng = tm.box_expected(rng, box_with_array)
+        assert_invalid_comparison(rng, other, box_with_array)
+
+    def test_td64arr_cmp_mixed_invalid(self):
+        rng = timedelta_range("1 days", periods=5)._data
+        other = np.array([0, 1, 2, rng[3], Timestamp("2021-01-01")])
+
+        result = rng == other
+        expected = np.array([False, False, False, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng != other
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        msg = "Invalid comparison between|Cannot compare type|not supported between"
+        with pytest.raises(TypeError, match=msg):
+            rng < other
+        with pytest.raises(TypeError, match=msg):
+            rng > other
+        with pytest.raises(TypeError, match=msg):
+            rng <= other
+        with pytest.raises(TypeError, match=msg):
+            rng >= other
+
+
+class TestTimedelta64ArrayComparisons:
+    # TODO: All of these need to be parametrized over box
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_comp_nat(self, dtype):
+        left = TimedeltaIndex([Timedelta("1 days"), NaT, Timedelta("3 days")])
+        right = TimedeltaIndex([NaT, NaT, Timedelta("3 days")])
+
+        lhs, rhs = left, right
+        if dtype is object:
+            lhs, rhs = left.astype(object), right.astype(object)
+
+        result = rhs == lhs
+        expected = np.array([False, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rhs != lhs
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(lhs == NaT, expected)
+        tm.assert_numpy_array_equal(NaT == rhs, expected)
+
+        expected = np.array([True, True, True])
+        tm.assert_numpy_array_equal(lhs != NaT, expected)
+        tm.assert_numpy_array_equal(NaT != lhs, expected)
+
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(lhs < NaT, expected)
+        tm.assert_numpy_array_equal(NaT > lhs, expected)
+
+    @pytest.mark.parametrize(
+        "idx2",
+        [
+            TimedeltaIndex(
+                ["2 day", "2 day", NaT, NaT, "1 day 00:00:02", "5 days 00:00:03"]
+            ),
+            np.array(
+                [
+                    np.timedelta64(2, "D"),
+                    np.timedelta64(2, "D"),
+                    np.timedelta64("nat"),
+                    np.timedelta64("nat"),
+                    np.timedelta64(1, "D") + np.timedelta64(2, "s"),
+                    np.timedelta64(5, "D") + np.timedelta64(3, "s"),
+                ]
+            ),
+        ],
+    )
+    def test_comparisons_nat(self, idx2):
+        idx1 = TimedeltaIndex(
+            [
+                "1 day",
+                NaT,
+                "1 day 00:00:01",
+                NaT,
+                "1 day 00:00:01",
+                "5 day 00:00:03",
+            ]
+        )
+        # Check pd.NaT is handles as the same as np.nan
+        result = idx1 < idx2
+        expected = np.array([True, False, False, False, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx2 > idx1
+        expected = np.array([True, False, False, False, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx1 <= idx2
+        expected = np.array([True, False, False, False, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx2 >= idx1
+        expected = np.array([True, False, False, False, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx1 == idx2
+        expected = np.array([False, False, False, False, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx1 != idx2
+        expected = np.array([True, True, True, True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    # TODO: better name
+    def test_comparisons_coverage(self):
+        rng = timedelta_range("1 days", periods=10)
+
+        result = rng < rng[3]
+        expected = np.array([True, True, True] + [False] * 7)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = rng == list(rng)
+        exp = rng == rng
+        tm.assert_numpy_array_equal(result, exp)
+
+
+# ------------------------------------------------------------------
+# Timedelta64[ns] dtype Arithmetic Operations
+
+
+class TestTimedelta64ArithmeticUnsorted:
+    # Tests moved from type-specific test files but not
+    #  yet sorted/parametrized/de-duplicated
+
+    def test_ufunc_coercions(self):
+        # normal ops are also tested in tseries/test_timedeltas.py
+        idx = TimedeltaIndex(["2h", "4h", "6h", "8h", "10h"], freq="2h", name="x")
+
+        for result in [idx * 2, np.multiply(idx, 2)]:
+            assert isinstance(result, TimedeltaIndex)
+            exp = TimedeltaIndex(["4h", "8h", "12h", "16h", "20h"], freq="4h", name="x")
+            tm.assert_index_equal(result, exp)
+            assert result.freq == "4h"
+
+        for result in [idx / 2, np.divide(idx, 2)]:
+            assert isinstance(result, TimedeltaIndex)
+            exp = TimedeltaIndex(["1h", "2h", "3h", "4h", "5h"], freq="h", name="x")
+            tm.assert_index_equal(result, exp)
+            assert result.freq == "h"
+
+        for result in [-idx, np.negative(idx)]:
+            assert isinstance(result, TimedeltaIndex)
+            exp = TimedeltaIndex(
+                ["-2h", "-4h", "-6h", "-8h", "-10h"], freq="-2h", name="x"
+            )
+            tm.assert_index_equal(result, exp)
+            assert result.freq == "-2h"
+
+        idx = TimedeltaIndex(["-2h", "-1h", "0h", "1h", "2h"], freq="h", name="x")
+        for result in [abs(idx), np.absolute(idx)]:
+            assert isinstance(result, TimedeltaIndex)
+            exp = TimedeltaIndex(["2h", "1h", "0h", "1h", "2h"], freq=None, name="x")
+            tm.assert_index_equal(result, exp)
+            assert result.freq is None
+
+    def test_subtraction_ops(self):
+        # with datetimes/timedelta and tdi/dti
+        tdi = TimedeltaIndex(["1 days", NaT, "2 days"], name="foo")
+        dti = pd.date_range("20130101", periods=3, name="bar")
+        td = Timedelta("1 days")
+        dt = Timestamp("20130101")
+
+        msg = "cannot subtract a datelike from a TimedeltaArray"
+        with pytest.raises(TypeError, match=msg):
+            tdi - dt
+        with pytest.raises(TypeError, match=msg):
+            tdi - dti
+
+        msg = r"unsupported operand type\(s\) for -"
+        with pytest.raises(TypeError, match=msg):
+            td - dt
+
+        msg = "(bad|unsupported) operand type for unary"
+        with pytest.raises(TypeError, match=msg):
+            td - dti
+
+        result = dt - dti
+        expected = TimedeltaIndex(["0 days", "-1 days", "-2 days"], name="bar")
+        tm.assert_index_equal(result, expected)
+
+        result = dti - dt
+        expected = TimedeltaIndex(["0 days", "1 days", "2 days"], name="bar")
+        tm.assert_index_equal(result, expected)
+
+        result = tdi - td
+        expected = TimedeltaIndex(["0 days", NaT, "1 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = td - tdi
+        expected = TimedeltaIndex(["0 days", NaT, "-1 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = dti - td
+        expected = DatetimeIndex(
+            ["20121231", "20130101", "20130102"], dtype="M8[ns]", freq="D", name="bar"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = dt - tdi
+        expected = DatetimeIndex(
+            ["20121231", NaT, "20121230"], dtype="M8[ns]", name="foo"
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_subtraction_ops_with_tz(self, box_with_array):
+        # check that dt/dti subtraction ops with tz are validated
+        dti = pd.date_range("20130101", periods=3)
+        dti = tm.box_expected(dti, box_with_array)
+        ts = Timestamp("20130101")
+        dt = ts.to_pydatetime()
+        dti_tz = pd.date_range("20130101", periods=3).tz_localize("US/Eastern")
+        dti_tz = tm.box_expected(dti_tz, box_with_array)
+        ts_tz = Timestamp("20130101").tz_localize("US/Eastern")
+        ts_tz2 = Timestamp("20130101").tz_localize("CET")
+        dt_tz = ts_tz.to_pydatetime()
+        td = Timedelta("1 days")
+
+        def _check(result, expected):
+            assert result == expected
+            assert isinstance(result, Timedelta)
+
+        # scalars
+        result = ts - ts
+        expected = Timedelta("0 days")
+        _check(result, expected)
+
+        result = dt_tz - ts_tz
+        expected = Timedelta("0 days")
+        _check(result, expected)
+
+        result = ts_tz - dt_tz
+        expected = Timedelta("0 days")
+        _check(result, expected)
+
+        # tz mismatches
+        msg = "Cannot subtract tz-naive and tz-aware datetime-like objects."
+        with pytest.raises(TypeError, match=msg):
+            dt_tz - ts
+        msg = "can't subtract offset-naive and offset-aware datetimes"
+        with pytest.raises(TypeError, match=msg):
+            dt_tz - dt
+        msg = "can't subtract offset-naive and offset-aware datetimes"
+        with pytest.raises(TypeError, match=msg):
+            dt - dt_tz
+        msg = "Cannot subtract tz-naive and tz-aware datetime-like objects."
+        with pytest.raises(TypeError, match=msg):
+            ts - dt_tz
+        with pytest.raises(TypeError, match=msg):
+            ts_tz2 - ts
+        with pytest.raises(TypeError, match=msg):
+            ts_tz2 - dt
+
+        msg = "Cannot subtract tz-naive and tz-aware"
+        # with dti
+        with pytest.raises(TypeError, match=msg):
+            dti - ts_tz
+        with pytest.raises(TypeError, match=msg):
+            dti_tz - ts
+
+        result = dti_tz - dt_tz
+        expected = TimedeltaIndex(["0 days", "1 days", "2 days"])
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = dt_tz - dti_tz
+        expected = TimedeltaIndex(["0 days", "-1 days", "-2 days"])
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = dti_tz - ts_tz
+        expected = TimedeltaIndex(["0 days", "1 days", "2 days"])
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = ts_tz - dti_tz
+        expected = TimedeltaIndex(["0 days", "-1 days", "-2 days"])
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+        result = td - td
+        expected = Timedelta("0 days")
+        _check(result, expected)
+
+        result = dti_tz - td
+        expected = DatetimeIndex(
+            ["20121231", "20130101", "20130102"], tz="US/Eastern"
+        ).as_unit("ns")
+        expected = tm.box_expected(expected, box_with_array)
+        tm.assert_equal(result, expected)
+
+    def test_dti_tdi_numeric_ops(self):
+        # These are normally union/diff set-like ops
+        tdi = TimedeltaIndex(["1 days", NaT, "2 days"], name="foo")
+        dti = pd.date_range("20130101", periods=3, name="bar")
+
+        result = tdi - tdi
+        expected = TimedeltaIndex(["0 days", NaT, "0 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = tdi + tdi
+        expected = TimedeltaIndex(["2 days", NaT, "4 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = dti - tdi  # name will be reset
+        expected = DatetimeIndex(["20121231", NaT, "20130101"], dtype="M8[ns]")
+        tm.assert_index_equal(result, expected)
+
+    def test_addition_ops(self):
+        # with datetimes/timedelta and tdi/dti
+        tdi = TimedeltaIndex(["1 days", NaT, "2 days"], name="foo")
+        dti = pd.date_range("20130101", periods=3, name="bar")
+        td = Timedelta("1 days")
+        dt = Timestamp("20130101")
+
+        result = tdi + dt
+        expected = DatetimeIndex(
+            ["20130102", NaT, "20130103"], dtype="M8[ns]", name="foo"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = dt + tdi
+        expected = DatetimeIndex(
+            ["20130102", NaT, "20130103"], dtype="M8[ns]", name="foo"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = td + tdi
+        expected = TimedeltaIndex(["2 days", NaT, "3 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = tdi + td
+        expected = TimedeltaIndex(["2 days", NaT, "3 days"], name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # unequal length
+        msg = "cannot add indices of unequal length"
+        with pytest.raises(ValueError, match=msg):
+            tdi + dti[0:1]
+        with pytest.raises(ValueError, match=msg):
+            tdi[0:1] + dti
+
+        # random indexes
+        msg = "Addition/subtraction of integers and integer-arrays"
+        with pytest.raises(TypeError, match=msg):
+            tdi + Index([1, 2, 3], dtype=np.int64)
+
+        # this is a union!
+        # FIXME: don't leave commented-out
+        # pytest.raises(TypeError, lambda : Index([1,2,3]) + tdi)
+
+        result = tdi + dti  # name will be reset
+        expected = DatetimeIndex(["20130102", NaT, "20130105"], dtype="M8[ns]")
+        tm.assert_index_equal(result, expected)
+
+        result = dti + tdi  # name will be reset
+        expected = DatetimeIndex(["20130102", NaT, "20130105"], dtype="M8[ns]")
+        tm.assert_index_equal(result, expected)
+
+        result = dt + td
+        expected = Timestamp("20130102")
+        assert result == expected
+
+        result = td + dt
+        expected = Timestamp("20130102")
+        assert result == expected
+
+    # TODO: Needs more informative name, probably split up into
+    # more targeted tests
+    @pytest.mark.parametrize("freq", ["D", "B"])
+    def test_timedelta(self, freq):
+        index = pd.date_range("1/1/2000", periods=50, freq=freq)
+
+        shifted = index + timedelta(1)
+        back = shifted + timedelta(-1)
+        back = back._with_freq("infer")
+        tm.assert_index_equal(index, back)
+
+        if freq == "D":
+            expected = pd.tseries.offsets.Day(1)
+            assert index.freq == expected
+            assert shifted.freq == expected
+            assert back.freq == expected
+        else:  # freq == 'B'
+            assert index.freq == pd.tseries.offsets.BusinessDay(1)
+            assert shifted.freq is None
+            assert back.freq == pd.tseries.offsets.BusinessDay(1)
+
+        result = index - timedelta(1)
+        expected = index + timedelta(-1)
+        tm.assert_index_equal(result, expected)
+
+    def test_timedelta_tick_arithmetic(self):
+        # GH#4134, buggy with timedeltas
+        rng = pd.date_range("2013", "2014")
+        s = Series(rng)
+        result1 = rng - offsets.Hour(1)
+        result2 = DatetimeIndex(s - np.timedelta64(100000000))
+        result3 = rng - np.timedelta64(100000000)
+        result4 = DatetimeIndex(s - offsets.Hour(1))
+
+        assert result1.freq == rng.freq
+        result1 = result1._with_freq(None)
+        tm.assert_index_equal(result1, result4)
+
+        assert result3.freq == rng.freq
+        result3 = result3._with_freq(None)
+        tm.assert_index_equal(result2, result3)
+
+    def test_tda_add_sub_index(self):
+        # Check that TimedeltaArray defers to Index on arithmetic ops
+        tdi = TimedeltaIndex(["1 days", NaT, "2 days"])
+        tda = tdi.array
+
+        dti = pd.date_range("1999-12-31", periods=3, freq="D")
+
+        result = tda + dti
+        expected = tdi + dti
+        tm.assert_index_equal(result, expected)
+
+        result = tda + tdi
+        expected = tdi + tdi
+        tm.assert_index_equal(result, expected)
+
+        result = tda - tdi
+        expected = tdi - tdi
+        tm.assert_index_equal(result, expected)
+
+    def test_tda_add_dt64_object_array(
+        self, performance_warning, box_with_array, tz_naive_fixture
+    ):
+        # Result should be cast back to DatetimeArray
+        box = box_with_array
+
+        dti = pd.date_range("2016-01-01", periods=3, tz=tz_naive_fixture)
+        dti = dti._with_freq(None)
+        tdi = dti - dti
+
+        obj = tm.box_expected(tdi, box)
+        other = tm.box_expected(dti, box)
+
+        with tm.assert_produces_warning(performance_warning):
+            result = obj + other.astype(object)
+        tm.assert_equal(result, other.astype(object))
+
+    # -------------------------------------------------------------
+    # Binary operations TimedeltaIndex and timedelta-like
+
+    def test_tdi_iadd_timedeltalike(self, two_hours, box_with_array):
+        # only test adding/sub offsets as + is now numeric
+        rng = timedelta_range("1 days", "10 days")
+        expected = timedelta_range("1 days 02:00:00", "10 days 02:00:00", freq="D")
+
+        rng = tm.box_expected(rng, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        orig_rng = rng
+        rng += two_hours
+        tm.assert_equal(rng, expected)
+        if box_with_array is not Index:
+            # Check that operation is actually inplace
+            tm.assert_equal(orig_rng, expected)
+
+    def test_tdi_isub_timedeltalike(self, two_hours, box_with_array):
+        # only test adding/sub offsets as - is now numeric
+        rng = timedelta_range("1 days", "10 days")
+        expected = timedelta_range("0 days 22:00:00", "9 days 22:00:00")
+
+        rng = tm.box_expected(rng, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        orig_rng = rng
+        rng -= two_hours
+        tm.assert_equal(rng, expected)
+        if box_with_array is not Index:
+            # Check that operation is actually inplace
+            tm.assert_equal(orig_rng, expected)
+
+    # -------------------------------------------------------------
+
+    def test_tdi_ops_attributes(self):
+        rng = timedelta_range("2 days", periods=5, freq="2D", name="x")
+
+        result = rng + 1 * rng.freq
+        exp = timedelta_range("4 days", periods=5, freq="2D", name="x")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == "2D"
+
+        result = rng - 2 * rng.freq
+        exp = timedelta_range("-2 days", periods=5, freq="2D", name="x")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == "2D"
+
+        result = rng * 2
+        exp = timedelta_range("4 days", periods=5, freq="4D", name="x")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == "4D"
+
+        result = rng / 2
+        exp = timedelta_range("1 days", periods=5, freq="D", name="x")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == "D"
+
+        result = -rng
+        exp = timedelta_range("-2 days", periods=5, freq="-2D", name="x")
+        tm.assert_index_equal(result, exp)
+        assert result.freq == "-2D"
+
+        rng = timedelta_range("-2 days", periods=5, freq="D", name="x")
+
+        result = abs(rng)
+        exp = TimedeltaIndex(
+            ["2 days", "1 days", "0 days", "1 days", "2 days"], name="x"
+        )
+        tm.assert_index_equal(result, exp)
+        assert result.freq is None
+
+
+class TestAddSubNaTMasking:
+    # TODO: parametrize over boxes
+
+    @pytest.mark.parametrize("str_ts", ["1950-01-01", "1980-01-01"])
+    def test_tdarr_add_timestamp_nat_masking(self, box_with_array, str_ts):
+        # GH#17991 checking for overflow-masking with NaT
+        tdinat = pd.to_timedelta(["24658 days 11:15:00", "NaT"])
+        tdobj = tm.box_expected(tdinat, box_with_array)
+
+        ts = Timestamp(str_ts)
+        ts_variants = [
+            ts,
+            ts.to_pydatetime(),
+            ts.to_datetime64().astype("datetime64[ns]"),
+            ts.to_datetime64().astype("datetime64[D]"),
+        ]
+
+        for variant in ts_variants:
+            res = tdobj + variant
+            if box_with_array is DataFrame:
+                assert res.iloc[1, 1] is NaT
+            else:
+                assert res[1] is NaT
+
+    def test_tdi_add_overflow(self):
+        # See GH#14068
+        # preliminary test scalar analogue of vectorized tests below
+        # TODO: Make raised error message more informative and test
+        with pytest.raises(OutOfBoundsDatetime, match="10155196800000000000"):
+            pd.to_timedelta(106580, "D") + Timestamp("2000")
+        with pytest.raises(OutOfBoundsDatetime, match="10155196800000000000"):
+            Timestamp("2000") + pd.to_timedelta(106580, "D")
+
+        _NaT = NaT._value + 1
+        msg = "Overflow in int64 addition"
+        with pytest.raises(OverflowError, match=msg):
+            pd.to_timedelta([106580], "D") + Timestamp("2000")
+        with pytest.raises(OverflowError, match=msg):
+            Timestamp("2000") + pd.to_timedelta([106580], "D")
+        with pytest.raises(OverflowError, match=msg):
+            pd.to_timedelta([_NaT]) - Timedelta("1 days")
+        with pytest.raises(OverflowError, match=msg):
+            pd.to_timedelta(["5 days", _NaT]) - Timedelta("1 days")
+        with pytest.raises(OverflowError, match=msg):
+            (
+                pd.to_timedelta([_NaT, "5 days", "1 hours"])
+                - pd.to_timedelta(["7 seconds", _NaT, "4 hours"])
+            )
+
+        # These should not overflow!
+        exp = TimedeltaIndex([NaT])
+        result = pd.to_timedelta([NaT]) - Timedelta("1 days")
+        tm.assert_index_equal(result, exp)
+
+        exp = TimedeltaIndex(["4 days", NaT])
+        result = pd.to_timedelta(["5 days", NaT]) - Timedelta("1 days")
+        tm.assert_index_equal(result, exp)
+
+        exp = TimedeltaIndex([NaT, NaT, "5 hours"])
+        result = pd.to_timedelta([NaT, "5 days", "1 hours"]) + pd.to_timedelta(
+            ["7 seconds", NaT, "4 hours"]
+        )
+        tm.assert_index_equal(result, exp)
+
+
+class TestTimedeltaArraylikeAddSubOps:
+    # Tests for timedelta64[ns] __add__, __sub__, __radd__, __rsub__
+
+    def test_sub_nat_retain_unit(self):
+        ser = pd.to_timedelta(Series(["00:00:01"])).astype("m8[s]")
+
+        result = ser - NaT
+        expected = Series([NaT], dtype="m8[s]")
+        tm.assert_series_equal(result, expected)
+
+    # TODO: moved from tests.indexes.timedeltas.test_arithmetic; needs
+    #  parametrization+de-duplication
+    def test_timedelta_ops_with_missing_values(self):
+        # setup
+        s1 = pd.to_timedelta(Series(["00:00:01"]))
+        s2 = pd.to_timedelta(Series(["00:00:02"]))
+
+        sn = pd.to_timedelta(Series([NaT], dtype="m8[ns]"))
+
+        df1 = DataFrame(["00:00:01"]).apply(pd.to_timedelta)
+        df2 = DataFrame(["00:00:02"]).apply(pd.to_timedelta)
+
+        dfn = DataFrame([NaT._value]).apply(pd.to_timedelta)
+
+        scalar1 = pd.to_timedelta("00:00:01")
+        scalar2 = pd.to_timedelta("00:00:02")
+        timedelta_NaT = pd.to_timedelta("NaT")
+
+        actual = scalar1 + scalar1
+        assert actual == scalar2
+        actual = scalar2 - scalar1
+        assert actual == scalar1
+
+        actual = s1 + s1
+        tm.assert_series_equal(actual, s2)
+        actual = s2 - s1
+        tm.assert_series_equal(actual, s1)
+
+        actual = s1 + scalar1
+        tm.assert_series_equal(actual, s2)
+        actual = scalar1 + s1
+        tm.assert_series_equal(actual, s2)
+        actual = s2 - scalar1
+        tm.assert_series_equal(actual, s1)
+        actual = -scalar1 + s2
+        tm.assert_series_equal(actual, s1)
+
+        actual = s1 + timedelta_NaT
+        tm.assert_series_equal(actual, sn)
+        actual = timedelta_NaT + s1
+        tm.assert_series_equal(actual, sn)
+        actual = s1 - timedelta_NaT
+        tm.assert_series_equal(actual, sn)
+        actual = -timedelta_NaT + s1
+        tm.assert_series_equal(actual, sn)
+
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            s1 + np.nan
+        with pytest.raises(TypeError, match=msg):
+            np.nan + s1
+        with pytest.raises(TypeError, match=msg):
+            s1 - np.nan
+        with pytest.raises(TypeError, match=msg):
+            -np.nan + s1
+
+        actual = s1 + NaT
+        tm.assert_series_equal(actual, sn)
+        actual = s2 - NaT
+        tm.assert_series_equal(actual, sn)
+
+        actual = s1 + df1
+        tm.assert_frame_equal(actual, df2)
+        actual = s2 - df1
+        tm.assert_frame_equal(actual, df1)
+        actual = df1 + s1
+        tm.assert_frame_equal(actual, df2)
+        actual = df2 - s1
+        tm.assert_frame_equal(actual, df1)
+
+        actual = df1 + df1
+        tm.assert_frame_equal(actual, df2)
+        actual = df2 - df1
+        tm.assert_frame_equal(actual, df1)
+
+        actual = df1 + scalar1
+        tm.assert_frame_equal(actual, df2)
+        actual = df2 - scalar1
+        tm.assert_frame_equal(actual, df1)
+
+        actual = df1 + timedelta_NaT
+        tm.assert_frame_equal(actual, dfn)
+        actual = df1 - timedelta_NaT
+        tm.assert_frame_equal(actual, dfn)
+
+        msg = "cannot subtract a datelike from|unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            df1 + np.nan
+        with pytest.raises(TypeError, match=msg):
+            df1 - np.nan
+
+        actual = df1 + NaT  # NaT is datetime, not timedelta
+        tm.assert_frame_equal(actual, dfn)
+        actual = df1 - NaT
+        tm.assert_frame_equal(actual, dfn)
+
+    # TODO: moved from tests.series.test_operators, needs splitting, cleanup,
+    # de-duplication, box-parametrization...
+    def test_operators_timedelta64(self):
+        # series ops
+        v1 = pd.date_range("2012-1-1", periods=3, freq="D")
+        v2 = pd.date_range("2012-1-2", periods=3, freq="D")
+        rs = Series(v2) - Series(v1)
+        xp = Series(1e9 * 3600 * 24, rs.index).astype("int64").astype("timedelta64[ns]")
+        tm.assert_series_equal(rs, xp)
+        assert rs.dtype == "timedelta64[ns]"
+
+        df = DataFrame({"A": v1})
+        td = Series([timedelta(days=i) for i in range(3)])
+        assert td.dtype == "timedelta64[ns]"
+
+        # series on the rhs
+        result = df["A"] - df["A"].shift()
+        assert result.dtype == "timedelta64[ns]"
+
+        result = df["A"] + td
+        assert result.dtype == "M8[ns]"
+
+        # scalar Timestamp on rhs
+        maxa = df["A"].max()
+        assert isinstance(maxa, Timestamp)
+
+        resultb = df["A"] - df["A"].max()
+        assert resultb.dtype == "timedelta64[ns]"
+
+        # timestamp on lhs
+        result = resultb + df["A"]
+        values = [Timestamp("20111230"), Timestamp("20120101"), Timestamp("20120103")]
+        expected = Series(values, dtype="M8[ns]", name="A")
+        tm.assert_series_equal(result, expected)
+
+        # datetimes on rhs
+        result = df["A"] - datetime(2001, 1, 1)
+        expected = Series([timedelta(days=4017 + i) for i in range(3)], name="A")
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "m8[ns]"
+
+        d = datetime(2001, 1, 1, 3, 4)
+        resulta = df["A"] - d
+        assert resulta.dtype == "m8[ns]"
+
+        # roundtrip
+        resultb = resulta + d
+        tm.assert_series_equal(df["A"], resultb)
+
+        # timedeltas on rhs
+        td = timedelta(days=1)
+        resulta = df["A"] + td
+        resultb = resulta - td
+        tm.assert_series_equal(resultb, df["A"])
+        assert resultb.dtype == "M8[ns]"
+
+        # roundtrip
+        td = timedelta(minutes=5, seconds=3)
+        resulta = df["A"] + td
+        resultb = resulta - td
+        tm.assert_series_equal(df["A"], resultb)
+        assert resultb.dtype == "M8[ns]"
+
+        # inplace
+        value = rs[2] + np.timedelta64(timedelta(minutes=5, seconds=1))
+        rs[2] += np.timedelta64(timedelta(minutes=5, seconds=1))
+        assert rs[2] == value
+
+    def test_timedelta64_ops_nat(self):
+        # GH 11349
+        timedelta_series = Series([NaT, Timedelta("1s")])
+        nat_series_dtype_timedelta = Series([NaT, NaT], dtype="timedelta64[ns]")
+        single_nat_dtype_timedelta = Series([NaT], dtype="timedelta64[ns]")
+
+        # subtraction
+        tm.assert_series_equal(timedelta_series - NaT, nat_series_dtype_timedelta)
+        tm.assert_series_equal(-NaT + timedelta_series, nat_series_dtype_timedelta)
+
+        tm.assert_series_equal(
+            timedelta_series - single_nat_dtype_timedelta, nat_series_dtype_timedelta
+        )
+        tm.assert_series_equal(
+            -single_nat_dtype_timedelta + timedelta_series, nat_series_dtype_timedelta
+        )
+
+        # addition
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta + NaT, nat_series_dtype_timedelta
+        )
+        tm.assert_series_equal(
+            NaT + nat_series_dtype_timedelta, nat_series_dtype_timedelta
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta + single_nat_dtype_timedelta,
+            nat_series_dtype_timedelta,
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_timedelta + nat_series_dtype_timedelta,
+            nat_series_dtype_timedelta,
+        )
+
+        tm.assert_series_equal(timedelta_series + NaT, nat_series_dtype_timedelta)
+        tm.assert_series_equal(NaT + timedelta_series, nat_series_dtype_timedelta)
+
+        tm.assert_series_equal(
+            timedelta_series + single_nat_dtype_timedelta, nat_series_dtype_timedelta
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_timedelta + timedelta_series, nat_series_dtype_timedelta
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta + NaT, nat_series_dtype_timedelta
+        )
+        tm.assert_series_equal(
+            NaT + nat_series_dtype_timedelta, nat_series_dtype_timedelta
+        )
+
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta + single_nat_dtype_timedelta,
+            nat_series_dtype_timedelta,
+        )
+        tm.assert_series_equal(
+            single_nat_dtype_timedelta + nat_series_dtype_timedelta,
+            nat_series_dtype_timedelta,
+        )
+
+        # multiplication
+        tm.assert_series_equal(
+            nat_series_dtype_timedelta * 1.0, nat_series_dtype_timedelta
+        )
+        tm.assert_series_equal(
+            1.0 * nat_series_dtype_timedelta, nat_series_dtype_timedelta
+        )
+
+        tm.assert_series_equal(timedelta_series * 1, timedelta_series)
+        tm.assert_series_equal(1 * timedelta_series, timedelta_series)
+
+        tm.assert_series_equal(timedelta_series * 1.5, Series([NaT, Timedelta("1.5s")]))
+        tm.assert_series_equal(1.5 * timedelta_series, Series([NaT, Timedelta("1.5s")]))
+
+        tm.assert_series_equal(timedelta_series * np.nan, nat_series_dtype_timedelta)
+        tm.assert_series_equal(np.nan * timedelta_series, nat_series_dtype_timedelta)
+
+        # division
+        tm.assert_series_equal(timedelta_series / 2, Series([NaT, Timedelta("0.5s")]))
+        tm.assert_series_equal(timedelta_series / 2.0, Series([NaT, Timedelta("0.5s")]))
+        tm.assert_series_equal(timedelta_series / np.nan, nat_series_dtype_timedelta)
+
+    # -------------------------------------------------------------
+    # Binary operations td64 arraylike and datetime-like
+
+    @pytest.mark.parametrize("cls", [Timestamp, datetime, np.datetime64])
+    def test_td64arr_add_sub_datetimelike_scalar(
+        self, cls, box_with_array, tz_naive_fixture
+    ):
+        # GH#11925, GH#29558, GH#23215
+        tz = tz_naive_fixture
+
+        dt_scalar = Timestamp("2012-01-01", tz=tz)
+        if cls is datetime:
+            ts = dt_scalar.to_pydatetime()
+        elif cls is np.datetime64:
+            if tz_naive_fixture is not None:
+                pytest.skip(f"{cls} doesn support {tz_naive_fixture}")
+            ts = dt_scalar.to_datetime64()
+        else:
+            ts = dt_scalar
+
+        tdi = timedelta_range("1 day", periods=3)
+        expected = pd.date_range("2012-01-02", periods=3, tz=tz)
+
+        tdarr = tm.box_expected(tdi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        tm.assert_equal(ts + tdarr, expected)
+        tm.assert_equal(tdarr + ts, expected)
+
+        expected2 = pd.date_range("2011-12-31", periods=3, freq="-1D", tz=tz)
+        expected2 = tm.box_expected(expected2, box_with_array)
+
+        tm.assert_equal(ts - tdarr, expected2)
+        tm.assert_equal(ts + (-tdarr), expected2)
+
+        msg = "cannot subtract a datelike"
+        with pytest.raises(TypeError, match=msg):
+            tdarr - ts
+
+    def test_td64arr_add_datetime64_nat(self, box_with_array):
+        # GH#23215
+        other = np.datetime64("NaT")
+
+        tdi = timedelta_range("1 day", periods=3)
+        expected = DatetimeIndex(["NaT", "NaT", "NaT"], dtype="M8[ns]")
+
+        tdser = tm.box_expected(tdi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        tm.assert_equal(tdser + other, expected)
+        tm.assert_equal(other + tdser, expected)
+
+    def test_td64arr_sub_dt64_array(self, box_with_array):
+        dti = pd.date_range("2016-01-01", periods=3)
+        tdi = TimedeltaIndex(["-1 Day"] * 3)
+        dtarr = dti.values
+        expected = DatetimeIndex(dtarr) - tdi
+
+        tdi = tm.box_expected(tdi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        msg = "cannot subtract a datelike from"
+        with pytest.raises(TypeError, match=msg):
+            tdi - dtarr
+
+        # TimedeltaIndex.__rsub__
+        result = dtarr - tdi
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_add_dt64_array(self, box_with_array):
+        dti = pd.date_range("2016-01-01", periods=3)
+        tdi = TimedeltaIndex(["-1 Day"] * 3)
+        dtarr = dti.values
+        expected = DatetimeIndex(dtarr) + tdi
+
+        tdi = tm.box_expected(tdi, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdi + dtarr
+        tm.assert_equal(result, expected)
+        result = dtarr + tdi
+        tm.assert_equal(result, expected)
+
+    # ------------------------------------------------------------------
+    # Invalid __add__/__sub__ operations
+
+    @pytest.mark.parametrize("pi_freq", ["D", "W", "Q", "h"])
+    @pytest.mark.parametrize("tdi_freq", [None, "h"])
+    def test_td64arr_sub_periodlike(
+        self, box_with_array, box_with_array2, tdi_freq, pi_freq
+    ):
+        # GH#20049 subtracting PeriodIndex should raise TypeError
+        tdi = TimedeltaIndex(["1 hours", "2 hours"], freq=tdi_freq)
+        dti = Timestamp("2018-03-07 17:16:40") + tdi
+        pi = dti.to_period(pi_freq)
+        per = pi[0]
+
+        tdi = tm.box_expected(tdi, box_with_array)
+        pi = tm.box_expected(pi, box_with_array2)
+        msg = "cannot subtract|unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            tdi - pi
+
+        # GH#13078 subtraction of Period scalar not supported
+        with pytest.raises(TypeError, match=msg):
+            tdi - per
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            # GH#12624 for str case
+            "a",
+            # GH#19123
+            1,
+            1.5,
+            np.array(2),
+        ],
+    )
+    def test_td64arr_addsub_numeric_scalar_invalid(self, box_with_array, other):
+        # vector-like others are tested in test_td64arr_add_sub_numeric_arr_invalid
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        tdarr = tm.box_expected(tdser, box_with_array)
+
+        assert_invalid_addsub_type(tdarr, other)
+
+    @pytest.mark.parametrize(
+        "vec",
+        [
+            np.array([1, 2, 3]),
+            Index([1, 2, 3]),
+            Series([1, 2, 3]),
+            DataFrame([[1, 2, 3]]),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_td64arr_addsub_numeric_arr_invalid(
+        self, box_with_array, vec, any_real_numpy_dtype
+    ):
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        tdarr = tm.box_expected(tdser, box_with_array)
+
+        vector = vec.astype(any_real_numpy_dtype)
+        assert_invalid_addsub_type(tdarr, vector)
+
+    def test_td64arr_add_sub_int(self, box_with_array, one):
+        # Variants of `one` for #19012, deprecated GH#22535
+        rng = timedelta_range("1 days 09:00:00", freq="h", periods=10)
+        tdarr = tm.box_expected(rng, box_with_array)
+
+        msg = "Addition/subtraction of integers"
+        assert_invalid_addsub_type(tdarr, one, msg)
+
+        # TODO: get inplace ops into assert_invalid_addsub_type
+        with pytest.raises(TypeError, match=msg):
+            tdarr += one
+        with pytest.raises(TypeError, match=msg):
+            tdarr -= one
+
+    def test_td64arr_add_sub_integer_array(self, box_with_array):
+        # GH#19959, deprecated GH#22535
+        # GH#22696 for DataFrame case, check that we don't dispatch to numpy
+        #  implementation, which treats int64 as m8[ns]
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        rng = timedelta_range("1 days 09:00:00", freq="h", periods=3)
+        tdarr = tm.box_expected(rng, box)
+        other = tm.box_expected([4, 3, 2], xbox)
+
+        msg = "Addition/subtraction of integers and integer-arrays"
+        assert_invalid_addsub_type(tdarr, other, msg)
+
+    def test_td64arr_addsub_integer_array_no_freq(self, box_with_array):
+        # GH#19959
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        tdi = TimedeltaIndex(["1 Day", "NaT", "3 Hours"])
+        tdarr = tm.box_expected(tdi, box)
+        other = tm.box_expected([14, -1, 16], xbox)
+
+        msg = "Addition/subtraction of integers"
+        assert_invalid_addsub_type(tdarr, other, msg)
+
+    # ------------------------------------------------------------------
+    # Operations with timedelta-like others
+
+    def test_td64arr_add_sub_td64_array(self, box_with_array):
+        box = box_with_array
+        dti = pd.date_range("2016-01-01", periods=3)
+        tdi = dti - dti.shift(1)
+        tdarr = tdi.values
+
+        expected = 2 * tdi
+        tdi = tm.box_expected(tdi, box)
+        expected = tm.box_expected(expected, box)
+
+        result = tdi + tdarr
+        tm.assert_equal(result, expected)
+        result = tdarr + tdi
+        tm.assert_equal(result, expected)
+
+        expected_sub = 0 * tdi
+        result = tdi - tdarr
+        tm.assert_equal(result, expected_sub)
+        result = tdarr - tdi
+        tm.assert_equal(result, expected_sub)
+
+    def test_td64arr_add_sub_tdi(self, box_with_array, names):
+        # GH#17250 make sure result dtype is correct
+        # GH#19043 make sure names are propagated correctly
+        box = box_with_array
+        exname = get_expected_name(box, names)
+
+        tdi = TimedeltaIndex(["0 days", "1 day"], name=names[1])
+        tdi = np.array(tdi) if box in [tm.to_array, pd.array] else tdi
+        ser = Series([Timedelta(hours=3), Timedelta(hours=4)], name=names[0])
+        expected = Series([Timedelta(hours=3), Timedelta(days=1, hours=4)], name=exname)
+
+        ser = tm.box_expected(ser, box)
+        expected = tm.box_expected(expected, box)
+
+        result = tdi + ser
+        tm.assert_equal(result, expected)
+        assert_dtype(result, "timedelta64[ns]")
+
+        result = ser + tdi
+        tm.assert_equal(result, expected)
+        assert_dtype(result, "timedelta64[ns]")
+
+        expected = Series(
+            [Timedelta(hours=-3), Timedelta(days=1, hours=-4)], name=exname
+        )
+        expected = tm.box_expected(expected, box)
+
+        result = tdi - ser
+        tm.assert_equal(result, expected)
+        assert_dtype(result, "timedelta64[ns]")
+
+        result = ser - tdi
+        tm.assert_equal(result, -expected)
+        assert_dtype(result, "timedelta64[ns]")
+
+    @pytest.mark.parametrize("tdnat", [np.timedelta64("NaT"), NaT])
+    def test_td64arr_add_sub_td64_nat(self, box_with_array, tdnat):
+        # GH#18808, GH#23320 special handling for timedelta64("NaT")
+        box = box_with_array
+        tdi = TimedeltaIndex([NaT, Timedelta("1s")])
+        expected = TimedeltaIndex(["NaT"] * 2)
+
+        obj = tm.box_expected(tdi, box)
+        expected = tm.box_expected(expected, box)
+
+        result = obj + tdnat
+        tm.assert_equal(result, expected)
+        result = tdnat + obj
+        tm.assert_equal(result, expected)
+        result = obj - tdnat
+        tm.assert_equal(result, expected)
+        result = tdnat - obj
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_add_timedeltalike(self, two_hours, box_with_array):
+        # only test adding/sub offsets as + is now numeric
+        # GH#10699 for Tick cases
+        box = box_with_array
+        rng = timedelta_range("1 days", "10 days")
+        expected = timedelta_range("1 days 02:00:00", "10 days 02:00:00", freq="D")
+        rng = tm.box_expected(rng, box)
+        expected = tm.box_expected(expected, box)
+
+        result = rng + two_hours
+        tm.assert_equal(result, expected)
+
+        result = two_hours + rng
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_sub_timedeltalike(self, two_hours, box_with_array):
+        # only test adding/sub offsets as - is now numeric
+        # GH#10699 for Tick cases
+        box = box_with_array
+        rng = timedelta_range("1 days", "10 days")
+        expected = timedelta_range("0 days 22:00:00", "9 days 22:00:00")
+
+        rng = tm.box_expected(rng, box)
+        expected = tm.box_expected(expected, box)
+
+        result = rng - two_hours
+        tm.assert_equal(result, expected)
+
+        result = two_hours - rng
+        tm.assert_equal(result, -expected)
+
+    # ------------------------------------------------------------------
+    # __add__/__sub__ with DateOffsets and arrays of DateOffsets
+
+    def test_td64arr_add_sub_offset_index(
+        self, performance_warning, names, box_with_array
+    ):
+        # GH#18849, GH#19744
+        box = box_with_array
+        exname = get_expected_name(box, names)
+
+        tdi = TimedeltaIndex(["1 days 00:00:00", "3 days 04:00:00"], name=names[0])
+        other = Index([offsets.Hour(n=1), offsets.Minute(n=-2)], name=names[1])
+        other = np.array(other) if box in [tm.to_array, pd.array] else other
+
+        expected = TimedeltaIndex(
+            [tdi[n] + other[n] for n in range(len(tdi))], freq="infer", name=exname
+        )
+        expected_sub = TimedeltaIndex(
+            [tdi[n] - other[n] for n in range(len(tdi))], freq="infer", name=exname
+        )
+
+        tdi = tm.box_expected(tdi, box)
+        expected = tm.box_expected(expected, box).astype(object)
+        expected_sub = tm.box_expected(expected_sub, box).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = tdi + other
+        tm.assert_equal(res, expected)
+
+        with tm.assert_produces_warning(performance_warning):
+            res2 = other + tdi
+        tm.assert_equal(res2, expected)
+
+        with tm.assert_produces_warning(performance_warning):
+            res_sub = tdi - other
+        tm.assert_equal(res_sub, expected_sub)
+
+    def test_td64arr_add_sub_offset_array(self, performance_warning, box_with_array):
+        # GH#18849, GH#18824
+        box = box_with_array
+        tdi = TimedeltaIndex(["1 days 00:00:00", "3 days 04:00:00"])
+        other = np.array([offsets.Hour(n=1), offsets.Minute(n=-2)])
+
+        expected = TimedeltaIndex(
+            [tdi[n] + other[n] for n in range(len(tdi))], freq="infer"
+        )
+        expected_sub = TimedeltaIndex(
+            [tdi[n] - other[n] for n in range(len(tdi))], freq="infer"
+        )
+
+        tdi = tm.box_expected(tdi, box)
+        expected = tm.box_expected(expected, box).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = tdi + other
+        tm.assert_equal(res, expected)
+
+        with tm.assert_produces_warning(performance_warning):
+            res2 = other + tdi
+        tm.assert_equal(res2, expected)
+
+        expected_sub = tm.box_expected(expected_sub, box_with_array).astype(object)
+        with tm.assert_produces_warning(performance_warning):
+            res_sub = tdi - other
+        tm.assert_equal(res_sub, expected_sub)
+
+    def test_td64arr_with_offset_series(
+        self, performance_warning, names, box_with_array
+    ):
+        # GH#18849
+        box = box_with_array
+        box2 = Series if box in [Index, tm.to_array, pd.array] else box
+        exname = get_expected_name(box, names)
+
+        tdi = TimedeltaIndex(["1 days 00:00:00", "3 days 04:00:00"], name=names[0])
+        other = Series([offsets.Hour(n=1), offsets.Minute(n=-2)], name=names[1])
+
+        expected_add = Series(
+            [tdi[n] + other[n] for n in range(len(tdi))], name=exname, dtype=object
+        )
+        obj = tm.box_expected(tdi, box)
+        expected_add = tm.box_expected(expected_add, box2).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res = obj + other
+        tm.assert_equal(res, expected_add)
+
+        with tm.assert_produces_warning(performance_warning):
+            res2 = other + obj
+        tm.assert_equal(res2, expected_add)
+
+        expected_sub = Series(
+            [tdi[n] - other[n] for n in range(len(tdi))], name=exname, dtype=object
+        )
+        expected_sub = tm.box_expected(expected_sub, box2).astype(object)
+
+        with tm.assert_produces_warning(performance_warning):
+            res3 = obj - other
+        tm.assert_equal(res3, expected_sub)
+
+    @pytest.mark.parametrize("obox", [np.array, Index, Series])
+    def test_td64arr_addsub_anchored_offset_arraylike(
+        self, performance_warning, obox, box_with_array
+    ):
+        # GH#18824
+        tdi = TimedeltaIndex(["1 days 00:00:00", "3 days 04:00:00"])
+        tdi = tm.box_expected(tdi, box_with_array)
+
+        anchored = obox([offsets.MonthEnd(), offsets.Day(n=2)])
+
+        # addition/subtraction ops with anchored offsets should issue
+        # a PerformanceWarning and _then_ raise a TypeError.
+        msg = "has incorrect type|cannot add the type MonthEnd"
+        with pytest.raises(TypeError, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                tdi + anchored
+        with pytest.raises(TypeError, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                anchored + tdi
+        with pytest.raises(TypeError, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                tdi - anchored
+        with pytest.raises(TypeError, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                anchored - tdi
+
+    # ------------------------------------------------------------------
+    # Unsorted
+
+    def test_td64arr_add_sub_object_array(self, performance_warning, box_with_array):
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        tdi = timedelta_range("1 day", periods=3, freq="D")
+        tdarr = tm.box_expected(tdi, box)
+
+        other = np.array([Timedelta(days=1), offsets.Day(2), Timestamp("2000-01-04")])
+
+        with tm.assert_produces_warning(performance_warning):
+            result = tdarr + other
+
+        expected = Index(
+            [Timedelta(days=2), Timedelta(days=4), Timestamp("2000-01-07")]
+        )
+        expected = tm.box_expected(expected, xbox).astype(object)
+        tm.assert_equal(result, expected)
+
+        msg = "unsupported operand type|cannot subtract a datelike"
+        with pytest.raises(TypeError, match=msg):
+            with tm.assert_produces_warning(performance_warning):
+                tdarr - other
+
+        with tm.assert_produces_warning(performance_warning):
+            result = other - tdarr
+
+        expected = Index([Timedelta(0), Timedelta(0), Timestamp("2000-01-01")])
+        expected = tm.box_expected(expected, xbox).astype(object)
+        tm.assert_equal(result, expected)
+
+
+class TestTimedeltaArraylikeMulDivOps:
+    # Tests for timedelta64[ns]
+    # __mul__, __rmul__, __div__, __rdiv__, __floordiv__, __rfloordiv__
+
+    # ------------------------------------------------------------------
+    # Multiplication
+    # organized with scalar others first, then array-like
+
+    def test_td64arr_mul_int(self, box_with_array):
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box_with_array)
+
+        result = idx * 1
+        tm.assert_equal(result, idx)
+
+        result = 1 * idx
+        tm.assert_equal(result, idx)
+
+    def test_td64arr_mul_tdlike_scalar_raises(self, two_hours, box_with_array):
+        rng = timedelta_range("1 days", "10 days", name="foo")
+        rng = tm.box_expected(rng, box_with_array)
+        msg = "|".join(
+            [
+                "argument must be an integer",
+                "cannot use operands with types dtype",
+                "Cannot multiply with",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            rng * two_hours
+
+    def test_tdi_mul_int_array_zerodim(self, box_with_array):
+        rng5 = np.arange(5, dtype="int64")
+        idx = TimedeltaIndex(rng5)
+        expected = TimedeltaIndex(rng5 * 5)
+
+        idx = tm.box_expected(idx, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = idx * np.array(5, dtype="int64")
+        tm.assert_equal(result, expected)
+
+    def test_tdi_mul_int_array(self, box_with_array):
+        rng5 = np.arange(5, dtype="int64")
+        idx = TimedeltaIndex(rng5)
+        expected = TimedeltaIndex(rng5**2)
+
+        idx = tm.box_expected(idx, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = idx * rng5
+        tm.assert_equal(result, expected)
+
+    def test_tdi_mul_int_series(self, box_with_array):
+        box = box_with_array
+        xbox = Series if box in [Index, tm.to_array, pd.array] else box
+
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        expected = TimedeltaIndex(np.arange(5, dtype="int64") ** 2)
+
+        idx = tm.box_expected(idx, box)
+        expected = tm.box_expected(expected, xbox)
+
+        result = idx * Series(np.arange(5, dtype="int64"))
+        tm.assert_equal(result, expected)
+
+    def test_tdi_mul_float_series(self, box_with_array):
+        box = box_with_array
+        xbox = Series if box in [Index, tm.to_array, pd.array] else box
+
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box)
+
+        rng5f = np.arange(5, dtype="float64")
+        expected = TimedeltaIndex(rng5f * (rng5f + 1.0))
+        expected = tm.box_expected(expected, xbox)
+
+        result = idx * Series(rng5f + 1.0)
+        tm.assert_equal(result, expected)
+
+    # TODO: Put Series/DataFrame in others?
+    @pytest.mark.parametrize(
+        "other",
+        [
+            np.arange(1, 11),
+            Index(np.arange(1, 11), np.int64),
+            Index(range(1, 11), np.uint64),
+            Index(range(1, 11), np.float64),
+            pd.RangeIndex(1, 11),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_tdi_rmul_arraylike(self, other, box_with_array):
+        box = box_with_array
+
+        tdi = TimedeltaIndex(["1 Day"] * 10)
+        expected = timedelta_range("1 days", "10 days")._with_freq(None)
+
+        tdi = tm.box_expected(tdi, box)
+        xbox = get_upcast_box(tdi, other)
+
+        expected = tm.box_expected(expected, xbox)
+
+        result = other * tdi
+        tm.assert_equal(result, expected)
+        commute = tdi * other
+        tm.assert_equal(commute, expected)
+
+    # ------------------------------------------------------------------
+    # __div__, __rdiv__
+
+    def test_td64arr_div_nat_invalid(self, box_with_array):
+        # don't allow division by NaT (maybe could in the future)
+        rng = timedelta_range("1 days", "10 days", name="foo")
+        rng = tm.box_expected(rng, box_with_array)
+
+        with pytest.raises(TypeError, match="unsupported operand type"):
+            rng / NaT
+        with pytest.raises(TypeError, match="Cannot divide NaTType by"):
+            NaT / rng
+
+        dt64nat = np.datetime64("NaT", "ns")
+        msg = "|".join(
+            [
+                # 'divide' on npdev as of 2021-12-18
+                "ufunc '(true_divide|divide)' cannot use operands",
+                "cannot perform __r?truediv__",
+                "Cannot divide datetime64 by TimedeltaArray",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            rng / dt64nat
+        with pytest.raises(TypeError, match=msg):
+            dt64nat / rng
+
+    def test_td64arr_div_td64nat(self, box_with_array):
+        # GH#23829
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        rng = timedelta_range("1 days", "10 days")
+        rng = tm.box_expected(rng, box)
+
+        other = np.timedelta64("NaT")
+
+        expected = np.array([np.nan] * 10)
+        expected = tm.box_expected(expected, xbox)
+
+        result = rng / other
+        tm.assert_equal(result, expected)
+
+        result = other / rng
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_div_int(self, box_with_array):
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box_with_array)
+
+        result = idx / 1
+        tm.assert_equal(result, idx)
+
+        with pytest.raises(TypeError, match="Cannot divide"):
+            # GH#23829
+            1 / idx
+
+    def test_td64arr_div_tdlike_scalar(self, two_hours, box_with_array):
+        # GH#20088, GH#22163 ensure DataFrame returns correct dtype
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        rng = timedelta_range("1 days", "10 days", name="foo")
+        expected = Index((np.arange(10) + 1) * 12, dtype=np.float64, name="foo")
+
+        rng = tm.box_expected(rng, box)
+        expected = tm.box_expected(expected, xbox)
+
+        result = rng / two_hours
+        tm.assert_equal(result, expected)
+
+        result = two_hours / rng
+        expected = 1 / expected
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("m", [1, 3, 10])
+    @pytest.mark.parametrize("unit", ["D", "h", "m", "s", "ms", "us", "ns"])
+    def test_td64arr_div_td64_scalar(self, m, unit, box_with_array):
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        ser = Series([Timedelta(days=59)] * 3)
+        ser[2] = np.nan
+        flat = ser
+        ser = tm.box_expected(ser, box)
+
+        # op
+        expected = Series([x / np.timedelta64(m, unit) for x in flat])
+        expected = tm.box_expected(expected, xbox)
+        result = ser / np.timedelta64(m, unit)
+        tm.assert_equal(result, expected)
+
+        # reverse op
+        expected = Series([Timedelta(np.timedelta64(m, unit)) / x for x in flat])
+        expected = tm.box_expected(expected, xbox)
+        result = np.timedelta64(m, unit) / ser
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_div_tdlike_scalar_with_nat(self, two_hours, box_with_array):
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        rng = TimedeltaIndex(["1 days", NaT, "2 days"], name="foo")
+        expected = Index([12, np.nan, 24], dtype=np.float64, name="foo")
+
+        rng = tm.box_expected(rng, box)
+        expected = tm.box_expected(expected, xbox)
+
+        result = rng / two_hours
+        tm.assert_equal(result, expected)
+
+        result = two_hours / rng
+        expected = 1 / expected
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_div_td64_ndarray(self, box_with_array):
+        # GH#22631
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        rng = TimedeltaIndex(["1 days", NaT, "2 days"])
+        expected = Index([12, np.nan, 24], dtype=np.float64)
+
+        rng = tm.box_expected(rng, box)
+        expected = tm.box_expected(expected, xbox)
+
+        other = np.array([2, 4, 2], dtype="m8[h]")
+        result = rng / other
+        tm.assert_equal(result, expected)
+
+        result = rng / tm.box_expected(other, box)
+        tm.assert_equal(result, expected)
+
+        result = rng / other.astype(object)
+        tm.assert_equal(result, expected.astype(object))
+
+        result = rng / list(other)
+        tm.assert_equal(result, expected)
+
+        # reversed op
+        expected = 1 / expected
+        result = other / rng
+        tm.assert_equal(result, expected)
+
+        result = tm.box_expected(other, box) / rng
+        tm.assert_equal(result, expected)
+
+        result = other.astype(object) / rng
+        tm.assert_equal(result, expected)
+
+        result = list(other) / rng
+        tm.assert_equal(result, expected)
+
+    def test_tdarr_div_length_mismatch(self, box_with_array):
+        rng = TimedeltaIndex(["1 days", NaT, "2 days"])
+        mismatched = [1, 2, 3, 4]
+
+        rng = tm.box_expected(rng, box_with_array)
+        msg = "Cannot divide vectors|Unable to coerce to Series"
+        for obj in [mismatched, mismatched[:2]]:
+            # one shorter, one longer
+            for other in [obj, np.array(obj), Index(obj)]:
+                with pytest.raises(ValueError, match=msg):
+                    rng / other
+                with pytest.raises(ValueError, match=msg):
+                    other / rng
+
+    def test_td64_div_object_mixed_result(self, box_with_array):
+        # Case where we having a NaT in the result inseat of timedelta64("NaT")
+        #  is misleading
+        orig = timedelta_range("1 Day", periods=3).insert(1, NaT)
+        tdi = tm.box_expected(orig, box_with_array, transpose=False)
+
+        other = np.array([orig[0], 1.5, 2.0, orig[2]], dtype=object)
+        other = tm.box_expected(other, box_with_array, transpose=False)
+
+        res = tdi / other
+
+        expected = Index([1.0, np.timedelta64("NaT", "ns"), orig[0], 1.5], dtype=object)
+        expected = tm.box_expected(expected, box_with_array, transpose=False)
+        if isinstance(expected, NumpyExtensionArray):
+            expected = expected.to_numpy()
+        tm.assert_equal(res, expected)
+        if box_with_array is DataFrame:
+            # We have a np.timedelta64(NaT), not pd.NaT
+            assert isinstance(res.iloc[1, 0], np.timedelta64)
+
+        res = tdi // other
+
+        expected = Index([1, np.timedelta64("NaT", "ns"), orig[0], 1], dtype=object)
+        expected = tm.box_expected(expected, box_with_array, transpose=False)
+        if isinstance(expected, NumpyExtensionArray):
+            expected = expected.to_numpy()
+        tm.assert_equal(res, expected)
+        if box_with_array is DataFrame:
+            # We have a np.timedelta64(NaT), not pd.NaT
+            assert isinstance(res.iloc[1, 0], np.timedelta64)
+
+    # ------------------------------------------------------------------
+    # __floordiv__, __rfloordiv__
+
+    @pytest.mark.skipif(WASM, reason="no fp exception support in wasm")
+    def test_td64arr_floordiv_td64arr_with_nat(self, box_with_array):
+        # GH#35529
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+
+        left = Series([1000, 222330, 30], dtype="timedelta64[ns]")
+        right = Series([1000, 222330, None], dtype="timedelta64[ns]")
+
+        left = tm.box_expected(left, box)
+        right = tm.box_expected(right, box)
+
+        expected = np.array([1.0, 1.0, np.nan], dtype=np.float64)
+        expected = tm.box_expected(expected, xbox)
+
+        with tm.maybe_produces_warning(
+            RuntimeWarning, box is pd.array, check_stacklevel=False
+        ):
+            result = left // right
+
+        tm.assert_equal(result, expected)
+
+        # case that goes through __rfloordiv__ with arraylike
+        with tm.maybe_produces_warning(
+            RuntimeWarning, box is pd.array, check_stacklevel=False
+        ):
+            result = np.asarray(left) // right
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:invalid value encountered:RuntimeWarning")
+    def test_td64arr_floordiv_tdscalar(self, box_with_array, scalar_td):
+        # GH#18831, GH#19125
+        box = box_with_array
+        xbox = np.ndarray if box is pd.array else box
+        td = Timedelta("5m3s")  # i.e. (scalar_td - 1sec) / 2
+
+        td1 = Series([td, td, NaT], dtype="m8[ns]")
+        td1 = tm.box_expected(td1, box, transpose=False)
+
+        expected = Series([0, 0, np.nan])
+        expected = tm.box_expected(expected, xbox, transpose=False)
+
+        result = td1 // scalar_td
+        tm.assert_equal(result, expected)
+
+        # Reversed op
+        expected = Series([2, 2, np.nan])
+        expected = tm.box_expected(expected, xbox, transpose=False)
+
+        result = scalar_td // td1
+        tm.assert_equal(result, expected)
+
+        # same thing buts let's be explicit about calling __rfloordiv__
+        result = td1.__rfloordiv__(scalar_td)
+        tm.assert_equal(result, expected)
+
+    def test_td64arr_floordiv_int(self, box_with_array):
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box_with_array)
+        result = idx // 1
+        tm.assert_equal(result, idx)
+
+        pattern = "floor_divide cannot use operands|Cannot divide int by Timedelta*"
+        with pytest.raises(TypeError, match=pattern):
+            1 // idx
+
+    # ------------------------------------------------------------------
+    # mod, divmod
+    # TODO: operations with timedelta-like arrays, numeric arrays,
+    #  reversed ops
+
+    def test_td64arr_mod_tdscalar(
+        self, performance_warning, box_with_array, three_days
+    ):
+        tdi = timedelta_range("1 Day", "9 days")
+        tdarr = tm.box_expected(tdi, box_with_array)
+
+        expected = TimedeltaIndex(["1 Day", "2 Days", "0 Days"] * 3)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdarr % three_days
+        tm.assert_equal(result, expected)
+
+        if box_with_array is DataFrame and isinstance(three_days, pd.DateOffset):
+            # TODO: making expected be object here a result of DataFrame.__divmod__
+            #  being defined in a naive way that does not dispatch to the underlying
+            #  array's __divmod__
+            expected = expected.astype(object)
+        else:
+            performance_warning = False
+
+        with tm.assert_produces_warning(performance_warning):
+            result = divmod(tdarr, three_days)
+
+        tm.assert_equal(result[1], expected)
+        tm.assert_equal(result[0], tdarr // three_days)
+
+    def test_td64arr_mod_int(self, box_with_array):
+        tdi = timedelta_range("1 ns", "10 ns", periods=10)
+        tdarr = tm.box_expected(tdi, box_with_array)
+
+        expected = TimedeltaIndex(["1 ns", "0 ns"] * 5)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdarr % 2
+        tm.assert_equal(result, expected)
+
+        msg = "Cannot divide int by"
+        with pytest.raises(TypeError, match=msg):
+            2 % tdarr
+
+        result = divmod(tdarr, 2)
+        tm.assert_equal(result[1], expected)
+        tm.assert_equal(result[0], tdarr // 2)
+
+    def test_td64arr_rmod_tdscalar(self, box_with_array, three_days):
+        tdi = timedelta_range("1 Day", "9 days")
+        tdarr = tm.box_expected(tdi, box_with_array)
+
+        expected = ["0 Days", "1 Day", "0 Days"] + ["3 Days"] * 6
+        expected = TimedeltaIndex(expected)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = three_days % tdarr
+        tm.assert_equal(result, expected)
+
+        result = divmod(three_days, tdarr)
+        tm.assert_equal(result[1], expected)
+        tm.assert_equal(result[0], three_days // tdarr)
+
+    # ------------------------------------------------------------------
+    # Operations with invalid others
+
+    def test_td64arr_mul_tdscalar_invalid(self, box_with_array, scalar_td):
+        td1 = Series([timedelta(minutes=5, seconds=3)] * 3)
+        td1.iloc[2] = np.nan
+
+        td1 = tm.box_expected(td1, box_with_array)
+
+        # check that we are getting a TypeError
+        # with 'operate' (from core/ops.py) for the ops that are not
+        # defined
+        pattern = "operate|unsupported|cannot|not supported"
+        with pytest.raises(TypeError, match=pattern):
+            td1 * scalar_td
+        with pytest.raises(TypeError, match=pattern):
+            scalar_td * td1
+
+    def test_td64arr_mul_too_short_raises(self, box_with_array):
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box_with_array)
+        msg = "|".join(
+            [
+                "cannot use operands with types dtype",
+                "Cannot multiply with unequal lengths",
+                "Unable to coerce to Series",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            # length check before dtype check
+            idx * idx[:3]
+        with pytest.raises(ValueError, match=msg):
+            idx * np.array([1, 2])
+
+    def test_td64arr_mul_td64arr_raises(self, box_with_array):
+        idx = TimedeltaIndex(np.arange(5, dtype="int64"))
+        idx = tm.box_expected(idx, box_with_array)
+        msg = "cannot use operands with types dtype"
+        with pytest.raises(TypeError, match=msg):
+            idx * idx
+
+    # ------------------------------------------------------------------
+    # Operations with numeric others
+
+    def test_td64arr_mul_numeric_scalar(self, box_with_array, one):
+        # GH#4521
+        # divide/multiply by integers
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        expected = Series(["-59 Days", "-59 Days", "NaT"], dtype="timedelta64[ns]")
+
+        tdser = tm.box_expected(tdser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdser * (-one)
+        tm.assert_equal(result, expected)
+        result = (-one) * tdser
+        tm.assert_equal(result, expected)
+
+        expected = Series(["118 Days", "118 Days", "NaT"], dtype="timedelta64[ns]")
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdser * (2 * one)
+        tm.assert_equal(result, expected)
+        result = (2 * one) * tdser
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("two", [2, 2.0, np.array(2), np.array(2.0)])
+    def test_td64arr_div_numeric_scalar(self, box_with_array, two):
+        # GH#4521
+        # divide/multiply by integers
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        expected = Series(["29.5D", "29.5D", "NaT"], dtype="timedelta64[ns]")
+
+        tdser = tm.box_expected(tdser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdser / two
+        tm.assert_equal(result, expected)
+
+        with pytest.raises(TypeError, match="Cannot divide"):
+            two / tdser
+
+    @pytest.mark.parametrize("two", [2, 2.0, np.array(2), np.array(2.0)])
+    def test_td64arr_floordiv_numeric_scalar(self, box_with_array, two):
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        expected = Series(["29.5D", "29.5D", "NaT"], dtype="timedelta64[ns]")
+
+        tdser = tm.box_expected(tdser, box_with_array)
+        expected = tm.box_expected(expected, box_with_array)
+
+        result = tdser // two
+        tm.assert_equal(result, expected)
+
+        with pytest.raises(TypeError, match="Cannot divide"):
+            two // tdser
+
+    @pytest.mark.parametrize(
+        "klass",
+        [np.array, Index, Series],
+        ids=lambda x: x.__name__,
+    )
+    def test_td64arr_rmul_numeric_array(
+        self,
+        box_with_array,
+        klass,
+        any_real_numpy_dtype,
+    ):
+        # GH#4521
+        # divide/multiply by integers
+
+        vector = klass([20, 30, 40])
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        vector = vector.astype(any_real_numpy_dtype)
+
+        expected = Series(["1180 Days", "1770 Days", "NaT"], dtype="timedelta64[ns]")
+
+        tdser = tm.box_expected(tdser, box_with_array)
+        xbox = get_upcast_box(tdser, vector)
+
+        expected = tm.box_expected(expected, xbox)
+
+        result = tdser * vector
+        tm.assert_equal(result, expected)
+
+        result = vector * tdser
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "klass",
+        [np.array, Index, Series],
+        ids=lambda x: x.__name__,
+    )
+    def test_td64arr_div_numeric_array(
+        self, box_with_array, klass, any_real_numpy_dtype
+    ):
+        # GH#4521
+        # divide/multiply by integers
+
+        vector = klass([20, 30, 40])
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="m8[ns]")
+        vector = vector.astype(any_real_numpy_dtype)
+
+        expected = Series(["2.95D", "1D 23h 12m", "NaT"], dtype="timedelta64[ns]")
+
+        tdser = tm.box_expected(tdser, box_with_array)
+        xbox = get_upcast_box(tdser, vector)
+        expected = tm.box_expected(expected, xbox)
+
+        result = tdser / vector
+        tm.assert_equal(result, expected)
+
+        pattern = "|".join(
+            [
+                "true_divide'? cannot use operands",
+                "cannot perform __div__",
+                "cannot perform __truediv__",
+                "unsupported operand",
+                "Cannot divide",
+                "ufunc 'divide' cannot use operands with types",
+            ]
+        )
+        with pytest.raises(TypeError, match=pattern):
+            vector / tdser
+
+        result = tdser / vector.astype(object)
+        if box_with_array is DataFrame:
+            expected = [tdser.iloc[0, n] / vector[n] for n in range(len(vector))]
+            expected = tm.box_expected(expected, xbox).astype(object)
+            expected[2] = expected[2].fillna(np.timedelta64("NaT", "ns"))
+        else:
+            expected = [tdser[n] / vector[n] for n in range(len(tdser))]
+            expected = [
+                x if x is not NaT else np.timedelta64("NaT", "ns") for x in expected
+            ]
+            if xbox is tm.to_array:
+                expected = tm.to_array(expected).astype(object)
+            else:
+                expected = xbox(expected, dtype=object)
+
+        tm.assert_equal(result, expected)
+
+        with pytest.raises(TypeError, match=pattern):
+            vector.astype(object) / tdser
+
+    def test_td64arr_mul_int_series(self, box_with_array, names):
+        # GH#19042 test for correct name attachment
+        box = box_with_array
+        exname = get_expected_name(box, names)
+
+        tdi = TimedeltaIndex(
+            ["0days", "1day", "2days", "3days", "4days"], name=names[0]
+        )
+        # TODO: Should we be parametrizing over types for `ser` too?
+        ser = Series([0, 1, 2, 3, 4], dtype=np.int64, name=names[1])
+
+        expected = Series(
+            ["0days", "1day", "4days", "9days", "16days"],
+            dtype="timedelta64[ns]",
+            name=exname,
+        )
+
+        tdi = tm.box_expected(tdi, box)
+        xbox = get_upcast_box(tdi, ser)
+
+        expected = tm.box_expected(expected, xbox)
+
+        result = ser * tdi
+        tm.assert_equal(result, expected)
+
+        result = tdi * ser
+        tm.assert_equal(result, expected)
+
+    # TODO: Should we be parametrizing over types for `ser` too?
+    def test_float_series_rdiv_td64arr(self, box_with_array, names):
+        # GH#19042 test for correct name attachment
+        box = box_with_array
+        tdi = TimedeltaIndex(
+            ["0days", "1day", "2days", "3days", "4days"], name=names[0]
+        )
+        ser = Series([1.5, 3, 4.5, 6, 7.5], dtype=np.float64, name=names[1])
+
+        xname = names[2] if box not in [tm.to_array, pd.array] else names[1]
+        expected = Series(
+            [tdi[n] / ser[n] for n in range(len(ser))],
+            dtype="timedelta64[ns]",
+            name=xname,
+        )
+
+        tdi = tm.box_expected(tdi, box)
+        xbox = get_upcast_box(tdi, ser)
+        expected = tm.box_expected(expected, xbox)
+
+        result = ser.__rtruediv__(tdi)
+        if box is DataFrame:
+            assert result is NotImplemented
+        else:
+            tm.assert_equal(result, expected)
+
+    def test_td64arr_all_nat_div_object_dtype_numeric(self, box_with_array):
+        # GH#39750 make sure we infer the result as td64
+        tdi = TimedeltaIndex([NaT, NaT])
+
+        left = tm.box_expected(tdi, box_with_array)
+        right = np.array([2, 2.0], dtype=object)
+
+        tdnat = np.timedelta64("NaT", "ns")
+        expected = Index([tdnat] * 2, dtype=object)
+        if box_with_array is not Index:
+            expected = tm.box_expected(expected, box_with_array).astype(object)
+            if box_with_array in [Series, DataFrame]:
+                expected = expected.fillna(tdnat)  # GH#18463
+
+        result = left / right
+        tm.assert_equal(result, expected)
+
+        result = left // right
+        tm.assert_equal(result, expected)
+
+
+class TestTimedelta64ArrayLikeArithmetic:
+    # Arithmetic tests for timedelta64[ns] vectors fully parametrized over
+    #  DataFrame/Series/TimedeltaIndex/TimedeltaArray.  Ideally all arithmetic
+    #  tests will eventually end up here.
+
+    def test_td64arr_pow_invalid(self, scalar_td, box_with_array):
+        td1 = Series([timedelta(minutes=5, seconds=3)] * 3)
+        td1.iloc[2] = np.nan
+
+        td1 = tm.box_expected(td1, box_with_array)
+
+        # check that we are getting a TypeError
+        # with 'operate' (from core/ops.py) for the ops that are not
+        # defined
+        pattern = "operate|unsupported|cannot|not supported"
+        with pytest.raises(TypeError, match=pattern):
+            scalar_td**td1
+
+        with pytest.raises(TypeError, match=pattern):
+            td1**scalar_td
+
+
+def test_add_timestamp_to_timedelta():
+    # GH: 35897
+    timestamp = Timestamp("2021-01-01")
+    result = timestamp + timedelta_range("0s", "1s", periods=31)
+    expected = DatetimeIndex(
+        [
+            timestamp
+            + (
+                pd.to_timedelta("0.033333333s") * i
+                + pd.to_timedelta("0.000000001s") * divmod(i, 3)[0]
+            )
+            for i in range(31)
+        ]
+    )
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/__init__.py b/pandas/pandas/tests/arrays/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/boolean/__init__.py b/pandas/pandas/tests/arrays/boolean/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/boolean/test_arithmetic.py b/pandas/pandas/tests/arrays/boolean/test_arithmetic.py
new file mode 100644
index 0000000000000..9ff690cdc914d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_arithmetic.py
@@ -0,0 +1,134 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.fixture
+def data():
+    """Fixture returning boolean array with valid and missing values."""
+    return pd.array(
+        [True, False] * 4 + [np.nan] + [True, False] * 44 + [np.nan] + [True, False],
+        dtype="boolean",
+    )
+
+
+@pytest.fixture
+def left_array():
+    """Fixture returning boolean array with valid and missing values."""
+    return pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+
+
+@pytest.fixture
+def right_array():
+    """Fixture returning boolean array with valid and missing values."""
+    return pd.array([True, False, None] * 3, dtype="boolean")
+
+
+# Basic test for the arithmetic array ops
+# -----------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize(
+    "opname, exp",
+    [
+        ("add", [True, True, None, True, False, None, None, None, None]),
+        ("mul", [True, False, None, False, False, None, None, None, None]),
+    ],
+    ids=["add", "mul"],
+)
+def test_add_mul(left_array, right_array, opname, exp):
+    op = getattr(operator, opname)
+    result = op(left_array, right_array)
+    expected = pd.array(exp, dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_sub(left_array, right_array):
+    msg = (
+        r"numpy boolean subtract, the `-` operator, is (?:deprecated|not supported), "
+        r"use the bitwise_xor, the `\^` operator, or the logical_xor function instead\."
+    )
+    with pytest.raises(TypeError, match=msg):
+        left_array - right_array
+
+
+def test_div(left_array, right_array):
+    msg = "operator '.*' not implemented for bool dtypes"
+    with pytest.raises(NotImplementedError, match=msg):
+        # check that we are matching the non-masked Series behavior
+        pd.Series(left_array._data) / pd.Series(right_array._data)
+
+    with pytest.raises(NotImplementedError, match=msg):
+        left_array / right_array
+
+
+@pytest.mark.parametrize(
+    "opname",
+    [
+        "floordiv",
+        "mod",
+        "pow",
+    ],
+)
+def test_op_int8(left_array, right_array, opname):
+    op = getattr(operator, opname)
+    if opname != "mod":
+        msg = "operator '.*' not implemented for bool dtypes"
+        with pytest.raises(NotImplementedError, match=msg):
+            result = op(left_array, right_array)
+        return
+    result = op(left_array, right_array)
+    expected = op(left_array.astype("Int8"), right_array.astype("Int8"))
+    tm.assert_extension_array_equal(result, expected)
+
+
+# Test generic characteristics / errors
+# -----------------------------------------------------------------------------
+
+
+def test_error_invalid_values(data, all_arithmetic_operators):
+    # invalid ops
+    op = all_arithmetic_operators
+    s = pd.Series(data)
+    ops = getattr(s, op)
+
+    # invalid scalars
+    msg = (
+        "did not contain a loop with signature matching types|"
+        "BooleanArray cannot perform the operation|"
+        "not supported for the input types, and the inputs could not be safely coerced "
+        "to any supported types according to the casting rule ''safe''|"
+        "not supported for dtype"
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops("foo")
+    msg = "|".join(
+        [
+            r"unsupported operand type\(s\) for",
+            "Concatenation operation is not implemented for NumPy arrays",
+            "has no kernel",
+            "not supported for dtype",
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Timestamp("20180101"))
+
+    # invalid array-likes
+    if op not in ("__mul__", "__rmul__"):
+        # TODO(extension) numpy's mul with object array sees booleans as numbers
+        msg = "|".join(
+            [
+                r"unsupported operand type\(s\) for",
+                "can only concatenate str",
+                "not all arguments converted during string formatting",
+                "has no kernel",
+                "not implemented",
+                "not supported for dtype",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            ops(pd.Series("foo", index=s.index))
diff --git a/pandas/pandas/tests/arrays/boolean/test_astype.py b/pandas/pandas/tests/arrays/boolean/test_astype.py
new file mode 100644
index 0000000000000..8c2672218f273
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_astype.py
@@ -0,0 +1,59 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_astype(using_infer_string):
+    # with missing values
+    arr = pd.array([True, False, None], dtype="boolean")
+
+    with pytest.raises(ValueError, match="cannot convert NA to integer"):
+        arr.astype("int64")
+
+    with pytest.raises(ValueError, match="cannot convert float NaN to"):
+        arr.astype("bool")
+
+    result = arr.astype("float64")
+    expected = np.array([1, 0, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.astype("str")
+    if using_infer_string:
+        expected = pd.array(
+            ["True", "False", None], dtype=pd.StringDtype(na_value=np.nan)
+        )
+        tm.assert_extension_array_equal(result, expected)
+    else:
+        expected = np.array(["True", "False", "<NA>"], dtype=f"{tm.ENDIAN}U5")
+        tm.assert_numpy_array_equal(result, expected)
+
+    # no missing values
+    arr = pd.array([True, False, True], dtype="boolean")
+    result = arr.astype("int64")
+    expected = np.array([1, 0, 1], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.astype("bool")
+    expected = np.array([True, False, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_astype_to_boolean_array():
+    # astype to BooleanArray
+    arr = pd.array([True, False, None], dtype="boolean")
+
+    result = arr.astype("boolean")
+    tm.assert_extension_array_equal(result, arr)
+    result = arr.astype(pd.BooleanDtype())
+    tm.assert_extension_array_equal(result, arr)
+
+
+def test_astype_to_integer_array():
+    # astype to IntegerArray
+    arr = pd.array([True, False, None], dtype="boolean")
+
+    result = arr.astype("Int64")
+    expected = pd.array([1, 0, None], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/boolean/test_comparison.py b/pandas/pandas/tests/arrays/boolean/test_comparison.py
new file mode 100644
index 0000000000000..2eeb9da574b1e
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_comparison.py
@@ -0,0 +1,60 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.arrays import BooleanArray
+from pandas.tests.arrays.masked_shared import ComparisonOps
+
+
+@pytest.fixture
+def data():
+    """Fixture returning boolean array with valid and missing data"""
+    return pd.array(
+        [True, False] * 4 + [np.nan] + [True, False] * 44 + [np.nan] + [True, False],
+        dtype="boolean",
+    )
+
+
+@pytest.fixture
+def dtype():
+    """Fixture returning BooleanDtype"""
+    return pd.BooleanDtype()
+
+
+class TestComparisonOps(ComparisonOps):
+    def test_compare_scalar(self, data, comparison_op):
+        self._compare_other(data, comparison_op, True)
+
+    def test_compare_array(self, data, comparison_op):
+        other = pd.array([True] * len(data), dtype="boolean")
+        self._compare_other(data, comparison_op, other)
+        other = np.array([True] * len(data))
+        self._compare_other(data, comparison_op, other)
+        other = pd.Series([True] * len(data))
+        self._compare_other(data, comparison_op, other)
+
+    @pytest.mark.parametrize("other", [True, False, pd.NA])
+    def test_scalar(self, other, comparison_op, dtype):
+        ComparisonOps.test_scalar(self, other, comparison_op, dtype)
+
+    def test_array(self, comparison_op):
+        op = comparison_op
+        a = pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        b = pd.array([True, False, None] * 3, dtype="boolean")
+
+        result = op(a, b)
+
+        values = op(a._data, b._data)
+        mask = a._mask | b._mask
+        expected = BooleanArray(values, mask)
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        result[0] = None
+        tm.assert_extension_array_equal(
+            a, pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        )
+        tm.assert_extension_array_equal(
+            b, pd.array([True, False, None] * 3, dtype="boolean")
+        )
diff --git a/pandas/pandas/tests/arrays/boolean/test_construction.py b/pandas/pandas/tests/arrays/boolean/test_construction.py
new file mode 100644
index 0000000000000..d821c52d3becb
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_construction.py
@@ -0,0 +1,325 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.arrays import BooleanArray
+from pandas.core.arrays.boolean import coerce_to_array
+
+
+def test_boolean_array_constructor():
+    values = np.array([True, False, True, False], dtype="bool")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = BooleanArray(values, mask)
+    expected = pd.array([True, False, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+    with pytest.raises(TypeError, match="values should be boolean numpy array"):
+        BooleanArray(values.tolist(), mask)
+
+    with pytest.raises(TypeError, match="mask should be boolean numpy array"):
+        BooleanArray(values, mask.tolist())
+
+    with pytest.raises(TypeError, match="values should be boolean numpy array"):
+        BooleanArray(values.astype(int), mask)
+
+    with pytest.raises(TypeError, match="mask should be boolean numpy array"):
+        BooleanArray(values, None)
+
+    with pytest.raises(ValueError, match="values.shape must match mask.shape"):
+        BooleanArray(values.reshape(1, -1), mask)
+
+    with pytest.raises(ValueError, match="values.shape must match mask.shape"):
+        BooleanArray(values, mask.reshape(1, -1))
+
+
+def test_boolean_array_constructor_copy():
+    values = np.array([True, False, True, False], dtype="bool")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = BooleanArray(values, mask)
+    assert result._data is values
+    assert result._mask is mask
+
+    result = BooleanArray(values, mask, copy=True)
+    assert result._data is not values
+    assert result._mask is not mask
+
+
+def test_to_boolean_array():
+    expected = BooleanArray(
+        np.array([True, False, True]), np.array([False, False, False])
+    )
+
+    result = pd.array([True, False, True], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+    result = pd.array(np.array([True, False, True]), dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+    result = pd.array(np.array([True, False, True], dtype=object), dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+    # with missing values
+    expected = BooleanArray(
+        np.array([True, False, True]), np.array([False, False, True])
+    )
+
+    result = pd.array([True, False, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+    result = pd.array(np.array([True, False, None], dtype=object), dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_boolean_array_all_none():
+    expected = BooleanArray(np.array([True, True, True]), np.array([True, True, True]))
+
+    result = pd.array([None, None, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+    result = pd.array(np.array([None, None, None], dtype=object), dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    [
+        ([True, False, None, np.nan, pd.NA], [True, False, None, None, None]),
+        ([True, np.nan], [True, None]),
+        ([True, pd.NA], [True, None]),
+        ([np.nan, np.nan], [None, None]),
+        (np.array([np.nan, np.nan], dtype=float), [None, None]),
+    ],
+)
+def test_to_boolean_array_missing_indicators(a, b):
+    result = pd.array(a, dtype="boolean")
+    expected = pd.array(b, dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        ["foo", "bar"],
+        ["1", "2"],
+        # "foo",
+        [1, 2],
+        [1.0, 2.0],
+        pd.date_range("20130101", periods=2),
+        np.array(["foo"]),
+        np.array([1, 2]),
+        np.array([1.0, 2.0]),
+        [np.nan, {"a": 1}],
+    ],
+)
+def test_to_boolean_array_error(values):
+    # error in converting existing arrays to BooleanArray
+    msg = "Need to pass bool-like value"
+    with pytest.raises(TypeError, match=msg):
+        pd.array(values, dtype="boolean")
+
+
+def test_to_boolean_array_from_integer_array():
+    result = pd.array(np.array([1, 0, 1, 0]), dtype="boolean")
+    expected = pd.array([True, False, True, False], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+    # with missing values
+    result = pd.array(np.array([1, 0, 1, None]), dtype="boolean")
+    expected = pd.array([True, False, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_boolean_array_from_float_array():
+    result = pd.array(np.array([1.0, 0.0, 1.0, 0.0]), dtype="boolean")
+    expected = pd.array([True, False, True, False], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+    # with missing values
+    result = pd.array(np.array([1.0, 0.0, 1.0, np.nan]), dtype="boolean")
+    expected = pd.array([True, False, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_boolean_array_integer_like():
+    # integers of 0's and 1's
+    result = pd.array([1, 0, 1, 0], dtype="boolean")
+    expected = pd.array([True, False, True, False], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+    # with missing values
+    result = pd.array([1, 0, 1, None], dtype="boolean")
+    expected = pd.array([True, False, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_coerce_to_array():
+    # TODO this is currently not public API
+    values = np.array([True, False, True, False], dtype="bool")
+    mask = np.array([False, False, False, True], dtype="bool")
+    result = BooleanArray(*coerce_to_array(values, mask=mask))
+    expected = BooleanArray(values, mask)
+    tm.assert_extension_array_equal(result, expected)
+    assert result._data is values
+    assert result._mask is mask
+    result = BooleanArray(*coerce_to_array(values, mask=mask, copy=True))
+    expected = BooleanArray(values, mask)
+    tm.assert_extension_array_equal(result, expected)
+    assert result._data is not values
+    assert result._mask is not mask
+
+    # mixed missing from values and mask
+    values = [True, False, None, False]
+    mask = np.array([False, False, False, True], dtype="bool")
+    result = BooleanArray(*coerce_to_array(values, mask=mask))
+    expected = BooleanArray(
+        np.array([True, False, True, True]), np.array([False, False, True, True])
+    )
+    tm.assert_extension_array_equal(result, expected)
+    result = BooleanArray(*coerce_to_array(np.array(values, dtype=object), mask=mask))
+    tm.assert_extension_array_equal(result, expected)
+    result = BooleanArray(*coerce_to_array(values, mask=mask.tolist()))
+    tm.assert_extension_array_equal(result, expected)
+
+    # raise errors for wrong dimension
+    values = np.array([True, False, True, False], dtype="bool")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    # passing 2D values is OK as long as no mask
+    coerce_to_array(values.reshape(1, -1))
+
+    with pytest.raises(ValueError, match="values.shape and mask.shape must match"):
+        coerce_to_array(values.reshape(1, -1), mask=mask)
+
+    with pytest.raises(ValueError, match="values.shape and mask.shape must match"):
+        coerce_to_array(values, mask=mask.reshape(1, -1))
+
+
+def test_coerce_to_array_from_boolean_array():
+    # passing BooleanArray to coerce_to_array
+    values = np.array([True, False, True, False], dtype="bool")
+    mask = np.array([False, False, False, True], dtype="bool")
+    arr = BooleanArray(values, mask)
+    result = BooleanArray(*coerce_to_array(arr))
+    tm.assert_extension_array_equal(result, arr)
+    # no copy
+    assert result._data is arr._data
+    assert result._mask is arr._mask
+
+    result = BooleanArray(*coerce_to_array(arr), copy=True)
+    tm.assert_extension_array_equal(result, arr)
+    assert result._data is not arr._data
+    assert result._mask is not arr._mask
+
+    with pytest.raises(ValueError, match="cannot pass mask for BooleanArray input"):
+        coerce_to_array(arr, mask=mask)
+
+
+def test_coerce_to_numpy_array():
+    # with missing values -> object dtype
+    arr = pd.array([True, False, None], dtype="boolean")
+    result = np.array(arr)
+    expected = np.array([True, False, pd.NA], dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # also with no missing values -> object dtype
+    arr = pd.array([True, False, True], dtype="boolean")
+    result = np.array(arr)
+    expected = np.array([True, False, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # force bool dtype
+    result = np.array(arr, dtype="bool")
+    expected = np.array([True, False, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+    # with missing values will raise error
+    arr = pd.array([True, False, None], dtype="boolean")
+    msg = (
+        "cannot convert to 'bool'-dtype NumPy array with missing values. "
+        "Specify an appropriate 'na_value' for this dtype."
+    )
+    with pytest.raises(ValueError, match=msg):
+        np.array(arr, dtype="bool")
+
+
+def test_to_boolean_array_from_strings():
+    result = BooleanArray._from_sequence_of_strings(
+        np.array(["True", "False", "1", "1.0", "0", "0.0", np.nan], dtype=object),
+        dtype=pd.BooleanDtype(),
+    )
+    expected = BooleanArray(
+        np.array([True, False, True, True, False, False, False]),
+        np.array([False, False, False, False, False, False, True]),
+    )
+
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_boolean_array_from_strings_invalid_string():
+    with pytest.raises(ValueError, match="cannot be cast"):
+        BooleanArray._from_sequence_of_strings(["donkey"], dtype=pd.BooleanDtype())
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy(box):
+    con = pd.Series if box else pd.array
+    # default (with or without missing values) -> object dtype
+    arr = con([True, False, True], dtype="boolean")
+    result = arr.to_numpy()
+    expected = np.array([True, False, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([True, False, None], dtype="boolean")
+    result = arr.to_numpy()
+    expected = np.array([True, False, pd.NA], dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([True, False, None], dtype="boolean")
+    result = arr.to_numpy(dtype="str")
+    expected = np.array([True, False, pd.NA], dtype=f"{tm.ENDIAN}U5")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # no missing values -> can convert to bool, otherwise raises
+    arr = con([True, False, True], dtype="boolean")
+    result = arr.to_numpy(dtype="bool")
+    expected = np.array([True, False, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([True, False, None], dtype="boolean")
+    with pytest.raises(ValueError, match="cannot convert to 'bool'-dtype"):
+        result = arr.to_numpy(dtype="bool")
+
+    # specify dtype and na_value
+    arr = con([True, False, None], dtype="boolean")
+    result = arr.to_numpy(dtype=object, na_value=None)
+    expected = np.array([True, False, None], dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype=bool, na_value=False)
+    expected = np.array([True, False, False], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype="int64", na_value=-99)
+    expected = np.array([1, 0, -99], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype="float64", na_value=np.nan)
+    expected = np.array([1, 0, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # converting to int or float without specifying na_value raises
+    with pytest.raises(ValueError, match="cannot convert to 'int64'-dtype"):
+        arr.to_numpy(dtype="int64")
+
+
+def test_to_numpy_copy():
+    # to_numpy can be zero-copy if no missing values
+    arr = pd.array([True, False, True], dtype="boolean")
+    result = arr.to_numpy(dtype=bool)
+    result[0] = False
+    tm.assert_extension_array_equal(
+        arr, pd.array([False, False, True], dtype="boolean")
+    )
+
+    arr = pd.array([True, False, True], dtype="boolean")
+    result = arr.to_numpy(dtype=bool, copy=True)
+    result[0] = False
+    tm.assert_extension_array_equal(arr, pd.array([True, False, True], dtype="boolean"))
diff --git a/pandas/pandas/tests/arrays/boolean/test_function.py b/pandas/pandas/tests/arrays/boolean/test_function.py
new file mode 100644
index 0000000000000..2b3f3d3d16ac6
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_function.py
@@ -0,0 +1,126 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "ufunc", [np.add, np.logical_or, np.logical_and, np.logical_xor]
+)
+def test_ufuncs_binary(ufunc):
+    # two BooleanArrays
+    a = pd.array([True, False, None], dtype="boolean")
+    result = ufunc(a, a)
+    expected = pd.array(ufunc(a._data, a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    s = pd.Series(a)
+    result = ufunc(s, a)
+    expected = pd.Series(ufunc(a._data, a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_series_equal(result, expected)
+
+    # Boolean with numpy array
+    arr = np.array([True, True, False])
+    result = ufunc(a, arr)
+    expected = pd.array(ufunc(a._data, arr), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(arr, a)
+    expected = pd.array(ufunc(arr, a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    # BooleanArray with scalar
+    result = ufunc(a, True)
+    expected = pd.array(ufunc(a._data, True), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(True, a)
+    expected = pd.array(ufunc(True, a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    # not handled types
+    msg = r"operand type\(s\) all returned NotImplemented from __array_ufunc__"
+    with pytest.raises(TypeError, match=msg):
+        ufunc(a, "test")
+
+
+@pytest.mark.parametrize("ufunc", [np.logical_not])
+def test_ufuncs_unary(ufunc):
+    a = pd.array([True, False, None], dtype="boolean")
+    result = ufunc(a)
+    expected = pd.array(ufunc(a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_extension_array_equal(result, expected)
+
+    ser = pd.Series(a)
+    result = ufunc(ser)
+    expected = pd.Series(ufunc(a._data), dtype="boolean")
+    expected[a._mask] = np.nan
+    tm.assert_series_equal(result, expected)
+
+
+def test_ufunc_numeric():
+    # np.sqrt on np.bool_ returns float16, which we upcast to Float32
+    #  bc we do not have Float16
+    arr = pd.array([True, False, None], dtype="boolean")
+
+    res = np.sqrt(arr)
+
+    expected = pd.array([1, 0, None], dtype="Float32")
+    tm.assert_extension_array_equal(res, expected)
+
+
+@pytest.mark.parametrize("values", [[True, False], [True, None]])
+def test_ufunc_reduce_raises(values):
+    arr = pd.array(values, dtype="boolean")
+
+    res = np.add.reduce(arr)
+    if arr[-1] is pd.NA:
+        expected = pd.NA
+    else:
+        expected = arr._data.sum()
+    tm.assert_almost_equal(res, expected)
+
+
+def test_value_counts_na():
+    arr = pd.array([True, False, pd.NA], dtype="boolean")
+    result = arr.value_counts(dropna=False)
+    expected = pd.Series([1, 1, 1], index=arr, dtype="Int64", name="count")
+    assert expected.index.dtype == arr.dtype
+    tm.assert_series_equal(result, expected)
+
+    result = arr.value_counts(dropna=True)
+    expected = pd.Series([1, 1], index=arr[:-1], dtype="Int64", name="count")
+    assert expected.index.dtype == arr.dtype
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_with_normalize():
+    ser = pd.Series([True, False, pd.NA], dtype="boolean")
+    result = ser.value_counts(normalize=True)
+    expected = pd.Series([1, 1], index=ser[:-1], dtype="Float64", name="proportion") / 2
+    assert expected.index.dtype == "boolean"
+    tm.assert_series_equal(result, expected)
+
+
+def test_diff():
+    a = pd.array(
+        [True, True, False, False, True, None, True, None, False], dtype="boolean"
+    )
+    result = pd.core.algorithms.diff(a, 1)
+    expected = pd.array(
+        [None, False, True, False, True, None, None, None, None], dtype="boolean"
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    ser = pd.Series(a)
+    result = ser.diff()
+    expected = pd.Series(expected)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/boolean/test_indexing.py b/pandas/pandas/tests/arrays/boolean/test_indexing.py
new file mode 100644
index 0000000000000..6a7daea16963c
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_indexing.py
@@ -0,0 +1,13 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("na", [None, np.nan, pd.NA])
+def test_setitem_missing_values(na):
+    arr = pd.array([True, False, None], dtype="boolean")
+    expected = pd.array([True, None, None], dtype="boolean")
+    arr[1] = na
+    tm.assert_extension_array_equal(arr, expected)
diff --git a/pandas/pandas/tests/arrays/boolean/test_logical.py b/pandas/pandas/tests/arrays/boolean/test_logical.py
new file mode 100644
index 0000000000000..97a24e0f24756
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_logical.py
@@ -0,0 +1,255 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.arrays import BooleanArray
+from pandas.core.ops.mask_ops import (
+    kleene_and,
+    kleene_or,
+    kleene_xor,
+)
+from pandas.tests.extension.base import BaseOpsUtil
+
+
+class TestLogicalOps(BaseOpsUtil):
+    def test_numpy_scalars_ok(self, all_logical_operators):
+        a = pd.array([True, False, None], dtype="boolean")
+        op = getattr(a, all_logical_operators)
+
+        tm.assert_extension_array_equal(op(True), op(np.bool_(True)))
+        tm.assert_extension_array_equal(op(False), op(np.bool_(False)))
+
+    def get_op_from_name(self, op_name):
+        short_opname = op_name.strip("_")
+        short_opname = short_opname if "xor" in short_opname else short_opname + "_"
+        try:
+            op = getattr(operator, short_opname)
+        except AttributeError:
+            # Assume it is the reverse operator
+            rop = getattr(operator, short_opname[1:])
+            op = lambda x, y: rop(y, x)
+
+        return op
+
+    def test_empty_ok(self, all_logical_operators):
+        a = pd.array([], dtype="boolean")
+        op_name = all_logical_operators
+        result = getattr(a, op_name)(True)
+        tm.assert_extension_array_equal(a, result)
+
+        result = getattr(a, op_name)(False)
+        tm.assert_extension_array_equal(a, result)
+
+        result = getattr(a, op_name)(pd.NA)
+        tm.assert_extension_array_equal(a, result)
+
+    @pytest.mark.parametrize(
+        "other", ["a", pd.Timestamp(2017, 1, 1, 12), np.timedelta64(4)]
+    )
+    def test_eq_mismatched_type(self, other):
+        # GH-44499
+        arr = pd.array([True, False])
+        result = arr == other
+        expected = pd.array([False, False])
+        tm.assert_extension_array_equal(result, expected)
+
+        result = arr != other
+        expected = pd.array([True, True])
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize("other", [[True, False], [True, False, True, False]])
+    def test_logical_length_mismatch_raises(self, other, all_logical_operators):
+        op_name = all_logical_operators
+        a = pd.array([True, False, None], dtype="boolean")
+        msg = "Lengths must match"
+
+        with pytest.raises(ValueError, match=msg):
+            getattr(a, op_name)(other)
+
+        with pytest.raises(ValueError, match=msg):
+            getattr(a, op_name)(np.array(other))
+
+        with pytest.raises(ValueError, match=msg):
+            getattr(a, op_name)(pd.array(other, dtype="boolean"))
+
+    def test_logical_nan_raises(self, all_logical_operators):
+        op_name = all_logical_operators
+        a = pd.array([True, False, None], dtype="boolean")
+        msg = "Got float instead"
+
+        with pytest.raises(TypeError, match=msg):
+            getattr(a, op_name)(np.nan)
+
+    @pytest.mark.parametrize("other", ["a", 1])
+    def test_non_bool_or_na_other_raises(self, other, all_logical_operators):
+        a = pd.array([True, False], dtype="boolean")
+        with pytest.raises(TypeError, match=str(type(other).__name__)):
+            getattr(a, all_logical_operators)(other)
+
+    def test_kleene_or(self):
+        # A clear test of behavior.
+        a = pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        b = pd.array([True, False, None] * 3, dtype="boolean")
+        result = a | b
+        expected = pd.array(
+            [True, True, True, True, False, None, True, None, None], dtype="boolean"
+        )
+        tm.assert_extension_array_equal(result, expected)
+
+        result = b | a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        )
+        tm.assert_extension_array_equal(
+            b, pd.array([True, False, None] * 3, dtype="boolean")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (pd.NA, [True, None, None]),
+            (True, [True, True, True]),
+            (np.bool_(True), [True, True, True]),
+            (False, [True, False, None]),
+            (np.bool_(False), [True, False, None]),
+        ],
+    )
+    def test_kleene_or_scalar(self, other, expected):
+        # TODO: test True & False
+        a = pd.array([True, False, None], dtype="boolean")
+        result = a | other
+        expected = pd.array(expected, dtype="boolean")
+        tm.assert_extension_array_equal(result, expected)
+
+        result = other | a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True, False, None], dtype="boolean")
+        )
+
+    def test_kleene_and(self):
+        # A clear test of behavior.
+        a = pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        b = pd.array([True, False, None] * 3, dtype="boolean")
+        result = a & b
+        expected = pd.array(
+            [True, False, None, False, False, False, None, False, None], dtype="boolean"
+        )
+        tm.assert_extension_array_equal(result, expected)
+
+        result = b & a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        )
+        tm.assert_extension_array_equal(
+            b, pd.array([True, False, None] * 3, dtype="boolean")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (pd.NA, [None, False, None]),
+            (True, [True, False, None]),
+            (False, [False, False, False]),
+            (np.bool_(True), [True, False, None]),
+            (np.bool_(False), [False, False, False]),
+        ],
+    )
+    def test_kleene_and_scalar(self, other, expected):
+        a = pd.array([True, False, None], dtype="boolean")
+        result = a & other
+        expected = pd.array(expected, dtype="boolean")
+        tm.assert_extension_array_equal(result, expected)
+
+        result = other & a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True, False, None], dtype="boolean")
+        )
+
+    def test_kleene_xor(self):
+        a = pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        b = pd.array([True, False, None] * 3, dtype="boolean")
+        result = a ^ b
+        expected = pd.array(
+            [False, True, None, True, False, None, None, None, None], dtype="boolean"
+        )
+        tm.assert_extension_array_equal(result, expected)
+
+        result = b ^ a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        )
+        tm.assert_extension_array_equal(
+            b, pd.array([True, False, None] * 3, dtype="boolean")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (pd.NA, [None, None, None]),
+            (True, [False, True, None]),
+            (np.bool_(True), [False, True, None]),
+            (np.bool_(False), [True, False, None]),
+        ],
+    )
+    def test_kleene_xor_scalar(self, other, expected):
+        a = pd.array([True, False, None], dtype="boolean")
+        result = a ^ other
+        expected = pd.array(expected, dtype="boolean")
+        tm.assert_extension_array_equal(result, expected)
+
+        result = other ^ a
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_extension_array_equal(
+            a, pd.array([True, False, None], dtype="boolean")
+        )
+
+    @pytest.mark.parametrize("other", [True, False, pd.NA, [True, False, None] * 3])
+    def test_no_masked_assumptions(self, other, all_logical_operators):
+        # The logical operations should not assume that masked values are False!
+        a = pd.arrays.BooleanArray(
+            np.array([True, True, True, False, False, False, True, False, True]),
+            np.array([False] * 6 + [True, True, True]),
+        )
+        b = pd.array([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean")
+        if isinstance(other, list):
+            other = pd.array(other, dtype="boolean")
+
+        result = getattr(a, all_logical_operators)(other)
+        expected = getattr(b, all_logical_operators)(other)
+        tm.assert_extension_array_equal(result, expected)
+
+        if isinstance(other, BooleanArray):
+            other._data[other._mask] = True
+            a._data[a._mask] = False
+
+            result = getattr(a, all_logical_operators)(other)
+            expected = getattr(b, all_logical_operators)(other)
+            tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("operation", [kleene_or, kleene_xor, kleene_and])
+def test_error_both_scalar(operation):
+    msg = r"Either `left` or `right` need to be a np\.ndarray."
+    with pytest.raises(TypeError, match=msg):
+        # masks need to be non-None, otherwise it ends up in an infinite recursion
+        operation(True, True, np.zeros(1), np.zeros(1))
diff --git a/pandas/pandas/tests/arrays/boolean/test_ops.py b/pandas/pandas/tests/arrays/boolean/test_ops.py
new file mode 100644
index 0000000000000..95ebe8528c2e5
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_ops.py
@@ -0,0 +1,27 @@
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestUnaryOps:
+    def test_invert(self):
+        a = pd.array([True, False, None], dtype="boolean")
+        expected = pd.array([False, True, None], dtype="boolean")
+        tm.assert_extension_array_equal(~a, expected)
+
+        expected = pd.Series(expected, index=["a", "b", "c"], name="name")
+        result = ~pd.Series(a, index=["a", "b", "c"], name="name")
+        tm.assert_series_equal(result, expected)
+
+        df = pd.DataFrame({"A": a, "B": [True, False, False]}, index=["a", "b", "c"])
+        result = ~df
+        expected = pd.DataFrame(
+            {"A": expected, "B": [False, True, True]}, index=["a", "b", "c"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_abs(self):
+        # matching numpy behavior, abs is the identity function
+        arr = pd.array([True, False, None], dtype="boolean")
+        result = abs(arr)
+
+        tm.assert_extension_array_equal(result, arr)
diff --git a/pandas/pandas/tests/arrays/boolean/test_reduction.py b/pandas/pandas/tests/arrays/boolean/test_reduction.py
new file mode 100644
index 0000000000000..7071c6f8844e4
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_reduction.py
@@ -0,0 +1,61 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+
+
+@pytest.fixture
+def data():
+    """Fixture returning boolean array, with valid and missing values."""
+    return pd.array(
+        [True, False] * 4 + [np.nan] + [True, False] * 44 + [np.nan] + [True, False],
+        dtype="boolean",
+    )
+
+
+@pytest.mark.parametrize(
+    "values, exp_any, exp_all, exp_any_noskip, exp_all_noskip",
+    [
+        ([True, pd.NA], True, True, True, pd.NA),
+        ([False, pd.NA], False, False, pd.NA, False),
+        ([pd.NA], False, True, pd.NA, pd.NA),
+        ([], False, True, False, True),
+        # GH-33253: all True / all False values buggy with skipna=False
+        ([True, True], True, True, True, True),
+        ([False, False], False, False, False, False),
+    ],
+)
+def test_any_all(values, exp_any, exp_all, exp_any_noskip, exp_all_noskip):
+    # the methods return numpy scalars
+    exp_any = pd.NA if exp_any is pd.NA else np.bool_(exp_any)
+    exp_all = pd.NA if exp_all is pd.NA else np.bool_(exp_all)
+    exp_any_noskip = pd.NA if exp_any_noskip is pd.NA else np.bool_(exp_any_noskip)
+    exp_all_noskip = pd.NA if exp_all_noskip is pd.NA else np.bool_(exp_all_noskip)
+
+    for con in [pd.array, pd.Series]:
+        a = con(values, dtype="boolean")
+        assert a.any() is exp_any
+        assert a.all() is exp_all
+        assert a.any(skipna=False) is exp_any_noskip
+        assert a.all(skipna=False) is exp_all_noskip
+
+        assert np.any(a.any()) is exp_any
+        assert np.all(a.all()) is exp_all
+
+
+def test_reductions_return_types(dropna, data, all_numeric_reductions):
+    op = all_numeric_reductions
+    s = pd.Series(data)
+    if dropna:
+        s = s.dropna()
+
+    if op in ("sum", "prod"):
+        assert isinstance(getattr(s, op)(), np.int_)
+    elif op == "count":
+        # Oddly on the 32 bit build (but not Windows), this is intc (!= intp)
+        assert isinstance(getattr(s, op)(), np.integer)
+    elif op in ("min", "max"):
+        assert isinstance(getattr(s, op)(), np.bool_)
+    else:
+        # "mean", "std", "var", "median", "kurt", "skew"
+        assert isinstance(getattr(s, op)(), np.float64)
diff --git a/pandas/pandas/tests/arrays/boolean/test_repr.py b/pandas/pandas/tests/arrays/boolean/test_repr.py
new file mode 100644
index 0000000000000..0ee904b18cc9e
--- /dev/null
+++ b/pandas/pandas/tests/arrays/boolean/test_repr.py
@@ -0,0 +1,13 @@
+import pandas as pd
+
+
+def test_repr():
+    df = pd.DataFrame({"A": pd.array([True, False, None], dtype="boolean")})
+    expected = "       A\n0   True\n1  False\n2   <NA>"
+    assert repr(df) == expected
+
+    expected = "0     True\n1    False\n2     <NA>\nName: A, dtype: boolean"
+    assert repr(df.A) == expected
+
+    expected = "<BooleanArray>\n[True, False, <NA>]\nLength: 3, dtype: boolean"
+    assert repr(df.A.array) == expected
diff --git a/pandas/pandas/tests/arrays/categorical/__init__.py b/pandas/pandas/tests/arrays/categorical/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/categorical/test_algos.py b/pandas/pandas/tests/arrays/categorical/test_algos.py
new file mode 100644
index 0000000000000..a7d0becc30dd9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_algos.py
@@ -0,0 +1,96 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("categories", [["b", "a", "c"], ["a", "b", "c", "d"]])
+def test_factorize(categories, ordered):
+    cat = pd.Categorical(
+        ["b", "b", "a", "c", None], categories=categories, ordered=ordered
+    )
+    codes, uniques = pd.factorize(cat)
+    expected_codes = np.array([0, 0, 1, 2, -1], dtype=np.intp)
+    expected_uniques = pd.Categorical(
+        ["b", "a", "c"], categories=categories, ordered=ordered
+    )
+
+    tm.assert_numpy_array_equal(codes, expected_codes)
+    tm.assert_categorical_equal(uniques, expected_uniques)
+
+
+def test_factorized_sort():
+    cat = pd.Categorical(["b", "b", None, "a"])
+    codes, uniques = pd.factorize(cat, sort=True)
+    expected_codes = np.array([1, 1, -1, 0], dtype=np.intp)
+    expected_uniques = pd.Categorical(["a", "b"])
+
+    tm.assert_numpy_array_equal(codes, expected_codes)
+    tm.assert_categorical_equal(uniques, expected_uniques)
+
+
+def test_factorized_sort_ordered():
+    cat = pd.Categorical(
+        ["b", "b", None, "a"], categories=["c", "b", "a"], ordered=True
+    )
+
+    codes, uniques = pd.factorize(cat, sort=True)
+    expected_codes = np.array([0, 0, -1, 1], dtype=np.intp)
+    expected_uniques = pd.Categorical(
+        ["b", "a"], categories=["c", "b", "a"], ordered=True
+    )
+
+    tm.assert_numpy_array_equal(codes, expected_codes)
+    tm.assert_categorical_equal(uniques, expected_uniques)
+
+
+def test_isin_cats():
+    # GH2003
+    cat = pd.Categorical(["a", "b", np.nan])
+
+    result = cat.isin(["a", np.nan])
+    expected = np.array([True, False, True], dtype=bool)
+    tm.assert_numpy_array_equal(expected, result)
+
+    result = cat.isin(["a", "c"])
+    expected = np.array([True, False, False], dtype=bool)
+    tm.assert_numpy_array_equal(expected, result)
+
+
+@pytest.mark.parametrize("value", [[""], [None, ""], [pd.NaT, ""]])
+def test_isin_cats_corner_cases(value):
+    # GH36550
+    cat = pd.Categorical([""])
+    result = cat.isin(value)
+    expected = np.array([True], dtype=bool)
+    tm.assert_numpy_array_equal(expected, result)
+
+
+@pytest.mark.parametrize("empty", [[], pd.Series(dtype=object), np.array([])])
+def test_isin_empty(empty):
+    s = pd.Categorical(["a", "b"])
+    expected = np.array([False, False], dtype=bool)
+
+    result = s.isin(empty)
+    tm.assert_numpy_array_equal(expected, result)
+
+
+def test_diff():
+    ser = pd.Series([1, 2, 3], dtype="category")
+
+    msg = "Convert to a suitable dtype"
+    with pytest.raises(TypeError, match=msg):
+        ser.diff()
+
+    df = ser.to_frame(name="A")
+    with pytest.raises(TypeError, match=msg):
+        df.diff()
+
+
+def test_hash_read_only_categorical():
+    # GH#58481
+    idx = pd.Index(pd.Index(["a", "b", "c"], dtype="object").values)
+    cat = pd.CategoricalDtype(idx)
+    arr = pd.Series(["a", "b"], dtype=cat).values
+    assert hash(arr.dtype) == hash(arr.dtype)
diff --git a/pandas/pandas/tests/arrays/categorical/test_analytics.py b/pandas/pandas/tests/arrays/categorical/test_analytics.py
new file mode 100644
index 0000000000000..47fa354e12393
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_analytics.py
@@ -0,0 +1,353 @@
+import re
+import sys
+
+import numpy as np
+import pytest
+
+from pandas.compat import PYPY
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.types import is_scalar
+
+
+class TestCategoricalAnalytics:
+    @pytest.mark.parametrize("aggregation", ["min", "max"])
+    def test_min_max_not_ordered_raises(self, aggregation):
+        # unordered cats have no min/max
+        cat = Categorical(["a", "b", "c", "d"], ordered=False)
+        msg = f"Categorical is not ordered for operation {aggregation}"
+        agg_func = getattr(cat, aggregation)
+
+        with pytest.raises(TypeError, match=msg):
+            agg_func()
+
+        ufunc = np.minimum if aggregation == "min" else np.maximum
+        with pytest.raises(TypeError, match=msg):
+            ufunc.reduce(cat)
+
+    def test_min_max_ordered(self, index_or_series_or_array):
+        cat = Categorical(["a", "b", "c", "d"], ordered=True)
+        obj = index_or_series_or_array(cat)
+        _min = obj.min()
+        _max = obj.max()
+        assert _min == "a"
+        assert _max == "d"
+
+        assert np.minimum.reduce(obj) == "a"
+        assert np.maximum.reduce(obj) == "d"
+        # TODO: raises if we pass axis=0  (on Index and Categorical, not Series)
+
+        cat = Categorical(
+            ["a", "b", "c", "d"], categories=["d", "c", "b", "a"], ordered=True
+        )
+        obj = index_or_series_or_array(cat)
+        _min = obj.min()
+        _max = obj.max()
+        assert _min == "d"
+        assert _max == "a"
+        assert np.minimum.reduce(obj) == "d"
+        assert np.maximum.reduce(obj) == "a"
+
+    def test_min_max_reduce(self):
+        # GH52788
+        cat = Categorical(["a", "b", "c", "d"], ordered=True)
+        df = DataFrame(cat)
+
+        result_max = df.agg("max")
+        expected_max = Series(Categorical(["d"], dtype=cat.dtype))
+        tm.assert_series_equal(result_max, expected_max)
+
+        result_min = df.agg("min")
+        expected_min = Series(Categorical(["a"], dtype=cat.dtype))
+        tm.assert_series_equal(result_min, expected_min)
+
+    @pytest.mark.parametrize(
+        "categories,expected",
+        [
+            (list("ABC"), np.nan),
+            ([1, 2, 3], np.nan),
+            pytest.param(
+                Series(date_range("2020-01-01", periods=3), dtype="category"),
+                NaT,
+                marks=pytest.mark.xfail(
+                    reason="https://github.com/pandas-dev/pandas/issues/29962"
+                ),
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("aggregation", ["min", "max"])
+    def test_min_max_ordered_empty(self, categories, expected, aggregation):
+        # GH 30227
+        cat = Categorical([], categories=categories, ordered=True)
+
+        agg_func = getattr(cat, aggregation)
+        result = agg_func()
+        assert result is expected
+
+    @pytest.mark.parametrize(
+        "values, categories",
+        [(["a", "b", "c", np.nan], list("cba")), ([1, 2, 3, np.nan], [3, 2, 1])],
+    )
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_with_nan(self, values, categories, function, skipna):
+        # GH 25303
+        cat = Categorical(values, categories=categories, ordered=True)
+        result = getattr(cat, function)(skipna=skipna)
+
+        if skipna is False:
+            assert result is np.nan
+        else:
+            expected = categories[0] if function == "min" else categories[2]
+            assert result == expected
+
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_only_nan(self, function, skipna):
+        # https://github.com/pandas-dev/pandas/issues/33450
+        cat = Categorical([np.nan], categories=[1, 2], ordered=True)
+        result = getattr(cat, function)(skipna=skipna)
+        assert result is np.nan
+
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_numeric_only_min_max_raises(self, method):
+        # GH 25303
+        cat = Categorical(
+            [np.nan, 1, 2, np.nan], categories=[5, 4, 3, 2, 1], ordered=True
+        )
+        with pytest.raises(TypeError, match=".* got an unexpected keyword"):
+            getattr(cat, method)(numeric_only=True)
+
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_numpy_min_max_raises(self, method):
+        cat = Categorical(["a", "b", "c", "b"], ordered=False)
+        msg = (
+            f"Categorical is not ordered for operation {method}\n"
+            "you can use .as_ordered() to change the Categorical to an ordered one"
+        )
+        method = getattr(np, method)
+        with pytest.raises(TypeError, match=re.escape(msg)):
+            method(cat)
+
+    @pytest.mark.parametrize("kwarg", ["axis", "out", "keepdims"])
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_numpy_min_max_unsupported_kwargs_raises(self, method, kwarg):
+        cat = Categorical(["a", "b", "c", "b"], ordered=True)
+        msg = (
+            f"the '{kwarg}' parameter is not supported in the pandas implementation "
+            f"of {method}"
+        )
+        if kwarg == "axis":
+            msg = r"`axis` must be fewer than the number of dimensions \(1\)"
+        kwargs = {kwarg: 42}
+        method = getattr(np, method)
+        with pytest.raises(ValueError, match=msg):
+            method(cat, **kwargs)
+
+    @pytest.mark.parametrize("method, expected", [("min", "a"), ("max", "c")])
+    def test_numpy_min_max_axis_equals_none(self, method, expected):
+        cat = Categorical(["a", "b", "c", "b"], ordered=True)
+        method = getattr(np, method)
+        result = method(cat, axis=None)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "values,categories,exp_mode",
+        [
+            ([1, 1, 2, 4, 5, 5, 5], [5, 4, 3, 2, 1], [5]),
+            ([1, 1, 1, 4, 5, 5, 5], [5, 4, 3, 2, 1], [5, 1]),
+            ([1, 2, 3, 4, 5], [5, 4, 3, 2, 1], [5, 4, 3, 2, 1]),
+            ([np.nan, np.nan, np.nan, 4, 5], [5, 4, 3, 2, 1], [5, 4]),
+            ([np.nan, np.nan, np.nan, 4, 5, 4], [5, 4, 3, 2, 1], [4]),
+            ([np.nan, np.nan, 4, 5, 4], [5, 4, 3, 2, 1], [4]),
+        ],
+    )
+    def test_mode(self, values, categories, exp_mode):
+        cat = Categorical(values, categories=categories, ordered=True)
+        res = Series(cat).mode()._values
+        exp = Categorical(exp_mode, categories=categories, ordered=True)
+        tm.assert_categorical_equal(res, exp)
+
+    def test_searchsorted(self, ordered):
+        # https://github.com/pandas-dev/pandas/issues/8420
+        # https://github.com/pandas-dev/pandas/issues/14522
+
+        cat = Categorical(
+            ["cheese", "milk", "apple", "bread", "bread"],
+            categories=["cheese", "milk", "apple", "bread"],
+            ordered=ordered,
+        )
+        ser = Series(cat)
+
+        # Searching for single item argument, side='left' (default)
+        res_cat = cat.searchsorted("apple")
+        assert res_cat == 2
+        assert is_scalar(res_cat)
+
+        res_ser = ser.searchsorted("apple")
+        assert res_ser == 2
+        assert is_scalar(res_ser)
+
+        # Searching for single item array, side='left' (default)
+        res_cat = cat.searchsorted(["bread"])
+        res_ser = ser.searchsorted(["bread"])
+        exp = np.array([3], dtype=np.intp)
+        tm.assert_numpy_array_equal(res_cat, exp)
+        tm.assert_numpy_array_equal(res_ser, exp)
+
+        # Searching for several items array, side='right'
+        res_cat = cat.searchsorted(["apple", "bread"], side="right")
+        res_ser = ser.searchsorted(["apple", "bread"], side="right")
+        exp = np.array([3, 5], dtype=np.intp)
+        tm.assert_numpy_array_equal(res_cat, exp)
+        tm.assert_numpy_array_equal(res_ser, exp)
+
+        # Searching for a single value that is not from the Categorical
+        with pytest.raises(TypeError, match="cucumber"):
+            cat.searchsorted("cucumber")
+        with pytest.raises(TypeError, match="cucumber"):
+            ser.searchsorted("cucumber")
+
+        # Searching for multiple values one of each is not from the Categorical
+        msg = (
+            "Cannot setitem on a Categorical with a new category, "
+            "set the categories first"
+        )
+        with pytest.raises(TypeError, match=msg):
+            cat.searchsorted(["bread", "cucumber"])
+        with pytest.raises(TypeError, match=msg):
+            ser.searchsorted(["bread", "cucumber"])
+
+    def test_unique(self, ordered):
+        # GH38140
+        dtype = CategoricalDtype(["a", "b", "c"], ordered=ordered)
+
+        # categories are reordered based on value when ordered=False
+        cat = Categorical(["a", "b", "c"], dtype=dtype)
+        res = cat.unique()
+        tm.assert_categorical_equal(res, cat)
+
+        cat = Categorical(["a", "b", "a", "a"], dtype=dtype)
+        res = cat.unique()
+        tm.assert_categorical_equal(res, Categorical(["a", "b"], dtype=dtype))
+
+        cat = Categorical(["c", "a", "b", "a", "a"], dtype=dtype)
+        res = cat.unique()
+        exp_cat = Categorical(["c", "a", "b"], dtype=dtype)
+        tm.assert_categorical_equal(res, exp_cat)
+
+        # nan must be removed
+        cat = Categorical(["b", np.nan, "b", np.nan, "a"], dtype=dtype)
+        res = cat.unique()
+        exp_cat = Categorical(["b", np.nan, "a"], dtype=dtype)
+        tm.assert_categorical_equal(res, exp_cat)
+
+    def test_unique_index_series(self, ordered):
+        # GH38140
+        dtype = CategoricalDtype([3, 2, 1], ordered=ordered)
+
+        c = Categorical([3, 1, 2, 2, 1], dtype=dtype)
+        # Categorical.unique sorts categories by appearance order
+        # if ordered=False
+        exp = Categorical([3, 1, 2], dtype=dtype)
+        tm.assert_categorical_equal(c.unique(), exp)
+
+        tm.assert_index_equal(Index(c).unique(), Index(exp))
+        tm.assert_categorical_equal(Series(c).unique(), exp)
+
+        c = Categorical([1, 1, 2, 2], dtype=dtype)
+        exp = Categorical([1, 2], dtype=dtype)
+        tm.assert_categorical_equal(c.unique(), exp)
+        tm.assert_index_equal(Index(c).unique(), Index(exp))
+        tm.assert_categorical_equal(Series(c).unique(), exp)
+
+    def test_shift(self):
+        # GH 9416
+        cat = Categorical(["a", "b", "c", "d", "a"])
+
+        # shift forward
+        sp1 = cat.shift(1)
+        xp1 = Categorical([np.nan, "a", "b", "c", "d"])
+        tm.assert_categorical_equal(sp1, xp1)
+        tm.assert_categorical_equal(cat[:-1], sp1[1:])
+
+        # shift back
+        sn2 = cat.shift(-2)
+        xp2 = Categorical(
+            ["c", "d", "a", np.nan, np.nan], categories=["a", "b", "c", "d"]
+        )
+        tm.assert_categorical_equal(sn2, xp2)
+        tm.assert_categorical_equal(cat[2:], sn2[:-2])
+
+        # shift by zero
+        tm.assert_categorical_equal(cat, cat.shift(0))
+
+    def test_nbytes(self):
+        cat = Categorical([1, 2, 3])
+        exp = 3 + 3 * 8  # 3 int8s for values + 3 int64s for categories
+        assert cat.nbytes == exp
+
+    def test_memory_usage(self, using_infer_string):
+        cat = Categorical([1, 2, 3])
+
+        # .categories is an index, so we include the hashtable
+        assert 0 < cat.nbytes <= cat.memory_usage()
+        assert 0 < cat.nbytes <= cat.memory_usage(deep=True)
+
+        cat = Categorical(["foo", "foo", "bar"])
+        if using_infer_string:
+            if cat.categories.dtype.storage == "python":
+                assert cat.memory_usage(deep=True) > cat.nbytes
+            else:
+                assert cat.memory_usage(deep=True) >= cat.nbytes
+        else:
+            assert cat.memory_usage(deep=True) > cat.nbytes
+
+        if not PYPY:
+            # sys.getsizeof will call the .memory_usage with
+            # deep=True, and add on some GC overhead
+            diff = cat.memory_usage(deep=True) - sys.getsizeof(cat)
+            assert abs(diff) < 100
+
+    def test_map(self):
+        c = Categorical(list("ABABC"), categories=list("CBA"), ordered=True)
+        result = c.map(lambda x: x.lower(), na_action=None)
+        exp = Categorical(list("ababc"), categories=list("cba"), ordered=True)
+        tm.assert_categorical_equal(result, exp)
+
+        c = Categorical(list("ABABC"), categories=list("ABC"), ordered=False)
+        result = c.map(lambda x: x.lower(), na_action=None)
+        exp = Categorical(list("ababc"), categories=list("abc"), ordered=False)
+        tm.assert_categorical_equal(result, exp)
+
+        result = c.map(lambda x: 1, na_action=None)
+        # GH 12766: Return an index not an array
+        tm.assert_index_equal(result, Index(np.array([1] * 5, dtype=np.int64)))
+
+    @pytest.mark.parametrize("value", [1, "True", [1, 2, 3], 5.0])
+    def test_validate_inplace_raises(self, value):
+        cat = Categorical(["A", "B", "B", "C", "A"])
+        msg = (
+            'For argument "inplace" expected type bool, '
+            f"received type {type(value).__name__}"
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            cat.sort_values(inplace=value)
+
+    def test_quantile_empty(self):
+        # make sure we have correct itemsize on resulting codes
+        cat = Categorical(["A", "B"])
+        idx = Index([0.0, 0.5])
+        result = cat[:0]._quantile(idx, interpolation="linear")
+        assert result._codes.dtype == np.int8
+
+        expected = cat.take([-1, -1], allow_fill=True)
+        tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/categorical/test_api.py b/pandas/pandas/tests/arrays/categorical/test_api.py
new file mode 100644
index 0000000000000..2791fd55f54d7
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_api.py
@@ -0,0 +1,493 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import PY311
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    Series,
+    StringDtype,
+)
+import pandas._testing as tm
+from pandas.core.arrays.categorical import recode_for_categories
+
+
+class TestCategoricalAPI:
+    def test_ordered_api(self):
+        # GH 9347
+        cat1 = Categorical(list("acb"), ordered=False)
+        tm.assert_index_equal(cat1.categories, Index(["a", "b", "c"]))
+        assert not cat1.ordered
+
+        cat2 = Categorical(list("acb"), categories=list("bca"), ordered=False)
+        tm.assert_index_equal(cat2.categories, Index(["b", "c", "a"]))
+        assert not cat2.ordered
+
+        cat3 = Categorical(list("acb"), ordered=True)
+        tm.assert_index_equal(cat3.categories, Index(["a", "b", "c"]))
+        assert cat3.ordered
+
+        cat4 = Categorical(list("acb"), categories=list("bca"), ordered=True)
+        tm.assert_index_equal(cat4.categories, Index(["b", "c", "a"]))
+        assert cat4.ordered
+
+    def test_set_ordered(self):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        cat2 = cat.as_unordered()
+        assert not cat2.ordered
+        cat2 = cat.as_ordered()
+        assert cat2.ordered
+
+        assert cat2.set_ordered(True).ordered
+        assert not cat2.set_ordered(False).ordered
+
+        # removed in 0.19.0
+        msg = (
+            "property 'ordered' of 'Categorical' object has no setter"
+            if PY311
+            else "can't set attribute"
+        )
+        with pytest.raises(AttributeError, match=msg):
+            cat.ordered = True
+        with pytest.raises(AttributeError, match=msg):
+            cat.ordered = False
+
+    def test_rename_categories(self):
+        cat = Categorical(["a", "b", "c", "a"])
+
+        # inplace=False: the old one must not be changed
+        res = cat.rename_categories([1, 2, 3])
+        tm.assert_numpy_array_equal(
+            res.__array__(), np.array([1, 2, 3, 1], dtype=np.int64)
+        )
+        tm.assert_index_equal(res.categories, Index([1, 2, 3]))
+
+        exp_cat = np.array(["a", "b", "c", "a"], dtype=np.object_)
+        tm.assert_numpy_array_equal(cat.__array__(), exp_cat)
+
+        exp_cat = Index(["a", "b", "c"])
+        tm.assert_index_equal(cat.categories, exp_cat)
+
+        # GH18862 (let rename_categories take callables)
+        result = cat.rename_categories(lambda x: x.upper())
+        expected = Categorical(["A", "B", "C", "A"])
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("new_categories", [[1, 2, 3, 4], [1, 2]])
+    def test_rename_categories_wrong_length_raises(self, new_categories):
+        cat = Categorical(["a", "b", "c", "a"])
+        msg = (
+            "new categories need to have the same number of items as the "
+            "old categories!"
+        )
+        with pytest.raises(ValueError, match=msg):
+            cat.rename_categories(new_categories)
+
+    def test_rename_categories_series(self):
+        # https://github.com/pandas-dev/pandas/issues/17981
+        c = Categorical(["a", "b"])
+        result = c.rename_categories(Series([0, 1], index=["a", "b"]))
+        expected = Categorical([0, 1])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_rename_categories_dict(self):
+        # GH 17336
+        cat = Categorical(["a", "b", "c", "d"])
+        res = cat.rename_categories({"a": 4, "b": 3, "c": 2, "d": 1})
+        expected = Index([4, 3, 2, 1])
+        tm.assert_index_equal(res.categories, expected)
+
+        # Test for dicts of smaller length
+        cat = Categorical(["a", "b", "c", "d"])
+        res = cat.rename_categories({"a": 1, "c": 3})
+
+        expected = Index([1, "b", 3, "d"])
+        tm.assert_index_equal(res.categories, expected)
+
+        # Test for dicts with bigger length
+        cat = Categorical(["a", "b", "c", "d"])
+        res = cat.rename_categories({"a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6})
+        expected = Index([1, 2, 3, 4])
+        tm.assert_index_equal(res.categories, expected)
+
+        # Test for dicts with no items from old categories
+        cat = Categorical(["a", "b", "c", "d"])
+        res = cat.rename_categories({"f": 1, "g": 3})
+
+        expected = Index(["a", "b", "c", "d"])
+        tm.assert_index_equal(res.categories, expected)
+
+    def test_reorder_categories(self):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        old = cat.copy()
+        new = Categorical(
+            ["a", "b", "c", "a"], categories=["c", "b", "a"], ordered=True
+        )
+
+        res = cat.reorder_categories(["c", "b", "a"])
+        # cat must be the same as before
+        tm.assert_categorical_equal(cat, old)
+        # only res is changed
+        tm.assert_categorical_equal(res, new)
+
+    @pytest.mark.parametrize(
+        "new_categories",
+        [
+            ["a"],  # not all "old" included in "new"
+            ["a", "b", "d"],  # still not all "old" in "new"
+            ["a", "b", "c", "d"],  # all "old" included in "new", but too long
+        ],
+    )
+    def test_reorder_categories_raises(self, new_categories):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        msg = "items in new_categories are not the same as in old categories"
+        with pytest.raises(ValueError, match=msg):
+            cat.reorder_categories(new_categories)
+
+    def test_add_categories(self):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        old = cat.copy()
+        new = Categorical(
+            ["a", "b", "c", "a"], categories=["a", "b", "c", "d"], ordered=True
+        )
+
+        res = cat.add_categories("d")
+        tm.assert_categorical_equal(cat, old)
+        tm.assert_categorical_equal(res, new)
+
+        res = cat.add_categories(["d"])
+        tm.assert_categorical_equal(cat, old)
+        tm.assert_categorical_equal(res, new)
+
+        # GH 9927
+        cat = Categorical(list("abc"), ordered=True)
+        expected = Categorical(list("abc"), categories=list("abcde"), ordered=True)
+        # test with Series, np.array, index, list
+        res = cat.add_categories(Series(["d", "e"]))
+        tm.assert_categorical_equal(res, expected)
+        res = cat.add_categories(np.array(["d", "e"]))
+        tm.assert_categorical_equal(res, expected)
+        res = cat.add_categories(Index(["d", "e"]))
+        tm.assert_categorical_equal(res, expected)
+        res = cat.add_categories(["d", "e"])
+        tm.assert_categorical_equal(res, expected)
+
+    def test_add_categories_existing_raises(self):
+        # new is in old categories
+        cat = Categorical(["a", "b", "c", "d"], ordered=True)
+        msg = re.escape("new categories must not include old categories: {'d'}")
+        with pytest.raises(ValueError, match=msg):
+            cat.add_categories(["d"])
+
+    def test_add_categories_losing_dtype_information(self):
+        # GH#48812
+        cat = Categorical(Series([1, 2], dtype="Int64"))
+        ser = Series([4], dtype="Int64")
+        result = cat.add_categories(ser)
+        expected = Categorical(
+            Series([1, 2], dtype="Int64"), categories=Series([1, 2, 4], dtype="Int64")
+        )
+        tm.assert_categorical_equal(result, expected)
+
+        cat = Categorical(Series(["a", "b", "a"], dtype=StringDtype()))
+        ser = Series(["d"], dtype=StringDtype())
+        result = cat.add_categories(ser)
+        expected = Categorical(
+            Series(["a", "b", "a"], dtype=StringDtype()),
+            categories=Series(["a", "b", "d"], dtype=StringDtype()),
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_set_categories(self):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        exp_categories = Index(["c", "b", "a"])
+        exp_values = np.array(["a", "b", "c", "a"], dtype=np.object_)
+
+        cat = cat.set_categories(["c", "b", "a"])
+        res = cat.set_categories(["a", "b", "c"])
+        # cat must be the same as before
+        tm.assert_index_equal(cat.categories, exp_categories)
+        tm.assert_numpy_array_equal(cat.__array__(), exp_values)
+        # only res is changed
+        exp_categories_back = Index(["a", "b", "c"])
+        tm.assert_index_equal(res.categories, exp_categories_back)
+        tm.assert_numpy_array_equal(res.__array__(), exp_values)
+
+        # not all "old" included in "new" -> all not included ones are now
+        # np.nan
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        res = cat.set_categories(["a"])
+        tm.assert_numpy_array_equal(res.codes, np.array([0, -1, -1, 0], dtype=np.int8))
+
+        # still not all "old" in "new"
+        res = cat.set_categories(["a", "b", "d"])
+        tm.assert_numpy_array_equal(res.codes, np.array([0, 1, -1, 0], dtype=np.int8))
+        tm.assert_index_equal(res.categories, Index(["a", "b", "d"]))
+
+        # all "old" included in "new"
+        cat = cat.set_categories(["a", "b", "c", "d"])
+        exp_categories = Index(["a", "b", "c", "d"])
+        tm.assert_index_equal(cat.categories, exp_categories)
+
+        # internals...
+        c = Categorical([1, 2, 3, 4, 1], categories=[1, 2, 3, 4], ordered=True)
+        tm.assert_numpy_array_equal(c._codes, np.array([0, 1, 2, 3, 0], dtype=np.int8))
+        tm.assert_index_equal(c.categories, Index([1, 2, 3, 4]))
+
+        exp = np.array([1, 2, 3, 4, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(np.asarray(c), exp)
+
+        # all "pointers" to '4' must be changed from 3 to 0,...
+        c = c.set_categories([4, 3, 2, 1])
+
+        # positions are changed
+        tm.assert_numpy_array_equal(c._codes, np.array([3, 2, 1, 0, 3], dtype=np.int8))
+
+        # categories are now in new order
+        tm.assert_index_equal(c.categories, Index([4, 3, 2, 1]))
+
+        # output is the same
+        exp = np.array([1, 2, 3, 4, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(np.asarray(c), exp)
+        assert c.min() == 4
+        assert c.max() == 1
+
+        # set_categories should set the ordering if specified
+        c2 = c.set_categories([4, 3, 2, 1], ordered=False)
+        assert not c2.ordered
+
+        tm.assert_numpy_array_equal(np.asarray(c), np.asarray(c2))
+
+        # set_categories should pass thru the ordering
+        c2 = c.set_ordered(False).set_categories([4, 3, 2, 1])
+        assert not c2.ordered
+
+        tm.assert_numpy_array_equal(np.asarray(c), np.asarray(c2))
+
+    @pytest.mark.parametrize(
+        "values, categories, new_categories",
+        [
+            # No NaNs, same cats, same order
+            (["a", "b", "a"], ["a", "b"], ["a", "b"]),
+            # No NaNs, same cats, different order
+            (["a", "b", "a"], ["a", "b"], ["b", "a"]),
+            # Same, unsorted
+            (["b", "a", "a"], ["a", "b"], ["a", "b"]),
+            # No NaNs, same cats, different order
+            (["b", "a", "a"], ["a", "b"], ["b", "a"]),
+            # NaNs
+            (["a", "b", "c"], ["a", "b"], ["a", "b"]),
+            (["a", "b", "c"], ["a", "b"], ["b", "a"]),
+            (["b", "a", "c"], ["a", "b"], ["a", "b"]),
+            (["b", "a", "c"], ["a", "b"], ["b", "a"]),
+            # Introduce NaNs
+            (["a", "b", "c"], ["a", "b"], ["a"]),
+            (["a", "b", "c"], ["a", "b"], ["b"]),
+            (["b", "a", "c"], ["a", "b"], ["a"]),
+            (["b", "a", "c"], ["a", "b"], ["b"]),
+            # No overlap
+            (["a", "b", "c"], ["a", "b"], ["d", "e"]),
+        ],
+    )
+    def test_set_categories_many(self, values, categories, new_categories, ordered):
+        c = Categorical(values, categories)
+        expected = Categorical(values, new_categories, ordered)
+        result = c.set_categories(new_categories, ordered=ordered)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_set_categories_rename_less(self):
+        # GH 24675
+        cat = Categorical(["A", "B"])
+        result = cat.set_categories(["A"], rename=True)
+        expected = Categorical(["A", np.nan])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_set_categories_private(self):
+        cat = Categorical(["a", "b", "c"], categories=["a", "b", "c", "d"])
+        cat._set_categories(["a", "c", "d", "e"])
+        expected = Categorical(["a", "c", "d"], categories=list("acde"))
+        tm.assert_categorical_equal(cat, expected)
+
+        # fastpath
+        cat = Categorical(["a", "b", "c"], categories=["a", "b", "c", "d"])
+        cat._set_categories(["a", "c", "d", "e"], fastpath=True)
+        expected = Categorical(["a", "c", "d"], categories=list("acde"))
+        tm.assert_categorical_equal(cat, expected)
+
+    def test_remove_categories(self):
+        cat = Categorical(["a", "b", "c", "a"], ordered=True)
+        old = cat.copy()
+        new = Categorical(["a", "b", np.nan, "a"], categories=["a", "b"], ordered=True)
+
+        res = cat.remove_categories("c")
+        tm.assert_categorical_equal(cat, old)
+        tm.assert_categorical_equal(res, new)
+
+        res = cat.remove_categories(["c"])
+        tm.assert_categorical_equal(cat, old)
+        tm.assert_categorical_equal(res, new)
+
+    @pytest.mark.parametrize("removals", [["c"], ["c", np.nan], "c", ["c", "c"]])
+    def test_remove_categories_raises(self, removals):
+        cat = Categorical(["a", "b", "a"])
+        message = re.escape("removals must all be in old categories: {'c'}")
+
+        with pytest.raises(ValueError, match=message):
+            cat.remove_categories(removals)
+
+    def test_remove_unused_categories(self):
+        c = Categorical(["a", "b", "c", "d", "a"], categories=["a", "b", "c", "d", "e"])
+        exp_categories_all = Index(["a", "b", "c", "d", "e"])
+        exp_categories_dropped = Index(["a", "b", "c", "d"])
+
+        tm.assert_index_equal(c.categories, exp_categories_all)
+
+        res = c.remove_unused_categories()
+        tm.assert_index_equal(res.categories, exp_categories_dropped)
+        tm.assert_index_equal(c.categories, exp_categories_all)
+
+        # with NaN values (GH11599)
+        c = Categorical(["a", "b", "c", np.nan], categories=["a", "b", "c", "d", "e"])
+        res = c.remove_unused_categories()
+        tm.assert_index_equal(res.categories, Index(np.array(["a", "b", "c"])))
+        exp_codes = np.array([0, 1, 2, -1], dtype=np.int8)
+        tm.assert_numpy_array_equal(res.codes, exp_codes)
+        tm.assert_index_equal(c.categories, exp_categories_all)
+
+        val = ["F", np.nan, "D", "B", "D", "F", np.nan]
+        cat = Categorical(values=val, categories=list("ABCDEFG"))
+        out = cat.remove_unused_categories()
+        tm.assert_index_equal(out.categories, Index(["B", "D", "F"]))
+        exp_codes = np.array([2, -1, 1, 0, 1, 2, -1], dtype=np.int8)
+        tm.assert_numpy_array_equal(out.codes, exp_codes)
+        assert out.tolist() == val
+
+        alpha = list("abcdefghijklmnopqrstuvwxyz")
+        val = np.random.default_rng(2).choice(alpha[::2], 10000).astype("object")
+        val[np.random.default_rng(2).choice(len(val), 100)] = np.nan
+
+        cat = Categorical(values=val, categories=alpha)
+        out = cat.remove_unused_categories()
+        assert out.tolist() == val.tolist()
+
+
+class TestCategoricalAPIWithFactor:
+    def test_describe(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        # string type
+        desc = factor.describe()
+        assert factor.ordered
+        exp_index = CategoricalIndex(
+            ["a", "b", "c"], name="categories", ordered=factor.ordered
+        )
+        expected = DataFrame(
+            {"counts": [3, 2, 3], "freqs": [3 / 8.0, 2 / 8.0, 3 / 8.0]}, index=exp_index
+        )
+        tm.assert_frame_equal(desc, expected)
+
+        # check unused categories
+        cat = factor.copy()
+        cat = cat.set_categories(["a", "b", "c", "d"])
+        desc = cat.describe()
+
+        exp_index = CategoricalIndex(
+            list("abcd"), ordered=factor.ordered, name="categories"
+        )
+        expected = DataFrame(
+            {"counts": [3, 2, 3, 0], "freqs": [3 / 8.0, 2 / 8.0, 3 / 8.0, 0]},
+            index=exp_index,
+        )
+        tm.assert_frame_equal(desc, expected)
+
+        # check an integer one
+        cat = Categorical([1, 2, 3, 1, 2, 3, 3, 2, 1, 1, 1])
+        desc = cat.describe()
+        exp_index = CategoricalIndex([1, 2, 3], ordered=cat.ordered, name="categories")
+        expected = DataFrame(
+            {"counts": [5, 3, 3], "freqs": [5 / 11.0, 3 / 11.0, 3 / 11.0]},
+            index=exp_index,
+        )
+        tm.assert_frame_equal(desc, expected)
+
+        # https://github.com/pandas-dev/pandas/issues/3678
+        # describe should work with NaN
+        cat = Categorical([np.nan, 1, 2, 2])
+        desc = cat.describe()
+        expected = DataFrame(
+            {"counts": [1, 2, 1], "freqs": [1 / 4.0, 2 / 4.0, 1 / 4.0]},
+            index=CategoricalIndex(
+                [1, 2, np.nan], categories=[1, 2], name="categories"
+            ),
+        )
+        tm.assert_frame_equal(desc, expected)
+
+
+class TestPrivateCategoricalAPI:
+    def test_codes_immutable(self):
+        # Codes should be read only
+        c = Categorical(["a", "b", "c", "a", np.nan])
+        exp = np.array([0, 1, 2, 0, -1], dtype="int8")
+        tm.assert_numpy_array_equal(c.codes, exp)
+
+        # Assignments to codes should raise
+        msg = (
+            "property 'codes' of 'Categorical' object has no setter"
+            if PY311
+            else "can't set attribute"
+        )
+        with pytest.raises(AttributeError, match=msg):
+            c.codes = np.array([0, 1, 2, 0, 1], dtype="int8")
+
+        # changes in the codes array should raise
+        codes = c.codes
+
+        with pytest.raises(ValueError, match="assignment destination is read-only"):
+            codes[4] = 1
+
+        # But even after getting the codes, the original array should still be
+        # writeable!
+        c[4] = "a"
+        exp = np.array([0, 1, 2, 0, 0], dtype="int8")
+        tm.assert_numpy_array_equal(c.codes, exp)
+        c._codes[4] = 2
+        exp = np.array([0, 1, 2, 0, 2], dtype="int8")
+        tm.assert_numpy_array_equal(c.codes, exp)
+
+    @pytest.mark.parametrize(
+        "codes, old, new, expected",
+        [
+            ([0, 1], ["a", "b"], ["a", "b"], [0, 1]),
+            ([0, 1], ["b", "a"], ["b", "a"], [0, 1]),
+            ([0, 1], ["a", "b"], ["b", "a"], [1, 0]),
+            ([0, 1], ["b", "a"], ["a", "b"], [1, 0]),
+            ([0, 1, 0, 1], ["a", "b"], ["a", "b", "c"], [0, 1, 0, 1]),
+            ([0, 1, 2, 2], ["a", "b", "c"], ["a", "b"], [0, 1, -1, -1]),
+            ([0, 1, -1], ["a", "b", "c"], ["a", "b", "c"], [0, 1, -1]),
+            ([0, 1, -1], ["a", "b", "c"], ["b"], [-1, 0, -1]),
+            ([0, 1, -1], ["a", "b", "c"], ["d"], [-1, -1, -1]),
+            ([0, 1, -1], ["a", "b", "c"], [], [-1, -1, -1]),
+            ([-1, -1], [], ["a", "b"], [-1, -1]),
+            ([1, 0], ["b", "a"], ["a", "b"], [0, 1]),
+        ],
+    )
+    def test_recode_to_categories(self, codes, old, new, expected):
+        codes = np.asanyarray(codes, dtype=np.int8)
+        expected = np.asanyarray(expected, dtype=np.int8)
+        old = Index(old)
+        new = Index(new)
+        result = recode_for_categories(codes, old, new)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_recode_to_categories_large(self):
+        N = 1000
+        codes = np.arange(N)
+        old = Index(codes)
+        expected = np.arange(N - 1, -1, -1, dtype=np.int16)
+        new = Index(expected)
+        result = recode_for_categories(codes, old, new)
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/categorical/test_astype.py b/pandas/pandas/tests/arrays/categorical/test_astype.py
new file mode 100644
index 0000000000000..00999d491b242
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_astype.py
@@ -0,0 +1,153 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DatetimeIndex,
+    Interval,
+    NaT,
+    Period,
+    Timestamp,
+    array,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+class TestAstype:
+    @pytest.mark.parametrize("cls", [Categorical, CategoricalIndex])
+    @pytest.mark.parametrize("values", [[1, np.nan], [Timestamp("2000"), NaT]])
+    def test_astype_nan_to_int(self, cls, values):
+        # GH#28406
+        obj = cls(values)
+
+        msg = "Cannot (cast|convert)"
+        with pytest.raises((ValueError, TypeError), match=msg):
+            obj.astype(int)
+
+    @pytest.mark.parametrize(
+        "expected",
+        [
+            array(["2019", "2020"], dtype="datetime64[ns, UTC]"),
+            array([0, 0], dtype="timedelta64[ns]"),
+            array([Period("2019"), Period("2020")], dtype="period[Y-DEC]"),
+            array([Interval(0, 1), Interval(1, 2)], dtype="interval"),
+            array([1, np.nan], dtype="Int64"),
+        ],
+    )
+    def test_astype_category_to_extension_dtype(self, expected):
+        # GH#28668
+        result = expected.astype("category").astype(expected.dtype)
+
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, expected",
+        [
+            (
+                "datetime64[ns]",
+                np.array(["2015-01-01T00:00:00.000000000"], dtype="datetime64[ns]"),
+            ),
+            (
+                "datetime64[ns, MET]",
+                DatetimeIndex([Timestamp("2015-01-01 00:00:00+0100", tz="MET")]).array,
+            ),
+        ],
+    )
+    def test_astype_to_datetime64(self, dtype, expected):
+        # GH#28448
+        result = Categorical(["2015-01-01"]).astype(dtype)
+        assert result == expected
+
+    def test_astype_str_int_categories_to_nullable_int(self):
+        # GH#39616
+        dtype = CategoricalDtype([str(i) for i in range(5)])
+        codes = np.random.default_rng(2).integers(5, size=20)
+        arr = Categorical.from_codes(codes, dtype=dtype)
+
+        res = arr.astype("Int64")
+        expected = array(codes, dtype="Int64")
+        tm.assert_extension_array_equal(res, expected)
+
+    def test_astype_str_int_categories_to_nullable_float(self):
+        # GH#39616
+        dtype = CategoricalDtype([str(i / 2) for i in range(5)])
+        codes = np.random.default_rng(2).integers(5, size=20)
+        arr = Categorical.from_codes(codes, dtype=dtype)
+
+        res = arr.astype("Float64")
+        expected = array(codes, dtype="Float64") / 2
+        tm.assert_extension_array_equal(res, expected)
+
+    def test_astype(self, ordered):
+        # string
+        cat = Categorical(list("abbaaccc"), ordered=ordered)
+        result = cat.astype(object)
+        expected = np.array(cat)
+        tm.assert_numpy_array_equal(result, expected)
+
+        msg = r"Cannot cast object|str dtype to float64"
+        with pytest.raises(ValueError, match=msg):
+            cat.astype(float)
+
+        # numeric
+        cat = Categorical([0, 1, 2, 2, 1, 0, 1, 0, 2], ordered=ordered)
+        result = cat.astype(object)
+        expected = np.array(cat, dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = cat.astype(int)
+        expected = np.array(cat, dtype="int")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = cat.astype(float)
+        expected = np.array(cat, dtype=float)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype_ordered", [True, False])
+    def test_astype_category(self, dtype_ordered, ordered):
+        # GH#10696/GH#18593
+        data = list("abcaacbab")
+        cat = Categorical(data, categories=list("bac"), ordered=ordered)
+
+        # standard categories
+        dtype = CategoricalDtype(ordered=dtype_ordered)
+        result = cat.astype(dtype)
+        expected = Categorical(data, categories=cat.categories, ordered=dtype_ordered)
+        tm.assert_categorical_equal(result, expected)
+
+        # non-standard categories
+        dtype = CategoricalDtype(list("adc"), dtype_ordered)
+        result = cat.astype(dtype)
+        expected = Categorical(data, dtype=dtype)
+        tm.assert_categorical_equal(result, expected)
+
+        if dtype_ordered is False:
+            # dtype='category' can't specify ordered, so only test once
+            result = cat.astype("category")
+            expected = cat
+            tm.assert_categorical_equal(result, expected)
+
+    def test_astype_object_datetime_categories(self):
+        # GH#40754
+        cat = Categorical(to_datetime(["2021-03-27", NaT]))
+        result = cat.astype(object)
+        expected = np.array([Timestamp("2021-03-27 00:00:00"), NaT], dtype="object")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_object_timestamp_categories(self):
+        # GH#18024
+        cat = Categorical([Timestamp("2014-01-01")])
+        result = cat.astype(object)
+        expected = np.array([Timestamp("2014-01-01 00:00:00")], dtype="object")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_category_readonly_mask_values(self):
+        # GH#53658
+        arr = array([0, 1, 2], dtype="Int64")
+        arr._mask.flags["WRITEABLE"] = False
+        result = arr.astype("category")
+        expected = array([0, 1, 2], dtype="Int64").astype("category")
+        tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/categorical/test_constructors.py b/pandas/pandas/tests/arrays/categorical/test_constructors.py
new file mode 100644
index 0000000000000..d7eb6800e5d07
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_constructors.py
@@ -0,0 +1,793 @@
+from datetime import (
+    date,
+    datetime,
+)
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import HAS_PYARROW
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_integer_dtype,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    RangeIndex,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalConstructors:
+    def test_categorical_from_cat_and_dtype_str_preserve_ordered(self):
+        # GH#49309 we should preserve orderedness in `res`
+        cat = Categorical([3, 1], categories=[3, 2, 1], ordered=True)
+
+        res = Categorical(cat, dtype="category")
+        assert res.dtype.ordered
+
+    def test_categorical_disallows_scalar(self):
+        # GH#38433
+        with pytest.raises(TypeError, match="Categorical input must be list-like"):
+            Categorical("A", categories=["A", "B"])
+
+    def test_categorical_1d_only(self):
+        # ndim > 1
+        msg = "> 1 ndim Categorical are not supported at this time"
+        with pytest.raises(NotImplementedError, match=msg):
+            Categorical(np.array([list("abcd")]))
+
+    def test_validate_ordered(self):
+        # see gh-14058
+        exp_msg = "'ordered' must either be 'True' or 'False'"
+        exp_err = TypeError
+
+        # This should be a boolean.
+        ordered = np.array([0, 1, 2])
+
+        with pytest.raises(exp_err, match=exp_msg):
+            Categorical([1, 2, 3], ordered=ordered)
+
+        with pytest.raises(exp_err, match=exp_msg):
+            Categorical.from_codes(
+                [0, 0, 1], categories=["a", "b", "c"], ordered=ordered
+            )
+
+    def test_constructor_empty(self):
+        # GH 17248
+        c = Categorical([])
+        expected = Index([])
+        tm.assert_index_equal(c.categories, expected)
+
+        c = Categorical([], categories=[1, 2, 3])
+        expected = Index([1, 2, 3], dtype=np.int64)
+        tm.assert_index_equal(c.categories, expected)
+
+    def test_constructor_empty_boolean(self):
+        # see gh-22702
+        cat = Categorical([], categories=[True, False])
+        categories = sorted(cat.categories.tolist())
+        assert categories == [False, True]
+
+    def test_constructor_tuples(self):
+        values = np.array([(1,), (1, 2), (1,), (1, 2)], dtype=object)
+        result = Categorical(values)
+        expected = Index([(1,), (1, 2)], tupleize_cols=False)
+        tm.assert_index_equal(result.categories, expected)
+        assert result.ordered is False
+
+    def test_constructor_tuples_datetimes(self):
+        # numpy will auto reshape when all of the tuples are the
+        # same len, so add an extra one with 2 items and slice it off
+        values = np.array(
+            [
+                (Timestamp("2010-01-01"),),
+                (Timestamp("2010-01-02"),),
+                (Timestamp("2010-01-01"),),
+                (Timestamp("2010-01-02"),),
+                ("a", "b"),
+            ],
+            dtype=object,
+        )[:-1]
+        result = Categorical(values)
+        expected = Index(
+            [(Timestamp("2010-01-01"),), (Timestamp("2010-01-02"),)],
+            tupleize_cols=False,
+        )
+        tm.assert_index_equal(result.categories, expected)
+
+    def test_constructor_unsortable(self):
+        # it works!
+        arr = np.array([1, 2, 3, datetime.now()], dtype="O")
+        factor = Categorical(arr, ordered=False)
+        assert not factor.ordered
+
+        # this however will raise as cannot be sorted
+        msg = (
+            "'values' is not ordered, please explicitly specify the "
+            "categories order by passing in a categories argument."
+        )
+        with pytest.raises(TypeError, match=msg):
+            Categorical(arr, ordered=True)
+
+    def test_constructor_interval(self):
+        result = Categorical(
+            [Interval(1, 2), Interval(2, 3), Interval(3, 6)], ordered=True
+        )
+        ii = IntervalIndex([Interval(1, 2), Interval(2, 3), Interval(3, 6)])
+        exp = Categorical(ii, ordered=True)
+        tm.assert_categorical_equal(result, exp)
+        tm.assert_index_equal(result.categories, ii)
+
+    def test_constructor(self):
+        exp_arr = np.array(["a", "b", "c", "a", "b", "c"], dtype=np.object_)
+        c1 = Categorical(exp_arr)
+        tm.assert_numpy_array_equal(c1.__array__(), exp_arr)
+        c2 = Categorical(exp_arr, categories=["a", "b", "c"])
+        tm.assert_numpy_array_equal(c2.__array__(), exp_arr)
+        c2 = Categorical(exp_arr, categories=["c", "b", "a"])
+        tm.assert_numpy_array_equal(c2.__array__(), exp_arr)
+
+        # categories must be unique
+        msg = "Categorical categories must be unique"
+        with pytest.raises(ValueError, match=msg):
+            Categorical([1, 2], [1, 2, 2])
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical(["a", "b"], ["a", "b", "b"])
+
+        # The default should be unordered
+        c1 = Categorical(["a", "b", "c", "a"])
+        assert not c1.ordered
+
+        # Categorical as input
+        c1 = Categorical(["a", "b", "c", "a"])
+        c2 = Categorical(c1)
+        tm.assert_categorical_equal(c1, c2)
+
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "b", "c", "d"])
+        c2 = Categorical(c1)
+        tm.assert_categorical_equal(c1, c2)
+
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "c", "b"])
+        c2 = Categorical(c1)
+        tm.assert_categorical_equal(c1, c2)
+
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "c", "b"])
+        c2 = Categorical(c1, categories=["a", "b", "c"])
+        tm.assert_numpy_array_equal(c1.__array__(), c2.__array__())
+        tm.assert_index_equal(c2.categories, Index(["a", "b", "c"]))
+
+        # Series of dtype category
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "b", "c", "d"])
+        c2 = Categorical(Series(c1))
+        tm.assert_categorical_equal(c1, c2)
+
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "c", "b"])
+        c2 = Categorical(Series(c1))
+        tm.assert_categorical_equal(c1, c2)
+
+        # Series
+        c1 = Categorical(["a", "b", "c", "a"])
+        c2 = Categorical(Series(["a", "b", "c", "a"]))
+        tm.assert_categorical_equal(c1, c2)
+
+        c1 = Categorical(["a", "b", "c", "a"], categories=["a", "b", "c", "d"])
+        c2 = Categorical(Series(["a", "b", "c", "a"]), categories=["a", "b", "c", "d"])
+        tm.assert_categorical_equal(c1, c2)
+
+        # This should result in integer categories, not float!
+        cat = Categorical([1, 2, 3, np.nan], categories=[1, 2, 3])
+        assert is_integer_dtype(cat.categories)
+
+        # https://github.com/pandas-dev/pandas/issues/3678
+        cat = Categorical([np.nan, 1, 2, 3])
+        assert is_integer_dtype(cat.categories)
+
+        # this should result in floats
+        cat = Categorical([np.nan, 1, 2.0, 3])
+        assert is_float_dtype(cat.categories)
+
+        cat = Categorical([np.nan, 1.0, 2.0, 3.0])
+        assert is_float_dtype(cat.categories)
+
+        # This doesn't work -> this would probably need some kind of "remember
+        # the original type" feature to try to cast the array interface result
+        # to...
+
+        # vals = np.asarray(cat[cat.notna()])
+        # assert is_integer_dtype(vals)
+
+        # corner cases
+        cat = Categorical([1])
+        assert len(cat.categories) == 1
+        assert cat.categories[0] == 1
+        assert len(cat.codes) == 1
+        assert cat.codes[0] == 0
+
+        cat = Categorical(["a"])
+        assert len(cat.categories) == 1
+        assert cat.categories[0] == "a"
+        assert len(cat.codes) == 1
+        assert cat.codes[0] == 0
+
+        # two arrays
+        #  - when the first is an integer dtype and the second is not
+        #  - when the resulting codes are all -1/NaN
+        with tm.assert_produces_warning(None):
+            Categorical([0, 1, 2, 0, 1, 2], categories=["a", "b", "c"])
+
+        with tm.assert_produces_warning(None):
+            Categorical([0, 1, 2, 0, 1, 2], categories=[3, 4, 5])
+
+        # the next one are from the old docs
+        with tm.assert_produces_warning(None):
+            Categorical([0, 1, 2, 0, 1, 2], [1, 2, 3])
+            cat = Categorical([1, 2], categories=[1, 2, 3])
+
+        # this is a legitimate constructor
+        with tm.assert_produces_warning(None):
+            Categorical(np.array([], dtype="int64"), categories=[3, 2, 1], ordered=True)
+
+    def test_constructor_with_existing_categories(self):
+        # GH25318: constructing with pd.Series used to bogusly skip recoding
+        # categories
+        c0 = Categorical(["a", "b", "c", "a"])
+        c1 = Categorical(["a", "b", "c", "a"], categories=["b", "c"])
+
+        c2 = Categorical(c0, categories=c1.categories)
+        tm.assert_categorical_equal(c1, c2)
+
+        c3 = Categorical(Series(c0), categories=c1.categories)
+        tm.assert_categorical_equal(c1, c3)
+
+    def test_constructor_not_sequence(self):
+        # https://github.com/pandas-dev/pandas/issues/16022
+        msg = r"^Parameter 'categories' must be list-like, was"
+        with pytest.raises(TypeError, match=msg):
+            Categorical(["a", "b"], categories="a")
+
+    def test_constructor_with_null(self):
+        # Cannot have NaN in categories
+        msg = "Categorical categories cannot be null"
+        with pytest.raises(ValueError, match=msg):
+            Categorical([np.nan, "a", "b", "c"], categories=[np.nan, "a", "b", "c"])
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical([None, "a", "b", "c"], categories=[None, "a", "b", "c"])
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical(
+                DatetimeIndex(["nat", "20160101"]),
+                categories=[NaT, Timestamp("20160101")],
+            )
+
+    def test_constructor_with_index(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"))
+        tm.assert_categorical_equal(ci.values, Categorical(ci))
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"))
+        tm.assert_categorical_equal(
+            ci.values, Categorical(ci.astype(object), categories=ci.categories)
+        )
+
+    def test_constructor_with_generator(self):
+        # This was raising an Error in isna(single_val).any() because isna
+        # returned a scalar for a generator
+
+        exp = Categorical([0, 1, 2])
+        cat = Categorical(x for x in [0, 1, 2])
+        tm.assert_categorical_equal(cat, exp)
+        cat = Categorical(range(3))
+        tm.assert_categorical_equal(cat, exp)
+
+        MultiIndex.from_product([range(5), ["a", "b", "c"]])
+
+        # check that categories accept generators and sequences
+        cat = Categorical([0, 1, 2], categories=(x for x in [0, 1, 2]))
+        tm.assert_categorical_equal(cat, exp)
+        cat = Categorical([0, 1, 2], categories=range(3))
+        tm.assert_categorical_equal(cat, exp)
+
+    def test_constructor_with_rangeindex(self):
+        # RangeIndex is preserved in Categories
+        rng = Index(range(3))
+
+        cat = Categorical(rng)
+        tm.assert_index_equal(cat.categories, rng, exact=True)
+
+        cat = Categorical([1, 2, 0], categories=rng)
+        tm.assert_index_equal(cat.categories, rng, exact=True)
+
+    @pytest.mark.parametrize(
+        "dtl",
+        [
+            date_range("1995-01-01 00:00:00", periods=5, freq="s"),
+            date_range("1995-01-01 00:00:00", periods=5, freq="s", tz="US/Eastern"),
+            timedelta_range("1 day", periods=5, freq="s"),
+        ],
+    )
+    def test_constructor_with_datetimelike(self, dtl):
+        # see gh-12077
+        # constructor with a datetimelike and NaT
+
+        s = Series(dtl)
+        c = Categorical(s)
+
+        expected = type(dtl)(s)
+        expected._data.freq = None
+
+        tm.assert_index_equal(c.categories, expected)
+        tm.assert_numpy_array_equal(c.codes, np.arange(5, dtype="int8"))
+
+        # with NaT
+        s2 = s.copy()
+        s2.iloc[-1] = NaT
+        c = Categorical(s2)
+
+        expected = type(dtl)(s2.dropna())
+        expected._data.freq = None
+
+        tm.assert_index_equal(c.categories, expected)
+
+        exp = np.array([0, 1, 2, 3, -1], dtype=np.int8)
+        tm.assert_numpy_array_equal(c.codes, exp)
+
+        result = repr(c)
+        assert "NaT" in result
+
+    def test_constructor_from_index_series_datetimetz(self):
+        idx = date_range("2015-01-01 10:00", freq="D", periods=3, tz="US/Eastern")
+        idx = idx._with_freq(None)  # freq not preserved in result.categories
+        result = Categorical(idx)
+        tm.assert_index_equal(result.categories, idx)
+
+        result = Categorical(Series(idx))
+        tm.assert_index_equal(result.categories, idx)
+
+    def test_constructor_date_objects(self):
+        # we dont cast date objects to timestamps, matching Index constructor
+        v = date.today()
+
+        cat = Categorical([v, v])
+        assert cat.categories.dtype == object
+        assert type(cat.categories[0]) is date
+
+    def test_constructor_from_index_series_timedelta(self):
+        idx = timedelta_range("1 days", freq="D", periods=3)
+        idx = idx._with_freq(None)  # freq not preserved in result.categories
+        result = Categorical(idx)
+        tm.assert_index_equal(result.categories, idx)
+
+        result = Categorical(Series(idx))
+        tm.assert_index_equal(result.categories, idx)
+
+    def test_constructor_from_index_series_period(self):
+        idx = period_range("2015-01-01", freq="D", periods=3)
+        result = Categorical(idx)
+        tm.assert_index_equal(result.categories, idx)
+
+        result = Categorical(Series(idx))
+        tm.assert_index_equal(result.categories, idx)
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            np.array([1.0, 1.2, 1.8, np.nan]),
+            np.array([1, 2, 3], dtype="int64"),
+            ["a", "b", "c", np.nan],
+            [pd.Period("2014-01"), pd.Period("2014-02"), NaT],
+            [Timestamp("2014-01-01"), Timestamp("2014-01-02"), NaT],
+            [
+                Timestamp("2014-01-01", tz="US/Eastern"),
+                Timestamp("2014-01-02", tz="US/Eastern"),
+                NaT,
+            ],
+        ],
+    )
+    def test_constructor_invariant(self, values):
+        # GH 14190
+        c = Categorical(values)
+        c2 = Categorical(c)
+        tm.assert_categorical_equal(c, c2)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_constructor_with_dtype(self, ordered):
+        categories = ["b", "a", "c"]
+        dtype = CategoricalDtype(categories, ordered=ordered)
+        result = Categorical(["a", "b", "a", "c"], dtype=dtype)
+        expected = Categorical(
+            ["a", "b", "a", "c"], categories=categories, ordered=ordered
+        )
+        tm.assert_categorical_equal(result, expected)
+        assert result.ordered is ordered
+
+    def test_constructor_dtype_and_others_raises(self):
+        dtype = CategoricalDtype(["a", "b"], ordered=True)
+        msg = "Cannot specify `categories` or `ordered` together with `dtype`."
+        with pytest.raises(ValueError, match=msg):
+            Categorical(["a", "b"], categories=["a", "b"], dtype=dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical(["a", "b"], ordered=True, dtype=dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical(["a", "b"], ordered=False, dtype=dtype)
+
+    @pytest.mark.parametrize("categories", [None, ["a", "b"], ["a", "c"]])
+    def test_constructor_str_category(self, categories, ordered):
+        result = Categorical(
+            ["a", "b"], categories=categories, ordered=ordered, dtype="category"
+        )
+        expected = Categorical(["a", "b"], categories=categories, ordered=ordered)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_constructor_str_unknown(self):
+        with pytest.raises(ValueError, match="Unknown dtype"):
+            Categorical([1, 2], dtype="foo")
+
+    @pytest.mark.xfail(
+        using_string_dtype() and HAS_PYARROW, reason="Can't be NumPy strings"
+    )
+    def test_constructor_np_strs(self):
+        # GH#31499 Hashtable.map_locations needs to work on np.str_ objects
+        cat = Categorical(["1", "0", "1"], [np.str_("0"), np.str_("1")])
+        assert all(isinstance(x, np.str_) for x in cat.categories)
+
+    def test_constructor_from_categorical_with_dtype(self):
+        dtype = CategoricalDtype(["a", "b", "c"], ordered=True)
+        values = Categorical(["a", "b", "d"])
+        result = Categorical(values, dtype=dtype)
+        # We use dtype.categories, not values.categories
+        expected = Categorical(
+            ["a", "b", "d"], categories=["a", "b", "c"], ordered=True
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_constructor_from_categorical_with_unknown_dtype(self):
+        dtype = CategoricalDtype(None, ordered=True)
+        values = Categorical(["a", "b", "d"])
+        result = Categorical(values, dtype=dtype)
+        # We use values.categories, not dtype.categories
+        expected = Categorical(
+            ["a", "b", "d"], categories=["a", "b", "d"], ordered=True
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_constructor_from_categorical_string(self):
+        values = Categorical(["a", "b", "d"])
+        # use categories, ordered
+        result = Categorical(
+            values, categories=["a", "b", "c"], ordered=True, dtype="category"
+        )
+        expected = Categorical(
+            ["a", "b", "d"], categories=["a", "b", "c"], ordered=True
+        )
+        tm.assert_categorical_equal(result, expected)
+
+        # No string
+        result = Categorical(values, categories=["a", "b", "c"], ordered=True)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_constructor_with_categorical_categories(self):
+        # GH17884
+        expected = Categorical(["a", "b"], categories=["a", "b", "c"])
+
+        result = Categorical(["a", "b"], categories=Categorical(["a", "b", "c"]))
+        tm.assert_categorical_equal(result, expected)
+
+        result = Categorical(["a", "b"], categories=CategoricalIndex(["a", "b", "c"]))
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [lambda x: np.array(x, dtype=object), list])
+    def test_construction_with_null(self, klass, nulls_fixture):
+        # https://github.com/pandas-dev/pandas/issues/31927
+        values = klass(["a", nulls_fixture, "b"])
+        result = Categorical(values)
+
+        dtype = CategoricalDtype(["a", "b"])
+        codes = [0, -1, 1]
+        expected = Categorical.from_codes(codes=codes, dtype=dtype)
+
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("validate", [True, False])
+    def test_from_codes_nullable_int_categories(self, any_numeric_ea_dtype, validate):
+        # GH#39649
+        cats = pd.array(range(5), dtype=any_numeric_ea_dtype)
+        codes = np.random.default_rng(2).integers(5, size=3)
+        dtype = CategoricalDtype(cats)
+        arr = Categorical.from_codes(codes, dtype=dtype, validate=validate)
+        assert arr.categories.dtype == cats.dtype
+        tm.assert_index_equal(arr.categories, Index(cats))
+
+    def test_from_codes_empty(self):
+        cat = ["a", "b", "c"]
+        result = Categorical.from_codes([], categories=cat)
+        expected = Categorical([], categories=cat)
+
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("validate", [True, False])
+    def test_from_codes_validate(self, validate):
+        # GH53122
+        dtype = CategoricalDtype(["a", "b"])
+        if validate:
+            with pytest.raises(ValueError, match="codes need to be between "):
+                Categorical.from_codes([4, 5], dtype=dtype, validate=validate)
+        else:
+            # passes, though has incorrect codes, but that's the user responsibility
+            Categorical.from_codes([4, 5], dtype=dtype, validate=validate)
+
+    def test_from_codes_too_few_categories(self):
+        dtype = CategoricalDtype(categories=[1, 2])
+        msg = "codes need to be between "
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes([1, 2], categories=dtype.categories)
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes([1, 2], dtype=dtype)
+
+    def test_from_codes_non_int_codes(self):
+        dtype = CategoricalDtype(categories=[1, 2])
+        msg = "codes need to be array-like integers"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(["a"], categories=dtype.categories)
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(["a"], dtype=dtype)
+
+    def test_from_codes_non_unique_categories(self):
+        with pytest.raises(ValueError, match="Categorical categories must be unique"):
+            Categorical.from_codes([0, 1, 2], categories=["a", "a", "b"])
+
+    def test_from_codes_nan_cat_included(self):
+        with pytest.raises(ValueError, match="Categorical categories cannot be null"):
+            Categorical.from_codes([0, 1, 2], categories=["a", "b", np.nan])
+
+    def test_from_codes_too_negative(self):
+        dtype = CategoricalDtype(categories=["a", "b", "c"])
+        msg = r"codes need to be between -1 and len\(categories\)-1"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes([-2, 1, 2], categories=dtype.categories)
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes([-2, 1, 2], dtype=dtype)
+
+    def test_from_codes(self):
+        dtype = CategoricalDtype(categories=["a", "b", "c"])
+        exp = Categorical(["a", "b", "c"], ordered=False)
+        res = Categorical.from_codes([0, 1, 2], categories=dtype.categories)
+        tm.assert_categorical_equal(exp, res)
+
+        res = Categorical.from_codes([0, 1, 2], dtype=dtype)
+        tm.assert_categorical_equal(exp, res)
+
+    @pytest.mark.parametrize("klass", [Categorical, CategoricalIndex])
+    def test_from_codes_with_categorical_categories(self, klass):
+        # GH17884
+        expected = Categorical(["a", "b"], categories=["a", "b", "c"])
+
+        result = Categorical.from_codes([0, 1], categories=klass(["a", "b", "c"]))
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [Categorical, CategoricalIndex])
+    def test_from_codes_with_non_unique_categorical_categories(self, klass):
+        with pytest.raises(ValueError, match="Categorical categories must be unique"):
+            Categorical.from_codes([0, 1], klass(["a", "b", "a"]))
+
+    def test_from_codes_with_nan_code(self):
+        # GH21767
+        codes = [1, 2, np.nan]
+        dtype = CategoricalDtype(categories=["a", "b", "c"])
+        with pytest.raises(ValueError, match="codes need to be array-like integers"):
+            Categorical.from_codes(codes, categories=dtype.categories)
+        with pytest.raises(ValueError, match="codes need to be array-like integers"):
+            Categorical.from_codes(codes, dtype=dtype)
+
+    @pytest.mark.parametrize("codes", [[1.0, 2.0, 0], [1.1, 2.0, 0]])
+    def test_from_codes_with_float(self, codes):
+        # GH21767
+        # float codes should raise even if values are equal to integers
+        dtype = CategoricalDtype(categories=["a", "b", "c"])
+
+        msg = "codes need to be array-like integers"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(codes, dtype.categories)
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(codes, dtype=dtype)
+
+    def test_from_codes_with_dtype_raises(self):
+        msg = "Cannot specify"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(
+                [0, 1], categories=["a", "b"], dtype=CategoricalDtype(["a", "b"])
+            )
+
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(
+                [0, 1], ordered=True, dtype=CategoricalDtype(["a", "b"])
+            )
+
+    def test_from_codes_neither(self):
+        msg = "Both were None"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes([0, 1])
+
+    def test_from_codes_with_nullable_int(self):
+        codes = pd.array([0, 1], dtype="Int64")
+        categories = ["a", "b"]
+
+        result = Categorical.from_codes(codes, categories=categories)
+        expected = Categorical.from_codes(codes.to_numpy(int), categories=categories)
+
+        tm.assert_categorical_equal(result, expected)
+
+    def test_from_codes_with_nullable_int_na_raises(self):
+        codes = pd.array([0, None], dtype="Int64")
+        categories = ["a", "b"]
+
+        msg = "codes cannot contain NA values"
+        with pytest.raises(ValueError, match=msg):
+            Categorical.from_codes(codes, categories=categories)
+
+    @pytest.mark.parametrize("dtype", [None, "category"])
+    def test_from_inferred_categories(self, dtype):
+        cats = ["a", "b"]
+        codes = np.array([0, 0, 1, 1], dtype="i8")
+        result = Categorical._from_inferred_categories(cats, codes, dtype)
+        expected = Categorical.from_codes(codes, cats)
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [None, "category"])
+    def test_from_inferred_categories_sorts(self, dtype):
+        cats = ["b", "a"]
+        codes = np.array([0, 1, 1, 1], dtype="i8")
+        result = Categorical._from_inferred_categories(cats, codes, dtype)
+        expected = Categorical.from_codes([1, 0, 0, 0], ["a", "b"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_from_inferred_categories_dtype(self):
+        cats = ["a", "b", "d"]
+        codes = np.array([0, 1, 0, 2], dtype="i8")
+        dtype = CategoricalDtype(["c", "b", "a"], ordered=True)
+        result = Categorical._from_inferred_categories(cats, codes, dtype)
+        expected = Categorical(
+            ["a", "b", "a", "d"], categories=["c", "b", "a"], ordered=True
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_from_inferred_categories_coerces(self):
+        cats = ["1", "2", "bad"]
+        codes = np.array([0, 0, 1, 2], dtype="i8")
+        dtype = CategoricalDtype([1, 2])
+        result = Categorical._from_inferred_categories(cats, codes, dtype)
+        expected = Categorical([1, 1, 2, np.nan])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_construction_with_ordered(self, ordered):
+        # GH 9347, 9190
+        cat = Categorical([0, 1, 2], ordered=ordered)
+        assert cat.ordered == bool(ordered)
+
+    def test_constructor_imaginary(self):
+        values = [1, 2, 3 + 1j]
+        c1 = Categorical(values)
+        tm.assert_index_equal(c1.categories, Index(values))
+        tm.assert_numpy_array_equal(np.array(c1), np.array(values))
+
+    def test_constructor_string_and_tuples(self):
+        # GH 21416
+        c = Categorical(np.array(["c", ("a", "b"), ("b", "a"), "c"], dtype=object))
+        expected_index = Index([("a", "b"), ("b", "a"), "c"])
+        assert c.categories.equals(expected_index)
+
+    def test_interval(self):
+        idx = pd.interval_range(0, 10, periods=10)
+        cat = Categorical(idx, categories=idx)
+        expected_codes = np.arange(10, dtype="int8")
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+        # infer categories
+        cat = Categorical(idx)
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+        # list values
+        cat = Categorical(list(idx))
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+        # list values, categories
+        cat = Categorical(list(idx), categories=list(idx))
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+        # shuffled
+        values = idx.take([1, 2, 0])
+        cat = Categorical(values, categories=idx)
+        tm.assert_numpy_array_equal(cat.codes, np.array([1, 2, 0], dtype="int8"))
+        tm.assert_index_equal(cat.categories, idx)
+
+        # extra
+        values = pd.interval_range(8, 11, periods=3)
+        cat = Categorical(values, categories=idx)
+        expected_codes = np.array([8, 9, -1], dtype="int8")
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+        # overlapping
+        idx = IntervalIndex([Interval(0, 2), Interval(0, 1)])
+        cat = Categorical(idx, categories=idx)
+        expected_codes = np.array([0, 1], dtype="int8")
+        tm.assert_numpy_array_equal(cat.codes, expected_codes)
+        tm.assert_index_equal(cat.categories, idx)
+
+    def test_categorical_extension_array_nullable(self, nulls_fixture):
+        # GH:
+        arr = pd.arrays.StringArray._from_sequence(
+            [nulls_fixture] * 2, dtype=pd.StringDtype()
+        )
+        result = Categorical(arr)
+        assert arr.dtype == result.categories.dtype
+        expected = Categorical(Series([pd.NA, pd.NA], dtype=arr.dtype))
+        tm.assert_categorical_equal(result, expected)
+
+    def test_from_sequence_copy(self):
+        cat = Categorical(np.arange(5).repeat(2))
+        result = Categorical._from_sequence(cat, dtype=cat.dtype, copy=False)
+
+        # more generally, we'd be OK with a view
+        assert result._codes is cat._codes
+
+        result = Categorical._from_sequence(cat, dtype=cat.dtype, copy=True)
+
+        assert not tm.shares_memory(result, cat)
+
+    def test_constructor_datetime64_non_nano(self):
+        categories = np.arange(10).view("M8[D]")
+        values = categories[::2].copy()
+
+        cat = Categorical(values, categories=categories)
+        assert (cat == values).all()
+
+    def test_constructor_preserves_freq(self):
+        # GH33830 freq retention in categorical
+        dti = date_range("2016-01-01", periods=5)
+
+        expected = dti.freq
+
+        cat = Categorical(dti)
+        result = cat.categories.freq
+
+        assert expected == result
+
+    @pytest.mark.parametrize(
+        "values, categories",
+        [
+            [range(5), None],
+            [range(4), range(5)],
+            [[0, 1, 2, 3], range(5)],
+            [[], range(5)],
+        ],
+    )
+    def test_range_values_preserves_rangeindex_categories(self, values, categories):
+        result = Categorical(values=values, categories=categories).categories
+        expected = RangeIndex(range(5))
+        tm.assert_index_equal(result, expected, exact=True)
diff --git a/pandas/pandas/tests/arrays/categorical/test_dtypes.py b/pandas/pandas/tests/arrays/categorical/test_dtypes.py
new file mode 100644
index 0000000000000..ec1d501ddba16
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_dtypes.py
@@ -0,0 +1,138 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    Index,
+    IntervalIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalDtypes:
+    def test_categories_match_up_to_permutation(self):
+        # test dtype comparisons between cats
+
+        c1 = Categorical(list("aabca"), categories=list("abc"), ordered=False)
+        c2 = Categorical(list("aabca"), categories=list("cab"), ordered=False)
+        c3 = Categorical(list("aabca"), categories=list("cab"), ordered=True)
+        assert c1._categories_match_up_to_permutation(c1)
+        assert c2._categories_match_up_to_permutation(c2)
+        assert c3._categories_match_up_to_permutation(c3)
+        assert c1._categories_match_up_to_permutation(c2)
+        assert not c1._categories_match_up_to_permutation(c3)
+        assert not c1._categories_match_up_to_permutation(Index(list("aabca")))
+        assert not c1._categories_match_up_to_permutation(c1.astype(object))
+        assert c1._categories_match_up_to_permutation(CategoricalIndex(c1))
+        assert c1._categories_match_up_to_permutation(
+            CategoricalIndex(c1, categories=list("cab"))
+        )
+        assert not c1._categories_match_up_to_permutation(
+            CategoricalIndex(c1, ordered=True)
+        )
+
+        # GH 16659
+        s1 = Series(c1)
+        s2 = Series(c2)
+        s3 = Series(c3)
+        assert c1._categories_match_up_to_permutation(s1)
+        assert c2._categories_match_up_to_permutation(s2)
+        assert c3._categories_match_up_to_permutation(s3)
+        assert c1._categories_match_up_to_permutation(s2)
+        assert not c1._categories_match_up_to_permutation(s3)
+        assert not c1._categories_match_up_to_permutation(s1.astype(object))
+
+    def test_set_dtype_same(self):
+        c = Categorical(["a", "b", "c"])
+        result = c._set_dtype(CategoricalDtype(["a", "b", "c"]))
+        tm.assert_categorical_equal(result, c)
+
+    def test_set_dtype_new_categories(self):
+        c = Categorical(["a", "b", "c"])
+        result = c._set_dtype(CategoricalDtype(list("abcd")))
+        tm.assert_numpy_array_equal(result.codes, c.codes)
+        tm.assert_index_equal(result.dtype.categories, Index(list("abcd")))
+
+    @pytest.mark.parametrize(
+        "values, categories, new_categories",
+        [
+            # No NaNs, same cats, same order
+            (["a", "b", "a"], ["a", "b"], ["a", "b"]),
+            # No NaNs, same cats, different order
+            (["a", "b", "a"], ["a", "b"], ["b", "a"]),
+            # Same, unsorted
+            (["b", "a", "a"], ["a", "b"], ["a", "b"]),
+            # No NaNs, same cats, different order
+            (["b", "a", "a"], ["a", "b"], ["b", "a"]),
+            # NaNs
+            (["a", "b", "c"], ["a", "b"], ["a", "b"]),
+            (["a", "b", "c"], ["a", "b"], ["b", "a"]),
+            (["b", "a", "c"], ["a", "b"], ["a", "b"]),
+            (["b", "a", "c"], ["a", "b"], ["b", "a"]),
+            # Introduce NaNs
+            (["a", "b", "c"], ["a", "b"], ["a"]),
+            (["a", "b", "c"], ["a", "b"], ["b"]),
+            (["b", "a", "c"], ["a", "b"], ["a"]),
+            (["b", "a", "c"], ["a", "b"], ["b"]),
+            # No overlap
+            (["a", "b", "c"], ["a", "b"], ["d", "e"]),
+        ],
+    )
+    def test_set_dtype_many(self, values, categories, new_categories, ordered):
+        c = Categorical(values, categories)
+        expected = Categorical(values, new_categories, ordered)
+        result = c._set_dtype(expected.dtype)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_set_dtype_no_overlap(self):
+        c = Categorical(["a", "b", "c"], ["d", "e"])
+        result = c._set_dtype(CategoricalDtype(["a", "b"]))
+        expected = Categorical([None, None, None], categories=["a", "b"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_codes_dtypes(self):
+        # GH 8453
+        result = Categorical(["foo", "bar", "baz"])
+        assert result.codes.dtype == "int8"
+
+        result = Categorical([f"foo{i:05d}" for i in range(400)])
+        assert result.codes.dtype == "int16"
+
+        result = Categorical([f"foo{i:05d}" for i in range(40000)])
+        assert result.codes.dtype == "int32"
+
+        # adding cats
+        result = Categorical(["foo", "bar", "baz"])
+        assert result.codes.dtype == "int8"
+        result = result.add_categories([f"foo{i:05d}" for i in range(400)])
+        assert result.codes.dtype == "int16"
+
+        # removing cats
+        result = result.remove_categories([f"foo{i:05d}" for i in range(300)])
+        assert result.codes.dtype == "int8"
+
+    def test_iter_python_types(self):
+        # GH-19909
+        cat = Categorical([1, 2])
+        assert isinstance(next(iter(cat)), int)
+        assert isinstance(cat.tolist()[0], int)
+
+    def test_iter_python_types_datetime(self):
+        cat = Categorical([Timestamp("2017-01-01"), Timestamp("2017-01-02")])
+        assert isinstance(next(iter(cat)), Timestamp)
+        assert isinstance(cat.tolist()[0], Timestamp)
+
+    def test_interval_index_category(self):
+        # GH 38316
+        index = IntervalIndex.from_breaks(np.arange(3, dtype="uint64"))
+
+        result = CategoricalIndex(index).dtype.categories
+        expected = IntervalIndex.from_arrays(
+            [0, 1], [1, 2], dtype="interval[uint64, right]"
+        )
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/categorical/test_indexing.py b/pandas/pandas/tests/arrays/categorical/test_indexing.py
new file mode 100644
index 0000000000000..33c55b2090bd6
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_indexing.py
@@ -0,0 +1,386 @@
+import math
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    Categorical,
+    CategoricalIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+import pandas.core.common as com
+
+
+class TestCategoricalIndexingWithFactor:
+    def test_getitem(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        assert factor[0] == "a"
+        assert factor[-1] == "c"
+
+        subf = factor[[0, 1, 2]]
+        tm.assert_numpy_array_equal(subf._codes, np.array([0, 1, 1], dtype=np.int8))
+
+        subf = factor[np.asarray(factor) == "c"]
+        tm.assert_numpy_array_equal(subf._codes, np.array([2, 2, 2], dtype=np.int8))
+
+    def test_setitem(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        # int/positional
+        c = factor.copy()
+        c[0] = "b"
+        assert c[0] == "b"
+        c[-1] = "a"
+        assert c[-1] == "a"
+
+        # boolean
+        c = factor.copy()
+        indexer = np.zeros(len(c), dtype="bool")
+        indexer[0] = True
+        indexer[-1] = True
+        c[indexer] = "c"
+        expected = Categorical(["c", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+
+        tm.assert_categorical_equal(c, expected)
+
+    @pytest.mark.parametrize("categories", [None, ["b", "a"]])
+    def test_setitem_same_but_unordered(self, categories):
+        # GH-24142
+        other = Categorical(["b", "a"], categories=categories)
+        target = Categorical(["a", "b"], categories=["a", "b"])
+        mask = np.array([True, False])
+        target[mask] = other[mask]
+        expected = Categorical(["b", "b"], categories=["a", "b"])
+        tm.assert_categorical_equal(target, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            Categorical(["b", "a"], categories=["b", "a", "c"]),
+            Categorical(["b", "a"], categories=["a", "b", "c"]),
+            Categorical(["a", "a"], categories=["a"]),
+            Categorical(["b", "b"], categories=["b"]),
+        ],
+    )
+    def test_setitem_different_unordered_raises(self, other):
+        # GH-24142
+        target = Categorical(["a", "b"], categories=["a", "b"])
+        mask = np.array([True, False])
+        msg = "Cannot set a Categorical with another, without identical categories"
+        with pytest.raises(TypeError, match=msg):
+            target[mask] = other[mask]
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            Categorical(["b", "a"]),
+            Categorical(["b", "a"], categories=["b", "a"], ordered=True),
+            Categorical(["b", "a"], categories=["a", "b", "c"], ordered=True),
+        ],
+    )
+    def test_setitem_same_ordered_raises(self, other):
+        # Gh-24142
+        target = Categorical(["a", "b"], categories=["a", "b"], ordered=True)
+        mask = np.array([True, False])
+        msg = "Cannot set a Categorical with another, without identical categories"
+        with pytest.raises(TypeError, match=msg):
+            target[mask] = other[mask]
+
+    def test_setitem_tuple(self):
+        # GH#20439
+        cat = Categorical([(0, 1), (0, 2), (0, 1)])
+
+        # This should not raise
+        cat[1] = cat[0]
+        assert cat[1] == (0, 1)
+
+    def test_setitem_listlike(self):
+        # GH#9469
+        # properly coerce the input indexers
+
+        cat = Categorical(
+            np.random.default_rng(2).integers(0, 5, size=150000).astype(np.int8)
+        ).add_categories([-1000])
+        indexer = np.array([100000]).astype(np.int64)
+        cat[indexer] = -1000
+
+        # we are asserting the code result here
+        # which maps to the -1000 category
+        result = cat.codes[np.array([100000]).astype(np.int64)]
+        tm.assert_numpy_array_equal(result, np.array([5], dtype="int8"))
+
+
+class TestCategoricalIndexing:
+    def test_getitem_slice(self):
+        cat = Categorical(["a", "b", "c", "d", "a", "b", "c"])
+        sliced = cat[3]
+        assert sliced == "d"
+
+        sliced = cat[3:5]
+        expected = Categorical(["d", "a"], categories=["a", "b", "c", "d"])
+        tm.assert_categorical_equal(sliced, expected)
+
+    def test_getitem_listlike(self):
+        # GH 9469
+        # properly coerce the input indexers
+
+        c = Categorical(
+            np.random.default_rng(2).integers(0, 5, size=150000).astype(np.int8)
+        )
+        result = c.codes[np.array([100000]).astype(np.int64)]
+        expected = c[np.array([100000]).astype(np.int64)].codes
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_periodindex(self):
+        idx1 = PeriodIndex(
+            ["2014-01", "2014-01", "2014-02", "2014-02", "2014-03", "2014-03"],
+            freq="M",
+        )
+
+        cat1 = Categorical(idx1)
+        str(cat1)
+        exp_arr = np.array([0, 0, 1, 1, 2, 2], dtype=np.int8)
+        exp_idx = PeriodIndex(["2014-01", "2014-02", "2014-03"], freq="M")
+        tm.assert_numpy_array_equal(cat1._codes, exp_arr)
+        tm.assert_index_equal(cat1.categories, exp_idx)
+
+        idx2 = PeriodIndex(
+            ["2014-03", "2014-03", "2014-02", "2014-01", "2014-03", "2014-01"],
+            freq="M",
+        )
+        cat2 = Categorical(idx2, ordered=True)
+        str(cat2)
+        exp_arr = np.array([2, 2, 1, 0, 2, 0], dtype=np.int8)
+        exp_idx2 = PeriodIndex(["2014-01", "2014-02", "2014-03"], freq="M")
+        tm.assert_numpy_array_equal(cat2._codes, exp_arr)
+        tm.assert_index_equal(cat2.categories, exp_idx2)
+
+        idx3 = PeriodIndex(
+            [
+                "2013-12",
+                "2013-11",
+                "2013-10",
+                "2013-09",
+                "2013-08",
+                "2013-07",
+                "2013-05",
+            ],
+            freq="M",
+        )
+        cat3 = Categorical(idx3, ordered=True)
+        exp_arr = np.array([6, 5, 4, 3, 2, 1, 0], dtype=np.int8)
+        exp_idx = PeriodIndex(
+            [
+                "2013-05",
+                "2013-07",
+                "2013-08",
+                "2013-09",
+                "2013-10",
+                "2013-11",
+                "2013-12",
+            ],
+            freq="M",
+        )
+        tm.assert_numpy_array_equal(cat3._codes, exp_arr)
+        tm.assert_index_equal(cat3.categories, exp_idx)
+
+    @pytest.mark.parametrize(
+        "null_val",
+        [None, np.nan, NaT, NA, math.nan, "NaT", "nat", "NAT", "nan", "NaN", "NAN"],
+    )
+    def test_periodindex_on_null_types(self, null_val):
+        # GH 46673
+        result = PeriodIndex(["2022-04-06", "2022-04-07", null_val], freq="D")
+        expected = PeriodIndex(["2022-04-06", "2022-04-07", "NaT"], dtype="period[D]")
+        assert result[2] is NaT
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("new_categories", [[1, 2, 3, 4], [1, 2]])
+    def test_categories_assignments_wrong_length_raises(self, new_categories):
+        cat = Categorical(["a", "b", "c", "a"])
+        msg = (
+            "new categories need to have the same number of items "
+            "as the old categories!"
+        )
+        with pytest.raises(ValueError, match=msg):
+            cat.rename_categories(new_categories)
+
+    # Combinations of sorted/unique:
+    @pytest.mark.parametrize(
+        "idx_values", [[1, 2, 3, 4], [1, 3, 2, 4], [1, 3, 3, 4], [1, 2, 2, 4]]
+    )
+    # Combinations of missing/unique
+    @pytest.mark.parametrize("key_values", [[1, 2], [1, 5], [1, 1], [5, 5]])
+    @pytest.mark.parametrize("key_class", [Categorical, CategoricalIndex])
+    @pytest.mark.parametrize("dtype", [None, "category", "key"])
+    def test_get_indexer_non_unique(self, idx_values, key_values, key_class, dtype):
+        # GH 21448
+        key = key_class(key_values, categories=range(1, 5))
+
+        if dtype == "key":
+            dtype = key.dtype
+
+        # Test for flat index and CategoricalIndex with same/different cats:
+        idx = Index(idx_values, dtype=dtype)
+        expected, exp_miss = idx.get_indexer_non_unique(key_values)
+        result, res_miss = idx.get_indexer_non_unique(key)
+
+        tm.assert_numpy_array_equal(expected, result)
+        tm.assert_numpy_array_equal(exp_miss, res_miss)
+
+        exp_unique = idx.unique().get_indexer(key_values)
+        res_unique = idx.unique().get_indexer(key)
+        tm.assert_numpy_array_equal(res_unique, exp_unique)
+
+    def test_where_unobserved_nan(self):
+        ser = Series(Categorical(["a", "b"]))
+        result = ser.where([True, False])
+        expected = Series(Categorical(["a", None], categories=["a", "b"]))
+        tm.assert_series_equal(result, expected)
+
+        # all NA
+        ser = Series(Categorical(["a", "b"]))
+        result = ser.where([False, False])
+        expected = Series(Categorical([None, None], categories=["a", "b"]))
+        tm.assert_series_equal(result, expected)
+
+    def test_where_unobserved_categories(self):
+        ser = Series(Categorical(["a", "b", "c"], categories=["d", "c", "b", "a"]))
+        result = ser.where([True, True, False], other="b")
+        expected = Series(Categorical(["a", "b", "b"], categories=ser.cat.categories))
+        tm.assert_series_equal(result, expected)
+
+    def test_where_other_categorical(self):
+        ser = Series(Categorical(["a", "b", "c"], categories=["d", "c", "b", "a"]))
+        other = Categorical(["b", "c", "a"], categories=["a", "c", "b", "d"])
+        result = ser.where([True, False, True], other)
+        expected = Series(Categorical(["a", "c", "c"], dtype=ser.dtype))
+        tm.assert_series_equal(result, expected)
+
+    def test_where_new_category_raises(self):
+        ser = Series(Categorical(["a", "b", "c"]))
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            ser.where([True, False, True], "d")
+
+    def test_where_ordered_differs_rasies(self):
+        ser = Series(
+            Categorical(["a", "b", "c"], categories=["d", "c", "b", "a"], ordered=True)
+        )
+        other = Categorical(
+            ["b", "c", "a"], categories=["a", "c", "b", "d"], ordered=True
+        )
+        with pytest.raises(TypeError, match="without identical categories"):
+            ser.where([True, False, True], other)
+
+
+class TestContains:
+    def test_contains(self):
+        # GH#21508
+        cat = Categorical(list("aabbca"), categories=list("cab"))
+
+        assert "b" in cat
+        assert "z" not in cat
+        assert np.nan not in cat
+        with pytest.raises(TypeError, match="unhashable type: 'list'"):
+            assert [1] in cat
+
+        # assert codes NOT in index
+        assert 0 not in cat
+        assert 1 not in cat
+
+        cat = Categorical(list("aabbca") + [np.nan], categories=list("cab"))
+        assert np.nan in cat
+
+    @pytest.mark.parametrize(
+        "item, expected",
+        [
+            (Interval(0, 1), True),
+            (1.5, True),
+            (Interval(0.5, 1.5), False),
+            ("a", False),
+            (Timestamp(1), False),
+            (Timedelta(1), False),
+        ],
+        ids=str,
+    )
+    def test_contains_interval(self, item, expected):
+        # GH#23705
+        cat = Categorical(IntervalIndex.from_breaks(range(3)))
+        result = item in cat
+        assert result is expected
+
+    def test_contains_list(self):
+        # GH#21729
+        cat = Categorical([1, 2, 3])
+
+        assert "a" not in cat
+
+        with pytest.raises(TypeError, match="unhashable type"):
+            ["a"] in cat
+
+        with pytest.raises(TypeError, match="unhashable type"):
+            ["a", "b"] in cat
+
+
+@pytest.mark.parametrize("index", [True, False])
+def test_mask_with_boolean(index):
+    ser = Series(range(3))
+    idx = Categorical([True, False, True])
+    if index:
+        idx = CategoricalIndex(idx)
+
+    assert com.is_bool_indexer(idx)
+    result = ser[idx]
+    expected = ser[idx.astype("object")]
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("index", [True, False])
+def test_mask_with_boolean_na_treated_as_false(index):
+    # https://github.com/pandas-dev/pandas/issues/31503
+    ser = Series(range(3))
+    idx = Categorical([True, False, None])
+    if index:
+        idx = CategoricalIndex(idx)
+
+    result = ser[idx]
+    expected = ser[idx.fillna(False)]
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.fixture
+def non_coercible_categorical(monkeypatch):
+    """
+    Monkeypatch Categorical.__array__ to ensure no implicit conversion.
+
+    Raises
+    ------
+    ValueError
+        When Categorical.__array__ is called.
+    """
+
+    # TODO(Categorical): identify other places where this may be
+    # useful and move to a conftest.py
+    def array(self, dtype=None):
+        raise ValueError("I cannot be converted.")
+
+    with monkeypatch.context() as m:
+        m.setattr(Categorical, "__array__", array)
+        yield
+
+
+def test_series_at():
+    arr = Categorical(["a", "b", "c"])
+    ser = Series(arr)
+    result = ser.at[0]
+    assert result == "a"
diff --git a/pandas/pandas/tests/arrays/categorical/test_map.py b/pandas/pandas/tests/arrays/categorical/test_map.py
new file mode 100644
index 0000000000000..585b207c9b241
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_map.py
@@ -0,0 +1,136 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "data, categories",
+    [
+        (list("abcbca"), list("cab")),
+        (pd.interval_range(0, 3).repeat(3), pd.interval_range(0, 3)),
+    ],
+    ids=["string", "interval"],
+)
+def test_map_str(data, categories, ordered, na_action):
+    # GH 31202 - override base class since we want to maintain categorical/ordered
+    cat = Categorical(data, categories=categories, ordered=ordered)
+    result = cat.map(str, na_action=na_action)
+    expected = Categorical(
+        map(str, data), categories=map(str, categories), ordered=ordered
+    )
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_map(na_action):
+    cat = Categorical(list("ABABC"), categories=list("CBA"), ordered=True)
+    result = cat.map(lambda x: x.lower(), na_action=na_action)
+    exp = Categorical(list("ababc"), categories=list("cba"), ordered=True)
+    tm.assert_categorical_equal(result, exp)
+
+    cat = Categorical(list("ABABC"), categories=list("BAC"), ordered=False)
+    result = cat.map(lambda x: x.lower(), na_action=na_action)
+    exp = Categorical(list("ababc"), categories=list("bac"), ordered=False)
+    tm.assert_categorical_equal(result, exp)
+
+    # GH 12766: Return an index not an array
+    result = cat.map(lambda x: 1, na_action=na_action)
+    exp = Index(np.array([1] * 5, dtype=np.int64))
+    tm.assert_index_equal(result, exp)
+
+    # change categories dtype
+    cat = Categorical(list("ABABC"), categories=list("BAC"), ordered=False)
+
+    def f(x):
+        return {"A": 10, "B": 20, "C": 30}.get(x)
+
+    result = cat.map(f, na_action=na_action)
+    exp = Categorical([10, 20, 10, 20, 30], categories=[20, 10, 30], ordered=False)
+    tm.assert_categorical_equal(result, exp)
+
+    mapper = Series([10, 20, 30], index=["A", "B", "C"])
+    result = cat.map(mapper, na_action=na_action)
+    tm.assert_categorical_equal(result, exp)
+
+    result = cat.map({"A": 10, "B": 20, "C": 30}, na_action=na_action)
+    tm.assert_categorical_equal(result, exp)
+
+
+@pytest.mark.parametrize(
+    ("data", "f", "expected"),
+    (
+        ([1, 1, np.nan], pd.isna, Index([False, False, True])),
+        ([1, 2, np.nan], pd.isna, Index([False, False, True])),
+        ([1, 1, np.nan], {1: False}, Categorical([False, False, np.nan])),
+        ([1, 2, np.nan], {1: False, 2: False}, Index([False, False, np.nan])),
+        (
+            [1, 1, np.nan],
+            Series([False, False]),
+            Categorical([False, False, np.nan]),
+        ),
+        (
+            [1, 2, np.nan],
+            Series([False] * 3),
+            Index([False, False, np.nan]),
+        ),
+    ),
+)
+def test_map_with_nan_none(data, f, expected):  # GH 24241
+    values = Categorical(data)
+    result = values.map(f, na_action=None)
+    if isinstance(expected, Categorical):
+        tm.assert_categorical_equal(result, expected)
+    else:
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("data", "f", "expected"),
+    (
+        ([1, 1, np.nan], pd.isna, Categorical([False, False, np.nan])),
+        ([1, 2, np.nan], pd.isna, Index([False, False, np.nan])),
+        ([1, 1, np.nan], {1: False}, Categorical([False, False, np.nan])),
+        ([1, 2, np.nan], {1: False, 2: False}, Index([False, False, np.nan])),
+        (
+            [1, 1, np.nan],
+            Series([False, False]),
+            Categorical([False, False, np.nan]),
+        ),
+        (
+            [1, 2, np.nan],
+            Series([False, False, False]),
+            Index([False, False, np.nan]),
+        ),
+    ),
+)
+def test_map_with_nan_ignore(data, f, expected):  # GH 24241
+    values = Categorical(data)
+    result = values.map(f, na_action="ignore")
+    if data[1] == 1:
+        tm.assert_categorical_equal(result, expected)
+    else:
+        tm.assert_index_equal(result, expected)
+
+
+def test_map_with_dict_or_series(na_action):
+    orig_values = ["a", "B", 1, "a"]
+    new_values = ["one", 2, 3.0, "one"]
+    cat = Categorical(orig_values)
+
+    mapper = Series(new_values[:-1], index=orig_values[:-1])
+    result = cat.map(mapper, na_action=na_action)
+
+    # Order of categories in result can be different
+    expected = Categorical(new_values, categories=[3.0, 2, "one"])
+    tm.assert_categorical_equal(result, expected)
+
+    mapper = dict(zip(orig_values[:-1], new_values[:-1]))
+    result = cat.map(mapper, na_action=na_action)
+    # Order of categories in result can be different
+    tm.assert_categorical_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/categorical/test_missing.py b/pandas/pandas/tests/arrays/categorical/test_missing.py
new file mode 100644
index 0000000000000..e3cb9664e19f2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_missing.py
@@ -0,0 +1,133 @@
+import collections
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    Index,
+    Series,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalMissing:
+    def test_isna(self):
+        exp = np.array([False, False, True])
+        cat = Categorical(["a", "b", np.nan])
+        res = cat.isna()
+
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_na_flags_int_categories(self):
+        # #1457
+
+        categories = list(range(10))
+        labels = np.random.default_rng(2).integers(0, 10, 20)
+        labels[::5] = -1
+
+        cat = Categorical(labels, categories)
+        repr(cat)
+
+        tm.assert_numpy_array_equal(isna(cat), labels == -1)
+
+    def test_nan_handling(self):
+        # Nans are represented as -1 in codes
+        c = Categorical(["a", "b", np.nan, "a"])
+        tm.assert_index_equal(c.categories, Index(["a", "b"]))
+        tm.assert_numpy_array_equal(c._codes, np.array([0, 1, -1, 0], dtype=np.int8))
+        c[1] = np.nan
+        tm.assert_index_equal(c.categories, Index(["a", "b"]))
+        tm.assert_numpy_array_equal(c._codes, np.array([0, -1, -1, 0], dtype=np.int8))
+
+        # Adding nan to categories should make assigned nan point to the
+        # category!
+        c = Categorical(["a", "b", np.nan, "a"])
+        tm.assert_index_equal(c.categories, Index(["a", "b"]))
+        tm.assert_numpy_array_equal(c._codes, np.array([0, 1, -1, 0], dtype=np.int8))
+
+    def test_set_dtype_nans(self):
+        c = Categorical(["a", "b", np.nan])
+        result = c._set_dtype(CategoricalDtype(["a", "c"]))
+        tm.assert_numpy_array_equal(result.codes, np.array([0, -1, -1], dtype="int8"))
+
+    def test_set_item_nan(self):
+        cat = Categorical([1, 2, 3])
+        cat[1] = np.nan
+
+        exp = Categorical([1, np.nan, 3], categories=[1, 2, 3])
+        tm.assert_categorical_equal(cat, exp)
+
+    @pytest.mark.parametrize("named", [True, False])
+    def test_fillna_iterable_category(self, named):
+        # https://github.com/pandas-dev/pandas/issues/21097
+        if named:
+            Point = collections.namedtuple("Point", "x y")
+        else:
+            Point = lambda *args: args  # tuple
+        cat = Categorical(np.array([Point(0, 0), Point(0, 1), None], dtype=object))
+        result = cat.fillna(Point(0, 0))
+        expected = Categorical([Point(0, 0), Point(0, 1), Point(0, 0)])
+
+        tm.assert_categorical_equal(result, expected)
+
+        # Case where the Point is not among our categories; we want ValueError,
+        #  not NotImplementedError GH#41914
+        cat = Categorical(np.array([Point(1, 0), Point(0, 1), None], dtype=object))
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            cat.fillna(Point(0, 0))
+
+    def test_fillna_array(self):
+        # accept Categorical or ndarray value if it holds appropriate values
+        cat = Categorical(["A", "B", "C", None, None])
+
+        other = cat.fillna("C")
+        result = cat.fillna(other)
+        tm.assert_categorical_equal(result, other)
+        assert isna(cat[-1])  # didn't modify original inplace
+
+        other = np.array(["A", "B", "C", "B", "A"])
+        result = cat.fillna(other)
+        expected = Categorical(["A", "B", "C", "B", "A"], dtype=cat.dtype)
+        tm.assert_categorical_equal(result, expected)
+        assert isna(cat[-1])  # didn't modify original inplace
+
+    @pytest.mark.parametrize(
+        "a1, a2, categories",
+        [
+            (["a", "b", "c"], [np.nan, "a", "b"], ["a", "b", "c"]),
+            ([1, 2, 3], [np.nan, 1, 2], [1, 2, 3]),
+        ],
+    )
+    def test_compare_categorical_with_missing(self, a1, a2, categories):
+        # GH 28384
+        cat_type = CategoricalDtype(categories)
+
+        # !=
+        result = Series(a1, dtype=cat_type) != Series(a2, dtype=cat_type)
+        expected = Series(a1) != Series(a2)
+        tm.assert_series_equal(result, expected)
+
+        # ==
+        result = Series(a1, dtype=cat_type) == Series(a2, dtype=cat_type)
+        expected = Series(a1) == Series(a2)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "na_value, dtype",
+        [
+            (pd.NaT, "datetime64[s]"),
+            (None, "float64"),
+            (np.nan, "float64"),
+            (pd.NA, "float64"),
+        ],
+    )
+    def test_categorical_only_missing_values_no_cast(self, na_value, dtype):
+        # GH#44900
+        result = Categorical([na_value, na_value])
+        tm.assert_index_equal(result.categories, Index([], dtype=dtype))
diff --git a/pandas/pandas/tests/arrays/categorical/test_operators.py b/pandas/pandas/tests/arrays/categorical/test_operators.py
new file mode 100644
index 0000000000000..dbc6cc7715744
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_operators.py
@@ -0,0 +1,408 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalOpsWithFactor:
+    def test_categories_none_comparisons(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        tm.assert_categorical_equal(factor, factor)
+
+    def test_comparisons(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        result = factor[factor == "a"]
+        expected = factor[np.asarray(factor) == "a"]
+        tm.assert_categorical_equal(result, expected)
+
+        result = factor[factor != "a"]
+        expected = factor[np.asarray(factor) != "a"]
+        tm.assert_categorical_equal(result, expected)
+
+        result = factor[factor < "c"]
+        expected = factor[np.asarray(factor) < "c"]
+        tm.assert_categorical_equal(result, expected)
+
+        result = factor[factor > "a"]
+        expected = factor[np.asarray(factor) > "a"]
+        tm.assert_categorical_equal(result, expected)
+
+        result = factor[factor >= "b"]
+        expected = factor[np.asarray(factor) >= "b"]
+        tm.assert_categorical_equal(result, expected)
+
+        result = factor[factor <= "b"]
+        expected = factor[np.asarray(factor) <= "b"]
+        tm.assert_categorical_equal(result, expected)
+
+        n = len(factor)
+
+        other = factor[np.random.default_rng(2).permutation(n)]
+        result = factor == other
+        expected = np.asarray(factor) == np.asarray(other)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = factor == "d"
+        expected = np.zeros(len(factor), dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # comparisons with categoricals
+        cat_rev = Categorical(["a", "b", "c"], categories=["c", "b", "a"], ordered=True)
+        cat_rev_base = Categorical(
+            ["b", "b", "b"], categories=["c", "b", "a"], ordered=True
+        )
+        cat = Categorical(["a", "b", "c"], ordered=True)
+        cat_base = Categorical(["b", "b", "b"], categories=cat.categories, ordered=True)
+
+        # comparisons need to take categories ordering into account
+        res_rev = cat_rev > cat_rev_base
+        exp_rev = np.array([True, False, False])
+        tm.assert_numpy_array_equal(res_rev, exp_rev)
+
+        res_rev = cat_rev < cat_rev_base
+        exp_rev = np.array([False, False, True])
+        tm.assert_numpy_array_equal(res_rev, exp_rev)
+
+        res = cat > cat_base
+        exp = np.array([False, False, True])
+        tm.assert_numpy_array_equal(res, exp)
+
+        # Only categories with same categories can be compared
+        msg = "Categoricals can only be compared if 'categories' are the same"
+        with pytest.raises(TypeError, match=msg):
+            cat > cat_rev
+
+        cat_rev_base2 = Categorical(["b", "b", "b"], categories=["c", "b", "a", "d"])
+
+        with pytest.raises(TypeError, match=msg):
+            cat_rev > cat_rev_base2
+
+        # Only categories with same ordering information can be compared
+        cat_unordered = cat.set_ordered(False)
+        assert not (cat > cat).any()
+
+        with pytest.raises(TypeError, match=msg):
+            cat > cat_unordered
+
+        # comparison (in both directions) with Series will raise
+        s = Series(["b", "b", "b"], dtype=object)
+        msg = (
+            "Cannot compare a Categorical for op __gt__ with type "
+            r"<class 'numpy\.ndarray'>"
+        )
+        with pytest.raises(TypeError, match=msg):
+            cat > s
+        with pytest.raises(TypeError, match=msg):
+            cat_rev > s
+        with pytest.raises(TypeError, match=msg):
+            s < cat
+        with pytest.raises(TypeError, match=msg):
+            s < cat_rev
+
+        # comparison with numpy.array will raise in both direction, but only on
+        # newer numpy versions
+        a = np.array(["b", "b", "b"], dtype=object)
+        with pytest.raises(TypeError, match=msg):
+            cat > a
+        with pytest.raises(TypeError, match=msg):
+            cat_rev > a
+
+        # Make sure that unequal comparison take the categories order in
+        # account
+        cat_rev = Categorical(list("abc"), categories=list("cba"), ordered=True)
+        exp = np.array([True, False, False])
+        res = cat_rev > "b"
+        tm.assert_numpy_array_equal(res, exp)
+
+        # check that zero-dim array gets unboxed
+        res = cat_rev > np.array("b")
+        tm.assert_numpy_array_equal(res, exp)
+
+
+class TestCategoricalOps:
+    @pytest.mark.parametrize(
+        "categories",
+        [["a", "b"], [0, 1], [Timestamp("2019"), Timestamp("2020")]],
+    )
+    def test_not_equal_with_na(self, categories):
+        # https://github.com/pandas-dev/pandas/issues/32276
+        c1 = Categorical.from_codes([-1, 0], categories=categories)
+        c2 = Categorical.from_codes([0, 1], categories=categories)
+
+        result = c1 != c2
+
+        assert result.all()
+
+    def test_compare_frame(self):
+        # GH#24282 check that Categorical.__cmp__(DataFrame) defers to frame
+        data = ["a", "b", 2, "a"]
+        cat = Categorical(data)
+
+        df = DataFrame(cat)
+
+        result = cat == df.T
+        expected = DataFrame([[True, True, True, True]])
+        tm.assert_frame_equal(result, expected)
+
+        result = cat[::-1] != df.T
+        expected = DataFrame([[False, True, True, False]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_compare_frame_raises(self, comparison_op):
+        # alignment raises unless we transpose
+        op = comparison_op
+        cat = Categorical(["a", "b", 2, "a"])
+        df = DataFrame(cat)
+        msg = "Unable to coerce to Series, length must be 1: given 4"
+        with pytest.raises(ValueError, match=msg):
+            op(cat, df)
+
+    def test_datetime_categorical_comparison(self):
+        dt_cat = Categorical(date_range("2014-01-01", periods=3), ordered=True)
+        tm.assert_numpy_array_equal(dt_cat > dt_cat[0], np.array([False, True, True]))
+        tm.assert_numpy_array_equal(dt_cat[0] < dt_cat, np.array([False, True, True]))
+
+    def test_reflected_comparison_with_scalars(self):
+        # GH8658
+        cat = Categorical([1, 2, 3], ordered=True)
+        tm.assert_numpy_array_equal(cat > cat[0], np.array([False, True, True]))
+        tm.assert_numpy_array_equal(cat[0] < cat, np.array([False, True, True]))
+
+    def test_comparison_with_unknown_scalars(self):
+        # https://github.com/pandas-dev/pandas/issues/9836#issuecomment-92123057
+        # and following comparisons with scalars not in categories should raise
+        # for unequal comps, but not for equal/not equal
+        cat = Categorical([1, 2, 3], ordered=True)
+
+        msg = "Invalid comparison between dtype=category and int"
+        with pytest.raises(TypeError, match=msg):
+            cat < 4
+        with pytest.raises(TypeError, match=msg):
+            cat > 4
+        with pytest.raises(TypeError, match=msg):
+            4 < cat
+        with pytest.raises(TypeError, match=msg):
+            4 > cat
+
+        tm.assert_numpy_array_equal(cat == 4, np.array([False, False, False]))
+        tm.assert_numpy_array_equal(cat != 4, np.array([True, True, True]))
+
+    def test_comparison_with_tuple(self):
+        cat = Categorical(np.array(["foo", (0, 1), 3, (0, 1)], dtype=object))
+
+        result = cat == "foo"
+        expected = np.array([True, False, False, False], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = cat == (0, 1)
+        expected = np.array([False, True, False, True], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = cat != (0, 1)
+        tm.assert_numpy_array_equal(result, ~expected)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_comparison_of_ordered_categorical_with_nan_to_scalar(
+        self, compare_operators_no_eq_ne
+    ):
+        # https://github.com/pandas-dev/pandas/issues/26504
+        # BUG: fix ordered categorical comparison with missing values (#26504 )
+        # and following comparisons with scalars in categories with missing
+        # values should be evaluated as False
+
+        cat = Categorical([1, 2, 3, None], categories=[1, 2, 3], ordered=True)
+        scalar = 2
+        expected = getattr(np.array(cat), compare_operators_no_eq_ne)(scalar)
+        actual = getattr(cat, compare_operators_no_eq_ne)(scalar)
+        tm.assert_numpy_array_equal(actual, expected)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_comparison_of_ordered_categorical_with_nan_to_listlike(
+        self, compare_operators_no_eq_ne
+    ):
+        # https://github.com/pandas-dev/pandas/issues/26504
+        # and following comparisons of missing values in ordered Categorical
+        # with listlike should be evaluated as False
+
+        cat = Categorical([1, 2, 3, None], categories=[1, 2, 3], ordered=True)
+        other = Categorical([2, 2, 2, 2], categories=[1, 2, 3], ordered=True)
+        expected = getattr(np.array(cat), compare_operators_no_eq_ne)(2)
+        actual = getattr(cat, compare_operators_no_eq_ne)(other)
+        tm.assert_numpy_array_equal(actual, expected)
+
+    @pytest.mark.parametrize(
+        "data,reverse,base",
+        [(list("abc"), list("cba"), list("bbb")), ([1, 2, 3], [3, 2, 1], [2, 2, 2])],
+    )
+    def test_comparisons(self, data, reverse, base):
+        cat_rev = Series(Categorical(data, categories=reverse, ordered=True))
+        cat_rev_base = Series(Categorical(base, categories=reverse, ordered=True))
+        cat = Series(Categorical(data, ordered=True))
+        cat_base = Series(
+            Categorical(base, categories=cat.cat.categories, ordered=True)
+        )
+        s = Series(base, dtype=object if base == list("bbb") else None)
+        a = np.array(base)
+
+        # comparisons need to take categories ordering into account
+        res_rev = cat_rev > cat_rev_base
+        exp_rev = Series([True, False, False])
+        tm.assert_series_equal(res_rev, exp_rev)
+
+        res_rev = cat_rev < cat_rev_base
+        exp_rev = Series([False, False, True])
+        tm.assert_series_equal(res_rev, exp_rev)
+
+        res = cat > cat_base
+        exp = Series([False, False, True])
+        tm.assert_series_equal(res, exp)
+
+        scalar = base[1]
+        res = cat > scalar
+        exp = Series([False, False, True])
+        exp2 = cat.values > scalar
+        tm.assert_series_equal(res, exp)
+        tm.assert_numpy_array_equal(res.values, exp2)
+        res_rev = cat_rev > scalar
+        exp_rev = Series([True, False, False])
+        exp_rev2 = cat_rev.values > scalar
+        tm.assert_series_equal(res_rev, exp_rev)
+        tm.assert_numpy_array_equal(res_rev.values, exp_rev2)
+
+        # Only categories with same categories can be compared
+        msg = "Categoricals can only be compared if 'categories' are the same"
+        with pytest.raises(TypeError, match=msg):
+            cat > cat_rev
+
+        # categorical cannot be compared to Series or numpy array, and also
+        # not the other way around
+        msg = (
+            "Cannot compare a Categorical for op __gt__ with type "
+            r"<class 'numpy\.ndarray'>"
+        )
+        with pytest.raises(TypeError, match=msg):
+            cat > s
+        with pytest.raises(TypeError, match=msg):
+            cat_rev > s
+        with pytest.raises(TypeError, match=msg):
+            cat > a
+        with pytest.raises(TypeError, match=msg):
+            cat_rev > a
+
+        with pytest.raises(TypeError, match=msg):
+            s < cat
+        with pytest.raises(TypeError, match=msg):
+            s < cat_rev
+
+        with pytest.raises(TypeError, match=msg):
+            a < cat
+        with pytest.raises(TypeError, match=msg):
+            a < cat_rev
+
+    @pytest.mark.parametrize("box", [lambda x: x, Series])
+    def test_unordered_different_order_equal(self, box):
+        # https://github.com/pandas-dev/pandas/issues/16014
+        c1 = box(Categorical(["a", "b"], categories=["a", "b"], ordered=False))
+        c2 = box(Categorical(["a", "b"], categories=["b", "a"], ordered=False))
+        assert (c1 == c2).all()
+
+        c1 = box(Categorical(["a", "b"], categories=["a", "b"], ordered=False))
+        c2 = box(Categorical(["b", "a"], categories=["b", "a"], ordered=False))
+        assert (c1 != c2).all()
+
+        c1 = box(Categorical(["a", "a"], categories=["a", "b"], ordered=False))
+        c2 = box(Categorical(["b", "b"], categories=["b", "a"], ordered=False))
+        assert (c1 != c2).all()
+
+        c1 = box(Categorical(["a", "a"], categories=["a", "b"], ordered=False))
+        c2 = box(Categorical(["a", "b"], categories=["b", "a"], ordered=False))
+        result = c1 == c2
+        tm.assert_numpy_array_equal(np.array(result), np.array([True, False]))
+
+    def test_unordered_different_categories_raises(self):
+        c1 = Categorical(["a", "b"], categories=["a", "b"], ordered=False)
+        c2 = Categorical(["a", "c"], categories=["c", "a"], ordered=False)
+
+        with pytest.raises(TypeError, match=("Categoricals can only be compared")):
+            c1 == c2
+
+    def test_compare_different_lengths(self):
+        c1 = Categorical([], categories=["a", "b"])
+        c2 = Categorical([], categories=["a"])
+
+        msg = "Categoricals can only be compared if 'categories' are the same."
+        with pytest.raises(TypeError, match=msg):
+            c1 == c2
+
+    def test_compare_unordered_different_order(self):
+        # https://github.com/pandas-dev/pandas/issues/16603#issuecomment-
+        # 349290078
+        a = Categorical(["a"], categories=["a", "b"])
+        b = Categorical(["b"], categories=["b", "a"])
+        assert not a.equals(b)
+
+    def test_numeric_like_ops(self):
+        df = DataFrame({"value": np.random.default_rng(2).integers(0, 10000, 100)})
+        labels = [f"{i} - {i + 499}" for i in range(0, 10000, 500)]
+        cat_labels = Categorical(labels, labels)
+
+        df = df.sort_values(by=["value"], ascending=True)
+        df["value_group"] = pd.cut(
+            df.value, range(0, 10500, 500), right=False, labels=cat_labels
+        )
+
+        # numeric ops should not succeed
+        for op, str_rep in [
+            ("__add__", r"\+"),
+            ("__sub__", "-"),
+            ("__mul__", r"\*"),
+            ("__truediv__", "/"),
+        ]:
+            msg = f"Series cannot perform the operation {str_rep}|unsupported operand"
+            with pytest.raises(TypeError, match=msg):
+                getattr(df, op)(df)
+
+        # reduction ops should not succeed (unless specifically defined, e.g.
+        # min/max)
+        s = df["value_group"]
+        for op in ["kurt", "skew", "var", "std", "mean", "sum", "median"]:
+            msg = f"does not support operation '{op}'"
+            with pytest.raises(TypeError, match=msg):
+                getattr(s, op)(numeric_only=False)
+
+    def test_numeric_like_ops_series(self):
+        # numpy ops
+        s = Series(Categorical([1, 2, 3, 4]))
+        with pytest.raises(TypeError, match="does not support operation 'sum'"):
+            np.sum(s)
+
+    @pytest.mark.parametrize(
+        "op, str_rep",
+        [
+            ("__add__", r"\+"),
+            ("__sub__", "-"),
+            ("__mul__", r"\*"),
+            ("__truediv__", "/"),
+        ],
+    )
+    def test_numeric_like_ops_series_arith(self, op, str_rep):
+        # numeric ops on a Series
+        s = Series(Categorical([1, 2, 3, 4]))
+        msg = f"Series cannot perform the operation {str_rep}|unsupported operand"
+        with pytest.raises(TypeError, match=msg):
+            getattr(s, op)(2)
+
+    def test_numeric_like_ops_series_invalid(self):
+        # invalid ufunc
+        s = Series(Categorical([1, 2, 3, 4]))
+        msg = "Object with dtype category cannot perform the numpy op log"
+        with pytest.raises(TypeError, match=msg):
+            np.log(s)
diff --git a/pandas/pandas/tests/arrays/categorical/test_replace.py b/pandas/pandas/tests/arrays/categorical/test_replace.py
new file mode 100644
index 0000000000000..7f3e8d3ed6e6e
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_replace.py
@@ -0,0 +1,71 @@
+import pytest
+
+import pandas as pd
+from pandas import Categorical
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "to_replace,value,expected",
+    [
+        # one-to-one
+        (4, 1, [1, 2, 3]),
+        (3, 1, [1, 2, 1]),
+        # many-to-one
+        ((5, 6), 2, [1, 2, 3]),
+        ((3, 2), 1, [1, 1, 1]),
+    ],
+)
+def test_replace_categorical_series(to_replace, value, expected):
+    # GH 31720
+    ser = pd.Series([1, 2, 3], dtype="category")
+    result = ser.replace(to_replace, value)
+    expected = pd.Series(Categorical(expected, categories=[1, 2, 3]))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "to_replace,value",
+    [
+        # one-to-one
+        (3, 5),
+        # many-to-one
+        ((3, 2), 5),
+    ],
+)
+def test_replace_categorical_series_new_category_raises(to_replace, value):
+    # GH 31720
+    ser = pd.Series([1, 2, 3], dtype="category")
+    with pytest.raises(
+        TypeError, match="Cannot setitem on a Categorical with a new category"
+    ):
+        ser.replace(to_replace, value)
+
+
+def test_replace_maintain_ordering():
+    # GH51016
+    dtype = pd.CategoricalDtype([0, 1, 2], ordered=True)
+    ser = pd.Series([0, 1, 2], dtype=dtype)
+    result = ser.replace(0, 2)
+    expected = pd.Series([2, 1, 2], dtype=dtype)
+    tm.assert_series_equal(expected, result, check_category_order=True)
+
+
+def test_replace_categorical_ea_dtype():
+    # GH49404
+    cat = Categorical(pd.array(["a", "b", "c"], dtype="string"))
+    result = pd.Series(cat).replace(["a", "b"], ["c", "c"])._values
+    expected = Categorical(
+        pd.array(["c"] * 3, dtype="string"),
+        categories=pd.array(["a", "b", "c"], dtype="string"),
+    )
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_replace_categorical_ea_dtype_different_cats_raises():
+    # GH49404
+    cat = Categorical(pd.array(["a", "b"], dtype="string"))
+    with pytest.raises(
+        TypeError, match="Cannot setitem on a Categorical with a new category"
+    ):
+        pd.Series(cat).replace(["a", "b"], ["c", pd.NA])
diff --git a/pandas/pandas/tests/arrays/categorical/test_repr.py b/pandas/pandas/tests/arrays/categorical/test_repr.py
new file mode 100644
index 0000000000000..3a2c489920eb0
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_repr.py
@@ -0,0 +1,550 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    Index,
+    Series,
+    date_range,
+    option_context,
+    period_range,
+    timedelta_range,
+)
+
+
+class TestCategoricalReprWithFactor:
+    def test_print(self, using_infer_string):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"], ordered=True)
+        if using_infer_string:
+            expected = [
+                "['a', 'b', 'b', 'a', 'a', 'c', 'c', 'c']",
+                "Categories (3, str): [a < b < c]",
+            ]
+        else:
+            expected = [
+                "['a', 'b', 'b', 'a', 'a', 'c', 'c', 'c']",
+                "Categories (3, object): ['a' < 'b' < 'c']",
+            ]
+        expected = "\n".join(expected)
+        actual = repr(factor)
+        assert actual == expected
+
+
+class TestCategoricalRepr:
+    def test_big_print(self):
+        codes = np.array([0, 1, 2, 0, 1, 2] * 100)
+        dtype = CategoricalDtype(categories=Index(["a", "b", "c"], dtype=object))
+        factor = Categorical.from_codes(codes, dtype=dtype)
+        expected = [
+            "['a', 'b', 'c', 'a', 'b', ..., 'b', 'c', 'a', 'b', 'c']",
+            "Length: 600",
+            "Categories (3, object): ['a', 'b', 'c']",
+        ]
+        expected = "\n".join(expected)
+
+        actual = repr(factor)
+
+        assert actual == expected
+
+    def test_empty_print(self):
+        factor = Categorical([], Index(["a", "b", "c"], dtype=object))
+        expected = "[], Categories (3, object): ['a', 'b', 'c']"
+        actual = repr(factor)
+        assert actual == expected
+
+        assert expected == actual
+        factor = Categorical([], Index(["a", "b", "c"], dtype=object), ordered=True)
+        expected = "[], Categories (3, object): ['a' < 'b' < 'c']"
+        actual = repr(factor)
+        assert expected == actual
+
+        factor = Categorical([], [])
+        expected = "[], Categories (0, object): []"
+        assert expected == repr(factor)
+
+    def test_print_none_width(self):
+        # GH10087
+        a = Series(Categorical([1, 2, 3, 4]))
+        exp = (
+            "0    1\n1    2\n2    3\n3    4\n"
+            "dtype: category\nCategories (4, int64): [1, 2, 3, 4]"
+        )
+
+        with option_context("display.width", None):
+            assert exp == repr(a)
+
+    @pytest.mark.skipif(
+        using_string_dtype(),
+        reason="Change once infer_string is set to True by default",
+    )
+    def test_unicode_print(self):
+        c = Categorical(["aaaaa", "bb", "cccc"] * 20)
+        expected = """\
+['aaaaa', 'bb', 'cccc', 'aaaaa', 'bb', ..., 'bb', 'cccc', 'aaaaa', 'bb', 'cccc']
+Length: 60
+Categories (3, object): ['aaaaa', 'bb', 'cccc']"""
+
+        assert repr(c) == expected
+
+        c = Categorical(["ああああ", "いいいいい", "ううううううう"] * 20)
+        expected = """\
+['ああああ', 'いいいいい', 'ううううううう', 'ああああ', 'いいいいい', ..., 'いいいいい', 'ううううううう', 'ああああ', 'いいいいい', 'ううううううう']
+Length: 60
+Categories (3, object): ['ああああ', 'いいいいい', 'ううううううう']"""  # noqa: E501
+
+        assert repr(c) == expected
+
+        # unicode option should not affect to Categorical, as it doesn't care
+        # the repr width
+        with option_context("display.unicode.east_asian_width", True):
+            c = Categorical(["ああああ", "いいいいい", "ううううううう"] * 20)
+            expected = """['ああああ', 'いいいいい', 'ううううううう', 'ああああ', 'いいいいい', ..., 'いいいいい', 'ううううううう', 'ああああ', 'いいいいい', 'ううううううう']
+Length: 60
+Categories (3, object): ['ああああ', 'いいいいい', 'ううううううう']"""  # noqa: E501
+
+            assert repr(c) == expected
+
+    def test_categorical_repr(self):
+        c = Categorical([1, 2, 3])
+        exp = """[1, 2, 3]
+Categories (3, int64): [1, 2, 3]"""
+
+        assert repr(c) == exp
+
+        c = Categorical([1, 2, 3, 1, 2, 3], categories=[1, 2, 3])
+        exp = """[1, 2, 3, 1, 2, 3]
+Categories (3, int64): [1, 2, 3]"""
+
+        assert repr(c) == exp
+
+        c = Categorical([1, 2, 3, 4, 5] * 10)
+        exp = """[1, 2, 3, 4, 5, ..., 1, 2, 3, 4, 5]
+Length: 50
+Categories (5, int64): [1, 2, 3, 4, 5]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(np.arange(20, dtype=np.int64))
+        exp = """[0, 1, 2, 3, 4, ..., 15, 16, 17, 18, 19]
+Length: 20
+Categories (20, int64): [0, 1, 2, 3, ..., 16, 17, 18, 19]"""
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_ordered(self):
+        c = Categorical([1, 2, 3], ordered=True)
+        exp = """[1, 2, 3]
+Categories (3, int64): [1 < 2 < 3]"""
+
+        assert repr(c) == exp
+
+        c = Categorical([1, 2, 3, 1, 2, 3], categories=[1, 2, 3], ordered=True)
+        exp = """[1, 2, 3, 1, 2, 3]
+Categories (3, int64): [1 < 2 < 3]"""
+
+        assert repr(c) == exp
+
+        c = Categorical([1, 2, 3, 4, 5] * 10, ordered=True)
+        exp = """[1, 2, 3, 4, 5, ..., 1, 2, 3, 4, 5]
+Length: 50
+Categories (5, int64): [1 < 2 < 3 < 4 < 5]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(np.arange(20, dtype=np.int64), ordered=True)
+        exp = """[0, 1, 2, 3, 4, ..., 15, 16, 17, 18, 19]
+Length: 20
+Categories (20, int64): [0 < 1 < 2 < 3 ... 16 < 17 < 18 < 19]"""
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_datetime(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        c = Categorical(idx)
+
+        exp = (
+            "[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, "
+            "2011-01-01 12:00:00, 2011-01-01 13:00:00]\n"
+            "Categories (5, datetime64[ns]): [2011-01-01 09:00:00, "
+            "2011-01-01 10:00:00, 2011-01-01 11:00:00,\n"
+            "                                 2011-01-01 12:00:00, "
+            "2011-01-01 13:00:00]"
+            ""
+        )
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = (
+            "[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, "
+            "2011-01-01 12:00:00, 2011-01-01 13:00:00, 2011-01-01 09:00:00, "
+            "2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, "
+            "2011-01-01 13:00:00]\n"
+            "Categories (5, datetime64[ns]): [2011-01-01 09:00:00, "
+            "2011-01-01 10:00:00, 2011-01-01 11:00:00,\n"
+            "                                 2011-01-01 12:00:00, "
+            "2011-01-01 13:00:00]"
+        )
+
+        assert repr(c) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        c = Categorical(idx)
+        exp = (
+            "[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, "
+            "2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, "
+            "2011-01-01 13:00:00-05:00]\n"
+            "Categories (5, datetime64[ns, US/Eastern]): "
+            "[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00,\n"
+            "                                             "
+            "2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00,\n"
+            "                                             "
+            "2011-01-01 13:00:00-05:00]"
+        )
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = (
+            "[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, "
+            "2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, "
+            "2011-01-01 13:00:00-05:00, 2011-01-01 09:00:00-05:00, "
+            "2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, "
+            "2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00]\n"
+            "Categories (5, datetime64[ns, US/Eastern]): "
+            "[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00,\n"
+            "                                             "
+            "2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00,\n"
+            "                                             "
+            "2011-01-01 13:00:00-05:00]"
+        )
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_datetime_ordered(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        c = Categorical(idx, ordered=True)
+        exp = """[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, 2011-01-01 13:00:00]
+Categories (5, datetime64[ns]): [2011-01-01 09:00:00 < 2011-01-01 10:00:00 < 2011-01-01 11:00:00 <
+                                 2011-01-01 12:00:00 < 2011-01-01 13:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, 2011-01-01 13:00:00, 2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, 2011-01-01 13:00:00]
+Categories (5, datetime64[ns]): [2011-01-01 09:00:00 < 2011-01-01 10:00:00 < 2011-01-01 11:00:00 <
+                                 2011-01-01 12:00:00 < 2011-01-01 13:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        c = Categorical(idx, ordered=True)
+        exp = """[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00]
+Categories (5, datetime64[ns, US/Eastern]): [2011-01-01 09:00:00-05:00 < 2011-01-01 10:00:00-05:00 <
+                                             2011-01-01 11:00:00-05:00 < 2011-01-01 12:00:00-05:00 <
+                                             2011-01-01 13:00:00-05:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00, 2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00]
+Categories (5, datetime64[ns, US/Eastern]): [2011-01-01 09:00:00-05:00 < 2011-01-01 10:00:00-05:00 <
+                                             2011-01-01 11:00:00-05:00 < 2011-01-01 12:00:00-05:00 <
+                                             2011-01-01 13:00:00-05:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_int_with_nan(self):
+        c = Categorical([1, 2, np.nan])
+        c_exp = """[1, 2, NaN]\nCategories (2, int64): [1, 2]"""
+        assert repr(c) == c_exp
+
+        s = Series([1, 2, np.nan], dtype="object").astype("category")
+        s_exp = """0      1\n1      2\n2    NaN
+dtype: category
+Categories (2, int64): [1, 2]"""
+        assert repr(s) == s_exp
+
+    def test_categorical_repr_period(self):
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        c = Categorical(idx)
+        exp = """[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00]
+Categories (5, period[h]): [2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00,
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = """[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00, 2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00]
+Categories (5, period[h]): [2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00,
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        c = Categorical(idx)
+        exp = """[2011-01, 2011-02, 2011-03, 2011-04, 2011-05]
+Categories (5, period[M]): [2011-01, 2011-02, 2011-03, 2011-04, 2011-05]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = """[2011-01, 2011-02, 2011-03, 2011-04, 2011-05, 2011-01, 2011-02, 2011-03, 2011-04, 2011-05]
+Categories (5, period[M]): [2011-01, 2011-02, 2011-03, 2011-04, 2011-05]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_period_ordered(self):
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        c = Categorical(idx, ordered=True)
+        exp = """[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00]
+Categories (5, period[h]): [2011-01-01 09:00 < 2011-01-01 10:00 < 2011-01-01 11:00 < 2011-01-01 12:00 <
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00, 2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00]
+Categories (5, period[h]): [2011-01-01 09:00 < 2011-01-01 10:00 < 2011-01-01 11:00 < 2011-01-01 12:00 <
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        c = Categorical(idx, ordered=True)
+        exp = """[2011-01, 2011-02, 2011-03, 2011-04, 2011-05]
+Categories (5, period[M]): [2011-01 < 2011-02 < 2011-03 < 2011-04 < 2011-05]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[2011-01, 2011-02, 2011-03, 2011-04, 2011-05, 2011-01, 2011-02, 2011-03, 2011-04, 2011-05]
+Categories (5, period[M]): [2011-01 < 2011-02 < 2011-03 < 2011-04 < 2011-05]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_timedelta(self):
+        idx = timedelta_range("1 days", periods=5)
+        c = Categorical(idx)
+        exp = """[1 days, 2 days, 3 days, 4 days, 5 days]
+Categories (5, timedelta64[ns]): [1 days, 2 days, 3 days, 4 days, 5 days]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = """[1 days, 2 days, 3 days, 4 days, 5 days, 1 days, 2 days, 3 days, 4 days, 5 days]
+Categories (5, timedelta64[ns]): [1 days, 2 days, 3 days, 4 days, 5 days]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        idx = timedelta_range("1 hours", periods=20)
+        c = Categorical(idx)
+        exp = """[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, 4 days 01:00:00, ..., 15 days 01:00:00, 16 days 01:00:00, 17 days 01:00:00, 18 days 01:00:00, 19 days 01:00:00]
+Length: 20
+Categories (20, timedelta64[ns]): [0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00,
+                                   3 days 01:00:00, ..., 16 days 01:00:00, 17 days 01:00:00,
+                                   18 days 01:00:00, 19 days 01:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx)
+        exp = """[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, 4 days 01:00:00, ..., 15 days 01:00:00, 16 days 01:00:00, 17 days 01:00:00, 18 days 01:00:00, 19 days 01:00:00]
+Length: 40
+Categories (20, timedelta64[ns]): [0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00,
+                                   3 days 01:00:00, ..., 16 days 01:00:00, 17 days 01:00:00,
+                                   18 days 01:00:00, 19 days 01:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+    def test_categorical_repr_timedelta_ordered(self):
+        idx = timedelta_range("1 days", periods=5)
+        c = Categorical(idx, ordered=True)
+        exp = """[1 days, 2 days, 3 days, 4 days, 5 days]
+Categories (5, timedelta64[ns]): [1 days < 2 days < 3 days < 4 days < 5 days]"""
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[1 days, 2 days, 3 days, 4 days, 5 days, 1 days, 2 days, 3 days, 4 days, 5 days]
+Categories (5, timedelta64[ns]): [1 days < 2 days < 3 days < 4 days < 5 days]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        idx = timedelta_range("1 hours", periods=20)
+        c = Categorical(idx, ordered=True)
+        exp = """[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, 4 days 01:00:00, ..., 15 days 01:00:00, 16 days 01:00:00, 17 days 01:00:00, 18 days 01:00:00, 19 days 01:00:00]
+Length: 20
+Categories (20, timedelta64[ns]): [0 days 01:00:00 < 1 days 01:00:00 < 2 days 01:00:00 <
+                                   3 days 01:00:00 ... 16 days 01:00:00 < 17 days 01:00:00 <
+                                   18 days 01:00:00 < 19 days 01:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+        c = Categorical(idx.append(idx), categories=idx, ordered=True)
+        exp = """[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, 4 days 01:00:00, ..., 15 days 01:00:00, 16 days 01:00:00, 17 days 01:00:00, 18 days 01:00:00, 19 days 01:00:00]
+Length: 40
+Categories (20, timedelta64[ns]): [0 days 01:00:00 < 1 days 01:00:00 < 2 days 01:00:00 <
+                                   3 days 01:00:00 ... 16 days 01:00:00 < 17 days 01:00:00 <
+                                   18 days 01:00:00 < 19 days 01:00:00]"""  # noqa: E501
+
+        assert repr(c) == exp
+
+    def test_categorical_index_repr(self):
+        idx = CategoricalIndex(Categorical([1, 2, 3]))
+        exp = """CategoricalIndex([1, 2, 3], categories=[1, 2, 3], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(idx) == exp
+
+        i = CategoricalIndex(Categorical(np.arange(10, dtype=np.int64)))
+        exp = """CategoricalIndex([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], categories=[0, 1, 2, 3, ..., 6, 7, 8, 9], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_ordered(self):
+        i = CategoricalIndex(Categorical([1, 2, 3], ordered=True))
+        exp = """CategoricalIndex([1, 2, 3], categories=[1, 2, 3], ordered=True, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        i = CategoricalIndex(Categorical(np.arange(10, dtype=np.int64), ordered=True))
+        exp = """CategoricalIndex([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], categories=[0, 1, 2, 3, ..., 6, 7, 8, 9], ordered=True, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_datetime(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00:00', '2011-01-01 10:00:00',
+                  '2011-01-01 11:00:00', '2011-01-01 12:00:00',
+                  '2011-01-01 13:00:00'],
+                 categories=[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, 2011-01-01 13:00:00], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00:00-05:00', '2011-01-01 10:00:00-05:00',
+                  '2011-01-01 11:00:00-05:00', '2011-01-01 12:00:00-05:00',
+                  '2011-01-01 13:00:00-05:00'],
+                 categories=[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_datetime_ordered(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['2011-01-01 09:00:00', '2011-01-01 10:00:00',
+                  '2011-01-01 11:00:00', '2011-01-01 12:00:00',
+                  '2011-01-01 13:00:00'],
+                 categories=[2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00, 2011-01-01 12:00:00, 2011-01-01 13:00:00], ordered=True, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['2011-01-01 09:00:00-05:00', '2011-01-01 10:00:00-05:00',
+                  '2011-01-01 11:00:00-05:00', '2011-01-01 12:00:00-05:00',
+                  '2011-01-01 13:00:00-05:00'],
+                 categories=[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00], ordered=True, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        i = CategoricalIndex(Categorical(idx.append(idx), ordered=True))
+        exp = """CategoricalIndex(['2011-01-01 09:00:00-05:00', '2011-01-01 10:00:00-05:00',
+                  '2011-01-01 11:00:00-05:00', '2011-01-01 12:00:00-05:00',
+                  '2011-01-01 13:00:00-05:00', '2011-01-01 09:00:00-05:00',
+                  '2011-01-01 10:00:00-05:00', '2011-01-01 11:00:00-05:00',
+                  '2011-01-01 12:00:00-05:00', '2011-01-01 13:00:00-05:00'],
+                 categories=[2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00, 2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00, 2011-01-01 13:00:00-05:00], ordered=True, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_period(self):
+        # test all length
+        idx = period_range("2011-01-01 09:00", freq="h", periods=1)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00'], categories=[2011-01-01 09:00], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        idx = period_range("2011-01-01 09:00", freq="h", periods=2)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00', '2011-01-01 10:00'], categories=[2011-01-01 09:00, 2011-01-01 10:00], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        idx = period_range("2011-01-01 09:00", freq="h", periods=3)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00', '2011-01-01 10:00', '2011-01-01 11:00'], categories=[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01-01 09:00', '2011-01-01 10:00', '2011-01-01 11:00',
+                  '2011-01-01 12:00', '2011-01-01 13:00'],
+                 categories=[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        i = CategoricalIndex(Categorical(idx.append(idx)))
+        exp = """CategoricalIndex(['2011-01-01 09:00', '2011-01-01 10:00', '2011-01-01 11:00',
+                  '2011-01-01 12:00', '2011-01-01 13:00', '2011-01-01 09:00',
+                  '2011-01-01 10:00', '2011-01-01 11:00', '2011-01-01 12:00',
+                  '2011-01-01 13:00'],
+                 categories=[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['2011-01', '2011-02', '2011-03', '2011-04', '2011-05'], categories=[2011-01, 2011-02, 2011-03, 2011-04, 2011-05], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_period_ordered(self):
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['2011-01-01 09:00', '2011-01-01 10:00', '2011-01-01 11:00',
+                  '2011-01-01 12:00', '2011-01-01 13:00'],
+                 categories=[2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00, 2011-01-01 13:00], ordered=True, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['2011-01', '2011-02', '2011-03', '2011-04', '2011-05'], categories=[2011-01, 2011-02, 2011-03, 2011-04, 2011-05], ordered=True, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_timedelta(self):
+        idx = timedelta_range("1 days", periods=5)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['1 days', '2 days', '3 days', '4 days', '5 days'], categories=[1 days, 2 days, 3 days, 4 days, 5 days], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        idx = timedelta_range("1 hours", periods=10)
+        i = CategoricalIndex(Categorical(idx))
+        exp = """CategoricalIndex(['0 days 01:00:00', '1 days 01:00:00', '2 days 01:00:00',
+                  '3 days 01:00:00', '4 days 01:00:00', '5 days 01:00:00',
+                  '6 days 01:00:00', '7 days 01:00:00', '8 days 01:00:00',
+                  '9 days 01:00:00'],
+                 categories=[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, ..., 6 days 01:00:00, 7 days 01:00:00, 8 days 01:00:00, 9 days 01:00:00], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+    def test_categorical_index_repr_timedelta_ordered(self):
+        idx = timedelta_range("1 days", periods=5)
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['1 days', '2 days', '3 days', '4 days', '5 days'], categories=[1 days, 2 days, 3 days, 4 days, 5 days], ordered=True, dtype='category')"""  # noqa: E501
+        assert repr(i) == exp
+
+        idx = timedelta_range("1 hours", periods=10)
+        i = CategoricalIndex(Categorical(idx, ordered=True))
+        exp = """CategoricalIndex(['0 days 01:00:00', '1 days 01:00:00', '2 days 01:00:00',
+                  '3 days 01:00:00', '4 days 01:00:00', '5 days 01:00:00',
+                  '6 days 01:00:00', '7 days 01:00:00', '8 days 01:00:00',
+                  '9 days 01:00:00'],
+                 categories=[0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00, 3 days 01:00:00, ..., 6 days 01:00:00, 7 days 01:00:00, 8 days 01:00:00, 9 days 01:00:00], ordered=True, dtype='category')"""  # noqa: E501
+
+        assert repr(i) == exp
+
+    def test_categorical_str_repr(self):
+        # GH 33676
+        result = repr(Categorical([1, "2", 3, 4]))
+        expected = "[1, '2', 3, 4]\nCategories (4, object): [1, 3, 4, '2']"
+        assert result == expected
diff --git a/pandas/pandas/tests/arrays/categorical/test_sorting.py b/pandas/pandas/tests/arrays/categorical/test_sorting.py
new file mode 100644
index 0000000000000..ae527065b3fb9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_sorting.py
@@ -0,0 +1,128 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    Index,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalSort:
+    def test_argsort(self):
+        c = Categorical([5, 3, 1, 4, 2], ordered=True)
+
+        expected = np.array([2, 4, 1, 3, 0])
+        tm.assert_numpy_array_equal(
+            c.argsort(ascending=True), expected, check_dtype=False
+        )
+
+        expected = expected[::-1]
+        tm.assert_numpy_array_equal(
+            c.argsort(ascending=False), expected, check_dtype=False
+        )
+
+    def test_numpy_argsort(self):
+        c = Categorical([5, 3, 1, 4, 2], ordered=True)
+
+        expected = np.array([2, 4, 1, 3, 0])
+        tm.assert_numpy_array_equal(np.argsort(c), expected, check_dtype=False)
+
+        tm.assert_numpy_array_equal(
+            np.argsort(c, kind="mergesort"), expected, check_dtype=False
+        )
+
+        msg = "the 'axis' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argsort(c, axis=0)
+
+        msg = "the 'order' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argsort(c, order="C")
+
+    def test_sort_values(self):
+        # unordered cats are sortable
+        cat = Categorical(["a", "b", "b", "a"], ordered=False)
+        cat.sort_values()
+
+        cat = Categorical(["a", "c", "b", "d"], ordered=True)
+
+        # sort_values
+        res = cat.sort_values()
+        exp = np.array(["a", "b", "c", "d"], dtype=object)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, cat.categories)
+
+        cat = Categorical(
+            ["a", "c", "b", "d"], categories=["a", "b", "c", "d"], ordered=True
+        )
+        res = cat.sort_values()
+        exp = np.array(["a", "b", "c", "d"], dtype=object)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, cat.categories)
+
+        res = cat.sort_values(ascending=False)
+        exp = np.array(["d", "c", "b", "a"], dtype=object)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, cat.categories)
+
+        # sort (inplace order)
+        cat1 = cat.copy()
+        orig_codes = cat1._codes
+        cat1.sort_values(inplace=True)
+        assert cat1._codes is orig_codes
+        exp = np.array(["a", "b", "c", "d"], dtype=object)
+        tm.assert_numpy_array_equal(cat1.__array__(), exp)
+        tm.assert_index_equal(res.categories, cat.categories)
+
+        # reverse
+        cat = Categorical(["a", "c", "c", "b", "d"], ordered=True)
+        res = cat.sort_values(ascending=False)
+        exp_val = np.array(["d", "c", "c", "b", "a"], dtype=object)
+        exp_categories = Index(["a", "b", "c", "d"])
+        tm.assert_numpy_array_equal(res.__array__(), exp_val)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+    def test_sort_values_na_position(self):
+        # see gh-12882
+        cat = Categorical([5, 2, np.nan, 2, np.nan], ordered=True)
+        exp_categories = Index([2, 5])
+
+        exp = np.array([2.0, 2.0, 5.0, np.nan, np.nan])
+        res = cat.sort_values()  # default arguments
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        exp = np.array([np.nan, np.nan, 2.0, 2.0, 5.0])
+        res = cat.sort_values(ascending=True, na_position="first")
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        exp = np.array([np.nan, np.nan, 5.0, 2.0, 2.0])
+        res = cat.sort_values(ascending=False, na_position="first")
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        exp = np.array([2.0, 2.0, 5.0, np.nan, np.nan])
+        res = cat.sort_values(ascending=True, na_position="last")
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        exp = np.array([5.0, 2.0, 2.0, np.nan, np.nan])
+        res = cat.sort_values(ascending=False, na_position="last")
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        cat = Categorical(["a", "c", "b", "d", np.nan], ordered=True)
+        res = cat.sort_values(ascending=False, na_position="last")
+        exp_val = np.array(["d", "c", "b", "a", np.nan], dtype=object)
+        exp_categories = Index(["a", "b", "c", "d"])
+        tm.assert_numpy_array_equal(res.__array__(), exp_val)
+        tm.assert_index_equal(res.categories, exp_categories)
+
+        cat = Categorical(["a", "c", "b", "d", np.nan], ordered=True)
+        res = cat.sort_values(ascending=False, na_position="first")
+        exp_val = np.array([np.nan, "d", "c", "b", "a"], dtype=object)
+        exp_categories = Index(["a", "b", "c", "d"])
+        tm.assert_numpy_array_equal(res.__array__(), exp_val)
+        tm.assert_index_equal(res.categories, exp_categories)
diff --git a/pandas/pandas/tests/arrays/categorical/test_subclass.py b/pandas/pandas/tests/arrays/categorical/test_subclass.py
new file mode 100644
index 0000000000000..5b0c0a44e655d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_subclass.py
@@ -0,0 +1,26 @@
+from pandas import Categorical
+import pandas._testing as tm
+
+
+class SubclassedCategorical(Categorical):
+    pass
+
+
+class TestCategoricalSubclassing:
+    def test_constructor(self):
+        sc = SubclassedCategorical(["a", "b", "c"])
+        assert isinstance(sc, SubclassedCategorical)
+        tm.assert_categorical_equal(sc, Categorical(["a", "b", "c"]))
+
+    def test_from_codes(self):
+        sc = SubclassedCategorical.from_codes([1, 0, 2], ["a", "b", "c"])
+        assert isinstance(sc, SubclassedCategorical)
+        exp = Categorical.from_codes([1, 0, 2], ["a", "b", "c"])
+        tm.assert_categorical_equal(sc, exp)
+
+    def test_map(self):
+        sc = SubclassedCategorical(["a", "b", "c"])
+        res = sc.map(lambda x: x.upper(), na_action=None)
+        assert isinstance(res, SubclassedCategorical)
+        exp = Categorical(["A", "B", "C"])
+        tm.assert_categorical_equal(res, exp)
diff --git a/pandas/pandas/tests/arrays/categorical/test_take.py b/pandas/pandas/tests/arrays/categorical/test_take.py
new file mode 100644
index 0000000000000..373f1b30a13c2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_take.py
@@ -0,0 +1,89 @@
+import numpy as np
+import pytest
+
+from pandas import Categorical
+import pandas._testing as tm
+
+
+@pytest.fixture(params=[True, False])
+def allow_fill(request):
+    """Boolean 'allow_fill' parameter for Categorical.take"""
+    return request.param
+
+
+class TestTake:
+    # https://github.com/pandas-dev/pandas/issues/20664
+
+    def test_take_default_allow_fill(self):
+        cat = Categorical(["a", "b"])
+        with tm.assert_produces_warning(None):
+            result = cat.take([0, -1])
+
+        assert result.equals(cat)
+
+    def test_take_positive_no_warning(self):
+        cat = Categorical(["a", "b"])
+        with tm.assert_produces_warning(None):
+            cat.take([0, 0])
+
+    def test_take_bounds(self, allow_fill):
+        # https://github.com/pandas-dev/pandas/issues/20664
+        cat = Categorical(["a", "b", "a"])
+        if allow_fill:
+            msg = "indices are out-of-bounds"
+        else:
+            msg = "index 4 is out of bounds for( axis 0 with)? size 3"
+        with pytest.raises(IndexError, match=msg):
+            cat.take([4, 5], allow_fill=allow_fill)
+
+    def test_take_empty(self, allow_fill):
+        # https://github.com/pandas-dev/pandas/issues/20664
+        cat = Categorical([], categories=["a", "b"])
+        if allow_fill:
+            msg = "indices are out-of-bounds"
+        else:
+            msg = "cannot do a non-empty take from an empty axes"
+        with pytest.raises(IndexError, match=msg):
+            cat.take([0], allow_fill=allow_fill)
+
+    def test_positional_take(self, ordered):
+        cat = Categorical(["a", "a", "b", "b"], categories=["b", "a"], ordered=ordered)
+        result = cat.take([0, 1, 2], allow_fill=False)
+        expected = Categorical(
+            ["a", "a", "b"], categories=cat.categories, ordered=ordered
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_positional_take_unobserved(self, ordered):
+        cat = Categorical(["a", "b"], categories=["a", "b", "c"], ordered=ordered)
+        result = cat.take([1, 0], allow_fill=False)
+        expected = Categorical(["b", "a"], categories=cat.categories, ordered=ordered)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_take_allow_fill(self):
+        # https://github.com/pandas-dev/pandas/issues/23296
+        cat = Categorical(["a", "a", "b"])
+        result = cat.take([0, -1, -1], allow_fill=True)
+        expected = Categorical(["a", np.nan, np.nan], categories=["a", "b"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_take_fill_with_negative_one(self):
+        # -1 was a category
+        cat = Categorical([-1, 0, 1])
+        result = cat.take([0, -1, 1], allow_fill=True, fill_value=-1)
+        expected = Categorical([-1, -1, 0], categories=[-1, 0, 1])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_take_fill_value(self):
+        # https://github.com/pandas-dev/pandas/issues/23296
+        cat = Categorical(["a", "b", "c"])
+        result = cat.take([0, 1, -1], fill_value="a", allow_fill=True)
+        expected = Categorical(["a", "b", "a"], categories=["a", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_take_fill_value_new_raises(self):
+        # https://github.com/pandas-dev/pandas/issues/23296
+        cat = Categorical(["a", "b", "c"])
+        xpr = r"Cannot setitem on a Categorical with a new category \(d\)"
+        with pytest.raises(TypeError, match=xpr):
+            cat.take([0, 1, -1], fill_value="d", allow_fill=True)
diff --git a/pandas/pandas/tests/arrays/categorical/test_warnings.py b/pandas/pandas/tests/arrays/categorical/test_warnings.py
new file mode 100644
index 0000000000000..68c59706a6c3b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/categorical/test_warnings.py
@@ -0,0 +1,19 @@
+import pytest
+
+import pandas._testing as tm
+
+
+class TestCategoricalWarnings:
+    def test_tab_complete_warning(self, ip):
+        # https://github.com/pandas-dev/pandas/issues/16409
+        pytest.importorskip("IPython", minversion="6.0.0")
+        from IPython.core.completer import provisionalcompleter
+
+        code = "import pandas as pd; c = pd.Categorical([])"
+        ip.run_cell(code)
+
+        # GH 31324 newer jedi version raises Deprecation warning;
+        #  appears resolved 2021-02-02
+        with tm.assert_produces_warning(None, raise_on_extra_warnings=False):
+            with provisionalcompleter("ignore"):
+                list(ip.Completer.completions("c.", 1))
diff --git a/pandas/pandas/tests/arrays/datetimes/__init__.py b/pandas/pandas/tests/arrays/datetimes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/datetimes/test_constructors.py b/pandas/pandas/tests/arrays/datetimes/test_constructors.py
new file mode 100644
index 0000000000000..74cc3e991bb76
--- /dev/null
+++ b/pandas/pandas/tests/arrays/datetimes/test_constructors.py
@@ -0,0 +1,193 @@
+import numpy as np
+import pytest
+
+from pandas._libs import iNaT
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+
+
+class TestDatetimeArrayConstructor:
+    def test_from_sequence_invalid_type(self):
+        mi = pd.MultiIndex.from_product([np.arange(5), np.arange(5)])
+        with pytest.raises(TypeError, match="Cannot create a DatetimeArray"):
+            DatetimeArray._from_sequence(mi, dtype="M8[ns]")
+
+    @pytest.mark.parametrize(
+        "meth",
+        [
+            DatetimeArray._from_sequence,
+            pd.to_datetime,
+            pd.DatetimeIndex,
+        ],
+    )
+    def test_mixing_naive_tzaware_raises(self, meth):
+        # GH#24569
+        arr = np.array([pd.Timestamp("2000"), pd.Timestamp("2000", tz="CET")])
+
+        msg = "|".join(
+            [
+                "Cannot mix tz-aware with tz-naive values",
+                "Tz-aware datetime.datetime cannot be converted "
+                "to datetime64 unless utc=True",
+            ]
+        )
+
+        for obj in [arr, arr[::-1]]:
+            # check that we raise regardless of whether naive is found
+            #  before aware or vice-versa
+            with pytest.raises(ValueError, match=msg):
+                meth(obj)
+
+    def test_from_pandas_array(self):
+        arr = pd.array(np.arange(5, dtype=np.int64)) * 3600 * 10**9
+
+        result = DatetimeArray._from_sequence(arr, dtype="M8[ns]")._with_freq("infer")
+
+        expected = pd.date_range("1970-01-01", periods=5, freq="h")._data
+        tm.assert_datetime_array_equal(result, expected)
+
+    def test_bool_dtype_raises(self):
+        arr = np.array([1, 2, 3], dtype="bool")
+
+        msg = r"dtype bool cannot be converted to datetime64\[ns\]"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeArray._from_sequence(arr, dtype="M8[ns]")
+
+        with pytest.raises(TypeError, match=msg):
+            pd.DatetimeIndex(arr)
+
+        with pytest.raises(TypeError, match=msg):
+            pd.to_datetime(arr)
+
+    def test_copy(self):
+        data = np.array([1, 2, 3], dtype="M8[ns]")
+        arr = DatetimeArray._from_sequence(data, dtype=data.dtype, copy=False)
+        assert arr._ndarray is data
+
+        arr = DatetimeArray._from_sequence(data, dtype=data.dtype, copy=True)
+        assert arr._ndarray is not data
+
+    def test_numpy_datetime_unit(self, unit):
+        data = np.array([1, 2, 3], dtype=f"M8[{unit}]")
+        arr = DatetimeArray._from_sequence(data)
+        assert arr.unit == unit
+        assert arr[0].unit == unit
+
+
+class TestSequenceToDT64NS:
+    def test_tz_dtype_mismatch_raises(self):
+        arr = DatetimeArray._from_sequence(
+            ["2000"], dtype=DatetimeTZDtype(tz="US/Central")
+        )
+        with pytest.raises(TypeError, match="data is already tz-aware"):
+            DatetimeArray._from_sequence(arr, dtype=DatetimeTZDtype(tz="UTC"))
+
+    def test_tz_dtype_matches(self):
+        dtype = DatetimeTZDtype(tz="US/Central")
+        arr = DatetimeArray._from_sequence(["2000"], dtype=dtype)
+        result = DatetimeArray._from_sequence(arr, dtype=dtype)
+        tm.assert_equal(arr, result)
+
+    @pytest.mark.parametrize("order", ["F", "C"])
+    def test_2d(self, order):
+        dti = pd.date_range("2016-01-01", periods=6, tz="US/Pacific")
+        arr = np.array(dti, dtype=object).reshape(3, 2)
+        if order == "F":
+            arr = arr.T
+
+        res = DatetimeArray._from_sequence(arr, dtype=dti.dtype)
+        expected = DatetimeArray._from_sequence(arr.ravel(), dtype=dti.dtype).reshape(
+            arr.shape
+        )
+        tm.assert_datetime_array_equal(res, expected)
+
+
+# ----------------------------------------------------------------------------
+# Arrow interaction
+
+
+EXTREME_VALUES = [0, 123456789, None, iNaT, 2**63 - 1, -(2**63) + 1]
+FINE_TO_COARSE_SAFE = [123_000_000_000, None, -123_000_000_000]
+COARSE_TO_FINE_SAFE = [123, None, -123]
+
+
+@pytest.mark.parametrize(
+    ("pa_unit", "pd_unit", "pa_tz", "pd_tz", "data"),
+    [
+        ("s", "s", "UTC", "UTC", EXTREME_VALUES),
+        ("ms", "ms", "UTC", "Europe/Berlin", EXTREME_VALUES),
+        ("us", "us", "US/Eastern", "UTC", EXTREME_VALUES),
+        ("ns", "ns", "US/Central", "Asia/Kolkata", EXTREME_VALUES),
+        ("ns", "s", "UTC", "UTC", FINE_TO_COARSE_SAFE),
+        ("us", "ms", "UTC", "Europe/Berlin", FINE_TO_COARSE_SAFE),
+        ("ms", "us", "US/Eastern", "UTC", COARSE_TO_FINE_SAFE),
+        ("s", "ns", "US/Central", "Asia/Kolkata", COARSE_TO_FINE_SAFE),
+    ],
+)
+def test_from_arrow_with_different_units_and_timezones_with(
+    pa_unit, pd_unit, pa_tz, pd_tz, data
+):
+    pa = pytest.importorskip("pyarrow")
+
+    pa_type = pa.timestamp(pa_unit, tz=pa_tz)
+    arr = pa.array(data, type=pa_type)
+    dtype = DatetimeTZDtype(unit=pd_unit, tz=pd_tz)
+
+    result = dtype.__from_arrow__(arr)
+    expected = DatetimeArray._from_sequence(data, dtype=f"M8[{pa_unit}, UTC]").astype(
+        dtype, copy=False
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    result = dtype.__from_arrow__(pa.chunked_array([arr]))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("unit", "tz"),
+    [
+        ("s", "UTC"),
+        ("ms", "Europe/Berlin"),
+        ("us", "US/Eastern"),
+        ("ns", "Asia/Kolkata"),
+        ("ns", "UTC"),
+    ],
+)
+def test_from_arrow_from_empty(unit, tz):
+    pa = pytest.importorskip("pyarrow")
+
+    data = []
+    arr = pa.array(data)
+    dtype = DatetimeTZDtype(unit=unit, tz=tz)
+
+    result = dtype.__from_arrow__(arr)
+    expected = DatetimeArray._from_sequence(
+        np.array(data, dtype=f"datetime64[{unit}]"), dtype=np.dtype(f"M8[{unit}]")
+    )
+    expected = expected.tz_localize(tz=tz)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = dtype.__from_arrow__(pa.chunked_array([arr]))
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_from_arrow_from_integers():
+    pa = pytest.importorskip("pyarrow")
+
+    data = [0, 123456789, None, 2**63 - 1, iNaT, -123456789]
+    arr = pa.array(data)
+    dtype = DatetimeTZDtype(unit="ns", tz="UTC")
+
+    result = dtype.__from_arrow__(arr)
+    expected = DatetimeArray._from_sequence(
+        np.array(data, dtype="datetime64[ns]"), dtype=np.dtype("M8[ns]")
+    )
+    expected = expected.tz_localize("UTC")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = dtype.__from_arrow__(pa.chunked_array([arr]))
+    tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/datetimes/test_cumulative.py b/pandas/pandas/tests/arrays/datetimes/test_cumulative.py
new file mode 100644
index 0000000000000..e9d2dfdd0048a
--- /dev/null
+++ b/pandas/pandas/tests/arrays/datetimes/test_cumulative.py
@@ -0,0 +1,44 @@
+import pytest
+
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+
+
+class TestAccumulator:
+    def test_accumulators_freq(self):
+        # GH#50297
+        arr = DatetimeArray._from_sequence(
+            [
+                "2000-01-01",
+                "2000-01-02",
+                "2000-01-03",
+            ],
+            dtype="M8[ns]",
+        )._with_freq("infer")
+        result = arr._accumulate("cummin")
+        expected = DatetimeArray._from_sequence(["2000-01-01"] * 3, dtype="M8[ns]")
+        tm.assert_datetime_array_equal(result, expected)
+
+        result = arr._accumulate("cummax")
+        expected = DatetimeArray._from_sequence(
+            [
+                "2000-01-01",
+                "2000-01-02",
+                "2000-01-03",
+            ],
+            dtype="M8[ns]",
+        )
+        tm.assert_datetime_array_equal(result, expected)
+
+    @pytest.mark.parametrize("func", ["cumsum", "cumprod"])
+    def test_accumulators_disallowed(self, func):
+        # GH#50297
+        arr = DatetimeArray._from_sequence(
+            [
+                "2000-01-01",
+                "2000-01-02",
+            ],
+            dtype="M8[ns]",
+        )._with_freq("infer")
+        with pytest.raises(TypeError, match=f"Accumulation {func}"):
+            arr._accumulate(func)
diff --git a/pandas/pandas/tests/arrays/datetimes/test_reductions.py b/pandas/pandas/tests/arrays/datetimes/test_reductions.py
new file mode 100644
index 0000000000000..a58b0f57bf0da
--- /dev/null
+++ b/pandas/pandas/tests/arrays/datetimes/test_reductions.py
@@ -0,0 +1,176 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+from pandas import NaT
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+
+
+class TestReductions:
+    @pytest.fixture
+    def arr1d(self, tz_naive_fixture):
+        """Fixture returning DatetimeArray with parametrized timezones"""
+        tz = tz_naive_fixture
+        dtype = DatetimeTZDtype(tz=tz) if tz is not None else np.dtype("M8[ns]")
+        arr = DatetimeArray._from_sequence(
+            [
+                "2000-01-03",
+                "2000-01-03",
+                "NaT",
+                "2000-01-02",
+                "2000-01-05",
+                "2000-01-04",
+            ],
+            dtype=dtype,
+        )
+        return arr
+
+    def test_min_max(self, arr1d, unit):
+        arr = arr1d
+        arr = arr.as_unit(unit)
+        tz = arr.tz
+
+        result = arr.min()
+        expected = pd.Timestamp("2000-01-02", tz=tz).as_unit(unit)
+        assert result == expected
+        assert result.unit == expected.unit
+
+        result = arr.max()
+        expected = pd.Timestamp("2000-01-05", tz=tz).as_unit(unit)
+        assert result == expected
+        assert result.unit == expected.unit
+
+        result = arr.min(skipna=False)
+        assert result is NaT
+
+        result = arr.max(skipna=False)
+        assert result is NaT
+
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_min_max_empty(self, skipna, tz):
+        dtype = DatetimeTZDtype(tz=tz) if tz is not None else np.dtype("M8[ns]")
+        arr = DatetimeArray._from_sequence([], dtype=dtype)
+        result = arr.min(skipna=skipna)
+        assert result is NaT
+
+        result = arr.max(skipna=skipna)
+        assert result is NaT
+
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_median_empty(self, skipna, tz):
+        dtype = DatetimeTZDtype(tz=tz) if tz is not None else np.dtype("M8[ns]")
+        arr = DatetimeArray._from_sequence([], dtype=dtype)
+        result = arr.median(skipna=skipna)
+        assert result is NaT
+
+        arr = arr.reshape(0, 3)
+        result = arr.median(axis=0, skipna=skipna)
+        expected = type(arr)._from_sequence([NaT, NaT, NaT], dtype=arr.dtype)
+        tm.assert_equal(result, expected)
+
+        result = arr.median(axis=1, skipna=skipna)
+        expected = type(arr)._from_sequence([], dtype=arr.dtype)
+        tm.assert_equal(result, expected)
+
+    def test_median(self, arr1d):
+        arr = arr1d
+
+        result = arr.median()
+        assert result == arr[0]
+        result = arr.median(skipna=False)
+        assert result is NaT
+
+        result = arr.dropna().median(skipna=False)
+        assert result == arr[0]
+
+        result = arr.median(axis=0)
+        assert result == arr[0]
+
+    def test_median_axis(self, arr1d):
+        arr = arr1d
+        assert arr.median(axis=0) == arr.median()
+        assert arr.median(axis=0, skipna=False) is NaT
+
+        msg = r"abs\(axis\) must be less than ndim"
+        with pytest.raises(ValueError, match=msg):
+            arr.median(axis=1)
+
+    @pytest.mark.filterwarnings("ignore:All-NaN slice encountered:RuntimeWarning")
+    def test_median_2d(self, arr1d):
+        arr = arr1d.reshape(1, -1)
+
+        # axis = None
+        assert arr.median() == arr1d.median()
+        assert arr.median(skipna=False) is NaT
+
+        # axis = 0
+        result = arr.median(axis=0)
+        expected = arr1d
+        tm.assert_equal(result, expected)
+
+        # Since column 3 is all-NaT, we get NaT there with or without skipna
+        result = arr.median(axis=0, skipna=False)
+        expected = arr1d
+        tm.assert_equal(result, expected)
+
+        # axis = 1
+        result = arr.median(axis=1)
+        expected = type(arr)._from_sequence([arr1d.median()], dtype=arr.dtype)
+        tm.assert_equal(result, expected)
+
+        result = arr.median(axis=1, skipna=False)
+        expected = type(arr)._from_sequence([NaT], dtype=arr.dtype)
+        tm.assert_equal(result, expected)
+
+    def test_mean(self, arr1d):
+        arr = arr1d
+
+        # manually verified result
+        expected = arr[0] + 0.4 * pd.Timedelta(days=1)
+
+        result = arr.mean()
+        assert result == expected
+        result = arr.mean(skipna=False)
+        assert result is NaT
+
+        result = arr.dropna().mean(skipna=False)
+        assert result == expected
+
+        result = arr.mean(axis=0)
+        assert result == expected
+
+    def test_mean_2d(self):
+        dti = pd.date_range("2016-01-01", periods=6, tz="US/Pacific")
+        dta = dti._data.reshape(3, 2)
+
+        result = dta.mean(axis=0)
+        expected = dta[1]
+        tm.assert_datetime_array_equal(result, expected)
+
+        result = dta.mean(axis=1)
+        expected = dta[:, 0] + pd.Timedelta(hours=12)
+        tm.assert_datetime_array_equal(result, expected)
+
+        result = dta.mean(axis=None)
+        expected = dti.mean()
+        assert result == expected
+
+    def test_mean_empty(self, arr1d, skipna):
+        arr = arr1d[:0]
+
+        assert arr.mean(skipna=skipna) is NaT
+
+        arr2d = arr.reshape(0, 3)
+        result = arr2d.mean(axis=0, skipna=skipna)
+        expected = DatetimeArray._from_sequence([NaT, NaT, NaT], dtype=arr.dtype)
+        tm.assert_datetime_array_equal(result, expected)
+
+        result = arr2d.mean(axis=1, skipna=skipna)
+        expected = arr  # i.e. 1D, empty
+        tm.assert_datetime_array_equal(result, expected)
+
+        result = arr2d.mean(axis=None, skipna=skipna)
+        assert result is NaT
diff --git a/pandas/pandas/tests/arrays/floating/__init__.py b/pandas/pandas/tests/arrays/floating/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/floating/conftest.py b/pandas/pandas/tests/arrays/floating/conftest.py
new file mode 100644
index 0000000000000..5e971c66029d5
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/conftest.py
@@ -0,0 +1,48 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas.core.arrays.floating import (
+    Float32Dtype,
+    Float64Dtype,
+)
+
+
+@pytest.fixture(params=[Float32Dtype, Float64Dtype])
+def dtype(request):
+    """Parametrized fixture returning a float 'dtype'"""
+    return request.param()
+
+
+@pytest.fixture
+def data(dtype):
+    """Fixture returning 'data' array according to parametrized float 'dtype'"""
+    return pd.array(
+        list(np.arange(0.1, 0.9, 0.1))
+        + [pd.NA]
+        + list(np.arange(1, 9.8, 0.1))
+        + [pd.NA]
+        + [9.9, 10.0],
+        dtype=dtype,
+    )
+
+
+@pytest.fixture
+def data_missing(dtype):
+    """
+    Fixture returning array with missing data according to parametrized float
+    'dtype'.
+    """
+    return pd.array([np.nan, 0.1], dtype=dtype)
+
+
+@pytest.fixture(params=["data", "data_missing"])
+def all_data(request, data, data_missing):
+    """Parametrized fixture returning 'data' or 'data_missing' float arrays.
+
+    Used to test dtype conversion with and without missing values.
+    """
+    if request.param == "data":
+        return data
+    elif request.param == "data_missing":
+        return data_missing
diff --git a/pandas/pandas/tests/arrays/floating/test_arithmetic.py b/pandas/pandas/tests/arrays/floating/test_arithmetic.py
new file mode 100644
index 0000000000000..009fac4c2f5ed
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_arithmetic.py
@@ -0,0 +1,240 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import FloatingArray
+
+# Basic test for the arithmetic array ops
+# -----------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize(
+    "opname, exp",
+    [
+        ("add", [1.1, 2.2, None, None, 5.5]),
+        ("mul", [0.1, 0.4, None, None, 2.5]),
+        ("sub", [0.9, 1.8, None, None, 4.5]),
+        ("truediv", [10.0, 10.0, None, None, 10.0]),
+        ("floordiv", [9.0, 9.0, None, None, 10.0]),
+        ("mod", [0.1, 0.2, None, None, 0.0]),
+    ],
+    ids=["add", "mul", "sub", "div", "floordiv", "mod"],
+)
+def test_array_op(dtype, opname, exp):
+    a = pd.array([1.0, 2.0, None, 4.0, 5.0], dtype=dtype)
+    b = pd.array([0.1, 0.2, 0.3, None, 0.5], dtype=dtype)
+
+    op = getattr(operator, opname)
+
+    result = op(a, b)
+    expected = pd.array(exp, dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("zero, negative", [(0, False), (0.0, False), (-0.0, True)])
+def test_divide_by_zero(dtype, zero, negative):
+    # TODO pending NA/NaN discussion
+    # https://github.com/pandas-dev/pandas/issues/32265/
+    a = pd.array([0, 1, -1, None], dtype=dtype)
+    result = a / zero
+    expected = FloatingArray(
+        np.array([np.nan, np.inf, -np.inf, np.nan], dtype=dtype.numpy_dtype),
+        np.array([False, False, False, True]),
+    )
+    if negative:
+        expected *= -1
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_pow_scalar(dtype):
+    a = pd.array([-1, 0, 1, None, 2], dtype=dtype)
+    result = a**0
+    expected = pd.array([1, 1, 1, 1, 1], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**1
+    expected = pd.array([-1, 0, 1, None, 2], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**pd.NA
+    expected = pd.array([None, None, 1, None, None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**np.nan
+    # TODO np.nan should be converted to pd.NA / missing before operation?
+    expected = FloatingArray(
+        np.array([np.nan, np.nan, 1, np.nan, np.nan], dtype=dtype.numpy_dtype),
+        mask=a._mask,
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    # reversed
+    a = a[1:]  # Can't raise integers to negative powers.
+
+    result = 0**a
+    expected = pd.array([1, 0, None, 0], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = 1**a
+    expected = pd.array([1, 1, 1, 1], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = pd.NA**a
+    expected = pd.array([1, None, None, None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = np.nan**a
+    expected = FloatingArray(
+        np.array([1, np.nan, np.nan, np.nan], dtype=dtype.numpy_dtype), mask=a._mask
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_pow_array(dtype):
+    a = pd.array([0, 0, 0, 1, 1, 1, None, None, None], dtype=dtype)
+    b = pd.array([0, 1, None, 0, 1, None, 0, 1, None], dtype=dtype)
+    result = a**b
+    expected = pd.array([1, 0, None, 1, 1, 1, 1, None, None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_rpow_one_to_na():
+    # https://github.com/pandas-dev/pandas/issues/22022
+    # https://github.com/pandas-dev/pandas/issues/29997
+    arr = pd.array([np.nan, np.nan], dtype="Float64")
+    result = np.array([1.0, 2.0]) ** arr
+    expected = pd.array([1.0, np.nan], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("other", [0, 0.5])
+def test_arith_zero_dim_ndarray(other):
+    arr = pd.array([1, None, 2], dtype="Float64")
+    result = arr + np.array(other)
+    expected = arr + other
+    tm.assert_equal(result, expected)
+
+
+# Test generic characteristics / errors
+# -----------------------------------------------------------------------------
+
+
+def test_error_invalid_values(data, all_arithmetic_operators):
+    op = all_arithmetic_operators
+    s = pd.Series(data)
+    ops = getattr(s, op)
+
+    # invalid scalars
+    msg = "|".join(
+        [
+            r"can only perform ops with numeric values",
+            r"FloatingArray cannot perform the operation mod",
+            "unsupported operand type",
+            "not all arguments converted during string formatting",
+            "can't multiply sequence by non-int of type 'float'",
+            "ufunc 'subtract' cannot use operands with types dtype",
+            r"can only concatenate str \(not \"float\"\) to str",
+            "ufunc '.*' not supported for the input types, and the inputs could not",
+            "ufunc '.*' did not contain a loop with signature matching types",
+            "Concatenation operation is not implemented for NumPy arrays",
+            "has no kernel",
+            "not implemented",
+            "not supported for dtype",
+            "Can only string multiply by an integer",
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops("foo")
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Timestamp("20180101"))
+
+    # invalid array-likes
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Series("foo", index=s.index))
+
+    msg = "|".join(
+        [
+            "can only perform ops with numeric values",
+            "cannot perform .* with this index type: DatetimeArray",
+            "Addition/subtraction of integers and integer-arrays "
+            "with DatetimeArray is no longer supported. *",
+            "unsupported operand type",
+            "not all arguments converted during string formatting",
+            "can't multiply sequence by non-int of type 'float'",
+            "ufunc 'subtract' cannot use operands with types dtype",
+            (
+                "ufunc 'add' cannot use operands with types "
+                rf"dtype\('{tm.ENDIAN}M8\[ns\]'\)"
+            ),
+            r"ufunc 'add' cannot use operands with types dtype\('float\d{2}'\)",
+            "cannot subtract DatetimeArray from ndarray",
+            "has no kernel",
+            "not implemented",
+            "not supported for dtype",
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Series(pd.date_range("20180101", periods=len(s))))
+
+
+# Various
+# -----------------------------------------------------------------------------
+
+
+def test_cross_type_arithmetic():
+    df = pd.DataFrame(
+        {
+            "A": pd.array([1, 2, np.nan], dtype="Float64"),
+            "B": pd.array([1, np.nan, 3], dtype="Float32"),
+            "C": np.array([1, 2, 3], dtype="float64"),
+        }
+    )
+
+    result = df.A + df.C
+    expected = pd.Series([2, 4, np.nan], dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+    result = (df.A + df.C) * 3 == 12
+    expected = pd.Series([False, True, None], dtype="boolean")
+    tm.assert_series_equal(result, expected)
+
+    result = df.A + df.B
+    expected = pd.Series([2, np.nan, np.nan], dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "source, neg_target, abs_target",
+    [
+        ([1.1, 2.2, 3.3], [-1.1, -2.2, -3.3], [1.1, 2.2, 3.3]),
+        ([1.1, 2.2, None], [-1.1, -2.2, None], [1.1, 2.2, None]),
+        ([-1.1, 0.0, 1.1], [1.1, 0.0, -1.1], [1.1, 0.0, 1.1]),
+    ],
+)
+def test_unary_float_operators(float_ea_dtype, source, neg_target, abs_target):
+    # GH38794
+    dtype = float_ea_dtype
+    arr = pd.array(source, dtype=dtype)
+    neg_result, pos_result, abs_result = -arr, +arr, abs(arr)
+    neg_target = pd.array(neg_target, dtype=dtype)
+    abs_target = pd.array(abs_target, dtype=dtype)
+
+    tm.assert_extension_array_equal(neg_result, neg_target)
+    tm.assert_extension_array_equal(pos_result, arr)
+    assert not tm.shares_memory(pos_result, arr)
+    tm.assert_extension_array_equal(abs_result, abs_target)
+
+
+def test_bitwise(dtype):
+    left = pd.array([1, None, 3, 4], dtype=dtype)
+    right = pd.array([None, 3, 5, 4], dtype=dtype)
+
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left | right
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left & right
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left ^ right
diff --git a/pandas/pandas/tests/arrays/floating/test_astype.py b/pandas/pandas/tests/arrays/floating/test_astype.py
new file mode 100644
index 0000000000000..752ebe194ffcf
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_astype.py
@@ -0,0 +1,135 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_astype():
+    # with missing values
+    arr = pd.array([0.1, 0.2, None], dtype="Float64")
+
+    with pytest.raises(ValueError, match="cannot convert NA to integer"):
+        arr.astype("int64")
+
+    with pytest.raises(ValueError, match="cannot convert float NaN to bool"):
+        arr.astype("bool")
+
+    result = arr.astype("float64")
+    expected = np.array([0.1, 0.2, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # no missing values
+    arr = pd.array([0.0, 1.0, 0.5], dtype="Float64")
+    result = arr.astype("int64")
+    expected = np.array([0, 1, 0], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.astype("bool")
+    expected = np.array([False, True, True], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_astype_to_floating_array():
+    # astype to FloatingArray
+    arr = pd.array([0.0, 1.0, None], dtype="Float64")
+
+    result = arr.astype("Float64")
+    tm.assert_extension_array_equal(result, arr)
+    result = arr.astype(pd.Float64Dtype())
+    tm.assert_extension_array_equal(result, arr)
+    result = arr.astype("Float32")
+    expected = pd.array([0.0, 1.0, None], dtype="Float32")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_astype_to_boolean_array():
+    # astype to BooleanArray
+    arr = pd.array([0.0, 1.0, None], dtype="Float64")
+
+    result = arr.astype("boolean")
+    expected = pd.array([False, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
+    result = arr.astype(pd.BooleanDtype())
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_astype_to_integer_array():
+    # astype to IntegerArray
+    arr = pd.array([0.0, 1.5, None], dtype="Float64")
+
+    result = arr.astype("Int64")
+    expected = pd.array([0, 1, None], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_astype_str(using_infer_string):
+    a = pd.array([0.1, 0.2, None], dtype="Float64")
+
+    if using_infer_string:
+        expected = pd.array(["0.1", "0.2", None], dtype=pd.StringDtype(na_value=np.nan))
+
+        tm.assert_extension_array_equal(a.astype(str), expected)
+        tm.assert_extension_array_equal(a.astype("str"), expected)
+    else:
+        expected = np.array(["0.1", "0.2", "<NA>"], dtype="U32")
+
+        tm.assert_numpy_array_equal(a.astype(str), expected)
+        tm.assert_numpy_array_equal(a.astype("str"), expected)
+
+
+def test_astype_copy():
+    arr = pd.array([0.1, 0.2, None], dtype="Float64")
+    orig = pd.array([0.1, 0.2, None], dtype="Float64")
+
+    # copy=True -> ensure both data and mask are actual copies
+    result = arr.astype("Float64", copy=True)
+    assert result is not arr
+    assert not tm.shares_memory(result, arr)
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, orig)
+    result[0] = pd.NA
+    tm.assert_extension_array_equal(arr, orig)
+
+    # copy=False
+    result = arr.astype("Float64", copy=False)
+    assert result is arr
+    assert np.shares_memory(result._data, arr._data)
+    assert np.shares_memory(result._mask, arr._mask)
+    result[0] = 10
+    assert arr[0] == 10
+    result[0] = pd.NA
+    assert arr[0] is pd.NA
+
+    # astype to different dtype -> always needs a copy -> even with copy=False
+    # we need to ensure that also the mask is actually copied
+    arr = pd.array([0.1, 0.2, None], dtype="Float64")
+    orig = pd.array([0.1, 0.2, None], dtype="Float64")
+
+    result = arr.astype("Float32", copy=False)
+    assert not tm.shares_memory(result, arr)
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, orig)
+    result[0] = pd.NA
+    tm.assert_extension_array_equal(arr, orig)
+
+
+def test_astype_object(dtype):
+    arr = pd.array([1.0, pd.NA], dtype=dtype)
+
+    result = arr.astype(object)
+    expected = np.array([1.0, pd.NA], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+    # check exact element types
+    assert isinstance(result[0], float)
+    assert result[1] is pd.NA
+
+
+def test_Float64_conversion():
+    # GH#40729
+    testseries = pd.Series(["1", "2", "3", "4"], dtype="object")
+    result = testseries.astype(pd.Float64Dtype())
+
+    expected = pd.Series([1.0, 2.0, 3.0, 4.0], dtype=pd.Float64Dtype())
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/floating/test_comparison.py b/pandas/pandas/tests/arrays/floating/test_comparison.py
new file mode 100644
index 0000000000000..a429649f1ce1d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_comparison.py
@@ -0,0 +1,65 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import FloatingArray
+from pandas.tests.arrays.masked_shared import (
+    ComparisonOps,
+    NumericOps,
+)
+
+
+class TestComparisonOps(NumericOps, ComparisonOps):
+    @pytest.mark.parametrize("other", [True, False, pd.NA, -1.0, 0.0, 1])
+    def test_scalar(self, other, comparison_op, dtype):
+        ComparisonOps.test_scalar(self, other, comparison_op, dtype)
+
+    def test_compare_with_integerarray(self, comparison_op):
+        op = comparison_op
+        a = pd.array([0, 1, None] * 3, dtype="Int64")
+        b = pd.array([0] * 3 + [1] * 3 + [None] * 3, dtype="Float64")
+        other = b.astype("Int64")
+        expected = op(a, other)
+        result = op(a, b)
+        tm.assert_extension_array_equal(result, expected)
+        expected = op(other, a)
+        result = op(b, a)
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_equals():
+    # GH-30652
+    # equals is generally tested in /tests/extension/base/methods, but this
+    # specifically tests that two arrays of the same class but different dtype
+    # do not evaluate equal
+    a1 = pd.array([1, 2, None], dtype="Float64")
+    a2 = pd.array([1, 2, None], dtype="Float32")
+    assert a1.equals(a2) is False
+
+
+def test_equals_nan_vs_na():
+    # GH#44382
+
+    mask = np.zeros(3, dtype=bool)
+    data = np.array([1.0, np.nan, 3.0], dtype=np.float64)
+
+    left = FloatingArray(data, mask)
+    assert left.equals(left)
+    tm.assert_extension_array_equal(left, left)
+
+    assert left.equals(left.copy())
+    assert left.equals(FloatingArray(data.copy(), mask.copy()))
+
+    mask2 = np.array([False, True, False], dtype=bool)
+    data2 = np.array([1.0, 2.0, 3.0], dtype=np.float64)
+    right = FloatingArray(data2, mask2)
+    assert right.equals(right)
+    tm.assert_extension_array_equal(right, right)
+
+    assert not left.equals(right)
+
+    # with mask[1] = True, the only difference is data[1], which should
+    #  not matter for equals
+    mask[1] = True
+    assert left.equals(right)
diff --git a/pandas/pandas/tests/arrays/floating/test_concat.py b/pandas/pandas/tests/arrays/floating/test_concat.py
new file mode 100644
index 0000000000000..2174a834aa959
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_concat.py
@@ -0,0 +1,20 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "to_concat_dtypes, result_dtype",
+    [
+        (["Float64", "Float64"], "Float64"),
+        (["Float32", "Float64"], "Float64"),
+        (["Float32", "Float32"], "Float32"),
+    ],
+)
+def test_concat_series(to_concat_dtypes, result_dtype):
+    result = pd.concat([pd.Series([1, 2, pd.NA], dtype=t) for t in to_concat_dtypes])
+    expected = pd.concat([pd.Series([1, 2, pd.NA], dtype=object)] * 2).astype(
+        result_dtype
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/floating/test_construction.py b/pandas/pandas/tests/arrays/floating/test_construction.py
new file mode 100644
index 0000000000000..4007ee6b415c9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_construction.py
@@ -0,0 +1,204 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import FloatingArray
+from pandas.core.arrays.floating import (
+    Float32Dtype,
+    Float64Dtype,
+)
+
+
+def test_uses_pandas_na():
+    a = pd.array([1, None], dtype=Float64Dtype())
+    assert a[1] is pd.NA
+
+
+def test_floating_array_constructor():
+    values = np.array([1, 2, 3, 4], dtype="float64")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = FloatingArray(values, mask)
+    expected = pd.array([1, 2, 3, np.nan], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+    tm.assert_numpy_array_equal(result._data, values)
+    tm.assert_numpy_array_equal(result._mask, mask)
+
+    msg = r".* should be .* numpy array. Use the 'pd.array' function instead"
+    with pytest.raises(TypeError, match=msg):
+        FloatingArray(values.tolist(), mask)
+
+    with pytest.raises(TypeError, match=msg):
+        FloatingArray(values, mask.tolist())
+
+    with pytest.raises(TypeError, match=msg):
+        FloatingArray(values.astype(int), mask)
+
+    msg = r"__init__\(\) missing 1 required positional argument: 'mask'"
+    with pytest.raises(TypeError, match=msg):
+        FloatingArray(values)
+
+
+def test_floating_array_disallows_float16():
+    # GH#44715
+    arr = np.array([1, 2], dtype=np.float16)
+    mask = np.array([False, False])
+
+    msg = "FloatingArray does not support np.float16 dtype"
+    with pytest.raises(TypeError, match=msg):
+        FloatingArray(arr, mask)
+
+
+def test_floating_array_disallows_Float16_dtype(request):
+    # GH#44715
+    with pytest.raises(TypeError, match="data type 'Float16' not understood"):
+        pd.array([1.0, 2.0], dtype="Float16")
+
+
+def test_floating_array_constructor_copy():
+    values = np.array([1, 2, 3, 4], dtype="float64")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = FloatingArray(values, mask)
+    assert result._data is values
+    assert result._mask is mask
+
+    result = FloatingArray(values, mask, copy=True)
+    assert result._data is not values
+    assert result._mask is not mask
+
+
+def test_to_array():
+    result = pd.array([0.1, 0.2, 0.3, 0.4])
+    expected = pd.array([0.1, 0.2, 0.3, 0.4], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    [
+        ([1, None], [1, pd.NA]),
+        ([None], [pd.NA]),
+        ([None, np.nan], [pd.NA, pd.NA]),
+        ([1, np.nan], [1, pd.NA]),
+        ([np.nan], [pd.NA]),
+    ],
+)
+def test_to_array_none_is_nan(a, b):
+    result = pd.array(a, dtype="Float64")
+    expected = pd.array(b, dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_array_mixed_integer_float():
+    result = pd.array([1, 2.0])
+    expected = pd.array([1.0, 2.0], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = pd.array([1, None, 2.0])
+    expected = pd.array([1.0, None, 2.0], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        ["foo", "bar"],
+        "foo",
+        1,
+        1.0,
+        pd.date_range("20130101", periods=2),
+        np.array(["foo"]),
+        [[1, 2], [3, 4]],
+        [np.nan, {"a": 1}],
+        # GH#44514 all-NA case used to get quietly swapped out before checking ndim
+        np.array([pd.NA] * 6, dtype=object).reshape(3, 2),
+    ],
+)
+def test_to_array_error(values):
+    # error in converting existing arrays to FloatingArray
+    msg = "|".join(
+        [
+            "cannot be converted to FloatingDtype",
+            "values must be a 1D list-like",
+            "Cannot pass scalar",
+            r"float\(\) argument must be a string or a (real )?number, not 'dict'",
+            "could not convert string to float: 'foo'",
+            r"could not convert string to float: np\.str_\('foo'\)",
+        ]
+    )
+    with pytest.raises((TypeError, ValueError), match=msg):
+        pd.array(values, dtype="Float64")
+
+
+@pytest.mark.parametrize("values", [["1", "2", None], ["1.5", "2", None]])
+def test_construct_from_float_strings(values):
+    # see also test_to_integer_array_str
+    expected = pd.array([float(values[0]), 2, None], dtype="Float64")
+
+    res = pd.array(values, dtype="Float64")
+    tm.assert_extension_array_equal(res, expected)
+
+    res = FloatingArray._from_sequence(values)
+    tm.assert_extension_array_equal(res, expected)
+
+
+def test_to_array_inferred_dtype():
+    # if values has dtype -> respect it
+    result = pd.array(np.array([1, 2], dtype="float32"))
+    assert result.dtype == Float32Dtype()
+
+    # if values have no dtype -> always float64
+    result = pd.array([1.0, 2.0])
+    assert result.dtype == Float64Dtype()
+
+
+def test_to_array_dtype_keyword():
+    result = pd.array([1, 2], dtype="Float32")
+    assert result.dtype == Float32Dtype()
+
+    # if values has dtype -> override it
+    result = pd.array(np.array([1, 2], dtype="float32"), dtype="Float64")
+    assert result.dtype == Float64Dtype()
+
+
+def test_to_array_integer():
+    result = pd.array([1, 2], dtype="Float64")
+    expected = pd.array([1.0, 2.0], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    # for integer dtypes, the itemsize is not preserved
+    # TODO can we specify "floating" in general?
+    result = pd.array(np.array([1, 2], dtype="int32"), dtype="Float64")
+    assert result.dtype == Float64Dtype()
+
+
+@pytest.mark.parametrize(
+    "bool_values, values, target_dtype, expected_dtype",
+    [
+        ([False, True], [0, 1], Float64Dtype(), Float64Dtype()),
+        ([False, True], [0, 1], "Float64", Float64Dtype()),
+        ([False, True, np.nan], [0, 1, np.nan], Float64Dtype(), Float64Dtype()),
+    ],
+)
+def test_to_array_bool(bool_values, values, target_dtype, expected_dtype):
+    result = pd.array(bool_values, dtype=target_dtype)
+    assert result.dtype == expected_dtype
+    expected = pd.array(values, dtype=target_dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_series_from_float(data):
+    # construct from our dtype & string dtype
+    dtype = data.dtype
+
+    # from float
+    expected = pd.Series(data)
+    result = pd.Series(data.to_numpy(na_value=np.nan, dtype="float"), dtype=str(dtype))
+    tm.assert_series_equal(result, expected)
+
+    # from list
+    expected = pd.Series(data)
+    result = pd.Series(np.array(data).tolist(), dtype=str(dtype))
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/floating/test_contains.py b/pandas/pandas/tests/arrays/floating/test_contains.py
new file mode 100644
index 0000000000000..956642697bf32
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_contains.py
@@ -0,0 +1,12 @@
+import numpy as np
+
+import pandas as pd
+
+
+def test_contains_nan():
+    # GH#52840
+    arr = pd.array(range(5)) / 0
+
+    assert np.isnan(arr._data[0])
+    assert not arr.isna()[0]
+    assert np.nan in arr
diff --git a/pandas/pandas/tests/arrays/floating/test_function.py b/pandas/pandas/tests/arrays/floating/test_function.py
new file mode 100644
index 0000000000000..9ea48bfb2413f
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_function.py
@@ -0,0 +1,191 @@
+import numpy as np
+import pytest
+
+from pandas.compat import IS64
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("ufunc", [np.abs, np.sign])
+# np.sign emits a warning with nans, <https://github.com/numpy/numpy/issues/15127>
+@pytest.mark.filterwarnings("ignore:invalid value encountered in sign:RuntimeWarning")
+def test_ufuncs_single(ufunc):
+    a = pd.array([1, 2, -3, np.nan], dtype="Float64")
+    result = ufunc(a)
+    expected = pd.array(ufunc(a.astype(float)), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    s = pd.Series(a)
+    result = ufunc(s)
+    expected = pd.Series(expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.log, np.exp, np.sin, np.cos, np.sqrt])
+def test_ufuncs_single_float(ufunc):
+    a = pd.array([1.0, 0.2, 3.0, np.nan], dtype="Float64")
+    with np.errstate(invalid="ignore"):
+        result = ufunc(a)
+        expected = pd.array(ufunc(a.astype(float)), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    s = pd.Series(a)
+    with np.errstate(invalid="ignore"):
+        result = ufunc(s)
+        expected = pd.Series(ufunc(s.astype(float)), dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.add, np.subtract])
+def test_ufuncs_binary_float(ufunc):
+    # two FloatingArrays
+    a = pd.array([1, 0.2, -3, np.nan], dtype="Float64")
+    result = ufunc(a, a)
+    expected = pd.array(ufunc(a.astype(float), a.astype(float)), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    # FloatingArray with numpy array
+    arr = np.array([1, 2, 3, 4])
+    result = ufunc(a, arr)
+    expected = pd.array(ufunc(a.astype(float), arr), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(arr, a)
+    expected = pd.array(ufunc(arr, a.astype(float)), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    # FloatingArray with scalar
+    result = ufunc(a, 1)
+    expected = pd.array(ufunc(a.astype(float), 1), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(1, a)
+    expected = pd.array(ufunc(1, a.astype(float)), dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("values", [[0, 1], [0, None]])
+def test_ufunc_reduce_raises(values):
+    arr = pd.array(values, dtype="Float64")
+
+    res = np.add.reduce(arr)
+    expected = arr.sum(skipna=False)
+    tm.assert_almost_equal(res, expected)
+
+
+@pytest.mark.skipif(not IS64, reason="GH 36579: fail on 32-bit system")
+@pytest.mark.parametrize(
+    "pandasmethname, kwargs",
+    [
+        ("var", {"ddof": 0}),
+        ("var", {"ddof": 1}),
+        ("std", {"ddof": 0}),
+        ("std", {"ddof": 1}),
+        ("kurtosis", {}),
+        ("skew", {}),
+        ("sem", {}),
+    ],
+)
+def test_stat_method(pandasmethname, kwargs):
+    s = pd.Series(data=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, np.nan, np.nan], dtype="Float64")
+    pandasmeth = getattr(s, pandasmethname)
+    result = pandasmeth(**kwargs)
+    s2 = pd.Series(data=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6], dtype="float64")
+    pandasmeth = getattr(s2, pandasmethname)
+    expected = pandasmeth(**kwargs)
+    assert expected == result
+
+
+def test_value_counts_na():
+    arr = pd.array([0.1, 0.2, 0.1, pd.NA], dtype="Float64")
+    result = arr.value_counts(dropna=False)
+    idx = pd.Index([0.1, 0.2, pd.NA], dtype=arr.dtype)
+    assert idx.dtype == arr.dtype
+    expected = pd.Series([2, 1, 1], index=idx, dtype="Int64", name="count")
+    tm.assert_series_equal(result, expected)
+
+    result = arr.value_counts(dropna=True)
+    expected = pd.Series([2, 1], index=idx[:-1], dtype="Int64", name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_empty():
+    ser = pd.Series([], dtype="Float64")
+    result = ser.value_counts()
+    idx = pd.Index([], dtype="Float64")
+    assert idx.dtype == "Float64"
+    expected = pd.Series([], index=idx, dtype="Int64", name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_with_normalize():
+    ser = pd.Series([0.1, 0.2, 0.1, pd.NA], dtype="Float64")
+    result = ser.value_counts(normalize=True)
+    expected = pd.Series([2, 1], index=ser[:2], dtype="Float64", name="proportion") / 3
+    assert expected.index.dtype == ser.dtype
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("min_count", [0, 4])
+def test_floating_array_sum(skipna, min_count, dtype):
+    arr = pd.array([1, 2, 3, None], dtype=dtype)
+    result = arr.sum(skipna=skipna, min_count=min_count)
+    if skipna and min_count == 0:
+        assert result == 6.0
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize(
+    "values, expected", [([1, 2, 3], 6.0), ([1, 2, 3, None], 6.0), ([None], 0.0)]
+)
+def test_floating_array_numpy_sum(values, expected):
+    arr = pd.array(values, dtype="Float64")
+    result = np.sum(arr)
+    assert result == expected
+
+
+@pytest.mark.parametrize("op", ["sum", "min", "max", "prod"])
+def test_preserve_dtypes(op):
+    df = pd.DataFrame(
+        {
+            "A": ["a", "b", "b"],
+            "B": [1, None, 3],
+            "C": pd.array([0.1, None, 3.0], dtype="Float64"),
+        }
+    )
+
+    # op
+    result = getattr(df.C, op)()
+    assert isinstance(result, np.float64)
+
+    # groupby
+    result = getattr(df.groupby("A"), op)()
+
+    expected = pd.DataFrame(
+        {"B": np.array([1.0, 3.0]), "C": pd.array([0.1, 3], dtype="Float64")},
+        index=pd.Index(["a", "b"], name="A"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["min", "max"])
+def test_floating_array_min_max(skipna, method, dtype):
+    arr = pd.array([0.0, 1.0, None], dtype=dtype)
+    func = getattr(arr, method)
+    result = func(skipna=skipna)
+    if skipna:
+        assert result == (0 if method == "min" else 1)
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize("min_count", [0, 9])
+def test_floating_array_prod(skipna, min_count, dtype):
+    arr = pd.array([1.0, 2.0, None], dtype=dtype)
+    result = arr.prod(skipna=skipna, min_count=min_count)
+    if skipna and min_count == 0:
+        assert result == 2
+    else:
+        assert result is pd.NA
diff --git a/pandas/pandas/tests/arrays/floating/test_repr.py b/pandas/pandas/tests/arrays/floating/test_repr.py
new file mode 100644
index 0000000000000..ea2cdd4fab86a
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_repr.py
@@ -0,0 +1,47 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas.core.arrays.floating import (
+    Float32Dtype,
+    Float64Dtype,
+)
+
+
+def test_dtypes(dtype):
+    # smoke tests on auto dtype construction
+
+    np.dtype(dtype.type).kind == "f"
+    assert dtype.name is not None
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [(Float32Dtype(), "Float32Dtype()"), (Float64Dtype(), "Float64Dtype()")],
+)
+def test_repr_dtype(dtype, expected):
+    assert repr(dtype) == expected
+
+
+def test_repr_array():
+    result = repr(pd.array([1.0, None, 3.0]))
+    expected = "<FloatingArray>\n[1.0, <NA>, 3.0]\nLength: 3, dtype: Float64"
+    assert result == expected
+
+
+def test_repr_array_long():
+    data = pd.array([1.0, 2.0, None] * 1000)
+    expected = """<FloatingArray>
+[ 1.0,  2.0, <NA>,  1.0,  2.0, <NA>,  1.0,  2.0, <NA>,  1.0,
+ ...
+ <NA>,  1.0,  2.0, <NA>,  1.0,  2.0, <NA>,  1.0,  2.0, <NA>]
+Length: 3000, dtype: Float64"""
+    result = repr(data)
+    assert result == expected
+
+
+def test_frame_repr(data_missing):
+    df = pd.DataFrame({"A": data_missing})
+    result = repr(df)
+    expected = "      A\n0  <NA>\n1   0.1"
+    assert result == expected
diff --git a/pandas/pandas/tests/arrays/floating/test_to_numpy.py b/pandas/pandas/tests/arrays/floating/test_to_numpy.py
new file mode 100644
index 0000000000000..e954cecba417a
--- /dev/null
+++ b/pandas/pandas/tests/arrays/floating/test_to_numpy.py
@@ -0,0 +1,132 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import FloatingArray
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy(box):
+    con = pd.Series if box else pd.array
+
+    # default (with or without missing values) -> object dtype
+    arr = con([0.1, 0.2, 0.3], dtype="Float64")
+    result = arr.to_numpy()
+    expected = np.array([0.1, 0.2, 0.3], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([0.1, 0.2, None], dtype="Float64")
+    result = arr.to_numpy()
+    expected = np.array([0.1, 0.2, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_float(box):
+    con = pd.Series if box else pd.array
+
+    # no missing values -> can convert to float, otherwise raises
+    arr = con([0.1, 0.2, 0.3], dtype="Float64")
+    result = arr.to_numpy(dtype="float64")
+    expected = np.array([0.1, 0.2, 0.3], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([0.1, 0.2, None], dtype="Float64")
+    result = arr.to_numpy(dtype="float64")
+    expected = np.array([0.1, 0.2, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype="float64", na_value=np.nan)
+    expected = np.array([0.1, 0.2, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_int(box):
+    con = pd.Series if box else pd.array
+
+    # no missing values -> can convert to int, otherwise raises
+    arr = con([1.0, 2.0, 3.0], dtype="Float64")
+    result = arr.to_numpy(dtype="int64")
+    expected = np.array([1, 2, 3], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = con([1.0, 2.0, None], dtype="Float64")
+    with pytest.raises(ValueError, match="cannot convert to 'int64'-dtype"):
+        result = arr.to_numpy(dtype="int64")
+
+    # automatic casting (floors the values)
+    arr = con([0.1, 0.9, 1.1], dtype="Float64")
+    result = arr.to_numpy(dtype="int64")
+    expected = np.array([0, 0, 1], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_na_value(box):
+    con = pd.Series if box else pd.array
+
+    arr = con([0.0, 1.0, None], dtype="Float64")
+    result = arr.to_numpy(dtype=object, na_value=None)
+    expected = np.array([0.0, 1.0, None], dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype=bool, na_value=False)
+    expected = np.array([False, True, False], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.to_numpy(dtype="int64", na_value=-99)
+    expected = np.array([0, 1, -99], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_na_value_with_nan():
+    # array with both NaN and NA -> only fill NA with `na_value`
+    arr = FloatingArray(np.array([0.0, np.nan, 0.0]), np.array([False, False, True]))
+    result = arr.to_numpy(dtype="float64", na_value=-1)
+    expected = np.array([0.0, np.nan, -1.0], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["float64", "float32", "int32", "int64", "bool"])
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_dtype(box, dtype):
+    con = pd.Series if box else pd.array
+    arr = con([0.0, 1.0], dtype="Float64")
+
+    result = arr.to_numpy(dtype=dtype)
+    expected = np.array([0, 1], dtype=dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int32", "int64", "bool"])
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_na_raises(box, dtype):
+    con = pd.Series if box else pd.array
+    arr = con([0.0, 1.0, None], dtype="Float64")
+    with pytest.raises(ValueError, match=dtype):
+        arr.to_numpy(dtype=dtype)
+
+
+@pytest.mark.parametrize("box", [True, False], ids=["series", "array"])
+def test_to_numpy_string(box, dtype):
+    con = pd.Series if box else pd.array
+    arr = con([0.0, 1.0, None], dtype="Float64")
+
+    result = arr.to_numpy(dtype="str")
+    expected = np.array([0.0, 1.0, pd.NA], dtype=f"{tm.ENDIAN}U32")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_copy():
+    # to_numpy can be zero-copy if no missing values
+    arr = pd.array([0.1, 0.2, 0.3], dtype="Float64")
+    result = arr.to_numpy(dtype="float64")
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, pd.array([10, 0.2, 0.3], dtype="Float64"))
+
+    arr = pd.array([0.1, 0.2, 0.3], dtype="Float64")
+    result = arr.to_numpy(dtype="float64", copy=True)
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, pd.array([0.1, 0.2, 0.3], dtype="Float64"))
diff --git a/pandas/pandas/tests/arrays/integer/__init__.py b/pandas/pandas/tests/arrays/integer/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/integer/conftest.py b/pandas/pandas/tests/arrays/integer/conftest.py
new file mode 100644
index 0000000000000..f73400dfe689e
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/conftest.py
@@ -0,0 +1,68 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    Int16Dtype,
+    Int32Dtype,
+    Int64Dtype,
+    UInt8Dtype,
+    UInt16Dtype,
+    UInt32Dtype,
+    UInt64Dtype,
+)
+
+
+@pytest.fixture(
+    params=[
+        Int8Dtype,
+        Int16Dtype,
+        Int32Dtype,
+        Int64Dtype,
+        UInt8Dtype,
+        UInt16Dtype,
+        UInt32Dtype,
+        UInt64Dtype,
+    ]
+)
+def dtype(request):
+    """Parametrized fixture returning integer 'dtype'"""
+    return request.param()
+
+
+@pytest.fixture
+def data(dtype):
+    """
+    Fixture returning 'data' array with valid and missing values according to
+    parametrized integer 'dtype'.
+
+    Used to test dtype conversion with and without missing values.
+    """
+    return pd.array(
+        list(range(8)) + [np.nan] + list(range(10, 98)) + [np.nan] + [99, 100],
+        dtype=dtype,
+    )
+
+
+@pytest.fixture
+def data_missing(dtype):
+    """
+    Fixture returning array with exactly one NaN and one valid integer,
+    according to parametrized integer 'dtype'.
+
+    Used to test dtype conversion with and without missing values.
+    """
+    return pd.array([np.nan, 1], dtype=dtype)
+
+
+@pytest.fixture(params=["data", "data_missing"])
+def all_data(request, data, data_missing):
+    """Parametrized fixture returning 'data' or 'data_missing' integer arrays.
+
+    Used to test dtype conversion with and without missing values.
+    """
+    if request.param == "data":
+        return data
+    elif request.param == "data_missing":
+        return data_missing
diff --git a/pandas/pandas/tests/arrays/integer/test_arithmetic.py b/pandas/pandas/tests/arrays/integer/test_arithmetic.py
new file mode 100644
index 0000000000000..dee3deeee0f2f
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_arithmetic.py
@@ -0,0 +1,376 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core import ops
+from pandas.core.arrays import FloatingArray
+
+# Basic test for the arithmetic array ops
+# -----------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize(
+    "opname, exp",
+    [("add", [1, 3, None, None, 9]), ("mul", [0, 2, None, None, 20])],
+    ids=["add", "mul"],
+)
+def test_add_mul(dtype, opname, exp):
+    a = pd.array([0, 1, None, 3, 4], dtype=dtype)
+    b = pd.array([1, 2, 3, None, 5], dtype=dtype)
+
+    # array / array
+    expected = pd.array(exp, dtype=dtype)
+
+    op = getattr(operator, opname)
+    result = op(a, b)
+    tm.assert_extension_array_equal(result, expected)
+
+    op = getattr(ops, "r" + opname)
+    result = op(a, b)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_sub(dtype):
+    a = pd.array([1, 2, 3, None, 5], dtype=dtype)
+    b = pd.array([0, 1, None, 3, 4], dtype=dtype)
+
+    result = a - b
+    expected = pd.array([1, 1, None, None, 1], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_div(dtype):
+    a = pd.array([1, 2, 3, None, 5], dtype=dtype)
+    b = pd.array([0, 1, None, 3, 4], dtype=dtype)
+
+    result = a / b
+    expected = pd.array([np.inf, 2, None, None, 1.25], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("zero, negative", [(0, False), (0.0, False), (-0.0, True)])
+def test_divide_by_zero(zero, negative):
+    # https://github.com/pandas-dev/pandas/issues/27398, GH#22793
+    a = pd.array([0, 1, -1, None], dtype="Int64")
+    result = a / zero
+    expected = FloatingArray(
+        np.array([np.nan, np.inf, -np.inf, 1], dtype="float64"),
+        np.array([False, False, False, True]),
+    )
+    if negative:
+        expected *= -1
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_floordiv(dtype):
+    a = pd.array([1, 2, 3, None, 5], dtype=dtype)
+    b = pd.array([0, 1, None, 3, 4], dtype=dtype)
+
+    result = a // b
+    # Series op sets 1//0 to np.inf, which IntegerArray does not do (yet)
+    expected = pd.array([0, 2, None, None, 1], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_floordiv_by_int_zero_no_mask(any_int_ea_dtype):
+    # GH 48223: Aligns with non-masked floordiv
+    # but differs from numpy
+    # https://github.com/pandas-dev/pandas/issues/30188#issuecomment-564452740
+    ser = pd.Series([0, 1], dtype=any_int_ea_dtype)
+    result = 1 // ser
+    expected = pd.Series([np.inf, 1.0], dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+    ser_non_nullable = ser.astype(ser.dtype.numpy_dtype)
+    result = 1 // ser_non_nullable
+    expected = expected.astype(np.float64)
+    tm.assert_series_equal(result, expected)
+
+
+def test_mod(dtype):
+    a = pd.array([1, 2, 3, None, 5], dtype=dtype)
+    b = pd.array([0, 1, None, 3, 4], dtype=dtype)
+
+    result = a % b
+    expected = pd.array([0, 0, None, None, 1], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_pow_scalar():
+    a = pd.array([-1, 0, 1, None, 2], dtype="Int64")
+    result = a**0
+    expected = pd.array([1, 1, 1, 1, 1], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**1
+    expected = pd.array([-1, 0, 1, None, 2], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**pd.NA
+    expected = pd.array([None, None, 1, None, None], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = a**np.nan
+    expected = FloatingArray(
+        np.array([np.nan, np.nan, 1, np.nan, np.nan], dtype="float64"),
+        np.array([False, False, False, True, False]),
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    # reversed
+    a = a[1:]  # Can't raise integers to negative powers.
+
+    result = 0**a
+    expected = pd.array([1, 0, None, 0], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = 1**a
+    expected = pd.array([1, 1, 1, 1], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = pd.NA**a
+    expected = pd.array([1, None, None, None], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = np.nan**a
+    expected = FloatingArray(
+        np.array([1, np.nan, np.nan, np.nan], dtype="float64"),
+        np.array([False, False, True, False]),
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_pow_array():
+    a = pd.array([0, 0, 0, 1, 1, 1, None, None, None])
+    b = pd.array([0, 1, None, 0, 1, None, 0, 1, None])
+    result = a**b
+    expected = pd.array([1, 0, None, 1, 1, 1, 1, None, None])
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_rpow_one_to_na():
+    # https://github.com/pandas-dev/pandas/issues/22022
+    # https://github.com/pandas-dev/pandas/issues/29997
+    arr = pd.array([np.nan, np.nan], dtype="Int64")
+    result = np.array([1.0, 2.0]) ** arr
+    expected = pd.array([1.0, np.nan], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("other", [0, 0.5])
+def test_numpy_zero_dim_ndarray(other):
+    arr = pd.array([1, None, 2])
+    result = arr + np.array(other)
+    expected = arr + other
+    tm.assert_equal(result, expected)
+
+
+# Test generic characteristics / errors
+# -----------------------------------------------------------------------------
+
+
+def test_error_invalid_values(data, all_arithmetic_operators):
+    op = all_arithmetic_operators
+    s = pd.Series(data)
+    ops = getattr(s, op)
+
+    # invalid scalars
+    msg = "|".join(
+        [
+            r"can only perform ops with numeric values",
+            r"IntegerArray cannot perform the operation mod",
+            r"unsupported operand type",
+            r"can only concatenate str \(not \"int\"\) to str",
+            "not all arguments converted during string",
+            "ufunc '.*' not supported for the input types, and the inputs could not",
+            "ufunc '.*' did not contain a loop with signature matching types",
+            "Addition/subtraction of integers and integer-arrays with Timestamp",
+            "has no kernel",
+            "not implemented",
+            "The 'out' kwarg is necessary. Use numpy.strings.multiply without it.",
+            "not supported for dtype",
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops("foo")
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Timestamp("20180101"))
+
+    # invalid array-likes
+    str_ser = pd.Series("foo", index=s.index)
+    # with pytest.raises(TypeError, match=msg):
+    if all_arithmetic_operators in [
+        "__mul__",
+        "__rmul__",
+    ]:  # (data[~data.isna()] >= 0).all():
+        res = ops(str_ser)
+        expected = pd.Series(["foo" * x for x in data], index=s.index)
+        expected = expected.fillna(np.nan)
+        # TODO: doing this fillna to keep tests passing as we make
+        #  assert_almost_equal stricter, but the expected with pd.NA seems
+        #  more-correct than np.nan here.
+        tm.assert_series_equal(res, expected)
+    else:
+        with pytest.raises(TypeError, match=msg):
+            ops(str_ser)
+
+    msg = "|".join(
+        [
+            "can only perform ops with numeric values",
+            "cannot perform .* with this index type: DatetimeArray",
+            "Addition/subtraction of integers and integer-arrays "
+            "with DatetimeArray is no longer supported. *",
+            "unsupported operand type",
+            r"can only concatenate str \(not \"int\"\) to str",
+            "not all arguments converted during string",
+            "cannot subtract DatetimeArray from ndarray",
+            "has no kernel",
+            "not implemented",
+            "not supported for dtype",
+        ]
+    )
+    with pytest.raises(TypeError, match=msg):
+        ops(pd.Series(pd.date_range("20180101", periods=len(s))))
+
+
+# Various
+# -----------------------------------------------------------------------------
+
+
+# TODO test unsigned overflow
+
+
+def test_arith_coerce_scalar(data, all_arithmetic_operators):
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    s = pd.Series(data)
+    other = 0.01
+
+    result = op(s, other)
+    expected = op(s.astype(float), other)
+    expected = expected.astype("Float64")
+
+    # rmod results in NaN that wasn't NA in original nullable Series -> unmask it
+    if all_arithmetic_operators == "__rmod__":
+        mask = (s == 0).fillna(False).to_numpy(bool)
+        expected.array._mask[mask] = False
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("other", [1.0, np.array(1.0)])
+def test_arithmetic_conversion(all_arithmetic_operators, other):
+    # if we have a float operand we should have a float result
+    # if that is equal to an integer
+    op = tm.get_op_from_name(all_arithmetic_operators)
+
+    s = pd.Series([1, 2, 3], dtype="Int64")
+    result = op(s, other)
+    assert result.dtype == "Float64"
+
+
+def test_cross_type_arithmetic():
+    df = pd.DataFrame(
+        {
+            "A": pd.Series([1, 2, np.nan], dtype="Int64"),
+            "B": pd.Series([1, np.nan, 3], dtype="UInt8"),
+            "C": [1, 2, 3],
+        }
+    )
+
+    result = df.A + df.C
+    expected = pd.Series([2, 4, np.nan], dtype="Int64")
+    tm.assert_series_equal(result, expected)
+
+    result = (df.A + df.C) * 3 == 12
+    expected = pd.Series([False, True, None], dtype="boolean")
+    tm.assert_series_equal(result, expected)
+
+    result = df.A + df.B
+    expected = pd.Series([2, np.nan, np.nan], dtype="Int64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", ["mean"])
+def test_reduce_to_float(op):
+    # some reduce ops always return float, even if the result
+    # is a rounded number
+    df = pd.DataFrame(
+        {
+            "A": ["a", "b", "b"],
+            "B": [1, None, 3],
+            "C": pd.array([1, None, 3], dtype="Int64"),
+        }
+    )
+
+    # op
+    result = getattr(df.C, op)()
+    assert isinstance(result, float)
+
+    # groupby
+    result = getattr(df.groupby("A"), op)()
+
+    expected = pd.DataFrame(
+        {"B": np.array([1.0, 3.0]), "C": pd.array([1, 3], dtype="Float64")},
+        index=pd.Index(["a", "b"], name="A"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "source, neg_target, abs_target",
+    [
+        ([1, 2, 3], [-1, -2, -3], [1, 2, 3]),
+        ([1, 2, None], [-1, -2, None], [1, 2, None]),
+        ([-1, 0, 1], [1, 0, -1], [1, 0, 1]),
+    ],
+)
+def test_unary_int_operators(any_signed_int_ea_dtype, source, neg_target, abs_target):
+    dtype = any_signed_int_ea_dtype
+    arr = pd.array(source, dtype=dtype)
+    neg_result, pos_result, abs_result = -arr, +arr, abs(arr)
+    neg_target = pd.array(neg_target, dtype=dtype)
+    abs_target = pd.array(abs_target, dtype=dtype)
+
+    tm.assert_extension_array_equal(neg_result, neg_target)
+    tm.assert_extension_array_equal(pos_result, arr)
+    assert not tm.shares_memory(pos_result, arr)
+    tm.assert_extension_array_equal(abs_result, abs_target)
+
+
+def test_values_multiplying_large_series_by_NA():
+    # GH#33701
+
+    result = pd.NA * pd.Series(np.zeros(10001))
+    expected = pd.Series([pd.NA] * 10001)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_bitwise(dtype):
+    left = pd.array([1, None, 3, 4], dtype=dtype)
+    right = pd.array([None, 3, 5, 4], dtype=dtype)
+
+    result = left | right
+    expected = pd.array([None, None, 3 | 5, 4 | 4], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = left & right
+    expected = pd.array([None, None, 3 & 5, 4 & 4], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = left ^ right
+    expected = pd.array([None, None, 3 ^ 5, 4 ^ 4], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    # TODO: desired behavior when operating with boolean?  defer?
+
+    floats = right.astype("Float64")
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left | floats
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left & floats
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        left ^ floats
diff --git a/pandas/pandas/tests/arrays/integer/test_comparison.py b/pandas/pandas/tests/arrays/integer/test_comparison.py
new file mode 100644
index 0000000000000..568b0b087bf1d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_comparison.py
@@ -0,0 +1,39 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.arrays.masked_shared import (
+    ComparisonOps,
+    NumericOps,
+)
+
+
+class TestComparisonOps(NumericOps, ComparisonOps):
+    @pytest.mark.parametrize("other", [True, False, pd.NA, -1, 0, 1])
+    def test_scalar(self, other, comparison_op, dtype):
+        ComparisonOps.test_scalar(self, other, comparison_op, dtype)
+
+    def test_compare_to_int(self, dtype, comparison_op):
+        # GH 28930
+        op_name = f"__{comparison_op.__name__}__"
+        s1 = pd.Series([1, None, 3], dtype=dtype)
+        s2 = pd.Series([1, None, 3], dtype="float")
+
+        method = getattr(s1, op_name)
+        result = method(2)
+
+        method = getattr(s2, op_name)
+        expected = method(2).astype("boolean")
+        expected[s2.isna()] = pd.NA
+
+        tm.assert_series_equal(result, expected)
+
+
+def test_equals():
+    # GH-30652
+    # equals is generally tested in /tests/extension/base/methods, but this
+    # specifically tests that two arrays of the same class but different dtype
+    # do not evaluate equal
+    a1 = pd.array([1, 2, None], dtype="Int64")
+    a2 = pd.array([1, 2, None], dtype="Int32")
+    assert a1.equals(a2) is False
diff --git a/pandas/pandas/tests/arrays/integer/test_concat.py b/pandas/pandas/tests/arrays/integer/test_concat.py
new file mode 100644
index 0000000000000..feba574da548f
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_concat.py
@@ -0,0 +1,69 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "to_concat_dtypes, result_dtype",
+    [
+        (["Int64", "Int64"], "Int64"),
+        (["UInt64", "UInt64"], "UInt64"),
+        (["Int8", "Int8"], "Int8"),
+        (["Int8", "Int16"], "Int16"),
+        (["UInt8", "Int8"], "Int16"),
+        (["Int32", "UInt32"], "Int64"),
+        (["Int64", "UInt64"], "Float64"),
+        (["Int64", "boolean"], "object"),
+        (["UInt8", "boolean"], "object"),
+    ],
+)
+def test_concat_series(to_concat_dtypes, result_dtype):
+    # we expect the same dtypes as we would get with non-masked inputs,
+    #  just masked where available.
+
+    result = pd.concat([pd.Series([0, 1, pd.NA], dtype=t) for t in to_concat_dtypes])
+    expected = pd.concat([pd.Series([0, 1, pd.NA], dtype=object)] * 2).astype(
+        result_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    # order doesn't matter for result
+    result = pd.concat(
+        [pd.Series([0, 1, pd.NA], dtype=t) for t in to_concat_dtypes[::-1]]
+    )
+    expected = pd.concat([pd.Series([0, 1, pd.NA], dtype=object)] * 2).astype(
+        result_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "to_concat_dtypes, result_dtype",
+    [
+        (["Int64", "int64"], "Int64"),
+        (["UInt64", "uint64"], "UInt64"),
+        (["Int8", "int8"], "Int8"),
+        (["Int8", "int16"], "Int16"),
+        (["UInt8", "int8"], "Int16"),
+        (["Int32", "uint32"], "Int64"),
+        (["Int64", "uint64"], "Float64"),
+        (["Int64", "bool"], "object"),
+        (["UInt8", "bool"], "object"),
+    ],
+)
+def test_concat_series_with_numpy(to_concat_dtypes, result_dtype):
+    # we expect the same dtypes as we would get with non-masked inputs,
+    #  just masked where available.
+
+    s1 = pd.Series([0, 1, pd.NA], dtype=to_concat_dtypes[0])
+    s2 = pd.Series(np.array([0, 1], dtype=to_concat_dtypes[1]))
+    result = pd.concat([s1, s2], ignore_index=True)
+    expected = pd.Series([0, 1, pd.NA, 0, 1], dtype=object).astype(result_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # order doesn't matter for result
+    result = pd.concat([s2, s1], ignore_index=True)
+    expected = pd.Series([0, 1, 0, 1, pd.NA], dtype=object).astype(result_dtype)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/integer/test_construction.py b/pandas/pandas/tests/arrays/integer/test_construction.py
new file mode 100644
index 0000000000000..64fe40e53a9d2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_construction.py
@@ -0,0 +1,245 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import is_integer
+from pandas.core.arrays import IntegerArray
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    Int32Dtype,
+    Int64Dtype,
+)
+
+
+@pytest.fixture(params=[pd.array, IntegerArray._from_sequence])
+def constructor(request):
+    """Fixture returning parametrized IntegerArray from given sequence.
+
+    Used to test dtype conversions.
+    """
+    return request.param
+
+
+def test_uses_pandas_na():
+    a = pd.array([1, None], dtype=Int64Dtype())
+    assert a[1] is pd.NA
+
+
+def test_from_dtype_from_float(data):
+    # construct from our dtype & string dtype
+    dtype = data.dtype
+
+    # from float
+    expected = pd.Series(data)
+    result = pd.Series(data.to_numpy(na_value=np.nan, dtype="float"), dtype=str(dtype))
+    tm.assert_series_equal(result, expected)
+
+    # from int / list
+    expected = pd.Series(data)
+    result = pd.Series(np.array(data).tolist(), dtype=str(dtype))
+    tm.assert_series_equal(result, expected)
+
+    # from int / array
+    expected = pd.Series(data).dropna().reset_index(drop=True)
+    dropped = np.array(data.dropna()).astype(np.dtype(dtype.type))
+    result = pd.Series(dropped, dtype=str(dtype))
+    tm.assert_series_equal(result, expected)
+
+
+def test_conversions(data_missing):
+    # astype to object series
+    df = pd.DataFrame({"A": data_missing})
+    result = df["A"].astype("object")
+    expected = pd.Series(np.array([pd.NA, 1], dtype=object), name="A")
+    tm.assert_series_equal(result, expected)
+
+    # convert to object ndarray
+    # we assert that we are exactly equal
+    # including type conversions of scalars
+    result = df["A"].astype("object").values
+    expected = np.array([pd.NA, 1], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+    for r, e in zip(result, expected):
+        if pd.isnull(r):
+            assert pd.isnull(e)
+        elif is_integer(r):
+            assert r == e
+            assert is_integer(e)
+        else:
+            assert r == e
+            assert type(r) == type(e)
+
+
+def test_integer_array_constructor():
+    values = np.array([1, 2, 3, 4], dtype="int64")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = IntegerArray(values, mask)
+    expected = pd.array([1, 2, 3, np.nan], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    msg = r".* should be .* numpy array. Use the 'pd.array' function instead"
+    with pytest.raises(TypeError, match=msg):
+        IntegerArray(values.tolist(), mask)
+
+    with pytest.raises(TypeError, match=msg):
+        IntegerArray(values, mask.tolist())
+
+    with pytest.raises(TypeError, match=msg):
+        IntegerArray(values.astype(float), mask)
+    msg = r"__init__\(\) missing 1 required positional argument: 'mask'"
+    with pytest.raises(TypeError, match=msg):
+        IntegerArray(values)
+
+
+def test_integer_array_constructor_copy():
+    values = np.array([1, 2, 3, 4], dtype="int64")
+    mask = np.array([False, False, False, True], dtype="bool")
+
+    result = IntegerArray(values, mask)
+    assert result._data is values
+    assert result._mask is mask
+
+    result = IntegerArray(values, mask, copy=True)
+    assert result._data is not values
+    assert result._mask is not mask
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    [
+        ([1, None], [1, np.nan]),
+        ([None], [np.nan]),
+        ([None, np.nan], [np.nan, np.nan]),
+        ([np.nan, np.nan], [np.nan, np.nan]),
+    ],
+)
+def test_to_integer_array_none_is_nan(a, b):
+    result = pd.array(a, dtype="Int64")
+    expected = pd.array(b, dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        ["foo", "bar"],
+        "foo",
+        1,
+        1.0,
+        pd.date_range("20130101", periods=2),
+        np.array(["foo"]),
+        [[1, 2], [3, 4]],
+        [np.nan, {"a": 1}],
+    ],
+)
+def test_to_integer_array_error(values):
+    # error in converting existing arrays to IntegerArrays
+    msg = "|".join(
+        [
+            r"cannot be converted to IntegerDtype",
+            r"invalid literal for int\(\) with base 10:",
+            r"values must be a 1D list-like",
+            r"Cannot pass scalar",
+            r"int\(\) argument must be a string",
+        ]
+    )
+    with pytest.raises((ValueError, TypeError), match=msg):
+        pd.array(values, dtype="Int64")
+
+    with pytest.raises((ValueError, TypeError), match=msg):
+        IntegerArray._from_sequence(values)
+
+
+def test_to_integer_array_inferred_dtype(constructor):
+    # if values has dtype -> respect it
+    result = constructor(np.array([1, 2], dtype="int8"))
+    assert result.dtype == Int8Dtype()
+    result = constructor(np.array([1, 2], dtype="int32"))
+    assert result.dtype == Int32Dtype()
+
+    # if values have no dtype -> always int64
+    result = constructor([1, 2])
+    assert result.dtype == Int64Dtype()
+
+
+def test_to_integer_array_dtype_keyword(constructor):
+    result = constructor([1, 2], dtype="Int8")
+    assert result.dtype == Int8Dtype()
+
+    # if values has dtype -> override it
+    result = constructor(np.array([1, 2], dtype="int8"), dtype="Int32")
+    assert result.dtype == Int32Dtype()
+
+
+def test_to_integer_array_float():
+    result = IntegerArray._from_sequence([1.0, 2.0], dtype="Int64")
+    expected = pd.array([1, 2], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    with pytest.raises(TypeError, match="cannot safely cast non-equivalent"):
+        IntegerArray._from_sequence([1.5, 2.0], dtype="Int64")
+
+    # for float dtypes, the itemsize is not preserved
+    result = IntegerArray._from_sequence(
+        np.array([1.0, 2.0], dtype="float32"), dtype="Int64"
+    )
+    assert result.dtype == Int64Dtype()
+
+
+def test_to_integer_array_str():
+    result = IntegerArray._from_sequence(["1", "2", None], dtype="Int64")
+    expected = pd.array([1, 2, np.nan], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    with pytest.raises(
+        ValueError, match=r"invalid literal for int\(\) with base 10: .*"
+    ):
+        IntegerArray._from_sequence(["1", "2", ""], dtype="Int64")
+
+    with pytest.raises(
+        ValueError, match=r"invalid literal for int\(\) with base 10: .*"
+    ):
+        IntegerArray._from_sequence(["1.5", "2.0"], dtype="Int64")
+
+
+@pytest.mark.parametrize(
+    "bool_values, int_values, target_dtype, expected_dtype",
+    [
+        ([False, True], [0, 1], Int64Dtype(), Int64Dtype()),
+        ([False, True], [0, 1], "Int64", Int64Dtype()),
+        ([False, True, np.nan], [0, 1, np.nan], Int64Dtype(), Int64Dtype()),
+    ],
+)
+def test_to_integer_array_bool(
+    constructor, bool_values, int_values, target_dtype, expected_dtype
+):
+    result = constructor(bool_values, dtype=target_dtype)
+    assert result.dtype == expected_dtype
+    expected = pd.array(int_values, dtype=target_dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values, to_dtype, result_dtype",
+    [
+        (np.array([1], dtype="int64"), None, Int64Dtype),
+        (np.array([1, np.nan]), None, Int64Dtype),
+        (np.array([1, np.nan]), "int8", Int8Dtype),
+    ],
+)
+def test_to_integer_array(values, to_dtype, result_dtype):
+    # convert existing arrays to IntegerArrays
+    result = IntegerArray._from_sequence(values, dtype=to_dtype)
+    assert result.dtype == result_dtype()
+    expected = pd.array(values, dtype=result_dtype())
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_integer_array_from_boolean():
+    # GH31104
+    expected = pd.array(np.array([True, False]), dtype="Int64")
+    result = pd.array(np.array([True, False], dtype=object), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/integer/test_dtypes.py b/pandas/pandas/tests/arrays/integer/test_dtypes.py
new file mode 100644
index 0000000000000..7972ba7b9fb0f
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_dtypes.py
@@ -0,0 +1,299 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.generic import ABCIndex
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    UInt32Dtype,
+)
+
+
+def test_dtypes(dtype):
+    # smoke tests on auto dtype construction
+
+    if dtype.is_signed_integer:
+        assert np.dtype(dtype.type).kind == "i"
+    else:
+        assert np.dtype(dtype.type).kind == "u"
+    assert dtype.name is not None
+
+
+@pytest.mark.parametrize("op", ["sum", "min", "max", "prod"])
+def test_preserve_dtypes(op):
+    # for ops that enable (mean would actually work here
+    # but generally it is a float return value)
+    df = pd.DataFrame(
+        {
+            "A": ["a", "b", "b"],
+            "B": [1, None, 3],
+            "C": pd.array([1, None, 3], dtype="Int64"),
+        }
+    )
+
+    # op
+    result = getattr(df.C, op)()
+    if op in {"sum", "prod", "min", "max"}:
+        assert isinstance(result, np.int64)
+    else:
+        assert isinstance(result, int)
+
+    # groupby
+    result = getattr(df.groupby("A"), op)()
+
+    expected = pd.DataFrame(
+        {"B": np.array([1.0, 3.0]), "C": pd.array([1, 3], dtype="Int64")},
+        index=pd.Index(["a", "b"], name="A"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_astype_nansafe():
+    # see gh-22343
+    arr = pd.array([np.nan, 1, 2], dtype="Int8")
+    msg = "cannot convert NA to integer"
+
+    with pytest.raises(ValueError, match=msg):
+        arr.astype("uint32")
+
+
+def test_construct_index(all_data, dropna):
+    # ensure that we do not coerce to different Index dtype or non-index
+
+    all_data = all_data[:10]
+    if dropna:
+        other = np.array(all_data[~all_data.isna()])
+    else:
+        other = all_data
+
+    result = pd.Index(pd.array(other, dtype=all_data.dtype))
+    expected = pd.Index(other, dtype=all_data.dtype)
+    assert all_data.dtype == expected.dtype  # dont coerce to object
+
+    tm.assert_index_equal(result, expected)
+
+
+def test_astype_index(all_data, dropna):
+    # as an int/uint index to Index
+
+    all_data = all_data[:10]
+    if dropna:
+        other = all_data[~all_data.isna()]
+    else:
+        other = all_data
+
+    dtype = all_data.dtype
+    idx = pd.Index(np.array(other))
+    assert isinstance(idx, ABCIndex)
+
+    result = idx.astype(dtype)
+    expected = idx.astype(object).astype(dtype)
+    tm.assert_index_equal(result, expected)
+
+
+def test_astype(all_data):
+    all_data = all_data[:10]
+
+    ints = all_data[~all_data.isna()]
+    mixed = all_data
+    dtype = Int8Dtype()
+
+    # coerce to same type - ints
+    s = pd.Series(ints)
+    result = s.astype(all_data.dtype)
+    expected = pd.Series(ints)
+    tm.assert_series_equal(result, expected)
+
+    # coerce to same other - ints
+    s = pd.Series(ints)
+    result = s.astype(dtype)
+    expected = pd.Series(ints, dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    # coerce to same numpy_dtype - ints
+    s = pd.Series(ints)
+    result = s.astype(all_data.dtype.numpy_dtype)
+    expected = pd.Series(ints._data.astype(all_data.dtype.numpy_dtype))
+    tm.assert_series_equal(result, expected)
+
+    # coerce to same type - mixed
+    s = pd.Series(mixed)
+    result = s.astype(all_data.dtype)
+    expected = pd.Series(mixed)
+    tm.assert_series_equal(result, expected)
+
+    # coerce to same other - mixed
+    s = pd.Series(mixed)
+    result = s.astype(dtype)
+    expected = pd.Series(mixed, dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    # coerce to same numpy_dtype - mixed
+    s = pd.Series(mixed)
+    msg = "cannot convert NA to integer"
+    with pytest.raises(ValueError, match=msg):
+        s.astype(all_data.dtype.numpy_dtype)
+
+    # coerce to object
+    s = pd.Series(mixed)
+    result = s.astype("object")
+    expected = pd.Series(np.asarray(mixed, dtype=object))
+    tm.assert_series_equal(result, expected)
+
+
+def test_astype_copy():
+    arr = pd.array([1, 2, 3, None], dtype="Int64")
+    orig = pd.array([1, 2, 3, None], dtype="Int64")
+
+    # copy=True -> ensure both data and mask are actual copies
+    result = arr.astype("Int64", copy=True)
+    assert result is not arr
+    assert not tm.shares_memory(result, arr)
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, orig)
+    result[0] = pd.NA
+    tm.assert_extension_array_equal(arr, orig)
+
+    # copy=False
+    result = arr.astype("Int64", copy=False)
+    assert result is arr
+    assert np.shares_memory(result._data, arr._data)
+    assert np.shares_memory(result._mask, arr._mask)
+    result[0] = 10
+    assert arr[0] == 10
+    result[0] = pd.NA
+    assert arr[0] is pd.NA
+
+    # astype to different dtype -> always needs a copy -> even with copy=False
+    # we need to ensure that also the mask is actually copied
+    arr = pd.array([1, 2, 3, None], dtype="Int64")
+    orig = pd.array([1, 2, 3, None], dtype="Int64")
+
+    result = arr.astype("Int32", copy=False)
+    assert not tm.shares_memory(result, arr)
+    result[0] = 10
+    tm.assert_extension_array_equal(arr, orig)
+    result[0] = pd.NA
+    tm.assert_extension_array_equal(arr, orig)
+
+
+def test_astype_to_larger_numpy():
+    a = pd.array([1, 2], dtype="Int32")
+    result = a.astype("int64")
+    expected = np.array([1, 2], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    a = pd.array([1, 2], dtype="UInt32")
+    result = a.astype("uint64")
+    expected = np.array([1, 2], dtype="uint64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [Int8Dtype(), "Int8", UInt32Dtype(), "UInt32"])
+def test_astype_specific_casting(dtype):
+    s = pd.Series([1, 2, 3], dtype="Int64")
+    result = s.astype(dtype)
+    expected = pd.Series([1, 2, 3], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    s = pd.Series([1, 2, 3, None], dtype="Int64")
+    result = s.astype(dtype)
+    expected = pd.Series([1, 2, 3, None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_astype_floating():
+    arr = pd.array([1, 2, None], dtype="Int64")
+    result = arr.astype("Float64")
+    expected = pd.array([1.0, 2.0, None], dtype="Float64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_astype_dt64():
+    # GH#32435
+    arr = pd.array([1, 2, 3, pd.NA]) * 10**9
+
+    result = arr.astype("datetime64[ns]")
+
+    expected = np.array([1, 2, 3, "NaT"], dtype="M8[s]").astype("M8[ns]")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_construct_cast_invalid(dtype):
+    msg = "cannot safely"
+    arr = [1.2, 2.3, 3.7]
+    with pytest.raises(TypeError, match=msg):
+        pd.array(arr, dtype=dtype)
+
+    with pytest.raises(TypeError, match=msg):
+        pd.Series(arr).astype(dtype)
+
+    arr = [1.2, 2.3, 3.7, np.nan]
+    with pytest.raises(TypeError, match=msg):
+        pd.array(arr, dtype=dtype)
+
+    with pytest.raises(TypeError, match=msg):
+        pd.Series(arr).astype(dtype)
+
+
+@pytest.mark.parametrize("in_series", [True, False])
+def test_to_numpy_na_nan(in_series):
+    a = pd.array([0, 1, None], dtype="Int64")
+    if in_series:
+        a = pd.Series(a)
+
+    result = a.to_numpy(dtype="float64", na_value=np.nan)
+    expected = np.array([0.0, 1.0, np.nan], dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = a.to_numpy(dtype="int64", na_value=-1)
+    expected = np.array([0, 1, -1], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = a.to_numpy(dtype="bool", na_value=False)
+    expected = np.array([False, True, False], dtype="bool")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("in_series", [True, False])
+@pytest.mark.parametrize("dtype", ["int32", "int64", "bool"])
+def test_to_numpy_dtype(dtype, in_series):
+    a = pd.array([0, 1], dtype="Int64")
+    if in_series:
+        a = pd.Series(a)
+
+    result = a.to_numpy(dtype=dtype)
+    expected = np.array([0, 1], dtype=dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "bool"])
+def test_to_numpy_na_raises(dtype):
+    a = pd.array([0, 1, None], dtype="Int64")
+    with pytest.raises(ValueError, match=dtype):
+        a.to_numpy(dtype=dtype)
+
+
+def test_astype_str(using_infer_string):
+    a = pd.array([1, 2, None], dtype="Int64")
+
+    if using_infer_string:
+        expected = pd.array(["1", "2", None], dtype=pd.StringDtype(na_value=np.nan))
+
+        tm.assert_extension_array_equal(a.astype(str), expected)
+        tm.assert_extension_array_equal(a.astype("str"), expected)
+    else:
+        expected = np.array(["1", "2", "<NA>"], dtype=f"{tm.ENDIAN}U21")
+
+        tm.assert_numpy_array_equal(a.astype(str), expected)
+        tm.assert_numpy_array_equal(a.astype("str"), expected)
+
+
+def test_astype_boolean():
+    # https://github.com/pandas-dev/pandas/issues/31102
+    a = pd.array([1, 0, -1, 2, None], dtype="Int64")
+    result = a.astype("boolean")
+    expected = pd.array([True, False, True, True, None], dtype="boolean")
+    tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/integer/test_function.py b/pandas/pandas/tests/arrays/integer/test_function.py
new file mode 100644
index 0000000000000..33300fff925f6
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_function.py
@@ -0,0 +1,200 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import FloatingArray
+
+
+@pytest.mark.parametrize("ufunc", [np.abs, np.sign])
+# np.sign emits a warning with nans, <https://github.com/numpy/numpy/issues/15127>
+@pytest.mark.filterwarnings("ignore:invalid value encountered in sign:RuntimeWarning")
+def test_ufuncs_single_int(ufunc):
+    a = pd.array([1, 2, -3, np.nan])
+    result = ufunc(a)
+    expected = pd.array(ufunc(a.astype(float)), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    s = pd.Series(a)
+    result = ufunc(s)
+    expected = pd.Series(pd.array(ufunc(a.astype(float)), dtype="Int64"))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.log, np.exp, np.sin, np.cos, np.sqrt])
+def test_ufuncs_single_float(ufunc):
+    a = pd.array([1, 2, -3, np.nan])
+    with np.errstate(invalid="ignore"):
+        result = ufunc(a)
+        expected = FloatingArray(ufunc(a.astype(float)), mask=a._mask)
+    tm.assert_extension_array_equal(result, expected)
+
+    s = pd.Series(a)
+    with np.errstate(invalid="ignore"):
+        result = ufunc(s)
+    expected = pd.Series(expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.add, np.subtract])
+def test_ufuncs_binary_int(ufunc):
+    # two IntegerArrays
+    a = pd.array([1, 2, -3, np.nan])
+    result = ufunc(a, a)
+    expected = pd.array(ufunc(a.astype(float), a.astype(float)), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    # IntegerArray with numpy array
+    arr = np.array([1, 2, 3, 4])
+    result = ufunc(a, arr)
+    expected = pd.array(ufunc(a.astype(float), arr), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(arr, a)
+    expected = pd.array(ufunc(arr, a.astype(float)), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    # IntegerArray with scalar
+    result = ufunc(a, 1)
+    expected = pd.array(ufunc(a.astype(float), 1), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+    result = ufunc(1, a)
+    expected = pd.array(ufunc(1, a.astype(float)), dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_ufunc_binary_output():
+    a = pd.array([1, 2, np.nan])
+    result = np.modf(a)
+    expected = np.modf(a.to_numpy(na_value=np.nan, dtype="float"))
+    expected = (pd.array(expected[0]), pd.array(expected[1]))
+
+    assert isinstance(result, tuple)
+    assert len(result) == 2
+
+    for x, y in zip(result, expected):
+        tm.assert_extension_array_equal(x, y)
+
+
+@pytest.mark.parametrize("values", [[0, 1], [0, None]])
+def test_ufunc_reduce_raises(values):
+    arr = pd.array(values)
+
+    res = np.add.reduce(arr)
+    expected = arr.sum(skipna=False)
+    tm.assert_almost_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "pandasmethname, kwargs",
+    [
+        ("var", {"ddof": 0}),
+        ("var", {"ddof": 1}),
+        ("std", {"ddof": 0}),
+        ("std", {"ddof": 1}),
+        ("kurtosis", {}),
+        ("skew", {}),
+        ("sem", {}),
+    ],
+)
+def test_stat_method(pandasmethname, kwargs):
+    s = pd.Series(data=[1, 2, 3, 4, 5, 6, np.nan, np.nan], dtype="Int64")
+    pandasmeth = getattr(s, pandasmethname)
+    result = pandasmeth(**kwargs)
+    s2 = pd.Series(data=[1, 2, 3, 4, 5, 6], dtype="Int64")
+    pandasmeth = getattr(s2, pandasmethname)
+    expected = pandasmeth(**kwargs)
+    assert expected == result
+
+
+def test_value_counts_na():
+    arr = pd.array([1, 2, 1, pd.NA], dtype="Int64")
+    result = arr.value_counts(dropna=False)
+    ex_index = pd.Index([1, 2, pd.NA], dtype="Int64")
+    assert ex_index.dtype == "Int64"
+    expected = pd.Series([2, 1, 1], index=ex_index, dtype="Int64", name="count")
+    tm.assert_series_equal(result, expected)
+
+    result = arr.value_counts(dropna=True)
+    expected = pd.Series([2, 1], index=arr[:2], dtype="Int64", name="count")
+    assert expected.index.dtype == arr.dtype
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_empty():
+    # https://github.com/pandas-dev/pandas/issues/33317
+    ser = pd.Series([], dtype="Int64")
+    result = ser.value_counts()
+    idx = pd.Index([], dtype=ser.dtype)
+    assert idx.dtype == ser.dtype
+    expected = pd.Series([], index=idx, dtype="Int64", name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_with_normalize():
+    # GH 33172
+    ser = pd.Series([1, 2, 1, pd.NA], dtype="Int64")
+    result = ser.value_counts(normalize=True)
+    expected = pd.Series([2, 1], index=ser[:2], dtype="Float64", name="proportion") / 3
+    assert expected.index.dtype == ser.dtype
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("min_count", [0, 4])
+def test_integer_array_sum(skipna, min_count, any_int_ea_dtype):
+    dtype = any_int_ea_dtype
+    arr = pd.array([1, 2, 3, None], dtype=dtype)
+    result = arr.sum(skipna=skipna, min_count=min_count)
+    if skipna and min_count == 0:
+        assert result == 6
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize("method", ["min", "max"])
+def test_integer_array_min_max(skipna, method, any_int_ea_dtype):
+    dtype = any_int_ea_dtype
+    arr = pd.array([0, 1, None], dtype=dtype)
+    func = getattr(arr, method)
+    result = func(skipna=skipna)
+    if skipna:
+        assert result == (0 if method == "min" else 1)
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize("min_count", [0, 9])
+def test_integer_array_prod(skipna, min_count, any_int_ea_dtype):
+    dtype = any_int_ea_dtype
+    arr = pd.array([1, 2, None], dtype=dtype)
+    result = arr.prod(skipna=skipna, min_count=min_count)
+    if skipna and min_count == 0:
+        assert result == 2
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize(
+    "values, expected", [([1, 2, 3], 6), ([1, 2, 3, None], 6), ([None], 0)]
+)
+def test_integer_array_numpy_sum(values, expected):
+    arr = pd.array(values, dtype="Int64")
+    result = np.sum(arr)
+    assert result == expected
+
+
+@pytest.mark.parametrize("op", ["sum", "prod", "min", "max"])
+def test_dataframe_reductions(op):
+    # https://github.com/pandas-dev/pandas/pull/32867
+    # ensure the integers are not cast to float during reductions
+    df = pd.DataFrame({"a": pd.array([1, 2], dtype="Int64")})
+    result = df.max()
+    assert isinstance(result["a"], np.int64)
+
+
+# TODO(jreback) - these need testing / are broken
+
+# shift
+
+# set_index (destroys type)
diff --git a/pandas/pandas/tests/arrays/integer/test_indexing.py b/pandas/pandas/tests/arrays/integer/test_indexing.py
new file mode 100644
index 0000000000000..4b953d699108b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_indexing.py
@@ -0,0 +1,19 @@
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_array_setitem_nullable_boolean_mask():
+    # GH 31446
+    ser = pd.Series([1, 2], dtype="Int64")
+    result = ser.where(ser > 1)
+    expected = pd.Series([pd.NA, 2], dtype="Int64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_array_setitem():
+    # GH 31446
+    arr = pd.Series([1, 2], dtype="Int64").array
+    arr[arr > 1] = 1
+
+    expected = pd.array([1, 1], dtype="Int64")
+    tm.assert_extension_array_equal(arr, expected)
diff --git a/pandas/pandas/tests/arrays/integer/test_reduction.py b/pandas/pandas/tests/arrays/integer/test_reduction.py
new file mode 100644
index 0000000000000..1c91cd25ba69c
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_reduction.py
@@ -0,0 +1,123 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    array,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "op, expected",
+    [
+        ["sum", np.int64(3)],
+        ["prod", np.int64(2)],
+        ["min", np.int64(1)],
+        ["max", np.int64(2)],
+        ["mean", np.float64(1.5)],
+        ["median", np.float64(1.5)],
+        ["var", np.float64(0.5)],
+        ["std", np.float64(0.5**0.5)],
+        ["skew", pd.NA],
+        ["kurt", pd.NA],
+        ["any", True],
+        ["all", True],
+    ],
+)
+def test_series_reductions(op, expected):
+    ser = Series([1, 2], dtype="Int64")
+    result = getattr(ser, op)()
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op, expected",
+    [
+        ["sum", Series([3], index=["a"], dtype="Int64")],
+        ["prod", Series([2], index=["a"], dtype="Int64")],
+        ["min", Series([1], index=["a"], dtype="Int64")],
+        ["max", Series([2], index=["a"], dtype="Int64")],
+        ["mean", Series([1.5], index=["a"], dtype="Float64")],
+        ["median", Series([1.5], index=["a"], dtype="Float64")],
+        ["var", Series([0.5], index=["a"], dtype="Float64")],
+        ["std", Series([0.5**0.5], index=["a"], dtype="Float64")],
+        ["skew", Series([pd.NA], index=["a"], dtype="Float64")],
+        ["kurt", Series([pd.NA], index=["a"], dtype="Float64")],
+        ["any", Series([True], index=["a"], dtype="boolean")],
+        ["all", Series([True], index=["a"], dtype="boolean")],
+    ],
+)
+def test_dataframe_reductions(op, expected):
+    df = DataFrame({"a": array([1, 2], dtype="Int64")})
+    result = getattr(df, op)()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op, expected",
+    [
+        ["sum", array([1, 3], dtype="Int64")],
+        ["prod", array([1, 3], dtype="Int64")],
+        ["min", array([1, 3], dtype="Int64")],
+        ["max", array([1, 3], dtype="Int64")],
+        ["mean", array([1, 3], dtype="Float64")],
+        ["median", array([1, 3], dtype="Float64")],
+        ["var", array([pd.NA], dtype="Float64")],
+        ["std", array([pd.NA], dtype="Float64")],
+        ["skew", array([pd.NA], dtype="Float64")],
+        ["any", array([True, True], dtype="boolean")],
+        ["all", array([True, True], dtype="boolean")],
+    ],
+)
+def test_groupby_reductions(op, expected):
+    df = DataFrame(
+        {
+            "A": ["a", "b", "b"],
+            "B": array([1, None, 3], dtype="Int64"),
+        }
+    )
+    result = getattr(df.groupby("A"), op)()
+    expected = DataFrame(expected, index=pd.Index(["a", "b"], name="A"), columns=["B"])
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op, expected",
+    [
+        ["sum", Series([4, 4], index=["B", "C"], dtype="Float64")],
+        ["prod", Series([3, 3], index=["B", "C"], dtype="Float64")],
+        ["min", Series([1, 1], index=["B", "C"], dtype="Float64")],
+        ["max", Series([3, 3], index=["B", "C"], dtype="Float64")],
+        ["mean", Series([2, 2], index=["B", "C"], dtype="Float64")],
+        ["median", Series([2, 2], index=["B", "C"], dtype="Float64")],
+        ["var", Series([2, 2], index=["B", "C"], dtype="Float64")],
+        ["std", Series([2**0.5, 2**0.5], index=["B", "C"], dtype="Float64")],
+        ["skew", Series([pd.NA, pd.NA], index=["B", "C"], dtype="Float64")],
+        ["kurt", Series([pd.NA, pd.NA], index=["B", "C"], dtype="Float64")],
+        ["any", Series([True, True, True], index=["A", "B", "C"], dtype="boolean")],
+        ["all", Series([True, True, True], index=["A", "B", "C"], dtype="boolean")],
+    ],
+)
+def test_mixed_reductions(op, expected):
+    df = DataFrame(
+        {
+            "A": ["a", "b", "b"],
+            "B": [1, None, 3],
+            "C": array([1, None, 3], dtype="Int64"),
+        }
+    )
+
+    # series
+    result = getattr(df.C, op)()
+    tm.assert_equal(result, expected["C"])
+
+    # frame
+    if op in ["any", "all"]:
+        result = getattr(df, op)()
+    else:
+        result = getattr(df, op)(numeric_only=True)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/integer/test_repr.py b/pandas/pandas/tests/arrays/integer/test_repr.py
new file mode 100644
index 0000000000000..168210eed5d06
--- /dev/null
+++ b/pandas/pandas/tests/arrays/integer/test_repr.py
@@ -0,0 +1,67 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    Int16Dtype,
+    Int32Dtype,
+    Int64Dtype,
+    UInt8Dtype,
+    UInt16Dtype,
+    UInt32Dtype,
+    UInt64Dtype,
+)
+
+
+def test_dtypes(dtype):
+    # smoke tests on auto dtype construction
+
+    if dtype.is_signed_integer:
+        assert np.dtype(dtype.type).kind == "i"
+    else:
+        assert np.dtype(dtype.type).kind == "u"
+    assert dtype.name is not None
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        (Int8Dtype(), "Int8Dtype()"),
+        (Int16Dtype(), "Int16Dtype()"),
+        (Int32Dtype(), "Int32Dtype()"),
+        (Int64Dtype(), "Int64Dtype()"),
+        (UInt8Dtype(), "UInt8Dtype()"),
+        (UInt16Dtype(), "UInt16Dtype()"),
+        (UInt32Dtype(), "UInt32Dtype()"),
+        (UInt64Dtype(), "UInt64Dtype()"),
+    ],
+)
+def test_repr_dtype(dtype, expected):
+    assert repr(dtype) == expected
+
+
+def test_repr_array():
+    result = repr(pd.array([1, None, 3]))
+    expected = "<IntegerArray>\n[1, <NA>, 3]\nLength: 3, dtype: Int64"
+    assert result == expected
+
+
+def test_repr_array_long():
+    data = pd.array([1, 2, None] * 1000)
+    expected = (
+        "<IntegerArray>\n"
+        "[   1,    2, <NA>,    1,    2, <NA>,    1,    2, <NA>,    1,\n"
+        " ...\n"
+        " <NA>,    1,    2, <NA>,    1,    2, <NA>,    1,    2, <NA>]\n"
+        "Length: 3000, dtype: Int64"
+    )
+    result = repr(data)
+    assert result == expected
+
+
+def test_frame_repr(data_missing):
+    df = pd.DataFrame({"A": data_missing})
+    result = repr(df)
+    expected = "      A\n0  <NA>\n1     1"
+    assert result == expected
diff --git a/pandas/pandas/tests/arrays/interval/__init__.py b/pandas/pandas/tests/arrays/interval/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/interval/test_astype.py b/pandas/pandas/tests/arrays/interval/test_astype.py
new file mode 100644
index 0000000000000..d7a2140f817f3
--- /dev/null
+++ b/pandas/pandas/tests/arrays/interval/test_astype.py
@@ -0,0 +1,28 @@
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    Index,
+    IntervalIndex,
+)
+import pandas._testing as tm
+
+
+class TestAstype:
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_astype_categorical_retains_ordered(self, ordered):
+        index = IntervalIndex.from_breaks(range(5))
+        arr = index._data
+
+        dtype = CategoricalDtype(None, ordered=ordered)
+
+        expected = Categorical(list(arr), ordered=ordered)
+        result = arr.astype(dtype)
+        assert result.ordered is ordered
+        tm.assert_categorical_equal(result, expected)
+
+        # test IntervalIndex.astype while we're at it.
+        result = index.astype(dtype)
+        expected = Index(expected)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/interval/test_formats.py b/pandas/pandas/tests/arrays/interval/test_formats.py
new file mode 100644
index 0000000000000..88c9bf81d718c
--- /dev/null
+++ b/pandas/pandas/tests/arrays/interval/test_formats.py
@@ -0,0 +1,11 @@
+from pandas.core.arrays import IntervalArray
+
+
+def test_repr():
+    # GH#25022
+    arr = IntervalArray.from_tuples([(0, 1), (1, 2)])
+    result = repr(arr)
+    expected = (
+        "<IntervalArray>\n[(0, 1], (1, 2]]\nLength: 2, dtype: interval[int64, right]"
+    )
+    assert result == expected
diff --git a/pandas/pandas/tests/arrays/interval/test_interval.py b/pandas/pandas/tests/arrays/interval/test_interval.py
new file mode 100644
index 0000000000000..8e13dcf25ceba
--- /dev/null
+++ b/pandas/pandas/tests/arrays/interval/test_interval.py
@@ -0,0 +1,231 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    Interval,
+    IntervalIndex,
+    Timedelta,
+    Timestamp,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+
+
+@pytest.fixture(
+    params=[
+        (Index([0, 2, 4]), Index([1, 3, 5])),
+        (Index([0.0, 1.0, 2.0]), Index([1.0, 2.0, 3.0])),
+        (timedelta_range("0 days", periods=3), timedelta_range("1 day", periods=3)),
+        (date_range("20170101", periods=3), date_range("20170102", periods=3)),
+        (
+            date_range("20170101", periods=3, tz="US/Eastern"),
+            date_range("20170102", periods=3, tz="US/Eastern"),
+        ),
+    ],
+    ids=lambda x: str(x[0].dtype),
+)
+def left_right_dtypes(request):
+    """
+    Fixture for building an IntervalArray from various dtypes
+    """
+    return request.param
+
+
+class TestAttributes:
+    @pytest.mark.parametrize(
+        "left, right",
+        [
+            (0, 1),
+            (Timedelta("0 days"), Timedelta("1 day")),
+            (Timestamp("2018-01-01"), Timestamp("2018-01-02")),
+            (
+                Timestamp("2018-01-01", tz="US/Eastern"),
+                Timestamp("2018-01-02", tz="US/Eastern"),
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("constructor", [IntervalArray, IntervalIndex])
+    def test_is_empty(self, constructor, left, right, closed):
+        # GH27219
+        tuples = [(left, left), (left, right), np.nan]
+        expected = np.array([closed != "both", False, False])
+        result = constructor.from_tuples(tuples, closed=closed).is_empty
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestMethods:
+    def test_set_closed(self, closed, other_closed):
+        # GH 21670
+        array = IntervalArray.from_breaks(range(10), closed=closed)
+        result = array.set_closed(other_closed)
+        expected = IntervalArray.from_breaks(range(10), closed=other_closed)
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            Interval(0, 1, closed="right"),
+            IntervalArray.from_breaks([1, 2, 3, 4], closed="right"),
+        ],
+    )
+    def test_where_raises(self, other):
+        # GH#45768 The IntervalArray methods raises; the Series method coerces
+        ser = pd.Series(IntervalArray.from_breaks([1, 2, 3, 4], closed="left"))
+        mask = np.array([True, False, True])
+        match = "'value.closed' is 'right', expected 'left'."
+        with pytest.raises(ValueError, match=match):
+            ser.array._where(mask, other)
+
+        res = ser.where(mask, other=other)
+        expected = ser.astype(object).where(mask, other)
+        tm.assert_series_equal(res, expected)
+
+    def test_shift(self):
+        # https://github.com/pandas-dev/pandas/issues/31495, GH#22428, GH#31502
+        a = IntervalArray.from_breaks([1, 2, 3])
+        result = a.shift()
+        # int -> float
+        expected = IntervalArray.from_tuples([(np.nan, np.nan), (1.0, 2.0)])
+        tm.assert_interval_array_equal(result, expected)
+
+        msg = "can only insert Interval objects and NA into an IntervalArray"
+        with pytest.raises(TypeError, match=msg):
+            a.shift(1, fill_value=pd.NaT)
+
+    def test_shift_datetime(self):
+        # GH#31502, GH#31504
+        a = IntervalArray.from_breaks(date_range("2000", periods=4))
+        result = a.shift(2)
+        expected = a.take([-1, -1, 0], allow_fill=True)
+        tm.assert_interval_array_equal(result, expected)
+
+        result = a.shift(-1)
+        expected = a.take([1, 2, -1], allow_fill=True)
+        tm.assert_interval_array_equal(result, expected)
+
+        msg = "can only insert Interval objects and NA into an IntervalArray"
+        with pytest.raises(TypeError, match=msg):
+            a.shift(1, fill_value=np.timedelta64("NaT", "ns"))
+
+
+class TestSetitem:
+    def test_set_na(self, left_right_dtypes):
+        left, right = left_right_dtypes
+        left = left.copy(deep=True)
+        right = right.copy(deep=True)
+        result = IntervalArray.from_arrays(left, right)
+
+        if result.dtype.subtype.kind not in ["m", "M"]:
+            msg = "'value' should be an interval type, got <.*NaTType'> instead."
+            with pytest.raises(TypeError, match=msg):
+                result[0] = pd.NaT
+        if result.dtype.subtype.kind in ["i", "u"]:
+            msg = "Cannot set float NaN to integer-backed IntervalArray"
+            # GH#45484 TypeError, not ValueError, matches what we get with
+            # non-NA un-holdable value.
+            with pytest.raises(TypeError, match=msg):
+                result[0] = np.nan
+            return
+
+        result[0] = np.nan
+
+        expected_left = Index([left._na_value] + list(left[1:]))
+        expected_right = Index([right._na_value] + list(right[1:]))
+        expected = IntervalArray.from_arrays(expected_left, expected_right)
+
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_setitem_mismatched_closed(self):
+        arr = IntervalArray.from_breaks(range(4))
+        orig = arr.copy()
+        other = arr.set_closed("both")
+
+        msg = "'value.closed' is 'both', expected 'right'"
+        with pytest.raises(ValueError, match=msg):
+            arr[0] = other[0]
+        with pytest.raises(ValueError, match=msg):
+            arr[:1] = other[:1]
+        with pytest.raises(ValueError, match=msg):
+            arr[:0] = other[:0]
+        with pytest.raises(ValueError, match=msg):
+            arr[:] = other[::-1]
+        with pytest.raises(ValueError, match=msg):
+            arr[:] = list(other[::-1])
+        with pytest.raises(ValueError, match=msg):
+            arr[:] = other[::-1].astype(object)
+        with pytest.raises(ValueError, match=msg):
+            arr[:] = other[::-1].astype("category")
+
+        # empty list should be no-op
+        arr[:0] = []
+        tm.assert_interval_array_equal(arr, orig)
+
+
+class TestReductions:
+    def test_min_max_invalid_axis(self, left_right_dtypes):
+        left, right = left_right_dtypes
+        left = left.copy(deep=True)
+        right = right.copy(deep=True)
+        arr = IntervalArray.from_arrays(left, right)
+
+        msg = "`axis` must be fewer than the number of dimensions"
+        for axis in [-2, 1]:
+            with pytest.raises(ValueError, match=msg):
+                arr.min(axis=axis)
+            with pytest.raises(ValueError, match=msg):
+                arr.max(axis=axis)
+
+        msg = "'>=' not supported between"
+        with pytest.raises(TypeError, match=msg):
+            arr.min(axis="foo")
+        with pytest.raises(TypeError, match=msg):
+            arr.max(axis="foo")
+
+    def test_min_max(self, left_right_dtypes, index_or_series_or_array):
+        # GH#44746
+        left, right = left_right_dtypes
+        left = left.copy(deep=True)
+        right = right.copy(deep=True)
+        arr = IntervalArray.from_arrays(left, right)
+
+        # The expected results below are only valid if monotonic
+        assert left.is_monotonic_increasing
+        assert Index(arr).is_monotonic_increasing
+
+        MIN = arr[0]
+        MAX = arr[-1]
+
+        indexer = np.arange(len(arr))
+        np.random.default_rng(2).shuffle(indexer)
+        arr = arr.take(indexer)
+
+        arr_na = arr.insert(2, np.nan)
+
+        arr = index_or_series_or_array(arr)
+        arr_na = index_or_series_or_array(arr_na)
+
+        for skipna in [True, False]:
+            res = arr.min(skipna=skipna)
+            assert res == MIN
+            assert type(res) == type(MIN)
+
+            res = arr.max(skipna=skipna)
+            assert res == MAX
+            assert type(res) == type(MAX)
+
+        res = arr_na.min(skipna=False)
+        assert np.isnan(res)
+        res = arr_na.max(skipna=False)
+        assert np.isnan(res)
+
+        for kws in [{"skipna": True}, {}]:
+            res = arr_na.min(**kws)
+            assert res == MIN
+            assert type(res) == type(MIN)
+            res = arr_na.max(**kws)
+            assert res == MAX
+            assert type(res) == type(MAX)
diff --git a/pandas/pandas/tests/arrays/interval/test_interval_pyarrow.py b/pandas/pandas/tests/arrays/interval/test_interval_pyarrow.py
new file mode 100644
index 0000000000000..ef8701be81e2b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/interval/test_interval_pyarrow.py
@@ -0,0 +1,160 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+
+
+def test_arrow_extension_type():
+    pa = pytest.importorskip("pyarrow")
+
+    from pandas.core.arrays.arrow.extension_types import ArrowIntervalType
+
+    p1 = ArrowIntervalType(pa.int64(), "left")
+    p2 = ArrowIntervalType(pa.int64(), "left")
+    p3 = ArrowIntervalType(pa.int64(), "right")
+
+    assert p1.closed == "left"
+    assert p1 == p2
+    assert p1 != p3
+    assert hash(p1) == hash(p2)
+    assert hash(p1) != hash(p3)
+
+
+def test_arrow_array():
+    pa = pytest.importorskip("pyarrow")
+
+    from pandas.core.arrays.arrow.extension_types import ArrowIntervalType
+
+    intervals = pd.interval_range(1, 5, freq=1).array
+
+    result = pa.array(intervals)
+    assert isinstance(result.type, ArrowIntervalType)
+    assert result.type.closed == intervals.closed
+    assert result.type.subtype == pa.int64()
+    assert result.storage.field("left").equals(pa.array([1, 2, 3, 4], type="int64"))
+    assert result.storage.field("right").equals(pa.array([2, 3, 4, 5], type="int64"))
+
+    expected = pa.array([{"left": i, "right": i + 1} for i in range(1, 5)])
+    assert result.storage.equals(expected)
+
+    # convert to its storage type
+    result = pa.array(intervals, type=expected.type)
+    assert result.equals(expected)
+
+    # unsupported conversions
+    with pytest.raises(TypeError, match="Not supported to convert IntervalArray"):
+        pa.array(intervals, type="float64")
+
+    with pytest.raises(TypeError, match="Not supported to convert IntervalArray"):
+        pa.array(intervals, type=ArrowIntervalType(pa.float64(), "left"))
+
+
+def test_arrow_array_missing():
+    pa = pytest.importorskip("pyarrow")
+
+    from pandas.core.arrays.arrow.extension_types import ArrowIntervalType
+
+    arr = IntervalArray.from_breaks([0.0, 1.0, 2.0, 3.0])
+    arr[1] = None
+
+    result = pa.array(arr)
+    assert isinstance(result.type, ArrowIntervalType)
+    assert result.type.closed == arr.closed
+    assert result.type.subtype == pa.float64()
+
+    # fields have missing values (not NaN)
+    left = pa.array([0.0, None, 2.0], type="float64")
+    right = pa.array([1.0, None, 3.0], type="float64")
+    assert result.storage.field("left").equals(left)
+    assert result.storage.field("right").equals(right)
+
+    # structarray itself also has missing values on the array level
+    vals = [
+        {"left": 0.0, "right": 1.0},
+        {"left": None, "right": None},
+        {"left": 2.0, "right": 3.0},
+    ]
+    expected = pa.StructArray.from_pandas(vals, mask=np.array([False, True, False]))
+    assert result.storage.equals(expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+@pytest.mark.parametrize(
+    "breaks",
+    [[0.0, 1.0, 2.0, 3.0], pd.date_range("2017", periods=4, freq="D")],
+    ids=["float", "datetime64[ns]"],
+)
+def test_arrow_table_roundtrip(breaks):
+    pa = pytest.importorskip("pyarrow")
+
+    from pandas.core.arrays.arrow.extension_types import ArrowIntervalType
+
+    arr = IntervalArray.from_breaks(breaks)
+    arr[1] = None
+    df = pd.DataFrame({"a": arr})
+
+    table = pa.table(df)
+    assert isinstance(table.field("a").type, ArrowIntervalType)
+    result = table.to_pandas()
+    assert isinstance(result["a"].dtype, pd.IntervalDtype)
+    tm.assert_frame_equal(result, df)
+
+    table2 = pa.concat_tables([table, table])
+    result = table2.to_pandas()
+    expected = pd.concat([df, df], ignore_index=True)
+    tm.assert_frame_equal(result, expected)
+
+    # GH#41040
+    table = pa.table(
+        [pa.chunked_array([], type=table.column(0).type)], schema=table.schema
+    )
+    result = table.to_pandas()
+    tm.assert_frame_equal(result, expected[0:0])
+
+
+@pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+@pytest.mark.parametrize(
+    "breaks",
+    [[0.0, 1.0, 2.0, 3.0], pd.date_range("2017", periods=4, freq="D")],
+    ids=["float", "datetime64[ns]"],
+)
+def test_arrow_table_roundtrip_without_metadata(breaks):
+    pa = pytest.importorskip("pyarrow")
+
+    arr = IntervalArray.from_breaks(breaks)
+    arr[1] = None
+    df = pd.DataFrame({"a": arr})
+
+    table = pa.table(df)
+    # remove the metadata
+    table = table.replace_schema_metadata()
+    assert table.schema.metadata is None
+
+    result = table.to_pandas()
+    assert isinstance(result["a"].dtype, pd.IntervalDtype)
+    tm.assert_frame_equal(result, df)
+
+
+def test_from_arrow_from_raw_struct_array():
+    # in case pyarrow lost the Interval extension type (eg on parquet roundtrip
+    # with datetime64[ns] subtype, see GH-45881), still allow conversion
+    # from arrow to IntervalArray
+    pa = pytest.importorskip("pyarrow")
+
+    arr = pa.array([{"left": 0, "right": 1}, {"left": 1, "right": 2}])
+    dtype = pd.IntervalDtype(np.dtype("int64"), closed="neither")
+
+    result = dtype.__from_arrow__(arr)
+    expected = IntervalArray.from_breaks(
+        np.array([0, 1, 2], dtype="int64"), closed="neither"
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    result = dtype.__from_arrow__(pa.chunked_array([arr]))
+    tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/interval/test_overlaps.py b/pandas/pandas/tests/arrays/interval/test_overlaps.py
new file mode 100644
index 0000000000000..5a48cf024ec0d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/interval/test_overlaps.py
@@ -0,0 +1,94 @@
+"""Tests for Interval-Interval operations, such as overlaps, contains, etc."""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Interval,
+    IntervalIndex,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+
+
+@pytest.fixture(params=[IntervalArray, IntervalIndex])
+def constructor(request):
+    """
+    Fixture for testing both interval container classes.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        (Timedelta("0 days"), Timedelta("1 day")),
+        (Timestamp("2018-01-01"), Timedelta("1 day")),
+        (0, 1),
+    ],
+    ids=lambda x: type(x[0]).__name__,
+)
+def start_shift(request):
+    """
+    Fixture for generating intervals of different types from a start value
+    and a shift value that can be added to start to generate an endpoint.
+    """
+    return request.param
+
+
+class TestOverlaps:
+    def test_overlaps_interval(self, constructor, start_shift, closed, other_closed):
+        start, shift = start_shift
+        interval = Interval(start, start + 3 * shift, other_closed)
+
+        # intervals: identical, nested, spanning, partial, adjacent, disjoint
+        tuples = [
+            (start, start + 3 * shift),
+            (start + shift, start + 2 * shift),
+            (start - shift, start + 4 * shift),
+            (start + 2 * shift, start + 4 * shift),
+            (start + 3 * shift, start + 4 * shift),
+            (start + 4 * shift, start + 5 * shift),
+        ]
+        interval_container = constructor.from_tuples(tuples, closed)
+
+        adjacent = interval.closed_right and interval_container.closed_left
+        expected = np.array([True, True, True, True, adjacent, False])
+        result = interval_container.overlaps(interval)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("other_constructor", [IntervalArray, IntervalIndex])
+    def test_overlaps_interval_container(self, constructor, other_constructor):
+        # TODO: modify this test when implemented
+        interval_container = constructor.from_breaks(range(5))
+        other_container = other_constructor.from_breaks(range(5))
+        with pytest.raises(NotImplementedError, match="^$"):
+            interval_container.overlaps(other_container)
+
+    def test_overlaps_na(self, constructor, start_shift):
+        """NA values are marked as False"""
+        start, shift = start_shift
+        interval = Interval(start, start + shift)
+
+        tuples = [
+            (start, start + shift),
+            np.nan,
+            (start + 2 * shift, start + 3 * shift),
+        ]
+        interval_container = constructor.from_tuples(tuples)
+
+        expected = np.array([True, False, False])
+        result = interval_container.overlaps(interval)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other",
+        [10, True, "foo", Timedelta("1 day"), Timestamp("2018-01-01")],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_overlaps_invalid_type(self, constructor, other):
+        interval_container = constructor.from_breaks(range(5))
+        msg = f"`other` must be Interval-like, got {type(other).__name__}"
+        with pytest.raises(TypeError, match=msg):
+            interval_container.overlaps(other)
diff --git a/pandas/pandas/tests/arrays/masked/__init__.py b/pandas/pandas/tests/arrays/masked/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/masked/test_arithmetic.py b/pandas/pandas/tests/arrays/masked/test_arithmetic.py
new file mode 100644
index 0000000000000..ea018d2da4d26
--- /dev/null
+++ b/pandas/pandas/tests/arrays/masked/test_arithmetic.py
@@ -0,0 +1,248 @@
+from __future__ import annotations
+
+from typing import Any
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+# integer dtypes
+arrays = [pd.array([1, 2, 3, None], dtype=dtype) for dtype in tm.ALL_INT_EA_DTYPES]
+scalars: list[Any] = [2] * len(arrays)
+# floating dtypes
+arrays += [pd.array([0.1, 0.2, 0.3, None], dtype=dtype) for dtype in tm.FLOAT_EA_DTYPES]
+scalars += [0.2, 0.2]
+# boolean
+arrays += [pd.array([True, False, True, None], dtype="boolean")]
+scalars += [False]
+
+
+@pytest.fixture(params=zip(arrays, scalars), ids=[a.dtype.name for a in arrays])
+def data(request):
+    """Fixture returning parametrized (array, scalar) tuple.
+
+    Used to test equivalence of scalars, numpy arrays with array ops, and the
+    equivalence of DataFrame and Series ops.
+    """
+    return request.param
+
+
+def check_skip(data, op_name):
+    if isinstance(data.dtype, pd.BooleanDtype) and "sub" in op_name:
+        pytest.skip("subtract not implemented for boolean")
+
+
+def is_bool_not_implemented(data, op_name):
+    # match non-masked behavior
+    return data.dtype.kind == "b" and op_name.strip("_").lstrip("r") in [
+        "pow",
+        "truediv",
+        "floordiv",
+    ]
+
+
+# Test equivalence of scalars, numpy arrays with array ops
+# -----------------------------------------------------------------------------
+
+
+def test_array_scalar_like_equivalence(data, all_arithmetic_operators):
+    data, scalar = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    check_skip(data, all_arithmetic_operators)
+
+    scalar_array = pd.array([scalar] * len(data), dtype=data.dtype)
+
+    # TODO also add len-1 array (np.array([scalar], dtype=data.dtype.numpy_dtype))
+    for val in [scalar, data.dtype.type(scalar)]:
+        if is_bool_not_implemented(data, all_arithmetic_operators):
+            msg = "operator '.*' not implemented for bool dtypes"
+            with pytest.raises(NotImplementedError, match=msg):
+                op(data, val)
+            with pytest.raises(NotImplementedError, match=msg):
+                op(data, scalar_array)
+        else:
+            result = op(data, val)
+            expected = op(data, scalar_array)
+            tm.assert_extension_array_equal(result, expected)
+
+
+def test_array_NA(data, all_arithmetic_operators):
+    data, _ = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    check_skip(data, all_arithmetic_operators)
+
+    scalar = pd.NA
+    scalar_array = pd.array([pd.NA] * len(data), dtype=data.dtype)
+
+    mask = data._mask.copy()
+
+    if is_bool_not_implemented(data, all_arithmetic_operators):
+        msg = "operator '.*' not implemented for bool dtypes"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(data, scalar)
+        # GH#45421 check op doesn't alter data._mask inplace
+        tm.assert_numpy_array_equal(mask, data._mask)
+        return
+
+    result = op(data, scalar)
+    # GH#45421 check op doesn't alter data._mask inplace
+    tm.assert_numpy_array_equal(mask, data._mask)
+
+    expected = op(data, scalar_array)
+    tm.assert_numpy_array_equal(mask, data._mask)
+
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_numpy_array_equivalence(data, all_arithmetic_operators):
+    data, scalar = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    check_skip(data, all_arithmetic_operators)
+
+    numpy_array = np.array([scalar] * len(data), dtype=data.dtype.numpy_dtype)
+    pd_array = pd.array(numpy_array, dtype=data.dtype)
+
+    if is_bool_not_implemented(data, all_arithmetic_operators):
+        msg = "operator '.*' not implemented for bool dtypes"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(data, numpy_array)
+        with pytest.raises(NotImplementedError, match=msg):
+            op(data, pd_array)
+        return
+
+    result = op(data, numpy_array)
+    expected = op(data, pd_array)
+    tm.assert_extension_array_equal(result, expected)
+
+
+# Test equivalence with Series and DataFrame ops
+# -----------------------------------------------------------------------------
+
+
+def test_frame(data, all_arithmetic_operators):
+    data, scalar = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    check_skip(data, all_arithmetic_operators)
+
+    # DataFrame with scalar
+    df = pd.DataFrame({"A": data})
+
+    if is_bool_not_implemented(data, all_arithmetic_operators):
+        msg = "operator '.*' not implemented for bool dtypes"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(df, scalar)
+        with pytest.raises(NotImplementedError, match=msg):
+            op(data, scalar)
+        return
+
+    result = op(df, scalar)
+    expected = pd.DataFrame({"A": op(data, scalar)})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_series(data, all_arithmetic_operators):
+    data, scalar = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+    check_skip(data, all_arithmetic_operators)
+
+    ser = pd.Series(data)
+
+    others = [
+        scalar,
+        np.array([scalar] * len(data), dtype=data.dtype.numpy_dtype),
+        pd.array([scalar] * len(data), dtype=data.dtype),
+        pd.Series([scalar] * len(data), dtype=data.dtype),
+    ]
+
+    for other in others:
+        if is_bool_not_implemented(data, all_arithmetic_operators):
+            msg = "operator '.*' not implemented for bool dtypes"
+            with pytest.raises(NotImplementedError, match=msg):
+                op(ser, other)
+
+        else:
+            result = op(ser, other)
+            expected = pd.Series(op(data, other))
+            tm.assert_series_equal(result, expected)
+
+
+# Test generic characteristics / errors
+# -----------------------------------------------------------------------------
+
+
+def test_error_invalid_object(data, all_arithmetic_operators):
+    data, _ = data
+
+    op = all_arithmetic_operators
+    opa = getattr(data, op)
+
+    # 2d -> return NotImplemented
+    result = opa(pd.DataFrame({"A": data}))
+    assert result is NotImplemented
+
+    msg = r"can only perform ops with 1-d structures"
+    with pytest.raises(NotImplementedError, match=msg):
+        opa(np.arange(len(data)).reshape(-1, len(data)))
+
+
+def test_error_len_mismatch(data, all_arithmetic_operators):
+    # operating with a list-like with non-matching length raises
+    data, scalar = data
+    op = tm.get_op_from_name(all_arithmetic_operators)
+
+    other = [scalar] * (len(data) - 1)
+
+    err = ValueError
+    msg = "|".join(
+        [
+            r"operands could not be broadcast together with shapes \(3,\) \(4,\)",
+            r"operands could not be broadcast together with shapes \(4,\) \(3,\)",
+        ]
+    )
+    if data.dtype.kind == "b" and all_arithmetic_operators.strip("_") in [
+        "sub",
+        "rsub",
+    ]:
+        err = TypeError
+        msg = (
+            r"numpy boolean subtract, the `\-` operator, is not supported, use "
+            r"the bitwise_xor, the `\^` operator, or the logical_xor function instead"
+        )
+    elif is_bool_not_implemented(data, all_arithmetic_operators):
+        msg = "operator '.*' not implemented for bool dtypes"
+        err = NotImplementedError
+
+    for val in [other, np.array(other)]:
+        with pytest.raises(err, match=msg):
+            op(data, val)
+
+        s = pd.Series(data)
+        with pytest.raises(err, match=msg):
+            op(s, val)
+
+
+@pytest.mark.parametrize("op", ["__neg__", "__abs__", "__invert__"])
+def test_unary_op_does_not_propagate_mask(data, op):
+    # https://github.com/pandas-dev/pandas/issues/39943
+    data, _ = data
+    ser = pd.Series(data)
+
+    if op == "__invert__" and data.dtype.kind == "f":
+        # we follow numpy in raising
+        msg = "ufunc 'invert' not supported for the input types"
+        with pytest.raises(TypeError, match=msg):
+            getattr(ser, op)()
+        with pytest.raises(TypeError, match=msg):
+            getattr(data, op)()
+        with pytest.raises(TypeError, match=msg):
+            # Check that this is still the numpy behavior
+            getattr(data._data, op)()
+
+        return
+
+    result = getattr(ser, op)()
+    expected = result.copy(deep=True)
+    ser[0] = None
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/masked/test_arrow_compat.py b/pandas/pandas/tests/arrays/masked/test_arrow_compat.py
new file mode 100644
index 0000000000000..d99b1118444c9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/masked/test_arrow_compat.py
@@ -0,0 +1,210 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+pa = pytest.importorskip("pyarrow")
+
+from pandas.core.arrays.arrow._arrow_utils import pyarrow_array_to_numpy_and_mask
+
+arrays = [pd.array([1, 2, 3, None], dtype=dtype) for dtype in tm.ALL_INT_EA_DTYPES]
+arrays += [pd.array([0.1, 0.2, 0.3, None], dtype=dtype) for dtype in tm.FLOAT_EA_DTYPES]
+arrays += [pd.array([True, False, True, None], dtype="boolean")]
+
+
+@pytest.fixture(params=arrays, ids=[a.dtype.name for a in arrays])
+def data(request):
+    """
+    Fixture returning parametrized array from given dtype, including integer,
+    float and boolean
+    """
+    return request.param
+
+
+def test_arrow_array(data):
+    arr = pa.array(data)
+    expected = pa.array(
+        data.to_numpy(object, na_value=None),
+        type=pa.from_numpy_dtype(data.dtype.numpy_dtype),
+    )
+    assert arr.equals(expected)
+
+
+def test_arrow_roundtrip(data):
+    df = pd.DataFrame({"a": data})
+    table = pa.table(df)
+    assert table.field("a").type == str(data.dtype.numpy_dtype)
+
+    result = table.to_pandas()
+    assert result["a"].dtype == data.dtype
+    tm.assert_frame_equal(result, df)
+
+
+def test_dataframe_from_arrow_types_mapper():
+    def types_mapper(arrow_type):
+        if pa.types.is_boolean(arrow_type):
+            return pd.BooleanDtype()
+        elif pa.types.is_integer(arrow_type):
+            return pd.Int64Dtype()
+
+    bools_array = pa.array([True, None, False], type=pa.bool_())
+    ints_array = pa.array([1, None, 2], type=pa.int64())
+    small_ints_array = pa.array([-1, 0, 7], type=pa.int8())
+    record_batch = pa.RecordBatch.from_arrays(
+        [bools_array, ints_array, small_ints_array], ["bools", "ints", "small_ints"]
+    )
+    result = record_batch.to_pandas(types_mapper=types_mapper)
+    bools = pd.Series([True, None, False], dtype="boolean")
+    ints = pd.Series([1, None, 2], dtype="Int64")
+    small_ints = pd.Series([-1, 0, 7], dtype="Int64")
+    expected = pd.DataFrame({"bools": bools, "ints": ints, "small_ints": small_ints})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_arrow_load_from_zero_chunks(data):
+    # GH-41040
+
+    df = pd.DataFrame({"a": data[0:0]})
+    table = pa.table(df)
+    assert table.field("a").type == str(data.dtype.numpy_dtype)
+    table = pa.table(
+        [pa.chunked_array([], type=table.field("a").type)], schema=table.schema
+    )
+    result = table.to_pandas()
+    assert result["a"].dtype == data.dtype
+    tm.assert_frame_equal(result, df)
+
+
+def test_arrow_from_arrow_uint():
+    # https://github.com/pandas-dev/pandas/issues/31896
+    # possible mismatch in types
+
+    dtype = pd.UInt32Dtype()
+    result = dtype.__from_arrow__(pa.array([1, 2, 3, 4, None], type="int64"))
+    expected = pd.array([1, 2, 3, 4, None], dtype="UInt32")
+
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_arrow_sliced(data):
+    # https://github.com/pandas-dev/pandas/issues/38525
+
+    df = pd.DataFrame({"a": data})
+    table = pa.table(df)
+    result = table.slice(2, None).to_pandas()
+    expected = df.iloc[2:].reset_index(drop=True)
+    tm.assert_frame_equal(result, expected)
+
+    # no missing values
+    df2 = df.fillna(data[0])
+    table = pa.table(df2)
+    result = table.slice(2, None).to_pandas()
+    expected = df2.iloc[2:].reset_index(drop=True)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.fixture
+def np_dtype_to_arrays(any_real_numpy_dtype):
+    """
+    Fixture returning actual and expected dtype, pandas and numpy arrays and
+    mask from a given numpy dtype
+    """
+    np_dtype = np.dtype(any_real_numpy_dtype)
+    pa_type = pa.from_numpy_dtype(np_dtype)
+
+    # None ensures the creation of a bitmask buffer.
+    pa_array = pa.array([0, 1, 2, None], type=pa_type)
+    # Since masked Arrow buffer slots are not required to contain a specific
+    # value, assert only the first three values of the created np.array
+    np_expected = np.array([0, 1, 2], dtype=np_dtype)
+    mask_expected = np.array([True, True, True, False])
+    return np_dtype, pa_array, np_expected, mask_expected
+
+
+def test_pyarrow_array_to_numpy_and_mask(np_dtype_to_arrays):
+    """
+    Test conversion from pyarrow array to numpy array.
+
+    Modifies the pyarrow buffer to contain padding and offset, which are
+    considered valid buffers by pyarrow.
+
+    Also tests empty pyarrow arrays with non empty buffers.
+    See https://github.com/pandas-dev/pandas/issues/40896
+    """
+    np_dtype, pa_array, np_expected, mask_expected = np_dtype_to_arrays
+    data, mask = pyarrow_array_to_numpy_and_mask(pa_array, np_dtype)
+    tm.assert_numpy_array_equal(data[:3], np_expected)
+    tm.assert_numpy_array_equal(mask, mask_expected)
+
+    mask_buffer = pa_array.buffers()[0]
+    data_buffer = pa_array.buffers()[1]
+    data_buffer_bytes = pa_array.buffers()[1].to_pybytes()
+
+    # Add trailing padding to the buffer.
+    data_buffer_trail = pa.py_buffer(data_buffer_bytes + b"\x00")
+    pa_array_trail = pa.Array.from_buffers(
+        type=pa_array.type,
+        length=len(pa_array),
+        buffers=[mask_buffer, data_buffer_trail],
+        offset=pa_array.offset,
+    )
+    pa_array_trail.validate()
+    data, mask = pyarrow_array_to_numpy_and_mask(pa_array_trail, np_dtype)
+    tm.assert_numpy_array_equal(data[:3], np_expected)
+    tm.assert_numpy_array_equal(mask, mask_expected)
+
+    # Add offset to the buffer.
+    offset = b"\x00" * (pa_array.type.bit_width // 8)
+    data_buffer_offset = pa.py_buffer(offset + data_buffer_bytes)
+    mask_buffer_offset = pa.py_buffer(b"\x0e")
+    pa_array_offset = pa.Array.from_buffers(
+        type=pa_array.type,
+        length=len(pa_array),
+        buffers=[mask_buffer_offset, data_buffer_offset],
+        offset=pa_array.offset + 1,
+    )
+    pa_array_offset.validate()
+    data, mask = pyarrow_array_to_numpy_and_mask(pa_array_offset, np_dtype)
+    tm.assert_numpy_array_equal(data[:3], np_expected)
+    tm.assert_numpy_array_equal(mask, mask_expected)
+
+    # Empty array
+    np_expected_empty = np.array([], dtype=np_dtype)
+    mask_expected_empty = np.array([], dtype=np.bool_)
+
+    pa_array_offset = pa.Array.from_buffers(
+        type=pa_array.type,
+        length=0,
+        buffers=[mask_buffer, data_buffer],
+        offset=pa_array.offset,
+    )
+    pa_array_offset.validate()
+    data, mask = pyarrow_array_to_numpy_and_mask(pa_array_offset, np_dtype)
+    tm.assert_numpy_array_equal(data[:3], np_expected_empty)
+    tm.assert_numpy_array_equal(mask, mask_expected_empty)
+
+
+@pytest.mark.parametrize(
+    "arr", [pa.nulls(10), pa.chunked_array([pa.nulls(4), pa.nulls(6)])]
+)
+def test_from_arrow_null(data, arr):
+    res = data.dtype.__from_arrow__(arr)
+    assert res.isna().all()
+    assert len(res) == 10
+
+
+def test_from_arrow_type_error(data):
+    # ensure that __from_arrow__ returns a TypeError when getting a wrong
+    # array type
+
+    arr = pa.array(data).cast("string")
+    with pytest.raises(TypeError, match=None):
+        # we don't test the exact error message, only the fact that it raises
+        # a TypeError is relevant
+        data.dtype.__from_arrow__(arr)
diff --git a/pandas/pandas/tests/arrays/masked/test_function.py b/pandas/pandas/tests/arrays/masked/test_function.py
new file mode 100644
index 0000000000000..b4b1761217826
--- /dev/null
+++ b/pandas/pandas/tests/arrays/masked/test_function.py
@@ -0,0 +1,74 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_integer_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import BaseMaskedArray
+
+arrays = [pd.array([1, 2, 3, None], dtype=dtype) for dtype in tm.ALL_INT_EA_DTYPES]
+arrays += [
+    pd.array([0.141, -0.268, 5.895, None], dtype=dtype) for dtype in tm.FLOAT_EA_DTYPES
+]
+
+
+@pytest.fixture(params=arrays, ids=[a.dtype.name for a in arrays])
+def data(request):
+    """
+    Fixture returning parametrized 'data' array with different integer and
+    floating point types
+    """
+    return request.param
+
+
+@pytest.fixture
+def numpy_dtype(data):
+    """
+    Fixture returning numpy dtype from 'data' input array.
+    """
+    # For integer dtype, the numpy conversion must be done to float
+    if is_integer_dtype(data):
+        numpy_dtype = float
+    else:
+        numpy_dtype = data.dtype.type
+    return numpy_dtype
+
+
+def test_round(data, numpy_dtype):
+    # No arguments
+    result = data.round()
+    expected = pd.array(
+        np.round(data.to_numpy(dtype=numpy_dtype, na_value=None)), dtype=data.dtype
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    # Decimals argument
+    result = data.round(decimals=2)
+    expected = pd.array(
+        np.round(data.to_numpy(dtype=numpy_dtype, na_value=None), decimals=2),
+        dtype=data.dtype,
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_tolist(data):
+    result = data.tolist()
+    expected = list(data)
+    tm.assert_equal(result, expected)
+
+
+def test_to_numpy():
+    # GH#56991
+
+    class MyStringArray(BaseMaskedArray):
+        dtype = pd.StringDtype()
+        _dtype_cls = pd.StringDtype
+        _internal_fill_value = pd.NA
+
+    arr = MyStringArray(
+        values=np.array(["a", "b", "c"]), mask=np.array([False, True, False])
+    )
+    result = arr.to_numpy()
+    expected = np.array(["a", pd.NA, "c"])
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/masked/test_indexing.py b/pandas/pandas/tests/arrays/masked/test_indexing.py
new file mode 100644
index 0000000000000..753d562c87ffa
--- /dev/null
+++ b/pandas/pandas/tests/arrays/masked/test_indexing.py
@@ -0,0 +1,60 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+
+
+class TestSetitemValidation:
+    def _check_setitem_invalid(self, arr, invalid):
+        msg = f"Invalid value '{invalid!s}' for dtype '{arr.dtype}'"
+        msg = re.escape(msg)
+        with pytest.raises(TypeError, match=msg):
+            arr[0] = invalid
+
+        with pytest.raises(TypeError, match=msg):
+            arr[:] = invalid
+
+        with pytest.raises(TypeError, match=msg):
+            arr[[0]] = invalid
+
+        # FIXME: don't leave commented-out
+        # with pytest.raises(TypeError):
+        #    arr[[0]] = [invalid]
+
+        # with pytest.raises(TypeError):
+        #    arr[[0]] = np.array([invalid], dtype=object)
+
+        # Series non-coercion, behavior subject to change
+        ser = pd.Series(arr)
+        with pytest.raises(TypeError, match=msg):
+            ser[0] = invalid
+            # TODO: so, so many other variants of this...
+
+    _invalid_scalars = [
+        1 + 2j,
+        "True",
+        "1",
+        "1.0",
+        pd.NaT,
+        np.datetime64("NaT"),
+        np.timedelta64("NaT"),
+    ]
+
+    @pytest.mark.parametrize(
+        "invalid", _invalid_scalars + [1, 1.0, np.int64(1), np.float64(1)]
+    )
+    def test_setitem_validation_scalar_bool(self, invalid):
+        arr = pd.array([True, False, None], dtype="boolean")
+        self._check_setitem_invalid(arr, invalid)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True, 1.5, np.float64(1.5)])
+    def test_setitem_validation_scalar_int(self, invalid, any_int_ea_dtype):
+        arr = pd.array([1, 2, None], dtype=any_int_ea_dtype)
+        self._check_setitem_invalid(arr, invalid)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True])
+    def test_setitem_validation_scalar_float(self, invalid, float_ea_dtype):
+        arr = pd.array([1, 2, None], dtype=float_ea_dtype)
+        self._check_setitem_invalid(arr, invalid)
diff --git a/pandas/pandas/tests/arrays/masked_shared.py b/pandas/pandas/tests/arrays/masked_shared.py
new file mode 100644
index 0000000000000..545b14af2c98b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/masked_shared.py
@@ -0,0 +1,155 @@
+"""
+Tests shared by MaskedArray subclasses.
+"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.extension.base import BaseOpsUtil
+
+
+class ComparisonOps(BaseOpsUtil):
+    def _compare_other(self, data, op, other):
+        # array
+        result = pd.Series(op(data, other))
+        expected = pd.Series(op(data._data, other), dtype="boolean")
+
+        # fill the nan locations
+        expected[data._mask] = pd.NA
+
+        tm.assert_series_equal(result, expected)
+
+        # series
+        ser = pd.Series(data)
+        result = op(ser, other)
+
+        # Set nullable dtype here to avoid upcasting when setting to pd.NA below
+        expected = op(pd.Series(data._data), other).astype("boolean")
+
+        # fill the nan locations
+        expected[data._mask] = pd.NA
+
+        tm.assert_series_equal(result, expected)
+
+    # subclass will override to parametrize 'other'
+    def test_scalar(self, other, comparison_op, dtype):
+        op = comparison_op
+        left = pd.array([1, 0, None], dtype=dtype)
+
+        result = op(left, other)
+
+        if other is pd.NA:
+            expected = pd.array([None, None, None], dtype="boolean")
+        else:
+            values = op(left._data, other)
+            expected = pd.arrays.BooleanArray(values, left._mask, copy=True)
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        result[0] = pd.NA
+        tm.assert_extension_array_equal(left, pd.array([1, 0, None], dtype=dtype))
+
+
+class NumericOps:
+    # Shared by IntegerArray and FloatingArray, not BooleanArray
+
+    def test_searchsorted_nan(self, dtype):
+        # The base class casts to object dtype, for which searchsorted returns
+        #  0 from the left and 10 from the right.
+        arr = pd.array(range(10), dtype=dtype)
+
+        assert arr.searchsorted(np.nan, side="left") == 10
+        assert arr.searchsorted(np.nan, side="right") == 10
+
+    def test_no_shared_mask(self, data):
+        result = data + 1
+        assert not tm.shares_memory(result, data)
+
+    def test_array(self, comparison_op, dtype):
+        op = comparison_op
+
+        left = pd.array([0, 1, 2, None, None, None], dtype=dtype)
+        right = pd.array([0, 1, None, 0, 1, None], dtype=dtype)
+
+        result = op(left, right)
+        values = op(left._data, right._data)
+        mask = left._mask | right._mask
+
+        expected = pd.arrays.BooleanArray(values, mask)
+        tm.assert_extension_array_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        result[0] = pd.NA
+        tm.assert_extension_array_equal(
+            left, pd.array([0, 1, 2, None, None, None], dtype=dtype)
+        )
+        tm.assert_extension_array_equal(
+            right, pd.array([0, 1, None, 0, 1, None], dtype=dtype)
+        )
+
+    def test_compare_with_booleanarray(self, comparison_op, dtype):
+        op = comparison_op
+
+        left = pd.array([True, False, None] * 3, dtype="boolean")
+        right = pd.array([0] * 3 + [1] * 3 + [None] * 3, dtype=dtype)
+        other = pd.array([False] * 3 + [True] * 3 + [None] * 3, dtype="boolean")
+
+        expected = op(left, other)
+        result = op(left, right)
+        tm.assert_extension_array_equal(result, expected)
+
+        # reversed op
+        expected = op(other, left)
+        result = op(right, left)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_compare_to_string(self, dtype):
+        # GH#28930
+        ser = pd.Series([1, None], dtype=dtype)
+        result = ser == "a"
+        expected = pd.Series([False, pd.NA], dtype="boolean")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_ufunc_with_out(self, dtype):
+        arr = pd.array([1, 2, 3], dtype=dtype)
+        arr2 = pd.array([1, 2, pd.NA], dtype=dtype)
+
+        mask = arr == arr
+        mask2 = arr2 == arr2
+
+        result = np.zeros(3, dtype=bool)
+        result |= mask
+        # If MaskedArray.__array_ufunc__ handled "out" appropriately,
+        #  `result` should still be an ndarray.
+        assert isinstance(result, np.ndarray)
+        assert result.all()
+
+        # result |= mask worked because mask could be cast losslessly to
+        #  boolean ndarray. mask2 can't, so this raises
+        result = np.zeros(3, dtype=bool)
+        msg = "Specify an appropriate 'na_value' for this dtype"
+        with pytest.raises(ValueError, match=msg):
+            result |= mask2
+
+        # addition
+        res = np.add(arr, arr2)
+        expected = pd.array([2, 4, pd.NA], dtype=dtype)
+        tm.assert_extension_array_equal(res, expected)
+
+        # when passing out=arr, we will modify 'arr' inplace.
+        res = np.add(arr, arr2, out=arr)
+        assert res is arr
+        tm.assert_extension_array_equal(res, expected)
+        tm.assert_extension_array_equal(arr, expected)
+
+    def test_mul_td64_array(self, dtype):
+        # GH#45622
+        arr = pd.array([1, 2, pd.NA], dtype=dtype)
+        other = np.arange(3, dtype=np.int64).view("m8[ns]")
+
+        result = arr * other
+        expected = pd.array([pd.Timedelta(0), pd.Timedelta(2), pd.NaT])
+        tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/numpy_/__init__.py b/pandas/pandas/tests/arrays/numpy_/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/numpy_/test_indexing.py b/pandas/pandas/tests/arrays/numpy_/test_indexing.py
new file mode 100644
index 0000000000000..225d64ad7d258
--- /dev/null
+++ b/pandas/pandas/tests/arrays/numpy_/test_indexing.py
@@ -0,0 +1,41 @@
+import numpy as np
+
+from pandas.core.dtypes.common import is_scalar
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestSearchsorted:
+    def test_searchsorted_string(self, string_dtype):
+        arr = pd.array(["a", "b", "c"], dtype=string_dtype)
+
+        result = arr.searchsorted("a", side="left")
+        assert is_scalar(result)
+        assert result == 0
+
+        result = arr.searchsorted("a", side="right")
+        assert is_scalar(result)
+        assert result == 1
+
+    def test_searchsorted_numeric_dtypes_scalar(self, any_real_numpy_dtype):
+        arr = pd.array([1, 3, 90], dtype=any_real_numpy_dtype)
+        result = arr.searchsorted(30)
+        assert is_scalar(result)
+        assert result == 2
+
+        result = arr.searchsorted([30])
+        expected = np.array([2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_searchsorted_numeric_dtypes_vector(self, any_real_numpy_dtype):
+        arr = pd.array([1, 3, 90], dtype=any_real_numpy_dtype)
+        result = arr.searchsorted([2, 30])
+        expected = np.array([1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_searchsorted_sorter(self, any_real_numpy_dtype):
+        arr = pd.array([3, 1, 2], dtype=any_real_numpy_dtype)
+        result = arr.searchsorted([0, 3], sorter=np.argsort(arr))
+        expected = np.array([0, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/numpy_/test_numpy.py b/pandas/pandas/tests/arrays/numpy_/test_numpy.py
new file mode 100644
index 0000000000000..e86eb014465e1
--- /dev/null
+++ b/pandas/pandas/tests/arrays/numpy_/test_numpy.py
@@ -0,0 +1,325 @@
+"""
+Additional tests for NumpyExtensionArray that aren't covered by
+the interface tests.
+"""
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import NumpyEADtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.arrays import NumpyExtensionArray
+
+
+@pytest.fixture(
+    params=[
+        np.array(["a", "b"], dtype=object),
+        np.array([0, 1], dtype=float),
+        np.array([0, 1], dtype=int),
+        np.array([0, 1 + 2j], dtype=complex),
+        np.array([True, False], dtype=bool),
+        np.array([0, 1], dtype="datetime64[ns]"),
+        np.array([0, 1], dtype="timedelta64[ns]"),
+    ]
+)
+def any_numpy_array(request):
+    """
+    Parametrized fixture for NumPy arrays with different dtypes.
+
+    This excludes string and bytes.
+    """
+    return request.param
+
+
+# ----------------------------------------------------------------------------
+# NumpyEADtype
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        ("bool", True),
+        ("int", True),
+        ("uint", True),
+        ("float", True),
+        ("complex", True),
+        ("str", False),
+        ("bytes", False),
+        ("datetime64[ns]", False),
+        ("object", False),
+        ("void", False),
+    ],
+)
+def test_is_numeric(dtype, expected):
+    dtype = NumpyEADtype(dtype)
+    assert dtype._is_numeric is expected
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        ("bool", True),
+        ("int", False),
+        ("uint", False),
+        ("float", False),
+        ("complex", False),
+        ("str", False),
+        ("bytes", False),
+        ("datetime64[ns]", False),
+        ("object", False),
+        ("void", False),
+    ],
+)
+def test_is_boolean(dtype, expected):
+    dtype = NumpyEADtype(dtype)
+    assert dtype._is_boolean is expected
+
+
+def test_repr():
+    dtype = NumpyEADtype(np.dtype("int64"))
+    assert repr(dtype) == "NumpyEADtype('int64')"
+
+
+def test_constructor_from_string():
+    result = NumpyEADtype.construct_from_string("int64")
+    expected = NumpyEADtype(np.dtype("int64"))
+    assert result == expected
+
+
+def test_dtype_idempotent(any_numpy_dtype):
+    dtype = NumpyEADtype(any_numpy_dtype)
+
+    result = NumpyEADtype(dtype)
+    assert result == dtype
+
+
+# ----------------------------------------------------------------------------
+# Construction
+
+
+def test_constructor_no_coercion():
+    with pytest.raises(ValueError, match="NumPy array"):
+        NumpyExtensionArray([1, 2, 3])
+
+
+def test_series_constructor_with_copy():
+    ndarray = np.array([1, 2, 3])
+    ser = pd.Series(NumpyExtensionArray(ndarray), copy=True)
+
+    assert ser.values is not ndarray
+
+
+def test_series_constructor_with_astype():
+    ndarray = np.array([1, 2, 3])
+    result = pd.Series(NumpyExtensionArray(ndarray), dtype="float64")
+    expected = pd.Series([1.0, 2.0, 3.0], dtype="float64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_from_sequence_dtype():
+    arr = np.array([1, 2, 3], dtype="int64")
+    result = NumpyExtensionArray._from_sequence(arr, dtype="uint64")
+    expected = NumpyExtensionArray(np.array([1, 2, 3], dtype="uint64"))
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_constructor_copy():
+    arr = np.array([0, 1])
+    result = NumpyExtensionArray(arr, copy=True)
+
+    assert not tm.shares_memory(result, arr)
+
+
+def test_constructor_with_data(any_numpy_array):
+    nparr = any_numpy_array
+    arr = NumpyExtensionArray(nparr)
+    assert arr.dtype.numpy_dtype == nparr.dtype
+
+
+# ----------------------------------------------------------------------------
+# Conversion
+
+
+def test_to_numpy():
+    arr = NumpyExtensionArray(np.array([1, 2, 3]))
+    result = arr.to_numpy()
+    assert result is arr._ndarray
+
+    result = arr.to_numpy(copy=True)
+    assert result is not arr._ndarray
+
+    result = arr.to_numpy(dtype="f8")
+    expected = np.array([1, 2, 3], dtype="f8")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+# ----------------------------------------------------------------------------
+# Setitem
+
+
+def test_setitem_series():
+    ser = pd.Series([1, 2, 3])
+    ser.array[0] = 10
+    expected = pd.Series([10, 2, 3])
+    tm.assert_series_equal(ser, expected)
+
+
+def test_setitem(any_numpy_array):
+    nparr = any_numpy_array
+    arr = NumpyExtensionArray(nparr, copy=True)
+
+    arr[0] = arr[1]
+    nparr[0] = nparr[1]
+
+    tm.assert_numpy_array_equal(arr.to_numpy(), nparr)
+
+
+# ----------------------------------------------------------------------------
+# Reductions
+
+
+def test_bad_reduce_raises():
+    arr = np.array([1, 2, 3], dtype="int64")
+    arr = NumpyExtensionArray(arr)
+    msg = "cannot perform not_a_method with type int"
+    with pytest.raises(TypeError, match=msg):
+        arr._reduce(msg)
+
+
+def test_validate_reduction_keyword_args():
+    arr = NumpyExtensionArray(np.array([1, 2, 3]))
+    msg = "the 'keepdims' parameter is not supported .*all"
+    with pytest.raises(ValueError, match=msg):
+        arr.all(keepdims=True)
+
+
+def test_np_max_nested_tuples():
+    # case where checking in ufunc.nout works while checking for tuples
+    #  does not
+    vals = [
+        (("j", "k"), ("l", "m")),
+        (("l", "m"), ("o", "p")),
+        (("o", "p"), ("j", "k")),
+    ]
+    ser = pd.Series(vals)
+    arr = ser.array
+
+    assert arr.max() is arr[2]
+    assert ser.max() is arr[2]
+
+    result = np.maximum.reduce(arr)
+    assert result == arr[2]
+
+    result = np.maximum.reduce(ser)
+    assert result == arr[2]
+
+
+def test_np_reduce_2d():
+    raw = np.arange(12).reshape(4, 3)
+    arr = NumpyExtensionArray(raw)
+
+    res = np.maximum.reduce(arr, axis=0)
+    tm.assert_extension_array_equal(res, arr[-1])
+
+    alt = arr.max(axis=0)
+    tm.assert_extension_array_equal(alt, arr[-1])
+
+
+# ----------------------------------------------------------------------------
+# Ops
+
+
+@pytest.mark.parametrize("ufunc", [np.abs, np.negative, np.positive])
+def test_ufunc_unary(ufunc):
+    arr = NumpyExtensionArray(np.array([-1.0, 0.0, 1.0]))
+    result = ufunc(arr)
+    expected = NumpyExtensionArray(ufunc(arr._ndarray))
+    tm.assert_extension_array_equal(result, expected)
+
+    # same thing but with the 'out' keyword
+    out = NumpyExtensionArray(np.array([-9.0, -9.0, -9.0]))
+    ufunc(arr, out=out)
+    tm.assert_extension_array_equal(out, expected)
+
+
+def test_ufunc():
+    arr = NumpyExtensionArray(np.array([-1.0, 0.0, 1.0]))
+
+    r1, r2 = np.divmod(arr, np.add(arr, 2))
+    e1, e2 = np.divmod(arr._ndarray, np.add(arr._ndarray, 2))
+    e1 = NumpyExtensionArray(e1)
+    e2 = NumpyExtensionArray(e2)
+    tm.assert_extension_array_equal(r1, e1)
+    tm.assert_extension_array_equal(r2, e2)
+
+
+def test_basic_binop():
+    # Just a basic smoke test. The EA interface tests exercise this
+    # more thoroughly.
+    x = NumpyExtensionArray(np.array([1, 2, 3]))
+    result = x + x
+    expected = NumpyExtensionArray(np.array([2, 4, 6]))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [None, object])
+def test_setitem_object_typecode(dtype):
+    arr = NumpyExtensionArray(np.array(["a", "b", "c"], dtype=dtype))
+    arr[0] = "t"
+    expected = NumpyExtensionArray(np.array(["t", "b", "c"], dtype=dtype))
+    tm.assert_extension_array_equal(arr, expected)
+
+
+def test_setitem_no_coercion():
+    # https://github.com/pandas-dev/pandas/issues/28150
+    arr = NumpyExtensionArray(np.array([1, 2, 3]))
+    with pytest.raises(ValueError, match="int"):
+        arr[0] = "a"
+
+    # With a value that we do coerce, check that we coerce the value
+    #  and not the underlying array.
+    arr[0] = 2.5
+    assert isinstance(arr[0], (int, np.integer)), type(arr[0])
+
+
+def test_setitem_preserves_views():
+    # GH#28150, see also extension test of the same name
+    arr = NumpyExtensionArray(np.array([1, 2, 3]))
+    view1 = arr.view()
+    view2 = arr[:]
+    view3 = np.asarray(arr)
+
+    arr[0] = 9
+    assert view1[0] == 9
+    assert view2[0] == 9
+    assert view3[0] == 9
+
+    arr[-1] = 2.5
+    view1[-1] = 5
+    assert arr[-1] == 5
+
+
+@pytest.mark.parametrize("dtype", [np.int64, np.uint64])
+def test_quantile_empty(dtype):
+    # we should get back np.nans, not -1s
+    arr = NumpyExtensionArray(np.array([], dtype=dtype))
+    idx = pd.Index([0.0, 0.5])
+
+    result = arr._quantile(idx, interpolation="linear")
+    expected = NumpyExtensionArray(np.array([np.nan, np.nan]))
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_factorize_unsigned():
+    # don't raise when calling factorize on unsigned int NumpyExtensionArray
+    arr = np.array([1, 2, 3], dtype=np.uint64)
+    obj = NumpyExtensionArray(arr)
+
+    res_codes, res_unique = obj.factorize()
+    exp_codes, exp_unique = pd.factorize(arr)
+
+    tm.assert_numpy_array_equal(res_codes, exp_codes)
+
+    tm.assert_extension_array_equal(res_unique, NumpyExtensionArray(exp_unique))
diff --git a/pandas/pandas/tests/arrays/period/__init__.py b/pandas/pandas/tests/arrays/period/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/period/test_arrow_compat.py b/pandas/pandas/tests/arrays/period/test_arrow_compat.py
new file mode 100644
index 0000000000000..431309aca0df2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/period/test_arrow_compat.py
@@ -0,0 +1,130 @@
+import pytest
+
+from pandas.compat.pyarrow import pa_version_under10p1
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import (
+    PeriodArray,
+    period_array,
+)
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+pa = pytest.importorskip("pyarrow")
+
+
+def test_arrow_extension_type():
+    from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+    p1 = ArrowPeriodType("D")
+    p2 = ArrowPeriodType("D")
+    p3 = ArrowPeriodType("M")
+
+    assert p1.freq == "D"
+    assert p1 == p2
+    assert p1 != p3
+    assert hash(p1) == hash(p2)
+    assert hash(p1) != hash(p3)
+
+
+@pytest.mark.xfail(not pa_version_under10p1, reason="Wrong behavior with pyarrow 10")
+@pytest.mark.parametrize(
+    "data, freq",
+    [
+        (pd.date_range("2017", periods=3), "D"),
+        (pd.date_range("2017", periods=3, freq="YE"), "Y-DEC"),
+    ],
+)
+def test_arrow_array(data, freq):
+    from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+    periods = period_array(data, freq=freq)
+    result = pa.array(periods)
+    assert isinstance(result.type, ArrowPeriodType)
+    assert result.type.freq == freq
+    expected = pa.array(periods.asi8, type="int64")
+    assert result.storage.equals(expected)
+
+    # convert to its storage type
+    result = pa.array(periods, type=pa.int64())
+    assert result.equals(expected)
+
+    # unsupported conversions
+    msg = "Not supported to convert PeriodArray to 'double' type"
+    with pytest.raises(TypeError, match=msg):
+        pa.array(periods, type="float64")
+
+    with pytest.raises(TypeError, match="different 'freq'"):
+        pa.array(periods, type=ArrowPeriodType("T"))
+
+
+def test_arrow_array_missing():
+    from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+    arr = PeriodArray([1, 2, 3], dtype="period[D]")
+    arr[1] = pd.NaT
+
+    result = pa.array(arr)
+    assert isinstance(result.type, ArrowPeriodType)
+    assert result.type.freq == "D"
+    expected = pa.array([1, None, 3], type="int64")
+    assert result.storage.equals(expected)
+
+
+def test_arrow_table_roundtrip():
+    from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+    arr = PeriodArray([1, 2, 3], dtype="period[D]")
+    arr[1] = pd.NaT
+    df = pd.DataFrame({"a": arr})
+
+    table = pa.table(df)
+    assert isinstance(table.field("a").type, ArrowPeriodType)
+    result = table.to_pandas()
+    assert isinstance(result["a"].dtype, PeriodDtype)
+    tm.assert_frame_equal(result, df)
+
+    table2 = pa.concat_tables([table, table])
+    result = table2.to_pandas()
+    expected = pd.concat([df, df], ignore_index=True)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_arrow_load_from_zero_chunks():
+    # GH-41040
+
+    from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+    arr = PeriodArray([], dtype="period[D]")
+    df = pd.DataFrame({"a": arr})
+
+    table = pa.table(df)
+    assert isinstance(table.field("a").type, ArrowPeriodType)
+    table = pa.table(
+        [pa.chunked_array([], type=table.column(0).type)], schema=table.schema
+    )
+
+    result = table.to_pandas()
+    assert isinstance(result["a"].dtype, PeriodDtype)
+    tm.assert_frame_equal(result, df)
+
+
+def test_arrow_table_roundtrip_without_metadata():
+    arr = PeriodArray([1, 2, 3], dtype="period[h]")
+    arr[1] = pd.NaT
+    df = pd.DataFrame({"a": arr})
+
+    table = pa.table(df)
+    # remove the metadata
+    table = table.replace_schema_metadata()
+    assert table.schema.metadata is None
+
+    result = table.to_pandas()
+    assert isinstance(result["a"].dtype, PeriodDtype)
+    tm.assert_frame_equal(result, df)
diff --git a/pandas/pandas/tests/arrays/period/test_astype.py b/pandas/pandas/tests/arrays/period/test_astype.py
new file mode 100644
index 0000000000000..9976c3a32580d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/period/test_astype.py
@@ -0,0 +1,67 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import period_array
+
+
+@pytest.mark.parametrize("dtype", [int, np.int32, np.int64, "uint32", "uint64"])
+def test_astype_int(dtype):
+    # We choose to ignore the sign and size of integers for
+    # Period/Datetime/Timedelta astype
+    arr = period_array(["2000", "2001", None], freq="D")
+
+    if np.dtype(dtype) != np.int64:
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype(dtype)
+        return
+
+    result = arr.astype(dtype)
+    expected = arr._ndarray.view("i8")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_astype_copies():
+    arr = period_array(["2000", "2001", None], freq="D")
+    result = arr.astype(np.int64, copy=False)
+
+    # Add the `.base`, since we now use `.asi8` which returns a view.
+    # We could maybe override it in PeriodArray to return ._ndarray directly.
+    assert result.base is arr._ndarray
+
+    result = arr.astype(np.int64, copy=True)
+    assert result is not arr._ndarray
+    tm.assert_numpy_array_equal(result, arr._ndarray.view("i8"))
+
+
+def test_astype_categorical():
+    arr = period_array(["2000", "2001", "2001", None], freq="D")
+    result = arr.astype("category")
+    categories = pd.PeriodIndex(["2000", "2001"], freq="D")
+    expected = pd.Categorical.from_codes([0, 1, 1, -1], categories=categories)
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_astype_period():
+    arr = period_array(["2000", "2001", None], freq="D")
+    result = arr.astype(PeriodDtype("M"))
+    expected = period_array(["2000", "2001", None], freq="M")
+    tm.assert_period_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["datetime64[ns]", "timedelta64[ns]"])
+def test_astype_datetime(dtype):
+    arr = period_array(["2000", "2001", None], freq="D")
+    # slice off the [ns] so that the regex matches.
+    if dtype == "timedelta64[ns]":
+        with pytest.raises(TypeError, match=dtype[:-4]):
+            arr.astype(dtype)
+
+    else:
+        # GH#45038 allow period->dt64 because we allow dt64->period
+        result = arr.astype(dtype)
+        expected = pd.DatetimeIndex(["2000", "2001", pd.NaT], dtype=dtype)._data
+        tm.assert_datetime_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/period/test_constructors.py b/pandas/pandas/tests/arrays/period/test_constructors.py
new file mode 100644
index 0000000000000..63b0e456c4566
--- /dev/null
+++ b/pandas/pandas/tests/arrays/period/test_constructors.py
@@ -0,0 +1,145 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+from pandas._libs.tslibs.offsets import MonthEnd
+from pandas._libs.tslibs.period import IncompatibleFrequency
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import (
+    PeriodArray,
+    period_array,
+)
+
+
+@pytest.mark.parametrize(
+    "data, freq, expected",
+    [
+        ([pd.Period("2017", "D")], None, [17167]),
+        ([pd.Period("2017", "D")], "D", [17167]),
+        ([2017], "D", [17167]),
+        (["2017"], "D", [17167]),
+        ([pd.Period("2017", "D")], pd.tseries.offsets.Day(), [17167]),
+        ([pd.Period("2017", "D"), None], None, [17167, iNaT]),
+        (pd.Series(pd.date_range("2017", periods=3)), None, [17167, 17168, 17169]),
+        (pd.date_range("2017", periods=3), None, [17167, 17168, 17169]),
+        (pd.period_range("2017", periods=4, freq="Q"), None, [188, 189, 190, 191]),
+    ],
+)
+def test_period_array_ok(data, freq, expected):
+    result = period_array(data, freq=freq).asi8
+    expected = np.asarray(expected, dtype=np.int64)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_period_array_readonly_object():
+    # https://github.com/pandas-dev/pandas/issues/25403
+    pa = period_array([pd.Period("2019-01-01")])
+    arr = np.asarray(pa, dtype="object")
+    arr.setflags(write=False)
+
+    result = period_array(arr)
+    tm.assert_period_array_equal(result, pa)
+
+    result = pd.Series(arr)
+    tm.assert_series_equal(result, pd.Series(pa))
+
+    result = pd.DataFrame({"A": arr})
+    tm.assert_frame_equal(result, pd.DataFrame({"A": pa}))
+
+
+def test_from_datetime64_freq_changes():
+    # https://github.com/pandas-dev/pandas/issues/23438
+    arr = pd.date_range("2017", periods=3, freq="D")
+    result = PeriodArray._from_datetime64(arr, freq="M")
+    expected = period_array(["2017-01-01", "2017-01-01", "2017-01-01"], freq="M")
+    tm.assert_period_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("freq", ["2M", MonthEnd(2)])
+def test_from_datetime64_freq_2M(freq):
+    arr = np.array(
+        ["2020-01-01T00:00:00", "2020-01-02T00:00:00"], dtype="datetime64[ns]"
+    )
+    result = PeriodArray._from_datetime64(arr, freq)
+    expected = period_array(["2020-01", "2020-01"], freq=freq)
+    tm.assert_period_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, freq, msg",
+    [
+        (
+            [pd.Period("2017", "D"), pd.Period("2017", "Y")],
+            None,
+            "Input has different freq",
+        ),
+        ([pd.Period("2017", "D")], "Y", "Input has different freq"),
+    ],
+)
+def test_period_array_raises(data, freq, msg):
+    with pytest.raises(IncompatibleFrequency, match=msg):
+        period_array(data, freq)
+
+
+def test_period_array_non_period_series_raies():
+    ser = pd.Series([1, 2, 3])
+    with pytest.raises(TypeError, match="dtype"):
+        PeriodArray(ser, dtype="period[D]")
+
+
+def test_period_array_freq_mismatch():
+    arr = period_array(["2000", "2001"], freq="D")
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        PeriodArray(arr, dtype="period[M]")
+
+    dtype = pd.PeriodDtype(pd.tseries.offsets.MonthEnd())
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        PeriodArray(arr, dtype=dtype)
+
+
+def test_from_sequence_disallows_i8():
+    arr = period_array(["2000", "2001"], freq="D")
+
+    msg = str(arr[0].ordinal)
+    with pytest.raises(TypeError, match=msg):
+        PeriodArray._from_sequence(arr.asi8, dtype=arr.dtype)
+
+    with pytest.raises(TypeError, match=msg):
+        PeriodArray._from_sequence(list(arr.asi8), dtype=arr.dtype)
+
+
+def test_from_td64nat_sequence_raises():
+    # GH#44507
+    td = pd.NaT.to_numpy("m8[ns]")
+
+    dtype = pd.period_range("2005-01-01", periods=3, freq="D").dtype
+
+    arr = np.array([None], dtype=object)
+    arr[0] = td
+
+    msg = "Value must be Period, string, integer, or datetime"
+    with pytest.raises(ValueError, match=msg):
+        PeriodArray._from_sequence(arr, dtype=dtype)
+
+    with pytest.raises(ValueError, match=msg):
+        pd.PeriodIndex(arr, dtype=dtype)
+    with pytest.raises(ValueError, match=msg):
+        pd.Index(arr, dtype=dtype)
+    with pytest.raises(ValueError, match=msg):
+        pd.array(arr, dtype=dtype)
+    with pytest.raises(ValueError, match=msg):
+        pd.Series(arr, dtype=dtype)
+    with pytest.raises(ValueError, match=msg):
+        pd.DataFrame(arr, dtype=dtype)
+
+
+def test_period_array_from_datetime64():
+    arr = np.array(
+        ["2020-01-01T00:00:00", "2020-02-02T00:00:00"], dtype="datetime64[ns]"
+    )
+    result = PeriodArray._from_datetime64(arr, freq=MonthEnd(2))
+
+    expected = period_array(["2020-01-01", "2020-02-01"], freq=MonthEnd(2))
+    tm.assert_period_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/period/test_reductions.py b/pandas/pandas/tests/arrays/period/test_reductions.py
new file mode 100644
index 0000000000000..5b859d86eb6d0
--- /dev/null
+++ b/pandas/pandas/tests/arrays/period/test_reductions.py
@@ -0,0 +1,39 @@
+import pandas as pd
+from pandas.core.arrays import period_array
+
+
+class TestReductions:
+    def test_min_max(self):
+        arr = period_array(
+            [
+                "2000-01-03",
+                "2000-01-03",
+                "NaT",
+                "2000-01-02",
+                "2000-01-05",
+                "2000-01-04",
+            ],
+            freq="D",
+        )
+
+        result = arr.min()
+        expected = pd.Period("2000-01-02", freq="D")
+        assert result == expected
+
+        result = arr.max()
+        expected = pd.Period("2000-01-05", freq="D")
+        assert result == expected
+
+        result = arr.min(skipna=False)
+        assert result is pd.NaT
+
+        result = arr.max(skipna=False)
+        assert result is pd.NaT
+
+    def test_min_max_empty(self, skipna):
+        arr = period_array([], freq="D")
+        result = arr.min(skipna=skipna)
+        assert result is pd.NaT
+
+        result = arr.max(skipna=skipna)
+        assert result is pd.NaT
diff --git a/pandas/pandas/tests/arrays/sparse/__init__.py b/pandas/pandas/tests/arrays/sparse/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/sparse/test_accessor.py b/pandas/pandas/tests/arrays/sparse/test_accessor.py
new file mode 100644
index 0000000000000..08bfd5b69fdd9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_accessor.py
@@ -0,0 +1,258 @@
+import string
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import SparseDtype
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+class TestSeriesAccessor:
+    def test_to_dense(self):
+        ser = pd.Series([0, 1, 0, 10], dtype="Sparse[int64]")
+        result = ser.sparse.to_dense()
+        expected = pd.Series([0, 1, 0, 10])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("attr", ["npoints", "density", "fill_value", "sp_values"])
+    def test_get_attributes(self, attr):
+        arr = SparseArray([0, 1])
+        ser = pd.Series(arr)
+
+        result = getattr(ser.sparse, attr)
+        expected = getattr(arr, attr)
+        assert result == expected
+
+    def test_from_coo(self):
+        scipy_sparse = pytest.importorskip("scipy.sparse")
+
+        row = [0, 3, 1, 0]
+        col = [0, 3, 1, 2]
+        data = [4, 5, 7, 9]
+
+        sp_array = scipy_sparse.coo_matrix((data, (row, col)))
+        result = pd.Series.sparse.from_coo(sp_array)
+
+        index = pd.MultiIndex.from_arrays(
+            [
+                np.array([0, 0, 1, 3], dtype=np.int32),
+                np.array([0, 2, 1, 3], dtype=np.int32),
+            ],
+        )
+        expected = pd.Series([4, 9, 7, 5], index=index, dtype="Sparse[int]")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "sort_labels, expected_rows, expected_cols, expected_values_pos",
+        [
+            (
+                False,
+                [("b", 2), ("a", 2), ("b", 1), ("a", 1)],
+                [("z", 1), ("z", 2), ("x", 2), ("z", 0)],
+                {1: (1, 0), 3: (3, 3)},
+            ),
+            (
+                True,
+                [("a", 1), ("a", 2), ("b", 1), ("b", 2)],
+                [("x", 2), ("z", 0), ("z", 1), ("z", 2)],
+                {1: (1, 2), 3: (0, 1)},
+            ),
+        ],
+    )
+    def test_to_coo(
+        self, sort_labels, expected_rows, expected_cols, expected_values_pos
+    ):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        values = SparseArray([0, np.nan, 1, 0, None, 3], fill_value=0)
+        index = pd.MultiIndex.from_tuples(
+            [
+                ("b", 2, "z", 1),
+                ("a", 2, "z", 2),
+                ("a", 2, "z", 1),
+                ("a", 2, "x", 2),
+                ("b", 1, "z", 1),
+                ("a", 1, "z", 0),
+            ]
+        )
+        ss = pd.Series(values, index=index)
+
+        expected_A = np.zeros((4, 4))
+        for value, (row, col) in expected_values_pos.items():
+            expected_A[row, col] = value
+
+        A, rows, cols = ss.sparse.to_coo(
+            row_levels=(0, 1), column_levels=(2, 3), sort_labels=sort_labels
+        )
+        assert isinstance(A, sp_sparse.coo_matrix)
+        tm.assert_numpy_array_equal(A.toarray(), expected_A)
+        assert rows == expected_rows
+        assert cols == expected_cols
+
+    def test_non_sparse_raises(self):
+        ser = pd.Series([1, 2, 3])
+        with pytest.raises(AttributeError, match=".sparse"):
+            ser.sparse.density
+
+
+class TestFrameAccessor:
+    def test_accessor_raises(self):
+        df = pd.DataFrame({"A": [0, 1]})
+        with pytest.raises(AttributeError, match="sparse"):
+            df.sparse
+
+    @pytest.mark.parametrize("format", ["csc", "csr", "coo"])
+    @pytest.mark.parametrize("labels", [None, list(string.ascii_letters[:10])])
+    @pytest.mark.parametrize("dtype", [np.complex128, np.float64, np.int64, bool])
+    def test_from_spmatrix(self, format, labels, dtype):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        sp_dtype = SparseDtype(dtype)
+
+        sp_mat = sp_sparse.eye(10, format=format, dtype=dtype)
+        result = pd.DataFrame.sparse.from_spmatrix(sp_mat, index=labels, columns=labels)
+        mat = np.eye(10, dtype=dtype)
+        expected = pd.DataFrame(
+            np.ma.array(mat, mask=(mat == 0)).filled(sp_dtype.fill_value),
+            index=labels,
+            columns=labels,
+        ).astype(sp_dtype)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("format", ["csc", "csr", "coo"])
+    @pytest.mark.parametrize("dtype", [np.int64, bool])
+    def test_from_spmatrix_including_explicit_zero(self, format, dtype):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        sp_dtype = SparseDtype(dtype)
+
+        sp_mat = sp_sparse.random(10, 2, density=0.5, format=format, dtype=dtype)
+        sp_mat.data[0] = 0
+        result = pd.DataFrame.sparse.from_spmatrix(sp_mat)
+        mat = sp_mat.toarray()
+        expected = pd.DataFrame(
+            np.ma.array(mat, mask=(mat == 0)).filled(sp_dtype.fill_value)
+        ).astype(sp_dtype)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "columns",
+        [["a", "b"], pd.MultiIndex.from_product([["A"], ["a", "b"]]), ["a", "a"]],
+    )
+    def test_from_spmatrix_columns(self, columns):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        sp_dtype = SparseDtype(np.float64)
+
+        sp_mat = sp_sparse.random(10, 2, density=0.5)
+        result = pd.DataFrame.sparse.from_spmatrix(sp_mat, columns=columns)
+        mat = sp_mat.toarray()
+        expected = pd.DataFrame(
+            np.ma.array(mat, mask=(mat == 0)).filled(sp_dtype.fill_value),
+            columns=columns,
+        ).astype(sp_dtype)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "columns", [("A", "B"), (1, 2), (1, pd.NA), (0.1, 0.2), ("x", "x"), (0, 0)]
+    )
+    @pytest.mark.parametrize("dtype", [np.complex128, np.float64, np.int64, bool])
+    def test_to_coo(self, columns, dtype):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        sp_dtype = SparseDtype(dtype)
+
+        expected = sp_sparse.random(10, 2, density=0.5, format="coo", dtype=dtype)
+        mat = expected.toarray()
+        result = pd.DataFrame(
+            np.ma.array(mat, mask=(mat == 0)).filled(sp_dtype.fill_value),
+            columns=columns,
+            dtype=sp_dtype,
+        ).sparse.to_coo()
+        assert (result != expected).nnz == 0
+
+    def test_to_coo_midx_categorical(self):
+        # GH#50996
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        midx = pd.MultiIndex.from_arrays(
+            [
+                pd.CategoricalIndex(list("ab"), name="x"),
+                pd.CategoricalIndex([0, 1], name="y"),
+            ]
+        )
+
+        ser = pd.Series(1, index=midx, dtype="Sparse[int]")
+        result = ser.sparse.to_coo(row_levels=["x"], column_levels=["y"])[0]
+        expected = sp_sparse.coo_matrix(
+            (np.array([1, 1]), (np.array([0, 1]), np.array([0, 1]))), shape=(2, 2)
+        )
+        assert (result != expected).nnz == 0
+
+    def test_to_dense(self):
+        df = pd.DataFrame(
+            {
+                "A": SparseArray([1, 0], dtype=SparseDtype("int64", 0)),
+                "B": SparseArray([1, 0], dtype=SparseDtype("int64", 1)),
+                "C": SparseArray([1.0, 0.0], dtype=SparseDtype("float64", 0.0)),
+            },
+            index=["b", "a"],
+        )
+        result = df.sparse.to_dense()
+        expected = pd.DataFrame(
+            {"A": [1, 0], "B": [1, 0], "C": [1.0, 0.0]}, index=["b", "a"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_density(self):
+        df = pd.DataFrame(
+            {
+                "A": SparseArray([1, 0, 2, 1], fill_value=0),
+                "B": SparseArray([0, 1, 1, 1], fill_value=0),
+            }
+        )
+        res = df.sparse.density
+        expected = 0.75
+        assert res == expected
+
+    @pytest.mark.parametrize("dtype", ["int64", "float64"])
+    @pytest.mark.parametrize("dense_index", [True, False])
+    def test_series_from_coo(self, dtype, dense_index):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        A = sp_sparse.eye(3, format="coo", dtype=dtype)
+        result = pd.Series.sparse.from_coo(A, dense_index=dense_index)
+
+        index = pd.MultiIndex.from_tuples(
+            [
+                np.array([0, 0], dtype=np.int32),
+                np.array([1, 1], dtype=np.int32),
+                np.array([2, 2], dtype=np.int32),
+            ],
+        )
+        expected = pd.Series(SparseArray(np.array([1, 1, 1], dtype=dtype)), index=index)
+        if dense_index:
+            expected = expected.reindex(pd.MultiIndex.from_product(index.levels))
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_from_coo_incorrect_format_raises(self):
+        # gh-26554
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        m = sp_sparse.csr_matrix(np.array([[0, 1], [0, 0]]))
+        with pytest.raises(
+            TypeError, match="Expected coo_matrix. Got csr_matrix instead."
+        ):
+            pd.Series.sparse.from_coo(m)
+
+    def test_with_column_named_sparse(self):
+        # https://github.com/pandas-dev/pandas/issues/30758
+        df = pd.DataFrame({"sparse": pd.arrays.SparseArray([1, 2])})
+        assert isinstance(df.sparse, pd.core.arrays.sparse.accessor.SparseFrameAccessor)
+
+    def test_subclassing(self):
+        df = tm.SubclassedDataFrame({"sparse": pd.arrays.SparseArray([1, 2])})
+        assert isinstance(df.sparse.to_dense(), tm.SubclassedDataFrame)
diff --git a/pandas/pandas/tests/arrays/sparse/test_arithmetics.py b/pandas/pandas/tests/arrays/sparse/test_arithmetics.py
new file mode 100644
index 0000000000000..a47e73d49674d
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_arithmetics.py
@@ -0,0 +1,524 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import SparseDtype
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+@pytest.fixture(params=["integer", "block"])
+def kind(request):
+    """kind kwarg to pass to SparseArray"""
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def mix(request):
+    """
+    Fixture returning True or False, determining whether to operate
+    op(sparse, dense) instead of op(sparse, sparse)
+    """
+    return request.param
+
+
+class TestSparseArrayArithmetics:
+    def _assert(self, a, b):
+        # We have to use tm.assert_sp_array_equal. See GH #45126
+        tm.assert_numpy_array_equal(a, b)
+
+    def _check_numeric_ops(self, a, b, a_dense, b_dense, mix: bool, op):
+        # Check that arithmetic behavior matches non-Sparse Series arithmetic
+
+        if isinstance(a_dense, np.ndarray):
+            expected = op(pd.Series(a_dense), b_dense).values
+        elif isinstance(b_dense, np.ndarray):
+            expected = op(a_dense, pd.Series(b_dense)).values
+        else:
+            raise NotImplementedError
+
+        with np.errstate(invalid="ignore", divide="ignore"):
+            if mix:
+                result = op(a, b_dense).to_dense()
+            else:
+                result = op(a, b).to_dense()
+
+        self._assert(result, expected)
+
+    def _check_bool_result(self, res):
+        assert isinstance(res, SparseArray)
+        assert isinstance(res.dtype, SparseDtype)
+        assert res.dtype.subtype == np.bool_
+        assert isinstance(res.fill_value, bool)
+
+    def _check_comparison_ops(self, a, b, a_dense, b_dense):
+        with np.errstate(invalid="ignore"):
+            # Unfortunately, trying to wrap the computation of each expected
+            # value is with np.errstate() is too tedious.
+            #
+            # sparse & sparse
+            self._check_bool_result(a == b)
+            self._assert((a == b).to_dense(), a_dense == b_dense)
+
+            self._check_bool_result(a != b)
+            self._assert((a != b).to_dense(), a_dense != b_dense)
+
+            self._check_bool_result(a >= b)
+            self._assert((a >= b).to_dense(), a_dense >= b_dense)
+
+            self._check_bool_result(a <= b)
+            self._assert((a <= b).to_dense(), a_dense <= b_dense)
+
+            self._check_bool_result(a > b)
+            self._assert((a > b).to_dense(), a_dense > b_dense)
+
+            self._check_bool_result(a < b)
+            self._assert((a < b).to_dense(), a_dense < b_dense)
+
+            # sparse & dense
+            self._check_bool_result(a == b_dense)
+            self._assert((a == b_dense).to_dense(), a_dense == b_dense)
+
+            self._check_bool_result(a != b_dense)
+            self._assert((a != b_dense).to_dense(), a_dense != b_dense)
+
+            self._check_bool_result(a >= b_dense)
+            self._assert((a >= b_dense).to_dense(), a_dense >= b_dense)
+
+            self._check_bool_result(a <= b_dense)
+            self._assert((a <= b_dense).to_dense(), a_dense <= b_dense)
+
+            self._check_bool_result(a > b_dense)
+            self._assert((a > b_dense).to_dense(), a_dense > b_dense)
+
+            self._check_bool_result(a < b_dense)
+            self._assert((a < b_dense).to_dense(), a_dense < b_dense)
+
+    def _check_logical_ops(self, a, b, a_dense, b_dense):
+        # sparse & sparse
+        self._check_bool_result(a & b)
+        self._assert((a & b).to_dense(), a_dense & b_dense)
+
+        self._check_bool_result(a | b)
+        self._assert((a | b).to_dense(), a_dense | b_dense)
+        # sparse & dense
+        self._check_bool_result(a & b_dense)
+        self._assert((a & b_dense).to_dense(), a_dense & b_dense)
+
+        self._check_bool_result(a | b_dense)
+        self._assert((a | b_dense).to_dense(), a_dense | b_dense)
+
+    @pytest.mark.parametrize("scalar", [0, 1, 3])
+    @pytest.mark.parametrize("fill_value", [None, 0, 2])
+    def test_float_scalar(
+        self, kind, mix, all_arithmetic_functions, fill_value, scalar, request
+    ):
+        op = all_arithmetic_functions
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        a = SparseArray(values, kind=kind, fill_value=fill_value)
+        self._check_numeric_ops(a, scalar, values, scalar, mix, op)
+
+    def test_float_scalar_comparison(self, kind):
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+
+        a = SparseArray(values, kind=kind)
+        self._check_comparison_ops(a, 1, values, 1)
+        self._check_comparison_ops(a, 0, values, 0)
+        self._check_comparison_ops(a, 3, values, 3)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        self._check_comparison_ops(a, 1, values, 1)
+        self._check_comparison_ops(a, 0, values, 0)
+        self._check_comparison_ops(a, 3, values, 3)
+
+        a = SparseArray(values, kind=kind, fill_value=2)
+        self._check_comparison_ops(a, 1, values, 1)
+        self._check_comparison_ops(a, 0, values, 0)
+        self._check_comparison_ops(a, 3, values, 3)
+
+    def test_float_same_index_without_nans(self, kind, mix, all_arithmetic_functions):
+        # when sp_index are the same
+        op = all_arithmetic_functions
+
+        values = np.array([0.0, 1.0, 2.0, 6.0, 0.0, 0.0, 1.0, 2.0, 1.0, 0.0])
+        rvalues = np.array([0.0, 2.0, 3.0, 4.0, 0.0, 0.0, 1.0, 3.0, 2.0, 0.0])
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_float_same_index_with_nans(
+        self, kind, mix, all_arithmetic_functions, request
+    ):
+        # when sp_index are the same
+        op = all_arithmetic_functions
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([np.nan, 2, 3, 4, np.nan, 0, 1, 3, 2, np.nan])
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_float_same_index_comparison(self, kind):
+        # when sp_index are the same
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([np.nan, 2, 3, 4, np.nan, 0, 1, 3, 2, np.nan])
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        values = np.array([0.0, 1.0, 2.0, 6.0, 0.0, 0.0, 1.0, 2.0, 1.0, 0.0])
+        rvalues = np.array([0.0, 2.0, 3.0, 4.0, 0.0, 0.0, 1.0, 3.0, 2.0, 0.0])
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+    def test_float_array(self, kind, mix, all_arithmetic_functions):
+        op = all_arithmetic_functions
+
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([2, np.nan, 2, 3, np.nan, 0, 1, 5, 2, np.nan])
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+        self._check_numeric_ops(a, b * 0, values, rvalues * 0, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=1)
+        b = SparseArray(rvalues, kind=kind, fill_value=2)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_float_array_different_kind(self, mix, all_arithmetic_functions):
+        op = all_arithmetic_functions
+
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([2, np.nan, 2, 3, np.nan, 0, 1, 5, 2, np.nan])
+
+        a = SparseArray(values, kind="integer")
+        b = SparseArray(rvalues, kind="block")
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+        self._check_numeric_ops(a, b * 0, values, rvalues * 0, mix, op)
+
+        a = SparseArray(values, kind="integer", fill_value=0)
+        b = SparseArray(rvalues, kind="block")
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind="integer", fill_value=0)
+        b = SparseArray(rvalues, kind="block", fill_value=0)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind="integer", fill_value=1)
+        b = SparseArray(rvalues, kind="block", fill_value=2)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_float_array_comparison(self, kind):
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([2, np.nan, 2, 3, np.nan, 0, 1, 5, 2, np.nan])
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_comparison_ops(a, b, values, rvalues)
+        self._check_comparison_ops(a, b * 0, values, rvalues * 0)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, kind=kind, fill_value=1)
+        b = SparseArray(rvalues, kind=kind, fill_value=2)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+    def test_int_array(self, kind, mix, all_arithmetic_functions):
+        op = all_arithmetic_functions
+
+        # have to specify dtype explicitly until fixing GH 667
+        dtype = np.int64
+
+        values = np.array([0, 1, 2, 0, 0, 0, 1, 2, 1, 0], dtype=dtype)
+        rvalues = np.array([2, 0, 2, 3, 0, 0, 1, 5, 2, 0], dtype=dtype)
+
+        a = SparseArray(values, dtype=dtype, kind=kind)
+        assert a.dtype == SparseDtype(dtype)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind)
+        assert b.dtype == SparseDtype(dtype)
+
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+        self._check_numeric_ops(a, b * 0, values, rvalues * 0, mix, op)
+
+        a = SparseArray(values, fill_value=0, dtype=dtype, kind=kind)
+        assert a.dtype == SparseDtype(dtype)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind)
+        assert b.dtype == SparseDtype(dtype)
+
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, fill_value=0, dtype=dtype, kind=kind)
+        assert a.dtype == SparseDtype(dtype)
+        b = SparseArray(rvalues, fill_value=0, dtype=dtype, kind=kind)
+        assert b.dtype == SparseDtype(dtype)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, fill_value=1, dtype=dtype, kind=kind)
+        assert a.dtype == SparseDtype(dtype, fill_value=1)
+        b = SparseArray(rvalues, fill_value=2, dtype=dtype, kind=kind)
+        assert b.dtype == SparseDtype(dtype, fill_value=2)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_int_array_comparison(self, kind):
+        dtype = "int64"
+        # int32 NI ATM
+
+        values = np.array([0, 1, 2, 0, 0, 0, 1, 2, 1, 0], dtype=dtype)
+        rvalues = np.array([2, 0, 2, 3, 0, 0, 1, 5, 2, 0], dtype=dtype)
+
+        a = SparseArray(values, dtype=dtype, kind=kind)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind)
+        self._check_comparison_ops(a, b, values, rvalues)
+        self._check_comparison_ops(a, b * 0, values, rvalues * 0)
+
+        a = SparseArray(values, dtype=dtype, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, dtype=dtype, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind, fill_value=0)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, dtype=dtype, kind=kind, fill_value=1)
+        b = SparseArray(rvalues, dtype=dtype, kind=kind, fill_value=2)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+    @pytest.mark.parametrize("fill_value", [True, False, np.nan])
+    def test_bool_same_index(self, kind, fill_value):
+        # GH 14000
+        # when sp_index are the same
+        values = np.array([True, False, True, True], dtype=np.bool_)
+        rvalues = np.array([True, False, True, True], dtype=np.bool_)
+
+        a = SparseArray(values, kind=kind, dtype=np.bool_, fill_value=fill_value)
+        b = SparseArray(rvalues, kind=kind, dtype=np.bool_, fill_value=fill_value)
+        self._check_logical_ops(a, b, values, rvalues)
+
+    @pytest.mark.parametrize("fill_value", [True, False, np.nan])
+    def test_bool_array_logical(self, kind, fill_value):
+        # GH 14000
+        # when sp_index are the same
+        values = np.array([True, False, True, False, True, True], dtype=np.bool_)
+        rvalues = np.array([True, False, False, True, False, True], dtype=np.bool_)
+
+        a = SparseArray(values, kind=kind, dtype=np.bool_, fill_value=fill_value)
+        b = SparseArray(rvalues, kind=kind, dtype=np.bool_, fill_value=fill_value)
+        self._check_logical_ops(a, b, values, rvalues)
+
+    def test_mixed_array_float_int(self, kind, mix, all_arithmetic_functions, request):
+        op = all_arithmetic_functions
+        rdtype = "int64"
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([2, 0, 2, 3, 0, 0, 1, 5, 2, 0], dtype=rdtype)
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        assert b.dtype == SparseDtype(rdtype)
+
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+        self._check_numeric_ops(a, b * 0, values, rvalues * 0, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind)
+        assert b.dtype == SparseDtype(rdtype)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        assert b.dtype == SparseDtype(rdtype)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+        a = SparseArray(values, kind=kind, fill_value=1)
+        b = SparseArray(rvalues, kind=kind, fill_value=2)
+        assert b.dtype == SparseDtype(rdtype, fill_value=2)
+        self._check_numeric_ops(a, b, values, rvalues, mix, op)
+
+    def test_mixed_array_comparison(self, kind):
+        rdtype = "int64"
+        # int32 NI ATM
+
+        values = np.array([np.nan, 1, 2, 0, np.nan, 0, 1, 2, 1, np.nan])
+        rvalues = np.array([2, 0, 2, 3, 0, 0, 1, 5, 2, 0], dtype=rdtype)
+
+        a = SparseArray(values, kind=kind)
+        b = SparseArray(rvalues, kind=kind)
+        assert b.dtype == SparseDtype(rdtype)
+
+        self._check_comparison_ops(a, b, values, rvalues)
+        self._check_comparison_ops(a, b * 0, values, rvalues * 0)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind)
+        assert b.dtype == SparseDtype(rdtype)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, kind=kind, fill_value=0)
+        b = SparseArray(rvalues, kind=kind, fill_value=0)
+        assert b.dtype == SparseDtype(rdtype)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+        a = SparseArray(values, kind=kind, fill_value=1)
+        b = SparseArray(rvalues, kind=kind, fill_value=2)
+        assert b.dtype == SparseDtype(rdtype, fill_value=2)
+        self._check_comparison_ops(a, b, values, rvalues)
+
+    def test_xor(self):
+        s = SparseArray([True, True, False, False])
+        t = SparseArray([True, False, True, False])
+        result = s ^ t
+        sp_index = pd.core.arrays.sparse.IntIndex(4, np.array([0, 1, 2], dtype="int32"))
+        expected = SparseArray([False, True, True], sparse_index=sp_index)
+        tm.assert_sp_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", [operator.eq, operator.add])
+def test_with_list(op):
+    arr = SparseArray([0, 1], fill_value=0)
+    result = op(arr, [0, 1])
+    expected = op(arr, SparseArray([0, 1]))
+    tm.assert_sp_array_equal(result, expected)
+
+
+def test_with_dataframe():
+    # GH#27910
+    arr = SparseArray([0, 1], fill_value=0)
+    df = pd.DataFrame([[1, 2], [3, 4]])
+    result = arr.__add__(df)
+    assert result is NotImplemented
+
+
+def test_with_zerodim_ndarray():
+    # GH#27910
+    arr = SparseArray([0, 1], fill_value=0)
+
+    result = arr * np.array(2)
+    expected = arr * 2
+    tm.assert_sp_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.abs, np.exp])
+@pytest.mark.parametrize(
+    "arr", [SparseArray([0, 0, -1, 1]), SparseArray([None, None, -1, 1])]
+)
+def test_ufuncs(ufunc, arr):
+    result = ufunc(arr)
+    fill_value = ufunc(arr.fill_value)
+    expected = SparseArray(ufunc(np.asarray(arr)), fill_value=fill_value)
+    tm.assert_sp_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    [
+        (SparseArray([0, 0, 0]), np.array([0, 1, 2])),
+        (SparseArray([0, 0, 0], fill_value=1), np.array([0, 1, 2])),
+    ],
+)
+@pytest.mark.parametrize("ufunc", [np.add, np.greater])
+def test_binary_ufuncs(ufunc, a, b):
+    # can't say anything about fill value here.
+    result = ufunc(a, b)
+    expected = ufunc(np.asarray(a), np.asarray(b))
+    assert isinstance(result, SparseArray)
+    tm.assert_numpy_array_equal(np.asarray(result), expected)
+
+
+def test_ndarray_inplace():
+    sparray = SparseArray([0, 2, 0, 0])
+    ndarray = np.array([0, 1, 2, 3])
+    ndarray += sparray
+    expected = np.array([0, 3, 2, 3])
+    tm.assert_numpy_array_equal(ndarray, expected)
+
+
+def test_sparray_inplace():
+    sparray = SparseArray([0, 2, 0, 0])
+    ndarray = np.array([0, 1, 2, 3])
+    sparray += ndarray
+    expected = SparseArray([0, 3, 2, 3], fill_value=0)
+    tm.assert_sp_array_equal(sparray, expected)
+
+
+@pytest.mark.parametrize("cons", [list, np.array, SparseArray])
+def test_mismatched_length_cmp_op(cons):
+    left = SparseArray([True, True])
+    right = cons([True, True, True])
+    with pytest.raises(ValueError, match="operands have mismatched length"):
+        left & right
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    [
+        ([0, 1, 2], [0, 1, 2, 3]),
+        ([0, 1, 2, 3], [0, 1, 2]),
+    ],
+)
+def test_mismatched_length_arith_op(a, b, all_arithmetic_functions):
+    op = all_arithmetic_functions
+    with pytest.raises(AssertionError, match=f"length mismatch: {len(a)} vs. {len(b)}"):
+        op(SparseArray(a, fill_value=0), np.array(b))
+
+
+@pytest.mark.parametrize("op", ["add", "sub", "mul", "truediv", "floordiv", "pow"])
+@pytest.mark.parametrize("fill_value", [np.nan, 3])
+def test_binary_operators(op, fill_value):
+    op = getattr(operator, op)
+    data1 = np.random.default_rng(2).standard_normal(20)
+    data2 = np.random.default_rng(2).standard_normal(20)
+
+    data1[::2] = fill_value
+    data2[::3] = fill_value
+
+    first = SparseArray(data1, fill_value=fill_value)
+    second = SparseArray(data2, fill_value=fill_value)
+
+    with np.errstate(all="ignore"):
+        res = op(first, second)
+        exp = SparseArray(
+            op(first.to_dense(), second.to_dense()), fill_value=first.fill_value
+        )
+        assert isinstance(res, SparseArray)
+        tm.assert_almost_equal(res.to_dense(), exp.to_dense())
+
+        res2 = op(first, second.to_dense())
+        assert isinstance(res2, SparseArray)
+        tm.assert_sp_array_equal(res, res2)
+
+        res3 = op(first.to_dense(), second)
+        assert isinstance(res3, SparseArray)
+        tm.assert_sp_array_equal(res, res3)
+
+        res4 = op(first, 4)
+        assert isinstance(res4, SparseArray)
+
+        # Ignore this if the actual op raises (e.g. pow).
+        try:
+            exp = op(first.to_dense(), 4)
+            exp_fv = op(first.fill_value, 4)
+        except ValueError:
+            pass
+        else:
+            tm.assert_almost_equal(res4.fill_value, exp_fv)
+            tm.assert_almost_equal(res4.to_dense(), exp)
diff --git a/pandas/pandas/tests/arrays/sparse/test_array.py b/pandas/pandas/tests/arrays/sparse/test_array.py
new file mode 100644
index 0000000000000..1b685100e4931
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_array.py
@@ -0,0 +1,513 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.sparse import IntIndex
+from pandas.compat.numpy import np_version_gt2
+
+import pandas as pd
+from pandas import (
+    SparseDtype,
+    isna,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+@pytest.fixture
+def arr_data():
+    """Fixture returning numpy array with valid and missing entries"""
+    return np.array([np.nan, np.nan, 1, 2, 3, np.nan, 4, 5, np.nan, 6])
+
+
+@pytest.fixture
+def arr(arr_data):
+    """Fixture returning SparseArray from 'arr_data'"""
+    return SparseArray(arr_data)
+
+
+@pytest.fixture
+def zarr():
+    """Fixture returning SparseArray with integer entries and 'fill_value=0'"""
+    return SparseArray([0, 0, 1, 2, 3, 0, 4, 5, 0, 6], fill_value=0)
+
+
+class TestSparseArray:
+    @pytest.mark.parametrize("fill_value", [0, None, np.nan])
+    def test_shift_fill_value(self, fill_value):
+        # GH #24128
+        sparse = SparseArray(np.array([1, 0, 0, 3, 0]), fill_value=8.0)
+        res = sparse.shift(1, fill_value=fill_value)
+        if isna(fill_value):
+            fill_value = res.dtype.na_value
+        exp = SparseArray(np.array([fill_value, 1, 0, 0, 3]), fill_value=8.0)
+        tm.assert_sp_array_equal(res, exp)
+
+    def test_set_fill_value(self):
+        arr = SparseArray([1.0, np.nan, 2.0], fill_value=np.nan)
+        arr.fill_value = 2
+        assert arr.fill_value == 2
+
+        arr = SparseArray([1, 0, 2], fill_value=0, dtype=np.int64)
+        arr.fill_value = 2
+        assert arr.fill_value == 2
+
+        msg = "fill_value must be a valid value for the SparseDtype.subtype"
+        with pytest.raises(ValueError, match=msg):
+            # GH#53043
+            arr.fill_value = 3.1
+        assert arr.fill_value == 2
+
+        arr.fill_value = np.nan
+        assert np.isnan(arr.fill_value)
+
+        arr = SparseArray([True, False, True], fill_value=False, dtype=np.bool_)
+        arr.fill_value = True
+        assert arr.fill_value is True
+
+        with pytest.raises(ValueError, match=msg):
+            arr.fill_value = 0
+        assert arr.fill_value is True
+
+        arr.fill_value = np.nan
+        assert np.isnan(arr.fill_value)
+
+    @pytest.mark.parametrize("val", [[1, 2, 3], np.array([1, 2]), (1, 2, 3)])
+    def test_set_fill_invalid_non_scalar(self, val):
+        arr = SparseArray([True, False, True], fill_value=False, dtype=np.bool_)
+        msg = "fill_value must be a scalar"
+
+        with pytest.raises(ValueError, match=msg):
+            arr.fill_value = val
+
+    def test_copy(self, arr):
+        arr2 = arr.copy()
+        assert arr2.sp_values is not arr.sp_values
+        assert arr2.sp_index is arr.sp_index
+
+    def test_values_asarray(self, arr_data, arr):
+        tm.assert_almost_equal(arr.to_dense(), arr_data)
+
+    @pytest.mark.parametrize(
+        "data,shape,dtype",
+        [
+            ([0, 0, 0, 0, 0], (5,), None),
+            ([], (0,), None),
+            ([0], (1,), None),
+            (["A", "A", np.nan, "B"], (4,), object),
+        ],
+    )
+    def test_shape(self, data, shape, dtype):
+        # GH 21126
+        out = SparseArray(data, dtype=dtype)
+        assert out.shape == shape
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            [np.nan, np.nan, np.nan, np.nan, np.nan],
+            [1, np.nan, np.nan, 3, np.nan],
+            [1, np.nan, 0, 3, 0],
+        ],
+    )
+    @pytest.mark.parametrize("fill_value", [None, 0])
+    def test_dense_repr(self, vals, fill_value):
+        vals = np.array(vals)
+        arr = SparseArray(vals, fill_value=fill_value)
+
+        res = arr.to_dense()
+        tm.assert_numpy_array_equal(res, vals)
+
+    @pytest.mark.parametrize("fix", ["arr", "zarr"])
+    def test_pickle(self, fix, request):
+        obj = request.getfixturevalue(fix)
+        unpickled = tm.round_trip_pickle(obj)
+        tm.assert_sp_array_equal(unpickled, obj)
+
+    def test_generator_warnings(self):
+        sp_arr = SparseArray([1, 2, 3])
+        with tm.assert_produces_warning(None):
+            for _ in sp_arr:
+                pass
+
+    def test_where_retain_fill_value(self):
+        # GH#45691 don't lose fill_value on _where
+        arr = SparseArray([np.nan, 1.0], fill_value=0)
+
+        mask = np.array([True, False])
+
+        res = arr._where(~mask, 1)
+        exp = SparseArray([1, 1.0], fill_value=0)
+        tm.assert_sp_array_equal(res, exp)
+
+        ser = pd.Series(arr)
+        res = ser.where(~mask, 1)
+        tm.assert_series_equal(res, pd.Series(exp))
+
+    def test_fillna(self):
+        s = SparseArray([1, np.nan, np.nan, 3, np.nan])
+        res = s.fillna(-1)
+        exp = SparseArray([1, -1, -1, 3, -1], fill_value=-1, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        s = SparseArray([1, np.nan, np.nan, 3, np.nan], fill_value=0)
+        res = s.fillna(-1)
+        exp = SparseArray([1, -1, -1, 3, -1], fill_value=0, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        s = SparseArray([1, np.nan, 0, 3, 0])
+        res = s.fillna(-1)
+        exp = SparseArray([1, -1, 0, 3, 0], fill_value=-1, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        s = SparseArray([1, np.nan, 0, 3, 0], fill_value=0)
+        res = s.fillna(-1)
+        exp = SparseArray([1, -1, 0, 3, 0], fill_value=0, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        s = SparseArray([np.nan, np.nan, np.nan, np.nan])
+        res = s.fillna(-1)
+        exp = SparseArray([-1, -1, -1, -1], fill_value=-1, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        s = SparseArray([np.nan, np.nan, np.nan, np.nan], fill_value=0)
+        res = s.fillna(-1)
+        exp = SparseArray([-1, -1, -1, -1], fill_value=0, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+        # float dtype's fill_value is np.nan, replaced by -1
+        s = SparseArray([0.0, 0.0, 0.0, 0.0])
+        res = s.fillna(-1)
+        exp = SparseArray([0.0, 0.0, 0.0, 0.0], fill_value=-1)
+        tm.assert_sp_array_equal(res, exp)
+
+        # int dtype shouldn't have missing. No changes.
+        s = SparseArray([0, 0, 0, 0])
+        assert s.dtype == SparseDtype(np.int64)
+        assert s.fill_value == 0
+        res = s.fillna(-1)
+        tm.assert_sp_array_equal(res, s)
+
+        s = SparseArray([0, 0, 0, 0], fill_value=0)
+        assert s.dtype == SparseDtype(np.int64)
+        assert s.fill_value == 0
+        res = s.fillna(-1)
+        exp = SparseArray([0, 0, 0, 0], fill_value=0)
+        tm.assert_sp_array_equal(res, exp)
+
+        # fill_value can be nan if there is no missing hole.
+        # only fill_value will be changed
+        s = SparseArray([0, 0, 0, 0], fill_value=np.nan)
+        assert s.dtype == SparseDtype(np.int64, fill_value=np.nan)
+        assert np.isnan(s.fill_value)
+        res = s.fillna(-1)
+        exp = SparseArray([0, 0, 0, 0], fill_value=-1)
+        tm.assert_sp_array_equal(res, exp)
+
+    def test_fillna_overlap(self):
+        s = SparseArray([1, np.nan, np.nan, 3, np.nan])
+        # filling with existing value doesn't replace existing value with
+        # fill_value, i.e. existing 3 remains in sp_values
+        res = s.fillna(3)
+        exp = np.array([1, 3, 3, 3, 3], dtype=np.float64)
+        tm.assert_numpy_array_equal(res.to_dense(), exp)
+
+        s = SparseArray([1, np.nan, np.nan, 3, np.nan], fill_value=0)
+        res = s.fillna(3)
+        exp = SparseArray([1, 3, 3, 3, 3], fill_value=0, dtype=np.float64)
+        tm.assert_sp_array_equal(res, exp)
+
+    def test_nonzero(self):
+        # Tests regression #21172.
+        sa = SparseArray([float("nan"), float("nan"), 1, 0, 0, 2, 0, 0, 0, 3, 0, 0])
+        expected = np.array([2, 5, 9], dtype=np.int32)
+        (result,) = sa.nonzero()
+        tm.assert_numpy_array_equal(expected, result)
+
+        sa = SparseArray([0, 0, 1, 0, 0, 2, 0, 0, 0, 3, 0, 0])
+        (result,) = sa.nonzero()
+        tm.assert_numpy_array_equal(expected, result)
+
+
+class TestSparseArrayAnalytics:
+    @pytest.mark.parametrize(
+        "data,expected",
+        [
+            (
+                np.array([1, 2, 3, 4, 5], dtype=float),  # non-null data
+                SparseArray(np.array([1.0, 3.0, 6.0, 10.0, 15.0])),
+            ),
+            (
+                np.array([1, 2, np.nan, 4, 5], dtype=float),  # null data
+                SparseArray(np.array([1.0, 3.0, np.nan, 7.0, 12.0])),
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("numpy", [True, False])
+    def test_cumsum(self, data, expected, numpy):
+        cumsum = np.cumsum if numpy else lambda s: s.cumsum()
+
+        out = cumsum(SparseArray(data))
+        tm.assert_sp_array_equal(out, expected)
+
+        out = cumsum(SparseArray(data, fill_value=np.nan))
+        tm.assert_sp_array_equal(out, expected)
+
+        out = cumsum(SparseArray(data, fill_value=2))
+        tm.assert_sp_array_equal(out, expected)
+
+        if numpy:  # numpy compatibility checks.
+            msg = "the 'dtype' parameter is not supported"
+            with pytest.raises(ValueError, match=msg):
+                np.cumsum(SparseArray(data), dtype=np.int64)
+
+            msg = "the 'out' parameter is not supported"
+            with pytest.raises(ValueError, match=msg):
+                np.cumsum(SparseArray(data), out=out)
+        else:
+            axis = 1  # SparseArray currently 1-D, so only axis = 0 is valid.
+            msg = re.escape(f"axis(={axis}) out of bounds")
+            with pytest.raises(ValueError, match=msg):
+                SparseArray(data).cumsum(axis=axis)
+
+    def test_ufunc(self):
+        # GH 13853 make sure ufunc is applied to fill_value
+        sparse = SparseArray([1, np.nan, 2, np.nan, -2])
+        result = SparseArray([1, np.nan, 2, np.nan, 2])
+        tm.assert_sp_array_equal(abs(sparse), result)
+        tm.assert_sp_array_equal(np.abs(sparse), result)
+
+        sparse = SparseArray([1, -1, 2, -2], fill_value=1)
+        result = SparseArray([1, 2, 2], sparse_index=sparse.sp_index, fill_value=1)
+        tm.assert_sp_array_equal(abs(sparse), result)
+        tm.assert_sp_array_equal(np.abs(sparse), result)
+
+        sparse = SparseArray([1, -1, 2, -2], fill_value=-1)
+        exp = SparseArray([1, 1, 2, 2], fill_value=1)
+        tm.assert_sp_array_equal(abs(sparse), exp)
+        tm.assert_sp_array_equal(np.abs(sparse), exp)
+
+        sparse = SparseArray([1, np.nan, 2, np.nan, -2])
+        result = SparseArray(np.sin([1, np.nan, 2, np.nan, -2]))
+        tm.assert_sp_array_equal(np.sin(sparse), result)
+
+        sparse = SparseArray([1, -1, 2, -2], fill_value=1)
+        result = SparseArray(np.sin([1, -1, 2, -2]), fill_value=np.sin(1))
+        tm.assert_sp_array_equal(np.sin(sparse), result)
+
+        sparse = SparseArray([1, -1, 0, -2], fill_value=0)
+        result = SparseArray(np.sin([1, -1, 0, -2]), fill_value=np.sin(0))
+        tm.assert_sp_array_equal(np.sin(sparse), result)
+
+    def test_ufunc_args(self):
+        # GH 13853 make sure ufunc is applied to fill_value, including its arg
+        sparse = SparseArray([1, np.nan, 2, np.nan, -2])
+        result = SparseArray([2, np.nan, 3, np.nan, -1])
+        tm.assert_sp_array_equal(np.add(sparse, 1), result)
+
+        sparse = SparseArray([1, -1, 2, -2], fill_value=1)
+        result = SparseArray([2, 0, 3, -1], fill_value=2)
+        tm.assert_sp_array_equal(np.add(sparse, 1), result)
+
+        sparse = SparseArray([1, -1, 0, -2], fill_value=0)
+        result = SparseArray([2, 0, 1, -1], fill_value=1)
+        tm.assert_sp_array_equal(np.add(sparse, 1), result)
+
+    @pytest.mark.parametrize("fill_value", [0.0, np.nan])
+    def test_modf(self, fill_value):
+        # https://github.com/pandas-dev/pandas/issues/26946
+        sparse = SparseArray([fill_value] * 10 + [1.1, 2.2], fill_value=fill_value)
+        r1, r2 = np.modf(sparse)
+        e1, e2 = np.modf(np.asarray(sparse))
+        tm.assert_sp_array_equal(r1, SparseArray(e1, fill_value=fill_value))
+        tm.assert_sp_array_equal(r2, SparseArray(e2, fill_value=fill_value))
+
+    def test_nbytes_integer(self):
+        arr = SparseArray([1, 0, 0, 0, 2], kind="integer")
+        result = arr.nbytes
+        # (2 * 8) + 2 * 4
+        assert result == 24
+
+    def test_nbytes_block(self):
+        arr = SparseArray([1, 2, 0, 0, 0], kind="block")
+        result = arr.nbytes
+        # (2 * 8) + 4 + 4
+        # sp_values, blocs, blengths
+        assert result == 24
+
+    def test_asarray_datetime64(self):
+        s = SparseArray(pd.to_datetime(["2012", None, None, "2013"]))
+        np.asarray(s)
+
+    def test_density(self):
+        arr = SparseArray([0, 1])
+        assert arr.density == 0.5
+
+    def test_npoints(self):
+        arr = SparseArray([0, 1])
+        assert arr.npoints == 1
+
+
+def test_setting_fill_value_fillna_still_works():
+    # This is why letting users update fill_value / dtype is bad
+    # astype has the same problem.
+    arr = SparseArray([1.0, np.nan, 1.0], fill_value=0.0)
+    arr.fill_value = np.nan
+    result = arr.isna()
+    # Can't do direct comparison, since the sp_index will be different
+    # So let's convert to ndarray and check there.
+    result = np.asarray(result)
+
+    expected = np.array([False, True, False])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_setting_fill_value_updates():
+    arr = SparseArray([0.0, np.nan], fill_value=0)
+    arr.fill_value = np.nan
+    # use private constructor to get the index right
+    # otherwise both nans would be un-stored.
+    expected = SparseArray._simple_new(
+        sparse_array=np.array([np.nan]),
+        sparse_index=IntIndex(2, [1]),
+        dtype=SparseDtype(float, np.nan),
+    )
+    tm.assert_sp_array_equal(arr, expected)
+
+
+@pytest.mark.parametrize(
+    "arr,fill_value,loc",
+    [
+        ([None, 1, 2], None, 0),
+        ([0, None, 2], None, 1),
+        ([0, 1, None], None, 2),
+        ([0, 1, 1, None, None], None, 3),
+        ([1, 1, 1, 2], None, -1),
+        ([], None, -1),
+        ([None, 1, 0, 0, None, 2], None, 0),
+        ([None, 1, 0, 0, None, 2], 1, 1),
+        ([None, 1, 0, 0, None, 2], 2, 5),
+        ([None, 1, 0, 0, None, 2], 3, -1),
+        ([None, 0, 0, 1, 2, 1], 0, 1),
+        ([None, 0, 0, 1, 2, 1], 1, 3),
+    ],
+)
+def test_first_fill_value_loc(arr, fill_value, loc):
+    result = SparseArray(arr, fill_value=fill_value)._first_fill_value_loc()
+    assert result == loc
+
+
+@pytest.mark.parametrize(
+    "arr",
+    [
+        [1, 2, np.nan, np.nan],
+        [1, np.nan, 2, np.nan],
+        [1, 2, np.nan],
+        [np.nan, 1, 0, 0, np.nan, 2],
+        [np.nan, 0, 0, 1, 2, 1],
+    ],
+)
+@pytest.mark.parametrize("fill_value", [np.nan, 0, 1])
+def test_unique_na_fill(arr, fill_value):
+    a = SparseArray(arr, fill_value=fill_value).unique()
+    b = pd.Series(arr).unique()
+    assert isinstance(a, SparseArray)
+    a = np.asarray(a)
+    tm.assert_numpy_array_equal(a, b)
+
+
+def test_unique_all_sparse():
+    # https://github.com/pandas-dev/pandas/issues/23168
+    arr = SparseArray([0, 0])
+    result = arr.unique()
+    expected = SparseArray([0])
+    tm.assert_sp_array_equal(result, expected)
+
+
+def test_map():
+    arr = SparseArray([0, 1, 2])
+    expected = SparseArray([10, 11, 12], fill_value=10)
+
+    # dict
+    result = arr.map({0: 10, 1: 11, 2: 12})
+    tm.assert_sp_array_equal(result, expected)
+
+    # series
+    result = arr.map(pd.Series({0: 10, 1: 11, 2: 12}))
+    tm.assert_sp_array_equal(result, expected)
+
+    # function
+    result = arr.map(pd.Series({0: 10, 1: 11, 2: 12}))
+    expected = SparseArray([10, 11, 12], fill_value=10)
+    tm.assert_sp_array_equal(result, expected)
+
+
+def test_map_missing():
+    arr = SparseArray([0, 1, 2])
+    expected = SparseArray([10, 11, None], fill_value=10)
+
+    result = arr.map({0: 10, 1: 11})
+    tm.assert_sp_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("fill_value", [np.nan, 1])
+def test_dropna(fill_value):
+    # GH-28287
+    arr = SparseArray([np.nan, 1], fill_value=fill_value)
+    exp = SparseArray([1.0], fill_value=fill_value)
+    tm.assert_sp_array_equal(arr.dropna(), exp)
+
+    df = pd.DataFrame({"a": [0, 1], "b": arr})
+    expected_df = pd.DataFrame({"a": [1], "b": exp}, index=pd.Index([1]))
+    tm.assert_equal(df.dropna(), expected_df)
+
+
+def test_drop_duplicates_fill_value():
+    # GH 11726
+    df = pd.DataFrame(np.zeros((5, 5))).apply(lambda x: SparseArray(x, fill_value=0))
+    result = df.drop_duplicates()
+    expected = pd.DataFrame({i: SparseArray([0.0], fill_value=0) for i in range(5)})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_zero_sparse_column():
+    # GH 27781
+    df1 = pd.DataFrame({"A": SparseArray([0, 0, 0]), "B": [1, 2, 3]})
+    df2 = pd.DataFrame({"A": SparseArray([0, 1, 0]), "B": [1, 2, 3]})
+    result = df1.loc[df1["B"] != 2]
+    expected = df2.loc[df2["B"] != 2]
+    tm.assert_frame_equal(result, expected)
+
+    expected = pd.DataFrame({"A": SparseArray([0, 0]), "B": [1, 3]}, index=[0, 2])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_array_interface(arr_data, arr):
+    # https://github.com/pandas-dev/pandas/pull/60046
+    result = np.asarray(arr)
+    tm.assert_numpy_array_equal(result, arr_data)
+
+    # it always gives a copy by default
+    result_copy1 = np.asarray(arr)
+    result_copy2 = np.asarray(arr)
+    assert not np.may_share_memory(result_copy1, result_copy2)
+
+    # or with explicit copy=True
+    result_copy1 = np.array(arr, copy=True)
+    result_copy2 = np.array(arr, copy=True)
+    assert not np.may_share_memory(result_copy1, result_copy2)
+
+    if not np_version_gt2:
+        # copy=False semantics are only supported in NumPy>=2.
+        return
+
+    # for sparse arrays, copy=False is never allowed
+    with pytest.raises(ValueError, match="Unable to avoid copy while creating"):
+        np.array(arr, copy=False)
+
+    # except when there are actually no sparse filled values
+    arr2 = SparseArray(np.array([1, 2, 3]))
+    result_nocopy1 = np.array(arr2, copy=False)
+    result_nocopy2 = np.array(arr2, copy=False)
+    assert np.may_share_memory(result_nocopy1, result_nocopy2)
diff --git a/pandas/pandas/tests/arrays/sparse/test_astype.py b/pandas/pandas/tests/arrays/sparse/test_astype.py
new file mode 100644
index 0000000000000..e6e4a11a0f5ab
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_astype.py
@@ -0,0 +1,133 @@
+import numpy as np
+import pytest
+
+from pandas._libs.sparse import IntIndex
+
+from pandas import (
+    SparseDtype,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+class TestAstype:
+    def test_astype(self):
+        # float -> float
+        arr = SparseArray([None, None, 0, 2])
+        result = arr.astype("Sparse[float32]")
+        expected = SparseArray([None, None, 0, 2], dtype=np.dtype("float32"))
+        tm.assert_sp_array_equal(result, expected)
+
+        dtype = SparseDtype("float64", fill_value=0)
+        result = arr.astype(dtype)
+        expected = SparseArray._simple_new(
+            np.array([0.0, 2.0], dtype=dtype.subtype), IntIndex(4, [2, 3]), dtype
+        )
+        tm.assert_sp_array_equal(result, expected)
+
+        dtype = SparseDtype("int64", 0)
+        result = arr.astype(dtype)
+        expected = SparseArray._simple_new(
+            np.array([0, 2], dtype=np.int64), IntIndex(4, [2, 3]), dtype
+        )
+        tm.assert_sp_array_equal(result, expected)
+
+        arr = SparseArray([0, np.nan, 0, 1], fill_value=0)
+        with pytest.raises(ValueError, match="NA"):
+            arr.astype("Sparse[i8]")
+
+    def test_astype_bool(self):
+        a = SparseArray([1, 0, 0, 1], dtype=SparseDtype(int, 0))
+        result = a.astype(bool)
+        expected = np.array([1, 0, 0, 1], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # update fill value
+        result = a.astype(SparseDtype(bool, False))
+        expected = SparseArray(
+            [True, False, False, True], dtype=SparseDtype(bool, False)
+        )
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_astype_all(self, any_real_numpy_dtype):
+        vals = np.array([1, 2, 3])
+        arr = SparseArray(vals, fill_value=1)
+        typ = np.dtype(any_real_numpy_dtype)
+        res = arr.astype(typ)
+        tm.assert_numpy_array_equal(res, vals.astype(any_real_numpy_dtype))
+
+    @pytest.mark.parametrize(
+        "arr, dtype, expected",
+        [
+            (
+                SparseArray([0, 1]),
+                "float",
+                SparseArray([0.0, 1.0], dtype=SparseDtype(float, 0.0)),
+            ),
+            (SparseArray([0, 1]), bool, SparseArray([False, True])),
+            (
+                SparseArray([0, 1], fill_value=1),
+                bool,
+                SparseArray([False, True], dtype=SparseDtype(bool, True)),
+            ),
+            pytest.param(
+                SparseArray([0, 1]),
+                "datetime64[ns]",
+                SparseArray(
+                    np.array([0, 1], dtype="datetime64[ns]"),
+                    dtype=SparseDtype("datetime64[ns]", Timestamp("1970")),
+                ),
+            ),
+            (
+                SparseArray([0, 1, 10]),
+                np.str_,
+                SparseArray(["0", "1", "10"], dtype=SparseDtype(np.str_, "0")),
+            ),
+            (SparseArray(["10", "20"]), float, SparseArray([10.0, 20.0])),
+            (
+                SparseArray([0, 1, 0]),
+                object,
+                SparseArray([0, 1, 0], dtype=SparseDtype(object, 0)),
+            ),
+        ],
+    )
+    def test_astype_more(self, arr, dtype, expected):
+        result = arr.astype(arr.dtype.update_dtype(dtype))
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_astype_nan_raises(self):
+        arr = SparseArray([1.0, np.nan])
+        with pytest.raises(ValueError, match="Cannot convert non-finite"):
+            arr.astype(int)
+
+    def test_astype_copy_false(self):
+        # GH#34456 bug caused by using .view instead of .astype in astype_nansafe
+        arr = SparseArray([1, 2, 3])
+
+        dtype = SparseDtype(float, 0)
+
+        result = arr.astype(dtype, copy=False)
+        expected = SparseArray([1.0, 2.0, 3.0], fill_value=0.0)
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_astype_dt64_to_int64(self):
+        # GH#49631 match non-sparse behavior
+        values = np.array(["NaT", "2016-01-02", "2016-01-03"], dtype="M8[ns]")
+
+        arr = SparseArray(values)
+        result = arr.astype("int64")
+        expected = values.astype("int64")
+        tm.assert_numpy_array_equal(result, expected)
+
+        # we should also be able to cast to equivalent Sparse[int64]
+        dtype_int64 = SparseDtype("int64", np.iinfo(np.int64).min)
+        result2 = arr.astype(dtype_int64)
+        tm.assert_numpy_array_equal(result2.to_numpy(), expected)
+
+        # GH#50087 we should match the non-sparse behavior regardless of
+        #  if we have a fill_value other than NaT
+        dtype = SparseDtype("datetime64[ns]", values[1])
+        arr3 = SparseArray(values, dtype=dtype)
+        result3 = arr3.astype("int64")
+        tm.assert_numpy_array_equal(result3, expected)
diff --git a/pandas/pandas/tests/arrays/sparse/test_combine_concat.py b/pandas/pandas/tests/arrays/sparse/test_combine_concat.py
new file mode 100644
index 0000000000000..0f09af269148b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_combine_concat.py
@@ -0,0 +1,62 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+class TestSparseArrayConcat:
+    @pytest.mark.parametrize("kind", ["integer", "block"])
+    def test_basic(self, kind):
+        a = SparseArray([1, 0, 0, 2], kind=kind)
+        b = SparseArray([1, 0, 2, 2], kind=kind)
+
+        result = SparseArray._concat_same_type([a, b])
+        # Can't make any assertions about the sparse index itself
+        # since we aren't don't merge sparse blocs across arrays
+        # in to_concat
+        expected = np.array([1, 2, 1, 2, 2], dtype="int64")
+        tm.assert_numpy_array_equal(result.sp_values, expected)
+        assert result.kind == kind
+
+    @pytest.mark.parametrize("kind", ["integer", "block"])
+    def test_uses_first_kind(self, kind):
+        other = "integer" if kind == "block" else "block"
+        a = SparseArray([1, 0, 0, 2], kind=kind)
+        b = SparseArray([1, 0, 2, 2], kind=other)
+
+        result = SparseArray._concat_same_type([a, b])
+        expected = np.array([1, 2, 1, 2, 2], dtype="int64")
+        tm.assert_numpy_array_equal(result.sp_values, expected)
+        assert result.kind == kind
+
+
+@pytest.mark.parametrize(
+    "other, expected_dtype",
+    [
+        # compatible dtype -> preserve sparse
+        (pd.Series([3, 4, 5], dtype="int64"), pd.SparseDtype("int64", 0)),
+        # (pd.Series([3, 4, 5], dtype="Int64"), pd.SparseDtype("int64", 0)),
+        # incompatible dtype -> Sparse[common dtype]
+        (pd.Series([1.5, 2.5, 3.5], dtype="float64"), pd.SparseDtype("float64", 0)),
+        # incompatible dtype -> Sparse[object] dtype
+        (pd.Series(["a", "b", "c"], dtype=object), pd.SparseDtype(object, 0)),
+        # categorical with compatible categories -> dtype of the categories
+        (pd.Series([3, 4, 5], dtype="category"), np.dtype("int64")),
+        (pd.Series([1.5, 2.5, 3.5], dtype="category"), np.dtype("float64")),
+        # categorical with incompatible categories -> object dtype
+        (pd.Series(["a", "b", "c"], dtype="category"), np.dtype(object)),
+    ],
+)
+def test_concat_with_non_sparse(other, expected_dtype):
+    # https://github.com/pandas-dev/pandas/issues/34336
+    s_sparse = pd.Series([1, 0, 2], dtype=pd.SparseDtype("int64", 0))
+
+    result = pd.concat([s_sparse, other], ignore_index=True)
+    expected = pd.Series(list(s_sparse) + list(other)).astype(expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = pd.concat([other, s_sparse], ignore_index=True)
+    expected = pd.Series(list(other) + list(s_sparse)).astype(expected_dtype)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/sparse/test_constructors.py b/pandas/pandas/tests/arrays/sparse/test_constructors.py
new file mode 100644
index 0000000000000..0bf3ab77e9eed
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_constructors.py
@@ -0,0 +1,277 @@
+import numpy as np
+import pytest
+
+from pandas._libs.sparse import IntIndex
+
+import pandas as pd
+from pandas import (
+    SparseDtype,
+    isna,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+class TestConstructors:
+    def test_constructor_dtype(self):
+        arr = SparseArray([np.nan, 1, 2, np.nan])
+        assert arr.dtype == SparseDtype(np.float64, np.nan)
+        assert arr.dtype.subtype == np.float64
+        assert np.isnan(arr.fill_value)
+
+        arr = SparseArray([np.nan, 1, 2, np.nan], fill_value=0)
+        assert arr.dtype == SparseDtype(np.float64, 0)
+        assert arr.fill_value == 0
+
+        arr = SparseArray([0, 1, 2, 4], dtype=np.float64)
+        assert arr.dtype == SparseDtype(np.float64, np.nan)
+        assert np.isnan(arr.fill_value)
+
+        arr = SparseArray([0, 1, 2, 4], dtype=np.int64)
+        assert arr.dtype == SparseDtype(np.int64, 0)
+        assert arr.fill_value == 0
+
+        arr = SparseArray([0, 1, 2, 4], fill_value=0, dtype=np.int64)
+        assert arr.dtype == SparseDtype(np.int64, 0)
+        assert arr.fill_value == 0
+
+        arr = SparseArray([0, 1, 2, 4], dtype=None)
+        assert arr.dtype == SparseDtype(np.int64, 0)
+        assert arr.fill_value == 0
+
+        arr = SparseArray([0, 1, 2, 4], fill_value=0, dtype=None)
+        assert arr.dtype == SparseDtype(np.int64, 0)
+        assert arr.fill_value == 0
+
+    def test_constructor_dtype_str(self):
+        result = SparseArray([1, 2, 3], dtype="int")
+        expected = SparseArray([1, 2, 3], dtype=int)
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_constructor_sparse_dtype(self):
+        result = SparseArray([1, 0, 0, 1], dtype=SparseDtype("int64", -1))
+        expected = SparseArray([1, 0, 0, 1], fill_value=-1, dtype=np.int64)
+        tm.assert_sp_array_equal(result, expected)
+        assert result.sp_values.dtype == np.dtype("int64")
+
+    def test_constructor_sparse_dtype_str(self):
+        result = SparseArray([1, 0, 0, 1], dtype="Sparse[int32]")
+        expected = SparseArray([1, 0, 0, 1], dtype=np.int32)
+        tm.assert_sp_array_equal(result, expected)
+        assert result.sp_values.dtype == np.dtype("int32")
+
+    def test_constructor_object_dtype(self):
+        # GH#11856
+        arr = SparseArray(["A", "A", np.nan, "B"], dtype=object)
+        assert arr.dtype == SparseDtype(object)
+        assert np.isnan(arr.fill_value)
+
+        arr = SparseArray(["A", "A", np.nan, "B"], dtype=object, fill_value="A")
+        assert arr.dtype == SparseDtype(object, "A")
+        assert arr.fill_value == "A"
+
+    def test_constructor_object_dtype_bool_fill(self):
+        # GH#17574
+        data = [False, 0, 100.0, 0.0]
+        arr = SparseArray(data, dtype=object, fill_value=False)
+        assert arr.dtype == SparseDtype(object, False)
+        assert arr.fill_value is False
+        arr_expected = np.array(data, dtype=object)
+        it = (type(x) == type(y) and x == y for x, y in zip(arr, arr_expected))
+        assert np.fromiter(it, dtype=np.bool_).all()
+
+    @pytest.mark.parametrize("dtype", [SparseDtype(int, 0), int])
+    def test_constructor_na_dtype(self, dtype):
+        with pytest.raises(ValueError, match="Cannot convert"):
+            SparseArray([0, 1, np.nan], dtype=dtype)
+
+    def test_constructor_warns_when_losing_timezone(self):
+        # GH#32501 warn when losing timezone information
+        dti = pd.date_range("2016-01-01", periods=3, tz="US/Pacific")
+
+        expected = SparseArray(np.asarray(dti, dtype="datetime64[ns]"))
+        msg = "loses timezone information"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = SparseArray(dti)
+
+        tm.assert_sp_array_equal(result, expected)
+
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = SparseArray(pd.Series(dti))
+
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_constructor_spindex_dtype(self):
+        arr = SparseArray(data=[1, 2], sparse_index=IntIndex(4, [1, 2]))
+        # TODO: actionable?
+        # XXX: Behavior change: specifying SparseIndex no longer changes the
+        # fill_value
+        expected = SparseArray([0, 1, 2, 0], kind="integer")
+        tm.assert_sp_array_equal(arr, expected)
+        assert arr.dtype == SparseDtype(np.int64)
+        assert arr.fill_value == 0
+
+        arr = SparseArray(
+            data=[1, 2, 3],
+            sparse_index=IntIndex(4, [1, 2, 3]),
+            dtype=np.int64,
+            fill_value=0,
+        )
+        exp = SparseArray([0, 1, 2, 3], dtype=np.int64, fill_value=0)
+        tm.assert_sp_array_equal(arr, exp)
+        assert arr.dtype == SparseDtype(np.int64)
+        assert arr.fill_value == 0
+
+        arr = SparseArray(
+            data=[1, 2], sparse_index=IntIndex(4, [1, 2]), fill_value=0, dtype=np.int64
+        )
+        exp = SparseArray([0, 1, 2, 0], fill_value=0, dtype=np.int64)
+        tm.assert_sp_array_equal(arr, exp)
+        assert arr.dtype == SparseDtype(np.int64)
+        assert arr.fill_value == 0
+
+        arr = SparseArray(
+            data=[1, 2, 3],
+            sparse_index=IntIndex(4, [1, 2, 3]),
+            dtype=None,
+            fill_value=0,
+        )
+        exp = SparseArray([0, 1, 2, 3], dtype=None)
+        tm.assert_sp_array_equal(arr, exp)
+        assert arr.dtype == SparseDtype(np.int64)
+        assert arr.fill_value == 0
+
+    @pytest.mark.parametrize("sparse_index", [None, IntIndex(1, [0])])
+    def test_constructor_spindex_dtype_scalar(self, sparse_index):
+        # scalar input
+        msg = "Cannot construct SparseArray from scalar data. Pass a sequence instead"
+        with pytest.raises(TypeError, match=msg):
+            SparseArray(data=1, sparse_index=sparse_index, dtype=None)
+
+        with pytest.raises(TypeError, match=msg):
+            SparseArray(data=1, sparse_index=IntIndex(1, [0]), dtype=None)
+
+    def test_constructor_spindex_dtype_scalar_broadcasts(self):
+        arr = SparseArray(
+            data=[1, 2], sparse_index=IntIndex(4, [1, 2]), fill_value=0, dtype=None
+        )
+        exp = SparseArray([0, 1, 2, 0], fill_value=0, dtype=None)
+        tm.assert_sp_array_equal(arr, exp)
+        assert arr.dtype == SparseDtype(np.int64)
+        assert arr.fill_value == 0
+
+    @pytest.mark.parametrize(
+        "data, fill_value",
+        [
+            (np.array([1, 2]), 0),
+            (np.array([1.0, 2.0]), np.nan),
+            ([True, False], False),
+            ([pd.Timestamp("2017-01-01")], pd.NaT),
+        ],
+    )
+    def test_constructor_inferred_fill_value(self, data, fill_value):
+        result = SparseArray(data).fill_value
+
+        if isna(fill_value):
+            assert isna(result)
+        else:
+            assert result == fill_value
+
+    @pytest.mark.parametrize("format", ["coo", "csc", "csr"])
+    @pytest.mark.parametrize("size", [0, 10])
+    def test_from_spmatrix(self, size, format):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        mat = sp_sparse.random(size, 1, density=0.5, format=format)
+        result = SparseArray.from_spmatrix(mat)
+
+        result = np.asarray(result)
+        expected = mat.toarray().ravel()
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("format", ["coo", "csc", "csr"])
+    def test_from_spmatrix_including_explicit_zero(self, format):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        mat = sp_sparse.random(10, 1, density=0.5, format=format)
+        mat.data[0] = 0
+        result = SparseArray.from_spmatrix(mat)
+
+        result = np.asarray(result)
+        expected = mat.toarray().ravel()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_from_spmatrix_raises(self):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        mat = sp_sparse.eye(5, 4, format="csc")
+
+        with pytest.raises(ValueError, match="not '4'"):
+            SparseArray.from_spmatrix(mat)
+
+    def test_constructor_from_too_large_array(self):
+        with pytest.raises(TypeError, match="expected dimension <= 1 data"):
+            SparseArray(np.arange(10).reshape((2, 5)))
+
+    def test_constructor_from_sparse(self):
+        zarr = SparseArray([0, 0, 1, 2, 3, 0, 4, 5, 0, 6], fill_value=0)
+        res = SparseArray(zarr)
+        assert res.fill_value == 0
+        tm.assert_almost_equal(res.sp_values, zarr.sp_values)
+
+    def test_constructor_copy(self):
+        arr_data = np.array([np.nan, np.nan, 1, 2, 3, np.nan, 4, 5, np.nan, 6])
+        arr = SparseArray(arr_data)
+
+        cp = SparseArray(arr, copy=True)
+        cp.sp_values[:3] = 0
+        assert not (arr.sp_values[:3] == 0).any()
+
+        not_copy = SparseArray(arr)
+        not_copy.sp_values[:3] = 0
+        assert (arr.sp_values[:3] == 0).all()
+
+    def test_constructor_bool(self):
+        # GH#10648
+        data = np.array([False, False, True, True, False, False])
+        arr = SparseArray(data, fill_value=False, dtype=bool)
+
+        assert arr.dtype == SparseDtype(bool)
+        tm.assert_numpy_array_equal(arr.sp_values, np.array([True, True]))
+        # Behavior change: np.asarray densifies.
+        # tm.assert_numpy_array_equal(arr.sp_values, np.asarray(arr))
+        tm.assert_numpy_array_equal(arr.sp_index.indices, np.array([2, 3], np.int32))
+
+        dense = arr.to_dense()
+        assert dense.dtype == bool
+        tm.assert_numpy_array_equal(dense, data)
+
+    def test_constructor_bool_fill_value(self):
+        arr = SparseArray([True, False, True], dtype=None)
+        assert arr.dtype == SparseDtype(np.bool_)
+        assert not arr.fill_value
+
+        arr = SparseArray([True, False, True], dtype=np.bool_)
+        assert arr.dtype == SparseDtype(np.bool_)
+        assert not arr.fill_value
+
+        arr = SparseArray([True, False, True], dtype=np.bool_, fill_value=True)
+        assert arr.dtype == SparseDtype(np.bool_, True)
+        assert arr.fill_value
+
+    def test_constructor_float32(self):
+        # GH#10648
+        data = np.array([1.0, np.nan, 3], dtype=np.float32)
+        arr = SparseArray(data, dtype=np.float32)
+
+        assert arr.dtype == SparseDtype(np.float32)
+        tm.assert_numpy_array_equal(arr.sp_values, np.array([1, 3], dtype=np.float32))
+        # Behavior change: np.asarray densifies.
+        # tm.assert_numpy_array_equal(arr.sp_values, np.asarray(arr))
+        tm.assert_numpy_array_equal(
+            arr.sp_index.indices, np.array([0, 2], dtype=np.int32)
+        )
+
+        dense = arr.to_dense()
+        assert dense.dtype == np.float32
+        tm.assert_numpy_array_equal(dense, data)
diff --git a/pandas/pandas/tests/arrays/sparse/test_dtype.py b/pandas/pandas/tests/arrays/sparse/test_dtype.py
new file mode 100644
index 0000000000000..6143163735ab8
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_dtype.py
@@ -0,0 +1,231 @@
+import re
+import warnings
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import SparseDtype
+
+
+@pytest.mark.parametrize(
+    "dtype, fill_value",
+    [
+        ("int", 0),
+        ("float", np.nan),
+        ("bool", False),
+        ("object", np.nan),
+        ("datetime64[ns]", np.datetime64("NaT", "ns")),
+        ("timedelta64[ns]", np.timedelta64("NaT", "ns")),
+    ],
+)
+def test_inferred_dtype(dtype, fill_value):
+    sparse_dtype = SparseDtype(dtype)
+    result = sparse_dtype.fill_value
+    if pd.isna(fill_value):
+        assert pd.isna(result) and type(result) == type(fill_value)
+    else:
+        assert result == fill_value
+
+
+def test_from_sparse_dtype():
+    dtype = SparseDtype("float", 0)
+    result = SparseDtype(dtype)
+    assert result.fill_value == 0
+
+
+def test_from_sparse_dtype_fill_value():
+    dtype = SparseDtype("int", 1)
+    result = SparseDtype(dtype, fill_value=2)
+    expected = SparseDtype("int", 2)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "dtype, fill_value",
+    [
+        ("int", None),
+        ("float", None),
+        ("bool", None),
+        ("object", None),
+        ("datetime64[ns]", None),
+        ("timedelta64[ns]", None),
+        ("int", np.nan),
+        ("float", 0),
+    ],
+)
+def test_equal(dtype, fill_value):
+    a = SparseDtype(dtype, fill_value)
+    b = SparseDtype(dtype, fill_value)
+    assert a == b
+    assert b == a
+
+
+def test_nans_equal():
+    a = SparseDtype(float, float("nan"))
+    b = SparseDtype(float, np.nan)
+    assert a == b
+    assert b == a
+
+
+def test_nans_not_equal():
+    # GH 54770
+    a = SparseDtype(float, 0)
+    b = SparseDtype(float, pd.NA)
+    assert a != b
+    assert b != a
+
+
+with warnings.catch_warnings():
+    msg = "Allowing arbitrary scalar fill_value in SparseDtype is deprecated"
+    warnings.filterwarnings("ignore", msg, category=FutureWarning)
+
+    tups = [
+        (SparseDtype("float64"), SparseDtype("float32")),
+        (SparseDtype("float64"), SparseDtype("float64", 0)),
+        (SparseDtype("float64"), SparseDtype("datetime64[ns]", np.nan)),
+        (SparseDtype("float64"), np.dtype("float64")),
+    ]
+
+
+@pytest.mark.parametrize(
+    "a, b",
+    tups,
+)
+def test_not_equal(a, b):
+    assert a != b
+
+
+def test_construct_from_string_raises():
+    with pytest.raises(
+        TypeError, match="Cannot construct a 'SparseDtype' from 'not a dtype'"
+    ):
+        SparseDtype.construct_from_string("not a dtype")
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        (int, True),
+        (float, True),
+        (bool, True),
+        (object, False),
+        (str, False),
+    ],
+)
+def test_is_numeric(dtype, expected):
+    assert SparseDtype(dtype)._is_numeric is expected
+
+
+def test_str_uses_object():
+    result = SparseDtype(str).subtype
+    assert result == np.dtype("object")
+
+
+@pytest.mark.parametrize(
+    "string, expected",
+    [
+        ("Sparse[float64]", SparseDtype(np.dtype("float64"))),
+        ("Sparse[float32]", SparseDtype(np.dtype("float32"))),
+        ("Sparse[int]", SparseDtype(np.dtype("int"))),
+        ("Sparse[str]", SparseDtype(np.dtype("str"))),
+        ("Sparse[datetime64[ns]]", SparseDtype(np.dtype("datetime64[ns]"))),
+        ("Sparse", SparseDtype(np.dtype("float"), np.nan)),
+    ],
+)
+def test_construct_from_string(string, expected):
+    result = SparseDtype.construct_from_string(string)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "a, b, expected",
+    [
+        (SparseDtype(float, 0.0), SparseDtype(np.dtype("float"), 0.0), True),
+        (SparseDtype(int, 0), SparseDtype(int, 0), True),
+        (SparseDtype(float, float("nan")), SparseDtype(float, np.nan), True),
+        (SparseDtype(float, 0), SparseDtype(float, np.nan), False),
+        (SparseDtype(int, 0.0), SparseDtype(float, 0.0), False),
+    ],
+)
+def test_hash_equal(a, b, expected):
+    result = a == b
+    assert result is expected
+
+    result = hash(a) == hash(b)
+    assert result is expected
+
+
+@pytest.mark.parametrize(
+    "string, expected",
+    [
+        ("Sparse[int]", "int"),
+        ("Sparse[int, 0]", "int"),
+        ("Sparse[int64]", "int64"),
+        ("Sparse[int64, 0]", "int64"),
+        ("Sparse[datetime64[ns], 0]", "datetime64[ns]"),
+    ],
+)
+def test_parse_subtype(string, expected):
+    subtype, _ = SparseDtype._parse_subtype(string)
+    assert subtype == expected
+
+
+@pytest.mark.parametrize(
+    "string", ["Sparse[int, 1]", "Sparse[float, 0.0]", "Sparse[bool, True]"]
+)
+def test_construct_from_string_fill_value_raises(string):
+    with pytest.raises(TypeError, match="fill_value in the string is not"):
+        SparseDtype.construct_from_string(string)
+
+
+@pytest.mark.parametrize(
+    "original, dtype, expected",
+    [
+        (SparseDtype(int, 0), float, SparseDtype(float, 0.0)),
+        (SparseDtype(int, 1), float, SparseDtype(float, 1.0)),
+        (SparseDtype(int, 1), np.str_, SparseDtype(object, "1")),
+        (SparseDtype(float, 1.5), int, SparseDtype(int, 1)),
+    ],
+)
+def test_update_dtype(original, dtype, expected):
+    result = original.update_dtype(dtype)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "original, dtype, expected_error_msg",
+    [
+        (
+            SparseDtype(float, np.nan),
+            int,
+            re.escape("Cannot convert non-finite values (NA or inf) to integer"),
+        ),
+        (
+            SparseDtype(str, "abc"),
+            int,
+            r"invalid literal for int\(\) with base 10: ('abc'|np\.str_\('abc'\))",
+        ),
+    ],
+)
+def test_update_dtype_raises(original, dtype, expected_error_msg):
+    with pytest.raises(ValueError, match=expected_error_msg):
+        original.update_dtype(dtype)
+
+
+def test_repr():
+    # GH-34352
+    result = str(SparseDtype("int64", fill_value=0))
+    expected = "Sparse[int64, 0]"
+    assert result == expected
+
+    result = str(SparseDtype(object, fill_value="0"))
+    expected = "Sparse[object, '0']"
+    assert result == expected
+
+
+def test_sparse_dtype_subtype_must_be_numpy_dtype():
+    # GH#53160
+    msg = "SparseDtype subtype must be a numpy dtype"
+    with pytest.raises(TypeError, match=msg):
+        SparseDtype("category", fill_value="c")
diff --git a/pandas/pandas/tests/arrays/sparse/test_indexing.py b/pandas/pandas/tests/arrays/sparse/test_indexing.py
new file mode 100644
index 0000000000000..60029ac06ddb4
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_indexing.py
@@ -0,0 +1,302 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import SparseDtype
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+@pytest.fixture
+def arr_data():
+    return np.array([np.nan, np.nan, 1, 2, 3, np.nan, 4, 5, np.nan, 6])
+
+
+@pytest.fixture
+def arr(arr_data):
+    return SparseArray(arr_data)
+
+
+class TestGetitem:
+    def test_getitem(self, arr):
+        dense = arr.to_dense()
+        for i, value in enumerate(arr):
+            tm.assert_almost_equal(value, dense[i])
+            tm.assert_almost_equal(arr[-i], dense[-i])
+
+    def test_getitem_arraylike_mask(self, arr):
+        arr = SparseArray([0, 1, 2])
+        result = arr[[True, False, True]]
+        expected = SparseArray([0, 2])
+        tm.assert_sp_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "slc",
+        [
+            np.s_[:],
+            np.s_[1:10],
+            np.s_[1:100],
+            np.s_[10:1],
+            np.s_[:-3],
+            np.s_[-5:-4],
+            np.s_[:-12],
+            np.s_[-12:],
+            np.s_[2:],
+            np.s_[2::3],
+            np.s_[::2],
+            np.s_[::-1],
+            np.s_[::-2],
+            np.s_[1:6:2],
+            np.s_[:-6:-2],
+        ],
+    )
+    @pytest.mark.parametrize(
+        "as_dense", [[np.nan] * 10, [1] * 10, [np.nan] * 5 + [1] * 5, []]
+    )
+    def test_getslice(self, slc, as_dense):
+        as_dense = np.array(as_dense)
+        arr = SparseArray(as_dense)
+
+        result = arr[slc]
+        expected = SparseArray(as_dense[slc])
+
+        tm.assert_sp_array_equal(result, expected)
+
+    def test_getslice_tuple(self):
+        dense = np.array([np.nan, 0, 3, 4, 0, 5, np.nan, np.nan, 0])
+
+        sparse = SparseArray(dense)
+        res = sparse[(slice(4, None),)]
+        exp = SparseArray(dense[4:])
+        tm.assert_sp_array_equal(res, exp)
+
+        sparse = SparseArray(dense, fill_value=0)
+        res = sparse[(slice(4, None),)]
+        exp = SparseArray(dense[4:], fill_value=0)
+        tm.assert_sp_array_equal(res, exp)
+
+        msg = "too many indices for array"
+        with pytest.raises(IndexError, match=msg):
+            sparse[4:, :]
+
+        with pytest.raises(IndexError, match=msg):
+            # check numpy compat
+            dense[4:, :]
+
+    def test_boolean_slice_empty(self):
+        arr = SparseArray([0, 1, 2])
+        res = arr[[False, False, False]]
+        assert res.dtype == arr.dtype
+
+    def test_getitem_bool_sparse_array(self, arr):
+        # GH 23122
+        spar_bool = SparseArray([False, True] * 5, dtype=np.bool_, fill_value=True)
+        exp = SparseArray([np.nan, 2, np.nan, 5, 6])
+        tm.assert_sp_array_equal(arr[spar_bool], exp)
+
+        spar_bool = ~spar_bool
+        res = arr[spar_bool]
+        exp = SparseArray([np.nan, 1, 3, 4, np.nan])
+        tm.assert_sp_array_equal(res, exp)
+
+        spar_bool = SparseArray(
+            [False, True, np.nan] * 3, dtype=np.bool_, fill_value=np.nan
+        )
+        res = arr[spar_bool]
+        exp = SparseArray([np.nan, 3, 5])
+        tm.assert_sp_array_equal(res, exp)
+
+    def test_getitem_bool_sparse_array_as_comparison(self):
+        # GH 45110
+        arr = SparseArray([1, 2, 3, 4, np.nan, np.nan], fill_value=np.nan)
+        res = arr[arr > 2]
+        exp = SparseArray([3.0, 4.0], fill_value=np.nan)
+        tm.assert_sp_array_equal(res, exp)
+
+    def test_get_item(self, arr):
+        zarr = SparseArray([0, 0, 1, 2, 3, 0, 4, 5, 0, 6], fill_value=0)
+
+        assert np.isnan(arr[1])
+        assert arr[2] == 1
+        assert arr[7] == 5
+
+        assert zarr[0] == 0
+        assert zarr[2] == 1
+        assert zarr[7] == 5
+
+        errmsg = "must be an integer between -10 and 10"
+
+        with pytest.raises(IndexError, match=errmsg):
+            arr[11]
+
+        with pytest.raises(IndexError, match=errmsg):
+            arr[-11]
+
+        assert arr[-1] == arr[len(arr) - 1]
+
+
+class TestSetitem:
+    def test_set_item(self, arr_data):
+        arr = SparseArray(arr_data).copy()
+
+        def setitem():
+            arr[5] = 3
+
+        def setslice():
+            arr[1:5] = 2
+
+        with pytest.raises(TypeError, match="assignment via setitem"):
+            setitem()
+
+        with pytest.raises(TypeError, match="assignment via setitem"):
+            setslice()
+
+
+class TestTake:
+    def test_take_scalar_raises(self, arr):
+        msg = "'indices' must be an array, not a scalar '2'."
+        with pytest.raises(ValueError, match=msg):
+            arr.take(2)
+
+    def test_take(self, arr_data, arr):
+        exp = SparseArray(np.take(arr_data, [2, 3]))
+        tm.assert_sp_array_equal(arr.take([2, 3]), exp)
+
+        exp = SparseArray(np.take(arr_data, [0, 1, 2]))
+        tm.assert_sp_array_equal(arr.take([0, 1, 2]), exp)
+
+    def test_take_all_empty(self):
+        sparse = pd.array([0, 0], dtype=SparseDtype("int64"))
+        result = sparse.take([0, 1], allow_fill=True, fill_value=np.nan)
+        tm.assert_sp_array_equal(sparse, result)
+
+    def test_take_different_fill_value(self):
+        # Take with a different fill value shouldn't overwrite the original
+        sparse = pd.array([0.0], dtype=SparseDtype("float64", fill_value=0.0))
+        result = sparse.take([0, -1], allow_fill=True, fill_value=np.nan)
+        expected = pd.array([0, np.nan], dtype=sparse.dtype)
+        tm.assert_sp_array_equal(expected, result)
+
+    def test_take_fill_value(self):
+        data = np.array([1, np.nan, 0, 3, 0])
+        sparse = SparseArray(data, fill_value=0)
+
+        exp = SparseArray(np.take(data, [0]), fill_value=0)
+        tm.assert_sp_array_equal(sparse.take([0]), exp)
+
+        exp = SparseArray(np.take(data, [1, 3, 4]), fill_value=0)
+        tm.assert_sp_array_equal(sparse.take([1, 3, 4]), exp)
+
+    def test_take_negative(self, arr_data, arr):
+        exp = SparseArray(np.take(arr_data, [-1]))
+        tm.assert_sp_array_equal(arr.take([-1]), exp)
+
+        exp = SparseArray(np.take(arr_data, [-4, -3, -2]))
+        tm.assert_sp_array_equal(arr.take([-4, -3, -2]), exp)
+
+    def test_bad_take(self, arr):
+        with pytest.raises(IndexError, match="bounds"):
+            arr.take([11])
+
+    def test_take_filling(self):
+        # similar tests as GH 12631
+        sparse = SparseArray([np.nan, np.nan, 1, np.nan, 4])
+        result = sparse.take(np.array([1, 0, -1]))
+        expected = SparseArray([np.nan, np.nan, 4])
+        tm.assert_sp_array_equal(result, expected)
+
+        # TODO: actionable?
+        # XXX: test change: fill_value=True -> allow_fill=True
+        result = sparse.take(np.array([1, 0, -1]), allow_fill=True)
+        expected = SparseArray([np.nan, np.nan, np.nan])
+        tm.assert_sp_array_equal(result, expected)
+
+        # allow_fill=False
+        result = sparse.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = SparseArray([np.nan, np.nan, 4])
+        tm.assert_sp_array_equal(result, expected)
+
+        msg = "Invalid value in 'indices'"
+        with pytest.raises(ValueError, match=msg):
+            sparse.take(np.array([1, 0, -2]), allow_fill=True)
+
+        with pytest.raises(ValueError, match=msg):
+            sparse.take(np.array([1, 0, -5]), allow_fill=True)
+
+        msg = "out of bounds value in 'indices'"
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, -6]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]), allow_fill=True)
+
+    def test_take_filling_fill_value(self):
+        # same tests as GH#12631
+        sparse = SparseArray([np.nan, 0, 1, 0, 4], fill_value=0)
+        result = sparse.take(np.array([1, 0, -1]))
+        expected = SparseArray([0, np.nan, 4], fill_value=0)
+        tm.assert_sp_array_equal(result, expected)
+
+        # fill_value
+        result = sparse.take(np.array([1, 0, -1]), allow_fill=True)
+        # TODO: actionable?
+        # XXX: behavior change.
+        # the old way of filling self.fill_value doesn't follow EA rules.
+        # It's supposed to be self.dtype.na_value (nan in this case)
+        expected = SparseArray([0, np.nan, np.nan], fill_value=0)
+        tm.assert_sp_array_equal(result, expected)
+
+        # allow_fill=False
+        result = sparse.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = SparseArray([0, np.nan, 4], fill_value=0)
+        tm.assert_sp_array_equal(result, expected)
+
+        msg = "Invalid value in 'indices'."
+        with pytest.raises(ValueError, match=msg):
+            sparse.take(np.array([1, 0, -2]), allow_fill=True)
+        with pytest.raises(ValueError, match=msg):
+            sparse.take(np.array([1, 0, -5]), allow_fill=True)
+
+        msg = "out of bounds value in 'indices'"
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, -6]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]), fill_value=True)
+
+    @pytest.mark.parametrize("kind", ["block", "integer"])
+    def test_take_filling_all_nan(self, kind):
+        sparse = SparseArray([np.nan, np.nan, np.nan, np.nan, np.nan], kind=kind)
+        result = sparse.take(np.array([1, 0, -1]))
+        expected = SparseArray([np.nan, np.nan, np.nan], kind=kind)
+        tm.assert_sp_array_equal(result, expected)
+
+        result = sparse.take(np.array([1, 0, -1]), fill_value=True)
+        expected = SparseArray([np.nan, np.nan, np.nan], kind=kind)
+        tm.assert_sp_array_equal(result, expected)
+
+        msg = "out of bounds value in 'indices'"
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, -6]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]))
+        with pytest.raises(IndexError, match=msg):
+            sparse.take(np.array([1, 5]), fill_value=True)
+
+
+class TestWhere:
+    def test_where_retain_fill_value(self):
+        # GH#45691 don't lose fill_value on _where
+        arr = SparseArray([np.nan, 1.0], fill_value=0)
+
+        mask = np.array([True, False])
+
+        res = arr._where(~mask, 1)
+        exp = SparseArray([1, 1.0], fill_value=0)
+        tm.assert_sp_array_equal(res, exp)
+
+        ser = pd.Series(arr)
+        res = ser.where(~mask, 1)
+        tm.assert_series_equal(res, pd.Series(exp))
diff --git a/pandas/pandas/tests/arrays/sparse/test_libsparse.py b/pandas/pandas/tests/arrays/sparse/test_libsparse.py
new file mode 100644
index 0000000000000..7a77a2064e7e0
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_libsparse.py
@@ -0,0 +1,551 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas._libs.sparse as splib
+import pandas.util._test_decorators as td
+
+from pandas import Series
+import pandas._testing as tm
+from pandas.core.arrays.sparse import (
+    BlockIndex,
+    IntIndex,
+    make_sparse_index,
+)
+
+
+@pytest.fixture
+def test_length():
+    return 20
+
+
+@pytest.fixture(
+    params=[
+        [
+            [0, 7, 15],
+            [3, 5, 5],
+            [2, 9, 14],
+            [2, 3, 5],
+            [2, 9, 15],
+            [1, 3, 4],
+        ],
+        [
+            [0, 5],
+            [4, 4],
+            [1],
+            [4],
+            [1],
+            [3],
+        ],
+        [
+            [0],
+            [10],
+            [0, 5],
+            [3, 7],
+            [0, 5],
+            [3, 5],
+        ],
+        [
+            [10],
+            [5],
+            [0, 12],
+            [5, 3],
+            [12],
+            [3],
+        ],
+        [
+            [0, 10],
+            [4, 6],
+            [5, 17],
+            [4, 2],
+            [],
+            [],
+        ],
+        [
+            [0],
+            [5],
+            [],
+            [],
+            [],
+            [],
+        ],
+    ],
+    ids=[
+        "plain_case",
+        "delete_blocks",
+        "split_blocks",
+        "skip_block",
+        "no_intersect",
+        "one_empty",
+    ],
+)
+def cases(request):
+    return request.param
+
+
+class TestSparseIndexUnion:
+    @pytest.mark.parametrize(
+        "xloc, xlen, yloc, ylen, eloc, elen",
+        [
+            [[0], [5], [5], [4], [0], [9]],
+            [[0, 10], [5, 5], [2, 17], [5, 2], [0, 10, 17], [7, 5, 2]],
+            [[1], [5], [3], [5], [1], [7]],
+            [[2, 10], [4, 4], [4], [8], [2], [12]],
+            [[0, 5], [3, 5], [0], [7], [0], [10]],
+            [[2, 10], [4, 4], [4, 13], [8, 4], [2], [15]],
+            [[2], [15], [4, 9, 14], [3, 2, 2], [2], [15]],
+            [[0, 10], [3, 3], [5, 15], [2, 2], [0, 5, 10, 15], [3, 2, 3, 2]],
+        ],
+    )
+    def test_index_make_union(self, xloc, xlen, yloc, ylen, eloc, elen, test_length):
+        # Case 1
+        # x: ----
+        # y:     ----
+        # r: --------
+        # Case 2
+        # x: -----     -----
+        # y:   -----          --
+        # Case 3
+        # x: ------
+        # y:    -------
+        # r: ----------
+        # Case 4
+        # x: ------  -----
+        # y:    -------
+        # r: -------------
+        # Case 5
+        # x: ---  -----
+        # y: -------
+        # r: -------------
+        # Case 6
+        # x: ------  -----
+        # y:    -------  ---
+        # r: -------------
+        # Case 7
+        # x: ----------------------
+        # y:   ----  ----   ---
+        # r: ----------------------
+        # Case 8
+        # x: ----       ---
+        # y:       ---       ---
+        xindex = BlockIndex(test_length, xloc, xlen)
+        yindex = BlockIndex(test_length, yloc, ylen)
+        bresult = xindex.make_union(yindex)
+        assert isinstance(bresult, BlockIndex)
+        tm.assert_numpy_array_equal(bresult.blocs, np.array(eloc, dtype=np.int32))
+        tm.assert_numpy_array_equal(bresult.blengths, np.array(elen, dtype=np.int32))
+
+        ixindex = xindex.to_int_index()
+        iyindex = yindex.to_int_index()
+        iresult = ixindex.make_union(iyindex)
+        assert isinstance(iresult, IntIndex)
+        tm.assert_numpy_array_equal(iresult.indices, bresult.to_int_index().indices)
+
+    def test_int_index_make_union(self):
+        a = IntIndex(5, np.array([0, 3, 4], dtype=np.int32))
+        b = IntIndex(5, np.array([0, 2], dtype=np.int32))
+        res = a.make_union(b)
+        exp = IntIndex(5, np.array([0, 2, 3, 4], np.int32))
+        assert res.equals(exp)
+
+        a = IntIndex(5, np.array([], dtype=np.int32))
+        b = IntIndex(5, np.array([0, 2], dtype=np.int32))
+        res = a.make_union(b)
+        exp = IntIndex(5, np.array([0, 2], np.int32))
+        assert res.equals(exp)
+
+        a = IntIndex(5, np.array([], dtype=np.int32))
+        b = IntIndex(5, np.array([], dtype=np.int32))
+        res = a.make_union(b)
+        exp = IntIndex(5, np.array([], np.int32))
+        assert res.equals(exp)
+
+        a = IntIndex(5, np.array([0, 1, 2, 3, 4], dtype=np.int32))
+        b = IntIndex(5, np.array([0, 1, 2, 3, 4], dtype=np.int32))
+        res = a.make_union(b)
+        exp = IntIndex(5, np.array([0, 1, 2, 3, 4], np.int32))
+        assert res.equals(exp)
+
+        a = IntIndex(5, np.array([0, 1], dtype=np.int32))
+        b = IntIndex(4, np.array([0, 1], dtype=np.int32))
+
+        msg = "Indices must reference same underlying length"
+        with pytest.raises(ValueError, match=msg):
+            a.make_union(b)
+
+
+class TestSparseIndexIntersect:
+    @td.skip_if_windows
+    def test_intersect(self, cases, test_length):
+        xloc, xlen, yloc, ylen, eloc, elen = cases
+        xindex = BlockIndex(test_length, xloc, xlen)
+        yindex = BlockIndex(test_length, yloc, ylen)
+        expected = BlockIndex(test_length, eloc, elen)
+        longer_index = BlockIndex(test_length + 1, yloc, ylen)
+
+        result = xindex.intersect(yindex)
+        assert result.equals(expected)
+        result = xindex.to_int_index().intersect(yindex.to_int_index())
+        assert result.equals(expected.to_int_index())
+
+        msg = "Indices must reference same underlying length"
+        with pytest.raises(Exception, match=msg):
+            xindex.intersect(longer_index)
+        with pytest.raises(Exception, match=msg):
+            xindex.to_int_index().intersect(longer_index.to_int_index())
+
+    def test_intersect_empty(self):
+        xindex = IntIndex(4, np.array([], dtype=np.int32))
+        yindex = IntIndex(4, np.array([2, 3], dtype=np.int32))
+        assert xindex.intersect(yindex).equals(xindex)
+        assert yindex.intersect(xindex).equals(xindex)
+
+        xindex = xindex.to_block_index()
+        yindex = yindex.to_block_index()
+        assert xindex.intersect(yindex).equals(xindex)
+        assert yindex.intersect(xindex).equals(xindex)
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            # Argument 2 to "IntIndex" has incompatible type "ndarray[Any,
+            # dtype[signedinteger[_32Bit]]]"; expected "Sequence[int]"
+            IntIndex(5, np.array([1, 2], dtype=np.int32)),  # type: ignore[arg-type]
+            IntIndex(5, np.array([0, 2, 4], dtype=np.int32)),  # type: ignore[arg-type]
+            IntIndex(0, np.array([], dtype=np.int32)),  # type: ignore[arg-type]
+            IntIndex(5, np.array([], dtype=np.int32)),  # type: ignore[arg-type]
+        ],
+    )
+    def test_intersect_identical(self, case):
+        assert case.intersect(case).equals(case)
+        case = case.to_block_index()
+        assert case.intersect(case).equals(case)
+
+
+class TestSparseIndexCommon:
+    def test_int_internal(self):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind="integer")
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 2
+        tm.assert_numpy_array_equal(idx.indices, np.array([2, 3], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind="integer")
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 0
+        tm.assert_numpy_array_equal(idx.indices, np.array([], dtype=np.int32))
+
+        idx = make_sparse_index(
+            4, np.array([0, 1, 2, 3], dtype=np.int32), kind="integer"
+        )
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 4
+        tm.assert_numpy_array_equal(idx.indices, np.array([0, 1, 2, 3], dtype=np.int32))
+
+    def test_block_internal(self):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 2
+        tm.assert_numpy_array_equal(idx.blocs, np.array([2], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([2], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 0
+        tm.assert_numpy_array_equal(idx.blocs, np.array([], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([0, 1, 2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 4
+        tm.assert_numpy_array_equal(idx.blocs, np.array([0], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([4], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([0, 2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 3
+        tm.assert_numpy_array_equal(idx.blocs, np.array([0, 2], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([1, 2], dtype=np.int32))
+
+    @pytest.mark.parametrize("kind", ["integer", "block"])
+    def test_lookup(self, kind):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind=kind)
+        assert idx.lookup(-1) == -1
+        assert idx.lookup(0) == -1
+        assert idx.lookup(1) == -1
+        assert idx.lookup(2) == 0
+        assert idx.lookup(3) == 1
+        assert idx.lookup(4) == -1
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind=kind)
+
+        for i in range(-1, 5):
+            assert idx.lookup(i) == -1
+
+        idx = make_sparse_index(4, np.array([0, 1, 2, 3], dtype=np.int32), kind=kind)
+        assert idx.lookup(-1) == -1
+        assert idx.lookup(0) == 0
+        assert idx.lookup(1) == 1
+        assert idx.lookup(2) == 2
+        assert idx.lookup(3) == 3
+        assert idx.lookup(4) == -1
+
+        idx = make_sparse_index(4, np.array([0, 2, 3], dtype=np.int32), kind=kind)
+        assert idx.lookup(-1) == -1
+        assert idx.lookup(0) == 0
+        assert idx.lookup(1) == -1
+        assert idx.lookup(2) == 1
+        assert idx.lookup(3) == 2
+        assert idx.lookup(4) == -1
+
+    @pytest.mark.parametrize("kind", ["integer", "block"])
+    def test_lookup_array(self, kind):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind=kind)
+
+        res = idx.lookup_array(np.array([-1, 0, 2], dtype=np.int32))
+        exp = np.array([-1, -1, 0], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        res = idx.lookup_array(np.array([4, 2, 1, 3], dtype=np.int32))
+        exp = np.array([-1, 0, -1, 1], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind=kind)
+        res = idx.lookup_array(np.array([-1, 0, 2, 4], dtype=np.int32))
+        exp = np.array([-1, -1, -1, -1], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        idx = make_sparse_index(4, np.array([0, 1, 2, 3], dtype=np.int32), kind=kind)
+        res = idx.lookup_array(np.array([-1, 0, 2], dtype=np.int32))
+        exp = np.array([-1, 0, 2], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        res = idx.lookup_array(np.array([4, 2, 1, 3], dtype=np.int32))
+        exp = np.array([-1, 2, 1, 3], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        idx = make_sparse_index(4, np.array([0, 2, 3], dtype=np.int32), kind=kind)
+        res = idx.lookup_array(np.array([2, 1, 3, 0], dtype=np.int32))
+        exp = np.array([1, -1, 2, 0], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+        res = idx.lookup_array(np.array([1, 4, 2, 5], dtype=np.int32))
+        exp = np.array([-1, -1, 1, -1], dtype=np.int32)
+        tm.assert_numpy_array_equal(res, exp)
+
+    @pytest.mark.parametrize(
+        "idx, expected",
+        [
+            [0, -1],
+            [5, 0],
+            [7, 2],
+            [8, -1],
+            [9, -1],
+            [10, -1],
+            [11, -1],
+            [12, 3],
+            [17, 8],
+            [18, -1],
+        ],
+    )
+    def test_lookup_basics(self, idx, expected):
+        bindex = BlockIndex(20, [5, 12], [3, 6])
+        assert bindex.lookup(idx) == expected
+
+        iindex = bindex.to_int_index()
+        assert iindex.lookup(idx) == expected
+
+
+class TestBlockIndex:
+    def test_block_internal(self):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 2
+        tm.assert_numpy_array_equal(idx.blocs, np.array([2], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([2], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 0
+        tm.assert_numpy_array_equal(idx.blocs, np.array([], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([0, 1, 2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 4
+        tm.assert_numpy_array_equal(idx.blocs, np.array([0], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([4], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([0, 2, 3], dtype=np.int32), kind="block")
+        assert isinstance(idx, BlockIndex)
+        assert idx.npoints == 3
+        tm.assert_numpy_array_equal(idx.blocs, np.array([0, 2], dtype=np.int32))
+        tm.assert_numpy_array_equal(idx.blengths, np.array([1, 2], dtype=np.int32))
+
+    @pytest.mark.parametrize("i", [5, 10, 100, 101])
+    def test_make_block_boundary(self, i):
+        idx = make_sparse_index(i, np.arange(0, i, 2, dtype=np.int32), kind="block")
+
+        exp = np.arange(0, i, 2, dtype=np.int32)
+        tm.assert_numpy_array_equal(idx.blocs, exp)
+        tm.assert_numpy_array_equal(idx.blengths, np.ones(len(exp), dtype=np.int32))
+
+    def test_equals(self):
+        index = BlockIndex(10, [0, 4], [2, 5])
+
+        assert index.equals(index)
+        assert not index.equals(BlockIndex(10, [0, 4], [2, 6]))
+
+    def test_check_integrity(self):
+        locs = []
+        lengths = []
+
+        # 0-length OK
+        BlockIndex(0, locs, lengths)
+
+        # also OK even though empty
+        BlockIndex(1, locs, lengths)
+
+        msg = "Block 0 extends beyond end"
+        with pytest.raises(ValueError, match=msg):
+            BlockIndex(10, [5], [10])
+
+        msg = "Block 0 overlaps"
+        with pytest.raises(ValueError, match=msg):
+            BlockIndex(10, [2, 5], [5, 3])
+
+    def test_to_int_index(self):
+        locs = [0, 10]
+        lengths = [4, 6]
+        exp_inds = [0, 1, 2, 3, 10, 11, 12, 13, 14, 15]
+
+        block = BlockIndex(20, locs, lengths)
+        dense = block.to_int_index()
+
+        tm.assert_numpy_array_equal(dense.indices, np.array(exp_inds, dtype=np.int32))
+
+    def test_to_block_index(self):
+        index = BlockIndex(10, [0, 5], [4, 5])
+        assert index.to_block_index() is index
+
+
+class TestIntIndex:
+    def test_check_integrity(self):
+        # Too many indices than specified in self.length
+        msg = "Too many indices"
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=1, indices=[1, 2, 3])
+
+        # No index can be negative.
+        msg = "No index can be less than zero"
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, -2, 3])
+
+        # No index can be negative.
+        msg = "No index can be less than zero"
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, -2, 3])
+
+        # All indices must be less than the length.
+        msg = "All indices must be less than the length"
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, 2, 5])
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, 2, 6])
+
+        # Indices must be strictly ascending.
+        msg = "Indices must be strictly increasing"
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, 3, 2])
+
+        with pytest.raises(ValueError, match=msg):
+            IntIndex(length=5, indices=[1, 3, 3])
+
+    def test_int_internal(self):
+        idx = make_sparse_index(4, np.array([2, 3], dtype=np.int32), kind="integer")
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 2
+        tm.assert_numpy_array_equal(idx.indices, np.array([2, 3], dtype=np.int32))
+
+        idx = make_sparse_index(4, np.array([], dtype=np.int32), kind="integer")
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 0
+        tm.assert_numpy_array_equal(idx.indices, np.array([], dtype=np.int32))
+
+        idx = make_sparse_index(
+            4, np.array([0, 1, 2, 3], dtype=np.int32), kind="integer"
+        )
+        assert isinstance(idx, IntIndex)
+        assert idx.npoints == 4
+        tm.assert_numpy_array_equal(idx.indices, np.array([0, 1, 2, 3], dtype=np.int32))
+
+    def test_equals(self):
+        index = IntIndex(10, [0, 1, 2, 3, 4])
+        assert index.equals(index)
+        assert not index.equals(IntIndex(10, [0, 1, 2, 3]))
+
+    def test_to_block_index(self, cases, test_length):
+        xloc, xlen, yloc, ylen, _, _ = cases
+        xindex = BlockIndex(test_length, xloc, xlen)
+        yindex = BlockIndex(test_length, yloc, ylen)
+
+        # see if survive the round trip
+        xbindex = xindex.to_int_index().to_block_index()
+        ybindex = yindex.to_int_index().to_block_index()
+        assert isinstance(xbindex, BlockIndex)
+        assert xbindex.equals(xindex)
+        assert ybindex.equals(yindex)
+
+    def test_to_int_index(self):
+        index = IntIndex(10, [2, 3, 4, 5, 6])
+        assert index.to_int_index() is index
+
+
+class TestSparseOperators:
+    @pytest.mark.parametrize("opname", ["add", "sub", "mul", "truediv", "floordiv"])
+    def test_op(self, opname, cases, test_length):
+        xloc, xlen, yloc, ylen, _, _ = cases
+        sparse_op = getattr(splib, f"sparse_{opname}_float64")
+        python_op = getattr(operator, opname)
+
+        xindex = BlockIndex(test_length, xloc, xlen)
+        yindex = BlockIndex(test_length, yloc, ylen)
+
+        xdindex = xindex.to_int_index()
+        ydindex = yindex.to_int_index()
+
+        x = np.arange(xindex.npoints) * 10.0 + 1
+        y = np.arange(yindex.npoints) * 100.0 + 1
+
+        xfill = 0
+        yfill = 2
+
+        result_block_vals, rb_index, bfill = sparse_op(
+            x, xindex, xfill, y, yindex, yfill
+        )
+        result_int_vals, ri_index, ifill = sparse_op(
+            x, xdindex, xfill, y, ydindex, yfill
+        )
+
+        assert rb_index.to_int_index().equals(ri_index)
+        tm.assert_numpy_array_equal(result_block_vals, result_int_vals)
+        assert bfill == ifill
+
+        # check versus Series...
+        xseries = Series(x, xdindex.indices)
+        xseries = xseries.reindex(np.arange(test_length)).fillna(xfill)
+
+        yseries = Series(y, ydindex.indices)
+        yseries = yseries.reindex(np.arange(test_length)).fillna(yfill)
+
+        series_result = python_op(xseries, yseries)
+        series_result = series_result.reindex(ri_index.indices)
+
+        tm.assert_numpy_array_equal(result_block_vals, series_result.values)
+        tm.assert_numpy_array_equal(result_int_vals, series_result.values)
diff --git a/pandas/pandas/tests/arrays/sparse/test_reductions.py b/pandas/pandas/tests/arrays/sparse/test_reductions.py
new file mode 100644
index 0000000000000..4171d1213a0dc
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_reductions.py
@@ -0,0 +1,304 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NaT,
+    SparseDtype,
+    Timestamp,
+    isna,
+)
+from pandas.core.arrays.sparse import SparseArray
+
+
+class TestReductions:
+    @pytest.mark.parametrize(
+        "data,pos,neg",
+        [
+            ([True, True, True], True, False),
+            ([1, 2, 1], 1, 0),
+            ([1.0, 2.0, 1.0], 1.0, 0.0),
+        ],
+    )
+    def test_all(self, data, pos, neg):
+        # GH#17570
+        out = SparseArray(data).all()
+        assert out
+
+        out = SparseArray(data, fill_value=pos).all()
+        assert out
+
+        data[1] = neg
+        out = SparseArray(data).all()
+        assert not out
+
+        out = SparseArray(data, fill_value=pos).all()
+        assert not out
+
+    @pytest.mark.parametrize(
+        "data,pos,neg",
+        [
+            ([True, True, True], True, False),
+            ([1, 2, 1], 1, 0),
+            ([1.0, 2.0, 1.0], 1.0, 0.0),
+        ],
+    )
+    def test_numpy_all(self, data, pos, neg):
+        # GH#17570
+        out = np.all(SparseArray(data))
+        assert out
+
+        out = np.all(SparseArray(data, fill_value=pos))
+        assert out
+
+        data[1] = neg
+        out = np.all(SparseArray(data))
+        assert not out
+
+        out = np.all(SparseArray(data, fill_value=pos))
+        assert not out
+
+        # raises with a different message on py2.
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.all(SparseArray(data), out=np.array([]))
+
+    @pytest.mark.parametrize(
+        "data,pos,neg",
+        [
+            ([False, True, False], True, False),
+            ([0, 2, 0], 2, 0),
+            ([0.0, 2.0, 0.0], 2.0, 0.0),
+        ],
+    )
+    def test_any(self, data, pos, neg):
+        # GH#17570
+        out = SparseArray(data).any()
+        assert out
+
+        out = SparseArray(data, fill_value=pos).any()
+        assert out
+
+        data[1] = neg
+        out = SparseArray(data).any()
+        assert not out
+
+        out = SparseArray(data, fill_value=pos).any()
+        assert not out
+
+    @pytest.mark.parametrize(
+        "data,pos,neg",
+        [
+            ([False, True, False], True, False),
+            ([0, 2, 0], 2, 0),
+            ([0.0, 2.0, 0.0], 2.0, 0.0),
+        ],
+    )
+    def test_numpy_any(self, data, pos, neg):
+        # GH#17570
+        out = np.any(SparseArray(data))
+        assert out
+
+        out = np.any(SparseArray(data, fill_value=pos))
+        assert out
+
+        data[1] = neg
+        out = np.any(SparseArray(data))
+        assert not out
+
+        out = np.any(SparseArray(data, fill_value=pos))
+        assert not out
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.any(SparseArray(data), out=out)
+
+    def test_sum(self):
+        data = np.arange(10).astype(float)
+        out = SparseArray(data).sum()
+        assert out == 45.0
+
+        data[5] = np.nan
+        out = SparseArray(data, fill_value=2).sum()
+        assert out == 40.0
+
+        out = SparseArray(data, fill_value=np.nan).sum()
+        assert out == 40.0
+
+    @pytest.mark.parametrize(
+        "arr",
+        [[0, 1, np.nan, 1], [0, 1, 1]],
+    )
+    @pytest.mark.parametrize("fill_value", [0, 1, np.nan])
+    @pytest.mark.parametrize("min_count, expected", [(3, 2), (4, np.nan)])
+    def test_sum_min_count(self, arr, fill_value, min_count, expected):
+        # GH#25777
+        sparray = SparseArray(np.array(arr), fill_value=fill_value)
+        result = sparray.sum(min_count=min_count)
+        if np.isnan(expected):
+            assert np.isnan(result)
+        else:
+            assert result == expected
+
+    def test_bool_sum_min_count(self):
+        spar_bool = SparseArray([False, True] * 5, dtype=np.bool_, fill_value=True)
+        res = spar_bool.sum(min_count=1)
+        assert res == 5
+        res = spar_bool.sum(min_count=11)
+        assert isna(res)
+
+    def test_numpy_sum(self):
+        data = np.arange(10).astype(float)
+        out = np.sum(SparseArray(data))
+        assert out == 45.0
+
+        data[5] = np.nan
+        out = np.sum(SparseArray(data, fill_value=2))
+        assert out == 40.0
+
+        out = np.sum(SparseArray(data, fill_value=np.nan))
+        assert out == 40.0
+
+        msg = "the 'dtype' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.sum(SparseArray(data), dtype=np.int64)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.sum(SparseArray(data), out=out)
+
+    def test_mean(self):
+        data = np.arange(10).astype(float)
+        out = SparseArray(data).mean()
+        assert out == 4.5
+
+        data[5] = np.nan
+        out = SparseArray(data).mean()
+        assert out == 40.0 / 9
+
+    def test_numpy_mean(self):
+        data = np.arange(10).astype(float)
+        out = np.mean(SparseArray(data))
+        assert out == 4.5
+
+        data[5] = np.nan
+        out = np.mean(SparseArray(data))
+        assert out == 40.0 / 9
+
+        msg = "the 'dtype' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.mean(SparseArray(data), dtype=np.int64)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.mean(SparseArray(data), out=out)
+
+
+class TestMinMax:
+    @pytest.mark.parametrize(
+        "raw_data,max_expected,min_expected",
+        [
+            (np.arange(5.0), [4], [0]),
+            (-np.arange(5.0), [0], [-4]),
+            (np.array([0, 1, 2, np.nan, 4]), [4], [0]),
+            (np.array([np.nan] * 5), [np.nan], [np.nan]),
+            (np.array([]), [np.nan], [np.nan]),
+        ],
+    )
+    def test_nan_fill_value(self, raw_data, max_expected, min_expected):
+        arr = SparseArray(raw_data)
+        max_result = arr.max()
+        min_result = arr.min()
+        assert max_result in max_expected
+        assert min_result in min_expected
+
+        max_result = arr.max(skipna=False)
+        min_result = arr.min(skipna=False)
+        if np.isnan(raw_data).any():
+            assert np.isnan(max_result)
+            assert np.isnan(min_result)
+        else:
+            assert max_result in max_expected
+            assert min_result in min_expected
+
+    @pytest.mark.parametrize(
+        "fill_value,max_expected,min_expected",
+        [
+            (100, 100, 0),
+            (-100, 1, -100),
+        ],
+    )
+    def test_fill_value(self, fill_value, max_expected, min_expected):
+        arr = SparseArray(
+            np.array([fill_value, 0, 1]), dtype=SparseDtype("int", fill_value)
+        )
+        max_result = arr.max()
+        assert max_result == max_expected
+
+        min_result = arr.min()
+        assert min_result == min_expected
+
+    def test_only_fill_value(self):
+        fv = 100
+        arr = SparseArray(np.array([fv, fv, fv]), dtype=SparseDtype("int", fv))
+        assert len(arr._valid_sp_values) == 0
+
+        assert arr.max() == fv
+        assert arr.min() == fv
+        assert arr.max(skipna=False) == fv
+        assert arr.min(skipna=False) == fv
+
+    @pytest.mark.parametrize("func", ["min", "max"])
+    @pytest.mark.parametrize("data", [np.array([]), np.array([np.nan, np.nan])])
+    @pytest.mark.parametrize(
+        "dtype,expected",
+        [
+            (SparseDtype(np.float64, np.nan), np.nan),
+            (SparseDtype(np.float64, 5.0), np.nan),
+            (SparseDtype("datetime64[ns]", NaT), NaT),
+            (SparseDtype("datetime64[ns]", Timestamp("2018-05-05")), NaT),
+        ],
+    )
+    def test_na_value_if_no_valid_values(self, func, data, dtype, expected):
+        arr = SparseArray(data, dtype=dtype)
+        result = getattr(arr, func)()
+        if expected is NaT:
+            # TODO: pin down whether we wrap datetime64("NaT")
+            assert result is NaT or np.isnat(result)
+        else:
+            assert np.isnan(result)
+
+
+class TestArgmaxArgmin:
+    @pytest.mark.parametrize(
+        "arr,argmax_expected,argmin_expected",
+        [
+            (SparseArray([1, 2, 0, 1, 2]), 1, 2),
+            (SparseArray([-1, -2, 0, -1, -2]), 2, 1),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, -1]), 1, 5),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2]), 5, 2),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2], fill_value=-1), 5, 2),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2], fill_value=0), 5, 2),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2], fill_value=1), 5, 2),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2], fill_value=2), 5, 2),
+            (SparseArray([np.nan, 1, 0, 0, np.nan, 2], fill_value=3), 5, 2),
+            (SparseArray([0] * 10 + [-1], fill_value=0), 0, 10),
+            (SparseArray([0] * 10 + [-1], fill_value=-1), 0, 10),
+            (SparseArray([0] * 10 + [-1], fill_value=1), 0, 10),
+            (SparseArray([-1] + [0] * 10, fill_value=0), 1, 0),
+            (SparseArray([1] + [0] * 10, fill_value=0), 0, 1),
+            (SparseArray([-1] + [0] * 10, fill_value=-1), 1, 0),
+            (SparseArray([1] + [0] * 10, fill_value=1), 0, 1),
+        ],
+    )
+    def test_argmax_argmin(self, arr, argmax_expected, argmin_expected):
+        argmax_result = arr.argmax()
+        argmin_result = arr.argmin()
+        assert argmax_result == argmax_expected
+        assert argmin_result == argmin_expected
+
+    @pytest.mark.parametrize("method", ["argmax", "argmin"])
+    def test_empty_array(self, method):
+        msg = f"attempt to get {method} of an empty sequence"
+        arr = SparseArray([])
+        with pytest.raises(ValueError, match=msg):
+            getattr(arr, method)()
diff --git a/pandas/pandas/tests/arrays/sparse/test_unary.py b/pandas/pandas/tests/arrays/sparse/test_unary.py
new file mode 100644
index 0000000000000..c00a73773fdd4
--- /dev/null
+++ b/pandas/pandas/tests/arrays/sparse/test_unary.py
@@ -0,0 +1,79 @@
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import SparseArray
+
+
+@pytest.mark.filterwarnings("ignore:invalid value encountered in cast:RuntimeWarning")
+@pytest.mark.parametrize("fill_value", [0, np.nan])
+@pytest.mark.parametrize("op", [operator.pos, operator.neg])
+def test_unary_op(op, fill_value):
+    arr = np.array([0, 1, np.nan, 2])
+    sparray = SparseArray(arr, fill_value=fill_value)
+    result = op(sparray)
+    expected = SparseArray(op(arr), fill_value=op(fill_value))
+    tm.assert_sp_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("fill_value", [True, False])
+def test_invert(fill_value):
+    arr = np.array([True, False, False, True])
+    sparray = SparseArray(arr, fill_value=fill_value)
+    result = ~sparray
+    expected = SparseArray(~arr, fill_value=not fill_value)
+    tm.assert_sp_array_equal(result, expected)
+
+    result = ~pd.Series(sparray)
+    expected = pd.Series(expected)
+    tm.assert_series_equal(result, expected)
+
+    result = ~pd.DataFrame({"A": sparray})
+    expected = pd.DataFrame({"A": expected})
+    tm.assert_frame_equal(result, expected)
+
+
+class TestUnaryMethods:
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_neg_operator(self):
+        arr = SparseArray([-1, -2, np.nan, 3], fill_value=np.nan, dtype=np.int8)
+        res = -arr
+        exp = SparseArray([1, 2, np.nan, -3], fill_value=np.nan, dtype=np.int8)
+        tm.assert_sp_array_equal(exp, res)
+
+        arr = SparseArray([-1, -2, 1, 3], fill_value=-1, dtype=np.int8)
+        res = -arr
+        exp = SparseArray([1, 2, -1, -3], fill_value=1, dtype=np.int8)
+        tm.assert_sp_array_equal(exp, res)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_abs_operator(self):
+        arr = SparseArray([-1, -2, np.nan, 3], fill_value=np.nan, dtype=np.int8)
+        res = abs(arr)
+        exp = SparseArray([1, 2, np.nan, 3], fill_value=np.nan, dtype=np.int8)
+        tm.assert_sp_array_equal(exp, res)
+
+        arr = SparseArray([-1, -2, 1, 3], fill_value=-1, dtype=np.int8)
+        res = abs(arr)
+        exp = SparseArray([1, 2, 1, 3], fill_value=1, dtype=np.int8)
+        tm.assert_sp_array_equal(exp, res)
+
+    def test_invert_operator(self):
+        arr = SparseArray([False, True, False, True], fill_value=False, dtype=np.bool_)
+        exp = SparseArray(
+            np.invert([False, True, False, True]), fill_value=True, dtype=np.bool_
+        )
+        res = ~arr
+        tm.assert_sp_array_equal(exp, res)
+
+        arr = SparseArray([0, 1, 0, 2, 3, 0], fill_value=0, dtype=np.int32)
+        res = ~arr
+        exp = SparseArray([-1, -2, -1, -3, -4, -1], fill_value=-1, dtype=np.int32)
+        tm.assert_sp_array_equal(exp, res)
diff --git a/pandas/pandas/tests/arrays/string_/__init__.py b/pandas/pandas/tests/arrays/string_/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/string_/test_concat.py b/pandas/pandas/tests/arrays/string_/test_concat.py
new file mode 100644
index 0000000000000..320d700b2b6c3
--- /dev/null
+++ b/pandas/pandas/tests/arrays/string_/test_concat.py
@@ -0,0 +1,73 @@
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+
+from pandas.core.dtypes.cast import find_common_type
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+@pytest.mark.parametrize(
+    "to_concat_dtypes, result_dtype",
+    [
+        # same types
+        ([("pyarrow", pd.NA), ("pyarrow", pd.NA)], ("pyarrow", pd.NA)),
+        ([("pyarrow", np.nan), ("pyarrow", np.nan)], ("pyarrow", np.nan)),
+        ([("python", pd.NA), ("python", pd.NA)], ("python", pd.NA)),
+        ([("python", np.nan), ("python", np.nan)], ("python", np.nan)),
+        # pyarrow preference
+        ([("pyarrow", pd.NA), ("python", pd.NA)], ("pyarrow", pd.NA)),
+        # NA preference
+        ([("python", pd.NA), ("python", np.nan)], ("python", pd.NA)),
+    ],
+)
+def test_concat_series(request, to_concat_dtypes, result_dtype):
+    if any(storage == "pyarrow" for storage, _ in to_concat_dtypes) and not HAS_PYARROW:
+        pytest.skip("Could not import 'pyarrow'")
+
+    ser_list = [
+        pd.Series(["a", "b", None], dtype=pd.StringDtype(storage, na_value))
+        for storage, na_value in to_concat_dtypes
+    ]
+
+    result = pd.concat(ser_list, ignore_index=True)
+    expected = pd.Series(
+        ["a", "b", None, "a", "b", None], dtype=pd.StringDtype(*result_dtype)
+    )
+    tm.assert_series_equal(result, expected)
+
+    # order doesn't matter for result
+    result = pd.concat(ser_list[::1], ignore_index=True)
+    tm.assert_series_equal(result, expected)
+
+
+def test_concat_with_object(string_dtype_arguments):
+    # _get_common_dtype cannot inspect values, so object dtype with strings still
+    # results in object dtype
+    result = pd.concat(
+        [
+            pd.Series(["a", "b", None], dtype=pd.StringDtype(*string_dtype_arguments)),
+            pd.Series(["a", "b", None], dtype=object),
+        ]
+    )
+    assert result.dtype == np.dtype("object")
+
+
+def test_concat_with_numpy(string_dtype_arguments):
+    # common type with a numpy string dtype always preserves the pandas string dtype
+    dtype = pd.StringDtype(*string_dtype_arguments)
+    assert find_common_type([dtype, np.dtype("U")]) == dtype
+    assert find_common_type([np.dtype("U"), dtype]) == dtype
+    assert find_common_type([dtype, np.dtype("U10")]) == dtype
+    assert find_common_type([np.dtype("U10"), dtype]) == dtype
+
+    # with any other numpy dtype -> object
+    assert find_common_type([dtype, np.dtype("S")]) == np.dtype("object")
+    assert find_common_type([dtype, np.dtype("int64")]) == np.dtype("object")
+
+    if Version(np.__version__) >= Version("2"):
+        assert find_common_type([dtype, np.dtypes.StringDType()]) == dtype
+        assert find_common_type([np.dtypes.StringDType(), dtype]) == dtype
diff --git a/pandas/pandas/tests/arrays/string_/test_string.py b/pandas/pandas/tests/arrays/string_/test_string.py
new file mode 100644
index 0000000000000..336a0fef69170
--- /dev/null
+++ b/pandas/pandas/tests/arrays/string_/test_string.py
@@ -0,0 +1,766 @@
+"""
+This module tests the functionality of StringArray and ArrowStringArray.
+Tests for the str accessors are in pandas/tests/strings/test_string_array.py
+"""
+
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat.pyarrow import (
+    pa_version_under12p0,
+    pa_version_under19p0,
+)
+
+from pandas.core.dtypes.common import is_dtype_equal
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.string_ import StringArrayNumpySemantics
+from pandas.core.arrays.string_arrow import (
+    ArrowStringArray,
+    ArrowStringArrayNumpySemantics,
+)
+
+
+@pytest.fixture
+def dtype(string_dtype_arguments):
+    """Fixture giving StringDtype from parametrized storage and na_value arguments"""
+    storage, na_value = string_dtype_arguments
+    return pd.StringDtype(storage=storage, na_value=na_value)
+
+
+@pytest.fixture
+def dtype2(string_dtype_arguments2):
+    storage, na_value = string_dtype_arguments2
+    return pd.StringDtype(storage=storage, na_value=na_value)
+
+
+@pytest.fixture
+def cls(dtype):
+    """Fixture giving array type from parametrized 'dtype'"""
+    return dtype.construct_array_type()
+
+
+def test_dtype_constructor():
+    pytest.importorskip("pyarrow")
+
+    with tm.assert_produces_warning(FutureWarning):
+        dtype = pd.StringDtype("pyarrow_numpy")
+    assert dtype == pd.StringDtype("pyarrow", na_value=np.nan)
+
+
+def test_dtype_equality():
+    pytest.importorskip("pyarrow")
+
+    dtype1 = pd.StringDtype("python")
+    dtype2 = pd.StringDtype("pyarrow")
+    dtype3 = pd.StringDtype("pyarrow", na_value=np.nan)
+
+    assert dtype1 == pd.StringDtype("python", na_value=pd.NA)
+    assert dtype1 != dtype2
+    assert dtype1 != dtype3
+
+    assert dtype2 == pd.StringDtype("pyarrow", na_value=pd.NA)
+    assert dtype2 != dtype1
+    assert dtype2 != dtype3
+
+    assert dtype3 == pd.StringDtype("pyarrow", na_value=np.nan)
+    assert dtype3 == pd.StringDtype("pyarrow", na_value=float("nan"))
+    assert dtype3 != dtype1
+    assert dtype3 != dtype2
+
+
+def test_repr(dtype):
+    df = pd.DataFrame({"A": pd.array(["a", pd.NA, "b"], dtype=dtype)})
+    if dtype.na_value is np.nan:
+        expected = "     A\n0    a\n1  NaN\n2    b"
+    else:
+        expected = "      A\n0     a\n1  <NA>\n2     b"
+    assert repr(df) == expected
+
+    if dtype.na_value is np.nan:
+        expected = "0      a\n1    NaN\n2      b\nName: A, dtype: str"
+    else:
+        expected = "0       a\n1    <NA>\n2       b\nName: A, dtype: string"
+    assert repr(df.A) == expected
+
+    if dtype.storage == "pyarrow" and dtype.na_value is pd.NA:
+        arr_name = "ArrowStringArray"
+        expected = f"<{arr_name}>\n['a', <NA>, 'b']\nLength: 3, dtype: string"
+    elif dtype.storage == "pyarrow" and dtype.na_value is np.nan:
+        arr_name = "ArrowStringArrayNumpySemantics"
+        expected = f"<{arr_name}>\n['a', nan, 'b']\nLength: 3, dtype: str"
+    elif dtype.storage == "python" and dtype.na_value is np.nan:
+        arr_name = "StringArrayNumpySemantics"
+        expected = f"<{arr_name}>\n['a', nan, 'b']\nLength: 3, dtype: str"
+    else:
+        arr_name = "StringArray"
+        expected = f"<{arr_name}>\n['a', <NA>, 'b']\nLength: 3, dtype: string"
+    assert repr(df.A.array) == expected
+
+
+def test_none_to_nan(cls, dtype):
+    a = cls._from_sequence(["a", None, "b"], dtype=dtype)
+    assert a[1] is not None
+    assert a[1] is a.dtype.na_value
+
+
+def test_setitem_validates(cls, dtype):
+    arr = cls._from_sequence(["a", "b"], dtype=dtype)
+
+    msg = "Invalid value '10' for dtype 'str"
+    with pytest.raises(TypeError, match=msg):
+        arr[0] = 10
+
+    msg = "Invalid value for dtype 'str"
+    with pytest.raises(TypeError, match=msg):
+        arr[:] = np.array([1, 2])
+
+
+def test_setitem_with_scalar_string(dtype):
+    # is_float_dtype considers some strings, like 'd', to be floats
+    # which can cause issues.
+    arr = pd.array(["a", "c"], dtype=dtype)
+    arr[0] = "d"
+    expected = pd.array(["d", "c"], dtype=dtype)
+    tm.assert_extension_array_equal(arr, expected)
+
+
+def test_setitem_with_array_with_missing(dtype):
+    # ensure that when setting with an array of values, we don't mutate the
+    # array `value` in __setitem__(self, key, value)
+    arr = pd.array(["a", "b", "c"], dtype=dtype)
+    value = np.array(["A", None])
+    value_orig = value.copy()
+    arr[[0, 1]] = value
+
+    expected = pd.array(["A", pd.NA, "c"], dtype=dtype)
+    tm.assert_extension_array_equal(arr, expected)
+    tm.assert_numpy_array_equal(value, value_orig)
+
+
+def test_astype_roundtrip(dtype):
+    ser = pd.Series(pd.date_range("2000", periods=12))
+    ser[0] = None
+
+    casted = ser.astype(dtype)
+    assert is_dtype_equal(casted.dtype, dtype)
+
+    result = casted.astype("datetime64[ns]")
+    tm.assert_series_equal(result, ser)
+
+    # GH#38509 same thing for timedelta64
+    ser2 = ser - ser.iloc[-1]
+    casted2 = ser2.astype(dtype)
+    assert is_dtype_equal(casted2.dtype, dtype)
+
+    result2 = casted2.astype(ser2.dtype)
+    tm.assert_series_equal(result2, ser2)
+
+
+def test_add(dtype):
+    a = pd.Series(["a", "b", "c", None, None], dtype=dtype)
+    b = pd.Series(["x", "y", None, "z", None], dtype=dtype)
+
+    result = a + b
+    expected = pd.Series(["ax", "by", None, None, None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = a.add(b)
+    tm.assert_series_equal(result, expected)
+
+    result = a.radd(b)
+    expected = pd.Series(["xa", "yb", None, None, None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = a.add(b, fill_value="-")
+    expected = pd.Series(["ax", "by", "c-", "-z", None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_add_2d(dtype, request):
+    if dtype.storage == "pyarrow":
+        reason = "Failed: DID NOT RAISE <class 'ValueError'>"
+        mark = pytest.mark.xfail(raises=None, reason=reason)
+        request.applymarker(mark)
+
+    a = pd.array(["a", "b", "c"], dtype=dtype)
+    b = np.array([["a", "b", "c"]], dtype=object)
+    with pytest.raises(ValueError, match="3 != 1"):
+        a + b
+
+    s = pd.Series(a)
+    with pytest.raises(ValueError, match="3 != 1"):
+        s + b
+
+
+def test_add_sequence(dtype):
+    a = pd.array(["a", "b", None, None], dtype=dtype)
+    other = ["x", None, "y", None]
+
+    result = a + other
+    expected = pd.array(["ax", None, None, None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = other + a
+    expected = pd.array(["xa", None, None, None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_mul(dtype):
+    a = pd.array(["a", "b", None], dtype=dtype)
+    result = a * 2
+    expected = pd.array(["aa", "bb", None], dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    result = 2 * a
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.xfail(reason="GH-28527")
+def test_add_strings(dtype):
+    arr = pd.array(["a", "b", "c", "d"], dtype=dtype)
+    df = pd.DataFrame([["t", "y", "v", "w"]], dtype=object)
+    assert arr.__add__(df) is NotImplemented
+
+    result = arr + df
+    expected = pd.DataFrame([["at", "by", "cv", "dw"]]).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+    result = df + arr
+    expected = pd.DataFrame([["ta", "yb", "vc", "wd"]]).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.xfail(reason="GH-28527")
+def test_add_frame(dtype):
+    arr = pd.array(["a", "b", np.nan, np.nan], dtype=dtype)
+    df = pd.DataFrame([["x", np.nan, "y", np.nan]])
+
+    assert arr.__add__(df) is NotImplemented
+
+    result = arr + df
+    expected = pd.DataFrame([["ax", np.nan, np.nan, np.nan]]).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+    result = df + arr
+    expected = pd.DataFrame([["xa", np.nan, np.nan, np.nan]]).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comparison_methods_scalar(comparison_op, dtype):
+    op_name = f"__{comparison_op.__name__}__"
+    a = pd.array(["a", None, "c"], dtype=dtype)
+    other = "a"
+    result = getattr(a, op_name)(other)
+    if dtype.na_value is np.nan:
+        expected = np.array([getattr(item, op_name)(other) for item in a])
+        if comparison_op == operator.ne:
+            expected[1] = True
+        else:
+            expected[1] = False
+        tm.assert_numpy_array_equal(result, expected.astype(np.bool_))
+    else:
+        expected_dtype = "boolean[pyarrow]" if dtype.storage == "pyarrow" else "boolean"
+        expected = np.array([getattr(item, op_name)(other) for item in a], dtype=object)
+        expected = pd.array(expected, dtype=expected_dtype)
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_comparison_methods_scalar_pd_na(comparison_op, dtype):
+    op_name = f"__{comparison_op.__name__}__"
+    a = pd.array(["a", None, "c"], dtype=dtype)
+    result = getattr(a, op_name)(pd.NA)
+
+    if dtype.na_value is np.nan:
+        if operator.ne == comparison_op:
+            expected = np.array([True, True, True])
+        else:
+            expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+    else:
+        expected_dtype = "boolean[pyarrow]" if dtype.storage == "pyarrow" else "boolean"
+        expected = pd.array([None, None, None], dtype=expected_dtype)
+        tm.assert_extension_array_equal(result, expected)
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_comparison_methods_scalar_not_string(comparison_op, dtype):
+    op_name = f"__{comparison_op.__name__}__"
+
+    a = pd.array(["a", None, "c"], dtype=dtype)
+    other = 42
+
+    if op_name not in ["__eq__", "__ne__"]:
+        with pytest.raises(TypeError, match="Invalid comparison|not supported between"):
+            getattr(a, op_name)(other)
+
+        return
+
+    result = getattr(a, op_name)(other)
+
+    if dtype.na_value is np.nan:
+        expected_data = {
+            "__eq__": [False, False, False],
+            "__ne__": [True, True, True],
+        }[op_name]
+        expected = np.array(expected_data)
+        tm.assert_numpy_array_equal(result, expected)
+    else:
+        expected_data = {"__eq__": [False, None, False], "__ne__": [True, None, True]}[
+            op_name
+        ]
+        expected_dtype = "boolean[pyarrow]" if dtype.storage == "pyarrow" else "boolean"
+        expected = pd.array(expected_data, dtype=expected_dtype)
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_comparison_methods_array(comparison_op, dtype):
+    op_name = f"__{comparison_op.__name__}__"
+
+    a = pd.array(["a", None, "c"], dtype=dtype)
+    other = [None, None, "c"]
+    result = getattr(a, op_name)(other)
+    if dtype.na_value is np.nan:
+        if operator.ne == comparison_op:
+            expected = np.array([True, True, False])
+        else:
+            expected = np.array([False, False, False])
+            expected[-1] = getattr(other[-1], op_name)(a[-1])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = getattr(a, op_name)(pd.NA)
+        if operator.ne == comparison_op:
+            expected = np.array([True, True, True])
+        else:
+            expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    else:
+        expected_dtype = "boolean[pyarrow]" if dtype.storage == "pyarrow" else "boolean"
+        expected = np.full(len(a), fill_value=None, dtype="object")
+        expected[-1] = getattr(other[-1], op_name)(a[-1])
+        expected = pd.array(expected, dtype=expected_dtype)
+        tm.assert_extension_array_equal(result, expected)
+
+        result = getattr(a, op_name)(pd.NA)
+        expected = pd.array([None, None, None], dtype=expected_dtype)
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_constructor_raises(cls):
+    if cls is pd.arrays.StringArray:
+        msg = "StringArray requires a sequence of strings or pandas.NA"
+    elif cls is StringArrayNumpySemantics:
+        msg = "StringArrayNumpySemantics requires a sequence of strings or NaN"
+    else:
+        msg = "Unsupported type '<class 'numpy.ndarray'>' for ArrowExtensionArray"
+
+    with pytest.raises(ValueError, match=msg):
+        cls(np.array(["a", "b"], dtype="S1"))
+
+    with pytest.raises(ValueError, match=msg):
+        cls(np.array([]))
+
+    if cls is pd.arrays.StringArray or cls is StringArrayNumpySemantics:
+        # GH#45057 np.nan and None do NOT raise, as they are considered valid NAs
+        #  for string dtype
+        cls(np.array(["a", np.nan], dtype=object))
+        cls(np.array(["a", None], dtype=object))
+    else:
+        with pytest.raises(ValueError, match=msg):
+            cls(np.array(["a", np.nan], dtype=object))
+        with pytest.raises(ValueError, match=msg):
+            cls(np.array(["a", None], dtype=object))
+
+    with pytest.raises(ValueError, match=msg):
+        cls(np.array(["a", pd.NaT], dtype=object))
+
+    with pytest.raises(ValueError, match=msg):
+        cls(np.array(["a", np.datetime64("NaT", "ns")], dtype=object))
+
+    with pytest.raises(ValueError, match=msg):
+        cls(np.array(["a", np.timedelta64("NaT", "ns")], dtype=object))
+
+
+@pytest.mark.parametrize("na", [np.nan, np.float64("nan"), float("nan"), None, pd.NA])
+def test_constructor_nan_like(na):
+    expected = pd.arrays.StringArray(np.array(["a", pd.NA]))
+    tm.assert_extension_array_equal(
+        pd.arrays.StringArray(np.array(["a", na], dtype="object")), expected
+    )
+
+
+@pytest.mark.parametrize("copy", [True, False])
+def test_from_sequence_no_mutate(copy, cls, dtype):
+    nan_arr = np.array(["a", np.nan], dtype=object)
+    expected_input = nan_arr.copy()
+    na_arr = np.array(["a", pd.NA], dtype=object)
+
+    result = cls._from_sequence(nan_arr, dtype=dtype, copy=copy)
+
+    if cls in (ArrowStringArray, ArrowStringArrayNumpySemantics):
+        import pyarrow as pa
+
+        expected = cls(pa.array(na_arr, type=pa.string(), from_pandas=True))
+    elif cls is StringArrayNumpySemantics:
+        expected = cls(nan_arr)
+    else:
+        expected = cls(na_arr)
+
+    tm.assert_extension_array_equal(result, expected)
+    tm.assert_numpy_array_equal(nan_arr, expected_input)
+
+
+def test_astype_int(dtype):
+    arr = pd.array(["1", "2", "3"], dtype=dtype)
+    result = arr.astype("int64")
+    expected = np.array([1, 2, 3], dtype="int64")
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = pd.array(["1", pd.NA, "3"], dtype=dtype)
+    if dtype.na_value is np.nan:
+        err = ValueError
+        msg = "cannot convert float NaN to integer"
+    else:
+        err = TypeError
+        msg = (
+            r"int\(\) argument must be a string, a bytes-like "
+            r"object or a( real)? number"
+        )
+    with pytest.raises(err, match=msg):
+        arr.astype("int64")
+
+
+def test_astype_nullable_int(dtype):
+    arr = pd.array(["1", pd.NA, "3"], dtype=dtype)
+
+    result = arr.astype("Int64")
+    expected = pd.array([1, pd.NA, 3], dtype="Int64")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_astype_float(dtype, any_float_dtype):
+    # Don't compare arrays (37974)
+    ser = pd.Series(["1.1", pd.NA, "3.3"], dtype=dtype)
+    result = ser.astype(any_float_dtype)
+    expected = pd.Series([1.1, np.nan, 3.3], dtype=any_float_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reduce(skipna, dtype):
+    arr = pd.Series(["a", "b", "c"], dtype=dtype)
+    result = arr.sum(skipna=skipna)
+    assert result == "abc"
+
+
+def test_reduce_missing(skipna, dtype):
+    arr = pd.Series([None, "a", None, "b", "c", None], dtype=dtype)
+    result = arr.sum(skipna=skipna)
+    if skipna:
+        assert result == "abc"
+    else:
+        assert pd.isna(result)
+
+
+@pytest.mark.parametrize("method", ["min", "max"])
+def test_min_max(method, skipna, dtype):
+    arr = pd.Series(["a", "b", "c", None], dtype=dtype)
+    result = getattr(arr, method)(skipna=skipna)
+    if skipna:
+        expected = "a" if method == "min" else "c"
+        assert result == expected
+    else:
+        assert result is arr.dtype.na_value
+
+
+@pytest.mark.parametrize("method", ["min", "max"])
+@pytest.mark.parametrize("box", [pd.Series, pd.array])
+def test_min_max_numpy(method, box, dtype, request):
+    if dtype.storage == "pyarrow" and box is pd.array:
+        if box is pd.array:
+            reason = "'<=' not supported between instances of 'str' and 'NoneType'"
+        else:
+            reason = "'ArrowStringArray' object has no attribute 'max'"
+        mark = pytest.mark.xfail(raises=TypeError, reason=reason)
+        request.applymarker(mark)
+
+    arr = box(["a", "b", "c", None], dtype=dtype)
+    result = getattr(np, method)(arr)
+    expected = "a" if method == "min" else "c"
+    assert result == expected
+
+
+def test_fillna_args(dtype):
+    # GH 37987
+
+    arr = pd.array(["a", pd.NA], dtype=dtype)
+
+    res = arr.fillna(value="b")
+    expected = pd.array(["a", "b"], dtype=dtype)
+    tm.assert_extension_array_equal(res, expected)
+
+    res = arr.fillna(value=np.str_("b"))
+    expected = pd.array(["a", "b"], dtype=dtype)
+    tm.assert_extension_array_equal(res, expected)
+
+    msg = "Invalid value '1' for dtype 'str"
+    with pytest.raises(TypeError, match=msg):
+        arr.fillna(value=1)
+
+
+def test_arrow_array(dtype):
+    # protocol added in 0.15.0
+    pa = pytest.importorskip("pyarrow")
+    import pyarrow.compute as pc
+
+    data = pd.array(["a", "b", "c"], dtype=dtype)
+    arr = pa.array(data)
+    expected = pa.array(list(data), type=pa.large_string(), from_pandas=True)
+    if dtype.storage == "pyarrow" and pa_version_under12p0:
+        expected = pa.chunked_array(expected)
+    if dtype.storage == "python":
+        expected = pc.cast(expected, pa.string())
+    assert arr.equals(expected)
+
+
+@pytest.mark.filterwarnings("ignore:Passing a BlockManager:DeprecationWarning")
+def test_arrow_roundtrip(dtype, string_storage, using_infer_string):
+    # roundtrip possible from arrow 1.0.0
+    pa = pytest.importorskip("pyarrow")
+
+    data = pd.array(["a", "b", None], dtype=dtype)
+    df = pd.DataFrame({"a": data})
+    table = pa.table(df)
+    if dtype.storage == "python":
+        assert table.field("a").type == "string"
+    else:
+        assert table.field("a").type == "large_string"
+    with pd.option_context("string_storage", string_storage):
+        result = table.to_pandas()
+    if dtype.na_value is np.nan and not using_infer_string:
+        assert result["a"].dtype == "object"
+    else:
+        assert isinstance(result["a"].dtype, pd.StringDtype)
+        expected = df.astype(pd.StringDtype(string_storage, na_value=dtype.na_value))
+        if using_infer_string:
+            expected.columns = expected.columns.astype(
+                pd.StringDtype(string_storage, na_value=np.nan)
+            )
+        tm.assert_frame_equal(result, expected)
+        # ensure the missing value is represented by NA and not np.nan or None
+        assert result.loc[2, "a"] is result["a"].dtype.na_value
+
+
+@pytest.mark.filterwarnings("ignore:Passing a BlockManager:DeprecationWarning")
+def test_arrow_from_string(using_infer_string):
+    # not roundtrip,  but starting with pyarrow table without pandas metadata
+    pa = pytest.importorskip("pyarrow")
+    table = pa.table({"a": pa.array(["a", "b", None], type=pa.string())})
+
+    result = table.to_pandas()
+
+    if using_infer_string and not pa_version_under19p0:
+        expected = pd.DataFrame({"a": ["a", "b", None]}, dtype="str")
+    else:
+        expected = pd.DataFrame({"a": ["a", "b", None]}, dtype="object")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:Passing a BlockManager:DeprecationWarning")
+def test_arrow_load_from_zero_chunks(dtype, string_storage, using_infer_string):
+    # GH-41040
+    pa = pytest.importorskip("pyarrow")
+
+    data = pd.array([], dtype=dtype)
+    df = pd.DataFrame({"a": data})
+    table = pa.table(df)
+    if dtype.storage == "python":
+        assert table.field("a").type == "string"
+    else:
+        assert table.field("a").type == "large_string"
+    # Instantiate the same table with no chunks at all
+    table = pa.table([pa.chunked_array([], type=pa.string())], schema=table.schema)
+    with pd.option_context("string_storage", string_storage):
+        result = table.to_pandas()
+
+    if dtype.na_value is np.nan and not using_string_dtype():
+        assert result["a"].dtype == "object"
+    else:
+        assert isinstance(result["a"].dtype, pd.StringDtype)
+        expected = df.astype(pd.StringDtype(string_storage, na_value=dtype.na_value))
+        if using_infer_string:
+            expected.columns = expected.columns.astype(
+                pd.StringDtype(string_storage, na_value=np.nan)
+            )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_value_counts_na(dtype):
+    if dtype.na_value is np.nan:
+        exp_dtype = "int64"
+    elif dtype.storage == "pyarrow":
+        exp_dtype = "int64[pyarrow]"
+    else:
+        exp_dtype = "Int64"
+    arr = pd.array(["a", "b", "a", pd.NA], dtype=dtype)
+    result = arr.value_counts(dropna=False)
+    expected = pd.Series([2, 1, 1], index=arr[[0, 1, 3]], dtype=exp_dtype, name="count")
+    tm.assert_series_equal(result, expected)
+
+    result = arr.value_counts(dropna=True)
+    expected = pd.Series([2, 1], index=arr[:2], dtype=exp_dtype, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_with_normalize(dtype):
+    if dtype.na_value is np.nan:
+        exp_dtype = np.float64
+    elif dtype.storage == "pyarrow":
+        exp_dtype = "double[pyarrow]"
+    else:
+        exp_dtype = "Float64"
+    ser = pd.Series(["a", "b", "a", pd.NA], dtype=dtype)
+    result = ser.value_counts(normalize=True)
+    expected = pd.Series([2, 1], index=ser[:2], dtype=exp_dtype, name="proportion") / 3
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_sort_false(dtype):
+    if dtype.na_value is np.nan:
+        exp_dtype = "int64"
+    elif dtype.storage == "pyarrow":
+        exp_dtype = "int64[pyarrow]"
+    else:
+        exp_dtype = "Int64"
+    ser = pd.Series(["a", "b", "c", "b"], dtype=dtype)
+    result = ser.value_counts(sort=False)
+    expected = pd.Series([1, 2, 1], index=ser[:3], dtype=exp_dtype, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_memory_usage(dtype):
+    # GH 33963
+
+    if dtype.storage == "pyarrow":
+        pytest.skip(f"not applicable for {dtype.storage}")
+
+    series = pd.Series(["a", "b", "c"], dtype=dtype)
+
+    assert 0 < series.nbytes <= series.memory_usage() < series.memory_usage(deep=True)
+
+
+@pytest.mark.parametrize("float_dtype", [np.float16, np.float32, np.float64])
+def test_astype_from_float_dtype(float_dtype, dtype):
+    # https://github.com/pandas-dev/pandas/issues/36451
+    ser = pd.Series([0.1], dtype=float_dtype)
+    result = ser.astype(dtype)
+    expected = pd.Series(["0.1"], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_numpy_returns_pdna_default(dtype):
+    arr = pd.array(["a", pd.NA, "b"], dtype=dtype)
+    result = np.array(arr)
+    expected = np.array(["a", dtype.na_value, "b"], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_na_value(dtype, nulls_fixture):
+    na_value = nulls_fixture
+    arr = pd.array(["a", pd.NA, "b"], dtype=dtype)
+    result = arr.to_numpy(na_value=na_value)
+    expected = np.array(["a", na_value, "b"], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_isin(dtype, fixed_now_ts):
+    s = pd.Series(["a", "b", None], dtype=dtype)
+
+    result = s.isin(["a", "c"])
+    expected = pd.Series([True, False, False])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin(["a", pd.NA])
+    expected = pd.Series([True, False, True])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin([])
+    expected = pd.Series([False, False, False])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin(["a", fixed_now_ts])
+    expected = pd.Series([True, False, False])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin([fixed_now_ts])
+    expected = pd.Series([False, False, False])
+    tm.assert_series_equal(result, expected)
+
+
+def test_isin_string_array(dtype, dtype2):
+    s = pd.Series(["a", "b", None], dtype=dtype)
+
+    result = s.isin(pd.array(["a", "c"], dtype=dtype2))
+    expected = pd.Series([True, False, False])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin(pd.array(["a", None], dtype=dtype2))
+    expected = pd.Series([True, False, True])
+    tm.assert_series_equal(result, expected)
+
+
+def test_isin_arrow_string_array(dtype):
+    pa = pytest.importorskip("pyarrow")
+    s = pd.Series(["a", "b", None], dtype=dtype)
+
+    result = s.isin(pd.array(["a", "c"], dtype=pd.ArrowDtype(pa.string())))
+    expected = pd.Series([True, False, False])
+    tm.assert_series_equal(result, expected)
+
+    result = s.isin(pd.array(["a", None], dtype=pd.ArrowDtype(pa.string())))
+    expected = pd.Series([True, False, True])
+    tm.assert_series_equal(result, expected)
+
+
+def test_setitem_scalar_with_mask_validation(dtype):
+    # https://github.com/pandas-dev/pandas/issues/47628
+    # setting None with a boolean mask (through _putmaks) should still result
+    # in pd.NA values in the underlying array
+    ser = pd.Series(["a", "b", "c"], dtype=dtype)
+    mask = np.array([False, True, False])
+
+    ser[mask] = None
+    assert ser.array[1] is ser.dtype.na_value
+
+    # for other non-string we should also raise an error
+    ser = pd.Series(["a", "b", "c"], dtype=dtype)
+    msg = "Invalid value '1' for dtype 'str"
+    with pytest.raises(TypeError, match=msg):
+        ser[mask] = 1
+
+
+def test_from_numpy_str(dtype):
+    vals = ["a", "b", "c"]
+    arr = np.array(vals, dtype=np.str_)
+    result = pd.array(arr, dtype=dtype)
+    expected = pd.array(vals, dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_tolist(dtype):
+    vals = ["a", "b", "c"]
+    arr = pd.array(vals, dtype=dtype)
+    result = arr.tolist()
+    expected = vals
+    tm.assert_equal(result, expected)
+
+
+def test_string_array_view_type_error():
+    arr = pd.array(["a", "b", "c"], dtype="string")
+    with pytest.raises(TypeError, match="Cannot change data-type for string array."):
+        arr.view("i8")
diff --git a/pandas/pandas/tests/arrays/string_/test_string_arrow.py b/pandas/pandas/tests/arrays/string_/test_string_arrow.py
new file mode 100644
index 0000000000000..e6103da5021bb
--- /dev/null
+++ b/pandas/pandas/tests/arrays/string_/test_string_arrow.py
@@ -0,0 +1,282 @@
+import pickle
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.string_ import (
+    StringArray,
+    StringDtype,
+)
+from pandas.core.arrays.string_arrow import (
+    ArrowStringArray,
+    ArrowStringArrayNumpySemantics,
+)
+
+
+def test_eq_all_na():
+    pytest.importorskip("pyarrow")
+    a = pd.array([pd.NA, pd.NA], dtype=StringDtype("pyarrow"))
+    result = a == a
+    expected = pd.array([pd.NA, pd.NA], dtype="boolean[pyarrow]")
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_config(string_storage, using_infer_string):
+    # with the default string_storage setting
+    # always "python" at the moment
+    assert StringDtype().storage == "python"
+
+    with pd.option_context("string_storage", string_storage):
+        assert StringDtype().storage == string_storage
+        result = pd.array(["a", "b"])
+        assert result.dtype.storage == string_storage
+
+    # pd.array(..) by default always returns the NA-variant
+    dtype = StringDtype(string_storage, na_value=pd.NA)
+    expected = dtype.construct_array_type()._from_sequence(["a", "b"], dtype=dtype)
+    tm.assert_equal(result, expected)
+
+
+def test_config_bad_storage_raises():
+    msg = re.escape("Value must be one of python|pyarrow")
+    with pytest.raises(ValueError, match=msg):
+        pd.options.mode.string_storage = "foo"
+
+
+@pytest.mark.parametrize("chunked", [True, False])
+@pytest.mark.parametrize("array_lib", ["numpy", "pyarrow"])
+def test_constructor_not_string_type_raises(array_lib, chunked):
+    pa = pytest.importorskip("pyarrow")
+
+    array_lib = pa if array_lib == "pyarrow" else np
+
+    arr = array_lib.array([1, 2, 3])
+    if chunked:
+        if array_lib is np:
+            pytest.skip("chunked not applicable to numpy array")
+        arr = pa.chunked_array(arr)
+    if array_lib is np:
+        msg = "Unsupported type '<class 'numpy.ndarray'>' for ArrowExtensionArray"
+    else:
+        msg = re.escape(
+            "ArrowStringArray requires a PyArrow (chunked) array of large_string type"
+        )
+    with pytest.raises(ValueError, match=msg):
+        ArrowStringArray(arr)
+
+
+@pytest.mark.parametrize("chunked", [True, False])
+def test_constructor_not_string_type_value_dictionary_raises(chunked):
+    pa = pytest.importorskip("pyarrow")
+
+    arr = pa.array([1, 2, 3], pa.dictionary(pa.int32(), pa.int32()))
+    if chunked:
+        arr = pa.chunked_array(arr)
+
+    msg = re.escape(
+        "ArrowStringArray requires a PyArrow (chunked) array of large_string type"
+    )
+    with pytest.raises(ValueError, match=msg):
+        ArrowStringArray(arr)
+
+
+@pytest.mark.parametrize("string_type", ["string", "large_string"])
+@pytest.mark.parametrize("chunked", [True, False])
+def test_constructor_valid_string_type_value_dictionary(string_type, chunked):
+    pa = pytest.importorskip("pyarrow")
+
+    arr = pa.array(["1", "2", "3"], getattr(pa, string_type)()).dictionary_encode()
+    if chunked:
+        arr = pa.chunked_array(arr)
+
+    arr = ArrowStringArray(arr)
+    # dictionary type get converted to dense large string array
+    assert pa.types.is_large_string(arr._pa_array.type)
+
+
+@pytest.mark.parametrize("chunked", [True, False])
+def test_constructor_valid_string_view(chunked):
+    # requires pyarrow>=18 for casting string_view to string
+    pa = pytest.importorskip("pyarrow", minversion="18")
+
+    arr = pa.array(["1", "2", "3"], pa.string_view())
+    if chunked:
+        arr = pa.chunked_array(arr)
+
+    arr = ArrowStringArray(arr)
+    # dictionary type get converted to dense large string array
+    assert pa.types.is_large_string(arr._pa_array.type)
+
+
+def test_constructor_from_list():
+    # GH#27673
+    pytest.importorskip("pyarrow")
+    result = pd.Series(["E"], dtype=StringDtype(storage="pyarrow"))
+    assert isinstance(result.dtype, StringDtype)
+    assert result.dtype.storage == "pyarrow"
+
+
+def test_from_sequence_wrong_dtype_raises(using_infer_string):
+    pytest.importorskip("pyarrow")
+    with pd.option_context("string_storage", "python"):
+        ArrowStringArray._from_sequence(["a", None, "c"], dtype="string")
+
+    with pd.option_context("string_storage", "pyarrow"):
+        ArrowStringArray._from_sequence(["a", None, "c"], dtype="string")
+
+    with pytest.raises(AssertionError, match=None):
+        ArrowStringArray._from_sequence(["a", None, "c"], dtype="string[python]")
+
+    ArrowStringArray._from_sequence(["a", None, "c"], dtype="string[pyarrow]")
+
+    if not using_infer_string:
+        with pytest.raises(AssertionError, match=None):
+            with pd.option_context("string_storage", "python"):
+                ArrowStringArray._from_sequence(["a", None, "c"], dtype=StringDtype())
+
+    with pd.option_context("string_storage", "pyarrow"):
+        ArrowStringArray._from_sequence(["a", None, "c"], dtype=StringDtype())
+
+    if not using_infer_string:
+        with pytest.raises(AssertionError, match=None):
+            ArrowStringArray._from_sequence(
+                ["a", None, "c"], dtype=StringDtype("python")
+            )
+
+    ArrowStringArray._from_sequence(["a", None, "c"], dtype=StringDtype("pyarrow"))
+
+    with pd.option_context("string_storage", "python"):
+        StringArray._from_sequence(["a", None, "c"], dtype="string")
+
+    with pd.option_context("string_storage", "pyarrow"):
+        StringArray._from_sequence(["a", None, "c"], dtype="string")
+
+    StringArray._from_sequence(["a", None, "c"], dtype="string[python]")
+
+    with pytest.raises(AssertionError, match=None):
+        StringArray._from_sequence(["a", None, "c"], dtype="string[pyarrow]")
+
+    if not using_infer_string:
+        with pd.option_context("string_storage", "python"):
+            StringArray._from_sequence(["a", None, "c"], dtype=StringDtype())
+
+    if not using_infer_string:
+        with pytest.raises(AssertionError, match=None):
+            with pd.option_context("string_storage", "pyarrow"):
+                StringArray._from_sequence(["a", None, "c"], dtype=StringDtype())
+
+    StringArray._from_sequence(["a", None, "c"], dtype=StringDtype("python"))
+
+    with pytest.raises(AssertionError, match=None):
+        StringArray._from_sequence(["a", None, "c"], dtype=StringDtype("pyarrow"))
+
+
+@td.skip_if_installed("pyarrow")
+def test_pyarrow_not_installed_raises():
+    msg = re.escape("pyarrow>=10.0.1 is required for PyArrow backed")
+
+    with pytest.raises(ImportError, match=msg):
+        StringDtype(storage="pyarrow")
+
+    with pytest.raises(ImportError, match=msg):
+        ArrowStringArray([])
+
+    with pytest.raises(ImportError, match=msg):
+        ArrowStringArrayNumpySemantics([])
+
+    with pytest.raises(ImportError, match=msg):
+        ArrowStringArray._from_sequence(["a", None, "b"])
+
+
+@pytest.mark.parametrize("multiple_chunks", [False, True])
+@pytest.mark.parametrize(
+    "key, value, expected",
+    [
+        (-1, "XX", ["a", "b", "c", "d", "XX"]),
+        (1, "XX", ["a", "XX", "c", "d", "e"]),
+        (1, None, ["a", None, "c", "d", "e"]),
+        (1, pd.NA, ["a", None, "c", "d", "e"]),
+        ([1, 3], "XX", ["a", "XX", "c", "XX", "e"]),
+        ([1, 3], ["XX", "YY"], ["a", "XX", "c", "YY", "e"]),
+        ([1, 3], ["XX", None], ["a", "XX", "c", None, "e"]),
+        ([1, 3], ["XX", pd.NA], ["a", "XX", "c", None, "e"]),
+        ([0, -1], ["XX", "YY"], ["XX", "b", "c", "d", "YY"]),
+        ([-1, 0], ["XX", "YY"], ["YY", "b", "c", "d", "XX"]),
+        (slice(3, None), "XX", ["a", "b", "c", "XX", "XX"]),
+        (slice(2, 4), ["XX", "YY"], ["a", "b", "XX", "YY", "e"]),
+        (slice(3, 1, -1), ["XX", "YY"], ["a", "b", "YY", "XX", "e"]),
+        (slice(None), "XX", ["XX", "XX", "XX", "XX", "XX"]),
+        ([False, True, False, True, False], ["XX", "YY"], ["a", "XX", "c", "YY", "e"]),
+    ],
+)
+def test_setitem(multiple_chunks, key, value, expected):
+    pa = pytest.importorskip("pyarrow")
+
+    result = pa.array(list("abcde"))
+    expected = pa.array(expected)
+
+    if multiple_chunks:
+        result = pa.chunked_array([result[:3], result[3:]])
+        expected = pa.chunked_array([expected[:3], expected[3:]])
+
+    result = ArrowStringArray(result)
+    expected = ArrowStringArray(expected)
+
+    result[key] = value
+    tm.assert_equal(result, expected)
+
+
+def test_setitem_invalid_indexer_raises():
+    pa = pytest.importorskip("pyarrow")
+
+    arr = ArrowStringArray(pa.array(list("abcde")))
+
+    with tm.external_error_raised(IndexError):
+        arr[5] = "foo"
+
+    with tm.external_error_raised(IndexError):
+        arr[-6] = "foo"
+
+    with tm.external_error_raised(IndexError):
+        arr[[0, 5]] = "foo"
+
+    with tm.external_error_raised(IndexError):
+        arr[[0, -6]] = "foo"
+
+    with tm.external_error_raised(IndexError):
+        arr[[True, True, False]] = "foo"
+
+    with tm.external_error_raised(ValueError):
+        arr[[0, 1]] = ["foo", "bar", "baz"]
+
+
+@pytest.mark.parametrize("na_value", [pd.NA, np.nan])
+def test_pickle_roundtrip(na_value):
+    # GH 42600
+    pytest.importorskip("pyarrow")
+    dtype = StringDtype("pyarrow", na_value=na_value)
+    expected = pd.Series(range(10), dtype=dtype)
+    expected_sliced = expected.head(2)
+    full_pickled = pickle.dumps(expected)
+    sliced_pickled = pickle.dumps(expected_sliced)
+
+    assert len(full_pickled) > len(sliced_pickled)
+
+    result = pickle.loads(full_pickled)
+    tm.assert_series_equal(result, expected)
+
+    result_sliced = pickle.loads(sliced_pickled)
+    tm.assert_series_equal(result_sliced, expected_sliced)
+
+
+def test_string_dtype_error_message():
+    # GH#55051
+    pytest.importorskip("pyarrow")
+    msg = "Storage must be 'python' or 'pyarrow'."
+    with pytest.raises(ValueError, match=msg):
+        StringDtype("bla")
diff --git a/pandas/pandas/tests/arrays/test_array.py b/pandas/pandas/tests/arrays/test_array.py
new file mode 100644
index 0000000000000..3c0ef1e4d928b
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_array.py
@@ -0,0 +1,532 @@
+import datetime
+import decimal
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.extensions import register_extension_dtype
+from pandas.arrays import (
+    BooleanArray,
+    DatetimeArray,
+    FloatingArray,
+    IntegerArray,
+    IntervalArray,
+    SparseArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays import (
+    NumpyExtensionArray,
+    period_array,
+)
+from pandas.tests.extension.decimal import (
+    DecimalArray,
+    DecimalDtype,
+    to_decimal,
+)
+
+
+@pytest.mark.parametrize("dtype_unit", ["M8[h]", "M8[m]", "m8[h]"])
+def test_dt64_array(dtype_unit):
+    # GH#53817
+    dtype_var = np.dtype(dtype_unit)
+    msg = (
+        r"datetime64 and timedelta64 dtype resolutions other than "
+        r"'s', 'ms', 'us', and 'ns' are no longer supported."
+    )
+    with pytest.raises(ValueError, match=msg):
+        pd.array([], dtype=dtype_var)
+
+
+@pytest.mark.parametrize(
+    "data, dtype, expected",
+    [
+        # Basic NumPy defaults.
+        ([], None, FloatingArray._from_sequence([], dtype="Float64")),
+        ([1, 2], None, IntegerArray._from_sequence([1, 2], dtype="Int64")),
+        ([1, 2], object, NumpyExtensionArray(np.array([1, 2], dtype=object))),
+        (
+            [1, 2],
+            np.dtype("float32"),
+            NumpyExtensionArray(np.array([1.0, 2.0], dtype=np.dtype("float32"))),
+        ),
+        (
+            np.array([], dtype=object),
+            None,
+            NumpyExtensionArray(np.array([], dtype=object)),
+        ),
+        (
+            np.array([1, 2], dtype="int64"),
+            None,
+            IntegerArray._from_sequence([1, 2], dtype="Int64"),
+        ),
+        (
+            np.array([1.0, 2.0], dtype="float64"),
+            None,
+            FloatingArray._from_sequence([1.0, 2.0], dtype="Float64"),
+        ),
+        # String alias passes through to NumPy
+        ([1, 2], "float32", NumpyExtensionArray(np.array([1, 2], dtype="float32"))),
+        ([1, 2], "int64", NumpyExtensionArray(np.array([1, 2], dtype=np.int64))),
+        # GH#44715 FloatingArray does not support float16, so fall
+        #  back to NumpyExtensionArray
+        (
+            np.array([1, 2], dtype=np.float16),
+            None,
+            NumpyExtensionArray(np.array([1, 2], dtype=np.float16)),
+        ),
+        # idempotency with e.g. pd.array(pd.array([1, 2], dtype="int64"))
+        (
+            NumpyExtensionArray(np.array([1, 2], dtype=np.int32)),
+            None,
+            NumpyExtensionArray(np.array([1, 2], dtype=np.int32)),
+        ),
+        # Period alias
+        (
+            [pd.Period("2000", "D"), pd.Period("2001", "D")],
+            "Period[D]",
+            period_array(["2000", "2001"], freq="D"),
+        ),
+        # Period dtype
+        (
+            [pd.Period("2000", "D")],
+            pd.PeriodDtype("D"),
+            period_array(["2000"], freq="D"),
+        ),
+        # Datetime (naive)
+        (
+            [1, 2],
+            np.dtype("datetime64[ns]"),
+            DatetimeArray._from_sequence(
+                np.array([1, 2], dtype="M8[ns]"), dtype="M8[ns]"
+            ),
+        ),
+        (
+            [1, 2],
+            np.dtype("datetime64[s]"),
+            DatetimeArray._from_sequence(
+                np.array([1, 2], dtype="M8[s]"), dtype="M8[s]"
+            ),
+        ),
+        (
+            np.array([1, 2], dtype="datetime64[ns]"),
+            None,
+            DatetimeArray._from_sequence(
+                np.array([1, 2], dtype="M8[ns]"), dtype="M8[ns]"
+            ),
+        ),
+        (
+            pd.DatetimeIndex(["2000", "2001"]),
+            np.dtype("datetime64[ns]"),
+            DatetimeArray._from_sequence(["2000", "2001"], dtype="M8[ns]"),
+        ),
+        (
+            pd.DatetimeIndex(["2000", "2001"]),
+            None,
+            DatetimeArray._from_sequence(["2000", "2001"], dtype="M8[s]"),
+        ),
+        (
+            ["2000", "2001"],
+            np.dtype("datetime64[ns]"),
+            DatetimeArray._from_sequence(["2000", "2001"], dtype="M8[ns]"),
+        ),
+        # Datetime (tz-aware)
+        (
+            ["2000", "2001"],
+            pd.DatetimeTZDtype(tz="CET"),
+            DatetimeArray._from_sequence(
+                ["2000", "2001"], dtype=pd.DatetimeTZDtype(tz="CET")
+            ),
+        ),
+        # Timedelta
+        (
+            ["1h", "2h"],
+            np.dtype("timedelta64[ns]"),
+            TimedeltaArray._from_sequence(["1h", "2h"], dtype="m8[ns]"),
+        ),
+        (
+            pd.TimedeltaIndex(["1h", "2h"]),
+            np.dtype("timedelta64[ns]"),
+            TimedeltaArray._from_sequence(["1h", "2h"], dtype="m8[ns]"),
+        ),
+        (
+            np.array([1, 2], dtype="m8[s]"),
+            np.dtype("timedelta64[s]"),
+            TimedeltaArray._from_sequence(
+                np.array([1, 2], dtype="m8[s]"), dtype="m8[s]"
+            ),
+        ),
+        (
+            pd.TimedeltaIndex(["1h", "2h"]),
+            None,
+            TimedeltaArray._from_sequence(["1h", "2h"], dtype="m8[ns]"),
+        ),
+        (
+            # preserve non-nano, i.e. don't cast to NumpyExtensionArray
+            TimedeltaArray._simple_new(
+                np.arange(5, dtype=np.int64).view("m8[s]"), dtype=np.dtype("m8[s]")
+            ),
+            None,
+            TimedeltaArray._simple_new(
+                np.arange(5, dtype=np.int64).view("m8[s]"), dtype=np.dtype("m8[s]")
+            ),
+        ),
+        (
+            # preserve non-nano, i.e. don't cast to NumpyExtensionArray
+            TimedeltaArray._simple_new(
+                np.arange(5, dtype=np.int64).view("m8[s]"), dtype=np.dtype("m8[s]")
+            ),
+            np.dtype("m8[s]"),
+            TimedeltaArray._simple_new(
+                np.arange(5, dtype=np.int64).view("m8[s]"), dtype=np.dtype("m8[s]")
+            ),
+        ),
+        # Category
+        (["a", "b"], "category", pd.Categorical(["a", "b"])),
+        (
+            ["a", "b"],
+            pd.CategoricalDtype(None, ordered=True),
+            pd.Categorical(["a", "b"], ordered=True),
+        ),
+        # Interval
+        (
+            [pd.Interval(1, 2), pd.Interval(3, 4)],
+            "interval",
+            IntervalArray.from_tuples([(1, 2), (3, 4)]),
+        ),
+        # Sparse
+        ([0, 1], "Sparse[int64]", SparseArray([0, 1], dtype="int64")),
+        # IntegerNA
+        ([1, None], "Int16", pd.array([1, None], dtype="Int16")),
+        (
+            pd.Series([1, 2]),
+            None,
+            NumpyExtensionArray(np.array([1, 2], dtype=np.int64)),
+        ),
+        # String
+        (
+            ["a", None],
+            "string",
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", None], dtype=pd.StringDtype()),
+        ),
+        (
+            ["a", None],
+            "str",
+            pd.StringDtype(na_value=np.nan)
+            .construct_array_type()
+            ._from_sequence(["a", None], dtype=pd.StringDtype(na_value=np.nan))
+            if using_string_dtype()
+            else NumpyExtensionArray(np.array(["a", "None"])),
+        ),
+        (
+            ["a", None],
+            pd.StringDtype(),
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", None], dtype=pd.StringDtype()),
+        ),
+        (
+            ["a", None],
+            pd.StringDtype(na_value=np.nan),
+            pd.StringDtype(na_value=np.nan)
+            .construct_array_type()
+            ._from_sequence(["a", None], dtype=pd.StringDtype(na_value=np.nan)),
+        ),
+        (
+            # numpy array with string dtype
+            np.array(["a", "b"], dtype=str),
+            pd.StringDtype(),
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", "b"], dtype=pd.StringDtype()),
+        ),
+        (
+            # numpy array with string dtype
+            np.array(["a", "b"], dtype=str),
+            pd.StringDtype(na_value=np.nan),
+            pd.StringDtype(na_value=np.nan)
+            .construct_array_type()
+            ._from_sequence(["a", "b"], dtype=pd.StringDtype(na_value=np.nan)),
+        ),
+        # Boolean
+        (
+            [True, None],
+            "boolean",
+            BooleanArray._from_sequence([True, None], dtype="boolean"),
+        ),
+        (
+            [True, None],
+            pd.BooleanDtype(),
+            BooleanArray._from_sequence([True, None], dtype="boolean"),
+        ),
+        # Index
+        (pd.Index([1, 2]), None, NumpyExtensionArray(np.array([1, 2], dtype=np.int64))),
+        # Series[EA] returns the EA
+        (
+            pd.Series(pd.Categorical(["a", "b"], categories=["a", "b", "c"])),
+            None,
+            pd.Categorical(["a", "b"], categories=["a", "b", "c"]),
+        ),
+        # "3rd party" EAs work
+        ([decimal.Decimal(0), decimal.Decimal(1)], "decimal", to_decimal([0, 1])),
+        # pass an ExtensionArray, but a different dtype
+        (
+            period_array(["2000", "2001"], freq="D"),
+            "category",
+            pd.Categorical([pd.Period("2000", "D"), pd.Period("2001", "D")]),
+        ),
+        # Complex
+        (
+            np.array([complex(1), complex(2)], dtype=np.complex128),
+            None,
+            NumpyExtensionArray(
+                np.array([complex(1), complex(2)], dtype=np.complex128)
+            ),
+        ),
+    ],
+)
+def test_array(data, dtype, expected):
+    result = pd.array(data, dtype=dtype)
+    tm.assert_equal(result, expected)
+
+
+def test_array_copy():
+    a = np.array([1, 2])
+    # default is to copy
+    b = pd.array(a, dtype=a.dtype)
+    assert not tm.shares_memory(a, b)
+
+    # copy=True
+    b = pd.array(a, dtype=a.dtype, copy=True)
+    assert not tm.shares_memory(a, b)
+
+    # copy=False
+    b = pd.array(a, dtype=a.dtype, copy=False)
+    assert tm.shares_memory(a, b)
+
+
+@pytest.mark.parametrize(
+    "data, expected",
+    [
+        # period
+        (
+            [pd.Period("2000", "D"), pd.Period("2001", "D")],
+            period_array(["2000", "2001"], freq="D"),
+        ),
+        # interval
+        ([pd.Interval(0, 1), pd.Interval(1, 2)], IntervalArray.from_breaks([0, 1, 2])),
+        # datetime
+        (
+            [pd.Timestamp("2000"), pd.Timestamp("2001")],
+            DatetimeArray._from_sequence(["2000", "2001"], dtype="M8[s]"),
+        ),
+        (
+            [datetime.datetime(2000, 1, 1), datetime.datetime(2001, 1, 1)],
+            DatetimeArray._from_sequence(["2000", "2001"], dtype="M8[us]"),
+        ),
+        (
+            np.array([1, 2], dtype="M8[ns]"),
+            DatetimeArray._from_sequence(np.array([1, 2], dtype="M8[ns]")),
+        ),
+        (
+            np.array([1, 2], dtype="M8[us]"),
+            DatetimeArray._simple_new(
+                np.array([1, 2], dtype="M8[us]"), dtype=np.dtype("M8[us]")
+            ),
+        ),
+        # datetimetz
+        (
+            [pd.Timestamp("2000", tz="CET"), pd.Timestamp("2001", tz="CET")],
+            DatetimeArray._from_sequence(
+                ["2000", "2001"], dtype=pd.DatetimeTZDtype(tz="CET", unit="s")
+            ),
+        ),
+        (
+            [
+                datetime.datetime(
+                    2000, 1, 1, tzinfo=zoneinfo.ZoneInfo("Europe/Berlin")
+                ),
+                datetime.datetime(
+                    2001, 1, 1, tzinfo=zoneinfo.ZoneInfo("Europe/Berlin")
+                ),
+            ],
+            DatetimeArray._from_sequence(
+                ["2000", "2001"],
+                dtype=pd.DatetimeTZDtype(
+                    tz=zoneinfo.ZoneInfo("Europe/Berlin"), unit="us"
+                ),
+            ),
+        ),
+        # timedelta
+        (
+            [pd.Timedelta("1h"), pd.Timedelta("2h")],
+            TimedeltaArray._from_sequence(["1h", "2h"], dtype="m8[ns]"),
+        ),
+        (
+            np.array([1, 2], dtype="m8[ns]"),
+            TimedeltaArray._from_sequence(
+                np.array([1, 2], dtype="m8[ns]"), dtype=np.dtype("m8[ns]")
+            ),
+        ),
+        (
+            np.array([1, 2], dtype="m8[us]"),
+            TimedeltaArray._from_sequence(
+                np.array([1, 2], dtype="m8[us]"), dtype=np.dtype("m8[us]")
+            ),
+        ),
+        # integer
+        ([1, 2], IntegerArray._from_sequence([1, 2], dtype="Int64")),
+        ([1, None], IntegerArray._from_sequence([1, None], dtype="Int64")),
+        ([1, pd.NA], IntegerArray._from_sequence([1, pd.NA], dtype="Int64")),
+        ([1, np.nan], IntegerArray._from_sequence([1, np.nan], dtype="Int64")),
+        # float
+        ([0.1, 0.2], FloatingArray._from_sequence([0.1, 0.2], dtype="Float64")),
+        ([0.1, None], FloatingArray._from_sequence([0.1, pd.NA], dtype="Float64")),
+        ([0.1, np.nan], FloatingArray._from_sequence([0.1, pd.NA], dtype="Float64")),
+        ([0.1, pd.NA], FloatingArray._from_sequence([0.1, pd.NA], dtype="Float64")),
+        # integer-like float
+        ([1.0, 2.0], FloatingArray._from_sequence([1.0, 2.0], dtype="Float64")),
+        ([1.0, None], FloatingArray._from_sequence([1.0, pd.NA], dtype="Float64")),
+        ([1.0, np.nan], FloatingArray._from_sequence([1.0, pd.NA], dtype="Float64")),
+        ([1.0, pd.NA], FloatingArray._from_sequence([1.0, pd.NA], dtype="Float64")),
+        # mixed-integer-float
+        ([1, 2.0], FloatingArray._from_sequence([1.0, 2.0], dtype="Float64")),
+        (
+            [1, np.nan, 2.0],
+            FloatingArray._from_sequence([1.0, None, 2.0], dtype="Float64"),
+        ),
+        # string
+        (
+            ["a", "b"],
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", "b"], dtype=pd.StringDtype()),
+        ),
+        (
+            ["a", None],
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", None], dtype=pd.StringDtype()),
+        ),
+        (
+            # numpy array with string dtype
+            np.array(["a", "b"], dtype=str),
+            pd.StringDtype()
+            .construct_array_type()
+            ._from_sequence(["a", "b"], dtype=pd.StringDtype()),
+        ),
+        # Boolean
+        ([True, False], BooleanArray._from_sequence([True, False], dtype="boolean")),
+        ([True, None], BooleanArray._from_sequence([True, None], dtype="boolean")),
+    ],
+)
+def test_array_inference(data, expected):
+    result = pd.array(data)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        # mix of frequencies
+        [pd.Period("2000", "D"), pd.Period("2001", "Y")],
+        # mix of closed
+        [pd.Interval(0, 1, closed="left"), pd.Interval(1, 2, closed="right")],
+        # Mix of timezones
+        [pd.Timestamp("2000", tz="CET"), pd.Timestamp("2000", tz="UTC")],
+        # Mix of tz-aware and tz-naive
+        [pd.Timestamp("2000", tz="CET"), pd.Timestamp("2000")],
+        np.array([pd.Timestamp("2000"), pd.Timestamp("2000", tz="CET")]),
+    ],
+)
+def test_array_inference_fails(data):
+    result = pd.array(data)
+    expected = NumpyExtensionArray(np.array(data, dtype=object))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("data", [np.array(0)])
+def test_nd_raises(data):
+    with pytest.raises(ValueError, match="NumpyExtensionArray must be 1-dimensional"):
+        pd.array(data, dtype="int64")
+
+
+def test_scalar_raises():
+    with pytest.raises(ValueError, match="Cannot pass scalar '1'"):
+        pd.array(1)
+
+
+def test_dataframe_raises():
+    # GH#51167 don't accidentally cast to StringArray by doing inference on columns
+    df = pd.DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+    msg = "Cannot pass DataFrame to 'pandas.array'"
+    with pytest.raises(TypeError, match=msg):
+        pd.array(df)
+
+
+def test_bounds_check():
+    # GH21796
+    with pytest.raises(
+        TypeError, match=r"cannot safely cast non-equivalent int(32|64) to uint16"
+    ):
+        pd.array([-1, 2, 3], dtype="UInt16")
+
+
+# ---------------------------------------------------------------------------
+# A couple dummy classes to ensure that Series and Indexes are unboxed before
+# getting to the EA classes.
+
+
+@register_extension_dtype
+class DecimalDtype2(DecimalDtype):
+    name = "decimal2"
+
+    @classmethod
+    def construct_array_type(cls):
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return DecimalArray2
+
+
+class DecimalArray2(DecimalArray):
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        if isinstance(scalars, (pd.Series, pd.Index)):
+            raise TypeError("scalars should not be of type pd.Series or pd.Index")
+
+        return super()._from_sequence(scalars, dtype=dtype, copy=copy)
+
+
+def test_array_unboxes(index_or_series):
+    box = index_or_series
+
+    data = box([decimal.Decimal("1"), decimal.Decimal("2")])
+    dtype = DecimalDtype2()
+    # make sure it works
+    with pytest.raises(
+        TypeError, match="scalars should not be of type pd.Series or pd.Index"
+    ):
+        DecimalArray2._from_sequence(data, dtype=dtype)
+
+    result = pd.array(data, dtype="decimal2")
+    expected = DecimalArray2._from_sequence(data.values, dtype=dtype)
+    tm.assert_equal(result, expected)
+
+
+def test_array_to_numpy_na():
+    # GH#40638
+    arr = pd.array([pd.NA, 1], dtype="string[python]")
+    result = arr.to_numpy(na_value=True, dtype=bool)
+    expected = np.array([True, True])
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/test_datetimelike.py b/pandas/pandas/tests/arrays/test_datetimelike.py
new file mode 100644
index 0000000000000..d1ef29b0bf8a0
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_datetimelike.py
@@ -0,0 +1,1358 @@
+from __future__ import annotations
+
+import re
+import warnings
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    NaT,
+    OutOfBoundsDatetime,
+    Timestamp,
+)
+from pandas._libs.tslibs import to_offset
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Period,
+    PeriodIndex,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    DatetimeArray,
+    NumpyExtensionArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+
+
+# TODO: more freq variants
+@pytest.fixture(params=["D", "B", "W", "ME", "QE", "YE"])
+def freqstr(request):
+    """Fixture returning parametrized frequency in string format."""
+    return request.param
+
+
+@pytest.fixture
+def period_index(freqstr):
+    """
+    A fixture to provide PeriodIndex objects with different frequencies.
+
+    Most PeriodArray behavior is already tested in PeriodIndex tests,
+    so here we just test that the PeriodArray behavior matches
+    the PeriodIndex behavior.
+    """
+    # TODO: non-monotone indexes; NaTs, different start dates
+    with warnings.catch_warnings():
+        # suppress deprecation of Period[B]
+        warnings.filterwarnings(
+            "ignore", message="Period with BDay freq", category=FutureWarning
+        )
+        freqstr = PeriodDtype(to_offset(freqstr))._freqstr
+        pi = pd.period_range(start=Timestamp("2000-01-01"), periods=100, freq=freqstr)
+    return pi
+
+
+@pytest.fixture
+def datetime_index(freqstr):
+    """
+    A fixture to provide DatetimeIndex objects with different frequencies.
+
+    Most DatetimeArray behavior is already tested in DatetimeIndex tests,
+    so here we just test that the DatetimeArray behavior matches
+    the DatetimeIndex behavior.
+    """
+    # TODO: non-monotone indexes; NaTs, different start dates, timezones
+    dti = pd.date_range(start=Timestamp("2000-01-01"), periods=100, freq=freqstr)
+    return dti
+
+
+@pytest.fixture
+def timedelta_index():
+    """
+    A fixture to provide TimedeltaIndex objects with different frequencies.
+     Most TimedeltaArray behavior is already tested in TimedeltaIndex tests,
+    so here we just test that the TimedeltaArray behavior matches
+    the TimedeltaIndex behavior.
+    """
+    # TODO: flesh this out
+    return TimedeltaIndex(["1 Day", "3 Hours", "NaT"])
+
+
+class SharedTests:
+    index_cls: type[DatetimeIndex | PeriodIndex | TimedeltaIndex]
+
+    @pytest.fixture
+    def arr1d(self):
+        """Fixture returning DatetimeArray with daily frequency."""
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(data, freq="D")
+        else:
+            arr = self.index_cls(data, freq="D")._data
+        return arr
+
+    def test_compare_len1_raises(self, arr1d):
+        # make sure we raise when comparing with different lengths, specific
+        #  to the case where one has length-1, which numpy would broadcast
+        arr = arr1d
+        idx = self.index_cls(arr)
+
+        with pytest.raises(ValueError, match="Lengths must match"):
+            arr == arr[:1]
+
+        # test the index classes while we're at it, GH#23078
+        with pytest.raises(ValueError, match="Lengths must match"):
+            idx <= idx[[0]]
+
+    @pytest.mark.parametrize(
+        "result",
+        [
+            pd.date_range("2020", periods=3),
+            pd.date_range("2020", periods=3, tz="UTC"),
+            pd.timedelta_range("0 days", periods=3),
+            pd.period_range("2020Q1", periods=3, freq="Q"),
+        ],
+    )
+    def test_compare_with_Categorical(self, result):
+        expected = pd.Categorical(result)
+        assert all(result == expected)
+        assert not any(result != expected)
+
+    @pytest.mark.parametrize("reverse", [True, False])
+    @pytest.mark.parametrize("as_index", [True, False])
+    def test_compare_categorical_dtype(self, arr1d, as_index, reverse, ordered):
+        other = pd.Categorical(arr1d, ordered=ordered)
+        if as_index:
+            other = pd.CategoricalIndex(other)
+
+        left, right = arr1d, other
+        if reverse:
+            left, right = right, left
+
+        ones = np.ones(arr1d.shape, dtype=bool)
+        zeros = ~ones
+
+        result = left == right
+        tm.assert_numpy_array_equal(result, ones)
+
+        result = left != right
+        tm.assert_numpy_array_equal(result, zeros)
+
+        if not reverse and not as_index:
+            # Otherwise Categorical raises TypeError bc it is not ordered
+            # TODO: we should probably get the same behavior regardless?
+            result = left < right
+            tm.assert_numpy_array_equal(result, zeros)
+
+            result = left <= right
+            tm.assert_numpy_array_equal(result, ones)
+
+            result = left > right
+            tm.assert_numpy_array_equal(result, zeros)
+
+            result = left >= right
+            tm.assert_numpy_array_equal(result, ones)
+
+    def test_take(self):
+        data = np.arange(100, dtype="i8") * 24 * 3600 * 10**9
+        np.random.default_rng(2).shuffle(data)
+
+        if self.array_cls is PeriodArray:
+            arr = PeriodArray(data, dtype="period[D]")
+        else:
+            arr = self.index_cls(data)._data
+        idx = self.index_cls._simple_new(arr)
+
+        takers = [1, 4, 94]
+        result = arr.take(takers)
+        expected = idx.take(takers)
+
+        tm.assert_index_equal(self.index_cls(result), expected)
+
+        takers = np.array([1, 4, 94])
+        result = arr.take(takers)
+        expected = idx.take(takers)
+
+        tm.assert_index_equal(self.index_cls(result), expected)
+
+    @pytest.mark.parametrize("fill_value", [2, 2.0, Timestamp(2021, 1, 1, 12).time])
+    def test_take_fill_raises(self, fill_value, arr1d):
+        msg = f"value should be a '{arr1d._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            arr1d.take([0, 1], allow_fill=True, fill_value=fill_value)
+
+    def test_take_fill(self, arr1d):
+        arr = arr1d
+
+        result = arr.take([-1, 1], allow_fill=True, fill_value=None)
+        assert result[0] is NaT
+
+        result = arr.take([-1, 1], allow_fill=True, fill_value=np.nan)
+        assert result[0] is NaT
+
+        result = arr.take([-1, 1], allow_fill=True, fill_value=NaT)
+        assert result[0] is NaT
+
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_take_fill_str(self, arr1d):
+        # Cast str fill_value matching other fill_value-taking methods
+        result = arr1d.take([-1, 1], allow_fill=True, fill_value=str(arr1d[-1]))
+        expected = arr1d[[-1, 1]]
+        tm.assert_equal(result, expected)
+
+        msg = f"value should be a '{arr1d._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            arr1d.take([-1, 1], allow_fill=True, fill_value="foo")
+
+    def test_concat_same_type(self, arr1d):
+        arr = arr1d
+        idx = self.index_cls(arr)
+        idx = idx.insert(0, NaT)
+        arr = arr1d
+
+        result = arr._concat_same_type([arr[:-1], arr[1:], arr])
+        arr2 = arr.astype(object)
+        expected = self.index_cls(np.concatenate([arr2[:-1], arr2[1:], arr2]))
+
+        tm.assert_index_equal(self.index_cls(result), expected)
+
+    def test_unbox_scalar(self, arr1d):
+        result = arr1d._unbox_scalar(arr1d[0])
+        expected = arr1d._ndarray.dtype.type
+        assert isinstance(result, expected)
+
+        result = arr1d._unbox_scalar(NaT)
+        assert isinstance(result, expected)
+
+        msg = f"'value' should be a {self.scalar_type.__name__}."
+        with pytest.raises(ValueError, match=msg):
+            arr1d._unbox_scalar("foo")
+
+    def test_check_compatible_with(self, arr1d):
+        arr1d._check_compatible_with(arr1d[0])
+        arr1d._check_compatible_with(arr1d[:1])
+        arr1d._check_compatible_with(NaT)
+
+    def test_scalar_from_string(self, arr1d):
+        result = arr1d._scalar_from_string(str(arr1d[0]))
+        assert result == arr1d[0]
+
+    def test_reduce_invalid(self, arr1d):
+        msg = "does not support operation 'not a method'"
+        with pytest.raises(TypeError, match=msg):
+            arr1d._reduce("not a method")
+
+    @pytest.mark.parametrize("method", ["pad", "backfill"])
+    def test_fillna_method_doesnt_change_orig(self, method):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(data, dtype="period[D]")
+        else:
+            dtype = "M8[ns]" if self.array_cls is DatetimeArray else "m8[ns]"
+            arr = self.array_cls._from_sequence(data, dtype=np.dtype(dtype))
+        arr[4] = NaT
+
+        fill_value = arr[3] if method == "pad" else arr[5]
+
+        result = arr._pad_or_backfill(method=method)
+        assert result[4] == fill_value
+
+        # check that the original was not changed
+        assert arr[4] is NaT
+
+    def test_searchsorted(self):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(data, dtype="period[D]")
+        else:
+            dtype = "M8[ns]" if self.array_cls is DatetimeArray else "m8[ns]"
+            arr = self.array_cls._from_sequence(data, dtype=np.dtype(dtype))
+
+        # scalar
+        result = arr.searchsorted(arr[1])
+        assert result == 1
+
+        result = arr.searchsorted(arr[2], side="right")
+        assert result == 3
+
+        # own-type
+        result = arr.searchsorted(arr[1:3])
+        expected = np.array([1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = arr.searchsorted(arr[1:3], side="right")
+        expected = np.array([2, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # GH#29884 match numpy convention on whether NaT goes
+        #  at the end or the beginning
+        result = arr.searchsorted(NaT)
+        assert result == 10
+
+    @pytest.mark.parametrize("box", [None, "index", "series"])
+    def test_searchsorted_castable_strings(self, arr1d, box, string_storage):
+        arr = arr1d
+        if box is None:
+            pass
+        elif box == "index":
+            # Test the equivalent Index.searchsorted method while we're here
+            arr = self.index_cls(arr)
+        else:
+            # Test the equivalent Series.searchsorted method while we're here
+            arr = pd.Series(arr)
+
+        # scalar
+        result = arr.searchsorted(str(arr[1]))
+        assert result == 1
+
+        result = arr.searchsorted(str(arr[2]), side="right")
+        assert result == 3
+
+        result = arr.searchsorted([str(x) for x in arr[1:3]])
+        expected = np.array([1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        with pytest.raises(
+            TypeError,
+            match=re.escape(
+                f"value should be a '{arr1d._scalar_type.__name__}', 'NaT', "
+                "or array of those. Got 'str' instead."
+            ),
+        ):
+            arr.searchsorted("foo")
+
+        with pd.option_context("string_storage", string_storage):
+            with pytest.raises(
+                TypeError,
+                match=re.escape(
+                    f"value should be a '{arr1d._scalar_type.__name__}', 'NaT', "
+                    "or array of those. Got string array instead."
+                ),
+            ):
+                arr.searchsorted([str(arr[1]), "baz"])
+
+    def test_getitem_near_implementation_bounds(self):
+        # We only check tz-naive for DTA bc the bounds are slightly different
+        #  for other tzs
+        i8vals = np.asarray([NaT._value + n for n in range(1, 5)], dtype="i8")
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(i8vals, dtype="period[ns]")
+        else:
+            arr = self.index_cls(i8vals, freq="ns")._data
+        arr[0]  # should not raise OutOfBoundsDatetime
+
+        index = pd.Index(arr)
+        index[0]  # should not raise OutOfBoundsDatetime
+
+        ser = pd.Series(arr)
+        ser[0]  # should not raise OutOfBoundsDatetime
+
+    def test_getitem_2d(self, arr1d):
+        # 2d slicing on a 1D array
+        expected = type(arr1d)._simple_new(
+            arr1d._ndarray[:, np.newaxis], dtype=arr1d.dtype
+        )
+        result = arr1d[:, np.newaxis]
+        tm.assert_equal(result, expected)
+
+        # Lookup on a 2D array
+        arr2d = expected
+        expected = type(arr2d)._simple_new(arr2d._ndarray[:3, 0], dtype=arr2d.dtype)
+        result = arr2d[:3, 0]
+        tm.assert_equal(result, expected)
+
+        # Scalar lookup
+        result = arr2d[-1, 0]
+        expected = arr1d[-1]
+        assert result == expected
+
+    def test_iter_2d(self, arr1d):
+        data2d = arr1d._ndarray[:3, np.newaxis]
+        arr2d = type(arr1d)._simple_new(data2d, dtype=arr1d.dtype)
+        result = list(arr2d)
+        assert len(result) == 3
+        for x in result:
+            assert isinstance(x, type(arr1d))
+            assert x.ndim == 1
+            assert x.dtype == arr1d.dtype
+
+    def test_repr_2d(self, arr1d):
+        data2d = arr1d._ndarray[:3, np.newaxis]
+        arr2d = type(arr1d)._simple_new(data2d, dtype=arr1d.dtype)
+
+        result = repr(arr2d)
+
+        if isinstance(arr2d, TimedeltaArray):
+            expected = (
+                f"<{type(arr2d).__name__}>\n"
+                "[\n"
+                f"['{arr1d[0]._repr_base()}'],\n"
+                f"['{arr1d[1]._repr_base()}'],\n"
+                f"['{arr1d[2]._repr_base()}']\n"
+                "]\n"
+                f"Shape: (3, 1), dtype: {arr1d.dtype}"
+            )
+        else:
+            expected = (
+                f"<{type(arr2d).__name__}>\n"
+                "[\n"
+                f"['{arr1d[0]}'],\n"
+                f"['{arr1d[1]}'],\n"
+                f"['{arr1d[2]}']\n"
+                "]\n"
+                f"Shape: (3, 1), dtype: {arr1d.dtype}"
+            )
+
+        assert result == expected
+
+    def test_setitem(self):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(data, dtype="period[D]")
+        else:
+            arr = self.index_cls(data, freq="D")._data
+
+        arr[0] = arr[1]
+        expected = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        expected[0] = expected[1]
+
+        tm.assert_numpy_array_equal(arr.asi8, expected)
+
+        arr[:2] = arr[-2:]
+        expected[:2] = expected[-2:]
+        tm.assert_numpy_array_equal(arr.asi8, expected)
+
+    @pytest.mark.parametrize(
+        "box",
+        [
+            pd.Index,
+            pd.Series,
+            np.array,
+            list,
+            NumpyExtensionArray,
+        ],
+    )
+    def test_setitem_object_dtype(self, box, arr1d):
+        expected = arr1d.copy()[::-1]
+        if expected.dtype.kind in ["m", "M"]:
+            expected = expected._with_freq(None)
+
+        vals = expected
+        if box is list:
+            vals = list(vals)
+        elif box is np.array:
+            # if we do np.array(x).astype(object) then dt64 and td64 cast to ints
+            vals = np.array(vals.astype(object))
+        elif box is NumpyExtensionArray:
+            vals = box(np.asarray(vals, dtype=object))
+        else:
+            vals = box(vals).astype(object)
+
+        arr1d[:] = vals
+
+        tm.assert_equal(arr1d, expected)
+
+    def test_setitem_strs(self, arr1d):
+        # Check that we parse strs in both scalar and listlike
+
+        # Setting list-like of strs
+        expected = arr1d.copy()
+        expected[[0, 1]] = arr1d[-2:]
+
+        result = arr1d.copy()
+        result[:2] = [str(x) for x in arr1d[-2:]]
+        tm.assert_equal(result, expected)
+
+        # Same thing but now for just a scalar str
+        expected = arr1d.copy()
+        expected[0] = arr1d[-1]
+
+        result = arr1d.copy()
+        result[0] = str(arr1d[-1])
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("as_index", [True, False])
+    def test_setitem_categorical(self, arr1d, as_index):
+        expected = arr1d.copy()[::-1]
+        if not isinstance(expected, PeriodArray):
+            expected = expected._with_freq(None)
+
+        cat = pd.Categorical(arr1d)
+        if as_index:
+            cat = pd.CategoricalIndex(cat)
+
+        arr1d[:] = cat[::-1]
+
+        tm.assert_equal(arr1d, expected)
+
+    def test_setitem_raises(self, arr1d):
+        arr = arr1d[:10]
+        val = arr[0]
+
+        with pytest.raises(IndexError, match="index 12 is out of bounds"):
+            arr[12] = val
+
+        with pytest.raises(TypeError, match="value should be a.* 'object'"):
+            arr[0] = object()
+
+        msg = "cannot set using a list-like indexer with a different length"
+        with pytest.raises(ValueError, match=msg):
+            # GH#36339
+            arr[[]] = [arr[1]]
+
+        msg = "cannot set using a slice indexer with a different length than"
+        with pytest.raises(ValueError, match=msg):
+            # GH#36339
+            arr[1:1] = arr[:3]
+
+    @pytest.mark.parametrize("box", [list, np.array, pd.Index, pd.Series])
+    def test_setitem_numeric_raises(self, arr1d, box):
+        # We dont case e.g. int64 to our own dtype for setitem
+
+        msg = (
+            f"value should be a '{arr1d._scalar_type.__name__}', "
+            "'NaT', or array of those. Got"
+        )
+        with pytest.raises(TypeError, match=msg):
+            arr1d[:2] = box([0, 1])
+
+        with pytest.raises(TypeError, match=msg):
+            arr1d[:2] = box([0.0, 1.0])
+
+    def test_inplace_arithmetic(self):
+        # GH#24115 check that iadd and isub are actually in-place
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        if self.array_cls is PeriodArray:
+            arr = self.array_cls(data, dtype="period[D]")
+        else:
+            arr = self.index_cls(data, freq="D")._data
+
+        expected = arr + pd.Timedelta(days=1)
+        arr += pd.Timedelta(days=1)
+        tm.assert_equal(arr, expected)
+
+        expected = arr - pd.Timedelta(days=1)
+        arr -= pd.Timedelta(days=1)
+        tm.assert_equal(arr, expected)
+
+    def test_shift_fill_int_deprecated(self, arr1d):
+        # GH#31971, enforced in 2.0
+        with pytest.raises(TypeError, match="value should be a"):
+            arr1d.shift(1, fill_value=1)
+
+    def test_median(self, arr1d):
+        arr = arr1d
+        if len(arr) % 2 == 0:
+            # make it easier to define `expected`
+            arr = arr[:-1]
+
+        expected = arr[len(arr) // 2]
+
+        result = arr.median()
+        assert type(result) is type(expected)
+        assert result == expected
+
+        arr[len(arr) // 2] = NaT
+        if not isinstance(expected, Period):
+            expected = arr[len(arr) // 2 - 1 : len(arr) // 2 + 2].mean()
+
+        assert arr.median(skipna=False) is NaT
+
+        result = arr.median()
+        assert type(result) is type(expected)
+        assert result == expected
+
+        assert arr[:0].median() is NaT
+        assert arr[:0].median(skipna=False) is NaT
+
+        # 2d Case
+        arr2 = arr.reshape(-1, 1)
+
+        result = arr2.median(axis=None)
+        assert type(result) is type(expected)
+        assert result == expected
+
+        assert arr2.median(axis=None, skipna=False) is NaT
+
+        result = arr2.median(axis=0)
+        expected2 = type(arr)._from_sequence([expected], dtype=arr.dtype)
+        tm.assert_equal(result, expected2)
+
+        result = arr2.median(axis=0, skipna=False)
+        expected2 = type(arr)._from_sequence([NaT], dtype=arr.dtype)
+        tm.assert_equal(result, expected2)
+
+        result = arr2.median(axis=1)
+        tm.assert_equal(result, arr)
+
+        result = arr2.median(axis=1, skipna=False)
+        tm.assert_equal(result, arr)
+
+    def test_from_integer_array(self):
+        arr = np.array([1, 2, 3], dtype=np.int64)
+        data = pd.array(arr, dtype="Int64")
+        if self.array_cls is PeriodArray:
+            expected = self.array_cls(arr, dtype=self.example_dtype)
+            result = self.array_cls(data, dtype=self.example_dtype)
+        else:
+            expected = self.array_cls._from_sequence(arr, dtype=self.example_dtype)
+            result = self.array_cls._from_sequence(data, dtype=self.example_dtype)
+
+        tm.assert_extension_array_equal(result, expected)
+
+
+class TestDatetimeArray(SharedTests):
+    index_cls = DatetimeIndex
+    array_cls = DatetimeArray
+    scalar_type = Timestamp
+    example_dtype = "M8[ns]"
+
+    @pytest.fixture
+    def arr1d(self, tz_naive_fixture, freqstr):
+        """
+        Fixture returning DatetimeArray with parametrized frequency and
+        timezones
+        """
+        tz = tz_naive_fixture
+        dti = pd.date_range("2016-01-01 01:01:00", periods=5, freq=freqstr, tz=tz)
+        dta = dti._data
+        return dta
+
+    def test_round(self, arr1d):
+        # GH#24064
+        dti = self.index_cls(arr1d)
+
+        result = dti.round(freq="2min")
+        expected = dti - pd.Timedelta(minutes=1)
+        expected = expected._with_freq(None)
+        tm.assert_index_equal(result, expected)
+
+        dta = dti._data
+        result = dta.round(freq="2min")
+        expected = expected._data._with_freq(None)
+        tm.assert_datetime_array_equal(result, expected)
+
+    def test_array_interface(self, datetime_index):
+        arr = datetime_index._data
+        copy_false = None if np_version_gt2 else False
+
+        # default asarray gives the same underlying data (for tz naive)
+        result = np.asarray(arr)
+        expected = arr._ndarray
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, copy=copy_false)
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+
+        # specifying M8[ns] gives the same result as default
+        result = np.asarray(arr, dtype="datetime64[ns]")
+        expected = arr._ndarray
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, dtype="datetime64[ns]", copy=copy_false)
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, dtype="datetime64[ns]")
+        if not np_version_gt2:
+            # TODO: GH 57739
+            assert result is not expected
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to object dtype
+        result = np.asarray(arr, dtype=object)
+        expected = np.array(list(arr), dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to other dtype always copies
+        result = np.asarray(arr, dtype="int64")
+        assert result is not arr.asi8
+        assert not np.may_share_memory(arr, result)
+        expected = arr.asi8.copy()
+        tm.assert_numpy_array_equal(result, expected)
+
+        # other dtypes handled by numpy
+        for dtype in ["float64", str]:
+            result = np.asarray(arr, dtype=dtype)
+            expected = np.asarray(arr).astype(dtype)
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_object_dtype(self, arr1d):
+        # GH#23524
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+
+        expected = np.array(list(dti))
+
+        result = np.array(arr, dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # also test the DatetimeIndex method while we're at it
+        result = np.array(dti, dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_tz(self, arr1d):
+        # GH#23524
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+        copy_false = None if np_version_gt2 else False
+
+        expected = dti.asi8.view("M8[ns]")
+        result = np.array(arr, dtype="M8[ns]")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.array(arr, dtype="datetime64[ns]")
+        tm.assert_numpy_array_equal(result, expected)
+
+        # check that we are not making copies when setting copy=copy_false
+        result = np.array(arr, dtype="M8[ns]", copy=copy_false)
+        assert result.base is expected.base
+        assert result.base is not None
+        result = np.array(arr, dtype="datetime64[ns]", copy=copy_false)
+        assert result.base is expected.base
+        assert result.base is not None
+
+    def test_array_i8_dtype(self, arr1d):
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+        copy_false = None if np_version_gt2 else False
+
+        expected = dti.asi8
+        result = np.array(arr, dtype="i8")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.array(arr, dtype=np.int64)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # check that we are still making copies when setting copy=copy_false
+        result = np.array(arr, dtype="i8", copy=copy_false)
+        assert result.base is not expected.base
+        assert result.base is None
+
+    def test_from_array_keeps_base(self):
+        # Ensure that DatetimeArray._ndarray.base isn't lost.
+        arr = np.array(["2000-01-01", "2000-01-02"], dtype="M8[ns]")
+        dta = DatetimeArray._from_sequence(arr, dtype=arr.dtype)
+
+        assert dta._ndarray is arr
+        dta = DatetimeArray._from_sequence(arr[:0], dtype=arr.dtype)
+        assert dta._ndarray.base is arr
+
+    def test_from_dti(self, arr1d):
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+        assert list(dti) == list(arr)
+
+        # Check that Index.__new__ knows what to do with DatetimeArray
+        dti2 = pd.Index(arr)
+        assert isinstance(dti2, DatetimeIndex)
+        assert list(dti2) == list(arr)
+
+    def test_astype_object(self, arr1d):
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+
+        asobj = arr.astype("O")
+        assert isinstance(asobj, np.ndarray)
+        assert asobj.dtype == "O"
+        assert list(asobj) == list(dti)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_to_period(self, datetime_index, freqstr):
+        dti = datetime_index
+        arr = dti._data
+
+        freqstr = PeriodDtype(to_offset(freqstr))._freqstr
+        expected = dti.to_period(freq=freqstr)
+        result = arr.to_period(freq=freqstr)
+        assert isinstance(result, PeriodArray)
+
+        tm.assert_equal(result, expected._data)
+
+    def test_to_period_2d(self, arr1d):
+        arr2d = arr1d.reshape(1, -1)
+
+        warn = None if arr1d.tz is None else UserWarning
+        with tm.assert_produces_warning(warn, match="will drop timezone information"):
+            result = arr2d.to_period("D")
+            expected = arr1d.to_period("D").reshape(1, -1)
+        tm.assert_period_array_equal(result, expected)
+
+    @pytest.mark.parametrize("propname", DatetimeArray._bool_ops)
+    def test_bool_properties(self, arr1d, propname):
+        # in this case _bool_ops is just `is_leap_year`
+        dti = self.index_cls(arr1d)
+        arr = arr1d
+        assert dti.freq == arr.freq
+
+        result = getattr(arr, propname)
+        expected = np.array(getattr(dti, propname), dtype=result.dtype)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("propname", DatetimeArray._field_ops)
+    def test_int_properties(self, arr1d, propname):
+        dti = self.index_cls(arr1d)
+        arr = arr1d
+
+        result = getattr(arr, propname)
+        expected = np.array(getattr(dti, propname), dtype=result.dtype)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_take_fill_valid(self, arr1d, fixed_now_ts):
+        arr = arr1d
+        dti = self.index_cls(arr1d)
+
+        now = fixed_now_ts.tz_localize(dti.tz)
+        result = arr.take([-1, 1], allow_fill=True, fill_value=now)
+        assert result[0] == now
+
+        msg = f"value should be a '{arr1d._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            # fill_value Timedelta invalid
+            arr.take([-1, 1], allow_fill=True, fill_value=now - now)
+
+        with pytest.raises(TypeError, match=msg):
+            # fill_value Period invalid
+            arr.take([-1, 1], allow_fill=True, fill_value=Period("2014Q1"))
+
+        tz = None if dti.tz is not None else "US/Eastern"
+        now = fixed_now_ts.tz_localize(tz)
+        msg = "Cannot compare tz-naive and tz-aware datetime-like objects"
+        with pytest.raises(TypeError, match=msg):
+            # Timestamp with mismatched tz-awareness
+            arr.take([-1, 1], allow_fill=True, fill_value=now)
+
+        value = NaT._value
+        msg = f"value should be a '{arr1d._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            # require NaT, not iNaT, as it could be confused with an integer
+            arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+        value = np.timedelta64("NaT", "ns")
+        with pytest.raises(TypeError, match=msg):
+            # require appropriate-dtype if we have a NA value
+            arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+        if arr.tz is not None:
+            # GH#37356
+            # Assuming here that arr1d fixture does not include Australia/Melbourne
+            value = fixed_now_ts.tz_localize("Australia/Melbourne")
+            result = arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+            expected = arr.take(
+                [-1, 1],
+                allow_fill=True,
+                fill_value=value.tz_convert(arr.dtype.tz),
+            )
+            tm.assert_equal(result, expected)
+
+    def test_concat_same_type_invalid(self, arr1d):
+        # different timezones
+        arr = arr1d
+
+        if arr.tz is None:
+            other = arr.tz_localize("UTC")
+        else:
+            other = arr.tz_localize(None)
+
+        with pytest.raises(ValueError, match="to_concat must have the same"):
+            arr._concat_same_type([arr, other])
+
+    def test_concat_same_type_different_freq(self, unit):
+        # we *can* concatenate DTI with different freqs.
+        a = pd.date_range("2000", periods=2, freq="D", tz="US/Central", unit=unit)._data
+        b = pd.date_range("2000", periods=2, freq="h", tz="US/Central", unit=unit)._data
+        result = DatetimeArray._concat_same_type([a, b])
+        expected = (
+            pd.to_datetime(
+                [
+                    "2000-01-01 00:00:00",
+                    "2000-01-02 00:00:00",
+                    "2000-01-01 00:00:00",
+                    "2000-01-01 01:00:00",
+                ]
+            )
+            .tz_localize("US/Central")
+            .as_unit(unit)
+            ._data
+        )
+
+        tm.assert_datetime_array_equal(result, expected)
+
+    def test_strftime(self, arr1d, using_infer_string):
+        arr = arr1d
+
+        result = arr.strftime("%Y %b")
+        expected = np.array([ts.strftime("%Y %b") for ts in arr], dtype=object)
+        if using_infer_string:
+            expected = pd.array(expected, dtype=pd.StringDtype(na_value=np.nan))
+        tm.assert_equal(result, expected)
+
+    def test_strftime_nat(self, using_infer_string):
+        # GH 29578
+        arr = DatetimeIndex(["2019-01-01", NaT])._data
+
+        result = arr.strftime("%Y-%m-%d")
+        expected = np.array(["2019-01-01", np.nan], dtype=object)
+        if using_infer_string:
+            expected = pd.array(expected, dtype=pd.StringDtype(na_value=np.nan))
+        tm.assert_equal(result, expected)
+
+
+class TestTimedeltaArray(SharedTests):
+    index_cls = TimedeltaIndex
+    array_cls = TimedeltaArray
+    scalar_type = pd.Timedelta
+    example_dtype = "m8[ns]"
+
+    def test_from_tdi(self):
+        tdi = TimedeltaIndex(["1 Day", "3 Hours"])
+        arr = tdi._data
+        assert list(arr) == list(tdi)
+
+        # Check that Index.__new__ knows what to do with TimedeltaArray
+        tdi2 = pd.Index(arr)
+        assert isinstance(tdi2, TimedeltaIndex)
+        assert list(tdi2) == list(arr)
+
+    def test_astype_object(self):
+        tdi = TimedeltaIndex(["1 Day", "3 Hours"])
+        arr = tdi._data
+        asobj = arr.astype("O")
+        assert isinstance(asobj, np.ndarray)
+        assert asobj.dtype == "O"
+        assert list(asobj) == list(tdi)
+
+    def test_to_pytimedelta(self, timedelta_index):
+        tdi = timedelta_index
+        arr = tdi._data
+
+        expected = tdi.to_pytimedelta()
+        result = arr.to_pytimedelta()
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_total_seconds(self, timedelta_index):
+        tdi = timedelta_index
+        arr = tdi._data
+
+        expected = tdi.total_seconds()
+        result = arr.total_seconds()
+
+        tm.assert_numpy_array_equal(result, expected.values)
+
+    @pytest.mark.parametrize("propname", TimedeltaArray._field_ops)
+    def test_int_properties(self, timedelta_index, propname):
+        tdi = timedelta_index
+        arr = tdi._data
+
+        result = getattr(arr, propname)
+        expected = np.array(getattr(tdi, propname), dtype=result.dtype)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_interface(self, timedelta_index):
+        arr = timedelta_index._data
+        copy_false = None if np_version_gt2 else False
+
+        # default asarray gives the same underlying data
+        result = np.asarray(arr)
+        expected = arr._ndarray
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, copy=copy_false)
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+
+        # specifying m8[ns] gives the same result as default
+        result = np.asarray(arr, dtype="timedelta64[ns]")
+        expected = arr._ndarray
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, dtype="timedelta64[ns]", copy=copy_false)
+        assert result is expected
+        tm.assert_numpy_array_equal(result, expected)
+        result = np.array(arr, dtype="timedelta64[ns]")
+        if not np_version_gt2:
+            # TODO: GH 57739
+            assert result is not expected
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to object dtype
+        result = np.asarray(arr, dtype=object)
+        expected = np.array(list(arr), dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to other dtype always copies
+        result = np.asarray(arr, dtype="int64")
+        assert result is not arr.asi8
+        assert not np.may_share_memory(arr, result)
+        expected = arr.asi8.copy()
+        tm.assert_numpy_array_equal(result, expected)
+
+        # other dtypes handled by numpy
+        for dtype in ["float64", str]:
+            result = np.asarray(arr, dtype=dtype)
+            expected = np.asarray(arr).astype(dtype)
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_take_fill_valid(self, timedelta_index, fixed_now_ts):
+        tdi = timedelta_index
+        arr = tdi._data
+
+        td1 = pd.Timedelta(days=1)
+        result = arr.take([-1, 1], allow_fill=True, fill_value=td1)
+        assert result[0] == td1
+
+        value = fixed_now_ts
+        msg = f"value should be a '{arr._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            # fill_value Timestamp invalid
+            arr.take([0, 1], allow_fill=True, fill_value=value)
+
+        value = fixed_now_ts.to_period("D")
+        with pytest.raises(TypeError, match=msg):
+            # fill_value Period invalid
+            arr.take([0, 1], allow_fill=True, fill_value=value)
+
+        value = np.datetime64("NaT", "ns")
+        with pytest.raises(TypeError, match=msg):
+            # require appropriate-dtype if we have a NA value
+            arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+
+@pytest.mark.filterwarnings(r"ignore:Period with BDay freq is deprecated:FutureWarning")
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+class TestPeriodArray(SharedTests):
+    index_cls = PeriodIndex
+    array_cls = PeriodArray
+    scalar_type = Period
+    example_dtype = PeriodIndex([], freq="W").dtype
+
+    @pytest.fixture
+    def arr1d(self, period_index):
+        """
+        Fixture returning DatetimeArray from parametrized PeriodIndex objects
+        """
+        return period_index._data
+
+    def test_from_pi(self, arr1d):
+        pi = self.index_cls(arr1d)
+        arr = arr1d
+        assert list(arr) == list(pi)
+
+        # Check that Index.__new__ knows what to do with PeriodArray
+        pi2 = pd.Index(arr)
+        assert isinstance(pi2, PeriodIndex)
+        assert list(pi2) == list(arr)
+
+    def test_astype_object(self, arr1d):
+        pi = self.index_cls(arr1d)
+        arr = arr1d
+        asobj = arr.astype("O")
+        assert isinstance(asobj, np.ndarray)
+        assert asobj.dtype == "O"
+        assert list(asobj) == list(pi)
+
+    def test_take_fill_valid(self, arr1d):
+        arr = arr1d
+
+        value = NaT._value
+        msg = f"value should be a '{arr1d._scalar_type.__name__}' or 'NaT'. Got"
+        with pytest.raises(TypeError, match=msg):
+            # require NaT, not iNaT, as it could be confused with an integer
+            arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+        value = np.timedelta64("NaT", "ns")
+        with pytest.raises(TypeError, match=msg):
+            # require appropriate-dtype if we have a NA value
+            arr.take([-1, 1], allow_fill=True, fill_value=value)
+
+    @pytest.mark.parametrize("how", ["S", "E"])
+    def test_to_timestamp(self, how, arr1d):
+        pi = self.index_cls(arr1d)
+        arr = arr1d
+
+        expected = DatetimeIndex(pi.to_timestamp(how=how))._data
+        result = arr.to_timestamp(how=how)
+        assert isinstance(result, DatetimeArray)
+
+        tm.assert_equal(result, expected)
+
+    def test_to_timestamp_roundtrip_bday(self):
+        # Case where infer_freq inside would choose "D" instead of "B"
+        dta = pd.date_range("2021-10-18", periods=3, freq="B")._data
+        parr = dta.to_period()
+        result = parr.to_timestamp()
+        assert result.freq == "B"
+        tm.assert_extension_array_equal(result, dta)
+
+        dta2 = dta[::2]
+        parr2 = dta2.to_period()
+        result2 = parr2.to_timestamp()
+        assert result2.freq == "2B"
+        tm.assert_extension_array_equal(result2, dta2)
+
+        parr3 = dta.to_period("2B")
+        result3 = parr3.to_timestamp()
+        assert result3.freq == "B"
+        tm.assert_extension_array_equal(result3, dta)
+
+    def test_to_timestamp_out_of_bounds(self):
+        # GH#19643 previously overflowed silently
+        pi = pd.period_range("1500", freq="Y", periods=3)
+        msg = "Out of bounds nanosecond timestamp: 1500-01-01 00:00:00"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            pi.to_timestamp()
+
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            pi._data.to_timestamp()
+
+    @pytest.mark.parametrize("propname", PeriodArray._bool_ops)
+    def test_bool_properties(self, arr1d, propname):
+        # in this case _bool_ops is just `is_leap_year`
+        pi = self.index_cls(arr1d)
+        arr = arr1d
+
+        result = getattr(arr, propname)
+        expected = np.array(getattr(pi, propname))
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("propname", PeriodArray._field_ops)
+    def test_int_properties(self, arr1d, propname):
+        pi = self.index_cls(arr1d)
+        arr = arr1d
+
+        result = getattr(arr, propname)
+        expected = np.array(getattr(pi, propname))
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_interface(self, arr1d):
+        arr = arr1d
+
+        # default asarray gives objects
+        result = np.asarray(arr)
+        expected = np.array(list(arr), dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to object dtype (same as default)
+        result = np.asarray(arr, dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # to int64 gives the underlying representation
+        result = np.asarray(arr, dtype="int64")
+        tm.assert_numpy_array_equal(result, arr.asi8)
+
+        result2 = np.asarray(arr, dtype="int64")
+        assert np.may_share_memory(result, result2)
+
+        result_copy1 = np.array(arr, dtype="int64", copy=True)
+        result_copy2 = np.array(arr, dtype="int64", copy=True)
+        assert not np.may_share_memory(result_copy1, result_copy2)
+
+        # to other dtypes
+        msg = r"float\(\) argument must be a string or a( real)? number, not 'Period'"
+        with pytest.raises(TypeError, match=msg):
+            np.asarray(arr, dtype="float64")
+
+        result = np.asarray(arr, dtype="S20")
+        expected = np.asarray(arr).astype("S20")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_strftime(self, arr1d, using_infer_string):
+        arr = arr1d
+
+        result = arr.strftime("%Y")
+        expected = np.array([per.strftime("%Y") for per in arr], dtype=object)
+        if using_infer_string:
+            expected = pd.array(expected, dtype=pd.StringDtype(na_value=np.nan))
+        tm.assert_equal(result, expected)
+
+    def test_strftime_nat(self, using_infer_string):
+        # GH 29578
+        arr = PeriodArray(PeriodIndex(["2019-01-01", NaT], dtype="period[D]"))
+
+        result = arr.strftime("%Y-%m-%d")
+        expected = np.array(["2019-01-01", np.nan], dtype=object)
+        if using_infer_string:
+            expected = pd.array(expected, dtype=pd.StringDtype(na_value=np.nan))
+        tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arr,casting_nats",
+    [
+        (
+            TimedeltaIndex(["1 Day", "3 Hours", "NaT"])._data,
+            (NaT, np.timedelta64("NaT", "ns")),
+        ),
+        (
+            pd.date_range("2000-01-01", periods=3, freq="D")._data,
+            (NaT, np.datetime64("NaT", "ns")),
+        ),
+        (pd.period_range("2000-01-01", periods=3, freq="D")._data, (NaT,)),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def test_casting_nat_setitem_array(arr, casting_nats):
+    expected = type(arr)._from_sequence([NaT, arr[1], arr[2]], dtype=arr.dtype)
+
+    for nat in casting_nats:
+        arr = arr.copy()
+        arr[0] = nat
+        tm.assert_equal(arr, expected)
+
+
+@pytest.mark.parametrize(
+    "arr,non_casting_nats",
+    [
+        (
+            TimedeltaIndex(["1 Day", "3 Hours", "NaT"])._data,
+            (np.datetime64("NaT", "ns"), NaT._value),
+        ),
+        (
+            pd.date_range("2000-01-01", periods=3, freq="D")._data,
+            (np.timedelta64("NaT", "ns"), NaT._value),
+        ),
+        (
+            pd.period_range("2000-01-01", periods=3, freq="D")._data,
+            (np.datetime64("NaT", "ns"), np.timedelta64("NaT", "ns"), NaT._value),
+        ),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def test_invalid_nat_setitem_array(arr, non_casting_nats):
+    msg = (
+        "value should be a '(Timestamp|Timedelta|Period)', 'NaT', or array of those. "
+        "Got '(timedelta64|datetime64|int)' instead."
+    )
+
+    for nat in non_casting_nats:
+        with pytest.raises(TypeError, match=msg):
+            arr[0] = nat
+
+
+@pytest.mark.parametrize(
+    "arr",
+    [
+        pd.date_range("2000", periods=4).array,
+        pd.timedelta_range("2000", periods=4).array,
+    ],
+)
+def test_to_numpy_extra(arr):
+    arr[0] = NaT
+    original = arr.copy()
+
+    result = arr.to_numpy()
+    assert np.isnan(result[0])
+
+    result = arr.to_numpy(dtype="int64")
+    assert result[0] == -9223372036854775808
+
+    result = arr.to_numpy(dtype="int64", na_value=0)
+    assert result[0] == 0
+
+    result = arr.to_numpy(na_value=arr[1].to_numpy())
+    assert result[0] == result[1]
+
+    result = arr.to_numpy(na_value=arr[1].to_numpy(copy=False))
+    assert result[0] == result[1]
+
+    tm.assert_equal(arr, original)
+
+
+@pytest.mark.parametrize("as_index", [True, False])
+@pytest.mark.parametrize(
+    "values",
+    [
+        pd.to_datetime(["2020-01-01", "2020-02-01"]),
+        pd.to_timedelta([1, 2], unit="D"),
+        PeriodIndex(["2020-01-01", "2020-02-01"], freq="D"),
+    ],
+)
+@pytest.mark.parametrize(
+    "klass",
+    [
+        list,
+        np.array,
+        pd.array,
+        pd.Series,
+        pd.Index,
+        pd.Categorical,
+        pd.CategoricalIndex,
+    ],
+)
+def test_searchsorted_datetimelike_with_listlike(values, klass, as_index):
+    # https://github.com/pandas-dev/pandas/issues/32762
+    if not as_index:
+        values = values._data
+
+    result = values.searchsorted(klass(values))
+    expected = np.array([0, 1], dtype=result.dtype)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        pd.to_datetime(["2020-01-01", "2020-02-01"]),
+        pd.to_timedelta([1, 2], unit="D"),
+        PeriodIndex(["2020-01-01", "2020-02-01"], freq="D"),
+    ],
+)
+@pytest.mark.parametrize(
+    "arg", [[1, 2], ["a", "b"], [Timestamp("2020-01-01", tz="Europe/London")] * 2]
+)
+def test_searchsorted_datetimelike_with_listlike_invalid_dtype(values, arg):
+    # https://github.com/pandas-dev/pandas/issues/32762
+    msg = "[Unexpected type|Cannot compare]"
+    with pytest.raises(TypeError, match=msg):
+        values.searchsorted(arg)
+
+
+@pytest.mark.parametrize("klass", [list, tuple, np.array, pd.Series])
+def test_period_index_construction_from_strings(klass):
+    # https://github.com/pandas-dev/pandas/issues/26109
+    strings = ["2020Q1", "2020Q2"] * 2
+    data = klass(strings)
+    result = PeriodIndex(data, freq="Q")
+    expected = PeriodIndex([Period(s) for s in strings])
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+def test_from_pandas_array(dtype):
+    # GH#24615
+    data = np.array([1, 2, 3], dtype=dtype)
+    arr = NumpyExtensionArray(data)
+
+    cls = {"M8[ns]": DatetimeArray, "m8[ns]": TimedeltaArray}[dtype]
+
+    result = cls._from_sequence(arr, dtype=dtype)
+    expected = cls._from_sequence(data, dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+    func = {"M8[ns]": pd.to_datetime, "m8[ns]": pd.to_timedelta}[dtype]
+    result = func(arr).array
+    expected = func(data).array
+    tm.assert_equal(result, expected)
+
+    # Let's check the Indexes while we're here
+    idx_cls = {"M8[ns]": DatetimeIndex, "m8[ns]": TimedeltaIndex}[dtype]
+    result = idx_cls(arr)
+    expected = idx_cls(data)
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/test_datetimes.py b/pandas/pandas/tests/arrays/test_datetimes.py
new file mode 100644
index 0000000000000..e3f49d04a0ff2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_datetimes.py
@@ -0,0 +1,834 @@
+"""
+Tests for DatetimeArray
+"""
+
+from __future__ import annotations
+
+from datetime import timedelta
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import tz_compare
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import (
+    DatetimeArray,
+    TimedeltaArray,
+)
+
+
+class TestNonNano:
+    @pytest.fixture(params=["s", "ms", "us"])
+    def unit(self, request):
+        """Fixture returning parametrized time units"""
+        return request.param
+
+    @pytest.fixture
+    def dtype(self, unit, tz_naive_fixture):
+        tz = tz_naive_fixture
+        if tz is None:
+            return np.dtype(f"datetime64[{unit}]")
+        else:
+            return DatetimeTZDtype(unit=unit, tz=tz)
+
+    @pytest.fixture
+    def dta_dti(self, unit, dtype):
+        tz = getattr(dtype, "tz", None)
+
+        dti = pd.date_range("2016-01-01", periods=55, freq="D", tz=tz)
+        if tz is None:
+            arr = np.asarray(dti).astype(f"M8[{unit}]")
+        else:
+            arr = np.asarray(dti.tz_convert("UTC").tz_localize(None)).astype(
+                f"M8[{unit}]"
+            )
+
+        dta = DatetimeArray._simple_new(arr, dtype=dtype)
+        return dta, dti
+
+    @pytest.fixture
+    def dta(self, dta_dti):
+        dta, dti = dta_dti
+        return dta
+
+    def test_non_nano(self, unit, dtype):
+        arr = np.arange(5, dtype=np.int64).view(f"M8[{unit}]")
+        dta = DatetimeArray._simple_new(arr, dtype=dtype)
+
+        assert dta.dtype == dtype
+        assert dta[0].unit == unit
+        assert tz_compare(dta.tz, dta[0].tz)
+        assert (dta[0] == dta[:1]).all()
+
+    @pytest.mark.parametrize(
+        "field", DatetimeArray._field_ops + DatetimeArray._bool_ops
+    )
+    def test_fields(self, unit, field, dtype, dta_dti):
+        dta, dti = dta_dti
+
+        assert (dti == dta).all()
+
+        res = getattr(dta, field)
+        expected = getattr(dti._data, field)
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_normalize(self, unit):
+        dti = pd.date_range("2016-01-01 06:00:00", periods=55, freq="D")
+        arr = np.asarray(dti).astype(f"M8[{unit}]")
+
+        dta = DatetimeArray._simple_new(arr, dtype=arr.dtype)
+
+        assert not dta.is_normalized
+
+        # TODO: simplify once we can just .astype to other unit
+        exp = np.asarray(dti.normalize()).astype(f"M8[{unit}]")
+        expected = DatetimeArray._simple_new(exp, dtype=exp.dtype)
+
+        res = dta.normalize()
+        tm.assert_extension_array_equal(res, expected)
+
+    def test_simple_new_requires_match(self, unit):
+        arr = np.arange(5, dtype=np.int64).view(f"M8[{unit}]")
+        dtype = DatetimeTZDtype(unit, "UTC")
+
+        dta = DatetimeArray._simple_new(arr, dtype=dtype)
+        assert dta.dtype == dtype
+
+        wrong = DatetimeTZDtype("ns", "UTC")
+        with pytest.raises(AssertionError, match=""):
+            DatetimeArray._simple_new(arr, dtype=wrong)
+
+    def test_std_non_nano(self, unit):
+        dti = pd.date_range("2016-01-01", periods=55, freq="D")
+        arr = np.asarray(dti).astype(f"M8[{unit}]")
+
+        dta = DatetimeArray._simple_new(arr, dtype=arr.dtype)
+
+        # we should match the nano-reso std, but floored to our reso.
+        res = dta.std()
+        assert res._creso == dta._creso
+        assert res == dti.std().floor(unit)
+
+    @pytest.mark.filterwarnings("ignore:Converting to PeriodArray.*:UserWarning")
+    def test_to_period(self, dta_dti):
+        dta, dti = dta_dti
+        result = dta.to_period("D")
+        expected = dti._data.to_period("D")
+
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_iter(self, dta):
+        res = next(iter(dta))
+        expected = dta[0]
+
+        assert type(res) is pd.Timestamp
+        assert res._value == expected._value
+        assert res._creso == expected._creso
+        assert res == expected
+
+    def test_astype_object(self, dta):
+        result = dta.astype(object)
+        assert all(x._creso == dta._creso for x in result)
+        assert all(x == y for x, y in zip(result, dta))
+
+    def test_to_pydatetime(self, dta_dti):
+        dta, dti = dta_dti
+
+        result = dta.to_pydatetime()
+        expected = dti.to_pydatetime()
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("meth", ["time", "timetz", "date"])
+    def test_time_date(self, dta_dti, meth):
+        dta, dti = dta_dti
+
+        result = getattr(dta, meth)
+        expected = getattr(dti, meth)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_format_native_types(self, unit, dtype, dta_dti):
+        # In this case we should get the same formatted values with our nano
+        #  version dti._data as we do with the non-nano dta
+        dta, dti = dta_dti
+
+        res = dta._format_native_types()
+        exp = dti._data._format_native_types()
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_repr(self, dta_dti, unit):
+        dta, dti = dta_dti
+
+        assert repr(dta) == repr(dti._data).replace("[ns", f"[{unit}")
+
+    # TODO: tests with td64
+    def test_compare_mismatched_resolutions(self, comparison_op):
+        # comparison that numpy gets wrong bc of silent overflows
+        op = comparison_op
+
+        iinfo = np.iinfo(np.int64)
+        vals = np.array([iinfo.min, iinfo.min + 1, iinfo.max], dtype=np.int64)
+
+        # Construct so that arr2[1] < arr[1] < arr[2] < arr2[2]
+        arr = np.array(vals).view("M8[ns]")
+        arr2 = arr.view("M8[s]")
+
+        left = DatetimeArray._simple_new(arr, dtype=arr.dtype)
+        right = DatetimeArray._simple_new(arr2, dtype=arr2.dtype)
+
+        if comparison_op is operator.eq:
+            expected = np.array([False, False, False])
+        elif comparison_op is operator.ne:
+            expected = np.array([True, True, True])
+        elif comparison_op in [operator.lt, operator.le]:
+            expected = np.array([False, False, True])
+        else:
+            expected = np.array([False, True, False])
+
+        result = op(left, right)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = op(left[1], right)
+        tm.assert_numpy_array_equal(result, expected)
+
+        if op not in [operator.eq, operator.ne]:
+            # check that numpy still gets this wrong; if it is fixed we may be
+            #  able to remove compare_mismatched_resolutions
+            np_res = op(left._ndarray, right._ndarray)
+            tm.assert_numpy_array_equal(np_res[1:], ~expected[1:])
+
+    def test_add_mismatched_reso_doesnt_downcast(self):
+        # https://github.com/pandas-dev/pandas/pull/48748#issuecomment-1260181008
+        td = pd.Timedelta(microseconds=1)
+        dti = pd.date_range("2016-01-01", periods=3) - td
+        dta = dti._data.as_unit("us")
+
+        res = dta + td.as_unit("us")
+        # even though the result is an even number of days
+        #  (so we _could_ downcast to unit="s"), we do not.
+        assert res.unit == "us"
+
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            timedelta(hours=2),
+            pd.Timedelta(hours=2),
+            np.timedelta64(2, "h"),
+            np.timedelta64(2 * 3600 * 1000, "ms"),
+            pd.offsets.Minute(120),
+            pd.offsets.Hour(2),
+        ],
+    )
+    def test_add_timedeltalike_scalar_mismatched_reso(self, dta_dti, scalar):
+        dta, dti = dta_dti
+
+        td = pd.Timedelta(scalar)
+        exp_unit = tm.get_finest_unit(dta.unit, td.unit)
+
+        expected = (dti + td)._data.as_unit(exp_unit)
+        result = dta + scalar
+        tm.assert_extension_array_equal(result, expected)
+
+        result = scalar + dta
+        tm.assert_extension_array_equal(result, expected)
+
+        expected = (dti - td)._data.as_unit(exp_unit)
+        result = dta - scalar
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_sub_datetimelike_scalar_mismatch(self):
+        dti = pd.date_range("2016-01-01", periods=3)
+        dta = dti._data.as_unit("us")
+
+        ts = dta[0].as_unit("s")
+
+        result = dta - ts
+        expected = (dti - dti[0])._data.as_unit("us")
+        assert result.dtype == "m8[us]"
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_sub_datetime64_reso_mismatch(self):
+        dti = pd.date_range("2016-01-01", periods=3)
+        left = dti._data.as_unit("s")
+        right = left.as_unit("ms")
+
+        result = left - right
+        exp_values = np.array([0, 0, 0], dtype="m8[ms]")
+        expected = TimedeltaArray._simple_new(
+            exp_values,
+            dtype=exp_values.dtype,
+        )
+        tm.assert_extension_array_equal(result, expected)
+        result2 = right - left
+        tm.assert_extension_array_equal(result2, expected)
+
+
+class TestDatetimeArrayComparisons:
+    # TODO: merge this into tests/arithmetic/test_datetime64 once it is
+    #  sufficiently robust
+
+    def test_cmp_dt64_arraylike_tznaive(self, comparison_op):
+        # arbitrary tz-naive DatetimeIndex
+        op = comparison_op
+
+        dti = pd.date_range("2016-01-1", freq="MS", periods=9, tz=None)
+        arr = dti._data
+        assert arr.freq == dti.freq
+        assert arr.tz == dti.tz
+
+        right = dti
+
+        expected = np.ones(len(arr), dtype=bool)
+        if comparison_op.__name__ in ["ne", "gt", "lt"]:
+            # for these the comparisons should be all-False
+            expected = ~expected
+
+        result = op(arr, arr)
+        tm.assert_numpy_array_equal(result, expected)
+        for other in [
+            right,
+            np.array(right),
+            list(right),
+            tuple(right),
+            right.astype(object),
+        ]:
+            result = op(arr, other)
+            tm.assert_numpy_array_equal(result, expected)
+
+            result = op(other, arr)
+            tm.assert_numpy_array_equal(result, expected)
+
+
+class TestDatetimeArray:
+    def test_astype_ns_to_ms_near_bounds(self):
+        # GH#55979
+        ts = pd.Timestamp("1677-09-21 00:12:43.145225")
+        target = ts.as_unit("ms")
+
+        dta = DatetimeArray._from_sequence([ts], dtype="M8[ns]")
+        assert (dta.view("i8") == ts.as_unit("ns").value).all()
+
+        result = dta.astype("M8[ms]")
+        assert result[0] == target
+
+        expected = DatetimeArray._from_sequence([ts], dtype="M8[ms]")
+        assert (expected.view("i8") == target._value).all()
+
+        tm.assert_datetime_array_equal(result, expected)
+
+    def test_astype_non_nano_tznaive(self):
+        dti = pd.date_range("2016-01-01", periods=3)
+
+        res = dti.astype("M8[s]")
+        assert res.dtype == "M8[s]"
+
+        dta = dti._data
+        res = dta.astype("M8[s]")
+        assert res.dtype == "M8[s]"
+        assert isinstance(res, pd.core.arrays.DatetimeArray)  # used to be ndarray
+
+    def test_astype_non_nano_tzaware(self):
+        dti = pd.date_range("2016-01-01", periods=3, tz="UTC")
+
+        res = dti.astype("M8[s, US/Pacific]")
+        assert res.dtype == "M8[s, US/Pacific]"
+
+        dta = dti._data
+        res = dta.astype("M8[s, US/Pacific]")
+        assert res.dtype == "M8[s, US/Pacific]"
+
+        # from non-nano to non-nano, preserving reso
+        res2 = res.astype("M8[s, UTC]")
+        assert res2.dtype == "M8[s, UTC]"
+        assert not tm.shares_memory(res2, res)
+
+        res3 = res.astype("M8[s, UTC]", copy=False)
+        assert res2.dtype == "M8[s, UTC]"
+        assert tm.shares_memory(res3, res)
+
+    def test_astype_to_same(self):
+        arr = DatetimeArray._from_sequence(
+            ["2000"], dtype=DatetimeTZDtype(tz="US/Central")
+        )
+        result = arr.astype(DatetimeTZDtype(tz="US/Central"), copy=False)
+        assert result is arr
+
+    @pytest.mark.parametrize("dtype", ["datetime64[ns]", "datetime64[ns, UTC]"])
+    @pytest.mark.parametrize(
+        "other", ["datetime64[ns]", "datetime64[ns, UTC]", "datetime64[ns, CET]"]
+    )
+    def test_astype_copies(self, dtype, other):
+        # https://github.com/pandas-dev/pandas/pull/32490
+        ser = pd.Series([1, 2], dtype=dtype)
+        orig = ser.copy()
+
+        err = False
+        if (dtype == "datetime64[ns]") ^ (other == "datetime64[ns]"):
+            # deprecated in favor of tz_localize
+            err = True
+
+        if err:
+            if dtype == "datetime64[ns]":
+                msg = "Use obj.tz_localize instead or series.dt.tz_localize instead"
+            else:
+                msg = "from timezone-aware dtype to timezone-naive dtype"
+            with pytest.raises(TypeError, match=msg):
+                ser.astype(other)
+        else:
+            t = ser.astype(other)
+            t[:] = pd.NaT
+            tm.assert_series_equal(ser, orig)
+
+    @pytest.mark.parametrize("dtype", [int, np.int32, np.int64, "uint32", "uint64"])
+    def test_astype_int(self, dtype):
+        arr = DatetimeArray._from_sequence(
+            [pd.Timestamp("2000"), pd.Timestamp("2001")], dtype="M8[ns]"
+        )
+
+        if np.dtype(dtype) != np.int64:
+            with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+                arr.astype(dtype)
+            return
+
+        result = arr.astype(dtype)
+        expected = arr._ndarray.view("i8")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_to_sparse_dt64(self):
+        # GH#50082
+        dti = pd.date_range("2016-01-01", periods=4)
+        dta = dti._data
+        result = dta.astype("Sparse[datetime64[ns]]")
+
+        assert result.dtype == "Sparse[datetime64[ns]]"
+        assert (result == dta).all()
+
+    def test_tz_setter_raises(self):
+        arr = DatetimeArray._from_sequence(
+            ["2000"], dtype=DatetimeTZDtype(tz="US/Central")
+        )
+        with pytest.raises(AttributeError, match="tz_localize"):
+            arr.tz = "UTC"
+
+    def test_setitem_str_impute_tz(self, tz_naive_fixture):
+        # Like for getitem, if we are passed a naive-like string, we impute
+        #  our own timezone.
+        tz = tz_naive_fixture
+
+        data = np.array([1, 2, 3], dtype="M8[ns]")
+        dtype = data.dtype if tz is None else DatetimeTZDtype(tz=tz)
+        arr = DatetimeArray._from_sequence(data, dtype=dtype)
+        expected = arr.copy()
+
+        ts = pd.Timestamp("2020-09-08 16:50").tz_localize(tz)
+        setter = str(ts.tz_localize(None))
+
+        # Setting a scalar tznaive string
+        expected[0] = ts
+        arr[0] = setter
+        tm.assert_equal(arr, expected)
+
+        # Setting a listlike of tznaive strings
+        expected[1] = ts
+        arr[:2] = [setter, setter]
+        tm.assert_equal(arr, expected)
+
+    def test_setitem_different_tz_raises(self):
+        # pre-2.0 we required exact tz match, in 2.0 we require only
+        #  tzawareness-match
+        data = np.array([1, 2, 3], dtype="M8[ns]")
+        arr = DatetimeArray._from_sequence(
+            data, copy=False, dtype=DatetimeTZDtype(tz="US/Central")
+        )
+        with pytest.raises(TypeError, match="Cannot compare tz-naive and tz-aware"):
+            arr[0] = pd.Timestamp("2000")
+
+        ts = pd.Timestamp("2000", tz="US/Eastern")
+        arr[0] = ts
+        assert arr[0] == ts.tz_convert("US/Central")
+
+    def test_setitem_clears_freq(self):
+        a = pd.date_range("2000", periods=2, freq="D", tz="US/Central")._data
+        a[0] = pd.Timestamp("2000", tz="US/Central")
+        assert a.freq is None
+
+    @pytest.mark.parametrize(
+        "obj",
+        [
+            pd.Timestamp("2021-01-01"),
+            pd.Timestamp("2021-01-01").to_datetime64(),
+            pd.Timestamp("2021-01-01").to_pydatetime(),
+        ],
+    )
+    def test_setitem_objects(self, obj):
+        # make sure we accept datetime64 and datetime in addition to Timestamp
+        dti = pd.date_range("2000", periods=2, freq="D")
+        arr = dti._data
+
+        arr[0] = obj
+        assert arr[0] == obj
+
+    def test_repeat_preserves_tz(self):
+        dti = pd.date_range("2000", periods=2, freq="D", tz="US/Central")
+        arr = dti._data
+
+        repeated = arr.repeat([1, 1])
+
+        # preserves tz and values, but not freq
+        expected = DatetimeArray._from_sequence(arr.asi8, dtype=arr.dtype)
+        tm.assert_equal(repeated, expected)
+
+    def test_value_counts_preserves_tz(self):
+        dti = pd.date_range("2000", periods=2, freq="D", tz="US/Central")
+        arr = dti._data.repeat([4, 3])
+
+        result = arr.value_counts()
+
+        # Note: not tm.assert_index_equal, since `freq`s do not match
+        assert result.index.equals(dti)
+
+        arr[-2] = pd.NaT
+        result = arr.value_counts(dropna=False)
+        expected = pd.Series([4, 2, 1], index=[dti[0], dti[1], pd.NaT], name="count")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["pad", "backfill"])
+    def test_fillna_preserves_tz(self, method):
+        dti = pd.date_range("2000-01-01", periods=5, freq="D", tz="US/Central")
+        arr = DatetimeArray._from_sequence(dti, dtype=dti.dtype, copy=True)
+        arr[2] = pd.NaT
+
+        fill_val = dti[1] if method == "pad" else dti[3]
+        expected = DatetimeArray._from_sequence(
+            [dti[0], dti[1], fill_val, dti[3], dti[4]],
+            dtype=DatetimeTZDtype(tz="US/Central"),
+        )
+
+        result = arr._pad_or_backfill(method=method)
+        tm.assert_extension_array_equal(result, expected)
+
+        # assert that arr and dti were not modified in-place
+        assert arr[2] is pd.NaT
+        assert dti[2] == pd.Timestamp("2000-01-03", tz="US/Central")
+
+    def test_fillna_2d(self):
+        dti = pd.date_range("2016-01-01", periods=6, tz="US/Pacific")
+        dta = dti._data.reshape(3, 2).copy()
+        dta[0, 1] = pd.NaT
+        dta[1, 0] = pd.NaT
+
+        res1 = dta._pad_or_backfill(method="pad")
+        expected1 = dta.copy()
+        expected1[1, 0] = dta[0, 0]
+        tm.assert_extension_array_equal(res1, expected1)
+
+        res2 = dta._pad_or_backfill(method="backfill")
+        expected2 = dta.copy()
+        expected2 = dta.copy()
+        expected2[1, 0] = dta[2, 0]
+        expected2[0, 1] = dta[1, 1]
+        tm.assert_extension_array_equal(res2, expected2)
+
+        # with different ordering for underlying ndarray; behavior should
+        #  be unchanged
+        dta2 = dta._from_backing_data(dta._ndarray.copy(order="F"))
+        assert dta2._ndarray.flags["F_CONTIGUOUS"]
+        assert not dta2._ndarray.flags["C_CONTIGUOUS"]
+        tm.assert_extension_array_equal(dta, dta2)
+
+        res3 = dta2._pad_or_backfill(method="pad")
+        tm.assert_extension_array_equal(res3, expected1)
+
+        res4 = dta2._pad_or_backfill(method="backfill")
+        tm.assert_extension_array_equal(res4, expected2)
+
+        # test the DataFrame method while we're here
+        df = pd.DataFrame(dta)
+        res = df.ffill()
+        expected = pd.DataFrame(expected1)
+        tm.assert_frame_equal(res, expected)
+
+        res = df.bfill()
+        expected = pd.DataFrame(expected2)
+        tm.assert_frame_equal(res, expected)
+
+    def test_array_interface_tz(self):
+        tz = "US/Central"
+        data = pd.date_range("2017", periods=2, tz=tz)._data
+        result = np.asarray(data)
+
+        expected = np.array(
+            [
+                pd.Timestamp("2017-01-01T00:00:00", tz=tz),
+                pd.Timestamp("2017-01-02T00:00:00", tz=tz),
+            ],
+            dtype=object,
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.asarray(data, dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.asarray(data, dtype="M8[ns]")
+
+        expected = np.array(
+            ["2017-01-01T06:00:00", "2017-01-02T06:00:00"], dtype="M8[ns]"
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_interface(self):
+        data = pd.date_range("2017", periods=2)._data
+        expected = np.array(
+            ["2017-01-01T00:00:00", "2017-01-02T00:00:00"], dtype="datetime64[ns]"
+        )
+
+        result = np.asarray(data)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.asarray(data, dtype=object)
+        expected = np.array(
+            [pd.Timestamp("2017-01-01T00:00:00"), pd.Timestamp("2017-01-02T00:00:00")],
+            dtype=object,
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("index", [True, False])
+    def test_searchsorted_different_tz(self, index):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        arr = pd.DatetimeIndex(data, freq="D")._data.tz_localize("Asia/Tokyo")
+        if index:
+            arr = pd.Index(arr)
+
+        expected = arr.searchsorted(arr[2])
+        result = arr.searchsorted(arr[2].tz_convert("UTC"))
+        assert result == expected
+
+        expected = arr.searchsorted(arr[2:6])
+        result = arr.searchsorted(arr[2:6].tz_convert("UTC"))
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("index", [True, False])
+    def test_searchsorted_tzawareness_compat(self, index):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        arr = pd.DatetimeIndex(data, freq="D")._data
+        if index:
+            arr = pd.Index(arr)
+
+        mismatch = arr.tz_localize("Asia/Tokyo")
+
+        msg = "Cannot compare tz-naive and tz-aware datetime-like objects"
+        with pytest.raises(TypeError, match=msg):
+            arr.searchsorted(mismatch[0])
+        with pytest.raises(TypeError, match=msg):
+            arr.searchsorted(mismatch)
+
+        with pytest.raises(TypeError, match=msg):
+            mismatch.searchsorted(arr[0])
+        with pytest.raises(TypeError, match=msg):
+            mismatch.searchsorted(arr)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            1,
+            np.int64(1),
+            1.0,
+            np.timedelta64("NaT"),
+            pd.Timedelta(days=2),
+            "invalid",
+            np.arange(10, dtype="i8") * 24 * 3600 * 10**9,
+            np.arange(10).view("timedelta64[ns]") * 24 * 3600 * 10**9,
+            pd.Timestamp("2021-01-01").to_period("D"),
+        ],
+    )
+    @pytest.mark.parametrize("index", [True, False])
+    def test_searchsorted_invalid_types(self, other, index):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        arr = pd.DatetimeIndex(data, freq="D")._data
+        if index:
+            arr = pd.Index(arr)
+
+        msg = "|".join(
+            [
+                "searchsorted requires compatible dtype or scalar",
+                "value should be a 'Timestamp', 'NaT', or array of those. Got",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            arr.searchsorted(other)
+
+    def test_shift_fill_value(self):
+        dti = pd.date_range("2016-01-01", periods=3)
+
+        dta = dti._data
+        expected = DatetimeArray._from_sequence(
+            np.roll(dta._ndarray, 1), dtype=dti.dtype
+        )
+
+        fv = dta[-1]
+        for fill_value in [fv, fv.to_pydatetime(), fv.to_datetime64()]:
+            result = dta.shift(1, fill_value=fill_value)
+            tm.assert_datetime_array_equal(result, expected)
+
+        dta = dta.tz_localize("UTC")
+        expected = expected.tz_localize("UTC")
+        fv = dta[-1]
+        for fill_value in [fv, fv.to_pydatetime()]:
+            result = dta.shift(1, fill_value=fill_value)
+            tm.assert_datetime_array_equal(result, expected)
+
+    def test_shift_value_tzawareness_mismatch(self):
+        dti = pd.date_range("2016-01-01", periods=3)
+
+        dta = dti._data
+
+        fv = dta[-1].tz_localize("UTC")
+        for invalid in [fv, fv.to_pydatetime()]:
+            with pytest.raises(TypeError, match="Cannot compare"):
+                dta.shift(1, fill_value=invalid)
+
+        dta = dta.tz_localize("UTC")
+        fv = dta[-1].tz_localize(None)
+        for invalid in [fv, fv.to_pydatetime(), fv.to_datetime64()]:
+            with pytest.raises(TypeError, match="Cannot compare"):
+                dta.shift(1, fill_value=invalid)
+
+    def test_shift_requires_tzmatch(self):
+        # pre-2.0 we required exact tz match, in 2.0 we require just
+        #  matching tzawareness
+        dti = pd.date_range("2016-01-01", periods=3, tz="UTC")
+        dta = dti._data
+
+        fill_value = pd.Timestamp("2020-10-18 18:44", tz="US/Pacific")
+
+        result = dta.shift(1, fill_value=fill_value)
+        expected = dta.shift(1, fill_value=fill_value.tz_convert("UTC"))
+        tm.assert_equal(result, expected)
+
+    def test_tz_localize_t2d(self):
+        dti = pd.date_range("1994-05-12", periods=12, tz="US/Pacific")
+        dta = dti._data.reshape(3, 4)
+        result = dta.tz_localize(None)
+
+        expected = dta.ravel().tz_localize(None).reshape(dta.shape)
+        tm.assert_datetime_array_equal(result, expected)
+
+        roundtrip = expected.tz_localize("US/Pacific")
+        tm.assert_datetime_array_equal(roundtrip, dta)
+
+    @pytest.mark.parametrize(
+        "tz", ["US/Eastern", "dateutil/US/Eastern", "pytz/US/Eastern"]
+    )
+    def test_iter_zoneinfo_fold(self, tz):
+        # GH#49684
+        if tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        utc_vals = np.array(
+            [1320552000, 1320555600, 1320559200, 1320562800], dtype=np.int64
+        )
+        utc_vals *= 1_000_000_000
+
+        dta = (
+            DatetimeArray._from_sequence(utc_vals, dtype=np.dtype("M8[ns]"))
+            .tz_localize("UTC")
+            .tz_convert(tz)
+        )
+
+        left = dta[2]
+        right = list(dta)[2]
+        assert str(left) == str(right)
+        # previously there was a bug where with non-pytz right would be
+        #  Timestamp('2011-11-06 01:00:00-0400', tz='US/Eastern')
+        # while left would be
+        #  Timestamp('2011-11-06 01:00:00-0500', tz='US/Eastern')
+        # The .value's would match (so they would compare as equal),
+        #  but the folds would not
+        assert left.utcoffset() == right.utcoffset()
+
+        # The same bug in ints_to_pydatetime affected .astype, so we test
+        #  that here.
+        right2 = dta.astype(object)[2]
+        assert str(left) == str(right2)
+        assert left.utcoffset() == right2.utcoffset()
+
+    @pytest.mark.parametrize(
+        "freq",
+        ["2M", "2SM", "2sm", "2Q", "2Q-SEP", "1Y", "2Y-MAR", "2m", "2q-sep", "2y"],
+    )
+    def test_date_range_frequency_M_Q_Y_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            pd.date_range("1/1/2000", periods=4, freq=freq)
+
+    @pytest.mark.parametrize("freq_depr", ["2MIN", "2nS", "2Us"])
+    def test_date_range_uppercase_frequency_deprecated(self, freq_depr):
+        # GH#9586, GH#54939
+        depr_msg = f"'{freq_depr[1:]}' is deprecated and will be removed in a "
+        f"future version. Please use '{freq_depr.lower()[1:]}' instead."
+
+        expected = pd.date_range("1/1/2000", periods=4, freq=freq_depr.lower())
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            result = pd.date_range("1/1/2000", periods=4, freq=freq_depr)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "freq",
+        [
+            "2ye-mar",
+            "2ys",
+            "2qe",
+            "2qs-feb",
+            "2bqs",
+            "2sms",
+            "2bms",
+            "2cbme",
+            "2me",
+        ],
+    )
+    def test_date_range_lowercase_frequency_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            pd.date_range("1/1/2000", periods=4, freq=freq)
+
+    def test_date_range_lowercase_frequency_deprecated(self):
+        # GH#9586, GH#54939
+        depr_msg = "'w' is deprecated and will be removed in a future version"
+
+        expected = pd.date_range("1/1/2000", periods=4, freq="2W")
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            result = pd.date_range("1/1/2000", periods=4, freq="2w")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["1A", "2A-MAR", "2a-mar"])
+    def test_date_range_frequency_A_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            pd.date_range("1/1/2000", periods=4, freq=freq)
+
+    @pytest.mark.parametrize("freq", ["2H", "2CBH", "2S"])
+    def test_date_range_uppercase_frequency_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            pd.date_range("1/1/2000", periods=4, freq=freq)
+
+
+def test_factorize_sort_without_freq():
+    dta = DatetimeArray._from_sequence([0, 2, 1], dtype="M8[ns]")
+
+    msg = r"call pd.factorize\(obj, sort=True\) instead"
+    with pytest.raises(NotImplementedError, match=msg):
+        dta.factorize(sort=True)
+
+    # Do TimedeltaArray while we're here
+    tda = dta - dta[0]
+    with pytest.raises(NotImplementedError, match=msg):
+        tda.factorize(sort=True)
diff --git a/pandas/pandas/tests/arrays/test_ndarray_backed.py b/pandas/pandas/tests/arrays/test_ndarray_backed.py
new file mode 100644
index 0000000000000..2af59a03a5b3e
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_ndarray_backed.py
@@ -0,0 +1,76 @@
+"""
+Tests for subclasses of NDArrayBackedExtensionArray
+"""
+
+import numpy as np
+
+from pandas import (
+    CategoricalIndex,
+    date_range,
+)
+from pandas.core.arrays import (
+    Categorical,
+    DatetimeArray,
+    NumpyExtensionArray,
+    TimedeltaArray,
+)
+
+
+class TestEmpty:
+    def test_empty_categorical(self):
+        ci = CategoricalIndex(["a", "b", "c"], ordered=True)
+        dtype = ci.dtype
+
+        # case with int8 codes
+        shape = (4,)
+        result = Categorical._empty(shape, dtype=dtype)
+        assert isinstance(result, Categorical)
+        assert result.shape == shape
+        assert result._ndarray.dtype == np.int8
+
+        # case where repr would segfault if we didn't override base implementation
+        result = Categorical._empty((4096,), dtype=dtype)
+        assert isinstance(result, Categorical)
+        assert result.shape == (4096,)
+        assert result._ndarray.dtype == np.int8
+        repr(result)
+
+        # case with int16 codes
+        ci = CategoricalIndex(list(range(512)) * 4, ordered=False)
+        dtype = ci.dtype
+        result = Categorical._empty(shape, dtype=dtype)
+        assert isinstance(result, Categorical)
+        assert result.shape == shape
+        assert result._ndarray.dtype == np.int16
+
+    def test_empty_dt64tz(self):
+        dti = date_range("2016-01-01", periods=2, tz="Asia/Tokyo")
+        dtype = dti.dtype
+
+        shape = (0,)
+        result = DatetimeArray._empty(shape, dtype=dtype)
+        assert result.dtype == dtype
+        assert isinstance(result, DatetimeArray)
+        assert result.shape == shape
+
+    def test_empty_dt64(self):
+        shape = (3, 9)
+        result = DatetimeArray._empty(shape, dtype="datetime64[ns]")
+        assert isinstance(result, DatetimeArray)
+        assert result.shape == shape
+
+    def test_empty_td64(self):
+        shape = (3, 9)
+        result = TimedeltaArray._empty(shape, dtype="m8[ns]")
+        assert isinstance(result, TimedeltaArray)
+        assert result.shape == shape
+
+    def test_empty_pandas_array(self):
+        arr = NumpyExtensionArray(np.array([1, 2]))
+        dtype = arr.dtype
+
+        shape = (3, 9)
+        result = NumpyExtensionArray._empty(shape, dtype=dtype)
+        assert isinstance(result, NumpyExtensionArray)
+        assert result.dtype == dtype
+        assert result.shape == shape
diff --git a/pandas/pandas/tests/arrays/test_period.py b/pandas/pandas/tests/arrays/test_period.py
new file mode 100644
index 0000000000000..48453ba19e9a1
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_period.py
@@ -0,0 +1,184 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+from pandas._libs.tslibs.period import IncompatibleFrequency
+
+from pandas.core.dtypes.base import _registry as registry
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import PeriodArray
+
+# ----------------------------------------------------------------------------
+# Dtype
+
+
+def test_registered():
+    assert PeriodDtype in registry.dtypes
+    result = registry.find("Period[D]")
+    expected = PeriodDtype("D")
+    assert result == expected
+
+
+# ----------------------------------------------------------------------------
+# period_array
+
+
+def test_asi8():
+    result = PeriodArray._from_sequence(["2000", "2001", None], dtype="period[D]").asi8
+    expected = np.array([10957, 11323, iNaT])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_take_raises():
+    arr = PeriodArray._from_sequence(["2000", "2001"], dtype="period[D]")
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        arr.take([0, -1], allow_fill=True, fill_value=pd.Period("2000", freq="W"))
+
+    msg = "value should be a 'Period' or 'NaT'. Got 'str' instead"
+    with pytest.raises(TypeError, match=msg):
+        arr.take([0, -1], allow_fill=True, fill_value="foo")
+
+
+def test_fillna_raises():
+    arr = PeriodArray._from_sequence(["2000", "2001", "2002"], dtype="period[D]")
+    with pytest.raises(ValueError, match="Length"):
+        arr.fillna(arr[:2])
+
+
+def test_fillna_copies():
+    arr = PeriodArray._from_sequence(["2000", "2001", "2002"], dtype="period[D]")
+    result = arr.fillna(pd.Period("2000", "D"))
+    assert result is not arr
+
+
+# ----------------------------------------------------------------------------
+# setitem
+
+
+@pytest.mark.parametrize(
+    "key, value, expected",
+    [
+        ([0], pd.Period("2000", "D"), [10957, 1, 2]),
+        ([0], None, [iNaT, 1, 2]),
+        ([0], np.nan, [iNaT, 1, 2]),
+        ([0, 1, 2], pd.Period("2000", "D"), [10957] * 3),
+        (
+            [0, 1, 2],
+            [pd.Period("2000", "D"), pd.Period("2001", "D"), pd.Period("2002", "D")],
+            [10957, 11323, 11688],
+        ),
+    ],
+)
+def test_setitem(key, value, expected):
+    arr = PeriodArray(np.arange(3), dtype="period[D]")
+    expected = PeriodArray(expected, dtype="period[D]")
+    arr[key] = value
+    tm.assert_period_array_equal(arr, expected)
+
+
+def test_setitem_raises_incompatible_freq():
+    arr = PeriodArray(np.arange(3), dtype="period[D]")
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        arr[0] = pd.Period("2000", freq="Y")
+
+    other = PeriodArray._from_sequence(["2000", "2001"], dtype="period[Y]")
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        arr[[0, 1]] = other
+
+
+def test_setitem_raises_length():
+    arr = PeriodArray(np.arange(3), dtype="period[D]")
+    with pytest.raises(ValueError, match="length"):
+        arr[[0, 1]] = [pd.Period("2000", freq="D")]
+
+
+def test_setitem_raises_type():
+    arr = PeriodArray(np.arange(3), dtype="period[D]")
+    with pytest.raises(TypeError, match="int"):
+        arr[0] = 1
+
+
+# ----------------------------------------------------------------------------
+# Ops
+
+
+def test_sub_period():
+    arr = PeriodArray._from_sequence(["2000", "2001"], dtype="period[D]")
+    other = pd.Period("2000", freq="M")
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        arr - other
+
+
+def test_sub_period_overflow():
+    # GH#47538
+    dti = pd.date_range("1677-09-22", periods=2, freq="D")
+    pi = dti.to_period("ns")
+
+    per = pd.Period._from_ordinal(10**14, pi.freq)
+
+    with pytest.raises(OverflowError, match="Overflow in int64 addition"):
+        pi - per
+
+    with pytest.raises(OverflowError, match="Overflow in int64 addition"):
+        per - pi
+
+
+# ----------------------------------------------------------------------------
+# Methods
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        pd.Period("2000", freq="h"),
+        PeriodArray._from_sequence(["2000", "2001", "2000"], dtype="period[h]"),
+    ],
+)
+def test_where_different_freq_raises(other):
+    # GH#45768 The PeriodArray method raises, the Series method coerces
+    ser = pd.Series(
+        PeriodArray._from_sequence(["2000", "2001", "2002"], dtype="period[D]")
+    )
+    cond = np.array([True, False, True])
+
+    with pytest.raises(IncompatibleFrequency, match="freq"):
+        ser.array._where(cond, other)
+
+    res = ser.where(cond, other)
+    expected = ser.astype(object).where(cond, other)
+    tm.assert_series_equal(res, expected)
+
+
+# ----------------------------------------------------------------------------
+# Printing
+
+
+def test_repr_small():
+    arr = PeriodArray._from_sequence(["2000", "2001"], dtype="period[D]")
+    result = str(arr)
+    expected = (
+        "<PeriodArray>\n['2000-01-01', '2001-01-01']\nLength: 2, dtype: period[D]"
+    )
+    assert result == expected
+
+
+def test_repr_large():
+    arr = PeriodArray._from_sequence(["2000", "2001"] * 500, dtype="period[D]")
+    result = str(arr)
+    expected = (
+        "<PeriodArray>\n"
+        "['2000-01-01', '2001-01-01', '2000-01-01', '2001-01-01', "
+        "'2000-01-01',\n"
+        " '2001-01-01', '2000-01-01', '2001-01-01', '2000-01-01', "
+        "'2001-01-01',\n"
+        " ...\n"
+        " '2000-01-01', '2001-01-01', '2000-01-01', '2001-01-01', "
+        "'2000-01-01',\n"
+        " '2001-01-01', '2000-01-01', '2001-01-01', '2000-01-01', "
+        "'2001-01-01']\n"
+        "Length: 1000, dtype: period[D]"
+    )
+    assert result == expected
diff --git a/pandas/pandas/tests/arrays/test_timedeltas.py b/pandas/pandas/tests/arrays/test_timedeltas.py
new file mode 100644
index 0000000000000..fb7c7afdc6ff9
--- /dev/null
+++ b/pandas/pandas/tests/arrays/test_timedeltas.py
@@ -0,0 +1,312 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Timedelta
+import pandas._testing as tm
+from pandas.core.arrays import (
+    DatetimeArray,
+    TimedeltaArray,
+)
+
+
+class TestNonNano:
+    @pytest.fixture(params=["s", "ms", "us"])
+    def unit(self, request):
+        return request.param
+
+    @pytest.fixture
+    def tda(self, unit):
+        arr = np.arange(5, dtype=np.int64).view(f"m8[{unit}]")
+        return TimedeltaArray._simple_new(arr, dtype=arr.dtype)
+
+    def test_non_nano(self, unit):
+        arr = np.arange(5, dtype=np.int64).view(f"m8[{unit}]")
+        tda = TimedeltaArray._simple_new(arr, dtype=arr.dtype)
+
+        assert tda.dtype == arr.dtype
+        assert tda[0].unit == unit
+
+    def test_as_unit_raises(self, tda):
+        # GH#50616
+        with pytest.raises(ValueError, match="Supported units"):
+            tda.as_unit("D")
+
+        tdi = pd.Index(tda)
+        with pytest.raises(ValueError, match="Supported units"):
+            tdi.as_unit("D")
+
+    @pytest.mark.parametrize("field", TimedeltaArray._field_ops)
+    def test_fields(self, tda, field):
+        as_nano = tda._ndarray.astype("m8[ns]")
+        tda_nano = TimedeltaArray._simple_new(as_nano, dtype=as_nano.dtype)
+
+        result = getattr(tda, field)
+        expected = getattr(tda_nano, field)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_to_pytimedelta(self, tda):
+        as_nano = tda._ndarray.astype("m8[ns]")
+        tda_nano = TimedeltaArray._simple_new(as_nano, dtype=as_nano.dtype)
+
+        result = tda.to_pytimedelta()
+        expected = tda_nano.to_pytimedelta()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_total_seconds(self, unit, tda):
+        as_nano = tda._ndarray.astype("m8[ns]")
+        tda_nano = TimedeltaArray._simple_new(as_nano, dtype=as_nano.dtype)
+
+        result = tda.total_seconds()
+        expected = tda_nano.total_seconds()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_timedelta_array_total_seconds(self):
+        # GH34290
+        expected = Timedelta("2 min").total_seconds()
+
+        result = pd.array([Timedelta("2 min")]).total_seconds()[0]
+        assert result == expected
+
+    def test_total_seconds_nanoseconds(self):
+        # issue #48521
+        start_time = pd.Series(["2145-11-02 06:00:00"]).astype("datetime64[ns]")
+        end_time = pd.Series(["2145-11-02 07:06:00"]).astype("datetime64[ns]")
+        expected = (end_time - start_time).values / np.timedelta64(1, "s")
+        result = (end_time - start_time).dt.total_seconds().values
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "nat", [np.datetime64("NaT", "ns"), np.datetime64("NaT", "us")]
+    )
+    def test_add_nat_datetimelike_scalar(self, nat, tda):
+        result = tda + nat
+        assert isinstance(result, DatetimeArray)
+        assert result._creso == tda._creso
+        assert result.isna().all()
+
+        result = nat + tda
+        assert isinstance(result, DatetimeArray)
+        assert result._creso == tda._creso
+        assert result.isna().all()
+
+    def test_add_pdnat(self, tda):
+        result = tda + pd.NaT
+        assert isinstance(result, TimedeltaArray)
+        assert result._creso == tda._creso
+        assert result.isna().all()
+
+        result = pd.NaT + tda
+        assert isinstance(result, TimedeltaArray)
+        assert result._creso == tda._creso
+        assert result.isna().all()
+
+    # TODO: 2022-07-11 this is the only test that gets to DTA.tz_convert
+    #  or tz_localize with non-nano; implement tests specific to that.
+    def test_add_datetimelike_scalar(self, tda, tz_naive_fixture):
+        ts = pd.Timestamp("2016-01-01", tz=tz_naive_fixture).as_unit("ns")
+
+        expected = tda.as_unit("ns") + ts
+        res = tda + ts
+        tm.assert_extension_array_equal(res, expected)
+        res = ts + tda
+        tm.assert_extension_array_equal(res, expected)
+
+        ts += Timedelta(1)  # case where we can't cast losslessly
+
+        exp_values = tda._ndarray + ts.asm8
+        expected = (
+            DatetimeArray._simple_new(exp_values, dtype=exp_values.dtype)
+            .tz_localize("UTC")
+            .tz_convert(ts.tz)
+        )
+
+        result = tda + ts
+        tm.assert_extension_array_equal(result, expected)
+
+        result = ts + tda
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_mul_scalar(self, tda):
+        other = 2
+        result = tda * other
+        expected = TimedeltaArray._simple_new(tda._ndarray * other, dtype=tda.dtype)
+        tm.assert_extension_array_equal(result, expected)
+        assert result._creso == tda._creso
+
+    def test_mul_listlike(self, tda):
+        other = np.arange(len(tda))
+        result = tda * other
+        expected = TimedeltaArray._simple_new(tda._ndarray * other, dtype=tda.dtype)
+        tm.assert_extension_array_equal(result, expected)
+        assert result._creso == tda._creso
+
+    def test_mul_listlike_object(self, tda):
+        other = np.arange(len(tda))
+        result = tda * other.astype(object)
+        expected = TimedeltaArray._simple_new(tda._ndarray * other, dtype=tda.dtype)
+        tm.assert_extension_array_equal(result, expected)
+        assert result._creso == tda._creso
+
+    def test_div_numeric_scalar(self, tda):
+        other = 2
+        result = tda / other
+        expected = TimedeltaArray._simple_new(tda._ndarray / other, dtype=tda.dtype)
+        tm.assert_extension_array_equal(result, expected)
+        assert result._creso == tda._creso
+
+    def test_div_td_scalar(self, tda):
+        other = timedelta(seconds=1)
+        result = tda / other
+        expected = tda._ndarray / np.timedelta64(1, "s")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_div_numeric_array(self, tda):
+        other = np.arange(len(tda))
+        result = tda / other
+        expected = TimedeltaArray._simple_new(tda._ndarray / other, dtype=tda.dtype)
+        tm.assert_extension_array_equal(result, expected)
+        assert result._creso == tda._creso
+
+    def test_div_td_array(self, tda):
+        other = tda._ndarray + tda._ndarray[-1]
+        result = tda / other
+        expected = tda._ndarray / other
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_add_timedeltaarraylike(self, tda):
+        tda_nano = tda.astype("m8[ns]")
+
+        expected = tda_nano * 2
+        res = tda_nano + tda
+        tm.assert_extension_array_equal(res, expected)
+        res = tda + tda_nano
+        tm.assert_extension_array_equal(res, expected)
+
+        expected = tda_nano * 0
+        res = tda - tda_nano
+        tm.assert_extension_array_equal(res, expected)
+
+        res = tda_nano - tda
+        tm.assert_extension_array_equal(res, expected)
+
+
+class TestTimedeltaArray:
+    def test_astype_int(self, any_int_numpy_dtype):
+        arr = TimedeltaArray._from_sequence(
+            [Timedelta("1h"), Timedelta("2h")], dtype="m8[ns]"
+        )
+
+        if np.dtype(any_int_numpy_dtype) != np.int64:
+            with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+                arr.astype(any_int_numpy_dtype)
+            return
+
+        result = arr.astype(any_int_numpy_dtype)
+        expected = arr._ndarray.view("i8")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_setitem_clears_freq(self):
+        a = pd.timedelta_range("1h", periods=2, freq="h")._data
+        a[0] = Timedelta("1h")
+        assert a.freq is None
+
+    @pytest.mark.parametrize(
+        "obj",
+        [
+            Timedelta(seconds=1),
+            Timedelta(seconds=1).to_timedelta64(),
+            Timedelta(seconds=1).to_pytimedelta(),
+        ],
+    )
+    def test_setitem_objects(self, obj):
+        # make sure we accept timedelta64 and timedelta in addition to Timedelta
+        tdi = pd.timedelta_range("2 Days", periods=4, freq="h")
+        arr = tdi._data
+
+        arr[0] = obj
+        assert arr[0] == Timedelta(seconds=1)
+
+    @pytest.mark.parametrize(
+        "other",
+        [
+            1,
+            np.int64(1),
+            1.0,
+            np.datetime64("NaT"),
+            pd.Timestamp("2021-01-01"),
+            "invalid",
+            np.arange(10, dtype="i8") * 24 * 3600 * 10**9,
+            (np.arange(10) * 24 * 3600 * 10**9).view("datetime64[ns]"),
+            pd.Timestamp("2021-01-01").to_period("D"),
+        ],
+    )
+    @pytest.mark.parametrize("index", [True, False])
+    def test_searchsorted_invalid_types(self, other, index):
+        data = np.arange(10, dtype="i8") * 24 * 3600 * 10**9
+        arr = pd.TimedeltaIndex(data, freq="D")._data
+        if index:
+            arr = pd.Index(arr)
+
+        msg = "|".join(
+            [
+                "searchsorted requires compatible dtype or scalar",
+                "value should be a 'Timedelta', 'NaT', or array of those. Got",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            arr.searchsorted(other)
+
+
+class TestUnaryOps:
+    def test_abs(self):
+        vals = np.array([-3600 * 10**9, "NaT", 7200 * 10**9], dtype="m8[ns]")
+        arr = TimedeltaArray._from_sequence(vals, dtype=vals.dtype)
+
+        evals = np.array([3600 * 10**9, "NaT", 7200 * 10**9], dtype="m8[ns]")
+        expected = TimedeltaArray._from_sequence(evals, dtype=evals.dtype)
+
+        result = abs(arr)
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result2 = np.abs(arr)
+        tm.assert_timedelta_array_equal(result2, expected)
+
+    def test_pos(self):
+        vals = np.array([-3600 * 10**9, "NaT", 7200 * 10**9], dtype="m8[ns]")
+        arr = TimedeltaArray._from_sequence(vals, dtype=vals.dtype)
+
+        result = +arr
+        tm.assert_timedelta_array_equal(result, arr)
+        assert not tm.shares_memory(result, arr)
+
+        result2 = np.positive(arr)
+        tm.assert_timedelta_array_equal(result2, arr)
+        assert not tm.shares_memory(result2, arr)
+
+    def test_neg(self):
+        vals = np.array([-3600 * 10**9, "NaT", 7200 * 10**9], dtype="m8[ns]")
+        arr = TimedeltaArray._from_sequence(vals, dtype=vals.dtype)
+
+        evals = np.array([3600 * 10**9, "NaT", -7200 * 10**9], dtype="m8[ns]")
+        expected = TimedeltaArray._from_sequence(evals)
+
+        result = -arr
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result2 = np.negative(arr)
+        tm.assert_timedelta_array_equal(result2, expected)
+
+    def test_neg_freq(self):
+        tdi = pd.timedelta_range("2 Days", periods=4, freq="h")
+        arr = tdi._data
+
+        expected = -tdi._data
+
+        result = -arr
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result2 = np.negative(arr)
+        tm.assert_timedelta_array_equal(result2, expected)
diff --git a/pandas/pandas/tests/arrays/timedeltas/__init__.py b/pandas/pandas/tests/arrays/timedeltas/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/arrays/timedeltas/test_constructors.py b/pandas/pandas/tests/arrays/timedeltas/test_constructors.py
new file mode 100644
index 0000000000000..ee29f505fd7b1
--- /dev/null
+++ b/pandas/pandas/tests/arrays/timedeltas/test_constructors.py
@@ -0,0 +1,58 @@
+import numpy as np
+import pytest
+
+from pandas.core.arrays import TimedeltaArray
+
+
+class TestTimedeltaArrayConstructor:
+    def test_other_type_raises(self):
+        msg = r"dtype bool cannot be converted to timedelta64\[ns\]"
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaArray._from_sequence(np.array([1, 2, 3], dtype="bool"))
+
+    def test_incorrect_dtype_raises(self):
+        msg = "dtype 'category' is invalid, should be np.timedelta64 dtype"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence(
+                np.array([1, 2, 3], dtype="i8"), dtype="category"
+            )
+
+        msg = "dtype 'int64' is invalid, should be np.timedelta64 dtype"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence(
+                np.array([1, 2, 3], dtype="i8"), dtype=np.dtype("int64")
+            )
+
+        msg = r"dtype 'datetime64\[ns\]' is invalid, should be np.timedelta64 dtype"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence(
+                np.array([1, 2, 3], dtype="i8"), dtype=np.dtype("M8[ns]")
+            )
+
+        msg = (
+            r"dtype 'datetime64\[us, UTC\]' is invalid, should be np.timedelta64 dtype"
+        )
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence(
+                np.array([1, 2, 3], dtype="i8"), dtype="M8[us, UTC]"
+            )
+
+        msg = "Supported timedelta64 resolutions are 's', 'ms', 'us', 'ns'"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence(
+                np.array([1, 2, 3], dtype="i8"), dtype=np.dtype("m8[Y]")
+            )
+
+    def test_copy(self):
+        data = np.array([1, 2, 3], dtype="m8[ns]")
+        arr = TimedeltaArray._from_sequence(data, copy=False)
+        assert arr._ndarray is data
+
+        arr = TimedeltaArray._from_sequence(data, copy=True)
+        assert arr._ndarray is not data
+        assert arr._ndarray.base is not data
+
+    def test_from_sequence_dtype(self):
+        msg = "dtype 'object' is invalid, should be np.timedelta64 dtype"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaArray._from_sequence([], dtype=object)
diff --git a/pandas/pandas/tests/arrays/timedeltas/test_cumulative.py b/pandas/pandas/tests/arrays/timedeltas/test_cumulative.py
new file mode 100644
index 0000000000000..2d8fe65f807e4
--- /dev/null
+++ b/pandas/pandas/tests/arrays/timedeltas/test_cumulative.py
@@ -0,0 +1,20 @@
+import pytest
+
+import pandas._testing as tm
+from pandas.core.arrays import TimedeltaArray
+
+
+class TestAccumulator:
+    def test_accumulators_disallowed(self):
+        # GH#50297
+        arr = TimedeltaArray._from_sequence(["1D", "2D"], dtype="m8[ns]")
+        with pytest.raises(TypeError, match="cumprod not supported"):
+            arr._accumulate("cumprod")
+
+    def test_cumsum(self, unit):
+        # GH#50297
+        dtype = f"m8[{unit}]"
+        arr = TimedeltaArray._from_sequence(["1D", "2D"], dtype=dtype)
+        result = arr._accumulate("cumsum")
+        expected = TimedeltaArray._from_sequence(["1D", "3D"], dtype=dtype)
+        tm.assert_timedelta_array_equal(result, expected)
diff --git a/pandas/pandas/tests/arrays/timedeltas/test_reductions.py b/pandas/pandas/tests/arrays/timedeltas/test_reductions.py
new file mode 100644
index 0000000000000..3565f2e8690e2
--- /dev/null
+++ b/pandas/pandas/tests/arrays/timedeltas/test_reductions.py
@@ -0,0 +1,216 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Timedelta
+import pandas._testing as tm
+from pandas.core import nanops
+from pandas.core.arrays import TimedeltaArray
+
+
+class TestReductions:
+    @pytest.mark.parametrize("name", ["std", "min", "max", "median", "mean"])
+    def test_reductions_empty(self, name, skipna):
+        tdi = pd.TimedeltaIndex([])
+        arr = tdi.array
+
+        result = getattr(tdi, name)(skipna=skipna)
+        assert result is pd.NaT
+
+        result = getattr(arr, name)(skipna=skipna)
+        assert result is pd.NaT
+
+    def test_sum_empty(self, skipna):
+        tdi = pd.TimedeltaIndex([])
+        arr = tdi.array
+
+        result = tdi.sum(skipna=skipna)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(0)
+
+        result = arr.sum(skipna=skipna)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(0)
+
+    def test_min_max(self, unit):
+        dtype = f"m8[{unit}]"
+        arr = TimedeltaArray._from_sequence(
+            ["3h", "3h", "NaT", "2h", "5h", "4h"], dtype=dtype
+        )
+
+        result = arr.min()
+        expected = Timedelta("2h")
+        assert result == expected
+
+        result = arr.max()
+        expected = Timedelta("5h")
+        assert result == expected
+
+        result = arr.min(skipna=False)
+        assert result is pd.NaT
+
+        result = arr.max(skipna=False)
+        assert result is pd.NaT
+
+    def test_sum(self):
+        tdi = pd.TimedeltaIndex(["3h", "3h", "NaT", "2h", "5h", "4h"])
+        arr = tdi.array
+
+        result = arr.sum(skipna=True)
+        expected = Timedelta(hours=17)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = tdi.sum(skipna=True)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = arr.sum(skipna=False)
+        assert result is pd.NaT
+
+        result = tdi.sum(skipna=False)
+        assert result is pd.NaT
+
+        result = arr.sum(min_count=9)
+        assert result is pd.NaT
+
+        result = tdi.sum(min_count=9)
+        assert result is pd.NaT
+
+        result = arr.sum(min_count=1)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = tdi.sum(min_count=1)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+    def test_npsum(self):
+        # GH#25282, GH#25335 np.sum should return a Timedelta, not timedelta64
+        tdi = pd.TimedeltaIndex(["3h", "3h", "2h", "5h", "4h"])
+        arr = tdi.array
+
+        result = np.sum(tdi)
+        expected = Timedelta(hours=17)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = np.sum(arr)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+    def test_sum_2d_skipna_false(self):
+        arr = np.arange(8).astype(np.int64).view("m8[s]").astype("m8[ns]").reshape(4, 2)
+        arr[-1, -1] = "Nat"
+
+        tda = TimedeltaArray._from_sequence(arr)
+
+        result = tda.sum(skipna=False)
+        assert result is pd.NaT
+
+        result = tda.sum(axis=0, skipna=False)
+        expected = pd.TimedeltaIndex([Timedelta(seconds=12), pd.NaT])._values
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result = tda.sum(axis=1, skipna=False)
+        expected = pd.TimedeltaIndex(
+            [
+                Timedelta(seconds=1),
+                Timedelta(seconds=5),
+                Timedelta(seconds=9),
+                pd.NaT,
+            ]
+        )._values
+        tm.assert_timedelta_array_equal(result, expected)
+
+    # Adding a Timestamp makes this a test for DatetimeArray.std
+    @pytest.mark.parametrize(
+        "add",
+        [
+            Timedelta(0),
+            pd.Timestamp("2021-01-01"),
+            pd.Timestamp("2021-01-01", tz="UTC"),
+            pd.Timestamp("2021-01-01", tz="Asia/Tokyo"),
+        ],
+    )
+    def test_std(self, add):
+        tdi = pd.TimedeltaIndex(["0h", "4h", "NaT", "4h", "0h", "2h"]) + add
+        arr = tdi.array
+
+        result = arr.std(skipna=True)
+        expected = Timedelta(hours=2)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = tdi.std(skipna=True)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        if getattr(arr, "tz", None) is None:
+            result = nanops.nanstd(np.asarray(arr), skipna=True)
+            assert isinstance(result, np.timedelta64)
+            assert result == expected
+
+        result = arr.std(skipna=False)
+        assert result is pd.NaT
+
+        result = tdi.std(skipna=False)
+        assert result is pd.NaT
+
+        if getattr(arr, "tz", None) is None:
+            result = nanops.nanstd(np.asarray(arr), skipna=False)
+            assert isinstance(result, np.timedelta64)
+            assert np.isnat(result)
+
+    def test_median(self):
+        tdi = pd.TimedeltaIndex(["0h", "3h", "NaT", "5h06m", "0h", "2h"])
+        arr = tdi.array
+
+        result = arr.median(skipna=True)
+        expected = Timedelta(hours=2)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = tdi.median(skipna=True)
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = arr.median(skipna=False)
+        assert result is pd.NaT
+
+        result = tdi.median(skipna=False)
+        assert result is pd.NaT
+
+    def test_mean(self):
+        tdi = pd.TimedeltaIndex(["0h", "3h", "NaT", "5h06m", "0h", "2h"])
+        arr = tdi._data
+
+        # manually verified result
+        expected = Timedelta(arr.dropna()._ndarray.mean())
+
+        result = arr.mean()
+        assert result == expected
+        result = arr.mean(skipna=False)
+        assert result is pd.NaT
+
+        result = arr.dropna().mean(skipna=False)
+        assert result == expected
+
+        result = arr.mean(axis=0)
+        assert result == expected
+
+    def test_mean_2d(self):
+        tdi = pd.timedelta_range("14 days", periods=6)
+        tda = tdi._data.reshape(3, 2)
+
+        result = tda.mean(axis=0)
+        expected = tda[1]
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result = tda.mean(axis=1)
+        expected = tda[:, 0] + Timedelta(hours=12)
+        tm.assert_timedelta_array_equal(result, expected)
+
+        result = tda.mean(axis=None)
+        expected = tdi.mean()
+        assert result == expected
diff --git a/pandas/pandas/tests/base/__init__.py b/pandas/pandas/tests/base/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/base/common.py b/pandas/pandas/tests/base/common.py
new file mode 100644
index 0000000000000..ad0b394105742
--- /dev/null
+++ b/pandas/pandas/tests/base/common.py
@@ -0,0 +1,9 @@
+from typing import Any
+
+from pandas import Index
+
+
+def allow_na_ops(obj: Any) -> bool:
+    """Whether to skip test cases including NaN"""
+    is_bool_index = isinstance(obj, Index) and obj.inferred_type == "boolean"
+    return not is_bool_index and obj._can_hold_na
diff --git a/pandas/pandas/tests/base/test_constructors.py b/pandas/pandas/tests/base/test_constructors.py
new file mode 100644
index 0000000000000..dffd2009ef373
--- /dev/null
+++ b/pandas/pandas/tests/base/test_constructors.py
@@ -0,0 +1,192 @@
+from datetime import datetime
+import sys
+
+import numpy as np
+import pytest
+
+from pandas.compat import PYPY
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.accessor import PandasDelegate
+from pandas.core.base import (
+    NoNewAttributesMixin,
+    PandasObject,
+)
+
+
+def series_via_frame_from_dict(x, **kwargs):
+    return DataFrame({"a": x}, **kwargs)["a"]
+
+
+def series_via_frame_from_scalar(x, **kwargs):
+    return DataFrame(x, **kwargs)[0]
+
+
+@pytest.fixture(
+    params=[
+        Series,
+        series_via_frame_from_dict,
+        series_via_frame_from_scalar,
+        Index,
+    ],
+    ids=["Series", "DataFrame-dict", "DataFrame-array", "Index"],
+)
+def constructor(request):
+    return request.param
+
+
+class TestPandasDelegate:
+    class Delegator:
+        _properties = ["prop"]
+        _methods = ["test_method"]
+
+        def _set_prop(self, value):
+            self.prop = value
+
+        def _get_prop(self):
+            return self.prop
+
+        prop = property(_get_prop, _set_prop, doc="foo property")
+
+        def test_method(self, *args, **kwargs):
+            """a test method"""
+
+    class Delegate(PandasDelegate, PandasObject):
+        def __init__(self, obj) -> None:
+            self.obj = obj
+
+    def test_invalid_delegation(self):
+        # these show that in order for the delegation to work
+        # the _delegate_* methods need to be overridden to not raise
+        # a TypeError
+
+        self.Delegate._add_delegate_accessors(
+            delegate=self.Delegator,
+            accessors=self.Delegator._properties,
+            typ="property",
+        )
+        self.Delegate._add_delegate_accessors(
+            delegate=self.Delegator, accessors=self.Delegator._methods, typ="method"
+        )
+
+        delegate = self.Delegate(self.Delegator())
+
+        msg = "You cannot access the property prop"
+        with pytest.raises(TypeError, match=msg):
+            delegate.prop
+
+        msg = "The property prop cannot be set"
+        with pytest.raises(TypeError, match=msg):
+            delegate.prop = 5
+
+        msg = "You cannot access the property prop"
+        with pytest.raises(TypeError, match=msg):
+            delegate.prop
+
+    @pytest.mark.skipif(PYPY, reason="not relevant for PyPy")
+    def test_memory_usage(self):
+        # Delegate does not implement memory_usage.
+        # Check that we fall back to in-built `__sizeof__`
+        # GH 12924
+        delegate = self.Delegate(self.Delegator())
+        sys.getsizeof(delegate)
+
+
+class TestNoNewAttributesMixin:
+    def test_mixin(self):
+        class T(NoNewAttributesMixin):
+            pass
+
+        t = T()
+        assert not hasattr(t, "__frozen")
+
+        t.a = "test"
+        assert t.a == "test"
+
+        t._freeze()
+        assert "__frozen" in dir(t)
+        assert getattr(t, "__frozen")
+        msg = "You cannot add any new attribute"
+        with pytest.raises(AttributeError, match=msg):
+            t.b = "test"
+
+        assert not hasattr(t, "b")
+
+
+class TestConstruction:
+    # test certain constructor behaviours on dtype inference across Series,
+    # Index and DataFrame
+
+    @pytest.mark.parametrize(
+        "a",
+        [
+            np.array(["2263-01-01"], dtype="datetime64[D]"),
+            np.array([datetime(2263, 1, 1)], dtype=object),
+            np.array([np.datetime64("2263-01-01", "D")], dtype=object),
+            np.array(["2263-01-01"], dtype=object),
+        ],
+        ids=[
+            "datetime64[D]",
+            "object-datetime.datetime",
+            "object-numpy-scalar",
+            "object-string",
+        ],
+    )
+    def test_constructor_datetime_outofbound(
+        self, a, constructor, request, using_infer_string
+    ):
+        # GH-26853 (+ bug GH-26206 out of bound non-ns unit)
+
+        # No dtype specified (dtype inference)
+        # datetime64[non-ns] raise error, other cases result in object dtype
+        # and preserve original data
+        result = constructor(a)
+        if a.dtype.kind == "M" or isinstance(a[0], np.datetime64):
+            # Can't fit in nanosecond bounds -> get the nearest supported unit
+            assert result.dtype == "M8[s]"
+        elif isinstance(a[0], datetime):
+            assert result.dtype == "M8[us]", result.dtype
+        else:
+            result = constructor(a)
+            if using_infer_string and "object-string" in request.node.callspec.id:
+                assert result.dtype == "string"
+            else:
+                assert result.dtype == "object"
+            tm.assert_numpy_array_equal(result.to_numpy(), a)
+
+        # Explicit dtype specified
+        # Forced conversion fails for all -> all cases raise error
+        msg = "Out of bounds|Out of bounds .* present at position 0"
+        with pytest.raises(pd.errors.OutOfBoundsDatetime, match=msg):
+            constructor(a, dtype="datetime64[ns]")
+
+    def test_constructor_datetime_nonns(self, constructor):
+        arr = np.array(["2020-01-01T00:00:00.000000"], dtype="datetime64[us]")
+        dta = pd.core.arrays.DatetimeArray._simple_new(arr, dtype=arr.dtype)
+        expected = constructor(dta)
+        assert expected.dtype == arr.dtype
+
+        result = constructor(arr)
+        tm.assert_equal(result, expected)
+
+        # https://github.com/pandas-dev/pandas/issues/34843
+        arr.flags.writeable = False
+        result = constructor(arr)
+        tm.assert_equal(result, expected)
+
+    def test_constructor_from_dict_keys(self, constructor, using_infer_string):
+        # https://github.com/pandas-dev/pandas/issues/60343
+        d = {"a": 1, "b": 2}
+        result = constructor(d.keys(), dtype="str")
+        if using_infer_string:
+            assert result.dtype == "str"
+        else:
+            assert result.dtype == "object"
+        expected = constructor(list(d.keys()), dtype="str")
+        tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/base/test_conversion.py b/pandas/pandas/tests/base/test_conversion.py
new file mode 100644
index 0000000000000..e3a821519c638
--- /dev/null
+++ b/pandas/pandas/tests/base/test_conversion.py
@@ -0,0 +1,598 @@
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    DatetimeArray,
+    IntervalArray,
+    NumpyExtensionArray,
+    PeriodArray,
+    SparseArray,
+    TimedeltaArray,
+)
+from pandas.core.arrays.string_ import StringArrayNumpySemantics
+from pandas.core.arrays.string_arrow import ArrowStringArrayNumpySemantics
+
+
+class TestToIterable:
+    # test that we convert an iterable to python types
+
+    dtypes = [
+        ("int8", int),
+        ("int16", int),
+        ("int32", int),
+        ("int64", int),
+        ("uint8", int),
+        ("uint16", int),
+        ("uint32", int),
+        ("uint64", int),
+        ("float16", float),
+        ("float32", float),
+        ("float64", float),
+        ("datetime64[ns]", Timestamp),
+        ("datetime64[ns, US/Eastern]", Timestamp),
+        ("timedelta64[ns]", Timedelta),
+    ]
+
+    @pytest.mark.parametrize("dtype, rdtype", dtypes)
+    @pytest.mark.parametrize(
+        "method",
+        [
+            lambda x: x.tolist(),
+            lambda x: x.to_list(),
+            lambda x: list(x),
+            lambda x: list(x.__iter__()),
+        ],
+        ids=["tolist", "to_list", "list", "iter"],
+    )
+    def test_iterable(self, index_or_series, method, dtype, rdtype):
+        # gh-10904
+        # gh-13258
+        # coerce iteration to underlying python / pandas types
+        typ = index_or_series
+        if dtype == "float16" and issubclass(typ, pd.Index):
+            with pytest.raises(NotImplementedError, match="float16 indexes are not "):
+                typ([1], dtype=dtype)
+            return
+        s = typ([1], dtype=dtype)
+        result = method(s)[0]
+        assert isinstance(result, rdtype)
+
+    @pytest.mark.parametrize(
+        "dtype, rdtype, obj",
+        [
+            ("object", object, "a"),
+            ("object", int, 1),
+            ("category", object, "a"),
+            ("category", int, 1),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "method",
+        [
+            lambda x: x.tolist(),
+            lambda x: x.to_list(),
+            lambda x: list(x),
+            lambda x: list(x.__iter__()),
+        ],
+        ids=["tolist", "to_list", "list", "iter"],
+    )
+    def test_iterable_object_and_category(
+        self, index_or_series, method, dtype, rdtype, obj
+    ):
+        # gh-10904
+        # gh-13258
+        # coerce iteration to underlying python / pandas types
+        typ = index_or_series
+        s = typ([obj], dtype=dtype)
+        result = method(s)[0]
+        assert isinstance(result, rdtype)
+
+    @pytest.mark.parametrize("dtype, rdtype", dtypes)
+    def test_iterable_items(self, dtype, rdtype):
+        # gh-13258
+        # test if items yields the correct boxed scalars
+        # this only applies to series
+        s = Series([1], dtype=dtype)
+        _, result = next(iter(s.items()))
+        assert isinstance(result, rdtype)
+
+        _, result = next(iter(s.items()))
+        assert isinstance(result, rdtype)
+
+    @pytest.mark.parametrize(
+        "dtype, rdtype", dtypes + [("object", int), ("category", int)]
+    )
+    def test_iterable_map(self, index_or_series, dtype, rdtype):
+        # gh-13236
+        # coerce iteration to underlying python / pandas types
+        typ = index_or_series
+        if dtype == "float16" and issubclass(typ, pd.Index):
+            with pytest.raises(NotImplementedError, match="float16 indexes are not "):
+                typ([1], dtype=dtype)
+            return
+        s = typ([1], dtype=dtype)
+        result = s.map(type)[0]
+        if not isinstance(rdtype, tuple):
+            rdtype = (rdtype,)
+        assert result in rdtype
+
+    @pytest.mark.parametrize(
+        "method",
+        [
+            lambda x: x.tolist(),
+            lambda x: x.to_list(),
+            lambda x: list(x),
+            lambda x: list(x.__iter__()),
+        ],
+        ids=["tolist", "to_list", "list", "iter"],
+    )
+    def test_categorial_datetimelike(self, method):
+        i = CategoricalIndex([Timestamp("1999-12-31"), Timestamp("2000-12-31")])
+
+        result = method(i)[0]
+        assert isinstance(result, Timestamp)
+
+    def test_iter_box_dt64(self, unit):
+        vals = [Timestamp("2011-01-01"), Timestamp("2011-01-02")]
+        ser = Series(vals).dt.as_unit(unit)
+        assert ser.dtype == f"datetime64[{unit}]"
+        for res, exp in zip(ser, vals):
+            assert isinstance(res, Timestamp)
+            assert res.tz is None
+            assert res == exp
+            assert res.unit == unit
+
+    def test_iter_box_dt64tz(self, unit):
+        vals = [
+            Timestamp("2011-01-01", tz="US/Eastern"),
+            Timestamp("2011-01-02", tz="US/Eastern"),
+        ]
+        ser = Series(vals).dt.as_unit(unit)
+
+        assert ser.dtype == f"datetime64[{unit}, US/Eastern]"
+        for res, exp in zip(ser, vals):
+            assert isinstance(res, Timestamp)
+            assert res.tz == exp.tz
+            assert res == exp
+            assert res.unit == unit
+
+    def test_iter_box_timedelta64(self, unit):
+        # timedelta
+        vals = [Timedelta("1 days"), Timedelta("2 days")]
+        ser = Series(vals).dt.as_unit(unit)
+        assert ser.dtype == f"timedelta64[{unit}]"
+        for res, exp in zip(ser, vals):
+            assert isinstance(res, Timedelta)
+            assert res == exp
+            assert res.unit == unit
+
+    def test_iter_box_period(self):
+        # period
+        vals = [pd.Period("2011-01-01", freq="M"), pd.Period("2011-01-02", freq="M")]
+        s = Series(vals)
+        assert s.dtype == "Period[M]"
+        for res, exp in zip(s, vals):
+            assert isinstance(res, pd.Period)
+            assert res.freq == "ME"
+            assert res == exp
+
+
+@pytest.mark.parametrize(
+    "arr, expected_type, dtype",
+    [
+        (np.array([0, 1], dtype=np.int64), np.ndarray, "int64"),
+        (np.array(["a", "b"]), np.ndarray, "object"),
+        (pd.Categorical(["a", "b"]), pd.Categorical, "category"),
+        (
+            pd.DatetimeIndex(["2017", "2018"], tz="US/Central"),
+            DatetimeArray,
+            "datetime64[ns, US/Central]",
+        ),
+        (
+            pd.PeriodIndex([2018, 2019], freq="Y"),
+            PeriodArray,
+            pd.core.dtypes.dtypes.PeriodDtype("Y-DEC"),
+        ),
+        (pd.IntervalIndex.from_breaks([0, 1, 2]), IntervalArray, "interval"),
+        (
+            pd.DatetimeIndex(["2017", "2018"]),
+            DatetimeArray,
+            "datetime64[ns]",
+        ),
+        (
+            pd.TimedeltaIndex([10**10]),
+            TimedeltaArray,
+            "m8[ns]",
+        ),
+    ],
+)
+def test_values_consistent(arr, expected_type, dtype, using_infer_string):
+    if using_infer_string and dtype == "object":
+        expected_type = (
+            ArrowStringArrayNumpySemantics if HAS_PYARROW else StringArrayNumpySemantics
+        )
+    l_values = Series(arr)._values
+    r_values = pd.Index(arr)._values
+    assert type(l_values) is expected_type
+    assert type(l_values) is type(r_values)
+
+    tm.assert_equal(l_values, r_values)
+
+
+@pytest.mark.parametrize("arr", [np.array([1, 2, 3])])
+def test_numpy_array(arr):
+    ser = Series(arr)
+    result = ser.array
+    expected = NumpyExtensionArray(arr)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_numpy_array_all_dtypes(any_numpy_dtype):
+    ser = Series(dtype=any_numpy_dtype)
+    result = ser.array
+    if np.dtype(any_numpy_dtype).kind == "M":
+        assert isinstance(result, DatetimeArray)
+    elif np.dtype(any_numpy_dtype).kind == "m":
+        assert isinstance(result, TimedeltaArray)
+    else:
+        assert isinstance(result, NumpyExtensionArray)
+
+
+@pytest.mark.parametrize(
+    "arr, attr",
+    [
+        (pd.Categorical(["a", "b"]), "_codes"),
+        (PeriodArray._from_sequence(["2000", "2001"], dtype="period[D]"), "_ndarray"),
+        (pd.array([0, np.nan], dtype="Int64"), "_data"),
+        (IntervalArray.from_breaks([0, 1]), "_left"),
+        (SparseArray([0, 1]), "_sparse_values"),
+        (
+            DatetimeArray._from_sequence(np.array([1, 2], dtype="datetime64[ns]")),
+            "_ndarray",
+        ),
+        # tz-aware Datetime
+        (
+            DatetimeArray._from_sequence(
+                np.array(
+                    ["2000-01-01T12:00:00", "2000-01-02T12:00:00"], dtype="M8[ns]"
+                ),
+                dtype=DatetimeTZDtype(tz="US/Central"),
+            ),
+            "_ndarray",
+        ),
+    ],
+)
+def test_array(arr, attr, index_or_series):
+    box = index_or_series
+
+    result = box(arr, copy=False).array
+
+    if attr:
+        arr = getattr(arr, attr)
+        result = getattr(result, attr)
+
+    assert result is arr
+
+
+def test_array_multiindex_raises():
+    idx = pd.MultiIndex.from_product([["A"], ["a", "b"]])
+    msg = "MultiIndex has no single backing array"
+    with pytest.raises(ValueError, match=msg):
+        idx.array
+
+
+@pytest.mark.parametrize(
+    "arr, expected, zero_copy",
+    [
+        (np.array([1, 2], dtype=np.int64), np.array([1, 2], dtype=np.int64), True),
+        (pd.Categorical(["a", "b"]), np.array(["a", "b"], dtype=object), False),
+        (
+            pd.core.arrays.period_array(["2000", "2001"], freq="D"),
+            np.array([pd.Period("2000", freq="D"), pd.Period("2001", freq="D")]),
+            False,
+        ),
+        (pd.array([0, np.nan], dtype="Int64"), np.array([0, np.nan]), False),
+        (
+            IntervalArray.from_breaks([0, 1, 2]),
+            np.array([pd.Interval(0, 1), pd.Interval(1, 2)], dtype=object),
+            False,
+        ),
+        (SparseArray([0, 1]), np.array([0, 1], dtype=np.int64), False),
+        # tz-naive datetime
+        (
+            DatetimeArray._from_sequence(np.array(["2000", "2001"], dtype="M8[ns]")),
+            np.array(["2000", "2001"], dtype="M8[ns]"),
+            True,
+        ),
+        # tz-aware stays tz`-aware
+        (
+            DatetimeArray._from_sequence(
+                np.array(["2000-01-01T06:00:00", "2000-01-02T06:00:00"], dtype="M8[ns]")
+            )
+            .tz_localize("UTC")
+            .tz_convert("US/Central"),
+            np.array(
+                [
+                    Timestamp("2000-01-01", tz="US/Central"),
+                    Timestamp("2000-01-02", tz="US/Central"),
+                ]
+            ),
+            False,
+        ),
+        # Timedelta
+        (
+            TimedeltaArray._from_sequence(
+                np.array([0, 3600000000000], dtype="i8").view("m8[ns]"),
+                dtype=np.dtype("m8[ns]"),
+            ),
+            np.array([0, 3600000000000], dtype="m8[ns]"),
+            True,
+        ),
+        # GH#26406 tz is preserved in Categorical[dt64tz]
+        (
+            pd.Categorical(date_range("2016-01-01", periods=2, tz="US/Pacific")),
+            np.array(
+                [
+                    Timestamp("2016-01-01", tz="US/Pacific"),
+                    Timestamp("2016-01-02", tz="US/Pacific"),
+                ]
+            ),
+            False,
+        ),
+    ],
+)
+def test_to_numpy(arr, expected, zero_copy, index_or_series_or_array):
+    box = index_or_series_or_array
+
+    with tm.assert_produces_warning(None):
+        thing = box(arr)
+
+    result = thing.to_numpy()
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = np.asarray(thing)
+    tm.assert_numpy_array_equal(result, expected)
+
+    # Additionally, we check the `copy=` semantics for array/asarray
+    # (these are implemented by us via `__array__`).
+    result_cp1 = np.array(thing, copy=True)
+    result_cp2 = np.array(thing, copy=True)
+    # When called with `copy=True` NumPy/we should ensure a copy was made
+    assert not np.may_share_memory(result_cp1, result_cp2)
+
+    if not np_version_gt2:
+        # copy=False semantics are only supported in NumPy>=2.
+        return
+
+    if not zero_copy:
+        with pytest.raises(ValueError, match="Unable to avoid copy while creating"):
+            # An error is always acceptable for `copy=False`
+            np.array(thing, copy=False)
+
+    else:
+        result_nocopy1 = np.array(thing, copy=False)
+        result_nocopy2 = np.array(thing, copy=False)
+        # If copy=False was given, these must share the same data
+        assert np.may_share_memory(result_nocopy1, result_nocopy2)
+
+
+@pytest.mark.parametrize("as_series", [True, False])
+@pytest.mark.parametrize(
+    "arr", [np.array([1, 2, 3], dtype="int64"), np.array(["a", "b", "c"], dtype=object)]
+)
+def test_to_numpy_copy(arr, as_series, using_infer_string):
+    obj = pd.Index(arr, copy=False)
+    if as_series:
+        obj = Series(obj.values, copy=False)
+
+    # no copy by default
+    result = obj.to_numpy()
+    if using_infer_string and arr.dtype == object and obj.dtype.storage == "pyarrow":
+        assert np.shares_memory(arr, result) is False
+    else:
+        assert np.shares_memory(arr, result) is True
+
+    result = obj.to_numpy(copy=False)
+    if using_infer_string and arr.dtype == object and obj.dtype.storage == "pyarrow":
+        assert np.shares_memory(arr, result) is False
+    else:
+        assert np.shares_memory(arr, result) is True
+
+    # copy=True
+    result = obj.to_numpy(copy=True)
+    assert np.shares_memory(arr, result) is False
+
+
+@pytest.mark.parametrize("as_series", [True, False])
+def test_to_numpy_dtype(as_series):
+    tz = "US/Eastern"
+    obj = pd.DatetimeIndex(["2000", "2001"], tz=tz)
+    if as_series:
+        obj = Series(obj)
+
+    # preserve tz by default
+    result = obj.to_numpy()
+    expected = np.array(
+        [Timestamp("2000", tz=tz), Timestamp("2001", tz=tz)], dtype=object
+    )
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = obj.to_numpy(dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = obj.to_numpy(dtype="M8[ns]")
+    expected = np.array(["2000-01-01T05", "2001-01-01T05"], dtype="M8[ns]")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values, dtype, na_value, expected",
+    [
+        ([1, 2, None], "float64", 0, [1.0, 2.0, 0.0]),
+        (
+            [Timestamp("2000"), Timestamp("2000"), pd.NaT],
+            None,
+            Timestamp("2000"),
+            [np.datetime64("2000-01-01T00:00:00", "s")] * 3,
+        ),
+    ],
+)
+def test_to_numpy_na_value_numpy_dtype(
+    index_or_series, values, dtype, na_value, expected
+):
+    obj = index_or_series(values)
+    result = obj.to_numpy(dtype=dtype, na_value=na_value)
+    expected = np.array(expected)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, multiindex, dtype, na_value, expected",
+    [
+        (
+            [1, 2, None, 4],
+            [(0, "a"), (0, "b"), (1, "b"), (1, "c")],
+            float,
+            None,
+            [1.0, 2.0, np.nan, 4.0],
+        ),
+        (
+            [1, 2, None, 4],
+            [(0, "a"), (0, "b"), (1, "b"), (1, "c")],
+            float,
+            np.nan,
+            [1.0, 2.0, np.nan, 4.0],
+        ),
+        (
+            [1.0, 2.0, np.nan, 4.0],
+            [("a", 0), ("a", 1), ("a", 2), ("b", 0)],
+            int,
+            0,
+            [1, 2, 0, 4],
+        ),
+        (
+            [Timestamp("2000"), Timestamp("2000"), pd.NaT],
+            [(0, Timestamp("2021")), (0, Timestamp("2022")), (1, Timestamp("2000"))],
+            None,
+            Timestamp("2000"),
+            [np.datetime64("2000-01-01T00:00:00", "s")] * 3,
+        ),
+    ],
+)
+def test_to_numpy_multiindex_series_na_value(
+    data, multiindex, dtype, na_value, expected
+):
+    index = pd.MultiIndex.from_tuples(multiindex)
+    series = Series(data, index=index)
+    result = series.to_numpy(dtype=dtype, na_value=na_value)
+    expected = np.array(expected)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_kwargs_raises():
+    # numpy
+    s = Series([1, 2, 3])
+    msg = r"to_numpy\(\) got an unexpected keyword argument 'foo'"
+    with pytest.raises(TypeError, match=msg):
+        s.to_numpy(foo=True)
+
+    # extension
+    s = Series([1, 2, 3], dtype="Int64")
+    with pytest.raises(TypeError, match=msg):
+        s.to_numpy(foo=True)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        {"a": [1, 2, 3], "b": [1, 2, None]},
+        {"a": np.array([1, 2, 3]), "b": np.array([1, 2, np.nan])},
+        {"a": pd.array([1, 2, 3]), "b": pd.array([1, 2, None])},
+    ],
+)
+@pytest.mark.parametrize("dtype, na_value", [(float, np.nan), (object, None)])
+def test_to_numpy_dataframe_na_value(data, dtype, na_value):
+    # https://github.com/pandas-dev/pandas/issues/33820
+    df = pd.DataFrame(data)
+    result = df.to_numpy(dtype=dtype, na_value=na_value)
+    expected = np.array([[1, 1], [2, 2], [3, na_value]], dtype=dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected_data",
+    [
+        (
+            {"a": pd.array([1, 2, None])},
+            [[1.0], [2.0], [np.nan]],
+        ),
+        (
+            {"a": [1, 2, 3], "b": [1, 2, 3]},
+            [[1, 1], [2, 2], [3, 3]],
+        ),
+    ],
+)
+def test_to_numpy_dataframe_single_block(data, expected_data):
+    # https://github.com/pandas-dev/pandas/issues/33820
+    df = pd.DataFrame(data)
+    result = df.to_numpy(dtype=float, na_value=np.nan)
+    expected = np.array(expected_data, dtype=float)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_dataframe_single_block_no_mutate():
+    # https://github.com/pandas-dev/pandas/issues/33820
+    result = pd.DataFrame(np.array([1.0, 2.0, np.nan]))
+    expected = pd.DataFrame(np.array([1.0, 2.0, np.nan]))
+    result.to_numpy(na_value=0.0)
+    tm.assert_frame_equal(result, expected)
+
+
+class TestAsArray:
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_asarray_object_dt64(self, tz):
+        ser = Series(date_range("2000", periods=2, tz=tz))
+
+        with tm.assert_produces_warning(None):
+            # Future behavior (for tzaware case) with no warning
+            result = np.asarray(ser, dtype=object)
+
+        expected = np.array(
+            [Timestamp("2000-01-01", tz=tz), Timestamp("2000-01-02", tz=tz)]
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_asarray_tz_naive(self):
+        # This shouldn't produce a warning.
+        ser = Series(date_range("2000", periods=2))
+        expected = np.array(["2000-01-01", "2000-01-02"], dtype="M8[ns]")
+        result = np.asarray(ser)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_asarray_tz_aware(self):
+        tz = "US/Central"
+        ser = Series(date_range("2000", periods=2, tz=tz))
+        expected = np.array(["2000-01-01T06", "2000-01-02T06"], dtype="M8[ns]")
+        result = np.asarray(ser, dtype="datetime64[ns]")
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        # Old behavior with no warning
+        result = np.asarray(ser, dtype="M8[ns]")
+
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/base/test_fillna.py b/pandas/pandas/tests/base/test_fillna.py
new file mode 100644
index 0000000000000..8c56bcc169d8e
--- /dev/null
+++ b/pandas/pandas/tests/base/test_fillna.py
@@ -0,0 +1,60 @@
+"""
+Though Index.fillna and Series.fillna has separate impl,
+test here to confirm these works as the same
+"""
+
+import numpy as np
+import pytest
+
+from pandas import MultiIndex
+import pandas._testing as tm
+from pandas.tests.base.common import allow_na_ops
+
+
+def test_fillna(index_or_series_obj):
+    # GH 11343
+    obj = index_or_series_obj
+
+    if isinstance(obj, MultiIndex):
+        msg = "fillna is not defined for MultiIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            obj.fillna(0)
+        return
+
+    # values will not be changed
+    fill_value = obj.values[0] if len(obj) > 0 else 0
+    result = obj.fillna(fill_value)
+
+    tm.assert_equal(obj, result)
+
+    # check shallow_copied
+    assert obj is not result
+
+
+@pytest.mark.parametrize("null_obj", [np.nan, None])
+def test_fillna_null(null_obj, index_or_series_obj):
+    # GH 11343
+    obj = index_or_series_obj
+    klass = type(obj)
+
+    if not allow_na_ops(obj):
+        pytest.skip(f"{klass} doesn't allow for NA operations")
+    elif len(obj) < 1:
+        pytest.skip("Test doesn't make sense on empty data")
+    elif isinstance(obj, MultiIndex):
+        pytest.skip(f"MultiIndex can't hold '{null_obj}'")
+
+    values = obj._values
+    fill_value = values[0]
+    expected = values.copy()
+    values[0:2] = null_obj
+    expected[0:2] = fill_value
+
+    expected = klass(expected)
+    obj = klass(values)
+
+    result = obj.fillna(fill_value)
+    tm.assert_equal(result, expected)
+
+    # check shallow_copied
+    assert obj is not result
diff --git a/pandas/pandas/tests/base/test_misc.py b/pandas/pandas/tests/base/test_misc.py
new file mode 100644
index 0000000000000..7819b7b75f065
--- /dev/null
+++ b/pandas/pandas/tests/base/test_misc.py
@@ -0,0 +1,192 @@
+import sys
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import PYPY
+
+from pandas.core.dtypes.common import (
+    is_dtype_equal,
+    is_object_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    Index,
+    Series,
+)
+
+
+def test_isnull_notnull_docstrings():
+    # GH#41855 make sure its clear these are aliases
+    doc = pd.DataFrame.notnull.__doc__
+    assert doc.startswith("\nDataFrame.notnull is an alias for DataFrame.notna.\n")
+    doc = pd.DataFrame.isnull.__doc__
+    assert doc.startswith("\nDataFrame.isnull is an alias for DataFrame.isna.\n")
+
+    doc = Series.notnull.__doc__
+    assert doc.startswith("\nSeries.notnull is an alias for Series.notna.\n")
+    doc = Series.isnull.__doc__
+    assert doc.startswith("\nSeries.isnull is an alias for Series.isna.\n")
+
+
+@pytest.mark.parametrize(
+    "op_name, op",
+    [
+        ("add", "+"),
+        ("sub", "-"),
+        ("mul", "*"),
+        ("mod", "%"),
+        ("pow", "**"),
+        ("truediv", "/"),
+        ("floordiv", "//"),
+    ],
+)
+def test_binary_ops_docstring(frame_or_series, op_name, op):
+    # not using the all_arithmetic_functions fixture with _get_opstr
+    # as _get_opstr is used internally in the dynamic implementation of the docstring
+    klass = frame_or_series
+
+    operand1 = klass.__name__.lower()
+    operand2 = "other"
+    expected_str = " ".join([operand1, op, operand2])
+    assert expected_str in getattr(klass, op_name).__doc__
+
+    # reverse version of the binary ops
+    expected_str = " ".join([operand2, op, operand1])
+    assert expected_str in getattr(klass, "r" + op_name).__doc__
+
+
+def test_ndarray_compat_properties(index_or_series_obj):
+    obj = index_or_series_obj
+
+    # Check that we work.
+    for p in ["shape", "dtype", "T", "nbytes"]:
+        assert getattr(obj, p, None) is not None
+
+    # deprecated properties
+    for p in ["strides", "itemsize", "base", "data"]:
+        assert not hasattr(obj, p)
+
+    msg = "can only convert an array of size 1 to a Python scalar"
+    with pytest.raises(ValueError, match=msg):
+        obj.item()  # len > 1
+
+    assert obj.ndim == 1
+    assert obj.size == len(obj)
+
+    assert Index([1]).item() == 1
+    assert Series([1]).item() == 1
+
+
+@pytest.mark.skipif(
+    PYPY or using_string_dtype(),
+    reason="not relevant for PyPy doesn't work properly for arrow strings",
+)
+def test_memory_usage(index_or_series_memory_obj):
+    obj = index_or_series_memory_obj
+    # Clear index caches so that len(obj) == 0 report 0 memory usage
+    if isinstance(obj, Series):
+        is_ser = True
+        obj.index._engine.clear_mapping()
+    else:
+        is_ser = False
+        obj._engine.clear_mapping()
+
+    res = obj.memory_usage()
+    res_deep = obj.memory_usage(deep=True)
+
+    is_object = is_object_dtype(obj) or (is_ser and is_object_dtype(obj.index))
+    is_categorical = isinstance(obj.dtype, pd.CategoricalDtype) or (
+        is_ser and isinstance(obj.index.dtype, pd.CategoricalDtype)
+    )
+    is_object_string = is_dtype_equal(obj, "string[python]") or (
+        is_ser and is_dtype_equal(obj.index.dtype, "string[python]")
+    )
+
+    if len(obj) == 0:
+        expected = 0
+        assert res_deep == res == expected
+    elif is_object or is_categorical or is_object_string:
+        # only deep will pick them up
+        assert res_deep > res
+    else:
+        assert res == res_deep
+
+    # sys.getsizeof will call the .memory_usage with
+    # deep=True, and add on some GC overhead
+    diff = res_deep - sys.getsizeof(obj)
+    assert abs(diff) < 100
+
+
+def test_memory_usage_components_series(series_with_simple_index):
+    series = series_with_simple_index
+    total_usage = series.memory_usage(index=True)
+    non_index_usage = series.memory_usage(index=False)
+    index_usage = series.index.memory_usage()
+    assert total_usage == non_index_usage + index_usage
+
+
+def test_memory_usage_components_narrow_series(any_real_numpy_dtype):
+    series = Series(
+        range(5),
+        dtype=any_real_numpy_dtype,
+        index=[f"i-{i}" for i in range(5)],
+        name="a",
+    )
+    total_usage = series.memory_usage(index=True)
+    non_index_usage = series.memory_usage(index=False)
+    index_usage = series.index.memory_usage()
+    assert total_usage == non_index_usage + index_usage
+
+
+def test_searchsorted(request, index_or_series_obj):
+    # numpy.searchsorted calls obj.searchsorted under the hood.
+    # See gh-12238
+    obj = index_or_series_obj
+
+    if isinstance(obj, pd.MultiIndex):
+        # See gh-14833
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="np.searchsorted doesn't work on pd.MultiIndex: GH 14833"
+            )
+        )
+    elif obj.dtype.kind == "c" and isinstance(obj, Index):
+        # TODO: Should Series cases also raise? Looks like they use numpy
+        #  comparison semantics https://github.com/numpy/numpy/issues/15981
+        mark = pytest.mark.xfail(reason="complex objects are not comparable")
+        request.applymarker(mark)
+
+    max_obj = max(obj, default=0)
+    index = np.searchsorted(obj, max_obj)
+    assert 0 <= index <= len(obj)
+
+    index = np.searchsorted(obj, max_obj, sorter=range(len(obj)))
+    assert 0 <= index <= len(obj)
+
+
+def test_access_by_position(index_flat):
+    index = index_flat
+
+    if len(index) == 0:
+        pytest.skip("Test doesn't make sense on empty data")
+
+    series = Series(index)
+    assert index[0] == series.iloc[0]
+    assert index[5] == series.iloc[5]
+    assert index[-1] == series.iloc[-1]
+
+    size = len(index)
+    assert index[-1] == index[size - 1]
+
+    msg = f"index {size} is out of bounds for axis 0 with size {size}"
+    if isinstance(index.dtype, pd.StringDtype) and index.dtype.storage == "pyarrow":
+        msg = "index out of bounds"
+    with pytest.raises(IndexError, match=msg):
+        index[size]
+    msg = "single positional indexer is out-of-bounds"
+    with pytest.raises(IndexError, match=msg):
+        series.iloc[size]
diff --git a/pandas/pandas/tests/base/test_transpose.py b/pandas/pandas/tests/base/test_transpose.py
new file mode 100644
index 0000000000000..246f33d27476c
--- /dev/null
+++ b/pandas/pandas/tests/base/test_transpose.py
@@ -0,0 +1,56 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalDtype,
+    DataFrame,
+)
+import pandas._testing as tm
+
+
+def test_transpose(index_or_series_obj):
+    obj = index_or_series_obj
+    tm.assert_equal(obj.transpose(), obj)
+
+
+def test_transpose_non_default_axes(index_or_series_obj):
+    msg = "the 'axes' parameter is not supported"
+    obj = index_or_series_obj
+    with pytest.raises(ValueError, match=msg):
+        obj.transpose(1)
+    with pytest.raises(ValueError, match=msg):
+        obj.transpose(axes=1)
+
+
+def test_numpy_transpose(index_or_series_obj):
+    msg = "the 'axes' parameter is not supported"
+    obj = index_or_series_obj
+    tm.assert_equal(np.transpose(obj), obj)
+
+    with pytest.raises(ValueError, match=msg):
+        np.transpose(obj, axes=1)
+
+
+@pytest.mark.parametrize(
+    "data, transposed_data, index, columns, dtype",
+    [
+        ([[1], [2]], [[1, 2]], ["a", "a"], ["b"], int),
+        ([[1], [2]], [[1, 2]], ["a", "a"], ["b"], CategoricalDtype([1, 2])),
+        ([[1, 2]], [[1], [2]], ["b"], ["a", "a"], int),
+        ([[1, 2]], [[1], [2]], ["b"], ["a", "a"], CategoricalDtype([1, 2])),
+        ([[1, 2], [3, 4]], [[1, 3], [2, 4]], ["a", "a"], ["b", "b"], int),
+        (
+            [[1, 2], [3, 4]],
+            [[1, 3], [2, 4]],
+            ["a", "a"],
+            ["b", "b"],
+            CategoricalDtype([1, 2, 3, 4]),
+        ),
+    ],
+)
+def test_duplicate_labels(data, transposed_data, index, columns, dtype):
+    # GH 42380
+    df = DataFrame(data, index=index, columns=columns, dtype=dtype)
+    result = df.T
+    expected = DataFrame(transposed_data, index=columns, columns=index, dtype=dtype)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/base/test_unique.py b/pandas/pandas/tests/base/test_unique.py
new file mode 100644
index 0000000000000..7f094db6ea524
--- /dev/null
+++ b/pandas/pandas/tests/base/test_unique.py
@@ -0,0 +1,120 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.base.common import allow_na_ops
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_unique(index_or_series_obj):
+    obj = index_or_series_obj
+    obj = np.repeat(obj, range(1, len(obj) + 1))
+    result = obj.unique()
+
+    # dict.fromkeys preserves the order
+    unique_values = list(dict.fromkeys(obj.values))
+    if isinstance(obj, pd.MultiIndex):
+        expected = pd.MultiIndex.from_tuples(unique_values)
+        expected.names = obj.names
+        tm.assert_index_equal(result, expected, exact=True)
+    elif isinstance(obj, pd.Index):
+        expected = pd.Index(unique_values, dtype=obj.dtype)
+        if isinstance(obj.dtype, pd.DatetimeTZDtype):
+            expected = expected.normalize()
+        tm.assert_index_equal(result, expected, exact=True)
+    else:
+        expected = np.array(unique_values)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.parametrize("null_obj", [np.nan, None])
+def test_unique_null(null_obj, index_or_series_obj):
+    obj = index_or_series_obj
+
+    if not allow_na_ops(obj):
+        pytest.skip("type doesn't allow for NA operations")
+    elif len(obj) < 1:
+        pytest.skip("Test doesn't make sense on empty data")
+    elif isinstance(obj, pd.MultiIndex):
+        pytest.skip(f"MultiIndex can't hold '{null_obj}'")
+
+    values = obj._values
+    values[0:2] = null_obj
+
+    klass = type(obj)
+    repeated_values = np.repeat(values, range(1, len(values) + 1))
+    obj = klass(repeated_values, dtype=obj.dtype)
+    result = obj.unique()
+
+    unique_values_raw = dict.fromkeys(obj.values)
+    # because np.nan == np.nan is False, but None == None is True
+    # np.nan would be duplicated, whereas None wouldn't
+    unique_values_not_null = [val for val in unique_values_raw if not pd.isnull(val)]
+    unique_values = [null_obj] + unique_values_not_null
+
+    if isinstance(obj, pd.Index):
+        expected = pd.Index(unique_values, dtype=obj.dtype)
+        if isinstance(obj.dtype, pd.DatetimeTZDtype):
+            result = result.normalize()
+            expected = expected.normalize()
+        tm.assert_index_equal(result, expected, exact=True)
+    else:
+        expected = np.array(unique_values, dtype=obj.dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+def test_nunique(index_or_series_obj):
+    obj = index_or_series_obj
+    obj = np.repeat(obj, range(1, len(obj) + 1))
+    expected = len(obj.unique())
+    assert obj.nunique(dropna=False) == expected
+
+
+@pytest.mark.parametrize("null_obj", [np.nan, None])
+def test_nunique_null(null_obj, index_or_series_obj):
+    obj = index_or_series_obj
+
+    if not allow_na_ops(obj):
+        pytest.skip("type doesn't allow for NA operations")
+    elif isinstance(obj, pd.MultiIndex):
+        pytest.skip(f"MultiIndex can't hold '{null_obj}'")
+
+    values = obj._values
+    values[0:2] = null_obj
+
+    klass = type(obj)
+    repeated_values = np.repeat(values, range(1, len(values) + 1))
+    obj = klass(repeated_values, dtype=obj.dtype)
+
+    if isinstance(obj, pd.CategoricalIndex):
+        assert obj.nunique() == len(obj.categories)
+        assert obj.nunique(dropna=False) == len(obj.categories) + 1
+    else:
+        num_unique_values = len(obj.unique())
+        assert obj.nunique() == max(0, num_unique_values - 1)
+        assert obj.nunique(dropna=False) == max(0, num_unique_values)
+
+
+@pytest.mark.single_cpu
+def test_unique_bad_unicode(index_or_series):
+    # regression test for #34550
+    uval = "\ud83d"  # smiley emoji
+
+    obj = index_or_series([uval] * 2, dtype=object)
+    result = obj.unique()
+
+    if isinstance(obj, pd.Index):
+        expected = pd.Index(["\ud83d"], dtype=object)
+        tm.assert_index_equal(result, expected, exact=True)
+    else:
+        expected = np.array(["\ud83d"], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+def test_nunique_dropna(dropna):
+    # GH37566
+    ser = pd.Series(["yes", "yes", pd.NA, np.nan, None, pd.NaT])
+    res = ser.nunique(dropna)
+    assert res == 1 if dropna else 5
diff --git a/pandas/pandas/tests/base/test_value_counts.py b/pandas/pandas/tests/base/test_value_counts.py
new file mode 100644
index 0000000000000..bcb31829a201f
--- /dev/null
+++ b/pandas/pandas/tests/base/test_value_counts.py
@@ -0,0 +1,341 @@
+import collections
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    array,
+)
+import pandas._testing as tm
+from pandas.tests.base.common import allow_na_ops
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_value_counts(index_or_series_obj):
+    obj = index_or_series_obj
+    obj = np.repeat(obj, range(1, len(obj) + 1))
+    result = obj.value_counts()
+
+    counter = collections.Counter(obj)
+    expected = Series(dict(counter.most_common()), dtype=np.int64, name="count")
+
+    if obj.dtype != np.float16:
+        expected.index = expected.index.astype(obj.dtype)
+    else:
+        with pytest.raises(NotImplementedError, match="float16 indexes are not "):
+            expected.index.astype(obj.dtype)
+        return
+    if isinstance(expected.index, MultiIndex):
+        expected.index.names = obj.names
+    else:
+        expected.index.name = obj.name
+
+    if not isinstance(result.dtype, np.dtype):
+        if getattr(obj.dtype, "storage", "") == "pyarrow":
+            expected = expected.astype("int64[pyarrow]")
+        else:
+            # i.e IntegerDtype
+            expected = expected.astype("Int64")
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("null_obj", [np.nan, None])
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_value_counts_null(null_obj, index_or_series_obj):
+    orig = index_or_series_obj
+    obj = orig.copy()
+
+    if not allow_na_ops(obj):
+        pytest.skip("type doesn't allow for NA operations")
+    elif len(obj) < 1:
+        pytest.skip("Test doesn't make sense on empty data")
+    elif isinstance(orig, MultiIndex):
+        pytest.skip(f"MultiIndex can't hold '{null_obj}'")
+
+    values = obj._values
+    values[0:2] = null_obj
+
+    klass = type(obj)
+    repeated_values = np.repeat(values, range(1, len(values) + 1))
+    obj = klass(repeated_values, dtype=obj.dtype)
+
+    # because np.nan == np.nan is False, but None == None is True
+    # np.nan would be duplicated, whereas None wouldn't
+    counter = collections.Counter(obj.dropna())
+    expected = Series(dict(counter.most_common()), dtype=np.int64, name="count")
+
+    if obj.dtype != np.float16:
+        expected.index = expected.index.astype(obj.dtype)
+    else:
+        with pytest.raises(NotImplementedError, match="float16 indexes are not "):
+            expected.index.astype(obj.dtype)
+        return
+    expected.index.name = obj.name
+
+    result = obj.value_counts()
+
+    if not isinstance(result.dtype, np.dtype):
+        if getattr(obj.dtype, "storage", "") == "pyarrow":
+            expected = expected.astype("int64[pyarrow]")
+        else:
+            # i.e IntegerDtype
+            expected = expected.astype("Int64")
+    tm.assert_series_equal(result, expected)
+
+    expected[null_obj] = 3
+
+    result = obj.value_counts(dropna=False)
+    expected = expected.sort_index()
+    result = result.sort_index()
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_inferred(index_or_series, using_infer_string):
+    klass = index_or_series
+    s_values = ["a", "b", "b", "b", "b", "c", "d", "d", "a", "a"]
+    s = klass(s_values)
+    expected = Series([4, 3, 2, 1], index=["b", "a", "d", "c"], name="count")
+    tm.assert_series_equal(s.value_counts(), expected)
+
+    if isinstance(s, Index):
+        exp = Index(np.unique(np.array(s_values, dtype=np.object_)))
+        tm.assert_index_equal(s.unique(), exp)
+    else:
+        exp = np.unique(np.array(s_values, dtype=np.object_))
+        if using_infer_string:
+            exp = array(exp, dtype="str")
+        tm.assert_equal(s.unique(), exp)
+
+    assert s.nunique() == 4
+    # don't sort, have to sort after the fact as not sorting is
+    # platform-dep
+    hist = s.value_counts(sort=False).sort_values()
+    expected = Series([3, 1, 4, 2], index=list("acbd"), name="count").sort_values()
+    tm.assert_series_equal(hist, expected)
+
+    # sort ascending
+    hist = s.value_counts(ascending=True)
+    expected = Series([1, 2, 3, 4], index=list("cdab"), name="count")
+    tm.assert_series_equal(hist, expected)
+
+    # relative histogram.
+    hist = s.value_counts(normalize=True)
+    expected = Series(
+        [0.4, 0.3, 0.2, 0.1], index=["b", "a", "d", "c"], name="proportion"
+    )
+    tm.assert_series_equal(hist, expected)
+
+
+def test_value_counts_bins(index_or_series, using_infer_string):
+    klass = index_or_series
+    s_values = ["a", "b", "b", "b", "b", "c", "d", "d", "a", "a"]
+    s = klass(s_values)
+
+    # bins
+    msg = "bins argument only works with numeric data"
+    with pytest.raises(TypeError, match=msg):
+        s.value_counts(bins=1)
+
+    s1 = Series([1, 1, 2, 3])
+    res1 = s1.value_counts(bins=1)
+    exp1 = Series({Interval(0.997, 3.0): 4}, name="count")
+    tm.assert_series_equal(res1, exp1)
+    res1n = s1.value_counts(bins=1, normalize=True)
+    exp1n = Series({Interval(0.997, 3.0): 1.0}, name="proportion")
+    tm.assert_series_equal(res1n, exp1n)
+
+    if isinstance(s1, Index):
+        tm.assert_index_equal(s1.unique(), Index([1, 2, 3]))
+    else:
+        exp = np.array([1, 2, 3], dtype=np.int64)
+        tm.assert_numpy_array_equal(s1.unique(), exp)
+
+    assert s1.nunique() == 3
+
+    # these return the same
+    res4 = s1.value_counts(bins=4, dropna=True)
+    intervals = IntervalIndex.from_breaks([0.997, 1.5, 2.0, 2.5, 3.0])
+    exp4 = Series([2, 1, 1, 0], index=intervals.take([0, 1, 3, 2]), name="count")
+    tm.assert_series_equal(res4, exp4)
+
+    res4 = s1.value_counts(bins=4, dropna=False)
+    intervals = IntervalIndex.from_breaks([0.997, 1.5, 2.0, 2.5, 3.0])
+    exp4 = Series([2, 1, 1, 0], index=intervals.take([0, 1, 3, 2]), name="count")
+    tm.assert_series_equal(res4, exp4)
+
+    res4n = s1.value_counts(bins=4, normalize=True)
+    exp4n = Series(
+        [0.5, 0.25, 0.25, 0], index=intervals.take([0, 1, 3, 2]), name="proportion"
+    )
+    tm.assert_series_equal(res4n, exp4n)
+
+    # handle NA's properly
+    s_values = ["a", "b", "b", "b", np.nan, np.nan, "d", "d", "a", "a", "b"]
+    s = klass(s_values)
+    expected = Series([4, 3, 2], index=["b", "a", "d"], name="count")
+    tm.assert_series_equal(s.value_counts(), expected)
+
+    if isinstance(s, Index):
+        exp = Index(["a", "b", np.nan, "d"])
+        tm.assert_index_equal(s.unique(), exp)
+    else:
+        exp = np.array(["a", "b", np.nan, "d"], dtype=object)
+        if using_infer_string:
+            exp = array(exp, dtype="str")
+        tm.assert_equal(s.unique(), exp)
+    assert s.nunique() == 3
+
+    s = klass({}) if klass is dict else klass({}, dtype=object)
+    expected = Series([], dtype=np.int64, name="count")
+    tm.assert_series_equal(s.value_counts(), expected, check_index_type=False)
+    # returned dtype differs depending on original
+    if isinstance(s, Index):
+        tm.assert_index_equal(s.unique(), Index([]), exact=False)
+    else:
+        tm.assert_numpy_array_equal(s.unique(), np.array([]), check_dtype=False)
+
+    assert s.nunique() == 0
+
+
+def test_value_counts_datetime64(index_or_series, unit):
+    klass = index_or_series
+
+    # GH 3002, datetime64[ns]
+    # don't test names though
+    df = pd.DataFrame(
+        {
+            "person_id": ["xxyyzz", "xxyyzz", "xxyyzz", "xxyyww", "foofoo", "foofoo"],
+            "dt": pd.to_datetime(
+                [
+                    "2010-01-01",
+                    "2010-01-01",
+                    "2010-01-01",
+                    "2009-01-01",
+                    "2008-09-09",
+                    "2008-09-09",
+                ]
+            ).as_unit(unit),
+            "food": ["PIE", "GUM", "EGG", "EGG", "PIE", "GUM"],
+        }
+    )
+
+    s = klass(df["dt"].copy())
+    s.name = None
+    idx = pd.to_datetime(
+        ["2010-01-01 00:00:00", "2008-09-09 00:00:00", "2009-01-01 00:00:00"]
+    ).as_unit(unit)
+    expected_s = Series([3, 2, 1], index=idx, name="count")
+    tm.assert_series_equal(s.value_counts(), expected_s)
+
+    expected = array(
+        np.array(
+            ["2010-01-01 00:00:00", "2009-01-01 00:00:00", "2008-09-09 00:00:00"],
+            dtype=f"datetime64[{unit}]",
+        )
+    )
+    result = s.unique()
+    if isinstance(s, Index):
+        tm.assert_index_equal(result, DatetimeIndex(expected))
+    else:
+        tm.assert_extension_array_equal(result, expected)
+
+    assert s.nunique() == 3
+
+    # with NaT
+    s = df["dt"].copy()
+    s = klass(list(s.values) + [pd.NaT] * 4)
+    if klass is Series:
+        s = s.dt.as_unit(unit)
+    else:
+        s = s.as_unit(unit)
+
+    result = s.value_counts()
+    assert result.index.dtype == f"datetime64[{unit}]"
+    tm.assert_series_equal(result, expected_s)
+
+    result = s.value_counts(dropna=False)
+    expected_s = pd.concat(
+        [
+            Series([4], index=DatetimeIndex([pd.NaT]).as_unit(unit), name="count"),
+            expected_s,
+        ]
+    )
+    tm.assert_series_equal(result, expected_s)
+
+    assert s.dtype == f"datetime64[{unit}]"
+    unique = s.unique()
+    assert unique.dtype == f"datetime64[{unit}]"
+
+    # numpy_array_equal cannot compare pd.NaT
+    if isinstance(s, Index):
+        exp_idx = DatetimeIndex(expected.tolist() + [pd.NaT]).as_unit(unit)
+        tm.assert_index_equal(unique, exp_idx)
+    else:
+        tm.assert_extension_array_equal(unique[:3], expected)
+        assert pd.isna(unique[3])
+
+    assert s.nunique() == 3
+    assert s.nunique(dropna=False) == 4
+
+
+def test_value_counts_timedelta64(index_or_series, unit):
+    # timedelta64[ns]
+    klass = index_or_series
+
+    day = Timedelta(timedelta(1)).as_unit(unit)
+    tdi = TimedeltaIndex([day], name="dt").as_unit(unit)
+
+    tdvals = np.zeros(6, dtype=f"m8[{unit}]") + day
+    td = klass(tdvals, name="dt")
+
+    result = td.value_counts()
+    expected_s = Series([6], index=tdi, name="count")
+    tm.assert_series_equal(result, expected_s)
+
+    expected = tdi
+    result = td.unique()
+    if isinstance(td, Index):
+        tm.assert_index_equal(result, expected)
+    else:
+        tm.assert_extension_array_equal(result, expected._values)
+
+    td2 = day + np.zeros(6, dtype=f"m8[{unit}]")
+    td2 = klass(td2, name="dt")
+    result2 = td2.value_counts()
+    tm.assert_series_equal(result2, expected_s)
+
+
+def test_value_counts_with_nan(dropna, index_or_series):
+    # GH31944
+    klass = index_or_series
+    values = [True, pd.NA, np.nan]
+    obj = klass(values)
+    res = obj.value_counts(dropna=dropna)
+    if dropna is True:
+        expected = Series([1], index=Index([True], dtype=obj.dtype), name="count")
+    else:
+        expected = Series([1, 1, 1], index=[True, pd.NA, np.nan], name="count")
+    tm.assert_series_equal(res, expected)
+
+
+def test_value_counts_object_inference_deprecated():
+    # GH#56161
+    dti = pd.date_range("2016-01-01", periods=3, tz="UTC")
+
+    idx = dti.astype(object)
+    res = idx.value_counts()
+
+    exp = dti.value_counts()
+    exp.index = exp.index.astype(object)
+    tm.assert_series_equal(res, exp)
diff --git a/pandas/pandas/tests/computation/__init__.py b/pandas/pandas/tests/computation/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/computation/test_compat.py b/pandas/pandas/tests/computation/test_compat.py
new file mode 100644
index 0000000000000..856a5b3a22a95
--- /dev/null
+++ b/pandas/pandas/tests/computation/test_compat.py
@@ -0,0 +1,32 @@
+import pytest
+
+from pandas.compat._optional import VERSIONS
+
+import pandas as pd
+from pandas.core.computation import expr
+from pandas.core.computation.engines import ENGINES
+from pandas.util.version import Version
+
+
+def test_compat():
+    # test we have compat with our version of numexpr
+
+    from pandas.core.computation.check import NUMEXPR_INSTALLED
+
+    ne = pytest.importorskip("numexpr")
+
+    ver = ne.__version__
+    if Version(ver) < Version(VERSIONS["numexpr"]):
+        assert not NUMEXPR_INSTALLED
+    else:
+        assert NUMEXPR_INSTALLED
+
+
+@pytest.mark.parametrize("engine", ENGINES)
+@pytest.mark.parametrize("parser", expr.PARSERS)
+def test_invalid_numexpr_version(engine, parser):
+    if engine == "numexpr":
+        pytest.importorskip("numexpr")
+    a, b = 1, 2  # noqa: F841
+    res = pd.eval("a + b", engine=engine, parser=parser)
+    assert res == 3
diff --git a/pandas/pandas/tests/computation/test_eval.py b/pandas/pandas/tests/computation/test_eval.py
new file mode 100644
index 0000000000000..3c0bf6c35866c
--- /dev/null
+++ b/pandas/pandas/tests/computation/test_eval.py
@@ -0,0 +1,2008 @@
+from __future__ import annotations
+
+from functools import reduce
+from itertools import product
+import operator
+
+import numpy as np
+import pytest
+
+from pandas.compat import PY312
+from pandas.errors import (
+    NumExprClobberingError,
+    PerformanceWarning,
+    UndefinedVariableError,
+)
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_float,
+    is_list_like,
+    is_scalar,
+)
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.computation import (
+    expr,
+    pytables,
+)
+from pandas.core.computation.engines import ENGINES
+from pandas.core.computation.expr import (
+    BaseExprVisitor,
+    PandasExprVisitor,
+    PythonExprVisitor,
+)
+from pandas.core.computation.expressions import (
+    NUMEXPR_INSTALLED,
+    USE_NUMEXPR,
+)
+from pandas.core.computation.ops import (
+    ARITH_OPS_SYMS,
+    _binary_math_ops,
+    _binary_ops_dict,
+    _unary_math_ops,
+)
+from pandas.core.computation.scope import DEFAULT_GLOBALS
+
+
+@pytest.fixture(
+    params=(
+        pytest.param(
+            engine,
+            marks=[
+                pytest.mark.skipif(
+                    engine == "numexpr" and not USE_NUMEXPR,
+                    reason=f"numexpr enabled->{USE_NUMEXPR}, "
+                    f"installed->{NUMEXPR_INSTALLED}",
+                ),
+                td.skip_if_no("numexpr"),
+            ],
+        )
+        for engine in ENGINES
+    )
+)
+def engine(request):
+    return request.param
+
+
+@pytest.fixture(params=expr.PARSERS)
+def parser(request):
+    return request.param
+
+
+def _eval_single_bin(lhs, cmp1, rhs, engine):
+    c = _binary_ops_dict[cmp1]
+    if ENGINES[engine].has_neg_frac:
+        try:
+            return c(lhs, rhs)
+        except ValueError as e:
+            if str(e).startswith(
+                "negative number cannot be raised to a fractional power"
+            ):
+                return np.nan
+            raise
+    return c(lhs, rhs)
+
+
+# TODO: using range(5) here is a kludge
+@pytest.fixture(
+    params=list(range(5)),
+    ids=["DataFrame", "Series", "SeriesNaN", "DataFrameNaN", "float"],
+)
+def lhs(request):
+    nan_df1 = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+    nan_df1[nan_df1 > 0.5] = np.nan
+
+    opts = (
+        DataFrame(np.random.default_rng(2).standard_normal((10, 5))),
+        Series(np.random.default_rng(2).standard_normal(5)),
+        Series([1, 2, np.nan, np.nan, 5]),
+        nan_df1,
+        np.random.default_rng(2).standard_normal(),
+    )
+    return opts[request.param]
+
+
+rhs = lhs
+midhs = lhs
+
+
+@pytest.fixture
+def idx_func_dict():
+    return {
+        "i": lambda n: Index(np.arange(n), dtype=np.int64),
+        "f": lambda n: Index(np.arange(n), dtype=np.float64),
+        "s": lambda n: Index([f"{i}_{chr(i)}" for i in range(97, 97 + n)]),
+        "dt": lambda n: date_range("2020-01-01", periods=n),
+        "td": lambda n: timedelta_range("1 day", periods=n),
+        "p": lambda n: period_range("2020-01-01", periods=n, freq="D"),
+    }
+
+
+class TestEval:
+    @pytest.mark.parametrize(
+        "cmp1",
+        ["!=", "==", "<=", ">=", "<", ">"],
+        ids=["ne", "eq", "le", "ge", "lt", "gt"],
+    )
+    @pytest.mark.parametrize("cmp2", [">", "<"], ids=["gt", "lt"])
+    @pytest.mark.parametrize("binop", expr.BOOL_OPS_SYMS)
+    def test_complex_cmp_ops(self, cmp1, cmp2, binop, lhs, rhs, engine, parser):
+        if parser == "python" and binop in ["and", "or"]:
+            msg = "'BoolOp' nodes are not implemented"
+            ex = f"(lhs {cmp1} rhs) {binop} (lhs {cmp2} rhs)"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(ex, engine=engine, parser=parser)
+            return
+
+        lhs_new = _eval_single_bin(lhs, cmp1, rhs, engine)
+        rhs_new = _eval_single_bin(lhs, cmp2, rhs, engine)
+        expected = _eval_single_bin(lhs_new, binop, rhs_new, engine)
+
+        ex = f"(lhs {cmp1} rhs) {binop} (lhs {cmp2} rhs)"
+        result = pd.eval(ex, engine=engine, parser=parser)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("cmp_op", expr.CMP_OPS_SYMS)
+    def test_simple_cmp_ops(self, cmp_op, lhs, rhs, engine, parser):
+        lhs = lhs < 0
+        rhs = rhs < 0
+
+        if parser == "python" and cmp_op in ["in", "not in"]:
+            msg = "'(In|NotIn)' nodes are not implemented"
+            ex = f"lhs {cmp_op} rhs"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(ex, engine=engine, parser=parser)
+            return
+
+        ex = f"lhs {cmp_op} rhs"
+        msg = "|".join(
+            [
+                r"only list-like( or dict-like)? objects are allowed to be "
+                r"passed to (DataFrame\.)?isin\(\), you passed a "
+                r"(`|')bool(`|')",
+                "argument of type 'bool' is not iterable",
+            ]
+        )
+        if cmp_op in ("in", "not in") and not is_list_like(rhs):
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(
+                    ex,
+                    engine=engine,
+                    parser=parser,
+                    local_dict={"lhs": lhs, "rhs": rhs},
+                )
+        else:
+            expected = _eval_single_bin(lhs, cmp_op, rhs, engine)
+            result = pd.eval(ex, engine=engine, parser=parser)
+            tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("op", expr.CMP_OPS_SYMS)
+    def test_compound_invert_op(self, op, lhs, rhs, request, engine, parser):
+        if parser == "python" and op in ["in", "not in"]:
+            msg = "'(In|NotIn)' nodes are not implemented"
+            ex = f"~(lhs {op} rhs)"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(ex, engine=engine, parser=parser)
+            return
+
+        if (
+            is_float(lhs)
+            and not is_float(rhs)
+            and op in ["in", "not in"]
+            and engine == "python"
+            and parser == "pandas"
+        ):
+            mark = pytest.mark.xfail(
+                reason="Looks like expected is negative, unclear whether "
+                "expected is incorrect or result is incorrect"
+            )
+            request.applymarker(mark)
+        skip_these = ["in", "not in"]
+        ex = f"~(lhs {op} rhs)"
+
+        msg = "|".join(
+            [
+                r"only list-like( or dict-like)? objects are allowed to be "
+                r"passed to (DataFrame\.)?isin\(\), you passed a "
+                r"(`|')float(`|')",
+                "argument of type 'float' is not iterable",
+            ]
+        )
+        if is_scalar(rhs) and op in skip_these:
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(
+                    ex,
+                    engine=engine,
+                    parser=parser,
+                    local_dict={"lhs": lhs, "rhs": rhs},
+                )
+        else:
+            # compound
+            if is_scalar(lhs) and is_scalar(rhs):
+                lhs, rhs = (np.array([x]) for x in (lhs, rhs))
+            expected = _eval_single_bin(lhs, op, rhs, engine)
+            if is_scalar(expected):
+                expected = not expected
+            else:
+                expected = ~expected
+            result = pd.eval(ex, engine=engine, parser=parser)
+            tm.assert_almost_equal(expected, result)
+
+    @pytest.mark.parametrize("cmp1", ["<", ">"])
+    @pytest.mark.parametrize("cmp2", ["<", ">"])
+    def test_chained_cmp_op(self, cmp1, cmp2, lhs, midhs, rhs, engine, parser):
+        mid = midhs
+        if parser == "python":
+            ex1 = f"lhs {cmp1} mid {cmp2} rhs"
+            msg = "'BoolOp' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(ex1, engine=engine, parser=parser)
+            return
+
+        lhs_new = _eval_single_bin(lhs, cmp1, mid, engine)
+        rhs_new = _eval_single_bin(mid, cmp2, rhs, engine)
+
+        if lhs_new is not None and rhs_new is not None:
+            ex1 = f"lhs {cmp1} mid {cmp2} rhs"
+            ex2 = f"lhs {cmp1} mid and mid {cmp2} rhs"
+            ex3 = f"(lhs {cmp1} mid) & (mid {cmp2} rhs)"
+            expected = _eval_single_bin(lhs_new, "&", rhs_new, engine)
+
+            for ex in (ex1, ex2, ex3):
+                result = pd.eval(ex, engine=engine, parser=parser)
+
+                tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arith1", sorted(set(ARITH_OPS_SYMS).difference({"**", "//", "%"}))
+    )
+    def test_binary_arith_ops(self, arith1, lhs, rhs, engine, parser):
+        ex = f"lhs {arith1} rhs"
+        result = pd.eval(ex, engine=engine, parser=parser)
+        expected = _eval_single_bin(lhs, arith1, rhs, engine)
+
+        tm.assert_almost_equal(result, expected)
+        ex = f"lhs {arith1} rhs {arith1} rhs"
+        result = pd.eval(ex, engine=engine, parser=parser)
+        nlhs = _eval_single_bin(lhs, arith1, rhs, engine)
+        try:
+            nlhs, ghs = nlhs.align(rhs)
+        except (ValueError, TypeError, AttributeError):
+            # ValueError: series frame or frame series align
+            # TypeError, AttributeError: series or frame with scalar align
+            return
+        else:
+            if engine == "numexpr":
+                import numexpr as ne
+
+                # direct numpy comparison
+                expected = ne.evaluate(f"nlhs {arith1} ghs")
+                # Update assert statement due to unreliable numerical
+                # precision component (GH37328)
+                # TODO: update testing code so that assert_almost_equal statement
+                #  can be replaced again by the assert_numpy_array_equal statement
+                tm.assert_almost_equal(result.values, expected)
+            else:
+                expected = eval(f"nlhs {arith1} ghs")
+                tm.assert_almost_equal(result, expected)
+
+    # modulus, pow, and floor division require special casing
+
+    def test_modulus(self, lhs, rhs, engine, parser):
+        ex = r"lhs % rhs"
+        result = pd.eval(ex, engine=engine, parser=parser)
+        expected = lhs % rhs
+        tm.assert_almost_equal(result, expected)
+
+        if engine == "numexpr":
+            import numexpr as ne
+
+            expected = ne.evaluate(r"expected % rhs")
+            if isinstance(result, (DataFrame, Series)):
+                tm.assert_almost_equal(result.values, expected)
+            else:
+                tm.assert_almost_equal(result, expected.item())
+        else:
+            expected = _eval_single_bin(expected, "%", rhs, engine)
+            tm.assert_almost_equal(result, expected)
+
+    def test_floor_division(self, lhs, rhs, engine, parser):
+        ex = "lhs // rhs"
+
+        if engine == "python":
+            res = pd.eval(ex, engine=engine, parser=parser)
+            expected = lhs // rhs
+            tm.assert_equal(res, expected)
+        else:
+            msg = (
+                r"unsupported operand type\(s\) for //: 'VariableNode' and "
+                "'VariableNode'"
+            )
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(
+                    ex,
+                    local_dict={"lhs": lhs, "rhs": rhs},
+                    engine=engine,
+                    parser=parser,
+                )
+
+    @td.skip_if_windows
+    def test_pow(self, lhs, rhs, engine, parser):
+        # odd failure on win32 platform, so skip
+        ex = "lhs ** rhs"
+        expected = _eval_single_bin(lhs, "**", rhs, engine)
+        result = pd.eval(ex, engine=engine, parser=parser)
+
+        if (
+            is_scalar(lhs)
+            and is_scalar(rhs)
+            and isinstance(expected, (complex, np.complexfloating))
+            and np.isnan(result)
+        ):
+            msg = "(DataFrame.columns|numpy array) are different"
+            with pytest.raises(AssertionError, match=msg):
+                tm.assert_numpy_array_equal(result, expected)
+        else:
+            tm.assert_almost_equal(result, expected)
+
+            ex = "(lhs ** rhs) ** rhs"
+            result = pd.eval(ex, engine=engine, parser=parser)
+
+            middle = _eval_single_bin(lhs, "**", rhs, engine)
+            expected = _eval_single_bin(middle, "**", rhs, engine)
+            tm.assert_almost_equal(result, expected)
+
+    def test_check_single_invert_op(self, lhs, engine, parser):
+        # simple
+        try:
+            elb = lhs.astype(bool)
+        except AttributeError:
+            elb = np.array([bool(lhs)])
+        expected = ~elb
+        result = pd.eval("~elb", engine=engine, parser=parser)
+        tm.assert_almost_equal(expected, result)
+
+    def test_frame_invert(self, engine, parser):
+        expr = "~lhs"
+
+        # ~ ##
+        # frame
+        # float always raises
+        lhs = DataFrame(np.random.default_rng(2).standard_normal((5, 2)))
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'invert_dd'"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            msg = "ufunc 'invert' not supported for the input types"
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+
+        # int raises on numexpr
+        lhs = DataFrame(np.random.default_rng(2).integers(5, size=(5, 2)))
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'invert"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            expect = ~lhs
+            result = pd.eval(expr, engine=engine, parser=parser)
+            tm.assert_frame_equal(expect, result)
+
+        # bool always works
+        lhs = DataFrame(np.random.default_rng(2).standard_normal((5, 2)) > 0.5)
+        expect = ~lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_frame_equal(expect, result)
+
+        # object raises
+        lhs = DataFrame(
+            {"b": ["a", 1, 2.0], "c": np.random.default_rng(2).standard_normal(3) > 0.5}
+        )
+        if engine == "numexpr":
+            with pytest.raises(ValueError, match="unknown type object"):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            msg = "bad operand type for unary ~: 'str'"
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+
+    def test_series_invert(self, engine, parser):
+        # ~ ####
+        expr = "~lhs"
+
+        # series
+        # float raises
+        lhs = Series(np.random.default_rng(2).standard_normal(5))
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'invert_dd'"
+            with pytest.raises(NotImplementedError, match=msg):
+                result = pd.eval(expr, engine=engine, parser=parser)
+        else:
+            msg = "ufunc 'invert' not supported for the input types"
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+
+        # int raises on numexpr
+        lhs = Series(np.random.default_rng(2).integers(5, size=5))
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'invert"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            expect = ~lhs
+            result = pd.eval(expr, engine=engine, parser=parser)
+            tm.assert_series_equal(expect, result)
+
+        # bool
+        lhs = Series(np.random.default_rng(2).standard_normal(5) > 0.5)
+        expect = ~lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_series_equal(expect, result)
+
+        # float
+        # int
+        # bool
+
+        # object
+        lhs = Series(["a", 1, 2.0])
+        if engine == "numexpr":
+            with pytest.raises(ValueError, match="unknown type object"):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            msg = "bad operand type for unary ~: 'str'"
+            with pytest.raises(TypeError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+
+    def test_frame_negate(self, engine, parser):
+        expr = "-lhs"
+
+        # float
+        lhs = DataFrame(np.random.default_rng(2).standard_normal((5, 2)))
+        expect = -lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_frame_equal(expect, result)
+
+        # int
+        lhs = DataFrame(np.random.default_rng(2).integers(5, size=(5, 2)))
+        expect = -lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_frame_equal(expect, result)
+
+        # bool doesn't work with numexpr but works elsewhere
+        lhs = DataFrame(np.random.default_rng(2).standard_normal((5, 2)) > 0.5)
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'neg_bb'"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            expect = -lhs
+            result = pd.eval(expr, engine=engine, parser=parser)
+            tm.assert_frame_equal(expect, result)
+
+    def test_series_negate(self, engine, parser):
+        expr = "-lhs"
+
+        # float
+        lhs = Series(np.random.default_rng(2).standard_normal(5))
+        expect = -lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_series_equal(expect, result)
+
+        # int
+        lhs = Series(np.random.default_rng(2).integers(5, size=5))
+        expect = -lhs
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_series_equal(expect, result)
+
+        # bool doesn't work with numexpr but works elsewhere
+        lhs = Series(np.random.default_rng(2).standard_normal(5) > 0.5)
+        if engine == "numexpr":
+            msg = "couldn't find matching opcode for 'neg_bb'"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(expr, engine=engine, parser=parser)
+        else:
+            expect = -lhs
+            result = pd.eval(expr, engine=engine, parser=parser)
+            tm.assert_series_equal(expect, result)
+
+    @pytest.mark.parametrize(
+        "lhs",
+        [
+            # Float
+            np.random.default_rng(2).standard_normal((5, 2)),
+            # Int
+            np.random.default_rng(2).integers(5, size=(5, 2)),
+            # bool doesn't work with numexpr but works elsewhere
+            np.array([True, False, True, False, True], dtype=np.bool_),
+        ],
+    )
+    def test_frame_pos(self, lhs, engine, parser):
+        lhs = DataFrame(lhs)
+        expr = "+lhs"
+        expect = lhs
+
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_frame_equal(expect, result)
+
+    @pytest.mark.parametrize(
+        "lhs",
+        [
+            # Float
+            np.random.default_rng(2).standard_normal(5),
+            # Int
+            np.random.default_rng(2).integers(5, size=5),
+            # bool doesn't work with numexpr but works elsewhere
+            np.array([True, False, True, False, True], dtype=np.bool_),
+        ],
+    )
+    def test_series_pos(self, lhs, engine, parser):
+        lhs = Series(lhs)
+        expr = "+lhs"
+        expect = lhs
+
+        result = pd.eval(expr, engine=engine, parser=parser)
+        tm.assert_series_equal(expect, result)
+
+    def test_scalar_unary(self, engine, parser):
+        msg = "bad operand type for unary ~: 'float'"
+        warn = None
+        if PY312 and not (engine == "numexpr" and parser == "pandas"):
+            warn = DeprecationWarning
+        with pytest.raises(TypeError, match=msg):
+            pd.eval("~1.0", engine=engine, parser=parser)
+
+        assert pd.eval("-1.0", parser=parser, engine=engine) == -1.0
+        assert pd.eval("+1.0", parser=parser, engine=engine) == +1.0
+        assert pd.eval("~1", parser=parser, engine=engine) == ~1
+        assert pd.eval("-1", parser=parser, engine=engine) == -1
+        assert pd.eval("+1", parser=parser, engine=engine) == +1
+        with tm.assert_produces_warning(
+            warn, match="Bitwise inversion", check_stacklevel=False
+        ):
+            assert pd.eval("~True", parser=parser, engine=engine) == ~True
+        with tm.assert_produces_warning(
+            warn, match="Bitwise inversion", check_stacklevel=False
+        ):
+            assert pd.eval("~False", parser=parser, engine=engine) == ~False
+        assert pd.eval("-True", parser=parser, engine=engine) == -True
+        assert pd.eval("-False", parser=parser, engine=engine) == -False
+        assert pd.eval("+True", parser=parser, engine=engine) == +True
+        assert pd.eval("+False", parser=parser, engine=engine) == +False
+
+    def test_unary_in_array(self):
+        # GH 11235
+        # TODO: 2022-01-29: result return list with numexpr 2.7.3 in CI
+        # but cannot reproduce locally
+        result = np.array(
+            pd.eval("[-True, True, +True, -False, False, +False, -37, 37, ~37, +37]"),
+            dtype=np.object_,
+        )
+        expected = np.array(
+            [
+                -True,
+                True,
+                +True,
+                -False,
+                False,
+                +False,
+                -37,
+                37,
+                ~37,
+                +37,
+            ],
+            dtype=np.object_,
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("expr", ["x < -0.1", "-5 > x"])
+    def test_float_comparison_bin_op(self, float_numpy_dtype, expr):
+        # GH 16363
+        df = DataFrame({"x": np.array([0], dtype=float_numpy_dtype)})
+        res = df.eval(expr)
+        assert res.values == np.array([False])
+
+    def test_unary_in_function(self):
+        # GH 46471
+        df = DataFrame({"x": [0, 1, np.nan]})
+
+        result = df.eval("x.fillna(-1)")
+        expected = df.x.fillna(-1)
+        # column name becomes None if using numexpr
+        # only check names when the engine is not numexpr
+        tm.assert_series_equal(result, expected, check_names=not USE_NUMEXPR)
+
+        result = df.eval("x.shift(1, fill_value=-1)")
+        expected = df.x.shift(1, fill_value=-1)
+        tm.assert_series_equal(result, expected, check_names=not USE_NUMEXPR)
+
+    @pytest.mark.parametrize(
+        "ex",
+        (
+            "1 or 2",
+            "1 and 2",
+            "a and b",
+            "a or b",
+            "1 or 2 and (3 + 2) > 3",
+            "2 * x > 2 or 1 and 2",
+            "2 * df > 3 and 1 or a",
+        ),
+    )
+    def test_disallow_scalar_bool_ops(self, ex, engine, parser):
+        x, a, b = np.random.default_rng(2).standard_normal(3), 1, 2  # noqa: F841
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 2)))  # noqa: F841
+
+        msg = "cannot evaluate scalar only bool ops|'BoolOp' nodes are not"
+        with pytest.raises(NotImplementedError, match=msg):
+            pd.eval(ex, engine=engine, parser=parser)
+
+    def test_identical(self, engine, parser):
+        # see gh-10546
+        x = 1
+        result = pd.eval("x", engine=engine, parser=parser)
+        assert result == 1
+        assert is_scalar(result)
+
+        x = 1.5
+        result = pd.eval("x", engine=engine, parser=parser)
+        assert result == 1.5
+        assert is_scalar(result)
+
+        x = False
+        result = pd.eval("x", engine=engine, parser=parser)
+        assert not result
+        assert is_bool(result)
+        assert is_scalar(result)
+
+        x = np.array([1])
+        result = pd.eval("x", engine=engine, parser=parser)
+        tm.assert_numpy_array_equal(result, np.array([1]))
+        assert result.shape == (1,)
+
+        x = np.array([1.5])
+        result = pd.eval("x", engine=engine, parser=parser)
+        tm.assert_numpy_array_equal(result, np.array([1.5]))
+        assert result.shape == (1,)
+
+        x = np.array([False])  # noqa: F841
+        result = pd.eval("x", engine=engine, parser=parser)
+        tm.assert_numpy_array_equal(result, np.array([False]))
+        assert result.shape == (1,)
+
+    def test_line_continuation(self, engine, parser):
+        # GH 11149
+        exp = """1 + 2 * \
+        5 - 1 + 2 """
+        result = pd.eval(exp, engine=engine, parser=parser)
+        assert result == 12
+
+    def test_float_truncation(self, engine, parser):
+        # GH 14241
+        exp = "1000000000.006"
+        result = pd.eval(exp, engine=engine, parser=parser)
+        expected = np.float64(exp)
+        assert result == expected
+
+        df = DataFrame({"A": [1000000000.0009, 1000000000.0011, 1000000000.0015]})
+        cutoff = 1000000000.0006
+        result = df.query(f"A < {cutoff:.4f}")
+        assert result.empty
+
+        cutoff = 1000000000.0010
+        result = df.query(f"A > {cutoff:.4f}")
+        expected = df.loc[[1, 2], :]
+        tm.assert_frame_equal(expected, result)
+
+        exact = 1000000000.0011
+        result = df.query(f"A == {exact:.4f}")
+        expected = df.loc[[1], :]
+        tm.assert_frame_equal(expected, result)
+
+    def test_disallow_python_keywords(self):
+        # GH 18221
+        df = DataFrame([[0, 0, 0]], columns=["foo", "bar", "class"])
+        msg = "Python keyword not valid identifier in numexpr query"
+        with pytest.raises(SyntaxError, match=msg):
+            df.query("class == 0")
+
+        df = DataFrame()
+        df.index.name = "lambda"
+        with pytest.raises(SyntaxError, match=msg):
+            df.query("lambda == 0")
+
+    def test_true_false_logic(self):
+        # GH 25823
+        # This behavior is deprecated in Python 3.12
+        with tm.maybe_produces_warning(
+            DeprecationWarning, PY312, check_stacklevel=False
+        ):
+            assert pd.eval("not True") == -2
+            assert pd.eval("not False") == -1
+            assert pd.eval("True and not True") == 0
+
+    def test_and_logic_string_match(self):
+        # GH 25823
+        event = Series({"a": "hello"})
+        assert pd.eval(f"{event.str.match('hello').a}")
+        assert pd.eval(f"{event.str.match('hello').a and event.str.match('hello').a}")
+
+    def test_eval_keep_name(self, engine, parser):
+        df = Series([2, 15, 28], name="a").to_frame()
+        res = df.eval("a + a", engine=engine, parser=parser)
+        expected = Series([4, 30, 56], name="a")
+        tm.assert_series_equal(expected, res)
+
+    def test_eval_unmatching_names(self, engine, parser):
+        variable_name = Series([42], name="series_name")
+        res = pd.eval("variable_name + 0", engine=engine, parser=parser)
+        tm.assert_series_equal(variable_name, res)
+
+
+# -------------------------------------
+# gh-12388: Typecasting rules consistency with python
+
+
+class TestTypeCasting:
+    @pytest.mark.parametrize("op", ["+", "-", "*", "**", "/"])
+    # maybe someday... numexpr has too many upcasting rules now
+    # chain(*(np.core.sctypes[x] for x in ['uint', 'int', 'float']))
+    @pytest.mark.parametrize("left_right", [("df", "3"), ("3", "df")])
+    def test_binop_typecasting(
+        self, engine, parser, op, complex_or_float_dtype, left_right, request
+    ):
+        # GH#21374
+        dtype = complex_or_float_dtype
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)), dtype=dtype)
+        left, right = left_right
+        s = f"{left} {op} {right}"
+        res = pd.eval(s, engine=engine, parser=parser)
+        if dtype == "complex64" and engine == "numexpr":
+            mark = pytest.mark.xfail(
+                reason="numexpr issue with complex that are upcast "
+                "to complex 128 "
+                "https://github.com/pydata/numexpr/issues/492"
+            )
+            request.applymarker(mark)
+        assert df.values.dtype == dtype
+        assert res.values.dtype == dtype
+        tm.assert_frame_equal(res, eval(s), check_exact=False)
+
+
+# -------------------------------------
+# Basic and complex alignment
+
+
+def should_warn(*args):
+    not_mono = not any(map(operator.attrgetter("is_monotonic_increasing"), args))
+    only_one_dt = reduce(
+        operator.xor, (issubclass(x.dtype.type, np.datetime64) for x in args)
+    )
+    return not_mono and only_one_dt
+
+
+class TestAlignment:
+    index_types = ["i", "s", "dt"]
+    lhs_index_types = index_types + ["s"]  # 'p'
+
+    def test_align_nested_unary_op(self, engine, parser):
+        s = "df * ~2"
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        res = pd.eval(s, engine=engine, parser=parser)
+        tm.assert_frame_equal(res, df * ~2)
+
+    @pytest.mark.filterwarnings("always::RuntimeWarning")
+    @pytest.mark.parametrize("lr_idx_type", lhs_index_types)
+    @pytest.mark.parametrize("rr_idx_type", index_types)
+    @pytest.mark.parametrize("c_idx_type", index_types)
+    def test_basic_frame_alignment(
+        self, engine, parser, lr_idx_type, rr_idx_type, c_idx_type, idx_func_dict
+    ):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 10)),
+            index=idx_func_dict[lr_idx_type](10),
+            columns=idx_func_dict[c_idx_type](10),
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 10)),
+            index=idx_func_dict[rr_idx_type](20),
+            columns=idx_func_dict[c_idx_type](10),
+        )
+        # only warns if not monotonic and not sortable
+        if should_warn(df.index, df2.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                res = pd.eval("df + df2", engine=engine, parser=parser)
+        else:
+            res = pd.eval("df + df2", engine=engine, parser=parser)
+        tm.assert_frame_equal(res, df + df2)
+
+    @pytest.mark.parametrize("r_idx_type", lhs_index_types)
+    @pytest.mark.parametrize("c_idx_type", lhs_index_types)
+    def test_frame_comparison(
+        self, engine, parser, r_idx_type, c_idx_type, idx_func_dict
+    ):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 10)),
+            index=idx_func_dict[r_idx_type](10),
+            columns=idx_func_dict[c_idx_type](10),
+        )
+        res = pd.eval("df < 2", engine=engine, parser=parser)
+        tm.assert_frame_equal(res, df < 2)
+
+        df3 = DataFrame(
+            np.random.default_rng(2).standard_normal(df.shape),
+            index=df.index,
+            columns=df.columns,
+        )
+        res = pd.eval("df < df3", engine=engine, parser=parser)
+        tm.assert_frame_equal(res, df < df3)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("r1", lhs_index_types)
+    @pytest.mark.parametrize("c1", index_types)
+    @pytest.mark.parametrize("r2", index_types)
+    @pytest.mark.parametrize("c2", index_types)
+    def test_medium_complex_frame_alignment(
+        self, engine, parser, r1, c1, r2, c2, idx_func_dict
+    ):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 2)),
+            index=idx_func_dict[r1](3),
+            columns=idx_func_dict[c1](2),
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)),
+            index=idx_func_dict[r2](4),
+            columns=idx_func_dict[c2](2),
+        )
+        df3 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)),
+            index=idx_func_dict[r2](5),
+            columns=idx_func_dict[c2](2),
+        )
+        if should_warn(df.index, df2.index, df3.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                res = pd.eval("df + df2 + df3", engine=engine, parser=parser)
+        else:
+            res = pd.eval("df + df2 + df3", engine=engine, parser=parser)
+        tm.assert_frame_equal(res, df + df2 + df3)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("index_name", ["index", "columns"])
+    @pytest.mark.parametrize("c_idx_type", index_types)
+    @pytest.mark.parametrize("r_idx_type", lhs_index_types)
+    def test_basic_frame_series_alignment(
+        self, engine, parser, index_name, r_idx_type, c_idx_type, idx_func_dict
+    ):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 10)),
+            index=idx_func_dict[r_idx_type](10),
+            columns=idx_func_dict[c_idx_type](10),
+        )
+        index = getattr(df, index_name)
+        s = Series(np.random.default_rng(2).standard_normal(5), index[:5])
+
+        if should_warn(df.index, s.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                res = pd.eval("df + s", engine=engine, parser=parser)
+        else:
+            res = pd.eval("df + s", engine=engine, parser=parser)
+
+        if r_idx_type == "dt" or c_idx_type == "dt":
+            expected = df.add(s) if engine == "numexpr" else df + s
+        else:
+            expected = df + s
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.parametrize("index_name", ["index", "columns"])
+    @pytest.mark.parametrize(
+        "r_idx_type, c_idx_type",
+        list(product(["i", "s"], ["i", "s"])) + [("dt", "dt")],
+    )
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_basic_series_frame_alignment(
+        self, request, engine, parser, index_name, r_idx_type, c_idx_type, idx_func_dict
+    ):
+        if (
+            engine == "numexpr"
+            and parser in ("pandas", "python")
+            and index_name == "index"
+            and r_idx_type == "i"
+            and c_idx_type == "s"
+        ):
+            reason = (
+                f"Flaky column ordering when engine={engine}, "
+                f"parser={parser}, index_name={index_name}, "
+                f"r_idx_type={r_idx_type}, c_idx_type={c_idx_type}"
+            )
+            request.applymarker(pytest.mark.xfail(reason=reason, strict=False))
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 7)),
+            index=idx_func_dict[r_idx_type](10),
+            columns=idx_func_dict[c_idx_type](7),
+        )
+        index = getattr(df, index_name)
+        s = Series(np.random.default_rng(2).standard_normal(5), index[:5])
+        if should_warn(s.index, df.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                res = pd.eval("s + df", engine=engine, parser=parser)
+        else:
+            res = pd.eval("s + df", engine=engine, parser=parser)
+
+        if r_idx_type == "dt" or c_idx_type == "dt":
+            expected = df.add(s) if engine == "numexpr" else s + df
+        else:
+            expected = s + df
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("c_idx_type", index_types)
+    @pytest.mark.parametrize("r_idx_type", lhs_index_types)
+    @pytest.mark.parametrize("index_name", ["index", "columns"])
+    @pytest.mark.parametrize("op", ["+", "*"])
+    def test_series_frame_commutativity(
+        self, engine, parser, index_name, op, r_idx_type, c_idx_type, idx_func_dict
+    ):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 10)),
+            index=idx_func_dict[r_idx_type](10),
+            columns=idx_func_dict[c_idx_type](10),
+        )
+        index = getattr(df, index_name)
+        s = Series(np.random.default_rng(2).standard_normal(5), index[:5])
+
+        lhs = f"s {op} df"
+        rhs = f"df {op} s"
+        if should_warn(df.index, s.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                a = pd.eval(lhs, engine=engine, parser=parser)
+            with tm.assert_produces_warning(RuntimeWarning):
+                b = pd.eval(rhs, engine=engine, parser=parser)
+        else:
+            a = pd.eval(lhs, engine=engine, parser=parser)
+            b = pd.eval(rhs, engine=engine, parser=parser)
+
+        if r_idx_type != "dt" and c_idx_type != "dt":
+            if engine == "numexpr":
+                tm.assert_frame_equal(a, b)
+
+    @pytest.mark.filterwarnings("always::RuntimeWarning")
+    @pytest.mark.parametrize("r1", lhs_index_types)
+    @pytest.mark.parametrize("c1", index_types)
+    @pytest.mark.parametrize("r2", index_types)
+    @pytest.mark.parametrize("c2", index_types)
+    def test_complex_series_frame_alignment(
+        self, engine, parser, r1, c1, r2, c2, idx_func_dict
+    ):
+        n = 3
+        m1 = 5
+        m2 = 2 * m1
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((m1, n)),
+            index=idx_func_dict[r1](m1),
+            columns=idx_func_dict[c1](n),
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((m2, n)),
+            index=idx_func_dict[r2](m2),
+            columns=idx_func_dict[c2](n),
+        )
+        index = df2.columns
+        ser = Series(np.random.default_rng(2).standard_normal(n), index[:n])
+
+        if r2 == "dt" or c2 == "dt":
+            if engine == "numexpr":
+                expected2 = df2.add(ser)
+            else:
+                expected2 = df2 + ser
+        else:
+            expected2 = df2 + ser
+
+        if r1 == "dt" or c1 == "dt":
+            if engine == "numexpr":
+                expected = expected2.add(df)
+            else:
+                expected = expected2 + df
+        else:
+            expected = expected2 + df
+
+        if should_warn(df2.index, ser.index, df.index):
+            with tm.assert_produces_warning(RuntimeWarning):
+                res = pd.eval("df2 + ser + df", engine=engine, parser=parser)
+        else:
+            res = pd.eval("df2 + ser + df", engine=engine, parser=parser)
+        assert res.shape == expected.shape
+        tm.assert_frame_equal(res, expected)
+
+    def test_performance_warning_for_poor_alignment(
+        self, performance_warning, engine, parser
+    ):
+        df = DataFrame(np.random.default_rng(2).standard_normal((1000, 10)))
+        s = Series(np.random.default_rng(2).standard_normal(10000))
+        if engine == "numexpr" and performance_warning:
+            seen = PerformanceWarning
+        else:
+            seen = False
+
+        msg = "Alignment difference on axis 1 is larger than an order of magnitude"
+        with tm.assert_produces_warning(seen, match=msg):
+            pd.eval("df + s", engine=engine, parser=parser)
+
+        s = Series(np.random.default_rng(2).standard_normal(1000))
+        with tm.assert_produces_warning(False):
+            pd.eval("df + s", engine=engine, parser=parser)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 10000)))
+        s = Series(np.random.default_rng(2).standard_normal(10000))
+        with tm.assert_produces_warning(False):
+            pd.eval("df + s", engine=engine, parser=parser)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 10)))
+        s = Series(np.random.default_rng(2).standard_normal(10000))
+
+        is_python_engine = engine == "python"
+
+        if not is_python_engine and performance_warning:
+            wrn = PerformanceWarning
+        else:
+            wrn = False
+
+        with tm.assert_produces_warning(wrn, match=msg) as w:
+            pd.eval("df + s", engine=engine, parser=parser)
+
+            if not is_python_engine and performance_warning:
+                assert len(w) == 1
+                msg = str(w[0].message)
+                logged = np.log10(s.size - df.shape[1])
+                expected = (
+                    f"Alignment difference on axis 1 is larger "
+                    f"than an order of magnitude on term 'df', "
+                    f"by more than {logged:.4g}; performance may suffer."
+                )
+                assert msg == expected
+
+
+# ------------------------------------
+# Slightly more complex ops
+
+
+class TestOperations:
+    def eval(self, *args, **kwargs):
+        kwargs["level"] = kwargs.pop("level", 0) + 1
+        return pd.eval(*args, **kwargs)
+
+    def test_simple_arith_ops(self, engine, parser):
+        exclude_arith = []
+        if parser == "python":
+            exclude_arith = ["in", "not in"]
+
+        arith_ops = [
+            op
+            for op in expr.ARITH_OPS_SYMS + expr.CMP_OPS_SYMS
+            if op not in exclude_arith
+        ]
+
+        ops = (op for op in arith_ops if op != "//")
+
+        for op in ops:
+            ex = f"1 {op} 1"
+            ex2 = f"x {op} 1"
+            ex3 = f"1 {op} (x + 1)"
+
+            if op in ("in", "not in"):
+                msg = "argument of type 'int' is not iterable"
+                with pytest.raises(TypeError, match=msg):
+                    pd.eval(ex, engine=engine, parser=parser)
+            else:
+                expec = _eval_single_bin(1, op, 1, engine)
+                x = self.eval(ex, engine=engine, parser=parser)
+                assert x == expec
+
+                expec = _eval_single_bin(x, op, 1, engine)
+                y = self.eval(ex2, local_dict={"x": x}, engine=engine, parser=parser)
+                assert y == expec
+
+                expec = _eval_single_bin(1, op, x + 1, engine)
+                y = self.eval(ex3, local_dict={"x": x}, engine=engine, parser=parser)
+                assert y == expec
+
+    @pytest.mark.parametrize("rhs", [True, False])
+    @pytest.mark.parametrize("lhs", [True, False])
+    @pytest.mark.parametrize("op", expr.BOOL_OPS_SYMS)
+    def test_simple_bool_ops(self, rhs, lhs, op):
+        ex = f"{lhs} {op} {rhs}"
+
+        if parser == "python" and op in ["and", "or"]:
+            msg = "'BoolOp' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                self.eval(ex)
+            return
+
+        res = self.eval(ex)
+        exp = eval(ex)
+        assert res == exp
+
+    @pytest.mark.parametrize("rhs", [True, False])
+    @pytest.mark.parametrize("lhs", [True, False])
+    @pytest.mark.parametrize("op", expr.BOOL_OPS_SYMS)
+    def test_bool_ops_with_constants(self, rhs, lhs, op):
+        ex = f"{lhs} {op} {rhs}"
+
+        if parser == "python" and op in ["and", "or"]:
+            msg = "'BoolOp' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                self.eval(ex)
+            return
+
+        res = self.eval(ex)
+        exp = eval(ex)
+        assert res == exp
+
+    def test_4d_ndarray_fails(self):
+        x = np.random.default_rng(2).standard_normal((3, 4, 5, 6))
+        y = Series(np.random.default_rng(2).standard_normal(10))
+        msg = "N-dimensional objects, where N > 2, are not supported with eval"
+        with pytest.raises(NotImplementedError, match=msg):
+            self.eval("x + y", local_dict={"x": x, "y": y})
+
+    def test_constant(self):
+        x = self.eval("1")
+        assert x == 1
+
+    def test_single_variable(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        df2 = self.eval("df", local_dict={"df": df})
+        tm.assert_frame_equal(df, df2)
+
+    def test_failing_subscript_with_name_error(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))  # noqa: F841
+        with pytest.raises(NameError, match="name 'x' is not defined"):
+            self.eval("df[x > 2] > 2")
+
+    def test_lhs_expression_subscript(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        result = self.eval("(df + 1)[df > 2]", local_dict={"df": df})
+        expected = (df + 1)[df > 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_attr_expression(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=list("abc")
+        )
+        expr1 = "df.a < df.b"
+        expec1 = df.a < df.b
+        expr2 = "df.a + df.b + df.c"
+        expec2 = df.a + df.b + df.c
+        expr3 = "df.a + df.b + df.c[df.b < 0]"
+        expec3 = df.a + df.b + df.c[df.b < 0]
+        exprs = expr1, expr2, expr3
+        expecs = expec1, expec2, expec3
+        for e, expec in zip(exprs, expecs):
+            tm.assert_series_equal(expec, self.eval(e, local_dict={"df": df}))
+
+    def test_assignment_fails(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=list("abc")
+        )
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        expr1 = "df = df2"
+        msg = "cannot assign without a target object"
+        with pytest.raises(ValueError, match=msg):
+            self.eval(expr1, local_dict={"df": df, "df2": df2})
+
+    def test_assignment_column_multiple_raise(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # multiple assignees
+        with pytest.raises(SyntaxError, match="invalid syntax"):
+            df.eval("d c = a + b")
+
+    def test_assignment_column_invalid_assign(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # invalid assignees
+        msg = "left hand side of an assignment must be a single name"
+        with pytest.raises(SyntaxError, match=msg):
+            df.eval("d,c = a + b")
+
+    def test_assignment_column_invalid_assign_function_call(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        msg = "cannot assign to function call"
+        with pytest.raises(SyntaxError, match=msg):
+            df.eval('Timestamp("20131001") = a + b')
+
+    def test_assignment_single_assign_existing(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # single assignment - existing variable
+        expected = df.copy()
+        expected["a"] = expected["a"] + expected["b"]
+        df.eval("a = a + b", inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+    def test_assignment_single_assign_new(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # single assignment - new variable
+        expected = df.copy()
+        expected["c"] = expected["a"] + expected["b"]
+        df.eval("c = a + b", inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+    def test_assignment_single_assign_local_overlap(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        df = df.copy()
+        a = 1  # noqa: F841
+        df.eval("a = 1 + b", inplace=True)
+
+        expected = df.copy()
+        expected["a"] = 1 + expected["b"]
+        tm.assert_frame_equal(df, expected)
+
+    def test_assignment_single_assign_name(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+
+        a = 1  # noqa: F841
+        old_a = df.a.copy()
+        df.eval("a = a + b", inplace=True)
+        result = old_a + df.b
+        tm.assert_series_equal(result, df.a, check_names=False)
+        assert result.name is None
+
+    def test_assignment_multiple_raises(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # multiple assignment
+        df.eval("c = a + b", inplace=True)
+        msg = "can only assign a single expression"
+        with pytest.raises(SyntaxError, match=msg):
+            df.eval("c = a = b")
+
+    def test_assignment_explicit(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+        # explicit targets
+        self.eval("c = df.a + df.b", local_dict={"df": df}, target=df, inplace=True)
+        expected = df.copy()
+        expected["c"] = expected["a"] + expected["b"]
+        tm.assert_frame_equal(df, expected)
+
+    def test_column_in(self, engine):
+        # GH 11235
+        df = DataFrame({"a": [11], "b": [-32]})
+        result = df.eval("a in [11, -32]", engine=engine)
+        expected = Series([True], name="a")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.xfail(reason="Unknown: Omitted test_ in name prior.")
+    def test_assignment_not_inplace(self):
+        # see gh-9297
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("ab")
+        )
+
+        actual = df.eval("c = a + b", inplace=False)
+        assert actual is not None
+
+        expected = df.copy()
+        expected["c"] = expected["a"] + expected["b"]
+        tm.assert_frame_equal(df, expected)
+
+    def test_multi_line_expression(self):
+        # GH 11149
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        expected = df.copy()
+
+        expected["c"] = expected["a"] + expected["b"]
+        expected["d"] = expected["c"] + expected["b"]
+        answer = df.eval(
+            """
+        c = a + b
+        d = c + b""",
+            inplace=True,
+        )
+        tm.assert_frame_equal(expected, df)
+        assert answer is None
+
+        expected["a"] = expected["a"] - 1
+        expected["e"] = expected["a"] + 2
+        answer = df.eval(
+            """
+        a = a - 1
+        e = a + 2""",
+            inplace=True,
+        )
+        tm.assert_frame_equal(expected, df)
+        assert answer is None
+
+        # multi-line not valid if not all assignments
+        msg = "Multi-line expressions are only valid if all expressions contain"
+        with pytest.raises(ValueError, match=msg):
+            df.eval(
+                """
+            a = b + 2
+            b - 2""",
+                inplace=False,
+            )
+
+    def test_multi_line_expression_not_inplace(self):
+        # GH 11149
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        expected = df.copy()
+
+        expected["c"] = expected["a"] + expected["b"]
+        expected["d"] = expected["c"] + expected["b"]
+        df = df.eval(
+            """
+        c = a + b
+        d = c + b""",
+            inplace=False,
+        )
+        tm.assert_frame_equal(expected, df)
+
+        expected["a"] = expected["a"] - 1
+        expected["e"] = expected["a"] + 2
+        df = df.eval(
+            """
+        a = a - 1
+        e = a + 2""",
+            inplace=False,
+        )
+        tm.assert_frame_equal(expected, df)
+
+    def test_multi_line_expression_local_variable(self):
+        # GH 15342
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        expected = df.copy()
+
+        local_var = 7
+        expected["c"] = expected["a"] * local_var
+        expected["d"] = expected["c"] + local_var
+        answer = df.eval(
+            """
+        c = a * @local_var
+        d = c + @local_var
+        """,
+            inplace=True,
+        )
+        tm.assert_frame_equal(expected, df)
+        assert answer is None
+
+    def test_multi_line_expression_callable_local_variable(self):
+        # 26426
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+        def local_func(a, b):
+            return b
+
+        expected = df.copy()
+        expected["c"] = expected["a"] * local_func(1, 7)
+        expected["d"] = expected["c"] + local_func(1, 7)
+        answer = df.eval(
+            """
+        c = a * @local_func(1, 7)
+        d = c + @local_func(1, 7)
+        """,
+            inplace=True,
+        )
+        tm.assert_frame_equal(expected, df)
+        assert answer is None
+
+    def test_multi_line_expression_callable_local_variable_with_kwargs(self):
+        # 26426
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+        def local_func(a, b):
+            return b
+
+        expected = df.copy()
+        expected["c"] = expected["a"] * local_func(b=7, a=1)
+        expected["d"] = expected["c"] + local_func(b=7, a=1)
+        answer = df.eval(
+            """
+        c = a * @local_func(b=7, a=1)
+        d = c + @local_func(b=7, a=1)
+        """,
+            inplace=True,
+        )
+        tm.assert_frame_equal(expected, df)
+        assert answer is None
+
+    def test_assignment_in_query(self):
+        # GH 8664
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df_orig = df.copy()
+        msg = "cannot assign without a target object"
+        with pytest.raises(ValueError, match=msg):
+            df.query("a = 1")
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_query_inplace(self):
+        # see gh-11149
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        expected = df.copy()
+        expected = expected[expected["a"] == 2]
+        df.query("a == 2", inplace=True)
+        tm.assert_frame_equal(expected, df)
+
+        df = {}
+        expected = {"a": 3}
+
+        self.eval("a = 1 + 2", target=df, inplace=True)
+        tm.assert_dict_equal(df, expected)
+
+    @pytest.mark.parametrize("invalid_target", [1, "cat", [1, 2], np.array([]), (1, 3)])
+    def test_cannot_item_assign(self, invalid_target):
+        msg = "Cannot assign expression output to target"
+        expression = "a = 1 + 2"
+
+        with pytest.raises(ValueError, match=msg):
+            self.eval(expression, target=invalid_target, inplace=True)
+
+        if hasattr(invalid_target, "copy"):
+            with pytest.raises(ValueError, match=msg):
+                self.eval(expression, target=invalid_target, inplace=False)
+
+    @pytest.mark.parametrize("invalid_target", [1, "cat", (1, 3)])
+    def test_cannot_copy_item(self, invalid_target):
+        msg = "Cannot return a copy of the target"
+        expression = "a = 1 + 2"
+
+        with pytest.raises(ValueError, match=msg):
+            self.eval(expression, target=invalid_target, inplace=False)
+
+    @pytest.mark.parametrize("target", [1, "cat", [1, 2], np.array([]), (1, 3), {1: 2}])
+    def test_inplace_no_assignment(self, target):
+        expression = "1 + 2"
+
+        assert self.eval(expression, target=target, inplace=False) == 3
+
+        msg = "Cannot operate inplace if there is no assignment"
+        with pytest.raises(ValueError, match=msg):
+            self.eval(expression, target=target, inplace=True)
+
+    def test_basic_period_index_boolean_expression(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((2, 2)),
+            columns=period_range("2020-01-01", freq="D", periods=2),
+        )
+        e = df < 2
+        r = self.eval("df < 2", local_dict={"df": df})
+        x = df < 2
+
+        tm.assert_frame_equal(r, e)
+        tm.assert_frame_equal(x, e)
+
+    def test_basic_period_index_subscript_expression(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((2, 2)),
+            columns=period_range("2020-01-01", freq="D", periods=2),
+        )
+        r = self.eval("df[df < 2 + 3]", local_dict={"df": df})
+        e = df[df < 2 + 3]
+        tm.assert_frame_equal(r, e)
+
+    def test_nested_period_index_subscript_expression(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((2, 2)),
+            columns=period_range("2020-01-01", freq="D", periods=2),
+        )
+        r = self.eval("df[df[df < 2] < 2] + df * 2", local_dict={"df": df})
+        e = df[df[df < 2] < 2] + df * 2
+        tm.assert_frame_equal(r, e)
+
+    def test_date_boolean(self, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=5)
+        res = self.eval(
+            "df.dates1 < 20130101",
+            local_dict={"df": df},
+            engine=engine,
+            parser=parser,
+        )
+        expec = df.dates1 < "20130101"
+        tm.assert_series_equal(res, expec)
+
+    def test_simple_in_ops(self, engine, parser):
+        if parser != "python":
+            res = pd.eval("1 in [1, 2]", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("2 in (1, 2)", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("3 in (1, 2)", engine=engine, parser=parser)
+            assert not res
+
+            res = pd.eval("3 not in (1, 2)", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("[3] not in (1, 2)", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("[3] in ([3], 2)", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("[[3]] in [[[3]], 2]", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("(3,) in [(3,), 2]", engine=engine, parser=parser)
+            assert res
+
+            res = pd.eval("(3,) not in [(3,), 2]", engine=engine, parser=parser)
+            assert not res
+
+            res = pd.eval("[(3,)] in [[(3,)], 2]", engine=engine, parser=parser)
+            assert res
+        else:
+            msg = "'In' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("1 in [1, 2]", engine=engine, parser=parser)
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("2 in (1, 2)", engine=engine, parser=parser)
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("3 in (1, 2)", engine=engine, parser=parser)
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("[(3,)] in (1, 2, [(3,)])", engine=engine, parser=parser)
+            msg = "'NotIn' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("3 not in (1, 2)", engine=engine, parser=parser)
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval("[3] not in (1, 2, [[3]])", engine=engine, parser=parser)
+
+    def test_check_many_exprs(self, engine, parser):
+        a = 1  # noqa: F841
+        expr = " * ".join("a" * 33)
+        expected = 1
+        res = pd.eval(expr, engine=engine, parser=parser)
+        assert res == expected
+
+    @pytest.mark.parametrize(
+        "expr",
+        [
+            "df > 2 and df > 3",
+            "df > 2 or df > 3",
+            "not df > 2",
+        ],
+    )
+    def test_fails_and_or_not(self, expr, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        if parser == "python":
+            msg = "'BoolOp' nodes are not implemented"
+            if "not" in expr:
+                msg = "'Not' nodes are not implemented"
+
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(
+                    expr,
+                    local_dict={"df": df},
+                    parser=parser,
+                    engine=engine,
+                )
+        else:
+            # smoke-test, should not raise
+            pd.eval(
+                expr,
+                local_dict={"df": df},
+                parser=parser,
+                engine=engine,
+            )
+
+    @pytest.mark.parametrize("char", ["|", "&"])
+    def test_fails_ampersand_pipe(self, char, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))  # noqa: F841
+        ex = f"(df + 2)[df > 1] > 0 {char} (df > 0)"
+        if parser == "python":
+            msg = "cannot evaluate scalar only bool ops"
+            with pytest.raises(NotImplementedError, match=msg):
+                pd.eval(ex, parser=parser, engine=engine)
+        else:
+            # smoke-test, should not raise
+            pd.eval(ex, parser=parser, engine=engine)
+
+
+class TestMath:
+    def eval(self, *args, **kwargs):
+        kwargs["level"] = kwargs.pop("level", 0) + 1
+        return pd.eval(*args, **kwargs)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("fn", _unary_math_ops)
+    def test_unary_functions(self, fn, engine, parser):
+        df = DataFrame({"a": np.random.default_rng(2).standard_normal(10)})
+        a = df.a
+
+        expr = f"{fn}(a)"
+        got = self.eval(expr, engine=engine, parser=parser)
+        with np.errstate(all="ignore"):
+            expect = getattr(np, fn)(a)
+        tm.assert_series_equal(got, expect)
+
+    @pytest.mark.parametrize("fn", _binary_math_ops)
+    def test_binary_functions(self, fn, engine, parser):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+        a = df.a
+        b = df.b
+
+        expr = f"{fn}(a, b)"
+        got = self.eval(expr, engine=engine, parser=parser)
+        with np.errstate(all="ignore"):
+            expect = getattr(np, fn)(a, b)
+        tm.assert_almost_equal(got, expect)
+
+    def test_df_use_case(self, engine, parser):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+        df.eval(
+            "e = arctan2(sin(a), b)",
+            engine=engine,
+            parser=parser,
+            inplace=True,
+        )
+        got = df.e
+        expect = np.arctan2(np.sin(df.a), df.b).rename("e")
+        tm.assert_series_equal(got, expect)
+
+    def test_df_arithmetic_subexpression(self, engine, parser):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+        df.eval("e = sin(a + b)", engine=engine, parser=parser, inplace=True)
+        got = df.e
+        expect = np.sin(df.a + df.b).rename("e")
+        tm.assert_series_equal(got, expect)
+
+    @pytest.mark.parametrize(
+        "dtype, expect_dtype",
+        [
+            (np.int32, np.float64),
+            (np.int64, np.float64),
+            (np.float32, np.float32),
+            (np.float64, np.float64),
+            pytest.param(np.complex128, np.complex128, marks=td.skip_if_windows),
+        ],
+    )
+    def test_result_types(self, dtype, expect_dtype, engine, parser):
+        # xref https://github.com/pandas-dev/pandas/issues/12293
+        #  this fails on Windows, apparently a floating point precision issue
+
+        # Did not test complex64 because DataFrame is converting it to
+        # complex128. Due to https://github.com/pandas-dev/pandas/issues/10952
+        df = DataFrame(
+            {"a": np.random.default_rng(2).standard_normal(10).astype(dtype)}
+        )
+        assert df.a.dtype == dtype
+        df.eval("b = sin(a)", engine=engine, parser=parser, inplace=True)
+        got = df.b
+        expect = np.sin(df.a).rename("b")
+        assert expect.dtype == got.dtype
+        assert expect_dtype == got.dtype
+        tm.assert_series_equal(got, expect)
+
+    def test_undefined_func(self, engine, parser):
+        df = DataFrame({"a": np.random.default_rng(2).standard_normal(10)})
+        msg = '"mysin" is not a supported function'
+
+        with pytest.raises(ValueError, match=msg):
+            df.eval("mysin(a)", engine=engine, parser=parser)
+
+    def test_keyword_arg(self, engine, parser):
+        df = DataFrame({"a": np.random.default_rng(2).standard_normal(10)})
+        msg = 'Function "sin" does not support keyword arguments'
+
+        with pytest.raises(TypeError, match=msg):
+            df.eval("sin(x=a)", engine=engine, parser=parser)
+
+
+_var_s = np.random.default_rng(2).standard_normal(10)
+
+
+class TestScope:
+    def test_global_scope(self, engine, parser):
+        e = "_var_s * 2"
+        tm.assert_numpy_array_equal(
+            _var_s * 2, pd.eval(e, engine=engine, parser=parser)
+        )
+
+    def test_no_new_locals(self, engine, parser):
+        x = 1
+        lcls = locals().copy()
+        pd.eval("x + 1", local_dict=lcls, engine=engine, parser=parser)
+        lcls2 = locals().copy()
+        lcls2.pop("lcls")
+        assert lcls == lcls2
+
+    def test_no_new_globals(self, engine, parser):
+        x = 1  # noqa: F841
+        gbls = globals().copy()
+        pd.eval("x + 1", engine=engine, parser=parser)
+        gbls2 = globals().copy()
+        assert gbls == gbls2
+
+    def test_empty_locals(self, engine, parser):
+        # GH 47084
+        x = 1  # noqa: F841
+        msg = "name 'x' is not defined"
+        with pytest.raises(UndefinedVariableError, match=msg):
+            pd.eval("x + 1", engine=engine, parser=parser, local_dict={})
+
+    def test_empty_globals(self, engine, parser):
+        # GH 47084
+        msg = "name '_var_s' is not defined"
+        e = "_var_s * 2"
+        with pytest.raises(UndefinedVariableError, match=msg):
+            pd.eval(e, engine=engine, parser=parser, global_dict={})
+
+
+@td.skip_if_no("numexpr")
+def test_invalid_engine():
+    msg = "Invalid engine 'asdf' passed"
+    with pytest.raises(KeyError, match=msg):
+        pd.eval("x + y", local_dict={"x": 1, "y": 2}, engine="asdf")
+
+
+@td.skip_if_no("numexpr")
+@pytest.mark.parametrize(
+    ("use_numexpr", "expected"),
+    (
+        (True, "numexpr"),
+        (False, "python"),
+    ),
+)
+def test_numexpr_option_respected(use_numexpr, expected):
+    # GH 32556
+    from pandas.core.computation.eval import _check_engine
+
+    with pd.option_context("compute.use_numexpr", use_numexpr):
+        result = _check_engine(None)
+        assert result == expected
+
+
+@td.skip_if_no("numexpr")
+def test_numexpr_option_incompatible_op():
+    # GH 32556
+    with pd.option_context("compute.use_numexpr", False):
+        df = DataFrame(
+            {"A": [True, False, True, False, None, None], "B": [1, 2, 3, 4, 5, 6]}
+        )
+        result = df.query("A.isnull()")
+        expected = DataFrame({"A": [None, None], "B": [5, 6]}, index=range(4, 6))
+        tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("numexpr")
+def test_invalid_parser():
+    msg = "Invalid parser 'asdf' passed"
+    with pytest.raises(KeyError, match=msg):
+        pd.eval("x + y", local_dict={"x": 1, "y": 2}, parser="asdf")
+
+
+_parsers: dict[str, type[BaseExprVisitor]] = {
+    "python": PythonExprVisitor,
+    "pytables": pytables.PyTablesExprVisitor,
+    "pandas": PandasExprVisitor,
+}
+
+
+@pytest.mark.parametrize("engine", ENGINES)
+@pytest.mark.parametrize("parser", _parsers)
+def test_disallowed_nodes(engine, parser):
+    VisitorClass = _parsers[parser]
+    inst = VisitorClass("x + 1", engine, parser)
+
+    for ops in VisitorClass.unsupported_nodes:
+        msg = "nodes are not implemented"
+        with pytest.raises(NotImplementedError, match=msg):
+            getattr(inst, ops)()
+
+
+def test_syntax_error_exprs(engine, parser):
+    e = "s +"
+    with pytest.raises(SyntaxError, match="invalid syntax"):
+        pd.eval(e, engine=engine, parser=parser)
+
+
+def test_name_error_exprs(engine, parser):
+    e = "s + t"
+    msg = "name 's' is not defined"
+    with pytest.raises(NameError, match=msg):
+        pd.eval(e, engine=engine, parser=parser)
+
+
+@pytest.mark.parametrize("express", ["a + @b", "@a + b", "@a + @b"])
+def test_invalid_local_variable_reference(engine, parser, express):
+    a, b = 1, 2  # noqa: F841
+
+    if parser != "pandas":
+        with pytest.raises(SyntaxError, match="The '@' prefix is only"):
+            pd.eval(express, engine=engine, parser=parser)
+    else:
+        with pytest.raises(SyntaxError, match="The '@' prefix is not"):
+            pd.eval(express, engine=engine, parser=parser)
+
+
+def test_numexpr_builtin_raises(engine, parser):
+    sin, dotted_line = 1, 2
+    if engine == "numexpr":
+        msg = "Variables in expression .+"
+        with pytest.raises(NumExprClobberingError, match=msg):
+            pd.eval("sin + dotted_line", engine=engine, parser=parser)
+    else:
+        res = pd.eval("sin + dotted_line", engine=engine, parser=parser)
+        assert res == sin + dotted_line
+
+
+def test_bad_resolver_raises(engine, parser):
+    cannot_resolve = 42, 3.0
+    with pytest.raises(TypeError, match="Resolver of type .+"):
+        pd.eval("1 + 2", resolvers=cannot_resolve, engine=engine, parser=parser)
+
+
+def test_empty_string_raises(engine, parser):
+    # GH 13139
+    with pytest.raises(ValueError, match="expr cannot be an empty string"):
+        pd.eval("", engine=engine, parser=parser)
+
+
+def test_more_than_one_expression_raises(engine, parser):
+    with pytest.raises(SyntaxError, match="only a single expression is allowed"):
+        pd.eval("1 + 1; 2 + 2", engine=engine, parser=parser)
+
+
+@pytest.mark.parametrize("cmp", ("and", "or"))
+@pytest.mark.parametrize("lhs", (int, float))
+@pytest.mark.parametrize("rhs", (int, float))
+def test_bool_ops_fails_on_scalars(lhs, cmp, rhs, engine, parser):
+    gen = {
+        int: lambda: np.random.default_rng(2).integers(10),
+        float: np.random.default_rng(2).standard_normal,
+    }
+
+    mid = gen[lhs]()  # noqa: F841
+    lhs = gen[lhs]()
+    rhs = gen[rhs]()
+
+    ex1 = f"lhs {cmp} mid {cmp} rhs"
+    ex2 = f"lhs {cmp} mid and mid {cmp} rhs"
+    ex3 = f"(lhs {cmp} mid) & (mid {cmp} rhs)"
+    for ex in (ex1, ex2, ex3):
+        msg = "cannot evaluate scalar only bool ops|'BoolOp' nodes are not"
+        with pytest.raises(NotImplementedError, match=msg):
+            pd.eval(ex, engine=engine, parser=parser)
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        "'x'",
+        "...",
+    ],
+)
+def test_equals_various(other):
+    df = DataFrame({"A": ["a", "b", "c"]}, dtype=object)
+    result = df.eval(f"A == {other}")
+    expected = Series([False, False, False], name="A")
+    tm.assert_series_equal(result, expected)
+
+
+def test_inf(engine, parser):
+    s = "inf + 1"
+    expected = np.inf
+    result = pd.eval(s, engine=engine, parser=parser)
+    assert result == expected
+
+
+@pytest.mark.parametrize("column", ["Temp(°C)", "Capacitance(μF)"])
+def test_query_token(engine, column):
+    # See: https://github.com/pandas-dev/pandas/pull/42826
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 2)), columns=[column, "b"]
+    )
+    expected = df[df[column] > 5]
+    query_string = f"`{column}` > 5"
+    result = df.query(query_string, engine=engine)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_negate_lt_eq_le(engine, parser):
+    df = DataFrame([[0, 10], [1, 20]], columns=["cat", "count"])
+    expected = df[~(df.cat > 0)]
+
+    result = df.query("~(cat > 0)", engine=engine, parser=parser)
+    tm.assert_frame_equal(result, expected)
+
+    if parser == "python":
+        msg = "'Not' nodes are not implemented"
+        with pytest.raises(NotImplementedError, match=msg):
+            df.query("not (cat > 0)", engine=engine, parser=parser)
+    else:
+        result = df.query("not (cat > 0)", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "column",
+    DEFAULT_GLOBALS.keys(),
+)
+def test_eval_no_support_column_name(request, column):
+    # GH 44603
+    if column in ["True", "False", "inf", "Inf"]:
+        request.applymarker(
+            pytest.mark.xfail(
+                raises=KeyError,
+                reason=f"GH 47859 DataFrame eval not supported with {column}",
+            )
+        )
+
+    df = DataFrame(
+        np.random.default_rng(2).integers(0, 100, size=(10, 2)),
+        columns=[column, "col1"],
+    )
+    expected = df[df[column] > 6]
+    result = df.query(f"{column}>6")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_set_inplace():
+    # https://github.com/pandas-dev/pandas/issues/47449
+    # Ensure we don't only update the DataFrame inplace, but also the actual
+    # column values, such that references to this column also get updated
+    df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+    result_view = df[:]
+    ser = df["A"]
+    df.eval("A = B + C", inplace=True)
+    expected = DataFrame({"A": [11, 13, 15], "B": [4, 5, 6], "C": [7, 8, 9]})
+    tm.assert_frame_equal(df, expected)
+    expected = Series([1, 2, 3], name="A")
+    tm.assert_series_equal(ser, expected)
+    tm.assert_series_equal(result_view["A"], expected)
+
+
+@pytest.mark.parametrize("value", [1, "True", [1, 2, 3], 5.0])
+def test_validate_bool_args(value):
+    msg = 'For argument "inplace" expected type bool, received type'
+    with pytest.raises(ValueError, match=msg):
+        pd.eval("2+2", inplace=value)
+
+
+@td.skip_if_no("numexpr")
+def test_eval_float_div_numexpr():
+    # GH 59736
+    result = pd.eval("1 / 2", engine="numexpr")
+    expected = 0.5
+    assert result == expected
diff --git a/pandas/pandas/tests/config/__init__.py b/pandas/pandas/tests/config/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/config/test_config.py b/pandas/pandas/tests/config/test_config.py
new file mode 100644
index 0000000000000..aaf6178866ecd
--- /dev/null
+++ b/pandas/pandas/tests/config/test_config.py
@@ -0,0 +1,437 @@
+import pytest
+
+from pandas._config import config as cf
+from pandas._config.config import OptionError
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestConfig:
+    @pytest.fixture(autouse=True)
+    def clean_config(self, monkeypatch):
+        with monkeypatch.context() as m:
+            m.setattr(cf, "_global_config", {})
+            m.setattr(cf, "options", cf.DictWrapper(cf._global_config))
+            m.setattr(cf, "_deprecated_options", {})
+            m.setattr(cf, "_registered_options", {})
+
+            # Our test fixture in conftest.py sets "chained_assignment"
+            # to "raise" only after all test methods have been setup.
+            # However, after this setup, there is no longer any
+            # "chained_assignment" option, so re-register it.
+            cf.register_option("chained_assignment", "raise")
+            yield
+
+    def test_api(self):
+        # the pandas object exposes the user API
+        assert hasattr(pd, "get_option")
+        assert hasattr(pd, "set_option")
+        assert hasattr(pd, "reset_option")
+        assert hasattr(pd, "describe_option")
+
+    def test_is_one_of_factory(self):
+        v = cf.is_one_of_factory([None, 12])
+
+        v(12)
+        v(None)
+        msg = r"Value must be one of None\|12"
+        with pytest.raises(ValueError, match=msg):
+            v(1.1)
+
+    def test_register_option(self):
+        cf.register_option("a", 1, "doc")
+
+        # can't register an already registered option
+        msg = "Option 'a' has already been registered"
+        with pytest.raises(OptionError, match=msg):
+            cf.register_option("a", 1, "doc")
+
+        # can't register an already registered option
+        msg = "Path prefix to option 'a' is already an option"
+        with pytest.raises(OptionError, match=msg):
+            cf.register_option("a.b.c.d1", 1, "doc")
+        with pytest.raises(OptionError, match=msg):
+            cf.register_option("a.b.c.d2", 1, "doc")
+
+        # no python keywords
+        msg = "for is a python keyword"
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("for", 0)
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("a.for.b", 0)
+        # must be valid identifier (ensure attribute access works)
+        msg = "oh my goddess! is not a valid identifier"
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("Oh my Goddess!", 0)
+
+        # we can register options several levels deep
+        # without predefining the intermediate steps
+        # and we can define differently named options
+        # in the same namespace
+        cf.register_option("k.b.c.d1", 1, "doc")
+        cf.register_option("k.b.c.d2", 1, "doc")
+
+    def test_describe_option(self):
+        cf.register_option("a", 1, "doc")
+        cf.register_option("b", 1, "doc2")
+        cf.deprecate_option("b")
+
+        cf.register_option("c.d.e1", 1, "doc3")
+        cf.register_option("c.d.e2", 1, "doc4")
+        cf.register_option("f", 1)
+        cf.register_option("g.h", 1)
+        cf.register_option("k", 2)
+        cf.deprecate_option("g.h", rkey="k")
+        cf.register_option("l", "foo")
+
+        # non-existent keys raise KeyError
+        msg = r"No such keys\(s\)"
+        with pytest.raises(OptionError, match=msg):
+            cf.describe_option("no.such.key")
+
+        # we can get the description for any key we registered
+        assert "doc" in cf.describe_option("a", _print_desc=False)
+        assert "doc2" in cf.describe_option("b", _print_desc=False)
+        assert "precated" in cf.describe_option("b", _print_desc=False)
+        assert "doc3" in cf.describe_option("c.d.e1", _print_desc=False)
+        assert "doc4" in cf.describe_option("c.d.e2", _print_desc=False)
+
+        # if no doc is specified we get a default message
+        # saying "description not available"
+        assert "available" in cf.describe_option("f", _print_desc=False)
+        assert "available" in cf.describe_option("g.h", _print_desc=False)
+        assert "precated" in cf.describe_option("g.h", _print_desc=False)
+        assert "k" in cf.describe_option("g.h", _print_desc=False)
+
+        # default is reported
+        assert "foo" in cf.describe_option("l", _print_desc=False)
+        # current value is reported
+        assert "bar" not in cf.describe_option("l", _print_desc=False)
+        cf.set_option("l", "bar")
+        assert "bar" in cf.describe_option("l", _print_desc=False)
+
+    def test_case_insensitive(self):
+        cf.register_option("KanBAN", 1, "doc")
+
+        assert "doc" in cf.describe_option("kanbaN", _print_desc=False)
+        assert cf.get_option("kanBaN") == 1
+        cf.set_option("KanBan", 2)
+        assert cf.get_option("kAnBaN") == 2
+
+        # gets of non-existent keys fail
+        msg = r"No such keys\(s\): 'no_such_option'"
+        with pytest.raises(OptionError, match=msg):
+            cf.get_option("no_such_option")
+
+        cf.deprecate_option("KanBan")
+        msg = "'kanban' is deprecated, please refrain from using it."
+        with pytest.raises(FutureWarning, match=msg):
+            cf.get_option("kAnBaN")
+
+    def test_get_option(self):
+        cf.register_option("a", 1, "doc")
+        cf.register_option("b.c", "hullo", "doc2")
+        cf.register_option("b.b", None, "doc2")
+
+        # gets of existing keys succeed
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+        assert cf.get_option("b.b") is None
+
+        # gets of non-existent keys fail
+        msg = r"No such keys\(s\): 'no_such_option'"
+        with pytest.raises(OptionError, match=msg):
+            cf.get_option("no_such_option")
+
+    def test_set_option(self):
+        cf.register_option("a", 1, "doc")
+        cf.register_option("b.c", "hullo", "doc2")
+        cf.register_option("b.b", None, "doc2")
+
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+        assert cf.get_option("b.b") is None
+
+        cf.set_option("a", 2)
+        cf.set_option("b.c", "wurld")
+        cf.set_option("b.b", 1.1)
+
+        assert cf.get_option("a") == 2
+        assert cf.get_option("b.c") == "wurld"
+        assert cf.get_option("b.b") == 1.1
+
+        msg = r"No such keys\(s\): 'no.such.key'"
+        with pytest.raises(OptionError, match=msg):
+            cf.set_option("no.such.key", None)
+
+    def test_set_option_empty_args(self):
+        msg = "Must provide an even number of non-keyword arguments"
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option()
+
+    def test_set_option_uneven_args(self):
+        msg = "Must provide an even number of non-keyword arguments"
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option("a.b", 2, "b.c")
+
+    def test_set_option_invalid_single_argument_type(self):
+        msg = "Must provide an even number of non-keyword arguments"
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option(2)
+
+    def test_set_option_multiple(self):
+        cf.register_option("a", 1, "doc")
+        cf.register_option("b.c", "hullo", "doc2")
+        cf.register_option("b.b", None, "doc2")
+
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+        assert cf.get_option("b.b") is None
+
+        cf.set_option("a", "2", "b.c", None, "b.b", 10.0)
+
+        assert cf.get_option("a") == "2"
+        assert cf.get_option("b.c") is None
+        assert cf.get_option("b.b") == 10.0
+
+    def test_validation(self):
+        cf.register_option("a", 1, "doc", validator=cf.is_int)
+        cf.register_option("d", 1, "doc", validator=cf.is_nonnegative_int)
+        cf.register_option("b.c", "hullo", "doc2", validator=cf.is_text)
+
+        msg = "Value must have type '<class 'int'>'"
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("a.b.c.d2", "NO", "doc", validator=cf.is_int)
+
+        cf.set_option("a", 2)  # int is_int
+        cf.set_option("b.c", "wurld")  # str is_str
+        cf.set_option("d", 2)
+        cf.set_option("d", None)  # non-negative int can be None
+
+        # None not is_int
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option("a", None)
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option("a", "ab")
+
+        msg = "Value must be a nonnegative integer or None"
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("a.b.c.d3", "NO", "doc", validator=cf.is_nonnegative_int)
+        with pytest.raises(ValueError, match=msg):
+            cf.register_option("a.b.c.d3", -2, "doc", validator=cf.is_nonnegative_int)
+
+        msg = r"Value must be an instance of <class 'str'>\|<class 'bytes'>"
+        with pytest.raises(ValueError, match=msg):
+            cf.set_option("b.c", 1)
+
+        validator = cf.is_one_of_factory([None, cf.is_callable])
+        cf.register_option("b", lambda: None, "doc", validator=validator)
+        cf.set_option("b", "%.1f".format)  # Formatter is callable
+        cf.set_option("b", None)  # Formatter is none (default)
+        with pytest.raises(ValueError, match="Value must be a callable"):
+            cf.set_option("b", "%.1f")
+
+    def test_reset_option(self):
+        cf.register_option("a", 1, "doc", validator=cf.is_int)
+        cf.register_option("b.c", "hullo", "doc2", validator=cf.is_str)
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+
+        cf.set_option("a", 2)
+        cf.set_option("b.c", "wurld")
+        assert cf.get_option("a") == 2
+        assert cf.get_option("b.c") == "wurld"
+
+        cf.reset_option("a")
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "wurld"
+        cf.reset_option("b.c")
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+
+    def test_reset_option_all(self):
+        cf.register_option("a", 1, "doc", validator=cf.is_int)
+        cf.register_option("b.c", "hullo", "doc2", validator=cf.is_str)
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+
+        cf.set_option("a", 2)
+        cf.set_option("b.c", "wurld")
+        assert cf.get_option("a") == 2
+        assert cf.get_option("b.c") == "wurld"
+
+        cf.reset_option("all")
+        assert cf.get_option("a") == 1
+        assert cf.get_option("b.c") == "hullo"
+
+    def test_deprecate_option(self):
+        # we can deprecate non-existent options
+        cf.deprecate_option("foo")
+
+        with tm.assert_produces_warning(FutureWarning, match="deprecated"):
+            with pytest.raises(KeyError, match="No such keys.s.: 'foo'"):
+                cf.get_option("foo")
+
+        cf.register_option("a", 1, "doc", validator=cf.is_int)
+        cf.register_option("b.c", "hullo", "doc2")
+        cf.register_option("foo", "hullo", "doc2")
+
+        cf.deprecate_option("a", removal_ver="nifty_ver")
+        with tm.assert_produces_warning(FutureWarning, match="eprecated.*nifty_ver"):
+            cf.get_option("a")
+
+            msg = "Option 'a' has already been defined as deprecated"
+            with pytest.raises(OptionError, match=msg):
+                cf.deprecate_option("a")
+
+        cf.deprecate_option("b.c", "zounds!")
+        with tm.assert_produces_warning(FutureWarning, match="zounds!"):
+            cf.get_option("b.c")
+
+        # test rerouting keys
+        cf.register_option("d.a", "foo", "doc2")
+        cf.register_option("d.dep", "bar", "doc2")
+        assert cf.get_option("d.a") == "foo"
+        assert cf.get_option("d.dep") == "bar"
+
+        cf.deprecate_option("d.dep", rkey="d.a")  # reroute d.dep to d.a
+        with tm.assert_produces_warning(FutureWarning, match="eprecated"):
+            assert cf.get_option("d.dep") == "foo"
+
+        with tm.assert_produces_warning(FutureWarning, match="eprecated"):
+            cf.set_option("d.dep", "baz")  # should overwrite "d.a"
+
+        with tm.assert_produces_warning(FutureWarning, match="eprecated"):
+            assert cf.get_option("d.dep") == "baz"
+
+    def test_config_prefix(self):
+        with cf.config_prefix("base"):
+            cf.register_option("a", 1, "doc1")
+            cf.register_option("b", 2, "doc2")
+            assert cf.get_option("a") == 1
+            assert cf.get_option("b") == 2
+
+            cf.set_option("a", 3)
+            cf.set_option("b", 4)
+            assert cf.get_option("a") == 3
+            assert cf.get_option("b") == 4
+
+        assert cf.get_option("base.a") == 3
+        assert cf.get_option("base.b") == 4
+        assert "doc1" in cf.describe_option("base.a", _print_desc=False)
+        assert "doc2" in cf.describe_option("base.b", _print_desc=False)
+
+        cf.reset_option("base.a")
+        cf.reset_option("base.b")
+
+        with cf.config_prefix("base"):
+            assert cf.get_option("a") == 1
+            assert cf.get_option("b") == 2
+
+    def test_callback(self):
+        k = [None]
+        v = [None]
+
+        def callback(key):
+            k.append(key)
+            v.append(cf.get_option(key))
+
+        cf.register_option("d.a", "foo", cb=callback)
+        cf.register_option("d.b", "foo", cb=callback)
+
+        del k[-1], v[-1]
+        cf.set_option("d.a", "fooz")
+        assert k[-1] == "d.a"
+        assert v[-1] == "fooz"
+
+        del k[-1], v[-1]
+        cf.set_option("d.b", "boo")
+        assert k[-1] == "d.b"
+        assert v[-1] == "boo"
+
+        del k[-1], v[-1]
+        cf.reset_option("d.b")
+        assert k[-1] == "d.b"
+
+    def test_set_ContextManager(self):
+        def eq(val):
+            assert cf.get_option("a") == val
+
+        cf.register_option("a", 0)
+        eq(0)
+        with cf.option_context("a", 15):
+            eq(15)
+            with cf.option_context("a", 25):
+                eq(25)
+            eq(15)
+        eq(0)
+
+        cf.set_option("a", 17)
+        eq(17)
+
+        # Test that option_context can be used as a decorator too (#34253).
+        @cf.option_context("a", 123)
+        def f():
+            eq(123)
+
+        f()
+
+    def test_attribute_access(self):
+        holder = []
+
+        def f3(key):
+            holder.append(True)
+
+        cf.register_option("a", 0)
+        cf.register_option("c", 0, cb=f3)
+        options = cf.options
+
+        assert options.a == 0
+        with cf.option_context("a", 15):
+            assert options.a == 15
+
+        options.a = 500
+        assert cf.get_option("a") == 500
+
+        cf.reset_option("a")
+        assert options.a == cf.get_option("a")
+
+        msg = "You can only set the value of existing options"
+        with pytest.raises(OptionError, match=msg):
+            options.b = 1
+        with pytest.raises(OptionError, match=msg):
+            options.display = 1
+
+        # make sure callback kicks when using this form of setting
+        options.c = 1
+        assert len(holder) == 1
+
+    def test_option_context_scope(self):
+        # Ensure that creating a context does not affect the existing
+        # environment as it is supposed to be used with the `with` statement.
+        # See https://github.com/pandas-dev/pandas/issues/8514
+
+        original_value = 60
+        context_value = 10
+        option_name = "a"
+
+        cf.register_option(option_name, original_value)
+
+        # Ensure creating contexts didn't affect the current context.
+        ctx = cf.option_context(option_name, context_value)
+        assert cf.get_option(option_name) == original_value
+
+        # Ensure the correct value is available inside the context.
+        with ctx:
+            assert cf.get_option(option_name) == context_value
+
+        # Ensure the current context is reset
+        assert cf.get_option(option_name) == original_value
+
+    def test_dictwrapper_getattr(self):
+        options = cf.options
+        # GH 19789
+        with pytest.raises(OptionError, match="No such option"):
+            options.bananas
+        assert not hasattr(options, "bananas")
diff --git a/pandas/pandas/tests/config/test_localization.py b/pandas/pandas/tests/config/test_localization.py
new file mode 100644
index 0000000000000..b9a0a44bf8c89
--- /dev/null
+++ b/pandas/pandas/tests/config/test_localization.py
@@ -0,0 +1,155 @@
+import codecs
+import locale
+import os
+
+import pytest
+
+from pandas._config.localization import (
+    can_set_locale,
+    get_locales,
+    set_locale,
+)
+
+from pandas.compat import ISMUSL
+
+import pandas as pd
+
+_all_locales = get_locales()
+
+# Don't run any of these tests if we have no locales.
+pytestmark = pytest.mark.skipif(not _all_locales, reason="Need locales")
+
+_skip_if_only_one_locale = pytest.mark.skipif(
+    len(_all_locales) <= 1, reason="Need multiple locales for meaningful test"
+)
+
+
+def _get_current_locale(lc_var: int = locale.LC_ALL) -> str:
+    # getlocale is not always compliant with setlocale, use setlocale. GH#46595
+    return locale.setlocale(lc_var)
+
+
+@pytest.mark.parametrize("lc_var", (locale.LC_ALL, locale.LC_CTYPE, locale.LC_TIME))
+def test_can_set_current_locale(lc_var):
+    # Can set the current locale
+    before_locale = _get_current_locale(lc_var)
+    assert can_set_locale(before_locale, lc_var=lc_var)
+    after_locale = _get_current_locale(lc_var)
+    assert before_locale == after_locale
+
+
+@pytest.mark.parametrize("lc_var", (locale.LC_ALL, locale.LC_CTYPE, locale.LC_TIME))
+def test_can_set_locale_valid_set(lc_var):
+    # Can set the default locale.
+    before_locale = _get_current_locale(lc_var)
+    assert can_set_locale("", lc_var=lc_var)
+    after_locale = _get_current_locale(lc_var)
+    assert before_locale == after_locale
+
+
+@pytest.mark.parametrize(
+    "lc_var",
+    (
+        locale.LC_ALL,
+        locale.LC_CTYPE,
+        pytest.param(
+            locale.LC_TIME,
+            marks=pytest.mark.skipif(
+                ISMUSL, reason="MUSL allows setting invalid LC_TIME."
+            ),
+        ),
+    ),
+)
+def test_can_set_locale_invalid_set(lc_var):
+    # Cannot set an invalid locale.
+    before_locale = _get_current_locale(lc_var)
+    assert not can_set_locale("non-existent_locale", lc_var=lc_var)
+    after_locale = _get_current_locale(lc_var)
+    assert before_locale == after_locale
+
+
+@pytest.mark.parametrize(
+    "lang,enc",
+    [
+        ("it_CH", "UTF-8"),
+        ("en_US", "ascii"),
+        ("zh_CN", "GB2312"),
+        ("it_IT", "ISO-8859-1"),
+    ],
+)
+@pytest.mark.parametrize("lc_var", (locale.LC_ALL, locale.LC_CTYPE, locale.LC_TIME))
+def test_can_set_locale_no_leak(lang, enc, lc_var):
+    # Test that can_set_locale does not leak even when returning False. See GH#46595
+    before_locale = _get_current_locale(lc_var)
+    can_set_locale((lang, enc), locale.LC_ALL)
+    after_locale = _get_current_locale(lc_var)
+    assert before_locale == after_locale
+
+
+def test_can_set_locale_invalid_get(monkeypatch):
+    # see GH#22129
+    # In some cases, an invalid locale can be set,
+    #  but a subsequent getlocale() raises a ValueError.
+
+    def mock_get_locale():
+        raise ValueError
+
+    with monkeypatch.context() as m:
+        m.setattr(locale, "getlocale", mock_get_locale)
+        assert not can_set_locale("")
+
+
+def test_get_locales_at_least_one():
+    # see GH#9744
+    assert len(_all_locales) > 0
+
+
+@_skip_if_only_one_locale
+def test_get_locales_prefix():
+    first_locale = _all_locales[0]
+    assert len(get_locales(prefix=first_locale[:2])) > 0
+
+
+@_skip_if_only_one_locale
+@pytest.mark.parametrize(
+    "lang,enc",
+    [
+        ("it_CH", "UTF-8"),
+        ("en_US", "ascii"),
+        ("zh_CN", "GB2312"),
+        ("it_IT", "ISO-8859-1"),
+    ],
+)
+def test_set_locale(lang, enc):
+    before_locale = _get_current_locale()
+
+    enc = codecs.lookup(enc).name
+    new_locale = lang, enc
+
+    if not can_set_locale(new_locale):
+        msg = "unsupported locale setting"
+
+        with pytest.raises(locale.Error, match=msg):
+            with set_locale(new_locale):
+                pass
+    else:
+        with set_locale(new_locale) as normalized_locale:
+            new_lang, new_enc = normalized_locale.split(".")
+            new_enc = codecs.lookup(enc).name
+
+            normalized_locale = new_lang, new_enc
+            assert normalized_locale == new_locale
+
+    # Once we exit the "with" statement, locale should be back to what it was.
+    after_locale = _get_current_locale()
+    assert before_locale == after_locale
+
+
+def test_encoding_detected():
+    system_locale = os.environ.get("LC_ALL")
+    system_encoding = system_locale.split(".")[-1] if system_locale else "utf-8"
+
+    assert (
+        codecs.lookup(pd.options.display.encoding).name
+        == codecs.lookup(system_encoding).name
+    )
diff --git a/pandas/pandas/tests/construction/__init__.py b/pandas/pandas/tests/construction/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/construction/test_extract_array.py b/pandas/pandas/tests/construction/test_extract_array.py
new file mode 100644
index 0000000000000..4dd3eda8c995c
--- /dev/null
+++ b/pandas/pandas/tests/construction/test_extract_array.py
@@ -0,0 +1,18 @@
+from pandas import Index
+import pandas._testing as tm
+from pandas.core.construction import extract_array
+
+
+def test_extract_array_rangeindex():
+    ri = Index(range(5))
+
+    expected = ri._values
+    res = extract_array(ri, extract_numpy=True, extract_range=True)
+    tm.assert_numpy_array_equal(res, expected)
+    res = extract_array(ri, extract_numpy=False, extract_range=True)
+    tm.assert_numpy_array_equal(res, expected)
+
+    res = extract_array(ri, extract_numpy=True, extract_range=False)
+    tm.assert_index_equal(res, ri)
+    res = extract_array(ri, extract_numpy=False, extract_range=False)
+    tm.assert_index_equal(res, ri)
diff --git a/pandas/pandas/tests/copy_view/__init__.py b/pandas/pandas/tests/copy_view/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/copy_view/index/__init__.py b/pandas/pandas/tests/copy_view/index/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/copy_view/index/test_datetimeindex.py b/pandas/pandas/tests/copy_view/index/test_datetimeindex.py
new file mode 100644
index 0000000000000..6194ea8b122c9
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/index/test_datetimeindex.py
@@ -0,0 +1,56 @@
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Setting a value on a view:FutureWarning"
+)
+
+
+@pytest.mark.parametrize("box", [lambda x: x, DatetimeIndex])
+def test_datetimeindex(box):
+    dt = date_range("2019-12-31", periods=3, freq="D")
+    ser = Series(dt)
+    idx = box(DatetimeIndex(ser))
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = Timestamp("2020-12-31")
+    tm.assert_index_equal(idx, expected)
+
+
+def test_datetimeindex_tz_convert():
+    dt = date_range("2019-12-31", periods=3, freq="D", tz="Europe/Berlin")
+    ser = Series(dt)
+    idx = DatetimeIndex(ser).tz_convert("US/Eastern")
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = Timestamp("2020-12-31", tz="Europe/Berlin")
+    tm.assert_index_equal(idx, expected)
+
+
+def test_datetimeindex_tz_localize():
+    dt = date_range("2019-12-31", periods=3, freq="D")
+    ser = Series(dt)
+    idx = DatetimeIndex(ser).tz_localize("Europe/Berlin")
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = Timestamp("2020-12-31")
+    tm.assert_index_equal(idx, expected)
+
+
+def test_datetimeindex_isocalendar():
+    dt = date_range("2019-12-31", periods=3, freq="D")
+    ser = Series(dt)
+    df = DatetimeIndex(ser).isocalendar()
+    expected = df.index.copy(deep=True)
+    ser.iloc[0] = Timestamp("2020-12-31")
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_index_values():
+    idx = date_range("2019-12-31", periods=3, freq="D")
+    result = idx.values
+    assert result.flags.writeable is False
diff --git a/pandas/pandas/tests/copy_view/index/test_index.py b/pandas/pandas/tests/copy_view/index/test_index.py
new file mode 100644
index 0000000000000..e51f5658cf437
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/index/test_index.py
@@ -0,0 +1,152 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def index_view(index_data):
+    df = DataFrame({"a": index_data, "b": 1.5})
+    view = df[:]
+    df = df.set_index("a", drop=True)
+    idx = df.index
+    # df = None
+    return idx, view
+
+
+def test_set_index_update_column():
+    df = DataFrame({"a": [1, 2], "b": 1})
+    df = df.set_index("a", drop=False)
+    expected = df.index.copy(deep=True)
+    df.iloc[0, 0] = 100
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_set_index_drop_update_column():
+    df = DataFrame({"a": [1, 2], "b": 1.5})
+    view = df[:]
+    df = df.set_index("a", drop=True)
+    expected = df.index.copy(deep=True)
+    view.iloc[0, 0] = 100
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_set_index_series():
+    df = DataFrame({"a": [1, 2], "b": 1.5})
+    ser = Series([10, 11])
+    df = df.set_index(ser)
+    expected = df.index.copy(deep=True)
+    ser.iloc[0] = 100
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_assign_index_as_series():
+    df = DataFrame({"a": [1, 2], "b": 1.5})
+    ser = Series([10, 11])
+    df.index = ser
+    expected = df.index.copy(deep=True)
+    ser.iloc[0] = 100
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_assign_index_as_index():
+    df = DataFrame({"a": [1, 2], "b": 1.5})
+    ser = Series([10, 11])
+    rhs_index = Index(ser)
+    df.index = rhs_index
+    rhs_index = None  # overwrite to clear reference
+    expected = df.index.copy(deep=True)
+    ser.iloc[0] = 100
+    tm.assert_index_equal(df.index, expected)
+
+
+def test_index_from_series():
+    ser = Series([1, 2])
+    idx = Index(ser)
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = 100
+    tm.assert_index_equal(idx, expected)
+
+
+def test_index_from_series_copy():
+    ser = Series([1, 2])
+    idx = Index(ser, copy=True)  # noqa: F841
+    arr = get_array(ser)
+    ser.iloc[0] = 100
+    assert np.shares_memory(get_array(ser), arr)
+
+
+def test_index_from_index():
+    ser = Series([1, 2])
+    idx = Index(ser)
+    idx = Index(idx)
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = 100
+    tm.assert_index_equal(idx, expected)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x: x._shallow_copy(x._values),
+        lambda x: x.view(),
+        lambda x: x.take([0, 1]),
+        lambda x: x.repeat([1, 1]),
+        lambda x: x[slice(0, 2)],
+        lambda x: x[[0, 1]],
+        lambda x: x._getitem_slice(slice(0, 2)),
+        lambda x: x.delete([]),
+        lambda x: x.rename("b"),
+        lambda x: x.astype("Int64", copy=False),
+    ],
+    ids=[
+        "_shallow_copy",
+        "view",
+        "take",
+        "repeat",
+        "getitem_slice",
+        "getitem_list",
+        "_getitem_slice",
+        "delete",
+        "rename",
+        "astype",
+    ],
+)
+def test_index_ops(func, request):
+    idx, view_ = index_view([1, 2])
+    expected = idx.copy(deep=True)
+    if "astype" in request.node.callspec.id:
+        expected = expected.astype("Int64")
+    idx = func(idx)
+    view_.iloc[0, 0] = 100
+    tm.assert_index_equal(idx, expected, check_names=False)
+
+
+def test_infer_objects():
+    idx, view_ = index_view(["a", "b"])
+    expected = idx.copy(deep=True)
+    idx = idx.infer_objects(copy=False)
+    view_.iloc[0, 0] = "aaaa"
+    tm.assert_index_equal(idx, expected, check_names=False)
+
+
+def test_index_to_frame():
+    idx = Index([1, 2, 3], name="a")
+    expected = idx.copy(deep=True)
+    df = idx.to_frame()
+    assert np.shares_memory(get_array(df, "a"), idx._values)
+    assert not df._mgr._has_no_reference(0)
+
+    df.iloc[0, 0] = 100
+    tm.assert_index_equal(idx, expected)
+
+
+def test_index_values():
+    idx = Index([1, 2, 3])
+    result = idx.values
+    assert result.flags.writeable is False
diff --git a/pandas/pandas/tests/copy_view/index/test_periodindex.py b/pandas/pandas/tests/copy_view/index/test_periodindex.py
new file mode 100644
index 0000000000000..2887b191038d2
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/index/test_periodindex.py
@@ -0,0 +1,23 @@
+import pytest
+
+from pandas import (
+    Period,
+    PeriodIndex,
+    Series,
+    period_range,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Setting a value on a view:FutureWarning"
+)
+
+
+@pytest.mark.parametrize("box", [lambda x: x, PeriodIndex])
+def test_periodindex(box):
+    dt = period_range("2019-12-31", periods=3, freq="D")
+    ser = Series(dt)
+    idx = box(PeriodIndex(ser))
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = Period("2020-12-31")
+    tm.assert_index_equal(idx, expected)
diff --git a/pandas/pandas/tests/copy_view/index/test_timedeltaindex.py b/pandas/pandas/tests/copy_view/index/test_timedeltaindex.py
new file mode 100644
index 0000000000000..6984df86b00e3
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/index/test_timedeltaindex.py
@@ -0,0 +1,29 @@
+import pytest
+
+from pandas import (
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Setting a value on a view:FutureWarning"
+)
+
+
+@pytest.mark.parametrize(
+    "cons",
+    [
+        lambda x: TimedeltaIndex(x),
+        lambda x: TimedeltaIndex(TimedeltaIndex(x)),
+    ],
+)
+def test_timedeltaindex(cons):
+    dt = timedelta_range("1 day", periods=3)
+    ser = Series(dt)
+    idx = cons(ser)
+    expected = idx.copy(deep=True)
+    ser.iloc[0] = Timedelta("5 days")
+    tm.assert_index_equal(idx, expected)
diff --git a/pandas/pandas/tests/copy_view/test_array.py b/pandas/pandas/tests/copy_view/test_array.py
new file mode 100644
index 0000000000000..2b3ef9201d918
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_array.py
@@ -0,0 +1,170 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gt2
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+# -----------------------------------------------------------------------------
+# Copy/view behaviour for accessing underlying array of Series/DataFrame
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda ser: ser.values,
+        lambda ser: np.asarray(ser),
+        lambda ser: np.array(ser, copy=False),
+    ],
+    ids=["values", "asarray", "array"],
+)
+def test_series_values(method):
+    ser = Series([1, 2, 3], name="name")
+    ser_orig = ser.copy()
+
+    arr = method(ser)
+
+    # .values still gives a view but is read-only
+    assert np.shares_memory(arr, get_array(ser, "name"))
+    assert arr.flags.writeable is False
+
+    # mutating series through arr therefore doesn't work
+    with pytest.raises(ValueError, match="read-only"):
+        arr[0] = 0
+    tm.assert_series_equal(ser, ser_orig)
+
+    # mutating the series itself still works
+    ser.iloc[0] = 0
+    assert ser.values[0] == 0
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df: df.values,
+        lambda df: np.asarray(df),
+        lambda ser: np.array(ser, copy=False),
+    ],
+    ids=["values", "asarray", "array"],
+)
+def test_dataframe_values(method):
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df_orig = df.copy()
+
+    arr = method(df)
+
+    # .values still gives a view but is read-only
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is False
+
+    # mutating series through arr therefore doesn't work
+    with pytest.raises(ValueError, match="read-only"):
+        arr[0, 0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+    # mutating the series itself still works
+    df.iloc[0, 0] = 0
+    assert df.values[0, 0] == 0
+
+
+def test_series_to_numpy():
+    ser = Series([1, 2, 3], name="name")
+    ser_orig = ser.copy()
+
+    # default: copy=False, no dtype or NAs
+    arr = ser.to_numpy()
+    # to_numpy still gives a view but is read-only
+    assert np.shares_memory(arr, get_array(ser, "name"))
+    assert arr.flags.writeable is False
+
+    # mutating series through arr therefore doesn't work
+    with pytest.raises(ValueError, match="read-only"):
+        arr[0] = 0
+    tm.assert_series_equal(ser, ser_orig)
+
+    # mutating the series itself still works
+    ser.iloc[0] = 0
+    assert ser.values[0] == 0
+
+    # specify copy=True gives a writeable array
+    ser = Series([1, 2, 3], name="name")
+    arr = ser.to_numpy(copy=True)
+    assert not np.shares_memory(arr, get_array(ser, "name"))
+    assert arr.flags.writeable is True
+
+    # specifying a dtype that already causes a copy also gives a writeable array
+    ser = Series([1, 2, 3], name="name")
+    arr = ser.to_numpy(dtype="float64")
+    assert not np.shares_memory(arr, get_array(ser, "name"))
+    assert arr.flags.writeable is True
+
+
+def test_series_array_ea_dtypes():
+    ser = Series([1, 2, 3], dtype="Int64")
+    arr = np.asarray(ser, dtype="int64")
+    assert np.shares_memory(arr, get_array(ser))
+    assert arr.flags.writeable is False
+
+    arr = np.asarray(ser)
+    assert np.shares_memory(arr, get_array(ser))
+    assert arr.flags.writeable is False
+
+
+def test_dataframe_array_ea_dtypes():
+    df = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    arr = np.asarray(df, dtype="int64")
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is False
+
+    arr = np.asarray(df)
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is False
+
+
+def test_dataframe_array_string_dtype():
+    df = DataFrame({"a": ["a", "b"]}, dtype="string")
+    arr = np.asarray(df)
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is False
+
+
+def test_dataframe_multiple_numpy_dtypes():
+    df = DataFrame({"a": [1, 2, 3], "b": 1.5})
+    arr = np.asarray(df)
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is True
+
+    if np_version_gt2:
+        # copy=False semantics are only supported in NumPy>=2.
+
+        with pytest.raises(ValueError, match="Unable to avoid copy while creating"):
+            arr = np.array(df, copy=False)
+
+    arr = np.array(df, copy=True)
+    assert arr.flags.writeable is True
+
+
+def test_dataframe_single_block_copy_true():
+    # the copy=False/None cases are tested above in test_dataframe_values
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    arr = np.array(df, copy=True)
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    assert arr.flags.writeable is True
+
+
+def test_values_is_ea():
+    df = DataFrame({"a": date_range("2012-01-01", periods=3)})
+    arr = np.asarray(df)
+    assert arr.flags.writeable is False
+
+
+def test_empty_dataframe():
+    df = DataFrame()
+    arr = np.asarray(df)
+    assert arr.flags.writeable is True
diff --git a/pandas/pandas/tests/copy_view/test_astype.py b/pandas/pandas/tests/copy_view/test_astype.py
new file mode 100644
index 0000000000000..91f5badeb9728
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_astype.py
@@ -0,0 +1,238 @@
+import pickle
+
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+from pandas.compat.pyarrow import pa_version_under12p0
+
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_astype_single_dtype():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": 1.5})
+    df_orig = df.copy()
+    df2 = df.astype("float64")
+
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column/block
+    df2.iloc[0, 2] = 5.5
+    assert not np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+    # mutating parent also doesn't update result
+    df2 = df.astype("float64")
+    df.iloc[0, 2] = 5.5
+    tm.assert_frame_equal(df2, df_orig.astype("float64"))
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+@pytest.mark.parametrize("new_dtype", ["int64", "Int64", "int64[pyarrow]"])
+def test_astype_avoids_copy(dtype, new_dtype):
+    if new_dtype == "int64[pyarrow]":
+        pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [1, 2, 3]}, dtype=dtype)
+    df_orig = df.copy()
+    df2 = df.astype(new_dtype)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column/block
+    df2.iloc[0, 0] = 10
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+    # mutating parent also doesn't update result
+    df2 = df.astype(new_dtype)
+    df.iloc[0, 0] = 100
+    tm.assert_frame_equal(df2, df_orig.astype(new_dtype))
+
+
+@pytest.mark.parametrize("dtype", ["float64", "int32", "Int32", "int32[pyarrow]"])
+def test_astype_different_target_dtype(dtype):
+    if dtype == "int32[pyarrow]":
+        pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [1, 2, 3]})
+    df_orig = df.copy()
+    df2 = df.astype(dtype)
+
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert df2._mgr._has_no_reference(0)
+
+    df2.iloc[0, 0] = 5
+    tm.assert_frame_equal(df, df_orig)
+
+    # mutating parent also doesn't update result
+    df2 = df.astype(dtype)
+    df.iloc[0, 0] = 100
+    tm.assert_frame_equal(df2, df_orig.astype(dtype))
+
+
+def test_astype_numpy_to_ea():
+    ser = Series([1, 2, 3])
+    result = ser.astype("Int64")
+    assert np.shares_memory(get_array(ser), get_array(result))
+
+
+@pytest.mark.parametrize(
+    "dtype, new_dtype", [("object", "string"), ("string", "object")]
+)
+def test_astype_string_and_object(dtype, new_dtype):
+    df = DataFrame({"a": ["a", "b", "c"]}, dtype=dtype)
+    df_orig = df.copy()
+    df2 = df.astype(new_dtype)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = "x"
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype, new_dtype", [("object", "string"), ("string", "object")]
+)
+def test_astype_string_and_object_update_original(dtype, new_dtype):
+    df = DataFrame({"a": ["a", "b", "c"]}, dtype=dtype)
+    df2 = df.astype(new_dtype)
+    df_orig = df2.copy()
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df.iloc[0, 0] = "x"
+    tm.assert_frame_equal(df2, df_orig)
+
+
+def test_astype_str_copy_on_pickle_roundrip():
+    # TODO(infer_string) this test can be removed after 3.0 (once str is the default)
+    # https://github.com/pandas-dev/pandas/issues/54654
+    # ensure_string_array may alter array inplace
+    base = Series(np.array([(1, 2), None, 1], dtype="object"))
+    base_copy = pickle.loads(pickle.dumps(base))
+    base_copy.astype(str)
+    tm.assert_series_equal(base, base_copy)
+
+
+def test_astype_string_copy_on_pickle_roundrip(any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/54654
+    # ensure_string_array may alter array inplace
+    base = Series(np.array([(1, 2), None, 1], dtype="object"))
+    base_copy = pickle.loads(pickle.dumps(base))
+    base_copy.astype(any_string_dtype)
+    tm.assert_series_equal(base, base_copy)
+
+
+def test_astype_string_read_only_on_pickle_roundrip(any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/54654
+    # ensure_string_array may alter read-only array inplace
+    base = Series(np.array([(1, 2), None, 1], dtype="object"))
+    base_copy = pickle.loads(pickle.dumps(base))
+    base_copy._values.flags.writeable = False
+    base_copy.astype(any_string_dtype)
+    tm.assert_series_equal(base, base_copy)
+
+
+def test_astype_dict_dtypes():
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": Series([1.5, 1.5, 1.5], dtype="float64")}
+    )
+    df_orig = df.copy()
+    df2 = df.astype({"a": "float64", "c": "float64"})
+
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column/block
+    df2.iloc[0, 2] = 5.5
+    assert not np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+
+    df2.iloc[0, 1] = 10
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_astype_different_datetime_resos():
+    df = DataFrame({"a": date_range("2019-12-31", periods=2, freq="D")})
+    result = df.astype("datetime64[ms]")
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+    assert result._mgr._has_no_reference(0)
+
+
+def test_astype_different_timezones():
+    df = DataFrame(
+        {"a": date_range("2019-12-31", periods=5, freq="D", tz="US/Pacific")}
+    )
+    result = df.astype("datetime64[ns, Europe/Berlin]")
+    assert not result._mgr._has_no_reference(0)
+    assert np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+
+
+def test_astype_different_timezones_different_reso():
+    df = DataFrame(
+        {"a": date_range("2019-12-31", periods=5, freq="D", tz="US/Pacific")}
+    )
+    result = df.astype("datetime64[ms, Europe/Berlin]")
+    assert result._mgr._has_no_reference(0)
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+
+
+def test_astype_arrow_timestamp():
+    pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {
+            "a": [
+                Timestamp("2020-01-01 01:01:01.000001"),
+                Timestamp("2020-01-01 01:01:01.000001"),
+            ]
+        },
+        dtype="M8[ns]",
+    )
+    result = df.astype("timestamp[ns][pyarrow]")
+    assert not result._mgr._has_no_reference(0)
+    if pa_version_under12p0:
+        assert not np.shares_memory(
+            get_array(df, "a"), get_array(result, "a")._pa_array
+        )
+    else:
+        assert np.shares_memory(get_array(df, "a"), get_array(result, "a")._pa_array)
+
+
+def test_convert_dtypes_infer_objects():
+    ser = Series(["a", "b", "c"])
+    ser_orig = ser.copy()
+    result = ser.convert_dtypes(
+        convert_integer=False,
+        convert_boolean=False,
+        convert_floating=False,
+        convert_string=False,
+    )
+
+    assert tm.shares_memory(get_array(ser), get_array(result))
+    result.iloc[0] = "x"
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_convert_dtypes(using_infer_string):
+    df = DataFrame({"a": ["a", "b"], "b": [1, 2], "c": [1.5, 2.5], "d": [True, False]})
+    df_orig = df.copy()
+    df2 = df.convert_dtypes()
+
+    if using_infer_string and HAS_PYARROW:
+        # TODO the default nullable string dtype still uses python storage
+        # this should be changed to pyarrow if installed
+        assert not tm.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    else:
+        assert tm.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert tm.shares_memory(get_array(df2, "d"), get_array(df, "d"))
+    assert tm.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert tm.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    df2.iloc[0, 0] = "x"
+    df2.iloc[0, 1] = 10
+    tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_chained_assignment_deprecation.py b/pandas/pandas/tests/copy_view/test_chained_assignment_deprecation.py
new file mode 100644
index 0000000000000..4aef69a6fde98
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_chained_assignment_deprecation.py
@@ -0,0 +1,33 @@
+import numpy as np
+import pytest
+
+from pandas.errors import ChainedAssignmentError
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "indexer", [0, [0, 1], slice(0, 2), np.array([True, False, True])]
+)
+def test_series_setitem(indexer):
+    # ensure we only get a single warning for those typical cases of chained
+    # assignment
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+
+    # using custom check instead of tm.assert_produces_warning because that doesn't
+    # fail if multiple warnings are raised
+    with pytest.warns() as record:  # noqa: TID251
+        df["a"][indexer] = 0
+    assert len(record) == 1
+    assert record[0].category == ChainedAssignmentError
+
+
+@pytest.mark.parametrize(
+    "indexer", ["a", ["a", "b"], slice(0, 2), np.array([True, False, True])]
+)
+def test_frame_setitem(indexer):
+    df = DataFrame({"a": [1, 2, 3, 4, 5], "b": 1})
+
+    with tm.raises_chained_assignment_error():
+        df[0:3][indexer] = 10
diff --git a/pandas/pandas/tests/copy_view/test_clip.py b/pandas/pandas/tests/copy_view/test_clip.py
new file mode 100644
index 0000000000000..56df33db6d416
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_clip.py
@@ -0,0 +1,72 @@
+import numpy as np
+
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_clip_inplace_reference():
+    df = DataFrame({"a": [1.5, 2, 3]})
+    df_copy = df.copy()
+    arr_a = get_array(df, "a")
+    view = df[:]
+    df.clip(lower=2, inplace=True)
+
+    assert not np.shares_memory(get_array(df, "a"), arr_a)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+    tm.assert_frame_equal(df_copy, view)
+
+
+def test_clip_inplace_reference_no_op():
+    df = DataFrame({"a": [1.5, 2, 3]})
+    df_copy = df.copy()
+    arr_a = get_array(df, "a")
+    view = df[:]
+    df.clip(lower=0, inplace=True)
+
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+
+    assert not df._mgr._has_no_reference(0)
+    assert not view._mgr._has_no_reference(0)
+    tm.assert_frame_equal(df_copy, view)
+
+
+def test_clip_inplace():
+    df = DataFrame({"a": [1.5, 2, 3]})
+    arr_a = get_array(df, "a")
+    df.clip(lower=2, inplace=True)
+
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+    assert df._mgr._has_no_reference(0)
+
+
+def test_clip():
+    df = DataFrame({"a": [1.5, 2, 3]})
+    df_orig = df.copy()
+    df2 = df.clip(lower=2)
+
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    assert df._mgr._has_no_reference(0)
+    tm.assert_frame_equal(df_orig, df)
+
+
+def test_clip_no_op():
+    df = DataFrame({"a": [1.5, 2, 3]})
+    df2 = df.clip(lower=0)
+
+    assert not df._mgr._has_no_reference(0)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+
+def test_clip_chained_inplace():
+    df = DataFrame({"a": [1, 4, 2], "b": 1})
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        df["a"].clip(1, 2, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        df[["a"]].clip(1, 2, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_constructors.py b/pandas/pandas/tests/copy_view/test_constructors.py
new file mode 100644
index 0000000000000..4e4df4ba3cf22
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_constructors.py
@@ -0,0 +1,275 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Period,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+# -----------------------------------------------------------------------------
+# Copy/view behaviour for Series / DataFrame constructors
+
+
+@pytest.mark.parametrize("dtype", [None, "int64"])
+def test_series_from_series(dtype):
+    # Case: constructing a Series from another Series object follows CoW rules:
+    # a new object is returned and thus mutations are not propagated
+    ser = Series([1, 2, 3], name="name")
+
+    # default is copy=False -> new Series is a shallow copy / view of original
+    result = Series(ser, dtype=dtype)
+
+    # the shallow copy still shares memory
+    assert np.shares_memory(get_array(ser), get_array(result))
+
+    assert result._mgr.blocks[0].refs.has_reference()
+
+    # mutating new series copy doesn't mutate original
+    result.iloc[0] = 0
+    assert ser.iloc[0] == 1
+    # mutating triggered a copy-on-write -> no longer shares memory
+    assert not np.shares_memory(get_array(ser), get_array(result))
+
+    # the same when modifying the parent
+    result = Series(ser, dtype=dtype)
+
+    # mutating original doesn't mutate new series
+    ser.iloc[0] = 0
+    assert result.iloc[0] == 1
+
+
+def test_series_from_series_with_reindex():
+    # Case: constructing a Series from another Series with specifying an index
+    # that potentially requires a reindex of the values
+    ser = Series([1, 2, 3], name="name")
+
+    # passing an index that doesn't actually require a reindex of the values
+    # -> without CoW we get an actual mutating view
+    for index in [
+        ser.index,
+        ser.index.copy(),
+        list(ser.index),
+        ser.index.rename("idx"),
+    ]:
+        result = Series(ser, index=index)
+        assert np.shares_memory(ser.values, result.values)
+        result.iloc[0] = 0
+        assert ser.iloc[0] == 1
+
+    # ensure that if an actual reindex is needed, we don't have any refs
+    # (mutating the result wouldn't trigger CoW)
+    result = Series(ser, index=[0, 1, 2, 3])
+    assert not np.shares_memory(ser.values, result.values)
+    assert not result._mgr.blocks[0].refs.has_reference()
+
+
+@pytest.mark.parametrize("dtype", [None, "int64"])
+@pytest.mark.parametrize("idx", [None, pd.RangeIndex(start=0, stop=3, step=1)])
+@pytest.mark.parametrize(
+    "arr", [np.array([1, 2, 3], dtype="int64"), pd.array([1, 2, 3], dtype="Int64")]
+)
+def test_series_from_array(idx, dtype, arr):
+    ser = Series(arr, dtype=dtype, index=idx)
+    ser_orig = ser.copy()
+    data = getattr(arr, "_data", arr)
+    assert not np.shares_memory(get_array(ser), data)
+
+    arr[0] = 100
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize("copy", [True, False, None])
+def test_series_from_array_different_dtype(copy):
+    arr = np.array([1, 2, 3], dtype="int64")
+    ser = Series(arr, dtype="int32", copy=copy)
+    assert not np.shares_memory(get_array(ser), arr)
+
+
+@pytest.mark.parametrize(
+    "idx",
+    [
+        Index([1, 2]),
+        DatetimeIndex([Timestamp("2019-12-31"), Timestamp("2020-12-31")]),
+        PeriodIndex([Period("2019-12-31"), Period("2020-12-31")]),
+        TimedeltaIndex([Timedelta("1 days"), Timedelta("2 days")]),
+    ],
+)
+def test_series_from_index(idx):
+    ser = Series(idx)
+    expected = idx.copy(deep=True)
+    assert np.shares_memory(get_array(ser), get_array(idx))
+    assert not ser._mgr._has_no_reference(0)
+    ser.iloc[0] = ser.iloc[1]
+    tm.assert_index_equal(idx, expected)
+
+
+def test_series_from_index_different_dtypes():
+    idx = Index([1, 2, 3], dtype="int64")
+    ser = Series(idx, dtype="int32")
+    assert not np.shares_memory(get_array(ser), get_array(idx))
+    assert ser._mgr._has_no_reference(0)
+
+
+def test_series_from_block_manager_different_dtype():
+    ser = Series([1, 2, 3], dtype="int64")
+    msg = "Passing a SingleBlockManager to Series"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        ser2 = Series(ser._mgr, dtype="int32")
+    assert not np.shares_memory(get_array(ser), get_array(ser2))
+    assert ser2._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("use_mgr", [True, False])
+@pytest.mark.parametrize("columns", [None, ["a"]])
+def test_dataframe_constructor_mgr_or_df(columns, use_mgr):
+    df = DataFrame({"a": [1, 2, 3]})
+    df_orig = df.copy()
+
+    if use_mgr:
+        data = df._mgr
+        warn = DeprecationWarning
+    else:
+        data = df
+        warn = None
+    msg = "Passing a BlockManager to DataFrame"
+    with tm.assert_produces_warning(warn, match=msg, check_stacklevel=False):
+        new_df = DataFrame(data)
+
+    assert np.shares_memory(get_array(df, "a"), get_array(new_df, "a"))
+    new_df.iloc[0] = 100
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(new_df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("dtype", [None, "int64", "Int64"])
+@pytest.mark.parametrize("index", [None, [0, 1, 2]])
+@pytest.mark.parametrize("columns", [None, ["a", "b"], ["a", "b", "c"]])
+def test_dataframe_from_dict_of_series(columns, index, dtype):
+    # Case: constructing a DataFrame from Series objects with copy=False
+    # has to do a lazy following CoW rules
+    # (the default for DataFrame(dict) is still to copy to ensure consolidation)
+    s1 = Series([1, 2, 3])
+    s2 = Series([4, 5, 6])
+    s1_orig = s1.copy()
+    expected = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6]}, index=index, columns=columns, dtype=dtype
+    )
+
+    result = DataFrame(
+        {"a": s1, "b": s2}, index=index, columns=columns, dtype=dtype, copy=False
+    )
+
+    # the shallow copy still shares memory
+    assert np.shares_memory(get_array(result, "a"), get_array(s1))
+
+    # mutating the new dataframe doesn't mutate original
+    result.iloc[0, 0] = 10
+    assert not np.shares_memory(get_array(result, "a"), get_array(s1))
+    tm.assert_series_equal(s1, s1_orig)
+
+    # the same when modifying the parent series
+    s1 = Series([1, 2, 3])
+    s2 = Series([4, 5, 6])
+    result = DataFrame(
+        {"a": s1, "b": s2}, index=index, columns=columns, dtype=dtype, copy=False
+    )
+    s1.iloc[0] = 10
+    assert not np.shares_memory(get_array(result, "a"), get_array(s1))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [None, "int64"])
+def test_dataframe_from_dict_of_series_with_reindex(dtype):
+    # Case: constructing a DataFrame from Series objects with copy=False
+    # and passing an index that requires an actual (no-view) reindex -> need
+    # to ensure the result doesn't have refs set up to unnecessarily trigger
+    # a copy on write
+    s1 = Series([1, 2, 3])
+    s2 = Series([4, 5, 6])
+    df = DataFrame({"a": s1, "b": s2}, index=[1, 2, 3], dtype=dtype, copy=False)
+
+    # df should own its memory, so mutating shouldn't trigger a copy
+    arr_before = get_array(df, "a")
+    assert not np.shares_memory(arr_before, get_array(s1))
+    df.iloc[0, 0] = 100
+    arr_after = get_array(df, "a")
+    assert np.shares_memory(arr_before, arr_after)
+
+
+@pytest.mark.parametrize(
+    "data, dtype", [([1, 2], None), ([1, 2], "int64"), (["a", "b"], object)]
+)
+def test_dataframe_from_series_or_index(data, dtype, index_or_series):
+    obj = index_or_series(data, dtype=dtype)
+    obj_orig = obj.copy()
+    df = DataFrame(obj, dtype=dtype)
+    assert np.shares_memory(get_array(obj), get_array(df, 0))
+    assert not df._mgr._has_no_reference(0)
+
+    df.iloc[0, 0] = data[-1]
+    tm.assert_equal(obj, obj_orig)
+
+
+def test_dataframe_from_series_or_index_different_dtype(index_or_series):
+    obj = index_or_series([1, 2], dtype="int64")
+    df = DataFrame(obj, dtype="int32")
+    assert not np.shares_memory(get_array(obj), get_array(df, 0))
+    assert df._mgr._has_no_reference(0)
+
+
+def test_dataframe_from_series_dont_infer_datetime():
+    ser = Series([Timestamp("2019-12-31"), Timestamp("2020-12-31")], dtype=object)
+    df = DataFrame(ser)
+    assert df.dtypes.iloc[0] == np.dtype(object)
+    assert np.shares_memory(get_array(ser), get_array(df, 0))
+    assert not df._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("index", [None, [0, 1, 2]])
+def test_dataframe_from_dict_of_series_with_dtype(index):
+    # Variant of above, but now passing a dtype that causes a copy
+    # -> need to ensure the result doesn't have refs set up to unnecessarily
+    # trigger a copy on write
+    s1 = Series([1.0, 2.0, 3.0])
+    s2 = Series([4, 5, 6])
+    df = DataFrame({"a": s1, "b": s2}, index=index, dtype="int64", copy=False)
+
+    # df should own its memory, so mutating shouldn't trigger a copy
+    arr_before = get_array(df, "a")
+    assert not np.shares_memory(arr_before, get_array(s1))
+    df.iloc[0, 0] = 100
+    arr_after = get_array(df, "a")
+    assert np.shares_memory(arr_before, arr_after)
+
+
+@pytest.mark.parametrize("copy", [False, None, True])
+def test_frame_from_numpy_array(copy):
+    arr = np.array([[1, 2], [3, 4]])
+    df = DataFrame(arr, copy=copy)
+
+    if copy is not False or copy is True:
+        assert not np.shares_memory(get_array(df, 0), arr)
+    else:
+        assert np.shares_memory(get_array(df, 0), arr)
+
+
+def test_frame_from_dict_of_index():
+    idx = Index([1, 2, 3])
+    expected = idx.copy(deep=True)
+    df = DataFrame({"a": idx}, copy=False)
+    assert np.shares_memory(get_array(df, "a"), idx._values)
+    assert not df._mgr._has_no_reference(0)
+
+    df.iloc[0, 0] = 100
+    tm.assert_index_equal(idx, expected)
diff --git a/pandas/pandas/tests/copy_view/test_copy_deprecation.py b/pandas/pandas/tests/copy_view/test_copy_deprecation.py
new file mode 100644
index 0000000000000..8ee37213b92ab
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_copy_deprecation.py
@@ -0,0 +1,89 @@
+import pytest
+
+import pandas as pd
+from pandas import (
+    concat,
+    merge,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "meth, kwargs",
+    [
+        ("truncate", {}),
+        ("tz_convert", {"tz": "UTC"}),
+        ("tz_localize", {"tz": "UTC"}),
+        ("infer_objects", {}),
+        ("astype", {"dtype": "float64"}),
+        ("reindex", {"index": [2, 0, 1]}),
+        ("transpose", {}),
+        ("set_axis", {"labels": [1, 2, 3]}),
+        ("rename", {"index": {1: 2}}),
+        ("set_flags", {}),
+        ("to_period", {}),
+        ("to_timestamp", {}),
+        ("swaplevel", {"i": 0, "j": 1}),
+    ],
+)
+def test_copy_deprecation(meth, kwargs):
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": 1})
+
+    if meth in ("tz_convert", "tz_localize", "to_period"):
+        tz = None if meth in ("tz_localize", "to_period") else "US/Eastern"
+        df.index = pd.date_range("2020-01-01", freq="D", periods=len(df), tz=tz)
+    elif meth == "to_timestamp":
+        df.index = pd.period_range("2020-01-01", freq="D", periods=len(df))
+    elif meth == "swaplevel":
+        df = df.set_index(["b", "c"])
+
+    if meth != "swaplevel":
+        with tm.assert_produces_warning(DeprecationWarning, match="copy"):
+            getattr(df, meth)(copy=False, **kwargs)
+
+    if meth != "transpose":
+        with tm.assert_produces_warning(DeprecationWarning, match="copy"):
+            getattr(df.a, meth)(copy=False, **kwargs)
+
+
+def test_copy_deprecation_reindex_like_align():
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    # Somehow the stack level check is incorrect here
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        df.reindex_like(df, copy=False)
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        df.a.reindex_like(df.a, copy=False)
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        df.align(df, copy=False)
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        df.a.align(df.a, copy=False)
+
+
+def test_copy_deprecation_merge_concat():
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        df.merge(df, copy=False)
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        merge(df, df, copy=False)
+
+    with tm.assert_produces_warning(
+        DeprecationWarning, match="copy", check_stacklevel=False
+    ):
+        concat([df, df], copy=False)
diff --git a/pandas/pandas/tests/copy_view/test_core_functionalities.py b/pandas/pandas/tests/copy_view/test_core_functionalities.py
new file mode 100644
index 0000000000000..ad16bafdf0ee4
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_core_functionalities.py
@@ -0,0 +1,93 @@
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_assigning_to_same_variable_removes_references():
+    df = DataFrame({"a": [1, 2, 3]})
+    df = df.reset_index()
+    assert df._mgr._has_no_reference(1)
+    arr = get_array(df, "a")
+    df.iloc[0, 1] = 100  # Write into a
+
+    assert np.shares_memory(arr, get_array(df, "a"))
+
+
+def test_setitem_dont_track_unnecessary_references():
+    df = DataFrame({"a": [1, 2, 3], "b": 1, "c": 1})
+
+    df["b"] = 100
+    arr = get_array(df, "a")
+    # We split the block in setitem, if we are not careful the new blocks will
+    # reference each other triggering a copy
+    df.iloc[0, 0] = 100
+    assert np.shares_memory(arr, get_array(df, "a"))
+
+
+def test_setitem_with_view_copies():
+    df = DataFrame({"a": [1, 2, 3], "b": 1, "c": 1})
+    view = df[:]
+    expected = df.copy()
+
+    df["b"] = 100
+    arr = get_array(df, "a")
+    df.iloc[0, 0] = 100  # Check that we correctly track reference
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    tm.assert_frame_equal(view, expected)
+
+
+def test_setitem_with_view_invalidated_does_not_copy(request):
+    df = DataFrame({"a": [1, 2, 3], "b": 1, "c": 1})
+    view = df[:]
+
+    df["b"] = 100
+    arr = get_array(df, "a")
+    view = None  # noqa: F841
+    # TODO(CoW) block gets split because of `df["b"] = 100`
+    # which introduces additional refs, even when those of `view` go out of scopes
+    df.iloc[0, 0] = 100
+    # Setitem split the block. Since the old block shared data with view
+    # all the new blocks are referencing view and each other. When view
+    # goes out of scope, they don't share data with any other block,
+    # so we should not trigger a copy
+    mark = pytest.mark.xfail(reason="blk.delete does not track references correctly")
+    request.applymarker(mark)
+    assert np.shares_memory(arr, get_array(df, "a"))
+
+
+def test_out_of_scope():
+    def func():
+        df = DataFrame({"a": [1, 2], "b": 1.5, "c": 1})
+        # create some subset
+        result = df[["a", "b"]]
+        return result
+
+    result = func()
+    assert not result._mgr.blocks[0].refs.has_reference()
+    assert not result._mgr.blocks[1].refs.has_reference()
+
+
+def test_delete():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((4, 3)), columns=["a", "b", "c"]
+    )
+    del df["b"]
+    assert not df._mgr.blocks[0].refs.has_reference()
+    assert not df._mgr.blocks[1].refs.has_reference()
+
+    df = df[["a"]]
+    assert not df._mgr.blocks[0].refs.has_reference()
+
+
+def test_delete_reference():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((4, 3)), columns=["a", "b", "c"]
+    )
+    x = df[:]
+    del df["b"]
+    assert df._mgr.blocks[0].refs.has_reference()
+    assert df._mgr.blocks[1].refs.has_reference()
+    assert x._mgr.blocks[0].refs.has_reference()
diff --git a/pandas/pandas/tests/copy_view/test_functions.py b/pandas/pandas/tests/copy_view/test_functions.py
new file mode 100644
index 0000000000000..32fea794975b6
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_functions.py
@@ -0,0 +1,317 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import HAS_PYARROW
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    concat,
+    merge,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_concat_frames():
+    df = DataFrame({"b": ["a"] * 3}, dtype=object)
+    df2 = DataFrame({"a": ["a"] * 3}, dtype=object)
+    df_orig = df.copy()
+    result = concat([df, df2], axis=1)
+
+    assert np.shares_memory(get_array(result, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+
+    result.iloc[0, 0] = "d"
+    assert not np.shares_memory(get_array(result, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+
+    result.iloc[0, 1] = "d"
+    assert not np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_concat_frames_updating_input():
+    df = DataFrame({"b": ["a"] * 3}, dtype=object)
+    df2 = DataFrame({"a": ["a"] * 3}, dtype=object)
+    result = concat([df, df2], axis=1)
+
+    assert np.shares_memory(get_array(result, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+
+    expected = result.copy()
+    df.iloc[0, 0] = "d"
+    assert not np.shares_memory(get_array(result, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+
+    df2.iloc[0, 0] = "d"
+    assert not np.shares_memory(get_array(result, "a"), get_array(df2, "a"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_series():
+    ser = Series([1, 2], name="a")
+    ser2 = Series([3, 4], name="b")
+    ser_orig = ser.copy()
+    ser2_orig = ser2.copy()
+    result = concat([ser, ser2], axis=1)
+
+    assert np.shares_memory(get_array(result, "a"), ser.values)
+    assert np.shares_memory(get_array(result, "b"), ser2.values)
+
+    result.iloc[0, 0] = 100
+    assert not np.shares_memory(get_array(result, "a"), ser.values)
+    assert np.shares_memory(get_array(result, "b"), ser2.values)
+
+    result.iloc[0, 1] = 1000
+    assert not np.shares_memory(get_array(result, "b"), ser2.values)
+    tm.assert_series_equal(ser, ser_orig)
+    tm.assert_series_equal(ser2, ser2_orig)
+
+
+def test_concat_frames_chained():
+    df1 = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+    df2 = DataFrame({"c": [4, 5, 6]})
+    df3 = DataFrame({"d": [4, 5, 6]})
+    result = concat([concat([df1, df2], axis=1), df3], axis=1)
+    expected = result.copy()
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "c"), get_array(df2, "c"))
+    assert np.shares_memory(get_array(result, "d"), get_array(df3, "d"))
+
+    df1.iloc[0, 0] = 100
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_series_chained():
+    ser1 = Series([1, 2, 3], name="a")
+    ser2 = Series([4, 5, 6], name="c")
+    ser3 = Series([4, 5, 6], name="d")
+    result = concat([concat([ser1, ser2], axis=1), ser3], axis=1)
+    expected = result.copy()
+
+    assert np.shares_memory(get_array(result, "a"), get_array(ser1, "a"))
+    assert np.shares_memory(get_array(result, "c"), get_array(ser2, "c"))
+    assert np.shares_memory(get_array(result, "d"), get_array(ser3, "d"))
+
+    ser1.iloc[0] = 100
+    assert not np.shares_memory(get_array(result, "a"), get_array(ser1, "a"))
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_series_updating_input():
+    ser = Series([1, 2], name="a")
+    ser2 = Series([3, 4], name="b")
+    expected = DataFrame({"a": [1, 2], "b": [3, 4]})
+    result = concat([ser, ser2], axis=1)
+
+    assert np.shares_memory(get_array(result, "a"), get_array(ser, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(ser2, "b"))
+
+    ser.iloc[0] = 100
+    assert not np.shares_memory(get_array(result, "a"), get_array(ser, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(ser2, "b"))
+    tm.assert_frame_equal(result, expected)
+
+    ser2.iloc[0] = 1000
+    assert not np.shares_memory(get_array(result, "b"), get_array(ser2, "b"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_mixed_series_frame():
+    df = DataFrame({"a": [1, 2, 3], "c": 1})
+    ser = Series([4, 5, 6], name="d")
+    result = concat([df, ser], axis=1)
+    expected = result.copy()
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(result, "c"), get_array(df, "c"))
+    assert np.shares_memory(get_array(result, "d"), get_array(ser, "d"))
+
+    ser.iloc[0] = 100
+    assert not np.shares_memory(get_array(result, "d"), get_array(ser, "d"))
+
+    df.iloc[0, 0] = 100
+    assert not np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_copy_keyword():
+    df = DataFrame({"a": [1, 2]})
+    df2 = DataFrame({"b": [1.5, 2.5]})
+
+    result = concat([df, df2], axis=1)
+
+    assert np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(result, "b"))
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda df1, df2, **kwargs: df1.merge(df2, **kwargs),
+        lambda df1, df2, **kwargs: merge(df1, df2, **kwargs),
+    ],
+)
+def test_merge_on_key(func):
+    df1 = DataFrame({"key": Series(["a", "b", "c"], dtype=object), "a": [1, 2, 3]})
+    df2 = DataFrame({"key": Series(["a", "b", "c"], dtype=object), "b": [4, 5, 6]})
+    df1_orig = df1.copy()
+    df2_orig = df2.copy()
+
+    result = func(df1, df2, on="key")
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+    assert np.shares_memory(get_array(result, "key"), get_array(df1, "key"))
+    assert not np.shares_memory(get_array(result, "key"), get_array(df2, "key"))
+
+    result.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+
+    result.iloc[0, 2] = 0
+    assert not np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+    tm.assert_frame_equal(df1, df1_orig)
+    tm.assert_frame_equal(df2, df2_orig)
+
+
+def test_merge_on_index():
+    df1 = DataFrame({"a": [1, 2, 3]})
+    df2 = DataFrame({"b": [4, 5, 6]})
+    df1_orig = df1.copy()
+    df2_orig = df2.copy()
+
+    result = merge(df1, df2, left_index=True, right_index=True)
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+
+    result.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+
+    result.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+    tm.assert_frame_equal(df1, df1_orig)
+    tm.assert_frame_equal(df2, df2_orig)
+
+
+@pytest.mark.parametrize(
+    "func, how",
+    [
+        (lambda df1, df2, **kwargs: merge(df2, df1, on="key", **kwargs), "right"),
+        (lambda df1, df2, **kwargs: merge(df1, df2, on="key", **kwargs), "left"),
+    ],
+)
+def test_merge_on_key_enlarging_one(func, how):
+    df1 = DataFrame({"key": Series(["a", "b", "c"], dtype=object), "a": [1, 2, 3]})
+    df2 = DataFrame({"key": Series(["a", "b"], dtype=object), "b": [4, 5]})
+    df1_orig = df1.copy()
+    df2_orig = df2.copy()
+
+    result = func(df1, df2, how=how)
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert not np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+    assert df2._mgr._has_no_reference(1)
+    assert df2._mgr._has_no_reference(0)
+    assert np.shares_memory(get_array(result, "key"), get_array(df1, "key")) is (
+        how == "left"
+    )
+    assert not np.shares_memory(get_array(result, "key"), get_array(df2, "key"))
+
+    if how == "left":
+        result.iloc[0, 1] = 0
+    else:
+        result.iloc[0, 2] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    tm.assert_frame_equal(df1, df1_orig)
+    tm.assert_frame_equal(df2, df2_orig)
+
+
+def test_merge_copy_keyword():
+    df = DataFrame({"a": [1, 2]})
+    df2 = DataFrame({"b": [3, 4.5]})
+
+    result = df.merge(df2, left_index=True, right_index=True)
+
+    assert np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(result, "b"))
+
+
+@pytest.mark.xfail(
+    using_string_dtype() and HAS_PYARROW,
+    reason="TODO(infer_string); result.index infers str dtype while both "
+    "df1 and df2 index are object.",
+)
+def test_join_on_key():
+    df_index = Index(["a", "b", "c"], name="key", dtype=object)
+
+    df1 = DataFrame({"a": [1, 2, 3]}, index=df_index.copy(deep=True))
+    df2 = DataFrame({"b": [4, 5, 6]}, index=df_index.copy(deep=True))
+
+    df1_orig = df1.copy()
+    df2_orig = df2.copy()
+
+    result = df1.join(df2, on="key")
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+    assert np.shares_memory(get_array(result.index), get_array(df1.index))
+    assert not np.shares_memory(get_array(result.index), get_array(df2.index))
+
+    result.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+
+    result.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(result, "b"), get_array(df2, "b"))
+
+    tm.assert_frame_equal(df1, df1_orig)
+    tm.assert_frame_equal(df2, df2_orig)
+
+
+def test_join_multiple_dataframes_on_key():
+    df_index = Index(["a", "b", "c"], name="key", dtype=object)
+
+    df1 = DataFrame({"a": [1, 2, 3]}, index=df_index.copy(deep=True))
+    dfs_list = [
+        DataFrame({"b": [4, 5, 6]}, index=df_index.copy(deep=True)),
+        DataFrame({"c": [7, 8, 9]}, index=df_index.copy(deep=True)),
+    ]
+
+    df1_orig = df1.copy()
+    dfs_list_orig = [df.copy() for df in dfs_list]
+
+    result = df1.join(dfs_list)
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(dfs_list[0], "b"))
+    assert np.shares_memory(get_array(result, "c"), get_array(dfs_list[1], "c"))
+    assert np.shares_memory(get_array(result.index), get_array(df1.index))
+    assert not np.shares_memory(get_array(result.index), get_array(dfs_list[0].index))
+    assert not np.shares_memory(get_array(result.index), get_array(dfs_list[1].index))
+
+    result.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df1, "a"))
+    assert np.shares_memory(get_array(result, "b"), get_array(dfs_list[0], "b"))
+    assert np.shares_memory(get_array(result, "c"), get_array(dfs_list[1], "c"))
+
+    result.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(result, "b"), get_array(dfs_list[0], "b"))
+    assert np.shares_memory(get_array(result, "c"), get_array(dfs_list[1], "c"))
+
+    result.iloc[0, 2] = 0
+    assert not np.shares_memory(get_array(result, "c"), get_array(dfs_list[1], "c"))
+
+    tm.assert_frame_equal(df1, df1_orig)
+    for df, df_orig in zip(dfs_list, dfs_list_orig):
+        tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_indexing.py b/pandas/pandas/tests/copy_view/test_indexing.py
new file mode 100644
index 0000000000000..37a21e1098e78
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_indexing.py
@@ -0,0 +1,883 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_float_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+@pytest.fixture(params=["numpy", "nullable"])
+def backend(request):
+    if request.param == "numpy":
+
+        def make_dataframe(*args, **kwargs):
+            return DataFrame(*args, **kwargs)
+
+        def make_series(*args, **kwargs):
+            return Series(*args, **kwargs)
+
+    elif request.param == "nullable":
+
+        def make_dataframe(*args, **kwargs):
+            df = DataFrame(*args, **kwargs)
+            df_nullable = df.convert_dtypes()
+            # convert_dtypes will try to cast float to int if there is no loss in
+            # precision -> undo that change
+            for col in df.columns:
+                if is_float_dtype(df[col].dtype) and not is_float_dtype(
+                    df_nullable[col].dtype
+                ):
+                    df_nullable[col] = df_nullable[col].astype("Float64")
+            # copy final result to ensure we start with a fully self-owning DataFrame
+            return df_nullable.copy()
+
+        def make_series(*args, **kwargs):
+            ser = Series(*args, **kwargs)
+            return ser.convert_dtypes().copy()
+
+    return request.param, make_dataframe, make_series
+
+
+# -----------------------------------------------------------------------------
+# Indexing operations taking subset + modifying the subset/parent
+
+
+def test_subset_column_selection(backend):
+    # Case: taking a subset of the columns of a DataFrame
+    # + afterwards modifying the subset
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    subset = df[["a", "c"]]
+
+    # the subset shares memory ...
+    assert np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+    # ... but uses CoW when being modified
+    subset.iloc[0, 0] = 0
+
+    assert not np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+
+    expected = DataFrame({"a": [0, 2, 3], "c": [0.1, 0.2, 0.3]})
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_subset_column_selection_modify_parent(backend):
+    # Case: taking a subset of the columns of a DataFrame
+    # + afterwards modifying the parent
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+
+    subset = df[["a", "c"]]
+
+    # the subset shares memory ...
+    assert np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+    # ... but parent uses CoW parent when it is modified
+    df.iloc[0, 0] = 0
+
+    assert not np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+    # different column/block still shares memory
+    assert np.shares_memory(get_array(subset, "c"), get_array(df, "c"))
+
+    expected = DataFrame({"a": [1, 2, 3], "c": [0.1, 0.2, 0.3]})
+    tm.assert_frame_equal(subset, expected)
+
+
+def test_subset_row_slice(backend):
+    # Case: taking a subset of the rows of a DataFrame using a slice
+    # + afterwards modifying the subset
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    subset = df[1:3]
+    subset._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+
+    subset.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(subset, "a"), get_array(df, "a"))
+
+    subset._mgr._verify_integrity()
+
+    expected = DataFrame({"a": [0, 3], "b": [5, 6], "c": [0.2, 0.3]}, index=range(1, 3))
+    tm.assert_frame_equal(subset, expected)
+    # original parent dataframe is not modified (CoW)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_subset_column_slice(backend, dtype):
+    # Case: taking a subset of the columns of a DataFrame using a slice
+    # + afterwards modifying the subset
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    subset = df.iloc[:, 1:]
+    subset._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(subset, "b"), get_array(df, "b"))
+
+    subset.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(subset, "b"), get_array(df, "b"))
+
+    expected = DataFrame({"b": [0, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)})
+    tm.assert_frame_equal(subset, expected)
+    # original parent dataframe is not modified (also not for BlockManager case,
+    # except for single block)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+@pytest.mark.parametrize(
+    "row_indexer",
+    [slice(1, 2), np.array([False, True, True]), np.array([1, 2])],
+    ids=["slice", "mask", "array"],
+)
+@pytest.mark.parametrize(
+    "column_indexer",
+    [slice("b", "c"), np.array([False, True, True]), ["b", "c"]],
+    ids=["slice", "mask", "array"],
+)
+def test_subset_loc_rows_columns(
+    backend,
+    dtype,
+    row_indexer,
+    column_indexer,
+):
+    # Case: taking a subset of the rows+columns of a DataFrame using .loc
+    # + afterwards modifying the subset
+    # Generic test for several combinations of row/column indexers, not all
+    # of those could actually return a view / need CoW (so this test is not
+    # checking memory sharing, only ensuring subsequent mutation doesn't
+    # affect the parent dataframe)
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    subset = df.loc[row_indexer, column_indexer]
+
+    # modifying the subset never modifies the parent
+    subset.iloc[0, 0] = 0
+
+    expected = DataFrame(
+        {"b": [0, 6], "c": np.array([8, 9], dtype=dtype)}, index=range(1, 3)
+    )
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+@pytest.mark.parametrize(
+    "row_indexer",
+    [slice(1, 3), np.array([False, True, True]), np.array([1, 2])],
+    ids=["slice", "mask", "array"],
+)
+@pytest.mark.parametrize(
+    "column_indexer",
+    [slice(1, 3), np.array([False, True, True]), [1, 2]],
+    ids=["slice", "mask", "array"],
+)
+def test_subset_iloc_rows_columns(
+    backend,
+    dtype,
+    row_indexer,
+    column_indexer,
+):
+    # Case: taking a subset of the rows+columns of a DataFrame using .iloc
+    # + afterwards modifying the subset
+    # Generic test for several combinations of row/column indexers, not all
+    # of those could actually return a view / need CoW (so this test is not
+    # checking memory sharing, only ensuring subsequent mutation doesn't
+    # affect the parent dataframe)
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    subset = df.iloc[row_indexer, column_indexer]
+
+    # modifying the subset never modifies the parent
+    subset.iloc[0, 0] = 0
+
+    expected = DataFrame(
+        {"b": [0, 6], "c": np.array([8, 9], dtype=dtype)}, index=range(1, 3)
+    )
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "indexer",
+    [slice(0, 2), np.array([True, True, False]), np.array([0, 1])],
+    ids=["slice", "mask", "array"],
+)
+def test_subset_set_with_row_indexer(backend, indexer_si, indexer):
+    # Case: setting values with a row indexer on a viewing subset
+    # subset[indexer] = value and subset.iloc[indexer] = value
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3, 4], "b": [4, 5, 6, 7], "c": [0.1, 0.2, 0.3, 0.4]})
+    df_orig = df.copy()
+    subset = df[1:4]
+
+    if (
+        indexer_si is tm.setitem
+        and isinstance(indexer, np.ndarray)
+        and indexer.dtype == "int"
+    ):
+        pytest.skip("setitem with labels selects on columns")
+
+    indexer_si(subset)[indexer] = 0
+
+    expected = DataFrame(
+        {"a": [0, 0, 4], "b": [0, 0, 7], "c": [0.0, 0.0, 0.4]}, index=range(1, 4)
+    )
+    tm.assert_frame_equal(subset, expected)
+    # original parent dataframe is not modified (CoW)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_subset_set_with_mask(backend):
+    # Case: setting values with a mask on a viewing subset: subset[mask] = value
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3, 4], "b": [4, 5, 6, 7], "c": [0.1, 0.2, 0.3, 0.4]})
+    df_orig = df.copy()
+    subset = df[1:4]
+
+    mask = subset > 3
+
+    subset[mask] = 0
+
+    expected = DataFrame(
+        {"a": [2, 3, 0], "b": [0, 0, 0], "c": [0.20, 0.3, 0.4]}, index=range(1, 4)
+    )
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_subset_set_column(backend):
+    # Case: setting a single column on a viewing subset -> subset[col] = value
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    subset = df[1:3]
+
+    if dtype_backend == "numpy":
+        arr = np.array([10, 11], dtype="int64")
+    else:
+        arr = pd.array([10, 11], dtype="Int64")
+
+    subset["a"] = arr
+    subset._mgr._verify_integrity()
+    expected = DataFrame(
+        {"a": [10, 11], "b": [5, 6], "c": [0.2, 0.3]}, index=range(1, 3)
+    )
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_subset_set_column_with_loc(backend, dtype):
+    # Case: setting a single column with loc on a viewing subset
+    # -> subset.loc[:, col] = value
+    _, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+    subset = df[1:3]
+
+    subset.loc[:, "a"] = np.array([10, 11], dtype="int64")
+
+    subset._mgr._verify_integrity()
+    expected = DataFrame(
+        {"a": [10, 11], "b": [5, 6], "c": np.array([8, 9], dtype=dtype)},
+        index=range(1, 3),
+    )
+    tm.assert_frame_equal(subset, expected)
+    # original parent dataframe is not modified (CoW)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_subset_set_column_with_loc2(backend):
+    # Case: setting a single column with loc on a viewing subset
+    # -> subset.loc[:, col] = value
+    # separate test for case of DataFrame of a single column -> takes a separate
+    # code path
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3]})
+    df_orig = df.copy()
+    subset = df[1:3]
+
+    subset.loc[:, "a"] = 0
+
+    subset._mgr._verify_integrity()
+    expected = DataFrame({"a": [0, 0]}, index=range(1, 3))
+    tm.assert_frame_equal(subset, expected)
+    # original parent dataframe is not modified (CoW)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_subset_set_columns(backend, dtype):
+    # Case: setting multiple columns on a viewing subset
+    # -> subset[[col1, col2]] = value
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+    subset = df[1:3]
+
+    subset[["a", "c"]] = 0
+
+    subset._mgr._verify_integrity()
+    # first and third column should certainly have no references anymore
+    assert all(subset._mgr._has_no_reference(i) for i in [0, 2])
+    expected = DataFrame({"a": [0, 0], "b": [5, 6], "c": [0, 0]}, index=range(1, 3))
+    if dtype_backend == "nullable":
+        # there is not yet a global option, so overriding a column by setting a scalar
+        # defaults to numpy dtype even if original column was nullable
+        expected["a"] = expected["a"].astype("int64")
+        expected["c"] = expected["c"].astype("int64")
+
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "indexer",
+    [slice("a", "b"), np.array([True, True, False]), ["a", "b"]],
+    ids=["slice", "mask", "array"],
+)
+def test_subset_set_with_column_indexer(backend, indexer):
+    # Case: setting multiple columns with a column indexer on a viewing subset
+    # -> subset.loc[:, [col1, col2]] = value
+    _, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3], "c": [4, 5, 6]})
+    df_orig = df.copy()
+    subset = df[1:3]
+
+    subset.loc[:, indexer] = 0
+
+    subset._mgr._verify_integrity()
+    expected = DataFrame({"a": [0, 0], "b": [0.0, 0.0], "c": [5, 6]}, index=range(1, 3))
+    tm.assert_frame_equal(subset, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df: df[["a", "b"]][0:2],
+        lambda df: df[0:2][["a", "b"]],
+        lambda df: df[["a", "b"]].iloc[0:2],
+        lambda df: df[["a", "b"]].loc[0:1],
+        lambda df: df[0:2].iloc[:, 0:2],
+        lambda df: df[0:2].loc[:, "a":"b"],  # type: ignore[misc]
+    ],
+    ids=[
+        "row-getitem-slice",
+        "column-getitem",
+        "row-iloc-slice",
+        "row-loc-slice",
+        "column-iloc-slice",
+        "column-loc-slice",
+    ],
+)
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_subset_chained_getitem(
+    request,
+    backend,
+    method,
+    dtype,
+):
+    # Case: creating a subset using multiple, chained getitem calls using views
+    # still needs to guarantee proper CoW behaviour
+    _, DataFrame, _ = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    # modify subset -> don't modify parent
+    subset = method(df)
+
+    subset.iloc[0, 0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+    # modify parent -> don't modify subset
+    subset = method(df)
+    df.iloc[0, 0] = 0
+    expected = DataFrame({"a": [1, 2], "b": [4, 5]})
+    tm.assert_frame_equal(subset, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_subset_chained_getitem_column(backend, dtype):
+    # Case: creating a subset using multiple, chained getitem calls using views
+    # still needs to guarantee proper CoW behaviour
+    dtype_backend, DataFrame, Series = backend
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    # modify subset -> don't modify parent
+    subset = df[:]["a"][0:2]
+    subset.iloc[0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+    # modify parent -> don't modify subset
+    subset = df[:]["a"][0:2]
+    df.iloc[0, 0] = 0
+    expected = Series([1, 2], name="a")
+    tm.assert_series_equal(subset, expected)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda s: s["a":"c"]["a":"b"],  # type: ignore[misc]
+        lambda s: s.iloc[0:3].iloc[0:2],
+        lambda s: s.loc["a":"c"].loc["a":"b"],  # type: ignore[misc]
+        lambda s: s.loc["a":"c"]  # type: ignore[misc]
+        .iloc[0:3]
+        .iloc[0:2]
+        .loc["a":"b"]  # type: ignore[misc]
+        .iloc[0:1],
+    ],
+    ids=["getitem", "iloc", "loc", "long-chain"],
+)
+def test_subset_chained_getitem_series(backend, method):
+    # Case: creating a subset using multiple, chained getitem calls using views
+    # still needs to guarantee proper CoW behaviour
+    _, _, Series = backend
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    s_orig = s.copy()
+
+    # modify subset -> don't modify parent
+    subset = method(s)
+    subset.iloc[0] = 0
+    tm.assert_series_equal(s, s_orig)
+
+    # modify parent -> don't modify subset
+    subset = s.iloc[0:3].iloc[0:2]
+    s.iloc[0] = 0
+    expected = Series([1, 2], index=["a", "b"])
+    tm.assert_series_equal(subset, expected)
+
+
+def test_subset_chained_single_block_row():
+    # not parametrizing this for dtype backend, since this explicitly tests single block
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+    df_orig = df.copy()
+
+    # modify subset -> don't modify parent
+    subset = df[:].iloc[0].iloc[0:2]
+    subset.iloc[0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+    # modify parent -> don't modify subset
+    subset = df[:].iloc[0].iloc[0:2]
+    df.iloc[0, 0] = 0
+    expected = Series([1, 4], index=["a", "b"], name=0)
+    tm.assert_series_equal(subset, expected)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df: df[:],
+        lambda df: df.loc[:, :],
+        lambda df: df.loc[:],
+        lambda df: df.iloc[:, :],
+        lambda df: df.iloc[:],
+    ],
+    ids=["getitem", "loc", "loc-rows", "iloc", "iloc-rows"],
+)
+def test_null_slice(backend, method):
+    # Case: also all variants of indexing with a null slice (:) should return
+    # new objects to ensure we correctly use CoW for the results
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+    df_orig = df.copy()
+
+    df2 = method(df)
+
+    # we always return new objects (shallow copy), regardless of CoW or not
+    assert df2 is not df
+
+    # and those trigger CoW when mutated
+    df2.iloc[0, 0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda s: s[:],
+        lambda s: s.loc[:],
+        lambda s: s.iloc[:],
+    ],
+    ids=["getitem", "loc", "iloc"],
+)
+def test_null_slice_series(backend, method):
+    _, _, Series = backend
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    s_orig = s.copy()
+
+    s2 = method(s)
+
+    # we always return new objects, regardless of CoW or not
+    assert s2 is not s
+
+    # and those trigger CoW when mutated
+    s2.iloc[0] = 0
+    tm.assert_series_equal(s, s_orig)
+
+
+# TODO add more tests modifying the parent
+
+
+# -----------------------------------------------------------------------------
+# Series -- Indexing operations taking subset + modifying the subset/parent
+
+
+def test_series_getitem_slice(backend):
+    # Case: taking a slice of a Series + afterwards modifying the subset
+    _, _, Series = backend
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    s_orig = s.copy()
+
+    subset = s[:]
+    assert np.shares_memory(get_array(subset), get_array(s))
+
+    subset.iloc[0] = 0
+
+    assert not np.shares_memory(get_array(subset), get_array(s))
+
+    expected = Series([0, 2, 3], index=["a", "b", "c"])
+    tm.assert_series_equal(subset, expected)
+
+    # original parent series is not modified (CoW)
+    tm.assert_series_equal(s, s_orig)
+
+
+def test_series_getitem_ellipsis():
+    # Case: taking a view of a Series using Ellipsis + afterwards modifying the subset
+    s = Series([1, 2, 3])
+    s_orig = s.copy()
+
+    subset = s[...]
+    assert np.shares_memory(get_array(subset), get_array(s))
+
+    subset.iloc[0] = 0
+
+    assert not np.shares_memory(get_array(subset), get_array(s))
+
+    expected = Series([0, 2, 3])
+    tm.assert_series_equal(subset, expected)
+
+    # original parent series is not modified (CoW)
+    tm.assert_series_equal(s, s_orig)
+
+
+@pytest.mark.parametrize(
+    "indexer",
+    [slice(0, 2), np.array([True, True, False]), np.array([0, 1])],
+    ids=["slice", "mask", "array"],
+)
+def test_series_subset_set_with_indexer(backend, indexer_si, indexer):
+    # Case: setting values in a viewing Series with an indexer
+    _, _, Series = backend
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    s_orig = s.copy()
+    subset = s[:]
+
+    if (
+        indexer_si is tm.setitem
+        and isinstance(indexer, np.ndarray)
+        and indexer.dtype.kind == "i"
+    ):
+        # In 3.0 we treat integers as always-labels
+        with pytest.raises(KeyError):
+            indexer_si(subset)[indexer] = 0
+        return
+
+    indexer_si(subset)[indexer] = 0
+    expected = Series([0, 0, 3], index=["a", "b", "c"])
+    tm.assert_series_equal(subset, expected)
+
+    tm.assert_series_equal(s, s_orig)
+
+
+# -----------------------------------------------------------------------------
+# del operator
+
+
+def test_del_frame(backend):
+    # Case: deleting a column with `del` on a viewing child dataframe should
+    # not modify parent + update the references
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df[:]
+
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    del df2["b"]
+
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+    tm.assert_frame_equal(df, df_orig)
+    tm.assert_frame_equal(df2, df_orig[["a", "c"]])
+    df2._mgr._verify_integrity()
+
+    df.loc[0, "b"] = 200
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+    df_orig = df.copy()
+
+    df2.loc[0, "a"] = 100
+    # modifying child after deleting a column still doesn't update parent
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_del_series(backend):
+    _, _, Series = backend
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    s_orig = s.copy()
+    s2 = s[:]
+
+    assert np.shares_memory(get_array(s), get_array(s2))
+
+    del s2["a"]
+
+    assert not np.shares_memory(get_array(s), get_array(s2))
+    tm.assert_series_equal(s, s_orig)
+    tm.assert_series_equal(s2, s_orig[["b", "c"]])
+
+    # modifying s2 doesn't need copy on write (due to `del`, s2 is backed by new array)
+    values = s2.values
+    s2.loc["b"] = 100
+    assert values[0] == 100
+
+
+# -----------------------------------------------------------------------------
+# Accessing column as Series
+
+
+def test_column_as_series(backend):
+    # Case: selecting a single column now also uses Copy-on-Write
+    dtype_backend, DataFrame, Series = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    s = df["a"]
+
+    assert np.shares_memory(get_array(s, "a"), get_array(df, "a"))
+    s[0] = 0
+
+    expected = Series([0, 2, 3], name="a")
+    tm.assert_series_equal(s, expected)
+    # assert not np.shares_memory(s.values, get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+    # ensure cached series on getitem is not the changed series
+    tm.assert_series_equal(df["a"], df_orig["a"])
+
+
+def test_column_as_series_set_with_upcast(backend):
+    # Case: selecting a single column now also uses Copy-on-Write -> when
+    # setting a value causes an upcast, we don't need to update the parent
+    # DataFrame through the cache mechanism
+    dtype_backend, DataFrame, Series = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    s = df["a"]
+    if dtype_backend == "nullable":
+        with pytest.raises(TypeError, match="Invalid value"):
+            s[0] = "foo"
+        expected = Series([1, 2, 3], name="a")
+        tm.assert_series_equal(s, expected)
+        tm.assert_frame_equal(df, df_orig)
+        # ensure cached series on getitem is not the changed series
+        tm.assert_series_equal(df["a"], df_orig["a"])
+    else:
+        with pytest.raises(TypeError, match="Invalid value"):
+            s[0] = "foo"
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df: df["a"],
+        lambda df: df.loc[:, "a"],
+        lambda df: df.iloc[:, 0],
+    ],
+    ids=["getitem", "loc", "iloc"],
+)
+def test_column_as_series_no_item_cache(request, backend, method):
+    # Case: selecting a single column (which now also uses Copy-on-Write to protect
+    # the view) should always give a new object (i.e. not make use of a cache)
+    dtype_backend, DataFrame, _ = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    s1 = method(df)
+    s2 = method(df)
+
+    assert s1 is not s2
+
+    s1.iloc[0] = 0
+
+    tm.assert_series_equal(s2, df_orig["a"])
+    tm.assert_frame_equal(df, df_orig)
+
+
+# TODO add tests for other indexing methods on the Series
+
+
+def test_dataframe_add_column_from_series(backend):
+    # Case: adding a new column to a DataFrame from an existing column/series
+    # -> delays copy under CoW
+    _, DataFrame, Series = backend
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+
+    s = Series([10, 11, 12])
+    df["new"] = s
+    assert np.shares_memory(get_array(df, "new"), get_array(s))
+
+    # editing series -> doesn't modify column in frame
+    s[0] = 0
+    expected = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3], "new": [10, 11, 12]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("val", [100, "a"])
+@pytest.mark.parametrize(
+    "indexer_func, indexer",
+    [
+        (tm.loc, (0, "a")),
+        (tm.iloc, (0, 0)),
+        (tm.loc, ([0], "a")),
+        (tm.iloc, ([0], 0)),
+        (tm.loc, (slice(None), "a")),
+        (tm.iloc, (slice(None), 0)),
+    ],
+)
+@pytest.mark.parametrize(
+    "col", [[0.1, 0.2, 0.3], [7, 8, 9]], ids=["mixed-block", "single-block"]
+)
+def test_set_value_copy_only_necessary_column(indexer_func, indexer, val, col):
+    # When setting inplace, only copy column that is modified instead of the whole
+    # block (by splitting the block)
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": col})
+    df_orig = df.copy()
+    view = df[:]
+
+    if val == "a":
+        with pytest.raises(TypeError, match="Invalid value"):
+            indexer_func(df)[indexer] = val
+    else:
+        indexer_func(df)[indexer] = val
+
+        assert np.shares_memory(get_array(df, "b"), get_array(view, "b"))
+        assert not np.shares_memory(get_array(df, "a"), get_array(view, "a"))
+        tm.assert_frame_equal(view, df_orig)
+
+
+def test_series_midx_slice():
+    ser = Series([1, 2, 3], index=pd.MultiIndex.from_arrays([[1, 1, 2], [3, 4, 5]]))
+    ser_orig = ser.copy()
+    result = ser[1]
+    assert np.shares_memory(get_array(ser), get_array(result))
+    result.iloc[0] = 100
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_getitem_midx_slice():
+    df = DataFrame({("a", "x"): [1, 2], ("a", "y"): 1, ("b", "x"): 2})
+    df_orig = df.copy()
+    new_df = df[("a",)]
+
+    assert not new_df._mgr._has_no_reference(0)
+
+    assert np.shares_memory(get_array(df, ("a", "x")), get_array(new_df, "x"))
+    new_df.iloc[0, 0] = 100
+    tm.assert_frame_equal(df_orig, df)
+
+
+def test_series_midx_tuples_slice():
+    ser = Series(
+        [1, 2, 3],
+        index=pd.MultiIndex.from_tuples([((1, 2), 3), ((1, 2), 4), ((2, 3), 4)]),
+    )
+    result = ser[(1, 2)]
+    assert np.shares_memory(get_array(ser), get_array(result))
+    result.iloc[0] = 100
+    expected = Series(
+        [1, 2, 3],
+        index=pd.MultiIndex.from_tuples([((1, 2), 3), ((1, 2), 4), ((2, 3), 4)]),
+    )
+    tm.assert_series_equal(ser, expected)
+
+
+def test_midx_read_only_bool_indexer():
+    # GH#56635
+    def mklbl(prefix, n):
+        return [f"{prefix}{i}" for i in range(n)]
+
+    idx = pd.MultiIndex.from_product(
+        [mklbl("A", 4), mklbl("B", 2), mklbl("C", 4), mklbl("D", 2)]
+    )
+    cols = pd.MultiIndex.from_tuples(
+        [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")], names=["lvl0", "lvl1"]
+    )
+    df = DataFrame(1, index=idx, columns=cols).sort_index().sort_index(axis=1)
+
+    mask = df[("a", "foo")] == 1
+    expected_mask = mask.copy()
+    result = df.loc[pd.IndexSlice[mask, :, ["C1", "C3"]], :]
+    expected = df.loc[pd.IndexSlice[:, :, ["C1", "C3"]], :]
+    tm.assert_frame_equal(result, expected)
+    tm.assert_series_equal(mask, expected_mask)
+
+
+def test_loc_enlarging_with_dataframe():
+    df = DataFrame({"a": [1, 2, 3]})
+    rhs = DataFrame({"b": [1, 2, 3], "c": [4, 5, 6]})
+    rhs_orig = rhs.copy()
+    df.loc[:, ["b", "c"]] = rhs
+    assert np.shares_memory(get_array(df, "b"), get_array(rhs, "b"))
+    assert np.shares_memory(get_array(df, "c"), get_array(rhs, "c"))
+    assert not df._mgr._has_no_reference(1)
+
+    df.iloc[0, 1] = 100
+    tm.assert_frame_equal(rhs, rhs_orig)
diff --git a/pandas/pandas/tests/copy_view/test_internals.py b/pandas/pandas/tests/copy_view/test_internals.py
new file mode 100644
index 0000000000000..b7baf01ecc36e
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_internals.py
@@ -0,0 +1,112 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_consolidate():
+    # create unconsolidated DataFrame
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+    df["c"] = [4, 5, 6]
+
+    # take a viewing subset
+    subset = df[:]
+
+    # each block of subset references a block of df
+    assert all(blk.refs.has_reference() for blk in subset._mgr.blocks)
+
+    # consolidate the two int64 blocks
+    subset._consolidate_inplace()
+
+    # the float64 block still references the parent one because it still a view
+    assert subset._mgr.blocks[0].refs.has_reference()
+    # equivalent of assert np.shares_memory(df["b"].values, subset["b"].values)
+    # but avoids caching df["b"]
+    assert np.shares_memory(get_array(df, "b"), get_array(subset, "b"))
+
+    # the new consolidated int64 block does not reference another
+    assert not subset._mgr.blocks[1].refs.has_reference()
+
+    # the parent dataframe now also only is linked for the float column
+    assert not df._mgr.blocks[0].refs.has_reference()
+    assert df._mgr.blocks[1].refs.has_reference()
+    assert not df._mgr.blocks[2].refs.has_reference()
+
+    # and modifying subset still doesn't modify parent
+    subset.iloc[0, 1] = 0.0
+    assert not df._mgr.blocks[1].refs.has_reference()
+    assert df.loc[0, "b"] == 0.1
+
+
+@pytest.mark.parametrize("dtype", [np.intp, np.int8])
+@pytest.mark.parametrize(
+    "locs, arr",
+    [
+        ([0], np.array([-1, -2, -3])),
+        ([1], np.array([-1, -2, -3])),
+        ([5], np.array([-1, -2, -3])),
+        ([0, 1], np.array([[-1, -2, -3], [-4, -5, -6]]).T),
+        ([0, 2], np.array([[-1, -2, -3], [-4, -5, -6]]).T),
+        ([0, 1, 2], np.array([[-1, -2, -3], [-4, -5, -6], [-4, -5, -6]]).T),
+        ([1, 2], np.array([[-1, -2, -3], [-4, -5, -6]]).T),
+        ([1, 3], np.array([[-1, -2, -3], [-4, -5, -6]]).T),
+    ],
+)
+def test_iset_splits_blocks_inplace(locs, arr, dtype):
+    # Nothing currently calls iset with
+    # more than 1 loc with inplace=True (only happens with inplace=False)
+    # but ensure that it works
+    df = DataFrame(
+        {
+            "a": [1, 2, 3],
+            "b": [4, 5, 6],
+            "c": [7, 8, 9],
+            "d": [10, 11, 12],
+            "e": [13, 14, 15],
+            "f": Series(["a", "b", "c"], dtype=object),
+        },
+    )
+    arr = arr.astype(dtype)
+    df_orig = df.copy()
+    df2 = df.copy(deep=False)  # Trigger a CoW (if enabled, otherwise makes copy)
+    df2._mgr.iset(locs, arr, inplace=True)
+
+    tm.assert_frame_equal(df, df_orig)
+    for i, col in enumerate(df.columns):
+        if i not in locs:
+            assert np.shares_memory(get_array(df, col), get_array(df2, col))
+
+
+def test_exponential_backoff():
+    # GH#55518
+    df = DataFrame({"a": [1, 2, 3]})
+    for i in range(490):
+        df.copy(deep=False)
+
+    assert len(df._mgr.blocks[0].refs.referenced_blocks) == 491
+
+    df = DataFrame({"a": [1, 2, 3]})
+    dfs = [df.copy(deep=False) for i in range(510)]
+
+    for i in range(20):
+        df.copy(deep=False)
+    assert len(df._mgr.blocks[0].refs.referenced_blocks) == 531
+    assert df._mgr.blocks[0].refs.clear_counter == 1000
+
+    for i in range(500):
+        df.copy(deep=False)
+
+    # Don't reduce since we still have over 500 objects alive
+    assert df._mgr.blocks[0].refs.clear_counter == 1000
+
+    dfs = dfs[:300]
+    for i in range(500):
+        df.copy(deep=False)
+
+    # Reduce since there are less than 500 objects alive
+    assert df._mgr.blocks[0].refs.clear_counter == 500
diff --git a/pandas/pandas/tests/copy_view/test_interp_fillna.py b/pandas/pandas/tests/copy_view/test_interp_fillna.py
new file mode 100644
index 0000000000000..6bcda0ef2c35a
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_interp_fillna.py
@@ -0,0 +1,300 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    Interval,
+    NaT,
+    Series,
+    Timestamp,
+    interval_range,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+@pytest.mark.parametrize("method", ["pad", "nearest", "linear"])
+def test_interpolate_no_op(method):
+    df = DataFrame({"a": [1, 2]})
+    df_orig = df.copy()
+
+    if method == "pad":
+        msg = f"Can not interpolate with method={method}"
+        with pytest.raises(ValueError, match=msg):
+            df.interpolate(method=method)
+    else:
+        result = df.interpolate(method=method)
+        assert np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+
+        result.iloc[0, 0] = 100
+
+        assert not np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+        tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill"])
+def test_interp_fill_functions(func):
+    # Check that these takes the same code paths as interpolate
+    df = DataFrame({"a": [1, 2]})
+    df_orig = df.copy()
+
+    result = getattr(df, func)()
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    result.iloc[0, 0] = 100
+
+    assert not np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill"])
+@pytest.mark.parametrize(
+    "vals", [[1, np.nan, 2], [Timestamp("2019-12-31"), NaT, Timestamp("2020-12-31")]]
+)
+def test_interpolate_triggers_copy(vals, func):
+    df = DataFrame({"a": vals})
+    result = getattr(df, func)()
+
+    assert not np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    # Check that we don't have references when triggering a copy
+    assert result._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize(
+    "vals", [[1, np.nan, 2], [Timestamp("2019-12-31"), NaT, Timestamp("2020-12-31")]]
+)
+def test_interpolate_inplace_no_reference_no_copy(vals):
+    df = DataFrame({"a": vals})
+    arr = get_array(df, "a")
+    df.interpolate(method="linear", inplace=True)
+
+    assert np.shares_memory(arr, get_array(df, "a"))
+    # Check that we don't have references when triggering a copy
+    assert df._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize(
+    "vals", [[1, np.nan, 2], [Timestamp("2019-12-31"), NaT, Timestamp("2020-12-31")]]
+)
+def test_interpolate_inplace_with_refs(vals):
+    df = DataFrame({"a": [1, np.nan, 2]})
+    df_orig = df.copy()
+    arr = get_array(df, "a")
+    view = df[:]
+    df.interpolate(method="linear", inplace=True)
+    # Check that copy was triggered in interpolate and that we don't
+    # have any references left
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    tm.assert_frame_equal(df_orig, view)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill"])
+@pytest.mark.parametrize("dtype", ["float64", "Float64"])
+def test_interp_fill_functions_inplace(func, dtype):
+    # Check that these takes the same code paths as interpolate
+    df = DataFrame({"a": [1, np.nan, 2]}, dtype=dtype)
+    df_orig = df.copy()
+    arr = get_array(df, "a")
+    view = df[:]
+
+    getattr(df, func)(inplace=True)
+
+    # Check that copy was triggered in interpolate and that we don't
+    # have any references left
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    tm.assert_frame_equal(df_orig, view)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+
+
+def test_interpolate_cannot_with_object_dtype():
+    df = DataFrame({"a": ["a", np.nan, "c"], "b": 1})
+    df["a"] = df["a"].astype(object)
+
+    msg = "DataFrame cannot interpolate with object dtype"
+    with pytest.raises(TypeError, match=msg):
+        df.interpolate()
+
+
+def test_interpolate_object_convert_no_op():
+    df = DataFrame({"a": ["a", "b", "c"], "b": 1})
+    df["a"] = df["a"].astype(object)
+    arr_a = get_array(df, "a")
+
+    # Now CoW makes a copy, it should not!
+    assert df._mgr._has_no_reference(0)
+    assert np.shares_memory(arr_a, get_array(df, "a"))
+
+
+def test_interpolate_object_convert_copies():
+    df = DataFrame({"a": [1, np.nan, 2.5], "b": 1})
+    arr_a = get_array(df, "a")
+    msg = "Can not interpolate with method=pad"
+    with pytest.raises(ValueError, match=msg):
+        df.interpolate(method="pad", inplace=True, downcast="infer")
+
+    assert df._mgr._has_no_reference(0)
+    assert np.shares_memory(arr_a, get_array(df, "a"))
+
+
+def test_interpolate_downcast_reference_triggers_copy():
+    df = DataFrame({"a": [1, np.nan, 2.5], "b": 1})
+    df_orig = df.copy()
+    arr_a = get_array(df, "a")
+    view = df[:]
+
+    msg = "Can not interpolate with method=pad"
+    with pytest.raises(ValueError, match=msg):
+        df.interpolate(method="pad", inplace=True, downcast="infer")
+        assert df._mgr._has_no_reference(0)
+        assert not np.shares_memory(arr_a, get_array(df, "a"))
+
+    tm.assert_frame_equal(df_orig, view)
+
+
+def test_fillna():
+    df = DataFrame({"a": [1.5, np.nan], "b": 1})
+    df_orig = df.copy()
+
+    df2 = df.fillna(5.5)
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    df2.iloc[0, 1] = 100
+    tm.assert_frame_equal(df_orig, df)
+
+
+def test_fillna_dict():
+    df = DataFrame({"a": [1.5, np.nan], "b": 1})
+    df_orig = df.copy()
+
+    df2 = df.fillna({"a": 100.5})
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.iloc[0, 1] = 100
+    tm.assert_frame_equal(df_orig, df)
+
+
+def test_fillna_inplace():
+    df = DataFrame({"a": [1.5, np.nan], "b": 1})
+    arr_a = get_array(df, "a")
+    arr_b = get_array(df, "b")
+
+    df.fillna(5.5, inplace=True)
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+    assert np.shares_memory(get_array(df, "b"), arr_b)
+    assert df._mgr._has_no_reference(0)
+    assert df._mgr._has_no_reference(1)
+
+
+def test_fillna_inplace_reference():
+    df = DataFrame({"a": [1.5, np.nan], "b": 1})
+    df_orig = df.copy()
+    arr_a = get_array(df, "a")
+    arr_b = get_array(df, "b")
+    view = df[:]
+
+    df.fillna(5.5, inplace=True)
+    assert not np.shares_memory(get_array(df, "a"), arr_a)
+    assert np.shares_memory(get_array(df, "b"), arr_b)
+    assert view._mgr._has_no_reference(0)
+    assert df._mgr._has_no_reference(0)
+    tm.assert_frame_equal(view, df_orig)
+    expected = DataFrame({"a": [1.5, 5.5], "b": 1})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_fillna_interval_inplace_reference():
+    # Set dtype explicitly to avoid implicit cast when setting nan
+    ser = Series(
+        interval_range(start=0, end=5), name="a", dtype="interval[float64, right]"
+    )
+    ser.iloc[1] = np.nan
+
+    ser_orig = ser.copy()
+    view = ser[:]
+    ser.fillna(value=Interval(left=0, right=5), inplace=True)
+
+    assert not np.shares_memory(
+        get_array(ser, "a").left.values, get_array(view, "a").left.values
+    )
+    tm.assert_series_equal(view, ser_orig)
+
+
+def test_fillna_series_empty_arg():
+    ser = Series([1, np.nan, 2])
+    ser_orig = ser.copy()
+    result = ser.fillna({})
+    assert np.shares_memory(get_array(ser), get_array(result))
+
+    ser.iloc[0] = 100.5
+    tm.assert_series_equal(ser_orig, result)
+
+
+def test_fillna_series_empty_arg_inplace():
+    ser = Series([1, np.nan, 2])
+    arr = get_array(ser)
+    ser.fillna({}, inplace=True)
+
+    assert np.shares_memory(get_array(ser), arr)
+    assert ser._mgr._has_no_reference(0)
+
+
+def test_fillna_ea_noop_shares_memory(any_numeric_ea_and_arrow_dtype):
+    df = DataFrame({"a": [1, NA, 3], "b": 1}, dtype=any_numeric_ea_and_arrow_dtype)
+    df_orig = df.copy()
+    df2 = df.fillna(100)
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert not df2._mgr._has_no_reference(1)
+    tm.assert_frame_equal(df_orig, df)
+
+    df2.iloc[0, 1] = 100
+    assert not np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert df2._mgr._has_no_reference(1)
+    assert df._mgr._has_no_reference(1)
+    tm.assert_frame_equal(df_orig, df)
+
+
+def test_fillna_inplace_ea_noop_shares_memory(any_numeric_ea_and_arrow_dtype):
+    df = DataFrame({"a": [1, NA, 3], "b": 1}, dtype=any_numeric_ea_and_arrow_dtype)
+    df_orig = df.copy()
+    view = df[:]
+    df.fillna(100, inplace=True)
+    assert not np.shares_memory(get_array(df, "a"), get_array(view, "a"))
+
+    assert np.shares_memory(get_array(df, "b"), get_array(view, "b"))
+    assert not df._mgr._has_no_reference(1)
+    assert not view._mgr._has_no_reference(1)
+
+    df.iloc[0, 1] = 100
+    tm.assert_frame_equal(df_orig, view)
+
+
+def test_fillna_chained_assignment():
+    df = DataFrame({"a": [1, np.nan, 2], "b": 1})
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        df["a"].fillna(100, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        df[["a"]].fillna(100, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("func", ["interpolate", "ffill", "bfill"])
+def test_interpolate_chained_assignment(func):
+    df = DataFrame({"a": [1, np.nan, 2], "b": 1})
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        getattr(df["a"], func)(inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        getattr(df[["a"]], func)(inplace=True)
+    tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_methods.py b/pandas/pandas/tests/copy_view/test_methods.py
new file mode 100644
index 0000000000000..250697c91ff13
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_methods.py
@@ -0,0 +1,1539 @@
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+def test_copy():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_copy = df.copy()
+
+    # the deep copy by defaults takes a shallow copy of the Index
+    assert df_copy.index is not df.index
+    assert df_copy.columns is not df.columns
+    assert df_copy.index.is_(df.index)
+    assert df_copy.columns.is_(df.columns)
+
+    # the deep copy doesn't share memory
+    assert not np.shares_memory(get_array(df_copy, "a"), get_array(df, "a"))
+    assert not df_copy._mgr.blocks[0].refs.has_reference()
+    assert not df_copy._mgr.blocks[1].refs.has_reference()
+
+    # mutating copy doesn't mutate original
+    df_copy.iloc[0, 0] = 0
+    assert df.iloc[0, 0] == 1
+
+
+def test_copy_shallow():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_copy = df.copy(deep=False)
+
+    # the shallow copy also makes a shallow copy of the index
+    assert df_copy.index is not df.index
+    assert df_copy.columns is not df.columns
+    assert df_copy.index.is_(df.index)
+    assert df_copy.columns.is_(df.columns)
+
+    # the shallow copy still shares memory
+    assert np.shares_memory(get_array(df_copy, "a"), get_array(df, "a"))
+    assert df_copy._mgr.blocks[0].refs.has_reference()
+    assert df_copy._mgr.blocks[1].refs.has_reference()
+
+    # mutating shallow copy doesn't mutate original
+    df_copy.iloc[0, 0] = 0
+    assert df.iloc[0, 0] == 1
+    # mutating triggered a copy-on-write -> no longer shares memory
+    assert not np.shares_memory(get_array(df_copy, "a"), get_array(df, "a"))
+    # but still shares memory for the other columns/blocks
+    assert np.shares_memory(get_array(df_copy, "c"), get_array(df, "c"))
+
+
+@pytest.mark.filterwarnings("ignore::DeprecationWarning")
+@pytest.mark.parametrize("copy", [True, None, False])
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df, copy: df.rename(columns=str.lower, copy=copy),
+        lambda df, copy: df.reindex(columns=["a", "c"], copy=copy),
+        lambda df, copy: df.reindex_like(df, copy=copy),
+        lambda df, copy: df.align(df, copy=copy)[0],
+        lambda df, copy: df.set_axis(["a", "b", "c"], axis="index", copy=copy),
+        lambda df, copy: df.rename_axis(index="test", copy=copy),
+        lambda df, copy: df.rename_axis(columns="test", copy=copy),
+        lambda df, copy: df.astype({"b": "int64"}, copy=copy),
+        # lambda df, copy: df.swaplevel(0, 0, copy=copy),
+        lambda df, copy: df.truncate(0, 5, copy=copy),
+        lambda df, copy: df.infer_objects(copy=copy),
+        lambda df, copy: df.to_timestamp(copy=copy),
+        lambda df, copy: df.to_period(freq="D", copy=copy),
+        lambda df, copy: df.tz_localize("US/Central", copy=copy),
+        lambda df, copy: df.tz_convert("US/Central", copy=copy),
+        lambda df, copy: df.set_flags(allows_duplicate_labels=False, copy=copy),
+    ],
+    ids=[
+        "rename",
+        "reindex",
+        "reindex_like",
+        "align",
+        "set_axis",
+        "rename_axis0",
+        "rename_axis1",
+        "astype",
+        # "swaplevel",  # only series
+        "truncate",
+        "infer_objects",
+        "to_timestamp",
+        "to_period",
+        "tz_localize",
+        "tz_convert",
+        "set_flags",
+    ],
+)
+def test_methods_copy_keyword(request, method, copy):
+    index = None
+    if "to_timestamp" in request.node.callspec.id:
+        index = period_range("2012-01-01", freq="D", periods=3)
+    elif "to_period" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3)
+    elif "tz_localize" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3)
+    elif "tz_convert" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3, tz="Europe/Brussels")
+
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]}, index=index)
+    df2 = method(df, copy=copy)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+
+@pytest.mark.filterwarnings("ignore::DeprecationWarning")
+@pytest.mark.parametrize("copy", [True, None, False])
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda ser, copy: ser.rename(index={0: 100}, copy=copy),
+        lambda ser, copy: ser.rename(None, copy=copy),
+        lambda ser, copy: ser.reindex(index=ser.index, copy=copy),
+        lambda ser, copy: ser.reindex_like(ser, copy=copy),
+        lambda ser, copy: ser.align(ser, copy=copy)[0],
+        lambda ser, copy: ser.set_axis(["a", "b", "c"], axis="index", copy=copy),
+        lambda ser, copy: ser.rename_axis(index="test", copy=copy),
+        lambda ser, copy: ser.astype("int64", copy=copy),
+        lambda ser, copy: ser.swaplevel(0, 1, copy=copy),
+        lambda ser, copy: ser.truncate(0, 5, copy=copy),
+        lambda ser, copy: ser.infer_objects(copy=copy),
+        lambda ser, copy: ser.to_timestamp(copy=copy),
+        lambda ser, copy: ser.to_period(freq="D", copy=copy),
+        lambda ser, copy: ser.tz_localize("US/Central", copy=copy),
+        lambda ser, copy: ser.tz_convert("US/Central", copy=copy),
+        lambda ser, copy: ser.set_flags(allows_duplicate_labels=False, copy=copy),
+    ],
+    ids=[
+        "rename (dict)",
+        "rename",
+        "reindex",
+        "reindex_like",
+        "align",
+        "set_axis",
+        "rename_axis0",
+        "astype",
+        "swaplevel",
+        "truncate",
+        "infer_objects",
+        "to_timestamp",
+        "to_period",
+        "tz_localize",
+        "tz_convert",
+        "set_flags",
+    ],
+)
+def test_methods_series_copy_keyword(request, method, copy):
+    index = None
+    if "to_timestamp" in request.node.callspec.id:
+        index = period_range("2012-01-01", freq="D", periods=3)
+    elif "to_period" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3)
+    elif "tz_localize" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3)
+    elif "tz_convert" in request.node.callspec.id:
+        index = date_range("2012-01-01", freq="D", periods=3, tz="Europe/Brussels")
+    elif "swaplevel" in request.node.callspec.id:
+        index = MultiIndex.from_arrays([[1, 2, 3], [4, 5, 6]])
+
+    ser = Series([1, 2, 3], index=index)
+    ser2 = method(ser, copy=copy)
+    assert np.shares_memory(get_array(ser2), get_array(ser))
+
+
+# -----------------------------------------------------------------------------
+# DataFrame methods returning new DataFrame using shallow copy
+
+
+def test_reset_index():
+    # Case: resetting the index (i.e. adding a new column) + mutating the
+    # resulting dataframe
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]}, index=[10, 11, 12]
+    )
+    df_orig = df.copy()
+    df2 = df.reset_index()
+    df2._mgr._verify_integrity()
+
+    # still shares memory (df2 is a shallow copy)
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    # mutating df2 triggers a copy-on-write for that column / block
+    df2.iloc[0, 2] = 0
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("index", [pd.RangeIndex(0, 2), Index([1, 2])])
+def test_reset_index_series_drop(index):
+    ser = Series([1, 2], index=index)
+    ser_orig = ser.copy()
+    ser2 = ser.reset_index(drop=True)
+    assert np.shares_memory(get_array(ser), get_array(ser2))
+    assert not ser._mgr._has_no_reference(0)
+
+    ser2.iloc[0] = 100
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_groupby_column_index_in_references():
+    df = DataFrame(
+        {"A": ["a", "b", "c", "d"], "B": [1, 2, 3, 4], "C": ["a", "a", "b", "b"]}
+    )
+    df = df.set_index("A")
+    key = df["C"]
+    result = df.groupby(key, observed=True).sum()
+    expected = df.groupby("C", observed=True).sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rename_columns():
+    # Case: renaming columns returns a new dataframe
+    # + afterwards modifying the result
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.rename(columns=str.upper)
+
+    assert np.shares_memory(get_array(df2, "A"), get_array(df, "a"))
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "A"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "C"), get_array(df, "c"))
+    expected = DataFrame({"A": [0, 2, 3], "B": [4, 5, 6], "C": [0.1, 0.2, 0.3]})
+    tm.assert_frame_equal(df2, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_rename_columns_modify_parent():
+    # Case: renaming columns returns a new dataframe
+    # + afterwards modifying the original (parent) dataframe
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df2 = df.rename(columns=str.upper)
+    df2_orig = df2.copy()
+
+    assert np.shares_memory(get_array(df2, "A"), get_array(df, "a"))
+    df.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "A"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "C"), get_array(df, "c"))
+    expected = DataFrame({"a": [0, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    tm.assert_frame_equal(df, expected)
+    tm.assert_frame_equal(df2, df2_orig)
+
+
+def test_pipe():
+    df = DataFrame({"a": [1, 2, 3], "b": 1.5})
+    df_orig = df.copy()
+
+    def testfunc(df):
+        return df
+
+    df2 = df.pipe(testfunc)
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column
+    df2.iloc[0, 0] = 0
+    tm.assert_frame_equal(df, df_orig)
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+
+def test_pipe_modify_df():
+    df = DataFrame({"a": [1, 2, 3], "b": 1.5})
+    df_orig = df.copy()
+
+    def testfunc(df):
+        df.iloc[0, 0] = 100
+        return df
+
+    df2 = df.pipe(testfunc)
+
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+    tm.assert_frame_equal(df, df_orig)
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+
+def test_reindex_columns():
+    # Case: reindexing the column returns a new dataframe
+    # + afterwards modifying the result
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.reindex(columns=["a", "c"])
+
+    # still shares memory (df2 is a shallow copy)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    # mutating df2 triggers a copy-on-write for that column
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        lambda idx: idx,
+        lambda idx: idx.view(),
+        lambda idx: idx.copy(),
+        lambda idx: list(idx),
+    ],
+    ids=["identical", "view", "copy", "values"],
+)
+def test_reindex_rows(index):
+    # Case: reindexing the rows with an index that matches the current index
+    # can use a shallow copy
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.reindex(index=index(df.index))
+
+    # still shares memory (df2 is a shallow copy)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    # mutating df2 triggers a copy-on-write for that column
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_drop_on_column():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.drop(columns="a")
+    df2._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_select_dtypes():
+    # Case: selecting columns using `select_dtypes()` returns a new dataframe
+    # + afterwards modifying the result
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.select_dtypes("int64")
+    df2._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    # mutating df2 triggers a copy-on-write for that column/block
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "filter_kwargs", [{"items": ["a"]}, {"like": "a"}, {"regex": "a"}]
+)
+def test_filter(filter_kwargs):
+    # Case: selecting columns using `filter()` returns a new dataframe
+    # + afterwards modifying the result
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.filter(**filter_kwargs)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column/block
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_shift_no_op():
+    df = DataFrame(
+        [[1, 2], [3, 4], [5, 6]],
+        index=date_range("2020-01-01", "2020-01-03"),
+        columns=["a", "b"],
+    )
+    df_orig = df.copy()
+    df2 = df.shift(periods=0)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    tm.assert_frame_equal(df2, df_orig)
+
+
+def test_shift_index():
+    df = DataFrame(
+        [[1, 2], [3, 4], [5, 6]],
+        index=date_range("2020-01-01", "2020-01-03"),
+        columns=["a", "b"],
+    )
+    df2 = df.shift(periods=1, axis=0)
+
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+
+def test_shift_rows_freq():
+    df = DataFrame(
+        [[1, 2], [3, 4], [5, 6]],
+        index=date_range("2020-01-01", "2020-01-03"),
+        columns=["a", "b"],
+    )
+    df_orig = df.copy()
+    df_orig.index = date_range("2020-01-02", "2020-01-04")
+    df2 = df.shift(periods=1, freq="1D")
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    df.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+    tm.assert_frame_equal(df2, df_orig)
+
+
+def test_shift_columns():
+    df = DataFrame(
+        [[1, 2], [3, 4], [5, 6]], columns=date_range("2020-01-01", "2020-01-02")
+    )
+    df2 = df.shift(periods=1, axis=1)
+
+    assert np.shares_memory(get_array(df2, "2020-01-02"), get_array(df, "2020-01-01"))
+    df.iloc[0, 0] = 0
+    assert not np.shares_memory(
+        get_array(df2, "2020-01-02"), get_array(df, "2020-01-01")
+    )
+    expected = DataFrame(
+        [[np.nan, 1], [np.nan, 3], [np.nan, 5]],
+        columns=date_range("2020-01-01", "2020-01-02"),
+    )
+    tm.assert_frame_equal(df2, expected)
+
+
+def test_pop():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    view_original = df[:]
+    result = df.pop("a")
+
+    assert np.shares_memory(result.values, get_array(view_original, "a"))
+    assert np.shares_memory(get_array(df, "b"), get_array(view_original, "b"))
+
+    result.iloc[0] = 0
+    assert not np.shares_memory(result.values, get_array(view_original, "a"))
+    df.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df, "b"), get_array(view_original, "b"))
+    tm.assert_frame_equal(view_original, df_orig)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x, y: x.align(y),
+        lambda x, y: x.align(y.a, axis=0),
+        lambda x, y: x.align(y.a.iloc[slice(0, 1)], axis=1),
+    ],
+)
+def test_align_frame(func):
+    df = DataFrame({"a": [1, 2, 3], "b": "a"})
+    df_orig = df.copy()
+    df_changed = df[["b", "a"]].copy()
+    df2, _ = func(df, df_changed)
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_align_series():
+    ser = Series([1, 2])
+    ser_orig = ser.copy()
+    ser_other = ser.copy()
+    ser2, ser_other_result = ser.align(ser_other)
+
+    assert np.shares_memory(ser2.values, ser.values)
+    assert np.shares_memory(ser_other_result.values, ser_other.values)
+    ser2.iloc[0] = 0
+    ser_other_result.iloc[0] = 0
+    assert not np.shares_memory(ser2.values, ser.values)
+    assert not np.shares_memory(ser_other_result.values, ser_other.values)
+    tm.assert_series_equal(ser, ser_orig)
+    tm.assert_series_equal(ser_other, ser_orig)
+
+
+def test_align_copy_false():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df_orig = df.copy()
+    df2, df3 = df.align(df)
+
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.loc[0, "a"] = 0
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+
+    df3.loc[0, "a"] = 0
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+
+
+def test_align_with_series_copy_false():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    ser = Series([1, 2, 3], name="x")
+    ser_orig = ser.copy()
+    df_orig = df.copy()
+    df2, ser2 = df.align(ser, axis=0)
+
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+    assert np.shares_memory(get_array(ser, "x"), get_array(ser2, "x"))
+
+    df2.loc[0, "a"] = 0
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+
+    ser2.loc[0] = 0
+    tm.assert_series_equal(ser, ser_orig)  # Original is unchanged
+
+
+def test_to_frame():
+    # Case: converting a Series to a DataFrame with to_frame
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+
+    df = ser[:].to_frame()
+
+    # currently this always returns a "view"
+    assert np.shares_memory(ser.values, get_array(df, 0))
+
+    df.iloc[0, 0] = 0
+
+    # mutating df triggers a copy-on-write for that column
+    assert not np.shares_memory(ser.values, get_array(df, 0))
+    tm.assert_series_equal(ser, ser_orig)
+
+    # modify original series -> don't modify dataframe
+    df = ser[:].to_frame()
+    ser.iloc[0] = 0
+
+    tm.assert_frame_equal(df, ser_orig.to_frame())
+
+
+@pytest.mark.parametrize(
+    "method, idx",
+    [
+        (lambda df: df.copy(deep=False).copy(deep=False), 0),
+        (lambda df: df.reset_index().reset_index(), 2),
+        (lambda df: df.rename(columns=str.upper).rename(columns=str.lower), 0),
+        (lambda df: df.copy(deep=False).select_dtypes(include="number"), 0),
+    ],
+    ids=["shallow-copy", "reset_index", "rename", "select_dtypes"],
+)
+def test_chained_methods(request, method, idx):
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    # modify df2 -> don't modify df
+    df2 = method(df)
+    df2.iloc[0, idx] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+    # modify df -> don't modify df2
+    df2 = method(df)
+    df.iloc[0, 0] = 0
+    tm.assert_frame_equal(df2.iloc[:, idx:], df_orig)
+
+
+@pytest.mark.parametrize("obj", [Series([1, 2], name="a"), DataFrame({"a": [1, 2]})])
+def test_to_timestamp(obj):
+    obj.index = Index([Period("2012-1-1", freq="D"), Period("2012-1-2", freq="D")])
+
+    obj_orig = obj.copy()
+    obj2 = obj.to_timestamp()
+
+    assert np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+
+    # mutating obj2 triggers a copy-on-write for that column / block
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+    tm.assert_equal(obj, obj_orig)
+
+
+@pytest.mark.parametrize("obj", [Series([1, 2], name="a"), DataFrame({"a": [1, 2]})])
+def test_to_period(obj):
+    obj.index = Index([Timestamp("2019-12-31"), Timestamp("2020-12-31")])
+
+    obj_orig = obj.copy()
+    obj2 = obj.to_period(freq="Y")
+
+    assert np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+
+    # mutating obj2 triggers a copy-on-write for that column / block
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+    tm.assert_equal(obj, obj_orig)
+
+
+def test_set_index():
+    # GH 49473
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.set_index("a")
+
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+    # mutating df2 triggers a copy-on-write for that column / block
+    df2.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_set_index_mutating_parent_does_not_mutate_index():
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    result = df.set_index("a")
+    expected = result.copy()
+
+    df.iloc[0, 0] = 100
+    tm.assert_frame_equal(result, expected)
+
+
+def test_add_prefix():
+    # GH 49473
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.add_prefix("CoW_")
+
+    assert np.shares_memory(get_array(df2, "CoW_a"), get_array(df, "a"))
+    df2.iloc[0, 0] = 0
+
+    assert not np.shares_memory(get_array(df2, "CoW_a"), get_array(df, "a"))
+
+    assert np.shares_memory(get_array(df2, "CoW_c"), get_array(df, "c"))
+    expected = DataFrame(
+        {"CoW_a": [0, 2, 3], "CoW_b": [4, 5, 6], "CoW_c": [0.1, 0.2, 0.3]}
+    )
+    tm.assert_frame_equal(df2, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_add_suffix():
+    # GH 49473
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.add_suffix("_CoW")
+    assert np.shares_memory(get_array(df2, "a_CoW"), get_array(df, "a"))
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a_CoW"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "c_CoW"), get_array(df, "c"))
+    expected = DataFrame(
+        {"a_CoW": [0, 2, 3], "b_CoW": [4, 5, 6], "c_CoW": [0.1, 0.2, 0.3]}
+    )
+    tm.assert_frame_equal(df2, expected)
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("axis, val", [(0, 5.5), (1, np.nan)])
+def test_dropna(axis, val):
+    df = DataFrame({"a": [1, 2, 3], "b": [4, val, 6], "c": "d"})
+    df_orig = df.copy()
+    df2 = df.dropna(axis=axis)
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("val", [5, 5.5])
+def test_dropna_series(val):
+    ser = Series([1, val, 4])
+    ser_orig = ser.copy()
+    ser2 = ser.dropna()
+    assert np.shares_memory(ser2.values, ser.values)
+
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2.values, ser.values)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda df: df.head(),
+        lambda df: df.head(2),
+        lambda df: df.tail(),
+        lambda df: df.tail(3),
+    ],
+)
+def test_head_tail(method):
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = method(df)
+    df2._mgr._verify_integrity()
+
+    # We are explicitly deviating for CoW here to make an eager copy (avoids
+    # tracking references for very cheap ops)
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+    # modify df2 to trigger CoW for that block
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_infer_objects(using_infer_string):
+    df = DataFrame(
+        {"a": [1, 2], "b": Series(["x", "y"], dtype=object), "c": 1, "d": "x"}
+    )
+    df_orig = df.copy()
+    df2 = df.infer_objects()
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    if using_infer_string and HAS_PYARROW:
+        assert not tm.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    else:
+        assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+    df2.iloc[0, 0] = 0
+    df2.iloc[0, 1] = "d"
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_infer_objects_no_reference(using_infer_string):
+    df = DataFrame(
+        {
+            "a": [1, 2],
+            "b": Series(["x", "y"], dtype=object),
+            "c": 1,
+            "d": Series(
+                [Timestamp("2019-12-31"), Timestamp("2020-12-31")], dtype="object"
+            ),
+            "e": Series(["z", "w"], dtype=object),
+        }
+    )
+    df = df.infer_objects()
+
+    arr_a = get_array(df, "a")
+    arr_b = get_array(df, "b")
+    arr_d = get_array(df, "d")
+
+    df.iloc[0, 0] = 0
+    df.iloc[0, 1] = "d"
+    df.iloc[0, 3] = Timestamp("2018-12-31")
+    assert np.shares_memory(arr_a, get_array(df, "a"))
+    if using_infer_string and HAS_PYARROW:
+        # note that the underlying memory of arr_b has been copied anyway
+        # because of the assignment, but the EA is updated inplace so still
+        # appears the share memory
+        assert tm.shares_memory(arr_b, get_array(df, "b"))
+    else:
+        # TODO(CoW): Block splitting causes references here
+        assert not np.shares_memory(arr_b, get_array(df, "b"))
+    assert np.shares_memory(arr_d, get_array(df, "d"))
+
+
+def test_infer_objects_reference():
+    df = DataFrame(
+        {
+            "a": [1, 2],
+            "b": Series(["x", "y"], dtype=object),
+            "c": 1,
+            "d": Series(
+                [Timestamp("2019-12-31"), Timestamp("2020-12-31")], dtype="object"
+            ),
+        }
+    )
+    view = df[:]  # noqa: F841
+    df = df.infer_objects()
+
+    arr_a = get_array(df, "a")
+    arr_b = get_array(df, "b")
+    arr_d = get_array(df, "d")
+
+    df.iloc[0, 0] = 0
+    df.iloc[0, 1] = "d"
+    df.iloc[0, 3] = Timestamp("2018-12-31")
+    assert not np.shares_memory(arr_a, get_array(df, "a"))
+    assert not np.shares_memory(arr_b, get_array(df, "b"))
+    assert np.shares_memory(arr_d, get_array(df, "d"))
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"before": "a", "after": "b", "axis": 1},
+        {"before": 0, "after": 1, "axis": 0},
+    ],
+)
+def test_truncate(kwargs):
+    df = DataFrame({"a": [1, 2, 3], "b": 1, "c": 2})
+    df_orig = df.copy()
+    df2 = df.truncate(**kwargs)
+    df2._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("method", ["assign", "drop_duplicates"])
+def test_assign_drop_duplicates(method):
+    df = DataFrame({"a": [1, 2, 3]})
+    df_orig = df.copy()
+    df2 = getattr(df, method)()
+    df2._mgr._verify_integrity()
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("obj", [Series([1, 2]), DataFrame({"a": [1, 2]})])
+def test_take(obj):
+    # Check that no copy is made when we take all rows in original order
+    obj_orig = obj.copy()
+    obj2 = obj.take([0, 1])
+    assert np.shares_memory(obj2.values, obj.values)
+
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(obj2.values, obj.values)
+    tm.assert_equal(obj, obj_orig)
+
+
+@pytest.mark.parametrize("obj", [Series([1, 2]), DataFrame({"a": [1, 2]})])
+def test_between_time(obj):
+    obj.index = date_range("2018-04-09", periods=2, freq="1D20min")
+    obj_orig = obj.copy()
+    obj2 = obj.between_time("0:00", "1:00")
+    assert np.shares_memory(obj2.values, obj.values)
+
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(obj2.values, obj.values)
+    tm.assert_equal(obj, obj_orig)
+
+
+def test_reindex_like():
+    df = DataFrame({"a": [1, 2], "b": "a"})
+    other = DataFrame({"b": "a", "a": [1, 2]})
+
+    df_orig = df.copy()
+    df2 = df.reindex_like(other)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 1] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_sort_index():
+    # GH 49473
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    ser2 = ser.sort_index()
+    assert np.shares_memory(ser.values, ser2.values)
+
+    # mutating ser triggers a copy-on-write for the column / block
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2.values, ser.values)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize(
+    "obj, kwargs",
+    [(Series([1, 2, 3], name="a"), {}), (DataFrame({"a": [1, 2, 3]}), {"by": "a"})],
+)
+def test_sort_values(obj, kwargs):
+    obj_orig = obj.copy()
+    obj2 = obj.sort_values(**kwargs)
+    assert np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+
+    # mutating df triggers a copy-on-write for the column / block
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(get_array(obj2, "a"), get_array(obj, "a"))
+    tm.assert_equal(obj, obj_orig)
+
+
+@pytest.mark.parametrize(
+    "obj, kwargs",
+    [(Series([1, 2, 3], name="a"), {}), (DataFrame({"a": [1, 2, 3]}), {"by": "a"})],
+)
+def test_sort_values_inplace(obj, kwargs):
+    obj_orig = obj.copy()
+    view = obj[:]
+    obj.sort_values(inplace=True, **kwargs)
+
+    assert np.shares_memory(get_array(obj, "a"), get_array(view, "a"))
+
+    # mutating obj triggers a copy-on-write for the column / block
+    obj.iloc[0] = 0
+    assert not np.shares_memory(get_array(obj, "a"), get_array(view, "a"))
+    tm.assert_equal(view, obj_orig)
+
+
+@pytest.mark.parametrize("decimals", [-1, 0, 1])
+def test_round(decimals):
+    df = DataFrame({"a": [1, 2], "b": "c"})
+    df_orig = df.copy()
+    df2 = df.round(decimals=decimals)
+
+    assert tm.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    # TODO: Make inplace by using out parameter of ndarray.round?
+    if decimals >= 0:
+        # Ensure lazy copy if no-op
+        assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    else:
+        assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 1] = "d"
+    df2.iloc[0, 0] = 4
+    assert not np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_reorder_levels():
+    index = MultiIndex.from_tuples(
+        [(1, 1), (1, 2), (2, 1), (2, 2)], names=["one", "two"]
+    )
+    df = DataFrame({"a": [1, 2, 3, 4]}, index=index)
+    df_orig = df.copy()
+    df2 = df.reorder_levels(order=["two", "one"])
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_series_reorder_levels():
+    index = MultiIndex.from_tuples(
+        [(1, 1), (1, 2), (2, 1), (2, 2)], names=["one", "two"]
+    )
+    ser = Series([1, 2, 3, 4], index=index)
+    ser_orig = ser.copy()
+    ser2 = ser.reorder_levels(order=["two", "one"])
+    assert np.shares_memory(ser2.values, ser.values)
+
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2.values, ser.values)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize("obj", [Series([1, 2, 3]), DataFrame({"a": [1, 2, 3]})])
+def test_swaplevel(obj):
+    index = MultiIndex.from_tuples([(1, 1), (1, 2), (2, 1)], names=["one", "two"])
+    obj.index = index
+    obj_orig = obj.copy()
+    obj2 = obj.swaplevel()
+    assert np.shares_memory(obj2.values, obj.values)
+
+    obj2.iloc[0] = 0
+    assert not np.shares_memory(obj2.values, obj.values)
+    tm.assert_equal(obj, obj_orig)
+
+
+def test_frame_set_axis():
+    # GH 49473
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+    df2 = df.set_axis(["a", "b", "c"], axis="index")
+
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column / block
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_series_set_axis():
+    # GH 49473
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    ser2 = ser.set_axis(["a", "b", "c"], axis="index")
+    assert np.shares_memory(ser, ser2)
+
+    # mutating ser triggers a copy-on-write for the column / block
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2, ser)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_set_flags():
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    ser2 = ser.set_flags(allows_duplicate_labels=False)
+
+    assert np.shares_memory(ser, ser2)
+
+    # mutating ser triggers a copy-on-write for the column / block
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2, ser)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize("kwargs", [{"mapper": "test"}, {"index": "test"}])
+def test_rename_axis(kwargs):
+    df = DataFrame({"a": [1, 2, 3, 4]}, index=Index([1, 2, 3, 4], name="a"))
+    df_orig = df.copy()
+    df2 = df.rename_axis(**kwargs)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    df2.iloc[0, 0] = 0
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize(
+    "func, tz", [("tz_convert", "Europe/Berlin"), ("tz_localize", None)]
+)
+def test_tz_convert_localize(func, tz):
+    # GH 49473
+    ser = Series(
+        [1, 2], index=date_range(start="2014-08-01 09:00", freq="h", periods=2, tz=tz)
+    )
+    ser_orig = ser.copy()
+    ser2 = getattr(ser, func)("US/Central")
+    assert np.shares_memory(ser.values, ser2.values)
+
+    # mutating ser triggers a copy-on-write for the column / block
+    ser2.iloc[0] = 0
+    assert not np.shares_memory(ser2.values, ser.values)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_droplevel():
+    # GH 49473
+    index = MultiIndex.from_tuples([(1, 1), (1, 2), (2, 1)], names=["one", "two"])
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]}, index=index)
+    df_orig = df.copy()
+    df2 = df.droplevel(0)
+
+    assert np.shares_memory(get_array(df2, "c"), get_array(df, "c"))
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column / block
+    df2.iloc[0, 0] = 0
+
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert np.shares_memory(get_array(df2, "b"), get_array(df, "b"))
+
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_squeeze():
+    df = DataFrame({"a": [1, 2, 3]})
+    df_orig = df.copy()
+    series = df.squeeze()
+
+    # Should share memory regardless of CoW since squeeze is just an iloc
+    assert np.shares_memory(series.values, get_array(df, "a"))
+
+    # mutating squeezed df triggers a copy-on-write for that column/block
+    series.iloc[0] = 0
+    assert not np.shares_memory(series.values, get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_items():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+    df_orig = df.copy()
+
+    # Test this twice, since the second time, the item cache will be
+    # triggered, and we want to make sure it still works then.
+    for i in range(2):
+        for name, ser in df.items():
+            assert np.shares_memory(get_array(ser, name), get_array(df, name))
+
+            # mutating df triggers a copy-on-write for that column / block
+            ser.iloc[0] = 0
+
+            assert not np.shares_memory(get_array(ser, name), get_array(df, name))
+            tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+def test_putmask(dtype):
+    df = DataFrame({"a": [1, 2], "b": 1, "c": 2}, dtype=dtype)
+    view = df[:]
+    df_orig = df.copy()
+    df[df == df] = 5
+
+    assert not np.shares_memory(get_array(view, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(view, df_orig)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+def test_putmask_no_reference(dtype):
+    df = DataFrame({"a": [1, 2], "b": 1, "c": 2}, dtype=dtype)
+    arr_a = get_array(df, "a")
+    df[df == df] = 5
+    assert np.shares_memory(arr_a, get_array(df, "a"))
+
+
+@pytest.mark.parametrize("dtype", ["float64", "Float64"])
+def test_putmask_aligns_rhs_no_reference(dtype):
+    df = DataFrame({"a": [1.5, 2], "b": 1.5}, dtype=dtype)
+    arr_a = get_array(df, "a")
+    df[df == df] = DataFrame({"a": [5.5, 5]})
+    assert np.shares_memory(arr_a, get_array(df, "a"))
+
+
+@pytest.mark.parametrize("val, exp, raises", [(5.5, True, True), (5, False, False)])
+def test_putmask_dont_copy_some_blocks(val, exp, raises: bool):
+    df = DataFrame({"a": [1, 2], "b": 1, "c": 1.5})
+    view = df[:]
+    df_orig = df.copy()
+    indexer = DataFrame(
+        [[True, False, False], [True, False, False]], columns=list("abc")
+    )
+    if raises:
+        with pytest.raises(TypeError, match="Invalid value"):
+            df[indexer] = val
+    else:
+        df[indexer] = val
+        assert not np.shares_memory(get_array(view, "a"), get_array(df, "a"))
+        # TODO(CoW): Could split blocks to avoid copying the whole block
+        assert np.shares_memory(get_array(view, "b"), get_array(df, "b")) is exp
+        assert np.shares_memory(get_array(view, "c"), get_array(df, "c"))
+        assert df._mgr._has_no_reference(1) is not exp
+        assert not df._mgr._has_no_reference(2)
+        tm.assert_frame_equal(view, df_orig)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda ser: ser.where(ser > 0, 10),
+        lambda ser: ser.mask(ser <= 0, 10),
+    ],
+)
+def test_where_mask_noop(dtype, func):
+    ser = Series([1, 2, 3], dtype=dtype)
+    ser_orig = ser.copy()
+
+    result = func(ser)
+    assert np.shares_memory(get_array(ser), get_array(result))
+
+    result.iloc[0] = 10
+    assert not np.shares_memory(get_array(ser), get_array(result))
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda ser: ser.where(ser < 0, 10),
+        lambda ser: ser.mask(ser >= 0, 10),
+    ],
+)
+def test_where_mask(dtype, func):
+    ser = Series([1, 2, 3], dtype=dtype)
+    ser_orig = ser.copy()
+
+    result = func(ser)
+
+    assert not np.shares_memory(get_array(ser), get_array(result))
+    tm.assert_series_equal(ser, ser_orig)
+
+
+@pytest.mark.parametrize("dtype, val", [("int64", 10.5), ("Int64", 10)])
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda df, val: df.where(df < 0, val),
+        lambda df, val: df.mask(df >= 0, val),
+    ],
+)
+def test_where_mask_noop_on_single_column(dtype, val, func):
+    df = DataFrame({"a": [1, 2, 3], "b": [-4, -5, -6]}, dtype=dtype)
+    df_orig = df.copy()
+
+    result = func(df, val)
+    assert np.shares_memory(get_array(df, "b"), get_array(result, "b"))
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+
+    result.iloc[0, 1] = 10
+    assert not np.shares_memory(get_array(df, "b"), get_array(result, "b"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("func", ["mask", "where"])
+def test_chained_where_mask(func):
+    df = DataFrame({"a": [1, 4, 2], "b": 1})
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        getattr(df["a"], func)(df["a"] > 2, 5, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        getattr(df[["a"]], func)(df["a"] > 2, 5, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_asfreq_noop():
+    df = DataFrame(
+        {"a": [0.0, None, 2.0, 3.0]},
+        index=date_range("1/1/2000", periods=4, freq="min"),
+    )
+    df_orig = df.copy()
+    df2 = df.asfreq(freq="min")
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+    # mutating df2 triggers a copy-on-write for that column / block
+    df2.iloc[0, 0] = 0
+
+    assert not np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_iterrows():
+    df = DataFrame({"a": 0, "b": 1}, index=[1, 2, 3])
+    df_orig = df.copy()
+
+    for _, sub in df.iterrows():
+        sub.iloc[0] = 100
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_interpolate_creates_copy():
+    # GH#51126
+    df = DataFrame({"a": [1.5, np.nan, 3]})
+    view = df[:]
+    expected = df.copy()
+
+    df.ffill(inplace=True)
+    df.iloc[0, 0] = 100.5
+    tm.assert_frame_equal(view, expected)
+
+
+def test_isetitem():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+    df_orig = df.copy()
+    df2 = df.copy(deep=False)  # Trigger a CoW
+    df2.isetitem(1, np.array([-1, -2, -3]))  # This is inplace
+    assert np.shares_memory(get_array(df, "c"), get_array(df2, "c"))
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.loc[0, "a"] = 0
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+    assert np.shares_memory(get_array(df, "c"), get_array(df2, "c"))
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_isetitem_series(dtype):
+    df = DataFrame({"a": [1, 2, 3], "b": np.array([4, 5, 6], dtype=dtype)})
+    ser = Series([7, 8, 9])
+    ser_orig = ser.copy()
+    df.isetitem(0, ser)
+
+    assert np.shares_memory(get_array(df, "a"), get_array(ser))
+    assert not df._mgr._has_no_reference(0)
+
+    # mutating dataframe doesn't update series
+    df.loc[0, "a"] = 0
+    tm.assert_series_equal(ser, ser_orig)
+
+    # mutating series doesn't update dataframe
+    df = DataFrame({"a": [1, 2, 3], "b": np.array([4, 5, 6], dtype=dtype)})
+    ser = Series([7, 8, 9])
+    df.isetitem(0, ser)
+
+    ser.loc[0] = 0
+    expected = DataFrame({"a": [7, 8, 9], "b": np.array([4, 5, 6], dtype=dtype)})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_isetitem_frame():
+    df = DataFrame({"a": [1, 2, 3], "b": 1, "c": 2})
+    rhs = DataFrame({"a": [4, 5, 6], "b": 2})
+    df.isetitem([0, 1], rhs)
+    assert np.shares_memory(get_array(df, "a"), get_array(rhs, "a"))
+    assert np.shares_memory(get_array(df, "b"), get_array(rhs, "b"))
+    assert not df._mgr._has_no_reference(0)
+    expected = df.copy()
+    rhs.iloc[0, 0] = 100
+    rhs.iloc[0, 1] = 100
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("key", ["a", ["a"]])
+def test_get(key):
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df_orig = df.copy()
+
+    result = df.get(key)
+
+    assert np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    result.iloc[0] = 0
+    assert not np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("axis, key", [(0, 0), (1, "a")])
+@pytest.mark.parametrize(
+    "dtype", ["int64", "float64"], ids=["single-block", "mixed-block"]
+)
+def test_xs(axis, key, dtype):
+    single_block = dtype == "int64"
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": [4, 5, 6], "c": np.array([7, 8, 9], dtype=dtype)}
+    )
+    df_orig = df.copy()
+
+    result = df.xs(key, axis=axis)
+
+    if axis == 1 or single_block:
+        assert np.shares_memory(get_array(df, "a"), get_array(result))
+    else:
+        assert result._mgr._has_no_reference(0)
+
+    result.iloc[0] = 0
+    tm.assert_frame_equal(df, df_orig)
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+@pytest.mark.parametrize("key, level", [("l1", 0), (2, 1)])
+def test_xs_multiindex(key, level, axis):
+    arr = np.arange(18).reshape(6, 3)
+    index = MultiIndex.from_product([["l1", "l2"], [1, 2, 3]], names=["lev1", "lev2"])
+    df = DataFrame(arr, index=index, columns=list("abc"))
+    if axis == 1:
+        df = df.transpose().copy()
+    df_orig = df.copy()
+
+    result = df.xs(key, level=level, axis=axis)
+
+    if level == 0:
+        assert np.shares_memory(
+            get_array(df, df.columns[0]), get_array(result, result.columns[0])
+        )
+    result.iloc[0, 0] = 0
+
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_update_frame():
+    df1 = DataFrame({"a": [1.0, 2.0, 3.0], "b": [4.0, 5.0, 6.0]})
+    df2 = DataFrame({"b": [100.0]}, index=[1])
+    df1_orig = df1.copy()
+    view = df1[:]
+    df1.update(df2)
+
+    expected = DataFrame({"a": [1.0, 2.0, 3.0], "b": [4.0, 100.0, 6.0]})
+    tm.assert_frame_equal(df1, expected)
+    # df1 is updated, but its view not
+    tm.assert_frame_equal(view, df1_orig)
+    assert np.shares_memory(get_array(df1, "a"), get_array(view, "a"))
+    assert not np.shares_memory(get_array(df1, "b"), get_array(view, "b"))
+
+
+def test_update_series():
+    ser1 = Series([1.0, 2.0, 3.0])
+    ser2 = Series([100.0], index=[1])
+    ser1_orig = ser1.copy()
+    view = ser1[:]
+
+    ser1.update(ser2)
+
+    expected = Series([1.0, 100.0, 3.0])
+    tm.assert_series_equal(ser1, expected)
+    # ser1 is updated, but its view not
+    tm.assert_series_equal(view, ser1_orig)
+
+
+def test_update_chained_assignment():
+    df = DataFrame({"a": [1, 2, 3]})
+    ser2 = Series([100.0], index=[1])
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        df["a"].update(ser2)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        df[["a"]].update(ser2.to_frame())
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_inplace_arithmetic_series():
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    data = get_array(ser)
+    ser *= 2
+    # https://github.com/pandas-dev/pandas/pull/55745
+    # changed to NOT update inplace because there is no benefit (actual
+    # operation already done non-inplace). This was only for the optics
+    # of updating the backing array inplace, but we no longer want to make
+    # that guarantee
+    assert not np.shares_memory(get_array(ser), data)
+    tm.assert_numpy_array_equal(data, get_array(ser_orig))
+
+
+def test_inplace_arithmetic_series_with_reference():
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    view = ser[:]
+    ser *= 2
+    assert not np.shares_memory(get_array(ser), get_array(view))
+    tm.assert_series_equal(ser_orig, view)
+
+
+def test_transpose():
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    df_orig = df.copy()
+    result = df.transpose()
+    assert np.shares_memory(get_array(df, "a"), get_array(result, 0))
+
+    result.iloc[0, 0] = 100
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_transpose_different_dtypes():
+    df = DataFrame({"a": [1, 2, 3], "b": 1.5})
+    df_orig = df.copy()
+    result = df.T
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, 0))
+    result.iloc[0, 0] = 100
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_transpose_ea_single_column():
+    df = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    result = df.T
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, 0))
+
+
+def test_transform_frame():
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    df_orig = df.copy()
+
+    def func(ser):
+        ser.iloc[0] = 100
+        return ser
+
+    df.transform(func)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_transform_series():
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+
+    def func(ser):
+        ser.iloc[0] = 100
+        return ser
+
+    ser.transform(func)
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_count_read_only_array():
+    df = DataFrame({"a": [1, 2], "b": 3})
+    result = df.count()
+    result.iloc[0] = 100
+    expected = Series([100, 2], index=["a", "b"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_insert_series():
+    df = DataFrame({"a": [1, 2, 3]})
+    ser = Series([1, 2, 3])
+    ser_orig = ser.copy()
+    df.insert(loc=1, value=ser, column="b")
+    assert np.shares_memory(get_array(ser), get_array(df, "b"))
+    assert not df._mgr._has_no_reference(1)
+
+    df.iloc[0, 1] = 100
+    tm.assert_series_equal(ser, ser_orig)
+
+
+def test_eval():
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    df_orig = df.copy()
+
+    result = df.eval("c = a+b")
+    assert np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+
+    result.iloc[0, 0] = 100
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_eval_inplace():
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    df_orig = df.copy()
+    df_view = df[:]
+
+    df.eval("c = a+b", inplace=True)
+    assert np.shares_memory(get_array(df, "a"), get_array(df_view, "a"))
+
+    df.iloc[0, 0] = 100
+    tm.assert_frame_equal(df_view, df_orig)
+
+
+def test_apply_modify_row():
+    # Case: applying a function on each row as a Series object, where the
+    # function mutates the row object (which needs to trigger CoW if row is a view)
+    df = DataFrame({"A": [1, 2], "B": [3, 4]})
+    df_orig = df.copy()
+
+    def transform(row):
+        row["B"] = 100
+        return row
+
+    df.apply(transform, axis=1)
+
+    tm.assert_frame_equal(df, df_orig)
+
+    # row Series is a copy
+    df = DataFrame({"A": [1, 2], "B": ["b", "c"]})
+    df_orig = df.copy()
+
+    with tm.assert_produces_warning(None):
+        df.apply(transform, axis=1)
+
+    tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_replace.py b/pandas/pandas/tests/copy_view/test_replace.py
new file mode 100644
index 0000000000000..d4838a5e68ab8
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_replace.py
@@ -0,0 +1,356 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+
+@pytest.mark.parametrize(
+    "replace_kwargs",
+    [
+        {"to_replace": {"a": 1, "b": 4}, "value": -1},
+        # Test CoW splits blocks to avoid copying unchanged columns
+        {"to_replace": {"a": 1}, "value": -1},
+        {"to_replace": {"b": 4}, "value": -1},
+        {"to_replace": {"b": {4: 1}}},
+        # TODO: Add these in a further optimization
+        # We would need to see which columns got replaced in the mask
+        # which could be expensive
+        # {"to_replace": {"b": 1}},
+        # 1
+    ],
+)
+def test_replace(replace_kwargs):
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [0.1, 0.2, 0.3]})
+    df_orig = df.copy()
+
+    df_replaced = df.replace(**replace_kwargs)
+
+    if (df_replaced["b"] == df["b"]).all():
+        assert np.shares_memory(get_array(df_replaced, "b"), get_array(df, "b"))
+    assert tm.shares_memory(get_array(df_replaced, "c"), get_array(df, "c"))
+
+    # mutating squeezed df triggers a copy-on-write for that column/block
+    df_replaced.loc[0, "c"] = -1
+    assert not np.shares_memory(get_array(df_replaced, "c"), get_array(df, "c"))
+
+    if "a" in replace_kwargs["to_replace"]:
+        arr = get_array(df_replaced, "a")
+        df_replaced.loc[0, "a"] = 100
+        assert np.shares_memory(get_array(df_replaced, "a"), arr)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_replace_regex_inplace_refs():
+    df = DataFrame({"a": ["aaa", "bbb"]})
+    df_orig = df.copy()
+    view = df[:]
+    arr = get_array(df, "a")
+    df.replace(to_replace=r"^a.*$", value="new", inplace=True, regex=True)
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    assert df._mgr._has_no_reference(0)
+    tm.assert_frame_equal(view, df_orig)
+
+
+def test_replace_regex_inplace():
+    df = DataFrame({"a": ["aaa", "bbb"]})
+    arr = get_array(df, "a")
+    df.replace(to_replace=r"^a.*$", value="new", inplace=True, regex=True)
+    assert df._mgr._has_no_reference(0)
+    assert tm.shares_memory(arr, get_array(df, "a"))
+
+    df_orig = df.copy()
+    df2 = df.replace(to_replace=r"^b.*$", value="new", regex=True)
+    tm.assert_frame_equal(df_orig, df)
+    assert not tm.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+
+def test_replace_regex_inplace_no_op():
+    df = DataFrame({"a": [1, 2]})
+    arr = get_array(df, "a")
+    df.replace(to_replace=r"^a.$", value="new", inplace=True, regex=True)
+    assert df._mgr._has_no_reference(0)
+    assert np.shares_memory(arr, get_array(df, "a"))
+
+    df_orig = df.copy()
+    df2 = df.replace(to_replace=r"^x.$", value="new", regex=True)
+    tm.assert_frame_equal(df_orig, df)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+
+
+def test_replace_mask_all_false_second_block():
+    df = DataFrame({"a": [1.5, 2, 3], "b": 100.5, "c": 1, "d": 2})
+    df_orig = df.copy()
+
+    df2 = df.replace(to_replace=1.5, value=55.5)
+
+    # TODO: Block splitting would allow us to avoid copying b
+    assert np.shares_memory(get_array(df, "c"), get_array(df2, "c"))
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.loc[0, "c"] = 1
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+
+    assert not np.shares_memory(get_array(df, "c"), get_array(df2, "c"))
+    assert np.shares_memory(get_array(df, "d"), get_array(df2, "d"))
+
+
+def test_replace_coerce_single_column():
+    df = DataFrame({"a": [1.5, 2, 3], "b": 100.5})
+    df_orig = df.copy()
+
+    df2 = df.replace(to_replace=1.5, value="a")
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.loc[0, "b"] = 0.5
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+    assert not np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+
+
+def test_replace_to_replace_wrong_dtype():
+    df = DataFrame({"a": [1.5, 2, 3], "b": 100.5})
+    df_orig = df.copy()
+
+    df2 = df.replace(to_replace="xxx", value=1.5)
+
+    assert np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+    assert np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    df2.loc[0, "b"] = 0.5
+    tm.assert_frame_equal(df, df_orig)  # Original is unchanged
+    assert not np.shares_memory(get_array(df, "b"), get_array(df2, "b"))
+
+
+def test_replace_list_categorical():
+    df = DataFrame({"a": ["a", "b", "c"]}, dtype="category")
+    arr = get_array(df, "a")
+
+    df.replace(["c"], value="a", inplace=True)
+    assert np.shares_memory(arr.codes, get_array(df, "a").codes)
+    assert df._mgr._has_no_reference(0)
+
+    df_orig = df.copy()
+    df.replace(["b"], value="a")
+    df2 = df.apply(lambda x: x.cat.rename_categories({"b": "d"}))
+    assert not np.shares_memory(arr.codes, get_array(df2, "a").codes)
+
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_replace_list_inplace_refs_categorical():
+    df = DataFrame({"a": ["a", "b", "c"]}, dtype="category")
+    view = df[:]
+    df_orig = df.copy()
+    df.replace(["c"], value="a", inplace=True)
+    tm.assert_frame_equal(df_orig, view)
+
+
+@pytest.mark.parametrize("to_replace", [1.5, [1.5], []])
+def test_replace_inplace(to_replace):
+    df = DataFrame({"a": [1.5, 2, 3]})
+    arr_a = get_array(df, "a")
+    df.replace(to_replace=1.5, value=15.5, inplace=True)
+
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+    assert df._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("to_replace", [1.5, [1.5]])
+def test_replace_inplace_reference(to_replace):
+    df = DataFrame({"a": [1.5, 2, 3]})
+    arr_a = get_array(df, "a")
+    view = df[:]
+    df.replace(to_replace=to_replace, value=15.5, inplace=True)
+
+    assert not np.shares_memory(get_array(df, "a"), arr_a)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("to_replace", ["a", 100.5])
+def test_replace_inplace_reference_no_op(to_replace):
+    df = DataFrame({"a": [1.5, 2, 3]})
+    arr_a = get_array(df, "a")
+    view = df[:]
+    df.replace(to_replace=to_replace, value=15.5, inplace=True)
+
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+    assert not df._mgr._has_no_reference(0)
+    assert not view._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("to_replace", [1, [1]])
+def test_replace_categorical_inplace_reference(to_replace):
+    df = DataFrame({"a": Categorical([1, 2, 3])})
+    df_orig = df.copy()
+    arr_a = get_array(df, "a")
+    view = df[:]
+    df.replace(to_replace=to_replace, value=1, inplace=True)
+    assert not np.shares_memory(get_array(df, "a").codes, arr_a.codes)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+    tm.assert_frame_equal(view, df_orig)
+
+
+def test_replace_categorical_inplace():
+    df = DataFrame({"a": Categorical([1, 2, 3])})
+    arr_a = get_array(df, "a")
+    df.replace(to_replace=1, value=1, inplace=True)
+
+    assert np.shares_memory(get_array(df, "a").codes, arr_a.codes)
+    assert df._mgr._has_no_reference(0)
+
+    expected = DataFrame({"a": Categorical([1, 2, 3])})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_replace_categorical():
+    df = DataFrame({"a": Categorical([1, 2, 3])})
+    df_orig = df.copy()
+    df2 = df.replace(to_replace=1, value=1)
+
+    assert df._mgr._has_no_reference(0)
+    assert df2._mgr._has_no_reference(0)
+    assert not np.shares_memory(get_array(df, "a").codes, get_array(df2, "a").codes)
+    tm.assert_frame_equal(df, df_orig)
+
+    arr_a = get_array(df2, "a").codes
+    df2.iloc[0, 0] = 2.0
+    assert np.shares_memory(get_array(df2, "a").codes, arr_a)
+
+
+@pytest.mark.parametrize("method", ["where", "mask"])
+def test_masking_inplace(method):
+    df = DataFrame({"a": [1.5, 2, 3]})
+    df_orig = df.copy()
+    arr_a = get_array(df, "a")
+    view = df[:]
+
+    method = getattr(df, method)
+    method(df["a"] > 1.6, -1, inplace=True)
+
+    assert not np.shares_memory(get_array(df, "a"), arr_a)
+    assert df._mgr._has_no_reference(0)
+    assert view._mgr._has_no_reference(0)
+    tm.assert_frame_equal(view, df_orig)
+
+
+def test_replace_empty_list():
+    df = DataFrame({"a": [1, 2]})
+
+    df2 = df.replace([], [])
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    assert not df._mgr._has_no_reference(0)
+    arr_a = get_array(df, "a")
+    df.replace([], [])
+    assert np.shares_memory(get_array(df, "a"), arr_a)
+    assert not df._mgr._has_no_reference(0)
+    assert not df2._mgr._has_no_reference(0)
+
+
+@pytest.mark.parametrize("value", ["d", None])
+def test_replace_object_list_inplace(value):
+    df = DataFrame({"a": ["a", "b", "c"]}, dtype=object)
+    arr = get_array(df, "a")
+    df.replace(["c"], value, inplace=True)
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert df._mgr._has_no_reference(0)
+
+
+def test_replace_list_multiple_elements_inplace():
+    df = DataFrame({"a": [1, 2, 3]})
+    arr = get_array(df, "a")
+    df.replace([1, 2], 4, inplace=True)
+    assert np.shares_memory(arr, get_array(df, "a"))
+    assert df._mgr._has_no_reference(0)
+
+
+def test_replace_list_none():
+    df = DataFrame({"a": ["a", "b", "c"]})
+
+    df_orig = df.copy()
+    df2 = df.replace(["b"], value=None)
+    tm.assert_frame_equal(df, df_orig)
+
+    assert not np.shares_memory(get_array(df, "a"), get_array(df2, "a"))
+
+    # replace multiple values that don't actually replace anything with None
+    # https://github.com/pandas-dev/pandas/issues/59770
+    df3 = df.replace(["d", "e", "f"], value=None)
+    tm.assert_frame_equal(df3, df_orig)
+    assert tm.shares_memory(get_array(df, "a"), get_array(df3, "a"))
+
+
+def test_replace_list_none_inplace_refs():
+    df = DataFrame({"a": ["a", "b", "c"]})
+    arr = get_array(df, "a")
+    df_orig = df.copy()
+    view = df[:]
+    df.replace(["a"], value=None, inplace=True)
+    assert df._mgr._has_no_reference(0)
+    assert not np.shares_memory(arr, get_array(df, "a"))
+    tm.assert_frame_equal(df_orig, view)
+
+
+def test_replace_columnwise_no_op_inplace():
+    df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3]})
+    view = df[:]
+    df_orig = df.copy()
+    df.replace({"a": 10}, 100, inplace=True)
+    assert np.shares_memory(get_array(view, "a"), get_array(df, "a"))
+    df.iloc[0, 0] = 100
+    tm.assert_frame_equal(view, df_orig)
+
+
+def test_replace_columnwise_no_op():
+    df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3]})
+    df_orig = df.copy()
+    df2 = df.replace({"a": 10}, 100)
+    assert np.shares_memory(get_array(df2, "a"), get_array(df, "a"))
+    df2.iloc[0, 0] = 100
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_replace_chained_assignment():
+    df = DataFrame({"a": [1, np.nan, 2], "b": 1})
+    df_orig = df.copy()
+    with tm.raises_chained_assignment_error():
+        df["a"].replace(1, 100, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+    with tm.raises_chained_assignment_error():
+        df[["a"]].replace(1, 100, inplace=True)
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_replace_listlike():
+    df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3]})
+    df_orig = df.copy()
+
+    result = df.replace([200, 201], [11, 11])
+    assert np.shares_memory(get_array(result, "a"), get_array(df, "a"))
+
+    result.iloc[0, 0] = 100
+    tm.assert_frame_equal(df, df)
+
+    result = df.replace([200, 2], [10, 10])
+    assert not np.shares_memory(get_array(df, "a"), get_array(result, "a"))
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_replace_listlike_inplace():
+    df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3]})
+    arr = get_array(df, "a")
+    df.replace([200, 2], [10, 11], inplace=True)
+    assert np.shares_memory(get_array(df, "a"), arr)
+
+    view = df[:]
+    df_orig = df.copy()
+    df.replace([200, 3], [10, 11], inplace=True)
+    assert not np.shares_memory(get_array(df, "a"), arr)
+    tm.assert_frame_equal(view, df_orig)
diff --git a/pandas/pandas/tests/copy_view/test_setitem.py b/pandas/pandas/tests/copy_view/test_setitem.py
new file mode 100644
index 0000000000000..2f28e9826c7a1
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_setitem.py
@@ -0,0 +1,142 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.copy_view.util import get_array
+
+# -----------------------------------------------------------------------------
+# Copy/view behaviour for the values that are set in a DataFrame
+
+
+def test_set_column_with_array():
+    # Case: setting an array as a new column (df[col] = arr) copies that data
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    arr = np.array([1, 2, 3], dtype="int64")
+
+    df["c"] = arr
+
+    # the array data is copied
+    assert not np.shares_memory(get_array(df, "c"), arr)
+    # and thus modifying the array does not modify the DataFrame
+    arr[0] = 0
+    tm.assert_series_equal(df["c"], Series([1, 2, 3], name="c"))
+
+
+def test_set_column_with_series():
+    # Case: setting a series as a new column (df[col] = s) copies that data
+    # (with delayed copy with CoW)
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    ser = Series([1, 2, 3])
+
+    df["c"] = ser
+
+    assert np.shares_memory(get_array(df, "c"), get_array(ser))
+
+    # and modifying the series does not modify the DataFrame
+    ser.iloc[0] = 0
+    assert ser.iloc[0] == 0
+    tm.assert_series_equal(df["c"], Series([1, 2, 3], name="c"))
+
+
+def test_set_column_with_index():
+    # Case: setting an index as a new column (df[col] = idx) copies that data
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    idx = Index([1, 2, 3])
+
+    df["c"] = idx
+
+    # the index data is copied
+    assert not np.shares_memory(get_array(df, "c"), idx.values)
+
+    idx = RangeIndex(1, 4)
+    arr = idx.values
+
+    df["d"] = idx
+
+    assert not np.shares_memory(get_array(df, "d"), arr)
+
+
+def test_set_columns_with_dataframe():
+    # Case: setting a DataFrame as new columns copies that data
+    # (with delayed copy with CoW)
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df2 = DataFrame({"c": [7, 8, 9], "d": [10, 11, 12]})
+
+    df[["c", "d"]] = df2
+
+    assert np.shares_memory(get_array(df, "c"), get_array(df2, "c"))
+    # and modifying the set DataFrame does not modify the original DataFrame
+    df2.iloc[0, 0] = 0
+    tm.assert_series_equal(df["c"], Series([7, 8, 9], name="c"))
+
+
+def test_setitem_series_no_copy():
+    # Case: setting a Series as column into a DataFrame can delay copying that data
+    df = DataFrame({"a": [1, 2, 3]})
+    rhs = Series([4, 5, 6])
+    rhs_orig = rhs.copy()
+
+    # adding a new column
+    df["b"] = rhs
+    assert np.shares_memory(get_array(rhs), get_array(df, "b"))
+
+    df.iloc[0, 1] = 100
+    tm.assert_series_equal(rhs, rhs_orig)
+
+
+def test_setitem_series_no_copy_single_block():
+    # Overwriting an existing column that is a single block
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+    rhs = Series([4, 5, 6])
+    rhs_orig = rhs.copy()
+
+    df["a"] = rhs
+    assert np.shares_memory(get_array(rhs), get_array(df, "a"))
+
+    df.iloc[0, 0] = 100
+    tm.assert_series_equal(rhs, rhs_orig)
+
+
+def test_setitem_series_no_copy_split_block():
+    # Overwriting an existing column that is part of a larger block
+    df = DataFrame({"a": [1, 2, 3], "b": 1})
+    rhs = Series([4, 5, 6])
+    rhs_orig = rhs.copy()
+
+    df["b"] = rhs
+    assert np.shares_memory(get_array(rhs), get_array(df, "b"))
+
+    df.iloc[0, 1] = 100
+    tm.assert_series_equal(rhs, rhs_orig)
+
+
+def test_setitem_series_column_midx_broadcasting():
+    # Setting a Series to multiple columns will repeat the data
+    # (currently copying the data eagerly)
+    df = DataFrame(
+        [[1, 2, 3], [3, 4, 5]],
+        columns=MultiIndex.from_arrays([["a", "a", "b"], [1, 2, 3]]),
+    )
+    rhs = Series([10, 11])
+    df["a"] = rhs
+    assert not np.shares_memory(get_array(rhs), df._get_column_array(0))
+    assert df._mgr._has_no_reference(0)
+
+
+def test_set_column_with_inplace_operator():
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+    # this should not raise any warning
+    with tm.assert_produces_warning(None):
+        df["a"] += 1
+
+    # when it is not in a chain, then it should produce a warning
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    ser = df["a"]
+    ser += 1
diff --git a/pandas/pandas/tests/copy_view/test_util.py b/pandas/pandas/tests/copy_view/test_util.py
new file mode 100644
index 0000000000000..ff55330d70b28
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/test_util.py
@@ -0,0 +1,14 @@
+import numpy as np
+
+from pandas import DataFrame
+from pandas.tests.copy_view.util import get_array
+
+
+def test_get_array_numpy():
+    df = DataFrame({"a": [1, 2, 3]})
+    assert np.shares_memory(get_array(df, "a"), get_array(df, "a"))
+
+
+def test_get_array_masked():
+    df = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    assert np.shares_memory(get_array(df, "a"), get_array(df, "a"))
diff --git a/pandas/pandas/tests/copy_view/util.py b/pandas/pandas/tests/copy_view/util.py
new file mode 100644
index 0000000000000..9693344249365
--- /dev/null
+++ b/pandas/pandas/tests/copy_view/util.py
@@ -0,0 +1,30 @@
+from pandas import (
+    Categorical,
+    Index,
+    Series,
+)
+from pandas.core.arrays import BaseMaskedArray
+
+
+def get_array(obj, col=None):
+    """
+    Helper method to get array for a DataFrame column or a Series.
+
+    Equivalent of df[col].values, but without going through normal getitem,
+    which triggers tracking references / CoW (and we might be testing that
+    this is done by some other operation).
+    """
+    if isinstance(obj, Index):
+        arr = obj._values
+    elif isinstance(obj, Series) and (col is None or obj.name == col):
+        arr = obj._values
+    else:
+        assert col is not None
+        icol = obj.columns.get_loc(col)
+        assert isinstance(icol, int)
+        arr = obj._get_column_array(icol)
+    if isinstance(arr, BaseMaskedArray):
+        return arr._data
+    elif isinstance(arr, Categorical):
+        return arr
+    return getattr(arr, "_ndarray", arr)
diff --git a/pandas/pandas/tests/dtypes/__init__.py b/pandas/pandas/tests/dtypes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/dtypes/cast/__init__.py b/pandas/pandas/tests/dtypes/cast/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/dtypes/cast/test_can_hold_element.py b/pandas/pandas/tests/dtypes/cast/test_can_hold_element.py
new file mode 100644
index 0000000000000..3b7d76ead119a
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_can_hold_element.py
@@ -0,0 +1,79 @@
+import numpy as np
+
+from pandas.core.dtypes.cast import can_hold_element
+
+
+def test_can_hold_element_range(any_int_numpy_dtype):
+    # GH#44261
+    dtype = np.dtype(any_int_numpy_dtype)
+    arr = np.array([], dtype=dtype)
+
+    rng = range(2, 127)
+    assert can_hold_element(arr, rng)
+
+    # negatives -> can't be held by uint dtypes
+    rng = range(-2, 127)
+    if dtype.kind == "i":
+        assert can_hold_element(arr, rng)
+    else:
+        assert not can_hold_element(arr, rng)
+
+    rng = range(2, 255)
+    if dtype == "int8":
+        assert not can_hold_element(arr, rng)
+    else:
+        assert can_hold_element(arr, rng)
+
+    rng = range(-255, 65537)
+    if dtype.kind == "u":
+        assert not can_hold_element(arr, rng)
+    elif dtype.itemsize < 4:
+        assert not can_hold_element(arr, rng)
+    else:
+        assert can_hold_element(arr, rng)
+
+    # empty
+    rng = range(-(10**10), -(10**10))
+    assert len(rng) == 0
+    # assert can_hold_element(arr, rng)
+
+    rng = range(10**10, 10**10)
+    assert len(rng) == 0
+    assert can_hold_element(arr, rng)
+
+
+def test_can_hold_element_int_values_float_ndarray():
+    arr = np.array([], dtype=np.int64)
+
+    element = np.array([1.0, 2.0])
+    assert can_hold_element(arr, element)
+
+    assert not can_hold_element(arr, element + 0.5)
+
+    # integer but not losslessly castable to int64
+    element = np.array([3, 2**65], dtype=np.float64)
+    assert not can_hold_element(arr, element)
+
+
+def test_can_hold_element_int8_int():
+    arr = np.array([], dtype=np.int8)
+
+    element = 2
+    assert can_hold_element(arr, element)
+    assert can_hold_element(arr, np.int8(element))
+    assert can_hold_element(arr, np.uint8(element))
+    assert can_hold_element(arr, np.int16(element))
+    assert can_hold_element(arr, np.uint16(element))
+    assert can_hold_element(arr, np.int32(element))
+    assert can_hold_element(arr, np.uint32(element))
+    assert can_hold_element(arr, np.int64(element))
+    assert can_hold_element(arr, np.uint64(element))
+
+    element = 2**9
+    assert not can_hold_element(arr, element)
+    assert not can_hold_element(arr, np.int16(element))
+    assert not can_hold_element(arr, np.uint16(element))
+    assert not can_hold_element(arr, np.int32(element))
+    assert not can_hold_element(arr, np.uint32(element))
+    assert not can_hold_element(arr, np.int64(element))
+    assert not can_hold_element(arr, np.uint64(element))
diff --git a/pandas/pandas/tests/dtypes/cast/test_construct_from_scalar.py b/pandas/pandas/tests/dtypes/cast/test_construct_from_scalar.py
new file mode 100644
index 0000000000000..0ce04ce2e64cd
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_construct_from_scalar.py
@@ -0,0 +1,55 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import construct_1d_arraylike_from_scalar
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+from pandas import (
+    Categorical,
+    Timedelta,
+)
+import pandas._testing as tm
+
+
+def test_cast_1d_array_like_from_scalar_categorical():
+    # see gh-19565
+    #
+    # Categorical result from scalar did not maintain
+    # categories and ordering of the passed dtype.
+    cats = ["a", "b", "c"]
+    cat_type = CategoricalDtype(categories=cats, ordered=False)
+    expected = Categorical(["a", "a"], categories=cats)
+
+    result = construct_1d_arraylike_from_scalar("a", len(expected), cat_type)
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_cast_1d_array_like_from_timestamp(fixed_now_ts):
+    # check we dont lose nanoseconds
+    ts = fixed_now_ts + Timedelta(1)
+    res = construct_1d_arraylike_from_scalar(ts, 2, np.dtype("M8[ns]"))
+    assert res[0] == ts
+
+
+def test_cast_1d_array_like_from_timedelta():
+    # check we dont lose nanoseconds
+    td = Timedelta(1)
+    res = construct_1d_arraylike_from_scalar(td, 2, np.dtype("m8[ns]"))
+    assert res[0] == td
+
+
+def test_cast_1d_array_like_mismatched_datetimelike():
+    td = np.timedelta64("NaT", "ns")
+    dt = np.datetime64("NaT", "ns")
+
+    with pytest.raises(TypeError, match="Cannot cast"):
+        construct_1d_arraylike_from_scalar(td, 2, dt.dtype)
+
+    with pytest.raises(TypeError, match="Cannot cast"):
+        construct_1d_arraylike_from_scalar(np.timedelta64(4, "ns"), 2, dt.dtype)
+
+    with pytest.raises(TypeError, match="Cannot cast"):
+        construct_1d_arraylike_from_scalar(dt, 2, td.dtype)
+
+    with pytest.raises(TypeError, match="Cannot cast"):
+        construct_1d_arraylike_from_scalar(np.datetime64(4, "ns"), 2, td.dtype)
diff --git a/pandas/pandas/tests/dtypes/cast/test_construct_ndarray.py b/pandas/pandas/tests/dtypes/cast/test_construct_ndarray.py
new file mode 100644
index 0000000000000..6b9b2dfda6e8b
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_construct_ndarray.py
@@ -0,0 +1,36 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.construction import sanitize_array
+
+
+@pytest.mark.parametrize(
+    "values, dtype, expected",
+    [
+        ([1, 2, 3], None, np.array([1, 2, 3], dtype=np.int64)),
+        (np.array([1, 2, 3]), None, np.array([1, 2, 3])),
+        (["1", "2", None], None, np.array(["1", "2", None])),
+        (["1", "2", None], np.dtype("str"), np.array(["1", "2", None])),
+        ([1, 2, None], np.dtype("str"), np.array(["1", "2", None])),
+    ],
+)
+def test_construct_1d_ndarray_preserving_na(
+    values, dtype, expected, using_infer_string
+):
+    result = sanitize_array(values, index=None, dtype=dtype)
+    if using_infer_string and expected.dtype == object and dtype is None:
+        tm.assert_extension_array_equal(result, pd.array(expected, dtype="str"))
+    else:
+        tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["m8[ns]", "M8[ns]"])
+def test_construct_1d_ndarray_preserving_na_datetimelike(dtype):
+    arr = np.arange(5, dtype=np.int64).view(dtype)
+    expected = np.array(list(arr), dtype=object)
+    assert all(isinstance(x, type(arr[0])) for x in expected)
+
+    result = sanitize_array(arr, index=None, dtype=np.dtype(object))
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/dtypes/cast/test_construct_object_arr.py b/pandas/pandas/tests/dtypes/cast/test_construct_object_arr.py
new file mode 100644
index 0000000000000..cb44f91f34dec
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_construct_object_arr.py
@@ -0,0 +1,20 @@
+import pytest
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+
+
+@pytest.mark.parametrize("datum1", [1, 2.0, "3", (4, 5), [6, 7], None])
+@pytest.mark.parametrize("datum2", [8, 9.0, "10", (11, 12), [13, 14], None])
+def test_cast_1d_array(datum1, datum2):
+    data = [datum1, datum2]
+    result = construct_1d_object_array_from_listlike(data)
+
+    # Direct comparison fails: https://github.com/numpy/numpy/issues/10218
+    assert result.dtype == "object"
+    assert list(result) == data
+
+
+@pytest.mark.parametrize("val", [1, 2.0, None])
+def test_cast_1d_array_invalid_scalar(val):
+    with pytest.raises(TypeError, match="has no len()"):
+        construct_1d_object_array_from_listlike(val)
diff --git a/pandas/pandas/tests/dtypes/cast/test_dict_compat.py b/pandas/pandas/tests/dtypes/cast/test_dict_compat.py
new file mode 100644
index 0000000000000..13dc82d779f95
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_dict_compat.py
@@ -0,0 +1,14 @@
+import numpy as np
+
+from pandas.core.dtypes.cast import dict_compat
+
+from pandas import Timestamp
+
+
+def test_dict_compat():
+    data_datetime64 = {np.datetime64("1990-03-15"): 1, np.datetime64("2015-03-15"): 2}
+    data_unchanged = {1: 2, 3: 4, 5: 6}
+    expected = {Timestamp("1990-3-15"): 1, Timestamp("2015-03-15"): 2}
+    assert dict_compat(data_datetime64) == expected
+    assert dict_compat(expected) == expected
+    assert dict_compat(data_unchanged) == data_unchanged
diff --git a/pandas/pandas/tests/dtypes/cast/test_downcast.py b/pandas/pandas/tests/dtypes/cast/test_downcast.py
new file mode 100644
index 0000000000000..69200b2e5fc96
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_downcast.py
@@ -0,0 +1,97 @@
+import decimal
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import maybe_downcast_to_dtype
+
+from pandas import (
+    Series,
+    Timedelta,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "arr,dtype,expected",
+    [
+        (
+            np.array([8.5, 8.6, 8.7, 8.8, 8.9999999999995]),
+            "infer",
+            np.array([8.5, 8.6, 8.7, 8.8, 8.9999999999995]),
+        ),
+        (
+            np.array([8.0, 8.0, 8.0, 8.0, 8.9999999999995]),
+            "infer",
+            np.array([8, 8, 8, 8, 9], dtype=np.int64),
+        ),
+        (
+            np.array([8.0, 8.0, 8.0, 8.0, 9.0000000000005]),
+            "infer",
+            np.array([8, 8, 8, 8, 9], dtype=np.int64),
+        ),
+        (
+            # This is a judgement call, but we do _not_ downcast Decimal
+            #  objects
+            np.array([decimal.Decimal("0.0")]),
+            "int64",
+            np.array([decimal.Decimal("0.0")]),
+        ),
+        (
+            # GH#45837
+            np.array([Timedelta(days=1), Timedelta(days=2)], dtype=object),
+            "infer",
+            np.array([1, 2], dtype="m8[D]").astype("m8[ns]"),
+        ),
+        # TODO: similar for dt64, dt64tz, Period, Interval?
+    ],
+)
+def test_downcast(arr, expected, dtype):
+    result = maybe_downcast_to_dtype(arr, dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_downcast_booleans():
+    # see gh-16875: coercing of booleans.
+    ser = Series([True, True, False])
+    result = maybe_downcast_to_dtype(ser, np.dtype(np.float64))
+
+    expected = ser.values
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_downcast_conversion_no_nan(any_real_numpy_dtype):
+    dtype = any_real_numpy_dtype
+    expected = np.array([1, 2])
+    arr = np.array([1.0, 2.0], dtype=dtype)
+
+    result = maybe_downcast_to_dtype(arr, "infer")
+    tm.assert_almost_equal(result, expected, check_dtype=False)
+
+
+def test_downcast_conversion_nan(float_numpy_dtype):
+    dtype = float_numpy_dtype
+    data = [1.0, 2.0, np.nan]
+
+    expected = np.array(data, dtype=dtype)
+    arr = np.array(data, dtype=dtype)
+
+    result = maybe_downcast_to_dtype(arr, "infer")
+    tm.assert_almost_equal(result, expected)
+
+
+def test_downcast_conversion_empty(any_real_numpy_dtype):
+    dtype = any_real_numpy_dtype
+    arr = np.array([], dtype=dtype)
+    result = maybe_downcast_to_dtype(arr, np.dtype("int64"))
+    tm.assert_numpy_array_equal(result, np.array([], dtype=np.int64))
+
+
+@pytest.mark.parametrize("klass", [np.datetime64, np.timedelta64])
+def test_datetime_likes_nan(klass):
+    dtype = klass.__name__ + "[ns]"
+    arr = np.array([1, 2, np.nan])
+
+    exp = np.array([1, 2, klass("NaT")], dtype)
+    res = maybe_downcast_to_dtype(arr, dtype)
+    tm.assert_numpy_array_equal(res, exp)
diff --git a/pandas/pandas/tests/dtypes/cast/test_find_common_type.py b/pandas/pandas/tests/dtypes/cast/test_find_common_type.py
new file mode 100644
index 0000000000000..83ef7382fbe8a
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_find_common_type.py
@@ -0,0 +1,175 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import find_common_type
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+
+from pandas import (
+    Categorical,
+    Index,
+)
+
+
+@pytest.mark.parametrize(
+    "source_dtypes,expected_common_dtype",
+    [
+        ((np.int64,), np.int64),
+        ((np.uint64,), np.uint64),
+        ((np.float32,), np.float32),
+        ((object,), object),
+        # Into ints.
+        ((np.int16, np.int64), np.int64),
+        ((np.int32, np.uint32), np.int64),
+        ((np.uint16, np.uint64), np.uint64),
+        # Into floats.
+        ((np.float16, np.float32), np.float32),
+        ((np.float16, np.int16), np.float32),
+        ((np.float32, np.int16), np.float32),
+        ((np.uint64, np.int64), np.float64),
+        ((np.int16, np.float64), np.float64),
+        ((np.float16, np.int64), np.float64),
+        # Into others.
+        ((np.complex128, np.int32), np.complex128),
+        ((object, np.float32), object),
+        ((object, np.int16), object),
+        # Bool with int.
+        ((np.dtype("bool"), np.int64), object),
+        ((np.dtype("bool"), np.int32), object),
+        ((np.dtype("bool"), np.int16), object),
+        ((np.dtype("bool"), np.int8), object),
+        ((np.dtype("bool"), np.uint64), object),
+        ((np.dtype("bool"), np.uint32), object),
+        ((np.dtype("bool"), np.uint16), object),
+        ((np.dtype("bool"), np.uint8), object),
+        # Bool with float.
+        ((np.dtype("bool"), np.float64), object),
+        ((np.dtype("bool"), np.float32), object),
+        (
+            (np.dtype("datetime64[ns]"), np.dtype("datetime64[ns]")),
+            np.dtype("datetime64[ns]"),
+        ),
+        (
+            (np.dtype("timedelta64[ns]"), np.dtype("timedelta64[ns]")),
+            np.dtype("timedelta64[ns]"),
+        ),
+        (
+            (np.dtype("datetime64[ns]"), np.dtype("datetime64[ms]")),
+            np.dtype("datetime64[ns]"),
+        ),
+        (
+            (np.dtype("timedelta64[ms]"), np.dtype("timedelta64[ns]")),
+            np.dtype("timedelta64[ns]"),
+        ),
+        ((np.dtype("datetime64[ns]"), np.dtype("timedelta64[ns]")), object),
+        ((np.dtype("datetime64[ns]"), np.int64), object),
+    ],
+)
+def test_numpy_dtypes(source_dtypes, expected_common_dtype):
+    source_dtypes = [pandas_dtype(x) for x in source_dtypes]
+    assert find_common_type(source_dtypes) == expected_common_dtype
+
+
+def test_raises_empty_input():
+    with pytest.raises(ValueError, match="no types given"):
+        find_common_type([])
+
+
+@pytest.mark.parametrize(
+    "dtypes,exp_type",
+    [
+        ([CategoricalDtype()], "category"),
+        ([object, CategoricalDtype()], object),
+        ([CategoricalDtype(), CategoricalDtype()], "category"),
+    ],
+)
+def test_categorical_dtype(dtypes, exp_type):
+    assert find_common_type(dtypes) == exp_type
+
+
+def test_datetimetz_dtype_match():
+    dtype = DatetimeTZDtype(unit="ns", tz="US/Eastern")
+    assert find_common_type([dtype, dtype]) == "datetime64[ns, US/Eastern]"
+
+
+@pytest.mark.parametrize(
+    "dtype2",
+    [
+        DatetimeTZDtype(unit="ns", tz="Asia/Tokyo"),
+        np.dtype("datetime64[ns]"),
+        object,
+        np.int64,
+    ],
+)
+def test_datetimetz_dtype_mismatch(dtype2):
+    dtype = DatetimeTZDtype(unit="ns", tz="US/Eastern")
+    assert find_common_type([dtype, dtype2]) == object
+    assert find_common_type([dtype2, dtype]) == object
+
+
+def test_period_dtype_match():
+    dtype = PeriodDtype(freq="D")
+    assert find_common_type([dtype, dtype]) == "period[D]"
+
+
+@pytest.mark.parametrize(
+    "dtype2",
+    [
+        DatetimeTZDtype(unit="ns", tz="Asia/Tokyo"),
+        PeriodDtype(freq="2D"),
+        PeriodDtype(freq="h"),
+        np.dtype("datetime64[ns]"),
+        object,
+        np.int64,
+    ],
+)
+def test_period_dtype_mismatch(dtype2):
+    dtype = PeriodDtype(freq="D")
+    assert find_common_type([dtype, dtype2]) == object
+    assert find_common_type([dtype2, dtype]) == object
+
+
+interval_dtypes = [
+    IntervalDtype(np.int64, "right"),
+    IntervalDtype(np.float64, "right"),
+    IntervalDtype(np.uint64, "right"),
+    IntervalDtype(DatetimeTZDtype(unit="ns", tz="US/Eastern"), "right"),
+    IntervalDtype("M8[ns]", "right"),
+    IntervalDtype("m8[ns]", "right"),
+]
+
+
+@pytest.mark.parametrize("left", interval_dtypes)
+@pytest.mark.parametrize("right", interval_dtypes)
+def test_interval_dtype(left, right):
+    result = find_common_type([left, right])
+
+    if left is right:
+        assert result is left
+
+    elif left.subtype.kind in ["i", "u", "f"]:
+        # i.e. numeric
+        if right.subtype.kind in ["i", "u", "f"]:
+            # both numeric -> common numeric subtype
+            expected = IntervalDtype(np.float64, "right")
+            assert result == expected
+        else:
+            assert result == object
+
+    else:
+        assert result == object
+
+
+@pytest.mark.parametrize("dtype", interval_dtypes)
+def test_interval_dtype_with_categorical(dtype):
+    obj = Index([], dtype=dtype)
+
+    cat = Categorical([], categories=obj)
+
+    result = find_common_type([dtype, cat.dtype])
+    assert result == dtype
diff --git a/pandas/pandas/tests/dtypes/cast/test_infer_datetimelike.py b/pandas/pandas/tests/dtypes/cast/test_infer_datetimelike.py
new file mode 100644
index 0000000000000..3c3844e69586d
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_infer_datetimelike.py
@@ -0,0 +1,28 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    NaT,
+    Series,
+    Timestamp,
+)
+
+
+@pytest.mark.parametrize(
+    "data,exp_size",
+    [
+        # see gh-16362.
+        ([[NaT, "a", "b", 0], [NaT, "b", "c", 1]], 8),
+        ([[NaT, "a", 0], [NaT, "b", 1]], 6),
+    ],
+)
+def test_maybe_infer_to_datetimelike_df_construct(data, exp_size):
+    result = DataFrame(np.array(data))
+    assert result.size == exp_size
+
+
+def test_maybe_infer_to_datetimelike_ser_construct():
+    # see gh-19671.
+    result = Series(["M1701", Timestamp("20130101")])
+    assert result.dtype.kind == "O"
diff --git a/pandas/pandas/tests/dtypes/cast/test_infer_dtype.py b/pandas/pandas/tests/dtypes/cast/test_infer_dtype.py
new file mode 100644
index 0000000000000..679031a625c2d
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_infer_dtype.py
@@ -0,0 +1,216 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import (
+    infer_dtype_from,
+    infer_dtype_from_array,
+    infer_dtype_from_scalar,
+)
+from pandas.core.dtypes.common import is_dtype_equal
+
+from pandas import (
+    Categorical,
+    Interval,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+
+
+def test_infer_dtype_from_int_scalar(any_int_numpy_dtype):
+    # Test that infer_dtype_from_scalar is
+    # returning correct dtype for int and float.
+    data = np.dtype(any_int_numpy_dtype).type(12)
+    dtype, val = infer_dtype_from_scalar(data)
+    assert dtype == type(data)
+
+
+def test_infer_dtype_from_float_scalar(float_numpy_dtype):
+    float_numpy_dtype = np.dtype(float_numpy_dtype).type
+    data = float_numpy_dtype(12)
+
+    dtype, val = infer_dtype_from_scalar(data)
+    assert dtype == float_numpy_dtype
+
+
+@pytest.mark.parametrize(
+    "data,exp_dtype", [(12, np.int64), (np.float64(12), np.float64)]
+)
+def test_infer_dtype_from_python_scalar(data, exp_dtype):
+    dtype, val = infer_dtype_from_scalar(data)
+    assert dtype == exp_dtype
+
+
+@pytest.mark.parametrize("bool_val", [True, False])
+def test_infer_dtype_from_boolean(bool_val):
+    dtype, val = infer_dtype_from_scalar(bool_val)
+    assert dtype == np.bool_
+
+
+def test_infer_dtype_from_complex(complex_dtype):
+    data = np.dtype(complex_dtype).type(1)
+    dtype, val = infer_dtype_from_scalar(data)
+    assert dtype == np.complex128
+
+
+def test_infer_dtype_from_datetime():
+    dt64 = np.datetime64(1, "ns")
+    dtype, val = infer_dtype_from_scalar(dt64)
+    assert dtype == "M8[ns]"
+
+    ts = Timestamp(1)
+    dtype, val = infer_dtype_from_scalar(ts)
+    assert dtype == "M8[ns]"
+
+    dt = datetime(2000, 1, 1, 0, 0)
+    dtype, val = infer_dtype_from_scalar(dt)
+    assert dtype == "M8[us]"
+
+
+def test_infer_dtype_from_timedelta():
+    td64 = np.timedelta64(1, "ns")
+    dtype, val = infer_dtype_from_scalar(td64)
+    assert dtype == "m8[ns]"
+
+    pytd = timedelta(1)
+    dtype, val = infer_dtype_from_scalar(pytd)
+    assert dtype == "m8[us]"
+
+    td = Timedelta(1)
+    dtype, val = infer_dtype_from_scalar(td)
+    assert dtype == "m8[ns]"
+
+
+@pytest.mark.parametrize("freq", ["M", "D"])
+def test_infer_dtype_from_period(freq):
+    p = Period("2011-01-01", freq=freq)
+    dtype, val = infer_dtype_from_scalar(p)
+
+    exp_dtype = f"period[{freq}]"
+
+    assert dtype == exp_dtype
+    assert val == p
+
+
+def test_infer_dtype_misc():
+    dt = date(2000, 1, 1)
+    dtype, val = infer_dtype_from_scalar(dt)
+    assert dtype == np.object_
+
+    ts = Timestamp(1, tz="US/Eastern")
+    dtype, val = infer_dtype_from_scalar(ts)
+    assert dtype == "datetime64[ns, US/Eastern]"
+
+
+@pytest.mark.parametrize("tz", ["UTC", "US/Eastern", "Asia/Tokyo"])
+def test_infer_from_scalar_tz(tz):
+    dt = Timestamp(1, tz=tz)
+    dtype, val = infer_dtype_from_scalar(dt)
+
+    exp_dtype = f"datetime64[ns, {tz}]"
+
+    assert dtype == exp_dtype
+    assert val == dt
+
+
+@pytest.mark.parametrize(
+    "left, right, subtype",
+    [
+        (0, 1, "int64"),
+        (0.0, 1.0, "float64"),
+        (Timestamp(0), Timestamp(1), "datetime64[ns]"),
+        (Timestamp(0, tz="UTC"), Timestamp(1, tz="UTC"), "datetime64[ns, UTC]"),
+        (Timedelta(0), Timedelta(1), "timedelta64[ns]"),
+    ],
+)
+def test_infer_from_interval(left, right, subtype, closed):
+    # GH 30337
+    interval = Interval(left, right, closed)
+    result_dtype, result_value = infer_dtype_from_scalar(interval)
+    expected_dtype = f"interval[{subtype}, {closed}]"
+    assert result_dtype == expected_dtype
+    assert result_value == interval
+
+
+def test_infer_dtype_from_scalar_errors():
+    msg = "invalid ndarray passed to infer_dtype_from_scalar"
+
+    with pytest.raises(ValueError, match=msg):
+        infer_dtype_from_scalar(np.array([1]))
+
+
+@pytest.mark.parametrize(
+    "value, expected",
+    [
+        ("foo", np.object_),
+        (b"foo", np.object_),
+        (1, np.int64),
+        (1.5, np.float64),
+        (np.datetime64("2016-01-01"), np.dtype("M8[s]")),
+        (Timestamp("20160101"), np.dtype("M8[s]")),
+        (Timestamp("20160101", tz="UTC"), "datetime64[s, UTC]"),
+    ],
+)
+def test_infer_dtype_from_scalar(value, expected, using_infer_string):
+    dtype, _ = infer_dtype_from_scalar(value)
+    if using_infer_string and value == "foo":
+        expected = "string"
+    assert is_dtype_equal(dtype, expected)
+
+    with pytest.raises(TypeError, match="must be list-like"):
+        infer_dtype_from_array(value)
+
+
+@pytest.mark.parametrize(
+    "arr, expected",
+    [
+        ([1], np.dtype(int)),
+        (np.array([1], dtype=np.int64), np.int64),
+        ([np.nan, 1, ""], np.object_),
+        (np.array([[1.0, 2.0]]), np.float64),
+        (Categorical(list("aabc")), "category"),
+        (Categorical([1, 2, 3]), "category"),
+        (date_range("20160101", periods=3), np.dtype("=M8[ns]")),
+        (
+            date_range("20160101", periods=3, tz="US/Eastern"),
+            "datetime64[ns, US/Eastern]",
+        ),
+        (Series([1.0, 2, 3]), np.float64),
+        (Series(list("abc")), np.object_),
+        (
+            Series(date_range("20160101", periods=3, tz="US/Eastern")),
+            "datetime64[ns, US/Eastern]",
+        ),
+    ],
+)
+def test_infer_dtype_from_array(arr, expected, using_infer_string):
+    dtype, _ = infer_dtype_from_array(arr)
+    if (
+        using_infer_string
+        and isinstance(arr, Series)
+        and arr.tolist() == ["a", "b", "c"]
+    ):
+        expected = "string"
+    assert is_dtype_equal(dtype, expected)
+
+
+@pytest.mark.parametrize("cls", [np.datetime64, np.timedelta64])
+def test_infer_dtype_from_scalar_zerodim_datetimelike(cls):
+    # ndarray.item() can incorrectly return int instead of td64/dt64
+    val = cls(1234, "ns")
+    arr = np.array(val)
+
+    dtype, res = infer_dtype_from_scalar(arr)
+    assert dtype.type is cls
+    assert isinstance(res, cls)
+
+    dtype, res = infer_dtype_from(arr)
+    assert dtype.type is cls
diff --git a/pandas/pandas/tests/dtypes/cast/test_maybe_box_native.py b/pandas/pandas/tests/dtypes/cast/test_maybe_box_native.py
new file mode 100644
index 0000000000000..3f62f31dac219
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_maybe_box_native.py
@@ -0,0 +1,40 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import maybe_box_native
+
+from pandas import (
+    Interval,
+    Period,
+    Timedelta,
+    Timestamp,
+)
+
+
+@pytest.mark.parametrize(
+    "obj,expected_dtype",
+    [
+        (b"\x00\x10", bytes),
+        (int(4), int),
+        (np.uint(4), int),
+        (np.int32(-4), int),
+        (np.uint8(4), int),
+        (float(454.98), float),
+        (np.float16(0.4), float),
+        (np.float64(1.4), float),
+        (np.bool_(False), bool),
+        (datetime(2005, 2, 25), datetime),
+        (np.datetime64("2005-02-25"), Timestamp),
+        (Timestamp("2005-02-25"), Timestamp),
+        (np.timedelta64(1, "D"), Timedelta),
+        (Timedelta(1, "D"), Timedelta),
+        (Interval(0, 1), Interval),
+        (Period("4Q2005"), Period),
+    ],
+)
+def test_maybe_box_native(obj, expected_dtype):
+    boxed_obj = maybe_box_native(obj)
+    result_dtype = type(boxed_obj)
+    assert result_dtype is expected_dtype
diff --git a/pandas/pandas/tests/dtypes/cast/test_promote.py b/pandas/pandas/tests/dtypes/cast/test_promote.py
new file mode 100644
index 0000000000000..021107724bef7
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/cast/test_promote.py
@@ -0,0 +1,530 @@
+"""
+These test the method maybe_promote from core/dtypes/cast.py
+"""
+
+import datetime
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import NaT
+
+from pandas.core.dtypes.cast import maybe_promote
+from pandas.core.dtypes.common import is_scalar
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+from pandas.core.dtypes.missing import isna
+
+import pandas as pd
+
+
+def _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar=None):
+    """
+    Auxiliary function to unify testing of scalar/array promotion.
+
+    Parameters
+    ----------
+    dtype : dtype
+        The value to pass on as the first argument to maybe_promote.
+    fill_value : scalar
+        The value to pass on as the second argument to maybe_promote as
+        a scalar.
+    expected_dtype : dtype
+        The expected dtype returned by maybe_promote (by design this is the
+        same regardless of whether fill_value was passed as a scalar or in an
+        array!).
+    exp_val_for_scalar : scalar
+        The expected value for the (potentially upcast) fill_value returned by
+        maybe_promote.
+    """
+    assert is_scalar(fill_value)
+
+    # here, we pass on fill_value as a scalar directly; the expected value
+    # returned from maybe_promote is fill_value, potentially upcast to the
+    # returned dtype.
+    result_dtype, result_fill_value = maybe_promote(dtype, fill_value)
+    expected_fill_value = exp_val_for_scalar
+
+    assert result_dtype == expected_dtype
+    _assert_match(result_fill_value, expected_fill_value)
+
+
+def _assert_match(result_fill_value, expected_fill_value):
+    # GH#23982/25425 require the same type in addition to equality/NA-ness
+    res_type = type(result_fill_value)
+    ex_type = type(expected_fill_value)
+
+    if hasattr(result_fill_value, "dtype"):
+        # Compare types in a way that is robust to platform-specific
+        #  idiosyncrasies where e.g. sometimes we get "ulonglong" as an alias
+        #  for "uint64" or "intc" as an alias for "int32"
+        assert result_fill_value.dtype.kind == expected_fill_value.dtype.kind
+        assert result_fill_value.dtype.itemsize == expected_fill_value.dtype.itemsize
+    else:
+        # On some builds, type comparison fails, e.g. np.int32 != np.int32
+        assert res_type == ex_type or res_type.__name__ == ex_type.__name__
+
+    match_value = result_fill_value == expected_fill_value
+    if match_value is pd.NA:
+        match_value = False
+
+    # Note: type check above ensures that we have the _same_ NA value
+    # for missing values, None == None (which is checked
+    # through match_value above), but np.nan != np.nan and pd.NaT != pd.NaT
+    match_missing = isna(result_fill_value) and isna(expected_fill_value)
+
+    assert match_value or match_missing
+
+
+@pytest.mark.parametrize(
+    "dtype, fill_value, expected_dtype",
+    [
+        # size 8
+        ("int8", 1, "int8"),
+        ("int8", np.iinfo("int8").max + 1, "int16"),
+        ("int8", np.iinfo("int16").max + 1, "int32"),
+        ("int8", np.iinfo("int32").max + 1, "int64"),
+        ("int8", np.iinfo("int64").max + 1, "object"),
+        ("int8", -1, "int8"),
+        ("int8", np.iinfo("int8").min - 1, "int16"),
+        ("int8", np.iinfo("int16").min - 1, "int32"),
+        ("int8", np.iinfo("int32").min - 1, "int64"),
+        ("int8", np.iinfo("int64").min - 1, "object"),
+        # keep signed-ness as long as possible
+        ("uint8", 1, "uint8"),
+        ("uint8", np.iinfo("int8").max + 1, "uint8"),
+        ("uint8", np.iinfo("uint8").max + 1, "uint16"),
+        ("uint8", np.iinfo("int16").max + 1, "uint16"),
+        ("uint8", np.iinfo("uint16").max + 1, "uint32"),
+        ("uint8", np.iinfo("int32").max + 1, "uint32"),
+        ("uint8", np.iinfo("uint32").max + 1, "uint64"),
+        ("uint8", np.iinfo("int64").max + 1, "uint64"),
+        ("uint8", np.iinfo("uint64").max + 1, "object"),
+        # max of uint8 cannot be contained in int8
+        ("uint8", -1, "int16"),
+        ("uint8", np.iinfo("int8").min - 1, "int16"),
+        ("uint8", np.iinfo("int16").min - 1, "int32"),
+        ("uint8", np.iinfo("int32").min - 1, "int64"),
+        ("uint8", np.iinfo("int64").min - 1, "object"),
+        # size 16
+        ("int16", 1, "int16"),
+        ("int16", np.iinfo("int8").max + 1, "int16"),
+        ("int16", np.iinfo("int16").max + 1, "int32"),
+        ("int16", np.iinfo("int32").max + 1, "int64"),
+        ("int16", np.iinfo("int64").max + 1, "object"),
+        ("int16", -1, "int16"),
+        ("int16", np.iinfo("int8").min - 1, "int16"),
+        ("int16", np.iinfo("int16").min - 1, "int32"),
+        ("int16", np.iinfo("int32").min - 1, "int64"),
+        ("int16", np.iinfo("int64").min - 1, "object"),
+        ("uint16", 1, "uint16"),
+        ("uint16", np.iinfo("int8").max + 1, "uint16"),
+        ("uint16", np.iinfo("uint8").max + 1, "uint16"),
+        ("uint16", np.iinfo("int16").max + 1, "uint16"),
+        ("uint16", np.iinfo("uint16").max + 1, "uint32"),
+        ("uint16", np.iinfo("int32").max + 1, "uint32"),
+        ("uint16", np.iinfo("uint32").max + 1, "uint64"),
+        ("uint16", np.iinfo("int64").max + 1, "uint64"),
+        ("uint16", np.iinfo("uint64").max + 1, "object"),
+        ("uint16", -1, "int32"),
+        ("uint16", np.iinfo("int8").min - 1, "int32"),
+        ("uint16", np.iinfo("int16").min - 1, "int32"),
+        ("uint16", np.iinfo("int32").min - 1, "int64"),
+        ("uint16", np.iinfo("int64").min - 1, "object"),
+        # size 32
+        ("int32", 1, "int32"),
+        ("int32", np.iinfo("int8").max + 1, "int32"),
+        ("int32", np.iinfo("int16").max + 1, "int32"),
+        ("int32", np.iinfo("int32").max + 1, "int64"),
+        ("int32", np.iinfo("int64").max + 1, "object"),
+        ("int32", -1, "int32"),
+        ("int32", np.iinfo("int8").min - 1, "int32"),
+        ("int32", np.iinfo("int16").min - 1, "int32"),
+        ("int32", np.iinfo("int32").min - 1, "int64"),
+        ("int32", np.iinfo("int64").min - 1, "object"),
+        ("uint32", 1, "uint32"),
+        ("uint32", np.iinfo("int8").max + 1, "uint32"),
+        ("uint32", np.iinfo("uint8").max + 1, "uint32"),
+        ("uint32", np.iinfo("int16").max + 1, "uint32"),
+        ("uint32", np.iinfo("uint16").max + 1, "uint32"),
+        ("uint32", np.iinfo("int32").max + 1, "uint32"),
+        ("uint32", np.iinfo("uint32").max + 1, "uint64"),
+        ("uint32", np.iinfo("int64").max + 1, "uint64"),
+        ("uint32", np.iinfo("uint64").max + 1, "object"),
+        ("uint32", -1, "int64"),
+        ("uint32", np.iinfo("int8").min - 1, "int64"),
+        ("uint32", np.iinfo("int16").min - 1, "int64"),
+        ("uint32", np.iinfo("int32").min - 1, "int64"),
+        ("uint32", np.iinfo("int64").min - 1, "object"),
+        # size 64
+        ("int64", 1, "int64"),
+        ("int64", np.iinfo("int8").max + 1, "int64"),
+        ("int64", np.iinfo("int16").max + 1, "int64"),
+        ("int64", np.iinfo("int32").max + 1, "int64"),
+        ("int64", np.iinfo("int64").max + 1, "object"),
+        ("int64", -1, "int64"),
+        ("int64", np.iinfo("int8").min - 1, "int64"),
+        ("int64", np.iinfo("int16").min - 1, "int64"),
+        ("int64", np.iinfo("int32").min - 1, "int64"),
+        ("int64", np.iinfo("int64").min - 1, "object"),
+        ("uint64", 1, "uint64"),
+        ("uint64", np.iinfo("int8").max + 1, "uint64"),
+        ("uint64", np.iinfo("uint8").max + 1, "uint64"),
+        ("uint64", np.iinfo("int16").max + 1, "uint64"),
+        ("uint64", np.iinfo("uint16").max + 1, "uint64"),
+        ("uint64", np.iinfo("int32").max + 1, "uint64"),
+        ("uint64", np.iinfo("uint32").max + 1, "uint64"),
+        ("uint64", np.iinfo("int64").max + 1, "uint64"),
+        ("uint64", np.iinfo("uint64").max + 1, "object"),
+        ("uint64", -1, "object"),
+        ("uint64", np.iinfo("int8").min - 1, "object"),
+        ("uint64", np.iinfo("int16").min - 1, "object"),
+        ("uint64", np.iinfo("int32").min - 1, "object"),
+        ("uint64", np.iinfo("int64").min - 1, "object"),
+    ],
+)
+def test_maybe_promote_int_with_int(dtype, fill_value, expected_dtype):
+    dtype = np.dtype(dtype)
+    expected_dtype = np.dtype(expected_dtype)
+
+    # output is not a generic int, but corresponds to expected_dtype
+    exp_val_for_scalar = np.array([fill_value], dtype=expected_dtype)[0]
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_int_with_float(any_int_numpy_dtype, float_numpy_dtype):
+    dtype = np.dtype(any_int_numpy_dtype)
+    fill_dtype = np.dtype(float_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling int with float always upcasts to float64
+    expected_dtype = np.float64
+    # fill_value can be different float type
+    exp_val_for_scalar = np.float64(fill_value)
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_float_with_int(float_numpy_dtype, any_int_numpy_dtype):
+    dtype = np.dtype(float_numpy_dtype)
+    fill_dtype = np.dtype(any_int_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling float with int always keeps float dtype
+    # because: np.finfo('float32').max > np.iinfo('uint64').max
+    expected_dtype = dtype
+    # output is not a generic float, but corresponds to expected_dtype
+    exp_val_for_scalar = np.array([fill_value], dtype=expected_dtype)[0]
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+@pytest.mark.parametrize(
+    "dtype, fill_value, expected_dtype",
+    [
+        # float filled with float
+        ("float32", 1, "float32"),
+        ("float32", float(np.finfo("float32").max) * 1.1, "float64"),
+        ("float64", 1, "float64"),
+        ("float64", float(np.finfo("float32").max) * 1.1, "float64"),
+        # complex filled with float
+        ("complex64", 1, "complex64"),
+        ("complex64", float(np.finfo("float32").max) * 1.1, "complex128"),
+        ("complex128", 1, "complex128"),
+        ("complex128", float(np.finfo("float32").max) * 1.1, "complex128"),
+        # float filled with complex
+        ("float32", 1 + 1j, "complex64"),
+        ("float32", float(np.finfo("float32").max) * (1.1 + 1j), "complex128"),
+        ("float64", 1 + 1j, "complex128"),
+        ("float64", float(np.finfo("float32").max) * (1.1 + 1j), "complex128"),
+        # complex filled with complex
+        ("complex64", 1 + 1j, "complex64"),
+        ("complex64", float(np.finfo("float32").max) * (1.1 + 1j), "complex128"),
+        ("complex128", 1 + 1j, "complex128"),
+        ("complex128", float(np.finfo("float32").max) * (1.1 + 1j), "complex128"),
+    ],
+)
+def test_maybe_promote_float_with_float(dtype, fill_value, expected_dtype):
+    dtype = np.dtype(dtype)
+    expected_dtype = np.dtype(expected_dtype)
+
+    # output is not a generic float, but corresponds to expected_dtype
+    exp_val_for_scalar = np.array([fill_value], dtype=expected_dtype)[0]
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_bool_with_any(any_numpy_dtype):
+    dtype = np.dtype(bool)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling bool with anything but bool casts to object
+    expected_dtype = np.dtype(object) if fill_dtype != bool else fill_dtype
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_any_with_bool(any_numpy_dtype):
+    dtype = np.dtype(any_numpy_dtype)
+    fill_value = True
+
+    # filling anything but bool with bool casts to object
+    expected_dtype = np.dtype(object) if dtype != bool else dtype
+    # output is not a generic bool, but corresponds to expected_dtype
+    exp_val_for_scalar = np.array([fill_value], dtype=expected_dtype)[0]
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_bytes_with_any(bytes_dtype, any_numpy_dtype):
+    dtype = np.dtype(bytes_dtype)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # we never use bytes dtype internally, always promote to object
+    expected_dtype = np.dtype(np.object_)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_any_with_bytes(any_numpy_dtype):
+    dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype
+    fill_value = b"abc"
+
+    # we never use bytes dtype internally, always promote to object
+    expected_dtype = np.dtype(np.object_)
+    # output is not a generic bytes, but corresponds to expected_dtype
+    exp_val_for_scalar = np.array([fill_value], dtype=expected_dtype)[0]
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_datetime64_with_any(datetime64_dtype, any_numpy_dtype):
+    dtype = np.dtype(datetime64_dtype)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling datetime with anything but datetime casts to object
+    if fill_dtype.kind == "M":
+        expected_dtype = dtype
+        # for datetime dtypes, scalar values get cast to to_datetime64
+        exp_val_for_scalar = pd.Timestamp(fill_value).to_datetime64()
+    else:
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+@pytest.mark.parametrize(
+    "fill_value",
+    [
+        pd.Timestamp("now"),
+        np.datetime64("now"),
+        datetime.datetime.now(),
+        datetime.date.today(),
+    ],
+    ids=["pd.Timestamp", "np.datetime64", "datetime.datetime", "datetime.date"],
+)
+def test_maybe_promote_any_with_datetime64(any_numpy_dtype, fill_value):
+    dtype = np.dtype(any_numpy_dtype)
+
+    # filling datetime with anything but datetime casts to object
+    if dtype.kind == "M":
+        expected_dtype = dtype
+        # for datetime dtypes, scalar values get cast to pd.Timestamp.value
+        exp_val_for_scalar = pd.Timestamp(fill_value).to_datetime64()
+    else:
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+
+    if type(fill_value) is datetime.date and dtype.kind == "M":
+        # Casting date to dt64 is deprecated, in 2.0 enforced to cast to object
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+@pytest.mark.parametrize(
+    "fill_value",
+    [
+        pd.Timestamp(2023, 1, 1),
+        np.datetime64("2023-01-01"),
+        datetime.datetime(2023, 1, 1),
+        datetime.date(2023, 1, 1),
+    ],
+    ids=["pd.Timestamp", "np.datetime64", "datetime.datetime", "datetime.date"],
+)
+def test_maybe_promote_any_numpy_dtype_with_datetimetz(
+    any_numpy_dtype, tz_aware_fixture, fill_value
+):
+    dtype = np.dtype(any_numpy_dtype)
+    fill_dtype = DatetimeTZDtype(tz=tz_aware_fixture)
+
+    fill_value = pd.Series([fill_value], dtype=fill_dtype)[0]
+
+    # filling any numpy dtype with datetimetz casts to object
+    expected_dtype = np.dtype(object)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_timedelta64_with_any(timedelta64_dtype, any_numpy_dtype):
+    dtype = np.dtype(timedelta64_dtype)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling timedelta with anything but timedelta casts to object
+    if fill_dtype.kind == "m":
+        expected_dtype = dtype
+        # for timedelta dtypes, scalar values get cast to pd.Timedelta.value
+        exp_val_for_scalar = pd.Timedelta(fill_value).to_timedelta64()
+    else:
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+@pytest.mark.parametrize(
+    "fill_value",
+    [pd.Timedelta(days=1), np.timedelta64(24, "h"), datetime.timedelta(1)],
+    ids=["pd.Timedelta", "np.timedelta64", "datetime.timedelta"],
+)
+def test_maybe_promote_any_with_timedelta64(any_numpy_dtype, fill_value):
+    dtype = np.dtype(any_numpy_dtype)
+
+    # filling anything but timedelta with timedelta casts to object
+    if dtype.kind == "m":
+        expected_dtype = dtype
+        # for timedelta dtypes, scalar values get cast to pd.Timedelta.value
+        exp_val_for_scalar = pd.Timedelta(fill_value).to_timedelta64()
+    else:
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_string_with_any(string_dtype, any_numpy_dtype):
+    dtype = np.dtype(string_dtype)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling string with anything casts to object
+    expected_dtype = np.dtype(object)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_any_with_string(any_numpy_dtype):
+    dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype
+    fill_value = "abc"
+
+    # filling anything with a string casts to object
+    expected_dtype = np.dtype(object)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_object_with_any(object_dtype, any_numpy_dtype):
+    dtype = np.dtype(object_dtype)
+    fill_dtype = np.dtype(any_numpy_dtype)
+
+    # create array of given dtype; casts "1" to correct dtype
+    fill_value = np.array([1], dtype=fill_dtype)[0]
+
+    # filling object with anything stays object
+    expected_dtype = np.dtype(object)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_any_with_object(any_numpy_dtype):
+    dtype = np.dtype(any_numpy_dtype)
+
+    # create array of object dtype from a scalar value (i.e. passing
+    # dtypes.common.is_scalar), which can however not be cast to int/float etc.
+    fill_value = pd.DateOffset(1)
+
+    # filling object with anything stays object
+    expected_dtype = np.dtype(object)
+    exp_val_for_scalar = fill_value
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
+
+
+def test_maybe_promote_any_numpy_dtype_with_na(any_numpy_dtype, nulls_fixture):
+    fill_value = nulls_fixture
+    dtype = np.dtype(any_numpy_dtype)
+
+    if isinstance(fill_value, Decimal):
+        # Subject to change, but ATM (When Decimal(NAN) is being added to nulls_fixture)
+        #  this is the existing behavior in maybe_promote,
+        #  hinges on is_valid_na_for_dtype
+        if dtype.kind in "iufc":
+            if dtype.kind in "iu":
+                expected_dtype = np.dtype(np.float64)
+            else:
+                expected_dtype = dtype
+            exp_val_for_scalar = np.nan
+        else:
+            expected_dtype = np.dtype(object)
+            exp_val_for_scalar = fill_value
+    elif dtype.kind in "iu" and fill_value is not NaT:
+        # integer + other missing value (np.nan / None) casts to float
+        expected_dtype = np.float64
+        exp_val_for_scalar = np.nan
+    elif dtype == object and fill_value is NaT:
+        # inserting into object does not cast the value
+        # but *does* cast None to np.nan
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = fill_value
+    elif dtype.kind in "mM":
+        # datetime / timedelta cast all missing values to dtyped-NaT
+        expected_dtype = dtype
+        exp_val_for_scalar = dtype.type("NaT", "ns")
+    elif fill_value is NaT:
+        # NaT upcasts everything that's not datetime/timedelta to object
+        expected_dtype = np.dtype(object)
+        exp_val_for_scalar = NaT
+    elif dtype.kind in "fc":
+        # float / complex + missing value (!= NaT) stays the same
+        expected_dtype = dtype
+        exp_val_for_scalar = np.nan
+    else:
+        # all other cases cast to object, and use np.nan as missing value
+        expected_dtype = np.dtype(object)
+        if fill_value is pd.NA:
+            exp_val_for_scalar = pd.NA
+        else:
+            exp_val_for_scalar = np.nan
+
+    _check_promote(dtype, fill_value, expected_dtype, exp_val_for_scalar)
diff --git a/pandas/pandas/tests/dtypes/test_common.py b/pandas/pandas/tests/dtypes/test_common.py
new file mode 100644
index 0000000000000..d30fa9fc2ea0f
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_common.py
@@ -0,0 +1,872 @@
+from __future__ import annotations
+
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.astype import astype_array
+import pandas.core.dtypes.common as com
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    CategoricalDtypeType,
+    DatetimeTZDtype,
+    ExtensionDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import isna
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import pandas_dtype
+from pandas.arrays import SparseArray
+from pandas.util.version import Version
+
+
+# EA & Actual Dtypes
+def to_ea_dtypes(dtypes):
+    """convert list of string dtypes to EA dtype"""
+    return [getattr(pd, dt + "Dtype") for dt in dtypes]
+
+
+def to_numpy_dtypes(dtypes):
+    """convert list of string dtypes to numpy dtype"""
+    return [getattr(np, dt) for dt in dtypes if isinstance(dt, str)]
+
+
+class TestNumpyEADtype:
+    # Passing invalid dtype, both as a string or object, must raise TypeError
+    # Per issue GH15520
+    @pytest.mark.parametrize("box", [pd.Timestamp, "pd.Timestamp", list])
+    def test_invalid_dtype_error(self, box):
+        with pytest.raises(TypeError, match="not understood"):
+            com.pandas_dtype(box)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            object,
+            "float64",
+            np.object_,
+            np.dtype("object"),
+            "O",
+            np.float64,
+            float,
+            np.dtype("float64"),
+            "object_",
+        ],
+    )
+    def test_pandas_dtype_valid(self, dtype):
+        assert com.pandas_dtype(dtype) == dtype
+
+    @pytest.mark.parametrize(
+        "dtype", ["M8[ns]", "m8[ns]", "object", "float64", "int64"]
+    )
+    def test_numpy_dtype(self, dtype):
+        assert com.pandas_dtype(dtype) == np.dtype(dtype)
+
+    def test_numpy_string_dtype(self):
+        # do not parse freq-like string as period dtype
+        assert com.pandas_dtype("U") == np.dtype("U")
+        assert com.pandas_dtype("S") == np.dtype("S")
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "datetime64[ns, US/Eastern]",
+            "datetime64[ns, Asia/Tokyo]",
+            "datetime64[ns, UTC]",
+            # GH#33885 check that the M8 alias is understood
+            "M8[ns, US/Eastern]",
+            "M8[ns, Asia/Tokyo]",
+            "M8[ns, UTC]",
+        ],
+    )
+    def test_datetimetz_dtype(self, dtype):
+        assert com.pandas_dtype(dtype) == DatetimeTZDtype.construct_from_string(dtype)
+        assert com.pandas_dtype(dtype) == dtype
+
+    def test_categorical_dtype(self):
+        assert com.pandas_dtype("category") == CategoricalDtype()
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "period[D]",
+            "period[3M]",
+            "period[us]",
+            "Period[D]",
+            "Period[3M]",
+            "Period[us]",
+        ],
+    )
+    def test_period_dtype(self, dtype):
+        assert com.pandas_dtype(dtype) is not PeriodDtype(dtype)
+        assert com.pandas_dtype(dtype) == PeriodDtype(dtype)
+        assert com.pandas_dtype(dtype) == dtype
+
+
+dtypes = {
+    "datetime_tz": com.pandas_dtype("datetime64[ns, US/Eastern]"),
+    "datetime": com.pandas_dtype("datetime64[ns]"),
+    "timedelta": com.pandas_dtype("timedelta64[ns]"),
+    "period": PeriodDtype("D"),
+    "integer": np.dtype(np.int64),
+    "float": np.dtype(np.float64),
+    "object": np.dtype(object),
+    "category": com.pandas_dtype("category"),
+    "string": pd.StringDtype(),
+}
+
+
+@pytest.mark.parametrize("name1,dtype1", list(dtypes.items()), ids=lambda x: str(x))
+@pytest.mark.parametrize("name2,dtype2", list(dtypes.items()), ids=lambda x: str(x))
+def test_dtype_equal(name1, dtype1, name2, dtype2):
+    # match equal to self, but not equal to other
+    assert com.is_dtype_equal(dtype1, dtype1)
+    if name1 != name2:
+        assert not com.is_dtype_equal(dtype1, dtype2)
+
+
+@pytest.mark.parametrize("name,dtype", list(dtypes.items()), ids=lambda x: str(x))
+def test_pyarrow_string_import_error(name, dtype):
+    # GH-44276
+    assert not com.is_dtype_equal(dtype, "string[pyarrow]")
+
+
+@pytest.mark.parametrize(
+    "dtype1,dtype2",
+    [
+        (np.int8, np.int64),
+        (np.int16, np.int64),
+        (np.int32, np.int64),
+        (np.float32, np.float64),
+        (PeriodDtype("D"), PeriodDtype("2D")),  # PeriodType
+        (
+            com.pandas_dtype("datetime64[ns, US/Eastern]"),
+            com.pandas_dtype("datetime64[ns, CET]"),
+        ),  # Datetime
+        (None, None),  # gh-15941: no exception should be raised.
+    ],
+)
+def test_dtype_equal_strict(dtype1, dtype2):
+    assert not com.is_dtype_equal(dtype1, dtype2)
+
+
+def get_is_dtype_funcs():
+    """
+    Get all functions in pandas.core.dtypes.common that
+    begin with 'is_' and end with 'dtype'
+
+    """
+    fnames = [f for f in dir(com) if (f.startswith("is_") and f.endswith("dtype"))]
+    fnames.remove("is_string_or_object_np_dtype")  # fastpath requires np.dtype obj
+    return [getattr(com, fname) for fname in fnames]
+
+
+@pytest.mark.filterwarnings(
+    "ignore:is_categorical_dtype is deprecated:DeprecationWarning"
+)
+@pytest.mark.parametrize("func", get_is_dtype_funcs(), ids=lambda x: x.__name__)
+def test_get_dtype_error_catch(func):
+    # see gh-15941
+    #
+    # No exception should be raised.
+
+    msg = f"{func.__name__} is deprecated"
+    warn = None
+    if (
+        func is com.is_int64_dtype
+        or func is com.is_interval_dtype
+        or func is com.is_datetime64tz_dtype
+        or func is com.is_categorical_dtype
+        or func is com.is_period_dtype
+    ):
+        warn = DeprecationWarning
+
+    with tm.assert_produces_warning(warn, match=msg):
+        assert not func(None)
+
+
+def test_is_object():
+    assert com.is_object_dtype(object)
+    assert com.is_object_dtype(np.array([], dtype=object))
+
+    assert not com.is_object_dtype(int)
+    assert not com.is_object_dtype(np.array([], dtype=int))
+    assert not com.is_object_dtype([1, 2, 3])
+
+
+@pytest.mark.parametrize(
+    "check_scipy", [False, pytest.param(True, marks=td.skip_if_no("scipy"))]
+)
+def test_is_sparse(check_scipy):
+    msg = "is_sparse is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert com.is_sparse(SparseArray([1, 2, 3]))
+
+        assert not com.is_sparse(np.array([1, 2, 3]))
+
+        if check_scipy:
+            import scipy.sparse
+
+            assert not com.is_sparse(scipy.sparse.bsr_matrix([1, 2, 3]))
+
+
+def test_is_scipy_sparse():
+    sp_sparse = pytest.importorskip("scipy.sparse")
+
+    assert com.is_scipy_sparse(sp_sparse.bsr_matrix([1, 2, 3]))
+
+    assert not com.is_scipy_sparse(SparseArray([1, 2, 3]))
+
+
+def test_is_datetime64_dtype():
+    assert not com.is_datetime64_dtype(object)
+    assert not com.is_datetime64_dtype([1, 2, 3])
+    assert not com.is_datetime64_dtype(np.array([], dtype=int))
+
+    assert com.is_datetime64_dtype(np.datetime64)
+    assert com.is_datetime64_dtype(np.array([], dtype=np.datetime64))
+
+
+def test_is_datetime64tz_dtype():
+    msg = "is_datetime64tz_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_datetime64tz_dtype(object)
+        assert not com.is_datetime64tz_dtype([1, 2, 3])
+        assert not com.is_datetime64tz_dtype(pd.DatetimeIndex([1, 2, 3]))
+        assert com.is_datetime64tz_dtype(pd.DatetimeIndex(["2000"], tz="US/Eastern"))
+
+
+def test_custom_ea_kind_M_not_datetime64tz():
+    # GH 34986
+    class NotTZDtype(ExtensionDtype):
+        @property
+        def kind(self) -> str:
+            return "M"
+
+    not_tz_dtype = NotTZDtype()
+    msg = "is_datetime64tz_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_datetime64tz_dtype(not_tz_dtype)
+        assert not com.needs_i8_conversion(not_tz_dtype)
+
+
+def test_is_timedelta64_dtype():
+    assert not com.is_timedelta64_dtype(object)
+    assert not com.is_timedelta64_dtype(None)
+    assert not com.is_timedelta64_dtype([1, 2, 3])
+    assert not com.is_timedelta64_dtype(np.array([], dtype=np.datetime64))
+    assert not com.is_timedelta64_dtype("0 days")
+    assert not com.is_timedelta64_dtype("0 days 00:00:00")
+    assert not com.is_timedelta64_dtype(["0 days 00:00:00"])
+    assert not com.is_timedelta64_dtype("NO DATE")
+
+    assert com.is_timedelta64_dtype(np.timedelta64)
+    assert com.is_timedelta64_dtype(pd.Series([], dtype="timedelta64[ns]"))
+    assert com.is_timedelta64_dtype(pd.to_timedelta(["0 days", "1 days"]))
+
+
+def test_is_period_dtype():
+    msg = "is_period_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_period_dtype(object)
+        assert not com.is_period_dtype([1, 2, 3])
+        assert not com.is_period_dtype(pd.Period("2017-01-01"))
+
+        assert com.is_period_dtype(PeriodDtype(freq="D"))
+        assert com.is_period_dtype(pd.PeriodIndex([], freq="Y"))
+
+
+def test_is_interval_dtype():
+    msg = "is_interval_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_interval_dtype(object)
+        assert not com.is_interval_dtype([1, 2, 3])
+
+        assert com.is_interval_dtype(IntervalDtype())
+
+        interval = pd.Interval(1, 2, closed="right")
+        assert not com.is_interval_dtype(interval)
+        assert com.is_interval_dtype(pd.IntervalIndex([interval]))
+
+
+def test_is_categorical_dtype():
+    msg = "is_categorical_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_categorical_dtype(object)
+        assert not com.is_categorical_dtype([1, 2, 3])
+
+        assert com.is_categorical_dtype(CategoricalDtype())
+        assert com.is_categorical_dtype(pd.Categorical([1, 2, 3]))
+        assert com.is_categorical_dtype(pd.CategoricalIndex([1, 2, 3]))
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        (int, False),
+        (pd.Series([1, 2]), False),
+        (str, True),
+        (object, True),
+        (np.array(["a", "b"]), True),
+        (pd.StringDtype(), True),
+        (pd.Index([], dtype="O"), True),
+    ],
+)
+def test_is_string_dtype(dtype, expected):
+    # GH#54661
+
+    result = com.is_string_dtype(dtype)
+    assert result is expected
+
+
+@pytest.mark.parametrize(
+    "data",
+    [[(0, 1), (1, 1)], pd.Categorical([1, 2, 3]), np.array([1, 2], dtype=object)],
+)
+def test_is_string_dtype_arraylike_with_object_elements_not_strings(data):
+    # GH 15585
+    assert not com.is_string_dtype(pd.Series(data))
+
+
+def test_is_string_dtype_nullable(nullable_string_dtype):
+    assert com.is_string_dtype(pd.array(["a", "b"], dtype=nullable_string_dtype))
+
+
+integer_dtypes: list = []
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    integer_dtypes
+    + [pd.Series([1, 2])]
+    + tm.ALL_INT_NUMPY_DTYPES
+    + to_numpy_dtypes(tm.ALL_INT_NUMPY_DTYPES)
+    + tm.ALL_INT_EA_DTYPES
+    + to_ea_dtypes(tm.ALL_INT_EA_DTYPES),
+)
+def test_is_integer_dtype(dtype):
+    assert com.is_integer_dtype(dtype)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        str,
+        float,
+        np.datetime64,
+        np.timedelta64,
+        pd.Index([1, 2.0]),
+        np.array(["a", "b"]),
+        np.array([], dtype=np.timedelta64),
+    ],
+)
+def test_is_not_integer_dtype(dtype):
+    assert not com.is_integer_dtype(dtype)
+
+
+signed_integer_dtypes: list = []
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    signed_integer_dtypes
+    + [pd.Series([1, 2])]
+    + tm.SIGNED_INT_NUMPY_DTYPES
+    + to_numpy_dtypes(tm.SIGNED_INT_NUMPY_DTYPES)
+    + tm.SIGNED_INT_EA_DTYPES
+    + to_ea_dtypes(tm.SIGNED_INT_EA_DTYPES),
+)
+def test_is_signed_integer_dtype(dtype):
+    assert com.is_integer_dtype(dtype)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        str,
+        float,
+        np.datetime64,
+        np.timedelta64,
+        pd.Index([1, 2.0]),
+        np.array(["a", "b"]),
+        np.array([], dtype=np.timedelta64),
+    ]
+    + tm.UNSIGNED_INT_NUMPY_DTYPES
+    + to_numpy_dtypes(tm.UNSIGNED_INT_NUMPY_DTYPES)
+    + tm.UNSIGNED_INT_EA_DTYPES
+    + to_ea_dtypes(tm.UNSIGNED_INT_EA_DTYPES),
+)
+def test_is_not_signed_integer_dtype(dtype):
+    assert not com.is_signed_integer_dtype(dtype)
+
+
+unsigned_integer_dtypes: list = []
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    unsigned_integer_dtypes
+    + [pd.Series([1, 2], dtype=np.uint32)]
+    + tm.UNSIGNED_INT_NUMPY_DTYPES
+    + to_numpy_dtypes(tm.UNSIGNED_INT_NUMPY_DTYPES)
+    + tm.UNSIGNED_INT_EA_DTYPES
+    + to_ea_dtypes(tm.UNSIGNED_INT_EA_DTYPES),
+)
+def test_is_unsigned_integer_dtype(dtype):
+    assert com.is_unsigned_integer_dtype(dtype)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        str,
+        float,
+        np.datetime64,
+        np.timedelta64,
+        pd.Index([1, 2.0]),
+        np.array(["a", "b"]),
+        np.array([], dtype=np.timedelta64),
+    ]
+    + tm.SIGNED_INT_NUMPY_DTYPES
+    + to_numpy_dtypes(tm.SIGNED_INT_NUMPY_DTYPES)
+    + tm.SIGNED_INT_EA_DTYPES
+    + to_ea_dtypes(tm.SIGNED_INT_EA_DTYPES),
+)
+def test_is_not_unsigned_integer_dtype(dtype):
+    assert not com.is_unsigned_integer_dtype(dtype)
+
+
+@pytest.mark.parametrize(
+    "dtype", [np.int64, np.array([1, 2], dtype=np.int64), "Int64", pd.Int64Dtype]
+)
+def test_is_int64_dtype(dtype):
+    msg = "is_int64_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert com.is_int64_dtype(dtype)
+
+
+def test_type_comparison_with_numeric_ea_dtype(any_numeric_ea_dtype):
+    # GH#43038
+    assert pandas_dtype(any_numeric_ea_dtype) == any_numeric_ea_dtype
+
+
+def test_type_comparison_with_real_numpy_dtype(any_real_numpy_dtype):
+    # GH#43038
+    assert pandas_dtype(any_real_numpy_dtype) == any_real_numpy_dtype
+
+
+def test_type_comparison_with_signed_int_ea_dtype_and_signed_int_numpy_dtype(
+    any_signed_int_ea_dtype, any_signed_int_numpy_dtype
+):
+    # GH#43038
+    assert not pandas_dtype(any_signed_int_ea_dtype) == any_signed_int_numpy_dtype
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        str,
+        float,
+        np.int32,
+        np.uint64,
+        pd.Index([1, 2.0]),
+        np.array(["a", "b"]),
+        np.array([1, 2], dtype=np.uint32),
+        "int8",
+        "Int8",
+        pd.Int8Dtype,
+    ],
+)
+def test_is_not_int64_dtype(dtype):
+    msg = "is_int64_dtype is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        assert not com.is_int64_dtype(dtype)
+
+
+def test_is_datetime64_any_dtype():
+    assert not com.is_datetime64_any_dtype(int)
+    assert not com.is_datetime64_any_dtype(str)
+    assert not com.is_datetime64_any_dtype(np.array([1, 2]))
+    assert not com.is_datetime64_any_dtype(np.array(["a", "b"]))
+
+    assert com.is_datetime64_any_dtype(np.datetime64)
+    assert com.is_datetime64_any_dtype(np.array([], dtype=np.datetime64))
+    assert com.is_datetime64_any_dtype(DatetimeTZDtype("ns", "US/Eastern"))
+    assert com.is_datetime64_any_dtype(
+        pd.DatetimeIndex([1, 2, 3], dtype="datetime64[ns]")
+    )
+
+
+def test_is_datetime64_ns_dtype():
+    assert not com.is_datetime64_ns_dtype(int)
+    assert not com.is_datetime64_ns_dtype(str)
+    assert not com.is_datetime64_ns_dtype(np.datetime64)
+    assert not com.is_datetime64_ns_dtype(np.array([1, 2]))
+    assert not com.is_datetime64_ns_dtype(np.array(["a", "b"]))
+    assert not com.is_datetime64_ns_dtype(np.array([], dtype=np.datetime64))
+
+    # This datetime array has the wrong unit (ps instead of ns)
+    assert not com.is_datetime64_ns_dtype(np.array([], dtype="datetime64[ps]"))
+
+    assert com.is_datetime64_ns_dtype(DatetimeTZDtype("ns", "US/Eastern"))
+    assert com.is_datetime64_ns_dtype(
+        pd.DatetimeIndex([1, 2, 3], dtype=np.dtype("datetime64[ns]"))
+    )
+
+    # non-nano dt64tz
+    assert not com.is_datetime64_ns_dtype(DatetimeTZDtype("us", "US/Eastern"))
+
+
+def test_is_timedelta64_ns_dtype():
+    assert not com.is_timedelta64_ns_dtype(np.dtype("m8[ps]"))
+    assert not com.is_timedelta64_ns_dtype(np.array([1, 2], dtype=np.timedelta64))
+
+    assert com.is_timedelta64_ns_dtype(np.dtype("m8[ns]"))
+    assert com.is_timedelta64_ns_dtype(np.array([1, 2], dtype="m8[ns]"))
+
+
+def test_is_numeric_v_string_like():
+    assert not com.is_numeric_v_string_like(np.array([1]), 1)
+    assert not com.is_numeric_v_string_like(np.array([1]), np.array([2]))
+    assert not com.is_numeric_v_string_like(np.array(["foo"]), np.array(["foo"]))
+
+    assert com.is_numeric_v_string_like(np.array([1]), "foo")
+    assert com.is_numeric_v_string_like(np.array([1, 2]), np.array(["foo"]))
+    assert com.is_numeric_v_string_like(np.array(["foo"]), np.array([1, 2]))
+
+
+def test_needs_i8_conversion():
+    assert not com.needs_i8_conversion(str)
+    assert not com.needs_i8_conversion(np.int64)
+    assert not com.needs_i8_conversion(pd.Series([1, 2]))
+    assert not com.needs_i8_conversion(np.array(["a", "b"]))
+
+    assert not com.needs_i8_conversion(np.datetime64)
+    assert com.needs_i8_conversion(np.dtype(np.datetime64))
+    assert not com.needs_i8_conversion(pd.Series([], dtype="timedelta64[ns]"))
+    assert com.needs_i8_conversion(pd.Series([], dtype="timedelta64[ns]").dtype)
+    assert not com.needs_i8_conversion(pd.DatetimeIndex(["2000"], tz="US/Eastern"))
+    assert com.needs_i8_conversion(pd.DatetimeIndex(["2000"], tz="US/Eastern").dtype)
+
+
+def test_is_numeric_dtype():
+    assert not com.is_numeric_dtype(str)
+    assert not com.is_numeric_dtype(np.datetime64)
+    assert not com.is_numeric_dtype(np.timedelta64)
+    assert not com.is_numeric_dtype(np.array(["a", "b"]))
+    assert not com.is_numeric_dtype(np.array([], dtype=np.timedelta64))
+
+    assert com.is_numeric_dtype(int)
+    assert com.is_numeric_dtype(float)
+    assert com.is_numeric_dtype(np.uint64)
+    assert com.is_numeric_dtype(pd.Series([1, 2]))
+    assert com.is_numeric_dtype(pd.Index([1, 2.0]))
+
+    class MyNumericDType(ExtensionDtype):
+        @property
+        def type(self):
+            return str
+
+        @property
+        def name(self):
+            raise NotImplementedError
+
+        @classmethod
+        def construct_array_type(cls):
+            raise NotImplementedError
+
+        def _is_numeric(self) -> bool:
+            return True
+
+    assert com.is_numeric_dtype(MyNumericDType())
+
+
+def test_is_any_real_numeric_dtype():
+    assert not com.is_any_real_numeric_dtype(str)
+    assert not com.is_any_real_numeric_dtype(bool)
+    assert not com.is_any_real_numeric_dtype(complex)
+    assert not com.is_any_real_numeric_dtype(object)
+    assert not com.is_any_real_numeric_dtype(np.datetime64)
+    assert not com.is_any_real_numeric_dtype(np.array(["a", "b", complex(1, 2)]))
+    assert not com.is_any_real_numeric_dtype(pd.DataFrame([complex(1, 2), True]))
+
+    assert com.is_any_real_numeric_dtype(int)
+    assert com.is_any_real_numeric_dtype(float)
+    assert com.is_any_real_numeric_dtype(np.array([1, 2.5]))
+
+
+def test_is_float_dtype():
+    assert not com.is_float_dtype(str)
+    assert not com.is_float_dtype(int)
+    assert not com.is_float_dtype(pd.Series([1, 2]))
+    assert not com.is_float_dtype(np.array(["a", "b"]))
+
+    assert com.is_float_dtype(float)
+    assert com.is_float_dtype(pd.Index([1, 2.0]))
+
+
+def test_is_bool_dtype():
+    assert not com.is_bool_dtype(int)
+    assert not com.is_bool_dtype(str)
+    assert not com.is_bool_dtype(pd.Series([1, 2]))
+    assert not com.is_bool_dtype(pd.Series(["a", "b"], dtype="category"))
+    assert not com.is_bool_dtype(np.array(["a", "b"]))
+    assert not com.is_bool_dtype(pd.Index(["a", "b"]))
+    assert not com.is_bool_dtype("Int64")
+
+    assert com.is_bool_dtype(bool)
+    assert com.is_bool_dtype(np.bool_)
+    assert com.is_bool_dtype(pd.Series([True, False], dtype="category"))
+    assert com.is_bool_dtype(np.array([True, False]))
+    assert com.is_bool_dtype(pd.Index([True, False]))
+
+    assert com.is_bool_dtype(pd.BooleanDtype())
+    assert com.is_bool_dtype(pd.array([True, False, None], dtype="boolean"))
+    assert com.is_bool_dtype("boolean")
+
+
+def test_is_bool_dtype_numpy_error():
+    # GH39010
+    assert not com.is_bool_dtype("0 - Name")
+
+
+@pytest.mark.parametrize(
+    "check_scipy", [False, pytest.param(True, marks=td.skip_if_no("scipy"))]
+)
+def test_is_extension_array_dtype(check_scipy):
+    assert not com.is_extension_array_dtype([1, 2, 3])
+    assert not com.is_extension_array_dtype(np.array([1, 2, 3]))
+    assert not com.is_extension_array_dtype(pd.DatetimeIndex([1, 2, 3]))
+
+    cat = pd.Categorical([1, 2, 3])
+    assert com.is_extension_array_dtype(cat)
+    assert com.is_extension_array_dtype(pd.Series(cat))
+    assert com.is_extension_array_dtype(SparseArray([1, 2, 3]))
+    assert com.is_extension_array_dtype(pd.DatetimeIndex(["2000"], tz="US/Eastern"))
+
+    dtype = DatetimeTZDtype("ns", tz="US/Eastern")
+    s = pd.Series([], dtype=dtype)
+    assert com.is_extension_array_dtype(s)
+
+    if check_scipy:
+        import scipy.sparse
+
+        assert not com.is_extension_array_dtype(scipy.sparse.bsr_matrix([1, 2, 3]))
+
+
+def test_is_complex_dtype():
+    assert not com.is_complex_dtype(int)
+    assert not com.is_complex_dtype(str)
+    assert not com.is_complex_dtype(pd.Series([1, 2]))
+    assert not com.is_complex_dtype(np.array(["a", "b"]))
+
+    assert com.is_complex_dtype(np.complex128)
+    assert com.is_complex_dtype(complex)
+    assert com.is_complex_dtype(np.array([1 + 1j, 5]))
+
+
+@pytest.mark.parametrize(
+    "input_param,result",
+    [
+        (int, np.dtype(int)),
+        ("int32", np.dtype("int32")),
+        (float, np.dtype(float)),
+        ("float64", np.dtype("float64")),
+        (np.dtype("float64"), np.dtype("float64")),
+        (str, np.dtype(str)),
+        (pd.Series([1, 2], dtype=np.dtype("int16")), np.dtype("int16")),
+        (pd.Series(["a", "b"], dtype=object), np.dtype(object)),
+        (pd.Index([1, 2]), np.dtype("int64")),
+        (pd.Index(["a", "b"], dtype=object), np.dtype(object)),
+        ("category", "category"),
+        (pd.Categorical(["a", "b"]).dtype, CategoricalDtype(["a", "b"])),
+        (pd.Categorical(["a", "b"]), CategoricalDtype(["a", "b"])),
+        (pd.CategoricalIndex(["a", "b"]).dtype, CategoricalDtype(["a", "b"])),
+        (pd.CategoricalIndex(["a", "b"]), CategoricalDtype(["a", "b"])),
+        (CategoricalDtype(), CategoricalDtype()),
+        (pd.DatetimeIndex([1, 2]), np.dtype("=M8[ns]")),
+        (pd.DatetimeIndex([1, 2]).dtype, np.dtype("=M8[ns]")),
+        ("<M8[ns]", np.dtype("<M8[ns]")),
+        ("datetime64[ns, Europe/London]", DatetimeTZDtype("ns", "Europe/London")),
+        (PeriodDtype(freq="D"), PeriodDtype(freq="D")),
+        ("period[D]", PeriodDtype(freq="D")),
+        (IntervalDtype(), IntervalDtype()),
+    ],
+)
+def test_get_dtype(input_param, result):
+    assert com._get_dtype(input_param) == result
+
+
+@pytest.mark.parametrize(
+    "input_param,expected_error_message",
+    [
+        (None, "Cannot deduce dtype from null object"),
+        (1, "data type not understood"),
+        (1.2, "data type not understood"),
+        # numpy dev changed from double-quotes to single quotes
+        ("random string", "data type [\"']random string[\"'] not understood"),
+        (pd.DataFrame([1, 2]), "data type not understood"),
+    ],
+)
+def test_get_dtype_fails(input_param, expected_error_message):
+    # python objects
+    # 2020-02-02 npdev changed error message
+    expected_error_message += f"|Cannot interpret '{input_param}' as a data type"
+    with pytest.raises(TypeError, match=expected_error_message):
+        com._get_dtype(input_param)
+
+
+@pytest.mark.parametrize(
+    "input_param,result",
+    [
+        (int, np.dtype(int).type),
+        ("int32", np.int32),
+        (float, np.dtype(float).type),
+        ("float64", np.float64),
+        (np.dtype("float64"), np.float64),
+        (str, np.dtype(str).type),
+        (pd.Series([1, 2], dtype=np.dtype("int16")), np.int16),
+        (pd.Series(["a", "b"], dtype=object), np.object_),
+        (pd.Index([1, 2], dtype="int64"), np.int64),
+        (pd.Index(["a", "b"], dtype=object), np.object_),
+        ("category", CategoricalDtypeType),
+        (pd.Categorical(["a", "b"]).dtype, CategoricalDtypeType),
+        (pd.Categorical(["a", "b"]), CategoricalDtypeType),
+        (pd.CategoricalIndex(["a", "b"]).dtype, CategoricalDtypeType),
+        (pd.CategoricalIndex(["a", "b"]), CategoricalDtypeType),
+        (pd.DatetimeIndex([1, 2]), np.datetime64),
+        (pd.DatetimeIndex([1, 2]).dtype, np.datetime64),
+        ("<M8[ns]", np.datetime64),
+        (pd.DatetimeIndex(["2000"], tz="Europe/London"), pd.Timestamp),
+        (pd.DatetimeIndex(["2000"], tz="Europe/London").dtype, pd.Timestamp),
+        ("datetime64[ns, Europe/London]", pd.Timestamp),
+        (PeriodDtype(freq="D"), pd.Period),
+        ("period[D]", pd.Period),
+        (IntervalDtype(), pd.Interval),
+        (None, type(None)),
+        (1, type(None)),
+        (1.2, type(None)),
+        (pd.DataFrame([1, 2]), type(None)),  # composite dtype
+    ],
+)
+def test__is_dtype_type(input_param, result):
+    assert com._is_dtype_type(input_param, lambda tipo: tipo == result)
+
+
+def test_astype_nansafe_copy_false(any_int_numpy_dtype):
+    # GH#34457 use astype, not view
+    arr = np.array([1, 2, 3], dtype=any_int_numpy_dtype)
+
+    dtype = np.dtype("float64")
+    result = astype_array(arr, dtype, copy=False)
+
+    expected = np.array([1.0, 2.0, 3.0], dtype=dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("from_type", [np.datetime64, np.timedelta64])
+def test_astype_object_preserves_datetime_na(from_type):
+    arr = np.array([from_type("NaT", "ns")])
+    result = astype_array(arr, dtype=np.dtype("object"))
+
+    assert isna(result)[0]
+
+
+def test_validate_allhashable():
+    assert com.validate_all_hashable(1, "a") is None
+
+    with pytest.raises(TypeError, match="All elements must be hashable"):
+        com.validate_all_hashable([])
+
+    with pytest.raises(TypeError, match="list must be a hashable type"):
+        com.validate_all_hashable([], error_name="list")
+
+
+def test_pandas_dtype_numpy_warning():
+    # GH#51523
+    if Version(np.__version__) < Version("2.3.0.dev0"):
+        ctx = tm.assert_produces_warning(
+            DeprecationWarning,
+            check_stacklevel=False,
+            match=(
+                "Converting `np.integer` or `np.signedinteger` to a dtype is deprecated"
+            ),
+        )
+    else:
+        ctx = tm.external_error_raised(TypeError)
+
+    with ctx:
+        pandas_dtype(np.integer)
+
+
+def test_pandas_dtype_ea_not_instance():
+    # GH 31356 GH 54592
+    with tm.assert_produces_warning(UserWarning, match="without any arguments"):
+        assert pandas_dtype(CategoricalDtype) == CategoricalDtype()
+
+
+def test_pandas_dtype_string_dtypes(string_storage):
+    with pd.option_context("future.infer_string", True):
+        # with the default string_storage setting
+        result = pandas_dtype("str")
+    assert result == pd.StringDtype(
+        "pyarrow" if HAS_PYARROW else "python", na_value=np.nan
+    )
+
+    with pd.option_context("future.infer_string", True):
+        # with the default string_storage setting
+        result = pandas_dtype(str)
+    assert result == pd.StringDtype(
+        "pyarrow" if HAS_PYARROW else "python", na_value=np.nan
+    )
+
+    with pd.option_context("future.infer_string", True):
+        with pd.option_context("string_storage", string_storage):
+            result = pandas_dtype("str")
+    assert result == pd.StringDtype(string_storage, na_value=np.nan)
+
+    with pd.option_context("future.infer_string", True):
+        with pd.option_context("string_storage", string_storage):
+            result = pandas_dtype(str)
+    assert result == pd.StringDtype(string_storage, na_value=np.nan)
+
+    with pd.option_context("future.infer_string", False):
+        with pd.option_context("string_storage", string_storage):
+            result = pandas_dtype("str")
+    assert result == np.dtype("U")
+
+    with pd.option_context("string_storage", string_storage):
+        result = pandas_dtype("string")
+    assert result == pd.StringDtype(string_storage, na_value=pd.NA)
+
+
+def test_pandas_dtype_string_dtype_alias_with_storage():
+    with pytest.raises(TypeError, match="not understood"):
+        pandas_dtype("str[python]")
+
+    with pytest.raises(TypeError, match="not understood"):
+        pandas_dtype("str[pyarrow]")
+
+    result = pandas_dtype("string[python]")
+    assert result == pd.StringDtype("python", na_value=pd.NA)
+
+    if HAS_PYARROW:
+        result = pandas_dtype("string[pyarrow]")
+        assert result == pd.StringDtype("pyarrow", na_value=pd.NA)
+    else:
+        with pytest.raises(
+            ImportError, match="required for PyArrow backed StringArray"
+        ):
+            pandas_dtype("string[pyarrow]")
+
+
+@td.skip_if_installed("pyarrow")
+def test_construct_from_string_without_pyarrow_installed():
+    # GH 57928
+    with pytest.raises(ImportError, match="pyarrow>=10.0.1 is required"):
+        pd.Series([-1.5, 0.2, None], dtype="float32[pyarrow]")
diff --git a/pandas/pandas/tests/dtypes/test_concat.py b/pandas/pandas/tests/dtypes/test_concat.py
new file mode 100644
index 0000000000000..571e12d0c3303
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_concat.py
@@ -0,0 +1,66 @@
+import pytest
+
+import pandas.core.dtypes.concat as _concat
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+
+def test_concat_mismatched_categoricals_with_empty():
+    # concat_compat behavior on series._values should match pd.concat on series
+    ser1 = Series(["a", "b", "c"], dtype="category")
+    ser2 = Series([], dtype="category")
+
+    result = _concat.concat_compat([ser1._values, ser2._values])
+    expected = pd.concat([ser1, ser2])._values
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_concat_single_dataframe_tz_aware():
+    # https://github.com/pandas-dev/pandas/issues/25257
+    df = pd.DataFrame(
+        {"timestamp": [pd.Timestamp("2020-04-08 09:00:00.709949+0000", tz="UTC")]}
+    )
+    expected = df.copy()
+    result = pd.concat([df])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_periodarray_2d():
+    pi = pd.period_range("2016-01-01", periods=36, freq="D")
+    arr = pi._data.reshape(6, 6)
+
+    result = _concat.concat_compat([arr[:2], arr[2:]], axis=0)
+    tm.assert_period_array_equal(result, arr)
+
+    result = _concat.concat_compat([arr[:, :2], arr[:, 2:]], axis=1)
+    tm.assert_period_array_equal(result, arr)
+
+    msg = (
+        "all the input array dimensions.* for the concatenation axis must match exactly"
+    )
+    with pytest.raises(ValueError, match=msg):
+        _concat.concat_compat([arr[:, :2], arr[:, 2:]], axis=0)
+
+    with pytest.raises(ValueError, match=msg):
+        _concat.concat_compat([arr[:2], arr[2:]], axis=1)
+
+
+def test_concat_series_between_empty_and_tzaware_series(using_infer_string):
+    tzaware_time = pd.Timestamp("2020-01-01T00:00:00+00:00")
+    ser1 = Series(index=[tzaware_time], data=0, dtype=float)
+    ser2 = Series(dtype=float)
+
+    result = pd.concat([ser1, ser2], axis=1)
+    expected = pd.DataFrame(
+        data=[
+            (0.0, None),
+        ],
+        index=[tzaware_time]
+        if using_infer_string
+        else pd.Index([tzaware_time], dtype=object),
+        columns=[0, 1],
+        dtype=float,
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/dtypes/test_dtypes.py b/pandas/pandas/tests/dtypes/test_dtypes.py
new file mode 100644
index 0000000000000..621217a8c9317
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_dtypes.py
@@ -0,0 +1,1253 @@
+import re
+import weakref
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+
+from pandas.core.dtypes.base import _registry as registry
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_categorical_dtype,
+    is_datetime64_any_dtype,
+    is_datetime64_dtype,
+    is_datetime64_ns_dtype,
+    is_datetime64tz_dtype,
+    is_dtype_equal,
+    is_interval_dtype,
+    is_period_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DatetimeIndex,
+    IntervalIndex,
+    Series,
+    SparseDtype,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+class Base:
+    def test_hash(self, dtype):
+        hash(dtype)
+
+    def test_equality_invalid(self, dtype):
+        assert not dtype == "foo"
+        assert not is_dtype_equal(dtype, np.int64)
+
+    def test_numpy_informed(self, dtype):
+        # npdev 2020-02-02 changed from "data type not understood" to
+        #  "Cannot interpret 'foo' as a data type"
+        msg = "|".join(
+            ["data type not understood", "Cannot interpret '.*' as a data type"]
+        )
+        with pytest.raises(TypeError, match=msg):
+            np.dtype(dtype)
+
+        assert not dtype == np.str_
+        assert not np.str_ == dtype
+
+    def test_pickle(self, dtype):
+        # make sure our cache is NOT pickled
+
+        # clear the cache
+        type(dtype).reset_cache()
+        assert not len(dtype._cache_dtypes)
+
+        # force back to the cache
+        result = tm.round_trip_pickle(dtype)
+        if not isinstance(dtype, PeriodDtype):
+            # Because PeriodDtype has a cython class as a base class,
+            #  it has different pickle semantics, and its cache is re-populated
+            #  on un-pickling.
+            assert not len(dtype._cache_dtypes)
+        assert result == dtype
+
+
+class TestCategoricalDtype(Base):
+    @pytest.fixture
+    def dtype(self):
+        """
+        Class level fixture of dtype for TestCategoricalDtype
+        """
+        return CategoricalDtype()
+
+    def test_hash_vs_equality(self, dtype):
+        dtype2 = CategoricalDtype()
+        assert dtype == dtype2
+        assert dtype2 == dtype
+        assert hash(dtype) == hash(dtype2)
+
+    def test_equality(self, dtype):
+        assert dtype == "category"
+        assert is_dtype_equal(dtype, "category")
+        assert "category" == dtype
+        assert is_dtype_equal("category", dtype)
+
+        assert dtype == CategoricalDtype()
+        assert is_dtype_equal(dtype, CategoricalDtype())
+        assert CategoricalDtype() == dtype
+        assert is_dtype_equal(CategoricalDtype(), dtype)
+
+        assert dtype != "foo"
+        assert not is_dtype_equal(dtype, "foo")
+        assert "foo" != dtype
+        assert not is_dtype_equal("foo", dtype)
+
+    def test_construction_from_string(self, dtype):
+        result = CategoricalDtype.construct_from_string("category")
+        assert is_dtype_equal(dtype, result)
+        msg = "Cannot construct a 'CategoricalDtype' from 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            CategoricalDtype.construct_from_string("foo")
+
+    def test_constructor_invalid(self):
+        msg = "Parameter 'categories' must be list-like"
+        with pytest.raises(TypeError, match=msg):
+            CategoricalDtype("category")
+
+    dtype1 = CategoricalDtype(["a", "b"], ordered=True)
+    dtype2 = CategoricalDtype(["x", "y"], ordered=False)
+    c = Categorical([0, 1], dtype=dtype1)
+
+    @pytest.mark.parametrize(
+        "values, categories, ordered, dtype, expected",
+        [
+            [None, None, None, None, CategoricalDtype()],
+            [None, ["a", "b"], True, None, dtype1],
+            [c, None, None, dtype2, dtype2],
+            [c, ["x", "y"], False, None, dtype2],
+        ],
+    )
+    def test_from_values_or_dtype(self, values, categories, ordered, dtype, expected):
+        result = CategoricalDtype._from_values_or_dtype(
+            values, categories, ordered, dtype
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "values, categories, ordered, dtype",
+        [
+            [None, ["a", "b"], True, dtype2],
+            [None, ["a", "b"], None, dtype2],
+            [None, None, True, dtype2],
+        ],
+    )
+    def test_from_values_or_dtype_raises(self, values, categories, ordered, dtype):
+        msg = "Cannot specify `categories` or `ordered` together with `dtype`."
+        with pytest.raises(ValueError, match=msg):
+            CategoricalDtype._from_values_or_dtype(values, categories, ordered, dtype)
+
+    def test_from_values_or_dtype_invalid_dtype(self):
+        msg = "Cannot not construct CategoricalDtype from <class 'object'>"
+        with pytest.raises(ValueError, match=msg):
+            CategoricalDtype._from_values_or_dtype(None, None, None, object)
+
+    def test_is_dtype(self, dtype):
+        assert CategoricalDtype.is_dtype(dtype)
+        assert CategoricalDtype.is_dtype("category")
+        assert CategoricalDtype.is_dtype(CategoricalDtype())
+        assert not CategoricalDtype.is_dtype("foo")
+        assert not CategoricalDtype.is_dtype(np.float64)
+
+    def test_basic(self, dtype):
+        msg = "is_categorical_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_categorical_dtype(dtype)
+
+            factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"])
+
+            s = Series(factor, name="A")
+
+            # dtypes
+            assert is_categorical_dtype(s.dtype)
+            assert is_categorical_dtype(s)
+            assert not is_categorical_dtype(np.dtype("float64"))
+
+    def test_tuple_categories(self):
+        categories = [(1, "a"), (2, "b"), (3, "c")]
+        result = CategoricalDtype(categories)
+        assert all(result.categories == categories)
+
+    @pytest.mark.parametrize(
+        "categories, expected",
+        [
+            ([True, False], True),
+            ([True, False, None], True),
+            ([True, False, "a", "b'"], False),
+            ([0, 1], False),
+        ],
+    )
+    def test_is_boolean(self, categories, expected):
+        cat = Categorical(categories)
+        assert cat.dtype._is_boolean is expected
+        assert is_bool_dtype(cat) is expected
+        assert is_bool_dtype(cat.dtype) is expected
+
+    def test_dtype_specific_categorical_dtype(self):
+        expected = "datetime64[ns]"
+        dti = DatetimeIndex([], dtype=expected)
+        result = str(Categorical(dti).categories.dtype)
+        assert result == expected
+
+    def test_not_string(self):
+        # though CategoricalDtype has object kind, it cannot be string
+        assert not is_string_dtype(CategoricalDtype())
+
+    def test_repr_range_categories(self):
+        rng = pd.Index(range(3))
+        dtype = CategoricalDtype(categories=rng, ordered=False)
+        result = repr(dtype)
+
+        expected = (
+            "CategoricalDtype(categories=range(0, 3), ordered=False, "
+            "categories_dtype=int64)"
+        )
+        assert result == expected
+
+    def test_update_dtype(self):
+        # GH 27338
+        result = CategoricalDtype(["a"]).update_dtype(Categorical(["b"], ordered=True))
+        expected = CategoricalDtype(["b"], ordered=True)
+        assert result == expected
+
+    def test_repr(self):
+        cat = Categorical(pd.Index([1, 2, 3], dtype="int32"))
+        result = cat.dtype.__repr__()
+        expected = (
+            "CategoricalDtype(categories=[1, 2, 3], ordered=False, "
+            "categories_dtype=int32)"
+        )
+        assert result == expected
+
+
+class TestDatetimeTZDtype(Base):
+    @pytest.fixture
+    def dtype(self):
+        """
+        Class level fixture of dtype for TestDatetimeTZDtype
+        """
+        return DatetimeTZDtype("ns", "US/Eastern")
+
+    def test_alias_to_unit_raises(self):
+        # 23990
+        with pytest.raises(ValueError, match="Passing a dtype alias"):
+            DatetimeTZDtype("datetime64[ns, US/Central]")
+
+    def test_alias_to_unit_bad_alias_raises(self):
+        # 23990
+        with pytest.raises(TypeError, match=""):
+            DatetimeTZDtype("this is a bad string")
+
+        with pytest.raises(TypeError, match=""):
+            DatetimeTZDtype("datetime64[ns, US/NotATZ]")
+
+    def test_hash_vs_equality(self, dtype):
+        # make sure that we satisfy is semantics
+        dtype2 = DatetimeTZDtype("ns", "US/Eastern")
+        dtype3 = DatetimeTZDtype(dtype2)
+        assert dtype == dtype2
+        assert dtype2 == dtype
+        assert dtype3 == dtype
+        assert hash(dtype) == hash(dtype2)
+        assert hash(dtype) == hash(dtype3)
+
+        dtype4 = DatetimeTZDtype("ns", "US/Central")
+        assert dtype2 != dtype4
+        assert hash(dtype2) != hash(dtype4)
+
+    def test_construction_non_nanosecond(self):
+        res = DatetimeTZDtype("ms", "US/Eastern")
+        assert res.unit == "ms"
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+        assert res.str == "|M8[ms]"
+        assert str(res) == "datetime64[ms, US/Eastern]"
+        assert res.base == np.dtype("M8[ms]")
+
+    def test_day_not_supported(self):
+        msg = "DatetimeTZDtype only supports s, ms, us, ns units"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeTZDtype("D", "US/Eastern")
+
+    def test_subclass(self):
+        a = DatetimeTZDtype.construct_from_string("datetime64[ns, US/Eastern]")
+        b = DatetimeTZDtype.construct_from_string("datetime64[ns, CET]")
+
+        assert issubclass(type(a), type(a))
+        assert issubclass(type(a), type(b))
+
+    def test_compat(self, dtype):
+        msg = "is_datetime64tz_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_datetime64tz_dtype(dtype)
+            assert is_datetime64tz_dtype("datetime64[ns, US/Eastern]")
+        assert is_datetime64_any_dtype(dtype)
+        assert is_datetime64_any_dtype("datetime64[ns, US/Eastern]")
+        assert is_datetime64_ns_dtype(dtype)
+        assert is_datetime64_ns_dtype("datetime64[ns, US/Eastern]")
+        assert not is_datetime64_dtype(dtype)
+        assert not is_datetime64_dtype("datetime64[ns, US/Eastern]")
+
+    def test_construction_from_string(self, dtype):
+        result = DatetimeTZDtype.construct_from_string("datetime64[ns, US/Eastern]")
+        assert is_dtype_equal(dtype, result)
+
+    @pytest.mark.parametrize(
+        "string",
+        [
+            "foo",
+            "datetime64[ns, notatz]",
+            # non-nano unit
+            "datetime64[ps, UTC]",
+            # dateutil str that returns None from gettz
+            "datetime64[ns, dateutil/invalid]",
+        ],
+    )
+    def test_construct_from_string_invalid_raises(self, string):
+        msg = f"Cannot construct a 'DatetimeTZDtype' from '{string}'"
+        with pytest.raises(TypeError, match=re.escape(msg)):
+            DatetimeTZDtype.construct_from_string(string)
+
+    def test_construct_from_string_wrong_type_raises(self):
+        msg = "'construct_from_string' expects a string, got <class 'list'>"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeTZDtype.construct_from_string(["datetime64[ns, notatz]"])
+
+    def test_is_dtype(self, dtype):
+        assert not DatetimeTZDtype.is_dtype(None)
+        assert DatetimeTZDtype.is_dtype(dtype)
+        assert DatetimeTZDtype.is_dtype("datetime64[ns, US/Eastern]")
+        assert DatetimeTZDtype.is_dtype("M8[ns, US/Eastern]")
+        assert not DatetimeTZDtype.is_dtype("foo")
+        assert DatetimeTZDtype.is_dtype(DatetimeTZDtype("ns", "US/Pacific"))
+        assert not DatetimeTZDtype.is_dtype(np.float64)
+
+    def test_equality(self, dtype):
+        assert is_dtype_equal(dtype, "datetime64[ns, US/Eastern]")
+        assert is_dtype_equal(dtype, "M8[ns, US/Eastern]")
+        assert is_dtype_equal(dtype, DatetimeTZDtype("ns", "US/Eastern"))
+        assert not is_dtype_equal(dtype, "foo")
+        assert not is_dtype_equal(dtype, DatetimeTZDtype("ns", "CET"))
+        assert not is_dtype_equal(
+            DatetimeTZDtype("ns", "US/Eastern"), DatetimeTZDtype("ns", "US/Pacific")
+        )
+
+        # numpy compat
+        assert is_dtype_equal(np.dtype("M8[ns]"), "datetime64[ns]")
+
+        assert dtype == "M8[ns, US/Eastern]"
+
+    def test_basic(self, dtype):
+        msg = "is_datetime64tz_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_datetime64tz_dtype(dtype)
+
+        dr = date_range("20130101", periods=3, tz="US/Eastern")
+        s = Series(dr, name="A")
+
+        # dtypes
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_datetime64tz_dtype(s.dtype)
+            assert is_datetime64tz_dtype(s)
+            assert not is_datetime64tz_dtype(np.dtype("float64"))
+            assert not is_datetime64tz_dtype(1.0)
+
+    def test_dst(self):
+        dr1 = date_range("2013-01-01", periods=3, tz="US/Eastern")
+        s1 = Series(dr1, name="A")
+        assert isinstance(s1.dtype, DatetimeTZDtype)
+
+        dr2 = date_range("2013-08-01", periods=3, tz="US/Eastern")
+        s2 = Series(dr2, name="A")
+        assert isinstance(s2.dtype, DatetimeTZDtype)
+        assert s1.dtype == s2.dtype
+
+    @pytest.mark.parametrize("tz", ["UTC", "US/Eastern"])
+    @pytest.mark.parametrize("constructor", ["M8", "datetime64"])
+    def test_parser(self, tz, constructor):
+        # pr #11245
+        dtz_str = f"{constructor}[ns, {tz}]"
+        result = DatetimeTZDtype.construct_from_string(dtz_str)
+        expected = DatetimeTZDtype("ns", tz)
+        assert result == expected
+
+    def test_empty(self):
+        with pytest.raises(TypeError, match="A 'tz' is required."):
+            DatetimeTZDtype()
+
+    def test_tz_standardize(self):
+        # GH 24713
+        pytz = pytest.importorskip("pytz")
+        tz = pytz.timezone("US/Eastern")
+        dr = date_range("2013-01-01", periods=3, tz=tz)
+        dtype = DatetimeTZDtype("ns", dr.tz)
+        assert dtype.tz == tz
+        dtype = DatetimeTZDtype("ns", dr[0].tz)
+        assert dtype.tz == tz
+
+
+class TestPeriodDtype(Base):
+    @pytest.fixture
+    def dtype(self):
+        """
+        Class level fixture of dtype for TestPeriodDtype
+        """
+        return PeriodDtype("D")
+
+    def test_hash_vs_equality(self, dtype):
+        # make sure that we satisfy is semantics
+        dtype2 = PeriodDtype("D")
+        dtype3 = PeriodDtype(dtype2)
+        assert dtype == dtype2
+        assert dtype2 == dtype
+        assert dtype3 == dtype
+        assert dtype is not dtype2
+        assert dtype2 is not dtype
+        assert dtype3 is not dtype
+        assert hash(dtype) == hash(dtype2)
+        assert hash(dtype) == hash(dtype3)
+
+    def test_construction(self):
+        with pytest.raises(ValueError, match="Invalid frequency: xx"):
+            PeriodDtype("xx")
+
+        for s in ["period[D]", "Period[D]", "D"]:
+            dt = PeriodDtype(s)
+            assert dt.freq == pd.tseries.offsets.Day()
+
+        for s in ["period[3D]", "Period[3D]", "3D"]:
+            dt = PeriodDtype(s)
+            assert dt.freq == pd.tseries.offsets.Day(3)
+
+        for s in [
+            "period[26h]",
+            "Period[26h]",
+            "26h",
+            "period[1D2h]",
+            "Period[1D2h]",
+            "1D2h",
+        ]:
+            dt = PeriodDtype(s)
+            assert dt.freq == pd.tseries.offsets.Hour(26)
+
+    def test_cannot_use_custom_businessday(self):
+        # GH#52534
+        msg = "C is not supported as period frequency"
+        msg1 = "<CustomBusinessDay> is not supported as period frequency"
+        msg2 = r"PeriodDtype\[B\] is deprecated"
+        with pytest.raises(ValueError, match=msg):
+            PeriodDtype("C")
+        with pytest.raises(ValueError, match=msg1):
+            with tm.assert_produces_warning(FutureWarning, match=msg2):
+                PeriodDtype(pd.offsets.CustomBusinessDay())
+
+    def test_subclass(self):
+        a = PeriodDtype("period[D]")
+        b = PeriodDtype("period[3D]")
+
+        assert issubclass(type(a), type(a))
+        assert issubclass(type(a), type(b))
+
+    def test_identity(self):
+        assert PeriodDtype("period[D]") == PeriodDtype("period[D]")
+        assert PeriodDtype("period[D]") is not PeriodDtype("period[D]")
+
+        assert PeriodDtype("period[3D]") == PeriodDtype("period[3D]")
+        assert PeriodDtype("period[3D]") is not PeriodDtype("period[3D]")
+
+        assert PeriodDtype("period[1s1us]") == PeriodDtype("period[1000001us]")
+        assert PeriodDtype("period[1s1us]") is not PeriodDtype("period[1000001us]")
+
+    def test_compat(self, dtype):
+        assert not is_datetime64_ns_dtype(dtype)
+        assert not is_datetime64_ns_dtype("period[D]")
+        assert not is_datetime64_dtype(dtype)
+        assert not is_datetime64_dtype("period[D]")
+
+    def test_construction_from_string(self, dtype):
+        result = PeriodDtype("period[D]")
+        assert is_dtype_equal(dtype, result)
+        result = PeriodDtype.construct_from_string("period[D]")
+        assert is_dtype_equal(dtype, result)
+
+        with pytest.raises(TypeError, match="list"):
+            PeriodDtype.construct_from_string([1, 2, 3])
+
+    @pytest.mark.parametrize(
+        "string",
+        [
+            "foo",
+            "period[foo]",
+            "foo[D]",
+            "datetime64[ns]",
+            "datetime64[ns, US/Eastern]",
+        ],
+    )
+    def test_construct_dtype_from_string_invalid_raises(self, string):
+        msg = f"Cannot construct a 'PeriodDtype' from '{string}'"
+        with pytest.raises(TypeError, match=re.escape(msg)):
+            PeriodDtype.construct_from_string(string)
+
+    def test_is_dtype(self, dtype):
+        assert PeriodDtype.is_dtype(dtype)
+        assert PeriodDtype.is_dtype("period[D]")
+        assert PeriodDtype.is_dtype("period[3D]")
+        assert PeriodDtype.is_dtype(PeriodDtype("3D"))
+        assert PeriodDtype.is_dtype("period[us]")
+        assert PeriodDtype.is_dtype("period[s]")
+        assert PeriodDtype.is_dtype(PeriodDtype("us"))
+        assert PeriodDtype.is_dtype(PeriodDtype("s"))
+
+        assert not PeriodDtype.is_dtype("D")
+        assert not PeriodDtype.is_dtype("3D")
+        assert not PeriodDtype.is_dtype("U")
+        assert not PeriodDtype.is_dtype("s")
+        assert not PeriodDtype.is_dtype("foo")
+        assert not PeriodDtype.is_dtype(np.object_)
+        assert not PeriodDtype.is_dtype(np.int64)
+        assert not PeriodDtype.is_dtype(np.float64)
+
+    def test_equality(self, dtype):
+        assert is_dtype_equal(dtype, "period[D]")
+        assert is_dtype_equal(dtype, PeriodDtype("D"))
+        assert is_dtype_equal(dtype, PeriodDtype("D"))
+        assert is_dtype_equal(PeriodDtype("D"), PeriodDtype("D"))
+
+        assert not is_dtype_equal(dtype, "D")
+        assert not is_dtype_equal(PeriodDtype("D"), PeriodDtype("2D"))
+
+    def test_basic(self, dtype):
+        msg = "is_period_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_period_dtype(dtype)
+
+            pidx = pd.period_range("2013-01-01 09:00", periods=5, freq="h")
+
+            assert is_period_dtype(pidx.dtype)
+            assert is_period_dtype(pidx)
+
+            s = Series(pidx, name="A")
+
+            assert is_period_dtype(s.dtype)
+            assert is_period_dtype(s)
+
+            assert not is_period_dtype(np.dtype("float64"))
+            assert not is_period_dtype(1.0)
+
+    def test_freq_argument_required(self):
+        # GH#27388
+        msg = "missing 1 required positional argument: 'freq'"
+        with pytest.raises(TypeError, match=msg):
+            PeriodDtype()
+
+        msg = "PeriodDtype argument should be string or BaseOffset, got NoneType"
+        with pytest.raises(TypeError, match=msg):
+            # GH#51790
+            PeriodDtype(None)
+
+    def test_not_string(self):
+        # though PeriodDtype has object kind, it cannot be string
+        assert not is_string_dtype(PeriodDtype("D"))
+
+    def test_perioddtype_caching_dateoffset_normalize(self):
+        # GH 24121
+        per_d = PeriodDtype(pd.offsets.YearEnd(normalize=True))
+        assert per_d.freq.normalize
+
+        per_d2 = PeriodDtype(pd.offsets.YearEnd(normalize=False))
+        assert not per_d2.freq.normalize
+
+    def test_dont_keep_ref_after_del(self):
+        # GH 54184
+        dtype = PeriodDtype("D")
+        ref = weakref.ref(dtype)
+        del dtype
+        assert ref() is None
+
+
+class TestIntervalDtype(Base):
+    @pytest.fixture
+    def dtype(self):
+        """
+        Class level fixture of dtype for TestIntervalDtype
+        """
+        return IntervalDtype("int64", "right")
+
+    def test_hash_vs_equality(self, dtype):
+        # make sure that we satisfy is semantics
+        dtype2 = IntervalDtype("int64", "right")
+        dtype3 = IntervalDtype(dtype2)
+        assert dtype == dtype2
+        assert dtype2 == dtype
+        assert dtype3 == dtype
+        assert dtype is not dtype2
+        assert dtype2 is not dtype3
+        assert dtype3 is not dtype
+        assert hash(dtype) == hash(dtype2)
+        assert hash(dtype) == hash(dtype3)
+
+        dtype1 = IntervalDtype("interval")
+        dtype2 = IntervalDtype(dtype1)
+        dtype3 = IntervalDtype("interval")
+        assert dtype2 == dtype1
+        assert dtype2 == dtype2
+        assert dtype2 == dtype3
+        assert dtype2 is not dtype1
+        assert dtype2 is dtype2
+        assert dtype2 is not dtype3
+        assert hash(dtype2) == hash(dtype1)
+        assert hash(dtype2) == hash(dtype2)
+        assert hash(dtype2) == hash(dtype3)
+
+    @pytest.mark.parametrize(
+        "subtype", ["interval[int64]", "Interval[int64]", "int64", np.dtype("int64")]
+    )
+    def test_construction(self, subtype):
+        i = IntervalDtype(subtype, closed="right")
+        assert i.subtype == np.dtype("int64")
+        msg = "is_interval_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_interval_dtype(i)
+
+    @pytest.mark.parametrize(
+        "subtype", ["interval[int64]", "Interval[int64]", "int64", np.dtype("int64")]
+    )
+    def test_construction_allows_closed_none(self, subtype):
+        # GH#38394
+        dtype = IntervalDtype(subtype)
+
+        assert dtype.closed is None
+
+    def test_closed_mismatch(self):
+        msg = "'closed' keyword does not match value specified in dtype string"
+        with pytest.raises(ValueError, match=msg):
+            IntervalDtype("interval[int64, left]", "right")
+
+    @pytest.mark.parametrize("subtype", [None, "interval", "Interval"])
+    def test_construction_generic(self, subtype):
+        # generic
+        i = IntervalDtype(subtype)
+        assert i.subtype is None
+        msg = "is_interval_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_interval_dtype(i)
+
+    @pytest.mark.parametrize(
+        "subtype",
+        [
+            CategoricalDtype(list("abc"), False),
+            CategoricalDtype(list("wxyz"), True),
+            object,
+            str,
+            "<U10",
+            "interval[category]",
+            "interval[object]",
+        ],
+    )
+    def test_construction_not_supported(self, subtype):
+        # GH 19016
+        msg = (
+            "category, object, and string subtypes are not supported for IntervalDtype"
+        )
+        with pytest.raises(TypeError, match=msg):
+            IntervalDtype(subtype)
+
+    @pytest.mark.parametrize("subtype", ["xx", "IntervalA", "Interval[foo]"])
+    def test_construction_errors(self, subtype):
+        msg = "could not construct IntervalDtype"
+        with pytest.raises(TypeError, match=msg):
+            IntervalDtype(subtype)
+
+    def test_closed_must_match(self):
+        # GH#37933
+        dtype = IntervalDtype(np.float64, "left")
+
+        msg = "dtype.closed and 'closed' do not match"
+        with pytest.raises(ValueError, match=msg):
+            IntervalDtype(dtype, closed="both")
+
+    def test_closed_invalid(self):
+        with pytest.raises(ValueError, match="closed must be one of"):
+            IntervalDtype(np.float64, "foo")
+
+    def test_construction_from_string(self, dtype):
+        result = IntervalDtype("interval[int64, right]")
+        assert is_dtype_equal(dtype, result)
+        result = IntervalDtype.construct_from_string("interval[int64, right]")
+        assert is_dtype_equal(dtype, result)
+
+    @pytest.mark.parametrize("string", [0, 3.14, ("a", "b"), None])
+    def test_construction_from_string_errors(self, string):
+        # these are invalid entirely
+        msg = f"'construct_from_string' expects a string, got {type(string)}"
+
+        with pytest.raises(TypeError, match=re.escape(msg)):
+            IntervalDtype.construct_from_string(string)
+
+    @pytest.mark.parametrize("string", ["foo", "foo[int64]", "IntervalA"])
+    def test_construction_from_string_error_subtype(self, string):
+        # this is an invalid subtype
+        msg = (
+            "Incorrectly formatted string passed to constructor. "
+            r"Valid formats include Interval or Interval\[dtype\] "
+            "where dtype is numeric, datetime, or timedelta"
+        )
+
+        with pytest.raises(TypeError, match=msg):
+            IntervalDtype.construct_from_string(string)
+
+    def test_subclass(self):
+        a = IntervalDtype("interval[int64, right]")
+        b = IntervalDtype("interval[int64, right]")
+
+        assert issubclass(type(a), type(a))
+        assert issubclass(type(a), type(b))
+
+    def test_is_dtype(self, dtype):
+        assert IntervalDtype.is_dtype(dtype)
+        assert IntervalDtype.is_dtype("interval")
+        assert IntervalDtype.is_dtype(IntervalDtype("float64"))
+        assert IntervalDtype.is_dtype(IntervalDtype("int64"))
+        assert IntervalDtype.is_dtype(IntervalDtype(np.int64))
+        assert IntervalDtype.is_dtype(IntervalDtype("float64", "left"))
+        assert IntervalDtype.is_dtype(IntervalDtype("int64", "right"))
+        assert IntervalDtype.is_dtype(IntervalDtype(np.int64, "both"))
+
+        assert not IntervalDtype.is_dtype("D")
+        assert not IntervalDtype.is_dtype("3D")
+        assert not IntervalDtype.is_dtype("us")
+        assert not IntervalDtype.is_dtype("S")
+        assert not IntervalDtype.is_dtype("foo")
+        assert not IntervalDtype.is_dtype("IntervalA")
+        assert not IntervalDtype.is_dtype(np.object_)
+        assert not IntervalDtype.is_dtype(np.int64)
+        assert not IntervalDtype.is_dtype(np.float64)
+
+    def test_equality(self, dtype):
+        assert is_dtype_equal(dtype, "interval[int64, right]")
+        assert is_dtype_equal(dtype, IntervalDtype("int64", "right"))
+        assert is_dtype_equal(
+            IntervalDtype("int64", "right"), IntervalDtype("int64", "right")
+        )
+
+        assert not is_dtype_equal(dtype, "interval[int64]")
+        assert not is_dtype_equal(dtype, IntervalDtype("int64"))
+        assert not is_dtype_equal(
+            IntervalDtype("int64", "right"), IntervalDtype("int64")
+        )
+
+        assert not is_dtype_equal(dtype, "int64")
+        assert not is_dtype_equal(
+            IntervalDtype("int64", "neither"), IntervalDtype("float64", "right")
+        )
+        assert not is_dtype_equal(
+            IntervalDtype("int64", "both"), IntervalDtype("int64", "left")
+        )
+
+        # invalid subtype comparisons do not raise when directly compared
+        dtype1 = IntervalDtype("float64", "left")
+        dtype2 = IntervalDtype("datetime64[ns, US/Eastern]", "left")
+        assert dtype1 != dtype2
+        assert dtype2 != dtype1
+
+    @pytest.mark.parametrize(
+        "subtype",
+        [
+            None,
+            "interval",
+            "Interval",
+            "int64",
+            "uint64",
+            "float64",
+            "complex128",
+            "datetime64",
+            "timedelta64",
+            PeriodDtype("Q"),
+        ],
+    )
+    def test_equality_generic(self, subtype):
+        # GH 18980
+        closed = "right" if subtype is not None else None
+        dtype = IntervalDtype(subtype, closed=closed)
+        assert is_dtype_equal(dtype, "interval")
+        assert is_dtype_equal(dtype, IntervalDtype())
+
+    @pytest.mark.parametrize(
+        "subtype",
+        [
+            "int64",
+            "uint64",
+            "float64",
+            "complex128",
+            "datetime64",
+            "timedelta64",
+            PeriodDtype("Q"),
+        ],
+    )
+    def test_name_repr(self, subtype):
+        # GH 18980
+        closed = "right" if subtype is not None else None
+        dtype = IntervalDtype(subtype, closed=closed)
+        expected = f"interval[{subtype}, {closed}]"
+        assert str(dtype) == expected
+        assert dtype.name == "interval"
+
+    @pytest.mark.parametrize("subtype", [None, "interval", "Interval"])
+    def test_name_repr_generic(self, subtype):
+        # GH 18980
+        dtype = IntervalDtype(subtype)
+        assert str(dtype) == "interval"
+        assert dtype.name == "interval"
+
+    def test_basic(self, dtype):
+        msg = "is_interval_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_interval_dtype(dtype)
+
+            ii = IntervalIndex.from_breaks(range(3))
+
+            assert is_interval_dtype(ii.dtype)
+            assert is_interval_dtype(ii)
+
+            s = Series(ii, name="A")
+
+            assert is_interval_dtype(s.dtype)
+            assert is_interval_dtype(s)
+
+    def test_basic_dtype(self):
+        msg = "is_interval_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_interval_dtype("interval[int64, both]")
+            assert is_interval_dtype(IntervalIndex.from_tuples([(0, 1)]))
+            assert is_interval_dtype(IntervalIndex.from_breaks(np.arange(4)))
+            assert is_interval_dtype(
+                IntervalIndex.from_breaks(date_range("20130101", periods=3))
+            )
+            assert not is_interval_dtype("U")
+            assert not is_interval_dtype("S")
+            assert not is_interval_dtype("foo")
+            assert not is_interval_dtype(np.object_)
+            assert not is_interval_dtype(np.int64)
+            assert not is_interval_dtype(np.float64)
+
+    def test_caching(self):
+        # GH 54184: Caching not shown to improve performance
+        IntervalDtype.reset_cache()
+        dtype = IntervalDtype("int64", "right")
+        assert len(IntervalDtype._cache_dtypes) == 0
+
+        IntervalDtype("interval")
+        assert len(IntervalDtype._cache_dtypes) == 0
+
+        IntervalDtype.reset_cache()
+        tm.round_trip_pickle(dtype)
+        assert len(IntervalDtype._cache_dtypes) == 0
+
+    def test_not_string(self):
+        # GH30568: though IntervalDtype has object kind, it cannot be string
+        assert not is_string_dtype(IntervalDtype())
+
+    def test_unpickling_without_closed(self):
+        # GH#38394
+        dtype = IntervalDtype("interval")
+
+        assert dtype._closed is None
+
+        tm.round_trip_pickle(dtype)
+
+    def test_dont_keep_ref_after_del(self):
+        # GH 54184
+        dtype = IntervalDtype("int64", "right")
+        ref = weakref.ref(dtype)
+        del dtype
+        assert ref() is None
+
+
+class TestCategoricalDtypeParametrized:
+    @pytest.mark.parametrize(
+        "categories",
+        [
+            list("abcd"),
+            np.arange(1000),
+            ["a", "b", 10, 2, 1.3, True],
+            [True, False],
+            date_range("2017", periods=4),
+        ],
+    )
+    def test_basic(self, categories, ordered):
+        c1 = CategoricalDtype(categories, ordered=ordered)
+        tm.assert_index_equal(c1.categories, pd.Index(categories))
+        assert c1.ordered is ordered
+
+    def test_order_matters(self):
+        categories = ["a", "b"]
+        c1 = CategoricalDtype(categories, ordered=True)
+        c2 = CategoricalDtype(categories, ordered=False)
+        c3 = CategoricalDtype(categories, ordered=None)
+        assert c1 is not c2
+        assert c1 is not c3
+
+    @pytest.mark.parametrize("ordered", [False, None])
+    def test_unordered_same(self, ordered):
+        c1 = CategoricalDtype(["a", "b"], ordered=ordered)
+        c2 = CategoricalDtype(["b", "a"], ordered=ordered)
+        assert hash(c1) == hash(c2)
+
+    def test_categories(self):
+        result = CategoricalDtype(["a", "b", "c"])
+        tm.assert_index_equal(result.categories, pd.Index(["a", "b", "c"]))
+        assert result.ordered is False
+
+    def test_equal_but_different(self):
+        c1 = CategoricalDtype([1, 2, 3])
+        c2 = CategoricalDtype([1.0, 2.0, 3.0])
+        assert c1 is not c2
+        assert c1 != c2
+
+    def test_equal_but_different_mixed_dtypes(self):
+        c1 = CategoricalDtype([1, 2, "3"])
+        c2 = CategoricalDtype(["3", 1, 2])
+        assert c1 is not c2
+        assert c1 == c2
+
+    def test_equal_empty_ordered(self):
+        c1 = CategoricalDtype([], ordered=True)
+        c2 = CategoricalDtype([], ordered=True)
+        assert c1 is not c2
+        assert c1 == c2
+
+    def test_equal_empty_unordered(self):
+        c1 = CategoricalDtype([])
+        c2 = CategoricalDtype([])
+        assert c1 is not c2
+        assert c1 == c2
+
+    @pytest.mark.parametrize("v1, v2", [([1, 2, 3], [1, 2, 3]), ([1, 2, 3], [3, 2, 1])])
+    def test_order_hashes_different(self, v1, v2):
+        c1 = CategoricalDtype(v1, ordered=False)
+        c2 = CategoricalDtype(v2, ordered=True)
+        c3 = CategoricalDtype(v1, ordered=None)
+        assert c1 is not c2
+        assert c1 is not c3
+
+    def test_nan_invalid(self):
+        msg = "Categorical categories cannot be null"
+        with pytest.raises(ValueError, match=msg):
+            CategoricalDtype([1, 2, np.nan])
+
+    def test_non_unique_invalid(self):
+        msg = "Categorical categories must be unique"
+        with pytest.raises(ValueError, match=msg):
+            CategoricalDtype([1, 2, 1])
+
+    def test_same_categories_different_order(self):
+        c1 = CategoricalDtype(["a", "b"], ordered=True)
+        c2 = CategoricalDtype(["b", "a"], ordered=True)
+        assert c1 is not c2
+
+    @pytest.mark.parametrize("ordered2", [True, False, None])
+    def test_categorical_equality(self, ordered, ordered2):
+        # same categories, same order
+        # any combination of None/False are equal
+        # True/True is the only combination with True that are equal
+        c1 = CategoricalDtype(list("abc"), ordered)
+        c2 = CategoricalDtype(list("abc"), ordered2)
+        result = c1 == c2
+        expected = bool(ordered) is bool(ordered2)
+        assert result is expected
+
+        # same categories, different order
+        # any combination of None/False are equal (order doesn't matter)
+        # any combination with True are not equal (different order of cats)
+        c1 = CategoricalDtype(list("abc"), ordered)
+        c2 = CategoricalDtype(list("cab"), ordered2)
+        result = c1 == c2
+        expected = (bool(ordered) is False) and (bool(ordered2) is False)
+        assert result is expected
+
+        # different categories
+        c2 = CategoricalDtype([1, 2, 3], ordered2)
+        assert c1 != c2
+
+        # none categories
+        c1 = CategoricalDtype(list("abc"), ordered)
+        c2 = CategoricalDtype(None, ordered2)
+        c3 = CategoricalDtype(None, ordered)
+        assert c1 != c2
+        assert c2 != c1
+        assert c2 == c3
+
+    def test_categorical_dtype_equality_requires_categories(self):
+        # CategoricalDtype with categories=None is *not* equal to
+        #  any fully-initialized CategoricalDtype
+        first = CategoricalDtype(["a", "b"])
+        second = CategoricalDtype()
+        third = CategoricalDtype(ordered=True)
+
+        assert second == second
+        assert third == third
+
+        assert first != second
+        assert second != first
+        assert first != third
+        assert third != first
+        assert second == third
+        assert third == second
+
+    @pytest.mark.parametrize("categories", [list("abc"), None])
+    @pytest.mark.parametrize("other", ["category", "not a category"])
+    def test_categorical_equality_strings(self, categories, ordered, other):
+        c1 = CategoricalDtype(categories, ordered)
+        result = c1 == other
+        expected = other == "category"
+        assert result is expected
+
+    def test_invalid_raises(self):
+        with pytest.raises(TypeError, match="ordered"):
+            CategoricalDtype(["a", "b"], ordered="foo")
+
+        with pytest.raises(TypeError, match="'categories' must be list-like"):
+            CategoricalDtype("category")
+
+    def test_mixed(self):
+        a = CategoricalDtype(["a", "b", 1, 2])
+        b = CategoricalDtype(["a", "b", "1", "2"])
+        assert hash(a) != hash(b)
+
+    def test_from_categorical_dtype_identity(self):
+        c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)
+        # Identity test for no changes
+        c2 = CategoricalDtype._from_categorical_dtype(c1)
+        assert c2 is c1
+
+    def test_from_categorical_dtype_categories(self):
+        c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)
+        # override categories
+        result = CategoricalDtype._from_categorical_dtype(c1, categories=[2, 3])
+        assert result == CategoricalDtype([2, 3], ordered=True)
+
+    def test_from_categorical_dtype_ordered(self):
+        c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)
+        # override ordered
+        result = CategoricalDtype._from_categorical_dtype(c1, ordered=False)
+        assert result == CategoricalDtype([1, 2, 3], ordered=False)
+
+    def test_from_categorical_dtype_both(self):
+        c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)
+        # override ordered
+        result = CategoricalDtype._from_categorical_dtype(
+            c1, categories=[1, 2], ordered=False
+        )
+        assert result == CategoricalDtype([1, 2], ordered=False)
+
+    def test_str_vs_repr(self, ordered, using_infer_string):
+        c1 = CategoricalDtype(["a", "b"], ordered=ordered)
+        assert str(c1) == "category"
+        # Py2 will have unicode prefixes
+        dtype = "str" if using_infer_string else "object"
+        pat = (
+            r"CategoricalDtype\(categories=\[.*\], ordered={ordered}, "
+            rf"categories_dtype={dtype}\)"
+        )
+        assert re.match(pat.format(ordered=ordered), repr(c1))
+
+    def test_categorical_categories(self):
+        # GH17884
+        c1 = CategoricalDtype(Categorical(["a", "b"]))
+        tm.assert_index_equal(c1.categories, pd.Index(["a", "b"]))
+        c1 = CategoricalDtype(CategoricalIndex(["a", "b"]))
+        tm.assert_index_equal(c1.categories, pd.Index(["a", "b"]))
+
+    @pytest.mark.parametrize(
+        "new_categories", [list("abc"), list("cba"), list("wxyz"), None]
+    )
+    @pytest.mark.parametrize("new_ordered", [True, False, None])
+    def test_update_dtype(self, ordered, new_categories, new_ordered):
+        original_categories = list("abc")
+        dtype = CategoricalDtype(original_categories, ordered)
+        new_dtype = CategoricalDtype(new_categories, new_ordered)
+
+        result = dtype.update_dtype(new_dtype)
+        expected_categories = pd.Index(new_categories or original_categories)
+        expected_ordered = new_ordered if new_ordered is not None else dtype.ordered
+
+        tm.assert_index_equal(result.categories, expected_categories)
+        assert result.ordered is expected_ordered
+
+    def test_update_dtype_string(self, ordered):
+        dtype = CategoricalDtype(list("abc"), ordered)
+        expected_categories = dtype.categories
+        expected_ordered = dtype.ordered
+        result = dtype.update_dtype("category")
+        tm.assert_index_equal(result.categories, expected_categories)
+        assert result.ordered is expected_ordered
+
+    @pytest.mark.parametrize("bad_dtype", ["foo", object, np.int64, PeriodDtype("Q")])
+    def test_update_dtype_errors(self, bad_dtype):
+        dtype = CategoricalDtype(list("abc"), False)
+        msg = "a CategoricalDtype must be passed to perform an update, "
+        with pytest.raises(ValueError, match=msg):
+            dtype.update_dtype(bad_dtype)
+
+
+@pytest.mark.parametrize(
+    "dtype", [CategoricalDtype, IntervalDtype, DatetimeTZDtype, PeriodDtype]
+)
+def test_registry(dtype):
+    assert dtype in registry.dtypes
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        ("int64", None),
+        ("interval", IntervalDtype()),
+        ("interval[int64, neither]", IntervalDtype()),
+        ("interval[datetime64[ns], left]", IntervalDtype("datetime64[ns]", "left")),
+        ("period[D]", PeriodDtype("D")),
+        ("category", CategoricalDtype()),
+        ("datetime64[ns, US/Eastern]", DatetimeTZDtype("ns", "US/Eastern")),
+    ],
+)
+def test_registry_find(dtype, expected):
+    assert registry.find(dtype) == expected
+
+
+@pytest.mark.parametrize(
+    "dtype, expected",
+    [
+        (str, False),
+        (int, False),
+        (bool, True),
+        (np.bool_, True),
+        (np.array(["a", "b"]), False),
+        (Series([1, 2]), False),
+        (np.array([True, False]), True),
+        (Series([True, False]), True),
+        (SparseArray([True, False]), True),
+        (SparseDtype(bool), True),
+    ],
+)
+def test_is_bool_dtype(dtype, expected):
+    result = is_bool_dtype(dtype)
+    assert result is expected
+
+
+def test_is_bool_dtype_sparse():
+    result = is_bool_dtype(Series(SparseArray([True, False])))
+    assert result is True
+
+
+@pytest.mark.parametrize(
+    "check",
+    [
+        is_categorical_dtype,
+        is_datetime64tz_dtype,
+        is_period_dtype,
+        is_datetime64_ns_dtype,
+        is_datetime64_dtype,
+        is_interval_dtype,
+        is_datetime64_any_dtype,
+        is_string_dtype,
+        is_bool_dtype,
+    ],
+)
+def test_is_dtype_no_warning(check):
+    data = pd.DataFrame({"A": [1, 2]})
+
+    warn = None
+    msg = f"{check.__name__} is deprecated"
+    if (
+        check is is_categorical_dtype
+        or check is is_interval_dtype
+        or check is is_datetime64tz_dtype
+        or check is is_period_dtype
+    ):
+        warn = DeprecationWarning
+
+    with tm.assert_produces_warning(warn, match=msg):
+        check(data)
+
+    with tm.assert_produces_warning(warn, match=msg):
+        check(data["A"])
+
+
+def test_period_dtype_compare_to_string():
+    # https://github.com/pandas-dev/pandas/issues/37265
+    dtype = PeriodDtype(freq="M")
+    assert (dtype == "period[M]") is True
+    assert (dtype != "period[M]") is False
+
+
+def test_compare_complex_dtypes():
+    # GH 28050
+    df = pd.DataFrame(np.arange(5).astype(np.complex128))
+    msg = "'<' not supported between instances of 'complex' and 'complex'"
+
+    with pytest.raises(TypeError, match=msg):
+        df < df.astype(object)
+
+    with pytest.raises(TypeError, match=msg):
+        df.lt(df.astype(object))
+
+
+def test_cast_string_to_complex():
+    # GH 4895
+    expected = pd.DataFrame(["1.0+5j", "1.5-3j"], dtype=complex)
+    result = pd.DataFrame(["1.0+5j", "1.5-3j"]).astype(complex)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_complex():
+    result = Categorical([1, 2 + 2j])
+    expected = Categorical([1.0 + 0.0j, 2.0 + 2.0j])
+    tm.assert_categorical_equal(result, expected)
+    result = Categorical([1, 2, 2 + 2j])
+    expected = Categorical([1.0 + 0.0j, 2.0 + 0.0j, 2.0 + 2.0j])
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_multi_column_dtype_assignment():
+    # GH #27583
+    df = pd.DataFrame({"a": [0.0], "b": 0.0})
+    expected = pd.DataFrame({"a": [0], "b": 0})
+
+    df[["a", "b"]] = 0
+    tm.assert_frame_equal(df, expected)
+
+    df["b"] = 0
+    tm.assert_frame_equal(df, expected)
+
+
+def test_loc_setitem_empty_labels_no_dtype_conversion():
+    # GH 29707
+
+    df = pd.DataFrame({"a": [2, 3]})
+    expected = df.copy()
+    assert df.a.dtype == "int64"
+    df.loc[[]] = 0.1
+
+    assert df.a.dtype == "int64"
+    tm.assert_frame_equal(df, expected)
+
+
+def test_categorical_nan_no_dtype_conversion():
+    # GH 43996
+
+    df = pd.DataFrame({"a": Categorical([np.nan], [1]), "b": [1]})
+    expected = pd.DataFrame({"a": Categorical([1], [1]), "b": [1]})
+    df.loc[0, "a"] = np.array([1])
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/dtypes/test_generic.py b/pandas/pandas/tests/dtypes/test_generic.py
new file mode 100644
index 0000000000000..2b90886a8d070
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_generic.py
@@ -0,0 +1,130 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes import generic as gt
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestABCClasses:
+    tuples = [[1, 2, 2], ["red", "blue", "red"]]
+    multi_index = pd.MultiIndex.from_arrays(tuples, names=("number", "color"))
+    datetime_index = pd.to_datetime(["2000/1/1", "2010/1/1"])
+    timedelta_index = pd.to_timedelta(np.arange(5), unit="s")
+    period_index = pd.period_range("2000/1/1", "2010/1/1/", freq="M")
+    categorical = pd.Categorical([1, 2, 3], categories=[2, 3, 1])
+    categorical_df = pd.DataFrame({"values": [1, 2, 3]}, index=categorical)
+    df = pd.DataFrame({"names": ["a", "b", "c"]}, index=multi_index)
+    sparse_array = pd.arrays.SparseArray(np.random.default_rng(2).standard_normal(10))
+
+    datetime_array = datetime_index.array
+    timedelta_array = timedelta_index.array
+
+    abc_pairs = [
+        ("ABCMultiIndex", multi_index),
+        ("ABCDatetimeIndex", datetime_index),
+        ("ABCRangeIndex", pd.RangeIndex(3)),
+        ("ABCTimedeltaIndex", timedelta_index),
+        ("ABCIntervalIndex", pd.interval_range(start=0, end=3)),
+        (
+            "ABCPeriodArray",
+            pd.arrays.PeriodArray([2000, 2001, 2002], dtype="period[D]"),
+        ),
+        ("ABCNumpyExtensionArray", pd.arrays.NumpyExtensionArray(np.array([0, 1, 2]))),
+        ("ABCPeriodIndex", period_index),
+        ("ABCCategoricalIndex", categorical_df.index),
+        ("ABCSeries", pd.Series([1, 2, 3])),
+        ("ABCDataFrame", df),
+        ("ABCCategorical", categorical),
+        ("ABCDatetimeArray", datetime_array),
+        ("ABCTimedeltaArray", timedelta_array),
+    ]
+
+    @pytest.mark.parametrize("abctype1, inst", abc_pairs)
+    @pytest.mark.parametrize("abctype2, _", abc_pairs)
+    def test_abc_pairs_instance_check(self, abctype1, abctype2, inst, _):
+        # GH 38588, 46719
+        if abctype1 == abctype2:
+            assert isinstance(inst, getattr(gt, abctype2))
+            assert not isinstance(type(inst), getattr(gt, abctype2))
+        else:
+            assert not isinstance(inst, getattr(gt, abctype2))
+
+    @pytest.mark.parametrize("abctype1, inst", abc_pairs)
+    @pytest.mark.parametrize("abctype2, _", abc_pairs)
+    def test_abc_pairs_subclass_check(self, abctype1, abctype2, inst, _):
+        # GH 38588, 46719
+        if abctype1 == abctype2:
+            assert issubclass(type(inst), getattr(gt, abctype2))
+
+            with pytest.raises(
+                TypeError, match=re.escape("issubclass() arg 1 must be a class")
+            ):
+                issubclass(inst, getattr(gt, abctype2))
+        else:
+            assert not issubclass(type(inst), getattr(gt, abctype2))
+
+    abc_subclasses = {
+        "ABCIndex": [
+            abctype
+            for abctype, _ in abc_pairs
+            if "Index" in abctype and abctype != "ABCIndex"
+        ],
+        "ABCNDFrame": ["ABCSeries", "ABCDataFrame"],
+        "ABCExtensionArray": [
+            "ABCCategorical",
+            "ABCDatetimeArray",
+            "ABCPeriodArray",
+            "ABCTimedeltaArray",
+        ],
+    }
+
+    @pytest.mark.parametrize("parent, subs", abc_subclasses.items())
+    @pytest.mark.parametrize("abctype, inst", abc_pairs)
+    def test_abc_hierarchy(self, parent, subs, abctype, inst):
+        # GH 38588
+        if abctype in subs:
+            assert isinstance(inst, getattr(gt, parent))
+        else:
+            assert not isinstance(inst, getattr(gt, parent))
+
+    @pytest.mark.parametrize("abctype", [e for e in gt.__dict__ if e.startswith("ABC")])
+    def test_abc_coverage(self, abctype):
+        # GH 38588
+        assert (
+            abctype in (e for e, _ in self.abc_pairs) or abctype in self.abc_subclasses
+        )
+
+
+def test_setattr_warnings():
+    # GH7175 - GOTCHA: You can't use dot notation to add a column...
+    d = {
+        "one": pd.Series([1.0, 2.0, 3.0], index=["a", "b", "c"]),
+        "two": pd.Series([1.0, 2.0, 3.0, 4.0], index=["a", "b", "c", "d"]),
+    }
+    df = pd.DataFrame(d)
+
+    with tm.assert_produces_warning(None):
+        #  successfully add new column
+        #  this should not raise a warning
+        df["three"] = df.two + 1
+        assert df.three.sum() > df.two.sum()
+
+    with tm.assert_produces_warning(None):
+        #  successfully modify column in place
+        #  this should not raise a warning
+        df.one += 1
+        assert df.one.iloc[0] == 2
+
+    with tm.assert_produces_warning(None):
+        #  successfully add an attribute to a series
+        #  this should not raise a warning
+        df.two.not_an_index = [1, 2]
+
+    with tm.assert_produces_warning(UserWarning, match="doesn't allow columns"):
+        #  warn when setting column to nonexistent name
+        df.four = df.two + 2
+        assert df.four.sum() > df.two.sum()
diff --git a/pandas/pandas/tests/dtypes/test_inference.py b/pandas/pandas/tests/dtypes/test_inference.py
new file mode 100644
index 0000000000000..db98751324ebc
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_inference.py
@@ -0,0 +1,2052 @@
+"""
+These the test the public routines exposed in types/common.py
+related to inference and not otherwise tested in types/test_common.py
+
+"""
+
+import collections
+from collections import namedtuple
+from collections.abc import Iterator
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+    timezone,
+)
+from decimal import Decimal
+from fractions import Fraction
+from io import StringIO
+import itertools
+from numbers import Number
+import re
+import sys
+from typing import (
+    Generic,
+    TypeVar,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+    ops as libops,
+)
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes import inference
+from pandas.core.dtypes.cast import find_result_type
+from pandas.core.dtypes.common import (
+    ensure_int32,
+    is_bool,
+    is_complex,
+    is_datetime64_any_dtype,
+    is_datetime64_dtype,
+    is_datetime64_ns_dtype,
+    is_datetime64tz_dtype,
+    is_float,
+    is_integer,
+    is_number,
+    is_scalar,
+    is_scipy_sparse,
+    is_timedelta64_dtype,
+    is_timedelta64_ns_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DateOffset,
+    DatetimeIndex,
+    Index,
+    Interval,
+    Period,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    BooleanArray,
+    FloatingArray,
+    IntegerArray,
+)
+
+
+@pytest.fixture(params=[True, False], ids=str)
+def coerce(request):
+    return request.param
+
+
+class MockNumpyLikeArray:
+    """
+    A class which is numpy-like (e.g. Pint's Quantity) but not actually numpy
+
+    The key is that it is not actually a numpy array so
+    ``util.is_array(mock_numpy_like_array_instance)`` returns ``False``. Other
+    important properties are that the class defines a :meth:`__iter__` method
+    (so that ``isinstance(abc.Iterable)`` returns ``True``) and has a
+    :meth:`ndim` property, as pandas special-cases 0-dimensional arrays in some
+    cases.
+
+    We expect pandas to behave with respect to such duck arrays exactly as
+    with real numpy arrays. In particular, a 0-dimensional duck array is *NOT*
+    a scalar (`is_scalar(np.array(1)) == False`), but it is not list-like either.
+    """
+
+    def __init__(self, values) -> None:
+        self._values = values
+
+    def __iter__(self) -> Iterator:
+        iter_values = iter(self._values)
+
+        def it_outer():
+            yield from iter_values
+
+        return it_outer()
+
+    def __len__(self) -> int:
+        return len(self._values)
+
+    def __array__(self, dtype=None, copy=None):
+        return np.asarray(self._values, dtype=dtype)
+
+    @property
+    def ndim(self):
+        return self._values.ndim
+
+    @property
+    def dtype(self):
+        return self._values.dtype
+
+    @property
+    def size(self):
+        return self._values.size
+
+    @property
+    def shape(self):
+        return self._values.shape
+
+
+# collect all objects to be tested for list-like-ness; use tuples of objects,
+# whether they are list-like or not (special casing for sets), and their ID
+ll_params = [
+    ([1], True, "list"),
+    ([], True, "list-empty"),
+    ((1,), True, "tuple"),
+    ((), True, "tuple-empty"),
+    ({"a": 1}, True, "dict"),
+    ({}, True, "dict-empty"),
+    ({"a", 1}, "set", "set"),
+    (set(), "set", "set-empty"),
+    (frozenset({"a", 1}), "set", "frozenset"),
+    (frozenset(), "set", "frozenset-empty"),
+    (iter([1, 2]), True, "iterator"),
+    (iter([]), True, "iterator-empty"),
+    ((x for x in [1, 2]), True, "generator"),
+    ((_ for _ in []), True, "generator-empty"),
+    (Series([1]), True, "Series"),
+    (Series([], dtype=object), True, "Series-empty"),
+    # Series.str will still raise a TypeError if iterated
+    (Series(["a"]).str, True, "StringMethods"),
+    (Series([], dtype="O").str, True, "StringMethods-empty"),
+    (Index([1]), True, "Index"),
+    (Index([]), True, "Index-empty"),
+    (DataFrame([[1]]), True, "DataFrame"),
+    (DataFrame(), True, "DataFrame-empty"),
+    (np.ndarray((2,) * 1), True, "ndarray-1d"),
+    (np.array([]), True, "ndarray-1d-empty"),
+    (np.ndarray((2,) * 2), True, "ndarray-2d"),
+    (np.array([[]]), True, "ndarray-2d-empty"),
+    (np.ndarray((2,) * 3), True, "ndarray-3d"),
+    (np.array([[[]]]), True, "ndarray-3d-empty"),
+    (np.ndarray((2,) * 4), True, "ndarray-4d"),
+    (np.array([[[[]]]]), True, "ndarray-4d-empty"),
+    (np.array(2), False, "ndarray-0d"),
+    (MockNumpyLikeArray(np.ndarray((2,) * 1)), True, "duck-ndarray-1d"),
+    (MockNumpyLikeArray(np.array([])), True, "duck-ndarray-1d-empty"),
+    (MockNumpyLikeArray(np.ndarray((2,) * 2)), True, "duck-ndarray-2d"),
+    (MockNumpyLikeArray(np.array([[]])), True, "duck-ndarray-2d-empty"),
+    (MockNumpyLikeArray(np.ndarray((2,) * 3)), True, "duck-ndarray-3d"),
+    (MockNumpyLikeArray(np.array([[[]]])), True, "duck-ndarray-3d-empty"),
+    (MockNumpyLikeArray(np.ndarray((2,) * 4)), True, "duck-ndarray-4d"),
+    (MockNumpyLikeArray(np.array([[[[]]]])), True, "duck-ndarray-4d-empty"),
+    (MockNumpyLikeArray(np.array(2)), False, "duck-ndarray-0d"),
+    (1, False, "int"),
+    (b"123", False, "bytes"),
+    (b"", False, "bytes-empty"),
+    ("123", False, "string"),
+    ("", False, "string-empty"),
+    (str, False, "string-type"),
+    (object(), False, "object"),
+    (np.nan, False, "NaN"),
+    (None, False, "None"),
+]
+objs, expected, ids = zip(*ll_params)
+
+
+@pytest.fixture(params=zip(objs, expected), ids=ids)
+def maybe_list_like(request):
+    return request.param
+
+
+def test_is_list_like(maybe_list_like):
+    obj, expected = maybe_list_like
+    expected = True if expected == "set" else expected
+    assert inference.is_list_like(obj) == expected
+
+
+def test_is_list_like_disallow_sets(maybe_list_like):
+    obj, expected = maybe_list_like
+    expected = False if expected == "set" else expected
+    assert inference.is_list_like(obj, allow_sets=False) == expected
+
+
+def test_is_list_like_recursion():
+    # GH 33721
+    # interpreter would crash with SIGABRT
+    def list_like():
+        inference.is_list_like([])
+        list_like()
+
+    rec_limit = sys.getrecursionlimit()
+    try:
+        # Limit to avoid stack overflow on Windows CI
+        sys.setrecursionlimit(100)
+        with tm.external_error_raised(RecursionError):
+            list_like()
+    finally:
+        sys.setrecursionlimit(rec_limit)
+
+
+def test_is_list_like_iter_is_none():
+    # GH 43373
+    # is_list_like was yielding false positives with __iter__ == None
+    class NotListLike:
+        def __getitem__(self, item):
+            return self
+
+        __iter__ = None
+
+    assert not inference.is_list_like(NotListLike())
+
+
+def test_is_list_like_generic():
+    # GH 49649
+    # is_list_like was yielding false positives for Generic classes in python 3.11
+    T = TypeVar("T")
+
+    class MyDataFrame(DataFrame, Generic[T]): ...
+
+    tstc = MyDataFrame[int]
+    tst = MyDataFrame[int]({"x": [1, 2, 3]})
+
+    assert not inference.is_list_like(tstc)
+    assert isinstance(tst, DataFrame)
+    assert inference.is_list_like(tst)
+
+
+def test_is_sequence():
+    is_seq = inference.is_sequence
+    assert is_seq((1, 2))
+    assert is_seq([1, 2])
+    assert not is_seq("abcd")
+    assert not is_seq(np.int64)
+
+    class A:
+        def __getitem__(self, item):
+            return 1
+
+    assert not is_seq(A())
+
+
+def test_is_array_like():
+    assert inference.is_array_like(Series([], dtype=object))
+    assert inference.is_array_like(Series([1, 2]))
+    assert inference.is_array_like(np.array(["a", "b"]))
+    assert inference.is_array_like(Index(["2016-01-01"]))
+    assert inference.is_array_like(np.array([2, 3]))
+    assert inference.is_array_like(MockNumpyLikeArray(np.array([2, 3])))
+
+    class DtypeList(list):
+        dtype = "special"
+
+    assert inference.is_array_like(DtypeList())
+
+    assert not inference.is_array_like([1, 2, 3])
+    assert not inference.is_array_like(())
+    assert not inference.is_array_like("foo")
+    assert not inference.is_array_like(123)
+
+
+@pytest.mark.parametrize(
+    "inner",
+    [
+        [],
+        [1],
+        (1,),
+        (1, 2),
+        {"a": 1},
+        {1, "a"},
+        Series([1]),
+        Series([], dtype=object),
+        Series(["a"]).str,
+        (x for x in range(5)),
+    ],
+)
+@pytest.mark.parametrize("outer", [list, Series, np.array, tuple])
+def test_is_nested_list_like_passes(inner, outer):
+    result = outer([inner for _ in range(5)])
+    assert inference.is_list_like(result)
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        "abc",
+        [],
+        [1],
+        (1,),
+        ["a"],
+        "a",
+        {"a"},
+        [1, 2, 3],
+        Series([1]),
+        DataFrame({"A": [1]}),
+        ([1, 2] for _ in range(5)),
+    ],
+)
+def test_is_nested_list_like_fails(obj):
+    assert not inference.is_nested_list_like(obj)
+
+
+@pytest.mark.parametrize("ll", [{}, {"A": 1}, Series([1]), collections.defaultdict()])
+def test_is_dict_like_passes(ll):
+    assert inference.is_dict_like(ll)
+
+
+@pytest.mark.parametrize(
+    "ll",
+    [
+        "1",
+        1,
+        [1, 2],
+        (1, 2),
+        range(2),
+        Index([1]),
+        dict,
+        collections.defaultdict,
+        Series,
+    ],
+)
+def test_is_dict_like_fails(ll):
+    assert not inference.is_dict_like(ll)
+
+
+@pytest.mark.parametrize("has_keys", [True, False])
+@pytest.mark.parametrize("has_getitem", [True, False])
+@pytest.mark.parametrize("has_contains", [True, False])
+def test_is_dict_like_duck_type(has_keys, has_getitem, has_contains):
+    class DictLike:
+        def __init__(self, d) -> None:
+            self.d = d
+
+        if has_keys:
+
+            def keys(self):
+                return self.d.keys()
+
+        if has_getitem:
+
+            def __getitem__(self, key):
+                return self.d.__getitem__(key)
+
+        if has_contains:
+
+            def __contains__(self, key) -> bool:
+                return self.d.__contains__(key)
+
+    d = DictLike({1: 2})
+    result = inference.is_dict_like(d)
+    expected = has_keys and has_getitem and has_contains
+
+    assert result is expected
+
+
+def test_is_file_like():
+    class MockFile:
+        pass
+
+    is_file = inference.is_file_like
+
+    data = StringIO("data")
+    assert is_file(data)
+
+    # No read / write attributes
+    # No iterator attributes
+    m = MockFile()
+    assert not is_file(m)
+
+    MockFile.write = lambda self: 0
+
+    # Write attribute but not an iterator
+    m = MockFile()
+    assert not is_file(m)
+
+    # gh-16530: Valid iterator just means we have the
+    # __iter__ attribute for our purposes.
+    MockFile.__iter__ = lambda self: self
+
+    # Valid write-only file
+    m = MockFile()
+    assert is_file(m)
+
+    del MockFile.write
+    MockFile.read = lambda self: 0
+
+    # Valid read-only file
+    m = MockFile()
+    assert is_file(m)
+
+    # Iterator but no read / write attributes
+    data = [1, 2, 3]
+    assert not is_file(data)
+
+
+test_tuple = collections.namedtuple("test_tuple", ["a", "b", "c"])
+
+
+@pytest.mark.parametrize("ll", [test_tuple(1, 2, 3)])
+def test_is_names_tuple_passes(ll):
+    assert inference.is_named_tuple(ll)
+
+
+@pytest.mark.parametrize("ll", [(1, 2, 3), "a", Series({"pi": 3.14})])
+def test_is_names_tuple_fails(ll):
+    assert not inference.is_named_tuple(ll)
+
+
+def test_is_hashable():
+    # all new-style classes are hashable by default
+    class HashableClass:
+        pass
+
+    class UnhashableClass1:
+        __hash__ = None
+
+    class UnhashableClass2:
+        def __hash__(self):
+            raise TypeError("Not hashable")
+
+    hashable = (1, 3.14, np.float64(3.14), "a", (), (1,), HashableClass())
+    not_hashable = ([], UnhashableClass1())
+    abc_hashable_not_really_hashable = (([],), UnhashableClass2())
+
+    for i in hashable:
+        assert inference.is_hashable(i)
+    for i in not_hashable:
+        assert not inference.is_hashable(i)
+    for i in abc_hashable_not_really_hashable:
+        assert not inference.is_hashable(i)
+
+    # numpy.array is no longer collections.abc.Hashable as of
+    # https://github.com/numpy/numpy/pull/5326, just test
+    # is_hashable()
+    assert not inference.is_hashable(np.array([]))
+
+
+@pytest.mark.parametrize("ll", [re.compile("ad")])
+def test_is_re_passes(ll):
+    assert inference.is_re(ll)
+
+
+@pytest.mark.parametrize("ll", ["x", 2, 3, object()])
+def test_is_re_fails(ll):
+    assert not inference.is_re(ll)
+
+
+@pytest.mark.parametrize(
+    "ll", [r"a", "x", r"asdf", re.compile("adsf"), r"\u2233\s*", re.compile(r"")]
+)
+def test_is_recompilable_passes(ll):
+    assert inference.is_re_compilable(ll)
+
+
+@pytest.mark.parametrize("ll", [1, [], object()])
+def test_is_recompilable_fails(ll):
+    assert not inference.is_re_compilable(ll)
+
+
+class TestInference:
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            np.array(list("abc"), dtype="S1"),
+            np.array(list("abc"), dtype="S1").astype(object),
+            [b"a", np.nan, b"c"],
+        ],
+    )
+    def test_infer_dtype_bytes(self, arr):
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "bytes"
+
+    @pytest.mark.parametrize(
+        "value, expected",
+        [
+            (float("inf"), True),
+            (np.inf, True),
+            (-np.inf, False),
+            (1, False),
+            ("a", False),
+        ],
+    )
+    def test_isposinf_scalar(self, value, expected):
+        # GH 11352
+        result = libmissing.isposinf_scalar(value)
+        assert result is expected
+
+    @pytest.mark.parametrize(
+        "value, expected",
+        [
+            (float("-inf"), True),
+            (-np.inf, True),
+            (np.inf, False),
+            (1, False),
+            ("a", False),
+        ],
+    )
+    def test_isneginf_scalar(self, value, expected):
+        result = libmissing.isneginf_scalar(value)
+        assert result is expected
+
+    @pytest.mark.parametrize(
+        "convert_to_masked_nullable, exp",
+        [
+            (
+                True,
+                BooleanArray(
+                    np.array([True, False], dtype="bool"), np.array([False, True])
+                ),
+            ),
+            (False, np.array([True, np.nan], dtype="object")),
+        ],
+    )
+    def test_maybe_convert_nullable_boolean(self, convert_to_masked_nullable, exp):
+        # GH 40687
+        arr = np.array([True, np.nan], dtype=object)
+        result = libops.maybe_convert_bool(
+            arr, set(), convert_to_masked_nullable=convert_to_masked_nullable
+        )
+        if convert_to_masked_nullable:
+            tm.assert_extension_array_equal(BooleanArray(*result), exp)
+        else:
+            result = result[0]
+            tm.assert_numpy_array_equal(result, exp)
+
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    @pytest.mark.parametrize("coerce_numeric", [True, False])
+    @pytest.mark.parametrize(
+        "infinity", ["inf", "inF", "iNf", "Inf", "iNF", "InF", "INf", "INF"]
+    )
+    @pytest.mark.parametrize("prefix", ["", "-", "+"])
+    def test_maybe_convert_numeric_infinities(
+        self, coerce_numeric, infinity, prefix, convert_to_masked_nullable
+    ):
+        # see gh-13274
+        result, _ = lib.maybe_convert_numeric(
+            np.array([prefix + infinity], dtype=object),
+            na_values={"", "NULL", "nan"},
+            coerce_numeric=coerce_numeric,
+            convert_to_masked_nullable=convert_to_masked_nullable,
+        )
+        expected = np.array([np.inf if prefix in ["", "+"] else -np.inf])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    def test_maybe_convert_numeric_infinities_raises(self, convert_to_masked_nullable):
+        msg = "Unable to parse string"
+        with pytest.raises(ValueError, match=msg):
+            lib.maybe_convert_numeric(
+                np.array(["foo_inf"], dtype=object),
+                na_values={"", "NULL", "nan"},
+                coerce_numeric=False,
+                convert_to_masked_nullable=convert_to_masked_nullable,
+            )
+
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    def test_maybe_convert_numeric_post_floatify_nan(
+        self, coerce, convert_to_masked_nullable
+    ):
+        # see gh-13314
+        data = np.array(["1.200", "-999.000", "4.500"], dtype=object)
+        expected = np.array([1.2, np.nan, 4.5], dtype=np.float64)
+        nan_values = {-999, -999.0}
+
+        out = lib.maybe_convert_numeric(
+            data,
+            nan_values,
+            coerce,
+            convert_to_masked_nullable=convert_to_masked_nullable,
+        )
+        if convert_to_masked_nullable:
+            expected = FloatingArray(expected, np.isnan(expected))
+            tm.assert_extension_array_equal(expected, FloatingArray(*out))
+        else:
+            out = out[0]
+            tm.assert_numpy_array_equal(out, expected)
+
+    def test_convert_infs(self):
+        arr = np.array(["inf", "inf", "inf"], dtype="O")
+        result, _ = lib.maybe_convert_numeric(arr, set(), False)
+        assert result.dtype == np.float64
+
+        arr = np.array(["-inf", "-inf", "-inf"], dtype="O")
+        result, _ = lib.maybe_convert_numeric(arr, set(), False)
+        assert result.dtype == np.float64
+
+    def test_scientific_no_exponent(self):
+        # See PR 12215
+        arr = np.array(["42E", "2E", "99e", "6e"], dtype="O")
+        result, _ = lib.maybe_convert_numeric(arr, set(), False, True)
+        assert np.all(np.isnan(result))
+
+    def test_convert_non_hashable(self):
+        # GH13324
+        # make sure that we are handing non-hashables
+        arr = np.array([[10.0, 2], 1.0, "apple"], dtype=object)
+        result, _ = lib.maybe_convert_numeric(arr, set(), False, True)
+        tm.assert_numpy_array_equal(result, np.array([np.nan, 1.0, np.nan]))
+
+    def test_convert_numeric_uint64(self):
+        arr = np.array([2**63], dtype=object)
+        exp = np.array([2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(lib.maybe_convert_numeric(arr, set())[0], exp)
+
+        arr = np.array([str(2**63)], dtype=object)
+        exp = np.array([2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(lib.maybe_convert_numeric(arr, set())[0], exp)
+
+        arr = np.array([np.uint64(2**63)], dtype=object)
+        exp = np.array([2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(lib.maybe_convert_numeric(arr, set())[0], exp)
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            np.array([2**63, np.nan], dtype=object),
+            np.array([str(2**63), np.nan], dtype=object),
+            np.array([np.nan, 2**63], dtype=object),
+            np.array([np.nan, str(2**63)], dtype=object),
+        ],
+    )
+    def test_convert_numeric_uint64_nan(self, coerce, arr):
+        expected = arr.astype(float) if coerce else arr.copy()
+        result, _ = lib.maybe_convert_numeric(arr, set(), coerce_numeric=coerce)
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    def test_convert_numeric_uint64_nan_values(
+        self, coerce, convert_to_masked_nullable
+    ):
+        arr = np.array([2**63, 2**63 + 1], dtype=object)
+        na_values = {2**63}
+
+        expected = np.array([np.nan, 2**63 + 1], dtype=float) if coerce else arr.copy()
+        result = lib.maybe_convert_numeric(
+            arr,
+            na_values,
+            coerce_numeric=coerce,
+            convert_to_masked_nullable=convert_to_masked_nullable,
+        )
+        if convert_to_masked_nullable and coerce:
+            expected = IntegerArray(
+                np.array([0, 2**63 + 1], dtype="u8"),
+                np.array([True, False], dtype="bool"),
+            )
+            result = IntegerArray(*result)
+        else:
+            result = result[0]  # discard mask
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            np.array([2**63, -1], dtype=object),
+            np.array([str(2**63), -1], dtype=object),
+            np.array([str(2**63), str(-1)], dtype=object),
+            np.array([-1, 2**63], dtype=object),
+            np.array([-1, str(2**63)], dtype=object),
+            np.array([str(-1), str(2**63)], dtype=object),
+        ],
+    )
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    def test_convert_numeric_int64_uint64(
+        self, case, coerce, convert_to_masked_nullable
+    ):
+        expected = case.astype(float) if coerce else case.copy()
+        result, _ = lib.maybe_convert_numeric(
+            case,
+            set(),
+            coerce_numeric=coerce,
+            convert_to_masked_nullable=convert_to_masked_nullable,
+        )
+
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize("convert_to_masked_nullable", [True, False])
+    def test_convert_numeric_string_uint64(self, convert_to_masked_nullable):
+        # GH32394
+        result = lib.maybe_convert_numeric(
+            np.array(["uint64"], dtype=object),
+            set(),
+            coerce_numeric=True,
+            convert_to_masked_nullable=convert_to_masked_nullable,
+        )
+        if convert_to_masked_nullable:
+            result = FloatingArray(*result)
+        else:
+            result = result[0]
+        assert np.isnan(result)
+
+    @pytest.mark.parametrize("value", [-(2**63) - 1, 2**64])
+    def test_convert_int_overflow(self, value):
+        # see gh-18584
+        arr = np.array([value], dtype=object)
+        result = lib.maybe_convert_objects(arr)
+        tm.assert_numpy_array_equal(arr, result)
+
+    @pytest.mark.parametrize("val", [None, np.nan, float("nan")])
+    @pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+    def test_maybe_convert_objects_nat_inference(self, val, dtype):
+        dtype = np.dtype(dtype)
+        vals = np.array([pd.NaT, val], dtype=object)
+        result = lib.maybe_convert_objects(
+            vals,
+            convert_non_numeric=True,
+            dtype_if_all_nat=dtype,
+        )
+        assert result.dtype == dtype
+        assert np.isnat(result).all()
+
+        result = lib.maybe_convert_objects(
+            vals[::-1],
+            convert_non_numeric=True,
+            dtype_if_all_nat=dtype,
+        )
+        assert result.dtype == dtype
+        assert np.isnat(result).all()
+
+    @pytest.mark.parametrize(
+        "value, expected_dtype",
+        [
+            # see gh-4471
+            ([2**63], np.uint64),
+            # NumPy bug: can't compare uint64 to int64, as that
+            # results in both casting to float64, so we should
+            # make sure that this function is robust against it
+            ([np.uint64(2**63)], np.uint64),
+            ([2, -1], np.int64),
+            ([2**63, -1], object),
+            # GH#47294
+            ([np.uint8(1)], np.uint8),
+            ([np.uint16(1)], np.uint16),
+            ([np.uint32(1)], np.uint32),
+            ([np.uint64(1)], np.uint64),
+            ([np.uint8(2), np.uint16(1)], np.uint16),
+            ([np.uint32(2), np.uint16(1)], np.uint32),
+            ([np.uint32(2), -1], object),
+            ([np.uint32(2), 1], np.uint64),
+            ([np.uint32(2), np.int32(1)], object),
+        ],
+    )
+    def test_maybe_convert_objects_uint(self, value, expected_dtype):
+        arr = np.array(value, dtype=object)
+        exp = np.array(value, dtype=expected_dtype)
+        tm.assert_numpy_array_equal(lib.maybe_convert_objects(arr), exp)
+
+    def test_maybe_convert_objects_datetime(self):
+        # GH27438
+        arr = np.array(
+            [np.datetime64("2000-01-01"), np.timedelta64(1, "s")], dtype=object
+        )
+        exp = arr.copy()
+        out = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        tm.assert_numpy_array_equal(out, exp)
+
+        arr = np.array([pd.NaT, np.timedelta64(1, "s")], dtype=object)
+        exp = np.array([np.timedelta64("NaT"), np.timedelta64(1, "s")], dtype="m8[ns]")
+        out = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        tm.assert_numpy_array_equal(out, exp)
+
+        # with convert_non_numeric=True, the nan is a valid NA value for td64
+        arr = np.array([np.timedelta64(1, "s"), np.nan], dtype=object)
+        exp = exp[::-1]
+        out = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        tm.assert_numpy_array_equal(out, exp)
+
+    def test_maybe_convert_objects_dtype_if_all_nat(self):
+        arr = np.array([pd.NaT, pd.NaT], dtype=object)
+        out = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        # no dtype_if_all_nat passed -> we dont guess
+        tm.assert_numpy_array_equal(out, arr)
+
+        out = lib.maybe_convert_objects(
+            arr,
+            convert_non_numeric=True,
+            dtype_if_all_nat=np.dtype("timedelta64[ns]"),
+        )
+        exp = np.array(["NaT", "NaT"], dtype="timedelta64[ns]")
+        tm.assert_numpy_array_equal(out, exp)
+
+        out = lib.maybe_convert_objects(
+            arr,
+            convert_non_numeric=True,
+            dtype_if_all_nat=np.dtype("datetime64[ns]"),
+        )
+        exp = np.array(["NaT", "NaT"], dtype="datetime64[ns]")
+        tm.assert_numpy_array_equal(out, exp)
+
+    def test_maybe_convert_objects_dtype_if_all_nat_invalid(self):
+        # we accept datetime64[ns], timedelta64[ns], and EADtype
+        arr = np.array([pd.NaT, pd.NaT], dtype=object)
+
+        with pytest.raises(ValueError, match="int64"):
+            lib.maybe_convert_objects(
+                arr,
+                convert_non_numeric=True,
+                dtype_if_all_nat=np.dtype("int64"),
+            )
+
+    @pytest.mark.parametrize("dtype", ["datetime64[ns]", "timedelta64[ns]"])
+    def test_maybe_convert_objects_datetime_overflow_safe(self, dtype):
+        stamp = datetime(2363, 10, 4)  # Enterprise-D launch date
+        if dtype == "timedelta64[ns]":
+            stamp = stamp - datetime(1970, 1, 1)
+        arr = np.array([stamp], dtype=object)
+
+        out = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        # no OutOfBoundsDatetime/OutOfBoundsTimedeltas
+        if dtype == "datetime64[ns]":
+            expected = np.array(["2363-10-04"], dtype="M8[us]")
+        else:
+            expected = arr
+        tm.assert_numpy_array_equal(out, expected)
+
+    def test_maybe_convert_objects_mixed_datetimes(self):
+        ts = Timestamp("now")
+        vals = [ts, ts.to_pydatetime(), ts.to_datetime64(), pd.NaT, np.nan, None]
+
+        for data in itertools.permutations(vals):
+            data = np.array(list(data), dtype=object)
+            expected = DatetimeIndex(data)._data._ndarray
+            result = lib.maybe_convert_objects(data, convert_non_numeric=True)
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_maybe_convert_objects_timedelta64_nat(self):
+        obj = np.timedelta64("NaT", "ns")
+        arr = np.array([obj], dtype=object)
+        assert arr[0] is obj
+
+        result = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+
+        expected = np.array([obj], dtype="m8[ns]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "exp",
+        [
+            IntegerArray(np.array([2, 0], dtype="i8"), np.array([False, True])),
+            IntegerArray(np.array([2, 0], dtype="int64"), np.array([False, True])),
+        ],
+    )
+    def test_maybe_convert_objects_nullable_integer(self, exp):
+        # GH27335
+        arr = np.array([2, np.nan], dtype=object)
+        result = lib.maybe_convert_objects(arr, convert_to_nullable_dtype=True)
+
+        tm.assert_extension_array_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "dtype, val", [("int64", 1), ("uint64", np.iinfo(np.int64).max + 1)]
+    )
+    def test_maybe_convert_objects_nullable_none(self, dtype, val):
+        # GH#50043
+        arr = np.array([val, None, 3], dtype="object")
+        result = lib.maybe_convert_objects(arr, convert_to_nullable_dtype=True)
+        expected = IntegerArray(
+            np.array([val, 0, 3], dtype=dtype), np.array([False, True, False])
+        )
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "convert_to_masked_nullable, exp",
+        [
+            (True, IntegerArray(np.array([2, 0], dtype="i8"), np.array([False, True]))),
+            (False, np.array([2, np.nan], dtype="float64")),
+        ],
+    )
+    def test_maybe_convert_numeric_nullable_integer(
+        self, convert_to_masked_nullable, exp
+    ):
+        # GH 40687
+        arr = np.array([2, np.nan], dtype=object)
+        result = lib.maybe_convert_numeric(
+            arr, set(), convert_to_masked_nullable=convert_to_masked_nullable
+        )
+        if convert_to_masked_nullable:
+            result = IntegerArray(*result)
+            tm.assert_extension_array_equal(result, exp)
+        else:
+            result = result[0]
+            tm.assert_numpy_array_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "convert_to_masked_nullable, exp",
+        [
+            (
+                True,
+                FloatingArray(
+                    np.array([2.0, 0.0], dtype="float64"), np.array([False, True])
+                ),
+            ),
+            (False, np.array([2.0, np.nan], dtype="float64")),
+        ],
+    )
+    def test_maybe_convert_numeric_floating_array(
+        self, convert_to_masked_nullable, exp
+    ):
+        # GH 40687
+        arr = np.array([2.0, np.nan], dtype=object)
+        result = lib.maybe_convert_numeric(
+            arr, set(), convert_to_masked_nullable=convert_to_masked_nullable
+        )
+        if convert_to_masked_nullable:
+            tm.assert_extension_array_equal(FloatingArray(*result), exp)
+        else:
+            result = result[0]
+            tm.assert_numpy_array_equal(result, exp)
+
+    def test_maybe_convert_objects_bool_nan(self):
+        # GH32146
+        ind = Index([True, False, np.nan], dtype=object)
+        exp = np.array([True, False, np.nan], dtype=object)
+        out = lib.maybe_convert_objects(ind.values, safe=1)
+        tm.assert_numpy_array_equal(out, exp)
+
+    def test_maybe_convert_objects_nullable_boolean(self):
+        # GH50047
+        arr = np.array([True, False], dtype=object)
+        exp = BooleanArray._from_sequence([True, False], dtype="boolean")
+        out = lib.maybe_convert_objects(arr, convert_to_nullable_dtype=True)
+        tm.assert_extension_array_equal(out, exp)
+
+        arr = np.array([True, False, pd.NaT], dtype=object)
+        exp = np.array([True, False, pd.NaT], dtype=object)
+        out = lib.maybe_convert_objects(arr, convert_to_nullable_dtype=True)
+        tm.assert_numpy_array_equal(out, exp)
+
+    @pytest.mark.parametrize("val", [None, np.nan])
+    def test_maybe_convert_objects_nullable_boolean_na(self, val):
+        # GH50047
+        arr = np.array([True, False, val], dtype=object)
+        exp = BooleanArray(
+            np.array([True, False, False]), np.array([False, False, True])
+        )
+        out = lib.maybe_convert_objects(arr, convert_to_nullable_dtype=True)
+        tm.assert_extension_array_equal(out, exp)
+
+    @pytest.mark.parametrize(
+        "data0",
+        [
+            True,
+            1,
+            1.0,
+            1.0 + 1.0j,
+            np.int8(1),
+            np.int16(1),
+            np.int32(1),
+            np.int64(1),
+            np.float16(1),
+            np.float32(1),
+            np.float64(1),
+            np.complex64(1),
+            np.complex128(1),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "data1",
+        [
+            True,
+            1,
+            1.0,
+            1.0 + 1.0j,
+            np.int8(1),
+            np.int16(1),
+            np.int32(1),
+            np.int64(1),
+            np.float16(1),
+            np.float32(1),
+            np.float64(1),
+            np.complex64(1),
+            np.complex128(1),
+        ],
+    )
+    def test_maybe_convert_objects_itemsize(self, data0, data1):
+        # GH 40908
+        data = [data0, data1]
+        arr = np.array(data, dtype="object")
+
+        common_kind = np.result_type(type(data0), type(data1)).kind
+        kind0 = "python" if not hasattr(data0, "dtype") else data0.dtype.kind
+        kind1 = "python" if not hasattr(data1, "dtype") else data1.dtype.kind
+        if kind0 != "python" and kind1 != "python":
+            kind = common_kind
+            itemsize = max(data0.dtype.itemsize, data1.dtype.itemsize)
+        elif is_bool(data0) or is_bool(data1):
+            kind = "bool" if (is_bool(data0) and is_bool(data1)) else "object"
+            itemsize = ""
+        elif is_complex(data0) or is_complex(data1):
+            kind = common_kind
+            itemsize = 16
+        else:
+            kind = common_kind
+            itemsize = 8
+
+        expected = np.array(data, dtype=f"{kind}{itemsize}")
+        result = lib.maybe_convert_objects(arr)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_mixed_dtypes_remain_object_array(self):
+        # GH14956
+        arr = np.array([datetime(2015, 1, 1, tzinfo=timezone.utc), 1], dtype=object)
+        result = lib.maybe_convert_objects(arr, convert_non_numeric=True)
+        tm.assert_numpy_array_equal(result, arr)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            pd.IntervalIndex.from_breaks(range(5), closed="both"),
+            pd.period_range("2016-01-01", periods=3, freq="D"),
+        ],
+    )
+    def test_maybe_convert_objects_ea(self, idx):
+        result = lib.maybe_convert_objects(
+            np.array(idx, dtype=object),
+            convert_non_numeric=True,
+        )
+        tm.assert_extension_array_equal(result, idx._data)
+
+
+class TestTypeInference:
+    # Dummy class used for testing with Python objects
+    class Dummy:
+        pass
+
+    def test_inferred_dtype_fixture(self, any_skipna_inferred_dtype):
+        # see pandas/conftest.py
+        inferred_dtype, values = any_skipna_inferred_dtype
+
+        # make sure the inferred dtype of the fixture is as requested
+        assert inferred_dtype == lib.infer_dtype(values, skipna=True)
+
+    def test_length_zero(self, skipna):
+        result = lib.infer_dtype(np.array([], dtype="i4"), skipna=skipna)
+        assert result == "integer"
+
+        result = lib.infer_dtype([], skipna=skipna)
+        assert result == "empty"
+
+        # GH 18004
+        arr = np.array([np.array([], dtype=object), np.array([], dtype=object)])
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "empty"
+
+    def test_integers(self):
+        arr = np.array([1, 2, 3, np.int64(4), np.int32(5)], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "integer"
+
+        arr = np.array([1, 2, 3, np.int64(4), np.int32(5), "foo"], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "mixed-integer"
+
+        arr = np.array([1, 2, 3, 4, 5], dtype="i4")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "integer"
+
+    @pytest.mark.parametrize(
+        "arr, skipna",
+        [
+            ([1, 2, np.nan, np.nan, 3], False),
+            ([1, 2, np.nan, np.nan, 3], True),
+            ([1, 2, 3, np.int64(4), np.int32(5), np.nan], False),
+            ([1, 2, 3, np.int64(4), np.int32(5), np.nan], True),
+        ],
+    )
+    def test_integer_na(self, arr, skipna):
+        # GH 27392
+        result = lib.infer_dtype(np.array(arr, dtype="O"), skipna=skipna)
+        expected = "integer" if skipna else "integer-na"
+        assert result == expected
+
+    def test_infer_dtype_skipna_default(self):
+        # infer_dtype `skipna` default deprecated in GH#24050,
+        #  changed to True in GH#29876
+        arr = np.array([1, 2, 3, np.nan], dtype=object)
+
+        result = lib.infer_dtype(arr)
+        assert result == "integer"
+
+    def test_bools(self):
+        arr = np.array([True, False, True, True, True], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "boolean"
+
+        arr = np.array([np.bool_(True), np.bool_(False)], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "boolean"
+
+        arr = np.array([True, False, True, "foo"], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "mixed"
+
+        arr = np.array([True, False, True], dtype=bool)
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "boolean"
+
+        arr = np.array([True, np.nan, False], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "boolean"
+
+        result = lib.infer_dtype(arr, skipna=False)
+        assert result == "mixed"
+
+    def test_floats(self):
+        arr = np.array([1.0, 2.0, 3.0, np.float64(4), np.float32(5)], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "floating"
+
+        arr = np.array([1, 2, 3, np.float64(4), np.float32(5), "foo"], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "mixed-integer"
+
+        arr = np.array([1, 2, 3, 4, 5], dtype="f4")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "floating"
+
+        arr = np.array([1, 2, 3, 4, 5], dtype="f8")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "floating"
+
+    def test_decimals(self):
+        # GH15690
+        arr = np.array([Decimal(1), Decimal(2), Decimal(3)])
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "decimal"
+
+        arr = np.array([1.0, 2.0, Decimal(3)])
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "mixed"
+
+        result = lib.infer_dtype(arr[::-1], skipna=True)
+        assert result == "mixed"
+
+        arr = np.array([Decimal(1), Decimal("NaN"), Decimal(3)])
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "decimal"
+
+        arr = np.array([Decimal(1), np.nan, Decimal(3)], dtype="O")
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "decimal"
+
+    # complex is compatible with nan, so skipna has no effect
+    def test_complex(self, skipna):
+        # gets cast to complex on array construction
+        arr = np.array([1.0, 2.0, 1 + 1j])
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "complex"
+
+        arr = np.array([1.0, 2.0, 1 + 1j], dtype="O")
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "mixed"
+
+        result = lib.infer_dtype(arr[::-1], skipna=skipna)
+        assert result == "mixed"
+
+        # gets cast to complex on array construction
+        arr = np.array([1, np.nan, 1 + 1j])
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "complex"
+
+        arr = np.array([1.0, np.nan, 1 + 1j], dtype="O")
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "mixed"
+
+        # complex with nans stays complex
+        arr = np.array([1 + 1j, np.nan, 3 + 3j], dtype="O")
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "complex"
+
+        # test smaller complex dtype; will pass through _try_infer_map fastpath
+        arr = np.array([1 + 1j, np.nan, 3 + 3j], dtype=np.complex64)
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == "complex"
+
+    def test_string(self):
+        pass
+
+    def test_unicode(self):
+        arr = ["a", np.nan, "c"]
+        result = lib.infer_dtype(arr, skipna=False)
+        # This currently returns "mixed", but it's not clear that's optimal.
+        # This could also return "string" or "mixed-string"
+        assert result == "mixed"
+
+        # even though we use skipna, we are only skipping those NAs that are
+        #  considered matching by is_string_array
+        arr = ["a", np.nan, "c"]
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "string"
+
+        arr = ["a", pd.NA, "c"]
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "string"
+
+        arr = ["a", pd.NaT, "c"]
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "mixed"
+
+        arr = ["a", "c"]
+        result = lib.infer_dtype(arr, skipna=False)
+        assert result == "string"
+
+    @pytest.mark.parametrize(
+        "dtype, missing, skipna, expected",
+        [
+            (float, np.nan, False, "floating"),
+            (float, np.nan, True, "floating"),
+            (object, np.nan, False, "floating"),
+            (object, np.nan, True, "empty"),
+            (object, None, False, "mixed"),
+            (object, None, True, "empty"),
+        ],
+    )
+    @pytest.mark.parametrize("box", [Series, np.array])
+    def test_object_empty(self, box, missing, dtype, skipna, expected):
+        # GH 23421
+        arr = box([missing, missing], dtype=dtype)
+
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == expected
+
+    def test_datetime(self):
+        dates = [datetime(2012, 1, x) for x in range(1, 20)]
+        index = Index(dates)
+        assert index.inferred_type == "datetime64"
+
+    def test_infer_dtype_datetime64(self):
+        arr = np.array(
+            [np.datetime64("2011-01-01"), np.datetime64("2011-01-01")], dtype=object
+        )
+        assert lib.infer_dtype(arr, skipna=True) == "datetime64"
+
+    @pytest.mark.parametrize("na_value", [pd.NaT, np.nan])
+    def test_infer_dtype_datetime64_with_na(self, na_value):
+        # starts with nan
+        arr = np.array([na_value, np.datetime64("2011-01-02")])
+        assert lib.infer_dtype(arr, skipna=True) == "datetime64"
+
+        arr = np.array([na_value, np.datetime64("2011-01-02"), na_value])
+        assert lib.infer_dtype(arr, skipna=True) == "datetime64"
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            np.array(
+                [np.timedelta64("nat"), np.datetime64("2011-01-02")], dtype=object
+            ),
+            np.array(
+                [np.datetime64("2011-01-02"), np.timedelta64("nat")], dtype=object
+            ),
+            np.array([np.datetime64("2011-01-01"), Timestamp("2011-01-02")]),
+            np.array([Timestamp("2011-01-02"), np.datetime64("2011-01-01")]),
+            np.array([np.nan, Timestamp("2011-01-02"), 1.1]),
+            np.array([np.nan, "2011-01-01", Timestamp("2011-01-02")], dtype=object),
+            np.array([np.datetime64("nat"), np.timedelta64(1, "D")], dtype=object),
+            np.array([np.timedelta64(1, "D"), np.datetime64("nat")], dtype=object),
+        ],
+    )
+    def test_infer_datetimelike_dtype_mixed(self, arr):
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+    def test_infer_dtype_mixed_integer(self):
+        arr = np.array([np.nan, Timestamp("2011-01-02"), 1])
+        assert lib.infer_dtype(arr, skipna=True) == "mixed-integer"
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            [Timestamp("2011-01-01"), Timestamp("2011-01-02")],
+            [datetime(2011, 1, 1), datetime(2012, 2, 1)],
+            [datetime(2011, 1, 1), Timestamp("2011-01-02")],
+        ],
+    )
+    def test_infer_dtype_datetime(self, arr):
+        assert lib.infer_dtype(np.array(arr), skipna=True) == "datetime"
+
+    @pytest.mark.parametrize("na_value", [pd.NaT, np.nan])
+    @pytest.mark.parametrize(
+        "time_stamp", [Timestamp("2011-01-01"), datetime(2011, 1, 1)]
+    )
+    def test_infer_dtype_datetime_with_na(self, na_value, time_stamp):
+        # starts with nan
+        arr = np.array([na_value, time_stamp])
+        assert lib.infer_dtype(arr, skipna=True) == "datetime"
+
+        arr = np.array([na_value, time_stamp, na_value])
+        assert lib.infer_dtype(arr, skipna=True) == "datetime"
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            np.array([Timedelta("1 days"), Timedelta("2 days")]),
+            np.array([np.timedelta64(1, "D"), np.timedelta64(2, "D")], dtype=object),
+            np.array([timedelta(1), timedelta(2)]),
+        ],
+    )
+    def test_infer_dtype_timedelta(self, arr):
+        assert lib.infer_dtype(arr, skipna=True) == "timedelta"
+
+    @pytest.mark.parametrize("na_value", [pd.NaT, np.nan])
+    @pytest.mark.parametrize(
+        "delta", [Timedelta("1 days"), np.timedelta64(1, "D"), timedelta(1)]
+    )
+    def test_infer_dtype_timedelta_with_na(self, na_value, delta):
+        # starts with nan
+        arr = np.array([na_value, delta])
+        assert lib.infer_dtype(arr, skipna=True) == "timedelta"
+
+        arr = np.array([na_value, delta, na_value])
+        assert lib.infer_dtype(arr, skipna=True) == "timedelta"
+
+    def test_infer_dtype_period(self):
+        # GH 13664
+        arr = np.array([Period("2011-01", freq="D"), Period("2011-02", freq="D")])
+        assert lib.infer_dtype(arr, skipna=True) == "period"
+
+        # non-homogeneous freqs -> mixed
+        arr = np.array([Period("2011-01", freq="D"), Period("2011-02", freq="M")])
+        assert lib.infer_dtype(arr, skipna=True) == "mixed"
+
+    def test_infer_dtype_period_array(self, index_or_series_or_array, skipna):
+        klass = index_or_series_or_array
+        # https://github.com/pandas-dev/pandas/issues/23553
+        values = klass(
+            [
+                Period("2011-01-01", freq="D"),
+                Period("2011-01-02", freq="D"),
+                pd.NaT,
+            ]
+        )
+        assert lib.infer_dtype(values, skipna=skipna) == "period"
+
+        # periods but mixed freq
+        values = klass(
+            [
+                Period("2011-01-01", freq="D"),
+                Period("2011-01-02", freq="M"),
+                pd.NaT,
+            ]
+        )
+        # with pd.array this becomes NumpyExtensionArray which ends up
+        #  as "unknown-array"
+        exp = "unknown-array" if klass is pd.array else "mixed"
+        assert lib.infer_dtype(values, skipna=skipna) == exp
+
+    def test_infer_dtype_period_mixed(self):
+        arr = np.array(
+            [Period("2011-01", freq="M"), np.datetime64("nat")], dtype=object
+        )
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        arr = np.array(
+            [np.datetime64("nat"), Period("2011-01", freq="M")], dtype=object
+        )
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+    @pytest.mark.parametrize("na_value", [pd.NaT, np.nan])
+    def test_infer_dtype_period_with_na(self, na_value):
+        # starts with nan
+        arr = np.array([na_value, Period("2011-01", freq="D")])
+        assert lib.infer_dtype(arr, skipna=True) == "period"
+
+        arr = np.array([na_value, Period("2011-01", freq="D"), na_value])
+        assert lib.infer_dtype(arr, skipna=True) == "period"
+
+    def test_infer_dtype_all_nan_nat_like(self):
+        arr = np.array([np.nan, np.nan])
+        assert lib.infer_dtype(arr, skipna=True) == "floating"
+
+        # nan and None mix are result in mixed
+        arr = np.array([np.nan, np.nan, None])
+        assert lib.infer_dtype(arr, skipna=True) == "empty"
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        arr = np.array([None, np.nan, np.nan])
+        assert lib.infer_dtype(arr, skipna=True) == "empty"
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        # pd.NaT
+        arr = np.array([pd.NaT])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime"
+
+        arr = np.array([pd.NaT, np.nan])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime"
+
+        arr = np.array([np.nan, pd.NaT])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime"
+
+        arr = np.array([np.nan, pd.NaT, np.nan])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime"
+
+        arr = np.array([None, pd.NaT, None])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime"
+
+        # np.datetime64(nat)
+        arr = np.array([np.datetime64("nat")])
+        assert lib.infer_dtype(arr, skipna=False) == "datetime64"
+
+        for n in [np.nan, pd.NaT, None]:
+            arr = np.array([n, np.datetime64("nat"), n])
+            assert lib.infer_dtype(arr, skipna=False) == "datetime64"
+
+            arr = np.array([pd.NaT, n, np.datetime64("nat"), n])
+            assert lib.infer_dtype(arr, skipna=False) == "datetime64"
+
+        arr = np.array([np.timedelta64("nat")], dtype=object)
+        assert lib.infer_dtype(arr, skipna=False) == "timedelta"
+
+        for n in [np.nan, pd.NaT, None]:
+            arr = np.array([n, np.timedelta64("nat"), n])
+            assert lib.infer_dtype(arr, skipna=False) == "timedelta"
+
+            arr = np.array([pd.NaT, n, np.timedelta64("nat"), n])
+            assert lib.infer_dtype(arr, skipna=False) == "timedelta"
+
+        # datetime / timedelta mixed
+        arr = np.array([pd.NaT, np.datetime64("nat"), np.timedelta64("nat"), np.nan])
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        arr = np.array([np.timedelta64("nat"), np.datetime64("nat")], dtype=object)
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+    def test_is_datetimelike_array_all_nan_nat_like(self):
+        arr = np.array([np.nan, pd.NaT, np.datetime64("nat")])
+        assert lib.is_datetime_array(arr)
+        assert lib.is_datetime64_array(arr)
+        assert not lib.is_timedelta_or_timedelta64_array(arr)
+
+        arr = np.array([np.nan, pd.NaT, np.timedelta64("nat")])
+        assert not lib.is_datetime_array(arr)
+        assert not lib.is_datetime64_array(arr)
+        assert lib.is_timedelta_or_timedelta64_array(arr)
+
+        arr = np.array([np.nan, pd.NaT, np.datetime64("nat"), np.timedelta64("nat")])
+        assert not lib.is_datetime_array(arr)
+        assert not lib.is_datetime64_array(arr)
+        assert not lib.is_timedelta_or_timedelta64_array(arr)
+
+        arr = np.array([np.nan, pd.NaT])
+        assert lib.is_datetime_array(arr)
+        assert lib.is_datetime64_array(arr)
+        assert lib.is_timedelta_or_timedelta64_array(arr)
+
+        arr = np.array([np.nan, np.nan], dtype=object)
+        assert not lib.is_datetime_array(arr)
+        assert not lib.is_datetime64_array(arr)
+        assert not lib.is_timedelta_or_timedelta64_array(arr)
+
+        assert lib.is_datetime_with_singletz_array(
+            np.array(
+                [
+                    Timestamp("20130101", tz="US/Eastern"),
+                    Timestamp("20130102", tz="US/Eastern"),
+                ],
+                dtype=object,
+            )
+        )
+        assert not lib.is_datetime_with_singletz_array(
+            np.array(
+                [
+                    Timestamp("20130101", tz="US/Eastern"),
+                    Timestamp("20130102", tz="CET"),
+                ],
+                dtype=object,
+            )
+        )
+
+    @pytest.mark.parametrize(
+        "func",
+        [
+            "is_datetime_array",
+            "is_datetime64_array",
+            "is_bool_array",
+            "is_timedelta_or_timedelta64_array",
+            "is_date_array",
+            "is_time_array",
+            "is_interval_array",
+        ],
+    )
+    def test_other_dtypes_for_array(self, func):
+        func = getattr(lib, func)
+        arr = np.array(["foo", "bar"])
+        assert not func(arr)
+        assert not func(arr.reshape(2, 1))
+
+        arr = np.array([1, 2])
+        assert not func(arr)
+        assert not func(arr.reshape(2, 1))
+
+    def test_date(self):
+        dates = [date(2012, 1, day) for day in range(1, 20)]
+        index = Index(dates)
+        assert index.inferred_type == "date"
+
+        dates = [date(2012, 1, day) for day in range(1, 20)] + [np.nan]
+        result = lib.infer_dtype(dates, skipna=False)
+        assert result == "mixed"
+
+        result = lib.infer_dtype(dates, skipna=True)
+        assert result == "date"
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            [date(2020, 1, 1), Timestamp("2020-01-01")],
+            [Timestamp("2020-01-01"), date(2020, 1, 1)],
+            [date(2020, 1, 1), pd.NaT],
+            [pd.NaT, date(2020, 1, 1)],
+        ],
+    )
+    def test_infer_dtype_date_order_invariant(self, values, skipna):
+        # https://github.com/pandas-dev/pandas/issues/33741
+        result = lib.infer_dtype(values, skipna=skipna)
+        assert result == "date"
+
+    def test_is_numeric_array(self):
+        assert lib.is_float_array(np.array([1, 2.0]))
+        assert lib.is_float_array(np.array([1, 2.0, np.nan]))
+        assert not lib.is_float_array(np.array([1, 2]))
+
+        assert lib.is_integer_array(np.array([1, 2]))
+        assert not lib.is_integer_array(np.array([1, 2.0]))
+
+    def test_is_string_array(self):
+        # We should only be accepting pd.NA, np.nan,
+        # other floating point nans e.g. float('nan')]
+        # when skipna is True.
+        assert lib.is_string_array(np.array(["foo", "bar"]))
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", pd.NA], dtype=object), skipna=False
+        )
+        assert lib.is_string_array(
+            np.array(["foo", "bar", pd.NA], dtype=object), skipna=True
+        )
+        # we allow NaN/None in the StringArray constructor, so its allowed here
+        assert lib.is_string_array(
+            np.array(["foo", "bar", None], dtype=object), skipna=True
+        )
+        assert lib.is_string_array(
+            np.array(["foo", "bar", np.nan], dtype=object), skipna=True
+        )
+        # But not e.g. datetimelike or Decimal NAs
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", pd.NaT], dtype=object), skipna=True
+        )
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", np.datetime64("NaT")], dtype=object), skipna=True
+        )
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", Decimal("NaN")], dtype=object), skipna=True
+        )
+
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", None], dtype=object), skipna=False
+        )
+        assert not lib.is_string_array(
+            np.array(["foo", "bar", np.nan], dtype=object), skipna=False
+        )
+        assert not lib.is_string_array(np.array([1, 2]))
+
+    @pytest.mark.parametrize(
+        "func",
+        [
+            "is_bool_array",
+            "is_date_array",
+            "is_datetime_array",
+            "is_datetime64_array",
+            "is_float_array",
+            "is_integer_array",
+            "is_interval_array",
+            "is_string_array",
+            "is_time_array",
+            "is_timedelta_or_timedelta64_array",
+        ],
+    )
+    def test_is_dtype_array_empty_obj(self, func):
+        # https://github.com/pandas-dev/pandas/pull/60796
+        func = getattr(lib, func)
+
+        arr = np.empty((2, 0), dtype=object)
+        assert not func(arr)
+
+        arr = np.empty((0, 2), dtype=object)
+        assert not func(arr)
+
+    def test_to_object_array_tuples(self):
+        r = (5, 6)
+        values = [r]
+        lib.to_object_array_tuples(values)
+
+        # make sure record array works
+        record = namedtuple("record", "x y")
+        r = record(5, 6)
+        values = [r]
+        lib.to_object_array_tuples(values)
+
+    def test_object(self):
+        # GH 7431
+        # cannot infer more than this as only a single element
+        arr = np.array([None], dtype="O")
+        result = lib.infer_dtype(arr, skipna=False)
+        assert result == "mixed"
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "empty"
+
+    def test_to_object_array_width(self):
+        # see gh-13320
+        rows = [[1, 2, 3], [4, 5, 6]]
+
+        expected = np.array(rows, dtype=object)
+        out = lib.to_object_array(rows)
+        tm.assert_numpy_array_equal(out, expected)
+
+        expected = np.array(rows, dtype=object)
+        out = lib.to_object_array(rows, min_width=1)
+        tm.assert_numpy_array_equal(out, expected)
+
+        expected = np.array(
+            [[1, 2, 3, None, None], [4, 5, 6, None, None]], dtype=object
+        )
+        out = lib.to_object_array(rows, min_width=5)
+        tm.assert_numpy_array_equal(out, expected)
+
+    def test_categorical(self):
+        # GH 8974
+        arr = Categorical(list("abc"))
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "categorical"
+
+        result = lib.infer_dtype(Series(arr), skipna=True)
+        assert result == "categorical"
+
+        arr = Categorical(list("abc"), categories=["cegfab"], ordered=True)
+        result = lib.infer_dtype(arr, skipna=True)
+        assert result == "categorical"
+
+        result = lib.infer_dtype(Series(arr), skipna=True)
+        assert result == "categorical"
+
+    @pytest.mark.parametrize("asobject", [True, False])
+    def test_interval(self, asobject):
+        idx = pd.IntervalIndex.from_breaks(range(5), closed="both")
+        if asobject:
+            idx = idx.astype(object)
+
+        inferred = lib.infer_dtype(idx, skipna=False)
+        assert inferred == "interval"
+
+        inferred = lib.infer_dtype(idx._data, skipna=False)
+        assert inferred == "interval"
+
+        inferred = lib.infer_dtype(Series(idx, dtype=idx.dtype), skipna=False)
+        assert inferred == "interval"
+
+    @pytest.mark.parametrize("value", [Timestamp(0), Timedelta(0), 0, 0.0])
+    def test_interval_mismatched_closed(self, value):
+        first = Interval(value, value, closed="left")
+        second = Interval(value, value, closed="right")
+
+        # if closed match, we should infer "interval"
+        arr = np.array([first, first], dtype=object)
+        assert lib.infer_dtype(arr, skipna=False) == "interval"
+
+        # if closed dont match, we should _not_ get "interval"
+        arr2 = np.array([first, second], dtype=object)
+        assert lib.infer_dtype(arr2, skipna=False) == "mixed"
+
+    def test_interval_mismatched_subtype(self):
+        first = Interval(0, 1, closed="left")
+        second = Interval(Timestamp(0), Timestamp(1), closed="left")
+        third = Interval(Timedelta(0), Timedelta(1), closed="left")
+
+        arr = np.array([first, second])
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        arr = np.array([second, third])
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        arr = np.array([first, third])
+        assert lib.infer_dtype(arr, skipna=False) == "mixed"
+
+        # float vs int subdtype are compatible
+        flt_interval = Interval(1.5, 2.5, closed="left")
+        arr = np.array([first, flt_interval], dtype=object)
+        assert lib.infer_dtype(arr, skipna=False) == "interval"
+
+    @pytest.mark.parametrize("data", [["a", "b", "c"], ["a", "b", pd.NA]])
+    def test_string_dtype(
+        self, data, skipna, index_or_series_or_array, nullable_string_dtype
+    ):
+        # StringArray
+        val = index_or_series_or_array(data, dtype=nullable_string_dtype)
+        inferred = lib.infer_dtype(val, skipna=skipna)
+        assert inferred == "string"
+
+    @pytest.mark.parametrize("data", [[True, False, True], [True, False, pd.NA]])
+    def test_boolean_dtype(self, data, skipna, index_or_series_or_array):
+        # BooleanArray
+        val = index_or_series_or_array(data, dtype="boolean")
+        inferred = lib.infer_dtype(val, skipna=skipna)
+        assert inferred == "boolean"
+
+
+class TestNumberScalar:
+    def test_is_number(self):
+        assert is_number(True)
+        assert is_number(1)
+        assert is_number(1.1)
+        assert is_number(1 + 3j)
+        assert is_number(np.int64(1))
+        assert is_number(np.float64(1.1))
+        assert is_number(np.complex128(1 + 3j))
+        assert is_number(np.nan)
+
+        assert not is_number(None)
+        assert not is_number("x")
+        assert not is_number(datetime(2011, 1, 1))
+        assert not is_number(np.datetime64("2011-01-01"))
+        assert not is_number(Timestamp("2011-01-01"))
+        assert not is_number(Timestamp("2011-01-01", tz="US/Eastern"))
+        assert not is_number(timedelta(1000))
+        assert not is_number(Timedelta("1 days"))
+
+        # questionable
+        assert not is_number(np.bool_(False))
+        assert is_number(np.timedelta64(1, "D"))
+
+    def test_is_bool(self):
+        assert is_bool(True)
+        assert is_bool(False)
+        assert is_bool(np.bool_(False))
+
+        assert not is_bool(1)
+        assert not is_bool(1.1)
+        assert not is_bool(1 + 3j)
+        assert not is_bool(np.int64(1))
+        assert not is_bool(np.float64(1.1))
+        assert not is_bool(np.complex128(1 + 3j))
+        assert not is_bool(np.nan)
+        assert not is_bool(None)
+        assert not is_bool("x")
+        assert not is_bool(datetime(2011, 1, 1))
+        assert not is_bool(np.datetime64("2011-01-01"))
+        assert not is_bool(Timestamp("2011-01-01"))
+        assert not is_bool(Timestamp("2011-01-01", tz="US/Eastern"))
+        assert not is_bool(timedelta(1000))
+        assert not is_bool(np.timedelta64(1, "D"))
+        assert not is_bool(Timedelta("1 days"))
+
+    def test_is_integer(self):
+        assert is_integer(1)
+        assert is_integer(np.int64(1))
+
+        assert not is_integer(True)
+        assert not is_integer(1.1)
+        assert not is_integer(1 + 3j)
+        assert not is_integer(False)
+        assert not is_integer(np.bool_(False))
+        assert not is_integer(np.float64(1.1))
+        assert not is_integer(np.complex128(1 + 3j))
+        assert not is_integer(np.nan)
+        assert not is_integer(None)
+        assert not is_integer("x")
+        assert not is_integer(datetime(2011, 1, 1))
+        assert not is_integer(np.datetime64("2011-01-01"))
+        assert not is_integer(Timestamp("2011-01-01"))
+        assert not is_integer(Timestamp("2011-01-01", tz="US/Eastern"))
+        assert not is_integer(timedelta(1000))
+        assert not is_integer(Timedelta("1 days"))
+        assert not is_integer(np.timedelta64(1, "D"))
+
+    def test_is_float(self):
+        assert is_float(1.1)
+        assert is_float(np.float64(1.1))
+        assert is_float(np.nan)
+
+        assert not is_float(True)
+        assert not is_float(1)
+        assert not is_float(1 + 3j)
+        assert not is_float(False)
+        assert not is_float(np.bool_(False))
+        assert not is_float(np.int64(1))
+        assert not is_float(np.complex128(1 + 3j))
+        assert not is_float(None)
+        assert not is_float("x")
+        assert not is_float(datetime(2011, 1, 1))
+        assert not is_float(np.datetime64("2011-01-01"))
+        assert not is_float(Timestamp("2011-01-01"))
+        assert not is_float(Timestamp("2011-01-01", tz="US/Eastern"))
+        assert not is_float(timedelta(1000))
+        assert not is_float(np.timedelta64(1, "D"))
+        assert not is_float(Timedelta("1 days"))
+
+    def test_is_datetime_dtypes(self):
+        ts = pd.date_range("20130101", periods=3)
+        tsa = pd.date_range("20130101", periods=3, tz="US/Eastern")
+
+        msg = "is_datetime64tz_dtype is deprecated"
+
+        assert is_datetime64_dtype("datetime64")
+        assert is_datetime64_dtype("datetime64[ns]")
+        assert is_datetime64_dtype(ts)
+        assert not is_datetime64_dtype(tsa)
+
+        assert not is_datetime64_ns_dtype("datetime64")
+        assert is_datetime64_ns_dtype("datetime64[ns]")
+        assert is_datetime64_ns_dtype(ts)
+        assert is_datetime64_ns_dtype(tsa)
+
+        assert is_datetime64_any_dtype("datetime64")
+        assert is_datetime64_any_dtype("datetime64[ns]")
+        assert is_datetime64_any_dtype(ts)
+        assert is_datetime64_any_dtype(tsa)
+
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert not is_datetime64tz_dtype("datetime64")
+            assert not is_datetime64tz_dtype("datetime64[ns]")
+            assert not is_datetime64tz_dtype(ts)
+            assert is_datetime64tz_dtype(tsa)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "UTC"])
+    def test_is_datetime_dtypes_with_tz(self, tz):
+        dtype = f"datetime64[ns, {tz}]"
+        assert not is_datetime64_dtype(dtype)
+
+        msg = "is_datetime64tz_dtype is deprecated"
+        with tm.assert_produces_warning(DeprecationWarning, match=msg):
+            assert is_datetime64tz_dtype(dtype)
+        assert is_datetime64_ns_dtype(dtype)
+        assert is_datetime64_any_dtype(dtype)
+
+    def test_is_timedelta(self):
+        assert is_timedelta64_dtype("timedelta64")
+        assert is_timedelta64_dtype("timedelta64[ns]")
+        assert not is_timedelta64_ns_dtype("timedelta64")
+        assert is_timedelta64_ns_dtype("timedelta64[ns]")
+
+        tdi = TimedeltaIndex([1e14, 2e14], dtype="timedelta64[ns]")
+        assert is_timedelta64_dtype(tdi)
+        assert is_timedelta64_ns_dtype(tdi)
+        assert is_timedelta64_ns_dtype(tdi.astype("timedelta64[ns]"))
+
+        assert not is_timedelta64_ns_dtype(Index([], dtype=np.float64))
+        assert not is_timedelta64_ns_dtype(Index([], dtype=np.int64))
+
+
+class TestIsScalar:
+    def test_is_scalar_builtin_scalars(self):
+        assert is_scalar(None)
+        assert is_scalar(True)
+        assert is_scalar(False)
+        assert is_scalar(Fraction())
+        assert is_scalar(0.0)
+        assert is_scalar(1)
+        assert is_scalar(complex(2))
+        assert is_scalar(float("NaN"))
+        assert is_scalar(np.nan)
+        assert is_scalar("foobar")
+        assert is_scalar(b"foobar")
+        assert is_scalar(datetime(2014, 1, 1))
+        assert is_scalar(date(2014, 1, 1))
+        assert is_scalar(time(12, 0))
+        assert is_scalar(timedelta(hours=1))
+        assert is_scalar(pd.NaT)
+        assert is_scalar(pd.NA)
+
+    def test_is_scalar_builtin_nonscalars(self):
+        assert not is_scalar({})
+        assert not is_scalar([])
+        assert not is_scalar([1])
+        assert not is_scalar(())
+        assert not is_scalar((1,))
+        assert not is_scalar(slice(None))
+        assert not is_scalar(Ellipsis)
+
+    def test_is_scalar_numpy_array_scalars(self):
+        assert is_scalar(np.int64(1))
+        assert is_scalar(np.float64(1.0))
+        assert is_scalar(np.int32(1))
+        assert is_scalar(np.complex64(2))
+        assert is_scalar(np.object_("foobar"))
+        assert is_scalar(np.str_("foobar"))
+        assert is_scalar(np.bytes_(b"foobar"))
+        assert is_scalar(np.datetime64("2014-01-01"))
+        assert is_scalar(np.timedelta64(1, "h"))
+
+    @pytest.mark.parametrize(
+        "zerodim",
+        [
+            1,
+            "foobar",
+            np.datetime64("2014-01-01"),
+            np.timedelta64(1, "h"),
+            np.datetime64("NaT"),
+        ],
+    )
+    def test_is_scalar_numpy_zerodim_arrays(self, zerodim):
+        zerodim = np.array(zerodim)
+        assert not is_scalar(zerodim)
+        assert is_scalar(lib.item_from_zerodim(zerodim))
+
+    @pytest.mark.parametrize("arr", [np.array([]), np.array([[]])])
+    def test_is_scalar_numpy_arrays(self, arr):
+        assert not is_scalar(arr)
+        assert not is_scalar(MockNumpyLikeArray(arr))
+
+    def test_is_scalar_pandas_scalars(self):
+        assert is_scalar(Timestamp("2014-01-01"))
+        assert is_scalar(Timedelta(hours=1))
+        assert is_scalar(Period("2014-01-01"))
+        assert is_scalar(Interval(left=0, right=1))
+        assert is_scalar(DateOffset(days=1))
+        assert is_scalar(pd.offsets.Minute(3))
+
+    def test_is_scalar_pandas_containers(self):
+        assert not is_scalar(Series(dtype=object))
+        assert not is_scalar(Series([1]))
+        assert not is_scalar(DataFrame())
+        assert not is_scalar(DataFrame([[1]]))
+        assert not is_scalar(Index([]))
+        assert not is_scalar(Index([1]))
+        assert not is_scalar(Categorical([]))
+        assert not is_scalar(DatetimeIndex([])._data)
+        assert not is_scalar(TimedeltaIndex([])._data)
+        assert not is_scalar(DatetimeIndex([])._data.to_period("D"))
+        assert not is_scalar(pd.array([1, 2, 3]))
+
+    def test_is_scalar_number(self):
+        # Number() is not recognized by PyNumber_Check, so by extension
+        #  is not recognized by is_scalar, but instances of non-abstract
+        #  subclasses are.
+
+        class Numeric(Number):
+            def __init__(self, value) -> None:
+                self.value = value
+
+            def __int__(self) -> int:
+                return self.value
+
+        num = Numeric(1)
+        assert is_scalar(num)
+
+
+@pytest.mark.parametrize("unit", ["ms", "us", "ns"])
+def test_datetimeindex_from_empty_datetime64_array(unit):
+    idx = DatetimeIndex(np.array([], dtype=f"datetime64[{unit}]"))
+    assert len(idx) == 0
+
+
+def test_nan_to_nat_conversions():
+    df = DataFrame(
+        {"A": np.asarray(range(10), dtype="float64"), "B": Timestamp("20010101")}
+    )
+    df.iloc[3:6, :] = np.nan
+    result = df.loc[4, "B"]
+    assert result is pd.NaT
+
+    s = df["B"].copy()
+    s[8:9] = np.nan
+    assert s[8] is pd.NaT
+
+
+@pytest.mark.filterwarnings("ignore::PendingDeprecationWarning")
+@pytest.mark.parametrize("spmatrix", ["bsr", "coo", "csc", "csr", "dia", "dok", "lil"])
+def test_is_scipy_sparse(spmatrix):
+    sparse = pytest.importorskip("scipy.sparse")
+
+    klass = getattr(sparse, spmatrix + "_matrix")
+    assert is_scipy_sparse(klass([[0, 1]]))
+    assert not is_scipy_sparse(np.array([1]))
+
+
+def test_ensure_int32():
+    values = np.arange(10, dtype=np.int32)
+    result = ensure_int32(values)
+    assert result.dtype == np.int32
+
+    values = np.arange(10, dtype=np.int64)
+    result = ensure_int32(values)
+    assert result.dtype == np.int32
+
+
+@pytest.mark.parametrize(
+    "right,result",
+    [
+        (0, np.uint8),
+        (-1, np.int16),
+        (300, np.uint16),
+        # For floats, we just upcast directly to float64 instead of trying to
+        # find a smaller floating dtype
+        (300.0, np.uint16),  # for integer floats, we convert them to ints
+        (300.1, np.float64),
+        (np.int16(300), np.int16 if np_version_gt2 else np.uint16),
+    ],
+)
+def test_find_result_type_uint_int(right, result):
+    left_dtype = np.dtype("uint8")
+    assert find_result_type(left_dtype, right) == result
+
+
+@pytest.mark.parametrize(
+    "right,result",
+    [
+        (0, np.int8),
+        (-1, np.int8),
+        (300, np.int16),
+        # For floats, we just upcast directly to float64 instead of trying to
+        # find a smaller floating dtype
+        (300.0, np.int16),  # for integer floats, we convert them to ints
+        (300.1, np.float64),
+        (np.int16(300), np.int16),
+    ],
+)
+def test_find_result_type_int_int(right, result):
+    left_dtype = np.dtype("int8")
+    assert find_result_type(left_dtype, right) == result
+
+
+@pytest.mark.parametrize(
+    "right,result",
+    [
+        (300.0, np.float64),
+        (np.float32(300), np.float32),
+    ],
+)
+def test_find_result_type_floats(right, result):
+    left_dtype = np.dtype("float16")
+    assert find_result_type(left_dtype, right) == result
diff --git a/pandas/pandas/tests/dtypes/test_missing.py b/pandas/pandas/tests/dtypes/test_missing.py
new file mode 100644
index 0000000000000..c61cda83cf6e0
--- /dev/null
+++ b/pandas/pandas/tests/dtypes/test_missing.py
@@ -0,0 +1,876 @@
+from datetime import datetime
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas._libs import missing as libmissing
+from pandas._libs.tslibs import iNaT
+
+from pandas.core.dtypes.common import (
+    is_float,
+    is_scalar,
+    pandas_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.missing import (
+    array_equivalent,
+    is_valid_na_for_dtype,
+    isna,
+    isnull,
+    na_value_for_dtype,
+    notna,
+    notnull,
+)
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    NaT,
+    Series,
+    TimedeltaIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+fix_now = pd.Timestamp("2021-01-01")
+fix_utcnow = pd.Timestamp("2021-01-01", tz="UTC")
+
+
+@pytest.mark.parametrize("notna_f", [notna, notnull])
+def test_notna_notnull(notna_f):
+    assert notna_f(1.0)
+    assert not notna_f(None)
+    assert not notna_f(np.nan)
+
+
+@pytest.mark.parametrize("null_func", [notna, notnull, isna, isnull])
+@pytest.mark.parametrize(
+    "ser",
+    [
+        Series(
+            [str(i) for i in range(5)],
+            index=Index([str(i) for i in range(5)], dtype=object),
+            dtype=object,
+        ),
+        Series(range(5), date_range("2020-01-01", periods=5)),
+        Series(range(5), period_range("2020-01-01", periods=5)),
+    ],
+)
+def test_null_check_is_series(null_func, ser):
+    assert isinstance(null_func(ser), Series)
+
+
+class TestIsNA:
+    def test_0d_array(self):
+        assert isna(np.array(np.nan))
+        assert not isna(np.array(0.0))
+        assert not isna(np.array(0))
+        # test object dtype
+        assert isna(np.array(np.nan, dtype=object))
+        assert not isna(np.array(0.0, dtype=object))
+        assert not isna(np.array(0, dtype=object))
+
+    @pytest.mark.parametrize("shape", [(4, 0), (4,)])
+    def test_empty_object(self, shape):
+        arr = np.empty(shape=shape, dtype=object)
+        result = isna(arr)
+        expected = np.ones(shape=shape, dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("isna_f", [isna, isnull])
+    def test_isna_isnull(self, isna_f):
+        assert not isna_f(1.0)
+        assert isna_f(None)
+        assert isna_f(np.nan)
+        assert float("nan")
+        assert not isna_f(np.inf)
+        assert not isna_f(-np.inf)
+
+        # type
+        assert not isna_f(type(Series(dtype=object)))
+        assert not isna_f(type(Series(dtype=np.float64)))
+        assert not isna_f(type(pd.DataFrame()))
+
+    @pytest.mark.parametrize("isna_f", [isna, isnull])
+    @pytest.mark.parametrize(
+        "data",
+        [
+            np.arange(4, dtype=float),
+            [0.0, 1.0, 0.0, 1.0],
+            Series(list("abcd"), dtype=object),
+            date_range("2020-01-01", periods=4),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("2020-01-01", periods=4),
+            range(4),
+            period_range("2020-01-01", periods=4),
+        ],
+    )
+    def test_isna_isnull_frame(self, isna_f, data, index):
+        # frame
+        df = pd.DataFrame(data, index=index)
+        result = isna_f(df)
+        expected = df.apply(isna_f)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isna_lists(self):
+        result = isna([[False]])
+        exp = np.array([[False]])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = isna([[1], [2]])
+        exp = np.array([[False], [False]])
+        tm.assert_numpy_array_equal(result, exp)
+
+        # list of strings / unicode
+        result = isna(["foo", "bar"])
+        exp = np.array([False, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = isna(["foo", "bar"])
+        exp = np.array([False, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+        # GH20675
+        result = isna([np.nan, "world"])
+        exp = np.array([True, False])
+        tm.assert_numpy_array_equal(result, exp)
+
+    def test_isna_nat(self):
+        result = isna([NaT])
+        exp = np.array([True])
+        tm.assert_numpy_array_equal(result, exp)
+
+        result = isna(np.array([NaT], dtype=object))
+        exp = np.array([True])
+        tm.assert_numpy_array_equal(result, exp)
+
+    def test_isna_numpy_nat(self):
+        arr = np.array(
+            [
+                NaT,
+                np.datetime64("NaT"),
+                np.timedelta64("NaT"),
+                np.datetime64("NaT", "s"),
+            ]
+        )
+        result = isna(arr)
+        expected = np.array([True] * 4)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_isna_datetime(self):
+        assert not isna(datetime.now())
+        assert notna(datetime.now())
+
+        idx = date_range("1/1/1990", periods=20)
+        exp = np.ones(len(idx), dtype=bool)
+        tm.assert_numpy_array_equal(notna(idx), exp)
+
+        idx = np.asarray(idx)
+        idx[0] = iNaT
+        idx = DatetimeIndex(idx)
+        mask = isna(idx)
+        assert mask[0]
+        exp = np.array([True] + [False] * (len(idx) - 1), dtype=bool)
+        tm.assert_numpy_array_equal(mask, exp)
+
+        # GH 9129
+        pidx = idx.to_period(freq="M")
+        mask = isna(pidx)
+        assert mask[0]
+        exp = np.array([True] + [False] * (len(idx) - 1), dtype=bool)
+        tm.assert_numpy_array_equal(mask, exp)
+
+        mask = isna(pidx[1:])
+        exp = np.zeros(len(mask), dtype=bool)
+        tm.assert_numpy_array_equal(mask, exp)
+
+    def test_isna_old_datetimelike(self):
+        # isna_old should work for dt64tz, td64, and period, not just tznaive
+        dti = date_range("2016-01-01", periods=3)
+        dta = dti._data
+        dta[-1] = NaT
+        expected = np.array([False, False, True], dtype=bool)
+
+        objs = [dta, dta.tz_localize("US/Eastern"), dta - dta, dta.to_period("D")]
+
+        for obj in objs:
+            result = isna(obj)
+
+            tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "value, expected",
+        [
+            (np.complex128(np.nan), True),
+            (np.float64(1), False),
+            (np.array([1, 1 + 0j, np.nan, 3]), np.array([False, False, True, False])),
+            (
+                np.array([1, 1 + 0j, np.nan, 3], dtype=object),
+                np.array([False, False, True, False]),
+            ),
+            (
+                np.array([1, 1 + 0j, np.nan, 3]).astype(object),
+                np.array([False, False, True, False]),
+            ),
+        ],
+    )
+    def test_complex(self, value, expected):
+        result = isna(value)
+        if is_scalar(result):
+            assert result is expected
+        else:
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_datetime_other_units(self):
+        idx = DatetimeIndex(["2011-01-01", "NaT", "2011-01-02"])
+        exp = np.array([False, True, False])
+        tm.assert_numpy_array_equal(isna(idx), exp)
+        tm.assert_numpy_array_equal(notna(idx), ~exp)
+        tm.assert_numpy_array_equal(isna(idx.values), exp)
+        tm.assert_numpy_array_equal(notna(idx.values), ~exp)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "datetime64[D]",
+            "datetime64[h]",
+            "datetime64[m]",
+            "datetime64[s]",
+            "datetime64[ms]",
+            "datetime64[us]",
+            "datetime64[ns]",
+        ],
+    )
+    def test_datetime_other_units_astype(self, dtype):
+        idx = DatetimeIndex(["2011-01-01", "NaT", "2011-01-02"])
+        values = idx.values.astype(dtype)
+
+        exp = np.array([False, True, False])
+        tm.assert_numpy_array_equal(isna(values), exp)
+        tm.assert_numpy_array_equal(notna(values), ~exp)
+
+        exp = Series([False, True, False])
+        s = Series(values)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+        s = Series(values, dtype=object)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+
+    def test_timedelta_other_units(self):
+        idx = TimedeltaIndex(["1 days", "NaT", "2 days"])
+        exp = np.array([False, True, False])
+        tm.assert_numpy_array_equal(isna(idx), exp)
+        tm.assert_numpy_array_equal(notna(idx), ~exp)
+        tm.assert_numpy_array_equal(isna(idx.values), exp)
+        tm.assert_numpy_array_equal(notna(idx.values), ~exp)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "timedelta64[D]",
+            "timedelta64[h]",
+            "timedelta64[m]",
+            "timedelta64[s]",
+            "timedelta64[ms]",
+            "timedelta64[us]",
+            "timedelta64[ns]",
+        ],
+    )
+    def test_timedelta_other_units_dtype(self, dtype):
+        idx = TimedeltaIndex(["1 days", "NaT", "2 days"])
+        values = idx.values.astype(dtype)
+
+        exp = np.array([False, True, False])
+        tm.assert_numpy_array_equal(isna(values), exp)
+        tm.assert_numpy_array_equal(notna(values), ~exp)
+
+        exp = Series([False, True, False])
+        s = Series(values)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+        s = Series(values, dtype=object)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+
+    def test_period(self):
+        idx = pd.PeriodIndex(["2011-01", "NaT", "2012-01"], freq="M")
+        exp = np.array([False, True, False])
+        tm.assert_numpy_array_equal(isna(idx), exp)
+        tm.assert_numpy_array_equal(notna(idx), ~exp)
+
+        exp = Series([False, True, False])
+        s = Series(idx)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+        s = Series(idx, dtype=object)
+        tm.assert_series_equal(isna(s), exp)
+        tm.assert_series_equal(notna(s), ~exp)
+
+    def test_decimal(self):
+        # scalars GH#23530
+        a = Decimal("1.0")
+        assert isna(a) is False
+        assert notna(a) is True
+
+        b = Decimal("NaN")
+        assert isna(b) is True
+        assert notna(b) is False
+
+        # array
+        arr = np.array([a, b])
+        expected = np.array([False, True])
+        result = isna(arr)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = notna(arr)
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        # series
+        ser = Series(arr)
+        expected = Series(expected)
+        result = isna(ser)
+        tm.assert_series_equal(result, expected)
+
+        result = notna(ser)
+        tm.assert_series_equal(result, ~expected)
+
+        # index
+        idx = Index(arr)
+        expected = np.array([False, True])
+        result = isna(idx)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = notna(idx)
+        tm.assert_numpy_array_equal(result, ~expected)
+
+
+@pytest.mark.parametrize("dtype_equal", [True, False])
+def test_array_equivalent(dtype_equal):
+    assert array_equivalent(
+        np.array([np.nan, np.nan]), np.array([np.nan, np.nan]), dtype_equal=dtype_equal
+    )
+    assert array_equivalent(
+        np.array([np.nan, 1, np.nan]),
+        np.array([np.nan, 1, np.nan]),
+        dtype_equal=dtype_equal,
+    )
+    assert array_equivalent(
+        np.array([np.nan, None], dtype="object"),
+        np.array([np.nan, None], dtype="object"),
+        dtype_equal=dtype_equal,
+    )
+    # Check the handling of nested arrays in array_equivalent_object
+    assert array_equivalent(
+        np.array([np.array([np.nan, None], dtype="object"), None], dtype="object"),
+        np.array([np.array([np.nan, None], dtype="object"), None], dtype="object"),
+        dtype_equal=dtype_equal,
+    )
+    assert array_equivalent(
+        np.array([np.nan, 1 + 1j], dtype="complex"),
+        np.array([np.nan, 1 + 1j], dtype="complex"),
+        dtype_equal=dtype_equal,
+    )
+    assert not array_equivalent(
+        np.array([np.nan, 1 + 1j], dtype="complex"),
+        np.array([np.nan, 1 + 2j], dtype="complex"),
+        dtype_equal=dtype_equal,
+    )
+    assert not array_equivalent(
+        np.array([np.nan, 1, np.nan]),
+        np.array([np.nan, 2, np.nan]),
+        dtype_equal=dtype_equal,
+    )
+    assert not array_equivalent(
+        np.array(["a", "b", "c", "d"]), np.array(["e", "e"]), dtype_equal=dtype_equal
+    )
+    assert array_equivalent(
+        Index([0, np.nan]), Index([0, np.nan]), dtype_equal=dtype_equal
+    )
+    assert not array_equivalent(
+        Index([0, np.nan]), Index([1, np.nan]), dtype_equal=dtype_equal
+    )
+
+
+@pytest.mark.parametrize("dtype_equal", [True, False])
+def test_array_equivalent_tdi(dtype_equal):
+    assert array_equivalent(
+        TimedeltaIndex([0, np.nan]),
+        TimedeltaIndex([0, np.nan]),
+        dtype_equal=dtype_equal,
+    )
+    assert not array_equivalent(
+        TimedeltaIndex([0, np.nan]),
+        TimedeltaIndex([1, np.nan]),
+        dtype_equal=dtype_equal,
+    )
+
+
+@pytest.mark.parametrize("dtype_equal", [True, False])
+def test_array_equivalent_dti(dtype_equal):
+    assert array_equivalent(
+        DatetimeIndex([0, np.nan]), DatetimeIndex([0, np.nan]), dtype_equal=dtype_equal
+    )
+    assert not array_equivalent(
+        DatetimeIndex([0, np.nan]), DatetimeIndex([1, np.nan]), dtype_equal=dtype_equal
+    )
+
+    dti1 = DatetimeIndex([0, np.nan], tz="US/Eastern")
+    dti2 = DatetimeIndex([0, np.nan], tz="CET")
+    dti3 = DatetimeIndex([1, np.nan], tz="US/Eastern")
+
+    assert array_equivalent(
+        dti1,
+        dti1,
+        dtype_equal=dtype_equal,
+    )
+    assert not array_equivalent(
+        dti1,
+        dti3,
+        dtype_equal=dtype_equal,
+    )
+    # The rest are not dtype_equal
+    assert not array_equivalent(DatetimeIndex([0, np.nan]), dti1)
+    assert array_equivalent(
+        dti2,
+        dti1,
+    )
+
+    assert not array_equivalent(DatetimeIndex([0, np.nan]), TimedeltaIndex([0, np.nan]))
+
+
+@pytest.mark.parametrize(
+    "val", [1, 1.1, 1 + 1j, True, "abc", [1, 2], (1, 2), {1, 2}, {"a": 1}, None]
+)
+def test_array_equivalent_series(val):
+    arr = np.array([1, 2])
+    assert not array_equivalent(Series([arr, arr]), Series([arr, val]))
+
+
+def test_array_equivalent_array_mismatched_shape():
+    # to trigger the motivating bug, the first N elements of the arrays need
+    #  to match
+    first = np.array([1, 2, 3])
+    second = np.array([1, 2])
+
+    left = Series([first, "a"], dtype=object)
+    right = Series([second, "a"], dtype=object)
+    assert not array_equivalent(left, right)
+
+
+def test_array_equivalent_array_mismatched_dtype():
+    # same shape, different dtype can still be equivalent
+    first = np.array([1, 2], dtype=np.float64)
+    second = np.array([1, 2])
+
+    left = Series([first, "a"], dtype=object)
+    right = Series([second, "a"], dtype=object)
+    assert array_equivalent(left, right)
+
+
+def test_array_equivalent_different_dtype_but_equal():
+    # Unclear if this is exposed anywhere in the public-facing API
+    assert array_equivalent(np.array([1, 2]), np.array([1.0, 2.0]))
+
+
+@pytest.mark.parametrize(
+    "lvalue, rvalue",
+    [
+        # There are 3 variants for each of lvalue and rvalue. We include all
+        #  three for the tz-naive `now` and exclude the datetim64 variant
+        #  for utcnow because it drops tzinfo.
+        (fix_now, fix_utcnow),
+        (fix_now.to_datetime64(), fix_utcnow),
+        (fix_now.to_pydatetime(), fix_utcnow),
+        (fix_now.to_datetime64(), fix_utcnow.to_pydatetime()),
+        (fix_now.to_pydatetime(), fix_utcnow.to_pydatetime()),
+    ],
+)
+def test_array_equivalent_tzawareness(lvalue, rvalue):
+    # we shouldn't raise if comparing tzaware and tznaive datetimes
+    left = np.array([lvalue], dtype=object)
+    right = np.array([rvalue], dtype=object)
+
+    assert not array_equivalent(left, right, strict_nan=True)
+    assert not array_equivalent(left, right, strict_nan=False)
+
+
+def test_array_equivalent_compat():
+    # see gh-13388
+    m = np.array([(1, 2), (3, 4)], dtype=[("a", int), ("b", float)])
+    n = np.array([(1, 2), (3, 4)], dtype=[("a", int), ("b", float)])
+    assert array_equivalent(m, n, strict_nan=True)
+    assert array_equivalent(m, n, strict_nan=False)
+
+    m = np.array([(1, 2), (3, 4)], dtype=[("a", int), ("b", float)])
+    n = np.array([(1, 2), (4, 3)], dtype=[("a", int), ("b", float)])
+    assert not array_equivalent(m, n, strict_nan=True)
+    assert not array_equivalent(m, n, strict_nan=False)
+
+    m = np.array([(1, 2), (3, 4)], dtype=[("a", int), ("b", float)])
+    n = np.array([(1, 2), (3, 4)], dtype=[("b", int), ("a", float)])
+    assert not array_equivalent(m, n, strict_nan=True)
+    assert not array_equivalent(m, n, strict_nan=False)
+
+
+@pytest.mark.parametrize("dtype", ["O", "S", "U"])
+def test_array_equivalent_str(dtype):
+    assert array_equivalent(
+        np.array(["A", "B"], dtype=dtype), np.array(["A", "B"], dtype=dtype)
+    )
+    assert not array_equivalent(
+        np.array(["A", "B"], dtype=dtype), np.array(["A", "X"], dtype=dtype)
+    )
+
+
+@pytest.mark.parametrize("strict_nan", [True, False])
+def test_array_equivalent_nested(strict_nan):
+    # reached in groupby aggregations, make sure we use np.any when checking
+    #  if the comparison is truthy
+    left = np.array([np.array([50, 70, 90]), np.array([20, 30])], dtype=object)
+    right = np.array([np.array([50, 70, 90]), np.array([20, 30])], dtype=object)
+
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    left = np.empty(2, dtype=object)
+    left[:] = [np.array([50, 70, 90]), np.array([20, 30, 40])]
+    right = np.empty(2, dtype=object)
+    right[:] = [np.array([50, 70, 90]), np.array([20, 30, 40])]
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    left = np.array([np.array([50, 50, 50]), np.array([40, 40])], dtype=object)
+    right = np.array([50, 40])
+    assert not array_equivalent(left, right, strict_nan=strict_nan)
+
+
+@pytest.mark.filterwarnings("ignore:elementwise comparison failed:DeprecationWarning")
+@pytest.mark.parametrize("strict_nan", [True, False])
+def test_array_equivalent_nested2(strict_nan):
+    # more than one level of nesting
+    left = np.array(
+        [
+            np.array([np.array([50, 70]), np.array([90])], dtype=object),
+            np.array([np.array([20, 30])], dtype=object),
+        ],
+        dtype=object,
+    )
+    right = np.array(
+        [
+            np.array([np.array([50, 70]), np.array([90])], dtype=object),
+            np.array([np.array([20, 30])], dtype=object),
+        ],
+        dtype=object,
+    )
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    left = np.array([np.array([np.array([50, 50, 50])], dtype=object)], dtype=object)
+    right = np.array([50])
+    assert not array_equivalent(left, right, strict_nan=strict_nan)
+
+
+@pytest.mark.parametrize("strict_nan", [True, False])
+def test_array_equivalent_nested_list(strict_nan):
+    left = np.array([[50, 70, 90], [20, 30]], dtype=object)
+    right = np.array([[50, 70, 90], [20, 30]], dtype=object)
+
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    left = np.array([[50, 50, 50], [40, 40]], dtype=object)
+    right = np.array([50, 40])
+    assert not array_equivalent(left, right, strict_nan=strict_nan)
+
+
+@pytest.mark.filterwarnings("ignore:elementwise comparison failed:DeprecationWarning")
+@pytest.mark.xfail(reason="failing")
+@pytest.mark.parametrize("strict_nan", [True, False])
+def test_array_equivalent_nested_mixed_list(strict_nan):
+    # mixed arrays / lists in left and right
+    # https://github.com/pandas-dev/pandas/issues/50360
+    left = np.array([np.array([1, 2, 3]), np.array([4, 5])], dtype=object)
+    right = np.array([[1, 2, 3], [4, 5]], dtype=object)
+
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    # multiple levels of nesting
+    left = np.array(
+        [
+            np.array([np.array([1, 2, 3]), np.array([4, 5])], dtype=object),
+            np.array([np.array([6]), np.array([7, 8]), np.array([9])], dtype=object),
+        ],
+        dtype=object,
+    )
+    right = np.array([[[1, 2, 3], [4, 5]], [[6], [7, 8], [9]]], dtype=object)
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    # same-length lists
+    subarr = np.empty(2, dtype=object)
+    subarr[:] = [
+        np.array([None, "b"], dtype=object),
+        np.array(["c", "d"], dtype=object),
+    ]
+    left = np.array([subarr, None], dtype=object)
+    right = np.array([[[None, "b"], ["c", "d"]], None], dtype=object)
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+
+@pytest.mark.xfail(reason="failing")
+@pytest.mark.parametrize("strict_nan", [True, False])
+def test_array_equivalent_nested_dicts(strict_nan):
+    left = np.array([{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object)
+    right = np.array(
+        [{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object
+    )
+    assert array_equivalent(left, right, strict_nan=strict_nan)
+    assert not array_equivalent(left, right[::-1], strict_nan=strict_nan)
+
+    right2 = np.array([{"f1": 1, "f2": ["a", "b"]}], dtype=object)
+    assert array_equivalent(left, right2, strict_nan=strict_nan)
+    assert not array_equivalent(left, right2[::-1], strict_nan=strict_nan)
+
+
+def test_array_equivalent_index_with_tuples():
+    # GH#48446
+    idx1 = Index(np.array([(pd.NA, 4), (1, 1)], dtype="object"))
+    idx2 = Index(np.array([(1, 1), (pd.NA, 4)], dtype="object"))
+    assert not array_equivalent(idx1, idx2)
+    assert not idx1.equals(idx2)
+    assert not array_equivalent(idx2, idx1)
+    assert not idx2.equals(idx1)
+
+    idx1 = Index(np.array([(4, pd.NA), (1, 1)], dtype="object"))
+    idx2 = Index(np.array([(1, 1), (4, pd.NA)], dtype="object"))
+    assert not array_equivalent(idx1, idx2)
+    assert not idx1.equals(idx2)
+    assert not array_equivalent(idx2, idx1)
+    assert not idx2.equals(idx1)
+
+
+@pytest.mark.parametrize(
+    "dtype, na_value",
+    [
+        # Datetime-like
+        (np.dtype("M8[ns]"), np.datetime64("NaT", "ns")),
+        (np.dtype("m8[ns]"), np.timedelta64("NaT", "ns")),
+        (DatetimeTZDtype.construct_from_string("datetime64[ns, US/Eastern]"), NaT),
+        (PeriodDtype("M"), NaT),
+        # Integer
+        ("u1", 0),
+        ("u2", 0),
+        ("u4", 0),
+        ("u8", 0),
+        ("i1", 0),
+        ("i2", 0),
+        ("i4", 0),
+        ("i8", 0),
+        # Bool
+        ("bool", False),
+        # Float
+        ("f2", np.nan),
+        ("f4", np.nan),
+        ("f8", np.nan),
+        # Complex
+        ("c8", np.nan),
+        ("c16", np.nan),
+        # Object
+        ("O", np.nan),
+        # Interval
+        (IntervalDtype(), np.nan),
+    ],
+)
+def test_na_value_for_dtype(dtype, na_value):
+    result = na_value_for_dtype(pandas_dtype(dtype))
+    # identify check doesn't work for datetime64/timedelta64("NaT") bc they
+    #  are not singletons
+    assert result is na_value or (
+        isna(result) and isna(na_value) and type(result) is type(na_value)
+    )
+
+
+class TestNAObj:
+    def _check_behavior(self, arr, expected):
+        result = libmissing.isnaobj(arr)
+        tm.assert_numpy_array_equal(result, expected)
+
+        arr = np.atleast_2d(arr)
+        expected = np.atleast_2d(expected)
+
+        result = libmissing.isnaobj(arr)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # Test fortran order
+        arr = arr.copy(order="F")
+        result = libmissing.isnaobj(arr)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_basic(self):
+        arr = np.array([1, None, "foo", -5.1, NaT, np.nan])
+        expected = np.array([False, True, False, False, True, True])
+
+        self._check_behavior(arr, expected)
+
+    def test_non_obj_dtype(self):
+        arr = np.array([1, 3, np.nan, 5], dtype=float)
+        expected = np.array([False, False, True, False])
+
+        self._check_behavior(arr, expected)
+
+    def test_empty_arr(self):
+        arr = np.array([])
+        expected = np.array([], dtype=bool)
+
+        self._check_behavior(arr, expected)
+
+    def test_empty_str_inp(self):
+        arr = np.array([""])  # empty but not na
+        expected = np.array([False])
+
+        self._check_behavior(arr, expected)
+
+    def test_empty_like(self):
+        # see gh-13717: no segfaults!
+        arr = np.empty_like([None])
+        expected = np.array([True])
+
+        self._check_behavior(arr, expected)
+
+
+m8_units = ["as", "ps", "ns", "us", "ms", "s", "m", "h", "D", "W", "M", "Y"]
+
+na_vals = (
+    [
+        None,
+        NaT,
+        float("NaN"),
+        complex("NaN"),
+        np.nan,
+        np.float64("NaN"),
+        np.float32("NaN"),
+        np.complex64(np.nan),
+        np.complex128(np.nan),
+        np.datetime64("NaT"),
+        np.timedelta64("NaT"),
+    ]
+    + [np.datetime64("NaT", unit) for unit in m8_units]
+    + [np.timedelta64("NaT", unit) for unit in m8_units]
+)
+
+inf_vals = [
+    float("inf"),
+    float("-inf"),
+    complex("inf"),
+    complex("-inf"),
+    np.inf,
+    -np.inf,
+]
+
+int_na_vals = [
+    # Values that match iNaT, which we treat as null in specific cases
+    np.int64(NaT._value),
+    int(NaT._value),
+]
+
+sometimes_na_vals = [Decimal("NaN")]
+
+never_na_vals = [
+    # float/complex values that when viewed as int64 match iNaT
+    -0.0,
+    np.float64("-0.0"),
+    -0j,
+    np.complex64(-0j),
+]
+
+
+class TestLibMissing:
+    @pytest.mark.parametrize("func", [libmissing.checknull, isna])
+    @pytest.mark.parametrize(
+        "value",
+        na_vals + sometimes_na_vals,  # type: ignore[operator]
+    )
+    def test_checknull_na_vals(self, func, value):
+        assert func(value)
+
+    @pytest.mark.parametrize("func", [libmissing.checknull, isna])
+    @pytest.mark.parametrize("value", inf_vals)
+    def test_checknull_inf_vals(self, func, value):
+        assert not func(value)
+
+    @pytest.mark.parametrize("func", [libmissing.checknull, isna])
+    @pytest.mark.parametrize("value", int_na_vals)
+    def test_checknull_intna_vals(self, func, value):
+        assert not func(value)
+
+    @pytest.mark.parametrize("func", [libmissing.checknull, isna])
+    @pytest.mark.parametrize("value", never_na_vals)
+    def test_checknull_never_na_vals(self, func, value):
+        assert not func(value)
+
+    @pytest.mark.parametrize(
+        "value",
+        na_vals + sometimes_na_vals,  # type: ignore[operator]
+    )
+    def test_checknull_old_na_vals(self, value):
+        assert libmissing.checknull(value)
+
+    @pytest.mark.parametrize("value", int_na_vals)
+    def test_checknull_old_intna_vals(self, value):
+        assert not libmissing.checknull(value)
+
+    def test_is_matching_na(self, nulls_fixture, nulls_fixture2):
+        left = nulls_fixture
+        right = nulls_fixture2
+
+        assert libmissing.is_matching_na(left, left)
+
+        if left is right:
+            assert libmissing.is_matching_na(left, right)
+        elif is_float(left) and is_float(right):
+            # np.nan vs float("NaN") we consider as matching
+            assert libmissing.is_matching_na(left, right)
+        elif type(left) is type(right):
+            # e.g. both Decimal("NaN")
+            assert libmissing.is_matching_na(left, right)
+        else:
+            assert not libmissing.is_matching_na(left, right)
+
+    def test_is_matching_na_nan_matches_none(self):
+        assert not libmissing.is_matching_na(None, np.nan)
+        assert not libmissing.is_matching_na(np.nan, None)
+
+        assert libmissing.is_matching_na(None, np.nan, nan_matches_none=True)
+        assert libmissing.is_matching_na(np.nan, None, nan_matches_none=True)
+
+
+class TestIsValidNAForDtype:
+    def test_is_valid_na_for_dtype_interval(self):
+        dtype = IntervalDtype("int64", "left")
+        assert not is_valid_na_for_dtype(NaT, dtype)
+
+        dtype = IntervalDtype("datetime64[ns]", "both")
+        assert not is_valid_na_for_dtype(NaT, dtype)
+
+    def test_is_valid_na_for_dtype_categorical(self):
+        dtype = CategoricalDtype(categories=[0, 1, 2])
+        assert is_valid_na_for_dtype(np.nan, dtype)
+
+        assert not is_valid_na_for_dtype(NaT, dtype)
+        assert not is_valid_na_for_dtype(np.datetime64("NaT", "ns"), dtype)
+        assert not is_valid_na_for_dtype(np.timedelta64("NaT", "ns"), dtype)
diff --git a/pandas/pandas/tests/extension/__init__.py b/pandas/pandas/tests/extension/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/extension/array_with_attr/__init__.py b/pandas/pandas/tests/extension/array_with_attr/__init__.py
new file mode 100644
index 0000000000000..49da6af024a31
--- /dev/null
+++ b/pandas/pandas/tests/extension/array_with_attr/__init__.py
@@ -0,0 +1,6 @@
+from pandas.tests.extension.array_with_attr.array import (
+    FloatAttrArray,
+    FloatAttrDtype,
+)
+
+__all__ = ["FloatAttrArray", "FloatAttrDtype"]
diff --git a/pandas/pandas/tests/extension/array_with_attr/array.py b/pandas/pandas/tests/extension/array_with_attr/array.py
new file mode 100644
index 0000000000000..4f65424ece145
--- /dev/null
+++ b/pandas/pandas/tests/extension/array_with_attr/array.py
@@ -0,0 +1,90 @@
+"""
+Test extension array that has custom attribute information (not stored on the dtype).
+
+"""
+
+from __future__ import annotations
+
+import numbers
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.base import ExtensionDtype
+
+import pandas as pd
+from pandas.core.arrays import ExtensionArray
+
+if TYPE_CHECKING:
+    from pandas._typing import type_t
+
+
+class FloatAttrDtype(ExtensionDtype):
+    type = float
+    name = "float_attr"
+    na_value = np.nan
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[FloatAttrArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return FloatAttrArray
+
+
+class FloatAttrArray(ExtensionArray):
+    dtype = FloatAttrDtype()
+    __array_priority__ = 1000
+
+    def __init__(self, values, attr=None) -> None:
+        if not isinstance(values, np.ndarray):
+            raise TypeError("Need to pass a numpy array of float64 dtype as values")
+        if not values.dtype == "float64":
+            raise TypeError("Need to pass a numpy array of float64 dtype as values")
+        self.data = values
+        self.attr = attr
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        if not copy:
+            data = np.asarray(scalars, dtype="float64")
+        else:
+            data = np.array(scalars, dtype="float64", copy=copy)
+        return cls(data)
+
+    def __getitem__(self, item):
+        if isinstance(item, numbers.Integral):
+            return self.data[item]
+        else:
+            # slice, list-like, mask
+            item = pd.api.indexers.check_array_indexer(self, item)
+            return type(self)(self.data[item], self.attr)
+
+    def __len__(self) -> int:
+        return len(self.data)
+
+    def isna(self):
+        return np.isnan(self.data)
+
+    def take(self, indexer, allow_fill=False, fill_value=None):
+        from pandas.api.extensions import take
+
+        data = self.data
+        if allow_fill and fill_value is None:
+            fill_value = self.dtype.na_value
+
+        result = take(data, indexer, fill_value=fill_value, allow_fill=allow_fill)
+        return type(self)(result, self.attr)
+
+    def copy(self):
+        return type(self)(self.data.copy(), self.attr)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat):
+        data = np.concatenate([x.data for x in to_concat])
+        attr = to_concat[0].attr if len(to_concat) else None
+        return cls(data, attr)
diff --git a/pandas/pandas/tests/extension/array_with_attr/test_array_with_attr.py b/pandas/pandas/tests/extension/array_with_attr/test_array_with_attr.py
new file mode 100644
index 0000000000000..3735fe40a0d67
--- /dev/null
+++ b/pandas/pandas/tests/extension/array_with_attr/test_array_with_attr.py
@@ -0,0 +1,33 @@
+import numpy as np
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.extension.array_with_attr import FloatAttrArray
+
+
+def test_concat_with_all_na():
+    # https://github.com/pandas-dev/pandas/pull/47762
+    # ensure that attribute of the column array is preserved (when it gets
+    # preserved in reindexing the array) during merge/concat
+    arr = FloatAttrArray(np.array([np.nan, np.nan], dtype="float64"), attr="test")
+
+    df1 = pd.DataFrame({"col": arr, "key": [0, 1]})
+    df2 = pd.DataFrame({"key": [0, 1], "col2": [1, 2]})
+    result = pd.merge(df1, df2, on="key")
+    expected = pd.DataFrame({"col": arr, "key": [0, 1], "col2": [1, 2]})
+    tm.assert_frame_equal(result, expected)
+    assert result["col"].array.attr == "test"
+
+    df1 = pd.DataFrame({"col": arr, "key": [0, 1]})
+    df2 = pd.DataFrame({"key": [0, 2], "col2": [1, 2]})
+    result = pd.merge(df1, df2, on="key")
+    expected = pd.DataFrame({"col": arr.take([0]), "key": [0], "col2": [1]})
+    tm.assert_frame_equal(result, expected)
+    assert result["col"].array.attr == "test"
+
+    result = pd.concat([df1.set_index("key"), df2.set_index("key")], axis=1)
+    expected = pd.DataFrame(
+        {"col": arr.take([0, 1, -1]), "col2": [1, np.nan, 2], "key": [0, 1, 2]}
+    ).set_index("key")
+    tm.assert_frame_equal(result, expected)
+    assert result["col"].array.attr == "test"
diff --git a/pandas/pandas/tests/extension/base/__init__.py b/pandas/pandas/tests/extension/base/__init__.py
new file mode 100644
index 0000000000000..2abf739d2428d
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/__init__.py
@@ -0,0 +1,89 @@
+"""
+Base test suite for extension arrays.
+
+These tests are intended for third-party libraries to subclass to validate
+that their extension arrays and dtypes satisfy the interface. Moving or
+renaming the tests should not be done lightly.
+
+Libraries are expected to implement a few pytest fixtures to provide data
+for the tests. The fixtures may be located in either
+
+* The same module as your test class.
+* A ``conftest.py`` in the same directory as your test class.
+
+The full list of fixtures may be found in the ``conftest.py`` next to this
+file.
+
+.. code-block:: python
+
+   import pytest
+   from pandas.tests.extension.base import BaseDtypeTests
+
+
+   @pytest.fixture
+   def dtype():
+       return MyDtype()
+
+
+   class TestMyDtype(BaseDtypeTests):
+       pass
+
+
+Your class ``TestDtype`` will inherit all the tests defined on
+``BaseDtypeTests``. pytest's fixture discover will supply your ``dtype``
+wherever the test requires it. You're free to implement additional tests.
+
+"""
+
+from pandas.tests.extension.base.accumulate import BaseAccumulateTests
+from pandas.tests.extension.base.casting import BaseCastingTests
+from pandas.tests.extension.base.constructors import BaseConstructorsTests
+from pandas.tests.extension.base.dim2 import (  # noqa: F401
+    Dim2CompatTests,
+    NDArrayBacked2DTests,
+)
+from pandas.tests.extension.base.dtype import BaseDtypeTests
+from pandas.tests.extension.base.getitem import BaseGetitemTests
+from pandas.tests.extension.base.groupby import BaseGroupbyTests
+from pandas.tests.extension.base.index import BaseIndexTests
+from pandas.tests.extension.base.interface import BaseInterfaceTests
+from pandas.tests.extension.base.io import BaseParsingTests
+from pandas.tests.extension.base.methods import BaseMethodsTests
+from pandas.tests.extension.base.missing import BaseMissingTests
+from pandas.tests.extension.base.ops import (  # noqa: F401
+    BaseArithmeticOpsTests,
+    BaseComparisonOpsTests,
+    BaseOpsUtil,
+    BaseUnaryOpsTests,
+)
+from pandas.tests.extension.base.printing import BasePrintingTests
+from pandas.tests.extension.base.reduce import BaseReduceTests
+from pandas.tests.extension.base.reshaping import BaseReshapingTests
+from pandas.tests.extension.base.setitem import BaseSetitemTests
+
+
+# One test class that you can inherit as an alternative to inheriting all the
+# test classes above.
+# Note this excludes Dim2CompatTests and NDArrayBacked2DTests.
+class ExtensionTests(
+    BaseAccumulateTests,
+    BaseCastingTests,
+    BaseConstructorsTests,
+    BaseDtypeTests,
+    BaseGetitemTests,
+    BaseGroupbyTests,
+    BaseIndexTests,
+    BaseInterfaceTests,
+    BaseParsingTests,
+    BaseMethodsTests,
+    BaseMissingTests,
+    BaseArithmeticOpsTests,
+    BaseComparisonOpsTests,
+    BaseUnaryOpsTests,
+    BasePrintingTests,
+    BaseReduceTests,
+    BaseReshapingTests,
+    BaseSetitemTests,
+    Dim2CompatTests,
+):
+    pass
diff --git a/pandas/pandas/tests/extension/base/accumulate.py b/pandas/pandas/tests/extension/base/accumulate.py
new file mode 100644
index 0000000000000..9a2f186c2a00b
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/accumulate.py
@@ -0,0 +1,40 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseAccumulateTests:
+    """
+    Accumulation specific tests. Generally these only
+    make sense for numeric/boolean operations.
+    """
+
+    def _supports_accumulation(self, ser: pd.Series, op_name: str) -> bool:
+        # Do we expect this accumulation to be supported for this dtype?
+        # We default to assuming "no"; subclass authors should override here.
+        return False
+
+    def check_accumulate(self, ser: pd.Series, op_name: str, skipna: bool):
+        try:
+            alt = ser.astype("float64")
+        except (TypeError, ValueError):
+            # e.g. Period can't be cast to float64 (TypeError)
+            #      String can't be cast to float64 (ValueError)
+            alt = ser.astype(object)
+
+        result = getattr(ser, op_name)(skipna=skipna)
+        expected = getattr(alt, op_name)(skipna=skipna)
+        tm.assert_series_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_accumulate_series(self, data, all_numeric_accumulations, skipna):
+        op_name = all_numeric_accumulations
+        ser = pd.Series(data)
+
+        if self._supports_accumulation(ser, op_name):
+            self.check_accumulate(ser, op_name, skipna)
+        else:
+            with pytest.raises((NotImplementedError, TypeError)):
+                # TODO: require TypeError for things that will _never_ work?
+                getattr(ser, op_name)(skipna=skipna)
diff --git a/pandas/pandas/tests/extension/base/base.py b/pandas/pandas/tests/extension/base/base.py
new file mode 100644
index 0000000000000..747ebee738c1e
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/base.py
@@ -0,0 +1,2 @@
+class BaseExtensionTests:
+    pass
diff --git a/pandas/pandas/tests/extension/base/casting.py b/pandas/pandas/tests/extension/base/casting.py
new file mode 100644
index 0000000000000..8e3f21e1a4f56
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/casting.py
@@ -0,0 +1,88 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.internals.blocks import NumpyBlock
+
+
+class BaseCastingTests:
+    """Casting to and from ExtensionDtypes"""
+
+    def test_astype_object_series(self, all_data):
+        ser = pd.Series(all_data, name="A")
+        result = ser.astype(object)
+        assert result.dtype == np.dtype(object)
+        if hasattr(result._mgr, "blocks"):
+            blk = result._mgr.blocks[0]
+            assert isinstance(blk, NumpyBlock)
+            assert blk.is_object
+        assert isinstance(result._mgr.array, np.ndarray)
+        assert result._mgr.array.dtype == np.dtype(object)
+
+    def test_astype_object_frame(self, all_data):
+        df = pd.DataFrame({"A": all_data})
+
+        result = df.astype(object)
+        if hasattr(result._mgr, "blocks"):
+            blk = result._mgr.blocks[0]
+            assert isinstance(blk, NumpyBlock), type(blk)
+            assert blk.is_object
+        arr = result._mgr.blocks[0].values
+        assert isinstance(arr, np.ndarray)
+        assert arr.dtype == np.dtype(object)
+
+        # check that we can compare the dtypes
+        comp = result.dtypes == df.dtypes
+        assert not comp.any()
+
+    def test_tolist(self, data):
+        result = pd.Series(data).tolist()
+        expected = list(data)
+        assert result == expected
+
+    def test_astype_str(self, data):
+        result = pd.Series(data[:2]).astype(str)
+        expected = pd.Series([str(x) for x in data[:2]], dtype=str)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "nullable_string_dtype",
+        [
+            "string[python]",
+            pytest.param("string[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        ],
+    )
+    def test_astype_string(self, data, nullable_string_dtype):
+        # GH-33465, GH#45326 as of 2.0 we decode bytes instead of calling str(obj)
+        result = pd.Series(data[:5]).astype(nullable_string_dtype)
+        expected = pd.Series(
+            [str(x) if not isinstance(x, bytes) else x.decode() for x in data[:5]],
+            dtype=nullable_string_dtype,
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_to_numpy(self, data):
+        expected = np.asarray(data)
+
+        result = data.to_numpy()
+        tm.assert_equal(result, expected)
+
+        result = pd.Series(data).to_numpy()
+        tm.assert_equal(result, expected)
+
+    def test_astype_empty_dataframe(self, dtype):
+        # https://github.com/pandas-dev/pandas/issues/33113
+        df = pd.DataFrame()
+        result = df.astype(dtype)
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize("copy", [True, False])
+    def test_astype_own_type(self, data, copy):
+        # ensure that astype returns the original object for equal dtype and copy=False
+        # https://github.com/pandas-dev/pandas/issues/28488
+        result = data.astype(data.dtype, copy=copy)
+        assert (result is data) is (not copy)
+        tm.assert_extension_array_equal(result, data)
diff --git a/pandas/pandas/tests/extension/base/constructors.py b/pandas/pandas/tests/extension/base/constructors.py
new file mode 100644
index 0000000000000..639dc874c9fb9
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/constructors.py
@@ -0,0 +1,142 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.extensions import ExtensionArray
+from pandas.core.internals.blocks import EABackedBlock
+
+
+class BaseConstructorsTests:
+    def test_from_sequence_from_cls(self, data):
+        result = type(data)._from_sequence(data, dtype=data.dtype)
+        tm.assert_extension_array_equal(result, data)
+
+        data = data[:0]
+        result = type(data)._from_sequence(data, dtype=data.dtype)
+        tm.assert_extension_array_equal(result, data)
+
+    def test_array_from_scalars(self, data):
+        scalars = [data[0], data[1], data[2]]
+        result = data._from_sequence(scalars, dtype=data.dtype)
+        assert isinstance(result, type(data))
+
+    def test_series_constructor(self, data):
+        result = pd.Series(data, copy=False)
+        assert result.dtype == data.dtype
+        assert len(result) == len(data)
+        if hasattr(result._mgr, "blocks"):
+            assert isinstance(result._mgr.blocks[0], EABackedBlock)
+        assert result._mgr.array is data
+
+        # Series[EA] is unboxed / boxed correctly
+        result2 = pd.Series(result)
+        assert result2.dtype == data.dtype
+        if hasattr(result._mgr, "blocks"):
+            assert isinstance(result2._mgr.blocks[0], EABackedBlock)
+
+    def test_series_constructor_no_data_with_index(self, dtype, na_value):
+        result = pd.Series(index=[1, 2, 3], dtype=dtype)
+        expected = pd.Series([na_value] * 3, index=[1, 2, 3], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        # GH 33559 - empty index
+        result = pd.Series(index=[], dtype=dtype)
+        expected = pd.Series([], index=pd.Index([], dtype="object"), dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_scalar_na_with_index(self, dtype, na_value):
+        result = pd.Series(na_value, index=[1, 2, 3], dtype=dtype)
+        expected = pd.Series([na_value] * 3, index=[1, 2, 3], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_scalar_with_index(self, data, dtype):
+        scalar = data[0]
+        result = pd.Series(scalar, index=[1, 2, 3], dtype=dtype)
+        expected = pd.Series([scalar] * 3, index=[1, 2, 3], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        result = pd.Series(scalar, index=["foo"], dtype=dtype)
+        expected = pd.Series([scalar], index=["foo"], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("from_series", [True, False])
+    def test_dataframe_constructor_from_dict(self, data, from_series):
+        if from_series:
+            data = pd.Series(data)
+        result = pd.DataFrame({"A": data})
+        assert result.dtypes["A"] == data.dtype
+        assert result.shape == (len(data), 1)
+        if hasattr(result._mgr, "blocks"):
+            assert isinstance(result._mgr.blocks[0], EABackedBlock)
+        assert isinstance(result._mgr.blocks[0].values, ExtensionArray)
+
+    def test_dataframe_from_series(self, data):
+        result = pd.DataFrame(pd.Series(data))
+        assert result.dtypes[0] == data.dtype
+        assert result.shape == (len(data), 1)
+        if hasattr(result._mgr, "blocks"):
+            assert isinstance(result._mgr.blocks[0], EABackedBlock)
+        assert isinstance(result._mgr.blocks[0].values, ExtensionArray)
+
+    def test_series_given_mismatched_index_raises(self, data):
+        msg = r"Length of values \(3\) does not match length of index \(5\)"
+        with pytest.raises(ValueError, match=msg):
+            pd.Series(data[:3], index=[0, 1, 2, 3, 4])
+
+    def test_from_dtype(self, data):
+        # construct from our dtype & string dtype
+        dtype = data.dtype
+
+        expected = pd.Series(data)
+        result = pd.Series(list(data), dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        result = pd.Series(list(data), dtype=str(dtype))
+        tm.assert_series_equal(result, expected)
+
+        # gh-30280
+
+        expected = pd.DataFrame(data).astype(dtype)
+        result = pd.DataFrame(list(data), dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.DataFrame(list(data), dtype=str(dtype))
+        tm.assert_frame_equal(result, expected)
+
+    def test_pandas_array(self, data):
+        # pd.array(extension_array) should be idempotent...
+        result = pd.array(data)
+        tm.assert_extension_array_equal(result, data)
+
+    def test_pandas_array_dtype(self, data):
+        # ... but specifying dtype will override idempotency
+        result = pd.array(data, dtype=np.dtype(object))
+        expected = pd.arrays.NumpyExtensionArray(np.asarray(data, dtype=object))
+        tm.assert_equal(result, expected)
+
+    def test_construct_empty_dataframe(self, dtype):
+        # GH 33623
+        result = pd.DataFrame(columns=["a"], dtype=dtype)
+        expected = pd.DataFrame(
+            {"a": pd.array([], dtype=dtype)}, index=pd.RangeIndex(0)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_empty(self, dtype):
+        cls = dtype.construct_array_type()
+        result = cls._empty((4,), dtype=dtype)
+        assert isinstance(result, cls)
+        assert result.dtype == dtype
+        assert result.shape == (4,)
+
+        # GH#19600 method on ExtensionDtype
+        result2 = dtype.empty((4,))
+        assert isinstance(result2, cls)
+        assert result2.dtype == dtype
+        assert result2.shape == (4,)
+
+        result2 = dtype.empty(4)
+        assert isinstance(result2, cls)
+        assert result2.dtype == dtype
+        assert result2.shape == (4,)
diff --git a/pandas/pandas/tests/extension/base/dim2.py b/pandas/pandas/tests/extension/base/dim2.py
new file mode 100644
index 0000000000000..8c7d8ff491cd3
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/dim2.py
@@ -0,0 +1,346 @@
+"""
+Tests for 2D compatibility.
+"""
+
+import numpy as np
+import pytest
+
+from pandas._libs.missing import is_matching_na
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_integer_dtype,
+)
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.integer import NUMPY_INT_TO_DTYPE
+
+
+class Dim2CompatTests:
+    # Note: these are ONLY for ExtensionArray subclasses that support 2D arrays.
+    #  i.e. not for pyarrow-backed EAs.
+
+    @pytest.fixture(autouse=True)
+    def skip_if_doesnt_support_2d(self, dtype, request):
+        if not dtype._supports_2d:
+            node = request.node
+            # In cases where we are mixed in to ExtensionTests, we only want to
+            #  skip tests that are defined in Dim2CompatTests
+            test_func = node._obj
+            if test_func.__qualname__.startswith("Dim2CompatTests"):
+                # TODO: is there a less hacky way of checking this?
+                pytest.skip(f"{dtype} does not support 2D.")
+
+    def test_transpose(self, data):
+        arr2d = data.repeat(2).reshape(-1, 2)
+        shape = arr2d.shape
+        assert shape[0] != shape[-1]  # otherwise the rest of the test is useless
+
+        assert arr2d.T.shape == shape[::-1]
+
+    def test_frame_from_2d_array(self, data):
+        arr2d = data.repeat(2).reshape(-1, 2)
+
+        df = pd.DataFrame(arr2d)
+        expected = pd.DataFrame({0: arr2d[:, 0], 1: arr2d[:, 1]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_swapaxes(self, data):
+        arr2d = data.repeat(2).reshape(-1, 2)
+
+        result = arr2d.swapaxes(0, 1)
+        expected = arr2d.T
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_delete_2d(self, data):
+        arr2d = data.repeat(3).reshape(-1, 3)
+
+        # axis = 0
+        result = arr2d.delete(1, axis=0)
+        expected = data.delete(1).repeat(3).reshape(-1, 3)
+        tm.assert_extension_array_equal(result, expected)
+
+        # axis = 1
+        result = arr2d.delete(1, axis=1)
+        expected = data.repeat(2).reshape(-1, 2)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_take_2d(self, data):
+        arr2d = data.reshape(-1, 1)
+
+        result = arr2d.take([0, 0, -1], axis=0)
+
+        expected = data.take([0, 0, -1]).reshape(-1, 1)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_repr_2d(self, data):
+        # this could fail in a corner case where an element contained the name
+        res = repr(data.reshape(1, -1))
+        assert res.count(f"<{type(data).__name__}") == 1
+
+        res = repr(data.reshape(-1, 1))
+        assert res.count(f"<{type(data).__name__}") == 1
+
+    def test_reshape(self, data):
+        arr2d = data.reshape(-1, 1)
+        assert arr2d.shape == (data.size, 1)
+        assert len(arr2d) == len(data)
+
+        arr2d = data.reshape((-1, 1))
+        assert arr2d.shape == (data.size, 1)
+        assert len(arr2d) == len(data)
+
+        with tm.external_error_raised(ValueError):
+            data.reshape((data.size, 2))
+        with tm.external_error_raised(ValueError):
+            data.reshape(data.size, 2)
+
+    def test_getitem_2d(self, data):
+        arr2d = data.reshape(1, -1)
+
+        result = arr2d[0]
+        tm.assert_extension_array_equal(result, data)
+
+        with pytest.raises(IndexError):
+            arr2d[1]
+
+        with pytest.raises(IndexError):
+            arr2d[-2]
+
+        result = arr2d[:]
+        tm.assert_extension_array_equal(result, arr2d)
+
+        result = arr2d[:, :]
+        tm.assert_extension_array_equal(result, arr2d)
+
+        result = arr2d[:, 0]
+        expected = data[[0]]
+        tm.assert_extension_array_equal(result, expected)
+
+        # dimension-expanding getitem on 1D
+        result = data[:, np.newaxis]
+        tm.assert_extension_array_equal(result, arr2d.T)
+
+    def test_iter_2d(self, data):
+        arr2d = data.reshape(1, -1)
+
+        objs = list(iter(arr2d))
+        assert len(objs) == arr2d.shape[0]
+
+        for obj in objs:
+            assert isinstance(obj, type(data))
+            assert obj.dtype == data.dtype
+            assert obj.ndim == 1
+            assert len(obj) == arr2d.shape[1]
+
+    def test_tolist_2d(self, data):
+        arr2d = data.reshape(1, -1)
+
+        result = arr2d.tolist()
+        expected = [data.tolist()]
+
+        assert isinstance(result, list)
+        assert all(isinstance(x, list) for x in result)
+
+        assert result == expected
+
+    def test_concat_2d(self, data):
+        left = type(data)._concat_same_type([data, data]).reshape(-1, 2)
+        right = left.copy()
+
+        # axis=0
+        result = left._concat_same_type([left, right], axis=0)
+        expected = data._concat_same_type([data] * 4).reshape(-1, 2)
+        tm.assert_extension_array_equal(result, expected)
+
+        # axis=1
+        result = left._concat_same_type([left, right], axis=1)
+        assert result.shape == (len(data), 4)
+        tm.assert_extension_array_equal(result[:, :2], left)
+        tm.assert_extension_array_equal(result[:, 2:], right)
+
+        # axis > 1 -> invalid
+        msg = "axis 2 is out of bounds for array of dimension 2"
+        with pytest.raises(ValueError, match=msg):
+            left._concat_same_type([left, right], axis=2)
+
+    @pytest.mark.parametrize("method", ["backfill", "pad"])
+    def test_fillna_2d_method(self, data_missing, method):
+        # pad_or_backfill is always along axis=0
+        arr = data_missing.repeat(2).reshape(2, 2)
+        assert arr[0].isna().all()
+        assert not arr[1].isna().any()
+
+        result = arr._pad_or_backfill(method=method, limit=None)
+
+        expected = data_missing._pad_or_backfill(method=method).repeat(2).reshape(2, 2)
+        tm.assert_extension_array_equal(result, expected)
+
+        # Reverse so that backfill is not a no-op.
+        arr2 = arr[::-1]
+        assert not arr2[0].isna().any()
+        assert arr2[1].isna().all()
+
+        result2 = arr2._pad_or_backfill(method=method, limit=None)
+
+        expected2 = (
+            data_missing[::-1]._pad_or_backfill(method=method).repeat(2).reshape(2, 2)
+        )
+        tm.assert_extension_array_equal(result2, expected2)
+
+    @pytest.mark.parametrize("method", ["mean", "median", "var", "std", "sum", "prod"])
+    def test_reductions_2d_axis_none(self, data, method):
+        arr2d = data.reshape(1, -1)
+
+        err_expected = None
+        err_result = None
+        try:
+            expected = getattr(data, method)()
+        except Exception as err:
+            # if the 1D reduction is invalid, the 2D reduction should be as well
+            err_expected = err
+            try:
+                result = getattr(arr2d, method)(axis=None)
+            except Exception as err2:
+                err_result = err2
+
+        else:
+            result = getattr(arr2d, method)(axis=None)
+
+        if err_result is not None or err_expected is not None:
+            assert type(err_result) == type(err_expected)
+            return
+
+        assert is_matching_na(result, expected) or result == expected
+
+    @pytest.mark.parametrize("method", ["mean", "median", "var", "std", "sum", "prod"])
+    @pytest.mark.parametrize("min_count", [0, 1])
+    def test_reductions_2d_axis0(self, data, method, min_count):
+        if min_count == 1 and method not in ["sum", "prod"]:
+            pytest.skip(f"min_count not relevant for {method}")
+
+        arr2d = data.reshape(1, -1)
+
+        kwargs = {}
+        if method in ["std", "var"]:
+            # pass ddof=0 so we get all-zero std instead of all-NA std
+            kwargs["ddof"] = 0
+        elif method in ["prod", "sum"]:
+            kwargs["min_count"] = min_count
+
+        try:
+            result = getattr(arr2d, method)(axis=0, **kwargs)
+        except Exception as err:
+            try:
+                getattr(data, method)()
+            except Exception as err2:
+                assert type(err) == type(err2)
+                return
+            else:
+                raise AssertionError("Both reductions should raise or neither")
+
+        def get_reduction_result_dtype(dtype):
+            # windows and 32bit builds will in some cases have int32/uint32
+            #  where other builds will have int64/uint64.
+            if dtype.itemsize == 8:
+                return dtype
+            elif dtype.kind in "ib":
+                return NUMPY_INT_TO_DTYPE[np.dtype(int)]
+            else:
+                # i.e. dtype.kind == "u"
+                return NUMPY_INT_TO_DTYPE[np.dtype("uint")]
+
+        if method in ["sum", "prod"]:
+            # std and var are not dtype-preserving
+            expected = data
+            if data.dtype.kind in "iub":
+                dtype = get_reduction_result_dtype(data.dtype)
+                expected = data.astype(dtype)
+                assert dtype == expected.dtype
+
+            if min_count == 0:
+                fill_value = 1 if method == "prod" else 0
+                expected = expected.fillna(fill_value)
+
+            tm.assert_extension_array_equal(result, expected)
+        elif method == "median":
+            # std and var are not dtype-preserving
+            expected = data
+            tm.assert_extension_array_equal(result, expected)
+        elif method in ["mean", "std", "var"]:
+            if is_integer_dtype(data) or is_bool_dtype(data):
+                data = data.astype("Float64")
+            if method == "mean":
+                tm.assert_extension_array_equal(result, data)
+            else:
+                tm.assert_extension_array_equal(result, data - data)
+
+    @pytest.mark.parametrize("method", ["mean", "median", "var", "std", "sum", "prod"])
+    def test_reductions_2d_axis1(self, data, method):
+        arr2d = data.reshape(1, -1)
+
+        try:
+            result = getattr(arr2d, method)(axis=1)
+        except Exception as err:
+            try:
+                getattr(data, method)()
+            except Exception as err2:
+                assert type(err) == type(err2)
+                return
+            else:
+                raise AssertionError("Both reductions should raise or neither")
+
+        # not necessarily type/dtype-preserving, so weaker assertions
+        assert result.shape == (1,)
+        expected_scalar = getattr(data, method)()
+        res = result[0]
+        assert is_matching_na(res, expected_scalar) or res == expected_scalar
+
+
+class NDArrayBacked2DTests(Dim2CompatTests):
+    # More specific tests for NDArrayBackedExtensionArray subclasses
+
+    def test_copy_order(self, data):
+        # We should be matching numpy semantics for the "order" keyword in 'copy'
+        arr2d = data.repeat(2).reshape(-1, 2)
+        assert arr2d._ndarray.flags["C_CONTIGUOUS"]
+
+        res = arr2d.copy()
+        assert res._ndarray.flags["C_CONTIGUOUS"]
+
+        res = arr2d[::2, ::2].copy()
+        assert res._ndarray.flags["C_CONTIGUOUS"]
+
+        res = arr2d.copy("F")
+        assert not res._ndarray.flags["C_CONTIGUOUS"]
+        assert res._ndarray.flags["F_CONTIGUOUS"]
+
+        res = arr2d.copy("K")
+        assert res._ndarray.flags["C_CONTIGUOUS"]
+
+        res = arr2d.T.copy("K")
+        assert not res._ndarray.flags["C_CONTIGUOUS"]
+        assert res._ndarray.flags["F_CONTIGUOUS"]
+
+        # order not accepted by numpy
+        msg = r"order must be one of 'C', 'F', 'A', or 'K' \(got 'Q'\)"
+        with pytest.raises(ValueError, match=msg):
+            arr2d.copy("Q")
+
+        # neither contiguity
+        arr_nc = arr2d[::2]
+        assert not arr_nc._ndarray.flags["C_CONTIGUOUS"]
+        assert not arr_nc._ndarray.flags["F_CONTIGUOUS"]
+
+        assert arr_nc.copy()._ndarray.flags["C_CONTIGUOUS"]
+        assert not arr_nc.copy()._ndarray.flags["F_CONTIGUOUS"]
+
+        assert arr_nc.copy("C")._ndarray.flags["C_CONTIGUOUS"]
+        assert not arr_nc.copy("C")._ndarray.flags["F_CONTIGUOUS"]
+
+        assert not arr_nc.copy("F")._ndarray.flags["C_CONTIGUOUS"]
+        assert arr_nc.copy("F")._ndarray.flags["F_CONTIGUOUS"]
+
+        assert arr_nc.copy("K")._ndarray.flags["C_CONTIGUOUS"]
+        assert not arr_nc.copy("K")._ndarray.flags["F_CONTIGUOUS"]
diff --git a/pandas/pandas/tests/extension/base/dtype.py b/pandas/pandas/tests/extension/base/dtype.py
new file mode 100644
index 0000000000000..38478ed3c40ae
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/dtype.py
@@ -0,0 +1,123 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import (
+    infer_dtype,
+    is_object_dtype,
+    is_string_dtype,
+)
+
+
+class BaseDtypeTests:
+    """Base class for ExtensionDtype classes"""
+
+    def test_name(self, dtype):
+        assert isinstance(dtype.name, str)
+
+    def test_kind(self, dtype):
+        valid = set("biufcmMOSUV")
+        assert dtype.kind in valid
+
+    def test_is_dtype_from_name(self, dtype):
+        result = type(dtype).is_dtype(dtype.name)
+        assert result is True
+
+    def test_is_dtype_unboxes_dtype(self, data, dtype):
+        assert dtype.is_dtype(data) is True
+
+    def test_is_dtype_from_self(self, dtype):
+        result = type(dtype).is_dtype(dtype)
+        assert result is True
+
+    def test_is_dtype_other_input(self, dtype):
+        assert dtype.is_dtype([1, 2, 3]) is False
+
+    def test_is_not_string_type(self, dtype):
+        assert not is_string_dtype(dtype)
+
+    def test_is_not_object_type(self, dtype):
+        assert not is_object_dtype(dtype)
+
+    def test_eq_with_str(self, dtype):
+        assert dtype == dtype.name
+        assert dtype != dtype.name + "-suffix"
+
+    def test_eq_with_numpy_object(self, dtype):
+        assert dtype != np.dtype("object")
+
+    def test_eq_with_self(self, dtype):
+        assert dtype == dtype
+        assert dtype != object()
+
+    def test_array_type(self, data, dtype):
+        assert dtype.construct_array_type() is type(data)
+
+    def test_check_dtype(self, data):
+        dtype = data.dtype
+
+        # check equivalency for using .dtypes
+        df = pd.DataFrame(
+            {
+                "A": pd.Series(data, dtype=dtype),
+                "B": data,
+                "C": pd.Series(["foo"] * len(data), dtype=object),
+                "D": 1,
+            }
+        )
+        result = df.dtypes == str(dtype)
+        assert np.dtype("int64") != "Int64"
+
+        expected = pd.Series([True, True, False, False], index=list("ABCD"))
+
+        tm.assert_series_equal(result, expected)
+
+        expected = pd.Series([True, True, False, False], index=list("ABCD"))
+        result = df.dtypes.apply(str) == str(dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_hashable(self, dtype):
+        hash(dtype)  # no error
+
+    def test_str(self, dtype):
+        assert str(dtype) == dtype.name
+
+    def test_eq(self, dtype):
+        assert dtype == dtype.name
+        assert dtype != "another_type"
+
+    def test_construct_from_string_own_name(self, dtype):
+        result = dtype.construct_from_string(dtype.name)
+        assert type(result) is type(dtype)
+
+        # check OK as classmethod
+        result = type(dtype).construct_from_string(dtype.name)
+        assert type(result) is type(dtype)
+
+    def test_construct_from_string_another_type_raises(self, dtype):
+        msg = f"Cannot construct a '{type(dtype).__name__}' from 'another_type'"
+        with pytest.raises(TypeError, match=msg):
+            type(dtype).construct_from_string("another_type")
+
+    def test_construct_from_string_wrong_type_raises(self, dtype):
+        with pytest.raises(
+            TypeError,
+            match="'construct_from_string' expects a string, got <class 'int'>",
+        ):
+            type(dtype).construct_from_string(0)
+
+    def test_get_common_dtype(self, dtype):
+        # in practice we will not typically call this with a 1-length list
+        # (we shortcut to just use that dtype as the common dtype), but
+        # still testing as good practice to have this working (and it is the
+        # only case we can test in general)
+        assert dtype._get_common_dtype([dtype]) == dtype
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_infer_dtype(self, data, data_missing, skipna):
+        # only testing that this works without raising an error
+        res = infer_dtype(data, skipna=skipna)
+        assert isinstance(res, str)
+        res = infer_dtype(data_missing, skipna=skipna)
+        assert isinstance(res, str)
diff --git a/pandas/pandas/tests/extension/base/getitem.py b/pandas/pandas/tests/extension/base/getitem.py
new file mode 100644
index 0000000000000..1f3680bf67e90
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/getitem.py
@@ -0,0 +1,469 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseGetitemTests:
+    """Tests for ExtensionArray.__getitem__."""
+
+    def test_iloc_series(self, data):
+        ser = pd.Series(data)
+        result = ser.iloc[:4]
+        expected = pd.Series(data[:4])
+        tm.assert_series_equal(result, expected)
+
+        result = ser.iloc[[0, 1, 2, 3]]
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_frame(self, data):
+        df = pd.DataFrame({"A": data, "B": np.arange(len(data), dtype="int64")})
+        expected = pd.DataFrame({"A": data[:4]})
+
+        # slice -> frame
+        result = df.iloc[:4, [0]]
+        tm.assert_frame_equal(result, expected)
+
+        # sequence -> frame
+        result = df.iloc[[0, 1, 2, 3], [0]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = pd.Series(data[:4], name="A")
+
+        # slice -> series
+        result = df.iloc[:4, 0]
+        tm.assert_series_equal(result, expected)
+
+        # sequence -> series
+        result = df.iloc[:4, 0]
+        tm.assert_series_equal(result, expected)
+
+        # GH#32959 slice columns with step
+        result = df.iloc[:, ::2]
+        tm.assert_frame_equal(result, df[["A"]])
+        result = df[["B", "A"]].iloc[:, ::2]
+        tm.assert_frame_equal(result, df[["B"]])
+
+    def test_iloc_frame_single_block(self, data):
+        # GH#32959 null slice along index, slice along columns with single-block
+        df = pd.DataFrame({"A": data})
+
+        result = df.iloc[:, :]
+        tm.assert_frame_equal(result, df)
+
+        result = df.iloc[:, :1]
+        tm.assert_frame_equal(result, df)
+
+        result = df.iloc[:, :2]
+        tm.assert_frame_equal(result, df)
+
+        result = df.iloc[:, ::2]
+        tm.assert_frame_equal(result, df)
+
+        result = df.iloc[:, 1:2]
+        tm.assert_frame_equal(result, df.iloc[:, :0])
+
+        result = df.iloc[:, -1:]
+        tm.assert_frame_equal(result, df)
+
+    def test_loc_series(self, data):
+        ser = pd.Series(data)
+        result = ser.loc[:3]
+        expected = pd.Series(data[:4])
+        tm.assert_series_equal(result, expected)
+
+        result = ser.loc[[0, 1, 2, 3]]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_frame(self, data):
+        df = pd.DataFrame({"A": data, "B": np.arange(len(data), dtype="int64")})
+        expected = pd.DataFrame({"A": data[:4]})
+
+        # slice -> frame
+        result = df.loc[:3, ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+        # sequence -> frame
+        result = df.loc[[0, 1, 2, 3], ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = pd.Series(data[:4], name="A")
+
+        # slice -> series
+        result = df.loc[:3, "A"]
+        tm.assert_series_equal(result, expected)
+
+        # sequence -> series
+        result = df.loc[:3, "A"]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_iloc_frame_single_dtype(self, data):
+        # GH#27110 bug in ExtensionBlock.iget caused df.iloc[n] to incorrectly
+        #  return a scalar
+        df = pd.DataFrame({"A": data})
+        expected = pd.Series([data[2]], index=["A"], name=2, dtype=data.dtype)
+
+        result = df.loc[2]
+        tm.assert_series_equal(result, expected)
+
+        expected = pd.Series(
+            [data[-1]], index=["A"], name=len(data) - 1, dtype=data.dtype
+        )
+        result = df.iloc[-1]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_scalar(self, data):
+        result = data[0]
+        assert isinstance(result, data.dtype.type)
+
+        result = pd.Series(data)[0]
+        assert isinstance(result, data.dtype.type)
+
+    def test_getitem_invalid(self, data):
+        # TODO: box over scalar, [scalar], (scalar,)?
+
+        msg = (
+            r"only integers, slices \(`:`\), ellipsis \(`...`\), numpy.newaxis "
+            r"\(`None`\) and integer or boolean arrays are valid indices"
+        )
+        with pytest.raises(IndexError, match=msg):
+            data["foo"]
+        with pytest.raises(IndexError, match=msg):
+            data[2.5]
+
+        ub = len(data)
+        msg = "|".join(
+            [
+                "list index out of range",  # json
+                "index out of bounds",  # pyarrow
+                "Out of bounds access",  # Sparse
+                f"loc must be an integer between -{ub} and {ub}",  # Sparse
+                f"index {ub + 1} is out of bounds for axis 0 with size {ub}",
+                f"index -{ub + 1} is out of bounds for axis 0 with size {ub}",
+            ]
+        )
+        with pytest.raises(IndexError, match=msg):
+            data[ub + 1]
+        with pytest.raises(IndexError, match=msg):
+            data[-ub - 1]
+
+    def test_getitem_scalar_na(self, data_missing, na_cmp, na_value):
+        result = data_missing[0]
+        assert na_cmp(result, na_value)
+
+    def test_getitem_empty(self, data):
+        # Indexing with empty list
+        result = data[[]]
+        assert len(result) == 0
+        assert isinstance(result, type(data))
+
+        expected = data[np.array([], dtype="int64")]
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_getitem_mask(self, data):
+        # Empty mask, raw array
+        mask = np.zeros(len(data), dtype=bool)
+        result = data[mask]
+        assert len(result) == 0
+        assert isinstance(result, type(data))
+
+        # Empty mask, in series
+        mask = np.zeros(len(data), dtype=bool)
+        result = pd.Series(data)[mask]
+        assert len(result) == 0
+        assert result.dtype == data.dtype
+
+        # non-empty mask, raw array
+        mask[0] = True
+        result = data[mask]
+        assert len(result) == 1
+        assert isinstance(result, type(data))
+
+        # non-empty mask, in series
+        result = pd.Series(data)[mask]
+        assert len(result) == 1
+        assert result.dtype == data.dtype
+
+    def test_getitem_mask_raises(self, data):
+        mask = np.array([True, False])
+        msg = f"Boolean index has wrong length: 2 instead of {len(data)}"
+        with pytest.raises(IndexError, match=msg):
+            data[mask]
+
+        mask = pd.array(mask, dtype="boolean")
+        with pytest.raises(IndexError, match=msg):
+            data[mask]
+
+    def test_getitem_boolean_array_mask(self, data):
+        mask = pd.array(np.zeros(data.shape, dtype="bool"), dtype="boolean")
+        result = data[mask]
+        assert len(result) == 0
+        assert isinstance(result, type(data))
+
+        result = pd.Series(data)[mask]
+        assert len(result) == 0
+        assert result.dtype == data.dtype
+
+        mask[:5] = True
+        expected = data.take([0, 1, 2, 3, 4])
+        result = data[mask]
+        tm.assert_extension_array_equal(result, expected)
+
+        expected = pd.Series(expected)
+        result = pd.Series(data)[mask]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_boolean_na_treated_as_false(self, data):
+        # https://github.com/pandas-dev/pandas/issues/31503
+        mask = pd.array(np.zeros(data.shape, dtype="bool"), dtype="boolean")
+        mask[:2] = pd.NA
+        mask[2:4] = True
+
+        result = data[mask]
+        expected = data[mask.fillna(False)]
+
+        tm.assert_extension_array_equal(result, expected)
+
+        s = pd.Series(data)
+
+        result = s[mask]
+        expected = s[mask.fillna(False)]
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2], pd.array([0, 1, 2], dtype="Int64"), np.array([0, 1, 2])],
+        ids=["list", "integer-array", "numpy-array"],
+    )
+    def test_getitem_integer_array(self, data, idx):
+        result = data[idx]
+        assert len(result) == 3
+        assert isinstance(result, type(data))
+        expected = data.take([0, 1, 2])
+        tm.assert_extension_array_equal(result, expected)
+
+        expected = pd.Series(expected)
+        result = pd.Series(data)[idx]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2, pd.NA], pd.array([0, 1, 2, pd.NA], dtype="Int64")],
+        ids=["list", "integer-array"],
+    )
+    def test_getitem_integer_with_missing_raises(self, data, idx):
+        msg = "Cannot index with an integer indexer containing NA values"
+        with pytest.raises(ValueError, match=msg):
+            data[idx]
+
+    @pytest.mark.xfail(
+        reason="Tries label-based and raises KeyError; "
+        "in some cases raises when calling np.asarray"
+    )
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2, pd.NA], pd.array([0, 1, 2, pd.NA], dtype="Int64")],
+        ids=["list", "integer-array"],
+    )
+    def test_getitem_series_integer_with_missing_raises(self, data, idx):
+        msg = "Cannot index with an integer indexer containing NA values"
+        # TODO: this raises KeyError about labels not found (it tries label-based)
+
+        ser = pd.Series(data, index=[chr(100 + i) for i in range(len(data))])
+        with pytest.raises(ValueError, match=msg):
+            ser[idx]
+
+    def test_getitem_slice(self, data):
+        # getitem[slice] should return an array
+        result = data[slice(0)]  # empty
+        assert isinstance(result, type(data))
+
+        result = data[slice(1)]  # scalar
+        assert isinstance(result, type(data))
+
+    def test_getitem_ellipsis_and_slice(self, data):
+        # GH#40353 this is called from slice_block_rows
+        result = data[..., :]
+        tm.assert_extension_array_equal(result, data)
+
+        result = data[:, ...]
+        tm.assert_extension_array_equal(result, data)
+
+        result = data[..., :3]
+        tm.assert_extension_array_equal(result, data[:3])
+
+        result = data[:3, ...]
+        tm.assert_extension_array_equal(result, data[:3])
+
+        result = data[..., ::2]
+        tm.assert_extension_array_equal(result, data[::2])
+
+        result = data[::2, ...]
+        tm.assert_extension_array_equal(result, data[::2])
+
+    def test_get(self, data):
+        # GH 20882
+        s = pd.Series(data, index=[2 * i for i in range(len(data))])
+        assert s.get(4) == s.iloc[2]
+
+        result = s.get([4, 6])
+        expected = s.iloc[[2, 3]]
+        tm.assert_series_equal(result, expected)
+
+        result = s.get(slice(2))
+        expected = s.iloc[[0, 1]]
+        tm.assert_series_equal(result, expected)
+
+        assert s.get(-1) is None
+        assert s.get(s.index.max() + 1) is None
+
+        s = pd.Series(data[:6], index=list("abcdef"))
+        assert s.get("c") == s.iloc[2]
+
+        result = s.get(slice("b", "d"))
+        expected = s.iloc[[1, 2, 3]]
+        tm.assert_series_equal(result, expected)
+
+        result = s.get("Z")
+        assert result is None
+
+        # As of 3.0, getitem with int keys treats them as labels
+        assert s.get(4) is None
+        assert s.get(-1) is None
+        assert s.get(len(s)) is None
+
+        # GH 21257
+        s = pd.Series(data)
+        with tm.assert_produces_warning(None):
+            # GH#45324 make sure we aren't giving a spurious FutureWarning
+            s2 = s[::2]
+        assert s2.get(1) is None
+
+    def test_take_sequence(self, data):
+        result = pd.Series(data)[[0, 1, 3]]
+        assert result.iloc[0] == data[0]
+        assert result.iloc[1] == data[1]
+        assert result.iloc[2] == data[3]
+
+    def test_take(self, data, na_value, na_cmp):
+        result = data.take([0, -1])
+        assert result.dtype == data.dtype
+        assert result[0] == data[0]
+        assert result[1] == data[-1]
+
+        result = data.take([0, -1], allow_fill=True, fill_value=na_value)
+        assert result[0] == data[0]
+        assert na_cmp(result[1], na_value)
+
+        with pytest.raises(IndexError, match="out of bounds"):
+            data.take([len(data) + 1])
+
+    def test_take_empty(self, data, na_value, na_cmp):
+        empty = data[:0]
+
+        result = empty.take([-1], allow_fill=True)
+        assert na_cmp(result[0], na_value)
+
+        msg = "cannot do a non-empty take from an empty axes|out of bounds"
+
+        with pytest.raises(IndexError, match=msg):
+            empty.take([-1])
+
+        with pytest.raises(IndexError, match="cannot do a non-empty take"):
+            empty.take([0, 1])
+
+    def test_take_negative(self, data):
+        # https://github.com/pandas-dev/pandas/issues/20640
+        n = len(data)
+        result = data.take([0, -n, n - 1, -1])
+        expected = data.take([0, 0, n - 1, n - 1])
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_take_non_na_fill_value(self, data_missing):
+        fill_value = data_missing[1]  # valid
+        na = data_missing[0]
+
+        arr = data_missing._from_sequence(
+            [na, fill_value, na], dtype=data_missing.dtype
+        )
+        result = arr.take([-1, 1], fill_value=fill_value, allow_fill=True)
+        expected = arr.take([1, 1])
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_take_pandas_style_negative_raises(self, data, na_value):
+        with tm.external_error_raised(ValueError):
+            data.take([0, -2], fill_value=na_value, allow_fill=True)
+
+    @pytest.mark.parametrize("allow_fill", [True, False])
+    def test_take_out_of_bounds_raises(self, data, allow_fill):
+        arr = data[:3]
+
+        with pytest.raises(IndexError, match="out of bounds|out-of-bounds"):
+            arr.take(np.asarray([0, 3]), allow_fill=allow_fill)
+
+    def test_take_series(self, data):
+        s = pd.Series(data)
+        result = s.take([0, -1])
+        expected = pd.Series(
+            data._from_sequence([data[0], data[len(data) - 1]], dtype=s.dtype),
+            index=range(0, 198, 99),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_reindex(self, data, na_value):
+        s = pd.Series(data)
+        result = s.reindex([0, 1, 3])
+        expected = pd.Series(data.take([0, 1, 3]), index=[0, 1, 3])
+        tm.assert_series_equal(result, expected)
+
+        n = len(data)
+        result = s.reindex([-1, 0, n])
+        expected = pd.Series(
+            data._from_sequence([na_value, data[0], na_value], dtype=s.dtype),
+            index=[-1, 0, n],
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = s.reindex([n, n + 1])
+        expected = pd.Series(
+            data._from_sequence([na_value, na_value], dtype=s.dtype),
+            index=range(n, n + 2, 1),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_reindex_non_na_fill_value(self, data_missing):
+        valid = data_missing[1]
+        na = data_missing[0]
+
+        arr = data_missing._from_sequence([na, valid], dtype=data_missing.dtype)
+        ser = pd.Series(arr)
+        result = ser.reindex([0, 1, 2], fill_value=valid)
+        expected = pd.Series(
+            data_missing._from_sequence([na, valid, valid], dtype=data_missing.dtype)
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_len1(self, data):
+        # see GH-27785 take_nd with indexer of len 1 resulting in wrong ndim
+        df = pd.DataFrame({"A": data})
+        res = df.loc[[0], "A"]
+        assert res.ndim == 1
+        assert res._mgr.blocks[0].ndim == 1
+        if hasattr(res._mgr, "blocks"):
+            assert res._mgr._block.ndim == 1
+
+    def test_item(self, data):
+        # https://github.com/pandas-dev/pandas/pull/30175
+        s = pd.Series(data)
+        result = s[:1].item()
+        assert result == data[0]
+
+        msg = "can only convert an array of size 1 to a Python scalar"
+        with pytest.raises(ValueError, match=msg):
+            s[:0].item()
+
+        with pytest.raises(ValueError, match=msg):
+            s.item()
diff --git a/pandas/pandas/tests/extension/base/groupby.py b/pandas/pandas/tests/extension/base/groupby.py
new file mode 100644
index 0000000000000..60cade97ab528
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/groupby.py
@@ -0,0 +1,172 @@
+import re
+
+import pytest
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_numeric_dtype,
+    is_object_dtype,
+    is_string_dtype,
+)
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.filterwarnings(
+    "ignore:The default of observed=False is deprecated:FutureWarning"
+)
+class BaseGroupbyTests:
+    """Groupby-specific tests."""
+
+    def test_grouping_grouper(self, data_for_grouping):
+        df = pd.DataFrame(
+            {
+                "A": pd.Series(
+                    ["B", "B", None, None, "A", "A", "B", "C"], dtype=object
+                ),
+                "B": data_for_grouping,
+            }
+        )
+        gr1 = df.groupby("A")._grouper.groupings[0]
+        gr2 = df.groupby("B")._grouper.groupings[0]
+
+        tm.assert_numpy_array_equal(gr1.grouping_vector, df.A.values)
+        tm.assert_extension_array_equal(gr2.grouping_vector, data_for_grouping)
+
+    @pytest.mark.parametrize("as_index", [True, False])
+    def test_groupby_extension_agg(self, as_index, data_for_grouping):
+        df = pd.DataFrame({"A": [1, 1, 2, 2, 3, 3, 1, 4], "B": data_for_grouping})
+
+        is_bool = data_for_grouping.dtype._is_boolean
+        if is_bool:
+            # only 2 unique values, and the final entry has c==b
+            #  (see data_for_grouping docstring)
+            df = df.iloc[:-1]
+
+        result = df.groupby("B", as_index=as_index).A.mean()
+        _, uniques = pd.factorize(data_for_grouping, sort=True)
+
+        exp_vals = [3.0, 1.0, 4.0]
+        if is_bool:
+            exp_vals = exp_vals[:-1]
+        if as_index:
+            index = pd.Index(uniques, name="B")
+            expected = pd.Series(exp_vals, index=index, name="A")
+            tm.assert_series_equal(result, expected)
+        else:
+            expected = pd.DataFrame({"B": uniques, "A": exp_vals})
+            tm.assert_frame_equal(result, expected)
+
+    def test_groupby_agg_extension(self, data_for_grouping):
+        # GH#38980 groupby agg on extension type fails for non-numeric types
+        df = pd.DataFrame({"A": [1, 1, 2, 2, 3, 3, 1, 4], "B": data_for_grouping})
+
+        expected = df.iloc[[0, 2, 4, 7]]
+        expected = expected.set_index("A")
+
+        result = df.groupby("A").agg({"B": "first"})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.groupby("A").agg("first")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.groupby("A").first()
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_extension_no_sort(self, data_for_grouping):
+        df = pd.DataFrame({"A": [1, 1, 2, 2, 3, 3, 1, 4], "B": data_for_grouping})
+
+        is_bool = data_for_grouping.dtype._is_boolean
+        if is_bool:
+            # only 2 unique values, and the final entry has c==b
+            #  (see data_for_grouping docstring)
+            df = df.iloc[:-1]
+
+        result = df.groupby("B", sort=False).A.mean()
+        _, index = pd.factorize(data_for_grouping, sort=False)
+
+        index = pd.Index(index, name="B")
+        exp_vals = [1.0, 3.0, 4.0]
+        if is_bool:
+            exp_vals = exp_vals[:-1]
+        expected = pd.Series(exp_vals, index=index, name="A")
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_extension_transform(self, data_for_grouping):
+        is_bool = data_for_grouping.dtype._is_boolean
+
+        valid = data_for_grouping[~data_for_grouping.isna()]
+        df = pd.DataFrame({"A": [1, 1, 3, 3, 1, 4], "B": valid})
+        is_bool = data_for_grouping.dtype._is_boolean
+        if is_bool:
+            # only 2 unique values, and the final entry has c==b
+            #  (see data_for_grouping docstring)
+            df = df.iloc[:-1]
+
+        result = df.groupby("B").A.transform(len)
+        expected = pd.Series([3, 3, 2, 2, 3, 1], name="A")
+        if is_bool:
+            expected = expected[:-1]
+
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_extension_apply(self, data_for_grouping, groupby_apply_op):
+        df = pd.DataFrame({"A": [1, 1, 2, 2, 3, 3, 1, 4], "B": data_for_grouping})
+        df.groupby("B", group_keys=False, observed=False).apply(groupby_apply_op)
+        df.groupby("B", group_keys=False, observed=False).A.apply(groupby_apply_op)
+        df.groupby("A", group_keys=False, observed=False).apply(groupby_apply_op)
+        df.groupby("A", group_keys=False, observed=False).B.apply(groupby_apply_op)
+
+    def test_groupby_apply_identity(self, data_for_grouping):
+        df = pd.DataFrame({"A": [1, 1, 2, 2, 3, 3, 1, 4], "B": data_for_grouping})
+        result = df.groupby("A").B.apply(lambda x: x.array)
+        expected = pd.Series(
+            [
+                df.B.iloc[[0, 1, 6]].array,
+                df.B.iloc[[2, 3]].array,
+                df.B.iloc[[4, 5]].array,
+                df.B.iloc[[7]].array,
+            ],
+            index=pd.Index([1, 2, 3, 4], name="A"),
+            name="B",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_in_numeric_groupby(self, data_for_grouping):
+        df = pd.DataFrame(
+            {
+                "A": [1, 1, 2, 2, 3, 3, 1, 4],
+                "B": data_for_grouping,
+                "C": [1, 1, 1, 1, 1, 1, 1, 1],
+            }
+        )
+
+        dtype = data_for_grouping.dtype
+        if (
+            is_numeric_dtype(dtype)
+            or is_bool_dtype(dtype)
+            or dtype.name == "decimal"
+            or is_string_dtype(dtype)
+            or is_object_dtype(dtype)
+            or dtype.kind == "m"  # in particular duration[*][pyarrow]
+        ):
+            expected = pd.Index(["B", "C"])
+            result = df.groupby("A").sum().columns
+        else:
+            expected = pd.Index(["C"])
+
+            msg = "|".join(
+                [
+                    # period
+                    "does not support sum operations",
+                    # datetime
+                    "does not support operation 'sum'",
+                    # all others
+                    re.escape(f"agg function failed [how->sum,dtype->{dtype}"),
+                ]
+            )
+            with pytest.raises(TypeError, match=msg):
+                df.groupby("A").sum()
+            result = df.groupby("A").sum(numeric_only=True).columns
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/extension/base/index.py b/pandas/pandas/tests/extension/base/index.py
new file mode 100644
index 0000000000000..e7bfebec92287
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/index.py
@@ -0,0 +1,20 @@
+"""
+Tests for Indexes backed by arbitrary ExtensionArrays.
+"""
+
+import pandas as pd
+
+
+class BaseIndexTests:
+    """Tests for Index object backed by an ExtensionArray"""
+
+    def test_index_from_array(self, data):
+        idx = pd.Index(data)
+        assert data.dtype == idx.dtype
+
+    def test_index_from_listlike_with_dtype(self, data):
+        idx = pd.Index(data, dtype=data.dtype)
+        assert idx.dtype == data.dtype
+
+        idx = pd.Index(list(data), dtype=data.dtype)
+        assert idx.dtype == data.dtype
diff --git a/pandas/pandas/tests/extension/base/interface.py b/pandas/pandas/tests/extension/base/interface.py
new file mode 100644
index 0000000000000..79eb64b5a654f
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/interface.py
@@ -0,0 +1,158 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+from pandas.core.dtypes.common import is_extension_array_dtype
+from pandas.core.dtypes.dtypes import ExtensionDtype
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseInterfaceTests:
+    """Tests that the basic interface is satisfied."""
+
+    # ------------------------------------------------------------------------
+    # Interface
+    # ------------------------------------------------------------------------
+
+    def test_len(self, data):
+        assert len(data) == 100
+
+    def test_size(self, data):
+        assert data.size == 100
+
+    def test_ndim(self, data):
+        assert data.ndim == 1
+
+    def test_can_hold_na_valid(self, data):
+        # GH-20761
+        assert data._can_hold_na is True
+
+    def test_contains(self, data, data_missing):
+        # GH-37867
+        # Tests for membership checks. Membership checks for nan-likes is tricky and
+        # the settled on rule is: `nan_like in arr` is True if nan_like is
+        # arr.dtype.na_value and arr.isna().any() is True. Else the check returns False.
+
+        na_value = data.dtype.na_value
+        # ensure data without missing values
+        data = data[~data.isna()]
+
+        # first elements are non-missing
+        assert data[0] in data
+        assert data_missing[0] in data_missing
+
+        # check the presence of na_value
+        assert na_value in data_missing
+        assert na_value not in data
+
+        # the data can never contain other nan-likes than na_value
+        for na_value_obj in tm.NULL_OBJECTS:
+            if na_value_obj is na_value or type(na_value_obj) == type(na_value):
+                # type check for e.g. two instances of Decimal("NAN")
+                continue
+            assert na_value_obj not in data
+            assert na_value_obj not in data_missing
+
+    def test_memory_usage(self, data):
+        s = pd.Series(data)
+        result = s.memory_usage(index=False)
+        assert result == s.nbytes
+
+    def test_array_interface(self, data):
+        result = np.array(data)
+        assert result[0] == data[0]
+
+        result = np.array(data, dtype=object)
+        expected = np.array(list(data), dtype=object)
+        if expected.ndim > 1:
+            # nested data, explicitly construct as 1D
+            expected = construct_1d_object_array_from_listlike(list(data))
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_array_interface_copy(self, data):
+        result_copy1 = np.array(data, copy=True)
+        result_copy2 = np.array(data, copy=True)
+        assert not np.may_share_memory(result_copy1, result_copy2)
+
+        if not np_version_gt2:
+            # copy=False semantics are only supported in NumPy>=2.
+            return
+
+        try:
+            result_nocopy1 = np.array(data, copy=False)
+        except ValueError:
+            # An error is always acceptable for `copy=False`
+            return
+
+        result_nocopy2 = np.array(data, copy=False)
+        # If copy=False was given and did not raise, these must share the same data
+        assert np.may_share_memory(result_nocopy1, result_nocopy2)
+
+    def test_is_extension_array_dtype(self, data):
+        assert is_extension_array_dtype(data)
+        assert is_extension_array_dtype(data.dtype)
+        assert is_extension_array_dtype(pd.Series(data))
+        assert isinstance(data.dtype, ExtensionDtype)
+
+    def test_no_values_attribute(self, data):
+        # GH-20735: EA's with .values attribute give problems with internal
+        # code, disallowing this for now until solved
+        assert not hasattr(data, "values")
+        assert not hasattr(data, "_values")
+
+    def test_is_numeric_honored(self, data):
+        result = pd.Series(data)
+        if hasattr(result._mgr, "blocks"):
+            assert result._mgr.blocks[0].is_numeric is data.dtype._is_numeric
+
+    def test_isna_extension_array(self, data_missing):
+        # If your `isna` returns an ExtensionArray, you must also implement
+        # _reduce. At the *very* least, you must implement any and all
+        na = data_missing.isna()
+        if is_extension_array_dtype(na):
+            assert na._reduce("any")
+            assert na.any()
+
+            assert not na._reduce("all")
+            assert not na.all()
+
+            assert na.dtype._is_boolean
+
+    def test_copy(self, data):
+        # GH#27083 removing deep keyword from EA.copy
+        assert data[0] != data[1]
+        result = data.copy()
+
+        if data.dtype._is_immutable:
+            pytest.skip(f"test_copy assumes mutability and {data.dtype} is immutable")
+
+        data[1] = data[0]
+        assert result[1] != result[0]
+
+    def test_view(self, data):
+        # view with no dtype should return a shallow copy, *not* the same
+        #  object
+        assert data[1] != data[0]
+
+        result = data.view()
+        assert result is not data
+        assert type(result) == type(data)
+
+        if data.dtype._is_immutable:
+            pytest.skip(f"test_view assumes mutability and {data.dtype} is immutable")
+
+        result[1] = result[0]
+        assert data[1] == data[0]
+
+        # check specifically that the `dtype` kwarg is accepted
+        data.view(dtype=None)
+
+    def test_tolist(self, data):
+        result = data.tolist()
+        expected = list(data)
+        assert isinstance(result, list)
+        assert result == expected
diff --git a/pandas/pandas/tests/extension/base/io.py b/pandas/pandas/tests/extension/base/io.py
new file mode 100644
index 0000000000000..f7367fceeb52f
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/io.py
@@ -0,0 +1,39 @@
+from io import StringIO
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import ExtensionArray
+
+
+class BaseParsingTests:
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_EA_types(self, engine, data, request):
+        if isinstance(data.dtype, pd.CategoricalDtype):
+            # in parsers.pyx _convert_with_dtype there is special-casing for
+            #  Categorical that preempts _from_sequence_of_strings
+            pass
+        elif isinstance(data.dtype, pd.core.dtypes.dtypes.NumpyEADtype):
+            # These get unwrapped internally so are treated as numpy dtypes
+            #  in the parsers.pyx code
+            pass
+        elif (
+            type(data)._from_sequence_of_strings.__func__
+            is ExtensionArray._from_sequence_of_strings.__func__
+        ):
+            # i.e. the EA hasn't overridden _from_sequence_of_strings
+            mark = pytest.mark.xfail(
+                reason="_from_sequence_of_strings not implemented",
+                raises=NotImplementedError,
+            )
+            request.node.add_marker(mark)
+
+        df = pd.DataFrame({"with_dtype": pd.Series(data, dtype=str(data.dtype))})
+        csv_output = df.to_csv(index=False, na_rep=np.nan)
+        result = pd.read_csv(
+            StringIO(csv_output), dtype={"with_dtype": str(data.dtype)}, engine=engine
+        )
+        expected = df
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/extension/base/methods.py b/pandas/pandas/tests/extension/base/methods.py
new file mode 100644
index 0000000000000..fd9fec0cb490c
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/methods.py
@@ -0,0 +1,728 @@
+import inspect
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._typing import Dtype
+
+from pandas.core.dtypes.common import is_bool_dtype
+from pandas.core.dtypes.dtypes import NumpyEADtype
+from pandas.core.dtypes.missing import na_value_for_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.sorting import nargsort
+
+
+class BaseMethodsTests:
+    """Various Series and DataFrame methods."""
+
+    def test_hash_pandas_object(self, data):
+        # _hash_pandas_object should return a uint64 ndarray of the same length
+        # as the data
+        from pandas.core.util.hashing import _default_hash_key
+
+        res = data._hash_pandas_object(
+            encoding="utf-8", hash_key=_default_hash_key, categorize=False
+        )
+        assert res.dtype == np.uint64
+        assert res.shape == data.shape
+
+    def test_value_counts_default_dropna(self, data):
+        # make sure we have consistent default dropna kwarg
+        if not hasattr(data, "value_counts"):
+            pytest.skip(f"value_counts is not implemented for {type(data)}")
+        sig = inspect.signature(data.value_counts)
+        kwarg = sig.parameters["dropna"]
+        assert kwarg.default is True
+
+    @pytest.mark.parametrize("dropna", [True, False])
+    def test_value_counts(self, all_data, dropna):
+        all_data = all_data[:10]
+        if dropna:
+            other = all_data[~all_data.isna()]
+        else:
+            other = all_data
+
+        result = pd.Series(all_data).value_counts(dropna=dropna).sort_index()
+        expected = pd.Series(other).value_counts(dropna=dropna).sort_index()
+
+        tm.assert_series_equal(result, expected)
+
+    def test_value_counts_with_normalize(self, data):
+        # GH 33172
+        data = data[:10].unique()
+        values = np.array(data[~data.isna()])
+        ser = pd.Series(data, dtype=data.dtype)
+
+        result = ser.value_counts(normalize=True).sort_index()
+
+        if not isinstance(data, pd.Categorical):
+            expected = pd.Series(
+                [1 / len(values)] * len(values), index=result.index, name="proportion"
+            )
+        else:
+            expected = pd.Series(0.0, index=result.index, name="proportion")
+            expected[result > 0] = 1 / len(values)
+
+        if isinstance(data.dtype, pd.StringDtype) and data.dtype.na_value is np.nan:
+            # TODO: avoid special-casing
+            expected = expected.astype("float64")
+        elif getattr(data.dtype, "storage", "") == "pyarrow" or isinstance(
+            data.dtype, pd.ArrowDtype
+        ):
+            # TODO: avoid special-casing
+            expected = expected.astype("double[pyarrow]")
+        elif na_value_for_dtype(data.dtype) is pd.NA:
+            # TODO(GH#44692): avoid special-casing
+            expected = expected.astype("Float64")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_count(self, data_missing):
+        df = pd.DataFrame({"A": data_missing})
+        result = df.count(axis="columns")
+        expected = pd.Series([0, 1])
+        tm.assert_series_equal(result, expected)
+
+    def test_series_count(self, data_missing):
+        # GH#26835
+        ser = pd.Series(data_missing)
+        result = ser.count()
+        expected = 1
+        assert result == expected
+
+    def test_apply_simple_series(self, data):
+        result = pd.Series(data).apply(id)
+        assert isinstance(result, pd.Series)
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data_missing, na_action):
+        result = data_missing.map(lambda x: x, na_action=na_action)
+        expected = data_missing.to_numpy()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_argsort(self, data_for_sorting):
+        result = pd.Series(data_for_sorting).argsort()
+        # argsort result gets passed to take, so should be np.intp
+        expected = pd.Series(np.array([2, 0, 1], dtype=np.intp))
+        tm.assert_series_equal(result, expected)
+
+    def test_argsort_missing_array(self, data_missing_for_sorting):
+        result = data_missing_for_sorting.argsort()
+        # argsort result gets passed to take, so should be np.intp
+        expected = np.array([2, 0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_argsort_missing(self, data_missing_for_sorting):
+        result = pd.Series(data_missing_for_sorting).argsort()
+        expected = pd.Series(np.array([2, 0, 1], dtype=np.intp))
+        tm.assert_series_equal(result, expected)
+
+    def test_argmin_argmax(self, data_for_sorting, data_missing_for_sorting, na_value):
+        # GH 24382
+        is_bool = data_for_sorting.dtype._is_boolean
+
+        exp_argmax = 1
+        exp_argmax_repeated = 3
+        if is_bool:
+            # See data_for_sorting docstring
+            exp_argmax = 0
+            exp_argmax_repeated = 1
+
+        # data_for_sorting -> [B, C, A] with A < B < C
+        assert data_for_sorting.argmax() == exp_argmax
+        assert data_for_sorting.argmin() == 2
+
+        # with repeated values -> first occurrence
+        data = data_for_sorting.take([2, 0, 0, 1, 1, 2])
+        assert data.argmax() == exp_argmax_repeated
+        assert data.argmin() == 0
+
+        # with missing values
+        # data_missing_for_sorting -> [B, NA, A] with A < B and NA missing.
+        assert data_missing_for_sorting.argmax() == 0
+        assert data_missing_for_sorting.argmin() == 2
+
+    @pytest.mark.parametrize("method", ["argmax", "argmin"])
+    def test_argmin_argmax_empty_array(self, method, data):
+        # GH 24382
+        err_msg = "attempt to get"
+        with pytest.raises(ValueError, match=err_msg):
+            getattr(data[:0], method)()
+
+    @pytest.mark.parametrize("method", ["argmax", "argmin"])
+    def test_argmin_argmax_all_na(self, method, data, na_value):
+        # all missing with skipna=True is the same as empty
+        err_msg = "attempt to get"
+        data_na = type(data)._from_sequence([na_value, na_value], dtype=data.dtype)
+        with pytest.raises(ValueError, match=err_msg):
+            getattr(data_na, method)()
+
+    @pytest.mark.parametrize(
+        "op_name, skipna, expected",
+        [
+            ("idxmax", True, 0),
+            ("idxmin", True, 2),
+            ("argmax", True, 0),
+            ("argmin", True, 2),
+            ("idxmax", False, -1),
+            ("idxmin", False, -1),
+            ("argmax", False, -1),
+            ("argmin", False, -1),
+        ],
+    )
+    def test_argreduce_series(
+        self, data_missing_for_sorting, op_name, skipna, expected
+    ):
+        # data_missing_for_sorting -> [B, NA, A] with A < B and NA missing.
+        ser = pd.Series(data_missing_for_sorting)
+        if expected == -1:
+            with pytest.raises(ValueError, match="Encountered an NA value"):
+                getattr(ser, op_name)(skipna=skipna)
+        else:
+            result = getattr(ser, op_name)(skipna=skipna)
+            tm.assert_almost_equal(result, expected)
+
+    def test_argmax_argmin_no_skipna_notimplemented(self, data_missing_for_sorting):
+        # GH#38733
+        data = data_missing_for_sorting
+
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            data.argmin(skipna=False)
+
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            data.argmax(skipna=False)
+
+    @pytest.mark.parametrize(
+        "na_position, expected",
+        [
+            ("last", np.array([2, 0, 1], dtype=np.dtype("intp"))),
+            ("first", np.array([1, 2, 0], dtype=np.dtype("intp"))),
+        ],
+    )
+    def test_nargsort(self, data_missing_for_sorting, na_position, expected):
+        # GH 25439
+        result = nargsort(data_missing_for_sorting, na_position=na_position)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("ascending", [True, False])
+    def test_sort_values(self, data_for_sorting, ascending, sort_by_key):
+        ser = pd.Series(data_for_sorting)
+        result = ser.sort_values(ascending=ascending, key=sort_by_key)
+        expected = ser.iloc[[2, 0, 1]]
+        if not ascending:
+            # GH 35922. Expect stable sort
+            if ser.nunique() == 2:
+                expected = ser.iloc[[0, 1, 2]]
+            else:
+                expected = ser.iloc[[1, 0, 2]]
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("ascending", [True, False])
+    def test_sort_values_missing(
+        self, data_missing_for_sorting, ascending, sort_by_key
+    ):
+        ser = pd.Series(data_missing_for_sorting)
+        result = ser.sort_values(ascending=ascending, key=sort_by_key)
+        if ascending:
+            expected = ser.iloc[[2, 0, 1]]
+        else:
+            expected = ser.iloc[[0, 2, 1]]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("ascending", [True, False])
+    def test_sort_values_frame(self, data_for_sorting, ascending):
+        df = pd.DataFrame({"A": [1, 2, 1], "B": data_for_sorting})
+        result = df.sort_values(["A", "B"])
+        expected = pd.DataFrame(
+            {"A": [1, 1, 2], "B": data_for_sorting.take([2, 0, 1])}, index=[2, 0, 1]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("keep", ["first", "last", False])
+    def test_duplicated(self, data, keep):
+        arr = data.take([0, 1, 0, 1])
+        result = arr.duplicated(keep=keep)
+        if keep == "first":
+            expected = np.array([False, False, True, True])
+        elif keep == "last":
+            expected = np.array([True, True, False, False])
+        else:
+            expected = np.array([True, True, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [pd.Series, lambda x: x])
+    @pytest.mark.parametrize("method", [lambda x: x.unique(), pd.unique])
+    def test_unique(self, data, box, method):
+        duplicated = box(data._from_sequence([data[0], data[0]], dtype=data.dtype))
+
+        result = method(duplicated)
+
+        assert len(result) == 1
+        assert isinstance(result, type(data))
+        assert result[0] == duplicated[0]
+
+    def test_factorize(self, data_for_grouping):
+        codes, uniques = pd.factorize(data_for_grouping, use_na_sentinel=True)
+
+        is_bool = data_for_grouping.dtype._is_boolean
+        if is_bool:
+            # only 2 unique values
+            expected_codes = np.array([0, 0, -1, -1, 1, 1, 0, 0], dtype=np.intp)
+            expected_uniques = data_for_grouping.take([0, 4])
+        else:
+            expected_codes = np.array([0, 0, -1, -1, 1, 1, 0, 2], dtype=np.intp)
+            expected_uniques = data_for_grouping.take([0, 4, 7])
+
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_extension_array_equal(uniques, expected_uniques)
+
+    def test_factorize_equivalence(self, data_for_grouping):
+        codes_1, uniques_1 = pd.factorize(data_for_grouping, use_na_sentinel=True)
+        codes_2, uniques_2 = data_for_grouping.factorize(use_na_sentinel=True)
+
+        tm.assert_numpy_array_equal(codes_1, codes_2)
+        tm.assert_extension_array_equal(uniques_1, uniques_2)
+        assert len(uniques_1) == len(pd.unique(uniques_1))
+        assert uniques_1.dtype == data_for_grouping.dtype
+
+    def test_factorize_empty(self, data):
+        codes, uniques = pd.factorize(data[:0])
+        expected_codes = np.array([], dtype=np.intp)
+        expected_uniques = type(data)._from_sequence([], dtype=data[:0].dtype)
+
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_extension_array_equal(uniques, expected_uniques)
+
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        df = pd.DataFrame({"A": data_missing.take([0, 1, 0, 1])})
+        expected = pd.DataFrame({"A": data_missing.take([1, 1, 0, 1])})
+        result = df.fillna(value=data_missing[1], limit=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        ser = pd.Series(data_missing.take([0, 1, 0, 1]))
+        expected = pd.Series(data_missing.take([1, 1, 0, 1]))
+        result = ser.fillna(value=data_missing[1], limit=1)
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_copy_frame(self, data_missing):
+        arr = data_missing.take([1, 1])
+        df = pd.DataFrame({"A": arr})
+        df_orig = df.copy()
+
+        filled_val = df.iloc[0, 0]
+        result = df.fillna(filled_val)
+
+        result.iloc[0, 0] = filled_val
+
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_fillna_copy_series(self, data_missing):
+        arr = data_missing.take([1, 1])
+        ser = pd.Series(arr, copy=False)
+        ser_orig = ser.copy()
+
+        filled_val = ser[0]
+        result = ser.fillna(filled_val)
+        result.iloc[0] = filled_val
+
+        tm.assert_series_equal(ser, ser_orig)
+
+    def test_fillna_length_mismatch(self, data_missing):
+        msg = "Length of 'value' does not match."
+        with pytest.raises(ValueError, match=msg):
+            data_missing.fillna(data_missing.take([1]))
+
+    # Subclasses can override if we expect e.g Sparse[bool], boolean, pyarrow[bool]
+    _combine_le_expected_dtype: Dtype = NumpyEADtype("bool")
+
+    def test_combine_le(self, data_repeated):
+        # GH 20825
+        # Test that combine works when doing a <= (le) comparison
+        orig_data1, orig_data2 = data_repeated(2)
+        s1 = pd.Series(orig_data1)
+        s2 = pd.Series(orig_data2)
+        result = s1.combine(s2, lambda x1, x2: x1 <= x2)
+        expected = pd.Series(
+            pd.array(
+                [a <= b for (a, b) in zip(list(orig_data1), list(orig_data2))],
+                dtype=self._combine_le_expected_dtype,
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+        val = s1.iloc[0]
+        result = s1.combine(val, lambda x1, x2: x1 <= x2)
+        expected = pd.Series(
+            pd.array(
+                [a <= val for a in list(orig_data1)],
+                dtype=self._combine_le_expected_dtype,
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_combine_add(self, data_repeated):
+        # GH 20825
+        orig_data1, orig_data2 = data_repeated(2)
+        s1 = pd.Series(orig_data1)
+        s2 = pd.Series(orig_data2)
+
+        # Check if the operation is supported pointwise for our scalars. If not,
+        #  we will expect Series.combine to raise as well.
+        try:
+            with np.errstate(over="ignore"):
+                expected = pd.Series(
+                    orig_data1._from_sequence(
+                        [a + b for (a, b) in zip(list(orig_data1), list(orig_data2))]
+                    )
+                )
+        except TypeError:
+            # If the operation is not supported pointwise for our scalars,
+            #  then Series.combine should also raise
+            with pytest.raises(TypeError):
+                s1.combine(s2, lambda x1, x2: x1 + x2)
+            return
+
+        result = s1.combine(s2, lambda x1, x2: x1 + x2)
+        tm.assert_series_equal(result, expected)
+
+        val = s1.iloc[0]
+        result = s1.combine(val, lambda x1, x2: x1 + x2)
+        expected = pd.Series(
+            orig_data1._from_sequence([a + val for a in list(orig_data1)])
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_combine_first(self, data):
+        # https://github.com/pandas-dev/pandas/issues/24147
+        a = pd.Series(data[:3])
+        b = pd.Series(data[2:5], index=[2, 3, 4])
+        result = a.combine_first(b)
+        expected = pd.Series(data[:5])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("frame", [True, False])
+    @pytest.mark.parametrize(
+        "periods, indices",
+        [(-2, [2, 3, 4, -1, -1]), (0, [0, 1, 2, 3, 4]), (2, [-1, -1, 0, 1, 2])],
+    )
+    def test_container_shift(self, data, frame, periods, indices):
+        # https://github.com/pandas-dev/pandas/issues/22386
+        subset = data[:5]
+        data = pd.Series(subset, name="A")
+        expected = pd.Series(subset.take(indices, allow_fill=True), name="A")
+
+        if frame:
+            result = data.to_frame(name="A").assign(B=1).shift(periods)
+            expected = pd.concat(
+                [expected, pd.Series([1] * 5, name="B").shift(periods)], axis=1
+            )
+            compare = tm.assert_frame_equal
+        else:
+            result = data.shift(periods)
+            compare = tm.assert_series_equal
+
+        compare(result, expected)
+
+    def test_shift_0_periods(self, data):
+        # GH#33856 shifting with periods=0 should return a copy, not same obj
+        result = data.shift(0)
+        assert data[0] != data[1]  # otherwise below is invalid
+        data[0] = data[1]
+        assert result[0] != result[1]  # i.e. not the same object/view
+
+    @pytest.mark.parametrize("periods", [1, -2])
+    def test_diff(self, data, periods):
+        data = data[:5]
+        if is_bool_dtype(data.dtype):
+            op = operator.xor
+        else:
+            op = operator.sub
+        try:
+            # does this array implement ops?
+            op(data, data)
+        except Exception:
+            pytest.skip(f"{type(data)} does not support diff")
+        s = pd.Series(data)
+        result = s.diff(periods)
+        expected = pd.Series(op(data, data.shift(periods)))
+        tm.assert_series_equal(result, expected)
+
+        df = pd.DataFrame({"A": data, "B": [1.0] * 5})
+        result = df.diff(periods)
+        if periods == 1:
+            b = [np.nan, 0, 0, 0, 0]
+        else:
+            b = [0, 0, 0, np.nan, np.nan]
+        expected = pd.DataFrame({"A": expected, "B": b})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "periods, indices",
+        [[-4, [-1, -1]], [-1, [1, -1]], [0, [0, 1]], [1, [-1, 0]], [4, [-1, -1]]],
+    )
+    def test_shift_non_empty_array(self, data, periods, indices):
+        # https://github.com/pandas-dev/pandas/issues/23911
+        subset = data[:2]
+        result = subset.shift(periods)
+        expected = subset.take(indices, allow_fill=True)
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize("periods", [-4, -1, 0, 1, 4])
+    def test_shift_empty_array(self, data, periods):
+        # https://github.com/pandas-dev/pandas/issues/23911
+        empty = data[:0]
+        result = empty.shift(periods)
+        expected = empty
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_shift_zero_copies(self, data):
+        # GH#31502
+        result = data.shift(0)
+        assert result is not data
+
+        result = data[:0].shift(2)
+        assert result is not data
+
+    def test_shift_fill_value(self, data):
+        arr = data[:4]
+        fill_value = data[0]
+        result = arr.shift(1, fill_value=fill_value)
+        expected = data.take([0, 0, 1, 2])
+        tm.assert_extension_array_equal(result, expected)
+
+        result = arr.shift(-2, fill_value=fill_value)
+        expected = data.take([2, 3, 0, 0])
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_not_hashable(self, data):
+        # We are in general mutable, so not hashable
+        with pytest.raises(TypeError, match="unhashable type"):
+            hash(data)
+
+    def test_hash_pandas_object_works(self, data, as_frame):
+        # https://github.com/pandas-dev/pandas/issues/23066
+        data = pd.Series(data)
+        if as_frame:
+            data = data.to_frame()
+        a = pd.util.hash_pandas_object(data)
+        b = pd.util.hash_pandas_object(data)
+        tm.assert_equal(a, b)
+
+    def test_searchsorted(self, data_for_sorting, as_series):
+        if data_for_sorting.dtype._is_boolean:
+            return self._test_searchsorted_bool_dtypes(data_for_sorting, as_series)
+
+        b, c, a = data_for_sorting
+        arr = data_for_sorting.take([2, 0, 1])  # to get [a, b, c]
+
+        if as_series:
+            arr = pd.Series(arr)
+        assert arr.searchsorted(a) == 0
+        assert arr.searchsorted(a, side="right") == 1
+
+        assert arr.searchsorted(b) == 1
+        assert arr.searchsorted(b, side="right") == 2
+
+        assert arr.searchsorted(c) == 2
+        assert arr.searchsorted(c, side="right") == 3
+
+        result = arr.searchsorted(arr.take([0, 2]))
+        expected = np.array([0, 2], dtype=np.intp)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        # sorter
+        sorter = np.array([1, 2, 0])
+        assert data_for_sorting.searchsorted(a, sorter=sorter) == 0
+
+    def _test_searchsorted_bool_dtypes(self, data_for_sorting, as_series):
+        # We call this from test_searchsorted in cases where we have a
+        #  boolean-like dtype. The non-bool test assumes we have more than 2
+        #  unique values.
+        dtype = data_for_sorting.dtype
+        data_for_sorting = pd.array([True, False], dtype=dtype)
+        b, a = data_for_sorting
+        arr = type(data_for_sorting)._from_sequence([a, b], dtype=dtype)
+
+        if as_series:
+            arr = pd.Series(arr)
+        assert arr.searchsorted(a) == 0
+        assert arr.searchsorted(a, side="right") == 1
+
+        assert arr.searchsorted(b) == 1
+        assert arr.searchsorted(b, side="right") == 2
+
+        result = arr.searchsorted(arr.take([0, 1]))
+        expected = np.array([0, 1], dtype=np.intp)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        # sorter
+        sorter = np.array([1, 0])
+        assert data_for_sorting.searchsorted(a, sorter=sorter) == 0
+
+    def test_where_series(self, data, na_value, as_frame):
+        assert data[0] != data[1]
+        cls = type(data)
+        a, b = data[:2]
+
+        orig = pd.Series(cls._from_sequence([a, a, b, b], dtype=data.dtype))
+        ser = orig.copy()
+        cond = np.array([True, True, False, False])
+
+        if as_frame:
+            ser = ser.to_frame(name="a")
+            cond = cond.reshape(-1, 1)
+
+        result = ser.where(cond)
+        expected = pd.Series(
+            cls._from_sequence([a, a, na_value, na_value], dtype=data.dtype)
+        )
+
+        if as_frame:
+            expected = expected.to_frame(name="a")
+        tm.assert_equal(result, expected)
+
+        ser.mask(~cond, inplace=True)
+        tm.assert_equal(ser, expected)
+
+        # array other
+        ser = orig.copy()
+        if as_frame:
+            ser = ser.to_frame(name="a")
+        cond = np.array([True, False, True, True])
+        other = cls._from_sequence([a, b, a, b], dtype=data.dtype)
+        if as_frame:
+            other = pd.DataFrame({"a": other})
+            cond = pd.DataFrame({"a": cond})
+        result = ser.where(cond, other)
+        expected = pd.Series(cls._from_sequence([a, b, b, b], dtype=data.dtype))
+        if as_frame:
+            expected = expected.to_frame(name="a")
+        tm.assert_equal(result, expected)
+
+        ser.mask(~cond, other, inplace=True)
+        tm.assert_equal(ser, expected)
+
+    @pytest.mark.parametrize("repeats", [0, 1, 2, [1, 2, 3]])
+    def test_repeat(self, data, repeats, as_series, use_numpy):
+        arr = type(data)._from_sequence(data[:3], dtype=data.dtype)
+        if as_series:
+            arr = pd.Series(arr)
+
+        result = np.repeat(arr, repeats) if use_numpy else arr.repeat(repeats)
+
+        repeats = [repeats] * 3 if isinstance(repeats, int) else repeats
+        expected = [x for x, n in zip(arr, repeats) for _ in range(n)]
+        expected = type(data)._from_sequence(expected, dtype=data.dtype)
+        if as_series:
+            expected = pd.Series(expected, index=arr.index.repeat(repeats))
+
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "repeats, kwargs, error, msg",
+        [
+            (2, {"axis": 1}, ValueError, "axis"),
+            (-1, {}, ValueError, "negative"),
+            ([1, 2], {}, ValueError, "shape"),
+            (2, {"foo": "bar"}, TypeError, "'foo'"),
+        ],
+    )
+    def test_repeat_raises(self, data, repeats, kwargs, error, msg, use_numpy):
+        with pytest.raises(error, match=msg):
+            if use_numpy:
+                np.repeat(data, repeats, **kwargs)
+            else:
+                data.repeat(repeats, **kwargs)
+
+    def test_delete(self, data):
+        result = data.delete(0)
+        expected = data[1:]
+        tm.assert_extension_array_equal(result, expected)
+
+        result = data.delete([1, 3])
+        expected = data._concat_same_type([data[[0]], data[[2]], data[4:]])
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_insert(self, data):
+        # insert at the beginning
+        result = data[1:].insert(0, data[0])
+        tm.assert_extension_array_equal(result, data)
+
+        result = data[1:].insert(-len(data[1:]), data[0])
+        tm.assert_extension_array_equal(result, data)
+
+        # insert at the middle
+        result = data[:-1].insert(4, data[-1])
+
+        taker = np.arange(len(data))
+        taker[5:] = taker[4:-1]
+        taker[4] = len(data) - 1
+        expected = data.take(taker)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_insert_invalid(self, data, invalid_scalar):
+        item = invalid_scalar
+
+        with pytest.raises((TypeError, ValueError)):
+            data.insert(0, item)
+
+        with pytest.raises((TypeError, ValueError)):
+            data.insert(4, item)
+
+        with pytest.raises((TypeError, ValueError)):
+            data.insert(len(data) - 1, item)
+
+    def test_insert_invalid_loc(self, data):
+        ub = len(data)
+
+        with pytest.raises(IndexError):
+            data.insert(ub + 1, data[0])
+
+        with pytest.raises(IndexError):
+            data.insert(-ub - 1, data[0])
+
+        with pytest.raises(TypeError):
+            # we expect TypeError here instead of IndexError to match np.insert
+            data.insert(1.5, data[0])
+
+    @pytest.mark.parametrize("box", [pd.array, pd.Series, pd.DataFrame])
+    def test_equals(self, data, na_value, as_series, box):
+        data2 = type(data)._from_sequence([data[0]] * len(data), dtype=data.dtype)
+        data_na = type(data)._from_sequence([na_value] * len(data), dtype=data.dtype)
+
+        data = tm.box_expected(data, box, transpose=False)
+        data2 = tm.box_expected(data2, box, transpose=False)
+        data_na = tm.box_expected(data_na, box, transpose=False)
+
+        # we are asserting with `is True/False` explicitly, to test that the
+        # result is an actual Python bool, and not something "truthy"
+
+        assert data.equals(data) is True
+        assert data.equals(data.copy()) is True
+
+        # unequal other data
+        assert data.equals(data2) is False
+        assert data.equals(data_na) is False
+
+        # different length
+        assert data[:2].equals(data[:3]) is False
+
+        # empty are equal
+        assert data[:0].equals(data[:0]) is True
+
+        # other types
+        assert data.equals(None) is False
+        assert data[[0]].equals(data[0]) is False
+
+    def test_equals_same_data_different_object(self, data):
+        # https://github.com/pandas-dev/pandas/issues/34660
+        assert pd.Series(data).equals(pd.Series(data))
diff --git a/pandas/pandas/tests/extension/base/missing.py b/pandas/pandas/tests/extension/base/missing.py
new file mode 100644
index 0000000000000..cee565d4f7c1e
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/missing.py
@@ -0,0 +1,181 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseMissingTests:
+    def test_isna(self, data_missing):
+        expected = np.array([True, False])
+
+        result = pd.isna(data_missing)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = pd.Series(data_missing).isna()
+        expected = pd.Series(expected)
+        tm.assert_series_equal(result, expected)
+
+        # GH 21189
+        result = pd.Series(data_missing).drop([0, 1]).isna()
+        expected = pd.Series([], dtype=bool)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("na_func", ["isna", "notna"])
+    def test_isna_returns_copy(self, data_missing, na_func):
+        result = pd.Series(data_missing)
+        expected = result.copy()
+        mask = getattr(result, na_func)()
+        if isinstance(mask.dtype, pd.SparseDtype):
+            # TODO: GH 57739
+            mask = np.array(mask)
+            mask.flags.writeable = True
+
+        mask[:] = True
+        tm.assert_series_equal(result, expected)
+
+    def test_dropna_array(self, data_missing):
+        result = data_missing.dropna()
+        expected = data_missing[[1]]
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_dropna_series(self, data_missing):
+        ser = pd.Series(data_missing)
+        result = ser.dropna()
+        expected = ser.iloc[[1]]
+        tm.assert_series_equal(result, expected)
+
+    def test_dropna_frame(self, data_missing):
+        df = pd.DataFrame({"A": data_missing}, columns=pd.Index(["A"], dtype=object))
+
+        # defaults
+        result = df.dropna()
+        expected = df.iloc[[1]]
+        tm.assert_frame_equal(result, expected)
+
+        # axis = 1
+        result = df.dropna(axis="columns")
+        expected = pd.DataFrame(index=pd.RangeIndex(2), columns=pd.Index([]))
+        tm.assert_frame_equal(result, expected)
+
+        # multiple
+        df = pd.DataFrame({"A": data_missing, "B": [1, np.nan]})
+        result = df.dropna()
+        expected = df.iloc[:0]
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_scalar(self, data_missing):
+        valid = data_missing[1]
+        result = data_missing.fillna(valid)
+        expected = data_missing.fillna(valid)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_fillna_with_none(self, data_missing):
+        # GH#57723
+        result = data_missing.fillna(None)
+        expected = data_missing
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_fillna_limit_pad(self, data_missing):
+        arr = data_missing.take([1, 0, 0, 0, 1])
+        result = pd.Series(arr).ffill(limit=2)
+        expected = pd.Series(data_missing.take([1, 1, 1, 0, 1]))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "limit_area, input_ilocs, expected_ilocs",
+        [
+            ("outside", [1, 0, 0, 0, 1], [1, 0, 0, 0, 1]),
+            ("outside", [1, 0, 1, 0, 1], [1, 0, 1, 0, 1]),
+            ("outside", [0, 1, 1, 1, 0], [0, 1, 1, 1, 1]),
+            ("outside", [0, 1, 0, 1, 0], [0, 1, 0, 1, 1]),
+            ("inside", [1, 0, 0, 0, 1], [1, 1, 1, 1, 1]),
+            ("inside", [1, 0, 1, 0, 1], [1, 1, 1, 1, 1]),
+            ("inside", [0, 1, 1, 1, 0], [0, 1, 1, 1, 0]),
+            ("inside", [0, 1, 0, 1, 0], [0, 1, 1, 1, 0]),
+        ],
+    )
+    def test_ffill_limit_area(
+        self, data_missing, limit_area, input_ilocs, expected_ilocs
+    ):
+        # GH#56616
+        arr = data_missing.take(input_ilocs)
+        result = pd.Series(arr).ffill(limit_area=limit_area)
+        expected = pd.Series(data_missing.take(expected_ilocs))
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_limit_backfill(self, data_missing):
+        arr = data_missing.take([1, 0, 0, 0, 1])
+        result = pd.Series(arr).bfill(limit=2)
+        expected = pd.Series(data_missing.take([1, 0, 1, 1, 1]))
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_no_op_returns_copy(self, data):
+        data = data[~data.isna()]
+
+        valid = data[0]
+        result = data.fillna(valid)
+        assert result is not data
+        tm.assert_extension_array_equal(result, data)
+
+        result = data._pad_or_backfill(method="backfill")
+        assert result is not data
+        tm.assert_extension_array_equal(result, data)
+
+    def test_fillna_series(self, data_missing):
+        fill_value = data_missing[1]
+        ser = pd.Series(data_missing)
+
+        result = ser.fillna(fill_value)
+        expected = pd.Series(
+            data_missing._from_sequence(
+                [fill_value, fill_value], dtype=data_missing.dtype
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+        # Fill with a series
+        result = ser.fillna(expected)
+        tm.assert_series_equal(result, expected)
+
+        # Fill with a series not affecting the missing values
+        result = ser.fillna(ser)
+        tm.assert_series_equal(result, ser)
+
+    def test_fillna_series_method(self, data_missing, fillna_method):
+        fill_value = data_missing[1]
+
+        if fillna_method == "ffill":
+            data_missing = data_missing[::-1]
+
+        result = getattr(pd.Series(data_missing), fillna_method)()
+        expected = pd.Series(
+            data_missing._from_sequence(
+                [fill_value, fill_value], dtype=data_missing.dtype
+            )
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_frame(self, data_missing):
+        fill_value = data_missing[1]
+
+        result = pd.DataFrame({"A": data_missing, "B": [1, 2]}).fillna(fill_value)
+
+        expected = pd.DataFrame(
+            {
+                "A": data_missing._from_sequence(
+                    [fill_value, fill_value], dtype=data_missing.dtype
+                ),
+                "B": [1, 2],
+            }
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_fill_other(self, data):
+        result = pd.DataFrame({"A": data, "B": [np.nan] * len(data)}).fillna({"B": 0.0})
+
+        expected = pd.DataFrame({"A": data, "B": [0.0] * len(result)})
+
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/extension/base/ops.py b/pandas/pandas/tests/extension/base/ops.py
new file mode 100644
index 0000000000000..222ff42d45052
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/ops.py
@@ -0,0 +1,289 @@
+from __future__ import annotations
+
+from typing import final
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_string_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core import ops
+
+
+class BaseOpsUtil:
+    series_scalar_exc: type[Exception] | None = TypeError
+    frame_scalar_exc: type[Exception] | None = TypeError
+    series_array_exc: type[Exception] | None = TypeError
+    divmod_exc: type[Exception] | None = TypeError
+
+    def _get_expected_exception(
+        self, op_name: str, obj, other
+    ) -> type[Exception] | tuple[type[Exception], ...] | None:
+        # Find the Exception, if any we expect to raise calling
+        #  obj.__op_name__(other)
+
+        # The self.obj_bar_exc pattern isn't great in part because it can depend
+        #  on op_name or dtypes, but we use it here for backward-compatibility.
+        if op_name in ["__divmod__", "__rdivmod__"]:
+            result = self.divmod_exc
+        elif isinstance(obj, pd.Series) and isinstance(other, pd.Series):
+            result = self.series_array_exc
+        elif isinstance(obj, pd.Series):
+            result = self.series_scalar_exc
+        else:
+            result = self.frame_scalar_exc
+
+        return result
+
+    def _cast_pointwise_result(self, op_name: str, obj, other, pointwise_result):
+        # In _check_op we check that the result of a pointwise operation
+        #  (found via _combine) matches the result of the vectorized
+        #  operation obj.__op_name__(other).
+        #  In some cases pandas dtype inference on the scalar result may not
+        #  give a matching dtype even if both operations are behaving "correctly".
+        #  In these cases, do extra required casting here.
+        return pointwise_result
+
+    def get_op_from_name(self, op_name: str):
+        return tm.get_op_from_name(op_name)
+
+    # Subclasses are not expected to need to override check_opname, _check_op,
+    #  _check_divmod_op, or _combine.
+    #  Ideally any relevant overriding can be done in _cast_pointwise_result,
+    #  get_op_from_name, and the specification of `exc`. If you find a use
+    #  case that still requires overriding _check_op or _combine, please let
+    #  us know at github.com/pandas-dev/pandas/issues
+    @final
+    def check_opname(self, ser: pd.Series, op_name: str, other):
+        exc = self._get_expected_exception(op_name, ser, other)
+        op = self.get_op_from_name(op_name)
+
+        self._check_op(ser, op, other, op_name, exc)
+
+    # see comment on check_opname
+    @final
+    def _combine(self, obj, other, op):
+        if isinstance(obj, pd.DataFrame):
+            if len(obj.columns) != 1:
+                raise NotImplementedError
+            expected = obj.iloc[:, 0].combine(other, op).to_frame()
+        else:
+            expected = obj.combine(other, op)
+        return expected
+
+    # see comment on check_opname
+    @final
+    def _check_op(
+        self, ser: pd.Series, op, other, op_name: str, exc=NotImplementedError
+    ):
+        # Check that the Series/DataFrame arithmetic/comparison method matches
+        #  the pointwise result from _combine.
+
+        if exc is None:
+            result = op(ser, other)
+            expected = self._combine(ser, other, op)
+            expected = self._cast_pointwise_result(op_name, ser, other, expected)
+            assert isinstance(result, type(ser))
+            tm.assert_equal(result, expected)
+        else:
+            with pytest.raises(exc):
+                op(ser, other)
+
+    # see comment on check_opname
+    @final
+    def _check_divmod_op(self, ser: pd.Series, op, other):
+        # check that divmod behavior matches behavior of floordiv+mod
+        if op is divmod:
+            exc = self._get_expected_exception("__divmod__", ser, other)
+        else:
+            exc = self._get_expected_exception("__rdivmod__", ser, other)
+        if exc is None:
+            result_div, result_mod = op(ser, other)
+            if op is divmod:
+                expected_div, expected_mod = ser // other, ser % other
+            else:
+                expected_div, expected_mod = other // ser, other % ser
+            tm.assert_series_equal(result_div, expected_div)
+            tm.assert_series_equal(result_mod, expected_mod)
+        else:
+            with pytest.raises(exc):
+                divmod(ser, other)
+
+
+class BaseArithmeticOpsTests(BaseOpsUtil):
+    """
+    Various Series and DataFrame arithmetic ops methods.
+
+    Subclasses supporting various ops should set the class variables
+    to indicate that they support ops of that kind
+
+    * series_scalar_exc = TypeError
+    * frame_scalar_exc = TypeError
+    * series_array_exc = TypeError
+    * divmod_exc = TypeError
+    """
+
+    series_scalar_exc: type[Exception] | None = TypeError
+    frame_scalar_exc: type[Exception] | None = TypeError
+    series_array_exc: type[Exception] | None = TypeError
+    divmod_exc: type[Exception] | None = TypeError
+
+    def test_arith_series_with_scalar(self, data, all_arithmetic_operators):
+        # series & scalar
+        if all_arithmetic_operators == "__rmod__" and is_string_dtype(data.dtype):
+            pytest.skip("Skip testing Python string formatting")
+
+        op_name = all_arithmetic_operators
+        ser = pd.Series(data)
+        self.check_opname(ser, op_name, ser.iloc[0])
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators):
+        # frame & scalar
+        if all_arithmetic_operators == "__rmod__" and is_string_dtype(data.dtype):
+            pytest.skip("Skip testing Python string formatting")
+
+        op_name = all_arithmetic_operators
+        df = pd.DataFrame({"A": data})
+        self.check_opname(df, op_name, data[0])
+
+    def test_arith_series_with_array(self, data, all_arithmetic_operators):
+        # ndarray & other series
+        op_name = all_arithmetic_operators
+        ser = pd.Series(data)
+        self.check_opname(ser, op_name, pd.Series([ser.iloc[0]] * len(ser)))
+
+    def test_divmod(self, data):
+        ser = pd.Series(data)
+        self._check_divmod_op(ser, divmod, 1)
+        self._check_divmod_op(1, ops.rdivmod, ser)
+
+    def test_divmod_series_array(self, data, data_for_twos):
+        ser = pd.Series(data)
+        self._check_divmod_op(ser, divmod, data)
+
+        other = data_for_twos
+        self._check_divmod_op(other, ops.rdivmod, ser)
+
+        other = pd.Series(other)
+        self._check_divmod_op(other, ops.rdivmod, ser)
+
+    def test_add_series_with_extension_array(self, data):
+        # Check adding an ExtensionArray to a Series of the same dtype matches
+        # the behavior of adding the arrays directly and then wrapping in a
+        # Series.
+
+        ser = pd.Series(data)
+
+        exc = self._get_expected_exception("__add__", ser, data)
+        if exc is not None:
+            with pytest.raises(exc):
+                ser + data
+            return
+
+        result = ser + data
+        expected = pd.Series(data + data)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [pd.Series, pd.DataFrame, pd.Index])
+    @pytest.mark.parametrize(
+        "op_name",
+        [
+            x
+            for x in tm.arithmetic_dunder_methods + tm.comparison_dunder_methods
+            if not x.startswith("__r")
+        ],
+    )
+    def test_direct_arith_with_ndframe_returns_not_implemented(
+        self, data, box, op_name
+    ):
+        # EAs should return NotImplemented for ops with Series/DataFrame/Index
+        # Pandas takes care of unboxing the series and calling the EA's op.
+        other = box(data)
+
+        if hasattr(data, op_name):
+            result = getattr(data, op_name)(other)
+            assert result is NotImplemented
+
+
+class BaseComparisonOpsTests(BaseOpsUtil):
+    """Various Series and DataFrame comparison ops methods."""
+
+    def _compare_other(self, ser: pd.Series, data, op, other):
+        if op.__name__ in ["eq", "ne"]:
+            # comparison should match point-wise comparisons
+            result = op(ser, other)
+            expected = ser.combine(other, op)
+            expected = self._cast_pointwise_result(op.__name__, ser, other, expected)
+            tm.assert_series_equal(result, expected)
+
+        else:
+            exc = None
+            try:
+                result = op(ser, other)
+            except Exception as err:
+                exc = err
+
+            if exc is None:
+                # Didn't error, then should match pointwise behavior
+                expected = ser.combine(other, op)
+                expected = self._cast_pointwise_result(
+                    op.__name__, ser, other, expected
+                )
+                tm.assert_series_equal(result, expected)
+            else:
+                with pytest.raises(type(exc)):
+                    ser.combine(other, op)
+
+    def test_compare_scalar(self, data, comparison_op):
+        ser = pd.Series(data)
+        self._compare_other(ser, data, comparison_op, 0)
+
+    def test_compare_array(self, data, comparison_op):
+        ser = pd.Series(data)
+        other = pd.Series([data[0]] * len(data), dtype=data.dtype)
+        self._compare_other(ser, data, comparison_op, other)
+
+
+class BaseUnaryOpsTests(BaseOpsUtil):
+    def test_invert(self, data):
+        ser = pd.Series(data, name="name")
+        try:
+            # 10 is an arbitrary choice here, just avoid iterating over
+            #  the whole array to trim test runtime
+            [~x for x in data[:10]]
+        except TypeError:
+            # scalars don't support invert -> we don't expect the vectorized
+            #  operation to succeed
+            with pytest.raises(TypeError):
+                ~ser
+            with pytest.raises(TypeError):
+                ~data
+        else:
+            # Note we do not reuse the pointwise result to construct expected
+            #  because python semantics for negating bools are weird see GH#54569
+            result = ~ser
+            expected = pd.Series(~data, name="name")
+            tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("ufunc", [np.positive, np.negative, np.abs])
+    def test_unary_ufunc_dunder_equivalence(self, data, ufunc):
+        # the dunder __pos__ works if and only if np.positive works,
+        #  same for __neg__/np.negative and __abs__/np.abs
+        attr = {np.positive: "__pos__", np.negative: "__neg__", np.abs: "__abs__"}[
+            ufunc
+        ]
+
+        exc = None
+        try:
+            result = getattr(data, attr)()
+        except Exception as err:
+            exc = err
+
+            # if __pos__ raised, then so should the ufunc
+            with pytest.raises((type(exc), TypeError)):
+                ufunc(data)
+        else:
+            alt = ufunc(data)
+            tm.assert_extension_array_equal(result, alt)
diff --git a/pandas/pandas/tests/extension/base/printing.py b/pandas/pandas/tests/extension/base/printing.py
new file mode 100644
index 0000000000000..b20236ec107b0
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/printing.py
@@ -0,0 +1,41 @@
+import io
+
+import pytest
+
+import pandas as pd
+
+
+class BasePrintingTests:
+    """Tests checking the formatting of your EA when printed."""
+
+    @pytest.mark.parametrize("size", ["big", "small"])
+    def test_array_repr(self, data, size):
+        if size == "small":
+            data = data[:5]
+        else:
+            data = type(data)._concat_same_type([data] * 5)
+
+        result = repr(data)
+        assert type(data).__name__ in result
+        assert f"Length: {len(data)}" in result
+        assert str(data.dtype) in result
+        if size == "big":
+            assert "..." in result
+
+    def test_array_repr_unicode(self, data):
+        result = str(data)
+        assert isinstance(result, str)
+
+    def test_series_repr(self, data):
+        ser = pd.Series(data)
+        assert data.dtype.name in repr(ser)
+
+    def test_dataframe_repr(self, data):
+        df = pd.DataFrame({"A": data})
+        repr(df)
+
+    def test_dtype_name_in_info(self, data):
+        buf = io.StringIO()
+        pd.DataFrame({"A": data}).info(buf=buf)
+        result = buf.getvalue()
+        assert data.dtype.name in result
diff --git a/pandas/pandas/tests/extension/base/reduce.py b/pandas/pandas/tests/extension/base/reduce.py
new file mode 100644
index 0000000000000..4b3431d938f96
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/reduce.py
@@ -0,0 +1,128 @@
+from typing import final
+
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseReduceTests:
+    """
+    Reduction specific tests. Generally these only
+    make sense for numeric/boolean operations.
+    """
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        # Specify if we expect this reduction to succeed.
+        return False
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        # We perform the same operation on the np.float64 data and check
+        #  that the results match. Override if you need to cast to something
+        #  other than float64.
+        res_op = getattr(ser, op_name)
+
+        try:
+            alt = ser.astype("float64")
+        except (TypeError, ValueError):
+            # e.g. Interval can't cast (TypeError), StringArray can't cast
+            #  (ValueError), so let's cast to object and do
+            #  the reduction pointwise
+            alt = ser.astype(object)
+
+        exp_op = getattr(alt, op_name)
+        if op_name == "count":
+            result = res_op()
+            expected = exp_op()
+        else:
+            result = res_op(skipna=skipna)
+            expected = exp_op(skipna=skipna)
+        tm.assert_almost_equal(result, expected)
+
+    def _get_expected_reduction_dtype(self, arr, op_name: str, skipna: bool):
+        # Find the expected dtype when the given reduction is done on a DataFrame
+        # column with this array.  The default assumes float64-like behavior,
+        # i.e. retains the dtype.
+        return arr.dtype
+
+    # We anticipate that authors should not need to override check_reduce_frame,
+    #  but should be able to do any necessary overriding in
+    #  _get_expected_reduction_dtype. If you have a use case where this
+    #  does not hold, please let us know at github.com/pandas-dev/pandas/issues.
+    @final
+    def check_reduce_frame(self, ser: pd.Series, op_name: str, skipna: bool):
+        # Check that the 2D reduction done in a DataFrame reduction "looks like"
+        # a wrapped version of the 1D reduction done by Series.
+        arr = ser.array
+        df = pd.DataFrame({"a": arr})
+
+        kwargs = {"ddof": 1} if op_name in ["var", "std", "sem"] else {}
+
+        cmp_dtype = self._get_expected_reduction_dtype(arr, op_name, skipna)
+
+        # The DataFrame method just calls arr._reduce with keepdims=True,
+        #  so this first check is perfunctory.
+        result1 = arr._reduce(op_name, skipna=skipna, keepdims=True, **kwargs)
+        result2 = getattr(df, op_name)(skipna=skipna, **kwargs).array
+        tm.assert_extension_array_equal(result1, result2)
+
+        # Check that the 2D reduction looks like a wrapped version of the
+        #  1D reduction
+        if not skipna and ser.isna().any():
+            expected = pd.array([pd.NA], dtype=cmp_dtype)
+        else:
+            exp_value = getattr(ser.dropna(), op_name)()
+            expected = pd.array([exp_value], dtype=cmp_dtype)
+
+        tm.assert_extension_array_equal(result1, expected)
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_series_boolean(self, data, all_boolean_reductions, skipna):
+        op_name = all_boolean_reductions
+        ser = pd.Series(data)
+
+        if not self._supports_reduction(ser, op_name):
+            # TODO: the message being checked here isn't actually checking anything
+            msg = (
+                "[Cc]annot perform|Categorical is not ordered for operation|"
+                "does not support operation|"
+            )
+
+            with pytest.raises(TypeError, match=msg):
+                getattr(ser, op_name)(skipna=skipna)
+
+        else:
+            self.check_reduce(ser, op_name, skipna)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_series_numeric(self, data, all_numeric_reductions, skipna):
+        op_name = all_numeric_reductions
+        ser = pd.Series(data)
+
+        if not self._supports_reduction(ser, op_name):
+            # TODO: the message being checked here isn't actually checking anything
+            msg = (
+                "[Cc]annot perform|Categorical is not ordered for operation|"
+                "does not support operation|"
+            )
+
+            with pytest.raises(TypeError, match=msg):
+                getattr(ser, op_name)(skipna=skipna)
+
+        else:
+            # min/max with empty produce numpy warnings
+            self.check_reduce(ser, op_name, skipna)
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_frame(self, data, all_numeric_reductions, skipna):
+        op_name = all_numeric_reductions
+        ser = pd.Series(data)
+
+        if op_name == "count":
+            pytest.skip(f"{op_name} not an array method")
+
+        if not self._supports_reduction(ser, op_name):
+            pytest.skip(f"Reduction {op_name} not supported for this dtype")
+
+        self.check_reduce_frame(ser, op_name, skipna)
diff --git a/pandas/pandas/tests/extension/base/reshaping.py b/pandas/pandas/tests/extension/base/reshaping.py
new file mode 100644
index 0000000000000..2915c0585f373
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/reshaping.py
@@ -0,0 +1,379 @@
+import itertools
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.extensions import ExtensionArray
+from pandas.core.internals.blocks import EABackedBlock
+
+
+class BaseReshapingTests:
+    """Tests for reshaping and concatenation."""
+
+    @pytest.mark.parametrize("in_frame", [True, False])
+    def test_concat(self, data, in_frame):
+        wrapped = pd.Series(data)
+        if in_frame:
+            wrapped = pd.DataFrame(wrapped)
+        result = pd.concat([wrapped, wrapped], ignore_index=True)
+
+        assert len(result) == len(data) * 2
+
+        if in_frame:
+            dtype = result.dtypes[0]
+        else:
+            dtype = result.dtype
+
+        assert dtype == data.dtype
+        if hasattr(result._mgr, "blocks"):
+            assert isinstance(result._mgr.blocks[0], EABackedBlock)
+        assert isinstance(result._mgr.blocks[0].values, ExtensionArray)
+
+    @pytest.mark.parametrize("in_frame", [True, False])
+    def test_concat_all_na_block(self, data_missing, in_frame):
+        valid_block = pd.Series(data_missing.take([1, 1]), index=range(2))
+        na_block = pd.Series(data_missing.take([0, 0]), index=range(2, 4))
+        if in_frame:
+            valid_block = pd.DataFrame({"a": valid_block})
+            na_block = pd.DataFrame({"a": na_block})
+        result = pd.concat([valid_block, na_block])
+        if in_frame:
+            expected = pd.DataFrame({"a": data_missing.take([1, 1, 0, 0])})
+            tm.assert_frame_equal(result, expected)
+        else:
+            expected = pd.Series(data_missing.take([1, 1, 0, 0]))
+            tm.assert_series_equal(result, expected)
+
+    def test_concat_mixed_dtypes(self, data):
+        # https://github.com/pandas-dev/pandas/issues/20762
+        df1 = pd.DataFrame({"A": data[:3]})
+        df2 = pd.DataFrame({"A": [1, 2, 3]})
+        df3 = pd.DataFrame({"A": ["a", "b", "c"]}).astype("category")
+        dfs = [df1, df2, df3]
+
+        # dataframes
+        result = pd.concat(dfs)
+        expected = pd.concat([x.astype(object) for x in dfs])
+        tm.assert_frame_equal(result, expected)
+
+        # series
+        result = pd.concat([x["A"] for x in dfs])
+        expected = pd.concat([x["A"].astype(object) for x in dfs])
+        tm.assert_series_equal(result, expected)
+
+        # simple test for just EA and one other
+        result = pd.concat([df1, df2.astype(object)])
+        expected = pd.concat([df1.astype("object"), df2.astype("object")])
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.concat([df1["A"], df2["A"].astype(object)])
+        expected = pd.concat([df1["A"].astype("object"), df2["A"].astype("object")])
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_columns(self, data, na_value):
+        df1 = pd.DataFrame({"A": data[:3]})
+        df2 = pd.DataFrame({"B": [1, 2, 3]})
+
+        expected = pd.DataFrame({"A": data[:3], "B": [1, 2, 3]})
+        result = pd.concat([df1, df2], axis=1)
+        tm.assert_frame_equal(result, expected)
+        result = pd.concat([df1["A"], df2["B"]], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # non-aligned
+        df2 = pd.DataFrame({"B": [1, 2, 3]}, index=[1, 2, 3])
+        expected = pd.DataFrame(
+            {
+                "A": data._from_sequence(list(data[:3]) + [na_value], dtype=data.dtype),
+                "B": [np.nan, 1, 2, 3],
+            }
+        )
+
+        result = pd.concat([df1, df2], axis=1)
+        tm.assert_frame_equal(result, expected)
+        result = pd.concat([df1["A"], df2["B"]], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_extension_arrays_copy_false(self, data, na_value):
+        # GH 20756
+        df1 = pd.DataFrame({"A": data[:3]})
+        df2 = pd.DataFrame({"B": data[3:7]})
+        expected = pd.DataFrame(
+            {
+                "A": data._from_sequence(list(data[:3]) + [na_value], dtype=data.dtype),
+                "B": data[3:7],
+            }
+        )
+        result = pd.concat([df1, df2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_with_reindex(self, data):
+        # GH-33027
+        a = pd.DataFrame({"a": data[:5]})
+        b = pd.DataFrame({"b": data[:5]})
+        result = pd.concat([a, b], ignore_index=True)
+        expected = pd.DataFrame(
+            {
+                "a": data.take(list(range(5)) + ([-1] * 5), allow_fill=True),
+                "b": data.take(([-1] * 5) + list(range(5)), allow_fill=True),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_align(self, data, na_value):
+        a = data[:3]
+        b = data[2:5]
+        r1, r2 = pd.Series(a).align(pd.Series(b, index=[1, 2, 3]))
+
+        # Assumes that the ctor can take a list of scalars of the type
+        e1 = pd.Series(data._from_sequence(list(a) + [na_value], dtype=data.dtype))
+        e2 = pd.Series(data._from_sequence([na_value] + list(b), dtype=data.dtype))
+        tm.assert_series_equal(r1, e1)
+        tm.assert_series_equal(r2, e2)
+
+    def test_align_frame(self, data, na_value):
+        a = data[:3]
+        b = data[2:5]
+        r1, r2 = pd.DataFrame({"A": a}).align(pd.DataFrame({"A": b}, index=[1, 2, 3]))
+
+        # Assumes that the ctor can take a list of scalars of the type
+        e1 = pd.DataFrame(
+            {"A": data._from_sequence(list(a) + [na_value], dtype=data.dtype)}
+        )
+        e2 = pd.DataFrame(
+            {"A": data._from_sequence([na_value] + list(b), dtype=data.dtype)}
+        )
+        tm.assert_frame_equal(r1, e1)
+        tm.assert_frame_equal(r2, e2)
+
+    def test_align_series_frame(self, data, na_value):
+        # https://github.com/pandas-dev/pandas/issues/20576
+        ser = pd.Series(data, name="a")
+        df = pd.DataFrame({"col": np.arange(len(ser) + 1)})
+        r1, r2 = ser.align(df)
+
+        e1 = pd.Series(
+            data._from_sequence(list(data) + [na_value], dtype=data.dtype),
+            name=ser.name,
+        )
+
+        tm.assert_series_equal(r1, e1)
+        tm.assert_frame_equal(r2, df)
+
+    def test_set_frame_expand_regular_with_extension(self, data):
+        df = pd.DataFrame({"A": [1] * len(data)})
+        df["B"] = data
+        expected = pd.DataFrame({"A": [1] * len(data), "B": data})
+        tm.assert_frame_equal(df, expected)
+
+    def test_set_frame_expand_extension_with_regular(self, data):
+        df = pd.DataFrame({"A": data})
+        df["B"] = [1] * len(data)
+        expected = pd.DataFrame({"A": data, "B": [1] * len(data)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_set_frame_overwrite_object(self, data):
+        # https://github.com/pandas-dev/pandas/issues/20555
+        df = pd.DataFrame({"A": [1] * len(data)}, dtype=object)
+        df["A"] = data
+        assert df.dtypes["A"] == data.dtype
+
+    def test_merge(self, data, na_value):
+        # GH-20743
+        df1 = pd.DataFrame({"ext": data[:3], "int1": [1, 2, 3], "key": [0, 1, 2]})
+        df2 = pd.DataFrame({"int2": [1, 2, 3, 4], "key": [0, 0, 1, 3]})
+
+        res = pd.merge(df1, df2)
+        exp = pd.DataFrame(
+            {
+                "int1": [1, 1, 2],
+                "int2": [1, 2, 3],
+                "key": [0, 0, 1],
+                "ext": data._from_sequence(
+                    [data[0], data[0], data[1]], dtype=data.dtype
+                ),
+            }
+        )
+        tm.assert_frame_equal(res, exp[["ext", "int1", "key", "int2"]])
+
+        res = pd.merge(df1, df2, how="outer")
+        exp = pd.DataFrame(
+            {
+                "int1": [1, 1, 2, 3, np.nan],
+                "int2": [1, 2, 3, np.nan, 4],
+                "key": [0, 0, 1, 2, 3],
+                "ext": data._from_sequence(
+                    [data[0], data[0], data[1], data[2], na_value], dtype=data.dtype
+                ),
+            }
+        )
+        tm.assert_frame_equal(res, exp[["ext", "int1", "key", "int2"]])
+
+    def test_merge_on_extension_array(self, data):
+        # GH 23020
+        a, b = data[:2]
+        key = type(data)._from_sequence([a, b], dtype=data.dtype)
+
+        df = pd.DataFrame({"key": key, "val": [1, 2]})
+        result = pd.merge(df, df, on="key")
+        expected = pd.DataFrame({"key": key, "val_x": [1, 2], "val_y": [1, 2]})
+        tm.assert_frame_equal(result, expected)
+
+        # order
+        result = pd.merge(df.iloc[[1, 0]], df, on="key")
+        expected = expected.iloc[[1, 0]].reset_index(drop=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_on_extension_array_duplicates(self, data):
+        # GH 23020
+        a, b = data[:2]
+        key = type(data)._from_sequence([a, b, a], dtype=data.dtype)
+        df1 = pd.DataFrame({"key": key, "val": [1, 2, 3]})
+        df2 = pd.DataFrame({"key": key, "val": [1, 2, 3]})
+
+        result = pd.merge(df1, df2, on="key")
+        expected = pd.DataFrame(
+            {
+                "key": key.take([0, 0, 1, 2, 2]),
+                "val_x": [1, 1, 2, 3, 3],
+                "val_y": [1, 3, 2, 1, 3],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "columns",
+        [
+            ["A", "B"],
+            pd.MultiIndex.from_tuples(
+                [("A", "a"), ("A", "b")], names=["outer", "inner"]
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("future_stack", [True, False])
+    def test_stack(self, data, columns, future_stack):
+        df = pd.DataFrame({"A": data[:5], "B": data[:5]})
+        df.columns = columns
+        result = df.stack(future_stack=future_stack)
+        expected = df.astype(object).stack(future_stack=future_stack)
+        # we need a second astype(object), in case the constructor inferred
+        # object -> specialized, as is done for period.
+        expected = expected.astype(object)
+
+        if isinstance(expected, pd.Series):
+            assert result.dtype == df.iloc[:, 0].dtype
+        else:
+            assert all(result.dtypes == df.iloc[:, 0].dtype)
+
+        result = result.astype(object)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            # Two levels, uniform.
+            pd.MultiIndex.from_product(([["A", "B"], ["a", "b"]]), names=["a", "b"]),
+            # non-uniform
+            pd.MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "b")]),
+            # three levels, non-uniform
+            pd.MultiIndex.from_product([("A", "B"), ("a", "b", "c"), (0, 1, 2)]),
+            pd.MultiIndex.from_tuples(
+                [
+                    ("A", "a", 1),
+                    ("A", "b", 0),
+                    ("A", "a", 0),
+                    ("B", "a", 0),
+                    ("B", "c", 1),
+                ]
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("obj", ["series", "frame"])
+    def test_unstack(self, data, index, obj):
+        data = data[: len(index)]
+        if obj == "series":
+            ser = pd.Series(data, index=index)
+        else:
+            ser = pd.DataFrame({"A": data, "B": data}, index=index)
+
+        n = index.nlevels
+        levels = list(range(n))
+        # [0, 1, 2]
+        # [(0,), (1,), (2,), (0, 1), (0, 2), (1, 0), (1, 2), (2, 0), (2, 1)]
+        combinations = itertools.chain.from_iterable(
+            itertools.permutations(levels, i) for i in range(1, n)
+        )
+
+        for level in combinations:
+            result = ser.unstack(level=level)
+            assert all(
+                isinstance(result[col].array, type(data)) for col in result.columns
+            )
+
+            if obj == "series":
+                # We should get the same result with to_frame+unstack+droplevel
+                df = ser.to_frame()
+
+                alt = df.unstack(level=level).droplevel(0, axis=1)
+                tm.assert_frame_equal(result, alt)
+
+            obj_ser = ser.astype(object)
+
+            expected = obj_ser.unstack(level=level, fill_value=data.dtype.na_value)
+            if obj == "series":
+                assert (expected.dtypes == object).all()
+
+            result = result.astype(object)
+            tm.assert_frame_equal(result, expected)
+
+    def test_ravel(self, data):
+        # as long as EA is 1D-only, ravel is a no-op
+        result = data.ravel()
+        assert type(result) == type(data)
+
+        if data.dtype._is_immutable:
+            pytest.skip(f"test_ravel assumes mutability and {data.dtype} is immutable")
+
+        # Check that we have a view, not a copy
+        result[0] = result[1]
+        assert data[0] == data[1]
+
+    def test_transpose(self, data):
+        result = data.transpose()
+        assert type(result) == type(data)
+
+        # check we get a new object
+        assert result is not data
+
+        # If we ever _did_ support 2D, shape should be reversed
+        assert result.shape == data.shape[::-1]
+
+        if data.dtype._is_immutable:
+            pytest.skip(
+                f"test_transpose assumes mutability and {data.dtype} is immutable"
+            )
+
+        # Check that we have a view, not a copy
+        result[0] = result[1]
+        assert data[0] == data[1]
+
+    def test_transpose_frame(self, data):
+        df = pd.DataFrame({"A": data[:4], "B": data[:4]}, index=["a", "b", "c", "d"])
+        result = df.T
+        expected = pd.DataFrame(
+            {
+                "a": type(data)._from_sequence([data[0]] * 2, dtype=data.dtype),
+                "b": type(data)._from_sequence([data[1]] * 2, dtype=data.dtype),
+                "c": type(data)._from_sequence([data[2]] * 2, dtype=data.dtype),
+                "d": type(data)._from_sequence([data[3]] * 2, dtype=data.dtype),
+            },
+            index=["A", "B"],
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(np.transpose(np.transpose(df)), df)
+        tm.assert_frame_equal(np.transpose(np.transpose(df[["A"]])), df[["A"]])
diff --git a/pandas/pandas/tests/extension/base/setitem.py b/pandas/pandas/tests/extension/base/setitem.py
new file mode 100644
index 0000000000000..1d613ced2c03f
--- /dev/null
+++ b/pandas/pandas/tests/extension/base/setitem.py
@@ -0,0 +1,456 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class BaseSetitemTests:
+    @pytest.fixture(
+        params=[
+            lambda x: x.index,
+            lambda x: list(x.index),
+            lambda x: slice(None),
+            lambda x: slice(0, len(x)),
+            lambda x: range(len(x)),
+            lambda x: list(range(len(x))),
+            lambda x: np.ones(len(x), dtype=bool),
+        ],
+        ids=[
+            "index",
+            "list[index]",
+            "null_slice",
+            "full_slice",
+            "range",
+            "list(range)",
+            "mask",
+        ],
+    )
+    def full_indexer(self, request):
+        """
+        Fixture for an indexer to pass to obj.loc to get/set the full length of the
+        object.
+
+        In some cases, assumes that obj.index is the default RangeIndex.
+        """
+        return request.param
+
+    @pytest.fixture(autouse=True)
+    def skip_if_immutable(self, dtype, request):
+        if dtype._is_immutable:
+            node = request.node
+            if node.name.split("[")[0] == "test_is_immutable":
+                # This fixture is auto-used, but we want to not-skip
+                # test_is_immutable.
+                return
+
+            # When BaseSetitemTests is mixed into ExtensionTests, we only
+            #  want this fixture to operate on the tests defined in this
+            #  class/file.
+            defined_in = node.function.__qualname__.split(".")[0]
+            if defined_in == "BaseSetitemTests":
+                pytest.skip("__setitem__ test not applicable with immutable dtype")
+
+    def test_is_immutable(self, data):
+        if data.dtype._is_immutable:
+            with pytest.raises(TypeError):
+                data[0] = data[0]
+        else:
+            data[0] = data[1]
+            assert data[0] == data[1]
+
+    def test_setitem_scalar_series(self, data, box_in_series):
+        if box_in_series:
+            data = pd.Series(data)
+        data[0] = data[1]
+        assert data[0] == data[1]
+
+    def test_setitem_sequence(self, data, box_in_series):
+        if box_in_series:
+            data = pd.Series(data)
+        original = data.copy()
+
+        data[[0, 1]] = [data[1], data[0]]
+        assert data[0] == original[1]
+        assert data[1] == original[0]
+
+    def test_setitem_sequence_mismatched_length_raises(self, data, as_array):
+        ser = pd.Series(data)
+        original = ser.copy()
+        value = [data[0]]
+        if as_array:
+            value = data._from_sequence(value, dtype=data.dtype)
+
+        xpr = "cannot set using a {} indexer with a different length"
+        with pytest.raises(ValueError, match=xpr.format("list-like")):
+            ser[[0, 1]] = value
+        # Ensure no modifications made before the exception
+        tm.assert_series_equal(ser, original)
+
+        with pytest.raises(ValueError, match=xpr.format("slice")):
+            ser[slice(3)] = value
+        tm.assert_series_equal(ser, original)
+
+    def test_setitem_empty_indexer(self, data, box_in_series):
+        if box_in_series:
+            data = pd.Series(data)
+        original = data.copy()
+        data[np.array([], dtype=int)] = []
+        tm.assert_equal(data, original)
+
+    def test_setitem_sequence_broadcasts(self, data, box_in_series):
+        if box_in_series:
+            data = pd.Series(data)
+        data[[0, 1]] = data[2]
+        assert data[0] == data[2]
+        assert data[1] == data[2]
+
+    @pytest.mark.parametrize("setter", ["loc", "iloc"])
+    def test_setitem_scalar(self, data, setter):
+        arr = pd.Series(data)
+        setter = getattr(arr, setter)
+        setter[0] = data[1]
+        assert arr[0] == data[1]
+
+    def test_setitem_loc_scalar_mixed(self, data):
+        df = pd.DataFrame({"A": np.arange(len(data)), "B": data})
+        df.loc[0, "B"] = data[1]
+        assert df.loc[0, "B"] == data[1]
+
+    def test_setitem_loc_scalar_single(self, data):
+        df = pd.DataFrame({"B": data})
+        df.loc[10, "B"] = data[1]
+        assert df.loc[10, "B"] == data[1]
+
+    def test_setitem_loc_scalar_multiple_homogoneous(self, data):
+        df = pd.DataFrame({"A": data, "B": data})
+        df.loc[10, "B"] = data[1]
+        assert df.loc[10, "B"] == data[1]
+
+    def test_setitem_iloc_scalar_mixed(self, data):
+        df = pd.DataFrame({"A": np.arange(len(data)), "B": data})
+        df.iloc[0, 1] = data[1]
+        assert df.loc[0, "B"] == data[1]
+
+    def test_setitem_iloc_scalar_single(self, data):
+        df = pd.DataFrame({"B": data})
+        df.iloc[10, 0] = data[1]
+        assert df.loc[10, "B"] == data[1]
+
+    def test_setitem_iloc_scalar_multiple_homogoneous(self, data):
+        df = pd.DataFrame({"A": data, "B": data})
+        df.iloc[10, 1] = data[1]
+        assert df.loc[10, "B"] == data[1]
+
+    @pytest.mark.parametrize(
+        "mask",
+        [
+            np.array([True, True, True, False, False]),
+            pd.array([True, True, True, False, False], dtype="boolean"),
+            pd.array([True, True, True, pd.NA, pd.NA], dtype="boolean"),
+        ],
+        ids=["numpy-array", "boolean-array", "boolean-array-na"],
+    )
+    def test_setitem_mask(self, data, mask, box_in_series):
+        arr = data[:5].copy()
+        expected = arr.take([0, 0, 0, 3, 4])
+        if box_in_series:
+            arr = pd.Series(arr)
+            expected = pd.Series(expected)
+        arr[mask] = data[0]
+        tm.assert_equal(expected, arr)
+
+    def test_setitem_mask_raises(self, data, box_in_series):
+        # wrong length
+        mask = np.array([True, False])
+
+        if box_in_series:
+            data = pd.Series(data)
+
+        with pytest.raises(IndexError, match="wrong length"):
+            data[mask] = data[0]
+
+        mask = pd.array(mask, dtype="boolean")
+        with pytest.raises(IndexError, match="wrong length"):
+            data[mask] = data[0]
+
+    def test_setitem_mask_boolean_array_with_na(self, data, box_in_series):
+        mask = pd.array(np.zeros(data.shape, dtype="bool"), dtype="boolean")
+        mask[:3] = True
+        mask[3:5] = pd.NA
+
+        if box_in_series:
+            data = pd.Series(data)
+
+        data[mask] = data[0]
+
+        assert (data[:3] == data[0]).all()
+
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2], pd.array([0, 1, 2], dtype="Int64"), np.array([0, 1, 2])],
+        ids=["list", "integer-array", "numpy-array"],
+    )
+    def test_setitem_integer_array(self, data, idx, box_in_series):
+        arr = data[:5].copy()
+        expected = data.take([0, 0, 0, 3, 4])
+
+        if box_in_series:
+            arr = pd.Series(arr)
+            expected = pd.Series(expected)
+
+        arr[idx] = arr[0]
+        tm.assert_equal(arr, expected)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 0, 1], pd.array([0, 0, 1], dtype="Int64"), np.array([0, 0, 1])],
+        ids=["list", "integer-array", "numpy-array"],
+    )
+    def test_setitem_integer_array_with_repeats(self, data, idx, box_in_series):
+        arr = data[:5].copy()
+        expected = data.take([2, 3, 2, 3, 4])
+
+        if box_in_series:
+            arr = pd.Series(arr)
+            expected = pd.Series(expected)
+
+        arr[idx] = [arr[2], arr[2], arr[3]]
+        tm.assert_equal(arr, expected)
+
+    @pytest.mark.parametrize(
+        "idx, box_in_series",
+        [
+            ([0, 1, 2, pd.NA], False),
+            pytest.param(
+                [0, 1, 2, pd.NA], True, marks=pytest.mark.xfail(reason="GH-31948")
+            ),
+            (pd.array([0, 1, 2, pd.NA], dtype="Int64"), False),
+            # TODO: change False to True?
+            (pd.array([0, 1, 2, pd.NA], dtype="Int64"), False),  # noqa: PT014
+        ],
+        ids=["list-False", "list-True", "integer-array-False", "integer-array-True"],
+    )
+    def test_setitem_integer_with_missing_raises(self, data, idx, box_in_series):
+        arr = data.copy()
+
+        # TODO(xfail) this raises KeyError about labels not found (it tries label-based)
+        # for list of labels with Series
+        if box_in_series:
+            arr = pd.Series(data, index=[chr(100 + i) for i in range(len(data))])
+
+        msg = "Cannot index with an integer indexer containing NA values"
+        with pytest.raises(ValueError, match=msg):
+            arr[idx] = arr[0]
+
+    @pytest.mark.parametrize("as_callable", [True, False])
+    @pytest.mark.parametrize("setter", ["loc", None])
+    def test_setitem_mask_aligned(self, data, as_callable, setter):
+        ser = pd.Series(data)
+        mask = np.zeros(len(data), dtype=bool)
+        mask[:2] = True
+
+        if as_callable:
+            mask2 = lambda x: mask
+        else:
+            mask2 = mask
+
+        if setter:
+            # loc
+            target = getattr(ser, setter)
+        else:
+            # Series.__setitem__
+            target = ser
+
+        target[mask2] = data[5:7]
+
+        ser[mask2] = data[5:7]
+        assert ser[0] == data[5]
+        assert ser[1] == data[6]
+
+    @pytest.mark.parametrize("setter", ["loc", None])
+    def test_setitem_mask_broadcast(self, data, setter):
+        ser = pd.Series(data)
+        mask = np.zeros(len(data), dtype=bool)
+        mask[:2] = True
+
+        if setter:  # loc
+            target = getattr(ser, setter)
+        else:  # __setitem__
+            target = ser
+
+        target[mask] = data[10]
+        assert ser[0] == data[10]
+        assert ser[1] == data[10]
+
+    def test_setitem_expand_columns(self, data):
+        df = pd.DataFrame({"A": data})
+        result = df.copy()
+        result["B"] = 1
+        expected = pd.DataFrame({"A": data, "B": [1] * len(data)})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.loc[:, "B"] = 1
+        tm.assert_frame_equal(result, expected)
+
+        # overwrite with new type
+        result["B"] = data
+        expected = pd.DataFrame({"A": data, "B": data})
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_expand_with_extension(self, data):
+        df = pd.DataFrame({"A": [1] * len(data)})
+        result = df.copy()
+        result["B"] = data
+        expected = pd.DataFrame({"A": [1] * len(data), "B": data})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.loc[:, "B"] = data
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_frame_invalid_length(self, data):
+        df = pd.DataFrame({"A": [1] * len(data)})
+        xpr = (
+            rf"Length of values \({len(data[:5])}\) "
+            rf"does not match length of index \({len(df)}\)"
+        )
+        with pytest.raises(ValueError, match=xpr):
+            df["B"] = data[:5]
+
+    def test_setitem_tuple_index(self, data):
+        ser = pd.Series(data[:2], index=[(0, 0), (0, 1)])
+        expected = pd.Series(data.take([1, 1]), index=ser.index)
+        ser[(0, 0)] = data[1]
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_slice(self, data, box_in_series):
+        arr = data[:5].copy()
+        expected = data.take([0, 0, 0, 3, 4])
+        if box_in_series:
+            arr = pd.Series(arr)
+            expected = pd.Series(expected)
+
+        arr[:3] = data[0]
+        tm.assert_equal(arr, expected)
+
+    def test_setitem_loc_iloc_slice(self, data):
+        arr = data[:5].copy()
+        s = pd.Series(arr, index=["a", "b", "c", "d", "e"])
+        expected = pd.Series(data.take([0, 0, 0, 3, 4]), index=s.index)
+
+        result = s.copy()
+        result.iloc[:3] = data[0]
+        tm.assert_equal(result, expected)
+
+        result = s.copy()
+        result.loc[:"c"] = data[0]
+        tm.assert_equal(result, expected)
+
+    def test_setitem_slice_mismatch_length_raises(self, data):
+        arr = data[:5]
+        with tm.external_error_raised(ValueError):
+            arr[:1] = arr[:2]
+
+    def test_setitem_slice_array(self, data):
+        arr = data[:5].copy()
+        arr[:5] = data[-5:]
+        tm.assert_extension_array_equal(arr, data[-5:])
+
+    def test_setitem_scalar_key_sequence_raise(self, data):
+        arr = data[:5].copy()
+        with tm.external_error_raised(ValueError):
+            arr[0] = arr[[0, 1]]
+
+    def test_setitem_preserves_views(self, data):
+        # GH#28150 setitem shouldn't swap the underlying data
+        view1 = data.view()
+        view2 = data[:]
+
+        data[0] = data[1]
+        assert view1[0] == data[1]
+        assert view2[0] == data[1]
+
+    def test_setitem_with_expansion_dataframe_column(self, data, full_indexer):
+        # https://github.com/pandas-dev/pandas/issues/32395
+        df = expected = pd.DataFrame(pd.Series(data))
+        result = pd.DataFrame(index=df.index)
+
+        key = full_indexer(df)
+        result.loc[key, 0] = df[0]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_with_expansion_row(self, data, na_value):
+        df = pd.DataFrame({"data": data[:1]})
+
+        df.loc[1, "data"] = data[1]
+        expected = pd.DataFrame({"data": data[:2]})
+        tm.assert_frame_equal(df, expected)
+
+        # https://github.com/pandas-dev/pandas/issues/47284
+        df.loc[2, "data"] = na_value
+        expected = pd.DataFrame(
+            {"data": pd.Series([data[0], data[1], na_value], dtype=data.dtype)}
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_series(self, data, full_indexer):
+        # https://github.com/pandas-dev/pandas/issues/32395
+        ser = pd.Series(data, name="data")
+        result = pd.Series(index=ser.index, dtype=object, name="data")
+
+        # because result has object dtype, the attempt to do setting inplace
+        #  is successful, and object dtype is retained
+        key = full_indexer(ser)
+        result.loc[key] = ser
+
+        expected = pd.Series(
+            data.astype(object), index=ser.index, name="data", dtype=object
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_frame_2d_values(self, data):
+        # GH#44514
+        df = pd.DataFrame({"A": data})
+        orig = df.copy()
+
+        df.iloc[:] = df.copy()
+        tm.assert_frame_equal(df, orig)
+
+        df.iloc[:-1] = df.iloc[:-1].copy()
+        tm.assert_frame_equal(df, orig)
+
+        df.iloc[:] = df.values
+        tm.assert_frame_equal(df, orig)
+
+        df.iloc[:-1] = df.values[:-1]
+        tm.assert_frame_equal(df, orig)
+
+    def test_delitem_series(self, data):
+        # GH#40763
+        ser = pd.Series(data, name="data")
+
+        taker = np.arange(len(ser))
+        taker = np.delete(taker, 1)
+
+        expected = ser[taker]
+        del ser[1]
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_invalid(self, data, invalid_scalar):
+        msg = ""  # messages vary by subclass, so we do not test it
+        with pytest.raises((ValueError, TypeError), match=msg):
+            data[0] = invalid_scalar
+
+        with pytest.raises((ValueError, TypeError), match=msg):
+            data[:] = invalid_scalar
+
+    def test_setitem_2d_values(self, data):
+        # GH50085
+        original = data.copy()
+        df = pd.DataFrame({"a": data, "b": data})
+        df.loc[[0, 1], :] = df.loc[[1, 0], :].values
+        assert (df.loc[0, :] == original[1]).all()
+        assert (df.loc[1, :] == original[0]).all()
diff --git a/pandas/pandas/tests/extension/conftest.py b/pandas/pandas/tests/extension/conftest.py
new file mode 100644
index 0000000000000..97fb5a0bc5066
--- /dev/null
+++ b/pandas/pandas/tests/extension/conftest.py
@@ -0,0 +1,214 @@
+import operator
+
+import pytest
+
+from pandas import Series
+
+
+@pytest.fixture
+def dtype():
+    """A fixture providing the ExtensionDtype to validate."""
+    raise NotImplementedError
+
+
+@pytest.fixture
+def data():
+    """
+    Length-100 array for this type.
+
+    * data[0] and data[1] should both be non missing
+    * data[0] and data[1] should not be equal
+    """
+    raise NotImplementedError
+
+
+@pytest.fixture
+def data_for_twos(dtype):
+    """
+    Length-100 array in which all the elements are two.
+
+    Call pytest.skip in your fixture if the dtype does not support divmod.
+    """
+    if not (dtype._is_numeric or dtype.kind == "m"):
+        # Object-dtypes may want to allow this, but for the most part
+        #  only numeric and timedelta-like dtypes will need to implement this.
+        pytest.skip(f"{dtype} is not a numeric dtype")
+
+    raise NotImplementedError
+
+
+@pytest.fixture
+def data_missing():
+    """Length-2 array with [NA, Valid]"""
+    raise NotImplementedError
+
+
+@pytest.fixture(params=["data", "data_missing"])
+def all_data(request, data, data_missing):
+    """Parametrized fixture giving 'data' and 'data_missing'"""
+    if request.param == "data":
+        return data
+    elif request.param == "data_missing":
+        return data_missing
+
+
+@pytest.fixture
+def data_repeated(data):
+    """
+    Generate many datasets.
+
+    Parameters
+    ----------
+    data : fixture implementing `data`
+
+    Returns
+    -------
+    Callable[[int], Generator]:
+        A callable that takes a `count` argument and
+        returns a generator yielding `count` datasets.
+    """
+
+    def gen(count):
+        for _ in range(count):
+            yield data
+
+    return gen
+
+
+@pytest.fixture
+def data_for_sorting():
+    """
+    Length-3 array with a known sort order.
+
+    This should be three items [B, C, A] with
+    A < B < C
+
+    For boolean dtypes (for which there are only 2 values available),
+    set B=C=True
+    """
+    raise NotImplementedError
+
+
+@pytest.fixture
+def data_missing_for_sorting():
+    """
+    Length-3 array with a known sort order.
+
+    This should be three items [B, NA, A] with
+    A < B and NA missing.
+    """
+    raise NotImplementedError
+
+
+@pytest.fixture
+def na_cmp():
+    """
+    Binary operator for comparing NA values.
+
+    Should return a function of two arguments that returns
+    True if both arguments are (scalar) NA for your type.
+
+    By default, uses ``operator.is_``
+    """
+    return operator.is_
+
+
+@pytest.fixture
+def na_value(dtype):
+    """
+    The scalar missing value for this type. Default dtype.na_value.
+
+    TODO: can be removed in 3.x (see https://github.com/pandas-dev/pandas/pull/54930)
+    """
+    return dtype.na_value
+
+
+@pytest.fixture
+def data_for_grouping():
+    """
+    Data for factorization, grouping, and unique tests.
+
+    Expected to be like [B, B, NA, NA, A, A, B, C]
+
+    Where A < B < C and NA is missing.
+
+    If a dtype has _is_boolean = True, i.e. only 2 unique non-NA entries,
+    then set C=B.
+    """
+    raise NotImplementedError
+
+
+@pytest.fixture(params=[True, False])
+def box_in_series(request):
+    """Whether to box the data in a Series"""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        lambda x: 1,
+        lambda x: [1] * len(x),
+        lambda x: Series([1] * len(x)),
+        lambda x: x,
+    ],
+    ids=["scalar", "list", "series", "object"],
+)
+def groupby_apply_op(request):
+    """
+    Functions to test groupby.apply().
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def as_frame(request):
+    """
+    Boolean fixture to support Series and Series.to_frame() comparison testing.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def as_series(request):
+    """
+    Boolean fixture to support arr and Series(arr) comparison testing.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def use_numpy(request):
+    """
+    Boolean fixture to support comparison testing of ExtensionDtype array
+    and numpy array.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["ffill", "bfill"])
+def fillna_method(request):
+    """
+    Parametrized fixture giving method parameters 'ffill' and 'bfill' for
+    Series.<method> testing.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def as_array(request):
+    """
+    Boolean fixture to support ExtensionDtype _from_sequence method testing.
+    """
+    return request.param
+
+
+@pytest.fixture
+def invalid_scalar(data):
+    """
+    A scalar that *cannot* be held by this ExtensionArray.
+
+    The default should work for most subclasses, but is not guaranteed.
+
+    If the array can hold any item (i.e. object dtype), then use pytest.skip.
+    """
+    return object.__new__(object)
diff --git a/pandas/pandas/tests/extension/date/__init__.py b/pandas/pandas/tests/extension/date/__init__.py
new file mode 100644
index 0000000000000..2a8c7e9f57a5d
--- /dev/null
+++ b/pandas/pandas/tests/extension/date/__init__.py
@@ -0,0 +1,6 @@
+from pandas.tests.extension.date.array import (
+    DateArray,
+    DateDtype,
+)
+
+__all__ = ["DateArray", "DateDtype"]
diff --git a/pandas/pandas/tests/extension/date/array.py b/pandas/pandas/tests/extension/date/array.py
new file mode 100644
index 0000000000000..2306f5974ba18
--- /dev/null
+++ b/pandas/pandas/tests/extension/date/array.py
@@ -0,0 +1,188 @@
+from __future__ import annotations
+
+import datetime as dt
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+
+import numpy as np
+
+from pandas.core.dtypes.dtypes import register_extension_dtype
+
+from pandas.api.extensions import (
+    ExtensionArray,
+    ExtensionDtype,
+)
+from pandas.api.types import pandas_dtype
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from pandas._typing import (
+        Dtype,
+        PositionalIndexer,
+    )
+
+
+@register_extension_dtype
+class DateDtype(ExtensionDtype):
+    @property
+    def type(self):
+        return dt.date
+
+    @property
+    def name(self):
+        return "DateDtype"
+
+    @classmethod
+    def construct_from_string(cls, string: str):
+        if not isinstance(string, str):
+            raise TypeError(
+                f"'construct_from_string' expects a string, got {type(string)}"
+            )
+
+        if string == cls.__name__:
+            return cls()
+        else:
+            raise TypeError(f"Cannot construct a '{cls.__name__}' from '{string}'")
+
+    @classmethod
+    def construct_array_type(cls):
+        return DateArray
+
+    @property
+    def na_value(self):
+        return dt.date.min
+
+    def __repr__(self) -> str:
+        return self.name
+
+
+class DateArray(ExtensionArray):
+    def __init__(
+        self,
+        dates: (
+            dt.date
+            | Sequence[dt.date]
+            | tuple[np.ndarray, np.ndarray, np.ndarray]
+            | np.ndarray
+        ),
+    ) -> None:
+        if isinstance(dates, dt.date):
+            self._year = np.array([dates.year])
+            self._month = np.array([dates.month])
+            self._day = np.array([dates.year])
+            return
+
+        ldates = len(dates)
+        if isinstance(dates, list):
+            # pre-allocate the arrays since we know the size before hand
+            self._year = np.zeros(ldates, dtype=np.uint16)  # 65535 (0, 9999)
+            self._month = np.zeros(ldates, dtype=np.uint8)  # 255 (1, 31)
+            self._day = np.zeros(ldates, dtype=np.uint8)  # 255 (1, 12)
+            # populate them
+            for i, (y, m, d) in enumerate(
+                (date.year, date.month, date.day) for date in dates
+            ):
+                self._year[i] = y
+                self._month[i] = m
+                self._day[i] = d
+
+        elif isinstance(dates, tuple):
+            # only support triples
+            if ldates != 3:
+                raise ValueError("only triples are valid")
+            # check if all elements have the same type
+            if any(not isinstance(x, np.ndarray) for x in dates):
+                raise TypeError("invalid type")
+            ly, lm, ld = (len(cast(np.ndarray, d)) for d in dates)
+            if not ly == lm == ld:
+                raise ValueError(
+                    f"tuple members must have the same length: {(ly, lm, ld)}"
+                )
+            self._year = dates[0].astype(np.uint16)
+            self._month = dates[1].astype(np.uint8)
+            self._day = dates[2].astype(np.uint8)
+
+        elif isinstance(dates, np.ndarray) and dates.dtype == "U10":
+            self._year = np.zeros(ldates, dtype=np.uint16)  # 65535 (0, 9999)
+            self._month = np.zeros(ldates, dtype=np.uint8)  # 255 (1, 31)
+            self._day = np.zeros(ldates, dtype=np.uint8)  # 255 (1, 12)
+
+            # error: "object_" object is not iterable
+            obj = np.char.split(dates, sep="-")
+            for (i,), (y, m, d) in np.ndenumerate(obj):  # type: ignore[misc]
+                self._year[i] = int(y)
+                self._month[i] = int(m)
+                self._day[i] = int(d)
+
+        else:
+            raise TypeError(f"{type(dates)} is not supported")
+
+    @property
+    def dtype(self) -> ExtensionDtype:
+        return DateDtype()
+
+    def astype(self, dtype, copy=True):
+        dtype = pandas_dtype(dtype)
+
+        if isinstance(dtype, DateDtype):
+            data = self.copy() if copy else self
+        else:
+            data = self.to_numpy(dtype=dtype, copy=copy, na_value=dt.date.min)
+
+        return data
+
+    @property
+    def nbytes(self) -> int:
+        return self._year.nbytes + self._month.nbytes + self._day.nbytes
+
+    def __len__(self) -> int:
+        return len(self._year)  # all 3 arrays are enforced to have the same length
+
+    def __getitem__(self, item: PositionalIndexer):
+        if isinstance(item, int):
+            return dt.date(self._year[item], self._month[item], self._day[item])
+        else:
+            raise NotImplementedError("only ints are supported as indexes")
+
+    def __setitem__(self, key: int | slice | np.ndarray, value: Any) -> None:
+        if not isinstance(key, int):
+            raise NotImplementedError("only ints are supported as indexes")
+
+        if not isinstance(value, dt.date):
+            raise TypeError("you can only set datetime.date types")
+
+        self._year[key] = value.year
+        self._month[key] = value.month
+        self._day[key] = value.day
+
+    def __repr__(self) -> str:
+        return f"DateArray{list(zip(self._year, self._month, self._day))}"
+
+    def copy(self) -> DateArray:
+        return DateArray((self._year.copy(), self._month.copy(), self._day.copy()))
+
+    def isna(self) -> np.ndarray:
+        return np.logical_and(
+            np.logical_and(
+                self._year == dt.date.min.year, self._month == dt.date.min.month
+            ),
+            self._day == dt.date.min.day,
+        )
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype: Dtype | None = None, copy=False):
+        if isinstance(scalars, dt.date):
+            raise TypeError
+        elif isinstance(scalars, DateArray):
+            if dtype is not None:
+                return scalars.astype(dtype, copy=copy)
+            if copy:
+                return scalars.copy()
+            return scalars[:]
+        elif isinstance(scalars, np.ndarray):
+            scalars = scalars.astype("U10")  # 10 chars for yyyy-mm-dd
+            return DateArray(scalars)
diff --git a/pandas/pandas/tests/extension/decimal/__init__.py b/pandas/pandas/tests/extension/decimal/__init__.py
new file mode 100644
index 0000000000000..47b1c7c57a47a
--- /dev/null
+++ b/pandas/pandas/tests/extension/decimal/__init__.py
@@ -0,0 +1,8 @@
+from pandas.tests.extension.decimal.array import (
+    DecimalArray,
+    DecimalDtype,
+    make_data,
+    to_decimal,
+)
+
+__all__ = ["DecimalArray", "DecimalDtype", "make_data", "to_decimal"]
diff --git a/pandas/pandas/tests/extension/decimal/array.py b/pandas/pandas/tests/extension/decimal/array.py
new file mode 100644
index 0000000000000..2ee6a73ec4054
--- /dev/null
+++ b/pandas/pandas/tests/extension/decimal/array.py
@@ -0,0 +1,303 @@
+from __future__ import annotations
+
+import decimal
+import numbers
+import sys
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.base import ExtensionDtype
+from pandas.core.dtypes.common import (
+    is_dtype_equal,
+    is_float,
+    is_integer,
+    pandas_dtype,
+)
+
+import pandas as pd
+from pandas.api.extensions import (
+    no_default,
+    register_extension_dtype,
+)
+from pandas.api.types import (
+    is_list_like,
+    is_scalar,
+)
+from pandas.core import arraylike
+from pandas.core.algorithms import value_counts_internal as value_counts
+from pandas.core.arraylike import OpsMixin
+from pandas.core.arrays import (
+    ExtensionArray,
+    ExtensionScalarOpsMixin,
+)
+from pandas.core.indexers import check_array_indexer
+
+if TYPE_CHECKING:
+    from pandas._typing import type_t
+
+
+@register_extension_dtype
+class DecimalDtype(ExtensionDtype):
+    type = decimal.Decimal
+    name = "decimal"
+    na_value = decimal.Decimal("NaN")
+    _metadata = ("context",)
+
+    def __init__(self, context=None) -> None:
+        self.context = context or decimal.getcontext()
+
+    def __repr__(self) -> str:
+        return f"DecimalDtype(context={self.context})"
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[DecimalArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return DecimalArray
+
+    @property
+    def _is_numeric(self) -> bool:
+        return True
+
+
+class DecimalArray(OpsMixin, ExtensionScalarOpsMixin, ExtensionArray):
+    __array_priority__ = 1000
+
+    def __init__(self, values, dtype=None, copy=False, context=None) -> None:
+        for i, val in enumerate(values):
+            if is_float(val) or is_integer(val):
+                if np.isnan(val):
+                    values[i] = DecimalDtype.na_value
+                else:
+                    # error: Argument 1 has incompatible type "float | int |
+                    # integer[Any]"; expected "Decimal | float | str | tuple[int,
+                    # Sequence[int], int]"
+                    values[i] = DecimalDtype.type(val)  # type: ignore[arg-type]
+            elif not isinstance(val, decimal.Decimal):
+                raise TypeError("All values must be of type " + str(decimal.Decimal))
+        values = np.asarray(values, dtype=object)
+
+        self._data = values
+        # Some aliases for common attribute names to ensure pandas supports
+        # these
+        self._items = self.data = self._data
+        # those aliases are currently not working due to assumptions
+        # in internal code (GH-20735)
+        # self._values = self.values = self.data
+        self._dtype = DecimalDtype(context)
+
+    @property
+    def dtype(self):
+        return self._dtype
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        return cls(scalars)
+
+    @classmethod
+    def _from_sequence_of_strings(cls, strings, *, dtype: ExtensionDtype, copy=False):
+        return cls._from_sequence(
+            [decimal.Decimal(x) for x in strings], dtype=dtype, copy=copy
+        )
+
+    @classmethod
+    def _from_factorized(cls, values, original):
+        return cls(values)
+
+    _HANDLED_TYPES = (decimal.Decimal, numbers.Number, np.ndarray)
+
+    def to_numpy(
+        self,
+        dtype=None,
+        copy: bool = False,
+        na_value: object = no_default,
+        decimals=None,
+    ) -> np.ndarray:
+        result = np.asarray(self, dtype=dtype)
+        if decimals is not None:
+            result = np.asarray([round(x, decimals) for x in result])
+        return result
+
+    def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
+        if not all(
+            isinstance(t, self._HANDLED_TYPES + (DecimalArray,)) for t in inputs
+        ):
+            return NotImplemented
+
+        result = arraylike.maybe_dispatch_ufunc_to_dunder_op(
+            self, ufunc, method, *inputs, **kwargs
+        )
+        if result is not NotImplemented:
+            # e.g. test_array_ufunc_series_scalar_other
+            return result
+
+        if "out" in kwargs:
+            return arraylike.dispatch_ufunc_with_out(
+                self, ufunc, method, *inputs, **kwargs
+            )
+
+        inputs = tuple(x._data if isinstance(x, DecimalArray) else x for x in inputs)
+        result = getattr(ufunc, method)(*inputs, **kwargs)
+
+        if method == "reduce":
+            result = arraylike.dispatch_reduction_ufunc(
+                self, ufunc, method, *inputs, **kwargs
+            )
+            if result is not NotImplemented:
+                return result
+
+        def reconstruct(x):
+            if isinstance(x, (decimal.Decimal, numbers.Number)):
+                return x
+            else:
+                return type(self)._from_sequence(x, dtype=self.dtype)
+
+        if ufunc.nout > 1:
+            return tuple(reconstruct(x) for x in result)
+        else:
+            return reconstruct(result)
+
+    def __getitem__(self, item):
+        if isinstance(item, numbers.Integral):
+            return self._data[item]
+        else:
+            # array, slice.
+            item = pd.api.indexers.check_array_indexer(self, item)
+            return type(self)(self._data[item])
+
+    def take(self, indexer, allow_fill=False, fill_value=None):
+        from pandas.api.extensions import take
+
+        data = self._data
+        if allow_fill and fill_value is None:
+            fill_value = self.dtype.na_value
+
+        result = take(data, indexer, fill_value=fill_value, allow_fill=allow_fill)
+        return self._from_sequence(result, dtype=self.dtype)
+
+    def copy(self):
+        return type(self)(self._data.copy(), dtype=self.dtype)
+
+    def astype(self, dtype, copy=True):
+        if is_dtype_equal(dtype, self._dtype):
+            if not copy:
+                return self
+        dtype = pandas_dtype(dtype)
+        if isinstance(dtype, type(self.dtype)):
+            return type(self)(self._data, copy=copy, context=dtype.context)
+
+        return super().astype(dtype, copy=copy)
+
+    def __setitem__(self, key, value) -> None:
+        if is_list_like(value):
+            if is_scalar(key):
+                raise ValueError("setting an array element with a sequence.")
+            value = [decimal.Decimal(v) for v in value]
+        else:
+            value = decimal.Decimal(value)
+
+        key = check_array_indexer(self, key)
+        self._data[key] = value
+
+    def __len__(self) -> int:
+        return len(self._data)
+
+    def __contains__(self, item) -> bool | np.bool_:
+        if not isinstance(item, decimal.Decimal):
+            return False
+        elif item.is_nan():
+            return self.isna().any()
+        else:
+            return super().__contains__(item)
+
+    @property
+    def nbytes(self) -> int:
+        n = len(self)
+        if n:
+            return n * sys.getsizeof(self[0])
+        return 0
+
+    def isna(self):
+        return np.array([x.is_nan() for x in self._data], dtype=bool)
+
+    @property
+    def _na_value(self):
+        return decimal.Decimal("NaN")
+
+    def _formatter(self, boxed=False):
+        if boxed:
+            return "Decimal: {}".format
+        return repr
+
+    @classmethod
+    def _concat_same_type(cls, to_concat):
+        return cls(np.concatenate([x._data for x in to_concat]))
+
+    def _reduce(
+        self, name: str, *, skipna: bool = True, keepdims: bool = False, **kwargs
+    ):
+        if skipna and self.isna().any():
+            # If we don't have any NAs, we can ignore skipna
+            other = self[~self.isna()]
+            result = other._reduce(name, **kwargs)
+        elif name == "sum" and len(self) == 0:
+            # GH#29630 avoid returning int 0 or np.bool_(False) on old numpy
+            result = decimal.Decimal(0)
+        else:
+            try:
+                op = getattr(self.data, name)
+            except AttributeError as err:
+                raise NotImplementedError(
+                    f"decimal does not support the {name} operation"
+                ) from err
+            result = op(axis=0)
+
+        if keepdims:
+            return type(self)([result])
+        else:
+            return result
+
+    def _cmp_method(self, other, op):
+        # For use with OpsMixin
+        def convert_values(param):
+            if isinstance(param, ExtensionArray) or is_list_like(param):
+                ovalues = param
+            else:
+                # Assume it's an object
+                ovalues = [param] * len(self)
+            return ovalues
+
+        lvalues = self
+        rvalues = convert_values(other)
+
+        # If the operator is not defined for the underlying objects,
+        # a TypeError should be raised
+        res = [op(a, b) for (a, b) in zip(lvalues, rvalues)]
+
+        return np.asarray(res, dtype=bool)
+
+    def value_counts(self, dropna: bool = True):
+        return value_counts(self.to_numpy(), dropna=dropna)
+
+    # We override fillna here to simulate a 3rd party EA that has done so. This
+    #  lets us test a 3rd-party EA that has not yet updated to include a "copy"
+    #  keyword in its fillna method.
+    def fillna(self, value=None, limit=None):
+        return super().fillna(value=value, limit=limit, copy=True)
+
+
+def to_decimal(values, context=None):
+    return DecimalArray([decimal.Decimal(x) for x in values], context=context)
+
+
+def make_data():
+    return [decimal.Decimal(val) for val in np.random.default_rng(2).random(100)]
+
+
+DecimalArray._add_arithmetic_ops()
diff --git a/pandas/pandas/tests/extension/decimal/test_decimal.py b/pandas/pandas/tests/extension/decimal/test_decimal.py
new file mode 100644
index 0000000000000..e0b35b7450303
--- /dev/null
+++ b/pandas/pandas/tests/extension/decimal/test_decimal.py
@@ -0,0 +1,477 @@
+from __future__ import annotations
+
+import decimal
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.extension import base
+from pandas.tests.extension.decimal.array import (
+    DecimalArray,
+    DecimalDtype,
+    make_data,
+    to_decimal,
+)
+
+
+@pytest.fixture
+def dtype():
+    return DecimalDtype()
+
+
+@pytest.fixture
+def data():
+    return DecimalArray(make_data())
+
+
+@pytest.fixture
+def data_for_twos():
+    return DecimalArray([decimal.Decimal(2) for _ in range(100)])
+
+
+@pytest.fixture
+def data_missing():
+    return DecimalArray([decimal.Decimal("NaN"), decimal.Decimal(1)])
+
+
+@pytest.fixture
+def data_for_sorting():
+    return DecimalArray(
+        [decimal.Decimal("1"), decimal.Decimal("2"), decimal.Decimal("0")]
+    )
+
+
+@pytest.fixture
+def data_missing_for_sorting():
+    return DecimalArray(
+        [decimal.Decimal("1"), decimal.Decimal("NaN"), decimal.Decimal("0")]
+    )
+
+
+@pytest.fixture
+def na_cmp():
+    return lambda x, y: x.is_nan() and y.is_nan()
+
+
+@pytest.fixture
+def data_for_grouping():
+    b = decimal.Decimal("1.0")
+    a = decimal.Decimal("0.0")
+    c = decimal.Decimal("2.0")
+    na = decimal.Decimal("NaN")
+    return DecimalArray([b, b, na, na, a, a, b, c])
+
+
+class TestDecimalArray(base.ExtensionTests):
+    def _get_expected_exception(
+        self, op_name: str, obj, other
+    ) -> type[Exception] | tuple[type[Exception], ...] | None:
+        return None
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        if op_name in ["kurt", "sem"]:
+            return False
+        return True
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        if op_name == "count":
+            return super().check_reduce(ser, op_name, skipna)
+        else:
+            result = getattr(ser, op_name)(skipna=skipna)
+            expected = getattr(np.asarray(ser), op_name)()
+            tm.assert_almost_equal(result, expected)
+
+    def test_reduce_series_numeric(self, data, all_numeric_reductions, skipna, request):
+        if all_numeric_reductions in ["kurt", "skew", "sem", "median"]:
+            mark = pytest.mark.xfail(raises=NotImplementedError)
+            request.applymarker(mark)
+        super().test_reduce_series_numeric(data, all_numeric_reductions, skipna)
+
+    def test_reduce_frame(self, data, all_numeric_reductions, skipna, request):
+        op_name = all_numeric_reductions
+        if op_name in ["skew", "median"]:
+            mark = pytest.mark.xfail(raises=NotImplementedError)
+            request.applymarker(mark)
+
+        return super().test_reduce_frame(data, all_numeric_reductions, skipna)
+
+    def test_compare_scalar(self, data, comparison_op):
+        ser = pd.Series(data)
+        self._compare_other(ser, data, comparison_op, 0.5)
+
+    def test_compare_array(self, data, comparison_op):
+        ser = pd.Series(data)
+
+        alter = np.random.default_rng(2).choice([-1, 0, 1], len(data))
+        # Randomly double, halve or keep same value
+        other = pd.Series(data) * [decimal.Decimal(pow(2.0, i)) for i in alter]
+        self._compare_other(ser, data, comparison_op, other)
+
+    def test_arith_series_with_array(self, data, all_arithmetic_operators):
+        op_name = all_arithmetic_operators
+        ser = pd.Series(data)
+
+        context = decimal.getcontext()
+        divbyzerotrap = context.traps[decimal.DivisionByZero]
+        invalidoptrap = context.traps[decimal.InvalidOperation]
+        context.traps[decimal.DivisionByZero] = 0
+        context.traps[decimal.InvalidOperation] = 0
+
+        # Decimal supports ops with int, but not float
+        other = pd.Series([int(d * 100) for d in data])
+        self.check_opname(ser, op_name, other)
+
+        if "mod" not in op_name:
+            self.check_opname(ser, op_name, ser * 2)
+
+        self.check_opname(ser, op_name, 0)
+        self.check_opname(ser, op_name, 5)
+        context.traps[decimal.DivisionByZero] = divbyzerotrap
+        context.traps[decimal.InvalidOperation] = invalidoptrap
+
+    def test_fillna_frame(self, data_missing):
+        msg = "ExtensionArray.fillna added a 'copy' keyword"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            super().test_fillna_frame(data_missing)
+
+    def test_fillna_series(self, data_missing):
+        msg = "ExtensionArray.fillna added a 'copy' keyword"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            super().test_fillna_series(data_missing)
+
+    def test_fillna_with_none(self, data_missing):
+        # GH#57723
+        # EAs that don't have special logic for None will raise, unlike pandas'
+        # which interpret None as the NA value for the dtype.
+        msg = "conversion from NoneType to Decimal is not supported"
+        with pytest.raises(TypeError, match=msg):
+            super().test_fillna_with_none(data_missing)
+
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        msg = "ExtensionArray.fillna added a 'copy' keyword"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            super().test_fillna_limit_frame(data_missing)
+
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        msg = "ExtensionArray.fillna added a 'copy' keyword"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            super().test_fillna_limit_series(data_missing)
+
+    @pytest.mark.parametrize("dropna", [True, False])
+    def test_value_counts(self, all_data, dropna):
+        all_data = all_data[:10]
+        if dropna:
+            other = np.array(all_data[~all_data.isna()])
+        else:
+            other = all_data
+
+        vcs = pd.Series(all_data).value_counts(dropna=dropna)
+        vcs_ex = pd.Series(other).value_counts(dropna=dropna)
+
+        with decimal.localcontext() as ctx:
+            # avoid raising when comparing Decimal("NAN") < Decimal(2)
+            ctx.traps[decimal.InvalidOperation] = False
+
+            result = vcs.sort_index()
+            expected = vcs_ex.sort_index()
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_repr(self, data):
+        # Overriding this base test to explicitly test that
+        # the custom _formatter is used
+        ser = pd.Series(data)
+        assert data.dtype.name in repr(ser)
+        assert "Decimal: " in repr(ser)
+
+    @pytest.mark.xfail(reason="Inconsistent array-vs-scalar behavior")
+    @pytest.mark.parametrize("ufunc", [np.positive, np.negative, np.abs])
+    def test_unary_ufunc_dunder_equivalence(self, data, ufunc):
+        super().test_unary_ufunc_dunder_equivalence(data, ufunc)
+
+
+def test_take_na_value_other_decimal():
+    arr = DecimalArray([decimal.Decimal("1.0"), decimal.Decimal("2.0")])
+    result = arr.take([0, -1], allow_fill=True, fill_value=decimal.Decimal("-1.0"))
+    expected = DecimalArray([decimal.Decimal("1.0"), decimal.Decimal("-1.0")])
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_series_constructor_coerce_data_to_extension_dtype():
+    dtype = DecimalDtype()
+    ser = pd.Series([0, 1, 2], dtype=dtype)
+
+    arr = DecimalArray(
+        [decimal.Decimal(0), decimal.Decimal(1), decimal.Decimal(2)],
+        dtype=dtype,
+    )
+    exp = pd.Series(arr)
+    tm.assert_series_equal(ser, exp)
+
+
+def test_series_constructor_with_dtype():
+    arr = DecimalArray([decimal.Decimal("10.0")])
+    result = pd.Series(arr, dtype=DecimalDtype())
+    expected = pd.Series(arr)
+    tm.assert_series_equal(result, expected)
+
+    result = pd.Series(arr, dtype="int64")
+    expected = pd.Series([10])
+    tm.assert_series_equal(result, expected)
+
+
+def test_dataframe_constructor_with_dtype():
+    arr = DecimalArray([decimal.Decimal("10.0")])
+
+    result = pd.DataFrame({"A": arr}, dtype=DecimalDtype())
+    expected = pd.DataFrame({"A": arr})
+    tm.assert_frame_equal(result, expected)
+
+    arr = DecimalArray([decimal.Decimal("10.0")])
+    result = pd.DataFrame({"A": arr}, dtype="int64")
+    expected = pd.DataFrame({"A": [10]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_astype_dispatches(frame_or_series):
+    # This is a dtype-specific test that ensures Series[decimal].astype
+    # gets all the way through to ExtensionArray.astype
+    # Designing a reliable smoke test that works for arbitrary data types
+    # is difficult.
+    data = pd.Series(DecimalArray([decimal.Decimal(2)]), name="a")
+    ctx = decimal.Context()
+    ctx.prec = 5
+
+    data = frame_or_series(data)
+
+    result = data.astype(DecimalDtype(ctx))
+
+    if frame_or_series is pd.DataFrame:
+        result = result["a"]
+
+    assert result.dtype.context.prec == ctx.prec
+
+
+class DecimalArrayWithoutFromSequence(DecimalArray):
+    """Helper class for testing error handling in _from_sequence."""
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        raise KeyError("For the test")
+
+
+class DecimalArrayWithoutCoercion(DecimalArrayWithoutFromSequence):
+    @classmethod
+    def _create_arithmetic_method(cls, op):
+        return cls._create_method(op, coerce_to_dtype=False)
+
+
+DecimalArrayWithoutCoercion._add_arithmetic_ops()
+
+
+def test_combine_from_sequence_raises(monkeypatch):
+    # https://github.com/pandas-dev/pandas/issues/22850
+    cls = DecimalArrayWithoutFromSequence
+
+    @classmethod
+    def construct_array_type(cls):
+        return DecimalArrayWithoutFromSequence
+
+    monkeypatch.setattr(DecimalDtype, "construct_array_type", construct_array_type)
+
+    arr = cls([decimal.Decimal("1.0"), decimal.Decimal("2.0")])
+    ser = pd.Series(arr)
+    result = ser.combine(ser, operator.add)
+
+    # note: object dtype
+    expected = pd.Series(
+        [decimal.Decimal("2.0"), decimal.Decimal("4.0")], dtype="object"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "class_", [DecimalArrayWithoutFromSequence, DecimalArrayWithoutCoercion]
+)
+def test_scalar_ops_from_sequence_raises(class_):
+    # op(EA, EA) should return an EA, or an ndarray if it's not possible
+    # to return an EA with the return values.
+    arr = class_([decimal.Decimal("1.0"), decimal.Decimal("2.0")])
+    result = arr + arr
+    expected = np.array(
+        [decimal.Decimal("2.0"), decimal.Decimal("4.0")], dtype="object"
+    )
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "reverse, expected_div, expected_mod",
+    [(False, [0, 1, 1, 2], [1, 0, 1, 0]), (True, [2, 1, 0, 0], [0, 0, 2, 2])],
+)
+def test_divmod_array(reverse, expected_div, expected_mod):
+    # https://github.com/pandas-dev/pandas/issues/22930
+    arr = to_decimal([1, 2, 3, 4])
+    if reverse:
+        div, mod = divmod(2, arr)
+    else:
+        div, mod = divmod(arr, 2)
+    expected_div = to_decimal(expected_div)
+    expected_mod = to_decimal(expected_mod)
+
+    tm.assert_extension_array_equal(div, expected_div)
+    tm.assert_extension_array_equal(mod, expected_mod)
+
+
+def test_ufunc_fallback(data):
+    a = data[:5]
+    s = pd.Series(a, index=range(3, 8))
+    result = np.abs(s)
+    expected = pd.Series(np.abs(a), index=range(3, 8))
+    tm.assert_series_equal(result, expected)
+
+
+def test_array_ufunc():
+    a = to_decimal([1, 2, 3])
+    result = np.exp(a)
+    expected = to_decimal(np.exp(a._data))
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_array_ufunc_series():
+    a = to_decimal([1, 2, 3])
+    s = pd.Series(a)
+    result = np.exp(s)
+    expected = pd.Series(to_decimal(np.exp(a._data)))
+    tm.assert_series_equal(result, expected)
+
+
+def test_array_ufunc_series_scalar_other():
+    # check _HANDLED_TYPES
+    a = to_decimal([1, 2, 3])
+    s = pd.Series(a)
+    result = np.add(s, decimal.Decimal(1))
+    expected = pd.Series(np.add(a, decimal.Decimal(1)))
+    tm.assert_series_equal(result, expected)
+
+
+def test_array_ufunc_series_defer():
+    a = to_decimal([1, 2, 3])
+    s = pd.Series(a)
+
+    expected = pd.Series(to_decimal([2, 4, 6]))
+    r1 = np.add(s, a)
+    r2 = np.add(a, s)
+
+    tm.assert_series_equal(r1, expected)
+    tm.assert_series_equal(r2, expected)
+
+
+def test_groupby_agg():
+    # Ensure that the result of agg is inferred to be decimal dtype
+    # https://github.com/pandas-dev/pandas/issues/29141
+
+    data = make_data()[:5]
+    df = pd.DataFrame(
+        {"id1": [0, 0, 0, 1, 1], "id2": [0, 1, 0, 1, 1], "decimals": DecimalArray(data)}
+    )
+
+    # single key, selected column
+    expected = pd.Series(to_decimal([data[0], data[3]]))
+    result = df.groupby("id1")["decimals"].agg(lambda x: x.iloc[0])
+    tm.assert_series_equal(result, expected, check_names=False)
+    result = df["decimals"].groupby(df["id1"]).agg(lambda x: x.iloc[0])
+    tm.assert_series_equal(result, expected, check_names=False)
+
+    # multiple keys, selected column
+    expected = pd.Series(
+        to_decimal([data[0], data[1], data[3]]),
+        index=pd.MultiIndex.from_tuples([(0, 0), (0, 1), (1, 1)]),
+    )
+    result = df.groupby(["id1", "id2"])["decimals"].agg(lambda x: x.iloc[0])
+    tm.assert_series_equal(result, expected, check_names=False)
+    result = df["decimals"].groupby([df["id1"], df["id2"]]).agg(lambda x: x.iloc[0])
+    tm.assert_series_equal(result, expected, check_names=False)
+
+    # multiple columns
+    expected = pd.DataFrame({"id2": [0, 1], "decimals": to_decimal([data[0], data[3]])})
+    result = df.groupby("id1").agg(lambda x: x.iloc[0])
+    tm.assert_frame_equal(result, expected, check_names=False)
+
+
+def test_groupby_agg_ea_method(monkeypatch):
+    # Ensure that the result of agg is inferred to be decimal dtype
+    # https://github.com/pandas-dev/pandas/issues/29141
+
+    def DecimalArray__my_sum(self):
+        return np.sum(np.array(self))
+
+    monkeypatch.setattr(DecimalArray, "my_sum", DecimalArray__my_sum, raising=False)
+
+    data = make_data()[:5]
+    df = pd.DataFrame({"id": [0, 0, 0, 1, 1], "decimals": DecimalArray(data)})
+    expected = pd.Series(to_decimal([data[0] + data[1] + data[2], data[3] + data[4]]))
+
+    result = df.groupby("id")["decimals"].agg(lambda x: x.values.my_sum())
+    tm.assert_series_equal(result, expected, check_names=False)
+    s = pd.Series(DecimalArray(data))
+    grouper = np.array([0, 0, 0, 1, 1], dtype=np.int64)
+    result = s.groupby(grouper).agg(lambda x: x.values.my_sum())
+    tm.assert_series_equal(result, expected, check_names=False)
+
+
+def test_indexing_no_materialize(monkeypatch):
+    # See https://github.com/pandas-dev/pandas/issues/29708
+    # Ensure that indexing operations do not materialize (convert to a numpy
+    # array) the ExtensionArray unnecessary
+
+    def DecimalArray__array__(self, dtype=None):
+        raise Exception("tried to convert a DecimalArray to a numpy array")
+
+    monkeypatch.setattr(DecimalArray, "__array__", DecimalArray__array__, raising=False)
+
+    data = make_data()
+    s = pd.Series(DecimalArray(data))
+    df = pd.DataFrame({"a": s, "b": range(len(s))})
+
+    # ensure the following operations do not raise an error
+    s[s > 0.5]
+    df[s > 0.5]
+    s.at[0]
+    df.at[0, "a"]
+
+
+def test_to_numpy_keyword():
+    # test the extra keyword
+    values = [decimal.Decimal("1.1111"), decimal.Decimal("2.2222")]
+    expected = np.array(
+        [decimal.Decimal("1.11"), decimal.Decimal("2.22")], dtype="object"
+    )
+    a = pd.array(values, dtype="decimal")
+    result = a.to_numpy(decimals=2)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = pd.Series(a).to_numpy(decimals=2)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_array_copy_on_write():
+    df = pd.DataFrame({"a": [decimal.Decimal(2), decimal.Decimal(3)]}, dtype="object")
+    df2 = df.astype(DecimalDtype())
+    df.iloc[0, 0] = 0
+    expected = pd.DataFrame(
+        {"a": [decimal.Decimal(2), decimal.Decimal(3)]}, dtype=DecimalDtype()
+    )
+    tm.assert_equal(df2.values, expected.values)
diff --git a/pandas/pandas/tests/extension/json/__init__.py b/pandas/pandas/tests/extension/json/__init__.py
new file mode 100644
index 0000000000000..7ebfd54a5b0d6
--- /dev/null
+++ b/pandas/pandas/tests/extension/json/__init__.py
@@ -0,0 +1,7 @@
+from pandas.tests.extension.json.array import (
+    JSONArray,
+    JSONDtype,
+    make_data,
+)
+
+__all__ = ["JSONArray", "JSONDtype", "make_data"]
diff --git a/pandas/pandas/tests/extension/json/array.py b/pandas/pandas/tests/extension/json/array.py
new file mode 100644
index 0000000000000..b110911bda400
--- /dev/null
+++ b/pandas/pandas/tests/extension/json/array.py
@@ -0,0 +1,263 @@
+"""
+Test extension array for storing nested data in a pandas container.
+
+The JSONArray stores lists of dictionaries. The storage mechanism is a list,
+not an ndarray.
+
+Note
+----
+We currently store lists of UserDicts. Pandas has a few places
+internally that specifically check for dicts, and does non-scalar things
+in that case. We *want* the dictionaries to be treated as scalars, so we
+hack around pandas by using UserDicts.
+"""
+
+from __future__ import annotations
+
+from collections import (
+    UserDict,
+    abc,
+)
+import itertools
+import numbers
+import string
+import sys
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+
+import numpy as np
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_list_like,
+    pandas_dtype,
+)
+
+import pandas as pd
+from pandas.api.extensions import (
+    ExtensionArray,
+    ExtensionDtype,
+)
+from pandas.core.indexers import unpack_tuple_and_ellipses
+
+if TYPE_CHECKING:
+    from collections.abc import Mapping
+
+    from pandas._typing import type_t
+
+
+class JSONDtype(ExtensionDtype):
+    type = abc.Mapping
+    name = "json"
+    na_value: Mapping[str, Any] = UserDict()
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[JSONArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return JSONArray
+
+
+class JSONArray(ExtensionArray):
+    dtype = JSONDtype()
+    __array_priority__ = 1000
+
+    def __init__(self, values, dtype=None, copy=False) -> None:
+        for val in values:
+            if not isinstance(val, self.dtype.type):
+                raise TypeError("All values must be of type " + str(self.dtype.type))
+        self.data = values
+
+        # Some aliases for common attribute names to ensure pandas supports
+        # these
+        self._items = self._data = self.data
+        # those aliases are currently not working due to assumptions
+        # in internal code (GH-20735)
+        # self._values = self.values = self.data
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        return cls(scalars)
+
+    @classmethod
+    def _from_factorized(cls, values, original):
+        return cls([UserDict(x) for x in values if x != ()])
+
+    def __getitem__(self, item):
+        if isinstance(item, tuple):
+            item = unpack_tuple_and_ellipses(item)
+
+        if isinstance(item, numbers.Integral):
+            return self.data[item]
+        elif isinstance(item, slice) and item == slice(None):
+            # Make sure we get a view
+            return type(self)(self.data)
+        elif isinstance(item, slice):
+            # slice
+            return type(self)(self.data[item])
+        elif not is_list_like(item):
+            # e.g. "foo" or 2.5
+            # exception message copied from numpy
+            raise IndexError(
+                r"only integers, slices (`:`), ellipsis (`...`), numpy.newaxis "
+                r"(`None`) and integer or boolean arrays are valid indices"
+            )
+        else:
+            item = pd.api.indexers.check_array_indexer(self, item)
+            if is_bool_dtype(item.dtype):
+                return type(self)._from_sequence(
+                    [x for x, m in zip(self, item) if m], dtype=self.dtype
+                )
+            # integer
+            return type(self)([self.data[i] for i in item])
+
+    def __setitem__(self, key, value) -> None:
+        if isinstance(key, numbers.Integral):
+            self.data[key] = value
+        else:
+            if not isinstance(value, (type(self), abc.Sequence)):
+                # broadcast value
+                value = itertools.cycle([value])
+
+            if isinstance(key, np.ndarray) and key.dtype == "bool":
+                # masking
+                for i, (k, v) in enumerate(zip(key, value)):
+                    if k:
+                        assert isinstance(v, self.dtype.type)
+                        self.data[i] = v
+            else:
+                for k, v in zip(key, value):
+                    assert isinstance(v, self.dtype.type)
+                    self.data[k] = v
+
+    def __len__(self) -> int:
+        return len(self.data)
+
+    def __eq__(self, other):
+        return NotImplemented
+
+    def __ne__(self, other):
+        return NotImplemented
+
+    def __array__(self, dtype=None, copy=None):
+        if copy is False:
+            raise ValueError(
+                "Unable to avoid copy while creating an array as requested."
+            )
+
+        if dtype is None:
+            dtype = object
+        if dtype == object:
+            # on py38 builds it looks like numpy is inferring to a non-1D array
+            return construct_1d_object_array_from_listlike(list(self))
+        if copy is None:
+            # Note: branch avoids `copy=None` for NumPy 1.x support
+            return np.asarray(self.data, dtype=dtype)
+        return np.asarray(self.data, dtype=dtype, copy=copy)
+
+    @property
+    def nbytes(self) -> int:
+        return sys.getsizeof(self.data)
+
+    def isna(self):
+        return np.array([x == self.dtype.na_value for x in self.data], dtype=bool)
+
+    def take(self, indexer, allow_fill=False, fill_value=None):
+        # re-implement here, since NumPy has trouble setting
+        # sized objects like UserDicts into scalar slots of
+        # an ndarary.
+        indexer = np.asarray(indexer)
+        msg = (
+            "Index is out of bounds or cannot do a non-empty take from an empty array."
+        )
+
+        if allow_fill:
+            if fill_value is None:
+                fill_value = self.dtype.na_value
+            # bounds check
+            if (indexer < -1).any():
+                raise ValueError
+            try:
+                output = [
+                    self.data[loc] if loc != -1 else fill_value for loc in indexer
+                ]
+            except IndexError as err:
+                raise IndexError(msg) from err
+        else:
+            try:
+                output = [self.data[loc] for loc in indexer]
+            except IndexError as err:
+                raise IndexError(msg) from err
+
+        return type(self)._from_sequence(output, dtype=self.dtype)
+
+    def copy(self):
+        return type(self)(self.data[:])
+
+    def astype(self, dtype, copy=True):
+        # NumPy has issues when all the dicts are the same length.
+        # np.array([UserDict(...), UserDict(...)]) fails,
+        # but np.array([{...}, {...}]) works, so cast.
+        from pandas.core.arrays.string_ import StringDtype
+
+        dtype = pandas_dtype(dtype)
+        # needed to add this check for the Series constructor
+        if isinstance(dtype, type(self.dtype)) and dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+        elif isinstance(dtype, StringDtype):
+            arr_cls = dtype.construct_array_type()
+            return arr_cls._from_sequence(self, dtype=dtype, copy=False)
+        elif not copy:
+            return np.asarray([dict(x) for x in self], dtype=dtype)
+        else:
+            return np.array([dict(x) for x in self], dtype=dtype, copy=copy)
+
+    def unique(self):
+        # Parent method doesn't work since np.array will try to infer
+        # a 2-dim object.
+        return type(self)([dict(x) for x in {tuple(d.items()) for d in self.data}])
+
+    @classmethod
+    def _concat_same_type(cls, to_concat):
+        data = list(itertools.chain.from_iterable(x.data for x in to_concat))
+        return cls(data)
+
+    def _values_for_factorize(self):
+        frozen = self._values_for_argsort()
+        if len(frozen) == 0:
+            # factorize_array expects 1-d array, this is a len-0 2-d array.
+            frozen = frozen.ravel()
+        return frozen, ()
+
+    def _values_for_argsort(self):
+        # Bypass NumPy's shape inference to get a (N,) array of tuples.
+        frozen = [tuple(x.items()) for x in self]
+        return construct_1d_object_array_from_listlike(frozen)
+
+    def _pad_or_backfill(self, *, method, limit=None, copy=True):
+        # GH#56616 - test EA method without limit_area argument
+        return super()._pad_or_backfill(method=method, limit=limit, copy=copy)
+
+
+def make_data():
+    # TODO: Use a regular dict. See _NDFrameIndexer._setitem_with_indexer
+    rng = np.random.default_rng(2)
+    return [
+        UserDict(
+            [
+                (rng.choice(list(string.ascii_letters)), rng.integers(0, 100))
+                for _ in range(rng.integers(0, 10))
+            ]
+        )
+        for _ in range(100)
+    ]
diff --git a/pandas/pandas/tests/extension/json/test_json.py b/pandas/pandas/tests/extension/json/test_json.py
new file mode 100644
index 0000000000000..4bc9562f1895d
--- /dev/null
+++ b/pandas/pandas/tests/extension/json/test_json.py
@@ -0,0 +1,491 @@
+import collections
+import operator
+import sys
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.extension import base
+from pandas.tests.extension.json.array import (
+    JSONArray,
+    JSONDtype,
+    make_data,
+)
+
+# We intentionally don't run base.BaseSetitemTests because pandas'
+# internals has trouble setting sequences of values into scalar positions.
+unhashable = pytest.mark.xfail(reason="Unhashable")
+
+
+@pytest.fixture
+def dtype():
+    return JSONDtype()
+
+
+@pytest.fixture
+def data():
+    """Length-100 PeriodArray for semantics test."""
+    data = make_data()
+
+    # Why the while loop? NumPy is unable to construct an ndarray from
+    # equal-length ndarrays. Many of our operations involve coercing the
+    # EA to an ndarray of objects. To avoid random test failures, we ensure
+    # that our data is coercible to an ndarray. Several tests deal with only
+    # the first two elements, so that's what we'll check.
+
+    while len(data[0]) == len(data[1]):
+        data = make_data()
+
+    return JSONArray(data)
+
+
+@pytest.fixture
+def data_missing():
+    """Length 2 array with [NA, Valid]"""
+    return JSONArray([{}, {"a": 10}])
+
+
+@pytest.fixture
+def data_for_sorting():
+    return JSONArray([{"b": 1}, {"c": 4}, {"a": 2, "c": 3}])
+
+
+@pytest.fixture
+def data_missing_for_sorting():
+    return JSONArray([{"b": 1}, {}, {"a": 4}])
+
+
+@pytest.fixture
+def na_cmp():
+    return operator.eq
+
+
+@pytest.fixture
+def data_for_grouping():
+    return JSONArray(
+        [
+            {"b": 1},
+            {"b": 1},
+            {},
+            {},
+            {"a": 0, "c": 2},
+            {"a": 0, "c": 2},
+            {"b": 1},
+            {"c": 2},
+        ]
+    )
+
+
+class TestJSONArray(base.ExtensionTests):
+    @pytest.mark.xfail(
+        reason="comparison method not implemented for JSONArray (GH-37867)"
+    )
+    def test_contains(self, data):
+        # GH-37867
+        super().test_contains(data)
+
+    @pytest.mark.xfail(reason="not implemented constructor from dtype")
+    def test_from_dtype(self, data):
+        # construct from our dtype & string dtype
+        super().test_from_dtype(data)
+
+    @pytest.mark.xfail(reason="RecursionError, GH-33900")
+    def test_series_constructor_no_data_with_index(self, dtype, na_value):
+        # RecursionError: maximum recursion depth exceeded in comparison
+        rec_limit = sys.getrecursionlimit()
+        try:
+            # Limit to avoid stack overflow on Windows CI
+            sys.setrecursionlimit(100)
+            super().test_series_constructor_no_data_with_index(dtype, na_value)
+        finally:
+            sys.setrecursionlimit(rec_limit)
+
+    @pytest.mark.xfail(reason="RecursionError, GH-33900")
+    def test_series_constructor_scalar_na_with_index(self, dtype, na_value):
+        # RecursionError: maximum recursion depth exceeded in comparison
+        rec_limit = sys.getrecursionlimit()
+        try:
+            # Limit to avoid stack overflow on Windows CI
+            sys.setrecursionlimit(100)
+            super().test_series_constructor_scalar_na_with_index(dtype, na_value)
+        finally:
+            sys.setrecursionlimit(rec_limit)
+
+    @pytest.mark.xfail(reason="collection as scalar, GH-33901")
+    def test_series_constructor_scalar_with_index(self, data, dtype):
+        # TypeError: All values must be of type <class 'collections.abc.Mapping'>
+        rec_limit = sys.getrecursionlimit()
+        try:
+            # Limit to avoid stack overflow on Windows CI
+            sys.setrecursionlimit(100)
+            super().test_series_constructor_scalar_with_index(data, dtype)
+        finally:
+            sys.setrecursionlimit(rec_limit)
+
+    @pytest.mark.xfail(reason="Different definitions of NA")
+    def test_stack(self):
+        """
+        The test does .astype(object).stack(). If we happen to have
+        any missing values in `data`, then we'll end up with different
+        rows since we consider `{}` NA, but `.astype(object)` doesn't.
+        """
+        super().test_stack()
+
+    @pytest.mark.xfail(reason="dict for NA")
+    def test_unstack(self, data, index):
+        # The base test has NaN for the expected NA value.
+        # this matches otherwise
+        return super().test_unstack(data, index)
+
+    @pytest.mark.xfail(reason="Setting a dict as a scalar")
+    def test_fillna_series(self):
+        """We treat dictionaries as a mapping in fillna, not a scalar."""
+        super().test_fillna_series()
+
+    @pytest.mark.xfail(reason="Setting a dict as a scalar")
+    def test_fillna_frame(self):
+        """We treat dictionaries as a mapping in fillna, not a scalar."""
+        super().test_fillna_frame()
+
+    def test_fillna_with_none(self, data_missing):
+        # GH#57723
+        # EAs that don't have special logic for None will raise, unlike pandas'
+        # which interpret None as the NA value for the dtype.
+        with pytest.raises(AssertionError):
+            super().test_fillna_with_none(data_missing)
+
+    @pytest.mark.xfail(reason="fill value is a dictionary, takes incorrect code path")
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        super().test_fillna_limit_frame(data_missing)
+
+    @pytest.mark.xfail(reason="fill value is a dictionary, takes incorrect code path")
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        super().test_fillna_limit_frame(data_missing)
+
+    @pytest.mark.parametrize(
+        "limit_area, input_ilocs, expected_ilocs",
+        [
+            ("outside", [1, 0, 0, 0, 1], [1, 0, 0, 0, 1]),
+            ("outside", [1, 0, 1, 0, 1], [1, 0, 1, 0, 1]),
+            ("outside", [0, 1, 1, 1, 0], [0, 1, 1, 1, 1]),
+            ("outside", [0, 1, 0, 1, 0], [0, 1, 0, 1, 1]),
+            ("inside", [1, 0, 0, 0, 1], [1, 1, 1, 1, 1]),
+            ("inside", [1, 0, 1, 0, 1], [1, 1, 1, 1, 1]),
+            ("inside", [0, 1, 1, 1, 0], [0, 1, 1, 1, 0]),
+            ("inside", [0, 1, 0, 1, 0], [0, 1, 1, 1, 0]),
+        ],
+    )
+    def test_ffill_limit_area(
+        self, data_missing, limit_area, input_ilocs, expected_ilocs
+    ):
+        # GH#56616
+        msg = "JSONArray does not implement limit_area"
+        with pytest.raises(NotImplementedError, match=msg):
+            super().test_ffill_limit_area(
+                data_missing, limit_area, input_ilocs, expected_ilocs
+            )
+
+    @unhashable
+    def test_value_counts(self, all_data, dropna):
+        super().test_value_counts(all_data, dropna)
+
+    @unhashable
+    def test_value_counts_with_normalize(self, data):
+        super().test_value_counts_with_normalize(data)
+
+    @unhashable
+    def test_sort_values_frame(self):
+        # TODO (EA.factorize): see if _values_for_factorize allows this.
+        super().test_sort_values_frame()
+
+    @pytest.mark.xfail(reason="combine for JSONArray not supported")
+    def test_combine_le(self, data_repeated):
+        super().test_combine_le(data_repeated)
+
+    @pytest.mark.xfail(
+        reason="combine for JSONArray not supported - "
+        "may pass depending on random data",
+        strict=False,
+        raises=AssertionError,
+    )
+    def test_combine_first(self, data):
+        super().test_combine_first(data)
+
+    @pytest.mark.xfail(reason="broadcasting error")
+    def test_where_series(self, data, na_value):
+        # Fails with
+        # *** ValueError: operands could not be broadcast together
+        # with shapes (4,) (4,) (0,)
+        super().test_where_series(data, na_value)
+
+    @pytest.mark.xfail(reason="Can't compare dicts.")
+    def test_searchsorted(self, data_for_sorting):
+        super().test_searchsorted(data_for_sorting)
+
+    @pytest.mark.xfail(reason="Can't compare dicts.")
+    def test_equals(self, data, na_value, as_series):
+        super().test_equals(data, na_value, as_series)
+
+    @pytest.mark.skip("fill-value is interpreted as a dict of values")
+    def test_fillna_copy_frame(self, data_missing):
+        super().test_fillna_copy_frame(data_missing)
+
+    @pytest.mark.xfail(reason="Fails with CoW")
+    def test_equals_same_data_different_object(self, data):
+        super().test_equals_same_data_different_object(data)
+
+    @pytest.mark.xfail(reason="failing on np.array(self, dtype=str)")
+    def test_astype_str(self):
+        """This currently fails in NumPy on np.array(self, dtype=str) with
+
+        *** ValueError: setting an array element with a sequence
+        """
+        super().test_astype_str()
+
+    @unhashable
+    def test_groupby_extension_transform(self):
+        """
+        This currently fails in Series.name.setter, since the
+        name must be hashable, but the value is a dictionary.
+        I think this is what we want, i.e. `.name` should be the original
+        values, and not the values for factorization.
+        """
+        super().test_groupby_extension_transform()
+
+    @unhashable
+    def test_groupby_extension_apply(self):
+        """
+        This fails in Index._do_unique_check with
+
+        >   hash(val)
+        E   TypeError: unhashable type: 'UserDict' with
+
+        I suspect that once we support Index[ExtensionArray],
+        we'll be able to dispatch unique.
+        """
+        super().test_groupby_extension_apply()
+
+    @unhashable
+    def test_groupby_extension_agg(self):
+        """
+        This fails when we get to tm.assert_series_equal when left.index
+        contains dictionaries, which are not hashable.
+        """
+        super().test_groupby_extension_agg()
+
+    @unhashable
+    def test_groupby_extension_no_sort(self):
+        """
+        This fails when we get to tm.assert_series_equal when left.index
+        contains dictionaries, which are not hashable.
+        """
+        super().test_groupby_extension_no_sort()
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators, request):
+        if len(data[0]) != 1:
+            mark = pytest.mark.xfail(reason="raises in coercing to Series")
+            request.applymarker(mark)
+        super().test_arith_frame_with_scalar(data, all_arithmetic_operators)
+
+    def test_compare_array(self, data, comparison_op, request):
+        if comparison_op.__name__ in ["eq", "ne"]:
+            mark = pytest.mark.xfail(reason="Comparison methods not implemented")
+            request.applymarker(mark)
+        super().test_compare_array(data, comparison_op)
+
+    @pytest.mark.xfail(reason="ValueError: Must have equal len keys and value")
+    def test_setitem_loc_scalar_mixed(self, data):
+        super().test_setitem_loc_scalar_mixed(data)
+
+    @pytest.mark.xfail(reason="ValueError: Must have equal len keys and value")
+    def test_setitem_loc_scalar_multiple_homogoneous(self, data):
+        super().test_setitem_loc_scalar_multiple_homogoneous(data)
+
+    @pytest.mark.xfail(reason="ValueError: Must have equal len keys and value")
+    def test_setitem_iloc_scalar_mixed(self, data):
+        super().test_setitem_iloc_scalar_mixed(data)
+
+    @pytest.mark.xfail(reason="ValueError: Must have equal len keys and value")
+    def test_setitem_iloc_scalar_multiple_homogoneous(self, data):
+        super().test_setitem_iloc_scalar_multiple_homogoneous(data)
+
+    @pytest.mark.parametrize(
+        "mask",
+        [
+            np.array([True, True, True, False, False]),
+            pd.array([True, True, True, False, False], dtype="boolean"),
+            pd.array([True, True, True, pd.NA, pd.NA], dtype="boolean"),
+        ],
+        ids=["numpy-array", "boolean-array", "boolean-array-na"],
+    )
+    def test_setitem_mask(self, data, mask, box_in_series, request):
+        if box_in_series:
+            mark = pytest.mark.xfail(
+                reason="cannot set using a list-like indexer with a different length"
+            )
+            request.applymarker(mark)
+        elif not isinstance(mask, np.ndarray):
+            mark = pytest.mark.xfail(reason="Issues unwanted DeprecationWarning")
+            request.applymarker(mark)
+        super().test_setitem_mask(data, mask, box_in_series)
+
+    def test_setitem_mask_raises(self, data, box_in_series, request):
+        if not box_in_series:
+            mark = pytest.mark.xfail(reason="Fails to raise")
+            request.applymarker(mark)
+
+        super().test_setitem_mask_raises(data, box_in_series)
+
+    @pytest.mark.xfail(
+        reason="cannot set using a list-like indexer with a different length"
+    )
+    def test_setitem_mask_boolean_array_with_na(self, data, box_in_series):
+        super().test_setitem_mask_boolean_array_with_na(data, box_in_series)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2], pd.array([0, 1, 2], dtype="Int64"), np.array([0, 1, 2])],
+        ids=["list", "integer-array", "numpy-array"],
+    )
+    def test_setitem_integer_array(self, data, idx, box_in_series, request):
+        if box_in_series:
+            mark = pytest.mark.xfail(
+                reason="cannot set using a list-like indexer with a different length"
+            )
+            request.applymarker(mark)
+        super().test_setitem_integer_array(data, idx, box_in_series)
+
+    @pytest.mark.xfail(reason="list indices must be integers or slices, not NAType")
+    @pytest.mark.parametrize(
+        "idx, box_in_series",
+        [
+            ([0, 1, 2, pd.NA], False),
+            pytest.param(
+                [0, 1, 2, pd.NA], True, marks=pytest.mark.xfail(reason="GH-31948")
+            ),
+            (pd.array([0, 1, 2, pd.NA], dtype="Int64"), False),
+            (pd.array([0, 1, 2, pd.NA], dtype="Int64"), True),
+        ],
+        ids=["list-False", "list-True", "integer-array-False", "integer-array-True"],
+    )
+    def test_setitem_integer_with_missing_raises(self, data, idx, box_in_series):
+        super().test_setitem_integer_with_missing_raises(data, idx, box_in_series)
+
+    @pytest.mark.xfail(reason="Fails to raise")
+    def test_setitem_scalar_key_sequence_raise(self, data):
+        super().test_setitem_scalar_key_sequence_raise(data)
+
+    def test_setitem_with_expansion_dataframe_column(self, data, full_indexer, request):
+        if "full_slice" in request.node.name:
+            mark = pytest.mark.xfail(reason="slice is not iterable")
+            request.applymarker(mark)
+        super().test_setitem_with_expansion_dataframe_column(data, full_indexer)
+
+    @pytest.mark.xfail(reason="slice is not iterable")
+    def test_setitem_frame_2d_values(self, data):
+        super().test_setitem_frame_2d_values(data)
+
+    @pytest.mark.xfail(
+        reason="cannot set using a list-like indexer with a different length"
+    )
+    @pytest.mark.parametrize("setter", ["loc", None])
+    def test_setitem_mask_broadcast(self, data, setter):
+        super().test_setitem_mask_broadcast(data, setter)
+
+    @pytest.mark.xfail(
+        reason="cannot set using a slice indexer with a different length"
+    )
+    def test_setitem_slice(self, data, box_in_series):
+        super().test_setitem_slice(data, box_in_series)
+
+    @pytest.mark.xfail(reason="slice object is not iterable")
+    def test_setitem_loc_iloc_slice(self, data):
+        super().test_setitem_loc_iloc_slice(data)
+
+    @pytest.mark.xfail(reason="slice object is not iterable")
+    def test_setitem_slice_mismatch_length_raises(self, data):
+        super().test_setitem_slice_mismatch_length_raises(data)
+
+    @pytest.mark.xfail(reason="slice object is not iterable")
+    def test_setitem_slice_array(self, data):
+        super().test_setitem_slice_array(data)
+
+    @pytest.mark.xfail(reason="Fail to raise")
+    def test_setitem_invalid(self, data, invalid_scalar):
+        super().test_setitem_invalid(data, invalid_scalar)
+
+    @pytest.mark.xfail(reason="only integer scalar arrays can be converted")
+    def test_setitem_2d_values(self, data):
+        super().test_setitem_2d_values(data)
+
+    @pytest.mark.xfail(reason="data type 'json' not understood")
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_EA_types(self, engine, data, request):
+        super().test_EA_types(engine, data, request)
+
+
+def custom_assert_series_equal(left, right, *args, **kwargs):
+    # NumPy doesn't handle an array of equal-length UserDicts.
+    # The default assert_series_equal eventually does a
+    # Series.values, which raises. We work around it by
+    # converting the UserDicts to dicts.
+    if left.dtype.name == "json":
+        assert left.dtype == right.dtype
+        left = pd.Series(
+            JSONArray(left.values.astype(object)), index=left.index, name=left.name
+        )
+        right = pd.Series(
+            JSONArray(right.values.astype(object)),
+            index=right.index,
+            name=right.name,
+        )
+    tm.assert_series_equal(left, right, *args, **kwargs)
+
+
+def custom_assert_frame_equal(left, right, *args, **kwargs):
+    obj_type = kwargs.get("obj", "DataFrame")
+    tm.assert_index_equal(
+        left.columns,
+        right.columns,
+        exact=kwargs.get("check_column_type", "equiv"),
+        check_names=kwargs.get("check_names", True),
+        check_exact=kwargs.get("check_exact", False),
+        check_categorical=kwargs.get("check_categorical", True),
+        obj=f"{obj_type}.columns",
+    )
+
+    jsons = (left.dtypes == "json").index
+
+    for col in jsons:
+        custom_assert_series_equal(left[col], right[col], *args, **kwargs)
+
+    left = left.drop(columns=jsons)
+    right = right.drop(columns=jsons)
+    tm.assert_frame_equal(left, right, *args, **kwargs)
+
+
+def test_custom_asserts():
+    # This would always trigger the KeyError from trying to put
+    # an array of equal-length UserDicts inside an ndarray.
+    data = JSONArray(
+        [
+            collections.UserDict({"a": 1}),
+            collections.UserDict({"b": 2}),
+            collections.UserDict({"c": 3}),
+        ]
+    )
+    a = pd.Series(data)
+    custom_assert_series_equal(a, a)
+    custom_assert_frame_equal(a.to_frame(), a.to_frame())
+
+    b = pd.Series(data.take([0, 0, 1]))
+    msg = r"Series are different"
+    with pytest.raises(AssertionError, match=msg):
+        custom_assert_series_equal(a, b)
+
+    with pytest.raises(AssertionError, match=msg):
+        custom_assert_frame_equal(a.to_frame(), b.to_frame())
diff --git a/pandas/pandas/tests/extension/list/__init__.py b/pandas/pandas/tests/extension/list/__init__.py
new file mode 100644
index 0000000000000..0f3f2f3537788
--- /dev/null
+++ b/pandas/pandas/tests/extension/list/__init__.py
@@ -0,0 +1,7 @@
+from pandas.tests.extension.list.array import (
+    ListArray,
+    ListDtype,
+    make_data,
+)
+
+__all__ = ["ListArray", "ListDtype", "make_data"]
diff --git a/pandas/pandas/tests/extension/list/array.py b/pandas/pandas/tests/extension/list/array.py
new file mode 100644
index 0000000000000..8b4728c7d6292
--- /dev/null
+++ b/pandas/pandas/tests/extension/list/array.py
@@ -0,0 +1,137 @@
+"""
+Test extension array for storing nested data in a pandas container.
+
+The ListArray stores an ndarray of lists.
+"""
+
+from __future__ import annotations
+
+import numbers
+import string
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.base import ExtensionDtype
+
+import pandas as pd
+from pandas.api.types import (
+    is_object_dtype,
+    is_string_dtype,
+)
+from pandas.core.arrays import ExtensionArray
+
+if TYPE_CHECKING:
+    from pandas._typing import type_t
+
+
+class ListDtype(ExtensionDtype):
+    type = list
+    name = "list"
+    na_value = np.nan
+
+    @classmethod
+    def construct_array_type(cls) -> type_t[ListArray]:
+        """
+        Return the array type associated with this dtype.
+
+        Returns
+        -------
+        type
+        """
+        return ListArray
+
+
+class ListArray(ExtensionArray):
+    dtype = ListDtype()
+    __array_priority__ = 1000
+
+    def __init__(self, values, dtype=None, copy=False) -> None:
+        if not isinstance(values, np.ndarray):
+            raise TypeError("Need to pass a numpy array as values")
+        for val in values:
+            if not isinstance(val, self.dtype.type) and not pd.isna(val):
+                raise TypeError("All values must be of type " + str(self.dtype.type))
+        self.data = values
+
+    @classmethod
+    def _from_sequence(cls, scalars, *, dtype=None, copy=False):
+        data = np.empty(len(scalars), dtype=object)
+        data[:] = scalars
+        return cls(data)
+
+    def __getitem__(self, item):
+        if isinstance(item, numbers.Integral):
+            return self.data[item]
+        else:
+            # slice, list-like, mask
+            return type(self)(self.data[item])
+
+    def __len__(self) -> int:
+        return len(self.data)
+
+    def isna(self):
+        return np.array(
+            [not isinstance(x, list) and np.isnan(x) for x in self.data], dtype=bool
+        )
+
+    def take(self, indexer, allow_fill=False, fill_value=None):
+        # re-implement here, since NumPy has trouble setting
+        # sized objects like UserDicts into scalar slots of
+        # an ndarary.
+        indexer = np.asarray(indexer)
+        msg = (
+            "Index is out of bounds or cannot do a non-empty take from an empty array."
+        )
+
+        if allow_fill:
+            if fill_value is None:
+                fill_value = self.dtype.na_value
+            # bounds check
+            if (indexer < -1).any():
+                raise ValueError
+            try:
+                output = [
+                    self.data[loc] if loc != -1 else fill_value for loc in indexer
+                ]
+            except IndexError as err:
+                raise IndexError(msg) from err
+        else:
+            try:
+                output = [self.data[loc] for loc in indexer]
+            except IndexError as err:
+                raise IndexError(msg) from err
+
+        return self._from_sequence(output)
+
+    def copy(self):
+        return type(self)(self.data[:])
+
+    def astype(self, dtype, copy=True):
+        if isinstance(dtype, type(self.dtype)) and dtype == self.dtype:
+            if copy:
+                return self.copy()
+            return self
+        elif is_string_dtype(dtype) and not is_object_dtype(dtype):
+            # numpy has problems with astype(str) for nested elements
+            return np.array([str(x) for x in self.data], dtype=dtype)
+        elif not copy:
+            return np.asarray(self.data, dtype=dtype)
+        else:
+            return np.array(self.data, dtype=dtype, copy=copy)
+
+    @classmethod
+    def _concat_same_type(cls, to_concat):
+        data = np.concatenate([x.data for x in to_concat])
+        return cls(data)
+
+
+def make_data():
+    # TODO: Use a regular dict. See _NDFrameIndexer._setitem_with_indexer
+    rng = np.random.default_rng(2)
+    data = np.empty(100, dtype=object)
+    data[:] = [
+        [rng.choice(list(string.ascii_letters)) for _ in range(rng.integers(0, 10))]
+        for _ in range(100)
+    ]
+    return data
diff --git a/pandas/pandas/tests/extension/list/test_list.py b/pandas/pandas/tests/extension/list/test_list.py
new file mode 100644
index 0000000000000..ac396cd3c60d4
--- /dev/null
+++ b/pandas/pandas/tests/extension/list/test_list.py
@@ -0,0 +1,33 @@
+import pytest
+
+import pandas as pd
+from pandas.tests.extension.list.array import (
+    ListArray,
+    ListDtype,
+    make_data,
+)
+
+
+@pytest.fixture
+def dtype():
+    return ListDtype()
+
+
+@pytest.fixture
+def data():
+    """Length-100 ListArray for semantics test."""
+    data = make_data()
+
+    while len(data[0]) == len(data[1]):
+        data = make_data()
+
+    return ListArray(data)
+
+
+def test_to_csv(data):
+    # https://github.com/pandas-dev/pandas/issues/28840
+    # array with list-likes fail when doing astype(str) on the numpy array
+    # which was done in get_values_for_csv
+    df = pd.DataFrame({"a": data})
+    res = df.to_csv()
+    assert str(data[0]) in res
diff --git a/pandas/pandas/tests/extension/test_arrow.py b/pandas/pandas/tests/extension/test_arrow.py
new file mode 100644
index 0000000000000..fbd3868f62899
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_arrow.py
@@ -0,0 +1,3542 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+from decimal import Decimal
+from io import (
+    BytesIO,
+    StringIO,
+)
+import operator
+import pickle
+import re
+import sys
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._libs.tslibs import timezones
+from pandas.compat import (
+    PY311,
+    PY312,
+    is_ci_environment,
+    is_platform_windows,
+    pa_version_under11p0,
+    pa_version_under13p0,
+    pa_version_under14p0,
+)
+
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    CategoricalDtypeType,
+)
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.extensions import no_default
+from pandas.api.types import (
+    is_bool_dtype,
+    is_datetime64_any_dtype,
+    is_float_dtype,
+    is_integer_dtype,
+    is_numeric_dtype,
+    is_signed_integer_dtype,
+    is_string_dtype,
+    is_unsigned_integer_dtype,
+)
+from pandas.tests.extension import base
+
+pa = pytest.importorskip("pyarrow")
+
+from pandas.core.arrays.arrow.array import (
+    ArrowExtensionArray,
+    get_unit_from_pa_dtype,
+)
+from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
+
+
+def _require_timezone_database(request):
+    if is_platform_windows() and is_ci_environment():
+        mark = pytest.mark.xfail(
+            raises=pa.ArrowInvalid,
+            reason=(
+                "TODO: Set ARROW_TIMEZONE_DATABASE environment variable "
+                "on CI to path to the tzdata for pyarrow."
+            ),
+        )
+        request.applymarker(mark)
+
+
+@pytest.fixture(params=tm.ALL_PYARROW_DTYPES, ids=str)
+def dtype(request):
+    return ArrowDtype(pyarrow_dtype=request.param)
+
+
+@pytest.fixture
+def data(dtype):
+    pa_dtype = dtype.pyarrow_dtype
+    if pa.types.is_boolean(pa_dtype):
+        data = [True, False] * 4 + [None] + [True, False] * 44 + [None] + [True, False]
+    elif pa.types.is_floating(pa_dtype):
+        data = [1.0, 0.0] * 4 + [None] + [-2.0, -1.0] * 44 + [None] + [0.5, 99.5]
+    elif pa.types.is_signed_integer(pa_dtype):
+        data = [1, 0] * 4 + [None] + [-2, -1] * 44 + [None] + [1, 99]
+    elif pa.types.is_unsigned_integer(pa_dtype):
+        data = [1, 0] * 4 + [None] + [2, 1] * 44 + [None] + [1, 99]
+    elif pa.types.is_decimal(pa_dtype):
+        data = (
+            [Decimal("1"), Decimal("0.0")] * 4
+            + [None]
+            + [Decimal("-2.0"), Decimal("-1.0")] * 44
+            + [None]
+            + [Decimal("0.5"), Decimal("33.123")]
+        )
+    elif pa.types.is_date(pa_dtype):
+        data = (
+            [date(2022, 1, 1), date(1999, 12, 31)] * 4
+            + [None]
+            + [date(2022, 1, 1), date(2022, 1, 1)] * 44
+            + [None]
+            + [date(1999, 12, 31), date(1999, 12, 31)]
+        )
+    elif pa.types.is_timestamp(pa_dtype):
+        data = (
+            [datetime(2020, 1, 1, 1, 1, 1, 1), datetime(1999, 1, 1, 1, 1, 1, 1)] * 4
+            + [None]
+            + [datetime(2020, 1, 1, 1), datetime(1999, 1, 1, 1)] * 44
+            + [None]
+            + [datetime(2020, 1, 1), datetime(1999, 1, 1)]
+        )
+    elif pa.types.is_duration(pa_dtype):
+        data = (
+            [timedelta(1), timedelta(1, 1)] * 4
+            + [None]
+            + [timedelta(-1), timedelta(0)] * 44
+            + [None]
+            + [timedelta(-10), timedelta(10)]
+        )
+    elif pa.types.is_time(pa_dtype):
+        data = (
+            [time(12, 0), time(0, 12)] * 4
+            + [None]
+            + [time(0, 0), time(1, 1)] * 44
+            + [None]
+            + [time(0, 5), time(5, 0)]
+        )
+    elif pa.types.is_string(pa_dtype):
+        data = ["a", "b"] * 4 + [None] + ["1", "2"] * 44 + [None] + ["!", ">"]
+    elif pa.types.is_binary(pa_dtype):
+        data = [b"a", b"b"] * 4 + [None] + [b"1", b"2"] * 44 + [None] + [b"!", b">"]
+    else:
+        raise NotImplementedError
+    return pd.array(data, dtype=dtype)
+
+
+@pytest.fixture
+def data_missing(data):
+    """Length-2 array with [NA, Valid]"""
+    return type(data)._from_sequence([None, data[0]], dtype=data.dtype)
+
+
+@pytest.fixture(params=["data", "data_missing"])
+def all_data(request, data, data_missing):
+    """Parametrized fixture returning 'data' or 'data_missing' integer arrays.
+
+    Used to test dtype conversion with and without missing values.
+    """
+    if request.param == "data":
+        return data
+    elif request.param == "data_missing":
+        return data_missing
+
+
+@pytest.fixture
+def data_for_grouping(dtype):
+    """
+    Data for factorization, grouping, and unique tests.
+
+    Expected to be like [B, B, NA, NA, A, A, B, C]
+
+    Where A < B < C and NA is missing
+    """
+    pa_dtype = dtype.pyarrow_dtype
+    if pa.types.is_boolean(pa_dtype):
+        A = False
+        B = True
+        C = True
+    elif pa.types.is_floating(pa_dtype):
+        A = -1.1
+        B = 0.0
+        C = 1.1
+    elif pa.types.is_signed_integer(pa_dtype):
+        A = -1
+        B = 0
+        C = 1
+    elif pa.types.is_unsigned_integer(pa_dtype):
+        A = 0
+        B = 1
+        C = 10
+    elif pa.types.is_date(pa_dtype):
+        A = date(1999, 12, 31)
+        B = date(2010, 1, 1)
+        C = date(2022, 1, 1)
+    elif pa.types.is_timestamp(pa_dtype):
+        A = datetime(1999, 1, 1, 1, 1, 1, 1)
+        B = datetime(2020, 1, 1)
+        C = datetime(2020, 1, 1, 1)
+    elif pa.types.is_duration(pa_dtype):
+        A = timedelta(-1)
+        B = timedelta(0)
+        C = timedelta(1, 4)
+    elif pa.types.is_time(pa_dtype):
+        A = time(0, 0)
+        B = time(0, 12)
+        C = time(12, 12)
+    elif pa.types.is_string(pa_dtype):
+        A = "a"
+        B = "b"
+        C = "c"
+    elif pa.types.is_binary(pa_dtype):
+        A = b"a"
+        B = b"b"
+        C = b"c"
+    elif pa.types.is_decimal(pa_dtype):
+        A = Decimal("-1.1")
+        B = Decimal("0.0")
+        C = Decimal("1.1")
+    else:
+        raise NotImplementedError
+    return pd.array([B, B, None, None, A, A, B, C], dtype=dtype)
+
+
+@pytest.fixture
+def data_for_sorting(data_for_grouping):
+    """
+    Length-3 array with a known sort order.
+
+    This should be three items [B, C, A] with
+    A < B < C
+    """
+    return type(data_for_grouping)._from_sequence(
+        [data_for_grouping[0], data_for_grouping[7], data_for_grouping[4]],
+        dtype=data_for_grouping.dtype,
+    )
+
+
+@pytest.fixture
+def data_missing_for_sorting(data_for_grouping):
+    """
+    Length-3 array with a known sort order.
+
+    This should be three items [B, NA, A] with
+    A < B and NA missing.
+    """
+    return type(data_for_grouping)._from_sequence(
+        [data_for_grouping[0], data_for_grouping[2], data_for_grouping[4]],
+        dtype=data_for_grouping.dtype,
+    )
+
+
+@pytest.fixture
+def data_for_twos(data):
+    """Length-100 array in which all the elements are two."""
+    pa_dtype = data.dtype.pyarrow_dtype
+    if (
+        pa.types.is_integer(pa_dtype)
+        or pa.types.is_floating(pa_dtype)
+        or pa.types.is_decimal(pa_dtype)
+        or pa.types.is_duration(pa_dtype)
+    ):
+        return pd.array([2] * 100, dtype=data.dtype)
+    # tests will be xfailed where 2 is not a valid scalar for pa_dtype
+    return data
+    # TODO: skip otherwise?
+
+
+class TestArrowArray(base.ExtensionTests):
+    def test_compare_scalar(self, data, comparison_op):
+        ser = pd.Series(data)
+        self._compare_other(ser, data, comparison_op, data[0])
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data_missing, na_action):
+        if data_missing.dtype.kind in "mM":
+            result = data_missing.map(lambda x: x, na_action=na_action)
+            expected = data_missing.to_numpy(dtype=object)
+            tm.assert_numpy_array_equal(result, expected)
+        else:
+            result = data_missing.map(lambda x: x, na_action=na_action)
+            if data_missing.dtype == "float32[pyarrow]":
+                # map roundtrips through objects, which converts to float64
+                expected = data_missing.to_numpy(dtype="float64", na_value=np.nan)
+            else:
+                expected = data_missing.to_numpy()
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_str(self, data, request, using_infer_string):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if pa.types.is_binary(pa_dtype):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=f"For {pa_dtype} .astype(str) decodes.",
+                )
+            )
+        elif not using_infer_string and (
+            (pa.types.is_timestamp(pa_dtype) and pa_dtype.tz is None)
+            or pa.types.is_duration(pa_dtype)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="pd.Timestamp/pd.Timedelta repr different from numpy repr",
+                )
+            )
+        super().test_astype_str(data)
+
+    def test_from_dtype(self, data, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if pa.types.is_string(pa_dtype) or pa.types.is_decimal(pa_dtype):
+            if pa.types.is_string(pa_dtype):
+                reason = "ArrowDtype(pa.string()) != StringDtype('pyarrow')"
+            else:
+                reason = f"pyarrow.type_for_alias cannot infer {pa_dtype}"
+
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=reason,
+                )
+            )
+        super().test_from_dtype(data)
+
+    def test_from_sequence_pa_array(self, data):
+        # https://github.com/pandas-dev/pandas/pull/47034#discussion_r955500784
+        # data._pa_array = pa.ChunkedArray
+        result = type(data)._from_sequence(data._pa_array, dtype=data.dtype)
+        tm.assert_extension_array_equal(result, data)
+        assert isinstance(result._pa_array, pa.ChunkedArray)
+
+        result = type(data)._from_sequence(
+            data._pa_array.combine_chunks(), dtype=data.dtype
+        )
+        tm.assert_extension_array_equal(result, data)
+        assert isinstance(result._pa_array, pa.ChunkedArray)
+
+    def test_from_sequence_pa_array_notimplemented(self, request):
+        dtype = ArrowDtype(pa.month_day_nano_interval())
+        with pytest.raises(NotImplementedError, match="Converting strings to"):
+            ArrowExtensionArray._from_sequence_of_strings(["12-1"], dtype=dtype)
+
+    def test_from_sequence_of_strings_pa_array(self, data, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if pa.types.is_time64(pa_dtype) and pa_dtype.equals("time64[ns]") and not PY311:
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="Nanosecond time parsing not supported.",
+                )
+            )
+        elif pa_version_under11p0 and (
+            pa.types.is_duration(pa_dtype) or pa.types.is_decimal(pa_dtype)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=pa.ArrowNotImplementedError,
+                    reason=f"pyarrow doesn't support parsing {pa_dtype}",
+                )
+            )
+        elif pa.types.is_timestamp(pa_dtype) and pa_dtype.tz is not None:
+            _require_timezone_database(request)
+
+        pa_array = data._pa_array.cast(pa.string())
+        result = type(data)._from_sequence_of_strings(pa_array, dtype=data.dtype)
+        tm.assert_extension_array_equal(result, data)
+
+        pa_array = pa_array.combine_chunks()
+        result = type(data)._from_sequence_of_strings(pa_array, dtype=data.dtype)
+        tm.assert_extension_array_equal(result, data)
+
+    def check_accumulate(self, ser, op_name, skipna):
+        result = getattr(ser, op_name)(skipna=skipna)
+
+        pa_type = ser.dtype.pyarrow_dtype
+        if pa.types.is_temporal(pa_type):
+            # Just check that we match the integer behavior.
+            if pa_type.bit_width == 32:
+                int_type = "int32[pyarrow]"
+            else:
+                int_type = "int64[pyarrow]"
+            ser = ser.astype(int_type)
+            result = result.astype(int_type)
+
+        result = result.astype("Float64")
+        expected = getattr(ser.astype("Float64"), op_name)(skipna=skipna)
+        tm.assert_series_equal(result, expected, check_dtype=False)
+
+    def _supports_accumulation(self, ser: pd.Series, op_name: str) -> bool:
+        # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype" has no
+        # attribute "pyarrow_dtype"
+        pa_type = ser.dtype.pyarrow_dtype  # type: ignore[union-attr]
+
+        if pa.types.is_binary(pa_type) or pa.types.is_decimal(pa_type):
+            if op_name in ["cumsum", "cumprod", "cummax", "cummin"]:
+                return False
+        elif pa.types.is_string(pa_type):
+            if op_name == "cumprod":
+                return False
+        elif pa.types.is_boolean(pa_type):
+            if op_name in ["cumprod", "cummax", "cummin"]:
+                return False
+        elif pa.types.is_temporal(pa_type):
+            if op_name == "cumsum" and not pa.types.is_duration(pa_type):
+                return False
+            elif op_name == "cumprod":
+                return False
+        return True
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_accumulate_series(self, data, all_numeric_accumulations, skipna, request):
+        pa_type = data.dtype.pyarrow_dtype
+        op_name = all_numeric_accumulations
+
+        if pa.types.is_string(pa_type) and op_name in ["cumsum", "cummin", "cummax"]:
+            # https://github.com/pandas-dev/pandas/pull/60633
+            # Doesn't fit test structure, tested in series/test_cumulative.py instead.
+            return
+
+        ser = pd.Series(data)
+
+        if not self._supports_accumulation(ser, op_name):
+            # The base class test will check that we raise
+            return super().test_accumulate_series(
+                data, all_numeric_accumulations, skipna
+            )
+
+        if pa_version_under13p0 and all_numeric_accumulations != "cumsum":
+            # xfailing takes a long time to run because pytest
+            # renders the exception messages even when not showing them
+            opt = request.config.option
+            if opt.markexpr and "not slow" in opt.markexpr:
+                pytest.skip(
+                    f"{all_numeric_accumulations} not implemented for pyarrow < 9"
+                )
+            mark = pytest.mark.xfail(
+                reason=f"{all_numeric_accumulations} not implemented for pyarrow < 9"
+            )
+            request.applymarker(mark)
+
+        elif all_numeric_accumulations == "cumsum" and (
+            pa.types.is_boolean(pa_type) or pa.types.is_decimal(pa_type)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=f"{all_numeric_accumulations} not implemented for {pa_type}",
+                    raises=TypeError,
+                )
+            )
+
+        self.check_accumulate(ser, op_name, skipna)
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        if op_name in ["kurt", "skew"]:
+            return False
+
+        dtype = ser.dtype
+        # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype" has
+        # no attribute "pyarrow_dtype"
+        pa_dtype = dtype.pyarrow_dtype  # type: ignore[union-attr]
+        if pa.types.is_temporal(pa_dtype) and op_name in ["sum", "var", "prod"]:
+            if pa.types.is_duration(pa_dtype) and op_name in ["sum"]:
+                # summing timedeltas is one case that *is* well-defined
+                pass
+            else:
+                return False
+        elif pa.types.is_binary(pa_dtype) and op_name == "sum":
+            return False
+        elif (
+            pa.types.is_string(pa_dtype) or pa.types.is_binary(pa_dtype)
+        ) and op_name in [
+            "mean",
+            "median",
+            "prod",
+            "std",
+            "sem",
+            "var",
+        ]:
+            return False
+
+        if (
+            pa.types.is_temporal(pa_dtype)
+            and not pa.types.is_duration(pa_dtype)
+            and op_name in ["any", "all"]
+        ):
+            # xref GH#34479 we support this in our non-pyarrow datetime64 dtypes,
+            #  but it isn't obvious we _should_.  For now, we keep the pyarrow
+            #  behavior which does not support this.
+            return False
+
+        if pa.types.is_boolean(pa_dtype) and op_name in [
+            "median",
+            "std",
+            "var",
+            "skew",
+            "kurt",
+            "sem",
+        ]:
+            return False
+
+        return True
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype" has no
+        # attribute "pyarrow_dtype"
+        pa_dtype = ser.dtype.pyarrow_dtype  # type: ignore[union-attr]
+        if pa.types.is_integer(pa_dtype) or pa.types.is_floating(pa_dtype):
+            alt = ser.astype("Float64")
+        else:
+            # TODO: in the opposite case, aren't we testing... nothing? For
+            # e.g. date/time dtypes trying to calculate 'expected' by converting
+            # to object will raise for mean, std etc
+            alt = ser
+
+        # TODO: in the opposite case, aren't we testing... nothing?
+        if op_name == "count":
+            result = getattr(ser, op_name)()
+            expected = getattr(alt, op_name)()
+        else:
+            result = getattr(ser, op_name)(skipna=skipna)
+            expected = getattr(alt, op_name)(skipna=skipna)
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_series_boolean(
+        self, data, all_boolean_reductions, skipna, na_value, request
+    ):
+        pa_dtype = data.dtype.pyarrow_dtype
+        xfail_mark = pytest.mark.xfail(
+            raises=TypeError,
+            reason=(
+                f"{all_boolean_reductions} is not implemented in "
+                f"pyarrow={pa.__version__} for {pa_dtype}"
+            ),
+        )
+        if pa.types.is_string(pa_dtype) or pa.types.is_binary(pa_dtype):
+            # We *might* want to make this behave like the non-pyarrow cases,
+            #  but have not yet decided.
+            request.applymarker(xfail_mark)
+
+        return super().test_reduce_series_boolean(data, all_boolean_reductions, skipna)
+
+    def _get_expected_reduction_dtype(self, arr, op_name: str, skipna: bool):
+        pa_type = arr._pa_array.type
+
+        if op_name in ["max", "min"]:
+            cmp_dtype = arr.dtype
+        elif pa.types.is_temporal(pa_type):
+            if op_name in ["std", "sem"]:
+                if pa.types.is_duration(pa_type):
+                    cmp_dtype = arr.dtype
+                elif pa.types.is_date(pa_type):
+                    cmp_dtype = ArrowDtype(pa.duration("s"))
+                elif pa.types.is_time(pa_type):
+                    unit = get_unit_from_pa_dtype(pa_type)
+                    cmp_dtype = ArrowDtype(pa.duration(unit))
+                else:
+                    cmp_dtype = ArrowDtype(pa.duration(pa_type.unit))
+            else:
+                cmp_dtype = arr.dtype
+        elif arr.dtype.name == "decimal128(7, 3)[pyarrow]":
+            if op_name not in ["median", "var", "std", "sem"]:
+                cmp_dtype = arr.dtype
+            else:
+                cmp_dtype = "float64[pyarrow]"
+        elif op_name in ["median", "var", "std", "mean", "skew", "sem"]:
+            cmp_dtype = "float64[pyarrow]"
+        elif op_name in ["sum", "prod"] and pa.types.is_boolean(pa_type):
+            cmp_dtype = "uint64[pyarrow]"
+        elif op_name == "sum" and pa.types.is_string(pa_type):
+            cmp_dtype = arr.dtype
+        else:
+            cmp_dtype = {
+                "i": "int64[pyarrow]",
+                "u": "uint64[pyarrow]",
+                "f": "float64[pyarrow]",
+            }[arr.dtype.kind]
+        return cmp_dtype
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_frame(self, data, all_numeric_reductions, skipna, request):
+        op_name = all_numeric_reductions
+        if op_name == "skew":
+            if data.dtype._is_numeric:
+                mark = pytest.mark.xfail(reason="skew not implemented")
+                request.applymarker(mark)
+        elif (
+            op_name in ["std", "sem"]
+            and pa.types.is_date64(data._pa_array.type)
+            and skipna
+        ):
+            # overflow
+            mark = pytest.mark.xfail(reason="Cannot cast")
+            request.applymarker(mark)
+        return super().test_reduce_frame(data, all_numeric_reductions, skipna)
+
+    @pytest.mark.parametrize("typ", ["int64", "uint64", "float64"])
+    def test_median_not_approximate(self, typ):
+        # GH 52679
+        result = pd.Series([1, 2], dtype=f"{typ}[pyarrow]").median()
+        assert result == 1.5
+
+    def test_construct_from_string_own_name(self, dtype, request):
+        pa_dtype = dtype.pyarrow_dtype
+        if pa.types.is_decimal(pa_dtype):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=NotImplementedError,
+                    reason=f"pyarrow.type_for_alias cannot infer {pa_dtype}",
+                )
+            )
+
+        if pa.types.is_string(pa_dtype):
+            # We still support StringDtype('pyarrow') over ArrowDtype(pa.string())
+            msg = r"string\[pyarrow\] should be constructed by StringDtype"
+            with pytest.raises(TypeError, match=msg):
+                dtype.construct_from_string(dtype.name)
+
+            return
+
+        super().test_construct_from_string_own_name(dtype)
+
+    def test_is_dtype_from_name(self, dtype, request):
+        pa_dtype = dtype.pyarrow_dtype
+        if pa.types.is_string(pa_dtype):
+            # We still support StringDtype('pyarrow') over ArrowDtype(pa.string())
+            assert not type(dtype).is_dtype(dtype.name)
+        else:
+            if pa.types.is_decimal(pa_dtype):
+                request.applymarker(
+                    pytest.mark.xfail(
+                        raises=NotImplementedError,
+                        reason=f"pyarrow.type_for_alias cannot infer {pa_dtype}",
+                    )
+                )
+            super().test_is_dtype_from_name(dtype)
+
+    def test_construct_from_string_another_type_raises(self, dtype):
+        msg = r"'another_type' must end with '\[pyarrow\]'"
+        with pytest.raises(TypeError, match=msg):
+            type(dtype).construct_from_string("another_type")
+
+    def test_get_common_dtype(self, dtype, request):
+        pa_dtype = dtype.pyarrow_dtype
+        if (
+            pa.types.is_date(pa_dtype)
+            or pa.types.is_time(pa_dtype)
+            or (pa.types.is_timestamp(pa_dtype) and pa_dtype.tz is not None)
+            or pa.types.is_binary(pa_dtype)
+            or pa.types.is_decimal(pa_dtype)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=(
+                        f"{pa_dtype} does not have associated numpy "
+                        f"dtype findable by find_common_type"
+                    )
+                )
+            )
+        super().test_get_common_dtype(dtype)
+
+    def test_is_not_string_type(self, dtype):
+        pa_dtype = dtype.pyarrow_dtype
+        if pa.types.is_string(pa_dtype):
+            assert is_string_dtype(dtype)
+        else:
+            super().test_is_not_string_type(dtype)
+
+    @pytest.mark.xfail(
+        reason="GH 45419: pyarrow.ChunkedArray does not support views.", run=False
+    )
+    def test_view(self, data):
+        super().test_view(data)
+
+    def test_fillna_no_op_returns_copy(self, data):
+        data = data[~data.isna()]
+
+        valid = data[0]
+        result = data.fillna(valid)
+        assert result is not data
+        tm.assert_extension_array_equal(result, data)
+
+    @pytest.mark.xfail(
+        reason="GH 45419: pyarrow.ChunkedArray does not support views", run=False
+    )
+    def test_transpose(self, data):
+        super().test_transpose(data)
+
+    @pytest.mark.xfail(
+        reason="GH 45419: pyarrow.ChunkedArray does not support views", run=False
+    )
+    def test_setitem_preserves_views(self, data):
+        super().test_setitem_preserves_views(data)
+
+    @pytest.mark.parametrize("dtype_backend", ["pyarrow", no_default])
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_EA_types(self, engine, data, dtype_backend, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if pa.types.is_decimal(pa_dtype):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=NotImplementedError,
+                    reason=f"Parameterized types {pa_dtype} not supported.",
+                )
+            )
+        elif pa.types.is_timestamp(pa_dtype) and pa_dtype.unit in ("us", "ns"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=ValueError,
+                    reason="https://github.com/pandas-dev/pandas/issues/49767",
+                )
+            )
+        elif pa.types.is_binary(pa_dtype):
+            request.applymarker(
+                pytest.mark.xfail(reason="CSV parsers don't correctly handle binary")
+            )
+        df = pd.DataFrame({"with_dtype": pd.Series(data, dtype=str(data.dtype))})
+        csv_output = df.to_csv(index=False, na_rep=np.nan)
+        if pa.types.is_binary(pa_dtype):
+            csv_output = BytesIO(csv_output)
+        else:
+            csv_output = StringIO(csv_output)
+        result = pd.read_csv(
+            csv_output,
+            dtype={"with_dtype": str(data.dtype)},
+            engine=engine,
+            dtype_backend=dtype_backend,
+        )
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+    def test_invert(self, data, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if not (
+            pa.types.is_boolean(pa_dtype)
+            or pa.types.is_integer(pa_dtype)
+            or pa.types.is_string(pa_dtype)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=pa.ArrowNotImplementedError,
+                    reason=f"pyarrow.compute.invert does support {pa_dtype}",
+                )
+            )
+        if PY312 and pa.types.is_boolean(pa_dtype):
+            with tm.assert_produces_warning(
+                DeprecationWarning, match="Bitwise inversion", check_stacklevel=False
+            ):
+                super().test_invert(data)
+        else:
+            super().test_invert(data)
+
+    @pytest.mark.parametrize("periods", [1, -2])
+    def test_diff(self, data, periods, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+        if pa.types.is_unsigned_integer(pa_dtype) and periods == 1:
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=pa.ArrowInvalid,
+                    reason=(
+                        f"diff with {pa_dtype} and periods={periods} will overflow"
+                    ),
+                )
+            )
+        super().test_diff(data, periods)
+
+    def test_value_counts_returns_pyarrow_int64(self, data):
+        # GH 51462
+        data = data[:10]
+        result = data.value_counts()
+        assert result.dtype == ArrowDtype(pa.int64())
+
+    _combine_le_expected_dtype = "bool[pyarrow]"
+
+    def get_op_from_name(self, op_name):
+        short_opname = op_name.strip("_")
+        if short_opname == "rtruediv":
+            # use the numpy version that won't raise on division by zero
+
+            def rtruediv(x, y):
+                return np.divide(y, x)
+
+            return rtruediv
+        elif short_opname == "rfloordiv":
+            return lambda x, y: np.floor_divide(y, x)
+
+        return tm.get_op_from_name(op_name)
+
+    def _cast_pointwise_result(self, op_name: str, obj, other, pointwise_result):
+        # BaseOpsUtil._combine can upcast expected dtype
+        # (because it generates expected on python scalars)
+        # while ArrowExtensionArray maintains original type
+        expected = pointwise_result
+
+        if op_name in ["eq", "ne", "lt", "le", "gt", "ge"]:
+            return pointwise_result.astype("boolean[pyarrow]")
+
+        was_frame = False
+        if isinstance(expected, pd.DataFrame):
+            was_frame = True
+            expected_data = expected.iloc[:, 0]
+            original_dtype = obj.iloc[:, 0].dtype
+        else:
+            expected_data = expected
+            original_dtype = obj.dtype
+
+        orig_pa_type = original_dtype.pyarrow_dtype
+        if not was_frame and isinstance(other, pd.Series):
+            # i.e. test_arith_series_with_array
+            if not (
+                pa.types.is_floating(orig_pa_type)
+                or (
+                    pa.types.is_integer(orig_pa_type)
+                    and op_name not in ["__truediv__", "__rtruediv__"]
+                )
+                or pa.types.is_duration(orig_pa_type)
+                or pa.types.is_timestamp(orig_pa_type)
+                or pa.types.is_date(orig_pa_type)
+                or pa.types.is_decimal(orig_pa_type)
+            ):
+                # base class _combine always returns int64, while
+                #  ArrowExtensionArray does not upcast
+                return expected
+        elif not (
+            (op_name == "__floordiv__" and pa.types.is_integer(orig_pa_type))
+            or pa.types.is_duration(orig_pa_type)
+            or pa.types.is_timestamp(orig_pa_type)
+            or pa.types.is_date(orig_pa_type)
+            or pa.types.is_decimal(orig_pa_type)
+        ):
+            # base class _combine always returns int64, while
+            #  ArrowExtensionArray does not upcast
+            return expected
+
+        pa_expected = pa.array(expected_data._values)
+
+        if pa.types.is_duration(pa_expected.type):
+            if pa.types.is_date(orig_pa_type):
+                if pa.types.is_date64(orig_pa_type):
+                    # TODO: why is this different vs date32?
+                    unit = "ms"
+                else:
+                    unit = "s"
+            else:
+                # pyarrow sees sequence of datetime/timedelta objects and defaults
+                #  to "us" but the non-pointwise op retains unit
+                # timestamp or duration
+                unit = orig_pa_type.unit
+                if type(other) in [datetime, timedelta] and unit in ["s", "ms"]:
+                    # pydatetime/pytimedelta objects have microsecond reso, so we
+                    #  take the higher reso of the original and microsecond. Note
+                    #  this matches what we would do with DatetimeArray/TimedeltaArray
+                    unit = "us"
+
+            pa_expected = pa_expected.cast(f"duration[{unit}]")
+
+        elif pa.types.is_decimal(pa_expected.type) and pa.types.is_decimal(
+            orig_pa_type
+        ):
+            # decimal precision can resize in the result type depending on data
+            # just compare the float values
+            alt = getattr(obj, op_name)(other)
+            alt_dtype = tm.get_dtype(alt)
+            assert isinstance(alt_dtype, ArrowDtype)
+            if op_name == "__pow__" and isinstance(other, Decimal):
+                # TODO: would it make more sense to retain Decimal here?
+                alt_dtype = ArrowDtype(pa.float64())
+            elif (
+                op_name == "__pow__"
+                and isinstance(other, pd.Series)
+                and other.dtype == original_dtype
+            ):
+                # TODO: would it make more sense to retain Decimal here?
+                alt_dtype = ArrowDtype(pa.float64())
+            else:
+                assert pa.types.is_decimal(alt_dtype.pyarrow_dtype)
+            return expected.astype(alt_dtype)
+
+        else:
+            pa_expected = pa_expected.cast(orig_pa_type)
+
+        pd_expected = type(expected_data._values)(pa_expected)
+        if was_frame:
+            expected = pd.DataFrame(
+                pd_expected, index=expected.index, columns=expected.columns
+            )
+        else:
+            expected = pd.Series(pd_expected)
+        return expected
+
+    def _is_temporal_supported(self, opname, pa_dtype):
+        return (
+            (
+                opname in ("__add__", "__radd__")
+                or (
+                    opname
+                    in ("__truediv__", "__rtruediv__", "__floordiv__", "__rfloordiv__")
+                    and not pa_version_under14p0
+                )
+            )
+            and pa.types.is_duration(pa_dtype)
+        ) or (opname in ("__sub__", "__rsub__") and pa.types.is_temporal(pa_dtype))
+
+    def _get_expected_exception(
+        self, op_name: str, obj, other
+    ) -> type[Exception] | tuple[type[Exception], ...] | None:
+        if op_name in ("__divmod__", "__rdivmod__"):
+            return (NotImplementedError, TypeError)
+
+        exc: type[Exception] | tuple[type[Exception], ...] | None
+        dtype = tm.get_dtype(obj)
+        # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype" has no
+        # attribute "pyarrow_dtype"
+        pa_dtype = dtype.pyarrow_dtype  # type: ignore[union-attr]
+
+        arrow_temporal_supported = self._is_temporal_supported(op_name, pa_dtype)
+        if op_name in {
+            "__mod__",
+            "__rmod__",
+        }:
+            exc = (NotImplementedError, TypeError)
+        elif arrow_temporal_supported:
+            exc = None
+        elif op_name in ["__add__", "__radd__"] and (
+            pa.types.is_string(pa_dtype) or pa.types.is_binary(pa_dtype)
+        ):
+            exc = None
+        elif not (
+            pa.types.is_floating(pa_dtype)
+            or pa.types.is_integer(pa_dtype)
+            or pa.types.is_decimal(pa_dtype)
+        ):
+            exc = TypeError
+        else:
+            exc = None
+        return exc
+
+    def _get_arith_xfail_marker(self, opname, pa_dtype):
+        mark = None
+
+        arrow_temporal_supported = self._is_temporal_supported(opname, pa_dtype)
+
+        if opname == "__rpow__" and (
+            pa.types.is_floating(pa_dtype)
+            or pa.types.is_integer(pa_dtype)
+            or pa.types.is_decimal(pa_dtype)
+        ):
+            mark = pytest.mark.xfail(
+                reason=(
+                    f"GH#29997: 1**pandas.NA == 1 while 1**pyarrow.NA == NULL "
+                    f"for {pa_dtype}"
+                )
+            )
+        elif arrow_temporal_supported and (
+            pa.types.is_time(pa_dtype)
+            or (
+                opname
+                in ("__truediv__", "__rtruediv__", "__floordiv__", "__rfloordiv__")
+                and pa.types.is_duration(pa_dtype)
+            )
+        ):
+            mark = pytest.mark.xfail(
+                raises=TypeError,
+                reason=(
+                    f"{opname} not supported betweenpd.NA and {pa_dtype} Python scalar"
+                ),
+            )
+        elif opname == "__rfloordiv__" and (
+            pa.types.is_integer(pa_dtype) or pa.types.is_decimal(pa_dtype)
+        ):
+            mark = pytest.mark.xfail(
+                raises=pa.ArrowInvalid,
+                reason="divide by 0",
+            )
+        elif opname == "__rtruediv__" and pa.types.is_decimal(pa_dtype):
+            mark = pytest.mark.xfail(
+                raises=pa.ArrowInvalid,
+                reason="divide by 0",
+            )
+
+        return mark
+
+    def test_arith_series_with_scalar(self, data, all_arithmetic_operators, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+
+        if all_arithmetic_operators == "__rmod__" and pa.types.is_binary(pa_dtype):
+            pytest.skip("Skip testing Python string formatting")
+
+        mark = self._get_arith_xfail_marker(all_arithmetic_operators, pa_dtype)
+        if mark is not None:
+            request.applymarker(mark)
+
+        super().test_arith_series_with_scalar(data, all_arithmetic_operators)
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+
+        if all_arithmetic_operators == "__rmod__" and (
+            pa.types.is_string(pa_dtype) or pa.types.is_binary(pa_dtype)
+        ):
+            pytest.skip("Skip testing Python string formatting")
+
+        mark = self._get_arith_xfail_marker(all_arithmetic_operators, pa_dtype)
+        if mark is not None:
+            request.applymarker(mark)
+
+        super().test_arith_frame_with_scalar(data, all_arithmetic_operators)
+
+    def test_arith_series_with_array(self, data, all_arithmetic_operators, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+
+        if all_arithmetic_operators in (
+            "__sub__",
+            "__rsub__",
+        ) and pa.types.is_unsigned_integer(pa_dtype):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=pa.ArrowInvalid,
+                    reason=(
+                        f"Implemented pyarrow.compute.subtract_checked "
+                        f"which raises on overflow for {pa_dtype}"
+                    ),
+                )
+            )
+
+        mark = self._get_arith_xfail_marker(all_arithmetic_operators, pa_dtype)
+        if mark is not None:
+            request.applymarker(mark)
+
+        op_name = all_arithmetic_operators
+        ser = pd.Series(data)
+        # pd.Series([ser.iloc[0]] * len(ser)) may not return ArrowExtensionArray
+        # since ser.iloc[0] is a python scalar
+        other = pd.Series(pd.array([ser.iloc[0]] * len(ser), dtype=data.dtype))
+
+        self.check_opname(ser, op_name, other)
+
+    def test_add_series_with_extension_array(self, data, request):
+        pa_dtype = data.dtype.pyarrow_dtype
+
+        if pa_dtype.equals("int8"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=pa.ArrowInvalid,
+                    reason=f"raises on overflow for {pa_dtype}",
+                )
+            )
+        super().test_add_series_with_extension_array(data)
+
+    def test_invalid_other_comp(self, data, comparison_op):
+        # GH 48833
+        with pytest.raises(
+            NotImplementedError, match=".* not implemented for <class 'object'>"
+        ):
+            comparison_op(data, object())
+
+    @pytest.mark.parametrize("masked_dtype", ["boolean", "Int64", "Float64"])
+    def test_comp_masked_numpy(self, masked_dtype, comparison_op):
+        # GH 52625
+        data = [1, 0, None]
+        ser_masked = pd.Series(data, dtype=masked_dtype)
+        ser_pa = pd.Series(data, dtype=f"{masked_dtype.lower()}[pyarrow]")
+        result = comparison_op(ser_pa, ser_masked)
+        if comparison_op in [operator.lt, operator.gt, operator.ne]:
+            exp = [False, False, None]
+        else:
+            exp = [True, True, None]
+        expected = pd.Series(exp, dtype=ArrowDtype(pa.bool_()))
+        tm.assert_series_equal(result, expected)
+
+
+class TestLogicalOps:
+    """Various Series and DataFrame logical ops methods."""
+
+    def test_kleene_or(self):
+        a = pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]")
+        b = pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        result = a | b
+        expected = pd.Series(
+            [True, True, True, True, False, None, True, None, None],
+            dtype="boolean[pyarrow]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = b | a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a,
+            pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]"),
+        )
+        tm.assert_series_equal(
+            b, pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (None, [True, None, None]),
+            (pd.NA, [True, None, None]),
+            (True, [True, True, True]),
+            (np.bool_(True), [True, True, True]),
+            (False, [True, False, None]),
+            (np.bool_(False), [True, False, None]),
+        ],
+    )
+    def test_kleene_or_scalar(self, other, expected):
+        a = pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        result = a | other
+        expected = pd.Series(expected, dtype="boolean[pyarrow]")
+        tm.assert_series_equal(result, expected)
+
+        result = other | a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a, pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        )
+
+    def test_kleene_and(self):
+        a = pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]")
+        b = pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        result = a & b
+        expected = pd.Series(
+            [True, False, None, False, False, False, None, False, None],
+            dtype="boolean[pyarrow]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = b & a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a,
+            pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]"),
+        )
+        tm.assert_series_equal(
+            b, pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (None, [None, False, None]),
+            (pd.NA, [None, False, None]),
+            (True, [True, False, None]),
+            (False, [False, False, False]),
+            (np.bool_(True), [True, False, None]),
+            (np.bool_(False), [False, False, False]),
+        ],
+    )
+    def test_kleene_and_scalar(self, other, expected):
+        a = pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        result = a & other
+        expected = pd.Series(expected, dtype="boolean[pyarrow]")
+        tm.assert_series_equal(result, expected)
+
+        result = other & a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a, pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        )
+
+    def test_kleene_xor(self):
+        a = pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]")
+        b = pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        result = a ^ b
+        expected = pd.Series(
+            [False, True, None, True, False, None, None, None, None],
+            dtype="boolean[pyarrow]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = b ^ a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a,
+            pd.Series([True] * 3 + [False] * 3 + [None] * 3, dtype="boolean[pyarrow]"),
+        )
+        tm.assert_series_equal(
+            b, pd.Series([True, False, None] * 3, dtype="boolean[pyarrow]")
+        )
+
+    @pytest.mark.parametrize(
+        "other, expected",
+        [
+            (None, [None, None, None]),
+            (pd.NA, [None, None, None]),
+            (True, [False, True, None]),
+            (np.bool_(True), [False, True, None]),
+            (np.bool_(False), [True, False, None]),
+        ],
+    )
+    def test_kleene_xor_scalar(self, other, expected):
+        a = pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        result = a ^ other
+        expected = pd.Series(expected, dtype="boolean[pyarrow]")
+        tm.assert_series_equal(result, expected)
+
+        result = other ^ a
+        tm.assert_series_equal(result, expected)
+
+        # ensure we haven't mutated anything inplace
+        tm.assert_series_equal(
+            a, pd.Series([True, False, None], dtype="boolean[pyarrow]")
+        )
+
+    @pytest.mark.parametrize(
+        "op, exp",
+        [
+            ["__and__", True],
+            ["__or__", True],
+            ["__xor__", False],
+        ],
+    )
+    def test_logical_masked_numpy(self, op, exp):
+        # GH 52625
+        data = [True, False, None]
+        ser_masked = pd.Series(data, dtype="boolean")
+        ser_pa = pd.Series(data, dtype="boolean[pyarrow]")
+        result = getattr(ser_pa, op)(ser_masked)
+        expected = pd.Series([exp, False, None], dtype=ArrowDtype(pa.bool_()))
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.ALL_INT_PYARROW_DTYPES)
+def test_bitwise(pa_type):
+    # GH 54495
+    dtype = ArrowDtype(pa_type)
+    left = pd.Series([1, None, 3, 4], dtype=dtype)
+    right = pd.Series([None, 3, 5, 4], dtype=dtype)
+
+    result = left | right
+    expected = pd.Series([None, None, 3 | 5, 4 | 4], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = left & right
+    expected = pd.Series([None, None, 3 & 5, 4 & 4], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = left ^ right
+    expected = pd.Series([None, None, 3 ^ 5, 4 ^ 4], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ~left
+    expected = ~(left.fillna(0).to_numpy())
+    expected = pd.Series(expected, dtype=dtype).mask(left.isnull())
+    tm.assert_series_equal(result, expected)
+
+
+def test_arrowdtype_construct_from_string_type_with_unsupported_parameters():
+    with pytest.raises(NotImplementedError, match="Passing pyarrow type"):
+        ArrowDtype.construct_from_string("not_a_real_dype[s, tz=UTC][pyarrow]")
+
+    with pytest.raises(NotImplementedError, match="Passing pyarrow type"):
+        ArrowDtype.construct_from_string("decimal(7, 2)[pyarrow]")
+
+
+def test_arrowdtype_construct_from_string_supports_dt64tz():
+    # as of GH#50689, timestamptz is supported
+    dtype = ArrowDtype.construct_from_string("timestamp[s, tz=UTC][pyarrow]")
+    expected = ArrowDtype(pa.timestamp("s", "UTC"))
+    assert dtype == expected
+
+
+def test_arrowdtype_construct_from_string_type_only_one_pyarrow():
+    # GH#51225
+    invalid = "int64[pyarrow]foobar[pyarrow]"
+    msg = (
+        r"Passing pyarrow type specific parameters \(\[pyarrow\]\) in the "
+        r"string is not supported\."
+    )
+    with pytest.raises(NotImplementedError, match=msg):
+        pd.Series(range(3), dtype=invalid)
+
+
+def test_arrow_string_multiplication():
+    # GH 56537
+    binary = pd.Series(["abc", "defg"], dtype=ArrowDtype(pa.string()))
+    repeat = pd.Series([2, -2], dtype="int64[pyarrow]")
+    result = binary * repeat
+    expected = pd.Series(["abcabc", ""], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+    reflected_result = repeat * binary
+    tm.assert_series_equal(result, reflected_result)
+
+
+def test_arrow_string_multiplication_scalar_repeat():
+    binary = pd.Series(["abc", "defg"], dtype=ArrowDtype(pa.string()))
+    result = binary * 2
+    expected = pd.Series(["abcabc", "defgdefg"], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+    reflected_result = 2 * binary
+    tm.assert_series_equal(reflected_result, expected)
+
+
+@pytest.mark.parametrize(
+    "interpolation", ["linear", "lower", "higher", "nearest", "midpoint"]
+)
+@pytest.mark.parametrize("quantile", [0.5, [0.5, 0.5]])
+def test_quantile(data, interpolation, quantile, request):
+    pa_dtype = data.dtype.pyarrow_dtype
+
+    data = data.take([0, 0, 0])
+    ser = pd.Series(data)
+
+    if (
+        pa.types.is_string(pa_dtype)
+        or pa.types.is_binary(pa_dtype)
+        or pa.types.is_boolean(pa_dtype)
+    ):
+        # For string, bytes, and bool, we don't *expect* to have quantile work
+        # Note this matches the non-pyarrow behavior
+        msg = r"Function 'quantile' has no kernel matching input types \(.*\)"
+        with pytest.raises(pa.ArrowNotImplementedError, match=msg):
+            ser.quantile(q=quantile, interpolation=interpolation)
+        return
+
+    if (
+        pa.types.is_integer(pa_dtype)
+        or pa.types.is_floating(pa_dtype)
+        or pa.types.is_decimal(pa_dtype)
+    ):
+        pass
+    elif pa.types.is_temporal(data._pa_array.type):
+        pass
+    else:
+        request.applymarker(
+            pytest.mark.xfail(
+                raises=pa.ArrowNotImplementedError,
+                reason=f"quantile not supported by pyarrow for {pa_dtype}",
+            )
+        )
+    data = data.take([0, 0, 0])
+    ser = pd.Series(data)
+    result = ser.quantile(q=quantile, interpolation=interpolation)
+
+    if pa.types.is_timestamp(pa_dtype) and interpolation not in ["lower", "higher"]:
+        # rounding error will make the check below fail
+        #  (e.g. '2020-01-01 01:01:01.000001' vs '2020-01-01 01:01:01.000001024'),
+        #  so we'll check for now that we match the numpy analogue
+        if pa_dtype.tz:
+            pd_dtype = f"M8[{pa_dtype.unit}, {pa_dtype.tz}]"
+        else:
+            pd_dtype = f"M8[{pa_dtype.unit}]"
+        ser_np = ser.astype(pd_dtype)
+
+        expected = ser_np.quantile(q=quantile, interpolation=interpolation)
+        if quantile == 0.5:
+            if pa_dtype.unit == "us":
+                expected = expected.to_pydatetime(warn=False)
+            assert result == expected
+        else:
+            if pa_dtype.unit == "us":
+                expected = expected.dt.floor("us")
+            tm.assert_series_equal(result, expected.astype(data.dtype))
+        return
+
+    if quantile == 0.5:
+        assert result == data[0]
+    else:
+        # Just check the values
+        expected = pd.Series(data.take([0, 0]), index=[0.5, 0.5])
+        if (
+            pa.types.is_integer(pa_dtype)
+            or pa.types.is_floating(pa_dtype)
+            or pa.types.is_decimal(pa_dtype)
+        ):
+            expected = expected.astype("float64[pyarrow]")
+            result = result.astype("float64[pyarrow]")
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "take_idx, exp_idx",
+    [[[0, 0, 2, 2, 4, 4], [4, 0]], [[0, 0, 0, 2, 4, 4], [0]]],
+    ids=["multi_mode", "single_mode"],
+)
+def test_mode_dropna_true(data_for_grouping, take_idx, exp_idx):
+    data = data_for_grouping.take(take_idx)
+    ser = pd.Series(data)
+    result = ser.mode(dropna=True)
+    expected = pd.Series(data_for_grouping.take(exp_idx))
+    tm.assert_series_equal(result, expected)
+
+
+def test_mode_dropna_false_mode_na(data):
+    # GH 50982
+    more_nans = pd.Series([None, None, data[0]], dtype=data.dtype)
+    result = more_nans.mode(dropna=False)
+    expected = pd.Series([None], dtype=data.dtype)
+    tm.assert_series_equal(result, expected)
+
+    expected = pd.Series([data[0], None], dtype=data.dtype)
+    result = expected.mode(dropna=False)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arrow_dtype, expected_type",
+    [
+        [pa.binary(), bytes],
+        [pa.binary(16), bytes],
+        [pa.large_binary(), bytes],
+        [pa.large_string(), str],
+        [pa.list_(pa.int64()), list],
+        [pa.large_list(pa.int64()), list],
+        [pa.map_(pa.string(), pa.int64()), list],
+        [pa.struct([("f1", pa.int8()), ("f2", pa.string())]), dict],
+        [pa.dictionary(pa.int64(), pa.int64()), CategoricalDtypeType],
+    ],
+)
+def test_arrow_dtype_type(arrow_dtype, expected_type):
+    # GH 51845
+    # TODO: Redundant with test_getitem_scalar once arrow_dtype exists in data fixture
+    assert ArrowDtype(arrow_dtype).type == expected_type
+
+
+def test_is_bool_dtype():
+    # GH 22667
+    data = ArrowExtensionArray(pa.array([True, False, True]))
+    assert is_bool_dtype(data)
+    assert pd.core.common.is_bool_indexer(data)
+    s = pd.Series(range(len(data)))
+    result = s[data]
+    expected = s[np.asarray(data)]
+    tm.assert_series_equal(result, expected)
+
+
+def test_is_numeric_dtype(data):
+    # GH 50563
+    pa_type = data.dtype.pyarrow_dtype
+    if (
+        pa.types.is_floating(pa_type)
+        or pa.types.is_integer(pa_type)
+        or pa.types.is_decimal(pa_type)
+    ):
+        assert is_numeric_dtype(data)
+    else:
+        assert not is_numeric_dtype(data)
+
+
+def test_is_integer_dtype(data):
+    # GH 50667
+    pa_type = data.dtype.pyarrow_dtype
+    if pa.types.is_integer(pa_type):
+        assert is_integer_dtype(data)
+    else:
+        assert not is_integer_dtype(data)
+
+
+def test_is_signed_integer_dtype(data):
+    pa_type = data.dtype.pyarrow_dtype
+    if pa.types.is_signed_integer(pa_type):
+        assert is_signed_integer_dtype(data)
+    else:
+        assert not is_signed_integer_dtype(data)
+
+
+def test_is_unsigned_integer_dtype(data):
+    pa_type = data.dtype.pyarrow_dtype
+    if pa.types.is_unsigned_integer(pa_type):
+        assert is_unsigned_integer_dtype(data)
+    else:
+        assert not is_unsigned_integer_dtype(data)
+
+
+def test_is_datetime64_any_dtype(data):
+    pa_type = data.dtype.pyarrow_dtype
+    if pa.types.is_timestamp(pa_type) or pa.types.is_date(pa_type):
+        assert is_datetime64_any_dtype(data)
+    else:
+        assert not is_datetime64_any_dtype(data)
+
+
+def test_is_float_dtype(data):
+    pa_type = data.dtype.pyarrow_dtype
+    if pa.types.is_floating(pa_type):
+        assert is_float_dtype(data)
+    else:
+        assert not is_float_dtype(data)
+
+
+def test_pickle_roundtrip(data):
+    # GH 42600
+    expected = pd.Series(data)
+    expected_sliced = expected.head(2)
+    full_pickled = pickle.dumps(expected)
+    sliced_pickled = pickle.dumps(expected_sliced)
+
+    assert len(full_pickled) > len(sliced_pickled)
+
+    result = pickle.loads(full_pickled)
+    tm.assert_series_equal(result, expected)
+
+    result_sliced = pickle.loads(sliced_pickled)
+    tm.assert_series_equal(result_sliced, expected_sliced)
+
+
+def test_astype_from_non_pyarrow(data):
+    # GH49795
+    pd_array = data._pa_array.to_pandas().array
+    result = pd_array.astype(data.dtype)
+    assert not isinstance(pd_array.dtype, ArrowDtype)
+    assert isinstance(result.dtype, ArrowDtype)
+    tm.assert_extension_array_equal(result, data)
+
+
+def test_astype_float_from_non_pyarrow_str():
+    # GH50430
+    ser = pd.Series(["1.0"])
+    result = ser.astype("float64[pyarrow]")
+    expected = pd.Series([1.0], dtype="float64[pyarrow]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_astype_errors_ignore():
+    # GH 55399
+    expected = pd.DataFrame({"col": [17000000]}, dtype="int32[pyarrow]")
+    result = expected.astype("float[pyarrow]", errors="ignore")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_to_numpy_with_defaults(data):
+    # GH49973
+    result = data.to_numpy()
+
+    pa_type = data._pa_array.type
+    if pa.types.is_duration(pa_type) or pa.types.is_timestamp(pa_type):
+        pytest.skip("Tested in test_to_numpy_temporal")
+    elif pa.types.is_date(pa_type):
+        expected = np.array(list(data))
+    else:
+        expected = np.array(data._pa_array)
+
+    if data._hasna and not is_numeric_dtype(data.dtype):
+        expected = expected.astype(object)
+        expected[pd.isna(data)] = pd.NA
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_int_with_na():
+    # GH51227: ensure to_numpy does not convert int to float
+    data = [1, None]
+    arr = pd.array(data, dtype="int64[pyarrow]")
+    result = arr.to_numpy()
+    expected = np.array([1, np.nan])
+    assert isinstance(result[0], float)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("na_val, exp", [(lib.no_default, np.nan), (1, 1)])
+def test_to_numpy_null_array(na_val, exp):
+    # GH#52443
+    arr = pd.array([pd.NA, pd.NA], dtype="null[pyarrow]")
+    result = arr.to_numpy(dtype="float64", na_value=na_val)
+    expected = np.array([exp] * 2, dtype="float64")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_null_array_no_dtype():
+    # GH#52443
+    arr = pd.array([pd.NA, pd.NA], dtype="null[pyarrow]")
+    result = arr.to_numpy(dtype=None)
+    expected = np.array([pd.NA] * 2, dtype="object")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_without_dtype():
+    # GH 54808
+    arr = pd.array([True, pd.NA], dtype="boolean[pyarrow]")
+    result = arr.to_numpy(na_value=False)
+    expected = np.array([True, False], dtype=np.bool_)
+    tm.assert_numpy_array_equal(result, expected)
+
+    arr = pd.array([1.0, pd.NA], dtype="float32[pyarrow]")
+    result = arr.to_numpy(na_value=0.0)
+    expected = np.array([1.0, 0.0], dtype=np.float32)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_setitem_null_slice(data):
+    # GH50248
+    orig = data.copy()
+
+    result = orig.copy()
+    result[:] = data[0]
+    expected = ArrowExtensionArray._from_sequence(
+        [data[0]] * len(data),
+        dtype=data.dtype,
+    )
+    tm.assert_extension_array_equal(result, expected)
+
+    result = orig.copy()
+    result[:] = data[::-1]
+    expected = data[::-1]
+    tm.assert_extension_array_equal(result, expected)
+
+    result = orig.copy()
+    result[:] = data.tolist()
+    expected = data
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_setitem_invalid_dtype(data):
+    # GH50248
+    pa_type = data._pa_array.type
+    if pa.types.is_string(pa_type) or pa.types.is_binary(pa_type):
+        fill_value = 123
+        err = TypeError
+        msg = "Invalid value '123' for dtype"
+    elif (
+        pa.types.is_integer(pa_type)
+        or pa.types.is_floating(pa_type)
+        or pa.types.is_boolean(pa_type)
+    ):
+        fill_value = "foo"
+        err = pa.ArrowInvalid
+        msg = "Could not convert"
+    else:
+        fill_value = "foo"
+        err = TypeError
+        msg = "Invalid value 'foo' for dtype"
+    with pytest.raises(err, match=msg):
+        data[:] = fill_value
+
+
+def test_from_arrow_respecting_given_dtype():
+    date_array = pa.array(
+        [pd.Timestamp("2019-12-31"), pd.Timestamp("2019-12-31")], type=pa.date32()
+    )
+    result = date_array.to_pandas(
+        types_mapper={pa.date32(): ArrowDtype(pa.date64())}.get
+    )
+    expected = pd.Series(
+        [pd.Timestamp("2019-12-31"), pd.Timestamp("2019-12-31")],
+        dtype=ArrowDtype(pa.date64()),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_from_arrow_respecting_given_dtype_unsafe():
+    array = pa.array([1.5, 2.5], type=pa.float64())
+    with tm.external_error_raised(pa.ArrowInvalid):
+        array.to_pandas(types_mapper={pa.float64(): ArrowDtype(pa.int64())}.get)
+
+
+def test_round():
+    dtype = "float64[pyarrow]"
+
+    ser = pd.Series([0.0, 1.23, 2.56, pd.NA], dtype=dtype)
+    result = ser.round(1)
+    expected = pd.Series([0.0, 1.2, 2.6, pd.NA], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+    ser = pd.Series([123.4, pd.NA, 56.78], dtype=dtype)
+    result = ser.round(-1)
+    expected = pd.Series([120.0, pd.NA, 60.0], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_searchsorted_with_na_raises(data_for_sorting, as_series):
+    # GH50447
+    b, c, a = data_for_sorting
+    arr = data_for_sorting.take([2, 0, 1])  # to get [a, b, c]
+    arr[-1] = pd.NA
+
+    if as_series:
+        arr = pd.Series(arr)
+
+    msg = (
+        "searchsorted requires array to be sorted, "
+        "which is impossible with NAs present."
+    )
+    with pytest.raises(ValueError, match=msg):
+        arr.searchsorted(b)
+
+
+def test_sort_values_dictionary():
+    df = pd.DataFrame(
+        {
+            "a": pd.Series(
+                ["x", "y"], dtype=ArrowDtype(pa.dictionary(pa.int32(), pa.string()))
+            ),
+            "b": [1, 2],
+        },
+    )
+    expected = df.copy()
+    result = df.sort_values(by=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("pat", ["abc", "a[a-z]{2}"])
+def test_str_count(pat):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.count(pat)
+    expected = pd.Series([1, None], dtype=ArrowDtype(pa.int32()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_count_flags_unsupported():
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(NotImplementedError, match="count not"):
+        ser.str.count("abc", flags=1)
+
+
+@pytest.mark.parametrize(
+    "side, str_func", [["left", "rjust"], ["right", "ljust"], ["both", "center"]]
+)
+def test_str_pad(side, str_func):
+    ser = pd.Series(["a", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.pad(width=3, side=side, fillchar="x")
+    expected = pd.Series(
+        [getattr("a", str_func)(3, "x"), None], dtype=ArrowDtype(pa.string())
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_pad_invalid_side():
+    ser = pd.Series(["a", None], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(ValueError, match="Invalid side: foo"):
+        ser.str.pad(3, "foo", "x")
+
+
+@pytest.mark.parametrize(
+    "pat, case, na, regex, exp",
+    [
+        ["ab", False, None, False, [True, None]],
+        ["Ab", True, None, False, [False, None]],
+        ["ab", False, True, False, [True, True]],
+        ["a[a-z]{1}", False, None, True, [True, None]],
+        ["A[a-z]{1}", True, None, True, [False, None]],
+    ],
+)
+def test_str_contains(pat, case, na, regex, exp):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.contains(pat, case=case, na=na, regex=regex)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_contains_flags_unsupported():
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(NotImplementedError, match="contains not"):
+        ser.str.contains("a", flags=1)
+
+
+@pytest.mark.parametrize(
+    "side, pat, na, exp",
+    [
+        ["startswith", "ab", None, [True, None, False]],
+        ["startswith", "b", False, [False, False, False]],
+        ["endswith", "b", True, [False, True, False]],
+        ["endswith", "bc", None, [True, None, False]],
+        ["startswith", ("a", "e", "g"), None, [True, None, True]],
+        ["endswith", ("a", "c", "g"), None, [True, None, True]],
+        ["startswith", (), None, [False, None, False]],
+        ["endswith", (), None, [False, None, False]],
+    ],
+)
+def test_str_start_ends_with(side, pat, na, exp):
+    ser = pd.Series(["abc", None, "efg"], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, side)(pat, na=na)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("side", ("startswith", "endswith"))
+def test_str_starts_ends_with_all_nulls_empty_tuple(side):
+    ser = pd.Series([None, None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, side)(())
+
+    # bool datatype preserved for all nulls.
+    expected = pd.Series([None, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg_name, arg",
+    [["pat", re.compile("b")], ["repl", str], ["case", False], ["flags", 1]],
+)
+def test_str_replace_unsupported(arg_name, arg):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    kwargs = {"pat": "b", "repl": "x", "regex": True}
+    kwargs[arg_name] = arg
+    with pytest.raises(NotImplementedError, match="replace is not supported"):
+        ser.str.replace(**kwargs)
+
+
+@pytest.mark.parametrize(
+    "pat, repl, n, regex, exp",
+    [
+        ["a", "x", -1, False, ["xbxc", None]],
+        ["a", "x", 1, False, ["xbac", None]],
+        ["[a-b]", "x", -1, True, ["xxxc", None]],
+    ],
+)
+def test_str_replace(pat, repl, n, regex, exp):
+    ser = pd.Series(["abac", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.replace(pat, repl, n=n, regex=regex)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_replace_negative_n():
+    # GH 56404
+    ser = pd.Series(["abc", "aaaaaa"], dtype=ArrowDtype(pa.string()))
+    actual = ser.str.replace("a", "", -3, True)
+    expected = pd.Series(["bc", ""], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(expected, actual)
+
+    # Same bug for pyarrow-backed StringArray GH#59628
+    ser2 = ser.astype(pd.StringDtype(storage="pyarrow"))
+    actual2 = ser2.str.replace("a", "", -3, True)
+    expected2 = expected.astype(ser2.dtype)
+    tm.assert_series_equal(expected2, actual2)
+
+    ser3 = ser.astype(pd.StringDtype(storage="pyarrow", na_value=np.nan))
+    actual3 = ser3.str.replace("a", "", -3, True)
+    expected3 = expected.astype(ser3.dtype)
+    tm.assert_series_equal(expected3, actual3)
+
+
+def test_str_repeat_unsupported():
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(NotImplementedError, match="repeat is not"):
+        ser.str.repeat([1, 2])
+
+
+def test_str_repeat():
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.repeat(2)
+    expected = pd.Series(["abcabc", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pat, case, na, exp",
+    [
+        ["ab", False, None, [True, None]],
+        ["Ab", True, None, [False, None]],
+        ["bc", True, None, [False, None]],
+        ["ab", False, True, [True, True]],
+        ["a[a-z]{1}", False, None, [True, None]],
+        ["A[a-z]{1}", True, None, [False, None]],
+    ],
+)
+def test_str_match(pat, case, na, exp):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.match(pat, case=case, na=na)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pat, case, na, exp",
+    [
+        ["abc", False, None, [True, True, False, None]],
+        ["Abc", True, None, [False, False, False, None]],
+        ["bc", True, None, [False, False, False, None]],
+        ["ab", False, None, [True, True, False, None]],
+        ["a[a-z]{2}", False, None, [True, True, False, None]],
+        ["A[a-z]{1}", True, None, [False, False, False, None]],
+        # GH Issue: #56652
+        ["abc$", False, None, [True, False, False, None]],
+        ["abc\\$", False, None, [False, True, False, None]],
+        ["Abc$", True, None, [False, False, False, None]],
+        ["Abc\\$", True, None, [False, False, False, None]],
+    ],
+)
+def test_str_fullmatch(pat, case, na, exp):
+    ser = pd.Series(["abc", "abc$", "$abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.match(pat, case=case, na=na)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "sub, start, end, exp, exp_type",
+    [
+        ["ab", 0, None, [0, None], pa.int32()],
+        ["bc", 1, 3, [1, None], pa.int64()],
+        ["ab", 1, 3, [-1, None], pa.int64()],
+        ["ab", -3, -3, [-1, None], pa.int64()],
+    ],
+)
+def test_str_find(sub, start, end, exp, exp_type):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find(sub, start=start, end=end)
+    expected = pd.Series(exp, dtype=ArrowDtype(exp_type))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_find_negative_start():
+    # GH 56411
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find(sub="b", start=-1000, end=3)
+    expected = pd.Series([1, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_find_no_end():
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find("ab", start=1)
+    expected = pd.Series([-1, None], dtype="int64[pyarrow]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_find_negative_start_negative_end():
+    # GH 56791
+    ser = pd.Series(["abcdefg", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find(sub="d", start=-6, end=-3)
+    expected = pd.Series([3, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_find_large_start():
+    # GH 56791
+    ser = pd.Series(["abcdefg", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find(sub="d", start=16)
+    expected = pd.Series([-1, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.skipif(
+    pa_version_under13p0, reason="https://github.com/apache/arrow/issues/36311"
+)
+@pytest.mark.parametrize("start", [-15, -3, 0, 1, 15, None])
+@pytest.mark.parametrize("end", [-15, -1, 0, 3, 15, None])
+@pytest.mark.parametrize("sub", ["", "az", "abce", "a", "caa"])
+def test_str_find_e2e(start, end, sub):
+    s = pd.Series(
+        ["abcaadef", "abc", "abcdeddefgj8292", "ab", "a", ""],
+        dtype=ArrowDtype(pa.string()),
+    )
+    object_series = s.astype(pd.StringDtype(storage="python"))
+    result = s.str.find(sub, start, end)
+    expected = object_series.str.find(sub, start, end).astype(result.dtype)
+    tm.assert_series_equal(result, expected)
+
+    arrow_str_series = s.astype(pd.StringDtype(storage="pyarrow"))
+    result2 = arrow_str_series.str.find(sub, start, end).astype(result.dtype)
+    tm.assert_series_equal(result2, expected)
+
+
+def test_str_find_negative_start_negative_end_no_match():
+    # GH 56791
+    ser = pd.Series(["abcdefg", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.find(sub="d", start=-3, end=-6)
+    expected = pd.Series([-1, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "i, exp",
+    [
+        [1, ["b", "e", None]],
+        [-1, ["c", "e", None]],
+        [2, ["c", None, None]],
+        [-3, ["a", None, None]],
+        [4, [None, None, None]],
+    ],
+)
+def test_str_get(i, exp):
+    ser = pd.Series(["abc", "de", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.get(i)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.xfail(
+    reason="TODO: StringMethods._validate should support Arrow list types",
+    raises=AttributeError,
+)
+def test_str_join():
+    ser = pd.Series(ArrowExtensionArray(pa.array([list("abc"), list("123"), None])))
+    result = ser.str.join("=")
+    expected = pd.Series(["a=b=c", "1=2=3", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_join_string_type():
+    ser = pd.Series(ArrowExtensionArray(pa.array(["abc", "123", None])))
+    result = ser.str.join("=")
+    expected = pd.Series(["a=b=c", "1=2=3", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start, stop, step, exp",
+    [
+        [None, 2, None, ["ab", None]],
+        [None, 2, 1, ["ab", None]],
+        [1, 3, 1, ["bc", None]],
+        (None, None, -1, ["dcba", None]),
+    ],
+)
+def test_str_slice(start, stop, step, exp):
+    ser = pd.Series(["abcd", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.slice(start, stop, step)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start, stop, repl, exp",
+    [
+        [1, 2, "x", ["axcd", None]],
+        [None, 2, "x", ["xcd", None]],
+        [None, 2, None, ["cd", None]],
+    ],
+)
+def test_str_slice_replace(start, stop, repl, exp):
+    ser = pd.Series(["abcd", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.slice_replace(start, stop, repl)
+    expected = pd.Series(exp, dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "value, method, exp",
+    [
+        ["a1c", "isalnum", True],
+        ["!|,", "isalnum", False],
+        ["aaa", "isalpha", True],
+        ["!!!", "isalpha", False],
+        ["٠", "isdecimal", True],  # noqa: RUF001
+        ["~!", "isdecimal", False],
+        ["2", "isdigit", True],
+        ["~", "isdigit", False],
+        ["aaa", "islower", True],
+        ["aaA", "islower", False],
+        ["123", "isnumeric", True],
+        ["11I", "isnumeric", False],
+        [" ", "isspace", True],
+        ["", "isspace", False],
+        ["The That", "istitle", True],
+        ["the That", "istitle", False],
+        ["AAA", "isupper", True],
+        ["AAc", "isupper", False],
+    ],
+)
+def test_str_is_functions(value, method, exp):
+    ser = pd.Series([value, None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, method)()
+    expected = pd.Series([exp, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        ["capitalize", "Abc def"],
+        ["title", "Abc Def"],
+        ["swapcase", "AbC Def"],
+        ["lower", "abc def"],
+        ["upper", "ABC DEF"],
+        ["casefold", "abc def"],
+    ],
+)
+def test_str_transform_functions(method, exp):
+    ser = pd.Series(["aBc dEF", None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, method)()
+    expected = pd.Series([exp, None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_len():
+    ser = pd.Series(["abcd", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.len()
+    expected = pd.Series([4, None], dtype=ArrowDtype(pa.int32()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, to_strip, val",
+    [
+        ["strip", None, " abc "],
+        ["strip", "x", "xabcx"],
+        ["lstrip", None, " abc"],
+        ["lstrip", "x", "xabc"],
+        ["rstrip", None, "abc "],
+        ["rstrip", "x", "abcx"],
+    ],
+)
+def test_str_strip(method, to_strip, val):
+    ser = pd.Series([val, None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, method)(to_strip=to_strip)
+    expected = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", ["abc123", "abc"])
+def test_str_removesuffix(val):
+    ser = pd.Series([val, None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.removesuffix("123")
+    expected = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", ["123abc", "abc"])
+def test_str_removeprefix(val):
+    ser = pd.Series([val, None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.removeprefix("123")
+    expected = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("errors", ["ignore", "strict"])
+@pytest.mark.parametrize(
+    "encoding, exp",
+    [
+        ("utf8", {"little": b"abc", "big": "abc"}),
+        (
+            "utf32",
+            {
+                "little": b"\xff\xfe\x00\x00a\x00\x00\x00b\x00\x00\x00c\x00\x00\x00",
+                "big": b"\x00\x00\xfe\xff\x00\x00\x00a\x00\x00\x00b\x00\x00\x00c",
+            },
+        ),
+    ],
+    ids=["utf8", "utf32"],
+)
+def test_str_encode(errors, encoding, exp):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.encode(encoding, errors)
+    expected = pd.Series([exp[sys.byteorder], None], dtype=ArrowDtype(pa.binary()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("flags", [0, 2])
+def test_str_findall(flags):
+    ser = pd.Series(["abc", "efg", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.findall("b", flags=flags)
+    expected = pd.Series([["b"], [], None], dtype=ArrowDtype(pa.list_(pa.string())))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["index", "rindex"])
+@pytest.mark.parametrize(
+    "start, end",
+    [
+        [0, None],
+        [1, 4],
+    ],
+)
+def test_str_r_index(method, start, end):
+    ser = pd.Series(["abcba", None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, method)("c", start, end)
+    expected = pd.Series([2, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+    with pytest.raises(ValueError, match="substring not found"):
+        getattr(ser.str, method)("foo", start, end)
+
+
+@pytest.mark.parametrize("form", ["NFC", "NFKC"])
+def test_str_normalize(form):
+    ser = pd.Series(["abc", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.normalize(form)
+    expected = ser.copy()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start, end",
+    [
+        [0, None],
+        [1, 4],
+    ],
+)
+def test_str_rfind(start, end):
+    ser = pd.Series(["abcba", "foo", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.rfind("c", start, end)
+    expected = pd.Series([2, -1, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_translate():
+    ser = pd.Series(["abcba", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.translate({97: "b"})
+    expected = pd.Series(["bbcbb", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_wrap():
+    ser = pd.Series(["abcba", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.wrap(3)
+    expected = pd.Series(["abc\nba", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_get_dummies():
+    ser = pd.Series(["a|b", None, "a|c"], dtype=ArrowDtype(pa.string()))
+    result = ser.str.get_dummies()
+    expected = pd.DataFrame(
+        [[True, True, False], [False, False, False], [True, False, True]],
+        dtype=ArrowDtype(pa.bool_()),
+        columns=["a", "b", "c"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_str_partition():
+    ser = pd.Series(["abcba", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.partition("b")
+    expected = pd.DataFrame(
+        [["a", "b", "cba"], [None, None, None]],
+        dtype=ArrowDtype(pa.string()),
+        columns=pd.RangeIndex(3),
+    )
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+    result = ser.str.partition("b", expand=False)
+    expected = pd.Series(ArrowExtensionArray(pa.array([["a", "b", "cba"], None])))
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rpartition("b")
+    expected = pd.DataFrame(
+        [["abc", "b", "a"], [None, None, None]],
+        dtype=ArrowDtype(pa.string()),
+        columns=pd.RangeIndex(3),
+    )
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+    result = ser.str.rpartition("b", expand=False)
+    expected = pd.Series(ArrowExtensionArray(pa.array([["abc", "b", "a"], None])))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["rsplit", "split"])
+def test_str_split_pat_none(method):
+    # GH 56271
+    ser = pd.Series(["a1 cbc\nb", None], dtype=ArrowDtype(pa.string()))
+    result = getattr(ser.str, method)()
+    expected = pd.Series(ArrowExtensionArray(pa.array([["a1", "cbc", "b"], None])))
+    tm.assert_series_equal(result, expected)
+
+
+def test_str_split():
+    # GH 52401
+    ser = pd.Series(["a1cbcb", "a2cbcb", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.split("c")
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([["a1", "b", "b"], ["a2", "b", "b"], None]))
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.split("c", n=1)
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([["a1", "bcb"], ["a2", "bcb"], None]))
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.split("[1-2]", regex=True)
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([["a", "cbcb"], ["a", "cbcb"], None]))
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.split("[1-2]", regex=True, expand=True)
+    expected = pd.DataFrame(
+        {
+            0: ArrowExtensionArray(pa.array(["a", "a", None])),
+            1: ArrowExtensionArray(pa.array(["cbcb", "cbcb", None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = ser.str.split("1", expand=True)
+    expected = pd.DataFrame(
+        {
+            0: ArrowExtensionArray(pa.array(["a", "a2cbcb", None])),
+            1: ArrowExtensionArray(pa.array(["cbcb", None, None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_str_rsplit():
+    # GH 52401
+    ser = pd.Series(["a1cbcb", "a2cbcb", None], dtype=ArrowDtype(pa.string()))
+    result = ser.str.rsplit("c")
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([["a1", "b", "b"], ["a2", "b", "b"], None]))
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rsplit("c", n=1)
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([["a1cb", "b"], ["a2cb", "b"], None]))
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rsplit("c", n=1, expand=True)
+    expected = pd.DataFrame(
+        {
+            0: ArrowExtensionArray(pa.array(["a1cb", "a2cb", None])),
+            1: ArrowExtensionArray(pa.array(["b", "b", None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = ser.str.rsplit("1", expand=True)
+    expected = pd.DataFrame(
+        {
+            0: ArrowExtensionArray(pa.array(["a", "a2cbcb", None])),
+            1: ArrowExtensionArray(pa.array(["cbcb", None, None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_str_extract_non_symbolic():
+    ser = pd.Series(["a1", "b2", "c3"], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(ValueError, match="pat=.* must contain a symbolic group name."):
+        ser.str.extract(r"[ab](\d)")
+
+
+@pytest.mark.parametrize("expand", [True, False])
+def test_str_extract(expand):
+    ser = pd.Series(["a1", "b2", "c3"], dtype=ArrowDtype(pa.string()))
+    result = ser.str.extract(r"(?P<letter>[ab])(?P<digit>\d)", expand=expand)
+    expected = pd.DataFrame(
+        {
+            "letter": ArrowExtensionArray(pa.array(["a", "b", None])),
+            "digit": ArrowExtensionArray(pa.array(["1", "2", None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_str_extract_expand():
+    ser = pd.Series(["a1", "b2", "c3"], dtype=ArrowDtype(pa.string()))
+    result = ser.str.extract(r"[ab](?P<digit>\d)", expand=True)
+    expected = pd.DataFrame(
+        {
+            "digit": ArrowExtensionArray(pa.array(["1", "2", None])),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = ser.str.extract(r"[ab](?P<digit>\d)", expand=False)
+    expected = pd.Series(ArrowExtensionArray(pa.array(["1", "2", None])), name="digit")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("unit", ["ns", "us", "ms", "s"])
+def test_duration_from_strings_with_nat(unit):
+    # GH51175
+    strings = ["1000", "NaT"]
+    pa_type = pa.duration(unit)
+    dtype = ArrowDtype(pa_type)
+    result = ArrowExtensionArray._from_sequence_of_strings(strings, dtype=dtype)
+    expected = ArrowExtensionArray(pa.array([1000, None], type=pa_type))
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_unsupported_dt(data):
+    pa_dtype = data.dtype.pyarrow_dtype
+    if not pa.types.is_temporal(pa_dtype):
+        with pytest.raises(
+            AttributeError, match="Can only use .dt accessor with datetimelike values"
+        ):
+            pd.Series(data).dt
+
+
+@pytest.mark.parametrize(
+    "prop, expected",
+    [
+        ["year", 2023],
+        ["day", 2],
+        ["day_of_week", 0],
+        ["dayofweek", 0],
+        ["weekday", 0],
+        ["day_of_year", 2],
+        ["dayofyear", 2],
+        ["hour", 3],
+        ["minute", 4],
+        ["is_leap_year", False],
+        ["microsecond", 2000],
+        ["month", 1],
+        ["nanosecond", 6],
+        ["quarter", 1],
+        ["second", 7],
+        ["date", date(2023, 1, 2)],
+        ["time", time(3, 4, 7, 2000)],
+    ],
+)
+def test_dt_properties(prop, expected):
+    ser = pd.Series(
+        [
+            pd.Timestamp(
+                year=2023,
+                month=1,
+                day=2,
+                hour=3,
+                minute=4,
+                second=7,
+                microsecond=2000,
+                nanosecond=6,
+            ),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    result = getattr(ser.dt, prop)
+    exp_type = None
+    if isinstance(expected, date):
+        exp_type = pa.date32()
+    elif isinstance(expected, time):
+        exp_type = pa.time64("ns")
+    expected = pd.Series(ArrowExtensionArray(pa.array([expected, None], type=exp_type)))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("microsecond", [2000, 5, 0])
+def test_dt_microsecond(microsecond):
+    # GH 59183
+    ser = pd.Series(
+        [
+            pd.Timestamp(
+                year=2024,
+                month=7,
+                day=7,
+                second=5,
+                microsecond=microsecond,
+                nanosecond=6,
+            ),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    result = ser.dt.microsecond
+    expected = pd.Series([microsecond, None], dtype="int64[pyarrow]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_is_month_start_end():
+    ser = pd.Series(
+        [
+            datetime(year=2023, month=12, day=2, hour=3),
+            datetime(year=2023, month=1, day=1, hour=3),
+            datetime(year=2023, month=3, day=31, hour=3),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    result = ser.dt.is_month_start
+    expected = pd.Series([False, True, False, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+    result = ser.dt.is_month_end
+    expected = pd.Series([False, False, True, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_is_year_start_end():
+    ser = pd.Series(
+        [
+            datetime(year=2023, month=12, day=31, hour=3),
+            datetime(year=2023, month=1, day=1, hour=3),
+            datetime(year=2023, month=3, day=31, hour=3),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    result = ser.dt.is_year_start
+    expected = pd.Series([False, True, False, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+    result = ser.dt.is_year_end
+    expected = pd.Series([True, False, False, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_is_quarter_start_end():
+    ser = pd.Series(
+        [
+            datetime(year=2023, month=11, day=30, hour=3),
+            datetime(year=2023, month=1, day=1, hour=3),
+            datetime(year=2023, month=3, day=31, hour=3),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    result = ser.dt.is_quarter_start
+    expected = pd.Series([False, True, False, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+    result = ser.dt.is_quarter_end
+    expected = pd.Series([False, False, True, None], dtype=ArrowDtype(pa.bool_()))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["days_in_month", "daysinmonth"])
+def test_dt_days_in_month(method):
+    ser = pd.Series(
+        [
+            datetime(year=2023, month=3, day=30, hour=3),
+            datetime(year=2023, month=4, day=1, hour=3),
+            datetime(year=2023, month=2, day=3, hour=3),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    result = getattr(ser.dt, method)
+    expected = pd.Series([31, 30, 28, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_normalize():
+    ser = pd.Series(
+        [
+            datetime(year=2023, month=3, day=30),
+            datetime(year=2023, month=4, day=1, hour=3),
+            datetime(year=2023, month=2, day=3, hour=23, minute=59, second=59),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    result = ser.dt.normalize()
+    expected = pd.Series(
+        [
+            datetime(year=2023, month=3, day=30),
+            datetime(year=2023, month=4, day=1),
+            datetime(year=2023, month=2, day=3),
+            None,
+        ],
+        dtype=ArrowDtype(pa.timestamp("us")),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("unit", ["us", "ns"])
+def test_dt_time_preserve_unit(unit):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp(unit)),
+    )
+    assert ser.dt.unit == unit
+
+    result = ser.dt.time
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([time(3, 0), None], type=pa.time64(unit)))
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("tz", [None, "UTC", "US/Pacific"])
+def test_dt_tz(tz):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns", tz=tz)),
+    )
+    result = ser.dt.tz
+    assert result == timezones.maybe_get_tz(tz)
+
+
+def test_dt_isocalendar():
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    result = ser.dt.isocalendar()
+    expected = pd.DataFrame(
+        [[2023, 1, 1], [0, 0, 0]],
+        columns=["year", "week", "day"],
+        dtype="int64[pyarrow]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp", [["day_name", "Sunday"], ["month_name", "January"]]
+)
+def test_dt_day_month_name(method, exp, request):
+    # GH 52388
+    _require_timezone_database(request)
+
+    ser = pd.Series([datetime(2023, 1, 1), None], dtype=ArrowDtype(pa.timestamp("ms")))
+    result = getattr(ser.dt, method)()
+    expected = pd.Series([exp, None], dtype=ArrowDtype(pa.string()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_strftime(request):
+    _require_timezone_database(request)
+
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    result = ser.dt.strftime("%Y-%m-%dT%H:%M:%S")
+    expected = pd.Series(
+        ["2023-01-02T03:00:00.000000000", None], dtype=ArrowDtype(pa.string())
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["ceil", "floor", "round"])
+def test_dt_roundlike_tz_options_not_supported(method):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    with pytest.raises(NotImplementedError, match="ambiguous is not supported."):
+        getattr(ser.dt, method)("1h", ambiguous="NaT")
+
+    with pytest.raises(NotImplementedError, match="nonexistent is not supported."):
+        getattr(ser.dt, method)("1h", nonexistent="NaT")
+
+
+@pytest.mark.parametrize("method", ["ceil", "floor", "round"])
+def test_dt_roundlike_unsupported_freq(method):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    with pytest.raises(ValueError, match="freq='1B' is not supported"):
+        getattr(ser.dt, method)("1B")
+
+    with pytest.raises(ValueError, match="Must specify a valid frequency: None"):
+        getattr(ser.dt, method)(None)
+
+
+@pytest.mark.parametrize("freq", ["D", "h", "min", "s", "ms", "us", "ns"])
+@pytest.mark.parametrize("method", ["ceil", "floor", "round"])
+def test_dt_ceil_year_floor(freq, method):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=1), None],
+    )
+    pa_dtype = ArrowDtype(pa.timestamp("ns"))
+    expected = getattr(ser.dt, method)(f"1{freq}").astype(pa_dtype)
+    result = getattr(ser.astype(pa_dtype).dt, method)(f"1{freq}")
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_to_pydatetime():
+    # GH 51859
+    data = [datetime(2022, 1, 1), datetime(2023, 1, 1)]
+    ser = pd.Series(data, dtype=ArrowDtype(pa.timestamp("ns")))
+    result = ser.dt.to_pydatetime()
+    expected = pd.Series(data, dtype=object)
+    tm.assert_series_equal(result, expected)
+    assert all(type(expected.iloc[i]) is datetime for i in range(len(expected)))
+
+    expected = ser.astype("datetime64[ns]").dt.to_pydatetime()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("date_type", [32, 64])
+def test_dt_to_pydatetime_date_error(date_type):
+    # GH 52812
+    ser = pd.Series(
+        [date(2022, 12, 31)],
+        dtype=ArrowDtype(getattr(pa, f"date{date_type}")()),
+    )
+    with pytest.raises(ValueError, match="to_pydatetime cannot be called with"):
+        ser.dt.to_pydatetime()
+
+
+def test_dt_tz_localize_unsupported_tz_options():
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    with pytest.raises(NotImplementedError, match="ambiguous='NaT' is not supported"):
+        ser.dt.tz_localize("UTC", ambiguous="NaT")
+
+    with pytest.raises(NotImplementedError, match="nonexistent='NaT' is not supported"):
+        ser.dt.tz_localize("UTC", nonexistent="NaT")
+
+
+def test_dt_tz_localize_none():
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns", tz="US/Pacific")),
+    )
+    result = ser.dt.tz_localize(None)
+    expected = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("unit", ["us", "ns"])
+def test_dt_tz_localize(unit, request):
+    _require_timezone_database(request)
+
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp(unit)),
+    )
+    result = ser.dt.tz_localize("US/Pacific")
+    exp_data = pa.array(
+        [datetime(year=2023, month=1, day=2, hour=3), None], type=pa.timestamp(unit)
+    )
+    exp_data = pa.compute.assume_timezone(exp_data, "US/Pacific")
+    expected = pd.Series(ArrowExtensionArray(exp_data))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "nonexistent, exp_date",
+    [
+        ["shift_forward", datetime(year=2023, month=3, day=12, hour=3)],
+        ["shift_backward", pd.Timestamp("2023-03-12 01:59:59.999999999")],
+    ],
+)
+def test_dt_tz_localize_nonexistent(nonexistent, exp_date, request):
+    _require_timezone_database(request)
+
+    ser = pd.Series(
+        [datetime(year=2023, month=3, day=12, hour=2, minute=30), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    result = ser.dt.tz_localize("US/Pacific", nonexistent=nonexistent)
+    exp_data = pa.array([exp_date, None], type=pa.timestamp("ns"))
+    exp_data = pa.compute.assume_timezone(exp_data, "US/Pacific")
+    expected = pd.Series(ArrowExtensionArray(exp_data))
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_tz_convert_not_tz_raises():
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    with pytest.raises(TypeError, match="Cannot convert tz-naive timestamps"):
+        ser.dt.tz_convert("UTC")
+
+
+def test_dt_tz_convert_none():
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns", "US/Pacific")),
+    )
+    result = ser.dt.tz_convert(None)
+    expected = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp("ns")),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("unit", ["us", "ns"])
+def test_dt_tz_convert(unit):
+    ser = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp(unit, "US/Pacific")),
+    )
+    result = ser.dt.tz_convert("US/Eastern")
+    expected = pd.Series(
+        [datetime(year=2023, month=1, day=2, hour=3), None],
+        dtype=ArrowDtype(pa.timestamp(unit, "US/Eastern")),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["timestamp[ms][pyarrow]", "duration[ms][pyarrow]"])
+def test_as_unit(dtype):
+    # GH 52284
+    ser = pd.Series([1000, None], dtype=dtype)
+    result = ser.dt.as_unit("ns")
+    expected = ser.astype(dtype.replace("ms", "ns"))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "prop, expected",
+    [
+        ["days", 1],
+        ["seconds", 2],
+        ["microseconds", 3],
+        ["nanoseconds", 4],
+    ],
+)
+def test_dt_timedelta_properties(prop, expected):
+    # GH 52284
+    ser = pd.Series(
+        [
+            pd.Timedelta(
+                days=1,
+                seconds=2,
+                microseconds=3,
+                nanoseconds=4,
+            ),
+            None,
+        ],
+        dtype=ArrowDtype(pa.duration("ns")),
+    )
+    result = getattr(ser.dt, prop)
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([expected, None], type=pa.int32()))
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_timedelta_total_seconds():
+    # GH 52284
+    ser = pd.Series(
+        [
+            pd.Timedelta(
+                days=1,
+                seconds=2,
+                microseconds=3,
+                nanoseconds=4,
+            ),
+            None,
+        ],
+        dtype=ArrowDtype(pa.duration("ns")),
+    )
+    result = ser.dt.total_seconds()
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([86402.000003, None], type=pa.float64()))
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_dt_to_pytimedelta():
+    # GH 52284
+    data = [timedelta(1, 2, 3), timedelta(1, 2, 4)]
+    ser = pd.Series(data, dtype=ArrowDtype(pa.duration("ns")))
+
+    msg = "The behavior of ArrowTemporalProperties.to_pytimedelta is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = ser.dt.to_pytimedelta()
+    expected = np.array(data, dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+    assert all(type(res) is timedelta for res in result)
+
+    msg = "The behavior of TimedeltaProperties.to_pytimedelta is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        expected = ser.astype("timedelta64[ns]").dt.to_pytimedelta()
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_dt_components():
+    # GH 52284
+    ser = pd.Series(
+        [
+            pd.Timedelta(
+                days=1,
+                seconds=2,
+                microseconds=3,
+                nanoseconds=4,
+            ),
+            None,
+        ],
+        dtype=ArrowDtype(pa.duration("ns")),
+    )
+    result = ser.dt.components
+    expected = pd.DataFrame(
+        [[1, 0, 0, 2, 0, 3, 4], [None, None, None, None, None, None, None]],
+        columns=[
+            "days",
+            "hours",
+            "minutes",
+            "seconds",
+            "milliseconds",
+            "microseconds",
+            "nanoseconds",
+        ],
+        dtype="int32[pyarrow]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dt_components_large_values():
+    ser = pd.Series(
+        [
+            pd.Timedelta("365 days 23:59:59.999000"),
+            None,
+        ],
+        dtype=ArrowDtype(pa.duration("ns")),
+    )
+    result = ser.dt.components
+    expected = pd.DataFrame(
+        [[365, 23, 59, 59, 999, 0, 0], [None, None, None, None, None, None, None]],
+        columns=[
+            "days",
+            "hours",
+            "minutes",
+            "seconds",
+            "milliseconds",
+            "microseconds",
+            "nanoseconds",
+        ],
+        dtype="int32[pyarrow]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("skipna", [True, False])
+def test_boolean_reduce_series_all_null(all_boolean_reductions, skipna):
+    # GH51624
+    ser = pd.Series([None], dtype="float64[pyarrow]")
+    result = getattr(ser, all_boolean_reductions)(skipna=skipna)
+    if skipna:
+        expected = all_boolean_reductions == "all"
+    else:
+        expected = pd.NA
+    assert result is expected
+
+
+def test_from_sequence_of_strings_boolean():
+    true_strings = ["true", "TRUE", "True", "1", "1.0"]
+    false_strings = ["false", "FALSE", "False", "0", "0.0"]
+    nulls = [None]
+    strings = true_strings + false_strings + nulls
+    bools = (
+        [True] * len(true_strings) + [False] * len(false_strings) + [None] * len(nulls)
+    )
+
+    dtype = ArrowDtype(pa.bool_())
+    result = ArrowExtensionArray._from_sequence_of_strings(strings, dtype=dtype)
+    expected = pd.array(bools, dtype="boolean[pyarrow]")
+    tm.assert_extension_array_equal(result, expected)
+
+    strings = ["True", "foo"]
+    with pytest.raises(pa.ArrowInvalid, match="Failed to parse"):
+        ArrowExtensionArray._from_sequence_of_strings(strings, dtype=dtype)
+
+
+def test_concat_empty_arrow_backed_series(dtype):
+    # GH#51734
+    ser = pd.Series([], dtype=dtype)
+    expected = ser.copy()
+    result = pd.concat([ser[np.array([], dtype=np.bool_)]])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["string", "string[pyarrow]"])
+def test_series_from_string_array(dtype):
+    arr = pa.array("the quick brown fox".split())
+    ser = pd.Series(arr, dtype=dtype)
+    expected = pd.Series(ArrowExtensionArray(arr), dtype=dtype)
+    tm.assert_series_equal(ser, expected)
+
+
+# _data was renamed to _pa_data
+class OldArrowExtensionArray(ArrowExtensionArray):
+    def __getstate__(self):
+        state = super().__getstate__()
+        state["_data"] = state.pop("_pa_array")
+        return state
+
+
+def test_pickle_old_arrowextensionarray():
+    data = pa.array([1])
+    expected = OldArrowExtensionArray(data)
+    result = pickle.loads(pickle.dumps(expected))
+    tm.assert_extension_array_equal(result, expected)
+    assert result._pa_array == pa.chunked_array(data)
+    assert not hasattr(result, "_data")
+
+
+def test_setitem_boolean_replace_with_mask_segfault():
+    # GH#52059
+    N = 145_000
+    arr = ArrowExtensionArray(pa.chunked_array([np.ones((N,), dtype=np.bool_)]))
+    expected = arr.copy()
+    arr[np.zeros((N,), dtype=np.bool_)] = False
+    assert arr._pa_array == expected._pa_array
+
+
+@pytest.mark.parametrize(
+    "data, arrow_dtype",
+    [
+        ([b"a", b"b"], pa.large_binary()),
+        (["a", "b"], pa.large_string()),
+    ],
+)
+def test_conversion_large_dtypes_from_numpy_array(data, arrow_dtype):
+    dtype = ArrowDtype(arrow_dtype)
+    result = pd.array(np.array(data), dtype=dtype)
+    expected = pd.array(data, dtype=dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_concat_null_array():
+    df = pd.DataFrame({"a": [None, None]}, dtype=ArrowDtype(pa.null()))
+    df2 = pd.DataFrame({"a": [0, 1]}, dtype="int64[pyarrow]")
+
+    result = pd.concat([df, df2], ignore_index=True)
+    expected = pd.DataFrame({"a": [None, None, 0, 1]}, dtype="int64[pyarrow]")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.ALL_INT_PYARROW_DTYPES + tm.FLOAT_PYARROW_DTYPES)
+def test_describe_numeric_data(pa_type):
+    # GH 52470
+    data = pd.Series([1, 2, 3], dtype=ArrowDtype(pa_type))
+    result = data.describe()
+    expected = pd.Series(
+        [3, 2, 1, 1, 1.5, 2.0, 2.5, 3],
+        dtype=ArrowDtype(pa.float64()),
+        index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.TIMEDELTA_PYARROW_DTYPES)
+def test_describe_timedelta_data(pa_type):
+    # GH53001
+    data = pd.Series(range(1, 10), dtype=ArrowDtype(pa_type))
+    result = data.describe()
+    expected = pd.Series(
+        [9] + pd.to_timedelta([5, 2, 1, 3, 5, 7, 9], unit=pa_type.unit).tolist(),
+        dtype=object,
+        index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.DATETIME_PYARROW_DTYPES)
+def test_describe_datetime_data(pa_type):
+    # GH53001
+    data = pd.Series(range(1, 10), dtype=ArrowDtype(pa_type))
+    result = data.describe()
+    expected = pd.Series(
+        [9]
+        + [
+            pd.Timestamp(v, tz=pa_type.tz, unit=pa_type.unit)
+            for v in [5, 1, 3, 5, 7, 9]
+        ],
+        dtype=object,
+        index=["count", "mean", "min", "25%", "50%", "75%", "max"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_quantile_temporal(pa_type):
+    # GH52678
+    data = [1, 2, 3]
+    ser = pd.Series(data, dtype=ArrowDtype(pa_type))
+    result = ser.quantile(0.1)
+    expected = ser[0]
+    assert result == expected
+
+
+def test_date32_repr():
+    # GH48238
+    arrow_dt = pa.array([date.fromisoformat("2020-01-01")], type=pa.date32())
+    ser = pd.Series(arrow_dt, dtype=ArrowDtype(arrow_dt.type))
+    assert repr(ser) == "0    2020-01-01\ndtype: date32[day][pyarrow]"
+
+
+def test_duration_overflow_from_ndarray_containing_nat():
+    # GH52843
+    data_ts = pd.to_datetime([1, None])
+    data_td = pd.to_timedelta([1, None])
+    ser_ts = pd.Series(data_ts, dtype=ArrowDtype(pa.timestamp("ns")))
+    ser_td = pd.Series(data_td, dtype=ArrowDtype(pa.duration("ns")))
+    result = ser_ts + ser_td
+    expected = pd.Series([2, None], dtype=ArrowDtype(pa.timestamp("ns")))
+    tm.assert_series_equal(result, expected)
+
+
+def test_infer_dtype_pyarrow_dtype(data, request):
+    res = lib.infer_dtype(data)
+    assert res != "unknown-array"
+
+    if data._hasna and res in ["floating", "datetime64", "timedelta64"]:
+        mark = pytest.mark.xfail(
+            reason="in infer_dtype pd.NA is not ignored in these cases "
+            "even with skipna=True in the list(data) check below"
+        )
+        request.applymarker(mark)
+
+    assert res == lib.infer_dtype(list(data), skipna=True)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_from_sequence_temporal(pa_type):
+    # GH 53171
+    val = 3
+    unit = pa_type.unit
+    if pa.types.is_duration(pa_type):
+        seq = [pd.Timedelta(val, unit=unit).as_unit(unit)]
+    else:
+        seq = [pd.Timestamp(val, unit=unit, tz=pa_type.tz).as_unit(unit)]
+
+    result = ArrowExtensionArray._from_sequence(seq, dtype=pa_type)
+    expected = ArrowExtensionArray(pa.array([val], type=pa_type))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_setitem_temporal(pa_type):
+    # GH 53171
+    unit = pa_type.unit
+    if pa.types.is_duration(pa_type):
+        val = pd.Timedelta(1, unit=unit).as_unit(unit)
+    else:
+        val = pd.Timestamp(1, unit=unit, tz=pa_type.tz).as_unit(unit)
+
+    arr = ArrowExtensionArray(pa.array([1, 2, 3], type=pa_type))
+
+    result = arr.copy()
+    result[:] = val
+    expected = ArrowExtensionArray(pa.array([1, 1, 1], type=pa_type))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_arithmetic_temporal(pa_type, request):
+    # GH 53171
+    arr = ArrowExtensionArray(pa.array([1, 2, 3], type=pa_type))
+    unit = pa_type.unit
+    result = arr - pd.Timedelta(1, unit=unit).as_unit(unit)
+    expected = ArrowExtensionArray(pa.array([0, 1, 2], type=pa_type))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_comparison_temporal(pa_type):
+    # GH 53171
+    unit = pa_type.unit
+    if pa.types.is_duration(pa_type):
+        val = pd.Timedelta(1, unit=unit).as_unit(unit)
+    else:
+        val = pd.Timestamp(1, unit=unit, tz=pa_type.tz).as_unit(unit)
+
+    arr = ArrowExtensionArray(pa.array([1, 2, 3], type=pa_type))
+
+    result = arr > val
+    expected = ArrowExtensionArray(pa.array([False, True, True], type=pa.bool_()))
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_getitem_temporal(pa_type):
+    # GH 53326
+    arr = ArrowExtensionArray(pa.array([1, 2, 3], type=pa_type))
+    result = arr[1]
+    if pa.types.is_duration(pa_type):
+        expected = pd.Timedelta(2, unit=pa_type.unit).as_unit(pa_type.unit)
+        assert isinstance(result, pd.Timedelta)
+    else:
+        expected = pd.Timestamp(2, unit=pa_type.unit, tz=pa_type.tz).as_unit(
+            pa_type.unit
+        )
+        assert isinstance(result, pd.Timestamp)
+    assert result.unit == expected.unit
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES
+)
+def test_iter_temporal(pa_type):
+    # GH 53326
+    arr = ArrowExtensionArray(pa.array([1, None], type=pa_type))
+    result = list(arr)
+    if pa.types.is_duration(pa_type):
+        expected = [
+            pd.Timedelta(1, unit=pa_type.unit).as_unit(pa_type.unit),
+            pd.NA,
+        ]
+        assert isinstance(result[0], pd.Timedelta)
+    else:
+        expected = [
+            pd.Timestamp(1, unit=pa_type.unit, tz=pa_type.tz).as_unit(pa_type.unit),
+            pd.NA,
+        ]
+        assert isinstance(result[0], pd.Timestamp)
+    assert result[0].unit == expected[0].unit
+    assert result == expected
+
+
+def test_groupby_series_size_returns_pa_int(data):
+    # GH 54132
+    ser = pd.Series(data[:3], index=["a", "a", "b"])
+    result = ser.groupby(level=0).size()
+    expected = pd.Series([2, 1], dtype="int64[pyarrow]", index=["a", "b"])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pa_type", tm.DATETIME_PYARROW_DTYPES + tm.TIMEDELTA_PYARROW_DTYPES, ids=repr
+)
+@pytest.mark.parametrize("dtype", [None, object])
+def test_to_numpy_temporal(pa_type, dtype):
+    # GH 53326
+    # GH 55997: Return datetime64/timedelta64 types with NaT if possible
+    arr = ArrowExtensionArray(pa.array([1, None], type=pa_type))
+    result = arr.to_numpy(dtype=dtype)
+    if pa.types.is_duration(pa_type):
+        value = pd.Timedelta(1, unit=pa_type.unit).as_unit(pa_type.unit)
+    else:
+        value = pd.Timestamp(1, unit=pa_type.unit, tz=pa_type.tz).as_unit(pa_type.unit)
+
+    if dtype == object or (pa.types.is_timestamp(pa_type) and pa_type.tz is not None):
+        if dtype == object:
+            na = pd.NA
+        else:
+            na = pd.NaT
+        expected = np.array([value, na], dtype=object)
+        assert result[0].unit == value.unit
+    else:
+        na = pa_type.to_pandas_dtype().type("nat", pa_type.unit)
+        value = value.to_numpy()
+        expected = np.array([value, na])
+        assert np.datetime_data(result[0])[0] == pa_type.unit
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_groupby_count_return_arrow_dtype(data_missing):
+    df = pd.DataFrame({"A": [1, 1], "B": data_missing, "C": data_missing})
+    result = df.groupby("A").count()
+    expected = pd.DataFrame(
+        [[1, 1]],
+        index=pd.Index([1], name="A"),
+        columns=["B", "C"],
+        dtype="int64[pyarrow]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_fixed_size_list():
+    # GH#55000
+    ser = pd.Series(
+        [[1, 2], [3, 4]], dtype=ArrowDtype(pa.list_(pa.int64(), list_size=2))
+    )
+    result = ser.dtype.type
+    assert result == list
+
+
+def test_arrowextensiondtype_dataframe_repr():
+    # GH 54062
+    df = pd.DataFrame(
+        pd.period_range("2012", periods=3),
+        columns=["col"],
+        dtype=ArrowDtype(ArrowPeriodType("D")),
+    )
+    result = repr(df)
+    # TODO: repr value may not be expected; address how
+    # pyarrow.ExtensionType values are displayed
+    expected = "     col\n0  15340\n1  15341\n2  15342"
+    assert result == expected
+
+
+def test_pow_missing_operand():
+    # GH 55512
+    k = pd.Series([2, None], dtype="int64[pyarrow]")
+    result = k.pow(None, fill_value=3)
+    expected = pd.Series([8, None], dtype="int64[pyarrow]")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.skipif(
+    pa_version_under11p0, reason="Decimal128 to string cast implemented in pyarrow 11"
+)
+def test_decimal_parse_raises():
+    # GH 56984
+    ser = pd.Series(["1.2345"], dtype=ArrowDtype(pa.string()))
+    with pytest.raises(
+        pa.lib.ArrowInvalid, match="Rescaling Decimal128 value would cause data loss"
+    ):
+        ser.astype(ArrowDtype(pa.decimal128(1, 0)))
+
+
+@pytest.mark.skipif(
+    pa_version_under11p0, reason="Decimal128 to string cast implemented in pyarrow 11"
+)
+def test_decimal_parse_succeeds():
+    # GH 56984
+    ser = pd.Series(["1.2345"], dtype=ArrowDtype(pa.string()))
+    dtype = ArrowDtype(pa.decimal128(5, 4))
+    result = ser.astype(dtype)
+    expected = pd.Series([Decimal("1.2345")], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.TIMEDELTA_PYARROW_DTYPES)
+def test_duration_fillna_numpy(pa_type):
+    # GH 54707
+    ser1 = pd.Series([None, 2], dtype=ArrowDtype(pa_type))
+    ser2 = pd.Series(np.array([1, 3], dtype=f"m8[{pa_type.unit}]"))
+    result = ser1.fillna(ser2)
+    expected = pd.Series([1, 2], dtype=ArrowDtype(pa_type))
+    tm.assert_series_equal(result, expected)
+
+
+def test_comparison_not_propagating_arrow_error():
+    # GH#54944
+    a = pd.Series([1 << 63], dtype="uint64[pyarrow]")
+    b = pd.Series([None], dtype="int64[pyarrow]")
+    with pytest.raises(pa.lib.ArrowInvalid, match="Integer value"):
+        a < b
+
+
+def test_factorize_chunked_dictionary():
+    # GH 54844
+    pa_array = pa.chunked_array(
+        [pa.array(["a"]).dictionary_encode(), pa.array(["b"]).dictionary_encode()]
+    )
+    ser = pd.Series(ArrowExtensionArray(pa_array))
+    res_indices, res_uniques = ser.factorize()
+    exp_indicies = np.array([0, 1], dtype=np.intp)
+    exp_uniques = pd.Index(ArrowExtensionArray(pa_array.combine_chunks()))
+    tm.assert_numpy_array_equal(res_indices, exp_indicies)
+    tm.assert_index_equal(res_uniques, exp_uniques)
+
+
+def test_factorize_dictionary_with_na():
+    # GH#60567
+    arr = pd.array(
+        ["a1", pd.NA], dtype=ArrowDtype(pa.dictionary(pa.int32(), pa.utf8()))
+    )
+    indices, uniques = arr.factorize(use_na_sentinel=False)
+    expected_indices = np.array([0, 1], dtype=np.intp)
+    expected_uniques = pd.array(["a1", None], dtype=ArrowDtype(pa.string()))
+    tm.assert_numpy_array_equal(indices, expected_indices)
+    tm.assert_extension_array_equal(uniques, expected_uniques)
+
+
+def test_dictionary_astype_categorical():
+    # GH#56672
+    arrs = [
+        pa.array(np.array(["a", "x", "c", "a"])).dictionary_encode(),
+        pa.array(np.array(["a", "d", "c"])).dictionary_encode(),
+    ]
+    ser = pd.Series(ArrowExtensionArray(pa.chunked_array(arrs)))
+    result = ser.astype("category")
+    categories = pd.Index(["a", "x", "c", "d"], dtype=ArrowDtype(pa.string()))
+    expected = pd.Series(
+        ["a", "x", "c", "a", "a", "d", "c"],
+        dtype=pd.CategoricalDtype(categories=categories),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_arrow_floordiv():
+    # GH 55561
+    a = pd.Series([-7], dtype="int64[pyarrow]")
+    b = pd.Series([4], dtype="int64[pyarrow]")
+    expected = pd.Series([-2], dtype="int64[pyarrow]")
+    result = a // b
+    tm.assert_series_equal(result, expected)
+
+
+def test_arrow_floordiv_large_values():
+    # GH 56645
+    a = pd.Series([1425801600000000000], dtype="int64[pyarrow]")
+    expected = pd.Series([1425801600000], dtype="int64[pyarrow]")
+    result = a // 1_000_000
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int64[pyarrow]", "uint64[pyarrow]"])
+def test_arrow_floordiv_large_integral_result(dtype):
+    # GH 56676
+    a = pd.Series([18014398509481983], dtype=dtype)
+    result = a // 1
+    tm.assert_series_equal(result, a)
+
+
+@pytest.mark.parametrize("pa_type", tm.SIGNED_INT_PYARROW_DTYPES)
+def test_arrow_floordiv_larger_divisor(pa_type):
+    # GH 56676
+    dtype = ArrowDtype(pa_type)
+    a = pd.Series([-23], dtype=dtype)
+    result = a // 24
+    expected = pd.Series([-1], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.SIGNED_INT_PYARROW_DTYPES)
+def test_arrow_floordiv_integral_invalid(pa_type):
+    # GH 56676
+    min_value = np.iinfo(pa_type.to_pandas_dtype()).min
+    a = pd.Series([min_value], dtype=ArrowDtype(pa_type))
+    with pytest.raises(pa.lib.ArrowInvalid, match="overflow|not in range"):
+        a // -1
+    with pytest.raises(pa.lib.ArrowInvalid, match="divide by zero"):
+        a // 0
+
+
+@pytest.mark.parametrize("dtype", tm.FLOAT_PYARROW_DTYPES_STR_REPR)
+def test_arrow_floordiv_floating_0_divisor(dtype):
+    # GH 56676
+    a = pd.Series([2], dtype=dtype)
+    result = a // 0
+    expected = pd.Series([float("inf")], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["float64", "datetime64[ns]", "timedelta64[ns]"])
+def test_astype_int_with_null_to_numpy_dtype(dtype):
+    # GH 57093
+    ser = pd.Series([1, None], dtype="int64[pyarrow]")
+    result = ser.astype(dtype)
+    expected = pd.Series([1, None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("pa_type", tm.ALL_INT_PYARROW_DTYPES)
+def test_arrow_integral_floordiv_large_values(pa_type):
+    # GH 56676
+    max_value = np.iinfo(pa_type.to_pandas_dtype()).max
+    dtype = ArrowDtype(pa_type)
+    a = pd.Series([max_value], dtype=dtype)
+    b = pd.Series([1], dtype=dtype)
+    result = a // b
+    tm.assert_series_equal(result, a)
+
+
+@pytest.mark.parametrize("dtype", ["int64[pyarrow]", "uint64[pyarrow]"])
+def test_arrow_true_division_large_divisor(dtype):
+    # GH 56706
+    a = pd.Series([0], dtype=dtype)
+    b = pd.Series([18014398509481983], dtype=dtype)
+    expected = pd.Series([0], dtype="float64[pyarrow]")
+    result = a / b
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int64[pyarrow]", "uint64[pyarrow]"])
+def test_arrow_floor_division_large_divisor(dtype):
+    # GH 56706
+    a = pd.Series([0], dtype=dtype)
+    b = pd.Series([18014398509481983], dtype=dtype)
+    expected = pd.Series([0], dtype=dtype)
+    result = a // b
+    tm.assert_series_equal(result, expected)
+
+
+def test_string_to_datetime_parsing_cast():
+    # GH 56266
+    string_dates = ["2020-01-01 04:30:00", "2020-01-02 00:00:00", "2020-01-03 00:00:00"]
+    result = pd.Series(string_dates, dtype="timestamp[s][pyarrow]")
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array(pd.to_datetime(string_dates), from_pandas=True))
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.skipif(
+    pa_version_under13p0, reason="pairwise_diff_checked not implemented in pyarrow"
+)
+def test_interpolate_not_numeric(data):
+    if not data.dtype._is_numeric:
+        ser = pd.Series(data)
+        msg = re.escape(f"Cannot interpolate with {ser.dtype} dtype")
+        with pytest.raises(TypeError, match=msg):
+            pd.Series(data).interpolate()
+
+
+@pytest.mark.skipif(
+    pa_version_under13p0, reason="pairwise_diff_checked not implemented in pyarrow"
+)
+@pytest.mark.parametrize("dtype", ["int64[pyarrow]", "float64[pyarrow]"])
+def test_interpolate_linear(dtype):
+    ser = pd.Series([None, 1, 2, None, 4, None], dtype=dtype)
+    result = ser.interpolate()
+    expected = pd.Series([None, 1, 2, 3, 4, None], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_string_to_time_parsing_cast():
+    # GH 56463
+    string_times = ["11:41:43.076160"]
+    result = pd.Series(string_times, dtype="time64[us][pyarrow]")
+    expected = pd.Series(
+        ArrowExtensionArray(pa.array([time(11, 41, 43, 76160)], from_pandas=True))
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_numpy_float():
+    # GH#56267
+    ser = pd.Series([32, 40, None], dtype="float[pyarrow]")
+    result = ser.astype("float64")
+    expected = pd.Series([32, 40, np.nan], dtype="float64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_numpy_timestamp_to_int():
+    # GH 55997
+    ser = pd.Series(["2020-01-01 04:30:00"], dtype="timestamp[ns][pyarrow]")
+    result = ser.to_numpy(dtype=np.int64)
+    expected = np.array([1577853000000000000])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("arrow_type", [pa.large_string(), pa.string()])
+def test_cast_dictionary_different_value_dtype(arrow_type):
+    df = pd.DataFrame({"a": ["x", "y"]}, dtype="string[pyarrow]")
+    data_type = ArrowDtype(pa.dictionary(pa.int32(), arrow_type))
+    result = df.astype({"a": data_type})
+    assert result.dtypes.iloc[0] == data_type
+
+
+def test_map_numeric_na_action():
+    ser = pd.Series([32, 40, None], dtype="int64[pyarrow]")
+    result = ser.map(lambda x: 42, na_action="ignore")
+    expected = pd.Series([42.0, 42.0, np.nan], dtype="float64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_categorical_from_arrow_dictionary():
+    # GH 60563
+    df = pd.DataFrame(
+        {"A": ["a1", "a2"]}, dtype=ArrowDtype(pa.dictionary(pa.int32(), pa.utf8()))
+    )
+    result = df.value_counts(dropna=False)
+    expected = pd.Series(
+        [1, 1],
+        index=pd.MultiIndex.from_arrays(
+            [pd.Index(["a1", "a2"], dtype=ArrowDtype(pa.string()), name="A")]
+        ),
+        name="count",
+        dtype="int64",
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/extension/test_categorical.py b/pandas/pandas/tests/extension/test_categorical.py
new file mode 100644
index 0000000000000..8f8af607585df
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_categorical.py
@@ -0,0 +1,192 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+import string
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas as pd
+from pandas import Categorical
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+from pandas.tests.extension import base
+
+
+def make_data():
+    while True:
+        values = np.random.default_rng(2).choice(list(string.ascii_letters), size=100)
+        # ensure we meet the requirements
+        # 1. first two not null
+        # 2. first and second are different
+        if values[0] != values[1]:
+            break
+    return values
+
+
+@pytest.fixture
+def dtype():
+    return CategoricalDtype()
+
+
+@pytest.fixture
+def data():
+    """Length-100 array for this type.
+
+    * data[0] and data[1] should both be non missing
+    * data[0] and data[1] should not be equal
+    """
+    return Categorical(make_data())
+
+
+@pytest.fixture
+def data_missing():
+    """Length 2 array with [NA, Valid]"""
+    return Categorical([np.nan, "A"])
+
+
+@pytest.fixture
+def data_for_sorting():
+    return Categorical(["A", "B", "C"], categories=["C", "A", "B"], ordered=True)
+
+
+@pytest.fixture
+def data_missing_for_sorting():
+    return Categorical(["A", None, "B"], categories=["B", "A"], ordered=True)
+
+
+@pytest.fixture
+def data_for_grouping():
+    return Categorical(["a", "a", None, None, "b", "b", "a", "c"])
+
+
+class TestCategorical(base.ExtensionTests):
+    def test_contains(self, data, data_missing):
+        # GH-37867
+        # na value handling in Categorical.__contains__ is deprecated.
+        # See base.BaseInterFaceTests.test_contains for more details.
+
+        na_value = data.dtype.na_value
+        # ensure data without missing values
+        data = data[~data.isna()]
+
+        # first elements are non-missing
+        assert data[0] in data
+        assert data_missing[0] in data_missing
+
+        # check the presence of na_value
+        assert na_value in data_missing
+        assert na_value not in data
+
+        # Categoricals can contain other nan-likes than na_value
+        for na_value_obj in tm.NULL_OBJECTS:
+            if na_value_obj is na_value:
+                continue
+            assert na_value_obj not in data
+            # this section suffers from super method
+            if not using_string_dtype():
+                assert na_value_obj in data_missing
+
+    def test_empty(self, dtype):
+        cls = dtype.construct_array_type()
+        result = cls._empty((4,), dtype=dtype)
+
+        assert isinstance(result, cls)
+        # the dtype we passed is not initialized, so will not match the
+        #  dtype on our result.
+        assert result.dtype == CategoricalDtype([])
+
+    @pytest.mark.skip(reason="Backwards compatibility")
+    def test_getitem_scalar(self, data):
+        # CategoricalDtype.type isn't "correct" since it should
+        # be a parent of the elements (object). But don't want
+        # to break things by changing.
+        super().test_getitem_scalar(data)
+
+    def test_combine_add(self, data_repeated):
+        # GH 20825
+        # When adding categoricals in combine, result is a string
+        orig_data1, orig_data2 = data_repeated(2)
+        s1 = pd.Series(orig_data1)
+        s2 = pd.Series(orig_data2)
+        result = s1.combine(s2, lambda x1, x2: x1 + x2)
+        expected = pd.Series(
+            [a + b for (a, b) in zip(list(orig_data1), list(orig_data2))]
+        )
+        tm.assert_series_equal(result, expected)
+
+        val = s1.iloc[0]
+        result = s1.combine(val, lambda x1, x2: x1 + x2)
+        expected = pd.Series([a + val for a in list(orig_data1)])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data, na_action):
+        result = data.map(lambda x: x, na_action=na_action)
+        tm.assert_extension_array_equal(result, data)
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators, request):
+        # frame & scalar
+        op_name = all_arithmetic_operators
+        if op_name == "__rmod__":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="rmod never called when string is first argument"
+                )
+            )
+        super().test_arith_frame_with_scalar(data, op_name)
+
+    def test_arith_series_with_scalar(self, data, all_arithmetic_operators, request):
+        op_name = all_arithmetic_operators
+        if op_name == "__rmod__":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="rmod never called when string is first argument"
+                )
+            )
+        super().test_arith_series_with_scalar(data, op_name)
+
+    def _compare_other(self, ser: pd.Series, data, op, other):
+        op_name = f"__{op.__name__}__"
+        if op_name not in ["__eq__", "__ne__"]:
+            msg = "Unordered Categoricals can only compare equality or not"
+            with pytest.raises(TypeError, match=msg):
+                op(data, other)
+        else:
+            return super()._compare_other(ser, data, op, other)
+
+    @pytest.mark.xfail(reason="Categorical overrides __repr__")
+    @pytest.mark.parametrize("size", ["big", "small"])
+    def test_array_repr(self, data, size):
+        super().test_array_repr(data, size)
+
+    @pytest.mark.xfail(reason="TBD")
+    @pytest.mark.parametrize("as_index", [True, False])
+    def test_groupby_extension_agg(self, as_index, data_for_grouping):
+        super().test_groupby_extension_agg(as_index, data_for_grouping)
+
+
+class Test2DCompat(base.NDArrayBacked2DTests):
+    def test_repr_2d(self, data):
+        # Categorical __repr__ doesn't include "Categorical", so we need
+        #  to special-case
+        res = repr(data.reshape(1, -1))
+        assert res.count("\nCategories") == 1
+
+        res = repr(data.reshape(-1, 1))
+        assert res.count("\nCategories") == 1
diff --git a/pandas/pandas/tests/extension/test_common.py b/pandas/pandas/tests/extension/test_common.py
new file mode 100644
index 0000000000000..5eda0f00f54ca
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_common.py
@@ -0,0 +1,105 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes import dtypes
+from pandas.core.dtypes.common import is_extension_array_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import ExtensionArray
+
+
+class DummyDtype(dtypes.ExtensionDtype):
+    pass
+
+
+class DummyArray(ExtensionArray):
+    def __init__(self, data) -> None:
+        self.data = data
+
+    def __array__(self, dtype=None, copy=None):
+        return self.data
+
+    @property
+    def dtype(self):
+        return DummyDtype()
+
+    def astype(self, dtype, copy=True):
+        # we don't support anything but a single dtype
+        if isinstance(dtype, DummyDtype):
+            if copy:
+                return type(self)(self.data)
+            return self
+        elif not copy:
+            return np.asarray(self, dtype=dtype)
+        else:
+            return np.array(self, dtype=dtype, copy=copy)
+
+
+class TestExtensionArrayDtype:
+    @pytest.mark.parametrize(
+        "values",
+        [
+            pd.Categorical([]),
+            pd.Categorical([]).dtype,
+            pd.Series(pd.Categorical([])),
+            DummyDtype(),
+            DummyArray(np.array([1, 2])),
+        ],
+    )
+    def test_is_extension_array_dtype(self, values):
+        assert is_extension_array_dtype(values)
+
+    @pytest.mark.parametrize("values", [np.array([]), pd.Series(np.array([]))])
+    def test_is_not_extension_array_dtype(self, values):
+        assert not is_extension_array_dtype(values)
+
+
+def test_astype():
+    arr = DummyArray(np.array([1, 2, 3]))
+    expected = np.array([1, 2, 3], dtype=object)
+
+    result = arr.astype(object)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = arr.astype("object")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_astype_no_copy():
+    arr = DummyArray(np.array([1, 2, 3], dtype=np.int64))
+    result = arr.astype(arr.dtype, copy=False)
+
+    assert arr is result
+
+    result = arr.astype(arr.dtype)
+    assert arr is not result
+
+
+@pytest.mark.parametrize("dtype", [dtypes.CategoricalDtype(), dtypes.IntervalDtype()])
+def test_is_extension_array_dtype(dtype):
+    assert isinstance(dtype, dtypes.ExtensionDtype)
+    assert is_extension_array_dtype(dtype)
+
+
+class CapturingStringArray(pd.arrays.StringArray):
+    """Extend StringArray to capture arguments to __getitem__"""
+
+    def __getitem__(self, item):
+        self.last_item_arg = item
+        return super().__getitem__(item)
+
+
+def test_ellipsis_index():
+    # GH#42430 1D slices over extension types turn into N-dimensional slices
+    #  over ExtensionArrays
+    df = pd.DataFrame(
+        {"col1": CapturingStringArray(np.array(["hello", "world"], dtype=object))}
+    )
+    _ = df.iloc[:1]
+
+    # String comparison because there's no native way to compare slices.
+    # Before the fix for GH#42430, last_item_arg would get set to the 2D slice
+    # (Ellipsis, slice(None, 1, None))
+    out = df["col1"].array.last_item_arg
+    assert str(out) == "slice(None, 1, None)"
diff --git a/pandas/pandas/tests/extension/test_datetime.py b/pandas/pandas/tests/extension/test_datetime.py
new file mode 100644
index 0000000000000..356d5352f41f4
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_datetime.py
@@ -0,0 +1,148 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+from pandas.tests.extension import base
+
+
+@pytest.fixture
+def dtype():
+    return DatetimeTZDtype(unit="ns", tz="US/Central")
+
+
+@pytest.fixture
+def data(dtype):
+    data = DatetimeArray._from_sequence(
+        pd.date_range("2000", periods=100, tz=dtype.tz), dtype=dtype
+    )
+    return data
+
+
+@pytest.fixture
+def data_missing(dtype):
+    return DatetimeArray._from_sequence(
+        np.array(["NaT", "2000-01-01"], dtype="datetime64[ns]"), dtype=dtype
+    )
+
+
+@pytest.fixture
+def data_for_sorting(dtype):
+    a = pd.Timestamp("2000-01-01")
+    b = pd.Timestamp("2000-01-02")
+    c = pd.Timestamp("2000-01-03")
+    return DatetimeArray._from_sequence(
+        np.array([b, c, a], dtype="datetime64[ns]"), dtype=dtype
+    )
+
+
+@pytest.fixture
+def data_missing_for_sorting(dtype):
+    a = pd.Timestamp("2000-01-01")
+    b = pd.Timestamp("2000-01-02")
+    return DatetimeArray._from_sequence(
+        np.array([b, "NaT", a], dtype="datetime64[ns]"), dtype=dtype
+    )
+
+
+@pytest.fixture
+def data_for_grouping(dtype):
+    """
+    Expected to be like [B, B, NA, NA, A, A, B, C]
+
+    Where A < B < C and NA is missing
+    """
+    a = pd.Timestamp("2000-01-01")
+    b = pd.Timestamp("2000-01-02")
+    c = pd.Timestamp("2000-01-03")
+    na = "NaT"
+    return DatetimeArray._from_sequence(
+        np.array([b, b, na, na, a, a, b, c], dtype="datetime64[ns]"), dtype=dtype
+    )
+
+
+@pytest.fixture
+def na_cmp():
+    def cmp(a, b):
+        return a is pd.NaT and a is b
+
+    return cmp
+
+
+# ----------------------------------------------------------------------------
+class TestDatetimeArray(base.ExtensionTests):
+    def _get_expected_exception(self, op_name, obj, other):
+        if op_name in ["__sub__", "__rsub__"]:
+            return None
+        return super()._get_expected_exception(op_name, obj, other)
+
+    def _get_expected_reduction_dtype(self, arr, op_name: str, skipna: bool):
+        if op_name == "std":
+            return "timedelta64[ns]"
+        return arr.dtype
+
+    def _supports_accumulation(self, ser, op_name: str) -> bool:
+        return op_name in ["cummin", "cummax"]
+
+    def _supports_reduction(self, obj, op_name: str) -> bool:
+        return op_name in ["min", "max", "median", "mean", "std", "any", "all"]
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_series_boolean(self, data, all_boolean_reductions, skipna):
+        meth = all_boolean_reductions
+        msg = f"datetime64 type does not support operation '{meth}'"
+        with pytest.raises(TypeError, match=msg):
+            super().test_reduce_series_boolean(data, all_boolean_reductions, skipna)
+
+    def test_series_constructor(self, data):
+        # Series construction drops any .freq attr
+        data = data._with_freq(None)
+        super().test_series_constructor(data)
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data, na_action):
+        result = data.map(lambda x: x, na_action=na_action)
+        tm.assert_extension_array_equal(result, data)
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        if op_name in ["median", "mean", "std"]:
+            alt = ser.astype("int64")
+
+            res_op = getattr(ser, op_name)
+            exp_op = getattr(alt, op_name)
+            result = res_op(skipna=skipna)
+            expected = exp_op(skipna=skipna)
+            if op_name in ["mean", "median"]:
+                # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype"
+                # has no attribute "tz"
+                tz = ser.dtype.tz  # type: ignore[union-attr]
+                expected = pd.Timestamp(expected, tz=tz)
+            else:
+                expected = pd.Timedelta(expected)
+            tm.assert_almost_equal(result, expected)
+
+        else:
+            return super().check_reduce(ser, op_name, skipna)
+
+
+class Test2DCompat(base.NDArrayBacked2DTests):
+    pass
diff --git a/pandas/pandas/tests/extension/test_extension.py b/pandas/pandas/tests/extension/test_extension.py
new file mode 100644
index 0000000000000..456f4863b1c31
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_extension.py
@@ -0,0 +1,27 @@
+"""
+Tests for behavior if an author does *not* implement EA methods.
+"""
+
+import numpy as np
+import pytest
+
+from pandas.core.arrays import ExtensionArray
+
+
+class MyEA(ExtensionArray):
+    def __init__(self, values) -> None:
+        self._values = values
+
+
+@pytest.fixture
+def data():
+    arr = np.arange(10)
+    return MyEA(arr)
+
+
+class TestExtensionArray:
+    def test_errors(self, data, all_arithmetic_operators):
+        # invalid ops
+        op_name = all_arithmetic_operators
+        with pytest.raises(AttributeError):
+            getattr(data, op_name)
diff --git a/pandas/pandas/tests/extension/test_interval.py b/pandas/pandas/tests/extension/test_interval.py
new file mode 100644
index 0000000000000..011bf0b2016b2
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_interval.py
@@ -0,0 +1,134 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import IntervalDtype
+
+from pandas import Interval
+from pandas.core.arrays import IntervalArray
+from pandas.tests.extension import base
+
+if TYPE_CHECKING:
+    import pandas as pd
+
+
+def make_data():
+    N = 100
+    left_array = np.random.default_rng(2).uniform(size=N).cumsum()
+    right_array = left_array + np.random.default_rng(2).uniform(size=N)
+    return [Interval(left, right) for left, right in zip(left_array, right_array)]
+
+
+@pytest.fixture
+def dtype():
+    return IntervalDtype()
+
+
+@pytest.fixture
+def data():
+    """Length-100 PeriodArray for semantics test."""
+    return IntervalArray(make_data())
+
+
+@pytest.fixture
+def data_missing():
+    """Length 2 array with [NA, Valid]"""
+    return IntervalArray.from_tuples([None, (0, 1)])
+
+
+@pytest.fixture
+def data_for_twos():
+    pytest.skip("Interval is not a numeric dtype")
+
+
+@pytest.fixture
+def data_for_sorting():
+    return IntervalArray.from_tuples([(1, 2), (2, 3), (0, 1)])
+
+
+@pytest.fixture
+def data_missing_for_sorting():
+    return IntervalArray.from_tuples([(1, 2), None, (0, 1)])
+
+
+@pytest.fixture
+def data_for_grouping():
+    a = (0, 1)
+    b = (1, 2)
+    c = (2, 3)
+    return IntervalArray.from_tuples([b, b, None, None, a, a, b, c])
+
+
+class TestIntervalArray(base.ExtensionTests):
+    divmod_exc = TypeError
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        return op_name in ["min", "max"]
+
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        with pytest.raises(ValueError, match="limit must be None"):
+            super().test_fillna_limit_frame(data_missing)
+
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        with pytest.raises(ValueError, match="limit must be None"):
+            super().test_fillna_limit_frame(data_missing)
+
+    @pytest.mark.xfail(
+        reason="Raises with incorrect message bc it disallows *all* listlikes "
+        "instead of just wrong-length listlikes"
+    )
+    def test_fillna_length_mismatch(self, data_missing):
+        super().test_fillna_length_mismatch(data_missing)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_hash_pandas_object(self, data):
+        super().test_hash_pandas_object(data)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_hash_pandas_object_works(self, data, as_frame):
+        super().test_hash_pandas_object_works(data, as_frame)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_EA_types(self, engine, data, request):
+        super().test_EA_types(engine, data, request)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_astype_str(self, data):
+        super().test_astype_str(data)
+
+
+# TODO: either belongs in tests.arrays.interval or move into base tests.
+def test_fillna_non_scalar_raises(data_missing):
+    msg = "can only insert Interval objects and NA into an IntervalArray"
+    with pytest.raises(TypeError, match=msg):
+        data_missing.fillna([1, 1])
diff --git a/pandas/pandas/tests/extension/test_masked.py b/pandas/pandas/tests/extension/test_masked.py
new file mode 100644
index 0000000000000..3b9079d06e231
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_masked.py
@@ -0,0 +1,408 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_signed_integer_dtype,
+    is_unsigned_integer_dtype,
+)
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays.boolean import BooleanDtype
+from pandas.core.arrays.floating import (
+    Float32Dtype,
+    Float64Dtype,
+)
+from pandas.core.arrays.integer import (
+    Int8Dtype,
+    Int16Dtype,
+    Int32Dtype,
+    Int64Dtype,
+    UInt8Dtype,
+    UInt16Dtype,
+    UInt32Dtype,
+    UInt64Dtype,
+)
+from pandas.tests.extension import base
+
+is_windows_or_32bit = (is_platform_windows() and not np_version_gt2) or not IS64
+
+pytestmark = [
+    pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in divide:RuntimeWarning"
+    ),
+    pytest.mark.filterwarnings("ignore:Mean of empty slice:RuntimeWarning"),
+    # overflow only relevant for Floating dtype cases cases
+    pytest.mark.filterwarnings("ignore:overflow encountered in reduce:RuntimeWarning"),
+]
+
+
+def make_data():
+    return list(range(1, 9)) + [pd.NA] + list(range(10, 98)) + [pd.NA] + [99, 100]
+
+
+def make_float_data():
+    return (
+        list(np.arange(0.1, 0.9, 0.1))
+        + [pd.NA]
+        + list(np.arange(1, 9.8, 0.1))
+        + [pd.NA]
+        + [9.9, 10.0]
+    )
+
+
+def make_bool_data():
+    return [True, False] * 4 + [np.nan] + [True, False] * 44 + [np.nan] + [True, False]
+
+
+@pytest.fixture(
+    params=[
+        Int8Dtype,
+        Int16Dtype,
+        Int32Dtype,
+        Int64Dtype,
+        UInt8Dtype,
+        UInt16Dtype,
+        UInt32Dtype,
+        UInt64Dtype,
+        Float32Dtype,
+        Float64Dtype,
+        BooleanDtype,
+    ]
+)
+def dtype(request):
+    return request.param()
+
+
+@pytest.fixture
+def data(dtype):
+    if dtype.kind == "f":
+        data = make_float_data()
+    elif dtype.kind == "b":
+        data = make_bool_data()
+    else:
+        data = make_data()
+    return pd.array(data, dtype=dtype)
+
+
+@pytest.fixture
+def data_for_twos(dtype):
+    if dtype.kind == "b":
+        return pd.array(np.ones(100), dtype=dtype)
+    return pd.array(np.ones(100) * 2, dtype=dtype)
+
+
+@pytest.fixture
+def data_missing(dtype):
+    if dtype.kind == "f":
+        return pd.array([pd.NA, 0.1], dtype=dtype)
+    elif dtype.kind == "b":
+        return pd.array([np.nan, True], dtype=dtype)
+    return pd.array([pd.NA, 1], dtype=dtype)
+
+
+@pytest.fixture
+def data_for_sorting(dtype):
+    if dtype.kind == "f":
+        return pd.array([0.1, 0.2, 0.0], dtype=dtype)
+    elif dtype.kind == "b":
+        return pd.array([True, True, False], dtype=dtype)
+    return pd.array([1, 2, 0], dtype=dtype)
+
+
+@pytest.fixture
+def data_missing_for_sorting(dtype):
+    if dtype.kind == "f":
+        return pd.array([0.1, pd.NA, 0.0], dtype=dtype)
+    elif dtype.kind == "b":
+        return pd.array([True, np.nan, False], dtype=dtype)
+    return pd.array([1, pd.NA, 0], dtype=dtype)
+
+
+@pytest.fixture
+def na_cmp():
+    # we are pd.NA
+    return lambda x, y: x is pd.NA and y is pd.NA
+
+
+@pytest.fixture
+def data_for_grouping(dtype):
+    if dtype.kind == "f":
+        b = 0.1
+        a = 0.0
+        c = 0.2
+    elif dtype.kind == "b":
+        b = True
+        a = False
+        c = b
+    else:
+        b = 1
+        a = 0
+        c = 2
+
+    na = pd.NA
+    return pd.array([b, b, na, na, a, a, b, c], dtype=dtype)
+
+
+class TestMaskedArrays(base.ExtensionTests):
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data_missing, na_action):
+        result = data_missing.map(lambda x: x, na_action=na_action)
+        if data_missing.dtype == Float32Dtype():
+            # map roundtrips through objects, which converts to float64
+            expected = data_missing.to_numpy(dtype="float64", na_value=np.nan)
+        else:
+            expected = data_missing.to_numpy()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_map_na_action_ignore(self, data_missing_for_sorting):
+        zero = data_missing_for_sorting[2]
+        result = data_missing_for_sorting.map(lambda x: zero, na_action="ignore")
+        if data_missing_for_sorting.dtype.kind == "b":
+            expected = np.array([False, pd.NA, False], dtype=object)
+        else:
+            expected = np.array([zero, np.nan, zero])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def _get_expected_exception(self, op_name, obj, other):
+        try:
+            dtype = tm.get_dtype(obj)
+        except AttributeError:
+            # passed arguments reversed
+            dtype = tm.get_dtype(other)
+
+        if dtype.kind == "b":
+            if op_name.strip("_").lstrip("r") in ["pow", "truediv", "floordiv"]:
+                # match behavior with non-masked bool dtype
+                return NotImplementedError
+            elif op_name in ["__sub__", "__rsub__"]:
+                # exception message would include "numpy boolean subtract""
+                return TypeError
+            return None
+        return None
+
+    def _cast_pointwise_result(self, op_name: str, obj, other, pointwise_result):
+        sdtype = tm.get_dtype(obj)
+        expected = pointwise_result
+
+        if op_name in ("eq", "ne", "le", "ge", "lt", "gt"):
+            return expected.astype("boolean")
+
+        if sdtype.kind in "iu":
+            if op_name in ("__rtruediv__", "__truediv__", "__div__"):
+                filled = expected.fillna(np.nan)
+                expected = filled.astype("Float64")
+            else:
+                # combine method result in 'biggest' (int64) dtype
+                expected = expected.astype(sdtype)
+        elif sdtype.kind == "b":
+            if op_name in (
+                "__floordiv__",
+                "__rfloordiv__",
+                "__pow__",
+                "__rpow__",
+                "__mod__",
+                "__rmod__",
+            ):
+                # combine keeps boolean type
+                expected = expected.astype("Int8")
+
+            elif op_name in ("__truediv__", "__rtruediv__"):
+                # combine with bools does not generate the correct result
+                #  (numpy behaviour for div is to regard the bools as numeric)
+                op = self.get_op_from_name(op_name)
+                expected = self._combine(obj.astype(float), other, op)
+                expected = expected.astype("Float64")
+
+            if op_name == "__rpow__":
+                # for rpow, combine does not propagate NaN
+                result = getattr(obj, op_name)(other)
+                expected[result.isna()] = np.nan
+        else:
+            # combine method result in 'biggest' (float64) dtype
+            expected = expected.astype(sdtype)
+        return expected
+
+    def test_divmod_series_array(self, data, data_for_twos, request):
+        if data.dtype.kind == "b":
+            mark = pytest.mark.xfail(
+                reason="Inconsistency between floordiv and divmod; we raise for "
+                "floordiv but not for divmod. This matches what we do for "
+                "non-masked bool dtype."
+            )
+            request.applymarker(mark)
+        super().test_divmod_series_array(data, data_for_twos)
+
+    def test_combine_le(self, data_repeated):
+        # TODO: patching self is a bad pattern here
+        orig_data1, orig_data2 = data_repeated(2)
+        if orig_data1.dtype.kind == "b":
+            self._combine_le_expected_dtype = "boolean"
+        else:
+            # TODO: can we make this boolean?
+            self._combine_le_expected_dtype = object
+        super().test_combine_le(data_repeated)
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        if op_name in ["any", "all"] and ser.dtype.kind != "b":
+            pytest.skip(reason="Tested in tests/reductions/test_reductions.py")
+        return True
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        # overwrite to ensure pd.NA is tested instead of np.nan
+        # https://github.com/pandas-dev/pandas/issues/30958
+
+        cmp_dtype = "int64"
+        if ser.dtype.kind == "f":
+            # Item "dtype[Any]" of "Union[dtype[Any], ExtensionDtype]" has
+            # no attribute "numpy_dtype"
+            cmp_dtype = ser.dtype.numpy_dtype  # type: ignore[union-attr]
+        elif ser.dtype.kind == "b":
+            if op_name in ["min", "max"]:
+                cmp_dtype = "bool"
+
+        # TODO: prod with integer dtypes does *not* match the result we would
+        #  get if we used object for cmp_dtype. In that cae the object result
+        #  is a large integer while the non-object case overflows and returns 0
+        alt = ser.dropna().astype(cmp_dtype)
+        if op_name == "count":
+            result = getattr(ser, op_name)()
+            expected = getattr(alt, op_name)()
+        else:
+            result = getattr(ser, op_name)(skipna=skipna)
+            expected = getattr(alt, op_name)(skipna=skipna)
+            if not skipna and ser.isna().any() and op_name not in ["any", "all"]:
+                expected = pd.NA
+        tm.assert_almost_equal(result, expected)
+
+    def _get_expected_reduction_dtype(self, arr, op_name: str, skipna: bool):
+        if is_float_dtype(arr.dtype):
+            cmp_dtype = arr.dtype.name
+        elif op_name in ["mean", "median", "var", "std", "skew", "kurt", "sem"]:
+            cmp_dtype = "Float64"
+        elif op_name in ["max", "min"]:
+            cmp_dtype = arr.dtype.name
+        elif arr.dtype in ["Int64", "UInt64"]:
+            cmp_dtype = arr.dtype.name
+        elif is_signed_integer_dtype(arr.dtype):
+            # TODO: Why does Window Numpy 2.0 dtype depend on skipna?
+            cmp_dtype = (
+                "Int32"
+                if (is_platform_windows() and (not np_version_gt2 or not skipna))
+                or not IS64
+                else "Int64"
+            )
+        elif is_unsigned_integer_dtype(arr.dtype):
+            cmp_dtype = (
+                "UInt32"
+                if (is_platform_windows() and (not np_version_gt2 or not skipna))
+                or not IS64
+                else "UInt64"
+            )
+        elif arr.dtype.kind == "b":
+            if op_name in ["min", "max"]:
+                cmp_dtype = "boolean"
+            elif op_name in ["sum", "prod"]:
+                cmp_dtype = (
+                    "Int32"
+                    if (is_platform_windows() and (not np_version_gt2 or not skipna))
+                    or not IS64
+                    else "Int64"
+                )
+            else:
+                raise TypeError("not supposed to reach this")
+        else:
+            raise TypeError("not supposed to reach this")
+        return cmp_dtype
+
+    def _supports_accumulation(self, ser: pd.Series, op_name: str) -> bool:
+        return True
+
+    def check_accumulate(self, ser: pd.Series, op_name: str, skipna: bool):
+        # overwrite to ensure pd.NA is tested instead of np.nan
+        # https://github.com/pandas-dev/pandas/issues/30958
+        length = 64
+        if is_windows_or_32bit:
+            # Item "ExtensionDtype" of "Union[dtype[Any], ExtensionDtype]" has
+            # no attribute "itemsize"
+            if not ser.dtype.itemsize == 8:  # type: ignore[union-attr]
+                length = 32
+
+        if ser.dtype.name.startswith("U"):
+            expected_dtype = f"UInt{length}"
+        elif ser.dtype.name.startswith("I"):
+            expected_dtype = f"Int{length}"
+        elif ser.dtype.name.startswith("F"):
+            # Incompatible types in assignment (expression has type
+            # "Union[dtype[Any], ExtensionDtype]", variable has type "str")
+            expected_dtype = ser.dtype  # type: ignore[assignment]
+        elif ser.dtype.kind == "b":
+            if op_name in ("cummin", "cummax"):
+                expected_dtype = "boolean"
+            else:
+                expected_dtype = f"Int{length}"
+
+        if expected_dtype == "Float32" and op_name == "cumprod" and skipna:
+            # TODO: xfail?
+            pytest.skip(
+                f"Float32 precision lead to large differences with op {op_name} "
+                f"and skipna={skipna}"
+            )
+
+        if op_name == "cumsum":
+            result = getattr(ser, op_name)(skipna=skipna)
+            expected = pd.Series(
+                pd.array(
+                    getattr(ser.astype("float64"), op_name)(skipna=skipna),
+                    dtype=expected_dtype,
+                )
+            )
+            tm.assert_series_equal(result, expected)
+        elif op_name in ["cummax", "cummin"]:
+            result = getattr(ser, op_name)(skipna=skipna)
+            expected = pd.Series(
+                pd.array(
+                    getattr(ser.astype("float64"), op_name)(skipna=skipna),
+                    dtype=ser.dtype,
+                )
+            )
+            tm.assert_series_equal(result, expected)
+        elif op_name == "cumprod":
+            result = getattr(ser[:12], op_name)(skipna=skipna)
+            expected = pd.Series(
+                pd.array(
+                    getattr(ser[:12].astype("float64"), op_name)(skipna=skipna),
+                    dtype=expected_dtype,
+                )
+            )
+            tm.assert_series_equal(result, expected)
+
+        else:
+            raise NotImplementedError(f"{op_name} not supported")
+
+
+class Test2DCompat(base.Dim2CompatTests):
+    pass
diff --git a/pandas/pandas/tests/extension/test_numpy.py b/pandas/pandas/tests/extension/test_numpy.py
new file mode 100644
index 0000000000000..79cfb736941d6
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_numpy.py
@@ -0,0 +1,426 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+Note: we do not bother with base.BaseIndexTests because NumpyExtensionArray
+will never be held in an Index.
+"""
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import NumpyEADtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import is_object_dtype
+from pandas.core.arrays.numpy_ import NumpyExtensionArray
+from pandas.tests.extension import base
+
+orig_assert_attr_equal = tm.assert_attr_equal
+
+
+def _assert_attr_equal(attr: str, left, right, obj: str = "Attributes"):
+    """
+    patch tm.assert_attr_equal so NumpyEADtype("object") is closed enough to
+    np.dtype("object")
+    """
+    if attr == "dtype":
+        lattr = getattr(left, "dtype", None)
+        rattr = getattr(right, "dtype", None)
+        if isinstance(lattr, NumpyEADtype) and not isinstance(rattr, NumpyEADtype):
+            left = left.astype(lattr.numpy_dtype)
+        elif isinstance(rattr, NumpyEADtype) and not isinstance(lattr, NumpyEADtype):
+            right = right.astype(rattr.numpy_dtype)
+
+    orig_assert_attr_equal(attr, left, right, obj)
+
+
+@pytest.fixture(params=["float", "object"])
+def dtype(request):
+    return NumpyEADtype(np.dtype(request.param))
+
+
+@pytest.fixture
+def allow_in_pandas(monkeypatch):
+    """
+    A monkeypatch to tells pandas to let us in.
+
+    By default, passing a NumpyExtensionArray to an index / series / frame
+    constructor will unbox that NumpyExtensionArray to an ndarray, and treat
+    it as a non-EA column. We don't want people using EAs without
+    reason.
+
+    The mechanism for this is a check against ABCNumpyExtensionArray
+    in each constructor.
+
+    But, for testing, we need to allow them in pandas. So we patch
+    the _typ of NumpyExtensionArray, so that we evade the ABCNumpyExtensionArray
+    check.
+    """
+    with monkeypatch.context() as m:
+        m.setattr(NumpyExtensionArray, "_typ", "extension")
+        m.setattr(tm.asserters, "assert_attr_equal", _assert_attr_equal)
+        yield
+
+
+@pytest.fixture
+def data(allow_in_pandas, dtype):
+    if dtype.numpy_dtype == "object":
+        return pd.Series([(i,) for i in range(100)]).array
+    return NumpyExtensionArray(np.arange(1, 101, dtype=dtype._dtype))
+
+
+@pytest.fixture
+def data_missing(allow_in_pandas, dtype):
+    if dtype.numpy_dtype == "object":
+        return NumpyExtensionArray(np.array([np.nan, (1,)], dtype=object))
+    return NumpyExtensionArray(np.array([np.nan, 1.0]))
+
+
+@pytest.fixture
+def na_cmp():
+    def cmp(a, b):
+        return np.isnan(a) and np.isnan(b)
+
+    return cmp
+
+
+@pytest.fixture
+def data_for_sorting(allow_in_pandas, dtype):
+    """Length-3 array with a known sort order.
+
+    This should be three items [B, C, A] with
+    A < B < C
+    """
+    if dtype.numpy_dtype == "object":
+        # Use an empty tuple for first element, then remove,
+        # to disable np.array's shape inference.
+        return NumpyExtensionArray(np.array([(), (2,), (3,), (1,)], dtype=object)[1:])
+    return NumpyExtensionArray(np.array([1, 2, 0]))
+
+
+@pytest.fixture
+def data_missing_for_sorting(allow_in_pandas, dtype):
+    """Length-3 array with a known sort order.
+
+    This should be three items [B, NA, A] with
+    A < B and NA missing.
+    """
+    if dtype.numpy_dtype == "object":
+        return NumpyExtensionArray(np.array([(1,), np.nan, (0,)], dtype=object))
+    return NumpyExtensionArray(np.array([1, np.nan, 0]))
+
+
+@pytest.fixture
+def data_for_grouping(allow_in_pandas, dtype):
+    """Data for factorization, grouping, and unique tests.
+
+    Expected to be like [B, B, NA, NA, A, A, B, C]
+
+    Where A < B < C and NA is missing
+    """
+    if dtype.numpy_dtype == "object":
+        a, b, c = (1,), (2,), (3,)
+    else:
+        a, b, c = np.arange(3)
+    return NumpyExtensionArray(
+        np.array([b, b, np.nan, np.nan, a, a, b, c], dtype=dtype.numpy_dtype)
+    )
+
+
+@pytest.fixture
+def data_for_twos(dtype):
+    if dtype.kind == "O":
+        pytest.skip(f"{dtype} is not a numeric dtype")
+    arr = np.ones(100) * 2
+    return NumpyExtensionArray._from_sequence(arr, dtype=dtype)
+
+
+@pytest.fixture
+def skip_numpy_object(dtype, request):
+    """
+    Tests for NumpyExtensionArray with nested data. Users typically won't create
+    these objects via `pd.array`, but they can show up through `.array`
+    on a Series with nested data. Many of the base tests fail, as they aren't
+    appropriate for nested data.
+
+    This fixture allows these tests to be skipped when used as a usefixtures
+    marker to either an individual test or a test class.
+    """
+    if dtype == "object":
+        mark = pytest.mark.xfail(reason="Fails for object dtype")
+        request.applymarker(mark)
+
+
+skip_nested = pytest.mark.usefixtures("skip_numpy_object")
+
+
+class TestNumpyExtensionArray(base.ExtensionTests):
+    @pytest.mark.skip(reason="We don't register our dtype")
+    # We don't want to register. This test should probably be split in two.
+    def test_from_dtype(self, data):
+        pass
+
+    @skip_nested
+    def test_series_constructor_scalar_with_index(self, data, dtype):
+        # ValueError: Length of passed values is 1, index implies 3.
+        super().test_series_constructor_scalar_with_index(data, dtype)
+
+    def test_check_dtype(self, data, request, using_infer_string):
+        if data.dtype.numpy_dtype == "object":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=f"NumpyExtensionArray expectedly clashes with a "
+                    f"NumPy name: {data.dtype.numpy_dtype}"
+                )
+            )
+        super().test_check_dtype(data)
+
+    def test_is_not_object_type(self, dtype, request):
+        if dtype.numpy_dtype == "object":
+            # Different from BaseDtypeTests.test_is_not_object_type
+            # because NumpyEADtype(object) is an object type
+            assert is_object_dtype(dtype)
+        else:
+            super().test_is_not_object_type(dtype)
+
+    @skip_nested
+    def test_getitem_scalar(self, data):
+        # AssertionError
+        super().test_getitem_scalar(data)
+
+    @skip_nested
+    def test_shift_fill_value(self, data):
+        # np.array shape inference. Shift implementation fails.
+        super().test_shift_fill_value(data)
+
+    @skip_nested
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        # The "scalar" for this array isn't a scalar.
+        super().test_fillna_limit_frame(data_missing)
+
+    @skip_nested
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        # The "scalar" for this array isn't a scalar.
+        super().test_fillna_limit_series(data_missing)
+
+    @skip_nested
+    def test_fillna_copy_frame(self, data_missing):
+        # The "scalar" for this array isn't a scalar.
+        super().test_fillna_copy_frame(data_missing)
+
+    @skip_nested
+    def test_fillna_copy_series(self, data_missing):
+        # The "scalar" for this array isn't a scalar.
+        super().test_fillna_copy_series(data_missing)
+
+    @skip_nested
+    def test_searchsorted(self, data_for_sorting, as_series):
+        # TODO: NumpyExtensionArray.searchsorted calls ndarray.searchsorted which
+        #  isn't quite what we want in nested data cases. Instead we need to
+        #  adapt something like libindex._bin_search.
+        super().test_searchsorted(data_for_sorting, as_series)
+
+    @pytest.mark.xfail(reason="NumpyExtensionArray.diff may fail on dtype")
+    def test_diff(self, data, periods):
+        return super().test_diff(data, periods)
+
+    def test_insert(self, data, request):
+        if data.dtype.numpy_dtype == object:
+            mark = pytest.mark.xfail(reason="Dimension mismatch in np.concatenate")
+            request.applymarker(mark)
+
+        super().test_insert(data)
+
+    @skip_nested
+    def test_insert_invalid(self, data, invalid_scalar):
+        # NumpyExtensionArray[object] can hold anything, so skip
+        super().test_insert_invalid(data, invalid_scalar)
+
+    divmod_exc = None
+    series_scalar_exc = None
+    frame_scalar_exc = None
+    series_array_exc = None
+
+    def test_divmod(self, data):
+        divmod_exc = None
+        if data.dtype.kind == "O":
+            divmod_exc = TypeError
+        self.divmod_exc = divmod_exc
+        super().test_divmod(data)
+
+    def test_divmod_series_array(self, data):
+        ser = pd.Series(data)
+        exc = None
+        if data.dtype.kind == "O":
+            exc = TypeError
+            self.divmod_exc = exc
+        self._check_divmod_op(ser, divmod, data)
+
+    def test_arith_series_with_scalar(self, data, all_arithmetic_operators, request):
+        opname = all_arithmetic_operators
+        series_scalar_exc = None
+        if data.dtype.numpy_dtype == object:
+            if opname in ["__mul__", "__rmul__"]:
+                mark = pytest.mark.xfail(
+                    reason="the Series.combine step raises but not the Series method."
+                )
+                request.node.add_marker(mark)
+            series_scalar_exc = TypeError
+        self.series_scalar_exc = series_scalar_exc
+        super().test_arith_series_with_scalar(data, all_arithmetic_operators)
+
+    def test_arith_series_with_array(self, data, all_arithmetic_operators):
+        opname = all_arithmetic_operators
+        series_array_exc = None
+        if data.dtype.numpy_dtype == object and opname not in ["__add__", "__radd__"]:
+            series_array_exc = TypeError
+        self.series_array_exc = series_array_exc
+        super().test_arith_series_with_array(data, all_arithmetic_operators)
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators, request):
+        opname = all_arithmetic_operators
+        frame_scalar_exc = None
+        if data.dtype.numpy_dtype == object:
+            if opname in ["__mul__", "__rmul__"]:
+                mark = pytest.mark.xfail(
+                    reason="the Series.combine step raises but not the Series method."
+                )
+                request.node.add_marker(mark)
+            frame_scalar_exc = TypeError
+        self.frame_scalar_exc = frame_scalar_exc
+        super().test_arith_frame_with_scalar(data, all_arithmetic_operators)
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        if ser.dtype.kind == "O":
+            return op_name in ["sum", "min", "max", "any", "all"]
+        return True
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        res_op = getattr(ser, op_name)
+        # avoid coercing int -> float. Just cast to the actual numpy type.
+        # error: Item "ExtensionDtype" of "dtype[Any] | ExtensionDtype" has
+        # no attribute "numpy_dtype"
+        cmp_dtype = ser.dtype.numpy_dtype  # type: ignore[union-attr]
+        alt = ser.astype(cmp_dtype)
+        exp_op = getattr(alt, op_name)
+        if op_name == "count":
+            result = res_op()
+            expected = exp_op()
+        else:
+            result = res_op(skipna=skipna)
+            expected = exp_op(skipna=skipna)
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.skip("TODO: tests not written yet")
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_frame(self, data, all_numeric_reductions, skipna):
+        pass
+
+    @skip_nested
+    def test_fillna_series(self, data_missing):
+        # Non-scalar "scalar" values.
+        super().test_fillna_series(data_missing)
+
+    @skip_nested
+    def test_fillna_frame(self, data_missing):
+        # Non-scalar "scalar" values.
+        super().test_fillna_frame(data_missing)
+
+    @skip_nested
+    def test_setitem_invalid(self, data, invalid_scalar):
+        # object dtype can hold anything, so doesn't raise
+        super().test_setitem_invalid(data, invalid_scalar)
+
+    @skip_nested
+    def test_setitem_sequence_broadcasts(self, data, box_in_series):
+        # ValueError: cannot set using a list-like indexer with a different
+        # length than the value
+        super().test_setitem_sequence_broadcasts(data, box_in_series)
+
+    @skip_nested
+    @pytest.mark.parametrize("setter", ["loc", None])
+    def test_setitem_mask_broadcast(self, data, setter):
+        # ValueError: cannot set using a list-like indexer with a different
+        # length than the value
+        super().test_setitem_mask_broadcast(data, setter)
+
+    @skip_nested
+    def test_setitem_scalar_key_sequence_raise(self, data):
+        # Failed: DID NOT RAISE <class 'ValueError'>
+        super().test_setitem_scalar_key_sequence_raise(data)
+
+    # TODO: there is some issue with NumpyExtensionArray, therefore,
+    #   skip the setitem test for now, and fix it later (GH 31446)
+
+    @skip_nested
+    @pytest.mark.parametrize(
+        "mask",
+        [
+            np.array([True, True, True, False, False]),
+            pd.array([True, True, True, False, False], dtype="boolean"),
+        ],
+        ids=["numpy-array", "boolean-array"],
+    )
+    def test_setitem_mask(self, data, mask, box_in_series):
+        super().test_setitem_mask(data, mask, box_in_series)
+
+    @skip_nested
+    @pytest.mark.parametrize(
+        "idx",
+        [[0, 1, 2], pd.array([0, 1, 2], dtype="Int64"), np.array([0, 1, 2])],
+        ids=["list", "integer-array", "numpy-array"],
+    )
+    def test_setitem_integer_array(self, data, idx, box_in_series):
+        super().test_setitem_integer_array(data, idx, box_in_series)
+
+    @skip_nested
+    def test_setitem_slice(self, data, box_in_series):
+        super().test_setitem_slice(data, box_in_series)
+
+    @skip_nested
+    def test_setitem_loc_iloc_slice(self, data):
+        super().test_setitem_loc_iloc_slice(data)
+
+    def test_setitem_with_expansion_dataframe_column(self, data, full_indexer):
+        # https://github.com/pandas-dev/pandas/issues/32395
+        df = expected = pd.DataFrame({"data": pd.Series(data)})
+        result = pd.DataFrame(index=df.index)
+
+        # because result has object dtype, the attempt to do setting inplace
+        #  is successful, and object dtype is retained
+        key = full_indexer(df)
+        result.loc[key, "data"] = df["data"]
+
+        # base class method has expected = df; NumpyExtensionArray behaves oddly because
+        #  we patch _typ for these tests.
+        if data.dtype.numpy_dtype != object:
+            if not isinstance(key, slice) or key != slice(None):
+                expected = pd.DataFrame({"data": data.to_numpy()})
+        tm.assert_frame_equal(result, expected, check_column_type=False)
+
+    @pytest.mark.xfail(reason="NumpyEADtype is unpacked")
+    def test_index_from_listlike_with_dtype(self, data):
+        super().test_index_from_listlike_with_dtype(data)
+
+    @skip_nested
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_EA_types(self, engine, data, request):
+        super().test_EA_types(engine, data, request)
+
+
+class Test2DCompat(base.NDArrayBacked2DTests):
+    pass
diff --git a/pandas/pandas/tests/extension/test_period.py b/pandas/pandas/tests/extension/test_period.py
new file mode 100644
index 0000000000000..142bad6db4f95
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_period.py
@@ -0,0 +1,120 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    Period,
+    iNaT,
+)
+from pandas.compat import is_platform_windows
+from pandas.compat.numpy import np_version_gte1p24
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+import pandas._testing as tm
+from pandas.core.arrays import PeriodArray
+from pandas.tests.extension import base
+
+if TYPE_CHECKING:
+    import pandas as pd
+
+
+@pytest.fixture(params=["D", "2D"])
+def dtype(request):
+    return PeriodDtype(freq=request.param)
+
+
+@pytest.fixture
+def data(dtype):
+    return PeriodArray(np.arange(1970, 2070), dtype=dtype)
+
+
+@pytest.fixture
+def data_for_sorting(dtype):
+    return PeriodArray([2018, 2019, 2017], dtype=dtype)
+
+
+@pytest.fixture
+def data_missing(dtype):
+    return PeriodArray([iNaT, 2017], dtype=dtype)
+
+
+@pytest.fixture
+def data_missing_for_sorting(dtype):
+    return PeriodArray([2018, iNaT, 2017], dtype=dtype)
+
+
+@pytest.fixture
+def data_for_grouping(dtype):
+    B = 2018
+    NA = iNaT
+    A = 2017
+    C = 2019
+    return PeriodArray([B, B, NA, NA, A, A, B, C], dtype=dtype)
+
+
+class TestPeriodArray(base.ExtensionTests):
+    def _get_expected_exception(self, op_name, obj, other):
+        if op_name in ("__sub__", "__rsub__"):
+            return None
+        return super()._get_expected_exception(op_name, obj, other)
+
+    def _supports_accumulation(self, ser, op_name: str) -> bool:
+        return op_name in ["cummin", "cummax"]
+
+    def _supports_reduction(self, obj, op_name: str) -> bool:
+        return op_name in ["min", "max", "median"]
+
+    def check_reduce(self, ser: pd.Series, op_name: str, skipna: bool):
+        if op_name == "median":
+            res_op = getattr(ser, op_name)
+
+            alt = ser.astype("int64")
+
+            exp_op = getattr(alt, op_name)
+            result = res_op(skipna=skipna)
+            expected = exp_op(skipna=skipna)
+            # error: Item "dtype[Any]" of "dtype[Any] | ExtensionDtype" has no
+            # attribute "freq"
+            freq = ser.dtype.freq  # type: ignore[union-attr]
+            expected = Period._from_ordinal(int(expected), freq=freq)
+            tm.assert_almost_equal(result, expected)
+
+        else:
+            return super().check_reduce(ser, op_name, skipna)
+
+    @pytest.mark.parametrize("periods", [1, -2])
+    def test_diff(self, data, periods):
+        if is_platform_windows() and np_version_gte1p24:
+            with tm.assert_produces_warning(RuntimeWarning, check_stacklevel=False):
+                super().test_diff(data, periods)
+        else:
+            super().test_diff(data, periods)
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map(self, data, na_action):
+        result = data.map(lambda x: x, na_action=na_action)
+        tm.assert_extension_array_equal(result, data)
+
+
+class Test2DCompat(base.NDArrayBacked2DTests):
+    pass
diff --git a/pandas/pandas/tests/extension/test_sparse.py b/pandas/pandas/tests/extension/test_sparse.py
new file mode 100644
index 0000000000000..b7685a61d4937
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_sparse.py
@@ -0,0 +1,497 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import SparseDtype
+import pandas._testing as tm
+from pandas.arrays import SparseArray
+from pandas.tests.extension import base
+
+
+def make_data(fill_value):
+    rng = np.random.default_rng(2)
+    if np.isnan(fill_value):
+        data = rng.uniform(size=100)
+    else:
+        data = rng.integers(1, 100, size=100, dtype=int)
+        if data[0] == data[1]:
+            data[0] += 1
+
+    data[2::3] = fill_value
+    return data
+
+
+@pytest.fixture
+def dtype():
+    return SparseDtype()
+
+
+@pytest.fixture(params=[0, np.nan])
+def data(request):
+    """Length-100 PeriodArray for semantics test."""
+    res = SparseArray(make_data(request.param), fill_value=request.param)
+    return res
+
+
+@pytest.fixture
+def data_for_twos():
+    return SparseArray(np.ones(100) * 2)
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_missing(request):
+    """Length 2 array with [NA, Valid]"""
+    return SparseArray([np.nan, 1], fill_value=request.param)
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_repeated(request):
+    """Return different versions of data for count times"""
+
+    def gen(count):
+        for _ in range(count):
+            yield SparseArray(make_data(request.param), fill_value=request.param)
+
+    return gen
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_for_sorting(request):
+    return SparseArray([2, 3, 1], fill_value=request.param)
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_missing_for_sorting(request):
+    return SparseArray([2, np.nan, 1], fill_value=request.param)
+
+
+@pytest.fixture
+def na_cmp():
+    return lambda left, right: pd.isna(left) and pd.isna(right)
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_for_grouping(request):
+    return SparseArray([1, 1, np.nan, np.nan, 2, 2, 1, 3], fill_value=request.param)
+
+
+@pytest.fixture(params=[0, np.nan])
+def data_for_compare(request):
+    return SparseArray([0, 0, np.nan, -2, -1, 4, 2, 3, 0, 0], fill_value=request.param)
+
+
+class TestSparseArray(base.ExtensionTests):
+    def _supports_reduction(self, obj, op_name: str) -> bool:
+        return True
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_series_numeric(self, data, all_numeric_reductions, skipna, request):
+        if all_numeric_reductions in [
+            "prod",
+            "median",
+            "var",
+            "std",
+            "sem",
+            "skew",
+            "kurt",
+        ]:
+            mark = pytest.mark.xfail(
+                reason="This should be viable but is not implemented"
+            )
+            request.node.add_marker(mark)
+        elif (
+            all_numeric_reductions in ["sum", "max", "min", "mean"]
+            and data.dtype.kind == "f"
+            and not skipna
+        ):
+            mark = pytest.mark.xfail(reason="getting a non-nan float")
+            request.node.add_marker(mark)
+
+        super().test_reduce_series_numeric(data, all_numeric_reductions, skipna)
+
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_reduce_frame(self, data, all_numeric_reductions, skipna, request):
+        if all_numeric_reductions in [
+            "prod",
+            "median",
+            "var",
+            "std",
+            "sem",
+            "skew",
+            "kurt",
+        ]:
+            mark = pytest.mark.xfail(
+                reason="This should be viable but is not implemented"
+            )
+            request.node.add_marker(mark)
+        elif (
+            all_numeric_reductions in ["sum", "max", "min", "mean"]
+            and data.dtype.kind == "f"
+            and not skipna
+        ):
+            mark = pytest.mark.xfail(reason="ExtensionArray NA mask are different")
+            request.node.add_marker(mark)
+
+        super().test_reduce_frame(data, all_numeric_reductions, skipna)
+
+    def _check_unsupported(self, data):
+        if data.dtype == SparseDtype(int, 0):
+            pytest.skip("Can't store nan in int array.")
+
+    def test_concat_mixed_dtypes(self, data):
+        # https://github.com/pandas-dev/pandas/issues/20762
+        # This should be the same, aside from concat([sparse, float])
+        df1 = pd.DataFrame({"A": data[:3]})
+        df2 = pd.DataFrame({"A": [1, 2, 3]})
+        df3 = pd.DataFrame({"A": ["a", "b", "c"]}).astype("category")
+        dfs = [df1, df2, df3]
+
+        # dataframes
+        result = pd.concat(dfs)
+        expected = pd.concat(
+            [x.apply(lambda s: np.asarray(s).astype(object)) for x in dfs]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "columns",
+        [
+            ["A", "B"],
+            pd.MultiIndex.from_tuples(
+                [("A", "a"), ("A", "b")], names=["outer", "inner"]
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("future_stack", [True, False])
+    def test_stack(self, data, columns, future_stack):
+        super().test_stack(data, columns, future_stack)
+
+    def test_concat_columns(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_concat_columns(data, na_value)
+
+    def test_concat_extension_arrays_copy_false(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_concat_extension_arrays_copy_false(data, na_value)
+
+    def test_align(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_align(data, na_value)
+
+    def test_align_frame(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_align_frame(data, na_value)
+
+    def test_align_series_frame(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_align_series_frame(data, na_value)
+
+    def test_merge(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_merge(data, na_value)
+
+    def test_get(self, data):
+        ser = pd.Series(data, index=[2 * i for i in range(len(data))])
+        if np.isnan(ser.values.fill_value):
+            assert np.isnan(ser.get(4)) and np.isnan(ser.iloc[2])
+        else:
+            assert ser.get(4) == ser.iloc[2]
+        assert ser.get(2) == ser.iloc[1]
+
+    def test_reindex(self, data, na_value):
+        self._check_unsupported(data)
+        super().test_reindex(data, na_value)
+
+    def test_isna(self, data_missing):
+        sarr = SparseArray(data_missing)
+        expected_dtype = SparseDtype(bool, pd.isna(data_missing.dtype.fill_value))
+        expected = SparseArray([True, False], dtype=expected_dtype)
+        result = sarr.isna()
+        tm.assert_sp_array_equal(result, expected)
+
+        # test isna for arr without na
+        sarr = sarr.fillna(0)
+        expected_dtype = SparseDtype(bool, pd.isna(data_missing.dtype.fill_value))
+        expected = SparseArray([False, False], fill_value=False, dtype=expected_dtype)
+        tm.assert_equal(sarr.isna(), expected)
+
+    def test_fillna_no_op_returns_copy(self, data, request):
+        super().test_fillna_no_op_returns_copy(data)
+
+    @pytest.mark.xfail(reason="Unsupported")
+    def test_fillna_series(self, data_missing):
+        # this one looks doable.
+        # TODO: this fails bc we do not pass through data_missing. If we did,
+        #  the 0-fill case would xpass
+        super().test_fillna_series()
+
+    def test_fillna_frame(self, data_missing):
+        # Have to override to specify that fill_value will change.
+        fill_value = data_missing[1]
+
+        result = pd.DataFrame({"A": data_missing, "B": [1, 2]}).fillna(fill_value)
+
+        if pd.isna(data_missing.fill_value):
+            dtype = SparseDtype(data_missing.dtype, fill_value)
+        else:
+            dtype = data_missing.dtype
+
+        expected = pd.DataFrame(
+            {
+                "A": data_missing._from_sequence([fill_value, fill_value], dtype=dtype),
+                "B": [1, 2],
+            }
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_limit_frame(self, data_missing):
+        # GH#58001
+        with pytest.raises(ValueError, match="limit must be None"):
+            super().test_fillna_limit_frame(data_missing)
+
+    def test_fillna_limit_series(self, data_missing):
+        # GH#58001
+        with pytest.raises(ValueError, match="limit must be None"):
+            super().test_fillna_limit_frame(data_missing)
+
+    _combine_le_expected_dtype = "Sparse[bool]"
+
+    def test_fillna_copy_frame(self, data_missing):
+        arr = data_missing.take([1, 1])
+        df = pd.DataFrame({"A": arr}, copy=False)
+
+        filled_val = df.iloc[0, 0]
+        result = df.fillna(filled_val)
+
+        if hasattr(df._mgr, "blocks"):
+            assert df.values.base is result.values.base
+        assert df.A._values.to_dense() is arr.to_dense()
+
+    def test_fillna_copy_series(self, data_missing):
+        arr = data_missing.take([1, 1])
+        ser = pd.Series(arr, copy=False)
+
+        filled_val = ser[0]
+        result = ser.fillna(filled_val)
+
+        assert ser._values is result._values
+        assert ser._values.to_dense() is arr.to_dense()
+
+    @pytest.mark.xfail(reason="Not Applicable")
+    def test_fillna_length_mismatch(self, data_missing):
+        super().test_fillna_length_mismatch(data_missing)
+
+    def test_where_series(self, data, na_value):
+        assert data[0] != data[1]
+        cls = type(data)
+        a, b = data[:2]
+
+        ser = pd.Series(cls._from_sequence([a, a, b, b], dtype=data.dtype))
+
+        cond = np.array([True, True, False, False])
+        result = ser.where(cond)
+
+        new_dtype = SparseDtype("float", 0.0)
+        expected = pd.Series(
+            cls._from_sequence([a, a, na_value, na_value], dtype=new_dtype)
+        )
+        tm.assert_series_equal(result, expected)
+
+        other = cls._from_sequence([a, b, a, b], dtype=data.dtype)
+        cond = np.array([True, False, True, True])
+        result = ser.where(cond, other)
+        expected = pd.Series(cls._from_sequence([a, b, b, b], dtype=data.dtype))
+        tm.assert_series_equal(result, expected)
+
+    def test_searchsorted(self, performance_warning, data_for_sorting, as_series):
+        with tm.assert_produces_warning(performance_warning, check_stacklevel=False):
+            super().test_searchsorted(data_for_sorting, as_series)
+
+    def test_shift_0_periods(self, data):
+        # GH#33856 shifting with periods=0 should return a copy, not same obj
+        result = data.shift(0)
+
+        data._sparse_values[0] = data._sparse_values[1]
+        assert result._sparse_values[0] != result._sparse_values[1]
+
+    @pytest.mark.parametrize("method", ["argmax", "argmin"])
+    def test_argmin_argmax_all_na(self, method, data, na_value):
+        # overriding because Sparse[int64, 0] cannot handle na_value
+        self._check_unsupported(data)
+        super().test_argmin_argmax_all_na(method, data, na_value)
+
+    @pytest.mark.fails_arm_wheels
+    @pytest.mark.parametrize("box", [pd.array, pd.Series, pd.DataFrame])
+    def test_equals(self, data, na_value, as_series, box):
+        self._check_unsupported(data)
+        super().test_equals(data, na_value, as_series, box)
+
+    @pytest.mark.fails_arm_wheels
+    def test_equals_same_data_different_object(self, data):
+        super().test_equals_same_data_different_object(data)
+
+    @pytest.mark.parametrize(
+        "func, na_action, expected",
+        [
+            (lambda x: x, None, SparseArray([1.0, np.nan])),
+            (lambda x: x, "ignore", SparseArray([1.0, np.nan])),
+            (str, None, SparseArray(["1.0", "nan"], fill_value="nan")),
+            (str, "ignore", SparseArray(["1.0", np.nan])),
+        ],
+    )
+    def test_map(self, func, na_action, expected):
+        # GH52096
+        data = SparseArray([1, np.nan])
+        result = data.map(func, na_action=na_action)
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize("na_action", [None, "ignore"])
+    def test_map_raises(self, data, na_action):
+        # GH52096
+        msg = "fill value in the sparse values not supported"
+        with pytest.raises(ValueError, match=msg):
+            data.map(lambda x: np.nan, na_action=na_action)
+
+    @pytest.mark.xfail(raises=TypeError, reason="no sparse StringDtype")
+    def test_astype_string(self, data, nullable_string_dtype):
+        # TODO: this fails bc we do not pass through nullable_string_dtype;
+        #  If we did, the 0-cases would xpass
+        super().test_astype_string(data)
+
+    series_scalar_exc = None
+    frame_scalar_exc = None
+    divmod_exc = None
+    series_array_exc = None
+
+    def _skip_if_different_combine(self, data):
+        if data.fill_value == 0:
+            # arith ops call on dtype.fill_value so that the sparsity
+            # is maintained. Combine can't be called on a dtype in
+            # general, so we can't make the expected. This is tested elsewhere
+            pytest.skip("Incorrected expected from Series.combine and tested elsewhere")
+
+    def test_arith_series_with_scalar(self, data, all_arithmetic_operators):
+        self._skip_if_different_combine(data)
+        super().test_arith_series_with_scalar(data, all_arithmetic_operators)
+
+    def test_arith_series_with_array(self, data, all_arithmetic_operators):
+        self._skip_if_different_combine(data)
+        super().test_arith_series_with_array(data, all_arithmetic_operators)
+
+    def test_arith_frame_with_scalar(self, data, all_arithmetic_operators, request):
+        if data.dtype.fill_value != 0:
+            pass
+        elif all_arithmetic_operators.strip("_") not in [
+            "mul",
+            "rmul",
+            "floordiv",
+            "rfloordiv",
+            "truediv",
+            "rtruediv",
+            "pow",
+            "mod",
+            "rmod",
+        ]:
+            mark = pytest.mark.xfail(reason="result dtype.fill_value mismatch")
+            request.applymarker(mark)
+        super().test_arith_frame_with_scalar(data, all_arithmetic_operators)
+
+    def _compare_other(
+        self, ser: pd.Series, data_for_compare: SparseArray, comparison_op, other
+    ):
+        op = comparison_op
+
+        result = op(data_for_compare, other)
+        if isinstance(other, pd.Series):
+            assert isinstance(result, pd.Series)
+            assert isinstance(result.dtype, SparseDtype)
+        else:
+            assert isinstance(result, SparseArray)
+        assert result.dtype.subtype == np.bool_
+
+        if isinstance(other, pd.Series):
+            fill_value = op(data_for_compare.fill_value, other._values.fill_value)
+            expected = SparseArray(
+                op(data_for_compare.to_dense(), np.asarray(other)),
+                fill_value=fill_value,
+                dtype=np.bool_,
+            )
+
+        else:
+            fill_value = np.all(
+                op(np.asarray(data_for_compare.fill_value), np.asarray(other))
+            )
+
+            expected = SparseArray(
+                op(data_for_compare.to_dense(), np.asarray(other)),
+                fill_value=fill_value,
+                dtype=np.bool_,
+            )
+        if isinstance(other, pd.Series):
+            # error: Incompatible types in assignment
+            expected = pd.Series(expected)  # type: ignore[assignment]
+        tm.assert_equal(result, expected)
+
+    def test_scalar(self, data_for_compare: SparseArray, comparison_op):
+        ser = pd.Series(data_for_compare)
+        self._compare_other(ser, data_for_compare, comparison_op, 0)
+        self._compare_other(ser, data_for_compare, comparison_op, 1)
+        self._compare_other(ser, data_for_compare, comparison_op, -1)
+        self._compare_other(ser, data_for_compare, comparison_op, np.nan)
+
+    def test_array(self, data_for_compare: SparseArray, comparison_op, request):
+        if data_for_compare.dtype.fill_value == 0 and comparison_op.__name__ in [
+            "eq",
+            "ge",
+            "le",
+        ]:
+            mark = pytest.mark.xfail(reason="Wrong fill_value")
+            request.applymarker(mark)
+
+        arr = np.linspace(-4, 5, 10)
+        ser = pd.Series(data_for_compare)
+        self._compare_other(ser, data_for_compare, comparison_op, arr)
+
+    def test_sparse_array(self, data_for_compare: SparseArray, comparison_op, request):
+        if data_for_compare.dtype.fill_value == 0 and comparison_op.__name__ != "gt":
+            mark = pytest.mark.xfail(reason="Wrong fill_value")
+            request.applymarker(mark)
+
+        ser = pd.Series(data_for_compare)
+        arr = data_for_compare + 1
+        self._compare_other(ser, data_for_compare, comparison_op, arr)
+        arr = data_for_compare * 2
+        self._compare_other(ser, data_for_compare, comparison_op, arr)
+
+    @pytest.mark.xfail(reason="Different repr")
+    def test_array_repr(self, data, size):
+        super().test_array_repr(data, size)
+
+    @pytest.mark.xfail(reason="result does not match expected")
+    @pytest.mark.parametrize("as_index", [True, False])
+    def test_groupby_extension_agg(self, as_index, data_for_grouping):
+        super().test_groupby_extension_agg(as_index, data_for_grouping)
+
+
+def test_array_type_with_arg(dtype):
+    assert dtype.construct_array_type() is SparseArray
diff --git a/pandas/pandas/tests/extension/test_string.py b/pandas/pandas/tests/extension/test_string.py
new file mode 100644
index 0000000000000..25129111180d6
--- /dev/null
+++ b/pandas/pandas/tests/extension/test_string.py
@@ -0,0 +1,272 @@
+"""
+This file contains a minimal set of tests for compliance with the extension
+array interface test suite, and should contain no other tests.
+The test suite for the full functionality of the array is located in
+`pandas/tests/arrays/`.
+
+The tests in this file are inherited from the BaseExtensionTests, and only
+minimal tweaks should be applied to get the tests passing (by overwriting a
+parent method).
+
+Additional tests should either be added to one of the BaseExtensionTests
+classes (if they are relevant for the extension interface for all dtypes), or
+be added to the array-specific tests in `pandas/tests/arrays/`.
+
+"""
+
+from __future__ import annotations
+
+import string
+from typing import cast
+
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+
+from pandas.core.dtypes.base import StorageExtensionDtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import is_string_dtype
+from pandas.core.arrays import ArrowStringArray
+from pandas.core.arrays.string_ import StringDtype
+from pandas.tests.extension import base
+
+
+def maybe_split_array(arr, chunked):
+    if not chunked:
+        return arr
+    elif arr.dtype.storage != "pyarrow":
+        return arr
+
+    pa = pytest.importorskip("pyarrow")
+
+    arrow_array = arr._pa_array
+    split = len(arrow_array) // 2
+    arrow_array = pa.chunked_array(
+        [*arrow_array[:split].chunks, *arrow_array[split:].chunks]
+    )
+    assert arrow_array.num_chunks == 2
+    return type(arr)(arrow_array)
+
+
+@pytest.fixture(params=[True, False])
+def chunked(request):
+    return request.param
+
+
+@pytest.fixture
+def dtype(string_dtype_arguments):
+    storage, na_value = string_dtype_arguments
+    return StringDtype(storage=storage, na_value=na_value)
+
+
+@pytest.fixture
+def data(dtype, chunked):
+    strings = np.random.default_rng(2).choice(list(string.ascii_letters), size=100)
+    while strings[0] == strings[1]:
+        strings = np.random.default_rng(2).choice(list(string.ascii_letters), size=100)
+
+    arr = dtype.construct_array_type()._from_sequence(strings, dtype=dtype)
+    return maybe_split_array(arr, chunked)
+
+
+@pytest.fixture
+def data_missing(dtype, chunked):
+    """Length 2 array with [NA, Valid]"""
+    arr = dtype.construct_array_type()._from_sequence([pd.NA, "A"], dtype=dtype)
+    return maybe_split_array(arr, chunked)
+
+
+@pytest.fixture
+def data_for_sorting(dtype, chunked):
+    arr = dtype.construct_array_type()._from_sequence(["B", "C", "A"], dtype=dtype)
+    return maybe_split_array(arr, chunked)
+
+
+@pytest.fixture
+def data_missing_for_sorting(dtype, chunked):
+    arr = dtype.construct_array_type()._from_sequence(["B", pd.NA, "A"], dtype=dtype)
+    return maybe_split_array(arr, chunked)
+
+
+@pytest.fixture
+def data_for_grouping(dtype, chunked):
+    arr = dtype.construct_array_type()._from_sequence(
+        ["B", "B", pd.NA, pd.NA, "A", "A", "B", "C"], dtype=dtype
+    )
+    return maybe_split_array(arr, chunked)
+
+
+class TestStringArray(base.ExtensionTests):
+    def test_eq_with_str(self, dtype):
+        super().test_eq_with_str(dtype)
+
+        if dtype.na_value is pd.NA:
+            # only the NA-variant supports parametrized string alias
+            assert dtype == f"string[{dtype.storage}]"
+        elif dtype.storage == "pyarrow":
+            with tm.assert_produces_warning(FutureWarning):
+                assert dtype == "string[pyarrow_numpy]"
+
+    def test_is_not_string_type(self, dtype):
+        # Different from BaseDtypeTests.test_is_not_string_type
+        # because StringDtype is a string type
+        assert is_string_dtype(dtype)
+
+    def test_is_dtype_from_name(self, dtype, using_infer_string):
+        if dtype.na_value is np.nan and not using_infer_string:
+            result = type(dtype).is_dtype(dtype.name)
+            assert result is False
+        else:
+            super().test_is_dtype_from_name(dtype)
+
+    def test_construct_from_string_own_name(self, dtype, using_infer_string):
+        if dtype.na_value is np.nan and not using_infer_string:
+            with pytest.raises(TypeError, match="Cannot construct a 'StringDtype'"):
+                dtype.construct_from_string(dtype.name)
+        else:
+            super().test_construct_from_string_own_name(dtype)
+
+    def test_view(self, data):
+        if data.dtype.storage == "pyarrow":
+            pytest.skip(reason="2D support not implemented for ArrowStringArray")
+        super().test_view(data)
+
+    def test_from_dtype(self, data):
+        # base test uses string representation of dtype
+        pass
+
+    def test_transpose(self, data):
+        if data.dtype.storage == "pyarrow":
+            pytest.skip(reason="2D support not implemented for ArrowStringArray")
+        super().test_transpose(data)
+
+    def test_setitem_preserves_views(self, data):
+        if data.dtype.storage == "pyarrow":
+            pytest.skip(reason="2D support not implemented for ArrowStringArray")
+        super().test_setitem_preserves_views(data)
+
+    def test_dropna_array(self, data_missing):
+        result = data_missing.dropna()
+        expected = data_missing[[1]]
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_fillna_no_op_returns_copy(self, data):
+        data = data[~data.isna()]
+
+        valid = data[0]
+        result = data.fillna(valid)
+        assert result is not data
+        tm.assert_extension_array_equal(result, data)
+
+    def _get_expected_exception(
+        self, op_name: str, obj, other
+    ) -> type[Exception] | tuple[type[Exception], ...] | None:
+        if op_name in [
+            "__mod__",
+            "__rmod__",
+            "__divmod__",
+            "__rdivmod__",
+            "__pow__",
+            "__rpow__",
+        ]:
+            return TypeError
+        elif op_name in ["__mul__", "__rmul__"]:
+            # Can only multiply strings by integers
+            return TypeError
+        elif op_name in [
+            "__truediv__",
+            "__rtruediv__",
+            "__floordiv__",
+            "__rfloordiv__",
+            "__sub__",
+            "__rsub__",
+        ]:
+            return TypeError
+
+        return None
+
+    def _supports_reduction(self, ser: pd.Series, op_name: str) -> bool:
+        return op_name in ["min", "max", "sum"] or (
+            ser.dtype.na_value is np.nan  # type: ignore[union-attr]
+            and op_name in ("any", "all")
+        )
+
+    def _supports_accumulation(self, ser: pd.Series, op_name: str) -> bool:
+        assert isinstance(ser.dtype, StorageExtensionDtype)
+        return op_name in ["cummin", "cummax", "cumsum"]
+
+    def _cast_pointwise_result(self, op_name: str, obj, other, pointwise_result):
+        dtype = cast(StringDtype, tm.get_dtype(obj))
+        if op_name in ["__add__", "__radd__"]:
+            cast_to = dtype
+        elif dtype.na_value is np.nan:
+            cast_to = np.bool_  # type: ignore[assignment]
+        elif dtype.storage == "pyarrow":
+            cast_to = "boolean[pyarrow]"  # type: ignore[assignment]
+        else:
+            cast_to = "boolean"  # type: ignore[assignment]
+        return pointwise_result.astype(cast_to)
+
+    def test_compare_scalar(self, data, comparison_op):
+        ser = pd.Series(data)
+        self._compare_other(ser, data, comparison_op, "abc")
+
+    def test_groupby_extension_apply(self, data_for_grouping, groupby_apply_op):
+        super().test_groupby_extension_apply(data_for_grouping, groupby_apply_op)
+
+    def test_combine_add(self, data_repeated, using_infer_string, request):
+        dtype = next(data_repeated(1)).dtype
+        if using_infer_string and (
+            (dtype.na_value is pd.NA) and dtype.storage == "python"
+        ):
+            mark = pytest.mark.xfail(
+                reason="The pointwise operation result will be inferred to "
+                "string[nan, pyarrow], which does not match the input dtype"
+            )
+            request.applymarker(mark)
+        super().test_combine_add(data_repeated)
+
+    def test_arith_series_with_array(
+        self, data, all_arithmetic_operators, using_infer_string, request
+    ):
+        dtype = data.dtype
+        if (
+            using_infer_string
+            and all_arithmetic_operators == "__radd__"
+            and (
+                (dtype.na_value is pd.NA) or (dtype.storage == "python" and HAS_PYARROW)
+            )
+        ):
+            mark = pytest.mark.xfail(
+                reason="The pointwise operation result will be inferred to "
+                "string[nan, pyarrow], which does not match the input dtype"
+            )
+            request.applymarker(mark)
+        super().test_arith_series_with_array(data, all_arithmetic_operators)
+
+
+class Test2DCompat(base.Dim2CompatTests):
+    @pytest.fixture(autouse=True)
+    def arrow_not_supported(self, data):
+        if isinstance(data, ArrowStringArray):
+            pytest.skip(reason="2D support not implemented for ArrowStringArray")
+
+
+def test_searchsorted_with_na_raises(data_for_sorting, as_series):
+    # GH50447
+    b, c, a = data_for_sorting
+    arr = data_for_sorting.take([2, 0, 1])  # to get [a, b, c]
+    arr[-1] = pd.NA
+
+    if as_series:
+        arr = pd.Series(arr)
+
+    msg = (
+        "searchsorted requires array to be sorted, "
+        "which is impossible with NAs present."
+    )
+    with pytest.raises(ValueError, match=msg):
+        arr.searchsorted(b)
diff --git a/pandas/pandas/tests/frame/__init__.py b/pandas/pandas/tests/frame/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/frame/common.py b/pandas/pandas/tests/frame/common.py
new file mode 100644
index 0000000000000..fc41d7907a240
--- /dev/null
+++ b/pandas/pandas/tests/frame/common.py
@@ -0,0 +1,63 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from pandas import (
+    DataFrame,
+    concat,
+)
+
+if TYPE_CHECKING:
+    from pandas._typing import AxisInt
+
+
+def _check_mixed_float(df, dtype=None):
+    # float16 are most likely to be upcasted to float32
+    dtypes = {"A": "float32", "B": "float32", "C": "float16", "D": "float64"}
+    if isinstance(dtype, str):
+        dtypes = {k: dtype for k, v in dtypes.items()}
+    elif isinstance(dtype, dict):
+        dtypes.update(dtype)
+    if dtypes.get("A"):
+        assert df.dtypes["A"] == dtypes["A"]
+    if dtypes.get("B"):
+        assert df.dtypes["B"] == dtypes["B"]
+    if dtypes.get("C"):
+        assert df.dtypes["C"] == dtypes["C"]
+    if dtypes.get("D"):
+        assert df.dtypes["D"] == dtypes["D"]
+
+
+def _check_mixed_int(df, dtype=None):
+    dtypes = {"A": "int32", "B": "uint64", "C": "uint8", "D": "int64"}
+    if isinstance(dtype, str):
+        dtypes = {k: dtype for k, v in dtypes.items()}
+    elif isinstance(dtype, dict):
+        dtypes.update(dtype)
+    if dtypes.get("A"):
+        assert df.dtypes["A"] == dtypes["A"]
+    if dtypes.get("B"):
+        assert df.dtypes["B"] == dtypes["B"]
+    if dtypes.get("C"):
+        assert df.dtypes["C"] == dtypes["C"]
+    if dtypes.get("D"):
+        assert df.dtypes["D"] == dtypes["D"]
+
+
+def zip_frames(frames: list[DataFrame], axis: AxisInt = 1) -> DataFrame:
+    """
+    take a list of frames, zip them together under the
+    assumption that these all have the first frames' index/columns.
+
+    Returns
+    -------
+    new_frame : DataFrame
+    """
+    if axis == 1:
+        columns = frames[0].columns
+        zipped = [f.loc[:, c] for c in columns for f in frames]
+        return concat(zipped, axis=1)
+    else:
+        index = frames[0].index
+        zipped = [f.loc[i, :] for i in index for f in frames]
+        return DataFrame(zipped)
diff --git a/pandas/pandas/tests/frame/conftest.py b/pandas/pandas/tests/frame/conftest.py
new file mode 100644
index 0000000000000..b3140bad8276b
--- /dev/null
+++ b/pandas/pandas/tests/frame/conftest.py
@@ -0,0 +1,100 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    NaT,
+    date_range,
+)
+
+
+@pytest.fixture
+def datetime_frame() -> DataFrame:
+    """
+    Fixture for DataFrame of floats with DatetimeIndex
+
+    Columns are ['A', 'B', 'C', 'D']
+    """
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+
+@pytest.fixture
+def float_string_frame():
+    """
+    Fixture for DataFrame of floats and strings with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D', 'foo'].
+    """
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        index=Index([f"foo_{i}" for i in range(30)], dtype=object),
+        columns=Index(list("ABCD")),
+    )
+    df["foo"] = "bar"
+    return df
+
+
+@pytest.fixture
+def mixed_float_frame():
+    """
+    Fixture for DataFrame of different float types with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D'].
+    """
+    df = DataFrame(
+        {
+            col: np.random.default_rng(2).random(30, dtype=dtype)
+            for col, dtype in zip(
+                list("ABCD"), ["float32", "float32", "float32", "float64"]
+            )
+        },
+        index=Index([f"foo_{i}" for i in range(30)], dtype=object),
+    )
+    # not supported by numpy random
+    df["C"] = df["C"].astype("float16")
+    return df
+
+
+@pytest.fixture
+def mixed_int_frame():
+    """
+    Fixture for DataFrame of different int types with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D'].
+    """
+    return DataFrame(
+        {
+            col: np.ones(30, dtype=dtype)
+            for col, dtype in zip(list("ABCD"), ["int32", "uint64", "uint8", "int64"])
+        },
+        index=Index([f"foo_{i}" for i in range(30)], dtype=object),
+    )
+
+
+@pytest.fixture
+def timezone_frame():
+    """
+    Fixture for DataFrame of date_range Series with different time zones
+
+    Columns are ['A', 'B', 'C']; some entries are missing
+
+               A                         B                         C
+    0 2013-01-01 2013-01-01 00:00:00-05:00 2013-01-01 00:00:00+01:00
+    1 2013-01-02                       NaT                       NaT
+    2 2013-01-03 2013-01-03 00:00:00-05:00 2013-01-03 00:00:00+01:00
+    """
+    df = DataFrame(
+        {
+            "A": date_range("20130101", periods=3),
+            "B": date_range("20130101", periods=3, tz="US/Eastern"),
+            "C": date_range("20130101", periods=3, tz="CET"),
+        }
+    )
+    df.iloc[1, 1] = NaT
+    df.iloc[1, 2] = NaT
+    return df
diff --git a/pandas/pandas/tests/frame/constructors/__init__.py b/pandas/pandas/tests/frame/constructors/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/frame/constructors/test_from_dict.py b/pandas/pandas/tests/frame/constructors/test_from_dict.py
new file mode 100644
index 0000000000000..845174bbf600e
--- /dev/null
+++ b/pandas/pandas/tests/frame/constructors/test_from_dict.py
@@ -0,0 +1,223 @@
+from collections import OrderedDict
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestFromDict:
+    # Note: these tests are specific to the from_dict method, not for
+    #  passing dictionaries to DataFrame.__init__
+
+    def test_constructor_list_of_odicts(self):
+        data = [
+            OrderedDict([["a", 1.5], ["b", 3], ["c", 4], ["d", 6]]),
+            OrderedDict([["a", 1.5], ["b", 3], ["d", 6]]),
+            OrderedDict([["a", 1.5], ["d", 6]]),
+            OrderedDict(),
+            OrderedDict([["a", 1.5], ["b", 3], ["c", 4]]),
+            OrderedDict([["b", 3], ["c", 4], ["d", 6]]),
+        ]
+
+        result = DataFrame(data)
+        expected = DataFrame.from_dict(
+            dict(zip(range(len(data)), data)), orient="index"
+        )
+        tm.assert_frame_equal(result, expected.reindex(result.index))
+
+    def test_constructor_single_row(self):
+        data = [OrderedDict([["a", 1.5], ["b", 3], ["c", 4], ["d", 6]])]
+
+        result = DataFrame(data)
+        expected = DataFrame.from_dict(dict(zip([0], data)), orient="index").reindex(
+            result.index
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_series(self):
+        data = [
+            OrderedDict([["a", 1.5], ["b", 3.0], ["c", 4.0]]),
+            OrderedDict([["a", 1.5], ["b", 3.0], ["c", 6.0]]),
+        ]
+        sdict = OrderedDict(zip(["x", "y"], data))
+        idx = Index(["a", "b", "c"])
+
+        # all named
+        data2 = [
+            Series([1.5, 3, 4], idx, dtype="O", name="x"),
+            Series([1.5, 3, 6], idx, name="y"),
+        ]
+        result = DataFrame(data2)
+        expected = DataFrame.from_dict(sdict, orient="index")
+        tm.assert_frame_equal(result, expected)
+
+        # some unnamed
+        data2 = [
+            Series([1.5, 3, 4], idx, dtype="O", name="x"),
+            Series([1.5, 3, 6], idx),
+        ]
+        result = DataFrame(data2)
+
+        sdict = OrderedDict(zip(["x", "Unnamed 0"], data))
+        expected = DataFrame.from_dict(sdict, orient="index")
+        tm.assert_frame_equal(result, expected)
+
+        # none named
+        data = [
+            OrderedDict([["a", 1.5], ["b", 3], ["c", 4], ["d", 6]]),
+            OrderedDict([["a", 1.5], ["b", 3], ["d", 6]]),
+            OrderedDict([["a", 1.5], ["d", 6]]),
+            OrderedDict(),
+            OrderedDict([["a", 1.5], ["b", 3], ["c", 4]]),
+            OrderedDict([["b", 3], ["c", 4], ["d", 6]]),
+        ]
+        data = [Series(d) for d in data]
+
+        result = DataFrame(data)
+        sdict = OrderedDict(zip(range(len(data)), data))
+        expected = DataFrame.from_dict(sdict, orient="index")
+        tm.assert_frame_equal(result, expected.reindex(result.index))
+
+        result2 = DataFrame(data, index=np.arange(6, dtype=np.int64))
+        tm.assert_frame_equal(result, result2)
+
+        result = DataFrame([Series(dtype=object)])
+        expected = DataFrame(index=[0])
+        tm.assert_frame_equal(result, expected)
+
+        data = [
+            OrderedDict([["a", 1.5], ["b", 3.0], ["c", 4.0]]),
+            OrderedDict([["a", 1.5], ["b", 3.0], ["c", 6.0]]),
+        ]
+        sdict = OrderedDict(zip(range(len(data)), data))
+
+        idx = Index(["a", "b", "c"])
+        data2 = [Series([1.5, 3, 4], idx, dtype="O"), Series([1.5, 3, 6], idx)]
+        result = DataFrame(data2)
+        expected = DataFrame.from_dict(sdict, orient="index")
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_orient(self, float_string_frame):
+        data_dict = float_string_frame.T._series
+        recons = DataFrame.from_dict(data_dict, orient="index")
+        expected = float_string_frame.reindex(index=recons.index)
+        tm.assert_frame_equal(recons, expected)
+
+        # dict of sequence
+        a = {"hi": [32, 3, 3], "there": [3, 5, 3]}
+        rs = DataFrame.from_dict(a, orient="index")
+        xp = DataFrame.from_dict(a).T.reindex(list(a.keys()))
+        tm.assert_frame_equal(rs, xp)
+
+    def test_constructor_from_ordered_dict(self):
+        # GH#8425
+        a = OrderedDict(
+            [
+                ("one", OrderedDict([("col_a", "foo1"), ("col_b", "bar1")])),
+                ("two", OrderedDict([("col_a", "foo2"), ("col_b", "bar2")])),
+                ("three", OrderedDict([("col_a", "foo3"), ("col_b", "bar3")])),
+            ]
+        )
+        expected = DataFrame.from_dict(a, orient="columns").T
+        result = DataFrame.from_dict(a, orient="index")
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_dict_columns_parameter(self):
+        # GH#18529
+        # Test new columns parameter for from_dict that was added to make
+        # from_items(..., orient='index', columns=[...]) easier to replicate
+        result = DataFrame.from_dict(
+            OrderedDict([("A", [1, 2]), ("B", [4, 5])]),
+            orient="index",
+            columns=["one", "two"],
+        )
+        expected = DataFrame([[1, 2], [4, 5]], index=["A", "B"], columns=["one", "two"])
+        tm.assert_frame_equal(result, expected)
+
+        msg = "cannot use columns parameter with orient='columns'"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame.from_dict(
+                {"A": [1, 2], "B": [4, 5]},
+                orient="columns",
+                columns=["one", "two"],
+            )
+        with pytest.raises(ValueError, match=msg):
+            DataFrame.from_dict({"A": [1, 2], "B": [4, 5]}, columns=["one", "two"])
+
+    @pytest.mark.parametrize(
+        "data_dict, orient, expected",
+        [
+            ({}, "index", RangeIndex(0)),
+            (
+                [{("a",): 1}, {("a",): 2}],
+                "columns",
+                Index([("a",)], tupleize_cols=False),
+            ),
+            (
+                [OrderedDict([(("a",), 1), (("b",), 2)])],
+                "columns",
+                Index([("a",), ("b",)], tupleize_cols=False),
+            ),
+            ([{("a", "b"): 1}], "columns", Index([("a", "b")], tupleize_cols=False)),
+        ],
+    )
+    def test_constructor_from_dict_tuples(self, data_dict, orient, expected):
+        # GH#16769
+        df = DataFrame.from_dict(data_dict, orient)
+        result = df.columns
+        tm.assert_index_equal(result, expected)
+
+    def test_frame_dict_constructor_empty_series(self):
+        s1 = Series(
+            [1, 2, 3, 4], index=MultiIndex.from_tuples([(1, 2), (1, 3), (2, 2), (2, 4)])
+        )
+        s2 = Series(
+            [1, 2, 3, 4], index=MultiIndex.from_tuples([(1, 2), (1, 3), (3, 2), (3, 4)])
+        )
+        s3 = Series(dtype=object)
+
+        # it works!
+        DataFrame({"foo": s1, "bar": s2, "baz": s3})
+        DataFrame.from_dict({"foo": s1, "baz": s3, "bar": s2})
+
+    def test_from_dict_scalars_requires_index(self):
+        msg = "If using all scalar values, you must pass an index"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame.from_dict(OrderedDict([("b", 8), ("a", 5), ("a", 6)]))
+
+    def test_from_dict_orient_invalid(self):
+        msg = (
+            "Expected 'index', 'columns' or 'tight' for orient parameter. "
+            "Got 'abc' instead"
+        )
+        with pytest.raises(ValueError, match=msg):
+            DataFrame.from_dict({"foo": 1, "baz": 3, "bar": 2}, orient="abc")
+
+    def test_from_dict_order_with_single_column(self):
+        data = {
+            "alpha": {
+                "value2": 123,
+                "value1": 532,
+                "animal": 222,
+                "plant": False,
+                "name": "test",
+            }
+        }
+        result = DataFrame.from_dict(
+            data,
+            orient="columns",
+        )
+        expected = DataFrame(
+            [[123], [532], [222], [False], ["test"]],
+            index=["value2", "value1", "animal", "plant", "name"],
+            columns=["alpha"],
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/constructors/test_from_records.py b/pandas/pandas/tests/frame/constructors/test_from_records.py
new file mode 100644
index 0000000000000..1d4a2c0075e3e
--- /dev/null
+++ b/pandas/pandas/tests/frame/constructors/test_from_records.py
@@ -0,0 +1,494 @@
+from collections.abc import Iterator
+from datetime import (
+    datetime,
+    timezone,
+)
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import is_platform_little_endian
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    Interval,
+    RangeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestFromRecords:
+    def test_from_records_dt64tz_frame(self):
+        # GH#51697
+        df = DataFrame({"a": [1, 2, 3]})
+        with pytest.raises(TypeError, match="not supported"):
+            DataFrame.from_records(df)
+
+    def test_from_records_with_datetimes(self):
+        # this may fail on certain platforms because of a numpy issue
+        # related GH#6140
+        if not is_platform_little_endian():
+            pytest.skip("known failure of test on non-little endian")
+
+        # construction with a null in a recarray
+        # GH#6140
+        expected = DataFrame({"EXPIRY": [datetime(2005, 3, 1, 0, 0), None]})
+
+        arrdata = [np.array([datetime(2005, 3, 1, 0, 0), None])]
+        dtypes = [("EXPIRY", "<M8[us]")]
+
+        recarray = np.rec.fromarrays(arrdata, dtype=dtypes)
+
+        result = DataFrame.from_records(recarray)
+        tm.assert_frame_equal(result, expected)
+
+        # coercion should work too
+        arrdata = [np.array([datetime(2005, 3, 1, 0, 0), None])]
+        dtypes = [("EXPIRY", "<M8[m]")]
+        recarray = np.rec.fromarrays(arrdata, dtype=dtypes)
+        result = DataFrame.from_records(recarray)
+        # we get the closest supported unit, "s"
+        expected["EXPIRY"] = expected["EXPIRY"].astype("M8[s]")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.xfail(using_string_dtype(), reason="dtype checking logic doesn't work")
+    def test_from_records_sequencelike(self):
+        df = DataFrame(
+            {
+                "A": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float64
+                ),
+                "A1": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float64
+                ),
+                "B": np.array(np.arange(6), dtype=np.int64),
+                "C": ["foo"] * 6,
+                "D": np.array([True, False] * 3, dtype=bool),
+                "E": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float32
+                ),
+                "E1": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float32
+                ),
+                "F": np.array(np.arange(6), dtype=np.int32),
+            }
+        )
+
+        # this is actually tricky to create the recordlike arrays and
+        # have the dtypes be intact
+        blocks = df._to_dict_of_blocks()
+        tuples = []
+        columns = []
+        dtypes = []
+        for dtype, b in blocks.items():
+            columns.extend(b.columns)
+            dtypes.extend([(c, np.dtype(dtype).descr[0][1]) for c in b.columns])
+        for i in range(len(df.index)):
+            tup = []
+            for _, b in blocks.items():
+                tup.extend(b.iloc[i].values)
+            tuples.append(tuple(tup))
+
+        recarray = np.array(tuples, dtype=dtypes).view(np.rec.recarray)
+        recarray2 = df.to_records()
+        lists = [list(x) for x in tuples]
+
+        # tuples (lose the dtype info)
+        result = DataFrame.from_records(tuples, columns=columns).reindex(
+            columns=df.columns
+        )
+
+        # created recarray and with to_records recarray (have dtype info)
+        result2 = DataFrame.from_records(recarray, columns=columns).reindex(
+            columns=df.columns
+        )
+        result3 = DataFrame.from_records(recarray2, columns=columns).reindex(
+            columns=df.columns
+        )
+
+        # list of tuples (no dtype info)
+        result4 = DataFrame.from_records(lists, columns=columns).reindex(
+            columns=df.columns
+        )
+
+        tm.assert_frame_equal(result, df, check_dtype=False)
+        tm.assert_frame_equal(result2, df)
+        tm.assert_frame_equal(result3, df)
+        tm.assert_frame_equal(result4, df, check_dtype=False)
+
+        # tuples is in the order of the columns
+        result = DataFrame.from_records(tuples)
+        tm.assert_index_equal(result.columns, RangeIndex(8))
+
+        # test exclude parameter & we are casting the results here (as we don't
+        # have dtype info to recover)
+        columns_to_test = [columns.index("C"), columns.index("E1")]
+
+        exclude = list(set(range(8)) - set(columns_to_test))
+        result = DataFrame.from_records(tuples, exclude=exclude)
+        result.columns = [columns[i] for i in sorted(columns_to_test)]
+        tm.assert_series_equal(result["C"], df["C"])
+        tm.assert_series_equal(result["E1"], df["E1"])
+
+    def test_from_records_sequencelike_empty(self):
+        # empty case
+        result = DataFrame.from_records([], columns=["foo", "bar", "baz"])
+        assert len(result) == 0
+        tm.assert_index_equal(result.columns, Index(["foo", "bar", "baz"]))
+
+        result = DataFrame.from_records([])
+        assert len(result) == 0
+        assert len(result.columns) == 0
+
+    def test_from_records_sequencelike_empty_index(self):
+        result = DataFrame.from_records([], index=[])
+        assert len(result) == 0
+        assert len(result.columns) == 0
+        assert len(result.index) == 0
+
+    def test_from_records_dictlike(self):
+        # test the dict methods
+        df = DataFrame(
+            {
+                "A": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float64
+                ),
+                "A1": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float64
+                ),
+                "B": np.array(np.arange(6), dtype=np.int64),
+                "C": ["foo"] * 6,
+                "D": np.array([True, False] * 3, dtype=bool),
+                "E": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float32
+                ),
+                "E1": np.array(
+                    np.random.default_rng(2).standard_normal(6), dtype=np.float32
+                ),
+                "F": np.array(np.arange(6), dtype=np.int32),
+            }
+        )
+
+        # columns is in a different order here than the actual items iterated
+        # from the dict
+        blocks = df._to_dict_of_blocks()
+        columns = []
+        for b in blocks.values():
+            columns.extend(b.columns)
+
+        asdict = dict(df.items())
+        asdict2 = {x: y.values for x, y in df.items()}
+
+        # dict of series & dict of ndarrays (have dtype info)
+        results = []
+        results.append(DataFrame.from_records(asdict).reindex(columns=df.columns))
+        results.append(
+            DataFrame.from_records(asdict, columns=columns).reindex(columns=df.columns)
+        )
+        results.append(
+            DataFrame.from_records(asdict2, columns=columns).reindex(columns=df.columns)
+        )
+
+        for r in results:
+            tm.assert_frame_equal(r, df)
+
+    def test_from_records_non_tuple(self):
+        class Record:
+            def __init__(self, *args) -> None:
+                self.args = args
+
+            def __getitem__(self, i):
+                return self.args[i]
+
+            def __iter__(self) -> Iterator:
+                return iter(self.args)
+
+        recs = [Record(1, 2, 3), Record(4, 5, 6), Record(7, 8, 9)]
+        tups = [tuple(rec) for rec in recs]
+
+        result = DataFrame.from_records(recs)
+        expected = DataFrame.from_records(tups)
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_len0_with_columns(self):
+        # GH#2633
+        result = DataFrame.from_records([], index="foo", columns=["foo", "bar"])
+        expected = Index(["bar"])
+
+        assert len(result) == 0
+        assert result.index.name == "foo"
+        tm.assert_index_equal(result.columns, expected)
+
+    def test_from_records_series_list_dict(self):
+        # GH#27358
+        expected = DataFrame([[{"a": 1, "b": 2}, {"a": 3, "b": 4}]]).T
+        data = Series([[{"a": 1, "b": 2}], [{"a": 3, "b": 4}]])
+        result = DataFrame.from_records(data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_series_categorical_index(self):
+        # GH#32805
+        index = CategoricalIndex(
+            [Interval(-20, -10), Interval(-10, 0), Interval(0, 10)]
+        )
+        series_of_dicts = Series([{"a": 1}, {"a": 2}, {"b": 3}], index=index)
+        frame = DataFrame.from_records(series_of_dicts, index=index)
+        expected = DataFrame(
+            {"a": [1, 2, np.nan], "b": [np.nan, np.nan, 3]}, index=index
+        )
+        tm.assert_frame_equal(frame, expected)
+
+    def test_frame_from_records_utc(self):
+        rec = {"datum": 1.5, "begin_time": datetime(2006, 4, 27, tzinfo=timezone.utc)}
+
+        # it works
+        DataFrame.from_records([rec], index="begin_time")
+
+    def test_from_records_to_records(self):
+        # from numpy documentation
+        arr = np.zeros((2,), dtype=("i4,f4,S10"))
+        arr[:] = [(1, 2.0, "Hello"), (2, 3.0, "World")]
+
+        DataFrame.from_records(arr)
+
+        index = Index(np.arange(len(arr))[::-1])
+        indexed_frame = DataFrame.from_records(arr, index=index)
+        tm.assert_index_equal(indexed_frame.index, index)
+
+        # without names, it should go to last ditch
+        arr2 = np.zeros((2, 3))
+        tm.assert_frame_equal(DataFrame.from_records(arr2), DataFrame(arr2))
+
+        # wrong length
+        msg = "|".join(
+            [
+                r"Length of values \(2\) does not match length of index \(1\)",
+            ]
+        )
+        with pytest.raises(ValueError, match=msg):
+            DataFrame.from_records(arr, index=index[:-1])
+
+        indexed_frame = DataFrame.from_records(arr, index="f1")
+
+        # what to do?
+        records = indexed_frame.to_records()
+        assert len(records.dtype.names) == 3
+
+        records = indexed_frame.to_records(index=False)
+        assert len(records.dtype.names) == 2
+        assert "index" not in records.dtype.names
+
+    def test_from_records_nones(self):
+        tuples = [(1, 2, None, 3), (1, 2, None, 3), (None, 2, 5, 3)]
+
+        df = DataFrame.from_records(tuples, columns=["a", "b", "c", "d"])
+        assert np.isnan(df["c"][0])
+
+    def test_from_records_iterator(self):
+        arr = np.array(
+            [(1.0, 1.0, 2, 2), (3.0, 3.0, 4, 4), (5.0, 5.0, 6, 6), (7.0, 7.0, 8, 8)],
+            dtype=[
+                ("x", np.float64),
+                ("u", np.float32),
+                ("y", np.int64),
+                ("z", np.int32),
+            ],
+        )
+        df = DataFrame.from_records(iter(arr), nrows=2)
+        xp = DataFrame(
+            {
+                "x": np.array([1.0, 3.0], dtype=np.float64),
+                "u": np.array([1.0, 3.0], dtype=np.float32),
+                "y": np.array([2, 4], dtype=np.int64),
+                "z": np.array([2, 4], dtype=np.int32),
+            }
+        )
+        tm.assert_frame_equal(df.reindex_like(xp), xp)
+
+        # no dtypes specified here, so just compare with the default
+        arr = [(1.0, 2), (3.0, 4), (5.0, 6), (7.0, 8)]
+        df = DataFrame.from_records(iter(arr), columns=["x", "y"], nrows=2)
+        tm.assert_frame_equal(df, xp.reindex(columns=["x", "y"]), check_dtype=False)
+
+    def test_from_records_tuples_generator(self):
+        def tuple_generator(length):
+            for i in range(length):
+                letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
+                yield (i, letters[i % len(letters)], i / length)
+
+        columns_names = ["Integer", "String", "Float"]
+        columns = [
+            [i[j] for i in tuple_generator(10)] for j in range(len(columns_names))
+        ]
+        data = {"Integer": columns[0], "String": columns[1], "Float": columns[2]}
+        expected = DataFrame(data, columns=columns_names)
+
+        generator = tuple_generator(10)
+        result = DataFrame.from_records(generator, columns=columns_names)
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_lists_generator(self):
+        def list_generator(length):
+            for i in range(length):
+                letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
+                yield [i, letters[i % len(letters)], i / length]
+
+        columns_names = ["Integer", "String", "Float"]
+        columns = [
+            [i[j] for i in list_generator(10)] for j in range(len(columns_names))
+        ]
+        data = {"Integer": columns[0], "String": columns[1], "Float": columns[2]}
+        expected = DataFrame(data, columns=columns_names)
+
+        generator = list_generator(10)
+        result = DataFrame.from_records(generator, columns=columns_names)
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_columns_not_modified(self):
+        tuples = [(1, 2, 3), (1, 2, 3), (2, 5, 3)]
+
+        columns = ["a", "b", "c"]
+        original_columns = list(columns)
+
+        DataFrame.from_records(tuples, columns=columns, index="a")
+
+        assert columns == original_columns
+
+    def test_from_records_decimal(self):
+        tuples = [(Decimal("1.5"),), (Decimal("2.5"),), (None,)]
+
+        df = DataFrame.from_records(tuples, columns=["a"])
+        assert df["a"].dtype == object
+
+        df = DataFrame.from_records(tuples, columns=["a"], coerce_float=True)
+        assert df["a"].dtype == np.float64
+        assert np.isnan(df["a"].values[-1])
+
+    def test_from_records_duplicates(self):
+        result = DataFrame.from_records([(1, 2, 3), (4, 5, 6)], columns=["a", "b", "a"])
+
+        expected = DataFrame([(1, 2, 3), (4, 5, 6)], columns=["a", "b", "a"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_set_index_name(self):
+        def create_dict(order_id):
+            return {
+                "order_id": order_id,
+                "quantity": np.random.default_rng(2).integers(1, 10),
+                "price": np.random.default_rng(2).integers(1, 10),
+            }
+
+        documents = [create_dict(i) for i in range(10)]
+        # demo missing data
+        documents.append({"order_id": 10, "quantity": 5})
+
+        result = DataFrame.from_records(documents, index="order_id")
+        assert result.index.name == "order_id"
+
+        # MultiIndex
+        result = DataFrame.from_records(documents, index=["order_id", "quantity"])
+        assert result.index.names == ("order_id", "quantity")
+
+    def test_from_records_misc_brokenness(self):
+        # GH#2179
+
+        data = {1: ["foo"], 2: ["bar"]}
+
+        result = DataFrame.from_records(data, columns=["a", "b"])
+        exp = DataFrame(data, columns=["a", "b"])
+        tm.assert_frame_equal(result, exp)
+
+        # overlap in index/index_names
+
+        data = {"a": [1, 2, 3], "b": [4, 5, 6]}
+
+        result = DataFrame.from_records(data, index=["a", "b", "c"])
+        exp = DataFrame(data, index=["a", "b", "c"])
+        tm.assert_frame_equal(result, exp)
+
+    def test_from_records_misc_brokenness2(self):
+        # GH#2623
+        rows = []
+        rows.append([datetime(2010, 1, 1), 1])
+        rows.append([datetime(2010, 1, 2), "hi"])  # test col upconverts to obj
+        result = DataFrame.from_records(rows, columns=["date", "test"])
+        expected = DataFrame(
+            {"date": [row[0] for row in rows], "test": [row[1] for row in rows]}
+        )
+        tm.assert_frame_equal(result, expected)
+        assert result.dtypes["test"] == np.dtype(object)
+
+    def test_from_records_misc_brokenness3(self):
+        rows = []
+        rows.append([datetime(2010, 1, 1), 1])
+        rows.append([datetime(2010, 1, 2), 1])
+        result = DataFrame.from_records(rows, columns=["date", "test"])
+        expected = DataFrame(
+            {"date": [row[0] for row in rows], "test": [row[1] for row in rows]}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_empty(self):
+        # GH#3562
+        result = DataFrame.from_records([], columns=["a", "b", "c"])
+        expected = DataFrame(columns=["a", "b", "c"])
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame.from_records([], columns=["a", "b", "b"])
+        expected = DataFrame(columns=["a", "b", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_records_empty_with_nonempty_fields_gh3682(self):
+        a = np.array([(1, 2)], dtype=[("id", np.int64), ("value", np.int64)])
+        df = DataFrame.from_records(a, index="id")
+
+        ex_index = Index([1], name="id")
+        expected = DataFrame({"value": [2]}, index=ex_index, columns=["value"])
+        tm.assert_frame_equal(df, expected)
+
+        b = a[:0]
+        df2 = DataFrame.from_records(b, index="id")
+        tm.assert_frame_equal(df2, df.iloc[:0])
+
+    def test_from_records_empty2(self):
+        # GH#42456
+        dtype = [("prop", int)]
+        shape = (0, len(dtype))
+        arr = np.empty(shape, dtype=dtype)
+
+        result = DataFrame.from_records(arr)
+        expected = DataFrame({"prop": np.array([], dtype=int)})
+        tm.assert_frame_equal(result, expected)
+
+        alt = DataFrame(arr)
+        tm.assert_frame_equal(alt, expected)
+
+    def test_from_records_structured_array(self):
+        # GH 59717
+        data = np.array(
+            [
+                ("John", 25, "New York", 50000),
+                ("Jane", 30, "San Francisco", 75000),
+                ("Bob", 35, "Chicago", 65000),
+                ("Alice", 28, "Los Angeles", 60000),
+            ],
+            dtype=[("name", "U10"), ("age", "i4"), ("city", "U15"), ("salary", "i4")],
+        )
+
+        actual_result = DataFrame.from_records(data, columns=["name", "salary", "city"])
+
+        modified_data = {
+            "name": ["John", "Jane", "Bob", "Alice"],
+            "salary": np.array([50000, 75000, 65000, 60000], dtype="int32"),
+            "city": ["New York", "San Francisco", "Chicago", "Los Angeles"],
+        }
+        expected_result = DataFrame(modified_data)
+
+        tm.assert_frame_equal(actual_result, expected_result)
diff --git a/pandas/pandas/tests/frame/indexing/__init__.py b/pandas/pandas/tests/frame/indexing/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/frame/indexing/test_coercion.py b/pandas/pandas/tests/frame/indexing/test_coercion.py
new file mode 100644
index 0000000000000..472bfb7772a80
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_coercion.py
@@ -0,0 +1,176 @@
+"""
+Tests for values coercion in setitem-like operations on DataFrame.
+
+For the most part, these should be multi-column DataFrames, otherwise
+we would share the tests with Series.
+"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameSetitemCoercion:
+    @pytest.mark.parametrize("consolidate", [True, False])
+    def test_loc_setitem_multiindex_columns(self, consolidate):
+        # GH#18415 Setting values in a single column preserves dtype,
+        #  while setting them in multiple columns did unwanted cast.
+
+        # Note that A here has 2 blocks, below we do the same thing
+        #  with a consolidated frame.
+        A = DataFrame(np.zeros((6, 5), dtype=np.float32))
+        A = pd.concat([A, A], axis=1, keys=[1, 2])
+        if consolidate:
+            A = A._consolidate()
+
+        A.loc[2:3, (1, slice(2, 3))] = np.ones((2, 2), dtype=np.float32)
+        assert (A.dtypes == np.float32).all()
+
+        A.loc[0:5, (1, slice(2, 3))] = np.ones((6, 2), dtype=np.float32)
+
+        assert (A.dtypes == np.float32).all()
+
+        A.loc[:, (1, slice(2, 3))] = np.ones((6, 2), dtype=np.float32)
+        assert (A.dtypes == np.float32).all()
+
+        # TODO: i think this isn't about MultiIndex and could be done with iloc?
+
+
+def test_37477():
+    # fixed by GH#45121
+    orig = DataFrame({"A": [1, 2, 3], "B": [3, 4, 5]})
+
+    df = orig.copy()
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.at[1, "B"] = 1.2
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.loc[1, "B"] = 1.2
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.iat[1, 1] = 1.2
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.iloc[1, 1] = 1.2
+
+
+def test_6942(indexer_al):
+    # check that the .at __setitem__ after setting "Live" actually sets the data
+    start = Timestamp("2014-04-01")
+    t1 = Timestamp("2014-04-23 12:42:38.883082")
+    t2 = Timestamp("2014-04-24 01:33:30.040039")
+
+    dti = date_range(start, periods=1)
+    orig = DataFrame(index=dti, columns=["timenow", "Live"])
+
+    df = orig.copy()
+    indexer_al(df)[start, "timenow"] = t1
+
+    df["Live"] = True
+
+    df.at[start, "timenow"] = t2
+    assert df.iloc[0, 0] == t2
+
+
+def test_26395(indexer_al):
+    # .at case fixed by GH#45121 (best guess)
+    df = DataFrame(index=["A", "B", "C"])
+    df["D"] = 0
+
+    indexer_al(df)["C", "D"] = 2
+    expected = DataFrame({"D": [0, 0, 2]}, index=["A", "B", "C"], dtype=np.int64)
+    tm.assert_frame_equal(df, expected)
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        indexer_al(df)["C", "D"] = 44.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        indexer_al(df)["C", "D"] = "hello"
+
+
+@pytest.mark.xfail(reason="unwanted upcast")
+def test_15231():
+    df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+    df.loc[2] = Series({"a": 5, "b": 6})
+    assert (df.dtypes == np.int64).all()
+
+    df.loc[3] = Series({"a": 7})
+
+    # df["a"] doesn't have any NaNs, should not have been cast
+    exp_dtypes = Series([np.int64, np.float64], dtype=object, index=["a", "b"])
+    tm.assert_series_equal(df.dtypes, exp_dtypes)
+
+
+def test_iloc_setitem_unnecesssary_float_upcasting():
+    # GH#12255
+    df = DataFrame(
+        {
+            0: np.array([1, 3], dtype=np.float32),
+            1: np.array([2, 4], dtype=np.float32),
+            2: ["a", "b"],
+        }
+    )
+    orig = df.copy()
+
+    values = df[0].values.reshape(2, 1)
+    df.iloc[:, 0:1] = values
+
+    tm.assert_frame_equal(df, orig)
+
+
+@pytest.mark.xfail(reason="unwanted casting to dt64")
+def test_12499():
+    # TODO: OP in GH#12499 used np.datetim64("NaT") instead of pd.NaT,
+    #  which has consequences for the expected df["two"] (though i think at
+    #  the time it might not have because of a separate bug). See if it makes
+    #  a difference which one we use here.
+    ts = Timestamp("2016-03-01 03:13:22.98986", tz="UTC")
+
+    data = [{"one": 0, "two": ts}]
+    orig = DataFrame(data)
+    df = orig.copy()
+    df.loc[1] = [np.nan, NaT]
+
+    expected = DataFrame(
+        {"one": [0, np.nan], "two": Series([ts, NaT], dtype="datetime64[ns, UTC]")}
+    )
+    tm.assert_frame_equal(df, expected)
+
+    data = [{"one": 0, "two": ts}]
+    df = orig.copy()
+    df.loc[1, :] = [np.nan, NaT]
+    tm.assert_frame_equal(df, expected)
+
+
+def test_20476():
+    mi = MultiIndex.from_product([["A", "B"], ["a", "b", "c"]])
+    df = DataFrame(-1, index=range(3), columns=mi)
+    filler = DataFrame([[1, 2, 3.0]] * 3, index=range(3), columns=["a", "b", "c"])
+    df["A"] = filler
+
+    expected = DataFrame(
+        {
+            0: [1, 1, 1],
+            1: [2, 2, 2],
+            2: [3.0, 3.0, 3.0],
+            3: [-1, -1, -1],
+            4: [-1, -1, -1],
+            5: [-1, -1, -1],
+        }
+    )
+    expected.columns = mi
+    exp_dtypes = Series(
+        [np.dtype(np.int64)] * 2 + [np.dtype(np.float64)] + [np.dtype(np.int64)] * 3,
+        index=mi,
+    )
+    tm.assert_series_equal(df.dtypes, exp_dtypes)
diff --git a/pandas/pandas/tests/frame/indexing/test_delitem.py b/pandas/pandas/tests/frame/indexing/test_delitem.py
new file mode 100644
index 0000000000000..daec991b7a8db
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_delitem.py
@@ -0,0 +1,60 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+
+
+class TestDataFrameDelItem:
+    def test_delitem(self, float_frame):
+        del float_frame["A"]
+        assert "A" not in float_frame
+
+    def test_delitem_multiindex(self):
+        midx = MultiIndex.from_product([["A", "B"], [1, 2]])
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 4)), columns=midx)
+        assert len(df.columns) == 4
+        assert ("A",) in df.columns
+        assert "A" in df.columns
+
+        result = df["A"]
+        assert isinstance(result, DataFrame)
+        del df["A"]
+
+        assert len(df.columns) == 2
+
+        # A still in the levels, BUT get a KeyError if trying
+        # to delete
+        assert ("A",) not in df.columns
+        with pytest.raises(KeyError, match=re.escape("('A',)")):
+            del df[("A",)]
+
+        # behavior of dropped/deleted MultiIndex levels changed from
+        # GH 2770 to GH 19027: MultiIndex no longer '.__contains__'
+        # levels which are dropped/deleted
+        assert "A" not in df.columns
+        with pytest.raises(KeyError, match=re.escape("('A',)")):
+            del df["A"]
+
+    def test_delitem_corner(self, float_frame):
+        f = float_frame.copy()
+        del f["D"]
+        assert len(f.columns) == 3
+        with pytest.raises(KeyError, match=r"^'D'$"):
+            del f["D"]
+        del f["B"]
+        assert len(f.columns) == 2
+
+    def test_delitem_col_still_multiindex(self):
+        arrays = [["a", "b", "c", "top"], ["", "", "", "OD"], ["", "", "", "wx"]]
+
+        tuples = sorted(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 4)), columns=index)
+        del df[("a", "", "")]
+        assert isinstance(df.columns, MultiIndex)
diff --git a/pandas/pandas/tests/frame/indexing/test_get.py b/pandas/pandas/tests/frame/indexing/test_get.py
new file mode 100644
index 0000000000000..75bad0ec1f159
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_get.py
@@ -0,0 +1,27 @@
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestGet:
+    def test_get(self, float_frame):
+        b = float_frame.get("B")
+        tm.assert_series_equal(b, float_frame["B"])
+
+        assert float_frame.get("foo") is None
+        tm.assert_series_equal(
+            float_frame.get("foo", float_frame["B"]), float_frame["B"]
+        )
+
+    @pytest.mark.parametrize(
+        "columns, index",
+        [
+            [None, None],
+            [list("AB"), None],
+            [list("AB"), range(3)],
+        ],
+    )
+    def test_get_none(self, columns, index):
+        # see gh-5652
+        assert DataFrame(columns=columns, index=index).get(None) is None
diff --git a/pandas/pandas/tests/frame/indexing/test_get_value.py b/pandas/pandas/tests/frame/indexing/test_get_value.py
new file mode 100644
index 0000000000000..65a1c64a1578a
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_get_value.py
@@ -0,0 +1,22 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+
+
+class TestGetValue:
+    def test_get_set_value_no_partial_indexing(self):
+        # partial w/ MultiIndex raise exception
+        index = MultiIndex.from_tuples([(0, 1), (0, 2), (1, 1), (1, 2)])
+        df = DataFrame(index=index, columns=range(4))
+        with pytest.raises(KeyError, match=r"^0$"):
+            df._get_value(0, 1)
+
+    def test_get_value(self, float_frame):
+        for idx in float_frame.index:
+            for col in float_frame.columns:
+                result = float_frame._get_value(idx, col)
+                expected = float_frame[col][idx]
+                assert result == expected
diff --git a/pandas/pandas/tests/frame/indexing/test_getitem.py b/pandas/pandas/tests/frame/indexing/test_getitem.py
new file mode 100644
index 0000000000000..25d6e06a4c675
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_getitem.py
@@ -0,0 +1,466 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    DateOffset,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    get_dummies,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import SparseArray
+
+
+class TestGetitem:
+    def test_getitem_unused_level_raises(self):
+        # GH#20410
+        mi = MultiIndex(
+            levels=[["a_lot", "onlyone", "notevenone"], [1970, ""]],
+            codes=[[1, 0], [1, 0]],
+        )
+        df = DataFrame(-1, index=range(3), columns=mi)
+
+        with pytest.raises(KeyError, match="notevenone"):
+            df["notevenone"]
+
+    def test_getitem_periodindex(self):
+        rng = period_range("1/1/2000", periods=5)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)), columns=rng)
+
+        ts = df[rng[0]]
+        tm.assert_series_equal(ts, df.iloc[:, 0])
+
+        ts = df["1/1/2000"]
+        tm.assert_series_equal(ts, df.iloc[:, 0])
+
+    def test_getitem_list_of_labels_categoricalindex_cols(self):
+        # GH#16115
+        cats = Categorical([Timestamp("12-31-1999"), Timestamp("12-31-2000")])
+
+        expected = DataFrame([[1, 0], [0, 1]], dtype="bool", index=[0, 1], columns=cats)
+        dummies = get_dummies(cats)
+        result = dummies[list(dummies.columns)]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_sparse_column_return_type_and_dtype(self):
+        # https://github.com/pandas-dev/pandas/issues/23559
+        data = SparseArray([0, 1])
+        df = DataFrame({"A": data})
+        expected = Series(data, name="A")
+        result = df["A"]
+        tm.assert_series_equal(result, expected)
+
+        # Also check iloc and loc while we're here
+        result = df.iloc[:, 0]
+        tm.assert_series_equal(result, expected)
+
+        result = df.loc[:, "A"]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_string_columns(self):
+        # GH#46185
+        df = DataFrame([[1, 2]], columns=Index(["A", "B"], dtype="string"))
+        result = df.A
+        expected = df["A"]
+        tm.assert_series_equal(result, expected)
+
+
+class TestGetitemListLike:
+    def test_getitem_list_missing_key(self):
+        # GH#13822, incorrect error string with non-unique columns when missing
+        # column is accessed
+        df = DataFrame({"x": [1.0], "y": [2.0], "z": [3.0]})
+        df.columns = ["x", "x", "z"]
+
+        # Check that we get the correct value in the KeyError
+        with pytest.raises(KeyError, match=r"\['y'\] not in index"):
+            df[["x", "y", "z"]]
+
+    def test_getitem_list_duplicates(self):
+        # GH#1943
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), columns=list("AABC")
+        )
+        df.columns.name = "foo"
+
+        result = df[["B", "C"]]
+        assert result.columns.name == "foo"
+
+        expected = df.iloc[:, 2:]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_dupe_cols(self):
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["a", "a", "b"])
+        msg = "\"None of [Index(['baf'], dtype="
+        with pytest.raises(KeyError, match=re.escape(msg)):
+            df[["baf"]]
+
+    @pytest.mark.parametrize(
+        "idx_type",
+        [
+            list,
+            iter,
+            Index,
+            set,
+            lambda keys: dict(zip(keys, range(len(keys)))),
+            lambda keys: dict(zip(keys, range(len(keys)))).keys(),
+        ],
+        ids=["list", "iter", "Index", "set", "dict", "dict_keys"],
+    )
+    @pytest.mark.parametrize("levels", [1, 2])
+    def test_getitem_listlike(self, idx_type, levels, float_frame):
+        # GH#21294
+
+        if levels == 1:
+            frame, missing = float_frame, "food"
+        else:
+            # MultiIndex columns
+            frame = DataFrame(
+                np.random.default_rng(2).standard_normal((8, 3)),
+                columns=Index(
+                    [("foo", "bar"), ("baz", "qux"), ("peek", "aboo")],
+                    name=("sth", "sth2"),
+                ),
+            )
+            missing = ("good", "food")
+
+        keys = [frame.columns[1], frame.columns[0]]
+        idx = idx_type(keys)
+        idx_check = list(idx_type(keys))
+
+        if isinstance(idx, (set, dict)):
+            with pytest.raises(TypeError, match="as an indexer is not supported"):
+                frame[idx]
+
+            return
+        else:
+            result = frame[idx]
+
+        expected = frame.loc[:, idx_check]
+        expected.columns.names = frame.columns.names
+
+        tm.assert_frame_equal(result, expected)
+
+        idx = idx_type(keys + [missing])
+        with pytest.raises(KeyError, match="not in index"):
+            frame[idx]
+
+    def test_getitem_iloc_generator(self):
+        # GH#39614
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        indexer = (x for x in [1, 2])
+        result = df.iloc[indexer]
+        expected = DataFrame({"a": [2, 3], "b": [5, 6]}, index=[1, 2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_iloc_two_dimensional_generator(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        indexer = (x for x in [1, 2])
+        result = df.iloc[indexer, 1]
+        expected = Series([5, 6], name="b", index=[1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_iloc_dateoffset_days(self):
+        # GH 46671
+        df = DataFrame(
+            list(range(10)),
+            index=date_range("01-01-2022", periods=10, freq=DateOffset(days=1)),
+        )
+        result = df.loc["2022-01-01":"2022-01-03"]
+        expected = DataFrame(
+            [0, 1, 2],
+            index=DatetimeIndex(
+                ["2022-01-01", "2022-01-02", "2022-01-03"],
+                dtype="datetime64[ns]",
+                freq=DateOffset(days=1),
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            list(range(10)),
+            index=date_range(
+                "01-01-2022", periods=10, freq=DateOffset(days=1, hours=2)
+            ),
+        )
+        result = df.loc["2022-01-01":"2022-01-03"]
+        expected = DataFrame(
+            [0, 1, 2],
+            index=DatetimeIndex(
+                ["2022-01-01 00:00:00", "2022-01-02 02:00:00", "2022-01-03 04:00:00"],
+                dtype="datetime64[ns]",
+                freq=DateOffset(days=1, hours=2),
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            list(range(10)),
+            index=date_range("01-01-2022", periods=10, freq=DateOffset(minutes=3)),
+        )
+        result = df.loc["2022-01-01":"2022-01-03"]
+        tm.assert_frame_equal(result, df)
+
+
+class TestGetitemCallable:
+    def test_getitem_callable(self, float_frame):
+        # GH#12533
+        result = float_frame[lambda x: "A"]
+        expected = float_frame.loc[:, "A"]
+        tm.assert_series_equal(result, expected)
+
+        result = float_frame[lambda x: ["A", "B"]]
+        expected = float_frame.loc[:, ["A", "B"]]
+        tm.assert_frame_equal(result, float_frame.loc[:, ["A", "B"]])
+
+        df = float_frame[:3]
+        result = df[lambda x: [True, False, True]]
+        expected = float_frame.iloc[[0, 2], :]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_multiindex_columns_one_level(self):
+        # GH#29749
+        df = DataFrame([[1, 2]], columns=[["a", "b"]])
+        expected = DataFrame([1], columns=[["a"]])
+
+        result = df["a"]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[:, "a"]
+        tm.assert_frame_equal(result, expected)
+
+
+class TestGetitemBooleanMask:
+    def test_getitem_bool_mask_categorical_index(self):
+        df3 = DataFrame(
+            {
+                "A": np.arange(6, dtype="int64"),
+            },
+            index=CategoricalIndex(
+                [1, 1, 2, 1, 3, 2],
+                dtype=CategoricalDtype([3, 2, 1], ordered=True),
+                name="B",
+            ),
+        )
+        df4 = DataFrame(
+            {
+                "A": np.arange(6, dtype="int64"),
+            },
+            index=CategoricalIndex(
+                [1, 1, 2, 1, 3, 2],
+                dtype=CategoricalDtype([3, 2, 1], ordered=False),
+                name="B",
+            ),
+        )
+
+        result = df3[df3.index == "a"]
+        expected = df3.iloc[[]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df4[df4.index == "a"]
+        expected = df4.iloc[[]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df3[df3.index == 1]
+        expected = df3.iloc[[0, 1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df4[df4.index == 1]
+        expected = df4.iloc[[0, 1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # since we have an ordered categorical
+
+        # CategoricalIndex([1, 1, 2, 1, 3, 2],
+        #         categories=[3, 2, 1],
+        #         ordered=True,
+        #         name='B')
+        result = df3[df3.index < 2]
+        expected = df3.iloc[[4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df3[df3.index > 1]
+        expected = df3.iloc[[]]
+        tm.assert_frame_equal(result, expected)
+
+        # unordered
+        # cannot be compared
+
+        # CategoricalIndex([1, 1, 2, 1, 3, 2],
+        #         categories=[3, 2, 1],
+        #         ordered=False,
+        #         name='B')
+        msg = "Unordered Categoricals can only compare equality or not"
+        with pytest.raises(TypeError, match=msg):
+            df4[df4.index < 2]
+        with pytest.raises(TypeError, match=msg):
+            df4[df4.index > 1]
+
+    @pytest.mark.parametrize(
+        "data1,data2,expected_data",
+        (
+            (
+                [[1, 2], [3, 4]],
+                [[0.5, 6], [7, 8]],
+                [[np.nan, 3.0], [np.nan, 4.0], [np.nan, 7.0], [6.0, 8.0]],
+            ),
+            (
+                [[1, 2], [3, 4]],
+                [[5, 6], [7, 8]],
+                [[np.nan, 3.0], [np.nan, 4.0], [5, 7], [6, 8]],
+            ),
+        ),
+    )
+    def test_getitem_bool_mask_duplicate_columns_mixed_dtypes(
+        self,
+        data1,
+        data2,
+        expected_data,
+    ):
+        # GH#31954
+
+        df1 = DataFrame(np.array(data1))
+        df2 = DataFrame(np.array(data2))
+        df = concat([df1, df2], axis=1)
+
+        result = df[df > 2]
+
+        exdict = {i: np.array(col) for i, col in enumerate(expected_data)}
+        expected = DataFrame(exdict).rename(columns={2: 0, 3: 1})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.fixture
+    def df_dup_cols(self):
+        dups = ["A", "A", "C", "D"]
+        df = DataFrame(np.arange(12).reshape(3, 4), columns=dups, dtype="float64")
+        return df
+
+    def test_getitem_boolean_frame_unaligned_with_duplicate_columns(self, df_dup_cols):
+        # `df.A > 6` is a DataFrame with a different shape from df
+
+        # boolean with the duplicate raises
+        df = df_dup_cols
+        msg = "cannot reindex on an axis with duplicate labels"
+        with pytest.raises(ValueError, match=msg):
+            df[df.A > 6]
+
+    def test_getitem_boolean_series_with_duplicate_columns(self, df_dup_cols):
+        # boolean indexing
+        # GH#4879
+        df = DataFrame(
+            np.arange(12).reshape(3, 4), columns=["A", "B", "C", "D"], dtype="float64"
+        )
+        expected = df[df.C > 6]
+        expected.columns = df_dup_cols.columns
+
+        df = df_dup_cols
+        result = df[df.C > 6]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_boolean_frame_with_duplicate_columns(self, df_dup_cols):
+        # where
+        df = DataFrame(
+            np.arange(12).reshape(3, 4), columns=["A", "B", "C", "D"], dtype="float64"
+        )
+        # `df > 6` is a DataFrame with the same shape+alignment as df
+        expected = df[df > 6]
+        expected.columns = df_dup_cols.columns
+
+        df = df_dup_cols
+        result = df[df > 6]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_empty_frame_with_boolean(self):
+        # Test for issue GH#11859
+
+        df = DataFrame()
+        df2 = df[df > 0]
+        tm.assert_frame_equal(df, df2)
+
+    def test_getitem_returns_view_when_column_is_unique_in_df(self):
+        # GH#45316
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["a", "a", "b"])
+        df_orig = df.copy()
+        view = df["b"]
+        view.loc[:] = 100
+        expected = df_orig
+        tm.assert_frame_equal(df, expected)
+
+    def test_getitem_frozenset_unique_in_column(self):
+        # GH#41062
+        df = DataFrame([[1, 2, 3, 4]], columns=[frozenset(["KEY"]), "B", "C", "C"])
+        result = df[frozenset(["KEY"])]
+        expected = Series([1], name=frozenset(["KEY"]))
+        tm.assert_series_equal(result, expected)
+
+
+class TestGetitemSlice:
+    def test_getitem_slice_float64(self, frame_or_series):
+        values = np.arange(10.0, 50.0, 2)
+        index = Index(values)
+
+        start, end = values[[5, 15]]
+
+        data = np.random.default_rng(2).standard_normal((20, 3))
+        if frame_or_series is not DataFrame:
+            data = data[:, 0]
+
+        obj = frame_or_series(data, index=index)
+
+        result = obj[start:end]
+        expected = obj.iloc[5:16]
+        tm.assert_equal(result, expected)
+
+        result = obj.loc[start:end]
+        tm.assert_equal(result, expected)
+
+    def test_getitem_datetime_slice(self):
+        # GH#43223
+        df = DataFrame(
+            {"a": 0},
+            index=DatetimeIndex(
+                [
+                    "11.01.2011 22:00",
+                    "11.01.2011 23:00",
+                    "12.01.2011 00:00",
+                    "2011-01-13 00:00",
+                ]
+            ),
+        )
+        with pytest.raises(
+            KeyError, match="Value based partial slicing on non-monotonic"
+        ):
+            df["2011-01-01":"2011-11-01"]
+
+    def test_getitem_slice_same_dim_only_one_axis(self):
+        # GH#54622
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 8)))
+        result = df.iloc[(slice(None, None, 2),)]
+        assert result.shape == (5, 8)
+        expected = df.iloc[slice(None, None, 2), slice(None)]
+        tm.assert_frame_equal(result, expected)
+
+
+class TestGetitemDeprecatedIndexers:
+    @pytest.mark.parametrize("key", [{"a", "b"}, {"a": "a"}])
+    def test_getitem_dict_and_set_deprecated(self, key):
+        # GH#42825 enforced in 2.0
+        df = DataFrame(
+            [[1, 2], [3, 4]], columns=MultiIndex.from_tuples([("a", 1), ("b", 2)])
+        )
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            df[key]
diff --git a/pandas/pandas/tests/frame/indexing/test_indexing.py b/pandas/pandas/tests/frame/indexing/test_indexing.py
new file mode 100644
index 0000000000000..0c99b08cb30c4
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_indexing.py
@@ -0,0 +1,1941 @@
+from collections import namedtuple
+from datetime import (
+    datetime,
+    timedelta,
+)
+from decimal import Decimal
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import iNaT
+from pandas.errors import InvalidIndexError
+
+from pandas.core.dtypes.common import is_integer
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+    to_datetime,
+)
+import pandas._testing as tm
+
+# We pass through a TypeError raised by numpy
+_slice_msg = "slice indices must be integers or None or have an __index__ method"
+
+
+class TestDataFrameIndexing:
+    def test_getitem(self, float_frame):
+        # Slicing
+        sl = float_frame[:20]
+        assert len(sl.index) == 20
+
+        # Column access
+        for _, series in sl.items():
+            assert len(series.index) == 20
+            tm.assert_index_equal(series.index, sl.index)
+
+        for key, _ in float_frame._series.items():
+            assert float_frame[key] is not None
+
+        assert "random" not in float_frame
+        with pytest.raises(KeyError, match="random"):
+            float_frame["random"]
+
+    def test_getitem_numeric_should_not_fallback_to_positional(self, any_numeric_dtype):
+        # GH51053
+        dtype = any_numeric_dtype
+        idx = Index([1, 0, 1], dtype=dtype)
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=idx)
+        result = df[1]
+        expected = DataFrame([[1, 3], [4, 6]], columns=Index([1, 1], dtype=dtype))
+        tm.assert_frame_equal(result, expected, check_exact=True)
+
+    def test_getitem2(self, float_frame):
+        df = float_frame.copy()
+        df["$10"] = np.random.default_rng(2).standard_normal(len(df))
+
+        ad = np.random.default_rng(2).standard_normal(len(df))
+        df["@awesome_domain"] = ad
+
+        with pytest.raises(KeyError, match=re.escape("'df[\"$10\"]'")):
+            df.__getitem__('df["$10"]')
+
+        res = df["@awesome_domain"]
+        tm.assert_numpy_array_equal(ad, res.values)
+
+    def test_setitem_numeric_should_not_fallback_to_positional(self, any_numeric_dtype):
+        # GH51053
+        dtype = any_numeric_dtype
+        idx = Index([1, 0, 1], dtype=dtype)
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=idx)
+        df[1] = 10
+        expected = DataFrame([[10, 2, 10], [10, 5, 10]], columns=idx)
+        tm.assert_frame_equal(df, expected, check_exact=True)
+
+    def test_setitem_list(self, float_frame):
+        float_frame["E"] = "foo"
+        data = float_frame[["A", "B"]]
+        float_frame[["B", "A"]] = data
+
+        tm.assert_series_equal(float_frame["B"], data["A"], check_names=False)
+        tm.assert_series_equal(float_frame["A"], data["B"], check_names=False)
+
+        msg = "Columns must be same length as key"
+        with pytest.raises(ValueError, match=msg):
+            data[["A"]] = float_frame[["A", "B"]]
+        newcolumndata = range(len(data.index) - 1)
+        msg = (
+            rf"Length of values \({len(newcolumndata)}\) "
+            rf"does not match length of index \({len(data)}\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            data["A"] = newcolumndata
+
+    def test_setitem_list2(self):
+        df = DataFrame(0, index=range(3), columns=["tt1", "tt2"], dtype=int)
+        df.loc[1, ["tt1", "tt2"]] = [1, 2]
+
+        result = df.loc[df.index[1], ["tt1", "tt2"]]
+        expected = Series([1, 2], df.columns, dtype=int, name=1)
+        tm.assert_series_equal(result, expected)
+
+        df["tt1"] = df["tt2"] = "0"
+        df.loc[df.index[1], ["tt1", "tt2"]] = ["1", "2"]
+        result = df.loc[df.index[1], ["tt1", "tt2"]]
+        expected = Series(["1", "2"], df.columns, name=1)
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_boolean(self, mixed_float_frame, mixed_int_frame, datetime_frame):
+        # boolean indexing
+        d = datetime_frame.index[len(datetime_frame) // 2]
+        indexer = datetime_frame.index > d
+        indexer_obj = indexer.astype(object)
+
+        subindex = datetime_frame.index[indexer]
+        subframe = datetime_frame[indexer]
+
+        tm.assert_index_equal(subindex, subframe.index)
+        with pytest.raises(ValueError, match="Item wrong length"):
+            datetime_frame[indexer[:-1]]
+
+        subframe_obj = datetime_frame[indexer_obj]
+        tm.assert_frame_equal(subframe_obj, subframe)
+
+        with pytest.raises(TypeError, match="Boolean array expected"):
+            datetime_frame[datetime_frame]
+
+        # test that Series work
+        indexer_obj = Series(indexer_obj, datetime_frame.index)
+
+        subframe_obj = datetime_frame[indexer_obj]
+        tm.assert_frame_equal(subframe_obj, subframe)
+
+        # test that Series indexers reindex
+        # we are producing a warning that since the passed boolean
+        # key is not the same as the given index, we will reindex
+        # not sure this is really necessary
+        with tm.assert_produces_warning(UserWarning, match="will be reindexed"):
+            indexer_obj = indexer_obj.reindex(datetime_frame.index[::-1])
+            subframe_obj = datetime_frame[indexer_obj]
+            tm.assert_frame_equal(subframe_obj, subframe)
+
+        # test df[df > 0]
+        for df in [
+            datetime_frame,
+            mixed_float_frame,
+            mixed_int_frame,
+        ]:
+            data = df._get_numeric_data()
+            bif = df[df > 0]
+            bifw = DataFrame(
+                {c: np.where(data[c] > 0, data[c], np.nan) for c in data.columns},
+                index=data.index,
+                columns=data.columns,
+            )
+
+            # add back other columns to compare
+            for c in df.columns:
+                if c not in bifw:
+                    bifw[c] = df[c]
+            bifw = bifw.reindex(columns=df.columns)
+
+            tm.assert_frame_equal(bif, bifw, check_dtype=False)
+            for c in df.columns:
+                if bif[c].dtype != bifw[c].dtype:
+                    assert bif[c].dtype == df[c].dtype
+
+    def test_getitem_boolean_casting(self, datetime_frame):
+        # don't upcast if we don't need to
+        df = datetime_frame.copy()
+        df["E"] = 1
+        df["E"] = df["E"].astype("int32")
+        df["E1"] = df["E"].copy()
+        df["F"] = 1
+        df["F"] = df["F"].astype("int64")
+        df["F1"] = df["F"].copy()
+
+        casted = df[df > 0]
+        result = casted.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 4
+            + [np.dtype("int32")] * 2
+            + [np.dtype("int64")] * 2,
+            index=["A", "B", "C", "D", "E", "E1", "F", "F1"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # int block splitting
+        df.loc[df.index[1:3], ["E1", "F1"]] = 0
+        casted = df[df > 0]
+        result = casted.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 4
+            + [np.dtype("int32")]
+            + [np.dtype("float64")]
+            + [np.dtype("int64")]
+            + [np.dtype("float64")],
+            index=["A", "B", "C", "D", "E", "E1", "F", "F1"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "lst", [[True, False, True], [True, True, True], [False, False, False]]
+    )
+    def test_getitem_boolean_list(self, lst):
+        df = DataFrame(np.arange(12).reshape(3, 4))
+        result = df[lst]
+        expected = df.loc[df.index[lst]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_boolean_iadd(self):
+        arr = np.random.default_rng(2).standard_normal((5, 5))
+
+        df = DataFrame(arr.copy(), columns=["A", "B", "C", "D", "E"])
+
+        df[df < 0] += 1
+        arr[arr < 0] += 1
+
+        tm.assert_almost_equal(df.values, arr)
+
+    def test_boolean_index_empty_corner(self):
+        # #2096
+        blah = DataFrame(np.empty([0, 1]), columns=["A"], index=DatetimeIndex([]))
+
+        # both of these should succeed trivially
+        k = np.array([], bool)
+
+        blah[k]
+        blah[k] = 0
+
+    def test_getitem_ix_mixed_integer(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 3)),
+            index=[1, 10, "C", "E"],
+            columns=[1, 2, 3],
+        )
+
+        result = df.iloc[:-1]
+        expected = df.loc[df.index[:-1]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[[1, 10]]
+        expected = df.loc[Index([1, 10])]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_ix_mixed_integer2(self):
+        # 11320
+        df = DataFrame(
+            {
+                "rna": (1.5, 2.2, 3.2, 4.5),
+                -1000: [11, 21, 36, 40],
+                0: [10, 22, 43, 34],
+                1000: [0, 10, 20, 30],
+            },
+            columns=["rna", -1000, 0, 1000],
+        )
+        result = df[[1000]]
+        expected = df.iloc[:, [3]]
+        tm.assert_frame_equal(result, expected)
+        result = df[[-1000]]
+        expected = df.iloc[:, [1]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getattr(self, float_frame):
+        tm.assert_series_equal(float_frame.A, float_frame["A"])
+        msg = "'DataFrame' object has no attribute 'NONEXISTENT_NAME'"
+        with pytest.raises(AttributeError, match=msg):
+            float_frame.NONEXISTENT_NAME
+
+    def test_setattr_column(self):
+        df = DataFrame({"foobar": 1}, index=range(10))
+
+        df.foobar = 5
+        assert (df.foobar == 5).all()
+
+    def test_setitem(self, float_frame, using_infer_string):
+        # not sure what else to do here
+        series = float_frame["A"][::2]
+        float_frame["col5"] = series
+        assert "col5" in float_frame
+
+        assert len(series) == 15
+        assert len(float_frame) == 30
+
+        exp = np.ravel(np.column_stack((series.values, [np.nan] * 15)))
+        exp = Series(exp, index=float_frame.index, name="col5")
+        tm.assert_series_equal(float_frame["col5"], exp)
+
+        series = float_frame["A"]
+        float_frame["col6"] = series
+        tm.assert_series_equal(series, float_frame["col6"], check_names=False)
+
+        # set ndarray
+        arr = np.random.default_rng(2).standard_normal(len(float_frame))
+        float_frame["col9"] = arr
+        assert (float_frame["col9"] == arr).all()
+
+        float_frame["col7"] = 5
+        assert (float_frame["col7"] == 5).all()
+
+        float_frame["col0"] = 3.14
+        assert (float_frame["col0"] == 3.14).all()
+
+        float_frame["col8"] = "foo"
+        assert (float_frame["col8"] == "foo").all()
+
+        # this is partially a view (e.g. some blocks are view)
+        # so raise/warn
+        smaller = float_frame[:2]
+
+        smaller["col10"] = ["1", "2"]
+
+        if using_infer_string:
+            assert smaller["col10"].dtype == "str"
+        else:
+            assert smaller["col10"].dtype == np.object_
+        assert (smaller["col10"] == ["1", "2"]).all()
+
+    def test_setitem2(self):
+        # dtype changing GH4204
+        df = DataFrame([[0, 0]])
+        df.iloc[0] = np.nan
+        expected = DataFrame([[np.nan, np.nan]])
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame([[0, 0]])
+        df.loc[0] = np.nan
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_boolean(self, float_frame):
+        df = float_frame.copy()
+        values = float_frame.values.copy()
+
+        df[df["A"] > 0] = 4
+        values[values[:, 0] > 0] = 4
+        tm.assert_almost_equal(df.values, values)
+
+        # test that column reindexing works
+        series = df["A"] == 4
+        series = series.reindex(df.index[::-1])
+        df[series] = 1
+        values[values[:, 0] == 4] = 1
+        tm.assert_almost_equal(df.values, values)
+
+        df[df > 0] = 5
+        values[values > 0] = 5
+        tm.assert_almost_equal(df.values, values)
+
+        df[df == 5] = 0
+        values[values == 5] = 0
+        tm.assert_almost_equal(df.values, values)
+
+        # a df that needs alignment first
+        df[df[:-1] < 0] = 2
+        np.putmask(values[:-1], values[:-1] < 0, 2)
+        tm.assert_almost_equal(df.values, values)
+
+        # indexed with same shape but rows-reversed df
+        df[df[::-1] == 2] = 3
+        values[values == 2] = 3
+        tm.assert_almost_equal(df.values, values)
+
+        msg = "Must pass DataFrame or 2-d ndarray with boolean values only"
+        with pytest.raises(TypeError, match=msg):
+            df[df * 0] = 2
+
+        # index with DataFrame
+        df_orig = df.copy()
+        mask = df > np.abs(df)
+        df[df > np.abs(df)] = np.nan
+        values = df_orig.values.copy()
+        values[mask.values] = np.nan
+        expected = DataFrame(values, index=df_orig.index, columns=df_orig.columns)
+        tm.assert_frame_equal(df, expected)
+
+        # set from DataFrame
+        df[df > np.abs(df)] = df * 2
+        np.putmask(values, mask.values, df.values * 2)
+        expected = DataFrame(values, index=df_orig.index, columns=df_orig.columns)
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_cast(self, float_frame):
+        float_frame["D"] = float_frame["D"].astype("i8")
+        assert float_frame["D"].dtype == np.int64
+
+        # #669, should not cast?
+        # this is now set to int64, which means a replacement of the column to
+        # the value dtype (and nothing to do with the existing dtype)
+        float_frame["B"] = 0
+        assert float_frame["B"].dtype == np.int64
+
+        # cast if pass array of course
+        float_frame["B"] = np.arange(len(float_frame))
+        assert issubclass(float_frame["B"].dtype.type, np.integer)
+
+        float_frame["foo"] = "bar"
+        float_frame["foo"] = 0
+        assert float_frame["foo"].dtype == np.int64
+
+        float_frame["foo"] = "bar"
+        float_frame["foo"] = 2.5
+        assert float_frame["foo"].dtype == np.float64
+
+        float_frame["something"] = 0
+        assert float_frame["something"].dtype == np.int64
+        float_frame["something"] = 2
+        assert float_frame["something"].dtype == np.int64
+        float_frame["something"] = 2.5
+        assert float_frame["something"].dtype == np.float64
+
+    def test_setitem_corner(self, float_frame, using_infer_string):
+        # corner case
+        df = DataFrame({"B": [1.0, 2.0, 3.0], "C": ["a", "b", "c"]}, index=np.arange(3))
+        del df["B"]
+        df["B"] = [1.0, 2.0, 3.0]
+        assert "B" in df
+        assert len(df.columns) == 2
+
+        df["A"] = "beginning"
+        df["E"] = "foo"
+        df["D"] = "bar"
+        df[datetime.now()] = "date"
+        df[datetime.now()] = 5.0
+
+        # what to do when empty frame with index
+        dm = DataFrame(index=float_frame.index)
+        dm["A"] = "foo"
+        dm["B"] = "bar"
+        assert len(dm.columns) == 2
+        assert dm.values.dtype == np.object_
+
+        # upcast
+        dm["C"] = 1
+        assert dm["C"].dtype == np.int64
+
+        dm["E"] = 1.0
+        assert dm["E"].dtype == np.float64
+
+        # set existing column
+        dm["A"] = "bar"
+        assert "bar" == dm["A"].iloc[0]
+
+        dm = DataFrame(index=np.arange(3))
+        dm["A"] = 1
+        dm["foo"] = "bar"
+        del dm["foo"]
+        dm["foo"] = "bar"
+        if using_infer_string:
+            assert dm["foo"].dtype == "str"
+        else:
+            assert dm["foo"].dtype == np.object_
+
+        dm["coercible"] = ["1", "2", "3"]
+        if using_infer_string:
+            assert dm["coercible"].dtype == "str"
+        else:
+            assert dm["coercible"].dtype == np.object_
+
+    def test_setitem_corner2(self):
+        data = {
+            "title": ["foobar", "bar", "foobar"] + ["foobar"] * 17,
+            "cruft": np.random.default_rng(2).random(20),
+        }
+
+        df = DataFrame(data)
+        ix = df[df["title"] == "bar"].index
+
+        df.loc[ix, ["title"]] = "foobar"
+        df.loc[ix, ["cruft"]] = 0
+
+        assert df.loc[1, "title"] == "foobar"
+        assert df.loc[1, "cruft"] == 0
+
+    def test_setitem_ambig(self, using_infer_string):
+        # Difficulties with mixed-type data
+        # Created as float type
+        dm = DataFrame(index=range(3), columns=range(3))
+
+        coercable_series = Series([Decimal(1) for _ in range(3)], index=range(3))
+        uncoercable_series = Series(["foo", "bzr", "baz"], index=range(3))
+
+        dm[0] = np.ones(3)
+        assert len(dm.columns) == 3
+
+        dm[1] = coercable_series
+        assert len(dm.columns) == 3
+
+        dm[2] = uncoercable_series
+        assert len(dm.columns) == 3
+        if using_infer_string:
+            assert dm[2].dtype == "str"
+        else:
+            assert dm[2].dtype == np.object_
+
+    def test_setitem_None(self, float_frame):
+        # GH #766
+        float_frame[None] = float_frame["A"]
+        tm.assert_series_equal(
+            float_frame.iloc[:, -1], float_frame["A"], check_names=False
+        )
+        tm.assert_series_equal(
+            float_frame.loc[:, None], float_frame["A"], check_names=False
+        )
+        tm.assert_series_equal(float_frame[None], float_frame["A"], check_names=False)
+
+    def test_loc_setitem_boolean_mask_allfalse(self):
+        # GH 9596
+        df = DataFrame(
+            {"a": ["1", "2", "3"], "b": ["11", "22", "33"], "c": ["111", "222", "333"]}
+        )
+
+        result = df.copy()
+        result.loc[result.b.isna(), "a"] = result.a.copy()
+        tm.assert_frame_equal(result, df)
+
+    def test_getitem_slice_empty(self):
+        df = DataFrame([[1]], columns=MultiIndex.from_product([["A"], ["a"]]))
+        result = df[:]
+
+        expected = DataFrame([[1]], columns=MultiIndex.from_product([["A"], ["a"]]))
+
+        tm.assert_frame_equal(result, expected)
+        # Ensure df[:] returns a view of df, not the same object
+        assert result is not df
+
+    def test_getitem_fancy_slice_integers_step(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+
+        # this is OK
+        df.iloc[:8:2]
+        df.iloc[:8:2] = np.nan
+        assert isna(df.iloc[:8:2]).values.all()
+
+    def test_getitem_setitem_integer_slice_keyerrors(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 5)), index=range(0, 20, 2)
+        )
+
+        # this is OK
+        cp = df.copy()
+        cp.iloc[4:10] = 0
+        assert (cp.iloc[4:10] == 0).values.all()
+
+        # so is this
+        cp = df.copy()
+        cp.iloc[3:11] = 0
+        assert (cp.iloc[3:11] == 0).values.all()
+
+        result = df.iloc[2:6]
+        result2 = df.loc[3:11]
+        expected = df.reindex([4, 6, 8, 10])
+
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        # non-monotonic, raise KeyError
+        df2 = df.iloc[list(range(5)) + list(range(5, 10))[::-1]]
+        with pytest.raises(KeyError, match=r"^3$"):
+            df2.loc[3:11]
+        with pytest.raises(KeyError, match=r"^3$"):
+            df2.loc[3:11] = 0
+
+    def test_fancy_getitem_slice_mixed(self, float_frame, float_string_frame):
+        sliced = float_string_frame.iloc[:, -3:]
+        assert sliced["D"].dtype == np.float64
+
+        # get view with single block
+        # setting it triggers setting with copy
+        original = float_frame.copy()
+        sliced = float_frame.iloc[:, -3:]
+
+        assert np.shares_memory(sliced["C"]._values, float_frame["C"]._values)
+
+        sliced.loc[:, "C"] = 4.0
+        tm.assert_frame_equal(float_frame, original)
+
+    def test_getitem_setitem_non_ix_labels(self):
+        df = DataFrame(range(20), index=date_range("2020-01-01", periods=20))
+
+        start, end = df.index[[5, 10]]
+
+        result = df.loc[start:end]
+        result2 = df[start:end]
+        expected = df[5:11]
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        result = df.copy()
+        result.loc[start:end] = 0
+        result2 = df.copy()
+        result2[start:end] = 0
+        expected = df.copy()
+        expected[5:11] = 0
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_ix_multi_take(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 2)))
+        rs = df.loc[df.index == 0, :]
+        xp = df.reindex([0])
+        tm.assert_frame_equal(rs, xp)
+
+        # GH#1321
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 2)))
+        rs = df.loc[df.index == 0, df.columns == 1]
+        xp = df.reindex(index=[0], columns=[1])
+        tm.assert_frame_equal(rs, xp)
+
+    def test_getitem_fancy_scalar(self, float_frame):
+        f = float_frame
+        ix = f.loc
+
+        # individual value
+        for col in f.columns:
+            ts = f[col]
+            for idx in f.index[::5]:
+                assert ix[idx, col] == ts[idx]
+
+    def test_setitem_fancy_scalar(self, float_frame):
+        f = float_frame
+        expected = float_frame.copy()
+        ix = f.loc
+
+        # individual value
+        for j, col in enumerate(f.columns):
+            f[col]
+            for idx in f.index[::5]:
+                i = f.index.get_loc(idx)
+                val = np.random.default_rng(2).standard_normal()
+                expected.iloc[i, j] = val
+
+                ix[idx, col] = val
+                tm.assert_frame_equal(f, expected)
+
+    def test_getitem_fancy_boolean(self, float_frame):
+        f = float_frame
+        ix = f.loc
+
+        expected = f.reindex(columns=["B", "D"])
+        result = ix[:, [False, True, False, True]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = f.reindex(index=f.index[5:10], columns=["B", "D"])
+        result = ix[f.index[5:10], [False, True, False, True]]
+        tm.assert_frame_equal(result, expected)
+
+        boolvec = f.index > f.index[7]
+        expected = f.reindex(index=f.index[boolvec])
+        result = ix[boolvec]
+        tm.assert_frame_equal(result, expected)
+        result = ix[boolvec, :]
+        tm.assert_frame_equal(result, expected)
+
+        result = ix[boolvec, f.columns[2:]]
+        expected = f.reindex(index=f.index[boolvec], columns=["C", "D"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_fancy_boolean(self, float_frame):
+        # from 2d, set with booleans
+        frame = float_frame.copy()
+        expected = float_frame.copy()
+        values = expected.values.copy()
+
+        mask = frame["A"] > 0
+        frame.loc[mask] = 0.0
+        values[mask.values] = 0.0
+        expected = DataFrame(values, index=expected.index, columns=expected.columns)
+        tm.assert_frame_equal(frame, expected)
+
+        frame = float_frame.copy()
+        expected = float_frame.copy()
+        values = expected.values.copy()
+        frame.loc[mask, ["A", "B"]] = 0.0
+        values[mask.values, :2] = 0.0
+        expected = DataFrame(values, index=expected.index, columns=expected.columns)
+        tm.assert_frame_equal(frame, expected)
+
+    def test_getitem_fancy_ints(self, float_frame):
+        result = float_frame.iloc[[1, 4, 7]]
+        expected = float_frame.loc[float_frame.index[[1, 4, 7]]]
+        tm.assert_frame_equal(result, expected)
+
+        result = float_frame.iloc[:, [2, 0, 1]]
+        expected = float_frame.loc[:, float_frame.columns[[2, 0, 1]]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_setitem_boolean_misaligned(self, float_frame):
+        # boolean index misaligned labels
+        mask = float_frame["A"][::-1] > 1
+
+        result = float_frame.loc[mask]
+        expected = float_frame.loc[mask[::-1]]
+        tm.assert_frame_equal(result, expected)
+
+        cp = float_frame.copy()
+        expected = float_frame.copy()
+        cp.loc[mask] = 0
+        expected.loc[mask] = 0
+        tm.assert_frame_equal(cp, expected)
+
+    def test_getitem_setitem_boolean_multi(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 2)))
+
+        # get
+        k1 = np.array([True, False, True])
+        k2 = np.array([False, True])
+        result = df.loc[k1, k2]
+        expected = df.loc[[0, 2], [1]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.copy()
+        df.loc[np.array([True, False, True]), np.array([False, True])] = 5
+        expected.loc[[0, 2], [1]] = 5
+        tm.assert_frame_equal(df, expected)
+
+    def test_getitem_float_label_positional(self):
+        # GH 53338
+        index = Index([1.5, 2])
+        df = DataFrame(range(2), index=index)
+        result = df[1:2]
+        expected = DataFrame([1], index=[2.0])
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_setitem_float_labels(self):
+        index = Index([1.5, 2, 3, 4, 5])
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)), index=index)
+
+        result = df.loc[1.5:4]
+        expected = df.reindex([1.5, 2, 3, 4])
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 4
+
+        result = df.loc[4:5]
+        expected = df.reindex([4, 5])  # reindex with int
+        tm.assert_frame_equal(result, expected, check_index_type=False)
+        assert len(result) == 2
+
+        result = df.loc[4:5]
+        expected = df.reindex([4.0, 5.0])  # reindex with float
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 2
+
+        # loc_float changes this to work properly
+        result = df.loc[1:2]
+        expected = df.iloc[0:2]
+        tm.assert_frame_equal(result, expected)
+
+        # #2727
+        index = Index([1.0, 2.5, 3.5, 4.5, 5.0])
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)), index=index)
+
+        # positional slicing only via iloc!
+        msg = (
+            "cannot do positional indexing on Index with "
+            r"these indexers \[1.0\] of type float"
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.iloc[1.0:5]
+
+        result = df.iloc[4:5]
+        expected = df.reindex([5.0])
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 1
+
+        cp = df.copy()
+
+        with pytest.raises(TypeError, match=_slice_msg):
+            cp.iloc[1.0:5] = 0
+
+        with pytest.raises(TypeError, match=msg):
+            result = cp.iloc[1.0:5] == 0
+
+        assert result.values.all()
+        assert (cp.iloc[0:1] == df.iloc[0:1]).values.all()
+
+        cp = df.copy()
+        cp.iloc[4:5] = 0
+        assert (cp.iloc[4:5] == 0).values.all()
+        assert (cp.iloc[0:4] == df.iloc[0:4]).values.all()
+
+        # float slicing
+        result = df.loc[1.0:5]
+        expected = df
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 5
+
+        result = df.loc[1.1:5]
+        expected = df.reindex([2.5, 3.5, 4.5, 5.0])
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 4
+
+        result = df.loc[4.51:5]
+        expected = df.reindex([5.0])
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 1
+
+        result = df.loc[1.0:5.0]
+        expected = df.reindex([1.0, 2.5, 3.5, 4.5, 5.0])
+        tm.assert_frame_equal(result, expected)
+        assert len(result) == 5
+
+        cp = df.copy()
+        cp.loc[1.0:5.0] = 0
+        result = cp.loc[1.0:5.0]
+        assert (result == 0).values.all()
+
+    def test_setitem_single_column_mixed_datetime(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=["a", "b", "c", "d", "e"],
+            columns=["foo", "bar", "baz"],
+        )
+
+        df["timestamp"] = Timestamp("20010102")
+
+        # check our dtypes
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 3 + [np.dtype("datetime64[s]")],
+            index=["foo", "bar", "baz", "timestamp"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # GH#16674 iNaT is treated as an integer when given by the user
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc["b", "timestamp"] = iNaT
+
+        # allow this syntax (as of GH#3216)
+        df.loc["c", "timestamp"] = np.nan
+        assert isna(df.loc["c", "timestamp"])
+
+        # allow this syntax
+        df.loc["d", :] = np.nan
+        assert not isna(df.loc["c", :]).all()
+
+    def test_setitem_mixed_datetime(self):
+        # GH 9336
+        df = DataFrame(0, columns=list("ab"), index=range(6))
+        df["b"] = pd.NaT
+        df.loc[0, "b"] = datetime(2012, 1, 1)
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[1, "b"] = 1
+
+    def test_setitem_frame_float(self, float_frame):
+        piece = float_frame.loc[float_frame.index[:2], ["A", "B"]]
+        float_frame.loc[float_frame.index[-2] :, ["A", "B"]] = piece.values
+        result = float_frame.loc[float_frame.index[-2:], ["A", "B"]].values
+        expected = piece.values
+        tm.assert_almost_equal(result, expected)
+
+    def test_setitem_frame_mixed(self, float_string_frame):
+        # GH 3216
+
+        # already aligned
+        f = float_string_frame.copy()
+        piece = DataFrame(
+            [[1.0, 2.0], [3.0, 4.0]], index=f.index[0:2], columns=["A", "B"]
+        )
+        key = (f.index[slice(None, 2)], ["A", "B"])
+        f.loc[key] = piece
+        tm.assert_almost_equal(f.loc[f.index[0:2], ["A", "B"]].values, piece.values)
+
+    def test_setitem_frame_mixed_rows_unaligned(self, float_string_frame):
+        # GH#3216 rows unaligned
+        f = float_string_frame.copy()
+        piece = DataFrame(
+            [[1.0, 2.0], [3.0, 4.0], [5.0, 6.0], [7.0, 8.0]],
+            index=list(f.index[0:2]) + ["foo", "bar"],
+            columns=["A", "B"],
+        )
+        key = (f.index[slice(None, 2)], ["A", "B"])
+        f.loc[key] = piece
+        tm.assert_almost_equal(
+            f.loc[f.index[0:2:], ["A", "B"]].values, piece.values[0:2]
+        )
+
+    def test_setitem_frame_mixed_key_unaligned(self, float_string_frame):
+        # GH#3216 key is unaligned with values
+        f = float_string_frame.copy()
+        piece = f.loc[f.index[:2], ["A"]]
+        piece.index = f.index[-2:]
+        key = (f.index[slice(-2, None)], ["A", "B"])
+        f.loc[key] = piece
+        piece["B"] = np.nan
+        tm.assert_almost_equal(f.loc[f.index[-2:], ["A", "B"]].values, piece.values)
+
+    def test_setitem_frame_mixed_ndarray(self, float_string_frame):
+        # GH#3216 ndarray
+        f = float_string_frame.copy()
+        piece = float_string_frame.loc[f.index[:2], ["A", "B"]]
+        key = (f.index[slice(-2, None)], ["A", "B"])
+        f.loc[key] = piece.values
+        tm.assert_almost_equal(f.loc[f.index[-2:], ["A", "B"]].values, piece.values)
+
+    def test_setitem_frame_upcast(self):
+        # needs upcasting
+        df = DataFrame([[1, 2, "foo"], [3, 4, "bar"]], columns=["A", "B", "C"])
+        df2 = df.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            df2.loc[:, ["A", "B"]] = df.loc[:, ["A", "B"]] + 0.5
+        # Manually upcast so we can add .5
+        df = df.astype({"A": "float64", "B": "float64"})
+        df2 = df2.astype({"A": "float64", "B": "float64"})
+        df2.loc[:, ["A", "B"]] = df.loc[:, ["A", "B"]] + 0.5
+        expected = df.reindex(columns=["A", "B"])
+        expected += 0.5
+        expected["C"] = df["C"]
+        tm.assert_frame_equal(df2, expected)
+
+    def test_setitem_frame_align(self, float_frame):
+        piece = float_frame.loc[float_frame.index[:2], ["A", "B"]]
+        piece.index = float_frame.index[-2:]
+        piece.columns = ["A", "B"]
+        float_frame.loc[float_frame.index[-2:], ["A", "B"]] = piece
+        result = float_frame.loc[float_frame.index[-2:], ["A", "B"]].values
+        expected = piece.values
+        tm.assert_almost_equal(result, expected)
+
+    def test_getitem_setitem_ix_duplicates(self):
+        # #1201
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=["foo", "foo", "bar", "baz", "bar"],
+        )
+
+        result = df.loc["foo"]
+        expected = df[:2]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc["bar"]
+        expected = df.iloc[[2, 4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc["baz"]
+        expected = df.iloc[3]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_ix_boolean_duplicates_multiple(self):
+        # #1201
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=["foo", "foo", "bar", "baz", "bar"],
+        )
+
+        result = df.loc[["bar"]]
+        exp = df.iloc[[2, 4]]
+        tm.assert_frame_equal(result, exp)
+
+        result = df.loc[df[1] > 0]
+        exp = df[df[1] > 0]
+        tm.assert_frame_equal(result, exp)
+
+        result = df.loc[df[0] > 0]
+        exp = df[df[0] > 0]
+        tm.assert_frame_equal(result, exp)
+
+    @pytest.mark.parametrize("bool_value", [True, False])
+    def test_getitem_setitem_ix_bool_keyerror(self, bool_value):
+        # #2199
+        df = DataFrame({"a": [1, 2, 3]})
+        message = f"{bool_value}: boolean label can not be used without a boolean index"
+        with pytest.raises(KeyError, match=message):
+            df.loc[bool_value]
+
+        msg = "cannot use a single bool to index into setitem"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[bool_value] = 0
+
+    # TODO: rename?  remove?
+    def test_single_element_ix_dont_upcast(self, float_frame):
+        float_frame["E"] = 1
+        assert issubclass(float_frame["E"].dtype.type, (int, np.integer))
+
+        result = float_frame.loc[float_frame.index[5], "E"]
+        assert is_integer(result)
+
+        # GH 11617
+        df = DataFrame({"a": [1.23]})
+        df["b"] = 666
+
+        result = df.loc[0, "b"]
+        assert is_integer(result)
+
+        expected = Series([666], index=range(1), name="b")
+        result = df.loc[[0], "b"]
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_callable_tuple_return_value_raises(self):
+        # GH53769: Enforced pandas 3.0
+        df = DataFrame(np.arange(40).reshape(10, 4), index=range(0, 20, 2))
+        msg = "Returning a tuple from"
+        with pytest.raises(ValueError, match=msg):
+            df.iloc[lambda _: (0,)]
+        with pytest.raises(ValueError, match=msg):
+            df.iloc[lambda _: (0,)] = 1
+
+    def test_iloc_row(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), index=range(0, 20, 2)
+        )
+
+        result = df.iloc[1]
+        exp = df.loc[2]
+        tm.assert_series_equal(result, exp)
+
+        result = df.iloc[2]
+        exp = df.loc[4]
+        tm.assert_series_equal(result, exp)
+
+        # slice
+        result = df.iloc[slice(4, 8)]
+        expected = df.loc[8:14]
+        tm.assert_frame_equal(result, expected)
+
+        # list of integers
+        result = df.iloc[[1, 2, 4, 6]]
+        expected = df.reindex(df.index[[1, 2, 4, 6]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_row_slice_view(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), index=range(0, 20, 2)
+        )
+        original = df.copy()
+
+        # verify slice is view
+        # setting it makes it raise/warn
+        subset = df.iloc[slice(4, 8)]
+
+        assert np.shares_memory(df[2], subset[2])
+
+        exp_col = original[2].copy()
+        subset.loc[:, 2] = 0.0
+        tm.assert_series_equal(df[2], exp_col)
+
+    def test_iloc_col(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 10)), columns=range(0, 20, 2)
+        )
+
+        result = df.iloc[:, 1]
+        exp = df.loc[:, 2]
+        tm.assert_series_equal(result, exp)
+
+        result = df.iloc[:, 2]
+        exp = df.loc[:, 4]
+        tm.assert_series_equal(result, exp)
+
+        # slice
+        result = df.iloc[:, slice(4, 8)]
+        expected = df.loc[:, 8:14]
+        tm.assert_frame_equal(result, expected)
+
+        # list of integers
+        result = df.iloc[:, [1, 2, 4, 6]]
+        expected = df.reindex(columns=df.columns[[1, 2, 4, 6]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_col_slice_view(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 10)), columns=range(0, 20, 2)
+        )
+        original = df.copy()
+        subset = df.iloc[:, slice(4, 8)]
+
+        # verify slice is view
+        assert np.shares_memory(df[8]._values, subset[8]._values)
+        subset[8] = 0.0
+        # subset changed
+        assert (subset[8] == 0).all()
+        # but df itself did not change (setitem replaces full column)
+        tm.assert_frame_equal(df, original)
+
+    def test_loc_duplicates(self):
+        # gh-17105
+
+        # insert a duplicate element to the index
+        trange = date_range(
+            start=Timestamp(year=2017, month=1, day=1),
+            end=Timestamp(year=2017, month=1, day=5),
+        )
+
+        trange = trange.insert(loc=5, item=Timestamp(year=2017, month=1, day=5))
+
+        df = DataFrame(0, index=trange, columns=["A", "B"])
+        bool_idx = np.array([False, False, False, False, False, True])
+
+        # assignment
+        df.loc[trange[bool_idx], "A"] = 6
+
+        expected = DataFrame(
+            {"A": [0, 0, 0, 0, 6, 6], "B": [0, 0, 0, 0, 0, 0]}, index=trange
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # in-place
+        df = DataFrame(0, index=trange, columns=["A", "B"])
+        df.loc[trange[bool_idx], "A"] += 6
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_with_unaligned_tz_aware_datetime_column(self):
+        # GH 12981
+        # Assignment of unaligned offset-aware datetime series.
+        # Make sure timezone isn't lost
+        column = Series(date_range("2015-01-01", periods=3, tz="utc"), name="dates")
+        df = DataFrame({"dates": column})
+        df["dates"] = column[[1, 0, 2]]
+        tm.assert_series_equal(df["dates"], column)
+
+        df = DataFrame({"dates": column})
+        df.loc[[0, 1, 2], "dates"] = column[[1, 0, 2]]
+        tm.assert_series_equal(df["dates"], column)
+
+    def test_loc_setitem_datetimelike_with_inference(self):
+        # GH 7592
+        # assignment of timedeltas with NaT
+
+        one_hour = timedelta(hours=1)
+        df = DataFrame(index=date_range("20130101", periods=4))
+        df["A"] = np.array([1 * one_hour] * 4, dtype="m8[ns]")
+        df.loc[:, "B"] = np.array([2 * one_hour] * 4, dtype="m8[ns]")
+        df.loc[df.index[:3], "C"] = np.array([3 * one_hour] * 3, dtype="m8[ns]")
+        df.loc[:, "D"] = np.array([4 * one_hour] * 4, dtype="m8[ns]")
+        df.loc[df.index[:3], "E"] = np.array([5 * one_hour] * 3, dtype="m8[ns]")
+        df["F"] = np.timedelta64("NaT")
+        df.loc[df.index[:-1], "F"] = np.array([6 * one_hour] * 3, dtype="m8[ns]")
+        df.loc[df.index[-3] :, "G"] = date_range("20130101", periods=3)
+        df["H"] = np.datetime64("NaT")
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("timedelta64[ns]")] * 6 + [np.dtype("datetime64[ns]")] * 2,
+            index=list("ABCDEFGH"),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_boolean_indexing_mixed(self):
+        df = DataFrame(
+            {
+                0: {35: np.nan, 40: np.nan, 43: np.nan, 49: np.nan, 50: np.nan},
+                1: {
+                    35: np.nan,
+                    40: 0.32632316859446198,
+                    43: np.nan,
+                    49: 0.32632316859446198,
+                    50: 0.39114724480578139,
+                },
+                2: {
+                    35: np.nan,
+                    40: np.nan,
+                    43: 0.29012581014105987,
+                    49: np.nan,
+                    50: np.nan,
+                },
+                3: {35: np.nan, 40: np.nan, 43: np.nan, 49: np.nan, 50: np.nan},
+                4: {
+                    35: 0.34215328467153283,
+                    40: np.nan,
+                    43: np.nan,
+                    49: np.nan,
+                    50: np.nan,
+                },
+                "y": {35: 0, 40: 0, 43: 0, 49: 0, 50: 1},
+            }
+        )
+
+        # mixed int/float ok
+        df2 = df.copy()
+        df2[df2 > 0.3] = 1
+        expected = df.copy()
+        expected.loc[40, 1] = 1
+        expected.loc[49, 1] = 1
+        expected.loc[50, 1] = 1
+        expected.loc[35, 4] = 1
+        tm.assert_frame_equal(df2, expected)
+
+        df["foo"] = "test"
+        msg = "not supported between instances|unorderable types|Invalid comparison"
+
+        with pytest.raises(TypeError, match=msg):
+            df[df > 0.3] = 1
+
+    def test_type_error_multiindex(self):
+        # See gh-12218
+        mi = MultiIndex.from_product([["x", "y"], [0, 1]], names=[None, "c"])
+        dg = DataFrame(
+            [[1, 1, 2, 2], [3, 3, 4, 4]], columns=mi, index=Index(range(2), name="i")
+        )
+        with pytest.raises(InvalidIndexError, match="slice"):
+            dg[:, 0]
+
+        index = Index(range(2), name="i")
+        columns = MultiIndex(
+            levels=[["x", "y"], [0, 1]], codes=[[0, 1], [0, 0]], names=[None, "c"]
+        )
+        expected = DataFrame([[1, 2], [3, 4]], columns=columns, index=index)
+
+        result = dg.loc[:, (slice(None), 0)]
+        tm.assert_frame_equal(result, expected)
+
+        name = ("x", 0)
+        index = Index(range(2), name="i")
+        expected = Series([1, 3], index=index, name=name)
+
+        result = dg["x", 0]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_interval_index_partial_indexing(self):
+        # GH#36490
+        df = DataFrame(
+            np.ones((3, 4)), columns=pd.IntervalIndex.from_breaks(np.arange(5))
+        )
+
+        expected = df.iloc[:, 0]
+
+        res = df[0.5]
+        tm.assert_series_equal(res, expected)
+
+        res = df.loc[:, 0.5]
+        tm.assert_series_equal(res, expected)
+
+    def test_setitem_array_as_cell_value(self):
+        # GH#43422
+        df = DataFrame(columns=["a", "b"], dtype=object)
+        df.loc[0] = {"a": np.zeros((2,)), "b": np.zeros((2, 2))}
+        expected = DataFrame({"a": [np.zeros((2,))], "b": [np.zeros((2, 2))]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_nullable_2d_values(self):
+        df = DataFrame({"A": [1, 2, 3]}, dtype="Int64")
+        orig = df.copy()
+
+        df.loc[:] = df.values[:, ::-1]
+        tm.assert_frame_equal(df, orig)
+
+        df.loc[:] = pd.core.arrays.NumpyExtensionArray(df.values[:, ::-1])
+        tm.assert_frame_equal(df, orig)
+
+        df.iloc[:] = df.iloc[:, :].copy()
+        tm.assert_frame_equal(df, orig)
+
+    def test_getitem_segfault_with_empty_like_object(self):
+        # GH#46848
+        df = DataFrame(np.empty((1, 1), dtype=object))
+        df[0] = np.empty_like(df[0])
+        # this produces the segfault
+        df[[0]]
+
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+    @pytest.mark.parametrize(
+        "null", [pd.NaT, pd.NaT.to_numpy("M8[ns]"), pd.NaT.to_numpy("m8[ns]")]
+    )
+    def test_setting_mismatched_na_into_nullable_fails(
+        self, null, any_numeric_ea_dtype
+    ):
+        # GH#44514 don't cast mismatched nulls to pd.NA
+        df = DataFrame({"A": [1, 2, 3]}, dtype=any_numeric_ea_dtype)
+        ser = df["A"].copy()
+        arr = ser._values
+
+        msg = "|".join(
+            [
+                r"timedelta64\[ns\] cannot be converted to (Floating|Integer)Dtype",
+                r"datetime64\[ns\] cannot be converted to (Floating|Integer)Dtype",
+                "'values' contains non-numeric NA",
+                r"Invalid value '.*' for dtype '(U?Int|Float)\d{1,2}'",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            arr[0] = null
+
+        with pytest.raises(TypeError, match=msg):
+            arr[:2] = [null, null]
+
+        with pytest.raises(TypeError, match=msg):
+            ser[0] = null
+
+        with pytest.raises(TypeError, match=msg):
+            ser[:2] = [null, null]
+
+        with pytest.raises(TypeError, match=msg):
+            ser.iloc[0] = null
+
+        with pytest.raises(TypeError, match=msg):
+            ser.iloc[:2] = [null, null]
+
+        with pytest.raises(TypeError, match=msg):
+            df.iloc[0, 0] = null
+
+        with pytest.raises(TypeError, match=msg):
+            df.iloc[:2, 0] = [null, null]
+
+        # Multi-Block
+        df2 = df.copy()
+        df2["B"] = ser.copy()
+        with pytest.raises(TypeError, match=msg):
+            df2.iloc[0, 0] = null
+
+        with pytest.raises(TypeError, match=msg):
+            df2.iloc[:2, 0] = [null, null]
+
+    def test_loc_expand_empty_frame_keep_index_name(self):
+        # GH#45621
+        df = DataFrame(columns=["b"], index=Index([], name="a"))
+        df.loc[0] = 1
+        expected = DataFrame({"b": [1]}, index=Index([0], name="a"))
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_expand_empty_frame_keep_midx_names(self):
+        # GH#46317
+        df = DataFrame(
+            columns=["d"], index=MultiIndex.from_tuples([], names=["a", "b", "c"])
+        )
+        df.loc[(1, 2, 3)] = "foo"
+        expected = DataFrame(
+            {"d": ["foo"]},
+            index=MultiIndex.from_tuples([(1, 2, 3)], names=["a", "b", "c"]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "val, idxr",
+        [
+            ("x", "a"),
+            ("x", ["a"]),
+            (1, "a"),
+            (1, ["a"]),
+        ],
+    )
+    def test_loc_setitem_rhs_frame(self, idxr, val):
+        # GH#47578
+        df = DataFrame({"a": [1, 2]})
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, idxr] = DataFrame({"a": [val, 11]}, index=[1, 2])
+
+    def test_iloc_setitem_enlarge_no_warning(self):
+        # GH#47381
+        df = DataFrame(columns=["a", "b"])
+        expected = df.copy()
+        view = df[:]
+        df.iloc[:, 0] = np.array([1, 2], dtype=np.float64)
+        tm.assert_frame_equal(view, expected)
+
+    def test_loc_internals_not_updated_correctly(self):
+        # GH#47867 all steps are necessary to reproduce the initial bug
+        df = DataFrame(
+            {"bool_col": True, "a": 1, "b": 2.5},
+            index=MultiIndex.from_arrays([[1, 2], [1, 2]], names=["idx1", "idx2"]),
+        )
+        idx = [(1, 1)]
+
+        df["c"] = 3
+        df.loc[idx, "c"] = 0
+
+        df.loc[idx, "c"]
+        df.loc[idx, ["a", "b"]]
+
+        df.loc[idx, "c"] = 15
+        result = df.loc[idx, "c"]
+        expected = df = Series(
+            15,
+            index=MultiIndex.from_arrays([[1], [1]], names=["idx1", "idx2"]),
+            name="c",
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("val", [None, [None], pd.NA, [pd.NA]])
+    def test_iloc_setitem_string_list_na(self, val):
+        # GH#45469
+        df = DataFrame({"a": ["a", "b", "c"]}, dtype="string")
+        df.iloc[[0], :] = val
+        expected = DataFrame({"a": [pd.NA, "b", "c"]}, dtype="string")
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("val", [None, pd.NA])
+    def test_iloc_setitem_string_na(self, val):
+        # GH#45469
+        df = DataFrame({"a": ["a", "b", "c"]}, dtype="string")
+        df.iloc[0, :] = val
+        expected = DataFrame({"a": [pd.NA, "b", "c"]}, dtype="string")
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("func", [list, Series, np.array])
+    def test_iloc_setitem_ea_null_slice_length_one_list(self, func):
+        # GH#48016
+        df = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+        df.iloc[:, func([0])] = 5
+        expected = DataFrame({"a": [5, 5, 5]}, dtype="Int64")
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_named_tuple_for_midx(self):
+        # GH#48124
+        df = DataFrame(
+            index=MultiIndex.from_product(
+                [["A", "B"], ["a", "b", "c"]], names=["first", "second"]
+            )
+        )
+        indexer_tuple = namedtuple("Indexer", df.index.names)
+        idxr = indexer_tuple(first="A", second=["a", "b"])
+        result = df.loc[idxr, :]
+        expected = DataFrame(
+            index=MultiIndex.from_tuples(
+                [("A", "a"), ("A", "b")], names=["first", "second"]
+            )
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("indexer", [["a"], "a"])
+    @pytest.mark.parametrize("col", [{}, {"b": 1}])
+    def test_set_2d_casting_date_to_int(self, col, indexer):
+        # GH#49159
+        df = DataFrame(
+            {"a": [Timestamp("2022-12-29"), Timestamp("2022-12-30")], **col},
+        )
+        df.loc[[1], indexer] = df["a"] + pd.Timedelta(days=1)
+        expected = DataFrame(
+            {"a": [Timestamp("2022-12-29"), Timestamp("2022-12-31")], **col},
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("col", [{}, {"name": "a"}])
+    def test_loc_setitem_reordering_with_all_true_indexer(self, col):
+        # GH#48701
+        n = 17
+        df = DataFrame({**col, "x": range(n), "y": range(n)})
+        expected = df.copy()
+        df.loc[n * [True], ["x", "y"]] = df[["x", "y"]]
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_rhs_empty_warning(self):
+        # GH48480
+        df = DataFrame(columns=["a", "b"])
+        expected = df.copy()
+        rhs = DataFrame(columns=["a"])
+        with tm.assert_produces_warning(None):
+            df.loc[:, "a"] = rhs
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_ea_series_indexer(self):
+        # GH#49521
+        df = DataFrame([[0, 1, 2, 3, 4], [5, 6, 7, 8, 9]])
+        indexer = Series([0, 1], dtype="Int64")
+        row_indexer = Series([1], dtype="Int64")
+        result = df.iloc[row_indexer, indexer]
+        expected = DataFrame([[5, 6]], index=range(1, 2))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[row_indexer.values, indexer.values]
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_ea_series_indexer_with_na(self):
+        # GH#49521
+        df = DataFrame([[0, 1, 2, 3, 4], [5, 6, 7, 8, 9]])
+        indexer = Series([0, pd.NA], dtype="Int64")
+        msg = "cannot convert"
+        with pytest.raises(ValueError, match=msg):
+            df.iloc[:, indexer]
+        with pytest.raises(ValueError, match=msg):
+            df.iloc[:, indexer.values]
+
+    @pytest.mark.parametrize("indexer", [True, (True,)])
+    @pytest.mark.parametrize("dtype", [bool, "boolean"])
+    def test_loc_bool_multiindex(self, performance_warning, dtype, indexer):
+        # GH#47687
+        midx = MultiIndex.from_arrays(
+            [
+                Series([True, True, False, False], dtype=dtype),
+                Series([True, False, True, False], dtype=dtype),
+            ],
+            names=["a", "b"],
+        )
+        df = DataFrame({"c": [1, 2, 3, 4]}, index=midx)
+        with tm.maybe_produces_warning(performance_warning, isinstance(indexer, tuple)):
+            result = df.loc[indexer]
+        expected = DataFrame(
+            {"c": [1, 2]}, index=Index([True, False], name="b", dtype=dtype)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("utc", [False, True])
+    @pytest.mark.parametrize("indexer", ["date", ["date"]])
+    def test_loc_datetime_assignment_dtype_does_not_change(self, utc, indexer):
+        # GH#49837
+        df = DataFrame(
+            {
+                "date": to_datetime(
+                    [datetime(2022, 1, 20), datetime(2022, 1, 22)], utc=utc
+                ),
+                "update": [True, False],
+            }
+        )
+        expected = df.copy(deep=True)
+
+        update_df = df[df["update"]]
+
+        df.loc[df["update"], indexer] = update_df["date"]
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("indexer, idx", [(tm.loc, 1), (tm.iloc, 2)])
+    def test_setitem_value_coercing_dtypes(self, indexer, idx):
+        # GH#50467
+        df = DataFrame([["1", np.nan], ["2", np.nan], ["3", np.nan]], dtype=object)
+        rhs = DataFrame([[1, np.nan], [2, np.nan]])
+        indexer(df)[:idx, :] = rhs
+        expected = DataFrame([[1, np.nan], [2, np.nan], ["3", np.nan]], dtype=object)
+        tm.assert_frame_equal(df, expected)
+
+    def test_big_endian_support_selecting_columns(self):
+        # GH#57457
+        columns = ["a"]
+        data = [np.array([1, 2], dtype=">f8")]
+        df = DataFrame(dict(zip(columns, data)))
+        result = df[df.columns]
+        dfexp = DataFrame({"a": [1, 2]}, dtype=">f8")
+        expected = dfexp[dfexp.columns]
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameIndexingUInt64:
+    def test_setitem(self):
+        df = DataFrame(
+            {"A": np.arange(3), "B": [2**63, 2**63 + 5, 2**63 + 10]},
+            dtype=np.uint64,
+        )
+        idx = df["A"].rename("foo")
+
+        # setitem
+        assert "C" not in df.columns
+        df["C"] = idx
+        tm.assert_series_equal(df["C"], Series(idx, name="C"))
+
+        assert "D" not in df.columns
+        df["D"] = "foo"
+        df["D"] = idx
+        tm.assert_series_equal(df["D"], Series(idx, name="D"))
+        del df["D"]
+
+        # With NaN: because uint64 has no NaN element,
+        # the column should be cast to object.
+        df2 = df.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            df2.iloc[1, 1] = pd.NaT
+            df2.iloc[1, 2] = pd.NaT
+
+
+def test_object_casting_indexing_wraps_datetimelike():
+    # GH#31649, check the indexing methods all the way down the stack
+    df = DataFrame(
+        {
+            "A": [1, 2],
+            "B": date_range("2000", periods=2),
+            "C": pd.timedelta_range("1 Day", periods=2),
+        }
+    )
+
+    ser = df.loc[0]
+    assert isinstance(ser.values[1], Timestamp)
+    assert isinstance(ser.values[2], pd.Timedelta)
+
+    ser = df.iloc[0]
+    assert isinstance(ser.values[1], Timestamp)
+    assert isinstance(ser.values[2], pd.Timedelta)
+
+    ser = df.xs(0, axis=0)
+    assert isinstance(ser.values[1], Timestamp)
+    assert isinstance(ser.values[2], pd.Timedelta)
+
+    mgr = df._mgr
+    mgr._rebuild_blknos_and_blklocs()
+    arr = mgr.fast_xs(0).array
+    assert isinstance(arr[1], Timestamp)
+    assert isinstance(arr[2], pd.Timedelta)
+
+    blk = mgr.blocks[mgr.blknos[1]]
+    assert blk.dtype == "M8[ns]"  # we got the right block
+    val = blk.iget((0, 0))
+    assert isinstance(val, Timestamp)
+
+    blk = mgr.blocks[mgr.blknos[2]]
+    assert blk.dtype == "m8[ns]"  # we got the right block
+    val = blk.iget((0, 0))
+    assert isinstance(val, pd.Timedelta)
+
+
+msg1 = r"Cannot setitem on a Categorical with a new category( \(.*\))?, set the"
+msg2 = "Cannot set a Categorical with another, without identical categories"
+
+
+class TestLocILocDataFrameCategorical:
+    @pytest.fixture
+    def orig(self):
+        cats = Categorical(["a", "a", "a", "a", "a", "a", "a"], categories=["a", "b"])
+        idx = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values = [1, 1, 1, 1, 1, 1, 1]
+        orig = DataFrame({"cats": cats, "values": values}, index=idx)
+        return orig
+
+    @pytest.fixture
+    def exp_parts_cats_col(self):
+        # changed part of the cats column
+        cats3 = Categorical(["a", "a", "b", "b", "a", "a", "a"], categories=["a", "b"])
+        idx3 = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values3 = [1, 1, 1, 1, 1, 1, 1]
+        exp_parts_cats_col = DataFrame({"cats": cats3, "values": values3}, index=idx3)
+        return exp_parts_cats_col
+
+    @pytest.fixture
+    def exp_single_cats_value(self):
+        # changed single value in cats col
+        cats4 = Categorical(["a", "a", "b", "a", "a", "a", "a"], categories=["a", "b"])
+        idx4 = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values4 = [1, 1, 1, 1, 1, 1, 1]
+        exp_single_cats_value = DataFrame(
+            {"cats": cats4, "values": values4}, index=idx4
+        )
+        return exp_single_cats_value
+
+    def test_loc_iloc_setitem_list_of_lists(self, orig, indexer_li):
+        #   - assign multiple rows (mixed values) -> exp_multi_row
+        df = orig.copy()
+
+        key = slice(2, 4)
+        if indexer_li is tm.loc:
+            key = slice("j", "k")
+
+        indexer_li(df)[key, :] = [["b", 2], ["b", 2]]
+
+        cats2 = Categorical(["a", "a", "b", "b", "a", "a", "a"], categories=["a", "b"])
+        idx2 = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values2 = [1, 1, 2, 2, 1, 1, 1]
+        exp_multi_row = DataFrame({"cats": cats2, "values": values2}, index=idx2)
+        tm.assert_frame_equal(df, exp_multi_row)
+
+        df = orig.copy()
+        with pytest.raises(TypeError, match=msg1):
+            indexer_li(df)[key, :] = [["c", 2], ["c", 2]]
+
+    @pytest.mark.parametrize("indexer", [tm.loc, tm.iloc, tm.at, tm.iat])
+    def test_loc_iloc_at_iat_setitem_single_value_in_categories(
+        self, orig, exp_single_cats_value, indexer
+    ):
+        #   - assign a single value -> exp_single_cats_value
+        df = orig.copy()
+
+        key = (2, 0)
+        if indexer in [tm.loc, tm.at]:
+            key = (df.index[2], df.columns[0])
+
+        # "b" is among the categories for df["cat"}]
+        indexer(df)[key] = "b"
+        tm.assert_frame_equal(df, exp_single_cats_value)
+
+        # "c" is not among the categories for df["cat"]
+        with pytest.raises(TypeError, match=msg1):
+            indexer(df)[key] = "c"
+
+    def test_loc_iloc_setitem_mask_single_value_in_categories(
+        self, orig, exp_single_cats_value, indexer_li
+    ):
+        # mask with single True
+        df = orig.copy()
+
+        mask = df.index == "j"
+        key = 0
+        if indexer_li is tm.loc:
+            key = df.columns[key]
+
+        indexer_li(df)[mask, key] = "b"
+        tm.assert_frame_equal(df, exp_single_cats_value)
+
+    def test_loc_iloc_setitem_full_row_non_categorical_rhs(self, orig, indexer_li):
+        #   - assign a complete row (mixed values) -> exp_single_row
+        df = orig.copy()
+
+        key = 2
+        if indexer_li is tm.loc:
+            key = df.index[2]
+
+        # not categorical dtype, but "b" _is_ among the categories for df["cat"]
+        indexer_li(df)[key, :] = ["b", 2]
+        cats1 = Categorical(["a", "a", "b", "a", "a", "a", "a"], categories=["a", "b"])
+        idx1 = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values1 = [1, 1, 2, 1, 1, 1, 1]
+        exp_single_row = DataFrame({"cats": cats1, "values": values1}, index=idx1)
+        tm.assert_frame_equal(df, exp_single_row)
+
+        # "c" is not among the categories for df["cat"]
+        with pytest.raises(TypeError, match=msg1):
+            indexer_li(df)[key, :] = ["c", 2]
+
+    def test_loc_iloc_setitem_partial_col_categorical_rhs(
+        self, orig, exp_parts_cats_col, indexer_li
+    ):
+        # assign a part of a column with dtype == categorical ->
+        # exp_parts_cats_col
+        df = orig.copy()
+
+        key = (slice(2, 4), 0)
+        if indexer_li is tm.loc:
+            key = (slice("j", "k"), df.columns[0])
+
+        # same categories as we currently have in df["cats"]
+        compat = Categorical(["b", "b"], categories=["a", "b"])
+        indexer_li(df)[key] = compat
+        tm.assert_frame_equal(df, exp_parts_cats_col)
+
+        # categories do not match df["cat"]'s, but "b" is among them
+        semi_compat = Categorical(list("bb"), categories=list("abc"))
+        with pytest.raises(TypeError, match=msg2):
+            # different categories but holdable values
+            #  -> not sure if this should fail or pass
+            indexer_li(df)[key] = semi_compat
+
+        # categories do not match df["cat"]'s, and "c" is not among them
+        incompat = Categorical(list("cc"), categories=list("abc"))
+        with pytest.raises(TypeError, match=msg2):
+            # different values
+            indexer_li(df)[key] = incompat
+
+    def test_loc_iloc_setitem_non_categorical_rhs(
+        self, orig, exp_parts_cats_col, indexer_li
+    ):
+        # assign a part of a column with dtype != categorical -> exp_parts_cats_col
+        df = orig.copy()
+
+        key = (slice(2, 4), 0)
+        if indexer_li is tm.loc:
+            key = (slice("j", "k"), df.columns[0])
+
+        # "b" is among the categories for df["cat"]
+        indexer_li(df)[key] = ["b", "b"]
+        tm.assert_frame_equal(df, exp_parts_cats_col)
+
+        # "c" not part of the categories
+        with pytest.raises(TypeError, match=msg1):
+            indexer_li(df)[key] = ["c", "c"]
+
+    @pytest.mark.parametrize("indexer", [tm.getitem, tm.loc, tm.iloc])
+    def test_getitem_preserve_object_index_with_dates(self, indexer):
+        # https://github.com/pandas-dev/pandas/pull/42950 - when selecting a column
+        # from dataframe, don't try to infer object dtype index on Series construction
+        idx = date_range("2012", periods=3).astype(object)
+        df = DataFrame({0: [1, 2, 3]}, index=idx)
+        assert df.index.dtype == object
+
+        if indexer is tm.getitem:
+            ser = indexer(df)[0]
+        else:
+            ser = indexer(df)[:, 0]
+
+        assert ser.index.dtype == object
+
+    def test_loc_on_multiindex_one_level(self):
+        # GH#45779
+        df = DataFrame(
+            data=[[0], [1]],
+            index=MultiIndex.from_tuples([("a",), ("b",)], names=["first"]),
+        )
+        expected = DataFrame(
+            data=[[0]], index=MultiIndex.from_tuples([("a",)], names=["first"])
+        )
+        result = df.loc["a"]
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDeprecatedIndexers:
+    @pytest.mark.parametrize(
+        "key", [{1}, {1: 1}, ({1}, "a"), ({1: 1}, "a"), (1, {"a"}), (1, {"a": "a"})]
+    )
+    def test_getitem_dict_and_set_deprecated(self, key):
+        # GH#42825 enforced in 2.0
+        df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            df.loc[key]
+
+    @pytest.mark.parametrize(
+        "key",
+        [
+            {1},
+            {1: 1},
+            (({1}, 2), "a"),
+            (({1: 1}, 2), "a"),
+            ((1, 2), {"a"}),
+            ((1, 2), {"a": "a"}),
+        ],
+    )
+    def test_getitem_dict_and_set_deprecated_multiindex(self, key):
+        # GH#42825 enforced in 2.0
+        df = DataFrame(
+            [[1, 2], [3, 4]],
+            columns=["a", "b"],
+            index=MultiIndex.from_tuples([(1, 2), (3, 4)]),
+        )
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            df.loc[key]
+
+    @pytest.mark.parametrize(
+        "key", [{1}, {1: 1}, ({1}, "a"), ({1: 1}, "a"), (1, {"a"}), (1, {"a": "a"})]
+    )
+    def test_setitem_dict_and_set_disallowed(self, key):
+        # GH#42825 enforced in 2.0
+        df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            df.loc[key] = 1
+
+    @pytest.mark.parametrize(
+        "key",
+        [
+            {1},
+            {1: 1},
+            (({1}, 2), "a"),
+            (({1: 1}, 2), "a"),
+            ((1, 2), {"a"}),
+            ((1, 2), {"a": "a"}),
+        ],
+    )
+    def test_setitem_dict_and_set_disallowed_multiindex(self, key):
+        # GH#42825 enforced in 2.0
+        df = DataFrame(
+            [[1, 2], [3, 4]],
+            columns=["a", "b"],
+            index=MultiIndex.from_tuples([(1, 2), (3, 4)]),
+        )
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            df.loc[key] = 1
+
+
+def test_adding_new_conditional_column() -> None:
+    # https://github.com/pandas-dev/pandas/issues/55025
+    df = DataFrame({"x": [1]})
+    df.loc[df["x"] == 1, "y"] = "1"
+    expected = DataFrame({"x": [1], "y": ["1"]})
+    tm.assert_frame_equal(df, expected)
+
+    df = DataFrame({"x": [1]})
+    # try inserting something which numpy would store as 'object'
+    value = lambda x: x
+    df.loc[df["x"] == 1, "y"] = value
+    expected = DataFrame({"x": [1], "y": [value]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize(
+    ("dtype", "infer_string"),
+    [
+        (object, False),
+        (pd.StringDtype(na_value=np.nan), True),
+    ],
+)
+def test_adding_new_conditional_column_with_string(dtype, infer_string) -> None:
+    # https://github.com/pandas-dev/pandas/issues/56204
+    df = DataFrame({"a": [1, 2], "b": [3, 4]})
+    with pd.option_context("future.infer_string", infer_string):
+        df.loc[df["a"] == 1, "c"] = "1"
+    expected = DataFrame({"a": [1, 2], "b": [3, 4], "c": ["1", float("nan")]}).astype(
+        {"a": "int64", "b": "int64", "c": dtype}
+    )
+    tm.assert_frame_equal(df, expected)
+
+
+def test_add_new_column_infer_string():
+    # GH#55366
+    df = DataFrame({"x": [1]})
+    with pd.option_context("future.infer_string", True):
+        df.loc[df["x"] == 1, "y"] = "1"
+    expected = DataFrame(
+        {"x": [1], "y": Series(["1"], dtype=pd.StringDtype(na_value=np.nan))},
+        columns=Index(["x", "y"], dtype="str"),
+    )
+    tm.assert_frame_equal(df, expected)
+
+
+class TestSetitemValidation:
+    # This is adapted from pandas/tests/arrays/masked/test_indexing.py
+    def _check_setitem_invalid(self, df, invalid, indexer):
+        orig_df = df.copy()
+
+        # iloc
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.iloc[indexer, 0] = invalid
+            df = orig_df.copy()
+
+        # loc
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[indexer, "a"] = invalid
+            df = orig_df.copy()
+
+    def _check_setitem_valid(self, df, value, indexer):
+        orig_df = df.copy()
+
+        # iloc
+        df.iloc[indexer, 0] = value
+        df = orig_df.copy()
+
+        # loc
+        df.loc[indexer, "a"] = value
+        df = orig_df.copy()
+
+    _invalid_scalars = [
+        1 + 2j,
+        "True",
+        "1",
+        "1.0",
+        pd.NaT,
+        np.datetime64("NaT"),
+        np.timedelta64("NaT"),
+    ]
+    _indexers = [0, [0], slice(0, 1), [True, False, False], slice(None, None, None)]
+
+    @pytest.mark.parametrize(
+        "invalid", _invalid_scalars + [1, 1.0, np.int64(1), np.float64(1)]
+    )
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_bool(self, invalid, indexer):
+        df = DataFrame({"a": [True, False, False]}, dtype="bool")
+        self._check_setitem_invalid(df, invalid, indexer)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True, 1.5, np.float64(1.5)])
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_int(self, invalid, any_int_numpy_dtype, indexer):
+        df = DataFrame({"a": [1, 2, 3]}, dtype=any_int_numpy_dtype)
+        if isna(invalid) and invalid is not pd.NaT and not np.isnat(invalid):
+            self._check_setitem_valid(df, invalid, indexer)
+        else:
+            self._check_setitem_invalid(df, invalid, indexer)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True])
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_float(self, invalid, float_numpy_dtype, indexer):
+        df = DataFrame({"a": [1, 2, None]}, dtype=float_numpy_dtype)
+        self._check_setitem_invalid(df, invalid, indexer)
diff --git a/pandas/pandas/tests/frame/indexing/test_insert.py b/pandas/pandas/tests/frame/indexing/test_insert.py
new file mode 100644
index 0000000000000..b530cb98ef46c
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_insert.py
@@ -0,0 +1,111 @@
+"""
+test_insert is specifically for the DataFrame.insert method; not to be
+confused with tests with "insert" in their names that are really testing
+__setitem__.
+"""
+
+import numpy as np
+import pytest
+
+from pandas.errors import PerformanceWarning
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameInsert:
+    def test_insert(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=np.arange(5),
+            columns=["c", "b", "a"],
+        )
+
+        df.insert(0, "foo", df["a"])
+        tm.assert_index_equal(df.columns, Index(["foo", "c", "b", "a"]))
+        tm.assert_series_equal(df["a"], df["foo"], check_names=False)
+
+        df.insert(2, "bar", df["c"])
+        tm.assert_index_equal(df.columns, Index(["foo", "c", "bar", "b", "a"]))
+        tm.assert_almost_equal(df["c"], df["bar"], check_names=False)
+
+        with pytest.raises(ValueError, match="already exists"):
+            df.insert(1, "a", df["b"])
+
+        msg = "cannot insert c, already exists"
+        with pytest.raises(ValueError, match=msg):
+            df.insert(1, "c", df["b"])
+
+        df.columns.name = "some_name"
+        # preserve columns name field
+        df.insert(0, "baz", df["c"])
+        assert df.columns.name == "some_name"
+
+    def test_insert_column_bug_4032(self):
+        # GH#4032, inserting a column and renaming causing errors
+        df = DataFrame({"b": [1.1, 2.2]})
+
+        df = df.rename(columns={})
+        df.insert(0, "a", [1, 2])
+        result = df.rename(columns={})
+
+        expected = DataFrame([[1, 1.1], [2, 2.2]], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+        df.insert(0, "c", [1.3, 2.3])
+        result = df.rename(columns={})
+
+        expected = DataFrame([[1.3, 1, 1.1], [2.3, 2, 2.2]], columns=["c", "a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_insert_with_columns_dups(self):
+        # GH#14291
+        df = DataFrame()
+        df.insert(0, "A", ["g", "h", "i"], allow_duplicates=True)
+        df.insert(0, "A", ["d", "e", "f"], allow_duplicates=True)
+        df.insert(0, "A", ["a", "b", "c"], allow_duplicates=True)
+        exp = DataFrame(
+            [["a", "d", "g"], ["b", "e", "h"], ["c", "f", "i"]], columns=["A", "A", "A"]
+        )
+        tm.assert_frame_equal(df, exp)
+
+    def test_insert_item_cache(self, performance_warning):
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 3)))
+        ser = df[0]
+        expected_warning = PerformanceWarning if performance_warning else None
+
+        with tm.assert_produces_warning(expected_warning):
+            for n in range(100):
+                df[n + 3] = df[1] * n
+
+        ser.iloc[0] = 99
+        assert df.iloc[0, 0] == df[0][0]
+        assert df.iloc[0, 0] != 99
+
+    def test_insert_EA_no_warning(self):
+        # PerformanceWarning about fragmented frame should not be raised when
+        # using EAs (https://github.com/pandas-dev/pandas/issues/44098)
+        df = DataFrame(
+            np.random.default_rng(2).integers(0, 100, size=(3, 100)), dtype="Int64"
+        )
+        with tm.assert_produces_warning(None):
+            df["a"] = np.array([1, 2, 3])
+
+    def test_insert_frame(self):
+        # GH#42403
+        df = DataFrame({"col1": [1, 2], "col2": [3, 4]})
+
+        msg = (
+            "Expected a one-dimensional object, got a DataFrame with 2 columns instead."
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.insert(1, "newcol", df)
+
+    def test_insert_int64_loc(self):
+        # GH#53193
+        df = DataFrame({"a": [1, 2]})
+        df.insert(np.int64(0), "b", 0)
+        tm.assert_frame_equal(df, DataFrame({"b": [0, 0], "a": [1, 2]}))
diff --git a/pandas/pandas/tests/frame/indexing/test_mask.py b/pandas/pandas/tests/frame/indexing/test_mask.py
new file mode 100644
index 0000000000000..ac6f0a1ac0f73
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_mask.py
@@ -0,0 +1,152 @@
+"""
+Tests for DataFrame.mask; tests DataFrame.where as a side-effect.
+"""
+
+import numpy as np
+
+from pandas import (
+    NA,
+    DataFrame,
+    Float64Dtype,
+    Series,
+    StringDtype,
+    Timedelta,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameMask:
+    def test_mask(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        cond = df > 0
+
+        rs = df.where(cond, np.nan)
+        tm.assert_frame_equal(rs, df.mask(df <= 0))
+        tm.assert_frame_equal(rs, df.mask(~cond))
+
+        other = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        rs = df.where(cond, other)
+        tm.assert_frame_equal(rs, df.mask(df <= 0, other))
+        tm.assert_frame_equal(rs, df.mask(~cond, other))
+
+    def test_mask2(self):
+        # see GH#21891
+        df = DataFrame([1, 2])
+        res = df.mask([[True], [False]])
+
+        exp = DataFrame([np.nan, 2])
+        tm.assert_frame_equal(res, exp)
+
+    def test_mask_inplace(self):
+        # GH#8801
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        cond = df > 0
+
+        rdf = df.copy()
+
+        return_value = rdf.where(cond, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(rdf, df.where(cond))
+        tm.assert_frame_equal(rdf, df.mask(~cond))
+
+        rdf = df.copy()
+        return_value = rdf.where(cond, -df, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(rdf, df.where(cond, -df))
+        tm.assert_frame_equal(rdf, df.mask(~cond, -df))
+
+    def test_mask_edge_case_1xN_frame(self):
+        # GH#4071
+        df = DataFrame([[1, 2]])
+        res = df.mask(DataFrame([[True, False]]))
+        expec = DataFrame([[np.nan, 2]])
+        tm.assert_frame_equal(res, expec)
+
+    def test_mask_callable(self):
+        # GH#12533
+        df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
+        result = df.mask(lambda x: x > 4, lambda x: x + 1)
+        exp = DataFrame([[1, 2, 3], [4, 6, 7], [8, 9, 10]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, df.mask(df > 4, df + 1))
+
+        # return ndarray and scalar
+        result = df.mask(lambda x: (x % 2 == 0).values, lambda x: 99)
+        exp = DataFrame([[1, 99, 3], [99, 5, 99], [7, 99, 9]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, df.mask(df % 2 == 0, 99))
+
+        # chain
+        result = (df + 2).mask(lambda x: x > 8, lambda x: x + 10)
+        exp = DataFrame([[3, 4, 5], [6, 7, 8], [19, 20, 21]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, (df + 2).mask((df + 2) > 8, (df + 2) + 10))
+
+    def test_mask_dtype_bool_conversion(self):
+        # GH#3733
+        df = DataFrame(data=np.random.default_rng(2).standard_normal((100, 50)))
+        df = df.where(df > 0)  # create nans
+        bools = df > 0
+        mask = isna(df)
+        expected = bools.astype(object).mask(mask)
+        result = bools.mask(mask)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_mask_stringdtype(frame_or_series):
+    # GH 40824
+    obj = DataFrame(
+        {"A": ["foo", "bar", "baz", NA]},
+        index=["id1", "id2", "id3", "id4"],
+        dtype=StringDtype(),
+    )
+    filtered_obj = DataFrame(
+        {"A": ["this", "that"]}, index=["id2", "id3"], dtype=StringDtype()
+    )
+    expected = DataFrame(
+        {"A": [NA, "this", "that", NA]},
+        index=["id1", "id2", "id3", "id4"],
+        dtype=StringDtype(),
+    )
+    if frame_or_series is Series:
+        obj = obj["A"]
+        filtered_obj = filtered_obj["A"]
+        expected = expected["A"]
+
+    filter_ser = Series([False, True, True, False])
+    result = obj.mask(filter_ser, filtered_obj)
+
+    tm.assert_equal(result, expected)
+
+
+def test_mask_where_dtype_timedelta():
+    # https://github.com/pandas-dev/pandas/issues/39548
+    df = DataFrame([Timedelta(i, unit="D") for i in range(5)])
+
+    expected = DataFrame(np.full(5, np.nan, dtype="timedelta64[ns]"))
+    tm.assert_frame_equal(df.mask(df.notna()), expected)
+
+    expected = DataFrame(
+        [np.nan, np.nan, np.nan, Timedelta("3 day"), Timedelta("4 day")]
+    )
+    tm.assert_frame_equal(df.where(df > Timedelta(2, unit="D")), expected)
+
+
+def test_mask_return_dtype():
+    # GH#50488
+    ser = Series([0.0, 1.0, 2.0, 3.0], dtype=Float64Dtype())
+    cond = ~ser.isna()
+    other = Series([True, False, True, False])
+    excepted = Series([1.0, 0.0, 1.0, 0.0], dtype=ser.dtype)
+    result = ser.mask(cond, other)
+    tm.assert_series_equal(result, excepted)
+
+
+def test_mask_inplace_no_other():
+    # GH#51685
+    df = DataFrame({"a": [1.0, 2.0], "b": ["x", "y"]})
+    cond = DataFrame({"a": [True, False], "b": [False, True]})
+    df.mask(cond, inplace=True)
+    expected = DataFrame({"a": [np.nan, 2], "b": ["x", np.nan]})
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/indexing/test_set_value.py b/pandas/pandas/tests/frame/indexing/test_set_value.py
new file mode 100644
index 0000000000000..124505f440e6c
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_set_value.py
@@ -0,0 +1,74 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_float_dtype
+
+from pandas import (
+    DataFrame,
+    isna,
+)
+
+
+class TestSetValue:
+    def test_set_value(self, float_frame):
+        for idx in float_frame.index:
+            for col in float_frame.columns:
+                float_frame._set_value(idx, col, 1)
+                assert float_frame[col][idx] == 1
+
+    def test_set_value_resize(self, float_frame, using_infer_string):
+        res = float_frame._set_value("foobar", "B", 0)
+        assert res is None
+        assert float_frame.index[-1] == "foobar"
+        assert float_frame._get_value("foobar", "B") == 0
+
+        float_frame.loc["foobar", "qux"] = 0
+        assert float_frame._get_value("foobar", "qux") == 0
+
+        res = float_frame.copy()
+        res._set_value("foobar", "baz", "sam")
+        if using_infer_string:
+            assert res["baz"].dtype == "str"
+        else:
+            assert res["baz"].dtype == np.object_
+        res = float_frame.copy()
+        res._set_value("foobar", "baz", True)
+        assert res["baz"].dtype == np.object_
+
+        res = float_frame.copy()
+        res._set_value("foobar", "baz", 5)
+        assert is_float_dtype(res["baz"])
+        assert isna(res["baz"].drop(["foobar"])).all()
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            res._set_value("foobar", "baz", "sam")
+
+    def test_set_value_with_index_dtype_change(self):
+        df_orig = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=range(3),
+            columns=list("ABC"),
+        )
+
+        # this is actually ambiguous as the 2 is interpreted as a positional
+        # so column is not created
+        df = df_orig.copy()
+        df._set_value("C", 2, 1.0)
+        assert list(df.index) == list(df_orig.index) + ["C"]
+        # assert list(df.columns) == list(df_orig.columns) + [2]
+
+        df = df_orig.copy()
+        df.loc["C", 2] = 1.0
+        assert list(df.index) == list(df_orig.index) + ["C"]
+        # assert list(df.columns) == list(df_orig.columns) + [2]
+
+        # create both new
+        df = df_orig.copy()
+        df._set_value("C", "D", 1.0)
+        assert list(df.index) == list(df_orig.index) + ["C"]
+        assert list(df.columns) == list(df_orig.columns) + ["D"]
+
+        df = df_orig.copy()
+        df.loc["C", "D"] = 1.0
+        assert list(df.index) == list(df_orig.index) + ["C"]
+        assert list(df.columns) == list(df_orig.columns) + ["D"]
diff --git a/pandas/pandas/tests/frame/indexing/test_setitem.py b/pandas/pandas/tests/frame/indexing/test_setitem.py
new file mode 100644
index 0000000000000..20dd7b0c4d3e7
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_setitem.py
@@ -0,0 +1,1401 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.base import _registry as ea_registry
+from pandas.core.dtypes.common import is_object_dtype
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    IntervalDtype,
+    PeriodDtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    Period,
+    PeriodIndex,
+    Series,
+    Timestamp,
+    cut,
+    date_range,
+    notna,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import SparseArray
+
+from pandas.tseries.offsets import BDay
+
+
+class TestDataFrameSetItem:
+    def test_setitem_str_subclass(self):
+        # GH#37366
+        class mystring(str):
+            __slots__ = ()
+
+        data = ["2020-10-22 01:21:00+00:00"]
+        index = DatetimeIndex(data)
+        df = DataFrame({"a": [1]}, index=index)
+        df["b"] = 2
+        df[mystring("c")] = 3
+        expected = DataFrame({"a": [1], "b": [2], mystring("c"): [3]}, index=index)
+        tm.assert_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "dtype", ["int32", "int64", "uint32", "uint64", "float32", "float64"]
+    )
+    def test_setitem_dtype(self, dtype, float_frame):
+        # Use integers since casting negative floats to uints is undefined
+        arr = np.random.default_rng(2).integers(1, 10, len(float_frame))
+
+        float_frame[dtype] = np.array(arr, dtype=dtype)
+        assert float_frame[dtype].dtype.name == dtype
+
+    def test_setitem_list_not_dataframe(self, float_frame):
+        data = np.random.default_rng(2).standard_normal((len(float_frame), 2))
+        float_frame[["A", "B"]] = data
+        tm.assert_almost_equal(float_frame[["A", "B"]].values, data)
+
+    def test_setitem_error_msmgs(self):
+        # GH 7432
+        df = DataFrame(
+            {"bar": [1, 2, 3], "baz": ["d", "e", "f"]},
+            index=Index(["a", "b", "c"], name="foo"),
+        )
+        ser = Series(
+            ["g", "h", "i", "j"],
+            index=Index(["a", "b", "c", "a"], name="foo"),
+            name="fiz",
+        )
+        msg = "cannot reindex on an axis with duplicate labels"
+        with pytest.raises(ValueError, match=msg):
+            df["newcol"] = ser
+
+        # GH 4107, more descriptive error message
+        df = DataFrame(
+            np.random.default_rng(2).integers(0, 2, (4, 4)),
+            columns=["a", "b", "c", "d"],
+        )
+
+        msg = "Cannot set a DataFrame with multiple columns to the single column gr"
+        with pytest.raises(ValueError, match=msg):
+            df["gr"] = df.groupby(["b", "c"]).count()
+
+        # GH 55956, specific message for zero columns
+        msg = "Cannot set a DataFrame without columns to the column gr"
+        with pytest.raises(ValueError, match=msg):
+            df["gr"] = DataFrame()
+
+    def test_setitem_benchmark(self):
+        # from the vb_suite/frame_methods/frame_insert_columns
+        N = 10
+        K = 5
+        df = DataFrame(index=range(N))
+        new_col = np.random.default_rng(2).standard_normal(N)
+        for i in range(K):
+            df[i] = new_col
+        expected = DataFrame(np.repeat(new_col, K).reshape(N, K), index=range(N))
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_different_dtype(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=np.arange(5),
+            columns=["c", "b", "a"],
+        )
+        df.insert(0, "foo", df["a"])
+        df.insert(2, "bar", df["c"])
+
+        # diff dtype
+
+        # new item
+        df["x"] = df["a"].astype("float32")
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 5 + [np.dtype("float32")],
+            index=["foo", "c", "bar", "b", "a", "x"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # replacing current (in different block)
+        df["a"] = df["a"].astype("float32")
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 4 + [np.dtype("float32")] * 2,
+            index=["foo", "c", "bar", "b", "a", "x"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        df["y"] = df["a"].astype("int32")
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 4 + [np.dtype("float32")] * 2 + [np.dtype("int32")],
+            index=["foo", "c", "bar", "b", "a", "x", "y"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_overwrite_index(self):
+        # GH 13522 - assign the index as a column and then overwrite the values
+        # -> should not affect the index
+        df = DataFrame(index=["A", "B", "C"])
+        df["X"] = df.index
+        df["X"] = ["x", "y", "z"]
+        exp = DataFrame(
+            data={"X": ["x", "y", "z"]}, index=["A", "B", "C"], columns=["X"]
+        )
+        tm.assert_frame_equal(df, exp)
+
+    def test_setitem_empty_columns(self):
+        # Starting from an empty DataFrame and setting a column should result
+        # in a default string dtype for the columns' Index
+        # https://github.com/pandas-dev/pandas/issues/60338
+
+        df = DataFrame()
+        df["foo"] = [1, 2, 3]
+        expected = DataFrame({"foo": [1, 2, 3]})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(columns=Index([]))
+        df["foo"] = [1, 2, 3]
+        expected = DataFrame({"foo": [1, 2, 3]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_dt64_index_empty_columns(self):
+        rng = date_range("1/1/2000 00:00:00", "1/1/2000 1:59:50", freq="10s")
+        df = DataFrame(index=np.arange(len(rng)))
+
+        df["A"] = rng
+        assert df["A"].dtype == np.dtype("M8[ns]")
+
+    def test_setitem_timestamp_empty_columns(self):
+        # GH#19843
+        df = DataFrame(index=range(3))
+        df["now"] = Timestamp("20130101", tz="UTC")
+
+        expected = DataFrame(
+            [[Timestamp("20130101", tz="UTC")]] * 3, index=range(3), columns=["now"]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_wrong_length_categorical_dtype_raises(self):
+        # GH#29523
+        cat = Categorical.from_codes([0, 1, 1, 0, 1, 2], ["a", "b", "c"])
+        df = DataFrame(range(10), columns=["bar"])
+
+        msg = (
+            rf"Length of values \({len(cat)}\) "
+            rf"does not match length of index \({len(df)}\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df["foo"] = cat
+
+    def test_setitem_with_sparse_value(self):
+        # GH#8131
+        df = DataFrame({"c_1": ["a", "b", "c"], "n_1": [1.0, 2.0, 3.0]})
+        sp_array = SparseArray([0, 0, 1])
+        df["new_column"] = sp_array
+
+        expected = Series(sp_array, name="new_column")
+        tm.assert_series_equal(df["new_column"], expected)
+
+    def test_setitem_with_unaligned_sparse_value(self):
+        df = DataFrame({"c_1": ["a", "b", "c"], "n_1": [1.0, 2.0, 3.0]})
+        sp_series = Series(SparseArray([0, 0, 1]), index=[2, 1, 0])
+
+        df["new_column"] = sp_series
+        expected = Series(SparseArray([1, 0, 0]), name="new_column")
+        tm.assert_series_equal(df["new_column"], expected)
+
+    def test_setitem_period_preserves_dtype(self):
+        # GH: 26861
+        data = [Period("2003-12", "D")]
+        result = DataFrame([])
+        result["a"] = data
+
+        expected = DataFrame({"a": data}, columns=["a"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_dict_preserves_dtypes(self):
+        # https://github.com/pandas-dev/pandas/issues/34573
+        expected = DataFrame(
+            {
+                "a": Series([0, 1, 2], dtype="int64"),
+                "b": Series([1, 2, 3], dtype=float),
+                "c": Series([1, 2, 3], dtype=float),
+                "d": Series([1, 2, 3], dtype="uint32"),
+            }
+        )
+        df = DataFrame(
+            {
+                "a": Series([], dtype="int64"),
+                "b": Series([], dtype=float),
+                "c": Series([], dtype=float),
+                "d": Series([], dtype="uint32"),
+            }
+        )
+        for idx, b in enumerate([1, 2, 3]):
+            df.loc[df.shape[0]] = {
+                "a": int(idx),
+                "b": float(b),
+                "c": float(b),
+                "d": np.uint32(b),
+            }
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "obj,dtype",
+        [
+            (Period("2020-01"), PeriodDtype("M")),
+            (Interval(left=0, right=5), IntervalDtype("int64", "right")),
+            (
+                Timestamp("2011-01-01", tz="US/Eastern"),
+                DatetimeTZDtype(unit="s", tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_setitem_extension_types(self, obj, dtype):
+        # GH: 34832
+        expected = DataFrame({"idx": [1, 2, 3], "obj": Series([obj] * 3, dtype=dtype)})
+
+        df = DataFrame({"idx": [1, 2, 3]})
+        df["obj"] = obj
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "ea_name",
+        [
+            dtype.name
+            for dtype in ea_registry.dtypes
+            # property would require instantiation
+            if not isinstance(dtype.name, property)
+        ]
+        + ["datetime64[ns, UTC]", "period[D]"],
+    )
+    def test_setitem_with_ea_name(self, ea_name):
+        # GH 38386
+        result = DataFrame([0])
+        result[ea_name] = [1]
+        expected = DataFrame({0: [0], ea_name: [1]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_dt64_ndarray_with_NaT_and_diff_time_units(self):
+        # GH#7492
+        data_ns = np.array([1, "nat"], dtype="datetime64[ns]")
+        result = Series(data_ns).to_frame()
+        result["new"] = data_ns
+        expected = DataFrame({0: [1, None], "new": [1, None]}, dtype="datetime64[ns]")
+        tm.assert_frame_equal(result, expected)
+
+        # OutOfBoundsDatetime error shouldn't occur; as of 2.0 we preserve "M8[s]"
+        data_s = np.array([1, "nat"], dtype="datetime64[s]")
+        result["new"] = data_s
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_numpy_array_equal(result["new"].to_numpy(), data_s)
+
+    @pytest.mark.parametrize("unit", ["h", "m", "s", "ms", "D", "M", "Y"])
+    def test_frame_setitem_datetime64_col_other_units(self, unit):
+        # Check that non-nano dt64 values get cast to dt64 on setitem
+        #  into a not-yet-existing column
+        n = 100
+
+        dtype = np.dtype(f"M8[{unit}]")
+        vals = np.arange(n, dtype=np.int64).view(dtype)
+        if unit in ["s", "ms"]:
+            # supported unit
+            ex_vals = vals
+        else:
+            # we get the nearest supported units, i.e. "s"
+            ex_vals = vals.astype("datetime64[s]")
+
+        df = DataFrame({"ints": np.arange(n)}, index=np.arange(n))
+        df[unit] = vals
+
+        assert df[unit].dtype == ex_vals.dtype
+        assert (df[unit].values == ex_vals).all()
+
+    @pytest.mark.parametrize("unit", ["h", "m", "s", "ms", "D", "M", "Y"])
+    def test_frame_setitem_existing_datetime64_col_other_units(self, unit):
+        # Check that non-nano dt64 values get cast to dt64 on setitem
+        #  into an already-existing dt64 column
+        n = 100
+
+        dtype = np.dtype(f"M8[{unit}]")
+        vals = np.arange(n, dtype=np.int64).view(dtype)
+        ex_vals = vals.astype("datetime64[ns]")
+
+        df = DataFrame({"ints": np.arange(n)}, index=np.arange(n))
+        df["dates"] = np.arange(n, dtype=np.int64).view("M8[ns]")
+
+        # We overwrite existing dt64 column with new, non-nano dt64 vals
+        df["dates"] = vals
+        assert (df["dates"].values == ex_vals).all()
+
+    def test_setitem_dt64tz(self, timezone_frame):
+        df = timezone_frame
+        idx = df["B"].rename("foo")
+
+        # setitem
+        df["C"] = idx
+        tm.assert_series_equal(df["C"], Series(idx, name="C"))
+
+        df["D"] = "foo"
+        df["D"] = idx
+        tm.assert_series_equal(df["D"], Series(idx, name="D"))
+        del df["D"]
+
+        # assert that A & C are not sharing the same base (e.g. they
+        # are copies)
+        # Note: This does not hold with Copy on Write (because of lazy copying)
+        v1 = df._mgr.blocks[1].values
+        v2 = df._mgr.blocks[2].values
+        tm.assert_extension_array_equal(v1, v2)
+        v1base = v1._ndarray.base
+        v2base = v2._ndarray.base
+        assert id(v1base) == id(v2base)
+
+        # with nan
+        df2 = df.copy()
+        df2.iloc[1, 1] = NaT
+        df2.iloc[1, 2] = NaT
+        result = df2["B"]
+        tm.assert_series_equal(notna(result), Series([True, False, True], name="B"))
+        tm.assert_series_equal(df2.dtypes, df.dtypes)
+
+    def test_setitem_periodindex(self):
+        rng = period_range("1/1/2000", periods=5, name="index")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)), index=rng)
+
+        df["Index"] = rng
+        rs = Index(df["Index"])
+        tm.assert_index_equal(rs, rng, check_names=False)
+        assert rs.name == "Index"
+        assert rng.name == "index"
+
+        rs = df.reset_index().set_index("index")
+        assert isinstance(rs.index, PeriodIndex)
+        tm.assert_index_equal(rs.index, rng)
+
+    def test_setitem_complete_column_with_array(self):
+        # GH#37954
+        df = DataFrame({"a": ["one", "two", "three"], "b": [1, 2, 3]})
+        arr = np.array([[1, 1], [3, 1], [5, 1]])
+        df[["c", "d"]] = arr
+        expected = DataFrame(
+            {
+                "a": ["one", "two", "three"],
+                "b": [1, 2, 3],
+                "c": [1, 3, 5],
+                "d": [1, 1, 1],
+            }
+        )
+        expected["c"] = expected["c"].astype(arr.dtype)
+        expected["d"] = expected["d"].astype(arr.dtype)
+        assert expected["c"].dtype == arr.dtype
+        assert expected["d"].dtype == arr.dtype
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_period_d_dtype(self):
+        # GH 39763
+        rng = period_range("2016-01-01", periods=9, freq="D", name="A")
+        result = DataFrame(rng)
+        expected = DataFrame(
+            {"A": ["NaT", "NaT", "NaT", "NaT", "NaT", "NaT", "NaT", "NaT", "NaT"]},
+            dtype="period[D]",
+        )
+        result.iloc[:] = rng._na_value
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["f8", "i8", "u8"])
+    def test_setitem_bool_with_numeric_index(self, dtype):
+        # GH#36319
+        cols = Index([1, 2, 3], dtype=dtype)
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 3)), columns=cols)
+
+        df[False] = ["a", "b", "c"]
+
+        expected_cols = Index([1, 2, 3, False], dtype=object)
+        if dtype == "f8":
+            expected_cols = Index([1.0, 2.0, 3.0, False], dtype=object)
+
+        tm.assert_index_equal(df.columns, expected_cols)
+
+    @pytest.mark.parametrize("indexer", ["B", ["B"]])
+    def test_setitem_frame_length_0_str_key(self, indexer):
+        # GH#38831
+        df = DataFrame(columns=["A", "B"])
+        other = DataFrame({"B": [1, 2]})
+        df[indexer] = other
+        expected = DataFrame({"A": [np.nan] * 2, "B": [1, 2]})
+        expected["A"] = expected["A"].astype("object")
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_frame_duplicate_columns(self):
+        # GH#15695
+        cols = ["A", "B", "C"] * 2
+        df = DataFrame(index=range(3), columns=cols)
+        df.loc[0, "A"] = (0, 3)
+        df.loc[:, "B"] = (1, 4)
+        df["C"] = (2, 5)
+        expected = DataFrame(
+            [
+                [0, 1, 2, 3, 4, 5],
+                [np.nan, 1, 2, np.nan, 4, 5],
+                [np.nan, 1, 2, np.nan, 4, 5],
+            ],
+            dtype="object",
+        )
+
+        # set these with unique columns to be extra-unambiguous
+        expected[2] = expected[2].astype(np.int64)
+        expected[5] = expected[5].astype(np.int64)
+        expected.columns = cols
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_frame_duplicate_columns_size_mismatch(self):
+        # GH#39510
+        cols = ["A", "B", "C"] * 2
+        df = DataFrame(index=range(3), columns=cols)
+        with pytest.raises(ValueError, match="Columns must be same length as key"):
+            df[["A"]] = (0, 3, 5)
+
+        df2 = df.iloc[:, :3]  # unique columns
+        with pytest.raises(ValueError, match="Columns must be same length as key"):
+            df2[["A"]] = (0, 3, 5)
+
+    @pytest.mark.parametrize("cols", [["a", "b", "c"], ["a", "a", "a"]])
+    def test_setitem_df_wrong_column_number(self, cols):
+        # GH#38604
+        df = DataFrame([[1, 2, 3]], columns=cols)
+        rhs = DataFrame([[10, 11]], columns=["d", "e"])
+        msg = "Columns must be same length as key"
+        with pytest.raises(ValueError, match=msg):
+            df["a"] = rhs
+
+    def test_setitem_listlike_indexer_duplicate_columns(self):
+        # GH#38604
+        df = DataFrame([[1, 2, 3]], columns=["a", "b", "b"])
+        rhs = DataFrame([[10, 11, 12]], columns=["a", "b", "b"])
+        df[["a", "b"]] = rhs
+        expected = DataFrame([[10, 11, 12]], columns=["a", "b", "b"])
+        tm.assert_frame_equal(df, expected)
+
+        df[["c", "b"]] = rhs
+        expected = DataFrame([[10, 11, 12, 10]], columns=["a", "b", "b", "c"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_listlike_indexer_duplicate_columns_not_equal_length(self):
+        # GH#39403
+        df = DataFrame([[1, 2, 3]], columns=["a", "b", "b"])
+        rhs = DataFrame([[10, 11]], columns=["a", "b"])
+        msg = "Columns must be same length as key"
+        with pytest.raises(ValueError, match=msg):
+            df[["a", "b"]] = rhs
+
+    def test_setitem_intervals(self):
+        df = DataFrame({"A": range(10)})
+        ser = cut(df["A"], 5)
+        assert isinstance(ser.cat.categories, IntervalIndex)
+
+        # B & D end up as Categoricals
+        # the remainder are converted to in-line objects
+        # containing an IntervalIndex.values
+        df["B"] = ser
+        df["C"] = np.array(ser)
+        df["D"] = ser.values
+        df["E"] = np.array(ser.values)
+        df["F"] = ser.astype(object)
+
+        assert isinstance(df["B"].dtype, CategoricalDtype)
+        assert isinstance(df["B"].cat.categories.dtype, IntervalDtype)
+        assert isinstance(df["D"].dtype, CategoricalDtype)
+        assert isinstance(df["D"].cat.categories.dtype, IntervalDtype)
+
+        # These go through the Series constructor and so get inferred back
+        #  to IntervalDtype
+        assert isinstance(df["C"].dtype, IntervalDtype)
+        assert isinstance(df["E"].dtype, IntervalDtype)
+
+        # But the Series constructor doesn't do inference on Series objects,
+        #  so setting df["F"] doesn't get cast back to IntervalDtype
+        assert is_object_dtype(df["F"])
+
+        # they compare equal as Index
+        # when converted to numpy objects
+        c = lambda x: Index(np.array(x))
+        tm.assert_index_equal(c(df.B), c(df.B))
+        tm.assert_index_equal(c(df.B), c(df.C), check_names=False)
+        tm.assert_index_equal(c(df.B), c(df.D), check_names=False)
+        tm.assert_index_equal(c(df.C), c(df.D), check_names=False)
+
+        # B & D are the same Series
+        tm.assert_series_equal(df["B"], df["B"])
+        tm.assert_series_equal(df["B"], df["D"], check_names=False)
+
+        # C & E are the same Series
+        tm.assert_series_equal(df["C"], df["C"])
+        tm.assert_series_equal(df["C"], df["E"], check_names=False)
+
+    def test_setitem_categorical(self):
+        # GH#35369
+        df = DataFrame({"h": Series(list("mn")).astype("category")})
+        df.h = df.h.cat.reorder_categories(["n", "m"])
+        expected = DataFrame(
+            {"h": Categorical(["m", "n"]).reorder_categories(["n", "m"])}
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_with_empty_listlike(self):
+        # GH#17101
+        index = Index([], name="idx")
+        result = DataFrame(columns=["A"], index=index)
+        result["A"] = []
+        expected = DataFrame(columns=["A"], index=index)
+        tm.assert_index_equal(result.index, expected.index)
+
+    @pytest.mark.parametrize(
+        "cols, values, expected",
+        [
+            (["C", "D", "D", "a"], [1, 2, 3, 4], 4),  # with duplicates
+            (["D", "C", "D", "a"], [1, 2, 3, 4], 4),  # mixed order
+            (["C", "B", "B", "a"], [1, 2, 3, 4], 4),  # other duplicate cols
+            (["C", "B", "a"], [1, 2, 3], 3),  # no duplicates
+            (["B", "C", "a"], [3, 2, 1], 1),  # alphabetical order
+            (["C", "a", "B"], [3, 2, 1], 2),  # in the middle
+        ],
+    )
+    def test_setitem_same_column(self, cols, values, expected):
+        # GH#23239
+        df = DataFrame([values], columns=cols)
+        df["a"] = df["a"]
+        result = df["a"].values[0]
+        assert result == expected
+
+    def test_setitem_multi_index(self):
+        # GH#7655, test that assigning to a sub-frame of a frame
+        # with multi-index columns aligns both rows and columns
+        it = ["jim", "joe", "jolie"], ["first", "last"], ["left", "center", "right"]
+
+        cols = MultiIndex.from_product(it)
+        index = date_range("20141006", periods=20)
+        vals = np.random.default_rng(2).integers(1, 1000, (len(index), len(cols)))
+        df = DataFrame(vals, columns=cols, index=index)
+
+        i, j = df.index.values.copy(), it[-1][:]
+
+        np.random.default_rng(2).shuffle(i)
+        df["jim"] = df["jolie"].loc[i, ::-1]
+        tm.assert_frame_equal(df["jim"], df["jolie"])
+
+        np.random.default_rng(2).shuffle(j)
+        df[("joe", "first")] = df[("jolie", "last")].loc[i, j]
+        tm.assert_frame_equal(df[("joe", "first")], df[("jolie", "last")])
+
+        np.random.default_rng(2).shuffle(j)
+        df[("joe", "last")] = df[("jolie", "first")].loc[i, j]
+        tm.assert_frame_equal(df[("joe", "last")], df[("jolie", "first")])
+
+    @pytest.mark.parametrize(
+        "columns,box,expected",
+        [
+            (
+                ["A", "B", "C", "D"],
+                7,
+                DataFrame(
+                    [[7, 7, 7, 7], [7, 7, 7, 7], [7, 7, 7, 7]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                ["C", "D"],
+                [7, 8],
+                DataFrame(
+                    [[1, 2, 7, 8], [3, 4, 7, 8], [5, 6, 7, 8]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                ["A", "B", "C"],
+                np.array([7, 8, 9], dtype=np.int64),
+                DataFrame([[7, 8, 9], [7, 8, 9], [7, 8, 9]], columns=["A", "B", "C"]),
+            ),
+            (
+                ["B", "C", "D"],
+                [[7, 8, 9], [10, 11, 12], [13, 14, 15]],
+                DataFrame(
+                    [[1, 7, 8, 9], [3, 10, 11, 12], [5, 13, 14, 15]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                ["C", "A", "D"],
+                np.array([[7, 8, 9], [10, 11, 12], [13, 14, 15]], dtype=np.int64),
+                DataFrame(
+                    [[8, 2, 7, 9], [11, 4, 10, 12], [14, 6, 13, 15]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                ["A", "C"],
+                DataFrame([[7, 8], [9, 10], [11, 12]], columns=["A", "C"]),
+                DataFrame(
+                    [[7, 2, 8], [9, 4, 10], [11, 6, 12]], columns=["A", "B", "C"]
+                ),
+            ),
+        ],
+    )
+    def test_setitem_list_missing_columns(self, columns, box, expected):
+        # GH#29334
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], columns=["A", "B"])
+        df[columns] = box
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_list_of_tuples(self, float_frame):
+        tuples = list(zip(float_frame["A"], float_frame["B"]))
+        float_frame["tuples"] = tuples
+
+        result = float_frame["tuples"]
+        expected = Series(tuples, index=float_frame.index, name="tuples")
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_iloc_generator(self):
+        # GH#39614
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        indexer = (x for x in [1, 2])
+        df.iloc[indexer] = 1
+        expected = DataFrame({"a": [1, 1, 1], "b": [4, 1, 1]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_iloc_two_dimensional_generator(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        indexer = (x for x in [1, 2])
+        df.iloc[indexer, 1] = 1
+        expected = DataFrame({"a": [1, 2, 3], "b": [4, 1, 1]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_dtypes_bytes_type_to_object(self):
+        # GH 20734
+        index = Series(name="id", dtype="S24")
+        df = DataFrame(index=index, columns=Index([], dtype="str"))
+        df["a"] = Series(name="a", index=index, dtype=np.uint32)
+        df["b"] = Series(name="b", index=index, dtype="S64")
+        df["c"] = Series(name="c", index=index, dtype="S64")
+        df["d"] = Series(name="d", index=index, dtype=np.uint8)
+        result = df.dtypes
+        expected = Series([np.uint32, object, object, np.uint8], index=list("abcd"))
+        tm.assert_series_equal(result, expected)
+
+    def test_boolean_mask_nullable_int64(self):
+        # GH 28928
+        result = DataFrame({"a": [3, 4], "b": [5, 6]}).astype(
+            {"a": "int64", "b": "Int64"}
+        )
+        mask = Series(False, index=result.index)
+        result.loc[mask, "a"] = result["a"]
+        result.loc[mask, "b"] = result["b"]
+        expected = DataFrame({"a": [3, 4], "b": [5, 6]}).astype(
+            {"a": "int64", "b": "Int64"}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_ea_dtype_rhs_series(self):
+        # GH#47425
+        df = DataFrame({"a": [1, 2]})
+        df["a"] = Series([1, 2], dtype="Int64")
+        expected = DataFrame({"a": [1, 2]}, dtype="Int64")
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_npmatrix_2d(self):
+        # GH#42376
+        # for use-case df["x"] = sparse.random((10, 10)).mean(axis=1)
+        expected = DataFrame(
+            {"np-array": np.ones(10), "np-matrix": np.ones(10)}, index=np.arange(10)
+        )
+
+        a = np.ones((10, 1))
+        df = DataFrame(index=np.arange(10), columns=Index([], dtype="str"))
+        df["np-array"] = a
+
+        # Instantiation of `np.matrix` gives PendingDeprecationWarning
+        with tm.assert_produces_warning(
+            PendingDeprecationWarning,
+            match="matrix subclass is not the recommended way to represent matrices",
+        ):
+            df["np-matrix"] = np.matrix(a)
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("vals", [{}, {"d": "a"}])
+    def test_setitem_aligning_dict_with_index(self, vals):
+        # GH#47216
+        df = DataFrame({"a": [1, 2], "b": [3, 4], **vals})
+        df.loc[:, "a"] = {1: 100, 0: 200}
+        df.loc[:, "c"] = {0: 5, 1: 6}
+        df.loc[:, "e"] = {1: 5}
+        expected = DataFrame(
+            {"a": [200, 100], "b": [3, 4], **vals, "c": [5, 6], "e": [np.nan, 5]}
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_rhs_dataframe(self):
+        # GH#47578
+        df = DataFrame({"a": [1, 2]})
+        df["a"] = DataFrame({"a": [10, 11]}, index=[1, 2])
+        expected = DataFrame({"a": [np.nan, 10]})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame({"a": [1, 2]})
+        df.isetitem(0, DataFrame({"a": [10, 11]}, index=[1, 2]))
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_frame_overwrite_with_ea_dtype(self, any_numeric_ea_dtype):
+        # GH#46896
+        df = DataFrame(columns=["a", "b"], data=[[1, 2], [3, 4]])
+        df["a"] = DataFrame({"a": [10, 11]}, dtype=any_numeric_ea_dtype)
+        expected = DataFrame(
+            {
+                "a": Series([10, 11], dtype=any_numeric_ea_dtype),
+                "b": [2, 4],
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_string_option_object_index(self):
+        # GH#55638
+        pytest.importorskip("pyarrow")
+        df = DataFrame({"a": [1, 2]})
+        with pd.option_context("future.infer_string", True):
+            df["b"] = Index(["a", "b"], dtype=object)
+        expected = DataFrame({"a": [1, 2], "b": Series(["a", "b"], dtype=object)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_frame_midx_columns(self):
+        # GH#49121
+        df = DataFrame({("a", "b"): [10]})
+        expected = df.copy()
+        col_name = ("a", "b")
+        df[col_name] = df[[col_name]]
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_ea_dtype(self):
+        # GH#55604
+        df = DataFrame({"a": np.array([10], dtype="i8")})
+        df.loc[:, "a"] = Series([11], dtype="Int64")
+        expected = DataFrame({"a": np.array([11], dtype="i8")})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame({"a": np.array([10], dtype="i8")})
+        df.iloc[:, 0] = Series([11], dtype="Int64")
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_index_object_dtype_not_inferring(self):
+        # GH#56102
+        idx = Index([Timestamp("2019-12-31")], dtype=object)
+        df = DataFrame({"a": [1]})
+        df.loc[:, "b"] = idx
+        df["c"] = idx
+
+        expected = DataFrame(
+            {
+                "a": [1],
+                "b": idx,
+                "c": idx,
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+
+class TestSetitemTZAwareValues:
+    @pytest.fixture
+    def idx(self):
+        naive = DatetimeIndex(["2013-1-1 13:00", "2013-1-2 14:00"], name="B")
+        idx = naive.tz_localize("US/Pacific")
+        return idx
+
+    @pytest.fixture
+    def expected(self, idx):
+        expected = Series(np.array(idx.tolist(), dtype="object"), name="B")
+        assert expected.dtype == idx.dtype
+        return expected
+
+    def test_setitem_dt64series(self, idx, expected):
+        # convert to utc
+        df = DataFrame(np.random.default_rng(2).standard_normal((2, 1)), columns=["A"])
+        df["B"] = idx
+        df["B"] = idx.to_series(index=[0, 1]).dt.tz_convert(None)
+
+        result = df["B"]
+        comp = Series(idx.tz_convert("UTC").tz_localize(None), name="B")
+        tm.assert_series_equal(result, comp)
+
+    def test_setitem_datetimeindex(self, idx, expected):
+        # setting a DataFrame column with a tzaware DTI retains the dtype
+        df = DataFrame(np.random.default_rng(2).standard_normal((2, 1)), columns=["A"])
+
+        # assign to frame
+        df["B"] = idx
+        result = df["B"]
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_object_array_of_tzaware_datetimes(self, idx, expected):
+        # setting a DataFrame column with a tzaware DTI retains the dtype
+        df = DataFrame(np.random.default_rng(2).standard_normal((2, 1)), columns=["A"])
+
+        # object array of datetimes with a tz
+        df["B"] = idx.to_pydatetime()
+        result = df["B"]
+        expected = expected.dt.as_unit("us")
+        tm.assert_series_equal(result, expected)
+
+
+class TestDataFrameSetItemWithExpansion:
+    def test_setitem_listlike_views(self):
+        # GH#38148
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 4, 6]})
+
+        # get one column as a view of df
+        ser = df["a"]
+
+        # add columns with list-like indexer
+        df[["c", "d"]] = np.array([[0.1, 0.2], [0.3, 0.4], [0.4, 0.5]])
+
+        # edit in place the first column to check view semantics
+        df.iloc[0, 0] = 100
+
+        expected = Series([1, 2, 3], name="a")
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_string_column_numpy_dtype_raising(self):
+        # GH#39010
+        df = DataFrame([[1, 2], [3, 4]])
+        df["0 - Name"] = [5, 6]
+        expected = DataFrame([[1, 2, 5], [3, 4, 6]], columns=[0, 1, "0 - Name"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_empty_df_duplicate_columns(self):
+        # GH#38521
+        df = DataFrame(columns=["a", "b", "b"], dtype="float64")
+        df.loc[:, "a"] = list(range(2))
+        expected = DataFrame(
+            [[0, np.nan, np.nan], [1, np.nan, np.nan]], columns=["a", "b", "b"]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_with_expansion_categorical_dtype(self):
+        # assignment
+        df = DataFrame(
+            {
+                "value": np.array(
+                    np.random.default_rng(2).integers(0, 10000, 100), dtype="int32"
+                )
+            }
+        )
+        labels = Categorical([f"{i} - {i + 499}" for i in range(0, 10000, 500)])
+
+        df = df.sort_values(by=["value"], ascending=True)
+        ser = cut(df.value, range(0, 10500, 500), right=False, labels=labels)
+        cat = ser.values
+
+        # setting with a Categorical
+        df["D"] = cat
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("int32"), CategoricalDtype(categories=labels, ordered=False)],
+            index=["value", "D"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # setting with a Series
+        df["E"] = ser
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("int32"),
+                CategoricalDtype(categories=labels, ordered=False),
+                CategoricalDtype(categories=labels, ordered=False),
+            ],
+            index=["value", "D", "E"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        result1 = df["D"]
+        result2 = df["E"]
+        tm.assert_categorical_equal(result1._mgr.array, cat)
+
+        # sorting
+        ser.name = "E"
+        tm.assert_series_equal(result2.sort_index(), ser.sort_index())
+
+    def test_setitem_scalars_no_index(self):
+        # GH#16823 / GH#17894
+        df = DataFrame()
+        df["foo"] = 1
+        expected = DataFrame(columns=["foo"]).astype(np.int64)
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_newcol_tuple_key(self, float_frame):
+        assert (
+            "A",
+            "B",
+        ) not in float_frame.columns
+        float_frame["A", "B"] = float_frame["A"]
+        assert ("A", "B") in float_frame.columns
+
+        result = float_frame["A", "B"]
+        expected = float_frame["A"]
+        tm.assert_series_equal(result, expected, check_names=False)
+
+    def test_frame_setitem_newcol_timestamp(self):
+        # GH#2155
+        columns = date_range(start="1/1/2012", end="2/1/2012", freq=BDay())
+        data = DataFrame(columns=columns, index=range(10))
+        t = datetime(2012, 11, 1)
+        ts = Timestamp(t)
+        data[ts] = np.nan  # works, mostly a smoke-test
+        assert np.isnan(data[ts]).all()
+
+    def test_frame_setitem_rangeindex_into_new_col(self):
+        # GH#47128
+        df = DataFrame({"a": ["a", "b"]})
+        df["b"] = df.index
+        df.loc[[False, True], "b"] = 100
+        result = df.loc[[1], :]
+        expected = DataFrame({"a": ["b"], "b": [100]}, index=[1])
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_frame_keep_ea_dtype(self, any_numeric_ea_dtype):
+        # GH#46896
+        df = DataFrame(columns=["a", "b"], data=[[1, 2], [3, 4]])
+        df["c"] = DataFrame({"a": [10, 11]}, dtype=any_numeric_ea_dtype)
+        expected = DataFrame(
+            {
+                "a": [1, 3],
+                "b": [2, 4],
+                "c": Series([10, 11], dtype=any_numeric_ea_dtype),
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_expansion_with_timedelta_type(self):
+        result = DataFrame(columns=list("abc"))
+        result.loc[0] = {
+            "a": pd.to_timedelta(5, unit="s"),
+            "b": pd.to_timedelta(72, unit="s"),
+            "c": "23",
+        }
+        expected = DataFrame(
+            [[pd.Timedelta("0 days 00:00:05"), pd.Timedelta("0 days 00:01:12"), "23"]],
+            index=Index([0]),
+            columns=(["a", "b", "c"]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameSetItemSlicing:
+    def test_setitem_slice_position(self):
+        # GH#31469
+        df = DataFrame(np.zeros((100, 1)))
+        df[-4:] = 1
+        arr = np.zeros((100, 1))
+        arr[-4:] = 1
+        expected = DataFrame(arr)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("box", [Series, np.array, list, pd.array])
+    @pytest.mark.parametrize("n", [1, 2, 3])
+    def test_setitem_slice_indexer_broadcasting_rhs(self, n, box, indexer_si):
+        # GH#40440
+        df = DataFrame([[1, 3, 5]] + [[2, 4, 6]] * n, columns=["a", "b", "c"])
+        indexer_si(df)[1:] = box([10, 11, 12])
+        expected = DataFrame([[1, 3, 5]] + [[10, 11, 12]] * n, columns=["a", "b", "c"])
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("box", [Series, np.array, list, pd.array])
+    @pytest.mark.parametrize("n", [1, 2, 3])
+    def test_setitem_list_indexer_broadcasting_rhs(self, n, box):
+        # GH#40440
+        df = DataFrame([[1, 3, 5]] + [[2, 4, 6]] * n, columns=["a", "b", "c"])
+        df.iloc[list(range(1, n + 1))] = box([10, 11, 12])
+        expected = DataFrame([[1, 3, 5]] + [[10, 11, 12]] * n, columns=["a", "b", "c"])
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("box", [Series, np.array, list, pd.array])
+    @pytest.mark.parametrize("n", [1, 2, 3])
+    def test_setitem_slice_broadcasting_rhs_mixed_dtypes(self, n, box, indexer_si):
+        # GH#40440
+        df = DataFrame(
+            [[1, 3, 5], ["x", "y", "z"]] + [[2, 4, 6]] * n, columns=["a", "b", "c"]
+        )
+        indexer_si(df)[1:] = box([10, 11, 12])
+        expected = DataFrame(
+            [[1, 3, 5]] + [[10, 11, 12]] * (n + 1),
+            columns=["a", "b", "c"],
+            dtype="object",
+        )
+        tm.assert_frame_equal(df, expected)
+
+
+class TestDataFrameSetItemCallable:
+    def test_setitem_callable(self):
+        # GH#12533
+        df = DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8]})
+        df[lambda x: "A"] = [11, 12, 13, 14]
+
+        exp = DataFrame({"A": [11, 12, 13, 14], "B": [5, 6, 7, 8]})
+        tm.assert_frame_equal(df, exp)
+
+    def test_setitem_other_callable(self):
+        # GH#13299
+        def inc(x):
+            return x + 1
+
+        # Set dtype object straight away to avoid upcast when setting inc below
+        df = DataFrame([[-1, 1], [1, -1]], dtype=object)
+        df[df > 0] = inc
+
+        expected = DataFrame([[-1, inc], [inc, -1]])
+        tm.assert_frame_equal(df, expected)
+
+
+class TestDataFrameSetItemBooleanMask:
+    @pytest.mark.parametrize(
+        "mask_type",
+        [lambda df: df > np.abs(df) / 2, lambda df: (df > np.abs(df) / 2).values],
+        ids=["dataframe", "array"],
+    )
+    def test_setitem_boolean_mask(self, mask_type, float_frame):
+        # Test for issue #18582
+        df = float_frame.copy()
+        mask = mask_type(df)
+
+        # index with boolean mask
+        result = df.copy()
+        result[mask] = np.nan
+
+        expected = df.values.copy()
+        expected[np.array(mask)] = np.nan
+        expected = DataFrame(expected, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.xfail(reason="Currently empty indexers are treated as all False")
+    @pytest.mark.parametrize("box", [list, np.array, Series])
+    def test_setitem_loc_empty_indexer_raises_with_non_empty_value(self, box):
+        # GH#37672
+        df = DataFrame({"a": ["a"], "b": [1], "c": [1]})
+        if box == Series:
+            indexer = box([], dtype="object")
+        else:
+            indexer = box([])
+        msg = "Must have equal len keys and value when setting with an iterable"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[indexer, ["b"]] = [1]
+
+    @pytest.mark.parametrize("box", [list, np.array, Series])
+    def test_setitem_loc_only_false_indexer_dtype_changed(self, box):
+        # GH#37550
+        # Dtype is only changed when value to set is a Series and indexer is
+        # empty/bool all False
+        df = DataFrame({"a": ["a"], "b": [1], "c": [1]})
+        indexer = box([False])
+        df.loc[indexer, ["b"]] = 10 - df["c"]
+        expected = DataFrame({"a": ["a"], "b": [1], "c": [1]})
+        tm.assert_frame_equal(df, expected)
+
+        df.loc[indexer, ["b"]] = 9
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_boolean_mask_aligning(self, indexer_sl):
+        # GH#39931
+        df = DataFrame({"a": [1, 4, 2, 3], "b": [5, 6, 7, 8]})
+        expected = df.copy()
+        mask = df["a"] >= 3
+        indexer_sl(df)[mask] = indexer_sl(df)[mask].sort_values("a")
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_mask_categorical(self):
+        # assign multiple rows (mixed values) (-> array) -> exp_multi_row
+        # changed multiple rows
+        cats2 = Categorical(["a", "a", "b", "b", "a", "a", "a"], categories=["a", "b"])
+        idx2 = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values2 = [1, 1, 2, 2, 1, 1, 1]
+        exp_multi_row = DataFrame({"cats": cats2, "values": values2}, index=idx2)
+
+        catsf = Categorical(
+            ["a", "a", "c", "c", "a", "a", "a"], categories=["a", "b", "c"]
+        )
+        idxf = Index(["h", "i", "j", "k", "l", "m", "n"])
+        valuesf = [1, 1, 3, 3, 1, 1, 1]
+        df = DataFrame({"cats": catsf, "values": valuesf}, index=idxf)
+
+        exp_fancy = exp_multi_row.copy()
+        exp_fancy["cats"] = exp_fancy["cats"].cat.set_categories(["a", "b", "c"])
+
+        mask = df["cats"] == "c"
+        df[mask] = ["b", 2]
+        # category c is kept in .categories
+        tm.assert_frame_equal(df, exp_fancy)
+
+    @pytest.mark.parametrize("dtype", ["float", "int64"])
+    @pytest.mark.parametrize("kwargs", [{}, {"index": [1]}, {"columns": ["A"]}])
+    def test_setitem_empty_frame_with_boolean(self, dtype, kwargs):
+        # see GH#10126
+        kwargs["dtype"] = dtype
+        df = DataFrame(**kwargs)
+
+        df2 = df.copy()
+        df[df > df2] = 47
+        tm.assert_frame_equal(df, df2)
+
+    def test_setitem_boolean_indexing(self):
+        idx = list(range(3))
+        cols = ["A", "B", "C"]
+        df1 = DataFrame(
+            index=idx,
+            columns=cols,
+            data=np.array(
+                [[0.0, 0.5, 1.0], [1.5, 2.0, 2.5], [3.0, 3.5, 4.0]], dtype=float
+            ),
+        )
+        df2 = DataFrame(index=idx, columns=cols, data=np.ones((len(idx), len(cols))))
+
+        expected = DataFrame(
+            index=idx,
+            columns=cols,
+            data=np.array([[0.0, 0.5, 1.0], [1.5, 2.0, -1], [-1, -1, -1]], dtype=float),
+        )
+
+        df1[df1 > 2.0 * df2] = -1
+        tm.assert_frame_equal(df1, expected)
+        with pytest.raises(ValueError, match="Item wrong length"):
+            df1[df1.index[:-1] > 2] = -1
+
+    def test_loc_setitem_all_false_boolean_two_blocks(self):
+        # GH#40885
+        df = DataFrame({"a": [1, 2], "b": [3, 4], "c": "a"})
+        expected = df.copy()
+        indexer = Series([False, False], name="c")
+        df.loc[indexer, ["b"]] = DataFrame({"b": [5, 6]}, index=[0, 1])
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_ea_boolean_mask(self):
+        # GH#47125
+        df = DataFrame([[-1, 2], [3, -4]])
+        expected = DataFrame([[0, 2], [3, 0]])
+        boolean_indexer = DataFrame(
+            {
+                0: Series([True, False], dtype="boolean"),
+                1: Series([pd.NA, True], dtype="boolean"),
+            }
+        )
+        df[boolean_indexer] = 0
+        tm.assert_frame_equal(df, expected)
+
+
+class TestDataFrameSetitemCopyViewSemantics:
+    def test_setitem_always_copy(self, float_frame):
+        assert "E" not in float_frame.columns
+        s = float_frame["A"].copy()
+        float_frame["E"] = s
+
+        float_frame.iloc[5:10, float_frame.columns.get_loc("E")] = np.nan
+        assert notna(s[5:10]).all()
+
+    @pytest.mark.parametrize("consolidate", [True, False])
+    def test_setitem_partial_column_inplace(self, consolidate):
+        # This setting should be in-place, regardless of whether frame is
+        #  single-block or multi-block
+        # GH#304 this used to be incorrectly not-inplace, in which case
+        #  we needed to ensure _item_cache was cleared.
+
+        df = DataFrame(
+            {"x": [1.1, 2.1, 3.1, 4.1], "y": [5.1, 6.1, 7.1, 8.1]}, index=[0, 1, 2, 3]
+        )
+        df.insert(2, "z", np.nan)
+        if consolidate:
+            df._consolidate_inplace()
+            assert len(df._mgr.blocks) == 1
+        else:
+            assert len(df._mgr.blocks) == 2
+
+        df.loc[2:, "z"] = 42
+
+        expected = Series([np.nan, np.nan, 42, 42], index=df.index, name="z")
+        tm.assert_series_equal(df["z"], expected)
+
+    def test_setitem_duplicate_columns_not_inplace(self):
+        # GH#39510
+        cols = ["A", "B"] * 2
+        df = DataFrame(0.0, index=[0], columns=cols)
+        df_copy = df.copy()
+        df_view = df[:]
+        df["B"] = (2, 5)
+
+        expected = DataFrame([[0.0, 2, 0.0, 5]], columns=cols)
+        tm.assert_frame_equal(df_view, df_copy)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "value", [1, np.array([[1], [1]], dtype="int64"), [[1], [1]]]
+    )
+    def test_setitem_same_dtype_not_inplace(self, value):
+        # GH#39510
+        cols = ["A", "B"]
+        df = DataFrame(0, index=[0, 1], columns=cols)
+        df_copy = df.copy()
+        df_view = df[:]
+        df[["B"]] = value
+
+        expected = DataFrame([[0, 1], [0, 1]], columns=cols)
+        tm.assert_frame_equal(df, expected)
+        tm.assert_frame_equal(df_view, df_copy)
+
+    @pytest.mark.parametrize("value", [1.0, np.array([[1.0], [1.0]]), [[1.0], [1.0]]])
+    def test_setitem_listlike_key_scalar_value_not_inplace(self, value):
+        # GH#39510
+        cols = ["A", "B"]
+        df = DataFrame(0, index=[0, 1], columns=cols)
+        df_copy = df.copy()
+        df_view = df[:]
+        df[["B"]] = value
+
+        expected = DataFrame([[0, 1.0], [0, 1.0]], columns=cols)
+        tm.assert_frame_equal(df_view, df_copy)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "indexer",
+        [
+            "a",
+            ["a"],
+            pytest.param(
+                [True, False],
+                marks=pytest.mark.xfail(
+                    reason="Boolean indexer incorrectly setting inplace",
+                    strict=False,  # passing on some builds, no obvious pattern
+                ),
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "value, set_value",
+        [
+            (1, 5),
+            (1.0, 5.0),
+            (Timestamp("2020-12-31"), Timestamp("2021-12-31")),
+            ("a", "b"),
+        ],
+    )
+    def test_setitem_not_operating_inplace(self, value, set_value, indexer):
+        # GH#43406
+        df = DataFrame({"a": value}, index=[0, 1])
+        expected = df.copy()
+        view = df[:]
+        df[indexer] = set_value
+        tm.assert_frame_equal(view, expected)
+
+    def test_setitem_column_update_inplace(self):
+        # https://github.com/pandas-dev/pandas/issues/47172
+
+        labels = [f"c{i}" for i in range(10)]
+        df = DataFrame({col: np.zeros(len(labels)) for col in labels}, index=labels)
+        values = df._mgr.blocks[0].values
+
+        with tm.raises_chained_assignment_error():
+            for label in df.columns:
+                df[label][label] = 1
+        # original dataframe not updated
+        assert np.all(values[np.arange(10), np.arange(10)] == 0)
+
+    def test_setitem_column_frame_as_category(self):
+        # GH31581
+        df = DataFrame([1, 2, 3])
+        df["col1"] = DataFrame([1, 2, 3], dtype="category")
+        df["col2"] = Series([1, 2, 3], dtype="category")
+
+        expected_types = Series(
+            ["int64", "category", "category"], index=[0, "col1", "col2"], dtype=object
+        )
+        tm.assert_series_equal(df.dtypes, expected_types)
+
+    @pytest.mark.parametrize("dtype", ["int64", "Int64"])
+    def test_setitem_iloc_with_numpy_array(self, dtype):
+        # GH-33828
+        df = DataFrame({"a": np.ones(3)}, dtype=dtype)
+        df.iloc[np.array([0]), np.array([0])] = np.array([[2]])
+
+        expected = DataFrame({"a": [2, 1, 1]}, dtype=dtype)
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_frame_dup_cols_dtype(self):
+        # GH#53143
+        df = DataFrame([[1, 2, 3, 4], [4, 5, 6, 7]], columns=["a", "b", "a", "c"])
+        rhs = DataFrame([[0, 1.5], [2, 2.5]], columns=["a", "a"])
+        df["a"] = rhs
+        expected = DataFrame(
+            [[0, 2, 1.5, 4], [2, 5, 2.5, 7]], columns=["a", "b", "a", "c"]
+        )
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["a", "a", "b"])
+        rhs = DataFrame([[0, 1.5], [2, 2.5]], columns=["a", "a"])
+        df["a"] = rhs
+        expected = DataFrame([[0, 1.5, 3], [2, 2.5, 6]], columns=["a", "a", "b"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_frame_setitem_empty_dataframe(self):
+        # GH#28871
+        dti = DatetimeIndex(["2000-01-01"], dtype="M8[ns]", name="date")
+        df = DataFrame({"date": dti}).set_index("date")
+        df = df[0:0].copy()
+
+        df["3010"] = None
+        df["2010"] = None
+
+        expected = DataFrame(
+            [],
+            columns=["3010", "2010"],
+            index=dti[:0],
+        )
+        tm.assert_frame_equal(df, expected)
+
+
+def test_full_setter_loc_incompatible_dtype():
+    # https://github.com/pandas-dev/pandas/issues/55791
+    df = DataFrame({"a": [1, 2]})
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.loc[:, "a"] = True
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.loc[:, "a"] = {0: 3.5, 1: 4.5}
+
+    df.loc[:, "a"] = {0: 3, 1: 4}
+    expected = DataFrame({"a": [3, 4]})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_setitem_partial_row_multiple_columns():
+    # https://github.com/pandas-dev/pandas/issues/56503
+    df = DataFrame({"A": [1, 2, 3], "B": [4.0, 5, 6]})
+    # should not warn
+    df.loc[df.index <= 1, ["F", "G"]] = (1, "abc")
+    expected = DataFrame(
+        {
+            "A": [1, 2, 3],
+            "B": [4.0, 5, 6],
+            "F": [1.0, 1, float("nan")],
+            "G": ["abc", "abc", float("nan")],
+        }
+    )
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/indexing/test_take.py b/pandas/pandas/tests/frame/indexing/test_take.py
new file mode 100644
index 0000000000000..56a7bfd003083
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_take.py
@@ -0,0 +1,92 @@
+import pytest
+
+import pandas._testing as tm
+
+
+class TestDataFrameTake:
+    def test_take_slices_not_supported(self, float_frame):
+        # GH#51539
+        df = float_frame
+
+        slc = slice(0, 4, 1)
+        with pytest.raises(TypeError, match="slice"):
+            df.take(slc, axis=0)
+        with pytest.raises(TypeError, match="slice"):
+            df.take(slc, axis=1)
+
+    def test_take(self, float_frame):
+        # homogeneous
+        order = [3, 1, 2, 0]
+        for df in [float_frame]:
+            result = df.take(order, axis=0)
+            expected = df.reindex(df.index.take(order))
+            tm.assert_frame_equal(result, expected)
+
+            # axis = 1
+            result = df.take(order, axis=1)
+            expected = df.loc[:, ["D", "B", "C", "A"]]
+            tm.assert_frame_equal(result, expected, check_names=False)
+
+        # negative indices
+        order = [2, 1, -1]
+        for df in [float_frame]:
+            result = df.take(order, axis=0)
+            expected = df.reindex(df.index.take(order))
+            tm.assert_frame_equal(result, expected)
+
+            result = df.take(order, axis=0)
+            tm.assert_frame_equal(result, expected)
+
+            # axis = 1
+            result = df.take(order, axis=1)
+            expected = df.loc[:, ["C", "B", "D"]]
+            tm.assert_frame_equal(result, expected, check_names=False)
+
+        # illegal indices
+        msg = "indices are out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            df.take([3, 1, 2, 30], axis=0)
+        with pytest.raises(IndexError, match=msg):
+            df.take([3, 1, 2, -31], axis=0)
+        with pytest.raises(IndexError, match=msg):
+            df.take([3, 1, 2, 5], axis=1)
+        with pytest.raises(IndexError, match=msg):
+            df.take([3, 1, 2, -5], axis=1)
+
+    def test_take_mixed_type(self, float_string_frame):
+        # mixed-dtype
+        order = [4, 1, 2, 0, 3]
+        for df in [float_string_frame]:
+            result = df.take(order, axis=0)
+            expected = df.reindex(df.index.take(order))
+            tm.assert_frame_equal(result, expected)
+
+            # axis = 1
+            result = df.take(order, axis=1)
+            expected = df.loc[:, ["foo", "B", "C", "A", "D"]]
+            tm.assert_frame_equal(result, expected)
+
+        # negative indices
+        order = [4, 1, -2]
+        for df in [float_string_frame]:
+            result = df.take(order, axis=0)
+            expected = df.reindex(df.index.take(order))
+            tm.assert_frame_equal(result, expected)
+
+            # axis = 1
+            result = df.take(order, axis=1)
+            expected = df.loc[:, ["foo", "B", "D"]]
+            tm.assert_frame_equal(result, expected)
+
+    def test_take_mixed_numeric(self, mixed_float_frame, mixed_int_frame):
+        # by dtype
+        order = [1, 2, 0, 3]
+        for df in [mixed_float_frame, mixed_int_frame]:
+            result = df.take(order, axis=0)
+            expected = df.reindex(df.index.take(order))
+            tm.assert_frame_equal(result, expected)
+
+            # axis = 1
+            result = df.take(order, axis=1)
+            expected = df.loc[:, ["B", "C", "A", "D"]]
+            tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/indexing/test_where.py b/pandas/pandas/tests/frame/indexing/test_where.py
new file mode 100644
index 0000000000000..d6570fcda2ee8
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_where.py
@@ -0,0 +1,1043 @@
+from datetime import datetime
+
+from hypothesis import given
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_scalar
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Series,
+    StringDtype,
+    Timestamp,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+from pandas._testing._hypothesis import OPTIONAL_ONE_OF_ALL
+
+
+@pytest.fixture(params=["default", "float_string", "mixed_float", "mixed_int"])
+def where_frame(request, float_string_frame, mixed_float_frame, mixed_int_frame):
+    if request.param == "default":
+        return DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=["A", "B", "C"]
+        )
+    if request.param == "float_string":
+        return float_string_frame
+    if request.param == "mixed_float":
+        return mixed_float_frame
+    if request.param == "mixed_int":
+        return mixed_int_frame
+
+
+def _safe_add(df):
+    # only add to the numeric items
+    def is_ok(s):
+        return (
+            issubclass(s.dtype.type, (np.integer, np.floating)) and s.dtype != "uint8"
+        )
+
+    return DataFrame(dict((c, s + 1) if is_ok(s) else (c, s) for c, s in df.items()))
+
+
+class TestDataFrameIndexingWhere:
+    def test_where_get(self, where_frame, float_string_frame):
+        def _check_get(df, cond, check_dtypes=True):
+            other1 = _safe_add(df)
+            rs = df.where(cond, other1)
+            rs2 = df.where(cond.values, other1)
+            for k, v in rs.items():
+                exp = Series(np.where(cond[k], df[k], other1[k]), index=v.index)
+                tm.assert_series_equal(v, exp, check_names=False)
+            tm.assert_frame_equal(rs, rs2)
+
+            # dtypes
+            if check_dtypes:
+                assert (rs.dtypes == df.dtypes).all()
+
+        # check getting
+        df = where_frame
+        if df is float_string_frame:
+            msg = (
+                "'>' not supported between instances of 'str' and 'int'"
+                "|Invalid comparison"
+            )
+            with pytest.raises(TypeError, match=msg):
+                df > 0
+            return
+        cond = df > 0
+        _check_get(df, cond)
+
+    def test_where_upcasting(self):
+        # upcasting case (GH # 2794)
+        df = DataFrame(
+            {
+                c: Series([1] * 3, dtype=c)
+                for c in ["float32", "float64", "int32", "int64"]
+            }
+        )
+        df.iloc[1, :] = 0
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("float32"),
+                np.dtype("float64"),
+                np.dtype("int32"),
+                np.dtype("int64"),
+            ],
+            index=["float32", "float64", "int32", "int64"],
+        )
+
+        # when we don't preserve boolean casts
+        #
+        # expected = Series({ 'float32' : 1, 'float64' : 3 })
+
+        tm.assert_series_equal(result, expected)
+
+    def test_where_alignment(self, where_frame, float_string_frame):
+        # aligning
+        def _check_align(df, cond, other, check_dtypes=True):
+            rs = df.where(cond, other)
+            for i, k in enumerate(rs.columns):
+                result = rs[k]
+                d = df[k].values
+                c = cond[k].reindex(df[k].index).fillna(False).values
+
+                if is_scalar(other):
+                    o = other
+                elif isinstance(other, np.ndarray):
+                    o = Series(other[:, i], index=result.index).values
+                else:
+                    o = other[k].values
+
+                new_values = d if c.all() else np.where(c, d, o)
+                expected = Series(new_values, index=result.index, name=k)
+
+                # since we can't always have the correct numpy dtype
+                # as numpy doesn't know how to downcast, don't check
+                tm.assert_series_equal(result, expected, check_dtype=False)
+
+            # dtypes
+            # can't check dtype when other is an ndarray
+
+            if check_dtypes and not isinstance(other, np.ndarray):
+                assert (rs.dtypes == df.dtypes).all()
+
+        df = where_frame
+        if df is float_string_frame:
+            msg = (
+                "'>' not supported between instances of 'str' and 'int'"
+                "|Invalid comparison"
+            )
+            with pytest.raises(TypeError, match=msg):
+                df > 0
+            return
+
+        # other is a frame
+        cond = (df > 0)[1:]
+        _check_align(df, cond, _safe_add(df))
+
+        # check other is ndarray
+        cond = df > 0
+        _check_align(df, cond, (_safe_add(df).values))
+
+        # integers are upcast, so don't check the dtypes
+        cond = df > 0
+        check_dtypes = all(not issubclass(s.type, np.integer) for s in df.dtypes)
+        _check_align(df, cond, np.nan, check_dtypes=check_dtypes)
+
+    # Ignore deprecation warning in Python 3.12 for inverting a bool
+    @pytest.mark.filterwarnings("ignore::DeprecationWarning")
+    def test_where_invalid(self):
+        # invalid conditions
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=["A", "B", "C"]
+        )
+        cond = df > 0
+
+        err1 = (df + 1).values[0:2, :]
+        msg = "other must be the same shape as self when an ndarray"
+        with pytest.raises(ValueError, match=msg):
+            df.where(cond, err1)
+
+        err2 = cond.iloc[:2, :].values
+        other1 = _safe_add(df)
+        msg = "Array conditional must be same shape as self"
+        with pytest.raises(ValueError, match=msg):
+            df.where(err2, other1)
+
+        with pytest.raises(ValueError, match=msg):
+            df.mask(True)
+        with pytest.raises(ValueError, match=msg):
+            df.mask(0)
+
+    def test_where_set(self, where_frame, float_string_frame, mixed_int_frame):
+        # where inplace
+
+        def _check_set(df, cond, check_dtypes=True):
+            dfi = df.copy()
+            econd = cond.reindex_like(df).fillna(True).infer_objects()
+            expected = dfi.mask(~econd)
+
+            return_value = dfi.where(cond, np.nan, inplace=True)
+            assert return_value is None
+            tm.assert_frame_equal(dfi, expected)
+
+            # dtypes (and confirm upcasts)x
+            if check_dtypes:
+                for k, v in df.dtypes.items():
+                    if issubclass(v.type, np.integer) and not cond[k].all():
+                        v = np.dtype("float64")
+                    assert dfi[k].dtype == v
+
+        df = where_frame
+        if df is float_string_frame:
+            msg = (
+                "'>' not supported between instances of 'str' and 'int'"
+                "|Invalid comparison"
+            )
+            with pytest.raises(TypeError, match=msg):
+                df > 0
+            return
+        if df is mixed_int_frame:
+            df = df.astype("float64")
+
+        cond = df > 0
+        _check_set(df, cond)
+
+        cond = df >= 0
+        _check_set(df, cond)
+
+        # aligning
+        cond = (df >= 0)[1:]
+        _check_set(df, cond)
+
+    def test_where_series_slicing(self):
+        # GH 10218
+        # test DataFrame.where with Series slicing
+        df = DataFrame({"a": range(3), "b": range(4, 7)})
+        result = df.where(df["a"] == 1)
+        expected = df[df["a"] == 1].reindex(df.index)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [list, tuple, np.array])
+    def test_where_array_like(self, klass):
+        # see gh-15414
+        df = DataFrame({"a": [1, 2, 3]})
+        cond = [[False], [True], [True]]
+        expected = DataFrame({"a": [np.nan, 2, 3]})
+
+        result = df.where(klass(cond))
+        tm.assert_frame_equal(result, expected)
+
+        df["b"] = 2
+        expected["b"] = [2, np.nan, 2]
+        cond = [[False, True], [True, False], [True, True]]
+
+        result = df.where(klass(cond))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "cond",
+        [
+            [[1], [0], [1]],
+            Series([[2], [5], [7]]),
+            DataFrame({"a": [2, 5, 7]}),
+            [["True"], ["False"], ["True"]],
+            [[Timestamp("2017-01-01")], [pd.NaT], [Timestamp("2017-01-02")]],
+        ],
+    )
+    def test_where_invalid_input_single(self, cond):
+        # see gh-15414: only boolean arrays accepted
+        df = DataFrame({"a": [1, 2, 3]})
+        msg = "Boolean array expected for the condition"
+
+        with pytest.raises(TypeError, match=msg):
+            df.where(cond)
+
+    @pytest.mark.parametrize(
+        "cond",
+        [
+            [[0, 1], [1, 0], [1, 1]],
+            Series([[0, 2], [5, 0], [4, 7]]),
+            [["False", "True"], ["True", "False"], ["True", "True"]],
+            DataFrame({"a": [2, 5, 7], "b": [4, 8, 9]}),
+            [
+                [pd.NaT, Timestamp("2017-01-01")],
+                [Timestamp("2017-01-02"), pd.NaT],
+                [Timestamp("2017-01-03"), Timestamp("2017-01-03")],
+            ],
+        ],
+    )
+    def test_where_invalid_input_multiple(self, cond):
+        # see gh-15414: only boolean arrays accepted
+        df = DataFrame({"a": [1, 2, 3], "b": [2, 2, 2]})
+        msg = "Boolean array expected for the condition"
+
+        with pytest.raises(TypeError, match=msg):
+            df.where(cond)
+
+    def test_where_dataframe_col_match(self):
+        df = DataFrame([[1, 2, 3], [4, 5, 6]])
+        cond = DataFrame([[True, False, True], [False, False, True]])
+
+        result = df.where(cond)
+        expected = DataFrame([[1.0, np.nan, 3], [np.nan, np.nan, 6]])
+        tm.assert_frame_equal(result, expected)
+
+        # this *does* align, though has no matching columns
+        cond.columns = ["a", "b", "c"]
+        result = df.where(cond)
+        expected = DataFrame(np.nan, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_ndframe_align(self):
+        msg = "Array conditional must be same shape as self"
+        df = DataFrame([[1, 2, 3], [4, 5, 6]])
+
+        cond = [True]
+        with pytest.raises(ValueError, match=msg):
+            df.where(cond)
+
+        expected = DataFrame([[1, 2, 3], [np.nan, np.nan, np.nan]])
+
+        out = df.where(Series(cond))
+        tm.assert_frame_equal(out, expected)
+
+        cond = np.array([False, True, False, True])
+        with pytest.raises(ValueError, match=msg):
+            df.where(cond)
+
+        expected = DataFrame([[np.nan, np.nan, np.nan], [4, 5, 6]])
+
+        out = df.where(Series(cond))
+        tm.assert_frame_equal(out, expected)
+
+    def test_where_bug(self):
+        # see gh-2793
+        df = DataFrame(
+            {"a": [1.0, 2.0, 3.0, 4.0], "b": [4.0, 3.0, 2.0, 1.0]}, dtype="float64"
+        )
+        expected = DataFrame(
+            {"a": [np.nan, np.nan, 3.0, 4.0], "b": [4.0, 3.0, np.nan, np.nan]},
+            dtype="float64",
+        )
+        result = df.where(df > 2, np.nan)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(result > 2, np.nan, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_bug_mixed(self, any_signed_int_numpy_dtype):
+        # see gh-2793
+        df = DataFrame(
+            {
+                "a": np.array([1, 2, 3, 4], dtype=any_signed_int_numpy_dtype),
+                "b": np.array([4.0, 3.0, 2.0, 1.0], dtype="float64"),
+            }
+        )
+
+        expected = DataFrame(
+            {"a": [-1, -1, 3, 4], "b": [4.0, 3.0, -1, -1]},
+        ).astype({"a": any_signed_int_numpy_dtype, "b": "float64"})
+
+        result = df.where(df > 2, -1)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(result > 2, -1, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_bug_transposition(self):
+        # see gh-7506
+        a = DataFrame({0: [1, 2], 1: [3, 4], 2: [5, 6]})
+        b = DataFrame({0: [np.nan, 8], 1: [9, np.nan], 2: [np.nan, np.nan]})
+        do_not_replace = b.isna() | (a > b)
+
+        expected = a.copy()
+        expected[~do_not_replace] = b
+        expected[[0, 1]] = expected[[0, 1]].astype("float64")
+
+        result = a.where(do_not_replace, b)
+        tm.assert_frame_equal(result, expected)
+
+        a = DataFrame({0: [4, 6], 1: [1, 0]})
+        b = DataFrame({0: [np.nan, 3], 1: [3, np.nan]})
+        do_not_replace = b.isna() | (a > b)
+
+        expected = a.copy()
+        expected[~do_not_replace] = b
+        expected[1] = expected[1].astype("float64")
+
+        result = a.where(do_not_replace, b)
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_datetime(self):
+        # GH 3311
+        df = DataFrame(
+            {
+                "A": date_range("20130102", periods=5),
+                "B": date_range("20130104", periods=5),
+                "C": np.random.default_rng(2).standard_normal(5),
+            }
+        )
+
+        stamp = datetime(2013, 1, 3)
+        msg = "'>' not supported between instances of 'float' and 'datetime.datetime'"
+        with pytest.raises(TypeError, match=msg):
+            df > stamp
+
+        result = df[df.iloc[:, :-1] > stamp]
+
+        expected = df.copy()
+        expected.loc[[0, 1], "A"] = np.nan
+
+        expected.loc[:, "C"] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_none(self):
+        # GH 4667
+        # setting with None changes dtype
+        df = DataFrame({"series": Series(range(10))}).astype(float)
+        df[df > 7] = None
+        expected = DataFrame(
+            {"series": Series([0, 1, 2, 3, 4, 5, 6, 7, np.nan, np.nan])}
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # GH 7656
+        df = DataFrame(
+            [
+                {"A": 1, "B": np.nan, "C": "Test"},
+                {"A": np.nan, "B": "Test", "C": np.nan},
+            ]
+        )
+
+        orig = df.copy()
+
+        mask = ~isna(df)
+        df.where(mask, None, inplace=True)
+        expected = DataFrame(
+            {
+                "A": [1.0, np.nan],
+                "B": [None, "Test"],
+                "C": ["Test", None],
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+        df = orig.copy()
+        df[~mask] = None
+        tm.assert_frame_equal(df, expected)
+
+    def test_where_empty_df_and_empty_cond_having_non_bool_dtypes(self):
+        # see gh-21947
+        df = DataFrame(columns=["a"])
+        cond = df
+        assert (cond.dtypes == object).all()
+
+        result = df.where(cond)
+        tm.assert_frame_equal(result, df)
+
+    def test_where_align(self):
+        def create():
+            df = DataFrame(np.random.default_rng(2).standard_normal((10, 3)))
+            df.iloc[3:5, 0] = np.nan
+            df.iloc[4:6, 1] = np.nan
+            df.iloc[5:8, 2] = np.nan
+            return df
+
+        # series
+        df = create()
+        expected = df.fillna(df.mean())
+        result = df.where(pd.notna(df), df.mean(), axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+        return_value = df.where(pd.notna(df), df.mean(), inplace=True, axis="columns")
+        assert return_value is None
+        tm.assert_frame_equal(df, expected)
+
+        df = create().fillna(0)
+        expected = df.apply(lambda x, y: x.where(x > 0, y), y=df[0])
+        result = df.where(df > 0, df[0], axis="index")
+        tm.assert_frame_equal(result, expected)
+        result = df.where(df > 0, df[0], axis="rows")
+        tm.assert_frame_equal(result, expected)
+
+        # frame
+        df = create()
+        expected = df.fillna(1)
+        result = df.where(
+            pd.notna(df), DataFrame(1, index=df.index, columns=df.columns)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_complex(self):
+        # GH 6345
+        expected = DataFrame([[1 + 1j, 2], [np.nan, 4 + 1j]], columns=["a", "b"])
+        df = DataFrame([[1 + 1j, 2], [5 + 1j, 4 + 1j]], columns=["a", "b"])
+        df[df.abs() >= 5] = np.nan
+        tm.assert_frame_equal(df, expected)
+
+    def test_where_axis(self):
+        # GH 9736
+        df = DataFrame(np.random.default_rng(2).standard_normal((2, 2)))
+        mask = DataFrame([[False, False], [False, False]])
+        ser = Series([0, 1])
+
+        expected = DataFrame([[0, 0], [1, 1]], dtype="float64")
+        result = df.where(mask, ser, axis="index")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(mask, ser, axis="index", inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([[0, 1], [0, 1]], dtype="float64")
+        result = df.where(mask, ser, axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(mask, ser, axis="columns", inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_axis_with_upcast(self):
+        # Upcast needed
+        df = DataFrame([[1, 2], [3, 4]], dtype="int64")
+        mask = DataFrame([[False, False], [False, False]])
+        ser = Series([0, np.nan])
+
+        expected = DataFrame([[0, 0], [np.nan, np.nan]], dtype="float64")
+        result = df.where(mask, ser, axis="index")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            result.where(mask, ser, axis="index", inplace=True)
+
+        expected = DataFrame([[0, np.nan], [0, np.nan]])
+        result = df.where(mask, ser, axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.where(mask, ser, axis="columns", inplace=True)
+
+    def test_where_axis_multiple_dtypes(self):
+        # Multiple dtypes (=> multiple Blocks)
+        df = pd.concat(
+            [
+                DataFrame(np.random.default_rng(2).standard_normal((10, 2))),
+                DataFrame(
+                    np.random.default_rng(2).integers(0, 10, size=(10, 2)),
+                    dtype="int64",
+                ),
+            ],
+            ignore_index=True,
+            axis=1,
+        )
+        mask = DataFrame(False, columns=df.columns, index=df.index)
+        s1 = Series(1, index=df.columns)
+        s2 = Series(2, index=df.index)
+
+        result = df.where(mask, s1, axis="columns")
+        expected = DataFrame(1.0, columns=df.columns, index=df.index)
+        expected[2] = expected[2].astype("int64")
+        expected[3] = expected[3].astype("int64")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(mask, s1, axis="columns", inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+        result = df.where(mask, s2, axis="index")
+        expected = DataFrame(2.0, columns=df.columns, index=df.index)
+        expected[2] = expected[2].astype("int64")
+        expected[3] = expected[3].astype("int64")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        return_value = result.where(mask, s2, axis="index", inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+        # DataFrame vs DataFrame
+        d1 = df.copy().drop(1, axis=0)
+        # Explicit cast to avoid implicit cast when setting value to np.nan
+        expected = df.copy().astype("float")
+        expected.loc[1, :] = np.nan
+
+        result = df.where(mask, d1)
+        tm.assert_frame_equal(result, expected)
+        result = df.where(mask, d1, axis="index")
+        tm.assert_frame_equal(result, expected)
+        result = df.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            result.where(mask, d1, inplace=True)
+        with pytest.raises(TypeError, match="Invalid value"):
+            return_value = result.where(mask, d1, inplace=True, axis="index")
+
+        d2 = df.copy().drop(1, axis=1)
+        expected = df.copy()
+        expected.loc[:, 1] = np.nan
+
+        result = df.where(mask, d2)
+        tm.assert_frame_equal(result, expected)
+        result = df.where(mask, d2, axis="columns")
+        tm.assert_frame_equal(result, expected)
+        result = df.copy()
+        return_value = result.where(mask, d2, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+        result = df.copy()
+        return_value = result.where(mask, d2, inplace=True, axis="columns")
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_callable(self):
+        # GH 12533
+        df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
+        result = df.where(lambda x: x > 4, lambda x: x + 1)
+        exp = DataFrame([[2, 3, 4], [5, 5, 6], [7, 8, 9]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, df.where(df > 4, df + 1))
+
+        # return ndarray and scalar
+        result = df.where(lambda x: (x % 2 == 0).values, lambda x: 99)
+        exp = DataFrame([[99, 2, 99], [4, 99, 6], [99, 8, 99]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, df.where(df % 2 == 0, 99))
+
+        # chain
+        result = (df + 2).where(lambda x: x > 8, lambda x: x + 10)
+        exp = DataFrame([[13, 14, 15], [16, 17, 18], [9, 10, 11]])
+        tm.assert_frame_equal(result, exp)
+        tm.assert_frame_equal(result, (df + 2).where((df + 2) > 8, (df + 2) + 10))
+
+    def test_where_tz_values(self, tz_naive_fixture, frame_or_series):
+        obj1 = DataFrame(
+            DatetimeIndex(["20150101", "20150102", "20150103"], tz=tz_naive_fixture),
+            columns=["date"],
+        )
+        obj2 = DataFrame(
+            DatetimeIndex(["20150103", "20150104", "20150105"], tz=tz_naive_fixture),
+            columns=["date"],
+        )
+        mask = DataFrame([True, True, False], columns=["date"])
+        exp = DataFrame(
+            DatetimeIndex(["20150101", "20150102", "20150105"], tz=tz_naive_fixture),
+            columns=["date"],
+        )
+        if frame_or_series is Series:
+            obj1 = obj1["date"]
+            obj2 = obj2["date"]
+            mask = mask["date"]
+            exp = exp["date"]
+
+        result = obj1.where(mask, obj2)
+        tm.assert_equal(exp, result)
+
+    def test_df_where_change_dtype(self):
+        # GH#16979
+        df = DataFrame(np.arange(2 * 3).reshape(2, 3), columns=list("ABC"))
+        mask = np.array([[True, False, False], [False, False, True]])
+
+        result = df.where(mask)
+        expected = DataFrame(
+            [[0, np.nan, np.nan], [np.nan, np.nan, 5]], columns=list("ABC")
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("kwargs", [{}, {"other": None}])
+    def test_df_where_with_category(self, kwargs):
+        # GH#16979
+        data = np.arange(2 * 3, dtype=np.int64).reshape(2, 3)
+        df = DataFrame(data, columns=list("ABC"))
+        mask = np.array([[True, False, False], [False, False, True]])
+
+        # change type to category
+        df.A = df.A.astype("category")
+        df.B = df.B.astype("category")
+        df.C = df.C.astype("category")
+
+        result = df.where(mask, **kwargs)
+        A = pd.Categorical([0, np.nan], categories=[0, 3])
+        B = pd.Categorical([np.nan, np.nan], categories=[1, 4])
+        C = pd.Categorical([np.nan, 5], categories=[2, 5])
+        expected = DataFrame({"A": A, "B": B, "C": C})
+
+        tm.assert_frame_equal(result, expected)
+
+        # Check Series.where while we're here
+        result = df.A.where(mask[:, 0], **kwargs)
+        expected = Series(A, name="A")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_where_categorical_filtering(self):
+        # GH#22609 Verify filtering operations on DataFrames with categorical Series
+        df = DataFrame(data=[[0, 0], [1, 1]], columns=["a", "b"])
+        df["b"] = df["b"].astype("category")
+
+        result = df.where(df["a"] > 0)
+        # Explicitly cast to 'float' to avoid implicit cast when setting np.nan
+        expected = df.copy().astype({"a": "float"})
+        expected.loc[0, :] = np.nan
+
+        tm.assert_equal(result, expected)
+
+    def test_where_ea_other(self):
+        # GH#38729/GH#38742
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        arr = pd.array([7, pd.NA, 9])
+        ser = Series(arr)
+        mask = np.ones(df.shape, dtype=bool)
+        mask[1, :] = False
+
+        # TODO: ideally we would get Int64 instead of object
+        result = df.where(mask, ser, axis=0)
+        expected = DataFrame({"A": [1, np.nan, 3], "B": [4, np.nan, 6]})
+        tm.assert_frame_equal(result, expected)
+
+        ser2 = Series(arr[:2], index=["A", "B"])
+        expected = DataFrame({"A": [1, 7, 3], "B": [4, np.nan, 6]})
+        result = df.where(mask, ser2, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_where_interval_noop(self):
+        # GH#44181
+        df = DataFrame([pd.Interval(0, 0)])
+        res = df.where(df.notna())
+        tm.assert_frame_equal(res, df)
+
+        ser = df[0]
+        res = ser.where(ser.notna())
+        tm.assert_series_equal(res, ser)
+
+    def test_where_interval_fullop_downcast(self, frame_or_series):
+        # GH#45768
+        obj = frame_or_series([pd.Interval(0, 0)] * 2)
+        other = frame_or_series([1.0, 2.0], dtype=object)
+        res = obj.where(~obj.notna(), other)
+        tm.assert_equal(res, other)
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            obj.mask(obj.notna(), other, inplace=True)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "timedelta64[ns]",
+            "datetime64[ns]",
+            "datetime64[ns, Asia/Tokyo]",
+        ],
+    )
+    def test_where_datetimelike_noop(self, dtype):
+        # GH#45135, analogue to GH#44181 for Period don't raise on no-op
+        # For td64/dt64/dt64tz we already don't raise, but also are
+        #  checking that we don't unnecessarily upcast to object.
+        ser = Series(np.arange(3) * 10**9, dtype=np.int64).astype(dtype)
+        df = ser.to_frame()
+        mask = np.array([False, False, False])
+
+        res = ser.where(~mask, "foo")
+        tm.assert_series_equal(res, ser)
+
+        mask2 = mask.reshape(-1, 1)
+        res2 = df.where(~mask2, "foo")
+        tm.assert_frame_equal(res2, df)
+
+        res3 = ser.mask(mask, "foo")
+        tm.assert_series_equal(res3, ser)
+
+        res4 = df.mask(mask2, "foo")
+        tm.assert_frame_equal(res4, df)
+
+        # unlike where, Block.putmask does not downcast
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.mask(~mask2, 4, inplace=True)
+
+
+def test_where_int_downcasting_deprecated():
+    # GH#44597
+    arr = np.arange(6).astype(np.int16).reshape(3, 2)
+    df = DataFrame(arr)
+
+    mask = np.zeros(arr.shape, dtype=bool)
+    mask[:, 0] = True
+
+    res = df.where(mask, 2**17)
+
+    expected = DataFrame({0: arr[:, 0], 1: np.array([2**17] * 3, dtype=np.int32)})
+    tm.assert_frame_equal(res, expected)
+
+
+def test_where_copies_with_noop(frame_or_series):
+    # GH-39595
+    result = frame_or_series([1, 2, 3, 4])
+    expected = result.copy()
+    col = result[0] if frame_or_series is DataFrame else result
+
+    where_res = result.where(col < 5)
+    where_res *= 2
+
+    tm.assert_equal(result, expected)
+
+    where_res = result.where(col > 5, [1, 2, 3, 4])
+    where_res *= 2
+
+    tm.assert_equal(result, expected)
+
+
+def test_where_string_dtype(frame_or_series):
+    # GH40824
+    obj = frame_or_series(
+        ["a", "b", "c", "d"], index=["id1", "id2", "id3", "id4"], dtype=StringDtype()
+    )
+    filtered_obj = frame_or_series(
+        ["b", "c"], index=["id2", "id3"], dtype=StringDtype()
+    )
+    filter_ser = Series([False, True, True, False])
+
+    result = obj.where(filter_ser, filtered_obj)
+    expected = frame_or_series(
+        [pd.NA, "b", "c", pd.NA],
+        index=["id1", "id2", "id3", "id4"],
+        dtype=StringDtype(),
+    )
+    tm.assert_equal(result, expected)
+
+    result = obj.mask(~filter_ser, filtered_obj)
+    tm.assert_equal(result, expected)
+
+    obj.mask(~filter_ser, filtered_obj, inplace=True)
+    tm.assert_equal(result, expected)
+
+
+def test_where_bool_comparison():
+    # GH 10336
+    df_mask = DataFrame(
+        {"AAA": [True] * 4, "BBB": [False] * 4, "CCC": [True, False, True, False]}
+    )
+    result = df_mask.where(df_mask == False)  # noqa: E712
+    expected = DataFrame(
+        {
+            "AAA": np.array([np.nan] * 4, dtype=object),
+            "BBB": [False] * 4,
+            "CCC": [np.nan, False, np.nan, False],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_where_none_nan_coerce():
+    # GH 15613
+    expected = DataFrame(
+        {
+            "A": [Timestamp("20130101"), pd.NaT, Timestamp("20130103")],
+            "B": [1, 2, np.nan],
+        }
+    )
+    result = expected.where(expected.notnull(), None)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_where_duplicate_axes_mixed_dtypes():
+    # GH 25399, verify manually masking is not affected anymore by dtype of column for
+    # duplicate axes.
+    result = DataFrame(data=[[0, np.nan]], columns=Index(["A", "A"]))
+    index, columns = result.axes
+    mask = DataFrame(data=[[True, True]], columns=columns, index=index)
+    a = result.astype(object).where(mask)
+    b = result.astype("f8").where(mask)
+    c = result.T.where(mask.T).T
+    d = result.where(mask)  # used to fail with "cannot reindex from a duplicate axis"
+    tm.assert_frame_equal(a.astype("f8"), b.astype("f8"))
+    tm.assert_frame_equal(b.astype("f8"), c.astype("f8"))
+    tm.assert_frame_equal(c.astype("f8"), d.astype("f8"))
+
+
+def test_where_columns_casting():
+    # GH 42295
+
+    df = DataFrame({"a": [1.0, 2.0], "b": [3, np.nan]})
+    expected = df.copy()
+    result = df.where(pd.notnull(df), None)
+    # make sure dtypes don't change
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize("as_cat", [True, False])
+def test_where_period_invalid_na(frame_or_series, as_cat, request):
+    # GH#44697
+    idx = pd.period_range("2016-01-01", periods=3, freq="D")
+    if as_cat:
+        idx = idx.astype("category")
+    obj = frame_or_series(idx)
+
+    # NA value that we should *not* cast to Period dtype
+    tdnat = pd.NaT.to_numpy("m8[ns]")
+
+    mask = np.array([True, True, False], ndmin=obj.ndim).T
+
+    if as_cat:
+        msg = (
+            r"Cannot setitem on a Categorical with a new category \(NaT\), "
+            "set the categories first"
+        )
+    else:
+        msg = "value should be a 'Period'"
+
+    if as_cat:
+        with pytest.raises(TypeError, match=msg):
+            obj.where(mask, tdnat)
+
+        with pytest.raises(TypeError, match=msg):
+            obj.mask(mask, tdnat)
+
+        with pytest.raises(TypeError, match=msg):
+            obj.mask(mask, tdnat, inplace=True)
+
+    else:
+        # With PeriodDtype, ser[i] = tdnat coerces instead of raising,
+        #  so for consistency, ser[mask] = tdnat must as well
+        expected = obj.astype(object).where(mask, tdnat)
+        result = obj.where(mask, tdnat)
+        tm.assert_equal(result, expected)
+
+        expected = obj.astype(object).mask(mask, tdnat)
+        result = obj.mask(mask, tdnat)
+        tm.assert_equal(result, expected)
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            obj.mask(mask, tdnat, inplace=True)
+
+
+def test_where_nullable_invalid_na(frame_or_series, any_numeric_ea_dtype):
+    # GH#44697
+    arr = pd.array([1, 2, 3], dtype=any_numeric_ea_dtype)
+    obj = frame_or_series(arr)
+
+    mask = np.array([True, True, False], ndmin=obj.ndim).T
+
+    msg = r"Invalid value '.*' for dtype '(U?Int|Float)\d{1,2}'"
+
+    for null in tm.NP_NAT_OBJECTS + [pd.NaT]:
+        # NaT is an NA value that we should *not* cast to pd.NA dtype
+        with pytest.raises(TypeError, match=msg):
+            obj.where(mask, null)
+
+        with pytest.raises(TypeError, match=msg):
+            obj.mask(mask, null)
+
+
+@given(data=OPTIONAL_ONE_OF_ALL)
+def test_where_inplace_casting(data):
+    # GH 22051
+    df = DataFrame({"a": data})
+    df_copy = df.where(pd.notnull(df), None).copy()
+    df.where(pd.notnull(df), None, inplace=True)
+    tm.assert_equal(df, df_copy)
+
+
+def test_where_downcast_to_td64():
+    ser = Series([1, 2, 3])
+
+    mask = np.array([False, False, False])
+
+    td = pd.Timedelta(days=1)
+    expected = Series([td, td, td], dtype="m8[ns]")
+
+    res2 = ser.where(mask, td)
+    expected2 = expected.astype(object)
+    tm.assert_series_equal(res2, expected2)
+
+
+def _check_where_equivalences(df, mask, other, expected):
+    # similar to tests.series.indexing.test_setitem.SetitemCastingEquivalences
+    #  but with DataFrame in mind and less fleshed-out
+    res = df.where(mask, other)
+    tm.assert_frame_equal(res, expected)
+
+    res = df.mask(~mask, other)
+    tm.assert_frame_equal(res, expected)
+
+    # Note: frame.mask(~mask, other, inplace=True) takes some more work bc
+    #  Block.putmask does *not* downcast.  The change to 'expected' here
+    #  is specific to the cases in test_where_dt64_2d.
+    df = df.copy()
+    df.mask(~mask, other, inplace=True)
+    if not mask.all():
+        # with mask.all(), Block.putmask is a no-op, so does not downcast
+        expected = expected.copy()
+        expected["A"] = expected["A"].astype(object)
+    tm.assert_frame_equal(df, expected)
+
+
+def test_where_dt64_2d():
+    dti = date_range("2016-01-01", periods=6)
+    dta = dti._data.reshape(3, 2)
+    other = dta - dta[0, 0]
+
+    df = DataFrame(dta, columns=["A", "B"])
+
+    mask = np.asarray(df.isna()).copy()
+    mask[:, 1] = True
+
+    # setting part of one column, none of the other
+    mask[1, 0] = True
+    expected = DataFrame(
+        {
+            "A": np.array([other[0, 0], dta[1, 0], other[2, 0]], dtype=object),
+            "B": dta[:, 1],
+        }
+    )
+    with pytest.raises(TypeError, match="Invalid value"):
+        _check_where_equivalences(df, mask, other, expected)
+
+    # setting nothing in either column
+    mask[:] = True
+    expected = df
+    _check_where_equivalences(df, mask, other, expected)
+
+
+def test_where_producing_ea_cond_for_np_dtype():
+    # GH#44014
+    df = DataFrame({"a": Series([1, pd.NA, 2], dtype="Int64"), "b": [1, 2, 3]})
+    result = df.where(lambda x: x.apply(lambda y: y > 1, axis=1))
+    expected = DataFrame(
+        {"a": Series([pd.NA, pd.NA, 2], dtype="Int64"), "b": [np.nan, 2, 3]}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "replacement", [0.001, True, "snake", None, datetime(2022, 5, 4)]
+)
+def test_where_int_overflow(replacement):
+    # GH 31687
+    df = DataFrame([[1.0, 2e25, "nine"], [np.nan, 0.1, None]])
+    result = df.where(pd.notnull(df), replacement)
+    expected = DataFrame([[1.0, 2e25, "nine"], [replacement, 0.1, replacement]])
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_where_inplace_no_other():
+    # GH#51685
+    df = DataFrame({"a": [1.0, 2.0], "b": ["x", "y"]})
+    cond = DataFrame({"a": [True, False], "b": [False, True]})
+    df.where(cond, inplace=True)
+    expected = DataFrame({"a": [1, np.nan], "b": [np.nan, "y"]})
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/indexing/test_xs.py b/pandas/pandas/tests/frame/indexing/test_xs.py
new file mode 100644
index 0000000000000..54733129b4d47
--- /dev/null
+++ b/pandas/pandas/tests/frame/indexing/test_xs.py
@@ -0,0 +1,387 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    IndexSlice,
+    MultiIndex,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import BDay
+
+
+@pytest.fixture
+def four_level_index_dataframe():
+    arr = np.array(
+        [
+            [-0.5109, -2.3358, -0.4645, 0.05076, 0.364],
+            [0.4473, 1.4152, 0.2834, 1.00661, 0.1744],
+            [-0.6662, -0.5243, -0.358, 0.89145, 2.5838],
+        ]
+    )
+    index = MultiIndex(
+        levels=[["a", "x"], ["b", "q"], [10.0032, 20.0, 30.0], [3, 4, 5]],
+        codes=[[0, 0, 1], [0, 1, 1], [0, 1, 2], [2, 1, 0]],
+        names=["one", "two", "three", "four"],
+    )
+    return DataFrame(arr, index=index, columns=list("ABCDE"))
+
+
+class TestXS:
+    def test_xs(self, float_frame):
+        idx = float_frame.index[5]
+        xs = float_frame.xs(idx)
+        for item, value in xs.items():
+            if np.isnan(value):
+                assert np.isnan(float_frame[item][idx])
+            else:
+                assert value == float_frame[item][idx]
+
+    def test_xs_mixed(self):
+        # mixed-type xs
+        test_data = {"A": {"1": 1, "2": 2}, "B": {"1": "1", "2": "2", "3": "3"}}
+        frame = DataFrame(test_data)
+        xs = frame.xs("1")
+        assert xs.dtype == np.object_
+        assert xs["A"] == 1
+        assert xs["B"] == "1"
+
+    def test_xs_dt_error(self, datetime_frame):
+        with pytest.raises(
+            KeyError, match=re.escape("Timestamp('1999-12-31 00:00:00')")
+        ):
+            datetime_frame.xs(datetime_frame.index[0] - BDay())
+
+    def test_xs_other(self, float_frame):
+        float_frame_orig = float_frame.copy()
+        # xs get column
+        series = float_frame.xs("A", axis=1)
+        expected = float_frame["A"]
+        tm.assert_series_equal(series, expected)
+
+        # view is returned if possible
+        series = float_frame.xs("A", axis=1)
+        series[:] = 5
+        # The view shouldn't propagate mutations
+        tm.assert_series_equal(float_frame["A"], float_frame_orig["A"])
+        assert not (expected == 5).all()
+
+    def test_xs_corner(self):
+        # pathological mixed-type reordering case
+        df = DataFrame(index=[0], columns=Index([], dtype="str"))
+        df["A"] = 1.0
+        df["B"] = "foo"
+        df["C"] = 2.0
+        df["D"] = "bar"
+        df["E"] = 3.0
+
+        xs = df.xs(0)
+        exp = Series([1.0, "foo", 2.0, "bar", 3.0], index=list("ABCDE"), name=0)
+        tm.assert_series_equal(xs, exp)
+
+        # no columns but Index(dtype=object)
+        df = DataFrame(index=["a", "b", "c"])
+        result = df.xs("a")
+        expected = Series([], name="a", dtype=np.float64)
+        tm.assert_series_equal(result, expected)
+
+    def test_xs_duplicates(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)),
+            index=["b", "b", "c", "b", "a"],
+        )
+
+        cross = df.xs("c")
+        exp = df.iloc[2]
+        tm.assert_series_equal(cross, exp)
+
+    def test_xs_keep_level(self):
+        df = DataFrame(
+            {
+                "day": {0: "sat", 1: "sun"},
+                "flavour": {0: "strawberry", 1: "strawberry"},
+                "sales": {0: 10, 1: 12},
+                "year": {0: 2008, 1: 2008},
+            }
+        ).set_index(["year", "flavour", "day"])
+        result = df.xs("sat", level="day", drop_level=False)
+        expected = df[:1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.xs((2008, "sat"), level=["year", "day"], drop_level=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_view(self):
+        # in 0.14 this will return a view if possible a copy otherwise, but
+        # this is numpy dependent
+
+        dm = DataFrame(np.arange(20.0).reshape(4, 5), index=range(4), columns=range(5))
+        df_orig = dm.copy()
+
+        with tm.raises_chained_assignment_error():
+            dm.xs(2)[:] = 20
+        tm.assert_frame_equal(dm, df_orig)
+
+
+class TestXSWithMultiIndex:
+    def test_xs_doc_example(self):
+        # TODO: more descriptive name
+        # based on example in advanced.rst
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = list(zip(*arrays))
+
+        index = MultiIndex.from_tuples(tuples, names=["first", "second"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 8)),
+            index=["A", "B", "C"],
+            columns=index,
+        )
+
+        result = df.xs(("one", "bar"), level=("second", "first"), axis=1)
+
+        expected = df.iloc[:, [0]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_integer_key(self):
+        # see GH#2107
+        dates = range(20111201, 20111205)
+        ids = list("abcde")
+        index = MultiIndex.from_product([dates, ids], names=["date", "secid"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 3)),
+            index,
+            ["X", "Y", "Z"],
+        )
+
+        result = df.xs(20111201, level="date")
+        expected = df.loc[20111201, :]
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_level(self, multiindex_dataframe_random_data):
+        df = multiindex_dataframe_random_data
+        result = df.xs("two", level="second")
+        expected = df[df.index.get_level_values(1) == "two"]
+        expected.index = Index(["foo", "bar", "baz", "qux"], name="first")
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_level_eq_2(self):
+        arr = np.random.default_rng(2).standard_normal((3, 5))
+        index = MultiIndex(
+            levels=[["a", "p", "x"], ["b", "q", "y"], ["c", "r", "z"]],
+            codes=[[2, 0, 1], [2, 0, 1], [2, 0, 1]],
+        )
+        df = DataFrame(arr, index=index)
+        expected = DataFrame(arr[1:2], index=[["a"], ["b"]])
+        result = df.xs("c", level=2)
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_setting_with_copy_error(self, multiindex_dataframe_random_data):
+        # this is a copy in 0.14
+        df = multiindex_dataframe_random_data
+        df_orig = df.copy()
+        result = df.xs("two", level="second")
+
+        result[:] = 10
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_xs_setting_with_copy_error_multiple(self, four_level_index_dataframe):
+        # this is a copy in 0.14
+        df = four_level_index_dataframe
+        df_orig = df.copy()
+        result = df.xs(("a", 4), level=["one", "four"])
+
+        result[:] = 10
+        tm.assert_frame_equal(df, df_orig)
+
+    @pytest.mark.parametrize("key, level", [("one", "second"), (["one"], ["second"])])
+    def test_xs_with_duplicates(self, key, level, multiindex_dataframe_random_data):
+        # see GH#13719
+        frame = multiindex_dataframe_random_data
+        df = concat([frame] * 2)
+        assert df.index.is_unique is False
+        expected = concat([frame.xs("one", level="second")] * 2)
+
+        if isinstance(key, list):
+            result = df.xs(tuple(key), level=level)
+        else:
+            result = df.xs(key, level=level)
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_missing_values_in_index(self):
+        # see GH#6574
+        # missing values in returned index should be preserved
+        acc = [
+            ("a", "abcde", 1),
+            ("b", "bbcde", 2),
+            ("y", "yzcde", 25),
+            ("z", "xbcde", 24),
+            ("z", None, 26),
+            ("z", "zbcde", 25),
+            ("z", "ybcde", 26),
+        ]
+        df = DataFrame(acc, columns=["a1", "a2", "cnt"]).set_index(["a1", "a2"])
+        expected = DataFrame(
+            {"cnt": [24, 26, 25, 26]},
+            index=Index(["xbcde", np.nan, "zbcde", "ybcde"], name="a2"),
+        )
+
+        result = df.xs("z", level="a1")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "key, level, exp_arr, exp_index",
+        [
+            ("a", "lvl0", lambda x: x[:, 0:2], Index(["bar", "foo"], name="lvl1")),
+            ("foo", "lvl1", lambda x: x[:, 1:2], Index(["a"], name="lvl0")),
+        ],
+    )
+    def test_xs_named_levels_axis_eq_1(self, key, level, exp_arr, exp_index):
+        # see GH#2903
+        arr = np.random.default_rng(2).standard_normal((4, 4))
+        index = MultiIndex(
+            levels=[["a", "b"], ["bar", "foo", "hello", "world"]],
+            codes=[[0, 0, 1, 1], [0, 1, 2, 3]],
+            names=["lvl0", "lvl1"],
+        )
+        df = DataFrame(arr, columns=index)
+        result = df.xs(key, level=level, axis=1)
+        expected = DataFrame(exp_arr(arr), columns=exp_index)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "indexer",
+        [
+            lambda df: df.xs(("a", 4), level=["one", "four"]),
+            lambda df: df.xs("a").xs(4, level="four"),
+        ],
+    )
+    def test_xs_level_multiple(self, indexer, four_level_index_dataframe):
+        df = four_level_index_dataframe
+        expected_values = [[0.4473, 1.4152, 0.2834, 1.00661, 0.1744]]
+        expected_index = MultiIndex(
+            levels=[["q"], [20.0]], codes=[[0], [0]], names=["two", "three"]
+        )
+        expected = DataFrame(
+            expected_values, index=expected_index, columns=list("ABCDE")
+        )
+        result = indexer(df)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "indexer", [lambda df: df.xs("a", level=0), lambda df: df.xs("a")]
+    )
+    def test_xs_level0(self, indexer, four_level_index_dataframe):
+        df = four_level_index_dataframe
+        expected_values = [
+            [-0.5109, -2.3358, -0.4645, 0.05076, 0.364],
+            [0.4473, 1.4152, 0.2834, 1.00661, 0.1744],
+        ]
+        expected_index = MultiIndex(
+            levels=[["b", "q"], [10.0032, 20.0], [4, 5]],
+            codes=[[0, 1], [0, 1], [1, 0]],
+            names=["two", "three", "four"],
+        )
+        expected = DataFrame(
+            expected_values, index=expected_index, columns=list("ABCDE")
+        )
+
+        result = indexer(df)
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_values(self, multiindex_dataframe_random_data):
+        df = multiindex_dataframe_random_data
+        result = df.xs(("bar", "two")).values
+        expected = df.values[4]
+        tm.assert_almost_equal(result, expected)
+
+    def test_xs_loc_equality(self, multiindex_dataframe_random_data):
+        df = multiindex_dataframe_random_data
+        result = df.xs(("bar", "two"))
+        expected = df.loc[("bar", "two")]
+        tm.assert_series_equal(result, expected)
+
+    def test_xs_IndexSlice_argument_not_implemented(self, frame_or_series):
+        # GH#35301
+
+        index = MultiIndex(
+            levels=[[("foo", "bar", 0), ("foo", "baz", 0), ("foo", "qux", 0)], [0, 1]],
+            codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+        )
+
+        obj = DataFrame(np.random.default_rng(2).standard_normal((6, 4)), index=index)
+        if frame_or_series is Series:
+            obj = obj[0]
+
+        expected = obj.iloc[-2:].droplevel(0)
+
+        result = obj.xs(IndexSlice[("foo", "qux", 0), :])
+        tm.assert_equal(result, expected)
+
+        result = obj.loc[IndexSlice[("foo", "qux", 0), :]]
+        tm.assert_equal(result, expected)
+
+    def test_xs_levels_raises(self, frame_or_series):
+        obj = DataFrame({"A": [1, 2, 3]})
+        if frame_or_series is Series:
+            obj = obj["A"]
+
+        msg = "Index must be a MultiIndex"
+        with pytest.raises(TypeError, match=msg):
+            obj.xs(0, level="as")
+
+    def test_xs_multiindex_droplevel_false(self):
+        # GH#19056
+        mi = MultiIndex.from_tuples(
+            [("a", "x"), ("a", "y"), ("b", "x")], names=["level1", "level2"]
+        )
+        df = DataFrame([[1, 2, 3]], columns=mi)
+        result = df.xs("a", axis=1, drop_level=False)
+        expected = DataFrame(
+            [[1, 2]],
+            columns=MultiIndex.from_tuples(
+                [("a", "x"), ("a", "y")], names=["level1", "level2"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_droplevel_false(self):
+        # GH#19056
+        df = DataFrame([[1, 2, 3]], columns=Index(["a", "b", "c"]))
+        result = df.xs("a", axis=1, drop_level=False)
+        expected = DataFrame({"a": [1]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_droplevel_false_view(self):
+        # GH#37832
+        df = DataFrame([[1, 2, 3]], columns=Index(["a", "b", "c"]))
+        result = df.xs("a", axis=1, drop_level=False)
+        # check that result still views the same data as df
+        assert np.shares_memory(result.iloc[:, 0]._values, df.iloc[:, 0]._values)
+
+        df.iloc[0, 0] = 2
+        # The subset is never modified
+        expected = DataFrame({"a": [1]})
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame([[1, 2.5, "a"]], columns=Index(["a", "b", "c"]))
+        result = df.xs("a", axis=1, drop_level=False)
+        df.iloc[0, 0] = 2
+        # The subset is never modified
+        expected = DataFrame({"a": [1]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_xs_list_indexer_droplevel_false(self):
+        # GH#41760
+        mi = MultiIndex.from_tuples([("x", "m", "a"), ("x", "n", "b"), ("y", "o", "c")])
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=mi)
+        with pytest.raises(KeyError, match="y"):
+            df.xs(("x", "y"), drop_level=False, axis=1)
diff --git a/pandas/pandas/tests/frame/methods/__init__.py b/pandas/pandas/tests/frame/methods/__init__.py
new file mode 100644
index 0000000000000..245594bfdc9e7
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/__init__.py
@@ -0,0 +1,7 @@
+"""
+Test files dedicated to individual (stand-alone) DataFrame methods
+
+Ideally these files/tests should correspond 1-to-1 with tests.series.methods
+
+These may also present opportunities for sharing/de-duplicating test code.
+"""
diff --git a/pandas/pandas/tests/frame/methods/test_add_prefix_suffix.py b/pandas/pandas/tests/frame/methods/test_add_prefix_suffix.py
new file mode 100644
index 0000000000000..92d7cdd7990e1
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_add_prefix_suffix.py
@@ -0,0 +1,49 @@
+import pytest
+
+from pandas import Index
+import pandas._testing as tm
+
+
+def test_add_prefix_suffix(float_frame):
+    with_prefix = float_frame.add_prefix("foo#")
+    expected = Index([f"foo#{c}" for c in float_frame.columns])
+    tm.assert_index_equal(with_prefix.columns, expected)
+
+    with_suffix = float_frame.add_suffix("#foo")
+    expected = Index([f"{c}#foo" for c in float_frame.columns])
+    tm.assert_index_equal(with_suffix.columns, expected)
+
+    with_pct_prefix = float_frame.add_prefix("%")
+    expected = Index([f"%{c}" for c in float_frame.columns])
+    tm.assert_index_equal(with_pct_prefix.columns, expected)
+
+    with_pct_suffix = float_frame.add_suffix("%")
+    expected = Index([f"{c}%" for c in float_frame.columns])
+    tm.assert_index_equal(with_pct_suffix.columns, expected)
+
+
+def test_add_prefix_suffix_axis(float_frame):
+    # GH 47819
+    with_prefix = float_frame.add_prefix("foo#", axis=0)
+    expected = Index([f"foo#{c}" for c in float_frame.index])
+    tm.assert_index_equal(with_prefix.index, expected)
+
+    with_prefix = float_frame.add_prefix("foo#", axis=1)
+    expected = Index([f"foo#{c}" for c in float_frame.columns])
+    tm.assert_index_equal(with_prefix.columns, expected)
+
+    with_pct_suffix = float_frame.add_suffix("#foo", axis=0)
+    expected = Index([f"{c}#foo" for c in float_frame.index])
+    tm.assert_index_equal(with_pct_suffix.index, expected)
+
+    with_pct_suffix = float_frame.add_suffix("#foo", axis=1)
+    expected = Index([f"{c}#foo" for c in float_frame.columns])
+    tm.assert_index_equal(with_pct_suffix.columns, expected)
+
+
+def test_add_prefix_suffix_invalid_axis(float_frame):
+    with pytest.raises(ValueError, match="No axis named 2 for object type DataFrame"):
+        float_frame.add_prefix("foo#", axis=2)
+
+    with pytest.raises(ValueError, match="No axis named 2 for object type DataFrame"):
+        float_frame.add_suffix("foo#", axis=2)
diff --git a/pandas/pandas/tests/frame/methods/test_align.py b/pandas/pandas/tests/frame/methods/test_align.py
new file mode 100644
index 0000000000000..2677b7332a65a
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_align.py
@@ -0,0 +1,338 @@
+from datetime import timezone
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameAlign:
+    def test_frame_align_aware(self):
+        idx1 = date_range("2001", periods=5, freq="h", tz="US/Eastern")
+        idx2 = date_range("2001", periods=5, freq="2h", tz="US/Eastern")
+        df1 = DataFrame(np.random.default_rng(2).standard_normal((len(idx1), 3)), idx1)
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((len(idx2), 3)), idx2)
+        new1, new2 = df1.align(df2)
+        assert df1.index.tz == new1.index.tz
+        assert df2.index.tz == new2.index.tz
+
+        # different timezones convert to UTC
+
+        # frame with frame
+        df1_central = df1.tz_convert("US/Central")
+        new1, new2 = df1.align(df1_central)
+        assert new1.index.tz is timezone.utc
+        assert new2.index.tz is timezone.utc
+
+        # frame with Series
+        new1, new2 = df1.align(df1_central[0], axis=0)
+        assert new1.index.tz is timezone.utc
+        assert new2.index.tz is timezone.utc
+
+        df1[0].align(df1_central, axis=0)
+        assert new1.index.tz is timezone.utc
+        assert new2.index.tz is timezone.utc
+
+    def test_align_float(self, float_frame):
+        af, bf = float_frame.align(float_frame)
+        assert af._mgr is not float_frame._mgr
+
+        af, bf = float_frame.align(float_frame)
+        assert af._mgr is not float_frame._mgr
+
+        # axis = 0
+        other = float_frame.iloc[:-5, :3]
+        af, bf = float_frame.align(other, axis=0, fill_value=-1)
+
+        tm.assert_index_equal(bf.columns, other.columns)
+
+        # test fill value
+        join_idx = float_frame.index.join(other.index)
+        diff_a = float_frame.index.difference(join_idx)
+        diff_a_vals = af.reindex(diff_a).values
+        assert (diff_a_vals == -1).all()
+
+        af, bf = float_frame.align(other, join="right", axis=0)
+        tm.assert_index_equal(bf.columns, other.columns)
+        tm.assert_index_equal(bf.index, other.index)
+        tm.assert_index_equal(af.index, other.index)
+
+        # axis = 1
+        other = float_frame.iloc[:-5, :3].copy()
+        af, bf = float_frame.align(other, axis=1)
+        tm.assert_index_equal(bf.columns, float_frame.columns)
+        tm.assert_index_equal(bf.index, other.index)
+
+        # test fill value
+        join_idx = float_frame.index.join(other.index)
+        diff_a = float_frame.index.difference(join_idx)
+        diff_a_vals = af.reindex(diff_a).values
+
+        assert (diff_a_vals == -1).all()
+
+        af, bf = float_frame.align(other, join="inner", axis=1)
+        tm.assert_index_equal(bf.columns, other.columns)
+
+        # Try to align DataFrame to Series along bad axis
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            float_frame.align(af.iloc[0, :3], join="inner", axis=2)
+
+    def test_align_frame_with_series(self, float_frame):
+        # align dataframe to series with broadcast or not
+        idx = float_frame.index
+        s = Series(range(len(idx)), index=idx)
+
+        left, right = float_frame.align(s, axis=0)
+        tm.assert_index_equal(left.index, float_frame.index)
+        tm.assert_index_equal(right.index, float_frame.index)
+        assert isinstance(right, Series)
+
+    def test_align_series_condition(self):
+        # see gh-9558
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        result = df[df["a"] == 2]
+        expected = DataFrame([[2, 5]], index=[1], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+        result = df.where(df["a"] == 2, 0)
+        expected = DataFrame({"a": [0, 2, 0], "b": [0, 5, 0]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_align_mixed_float(self, mixed_float_frame):
+        # mixed floats/ints
+        other = DataFrame(index=range(5), columns=["A", "B", "C"])
+        af, bf = mixed_float_frame.align(
+            other.iloc[:, 0], join="inner", axis=1, fill_value=0
+        )
+        tm.assert_index_equal(bf.index, Index([]))
+
+    def test_align_mixed_int(self, mixed_int_frame):
+        other = DataFrame(index=range(5), columns=["A", "B", "C"])
+        af, bf = mixed_int_frame.align(
+            other.iloc[:, 0], join="inner", axis=1, fill_value=0
+        )
+        tm.assert_index_equal(bf.index, Index([]))
+
+    @pytest.mark.parametrize(
+        "l_ordered,r_ordered,expected",
+        [
+            [True, True, pd.CategoricalIndex],
+            [True, False, Index],
+            [False, True, Index],
+            [False, False, pd.CategoricalIndex],
+        ],
+    )
+    def test_align_categorical(self, l_ordered, r_ordered, expected):
+        # GH-28397
+        df_1 = DataFrame(
+            {
+                "A": np.arange(6, dtype="int64"),
+                "B": Series(list("aabbca")).astype(
+                    pd.CategoricalDtype(list("cab"), ordered=l_ordered)
+                ),
+            }
+        ).set_index("B")
+        df_2 = DataFrame(
+            {
+                "A": np.arange(5, dtype="int64"),
+                "B": Series(list("babca")).astype(
+                    pd.CategoricalDtype(list("cab"), ordered=r_ordered)
+                ),
+            }
+        ).set_index("B")
+
+        aligned_1, aligned_2 = df_1.align(df_2)
+        assert isinstance(aligned_1.index, expected)
+        assert isinstance(aligned_2.index, expected)
+        tm.assert_index_equal(aligned_1.index, aligned_2.index)
+
+    def test_align_multiindex(self):
+        # GH#10665
+        # same test cases as test_align_multiindex in test_series.py
+
+        midx = pd.MultiIndex.from_product(
+            [range(2), range(3), range(2)], names=("a", "b", "c")
+        )
+        idx = Index(range(2), name="b")
+        df1 = DataFrame(np.arange(12, dtype="int64"), index=midx)
+        df2 = DataFrame(np.arange(2, dtype="int64"), index=idx)
+
+        # these must be the same results (but flipped)
+        res1l, res1r = df1.align(df2, join="left")
+        res2l, res2r = df2.align(df1, join="right")
+
+        expl = df1
+        tm.assert_frame_equal(expl, res1l)
+        tm.assert_frame_equal(expl, res2r)
+        expr = DataFrame([0, 0, 1, 1, np.nan, np.nan] * 2, index=midx)
+        tm.assert_frame_equal(expr, res1r)
+        tm.assert_frame_equal(expr, res2l)
+
+        res1l, res1r = df1.align(df2, join="right")
+        res2l, res2r = df2.align(df1, join="left")
+
+        exp_idx = pd.MultiIndex.from_product(
+            [range(2), range(2), range(2)], names=("a", "b", "c")
+        )
+        expl = DataFrame([0, 1, 2, 3, 6, 7, 8, 9], index=exp_idx)
+        tm.assert_frame_equal(expl, res1l)
+        tm.assert_frame_equal(expl, res2r)
+        expr = DataFrame([0, 0, 1, 1] * 2, index=exp_idx)
+        tm.assert_frame_equal(expr, res1r)
+        tm.assert_frame_equal(expr, res2l)
+
+    def test_align_series_combinations(self):
+        df = DataFrame({"a": [1, 3, 5], "b": [1, 3, 5]}, index=list("ACE"))
+        s = Series([1, 2, 4], index=list("ABD"), name="x")
+
+        # frame + series
+        res1, res2 = df.align(s, axis=0)
+        exp1 = DataFrame(
+            {"a": [1, np.nan, 3, np.nan, 5], "b": [1, np.nan, 3, np.nan, 5]},
+            index=list("ABCDE"),
+        )
+        exp2 = Series([1, 2, np.nan, 4, np.nan], index=list("ABCDE"), name="x")
+
+        tm.assert_frame_equal(res1, exp1)
+        tm.assert_series_equal(res2, exp2)
+
+        # series + frame
+        res1, res2 = s.align(df)
+        tm.assert_series_equal(res1, exp2)
+        tm.assert_frame_equal(res2, exp1)
+
+    def test_multiindex_align_to_series_with_common_index_level(self):
+        #  GH-46001
+        foo_index = Index([1, 2, 3], name="foo")
+        bar_index = Index([1, 2], name="bar")
+
+        series = Series([1, 2], index=bar_index, name="foo_series")
+        df = DataFrame(
+            {"col": np.arange(6)},
+            index=pd.MultiIndex.from_product([foo_index, bar_index]),
+        )
+
+        expected_r = Series([1, 2] * 3, index=df.index, name="foo_series")
+        result_l, result_r = df.align(series, axis=0)
+
+        tm.assert_frame_equal(result_l, df)
+        tm.assert_series_equal(result_r, expected_r)
+
+    def test_multiindex_align_to_series_with_common_index_level_missing_in_left(self):
+        #  GH-46001
+        foo_index = Index([1, 2, 3], name="foo")
+        bar_index = Index([1, 2], name="bar")
+
+        series = Series(
+            [1, 2, 3, 4], index=Index([1, 2, 3, 4], name="bar"), name="foo_series"
+        )
+        df = DataFrame(
+            {"col": np.arange(6)},
+            index=pd.MultiIndex.from_product([foo_index, bar_index]),
+        )
+
+        expected_r = Series([1, 2] * 3, index=df.index, name="foo_series")
+        result_l, result_r = df.align(series, axis=0)
+
+        tm.assert_frame_equal(result_l, df)
+        tm.assert_series_equal(result_r, expected_r)
+
+    def test_multiindex_align_to_series_with_common_index_level_missing_in_right(self):
+        #  GH-46001
+        foo_index = Index([1, 2, 3], name="foo")
+        bar_index = Index([1, 2, 3, 4], name="bar")
+
+        series = Series([1, 2], index=Index([1, 2], name="bar"), name="foo_series")
+        df = DataFrame(
+            {"col": np.arange(12)},
+            index=pd.MultiIndex.from_product([foo_index, bar_index]),
+        )
+
+        expected_r = Series(
+            [1, 2, np.nan, np.nan] * 3, index=df.index, name="foo_series"
+        )
+        result_l, result_r = df.align(series, axis=0)
+
+        tm.assert_frame_equal(result_l, df)
+        tm.assert_series_equal(result_r, expected_r)
+
+    def test_multiindex_align_to_series_with_common_index_level_missing_in_both(self):
+        #  GH-46001
+        foo_index = Index([1, 2, 3], name="foo")
+        bar_index = Index([1, 3, 4], name="bar")
+
+        series = Series(
+            [1, 2, 3], index=Index([1, 2, 4], name="bar"), name="foo_series"
+        )
+        df = DataFrame(
+            {"col": np.arange(9)},
+            index=pd.MultiIndex.from_product([foo_index, bar_index]),
+        )
+
+        expected_r = Series([1, np.nan, 3] * 3, index=df.index, name="foo_series")
+        result_l, result_r = df.align(series, axis=0)
+
+        tm.assert_frame_equal(result_l, df)
+        tm.assert_series_equal(result_r, expected_r)
+
+    def test_multiindex_align_to_series_with_common_index_level_non_unique_cols(self):
+        #  GH-46001
+        foo_index = Index([1, 2, 3], name="foo")
+        bar_index = Index([1, 2], name="bar")
+
+        series = Series([1, 2], index=bar_index, name="foo_series")
+        df = DataFrame(
+            np.arange(18).reshape(6, 3),
+            index=pd.MultiIndex.from_product([foo_index, bar_index]),
+        )
+        df.columns = ["cfoo", "cbar", "cfoo"]
+
+        expected = Series([1, 2] * 3, index=df.index, name="foo_series")
+        result_left, result_right = df.align(series, axis=0)
+
+        tm.assert_series_equal(result_right, expected)
+        tm.assert_index_equal(result_left.columns, df.columns)
+
+    def test_missing_axis_specification_exception(self):
+        df = DataFrame(np.arange(50).reshape((10, 5)))
+        series = Series(np.arange(5))
+
+        with pytest.raises(ValueError, match=r"axis=0 or 1"):
+            df.align(series)
+
+    def test_align_series_check_copy(self):
+        # GH#
+        df = DataFrame({0: [1, 2]})
+        ser = Series([1], name=0)
+        expected = ser.copy()
+        result, other = df.align(ser, axis=1)
+        ser.iloc[0] = 100
+        tm.assert_series_equal(other, expected)
+
+    def test_align_identical_different_object(self):
+        # GH#51032
+        df = DataFrame({"a": [1, 2]})
+        ser = Series([3, 4])
+        result, result2 = df.align(ser, axis=0)
+        tm.assert_frame_equal(result, df)
+        tm.assert_series_equal(result2, ser)
+        assert df is not result
+        assert ser is not result2
+
+    def test_align_identical_different_object_columns(self):
+        # GH#51032
+        df = DataFrame({"a": [1, 2]})
+        ser = Series([1], index=["a"])
+        result, result2 = df.align(ser, axis=1)
+        tm.assert_frame_equal(result, df)
+        tm.assert_series_equal(result2, ser)
+        assert df is not result
+        assert ser is not result2
diff --git a/pandas/pandas/tests/frame/methods/test_asfreq.py b/pandas/pandas/tests/frame/methods/test_asfreq.py
new file mode 100644
index 0000000000000..1c3c41e2e0299
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_asfreq.py
@@ -0,0 +1,296 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import MonthEnd
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+class TestAsFreq:
+    def test_asfreq2(self, frame_or_series):
+        ts = frame_or_series(
+            [0.0, 1.0, 2.0],
+            index=DatetimeIndex(
+                [
+                    datetime(2009, 10, 30),
+                    datetime(2009, 11, 30),
+                    datetime(2009, 12, 31),
+                ],
+                dtype="M8[ns]",
+                freq="BME",
+            ),
+        )
+
+        daily_ts = ts.asfreq("B")
+        monthly_ts = daily_ts.asfreq("BME")
+        tm.assert_equal(monthly_ts, ts)
+
+        daily_ts = ts.asfreq("B", method="pad")
+        monthly_ts = daily_ts.asfreq("BME")
+        tm.assert_equal(monthly_ts, ts)
+
+        daily_ts = ts.asfreq(offsets.BDay())
+        monthly_ts = daily_ts.asfreq(offsets.BMonthEnd())
+        tm.assert_equal(monthly_ts, ts)
+
+        result = ts[:0].asfreq("ME")
+        assert len(result) == 0
+        assert result is not ts
+
+        if frame_or_series is Series:
+            daily_ts = ts.asfreq("D", fill_value=-1)
+            result = daily_ts.value_counts().sort_index()
+            expected = Series(
+                [60, 1, 1, 1], index=[-1.0, 2.0, 1.0, 0.0], name="count"
+            ).sort_index()
+            tm.assert_series_equal(result, expected)
+
+    def test_asfreq_datetimeindex_empty(self, frame_or_series):
+        # GH#14320
+        index = DatetimeIndex(["2016-09-29 11:00"])
+        expected = frame_or_series(index=index, dtype=object).asfreq("h")
+        result = frame_or_series([3], index=index.copy()).asfreq("h")
+        tm.assert_index_equal(expected.index, result.index)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_tz_aware_asfreq_smoke(self, tz, frame_or_series):
+        dr = date_range("2011-12-01", "2012-07-20", freq="D", tz=tz)
+
+        obj = frame_or_series(
+            np.random.default_rng(2).standard_normal(len(dr)), index=dr
+        )
+
+        # it works!
+        obj.asfreq("min")
+
+    def test_asfreq_normalize(self, frame_or_series):
+        rng = date_range("1/1/2000 09:30", periods=20)
+        norm = date_range("1/1/2000", periods=20)
+
+        vals = np.random.default_rng(2).standard_normal((20, 3))
+
+        obj = DataFrame(vals, index=rng)
+        expected = DataFrame(vals, index=norm)
+        if frame_or_series is Series:
+            obj = obj[0]
+            expected = expected[0]
+
+        result = obj.asfreq("D", normalize=True)
+        tm.assert_equal(result, expected)
+
+    def test_asfreq_keep_index_name(self, frame_or_series):
+        # GH#9854
+        index_name = "bar"
+        index = date_range("20130101", periods=20, name=index_name)
+        obj = DataFrame(list(range(20)), columns=["foo"], index=index)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        assert index_name == obj.index.name
+        assert index_name == obj.asfreq("10D").index.name
+
+    def test_asfreq_ts(self, frame_or_series):
+        index = period_range(freq="Y", start="1/1/2001", end="12/31/2010")
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 3)), index=index
+        )
+        obj = tm.get_obj(obj, frame_or_series)
+
+        result = obj.asfreq("D", how="end")
+        exp_index = index.asfreq("D", how="end")
+        assert len(result) == len(obj)
+        tm.assert_index_equal(result.index, exp_index)
+
+        result = obj.asfreq("D", how="start")
+        exp_index = index.asfreq("D", how="start")
+        assert len(result) == len(obj)
+        tm.assert_index_equal(result.index, exp_index)
+
+    def test_asfreq_resample_set_correct_freq(self, frame_or_series):
+        # GH#5613
+        # we test if .asfreq() and .resample() set the correct value for .freq
+        dti = to_datetime(["2012-01-01", "2012-01-02", "2012-01-03"])
+        obj = DataFrame({"col": [1, 2, 3]}, index=dti)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        # testing the settings before calling .asfreq() and .resample()
+        assert obj.index.freq is None
+        assert obj.index.inferred_freq == "D"
+
+        # does .asfreq() set .freq correctly?
+        assert obj.asfreq("D").index.freq == "D"
+
+        # does .resample() set .freq correctly?
+        assert obj.resample("D").asfreq().index.freq == "D"
+
+    def test_asfreq_empty(self, datetime_frame):
+        # test does not blow up on length-0 DataFrame
+        zero_length = datetime_frame.reindex([])
+        result = zero_length.asfreq("BME")
+        assert result is not zero_length
+
+    def test_asfreq(self, datetime_frame):
+        offset_monthly = datetime_frame.asfreq(offsets.BMonthEnd())
+        rule_monthly = datetime_frame.asfreq("BME")
+
+        tm.assert_frame_equal(offset_monthly, rule_monthly)
+
+        rule_monthly.asfreq("B", method="pad")
+        # TODO: actually check that this worked.
+
+        # don't forget!
+        rule_monthly.asfreq("B", method="pad")
+
+    def test_asfreq_datetimeindex(self):
+        df = DataFrame(
+            {"A": [1, 2, 3]},
+            index=[datetime(2011, 11, 1), datetime(2011, 11, 2), datetime(2011, 11, 3)],
+        )
+        df = df.asfreq("B")
+        assert isinstance(df.index, DatetimeIndex)
+
+        ts = df["A"].asfreq("B")
+        assert isinstance(ts.index, DatetimeIndex)
+
+    def test_asfreq_fillvalue(self):
+        # test for fill value during upsampling, related to issue 3715
+
+        # setup
+        rng = date_range("1/1/2016", periods=10, freq="2s")
+        # Explicit cast to 'float' to avoid implicit cast when setting None
+        ts = Series(np.arange(len(rng)), index=rng, dtype="float")
+        df = DataFrame({"one": ts})
+
+        # insert pre-existing missing value
+        df.loc["2016-01-01 00:00:08", "one"] = None
+
+        actual_df = df.asfreq(freq="1s", fill_value=9.0)
+        expected_df = df.asfreq(freq="1s").fillna(9.0)
+        expected_df.loc["2016-01-01 00:00:08", "one"] = None
+        tm.assert_frame_equal(expected_df, actual_df)
+
+        expected_series = ts.asfreq(freq="1s").fillna(9.0)
+        actual_series = ts.asfreq(freq="1s", fill_value=9.0)
+        tm.assert_series_equal(expected_series, actual_series)
+
+    def test_asfreq_with_date_object_index(self, frame_or_series):
+        rng = date_range("1/1/2000", periods=20)
+        ts = frame_or_series(np.random.default_rng(2).standard_normal(20), index=rng)
+
+        ts2 = ts.copy()
+        ts2.index = [x.date() for x in ts2.index]
+
+        result = ts2.asfreq("4h", method="ffill")
+        expected = ts.asfreq("4h", method="ffill")
+        tm.assert_equal(result, expected)
+
+    def test_asfreq_with_unsorted_index(self, frame_or_series):
+        # GH#39805
+        # Test that rows are not dropped when the datetime index is out of order
+        index = to_datetime(["2021-01-04", "2021-01-02", "2021-01-03", "2021-01-01"])
+        result = frame_or_series(range(4), index=index)
+
+        expected = result.reindex(sorted(index))
+        expected.index = expected.index._with_freq("infer")
+
+        result = result.asfreq("D")
+        tm.assert_equal(result, expected)
+
+    def test_asfreq_after_normalize(self, unit):
+        # https://github.com/pandas-dev/pandas/issues/50727
+        result = DatetimeIndex(
+            date_range("2000", periods=2).as_unit(unit).normalize(), freq="D"
+        )
+        expected = DatetimeIndex(["2000-01-01", "2000-01-02"], freq="D").as_unit(unit)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "freq, freq_half",
+        [
+            ("2ME", "ME"),
+            (MonthEnd(2), MonthEnd(1)),
+        ],
+    )
+    def test_asfreq_2ME(self, freq, freq_half):
+        index = date_range("1/1/2000", periods=6, freq=freq_half)
+        df = DataFrame({"s": Series([0.0, 1.0, 2.0, 3.0, 4.0, 5.0], index=index)})
+        expected = df.asfreq(freq=freq)
+
+        index = date_range("1/1/2000", periods=3, freq=freq)
+        result = DataFrame({"s": Series([0.0, 2.0, 4.0], index=index)})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "freq, freq_depr",
+        [
+            ("2ME", "2M"),
+            ("2ME", "2m"),
+            ("2QE", "2Q"),
+            ("2QE-SEP", "2Q-SEP"),
+            ("1BQE", "1BQ"),
+            ("2BQE-SEP", "2BQ-SEP"),
+            ("2BQE-SEP", "2bq-sep"),
+            ("1YE", "1y"),
+            ("2YE-MAR", "2Y-MAR"),
+        ],
+    )
+    def test_asfreq_frequency_M_Q_Y_raises(self, freq, freq_depr):
+        msg = f"Invalid frequency: {freq_depr}"
+
+        index = date_range("1/1/2000", periods=4, freq=f"{freq[1:]}")
+        df = DataFrame({"s": Series([0.0, 1.0, 2.0, 3.0], index=index)})
+        with pytest.raises(ValueError, match=msg):
+            df.asfreq(freq=freq_depr)
+
+    @pytest.mark.parametrize(
+        "freq, error_msg",
+        [
+            (
+                "2MS",
+                "Invalid frequency: 2MS",
+            ),
+            (
+                offsets.MonthBegin(),
+                r"\<MonthBegin\> is not supported as period frequency",
+            ),
+            (
+                offsets.DateOffset(months=2),
+                r"\<DateOffset: months=2\> is not supported as period frequency",
+            ),
+        ],
+    )
+    def test_asfreq_unsupported_freq(self, freq, error_msg):
+        # https://github.com/pandas-dev/pandas/issues/56718
+        index = PeriodIndex(["2020-01-01", "2021-01-01"], freq="M")
+        df = DataFrame({"a": Series([0, 1], index=index)})
+
+        with pytest.raises(ValueError, match=error_msg):
+            df.asfreq(freq=freq)
+
+    @pytest.mark.parametrize(
+        "freq, freq_depr",
+        [
+            ("2YE", "2A"),
+            ("2BYE-MAR", "2BA-MAR"),
+        ],
+    )
+    def test_asfreq_frequency_A_BA_raises(self, freq, freq_depr):
+        msg = f"Invalid frequency: {freq_depr}"
+
+        index = date_range("1/1/2000", periods=4, freq=freq)
+        df = DataFrame({"s": Series([0.0, 1.0, 2.0, 3.0], index=index)})
+
+        with pytest.raises(ValueError, match=msg):
+            df.asfreq(freq=freq_depr)
diff --git a/pandas/pandas/tests/frame/methods/test_asof.py b/pandas/pandas/tests/frame/methods/test_asof.py
new file mode 100644
index 0000000000000..c510ef78d03aa
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_asof.py
@@ -0,0 +1,185 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import IncompatibleFrequency
+
+from pandas import (
+    DataFrame,
+    Period,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def date_range_frame():
+    """
+    Fixture for DataFrame of ints with date_range index
+
+    Columns are ['A', 'B'].
+    """
+    N = 50
+    rng = date_range("1/1/1990", periods=N, freq="53s")
+    return DataFrame({"A": np.arange(N), "B": np.arange(N)}, index=rng)
+
+
+class TestFrameAsof:
+    def test_basic(self, date_range_frame):
+        # Explicitly cast to float to avoid implicit cast when setting np.nan
+        df = date_range_frame.astype({"A": "float"})
+        N = 50
+        df.loc[df.index[15:30], "A"] = np.nan
+        dates = date_range("1/1/1990", periods=N * 3, freq="25s")
+
+        result = df.asof(dates)
+        assert result.notna().all(axis=1).all()
+        lb = df.index[14]
+        ub = df.index[30]
+
+        dates = list(dates)
+
+        result = df.asof(dates)
+        assert result.notna().all(axis=1).all()
+
+        mask = (result.index >= lb) & (result.index < ub)
+        rs = result[mask]
+        assert (rs == 14).all(axis=1).all()
+
+    def test_subset(self, date_range_frame):
+        N = 10
+        # explicitly cast to float to avoid implicit upcast when setting to np.nan
+        df = date_range_frame.iloc[:N].copy().astype({"A": "float"})
+        df.loc[df.index[4:8], "A"] = np.nan
+        dates = date_range("1/1/1990", periods=N * 3, freq="25s")
+
+        # with a subset of A should be the same
+        result = df.asof(dates, subset="A")
+        expected = df.asof(dates)
+        tm.assert_frame_equal(result, expected)
+
+        # same with A/B
+        result = df.asof(dates, subset=["A", "B"])
+        expected = df.asof(dates)
+        tm.assert_frame_equal(result, expected)
+
+        # B gives df.asof
+        result = df.asof(dates, subset="B")
+        expected = df.resample("25s", closed="right").ffill().reindex(dates)
+        expected.iloc[20:] = 9
+        # no "missing", so "B" can retain int dtype (df["A"].dtype platform-dependent)
+        expected["B"] = expected["B"].astype(df["B"].dtype)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_missing(self, date_range_frame):
+        # GH 15118
+        # no match found - `where` value before earliest date in index
+        N = 10
+        # Cast to 'float64' to avoid upcast when introducing nan in df.asof
+        df = date_range_frame.iloc[:N].copy().astype("float64")
+
+        result = df.asof("1989-12-31")
+
+        expected = Series(
+            index=["A", "B"], name=Timestamp("1989-12-31"), dtype=np.float64
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = df.asof(to_datetime(["1989-12-31"]))
+        expected = DataFrame(
+            index=to_datetime(["1989-12-31"]), columns=["A", "B"], dtype="float64"
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Check that we handle PeriodIndex correctly, dont end up with
+        #  period.ordinal for series name
+        df = df.to_period("D")
+        result = df.asof("1989-12-31")
+        assert isinstance(result.name, Period)
+
+    def test_asof_all_nans(self, frame_or_series):
+        # GH 15713
+        # DataFrame/Series is all nans
+        result = frame_or_series([np.nan]).asof([0])
+        expected = frame_or_series([np.nan])
+        tm.assert_equal(result, expected)
+
+    def test_all_nans(self, date_range_frame):
+        # GH 15713
+        # DataFrame is all nans
+
+        # testing non-default indexes, multiple inputs
+        N = 150
+        rng = date_range_frame.index
+        dates = date_range("1/1/1990", periods=N, freq="25s")
+        result = DataFrame(np.nan, index=rng, columns=["A"]).asof(dates)
+        expected = DataFrame(np.nan, index=dates, columns=["A"])
+        tm.assert_frame_equal(result, expected)
+
+        # testing multiple columns
+        dates = date_range("1/1/1990", periods=N, freq="25s")
+        result = DataFrame(np.nan, index=rng, columns=["A", "B", "C"]).asof(dates)
+        expected = DataFrame(np.nan, index=dates, columns=["A", "B", "C"])
+        tm.assert_frame_equal(result, expected)
+
+        # testing scalar input
+        result = DataFrame(np.nan, index=[1, 2], columns=["A", "B"]).asof([3])
+        expected = DataFrame(np.nan, index=[3], columns=["A", "B"])
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(np.nan, index=[1, 2], columns=["A", "B"]).asof(3)
+        expected = Series(np.nan, index=["A", "B"], name=3)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "stamp,expected",
+        [
+            (
+                Timestamp("2018-01-01 23:22:43.325+00:00"),
+                Series(2, name=Timestamp("2018-01-01 23:22:43.325+00:00")),
+            ),
+            (
+                Timestamp("2018-01-01 22:33:20.682+01:00"),
+                Series(1, name=Timestamp("2018-01-01 22:33:20.682+01:00")),
+            ),
+        ],
+    )
+    def test_time_zone_aware_index(self, stamp, expected):
+        # GH21194
+        # Testing awareness of DataFrame index considering different
+        # UTC and timezone
+        df = DataFrame(
+            data=[1, 2],
+            index=[
+                Timestamp("2018-01-01 21:00:05.001+00:00"),
+                Timestamp("2018-01-01 22:35:10.550+00:00"),
+            ],
+        )
+
+        result = df.asof(stamp)
+        tm.assert_series_equal(result, expected)
+
+    def test_asof_periodindex_mismatched_freq(self):
+        N = 50
+        rng = period_range("1/1/1990", periods=N, freq="h")
+        df = DataFrame(np.random.default_rng(2).standard_normal(N), index=rng)
+
+        # Mismatched freq
+        msg = "Input has different freq"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            df.asof(rng.asfreq("D"))
+
+    def test_asof_preserves_bool_dtype(self):
+        # GH#16063 was casting bools to floats
+        dti = date_range("2017-01-01", freq="MS", periods=4)
+        ser = Series([True, False, True], index=dti[:-1])
+
+        ts = dti[-1]
+        res = ser.asof([ts])
+
+        expected = Series([True], index=[ts])
+        tm.assert_series_equal(res, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_assign.py b/pandas/pandas/tests/frame/methods/test_assign.py
new file mode 100644
index 0000000000000..0ae501d43e742
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_assign.py
@@ -0,0 +1,84 @@
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestAssign:
+    def test_assign(self):
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        original = df.copy()
+        result = df.assign(C=df.B / df.A)
+        expected = df.copy()
+        expected["C"] = [4, 2.5, 2]
+        tm.assert_frame_equal(result, expected)
+
+        # lambda syntax
+        result = df.assign(C=lambda x: x.B / x.A)
+        tm.assert_frame_equal(result, expected)
+
+        # original is unmodified
+        tm.assert_frame_equal(df, original)
+
+        # Non-Series array-like
+        result = df.assign(C=[4, 2.5, 2])
+        tm.assert_frame_equal(result, expected)
+        # original is unmodified
+        tm.assert_frame_equal(df, original)
+
+        result = df.assign(B=df.B / df.A)
+        expected = expected.drop("B", axis=1).rename(columns={"C": "B"})
+        tm.assert_frame_equal(result, expected)
+
+        # overwrite
+        result = df.assign(A=df.A + df.B)
+        expected = df.copy()
+        expected["A"] = [5, 7, 9]
+        tm.assert_frame_equal(result, expected)
+
+        # lambda
+        result = df.assign(A=lambda x: x.A + x.B)
+        tm.assert_frame_equal(result, expected)
+
+    def test_assign_multiple(self):
+        df = DataFrame([[1, 4], [2, 5], [3, 6]], columns=["A", "B"])
+        result = df.assign(C=[7, 8, 9], D=df.A, E=lambda x: x.B)
+        expected = DataFrame(
+            [[1, 4, 7, 1, 4], [2, 5, 8, 2, 5], [3, 6, 9, 3, 6]], columns=list("ABCDE")
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_assign_order(self):
+        # GH 9818
+        df = DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        result = df.assign(D=df.A + df.B, C=df.A - df.B)
+
+        expected = DataFrame([[1, 2, 3, -1], [3, 4, 7, -1]], columns=list("ABDC"))
+        tm.assert_frame_equal(result, expected)
+        result = df.assign(C=df.A - df.B, D=df.A + df.B)
+
+        expected = DataFrame([[1, 2, -1, 3], [3, 4, -1, 7]], columns=list("ABCD"))
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_assign_bad(self):
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+
+        # non-keyword argument
+        msg = r"assign\(\) takes 1 positional argument but 2 were given"
+        with pytest.raises(TypeError, match=msg):
+            df.assign(lambda x: x.A)
+        msg = "'DataFrame' object has no attribute 'C'"
+        with pytest.raises(AttributeError, match=msg):
+            df.assign(C=df.A, D=df.A + df.C)
+
+    def test_assign_dependent(self):
+        df = DataFrame({"A": [1, 2], "B": [3, 4]})
+
+        result = df.assign(C=df.A, D=lambda x: x["A"] + x["C"])
+        expected = DataFrame([[1, 3, 1, 2], [2, 4, 2, 4]], columns=list("ABCD"))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.assign(C=lambda df: df.A, D=lambda df: df["A"] + df["C"])
+        expected = DataFrame([[1, 3, 1, 2], [2, 4, 2, 4]], columns=list("ABCD"))
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_astype.py b/pandas/pandas/tests/frame/methods/test_astype.py
new file mode 100644
index 0000000000000..eb1ee4e7b2970
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_astype.py
@@ -0,0 +1,909 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    DataFrame,
+    DatetimeTZDtype,
+    Index,
+    Interval,
+    IntervalDtype,
+    NaT,
+    Series,
+    Timedelta,
+    Timestamp,
+    concat,
+    date_range,
+    option_context,
+)
+import pandas._testing as tm
+
+
+def _check_cast(df, v):
+    """
+    Check if all dtypes of df are equal to v
+    """
+    assert all(s.dtype.name == v for _, s in df.items())
+
+
+class TestAstype:
+    def test_astype_float(self, float_frame):
+        casted = float_frame.astype(int)
+        expected = DataFrame(
+            float_frame.values.astype(int),
+            index=float_frame.index,
+            columns=float_frame.columns,
+        )
+        tm.assert_frame_equal(casted, expected)
+
+        casted = float_frame.astype(np.int32)
+        expected = DataFrame(
+            float_frame.values.astype(np.int32),
+            index=float_frame.index,
+            columns=float_frame.columns,
+        )
+        tm.assert_frame_equal(casted, expected)
+
+        float_frame["foo"] = "5"
+        casted = float_frame.astype(int)
+        expected = DataFrame(
+            float_frame.values.astype(int),
+            index=float_frame.index,
+            columns=float_frame.columns,
+        )
+        tm.assert_frame_equal(casted, expected)
+
+    def test_astype_mixed_float(self, mixed_float_frame):
+        # mixed casting
+        casted = mixed_float_frame.reindex(columns=["A", "B"]).astype("float32")
+        _check_cast(casted, "float32")
+
+        casted = mixed_float_frame.reindex(columns=["A", "B"]).astype("float16")
+        _check_cast(casted, "float16")
+
+    def test_astype_mixed_type(self):
+        # mixed casting
+        df = DataFrame(
+            {
+                "a": 1.0,
+                "b": 2,
+                "c": "foo",
+                "float32": np.array([1.0] * 10, dtype="float32"),
+                "int32": np.array([1] * 10, dtype="int32"),
+            },
+            index=np.arange(10),
+        )
+        mn = df._get_numeric_data().copy()
+        mn["little_float"] = np.array(12345.0, dtype="float16")
+        mn["big_float"] = np.array(123456789101112.0, dtype="float64")
+
+        casted = mn.astype("float64")
+        _check_cast(casted, "float64")
+
+        casted = mn.astype("int64")
+        _check_cast(casted, "int64")
+
+        casted = mn.reindex(columns=["little_float"]).astype("float16")
+        _check_cast(casted, "float16")
+
+        casted = mn.astype("float32")
+        _check_cast(casted, "float32")
+
+        casted = mn.astype("int32")
+        _check_cast(casted, "int32")
+
+        # to object
+        casted = mn.astype("O")
+        _check_cast(casted, "object")
+
+    def test_astype_with_exclude_string(self, float_frame):
+        df = float_frame.copy()
+        expected = float_frame.astype(int)
+        df["string"] = "foo"
+        casted = df.astype(int, errors="ignore")
+
+        expected["string"] = "foo"
+        tm.assert_frame_equal(casted, expected)
+
+        df = float_frame.copy()
+        expected = float_frame.astype(np.int32)
+        df["string"] = "foo"
+        casted = df.astype(np.int32, errors="ignore")
+
+        expected["string"] = "foo"
+        tm.assert_frame_equal(casted, expected)
+
+    def test_astype_with_view_float(self, float_frame):
+        # this is the only real reason to do it this way
+        tf = np.round(float_frame).astype(np.int32)
+        tf.astype(np.float32)
+
+        # TODO(wesm): verification?
+        tf = float_frame.astype(np.float64)
+        tf.astype(np.int64)
+
+    def test_astype_with_view_mixed_float(self, mixed_float_frame):
+        tf = mixed_float_frame.reindex(columns=["A", "B", "C"])
+
+        tf.astype(np.int64)
+        tf.astype(np.float32)
+
+    @pytest.mark.parametrize("val", [np.nan, np.inf])
+    def test_astype_cast_nan_inf_int(self, val, any_int_numpy_dtype):
+        # see GH#14265
+        #
+        # Check NaN and inf --> raise error when converting to int.
+        msg = "Cannot convert non-finite values \\(NA or inf\\) to integer"
+        df = DataFrame([val])
+
+        with pytest.raises(ValueError, match=msg):
+            df.astype(any_int_numpy_dtype)
+
+    def test_astype_str(self):
+        # see GH#9757
+        a = Series(date_range("2010-01-04", periods=5))
+        b = Series(date_range("3/6/2012 00:00", periods=5, tz="US/Eastern"))
+        c = Series([Timedelta(x, unit="D") for x in range(5)])
+        d = Series(range(5))
+        e = Series([0.0, 0.2, 0.4, 0.6, 0.8])
+
+        df = DataFrame({"a": a, "b": b, "c": c, "d": d, "e": e})
+
+        # Datetime-like
+        result = df.astype(str)
+
+        expected = DataFrame(
+            {
+                "a": list(map(str, (Timestamp(x)._date_repr for x in a._values))),
+                "b": list(map(str, map(Timestamp, b._values))),
+                "c": [Timedelta(x)._repr_base() for x in c._values],
+                "d": list(map(str, d._values)),
+                "e": list(map(str, e._values)),
+            },
+            dtype="str",
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_astype_str_float(self, using_infer_string):
+        # see GH#11302
+        result = DataFrame([np.nan]).astype(str)
+        expected = DataFrame([np.nan if using_infer_string else "nan"], dtype="str")
+
+        tm.assert_frame_equal(result, expected)
+        result = DataFrame([1.12345678901234567890]).astype(str)
+
+        val = "1.1234567890123457"
+        expected = DataFrame([val], dtype="str")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype_class", [dict, Series])
+    def test_astype_dict_like(self, dtype_class):
+        # GH7271 & GH16717
+        a = Series(date_range("2010-01-04", periods=5))
+        b = Series(range(5))
+        c = Series([0.0, 0.2, 0.4, 0.6, 0.8])
+        d = Series(["1.0", "2", "3.14", "4", "5.4"])
+        df = DataFrame({"a": a, "b": b, "c": c, "d": d})
+        original = df.copy(deep=True)
+
+        # change type of a subset of columns
+        dt1 = dtype_class({"b": "str", "d": "float32"})
+        result = df.astype(dt1)
+        expected = DataFrame(
+            {
+                "a": a,
+                "b": Series(["0", "1", "2", "3", "4"], dtype="str"),
+                "c": c,
+                "d": Series([1.0, 2.0, 3.14, 4.0, 5.4], dtype="float32"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(df, original)
+
+        dt2 = dtype_class({"b": np.float32, "c": "float32", "d": np.float64})
+        result = df.astype(dt2)
+        expected = DataFrame(
+            {
+                "a": a,
+                "b": Series([0.0, 1.0, 2.0, 3.0, 4.0], dtype="float32"),
+                "c": Series([0.0, 0.2, 0.4, 0.6, 0.8], dtype="float32"),
+                "d": Series([1.0, 2.0, 3.14, 4.0, 5.4], dtype="float64"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(df, original)
+
+        # change all columns
+        dt3 = dtype_class({"a": str, "b": str, "c": str, "d": str})
+        tm.assert_frame_equal(df.astype(dt3), df.astype(str))
+        tm.assert_frame_equal(df, original)
+
+        # error should be raised when using something other than column labels
+        # in the keys of the dtype dict
+        dt4 = dtype_class({"b": str, 2: str})
+        dt5 = dtype_class({"e": str})
+        msg_frame = (
+            "Only a column name can be used for the key in a dtype mappings argument. "
+            "'{}' not found in columns."
+        )
+        with pytest.raises(KeyError, match=msg_frame.format(2)):
+            df.astype(dt4)
+        with pytest.raises(KeyError, match=msg_frame.format("e")):
+            df.astype(dt5)
+        tm.assert_frame_equal(df, original)
+
+        # if the dtypes provided are the same as the original dtypes, the
+        # resulting DataFrame should be the same as the original DataFrame
+        dt6 = dtype_class({col: df[col].dtype for col in df.columns})
+        equiv = df.astype(dt6)
+        tm.assert_frame_equal(df, equiv)
+        tm.assert_frame_equal(df, original)
+
+        # GH#16717
+        # if dtypes provided is empty, the resulting DataFrame
+        # should be the same as the original DataFrame
+        dt7 = dtype_class({}) if dtype_class is dict else dtype_class({}, dtype=object)
+        equiv = df.astype(dt7)
+        tm.assert_frame_equal(df, equiv)
+        tm.assert_frame_equal(df, original)
+
+    def test_astype_duplicate_col(self):
+        a1 = Series([1, 2, 3, 4, 5], name="a")
+        b = Series([0.1, 0.2, 0.4, 0.6, 0.8], name="b")
+        a2 = Series([0, 1, 2, 3, 4], name="a")
+        df = concat([a1, b, a2], axis=1)
+
+        result = df.astype("str")
+        a1_str = Series(["1", "2", "3", "4", "5"], dtype="str", name="a")
+        b_str = Series(["0.1", "0.2", "0.4", "0.6", "0.8"], dtype="str", name="b")
+        a2_str = Series(["0", "1", "2", "3", "4"], dtype="str", name="a")
+        expected = concat([a1_str, b_str, a2_str], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.astype({"a": "str"})
+        expected = concat([a1_str, b, a2_str], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_astype_duplicate_col_series_arg(self):
+        # GH#44417
+        vals = np.random.default_rng(2).standard_normal((3, 4))
+        df = DataFrame(vals, columns=["A", "B", "C", "A"])
+        dtypes = df.dtypes
+        dtypes.iloc[0] = str
+        dtypes.iloc[2] = "Float64"
+
+        result = df.astype(dtypes)
+        expected = DataFrame(
+            {
+                0: Series(vals[:, 0].astype(str), dtype="str"),
+                1: vals[:, 1],
+                2: pd.array(vals[:, 2], dtype="Float64"),
+                3: vals[:, 3],
+            }
+        )
+        expected.columns = df.columns
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "category",
+            CategoricalDtype(),
+            CategoricalDtype(ordered=True),
+            CategoricalDtype(ordered=False),
+            CategoricalDtype(categories=list("abcdef")),
+            CategoricalDtype(categories=list("edba"), ordered=False),
+            CategoricalDtype(categories=list("edcb"), ordered=True),
+        ],
+        ids=repr,
+    )
+    def test_astype_categorical(self, dtype):
+        # GH#18099
+        d = {"A": list("abbc"), "B": list("bccd"), "C": list("cdde")}
+        df = DataFrame(d)
+        result = df.astype(dtype)
+        expected = DataFrame({k: Categorical(v, dtype=dtype) for k, v in d.items()})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("cls", [CategoricalDtype, DatetimeTZDtype, IntervalDtype])
+    def test_astype_categoricaldtype_class_raises(self, cls):
+        df = DataFrame({"A": ["a", "a", "b", "c"]})
+        xpr = f"Expected an instance of {cls.__name__}"
+        with pytest.raises(TypeError, match=xpr):
+            df.astype({"A": cls})
+
+        with pytest.raises(TypeError, match=xpr):
+            df["A"].astype(cls)
+
+    def test_astype_extension_dtypes(self, any_int_ea_dtype):
+        # GH#22578
+        dtype = any_int_ea_dtype
+        df = DataFrame([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]], columns=["a", "b"])
+
+        expected1 = DataFrame(
+            {
+                "a": pd.array([1, 3, 5], dtype=dtype),
+                "b": pd.array([2, 4, 6], dtype=dtype),
+            }
+        )
+        tm.assert_frame_equal(df.astype(dtype), expected1)
+        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)
+        tm.assert_frame_equal(df.astype(dtype).astype("float64"), df)
+
+        df = DataFrame([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]], columns=["a", "b"])
+        df["b"] = df["b"].astype(dtype)
+        expected2 = DataFrame(
+            {"a": [1.0, 3.0, 5.0], "b": pd.array([2, 4, 6], dtype=dtype)}
+        )
+        tm.assert_frame_equal(df, expected2)
+
+        tm.assert_frame_equal(df.astype(dtype), expected1)
+        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)
+
+    def test_astype_extension_dtypes_1d(self, any_int_ea_dtype):
+        # GH#22578
+        dtype = any_int_ea_dtype
+        df = DataFrame({"a": [1.0, 2.0, 3.0]})
+
+        expected1 = DataFrame({"a": pd.array([1, 2, 3], dtype=dtype)})
+        tm.assert_frame_equal(df.astype(dtype), expected1)
+        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)
+
+        df = DataFrame({"a": [1.0, 2.0, 3.0]})
+        df["a"] = df["a"].astype(dtype)
+        expected2 = DataFrame({"a": pd.array([1, 2, 3], dtype=dtype)})
+        tm.assert_frame_equal(df, expected2)
+
+        tm.assert_frame_equal(df.astype(dtype), expected1)
+        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)
+
+    @pytest.mark.parametrize("dtype", ["category", "Int64"])
+    def test_astype_extension_dtypes_duplicate_col(self, dtype):
+        # GH#24704
+        a1 = Series([0, np.nan, 4], name="a")
+        a2 = Series([np.nan, 3, 5], name="a")
+        df = concat([a1, a2], axis=1)
+
+        result = df.astype(dtype)
+        expected = concat([a1.astype(dtype), a2.astype(dtype)], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype", [{100: "float64", 200: "uint64"}, "category", "float64"]
+    )
+    def test_astype_column_metadata(self, dtype):
+        # GH#19920
+        columns = Index([100, 200, 300], dtype=np.uint64, name="foo")
+        df = DataFrame(np.arange(15).reshape(5, 3), columns=columns)
+        df = df.astype(dtype)
+        tm.assert_index_equal(df.columns, columns)
+
+    @pytest.mark.parametrize("unit", ["Y", "M", "W", "D", "h", "m"])
+    def test_astype_from_object_to_datetime_unit(self, unit):
+        vals = [
+            ["2015-01-01", "2015-01-02", "2015-01-03"],
+            ["2017-01-01", "2017-01-02", "2017-02-03"],
+        ]
+        df = DataFrame(vals, dtype=object)
+        msg = (
+            rf"Unexpected value for 'dtype': 'datetime64\[{unit}\]'. "
+            r"Must be 'datetime64\[s\]', 'datetime64\[ms\]', 'datetime64\[us\]', "
+            r"'datetime64\[ns\]' or DatetimeTZDtype"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.astype(f"M8[{unit}]")
+
+    @pytest.mark.parametrize("unit", ["Y", "M", "W", "D", "h", "m"])
+    def test_astype_from_object_to_timedelta_unit(self, unit):
+        vals = [
+            ["1 Day", "2 Days", "3 Days"],
+            ["4 Days", "5 Days", "6 Days"],
+        ]
+        df = DataFrame(vals, dtype=object)
+        msg = (
+            r"Cannot convert from timedelta64\[ns\] to timedelta64\[.*\]. "
+            "Supported resolutions are 's', 'ms', 'us', 'ns'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            # TODO: this is ValueError while for DatetimeArray it is TypeError;
+            #  get these consistent
+            df.astype(f"m8[{unit}]")
+
+    @pytest.mark.parametrize("dtype", ["M8", "m8"])
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
+    def test_astype_from_datetimelike_to_object(self, dtype, unit):
+        # tests astype to object dtype
+        # GH#19223 / GH#12425
+        dtype = f"{dtype}[{unit}]"
+        arr = np.array([[1, 2, 3]], dtype=dtype)
+        df = DataFrame(arr)
+        result = df.astype(object)
+        assert (result.dtypes == object).all()
+
+        if dtype.startswith("M8"):
+            assert result.iloc[0, 0] == Timestamp(1, unit=unit)
+        else:
+            assert result.iloc[0, 0] == Timedelta(1, unit=unit)
+
+    @pytest.mark.parametrize("dtype", ["M8", "m8"])
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
+    def test_astype_to_datetimelike_unit(self, any_real_numpy_dtype, dtype, unit):
+        # tests all units from numeric origination
+        # GH#19223 / GH#12425
+        dtype = f"{dtype}[{unit}]"
+        arr = np.array([[1, 2, 3]], dtype=any_real_numpy_dtype)
+        df = DataFrame(arr)
+        result = df.astype(dtype)
+        expected = DataFrame(arr.astype(dtype))
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
+    def test_astype_to_datetime_unit(self, unit):
+        # tests all units from datetime origination
+        # GH#19223
+        dtype = f"M8[{unit}]"
+        arr = np.array([[1, 2, 3]], dtype=dtype)
+        df = DataFrame(arr)
+        ser = df.iloc[:, 0]
+        idx = Index(ser)
+        dta = ser._values
+
+        if unit in ["ns", "us", "ms", "s"]:
+            # GH#48928
+            result = df.astype(dtype)
+        else:
+            # we use the nearest supported dtype (i.e. M8[s])
+            msg = rf"Cannot cast DatetimeArray to dtype datetime64\[{unit}\]"
+            with pytest.raises(TypeError, match=msg):
+                df.astype(dtype)
+
+            with pytest.raises(TypeError, match=msg):
+                ser.astype(dtype)
+
+            with pytest.raises(TypeError, match=msg.replace("Array", "Index")):
+                idx.astype(dtype)
+
+            with pytest.raises(TypeError, match=msg):
+                dta.astype(dtype)
+
+            return
+
+        exp_df = DataFrame(arr.astype(dtype))
+        assert (exp_df.dtypes == dtype).all()
+        tm.assert_frame_equal(result, exp_df)
+
+        res_ser = ser.astype(dtype)
+        exp_ser = exp_df.iloc[:, 0]
+        assert exp_ser.dtype == dtype
+        tm.assert_series_equal(res_ser, exp_ser)
+
+        exp_dta = exp_ser._values
+
+        res_index = idx.astype(dtype)
+        exp_index = Index(exp_ser)
+        assert exp_index.dtype == dtype
+        tm.assert_index_equal(res_index, exp_index)
+
+        res_dta = dta.astype(dtype)
+        assert exp_dta.dtype == dtype
+        tm.assert_extension_array_equal(res_dta, exp_dta)
+
+    def test_astype_to_timedelta_unit_ns(self):
+        # preserver the timedelta conversion
+        # GH#19223
+        dtype = "m8[ns]"
+        arr = np.array([[1, 2, 3]], dtype=dtype)
+        df = DataFrame(arr)
+        result = df.astype(dtype)
+        expected = DataFrame(arr.astype(dtype))
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("unit", ["us", "ms", "s", "h", "m", "D"])
+    def test_astype_to_timedelta_unit(self, unit):
+        # coerce to float
+        # GH#19223 until 2.0 used to coerce to float
+        dtype = f"m8[{unit}]"
+        arr = np.array([[1, 2, 3]], dtype=dtype)
+        df = DataFrame(arr)
+        ser = df.iloc[:, 0]
+        tdi = Index(ser)
+        tda = tdi._values
+
+        if unit in ["us", "ms", "s"]:
+            assert (df.dtypes == dtype).all()
+            result = df.astype(dtype)
+        else:
+            # We get the nearest supported unit, i.e. "s"
+            assert (df.dtypes == "m8[s]").all()
+
+            msg = (
+                rf"Cannot convert from timedelta64\[s\] to timedelta64\[{unit}\]. "
+                "Supported resolutions are 's', 'ms', 'us', 'ns'"
+            )
+            with pytest.raises(ValueError, match=msg):
+                df.astype(dtype)
+            with pytest.raises(ValueError, match=msg):
+                ser.astype(dtype)
+            with pytest.raises(ValueError, match=msg):
+                tdi.astype(dtype)
+            with pytest.raises(ValueError, match=msg):
+                tda.astype(dtype)
+
+            return
+
+        result = df.astype(dtype)
+        # The conversion is a no-op, so we just get a copy
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
+    def test_astype_to_incorrect_datetimelike(self, unit):
+        # trying to astype a m to a M, or vice-versa
+        # GH#19224
+        dtype = f"M8[{unit}]"
+        other = f"m8[{unit}]"
+
+        df = DataFrame(np.array([[1, 2, 3]], dtype=dtype))
+        msg = rf"Cannot cast DatetimeArray to dtype timedelta64\[{unit}\]"
+        with pytest.raises(TypeError, match=msg):
+            df.astype(other)
+
+        msg = rf"Cannot cast TimedeltaArray to dtype datetime64\[{unit}\]"
+        df = DataFrame(np.array([[1, 2, 3]], dtype=other))
+        with pytest.raises(TypeError, match=msg):
+            df.astype(dtype)
+
+    def test_astype_arg_for_errors(self):
+        # GH#14878
+
+        df = DataFrame([1, 2, 3])
+
+        msg = (
+            "Expected value of kwarg 'errors' to be one of "
+            "['raise', 'ignore']. Supplied value is 'True'"
+        )
+        with pytest.raises(ValueError, match=re.escape(msg)):
+            df.astype(np.float64, errors=True)
+
+        df.astype(np.int8, errors="ignore")
+
+    def test_astype_invalid_conversion(self):
+        # GH#47571
+        df = DataFrame({"a": [1, 2, "text"], "b": [1, 2, 3]})
+
+        msg = (
+            "invalid literal for int() with base 10: 'text': "
+            "Error while type casting for column 'a'"
+        )
+
+        with pytest.raises(ValueError, match=re.escape(msg)):
+            df.astype({"a": int})
+
+    def test_astype_arg_for_errors_dictlist(self):
+        # GH#25905
+        df = DataFrame(
+            [
+                {"a": "1", "b": "16.5%", "c": "test"},
+                {"a": "2.2", "b": "15.3", "c": "another_test"},
+            ]
+        )
+        expected = DataFrame(
+            [
+                {"a": 1.0, "b": "16.5%", "c": "test"},
+                {"a": 2.2, "b": "15.3", "c": "another_test"},
+            ]
+        )
+        expected["c"] = expected["c"].astype("object")
+        type_dict = {"a": "float64", "b": "float64", "c": "object"}
+
+        result = df.astype(dtype=type_dict, errors="ignore")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_astype_dt64tz(self, timezone_frame):
+        # astype
+        expected = np.array(
+            [
+                [
+                    Timestamp("2013-01-01 00:00:00"),
+                    Timestamp("2013-01-02 00:00:00"),
+                    Timestamp("2013-01-03 00:00:00"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00-0500", tz="US/Eastern"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00+0100", tz="CET"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00+0100", tz="CET"),
+                ],
+            ],
+            dtype=object,
+        ).T
+        expected = DataFrame(
+            expected,
+            index=timezone_frame.index,
+            columns=timezone_frame.columns,
+            dtype=object,
+        )
+        result = timezone_frame.astype(object)
+        tm.assert_frame_equal(result, expected)
+
+        msg = "Cannot use .astype to convert from timezone-aware dtype to timezone-"
+        with pytest.raises(TypeError, match=msg):
+            # dt64tz->dt64 deprecated
+            timezone_frame.astype("datetime64[ns]")
+
+    def test_astype_dt64tz_to_str(self, timezone_frame, using_infer_string):
+        # str formatting
+        result = timezone_frame.astype(str)
+        na_value = np.nan if using_infer_string else "NaT"
+        expected = DataFrame(
+            [
+                [
+                    "2013-01-01",
+                    "2013-01-01 00:00:00-05:00",
+                    "2013-01-01 00:00:00+01:00",
+                ],
+                ["2013-01-02", na_value, na_value],
+                [
+                    "2013-01-03",
+                    "2013-01-03 00:00:00-05:00",
+                    "2013-01-03 00:00:00+01:00",
+                ],
+            ],
+            columns=timezone_frame.columns,
+            dtype="str",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        with option_context("display.max_columns", 20):
+            result = str(timezone_frame)
+            assert (
+                "0 2013-01-01 2013-01-01 00:00:00-05:00 2013-01-01 00:00:00+01:00"
+            ) in result
+            assert (
+                "1 2013-01-02                       NaT                       NaT"
+            ) in result
+            assert (
+                "2 2013-01-03 2013-01-03 00:00:00-05:00 2013-01-03 00:00:00+01:00"
+            ) in result
+
+    def test_astype_empty_dtype_dict(self):
+        # issue mentioned further down in the following issue's thread
+        # https://github.com/pandas-dev/pandas/issues/33113
+        df = DataFrame()
+        result = df.astype({})
+        tm.assert_frame_equal(result, df)
+        assert result is not df
+
+    @pytest.mark.parametrize(
+        "data, dtype",
+        [
+            (["x", "y", "z"], "string[python]"),
+            pytest.param(
+                ["x", "y", "z"],
+                "string[pyarrow]",
+                marks=td.skip_if_no("pyarrow"),
+            ),
+            (["x", "y", "z"], "category"),
+            (3 * [Timestamp("2020-01-01", tz="UTC")], None),
+            (3 * [Interval(0, 1)], None),
+        ],
+    )
+    @pytest.mark.parametrize("errors", ["raise", "ignore"])
+    def test_astype_ignores_errors_for_extension_dtypes(self, data, dtype, errors):
+        # https://github.com/pandas-dev/pandas/issues/35471
+        df = DataFrame(Series(data, dtype=dtype))
+        if errors == "ignore":
+            expected = df
+            result = df.astype(float, errors=errors)
+            tm.assert_frame_equal(result, expected)
+        else:
+            msg = "(Cannot cast)|(could not convert)"
+            with pytest.raises((ValueError, TypeError), match=msg):
+                df.astype(float, errors=errors)
+
+    def test_astype_tz_conversion(self):
+        # GH 35973, GH#58998
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            val = {
+                "tz": date_range("2020-08-30", freq="d", periods=2, tz="Europe/London")
+            }
+        df = DataFrame(val)
+        result = df.astype({"tz": "datetime64[ns, Europe/Berlin]"})
+
+        expected = df
+        expected["tz"] = expected["tz"].dt.tz_convert("Europe/Berlin")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["UTC", "Europe/Berlin"])
+    def test_astype_tz_object_conversion(self, tz):
+        # GH 35973
+        val = {"tz": date_range("2020-08-30", freq="D", periods=2, tz="Europe/London")}
+        expected = DataFrame(val)
+
+        # convert expected to object dtype from other tz str (independently tested)
+        result = expected.astype({"tz": f"datetime64[ns, {tz}]"})
+        result = result.astype({"tz": "object"})
+
+        # do real test: object dtype to a specified tz, different from construction tz.
+        result = result.astype({"tz": "datetime64[ns, Europe/London]"})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string) GH#60639")
+    def test_astype_dt64_to_string(
+        self, frame_or_series, tz_naive_fixture, using_infer_string
+    ):
+        # GH#41409
+        tz = tz_naive_fixture
+
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        dta = dti._data
+        dta[0] = NaT
+
+        obj = frame_or_series(dta)
+        result = obj.astype("string")
+
+        # Check that Series/DataFrame.astype matches DatetimeArray.astype
+        expected = frame_or_series(dta.astype("string"))
+        tm.assert_equal(result, expected)
+
+        item = result.iloc[0]
+        if frame_or_series is DataFrame:
+            item = item.iloc[0]
+        if using_infer_string:
+            assert item is np.nan
+        else:
+            assert item is pd.NA
+
+        # For non-NA values, we should match what we get for non-EA str
+        alt = obj.astype(str)
+        assert np.all(alt.iloc[1:] == result.iloc[1:])
+
+    def test_astype_td64_to_string(self, frame_or_series):
+        # GH#41409
+        tdi = pd.timedelta_range("1 Day", periods=3)
+        obj = frame_or_series(tdi)
+
+        expected = frame_or_series(["1 days", "2 days", "3 days"], dtype="string")
+        result = obj.astype("string")
+        tm.assert_equal(result, expected)
+
+    def test_astype_bytes(self):
+        # GH#39474
+        result = DataFrame(["foo", "bar", "baz"]).astype(bytes)
+        assert result.dtypes[0] == np.dtype("S3")
+
+    @pytest.mark.parametrize(
+        "index_slice",
+        [
+            np.s_[:2, :2],
+            np.s_[:1, :2],
+            np.s_[:2, :1],
+            np.s_[::2, ::2],
+            np.s_[::1, ::2],
+            np.s_[::2, ::1],
+        ],
+    )
+    def test_astype_noncontiguous(self, index_slice):
+        # GH#42396
+        data = np.arange(16).reshape(4, 4)
+        df = DataFrame(data)
+
+        result = df.iloc[index_slice].astype("int16")
+        expected = df.iloc[index_slice]
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_astype_retain_attrs(self, any_numpy_dtype):
+        # GH#44414
+        df = DataFrame({"a": [0, 1, 2], "b": [3, 4, 5]})
+        df.attrs["Location"] = "Michigan"
+
+        result = df.astype({"a": any_numpy_dtype}).attrs
+        expected = df.attrs
+
+        tm.assert_dict_equal(expected, result)
+
+
+class TestAstypeCategorical:
+    def test_astype_from_categorical3(self):
+        df = DataFrame({"cats": [1, 2, 3, 4, 5, 6], "vals": [1, 2, 3, 4, 5, 6]})
+        cats = Categorical([1, 2, 3, 4, 5, 6])
+        exp_df = DataFrame({"cats": cats, "vals": [1, 2, 3, 4, 5, 6]})
+        df["cats"] = df["cats"].astype("category")
+        tm.assert_frame_equal(exp_df, df)
+
+    def test_astype_from_categorical4(self):
+        df = DataFrame(
+            {"cats": ["a", "b", "b", "a", "a", "d"], "vals": [1, 2, 3, 4, 5, 6]}
+        )
+        cats = Categorical(["a", "b", "b", "a", "a", "d"])
+        exp_df = DataFrame({"cats": cats, "vals": [1, 2, 3, 4, 5, 6]})
+        df["cats"] = df["cats"].astype("category")
+        tm.assert_frame_equal(exp_df, df)
+
+    def test_categorical_astype_to_int(self, any_int_dtype):
+        # GH#39402
+
+        df = DataFrame(data={"col1": pd.array([2.0, 1.0, 3.0])})
+        df.col1 = df.col1.astype("category")
+        df.col1 = df.col1.astype(any_int_dtype)
+        expected = DataFrame({"col1": pd.array([2, 1, 3], dtype=any_int_dtype)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_astype_categorical_to_string_missing(self):
+        # https://github.com/pandas-dev/pandas/issues/41797
+        df = DataFrame(["a", "b", np.nan])
+        expected = df.astype(str)
+        cat = df.astype("category")
+        result = cat.astype(str)
+        tm.assert_frame_equal(result, expected)
+
+
+class IntegerArrayNoCopy(pd.core.arrays.IntegerArray):
+    # GH 42501
+
+    def copy(self):
+        raise NotImplementedError
+
+
+class Int16DtypeNoCopy(pd.Int16Dtype):
+    # GH 42501
+
+    @classmethod
+    def construct_array_type(cls):
+        return IntegerArrayNoCopy
+
+
+def test_frame_astype_no_copy():
+    # GH 42501
+    df = DataFrame({"a": [1, 4, None, 5], "b": [6, 7, 8, 9]}, dtype=object)
+    result = df.astype({"a": Int16DtypeNoCopy()})
+
+    assert result.a.dtype == pd.Int16Dtype()
+    assert np.shares_memory(df.b.values, result.b.values)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+def test_astype_copies(dtype):
+    # GH#50984
+    pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [1, 2, 3]}, dtype=dtype)
+    result = df.astype("int64[pyarrow]")
+    df.iloc[0, 0] = 100
+    expected = DataFrame({"a": [1, 2, 3]}, dtype="int64[pyarrow]")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [None, 1, 1.5, np.nan, NaT])
+def test_astype_to_string_not_modifying_input(string_storage, val):
+    # GH#51073
+    df = DataFrame({"a": ["a", "b", val]})
+    expected = df.copy()
+    with option_context("mode.string_storage", string_storage):
+        df.astype("string")
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("val", [None, 1, 1.5, np.nan, NaT])
+def test_astype_to_string_dtype_not_modifying_input(any_string_dtype, val):
+    # GH#51073 - variant of the above test with explicit dtype instances
+    df = DataFrame({"a": ["a", "b", val]})
+    expected = df.copy()
+    df.astype(any_string_dtype)
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_at_time.py b/pandas/pandas/tests/frame/methods/test_at_time.py
new file mode 100644
index 0000000000000..b69db80dee446
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_at_time.py
@@ -0,0 +1,134 @@
+from datetime import (
+    time,
+    timezone,
+)
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import timezones
+
+from pandas import (
+    DataFrame,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestAtTime:
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_localized_at_time(self, tzstr, frame_or_series):
+        tz = timezones.maybe_get_tz(tzstr)
+
+        rng = date_range("4/16/2012", "5/1/2012", freq="h")
+        ts = frame_or_series(
+            np.random.default_rng(2).standard_normal(len(rng)), index=rng
+        )
+
+        ts_local = ts.tz_localize(tzstr)
+
+        result = ts_local.at_time(time(10, 0))
+        expected = ts.at_time(time(10, 0)).tz_localize(tzstr)
+        tm.assert_equal(result, expected)
+        assert timezones.tz_compare(result.index.tz, tz)
+
+    def test_at_time(self, frame_or_series):
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 2)), index=rng
+        )
+        ts = tm.get_obj(ts, frame_or_series)
+        rs = ts.at_time(rng[1])
+        assert (rs.index.hour == rng[1].hour).all()
+        assert (rs.index.minute == rng[1].minute).all()
+        assert (rs.index.second == rng[1].second).all()
+
+        result = ts.at_time("9:30")
+        expected = ts.at_time(time(9, 30))
+        tm.assert_equal(result, expected)
+
+    def test_at_time_midnight(self, frame_or_series):
+        # midnight, everything
+        rng = date_range("1/1/2000", "1/31/2000")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 3)), index=rng
+        )
+        ts = tm.get_obj(ts, frame_or_series)
+
+        result = ts.at_time(time(0, 0))
+        tm.assert_equal(result, ts)
+
+    def test_at_time_nonexistent(self, frame_or_series):
+        # time doesn't exist
+        rng = date_range("1/1/2012", freq="23Min", periods=384)
+        ts = DataFrame(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        ts = tm.get_obj(ts, frame_or_series)
+        rs = ts.at_time("16:00")
+        assert len(rs) == 0
+
+    @pytest.mark.parametrize(
+        "hour", ["1:00", "1:00AM", time(1), time(1, tzinfo=timezone.utc)]
+    )
+    def test_at_time_errors(self, hour):
+        # GH#24043
+        dti = date_range("2018", periods=3, freq="h")
+        df = DataFrame(list(range(len(dti))), index=dti)
+        if getattr(hour, "tzinfo", None) is None:
+            result = df.at_time(hour)
+            expected = df.iloc[1:2]
+            tm.assert_frame_equal(result, expected)
+        else:
+            with pytest.raises(ValueError, match="Index must be timezone"):
+                df.at_time(hour)
+
+    def test_at_time_tz(self):
+        # GH#24043
+        dti = date_range("2018", periods=3, freq="h", tz="US/Pacific")
+        df = DataFrame(list(range(len(dti))), index=dti)
+        result = df.at_time(time(4, tzinfo=zoneinfo.ZoneInfo("US/Eastern")))
+        expected = df.iloc[1:2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_at_time_raises(self, frame_or_series):
+        # GH#20725
+        obj = DataFrame([[1, 2, 3], [4, 5, 6]])
+        obj = tm.get_obj(obj, frame_or_series)
+        msg = "Index must be DatetimeIndex"
+        with pytest.raises(TypeError, match=msg):  # index is not a DatetimeIndex
+            obj.at_time("00:00")
+
+    def test_at_time_axis(self, axis):
+        # issue 8839
+        rng = date_range("1/1/2000", "1/2/2000", freq="5min")
+        ts = DataFrame(np.random.default_rng(2).standard_normal((len(rng), len(rng))))
+        ts.index, ts.columns = rng, rng
+
+        indices = rng[(rng.hour == 9) & (rng.minute == 30) & (rng.second == 0)]
+
+        if axis in ["index", 0]:
+            expected = ts.loc[indices, :]
+        elif axis in ["columns", 1]:
+            expected = ts.loc[:, indices]
+
+        result = ts.at_time("9:30", axis=axis)
+
+        # Without clearing freq, result has freq 1440T and expected 5T
+        result.index = result.index._with_freq(None)
+        expected.index = expected.index._with_freq(None)
+        tm.assert_frame_equal(result, expected)
+
+    def test_at_time_datetimeindex(self):
+        index = date_range("2012-01-01", "2012-01-05", freq="30min")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 5)), index=index
+        )
+        akey = time(12, 0, 0)
+        ainds = [24, 72, 120, 168]
+
+        result = df.at_time(akey)
+        expected = df.loc[akey]
+        expected2 = df.iloc[ainds]
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result, expected2)
+        assert len(result) == 4
diff --git a/pandas/pandas/tests/frame/methods/test_between_time.py b/pandas/pandas/tests/frame/methods/test_between_time.py
new file mode 100644
index 0000000000000..74d6291707e19
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_between_time.py
@@ -0,0 +1,227 @@
+from datetime import (
+    datetime,
+    time,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import timezones
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestBetweenTime:
+    @td.skip_if_not_us_locale
+    def test_between_time_formats(self, frame_or_series):
+        # GH#11818
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 2)), index=rng
+        )
+        ts = tm.get_obj(ts, frame_or_series)
+
+        strings = [
+            ("2:00", "2:30"),
+            ("0200", "0230"),
+            ("2:00am", "2:30am"),
+            ("0200am", "0230am"),
+            ("2:00:00", "2:30:00"),
+            ("020000", "023000"),
+            ("2:00:00am", "2:30:00am"),
+            ("020000am", "023000am"),
+        ]
+        expected_length = 28
+
+        for time_string in strings:
+            assert len(ts.between_time(*time_string)) == expected_length
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_localized_between_time(self, tzstr, frame_or_series):
+        tz = timezones.maybe_get_tz(tzstr)
+
+        rng = date_range("4/16/2012", "5/1/2012", freq="h")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+        if frame_or_series is DataFrame:
+            ts = ts.to_frame()
+
+        ts_local = ts.tz_localize(tzstr)
+
+        t1, t2 = time(10, 0), time(11, 0)
+        result = ts_local.between_time(t1, t2)
+        expected = ts.between_time(t1, t2).tz_localize(tzstr)
+        tm.assert_equal(result, expected)
+        assert timezones.tz_compare(result.index.tz, tz)
+
+    def test_between_time_types(self, frame_or_series):
+        # GH11818
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        obj = DataFrame({"A": 0}, index=rng)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = r"Cannot convert arg \[datetime\.datetime\(2010, 1, 2, 1, 0\)\] to a time"
+        with pytest.raises(ValueError, match=msg):
+            obj.between_time(datetime(2010, 1, 2, 1), datetime(2010, 1, 2, 5))
+
+    def test_between_time(self, inclusive_endpoints_fixture, frame_or_series):
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 2)), index=rng
+        )
+        ts = tm.get_obj(ts, frame_or_series)
+
+        stime = time(0, 0)
+        etime = time(1, 0)
+        inclusive = inclusive_endpoints_fixture
+
+        filtered = ts.between_time(stime, etime, inclusive=inclusive)
+        exp_len = 13 * 4 + 1
+
+        if inclusive in ["right", "neither"]:
+            exp_len -= 5
+        if inclusive in ["left", "neither"]:
+            exp_len -= 4
+
+        assert len(filtered) == exp_len
+        for rs in filtered.index:
+            t = rs.time()
+            if inclusive in ["left", "both"]:
+                assert t >= stime
+            else:
+                assert t > stime
+
+            if inclusive in ["right", "both"]:
+                assert t <= etime
+            else:
+                assert t < etime
+
+        result = ts.between_time("00:00", "01:00")
+        expected = ts.between_time(stime, etime)
+        tm.assert_equal(result, expected)
+
+        # across midnight
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 2)), index=rng
+        )
+        ts = tm.get_obj(ts, frame_or_series)
+        stime = time(22, 0)
+        etime = time(9, 0)
+
+        filtered = ts.between_time(stime, etime, inclusive=inclusive)
+        exp_len = (12 * 11 + 1) * 4 + 1
+        if inclusive in ["right", "neither"]:
+            exp_len -= 4
+        if inclusive in ["left", "neither"]:
+            exp_len -= 4
+
+        assert len(filtered) == exp_len
+        for rs in filtered.index:
+            t = rs.time()
+            if inclusive in ["left", "both"]:
+                assert (t >= stime) or (t <= etime)
+            else:
+                assert (t > stime) or (t <= etime)
+
+            if inclusive in ["right", "both"]:
+                assert (t <= etime) or (t >= stime)
+            else:
+                assert (t < etime) or (t >= stime)
+
+    def test_between_time_raises(self, frame_or_series):
+        # GH#20725
+        obj = DataFrame([[1, 2, 3], [4, 5, 6]])
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = "Index must be DatetimeIndex"
+        with pytest.raises(TypeError, match=msg):  # index is not a DatetimeIndex
+            obj.between_time(start_time="00:00", end_time="12:00")
+
+    def test_between_time_axis(self, frame_or_series):
+        # GH#8839
+        rng = date_range("1/1/2000", periods=100, freq="10min")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+        if frame_or_series is DataFrame:
+            ts = ts.to_frame()
+
+        stime, etime = ("08:00:00", "09:00:00")
+        expected_length = 7
+
+        assert len(ts.between_time(stime, etime)) == expected_length
+        assert len(ts.between_time(stime, etime, axis=0)) == expected_length
+        msg = f"No axis named {ts.ndim} for object type {type(ts).__name__}"
+        with pytest.raises(ValueError, match=msg):
+            ts.between_time(stime, etime, axis=ts.ndim)
+
+    def test_between_time_axis_aliases(self, axis):
+        # GH#8839
+        rng = date_range("1/1/2000", periods=100, freq="10min")
+        ts = DataFrame(np.random.default_rng(2).standard_normal((len(rng), len(rng))))
+        stime, etime = ("08:00:00", "09:00:00")
+        exp_len = 7
+
+        if axis in ["index", 0]:
+            ts.index = rng
+            assert len(ts.between_time(stime, etime)) == exp_len
+            assert len(ts.between_time(stime, etime, axis=0)) == exp_len
+
+        if axis in ["columns", 1]:
+            ts.columns = rng
+            selected = ts.between_time(stime, etime, axis=1).columns
+            assert len(selected) == exp_len
+
+    def test_between_time_axis_raises(self, axis):
+        # issue 8839
+        rng = date_range("1/1/2000", periods=100, freq="10min")
+        mask = np.arange(0, len(rng))
+        rand_data = np.random.default_rng(2).standard_normal((len(rng), len(rng)))
+        ts = DataFrame(rand_data, index=rng, columns=rng)
+        stime, etime = ("08:00:00", "09:00:00")
+
+        msg = "Index must be DatetimeIndex"
+        if axis in ["columns", 1]:
+            ts.index = mask
+            with pytest.raises(TypeError, match=msg):
+                ts.between_time(stime, etime)
+            with pytest.raises(TypeError, match=msg):
+                ts.between_time(stime, etime, axis=0)
+
+        if axis in ["index", 0]:
+            ts.columns = mask
+            with pytest.raises(TypeError, match=msg):
+                ts.between_time(stime, etime, axis=1)
+
+    def test_between_time_datetimeindex(self):
+        index = date_range("2012-01-01", "2012-01-05", freq="30min")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 5)), index=index
+        )
+        bkey = slice(time(13, 0, 0), time(14, 0, 0))
+        binds = [26, 27, 28, 74, 75, 76, 122, 123, 124, 170, 171, 172]
+
+        result = df.between_time(bkey.start, bkey.stop)
+        expected = df.loc[bkey]
+        expected2 = df.iloc[binds]
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result, expected2)
+        assert len(result) == 12
+
+    def test_between_time_incorrect_arg_inclusive(self):
+        # GH40245
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 2)), index=rng
+        )
+
+        stime = time(0, 0)
+        etime = time(1, 0)
+        inclusive = "bad_string"
+        msg = "Inclusive has to be either 'both', 'neither', 'left' or 'right'"
+        with pytest.raises(ValueError, match=msg):
+            ts.between_time(stime, etime, inclusive=inclusive)
diff --git a/pandas/pandas/tests/frame/methods/test_clip.py b/pandas/pandas/tests/frame/methods/test_clip.py
new file mode 100644
index 0000000000000..62a97271d208b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_clip.py
@@ -0,0 +1,200 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameClip:
+    def test_clip(self, float_frame):
+        median = float_frame.median().median()
+        original = float_frame.copy()
+
+        double = float_frame.clip(upper=median, lower=median)
+        assert not (double.values != median).any()
+
+        # Verify that float_frame was not changed inplace
+        assert (float_frame.values == original.values).all()
+
+    def test_inplace_clip(self, float_frame):
+        # GH#15388
+        median = float_frame.median().median()
+        frame_copy = float_frame.copy()
+
+        return_value = frame_copy.clip(upper=median, lower=median, inplace=True)
+        assert return_value is None
+        assert not (frame_copy.values != median).any()
+
+    def test_dataframe_clip(self):
+        # GH#2747
+        df = DataFrame(np.random.default_rng(2).standard_normal((1000, 2)))
+
+        for lb, ub in [(-1, 1), (1, -1)]:
+            clipped_df = df.clip(lb, ub)
+
+            lb, ub = min(lb, ub), max(ub, lb)
+            lb_mask = df.values <= lb
+            ub_mask = df.values >= ub
+            mask = ~lb_mask & ~ub_mask
+            assert (clipped_df.values[lb_mask] == lb).all()
+            assert (clipped_df.values[ub_mask] == ub).all()
+            assert (clipped_df.values[mask] == df.values[mask]).all()
+
+    def test_clip_mixed_numeric(self):
+        # clip on mixed integer or floats
+        # GH#24162, clipping now preserves numeric types per column
+        df = DataFrame({"A": [1, 2, 3], "B": [1.0, np.nan, 3.0]})
+        result = df.clip(1, 2)
+        expected = DataFrame({"A": [1, 2, 2], "B": [1.0, np.nan, 2.0]})
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame([[1, 2, 3.4], [3, 4, 5.6]], columns=["foo", "bar", "baz"])
+        expected = df.dtypes
+        result = df.clip(upper=3).dtypes
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    def test_clip_against_series(self, inplace):
+        # GH#6966
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((1000, 2)))
+        lb = Series(np.random.default_rng(2).standard_normal(1000))
+        ub = lb + 1
+
+        original = df.copy()
+        clipped_df = df.clip(lb, ub, axis=0, inplace=inplace)
+
+        if inplace:
+            clipped_df = df
+
+        for i in range(2):
+            lb_mask = original.iloc[:, i] <= lb
+            ub_mask = original.iloc[:, i] >= ub
+            mask = ~lb_mask & ~ub_mask
+
+            result = clipped_df.loc[lb_mask, i]
+            tm.assert_series_equal(result, lb[lb_mask], check_names=False)
+            assert result.name == i
+
+            result = clipped_df.loc[ub_mask, i]
+            tm.assert_series_equal(result, ub[ub_mask], check_names=False)
+            assert result.name == i
+
+            tm.assert_series_equal(clipped_df.loc[mask, i], df.loc[mask, i])
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize("lower", [[2, 3, 4], np.asarray([2, 3, 4])])
+    @pytest.mark.parametrize(
+        "axis,res",
+        [
+            (0, [[2.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 7.0, 7.0]]),
+            (1, [[2.0, 3.0, 4.0], [4.0, 5.0, 6.0], [5.0, 6.0, 7.0]]),
+        ],
+    )
+    def test_clip_against_list_like(self, inplace, lower, axis, res):
+        # GH#15390
+        arr = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0]])
+
+        original = DataFrame(
+            arr, columns=["one", "two", "three"], index=["a", "b", "c"]
+        )
+
+        result = original.clip(lower=lower, upper=[5, 6, 7], axis=axis, inplace=inplace)
+
+        expected = DataFrame(res, columns=original.columns, index=original.index)
+        if inplace:
+            result = original
+        tm.assert_frame_equal(result, expected, check_exact=True)
+
+    @pytest.mark.parametrize("axis", [0, 1, None])
+    def test_clip_against_frame(self, axis):
+        df = DataFrame(np.random.default_rng(2).standard_normal((1000, 2)))
+        lb = DataFrame(np.random.default_rng(2).standard_normal((1000, 2)))
+        ub = lb + 1
+
+        clipped_df = df.clip(lb, ub, axis=axis)
+
+        lb_mask = df <= lb
+        ub_mask = df >= ub
+        mask = ~lb_mask & ~ub_mask
+
+        tm.assert_frame_equal(clipped_df[lb_mask], lb[lb_mask])
+        tm.assert_frame_equal(clipped_df[ub_mask], ub[ub_mask])
+        tm.assert_frame_equal(clipped_df[mask], df[mask])
+
+    def test_clip_against_unordered_columns(self):
+        # GH#20911
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 4)),
+            columns=["A", "B", "C", "D"],
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 4)),
+            columns=["D", "A", "B", "C"],
+        )
+        df3 = DataFrame(df2.values - 1, columns=["B", "D", "C", "A"])
+        result_upper = df1.clip(lower=0, upper=df2)
+        expected_upper = df1.clip(lower=0, upper=df2[df1.columns])
+        result_lower = df1.clip(lower=df3, upper=3)
+        expected_lower = df1.clip(lower=df3[df1.columns], upper=3)
+        result_lower_upper = df1.clip(lower=df3, upper=df2)
+        expected_lower_upper = df1.clip(lower=df3[df1.columns], upper=df2[df1.columns])
+        tm.assert_frame_equal(result_upper, expected_upper)
+        tm.assert_frame_equal(result_lower, expected_lower)
+        tm.assert_frame_equal(result_lower_upper, expected_lower_upper)
+
+    def test_clip_with_na_args(self, float_frame):
+        """Should process np.nan argument as None"""
+        # GH#17276
+        tm.assert_frame_equal(float_frame.clip(np.nan), float_frame)
+        tm.assert_frame_equal(float_frame.clip(upper=np.nan, lower=np.nan), float_frame)
+
+        # GH#19992 and adjusted in GH#40420
+        df = DataFrame({"col_0": [1, 2, 3], "col_1": [4, 5, 6], "col_2": [7, 8, 9]})
+
+        result = df.clip(lower=[4, 5, np.nan], axis=0)
+        expected = DataFrame(
+            {
+                "col_0": Series([4, 5, 3], dtype="float"),
+                "col_1": [4, 5, 6],
+                "col_2": [7, 8, 9],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.clip(lower=[4, 5, np.nan], axis=1)
+        expected = DataFrame(
+            {"col_0": [4, 4, 4], "col_1": [5, 5, 6], "col_2": [7, 8, 9]}
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # GH#40420
+        data = {"col_0": [9, -3, 0, -1, 5], "col_1": [-2, -7, 6, 8, -5]}
+        df = DataFrame(data)
+        t = Series([2, -4, np.nan, 6, 3])
+        result = df.clip(lower=t, axis=0)
+        expected = DataFrame(
+            {"col_0": [9, -3, 0, 6, 5], "col_1": [2, -4, 6, 8, 3]}, dtype="float"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_clip_int_data_with_float_bound(self):
+        # GH51472
+        df = DataFrame({"a": [1, 2, 3]})
+        result = df.clip(lower=1.5)
+        expected = DataFrame({"a": [1.5, 2.0, 3.0]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_clip_with_list_bound(self):
+        # GH#54817
+        df = DataFrame([1, 5])
+        expected = DataFrame([3, 5])
+        result = df.clip([3])
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([1, 3])
+        result = df.clip(upper=[3])
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_combine.py b/pandas/pandas/tests/frame/methods/test_combine.py
new file mode 100644
index 0000000000000..bc6a67e4e1f32
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_combine.py
@@ -0,0 +1,47 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestCombine:
+    @pytest.mark.parametrize(
+        "data",
+        [
+            pd.date_range("2000", periods=4),
+            pd.date_range("2000", periods=4, tz="US/Central"),
+            pd.period_range("2000", periods=4),
+            pd.timedelta_range(0, periods=4),
+        ],
+    )
+    def test_combine_datetlike_udf(self, data):
+        # GH#23079
+        df = pd.DataFrame({"A": data})
+        other = df.copy()
+        df.iloc[1, 0] = None
+
+        def combiner(a, b):
+            return b
+
+        result = df.combine(other, combiner)
+        tm.assert_frame_equal(result, other)
+
+    def test_combine_generic(self, float_frame):
+        df1 = float_frame
+        df2 = float_frame.loc[float_frame.index[:-5], ["A", "B", "C"]]
+
+        combined = df1.combine(df2, np.add)
+        combined2 = df2.combine(df1, np.add)
+        assert combined["D"].isna().all()
+        assert combined2["D"].isna().all()
+
+        chunk = combined.loc[combined.index[:-5], ["A", "B", "C"]]
+        chunk2 = combined2.loc[combined2.index[:-5], ["A", "B", "C"]]
+
+        exp = (
+            float_frame.loc[float_frame.index[:-5], ["A", "B", "C"]].reindex_like(chunk)
+            * 2
+        )
+        tm.assert_frame_equal(chunk, exp)
+        tm.assert_frame_equal(chunk2, exp)
diff --git a/pandas/pandas/tests/frame/methods/test_combine_first.py b/pandas/pandas/tests/frame/methods/test_combine_first.py
new file mode 100644
index 0000000000000..1e594043510ea
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_combine_first.py
@@ -0,0 +1,567 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import find_common_type
+from pandas.core.dtypes.common import is_dtype_equal
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameCombineFirst:
+    def test_combine_first_mixed(self):
+        a = Series(["a", "b"], index=range(2))
+        b = Series(range(2), index=range(2))
+        f = DataFrame({"A": a, "B": b})
+
+        a = Series(["a", "b"], index=range(5, 7))
+        b = Series(range(2), index=range(5, 7))
+        g = DataFrame({"A": a, "B": b})
+
+        exp = DataFrame({"A": list("abab"), "B": [0, 1, 0, 1]}, index=[0, 1, 5, 6])
+        combined = f.combine_first(g)
+        tm.assert_frame_equal(combined, exp)
+
+    def test_combine_first(self, float_frame):
+        # disjoint
+        head, tail = float_frame[:5], float_frame[5:]
+
+        combined = head.combine_first(tail)
+        reordered_frame = float_frame.reindex(combined.index)
+        tm.assert_frame_equal(combined, reordered_frame)
+        tm.assert_index_equal(combined.columns, float_frame.columns)
+        tm.assert_series_equal(combined["A"], reordered_frame["A"])
+
+        # same index
+        fcopy = float_frame.copy()
+        fcopy["A"] = 1
+        del fcopy["C"]
+
+        fcopy2 = float_frame.copy()
+        fcopy2["B"] = 0
+        del fcopy2["D"]
+
+        combined = fcopy.combine_first(fcopy2)
+
+        assert (combined["A"] == 1).all()
+        tm.assert_series_equal(combined["B"], fcopy["B"])
+        tm.assert_series_equal(combined["C"], fcopy2["C"])
+        tm.assert_series_equal(combined["D"], fcopy["D"])
+
+        # overlap
+        head, tail = reordered_frame[:10].copy(), reordered_frame
+        head["A"] = 1
+
+        combined = head.combine_first(tail)
+        assert (combined["A"][:10] == 1).all()
+
+        # reverse overlap
+        tail.iloc[:10, tail.columns.get_loc("A")] = 0
+        combined = tail.combine_first(head)
+        assert (combined["A"][:10] == 0).all()
+
+        # no overlap
+        f = float_frame[:10]
+        g = float_frame[10:]
+        combined = f.combine_first(g)
+        tm.assert_series_equal(combined["A"].reindex(f.index), f["A"])
+        tm.assert_series_equal(combined["A"].reindex(g.index), g["A"])
+
+        # corner cases
+        comb = float_frame.combine_first(DataFrame())
+        tm.assert_frame_equal(comb, float_frame)
+
+        comb = DataFrame().combine_first(float_frame)
+        tm.assert_frame_equal(comb, float_frame.sort_index())
+
+        comb = float_frame.combine_first(DataFrame(index=["faz", "boo"]))
+        assert "faz" in comb.index
+
+        # #2525
+        df = DataFrame({"a": [1]}, index=[datetime(2012, 1, 1)])
+        df2 = DataFrame(columns=["b"])
+        result = df.combine_first(df2)
+        assert "b" in result
+
+    def test_combine_first_mixed_bug(self):
+        idx = Index(["a", "b", "c", "e"])
+        ser1 = Series([5.0, -9.0, 4.0, 100.0], index=idx)
+        ser2 = Series(["a", "b", "c", "e"], index=idx)
+        ser3 = Series([12, 4, 5, 97], index=idx)
+
+        frame1 = DataFrame({"col0": ser1, "col2": ser2, "col3": ser3})
+
+        idx = Index(["a", "b", "c", "f"])
+        ser1 = Series([5.0, -9.0, 4.0, 100.0], index=idx)
+        ser2 = Series(["a", "b", "c", "f"], index=idx)
+        ser3 = Series([12, 4, 5, 97], index=idx)
+
+        frame2 = DataFrame({"col1": ser1, "col2": ser2, "col5": ser3})
+
+        combined = frame1.combine_first(frame2)
+        assert len(combined.columns) == 5
+
+    def test_combine_first_same_as_in_update(self):
+        # gh 3016 (same as in update)
+        df = DataFrame(
+            [[1.0, 2.0, False, True], [4.0, 5.0, True, False]],
+            columns=["A", "B", "bool1", "bool2"],
+        )
+
+        other = DataFrame([[45, 45]], index=[0], columns=["A", "B"])
+        result = df.combine_first(other)
+        tm.assert_frame_equal(result, df)
+
+        df.loc[0, "A"] = np.nan
+        result = df.combine_first(other)
+        df.loc[0, "A"] = 45
+        tm.assert_frame_equal(result, df)
+
+    def test_combine_first_doc_example(self):
+        # doc example
+        df1 = DataFrame(
+            {"A": [1.0, np.nan, 3.0, 5.0, np.nan], "B": [np.nan, 2.0, 3.0, np.nan, 6.0]}
+        )
+
+        df2 = DataFrame(
+            {
+                "A": [5.0, 2.0, 4.0, np.nan, 3.0, 7.0],
+                "B": [np.nan, np.nan, 3.0, 4.0, 6.0, 8.0],
+            }
+        )
+
+        result = df1.combine_first(df2)
+        expected = DataFrame({"A": [1, 2, 3, 5, 3, 7.0], "B": [np.nan, 2, 3, 4, 6, 8]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_combine_first_return_obj_type_with_bools(self):
+        # GH3552
+
+        df1 = DataFrame(
+            [[np.nan, 3.0, True], [-4.6, np.nan, True], [np.nan, 7.0, False]]
+        )
+        df2 = DataFrame([[-42.6, np.nan, True], [-5.0, 1.6, False]], index=[1, 2])
+
+        expected = Series([True, True, False], name=2, dtype=bool)
+
+        result_12 = df1.combine_first(df2)[2]
+        tm.assert_series_equal(result_12, expected)
+
+        result_21 = df2.combine_first(df1)[2]
+        tm.assert_series_equal(result_21, expected)
+
+    @pytest.mark.parametrize(
+        "data1, data2, data_expected",
+        (
+            (
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                [pd.NaT, pd.NaT, pd.NaT],
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            ),
+            (
+                [pd.NaT, pd.NaT, pd.NaT],
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            ),
+            (
+                [datetime(2000, 1, 2), pd.NaT, pd.NaT],
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                [datetime(2000, 1, 2), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            ),
+            (
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                [datetime(2000, 1, 2), pd.NaT, pd.NaT],
+                [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            ),
+        ),
+    )
+    def test_combine_first_convert_datatime_correctly(
+        self, data1, data2, data_expected
+    ):
+        # GH 3593
+
+        df1, df2 = DataFrame({"a": data1}), DataFrame({"a": data2})
+        result = df1.combine_first(df2)
+        expected = DataFrame({"a": data_expected})
+        tm.assert_frame_equal(result, expected)
+
+    def test_combine_first_align_nan(self):
+        # GH 7509 (not fixed)
+        dfa = DataFrame([[pd.Timestamp("2011-01-01"), 2]], columns=["a", "b"])
+        dfb = DataFrame([[4], [5]], columns=["b"])
+        assert dfa["a"].dtype == "datetime64[s]"
+        assert dfa["b"].dtype == "int64"
+
+        res = dfa.combine_first(dfb)
+        exp = DataFrame(
+            {"a": [pd.Timestamp("2011-01-01"), pd.NaT], "b": [2, 5]},
+            columns=["a", "b"],
+        )
+        tm.assert_frame_equal(res, exp)
+        assert res["a"].dtype == "datetime64[s]"
+        # TODO: this must be int64
+        assert res["b"].dtype == "int64"
+
+        res = dfa.iloc[:0].combine_first(dfb)
+        exp = DataFrame({"a": [np.nan, np.nan], "b": [4, 5]}, columns=["a", "b"])
+        tm.assert_frame_equal(res, exp)
+        # TODO: this must be datetime64
+        assert res["a"].dtype == "float64"
+        # TODO: this must be int64
+        assert res["b"].dtype == "int64"
+
+    def test_combine_first_timezone(self, unit):
+        # see gh-7630
+        data1 = pd.to_datetime("20100101 01:01").tz_localize("UTC").as_unit(unit)
+        df1 = DataFrame(
+            columns=["UTCdatetime", "abc"],
+            data=data1,
+            index=pd.date_range("20140627", periods=1, unit=unit),
+        )
+        data2 = pd.to_datetime("20121212 12:12").tz_localize("UTC").as_unit(unit)
+        df2 = DataFrame(
+            columns=["UTCdatetime", "xyz"],
+            data=data2,
+            index=pd.date_range("20140628", periods=1, unit=unit),
+        )
+        res = df2[["UTCdatetime"]].combine_first(df1)
+        exp = DataFrame(
+            {
+                "UTCdatetime": [
+                    pd.Timestamp("2010-01-01 01:01", tz="UTC"),
+                    pd.Timestamp("2012-12-12 12:12", tz="UTC"),
+                ],
+                "abc": [pd.Timestamp("2010-01-01 01:01:00", tz="UTC"), pd.NaT],
+            },
+            columns=["UTCdatetime", "abc"],
+            index=pd.date_range("20140627", periods=2, freq="D", unit=unit),
+            dtype=f"datetime64[{unit}, UTC]",
+        )
+        assert res["UTCdatetime"].dtype == f"datetime64[{unit}, UTC]"
+        assert res["abc"].dtype == f"datetime64[{unit}, UTC]"
+
+        tm.assert_frame_equal(res, exp)
+
+    def test_combine_first_timezone2(self, unit):
+        # see gh-10567
+        dts1 = pd.date_range("2015-01-01", "2015-01-05", tz="UTC", unit=unit)
+        df1 = DataFrame({"DATE": dts1})
+        dts2 = pd.date_range("2015-01-03", "2015-01-05", tz="UTC", unit=unit)
+        df2 = DataFrame({"DATE": dts2})
+
+        res = df1.combine_first(df2)
+        tm.assert_frame_equal(res, df1)
+        assert res["DATE"].dtype == f"datetime64[{unit}, UTC]"
+
+    def test_combine_first_timezone3(self, unit):
+        dts1 = pd.DatetimeIndex(
+            ["2011-01-01", "NaT", "2011-01-03", "2011-01-04"], tz="US/Eastern"
+        ).as_unit(unit)
+        df1 = DataFrame({"DATE": dts1}, index=[1, 3, 5, 7])
+        dts2 = pd.DatetimeIndex(
+            ["2012-01-01", "2012-01-02", "2012-01-03"], tz="US/Eastern"
+        ).as_unit(unit)
+        df2 = DataFrame({"DATE": dts2}, index=[2, 4, 5])
+
+        res = df1.combine_first(df2)
+        exp_dts = pd.DatetimeIndex(
+            [
+                "2011-01-01",
+                "2012-01-01",
+                "NaT",
+                "2012-01-02",
+                "2011-01-03",
+                "2011-01-04",
+            ],
+            tz="US/Eastern",
+        ).as_unit(unit)
+        exp = DataFrame({"DATE": exp_dts}, index=[1, 2, 3, 4, 5, 7])
+        tm.assert_frame_equal(res, exp)
+
+    def test_combine_first_timezone4(self, unit):
+        # different tz
+        dts1 = pd.date_range("2015-01-01", "2015-01-05", tz="US/Eastern", unit=unit)
+        df1 = DataFrame({"DATE": dts1})
+        dts2 = pd.date_range("2015-01-03", "2015-01-05", unit=unit)
+        df2 = DataFrame({"DATE": dts2})
+
+        # if df1 doesn't have NaN, keep its dtype
+        res = df1.combine_first(df2)
+        tm.assert_frame_equal(res, df1)
+        assert res["DATE"].dtype == f"datetime64[{unit}, US/Eastern]"
+
+    def test_combine_first_timezone5(self, unit):
+        dts1 = pd.date_range("2015-01-01", "2015-01-02", tz="US/Eastern", unit=unit)
+        df1 = DataFrame({"DATE": dts1})
+        dts2 = pd.date_range("2015-01-01", "2015-01-03", unit=unit)
+        df2 = DataFrame({"DATE": dts2})
+
+        res = df1.combine_first(df2)
+        exp_dts = [
+            pd.Timestamp("2015-01-01", tz="US/Eastern"),
+            pd.Timestamp("2015-01-02", tz="US/Eastern"),
+            pd.Timestamp("2015-01-03"),
+        ]
+        exp = DataFrame({"DATE": exp_dts})
+        tm.assert_frame_equal(res, exp)
+        assert res["DATE"].dtype == "object"
+
+    def test_combine_first_timedelta(self):
+        data1 = pd.TimedeltaIndex(["1 day", "NaT", "3 day", "4day"])
+        df1 = DataFrame({"TD": data1}, index=[1, 3, 5, 7])
+        data2 = pd.TimedeltaIndex(["10 day", "11 day", "12 day"])
+        df2 = DataFrame({"TD": data2}, index=[2, 4, 5])
+
+        res = df1.combine_first(df2)
+        exp_dts = pd.TimedeltaIndex(
+            ["1 day", "10 day", "NaT", "11 day", "3 day", "4 day"]
+        )
+        exp = DataFrame({"TD": exp_dts}, index=[1, 2, 3, 4, 5, 7])
+        tm.assert_frame_equal(res, exp)
+        assert res["TD"].dtype == "timedelta64[ns]"
+
+    def test_combine_first_period(self):
+        data1 = pd.PeriodIndex(["2011-01", "NaT", "2011-03", "2011-04"], freq="M")
+        df1 = DataFrame({"P": data1}, index=[1, 3, 5, 7])
+        data2 = pd.PeriodIndex(["2012-01-01", "2012-02", "2012-03"], freq="M")
+        df2 = DataFrame({"P": data2}, index=[2, 4, 5])
+
+        res = df1.combine_first(df2)
+        exp_dts = pd.PeriodIndex(
+            ["2011-01", "2012-01", "NaT", "2012-02", "2011-03", "2011-04"], freq="M"
+        )
+        exp = DataFrame({"P": exp_dts}, index=[1, 2, 3, 4, 5, 7])
+        tm.assert_frame_equal(res, exp)
+        assert res["P"].dtype == data1.dtype
+
+        # different freq
+        dts2 = pd.PeriodIndex(["2012-01-01", "2012-01-02", "2012-01-03"], freq="D")
+        df2 = DataFrame({"P": dts2}, index=[2, 4, 5])
+
+        res = df1.combine_first(df2)
+        exp_dts = [
+            pd.Period("2011-01", freq="M"),
+            pd.Period("2012-01-01", freq="D"),
+            pd.NaT,
+            pd.Period("2012-01-02", freq="D"),
+            pd.Period("2011-03", freq="M"),
+            pd.Period("2011-04", freq="M"),
+        ]
+        exp = DataFrame({"P": exp_dts}, index=[1, 2, 3, 4, 5, 7])
+        tm.assert_frame_equal(res, exp)
+        assert res["P"].dtype == "object"
+
+    def test_combine_first_int(self):
+        # GH14687 - integer series that do no align exactly
+
+        df1 = DataFrame({"a": [0, 1, 3, 5]}, dtype="int64")
+        df2 = DataFrame({"a": [1, 4]}, dtype="int64")
+
+        result_12 = df1.combine_first(df2)
+        expected_12 = DataFrame({"a": [0, 1, 3, 5]})
+        tm.assert_frame_equal(result_12, expected_12)
+
+        result_21 = df2.combine_first(df1)
+        expected_21 = DataFrame({"a": [1, 4, 3, 5]})
+        tm.assert_frame_equal(result_21, expected_21)
+
+    @pytest.mark.parametrize("val", [1, 1.0])
+    def test_combine_first_with_asymmetric_other(self, val):
+        # see gh-20699
+        df1 = DataFrame({"isNum": [val]})
+        df2 = DataFrame({"isBool": [True]})
+
+        res = df1.combine_first(df2)
+        exp = DataFrame({"isNum": [val], "isBool": [True]})
+
+        tm.assert_frame_equal(res, exp)
+
+    def test_combine_first_string_dtype_only_na(self, nullable_string_dtype):
+        # GH: 37519
+        df = DataFrame(
+            {"a": ["962", "85"], "b": [pd.NA] * 2}, dtype=nullable_string_dtype
+        )
+        df2 = DataFrame({"a": ["85"], "b": [pd.NA]}, dtype=nullable_string_dtype)
+        df.set_index(["a", "b"], inplace=True)
+        df2.set_index(["a", "b"], inplace=True)
+        result = df.combine_first(df2)
+        expected = DataFrame(
+            {"a": ["962", "85"], "b": [pd.NA] * 2}, dtype=nullable_string_dtype
+        ).set_index(["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "scalar1, scalar2",
+    [
+        (datetime(2020, 1, 1), datetime(2020, 1, 2)),
+        (pd.Period("2020-01-01", "D"), pd.Period("2020-01-02", "D")),
+        (pd.Timedelta("89 days"), pd.Timedelta("60 min")),
+        (pd.Interval(left=0, right=1), pd.Interval(left=2, right=3, closed="left")),
+    ],
+)
+def test_combine_first_timestamp_bug(scalar1, scalar2, nulls_fixture):
+    # GH28481
+    na_value = nulls_fixture
+
+    frame = DataFrame([[na_value, na_value]], columns=["a", "b"])
+    other = DataFrame([[scalar1, scalar2]], columns=["b", "c"])
+
+    common_dtype = find_common_type([frame.dtypes["b"], other.dtypes["b"]])
+
+    if (
+        is_dtype_equal(common_dtype, "object")
+        or frame.dtypes["b"] == other.dtypes["b"]
+        or frame.dtypes["b"].kind == frame.dtypes["b"].kind == "M"
+    ):
+        val = scalar1
+    else:
+        val = na_value
+
+    result = frame.combine_first(other)
+
+    expected = DataFrame([[na_value, val, scalar2]], columns=["a", "b", "c"])
+
+    expected["b"] = expected["b"].astype(common_dtype)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_combine_first_timestamp_bug_NaT():
+    # GH28481
+    frame = DataFrame([[pd.NaT, pd.NaT]], columns=["a", "b"])
+    other = DataFrame(
+        [[datetime(2020, 1, 1), datetime(2020, 1, 2)]], columns=["b", "c"]
+    )
+
+    result = frame.combine_first(other)
+    expected = DataFrame(
+        [[pd.NaT, datetime(2020, 1, 1), datetime(2020, 1, 2)]], columns=["a", "b", "c"]
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_combine_first_with_nan_multiindex():
+    # gh-36562
+
+    mi1 = MultiIndex.from_arrays(
+        [["b", "b", "c", "a", "b", np.nan], [1, 2, 3, 4, 5, 6]], names=["a", "b"]
+    )
+    df = DataFrame({"c": [1, 1, 1, 1, 1, 1]}, index=mi1)
+    mi2 = MultiIndex.from_arrays(
+        [["a", "b", "c", "a", "b", "d"], [1, 1, 1, 1, 1, 1]], names=["a", "b"]
+    )
+    s = Series([1, 2, 3, 4, 5, 6], index=mi2)
+    res = df.combine_first(DataFrame({"d": s}))
+    mi_expected = MultiIndex.from_arrays(
+        [
+            ["a", "a", "a", "b", "b", "b", "b", "c", "c", "d", np.nan],
+            [1, 1, 4, 1, 1, 2, 5, 1, 3, 1, 6],
+        ],
+        names=["a", "b"],
+    )
+    expected = DataFrame(
+        {
+            "c": [np.nan, np.nan, 1, 1, 1, 1, 1, np.nan, 1, np.nan, 1],
+            "d": [1.0, 4.0, np.nan, 2.0, 5.0, np.nan, np.nan, 3.0, np.nan, 6.0, np.nan],
+        },
+        index=mi_expected,
+    )
+    tm.assert_frame_equal(res, expected)
+
+
+def test_combine_preserve_dtypes():
+    # GH7509
+    a_column = Series(["a", "b"], index=range(2))
+    b_column = Series(range(2), index=range(2))
+    df1 = DataFrame({"A": a_column, "B": b_column})
+
+    c_column = Series(["a", "b"], index=range(5, 7))
+    b_column = Series(range(-1, 1), index=range(5, 7))
+    df2 = DataFrame({"B": b_column, "C": c_column})
+
+    expected = DataFrame(
+        {
+            "A": ["a", "b", np.nan, np.nan],
+            "B": [0, 1, -1, 0],
+            "C": [np.nan, np.nan, "a", "b"],
+        },
+        index=[0, 1, 5, 6],
+    )
+    combined = df1.combine_first(df2)
+    tm.assert_frame_equal(combined, expected)
+
+
+def test_combine_first_duplicates_rows_for_nan_index_values():
+    # GH39881
+    df1 = DataFrame(
+        {"x": [9, 10, 11]},
+        index=MultiIndex.from_arrays([[1, 2, 3], [np.nan, 5, 6]], names=["a", "b"]),
+    )
+
+    df2 = DataFrame(
+        {"y": [12, 13, 14]},
+        index=MultiIndex.from_arrays([[1, 2, 4], [np.nan, 5, 7]], names=["a", "b"]),
+    )
+
+    expected = DataFrame(
+        {
+            "x": [9.0, 10.0, 11.0, np.nan],
+            "y": [12.0, 13.0, np.nan, 14.0],
+        },
+        index=MultiIndex.from_arrays(
+            [[1, 2, 3, 4], [np.nan, 5, 6, 7]], names=["a", "b"]
+        ),
+    )
+    combined = df1.combine_first(df2)
+    tm.assert_frame_equal(combined, expected)
+
+
+def test_combine_first_int64_not_cast_to_float64():
+    # GH 28613
+    df_1 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+    df_2 = DataFrame({"A": [1, 20, 30], "B": [40, 50, 60], "C": [12, 34, 65]})
+    result = df_1.combine_first(df_2)
+    expected = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [12, 34, 65]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_midx_losing_dtype():
+    # GH#49830
+    midx = MultiIndex.from_arrays([[0, 0], [np.nan, np.nan]])
+    midx2 = MultiIndex.from_arrays([[1, 1], [np.nan, np.nan]])
+    df1 = DataFrame({"a": [None, 4]}, index=midx)
+    df2 = DataFrame({"a": [3, 3]}, index=midx2)
+    result = df1.combine_first(df2)
+    expected_midx = MultiIndex.from_arrays(
+        [[0, 0, 1, 1], [np.nan, np.nan, np.nan, np.nan]]
+    )
+    expected = DataFrame({"a": [np.nan, 4, 3, 3]}, index=expected_midx)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_combine_first_empty_columns():
+    left = DataFrame(columns=["a", "b"])
+    right = DataFrame(columns=["a", "c"])
+    result = left.combine_first(right)
+    expected = DataFrame(columns=["a", "b", "c"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_combine_first_preserve_column_order():
+    # GH#60427
+    df1 = DataFrame({"B": [1, 2, 3], "A": [4, None, 6]})
+    df2 = DataFrame({"A": [5]}, index=[1])
+
+    result = df1.combine_first(df2)
+    expected = DataFrame({"B": [1, 2, 3], "A": [4.0, 5.0, 6.0]})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_compare.py b/pandas/pandas/tests/frame/methods/test_compare.py
new file mode 100644
index 0000000000000..2ffc3f933e246
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_compare.py
@@ -0,0 +1,306 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gte1p25
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("align_axis", [0, 1, "index", "columns"])
+def test_compare_axis(align_axis):
+    # GH#30429
+    df = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+        columns=["col1", "col2", "col3"],
+    )
+    df2 = df.copy()
+    df2.loc[0, "col1"] = "c"
+    df2.loc[2, "col3"] = 4.0
+
+    result = df.compare(df2, align_axis=align_axis)
+
+    if align_axis in (1, "columns"):
+        indices = pd.RangeIndex(0, 4, 2)
+        columns = pd.MultiIndex.from_product([["col1", "col3"], ["self", "other"]])
+        expected = pd.DataFrame(
+            [["a", "c", np.nan, np.nan], [np.nan, np.nan, 3.0, 4.0]],
+            index=indices,
+            columns=columns,
+        )
+    else:
+        indices = pd.MultiIndex.from_product([range(0, 4, 2), ["self", "other"]])
+        columns = pd.Index(["col1", "col3"])
+        expected = pd.DataFrame(
+            [["a", np.nan], ["c", np.nan], [np.nan, 3.0], [np.nan, 4.0]],
+            index=indices,
+            columns=columns,
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keep_shape, keep_equal",
+    [
+        (True, False),
+        (False, True),
+        (True, True),
+        # False, False case is already covered in test_compare_axis
+    ],
+)
+def test_compare_various_formats(keep_shape, keep_equal):
+    df = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+        columns=["col1", "col2", "col3"],
+    )
+    df2 = df.copy()
+    df2.loc[0, "col1"] = "c"
+    df2.loc[2, "col3"] = 4.0
+
+    result = df.compare(df2, keep_shape=keep_shape, keep_equal=keep_equal)
+
+    if keep_shape:
+        indices = pd.RangeIndex(3)
+        columns = pd.MultiIndex.from_product(
+            [["col1", "col2", "col3"], ["self", "other"]]
+        )
+        if keep_equal:
+            expected = pd.DataFrame(
+                [
+                    ["a", "c", 1.0, 1.0, 1.0, 1.0],
+                    ["b", "b", 2.0, 2.0, 2.0, 2.0],
+                    ["c", "c", np.nan, np.nan, 3.0, 4.0],
+                ],
+                index=indices,
+                columns=columns,
+            )
+        else:
+            expected = pd.DataFrame(
+                [
+                    ["a", "c", np.nan, np.nan, np.nan, np.nan],
+                    [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+                    [np.nan, np.nan, np.nan, np.nan, 3.0, 4.0],
+                ],
+                index=indices,
+                columns=columns,
+            )
+    else:
+        indices = pd.RangeIndex(0, 4, 2)
+        columns = pd.MultiIndex.from_product([["col1", "col3"], ["self", "other"]])
+        expected = pd.DataFrame(
+            [["a", "c", 1.0, 1.0], ["c", "c", 3.0, 4.0]], index=indices, columns=columns
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compare_with_equal_nulls():
+    # We want to make sure two NaNs are considered the same
+    # and dropped where applicable
+    df = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+        columns=["col1", "col2", "col3"],
+    )
+    df2 = df.copy()
+    df2.loc[0, "col1"] = "c"
+
+    result = df.compare(df2)
+    indices = pd.Index([0])
+    columns = pd.MultiIndex.from_product([["col1"], ["self", "other"]])
+    expected = pd.DataFrame([["a", "c"]], index=indices, columns=columns)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compare_with_non_equal_nulls():
+    # We want to make sure the relevant NaNs do not get dropped
+    # even if the entire row or column are NaNs
+    df = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+        columns=["col1", "col2", "col3"],
+    )
+    df2 = df.copy()
+    df2.loc[0, "col1"] = "c"
+    df2.loc[2, "col3"] = np.nan
+
+    result = df.compare(df2)
+
+    indices = pd.Index([0, 2])
+    columns = pd.MultiIndex.from_product([["col1", "col3"], ["self", "other"]])
+    expected = pd.DataFrame(
+        [["a", "c", np.nan, np.nan], [np.nan, np.nan, 3.0, np.nan]],
+        index=indices,
+        columns=columns,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("align_axis", [0, 1])
+def test_compare_multi_index(align_axis):
+    df = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]}
+    )
+    df.columns = pd.MultiIndex.from_arrays([["a", "a", "b"], ["col1", "col2", "col3"]])
+    df.index = pd.MultiIndex.from_arrays([["x", "x", "y"], [0, 1, 2]])
+
+    df2 = df.copy()
+    df2.iloc[0, 0] = "c"
+    df2.iloc[2, 2] = 4.0
+
+    result = df.compare(df2, align_axis=align_axis)
+
+    if align_axis == 0:
+        indices = pd.MultiIndex.from_arrays(
+            [["x", "x", "y", "y"], [0, 0, 2, 2], ["self", "other", "self", "other"]]
+        )
+        columns = pd.MultiIndex.from_arrays([["a", "b"], ["col1", "col3"]])
+        data = [["a", np.nan], ["c", np.nan], [np.nan, 3.0], [np.nan, 4.0]]
+    else:
+        indices = pd.MultiIndex.from_arrays([["x", "y"], [0, 2]])
+        columns = pd.MultiIndex.from_arrays(
+            [
+                ["a", "a", "b", "b"],
+                ["col1", "col1", "col3", "col3"],
+                ["self", "other", "self", "other"],
+            ]
+        )
+        data = [["a", "c", np.nan, np.nan], [np.nan, np.nan, 3.0, 4.0]]
+
+    expected = pd.DataFrame(data=data, index=indices, columns=columns)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compare_different_indices():
+    msg = (
+        r"Can only compare identically-labeled \(both index and columns\) DataFrame "
+        "objects"
+    )
+    df1 = pd.DataFrame([1, 2, 3], index=["a", "b", "c"])
+    df2 = pd.DataFrame([1, 2, 3], index=["a", "b", "d"])
+    with pytest.raises(ValueError, match=msg):
+        df1.compare(df2)
+
+
+def test_compare_different_shapes():
+    msg = (
+        r"Can only compare identically-labeled \(both index and columns\) DataFrame "
+        "objects"
+    )
+    df1 = pd.DataFrame(np.ones((3, 3)))
+    df2 = pd.DataFrame(np.zeros((2, 1)))
+    with pytest.raises(ValueError, match=msg):
+        df1.compare(df2)
+
+
+def test_compare_result_names():
+    # GH 44354
+    df1 = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+    )
+    df2 = pd.DataFrame(
+        {
+            "col1": ["c", "b", "c"],
+            "col2": [1.0, 2.0, np.nan],
+            "col3": [1.0, 2.0, np.nan],
+        },
+    )
+    result = df1.compare(df2, result_names=("left", "right"))
+    result.index = pd.Index([0, 2])
+    expected = pd.DataFrame(
+        {
+            ("col1", "left"): {0: "a", 2: np.nan},
+            ("col1", "right"): {0: "c", 2: np.nan},
+            ("col3", "left"): {0: np.nan, 2: 3.0},
+            ("col3", "right"): {0: np.nan, 2: np.nan},
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "result_names",
+    [
+        [1, 2],
+        "HK",
+        {"2": 2, "3": 3},
+        3,
+        3.0,
+    ],
+)
+def test_invalid_input_result_names(result_names):
+    # GH 44354
+    df1 = pd.DataFrame(
+        {"col1": ["a", "b", "c"], "col2": [1.0, 2.0, np.nan], "col3": [1.0, 2.0, 3.0]},
+    )
+    df2 = pd.DataFrame(
+        {
+            "col1": ["c", "b", "c"],
+            "col2": [1.0, 2.0, np.nan],
+            "col3": [1.0, 2.0, np.nan],
+        },
+    )
+    with pytest.raises(
+        TypeError,
+        match=(
+            f"Passing 'result_names' as a {type(result_names)} is not "
+            "supported. Provide 'result_names' as a tuple instead."
+        ),
+    ):
+        df1.compare(df2, result_names=result_names)
+
+
+@pytest.mark.parametrize(
+    "val1,val2",
+    [(4, pd.NA), (pd.NA, pd.NA), (pd.NA, 4)],
+)
+def test_compare_ea_and_np_dtype(val1, val2):
+    # GH 48966
+    arr = [4.0, val1]
+    ser = pd.Series([1, val2], dtype="Int64")
+
+    df1 = pd.DataFrame({"a": arr, "b": [1.0, 2]})
+    df2 = pd.DataFrame({"a": ser, "b": [1.0, 2]})
+    expected = pd.DataFrame(
+        {
+            ("a", "self"): arr,
+            ("a", "other"): ser,
+            ("b", "self"): np.nan,
+            ("b", "other"): np.nan,
+        }
+    )
+    if val1 is pd.NA and val2 is pd.NA:
+        # GH#18463 TODO: is this really the desired behavior?
+        expected.loc[1, ("a", "self")] = np.nan
+
+    if val1 is pd.NA and np_version_gte1p25:
+        # can't compare with numpy array if it contains pd.NA
+        with pytest.raises(TypeError, match="boolean value of NA is ambiguous"):
+            result = df1.compare(df2, keep_shape=True)
+    else:
+        result = df1.compare(df2, keep_shape=True)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df1_val,df2_val,diff_self,diff_other",
+    [
+        (4, 3, 4, 3),
+        (4, 4, pd.NA, pd.NA),
+        (4, pd.NA, 4, pd.NA),
+        (pd.NA, pd.NA, pd.NA, pd.NA),
+    ],
+)
+def test_compare_nullable_int64_dtype(df1_val, df2_val, diff_self, diff_other):
+    # GH 48966
+    df1 = pd.DataFrame({"a": pd.Series([df1_val, pd.NA], dtype="Int64"), "b": [1.0, 2]})
+    df2 = df1.copy()
+    df2.loc[0, "a"] = df2_val
+
+    expected = pd.DataFrame(
+        {
+            ("a", "self"): pd.Series([diff_self, pd.NA], dtype="Int64"),
+            ("a", "other"): pd.Series([diff_other, pd.NA], dtype="Int64"),
+            ("b", "self"): np.nan,
+            ("b", "other"): np.nan,
+        }
+    )
+    result = df1.compare(df2, keep_shape=True)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_convert_dtypes.py b/pandas/pandas/tests/frame/methods/test_convert_dtypes.py
new file mode 100644
index 0000000000000..d0f30204758d3
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_convert_dtypes.py
@@ -0,0 +1,213 @@
+import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestConvertDtypes:
+    @pytest.mark.parametrize(
+        "convert_integer, expected", [(False, np.dtype("int32")), (True, "Int32")]
+    )
+    def test_convert_dtypes(self, convert_integer, expected, string_storage):
+        # Specific types are tested in tests/series/test_dtypes.py
+        # Just check that it works for DataFrame here
+        df = pd.DataFrame(
+            {
+                "a": pd.Series([1, 2, 3], dtype=np.dtype("int32")),
+                "b": pd.Series(["x", "y", "z"], dtype=np.dtype("O")),
+            }
+        )
+        with pd.option_context("string_storage", string_storage):
+            result = df.convert_dtypes(True, True, convert_integer, False)
+        expected = pd.DataFrame(
+            {
+                "a": pd.Series([1, 2, 3], dtype=expected),
+                "b": pd.Series(["x", "y", "z"], dtype=f"string[{string_storage}]"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_convert_empty(self):
+        # Empty DataFrame can pass convert_dtypes, see GH#40393
+        empty_df = pd.DataFrame()
+        tm.assert_frame_equal(empty_df, empty_df.convert_dtypes())
+
+    def test_convert_dtypes_retain_column_names(self):
+        # GH#41435
+        df = pd.DataFrame({"a": [1, 2], "b": [3, 4]})
+        df.columns.name = "cols"
+
+        result = df.convert_dtypes()
+        tm.assert_index_equal(result.columns, df.columns)
+        assert result.columns.name == "cols"
+
+    def test_pyarrow_dtype_backend(self):
+        pa = pytest.importorskip("pyarrow")
+        df = pd.DataFrame(
+            {
+                "a": pd.Series([1, 2, 3], dtype=np.dtype("int32")),
+                "b": pd.Series(["x", "y", None], dtype=np.dtype("O")),
+                "c": pd.Series([True, False, None], dtype=np.dtype("O")),
+                "d": pd.Series([np.nan, 100.5, 200], dtype=np.dtype("float")),
+                "e": pd.Series(pd.date_range("2022", periods=3)),
+                "f": pd.Series(pd.date_range("2022", periods=3, tz="UTC").as_unit("s")),
+                "g": pd.Series(pd.timedelta_range("1D", periods=3)),
+            }
+        )
+        result = df.convert_dtypes(dtype_backend="pyarrow")
+        expected = pd.DataFrame(
+            {
+                "a": pd.arrays.ArrowExtensionArray(
+                    pa.array([1, 2, 3], type=pa.int32())
+                ),
+                "b": pd.arrays.ArrowExtensionArray(pa.array(["x", "y", None])),
+                "c": pd.arrays.ArrowExtensionArray(pa.array([True, False, None])),
+                "d": pd.arrays.ArrowExtensionArray(pa.array([None, 100.5, 200.0])),
+                "e": pd.arrays.ArrowExtensionArray(
+                    pa.array(
+                        [
+                            datetime.datetime(2022, 1, 1),
+                            datetime.datetime(2022, 1, 2),
+                            datetime.datetime(2022, 1, 3),
+                        ],
+                        type=pa.timestamp(unit="ns"),
+                    )
+                ),
+                "f": pd.arrays.ArrowExtensionArray(
+                    pa.array(
+                        [
+                            datetime.datetime(2022, 1, 1),
+                            datetime.datetime(2022, 1, 2),
+                            datetime.datetime(2022, 1, 3),
+                        ],
+                        type=pa.timestamp(unit="s", tz="UTC"),
+                    )
+                ),
+                "g": pd.arrays.ArrowExtensionArray(
+                    pa.array(
+                        [
+                            datetime.timedelta(1),
+                            datetime.timedelta(2),
+                            datetime.timedelta(3),
+                        ],
+                        type=pa.duration("ns"),
+                    )
+                ),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pyarrow_dtype_backend_already_pyarrow(self):
+        pytest.importorskip("pyarrow")
+        expected = pd.DataFrame([1, 2, 3], dtype="int64[pyarrow]")
+        result = expected.convert_dtypes(dtype_backend="pyarrow")
+        tm.assert_frame_equal(result, expected)
+
+    def test_pyarrow_dtype_backend_from_pandas_nullable(self):
+        pa = pytest.importorskip("pyarrow")
+        df = pd.DataFrame(
+            {
+                "a": pd.Series([1, 2, None], dtype="Int32"),
+                "b": pd.Series(["x", "y", None], dtype="string[python]"),
+                "c": pd.Series([True, False, None], dtype="boolean"),
+                "d": pd.Series([None, 100.5, 200], dtype="Float64"),
+            }
+        )
+        result = df.convert_dtypes(dtype_backend="pyarrow")
+        expected = pd.DataFrame(
+            {
+                "a": pd.arrays.ArrowExtensionArray(
+                    pa.array([1, 2, None], type=pa.int32())
+                ),
+                "b": pd.arrays.ArrowExtensionArray(pa.array(["x", "y", None])),
+                "c": pd.arrays.ArrowExtensionArray(pa.array([True, False, None])),
+                "d": pd.arrays.ArrowExtensionArray(pa.array([None, 100.5, 200.0])),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pyarrow_dtype_empty_object(self):
+        # GH 50970
+        pytest.importorskip("pyarrow")
+        expected = pd.DataFrame(columns=[0])
+        result = expected.convert_dtypes(dtype_backend="pyarrow")
+        tm.assert_frame_equal(result, expected)
+
+    def test_pyarrow_engine_lines_false(self):
+        # GH 48893
+        df = pd.DataFrame({"a": [1, 2, 3]})
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.convert_dtypes(dtype_backend="numpy")
+
+    def test_pyarrow_backend_no_conversion(self):
+        # GH#52872
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame({"a": [1, 2], "b": 1.5, "c": True, "d": "x"})
+        expected = df.copy()
+        result = df.convert_dtypes(
+            convert_floating=False,
+            convert_integer=False,
+            convert_boolean=False,
+            convert_string=False,
+            dtype_backend="pyarrow",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_convert_dtypes_pyarrow_to_np_nullable(self):
+        # GH 53648
+        pytest.importorskip("pyarrow")
+        ser = pd.DataFrame(range(2), dtype="int32[pyarrow]")
+        result = ser.convert_dtypes(dtype_backend="numpy_nullable")
+        expected = pd.DataFrame(range(2), dtype="Int32")
+        tm.assert_frame_equal(result, expected)
+
+    def test_convert_dtypes_pyarrow_timestamp(self):
+        # GH 54191
+        pytest.importorskip("pyarrow")
+        ser = pd.Series(pd.date_range("2020-01-01", "2020-01-02", freq="1min"))
+        expected = ser.astype("timestamp[ms][pyarrow]")
+        result = expected.convert_dtypes(dtype_backend="pyarrow")
+        tm.assert_series_equal(result, expected)
+
+    def test_convert_dtypes_avoid_block_splitting(self):
+        # GH#55341
+        df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": "a"})
+        result = df.convert_dtypes(convert_integer=False)
+        expected = pd.DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": [4, 5, 6],
+                "c": pd.Series(["a"] * 3, dtype="string[python]"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+        assert result._mgr.nblocks == 2
+
+    def test_convert_dtypes_from_arrow(self):
+        # GH#56581
+        df = pd.DataFrame([["a", datetime.time(18, 12)]], columns=["a", "b"])
+        result = df.convert_dtypes()
+        expected = df.astype({"a": "string[python]"})
+        tm.assert_frame_equal(result, expected)
+
+    def test_convert_dtype_pyarrow_timezone_preserve(self):
+        # GH 60237
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame(
+            {
+                "timestamps": pd.Series(
+                    pd.to_datetime(range(5), utc=True, unit="h"),
+                    dtype="timestamp[ns, tz=UTC][pyarrow]",
+                )
+            }
+        )
+        result = df.convert_dtypes(dtype_backend="pyarrow")
+        expected = df.copy()
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_copy.py b/pandas/pandas/tests/frame/methods/test_copy.py
new file mode 100644
index 0000000000000..5b72a84320c52
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_copy.py
@@ -0,0 +1,41 @@
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+
+
+class TestCopy:
+    @pytest.mark.parametrize("attr", ["index", "columns"])
+    def test_copy_index_name_checking(self, float_frame, attr):
+        # don't want to be able to modify the index stored elsewhere after
+        # making a copy
+        ind = getattr(float_frame, attr)
+        ind.name = None
+        cp = float_frame.copy()
+        getattr(cp, attr).name = "foo"
+        assert getattr(float_frame, attr).name is None
+
+    def test_copy(self, float_frame, float_string_frame):
+        cop = float_frame.copy()
+        cop["E"] = cop["A"]
+        assert "E" not in float_frame
+
+        # copy objects
+        copy = float_string_frame.copy()
+        assert copy._mgr is not float_string_frame._mgr
+
+    def test_copy_consolidates(self):
+        # GH#42477
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).integers(0, 100, size=55),
+                "b": np.random.default_rng(2).integers(0, 100, size=55),
+            }
+        )
+
+        for i in range(10):
+            df.loc[:, f"n_{i}"] = np.random.default_rng(2).integers(0, 100, size=55)
+
+        assert len(df._mgr.blocks) == 11
+        result = df.copy()
+        assert len(result._mgr.blocks) == 1
diff --git a/pandas/pandas/tests/frame/methods/test_count.py b/pandas/pandas/tests/frame/methods/test_count.py
new file mode 100644
index 0000000000000..1553a8a86305d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_count.py
@@ -0,0 +1,39 @@
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameCount:
+    def test_count(self):
+        # corner case
+        frame = DataFrame()
+        ct1 = frame.count(1)
+        assert isinstance(ct1, Series)
+
+        ct2 = frame.count(0)
+        assert isinstance(ct2, Series)
+
+        # GH#423
+        df = DataFrame(index=range(10))
+        result = df.count(1)
+        expected = Series(0, index=df.index)
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame(columns=range(10))
+        result = df.count(0)
+        expected = Series(0, index=df.columns)
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame()
+        result = df.count()
+        expected = Series(dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_count_objects(self, float_string_frame):
+        dm = DataFrame(float_string_frame._series)
+        df = DataFrame(float_string_frame._series)
+
+        tm.assert_series_equal(dm.count(), df.count())
+        tm.assert_series_equal(dm.count(1), df.count(1))
diff --git a/pandas/pandas/tests/frame/methods/test_cov_corr.py b/pandas/pandas/tests/frame/methods/test_cov_corr.py
new file mode 100644
index 0000000000000..d5e94382b8314
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_cov_corr.py
@@ -0,0 +1,487 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameCov:
+    def test_cov(self, float_frame, float_string_frame):
+        # min_periods no NAs (corner case)
+        expected = float_frame.cov()
+        result = float_frame.cov(min_periods=len(float_frame))
+
+        tm.assert_frame_equal(expected, result)
+
+        result = float_frame.cov(min_periods=len(float_frame) + 1)
+        assert isna(result.values).all()
+
+        # with NAs
+        frame = float_frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+        frame.iloc[5:10, frame.columns.get_loc("B")] = np.nan
+        result = frame.cov(min_periods=len(frame) - 8)
+        expected = frame.cov()
+        expected.loc["A", "B"] = np.nan
+        expected.loc["B", "A"] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        # regular
+        result = frame.cov()
+        expected = frame["A"].cov(frame["C"])
+        tm.assert_almost_equal(result["A"]["C"], expected)
+
+        # fails on non-numeric types
+        with pytest.raises(ValueError, match="could not convert string to float"):
+            float_string_frame.cov()
+        result = float_string_frame.cov(numeric_only=True)
+        expected = float_string_frame.loc[:, ["A", "B", "C", "D"]].cov()
+        tm.assert_frame_equal(result, expected)
+
+        # Single column frame
+        df = DataFrame(np.linspace(0.0, 1.0, 10))
+        result = df.cov()
+        expected = DataFrame(
+            np.cov(df.values.T).reshape((1, 1)), index=df.columns, columns=df.columns
+        )
+        tm.assert_frame_equal(result, expected)
+        df.loc[0] = np.nan
+        result = df.cov()
+        expected = DataFrame(
+            np.cov(df.values[1:].T).reshape((1, 1)),
+            index=df.columns,
+            columns=df.columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("test_ddof", [None, 0, 1, 2, 3])
+    def test_cov_ddof(self, test_ddof):
+        # GH#34611
+        np_array1 = np.random.default_rng(2).random(10)
+        np_array2 = np.random.default_rng(2).random(10)
+        df = DataFrame({0: np_array1, 1: np_array2})
+        result = df.cov(ddof=test_ddof)
+        expected_np = np.cov(np_array1, np_array2, ddof=test_ddof)
+        expected = DataFrame(expected_np)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "other_column", [pd.array([1, 2, 3]), np.array([1.0, 2.0, 3.0])]
+    )
+    def test_cov_nullable_integer(self, other_column):
+        # https://github.com/pandas-dev/pandas/issues/33803
+        data = DataFrame({"a": pd.array([1, 2, None]), "b": other_column})
+        result = data.cov()
+        arr = np.array([[0.5, 0.5], [0.5, 1.0]])
+        expected = DataFrame(arr, columns=["a", "b"], index=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("numeric_only", [True, False])
+    def test_cov_numeric_only(self, numeric_only):
+        # when dtypes of pandas series are different
+        # then ndarray will have dtype=object,
+        # so it need to be properly handled
+        df = DataFrame({"a": [1, 0], "c": ["x", "y"]})
+        expected = DataFrame(0.5, index=["a"], columns=["a"])
+        if numeric_only:
+            result = df.cov(numeric_only=numeric_only)
+            tm.assert_frame_equal(result, expected)
+        else:
+            with pytest.raises(ValueError, match="could not convert string to float"):
+                df.cov(numeric_only=numeric_only)
+
+
+class TestDataFrameCorr:
+    # DataFrame.corr(), as opposed to DataFrame.corrwith
+
+    @pytest.mark.parametrize("method", ["pearson", "kendall", "spearman"])
+    def test_corr_scipy_method(self, float_frame, method):
+        pytest.importorskip("scipy")
+        float_frame.loc[float_frame.index[:5], "A"] = np.nan
+        float_frame.loc[float_frame.index[5:10], "B"] = np.nan
+        float_frame.loc[float_frame.index[:10], "A"] = float_frame["A"][10:20].copy()
+
+        correls = float_frame.corr(method=method)
+        expected = float_frame["A"].corr(float_frame["C"], method=method)
+        tm.assert_almost_equal(correls["A"]["C"], expected)
+
+    # ---------------------------------------------------------------------
+
+    def test_corr_non_numeric(self, float_string_frame):
+        with pytest.raises(ValueError, match="could not convert string to float"):
+            float_string_frame.corr()
+        result = float_string_frame.corr(numeric_only=True)
+        expected = float_string_frame.loc[:, ["A", "B", "C", "D"]].corr()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("meth", ["pearson", "kendall", "spearman"])
+    def test_corr_nooverlap(self, meth):
+        # nothing in common
+        pytest.importorskip("scipy")
+        df = DataFrame(
+            {
+                "A": [1, 1.5, 1, np.nan, np.nan, np.nan],
+                "B": [np.nan, np.nan, np.nan, 1, 1.5, 1],
+                "C": [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+            }
+        )
+        rs = df.corr(meth)
+        assert isna(rs.loc["A", "B"])
+        assert isna(rs.loc["B", "A"])
+        assert rs.loc["A", "A"] == 1
+        assert rs.loc["B", "B"] == 1
+        assert isna(rs.loc["C", "C"])
+
+    @pytest.mark.parametrize("meth", ["pearson", "spearman"])
+    def test_corr_constant(self, meth):
+        # constant --> all NA
+        df = DataFrame(
+            {
+                "A": [1, 1, 1, np.nan, np.nan, np.nan],
+                "B": [np.nan, np.nan, np.nan, 1, 1, 1],
+            }
+        )
+        rs = df.corr(meth)
+        assert isna(rs.values).all()
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize("meth", ["pearson", "kendall", "spearman"])
+    def test_corr_int_and_boolean(self, meth):
+        # when dtypes of pandas series are different
+        # then ndarray will have dtype=object,
+        # so it need to be properly handled
+        pytest.importorskip("scipy")
+        df = DataFrame({"a": [True, False], "b": [1, 0]})
+
+        expected = DataFrame(np.ones((2, 2)), index=["a", "b"], columns=["a", "b"])
+        result = df.corr(meth)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["cov", "corr"])
+    def test_corr_cov_independent_index_column(self, method):
+        # GH#14617
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(4 * 10).reshape(10, 4),
+            columns=list("abcd"),
+        )
+        result = getattr(df, method)()
+        assert result.index is not result.columns
+        assert result.index.equals(result.columns)
+
+    def test_corr_invalid_method(self):
+        # GH#22298
+        df = DataFrame(np.random.default_rng(2).normal(size=(10, 2)))
+        msg = "method must be either 'pearson', 'spearman', 'kendall', or a callable, "
+        with pytest.raises(ValueError, match=msg):
+            df.corr(method="____")
+
+    def test_corr_int(self):
+        # dtypes other than float64 GH#1761
+        df = DataFrame({"a": [1, 2, 3, 4], "b": [1, 2, 3, 4]})
+
+        df.cov()
+        df.corr()
+
+    @pytest.mark.parametrize(
+        "nullable_column", [pd.array([1, 2, 3]), pd.array([1, 2, None])]
+    )
+    @pytest.mark.parametrize(
+        "other_column",
+        [pd.array([1, 2, 3]), np.array([1.0, 2.0, 3.0]), np.array([1.0, 2.0, np.nan])],
+    )
+    @pytest.mark.parametrize("method", ["pearson", "spearman", "kendall"])
+    def test_corr_nullable_integer(self, nullable_column, other_column, method):
+        # https://github.com/pandas-dev/pandas/issues/33803
+        pytest.importorskip("scipy")
+        data = DataFrame({"a": nullable_column, "b": other_column})
+        result = data.corr(method=method)
+        expected = DataFrame(np.ones((2, 2)), columns=["a", "b"], index=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_corr_item_cache(self):
+        # Check that corr does not lead to incorrect entries in item_cache
+
+        df = DataFrame({"A": range(10)})
+        df["B"] = range(10)[::-1]
+
+        ser = df["A"]  # populate item_cache
+        assert len(df._mgr.blocks) == 2
+
+        _ = df.corr(numeric_only=True)
+
+        ser.iloc[0] = 99
+        assert df.loc[0, "A"] == 0
+
+    @pytest.mark.parametrize("length", [2, 20, 200, 2000])
+    def test_corr_for_constant_columns(self, length):
+        # GH: 37448
+        df = DataFrame(length * [[0.4, 0.1]], columns=["A", "B"])
+        result = df.corr()
+        expected = DataFrame(
+            {"A": [np.nan, np.nan], "B": [np.nan, np.nan]}, index=["A", "B"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_calc_corr_small_numbers(self):
+        # GH: 37452
+        df = DataFrame(
+            {"A": [1.0e-20, 2.0e-20, 3.0e-20], "B": [1.0e-20, 2.0e-20, 3.0e-20]}
+        )
+        result = df.corr()
+        expected = DataFrame({"A": [1.0, 1.0], "B": [1.0, 1.0]}, index=["A", "B"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["pearson", "spearman", "kendall"])
+    def test_corr_min_periods_greater_than_length(self, method):
+        pytest.importorskip("scipy")
+        df = DataFrame({"A": [1, 2], "B": [1, 2]})
+        result = df.corr(method=method, min_periods=3)
+        expected = DataFrame(
+            {"A": [np.nan, np.nan], "B": [np.nan, np.nan]}, index=["A", "B"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("meth", ["pearson", "kendall", "spearman"])
+    @pytest.mark.parametrize("numeric_only", [True, False])
+    def test_corr_numeric_only(self, meth, numeric_only):
+        # when dtypes of pandas series are different
+        # then ndarray will have dtype=object,
+        # so it need to be properly handled
+        pytest.importorskip("scipy")
+        df = DataFrame({"a": [1, 0], "b": [1, 0], "c": ["x", "y"]})
+        expected = DataFrame(np.ones((2, 2)), index=["a", "b"], columns=["a", "b"])
+        if numeric_only:
+            result = df.corr(meth, numeric_only=numeric_only)
+            tm.assert_frame_equal(result, expected)
+        else:
+            with pytest.raises(ValueError, match="could not convert string to float"):
+                df.corr(meth, numeric_only=numeric_only)
+
+
+class TestDataFrameCorrWith:
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "float64",
+            "Float64",
+            pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        ],
+    )
+    def test_corrwith(self, datetime_frame, dtype):
+        datetime_frame = datetime_frame.astype(dtype)
+
+        a = datetime_frame
+        noise = Series(np.random.default_rng(2).standard_normal(len(a)), index=a.index)
+
+        b = datetime_frame.add(noise, axis=0)
+
+        # make sure order does not matter
+        b = b.reindex(columns=b.columns[::-1], index=b.index[::-1][len(a) // 2 :])
+        del b["B"]
+
+        colcorr = a.corrwith(b, axis=0)
+        tm.assert_almost_equal(colcorr["A"], a["A"].corr(b["A"]))
+
+        rowcorr = a.corrwith(b, axis=1)
+        tm.assert_series_equal(rowcorr, a.T.corrwith(b.T, axis=0))
+
+        dropped = a.corrwith(b, axis=0, drop=True)
+        tm.assert_almost_equal(dropped["A"], a["A"].corr(b["A"]))
+        assert "B" not in dropped
+
+        dropped = a.corrwith(b, axis=1, drop=True)
+        assert a.index[-1] not in dropped.index
+
+    def test_corrwith_non_timeseries_data(self):
+        index = ["a", "b", "c", "d", "e"]
+        columns = ["one", "two", "three", "four"]
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            index=index,
+            columns=columns,
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=index[:4],
+            columns=columns,
+        )
+        correls = df1.corrwith(df2, axis=1)
+        for row in index[:4]:
+            tm.assert_almost_equal(correls[row], df1.loc[row].corr(df2.loc[row]))
+
+    def test_corrwith_with_objects(self, using_infer_string):
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df2 = df1.copy()
+        cols = ["A", "B", "C", "D"]
+
+        df1["obj"] = "foo"
+        df2["obj"] = "bar"
+
+        if using_infer_string:
+            msg = "Cannot perform reduction 'mean' with string dtype"
+            with pytest.raises(TypeError, match=msg):
+                df1.corrwith(df2)
+        else:
+            with pytest.raises(TypeError, match="Could not convert"):
+                df1.corrwith(df2)
+        result = df1.corrwith(df2, numeric_only=True)
+        expected = df1.loc[:, cols].corrwith(df2.loc[:, cols])
+        tm.assert_series_equal(result, expected)
+
+        with pytest.raises(TypeError, match="unsupported operand type"):
+            df1.corrwith(df2, axis=1)
+        result = df1.corrwith(df2, axis=1, numeric_only=True)
+        expected = df1.loc[:, cols].corrwith(df2.loc[:, cols], axis=1)
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_series(self, datetime_frame):
+        result = datetime_frame.corrwith(datetime_frame["A"])
+        expected = datetime_frame.apply(datetime_frame["A"].corr)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_matches_corrcoef(self):
+        df1 = DataFrame(np.arange(100), columns=["a"])
+        df2 = DataFrame(np.arange(100) ** 2, columns=["a"])
+        c1 = df1.corrwith(df2)["a"]
+        c2 = np.corrcoef(df1["a"], df2["a"])[0][1]
+
+        tm.assert_almost_equal(c1, c2)
+        assert c1 < 1
+
+    @pytest.mark.parametrize("numeric_only", [True, False])
+    def test_corrwith_mixed_dtypes(self, numeric_only):
+        # GH#18570
+        df = DataFrame(
+            {"a": [1, 4, 3, 2], "b": [4, 6, 7, 3], "c": ["a", "b", "c", "d"]}
+        )
+        s = Series([0, 6, 7, 3])
+        if numeric_only:
+            result = df.corrwith(s, numeric_only=numeric_only)
+            corrs = [df["a"].corr(s), df["b"].corr(s)]
+            expected = Series(data=corrs, index=["a", "b"])
+            tm.assert_series_equal(result, expected)
+        else:
+            with pytest.raises(
+                ValueError,
+                match="could not convert string to float",
+            ):
+                df.corrwith(s, numeric_only=numeric_only)
+
+    def test_corrwith_index_intersection(self):
+        df1 = DataFrame(
+            np.random.default_rng(2).random(size=(10, 2)), columns=["a", "b"]
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).random(size=(10, 3)), columns=["a", "b", "c"]
+        )
+
+        result = df1.corrwith(df2, drop=True).index.sort_values()
+        expected = df1.columns.intersection(df2.columns).sort_values()
+        tm.assert_index_equal(result, expected)
+
+    def test_corrwith_index_union(self):
+        df1 = DataFrame(
+            np.random.default_rng(2).random(size=(10, 2)), columns=["a", "b"]
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).random(size=(10, 3)), columns=["a", "b", "c"]
+        )
+
+        result = df1.corrwith(df2, drop=False).index.sort_values()
+        expected = df1.columns.union(df2.columns).sort_values()
+        tm.assert_index_equal(result, expected)
+
+    def test_corrwith_dup_cols(self):
+        # GH#21925
+        df1 = DataFrame(np.vstack([np.arange(10)] * 3).T)
+        df2 = df1.copy()
+        df2 = pd.concat((df2, df2[0]), axis=1)
+
+        result = df1.corrwith(df2)
+        expected = Series(np.ones(4), index=[0, 0, 1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_corr_numerical_instabilities(self):
+        # GH#45640
+        df = DataFrame([[0.2, 0.4], [0.4, 0.2]])
+        result = df.corr()
+        expected = DataFrame({0: [1.0, -1.0], 1: [-1.0, 1.0]})
+        tm.assert_frame_equal(result - 1, expected - 1, atol=1e-17)
+
+    def test_corrwith_spearman(self):
+        # GH#21925
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).random(size=(100, 3)))
+        result = df.corrwith(df**2, method="spearman")
+        expected = Series(np.ones(len(result)))
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_kendall(self):
+        # GH#21925
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).random(size=(100, 3)))
+        result = df.corrwith(df**2, method="kendall")
+        expected = Series(np.ones(len(result)))
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_spearman_with_tied_data(self):
+        # GH#48826
+        pytest.importorskip("scipy")
+        df1 = DataFrame(
+            {
+                "A": [1, np.nan, 7, 8],
+                "B": [False, True, True, False],
+                "C": [10, 4, 9, 3],
+            }
+        )
+        df2 = df1[["B", "C"]]
+        result = (df1 + 1).corrwith(df2.B, method="spearman")
+        expected = Series([0.0, 1.0, 0.0], index=["A", "B", "C"])
+        tm.assert_series_equal(result, expected)
+
+        df_bool = DataFrame(
+            {"A": [True, True, False, False], "B": [True, False, False, True]}
+        )
+        ser_bool = Series([True, True, False, True])
+        result = df_bool.corrwith(ser_bool)
+        expected = Series([0.57735, 0.57735], index=["A", "B"])
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_min_periods_method(self):
+        # GH#9490
+        pytest.importorskip("scipy")
+        df1 = DataFrame(
+            {
+                "A": [1, np.nan, 7, 8],
+                "B": [False, True, True, False],
+                "C": [10, 4, 9, 3],
+            }
+        )
+        df2 = df1[["B", "C"]]
+        result = (df1 + 1).corrwith(df2.B, method="spearman", min_periods=2)
+        expected = Series([0.0, 1.0, 0.0], index=["A", "B", "C"])
+        tm.assert_series_equal(result, expected)
+
+    def test_corrwith_min_periods_boolean(self):
+        # GH#9490
+        df_bool = DataFrame(
+            {"A": [True, True, False, False], "B": [True, False, False, True]}
+        )
+        ser_bool = Series([True, True, False, True])
+        result = df_bool.corrwith(ser_bool, min_periods=3)
+        expected = Series([0.57735, 0.57735], index=["A", "B"])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_describe.py b/pandas/pandas/tests/frame/methods/test_describe.py
new file mode 100644
index 0000000000000..e9206e86b7b08
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_describe.py
@@ -0,0 +1,415 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameDescribe:
+    def test_describe_bool_in_mixed_frame(self):
+        df = DataFrame(
+            {
+                "string_data": ["a", "b", "c", "d", "e"],
+                "bool_data": [True, True, False, False, False],
+                "int_data": [10, 20, 30, 40, 50],
+            }
+        )
+
+        # Integer data are included in .describe() output,
+        # Boolean and string data are not.
+        result = df.describe()
+        expected = DataFrame(
+            {"int_data": [5, 30, df.int_data.std(), 10, 20, 30, 40, 50]},
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Top value is a boolean value that is False
+        result = df.describe(include=["bool"])
+
+        expected = DataFrame(
+            {"bool_data": [5, 2, False, 3]}, index=["count", "unique", "top", "freq"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_empty_object(self):
+        # GH#27183
+        df = DataFrame({"A": [None, None]}, dtype=object)
+        result = df.describe()
+        expected = DataFrame(
+            {"A": [0, 0, np.nan, np.nan]},
+            dtype=object,
+            index=["count", "unique", "top", "freq"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:0].describe()
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_bool_frame(self):
+        # GH#13891
+        df = DataFrame(
+            {
+                "bool_data_1": [False, False, True, True],
+                "bool_data_2": [False, True, True, True],
+            }
+        )
+        result = df.describe()
+        expected = DataFrame(
+            {"bool_data_1": [4, 2, False, 2], "bool_data_2": [4, 2, True, 3]},
+            index=["count", "unique", "top", "freq"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "bool_data": [False, False, True, True, False],
+                "int_data": [0, 1, 2, 3, 4],
+            }
+        )
+        result = df.describe()
+        expected = DataFrame(
+            {"int_data": [5, 2, df.int_data.std(), 0, 1, 2, 3, 4]},
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {"bool_data": [False, False, True, True], "str_data": ["a", "b", "c", "a"]}
+        )
+        result = df.describe()
+        expected = DataFrame(
+            {"bool_data": [4, 2, False, 2], "str_data": [4, 3, "a", 2]},
+            index=["count", "unique", "top", "freq"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_categorical(self):
+        df = DataFrame({"value": np.random.default_rng(2).integers(0, 10000, 100)})
+        labels = [f"{i} - {i + 499}" for i in range(0, 10000, 500)]
+        cat_labels = Categorical(labels, labels)
+
+        df = df.sort_values(by=["value"], ascending=True)
+        df["value_group"] = pd.cut(
+            df.value, range(0, 10500, 500), right=False, labels=cat_labels
+        )
+        cat = df
+
+        # Categoricals should not show up together with numerical columns
+        result = cat.describe()
+        assert len(result.columns) == 1
+
+        # In a frame, describe() for the cat should be the same as for string
+        # arrays (count, unique, top, freq)
+
+        cat = Categorical(
+            ["a", "b", "b", "b"], categories=["a", "b", "c"], ordered=True
+        )
+        s = Series(cat)
+        result = s.describe()
+        expected = Series([4, 2, "b", 3], index=["count", "unique", "top", "freq"])
+        tm.assert_series_equal(result, expected)
+
+        cat = Series(Categorical(["a", "b", "c", "c"]))
+        df3 = DataFrame({"cat": cat, "s": ["a", "b", "c", "c"]})
+        result = df3.describe()
+        tm.assert_numpy_array_equal(result["cat"].values, result["s"].values)
+
+    def test_describe_empty_categorical_column(self):
+        # GH#26397
+        # Ensure the index of an empty categorical DataFrame column
+        # also contains (count, unique, top, freq)
+        df = DataFrame({"empty_col": Categorical([])})
+        result = df.describe()
+        expected = DataFrame(
+            {"empty_col": [0, 0, np.nan, np.nan]},
+            index=["count", "unique", "top", "freq"],
+            dtype="object",
+        )
+        tm.assert_frame_equal(result, expected)
+        # ensure NaN, not None
+        assert np.isnan(result.iloc[2, 0])
+        assert np.isnan(result.iloc[3, 0])
+
+    def test_describe_categorical_columns(self):
+        # GH#11558
+        columns = pd.CategoricalIndex(["int1", "int2", "obj"], ordered=True, name="XXX")
+        df = DataFrame(
+            {
+                "int1": [10, 20, 30, 40, 50],
+                "int2": [10, 20, 30, 40, 50],
+                "obj": ["A", 0, None, "X", 1],
+            },
+            columns=columns,
+        )
+        result = df.describe()
+
+        exp_columns = pd.CategoricalIndex(
+            ["int1", "int2"],
+            categories=["int1", "int2", "obj"],
+            ordered=True,
+            name="XXX",
+        )
+        expected = DataFrame(
+            {
+                "int1": [5, 30, df.int1.std(), 10, 20, 30, 40, 50],
+                "int2": [5, 30, df.int2.std(), 10, 20, 30, 40, 50],
+            },
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+            columns=exp_columns,
+        )
+
+        tm.assert_frame_equal(result, expected)
+        tm.assert_categorical_equal(result.columns.values, expected.columns.values)
+
+    def test_describe_datetime_columns(self):
+        columns = pd.DatetimeIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01"],
+            freq="MS",
+            tz="US/Eastern",
+            name="XXX",
+        )
+        df = DataFrame(
+            {
+                0: [10, 20, 30, 40, 50],
+                1: [10, 20, 30, 40, 50],
+                2: ["A", 0, None, "X", 1],
+            }
+        )
+        df.columns = columns
+        result = df.describe()
+
+        exp_columns = pd.DatetimeIndex(
+            ["2011-01-01", "2011-02-01"], freq="MS", tz="US/Eastern", name="XXX"
+        )
+        expected = DataFrame(
+            {
+                0: [5, 30, df.iloc[:, 0].std(), 10, 20, 30, 40, 50],
+                1: [5, 30, df.iloc[:, 1].std(), 10, 20, 30, 40, 50],
+            },
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        expected.columns = exp_columns
+        tm.assert_frame_equal(result, expected)
+        assert result.columns.freq == "MS"
+        assert result.columns.tz == expected.columns.tz
+
+    def test_describe_timedelta_values(self):
+        # GH#6145
+        t1 = pd.timedelta_range("1 days", freq="D", periods=5)
+        t2 = pd.timedelta_range("1 hours", freq="h", periods=5)
+        df = DataFrame({"t1": t1, "t2": t2})
+
+        expected = DataFrame(
+            {
+                "t1": [
+                    5,
+                    pd.Timedelta("3 days"),
+                    df.iloc[:, 0].std(),
+                    pd.Timedelta("1 days"),
+                    pd.Timedelta("2 days"),
+                    pd.Timedelta("3 days"),
+                    pd.Timedelta("4 days"),
+                    pd.Timedelta("5 days"),
+                ],
+                "t2": [
+                    5,
+                    pd.Timedelta("3 hours"),
+                    df.iloc[:, 1].std(),
+                    pd.Timedelta("1 hours"),
+                    pd.Timedelta("2 hours"),
+                    pd.Timedelta("3 hours"),
+                    pd.Timedelta("4 hours"),
+                    pd.Timedelta("5 hours"),
+                ],
+            },
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+
+        result = df.describe()
+        tm.assert_frame_equal(result, expected)
+
+        exp_repr = (
+            "                              t1                         t2\n"
+            "count                          5                          5\n"
+            "mean             3 days 00:00:00            0 days 03:00:00\n"
+            "std    1 days 13:56:50.394919273  0 days 01:34:52.099788303\n"
+            "min              1 days 00:00:00            0 days 01:00:00\n"
+            "25%              2 days 00:00:00            0 days 02:00:00\n"
+            "50%              3 days 00:00:00            0 days 03:00:00\n"
+            "75%              4 days 00:00:00            0 days 04:00:00\n"
+            "max              5 days 00:00:00            0 days 05:00:00"
+        )
+        assert repr(result) == exp_repr
+
+    def test_describe_tz_values(self, tz_naive_fixture):
+        # GH#21332
+        tz = tz_naive_fixture
+        s1 = Series(range(5))
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s2 = Series(date_range(start, end, tz=tz))
+        df = DataFrame({"s1": s1, "s2": s2})
+
+        expected = DataFrame(
+            {
+                "s1": [5, 2, 0, 1, 2, 3, 4, 1.581139],
+                "s2": [
+                    5,
+                    Timestamp(2018, 1, 3).tz_localize(tz),
+                    start.tz_localize(tz),
+                    s2[1],
+                    s2[2],
+                    s2[3],
+                    end.tz_localize(tz),
+                    np.nan,
+                ],
+            },
+            index=["count", "mean", "min", "25%", "50%", "75%", "max", "std"],
+        )
+        result = df.describe(include="all")
+        tm.assert_frame_equal(result, expected)
+
+    def test_datetime_is_numeric_includes_datetime(self):
+        df = DataFrame({"a": date_range("2012", periods=3), "b": [1, 2, 3]})
+        result = df.describe()
+        expected = DataFrame(
+            {
+                "a": [
+                    3,
+                    Timestamp("2012-01-02"),
+                    Timestamp("2012-01-01"),
+                    Timestamp("2012-01-01T12:00:00"),
+                    Timestamp("2012-01-02"),
+                    Timestamp("2012-01-02T12:00:00"),
+                    Timestamp("2012-01-03"),
+                    np.nan,
+                ],
+                "b": [3, 2, 1, 1.5, 2, 2.5, 3, 1],
+            },
+            index=["count", "mean", "min", "25%", "50%", "75%", "max", "std"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_tz_values2(self):
+        tz = "CET"
+        s1 = Series(range(5))
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s2 = Series(date_range(start, end, tz=tz))
+        df = DataFrame({"s1": s1, "s2": s2})
+
+        s1_ = s1.describe()
+        s2_ = s2.describe()
+        idx = [
+            "count",
+            "mean",
+            "min",
+            "25%",
+            "50%",
+            "75%",
+            "max",
+            "std",
+        ]
+        expected = pd.concat([s1_, s2_], axis=1, keys=["s1", "s2"]).reindex(idx)
+
+        result = df.describe(include="all")
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_percentiles_integer_idx(self):
+        # GH#26660
+        df = DataFrame({"x": [1]})
+        pct = np.linspace(0, 1, 10 + 1)
+        result = df.describe(percentiles=pct)
+
+        expected = DataFrame(
+            {"x": [1.0, 1.0, np.nan, 1.0, *(1.0 for _ in pct), 1.0]},
+            index=[
+                "count",
+                "mean",
+                "std",
+                "min",
+                "0%",
+                "10%",
+                "20%",
+                "30%",
+                "40%",
+                "50%",
+                "60%",
+                "70%",
+                "80%",
+                "90%",
+                "100%",
+                "max",
+            ],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_does_not_raise_error_for_dictlike_elements(self):
+        # GH#32409
+        df = DataFrame([{"test": {"a": "1"}}, {"test": {"a": "2"}}])
+        expected = DataFrame(
+            {"test": [2, 2, {"a": "1"}, 1]}, index=["count", "unique", "top", "freq"]
+        )
+        result = df.describe()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("exclude", ["x", "y", ["x", "y"], ["x", "z"]])
+    def test_describe_when_include_all_exclude_not_allowed(self, exclude):
+        """
+        When include is 'all', then setting exclude != None is not allowed.
+        """
+        df = DataFrame({"x": [1], "y": [2], "z": [3]})
+        msg = "exclude must be None when include is 'all'"
+        with pytest.raises(ValueError, match=msg):
+            df.describe(include="all", exclude=exclude)
+
+    def test_describe_with_duplicate_columns(self):
+        df = DataFrame(
+            [[1, 1, 1], [2, 2, 2], [3, 3, 3]],
+            columns=["bar", "a", "a"],
+            dtype="float64",
+        )
+        result = df.describe()
+        ser = df.iloc[:, 0].describe()
+        expected = pd.concat([ser, ser, ser], keys=df.columns, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_ea_with_na(self, any_numeric_ea_dtype):
+        # GH#48778
+
+        df = DataFrame({"a": [1, pd.NA, pd.NA], "b": pd.NA}, dtype=any_numeric_ea_dtype)
+        result = df.describe()
+        expected = DataFrame(
+            {"a": [1.0, 1.0, pd.NA] + [1.0] * 5, "b": [0.0] + [pd.NA] * 7},
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+            dtype="Float64",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_describe_exclude_pa_dtype(self):
+        # GH#52570
+        pa = pytest.importorskip("pyarrow")
+        df = DataFrame(
+            {
+                "a": Series([1, 2, 3], dtype=pd.ArrowDtype(pa.int8())),
+                "b": Series([1, 2, 3], dtype=pd.ArrowDtype(pa.int16())),
+                "c": Series([1, 2, 3], dtype=pd.ArrowDtype(pa.int32())),
+            }
+        )
+        result = df.describe(
+            include=pd.ArrowDtype(pa.int8()), exclude=pd.ArrowDtype(pa.int32())
+        )
+        expected = DataFrame(
+            {"a": [3, 2, 1, 1, 1.5, 2, 2.5, 3]},
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+            dtype=pd.ArrowDtype(pa.float64()),
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_diff.py b/pandas/pandas/tests/frame/methods/test_diff.py
new file mode 100644
index 0000000000000..bef18dbaf8a8a
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_diff.py
@@ -0,0 +1,308 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameDiff:
+    def test_diff_requires_integer(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((2, 2)))
+        with pytest.raises(ValueError, match="periods must be an integer"):
+            df.diff(1.5)
+
+    # GH#44572 np.int64 is accepted
+    @pytest.mark.parametrize("num", [1, np.int64(1)])
+    def test_diff(self, datetime_frame, num):
+        df = datetime_frame
+        the_diff = df.diff(num)
+
+        expected = df["A"] - df["A"].shift(num)
+        tm.assert_series_equal(the_diff["A"], expected)
+
+    def test_diff_int_dtype(self):
+        # int dtype
+        a = 10_000_000_000_000_000
+        b = a + 1
+        ser = Series([a, b])
+
+        rs = DataFrame({"s": ser}).diff()
+        assert rs.s[1] == 1
+
+    def test_diff_mixed_numeric(self, datetime_frame):
+        # mixed numeric
+        tf = datetime_frame.astype("float32")
+        the_diff = tf.diff(1)
+        tm.assert_series_equal(the_diff["A"], tf["A"] - tf["A"].shift(1))
+
+    def test_diff_axis1_nonconsolidated(self):
+        # GH#10907
+        df = DataFrame({"y": Series([2]), "z": Series([3])})
+        df.insert(0, "x", 1)
+        result = df.diff(axis=1)
+        expected = DataFrame({"x": np.nan, "y": Series(1), "z": Series(1)})
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_timedelta64_with_nat(self):
+        # GH#32441
+        arr = np.arange(6).reshape(3, 2).astype("timedelta64[ns]")
+        arr[:, 0] = np.timedelta64("NaT", "ns")
+
+        df = DataFrame(arr)
+        result = df.diff(1, axis=0)
+
+        expected = DataFrame({0: df[0], 1: [pd.NaT, pd.Timedelta(2), pd.Timedelta(2)]})
+        tm.assert_equal(result, expected)
+
+        result = df.diff(0)
+        expected = df - df
+        assert expected[0].isna().all()
+        tm.assert_equal(result, expected)
+
+        result = df.diff(-1, axis=1)
+        expected = df * np.nan
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_diff_datetime_axis0_with_nat(self, tz, unit):
+        # GH#32441
+        dti = pd.DatetimeIndex(["NaT", "2019-01-01", "2019-01-02"], tz=tz).as_unit(unit)
+        ser = Series(dti)
+
+        df = ser.to_frame()
+
+        result = df.diff()
+        ex_index = pd.TimedeltaIndex([pd.NaT, pd.NaT, pd.Timedelta(days=1)]).as_unit(
+            unit
+        )
+        expected = Series(ex_index).to_frame()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_diff_datetime_with_nat_zero_periods(self, tz):
+        # diff on NaT values should give NaT, not timedelta64(0)
+        dti = date_range("2016-01-01", periods=4, tz=tz)
+        ser = Series(dti)
+        df = ser.to_frame().copy()
+
+        df[1] = ser.copy()
+
+        df.iloc[:, 0] = pd.NaT
+
+        expected = df - df
+        assert expected[0].isna().all()
+
+        result = df.diff(0, axis=0)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.diff(0, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_diff_datetime_axis0(self, tz):
+        # GH#18578
+        df = DataFrame(
+            {
+                0: date_range("2010", freq="D", periods=2, tz=tz),
+                1: date_range("2010", freq="D", periods=2, tz=tz),
+            }
+        )
+
+        result = df.diff(axis=0)
+        expected = DataFrame(
+            {
+                0: pd.TimedeltaIndex(["NaT", "1 days"]),
+                1: pd.TimedeltaIndex(["NaT", "1 days"]),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_diff_datetime_axis1(self, tz):
+        # GH#18578
+        df = DataFrame(
+            {
+                0: date_range("2010", freq="D", periods=2, tz=tz),
+                1: date_range("2010", freq="D", periods=2, tz=tz),
+            }
+        )
+
+        result = df.diff(axis=1)
+        expected = DataFrame(
+            {
+                0: pd.TimedeltaIndex(["NaT", "NaT"]),
+                1: pd.TimedeltaIndex(["0 days", "0 days"]),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_timedelta(self, unit):
+        # GH#4533
+        df = DataFrame(
+            {
+                "time": [Timestamp("20130101 9:01"), Timestamp("20130101 9:02")],
+                "value": [1.0, 2.0],
+            }
+        )
+        df["time"] = df["time"].dt.as_unit(unit)
+
+        res = df.diff()
+        exp = DataFrame(
+            [[pd.NaT, np.nan], [pd.Timedelta("00:01:00"), 1]], columns=["time", "value"]
+        )
+        exp["time"] = exp["time"].dt.as_unit(unit)
+        tm.assert_frame_equal(res, exp)
+
+    def test_diff_mixed_dtype(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df["A"] = np.array([1, 2, 3, 4, 5], dtype=object)
+
+        result = df.diff()
+        assert result[0].dtype == np.float64
+
+    def test_diff_neg_n(self, datetime_frame):
+        rs = datetime_frame.diff(-1)
+        xp = datetime_frame - datetime_frame.shift(-1)
+        tm.assert_frame_equal(rs, xp)
+
+    def test_diff_float_n(self, datetime_frame):
+        rs = datetime_frame.diff(1.0)
+        xp = datetime_frame.diff(1)
+        tm.assert_frame_equal(rs, xp)
+
+    def test_diff_axis(self):
+        # GH#9727
+        df = DataFrame([[1.0, 2.0], [3.0, 4.0]])
+        tm.assert_frame_equal(
+            df.diff(axis=1), DataFrame([[np.nan, 1.0], [np.nan, 1.0]])
+        )
+        tm.assert_frame_equal(
+            df.diff(axis=0), DataFrame([[np.nan, np.nan], [2.0, 2.0]])
+        )
+
+    def test_diff_period(self):
+        # GH#32995 Don't pass an incorrect axis
+        pi = date_range("2016-01-01", periods=3).to_period("D")
+        df = DataFrame({"A": pi})
+
+        result = df.diff(1, axis=1)
+
+        expected = (df - pd.NaT).astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_axis1_mixed_dtypes(self):
+        # GH#32995 operate column-wise when we have mixed dtypes and axis=1
+        df = DataFrame({"A": range(3), "B": 2 * np.arange(3, dtype=np.float64)})
+
+        expected = DataFrame({"A": [np.nan, np.nan, np.nan], "B": df["B"] / 2})
+
+        result = df.diff(axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # GH#21437 mixed-float-dtypes
+        df = DataFrame(
+            {"a": np.arange(3, dtype="float32"), "b": np.arange(3, dtype="float64")}
+        )
+        result = df.diff(axis=1)
+        expected = DataFrame({"a": df["a"] * np.nan, "b": df["b"] * 0})
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_axis1_mixed_dtypes_large_periods(self):
+        # GH#32995 operate column-wise when we have mixed dtypes and axis=1
+        df = DataFrame({"A": range(3), "B": 2 * np.arange(3, dtype=np.float64)})
+
+        expected = df * np.nan
+
+        result = df.diff(axis=1, periods=3)
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_axis1_mixed_dtypes_negative_periods(self):
+        # GH#32995 operate column-wise when we have mixed dtypes and axis=1
+        df = DataFrame({"A": range(3), "B": 2 * np.arange(3, dtype=np.float64)})
+
+        expected = DataFrame({"A": -1.0 * df["A"], "B": df["B"] * np.nan})
+
+        result = df.diff(axis=1, periods=-1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_sparse(self):
+        # GH#28813 .diff() should work for sparse dataframes as well
+        sparse_df = DataFrame([[0, 1], [1, 0]], dtype="Sparse[int]")
+
+        result = sparse_df.diff()
+        expected = DataFrame(
+            [[np.nan, np.nan], [1.0, -1.0]], dtype=pd.SparseDtype("float", 0.0)
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "axis,expected",
+        [
+            (
+                0,
+                DataFrame(
+                    {
+                        "a": [np.nan, 0, 1, 0, np.nan, np.nan, np.nan, 0],
+                        "b": [np.nan, 1, np.nan, np.nan, -2, 1, np.nan, np.nan],
+                        "c": np.repeat(np.nan, 8),
+                        "d": [np.nan, 3, 5, 7, 9, 11, 13, 15],
+                    },
+                    dtype="Int64",
+                ),
+            ),
+            (
+                1,
+                DataFrame(
+                    {
+                        "a": np.repeat(np.nan, 8),
+                        "b": [0, 1, np.nan, 1, np.nan, np.nan, np.nan, 0],
+                        "c": np.repeat(np.nan, 8),
+                        "d": np.repeat(np.nan, 8),
+                    },
+                    dtype="Int64",
+                ),
+            ),
+        ],
+    )
+    def test_diff_integer_na(self, axis, expected):
+        # GH#24171 IntegerNA Support for DataFrame.diff()
+        df = DataFrame(
+            {
+                "a": np.repeat([0, 1, np.nan, 2], 2),
+                "b": np.tile([0, 1, np.nan, 2], 2),
+                "c": np.repeat(np.nan, 8),
+                "d": np.arange(1, 9) ** 2,
+            },
+            dtype="Int64",
+        )
+
+        # Test case for default behaviour of diff
+        result = df.diff(axis=axis)
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_readonly(self):
+        # https://github.com/pandas-dev/pandas/issues/35559
+        arr = np.random.default_rng(2).standard_normal((5, 2))
+        arr.flags.writeable = False
+        df = DataFrame(arr)
+        result = df.diff()
+        expected = DataFrame(np.array(df)).diff()
+        tm.assert_frame_equal(result, expected)
+
+    def test_diff_all_int_dtype(self, any_int_numpy_dtype):
+        # GH 14773
+        df = DataFrame(range(5))
+        df = df.astype(any_int_numpy_dtype)
+        result = df.diff()
+        expected_dtype = (
+            "float32" if any_int_numpy_dtype in ("int8", "int16") else "float64"
+        )
+        expected = DataFrame([np.nan, 1.0, 1.0, 1.0, 1.0], dtype=expected_dtype)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_dot.py b/pandas/pandas/tests/frame/methods/test_dot.py
new file mode 100644
index 0000000000000..3e01f67c8794b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_dot.py
@@ -0,0 +1,155 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class DotSharedTests:
+    @pytest.fixture
+    def obj(self):
+        raise NotImplementedError
+
+    @pytest.fixture
+    def other(self) -> DataFrame:
+        """
+        other is a DataFrame that is indexed so that obj.dot(other) is valid
+        """
+        raise NotImplementedError
+
+    @pytest.fixture
+    def expected(self, obj, other) -> DataFrame:
+        """
+        The expected result of obj.dot(other)
+        """
+        raise NotImplementedError
+
+    @classmethod
+    def reduced_dim_assert(cls, result, expected):
+        """
+        Assertion about results with 1 fewer dimension that self.obj
+        """
+        raise NotImplementedError
+
+    def test_dot_equiv_values_dot(self, obj, other, expected):
+        # `expected` is constructed from obj.values.dot(other.values)
+        result = obj.dot(other)
+        tm.assert_equal(result, expected)
+
+    def test_dot_2d_ndarray(self, obj, other, expected):
+        # Check ndarray argument; in this case we get matching values,
+        #  but index/columns may not match
+        result = obj.dot(other.values)
+        assert np.all(result == expected.values)
+
+    def test_dot_1d_ndarray(self, obj, expected):
+        # can pass correct-length array
+        row = obj.iloc[0] if obj.ndim == 2 else obj
+
+        result = obj.dot(row.values)
+        expected = obj.dot(row)
+        self.reduced_dim_assert(result, expected)
+
+    def test_dot_series(self, obj, other, expected):
+        # Check series argument
+        result = obj.dot(other["1"])
+        self.reduced_dim_assert(result, expected["1"])
+
+    def test_dot_series_alignment(self, obj, other, expected):
+        result = obj.dot(other.iloc[::-1]["1"])
+        self.reduced_dim_assert(result, expected["1"])
+
+    def test_dot_aligns(self, obj, other, expected):
+        # Check index alignment
+        other2 = other.iloc[::-1]
+        result = obj.dot(other2)
+        tm.assert_equal(result, expected)
+
+    def test_dot_shape_mismatch(self, obj):
+        msg = "Dot product shape mismatch"
+        # exception raised is of type Exception
+        with pytest.raises(Exception, match=msg):
+            obj.dot(obj.values[:3])
+
+    def test_dot_misaligned(self, obj, other):
+        msg = "matrices are not aligned"
+        with pytest.raises(ValueError, match=msg):
+            obj.dot(other.T)
+
+
+class TestSeriesDot(DotSharedTests):
+    @pytest.fixture
+    def obj(self):
+        return Series(
+            np.random.default_rng(2).standard_normal(4), index=["p", "q", "r", "s"]
+        )
+
+    @pytest.fixture
+    def other(self):
+        return DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            index=["1", "2", "3"],
+            columns=["p", "q", "r", "s"],
+        ).T
+
+    @pytest.fixture
+    def expected(self, obj, other):
+        return Series(np.dot(obj.values, other.values), index=other.columns)
+
+    @classmethod
+    def reduced_dim_assert(cls, result, expected):
+        """
+        Assertion about results with 1 fewer dimension that self.obj
+        """
+        tm.assert_almost_equal(result, expected)
+
+
+class TestDataFrameDot(DotSharedTests):
+    @pytest.fixture
+    def obj(self):
+        return DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            index=["a", "b", "c"],
+            columns=["p", "q", "r", "s"],
+        )
+
+    @pytest.fixture
+    def other(self):
+        return DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)),
+            index=["p", "q", "r", "s"],
+            columns=["1", "2"],
+        )
+
+    @pytest.fixture
+    def expected(self, obj, other):
+        return DataFrame(
+            np.dot(obj.values, other.values), index=obj.index, columns=other.columns
+        )
+
+    @classmethod
+    def reduced_dim_assert(cls, result, expected):
+        """
+        Assertion about results with 1 fewer dimension that self.obj
+        """
+        tm.assert_series_equal(result, expected, check_names=False)
+        assert result.name is None
+
+
+@pytest.mark.parametrize(
+    "dtype,exp_dtype",
+    [("Float32", "Float64"), ("Int16", "Int32"), ("float[pyarrow]", "double[pyarrow]")],
+)
+def test_arrow_dtype(dtype, exp_dtype):
+    pytest.importorskip("pyarrow")
+
+    cols = ["a", "b"]
+    df_a = DataFrame([[1, 2], [3, 4], [5, 6]], columns=cols, dtype="int32")
+    df_b = DataFrame([[1, 0], [0, 1]], index=cols, dtype=dtype)
+    result = df_a.dot(df_b)
+    expected = DataFrame([[1, 2], [3, 4], [5, 6]], dtype=exp_dtype)
+
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_drop.py b/pandas/pandas/tests/frame/methods/test_drop.py
new file mode 100644
index 0000000000000..d9668ce46c943
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_drop.py
@@ -0,0 +1,542 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "msg,labels,level",
+    [
+        (r"labels \[4\] not found in level", 4, "a"),
+        (r"labels \[7\] not found in level", 7, "b"),
+    ],
+)
+def test_drop_raise_exception_if_labels_not_in_level(msg, labels, level):
+    # GH 8594
+    mi = MultiIndex.from_arrays([[1, 2, 3], [4, 5, 6]], names=["a", "b"])
+    s = Series([10, 20, 30], index=mi)
+    df = DataFrame([10, 20, 30], index=mi)
+
+    with pytest.raises(KeyError, match=msg):
+        s.drop(labels, level=level)
+    with pytest.raises(KeyError, match=msg):
+        df.drop(labels, level=level)
+
+
+@pytest.mark.parametrize("labels,level", [(4, "a"), (7, "b")])
+def test_drop_errors_ignore(labels, level):
+    # GH 8594
+    mi = MultiIndex.from_arrays([[1, 2, 3], [4, 5, 6]], names=["a", "b"])
+    s = Series([10, 20, 30], index=mi)
+    df = DataFrame([10, 20, 30], index=mi)
+
+    expected_s = s.drop(labels, level=level, errors="ignore")
+    tm.assert_series_equal(s, expected_s)
+
+    expected_df = df.drop(labels, level=level, errors="ignore")
+    tm.assert_frame_equal(df, expected_df)
+
+
+def test_drop_with_non_unique_datetime_index_and_invalid_keys():
+    # GH 30399
+
+    # define dataframe with unique datetime index
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 3)),
+        columns=["a", "b", "c"],
+        index=pd.date_range("2012", freq="h", periods=5),
+    )
+    # create dataframe with non-unique datetime index
+    df = df.iloc[[0, 2, 2, 3]].copy()
+
+    with pytest.raises(KeyError, match="not found in axis"):
+        df.drop(["a", "b"])  # Dropping with labels not exist in the index
+
+
+class TestDataFrameDrop:
+    def test_drop_names(self):
+        df = DataFrame(
+            [[1, 2, 3], [3, 4, 5], [5, 6, 7]],
+            index=["a", "b", "c"],
+            columns=["d", "e", "f"],
+        )
+        df.index.name, df.columns.name = "first", "second"
+        df_dropped_b = df.drop("b")
+        df_dropped_e = df.drop("e", axis=1)
+        df_inplace_b, df_inplace_e = df.copy(), df.copy()
+        return_value = df_inplace_b.drop("b", inplace=True)
+        assert return_value is None
+        return_value = df_inplace_e.drop("e", axis=1, inplace=True)
+        assert return_value is None
+        for obj in (df_dropped_b, df_dropped_e, df_inplace_b, df_inplace_e):
+            assert obj.index.name == "first"
+            assert obj.columns.name == "second"
+        assert list(df.columns) == ["d", "e", "f"]
+
+        msg = r"\['g'\] not found in axis"
+        with pytest.raises(KeyError, match=msg):
+            df.drop(["g"])
+        with pytest.raises(KeyError, match=msg):
+            df.drop(["g"], axis=1)
+
+        # errors = 'ignore'
+        dropped = df.drop(["g"], errors="ignore")
+        expected = Index(["a", "b", "c"], name="first")
+        tm.assert_index_equal(dropped.index, expected)
+
+        dropped = df.drop(["b", "g"], errors="ignore")
+        expected = Index(["a", "c"], name="first")
+        tm.assert_index_equal(dropped.index, expected)
+
+        dropped = df.drop(["g"], axis=1, errors="ignore")
+        expected = Index(["d", "e", "f"], name="second")
+        tm.assert_index_equal(dropped.columns, expected)
+
+        dropped = df.drop(["d", "g"], axis=1, errors="ignore")
+        expected = Index(["e", "f"], name="second")
+        tm.assert_index_equal(dropped.columns, expected)
+
+        # GH 16398
+        dropped = df.drop([], errors="ignore")
+        expected = Index(["a", "b", "c"], name="first")
+        tm.assert_index_equal(dropped.index, expected)
+
+    def test_drop(self):
+        simple = DataFrame({"A": [1, 2, 3, 4], "B": [0, 1, 2, 3]})
+        tm.assert_frame_equal(simple.drop("A", axis=1), simple[["B"]])
+        tm.assert_frame_equal(simple.drop(["A", "B"], axis="columns"), simple[[]])
+        tm.assert_frame_equal(simple.drop([0, 1, 3], axis=0), simple.loc[[2], :])
+        tm.assert_frame_equal(simple.drop([0, 3], axis="index"), simple.loc[[1, 2], :])
+
+        with pytest.raises(KeyError, match=r"\[5\] not found in axis"):
+            simple.drop(5)
+        with pytest.raises(KeyError, match=r"\['C'\] not found in axis"):
+            simple.drop("C", axis=1)
+        with pytest.raises(KeyError, match=r"\[5\] not found in axis"):
+            simple.drop([1, 5])
+        with pytest.raises(KeyError, match=r"\['C'\] not found in axis"):
+            simple.drop(["A", "C"], axis=1)
+
+        # GH 42881
+        with pytest.raises(KeyError, match=r"\['C', 'D', 'F'\] not found in axis"):
+            simple.drop(["C", "D", "F"], axis=1)
+
+        # errors = 'ignore'
+        tm.assert_frame_equal(simple.drop(5, errors="ignore"), simple)
+        tm.assert_frame_equal(
+            simple.drop([0, 5], errors="ignore"), simple.loc[[1, 2, 3], :]
+        )
+        tm.assert_frame_equal(simple.drop("C", axis=1, errors="ignore"), simple)
+        tm.assert_frame_equal(
+            simple.drop(["A", "C"], axis=1, errors="ignore"), simple[["B"]]
+        )
+
+        # non-unique - wheee!
+        nu_df = DataFrame(
+            list(zip(range(3), range(-3, 1), list("abc"))), columns=["a", "a", "b"]
+        )
+        tm.assert_frame_equal(nu_df.drop("a", axis=1), nu_df[["b"]])
+        tm.assert_frame_equal(nu_df.drop("b", axis="columns"), nu_df["a"])
+        tm.assert_frame_equal(nu_df.drop([]), nu_df)  # GH 16398
+
+        nu_df = nu_df.set_index(Index(["X", "Y", "X"]))
+        nu_df.columns = list("abc")
+        tm.assert_frame_equal(nu_df.drop("X", axis="rows"), nu_df.loc[["Y"], :])
+        tm.assert_frame_equal(nu_df.drop(["X", "Y"], axis=0), nu_df.loc[[], :])
+
+        # inplace cache issue
+        # GH#5628
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=list("abc")
+        )
+        expected = df[~(df.b > 0)]
+        return_value = df.drop(labels=df[df.b > 0].index, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(df, expected)
+
+    def test_drop_multiindex_not_lexsorted(self, performance_warning):
+        # GH#11640
+
+        # define the lexsorted version
+        lexsorted_mi = MultiIndex.from_tuples(
+            [("a", ""), ("b1", "c1"), ("b2", "c2")], names=["b", "c"]
+        )
+        lexsorted_df = DataFrame([[1, 3, 4]], columns=lexsorted_mi)
+        assert lexsorted_df.columns._is_lexsorted()
+
+        # define the non-lexsorted version
+        not_lexsorted_df = DataFrame(
+            columns=["a", "b", "c", "d"], data=[[1, "b1", "c1", 3], [1, "b2", "c2", 4]]
+        )
+        not_lexsorted_df = not_lexsorted_df.pivot_table(
+            index="a", columns=["b", "c"], values="d"
+        )
+        not_lexsorted_df = not_lexsorted_df.reset_index()
+        assert not not_lexsorted_df.columns._is_lexsorted()
+
+        expected = lexsorted_df.drop("a", axis=1).astype(float)
+        with tm.assert_produces_warning(performance_warning):
+            result = not_lexsorted_df.drop("a", axis=1)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_api_equivalence(self):
+        # equivalence of the labels/axis and index/columns API's (GH#12392)
+        df = DataFrame(
+            [[1, 2, 3], [3, 4, 5], [5, 6, 7]],
+            index=["a", "b", "c"],
+            columns=["d", "e", "f"],
+        )
+
+        res1 = df.drop("a")
+        res2 = df.drop(index="a")
+        tm.assert_frame_equal(res1, res2)
+
+        res1 = df.drop("d", axis=1)
+        res2 = df.drop(columns="d")
+        tm.assert_frame_equal(res1, res2)
+
+        res1 = df.drop(labels="e", axis=1)
+        res2 = df.drop(columns="e")
+        tm.assert_frame_equal(res1, res2)
+
+        res1 = df.drop(["a"], axis=0)
+        res2 = df.drop(index=["a"])
+        tm.assert_frame_equal(res1, res2)
+
+        res1 = df.drop(["a"], axis=0).drop(["d"], axis=1)
+        res2 = df.drop(index=["a"], columns=["d"])
+        tm.assert_frame_equal(res1, res2)
+
+        msg = "Cannot specify both 'labels' and 'index'/'columns'"
+        with pytest.raises(ValueError, match=msg):
+            df.drop(labels="a", index="b")
+
+        with pytest.raises(ValueError, match=msg):
+            df.drop(labels="a", columns="b")
+
+        msg = "Need to specify at least one of 'labels', 'index' or 'columns'"
+        with pytest.raises(ValueError, match=msg):
+            df.drop(axis=1)
+
+    @pytest.mark.parametrize(
+        "actual",
+        [
+            DataFrame([[1, 2, 3], [1, 2, 3]], index=["a", "a"]),
+            DataFrame([[1, 2, 3], [1, 2, 3]], index=["a", "b"]),
+            DataFrame([[1, 2, 3], [1, 2, 3]], index=["a", "b"]).set_index([0, 1]),
+            DataFrame([[1, 2, 3], [1, 2, 3]], index=["a", "a"]).set_index([0, 1]),
+        ],
+    )
+    def test_raise_on_drop_duplicate_index(self, actual):
+        # GH#19186
+        level = 0 if isinstance(actual.index, MultiIndex) else None
+        msg = re.escape("\"['c'] not found in axis\"")
+        with pytest.raises(KeyError, match=msg):
+            actual.drop("c", level=level, axis=0)
+        with pytest.raises(KeyError, match=msg):
+            actual.T.drop("c", level=level, axis=1)
+        expected_no_err = actual.drop("c", axis=0, level=level, errors="ignore")
+        tm.assert_frame_equal(expected_no_err, actual)
+        expected_no_err = actual.T.drop("c", axis=1, level=level, errors="ignore")
+        tm.assert_frame_equal(expected_no_err.T, actual)
+
+    @pytest.mark.parametrize("index", [[1, 2, 3], [1, 1, 2]])
+    @pytest.mark.parametrize("drop_labels", [[], [1], [2]])
+    def test_drop_empty_list(self, index, drop_labels):
+        # GH#21494
+        expected_index = [i for i in index if i not in drop_labels]
+        frame = DataFrame(index=index).drop(drop_labels)
+        tm.assert_frame_equal(frame, DataFrame(index=expected_index))
+
+    @pytest.mark.parametrize("index", [[1, 2, 3], [1, 2, 2]])
+    @pytest.mark.parametrize("drop_labels", [[1, 4], [4, 5]])
+    def test_drop_non_empty_list(self, index, drop_labels):
+        # GH# 21494
+        with pytest.raises(KeyError, match="not found in axis"):
+            DataFrame(index=index).drop(drop_labels)
+
+    @pytest.mark.parametrize(
+        "empty_listlike",
+        [
+            [],
+            {},
+            np.array([]),
+            Series([], dtype="datetime64[ns]"),
+            Index([]),
+            DatetimeIndex([]),
+        ],
+    )
+    def test_drop_empty_listlike_non_unique_datetime_index(self, empty_listlike):
+        # GH#27994
+        data = {"column_a": [5, 10], "column_b": ["one", "two"]}
+        index = [Timestamp("2021-01-01"), Timestamp("2021-01-01")]
+        df = DataFrame(data, index=index)
+
+        # Passing empty list-like should return the same DataFrame.
+        expected = df.copy()
+        result = df.drop(empty_listlike)
+        tm.assert_frame_equal(result, expected)
+
+    def test_mixed_depth_drop(self):
+        arrays = [
+            ["a", "top", "top", "routine1", "routine1", "routine2"],
+            ["", "OD", "OD", "result1", "result2", "result1"],
+            ["", "wx", "wy", "", "", ""],
+        ]
+
+        tuples = sorted(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples)
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 6)), columns=index)
+
+        result = df.drop("a", axis=1)
+        expected = df.drop([("a", "", "")], axis=1)
+        tm.assert_frame_equal(expected, result)
+
+        result = df.drop(["top"], axis=1)
+        expected = df.drop([("top", "OD", "wx")], axis=1)
+        expected = expected.drop([("top", "OD", "wy")], axis=1)
+        tm.assert_frame_equal(expected, result)
+
+        result = df.drop(("top", "OD", "wx"), axis=1)
+        expected = df.drop([("top", "OD", "wx")], axis=1)
+        tm.assert_frame_equal(expected, result)
+
+        expected = df.drop([("top", "OD", "wy")], axis=1)
+        expected = df.drop("top", axis=1)
+
+        result = df.drop("result1", level=1, axis=1)
+        expected = df.drop(
+            [("routine1", "result1", ""), ("routine2", "result1", "")], axis=1
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_drop_multiindex_other_level_nan(self):
+        # GH#12754
+        df = (
+            DataFrame(
+                {
+                    "A": ["one", "one", "two", "two"],
+                    "B": [np.nan, 0.0, 1.0, 2.0],
+                    "C": ["a", "b", "c", "c"],
+                    "D": [1, 2, 3, 4],
+                }
+            )
+            .set_index(["A", "B", "C"])
+            .sort_index()
+        )
+        result = df.drop("c", level="C")
+        expected = DataFrame(
+            [2, 1],
+            columns=["D"],
+            index=MultiIndex.from_tuples(
+                [("one", 0.0, "b"), ("one", np.nan, "a")], names=["A", "B", "C"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_nonunique(self):
+        df = DataFrame(
+            [
+                ["x-a", "x", "a", 1.5],
+                ["x-a", "x", "a", 1.2],
+                ["z-c", "z", "c", 3.1],
+                ["x-a", "x", "a", 4.1],
+                ["x-b", "x", "b", 5.1],
+                ["x-b", "x", "b", 4.1],
+                ["x-b", "x", "b", 2.2],
+                ["y-a", "y", "a", 1.2],
+                ["z-b", "z", "b", 2.1],
+            ],
+            columns=["var1", "var2", "var3", "var4"],
+        )
+
+        grp_size = df.groupby("var1").size()
+        drop_idx = grp_size.loc[grp_size == 1]
+
+        idf = df.set_index(["var1", "var2", "var3"])
+
+        # it works! GH#2101
+        result = idf.drop(drop_idx.index, level=0).reset_index()
+        expected = df[-df.var1.isin(drop_idx.index)]
+
+        result.index = expected.index
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_level(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        result = frame.drop(["bar", "qux"], level="first")
+        expected = frame.iloc[[0, 1, 2, 5, 6]]
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.drop(["two"], level="second")
+        expected = frame.iloc[[0, 2, 3, 6, 7, 9]]
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.T.drop(["bar", "qux"], axis=1, level="first")
+        expected = frame.iloc[[0, 1, 2, 5, 6]].T
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.T.drop(["two"], axis=1, level="second")
+        expected = frame.iloc[[0, 2, 3, 6, 7, 9]].T
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_level_nonunique_datetime(self):
+        # GH#12701
+        idx = Index([2, 3, 4, 4, 5], name="id")
+        idxdt = pd.to_datetime(
+            [
+                "2016-03-23 14:00",
+                "2016-03-23 15:00",
+                "2016-03-23 16:00",
+                "2016-03-23 16:00",
+                "2016-03-23 17:00",
+            ]
+        )
+        df = DataFrame(np.arange(10).reshape(5, 2), columns=list("ab"), index=idx)
+        df["tstamp"] = idxdt
+        df = df.set_index("tstamp", append=True)
+        ts = Timestamp("201603231600")
+        assert df.index.is_unique is False
+
+        result = df.drop(ts, level="tstamp")
+        expected = df.loc[idx != 4]
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_tz_aware_timestamp_across_dst(self, frame_or_series):
+        # GH#21761
+        start = Timestamp("2017-10-29", tz="Europe/Berlin")
+        end = Timestamp("2017-10-29 04:00:00", tz="Europe/Berlin")
+        index = pd.date_range(start, end, freq="15min")
+        data = frame_or_series(data=[1] * len(index), index=index)
+        result = data.drop(start)
+        expected_start = Timestamp("2017-10-29 00:15:00", tz="Europe/Berlin")
+        expected_idx = pd.date_range(expected_start, end, freq="15min")
+        expected = frame_or_series(data=[1] * len(expected_idx), index=expected_idx)
+        tm.assert_equal(result, expected)
+
+    def test_drop_preserve_names(self):
+        index = MultiIndex.from_arrays(
+            [[0, 0, 0, 1, 1, 1], [1, 2, 3, 1, 2, 3]], names=["one", "two"]
+        )
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((6, 3)), index=index)
+
+        result = df.drop([(0, 2)])
+        assert result.index.names == ("one", "two")
+
+    @pytest.mark.parametrize(
+        "operation", ["__iadd__", "__isub__", "__imul__", "__ipow__"]
+    )
+    @pytest.mark.parametrize("inplace", [False, True])
+    def test_inplace_drop_and_operation(self, operation, inplace):
+        # GH#30484
+        df = DataFrame({"x": range(5)})
+        expected = df.copy()
+        df["y"] = range(5)
+        y = df["y"]
+
+        with tm.assert_produces_warning(None):
+            if inplace:
+                df.drop("y", axis=1, inplace=inplace)
+            else:
+                df = df.drop("y", axis=1, inplace=inplace)
+
+            # Perform operation and check result
+            getattr(y, operation)(1)
+            tm.assert_frame_equal(df, expected)
+
+    def test_drop_with_non_unique_multiindex(self):
+        # GH#36293
+        mi = MultiIndex.from_arrays([["x", "y", "x"], ["i", "j", "i"]])
+        df = DataFrame([1, 2, 3], index=mi)
+        result = df.drop(index="x")
+        expected = DataFrame([2], index=MultiIndex.from_arrays([["y"], ["j"]]))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("indexer", [("a", "a"), [("a", "a")]])
+    def test_drop_tuple_with_non_unique_multiindex(self, indexer):
+        # GH#42771
+        idx = MultiIndex.from_product([["a", "b"], ["a", "a"]])
+        df = DataFrame({"x": range(len(idx))}, index=idx)
+        result = df.drop(index=[("a", "a")])
+        expected = DataFrame(
+            {"x": [2, 3]}, index=MultiIndex.from_tuples([("b", "a"), ("b", "a")])
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_with_duplicate_columns(self):
+        df = DataFrame(
+            [[1, 5, 7.0], [1, 5, 7.0], [1, 5, 7.0]], columns=["bar", "a", "a"]
+        )
+        result = df.drop(["a"], axis=1)
+        expected = DataFrame([[1], [1], [1]], columns=["bar"])
+        tm.assert_frame_equal(result, expected)
+        result = df.drop("a", axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_with_duplicate_columns2(self):
+        # drop buggy GH#6240
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(5),
+                "B": np.random.default_rng(2).standard_normal(5),
+                "C": np.random.default_rng(2).standard_normal(5),
+                "D": ["a", "b", "c", "d", "e"],
+            }
+        )
+
+        expected = df.take([0, 1, 1], axis=1)
+        df2 = df.take([2, 0, 1, 2, 1], axis=1)
+        result = df2.drop("C", axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_inplace_no_leftover_column_reference(self):
+        # GH 13934
+        df = DataFrame({"a": [1, 2, 3]}, columns=Index(["a"], dtype="object"))
+        a = df.a
+        df.drop(["a"], axis=1, inplace=True)
+        tm.assert_index_equal(df.columns, Index([], dtype="object"))
+        a -= a.mean()
+        tm.assert_index_equal(df.columns, Index([], dtype="object"))
+
+    def test_drop_level_missing_label_multiindex(self):
+        # GH 18561
+        df = DataFrame(index=MultiIndex.from_product([range(3), range(3)]))
+        with pytest.raises(KeyError, match="labels \\[5\\] not found in level"):
+            df.drop(5, level=0)
+
+    @pytest.mark.parametrize("idx, level", [(["a", "b"], 0), (["a"], None)])
+    def test_drop_index_ea_dtype(self, any_numeric_ea_dtype, idx, level):
+        # GH#45860
+        df = DataFrame(
+            {"a": [1, 2, 2, pd.NA], "b": 100}, dtype=any_numeric_ea_dtype
+        ).set_index(idx)
+        result = df.drop(Index([2, pd.NA]), level=level)
+        expected = DataFrame(
+            {"a": [1], "b": 100}, dtype=any_numeric_ea_dtype
+        ).set_index(idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_drop_parse_strings_datetime_index(self):
+        # GH #5355
+        df = DataFrame(
+            {"a": [1, 2], "b": [1, 2]},
+            index=[Timestamp("2000-01-03"), Timestamp("2000-01-04")],
+        )
+        result = df.drop("2000-01-03", axis=0)
+        expected = DataFrame({"a": [2], "b": [2]}, index=[Timestamp("2000-01-04")])
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_drop_duplicates.py b/pandas/pandas/tests/frame/methods/test_drop_duplicates.py
new file mode 100644
index 0000000000000..7feb3b6fd816d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_drop_duplicates.py
@@ -0,0 +1,516 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    NaT,
+    concat,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("subset", ["a", ["a"], ["a", "B"]])
+def test_drop_duplicates_with_misspelled_column_name(subset):
+    # GH 19730
+    df = DataFrame({"A": [0, 0, 1], "B": [0, 0, 1], "C": [0, 0, 1]})
+    msg = re.escape("Index(['a'], dtype=")
+
+    with pytest.raises(KeyError, match=msg):
+        df.drop_duplicates(subset)
+
+
+def test_drop_duplicates():
+    df = DataFrame(
+        {
+            "AAA": ["foo", "bar", "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1, 1, 2, 2, 2, 2, 1, 2],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates("AAA")
+    expected = df[:2]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("AAA", keep="last")
+    expected = df.loc[[6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("AAA", keep=False)
+    expected = df.loc[[]]
+    tm.assert_frame_equal(result, expected)
+    assert len(result) == 0
+
+    # multi column
+    expected = df.loc[[0, 1, 2, 3]]
+    result = df.drop_duplicates(np.array(["AAA", "B"]))
+    tm.assert_frame_equal(result, expected)
+    result = df.drop_duplicates(["AAA", "B"])
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(("AAA", "B"), keep="last")
+    expected = df.loc[[0, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(("AAA", "B"), keep=False)
+    expected = df.loc[[0]]
+    tm.assert_frame_equal(result, expected)
+
+    # consider everything
+    df2 = df.loc[:, ["AAA", "B", "C"]]
+
+    result = df2.drop_duplicates()
+    # in this case only
+    expected = df2.drop_duplicates(["AAA", "B"])
+    tm.assert_frame_equal(result, expected)
+
+    result = df2.drop_duplicates(keep="last")
+    expected = df2.drop_duplicates(["AAA", "B"], keep="last")
+    tm.assert_frame_equal(result, expected)
+
+    result = df2.drop_duplicates(keep=False)
+    expected = df2.drop_duplicates(["AAA", "B"], keep=False)
+    tm.assert_frame_equal(result, expected)
+
+    # integers
+    result = df.drop_duplicates("C")
+    expected = df.iloc[[0, 2]]
+    tm.assert_frame_equal(result, expected)
+    result = df.drop_duplicates("C", keep="last")
+    expected = df.iloc[[-2, -1]]
+    tm.assert_frame_equal(result, expected)
+
+    df["E"] = df["C"].astype("int8")
+    result = df.drop_duplicates("E")
+    expected = df.iloc[[0, 2]]
+    tm.assert_frame_equal(result, expected)
+    result = df.drop_duplicates("E", keep="last")
+    expected = df.iloc[[-2, -1]]
+    tm.assert_frame_equal(result, expected)
+
+    # GH 11376
+    df = DataFrame({"x": [7, 6, 3, 3, 4, 8, 0], "y": [0, 6, 5, 5, 9, 1, 2]})
+    expected = df.loc[df.index != 3]
+    tm.assert_frame_equal(df.drop_duplicates(), expected)
+
+    df = DataFrame([[1, 0], [0, 2]])
+    tm.assert_frame_equal(df.drop_duplicates(), df)
+
+    df = DataFrame([[-2, 0], [0, -4]])
+    tm.assert_frame_equal(df.drop_duplicates(), df)
+
+    x = np.iinfo(np.int64).max / 3 * 2
+    df = DataFrame([[-x, x], [0, x + 4]])
+    tm.assert_frame_equal(df.drop_duplicates(), df)
+
+    df = DataFrame([[-x, x], [x, x + 4]])
+    tm.assert_frame_equal(df.drop_duplicates(), df)
+
+    # GH 11864
+    df = DataFrame([i] * 9 for i in range(16))
+    df = concat([df, DataFrame([[1] + [0] * 8])], ignore_index=True)
+
+    for keep in ["first", "last", False]:
+        assert df.duplicated(keep=keep).sum() == 0
+
+
+def test_drop_duplicates_with_duplicate_column_names():
+    # GH17836
+    df = DataFrame([[1, 2, 5], [3, 4, 6], [3, 4, 7]], columns=["a", "a", "b"])
+
+    result0 = df.drop_duplicates()
+    tm.assert_frame_equal(result0, df)
+
+    result1 = df.drop_duplicates("a")
+    expected1 = df[:2]
+    tm.assert_frame_equal(result1, expected1)
+
+
+def test_drop_duplicates_for_take_all():
+    df = DataFrame(
+        {
+            "AAA": ["foo", "bar", "baz", "bar", "foo", "bar", "qux", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1, 1, 2, 2, 2, 2, 1, 2],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates("AAA")
+    expected = df.iloc[[0, 1, 2, 6]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("AAA", keep="last")
+    expected = df.iloc[[2, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("AAA", keep=False)
+    expected = df.iloc[[2, 6]]
+    tm.assert_frame_equal(result, expected)
+
+    # multiple columns
+    result = df.drop_duplicates(["AAA", "B"])
+    expected = df.iloc[[0, 1, 2, 3, 4, 6]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["AAA", "B"], keep="last")
+    expected = df.iloc[[0, 1, 2, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["AAA", "B"], keep=False)
+    expected = df.iloc[[0, 1, 2, 6]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_drop_duplicates_tuple():
+    df = DataFrame(
+        {
+            ("AA", "AB"): ["foo", "bar", "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1, 1, 2, 2, 2, 2, 1, 2],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates(("AA", "AB"))
+    expected = df[:2]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(("AA", "AB"), keep="last")
+    expected = df.loc[[6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(("AA", "AB"), keep=False)
+    expected = df.loc[[]]  # empty df
+    assert len(result) == 0
+    tm.assert_frame_equal(result, expected)
+
+    # multi column
+    expected = df.loc[[0, 1, 2, 3]]
+    result = df.drop_duplicates((("AA", "AB"), "B"))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df",
+    [
+        DataFrame(),
+        DataFrame(columns=[]),
+        DataFrame(columns=["A", "B", "C"]),
+        DataFrame(index=[]),
+        DataFrame(index=["A", "B", "C"]),
+    ],
+)
+def test_drop_duplicates_empty(df):
+    # GH 20516
+    result = df.drop_duplicates()
+    tm.assert_frame_equal(result, df)
+
+    result = df.copy()
+    result.drop_duplicates(inplace=True)
+    tm.assert_frame_equal(result, df)
+
+
+def test_drop_duplicates_NA():
+    # none
+    df = DataFrame(
+        {
+            "A": [None, None, "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1.0, np.nan, np.nan, np.nan, 1.0, 1.0, 1, 1.0],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates("A")
+    expected = df.loc[[0, 2, 3]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("A", keep="last")
+    expected = df.loc[[1, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("A", keep=False)
+    expected = df.loc[[]]  # empty df
+    tm.assert_frame_equal(result, expected)
+    assert len(result) == 0
+
+    # multi column
+    result = df.drop_duplicates(["A", "B"])
+    expected = df.loc[[0, 2, 3, 6]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["A", "B"], keep="last")
+    expected = df.loc[[1, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["A", "B"], keep=False)
+    expected = df.loc[[6]]
+    tm.assert_frame_equal(result, expected)
+
+    # nan
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1.0, np.nan, np.nan, np.nan, 1.0, 1.0, 1, 1.0],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates("C")
+    expected = df[:2]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("C", keep="last")
+    expected = df.loc[[3, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("C", keep=False)
+    expected = df.loc[[]]  # empty df
+    tm.assert_frame_equal(result, expected)
+    assert len(result) == 0
+
+    # multi column
+    result = df.drop_duplicates(["C", "B"])
+    expected = df.loc[[0, 1, 2, 4]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["C", "B"], keep="last")
+    expected = df.loc[[1, 3, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates(["C", "B"], keep=False)
+    expected = df.loc[[1]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_drop_duplicates_NA_for_take_all():
+    # none
+    df = DataFrame(
+        {
+            "A": [None, None, "foo", "bar", "foo", "baz", "bar", "qux"],
+            "C": [1.0, np.nan, np.nan, np.nan, 1.0, 2.0, 3, 1.0],
+        }
+    )
+
+    # single column
+    result = df.drop_duplicates("A")
+    expected = df.iloc[[0, 2, 3, 5, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("A", keep="last")
+    expected = df.iloc[[1, 4, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("A", keep=False)
+    expected = df.iloc[[5, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    # nan
+
+    # single column
+    result = df.drop_duplicates("C")
+    expected = df.iloc[[0, 1, 5, 6]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("C", keep="last")
+    expected = df.iloc[[3, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates("C", keep=False)
+    expected = df.iloc[[5, 6]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_drop_duplicates_inplace():
+    orig = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1, 1, 2, 2, 2, 2, 1, 2],
+            "D": range(8),
+        }
+    )
+    # single column
+    df = orig.copy()
+    return_value = df.drop_duplicates("A", inplace=True)
+    expected = orig[:2]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df = orig.copy()
+    return_value = df.drop_duplicates("A", keep="last", inplace=True)
+    expected = orig.loc[[6, 7]]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df = orig.copy()
+    return_value = df.drop_duplicates("A", keep=False, inplace=True)
+    expected = orig.loc[[]]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert len(df) == 0
+    assert return_value is None
+
+    # multi column
+    df = orig.copy()
+    return_value = df.drop_duplicates(["A", "B"], inplace=True)
+    expected = orig.loc[[0, 1, 2, 3]]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df = orig.copy()
+    return_value = df.drop_duplicates(["A", "B"], keep="last", inplace=True)
+    expected = orig.loc[[0, 5, 6, 7]]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df = orig.copy()
+    return_value = df.drop_duplicates(["A", "B"], keep=False, inplace=True)
+    expected = orig.loc[[0]]
+    result = df
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    # consider everything
+    orig2 = orig.loc[:, ["A", "B", "C"]].copy()
+
+    df2 = orig2.copy()
+    return_value = df2.drop_duplicates(inplace=True)
+    # in this case only
+    expected = orig2.drop_duplicates(["A", "B"])
+    result = df2
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df2 = orig2.copy()
+    return_value = df2.drop_duplicates(keep="last", inplace=True)
+    expected = orig2.drop_duplicates(["A", "B"], keep="last")
+    result = df2
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+    df2 = orig2.copy()
+    return_value = df2.drop_duplicates(keep=False, inplace=True)
+    expected = orig2.drop_duplicates(["A", "B"], keep=False)
+    result = df2
+    tm.assert_frame_equal(result, expected)
+    assert return_value is None
+
+
+@pytest.mark.parametrize("inplace", [True, False])
+@pytest.mark.parametrize(
+    "origin_dict, output_dict, ignore_index, output_index",
+    [
+        ({"A": [2, 2, 3]}, {"A": [2, 3]}, True, range(2)),
+        ({"A": [2, 2, 3]}, {"A": [2, 3]}, False, range(0, 4, 2)),
+        ({"A": [2, 2, 3], "B": [2, 2, 4]}, {"A": [2, 3], "B": [2, 4]}, True, range(2)),
+        (
+            {"A": [2, 2, 3], "B": [2, 2, 4]},
+            {"A": [2, 3], "B": [2, 4]},
+            False,
+            range(0, 4, 2),
+        ),
+    ],
+)
+def test_drop_duplicates_ignore_index(
+    inplace, origin_dict, output_dict, ignore_index, output_index
+):
+    # GH 30114
+    df = DataFrame(origin_dict)
+    expected = DataFrame(output_dict, index=output_index)
+
+    if inplace:
+        result_df = df.copy()
+        result_df.drop_duplicates(ignore_index=ignore_index, inplace=inplace)
+    else:
+        result_df = df.drop_duplicates(ignore_index=ignore_index, inplace=inplace)
+
+    tm.assert_frame_equal(result_df, expected)
+    tm.assert_frame_equal(df, DataFrame(origin_dict))
+
+
+def test_drop_duplicates_null_in_object_column(nulls_fixture):
+    # https://github.com/pandas-dev/pandas/issues/32992
+    df = DataFrame([[1, nulls_fixture], [2, "a"]], dtype=object)
+    result = df.drop_duplicates()
+    tm.assert_frame_equal(result, df)
+
+
+def test_drop_duplicates_series_vs_dataframe(keep):
+    # GH#14192
+    df = DataFrame(
+        {
+            "a": [1, 1, 1, "one", "one"],
+            "b": [2, 2, np.nan, np.nan, np.nan],
+            "c": [3, 3, np.nan, np.nan, "three"],
+            "d": [1, 2, 3, 4, 4],
+            "e": [
+                datetime(2015, 1, 1),
+                datetime(2015, 1, 1),
+                datetime(2015, 2, 1),
+                NaT,
+                NaT,
+            ],
+        }
+    )
+    for column in df.columns:
+        dropped_frame = df[[column]].drop_duplicates(keep=keep)
+        dropped_series = df[column].drop_duplicates(keep=keep)
+        tm.assert_frame_equal(dropped_frame, dropped_series.to_frame())
+
+
+@pytest.mark.parametrize("arg", [[1], 1, "True", [], 0])
+def test_drop_duplicates_non_boolean_ignore_index(arg):
+    # GH#38274
+    df = DataFrame({"a": [1, 2, 1, 3]})
+    msg = '^For argument "ignore_index" expected type bool, received type .*.$'
+    with pytest.raises(ValueError, match=msg):
+        df.drop_duplicates(ignore_index=arg)
+
+
+def test_drop_duplicates_set():
+    # GH#59237
+    df = DataFrame(
+        {
+            "AAA": ["foo", "bar", "foo", "bar", "foo", "bar", "bar", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": [1, 1, 2, 2, 2, 2, 1, 2],
+            "D": range(8),
+        }
+    )
+    # single column
+    result = df.drop_duplicates({"AAA"})
+    expected = df[:2]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates({"AAA"}, keep="last")
+    expected = df.loc[[6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates({"AAA"}, keep=False)
+    expected = df.loc[[]]
+    tm.assert_frame_equal(result, expected)
+    assert len(result) == 0
+
+    # multi column
+    expected = df.loc[[0, 1, 2, 3]]
+    result = df.drop_duplicates({"AAA", "B"})
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates({"AAA", "B"}, keep="last")
+    expected = df.loc[[0, 5, 6, 7]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.drop_duplicates({"AAA", "B"}, keep=False)
+    expected = df.loc[[0]]
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_droplevel.py b/pandas/pandas/tests/frame/methods/test_droplevel.py
new file mode 100644
index 0000000000000..e1302d4b73f2b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_droplevel.py
@@ -0,0 +1,36 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+class TestDropLevel:
+    def test_droplevel(self, frame_or_series):
+        # GH#20342
+        cols = MultiIndex.from_tuples(
+            [("c", "e"), ("d", "f")], names=["level_1", "level_2"]
+        )
+        mi = MultiIndex.from_tuples([(1, 2), (5, 6), (9, 10)], names=["a", "b"])
+        df = DataFrame([[3, 4], [7, 8], [11, 12]], index=mi, columns=cols)
+        if frame_or_series is not DataFrame:
+            df = df.iloc[:, 0]
+
+        # test that dropping of a level in index works
+        expected = df.reset_index("a", drop=True)
+        result = df.droplevel("a", axis="index")
+        tm.assert_equal(result, expected)
+
+        if frame_or_series is DataFrame:
+            # test that dropping of a level in columns works
+            expected = df.copy()
+            expected.columns = Index(["c", "d"], name="level_1")
+            result = df.droplevel("level_2", axis="columns")
+            tm.assert_equal(result, expected)
+        else:
+            # test that droplevel raises ValueError on axis != 0
+            with pytest.raises(ValueError, match="No axis named columns"):
+                df.droplevel(1, axis="columns")
diff --git a/pandas/pandas/tests/frame/methods/test_dropna.py b/pandas/pandas/tests/frame/methods/test_dropna.py
new file mode 100644
index 0000000000000..11893d7fac1a4
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_dropna.py
@@ -0,0 +1,285 @@
+import datetime
+
+import dateutil
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameMissingData:
+    def test_dropEmptyRows(self, float_frame):
+        N = len(float_frame.index)
+        mat = np.random.default_rng(2).standard_normal(N)
+        mat[:5] = np.nan
+
+        frame = DataFrame({"foo": mat}, index=float_frame.index)
+        original = Series(mat, index=float_frame.index, name="foo")
+        expected = original.dropna()
+        inplace_frame1, inplace_frame2 = frame.copy(), frame.copy()
+
+        smaller_frame = frame.dropna(how="all")
+        # check that original was preserved
+        tm.assert_series_equal(frame["foo"], original)
+        return_value = inplace_frame1.dropna(how="all", inplace=True)
+        tm.assert_series_equal(smaller_frame["foo"], expected)
+        tm.assert_series_equal(inplace_frame1["foo"], expected)
+        assert return_value is None
+
+        smaller_frame = frame.dropna(how="all", subset=["foo"])
+        return_value = inplace_frame2.dropna(how="all", subset=["foo"], inplace=True)
+        tm.assert_series_equal(smaller_frame["foo"], expected)
+        tm.assert_series_equal(inplace_frame2["foo"], expected)
+        assert return_value is None
+
+    def test_dropIncompleteRows(self, float_frame):
+        N = len(float_frame.index)
+        mat = np.random.default_rng(2).standard_normal(N)
+        mat[:5] = np.nan
+
+        frame = DataFrame({"foo": mat}, index=float_frame.index)
+        frame["bar"] = 5
+        original = Series(mat, index=float_frame.index, name="foo")
+        inp_frame1, inp_frame2 = frame.copy(), frame.copy()
+
+        smaller_frame = frame.dropna()
+        tm.assert_series_equal(frame["foo"], original)
+        return_value = inp_frame1.dropna(inplace=True)
+
+        exp = Series(mat[5:], index=float_frame.index[5:], name="foo")
+        tm.assert_series_equal(smaller_frame["foo"], exp)
+        tm.assert_series_equal(inp_frame1["foo"], exp)
+        assert return_value is None
+
+        samesize_frame = frame.dropna(subset=["bar"])
+        tm.assert_series_equal(frame["foo"], original)
+        assert (frame["bar"] == 5).all()
+        return_value = inp_frame2.dropna(subset=["bar"], inplace=True)
+        tm.assert_index_equal(samesize_frame.index, float_frame.index)
+        tm.assert_index_equal(inp_frame2.index, float_frame.index)
+        assert return_value is None
+
+    def test_dropna(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((6, 4)))
+        df.iloc[:2, 2] = np.nan
+
+        dropped = df.dropna(axis=1)
+        expected = df.loc[:, [0, 1, 3]]
+        inp = df.copy()
+        return_value = inp.dropna(axis=1, inplace=True)
+        tm.assert_frame_equal(dropped, expected)
+        tm.assert_frame_equal(inp, expected)
+        assert return_value is None
+
+        dropped = df.dropna(axis=0)
+        expected = df.loc[list(range(2, 6))]
+        inp = df.copy()
+        return_value = inp.dropna(axis=0, inplace=True)
+        tm.assert_frame_equal(dropped, expected)
+        tm.assert_frame_equal(inp, expected)
+        assert return_value is None
+
+        # threshold
+        dropped = df.dropna(axis=1, thresh=5)
+        expected = df.loc[:, [0, 1, 3]]
+        inp = df.copy()
+        return_value = inp.dropna(axis=1, thresh=5, inplace=True)
+        tm.assert_frame_equal(dropped, expected)
+        tm.assert_frame_equal(inp, expected)
+        assert return_value is None
+
+        dropped = df.dropna(axis=0, thresh=4)
+        expected = df.loc[range(2, 6)]
+        inp = df.copy()
+        return_value = inp.dropna(axis=0, thresh=4, inplace=True)
+        tm.assert_frame_equal(dropped, expected)
+        tm.assert_frame_equal(inp, expected)
+        assert return_value is None
+
+        dropped = df.dropna(axis=1, thresh=4)
+        tm.assert_frame_equal(dropped, df)
+
+        dropped = df.dropna(axis=1, thresh=3)
+        tm.assert_frame_equal(dropped, df)
+
+        # subset
+        dropped = df.dropna(axis=0, subset=[0, 1, 3])
+        inp = df.copy()
+        return_value = inp.dropna(axis=0, subset=[0, 1, 3], inplace=True)
+        tm.assert_frame_equal(dropped, df)
+        tm.assert_frame_equal(inp, df)
+        assert return_value is None
+
+        # all
+        dropped = df.dropna(axis=1, how="all")
+        tm.assert_frame_equal(dropped, df)
+
+        df[2] = np.nan
+        dropped = df.dropna(axis=1, how="all")
+        expected = df.loc[:, [0, 1, 3]]
+        tm.assert_frame_equal(dropped, expected)
+
+        # bad input
+        msg = "No axis named 3 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.dropna(axis=3)
+
+    def test_drop_and_dropna_caching(self):
+        # tst that cacher updates
+        original = Series([1, 2, np.nan], name="A")
+        expected = Series([1, 2], dtype=original.dtype, name="A")
+        df = DataFrame({"A": original.values.copy()})
+        df2 = df.copy()
+        df["A"].dropna()
+        tm.assert_series_equal(df["A"], original)
+
+        ser = df["A"]
+        return_value = ser.dropna(inplace=True)
+        tm.assert_series_equal(ser, expected)
+        tm.assert_series_equal(df["A"], original)
+        assert return_value is None
+
+        df2["A"].drop([1])
+        tm.assert_series_equal(df2["A"], original)
+
+        ser = df2["A"]
+        return_value = ser.drop([1], inplace=True)
+        tm.assert_series_equal(ser, original.drop([1]))
+        tm.assert_series_equal(df2["A"], original)
+        assert return_value is None
+
+    def test_dropna_corner(self, float_frame):
+        # bad input
+        msg = "invalid how option: foo"
+        with pytest.raises(ValueError, match=msg):
+            float_frame.dropna(how="foo")
+        # non-existent column - 8303
+        with pytest.raises(KeyError, match=r"^\['X'\]$"):
+            float_frame.dropna(subset=["A", "X"])
+
+    def test_dropna_multiple_axes(self):
+        df = DataFrame(
+            [
+                [1, np.nan, 2, 3],
+                [4, np.nan, 5, 6],
+                [np.nan, np.nan, np.nan, np.nan],
+                [7, np.nan, 8, 9],
+            ]
+        )
+
+        # GH20987
+        with pytest.raises(TypeError, match="supplying multiple axes"):
+            df.dropna(how="all", axis=[0, 1])
+        with pytest.raises(TypeError, match="supplying multiple axes"):
+            df.dropna(how="all", axis=(0, 1))
+
+        inp = df.copy()
+        with pytest.raises(TypeError, match="supplying multiple axes"):
+            inp.dropna(how="all", axis=(0, 1), inplace=True)
+
+    def test_dropna_tz_aware_datetime(self):
+        # GH13407
+        df = DataFrame()
+        dt1 = datetime.datetime(2015, 1, 1, tzinfo=dateutil.tz.tzutc())
+        dt2 = datetime.datetime(2015, 2, 2, tzinfo=dateutil.tz.tzutc())
+        df["Time"] = [dt1]
+        result = df.dropna(axis=0)
+        expected = DataFrame({"Time": [dt1]})
+        tm.assert_frame_equal(result, expected)
+
+        # Ex2
+        df = DataFrame({"Time": [dt1, None, np.nan, dt2]})
+        result = df.dropna(axis=0)
+        expected = DataFrame([dt1, dt2], columns=["Time"], index=range(0, 6, 3))
+        tm.assert_frame_equal(result, expected)
+
+    def test_dropna_categorical_interval_index(self):
+        # GH 25087
+        ii = pd.IntervalIndex.from_breaks([0, 2.78, 3.14, 6.28])
+        ci = pd.CategoricalIndex(ii)
+        df = DataFrame({"A": list("abc")}, index=ci)
+
+        expected = df
+        result = df.dropna()
+        tm.assert_frame_equal(result, expected)
+
+    def test_dropna_with_duplicate_columns(self):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(5),
+                "B": np.random.default_rng(2).standard_normal(5),
+                "C": np.random.default_rng(2).standard_normal(5),
+                "D": ["a", "b", "c", "d", "e"],
+            }
+        )
+        df.iloc[2, [0, 1, 2]] = np.nan
+        df.iloc[0, 0] = np.nan
+        df.iloc[1, 1] = np.nan
+        df.iloc[:, 3] = np.nan
+        expected = df.dropna(subset=["A", "B", "C"], how="all")
+        expected.columns = ["A", "A", "B", "C"]
+
+        df.columns = ["A", "A", "B", "C"]
+
+        result = df.dropna(subset=["A", "C"], how="all")
+        tm.assert_frame_equal(result, expected)
+
+    def test_set_single_column_subset(self):
+        # GH 41021
+        df = DataFrame({"A": [1, 2, 3], "B": list("abc"), "C": [4, np.nan, 5]})
+        expected = DataFrame(
+            {"A": [1, 3], "B": list("ac"), "C": [4.0, 5.0]}, index=range(0, 4, 2)
+        )
+        result = df.dropna(subset="C")
+        tm.assert_frame_equal(result, expected)
+
+    def test_single_column_not_present_in_axis(self):
+        # GH 41021
+        df = DataFrame({"A": [1, 2, 3]})
+
+        # Column not present
+        with pytest.raises(KeyError, match="['D']"):
+            df.dropna(subset="D", axis=0)
+
+    def test_subset_is_nparray(self):
+        # GH 41021
+        df = DataFrame({"A": [1, 2, np.nan], "B": list("abc"), "C": [4, np.nan, 5]})
+        expected = DataFrame({"A": [1.0], "B": ["a"], "C": [4.0]})
+        result = df.dropna(subset=np.array(["A", "C"]))
+        tm.assert_frame_equal(result, expected)
+
+    def test_no_nans_in_frame(self, axis):
+        # GH#41965
+        df = DataFrame([[1, 2], [3, 4]], columns=pd.RangeIndex(0, 2))
+        expected = df.copy()
+        result = df.dropna(axis=axis)
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_how_thresh_param_incompatible(self):
+        # GH46575
+        df = DataFrame([1, 2, pd.NA])
+        msg = "You cannot set both the how and thresh arguments at the same time"
+        with pytest.raises(TypeError, match=msg):
+            df.dropna(how="all", thresh=2)
+
+        with pytest.raises(TypeError, match=msg):
+            df.dropna(how="any", thresh=2)
+
+        with pytest.raises(TypeError, match=msg):
+            df.dropna(how=None, thresh=None)
+
+    @pytest.mark.parametrize("val", [1, 1.5])
+    def test_dropna_ignore_index(self, val):
+        # GH#31725
+        df = DataFrame({"a": [1, 2, val]}, index=[3, 2, 1])
+        result = df.dropna(ignore_index=True)
+        expected = DataFrame({"a": [1, 2, val]})
+        tm.assert_frame_equal(result, expected)
+
+        df.dropna(ignore_index=True, inplace=True)
+        tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_dtypes.py b/pandas/pandas/tests/frame/methods/test_dtypes.py
new file mode 100644
index 0000000000000..bf01ec73cf72b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_dtypes.py
@@ -0,0 +1,141 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameDataTypes:
+    def test_empty_frame_dtypes(self):
+        empty_df = DataFrame()
+        tm.assert_series_equal(empty_df.dtypes, Series(dtype=object))
+
+        nocols_df = DataFrame(index=[1, 2, 3])
+        tm.assert_series_equal(nocols_df.dtypes, Series(dtype=object))
+
+        norows_df = DataFrame(columns=list("abc"))
+        tm.assert_series_equal(norows_df.dtypes, Series(object, index=list("abc")))
+
+        norows_int_df = DataFrame(columns=list("abc")).astype(np.int32)
+        tm.assert_series_equal(
+            norows_int_df.dtypes, Series(np.dtype("int32"), index=list("abc"))
+        )
+
+        df = DataFrame({"a": 1, "b": True, "c": 1.0}, index=[1, 2, 3])
+        ex_dtypes = Series({"a": np.int64, "b": np.bool_, "c": np.float64})
+        tm.assert_series_equal(df.dtypes, ex_dtypes)
+
+        # same but for empty slice of df
+        tm.assert_series_equal(df[:0].dtypes, ex_dtypes)
+
+    def test_datetime_with_tz_dtypes(self):
+        tzframe = DataFrame(
+            {
+                "A": date_range("20130101", periods=3),
+                "B": date_range("20130101", periods=3, tz="US/Eastern"),
+                "C": date_range("20130101", periods=3, tz="CET"),
+            }
+        )
+        tzframe.iloc[1, 1] = pd.NaT
+        tzframe.iloc[1, 2] = pd.NaT
+        result = tzframe.dtypes.sort_index()
+        expected = Series(
+            [
+                np.dtype("datetime64[ns]"),
+                DatetimeTZDtype("ns", "US/Eastern"),
+                DatetimeTZDtype("ns", "CET"),
+            ],
+            ["A", "B", "C"],
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_dtypes_are_correct_after_column_slice(self):
+        # GH6525
+        df = DataFrame(index=range(5), columns=list("abc"), dtype=np.float64)
+        tm.assert_series_equal(
+            df.dtypes,
+            Series({"a": np.float64, "b": np.float64, "c": np.float64}),
+        )
+        tm.assert_series_equal(df.iloc[:, 2:].dtypes, Series({"c": np.float64}))
+        tm.assert_series_equal(
+            df.dtypes,
+            Series({"a": np.float64, "b": np.float64, "c": np.float64}),
+        )
+
+    @pytest.mark.parametrize(
+        "data",
+        [pd.NA, True],
+    )
+    def test_dtypes_are_correct_after_groupby_last(self, data):
+        # GH46409
+        df = DataFrame(
+            {"id": [1, 2, 3, 4], "test": [True, pd.NA, data, False]}
+        ).convert_dtypes()
+        result = df.groupby("id").last().test
+        expected = df.set_index("id").test
+        assert result.dtype == pd.BooleanDtype()
+        tm.assert_series_equal(expected, result)
+
+    def test_dtypes_gh8722(self, float_string_frame):
+        float_string_frame["bool"] = float_string_frame["A"] > 0
+        result = float_string_frame.dtypes
+        expected = Series(
+            {k: v.dtype for k, v in float_string_frame.items()}, index=result.index
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dtypes_timedeltas(self):
+        df = DataFrame(
+            {
+                "A": Series(date_range("2012-1-1", periods=3, freq="D")),
+                "B": Series([timedelta(days=i) for i in range(3)]),
+            }
+        )
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("datetime64[ns]"), np.dtype("timedelta64[ns]")], index=list("AB")
+        )
+        tm.assert_series_equal(result, expected)
+
+        df["C"] = df["A"] + df["B"]
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("datetime64[ns]"),
+                np.dtype("timedelta64[ns]"),
+                np.dtype("datetime64[ns]"),
+            ],
+            index=list("ABC"),
+        )
+        tm.assert_series_equal(result, expected)
+
+        # mixed int types
+        df["D"] = 1
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("datetime64[ns]"),
+                np.dtype("timedelta64[ns]"),
+                np.dtype("datetime64[ns]"),
+                np.dtype("int64"),
+            ],
+            index=list("ABCD"),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_frame_apply_np_array_return_type(self, using_infer_string):
+        # GH 35517
+        df = DataFrame([["foo"]])
+        result = df.apply(lambda col: np.array("bar"))
+        expected = Series(np.array("bar"))
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_duplicated.py b/pandas/pandas/tests/frame/methods/test_duplicated.py
new file mode 100644
index 0000000000000..6052b61ea8db5
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_duplicated.py
@@ -0,0 +1,117 @@
+import re
+import sys
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("subset", ["a", ["a"], ["a", "B"]])
+def test_duplicated_with_misspelled_column_name(subset):
+    # GH 19730
+    df = DataFrame({"A": [0, 0, 1], "B": [0, 0, 1], "C": [0, 0, 1]})
+    msg = re.escape("Index(['a'], dtype=")
+
+    with pytest.raises(KeyError, match=msg):
+        df.duplicated(subset)
+
+
+def test_duplicated_implemented_no_recursion():
+    # gh-21524
+    # Ensure duplicated isn't implemented using recursion that
+    # can fail on wide frames
+    df = DataFrame(np.random.default_rng(2).integers(0, 1000, (10, 1000)))
+    rec_limit = sys.getrecursionlimit()
+    try:
+        sys.setrecursionlimit(100)
+        result = df.duplicated()
+    finally:
+        sys.setrecursionlimit(rec_limit)
+
+    # Then duplicates produce the bool Series as a result and don't fail during
+    # calculation. Actual values doesn't matter here, though usually it's all
+    # False in this case
+    assert isinstance(result, Series)
+    assert result.dtype == np.bool_
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", Series([False, False, True, False, True])),
+        ("last", Series([True, True, False, False, False])),
+        (False, Series([True, True, True, False, True])),
+    ],
+)
+def test_duplicated_keep(keep, expected):
+    df = DataFrame({"A": [0, 1, 1, 2, 0], "B": ["a", "b", "b", "c", "a"]})
+
+    result = df.duplicated(keep=keep)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.xfail(reason="GH#21720; nan/None falsely considered equal")
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", Series([False, False, True, False, True])),
+        ("last", Series([True, True, False, False, False])),
+        (False, Series([True, True, True, False, True])),
+    ],
+)
+def test_duplicated_nan_none(keep, expected):
+    df = DataFrame({"C": [np.nan, 3, 3, None, np.nan], "x": 1}, dtype=object)
+
+    result = df.duplicated(keep=keep)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("subset", [None, ["A", "B"], "A"])
+def test_duplicated_subset(subset, keep):
+    df = DataFrame(
+        {
+            "A": [0, 1, 1, 2, 0],
+            "B": ["a", "b", "b", "c", "a"],
+            "C": [np.nan, 3, 3, None, np.nan],
+        }
+    )
+
+    if subset is None:
+        subset = list(df.columns)
+    elif isinstance(subset, str):
+        # need to have a DataFrame, not a Series
+        # -> select columns with singleton list, not string
+        subset = [subset]
+
+    expected = df[subset].duplicated(keep=keep)
+    result = df.duplicated(keep=keep, subset=subset)
+    tm.assert_series_equal(result, expected)
+
+
+def test_duplicated_on_empty_frame():
+    # GH 25184
+
+    df = DataFrame(columns=["a", "b"])
+    dupes = df.duplicated("a")
+
+    result = df[dupes]
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_datetime64_duplicated():
+    dates = date_range("2010-07-01", end="2010-08-05")
+
+    tst = DataFrame({"symbol": "AAA", "date": dates})
+    result = tst.duplicated(["date", "symbol"])
+    assert (-result).all()
+
+    tst = DataFrame({"date": dates})
+    result = tst.date.duplicated()
+    assert (-result).all()
diff --git a/pandas/pandas/tests/frame/methods/test_equals.py b/pandas/pandas/tests/frame/methods/test_equals.py
new file mode 100644
index 0000000000000..88b3fec02182b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_equals.py
@@ -0,0 +1,85 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestEquals:
+    def test_dataframe_not_equal(self):
+        # see GH#28839
+        df1 = DataFrame({"a": [1, 2], "b": ["s", "d"]})
+        df2 = DataFrame({"a": ["s", "d"], "b": [1, 2]})
+        assert df1.equals(df2) is False
+
+    def test_equals_different_blocks(self, using_infer_string):
+        # GH#9330
+        df0 = DataFrame({"A": ["x", "y"], "B": [1, 2], "C": ["w", "z"]})
+        df1 = df0.reset_index()[["A", "B", "C"]]
+        if not using_infer_string:
+            # this assert verifies that the above operations have
+            # induced a block rearrangement
+            assert df0._mgr.blocks[0].dtype != df1._mgr.blocks[0].dtype
+
+        # do the real tests
+        tm.assert_frame_equal(df0, df1)
+        assert df0.equals(df1)
+        assert df1.equals(df0)
+
+    def test_equals(self):
+        # Add object dtype column with nans
+        index = np.random.default_rng(2).random(10)
+        df1 = DataFrame(
+            np.random.default_rng(2).random(10), index=index, columns=["floats"]
+        )
+        df1["text"] = "the sky is so blue. we could use more chocolate.".split()
+        df1["start"] = date_range("2000-1-1", periods=10, freq="min")
+        df1["end"] = date_range("2000-1-1", periods=10, freq="D")
+        df1["diff"] = df1["end"] - df1["start"]
+        # Explicitly cast to object, to avoid implicit cast when setting np.nan
+        df1["bool"] = (np.arange(10) % 3 == 0).astype(object)
+        df1.loc[::2] = np.nan
+        df2 = df1.copy()
+        assert df1["text"].equals(df2["text"])
+        assert df1["start"].equals(df2["start"])
+        assert df1["end"].equals(df2["end"])
+        assert df1["diff"].equals(df2["diff"])
+        assert df1["bool"].equals(df2["bool"])
+        assert df1.equals(df2)
+        assert not df1.equals(object)
+
+        # different dtype
+        different = df1.copy()
+        different["floats"] = different["floats"].astype("float32")
+        assert not df1.equals(different)
+
+        # different index
+        different_index = -index
+        different = df2.set_index(different_index)
+        assert not df1.equals(different)
+
+        # different columns
+        different = df2.copy()
+        different.columns = df2.columns[::-1]
+        assert not df1.equals(different)
+
+        # DatetimeIndex
+        index = date_range("2000-1-1", periods=10, freq="min")
+        df1 = df1.set_index(index)
+        df2 = df1.copy()
+        assert df1.equals(df2)
+
+        # MultiIndex
+        df3 = df1.set_index(["text"], append=True)
+        df2 = df1.set_index(["text"], append=True)
+        assert df3.equals(df2)
+
+        df2 = df1.set_index(["floats"], append=True)
+        assert not df3.equals(df2)
+
+        # NaN in index
+        df3 = df1.set_index(["floats"], append=True)
+        df2 = df1.set_index(["floats"], append=True)
+        assert df3.equals(df2)
diff --git a/pandas/pandas/tests/frame/methods/test_explode.py b/pandas/pandas/tests/frame/methods/test_explode.py
new file mode 100644
index 0000000000000..876ad5539d603
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_explode.py
@@ -0,0 +1,299 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_error():
+    df = pd.DataFrame(
+        {"A": pd.Series([[0, 1, 2], np.nan, [], (3, 4)], index=list("abcd")), "B": 1}
+    )
+    with pytest.raises(
+        ValueError, match="column must be a scalar, tuple, or list thereof"
+    ):
+        df.explode([list("AA")])
+
+    with pytest.raises(ValueError, match="column must be unique"):
+        df.explode(list("AA"))
+
+    df.columns = list("AA")
+    with pytest.raises(
+        ValueError,
+        match=re.escape("DataFrame columns must be unique. Duplicate columns: ['A']"),
+    ):
+        df.explode("A")
+
+
+@pytest.mark.parametrize(
+    "input_subset, error_message",
+    [
+        (
+            list("AC"),
+            "columns must have matching element counts",
+        ),
+        (
+            [],
+            "column must be nonempty",
+        ),
+    ],
+)
+def test_error_multi_columns(input_subset, error_message):
+    # GH 39240
+    df = pd.DataFrame(
+        {
+            "A": [[0, 1, 2], np.nan, [], (3, 4)],
+            "B": 1,
+            "C": [["a", "b", "c"], "foo", [], ["d", "e", "f"]],
+        },
+        index=list("abcd"),
+    )
+    with pytest.raises(ValueError, match=error_message):
+        df.explode(input_subset)
+
+
+@pytest.mark.parametrize(
+    "scalar",
+    ["a", 0, 1.5, pd.Timedelta("1 days"), pd.Timestamp("2019-12-31")],
+)
+def test_basic(scalar):
+    df = pd.DataFrame(
+        {scalar: pd.Series([[0, 1, 2], np.nan, [], (3, 4)], index=list("abcd")), "B": 1}
+    )
+    result = df.explode(scalar)
+    expected = pd.DataFrame(
+        {
+            scalar: pd.Series(
+                [0, 1, 2, np.nan, np.nan, 3, 4], index=list("aaabcdd"), dtype=object
+            ),
+            "B": 1,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multi_index_rows():
+    df = pd.DataFrame(
+        {"A": np.array([[0, 1, 2], np.nan, [], (3, 4)], dtype=object), "B": 1},
+        index=pd.MultiIndex.from_tuples([("a", 1), ("a", 2), ("b", 1), ("b", 2)]),
+    )
+
+    result = df.explode("A")
+    expected = pd.DataFrame(
+        {
+            "A": pd.Series(
+                [0, 1, 2, np.nan, np.nan, 3, 4],
+                index=pd.MultiIndex.from_tuples(
+                    [
+                        ("a", 1),
+                        ("a", 1),
+                        ("a", 1),
+                        ("a", 2),
+                        ("b", 1),
+                        ("b", 2),
+                        ("b", 2),
+                    ]
+                ),
+                dtype=object,
+            ),
+            "B": 1,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multi_index_columns():
+    df = pd.DataFrame(
+        {("A", 1): np.array([[0, 1, 2], np.nan, [], (3, 4)], dtype=object), ("A", 2): 1}
+    )
+
+    result = df.explode(("A", 1))
+    expected = pd.DataFrame(
+        {
+            ("A", 1): pd.Series(
+                [0, 1, 2, np.nan, np.nan, 3, 4],
+                index=pd.Index([0, 0, 0, 1, 2, 3, 3]),
+                dtype=object,
+            ),
+            ("A", 2): 1,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecase():
+    # explode a single column
+    # gh-10511
+    df = pd.DataFrame(
+        [[11, range(5), 10], [22, range(3), 20]], columns=list("ABC")
+    ).set_index("C")
+    result = df.explode("B")
+
+    expected = pd.DataFrame(
+        {
+            "A": [11, 11, 11, 11, 11, 22, 22, 22],
+            "B": np.array([0, 1, 2, 3, 4, 0, 1, 2], dtype=object),
+            "C": [10, 10, 10, 10, 10, 20, 20, 20],
+        },
+        columns=list("ABC"),
+    ).set_index("C")
+
+    tm.assert_frame_equal(result, expected)
+
+    # gh-8517
+    df = pd.DataFrame(
+        [["2014-01-01", "Alice", "A B"], ["2014-01-02", "Bob", "C D"]],
+        columns=["dt", "name", "text"],
+    )
+    result = df.assign(text=df.text.str.split(" ")).explode("text")
+    expected = pd.DataFrame(
+        [
+            ["2014-01-01", "Alice", "A"],
+            ["2014-01-01", "Alice", "B"],
+            ["2014-01-02", "Bob", "C"],
+            ["2014-01-02", "Bob", "D"],
+        ],
+        columns=["dt", "name", "text"],
+        index=[0, 0, 1, 1],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "input_dict, input_index, expected_dict, expected_index",
+    [
+        (
+            {"col1": [[1, 2], [3, 4]], "col2": ["foo", "bar"]},
+            [0, 0],
+            {"col1": [1, 2, 3, 4], "col2": ["foo", "foo", "bar", "bar"]},
+            [0, 0, 0, 0],
+        ),
+        (
+            {"col1": [[1, 2], [3, 4]], "col2": ["foo", "bar"]},
+            pd.Index([0, 0], name="my_index"),
+            {"col1": [1, 2, 3, 4], "col2": ["foo", "foo", "bar", "bar"]},
+            pd.Index([0, 0, 0, 0], name="my_index"),
+        ),
+        (
+            {"col1": [[1, 2], [3, 4]], "col2": ["foo", "bar"]},
+            pd.MultiIndex.from_arrays(
+                [[0, 0], [1, 1]], names=["my_first_index", "my_second_index"]
+            ),
+            {"col1": [1, 2, 3, 4], "col2": ["foo", "foo", "bar", "bar"]},
+            pd.MultiIndex.from_arrays(
+                [[0, 0, 0, 0], [1, 1, 1, 1]],
+                names=["my_first_index", "my_second_index"],
+            ),
+        ),
+        (
+            {"col1": [[1, 2], [3, 4]], "col2": ["foo", "bar"]},
+            pd.MultiIndex.from_arrays([[0, 0], [1, 1]], names=["my_index", None]),
+            {"col1": [1, 2, 3, 4], "col2": ["foo", "foo", "bar", "bar"]},
+            pd.MultiIndex.from_arrays(
+                [[0, 0, 0, 0], [1, 1, 1, 1]], names=["my_index", None]
+            ),
+        ),
+    ],
+)
+def test_duplicate_index(input_dict, input_index, expected_dict, expected_index):
+    # GH 28005
+    df = pd.DataFrame(input_dict, index=input_index, dtype=object)
+    result = df.explode("col1")
+    expected = pd.DataFrame(expected_dict, index=expected_index, dtype=object)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ignore_index():
+    # GH 34932
+    df = pd.DataFrame({"id": range(0, 20, 10), "values": [list("ab"), list("cd")]})
+    result = df.explode("values", ignore_index=True)
+    expected = pd.DataFrame(
+        {"id": [0, 0, 10, 10], "values": list("abcd")}, index=range(4)
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_explode_sets():
+    # https://github.com/pandas-dev/pandas/issues/35614
+    df = pd.DataFrame({"a": [{"x", "y"}], "b": [1]}, index=[1])
+    result = df.explode(column="a").sort_values(by="a")
+    expected = pd.DataFrame({"a": ["x", "y"], "b": [1, 1]}, index=[1, 1])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "input_subset, expected_dict, expected_index",
+    [
+        (
+            list("AC"),
+            {
+                "A": pd.Series(
+                    [0, 1, 2, np.nan, np.nan, 3, 4, np.nan],
+                    index=list("aaabcdde"),
+                    dtype=object,
+                ),
+                "B": 1,
+                "C": ["a", "b", "c", "foo", np.nan, "d", "e", np.nan],
+            },
+            list("aaabcdde"),
+        ),
+        (
+            list("A"),
+            {
+                "A": pd.Series(
+                    [0, 1, 2, np.nan, np.nan, 3, 4, np.nan],
+                    index=list("aaabcdde"),
+                    dtype=object,
+                ),
+                "B": 1,
+                "C": [
+                    ["a", "b", "c"],
+                    ["a", "b", "c"],
+                    ["a", "b", "c"],
+                    "foo",
+                    [],
+                    ["d", "e"],
+                    ["d", "e"],
+                    np.nan,
+                ],
+            },
+            list("aaabcdde"),
+        ),
+    ],
+)
+def test_multi_columns(input_subset, expected_dict, expected_index):
+    # GH 39240
+    df = pd.DataFrame(
+        {
+            "A": [[0, 1, 2], np.nan, [], (3, 4), np.nan],
+            "B": 1,
+            "C": [["a", "b", "c"], "foo", [], ["d", "e"], np.nan],
+        },
+        index=list("abcde"),
+    )
+    result = df.explode(input_subset)
+    expected = pd.DataFrame(expected_dict, expected_index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multi_columns_nan_empty():
+    # GH 46084
+    df = pd.DataFrame(
+        {
+            "A": [[0, 1], [5], [], [2, 3]],
+            "B": [9, 8, 7, 6],
+            "C": [[1, 2], np.nan, [], [3, 4]],
+        }
+    )
+    result = df.explode(["A", "C"])
+    expected = pd.DataFrame(
+        {
+            "A": np.array([0, 1, 5, np.nan, 2, 3], dtype=object),
+            "B": [9, 9, 8, 7, 6, 6],
+            "C": np.array([1, 2, np.nan, np.nan, 3, 4], dtype=object),
+        },
+        index=[0, 0, 1, 2, 3, 3],
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_fillna.py b/pandas/pandas/tests/frame/methods/test_fillna.py
new file mode 100644
index 0000000000000..67d1d45af1cb3
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_fillna.py
@@ -0,0 +1,783 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    TimedeltaIndex,
+    Timestamp,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.tests.frame.common import _check_mixed_float
+
+
+class TestFillNA:
+    def test_fillna_dict_inplace_nonunique_columns(self):
+        df = DataFrame(
+            {"A": [np.nan] * 3, "B": [NaT, Timestamp(1), NaT], "C": [np.nan, "foo", 2]}
+        )
+        df.columns = ["A", "A", "A"]
+        orig = df[:]
+
+        df.fillna({"A": 2}, inplace=True)
+        # The first and third columns can be set inplace, while the second cannot.
+
+        expected = DataFrame(
+            {"A": [2.0] * 3, "B": [2, Timestamp(1), 2], "C": [2, "foo", 2]}
+        )
+        expected.columns = ["A", "A", "A"]
+        tm.assert_frame_equal(df, expected)
+        assert not tm.shares_memory(df.iloc[:, 1], orig.iloc[:, 1])
+
+    def test_fillna_on_column_view(self):
+        # GH#46149 avoid unnecessary copies
+        arr = np.full((40, 50), np.nan)
+        df = DataFrame(arr, copy=False)
+
+        with tm.raises_chained_assignment_error():
+            df[0].fillna(-1, inplace=True)
+        assert np.isnan(arr[:, 0]).all()
+
+        # i.e. we didn't create a new 49-column block
+        assert len(df._mgr.blocks) == 1
+        assert np.shares_memory(df.values, arr)
+
+    def test_fillna_datetime(self, datetime_frame):
+        tf = datetime_frame
+        tf.loc[tf.index[:5], "A"] = np.nan
+        tf.loc[tf.index[-5:], "A"] = np.nan
+
+        zero_filled = datetime_frame.fillna(0)
+        assert (zero_filled.loc[zero_filled.index[:5], "A"] == 0).all()
+
+        padded = datetime_frame.ffill()
+        assert np.isnan(padded.loc[padded.index[:5], "A"]).all()
+
+        msg = r"missing 1 required positional argument: 'value'"
+        with pytest.raises(TypeError, match=msg):
+            datetime_frame.fillna()
+
+    def test_fillna_mixed_type(self, float_string_frame, using_infer_string):
+        mf = float_string_frame
+        mf.loc[mf.index[5:20], "foo"] = np.nan
+        mf.loc[mf.index[-10:], "A"] = np.nan
+
+        result = mf.ffill()
+        assert (
+            result.loc[result.index[-10:], "A"] == result.loc[result.index[-11], "A"]
+        ).all()
+        assert (result.loc[result.index[5:20], "foo"] == "bar").all()
+
+        result = mf.fillna(value=0)
+        assert (result.loc[result.index[-10:], "A"] == 0).all()
+        assert (result.loc[result.index[5:20], "foo"] == 0).all()
+
+    def test_fillna_mixed_float(self, mixed_float_frame):
+        # mixed numeric (but no float16)
+        mf = mixed_float_frame.reindex(columns=["A", "B", "D"])
+        mf.loc[mf.index[-10:], "A"] = np.nan
+        result = mf.fillna(value=0)
+        _check_mixed_float(result, dtype={"C": None})
+        result = mf.ffill()
+        _check_mixed_float(result, dtype={"C": None})
+
+    def test_fillna_different_dtype(self):
+        # with different dtype (GH#3386)
+        df = DataFrame(
+            [["a", "a", np.nan, "a"], ["b", "b", np.nan, "b"], ["c", "c", np.nan, "c"]]
+        )
+
+        result = df.fillna({2: "foo"})
+        expected = DataFrame(
+            [["a", "a", "foo", "a"], ["b", "b", "foo", "b"], ["c", "c", "foo", "c"]]
+        )
+        # column is originally float (all-NaN) -> filling with string gives object dtype
+        expected[2] = expected[2].astype("object")
+        tm.assert_frame_equal(result, expected)
+
+        return_value = df.fillna({2: "foo"}, inplace=True)
+        tm.assert_frame_equal(df, expected)
+        assert return_value is None
+
+    def test_fillna_limit_and_value(self):
+        # limit and value
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 3)))
+        df.iloc[2:7, 0] = np.nan
+        df.iloc[3:5, 2] = np.nan
+
+        expected = df.copy()
+        expected.iloc[2, 0] = 999
+        expected.iloc[3, 2] = 999
+        result = df.fillna(999, limit=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_datelike(self):
+        # with datelike
+        # GH#6344
+        df = DataFrame(
+            {
+                "Date": [NaT, Timestamp("2014-1-1")],
+                "Date2": [Timestamp("2013-1-1"), NaT],
+            }
+        )
+
+        expected = df.copy()
+        expected["Date"] = expected["Date"].fillna(df.loc[df.index[0], "Date2"])
+        result = df.fillna(value={"Date": df["Date2"]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_tzaware(self):
+        # with timezone
+        # GH#15855
+        df = DataFrame({"A": [Timestamp("2012-11-11 00:00:00+01:00"), NaT]})
+        exp = DataFrame(
+            {
+                "A": [
+                    Timestamp("2012-11-11 00:00:00+01:00"),
+                    Timestamp("2012-11-11 00:00:00+01:00"),
+                ]
+            }
+        )
+        res = df.ffill()
+        tm.assert_frame_equal(res, exp)
+
+        df = DataFrame({"A": [NaT, Timestamp("2012-11-11 00:00:00+01:00")]})
+        exp = DataFrame(
+            {
+                "A": [
+                    Timestamp("2012-11-11 00:00:00+01:00"),
+                    Timestamp("2012-11-11 00:00:00+01:00"),
+                ]
+            }
+        )
+        res = df.bfill()
+        tm.assert_frame_equal(res, exp)
+
+    def test_fillna_tzaware_different_column(self):
+        # with timezone in another column
+        # GH#15522
+        df = DataFrame(
+            {
+                "A": date_range("20130101", periods=4, tz="US/Eastern"),
+                "B": [1, 2, np.nan, np.nan],
+            }
+        )
+        result = df.ffill()
+        expected = DataFrame(
+            {
+                "A": date_range("20130101", periods=4, tz="US/Eastern"),
+                "B": [1.0, 2.0, 2.0, 2.0],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_na_actions_categorical(self):
+        cat = Categorical([1, 2, 3, np.nan], categories=[1, 2, 3])
+        vals = ["a", "b", np.nan, "d"]
+        df = DataFrame({"cats": cat, "vals": vals})
+        cat2 = Categorical([1, 2, 3, 3], categories=[1, 2, 3])
+        vals2 = ["a", "b", "b", "d"]
+        df_exp_fill = DataFrame({"cats": cat2, "vals": vals2})
+        cat3 = Categorical([1, 2, 3], categories=[1, 2, 3])
+        vals3 = ["a", "b", np.nan]
+        df_exp_drop_cats = DataFrame({"cats": cat3, "vals": vals3})
+        cat4 = Categorical([1, 2], categories=[1, 2, 3])
+        vals4 = ["a", "b"]
+        df_exp_drop_all = DataFrame({"cats": cat4, "vals": vals4})
+
+        # fillna
+        res = df.fillna(value={"cats": 3, "vals": "b"})
+        tm.assert_frame_equal(res, df_exp_fill)
+
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            df.fillna(value={"cats": 4, "vals": "c"})
+
+        res = df.ffill()
+        tm.assert_frame_equal(res, df_exp_fill)
+
+        # dropna
+        res = df.dropna(subset=["cats"])
+        tm.assert_frame_equal(res, df_exp_drop_cats)
+
+        res = df.dropna()
+        tm.assert_frame_equal(res, df_exp_drop_all)
+
+        # make sure that fillna takes missing values into account
+        c = Categorical([np.nan, "b", np.nan], categories=["a", "b"])
+        df = DataFrame({"cats": c, "vals": [1, 2, 3]})
+
+        cat_exp = Categorical(["a", "b", "a"], categories=["a", "b"])
+        df_exp = DataFrame({"cats": cat_exp, "vals": [1, 2, 3]})
+
+        res = df.fillna("a")
+        tm.assert_frame_equal(res, df_exp)
+
+    def test_fillna_categorical_nan(self):
+        # GH#14021
+        # np.nan should always be a valid filler
+        cat = Categorical([np.nan, 2, np.nan])
+        val = Categorical([np.nan, np.nan, np.nan])
+        df = DataFrame({"cats": cat, "vals": val})
+
+        # GH#32950 df.median() is poorly behaved because there is no
+        #  Categorical.median
+        median = Series({"cats": 2.0, "vals": np.nan})
+
+        res = df.fillna(median)
+        v_exp = [np.nan, np.nan, np.nan]
+        df_exp = DataFrame({"cats": [2, 2, 2], "vals": v_exp}, dtype="category")
+        tm.assert_frame_equal(res, df_exp)
+
+        result = df.cats.fillna(np.nan)
+        tm.assert_series_equal(result, df.cats)
+
+        result = df.vals.fillna(np.nan)
+        tm.assert_series_equal(result, df.vals)
+
+        idx = DatetimeIndex(
+            ["2011-01-01 09:00", "2016-01-01 23:45", "2011-01-01 09:00", NaT, NaT]
+        )
+        df = DataFrame({"a": Categorical(idx)})
+        tm.assert_frame_equal(df.fillna(value=NaT), df)
+
+        idx = PeriodIndex(["2011-01", "2011-01", "2011-01", NaT, NaT], freq="M")
+        df = DataFrame({"a": Categorical(idx)})
+        tm.assert_frame_equal(df.fillna(value=NaT), df)
+
+        idx = TimedeltaIndex(["1 days", "2 days", "1 days", NaT, NaT])
+        df = DataFrame({"a": Categorical(idx)})
+        tm.assert_frame_equal(df.fillna(value=NaT), df)
+
+    def test_fillna_no_downcast(self, frame_or_series):
+        # GH#45603 preserve object dtype
+        obj = frame_or_series([1, 2, 3], dtype="object")
+        result = obj.fillna("")
+        tm.assert_equal(result, obj)
+
+    @pytest.mark.parametrize("columns", [["A", "A", "B"], ["A", "A"]])
+    def test_fillna_dictlike_value_duplicate_colnames(self, columns):
+        # GH#43476
+        df = DataFrame(np.nan, index=[0, 1], columns=columns)
+        with tm.assert_produces_warning(None):
+            result = df.fillna({"A": 0})
+
+        expected = df.copy()
+        expected["A"] = 0.0
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_dtype_conversion(self):
+        # make sure that fillna on an empty frame works
+        df = DataFrame(index=["A", "B", "C"], columns=[1, 2, 3, 4, 5])
+        result = df.dtypes
+        expected = Series([np.dtype("object")] * 5, index=[1, 2, 3, 4, 5])
+        tm.assert_series_equal(result, expected)
+        result = df.fillna(1)
+        expected = DataFrame(
+            1, index=["A", "B", "C"], columns=[1, 2, 3, 4, 5], dtype=object
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # empty block
+        df = DataFrame(index=range(3), columns=["A", "B"], dtype="float64")
+        result = df.fillna("nan")
+        expected = DataFrame("nan", dtype="object", index=range(3), columns=["A", "B"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("val", ["", 1, np.nan, 1.0])
+    def test_fillna_dtype_conversion_equiv_replace(self, val):
+        df = DataFrame({"A": [1, np.nan], "B": [1.0, 2.0]})
+        expected = df.replace(np.nan, val)
+        result = df.fillna(val)
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_datetime_columns(self):
+        # GH#7095
+        df = DataFrame(
+            {
+                "A": [-1, -2, np.nan],
+                "B": date_range("20130101", periods=3),
+                "C": ["foo", "bar", None],
+                "D": ["foo2", "bar2", None],
+            },
+            index=date_range("20130110", periods=3),
+        )
+        result = df.fillna("?")
+        expected = DataFrame(
+            {
+                "A": [-1, -2, "?"],
+                "B": date_range("20130101", periods=3),
+                "C": ["foo", "bar", "?"],
+                "D": ["foo2", "bar2", "?"],
+            },
+            index=date_range("20130110", periods=3),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "A": [-1, -2, np.nan],
+                "B": [Timestamp("2013-01-01"), Timestamp("2013-01-02"), NaT],
+                "C": ["foo", "bar", None],
+                "D": ["foo2", "bar2", None],
+            },
+            index=date_range("20130110", periods=3),
+        )
+        result = df.fillna("?")
+        expected = DataFrame(
+            {
+                "A": [-1, -2, "?"],
+                "B": [Timestamp("2013-01-01"), Timestamp("2013-01-02"), "?"],
+                "C": ["foo", "bar", "?"],
+                "D": ["foo2", "bar2", "?"],
+            },
+            index=date_range("20130110", periods=3),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_ffill(self, datetime_frame):
+        datetime_frame.loc[datetime_frame.index[:5], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[-5:], "A"] = np.nan
+
+        alt = datetime_frame.ffill()
+        tm.assert_frame_equal(datetime_frame.ffill(), alt)
+
+    def test_bfill(self, datetime_frame):
+        datetime_frame.loc[datetime_frame.index[:5], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[-5:], "A"] = np.nan
+
+        alt = datetime_frame.bfill()
+        tm.assert_frame_equal(datetime_frame.bfill(), alt)
+
+    def test_frame_pad_backfill_limit(self):
+        index = np.arange(10)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)), index=index)
+
+        result = df[:2].reindex(index, method="pad", limit=5)
+
+        expected = df[:2].reindex(index).ffill()
+        expected.iloc[-3:] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = df[-2:].reindex(index, method="backfill", limit=5)
+
+        expected = df[-2:].reindex(index).bfill()
+        expected.iloc[:3] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_fillna_limit(self):
+        index = np.arange(10)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)), index=index)
+
+        result = df[:2].reindex(index)
+        result = result.ffill(limit=5)
+
+        expected = df[:2].reindex(index).ffill()
+        expected.iloc[-3:] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = df[-2:].reindex(index)
+        result = result.bfill(limit=5)
+
+        expected = df[-2:].reindex(index).bfill()
+        expected.iloc[:3] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_skip_certain_blocks(self):
+        # don't try to fill boolean, int blocks
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)).astype(int))
+
+        # it works!
+        df.fillna(np.nan)
+
+    @pytest.mark.parametrize("type", [int, float])
+    def test_fillna_positive_limit(self, type):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4))).astype(type)
+
+        msg = "Limit must be greater than 0"
+        with pytest.raises(ValueError, match=msg):
+            df.fillna(0, limit=-5)
+
+    @pytest.mark.parametrize("type", [int, float])
+    def test_fillna_integer_limit(self, type):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4))).astype(type)
+
+        msg = "Limit must be an integer"
+        with pytest.raises(ValueError, match=msg):
+            df.fillna(0, limit=0.5)
+
+    def test_fillna_inplace(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        df.loc[:4, 1] = np.nan
+        df.loc[-4:, 3] = np.nan
+
+        expected = df.fillna(value=0)
+        assert expected is not df
+
+        df.fillna(value=0, inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+        expected = df.fillna(value={0: 0}, inplace=True)
+        assert expected is None
+
+        df.loc[:4, 1] = np.nan
+        df.loc[-4:, 3] = np.nan
+        expected = df.ffill()
+        assert expected is not df
+
+        df.ffill(inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+    def test_fillna_dict_series(self):
+        df = DataFrame(
+            {
+                "a": [np.nan, 1, 2, np.nan, np.nan],
+                "b": [1, 2, 3, np.nan, np.nan],
+                "c": [np.nan, 1, 2, 3, 4],
+            }
+        )
+
+        result = df.fillna({"a": 0, "b": 5})
+
+        expected = df.copy()
+        expected["a"] = expected["a"].fillna(0)
+        expected["b"] = expected["b"].fillna(5)
+        tm.assert_frame_equal(result, expected)
+
+        # it works
+        result = df.fillna({"a": 0, "b": 5, "d": 7})
+
+        # Series treated same as dict
+        result = df.fillna(df.max())
+        expected = df.fillna(df.max().to_dict())
+        tm.assert_frame_equal(result, expected)
+
+        # disable this for now
+        with pytest.raises(NotImplementedError, match="column by column"):
+            df.fillna(df.max(axis=1), axis=1)
+
+    def test_fillna_dataframe(self):
+        # GH#8377
+        df = DataFrame(
+            {
+                "a": [np.nan, 1, 2, np.nan, np.nan],
+                "b": [1, 2, 3, np.nan, np.nan],
+                "c": [np.nan, 1, 2, 3, 4],
+            },
+            index=list("VWXYZ"),
+        )
+
+        # df2 may have different index and columns
+        df2 = DataFrame(
+            {
+                "a": [np.nan, 10, 20, 30, 40],
+                "b": [50, 60, 70, 80, 90],
+                "foo": ["bar"] * 5,
+            },
+            index=list("VWXuZ"),
+        )
+
+        result = df.fillna(df2)
+
+        # only those columns and indices which are shared get filled
+        expected = DataFrame(
+            {
+                "a": [np.nan, 1, 2, np.nan, 40],
+                "b": [1, 2, 3, np.nan, 90],
+                "c": [np.nan, 1, 2, 3, 4],
+            },
+            index=list("VWXYZ"),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_columns(self):
+        arr = np.random.default_rng(2).standard_normal((10, 10))
+        arr[:, ::2] = np.nan
+        df = DataFrame(arr)
+
+        result = df.ffill(axis=1)
+        expected = df.T.ffill().T
+        tm.assert_frame_equal(result, expected)
+
+        df.insert(6, "foo", 5)
+        result = df.ffill(axis=1)
+        expected = df.astype(float).ffill(axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_invalid_value(self, float_frame):
+        # list
+        msg = '"value" parameter must be a scalar or dict, but you passed a "{}"'
+        with pytest.raises(TypeError, match=msg.format("list")):
+            float_frame.fillna([1, 2])
+        # tuple
+        with pytest.raises(TypeError, match=msg.format("tuple")):
+            float_frame.fillna((1, 2))
+        # frame with series
+        msg = (
+            '"value" parameter must be a scalar, dict or Series, but you '
+            'passed a "DataFrame"'
+        )
+        with pytest.raises(TypeError, match=msg):
+            float_frame.iloc[:, 0].fillna(float_frame)
+
+    def test_fillna_col_reordering(self):
+        cols = ["COL." + str(i) for i in range(5, 0, -1)]
+        data = np.random.default_rng(2).random((20, 5))
+        df = DataFrame(index=range(20), columns=cols, data=data)
+        filled = df.ffill()
+        assert df.columns.tolist() == filled.columns.tolist()
+
+    def test_fill_empty(self, float_frame):
+        df = float_frame.reindex(columns=[])
+        result = df.fillna(value=0)
+        tm.assert_frame_equal(result, df)
+
+    def test_fillna_with_columns_and_limit(self):
+        # GH40989
+        df = DataFrame(
+            [
+                [np.nan, 2, np.nan, 0],
+                [3, 4, np.nan, 1],
+                [np.nan, np.nan, np.nan, 5],
+                [np.nan, 3, np.nan, 4],
+            ],
+            columns=list("ABCD"),
+        )
+        result = df.fillna(axis=1, value=100, limit=1)
+        result2 = df.fillna(axis=1, value=100, limit=2)
+
+        expected = DataFrame(
+            {
+                "A": Series([100, 3, 100, 100], dtype="float64"),
+                "B": [2, 4, np.nan, 3],
+                "C": [np.nan, 100, np.nan, np.nan],
+                "D": Series([0, 1, 5, 4], dtype="float64"),
+            },
+            index=[0, 1, 2, 3],
+        )
+        expected2 = DataFrame(
+            {
+                "A": Series([100, 3, 100, 100], dtype="float64"),
+                "B": Series([2, 4, 100, 3], dtype="float64"),
+                "C": [100, 100, np.nan, 100],
+                "D": Series([0, 1, 5, 4], dtype="float64"),
+            },
+            index=[0, 1, 2, 3],
+        )
+
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected2)
+
+    def test_fillna_datetime_inplace(self):
+        # GH#48863
+        df = DataFrame(
+            {
+                "date1": to_datetime(["2018-05-30", None]),
+                "date2": to_datetime(["2018-09-30", None]),
+            }
+        )
+        expected = df.copy()
+        df.fillna(np.nan, inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+    def test_fillna_inplace_with_columns_limit_and_value(self):
+        # GH40989
+        df = DataFrame(
+            [
+                [np.nan, 2, np.nan, 0],
+                [3, 4, np.nan, 1],
+                [np.nan, np.nan, np.nan, 5],
+                [np.nan, 3, np.nan, 4],
+            ],
+            columns=list("ABCD"),
+        )
+
+        expected = df.fillna(axis=1, value=100, limit=1)
+        assert expected is not df
+
+        df.fillna(axis=1, value=100, limit=1, inplace=True)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("val", [-1, {"x": -1, "y": -1}])
+    def test_inplace_dict_update_view(self, val):
+        # GH#47188
+        df = DataFrame({"x": [np.nan, 2], "y": [np.nan, 2]})
+        df_orig = df.copy()
+        result_view = df[:]
+        df.fillna(val, inplace=True)
+        expected = DataFrame({"x": [-1, 2.0], "y": [-1.0, 2]})
+        tm.assert_frame_equal(df, expected)
+        tm.assert_frame_equal(result_view, df_orig)
+
+    def test_single_block_df_with_horizontal_axis(self):
+        # GH 47713
+        df = DataFrame(
+            {
+                "col1": [5, 0, np.nan, 10, np.nan],
+                "col2": [7, np.nan, np.nan, 5, 3],
+                "col3": [12, np.nan, 1, 2, 0],
+                "col4": [np.nan, 1, 1, np.nan, 18],
+            }
+        )
+        result = df.fillna(50, limit=1, axis=1)
+        expected = DataFrame(
+            [
+                [5.0, 7.0, 12.0, 50.0],
+                [0.0, 50.0, np.nan, 1.0],
+                [50.0, np.nan, 1.0, 1.0],
+                [10.0, 5.0, 2.0, 50.0],
+                [50.0, 3.0, 0.0, 18.0],
+            ],
+            columns=["col1", "col2", "col3", "col4"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_fillna_with_multi_index_frame(self):
+        # GH 47649
+        pdf = DataFrame(
+            {
+                ("x", "a"): [np.nan, 2.0, 3.0],
+                ("x", "b"): [1.0, 2.0, np.nan],
+                ("y", "c"): [1.0, 2.0, np.nan],
+            }
+        )
+        expected = DataFrame(
+            {
+                ("x", "a"): [-1.0, 2.0, 3.0],
+                ("x", "b"): [1.0, 2.0, -1.0],
+                ("y", "c"): [1.0, 2.0, np.nan],
+            }
+        )
+        tm.assert_frame_equal(pdf.fillna({"x": -1}), expected)
+        tm.assert_frame_equal(pdf.fillna({"x": -1, ("x", "b"): -2}), expected)
+
+        expected = DataFrame(
+            {
+                ("x", "a"): [-1.0, 2.0, 3.0],
+                ("x", "b"): [1.0, 2.0, -2.0],
+                ("y", "c"): [1.0, 2.0, np.nan],
+            }
+        )
+        tm.assert_frame_equal(pdf.fillna({("x", "b"): -2, "x": -1}), expected)
+
+
+def test_fillna_nonconsolidated_frame():
+    # https://github.com/pandas-dev/pandas/issues/36495
+    df = DataFrame(
+        [
+            [1, 1, 1, 1.0],
+            [2, 2, 2, 2.0],
+            [3, 3, 3, 3.0],
+        ],
+        columns=["i1", "i2", "i3", "f1"],
+    )
+    df_nonconsol = df.pivot(index="i1", columns="i2")
+    result = df_nonconsol.fillna(0)
+    assert result.isna().sum().sum() == 0
+
+
+def test_fillna_nones_inplace():
+    # GH 48480
+    df = DataFrame(
+        [[None, None], [None, None]],
+        columns=["A", "B"],
+    )
+    df.fillna(value={"A": 1, "B": 2}, inplace=True)
+
+    expected = DataFrame([[1, 2], [1, 2]], columns=["A", "B"], dtype=object)
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected_data, method, kwargs",
+    (
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 3.0, 3.0, 3.0, 7.0, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "inside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 3.0, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "inside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, 7.0],
+            "ffill",
+            {"limit_area": "outside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, np.nan],
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+            [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            range(5),
+            range(5),
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 7.0, 7.0, 7.0, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "inside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, 7.0, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "inside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [3.0, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "outside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+    ),
+)
+def test_ffill_bfill_limit_area(data, expected_data, method, kwargs):
+    # GH#56492
+    df = DataFrame(data)
+    expected = DataFrame(expected_data)
+    result = getattr(df, method)(**kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("test_frame", [True, False])
+@pytest.mark.parametrize("dtype", ["float", "object"])
+def test_fillna_with_none_object(test_frame, dtype):
+    # GH#57723
+    obj = Series([1, np.nan, 3], dtype=dtype)
+    if test_frame:
+        obj = obj.to_frame()
+    result = obj.fillna(value=None)
+    expected = Series([1, None, 3], dtype=dtype)
+    if test_frame:
+        expected = expected.to_frame()
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_filter.py b/pandas/pandas/tests/frame/methods/test_filter.py
new file mode 100644
index 0000000000000..dc84e2adf1239
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_filter.py
@@ -0,0 +1,153 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestDataFrameFilter:
+    def test_filter(self, float_frame, float_string_frame):
+        # Items
+        filtered = float_frame.filter(["A", "B", "E"])
+        assert len(filtered.columns) == 2
+        assert "E" not in filtered
+
+        filtered = float_frame.filter(["A", "B", "E"], axis="columns")
+        assert len(filtered.columns) == 2
+        assert "E" not in filtered
+
+        # Other axis
+        idx = float_frame.index[0:4]
+        filtered = float_frame.filter(idx, axis="index")
+        expected = float_frame.reindex(index=idx)
+        tm.assert_frame_equal(filtered, expected)
+
+        # like
+        fcopy = float_frame.copy()
+        fcopy["AA"] = 1
+
+        filtered = fcopy.filter(like="A")
+        assert len(filtered.columns) == 2
+        assert "AA" in filtered
+
+        # like with ints in column names
+        df = DataFrame(0.0, index=[0, 1, 2], columns=[0, 1, "_A", "_B"])
+        filtered = df.filter(like="_")
+        assert len(filtered.columns) == 2
+
+        # regex with ints in column names
+        # from PR #10384
+        df = DataFrame(0.0, index=[0, 1, 2], columns=["A1", 1, "B", 2, "C"])
+        expected = DataFrame(
+            0.0, index=[0, 1, 2], columns=pd.Index([1, 2], dtype=object)
+        )
+        filtered = df.filter(regex="^[0-9]+$")
+        tm.assert_frame_equal(filtered, expected)
+
+        expected = DataFrame(0.0, index=[0, 1, 2], columns=[0, "0", 1, "1"])
+        # shouldn't remove anything
+        filtered = expected.filter(regex="^[0-9]+$")
+        tm.assert_frame_equal(filtered, expected)
+
+        # pass in None
+        with pytest.raises(TypeError, match="Must pass"):
+            float_frame.filter()
+        with pytest.raises(TypeError, match="Must pass"):
+            float_frame.filter(items=None)
+        with pytest.raises(TypeError, match="Must pass"):
+            float_frame.filter(axis=1)
+
+        # test mutually exclusive arguments
+        with pytest.raises(TypeError, match="mutually exclusive"):
+            float_frame.filter(items=["one", "three"], regex="e$", like="bbi")
+        with pytest.raises(TypeError, match="mutually exclusive"):
+            float_frame.filter(items=["one", "three"], regex="e$", axis=1)
+        with pytest.raises(TypeError, match="mutually exclusive"):
+            float_frame.filter(items=["one", "three"], regex="e$")
+        with pytest.raises(TypeError, match="mutually exclusive"):
+            float_frame.filter(items=["one", "three"], like="bbi", axis=0)
+        with pytest.raises(TypeError, match="mutually exclusive"):
+            float_frame.filter(items=["one", "three"], like="bbi")
+
+        # objects
+        filtered = float_string_frame.filter(like="foo")
+        assert "foo" in filtered
+
+        # unicode columns, won't ascii-encode
+        df = float_frame.rename(columns={"B": "\u2202"})
+        filtered = df.filter(like="C")
+        assert "C" in filtered
+
+    def test_filter_regex_search(self, float_frame):
+        fcopy = float_frame.copy()
+        fcopy["AA"] = 1
+
+        # regex
+        filtered = fcopy.filter(regex="[A]+")
+        assert len(filtered.columns) == 2
+        assert "AA" in filtered
+
+        # doesn't have to be at beginning
+        df = DataFrame(
+            {"aBBa": [1, 2], "BBaBB": [1, 2], "aCCa": [1, 2], "aCCaBB": [1, 2]}
+        )
+
+        result = df.filter(regex="BB")
+        exp = df[[x for x in df.columns if "BB" in x]]
+        tm.assert_frame_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "name,expected_data",
+        [
+            ("a", {"a": [1, 2]}),
+            ("あ", {"あ": [3, 4]}),
+        ],
+    )
+    def test_filter_unicode(self, name, expected_data):
+        # GH13101
+        df = DataFrame({"a": [1, 2], "あ": [3, 4]})
+        expected = DataFrame(expected_data)
+
+        tm.assert_frame_equal(df.filter(like=name), expected)
+        tm.assert_frame_equal(df.filter(regex=name), expected)
+
+    def test_filter_bytestring(self):
+        # GH13101
+        name = "a"
+        df = DataFrame({b"a": [1, 2], b"b": [3, 4]})
+        expected = DataFrame({b"a": [1, 2]})
+
+        tm.assert_frame_equal(df.filter(like=name), expected)
+        tm.assert_frame_equal(df.filter(regex=name), expected)
+
+    def test_filter_corner(self):
+        empty = DataFrame()
+
+        result = empty.filter([])
+        tm.assert_frame_equal(result, empty)
+
+        result = empty.filter(like="foo")
+        tm.assert_frame_equal(result, empty)
+
+    def test_filter_regex_non_string(self):
+        # GH#5798 trying to filter on non-string columns should drop,
+        #  not raise
+        df = DataFrame(np.random.default_rng(2).random((3, 2)), columns=["STRING", 123])
+        result = df.filter(regex="STRING")
+        expected = df[["STRING"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_filter_keep_order(self):
+        # GH#54980
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        result = df.filter(items=["B", "A"])
+        expected = df[["B", "A"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_filter_different_dtype(self):
+        # GH#54980
+        df = DataFrame({1: [1, 2, 3], 2: [4, 5, 6]})
+        result = df.filter(items=["B", "A"])
+        expected = df[[]]
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_first_valid_index.py b/pandas/pandas/tests/frame/methods/test_first_valid_index.py
new file mode 100644
index 0000000000000..5855be2373ae2
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_first_valid_index.py
@@ -0,0 +1,79 @@
+"""
+Includes test for last_valid_index.
+"""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+)
+
+
+class TestFirstValidIndex:
+    def test_first_valid_index_single_nan(self, frame_or_series):
+        # GH#9752 Series/DataFrame should both return None, not raise
+        obj = frame_or_series([np.nan])
+
+        assert obj.first_valid_index() is None
+        assert obj.iloc[:0].first_valid_index() is None
+
+    @pytest.mark.parametrize(
+        "empty", [DataFrame(), Series(dtype=object), Series([], index=[], dtype=object)]
+    )
+    def test_first_valid_index_empty(self, empty):
+        # GH#12800
+        assert empty.last_valid_index() is None
+        assert empty.first_valid_index() is None
+
+    @pytest.mark.parametrize(
+        "data,idx,expected_first,expected_last",
+        [
+            ({"A": [1, 2, 3]}, [1, 1, 2], 1, 2),
+            ({"A": [1, 2, 3]}, [1, 2, 2], 1, 2),
+            ({"A": [1, 2, 3, 4]}, ["d", "d", "d", "d"], "d", "d"),
+            ({"A": [1, np.nan, 3]}, [1, 1, 2], 1, 2),
+            ({"A": [np.nan, np.nan, 3]}, [1, 1, 2], 2, 2),
+            ({"A": [1, np.nan, 3]}, [1, 2, 2], 1, 2),
+        ],
+    )
+    def test_first_last_valid_frame(self, data, idx, expected_first, expected_last):
+        # GH#21441
+        df = DataFrame(data, index=idx)
+        assert expected_first == df.first_valid_index()
+        assert expected_last == df.last_valid_index()
+
+    @pytest.mark.parametrize(
+        "index",
+        [Index([str(i) for i in range(20)]), date_range("2020-01-01", periods=20)],
+    )
+    def test_first_last_valid(self, index):
+        mat = np.random.default_rng(2).standard_normal(len(index))
+        mat[:5] = np.nan
+        mat[-5:] = np.nan
+
+        frame = DataFrame({"foo": mat}, index=index)
+        assert frame.first_valid_index() == frame.index[5]
+        assert frame.last_valid_index() == frame.index[-6]
+
+        ser = frame["foo"]
+        assert ser.first_valid_index() == frame.index[5]
+        assert ser.last_valid_index() == frame.index[-6]
+
+    @pytest.mark.parametrize(
+        "index",
+        [Index([str(i) for i in range(10)]), date_range("2020-01-01", periods=10)],
+    )
+    def test_first_last_valid_all_nan(self, index):
+        # GH#17400: no valid entries
+        frame = DataFrame(np.nan, columns=["foo"], index=index)
+
+        assert frame.last_valid_index() is None
+        assert frame.first_valid_index() is None
+
+        ser = frame["foo"]
+        assert ser.first_valid_index() is None
+        assert ser.last_valid_index() is None
diff --git a/pandas/pandas/tests/frame/methods/test_get_numeric_data.py b/pandas/pandas/tests/frame/methods/test_get_numeric_data.py
new file mode 100644
index 0000000000000..6d097e75f6703
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_get_numeric_data.py
@@ -0,0 +1,104 @@
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+
+
+class TestGetNumericData:
+    def test_get_numeric_data_preserve_dtype(self):
+        # get the numeric data
+        obj = DataFrame({"A": [1, "2", 3.0]}, columns=Index(["A"], dtype="object"))
+        result = obj._get_numeric_data()
+        expected = DataFrame(dtype=object, index=pd.RangeIndex(3), columns=[])
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_numeric_data(self, using_infer_string):
+        datetime64name = np.dtype("M8[s]").name
+        objectname = np.dtype(np.object_).name
+
+        df = DataFrame(
+            {"a": 1.0, "b": 2, "c": "foo", "f": Timestamp("20010102")},
+            index=np.arange(10),
+        )
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("float64"),
+                np.dtype("int64"),
+                np.dtype(objectname)
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan),
+                np.dtype(datetime64name),
+            ],
+            index=["a", "b", "c", "f"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "a": 1.0,
+                "b": 2,
+                "c": "foo",
+                "d": np.array([1.0] * 10, dtype="float32"),
+                "e": np.array([1] * 10, dtype="int32"),
+                "f": np.array([1] * 10, dtype="int16"),
+                "g": Timestamp("20010102"),
+            },
+            index=np.arange(10),
+        )
+
+        result = df._get_numeric_data()
+        expected = df.loc[:, ["a", "b", "d", "e", "f"]]
+        tm.assert_frame_equal(result, expected)
+
+        only_obj = df.loc[:, ["c", "g"]]
+        result = only_obj._get_numeric_data()
+        expected = df.loc[:, []]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame.from_dict({"a": [1, 2], "b": ["foo", "bar"], "c": [np.pi, np.e]})
+        result = df._get_numeric_data()
+        expected = DataFrame.from_dict({"a": [1, 2], "c": [np.pi, np.e]})
+        tm.assert_frame_equal(result, expected)
+
+        df = result.copy()
+        result = df._get_numeric_data()
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_numeric_data_mixed_dtype(self):
+        # numeric and object columns
+
+        df = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": [True, False, True],
+                "c": ["foo", "bar", "baz"],
+                "d": [None, None, None],
+                "e": [3.14, 0.577, 2.773],
+            }
+        )
+        result = df._get_numeric_data()
+        tm.assert_index_equal(result.columns, Index(["a", "b", "e"]))
+
+    def test_get_numeric_data_extension_dtype(self):
+        # GH#22290
+        df = DataFrame(
+            {
+                "A": pd.array([-10, np.nan, 0, 10, 20, 30], dtype="Int64"),
+                "B": Categorical(list("abcabc")),
+                "C": pd.array([0, 1, 2, 3, np.nan, 5], dtype="UInt8"),
+                "D": IntervalArray.from_breaks(range(7)),
+            }
+        )
+        result = df._get_numeric_data()
+        expected = df.loc[:, ["A", "C"]]
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_head_tail.py b/pandas/pandas/tests/frame/methods/test_head_tail.py
new file mode 100644
index 0000000000000..9363c4d79983f
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_head_tail.py
@@ -0,0 +1,57 @@
+import numpy as np
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+def test_head_tail_generic(index, frame_or_series):
+    # GH#5370
+
+    ndim = 2 if frame_or_series is DataFrame else 1
+    shape = (len(index),) * ndim
+    vals = np.random.default_rng(2).standard_normal(shape)
+    obj = frame_or_series(vals, index=index)
+
+    tm.assert_equal(obj.head(), obj.iloc[:5])
+    tm.assert_equal(obj.tail(), obj.iloc[-5:])
+
+    # 0-len
+    tm.assert_equal(obj.head(0), obj.iloc[0:0])
+    tm.assert_equal(obj.tail(0), obj.iloc[0:0])
+
+    # bounded
+    tm.assert_equal(obj.head(len(obj) + 1), obj)
+    tm.assert_equal(obj.tail(len(obj) + 1), obj)
+
+    # neg index
+    tm.assert_equal(obj.head(-3), obj.head(len(index) - 3))
+    tm.assert_equal(obj.tail(-3), obj.tail(len(index) - 3))
+
+
+def test_head_tail(float_frame):
+    tm.assert_frame_equal(float_frame.head(), float_frame[:5])
+    tm.assert_frame_equal(float_frame.tail(), float_frame[-5:])
+
+    tm.assert_frame_equal(float_frame.head(0), float_frame[0:0])
+    tm.assert_frame_equal(float_frame.tail(0), float_frame[0:0])
+
+    tm.assert_frame_equal(float_frame.head(-1), float_frame[:-1])
+    tm.assert_frame_equal(float_frame.tail(-1), float_frame[1:])
+    tm.assert_frame_equal(float_frame.head(1), float_frame[:1])
+    tm.assert_frame_equal(float_frame.tail(1), float_frame[-1:])
+    # with a float index
+    df = float_frame.copy()
+    df.index = np.arange(len(float_frame)) + 0.1
+    tm.assert_frame_equal(df.head(), df.iloc[:5])
+    tm.assert_frame_equal(df.tail(), df.iloc[-5:])
+    tm.assert_frame_equal(df.head(0), df[0:0])
+    tm.assert_frame_equal(df.tail(0), df[0:0])
+    tm.assert_frame_equal(df.head(-1), df.iloc[:-1])
+    tm.assert_frame_equal(df.tail(-1), df.iloc[1:])
+
+
+def test_head_tail_empty():
+    # test empty dataframe
+    empty_df = DataFrame()
+    tm.assert_frame_equal(empty_df.tail(), empty_df)
+    tm.assert_frame_equal(empty_df.head(), empty_df)
diff --git a/pandas/pandas/tests/frame/methods/test_infer_objects.py b/pandas/pandas/tests/frame/methods/test_infer_objects.py
new file mode 100644
index 0000000000000..c7cdcd177403b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_infer_objects.py
@@ -0,0 +1,42 @@
+from datetime import datetime
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestInferObjects:
+    def test_infer_objects(self):
+        # GH#11221
+        df = DataFrame(
+            {
+                "a": ["a", 1, 2, 3],
+                "b": ["b", 2.0, 3.0, 4.1],
+                "c": [
+                    "c",
+                    datetime(2016, 1, 1),
+                    datetime(2016, 1, 2),
+                    datetime(2016, 1, 3),
+                ],
+                "d": [1, 2, 3, "d"],
+            },
+            columns=["a", "b", "c", "d"],
+        )
+        df = df.iloc[1:].infer_objects()
+
+        assert df["a"].dtype == "int64"
+        assert df["b"].dtype == "float64"
+        assert df["c"].dtype == "M8[us]"
+        assert df["d"].dtype == "object"
+
+        expected = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": [2.0, 3.0, 4.1],
+                "c": [datetime(2016, 1, 1), datetime(2016, 1, 2), datetime(2016, 1, 3)],
+                "d": [2, 3, "d"],
+            },
+            columns=["a", "b", "c", "d"],
+        )
+        # reconstruct frame to verify inference is same
+        result = df.reset_index(drop=True)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_info.py b/pandas/pandas/tests/frame/methods/test_info.py
new file mode 100644
index 0000000000000..de6737ec3bc39
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_info.py
@@ -0,0 +1,586 @@
+from io import StringIO
+import re
+from string import ascii_uppercase
+import sys
+import textwrap
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    HAS_PYARROW,
+    IS64,
+    PYPY,
+    is_platform_arm,
+)
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+@pytest.fixture
+def duplicate_columns_frame():
+    """Dataframe with duplicate column names."""
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((1500, 4)),
+        columns=["a", "a", "b", "b"],
+    )
+
+
+def test_info_empty():
+    # GH #45494
+    df = DataFrame()
+    buf = StringIO()
+    df.info(buf=buf)
+    result = buf.getvalue()
+    expected = textwrap.dedent(
+        """\
+        <class 'pandas.DataFrame'>
+        RangeIndex: 0 entries
+        Empty DataFrame\n"""
+    )
+    assert result == expected
+
+
+def test_info_categorical_column_smoke_test():
+    n = 2500
+    df = DataFrame({"int64": np.random.default_rng(2).integers(100, size=n, dtype=int)})
+    df["category"] = Series(
+        np.array(list("abcdefghij")).take(
+            np.random.default_rng(2).integers(0, 10, size=n, dtype=int)
+        )
+    ).astype("category")
+    df.isna()
+    buf = StringIO()
+    df.info(buf=buf)
+
+    df2 = df[df["category"] == "d"]
+    buf = StringIO()
+    df2.info(buf=buf)
+
+
+@pytest.mark.parametrize(
+    "fixture_func_name",
+    [
+        "int_frame",
+        "float_frame",
+        "datetime_frame",
+        "duplicate_columns_frame",
+        "float_string_frame",
+    ],
+)
+def test_info_smoke_test(fixture_func_name, request):
+    frame = request.getfixturevalue(fixture_func_name)
+    buf = StringIO()
+    frame.info(buf=buf)
+    result = buf.getvalue().splitlines()
+    assert len(result) > 10
+
+    buf = StringIO()
+    frame.info(buf=buf, verbose=False)
+
+
+def test_info_smoke_test2(float_frame):
+    # pretty useless test, used to be mixed into the repr tests
+    buf = StringIO()
+    float_frame.reindex(columns=["A"]).info(verbose=False, buf=buf)
+    float_frame.reindex(columns=["A", "B"]).info(verbose=False, buf=buf)
+
+    # no columns or index
+    DataFrame().info(buf=buf)
+
+
+@pytest.mark.parametrize(
+    "num_columns, max_info_columns, verbose",
+    [
+        (10, 100, True),
+        (10, 11, True),
+        (10, 10, True),
+        (10, 9, False),
+        (10, 1, False),
+    ],
+)
+def test_info_default_verbose_selection(num_columns, max_info_columns, verbose):
+    frame = DataFrame(np.random.default_rng(2).standard_normal((5, num_columns)))
+    with option_context("display.max_info_columns", max_info_columns):
+        io_default = StringIO()
+        frame.info(buf=io_default)
+        result = io_default.getvalue()
+
+        io_explicit = StringIO()
+        frame.info(buf=io_explicit, verbose=verbose)
+        expected = io_explicit.getvalue()
+
+        assert result == expected
+
+
+def test_info_verbose_check_header_separator_body():
+    buf = StringIO()
+    size = 1001
+    start = 5
+    frame = DataFrame(np.random.default_rng(2).standard_normal((3, size)))
+    frame.info(verbose=True, buf=buf)
+
+    res = buf.getvalue()
+    header = " #     Column  Dtype  \n---    ------  -----  "
+    assert header in res
+
+    frame.info(verbose=True, buf=buf)
+    buf.seek(0)
+    lines = buf.readlines()
+    assert len(lines) > 0
+
+    for i, line in enumerate(lines):
+        if start <= i < start + size:
+            line_nr = f" {i - start} "
+            assert line.startswith(line_nr)
+
+
+@pytest.mark.parametrize(
+    "size, header_exp, separator_exp, first_line_exp, last_line_exp",
+    [
+        (
+            4,
+            " #   Column  Non-Null Count  Dtype  ",
+            "---  ------  --------------  -----  ",
+            " 0   0       3 non-null      float64",
+            " 3   3       3 non-null      float64",
+        ),
+        (
+            11,
+            " #   Column  Non-Null Count  Dtype  ",
+            "---  ------  --------------  -----  ",
+            " 0   0       3 non-null      float64",
+            " 10  10      3 non-null      float64",
+        ),
+        (
+            101,
+            " #    Column  Non-Null Count  Dtype  ",
+            "---   ------  --------------  -----  ",
+            " 0    0       3 non-null      float64",
+            " 100  100     3 non-null      float64",
+        ),
+        (
+            1001,
+            " #     Column  Non-Null Count  Dtype  ",
+            "---    ------  --------------  -----  ",
+            " 0     0       3 non-null      float64",
+            " 1000  1000    3 non-null      float64",
+        ),
+        (
+            10001,
+            " #      Column  Non-Null Count  Dtype  ",
+            "---     ------  --------------  -----  ",
+            " 0      0       3 non-null      float64",
+            " 10000  10000   3 non-null      float64",
+        ),
+    ],
+)
+def test_info_verbose_with_counts_spacing(
+    size, header_exp, separator_exp, first_line_exp, last_line_exp
+):
+    """Test header column, spacer, first line and last line in verbose mode."""
+    frame = DataFrame(np.random.default_rng(2).standard_normal((3, size)))
+    with StringIO() as buf:
+        frame.info(verbose=True, show_counts=True, buf=buf)
+        all_lines = buf.getvalue().splitlines()
+    # Here table would contain only header, separator and table lines
+    # dframe repr, index summary, memory usage and dtypes are excluded
+    table = all_lines[3:-2]
+    header, separator, first_line, *rest, last_line = table
+    assert header == header_exp
+    assert separator == separator_exp
+    assert first_line == first_line_exp
+    assert last_line == last_line_exp
+
+
+def test_info_memory():
+    # https://github.com/pandas-dev/pandas/issues/21056
+    df = DataFrame({"a": Series([1, 2], dtype="i8")})
+    buf = StringIO()
+    df.info(buf=buf)
+    result = buf.getvalue()
+    bytes = float(df.memory_usage().sum())
+    expected = textwrap.dedent(
+        f"""\
+    <class 'pandas.DataFrame'>
+    RangeIndex: 2 entries, 0 to 1
+    Data columns (total 1 columns):
+     #   Column  Non-Null Count  Dtype
+    ---  ------  --------------  -----
+     0   a       2 non-null      int64
+    dtypes: int64(1)
+    memory usage: {bytes} bytes
+    """
+    )
+    assert result == expected
+
+
+def test_info_wide():
+    io = StringIO()
+    df = DataFrame(np.random.default_rng(2).standard_normal((5, 101)))
+    df.info(buf=io)
+
+    io = StringIO()
+    df.info(buf=io, max_cols=101)
+    result = io.getvalue()
+    assert len(result.splitlines()) > 100
+
+    expected = result
+    with option_context("display.max_info_columns", 101):
+        io = StringIO()
+        df.info(buf=io)
+        result = io.getvalue()
+        assert result == expected
+
+
+def test_info_duplicate_columns_shows_correct_dtypes():
+    # GH11761
+    io = StringIO()
+    frame = DataFrame([[1, 2.0]], columns=["a", "a"])
+    frame.info(buf=io)
+    lines = io.getvalue().splitlines(True)
+    assert " 0   a       1 non-null      int64  \n" == lines[5]
+    assert " 1   a       1 non-null      float64\n" == lines[6]
+
+
+def test_info_shows_column_dtypes():
+    dtypes = [
+        "int64",
+        "float64",
+        "datetime64[ns]",
+        "timedelta64[ns]",
+        "complex128",
+        "object",
+        "bool",
+    ]
+    data = {}
+    n = 10
+    for i, dtype in enumerate(dtypes):
+        data[i] = np.random.default_rng(2).integers(2, size=n).astype(dtype)
+    df = DataFrame(data)
+    buf = StringIO()
+    df.info(buf=buf)
+    res = buf.getvalue()
+    header = (
+        " #   Column  Non-Null Count  Dtype          \n"
+        "---  ------  --------------  -----          "
+    )
+    assert header in res
+    for i, dtype in enumerate(dtypes):
+        name = f" {i:d}   {i:d}       {n:d} non-null     {dtype}"
+        assert name in res
+
+
+def test_info_max_cols():
+    df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+    for len_, verbose in [(5, None), (5, False), (12, True)]:
+        # For verbose always      ^ setting  ^ summarize ^ full output
+        with option_context("max_info_columns", 4):
+            buf = StringIO()
+            df.info(buf=buf, verbose=verbose)
+            res = buf.getvalue()
+            assert len(res.strip().split("\n")) == len_
+
+    for len_, verbose in [(12, None), (5, False), (12, True)]:
+        # max_cols not exceeded
+        with option_context("max_info_columns", 5):
+            buf = StringIO()
+            df.info(buf=buf, verbose=verbose)
+            res = buf.getvalue()
+            assert len(res.strip().split("\n")) == len_
+
+    for len_, max_cols in [(12, 5), (5, 4)]:
+        # setting truncates
+        with option_context("max_info_columns", 4):
+            buf = StringIO()
+            df.info(buf=buf, max_cols=max_cols)
+            res = buf.getvalue()
+            assert len(res.strip().split("\n")) == len_
+
+        # setting wouldn't truncate
+        with option_context("max_info_columns", 5):
+            buf = StringIO()
+            df.info(buf=buf, max_cols=max_cols)
+            res = buf.getvalue()
+            assert len(res.strip().split("\n")) == len_
+
+
+def test_info_memory_usage():
+    # Ensure memory usage is displayed, when asserted, on the last line
+    dtypes = [
+        "int64",
+        "float64",
+        "datetime64[ns]",
+        "timedelta64[ns]",
+        "complex128",
+        "object",
+        "bool",
+    ]
+    data = {}
+    n = 10
+    for i, dtype in enumerate(dtypes):
+        data[i] = np.random.default_rng(2).integers(2, size=n).astype(dtype)
+    df = DataFrame(data)
+    buf = StringIO()
+
+    # display memory usage case
+    df.info(buf=buf, memory_usage=True)
+    res = buf.getvalue().splitlines()
+    assert "memory usage: " in res[-1]
+
+    # do not display memory usage case
+    df.info(buf=buf, memory_usage=False)
+    res = buf.getvalue().splitlines()
+    assert "memory usage: " not in res[-1]
+
+    df.info(buf=buf, memory_usage=True)
+    res = buf.getvalue().splitlines()
+
+    # memory usage is a lower bound, so print it as XYZ+ MB
+    assert re.match(r"memory usage: [^+]+\+", res[-1])
+
+    df.iloc[:, :5].info(buf=buf, memory_usage=True)
+    res = buf.getvalue().splitlines()
+
+    # excluded column with object dtype, so estimate is accurate
+    assert not re.match(r"memory usage: [^+]+\+", res[-1])
+
+    # Test a DataFrame with duplicate columns
+    dtypes = ["int64", "int64", "int64", "float64"]
+    data = {}
+    n = 100
+    for i, dtype in enumerate(dtypes):
+        data[i] = np.random.default_rng(2).integers(2, size=n).astype(dtype)
+    df = DataFrame(data)
+    df.columns = dtypes
+
+    df_with_object_index = DataFrame({"a": [1]}, index=Index(["foo"], dtype=object))
+    df_with_object_index.info(buf=buf, memory_usage=True)
+    res = buf.getvalue().splitlines()
+    assert re.match(r"memory usage: [^+]+\+", res[-1])
+
+    df_with_object_index.info(buf=buf, memory_usage="deep")
+    res = buf.getvalue().splitlines()
+    assert re.match(r"memory usage: [^+]+$", res[-1])
+
+    # Ensure df size is as expected
+    # (cols * rows * bytes) + index size
+    df_size = df.memory_usage().sum()
+    exp_size = len(dtypes) * n * 8 + df.index.nbytes
+    assert df_size == exp_size
+
+    # Ensure number of cols in memory_usage is the same as df
+    size_df = np.size(df.columns.values) + 1  # index=True; default
+    assert size_df == np.size(df.memory_usage())
+
+    # assert deep works only on object
+    assert df.memory_usage().sum() == df.memory_usage(deep=True).sum()
+
+    # test for validity
+    DataFrame(1, index=["a"], columns=["A"]).memory_usage(index=True)
+    DataFrame(1, index=["a"], columns=["A"]).index.nbytes
+    df = DataFrame(
+        data=1, index=MultiIndex.from_product([["a"], range(1000)]), columns=["A"]
+    )
+    df.index.nbytes
+    df.memory_usage(index=True)
+    df.index.values.nbytes
+
+    mem = df.memory_usage(deep=True).sum()
+    assert mem > 0
+
+
+@pytest.mark.skipif(PYPY, reason="on PyPy deep=True doesn't change result")
+def test_info_memory_usage_deep_not_pypy():
+    df_with_object_index = DataFrame({"a": [1]}, index=Index(["foo"], dtype=object))
+    assert (
+        df_with_object_index.memory_usage(index=True, deep=True).sum()
+        > df_with_object_index.memory_usage(index=True).sum()
+    )
+
+    df_object = DataFrame({"a": Series(["a"], dtype=object)})
+    assert df_object.memory_usage(deep=True).sum() > df_object.memory_usage().sum()
+
+
+@pytest.mark.xfail(not PYPY, reason="on PyPy deep=True does not change result")
+def test_info_memory_usage_deep_pypy():
+    df_with_object_index = DataFrame({"a": [1]}, index=Index(["foo"], dtype=object))
+    assert (
+        df_with_object_index.memory_usage(index=True, deep=True).sum()
+        == df_with_object_index.memory_usage(index=True).sum()
+    )
+
+    df_object = DataFrame({"a": Series(["a"], dtype=object)})
+    assert df_object.memory_usage(deep=True).sum() == df_object.memory_usage().sum()
+
+
+@pytest.mark.skipif(PYPY, reason="PyPy getsizeof() fails by design")
+def test_usage_via_getsizeof():
+    df = DataFrame(
+        data=1, index=MultiIndex.from_product([["a"], range(1000)]), columns=["A"]
+    )
+    mem = df.memory_usage(deep=True).sum()
+    # sys.getsizeof will call the .memory_usage with
+    # deep=True, and add on some GC overhead
+    diff = mem - sys.getsizeof(df)
+    assert abs(diff) < 100
+
+
+def test_info_memory_usage_qualified(using_infer_string):
+    buf = StringIO()
+    df = DataFrame(1, columns=list("ab"), index=[1, 2, 3])
+    df.info(buf=buf)
+    assert "+" not in buf.getvalue()
+
+    buf = StringIO()
+    df = DataFrame(1, columns=list("ab"), index=Index(list("ABC"), dtype=object))
+    df.info(buf=buf)
+    assert "+" in buf.getvalue()
+
+    buf = StringIO()
+    df = DataFrame(1, columns=list("ab"), index=Index(list("ABC"), dtype="str"))
+    df.info(buf=buf)
+    if using_infer_string and HAS_PYARROW:
+        assert "+" not in buf.getvalue()
+    else:
+        assert "+" in buf.getvalue()
+
+    buf = StringIO()
+    df = DataFrame(
+        1, columns=list("ab"), index=MultiIndex.from_product([range(3), range(3)])
+    )
+    df.info(buf=buf)
+    assert "+" not in buf.getvalue()
+
+    buf = StringIO()
+    df = DataFrame(
+        1, columns=list("ab"), index=MultiIndex.from_product([range(3), ["foo", "bar"]])
+    )
+    df.info(buf=buf)
+    if using_infer_string and HAS_PYARROW:
+        assert "+" not in buf.getvalue()
+    else:
+        assert "+" in buf.getvalue()
+
+
+def test_info_memory_usage_bug_on_multiindex():
+    # GH 14308
+    # memory usage introspection should not materialize .values
+
+    def memory_usage(f):
+        return f.memory_usage(deep=True).sum()
+
+    N = 100
+    M = len(ascii_uppercase)
+    index = MultiIndex.from_product(
+        [list(ascii_uppercase), date_range("20160101", periods=N)],
+        names=["id", "date"],
+    )
+    df = DataFrame(
+        {"value": np.random.default_rng(2).standard_normal(N * M)}, index=index
+    )
+
+    unstacked = df.unstack("id")
+    assert df.values.nbytes == unstacked.values.nbytes
+    assert memory_usage(df) > memory_usage(unstacked)
+
+    # high upper bound
+    assert memory_usage(unstacked) - memory_usage(df) < 2000
+
+
+def test_info_categorical():
+    # GH14298
+    idx = CategoricalIndex(["a", "b"])
+    df = DataFrame(np.zeros((2, 2)), index=idx, columns=idx)
+
+    buf = StringIO()
+    df.info(buf=buf)
+
+
+@pytest.mark.xfail(not IS64, reason="GH 36579: fail on 32-bit system")
+def test_info_int_columns(using_infer_string):
+    # GH#37245
+    df = DataFrame({1: [1, 2], 2: [2, 3]}, index=["A", "B"])
+    buf = StringIO()
+    df.info(show_counts=True, buf=buf)
+    result = buf.getvalue()
+    expected = textwrap.dedent(
+        f"""\
+        <class 'pandas.DataFrame'>
+        Index: 2 entries, A to B
+        Data columns (total 2 columns):
+         #   Column  Non-Null Count  Dtype
+        ---  ------  --------------  -----
+         0   1       2 non-null      int64
+         1   2       2 non-null      int64
+        dtypes: int64(2)
+        memory usage: {"50.0" if using_infer_string and HAS_PYARROW else "48.0+"} bytes
+        """
+    )
+    assert result == expected
+
+
+def test_memory_usage_empty_no_warning(using_infer_string):
+    # GH#50066
+    df = DataFrame(index=["a", "b"])
+    with tm.assert_produces_warning(None):
+        result = df.memory_usage()
+    if using_infer_string and HAS_PYARROW:
+        value = 18
+    else:
+        value = 16 if IS64 else 8
+    expected = Series(value, index=["Index"])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.single_cpu
+def test_info_compute_numba():
+    # GH#51922
+    numba = pytest.importorskip("numba")
+    if Version(numba.__version__) == Version("0.61") and is_platform_arm():
+        pytest.skip(f"Segfaults on ARM platforms with numba {numba.__version__}")
+    df = DataFrame([[1, 2], [3, 4]])
+
+    with option_context("compute.use_numba", True):
+        buf = StringIO()
+        df.info(buf=buf)
+        result = buf.getvalue()
+
+    buf = StringIO()
+    df.info(buf=buf)
+    expected = buf.getvalue()
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "row, columns, show_counts, result",
+    [
+        [20, 20, None, True],
+        [20, 20, True, True],
+        [20, 20, False, False],
+        [5, 5, None, False],
+        [5, 5, True, False],
+        [5, 5, False, False],
+    ],
+)
+def test_info_show_counts(row, columns, show_counts, result):
+    # Explicit cast to float to avoid implicit cast when setting nan
+    df = DataFrame(1, columns=range(10), index=range(10)).astype({1: "float"})
+    df.iloc[1, 1] = np.nan
+
+    with option_context(
+        "display.max_info_rows", row, "display.max_info_columns", columns
+    ):
+        with StringIO() as buf:
+            df.info(buf=buf, show_counts=show_counts)
+            assert ("non-null" in buf.getvalue()) is result
diff --git a/pandas/pandas/tests/frame/methods/test_interpolate.py b/pandas/pandas/tests/frame/methods/test_interpolate.py
new file mode 100644
index 0000000000000..09d1cc9a479b2
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_interpolate.py
@@ -0,0 +1,442 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    NaT,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameInterpolate:
+    def test_interpolate_complex(self):
+        # GH#53635
+        ser = Series([complex("1+1j"), float("nan"), complex("2+2j")])
+        assert ser.dtype.kind == "c"
+
+        res = ser.interpolate()
+        expected = Series([ser[0], ser[0] * 1.5, ser[2]])
+        tm.assert_series_equal(res, expected)
+
+        df = ser.to_frame()
+        res = df.interpolate()
+        expected = expected.to_frame()
+        tm.assert_frame_equal(res, expected)
+
+    def test_interpolate_datetimelike_values(self, frame_or_series):
+        # GH#11312, GH#51005
+        orig = Series(date_range("2012-01-01", periods=5))
+        ser = orig.copy()
+        ser[2] = NaT
+
+        res = frame_or_series(ser).interpolate()
+        expected = frame_or_series(orig)
+        tm.assert_equal(res, expected)
+
+        # datetime64tz cast
+        ser_tz = ser.dt.tz_localize("US/Pacific")
+        res_tz = frame_or_series(ser_tz).interpolate()
+        expected_tz = frame_or_series(orig.dt.tz_localize("US/Pacific"))
+        tm.assert_equal(res_tz, expected_tz)
+
+        # timedelta64 cast
+        ser_td = ser - ser[0]
+        res_td = frame_or_series(ser_td).interpolate()
+        expected_td = frame_or_series(orig - orig[0])
+        tm.assert_equal(res_td, expected_td)
+
+    def test_interpolate_inplace(self, frame_or_series, request):
+        # GH#44749
+        obj = frame_or_series([1, np.nan, 2])
+        orig = obj.values
+
+        obj.interpolate(inplace=True)
+        expected = frame_or_series([1, 1.5, 2])
+        tm.assert_equal(obj, expected)
+
+        # check we operated *actually* inplace
+        assert np.shares_memory(orig, obj.values)
+        assert orig.squeeze()[1] == 1.5
+
+    def test_interp_basic(self, using_infer_string):
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 4],
+                "B": [1, 4, 9, np.nan],
+                "C": [1, 2, 3, 5],
+                "D": list("abcd"),
+            }
+        )
+        dtype = "str" if using_infer_string else "object"
+        msg = f"[Cc]annot interpolate with {dtype} dtype"
+        with pytest.raises(TypeError, match=msg):
+            df.interpolate()
+
+        cvalues = df["C"]._values
+        dvalues = df["D"].values
+        with pytest.raises(TypeError, match=msg):
+            df.interpolate(inplace=True)
+
+        # check we DID operate inplace
+        assert tm.shares_memory(df["C"]._values, cvalues)
+        assert tm.shares_memory(df["D"]._values, dvalues)
+
+    @pytest.mark.xfail(
+        using_string_dtype(), reason="interpolate doesn't work for string"
+    )
+    def test_interp_basic_with_non_range_index(self, using_infer_string):
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 4],
+                "B": [1, 4, 9, np.nan],
+                "C": [1, 2, 3, 5],
+                "D": list("abcd"),
+            }
+        )
+
+        msg = "DataFrame cannot interpolate with object dtype"
+        if not using_infer_string:
+            with pytest.raises(TypeError, match=msg):
+                df.set_index("C").interpolate()
+        else:
+            result = df.set_index("C").interpolate()
+            expected = df.set_index("C")
+            expected.loc[3, "A"] = 2.66667
+            expected.loc[5, "B"] = 9
+            tm.assert_frame_equal(result, expected)
+
+    def test_interp_empty(self):
+        # https://github.com/pandas-dev/pandas/issues/35598
+        df = DataFrame()
+        result = df.interpolate()
+        assert result is not df
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+    def test_interp_bad_method(self):
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 4],
+                "B": [1, 4, 9, np.nan],
+                "C": [1, 2, 3, 5],
+            }
+        )
+        msg = "Can not interpolate with method=not_a_method"
+        with pytest.raises(ValueError, match=msg):
+            df.interpolate(method="not_a_method")
+
+    def test_interp_combo(self):
+        df = DataFrame(
+            {
+                "A": [1.0, 2.0, np.nan, 4.0],
+                "B": [1, 4, 9, np.nan],
+                "C": [1, 2, 3, 5],
+                "D": list("abcd"),
+            }
+        )
+
+        result = df["A"].interpolate()
+        expected = Series([1.0, 2.0, 3.0, 4.0], name="A")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_nan_idx(self):
+        df = DataFrame({"A": [1, 2, np.nan, 4], "B": [np.nan, 2, 3, 4]})
+        df = df.set_index("A")
+        msg = (
+            "Interpolation with NaNs in the index has not been implemented. "
+            "Try filling those NaNs before interpolating."
+        )
+        with pytest.raises(NotImplementedError, match=msg):
+            df.interpolate(method="values")
+
+    def test_interp_various(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(
+            {"A": [1, 2, np.nan, 4, 5, np.nan, 7], "C": [1, 2, 3, 5, 8, 13, 21]}
+        )
+        df = df.set_index("C")
+        expected = df.copy()
+        result = df.interpolate(method="polynomial", order=1)
+
+        expected.loc[3, "A"] = 2.66666667
+        expected.loc[13, "A"] = 5.76923076
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(method="cubic")
+        # GH #15662.
+        expected.loc[3, "A"] = 2.81547781
+        expected.loc[13, "A"] = 5.52964175
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(method="nearest")
+        expected.loc[3, "A"] = 2
+        expected.loc[13, "A"] = 5
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+        result = df.interpolate(method="quadratic")
+        expected.loc[3, "A"] = 2.82150771
+        expected.loc[13, "A"] = 6.12648668
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(method="slinear")
+        expected.loc[3, "A"] = 2.66666667
+        expected.loc[13, "A"] = 5.76923077
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(method="zero")
+        expected.loc[3, "A"] = 2.0
+        expected.loc[13, "A"] = 5
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_interp_alt_scipy(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(
+            {"A": [1, 2, np.nan, 4, 5, np.nan, 7], "C": [1, 2, 3, 5, 8, 13, 21]}
+        )
+        result = df.interpolate(method="barycentric")
+        expected = df.copy()
+        expected.loc[2, "A"] = 3
+        expected.loc[5, "A"] = 6
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(method="krogh")
+        expectedk = df.copy()
+        expectedk["A"] = expected["A"]
+        tm.assert_frame_equal(result, expectedk)
+
+        result = df.interpolate(method="pchip")
+        expected.loc[2, "A"] = 3
+        expected.loc[5, "A"] = 6.0
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_interp_rowwise(self):
+        df = DataFrame(
+            {
+                0: [1, 2, np.nan, 4],
+                1: [2, 3, 4, np.nan],
+                2: [np.nan, 4, 5, 6],
+                3: [4, np.nan, 6, 7],
+                4: [1, 2, 3, 4],
+            }
+        )
+        result = df.interpolate(axis=1)
+        expected = df.copy()
+        expected.loc[3, 1] = 5
+        expected.loc[0, 2] = 3
+        expected.loc[1, 3] = 3
+        expected[4] = expected[4].astype(np.float64)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(axis=1, method="values")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.interpolate(axis=0)
+        expected = df.interpolate()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "axis_name, axis_number",
+        [
+            pytest.param("rows", 0, id="rows_0"),
+            pytest.param("index", 0, id="index_0"),
+            pytest.param("columns", 1, id="columns_1"),
+        ],
+    )
+    def test_interp_axis_names(self, axis_name, axis_number):
+        # GH 29132: test axis names
+        data = {0: [0, np.nan, 6], 1: [1, np.nan, 7], 2: [2, 5, 8]}
+
+        df = DataFrame(data, dtype=np.float64)
+        result = df.interpolate(axis=axis_name, method="linear")
+        expected = df.interpolate(axis=axis_number, method="linear")
+        tm.assert_frame_equal(result, expected)
+
+    def test_rowwise_alt(self):
+        df = DataFrame(
+            {
+                0: [0, 0.5, 1.0, np.nan, 4, 8, np.nan, np.nan, 64],
+                1: [1, 2, 3, 4, 3, 2, 1, 0, -1],
+            }
+        )
+        df.interpolate(axis=0)
+        # TODO: assert something?
+
+    @pytest.mark.parametrize(
+        "check_scipy", [False, pytest.param(True, marks=td.skip_if_no("scipy"))]
+    )
+    def test_interp_leading_nans(self, check_scipy):
+        df = DataFrame(
+            {"A": [np.nan, np.nan, 0.5, 0.25, 0], "B": [np.nan, -3, -3.5, np.nan, -4]}
+        )
+        result = df.interpolate()
+        expected = df.copy()
+        expected.loc[3, "B"] = -3.75
+        tm.assert_frame_equal(result, expected)
+
+        if check_scipy:
+            result = df.interpolate(method="polynomial", order=1)
+            tm.assert_frame_equal(result, expected)
+
+    def test_interp_raise_on_only_mixed(self, axis):
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 4],
+                "B": ["a", "b", "c", "d"],
+                "C": [np.nan, 2, 5, 7],
+                "D": [np.nan, np.nan, 9, 9],
+                "E": [1, 2, 3, 4],
+            }
+        )
+        msg = "DataFrame cannot interpolate with object dtype"
+        with pytest.raises(TypeError, match=msg):
+            df.astype("object").interpolate(axis=axis)
+
+    def test_interp_raise_on_all_object_dtype(self):
+        # GH 22985
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, dtype="object")
+        msg = "DataFrame cannot interpolate with object dtype"
+        with pytest.raises(TypeError, match=msg):
+            df.interpolate()
+
+    def test_interp_inplace(self):
+        df = DataFrame({"a": [1.0, 2.0, np.nan, 4.0]})
+        expected = df.copy()
+        result = df.copy()
+
+        with tm.raises_chained_assignment_error():
+            return_value = result["a"].interpolate(inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_interp_inplace_row(self):
+        # GH 10395
+        result = DataFrame(
+            {"a": [1.0, 2.0, 3.0, 4.0], "b": [np.nan, 2.0, 3.0, 4.0], "c": [3, 2, 2, 2]}
+        )
+        expected = result.interpolate(method="linear", axis=1, inplace=False)
+        return_value = result.interpolate(method="linear", axis=1, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_interp_ignore_all_good(self):
+        # GH
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 4],
+                "B": [1, 2, 3, 4],
+                "C": [1.0, 2.0, np.nan, 4.0],
+                "D": [1.0, 2.0, 3.0, 4.0],
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": np.array([1, 2, 3, 4], dtype="float64"),
+                "B": np.array([1, 2, 3, 4], dtype="int64"),
+                "C": np.array([1.0, 2.0, 3, 4.0], dtype="float64"),
+                "D": np.array([1.0, 2.0, 3.0, 4.0], dtype="float64"),
+            }
+        )
+        result = df.interpolate()
+        tm.assert_frame_equal(result, expected)
+
+        # all good
+        result = df[["B", "D"]].interpolate()
+        tm.assert_frame_equal(result, df[["B", "D"]])
+
+    def test_interp_time_inplace_axis(self):
+        # GH 9687
+        periods = 5
+        idx = date_range(start="2014-01-01", periods=periods)
+        data = np.random.default_rng(2).random((periods, periods))
+        data[data < 0.5] = np.nan
+        expected = DataFrame(index=idx, columns=idx, data=data)
+
+        result = expected.interpolate(axis=0, method="time")
+        return_value = expected.interpolate(axis=0, method="time", inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("axis_name, axis_number", [("index", 0), ("columns", 1)])
+    def test_interp_string_axis(self, axis_name, axis_number):
+        # https://github.com/pandas-dev/pandas/issues/25190
+        x = np.linspace(0, 100, 1000)
+        y = np.sin(x)
+        df = DataFrame(
+            data=np.tile(y, (10, 1)), index=np.arange(10), columns=x
+        ).reindex(columns=x * 1.005)
+        result = df.interpolate(method="linear", axis=axis_name)
+        expected = df.interpolate(method="linear", axis=axis_number)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("multiblock", [True, False])
+    @pytest.mark.parametrize("method", ["ffill", "bfill", "pad"])
+    def test_interp_fillna_methods(self, axis, multiblock, method):
+        # GH 12918
+        df = DataFrame(
+            {
+                "A": [1.0, 2.0, 3.0, 4.0, np.nan, 5.0],
+                "B": [2.0, 4.0, 6.0, np.nan, 8.0, 10.0],
+                "C": [3.0, 6.0, 9.0, np.nan, np.nan, 30.0],
+            }
+        )
+        if multiblock:
+            df["D"] = np.nan
+            df["E"] = 1.0
+
+        msg = f"Can not interpolate with method={method}"
+        with pytest.raises(ValueError, match=msg):
+            df.interpolate(method=method, axis=axis)
+
+    def test_interpolate_empty_df(self):
+        # GH#53199
+        df = DataFrame()
+        expected = df.copy()
+        result = df.interpolate(inplace=True)
+        assert result is None
+        tm.assert_frame_equal(df, expected)
+
+    def test_interpolate_ea(self, any_int_ea_dtype):
+        # GH#55347
+        df = DataFrame({"a": [1, None, None, None, 3]}, dtype=any_int_ea_dtype)
+        orig = df.copy()
+        result = df.interpolate(limit=2)
+        expected = DataFrame({"a": [1, 1.5, 2.0, None, 3]}, dtype="Float64")
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(df, orig)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "Float64",
+            "Float32",
+            pytest.param("float32[pyarrow]", marks=td.skip_if_no("pyarrow")),
+            pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        ],
+    )
+    def test_interpolate_ea_float(self, dtype):
+        # GH#55347
+        df = DataFrame({"a": [1, None, None, None, 3]}, dtype=dtype)
+        orig = df.copy()
+        result = df.interpolate(limit=2)
+        expected = DataFrame({"a": [1, 1.5, 2.0, None, 3]}, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(df, orig)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        ["int64", "uint64", "int32", "int16", "int8", "uint32", "uint16", "uint8"],
+    )
+    def test_interpolate_arrow(self, dtype):
+        # GH#55347
+        pytest.importorskip("pyarrow")
+        df = DataFrame({"a": [1, None, None, None, 3]}, dtype=dtype + "[pyarrow]")
+        result = df.interpolate(limit=2)
+        expected = DataFrame({"a": [1, 1.5, 2.0, None, 3]}, dtype="float64[pyarrow]")
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_is_homogeneous_dtype.py b/pandas/pandas/tests/frame/methods/test_is_homogeneous_dtype.py
new file mode 100644
index 0000000000000..086986702d24f
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_is_homogeneous_dtype.py
@@ -0,0 +1,53 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+)
+
+
+@pytest.mark.parametrize(
+    "data, expected",
+    [
+        # empty
+        (DataFrame(), True),
+        # multi-same
+        (DataFrame({"A": [1, 2], "B": [1, 2]}), True),
+        # multi-object
+        (
+            DataFrame(
+                {
+                    "A": np.array([1, 2], dtype=object),
+                    "B": np.array(["a", "b"], dtype=object),
+                },
+                dtype="object",
+            ),
+            True,
+        ),
+        # multi-extension
+        (
+            DataFrame({"A": Categorical(["a", "b"]), "B": Categorical(["a", "b"])}),
+            True,
+        ),
+        # differ types
+        (DataFrame({"A": [1, 2], "B": [1.0, 2.0]}), False),
+        # differ sizes
+        (
+            DataFrame(
+                {
+                    "A": np.array([1, 2], dtype=np.int32),
+                    "B": np.array([1, 2], dtype=np.int64),
+                }
+            ),
+            False,
+        ),
+        # multi-extension differ
+        (
+            DataFrame({"A": Categorical(["a", "b"]), "B": Categorical(["b", "c"])}),
+            False,
+        ),
+    ],
+)
+def test_is_homogeneous_type(data, expected):
+    assert data._is_homogeneous_type is expected
diff --git a/pandas/pandas/tests/frame/methods/test_isetitem.py b/pandas/pandas/tests/frame/methods/test_isetitem.py
new file mode 100644
index 0000000000000..69f394afb6519
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_isetitem.py
@@ -0,0 +1,50 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameSetItem:
+    def test_isetitem_ea_df(self):
+        # GH#49922
+        df = DataFrame([[1, 2, 3], [4, 5, 6]])
+        rhs = DataFrame([[11, 12], [13, 14]], dtype="Int64")
+
+        df.isetitem([0, 1], rhs)
+        expected = DataFrame(
+            {
+                0: Series([11, 13], dtype="Int64"),
+                1: Series([12, 14], dtype="Int64"),
+                2: [3, 6],
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_isetitem_ea_df_scalar_indexer(self):
+        # GH#49922
+        df = DataFrame([[1, 2, 3], [4, 5, 6]])
+        rhs = DataFrame([[11], [13]], dtype="Int64")
+
+        df.isetitem(2, rhs)
+        expected = DataFrame(
+            {
+                0: [1, 4],
+                1: [2, 5],
+                2: Series([11, 13], dtype="Int64"),
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_isetitem_dimension_mismatch(self):
+        # GH#51701
+        df = DataFrame({"a": [1, 2], "b": [3, 4], "c": [5, 6]})
+        value = df.copy()
+        with pytest.raises(ValueError, match="Got 2 positions but value has 3 columns"):
+            df.isetitem([1, 2], value)
+
+        value = df.copy()
+        with pytest.raises(ValueError, match="Got 2 positions but value has 1 columns"):
+            df.isetitem([1, 2], value[["a"]])
diff --git a/pandas/pandas/tests/frame/methods/test_isin.py b/pandas/pandas/tests/frame/methods/test_isin.py
new file mode 100644
index 0000000000000..b4511aad27a93
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_isin.py
@@ -0,0 +1,227 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameIsIn:
+    def test_isin(self):
+        # GH#4211
+        df = DataFrame(
+            {
+                "vals": [1, 2, 3, 4],
+                "ids": ["a", "b", "f", "n"],
+                "ids2": ["a", "n", "c", "n"],
+            },
+            index=["foo", "bar", "baz", "qux"],
+        )
+        other = ["a", "b", "c"]
+
+        result = df.isin(other)
+        expected = DataFrame([df.loc[s].isin(other) for s in df.index])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("empty", [[], Series(dtype=object), np.array([])])
+    def test_isin_empty(self, empty):
+        # GH#16991
+        df = DataFrame({"A": ["a", "b", "c"], "B": ["a", "e", "f"]})
+        expected = DataFrame(False, df.index, df.columns)
+
+        result = df.isin(empty)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_dict(self):
+        df = DataFrame({"A": ["a", "b", "c"], "B": ["a", "e", "f"]})
+        d = {"A": ["a"]}
+
+        expected = DataFrame(False, df.index, df.columns)
+        expected.loc[0, "A"] = True
+
+        result = df.isin(d)
+        tm.assert_frame_equal(result, expected)
+
+        # non unique columns
+        df = DataFrame({"A": ["a", "b", "c"], "B": ["a", "e", "f"]})
+        df.columns = ["A", "A"]
+        expected = DataFrame(False, df.index, df.columns)
+        expected.loc[0, "A"] = True
+        result = df.isin(d)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_with_string_scalar(self):
+        # GH#4763
+        df = DataFrame(
+            {
+                "vals": [1, 2, 3, 4],
+                "ids": ["a", "b", "f", "n"],
+                "ids2": ["a", "n", "c", "n"],
+            },
+            index=["foo", "bar", "baz", "qux"],
+        )
+        msg = (
+            r"only list-like or dict-like objects are allowed "
+            r"to be passed to DataFrame.isin\(\), you passed a 'str'"
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.isin("a")
+
+        with pytest.raises(TypeError, match=msg):
+            df.isin("aaa")
+
+    def test_isin_df(self):
+        df1 = DataFrame({"A": [1, 2, 3, 4], "B": [2, np.nan, 4, 4]})
+        df2 = DataFrame({"A": [0, 2, 12, 4], "B": [2, np.nan, 4, 5]})
+        expected = DataFrame(False, df1.index, df1.columns)
+        result = df1.isin(df2)
+        expected.loc[[1, 3], "A"] = True
+        expected.loc[[0, 2], "B"] = True
+        tm.assert_frame_equal(result, expected)
+
+        # partial overlapping columns
+        df2.columns = ["A", "C"]
+        result = df1.isin(df2)
+        expected["B"] = False
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_tuples(self):
+        # GH#16394
+        df = DataFrame({"A": [1, 2, 3], "B": ["a", "b", "f"]})
+        df["C"] = list(zip(df["A"], df["B"]))
+        result = df["C"].isin([(1, "a")])
+        tm.assert_series_equal(result, Series([True, False, False], name="C"))
+
+    def test_isin_df_dupe_values(self):
+        df1 = DataFrame({"A": [1, 2, 3, 4], "B": [2, np.nan, 4, 4]})
+        # just cols duped
+        df2 = DataFrame([[0, 2], [12, 4], [2, np.nan], [4, 5]], columns=["B", "B"])
+        msg = r"cannot compute isin with a duplicate axis\."
+        with pytest.raises(ValueError, match=msg):
+            df1.isin(df2)
+
+        # just index duped
+        df2 = DataFrame(
+            [[0, 2], [12, 4], [2, np.nan], [4, 5]],
+            columns=["A", "B"],
+            index=[0, 0, 1, 1],
+        )
+        with pytest.raises(ValueError, match=msg):
+            df1.isin(df2)
+
+        # cols and index:
+        df2.columns = ["B", "B"]
+        with pytest.raises(ValueError, match=msg):
+            df1.isin(df2)
+
+    def test_isin_dupe_self(self):
+        other = DataFrame({"A": [1, 0, 1, 0], "B": [1, 1, 0, 0]})
+        df = DataFrame([[1, 1], [1, 0], [0, 0]], columns=["A", "A"])
+        result = df.isin(other)
+        expected = DataFrame(False, index=df.index, columns=df.columns)
+        expected.loc[0] = True
+        expected.iloc[1, 1] = True
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_against_series(self):
+        df = DataFrame(
+            {"A": [1, 2, 3, 4], "B": [2, np.nan, 4, 4]}, index=["a", "b", "c", "d"]
+        )
+        s = Series([1, 3, 11, 4], index=["a", "b", "c", "d"])
+        expected = DataFrame(False, index=df.index, columns=df.columns)
+        expected.loc["a", "A"] = True
+        expected.loc["d"] = True
+        result = df.isin(s)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_multiIndex(self):
+        idx = MultiIndex.from_tuples(
+            [
+                (0, "a", "foo"),
+                (0, "a", "bar"),
+                (0, "b", "bar"),
+                (0, "b", "baz"),
+                (2, "a", "foo"),
+                (2, "a", "bar"),
+                (2, "c", "bar"),
+                (2, "c", "baz"),
+                (1, "b", "foo"),
+                (1, "b", "bar"),
+                (1, "c", "bar"),
+                (1, "c", "baz"),
+            ]
+        )
+        df1 = DataFrame({"A": np.ones(12), "B": np.zeros(12)}, index=idx)
+        df2 = DataFrame(
+            {
+                "A": [1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1],
+                "B": [1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 1, 1],
+            }
+        )
+        # against regular index
+        expected = DataFrame(False, index=df1.index, columns=df1.columns)
+        result = df1.isin(df2)
+        tm.assert_frame_equal(result, expected)
+
+        df2.index = idx
+        expected = df2.values.astype(bool)
+        expected[:, 1] = ~expected[:, 1]
+        expected = DataFrame(expected, columns=["A", "B"], index=idx)
+
+        result = df1.isin(df2)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_empty_datetimelike(self):
+        # GH#15473
+        df1_ts = DataFrame({"date": pd.to_datetime(["2014-01-01", "2014-01-02"])})
+        df1_td = DataFrame({"date": [pd.Timedelta(1, "s"), pd.Timedelta(2, "s")]})
+        df2 = DataFrame({"date": []})
+        df3 = DataFrame()
+
+        expected = DataFrame({"date": [False, False]})
+
+        result = df1_ts.isin(df2)
+        tm.assert_frame_equal(result, expected)
+        result = df1_ts.isin(df3)
+        tm.assert_frame_equal(result, expected)
+
+        result = df1_td.isin(df2)
+        tm.assert_frame_equal(result, expected)
+        result = df1_td.isin(df3)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            DataFrame({"a": [1, 2, 3]}, dtype="category"),
+            Series([1, 2, 3], dtype="category"),
+        ],
+    )
+    def test_isin_category_frame(self, values):
+        # GH#34256
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        expected = DataFrame({"a": [True, True, True], "b": [False, False, False]})
+
+        result = df.isin(values)
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_read_only(self):
+        # https://github.com/pandas-dev/pandas/issues/37174
+        arr = np.array([1, 2, 3])
+        arr.setflags(write=False)
+        df = DataFrame([1, 2, 3])
+        result = df.isin(arr)
+        expected = DataFrame([True, True, True])
+        tm.assert_frame_equal(result, expected)
+
+    def test_isin_not_lossy(self):
+        # GH 53514
+        val = 1666880195890293744
+        df = DataFrame({"a": [val], "b": [1.0]})
+        result = df.isin([val])
+        expected = DataFrame({"a": [True], "b": [False]})
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_iterrows.py b/pandas/pandas/tests/frame/methods/test_iterrows.py
new file mode 100644
index 0000000000000..0bd0bed76dc9d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_iterrows.py
@@ -0,0 +1,16 @@
+from pandas import (
+    DataFrame,
+    Timedelta,
+)
+
+
+def test_no_overflow_of_freq_and_time_in_dataframe():
+    # GH 35665
+    df = DataFrame(
+        {
+            "some_string": ["2222Y3"],
+            "time": [Timedelta("0 days 00:00:00.990000")],
+        }
+    )
+    for _, row in df.iterrows():
+        assert row.dtype == "object"
diff --git a/pandas/pandas/tests/frame/methods/test_join.py b/pandas/pandas/tests/frame/methods/test_join.py
new file mode 100644
index 0000000000000..aaa9485cab580
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_join.py
@@ -0,0 +1,577 @@
+from datetime import datetime
+import re
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas.errors import MergeError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core.reshape.concat import concat
+
+
+@pytest.fixture
+def left_no_dup():
+    return DataFrame(
+        {"a": ["a", "b", "c", "d"], "b": ["cat", "dog", "weasel", "horse"]},
+        index=range(4),
+    )
+
+
+@pytest.fixture
+def right_no_dup():
+    return DataFrame(
+        {
+            "a": ["a", "b", "c", "d", "e"],
+            "c": ["meow", "bark", "um... weasel noise?", "nay", "chirp"],
+        },
+        index=range(5),
+    ).set_index("a")
+
+
+@pytest.fixture
+def left_w_dups(left_no_dup):
+    return concat(
+        [left_no_dup, DataFrame({"a": ["a"], "b": ["cow"]}, index=[3])], sort=True
+    )
+
+
+@pytest.fixture
+def right_w_dups(right_no_dup):
+    return concat(
+        [right_no_dup, DataFrame({"a": ["e"], "c": ["moo"]}, index=[3])]
+    ).set_index("a")
+
+
+@pytest.mark.parametrize(
+    "how, sort, expected",
+    [
+        ("inner", False, DataFrame({"a": [20, 10], "b": [200, 100]}, index=[2, 1])),
+        ("inner", True, DataFrame({"a": [10, 20], "b": [100, 200]}, index=[1, 2])),
+        (
+            "left",
+            False,
+            DataFrame({"a": [20, 10, 0], "b": [200, 100, np.nan]}, index=[2, 1, 0]),
+        ),
+        (
+            "left",
+            True,
+            DataFrame({"a": [0, 10, 20], "b": [np.nan, 100, 200]}, index=[0, 1, 2]),
+        ),
+        (
+            "right",
+            False,
+            DataFrame({"a": [np.nan, 10, 20], "b": [300, 100, 200]}, index=[3, 1, 2]),
+        ),
+        (
+            "right",
+            True,
+            DataFrame({"a": [10, 20, np.nan], "b": [100, 200, 300]}, index=[1, 2, 3]),
+        ),
+        (
+            "outer",
+            False,
+            DataFrame(
+                {"a": [0, 10, 20, np.nan], "b": [np.nan, 100, 200, 300]},
+                index=[0, 1, 2, 3],
+            ),
+        ),
+        (
+            "outer",
+            True,
+            DataFrame(
+                {"a": [0, 10, 20, np.nan], "b": [np.nan, 100, 200, 300]},
+                index=[0, 1, 2, 3],
+            ),
+        ),
+    ],
+)
+def test_join(how, sort, expected):
+    left = DataFrame({"a": [20, 10, 0]}, index=[2, 1, 0])
+    right = DataFrame({"b": [300, 100, 200]}, index=[3, 1, 2])
+    result = left.join(right, how=how, sort=sort, validate="1:1")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_suffix_on_list_join():
+    first = DataFrame({"key": [1, 2, 3, 4, 5]})
+    second = DataFrame({"key": [1, 8, 3, 2, 5], "v1": [1, 2, 3, 4, 5]})
+    third = DataFrame({"keys": [5, 2, 3, 4, 1], "v2": [1, 2, 3, 4, 5]})
+
+    # check proper errors are raised
+    msg = "Suffixes not supported when joining multiple DataFrames"
+    with pytest.raises(ValueError, match=msg):
+        first.join([second], lsuffix="y")
+    with pytest.raises(ValueError, match=msg):
+        first.join([second, third], rsuffix="x")
+    with pytest.raises(ValueError, match=msg):
+        first.join([second, third], lsuffix="y", rsuffix="x")
+    with pytest.raises(ValueError, match="Indexes have overlapping values"):
+        first.join([second, third])
+
+    # no errors should be raised
+    arr_joined = first.join([third])
+    norm_joined = first.join(third)
+    tm.assert_frame_equal(arr_joined, norm_joined)
+
+
+def test_join_invalid_validate(left_no_dup, right_no_dup):
+    # GH 46622
+    # Check invalid arguments
+    msg = (
+        '"invalid" is not a valid argument. '
+        "Valid arguments are:\n"
+        '- "1:1"\n'
+        '- "1:m"\n'
+        '- "m:1"\n'
+        '- "m:m"\n'
+        '- "one_to_one"\n'
+        '- "one_to_many"\n'
+        '- "many_to_one"\n'
+        '- "many_to_many"'
+    )
+    with pytest.raises(ValueError, match=msg):
+        left_no_dup.merge(right_no_dup, on="a", validate="invalid")
+
+
+@pytest.mark.parametrize("dtype", ["object", "string[pyarrow]"])
+def test_join_on_single_col_dup_on_right(left_no_dup, right_w_dups, dtype):
+    # GH 46622
+    # Dups on right allowed by one_to_many constraint
+    if dtype == "string[pyarrow]":
+        pytest.importorskip("pyarrow")
+    left_no_dup = left_no_dup.astype(dtype)
+    right_w_dups.index = right_w_dups.index.astype(dtype)
+    left_no_dup.join(
+        right_w_dups,
+        on="a",
+        validate="one_to_many",
+    )
+
+    # Dups on right not allowed by one_to_one constraint
+    msg = "Merge keys are not unique in right dataset; not a one-to-one merge"
+    with pytest.raises(MergeError, match=msg):
+        left_no_dup.join(
+            right_w_dups,
+            on="a",
+            validate="one_to_one",
+        )
+
+
+def test_join_on_single_col_dup_on_left(left_w_dups, right_no_dup):
+    # GH 46622
+    # Dups on left allowed by many_to_one constraint
+    left_w_dups.join(
+        right_no_dup,
+        on="a",
+        validate="many_to_one",
+    )
+
+    # Dups on left not allowed by one_to_one constraint
+    msg = "Merge keys are not unique in left dataset; not a one-to-one merge"
+    with pytest.raises(MergeError, match=msg):
+        left_w_dups.join(
+            right_no_dup,
+            on="a",
+            validate="one_to_one",
+        )
+
+
+def test_join_on_single_col_dup_on_both(left_w_dups, right_w_dups):
+    # GH 46622
+    # Dups on both allowed by many_to_many constraint
+    left_w_dups.join(right_w_dups, on="a", validate="many_to_many")
+
+    # Dups on both not allowed by many_to_one constraint
+    msg = "Merge keys are not unique in right dataset; not a many-to-one merge"
+    with pytest.raises(MergeError, match=msg):
+        left_w_dups.join(
+            right_w_dups,
+            on="a",
+            validate="many_to_one",
+        )
+
+    # Dups on both not allowed by one_to_many constraint
+    msg = "Merge keys are not unique in left dataset; not a one-to-many merge"
+    with pytest.raises(MergeError, match=msg):
+        left_w_dups.join(
+            right_w_dups,
+            on="a",
+            validate="one_to_many",
+        )
+
+
+def test_join_on_multi_col_check_dup():
+    # GH 46622
+    # Two column join, dups in both, but jointly no dups
+    left = DataFrame(
+        {
+            "a": ["a", "a", "b", "b"],
+            "b": [0, 1, 0, 1],
+            "c": ["cat", "dog", "weasel", "horse"],
+        },
+        index=range(4),
+    ).set_index(["a", "b"])
+
+    right = DataFrame(
+        {
+            "a": ["a", "a", "b"],
+            "b": [0, 1, 0],
+            "d": ["meow", "bark", "um... weasel noise?"],
+        },
+        index=range(3),
+    ).set_index(["a", "b"])
+
+    expected_multi = DataFrame(
+        {
+            "a": ["a", "a", "b"],
+            "b": [0, 1, 0],
+            "c": ["cat", "dog", "weasel"],
+            "d": ["meow", "bark", "um... weasel noise?"],
+        },
+        index=range(3),
+    ).set_index(["a", "b"])
+
+    # Jointly no dups allowed by one_to_one constraint
+    result = left.join(right, how="inner", validate="1:1")
+    tm.assert_frame_equal(result, expected_multi)
+
+
+def test_join_index(float_frame):
+    # left / right
+
+    f = float_frame.loc[float_frame.index[:10], ["A", "B"]]
+    f2 = float_frame.loc[float_frame.index[5:], ["C", "D"]].iloc[::-1]
+
+    joined = f.join(f2)
+    tm.assert_index_equal(f.index, joined.index)
+    expected_columns = Index(["A", "B", "C", "D"])
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    joined = f.join(f2, how="left")
+    tm.assert_index_equal(joined.index, f.index)
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    joined = f.join(f2, how="right")
+    tm.assert_index_equal(joined.index, f2.index)
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    # inner
+
+    joined = f.join(f2, how="inner")
+    tm.assert_index_equal(joined.index, f.index[5:10])
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    # outer
+
+    joined = f.join(f2, how="outer")
+    tm.assert_index_equal(joined.index, float_frame.index.sort_values())
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    # left anti
+    joined = f.join(f2, how="left_anti")
+    tm.assert_index_equal(joined.index, float_frame.index[:5])
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    # right anti
+    joined = f.join(f2, how="right_anti")
+    tm.assert_index_equal(joined.index, float_frame.index[10:][::-1])
+    tm.assert_index_equal(joined.columns, expected_columns)
+
+    join_msg = (
+        "'foo' is not a valid Merge type: left, right, inner, outer, "
+        "left_anti, right_anti, cross, asof"
+    )
+    with pytest.raises(ValueError, match=re.escape(join_msg)):
+        f.join(f2, how="foo")
+
+    # corner case - overlapping columns
+    msg = "columns overlap but no suffix"
+    for how in ("outer", "left", "inner"):
+        with pytest.raises(ValueError, match=msg):
+            float_frame.join(float_frame, how=how)
+
+
+def test_join_index_more(float_frame):
+    af = float_frame.loc[:, ["A", "B"]]
+    bf = float_frame.loc[::2, ["C", "D"]]
+
+    expected = af.copy()
+    expected["C"] = float_frame["C"][::2]
+    expected["D"] = float_frame["D"][::2]
+
+    result = af.join(bf)
+    tm.assert_frame_equal(result, expected)
+
+    result = af.join(bf, how="right")
+    tm.assert_frame_equal(result, expected[::2])
+
+    result = bf.join(af, how="right")
+    tm.assert_frame_equal(result, expected.loc[:, result.columns])
+
+
+def test_join_index_series(float_frame):
+    df = float_frame.copy()
+    ser = df.pop(float_frame.columns[-1])
+    joined = df.join(ser)
+
+    tm.assert_frame_equal(joined, float_frame)
+
+    ser.name = None
+    with pytest.raises(ValueError, match="must have a name"):
+        df.join(ser)
+
+
+def test_join_overlap(float_frame):
+    df1 = float_frame.loc[:, ["A", "B", "C"]]
+    df2 = float_frame.loc[:, ["B", "C", "D"]]
+
+    joined = df1.join(df2, lsuffix="_df1", rsuffix="_df2")
+    df1_suf = df1.loc[:, ["B", "C"]].add_suffix("_df1")
+    df2_suf = df2.loc[:, ["B", "C"]].add_suffix("_df2")
+
+    no_overlap = float_frame.loc[:, ["A", "D"]]
+    expected = df1_suf.join(df2_suf).join(no_overlap)
+
+    # column order not necessarily sorted
+    tm.assert_frame_equal(joined, expected.loc[:, joined.columns])
+
+
+def test_join_period_index():
+    frame_with_period_index = DataFrame(
+        data=np.arange(20).reshape(4, 5),
+        columns=list("abcde"),
+        index=period_range(start="2000", freq="Y", periods=4),
+    )
+    other = frame_with_period_index.rename(columns=lambda key: f"{key}{key}")
+
+    joined_values = np.concatenate([frame_with_period_index.values] * 2, axis=1)
+
+    joined_cols = frame_with_period_index.columns.append(other.columns)
+
+    joined = frame_with_period_index.join(other)
+    expected = DataFrame(
+        data=joined_values, columns=joined_cols, index=frame_with_period_index.index
+    )
+
+    tm.assert_frame_equal(joined, expected)
+
+
+def test_join_left_sequence_non_unique_index():
+    # https://github.com/pandas-dev/pandas/issues/19607
+    df1 = DataFrame({"a": [0, 10, 20]}, index=[1, 2, 3])
+    df2 = DataFrame({"b": [100, 200, 300]}, index=[4, 3, 2])
+    df3 = DataFrame({"c": [400, 500, 600]}, index=[2, 2, 4])
+
+    joined = df1.join([df2, df3], how="left")
+
+    expected = DataFrame(
+        {
+            "a": [0, 10, 10, 20],
+            "b": [np.nan, 300, 300, 200],
+            "c": [np.nan, 400, 500, np.nan],
+        },
+        index=[1, 2, 2, 3],
+    )
+
+    tm.assert_frame_equal(joined, expected)
+
+
+def test_join_list_series(float_frame):
+    # GH#46850
+    # Join a DataFrame with a list containing both a Series and a DataFrame
+    left = float_frame.A.to_frame()
+    right = [float_frame.B, float_frame[["C", "D"]]]
+    result = left.join(right)
+    tm.assert_frame_equal(result, float_frame)
+
+
+def test_suppress_future_warning_with_sort_kw(sort):
+    sort_kw = sort
+    a = DataFrame({"col1": [1, 2]}, index=["c", "a"])
+
+    b = DataFrame({"col2": [4, 5]}, index=["b", "a"])
+
+    c = DataFrame({"col3": [7, 8]}, index=["a", "b"])
+
+    expected = DataFrame(
+        {
+            "col1": {"a": 2.0, "b": float("nan"), "c": 1.0},
+            "col2": {"a": 5.0, "b": 4.0, "c": float("nan")},
+            "col3": {"a": 7.0, "b": 8.0, "c": float("nan")},
+        }
+    )
+    if sort_kw is False:
+        expected = expected.reindex(index=["c", "a", "b"])
+
+    with tm.assert_produces_warning(None):
+        result = a.join([b, c], how="outer", sort=sort_kw)
+    tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameJoin:
+    def test_join(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        a = frame.loc[frame.index[:5], ["A"]]
+        b = frame.loc[frame.index[2:], ["B", "C"]]
+
+        joined = a.join(b, how="outer").reindex(frame.index)
+        expected = frame.copy().values.copy()
+        expected[np.isnan(joined.values)] = np.nan
+        expected = DataFrame(expected, index=frame.index, columns=frame.columns)
+
+        assert not np.isnan(joined.values).all()
+
+        tm.assert_frame_equal(joined, expected)
+
+    def test_join_segfault(self):
+        # GH#1532
+        df1 = DataFrame({"a": [1, 1], "b": [1, 2], "x": [1, 2]})
+        df2 = DataFrame({"a": [2, 2], "b": [1, 2], "y": [1, 2]})
+        df1 = df1.set_index(["a", "b"])
+        df2 = df2.set_index(["a", "b"])
+        # it works!
+        for how in ["left", "right", "outer"]:
+            df1.join(df2, how=how)
+
+    def test_join_str_datetime(self):
+        str_dates = ["20120209", "20120222"]
+        dt_dates = [datetime(2012, 2, 9), datetime(2012, 2, 22)]
+
+        A = DataFrame(str_dates, index=range(2), columns=["aa"])
+        C = DataFrame([[1, 2], [3, 4]], index=str_dates, columns=dt_dates)
+
+        tst = A.join(C, on="aa")
+
+        assert len(tst.columns) == 3
+
+    def test_join_multiindex_leftright(self):
+        # GH 10741
+        df1 = DataFrame(
+            [
+                ["a", "x", 0.471780],
+                ["a", "y", 0.774908],
+                ["a", "z", 0.563634],
+                ["b", "x", -0.353756],
+                ["b", "y", 0.368062],
+                ["b", "z", -1.721840],
+                ["c", "x", 1],
+                ["c", "y", 2],
+                ["c", "z", 3],
+            ],
+            columns=["first", "second", "value1"],
+        ).set_index(["first", "second"])
+
+        df2 = DataFrame([["a", 10], ["b", 20]], columns=["first", "value2"]).set_index(
+            ["first"]
+        )
+
+        exp = DataFrame(
+            [
+                [0.471780, 10],
+                [0.774908, 10],
+                [0.563634, 10],
+                [-0.353756, 20],
+                [0.368062, 20],
+                [-1.721840, 20],
+                [1.000000, np.nan],
+                [2.000000, np.nan],
+                [3.000000, np.nan],
+            ],
+            index=df1.index,
+            columns=["value1", "value2"],
+        )
+
+        # these must be the same results (but columns are flipped)
+        tm.assert_frame_equal(df1.join(df2, how="left"), exp)
+        tm.assert_frame_equal(df2.join(df1, how="right"), exp[["value2", "value1"]])
+
+        exp_idx = MultiIndex.from_product(
+            [["a", "b"], ["x", "y", "z"]], names=["first", "second"]
+        )
+        exp = DataFrame(
+            [
+                [0.471780, 10],
+                [0.774908, 10],
+                [0.563634, 10],
+                [-0.353756, 20],
+                [0.368062, 20],
+                [-1.721840, 20],
+            ],
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        tm.assert_frame_equal(df1.join(df2, how="right"), exp)
+        tm.assert_frame_equal(df2.join(df1, how="left"), exp[["value2", "value1"]])
+
+    def test_join_multiindex_dates(self):
+        # GH 33692
+        date = pd.Timestamp(2000, 1, 1).date()
+
+        df1_index = MultiIndex.from_tuples([(0, date)], names=["index_0", "date"])
+        df1 = DataFrame({"col1": [0]}, index=df1_index)
+        df2_index = MultiIndex.from_tuples([(0, date)], names=["index_0", "date"])
+        df2 = DataFrame({"col2": [0]}, index=df2_index)
+        df3_index = MultiIndex.from_tuples([(0, date)], names=["index_0", "date"])
+        df3 = DataFrame({"col3": [0]}, index=df3_index)
+
+        result = df1.join([df2, df3])
+
+        expected_index = MultiIndex.from_tuples([(0, date)], names=["index_0", "date"])
+        expected = DataFrame(
+            {"col1": [0], "col2": [0], "col3": [0]}, index=expected_index
+        )
+
+        tm.assert_equal(result, expected)
+
+    def test_merge_join_different_levels_raises(self):
+        # GH#9455
+        # GH 40993: For raising, enforced in 2.0
+
+        # first dataframe
+        df1 = DataFrame(columns=["a", "b"], data=[[1, 11], [0, 22]])
+
+        # second dataframe
+        columns = MultiIndex.from_tuples([("a", ""), ("c", "c1")])
+        df2 = DataFrame(columns=columns, data=[[1, 33], [0, 44]])
+
+        # merge
+        with pytest.raises(
+            MergeError, match="Not allowed to merge between different levels"
+        ):
+            pd.merge(df1, df2, on="a")
+
+        # join, see discussion in GH#12219
+        with pytest.raises(
+            MergeError, match="Not allowed to merge between different levels"
+        ):
+            df1.join(df2, on="a")
+
+    def test_frame_join_tzaware(self):
+        tz = zoneinfo.ZoneInfo("US/Central")
+        test1 = DataFrame(
+            np.zeros((6, 3)),
+            index=date_range("2012-11-15 00:00:00", periods=6, freq="100ms", tz=tz),
+        )
+        test2 = DataFrame(
+            np.zeros((3, 3)),
+            index=date_range("2012-11-15 00:00:00", periods=3, freq="250ms", tz=tz),
+            columns=range(3, 6),
+        )
+
+        result = test1.join(test2, how="outer")
+        expected = test1.index.union(test2.index)
+
+        tm.assert_index_equal(result.index, expected)
+        assert result.index.tz.key == "US/Central"
diff --git a/pandas/pandas/tests/frame/methods/test_map.py b/pandas/pandas/tests/frame/methods/test_map.py
new file mode 100644
index 0000000000000..9850de14b2092
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_map.py
@@ -0,0 +1,208 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import BDay
+
+
+def test_map(float_frame):
+    result = float_frame.map(lambda x: x * 2)
+    tm.assert_frame_equal(result, float_frame * 2)
+    float_frame.map(type)
+
+    # GH 465: function returning tuples
+    result = float_frame.map(lambda x: (x, x))["A"].iloc[0]
+    assert isinstance(result, tuple)
+
+
+@pytest.mark.parametrize("val", [1, 1.0])
+def test_map_float_object_conversion(val):
+    # GH 2909: object conversion to float in constructor?
+    df = DataFrame(data=[val, "a"])
+    result = df.map(lambda x: x).dtypes[0]
+    assert result == object
+
+
+def test_map_keeps_dtype(na_action):
+    # GH52219
+    arr = Series(["a", np.nan, "b"])
+    sparse_arr = arr.astype(pd.SparseDtype(object))
+    df = DataFrame(data={"a": arr, "b": sparse_arr})
+
+    def func(x):
+        return str.upper(x) if not pd.isna(x) else x
+
+    result = df.map(func, na_action=na_action)
+
+    expected_sparse = pd.array(["A", np.nan, "B"], dtype=pd.SparseDtype(object))
+    expected_arr = expected_sparse.astype(object)
+    expected = DataFrame({"a": expected_arr, "b": expected_sparse})
+
+    tm.assert_frame_equal(result, expected)
+
+    result_empty = df.iloc[:0, :].map(func, na_action=na_action)
+    expected_empty = expected.iloc[:0, :]
+    tm.assert_frame_equal(result_empty, expected_empty)
+
+
+def test_map_str():
+    # GH 2786
+    df = DataFrame(np.random.default_rng(2).random((3, 4)))
+    df2 = df.copy()
+    cols = ["a", "a", "a", "a"]
+    df.columns = cols
+
+    expected = df2.map(str)
+    expected.columns = cols
+    result = df.map(str)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "col, val",
+    [["datetime", Timestamp("20130101")], ["timedelta", pd.Timedelta("1 min")]],
+)
+def test_map_datetimelike(col, val):
+    # datetime/timedelta
+    df = DataFrame(np.random.default_rng(2).random((3, 4)))
+    df[col] = val
+    result = df.map(str)
+    assert result.loc[0, col] == str(df.loc[0, col])
+
+
+@pytest.mark.parametrize(
+    "expected",
+    [
+        DataFrame(),
+        DataFrame(columns=list("ABC")),
+        DataFrame(index=list("ABC")),
+        DataFrame({"A": [], "B": [], "C": []}),
+    ],
+)
+@pytest.mark.parametrize("func", [round, lambda x: x])
+def test_map_empty(expected, func):
+    # GH 8222
+    result = expected.map(func)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_map_kwargs():
+    # GH 40652
+    result = DataFrame([[1, 2], [3, 4]]).map(lambda x, y: x + y, y=2)
+    expected = DataFrame([[3, 4], [5, 6]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_map_na_ignore(float_frame):
+    # GH 23803
+    strlen_frame = float_frame.map(lambda x: len(str(x)))
+    float_frame_with_na = float_frame.copy()
+    mask = np.random.default_rng(2).integers(0, 2, size=float_frame.shape, dtype=bool)
+    float_frame_with_na[mask] = pd.NA
+    strlen_frame_na_ignore = float_frame_with_na.map(
+        lambda x: len(str(x)), na_action="ignore"
+    )
+    # Set float64 type to avoid upcast when setting NA below
+    strlen_frame_with_na = strlen_frame.copy().astype("float64")
+    strlen_frame_with_na[mask] = pd.NA
+    tm.assert_frame_equal(strlen_frame_na_ignore, strlen_frame_with_na)
+
+
+def test_map_box_timestamps():
+    # GH 2689, GH 2627
+    ser = Series(date_range("1/1/2000", periods=10))
+
+    def func(x):
+        return (x.hour, x.day, x.month)
+
+    # it works!
+    DataFrame(ser).map(func)
+
+
+def test_map_box():
+    # ufunc will not be boxed. Same test cases as the test_map_box
+    df = DataFrame(
+        {
+            "a": [Timestamp("2011-01-01"), Timestamp("2011-01-02")],
+            "b": [
+                Timestamp("2011-01-01", tz="US/Eastern"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+            ],
+            "c": [pd.Timedelta("1 days"), pd.Timedelta("2 days")],
+            "d": [
+                pd.Period("2011-01-01", freq="M"),
+                pd.Period("2011-01-02", freq="M"),
+            ],
+        }
+    )
+
+    result = df.map(lambda x: type(x).__name__)
+    expected = DataFrame(
+        {
+            "a": ["Timestamp", "Timestamp"],
+            "b": ["Timestamp", "Timestamp"],
+            "c": ["Timedelta", "Timedelta"],
+            "d": ["Period", "Period"],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_map_dont_convert_datetime64(unit):
+    dtype = f"M8[{unit}]"
+    df = DataFrame({"x1": [datetime(1996, 1, 1)]}, dtype=dtype)
+
+    df = df.map(lambda x: x + BDay())
+    df = df.map(lambda x: x + BDay())
+
+    result = df.x1.dtype
+    assert result == dtype
+
+
+def test_map_function_runs_once():
+    df = DataFrame({"a": [1, 2, 3]})
+    values = []  # Save values function is applied to
+
+    def reducing_function(val):
+        values.append(val)
+
+    def non_reducing_function(val):
+        values.append(val)
+        return val
+
+    for func in [reducing_function, non_reducing_function]:
+        del values[:]
+
+        df.map(func)
+        assert values == df.a.to_list()
+
+
+def test_map_type():
+    # GH 46719
+    df = DataFrame(
+        {"col1": [3, "string", float], "col2": [0.25, datetime(2020, 1, 1), np.nan]},
+        index=["a", "b", "c"],
+    )
+
+    result = df.map(type)
+    expected = DataFrame(
+        {"col1": [int, str, type], "col2": [float, datetime, float]},
+        index=["a", "b", "c"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_map_invalid_na_action(float_frame):
+    # GH 23803
+    with pytest.raises(ValueError, match="na_action must be .*Got 'abc'"):
+        float_frame.map(lambda x: len(str(x)), na_action="abc")
diff --git a/pandas/pandas/tests/frame/methods/test_matmul.py b/pandas/pandas/tests/frame/methods/test_matmul.py
new file mode 100644
index 0000000000000..be9462b64fa1b
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_matmul.py
@@ -0,0 +1,98 @@
+import operator
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestMatMul:
+    def test_matmul(self):
+        # matmul test is for GH#10259
+        a = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            index=["a", "b", "c"],
+            columns=["p", "q", "r", "s"],
+        )
+        b = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)),
+            index=["p", "q", "r", "s"],
+            columns=["one", "two"],
+        )
+
+        # DataFrame @ DataFrame
+        result = operator.matmul(a, b)
+        expected = DataFrame(
+            np.dot(a.values, b.values), index=["a", "b", "c"], columns=["one", "two"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # DataFrame @ Series
+        result = operator.matmul(a, b.one)
+        expected = Series(np.dot(a.values, b.one.values), index=["a", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+        # np.array @ DataFrame
+        result = operator.matmul(a.values, b)
+        assert isinstance(result, DataFrame)
+        assert result.columns.equals(b.columns)
+        assert result.index.equals(Index(range(3)))
+        expected = np.dot(a.values, b.values)
+        tm.assert_almost_equal(result.values, expected)
+
+        # nested list @ DataFrame (__rmatmul__)
+        result = operator.matmul(a.values.tolist(), b)
+        expected = DataFrame(
+            np.dot(a.values, b.values), index=["a", "b", "c"], columns=["one", "two"]
+        )
+        tm.assert_almost_equal(result.values, expected.values)
+
+        # mixed dtype DataFrame @ DataFrame
+        a["q"] = a.q.round().astype(int)
+        result = operator.matmul(a, b)
+        expected = DataFrame(
+            np.dot(a.values, b.values), index=["a", "b", "c"], columns=["one", "two"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # different dtypes DataFrame @ DataFrame
+        a = a.astype(int)
+        result = operator.matmul(a, b)
+        expected = DataFrame(
+            np.dot(a.values, b.values), index=["a", "b", "c"], columns=["one", "two"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # unaligned
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            index=[1, 2, 3],
+            columns=range(4),
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=range(5),
+            columns=[1, 2, 3],
+        )
+
+        with pytest.raises(ValueError, match="aligned"):
+            operator.matmul(df, df2)
+
+    def test_matmul_message_shapes(self):
+        # GH#21581 exception message should reflect original shapes,
+        #  not transposed shapes
+        a = np.random.default_rng(2).random((10, 4))
+        b = np.random.default_rng(2).random((5, 3))
+
+        df = DataFrame(b)
+
+        msg = r"shapes \(10, 4\) and \(5, 3\) not aligned"
+        with pytest.raises(ValueError, match=msg):
+            a @ df
+        with pytest.raises(ValueError, match=msg):
+            a.tolist() @ df
diff --git a/pandas/pandas/tests/frame/methods/test_nlargest.py b/pandas/pandas/tests/frame/methods/test_nlargest.py
new file mode 100644
index 0000000000000..c6e5304ae3cb4
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_nlargest.py
@@ -0,0 +1,244 @@
+"""
+Note: for naming purposes, most tests are title with as e.g. "test_nlargest_foo"
+but are implicitly also testing nsmallest_foo.
+"""
+
+from string import ascii_lowercase
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+@pytest.fixture
+def df_main_dtypes():
+    return pd.DataFrame(
+        {
+            "group": [1, 1, 2],
+            "int": [1, 2, 3],
+            "float": [4.0, 5.0, 6.0],
+            "string": list("abc"),
+            "category_string": pd.Series(list("abc")).astype("category"),
+            "category_int": [7, 8, 9],
+            "datetime": pd.date_range("20130101", periods=3),
+            "datetimetz": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+            "timedelta": pd.timedelta_range("1 s", periods=3, freq="s"),
+        },
+        columns=[
+            "group",
+            "int",
+            "float",
+            "string",
+            "category_string",
+            "category_int",
+            "datetime",
+            "datetimetz",
+            "timedelta",
+        ],
+    )
+
+
+class TestNLargestNSmallest:
+    # ----------------------------------------------------------------------
+    # Top / bottom
+    @pytest.mark.parametrize(
+        "order",
+        [
+            ["a"],
+            ["c"],
+            ["a", "b"],
+            ["a", "c"],
+            ["b", "a"],
+            ["b", "c"],
+            ["a", "b", "c"],
+            ["c", "a", "b"],
+            ["c", "b", "a"],
+            ["b", "c", "a"],
+            ["b", "a", "c"],
+            # dups!
+            ["b", "c", "c"],
+        ],
+    )
+    @pytest.mark.parametrize("n", range(1, 11))
+    def test_nlargest_n(self, nselect_method, n, order):
+        # GH#10393
+        df = pd.DataFrame(
+            {
+                "a": np.random.default_rng(2).permutation(10),
+                "b": list(ascii_lowercase[:10]),
+                "c": np.random.default_rng(2).permutation(10).astype("float64"),
+            }
+        )
+        if "b" in order:
+            error_msg = (
+                f"Column 'b' has dtype (object|str), "
+                f"cannot use method '{nselect_method}' with this dtype"
+            )
+            with pytest.raises(TypeError, match=error_msg):
+                getattr(df, nselect_method)(n, order)
+        else:
+            ascending = nselect_method == "nsmallest"
+            result = getattr(df, nselect_method)(n, order)
+            result.index = pd.Index(list(result.index))
+            expected = df.sort_values(order, ascending=ascending).head(n)
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "columns", [["group", "category_string"], ["group", "string"]]
+    )
+    def test_nlargest_error(self, df_main_dtypes, nselect_method, columns):
+        df = df_main_dtypes
+        col = columns[1]
+        error_msg = (
+            f"Column '{col}' has dtype {df[col].dtype}, "
+            f"cannot use method '{nselect_method}' with this dtype"
+        )
+        # escape some characters that may be in the repr
+        error_msg = (
+            error_msg.replace("(", "\\(")
+            .replace(")", "\\)")
+            .replace("[", "\\[")
+            .replace("]", "\\]")
+        )
+        with pytest.raises(TypeError, match=error_msg):
+            getattr(df, nselect_method)(2, columns)
+
+    def test_nlargest_all_dtypes(self, df_main_dtypes):
+        df = df_main_dtypes
+        df.nsmallest(2, list(set(df) - {"category_string", "string"}))
+        df.nlargest(2, list(set(df) - {"category_string", "string"}))
+
+    def test_nlargest_duplicates_on_starter_columns(self):
+        # regression test for GH#22752
+
+        df = pd.DataFrame({"a": [2, 2, 2, 1, 1, 1], "b": [1, 2, 3, 3, 2, 1]})
+
+        result = df.nlargest(4, columns=["a", "b"])
+        expected = pd.DataFrame(
+            {"a": [2, 2, 2, 1], "b": [3, 2, 1, 3]}, index=[2, 1, 0, 3]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.nsmallest(4, columns=["a", "b"])
+        expected = pd.DataFrame(
+            {"a": [1, 1, 1, 2], "b": [1, 2, 3, 1]}, index=[5, 4, 3, 0]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_nlargest_n_identical_values(self):
+        # GH#15297
+        df = pd.DataFrame({"a": [1] * 5, "b": [1, 2, 3, 4, 5]})
+
+        result = df.nlargest(3, "a")
+        expected = pd.DataFrame({"a": [1] * 3, "b": [1, 2, 3]}, index=range(3))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.nsmallest(3, "a")
+        expected = pd.DataFrame({"a": [1] * 3, "b": [1, 2, 3]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "order",
+        [["a", "b", "c"], ["c", "b", "a"], ["a"], ["b"], ["a", "b"], ["c", "b"]],
+    )
+    @pytest.mark.parametrize("n", range(1, 6))
+    def test_nlargest_n_duplicate_index(self, n, order, request):
+        # GH#13412
+
+        df = pd.DataFrame(
+            {"a": [1, 2, 3, 4, 4], "b": [1, 1, 1, 1, 1], "c": [0, 1, 2, 5, 4]},
+            index=[0, 0, 1, 1, 1],
+        )
+        result = df.nsmallest(n, order)
+        expected = df.sort_values(order).head(n)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.nlargest(n, order)
+        expected = df.sort_values(order, ascending=False).head(n)
+        if Version(np.__version__) >= Version("1.25") and (
+            (order == ["a"] and n in (1, 2, 3, 4)) or ((order == ["a", "b"]) and n == 5)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=(
+                        "pandas default unstable sorting of duplicates"
+                        "issue with numpy>=1.25 with AVX instructions"
+                    ),
+                    strict=False,
+                )
+            )
+        tm.assert_frame_equal(result, expected)
+
+    def test_nlargest_duplicate_keep_all_ties(self):
+        # GH#16818
+        df = pd.DataFrame(
+            {"a": [5, 4, 4, 2, 3, 3, 3, 3], "b": [10, 9, 8, 7, 5, 50, 10, 20]}
+        )
+        result = df.nlargest(4, "a", keep="all")
+        expected = pd.DataFrame(
+            {
+                "a": [5, 4, 4, 3, 3, 3, 3],
+                "b": [10, 9, 8, 5, 50, 10, 20],
+            },
+            index=[0, 1, 2, 4, 5, 6, 7],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.nsmallest(2, "a", keep="all")
+        expected = pd.DataFrame(
+            {
+                "a": [2, 3, 3, 3, 3],
+                "b": [7, 5, 50, 10, 20],
+            },
+            index=range(3, 8),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_nlargest_multiindex_column_lookup(self):
+        # Check whether tuples are correctly treated as multi-level lookups.
+        # GH#23033
+        df = pd.DataFrame(
+            columns=pd.MultiIndex.from_product([["x"], ["a", "b"]]),
+            data=[[0.33, 0.13], [0.86, 0.25], [0.25, 0.70], [0.85, 0.91]],
+        )
+
+        # nsmallest
+        result = df.nsmallest(3, ("x", "a"))
+        expected = df.iloc[[2, 0, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # nlargest
+        result = df.nlargest(3, ("x", "b"))
+        expected = df.iloc[[3, 2, 1]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_nlargest_nan(self):
+        # GH#43060
+        df = pd.DataFrame([np.nan, np.nan, 0, 1, 2, 3])
+        result = df.nlargest(5, 0)
+        expected = df.sort_values(0, ascending=False).head(5)
+        tm.assert_frame_equal(result, expected)
+
+    def test_nsmallest_nan_after_n_element(self):
+        # GH#46589
+        df = pd.DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, None, 7],
+                "b": [7, 6, 5, 4, 3, 2, 1],
+                "c": [1, 1, 2, 2, 3, 3, 3],
+            },
+            index=range(7),
+        )
+        result = df.nsmallest(5, columns=["a", "b"])
+        expected = pd.DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5],
+                "b": [7, 6, 5, 4, 3],
+                "c": [1, 1, 2, 2, 3],
+            },
+            index=range(5),
+        ).astype({"a": "float"})
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_pct_change.py b/pandas/pandas/tests/frame/methods/test_pct_change.py
new file mode 100644
index 0000000000000..7d4197577228e
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_pct_change.py
@@ -0,0 +1,125 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFramePctChange:
+    @pytest.mark.parametrize(
+        "periods, exp",
+        [
+            (1, [np.nan, np.nan, np.nan, 1, 1, 1.5, np.nan, np.nan]),
+            (-1, [np.nan, np.nan, -0.5, -0.5, -0.6, np.nan, np.nan, np.nan]),
+        ],
+    )
+    def test_pct_change_with_nas(self, periods, exp, frame_or_series):
+        vals = [np.nan, np.nan, 1, 2, 4, 10, np.nan, np.nan]
+        obj = frame_or_series(vals)
+
+        res = obj.pct_change(periods=periods)
+        tm.assert_equal(res, frame_or_series(exp))
+
+    def test_pct_change_numeric(self):
+        # GH#11150
+        pnl = DataFrame(
+            [np.arange(0, 40, 10), np.arange(0, 40, 10), np.arange(0, 40, 10)]
+        ).astype(np.float64)
+        pnl.iat[1, 0] = np.nan
+        pnl.iat[1, 1] = np.nan
+        pnl.iat[2, 3] = 60
+
+        for axis in range(2):
+            expected = pnl / pnl.shift(axis=axis) - 1
+            result = pnl.pct_change(axis=axis)
+            tm.assert_frame_equal(result, expected)
+
+    def test_pct_change(self, datetime_frame):
+        rs = datetime_frame.pct_change()
+        tm.assert_frame_equal(rs, datetime_frame / datetime_frame.shift(1) - 1)
+
+        rs = datetime_frame.pct_change(2)
+        filled = datetime_frame.ffill()
+        tm.assert_frame_equal(rs, filled / filled.shift(2) - 1)
+
+        rs = datetime_frame.pct_change()
+        tm.assert_frame_equal(rs, datetime_frame / datetime_frame.shift(1) - 1)
+
+        rs = datetime_frame.pct_change(freq="5D")
+        tm.assert_frame_equal(
+            rs,
+            (datetime_frame / datetime_frame.shift(freq="5D") - 1).reindex_like(
+                datetime_frame
+            ),
+        )
+
+    def test_pct_change_shift_over_nas(self):
+        s = Series([1.0, 1.5, np.nan, 2.5, 3.0])
+
+        df = DataFrame({"a": s, "b": s})
+
+        chg = df.pct_change()
+        expected = Series([np.nan, 0.5, np.nan, np.nan, 0.2])
+        edf = DataFrame({"a": expected, "b": expected})
+        tm.assert_frame_equal(chg, edf)
+
+    @pytest.mark.parametrize(
+        "freq, periods",
+        [
+            ("5B", 5),
+            ("3B", 3),
+            ("14B", 14),
+        ],
+    )
+    def test_pct_change_periods_freq(
+        self,
+        datetime_frame,
+        freq,
+        periods,
+    ):
+        # GH#7292
+        rs_freq = datetime_frame.pct_change(freq=freq)
+        rs_periods = datetime_frame.pct_change(periods)
+        tm.assert_frame_equal(rs_freq, rs_periods)
+
+        empty_ts = DataFrame(index=datetime_frame.index, columns=datetime_frame.columns)
+        rs_freq = empty_ts.pct_change(freq=freq)
+        rs_periods = empty_ts.pct_change(periods)
+        tm.assert_frame_equal(rs_freq, rs_periods)
+
+
+def test_pct_change_with_duplicated_indices():
+    # GH30463
+    data = DataFrame(
+        {0: [np.nan, 1, 2, 3, 9, 18], 1: [0, 1, np.nan, 3, 9, 18]}, index=["a", "b"] * 3
+    )
+
+    result = data.pct_change()
+
+    second_column = [np.nan, np.inf, np.nan, np.nan, 2.0, 1.0]
+    expected = DataFrame(
+        {0: [np.nan, np.nan, 1.0, 0.5, 2.0, 1.0], 1: second_column},
+        index=["a", "b"] * 3,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_pct_change_none_beginning():
+    # GH#54481
+    df = DataFrame(
+        [
+            [1, None],
+            [2, 1],
+            [3, 2],
+            [4, 3],
+            [5, 4],
+        ]
+    )
+    result = df.pct_change()
+    expected = DataFrame(
+        {0: [np.nan, 1, 0.5, 1 / 3, 0.25], 1: [np.nan, np.nan, 1, 0.5, 1 / 3]}
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_pipe.py b/pandas/pandas/tests/frame/methods/test_pipe.py
new file mode 100644
index 0000000000000..5bcc4360487f3
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_pipe.py
@@ -0,0 +1,39 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestPipe:
+    def test_pipe(self, frame_or_series):
+        obj = DataFrame({"A": [1, 2, 3]})
+        expected = DataFrame({"A": [1, 4, 9]})
+        if frame_or_series is Series:
+            obj = obj["A"]
+            expected = expected["A"]
+
+        f = lambda x, y: x**y
+        result = obj.pipe(f, 2)
+        tm.assert_equal(result, expected)
+
+    def test_pipe_tuple(self, frame_or_series):
+        obj = DataFrame({"A": [1, 2, 3]})
+        obj = tm.get_obj(obj, frame_or_series)
+
+        f = lambda x, y: y
+        result = obj.pipe((f, "y"), 0)
+        tm.assert_equal(result, obj)
+
+    def test_pipe_tuple_error(self, frame_or_series):
+        obj = DataFrame({"A": [1, 2, 3]})
+        obj = tm.get_obj(obj, frame_or_series)
+
+        f = lambda x, y: y
+
+        msg = "y is both the pipe target and a keyword argument"
+
+        with pytest.raises(ValueError, match=msg):
+            obj.pipe((f, "y"), x=1, y=0)
diff --git a/pandas/pandas/tests/frame/methods/test_pop.py b/pandas/pandas/tests/frame/methods/test_pop.py
new file mode 100644
index 0000000000000..617f0c3a27885
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_pop.py
@@ -0,0 +1,71 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestDataFramePop:
+    def test_pop(self, float_frame):
+        float_frame.columns.name = "baz"
+
+        float_frame.pop("A")
+        assert "A" not in float_frame
+
+        float_frame["foo"] = "bar"
+        float_frame.pop("foo")
+        assert "foo" not in float_frame
+        assert float_frame.columns.name == "baz"
+
+        # gh-10912: inplace ops cause caching issue
+        a = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"], index=["X", "Y"])
+        b = a.pop("B")
+        b += 1
+
+        # original frame
+        expected = DataFrame([[1, 3], [4, 6]], columns=["A", "C"], index=["X", "Y"])
+        tm.assert_frame_equal(a, expected)
+
+        # result
+        expected = Series([2, 5], index=["X", "Y"], name="B") + 1
+        tm.assert_series_equal(b, expected)
+
+    def test_pop_non_unique_cols(self):
+        df = DataFrame({0: [0, 1], 1: [0, 1], 2: [4, 5]})
+        df.columns = ["a", "b", "a"]
+
+        res = df.pop("a")
+        assert type(res) == DataFrame
+        assert len(res) == 2
+        assert len(df.columns) == 1
+        assert "b" in df.columns
+        assert "a" not in df.columns
+        assert len(df.index) == 2
+
+    def test_mixed_depth_pop(self):
+        arrays = [
+            ["a", "top", "top", "routine1", "routine1", "routine2"],
+            ["", "OD", "OD", "result1", "result2", "result1"],
+            ["", "wx", "wy", "", "", ""],
+        ]
+
+        tuples = sorted(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples)
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 6)), columns=index)
+
+        df1 = df.copy()
+        df2 = df.copy()
+        result = df1.pop("a")
+        expected = df2.pop(("a", "", ""))
+        tm.assert_series_equal(expected, result, check_names=False)
+        tm.assert_frame_equal(df1, df2)
+        assert result.name == "a"
+
+        expected = df1["top"]
+        df1 = df1.drop(["top"], axis=1)
+        result = df2.pop("top")
+        tm.assert_frame_equal(expected, result)
+        tm.assert_frame_equal(df1, df2)
diff --git a/pandas/pandas/tests/frame/methods/test_quantile.py b/pandas/pandas/tests/frame/methods/test_quantile.py
new file mode 100644
index 0000000000000..d7baac7264a1d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_quantile.py
@@ -0,0 +1,946 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(
+    params=[["linear", "single"], ["nearest", "table"]], ids=lambda x: "-".join(x)
+)
+def interp_method(request):
+    """(interpolation, method) arguments for quantile"""
+    return request.param
+
+
+class TestDataFrameQuantile:
+    @pytest.mark.parametrize(
+        "df,expected",
+        [
+            [
+                DataFrame(
+                    {
+                        0: Series(pd.arrays.SparseArray([1, 2])),
+                        1: Series(pd.arrays.SparseArray([3, 4])),
+                    }
+                ),
+                Series([1.5, 3.5], name=0.5),
+            ],
+            [
+                DataFrame(Series([0.0, None, 1.0, 2.0], dtype="Sparse[float]")),
+                Series([1.0], name=0.5),
+            ],
+        ],
+    )
+    def test_quantile_sparse(self, df, expected):
+        # GH#17198
+        # GH#24600
+        result = df.quantile()
+        expected = expected.astype("Sparse[float]")
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile(self, datetime_frame, interp_method, request):
+        interpolation, method = interp_method
+        df = datetime_frame
+        result = df.quantile(
+            0.1, axis=0, numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = Series(
+            [np.percentile(df[col], 10) for col in df.columns],
+            index=df.columns,
+            name=0.1,
+        )
+        if interpolation == "linear":
+            # np.percentile values only comparable to linear interpolation
+            tm.assert_series_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.index, expected.index)
+            assert result.name == expected.name
+
+        result = df.quantile(
+            0.9, axis=1, numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = Series(
+            [np.percentile(df.loc[date], 90) for date in df.index],
+            index=df.index,
+            name=0.9,
+        )
+        if interpolation == "linear":
+            # np.percentile values only comparable to linear interpolation
+            tm.assert_series_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.index, expected.index)
+            assert result.name == expected.name
+
+    def test_empty(self, interp_method):
+        interpolation, method = interp_method
+        q = DataFrame({"x": [], "y": []}).quantile(
+            0.1, axis=0, numeric_only=True, interpolation=interpolation, method=method
+        )
+        assert np.isnan(q["x"]) and np.isnan(q["y"])
+
+    def test_non_numeric_exclusion(self, interp_method, request):
+        interpolation, method = interp_method
+        df = DataFrame({"col1": ["A", "A", "B", "B"], "col2": [1, 2, 3, 4]})
+        rs = df.quantile(
+            0.5, numeric_only=True, interpolation=interpolation, method=method
+        )
+        xp = df.median(numeric_only=True).rename(0.5)
+        if interpolation == "nearest":
+            xp = (xp + 0.5).astype(np.int64)
+        tm.assert_series_equal(rs, xp)
+
+    def test_axis(self, interp_method):
+        # axis
+        interpolation, method = interp_method
+        df = DataFrame({"A": [1, 2, 3], "B": [2, 3, 4]}, index=[1, 2, 3])
+        result = df.quantile(0.5, axis=1, interpolation=interpolation, method=method)
+        expected = Series([1.5, 2.5, 3.5], index=[1, 2, 3], name=0.5)
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_series_equal(result, expected)
+
+        result = df.quantile(
+            [0.5, 0.75], axis=1, interpolation=interpolation, method=method
+        )
+        expected = DataFrame(
+            {1: [1.5, 1.75], 2: [2.5, 2.75], 3: [3.5, 3.75]}, index=[0.5, 0.75]
+        )
+        if interpolation == "nearest":
+            expected.iloc[0, :] -= 0.5
+            expected.iloc[1, :] += 0.25
+            expected = expected.astype(np.int64)
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_axis_numeric_only_true(self, interp_method):
+        # We may want to break API in the future to change this
+        # so that we exclude non-numeric along the same axis
+        # See GH #7312
+        interpolation, method = interp_method
+        df = DataFrame([[1, 2, 3], ["a", "b", 4]])
+        result = df.quantile(
+            0.5, axis=1, numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = Series([3.0, 4.0], index=range(2), name=0.5)
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_date_range(self, interp_method):
+        # GH 2460
+        interpolation, method = interp_method
+        dti = pd.date_range("2016-01-01", periods=3, tz="US/Pacific")
+        ser = Series(dti)
+        df = DataFrame(ser)
+
+        result = df.quantile(
+            numeric_only=False, interpolation=interpolation, method=method
+        )
+        expected = Series(
+            ["2016-01-02 00:00:00"], name=0.5, dtype="datetime64[ns, US/Pacific]"
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_axis_mixed(self, interp_method):
+        # mixed on axis=1
+        interpolation, method = interp_method
+        df = DataFrame(
+            {
+                "A": [1, 2, 3],
+                "B": [2.0, 3.0, 4.0],
+                "C": pd.date_range("20130101", periods=3),
+                "D": ["foo", "bar", "baz"],
+            }
+        )
+        result = df.quantile(
+            0.5, axis=1, numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = Series([1.5, 2.5, 3.5], name=0.5)
+        if interpolation == "nearest":
+            expected -= 0.5
+        tm.assert_series_equal(result, expected)
+
+        # must raise
+        msg = "'<' not supported between instances of 'Timestamp' and 'float'"
+        with pytest.raises(TypeError, match=msg):
+            df.quantile(0.5, axis=1, numeric_only=False)
+
+    def test_quantile_axis_parameter(self, interp_method):
+        # GH 9543/9544
+        interpolation, method = interp_method
+        df = DataFrame({"A": [1, 2, 3], "B": [2, 3, 4]}, index=[1, 2, 3])
+
+        result = df.quantile(0.5, axis=0, interpolation=interpolation, method=method)
+
+        expected = Series([2.0, 3.0], index=["A", "B"], name=0.5)
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_series_equal(result, expected)
+
+        expected = df.quantile(
+            0.5, axis="index", interpolation=interpolation, method=method
+        )
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_series_equal(result, expected)
+
+        result = df.quantile(0.5, axis=1, interpolation=interpolation, method=method)
+
+        expected = Series([1.5, 2.5, 3.5], index=[1, 2, 3], name=0.5)
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_series_equal(result, expected)
+
+        result = df.quantile(
+            0.5, axis="columns", interpolation=interpolation, method=method
+        )
+        tm.assert_series_equal(result, expected)
+
+        msg = "No axis named -1 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.quantile(0.1, axis=-1, interpolation=interpolation, method=method)
+        msg = "No axis named column for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.quantile(0.1, axis="column")
+
+    def test_quantile_interpolation(self):
+        # see gh-10174
+
+        # interpolation method other than default linear
+        df = DataFrame({"A": [1, 2, 3], "B": [2, 3, 4]}, index=[1, 2, 3])
+        result = df.quantile(0.5, axis=1, interpolation="nearest")
+        expected = Series([1, 2, 3], index=[1, 2, 3], name=0.5)
+        tm.assert_series_equal(result, expected)
+
+        # cross-check interpolation=nearest results in original dtype
+        exp = np.percentile(
+            np.array([[1, 2, 3], [2, 3, 4]]),
+            0.5,
+            axis=0,
+            method="nearest",
+        )
+        expected = Series(exp, index=[1, 2, 3], name=0.5, dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+        # float
+        df = DataFrame({"A": [1.0, 2.0, 3.0], "B": [2.0, 3.0, 4.0]}, index=[1, 2, 3])
+        result = df.quantile(0.5, axis=1, interpolation="nearest")
+        expected = Series([1.0, 2.0, 3.0], index=[1, 2, 3], name=0.5)
+        tm.assert_series_equal(result, expected)
+        exp = np.percentile(
+            np.array([[1.0, 2.0, 3.0], [2.0, 3.0, 4.0]]),
+            0.5,
+            axis=0,
+            method="nearest",
+        )
+        expected = Series(exp, index=[1, 2, 3], name=0.5, dtype="float64")
+        tm.assert_series_equal(result, expected)
+
+        # axis
+        result = df.quantile([0.5, 0.75], axis=1, interpolation="lower")
+        expected = DataFrame(
+            {1: [1.0, 1.0], 2: [2.0, 2.0], 3: [3.0, 3.0]}, index=[0.5, 0.75]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # test degenerate case
+        df = DataFrame({"x": [], "y": []})
+        q = df.quantile(0.1, axis=0, interpolation="higher")
+        assert np.isnan(q["x"]) and np.isnan(q["y"])
+
+        # multi
+        df = DataFrame([[1, 1, 1], [2, 2, 2], [3, 3, 3]], columns=["a", "b", "c"])
+        result = df.quantile([0.25, 0.5], interpolation="midpoint")
+
+        # https://github.com/numpy/numpy/issues/7163
+        expected = DataFrame(
+            [[1.5, 1.5, 1.5], [2.0, 2.0, 2.0]],
+            index=[0.25, 0.5],
+            columns=["a", "b", "c"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_quantile_interpolation_datetime(self, datetime_frame):
+        # see gh-10174
+
+        # interpolation = linear (default case)
+        df = datetime_frame
+        q = df.quantile(0.1, axis=0, numeric_only=True, interpolation="linear")
+        assert q["A"] == np.percentile(df["A"], 10)
+
+    def test_quantile_interpolation_int(self, int_frame):
+        # see gh-10174
+
+        df = int_frame
+        # interpolation = linear (default case)
+        q = df.quantile(0.1)
+        assert q["A"] == np.percentile(df["A"], 10)
+
+        # test with and without interpolation keyword
+        q1 = df.quantile(0.1, axis=0, interpolation="linear")
+        assert q1["A"] == np.percentile(df["A"], 10)
+        tm.assert_series_equal(q, q1)
+
+    def test_quantile_multi(self, interp_method):
+        interpolation, method = interp_method
+        df = DataFrame([[1, 1, 1], [2, 2, 2], [3, 3, 3]], columns=["a", "b", "c"])
+        result = df.quantile([0.25, 0.5], interpolation=interpolation, method=method)
+        expected = DataFrame(
+            [[1.5, 1.5, 1.5], [2.0, 2.0, 2.0]],
+            index=[0.25, 0.5],
+            columns=["a", "b", "c"],
+        )
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_frame_equal(result, expected)
+
+    def test_quantile_multi_axis_1(self, interp_method):
+        interpolation, method = interp_method
+        df = DataFrame([[1, 1, 1], [2, 2, 2], [3, 3, 3]], columns=["a", "b", "c"])
+        result = df.quantile(
+            [0.25, 0.5], axis=1, interpolation=interpolation, method=method
+        )
+        expected = DataFrame(
+            [[1.0, 2.0, 3.0]] * 2, index=[0.25, 0.5], columns=[0, 1, 2]
+        )
+        if interpolation == "nearest":
+            expected = expected.astype(np.int64)
+        tm.assert_frame_equal(result, expected)
+
+    def test_quantile_multi_empty(self, interp_method):
+        interpolation, method = interp_method
+        result = DataFrame({"x": [], "y": []}).quantile(
+            [0.1, 0.9], axis=0, interpolation=interpolation, method=method
+        )
+        expected = DataFrame(
+            {"x": [np.nan, np.nan], "y": [np.nan, np.nan]}, index=[0.1, 0.9]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_quantile_datetime(self, unit):
+        dti = pd.to_datetime(["2010", "2011"]).as_unit(unit)
+        df = DataFrame({"a": dti, "b": [0, 5]})
+
+        # exclude datetime
+        result = df.quantile(0.5, numeric_only=True)
+        expected = Series([2.5], index=["b"], name=0.5)
+        tm.assert_series_equal(result, expected)
+
+        # datetime
+        result = df.quantile(0.5, numeric_only=False)
+        expected = Series(
+            [Timestamp("2010-07-02 12:00:00"), 2.5], index=["a", "b"], name=0.5
+        )
+        tm.assert_series_equal(result, expected)
+
+        # datetime w/ multi
+        result = df.quantile([0.5], numeric_only=False)
+        expected = DataFrame(
+            {"a": Timestamp("2010-07-02 12:00:00").as_unit(unit), "b": 2.5},
+            index=[0.5],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # axis = 1
+        df["c"] = pd.to_datetime(["2011", "2012"]).as_unit(unit)
+        result = df[["a", "c"]].quantile(0.5, axis=1, numeric_only=False)
+        expected = Series(
+            [Timestamp("2010-07-02 12:00:00"), Timestamp("2011-07-02 12:00:00")],
+            index=[0, 1],
+            name=0.5,
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = df[["a", "c"]].quantile([0.5], axis=1, numeric_only=False)
+        expected = DataFrame(
+            [[Timestamp("2010-07-02 12:00:00"), Timestamp("2011-07-02 12:00:00")]],
+            index=[0.5],
+            columns=[0, 1],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # empty when numeric_only=True
+        result = df[["a", "c"]].quantile(0.5, numeric_only=True)
+        expected = Series([], index=Index([], dtype="str"), dtype=np.float64, name=0.5)
+        tm.assert_series_equal(result, expected)
+
+        result = df[["a", "c"]].quantile([0.5], numeric_only=True)
+        expected = DataFrame(index=[0.5], columns=Index([], dtype="str"))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "datetime64[ns]",
+            "datetime64[ns, US/Pacific]",
+            "timedelta64[ns]",
+            "Period[D]",
+        ],
+    )
+    def test_quantile_dt64_empty(self, dtype, interp_method):
+        # GH#41544
+        interpolation, method = interp_method
+        df = DataFrame(columns=["a", "b"], dtype=dtype)
+
+        res = df.quantile(
+            0.5, axis=1, numeric_only=False, interpolation=interpolation, method=method
+        )
+        expected = Series([], index=Index([], dtype="str"), name=0.5, dtype=dtype)
+        tm.assert_series_equal(res, expected)
+
+        # no columns in result, so no dtype preservation
+        res = df.quantile(
+            [0.5],
+            axis=1,
+            numeric_only=False,
+            interpolation=interpolation,
+            method=method,
+        )
+        expected = DataFrame(index=[0.5], columns=Index([], dtype="str"))
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.parametrize("invalid", [-1, 2, [0.5, -1], [0.5, 2]])
+    def test_quantile_invalid(self, invalid, datetime_frame, interp_method):
+        msg = "percentiles should all be in the interval \\[0, 1\\]"
+        interpolation, method = interp_method
+        with pytest.raises(ValueError, match=msg):
+            datetime_frame.quantile(invalid, interpolation=interpolation, method=method)
+
+    def test_quantile_box(self, interp_method):
+        interpolation, method = interp_method
+        df = DataFrame(
+            {
+                "A": [
+                    Timestamp("2011-01-01"),
+                    Timestamp("2011-01-02"),
+                    Timestamp("2011-01-03"),
+                ],
+                "B": [
+                    Timestamp("2011-01-01", tz="US/Eastern"),
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    Timestamp("2011-01-03", tz="US/Eastern"),
+                ],
+                "C": [
+                    pd.Timedelta("1 days"),
+                    pd.Timedelta("2 days"),
+                    pd.Timedelta("3 days"),
+                ],
+            }
+        )
+
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+
+        exp = Series(
+            [
+                Timestamp("2011-01-02"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+                pd.Timedelta("2 days"),
+            ],
+            name=0.5,
+            index=["A", "B", "C"],
+        )
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile(
+            [0.5], numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = DataFrame(
+            [
+                [
+                    Timestamp("2011-01-02"),
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    pd.Timedelta("2 days"),
+                ]
+            ],
+            index=[0.5],
+            columns=["A", "B", "C"],
+        )
+        tm.assert_frame_equal(res, exp)
+
+    def test_quantile_box_nat(self):
+        # DatetimeLikeBlock may be consolidated and contain NaT in different loc
+        df = DataFrame(
+            {
+                "A": [
+                    Timestamp("2011-01-01"),
+                    pd.NaT,
+                    Timestamp("2011-01-02"),
+                    Timestamp("2011-01-03"),
+                ],
+                "a": [
+                    Timestamp("2011-01-01"),
+                    Timestamp("2011-01-02"),
+                    pd.NaT,
+                    Timestamp("2011-01-03"),
+                ],
+                "B": [
+                    Timestamp("2011-01-01", tz="US/Eastern"),
+                    pd.NaT,
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    Timestamp("2011-01-03", tz="US/Eastern"),
+                ],
+                "b": [
+                    Timestamp("2011-01-01", tz="US/Eastern"),
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    pd.NaT,
+                    Timestamp("2011-01-03", tz="US/Eastern"),
+                ],
+                "C": [
+                    pd.Timedelta("1 days"),
+                    pd.Timedelta("2 days"),
+                    pd.Timedelta("3 days"),
+                    pd.NaT,
+                ],
+                "c": [
+                    pd.NaT,
+                    pd.Timedelta("1 days"),
+                    pd.Timedelta("2 days"),
+                    pd.Timedelta("3 days"),
+                ],
+            },
+            columns=list("AaBbCc"),
+        )
+
+        res = df.quantile(0.5, numeric_only=False)
+        exp = Series(
+            [
+                Timestamp("2011-01-02"),
+                Timestamp("2011-01-02"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+                pd.Timedelta("2 days"),
+                pd.Timedelta("2 days"),
+            ],
+            name=0.5,
+            index=list("AaBbCc"),
+        )
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile([0.5], numeric_only=False)
+        exp = DataFrame(
+            [
+                [
+                    Timestamp("2011-01-02"),
+                    Timestamp("2011-01-02"),
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    Timestamp("2011-01-02", tz="US/Eastern"),
+                    pd.Timedelta("2 days"),
+                    pd.Timedelta("2 days"),
+                ]
+            ],
+            index=[0.5],
+            columns=list("AaBbCc"),
+        )
+        tm.assert_frame_equal(res, exp)
+
+    def test_quantile_nan(self, interp_method):
+        interpolation, method = interp_method
+        # GH 14357 - float block where some cols have missing values
+        df = DataFrame({"a": np.arange(1, 6.0), "b": np.arange(1, 6.0)})
+        df.iloc[-1, 1] = np.nan
+
+        res = df.quantile(0.5, interpolation=interpolation, method=method)
+        exp = Series(
+            [3.0, 2.5 if interpolation == "linear" else 3.0], index=["a", "b"], name=0.5
+        )
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile([0.5, 0.75], interpolation=interpolation, method=method)
+        exp = DataFrame(
+            {
+                "a": [3.0, 4.0],
+                "b": [2.5, 3.25] if interpolation == "linear" else [3.0, 4.0],
+            },
+            index=[0.5, 0.75],
+        )
+        tm.assert_frame_equal(res, exp)
+
+        res = df.quantile(0.5, axis=1, interpolation=interpolation, method=method)
+        exp = Series(np.arange(1.0, 6.0), name=0.5)
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile(
+            [0.5, 0.75], axis=1, interpolation=interpolation, method=method
+        )
+        exp = DataFrame([np.arange(1.0, 6.0)] * 2, index=[0.5, 0.75])
+        if interpolation == "nearest":
+            exp.iloc[1, -1] = np.nan
+        tm.assert_frame_equal(res, exp)
+
+        # full-nan column
+        df["b"] = np.nan
+
+        res = df.quantile(0.5, interpolation=interpolation, method=method)
+        exp = Series([3.0, np.nan], index=["a", "b"], name=0.5)
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile([0.5, 0.75], interpolation=interpolation, method=method)
+        exp = DataFrame({"a": [3.0, 4.0], "b": [np.nan, np.nan]}, index=[0.5, 0.75])
+        tm.assert_frame_equal(res, exp)
+
+    def test_quantile_nat(self, interp_method, unit):
+        interpolation, method = interp_method
+        # full NaT column
+        df = DataFrame({"a": [pd.NaT, pd.NaT, pd.NaT]}, dtype=f"M8[{unit}]")
+
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = Series([pd.NaT], index=["a"], name=0.5, dtype=f"M8[{unit}]")
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile(
+            [0.5], numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = DataFrame({"a": [pd.NaT]}, index=[0.5], dtype=f"M8[{unit}]")
+        tm.assert_frame_equal(res, exp)
+
+        # mixed non-null / full null column
+        df = DataFrame(
+            {
+                "a": [
+                    Timestamp("2012-01-01"),
+                    Timestamp("2012-01-02"),
+                    Timestamp("2012-01-03"),
+                ],
+                "b": [pd.NaT, pd.NaT, pd.NaT],
+            },
+            dtype=f"M8[{unit}]",
+        )
+
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = Series(
+            [Timestamp("2012-01-02"), pd.NaT],
+            index=["a", "b"],
+            name=0.5,
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile(
+            [0.5], numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = DataFrame(
+            [[Timestamp("2012-01-02"), pd.NaT]],
+            index=[0.5],
+            columns=["a", "b"],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_frame_equal(res, exp)
+
+    def test_quantile_empty_no_rows_floats(self, interp_method):
+        interpolation, method = interp_method
+
+        df = DataFrame(columns=["a", "b"], dtype="float64")
+
+        res = df.quantile(0.5, interpolation=interpolation, method=method)
+        exp = Series([np.nan, np.nan], index=["a", "b"], name=0.5)
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile([0.5], interpolation=interpolation, method=method)
+        exp = DataFrame([[np.nan, np.nan]], columns=["a", "b"], index=[0.5])
+        tm.assert_frame_equal(res, exp)
+
+        res = df.quantile(0.5, axis=1, interpolation=interpolation, method=method)
+        exp = Series([], index=Index([], dtype="str"), dtype="float64", name=0.5)
+        tm.assert_series_equal(res, exp)
+
+        res = df.quantile([0.5], axis=1, interpolation=interpolation, method=method)
+        exp = DataFrame(columns=Index([], dtype="str"), index=[0.5])
+        tm.assert_frame_equal(res, exp)
+
+    def test_quantile_empty_no_rows_ints(self, interp_method):
+        interpolation, method = interp_method
+        df = DataFrame(columns=["a", "b"], dtype="int64")
+
+        res = df.quantile(0.5, interpolation=interpolation, method=method)
+        exp = Series([np.nan, np.nan], index=["a", "b"], name=0.5)
+        tm.assert_series_equal(res, exp)
+
+    def test_quantile_empty_no_rows_dt64(self, interp_method):
+        interpolation, method = interp_method
+        # datetimes
+        df = DataFrame(columns=["a", "b"], dtype="datetime64[ns]")
+
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = Series(
+            [pd.NaT, pd.NaT], index=["a", "b"], dtype="datetime64[ns]", name=0.5
+        )
+        tm.assert_series_equal(res, exp)
+
+        # Mixed dt64/dt64tz
+        df["a"] = df["a"].dt.tz_localize("US/Central")
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = exp.astype(object)
+        if interpolation == "nearest":
+            # GH#18463 TODO: would we prefer NaTs here?
+            exp = exp.fillna(np.nan)
+        tm.assert_series_equal(res, exp)
+
+        # both dt64tz
+        df["b"] = df["b"].dt.tz_localize("US/Central")
+        res = df.quantile(
+            0.5, numeric_only=False, interpolation=interpolation, method=method
+        )
+        exp = exp.astype(df["b"].dtype)
+        tm.assert_series_equal(res, exp)
+
+    def test_quantile_empty_no_columns(self, interp_method):
+        # GH#23925 _get_numeric_data may drop all columns
+        interpolation, method = interp_method
+        df = DataFrame(pd.date_range("1/1/18", periods=5))
+        df.columns.name = "captain tightpants"
+        result = df.quantile(
+            0.5, numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = Series([], name=0.5, dtype=np.float64)
+        expected.index.name = "captain tightpants"
+        tm.assert_series_equal(result, expected)
+
+        result = df.quantile(
+            [0.5], numeric_only=True, interpolation=interpolation, method=method
+        )
+        expected = DataFrame([], index=[0.5])
+        expected.columns.name = "captain tightpants"
+        tm.assert_frame_equal(result, expected)
+
+    def test_quantile_item_cache(self, interp_method):
+        # previous behavior incorrect retained an invalid _item_cache entry
+        interpolation, method = interp_method
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 3)), columns=["A", "B", "C"]
+        )
+        df["D"] = df["A"] * 2
+        ser = df["A"]
+        assert len(df._mgr.blocks) == 2
+
+        df.quantile(numeric_only=False, interpolation=interpolation, method=method)
+
+        ser.iloc[0] = 99
+        assert df.iloc[0, 0] == df["A"][0]
+        assert df.iloc[0, 0] != 99
+
+    def test_invalid_method(self):
+        with pytest.raises(ValueError, match="Invalid method: foo"):
+            DataFrame(range(1)).quantile(0.5, method="foo")
+
+    def test_table_invalid_interpolation(self):
+        with pytest.raises(ValueError, match="Invalid interpolation: foo"):
+            DataFrame(range(1)).quantile(0.5, method="table", interpolation="foo")
+
+
+class TestQuantileExtensionDtype:
+    # TODO: tests for axis=1?
+    # TODO: empty case?
+
+    @pytest.fixture(
+        params=[
+            pytest.param(
+                pd.IntervalIndex.from_breaks(range(10)),
+                marks=pytest.mark.xfail(reason="raises when trying to add Intervals"),
+            ),
+            pd.period_range("2016-01-01", periods=9, freq="D"),
+            pd.date_range("2016-01-01", periods=9, tz="US/Pacific"),
+            pd.timedelta_range("1 Day", periods=9),
+            pd.array(np.arange(9), dtype="Int64"),
+            pd.array(np.arange(9), dtype="Float64"),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def index(self, request):
+        # NB: not actually an Index object
+        idx = request.param
+        idx.name = "A"
+        return idx
+
+    @pytest.fixture
+    def obj(self, index, frame_or_series):
+        # bc index is not always an Index (yet), we need to re-patch .name
+        obj = frame_or_series(index).copy()
+
+        if frame_or_series is Series:
+            obj.name = "A"
+        else:
+            obj.columns = ["A"]
+        return obj
+
+    def compute_quantile(self, obj, qs):
+        if isinstance(obj, Series):
+            result = obj.quantile(qs)
+        else:
+            result = obj.quantile(qs, numeric_only=False)
+        return result
+
+    def test_quantile_ea(self, request, obj, index):
+        # result should be invariant to shuffling
+        indexer = np.arange(len(index), dtype=np.intp)
+        np.random.default_rng(2).shuffle(indexer)
+        obj = obj.iloc[indexer]
+
+        qs = [0.5, 0, 1]
+        result = self.compute_quantile(obj, qs)
+
+        exp_dtype = index.dtype
+        if index.dtype == "Int64":
+            # match non-nullable casting behavior
+            exp_dtype = "Float64"
+
+        # expected here assumes len(index) == 9
+        expected = Series(
+            [index[4], index[0], index[-1]], dtype=exp_dtype, index=qs, name="A"
+        )
+        expected = type(obj)(expected)
+
+        tm.assert_equal(result, expected)
+
+    def test_quantile_ea_with_na(self, obj, index):
+        obj.iloc[0] = index._na_value
+        obj.iloc[-1] = index._na_value
+
+        # result should be invariant to shuffling
+        indexer = np.arange(len(index), dtype=np.intp)
+        np.random.default_rng(2).shuffle(indexer)
+        obj = obj.iloc[indexer]
+
+        qs = [0.5, 0, 1]
+        result = self.compute_quantile(obj, qs)
+
+        # expected here assumes len(index) == 9
+        expected = Series(
+            [index[4], index[1], index[-2]], dtype=index.dtype, index=qs, name="A"
+        )
+        expected = type(obj)(expected)
+        tm.assert_equal(result, expected)
+
+    def test_quantile_ea_all_na(self, request, obj, index):
+        obj.iloc[:] = index._na_value
+        # Check dtypes were preserved; this was once a problem see GH#39763
+        assert np.all(obj.dtypes == index.dtype)
+
+        # result should be invariant to shuffling
+        indexer = np.arange(len(index), dtype=np.intp)
+        np.random.default_rng(2).shuffle(indexer)
+        obj = obj.iloc[indexer]
+
+        qs = [0.5, 0, 1]
+        result = self.compute_quantile(obj, qs)
+
+        expected = index.take([-1, -1, -1], allow_fill=True, fill_value=index._na_value)
+        expected = Series(expected, index=qs, name="A")
+        expected = type(obj)(expected)
+        tm.assert_equal(result, expected)
+
+    def test_quantile_ea_scalar(self, request, obj, index):
+        # scalar qs
+
+        # result should be invariant to shuffling
+        indexer = np.arange(len(index), dtype=np.intp)
+        np.random.default_rng(2).shuffle(indexer)
+        obj = obj.iloc[indexer]
+
+        qs = 0.5
+        result = self.compute_quantile(obj, qs)
+
+        exp_dtype = index.dtype
+        if index.dtype == "Int64":
+            exp_dtype = "Float64"
+
+        expected = Series({"A": index[4]}, dtype=exp_dtype, name=0.5)
+        if isinstance(obj, Series):
+            expected = expected["A"]
+            assert result == expected
+        else:
+            tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, expected_data, expected_index, axis",
+        [
+            ["float64", [], [], 1],
+            ["int64", [], [], 1],
+            ["float64", [np.nan, np.nan], ["a", "b"], 0],
+            ["int64", [np.nan, np.nan], ["a", "b"], 0],
+        ],
+    )
+    def test_empty_numeric(self, dtype, expected_data, expected_index, axis):
+        # GH 14564
+        df = DataFrame(columns=["a", "b"], dtype=dtype)
+        result = df.quantile(0.5, axis=axis)
+        expected = Series(
+            expected_data,
+            name=0.5,
+            index=Index(expected_index, dtype="str"),
+            dtype="float64",
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, expected_data, expected_index, axis, expected_dtype",
+        [
+            ["datetime64[ns]", [], [], 1, "datetime64[ns]"],
+            ["datetime64[ns]", [pd.NaT, pd.NaT], ["a", "b"], 0, "datetime64[ns]"],
+        ],
+    )
+    def test_empty_datelike(
+        self, dtype, expected_data, expected_index, axis, expected_dtype
+    ):
+        # GH 14564
+        df = DataFrame(columns=["a", "b"], dtype=dtype)
+        result = df.quantile(0.5, axis=axis, numeric_only=False)
+        expected = Series(
+            expected_data,
+            name=0.5,
+            index=Index(expected_index, dtype="str"),
+            dtype=expected_dtype,
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "expected_data, expected_index, axis",
+        [
+            [[np.nan, np.nan], range(2), 1],
+            [[], [], 0],
+        ],
+    )
+    def test_datelike_numeric_only(self, expected_data, expected_index, axis):
+        # GH 14564
+        df = DataFrame(
+            {
+                "a": pd.to_datetime(["2010", "2011"]),
+                "b": [0, 5],
+                "c": pd.to_datetime(["2011", "2012"]),
+            }
+        )
+        result = df[["a", "c"]].quantile(0.5, axis=axis, numeric_only=True)
+        expected = Series(
+            expected_data,
+            name=0.5,
+            index=Index(expected_index, dtype="str" if axis == 0 else "int64"),
+            dtype=np.float64,
+        )
+        tm.assert_series_equal(result, expected)
+
+
+def test_multi_quantile_numeric_only_retains_columns():
+    df = DataFrame(list("abc"))
+    result = df.quantile([0.5, 0.7], numeric_only=True)
+    expected = DataFrame(index=[0.5, 0.7])
+    tm.assert_frame_equal(
+        result, expected, check_index_type=True, check_column_type=True
+    )
diff --git a/pandas/pandas/tests/frame/methods/test_rank.py b/pandas/pandas/tests/frame/methods/test_rank.py
new file mode 100644
index 0000000000000..6c6c208ee0c78
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_rank.py
@@ -0,0 +1,500 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.algos import (
+    Infinity,
+    NegInfinity,
+)
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestRank:
+    s = Series([1, 3, 4, 2, np.nan, 2, 1, 5, np.nan, 3])
+    df = DataFrame({"A": s, "B": s})
+
+    results = {
+        "average": np.array([1.5, 5.5, 7.0, 3.5, np.nan, 3.5, 1.5, 8.0, np.nan, 5.5]),
+        "min": np.array([1, 5, 7, 3, np.nan, 3, 1, 8, np.nan, 5]),
+        "max": np.array([2, 6, 7, 4, np.nan, 4, 2, 8, np.nan, 6]),
+        "first": np.array([1, 5, 7, 3, np.nan, 4, 2, 8, np.nan, 6]),
+        "dense": np.array([1, 3, 4, 2, np.nan, 2, 1, 5, np.nan, 3]),
+    }
+
+    def test_rank(self, float_frame):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        float_frame.loc[::2, "A"] = np.nan
+        float_frame.loc[::3, "B"] = np.nan
+        float_frame.loc[::4, "C"] = np.nan
+        float_frame.loc[::5, "D"] = np.nan
+
+        ranks0 = float_frame.rank()
+        ranks1 = float_frame.rank(1)
+        mask = np.isnan(float_frame.values)
+
+        fvals = float_frame.fillna(np.inf).values
+
+        exp0 = np.apply_along_axis(sp_stats.rankdata, 0, fvals)
+        exp0[mask] = np.nan
+
+        exp1 = np.apply_along_axis(sp_stats.rankdata, 1, fvals)
+        exp1[mask] = np.nan
+
+        tm.assert_almost_equal(ranks0.values, exp0)
+        tm.assert_almost_equal(ranks1.values, exp1)
+
+        # integers
+        df = DataFrame(
+            np.random.default_rng(2).integers(0, 5, size=40).reshape((10, 4))
+        )
+
+        result = df.rank()
+        exp = df.astype(float).rank()
+        tm.assert_frame_equal(result, exp)
+
+        result = df.rank(1)
+        exp = df.astype(float).rank(1)
+        tm.assert_frame_equal(result, exp)
+
+    def test_rank2(self):
+        df = DataFrame([[1, 3, 2], [1, 2, 3]])
+        expected = DataFrame([[1.0, 3.0, 2.0], [1, 2, 3]]) / 3.0
+        result = df.rank(1, pct=True)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame([[1, 3, 2], [1, 2, 3]])
+        expected = df.rank(0) / 2.0
+        result = df.rank(0, pct=True)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame([["b", "c", "a"], ["a", "c", "b"]])
+        expected = DataFrame([[2.0, 3.0, 1.0], [1, 3, 2]])
+        result = df.rank(1, numeric_only=False)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([[2.0, 1.5, 1.0], [1, 1.5, 2]])
+        result = df.rank(0, numeric_only=False)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame([["b", np.nan, "a"], ["a", "c", "b"]])
+        expected = DataFrame([[2.0, np.nan, 1.0], [1.0, 3.0, 2.0]])
+        result = df.rank(1, numeric_only=False)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([[2.0, np.nan, 1.0], [1.0, 1.0, 2.0]])
+        result = df.rank(0, numeric_only=False)
+        tm.assert_frame_equal(result, expected)
+
+        # f7u12, this does not work without extensive workaround
+        data = [
+            [datetime(2001, 1, 5), np.nan, datetime(2001, 1, 2)],
+            [datetime(2000, 1, 2), datetime(2000, 1, 3), datetime(2000, 1, 1)],
+        ]
+        df = DataFrame(data)
+
+        # check the rank
+        expected = DataFrame([[2.0, np.nan, 1.0], [2.0, 3.0, 1.0]])
+        result = df.rank(1, numeric_only=False, ascending=True)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([[1.0, np.nan, 2.0], [2.0, 1.0, 3.0]])
+        result = df.rank(1, numeric_only=False, ascending=False)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({"a": [1e-20, -5, 1e-20 + 1e-40, 10, 1e60, 1e80, 1e-30]})
+        exp = DataFrame({"a": [3.5, 1.0, 3.5, 5.0, 6.0, 7.0, 2.0]})
+        tm.assert_frame_equal(df.rank(), exp)
+
+    def test_rank_does_not_mutate(self):
+        # GH#18521
+        # Check rank does not mutate DataFrame
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), dtype="float64"
+        )
+        expected = df.copy()
+        df.rank()
+        result = df
+        tm.assert_frame_equal(result, expected)
+
+    def test_rank_mixed_frame(self, float_string_frame):
+        float_string_frame["datetime"] = datetime.now()
+        float_string_frame["timedelta"] = timedelta(days=1, seconds=1)
+
+        float_string_frame.rank(numeric_only=False)
+        with pytest.raises(TypeError, match="not supported between instances of"):
+            float_string_frame.rank(axis=1)
+
+    def test_rank_na_option(self, float_frame):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        float_frame.loc[::2, "A"] = np.nan
+        float_frame.loc[::3, "B"] = np.nan
+        float_frame.loc[::4, "C"] = np.nan
+        float_frame.loc[::5, "D"] = np.nan
+
+        # bottom
+        ranks0 = float_frame.rank(na_option="bottom")
+        ranks1 = float_frame.rank(1, na_option="bottom")
+
+        fvals = float_frame.fillna(np.inf).values
+
+        exp0 = np.apply_along_axis(sp_stats.rankdata, 0, fvals)
+        exp1 = np.apply_along_axis(sp_stats.rankdata, 1, fvals)
+
+        tm.assert_almost_equal(ranks0.values, exp0)
+        tm.assert_almost_equal(ranks1.values, exp1)
+
+        # top
+        ranks0 = float_frame.rank(na_option="top")
+        ranks1 = float_frame.rank(1, na_option="top")
+
+        fval0 = float_frame.fillna((float_frame.min() - 1).to_dict()).values
+        fval1 = float_frame.T
+        fval1 = fval1.fillna((fval1.min() - 1).to_dict()).T
+        fval1 = fval1.fillna(np.inf).values
+
+        exp0 = np.apply_along_axis(sp_stats.rankdata, 0, fval0)
+        exp1 = np.apply_along_axis(sp_stats.rankdata, 1, fval1)
+
+        tm.assert_almost_equal(ranks0.values, exp0)
+        tm.assert_almost_equal(ranks1.values, exp1)
+
+        # descending
+
+        # bottom
+        ranks0 = float_frame.rank(na_option="top", ascending=False)
+        ranks1 = float_frame.rank(1, na_option="top", ascending=False)
+
+        fvals = float_frame.fillna(np.inf).values
+
+        exp0 = np.apply_along_axis(sp_stats.rankdata, 0, -fvals)
+        exp1 = np.apply_along_axis(sp_stats.rankdata, 1, -fvals)
+
+        tm.assert_almost_equal(ranks0.values, exp0)
+        tm.assert_almost_equal(ranks1.values, exp1)
+
+        # descending
+
+        # top
+        ranks0 = float_frame.rank(na_option="bottom", ascending=False)
+        ranks1 = float_frame.rank(1, na_option="bottom", ascending=False)
+
+        fval0 = float_frame.fillna((float_frame.min() - 1).to_dict()).values
+        fval1 = float_frame.T
+        fval1 = fval1.fillna((fval1.min() - 1).to_dict()).T
+        fval1 = fval1.fillna(np.inf).values
+
+        exp0 = np.apply_along_axis(sp_stats.rankdata, 0, -fval0)
+        exp1 = np.apply_along_axis(sp_stats.rankdata, 1, -fval1)
+
+        tm.assert_numpy_array_equal(ranks0.values, exp0)
+        tm.assert_numpy_array_equal(ranks1.values, exp1)
+
+        # bad values throw error
+        msg = "na_option must be one of 'keep', 'top', or 'bottom'"
+
+        with pytest.raises(ValueError, match=msg):
+            float_frame.rank(na_option="bad", ascending=False)
+
+        # invalid type
+        with pytest.raises(ValueError, match=msg):
+            float_frame.rank(na_option=True, ascending=False)
+
+    def test_rank_axis(self):
+        # check if using axes' names gives the same result
+        df = DataFrame([[2, 1], [4, 3]])
+        tm.assert_frame_equal(df.rank(axis=0), df.rank(axis="index"))
+        tm.assert_frame_equal(df.rank(axis=1), df.rank(axis="columns"))
+
+    @pytest.mark.parametrize("ax", [0, 1])
+    def test_rank_methods_frame(self, ax, rank_method):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        xs = np.random.default_rng(2).integers(0, 21, (100, 26))
+        xs = (xs - 10.0) / 10.0
+        cols = [chr(ord("z") - i) for i in range(xs.shape[1])]
+
+        for vals in [xs, xs + 1e6, xs * 1e-6]:
+            df = DataFrame(vals, columns=cols)
+
+            result = df.rank(axis=ax, method=rank_method)
+            sprank = np.apply_along_axis(
+                sp_stats.rankdata,
+                ax,
+                vals,
+                rank_method if rank_method != "first" else "ordinal",
+            )
+            sprank = sprank.astype(np.float64)
+            expected = DataFrame(sprank, columns=cols).astype("float64")
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["O", "f8", "i8"])
+    def test_rank_descending(self, rank_method, dtype):
+        if "i" in dtype:
+            df = self.df.dropna().astype(dtype)
+        else:
+            df = self.df.astype(dtype)
+
+        res = df.rank(ascending=False)
+        expected = (df.max() - df).rank()
+        tm.assert_frame_equal(res, expected)
+
+        expected = (df.max() - df).rank(method=rank_method)
+
+        if dtype != "O":
+            res2 = df.rank(method=rank_method, ascending=False, numeric_only=True)
+            tm.assert_frame_equal(res2, expected)
+
+        res3 = df.rank(method=rank_method, ascending=False, numeric_only=False)
+        tm.assert_frame_equal(res3, expected)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_rank_2d_tie_methods(self, rank_method, axis, dtype):
+        df = self.df
+
+        def _check2d(df, expected, method="average", axis=0):
+            exp_df = DataFrame({"A": expected, "B": expected})
+
+            if axis == 1:
+                df = df.T
+                exp_df = exp_df.T
+
+            result = df.rank(method=rank_method, axis=axis)
+            tm.assert_frame_equal(result, exp_df)
+
+        frame = df if dtype is None else df.astype(dtype)
+        _check2d(frame, self.results[rank_method], method=rank_method, axis=axis)
+
+    @pytest.mark.parametrize(
+        "rank_method,exp",
+        [
+            ("dense", [[1.0, 1.0, 1.0], [1.0, 0.5, 2.0 / 3], [1.0, 0.5, 1.0 / 3]]),
+            (
+                "min",
+                [
+                    [1.0 / 3, 1.0, 1.0],
+                    [1.0 / 3, 1.0 / 3, 2.0 / 3],
+                    [1.0 / 3, 1.0 / 3, 1.0 / 3],
+                ],
+            ),
+            (
+                "max",
+                [[1.0, 1.0, 1.0], [1.0, 2.0 / 3, 2.0 / 3], [1.0, 2.0 / 3, 1.0 / 3]],
+            ),
+            (
+                "average",
+                [[2.0 / 3, 1.0, 1.0], [2.0 / 3, 0.5, 2.0 / 3], [2.0 / 3, 0.5, 1.0 / 3]],
+            ),
+            (
+                "first",
+                [
+                    [1.0 / 3, 1.0, 1.0],
+                    [2.0 / 3, 1.0 / 3, 2.0 / 3],
+                    [3.0 / 3, 2.0 / 3, 1.0 / 3],
+                ],
+            ),
+        ],
+    )
+    def test_rank_pct_true(self, rank_method, exp):
+        # see gh-15630.
+
+        df = DataFrame([[2012, 66, 3], [2012, 65, 2], [2012, 65, 1]])
+        result = df.rank(method=rank_method, pct=True)
+
+        expected = DataFrame(exp)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.single_cpu
+    def test_pct_max_many_rows(self):
+        # GH 18271
+        df = DataFrame({"A": np.arange(2**24 + 1), "B": np.arange(2**24 + 1, 0, -1)})
+        result = df.rank(pct=True).max()
+        assert (result == 1).all()
+
+    @pytest.mark.parametrize(
+        "contents,dtype",
+        [
+            (
+                [
+                    -np.inf,
+                    -50,
+                    -1,
+                    -1e-20,
+                    -1e-25,
+                    -1e-50,
+                    0,
+                    1e-40,
+                    1e-20,
+                    1e-10,
+                    2,
+                    40,
+                    np.inf,
+                ],
+                "float64",
+            ),
+            (
+                [
+                    -np.inf,
+                    -50,
+                    -1,
+                    -1e-20,
+                    -1e-25,
+                    -1e-45,
+                    0,
+                    1e-40,
+                    1e-20,
+                    1e-10,
+                    2,
+                    40,
+                    np.inf,
+                ],
+                "float32",
+            ),
+            ([np.iinfo(np.uint8).min, 1, 2, 100, np.iinfo(np.uint8).max], "uint8"),
+            (
+                [
+                    np.iinfo(np.int64).min,
+                    -100,
+                    0,
+                    1,
+                    9999,
+                    100000,
+                    1e10,
+                    np.iinfo(np.int64).max,
+                ],
+                "int64",
+            ),
+            ([NegInfinity(), "1", "A", "BA", "Ba", "C", Infinity()], "object"),
+            (
+                [datetime(2001, 1, 1), datetime(2001, 1, 2), datetime(2001, 1, 5)],
+                "datetime64",
+            ),
+        ],
+    )
+    def test_rank_inf_and_nan(self, contents, dtype, frame_or_series):
+        dtype_na_map = {
+            "float64": np.nan,
+            "float32": np.nan,
+            "object": None,
+            "datetime64": np.datetime64("nat"),
+        }
+        # Insert nans at random positions if underlying dtype has missing
+        # value. Then adjust the expected order by adding nans accordingly
+        # This is for testing whether rank calculation is affected
+        # when values are intertwined with nan values.
+        values = np.array(contents, dtype=dtype)
+        exp_order = np.array(range(len(values)), dtype="float64") + 1.0
+        if dtype in dtype_na_map:
+            na_value = dtype_na_map[dtype]
+            nan_indices = np.random.default_rng(2).choice(range(len(values)), 5)
+            values = np.insert(values, nan_indices, na_value)
+            exp_order = np.insert(exp_order, nan_indices, np.nan)
+
+        # Shuffle the testing array and expected results in the same way
+        random_order = np.random.default_rng(2).permutation(len(values))
+        obj = frame_or_series(values[random_order])
+        expected = frame_or_series(exp_order[random_order], dtype="float64")
+        result = obj.rank()
+        tm.assert_equal(result, expected)
+
+    def test_df_series_inf_nan_consistency(self):
+        # GH#32593
+        index = [5, 4, 3, 2, 1, 6, 7, 8, 9, 10]
+        col1 = [5, 4, 3, 5, 8, 5, 2, 1, 6, 6]
+        col2 = [5, 4, np.nan, 5, 8, 5, np.inf, np.nan, 6, -np.inf]
+        df = DataFrame(
+            data={
+                "col1": col1,
+                "col2": col2,
+            },
+            index=index,
+            dtype="f8",
+        )
+        df_result = df.rank()
+
+        series_result = df.copy()
+        series_result["col1"] = df["col1"].rank()
+        series_result["col2"] = df["col2"].rank()
+
+        tm.assert_frame_equal(df_result, series_result)
+
+    def test_rank_both_inf(self):
+        # GH#32593
+        df = DataFrame({"a": [-np.inf, 0, np.inf]})
+        expected = DataFrame({"a": [1.0, 2.0, 3.0]})
+        result = df.rank()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "na_option,ascending,expected",
+        [
+            ("top", True, [3.0, 1.0, 2.0]),
+            ("top", False, [2.0, 1.0, 3.0]),
+            ("bottom", True, [2.0, 3.0, 1.0]),
+            ("bottom", False, [1.0, 3.0, 2.0]),
+        ],
+    )
+    def test_rank_inf_nans_na_option(
+        self, frame_or_series, rank_method, na_option, ascending, expected
+    ):
+        obj = frame_or_series([np.inf, np.nan, -np.inf])
+        result = obj.rank(method=rank_method, na_option=na_option, ascending=ascending)
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "na_option,ascending,expected",
+        [
+            ("bottom", True, [1.0, 2.0, 4.0, 3.0]),
+            ("bottom", False, [1.0, 2.0, 4.0, 3.0]),
+            ("top", True, [2.0, 3.0, 1.0, 4.0]),
+            ("top", False, [2.0, 3.0, 1.0, 4.0]),
+        ],
+    )
+    def test_rank_object_first(self, frame_or_series, na_option, ascending, expected):
+        obj = frame_or_series(["foo", "foo", None, "foo"])
+        result = obj.rank(method="first", na_option=na_option, ascending=ascending)
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data,expected",
+        [
+            (
+                {"a": [1, 2, "a"], "b": [4, 5, 6]},
+                DataFrame({"b": [1.0, 2.0, 3.0]}, columns=Index(["b"], dtype=object)),
+            ),
+            ({"a": [1, 2, "a"]}, DataFrame(index=range(3), columns=[])),
+        ],
+    )
+    def test_rank_mixed_axis_zero(self, data, expected):
+        df = DataFrame(data, columns=Index(list(data.keys()), dtype=object))
+        with pytest.raises(TypeError, match="'<' not supported between instances of"):
+            df.rank()
+        result = df.rank(numeric_only=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_rank_string_dtype(self, string_dtype_no_object):
+        # GH#55362
+        obj = Series(["foo", "foo", None, "foo"], dtype=string_dtype_no_object)
+        result = obj.rank(method="first")
+        exp_dtype = (
+            "Float64" if string_dtype_no_object == "string[pyarrow]" else "float64"
+        )
+        if string_dtype_no_object.storage == "python":
+            # TODO nullable string[python] should also return nullable Int64
+            exp_dtype = "float64"
+        expected = Series([1, 2, None, 3], dtype=exp_dtype)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_reindex.py b/pandas/pandas/tests/frame/methods/test_reindex.py
new file mode 100644
index 0000000000000..37adc31fb0f4d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_reindex.py
@@ -0,0 +1,1260 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+import inspect
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.timezones import dateutil_gettz as gettz
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+from pandas.compat.numpy import np_version_gt2
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+
+
+class TestReindexSetIndex:
+    # Tests that check both reindex and set_index
+
+    def test_dti_set_index_reindex_datetimeindex(self):
+        # GH#6631
+        df = DataFrame(np.random.default_rng(2).random(6))
+        idx1 = date_range("2011/01/01", periods=6, freq="ME", tz="US/Eastern")
+        idx2 = date_range("2013", periods=6, freq="YE", tz="Asia/Tokyo")
+
+        df = df.set_index(idx1)
+        tm.assert_index_equal(df.index, idx1)
+        df = df.reindex(idx2)
+        tm.assert_index_equal(df.index, idx2)
+
+    def test_dti_set_index_reindex_freq_with_tz(self):
+        # GH#11314 with tz
+        index = date_range(
+            datetime(2015, 10, 1), datetime(2015, 10, 1, 23), freq="h", tz="US/Eastern"
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((24, 1)),
+            columns=["a"],
+            index=index,
+        )
+        new_index = date_range(
+            datetime(2015, 10, 2), datetime(2015, 10, 2, 23), freq="h", tz="US/Eastern"
+        )
+
+        result = df.set_index(new_index)
+        assert result.index.freq == index.freq
+
+    def test_set_reset_index_intervalindex(self):
+        df = DataFrame({"A": range(10)})
+        ser = pd.cut(df.A, 5)
+        df["B"] = ser
+        df = df.set_index("B")
+
+        df = df.reset_index()
+
+    def test_setitem_reset_index_dtypes(self):
+        # GH 22060
+        df = DataFrame(columns=["a", "b", "c"]).astype(
+            {"a": "datetime64[ns]", "b": np.int64, "c": np.float64}
+        )
+        df1 = df.set_index(["a"])
+        df1["d"] = []
+        result = df1.reset_index()
+        expected = DataFrame(columns=["a", "b", "c", "d"], index=range(0)).astype(
+            {"a": "datetime64[ns]", "b": np.int64, "c": np.float64, "d": np.float64}
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df2 = df.set_index(["a", "b"])
+        df2["d"] = []
+        result = df2.reset_index()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "timezone, year, month, day, hour",
+        [["America/Chicago", 2013, 11, 3, 1], ["America/Santiago", 2021, 4, 3, 23]],
+    )
+    def test_reindex_timestamp_with_fold(self, timezone, year, month, day, hour):
+        # see gh-40817
+        test_timezone = gettz(timezone)
+        transition_1 = pd.Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=0,
+            tzinfo=test_timezone,
+        )
+        transition_2 = pd.Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=1,
+            tzinfo=test_timezone,
+        )
+        df = (
+            DataFrame({"index": [transition_1, transition_2], "vals": ["a", "b"]})
+            .set_index("index")
+            .reindex(["1", "2"])
+        )
+        exp = DataFrame({"index": ["1", "2"], "vals": [np.nan, np.nan]}).set_index(
+            "index"
+        )
+        exp = exp.astype(df.vals.dtype)
+        tm.assert_frame_equal(
+            df,
+            exp,
+        )
+
+
+class TestDataFrameSelectReindex:
+    # These are specific reindex-based tests; other indexing tests should go in
+    # test_indexing
+
+    @pytest.mark.xfail(
+        not IS64 or (is_platform_windows() and not np_version_gt2),
+        reason="Passes int32 values to DatetimeArray in make_na_array on "
+        "windows, 32bit linux builds",
+    )
+    def test_reindex_tzaware_fill_value(self):
+        # GH#52586
+        df = DataFrame([[1]])
+
+        ts = pd.Timestamp("2023-04-10 17:32", tz="US/Pacific")
+        res = df.reindex([0, 1], axis=1, fill_value=ts)
+        assert res.dtypes[1] == pd.DatetimeTZDtype(unit="s", tz="US/Pacific")
+        expected = DataFrame({0: [1], 1: [ts]})
+        expected[1] = expected[1].astype(res.dtypes[1])
+        tm.assert_frame_equal(res, expected)
+
+        per = ts.tz_localize(None).to_period("s")
+        res = df.reindex([0, 1], axis=1, fill_value=per)
+        assert res.dtypes[1] == pd.PeriodDtype("s")
+        expected = DataFrame({0: [1], 1: [per]})
+        tm.assert_frame_equal(res, expected)
+
+        interval = pd.Interval(ts, ts + pd.Timedelta(seconds=1))
+        res = df.reindex([0, 1], axis=1, fill_value=interval)
+        assert res.dtypes[1] == pd.IntervalDtype("datetime64[s, US/Pacific]", "right")
+        expected = DataFrame({0: [1], 1: [interval]})
+        expected[1] = expected[1].astype(res.dtypes[1])
+        tm.assert_frame_equal(res, expected)
+
+    def test_reindex_date_fill_value(self):
+        # passing date to dt64 is deprecated; enforced in 2.0 to cast to object
+        arr = date_range("2016-01-01", periods=6).values.reshape(3, 2)
+        df = DataFrame(arr, columns=["A", "B"], index=range(3))
+
+        ts = df.iloc[0, 0]
+        fv = ts.date()
+
+        res = df.reindex(index=range(4), columns=["A", "B", "C"], fill_value=fv)
+
+        expected = DataFrame(
+            {"A": df["A"].tolist() + [fv], "B": df["B"].tolist() + [fv], "C": [fv] * 4},
+            dtype=object,
+        )
+        tm.assert_frame_equal(res, expected)
+
+        # only reindexing rows
+        res = df.reindex(index=range(4), fill_value=fv)
+        tm.assert_frame_equal(res, expected[["A", "B"]])
+
+        # same with a datetime-castable str
+        res = df.reindex(
+            index=range(4), columns=["A", "B", "C"], fill_value="2016-01-01"
+        )
+        expected = DataFrame(
+            {"A": df["A"].tolist() + [ts], "B": df["B"].tolist() + [ts], "C": [ts] * 4},
+        )
+        tm.assert_frame_equal(res, expected)
+
+    def test_reindex_with_multi_index(self):
+        # https://github.com/pandas-dev/pandas/issues/29896
+        # tests for reindexing a multi-indexed DataFrame with a new MultiIndex
+        #
+        # confirms that we can reindex a multi-indexed DataFrame with a new
+        # MultiIndex object correctly when using no filling, backfilling, and
+        # padding
+        #
+        # The DataFrame, `df`, used in this test is:
+        #       c
+        #  a b
+        # -1 0  A
+        #    1  B
+        #    2  C
+        #    3  D
+        #    4  E
+        #    5  F
+        #    6  G
+        #  0 0  A
+        #    1  B
+        #    2  C
+        #    3  D
+        #    4  E
+        #    5  F
+        #    6  G
+        #  1 0  A
+        #    1  B
+        #    2  C
+        #    3  D
+        #    4  E
+        #    5  F
+        #    6  G
+        #
+        # and the other MultiIndex, `new_multi_index`, is:
+        # 0: 0 0.5
+        # 1:   2.0
+        # 2:   5.0
+        # 3:   5.8
+        df = DataFrame(
+            {
+                "a": [-1] * 7 + [0] * 7 + [1] * 7,
+                "b": list(range(7)) * 3,
+                "c": ["A", "B", "C", "D", "E", "F", "G"] * 3,
+            }
+        ).set_index(["a", "b"])
+        new_index = [0.5, 2.0, 5.0, 5.8]
+        new_multi_index = MultiIndex.from_product([[0], new_index], names=["a", "b"])
+
+        # reindexing w/o a `method` value
+        reindexed = df.reindex(new_multi_index)
+        expected = DataFrame(
+            {"a": [0] * 4, "b": new_index, "c": [np.nan, "C", "F", np.nan]}
+        ).set_index(["a", "b"])
+        tm.assert_frame_equal(expected, reindexed)
+
+        # reindexing with backfilling
+        expected = DataFrame(
+            {"a": [0] * 4, "b": new_index, "c": ["B", "C", "F", "G"]}
+        ).set_index(["a", "b"])
+        reindexed_with_backfilling = df.reindex(new_multi_index, method="bfill")
+        tm.assert_frame_equal(expected, reindexed_with_backfilling)
+
+        reindexed_with_backfilling = df.reindex(new_multi_index, method="backfill")
+        tm.assert_frame_equal(expected, reindexed_with_backfilling)
+
+        # reindexing with padding
+        expected = DataFrame(
+            {"a": [0] * 4, "b": new_index, "c": ["A", "C", "F", "F"]}
+        ).set_index(["a", "b"])
+        reindexed_with_padding = df.reindex(new_multi_index, method="pad")
+        tm.assert_frame_equal(expected, reindexed_with_padding)
+
+        reindexed_with_padding = df.reindex(new_multi_index, method="ffill")
+        tm.assert_frame_equal(expected, reindexed_with_padding)
+
+    @pytest.mark.parametrize(
+        "method,expected_values",
+        [
+            ("nearest", [0, 1, 1, 2]),
+            ("pad", [np.nan, 0, 1, 1]),
+            ("backfill", [0, 1, 2, 2]),
+        ],
+    )
+    def test_reindex_methods(self, method, expected_values):
+        df = DataFrame({"x": list(range(5))})
+        target = np.array([-0.1, 0.9, 1.1, 1.5])
+
+        expected = DataFrame({"x": expected_values}, index=target)
+        actual = df.reindex(target, method=method)
+        tm.assert_frame_equal(expected, actual)
+
+        actual = df.reindex(target, method=method, tolerance=1)
+        tm.assert_frame_equal(expected, actual)
+        actual = df.reindex(target, method=method, tolerance=[1, 1, 1, 1])
+        tm.assert_frame_equal(expected, actual)
+
+        e2 = expected[::-1]
+        actual = df.reindex(target[::-1], method=method)
+        tm.assert_frame_equal(e2, actual)
+
+        new_order = [3, 0, 2, 1]
+        e2 = expected.iloc[new_order]
+        actual = df.reindex(target[new_order], method=method)
+        tm.assert_frame_equal(e2, actual)
+
+        switched_method = (
+            "pad" if method == "backfill" else "backfill" if method == "pad" else method
+        )
+        actual = df[::-1].reindex(target, method=switched_method)
+        tm.assert_frame_equal(expected, actual)
+
+    def test_reindex_methods_nearest_special(self):
+        df = DataFrame({"x": list(range(5))})
+        target = np.array([-0.1, 0.9, 1.1, 1.5])
+
+        expected = DataFrame({"x": [0, 1, 1, np.nan]}, index=target)
+        actual = df.reindex(target, method="nearest", tolerance=0.2)
+        tm.assert_frame_equal(expected, actual)
+
+        expected = DataFrame({"x": [0, np.nan, 1, np.nan]}, index=target)
+        actual = df.reindex(target, method="nearest", tolerance=[0.5, 0.01, 0.4, 0.1])
+        tm.assert_frame_equal(expected, actual)
+
+    def test_reindex_nearest_tz(self, tz_aware_fixture):
+        # GH26683
+        tz = tz_aware_fixture
+        idx = date_range("2019-01-01", periods=5, tz=tz)
+        df = DataFrame({"x": list(range(5))}, index=idx)
+
+        expected = df.head(3)
+        actual = df.reindex(idx[:3], method="nearest")
+        tm.assert_frame_equal(expected, actual)
+
+    def test_reindex_nearest_tz_empty_frame(self):
+        # https://github.com/pandas-dev/pandas/issues/31964
+        dti = pd.DatetimeIndex(["2016-06-26 14:27:26+00:00"])
+        df = DataFrame(index=pd.DatetimeIndex(["2016-07-04 14:00:59+00:00"]))
+        expected = DataFrame(index=dti)
+        result = df.reindex(dti, method="nearest")
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_frame_add_nat(self):
+        rng = date_range("1/1/2000 00:00:00", periods=10, freq="10s")
+        df = DataFrame(
+            {"A": np.random.default_rng(2).standard_normal(len(rng)), "B": rng}
+        )
+
+        result = df.reindex(range(15))
+        assert np.issubdtype(result["B"].dtype, np.dtype("M8[ns]"))
+
+        mask = isna(result)["B"]
+        assert mask[-5:].all()
+        assert not mask[:-5].any()
+
+    @pytest.mark.parametrize(
+        "method, exp_values",
+        [("ffill", [0, 1, 2, 3]), ("bfill", [1.0, 2.0, 3.0, np.nan])],
+    )
+    def test_reindex_frame_tz_ffill_bfill(self, frame_or_series, method, exp_values):
+        # GH#38566
+        obj = frame_or_series(
+            [0, 1, 2, 3],
+            index=date_range("2020-01-01 00:00:00", periods=4, freq="h", tz="UTC"),
+        )
+        new_index = date_range("2020-01-01 00:01:00", periods=4, freq="h", tz="UTC")
+        result = obj.reindex(new_index, method=method, tolerance=pd.Timedelta("1 hour"))
+        expected = frame_or_series(exp_values, index=new_index)
+        tm.assert_equal(result, expected)
+
+    def test_reindex_limit(self):
+        # GH 28631
+        data = [["A", "A", "A"], ["B", "B", "B"], ["C", "C", "C"], ["D", "D", "D"]]
+        exp_data = [
+            ["A", "A", "A"],
+            ["B", "B", "B"],
+            ["C", "C", "C"],
+            ["D", "D", "D"],
+            ["D", "D", "D"],
+            [np.nan, np.nan, np.nan],
+        ]
+        df = DataFrame(data)
+        result = df.reindex([0, 1, 2, 3, 4, 5], method="ffill", limit=1)
+        expected = DataFrame(exp_data)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "idx, check_index_type",
+        [
+            [["C", "B", "A"], True],
+            [["F", "C", "A", "D"], True],
+            [["A"], True],
+            [["A", "B", "C"], True],
+            [["C", "A", "B"], True],
+            [["C", "B"], True],
+            [["C", "A"], True],
+            [["A", "B"], True],
+            [["B", "A", "C"], True],
+            # reindex by these causes different MultiIndex levels
+            [["D", "F"], False],
+            [["A", "C", "B"], False],
+        ],
+    )
+    def test_reindex_level_verify_first_level(self, idx, check_index_type):
+        df = DataFrame(
+            {
+                "jim": list("B" * 4 + "A" * 2 + "C" * 3),
+                "joe": list("abcdeabcd")[::-1],
+                "jolie": [10, 20, 30] * 3,
+                "joline": np.random.default_rng(2).integers(0, 1000, 9),
+            }
+        )
+        icol = ["jim", "joe", "jolie"]
+
+        def f(val):
+            return np.nonzero((df["jim"] == val).to_numpy())[0]
+
+        i = np.concatenate(list(map(f, idx)))
+        left = df.set_index(icol).reindex(idx, level="jim")
+        right = df.iloc[i].set_index(icol)
+        tm.assert_frame_equal(left, right, check_index_type=check_index_type)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            ("mid",),
+            ("mid", "btm"),
+            ("mid", "btm", "top"),
+            ("mid", "top"),
+            ("mid", "top", "btm"),
+            ("btm",),
+            ("btm", "mid"),
+            ("btm", "mid", "top"),
+            ("btm", "top"),
+            ("btm", "top", "mid"),
+            ("top",),
+            ("top", "mid"),
+            ("top", "mid", "btm"),
+            ("top", "btm"),
+            ("top", "btm", "mid"),
+        ],
+    )
+    def test_reindex_level_verify_first_level_repeats(self, idx):
+        df = DataFrame(
+            {
+                "jim": ["mid"] * 5 + ["btm"] * 8 + ["top"] * 7,
+                "joe": ["3rd"] * 2
+                + ["1st"] * 3
+                + ["2nd"] * 3
+                + ["1st"] * 2
+                + ["3rd"] * 3
+                + ["1st"] * 2
+                + ["3rd"] * 3
+                + ["2nd"] * 2,
+                # this needs to be jointly unique with jim and joe or
+                # reindexing will fail ~1.5% of the time, this works
+                # out to needing unique groups of same size as joe
+                "jolie": np.concatenate(
+                    [
+                        np.random.default_rng(2).choice(1000, x, replace=False)
+                        for x in [2, 3, 3, 2, 3, 2, 3, 2]
+                    ]
+                ),
+                "joline": np.random.default_rng(2).standard_normal(20).round(3) * 10,
+            }
+        )
+        icol = ["jim", "joe", "jolie"]
+
+        def f(val):
+            return np.nonzero((df["jim"] == val).to_numpy())[0]
+
+        i = np.concatenate(list(map(f, idx)))
+        left = df.set_index(icol).reindex(idx, level="jim")
+        right = df.iloc[i].set_index(icol)
+        tm.assert_frame_equal(left, right)
+
+    @pytest.mark.parametrize(
+        "idx, indexer",
+        [
+            [
+                ["1st", "2nd", "3rd"],
+                [2, 3, 4, 0, 1, 8, 9, 5, 6, 7, 10, 11, 12, 13, 14, 18, 19, 15, 16, 17],
+            ],
+            [
+                ["3rd", "2nd", "1st"],
+                [0, 1, 2, 3, 4, 10, 11, 12, 5, 6, 7, 8, 9, 15, 16, 17, 18, 19, 13, 14],
+            ],
+            [["2nd", "3rd"], [0, 1, 5, 6, 7, 10, 11, 12, 18, 19, 15, 16, 17]],
+            [["3rd", "1st"], [0, 1, 2, 3, 4, 10, 11, 12, 8, 9, 15, 16, 17, 13, 14]],
+        ],
+    )
+    def test_reindex_level_verify_repeats(self, idx, indexer):
+        df = DataFrame(
+            {
+                "jim": ["mid"] * 5 + ["btm"] * 8 + ["top"] * 7,
+                "joe": ["3rd"] * 2
+                + ["1st"] * 3
+                + ["2nd"] * 3
+                + ["1st"] * 2
+                + ["3rd"] * 3
+                + ["1st"] * 2
+                + ["3rd"] * 3
+                + ["2nd"] * 2,
+                # this needs to be jointly unique with jim and joe or
+                # reindexing will fail ~1.5% of the time, this works
+                # out to needing unique groups of same size as joe
+                "jolie": np.concatenate(
+                    [
+                        np.random.default_rng(2).choice(1000, x, replace=False)
+                        for x in [2, 3, 3, 2, 3, 2, 3, 2]
+                    ]
+                ),
+                "joline": np.random.default_rng(2).standard_normal(20).round(3) * 10,
+            }
+        )
+        icol = ["jim", "joe", "jolie"]
+        left = df.set_index(icol).reindex(idx, level="joe")
+        right = df.iloc[indexer].set_index(icol)
+        tm.assert_frame_equal(left, right)
+
+    @pytest.mark.parametrize(
+        "idx, indexer, check_index_type",
+        [
+            [list("abcde"), [3, 2, 1, 0, 5, 4, 8, 7, 6], True],
+            [list("abcd"), [3, 2, 1, 0, 5, 8, 7, 6], True],
+            [list("abc"), [3, 2, 1, 8, 7, 6], True],
+            [list("eca"), [1, 3, 4, 6, 8], True],
+            [list("edc"), [0, 1, 4, 5, 6], True],
+            [list("eadbc"), [3, 0, 2, 1, 4, 5, 8, 7, 6], True],
+            [list("edwq"), [0, 4, 5], True],
+            [list("wq"), [], False],
+        ],
+    )
+    def test_reindex_level_verify(self, idx, indexer, check_index_type):
+        df = DataFrame(
+            {
+                "jim": list("B" * 4 + "A" * 2 + "C" * 3),
+                "joe": list("abcdeabcd")[::-1],
+                "jolie": [10, 20, 30] * 3,
+                "joline": np.random.default_rng(2).integers(0, 1000, 9),
+            }
+        )
+        icol = ["jim", "joe", "jolie"]
+        left = df.set_index(icol).reindex(idx, level="joe")
+        right = df.iloc[indexer].set_index(icol)
+        tm.assert_frame_equal(left, right, check_index_type=check_index_type)
+
+    def test_non_monotonic_reindex_methods(self):
+        dr = date_range("2013-08-01", periods=6, freq="B")
+        data = np.random.default_rng(2).standard_normal((6, 1))
+        df = DataFrame(data, index=dr, columns=list("A"))
+        df_rev = DataFrame(data, index=dr[[3, 4, 5] + [0, 1, 2]], columns=list("A"))
+        # index is not monotonic increasing or decreasing
+        msg = "index must be monotonic increasing or decreasing"
+        with pytest.raises(ValueError, match=msg):
+            df_rev.reindex(df.index, method="pad")
+        with pytest.raises(ValueError, match=msg):
+            df_rev.reindex(df.index, method="ffill")
+        with pytest.raises(ValueError, match=msg):
+            df_rev.reindex(df.index, method="bfill")
+        with pytest.raises(ValueError, match=msg):
+            df_rev.reindex(df.index, method="nearest")
+
+    def test_reindex_sparse(self):
+        # https://github.com/pandas-dev/pandas/issues/35286
+        df = DataFrame(
+            {"A": [0, 1], "B": pd.array([0, 1], dtype=pd.SparseDtype("int64", 0))}
+        )
+        result = df.reindex([0, 2])
+        expected = DataFrame(
+            {
+                "A": [0.0, np.nan],
+                "B": pd.array([0.0, np.nan], dtype=pd.SparseDtype("float64", 0.0)),
+            },
+            index=[0, 2],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex(self, float_frame):
+        datetime_series = Series(
+            np.arange(30, dtype=np.float64), index=date_range("2020-01-01", periods=30)
+        )
+
+        newFrame = float_frame.reindex(datetime_series.index)
+
+        for col in newFrame.columns:
+            for idx, val in newFrame[col].items():
+                if idx in float_frame.index:
+                    if np.isnan(val):
+                        assert np.isnan(float_frame[col][idx])
+                    else:
+                        assert val == float_frame[col][idx]
+                else:
+                    assert np.isnan(val)
+
+        for col, series in newFrame.items():
+            tm.assert_index_equal(series.index, newFrame.index)
+        emptyFrame = float_frame.reindex(Index([]))
+        assert len(emptyFrame.index) == 0
+
+        # Cython code should be unit-tested directly
+        nonContigFrame = float_frame.reindex(datetime_series.index[::2])
+
+        for col in nonContigFrame.columns:
+            for idx, val in nonContigFrame[col].items():
+                if idx in float_frame.index:
+                    if np.isnan(val):
+                        assert np.isnan(float_frame[col][idx])
+                    else:
+                        assert val == float_frame[col][idx]
+                else:
+                    assert np.isnan(val)
+
+        for col, series in nonContigFrame.items():
+            tm.assert_index_equal(series.index, nonContigFrame.index)
+
+        # corner cases
+        newFrame = float_frame.reindex(float_frame.index)
+        assert newFrame.index.is_(float_frame.index)
+
+        # length zero
+        newFrame = float_frame.reindex([])
+        assert newFrame.empty
+        assert len(newFrame.columns) == len(float_frame.columns)
+
+        # length zero with columns reindexed with non-empty index
+        newFrame = float_frame.reindex([])
+        newFrame = newFrame.reindex(float_frame.index)
+        assert len(newFrame.index) == len(float_frame.index)
+        assert len(newFrame.columns) == len(float_frame.columns)
+
+        # pass non-Index
+        newFrame = float_frame.reindex(list(datetime_series.index))
+        expected = datetime_series.index._with_freq(None)
+        tm.assert_index_equal(newFrame.index, expected)
+
+        # copy with no axes
+        result = float_frame.reindex()
+        tm.assert_frame_equal(result, float_frame)
+        assert result is not float_frame
+
+    def test_reindex_nan(self):
+        df = DataFrame(
+            [[1, 2], [3, 5], [7, 11], [9, 23]],
+            index=[2, np.nan, 1, 5],
+            columns=["joe", "jim"],
+        )
+
+        i, j = [np.nan, 5, 5, np.nan, 1, 2, np.nan], [1, 3, 3, 1, 2, 0, 1]
+        tm.assert_frame_equal(df.reindex(i), df.iloc[j])
+
+        df.index = df.index.astype("object")
+        tm.assert_frame_equal(df.reindex(i), df.iloc[j], check_index_type=False)
+
+        # GH10388
+        df = DataFrame(
+            {
+                "other": ["a", "b", np.nan, "c"],
+                "date": ["2015-03-22", np.nan, "2012-01-08", np.nan],
+                "amount": [2, 3, 4, 5],
+            }
+        )
+
+        df["date"] = pd.to_datetime(df.date)
+        df["delta"] = (pd.to_datetime("2015-06-18") - df["date"]).shift(1)
+
+        left = df.set_index(["delta", "other", "date"]).reset_index()
+        right = df.reindex(columns=["delta", "other", "date", "amount"])
+        tm.assert_frame_equal(left, right)
+
+    def test_reindex_name_remains(self):
+        s = Series(np.random.default_rng(2).random(10))
+        df = DataFrame(s, index=np.arange(len(s)))
+        i = Series(np.arange(10), name="iname")
+
+        df = df.reindex(i)
+        assert df.index.name == "iname"
+
+        df = df.reindex(Index(np.arange(10), name="tmpname"))
+        assert df.index.name == "tmpname"
+
+        s = Series(np.random.default_rng(2).random(10))
+        df = DataFrame(s.T, index=np.arange(len(s)))
+        i = Series(np.arange(10), name="iname")
+        df = df.reindex(columns=i)
+        assert df.columns.name == "iname"
+
+    def test_reindex_int(self, int_frame):
+        smaller = int_frame.reindex(int_frame.index[::2])
+
+        assert smaller["A"].dtype == np.int64
+
+        bigger = smaller.reindex(int_frame.index)
+        assert bigger["A"].dtype == np.float64
+
+        smaller = int_frame.reindex(columns=["A", "B"])
+        assert smaller["A"].dtype == np.int64
+
+    def test_reindex_columns(self, float_frame):
+        new_frame = float_frame.reindex(columns=["A", "B", "E"])
+
+        tm.assert_series_equal(new_frame["B"], float_frame["B"])
+        assert np.isnan(new_frame["E"]).all()
+        assert "C" not in new_frame
+
+        # Length zero
+        new_frame = float_frame.reindex(columns=[])
+        assert new_frame.empty
+
+    def test_reindex_columns_method(self):
+        # GH 14992, reindexing over columns ignored method
+        df = DataFrame(
+            data=[[11, 12, 13], [21, 22, 23], [31, 32, 33]],
+            index=[1, 2, 4],
+            columns=[1, 2, 4],
+            dtype=float,
+        )
+
+        # default method
+        result = df.reindex(columns=range(6))
+        expected = DataFrame(
+            data=[
+                [np.nan, 11, 12, np.nan, 13, np.nan],
+                [np.nan, 21, 22, np.nan, 23, np.nan],
+                [np.nan, 31, 32, np.nan, 33, np.nan],
+            ],
+            index=[1, 2, 4],
+            columns=range(6),
+            dtype=float,
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # method='ffill'
+        result = df.reindex(columns=range(6), method="ffill")
+        expected = DataFrame(
+            data=[
+                [np.nan, 11, 12, 12, 13, 13],
+                [np.nan, 21, 22, 22, 23, 23],
+                [np.nan, 31, 32, 32, 33, 33],
+            ],
+            index=[1, 2, 4],
+            columns=range(6),
+            dtype=float,
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # method='bfill'
+        result = df.reindex(columns=range(6), method="bfill")
+        expected = DataFrame(
+            data=[
+                [11, 11, 12, 13, 13, np.nan],
+                [21, 21, 22, 23, 23, np.nan],
+                [31, 31, 32, 33, 33, np.nan],
+            ],
+            index=[1, 2, 4],
+            columns=range(6),
+            dtype=float,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_axes(self):
+        # GH 3317, reindexing by both axes loses freq of the index
+        df = DataFrame(
+            np.ones((3, 3)),
+            index=[datetime(2012, 1, 1), datetime(2012, 1, 2), datetime(2012, 1, 3)],
+            columns=["a", "b", "c"],
+        )
+
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            time_freq = date_range("2012-01-01", "2012-01-03", freq="d")
+        some_cols = ["a", "b"]
+
+        index_freq = df.reindex(index=time_freq).index.freq
+        both_freq = df.reindex(index=time_freq, columns=some_cols).index.freq
+        seq_freq = df.reindex(index=time_freq).reindex(columns=some_cols).index.freq
+        assert index_freq == both_freq
+        assert index_freq == seq_freq
+
+    def test_reindex_fill_value(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+
+        # axis=0
+        result = df.reindex(list(range(15)))
+        assert np.isnan(result.values[-5:]).all()
+
+        result = df.reindex(range(15), fill_value=0)
+        expected = df.reindex(range(15)).fillna(0)
+        tm.assert_frame_equal(result, expected)
+
+        # axis=1
+        result = df.reindex(columns=range(5), fill_value=0.0)
+        expected = df.copy()
+        expected[4] = 0.0
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex(columns=range(5), fill_value=0)
+        expected = df.copy()
+        expected[4] = 0
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex(columns=range(5), fill_value="foo")
+        expected = df.copy()
+        expected[4] = "foo"
+        tm.assert_frame_equal(result, expected)
+
+        # other dtypes
+        df["foo"] = "foo"
+        result = df.reindex(range(15), fill_value="0")
+        expected = df.reindex(range(15)).fillna("0")
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_uint_dtypes_fill_value(self, any_unsigned_int_numpy_dtype):
+        # GH#48184
+        df = DataFrame({"a": [1, 2], "b": [1, 2]}, dtype=any_unsigned_int_numpy_dtype)
+        result = df.reindex(columns=list("abcd"), index=[0, 1, 2, 3], fill_value=10)
+        expected = DataFrame(
+            {"a": [1, 2, 10, 10], "b": [1, 2, 10, 10], "c": 10, "d": 10},
+            dtype=any_unsigned_int_numpy_dtype,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_single_column_ea_index_and_columns(self, any_numeric_ea_dtype):
+        # GH#48190
+        df = DataFrame({"a": [1, 2]}, dtype=any_numeric_ea_dtype)
+        result = df.reindex(columns=list("ab"), index=[0, 1, 2], fill_value=10)
+        expected = DataFrame(
+            {"a": Series([1, 2, 10], dtype=any_numeric_ea_dtype), "b": 10}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_dups(self):
+        # GH4746, reindex on duplicate index error messages
+        arr = np.random.default_rng(2).standard_normal(10)
+        df = DataFrame(arr, index=[1, 2, 3, 4, 5, 1, 2, 3, 4, 5])
+
+        # set index is ok
+        result = df.copy()
+        result.index = list(range(len(df)))
+        expected = DataFrame(arr, index=list(range(len(df))))
+        tm.assert_frame_equal(result, expected)
+
+        # reindex fails
+        msg = "cannot reindex on an axis with duplicate labels"
+        with pytest.raises(ValueError, match=msg):
+            df.reindex(index=list(range(len(df))))
+
+    def test_reindex_with_duplicate_columns(self):
+        # reindex is invalid!
+        df = DataFrame(
+            [[1, 5, 7.0], [1, 5, 7.0], [1, 5, 7.0]], columns=["bar", "a", "a"]
+        )
+        msg = "cannot reindex on an axis with duplicate labels"
+        with pytest.raises(ValueError, match=msg):
+            df.reindex(columns=["bar"])
+        with pytest.raises(ValueError, match=msg):
+            df.reindex(columns=["bar", "foo"])
+
+    def test_reindex_axis_style(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        expected = DataFrame(
+            {"A": [1, 2, np.nan], "B": [4, 5, np.nan]}, index=[0, 1, 3]
+        )
+        result = df.reindex([0, 1, 3])
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex([0, 1, 3], axis=0)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex([0, 1, 3], axis="index")
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_positional_raises(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        # Enforced in 2.0
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        msg = r"reindex\(\) takes from 1 to 2 positional arguments but 3 were given"
+        with pytest.raises(TypeError, match=msg):
+            df.reindex([0, 1], ["A", "B", "C"])
+
+    def test_reindex_axis_style_raises(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex([0, 1], columns=["A"], axis=1)
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex([0, 1], columns=["A"], axis="index")
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(index=[0, 1], axis="index")
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(index=[0, 1], axis="columns")
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(columns=[0, 1], axis="columns")
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(index=[0, 1], columns=[0, 1], axis="columns")
+
+        with pytest.raises(TypeError, match="Cannot specify all"):
+            df.reindex(labels=[0, 1], index=[0], columns=["A"])
+
+        # Mixing styles
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(index=[0, 1], axis="index")
+
+        with pytest.raises(TypeError, match="Cannot specify both 'axis'"):
+            df.reindex(index=[0, 1], axis="columns")
+
+        # Duplicates
+        with pytest.raises(TypeError, match="multiple values"):
+            df.reindex([0, 1], labels=[0, 1])
+
+    def test_reindex_single_named_indexer(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        df = DataFrame({"A": [1, 2, 3], "B": [1, 2, 3]})
+        result = df.reindex([0, 1], columns=["A"])
+        expected = DataFrame({"A": [1, 2]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_api_equivalence(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        # equivalence of the labels/axis and index/columns API's
+        df = DataFrame(
+            [[1, 2, 3], [3, 4, 5], [5, 6, 7]],
+            index=["a", "b", "c"],
+            columns=["d", "e", "f"],
+        )
+
+        res1 = df.reindex(["b", "a"])
+        res2 = df.reindex(index=["b", "a"])
+        res3 = df.reindex(labels=["b", "a"])
+        res4 = df.reindex(labels=["b", "a"], axis=0)
+        res5 = df.reindex(["b", "a"], axis=0)
+        for res in [res2, res3, res4, res5]:
+            tm.assert_frame_equal(res1, res)
+
+        res1 = df.reindex(columns=["e", "d"])
+        res2 = df.reindex(["e", "d"], axis=1)
+        res3 = df.reindex(labels=["e", "d"], axis=1)
+        for res in [res2, res3]:
+            tm.assert_frame_equal(res1, res)
+
+        res1 = df.reindex(index=["b", "a"], columns=["e", "d"])
+        res2 = df.reindex(columns=["e", "d"], index=["b", "a"])
+        res3 = df.reindex(labels=["b", "a"], axis=0).reindex(labels=["e", "d"], axis=1)
+        for res in [res2, res3]:
+            tm.assert_frame_equal(res1, res)
+
+    def test_reindex_boolean(self):
+        frame = DataFrame(
+            np.ones((10, 2), dtype=bool), index=np.arange(0, 20, 2), columns=[0, 2]
+        )
+
+        reindexed = frame.reindex(np.arange(10))
+        assert reindexed.values.dtype == np.object_
+        assert isna(reindexed[0][1])
+
+        reindexed = frame.reindex(columns=range(3))
+        assert reindexed.values.dtype == np.object_
+        assert isna(reindexed[1]).all()
+
+    def test_reindex_objects(self, float_string_frame):
+        reindexed = float_string_frame.reindex(columns=["foo", "A", "B"])
+        assert "foo" in reindexed
+
+        reindexed = float_string_frame.reindex(columns=["A", "B"])
+        assert "foo" not in reindexed
+
+    def test_reindex_corner(self, int_frame):
+        index = Index(["a", "b", "c"])
+        dm = DataFrame({}).reindex(index=[1, 2, 3])
+        reindexed = dm.reindex(columns=index)
+        tm.assert_index_equal(reindexed.columns, index)
+
+        # ints are weird
+        smaller = int_frame.reindex(columns=["A", "B", "E"])
+        assert smaller["E"].dtype == np.float64
+
+    def test_reindex_with_nans(self):
+        df = DataFrame(
+            [[1, 2], [3, 4], [np.nan, np.nan], [7, 8], [9, 10]],
+            columns=["a", "b"],
+            index=[100.0, 101.0, np.nan, 102.0, 103.0],
+        )
+
+        result = df.reindex(index=[101.0, 102.0, 103.0])
+        expected = df.iloc[[1, 3, 4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex(index=[103.0])
+        expected = df.iloc[[4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.reindex(index=[101.0])
+        expected = df.iloc[[1]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_without_upcasting(self):
+        # GH45857
+        df = DataFrame(np.zeros((10, 10), dtype=np.float32))
+        result = df.reindex(columns=np.arange(5, 15))
+        assert result.dtypes.eq(np.float32).all()
+
+    def test_reindex_multi(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 3)))
+
+        result = df.reindex(index=range(4), columns=range(4))
+        expected = df.reindex(list(range(4))).reindex(columns=range(4))
+
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(np.random.default_rng(2).integers(0, 10, (3, 3)))
+
+        result = df.reindex(index=range(4), columns=range(4))
+        expected = df.reindex(list(range(4))).reindex(columns=range(4))
+
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(np.random.default_rng(2).integers(0, 10, (3, 3)))
+
+        result = df.reindex(index=range(2), columns=range(2))
+        expected = df.reindex(range(2)).reindex(columns=range(2))
+
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)) + 1j,
+            columns=["a", "b", "c"],
+        )
+
+        result = df.reindex(index=[0, 1], columns=["a", "b"])
+        expected = df.reindex([0, 1]).reindex(columns=["a", "b"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_multi_categorical_time(self):
+        # https://github.com/pandas-dev/pandas/issues/21390
+        midx = MultiIndex.from_product(
+            [
+                Categorical(["a", "b", "c"]),
+                Categorical(date_range("2012-01-01", periods=3, freq="h")),
+            ]
+        )
+        df = DataFrame({"a": range(len(midx))}, index=midx)
+        df2 = df.iloc[[0, 1, 2, 3, 4, 5, 6, 8]]
+
+        result = df2.reindex(midx)
+        expected = DataFrame({"a": [0, 1, 2, 3, 4, 5, 6, np.nan, 8]}, index=midx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_reindex_with_categoricalindex(self):
+        df = DataFrame(
+            {
+                "A": np.arange(3, dtype="int64"),
+            },
+            index=CategoricalIndex(
+                list("abc"), dtype=CategoricalDtype(list("cabe")), name="B"
+            ),
+        )
+
+        # reindexing
+        # convert to a regular index
+        result = df.reindex(["a", "b", "e"])
+        expected = DataFrame({"A": [0, 1, np.nan], "B": Series(list("abe"))}).set_index(
+            "B"
+        )
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["a", "b"])
+        expected = DataFrame({"A": [0, 1], "B": Series(list("ab"))}).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["e"])
+        expected = DataFrame({"A": [np.nan], "B": Series(["e"])}).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["d"])
+        expected = DataFrame({"A": [np.nan], "B": Series(["d"])}).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        # since we are actually reindexing with a Categorical
+        # then return a Categorical
+        cats = list("cabe")
+
+        result = df.reindex(Categorical(["a", "e"], categories=cats))
+        expected = DataFrame(
+            {"A": [0, np.nan], "B": Series(list("ae")).astype(CategoricalDtype(cats))}
+        ).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(Categorical(["a"], categories=cats))
+        expected = DataFrame(
+            {"A": [0], "B": Series(list("a")).astype(CategoricalDtype(cats))}
+        ).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["a", "b", "e"])
+        expected = DataFrame({"A": [0, 1, np.nan], "B": Series(list("abe"))}).set_index(
+            "B"
+        )
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["a", "b"])
+        expected = DataFrame({"A": [0, 1], "B": Series(list("ab"))}).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(["e"])
+        expected = DataFrame({"A": [np.nan], "B": Series(["e"])}).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        # give back the type of categorical that we received
+        result = df.reindex(Categorical(["a", "e"], categories=cats, ordered=True))
+        expected = DataFrame(
+            {
+                "A": [0, np.nan],
+                "B": Series(list("ae")).astype(CategoricalDtype(cats, ordered=True)),
+            }
+        ).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        result = df.reindex(Categorical(["a", "d"], categories=["a", "d"]))
+        expected = DataFrame(
+            {
+                "A": [0, np.nan],
+                "B": Series(list("ad")).astype(CategoricalDtype(["a", "d"])),
+            }
+        ).set_index("B")
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+        df2 = DataFrame(
+            {
+                "A": np.arange(6, dtype="int64"),
+            },
+            index=CategoricalIndex(
+                list("aabbca"), dtype=CategoricalDtype(list("cabe")), name="B"
+            ),
+        )
+        # passed duplicate indexers are not allowed
+        msg = "cannot reindex on an axis with duplicate labels"
+        with pytest.raises(ValueError, match=msg):
+            df2.reindex(["a", "b"])
+
+        # args NotImplemented ATM
+        msg = r"argument {} is not implemented for CategoricalIndex\.reindex"
+        with pytest.raises(NotImplementedError, match=msg.format("method")):
+            df.reindex(["a"], method="ffill")
+        with pytest.raises(NotImplementedError, match=msg.format("level")):
+            df.reindex(["a"], level=1)
+        with pytest.raises(NotImplementedError, match=msg.format("limit")):
+            df.reindex(["a"], limit=2)
+
+    def test_reindex_signature(self):
+        sig = inspect.signature(DataFrame.reindex)
+        parameters = set(sig.parameters)
+        assert parameters == {
+            "self",
+            "labels",
+            "index",
+            "columns",
+            "axis",
+            "limit",
+            "copy",
+            "level",
+            "method",
+            "fill_value",
+            "tolerance",
+        }
+
+    def test_reindex_multiindex_ffill_added_rows(self):
+        # GH#23693
+        # reindex added rows with nan values even when fill method was specified
+        mi = MultiIndex.from_tuples([("a", "b"), ("d", "e")])
+        df = DataFrame([[0, 7], [3, 4]], index=mi, columns=["x", "y"])
+        mi2 = MultiIndex.from_tuples([("a", "b"), ("d", "e"), ("h", "i")])
+        result = df.reindex(mi2, axis=0, method="ffill")
+        expected = DataFrame([[0, 7], [3, 4], [3, 4]], index=mi2, columns=["x", "y"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"method": "pad", "tolerance": timedelta(seconds=9)},
+            {"method": "backfill", "tolerance": timedelta(seconds=9)},
+            {"method": "nearest"},
+            {"method": None},
+        ],
+    )
+    def test_reindex_empty_frame(self, kwargs):
+        # GH#27315
+        idx = date_range(start="2020", freq="30s", periods=3)
+        df = DataFrame([], index=Index([], name="time"), columns=["a"])
+        result = df.reindex(idx, **kwargs)
+        expected = DataFrame({"a": [np.nan] * 3}, index=idx, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("src_idx", [Index, CategoricalIndex])
+    @pytest.mark.parametrize(
+        "cat_idx",
+        [
+            # No duplicates
+            Index([]),
+            CategoricalIndex([]),
+            Index(["A", "B"]),
+            CategoricalIndex(["A", "B"]),
+            # Duplicates: GH#38906
+            Index(["A", "A"]),
+            CategoricalIndex(["A", "A"]),
+        ],
+    )
+    def test_reindex_empty(self, src_idx, cat_idx):
+        df = DataFrame(columns=src_idx([]), index=["K"], dtype="f8")
+
+        result = df.reindex(columns=cat_idx)
+        expected = DataFrame(index=["K"], columns=cat_idx, dtype="f8")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["m8[ns]", "M8[ns]"])
+    def test_reindex_datetimelike_to_object(self, dtype):
+        # GH#39755 dont cast dt64/td64 to ints
+        mi = MultiIndex.from_product([list("ABCDE"), range(2)])
+
+        dti = date_range("2016-01-01", periods=10)
+        fv = np.timedelta64("NaT", "ns")
+        if dtype == "m8[ns]":
+            dti = dti - dti[0]
+            fv = np.datetime64("NaT", "ns")
+
+        ser = Series(dti, index=mi)
+        ser[::3] = pd.NaT
+
+        df = ser.unstack()
+
+        index = df.index.append(Index([1]))
+        columns = df.columns.append(Index(["foo"]))
+
+        res = df.reindex(index=index, columns=columns, fill_value=fv)
+
+        expected = DataFrame(
+            {
+                0: df[0].tolist() + [fv],
+                1: df[1].tolist() + [fv],
+                "foo": np.array(["NaT"] * 6, dtype=fv.dtype),
+            },
+            index=index,
+        )
+        assert (res.dtypes[[0, 1]] == object).all()
+        assert res.iloc[0, 0] is pd.NaT
+        assert res.iloc[-1, 0] is fv
+        assert res.iloc[-1, 1] is fv
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.parametrize("klass", [Index, CategoricalIndex])
+    @pytest.mark.parametrize("data", ["A", "B"])
+    def test_reindex_not_category(self, klass, data):
+        # GH#28690
+        df = DataFrame(index=CategoricalIndex([], categories=["A"]))
+        idx = klass([data])
+        result = df.reindex(index=idx)
+        expected = DataFrame(index=idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_invalid_method(self):
+        df = DataFrame({"A": [1, np.nan, 2]})
+
+        msg = "Invalid fill method"
+        with pytest.raises(ValueError, match=msg):
+            df.reindex([1, 0, 2], method="asfreq")
diff --git a/pandas/pandas/tests/frame/methods/test_reindex_like.py b/pandas/pandas/tests/frame/methods/test_reindex_like.py
new file mode 100644
index 0000000000000..03968dcbb6314
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_reindex_like.py
@@ -0,0 +1,41 @@
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestDataFrameReindexLike:
+    def test_reindex_like(self, float_frame):
+        other = float_frame.reindex(index=float_frame.index[:10], columns=["C", "B"])
+
+        tm.assert_frame_equal(other, float_frame.reindex_like(other))
+
+    @pytest.mark.parametrize(
+        "method,expected_values",
+        [
+            ("nearest", [0, 1, 1, 2]),
+            ("pad", [np.nan, 0, 1, 1]),
+            ("backfill", [0, 1, 2, 2]),
+        ],
+    )
+    def test_reindex_like_methods(self, method, expected_values):
+        df = DataFrame({"x": list(range(5))})
+
+        with tm.assert_produces_warning(FutureWarning):
+            result = df.reindex_like(df, method=method, tolerance=0)
+        tm.assert_frame_equal(df, result)
+        with tm.assert_produces_warning(FutureWarning):
+            result = df.reindex_like(df, method=method, tolerance=[0, 0, 0, 0])
+        tm.assert_frame_equal(df, result)
+
+    def test_reindex_like_subclass(self):
+        # https://github.com/pandas-dev/pandas/issues/31925
+        class MyDataFrame(DataFrame):
+            pass
+
+        expected = DataFrame()
+        df = MyDataFrame()
+        result = df.reindex_like(expected)
+
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_rename.py b/pandas/pandas/tests/frame/methods/test_rename.py
new file mode 100644
index 0000000000000..6153a168476d4
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_rename.py
@@ -0,0 +1,411 @@
+from collections import ChainMap
+import inspect
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    merge,
+)
+import pandas._testing as tm
+
+
+class TestRename:
+    def test_rename_signature(self):
+        sig = inspect.signature(DataFrame.rename)
+        parameters = set(sig.parameters)
+        assert parameters == {
+            "self",
+            "mapper",
+            "index",
+            "columns",
+            "axis",
+            "inplace",
+            "copy",
+            "level",
+            "errors",
+        }
+
+    def test_rename_mi(self, frame_or_series):
+        obj = frame_or_series(
+            [11, 21, 31],
+            index=MultiIndex.from_tuples([("A", x) for x in ["a", "B", "c"]]),
+        )
+        obj.rename(str.lower)
+
+    def test_rename(self, float_frame):
+        mapping = {"A": "a", "B": "b", "C": "c", "D": "d"}
+
+        renamed = float_frame.rename(columns=mapping)
+        renamed2 = float_frame.rename(columns=str.lower)
+
+        tm.assert_frame_equal(renamed, renamed2)
+        tm.assert_frame_equal(
+            renamed2.rename(columns=str.upper), float_frame, check_names=False
+        )
+
+        # index
+        data = {"A": {"foo": 0, "bar": 1}}
+
+        df = DataFrame(data)
+        renamed = df.rename(index={"foo": "bar", "bar": "foo"})
+        tm.assert_index_equal(renamed.index, Index(["bar", "foo"]))
+
+        renamed = df.rename(index=str.upper)
+        tm.assert_index_equal(renamed.index, Index(["FOO", "BAR"]))
+
+        # have to pass something
+        with pytest.raises(TypeError, match="must pass an index to rename"):
+            float_frame.rename()
+
+        # partial columns
+        renamed = float_frame.rename(columns={"C": "foo", "D": "bar"})
+        tm.assert_index_equal(renamed.columns, Index(["A", "B", "foo", "bar"]))
+
+        # other axis
+        renamed = float_frame.T.rename(index={"C": "foo", "D": "bar"})
+        tm.assert_index_equal(renamed.index, Index(["A", "B", "foo", "bar"]))
+
+        # index with name
+        index = Index(["foo", "bar"], name="name")
+        renamer = DataFrame(data, index=index)
+        renamed = renamer.rename(index={"foo": "bar", "bar": "foo"})
+        tm.assert_index_equal(renamed.index, Index(["bar", "foo"], name="name"))
+        assert renamed.index.name == renamer.index.name
+
+    @pytest.mark.parametrize(
+        "args,kwargs",
+        [
+            ((ChainMap({"A": "a"}, {"B": "b"}),), {"axis": "columns"}),
+            ((), {"columns": ChainMap({"A": "a"}, {"B": "b"})}),
+        ],
+    )
+    def test_rename_chainmap(self, args, kwargs):
+        # see gh-23859
+        colAData = range(1, 11)
+        colBdata = np.random.default_rng(2).standard_normal(10)
+
+        df = DataFrame({"A": colAData, "B": colBdata})
+        result = df.rename(*args, **kwargs)
+
+        expected = DataFrame({"a": colAData, "b": colBdata})
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_multiindex(self):
+        tuples_index = [("foo1", "bar1"), ("foo2", "bar2")]
+        tuples_columns = [("fizz1", "buzz1"), ("fizz2", "buzz2")]
+        index = MultiIndex.from_tuples(tuples_index, names=["foo", "bar"])
+        columns = MultiIndex.from_tuples(tuples_columns, names=["fizz", "buzz"])
+        df = DataFrame([(0, 0), (1, 1)], index=index, columns=columns)
+
+        #
+        # without specifying level -> across all levels
+
+        renamed = df.rename(
+            index={"foo1": "foo3", "bar2": "bar3"},
+            columns={"fizz1": "fizz3", "buzz2": "buzz3"},
+        )
+        new_index = MultiIndex.from_tuples(
+            [("foo3", "bar1"), ("foo2", "bar3")], names=["foo", "bar"]
+        )
+        new_columns = MultiIndex.from_tuples(
+            [("fizz3", "buzz1"), ("fizz2", "buzz3")], names=["fizz", "buzz"]
+        )
+        tm.assert_index_equal(renamed.index, new_index)
+        tm.assert_index_equal(renamed.columns, new_columns)
+        assert renamed.index.names == df.index.names
+        assert renamed.columns.names == df.columns.names
+
+        #
+        # with specifying a level (GH13766)
+
+        # dict
+        new_columns = MultiIndex.from_tuples(
+            [("fizz3", "buzz1"), ("fizz2", "buzz2")], names=["fizz", "buzz"]
+        )
+        renamed = df.rename(columns={"fizz1": "fizz3", "buzz2": "buzz3"}, level=0)
+        tm.assert_index_equal(renamed.columns, new_columns)
+        renamed = df.rename(columns={"fizz1": "fizz3", "buzz2": "buzz3"}, level="fizz")
+        tm.assert_index_equal(renamed.columns, new_columns)
+
+        new_columns = MultiIndex.from_tuples(
+            [("fizz1", "buzz1"), ("fizz2", "buzz3")], names=["fizz", "buzz"]
+        )
+        renamed = df.rename(columns={"fizz1": "fizz3", "buzz2": "buzz3"}, level=1)
+        tm.assert_index_equal(renamed.columns, new_columns)
+        renamed = df.rename(columns={"fizz1": "fizz3", "buzz2": "buzz3"}, level="buzz")
+        tm.assert_index_equal(renamed.columns, new_columns)
+
+        # function
+        func = str.upper
+        new_columns = MultiIndex.from_tuples(
+            [("FIZZ1", "buzz1"), ("FIZZ2", "buzz2")], names=["fizz", "buzz"]
+        )
+        renamed = df.rename(columns=func, level=0)
+        tm.assert_index_equal(renamed.columns, new_columns)
+        renamed = df.rename(columns=func, level="fizz")
+        tm.assert_index_equal(renamed.columns, new_columns)
+
+        new_columns = MultiIndex.from_tuples(
+            [("fizz1", "BUZZ1"), ("fizz2", "BUZZ2")], names=["fizz", "buzz"]
+        )
+        renamed = df.rename(columns=func, level=1)
+        tm.assert_index_equal(renamed.columns, new_columns)
+        renamed = df.rename(columns=func, level="buzz")
+        tm.assert_index_equal(renamed.columns, new_columns)
+
+        # index
+        new_index = MultiIndex.from_tuples(
+            [("foo3", "bar1"), ("foo2", "bar2")], names=["foo", "bar"]
+        )
+        renamed = df.rename(index={"foo1": "foo3", "bar2": "bar3"}, level=0)
+        tm.assert_index_equal(renamed.index, new_index)
+
+    def test_rename_nocopy(self, float_frame):
+        renamed = float_frame.rename(columns={"C": "foo"})
+
+        assert np.shares_memory(renamed["foo"]._values, float_frame["C"]._values)
+
+        renamed.loc[:, "foo"] = 1.0
+        assert not (float_frame["C"] == 1.0).all()
+
+    def test_rename_inplace(self, float_frame):
+        float_frame.rename(columns={"C": "foo"})
+        assert "C" in float_frame
+        assert "foo" not in float_frame
+
+        c_values = float_frame["C"]
+        float_frame = float_frame.copy()
+        return_value = float_frame.rename(columns={"C": "foo"}, inplace=True)
+        assert return_value is None
+
+        assert "C" not in float_frame
+        assert "foo" in float_frame
+        # GH 44153
+        # Used to be id(float_frame["foo"]) != c_id, but flaky in the CI
+        assert float_frame["foo"] is not c_values
+
+    def test_rename_bug(self):
+        # GH 5344
+        # rename set ref_locs, and set_index was not resetting
+        df = DataFrame({0: ["foo", "bar"], 1: ["bah", "bas"], 2: [1, 2]})
+        df = df.rename(columns={0: "a"})
+        df = df.rename(columns={1: "b"})
+        df = df.set_index(["a", "b"])
+        df.columns = ["2001-01-01"]
+        expected = DataFrame(
+            [[1], [2]],
+            index=MultiIndex.from_tuples(
+                [("foo", "bah"), ("bar", "bas")], names=["a", "b"]
+            ),
+            columns=["2001-01-01"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_rename_bug2(self):
+        # GH 19497
+        # rename was changing Index to MultiIndex if Index contained tuples
+
+        df = DataFrame(data=np.arange(3), index=[(0, 0), (1, 1), (2, 2)], columns=["a"])
+        df = df.rename({(1, 1): (5, 4)}, axis="index")
+        expected = DataFrame(
+            data=np.arange(3), index=[(0, 0), (5, 4), (2, 2)], columns=["a"]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_rename_errors_raises(self):
+        df = DataFrame(columns=["A", "B", "C", "D"])
+        with pytest.raises(KeyError, match="'E'] not found in axis"):
+            df.rename(columns={"A": "a", "E": "e"}, errors="raise")
+
+    @pytest.mark.parametrize(
+        "mapper, errors, expected_columns",
+        [
+            ({"A": "a", "E": "e"}, "ignore", ["a", "B", "C", "D"]),
+            ({"A": "a"}, "raise", ["a", "B", "C", "D"]),
+            (str.lower, "raise", ["a", "b", "c", "d"]),
+        ],
+    )
+    def test_rename_errors(self, mapper, errors, expected_columns):
+        # GH 13473
+        # rename now works with errors parameter
+        df = DataFrame(columns=["A", "B", "C", "D"])
+        result = df.rename(columns=mapper, errors=errors)
+        expected = DataFrame(columns=expected_columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_objects(self, float_string_frame):
+        renamed = float_string_frame.rename(columns=str.upper)
+
+        assert "FOO" in renamed
+        assert "foo" not in renamed
+
+    def test_rename_axis_style(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        df = DataFrame({"A": [1, 2], "B": [1, 2]}, index=["X", "Y"])
+        expected = DataFrame({"a": [1, 2], "b": [1, 2]}, index=["X", "Y"])
+
+        result = df.rename(str.lower, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename(str.lower, axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename({"A": "a", "B": "b"}, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename({"A": "a", "B": "b"}, axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+        # Index
+        expected = DataFrame({"A": [1, 2], "B": [1, 2]}, index=["x", "y"])
+        result = df.rename(str.lower, axis=0)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename(str.lower, axis="index")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename({"X": "x", "Y": "y"}, axis=0)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename({"X": "x", "Y": "y"}, axis="index")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rename(mapper=str.lower, axis="index")
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_mapper_multi(self):
+        df = DataFrame({"A": ["a", "b"], "B": ["c", "d"], "C": [1, 2]}).set_index(
+            ["A", "B"]
+        )
+        result = df.rename(str.upper)
+        expected = df.rename(index=str.upper)
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_positional_named(self):
+        # https://github.com/pandas-dev/pandas/issues/12392
+        df = DataFrame({"a": [1, 2], "b": [1, 2]}, index=["X", "Y"])
+        result = df.rename(index=str.lower, columns=str.upper)
+        expected = DataFrame({"A": [1, 2], "B": [1, 2]}, index=["x", "y"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_axis_style_raises(self):
+        # see gh-12392
+        df = DataFrame({"A": [1, 2], "B": [1, 2]}, index=["0", "1"])
+
+        # Named target and axis
+        over_spec_msg = "Cannot specify both 'axis' and any of 'index' or 'columns'"
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(index=str.lower, axis=1)
+
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(index=str.lower, axis="columns")
+
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(columns=str.lower, axis="columns")
+
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(index=str.lower, axis=0)
+
+        # Multiple targets and axis
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(str.lower, index=str.lower, axis="columns")
+
+        # Too many targets
+        over_spec_msg = "Cannot specify both 'mapper' and any of 'index' or 'columns'"
+        with pytest.raises(TypeError, match=over_spec_msg):
+            df.rename(str.lower, index=str.lower, columns=str.lower)
+
+        # Duplicates
+        with pytest.raises(TypeError, match="multiple values"):
+            df.rename(id, mapper=id)
+
+    def test_rename_positional_raises(self):
+        # GH 29136
+        df = DataFrame(columns=["A", "B"])
+        msg = r"rename\(\) takes from 1 to 2 positional arguments"
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename(None, str.lower)
+
+    def test_rename_no_mappings_raises(self):
+        # GH 29136
+        df = DataFrame([[1]])
+        msg = "must pass an index to rename"
+        with pytest.raises(TypeError, match=msg):
+            df.rename()
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename(None, index=None)
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename(None, columns=None)
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename(None, columns=None, index=None)
+
+    def test_rename_mapper_and_positional_arguments_raises(self):
+        # GH 29136
+        df = DataFrame([[1]])
+        msg = "Cannot specify both 'mapper' and any of 'index' or 'columns'"
+        with pytest.raises(TypeError, match=msg):
+            df.rename({}, index={})
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename({}, columns={})
+
+        with pytest.raises(TypeError, match=msg):
+            df.rename({}, columns={}, index={})
+
+    def test_rename_with_duplicate_columns(self):
+        # GH#4403
+        df4 = DataFrame(
+            {"RT": [0.0454], "TClose": [22.02], "TExg": [0.0422]},
+            index=MultiIndex.from_tuples(
+                [(600809, 20130331)], names=["STK_ID", "RPT_Date"]
+            ),
+        )
+
+        df5 = DataFrame(
+            {
+                "RPT_Date": [20120930, 20121231, 20130331],
+                "STK_ID": [600809] * 3,
+                "STK_Name": ["饡驦", "饡驦", "饡驦"],
+                "TClose": [38.05, 41.66, 30.01],
+            },
+            index=MultiIndex.from_tuples(
+                [(600809, 20120930), (600809, 20121231), (600809, 20130331)],
+                names=["STK_ID", "RPT_Date"],
+            ),
+        )
+        # TODO: can we construct this without merge?
+        k = merge(df4, df5, how="inner", left_index=True, right_index=True)
+        result = k.rename(columns={"TClose_x": "TClose", "TClose_y": "QT_Close"})
+
+        expected = DataFrame(
+            [[0.0454, 22.02, 0.0422, 20130331, 600809, "饡驦", 30.01]],
+            columns=[
+                "RT",
+                "TClose",
+                "TExg",
+                "RPT_Date",
+                "STK_ID",
+                "STK_Name",
+                "QT_Close",
+            ],
+        ).set_index(["STK_ID", "RPT_Date"], drop=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_boolean_index(self):
+        df = DataFrame(np.arange(15).reshape(3, 5), columns=[False, True, 2, 3, 4])
+        mapper = {0: "foo", 1: "bar", 2: "bah"}
+        res = df.rename(index=mapper)
+        exp = DataFrame(
+            np.arange(15).reshape(3, 5),
+            columns=[False, True, 2, 3, 4],
+            index=["foo", "bar", "bah"],
+        )
+        tm.assert_frame_equal(res, exp)
diff --git a/pandas/pandas/tests/frame/methods/test_rename_axis.py b/pandas/pandas/tests/frame/methods/test_rename_axis.py
new file mode 100644
index 0000000000000..dd4a77c6509b8
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_rename_axis.py
@@ -0,0 +1,111 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameRenameAxis:
+    def test_rename_axis_inplace(self, float_frame):
+        # GH#15704
+        expected = float_frame.rename_axis("foo")
+        result = float_frame.copy()
+        return_value = no_return = result.rename_axis("foo", inplace=True)
+        assert return_value is None
+
+        assert no_return is None
+        tm.assert_frame_equal(result, expected)
+
+        expected = float_frame.rename_axis("bar", axis=1)
+        result = float_frame.copy()
+        return_value = no_return = result.rename_axis("bar", axis=1, inplace=True)
+        assert return_value is None
+
+        assert no_return is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_axis_raises(self):
+        # GH#17833
+        df = DataFrame({"A": [1, 2], "B": [1, 2]})
+        with pytest.raises(ValueError, match="Use `.rename`"):
+            df.rename_axis(id, axis=0)
+
+        with pytest.raises(ValueError, match="Use `.rename`"):
+            df.rename_axis({0: 10, 1: 20}, axis=0)
+
+        with pytest.raises(ValueError, match="Use `.rename`"):
+            df.rename_axis(id, axis=1)
+
+        with pytest.raises(ValueError, match="Use `.rename`"):
+            df["A"].rename_axis(id)
+
+    def test_rename_axis_mapper(self):
+        # GH#19978
+        mi = MultiIndex.from_product([["a", "b", "c"], [1, 2]], names=["ll", "nn"])
+        df = DataFrame(
+            {"x": list(range(len(mi))), "y": [i * 10 for i in range(len(mi))]}, index=mi
+        )
+
+        # Test for rename of the Index object of columns
+        result = df.rename_axis("cols", axis=1)
+        tm.assert_index_equal(result.columns, Index(["x", "y"], name="cols"))
+
+        # Test for rename of the Index object of columns using dict
+        result = result.rename_axis(columns={"cols": "new"}, axis=1)
+        tm.assert_index_equal(result.columns, Index(["x", "y"], name="new"))
+
+        # Test for renaming index using dict
+        result = df.rename_axis(index={"ll": "foo"})
+        assert result.index.names == ["foo", "nn"]
+
+        # Test for renaming index using a function
+        result = df.rename_axis(index=str.upper, axis=0)
+        assert result.index.names == ["LL", "NN"]
+
+        # Test for renaming index providing complete list
+        result = df.rename_axis(index=["foo", "goo"])
+        assert result.index.names == ["foo", "goo"]
+
+        # Test for changing index and columns at same time
+        sdf = df.reset_index().set_index("nn").drop(columns=["ll", "y"])
+        result = sdf.rename_axis(index="foo", columns="meh")
+        assert result.index.name == "foo"
+        assert result.columns.name == "meh"
+
+        # Test different error cases
+        with pytest.raises(TypeError, match="Must pass"):
+            df.rename_axis(index="wrong")
+
+        with pytest.raises(ValueError, match="Length of names"):
+            df.rename_axis(index=["wrong"])
+
+        with pytest.raises(TypeError, match="bogus"):
+            df.rename_axis(bogus=None)
+
+    @pytest.mark.parametrize(
+        "kwargs, rename_index, rename_columns",
+        [
+            ({"mapper": None, "axis": 0}, True, False),
+            ({"mapper": None, "axis": 1}, False, True),
+            ({"index": None}, True, False),
+            ({"columns": None}, False, True),
+            ({"index": None, "columns": None}, True, True),
+            ({}, False, False),
+        ],
+    )
+    def test_rename_axis_none(self, kwargs, rename_index, rename_columns):
+        # GH 25034
+        index = Index(list("abc"), name="foo")
+        columns = Index(["col1", "col2"], name="bar")
+        data = np.arange(6).reshape(3, 2)
+        df = DataFrame(data, index, columns)
+
+        result = df.rename_axis(**kwargs)
+        expected_index = index.rename(None) if rename_index else index
+        expected_columns = columns.rename(None) if rename_columns else columns
+        expected = DataFrame(data, expected_index, expected_columns)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_reorder_levels.py b/pandas/pandas/tests/frame/methods/test_reorder_levels.py
new file mode 100644
index 0000000000000..5d6b65daae4d5
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_reorder_levels.py
@@ -0,0 +1,74 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+class TestReorderLevels:
+    def test_reorder_levels(self, frame_or_series):
+        index = MultiIndex(
+            levels=[["bar"], ["one", "two", "three"], [0, 1]],
+            codes=[[0, 0, 0, 0, 0, 0], [0, 1, 2, 0, 1, 2], [0, 1, 0, 1, 0, 1]],
+            names=["L0", "L1", "L2"],
+        )
+        df = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=index)
+        obj = tm.get_obj(df, frame_or_series)
+
+        # no change, position
+        result = obj.reorder_levels([0, 1, 2])
+        tm.assert_equal(obj, result)
+
+        # no change, labels
+        result = obj.reorder_levels(["L0", "L1", "L2"])
+        tm.assert_equal(obj, result)
+
+        # rotate, position
+        result = obj.reorder_levels([1, 2, 0])
+        e_idx = MultiIndex(
+            levels=[["one", "two", "three"], [0, 1], ["bar"]],
+            codes=[[0, 1, 2, 0, 1, 2], [0, 1, 0, 1, 0, 1], [0, 0, 0, 0, 0, 0]],
+            names=["L1", "L2", "L0"],
+        )
+        expected = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=e_idx)
+        expected = tm.get_obj(expected, frame_or_series)
+        tm.assert_equal(result, expected)
+
+        result = obj.reorder_levels([0, 0, 0])
+        e_idx = MultiIndex(
+            levels=[["bar"], ["bar"], ["bar"]],
+            codes=[[0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0]],
+            names=["L0", "L0", "L0"],
+        )
+        expected = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=e_idx)
+        expected = tm.get_obj(expected, frame_or_series)
+        tm.assert_equal(result, expected)
+
+        result = obj.reorder_levels(["L0", "L0", "L0"])
+        tm.assert_equal(result, expected)
+
+    def test_reorder_levels_swaplevel_equivalence(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        result = ymd.reorder_levels(["month", "day", "year"])
+        expected = ymd.swaplevel(0, 1).swaplevel(1, 2)
+        tm.assert_frame_equal(result, expected)
+
+        result = ymd["A"].reorder_levels(["month", "day", "year"])
+        expected = ymd["A"].swaplevel(0, 1).swaplevel(1, 2)
+        tm.assert_series_equal(result, expected)
+
+        result = ymd.T.reorder_levels(["month", "day", "year"], axis=1)
+        expected = ymd.T.swaplevel(0, 1, axis=1).swaplevel(1, 2, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(TypeError, match="hierarchical axis"):
+            ymd.reorder_levels([1, 2], axis=1)
+
+        with pytest.raises(IndexError, match="Too many levels"):
+            ymd.index.reorder_levels([1, 2, 3])
diff --git a/pandas/pandas/tests/frame/methods/test_replace.py b/pandas/pandas/tests/frame/methods/test_replace.py
new file mode 100644
index 0000000000000..9e302dc5f94ee
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_replace.py
@@ -0,0 +1,1530 @@
+from __future__ import annotations
+
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def mix_ab() -> dict[str, list[int | str]]:
+    return {"a": list(range(4)), "b": list("ab..")}
+
+
+@pytest.fixture
+def mix_abc() -> dict[str, list[float | str]]:
+    return {"a": list(range(4)), "b": list("ab.."), "c": ["a", "b", np.nan, "d"]}
+
+
+class TestDataFrameReplace:
+    def test_replace_inplace(self, datetime_frame, float_string_frame):
+        datetime_frame.loc[datetime_frame.index[:5], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[-5:], "A"] = np.nan
+
+        tsframe = datetime_frame.copy()
+        return_value = tsframe.replace(np.nan, 0, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(tsframe, datetime_frame.fillna(0))
+
+        # mixed type
+        mf = float_string_frame
+        mf.iloc[5:20, mf.columns.get_loc("foo")] = np.nan
+        mf.iloc[-10:, mf.columns.get_loc("A")] = np.nan
+
+        result = float_string_frame.replace(np.nan, 0)
+        expected = float_string_frame.copy()
+        expected["foo"] = expected["foo"].astype(object)
+        expected = expected.fillna(value=0)
+        tm.assert_frame_equal(result, expected)
+
+        tsframe = datetime_frame.copy()
+        return_value = tsframe.replace([np.nan], [0], inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(tsframe, datetime_frame.fillna(0))
+
+    @pytest.mark.parametrize(
+        "to_replace,values,expected",
+        [
+            # lists of regexes and values
+            # list of [re1, re2, ..., reN] -> [v1, v2, ..., vN]
+            (
+                [r"\s*\.\s*", r"e|f|g"],
+                [np.nan, "crap"],
+                {
+                    "a": ["a", "b", np.nan, np.nan],
+                    "b": ["crap"] * 3 + ["h"],
+                    "c": ["h", "crap", "l", "o"],
+                },
+            ),
+            # list of [re1, re2, ..., reN] -> [re1, re2, .., reN]
+            (
+                [r"\s*(\.)\s*", r"(e|f|g)"],
+                [r"\1\1", r"\1_crap"],
+                {
+                    "a": ["a", "b", "..", ".."],
+                    "b": ["e_crap", "f_crap", "g_crap", "h"],
+                    "c": ["h", "e_crap", "l", "o"],
+                },
+            ),
+            # list of [re1, re2, ..., reN] -> [(re1 or v1), (re2 or v2), ..., (reN
+            # or vN)]
+            (
+                [r"\s*(\.)\s*", r"e"],
+                [r"\1\1", r"crap"],
+                {
+                    "a": ["a", "b", "..", ".."],
+                    "b": ["crap", "f", "g", "h"],
+                    "c": ["h", "crap", "l", "o"],
+                },
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize("use_value_regex_args", [True, False])
+    def test_regex_replace_list_obj(
+        self, to_replace, values, expected, inplace, use_value_regex_args
+    ):
+        df = DataFrame({"a": list("ab.."), "b": list("efgh"), "c": list("helo")})
+
+        if use_value_regex_args:
+            result = df.replace(value=values, regex=to_replace, inplace=inplace)
+        else:
+            result = df.replace(to_replace, values, regex=True, inplace=inplace)
+
+        if inplace:
+            assert result is None
+            result = df
+
+        expected = DataFrame(expected)
+        tm.assert_frame_equal(result, expected)
+
+    def test_regex_replace_list_mixed(self, mix_ab):
+        # mixed frame to make sure this doesn't break things
+        dfmix = DataFrame(mix_ab)
+
+        # lists of regexes and values
+        # list of [re1, re2, ..., reN] -> [v1, v2, ..., vN]
+        to_replace_res = [r"\s*\.\s*", r"a"]
+        values = [np.nan, "crap"]
+        mix2 = {"a": list(range(4)), "b": list("ab.."), "c": list("halo")}
+        dfmix2 = DataFrame(mix2)
+        res = dfmix2.replace(to_replace_res, values, regex=True)
+        expec = DataFrame(
+            {
+                "a": mix2["a"],
+                "b": ["crap", "b", np.nan, np.nan],
+                "c": ["h", "crap", "l", "o"],
+            }
+        )
+        tm.assert_frame_equal(res, expec)
+
+        # list of [re1, re2, ..., reN] -> [re1, re2, .., reN]
+        to_replace_res = [r"\s*(\.)\s*", r"(a|b)"]
+        values = [r"\1\1", r"\1_crap"]
+        res = dfmix.replace(to_replace_res, values, regex=True)
+        expec = DataFrame({"a": mix_ab["a"], "b": ["a_crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+        # list of [re1, re2, ..., reN] -> [(re1 or v1), (re2 or v2), ..., (reN
+        # or vN)]
+        to_replace_res = [r"\s*(\.)\s*", r"a", r"(b)"]
+        values = [r"\1\1", r"crap", r"\1_crap"]
+        res = dfmix.replace(to_replace_res, values, regex=True)
+        expec = DataFrame({"a": mix_ab["a"], "b": ["crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+        to_replace_res = [r"\s*(\.)\s*", r"a", r"(b)"]
+        values = [r"\1\1", r"crap", r"\1_crap"]
+        res = dfmix.replace(regex=to_replace_res, value=values)
+        expec = DataFrame({"a": mix_ab["a"], "b": ["crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+    def test_regex_replace_list_mixed_inplace(self, mix_ab):
+        dfmix = DataFrame(mix_ab)
+        # the same inplace
+        # lists of regexes and values
+        # list of [re1, re2, ..., reN] -> [v1, v2, ..., vN]
+        to_replace_res = [r"\s*\.\s*", r"a"]
+        values = [np.nan, "crap"]
+        res = dfmix.copy()
+        return_value = res.replace(to_replace_res, values, inplace=True, regex=True)
+        assert return_value is None
+        expec = DataFrame({"a": mix_ab["a"], "b": ["crap", "b", np.nan, np.nan]})
+        tm.assert_frame_equal(res, expec)
+
+        # list of [re1, re2, ..., reN] -> [re1, re2, .., reN]
+        to_replace_res = [r"\s*(\.)\s*", r"(a|b)"]
+        values = [r"\1\1", r"\1_crap"]
+        res = dfmix.copy()
+        return_value = res.replace(to_replace_res, values, inplace=True, regex=True)
+        assert return_value is None
+        expec = DataFrame({"a": mix_ab["a"], "b": ["a_crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+        # list of [re1, re2, ..., reN] -> [(re1 or v1), (re2 or v2), ..., (reN
+        # or vN)]
+        to_replace_res = [r"\s*(\.)\s*", r"a", r"(b)"]
+        values = [r"\1\1", r"crap", r"\1_crap"]
+        res = dfmix.copy()
+        return_value = res.replace(to_replace_res, values, inplace=True, regex=True)
+        assert return_value is None
+        expec = DataFrame({"a": mix_ab["a"], "b": ["crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+        to_replace_res = [r"\s*(\.)\s*", r"a", r"(b)"]
+        values = [r"\1\1", r"crap", r"\1_crap"]
+        res = dfmix.copy()
+        return_value = res.replace(regex=to_replace_res, value=values, inplace=True)
+        assert return_value is None
+        expec = DataFrame({"a": mix_ab["a"], "b": ["crap", "b_crap", "..", ".."]})
+        tm.assert_frame_equal(res, expec)
+
+    def test_regex_replace_dict_mixed(self, mix_abc):
+        dfmix = DataFrame(mix_abc)
+
+        # dicts
+        # single dict {re1: v1}, search the whole frame
+        # need test for this...
+
+        # list of dicts {re1: v1, re2: v2, ..., re3: v3}, search the whole
+        # frame
+        res = dfmix.replace({"b": r"\s*\.\s*"}, {"b": np.nan}, regex=True)
+        res2 = dfmix.copy()
+        return_value = res2.replace(
+            {"b": r"\s*\.\s*"}, {"b": np.nan}, inplace=True, regex=True
+        )
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", "b", np.nan, np.nan], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+
+        # list of dicts {re1: re11, re2: re12, ..., reN: re1N}, search the
+        # whole frame
+        res = dfmix.replace({"b": r"\s*(\.)\s*"}, {"b": r"\1ty"}, regex=True)
+        res2 = dfmix.copy()
+        return_value = res2.replace(
+            {"b": r"\s*(\.)\s*"}, {"b": r"\1ty"}, inplace=True, regex=True
+        )
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", "b", ".ty", ".ty"], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+
+        res = dfmix.replace(regex={"b": r"\s*(\.)\s*"}, value={"b": r"\1ty"})
+        res2 = dfmix.copy()
+        return_value = res2.replace(
+            regex={"b": r"\s*(\.)\s*"}, value={"b": r"\1ty"}, inplace=True
+        )
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", "b", ".ty", ".ty"], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+
+        # scalar -> dict
+        # to_replace regex, {value: value}
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": [np.nan, "b", ".", "."], "c": mix_abc["c"]}
+        )
+        res = dfmix.replace("a", {"b": np.nan}, regex=True)
+        res2 = dfmix.copy()
+        return_value = res2.replace("a", {"b": np.nan}, regex=True, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+
+        res = dfmix.replace("a", {"b": np.nan}, regex=True)
+        res2 = dfmix.copy()
+        return_value = res2.replace(regex="a", value={"b": np.nan}, inplace=True)
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": [np.nan, "b", ".", "."], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+
+    def test_regex_replace_dict_nested(self, mix_abc):
+        # nested dicts will not work until this is implemented for Series
+        dfmix = DataFrame(mix_abc)
+        res = dfmix.replace({"b": {r"\s*\.\s*": np.nan}}, regex=True)
+        res2 = dfmix.copy()
+        res4 = dfmix.copy()
+        return_value = res2.replace(
+            {"b": {r"\s*\.\s*": np.nan}}, inplace=True, regex=True
+        )
+        assert return_value is None
+        res3 = dfmix.replace(regex={"b": {r"\s*\.\s*": np.nan}})
+        return_value = res4.replace(regex={"b": {r"\s*\.\s*": np.nan}}, inplace=True)
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", "b", np.nan, np.nan], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+        tm.assert_frame_equal(res3, expec)
+        tm.assert_frame_equal(res4, expec)
+
+    def test_regex_replace_dict_nested_non_first_character(
+        self, any_string_dtype, using_infer_string
+    ):
+        # GH 25259
+        dtype = any_string_dtype
+        df = DataFrame({"first": ["abc", "bca", "cab"]}, dtype=dtype)
+        result = df.replace({"a": "."}, regex=True)
+        expected = DataFrame({"first": [".bc", "bc.", "c.b"]}, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+    def test_regex_replace_dict_nested_gh4115(self):
+        df = DataFrame(
+            {"Type": Series(["Q", "T", "Q", "Q", "T"], dtype=object), "tmp": 2}
+        )
+        expected = DataFrame({"Type": Series([0, 1, 0, 0, 1], dtype=object), "tmp": 2})
+        result = df.replace({"Type": {"Q": 0, "T": 1}})
+        tm.assert_frame_equal(result, expected)
+
+    def test_regex_replace_list_to_scalar(self, mix_abc):
+        df = DataFrame(mix_abc)
+        expec = DataFrame(
+            {
+                "a": mix_abc["a"],
+                "b": Series([np.nan] * 4, dtype="str"),
+                "c": [np.nan, np.nan, np.nan, "d"],
+            }
+        )
+
+        res = df.replace([r"\s*\.\s*", "a|b"], np.nan, regex=True)
+        res2 = df.copy()
+        res3 = df.copy()
+        return_value = res2.replace(
+            [r"\s*\.\s*", "a|b"], np.nan, regex=True, inplace=True
+        )
+        assert return_value is None
+        return_value = res3.replace(
+            regex=[r"\s*\.\s*", "a|b"], value=np.nan, inplace=True
+        )
+        assert return_value is None
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+        tm.assert_frame_equal(res3, expec)
+
+    def test_regex_replace_str_to_numeric(self, mix_abc):
+        # what happens when you try to replace a numeric value with a regex?
+        df = DataFrame(mix_abc)
+        res = df.replace(r"\s*\.\s*", 0, regex=True)
+        res2 = df.copy()
+        return_value = res2.replace(r"\s*\.\s*", 0, inplace=True, regex=True)
+        assert return_value is None
+        res3 = df.copy()
+        return_value = res3.replace(regex=r"\s*\.\s*", value=0, inplace=True)
+        assert return_value is None
+        expec = DataFrame({"a": mix_abc["a"], "b": ["a", "b", 0, 0], "c": mix_abc["c"]})
+        expec["c"] = expec["c"].astype(object)
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+        tm.assert_frame_equal(res3, expec)
+
+    def test_regex_replace_regex_list_to_numeric(self, mix_abc):
+        df = DataFrame(mix_abc)
+        res = df.replace([r"\s*\.\s*", "b"], 0, regex=True)
+        res2 = df.copy()
+        return_value = res2.replace([r"\s*\.\s*", "b"], 0, regex=True, inplace=True)
+        assert return_value is None
+        res3 = df.copy()
+        return_value = res3.replace(regex=[r"\s*\.\s*", "b"], value=0, inplace=True)
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", 0, 0, 0], "c": ["a", 0, np.nan, "d"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+        tm.assert_frame_equal(res3, expec)
+
+    def test_regex_replace_series_of_regexes(self, mix_abc):
+        df = DataFrame(mix_abc)
+        s1 = Series({"b": r"\s*\.\s*"})
+        s2 = Series({"b": np.nan})
+        res = df.replace(s1, s2, regex=True)
+        res2 = df.copy()
+        return_value = res2.replace(s1, s2, inplace=True, regex=True)
+        assert return_value is None
+        res3 = df.copy()
+        return_value = res3.replace(regex=s1, value=s2, inplace=True)
+        assert return_value is None
+        expec = DataFrame(
+            {"a": mix_abc["a"], "b": ["a", "b", np.nan, np.nan], "c": mix_abc["c"]}
+        )
+        tm.assert_frame_equal(res, expec)
+        tm.assert_frame_equal(res2, expec)
+        tm.assert_frame_equal(res3, expec)
+
+    def test_regex_replace_numeric_to_object_conversion(self, mix_abc):
+        df = DataFrame(mix_abc)
+        expec = DataFrame({"a": ["a", 1, 2, 3], "b": mix_abc["b"], "c": mix_abc["c"]})
+        res = df.replace(0, "a")
+        tm.assert_frame_equal(res, expec)
+        assert res.a.dtype == np.object_
+
+    @pytest.mark.parametrize(
+        "to_replace", [{"": np.nan, ",": ""}, {",": "", "": np.nan}]
+    )
+    def test_joint_simple_replace_and_regex_replace(self, to_replace):
+        # GH-39338
+        df = DataFrame(
+            {
+                "col1": ["1,000", "a", "3"],
+                "col2": ["a", "", "b"],
+                "col3": ["a", "b", "c"],
+            }
+        )
+        result = df.replace(regex=to_replace)
+        expected = DataFrame(
+            {
+                "col1": ["1000", "a", "3"],
+                "col2": ["a", np.nan, "b"],
+                "col3": ["a", "b", "c"],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("metachar", ["[]", "()", r"\d", r"\w", r"\s"])
+    def test_replace_regex_metachar(self, metachar):
+        df = DataFrame({"a": [metachar, "else"]})
+        result = df.replace({"a": {metachar: "paren"}})
+        expected = DataFrame({"a": ["paren", "else"]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data,to_replace,expected",
+        [
+            (["xax", "xbx"], {"a": "c", "b": "d"}, ["xcx", "xdx"]),
+            (["d", "", ""], {r"^\s*$": pd.NA}, ["d", pd.NA, pd.NA]),
+        ],
+    )
+    def test_regex_replace_string_types(
+        self,
+        data,
+        to_replace,
+        expected,
+        frame_or_series,
+        any_string_dtype,
+        using_infer_string,
+        request,
+    ):
+        # GH-41333, GH-35977
+        dtype = any_string_dtype
+        obj = frame_or_series(data, dtype=dtype)
+        result = obj.replace(to_replace, regex=True)
+        expected = frame_or_series(expected, dtype=dtype)
+
+        tm.assert_equal(result, expected)
+
+    def test_replace(self, datetime_frame):
+        datetime_frame.loc[datetime_frame.index[:5], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[-5:], "A"] = np.nan
+
+        zero_filled = datetime_frame.replace(np.nan, -1e8)
+        tm.assert_frame_equal(zero_filled, datetime_frame.fillna(-1e8))
+        tm.assert_frame_equal(zero_filled.replace(-1e8, np.nan), datetime_frame)
+
+        datetime_frame.loc[datetime_frame.index[:5], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[-5:], "A"] = np.nan
+        datetime_frame.loc[datetime_frame.index[:5], "B"] = -1e8
+
+        # empty
+        df = DataFrame(index=["a", "b"])
+        tm.assert_frame_equal(df, df.replace(5, 7))
+
+        # GH 11698
+        # test for mixed data types.
+        df = DataFrame(
+            [("-", pd.to_datetime("20150101")), ("a", pd.to_datetime("20150102"))]
+        )
+        df1 = df.replace("-", np.nan)
+        expected_df = DataFrame(
+            [(np.nan, pd.to_datetime("20150101")), ("a", pd.to_datetime("20150102"))]
+        )
+        tm.assert_frame_equal(df1, expected_df)
+
+    def test_replace_list(self):
+        obj = {"a": list("ab.."), "b": list("efgh"), "c": list("helo")}
+        dfobj = DataFrame(obj)
+
+        # lists of regexes and values
+        # list of [v1, v2, ..., vN] -> [v1, v2, ..., vN]
+        to_replace_res = [r".", r"e"]
+        values = [np.nan, "crap"]
+        res = dfobj.replace(to_replace_res, values)
+        expec = DataFrame(
+            {
+                "a": ["a", "b", np.nan, np.nan],
+                "b": ["crap", "f", "g", "h"],
+                "c": ["h", "crap", "l", "o"],
+            }
+        )
+        tm.assert_frame_equal(res, expec)
+
+        # list of [v1, v2, ..., vN] -> [v1, v2, .., vN]
+        to_replace_res = [r".", r"f"]
+        values = [r"..", r"crap"]
+        res = dfobj.replace(to_replace_res, values)
+        expec = DataFrame(
+            {
+                "a": ["a", "b", "..", ".."],
+                "b": ["e", "crap", "g", "h"],
+                "c": ["h", "e", "l", "o"],
+            }
+        )
+        tm.assert_frame_equal(res, expec)
+
+    def test_replace_with_empty_list(self, frame_or_series):
+        # GH 21977
+        ser = Series([["a", "b"], [], np.nan, [1]])
+        obj = DataFrame({"col": ser})
+        obj = tm.get_obj(obj, frame_or_series)
+        expected = obj
+        result = obj.replace([], np.nan)
+        tm.assert_equal(result, expected)
+
+        # GH 19266
+        msg = (
+            "NumPy boolean array indexing assignment cannot assign {size} "
+            "input values to the 1 output values where the mask is true"
+        )
+        with pytest.raises(ValueError, match=msg.format(size=0)):
+            obj.replace({np.nan: []})
+        with pytest.raises(ValueError, match=msg.format(size=2)):
+            obj.replace({np.nan: ["dummy", "alt"]})
+
+    def test_replace_series_dict(self):
+        # from GH 3064
+        df = DataFrame({"zero": {"a": 0.0, "b": 1}, "one": {"a": 2.0, "b": 0}})
+        result = df.replace(0, {"zero": 0.5, "one": 1.0})
+        expected = DataFrame({"zero": {"a": 0.5, "b": 1}, "one": {"a": 2.0, "b": 1.0}})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.replace(0, df.mean())
+        tm.assert_frame_equal(result, expected)
+
+        # series to series/dict
+        df = DataFrame({"zero": {"a": 0.0, "b": 1}, "one": {"a": 2.0, "b": 0}})
+        s = Series({"zero": 0.0, "one": 2.0})
+        result = df.replace(s, {"zero": 0.5, "one": 1.0})
+        expected = DataFrame({"zero": {"a": 0.5, "b": 1}, "one": {"a": 1.0, "b": 0.0}})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.replace(s, df.mean())
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_convert(self, any_string_dtype):
+        # gh 3907 (pandas >= 3.0 no longer converts dtypes)
+        df = DataFrame(
+            [["foo", "bar", "bah"], ["bar", "foo", "bah"]], dtype=any_string_dtype
+        )
+        m = {"foo": 1, "bar": 2, "bah": 3}
+        rep = df.replace(m)
+        assert (rep.dtypes == object).all()
+
+    def test_replace_mixed(self, float_string_frame):
+        mf = float_string_frame
+        mf.iloc[5:20, mf.columns.get_loc("foo")] = np.nan
+        mf.iloc[-10:, mf.columns.get_loc("A")] = np.nan
+
+        result = float_string_frame.replace(np.nan, -18)
+        expected = float_string_frame.copy()
+        expected["foo"] = expected["foo"].astype(object)
+        expected = expected.fillna(value=-18)
+        tm.assert_frame_equal(result, expected)
+        expected2 = float_string_frame.copy()
+        expected2["foo"] = expected2["foo"].astype(object)
+        tm.assert_frame_equal(result.replace(-18, np.nan), expected2)
+
+        result = float_string_frame.replace(np.nan, -1e8)
+        expected = float_string_frame.copy()
+        expected["foo"] = expected["foo"].astype(object)
+        expected = expected.fillna(value=-1e8)
+        tm.assert_frame_equal(result, expected)
+        expected2 = float_string_frame.copy()
+        expected2["foo"] = expected2["foo"].astype(object)
+        tm.assert_frame_equal(result.replace(-1e8, np.nan), expected2)
+
+    def test_replace_mixed_int_block_upcasting(self):
+        # int block upcasting
+        df = DataFrame(
+            {
+                "A": Series([1.0, 2.0], dtype="float64"),
+                "B": Series([0, 1], dtype="int64"),
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": Series([1.0, 2.0], dtype="float64"),
+                "B": Series([0.5, 1], dtype="float64"),
+            }
+        )
+        result = df.replace(0, 0.5)
+        tm.assert_frame_equal(result, expected)
+
+        return_value = df.replace(0, 0.5, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(df, expected)
+
+    def test_replace_mixed_int_block_splitting(self):
+        # int block splitting
+        df = DataFrame(
+            {
+                "A": Series([1.0, 2.0], dtype="float64"),
+                "B": Series([0, 1], dtype="int64"),
+                "C": Series([1, 2], dtype="int64"),
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": Series([1.0, 2.0], dtype="float64"),
+                "B": Series([0.5, 1], dtype="float64"),
+                "C": Series([1, 2], dtype="int64"),
+            }
+        )
+        result = df.replace(0, 0.5)
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_mixed2(self):
+        # to object block upcasting
+        df = DataFrame(
+            {
+                "A": Series([1.0, 2.0], dtype="float64"),
+                "B": Series([0, 1], dtype="int64"),
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": Series([1, "foo"], dtype="object"),
+                "B": Series([0, 1], dtype="int64"),
+            }
+        )
+        result = df.replace(2, "foo")
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            {
+                "A": Series(["foo", "bar"], dtype="object"),
+                "B": Series([0, "foo"], dtype="object"),
+            }
+        )
+        result = df.replace([1, 2], ["foo", "bar"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_mixed3(self):
+        # test case from
+        df = DataFrame(
+            {"A": Series([3, 0], dtype="int64"), "B": Series([0, 3], dtype="int64")}
+        )
+        result = df.replace(3, df.mean().to_dict())
+        expected = df.copy().astype("float64")
+        m = df.mean()
+        expected.iloc[0, 0] = m.iloc[0]
+        expected.iloc[1, 1] = m.iloc[1]
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_nullable_int_with_string_doesnt_cast(self):
+        # GH#25438 don't cast df['a'] to float64
+        df = DataFrame({"a": [1, 2, 3, np.nan], "b": ["some", "strings", "here", "he"]})
+        df["a"] = df["a"].astype("Int64")
+
+        res = df.replace("", np.nan)
+        tm.assert_series_equal(res["a"], df["a"])
+
+    @pytest.mark.parametrize("dtype", ["boolean", "Int64", "Float64"])
+    def test_replace_with_nullable_column(self, dtype):
+        # GH-44499
+        nullable_ser = Series([1, 0, 1], dtype=dtype)
+        df = DataFrame({"A": ["A", "B", "x"], "B": nullable_ser})
+        result = df.replace("x", "X")
+        expected = DataFrame({"A": ["A", "B", "X"], "B": nullable_ser})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_simple_nested_dict(self):
+        df = DataFrame({"col": range(1, 5)})
+        expected = DataFrame({"col": ["a", 2, 3, "b"]})
+
+        result = df.replace({"col": {1: "a", 4: "b"}})
+        tm.assert_frame_equal(expected, result)
+
+        # in this case, should be the same as the not nested version
+        result = df.replace({1: "a", 4: "b"})
+        tm.assert_frame_equal(expected, result)
+
+    def test_replace_simple_nested_dict_with_nonexistent_value(self):
+        df = DataFrame({"col": range(1, 5)})
+        expected = DataFrame({"col": ["a", 2, 3, "b"]})
+
+        result = df.replace({-1: "-", 1: "a", 4: "b"})
+        tm.assert_frame_equal(expected, result)
+
+        result = df.replace({"col": {-1: "-", 1: "a", 4: "b"}})
+        tm.assert_frame_equal(expected, result)
+
+    def test_replace_NA_with_None(self):
+        # gh-45601
+        df = DataFrame({"value": [42, None]}).astype({"value": "Int64"})
+        result = df.replace({pd.NA: None})
+        expected = DataFrame({"value": [42, None]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_NAT_with_None(self):
+        # gh-45836
+        df = DataFrame([pd.NaT, pd.NaT])
+        result = df.replace({pd.NaT: None, np.nan: None})
+        expected = DataFrame([None, None])
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_with_None_keeps_categorical(self):
+        # gh-46634
+        cat_series = Series(["b", "b", "b", "d"], dtype="category")
+        df = DataFrame(
+            {
+                "id": Series([5, 4, 3, 2], dtype="float64"),
+                "col": cat_series,
+            }
+        )
+        result = df.replace({3: None})
+
+        expected = DataFrame(
+            {
+                "id": Series([5.0, 4.0, None, 2.0], dtype="object"),
+                "col": cat_series,
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_all_NA(self):
+        # GH#60688
+        df = DataFrame({"ticker": ["#1234#"], "name": [None]})
+        result = df.replace({col: {r"^#": "$"} for col in df.columns}, regex=True)
+        expected = DataFrame({"ticker": ["$1234#"], "name": [None]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_value_is_none(self, datetime_frame):
+        orig_value = datetime_frame.iloc[0, 0]
+        orig2 = datetime_frame.iloc[1, 0]
+
+        datetime_frame.iloc[0, 0] = np.nan
+        datetime_frame.iloc[1, 0] = 1
+
+        result = datetime_frame.replace(to_replace={np.nan: 0})
+        expected = datetime_frame.T.replace(to_replace={np.nan: 0}).T
+        tm.assert_frame_equal(result, expected)
+
+        result = datetime_frame.replace(to_replace={np.nan: 0, 1: -1e8})
+        tsframe = datetime_frame.copy()
+        tsframe.iloc[0, 0] = 0
+        tsframe.iloc[1, 0] = -1e8
+        expected = tsframe
+        tm.assert_frame_equal(expected, result)
+        datetime_frame.iloc[0, 0] = orig_value
+        datetime_frame.iloc[1, 0] = orig2
+
+    def test_replace_for_new_dtypes(self, datetime_frame):
+        # dtypes
+        tsframe = datetime_frame.copy().astype(np.float32)
+        tsframe.loc[tsframe.index[:5], "A"] = np.nan
+        tsframe.loc[tsframe.index[-5:], "A"] = np.nan
+
+        zero_filled = tsframe.replace(np.nan, -1e8)
+        tm.assert_frame_equal(zero_filled, tsframe.fillna(-1e8))
+        tm.assert_frame_equal(zero_filled.replace(-1e8, np.nan), tsframe)
+
+        tsframe.loc[tsframe.index[:5], "A"] = np.nan
+        tsframe.loc[tsframe.index[-5:], "A"] = np.nan
+        tsframe.loc[tsframe.index[:5], "B"] = np.nan
+
+    @pytest.mark.parametrize(
+        "frame, to_replace, value, expected",
+        [
+            (DataFrame({"ints": [1, 2, 3]}), 1, 0, DataFrame({"ints": [0, 2, 3]})),
+            (
+                DataFrame({"ints": [1, 2, 3]}, dtype=np.int32),
+                1,
+                0,
+                DataFrame({"ints": [0, 2, 3]}, dtype=np.int32),
+            ),
+            (
+                DataFrame({"ints": [1, 2, 3]}, dtype=np.int16),
+                1,
+                0,
+                DataFrame({"ints": [0, 2, 3]}, dtype=np.int16),
+            ),
+            (
+                DataFrame({"bools": [True, False, True]}),
+                False,
+                True,
+                DataFrame({"bools": [True, True, True]}),
+            ),
+            (
+                DataFrame({"complex": [1j, 2j, 3j]}),
+                1j,
+                0,
+                DataFrame({"complex": [0j, 2j, 3j]}),
+            ),
+            (
+                DataFrame(
+                    {
+                        "datetime64": Index(
+                            [
+                                datetime(2018, 5, 28),
+                                datetime(2018, 7, 28),
+                                datetime(2018, 5, 28),
+                            ]
+                        )
+                    }
+                ),
+                datetime(2018, 5, 28),
+                datetime(2018, 7, 28),
+                DataFrame({"datetime64": Index([datetime(2018, 7, 28)] * 3)}),
+            ),
+            # GH 20380
+            (
+                DataFrame({"dt": [datetime(3017, 12, 20)], "str": ["foo"]}),
+                "foo",
+                "bar",
+                DataFrame({"dt": [datetime(3017, 12, 20)], "str": ["bar"]}),
+            ),
+            (
+                DataFrame(
+                    {
+                        "A": date_range("20130101", periods=3, tz="US/Eastern"),
+                        "B": [0, np.nan, 2],
+                    }
+                ),
+                Timestamp("20130102", tz="US/Eastern"),
+                Timestamp("20130104", tz="US/Eastern"),
+                DataFrame(
+                    {
+                        "A": pd.DatetimeIndex(
+                            [
+                                Timestamp("20130101", tz="US/Eastern"),
+                                Timestamp("20130104", tz="US/Eastern"),
+                                Timestamp("20130103", tz="US/Eastern"),
+                            ]
+                        ).as_unit("ns"),
+                        "B": [0, np.nan, 2],
+                    }
+                ),
+            ),
+            # GH 35376
+            (
+                DataFrame([[1, 1.0], [2, 2.0]]),
+                1.0,
+                5,
+                DataFrame([[5, 5.0], [2, 2.0]]),
+            ),
+            (
+                DataFrame([[1, 1.0], [2, 2.0]]),
+                1,
+                5,
+                DataFrame([[5, 5.0], [2, 2.0]]),
+            ),
+            (
+                DataFrame([[1, 1.0], [2, 2.0]]),
+                1.0,
+                5.0,
+                DataFrame([[5, 5.0], [2, 2.0]]),
+            ),
+            (
+                DataFrame([[1, 1.0], [2, 2.0]]),
+                1,
+                5.0,
+                DataFrame([[5, 5.0], [2, 2.0]]),
+            ),
+        ],
+    )
+    def test_replace_dtypes(self, frame, to_replace, value, expected):
+        result = frame.replace(to_replace, value)
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_input_formats_listlike(self):
+        # both dicts
+        to_rep = {"A": np.nan, "B": 0, "C": ""}
+        values = {"A": 0, "B": -1, "C": "missing"}
+        df = DataFrame(
+            {"A": [np.nan, 0, np.inf], "B": [0, 2, 5], "C": ["", "asdf", "fd"]}
+        )
+        filled = df.replace(to_rep, values)
+        expected = {k: v.replace(to_rep[k], values[k]) for k, v in df.items()}
+        tm.assert_frame_equal(filled, DataFrame(expected))
+
+        result = df.replace([0, 2, 5], [5, 2, 0])
+        expected = DataFrame(
+            {"A": [np.nan, 5, np.inf], "B": [5, 2, 0], "C": ["", "asdf", "fd"]}
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # scalar to dict
+        values = {"A": 0, "B": -1, "C": "missing"}
+        df = DataFrame(
+            {"A": [np.nan, 0, np.nan], "B": [0, 2, 5], "C": ["", "asdf", "fd"]}
+        )
+        filled = df.replace(np.nan, values)
+        expected = {k: v.replace(np.nan, values[k]) for k, v in df.items()}
+        tm.assert_frame_equal(filled, DataFrame(expected))
+
+        # list to list
+        to_rep = [np.nan, 0, ""]
+        values = [-2, -1, "missing"]
+        result = df.replace(to_rep, values)
+        expected = df.copy()
+        for rep, value in zip(to_rep, values):
+            return_value = expected.replace(rep, value, inplace=True)
+            assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+        msg = r"Replacement lists must match in length\. Expecting 3 got 2"
+        with pytest.raises(ValueError, match=msg):
+            df.replace(to_rep, values[1:])
+
+    def test_replace_input_formats_scalar(self):
+        df = DataFrame(
+            {"A": [np.nan, 0, np.inf], "B": [0, 2, 5], "C": ["", "asdf", "fd"]}
+        )
+
+        # dict to scalar
+        to_rep = {"A": np.nan, "B": 0, "C": ""}
+        filled = df.replace(to_rep, 0)
+        expected = {k: v.replace(to_rep[k], 0) for k, v in df.items()}
+        tm.assert_frame_equal(filled, DataFrame(expected))
+
+        msg = "value argument must be scalar, dict, or Series"
+        with pytest.raises(TypeError, match=msg):
+            df.replace(to_rep, [np.nan, 0, ""])
+
+        # list to scalar
+        to_rep = [np.nan, 0, ""]
+        result = df.replace(to_rep, -1)
+        expected = df.copy()
+        for rep in to_rep:
+            return_value = expected.replace(rep, -1, inplace=True)
+            assert return_value is None
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_limit(self):
+        # TODO
+        pass
+
+    def test_replace_dict_no_regex(self, any_string_dtype):
+        answer = Series(
+            {
+                0: "Strongly Agree",
+                1: "Agree",
+                2: "Neutral",
+                3: "Disagree",
+                4: "Strongly Disagree",
+            },
+            dtype=any_string_dtype,
+        )
+        weights = {
+            "Agree": 4,
+            "Disagree": 2,
+            "Neutral": 3,
+            "Strongly Agree": 5,
+            "Strongly Disagree": 1,
+        }
+        expected = Series({0: 5, 1: 4, 2: 3, 3: 2, 4: 1}, dtype=object)
+        result = answer.replace(weights)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_series_no_regex(self, any_string_dtype):
+        answer = Series(
+            {
+                0: "Strongly Agree",
+                1: "Agree",
+                2: "Neutral",
+                3: "Disagree",
+                4: "Strongly Disagree",
+            },
+            dtype=any_string_dtype,
+        )
+        weights = Series(
+            {
+                "Agree": 4,
+                "Disagree": 2,
+                "Neutral": 3,
+                "Strongly Agree": 5,
+                "Strongly Disagree": 1,
+            }
+        )
+        expected = Series({0: 5, 1: 4, 2: 3, 3: 2, 4: 1}, dtype=object)
+        result = answer.replace(weights)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_dict_tuple_list_ordering_remains_the_same(self):
+        df = DataFrame({"A": [np.nan, 1]})
+        res1 = df.replace(to_replace={np.nan: 0, 1: -1e8})
+        res2 = df.replace(to_replace=(1, np.nan), value=[-1e8, 0])
+        res3 = df.replace(to_replace=[1, np.nan], value=[-1e8, 0])
+
+        expected = DataFrame({"A": [0, -1e8]})
+        tm.assert_frame_equal(res1, res2)
+        tm.assert_frame_equal(res2, res3)
+        tm.assert_frame_equal(res3, expected)
+
+    def test_replace_doesnt_replace_without_regex(self):
+        df = DataFrame(
+            {
+                "fol": [1, 2, 2, 3],
+                "T_opp": ["0", "vr", "0", "0"],
+                "T_Dir": ["0", "0", "0", "bt"],
+                "T_Enh": ["vo", "0", "0", "0"],
+            }
+        )
+        res = df.replace({r"\D": 1})
+        tm.assert_frame_equal(df, res)
+
+    def test_replace_bool_with_string(self):
+        df = DataFrame({"a": [True, False], "b": list("ab")})
+        result = df.replace(True, "a")
+        expected = DataFrame({"a": ["a", False], "b": df.b})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_pure_bool_with_string_no_op(self):
+        df = DataFrame(np.random.default_rng(2).random((2, 2)) > 0.5)
+        result = df.replace("asdf", "fdsa")
+        tm.assert_frame_equal(df, result)
+
+    def test_replace_bool_with_bool(self):
+        df = DataFrame(np.random.default_rng(2).random((2, 2)) > 0.5)
+        result = df.replace(False, True)
+        expected = DataFrame(np.ones((2, 2), dtype=bool))
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_with_dict_with_bool_keys(self):
+        df = DataFrame({0: [True, False], 1: [False, True]})
+        result = df.replace({"asdf": "asdb", True: "yes"})
+        expected = DataFrame({0: ["yes", False], 1: [False, "yes"]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_dict_strings_vs_ints(self):
+        # GH#34789
+        df = DataFrame({"Y0": [1, 2], "Y1": [3, 4]})
+        result = df.replace({"replace_string": "test"})
+
+        tm.assert_frame_equal(result, df)
+
+        result = df["Y0"].replace({"replace_string": "test"})
+        tm.assert_series_equal(result, df["Y0"])
+
+    def test_replace_truthy(self):
+        df = DataFrame({"a": [True, True]})
+        r = df.replace([np.inf, -np.inf], np.nan)
+        e = df
+        tm.assert_frame_equal(r, e)
+
+    def test_nested_dict_overlapping_keys_replace_int(self):
+        # GH 27660 keep behaviour consistent for simple dictionary and
+        # nested dictionary replacement
+        df = DataFrame({"a": list(range(1, 5))})
+
+        result = df.replace({"a": dict(zip(range(1, 5), range(2, 6)))})
+        expected = df.replace(dict(zip(range(1, 5), range(2, 6))))
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_dict_overlapping_keys_replace_str(self):
+        # GH 27660
+        a = np.arange(1, 5)
+        astr = a.astype(str)
+        bstr = np.arange(2, 6).astype(str)
+        df = DataFrame({"a": astr})
+        result = df.replace(dict(zip(astr, bstr)))
+        expected = df.replace({"a": dict(zip(astr, bstr))})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_swapping_bug(self):
+        df = DataFrame({"a": [True, False, True]})
+        res = df.replace({"a": {True: "Y", False: "N"}})
+        expect = DataFrame({"a": ["Y", "N", "Y"]}, dtype=object)
+        tm.assert_frame_equal(res, expect)
+
+        df = DataFrame({"a": [0, 1, 0]})
+        res = df.replace({"a": {0: "Y", 1: "N"}})
+        expect = DataFrame({"a": ["Y", "N", "Y"]}, dtype=object)
+        tm.assert_frame_equal(res, expect)
+
+    def test_replace_datetimetz(self):
+        # GH 11326
+        # behaving poorly when presented with a datetime64[ns, tz]
+        df = DataFrame(
+            {
+                "A": date_range("20130101", periods=3, tz="US/Eastern"),
+                "B": [0, np.nan, 2],
+            }
+        )
+        result = df.replace(np.nan, 1)
+        expected = DataFrame(
+            {
+                "A": date_range("20130101", periods=3, tz="US/Eastern"),
+                "B": Series([0, 1, 2], dtype="float64"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.fillna(1)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.replace(0, np.nan)
+        expected = DataFrame(
+            {
+                "A": date_range("20130101", periods=3, tz="US/Eastern"),
+                "B": [np.nan, np.nan, 2],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.replace(
+            Timestamp("20130102", tz="US/Eastern"),
+            Timestamp("20130104", tz="US/Eastern"),
+        )
+        expected = DataFrame(
+            {
+                "A": [
+                    Timestamp("20130101", tz="US/Eastern"),
+                    Timestamp("20130104", tz="US/Eastern"),
+                    Timestamp("20130103", tz="US/Eastern"),
+                ],
+                "B": [0, np.nan, 2],
+            }
+        )
+        expected["A"] = expected["A"].dt.as_unit("ns")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.iloc[1, 0] = np.nan
+        result = result.replace({"A": pd.NaT}, Timestamp("20130104", tz="US/Eastern"))
+        tm.assert_frame_equal(result, expected)
+
+        # pre-2.0 this would coerce to object with mismatched tzs
+        result = df.copy()
+        result.iloc[1, 0] = np.nan
+        result = result.replace({"A": pd.NaT}, Timestamp("20130104", tz="US/Pacific"))
+        expected = DataFrame(
+            {
+                "A": [
+                    Timestamp("20130101", tz="US/Eastern"),
+                    Timestamp("20130104", tz="US/Pacific").tz_convert("US/Eastern"),
+                    Timestamp("20130103", tz="US/Eastern"),
+                ],
+                "B": [0, np.nan, 2],
+            }
+        )
+        expected["A"] = expected["A"].dt.as_unit("ns")
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.iloc[1, 0] = np.nan
+        result = result.replace({"A": np.nan}, Timestamp("20130104"))
+        expected = DataFrame(
+            {
+                "A": [
+                    Timestamp("20130101", tz="US/Eastern"),
+                    Timestamp("20130104"),
+                    Timestamp("20130103", tz="US/Eastern"),
+                ],
+                "B": [0, np.nan, 2],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_with_empty_dictlike(self, mix_abc):
+        # GH 15289
+        df = DataFrame(mix_abc)
+        tm.assert_frame_equal(df, df.replace({}))
+        tm.assert_frame_equal(df, df.replace(Series([], dtype=object)))
+
+        tm.assert_frame_equal(df, df.replace({"b": {}}))
+        tm.assert_frame_equal(df, df.replace(Series({"b": {}})))
+
+    @pytest.mark.parametrize(
+        "df, to_replace, exp",
+        [
+            (
+                {"col1": [1, 2, 3], "col2": [4, 5, 6]},
+                {4: 5, 5: 6, 6: 7},
+                {"col1": [1, 2, 3], "col2": [5, 6, 7]},
+            ),
+            (
+                {"col1": [1, 2, 3], "col2": ["4", "5", "6"]},
+                {"4": "5", "5": "6", "6": "7"},
+                {"col1": [1, 2, 3], "col2": ["5", "6", "7"]},
+            ),
+        ],
+    )
+    def test_replace_commutative(self, df, to_replace, exp):
+        # GH 16051
+        # DataFrame.replace() overwrites when values are non-numeric
+        # also added to data frame whilst issue was for series
+
+        df = DataFrame(df)
+
+        expected = DataFrame(exp)
+        result = df.replace(to_replace)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "replacer",
+        [
+            Timestamp("20170827"),
+            np.int8(1),
+            np.int16(1),
+            np.float32(1),
+            np.float64(1),
+        ],
+    )
+    def test_replace_replacer_dtype(self, replacer):
+        # GH26632
+        df = DataFrame(["a"], dtype=object)
+        result = df.replace({"a": replacer, "b": replacer})
+        expected = DataFrame([replacer], dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_after_convert_dtypes(self):
+        # GH31517
+        df = DataFrame({"grp": [1, 2, 3, 4, 5]}, dtype="Int64")
+        result = df.replace(1, 10)
+        expected = DataFrame({"grp": [10, 2, 3, 4, 5]}, dtype="Int64")
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_invalid_to_replace(self):
+        # GH 18634
+        # API: replace() should raise an exception if invalid argument is given
+        df = DataFrame({"one": ["a", "b ", "c"], "two": ["d ", "e ", "f "]})
+        msg = (
+            r"Expecting 'to_replace' to be either a scalar, array-like, "
+            r"dict or None, got invalid type.*"
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.replace(lambda x: x.strip())
+
+    @pytest.mark.parametrize("dtype", ["float", "float64", "int64", "Int64", "boolean"])
+    @pytest.mark.parametrize("value", [np.nan, pd.NA])
+    def test_replace_no_replacement_dtypes(self, dtype, value):
+        # https://github.com/pandas-dev/pandas/issues/32988
+        df = DataFrame(np.eye(2), dtype=dtype)
+        result = df.replace(to_replace=[None, -np.inf, np.inf], value=value)
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize("replacement", [np.nan, 5])
+    def test_replace_with_duplicate_columns(self, replacement):
+        # GH 24798
+        result = DataFrame({"A": [1, 2, 3], "A1": [4, 5, 6], "B": [7, 8, 9]})
+        result.columns = list("AAB")
+
+        expected = DataFrame(
+            {"A": [1, 2, 3], "A1": [4, 5, 6], "B": [replacement, 8, 9]}
+        )
+        expected.columns = list("AAB")
+
+        result["B"] = result["B"].replace(7, replacement)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("value", [pd.Period("2020-01"), pd.Interval(0, 5)])
+    def test_replace_ea_ignore_float(self, frame_or_series, value):
+        # GH#34871
+        obj = DataFrame({"Per": [value] * 3})
+        obj = tm.get_obj(obj, frame_or_series)
+
+        expected = obj.copy()
+        result = obj.replace(1.0, 0.0)
+        tm.assert_equal(expected, result)
+
+    @pytest.mark.parametrize(
+        "replace_dict, final_data",
+        [({"a": 1, "b": 1}, [[2, 2], [2, 2]]), ({"a": 1, "b": 2}, [[2, 1], [2, 2]])],
+    )
+    def test_categorical_replace_with_dict(self, replace_dict, final_data):
+        # GH 26988
+        df = DataFrame([[1, 1], [2, 2]], columns=["a", "b"], dtype="category")
+
+        final_data = np.array(final_data)
+
+        a = pd.Categorical(final_data[:, 0], categories=[1, 2])
+        b = pd.Categorical(final_data[:, 1], categories=[1, 2])
+
+        expected = DataFrame({"a": a, "b": b})
+        result = df.replace(replace_dict, 2)
+        tm.assert_frame_equal(result, expected)
+        msg = r"DataFrame.iloc\[:, 0\] \(column name=\"a\"\) are " "different"
+        with pytest.raises(AssertionError, match=msg):
+            # ensure non-inplace call does not affect original
+            tm.assert_frame_equal(df, expected)
+        return_value = df.replace(replace_dict, 2, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(df, expected)
+
+    def test_replace_value_category_type(self):
+        """
+        Test for #23305: to ensure category dtypes are maintained
+        after replace with direct values
+        """
+
+        # create input data
+        input_dict = {
+            "col1": [1, 2, 3, 4],
+            "col2": ["a", "b", "c", "d"],
+            "col3": [1.5, 2.5, 3.5, 4.5],
+            "col4": ["cat1", "cat2", "cat3", "cat4"],
+            "col5": ["obj1", "obj2", "obj3", "obj4"],
+        }
+        # explicitly cast columns as category and order them
+        input_df = DataFrame(data=input_dict).astype(
+            {"col2": "category", "col4": "category"}
+        )
+        input_df["col2"] = input_df["col2"].cat.reorder_categories(
+            ["a", "b", "c", "d"], ordered=True
+        )
+        input_df["col4"] = input_df["col4"].cat.reorder_categories(
+            ["cat1", "cat2", "cat3", "cat4"], ordered=True
+        )
+
+        # create expected dataframe
+        expected_dict = {
+            "col1": [1, 2, 3, 4],
+            "col2": ["a", "b", "c", "z"],
+            "col3": [1.5, 2.5, 3.5, 4.5],
+            "col4": ["cat1", "catX", "cat3", "cat4"],
+            "col5": ["obj9", "obj2", "obj3", "obj4"],
+        }
+        # explicitly cast columns as category and order them
+        expected = DataFrame(data=expected_dict).astype(
+            {"col2": "category", "col4": "category"}
+        )
+        expected["col2"] = expected["col2"].cat.reorder_categories(
+            ["a", "b", "c", "z"], ordered=True
+        )
+        expected["col4"] = expected["col4"].cat.reorder_categories(
+            ["cat1", "catX", "cat3", "cat4"], ordered=True
+        )
+
+        # replace values in input dataframe
+        input_df = input_df.apply(
+            lambda x: x.astype("category").cat.rename_categories({"d": "z"})
+        )
+        input_df = input_df.apply(
+            lambda x: x.astype("category").cat.rename_categories({"obj1": "obj9"})
+        )
+        result = input_df.apply(
+            lambda x: x.astype("category").cat.rename_categories({"cat2": "catX"})
+        )
+
+        result = result.astype({"col1": "int64", "col3": "float64", "col5": "str"})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_dict_category_type(self):
+        """
+        Test to ensure category dtypes are maintained
+        after replace with dict values
+        """
+        # GH#35268, GH#44940
+
+        # create input dataframe
+        input_dict = {"col1": ["a"], "col2": ["obj1"], "col3": ["cat1"]}
+        # explicitly cast columns as category
+        input_df = DataFrame(data=input_dict).astype(
+            {"col1": "category", "col2": "category", "col3": "category"}
+        )
+
+        # create expected dataframe
+        expected_dict = {"col1": ["z"], "col2": ["obj9"], "col3": ["catX"]}
+        # explicitly cast columns as category
+        expected = DataFrame(data=expected_dict).astype(
+            {"col1": "category", "col2": "category", "col3": "category"}
+        )
+
+        # replace values in input dataframe using a dict
+        result = input_df.apply(
+            lambda x: x.cat.rename_categories(
+                {"a": "z", "obj1": "obj9", "cat1": "catX"}
+            )
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_with_compiled_regex(self):
+        # https://github.com/pandas-dev/pandas/issues/35680
+        df = DataFrame(["a", "b", "c"])
+        regex = re.compile("^a$")
+        result = df.replace({regex: "z"}, regex=True)
+        expected = DataFrame(["z", "b", "c"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_intervals(self):
+        # https://github.com/pandas-dev/pandas/issues/35931
+        df = DataFrame({"a": [pd.Interval(0, 1), pd.Interval(0, 1)]})
+        result = df.replace({"a": {pd.Interval(0, 1): "x"}})
+        expected = DataFrame({"a": ["x", "x"]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_unicode(self):
+        # GH: 16784
+        columns_values_map = {"positive": {"正面": 1, "中立": 1, "负面": 0}}
+        df1 = DataFrame({"positive": np.ones(3)})
+        result = df1.replace(columns_values_map)
+        expected = DataFrame({"positive": np.ones(3)})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_bytes(self, frame_or_series):
+        # GH#38900
+        obj = frame_or_series(["o"]).astype("|S")
+        expected = obj.copy()
+        obj = obj.replace({None: np.nan})
+        tm.assert_equal(obj, expected)
+
+    @pytest.mark.parametrize(
+        "data, to_replace, value, expected",
+        [
+            ([1], [1.0], [0], [0]),
+            ([1], [1], [0], [0]),
+            ([1.0], [1.0], [0], [0.0]),
+            ([1.0], [1], [0], [0.0]),
+        ],
+    )
+    @pytest.mark.parametrize("box", [list, tuple, np.array])
+    def test_replace_list_with_mixed_type(
+        self, data, to_replace, value, expected, box, frame_or_series
+    ):
+        # GH#40371
+        obj = frame_or_series(data)
+        expected = frame_or_series(expected)
+        result = obj.replace(box(to_replace), value)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("val", [2, np.nan, 2.0])
+    def test_replace_value_none_dtype_numeric(self, val):
+        # GH#48231
+        df = DataFrame({"a": [1, val]})
+        result = df.replace(val, None)
+        expected = DataFrame({"a": [1, None]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({"a": [1, val]})
+        result = df.replace({val: None})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_with_nil_na(self):
+        # GH 32075
+        ser = DataFrame({"a": ["nil", pd.NA]})
+        expected = DataFrame({"a": ["anything else", pd.NA]}, index=[0, 1])
+        result = ser.replace("nil", "anything else")
+        tm.assert_frame_equal(expected, result)
+
+
+class TestDataFrameReplaceRegex:
+    @pytest.mark.parametrize(
+        "data",
+        [
+            {"a": list("ab.."), "b": list("efgh")},
+            {"a": list("ab.."), "b": list(range(4))},
+        ],
+    )
+    @pytest.mark.parametrize(
+        "to_replace,value", [(r"\s*\.\s*", np.nan), (r"\s*(\.)\s*", r"\1\1\1")]
+    )
+    @pytest.mark.parametrize("compile_regex", [True, False])
+    @pytest.mark.parametrize("regex_kwarg", [True, False])
+    @pytest.mark.parametrize("inplace", [True, False])
+    def test_regex_replace_scalar(
+        self, data, to_replace, value, compile_regex, regex_kwarg, inplace
+    ):
+        df = DataFrame(data)
+        expected = df.copy()
+
+        if compile_regex:
+            to_replace = re.compile(to_replace)
+
+        if regex_kwarg:
+            regex = to_replace
+            to_replace = None
+        else:
+            regex = True
+
+        result = df.replace(to_replace, value, inplace=inplace, regex=regex)
+
+        if inplace:
+            assert result is None
+            result = df
+
+        if value is np.nan:
+            expected_replace_val = np.nan
+        else:
+            expected_replace_val = "..."
+
+        expected.loc[expected["a"] == ".", "a"] = expected_replace_val
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("regex", [False, True])
+    @pytest.mark.parametrize("value", [1, "1"])
+    def test_replace_regex_dtype_frame(self, regex, value):
+        # GH-48644
+        df1 = DataFrame({"A": ["0"], "B": ["0"]})
+        # When value is an integer, coerce result to object.
+        # When value is a string, infer the correct string dtype.
+        dtype = object if value == 1 else None
+
+        expected_df1 = DataFrame({"A": [value], "B": [value]}, dtype=dtype)
+        result_df1 = df1.replace(to_replace="0", value=value, regex=regex)
+        tm.assert_frame_equal(result_df1, expected_df1)
+
+        df2 = DataFrame({"A": ["0"], "B": ["1"]})
+        if regex:
+            expected_df2 = DataFrame({"A": [value], "B": ["1"]}, dtype=dtype)
+        else:
+            expected_df2 = DataFrame({"A": Series([value], dtype=dtype), "B": ["1"]})
+        result_df2 = df2.replace(to_replace="0", value=value, regex=regex)
+        tm.assert_frame_equal(result_df2, expected_df2)
+
+    def test_replace_with_value_also_being_replaced(self):
+        # GH46306
+        df = DataFrame({"A": [0, 1, 2], "B": [1, 0, 2]})
+        result = df.replace({0: 1, 1: np.nan})
+        expected = DataFrame({"A": [1, np.nan, 2], "B": [np.nan, 1, 2]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_categorical_no_replacement(self):
+        # GH#46672
+        df = DataFrame(
+            {
+                "a": ["one", "two", None, "three"],
+                "b": ["one", None, "two", "three"],
+            },
+            dtype="category",
+        )
+        expected = df.copy()
+
+        result = df.replace(to_replace=[".", "def"], value=["_", None])
+        tm.assert_frame_equal(result, expected)
+
+    def test_replace_object_splitting(self, using_infer_string):
+        # GH#53977
+        df = DataFrame({"a": ["a"], "b": "b"})
+        if using_infer_string:
+            assert len(df._mgr.blocks) == 2
+        else:
+            assert len(df._mgr.blocks) == 1
+        df.replace(to_replace=r"^\s*$", value="", inplace=True, regex=True)
+        if using_infer_string:
+            assert len(df._mgr.blocks) == 2
+        else:
+            assert len(df._mgr.blocks) == 1
diff --git a/pandas/pandas/tests/frame/methods/test_reset_index.py b/pandas/pandas/tests/frame/methods/test_reset_index.py
new file mode 100644
index 0000000000000..80227c0462329
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_reset_index.py
@@ -0,0 +1,811 @@
+from datetime import datetime
+from itertools import product
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_integer_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    Timestamp,
+    cut,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def multiindex_df():
+    levels = [["A", ""], ["B", "b"]]
+    return DataFrame([[0, 2], [1, 3]], columns=MultiIndex.from_tuples(levels))
+
+
+class TestResetIndex:
+    def test_reset_index_empty_rangeindex(self):
+        # GH#45230
+        df = DataFrame(
+            columns=["brand"], dtype=np.int64, index=RangeIndex(0, 0, 1, name="foo")
+        )
+
+        df2 = df.set_index([df.index, "brand"])
+
+        result = df2.reset_index([1], drop=True)
+        tm.assert_frame_equal(result, df[[]], check_index_type=True)
+
+    def test_set_reset(self):
+        idx = Index([2**63, 2**63 + 5, 2**63 + 10], name="foo")
+
+        # set/reset
+        df = DataFrame({"A": [0, 1, 2]}, index=idx)
+        result = df.reset_index()
+        assert result["foo"].dtype == np.dtype("uint64")
+
+        df = result.set_index("foo")
+        tm.assert_index_equal(df.index, idx)
+
+    def test_set_index_reset_index_dt64tz(self):
+        idx = Index(date_range("20130101", periods=3, tz="US/Eastern"), name="foo")
+
+        # set/reset
+        df = DataFrame({"A": [0, 1, 2]}, index=idx)
+        result = df.reset_index()
+        assert result["foo"].dtype == "datetime64[ns, US/Eastern]"
+
+        df = result.set_index("foo")
+        tm.assert_index_equal(df.index, idx)
+
+    def test_reset_index_tz(self, tz_aware_fixture):
+        # GH 3950
+        # reset_index with single level
+        tz = tz_aware_fixture
+        idx = date_range("1/1/2011", periods=5, freq="D", tz=tz, name="idx")
+        df = DataFrame({"a": range(5), "b": ["A", "B", "C", "D", "E"]}, index=idx)
+
+        expected = DataFrame(
+            {
+                "idx": idx,
+                "a": range(5),
+                "b": ["A", "B", "C", "D", "E"],
+            },
+            columns=["idx", "a", "b"],
+        )
+        result = df.reset_index()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_frame_reset_index_tzaware_index(self, tz):
+        dr = date_range("2012-06-02", periods=10, tz=tz)
+        df = DataFrame(np.random.default_rng(2).standard_normal(len(dr)), dr)
+        roundtripped = df.reset_index().set_index("index")
+        xp = df.index.tz
+        rs = roundtripped.index.tz
+        assert xp == rs
+
+    def test_reset_index_with_intervals(self):
+        idx = IntervalIndex.from_breaks(np.arange(11), name="x")
+        original = DataFrame({"x": idx, "y": np.arange(10)})[["x", "y"]]
+
+        result = original.set_index("x")
+        expected = DataFrame({"y": np.arange(10)}, index=idx)
+        tm.assert_frame_equal(result, expected)
+
+        result2 = result.reset_index()
+        tm.assert_frame_equal(result2, original)
+
+    def test_reset_index(self, float_frame):
+        stacked = float_frame.stack()[::2]
+        stacked = DataFrame({"foo": stacked, "bar": stacked})
+
+        names = ["first", "second"]
+        stacked.index.names = names
+        deleveled = stacked.reset_index()
+        for i, (lev, level_codes) in enumerate(
+            zip(stacked.index.levels, stacked.index.codes)
+        ):
+            values = lev.take(level_codes)
+            name = names[i]
+            tm.assert_index_equal(values, Index(deleveled[name]))
+
+        stacked.index.names = [None, None]
+        deleveled2 = stacked.reset_index()
+        tm.assert_series_equal(
+            deleveled["first"], deleveled2["level_0"], check_names=False
+        )
+        tm.assert_series_equal(
+            deleveled["second"], deleveled2["level_1"], check_names=False
+        )
+
+        # default name assigned
+        rdf = float_frame.reset_index()
+        exp = Series(float_frame.index.values, name="index")
+        tm.assert_series_equal(rdf["index"], exp)
+
+        # default name assigned, corner case
+        df = float_frame.copy()
+        df["index"] = "foo"
+        rdf = df.reset_index()
+        exp = Series(float_frame.index.values, name="level_0")
+        tm.assert_series_equal(rdf["level_0"], exp)
+
+        # but this is ok
+        float_frame.index.name = "index"
+        deleveled = float_frame.reset_index()
+        tm.assert_series_equal(deleveled["index"], Series(float_frame.index))
+        tm.assert_index_equal(deleveled.index, Index(range(len(deleveled))), exact=True)
+
+        # preserve column names
+        float_frame.columns.name = "columns"
+        reset = float_frame.reset_index()
+        assert reset.columns.name == "columns"
+
+        # only remove certain columns
+        df = float_frame.reset_index().set_index(["index", "A", "B"])
+        rs = df.reset_index(["A", "B"])
+
+        tm.assert_frame_equal(rs, float_frame)
+
+        rs = df.reset_index(["index", "A", "B"])
+        tm.assert_frame_equal(rs, float_frame.reset_index())
+
+        rs = df.reset_index(["index", "A", "B"])
+        tm.assert_frame_equal(rs, float_frame.reset_index())
+
+        rs = df.reset_index("A")
+        xp = float_frame.reset_index().set_index(["index", "B"])
+        tm.assert_frame_equal(rs, xp)
+
+        # test resetting in place
+        df = float_frame.copy()
+        reset = float_frame.reset_index()
+        return_value = df.reset_index(inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(df, reset)
+
+        df = float_frame.reset_index().set_index(["index", "A", "B"])
+        rs = df.reset_index("A", drop=True)
+        xp = float_frame.copy()
+        del xp["A"]
+        xp = xp.set_index(["B"], append=True)
+        tm.assert_frame_equal(rs, xp)
+
+    def test_reset_index_name(self):
+        df = DataFrame(
+            [[1, 2, 3, 4], [5, 6, 7, 8]],
+            columns=["A", "B", "C", "D"],
+            index=Index(range(2), name="x"),
+        )
+        assert df.reset_index().index.name is None
+        assert df.reset_index(drop=True).index.name is None
+        return_value = df.reset_index(inplace=True)
+        assert return_value is None
+        assert df.index.name is None
+
+    @pytest.mark.parametrize("levels", [["A", "B"], [0, 1]])
+    def test_reset_index_level(self, levels):
+        df = DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]], columns=["A", "B", "C", "D"])
+
+        # With MultiIndex
+        result = df.set_index(["A", "B"]).reset_index(level=levels[0])
+        tm.assert_frame_equal(result, df.set_index("B"))
+
+        result = df.set_index(["A", "B"]).reset_index(level=levels[:1])
+        tm.assert_frame_equal(result, df.set_index("B"))
+
+        result = df.set_index(["A", "B"]).reset_index(level=levels)
+        tm.assert_frame_equal(result, df)
+
+        result = df.set_index(["A", "B"]).reset_index(level=levels, drop=True)
+        tm.assert_frame_equal(result, df[["C", "D"]])
+
+        # With single-level Index (GH 16263)
+        result = df.set_index("A").reset_index(level=levels[0])
+        tm.assert_frame_equal(result, df)
+
+        result = df.set_index("A").reset_index(level=levels[:1])
+        tm.assert_frame_equal(result, df)
+
+        result = df.set_index(["A"]).reset_index(level=levels[0], drop=True)
+        tm.assert_frame_equal(result, df[["B", "C", "D"]])
+
+    @pytest.mark.parametrize("idx_lev", [["A", "B"], ["A"]])
+    def test_reset_index_level_missing(self, idx_lev):
+        # Missing levels - for both MultiIndex and single-level Index:
+        df = DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]], columns=["A", "B", "C", "D"])
+
+        with pytest.raises(KeyError, match=r"(L|l)evel \(?E\)?"):
+            df.set_index(idx_lev).reset_index(level=["A", "E"])
+        with pytest.raises(IndexError, match="Too many levels"):
+            df.set_index(idx_lev).reset_index(level=[0, 1, 2])
+
+    def test_reset_index_right_dtype(self):
+        time = np.arange(0.0, 10, np.sqrt(2) / 2)
+        s1 = Series((9.81 * time**2) / 2, index=Index(time, name="time"), name="speed")
+        df = DataFrame(s1)
+
+        reset = s1.reset_index()
+        assert reset["time"].dtype == np.float64
+
+        reset = df.reset_index()
+        assert reset["time"].dtype == np.float64
+
+    def test_reset_index_multiindex_col(self):
+        vals = np.random.default_rng(2).standard_normal((3, 3)).astype(object)
+        idx = ["x", "y", "z"]
+        full = np.hstack(([[x] for x in idx], vals))
+        df = DataFrame(
+            vals,
+            Index(idx, name="a"),
+            columns=[["b", "b", "c"], ["mean", "median", "mean"]],
+        )
+        rs = df.reset_index()
+        xp = DataFrame(
+            full, columns=[["a", "b", "b", "c"], ["", "mean", "median", "mean"]]
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        rs = df.reset_index(col_fill=None)
+        xp = DataFrame(
+            full, columns=[["a", "b", "b", "c"], ["a", "mean", "median", "mean"]]
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        rs = df.reset_index(col_level=1, col_fill="blah")
+        xp = DataFrame(
+            full, columns=[["blah", "b", "b", "c"], ["a", "mean", "median", "mean"]]
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        df = DataFrame(
+            vals,
+            MultiIndex.from_arrays([[0, 1, 2], ["x", "y", "z"]], names=["d", "a"]),
+            columns=[["b", "b", "c"], ["mean", "median", "mean"]],
+        )
+        rs = df.reset_index("a")
+        xp = DataFrame(
+            full,
+            Index([0, 1, 2], name="d"),
+            columns=[["a", "b", "b", "c"], ["", "mean", "median", "mean"]],
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        rs = df.reset_index("a", col_fill=None)
+        xp = DataFrame(
+            full,
+            Index(range(3), name="d"),
+            columns=[["a", "b", "b", "c"], ["a", "mean", "median", "mean"]],
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        rs = df.reset_index("a", col_fill="blah", col_level=1)
+        xp = DataFrame(
+            full,
+            Index(range(3), name="d"),
+            columns=[["blah", "b", "b", "c"], ["a", "mean", "median", "mean"]],
+        )
+        tm.assert_frame_equal(rs, xp)
+
+    def test_reset_index_multiindex_nan(self):
+        # GH#6322, testing reset_index on MultiIndexes
+        # when we have a nan or all nan
+        df = DataFrame(
+            {
+                "A": ["a", "b", "c"],
+                "B": [0, 1, np.nan],
+                "C": np.random.default_rng(2).random(3),
+            }
+        )
+        rs = df.set_index(["A", "B"]).reset_index()
+        tm.assert_frame_equal(rs, df)
+
+        df = DataFrame(
+            {
+                "A": [np.nan, "b", "c"],
+                "B": [0, 1, 2],
+                "C": np.random.default_rng(2).random(3),
+            }
+        )
+        rs = df.set_index(["A", "B"]).reset_index()
+        tm.assert_frame_equal(rs, df)
+
+        df = DataFrame({"A": ["a", "b", "c"], "B": [0, 1, 2], "C": [np.nan, 1.1, 2.2]})
+        rs = df.set_index(["A", "B"]).reset_index()
+        tm.assert_frame_equal(rs, df)
+
+        df = DataFrame(
+            {
+                "A": ["a", "b", "c"],
+                "B": [np.nan, np.nan, np.nan],
+                "C": np.random.default_rng(2).random(3),
+            }
+        )
+        rs = df.set_index(["A", "B"]).reset_index()
+        tm.assert_frame_equal(rs, df)
+
+    @pytest.mark.parametrize(
+        "name",
+        [
+            None,
+            "foo",
+            2,
+            3.0,
+            pd.Timedelta(6),
+            Timestamp("2012-12-30", tz="UTC"),
+            "2012-12-31",
+        ],
+    )
+    def test_reset_index_with_datetimeindex_cols(self, name):
+        # GH#5818
+        df = DataFrame(
+            [[1, 2], [3, 4]],
+            columns=date_range("1/1/2013", "1/2/2013"),
+            index=["A", "B"],
+        )
+        df.index.name = name
+
+        result = df.reset_index()
+
+        item = name if name is not None else "index"
+        columns = Index([item, datetime(2013, 1, 1), datetime(2013, 1, 2)])
+        if isinstance(item, str) and item == "2012-12-31":
+            columns = columns.astype("datetime64[ns]")
+        else:
+            assert columns.dtype == object
+
+        expected = DataFrame(
+            [["A", 1, 2], ["B", 3, 4]],
+            columns=columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reset_index_range(self):
+        # GH#12071
+        df = DataFrame([[0, 0], [1, 1]], columns=["A", "B"], index=RangeIndex(stop=2))
+        result = df.reset_index()
+        assert isinstance(result.index, RangeIndex)
+        expected = DataFrame(
+            [[0, 0, 0], [1, 1, 1]],
+            columns=["index", "A", "B"],
+            index=RangeIndex(stop=2),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_reset_index_multiindex_columns(self, multiindex_df):
+        result = multiindex_df[["B"]].rename_axis("A").reset_index()
+        tm.assert_frame_equal(result, multiindex_df)
+
+        # GH#16120: already existing column
+        msg = r"cannot insert \('A', ''\), already exists"
+        with pytest.raises(ValueError, match=msg):
+            multiindex_df.rename_axis("A").reset_index()
+
+        # GH#16164: multiindex (tuple) full key
+        result = multiindex_df.set_index([("A", "")]).reset_index()
+        tm.assert_frame_equal(result, multiindex_df)
+
+        # with additional (unnamed) index level
+        idx_col = DataFrame(
+            [[0], [1]], columns=MultiIndex.from_tuples([("level_0", "")])
+        )
+        expected = pd.concat([idx_col, multiindex_df[[("B", "b"), ("A", "")]]], axis=1)
+        result = multiindex_df.set_index([("B", "b")], append=True).reset_index()
+        tm.assert_frame_equal(result, expected)
+
+        # with index name which is a too long tuple...
+        msg = "Item must have length equal to number of levels."
+        with pytest.raises(ValueError, match=msg):
+            multiindex_df.rename_axis([("C", "c", "i")]).reset_index()
+
+        # or too short...
+        levels = [["A", "a", ""], ["B", "b", "i"]]
+        df2 = DataFrame([[0, 2], [1, 3]], columns=MultiIndex.from_tuples(levels))
+        idx_col = DataFrame(
+            [[0], [1]], columns=MultiIndex.from_tuples([("C", "c", "ii")])
+        )
+        expected = pd.concat([idx_col, df2], axis=1)
+        result = df2.rename_axis([("C", "c")]).reset_index(col_fill="ii")
+        tm.assert_frame_equal(result, expected)
+
+        # ... which is incompatible with col_fill=None
+        with pytest.raises(
+            ValueError,
+            match=(
+                "col_fill=None is incompatible with "
+                r"incomplete column name \('C', 'c'\)"
+            ),
+        ):
+            df2.rename_axis([("C", "c")]).reset_index(col_fill=None)
+
+        # with col_level != 0
+        result = df2.rename_axis([("c", "ii")]).reset_index(col_level=1, col_fill="C")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("flag", [False, True])
+    @pytest.mark.parametrize("allow_duplicates", [False, True])
+    def test_reset_index_duplicate_columns_allow(
+        self, multiindex_df, flag, allow_duplicates
+    ):
+        # GH#44755 reset_index with duplicate column labels
+        df = multiindex_df.rename_axis("A")
+        df = df.set_flags(allows_duplicate_labels=flag)
+
+        if flag and allow_duplicates:
+            result = df.reset_index(allow_duplicates=allow_duplicates)
+            levels = [["A", ""], ["A", ""], ["B", "b"]]
+            expected = DataFrame(
+                [[0, 0, 2], [1, 1, 3]], columns=MultiIndex.from_tuples(levels)
+            )
+            tm.assert_frame_equal(result, expected)
+        else:
+            if not flag and allow_duplicates:
+                msg = (
+                    "Cannot specify 'allow_duplicates=True' when "
+                    "'self.flags.allows_duplicate_labels' is False"
+                )
+            else:
+                msg = r"cannot insert \('A', ''\), already exists"
+            with pytest.raises(ValueError, match=msg):
+                df.reset_index(allow_duplicates=allow_duplicates)
+
+    @pytest.mark.parametrize("flag", [False, True])
+    def test_reset_index_duplicate_columns_default(self, multiindex_df, flag):
+        df = multiindex_df.rename_axis("A")
+        df = df.set_flags(allows_duplicate_labels=flag)
+
+        msg = r"cannot insert \('A', ''\), already exists"
+        with pytest.raises(ValueError, match=msg):
+            df.reset_index()
+
+    @pytest.mark.parametrize("allow_duplicates", ["bad value"])
+    def test_reset_index_allow_duplicates_check(self, multiindex_df, allow_duplicates):
+        with pytest.raises(ValueError, match="expected type bool"):
+            multiindex_df.reset_index(allow_duplicates=allow_duplicates)
+
+    def test_reset_index_datetime(self, tz_naive_fixture):
+        # GH#3950
+        tz = tz_naive_fixture
+        idx1 = date_range("1/1/2011", periods=5, freq="D", tz=tz, name="idx1")
+        idx2 = Index(range(5), name="idx2", dtype="int64")
+        idx = MultiIndex.from_arrays([idx1, idx2])
+        df = DataFrame(
+            {"a": np.arange(5, dtype="int64"), "b": ["A", "B", "C", "D", "E"]},
+            index=idx,
+        )
+
+        expected = DataFrame(
+            {
+                "idx1": idx1,
+                "idx2": np.arange(5, dtype="int64"),
+                "a": np.arange(5, dtype="int64"),
+                "b": ["A", "B", "C", "D", "E"],
+            },
+            columns=["idx1", "idx2", "a", "b"],
+        )
+
+        tm.assert_frame_equal(df.reset_index(), expected)
+
+    def test_reset_index_datetime2(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        idx1 = date_range("1/1/2011", periods=5, freq="D", tz=tz, name="idx1")
+        idx2 = Index(range(5), name="idx2", dtype="int64")
+        idx3 = date_range(
+            "1/1/2012", periods=5, freq="MS", tz="Europe/Paris", name="idx3"
+        )
+        idx = MultiIndex.from_arrays([idx1, idx2, idx3])
+        df = DataFrame(
+            {"a": np.arange(5, dtype="int64"), "b": ["A", "B", "C", "D", "E"]},
+            index=idx,
+        )
+
+        expected = DataFrame(
+            {
+                "idx1": idx1,
+                "idx2": np.arange(5, dtype="int64"),
+                "idx3": idx3,
+                "a": np.arange(5, dtype="int64"),
+                "b": ["A", "B", "C", "D", "E"],
+            },
+            columns=["idx1", "idx2", "idx3", "a", "b"],
+        )
+        result = df.reset_index()
+        tm.assert_frame_equal(result, expected)
+
+    def test_reset_index_datetime3(self, tz_naive_fixture):
+        # GH#7793
+        tz = tz_naive_fixture
+        dti = date_range("20130101", periods=3, tz=tz)
+        idx = MultiIndex.from_product([["a", "b"], dti])
+        df = DataFrame(
+            np.arange(6, dtype="int64").reshape(6, 1), columns=["a"], index=idx
+        )
+
+        expected = DataFrame(
+            {
+                "level_0": "a a a b b b".split(),
+                "level_1": dti.append(dti),
+                "a": np.arange(6, dtype="int64"),
+            },
+            columns=["level_0", "level_1", "a"],
+        )
+        result = df.reset_index()
+        tm.assert_frame_equal(result, expected)
+
+    def test_reset_index_period(self):
+        # GH#7746
+        idx = MultiIndex.from_product(
+            [pd.period_range("20130101", periods=3, freq="M"), list("abc")],
+            names=["month", "feature"],
+        )
+
+        df = DataFrame(
+            np.arange(9, dtype="int64").reshape(-1, 1), index=idx, columns=["a"]
+        )
+        expected = DataFrame(
+            {
+                "month": (
+                    [pd.Period("2013-01", freq="M")] * 3
+                    + [pd.Period("2013-02", freq="M")] * 3
+                    + [pd.Period("2013-03", freq="M")] * 3
+                ),
+                "feature": ["a", "b", "c"] * 3,
+                "a": np.arange(9, dtype="int64"),
+            },
+            columns=["month", "feature", "a"],
+        )
+        result = df.reset_index()
+        tm.assert_frame_equal(result, expected)
+
+    def test_reset_index_delevel_infer_dtype(self):
+        tuples = list(product(["foo", "bar"], [10, 20], [1.0, 1.1]))
+        index = MultiIndex.from_tuples(tuples, names=["prm0", "prm1", "prm2"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 3)),
+            columns=["A", "B", "C"],
+            index=index,
+        )
+        deleveled = df.reset_index()
+        assert is_integer_dtype(deleveled["prm1"])
+        assert is_float_dtype(deleveled["prm2"])
+
+    def test_reset_index_with_drop(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        deleveled = ymd.reset_index(drop=True)
+        assert len(deleveled.columns) == len(ymd.columns)
+        assert deleveled.index.name == ymd.index.name
+
+    @pytest.mark.parametrize(
+        "ix_data, exp_data",
+        [
+            (
+                [(pd.NaT, 1), (pd.NaT, 2)],
+                {"a": [pd.NaT, pd.NaT], "b": [1, 2], "x": [11, 12]},
+            ),
+            (
+                [(pd.NaT, 1), (Timestamp("2020-01-01"), 2)],
+                {"a": [pd.NaT, Timestamp("2020-01-01")], "b": [1, 2], "x": [11, 12]},
+            ),
+            (
+                [(pd.NaT, 1), (pd.Timedelta(123, "D"), 2)],
+                {"a": [pd.NaT, pd.Timedelta(123, "D")], "b": [1, 2], "x": [11, 12]},
+            ),
+        ],
+    )
+    def test_reset_index_nat_multiindex(self, ix_data, exp_data):
+        # GH#36541: that reset_index() does not raise ValueError
+        ix = MultiIndex.from_tuples(ix_data, names=["a", "b"])
+        result = DataFrame({"x": [11, 12]}, index=ix)
+        result = result.reset_index()
+
+        expected = DataFrame(exp_data)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "codes", ([[0, 0, 1, 1], [0, 1, 0, 1]], [[0, 0, -1, 1], [0, 1, 0, 1]])
+    )
+    def test_rest_index_multiindex_categorical_with_missing_values(self, codes):
+        # GH#24206
+
+        index = MultiIndex(
+            [CategoricalIndex(["A", "B"]), CategoricalIndex(["a", "b"])], codes
+        )
+        data = {"col": range(len(index))}
+        df = DataFrame(data=data, index=index)
+
+        expected = DataFrame(
+            {
+                "level_0": Categorical.from_codes(codes[0], categories=["A", "B"]),
+                "level_1": Categorical.from_codes(codes[1], categories=["a", "b"]),
+                "col": range(4),
+            }
+        )
+
+        res = df.reset_index()
+        tm.assert_frame_equal(res, expected)
+
+        # roundtrip
+        res = expected.set_index(["level_0", "level_1"]).reset_index()
+        tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "array, dtype",
+    [
+        (["a", "b"], object),
+        (
+            pd.period_range("12-1-2000", periods=2, freq="Q-DEC"),
+            pd.PeriodDtype(freq="Q-DEC"),
+        ),
+    ],
+)
+def test_reset_index_dtypes_on_empty_frame_with_multiindex(
+    array, dtype, using_infer_string
+):
+    # GH 19602 - Preserve dtype on empty DataFrame with MultiIndex
+    idx = MultiIndex.from_product([[0, 1], [0.5, 1.0], array])
+    result = DataFrame(index=idx)[:0].reset_index().dtypes
+    if using_infer_string and dtype == object:
+        dtype = pd.StringDtype(na_value=np.nan)
+    expected = Series({"level_0": np.int64, "level_1": np.float64, "level_2": dtype})
+    tm.assert_series_equal(result, expected)
+
+
+def test_reset_index_empty_frame_with_datetime64_multiindex():
+    # https://github.com/pandas-dev/pandas/issues/35606
+    dti = pd.DatetimeIndex(["2020-07-20 00:00:00"], dtype="M8[ns]")
+    idx = MultiIndex.from_product([dti, [3, 4]], names=["a", "b"])[:0]
+    df = DataFrame(index=idx, columns=["c", "d"])
+    result = df.reset_index()
+    expected = DataFrame(
+        columns=list("abcd"), index=RangeIndex(start=0, stop=0, step=1)
+    )
+    expected["a"] = expected["a"].astype("datetime64[ns]")
+    expected["b"] = expected["b"].astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reset_index_empty_frame_with_datetime64_multiindex_from_groupby(
+    using_infer_string,
+):
+    # https://github.com/pandas-dev/pandas/issues/35657
+    dti = pd.DatetimeIndex(["2020-01-01"], dtype="M8[ns]")
+    df = DataFrame({"c1": [10.0], "c2": ["a"], "c3": dti})
+    df = df.head(0).groupby(["c2", "c3"])[["c1"]].sum()
+    result = df.reset_index()
+    expected = DataFrame(
+        columns=["c2", "c3", "c1"], index=RangeIndex(start=0, stop=0, step=1)
+    )
+    expected["c3"] = expected["c3"].astype("datetime64[ns]")
+    expected["c1"] = expected["c1"].astype("float64")
+    if using_infer_string:
+        expected["c2"] = expected["c2"].astype("str")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reset_index_multiindex_nat():
+    # GH 11479
+    idx = range(3)
+    tstamp = date_range("2015-07-01", freq="D", periods=3)
+    df = DataFrame({"id": idx, "tstamp": tstamp, "a": list("abc")})
+    df.loc[2, "tstamp"] = pd.NaT
+    result = df.set_index(["id", "tstamp"]).reset_index("id")
+    exp_dti = pd.DatetimeIndex(
+        ["2015-07-01", "2015-07-02", "NaT"], dtype="M8[ns]", name="tstamp"
+    )
+    expected = DataFrame(
+        {"id": range(3), "a": list("abc")},
+        index=exp_dti,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reset_index_interval_columns_object_cast():
+    # GH 19136
+    df = DataFrame(
+        np.eye(2), index=Index([1, 2], name="Year"), columns=cut([1, 2], [0, 1, 2])
+    )
+    result = df.reset_index()
+    expected = DataFrame(
+        [[1, 1.0, 0.0], [2, 0.0, 1.0]],
+        columns=Index(["Year", Interval(0, 1), Interval(1, 2)]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reset_index_rename(float_frame):
+    # GH 6878
+    result = float_frame.reset_index(names="new_name")
+    expected = Series(float_frame.index.values, name="new_name")
+    tm.assert_series_equal(result["new_name"], expected)
+
+    result = float_frame.reset_index(names=123)
+    expected = Series(float_frame.index.values, name=123)
+    tm.assert_series_equal(result[123], expected)
+
+
+def test_reset_index_rename_multiindex(float_frame):
+    # GH 6878
+    stacked_df = float_frame.stack()[::2]
+    stacked_df = DataFrame({"foo": stacked_df, "bar": stacked_df})
+
+    names = ["first", "second"]
+    stacked_df.index.names = names
+
+    result = stacked_df.reset_index()
+    expected = stacked_df.reset_index(names=["new_first", "new_second"])
+    tm.assert_series_equal(result["first"], expected["new_first"], check_names=False)
+    tm.assert_series_equal(result["second"], expected["new_second"], check_names=False)
+
+
+def test_errorreset_index_rename(float_frame):
+    # GH 6878
+    stacked_df = float_frame.stack()[::2]
+    stacked_df = DataFrame({"first": stacked_df, "second": stacked_df})
+
+    with pytest.raises(
+        ValueError, match="Index names must be str or 1-dimensional list"
+    ):
+        stacked_df.reset_index(names={"first": "new_first", "second": "new_second"})
+
+    with pytest.raises(IndexError, match="list index out of range"):
+        stacked_df.reset_index(names=["new_first"])
+
+
+def test_reset_index_false_index_name():
+    result_series = Series(data=range(5, 10), index=range(5))
+    result_series.index.name = False
+    result_series.reset_index()
+    expected_series = Series(range(5, 10), RangeIndex(range(5), name=False))
+    tm.assert_series_equal(result_series, expected_series)
+
+    # GH 38147
+    result_frame = DataFrame(data=range(5, 10), index=range(5))
+    result_frame.index.name = False
+    result_frame.reset_index()
+    expected_frame = DataFrame(range(5, 10), RangeIndex(range(5), name=False))
+    tm.assert_frame_equal(result_frame, expected_frame)
+
+
+@pytest.mark.parametrize("columns", [None, Index([])])
+def test_reset_index_with_empty_frame(columns):
+    # Currently empty DataFrame has RangeIndex or object dtype Index, but when
+    # resetting the index we still want to end up with the default string dtype
+    # https://github.com/pandas-dev/pandas/issues/60338
+
+    index = Index([], name="foo")
+    df = DataFrame(index=index, columns=columns)
+    result = df.reset_index()
+    expected = DataFrame(columns=["foo"])
+    tm.assert_frame_equal(result, expected)
+
+    index = Index([1, 2, 3], name="foo")
+    df = DataFrame(index=index, columns=columns)
+    result = df.reset_index()
+    expected = DataFrame({"foo": [1, 2, 3]})
+    tm.assert_frame_equal(result, expected)
+
+    index = MultiIndex.from_tuples([], names=["foo", "bar"])
+    df = DataFrame(index=index, columns=columns)
+    result = df.reset_index()
+    expected = DataFrame(columns=["foo", "bar"])
+    tm.assert_frame_equal(result, expected)
+
+    index = MultiIndex.from_tuples([(1, 2), (2, 3)], names=["foo", "bar"])
+    df = DataFrame(index=index, columns=columns)
+    result = df.reset_index()
+    expected = DataFrame({"foo": [1, 2], "bar": [2, 3]})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_round.py b/pandas/pandas/tests/frame/methods/test_round.py
new file mode 100644
index 0000000000000..a96df27b48d7d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_round.py
@@ -0,0 +1,225 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameRound:
+    def test_round(self):
+        # GH#2665
+
+        # Test that rounding an empty DataFrame does nothing
+        df = DataFrame()
+        tm.assert_frame_equal(df, df.round())
+
+        # Here's the test frame we'll be working with
+        df = DataFrame({"col1": [1.123, 2.123, 3.123], "col2": [1.234, 2.234, 3.234]})
+
+        # Default round to integer (i.e. decimals=0)
+        expected_rounded = DataFrame({"col1": [1.0, 2.0, 3.0], "col2": [1.0, 2.0, 3.0]})
+        tm.assert_frame_equal(df.round(), expected_rounded)
+
+        # Round with an integer
+        decimals = 2
+        expected_rounded = DataFrame(
+            {"col1": [1.12, 2.12, 3.12], "col2": [1.23, 2.23, 3.23]}
+        )
+        tm.assert_frame_equal(df.round(decimals), expected_rounded)
+
+        # This should also work with np.round (since np.round dispatches to
+        # df.round)
+        tm.assert_frame_equal(np.round(df, decimals), expected_rounded)
+
+        # Round with a list
+        round_list = [1, 2]
+        msg = "decimals must be an integer, a dict-like or a Series"
+        with pytest.raises(TypeError, match=msg):
+            df.round(round_list)
+
+        # Round with a dictionary
+        expected_rounded = DataFrame(
+            {"col1": [1.1, 2.1, 3.1], "col2": [1.23, 2.23, 3.23]}
+        )
+        round_dict = {"col1": 1, "col2": 2}
+        tm.assert_frame_equal(df.round(round_dict), expected_rounded)
+
+        # Incomplete dict
+        expected_partially_rounded = DataFrame(
+            {"col1": [1.123, 2.123, 3.123], "col2": [1.2, 2.2, 3.2]}
+        )
+        partial_round_dict = {"col2": 1}
+        tm.assert_frame_equal(df.round(partial_round_dict), expected_partially_rounded)
+
+        # Dict with unknown elements
+        wrong_round_dict = {"col3": 2, "col2": 1}
+        tm.assert_frame_equal(df.round(wrong_round_dict), expected_partially_rounded)
+
+        # float input to `decimals`
+        non_int_round_dict = {"col1": 1, "col2": 0.5}
+        msg = "Values in decimals must be integers"
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_dict)
+
+        # String input
+        non_int_round_dict = {"col1": 1, "col2": "foo"}
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_dict)
+
+        non_int_round_Series = Series(non_int_round_dict)
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_Series)
+
+        # List input
+        non_int_round_dict = {"col1": 1, "col2": [1, 2]}
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_dict)
+
+        non_int_round_Series = Series(non_int_round_dict)
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_Series)
+
+        # Non integer Series inputs
+        non_int_round_Series = Series(non_int_round_dict)
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_Series)
+
+        non_int_round_Series = Series(non_int_round_dict)
+        with pytest.raises(TypeError, match=msg):
+            df.round(non_int_round_Series)
+
+        # Negative numbers
+        negative_round_dict = {"col1": -1, "col2": -2}
+        big_df = df * 100
+        expected_neg_rounded = DataFrame(
+            {"col1": [110.0, 210, 310], "col2": [100.0, 200, 300]}
+        )
+        tm.assert_frame_equal(big_df.round(negative_round_dict), expected_neg_rounded)
+
+        # nan in Series round
+        nan_round_Series = Series({"col1": np.nan, "col2": 1})
+
+        with pytest.raises(TypeError, match=msg):
+            df.round(nan_round_Series)
+
+        # Make sure this doesn't break existing Series.round
+        tm.assert_series_equal(df["col1"].round(1), expected_rounded["col1"])
+
+        # named columns
+        # GH#11986
+        decimals = 2
+        expected_rounded = DataFrame(
+            {"col1": [1.12, 2.12, 3.12], "col2": [1.23, 2.23, 3.23]}
+        )
+        df.columns.name = "cols"
+        expected_rounded.columns.name = "cols"
+        tm.assert_frame_equal(df.round(decimals), expected_rounded)
+
+        # interaction of named columns & series
+        tm.assert_series_equal(df["col1"].round(decimals), expected_rounded["col1"])
+        tm.assert_series_equal(df.round(decimals)["col1"], expected_rounded["col1"])
+
+    def test_round_numpy(self):
+        # GH#12600
+        df = DataFrame([[1.53, 1.36], [0.06, 7.01]])
+        out = np.round(df, decimals=0)
+        expected = DataFrame([[2.0, 1.0], [0.0, 7.0]])
+        tm.assert_frame_equal(out, expected)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.round(df, decimals=0, out=df)
+
+    def test_round_numpy_with_nan(self):
+        # See GH#14197
+        df = Series([1.53, np.nan, 0.06]).to_frame()
+        with tm.assert_produces_warning(None):
+            result = df.round()
+        expected = Series([2.0, np.nan, 0.0]).to_frame()
+        tm.assert_frame_equal(result, expected)
+
+    def test_round_mixed_type(self):
+        # GH#11885
+        df = DataFrame(
+            {
+                "col1": [1.1, 2.2, 3.3, 4.4],
+                "col2": ["1", "a", "c", "f"],
+                "col3": date_range("20111111", periods=4),
+            }
+        )
+        round_0 = DataFrame(
+            {
+                "col1": [1.0, 2.0, 3.0, 4.0],
+                "col2": ["1", "a", "c", "f"],
+                "col3": date_range("20111111", periods=4),
+            }
+        )
+        tm.assert_frame_equal(df.round(), round_0)
+        tm.assert_frame_equal(df.round(1), df)
+        tm.assert_frame_equal(df.round({"col1": 1}), df)
+        tm.assert_frame_equal(df.round({"col1": 0}), round_0)
+        tm.assert_frame_equal(df.round({"col1": 0, "col2": 1}), round_0)
+        tm.assert_frame_equal(df.round({"col3": 1}), df)
+
+    def test_round_with_duplicate_columns(self):
+        # GH#11611
+
+        df = DataFrame(
+            np.random.default_rng(2).random([3, 3]),
+            columns=["A", "B", "C"],
+            index=["first", "second", "third"],
+        )
+
+        dfs = pd.concat((df, df), axis=1)
+        rounded = dfs.round()
+        tm.assert_index_equal(rounded.index, dfs.index)
+
+        decimals = Series([1, 0, 2], index=["A", "B", "A"])
+        msg = "Index of decimals must be unique"
+        with pytest.raises(ValueError, match=msg):
+            df.round(decimals)
+
+    def test_round_builtin(self):
+        # GH#11763
+        # Here's the test frame we'll be working with
+        df = DataFrame({"col1": [1.123, 2.123, 3.123], "col2": [1.234, 2.234, 3.234]})
+
+        # Default round to integer (i.e. decimals=0)
+        expected_rounded = DataFrame({"col1": [1.0, 2.0, 3.0], "col2": [1.0, 2.0, 3.0]})
+        tm.assert_frame_equal(round(df), expected_rounded)
+
+    def test_round_nonunique_categorical(self):
+        # See GH#21809
+        idx = pd.CategoricalIndex(["low"] * 3 + ["hi"] * 3)
+        df = DataFrame(np.random.default_rng(2).random((6, 3)), columns=list("abc"))
+
+        expected = df.round(3)
+        expected.index = idx
+
+        df_categorical = df.copy().set_index(idx)
+        assert df_categorical.shape == (6, 3)
+        result = df_categorical.round(3)
+        assert result.shape == (6, 3)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_round_interval_category_columns(self):
+        # GH#30063
+        columns = pd.CategoricalIndex(pd.interval_range(0, 2))
+        df = DataFrame([[0.66, 1.1], [0.3, 0.25]], columns=columns)
+
+        result = df.round()
+        expected = DataFrame([[1.0, 1.0], [0.0, 0.0]], columns=columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_round_empty_not_input(self):
+        # GH#51032
+        df = DataFrame()
+        result = df.round()
+        tm.assert_frame_equal(df, result)
+        assert df is not result
diff --git a/pandas/pandas/tests/frame/methods/test_sample.py b/pandas/pandas/tests/frame/methods/test_sample.py
new file mode 100644
index 0000000000000..a9d56cbfd2b46
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_sample.py
@@ -0,0 +1,369 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+import pandas.core.common as com
+
+
+class TestSample:
+    @pytest.fixture
+    def obj(self, frame_or_series):
+        if frame_or_series is Series:
+            arr = np.random.default_rng(2).standard_normal(10)
+        else:
+            arr = np.random.default_rng(2).standard_normal((10, 10))
+        return frame_or_series(arr, dtype=None)
+
+    @pytest.mark.parametrize("test", list(range(10)))
+    def test_sample(self, test, obj):
+        # Fixes issue: 2419
+        # Check behavior of random_state argument
+        # Check for stability when receives seed or random state -- run 10
+        # times.
+
+        seed = np.random.default_rng(2).integers(0, 100)
+        tm.assert_equal(
+            obj.sample(n=4, random_state=seed), obj.sample(n=4, random_state=seed)
+        )
+
+        tm.assert_equal(
+            obj.sample(frac=0.7, random_state=seed),
+            obj.sample(frac=0.7, random_state=seed),
+        )
+
+        tm.assert_equal(
+            obj.sample(n=4, random_state=np.random.default_rng(test)),
+            obj.sample(n=4, random_state=np.random.default_rng(test)),
+        )
+
+        tm.assert_equal(
+            obj.sample(frac=0.7, random_state=np.random.default_rng(test)),
+            obj.sample(frac=0.7, random_state=np.random.default_rng(test)),
+        )
+
+        tm.assert_equal(
+            obj.sample(
+                frac=2,
+                replace=True,
+                random_state=np.random.default_rng(test),
+            ),
+            obj.sample(
+                frac=2,
+                replace=True,
+                random_state=np.random.default_rng(test),
+            ),
+        )
+
+        os1, os2 = [], []
+        for _ in range(2):
+            os1.append(obj.sample(n=4, random_state=test))
+            os2.append(obj.sample(frac=0.7, random_state=test))
+        tm.assert_equal(*os1)
+        tm.assert_equal(*os2)
+
+    def test_sample_lengths(self, obj):
+        # Check lengths are right
+        assert len(obj.sample(n=4) == 4)
+        assert len(obj.sample(frac=0.34) == 3)
+        assert len(obj.sample(frac=0.36) == 4)
+
+    def test_sample_invalid_random_state(self, obj):
+        # Check for error when random_state argument invalid.
+        msg = (
+            "random_state must be an integer, array-like, a BitGenerator, Generator, "
+            "a numpy RandomState, or None"
+        )
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(random_state="a_string")
+
+    def test_sample_wont_accept_n_and_frac(self, obj):
+        # Giving both frac and N throws error
+        msg = "Please enter a value for `frac` OR `n`, not both"
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, frac=0.3)
+
+    def test_sample_requires_positive_n_frac(self, obj):
+        with pytest.raises(
+            ValueError,
+            match="A negative number of rows requested. Please provide `n` >= 0",
+        ):
+            obj.sample(n=-3)
+        with pytest.raises(
+            ValueError,
+            match="A negative number of rows requested. Please provide `frac` >= 0",
+        ):
+            obj.sample(frac=-0.3)
+
+    def test_sample_requires_integer_n(self, obj):
+        # Make sure float values of `n` give error
+        with pytest.raises(ValueError, match="Only integers accepted as `n` values"):
+            obj.sample(n=3.2)
+
+    def test_sample_invalid_weight_lengths(self, obj):
+        # Weight length must be right
+        msg = "Weights and axis to be sampled must be of same length"
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, weights=[0, 1])
+
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, weights=[0.5] * 11)
+
+        with pytest.raises(ValueError, match="Fewer non-zero entries in p than size"):
+            obj.sample(n=4, weights=Series([0, 0, 0.2]))
+
+    def test_sample_negative_weights(self, obj):
+        # Check won't accept negative weights
+        bad_weights = [-0.1] * 10
+        msg = "weight vector many not include negative values"
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, weights=bad_weights)
+
+    def test_sample_inf_weights(self, obj):
+        # Check inf and -inf throw errors:
+
+        weights_with_inf = [0.1] * 10
+        weights_with_inf[0] = np.inf
+        msg = "weight vector may not include `inf` values"
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, weights=weights_with_inf)
+
+        weights_with_ninf = [0.1] * 10
+        weights_with_ninf[0] = -np.inf
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(n=3, weights=weights_with_ninf)
+
+    def test_sample_zero_weights(self, obj):
+        # All zeros raises errors
+
+        zero_weights = [0] * 10
+        with pytest.raises(ValueError, match="Invalid weights: weights sum to zero"):
+            obj.sample(n=3, weights=zero_weights)
+
+    def test_sample_missing_weights(self, obj):
+        # All missing weights
+
+        nan_weights = [np.nan] * 10
+        with pytest.raises(ValueError, match="Invalid weights: weights sum to zero"):
+            obj.sample(n=3, weights=nan_weights)
+
+    def test_sample_none_weights(self, obj):
+        # Check None are also replaced by zeros.
+        weights_with_None = [None] * 10
+        weights_with_None[5] = 0.5
+        tm.assert_equal(
+            obj.sample(n=1, axis=0, weights=weights_with_None), obj.iloc[5:6]
+        )
+
+    @pytest.mark.parametrize(
+        "func_str,arg",
+        [
+            ("np.array", [2, 3, 1, 0]),
+            ("np.random.MT19937", 3),
+            ("np.random.PCG64", 11),
+        ],
+    )
+    def test_sample_random_state(self, func_str, arg, frame_or_series):
+        # GH#32503
+        obj = DataFrame({"col1": range(10, 20), "col2": range(20, 30)})
+        obj = tm.get_obj(obj, frame_or_series)
+        result = obj.sample(n=3, random_state=eval(func_str)(arg))
+        expected = obj.sample(n=3, random_state=com.random_state(eval(func_str)(arg)))
+        tm.assert_equal(result, expected)
+
+    def test_sample_generator(self, frame_or_series):
+        # GH#38100
+        obj = frame_or_series(np.arange(100))
+        rng = np.random.default_rng(2)
+
+        # Consecutive calls should advance the seed
+        result1 = obj.sample(n=50, random_state=rng)
+        result2 = obj.sample(n=50, random_state=rng)
+        assert not (result1.index.values == result2.index.values).all()
+
+        # Matching generator initialization must give same result
+        # Consecutive calls should advance the seed
+        result1 = obj.sample(n=50, random_state=np.random.default_rng(11))
+        result2 = obj.sample(n=50, random_state=np.random.default_rng(11))
+        tm.assert_equal(result1, result2)
+
+    def test_sample_upsampling_without_replacement(self, frame_or_series):
+        # GH#27451
+
+        obj = DataFrame({"A": list("abc")})
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = (
+            "Replace has to be set to `True` when upsampling the population `frac` > 1."
+        )
+        with pytest.raises(ValueError, match=msg):
+            obj.sample(frac=2, replace=False)
+
+
+class TestSampleDataFrame:
+    # Tests which are relevant only for DataFrame, so these are
+    #  as fully parametrized as they can get.
+
+    def test_sample(self):
+        # GH#2419
+        # additional specific object based tests
+
+        # A few dataframe test with degenerate weights.
+        easy_weight_list = [0] * 10
+        easy_weight_list[5] = 1
+
+        df = DataFrame(
+            {
+                "col1": range(10, 20),
+                "col2": range(20, 30),
+                "colString": ["a"] * 10,
+                "easyweights": easy_weight_list,
+            }
+        )
+        sample1 = df.sample(n=1, weights="easyweights")
+        tm.assert_frame_equal(sample1, df.iloc[5:6])
+
+        # Ensure proper error if string given as weight for Series or
+        # DataFrame with axis = 1.
+        ser = Series(range(10))
+        msg = "Strings cannot be passed as weights when sampling from a Series."
+        with pytest.raises(ValueError, match=msg):
+            ser.sample(n=3, weights="weight_column")
+
+        msg = (
+            "Strings can only be passed to weights when sampling from rows on a "
+            "DataFrame"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.sample(n=1, weights="weight_column", axis=1)
+
+        # Check weighting key error
+        with pytest.raises(
+            KeyError, match="'String passed to weights not a valid column'"
+        ):
+            df.sample(n=3, weights="not_a_real_column_name")
+
+        # Check that re-normalizes weights that don't sum to one.
+        weights_less_than_1 = [0] * 10
+        weights_less_than_1[0] = 0.5
+        tm.assert_frame_equal(df.sample(n=1, weights=weights_less_than_1), df.iloc[:1])
+
+        ###
+        # Test axis argument
+        ###
+
+        # Test axis argument
+        df = DataFrame({"col1": range(10), "col2": ["a"] * 10})
+        second_column_weight = [0, 1]
+        tm.assert_frame_equal(
+            df.sample(n=1, axis=1, weights=second_column_weight), df[["col2"]]
+        )
+
+        # Different axis arg types
+        tm.assert_frame_equal(
+            df.sample(n=1, axis="columns", weights=second_column_weight), df[["col2"]]
+        )
+
+        weight = [0] * 10
+        weight[5] = 0.5
+        tm.assert_frame_equal(df.sample(n=1, axis="rows", weights=weight), df.iloc[5:6])
+        tm.assert_frame_equal(
+            df.sample(n=1, axis="index", weights=weight), df.iloc[5:6]
+        )
+
+        # Check out of range axis values
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.sample(n=1, axis=2)
+
+        msg = "No axis named not_a_name for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.sample(n=1, axis="not_a_name")
+
+        ser = Series(range(10))
+        with pytest.raises(ValueError, match="No axis named 1 for object type Series"):
+            ser.sample(n=1, axis=1)
+
+        # Test weight length compared to correct axis
+        msg = "Weights and axis to be sampled must be of same length"
+        with pytest.raises(ValueError, match=msg):
+            df.sample(n=1, axis=1, weights=[0.5] * 10)
+
+    def test_sample_axis1(self):
+        # Check weights with axis = 1
+        easy_weight_list = [0] * 3
+        easy_weight_list[2] = 1
+
+        df = DataFrame(
+            {"col1": range(10, 20), "col2": range(20, 30), "colString": ["a"] * 10}
+        )
+        sample1 = df.sample(n=1, axis=1, weights=easy_weight_list)
+        tm.assert_frame_equal(sample1, df[["colString"]])
+
+        # Test default axes
+        tm.assert_frame_equal(
+            df.sample(n=3, random_state=42), df.sample(n=3, axis=0, random_state=42)
+        )
+
+    def test_sample_aligns_weights_with_frame(self):
+        # Test that function aligns weights with frame
+        df = DataFrame({"col1": [5, 6, 7], "col2": ["a", "b", "c"]}, index=[9, 5, 3])
+        ser = Series([1, 0, 0], index=[3, 5, 9])
+        tm.assert_frame_equal(df.loc[[3]], df.sample(1, weights=ser))
+
+        # Weights have index values to be dropped because not in
+        # sampled DataFrame
+        ser2 = Series([0.001, 0, 10000], index=[3, 5, 10])
+        tm.assert_frame_equal(df.loc[[3]], df.sample(1, weights=ser2))
+
+        # Weights have empty values to be filed with zeros
+        ser3 = Series([0.01, 0], index=[3, 5])
+        tm.assert_frame_equal(df.loc[[3]], df.sample(1, weights=ser3))
+
+        # No overlap in weight and sampled DataFrame indices
+        ser4 = Series([1, 0], index=[1, 2])
+
+        with pytest.raises(ValueError, match="Invalid weights: weights sum to zero"):
+            df.sample(1, weights=ser4)
+
+    def test_sample_is_copy(self):
+        # GH#27357, GH#30784: ensure the result of sample is an actual copy and
+        # doesn't track the parent dataframe
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=["a", "b", "c"]
+        )
+        df2 = df.sample(3)
+
+        with tm.assert_produces_warning(None):
+            df2["d"] = 1
+
+    def test_sample_does_not_modify_weights(self):
+        # GH-42843
+        result = np.array([np.nan, 1, np.nan])
+        expected = result.copy()
+        ser = Series([1, 2, 3])
+
+        # Test numpy array weights won't be modified in place
+        ser.sample(weights=result)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # Test DataFrame column won't be modified in place
+        df = DataFrame({"values": [1, 1, 1], "weights": [1, np.nan, np.nan]})
+        expected = df["weights"].copy()
+
+        df.sample(frac=1.0, replace=True, weights="weights")
+        result = df["weights"]
+        tm.assert_series_equal(result, expected)
+
+    def test_sample_ignore_index(self):
+        # GH 38581
+        df = DataFrame(
+            {"col1": range(10, 20), "col2": range(20, 30), "colString": ["a"] * 10}
+        )
+        result = df.sample(3, ignore_index=True)
+        expected_index = Index(range(3))
+        tm.assert_index_equal(result.index, expected_index, exact=True)
diff --git a/pandas/pandas/tests/frame/methods/test_select_dtypes.py b/pandas/pandas/tests/frame/methods/test_select_dtypes.py
new file mode 100644
index 0000000000000..0354e9df3d168
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_select_dtypes.py
@@ -0,0 +1,485 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import ExtensionDtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import ExtensionArray
+
+
+class DummyDtype(ExtensionDtype):
+    type = int
+
+    def __init__(self, numeric) -> None:
+        self._numeric = numeric
+
+    @property
+    def name(self):
+        return "Dummy"
+
+    @property
+    def _is_numeric(self):
+        return self._numeric
+
+
+class DummyArray(ExtensionArray):
+    def __init__(self, data, dtype) -> None:
+        self.data = data
+        self._dtype = dtype
+
+    def __array__(self, dtype=None, copy=None):
+        return self.data
+
+    @property
+    def dtype(self):
+        return self._dtype
+
+    def __len__(self) -> int:
+        return len(self.data)
+
+    def __getitem__(self, item):
+        pass
+
+    def copy(self):
+        return self
+
+
+class TestSelectDtypes:
+    def test_select_dtypes_include_using_list_like(self, using_infer_string):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.Categorical(list("abc")),
+                "g": pd.date_range("20130101", periods=3),
+                "h": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+                "i": pd.date_range("20130101", periods=3, tz="CET"),
+                "j": pd.period_range("2013-01", periods=3, freq="M"),
+                "k": pd.timedelta_range("1 day", periods=3),
+            }
+        )
+
+        ri = df.select_dtypes(include=[np.number])
+        ei = df[["b", "c", "d", "k"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=[np.number], exclude=["timedelta"])
+        ei = df[["b", "c", "d"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=[np.number, "category"], exclude=["timedelta"])
+        ei = df[["b", "c", "d", "f"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=["datetime"])
+        ei = df[["g"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=["datetime64"])
+        ei = df[["g"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=["datetimetz"])
+        ei = df[["h", "i"]]
+        tm.assert_frame_equal(ri, ei)
+
+        with pytest.raises(NotImplementedError, match=r"^$"):
+            df.select_dtypes(include=["period"])
+
+        if using_infer_string:
+            ri = df.select_dtypes(include=["str"])
+            ei = df[["a"]]
+            tm.assert_frame_equal(ri, ei)
+
+            ri = df.select_dtypes(include=[str])
+            tm.assert_frame_equal(ri, ei)
+
+    def test_select_dtypes_exclude_using_list_like(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+            }
+        )
+        re = df.select_dtypes(exclude=[np.number])
+        ee = df[["a", "e"]]
+        tm.assert_frame_equal(re, ee)
+
+    def test_select_dtypes_exclude_include_using_list_like(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6, dtype="u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        exclude = (np.datetime64,)
+        include = np.bool_, "integer"
+        r = df.select_dtypes(include=include, exclude=exclude)
+        e = df[["b", "c", "e"]]
+        tm.assert_frame_equal(r, e)
+
+        exclude = ("datetime",)
+        include = "bool", "int64", "int32"
+        r = df.select_dtypes(include=include, exclude=exclude)
+        e = df[["b", "e"]]
+        tm.assert_frame_equal(r, e)
+
+    @pytest.mark.parametrize(
+        "include", [(np.bool_, "int"), (np.bool_, "integer"), ("bool", int)]
+    )
+    def test_select_dtypes_exclude_include_int(self, include):
+        # Fix select_dtypes(include='int') for Windows, FYI #36596
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6, dtype="int32"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        exclude = (np.datetime64,)
+        result = df.select_dtypes(include=include, exclude=exclude)
+        expected = df[["b", "c", "e"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_select_dtypes_include_using_scalars(self, using_infer_string):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.Categorical(list("abc")),
+                "g": pd.date_range("20130101", periods=3),
+                "h": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+                "i": pd.date_range("20130101", periods=3, tz="CET"),
+                "j": pd.period_range("2013-01", periods=3, freq="M"),
+                "k": pd.timedelta_range("1 day", periods=3),
+            }
+        )
+
+        ri = df.select_dtypes(include=np.number)
+        ei = df[["b", "c", "d", "k"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include="datetime")
+        ei = df[["g"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include="datetime64")
+        ei = df[["g"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include="category")
+        ei = df[["f"]]
+        tm.assert_frame_equal(ri, ei)
+
+        with pytest.raises(NotImplementedError, match=r"^$"):
+            df.select_dtypes(include="period")
+
+        if using_infer_string:
+            ri = df.select_dtypes(include="str")
+            ei = df[["a"]]
+            tm.assert_frame_equal(ri, ei)
+
+    def test_select_dtypes_exclude_using_scalars(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.Categorical(list("abc")),
+                "g": pd.date_range("20130101", periods=3),
+                "h": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+                "i": pd.date_range("20130101", periods=3, tz="CET"),
+                "j": pd.period_range("2013-01", periods=3, freq="M"),
+                "k": pd.timedelta_range("1 day", periods=3),
+            }
+        )
+
+        ri = df.select_dtypes(exclude=np.number)
+        ei = df[["a", "e", "f", "g", "h", "i", "j"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(exclude="category")
+        ei = df[["a", "b", "c", "d", "e", "g", "h", "i", "j", "k"]]
+        tm.assert_frame_equal(ri, ei)
+
+        with pytest.raises(NotImplementedError, match=r"^$"):
+            df.select_dtypes(exclude="period")
+
+    def test_select_dtypes_include_exclude_using_scalars(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.Categorical(list("abc")),
+                "g": pd.date_range("20130101", periods=3),
+                "h": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+                "i": pd.date_range("20130101", periods=3, tz="CET"),
+                "j": pd.period_range("2013-01", periods=3, freq="M"),
+                "k": pd.timedelta_range("1 day", periods=3),
+            }
+        )
+
+        ri = df.select_dtypes(include=np.number, exclude="floating")
+        ei = df[["b", "c", "k"]]
+        tm.assert_frame_equal(ri, ei)
+
+    def test_select_dtypes_include_exclude_mixed_scalars_lists(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.Categorical(list("abc")),
+                "g": pd.date_range("20130101", periods=3),
+                "h": pd.date_range("20130101", periods=3, tz="US/Eastern"),
+                "i": pd.date_range("20130101", periods=3, tz="CET"),
+                "j": pd.period_range("2013-01", periods=3, freq="M"),
+                "k": pd.timedelta_range("1 day", periods=3),
+            }
+        )
+
+        ri = df.select_dtypes(include=np.number, exclude=["floating", "timedelta"])
+        ei = df[["b", "c"]]
+        tm.assert_frame_equal(ri, ei)
+
+        ri = df.select_dtypes(include=[np.number, "category"], exclude="floating")
+        ei = df[["b", "c", "f", "k"]]
+        tm.assert_frame_equal(ri, ei)
+
+    def test_select_dtypes_duplicate_columns(self):
+        # GH20839
+        df = DataFrame(
+            {
+                "a": ["a", "b", "c"],
+                "b": [1, 2, 3],
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        df.columns = ["a", "a", "b", "b", "b", "c"]
+
+        expected = DataFrame(
+            {"a": list(range(1, 4)), "b": np.arange(3, 6).astype("u1")}
+        )
+
+        result = df.select_dtypes(include=[np.number], exclude=["floating"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_select_dtypes_not_an_attr_but_still_valid_dtype(self, using_infer_string):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        df["g"] = df.f.diff()
+        assert not hasattr(np, "u8")
+        r = df.select_dtypes(include=["i8", "O"], exclude=["timedelta"])
+        if using_infer_string:
+            e = df[["b"]]
+        else:
+            e = df[["a", "b"]]
+        tm.assert_frame_equal(r, e)
+
+        r = df.select_dtypes(include=["i8", "O", "timedelta64[ns]"])
+        if using_infer_string:
+            e = df[["b", "g"]]
+        else:
+            e = df[["a", "b", "g"]]
+        tm.assert_frame_equal(r, e)
+
+    def test_select_dtypes_empty(self):
+        df = DataFrame({"a": list("abc"), "b": list(range(1, 4))})
+        msg = "at least one of include or exclude must be nonempty"
+        with pytest.raises(ValueError, match=msg):
+            df.select_dtypes()
+
+    def test_select_dtypes_bad_datetime64(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        with pytest.raises(ValueError, match=".+ is too specific"):
+            df.select_dtypes(include=["datetime64[D]"])
+
+        with pytest.raises(ValueError, match=".+ is too specific"):
+            df.select_dtypes(exclude=["datetime64[as]"])
+
+    def test_select_dtypes_datetime_with_tz(self):
+        df2 = DataFrame(
+            {
+                "A": Timestamp("20130102", tz="US/Eastern"),
+                "B": Timestamp("20130603", tz="CET"),
+            },
+            index=range(5),
+        )
+        df3 = pd.concat([df2.A.to_frame(), df2.B.to_frame()], axis=1)
+        result = df3.select_dtypes(include=["datetime64[ns]"])
+        expected = df3.reindex(columns=[])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [str, "str", np.bytes_, "S1", np.str_, "U1"])
+    @pytest.mark.parametrize("arg", ["include", "exclude"])
+    def test_select_dtypes_str_raises(self, dtype, arg, using_infer_string):
+        if using_infer_string and (dtype == "str" or dtype is str):
+            # this is tested below
+            pytest.skip("Selecting string columns works with future strings")
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "g": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+        msg = "string dtypes are not allowed"
+        kwargs = {arg: [dtype]}
+
+        with pytest.raises(TypeError, match=msg):
+            df.select_dtypes(**kwargs)
+
+    def test_select_dtypes_bad_arg_raises(self):
+        df = DataFrame(
+            {
+                "a": list("abc"),
+                "g": list("abc"),
+                "b": list(range(1, 4)),
+                "c": np.arange(3, 6).astype("u1"),
+                "d": np.arange(4.0, 7.0, dtype="float64"),
+                "e": [True, False, True],
+                "f": pd.date_range("now", periods=3).values,
+            }
+        )
+
+        msg = "data type.*not understood"
+        with pytest.raises(TypeError, match=msg):
+            df.select_dtypes(["blargy, blarg, blarg"])
+
+    def test_select_dtypes_typecodes(self):
+        # GH 11990
+        df = DataFrame(np.random.default_rng(2).random((5, 3)))
+        FLOAT_TYPES = list(np.typecodes["AllFloat"])
+        tm.assert_frame_equal(df.select_dtypes(FLOAT_TYPES), df)
+
+    @pytest.mark.parametrize(
+        "arr,expected",
+        (
+            (np.array([1, 2], dtype=np.int32), True),
+            (pd.array([1, 2], dtype="Int32"), True),
+            (DummyArray([1, 2], dtype=DummyDtype(numeric=True)), True),
+            (DummyArray([1, 2], dtype=DummyDtype(numeric=False)), False),
+        ),
+    )
+    def test_select_dtypes_numeric(self, arr, expected):
+        # GH 35340
+
+        df = DataFrame(arr)
+        is_selected = df.select_dtypes(np.number).shape == df.shape
+        assert is_selected == expected
+
+    def test_select_dtypes_numeric_nullable_string(self, nullable_string_dtype):
+        arr = pd.array(["a", "b"], dtype=nullable_string_dtype)
+        df = DataFrame(arr)
+        is_selected = df.select_dtypes(np.number).shape == df.shape
+        assert not is_selected
+
+    @pytest.mark.parametrize(
+        "expected, float_dtypes",
+        [
+            [
+                DataFrame(
+                    {"A": range(3), "B": range(5, 8), "C": range(10, 7, -1)}
+                ).astype(dtype={"A": float, "B": np.float64, "C": np.float32}),
+                float,
+            ],
+            [
+                DataFrame(
+                    {"A": range(3), "B": range(5, 8), "C": range(10, 7, -1)}
+                ).astype(dtype={"A": float, "B": np.float64, "C": np.float32}),
+                "float",
+            ],
+            [DataFrame({"C": range(10, 7, -1)}, dtype=np.float32), np.float32],
+            [
+                DataFrame({"A": range(3), "B": range(5, 8)}).astype(
+                    dtype={"A": float, "B": np.float64}
+                ),
+                np.float64,
+            ],
+        ],
+    )
+    def test_select_dtypes_float_dtype(self, expected, float_dtypes):
+        # GH#42452
+        dtype_dict = {"A": float, "B": np.float64, "C": np.float32}
+        df = DataFrame(
+            {"A": range(3), "B": range(5, 8), "C": range(10, 7, -1)},
+        )
+        df = df.astype(dtype_dict)
+        result = df.select_dtypes(include=float_dtypes)
+        tm.assert_frame_equal(result, expected)
+
+    def test_np_bool_ea_boolean_include_number(self):
+        # GH 46870
+        df = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": pd.Series([True, False, True], dtype="boolean"),
+                "c": np.array([True, False, True]),
+                "d": pd.Categorical([True, False, True]),
+                "e": pd.arrays.SparseArray([True, False, True]),
+            }
+        )
+        result = df.select_dtypes(include="number")
+        expected = DataFrame({"a": [1, 2, 3]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_select_dtypes_no_view(self):
+        # https://github.com/pandas-dev/pandas/issues/48090
+        # result of this method is not a view on the original dataframe
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df_orig = df.copy()
+        result = df.select_dtypes(include=["number"])
+        result.iloc[0, 0] = 0
+        tm.assert_frame_equal(df, df_orig)
diff --git a/pandas/pandas/tests/frame/methods/test_set_axis.py b/pandas/pandas/tests/frame/methods/test_set_axis.py
new file mode 100644
index 0000000000000..7b75bcf4f348d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_set_axis.py
@@ -0,0 +1,120 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class SharedSetAxisTests:
+    @pytest.fixture
+    def obj(self):
+        raise NotImplementedError("Implemented by subclasses")
+
+    def test_set_axis(self, obj):
+        # GH14636; this tests setting index for both Series and DataFrame
+        new_index = list("abcd")[: len(obj)]
+        expected = obj.copy()
+        expected.index = new_index
+        result = obj.set_axis(new_index, axis=0)
+        tm.assert_equal(expected, result)
+
+    def test_set_axis_copy(self, obj):
+        # Test copy keyword GH#47932
+        new_index = list("abcd")[: len(obj)]
+
+        orig = obj.iloc[:]
+        expected = obj.copy()
+        expected.index = new_index
+
+        result = obj.set_axis(new_index, axis=0)
+        tm.assert_equal(expected, result)
+        assert result is not obj
+        # check we did NOT make a copy
+        if obj.ndim == 1:
+            assert tm.shares_memory(result, obj)
+        else:
+            assert all(
+                tm.shares_memory(result.iloc[:, i], obj.iloc[:, i])
+                for i in range(obj.shape[1])
+            )
+
+        result = obj.set_axis(new_index, axis=0)
+        tm.assert_equal(expected, result)
+        assert result is not obj
+        # check we DID NOT make a copy
+        if obj.ndim == 1:
+            assert tm.shares_memory(result, obj)
+        else:
+            assert any(
+                tm.shares_memory(result.iloc[:, i], obj.iloc[:, i])
+                for i in range(obj.shape[1])
+            )
+
+        res = obj.set_axis(new_index)
+        tm.assert_equal(expected, res)
+        # check we did NOT make a copy
+        if res.ndim == 1:
+            assert tm.shares_memory(res, orig)
+        else:
+            assert all(
+                tm.shares_memory(res.iloc[:, i], orig.iloc[:, i])
+                for i in range(res.shape[1])
+            )
+
+    def test_set_axis_unnamed_kwarg_warns(self, obj):
+        # omitting the "axis" parameter
+        new_index = list("abcd")[: len(obj)]
+
+        expected = obj.copy()
+        expected.index = new_index
+
+        result = obj.set_axis(new_index)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("axis", [3, "foo"])
+    def test_set_axis_invalid_axis_name(self, axis, obj):
+        # wrong values for the "axis" parameter
+        with pytest.raises(ValueError, match="No axis named"):
+            obj.set_axis(list("abc"), axis=axis)
+
+    def test_set_axis_setattr_index_not_collection(self, obj):
+        # wrong type
+        msg = (
+            r"Index\(\.\.\.\) must be called with a collection of some "
+            r"kind, None was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            obj.index = None
+
+    def test_set_axis_setattr_index_wrong_length(self, obj):
+        # wrong length
+        msg = (
+            f"Length mismatch: Expected axis has {len(obj)} elements, "
+            f"new values have {len(obj) - 1} elements"
+        )
+        with pytest.raises(ValueError, match=msg):
+            obj.index = np.arange(len(obj) - 1)
+
+        if obj.ndim == 2:
+            with pytest.raises(ValueError, match="Length mismatch"):
+                obj.columns = obj.columns[::2]
+
+
+class TestDataFrameSetAxis(SharedSetAxisTests):
+    @pytest.fixture
+    def obj(self):
+        df = DataFrame(
+            {"A": [1.1, 2.2, 3.3], "B": [5.0, 6.1, 7.2], "C": [4.4, 5.5, 6.6]},
+            index=[2010, 2011, 2012],
+        )
+        return df
+
+
+class TestSeriesSetAxis(SharedSetAxisTests):
+    @pytest.fixture
+    def obj(self):
+        ser = Series(np.arange(4), index=[1, 3, 5, 7], dtype="int64")
+        return ser
diff --git a/pandas/pandas/tests/frame/methods/test_set_index.py b/pandas/pandas/tests/frame/methods/test_set_index.py
new file mode 100644
index 0000000000000..198cab0e91eab
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_set_index.py
@@ -0,0 +1,734 @@
+"""
+See also: test_reindex.py:TestReindexSetIndex
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def frame_of_index_cols():
+    """
+    Fixture for DataFrame of columns that can be used for indexing
+
+    Columns are ['A', 'B', 'C', 'D', 'E', ('tuple', 'as', 'label')];
+    'A' & 'B' contain duplicates (but are jointly unique), the rest are unique.
+
+         A      B  C         D         E  (tuple, as, label)
+    0  foo    one  a  0.608477 -0.012500           -1.664297
+    1  foo    two  b -0.633460  0.249614           -0.364411
+    2  foo  three  c  0.615256  2.154968           -0.834666
+    3  bar    one  d  0.234246  1.085675            0.718445
+    4  bar    two  e  0.533841 -0.005702           -3.533912
+    """
+    df = DataFrame(
+        {
+            "A": ["foo", "foo", "foo", "bar", "bar"],
+            "B": ["one", "two", "three", "one", "two"],
+            "C": ["a", "b", "c", "d", "e"],
+            "D": np.random.default_rng(2).standard_normal(5),
+            "E": np.random.default_rng(2).standard_normal(5),
+            ("tuple", "as", "label"): np.random.default_rng(2).standard_normal(5),
+        }
+    )
+    return df
+
+
+class TestSetIndex:
+    def test_set_index_multiindex(self):
+        # segfault in GH#3308
+        d = {"t1": [2, 2.5, 3], "t2": [4, 5, 6]}
+        df = DataFrame(d)
+        tuples = [(0, 1), (0, 2), (1, 2)]
+        df["tuples"] = tuples
+
+        index = MultiIndex.from_tuples(df["tuples"])
+        # it works!
+        df.set_index(index)
+
+    def test_set_index_empty_column(self):
+        # GH#1971
+        df = DataFrame(
+            [
+                {"a": 1, "p": 0},
+                {"a": 2, "m": 10},
+                {"a": 3, "m": 11, "p": 20},
+                {"a": 4, "m": 12, "p": 21},
+            ],
+            columns=["a", "m", "p", "x"],
+        )
+
+        result = df.set_index(["a", "x"])
+
+        expected = df[["m", "p"]]
+        expected.index = MultiIndex.from_arrays([df["a"], df["x"]], names=["a", "x"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_set_index_empty_dataframe(self):
+        # GH#38419
+        df1 = DataFrame(
+            {"a": Series(dtype="datetime64[ns]"), "b": Series(dtype="int64"), "c": []}
+        )
+
+        df2 = df1.set_index(["a", "b"])
+        result = df2.index.to_frame().dtypes
+        expected = df1[["a", "b"]].dtypes
+        tm.assert_series_equal(result, expected)
+
+    def test_set_index_multiindexcolumns(self):
+        columns = MultiIndex.from_tuples([("foo", 1), ("foo", 2), ("bar", 1)])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)), columns=columns
+        )
+
+        result = df.set_index(df.columns[0])
+
+        expected = df.iloc[:, 1:]
+        expected.index = df.iloc[:, 0].values
+        expected.index.names = [df.columns[0]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_set_index_timezone(self):
+        # GH#12358
+        # tz-aware Series should retain the tz
+        idx = DatetimeIndex(["2014-01-01 10:10:10"], tz="UTC").tz_convert("Europe/Rome")
+        df = DataFrame({"A": idx})
+        assert df.set_index(idx).index[0].hour == 11
+        assert DatetimeIndex(Series(df.A))[0].hour == 11
+        assert df.set_index(df.A).index[0].hour == 11
+
+    def test_set_index_cast_datetimeindex(self):
+        df = DataFrame(
+            {
+                "A": [datetime(2000, 1, 1) + timedelta(i) for i in range(1000)],
+                "B": np.random.default_rng(2).standard_normal(1000),
+            }
+        )
+
+        idf = df.set_index("A")
+        assert isinstance(idf.index, DatetimeIndex)
+
+    def test_set_index_dst(self):
+        di = date_range("2006-10-29 00:00:00", periods=3, freq="h", tz="US/Pacific")
+
+        df = DataFrame(data={"a": [0, 1, 2], "b": [3, 4, 5]}, index=di).reset_index()
+        # single level
+        res = df.set_index("index")
+        exp = DataFrame(
+            data={"a": [0, 1, 2], "b": [3, 4, 5]},
+            index=Index(di, name="index"),
+        )
+        exp.index = exp.index._with_freq(None)
+        tm.assert_frame_equal(res, exp)
+
+        # GH#12920
+        res = df.set_index(["index", "a"])
+        exp_index = MultiIndex.from_arrays([di, [0, 1, 2]], names=["index", "a"])
+        exp = DataFrame({"b": [3, 4, 5]}, index=exp_index)
+        tm.assert_frame_equal(res, exp)
+
+    def test_set_index(self, float_string_frame):
+        df = float_string_frame
+        idx = Index(np.arange(len(df) - 1, -1, -1, dtype=np.int64))
+
+        df = df.set_index(idx)
+        tm.assert_index_equal(df.index, idx)
+        with pytest.raises(ValueError, match="Length mismatch"):
+            df.set_index(idx[::2])
+
+    def test_set_index_names(self):
+        df = DataFrame(
+            np.ones((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(10)], dtype=object),
+        )
+        df.index.name = "name"
+
+        assert df.set_index(df.index).index.names == ["name"]
+
+        mi = MultiIndex.from_arrays(df[["A", "B"]].T.values, names=["A", "B"])
+        mi2 = MultiIndex.from_arrays(
+            df[["A", "B", "A", "B"]].T.values, names=["A", "B", "C", "D"]
+        )
+
+        df = df.set_index(["A", "B"])
+
+        assert df.set_index(df.index).index.names == ["A", "B"]
+
+        # Check that set_index isn't converting a MultiIndex into an Index
+        assert isinstance(df.set_index(df.index).index, MultiIndex)
+
+        # Check actual equality
+        tm.assert_index_equal(df.set_index(df.index).index, mi)
+
+        idx2 = df.index.rename(["C", "D"])
+
+        # Check that [MultiIndex, MultiIndex] yields a MultiIndex rather
+        # than a pair of tuples
+        assert isinstance(df.set_index([df.index, idx2]).index, MultiIndex)
+
+        # Check equality
+        tm.assert_index_equal(df.set_index([df.index, idx2]).index, mi2)
+
+    # A has duplicate values, C does not
+    @pytest.mark.parametrize("keys", ["A", "C", ["A", "B"], ("tuple", "as", "label")])
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_drop_inplace(self, frame_of_index_cols, drop, inplace, keys):
+        df = frame_of_index_cols
+
+        if isinstance(keys, list):
+            idx = MultiIndex.from_arrays([df[x] for x in keys], names=keys)
+        else:
+            idx = Index(df[keys], name=keys)
+        expected = df.drop(keys, axis=1) if drop else df
+        expected.index = idx
+
+        if inplace:
+            result = df.copy()
+            return_value = result.set_index(keys, drop=drop, inplace=True)
+            assert return_value is None
+        else:
+            result = df.set_index(keys, drop=drop)
+
+        tm.assert_frame_equal(result, expected)
+
+    # A has duplicate values, C does not
+    @pytest.mark.parametrize("keys", ["A", "C", ["A", "B"], ("tuple", "as", "label")])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_append(self, frame_of_index_cols, drop, keys):
+        df = frame_of_index_cols
+
+        keys = keys if isinstance(keys, list) else [keys]
+        idx = MultiIndex.from_arrays(
+            [df.index] + [df[x] for x in keys], names=[None] + keys
+        )
+        expected = df.drop(keys, axis=1) if drop else df.copy()
+        expected.index = idx
+
+        result = df.set_index(keys, drop=drop, append=True)
+
+        tm.assert_frame_equal(result, expected)
+
+    # A has duplicate values, C does not
+    @pytest.mark.parametrize("keys", ["A", "C", ["A", "B"], ("tuple", "as", "label")])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_append_to_multiindex(self, frame_of_index_cols, drop, keys):
+        # append to existing multiindex
+        df = frame_of_index_cols.set_index(["D"], drop=drop, append=True)
+
+        keys = keys if isinstance(keys, list) else [keys]
+        expected = frame_of_index_cols.set_index(["D"] + keys, drop=drop, append=True)
+
+        result = df.set_index(keys, drop=drop, append=True)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_set_index_after_mutation(self):
+        # GH#1590
+        df = DataFrame({"val": [0, 1, 2], "key": ["a", "b", "c"]})
+        expected = DataFrame({"val": [1, 2]}, Index(["b", "c"], name="key"))
+
+        df2 = df.loc[df.index.map(lambda indx: indx >= 1)]
+        result = df2.set_index("key")
+        tm.assert_frame_equal(result, expected)
+
+    # MultiIndex constructor does not work directly on Series -> lambda
+    # Add list-of-list constructor because list is ambiguous -> lambda
+    # also test index name if append=True (name is duplicate here for B)
+    @pytest.mark.parametrize(
+        "box",
+        [
+            Series,
+            Index,
+            np.array,
+            list,
+            lambda x: [list(x)],
+            lambda x: MultiIndex.from_arrays([x]),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "append, index_name", [(True, None), (True, "B"), (True, "test"), (False, None)]
+    )
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_pass_single_array(
+        self, frame_of_index_cols, drop, append, index_name, box
+    ):
+        df = frame_of_index_cols
+        df.index.name = index_name
+
+        key = box(df["B"])
+        if box == list:
+            # list of strings gets interpreted as list of keys
+            msg = "['one', 'two', 'three', 'one', 'two']"
+            with pytest.raises(KeyError, match=msg):
+                df.set_index(key, drop=drop, append=append)
+        else:
+            # np.array/list-of-list "forget" the name of B
+            name_mi = getattr(key, "names", None)
+            name = [getattr(key, "name", None)] if name_mi is None else name_mi
+
+            result = df.set_index(key, drop=drop, append=append)
+
+            # only valid column keys are dropped
+            # since B is always passed as array above, nothing is dropped
+            expected = df.set_index(["B"], drop=False, append=append)
+            expected.index.names = [index_name] + name if append else name
+
+            tm.assert_frame_equal(result, expected)
+
+    # MultiIndex constructor does not work directly on Series -> lambda
+    # also test index name if append=True (name is duplicate here for A & B)
+    @pytest.mark.parametrize(
+        "box", [Series, Index, np.array, list, lambda x: MultiIndex.from_arrays([x])]
+    )
+    @pytest.mark.parametrize(
+        "append, index_name",
+        [(True, None), (True, "A"), (True, "B"), (True, "test"), (False, None)],
+    )
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_pass_arrays(
+        self, frame_of_index_cols, drop, append, index_name, box
+    ):
+        df = frame_of_index_cols
+        df.index.name = index_name
+
+        keys = ["A", box(df["B"])]
+        # np.array/list "forget" the name of B
+        names = ["A", None if box in [np.array, list, tuple, iter] else "B"]
+
+        result = df.set_index(keys, drop=drop, append=append)
+
+        # only valid column keys are dropped
+        # since B is always passed as array above, only A is dropped, if at all
+        expected = df.set_index(["A", "B"], drop=False, append=append)
+        expected = expected.drop("A", axis=1) if drop else expected
+        expected.index.names = [index_name] + names if append else names
+
+        tm.assert_frame_equal(result, expected)
+
+    # MultiIndex constructor does not work directly on Series -> lambda
+    # We also emulate a "constructor" for the label -> lambda
+    # also test index name if append=True (name is duplicate here for A)
+    @pytest.mark.parametrize(
+        "box2",
+        [
+            Series,
+            Index,
+            np.array,
+            list,
+            iter,
+            lambda x: MultiIndex.from_arrays([x]),
+            lambda x: x.name,
+        ],
+    )
+    @pytest.mark.parametrize(
+        "box1",
+        [
+            Series,
+            Index,
+            np.array,
+            list,
+            iter,
+            lambda x: MultiIndex.from_arrays([x]),
+            lambda x: x.name,
+        ],
+    )
+    @pytest.mark.parametrize(
+        "append, index_name", [(True, None), (True, "A"), (True, "test"), (False, None)]
+    )
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_pass_arrays_duplicate(
+        self, frame_of_index_cols, drop, append, index_name, box1, box2
+    ):
+        df = frame_of_index_cols
+        df.index.name = index_name
+
+        keys = [box1(df["A"]), box2(df["A"])]
+        result = df.set_index(keys, drop=drop, append=append)
+
+        # if either box is iter, it has been consumed; re-read
+        keys = [box1(df["A"]), box2(df["A"])]
+
+        # need to adapt first drop for case that both keys are 'A' --
+        # cannot drop the same column twice;
+        # plain == would give ambiguous Boolean error for containers
+        first_drop = (
+            False
+            if (
+                isinstance(keys[0], str)
+                and keys[0] == "A"
+                and isinstance(keys[1], str)
+                and keys[1] == "A"
+            )
+            else drop
+        )
+        # to test against already-tested behaviour, we add sequentially,
+        # hence second append always True; must wrap keys in list, otherwise
+        # box = list would be interpreted as keys
+        expected = df.set_index([keys[0]], drop=first_drop, append=append)
+        expected = expected.set_index([keys[1]], drop=drop, append=True)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("append", [True, False])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_pass_multiindex(self, frame_of_index_cols, drop, append):
+        df = frame_of_index_cols
+        keys = MultiIndex.from_arrays([df["A"], df["B"]], names=["A", "B"])
+
+        result = df.set_index(keys, drop=drop, append=append)
+
+        # setting with a MultiIndex will never drop columns
+        expected = df.set_index(["A", "B"], drop=False, append=append)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_construction_with_categorical_index(self):
+        ci = CategoricalIndex(list("ab") * 5, name="B")
+
+        # with Categorical
+        df = DataFrame(
+            {"A": np.random.default_rng(2).standard_normal(10), "B": ci.values}
+        )
+        idf = df.set_index("B")
+        tm.assert_index_equal(idf.index, ci)
+
+        # from a CategoricalIndex
+        df = DataFrame({"A": np.random.default_rng(2).standard_normal(10), "B": ci})
+        idf = df.set_index("B")
+        tm.assert_index_equal(idf.index, ci)
+
+        # round-trip
+        idf = idf.reset_index().set_index("B")
+        tm.assert_index_equal(idf.index, ci)
+
+    def test_set_index_preserve_categorical_dtype(self):
+        # GH#13743, GH#13854
+        df = DataFrame(
+            {
+                "A": [1, 2, 1, 1, 2],
+                "B": [10, 16, 22, 28, 34],
+                "C1": Categorical(list("abaab"), categories=list("bac"), ordered=False),
+                "C2": Categorical(list("abaab"), categories=list("bac"), ordered=True),
+            }
+        )
+        for cols in ["C1", "C2", ["A", "C1"], ["A", "C2"], ["C1", "C2"]]:
+            result = df.set_index(cols).reset_index()
+            result = result.reindex(columns=df.columns)
+            tm.assert_frame_equal(result, df)
+
+    def test_set_index_datetime(self):
+        # GH#3950
+        df = DataFrame(
+            {
+                "label": ["a", "a", "a", "b", "b", "b"],
+                "datetime": [
+                    "2011-07-19 07:00:00",
+                    "2011-07-19 08:00:00",
+                    "2011-07-19 09:00:00",
+                    "2011-07-19 07:00:00",
+                    "2011-07-19 08:00:00",
+                    "2011-07-19 09:00:00",
+                ],
+                "value": range(6),
+            }
+        )
+        df.index = to_datetime(df.pop("datetime"), utc=True)
+        df.index = df.index.tz_convert("US/Pacific")
+
+        expected = DatetimeIndex(
+            ["2011-07-19 07:00:00", "2011-07-19 08:00:00", "2011-07-19 09:00:00"],
+            name="datetime",
+        )
+        expected = expected.tz_localize("UTC").tz_convert("US/Pacific")
+
+        df = df.set_index("label", append=True)
+        tm.assert_index_equal(df.index.levels[0], expected)
+        tm.assert_index_equal(df.index.levels[1], Index(["a", "b"], name="label"))
+        assert df.index.names == ["datetime", "label"]
+
+        df = df.swaplevel(0, 1)
+        tm.assert_index_equal(df.index.levels[0], Index(["a", "b"], name="label"))
+        tm.assert_index_equal(df.index.levels[1], expected)
+        assert df.index.names == ["label", "datetime"]
+
+        df = DataFrame(np.random.default_rng(2).random(6))
+        idx1 = DatetimeIndex(
+            [
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+            ],
+            tz="US/Eastern",
+        )
+        idx2 = DatetimeIndex(
+            [
+                "2012-04-01 09:00",
+                "2012-04-01 09:00",
+                "2012-04-01 09:00",
+                "2012-04-02 09:00",
+                "2012-04-02 09:00",
+                "2012-04-02 09:00",
+            ],
+            tz="US/Eastern",
+        )
+        idx3 = date_range("2011-01-01 09:00", periods=6, tz="Asia/Tokyo")
+        idx3 = idx3._with_freq(None)
+
+        df = df.set_index(idx1)
+        df = df.set_index(idx2, append=True)
+        df = df.set_index(idx3, append=True)
+
+        expected1 = DatetimeIndex(
+            ["2011-07-19 07:00:00", "2011-07-19 08:00:00", "2011-07-19 09:00:00"],
+            tz="US/Eastern",
+        )
+        expected2 = DatetimeIndex(
+            ["2012-04-01 09:00", "2012-04-02 09:00"], tz="US/Eastern"
+        )
+
+        tm.assert_index_equal(df.index.levels[0], expected1)
+        tm.assert_index_equal(df.index.levels[1], expected2)
+        tm.assert_index_equal(df.index.levels[2], idx3)
+
+        # GH#7092
+        tm.assert_index_equal(df.index.get_level_values(0), idx1)
+        tm.assert_index_equal(df.index.get_level_values(1), idx2)
+        tm.assert_index_equal(df.index.get_level_values(2), idx3)
+
+    def test_set_index_period(self):
+        # GH#6631
+        df = DataFrame(np.random.default_rng(2).random(6))
+        idx1 = period_range("2011-01-01", periods=3, freq="M")
+        idx1 = idx1.append(idx1)
+        idx2 = period_range("2013-01-01 09:00", periods=2, freq="h")
+        idx2 = idx2.append(idx2).append(idx2)
+        idx3 = period_range("2005", periods=6, freq="Y")
+
+        df = df.set_index(idx1)
+        df = df.set_index(idx2, append=True)
+        df = df.set_index(idx3, append=True)
+
+        expected1 = period_range("2011-01-01", periods=3, freq="M")
+        expected2 = period_range("2013-01-01 09:00", periods=2, freq="h")
+
+        tm.assert_index_equal(df.index.levels[0], expected1)
+        tm.assert_index_equal(df.index.levels[1], expected2)
+        tm.assert_index_equal(df.index.levels[2], idx3)
+
+        tm.assert_index_equal(df.index.get_level_values(0), idx1)
+        tm.assert_index_equal(df.index.get_level_values(1), idx2)
+        tm.assert_index_equal(df.index.get_level_values(2), idx3)
+
+
+class TestSetIndexInvalid:
+    def test_set_index_verify_integrity(self, frame_of_index_cols):
+        df = frame_of_index_cols
+
+        with pytest.raises(ValueError, match="Index has duplicate keys"):
+            df.set_index("A", verify_integrity=True)
+        # with MultiIndex
+        with pytest.raises(ValueError, match="Index has duplicate keys"):
+            df.set_index([df["A"], df["A"]], verify_integrity=True)
+
+    @pytest.mark.parametrize("append", [True, False])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
+        df = frame_of_index_cols
+
+        with pytest.raises(KeyError, match="['foo', 'bar', 'baz']"):
+            # column names are A-E, as well as one tuple
+            df.set_index(["foo", "bar", "baz"], drop=drop, append=append)
+
+        # non-existent key in list with arrays
+        with pytest.raises(KeyError, match="X"):
+            df.set_index([df["A"], df["B"], "X"], drop=drop, append=append)
+
+        msg = "[('foo', 'foo', 'foo', 'bar', 'bar')]"
+        # tuples always raise KeyError
+        with pytest.raises(KeyError, match=msg):
+            df.set_index(tuple(df["A"]), drop=drop, append=append)
+
+        # also within a list
+        with pytest.raises(KeyError, match=msg):
+            df.set_index(["A", df["A"], tuple(df["A"])], drop=drop, append=append)
+
+    @pytest.mark.parametrize("append", [True, False])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_raise_on_type(self, frame_of_index_cols, drop, append):
+        box = set
+        df = frame_of_index_cols
+
+        msg = 'The parameter "keys" may be a column key, .*'
+        # forbidden type, e.g. set
+        with pytest.raises(TypeError, match=msg):
+            df.set_index(box(df["A"]), drop=drop, append=append)
+
+        # forbidden type in list, e.g. set
+        with pytest.raises(TypeError, match=msg):
+            df.set_index(["A", df["A"], box(df["A"])], drop=drop, append=append)
+
+    # MultiIndex constructor does not work directly on Series -> lambda
+    @pytest.mark.parametrize(
+        "box",
+        [Series, Index, np.array, iter, lambda x: MultiIndex.from_arrays([x])],
+        ids=["Series", "Index", "np.array", "iter", "MultiIndex"],
+    )
+    @pytest.mark.parametrize("length", [4, 6], ids=["too_short", "too_long"])
+    @pytest.mark.parametrize("append", [True, False])
+    @pytest.mark.parametrize("drop", [True, False])
+    def test_set_index_raise_on_len(
+        self, frame_of_index_cols, box, length, drop, append
+    ):
+        # GH 24984
+        df = frame_of_index_cols  # has length 5
+
+        values = np.random.default_rng(2).integers(0, 10, (length,))
+
+        msg = "Length mismatch: Expected 5 rows, received array of length.*"
+
+        # wrong length directly
+        with pytest.raises(ValueError, match=msg):
+            df.set_index(box(values), drop=drop, append=append)
+
+        # wrong length in list
+        with pytest.raises(ValueError, match=msg):
+            df.set_index(["A", df.A, box(values)], drop=drop, append=append)
+
+
+class TestSetIndexCustomLabelType:
+    def test_set_index_custom_label_type(self):
+        # GH#24969
+
+        class Thing:
+            def __init__(self, name, color) -> None:
+                self.name = name
+                self.color = color
+
+            def __str__(self) -> str:
+                return f"<Thing {self.name!r}>"
+
+            # necessary for pretty KeyError
+            __repr__ = __str__
+
+        thing1 = Thing("One", "red")
+        thing2 = Thing("Two", "blue")
+        df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
+        expected = DataFrame({thing1: [0, 1]}, index=Index([2, 3], name=thing2))
+
+        # use custom label directly
+        result = df.set_index(thing2)
+        tm.assert_frame_equal(result, expected)
+
+        # custom label wrapped in list
+        result = df.set_index([thing2])
+        tm.assert_frame_equal(result, expected)
+
+        # missing key
+        thing3 = Thing("Three", "pink")
+        msg = "<Thing 'Three'>"
+        with pytest.raises(KeyError, match=msg):
+            # missing label directly
+            df.set_index(thing3)
+
+        with pytest.raises(KeyError, match=msg):
+            # missing label in list
+            df.set_index([thing3])
+
+    def test_set_index_custom_label_hashable_iterable(self):
+        # GH#24969
+
+        # actual example discussed in GH 24984 was e.g. for shapely.geometry
+        # objects (e.g. a collection of Points) that can be both hashable and
+        # iterable; using frozenset as a stand-in for testing here
+
+        class Thing(frozenset):
+            # need to stabilize repr for KeyError (due to random order in sets)
+            def __repr__(self) -> str:
+                tmp = sorted(self)
+                joined_reprs = ", ".join(map(repr, tmp))
+                # double curly brace prints one brace in format string
+                return f"frozenset({{{joined_reprs}}})"
+
+        thing1 = Thing(["One", "red"])
+        thing2 = Thing(["Two", "blue"])
+        df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
+        expected = DataFrame({thing1: [0, 1]}, index=Index([2, 3], name=thing2))
+
+        # use custom label directly
+        result = df.set_index(thing2)
+        tm.assert_frame_equal(result, expected)
+
+        # custom label wrapped in list
+        result = df.set_index([thing2])
+        tm.assert_frame_equal(result, expected)
+
+        # missing key
+        thing3 = Thing(["Three", "pink"])
+        msg = r"frozenset\(\{'Three', 'pink'\}\)"
+        with pytest.raises(KeyError, match=msg):
+            # missing label directly
+            df.set_index(thing3)
+
+        with pytest.raises(KeyError, match=msg):
+            # missing label in list
+            df.set_index([thing3])
+
+    def test_set_index_custom_label_type_raises(self):
+        # GH#24969
+
+        # purposefully inherit from something unhashable
+        class Thing(set):
+            def __init__(self, name, color) -> None:
+                self.name = name
+                self.color = color
+
+            def __str__(self) -> str:
+                return f"<Thing {self.name!r}>"
+
+        thing1 = Thing("One", "red")
+        thing2 = Thing("Two", "blue")
+        df = DataFrame([[0, 2], [1, 3]], columns=[thing1, thing2])
+
+        msg = 'The parameter "keys" may be a column key, .*'
+
+        with pytest.raises(TypeError, match=msg):
+            # use custom label directly
+            df.set_index(thing2)
+
+        with pytest.raises(TypeError, match=msg):
+            # custom label wrapped in list
+            df.set_index([thing2])
+
+    def test_set_index_periodindex(self):
+        # GH#6631
+        df = DataFrame(np.random.default_rng(2).random(6))
+        idx1 = period_range("2011/01/01", periods=6, freq="M")
+        idx2 = period_range("2013", periods=6, freq="Y")
+
+        df = df.set_index(idx1)
+        tm.assert_index_equal(df.index, idx1)
+        df = df.set_index(idx2)
+        tm.assert_index_equal(df.index, idx2)
diff --git a/pandas/pandas/tests/frame/methods/test_shift.py b/pandas/pandas/tests/frame/methods/test_shift.py
new file mode 100644
index 0000000000000..b52240c208493
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_shift.py
@@ -0,0 +1,768 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameShift:
+    def test_shift_axis1_with_valid_fill_value_one_array(self):
+        # Case with axis=1 that does not go through the "len(arrays)>1" path
+        #  in DataFrame.shift
+        data = np.random.default_rng(2).standard_normal((5, 3))
+        df = DataFrame(data)
+        res = df.shift(axis=1, periods=1, fill_value=12345)
+        expected = df.T.shift(periods=1, fill_value=12345).T
+        tm.assert_frame_equal(res, expected)
+
+        # same but with an 1D ExtensionArray backing it
+        df2 = df[[0]].astype("Float64")
+        res2 = df2.shift(axis=1, periods=1, fill_value=12345)
+        expected2 = DataFrame([12345] * 5, dtype="Float64")
+        tm.assert_frame_equal(res2, expected2)
+
+    def test_shift_disallow_freq_and_fill_value(self, frame_or_series):
+        # Can't pass both!
+        obj = frame_or_series(
+            np.random.default_rng(2).standard_normal(5),
+            index=date_range("1/1/2000", periods=5, freq="h"),
+        )
+
+        msg = "Passing a 'freq' together with a 'fill_value'"
+        with pytest.raises(ValueError, match=msg):
+            obj.shift(1, fill_value=1, freq="h")
+
+        if frame_or_series is DataFrame:
+            obj.columns = date_range("1/1/2000", periods=1, freq="h")
+            with pytest.raises(ValueError, match=msg):
+                obj.shift(1, axis=1, fill_value=1, freq="h")
+
+    @pytest.mark.parametrize(
+        "input_data, output_data",
+        [(np.empty(shape=(0,)), []), (np.ones(shape=(2,)), [np.nan, 1.0])],
+    )
+    def test_shift_non_writable_array(self, input_data, output_data, frame_or_series):
+        # GH21049 Verify whether non writable numpy array is shiftable
+        input_data.setflags(write=False)
+
+        result = frame_or_series(input_data).shift(1)
+        if frame_or_series is not Series:
+            # need to explicitly specify columns in the empty case
+            expected = frame_or_series(
+                output_data,
+                index=range(len(output_data)),
+                columns=range(1),
+                dtype="float64",
+            )
+        else:
+            expected = frame_or_series(output_data, dtype="float64")
+
+        tm.assert_equal(result, expected)
+
+    def test_shift_mismatched_freq(self, frame_or_series):
+        ts = frame_or_series(
+            np.random.default_rng(2).standard_normal(5),
+            index=date_range("1/1/2000", periods=5, freq="h"),
+        )
+
+        result = ts.shift(1, freq="5min")
+        exp_index = ts.index.shift(1, freq="5min")
+        tm.assert_index_equal(result.index, exp_index)
+
+        # GH#1063, multiple of same base
+        result = ts.shift(1, freq="4h")
+        exp_index = ts.index + offsets.Hour(4)
+        tm.assert_index_equal(result.index, exp_index)
+
+    @pytest.mark.parametrize(
+        "obj",
+        [
+            Series([np.arange(5)]),
+            date_range("1/1/2011", periods=24, freq="h"),
+            Series(range(5), index=date_range("2017", periods=5)),
+        ],
+    )
+    @pytest.mark.parametrize("shift_size", [0, 1, 2])
+    def test_shift_always_copy(self, obj, shift_size, frame_or_series):
+        # GH#22397
+        if frame_or_series is not Series:
+            obj = obj.to_frame()
+        assert obj.shift(shift_size) is not obj
+
+    def test_shift_object_non_scalar_fill(self):
+        # shift requires scalar fill_value except for object dtype
+        ser = Series(range(3))
+        with pytest.raises(ValueError, match="fill_value must be a scalar"):
+            ser.shift(1, fill_value=[])
+
+        df = ser.to_frame()
+        with pytest.raises(ValueError, match="fill_value must be a scalar"):
+            df.shift(1, fill_value=np.arange(3))
+
+        obj_ser = ser.astype(object)
+        result = obj_ser.shift(1, fill_value={})
+        assert result[0] == {}
+
+        obj_df = obj_ser.to_frame()
+        result = obj_df.shift(1, fill_value={})
+        assert result.iloc[0, 0] == {}
+
+    def test_shift_int(self, datetime_frame, frame_or_series):
+        ts = tm.get_obj(datetime_frame, frame_or_series).astype(int)
+        shifted = ts.shift(1)
+        expected = ts.astype(float).shift(1)
+        tm.assert_equal(shifted, expected)
+
+    @pytest.mark.parametrize("dtype", ["int32", "int64"])
+    def test_shift_32bit_take(self, frame_or_series, dtype):
+        # 32-bit taking
+        # GH#8129
+        index = date_range("2000-01-01", periods=5)
+        arr = np.arange(5, dtype=dtype)
+        s1 = frame_or_series(arr, index=index)
+        p = arr[1]
+        result = s1.shift(periods=p)
+        expected = frame_or_series([np.nan, 0, 1, 2, 3], index=index)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("periods", [1, 2, 3, 4])
+    def test_shift_preserve_freqstr(self, periods, frame_or_series):
+        # GH#21275
+        obj = frame_or_series(
+            range(periods),
+            index=date_range("2016-1-1 00:00:00", periods=periods, freq="h"),
+        )
+
+        result = obj.shift(1, "2h")
+
+        expected = frame_or_series(
+            range(periods),
+            index=date_range("2016-1-1 02:00:00", periods=periods, freq="h"),
+        )
+        tm.assert_equal(result, expected)
+
+    def test_shift_dst(self, frame_or_series):
+        # GH#13926
+        dates = date_range("2016-11-06", freq="h", periods=10, tz="US/Eastern")
+        obj = frame_or_series(dates)
+
+        res = obj.shift(0)
+        tm.assert_equal(res, obj)
+        assert tm.get_dtype(res) == "datetime64[ns, US/Eastern]"
+
+        res = obj.shift(1)
+        exp_vals = [NaT] + dates.astype(object).values.tolist()[:9]
+        exp = frame_or_series(exp_vals)
+        tm.assert_equal(res, exp)
+        assert tm.get_dtype(res) == "datetime64[ns, US/Eastern]"
+
+        res = obj.shift(-2)
+        exp_vals = dates.astype(object).values.tolist()[2:] + [NaT, NaT]
+        exp = frame_or_series(exp_vals)
+        tm.assert_equal(res, exp)
+        assert tm.get_dtype(res) == "datetime64[ns, US/Eastern]"
+
+    @pytest.mark.parametrize("ex", [10, -10, 20, -20])
+    def test_shift_dst_beyond(self, frame_or_series, ex):
+        # GH#13926
+        dates = date_range("2016-11-06", freq="h", periods=10, tz="US/Eastern")
+        obj = frame_or_series(dates)
+        res = obj.shift(ex)
+        exp = frame_or_series([NaT] * 10, dtype="datetime64[ns, US/Eastern]")
+        tm.assert_equal(res, exp)
+        assert tm.get_dtype(res) == "datetime64[ns, US/Eastern]"
+
+    def test_shift_by_zero(self, datetime_frame, frame_or_series):
+        # shift by 0
+        obj = tm.get_obj(datetime_frame, frame_or_series)
+        unshifted = obj.shift(0)
+        tm.assert_equal(unshifted, obj)
+
+    def test_shift(self, datetime_frame):
+        # naive shift
+        ser = datetime_frame["A"]
+
+        shifted = datetime_frame.shift(5)
+        tm.assert_index_equal(shifted.index, datetime_frame.index)
+
+        shifted_ser = ser.shift(5)
+        tm.assert_series_equal(shifted["A"], shifted_ser)
+
+        shifted = datetime_frame.shift(-5)
+        tm.assert_index_equal(shifted.index, datetime_frame.index)
+
+        shifted_ser = ser.shift(-5)
+        tm.assert_series_equal(shifted["A"], shifted_ser)
+
+        unshifted = datetime_frame.shift(5).shift(-5)
+        tm.assert_numpy_array_equal(
+            unshifted.dropna().values, datetime_frame.values[:-5]
+        )
+
+        unshifted_ser = ser.shift(5).shift(-5)
+        tm.assert_numpy_array_equal(unshifted_ser.dropna().values, ser.values[:-5])
+
+    def test_shift_by_offset(self, datetime_frame, frame_or_series):
+        # shift by DateOffset
+        obj = tm.get_obj(datetime_frame, frame_or_series)
+        offset = offsets.BDay()
+
+        shifted = obj.shift(5, freq=offset)
+        assert len(shifted) == len(obj)
+        unshifted = shifted.shift(-5, freq=offset)
+        tm.assert_equal(unshifted, obj)
+
+        shifted2 = obj.shift(5, freq="B")
+        tm.assert_equal(shifted, shifted2)
+
+        unshifted = obj.shift(0, freq=offset)
+        tm.assert_equal(unshifted, obj)
+
+        d = obj.index[0]
+        shifted_d = d + offset * 5
+        if frame_or_series is DataFrame:
+            tm.assert_series_equal(obj.xs(d), shifted.xs(shifted_d), check_names=False)
+        else:
+            tm.assert_almost_equal(obj.at[d], shifted.at[shifted_d])
+
+    def test_shift_with_periodindex(self, frame_or_series):
+        # Shifting with PeriodIndex
+        ps = DataFrame(
+            np.arange(4, dtype=float), index=pd.period_range("2020-01-01", periods=4)
+        )
+        ps = tm.get_obj(ps, frame_or_series)
+
+        shifted = ps.shift(1)
+        unshifted = shifted.shift(-1)
+        tm.assert_index_equal(shifted.index, ps.index)
+        tm.assert_index_equal(unshifted.index, ps.index)
+        if frame_or_series is DataFrame:
+            tm.assert_numpy_array_equal(
+                unshifted.iloc[:, 0].dropna().values, ps.iloc[:-1, 0].values
+            )
+        else:
+            tm.assert_numpy_array_equal(unshifted.dropna().values, ps.values[:-1])
+
+        shifted2 = ps.shift(1, "D")
+        shifted3 = ps.shift(1, offsets.Day())
+        tm.assert_equal(shifted2, shifted3)
+        tm.assert_equal(ps, shifted2.shift(-1, "D"))
+
+        msg = "does not match PeriodIndex freq"
+        with pytest.raises(ValueError, match=msg):
+            ps.shift(freq="W")
+
+        # legacy support
+        shifted4 = ps.shift(1, freq="D")
+        tm.assert_equal(shifted2, shifted4)
+
+        shifted5 = ps.shift(1, freq=offsets.Day())
+        tm.assert_equal(shifted5, shifted4)
+
+    def test_shift_other_axis(self):
+        # shift other axis
+        # GH#6371
+        df = DataFrame(np.random.default_rng(2).random((10, 5)))
+        expected = pd.concat(
+            [DataFrame(np.nan, index=df.index, columns=[0]), df.iloc[:, 0:-1]],
+            ignore_index=True,
+            axis=1,
+        )
+        result = df.shift(1, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_shift_named_axis(self):
+        # shift named axis
+        df = DataFrame(np.random.default_rng(2).random((10, 5)))
+        expected = pd.concat(
+            [DataFrame(np.nan, index=df.index, columns=[0]), df.iloc[:, 0:-1]],
+            ignore_index=True,
+            axis=1,
+        )
+        result = df.shift(1, axis="columns")
+        tm.assert_frame_equal(result, expected)
+
+    def test_shift_other_axis_with_freq(self, datetime_frame):
+        obj = datetime_frame.T
+        offset = offsets.BDay()
+
+        # GH#47039
+        shifted = obj.shift(5, freq=offset, axis=1)
+        assert len(shifted) == len(obj)
+        unshifted = shifted.shift(-5, freq=offset, axis=1)
+        tm.assert_equal(unshifted, obj)
+
+    def test_shift_bool(self):
+        df = DataFrame({"high": [True, False], "low": [False, False]})
+        rs = df.shift(1)
+        xp = DataFrame(
+            np.array([[np.nan, np.nan], [True, False]], dtype=object),
+            columns=["high", "low"],
+        )
+        tm.assert_frame_equal(rs, xp)
+
+    def test_shift_categorical1(self, frame_or_series):
+        # GH#9416
+        obj = frame_or_series(["a", "b", "c", "d"], dtype="category")
+
+        rt = obj.shift(1).shift(-1)
+        tm.assert_equal(obj.iloc[:-1], rt.dropna())
+
+        def get_cat_values(ndframe):
+            # For Series we could just do ._values; for DataFrame
+            #  we may be able to do this if we ever have 2D Categoricals
+            return ndframe._mgr.blocks[0].values
+
+        cat = get_cat_values(obj)
+
+        sp1 = obj.shift(1)
+        tm.assert_index_equal(obj.index, sp1.index)
+        assert np.all(get_cat_values(sp1).codes[:1] == -1)
+        assert np.all(cat.codes[:-1] == get_cat_values(sp1).codes[1:])
+
+        sn2 = obj.shift(-2)
+        tm.assert_index_equal(obj.index, sn2.index)
+        assert np.all(get_cat_values(sn2).codes[-2:] == -1)
+        assert np.all(cat.codes[2:] == get_cat_values(sn2).codes[:-2])
+
+        tm.assert_index_equal(cat.categories, get_cat_values(sp1).categories)
+        tm.assert_index_equal(cat.categories, get_cat_values(sn2).categories)
+
+    def test_shift_categorical(self):
+        # GH#9416
+        s1 = Series(["a", "b", "c"], dtype="category")
+        s2 = Series(["A", "B", "C"], dtype="category")
+        df = DataFrame({"one": s1, "two": s2})
+        rs = df.shift(1)
+        xp = DataFrame({"one": s1.shift(1), "two": s2.shift(1)})
+        tm.assert_frame_equal(rs, xp)
+
+    def test_shift_categorical_fill_value(self, frame_or_series):
+        ts = frame_or_series(["a", "b", "c", "d"], dtype="category")
+        res = ts.shift(1, fill_value="a")
+        expected = frame_or_series(
+            pd.Categorical(
+                ["a", "a", "b", "c"], categories=["a", "b", "c", "d"], ordered=False
+            )
+        )
+        tm.assert_equal(res, expected)
+
+        # check for incorrect fill_value
+        msg = r"Cannot setitem on a Categorical with a new category \(f\)"
+        with pytest.raises(TypeError, match=msg):
+            ts.shift(1, fill_value="f")
+
+    def test_shift_fill_value(self, frame_or_series):
+        # GH#24128
+        dti = date_range("1/1/2000", periods=5, freq="h")
+
+        ts = frame_or_series([1.0, 2.0, 3.0, 4.0, 5.0], index=dti)
+        exp = frame_or_series([0.0, 1.0, 2.0, 3.0, 4.0], index=dti)
+        # check that fill value works
+        result = ts.shift(1, fill_value=0.0)
+        tm.assert_equal(result, exp)
+
+        exp = frame_or_series([0.0, 0.0, 1.0, 2.0, 3.0], index=dti)
+        result = ts.shift(2, fill_value=0.0)
+        tm.assert_equal(result, exp)
+
+        ts = frame_or_series([1, 2, 3])
+        res = ts.shift(2, fill_value=0)
+        assert tm.get_dtype(res) == tm.get_dtype(ts)
+
+        # retain integer dtype
+        obj = frame_or_series([1, 2, 3, 4, 5], index=dti)
+        exp = frame_or_series([0, 1, 2, 3, 4], index=dti)
+        result = obj.shift(1, fill_value=0)
+        tm.assert_equal(result, exp)
+
+        exp = frame_or_series([0, 0, 1, 2, 3], index=dti)
+        result = obj.shift(2, fill_value=0)
+        tm.assert_equal(result, exp)
+
+    def test_shift_empty(self):
+        # Regression test for GH#8019
+        df = DataFrame({"foo": []})
+        rs = df.shift(-1)
+
+        tm.assert_frame_equal(df, rs)
+
+    def test_shift_duplicate_columns(self):
+        # GH#9092; verify that position-based shifting works
+        # in the presence of duplicate columns
+        column_lists = [list(range(5)), [1] * 5, [1, 1, 2, 2, 1]]
+        data = np.random.default_rng(2).standard_normal((20, 5))
+
+        shifted = []
+        for columns in column_lists:
+            df = DataFrame(data.copy(), columns=columns)
+            for s in range(5):
+                df.iloc[:, s] = df.iloc[:, s].shift(s + 1)
+            df.columns = range(5)
+            shifted.append(df)
+
+        # sanity check the base case
+        nulls = shifted[0].isna().sum()
+        tm.assert_series_equal(nulls, Series(range(1, 6), dtype="int64"))
+
+        # check all answers are the same
+        tm.assert_frame_equal(shifted[0], shifted[1])
+        tm.assert_frame_equal(shifted[0], shifted[2])
+
+    def test_shift_axis1_multiple_blocks(self):
+        # GH#35488
+        df1 = DataFrame(np.random.default_rng(2).integers(1000, size=(5, 3)))
+        df2 = DataFrame(np.random.default_rng(2).integers(1000, size=(5, 2)))
+        df3 = pd.concat([df1, df2], axis=1)
+        assert len(df3._mgr.blocks) == 2
+
+        result = df3.shift(2, axis=1)
+
+        expected = df3.take([-1, -1, 0, 1, 2], axis=1)
+        # Explicit cast to float to avoid implicit cast when setting nan.
+        # Column names aren't unique, so directly calling `expected.astype` won't work.
+        expected = expected.pipe(
+            lambda df: df.set_axis(range(df.shape[1]), axis=1)
+            .astype({0: "float", 1: "float"})
+            .set_axis(df.columns, axis=1)
+        )
+        expected.iloc[:, :2] = np.nan
+        expected.columns = df3.columns
+
+        tm.assert_frame_equal(result, expected)
+
+        # Case with periods < 0
+        # rebuild df3 because `take` call above consolidated
+        df3 = pd.concat([df1, df2], axis=1)
+        assert len(df3._mgr.blocks) == 2
+        result = df3.shift(-2, axis=1)
+
+        expected = df3.take([2, 3, 4, -1, -1], axis=1)
+        # Explicit cast to float to avoid implicit cast when setting nan.
+        # Column names aren't unique, so directly calling `expected.astype` won't work.
+        expected = expected.pipe(
+            lambda df: df.set_axis(range(df.shape[1]), axis=1)
+            .astype({3: "float", 4: "float"})
+            .set_axis(df.columns, axis=1)
+        )
+        expected.iloc[:, -2:] = np.nan
+        expected.columns = df3.columns
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_shift_axis1_multiple_blocks_with_int_fill(self):
+        # GH#42719
+        rng = np.random.default_rng(2)
+        df1 = DataFrame(rng.integers(1000, size=(5, 3), dtype=int))
+        df2 = DataFrame(rng.integers(1000, size=(5, 2), dtype=int))
+        df3 = pd.concat([df1.iloc[:4, 1:3], df2.iloc[:4, :]], axis=1)
+        result = df3.shift(2, axis=1, fill_value=np.int_(0))
+        assert len(df3._mgr.blocks) == 2
+
+        expected = df3.take([-1, -1, 0, 1], axis=1)
+        expected.iloc[:, :2] = np.int_(0)
+        expected.columns = df3.columns
+
+        tm.assert_frame_equal(result, expected)
+
+        # Case with periods < 0
+        df3 = pd.concat([df1.iloc[:4, 1:3], df2.iloc[:4, :]], axis=1)
+        result = df3.shift(-2, axis=1, fill_value=np.int_(0))
+        assert len(df3._mgr.blocks) == 2
+
+        expected = df3.take([2, 3, -1, -1], axis=1)
+        expected.iloc[:, -2:] = np.int_(0)
+        expected.columns = df3.columns
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_period_index_frame_shift_with_freq(self, frame_or_series):
+        ps = DataFrame(range(4), index=pd.period_range("2020-01-01", periods=4))
+        ps = tm.get_obj(ps, frame_or_series)
+
+        shifted = ps.shift(1, freq="infer")
+        unshifted = shifted.shift(-1, freq="infer")
+        tm.assert_equal(unshifted, ps)
+
+        shifted2 = ps.shift(freq="D")
+        tm.assert_equal(shifted, shifted2)
+
+        shifted3 = ps.shift(freq=offsets.Day())
+        tm.assert_equal(shifted, shifted3)
+
+    def test_datetime_frame_shift_with_freq(self, datetime_frame, frame_or_series):
+        dtobj = tm.get_obj(datetime_frame, frame_or_series)
+        shifted = dtobj.shift(1, freq="infer")
+        unshifted = shifted.shift(-1, freq="infer")
+        tm.assert_equal(dtobj, unshifted)
+
+        shifted2 = dtobj.shift(freq=dtobj.index.freq)
+        tm.assert_equal(shifted, shifted2)
+
+        inferred_ts = DataFrame(
+            datetime_frame.values,
+            Index(np.asarray(datetime_frame.index)),
+            columns=datetime_frame.columns,
+        )
+        inferred_ts = tm.get_obj(inferred_ts, frame_or_series)
+        shifted = inferred_ts.shift(1, freq="infer")
+        expected = dtobj.shift(1, freq="infer")
+        expected.index = expected.index._with_freq(None)
+        tm.assert_equal(shifted, expected)
+
+        unshifted = shifted.shift(-1, freq="infer")
+        tm.assert_equal(unshifted, inferred_ts)
+
+    def test_period_index_frame_shift_with_freq_error(self, frame_or_series):
+        ps = DataFrame(range(4), index=pd.period_range("2020-01-01", periods=4))
+        ps = tm.get_obj(ps, frame_or_series)
+        msg = "Given freq M does not match PeriodIndex freq D"
+        with pytest.raises(ValueError, match=msg):
+            ps.shift(freq="M")
+
+    def test_datetime_frame_shift_with_freq_error(
+        self, datetime_frame, frame_or_series
+    ):
+        dtobj = tm.get_obj(datetime_frame, frame_or_series)
+        no_freq = dtobj.iloc[[0, 5, 7]]
+        msg = "Freq was not set in the index hence cannot be inferred"
+        with pytest.raises(ValueError, match=msg):
+            no_freq.shift(freq="infer")
+
+    def test_shift_dt64values_int_fill_deprecated(self):
+        # GH#31971
+        ser = Series([pd.Timestamp("2020-01-01"), pd.Timestamp("2020-01-02")])
+
+        with pytest.raises(TypeError, match="value should be a"):
+            ser.shift(1, fill_value=0)
+
+        df = ser.to_frame()
+        with pytest.raises(TypeError, match="value should be a"):
+            df.shift(1, fill_value=0)
+
+        # axis = 1
+        df2 = DataFrame({"A": ser, "B": ser})
+        df2._consolidate_inplace()
+
+        result = df2.shift(1, axis=1, fill_value=0)
+        expected = DataFrame({"A": [0, 0], "B": df2["A"]})
+        tm.assert_frame_equal(result, expected)
+
+        # same thing but not consolidated; pre-2.0 we got different behavior
+        df3 = DataFrame({"A": ser})
+        df3["B"] = ser
+        assert len(df3._mgr.blocks) == 2
+        result = df3.shift(1, axis=1, fill_value=0)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "as_cat",
+        [
+            pytest.param(
+                True,
+                marks=pytest.mark.xfail(
+                    reason="_can_hold_element incorrectly always returns True"
+                ),
+            ),
+            False,
+        ],
+    )
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            date_range("2020-01-01", periods=2),
+            date_range("2020-01-01", periods=2, tz="US/Pacific"),
+            pd.period_range("2020-01-01", periods=2, freq="D"),
+            pd.timedelta_range("2020 Days", periods=2, freq="D"),
+            pd.interval_range(0, 3, periods=2),
+            pytest.param(
+                pd.array([1, 2], dtype="Int64"),
+                marks=pytest.mark.xfail(
+                    reason="_can_hold_element incorrectly always returns True"
+                ),
+            ),
+            pytest.param(
+                pd.array([1, 2], dtype="Float32"),
+                marks=pytest.mark.xfail(
+                    reason="_can_hold_element incorrectly always returns True"
+                ),
+            ),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def test_shift_dt64values_axis1_invalid_fill(self, vals, as_cat):
+        # GH#44564
+        ser = Series(vals)
+        if as_cat:
+            ser = ser.astype("category")
+
+        df = DataFrame({"A": ser})
+        result = df.shift(-1, axis=1, fill_value="foo")
+        expected = DataFrame({"A": ["foo", "foo"]})
+        tm.assert_frame_equal(result, expected)
+
+        # same thing but multiple blocks
+        df2 = DataFrame({"A": ser, "B": ser})
+        df2._consolidate_inplace()
+
+        result = df2.shift(-1, axis=1, fill_value="foo")
+        expected = DataFrame({"A": df2["B"], "B": ["foo", "foo"]})
+        tm.assert_frame_equal(result, expected)
+
+        # same thing but not consolidated
+        df3 = DataFrame({"A": ser})
+        df3["B"] = ser
+        assert len(df3._mgr.blocks) == 2
+        result = df3.shift(-1, axis=1, fill_value="foo")
+        tm.assert_frame_equal(result, expected)
+
+    def test_shift_axis1_categorical_columns(self):
+        # GH#38434
+        ci = CategoricalIndex(["a", "b", "c"])
+        df = DataFrame(
+            {"a": [1, 3], "b": [2, 4], "c": [5, 6]}, index=ci[:-1], columns=ci
+        )
+        result = df.shift(axis=1)
+
+        expected = DataFrame(
+            {"a": [np.nan, np.nan], "b": [1, 3], "c": [2, 4]}, index=ci[:-1], columns=ci
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # periods != 1
+        result = df.shift(2, axis=1)
+        expected = DataFrame(
+            {"a": [np.nan, np.nan], "b": [np.nan, np.nan], "c": [1, 3]},
+            index=ci[:-1],
+            columns=ci,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_shift_axis1_many_periods(self):
+        # GH#44978 periods > len(columns)
+        df = DataFrame(np.random.default_rng(2).random((5, 3)))
+        shifted = df.shift(6, axis=1, fill_value=None)
+
+        expected = df * np.nan
+        tm.assert_frame_equal(shifted, expected)
+
+        shifted2 = df.shift(-6, axis=1, fill_value=None)
+        tm.assert_frame_equal(shifted2, expected)
+
+    def test_shift_with_offsets_freq(self):
+        df = DataFrame({"x": [1, 2, 3]}, index=date_range("2000", periods=3))
+        shifted = df.shift(freq="1MS")
+        expected = DataFrame(
+            {"x": [1, 2, 3]},
+            index=date_range(start="02/01/2000", end="02/01/2000", periods=3),
+        )
+        tm.assert_frame_equal(shifted, expected)
+
+    def test_shift_with_iterable_basic_functionality(self):
+        # GH#44424
+        data = {"a": [1, 2, 3], "b": [4, 5, 6]}
+        shifts = [0, 1, 2]
+
+        df = DataFrame(data)
+        shifted = df.shift(shifts)
+
+        expected = DataFrame(
+            {
+                "a_0": [1, 2, 3],
+                "b_0": [4, 5, 6],
+                "a_1": [np.nan, 1.0, 2.0],
+                "b_1": [np.nan, 4.0, 5.0],
+                "a_2": [np.nan, np.nan, 1.0],
+                "b_2": [np.nan, np.nan, 4.0],
+            }
+        )
+        tm.assert_frame_equal(expected, shifted)
+
+    def test_shift_with_iterable_series(self):
+        # GH#44424
+        data = {"a": [1, 2, 3]}
+        shifts = [0, 1, 2]
+
+        df = DataFrame(data)
+        s = df["a"]
+        tm.assert_frame_equal(s.shift(shifts), df.shift(shifts))
+
+    def test_shift_with_iterable_freq_and_fill_value(self):
+        # GH#44424
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(5),
+            index=date_range("1/1/2000", periods=5, freq="h"),
+        )
+
+        tm.assert_frame_equal(
+            # rename because shift with an iterable leads to str column names
+            df.shift([1], fill_value=1).rename(columns=lambda x: int(x[0])),
+            df.shift(1, fill_value=1),
+        )
+
+        tm.assert_frame_equal(
+            df.shift([1], freq="h").rename(columns=lambda x: int(x[0])),
+            df.shift(1, freq="h"),
+        )
+
+    def test_shift_with_iterable_check_other_arguments(self):
+        # GH#44424
+        data = {"a": [1, 2], "b": [4, 5]}
+        shifts = [0, 1]
+        df = DataFrame(data)
+
+        # test suffix
+        shifted = df[["a"]].shift(shifts, suffix="_suffix")
+        expected = DataFrame({"a_suffix_0": [1, 2], "a_suffix_1": [np.nan, 1.0]})
+        tm.assert_frame_equal(shifted, expected)
+
+        # check bad inputs when doing multiple shifts
+        msg = "If `periods` contains multiple shifts, `axis` cannot be 1."
+        with pytest.raises(ValueError, match=msg):
+            df.shift(shifts, axis=1)
+
+        msg = "Periods must be integer, but s is <class 'str'>."
+        with pytest.raises(TypeError, match=msg):
+            df.shift(["s"])
+
+        msg = "If `periods` is an iterable, it cannot be empty."
+        with pytest.raises(ValueError, match=msg):
+            df.shift([])
+
+        msg = "Cannot specify `suffix` if `periods` is an int."
+        with pytest.raises(ValueError, match=msg):
+            df.shift(1, suffix="fails")
+
+    def test_shift_axis_one_empty(self):
+        # GH#57301
+        df = DataFrame()
+        result = df.shift(1, axis=1)
+        tm.assert_frame_equal(result, df)
+
+    def test_shift_with_offsets_freq_empty(self):
+        # GH#60102
+        dates = date_range("2020-01-01", periods=3, freq="D")
+        offset = offsets.Day()
+        shifted_dates = dates + offset
+        df = DataFrame(index=dates)
+        df_shifted = DataFrame(index=shifted_dates)
+        result = df.shift(freq=offset)
+        tm.assert_frame_equal(result, df_shifted)
+
+    def test_series_shift_interval_preserves_closed(self):
+        # GH#60389
+        ser = Series(
+            [pd.Interval(1, 2, closed="right"), pd.Interval(2, 3, closed="right")]
+        )
+        result = ser.shift(1)
+        expected = Series([np.nan, pd.Interval(1, 2, closed="right")])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_size.py b/pandas/pandas/tests/frame/methods/test_size.py
new file mode 100644
index 0000000000000..0c8b6473c85ea
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_size.py
@@ -0,0 +1,21 @@
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+
+
+@pytest.mark.parametrize(
+    "data, index, expected",
+    [
+        ({"col1": [1], "col2": [3]}, None, 2),
+        ({}, None, 0),
+        ({"col1": [1, np.nan], "col2": [3, 4]}, None, 4),
+        ({"col1": [1, 2], "col2": [3, 4]}, [["a", "b"], [1, 2]], 4),
+        ({"col1": [1, 2, 3, 4], "col2": [3, 4, 5, 6]}, ["x", "y", "a", "b"], 8),
+    ],
+)
+def test_size(data, index, expected):
+    # GH#52897
+    df = DataFrame(data, index=index)
+    assert df.size == expected
+    assert isinstance(df.size, int)
diff --git a/pandas/pandas/tests/frame/methods/test_sort_index.py b/pandas/pandas/tests/frame/methods/test_sort_index.py
new file mode 100644
index 0000000000000..1a631e760208a
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_sort_index.py
@@ -0,0 +1,1032 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    IntervalIndex,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameSortIndex:
+    def test_sort_index_and_reconstruction_doc_example(self):
+        # doc example
+        df = DataFrame(
+            {"value": [1, 2, 3, 4]},
+            index=MultiIndex(
+                levels=[["a", "b"], ["bb", "aa"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+            ),
+        )
+        assert df.index._is_lexsorted()
+        assert not df.index.is_monotonic_increasing
+
+        # sort it
+        expected = DataFrame(
+            {"value": [2, 1, 4, 3]},
+            index=MultiIndex(
+                levels=[["a", "b"], ["aa", "bb"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+            ),
+        )
+        result = df.sort_index()
+        assert result.index.is_monotonic_increasing
+        tm.assert_frame_equal(result, expected)
+
+        # reconstruct
+        result = df.sort_index().copy()
+        result.index = result.index._sort_levels_monotonic()
+        assert result.index.is_monotonic_increasing
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_non_existent_label_multiindex(self):
+        # GH#12261
+        df = DataFrame(0, columns=[], index=MultiIndex.from_product([[], []]))
+        with tm.assert_produces_warning(None):
+            df.loc["b", "2"] = 1
+            df.loc["a", "3"] = 1
+        result = df.sort_index().index.is_monotonic_increasing
+        assert result is True
+
+    def test_sort_index_reorder_on_ops(self):
+        # GH#15687
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 2)),
+            index=MultiIndex.from_product(
+                [["a", "b"], ["big", "small"], ["red", "blu"]],
+                names=["letter", "size", "color"],
+            ),
+            columns=["near", "far"],
+        )
+        df = df.sort_index()
+
+        def my_func(group):
+            group.index = ["newz", "newa"]
+            return group
+
+        result = df.groupby(level=["letter", "size"]).apply(my_func).sort_index()
+        expected = MultiIndex.from_product(
+            [["a", "b"], ["big", "small"], ["newa", "newz"]],
+            names=["letter", "size", None],
+        )
+
+        tm.assert_index_equal(result.index, expected)
+
+    def test_sort_index_nan_multiindex(self):
+        # GH#14784
+        # incorrect sorting w.r.t. nans
+        tuples = [[12, 13], [np.nan, np.nan], [np.nan, 3], [1, 2]]
+        mi = MultiIndex.from_tuples(tuples)
+
+        df = DataFrame(np.arange(16).reshape(4, 4), index=mi, columns=list("ABCD"))
+        s = Series(np.arange(4), index=mi)
+
+        df2 = DataFrame(
+            {
+                "date": pd.DatetimeIndex(
+                    [
+                        "20121002",
+                        "20121007",
+                        "20130130",
+                        "20130202",
+                        "20130305",
+                        "20121002",
+                        "20121207",
+                        "20130130",
+                        "20130202",
+                        "20130305",
+                        "20130202",
+                        "20130305",
+                    ]
+                ),
+                "user_id": [1, 1, 1, 1, 1, 3, 3, 3, 5, 5, 5, 5],
+                "whole_cost": [
+                    1790,
+                    np.nan,
+                    280,
+                    259,
+                    np.nan,
+                    623,
+                    90,
+                    312,
+                    np.nan,
+                    301,
+                    359,
+                    801,
+                ],
+                "cost": [12, 15, 10, 24, 39, 1, 0, np.nan, 45, 34, 1, 12],
+            }
+        ).set_index(["date", "user_id"])
+
+        # sorting frame, default nan position is last
+        result = df.sort_index()
+        expected = df.iloc[[3, 0, 2, 1], :]
+        tm.assert_frame_equal(result, expected)
+
+        # sorting frame, nan position last
+        result = df.sort_index(na_position="last")
+        expected = df.iloc[[3, 0, 2, 1], :]
+        tm.assert_frame_equal(result, expected)
+
+        # sorting frame, nan position first
+        result = df.sort_index(na_position="first")
+        expected = df.iloc[[1, 2, 3, 0], :]
+        tm.assert_frame_equal(result, expected)
+
+        # sorting frame with removed rows
+        result = df2.dropna().sort_index()
+        expected = df2.sort_index().dropna()
+        tm.assert_frame_equal(result, expected)
+
+        # sorting series, default nan position is last
+        result = s.sort_index()
+        expected = s.iloc[[3, 0, 2, 1]]
+        tm.assert_series_equal(result, expected)
+
+        # sorting series, nan position last
+        result = s.sort_index(na_position="last")
+        expected = s.iloc[[3, 0, 2, 1]]
+        tm.assert_series_equal(result, expected)
+
+        # sorting series, nan position first
+        result = s.sort_index(na_position="first")
+        expected = s.iloc[[1, 2, 3, 0]]
+        tm.assert_series_equal(result, expected)
+
+    def test_sort_index_nan(self):
+        # GH#3917
+
+        # Test DataFrame with nan label
+        df = DataFrame(
+            {"A": [1, 2, np.nan, 1, 6, 8, 4], "B": [9, np.nan, 5, 2, 5, 4, 5]},
+            index=[1, 2, 3, 4, 5, 6, np.nan],
+        )
+
+        # NaN label, ascending=True, na_position='last'
+        sorted_df = df.sort_index(kind="quicksort", ascending=True, na_position="last")
+        expected = DataFrame(
+            {"A": [1, 2, np.nan, 1, 6, 8, 4], "B": [9, np.nan, 5, 2, 5, 4, 5]},
+            index=[1, 2, 3, 4, 5, 6, np.nan],
+        )
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # NaN label, ascending=True, na_position='first'
+        sorted_df = df.sort_index(na_position="first")
+        expected = DataFrame(
+            {"A": [4, 1, 2, np.nan, 1, 6, 8], "B": [5, 9, np.nan, 5, 2, 5, 4]},
+            index=[np.nan, 1, 2, 3, 4, 5, 6],
+        )
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # NaN label, ascending=False, na_position='last'
+        sorted_df = df.sort_index(kind="quicksort", ascending=False)
+        expected = DataFrame(
+            {"A": [8, 6, 1, np.nan, 2, 1, 4], "B": [4, 5, 2, 5, np.nan, 9, 5]},
+            index=[6, 5, 4, 3, 2, 1, np.nan],
+        )
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # NaN label, ascending=False, na_position='first'
+        sorted_df = df.sort_index(
+            kind="quicksort", ascending=False, na_position="first"
+        )
+        expected = DataFrame(
+            {"A": [4, 8, 6, 1, np.nan, 2, 1], "B": [5, 4, 5, 2, 5, np.nan, 9]},
+            index=[np.nan, 6, 5, 4, 3, 2, 1],
+        )
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_index_multi_index(self):
+        # GH#25775, testing that sorting by index works with a multi-index.
+        df = DataFrame(
+            {"a": [3, 1, 2], "b": [0, 0, 0], "c": [0, 1, 2], "d": list("abc")}
+        )
+        result = df.set_index(list("abc")).sort_index(level=list("ba"))
+
+        expected = DataFrame(
+            {"a": [1, 2, 3], "b": [0, 0, 0], "c": [1, 2, 0], "d": list("bca")}
+        )
+        expected = expected.set_index(list("abc"))
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_inplace(self):
+        frame = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=[1, 2, 3, 4],
+            columns=["A", "B", "C", "D"],
+        )
+
+        # axis=0
+        unordered = frame.loc[[3, 2, 4, 1]]
+        a_values = unordered["A"]
+        df = unordered.copy()
+        return_value = df.sort_index(inplace=True)
+        assert return_value is None
+        expected = frame
+        tm.assert_frame_equal(df, expected)
+        # GH 44153 related
+        # Used to be a_id != id(df["A"]), but flaky in the CI
+        assert a_values is not df["A"]
+
+        df = unordered.copy()
+        return_value = df.sort_index(ascending=False, inplace=True)
+        assert return_value is None
+        expected = frame[::-1]
+        tm.assert_frame_equal(df, expected)
+
+        # axis=1
+        unordered = frame.loc[:, ["D", "B", "C", "A"]]
+        df = unordered.copy()
+        return_value = df.sort_index(axis=1, inplace=True)
+        assert return_value is None
+        expected = frame
+        tm.assert_frame_equal(df, expected)
+
+        df = unordered.copy()
+        return_value = df.sort_index(axis=1, ascending=False, inplace=True)
+        assert return_value is None
+        expected = frame.iloc[:, ::-1]
+        tm.assert_frame_equal(df, expected)
+
+    def test_sort_index_different_sortorder(self):
+        A = np.arange(20).repeat(5)
+        B = np.tile(np.arange(5), 20)
+
+        indexer = np.random.default_rng(2).permutation(100)
+        A = A.take(indexer)
+        B = B.take(indexer)
+
+        df = DataFrame(
+            {"A": A, "B": B, "C": np.random.default_rng(2).standard_normal(100)}
+        )
+
+        ex_indexer = np.lexsort((df.B.max() - df.B, df.A))
+        expected = df.take(ex_indexer)
+
+        # test with multiindex, too
+        idf = df.set_index(["A", "B"])
+
+        result = idf.sort_index(ascending=[1, 0])
+        expected = idf.take(ex_indexer)
+        tm.assert_frame_equal(result, expected)
+
+        # also, Series!
+        result = idf["C"].sort_index(ascending=[1, 0])
+        tm.assert_series_equal(result, expected["C"])
+
+    def test_sort_index_level(self):
+        mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+        df = DataFrame([[1, 2], [3, 4]], mi)
+
+        result = df.sort_index(level="A", sort_remaining=False)
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(level=["A", "B"], sort_remaining=False)
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+        # Error thrown by sort_index when
+        # first index is sorted last (GH#26053)
+        result = df.sort_index(level=["C", "B", "A"])
+        expected = df.iloc[[1, 0]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(level=["B", "C", "A"])
+        expected = df.iloc[[1, 0]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(level=["C", "A"])
+        expected = df.iloc[[1, 0]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_categorical_index(self):
+        df = DataFrame(
+            {
+                "A": np.arange(6, dtype="int64"),
+                "B": Series(list("aabbca")).astype(CategoricalDtype(list("cab"))),
+            }
+        ).set_index("B")
+
+        result = df.sort_index()
+        expected = df.iloc[[4, 0, 1, 5, 2, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(ascending=False)
+        expected = df.iloc[[2, 3, 0, 1, 5, 4]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index(self):
+        # GH#13496
+
+        frame = DataFrame(
+            np.arange(16).reshape(4, 4),
+            index=[1, 2, 3, 4],
+            columns=["A", "B", "C", "D"],
+        )
+
+        # axis=0 : sort rows by index labels
+        unordered = frame.loc[[3, 2, 4, 1]]
+        result = unordered.sort_index(axis=0)
+        expected = frame
+        tm.assert_frame_equal(result, expected)
+
+        result = unordered.sort_index(ascending=False)
+        expected = frame[::-1]
+        tm.assert_frame_equal(result, expected)
+
+        # axis=1 : sort columns by column names
+        unordered = frame.iloc[:, [2, 1, 3, 0]]
+        result = unordered.sort_index(axis=1)
+        tm.assert_frame_equal(result, frame)
+
+        result = unordered.sort_index(axis=1, ascending=False)
+        expected = frame.iloc[:, ::-1]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("level", ["A", 0])  # GH#21052
+    def test_sort_index_multiindex(self, level):
+        # GH#13496
+
+        # sort rows by specified level of multi-index
+        mi = MultiIndex.from_tuples(
+            [[2, 1, 3], [2, 1, 2], [1, 1, 1]], names=list("ABC")
+        )
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], index=mi)
+
+        expected_mi = MultiIndex.from_tuples(
+            [[1, 1, 1], [2, 1, 2], [2, 1, 3]], names=list("ABC")
+        )
+        expected = DataFrame([[5, 6], [3, 4], [1, 2]], index=expected_mi)
+        result = df.sort_index(level=level)
+        tm.assert_frame_equal(result, expected)
+
+        # sort_remaining=False
+        expected_mi = MultiIndex.from_tuples(
+            [[1, 1, 1], [2, 1, 3], [2, 1, 2]], names=list("ABC")
+        )
+        expected = DataFrame([[5, 6], [1, 2], [3, 4]], index=expected_mi)
+        result = df.sort_index(level=level, sort_remaining=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_intervalindex(self):
+        # this is a de-facto sort via unstack
+        # confirming that we sort in the order of the bins
+        y = Series(np.random.default_rng(2).standard_normal(100))
+        x1 = Series(np.sign(np.random.default_rng(2).standard_normal(100)))
+        x2 = pd.cut(
+            Series(np.random.default_rng(2).standard_normal(100)),
+            bins=[-3, -0.5, 0, 0.5, 3],
+        )
+        model = pd.concat([y, x1, x2], axis=1, keys=["Y", "X1", "X2"])
+
+        result = model.groupby(["X1", "X2"], observed=True).mean().unstack()
+        expected = IntervalIndex.from_tuples(
+            [(-3.0, -0.5), (-0.5, 0.0), (0.0, 0.5), (0.5, 3.0)], closed="right"
+        )
+        result = result.columns.levels[1].categories
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize(
+        "original_dict, sorted_dict, ascending, ignore_index, output_index",
+        [
+            ({"A": [1, 2, 3]}, {"A": [2, 3, 1]}, False, True, [0, 1, 2]),
+            ({"A": [1, 2, 3]}, {"A": [1, 3, 2]}, True, True, [0, 1, 2]),
+            ({"A": [1, 2, 3]}, {"A": [2, 3, 1]}, False, False, [5, 3, 2]),
+            ({"A": [1, 2, 3]}, {"A": [1, 3, 2]}, True, False, [2, 3, 5]),
+        ],
+    )
+    def test_sort_index_ignore_index(
+        self, inplace, original_dict, sorted_dict, ascending, ignore_index, output_index
+    ):
+        # GH 30114
+        original_index = [2, 5, 3]
+        df = DataFrame(original_dict, index=original_index)
+        expected_df = DataFrame(sorted_dict, index=output_index)
+        kwargs = {
+            "ascending": ascending,
+            "ignore_index": ignore_index,
+            "inplace": inplace,
+        }
+
+        if inplace:
+            result_df = df.copy()
+            result_df.sort_index(**kwargs)
+        else:
+            result_df = df.sort_index(**kwargs)
+
+        tm.assert_frame_equal(result_df, expected_df)
+        tm.assert_frame_equal(df, DataFrame(original_dict, index=original_index))
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize("ignore_index", [True, False])
+    def test_respect_ignore_index(self, inplace, ignore_index):
+        # GH 43591
+        df = DataFrame({"a": [1, 2, 3]}, index=RangeIndex(4, -1, -2))
+        result = df.sort_index(
+            ascending=False, ignore_index=ignore_index, inplace=inplace
+        )
+
+        if inplace:
+            result = df
+        if ignore_index:
+            expected = DataFrame({"a": [1, 2, 3]})
+        else:
+            expected = DataFrame({"a": [1, 2, 3]}, index=RangeIndex(4, -1, -2))
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize(
+        "original_dict, sorted_dict, ascending, ignore_index, output_index",
+        [
+            (
+                {"M1": [1, 2], "M2": [3, 4]},
+                {"M1": [1, 2], "M2": [3, 4]},
+                True,
+                True,
+                [0, 1],
+            ),
+            (
+                {"M1": [1, 2], "M2": [3, 4]},
+                {"M1": [2, 1], "M2": [4, 3]},
+                False,
+                True,
+                [0, 1],
+            ),
+            (
+                {"M1": [1, 2], "M2": [3, 4]},
+                {"M1": [1, 2], "M2": [3, 4]},
+                True,
+                False,
+                MultiIndex.from_tuples([(2, 1), (3, 4)], names=list("AB")),
+            ),
+            (
+                {"M1": [1, 2], "M2": [3, 4]},
+                {"M1": [2, 1], "M2": [4, 3]},
+                False,
+                False,
+                MultiIndex.from_tuples([(3, 4), (2, 1)], names=list("AB")),
+            ),
+        ],
+    )
+    def test_sort_index_ignore_index_multi_index(
+        self, inplace, original_dict, sorted_dict, ascending, ignore_index, output_index
+    ):
+        # GH 30114, this is to test ignore_index on MultiIndex of index
+        mi = MultiIndex.from_tuples([(2, 1), (3, 4)], names=list("AB"))
+        df = DataFrame(original_dict, index=mi)
+        expected_df = DataFrame(sorted_dict, index=output_index)
+
+        kwargs = {
+            "ascending": ascending,
+            "ignore_index": ignore_index,
+            "inplace": inplace,
+        }
+
+        if inplace:
+            result_df = df.copy()
+            result_df.sort_index(**kwargs)
+        else:
+            result_df = df.sort_index(**kwargs)
+
+        tm.assert_frame_equal(result_df, expected_df)
+        tm.assert_frame_equal(df, DataFrame(original_dict, index=mi))
+
+    def test_sort_index_categorical_multiindex(self):
+        # GH#15058
+        df = DataFrame(
+            {
+                "a": range(6),
+                "l1": pd.Categorical(
+                    ["a", "a", "b", "b", "c", "c"],
+                    categories=["c", "a", "b"],
+                    ordered=True,
+                ),
+                "l2": [0, 1, 0, 1, 0, 1],
+            }
+        )
+        result = df.set_index(["l1", "l2"]).sort_index()
+        expected = DataFrame(
+            [4, 5, 0, 1, 2, 3],
+            columns=["a"],
+            index=MultiIndex(
+                levels=[
+                    CategoricalIndex(
+                        ["c", "a", "b"],
+                        categories=["c", "a", "b"],
+                        ordered=True,
+                        name="l1",
+                        dtype="category",
+                    ),
+                    [0, 1],
+                ],
+                codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+                names=["l1", "l2"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_and_reconstruction(self):
+        # GH#15622
+        # lexsortedness should be identical
+        # across MultiIndex construction methods
+
+        df = DataFrame([[1, 1], [2, 2]], index=list("ab"))
+        expected = DataFrame(
+            [[1, 1], [2, 2], [1, 1], [2, 2]],
+            index=MultiIndex.from_tuples(
+                [(0.5, "a"), (0.5, "b"), (0.8, "a"), (0.8, "b")]
+            ),
+        )
+        assert expected.index._is_lexsorted()
+
+        result = DataFrame(
+            [[1, 1], [2, 2], [1, 1], [2, 2]],
+            index=MultiIndex.from_product([[0.5, 0.8], list("ab")]),
+        )
+        result = result.sort_index()
+        assert result.index.is_monotonic_increasing
+
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(
+            [[1, 1], [2, 2], [1, 1], [2, 2]],
+            index=MultiIndex(
+                levels=[[0.5, 0.8], ["a", "b"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+            ),
+        )
+        result = result.sort_index()
+        assert result.index._is_lexsorted()
+
+        tm.assert_frame_equal(result, expected)
+
+        concatted = pd.concat([df, df], keys=[0.8, 0.5])
+        result = concatted.sort_index()
+
+        assert result.index.is_monotonic_increasing
+
+        tm.assert_frame_equal(result, expected)
+
+        # GH#14015
+        df = DataFrame(
+            [[1, 2], [6, 7]],
+            columns=MultiIndex.from_tuples(
+                [(0, "20160811 12:00:00"), (0, "20160809 12:00:00")],
+                names=["l1", "Date"],
+            ),
+        )
+
+        df.columns = df.columns.set_levels(
+            pd.to_datetime(df.columns.levels[1]), level=1
+        )
+        assert not df.columns.is_monotonic_increasing
+        result = df.sort_index(axis=1)
+        assert result.columns.is_monotonic_increasing
+        result = df.sort_index(axis=1, level=1)
+        assert result.columns.is_monotonic_increasing
+
+    # TODO: better name, de-duplicate with test_sort_index_level above
+    def test_sort_index_level2(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        df = frame.copy()
+        df.index = np.arange(len(df))
+
+        # axis=1
+
+        # series
+        a_sorted = frame["A"].sort_index(level=0)
+
+        # preserve names
+        assert a_sorted.index.names == frame.index.names
+
+        # inplace
+        rs = frame.copy()
+        return_value = rs.sort_index(level=0, inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(rs, frame.sort_index(level=0))
+
+    def test_sort_index_level_large_cardinality(self):
+        # GH#2684 (int64)
+        index = MultiIndex.from_arrays([np.arange(4000)] * 3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(4000).astype("int64"), index=index
+        )
+
+        # it works!
+        result = df.sort_index(level=0)
+        assert result.index._lexsort_depth == 3
+
+        # GH#2684 (int32)
+        index = MultiIndex.from_arrays([np.arange(4000)] * 3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(4000).astype("int32"), index=index
+        )
+
+        # it works!
+        result = df.sort_index(level=0)
+        assert (result.dtypes.values == df.dtypes.values).all()
+        assert result.index._lexsort_depth == 3
+
+    def test_sort_index_level_by_name(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        frame.index.names = ["first", "second"]
+        result = frame.sort_index(level="second")
+        expected = frame.sort_index(level=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_level_mixed(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        sorted_before = frame.sort_index(level=1)
+
+        df = frame.copy()
+        df["foo"] = "bar"
+        sorted_after = df.sort_index(level=1)
+        tm.assert_frame_equal(sorted_before, sorted_after.drop(["foo"], axis=1))
+
+        dft = frame.T
+        sorted_before = dft.sort_index(level=1, axis=1)
+        dft["foo", "three"] = "bar"
+
+        sorted_after = dft.sort_index(level=1, axis=1)
+        tm.assert_frame_equal(
+            sorted_before.drop([("foo", "three")], axis=1),
+            sorted_after.drop([("foo", "three")], axis=1),
+        )
+
+    def test_sort_index_preserve_levels(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        result = frame.sort_index()
+        assert result.index.names == frame.index.names
+
+    @pytest.mark.parametrize(
+        "gen,extra",
+        [
+            ([1.0, 3.0, 2.0, 5.0], 4.0),
+            ([1, 3, 2, 5], 4),
+            (
+                [
+                    Timestamp("20130101"),
+                    Timestamp("20130103"),
+                    Timestamp("20130102"),
+                    Timestamp("20130105"),
+                ],
+                Timestamp("20130104"),
+            ),
+            (["1one", "3one", "2one", "5one"], "4one"),
+        ],
+    )
+    def test_sort_index_multilevel_repr_8017(self, gen, extra):
+        data = np.random.default_rng(2).standard_normal((3, 4))
+
+        columns = MultiIndex.from_tuples([("red", i) for i in gen])
+        df = DataFrame(data, index=list("def"), columns=columns)
+        df2 = pd.concat(
+            [
+                df,
+                DataFrame(
+                    "world",
+                    index=list("def"),
+                    columns=MultiIndex.from_tuples([("red", extra)]),
+                ),
+            ],
+            axis=1,
+        )
+
+        # check that the repr is good
+        # make sure that we have a correct sparsified repr
+        # e.g. only 1 header of read
+        assert str(df2).splitlines()[0].split() == ["red"]
+
+        # GH 8017
+        # sorting fails after columns added
+
+        # construct single-dtype then sort
+        result = df.copy().sort_index(axis=1)
+        expected = df.iloc[:, [0, 2, 1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df2.sort_index(axis=1)
+        expected = df2.iloc[:, [0, 2, 1, 4, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # setitem then sort
+        result = df.copy()
+        result[("red", extra)] = "world"
+
+        result = result.sort_index(axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "categories",
+        [
+            pytest.param(["a", "b", "c"], id="str"),
+            pytest.param(
+                [pd.Interval(0, 1), pd.Interval(1, 2), pd.Interval(2, 3)],
+                id="pd.Interval",
+            ),
+        ],
+    )
+    def test_sort_index_with_categories(self, categories):
+        # GH#23452
+        df = DataFrame(
+            {"foo": range(len(categories))},
+            index=CategoricalIndex(
+                data=categories, categories=categories, ordered=True
+            ),
+        )
+        df.index = df.index.reorder_categories(df.index.categories[::-1])
+        result = df.sort_index()
+        expected = DataFrame(
+            {"foo": reversed(range(len(categories)))},
+            index=CategoricalIndex(
+                data=categories[::-1], categories=categories[::-1], ordered=True
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ascending",
+        [
+            None,
+            [True, None],
+            [False, "True"],
+        ],
+    )
+    def test_sort_index_ascending_bad_value_raises(self, ascending):
+        # GH 39434
+        df = DataFrame(np.arange(64))
+        length = len(df.index)
+        df.index = [(i - length / 2) % length for i in range(length)]
+        match = 'For argument "ascending" expected type bool'
+        with pytest.raises(ValueError, match=match):
+            df.sort_index(axis=0, ascending=ascending, na_position="first")
+
+    @pytest.mark.parametrize(
+        "ascending",
+        [(True, False), [True, False]],
+    )
+    def test_sort_index_ascending_tuple(self, ascending):
+        df = DataFrame(
+            {
+                "legs": [4, 2, 4, 2, 2],
+            },
+            index=MultiIndex.from_tuples(
+                [
+                    ("mammal", "dog"),
+                    ("bird", "duck"),
+                    ("mammal", "horse"),
+                    ("bird", "penguin"),
+                    ("mammal", "kangaroo"),
+                ],
+                names=["class", "animal"],
+            ),
+        )
+
+        # parameter `ascending`` is a tuple
+        result = df.sort_index(level=(0, 1), ascending=ascending)
+
+        expected = DataFrame(
+            {
+                "legs": [2, 2, 2, 4, 4],
+            },
+            index=MultiIndex.from_tuples(
+                [
+                    ("bird", "penguin"),
+                    ("bird", "duck"),
+                    ("mammal", "kangaroo"),
+                    ("mammal", "horse"),
+                    ("mammal", "dog"),
+                ],
+                names=["class", "animal"],
+            ),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameSortIndexKey:
+    def test_sort_multi_index_key(self):
+        # GH 25775, testing that sorting by index works with a multi-index.
+        df = DataFrame(
+            {"a": [3, 1, 2], "b": [0, 0, 0], "c": [0, 1, 2], "d": list("abc")}
+        ).set_index(list("abc"))
+
+        result = df.sort_index(level=list("ac"), key=lambda x: x)
+
+        expected = DataFrame(
+            {"a": [1, 2, 3], "b": [0, 0, 0], "c": [1, 2, 0], "d": list("bca")}
+        ).set_index(list("abc"))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(level=list("ac"), key=lambda x: -x)
+        expected = DataFrame(
+            {"a": [3, 2, 1], "b": [0, 0, 0], "c": [0, 2, 1], "d": list("acb")}
+        ).set_index(list("abc"))
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_key(self):  # issue 27237
+        df = DataFrame(np.arange(6, dtype="int64"), index=list("aaBBca"))
+
+        result = df.sort_index()
+        expected = df.iloc[[2, 3, 0, 1, 5, 4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(key=lambda x: x.str.lower())
+        expected = df.iloc[[0, 1, 5, 2, 3, 4]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(key=lambda x: x.str.lower(), ascending=False)
+        expected = df.iloc[[4, 2, 3, 0, 1, 5]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_key_int(self):
+        df = DataFrame(np.arange(6, dtype="int64"), index=np.arange(6, dtype="int64"))
+
+        result = df.sort_index()
+        tm.assert_frame_equal(result, df)
+
+        result = df.sort_index(key=lambda x: -x)
+        expected = df.sort_index(ascending=False)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(key=lambda x: 2 * x)
+        tm.assert_frame_equal(result, df)
+
+    def test_sort_multi_index_key_str(self):
+        # GH 25775, testing that sorting by index works with a multi-index.
+        df = DataFrame(
+            {"a": ["B", "a", "C"], "b": [0, 1, 0], "c": list("abc"), "d": [0, 1, 2]}
+        ).set_index(list("abc"))
+
+        result = df.sort_index(level="a", key=lambda x: x.str.lower())
+
+        expected = DataFrame(
+            {"a": ["a", "B", "C"], "b": [1, 0, 0], "c": list("bac"), "d": [1, 0, 2]}
+        ).set_index(list("abc"))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_index(
+            level=list("abc"),  # can refer to names
+            key=lambda x: x.str.lower() if x.name in ["a", "c"] else -x,
+        )
+
+        expected = DataFrame(
+            {"a": ["a", "B", "C"], "b": [1, 0, 0], "c": list("bac"), "d": [1, 0, 2]}
+        ).set_index(list("abc"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_changes_length_raises(self):
+        df = DataFrame({"A": [1, 2, 3]})
+        with pytest.raises(ValueError, match="change the shape"):
+            df.sort_index(key=lambda x: x[:1])
+
+    def test_sort_index_multiindex_sparse_column(self):
+        # GH 29735, testing that sort_index on a multiindexed frame with sparse
+        # columns fills with 0.
+        expected = DataFrame(
+            {
+                i: pd.array([0.0, 0.0, 0.0, 0.0], dtype=pd.SparseDtype("float64", 0.0))
+                for i in range(4)
+            },
+            index=MultiIndex.from_product([[1, 2], [1, 2]]),
+        )
+
+        result = expected.sort_index(level=0)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_na_position(self):
+        # GH#51612
+        df = DataFrame([1, 2], index=MultiIndex.from_tuples([(1, 1), (1, pd.NA)]))
+        expected = df.copy()
+        result = df.sort_index(level=[0, 1], na_position="last")
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_index_multiindex_sort_remaining(self, ascending):
+        # GH #24247
+        df = DataFrame(
+            {"A": [1, 2, 3, 4, 5], "B": [10, 20, 30, 40, 50]},
+            index=MultiIndex.from_tuples(
+                [("a", "x"), ("a", "y"), ("b", "x"), ("b", "y"), ("c", "x")]
+            ),
+        )
+
+        result = df.sort_index(level=1, sort_remaining=False, ascending=ascending)
+
+        if ascending:
+            expected = DataFrame(
+                {"A": [1, 3, 5, 2, 4], "B": [10, 30, 50, 20, 40]},
+                index=MultiIndex.from_tuples(
+                    [("a", "x"), ("b", "x"), ("c", "x"), ("a", "y"), ("b", "y")]
+                ),
+            )
+        else:
+            expected = DataFrame(
+                {"A": [2, 4, 1, 3, 5], "B": [20, 40, 10, 30, 50]},
+                index=MultiIndex.from_tuples(
+                    [("a", "y"), ("b", "y"), ("a", "x"), ("b", "x"), ("c", "x")]
+                ),
+            )
+
+        tm.assert_frame_equal(result, expected)
+
+
+def test_sort_index_with_sliced_multiindex():
+    # GH 55379
+    mi = MultiIndex.from_tuples(
+        [
+            ("a", "10"),
+            ("a", "18"),
+            ("a", "25"),
+            ("b", "16"),
+            ("b", "26"),
+            ("a", "45"),
+            ("b", "28"),
+            ("a", "5"),
+            ("a", "50"),
+            ("a", "51"),
+            ("b", "4"),
+        ],
+        names=["group", "str"],
+    )
+
+    df = DataFrame({"x": range(len(mi))}, index=mi)
+    result = df.iloc[0:6].sort_index()
+
+    expected = DataFrame(
+        {"x": [0, 1, 2, 5, 3, 4]},
+        index=MultiIndex.from_tuples(
+            [
+                ("a", "10"),
+                ("a", "18"),
+                ("a", "25"),
+                ("a", "45"),
+                ("b", "16"),
+                ("b", "26"),
+            ],
+            names=["group", "str"],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_axis_columns_ignore_index():
+    # GH 56478
+    df = DataFrame([[1, 2]], columns=["d", "c"])
+    result = df.sort_index(axis="columns", ignore_index=True)
+    expected = DataFrame([[2, 1]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_axis_columns_ignore_index_ascending_false():
+    # GH 57293
+    df = DataFrame(
+        {
+            "b": [1.0, 3.0, np.nan],
+            "a": [1, 4, 3],
+            1: ["a", "b", "c"],
+            "e": [3, 1, 4],
+            "d": [1, 2, 8],
+        }
+    ).set_index(["b", "a", 1])
+    result = df.sort_index(axis="columns", ignore_index=True, ascending=False)
+    expected = df.copy()
+    expected.columns = RangeIndex(2)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sort_index_stable_sort():
+    # GH 57151
+    df = DataFrame(
+        data=[
+            (Timestamp("2024-01-30 13:00:00"), 13.0),
+            (Timestamp("2024-01-30 13:00:00"), 13.1),
+            (Timestamp("2024-01-30 12:00:00"), 12.0),
+            (Timestamp("2024-01-30 12:00:00"), 12.1),
+        ],
+        columns=["dt", "value"],
+    ).set_index(["dt"])
+    result = df.sort_index(level="dt", kind="stable")
+    expected = DataFrame(
+        data=[
+            (Timestamp("2024-01-30 12:00:00"), 12.0),
+            (Timestamp("2024-01-30 12:00:00"), 12.1),
+            (Timestamp("2024-01-30 13:00:00"), 13.0),
+            (Timestamp("2024-01-30 13:00:00"), 13.1),
+        ],
+        columns=["dt", "value"],
+    ).set_index(["dt"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_sort_values.py b/pandas/pandas/tests/frame/methods/test_sort_values.py
new file mode 100644
index 0000000000000..e728526519e9d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_sort_values.py
@@ -0,0 +1,922 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    NaT,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+class TestDataFrameSortValues:
+    @pytest.mark.parametrize("dtype", [np.uint8, bool])
+    def test_sort_values_sparse_no_warning(self, dtype):
+        # GH#45618
+        ser = pd.Series(Categorical(["a", "b", "a"], categories=["a", "b", "c"]))
+        df = pd.get_dummies(ser, dtype=dtype, sparse=True)
+
+        with tm.assert_produces_warning(None):
+            # No warnings about constructing Index from SparseArray
+            df.sort_values(by=df.columns.tolist())
+
+    def test_sort_values(self):
+        frame = DataFrame(
+            [[1, 1, 2], [3, 1, 0], [4, 5, 6]], index=[1, 2, 3], columns=list("ABC")
+        )
+
+        # by column (axis=0)
+        sorted_df = frame.sort_values(by="A")
+        indexer = frame["A"].argsort().values
+        expected = frame.loc[frame.index[indexer]]
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by="A", ascending=False)
+        indexer = indexer[::-1]
+        expected = frame.loc[frame.index[indexer]]
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by="A", ascending=False)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # GH4839
+        sorted_df = frame.sort_values(by=["A"], ascending=[False])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # multiple bys
+        sorted_df = frame.sort_values(by=["B", "C"])
+        expected = frame.loc[[2, 1, 3]]
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by=["B", "C"], ascending=False)
+        tm.assert_frame_equal(sorted_df, expected[::-1])
+
+        sorted_df = frame.sort_values(by=["B", "A"], ascending=[True, False])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            frame.sort_values(by=["A", "B"], axis=2, inplace=True)
+
+        # by row (axis=1): GH#10806
+        sorted_df = frame.sort_values(by=3, axis=1)
+        expected = frame
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by=3, axis=1, ascending=False)
+        expected = frame.reindex(columns=["C", "B", "A"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by=[1, 2], axis="columns")
+        expected = frame.reindex(columns=["B", "A", "C"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by=[1, 3], axis=1, ascending=[True, False])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.sort_values(by=[1, 3], axis=1, ascending=False)
+        expected = frame.reindex(columns=["C", "B", "A"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        msg = r"Length of ascending \(5\) != length of by \(2\)"
+        with pytest.raises(ValueError, match=msg):
+            frame.sort_values(by=["A", "B"], axis=0, ascending=[True] * 5)
+
+    def test_sort_values_by_empty_list(self):
+        # https://github.com/pandas-dev/pandas/issues/40258
+        expected = DataFrame({"a": [1, 4, 2, 5, 3, 6]})
+        result = expected.sort_values(by=[])
+        tm.assert_frame_equal(result, expected)
+        assert result is not expected
+
+    def test_sort_values_inplace(self):
+        frame = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=[1, 2, 3, 4],
+            columns=["A", "B", "C", "D"],
+        )
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(by="A", inplace=True)
+        assert return_value is None
+        expected = frame.sort_values(by="A")
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(by=1, axis=1, inplace=True)
+        assert return_value is None
+        expected = frame.sort_values(by=1, axis=1)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(by="A", ascending=False, inplace=True)
+        assert return_value is None
+        expected = frame.sort_values(by="A", ascending=False)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(
+            by=["A", "B"], ascending=False, inplace=True
+        )
+        assert return_value is None
+        expected = frame.sort_values(by=["A", "B"], ascending=False)
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_multicolumn(self):
+        A = np.arange(5).repeat(20)
+        B = np.tile(np.arange(5), 20)
+        np.random.default_rng(2).shuffle(A)
+        np.random.default_rng(2).shuffle(B)
+        frame = DataFrame(
+            {"A": A, "B": B, "C": np.random.default_rng(2).standard_normal(100)}
+        )
+
+        result = frame.sort_values(by=["A", "B"])
+        indexer = np.lexsort((frame["B"], frame["A"]))
+        expected = frame.take(indexer)
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.sort_values(by=["A", "B"], ascending=False)
+        indexer = np.lexsort(
+            (frame["B"].rank(ascending=False), frame["A"].rank(ascending=False))
+        )
+        expected = frame.take(indexer)
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.sort_values(by=["B", "A"])
+        indexer = np.lexsort((frame["A"], frame["B"]))
+        expected = frame.take(indexer)
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_multicolumn_uint64(self):
+        # GH#9918
+        # uint64 multicolumn sort
+
+        df = DataFrame(
+            {
+                "a": pd.Series([18446637057563306014, 1162265347240853609]),
+                "b": pd.Series([1, 2]),
+            }
+        )
+        df["a"] = df["a"].astype(np.uint64)
+        result = df.sort_values(["a", "b"])
+
+        expected = DataFrame(
+            {
+                "a": pd.Series([18446637057563306014, 1162265347240853609]),
+                "b": pd.Series([1, 2]),
+            },
+            index=range(1, -1, -1),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_nan(self):
+        # GH#3917
+        df = DataFrame(
+            {"A": [1, 2, np.nan, 1, 6, 8, 4], "B": [9, np.nan, 5, 2, 5, 4, 5]}
+        )
+
+        # sort one column only
+        expected = DataFrame(
+            {"A": [np.nan, 1, 1, 2, 4, 6, 8], "B": [5, 9, 2, np.nan, 5, 5, 4]},
+            index=[2, 0, 3, 1, 6, 4, 5],
+        )
+        sorted_df = df.sort_values(["A"], na_position="first")
+        tm.assert_frame_equal(sorted_df, expected)
+
+        expected = DataFrame(
+            {"A": [np.nan, 8, 6, 4, 2, 1, 1], "B": [5, 4, 5, 5, np.nan, 9, 2]},
+            index=[2, 5, 4, 6, 1, 0, 3],
+        )
+        sorted_df = df.sort_values(["A"], na_position="first", ascending=False)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        expected = df.reindex(columns=["B", "A"])
+        sorted_df = df.sort_values(by=1, axis=1, na_position="first")
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # na_position='last', order
+        expected = DataFrame(
+            {"A": [1, 1, 2, 4, 6, 8, np.nan], "B": [2, 9, np.nan, 5, 5, 4, 5]},
+            index=[3, 0, 1, 6, 4, 5, 2],
+        )
+        sorted_df = df.sort_values(["A", "B"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # na_position='first', order
+        expected = DataFrame(
+            {"A": [np.nan, 1, 1, 2, 4, 6, 8], "B": [5, 2, 9, np.nan, 5, 5, 4]},
+            index=[2, 3, 0, 1, 6, 4, 5],
+        )
+        sorted_df = df.sort_values(["A", "B"], na_position="first")
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # na_position='first', not order
+        expected = DataFrame(
+            {"A": [np.nan, 1, 1, 2, 4, 6, 8], "B": [5, 9, 2, np.nan, 5, 5, 4]},
+            index=[2, 0, 3, 1, 6, 4, 5],
+        )
+        sorted_df = df.sort_values(["A", "B"], ascending=[1, 0], na_position="first")
+        tm.assert_frame_equal(sorted_df, expected)
+
+        # na_position='last', not order
+        expected = DataFrame(
+            {"A": [8, 6, 4, 2, 1, 1, np.nan], "B": [4, 5, 5, np.nan, 2, 9, 5]},
+            index=[5, 4, 6, 1, 3, 0, 2],
+        )
+        sorted_df = df.sort_values(["A", "B"], ascending=[0, 1], na_position="last")
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_stable_descending_sort(self):
+        # GH#6399
+        df = DataFrame(
+            [[2, "first"], [2, "second"], [1, "a"], [1, "b"]],
+            columns=["sort_col", "order"],
+        )
+        sorted_df = df.sort_values(by="sort_col", kind="mergesort", ascending=False)
+        tm.assert_frame_equal(df, sorted_df)
+
+    @pytest.mark.parametrize(
+        "expected_idx_non_na, ascending",
+        [
+            [
+                [3, 4, 5, 0, 1, 8, 6, 9, 7, 10, 13, 14],
+                [True, True],
+            ],
+            [
+                [0, 3, 4, 5, 1, 8, 6, 7, 10, 13, 14, 9],
+                [True, False],
+            ],
+            [
+                [9, 7, 10, 13, 14, 6, 8, 1, 3, 4, 5, 0],
+                [False, True],
+            ],
+            [
+                [7, 10, 13, 14, 9, 6, 8, 1, 0, 3, 4, 5],
+                [False, False],
+            ],
+        ],
+    )
+    @pytest.mark.parametrize("na_position", ["first", "last"])
+    def test_sort_values_stable_multicolumn_sort(
+        self, expected_idx_non_na, ascending, na_position
+    ):
+        # GH#38426 Clarify sort_values with mult. columns / labels is stable
+        df = DataFrame(
+            {
+                "A": [1, 2, np.nan, 1, 1, 1, 6, 8, 4, 8, 8, np.nan, np.nan, 8, 8],
+                "B": [9, np.nan, 5, 2, 2, 2, 5, 4, 5, 3, 4, np.nan, np.nan, 4, 4],
+            }
+        )
+        # All rows with NaN in col "B" only have unique values in "A", therefore,
+        # only the rows with NaNs in "A" have to be treated individually:
+        expected_idx = (
+            [11, 12, 2] + expected_idx_non_na
+            if na_position == "first"
+            else expected_idx_non_na + [2, 11, 12]
+        )
+        expected = df.take(expected_idx)
+        sorted_df = df.sort_values(
+            ["A", "B"], ascending=ascending, na_position=na_position
+        )
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_stable_categorial(self):
+        # GH#16793
+        df = DataFrame({"x": Categorical(np.repeat([1, 2, 3, 4], 5), ordered=True)})
+        expected = df.copy()
+        sorted_df = df.sort_values("x", kind="mergesort")
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_datetimes(self):
+        # GH#3461, argsort / lexsort differences for a datetime column
+        df = DataFrame(
+            ["a", "a", "a", "b", "c", "d", "e", "f", "g"],
+            columns=["A"],
+            index=date_range("20130101", periods=9),
+        )
+        dts = [
+            Timestamp(x)
+            for x in [
+                "2004-02-11",
+                "2004-01-21",
+                "2004-01-26",
+                "2005-09-20",
+                "2010-10-04",
+                "2009-05-12",
+                "2008-11-12",
+                "2010-09-28",
+                "2010-09-28",
+            ]
+        ]
+        df["B"] = dts[::2] + dts[1::2]
+        df["C"] = 2.0
+        df["A1"] = 3.0
+
+        df1 = df.sort_values(by="A")
+        df2 = df.sort_values(by=["A"])
+        tm.assert_frame_equal(df1, df2)
+
+        df1 = df.sort_values(by="B")
+        df2 = df.sort_values(by=["B"])
+        tm.assert_frame_equal(df1, df2)
+
+        df1 = df.sort_values(by="B")
+
+        df2 = df.sort_values(by=["C", "B"])
+        tm.assert_frame_equal(df1, df2)
+
+    def test_sort_values_frame_column_inplace_sort_exception(self, float_frame):
+        s = float_frame["A"]
+        float_frame_orig = float_frame.copy()
+        # INFO(CoW) Series is a new object, so can be changed inplace
+        # without modifying original datafame
+        s.sort_values(inplace=True)
+        tm.assert_series_equal(s, float_frame_orig["A"].sort_values())
+        # column in dataframe is not changed
+        tm.assert_frame_equal(float_frame, float_frame_orig)
+
+        cp = s.copy()
+        cp.sort_values()  # it works!
+
+    def test_sort_values_nat_values_in_int_column(self):
+        # GH#14922: "sorting with large float and multiple columns incorrect"
+
+        # cause was that the int64 value NaT was considered as "na". Which is
+        # only correct for datetime64 columns.
+
+        int_values = (2, int(NaT._value))
+        float_values = (2.0, -1.797693e308)
+
+        df = DataFrame(
+            {"int": int_values, "float": float_values}, columns=["int", "float"]
+        )
+
+        df_reversed = DataFrame(
+            {"int": int_values[::-1], "float": float_values[::-1]},
+            columns=["int", "float"],
+            index=range(1, -1, -1),
+        )
+
+        # NaT is not a "na" for int64 columns, so na_position must not
+        # influence the result:
+        df_sorted = df.sort_values(["int", "float"], na_position="last")
+        tm.assert_frame_equal(df_sorted, df_reversed)
+
+        df_sorted = df.sort_values(["int", "float"], na_position="first")
+        tm.assert_frame_equal(df_sorted, df_reversed)
+
+        # reverse sorting order
+        df_sorted = df.sort_values(["int", "float"], ascending=False)
+        tm.assert_frame_equal(df_sorted, df)
+
+        # and now check if NaT is still considered as "na" for datetime64
+        # columns:
+        df = DataFrame(
+            {"datetime": [Timestamp("2016-01-01"), NaT], "float": float_values},
+            columns=["datetime", "float"],
+        )
+
+        df_reversed = DataFrame(
+            {"datetime": [NaT, Timestamp("2016-01-01")], "float": float_values[::-1]},
+            columns=["datetime", "float"],
+            index=range(1, -1, -1),
+        )
+
+        df_sorted = df.sort_values(["datetime", "float"], na_position="first")
+        tm.assert_frame_equal(df_sorted, df_reversed)
+
+        df_sorted = df.sort_values(["datetime", "float"], na_position="last")
+        tm.assert_frame_equal(df_sorted, df)
+
+        # Ascending should not affect the results.
+        df_sorted = df.sort_values(["datetime", "float"], ascending=False)
+        tm.assert_frame_equal(df_sorted, df)
+
+    def test_sort_nat(self):
+        # GH 16836
+
+        d1 = [Timestamp(x) for x in ["2016-01-01", "2015-01-01", np.nan, "2016-01-01"]]
+        d2 = [
+            Timestamp(x)
+            for x in ["2017-01-01", "2014-01-01", "2016-01-01", "2015-01-01"]
+        ]
+        df = DataFrame({"a": d1, "b": d2}, index=[0, 1, 2, 3])
+
+        d3 = [Timestamp(x) for x in ["2015-01-01", "2016-01-01", "2016-01-01", np.nan]]
+        d4 = [
+            Timestamp(x)
+            for x in ["2014-01-01", "2015-01-01", "2017-01-01", "2016-01-01"]
+        ]
+        expected = DataFrame({"a": d3, "b": d4}, index=[1, 3, 0, 2])
+        sorted_df = df.sort_values(by=["a", "b"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_na_position_with_categories(self):
+        # GH#22556
+        # Positioning missing value properly when column is Categorical.
+        categories = ["A", "B", "C"]
+        category_indices = [0, 2, 4]
+        list_of_nans = [np.nan, np.nan]
+        na_indices = [1, 3]
+        na_position_first = "first"
+        na_position_last = "last"
+        column_name = "c"
+
+        reversed_categories = sorted(categories, reverse=True)
+        reversed_category_indices = sorted(category_indices, reverse=True)
+        reversed_na_indices = sorted(na_indices)
+
+        df = DataFrame(
+            {
+                column_name: Categorical(
+                    ["A", np.nan, "B", np.nan, "C"], categories=categories, ordered=True
+                )
+            }
+        )
+        # sort ascending with na first
+        result = df.sort_values(
+            by=column_name, ascending=True, na_position=na_position_first
+        )
+        expected = DataFrame(
+            {
+                column_name: Categorical(
+                    list_of_nans + categories, categories=categories, ordered=True
+                )
+            },
+            index=na_indices + category_indices,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        # sort ascending with na last
+        result = df.sort_values(
+            by=column_name, ascending=True, na_position=na_position_last
+        )
+        expected = DataFrame(
+            {
+                column_name: Categorical(
+                    categories + list_of_nans, categories=categories, ordered=True
+                )
+            },
+            index=category_indices + na_indices,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        # sort descending with na first
+        result = df.sort_values(
+            by=column_name, ascending=False, na_position=na_position_first
+        )
+        expected = DataFrame(
+            {
+                column_name: Categorical(
+                    list_of_nans + reversed_categories,
+                    categories=categories,
+                    ordered=True,
+                )
+            },
+            index=reversed_na_indices + reversed_category_indices,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        # sort descending with na last
+        result = df.sort_values(
+            by=column_name, ascending=False, na_position=na_position_last
+        )
+        expected = DataFrame(
+            {
+                column_name: Categorical(
+                    reversed_categories + list_of_nans,
+                    categories=categories,
+                    ordered=True,
+                )
+            },
+            index=reversed_category_indices + reversed_na_indices,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_nat(self):
+        # GH#16836
+
+        d1 = [Timestamp(x) for x in ["2016-01-01", "2015-01-01", np.nan, "2016-01-01"]]
+        d2 = [
+            Timestamp(x)
+            for x in ["2017-01-01", "2014-01-01", "2016-01-01", "2015-01-01"]
+        ]
+        df = DataFrame({"a": d1, "b": d2}, index=[0, 1, 2, 3])
+
+        d3 = [Timestamp(x) for x in ["2015-01-01", "2016-01-01", "2016-01-01", np.nan]]
+        d4 = [
+            Timestamp(x)
+            for x in ["2014-01-01", "2015-01-01", "2017-01-01", "2016-01-01"]
+        ]
+        expected = DataFrame({"a": d3, "b": d4}, index=[1, 3, 0, 2])
+        sorted_df = df.sort_values(by=["a", "b"])
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_na_position_with_categories_raises(self):
+        df = DataFrame(
+            {
+                "c": Categorical(
+                    ["A", np.nan, "B", np.nan, "C"],
+                    categories=["A", "B", "C"],
+                    ordered=True,
+                )
+            }
+        )
+
+        with pytest.raises(ValueError, match="invalid na_position: bad_position"):
+            df.sort_values(by="c", ascending=False, na_position="bad_position")
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize(
+        "original_dict, sorted_dict, ignore_index, output_index",
+        [
+            ({"A": [1, 2, 3]}, {"A": [3, 2, 1]}, True, range(3)),
+            ({"A": [1, 2, 3]}, {"A": [3, 2, 1]}, False, range(2, -1, -1)),
+            (
+                {"A": [1, 2, 3], "B": [2, 3, 4]},
+                {"A": [3, 2, 1], "B": [4, 3, 2]},
+                True,
+                range(3),
+            ),
+            (
+                {"A": [1, 2, 3], "B": [2, 3, 4]},
+                {"A": [3, 2, 1], "B": [4, 3, 2]},
+                False,
+                range(2, -1, -1),
+            ),
+        ],
+    )
+    def test_sort_values_ignore_index(
+        self, inplace, original_dict, sorted_dict, ignore_index, output_index
+    ):
+        # GH 30114
+        df = DataFrame(original_dict)
+        expected = DataFrame(sorted_dict, index=output_index)
+        kwargs = {"ignore_index": ignore_index, "inplace": inplace}
+
+        if inplace:
+            result_df = df.copy()
+            result_df.sort_values("A", ascending=False, **kwargs)
+        else:
+            result_df = df.sort_values("A", ascending=False, **kwargs)
+
+        tm.assert_frame_equal(result_df, expected)
+        tm.assert_frame_equal(df, DataFrame(original_dict))
+
+    def test_sort_values_nat_na_position_default(self):
+        # GH 13230
+        expected = DataFrame(
+            {
+                "A": [1, 2, 3, 4, 4],
+                "date": pd.DatetimeIndex(
+                    [
+                        "2010-01-01 09:00:00",
+                        "2010-01-01 09:00:01",
+                        "2010-01-01 09:00:02",
+                        "2010-01-01 09:00:03",
+                        "NaT",
+                    ]
+                ),
+            }
+        )
+        result = expected.sort_values(["A", "date"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_item_cache(self):
+        # previous behavior incorrect retained an invalid _item_cache entry
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 3)), columns=["A", "B", "C"]
+        )
+        df["D"] = df["A"] * 2
+        ser = df["A"]
+        assert len(df._mgr.blocks) == 2
+
+        df.sort_values(by="A")
+
+        ser.iloc[0] = 99
+        assert df.iloc[0, 0] == df["A"][0]
+        assert df.iloc[0, 0] != 99
+
+    def test_sort_values_reshaping(self):
+        # GH 39426
+        values = list(range(21))
+        expected = DataFrame([values], columns=values)
+        df = expected.sort_values(expected.index[0], axis=1, ignore_index=True)
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_sort_values_no_by_inplace(self):
+        # GH#50643
+        df = DataFrame({"a": [1, 2, 3]})
+        expected = df.copy()
+        result = df.sort_values(by=[], inplace=True)
+        tm.assert_frame_equal(df, expected)
+        assert result is None
+
+    def test_sort_values_no_op_reset_index(self):
+        # GH#52553
+        df = DataFrame({"A": [10, 20], "B": [1, 5]}, index=[2, 3])
+        result = df.sort_values(by="A", ignore_index=True)
+        expected = DataFrame({"A": [10, 20], "B": [1, 5]})
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameSortKey:  # test key sorting (issue 27237)
+    def test_sort_values_inplace_key(self, sort_by_key):
+        frame = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=[1, 2, 3, 4],
+            columns=["A", "B", "C", "D"],
+        )
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(by="A", inplace=True, key=sort_by_key)
+        assert return_value is None
+        expected = frame.sort_values(by="A", key=sort_by_key)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(
+            by=1, axis=1, inplace=True, key=sort_by_key
+        )
+        assert return_value is None
+        expected = frame.sort_values(by=1, axis=1, key=sort_by_key)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        return_value = sorted_df.sort_values(
+            by="A", ascending=False, inplace=True, key=sort_by_key
+        )
+        assert return_value is None
+        expected = frame.sort_values(by="A", ascending=False, key=sort_by_key)
+        tm.assert_frame_equal(sorted_df, expected)
+
+        sorted_df = frame.copy()
+        sorted_df.sort_values(
+            by=["A", "B"], ascending=False, inplace=True, key=sort_by_key
+        )
+        expected = frame.sort_values(by=["A", "B"], ascending=False, key=sort_by_key)
+        tm.assert_frame_equal(sorted_df, expected)
+
+    def test_sort_values_key(self):
+        df = DataFrame(np.array([0, 5, np.nan, 3, 2, np.nan]))
+
+        result = df.sort_values(0)
+        expected = df.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(0, key=lambda x: x + 5)
+        expected = df.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(0, key=lambda x: -x, ascending=False)
+        expected = df.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_by_key(self):
+        df = DataFrame(
+            {
+                "a": np.array([0, 3, np.nan, 3, 2, np.nan]),
+                "b": np.array([0, 2, np.nan, 5, 2, np.nan]),
+            }
+        )
+
+        result = df.sort_values("a", key=lambda x: -x)
+        expected = df.iloc[[1, 3, 4, 0, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(by=["a", "b"], key=lambda x: -x)
+        expected = df.iloc[[3, 1, 4, 0, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(by=["a", "b"], key=lambda x: -x, ascending=False)
+        expected = df.iloc[[0, 4, 1, 3, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_by_key_by_name(self):
+        df = DataFrame(
+            {
+                "a": np.array([0, 3, np.nan, 3, 2, np.nan]),
+                "b": np.array([0, 2, np.nan, 5, 2, np.nan]),
+            }
+        )
+
+        def key(col):
+            if col.name == "a":
+                return -col
+            else:
+                return col
+
+        result = df.sort_values(by="a", key=key)
+        expected = df.iloc[[1, 3, 4, 0, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(by=["a"], key=key)
+        expected = df.iloc[[1, 3, 4, 0, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(by="b", key=key)
+        expected = df.iloc[[0, 1, 4, 3, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(by=["a", "b"], key=key)
+        expected = df.iloc[[1, 3, 4, 0, 2, 5]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_key_string(self):
+        df = DataFrame(np.array([["hello", "goodbye"], ["hello", "Hello"]]))
+
+        result = df.sort_values(1)
+        expected = df[::-1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values([0, 1], key=lambda col: col.str.lower())
+        tm.assert_frame_equal(result, df)
+
+        result = df.sort_values(
+            [0, 1], key=lambda col: col.str.lower(), ascending=False
+        )
+        expected = df.sort_values(1, key=lambda col: col.str.lower(), ascending=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_key_empty(self, sort_by_key):
+        df = DataFrame(np.array([]))
+
+        df.sort_values(0, key=sort_by_key)
+        df.sort_index(key=sort_by_key)
+
+    def test_changes_length_raises(self):
+        df = DataFrame({"A": [1, 2, 3]})
+        with pytest.raises(ValueError, match="change the shape"):
+            df.sort_values("A", key=lambda x: x[:1])
+
+    def test_sort_values_key_axes(self):
+        df = DataFrame({0: ["Hello", "goodbye"], 1: [0, 1]})
+
+        result = df.sort_values(0, key=lambda col: col.str.lower())
+        expected = df[::-1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(1, key=lambda col: -col)
+        expected = df[::-1]
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_key_dict_axis(self):
+        df = DataFrame({0: ["Hello", 0], 1: ["goodbye", 1]})
+
+        result = df.sort_values(0, key=lambda col: col.str.lower(), axis=1)
+        expected = df.loc[:, ::-1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.sort_values(1, key=lambda col: -col, axis=1)
+        expected = df.loc[:, ::-1]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_sort_values_key_casts_to_categorical(self, ordered):
+        # https://github.com/pandas-dev/pandas/issues/36383
+        categories = ["c", "b", "a"]
+        df = DataFrame({"x": [1, 1, 1], "y": ["a", "b", "c"]})
+
+        def sorter(key):
+            if key.name == "y":
+                return pd.Series(
+                    Categorical(key, categories=categories, ordered=ordered)
+                )
+            return key
+
+        result = df.sort_values(by=["x", "y"], key=sorter)
+        expected = DataFrame(
+            {"x": [1, 1, 1], "y": ["c", "b", "a"]}, index=pd.Index([2, 1, 0])
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.fixture
+def df_none():
+    return DataFrame(
+        {
+            "outer": ["a", "a", "a", "b", "b", "b"],
+            "inner": [1, 2, 2, 2, 1, 1],
+            "A": np.arange(6, 0, -1),
+            ("B", 5): ["one", "one", "two", "two", "one", "one"],
+        }
+    )
+
+
+@pytest.fixture(params=[["outer"], ["outer", "inner"]])
+def df_idx(request, df_none):
+    levels = request.param
+    return df_none.set_index(levels)
+
+
+@pytest.fixture(
+    params=[
+        "inner",  # index level
+        ["outer"],  # list of index level
+        "A",  # column
+        [("B", 5)],  # list of column
+        ["inner", "outer"],  # two index levels
+        [("B", 5), "outer"],  # index level and column
+        ["A", ("B", 5)],  # Two columns
+        ["inner", "outer"],  # two index levels and column
+    ]
+)
+def sort_names(request):
+    return request.param
+
+
+class TestSortValuesLevelAsStr:
+    def test_sort_index_level_and_column_label(
+        self, df_none, df_idx, sort_names, ascending, request
+    ):
+        # GH#14353
+        if (
+            Version(np.__version__) >= Version("1.25")
+            and request.node.callspec.id == "df_idx0-inner-True"
+        ):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=(
+                        "pandas default unstable sorting of duplicates"
+                        "issue with numpy>=1.25 with AVX instructions"
+                    ),
+                    strict=False,
+                )
+            )
+
+        # Get index levels from df_idx
+        levels = df_idx.index.names
+
+        # Compute expected by sorting on columns and the setting index
+        expected = df_none.sort_values(
+            by=sort_names, ascending=ascending, axis=0
+        ).set_index(levels)
+
+        # Compute result sorting on mix on columns and index levels
+        result = df_idx.sort_values(by=sort_names, ascending=ascending, axis=0)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_column_level_and_index_label(
+        self, df_none, df_idx, sort_names, ascending, request
+    ):
+        # GH#14353
+
+        # Get levels from df_idx
+        levels = df_idx.index.names
+
+        # Compute expected by sorting on axis=0, setting index levels, and then
+        # transposing. For some cases this will result in a frame with
+        # multiple column levels
+        expected = (
+            df_none.sort_values(by=sort_names, ascending=ascending, axis=0)
+            .set_index(levels)
+            .T
+        )
+
+        # Compute result by transposing and sorting on axis=1.
+        result = df_idx.T.sort_values(by=sort_names, ascending=ascending, axis=1)
+
+        if Version(np.__version__) >= Version("1.25"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=(
+                        "pandas default unstable sorting of duplicates"
+                        "issue with numpy>=1.25 with AVX instructions"
+                    ),
+                    strict=False,
+                )
+            )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_validate_ascending_for_value_error(self):
+        # GH41634
+        df = DataFrame({"D": [23, 7, 21]})
+
+        msg = 'For argument "ascending" expected type bool, received type str.'
+        with pytest.raises(ValueError, match=msg):
+            df.sort_values(by="D", ascending="False")
+
+    def test_sort_values_validate_ascending_functional(self, ascending):
+        df = DataFrame({"D": [23, 7, 21]})
+        indexer = df["D"].argsort().values
+
+        if not ascending:
+            indexer = indexer[::-1]
+
+        expected = df.loc[df.index[indexer]]
+        result = df.sort_values(by="D", ascending=ascending)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_swaplevel.py b/pandas/pandas/tests/frame/methods/test_swaplevel.py
new file mode 100644
index 0000000000000..5511ac7d6b1b2
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_swaplevel.py
@@ -0,0 +1,36 @@
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestSwaplevel:
+    def test_swaplevel(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        swapped = frame["A"].swaplevel()
+        swapped2 = frame["A"].swaplevel(0)
+        swapped3 = frame["A"].swaplevel(0, 1)
+        swapped4 = frame["A"].swaplevel("first", "second")
+        assert not swapped.index.equals(frame.index)
+        tm.assert_series_equal(swapped, swapped2)
+        tm.assert_series_equal(swapped, swapped3)
+        tm.assert_series_equal(swapped, swapped4)
+
+        back = swapped.swaplevel()
+        back2 = swapped.swaplevel(0)
+        back3 = swapped.swaplevel(0, 1)
+        back4 = swapped.swaplevel("second", "first")
+        assert back.index.equals(frame.index)
+        tm.assert_series_equal(back, back2)
+        tm.assert_series_equal(back, back3)
+        tm.assert_series_equal(back, back4)
+
+        ft = frame.T
+        swapped = ft.swaplevel("first", "second", axis=1)
+        exp = frame.swaplevel("first", "second").T
+        tm.assert_frame_equal(swapped, exp)
+
+        msg = "Can only swap levels on a hierarchical axis."
+        with pytest.raises(TypeError, match=msg):
+            DataFrame(range(3)).swaplevel()
diff --git a/pandas/pandas/tests/frame/methods/test_to_csv.py b/pandas/pandas/tests/frame/methods/test_to_csv.py
new file mode 100644
index 0000000000000..9eafc69013ffe
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_csv.py
@@ -0,0 +1,1452 @@
+import csv
+from io import StringIO
+import os
+
+import numpy as np
+import pytest
+
+from pandas.errors import ParserError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    read_csv,
+    to_datetime,
+)
+import pandas._testing as tm
+import pandas.core.common as com
+
+from pandas.io.common import get_handle
+
+
+class TestDataFrameToCSV:
+    def read_csv(self, path, **kwargs):
+        params = {"index_col": 0}
+        params.update(**kwargs)
+
+        return read_csv(path, **params)
+
+    def test_to_csv_from_csv1(self, temp_file, float_frame):
+        path = str(temp_file)
+        float_frame.iloc[:5, float_frame.columns.get_loc("A")] = np.nan
+
+        float_frame.to_csv(path)
+        float_frame.to_csv(path, columns=["A", "B"])
+        float_frame.to_csv(path, header=False)
+        float_frame.to_csv(path, index=False)
+
+    def test_to_csv_from_csv1_datetime(self, temp_file, datetime_frame):
+        path = str(temp_file)
+        # test roundtrip
+        # freq does not roundtrip
+        datetime_frame.index = datetime_frame.index._with_freq(None)
+        datetime_frame.to_csv(path)
+        recons = self.read_csv(path, parse_dates=True)
+        expected = datetime_frame.copy()
+        expected.index = expected.index.as_unit("s")
+        tm.assert_frame_equal(expected, recons)
+
+        datetime_frame.to_csv(path, index_label="index")
+        recons = self.read_csv(path, index_col=None, parse_dates=True)
+
+        assert len(recons.columns) == len(datetime_frame.columns) + 1
+
+        # no index
+        datetime_frame.to_csv(path, index=False)
+        recons = self.read_csv(path, index_col=None, parse_dates=True)
+        tm.assert_almost_equal(datetime_frame.values, recons.values)
+
+    def test_to_csv_from_csv1_corner_case(self, temp_file):
+        path = str(temp_file)
+        dm = DataFrame(
+            {
+                "s1": Series(range(3), index=np.arange(3, dtype=np.int64)),
+                "s2": Series(range(2), index=np.arange(2, dtype=np.int64)),
+            }
+        )
+        dm.to_csv(path)
+
+        recons = self.read_csv(path)
+        tm.assert_frame_equal(dm, recons)
+
+    def test_to_csv_from_csv2(self, temp_file, float_frame):
+        path = str(temp_file)
+        # duplicate index
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=["a", "a", "b"],
+            columns=["x", "y", "z"],
+        )
+        df.to_csv(path)
+        result = self.read_csv(path)
+        tm.assert_frame_equal(result, df)
+
+        midx = MultiIndex.from_tuples([("A", 1, 2), ("A", 1, 2), ("B", 1, 2)])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=midx,
+            columns=["x", "y", "z"],
+        )
+
+        df.to_csv(path)
+        result = self.read_csv(path, index_col=[0, 1, 2], parse_dates=False)
+        tm.assert_frame_equal(result, df, check_names=False)
+
+        # column aliases
+        col_aliases = Index(["AA", "X", "Y", "Z"])
+        float_frame.to_csv(path, header=col_aliases)
+
+        rs = self.read_csv(path)
+        xp = float_frame.copy()
+        xp.columns = col_aliases
+        tm.assert_frame_equal(xp, rs)
+
+        msg = "Writing 4 cols but got 2 aliases"
+        with pytest.raises(ValueError, match=msg):
+            float_frame.to_csv(path, header=["AA", "X"])
+
+    def test_to_csv_from_csv3(self, temp_file):
+        path = str(temp_file)
+        df1 = DataFrame(np.random.default_rng(2).standard_normal((3, 1)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((3, 1)))
+
+        df1.to_csv(path)
+        df2.to_csv(path, mode="a", header=False)
+        xp = pd.concat([df1, df2])
+        rs = read_csv(path, index_col=0)
+        rs.columns = [int(label) for label in rs.columns]
+        xp.columns = [int(label) for label in xp.columns]
+        tm.assert_frame_equal(xp, rs)
+
+    def test_to_csv_from_csv4(self, temp_file):
+        path = str(temp_file)
+        # GH 10833 (TimedeltaIndex formatting)
+        dt = pd.Timedelta(seconds=1)
+        df = DataFrame(
+            {"dt_data": [i * dt for i in range(3)]},
+            index=Index([i * dt for i in range(3)], name="dt_index"),
+        )
+        df.to_csv(path)
+
+        result = read_csv(path, index_col="dt_index")
+        result.index = pd.to_timedelta(result.index)
+        result["dt_data"] = pd.to_timedelta(result["dt_data"])
+
+        tm.assert_frame_equal(df, result, check_index_type=True)
+
+    def test_to_csv_from_csv5(self, temp_file, timezone_frame):
+        # tz, 8260
+        path = str(temp_file)
+        timezone_frame.to_csv(path)
+        result = read_csv(path, index_col=0, parse_dates=["A"])
+
+        converter = (
+            lambda c: to_datetime(result[c])
+            .dt.tz_convert("UTC")
+            .dt.tz_convert(timezone_frame[c].dt.tz)
+            .dt.as_unit("ns")
+        )
+        result["B"] = converter("B")
+        result["C"] = converter("C")
+        result["A"] = result["A"].dt.as_unit("ns")
+        tm.assert_frame_equal(result, timezone_frame)
+
+    def test_to_csv_cols_reordering(self, temp_file):
+        # GH3454
+        chunksize = 5
+        N = int(chunksize * 2.5)
+
+        df = DataFrame(
+            np.ones((N, 3)),
+            index=Index([f"i-{i}" for i in range(N)], name="a"),
+            columns=Index([f"i-{i}" for i in range(3)], name="a"),
+        )
+        cs = df.columns
+        cols = [cs[2], cs[0]]
+
+        path = str(temp_file)
+        df.to_csv(path, columns=cols, chunksize=chunksize)
+        rs_c = read_csv(path, index_col=0)
+
+        tm.assert_frame_equal(df[cols], rs_c, check_names=False)
+
+    @pytest.mark.parametrize("cols", [None, ["b", "a"]])
+    def test_to_csv_new_dupe_cols(self, temp_file, cols):
+        chunksize = 5
+        N = int(chunksize * 2.5)
+
+        # dupe cols
+        df = DataFrame(
+            np.ones((N, 3)),
+            index=Index([f"i-{i}" for i in range(N)], name="a"),
+            columns=["a", "a", "b"],
+        )
+        path = str(temp_file)
+        df.to_csv(path, columns=cols, chunksize=chunksize)
+        rs_c = read_csv(path, index_col=0)
+
+        # we wrote them in a different order
+        # so compare them in that order
+        if cols is not None:
+            if df.columns.is_unique:
+                rs_c.columns = cols
+            else:
+                indexer, missing = df.columns.get_indexer_non_unique(cols)
+                rs_c.columns = df.columns.take(indexer)
+
+            for c in cols:
+                obj_df = df[c]
+                obj_rs = rs_c[c]
+                if isinstance(obj_df, Series):
+                    tm.assert_series_equal(obj_df, obj_rs)
+                else:
+                    tm.assert_frame_equal(obj_df, obj_rs, check_names=False)
+
+        # wrote in the same order
+        else:
+            rs_c.columns = df.columns
+            tm.assert_frame_equal(df, rs_c, check_names=False)
+
+    @pytest.mark.slow
+    def test_to_csv_dtnat(self, temp_file):
+        # GH3437
+        def make_dtnat_arr(n, nnat=None):
+            if nnat is None:
+                nnat = int(n * 0.1)  # 10%
+            s = list(date_range("2000", freq="5min", periods=n))
+            if nnat:
+                for i in np.random.default_rng(2).integers(0, len(s), nnat):
+                    s[i] = NaT
+                i = np.random.default_rng(2).integers(100)
+                s[-i] = NaT
+                s[i] = NaT
+            return s
+
+        chunksize = 1000
+        s1 = make_dtnat_arr(chunksize + 5)
+        s2 = make_dtnat_arr(chunksize + 5, 0)
+
+        path = str(temp_file)
+        df = DataFrame({"a": s1, "b": s2})
+        df.to_csv(path, chunksize=chunksize)
+
+        result = self.read_csv(path).apply(to_datetime)
+
+        expected = df[:]
+        expected["a"] = expected["a"].astype("M8[s]")
+        expected["b"] = expected["b"].astype("M8[s]")
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    def _return_result_expected(
+        self,
+        df,
+        chunksize,
+        r_dtype=None,
+        c_dtype=None,
+        rnlvl=None,
+        cnlvl=None,
+        dupe_col=False,
+    ):
+        kwargs = {"parse_dates": False}
+        if cnlvl:
+            if rnlvl is not None:
+                kwargs["index_col"] = list(range(rnlvl))
+            kwargs["header"] = list(range(cnlvl))
+
+            with tm.ensure_clean("__tmp_to_csv_moar__") as path:
+                df.to_csv(path, encoding="utf8", chunksize=chunksize)
+                recons = self.read_csv(path, **kwargs)
+        else:
+            kwargs["header"] = 0
+
+            with tm.ensure_clean("__tmp_to_csv_moar__") as path:
+                df.to_csv(path, encoding="utf8", chunksize=chunksize)
+                recons = self.read_csv(path, **kwargs)
+
+        def _to_uni(x):
+            if not isinstance(x, str):
+                return x.decode("utf8")
+            return x
+
+        if dupe_col:
+            # read_Csv disambiguates the columns by
+            # labeling them dupe.1,dupe.2, etc'. monkey patch columns
+            recons.columns = df.columns
+        if rnlvl and not cnlvl:
+            delta_lvl = [recons.iloc[:, i].values for i in range(rnlvl - 1)]
+            ix = MultiIndex.from_arrays([list(recons.index)] + delta_lvl)
+            recons.index = ix
+            recons = recons.iloc[:, rnlvl - 1 :]
+
+        type_map = {"i": "i", "f": "f", "s": "O", "u": "O", "dt": "O", "p": "O"}
+        if r_dtype:
+            if r_dtype == "u":  # unicode
+                r_dtype = "O"
+                recons.index = np.array(
+                    [_to_uni(label) for label in recons.index], dtype=r_dtype
+                )
+                df.index = np.array(
+                    [_to_uni(label) for label in df.index], dtype=r_dtype
+                )
+            elif r_dtype == "dt":  # unicode
+                r_dtype = "O"
+                recons.index = np.array(
+                    [Timestamp(label) for label in recons.index], dtype=r_dtype
+                )
+                df.index = np.array(
+                    [Timestamp(label) for label in df.index], dtype=r_dtype
+                )
+            elif r_dtype == "p":
+                r_dtype = "O"
+                idx_list = to_datetime(recons.index)
+                recons.index = np.array(
+                    [Timestamp(label) for label in idx_list], dtype=r_dtype
+                )
+                df.index = np.array(
+                    list(map(Timestamp, df.index.to_timestamp())), dtype=r_dtype
+                )
+            else:
+                r_dtype = type_map.get(r_dtype)
+                recons.index = np.array(recons.index, dtype=r_dtype)
+                df.index = np.array(df.index, dtype=r_dtype)
+        if c_dtype:
+            if c_dtype == "u":
+                c_dtype = "O"
+                recons.columns = np.array(
+                    [_to_uni(label) for label in recons.columns], dtype=c_dtype
+                )
+                df.columns = np.array(
+                    [_to_uni(label) for label in df.columns], dtype=c_dtype
+                )
+            elif c_dtype == "dt":
+                c_dtype = "O"
+                recons.columns = np.array(
+                    [Timestamp(label) for label in recons.columns], dtype=c_dtype
+                )
+                df.columns = np.array(
+                    [Timestamp(label) for label in df.columns], dtype=c_dtype
+                )
+            elif c_dtype == "p":
+                c_dtype = "O"
+                col_list = to_datetime(recons.columns)
+                recons.columns = np.array(
+                    [Timestamp(label) for label in col_list], dtype=c_dtype
+                )
+                col_list = df.columns.to_timestamp()
+                df.columns = np.array(
+                    [Timestamp(label) for label in col_list], dtype=c_dtype
+                )
+            else:
+                c_dtype = type_map.get(c_dtype)
+                recons.columns = np.array(recons.columns, dtype=c_dtype)
+                df.columns = np.array(df.columns, dtype=c_dtype)
+        return df, recons
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "nrows", [2, 10, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
+    )
+    def test_to_csv_nrows(self, nrows):
+        df = DataFrame(
+            np.ones((nrows, 4)),
+            index=date_range("2020-01-01", periods=nrows),
+            columns=Index(list("abcd"), dtype=object),
+        )
+        result, expected = self._return_result_expected(df, 1000, "dt", "s")
+        expected.index = expected.index.astype("M8[ns]")
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "nrows", [2, 10, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
+    )
+    @pytest.mark.parametrize(
+        "r_idx_type, c_idx_type", [("i", "i"), ("s", "s"), ("s", "dt"), ("p", "p")]
+    )
+    @pytest.mark.parametrize("ncols", [1, 2, 3, 4])
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_to_csv_idx_types(self, nrows, r_idx_type, c_idx_type, ncols):
+        axes = {
+            "i": lambda n: Index(np.arange(n), dtype=np.int64),
+            "s": lambda n: Index([f"{i}_{chr(i)}" for i in range(97, 97 + n)]),
+            "dt": lambda n: date_range("2020-01-01", periods=n),
+            "p": lambda n: period_range("2020-01-01", periods=n, freq="D"),
+        }
+        df = DataFrame(
+            np.ones((nrows, ncols)),
+            index=axes[r_idx_type](nrows),
+            columns=axes[c_idx_type](ncols),
+        )
+        result, expected = self._return_result_expected(
+            df,
+            1000,
+            r_idx_type,
+            c_idx_type,
+        )
+        if r_idx_type in ["dt", "p"]:
+            expected.index = expected.index.astype("M8[ns]")
+        if c_idx_type in ["dt", "p"]:
+            expected.columns = expected.columns.astype("M8[ns]")
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "nrows", [10, 98, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
+    )
+    @pytest.mark.parametrize("ncols", [1, 2, 3, 4])
+    def test_to_csv_idx_ncols(self, nrows, ncols):
+        df = DataFrame(
+            np.ones((nrows, ncols)),
+            index=Index([f"i-{i}" for i in range(nrows)], name="a"),
+            columns=Index([f"i-{i}" for i in range(ncols)], name="a"),
+        )
+        result, expected = self._return_result_expected(df, 1000)
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("nrows", [10, 98, 99, 100, 101, 102])
+    def test_to_csv_dup_cols(self, nrows):
+        df = DataFrame(
+            np.ones((nrows, 3)),
+            index=Index([f"i-{i}" for i in range(nrows)], name="a"),
+            columns=Index([f"i-{i}" for i in range(3)], name="a"),
+        )
+
+        cols = list(df.columns)
+        cols[:2] = ["dupe", "dupe"]
+        cols[-2:] = ["dupe", "dupe"]
+        ix = list(df.index)
+        ix[:2] = ["rdupe", "rdupe"]
+        ix[-2:] = ["rdupe", "rdupe"]
+        df.index = ix
+        df.columns = cols
+        result, expected = self._return_result_expected(df, 1000, dupe_col=True)
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    @pytest.mark.slow
+    def test_to_csv_empty(self):
+        df = DataFrame(index=np.arange(10, dtype=np.int64))
+        result, expected = self._return_result_expected(df, 1000)
+        tm.assert_frame_equal(result, expected, check_column_type=False)
+
+    @pytest.mark.slow
+    def test_to_csv_chunksize(self):
+        chunksize = 1000
+        rows = chunksize // 2 + 1
+        df = DataFrame(
+            np.ones((rows, 2)),
+            columns=Index(list("ab")),
+            index=MultiIndex.from_arrays([range(rows) for _ in range(2)]),
+        )
+        result, expected = self._return_result_expected(df, chunksize, rnlvl=2)
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "nrows", [2, 10, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
+    )
+    @pytest.mark.parametrize("ncols", [2, 3, 4])
+    @pytest.mark.parametrize(
+        "df_params, func_params",
+        [
+            [{"r_idx_nlevels": 2}, {"rnlvl": 2}],
+            [{"c_idx_nlevels": 2}, {"cnlvl": 2}],
+            [{"r_idx_nlevels": 2, "c_idx_nlevels": 2}, {"rnlvl": 2, "cnlvl": 2}],
+        ],
+    )
+    def test_to_csv_params(self, nrows, df_params, func_params, ncols):
+        if df_params.get("r_idx_nlevels"):
+            index = MultiIndex.from_arrays(
+                [f"i-{i}" for i in range(nrows)]
+                for _ in range(df_params["r_idx_nlevels"])
+            )
+        else:
+            index = None
+
+        if df_params.get("c_idx_nlevels"):
+            columns = MultiIndex.from_arrays(
+                [f"i-{i}" for i in range(ncols)]
+                for _ in range(df_params["c_idx_nlevels"])
+            )
+        else:
+            columns = Index([f"i-{i}" for i in range(ncols)])
+        df = DataFrame(np.ones((nrows, ncols)), index=index, columns=columns)
+        result, expected = self._return_result_expected(df, 1000, **func_params)
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    def test_to_csv_from_csv_w_some_infs(self, temp_file, float_frame):
+        # test roundtrip with inf, -inf, nan, as full columns and mix
+        float_frame["G"] = np.nan
+        f = lambda x: [np.inf, np.nan][np.random.default_rng(2).random() < 0.5]
+        float_frame["h"] = float_frame.index.map(f)
+
+        path = str(temp_file)
+        float_frame.to_csv(path)
+        recons = self.read_csv(path)
+
+        tm.assert_frame_equal(float_frame, recons)
+        tm.assert_frame_equal(np.isinf(float_frame), np.isinf(recons))
+
+    def test_to_csv_from_csv_w_all_infs(self, temp_file, float_frame):
+        # test roundtrip with inf, -inf, nan, as full columns and mix
+        float_frame["E"] = np.inf
+        float_frame["F"] = -np.inf
+
+        path = str(temp_file)
+        float_frame.to_csv(path)
+        recons = self.read_csv(path)
+
+        tm.assert_frame_equal(float_frame, recons)
+        tm.assert_frame_equal(np.isinf(float_frame), np.isinf(recons))
+
+    def test_to_csv_no_index(self, temp_file):
+        # GH 3624, after appending columns, to_csv fails
+        path = str(temp_file)
+        df = DataFrame({"c1": [1, 2, 3], "c2": [4, 5, 6]})
+        df.to_csv(path, index=False)
+        result = read_csv(path)
+        tm.assert_frame_equal(df, result)
+        df["c3"] = Series([7, 8, 9], dtype="int64")
+        df.to_csv(path, index=False)
+        result = read_csv(path)
+        tm.assert_frame_equal(df, result)
+
+    def test_to_csv_with_mix_columns(self):
+        # gh-11637: incorrect output when a mix of integer and string column
+        # names passed as columns parameter in to_csv
+
+        df = DataFrame({0: ["a", "b", "c"], 1: ["aa", "bb", "cc"]})
+        df["test"] = "txt"
+        assert df.to_csv() == df.to_csv(columns=[0, 1, "test"])
+
+    def test_to_csv_headers(self, temp_file):
+        # GH6186, the presence or absence of `index` incorrectly
+        # causes to_csv to have different header semantics.
+        from_df = DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+        to_df = DataFrame([[1, 2], [3, 4]], columns=["X", "Y"])
+        path = str(temp_file)
+        from_df.to_csv(path, header=["X", "Y"])
+        recons = self.read_csv(path)
+
+        tm.assert_frame_equal(to_df, recons)
+
+        from_df.to_csv(path, index=False, header=["X", "Y"])
+        recons = self.read_csv(path)
+
+        return_value = recons.reset_index(inplace=True)
+        assert return_value is None
+        tm.assert_frame_equal(to_df, recons)
+
+    def test_to_csv_multiindex(self, temp_file, float_frame, datetime_frame):
+        frame = float_frame
+        old_index = frame.index
+        arrays = np.arange(len(old_index) * 2, dtype=np.int64).reshape(2, -1)
+        new_index = MultiIndex.from_arrays(arrays, names=["first", "second"])
+        frame.index = new_index
+
+        path = str(temp_file)
+        frame.to_csv(path, header=False)
+        frame.to_csv(path, columns=["A", "B"])
+
+        # round trip
+        frame.to_csv(path)
+
+        df = self.read_csv(path, index_col=[0, 1], parse_dates=False)
+
+        # TODO to_csv drops column name
+        tm.assert_frame_equal(frame, df, check_names=False)
+        assert frame.index.names == df.index.names
+
+        # needed if setUp becomes a class method
+        float_frame.index = old_index
+
+        # try multiindex with dates
+        tsframe = datetime_frame
+        old_index = tsframe.index
+        new_index = [old_index, np.arange(len(old_index), dtype=np.int64)]
+        tsframe.index = MultiIndex.from_arrays(new_index)
+
+        tsframe.to_csv(path, index_label=["time", "foo"])
+        with tm.assert_produces_warning(UserWarning, match="Could not infer format"):
+            recons = self.read_csv(path, index_col=[0, 1], parse_dates=True)
+
+        # TODO to_csv drops column name
+        expected = tsframe.copy()
+        expected.index = MultiIndex.from_arrays([old_index.as_unit("s"), new_index[1]])
+        tm.assert_frame_equal(recons, expected, check_names=False)
+
+        # do not load index
+        tsframe.to_csv(path)
+        recons = self.read_csv(path, index_col=None)
+        assert len(recons.columns) == len(tsframe.columns) + 2
+
+        # no index
+        tsframe.to_csv(path, index=False)
+        recons = self.read_csv(path, index_col=None)
+        tm.assert_almost_equal(recons.values, datetime_frame.values)
+
+        # needed if setUp becomes class method
+        datetime_frame.index = old_index
+
+        with tm.ensure_clean("__tmp_to_csv_multiindex__") as path:
+            # GH3571, GH1651, GH3141
+
+            def _make_frame(names=None):
+                if names is True:
+                    names = ["first", "second"]
+                return DataFrame(
+                    np.random.default_rng(2).integers(0, 10, size=(3, 3)),
+                    columns=MultiIndex.from_tuples(
+                        [("bah", "foo"), ("bah", "bar"), ("ban", "baz")], names=names
+                    ),
+                    dtype="int64",
+                )
+
+            # column & index are multi-index
+            df = DataFrame(
+                np.ones((5, 3)),
+                columns=MultiIndex.from_arrays(
+                    [[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
+                ),
+                index=MultiIndex.from_arrays(
+                    [[f"i-{i}" for i in range(5)] for _ in range(2)], names=list("ab")
+                ),
+            )
+            df.to_csv(path)
+            result = read_csv(path, header=[0, 1, 2, 3], index_col=[0, 1])
+            tm.assert_frame_equal(df, result)
+
+            # column is mi
+            df = DataFrame(
+                np.ones((5, 3)),
+                columns=MultiIndex.from_arrays(
+                    [[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
+                ),
+            )
+            df.to_csv(path)
+            result = read_csv(path, header=[0, 1, 2, 3], index_col=0)
+            tm.assert_frame_equal(df, result)
+
+            # dup column names?
+            df = DataFrame(
+                np.ones((5, 3)),
+                columns=MultiIndex.from_arrays(
+                    [[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
+                ),
+                index=MultiIndex.from_arrays(
+                    [[f"i-{i}" for i in range(5)] for _ in range(3)], names=list("abc")
+                ),
+            )
+            df.to_csv(path)
+            result = read_csv(path, header=[0, 1, 2, 3], index_col=[0, 1, 2])
+            tm.assert_frame_equal(df, result)
+
+            # writing with no index
+            df = _make_frame()
+            df.to_csv(path, index=False)
+            result = read_csv(path, header=[0, 1])
+            tm.assert_frame_equal(df, result)
+
+            # we lose the names here
+            df = _make_frame(True)
+            df.to_csv(path, index=False)
+            result = read_csv(path, header=[0, 1])
+            assert com.all_none(*result.columns.names)
+            result.columns.names = df.columns.names
+            tm.assert_frame_equal(df, result)
+
+            # whatsnew example
+            df = _make_frame()
+            df.to_csv(path)
+            result = read_csv(path, header=[0, 1], index_col=[0])
+            tm.assert_frame_equal(df, result)
+
+            df = _make_frame(True)
+            df.to_csv(path)
+            result = read_csv(path, header=[0, 1], index_col=[0])
+            tm.assert_frame_equal(df, result)
+
+            # invalid options
+            df = _make_frame(True)
+            df.to_csv(path)
+
+            for i in [6, 7]:
+                msg = f"len of {i}, but only 5 lines in file"
+                with pytest.raises(ParserError, match=msg):
+                    read_csv(path, header=list(range(i)), index_col=0)
+
+            # write with cols
+            msg = "cannot specify cols with a MultiIndex"
+            with pytest.raises(TypeError, match=msg):
+                df.to_csv(path, columns=["foo", "bar"])
+
+        with tm.ensure_clean("__tmp_to_csv_multiindex__") as path:
+            # empty
+            tsframe[:0].to_csv(path)
+            recons = self.read_csv(path)
+
+            exp = tsframe[:0]
+            exp.index = []
+
+            tm.assert_index_equal(recons.columns, exp.columns)
+            assert len(recons) == 0
+
+    def test_to_csv_interval_index(self, temp_file, using_infer_string):
+        # GH 28210
+        df = DataFrame({"A": list("abc"), "B": range(3)}, index=pd.interval_range(0, 3))
+
+        path = str(temp_file)
+        df.to_csv(path)
+        result = self.read_csv(path, index_col=0)
+
+        # can't roundtrip intervalindex via read_csv so check string repr (GH 23595)
+        expected = df.copy()
+        expected.index = expected.index.astype("str")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_csv_float32_nanrep(self, temp_file):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1, 4)).astype(np.float32)
+        )
+        df[1] = np.nan
+
+        path = str(temp_file)
+        df.to_csv(path, na_rep=999)
+
+        with open(path, encoding="utf-8") as f:
+            lines = f.readlines()
+            assert lines[1].split(",")[2] == "999"
+
+    def test_to_csv_withcommas(self, temp_file):
+        # Commas inside fields should be correctly escaped when saving as CSV.
+        df = DataFrame({"A": [1, 2, 3], "B": ["5,6", "7,8", "9,0"]})
+
+        path = str(temp_file)
+        df.to_csv(path)
+        df2 = self.read_csv(path)
+        tm.assert_frame_equal(df2, df)
+
+    def test_to_csv_mixed(self, temp_file):
+        def create_cols(name):
+            return [f"{name}{i:03d}" for i in range(5)]
+
+        df_float = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 5)),
+            dtype="float64",
+            columns=create_cols("float"),
+        )
+        df_int = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 5)).astype("int64"),
+            dtype="int64",
+            columns=create_cols("int"),
+        )
+        df_bool = DataFrame(True, index=df_float.index, columns=create_cols("bool"))
+        df_object = DataFrame(
+            "foo", index=df_float.index, columns=create_cols("object"), dtype="object"
+        )
+        df_dt = DataFrame(
+            Timestamp("20010101"),
+            index=df_float.index,
+            columns=create_cols("date"),
+        )
+
+        # add in some nans
+        df_float.iloc[30:50, 1:3] = np.nan
+        df_dt.iloc[30:50, 1:3] = np.nan
+
+        df = pd.concat([df_float, df_int, df_bool, df_object, df_dt], axis=1)
+
+        # dtype
+        dtypes = {}
+        for n, dtype in [
+            ("float", np.float64),
+            ("int", np.int64),
+            ("bool", np.bool_),
+            ("object", object),
+        ]:
+            for c in create_cols(n):
+                dtypes[c] = dtype
+
+        path = str(temp_file)
+        df.to_csv(path)
+        rs = read_csv(path, index_col=0, dtype=dtypes, parse_dates=create_cols("date"))
+        tm.assert_frame_equal(rs, df)
+
+    def test_to_csv_dups_cols(self, temp_file):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 30)),
+            columns=list(range(15)) + list(range(15)),
+            dtype="float64",
+        )
+
+        path = str(temp_file)
+        df.to_csv(path)  # single dtype, fine
+        result = read_csv(path, index_col=0)
+        result.columns = df.columns
+        tm.assert_frame_equal(result, df)
+
+        df_float = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 3)), dtype="float64"
+        )
+        df_int = DataFrame(np.random.default_rng(2).standard_normal((1000, 3))).astype(
+            "int64"
+        )
+        df_bool = DataFrame(True, index=df_float.index, columns=range(3))
+        df_object = DataFrame("foo", index=df_float.index, columns=range(3))
+        df_dt = DataFrame(Timestamp("20010101"), index=df_float.index, columns=range(3))
+        df = pd.concat(
+            [df_float, df_int, df_bool, df_object, df_dt], axis=1, ignore_index=True
+        )
+
+        df.columns = [0, 1, 2] * 5
+
+        with tm.ensure_clean() as filename:
+            df.to_csv(filename)
+            result = read_csv(filename, index_col=0)
+
+            # date cols
+            for i in ["0.4", "1.4", "2.4"]:
+                result[i] = to_datetime(result[i])
+
+            result.columns = df.columns
+            tm.assert_frame_equal(result, df)
+
+    def test_to_csv_dups_cols2(self, temp_file):
+        # GH3457
+        df = DataFrame(
+            np.ones((5, 3)),
+            index=Index([f"i-{i}" for i in range(5)], name="foo"),
+            columns=Index(["a", "a", "b"]),
+        )
+
+        path = str(temp_file)
+        df.to_csv(path)
+
+        # read_csv will rename the dups columns
+        result = read_csv(path, index_col=0)
+        result = result.rename(columns={"a.1": "a"})
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize("chunksize", [10000, 50000, 100000])
+    def test_to_csv_chunking(self, chunksize, temp_file):
+        aa = DataFrame({"A": range(100000)})
+        aa["B"] = aa.A + 1.0
+        aa["C"] = aa.A + 2.0
+        aa["D"] = aa.A + 3.0
+
+        path = str(temp_file)
+        aa.to_csv(path, chunksize=chunksize)
+        rs = read_csv(path, index_col=0)
+        tm.assert_frame_equal(rs, aa)
+
+    @pytest.mark.slow
+    def test_to_csv_wide_frame_formatting(self, temp_file, monkeypatch):
+        # Issue #8621
+        chunksize = 100
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1, chunksize + 10)),
+            columns=None,
+            index=None,
+        )
+        path = str(temp_file)
+        with monkeypatch.context() as m:
+            m.setattr("pandas.io.formats.csvs._DEFAULT_CHUNKSIZE_CELLS", chunksize)
+            df.to_csv(path, header=False, index=False)
+        rs = read_csv(path, header=None)
+        tm.assert_frame_equal(rs, df)
+
+    def test_to_csv_bug(self, temp_file):
+        f1 = StringIO("a,1.0\nb,2.0")
+        df = self.read_csv(f1, header=None)
+        newdf = DataFrame({"t": df[df.columns[0]]})
+
+        path = str(temp_file)
+        newdf.to_csv(path)
+
+        recons = read_csv(path, index_col=0)
+        # don't check_names as t != 1
+        tm.assert_frame_equal(recons, newdf, check_names=False)
+
+    def test_to_csv_unicode(self, temp_file):
+        df = DataFrame({"c/\u03c3": [1, 2, 3]})
+        path = str(temp_file)
+        df.to_csv(path, encoding="UTF-8")
+        df2 = read_csv(path, index_col=0, encoding="UTF-8")
+        tm.assert_frame_equal(df, df2)
+
+        df.to_csv(path, encoding="UTF-8", index=False)
+        df2 = read_csv(path, index_col=None, encoding="UTF-8")
+        tm.assert_frame_equal(df, df2)
+
+    def test_to_csv_unicode_index_col(self):
+        buf = StringIO("")
+        df = DataFrame(
+            [["\u05d0", "d2", "d3", "d4"], ["a1", "a2", "a3", "a4"]],
+            columns=["\u05d0", "\u05d1", "\u05d2", "\u05d3"],
+            index=["\u05d0", "\u05d1"],
+        )
+
+        df.to_csv(buf, encoding="UTF-8")
+        buf.seek(0)
+
+        df2 = read_csv(buf, index_col=0, encoding="UTF-8")
+        tm.assert_frame_equal(df, df2)
+
+    def test_to_csv_stringio(self, float_frame):
+        buf = StringIO()
+        float_frame.to_csv(buf)
+        buf.seek(0)
+        recons = read_csv(buf, index_col=0)
+        tm.assert_frame_equal(recons, float_frame)
+
+    def test_to_csv_float_format(self, temp_file):
+        df = DataFrame(
+            [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+
+        path = str(temp_file)
+        df.to_csv(path, float_format="%.2f")
+
+        rs = read_csv(path, index_col=0)
+        xp = DataFrame(
+            [[0.12, 0.23, 0.57], [12.32, 123123.20, 321321.20]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        tm.assert_frame_equal(rs, xp)
+
+    def test_to_csv_float_format_over_decimal(self):
+        # GH#47436
+        df = DataFrame({"a": [0.5, 1.0]})
+        result = df.to_csv(
+            decimal=",",
+            float_format=lambda x: np.format_float_positional(x, trim="-"),
+            index=False,
+        )
+        expected_rows = ["a", "0.5", "1"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_to_csv_unicodewriter_quoting(self):
+        df = DataFrame({"A": [1, 2, 3], "B": ["foo", "bar", "baz"]})
+
+        buf = StringIO()
+        df.to_csv(buf, index=False, quoting=csv.QUOTE_NONNUMERIC, encoding="utf-8")
+
+        result = buf.getvalue()
+        expected_rows = ['"A","B"', '1,"foo"', '2,"bar"', '3,"baz"']
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    @pytest.mark.parametrize("encoding", [None, "utf-8"])
+    def test_to_csv_quote_none(self, encoding):
+        # GH4328
+        df = DataFrame({"A": ["hello", '{"hello"}']})
+        buf = StringIO()
+        df.to_csv(buf, quoting=csv.QUOTE_NONE, encoding=encoding, index=False)
+
+        result = buf.getvalue()
+        expected_rows = ["A", "hello", '{"hello"}']
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_to_csv_index_no_leading_comma(self):
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["one", "two", "three"])
+
+        buf = StringIO()
+        df.to_csv(buf, index_label=False)
+
+        expected_rows = ["A,B", "one,1,4", "two,2,5", "three,3,6"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert buf.getvalue() == expected
+
+    def test_to_csv_lineterminators(self, temp_file):
+        # see gh-20353
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["one", "two", "three"])
+
+        path = str(temp_file)
+        # case 1: CRLF as line terminator
+        df.to_csv(path, lineterminator="\r\n")
+        expected = b",A,B\r\none,1,4\r\ntwo,2,5\r\nthree,3,6\r\n"
+
+        with open(path, mode="rb") as f:
+            assert f.read() == expected
+
+    def test_to_csv_lineterminators2(self, temp_file):
+        # see gh-20353
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["one", "two", "three"])
+
+        path = str(temp_file)
+        # case 2: LF as line terminator
+        df.to_csv(path, lineterminator="\n")
+        expected = b",A,B\none,1,4\ntwo,2,5\nthree,3,6\n"
+
+        with open(path, mode="rb") as f:
+            assert f.read() == expected
+
+    def test_to_csv_lineterminators3(self, temp_file):
+        # see gh-20353
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["one", "two", "three"])
+        path = str(temp_file)
+        # case 3: The default line terminator(=os.linesep)(gh-21406)
+        df.to_csv(path)
+        os_linesep = os.linesep.encode("utf-8")
+        expected = (
+            b",A,B"
+            + os_linesep
+            + b"one,1,4"
+            + os_linesep
+            + b"two,2,5"
+            + os_linesep
+            + b"three,3,6"
+            + os_linesep
+        )
+
+        with open(path, mode="rb") as f:
+            assert f.read() == expected
+
+    def test_to_csv_from_csv_categorical(self):
+        # CSV with categoricals should result in the same output
+        # as when one would add a "normal" Series/DataFrame.
+        s = Series(pd.Categorical(["a", "b", "b", "a", "a", "c", "c", "c"]))
+        s2 = Series(["a", "b", "b", "a", "a", "c", "c", "c"])
+        res = StringIO()
+
+        s.to_csv(res, header=False)
+        exp = StringIO()
+
+        s2.to_csv(exp, header=False)
+        assert res.getvalue() == exp.getvalue()
+
+        df = DataFrame({"s": s})
+        df2 = DataFrame({"s": s2})
+
+        res = StringIO()
+        df.to_csv(res)
+
+        exp = StringIO()
+        df2.to_csv(exp)
+
+        assert res.getvalue() == exp.getvalue()
+
+    def test_to_csv_path_is_none(self, float_frame):
+        # GH 8215
+        # Make sure we return string for consistency with
+        # Series.to_csv()
+        csv_str = float_frame.to_csv(path_or_buf=None)
+        assert isinstance(csv_str, str)
+        recons = read_csv(StringIO(csv_str), index_col=0)
+        tm.assert_frame_equal(float_frame, recons)
+
+    @pytest.mark.parametrize(
+        "df,encoding",
+        [
+            (
+                DataFrame(
+                    [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+                    index=["A", "B"],
+                    columns=["X", "Y", "Z"],
+                ),
+                None,
+            ),
+            # GH 21241, 21118
+            (DataFrame([["abc", "def", "ghi"]], columns=["X", "Y", "Z"]), "ascii"),
+            (DataFrame(5 * [[123, "你好", "世界"]], columns=["X", "Y", "Z"]), "gb2312"),
+            (
+                DataFrame(
+                    5 * [[123, "Γειά σου", "Κόσμε"]],  # noqa: RUF001
+                    columns=["X", "Y", "Z"],
+                ),
+                "cp737",
+            ),
+        ],
+    )
+    def test_to_csv_compression(self, temp_file, df, encoding, compression):
+        path = str(temp_file)
+        df.to_csv(path, compression=compression, encoding=encoding)
+        # test the round trip - to_csv -> read_csv
+        result = read_csv(path, compression=compression, index_col=0, encoding=encoding)
+        tm.assert_frame_equal(df, result)
+
+        # test the round trip using file handle - to_csv -> read_csv
+        with get_handle(
+            path, "w", compression=compression, encoding=encoding
+        ) as handles:
+            df.to_csv(handles.handle, encoding=encoding)
+            assert not handles.handle.closed
+
+        result = read_csv(
+            path,
+            compression=compression,
+            encoding=encoding,
+            index_col=0,
+        ).squeeze("columns")
+        tm.assert_frame_equal(df, result)
+
+        # explicitly make sure file is compressed
+        with tm.decompress_file(path, compression) as fh:
+            text = fh.read().decode(encoding or "utf8")
+            for col in df.columns:
+                assert col in text
+
+        with tm.decompress_file(path, compression) as fh:
+            tm.assert_frame_equal(df, read_csv(fh, index_col=0, encoding=encoding))
+
+    def test_to_csv_date_format(self, temp_file, datetime_frame):
+        path = str(temp_file)
+        dt_index = datetime_frame.index
+        datetime_frame = DataFrame(
+            {"A": dt_index, "B": dt_index.shift(1)}, index=dt_index
+        )
+        datetime_frame.to_csv(path, date_format="%Y%m%d")
+
+        # Check that the data was put in the specified format
+        test = read_csv(path, index_col=0)
+
+        datetime_frame_int = datetime_frame.map(lambda x: int(x.strftime("%Y%m%d")))
+        datetime_frame_int.index = datetime_frame_int.index.map(
+            lambda x: int(x.strftime("%Y%m%d"))
+        )
+
+        tm.assert_frame_equal(test, datetime_frame_int)
+
+        datetime_frame.to_csv(path, date_format="%Y-%m-%d")
+
+        # Check that the data was put in the specified format
+        test = read_csv(path, index_col=0)
+        datetime_frame_str = datetime_frame.map(lambda x: x.strftime("%Y-%m-%d"))
+        datetime_frame_str.index = datetime_frame_str.index.map(
+            lambda x: x.strftime("%Y-%m-%d")
+        )
+
+        tm.assert_frame_equal(test, datetime_frame_str)
+
+        # Check that columns get converted
+        datetime_frame_columns = datetime_frame.T
+        datetime_frame_columns.to_csv(path, date_format="%Y%m%d")
+
+        test = read_csv(path, index_col=0)
+
+        datetime_frame_columns = datetime_frame_columns.map(
+            lambda x: int(x.strftime("%Y%m%d"))
+        )
+        # Columns don't get converted to ints by read_csv
+        datetime_frame_columns.columns = datetime_frame_columns.columns.map(
+            lambda x: x.strftime("%Y%m%d")
+        )
+
+        tm.assert_frame_equal(test, datetime_frame_columns)
+
+        # test NaTs
+        nat_index = to_datetime(
+            ["NaT"] * 10 + ["2000-01-01", "2000-01-01", "2000-01-01"]
+        )
+        nat_frame = DataFrame({"A": nat_index}, index=nat_index)
+        nat_frame.to_csv(path, date_format="%Y-%m-%d")
+
+        test = read_csv(path, parse_dates=[0, 1], index_col=0)
+
+        tm.assert_frame_equal(test, nat_frame)
+
+    @pytest.mark.parametrize("td", [pd.Timedelta(0), pd.Timedelta("10s")])
+    def test_to_csv_with_dst_transitions(self, td, temp_file):
+        path = str(temp_file)
+        # make sure we are not failing on transitions
+        times = date_range(
+            "2013-10-26 23:00",
+            "2013-10-27 01:00",
+            tz="Europe/London",
+            freq="h",
+            ambiguous="infer",
+        )
+        i = times + td
+        i = i._with_freq(None)  # freq is not preserved by read_csv
+        time_range = np.array(range(len(i)), dtype="int64")
+        df = DataFrame({"A": time_range}, index=i)
+        df.to_csv(path, index=True)
+        # we have to reconvert the index as we
+        # don't parse the tz's
+        result = read_csv(path, index_col=0)
+        result.index = (
+            to_datetime(result.index, utc=True)
+            .tz_convert("Europe/London")
+            .as_unit("ns")
+        )
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize(
+        "start,end",
+        [
+            ["2015-03-29", "2015-03-30"],
+            ["2015-10-25", "2015-10-26"],
+        ],
+    )
+    def test_to_csv_with_dst_transitions_with_pickle(self, start, end, temp_file):
+        # GH11619
+        idx = date_range(start, end, freq="h", tz="Europe/Paris")
+        idx = idx._with_freq(None)  # freq does not round-trip
+        idx._data._freq = None  # otherwise there is trouble on unpickle
+        df = DataFrame({"values": 1, "idx": idx}, index=idx)
+        with tm.ensure_clean("csv_date_format_with_dst") as path:
+            df.to_csv(path, index=True)
+            result = read_csv(path, index_col=0)
+            result.index = (
+                to_datetime(result.index, utc=True)
+                .tz_convert("Europe/Paris")
+                .as_unit("ns")
+            )
+            result["idx"] = to_datetime(result["idx"], utc=True).astype(
+                "datetime64[ns, Europe/Paris]"
+            )
+            tm.assert_frame_equal(result, df)
+
+        # assert working
+        df.astype(str)
+
+        path = str(temp_file)
+        df.to_pickle(path)
+        result = pd.read_pickle(path)
+        tm.assert_frame_equal(result, df)
+
+    def test_to_csv_quoting(self):
+        df = DataFrame(
+            {
+                "c_bool": [True, False],
+                "c_float": [1.0, 3.2],
+                "c_int": [42, np.nan],
+                "c_string": ["a", "b,c"],
+            }
+        )
+
+        expected_rows = [
+            ",c_bool,c_float,c_int,c_string",
+            "0,True,1.0,42.0,a",
+            '1,False,3.2,,"b,c"',
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        result = df.to_csv()
+        assert result == expected
+
+        result = df.to_csv(quoting=None)
+        assert result == expected
+
+        expected_rows = [
+            ",c_bool,c_float,c_int,c_string",
+            "0,True,1.0,42.0,a",
+            '1,False,3.2,,"b,c"',
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        result = df.to_csv(quoting=csv.QUOTE_MINIMAL)
+        assert result == expected
+
+        expected_rows = [
+            '"","c_bool","c_float","c_int","c_string"',
+            '"0","True","1.0","42.0","a"',
+            '"1","False","3.2","","b,c"',
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        result = df.to_csv(quoting=csv.QUOTE_ALL)
+        assert result == expected
+
+        # see gh-12922, gh-13259: make sure changes to
+        # the formatters do not break this behaviour
+        expected_rows = [
+            '"","c_bool","c_float","c_int","c_string"',
+            '0,True,1.0,42.0,"a"',
+            '1,False,3.2,"","b,c"',
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        result = df.to_csv(quoting=csv.QUOTE_NONNUMERIC)
+        assert result == expected
+
+        msg = "need to escape, but no escapechar set"
+        with pytest.raises(csv.Error, match=msg):
+            df.to_csv(quoting=csv.QUOTE_NONE)
+
+        with pytest.raises(csv.Error, match=msg):
+            df.to_csv(quoting=csv.QUOTE_NONE, escapechar=None)
+
+        expected_rows = [
+            ",c_bool,c_float,c_int,c_string",
+            "0,True,1.0,42.0,a",
+            "1,False,3.2,,b!,c",
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        result = df.to_csv(quoting=csv.QUOTE_NONE, escapechar="!")
+        assert result == expected
+
+        expected_rows = [
+            ",c_bool,c_ffloat,c_int,c_string",
+            "0,True,1.0,42.0,a",
+            "1,False,3.2,,bf,c",
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        result = df.to_csv(quoting=csv.QUOTE_NONE, escapechar="f")
+        assert result == expected
+
+        # see gh-3503: quoting Windows line terminators
+        # presents with encoding?
+        text_rows = ["a,b,c", '1,"test \r\n",3']
+        text = tm.convert_rows_list_to_csv_str(text_rows)
+        df = read_csv(StringIO(text))
+
+        buf = StringIO()
+        df.to_csv(buf, encoding="utf-8", index=False)
+        assert buf.getvalue() == text
+
+        # xref gh-7791: make sure the quoting parameter is passed through
+        # with multi-indexes
+        df = DataFrame({"a": [1, 2], "b": [3, 4], "c": [5, 6]})
+        df = df.set_index(["a", "b"])
+
+        expected_rows = ['"a","b","c"', '"1","3","5"', '"2","4","6"']
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv(quoting=csv.QUOTE_ALL) == expected
+
+    def test_period_index_date_overflow(self):
+        # see gh-15982
+
+        dates = ["1990-01-01", "2000-01-01", "3005-01-01"]
+        index = pd.PeriodIndex(dates, freq="D")
+
+        df = DataFrame([4, 5, 6], index=index)
+        result = df.to_csv()
+
+        expected_rows = [",0", "1990-01-01,4", "2000-01-01,5", "3005-01-01,6"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+        date_format = "%m-%d-%Y"
+        result = df.to_csv(date_format=date_format)
+
+        expected_rows = [",0", "01-01-1990,4", "01-01-2000,5", "01-01-3005,6"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+        # Overflow with pd.NaT
+        dates = ["1990-01-01", NaT, "3005-01-01"]
+        index = pd.PeriodIndex(dates, freq="D")
+
+        df = DataFrame([4, 5, 6], index=index)
+        result = df.to_csv()
+
+        expected_rows = [",0", "1990-01-01,4", ",5", "3005-01-01,6"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_multi_index_header(self):
+        # see gh-5539
+        columns = MultiIndex.from_tuples([("a", 1), ("a", 2), ("b", 1), ("b", 2)])
+        df = DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]])
+        df.columns = columns
+
+        header = ["a", "b", "c", "d"]
+        result = df.to_csv(header=header)
+
+        expected_rows = [",a,b,c,d", "0,1,2,3,4", "1,5,6,7,8"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_to_csv_single_level_multi_index(self):
+        # see gh-26303
+        index = Index([(1,), (2,), (3,)])
+        df = DataFrame([[1, 2, 3]], columns=index)
+        df = df.reindex(columns=[(1,), (3,)])
+        expected = ",1,3\n0,1,3\n"
+        result = df.to_csv(lineterminator="\n")
+        tm.assert_almost_equal(result, expected)
+
+    def test_gz_lineend(self, tmp_path):
+        # GH 25311
+        df = DataFrame({"a": [1, 2]})
+        expected_rows = ["a", "1", "2"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        file_path = tmp_path / "__test_gz_lineend.csv.gz"
+        file_path.touch()
+        path = str(file_path)
+        df.to_csv(path, index=False)
+        with tm.decompress_file(path, compression="gzip") as f:
+            result = f.read().decode("utf-8")
+
+        assert result == expected
+
+    def test_to_csv_numpy_16_bug(self):
+        frame = DataFrame({"a": date_range("1/1/2000", periods=10)})
+
+        buf = StringIO()
+        frame.to_csv(buf)
+
+        result = buf.getvalue()
+        assert "2000-01-01" in result
+
+    def test_to_csv_na_quoting(self):
+        # GH 15891
+        # Normalize carriage return for Windows OS
+        result = (
+            DataFrame([None, None])
+            .to_csv(None, header=False, index=False, na_rep="")
+            .replace("\r\n", "\n")
+        )
+        expected = '""\n""\n'
+        assert result == expected
+
+    def test_to_csv_categorical_and_ea(self):
+        # GH#46812
+        df = DataFrame({"a": "x", "b": [1, pd.NA]})
+        df["b"] = df["b"].astype("Int16")
+        df["b"] = df["b"].astype("category")
+        result = df.to_csv()
+        expected_rows = [",a,b", "0,x,1", "1,x,"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_to_csv_categorical_and_interval(self):
+        # GH#46297
+        df = DataFrame(
+            {
+                "a": [
+                    pd.Interval(
+                        Timestamp("2020-01-01"),
+                        Timestamp("2020-01-02"),
+                        closed="both",
+                    )
+                ]
+            }
+        )
+        df["a"] = df["a"].astype("category")
+        result = df.to_csv()
+        expected_rows = [",a", '0,"[2020-01-01 00:00:00, 2020-01-02 00:00:00]"']
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_to_csv_warn_when_zip_tar_and_append_mode(self, tmp_path):
+        # GH57875
+        df = DataFrame({"a": [1, 2, 3]})
+        msg = (
+            "zip and tar do not support mode 'a' properly. This combination will "
+            "result in multiple files with same name being added to the archive"
+        )
+        zip_path = tmp_path / "test.zip"
+        tar_path = tmp_path / "test.tar"
+        with tm.assert_produces_warning(
+            RuntimeWarning, match=msg, raise_on_extra_warnings=False
+        ):
+            df.to_csv(zip_path, mode="a")
+
+        with tm.assert_produces_warning(
+            RuntimeWarning, match=msg, raise_on_extra_warnings=False
+        ):
+            df.to_csv(tar_path, mode="a")
diff --git a/pandas/pandas/tests/frame/methods/test_to_dict.py b/pandas/pandas/tests/frame/methods/test_to_dict.py
new file mode 100644
index 0000000000000..c43d947b4877e
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_dict.py
@@ -0,0 +1,541 @@
+from collections import (
+    OrderedDict,
+    defaultdict,
+)
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+    Interval,
+    MultiIndex,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameToDict:
+    def test_to_dict_timestamp(self):
+        # GH#11247
+        # split/records producing np.datetime64 rather than Timestamps
+        # on datetime64[ns] dtypes only
+
+        tsmp = Timestamp("20130101")
+        test_data = DataFrame({"A": [tsmp, tsmp], "B": [tsmp, tsmp]})
+        test_data_mixed = DataFrame({"A": [tsmp, tsmp], "B": [1, 2]})
+
+        expected_records = [{"A": tsmp, "B": tsmp}, {"A": tsmp, "B": tsmp}]
+        expected_records_mixed = [{"A": tsmp, "B": 1}, {"A": tsmp, "B": 2}]
+
+        assert test_data.to_dict(orient="records") == expected_records
+        assert test_data_mixed.to_dict(orient="records") == expected_records_mixed
+
+        expected_series = {
+            "A": Series([tsmp, tsmp], name="A"),
+            "B": Series([tsmp, tsmp], name="B"),
+        }
+        expected_series_mixed = {
+            "A": Series([tsmp, tsmp], name="A"),
+            "B": Series([1, 2], name="B"),
+        }
+
+        tm.assert_dict_equal(test_data.to_dict(orient="series"), expected_series)
+        tm.assert_dict_equal(
+            test_data_mixed.to_dict(orient="series"), expected_series_mixed
+        )
+
+        expected_split = {
+            "index": [0, 1],
+            "data": [[tsmp, tsmp], [tsmp, tsmp]],
+            "columns": ["A", "B"],
+        }
+        expected_split_mixed = {
+            "index": [0, 1],
+            "data": [[tsmp, 1], [tsmp, 2]],
+            "columns": ["A", "B"],
+        }
+
+        tm.assert_dict_equal(test_data.to_dict(orient="split"), expected_split)
+        tm.assert_dict_equal(
+            test_data_mixed.to_dict(orient="split"), expected_split_mixed
+        )
+
+    def test_to_dict_index_not_unique_with_index_orient(self):
+        # GH#22801
+        # Data loss when indexes are not unique. Raise ValueError.
+        df = DataFrame({"a": [1, 2], "b": [0.5, 0.75]}, index=["A", "A"])
+        msg = "DataFrame index must be unique for orient='index'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_dict(orient="index")
+
+    def test_to_dict_invalid_orient(self):
+        df = DataFrame({"A": [0, 1]})
+        msg = "orient 'xinvalid' not understood"
+        with pytest.raises(ValueError, match=msg):
+            df.to_dict(orient="xinvalid")
+
+    @pytest.mark.parametrize("orient", ["d", "l", "r", "sp", "s", "i"])
+    def test_to_dict_short_orient_raises(self, orient):
+        # GH#32515
+        df = DataFrame({"A": [0, 1]})
+        with pytest.raises(ValueError, match="not understood"):
+            df.to_dict(orient=orient)
+
+    @pytest.mark.parametrize("mapping", [dict, defaultdict(list), OrderedDict])
+    def test_to_dict(self, mapping):
+        # orient= should only take the listed options
+        # see GH#32515
+        test_data = {"A": {"1": 1, "2": 2}, "B": {"1": "1", "2": "2", "3": "3"}}
+
+        # GH#16122
+        recons_data = DataFrame(test_data).to_dict(into=mapping)
+
+        for k, v in test_data.items():
+            for k2, v2 in v.items():
+                assert v2 == recons_data[k][k2]
+
+        recons_data = DataFrame(test_data).to_dict("list", into=mapping)
+
+        for k, v in test_data.items():
+            for k2, v2 in v.items():
+                assert v2 == recons_data[k][int(k2) - 1]
+
+        recons_data = DataFrame(test_data).to_dict("series", into=mapping)
+
+        for k, v in test_data.items():
+            for k2, v2 in v.items():
+                assert v2 == recons_data[k][k2]
+
+        recons_data = DataFrame(test_data).to_dict("split", into=mapping)
+        expected_split = {
+            "columns": ["A", "B"],
+            "index": ["1", "2", "3"],
+            "data": [[1.0, "1"], [2.0, "2"], [np.nan, "3"]],
+        }
+        tm.assert_dict_equal(recons_data, expected_split)
+
+        recons_data = DataFrame(test_data).to_dict("records", into=mapping)
+        expected_records = [
+            {"A": 1.0, "B": "1"},
+            {"A": 2.0, "B": "2"},
+            {"A": np.nan, "B": "3"},
+        ]
+        assert isinstance(recons_data, list)
+        assert len(recons_data) == 3
+        for left, right in zip(recons_data, expected_records):
+            tm.assert_dict_equal(left, right)
+
+        # GH#10844
+        recons_data = DataFrame(test_data).to_dict("index")
+
+        for k, v in test_data.items():
+            for k2, v2 in v.items():
+                assert v2 == recons_data[k2][k]
+
+        df = DataFrame(test_data)
+        df["duped"] = df[df.columns[0]]
+        recons_data = df.to_dict("index")
+        comp_data = test_data.copy()
+        comp_data["duped"] = comp_data[df.columns[0]]
+        for k, v in comp_data.items():
+            for k2, v2 in v.items():
+                assert v2 == recons_data[k2][k]
+
+    @pytest.mark.parametrize("mapping", [list, defaultdict, []])
+    def test_to_dict_errors(self, mapping):
+        # GH#16122
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 3)))
+        msg = "|".join(
+            [
+                "unsupported type: <class 'list'>",
+                r"to_dict\(\) only accepts initialized defaultdicts",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.to_dict(into=mapping)
+
+    def test_to_dict_not_unique_warning(self):
+        # GH#16927: When converting to a dict, if a column has a non-unique name
+        # it will be dropped, throwing a warning.
+        df = DataFrame([[1, 2, 3]], columns=["a", "a", "b"])
+        with tm.assert_produces_warning(UserWarning, match="columns will be omitted"):
+            df.to_dict()
+
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    @pytest.mark.parametrize(
+        "orient,expected",
+        [
+            ("list", {"A": [2, 5], "B": [3, 6]}),
+            ("dict", {"A": {0: 2, 1: 5}, "B": {0: 3, 1: 6}}),
+        ],
+    )
+    def test_to_dict_not_unique(self, orient, expected):
+        # GH#54824: This is to make sure that dataframes with non-unique column
+        # would have uniform behavior throughout different orients
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "A", "B"])
+        result = df.to_dict(orient)
+        assert result == expected
+
+    # orient - orient argument to to_dict function
+    # item_getter - function for extracting value from
+    # the resulting dict using column name and index
+    @pytest.mark.parametrize(
+        "orient,item_getter",
+        [
+            ("dict", lambda d, col, idx: d[col][idx]),
+            ("records", lambda d, col, idx: d[idx][col]),
+            ("list", lambda d, col, idx: d[col][idx]),
+            ("split", lambda d, col, idx: d["data"][idx][d["columns"].index(col)]),
+            ("index", lambda d, col, idx: d[idx][col]),
+        ],
+    )
+    def test_to_dict_box_scalars(self, orient, item_getter):
+        # GH#14216, GH#23753
+        # make sure that we are boxing properly
+        df = DataFrame({"a": [1, 2], "b": [0.1, 0.2]})
+        result = df.to_dict(orient=orient)
+        assert isinstance(item_getter(result, "a", 0), int)
+        assert isinstance(item_getter(result, "b", 0), float)
+
+    def test_to_dict_tz(self):
+        # GH#18372 When converting to dict with orient='records' columns of
+        # datetime that are tz-aware were not converted to required arrays
+        data = [
+            (datetime(2017, 11, 18, 21, 53, 0, 219225, tzinfo=timezone.utc),),
+            (datetime(2017, 11, 18, 22, 6, 30, 61810, tzinfo=timezone.utc),),
+        ]
+        df = DataFrame(list(data), columns=["d"])
+
+        result = df.to_dict(orient="records")
+        expected = [
+            {"d": Timestamp("2017-11-18 21:53:00.219225+0000", tz=timezone.utc)},
+            {"d": Timestamp("2017-11-18 22:06:30.061810+0000", tz=timezone.utc)},
+        ]
+        tm.assert_dict_equal(result[0], expected[0])
+        tm.assert_dict_equal(result[1], expected[1])
+
+    @pytest.mark.parametrize(
+        "into, expected",
+        [
+            (
+                dict,
+                {
+                    0: {"int_col": 1, "float_col": 1.0},
+                    1: {"int_col": 2, "float_col": 2.0},
+                    2: {"int_col": 3, "float_col": 3.0},
+                },
+            ),
+            (
+                OrderedDict,
+                OrderedDict(
+                    [
+                        (0, {"int_col": 1, "float_col": 1.0}),
+                        (1, {"int_col": 2, "float_col": 2.0}),
+                        (2, {"int_col": 3, "float_col": 3.0}),
+                    ]
+                ),
+            ),
+            (
+                defaultdict(dict),
+                defaultdict(
+                    dict,
+                    {
+                        0: {"int_col": 1, "float_col": 1.0},
+                        1: {"int_col": 2, "float_col": 2.0},
+                        2: {"int_col": 3, "float_col": 3.0},
+                    },
+                ),
+            ),
+        ],
+    )
+    def test_to_dict_index_dtypes(self, into, expected):
+        # GH#18580
+        # When using to_dict(orient='index') on a dataframe with int
+        # and float columns only the int columns were cast to float
+
+        df = DataFrame({"int_col": [1, 2, 3], "float_col": [1.0, 2.0, 3.0]})
+
+        result = df.to_dict(orient="index", into=into)
+        cols = ["int_col", "float_col"]
+        result = DataFrame.from_dict(result, orient="index")[cols]
+        expected = DataFrame.from_dict(expected, orient="index")[cols]
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_dict_numeric_names(self):
+        # GH#24940
+        df = DataFrame({str(i): [i] for i in range(5)})
+        result = set(df.to_dict("records")[0].keys())
+        expected = set(df.columns)
+        assert result == expected
+
+    def test_to_dict_wide(self):
+        # GH#24939
+        df = DataFrame({(f"A_{i:d}"): [i] for i in range(256)})
+        result = df.to_dict("records")[0]
+        expected = {f"A_{i:d}": i for i in range(256)}
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "data,dtype",
+        (
+            ([True, True, False], bool),
+            [
+                [
+                    datetime(2018, 1, 1),
+                    datetime(2019, 2, 2),
+                    datetime(2020, 3, 3),
+                ],
+                Timestamp,
+            ],
+            [[1.0, 2.0, 3.0], float],
+            [[1, 2, 3], int],
+            [["X", "Y", "Z"], str],
+        ),
+    )
+    def test_to_dict_orient_dtype(self, data, dtype):
+        # GH22620 & GH21256
+
+        df = DataFrame({"a": data})
+        d = df.to_dict(orient="records")
+        assert all(type(record["a"]) is dtype for record in d)
+
+    @pytest.mark.parametrize(
+        "data,expected_dtype",
+        (
+            [np.uint64(2), int],
+            [np.int64(-9), int],
+            [np.float64(1.1), float],
+            [np.bool_(True), bool],
+            [np.datetime64("2005-02-25"), Timestamp],
+        ),
+    )
+    def test_to_dict_scalar_constructor_orient_dtype(self, data, expected_dtype):
+        # GH22620 & GH21256
+
+        df = DataFrame({"a": data}, index=[0])
+        d = df.to_dict(orient="records")
+        result = type(d[0]["a"])
+        assert result is expected_dtype
+
+    def test_to_dict_mixed_numeric_frame(self):
+        # GH 12859
+        df = DataFrame({"a": [1.0], "b": [9.0]})
+        result = df.reset_index().to_dict("records")
+        expected = [{"index": 0, "a": 1.0, "b": 9.0}]
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            None,
+            Index(["aa", "bb"]),
+            Index(["aa", "bb"], name="cc"),
+            MultiIndex.from_tuples([("a", "b"), ("a", "c")]),
+            MultiIndex.from_tuples([("a", "b"), ("a", "c")], names=["n1", "n2"]),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "columns",
+        [
+            ["x", "y"],
+            Index(["x", "y"]),
+            Index(["x", "y"], name="z"),
+            MultiIndex.from_tuples([("x", 1), ("y", 2)]),
+            MultiIndex.from_tuples([("x", 1), ("y", 2)], names=["z1", "z2"]),
+        ],
+    )
+    def test_to_dict_orient_tight(self, index, columns):
+        df = DataFrame.from_records(
+            [[1, 3], [2, 4]],
+            columns=columns,
+            index=index,
+        )
+        roundtrip = DataFrame.from_dict(df.to_dict(orient="tight"), orient="tight")
+
+        tm.assert_frame_equal(df, roundtrip)
+
+    @pytest.mark.parametrize(
+        "orient",
+        ["dict", "list", "split", "records", "index", "tight"],
+    )
+    @pytest.mark.parametrize(
+        "data,expected_types",
+        (
+            (
+                {
+                    "a": [np.int64(1), 1, np.int64(3)],
+                    "b": [np.float64(1.0), 2.0, np.float64(3.0)],
+                    "c": [np.float64(1.0), 2, np.int64(3)],
+                    "d": [np.float64(1.0), "a", np.int64(3)],
+                    "e": [np.float64(1.0), ["a"], np.int64(3)],
+                    "f": [np.float64(1.0), ("a",), np.int64(3)],
+                },
+                {
+                    "a": [int, int, int],
+                    "b": [float, float, float],
+                    "c": [float, float, float],
+                    "d": [float, str, int],
+                    "e": [float, list, int],
+                    "f": [float, tuple, int],
+                },
+            ),
+            (
+                {
+                    "a": [1, 2, 3],
+                    "b": [1.1, 2.2, 3.3],
+                },
+                {
+                    "a": [int, int, int],
+                    "b": [float, float, float],
+                },
+            ),
+            (  # Make sure we have one df which is all object type cols
+                {
+                    "a": [1, "hello", 3],
+                    "b": [1.1, "world", 3.3],
+                },
+                {
+                    "a": [int, str, int],
+                    "b": [float, str, float],
+                },
+            ),
+        ),
+    )
+    def test_to_dict_returns_native_types(self, orient, data, expected_types):
+        # GH 46751
+        # Tests we get back native types for all orient types
+        df = DataFrame(data)
+        result = df.to_dict(orient)
+        if orient == "dict":
+            assertion_iterator = (
+                (i, key, value)
+                for key, index_value_map in result.items()
+                for i, value in index_value_map.items()
+            )
+        elif orient == "list":
+            assertion_iterator = (
+                (i, key, value)
+                for key, values in result.items()
+                for i, value in enumerate(values)
+            )
+        elif orient in {"split", "tight"}:
+            assertion_iterator = (
+                (i, key, result["data"][i][j])
+                for i in result["index"]
+                for j, key in enumerate(result["columns"])
+            )
+        elif orient == "records":
+            assertion_iterator = (
+                (i, key, value)
+                for i, record in enumerate(result)
+                for key, value in record.items()
+            )
+        elif orient == "index":
+            assertion_iterator = (
+                (i, key, value)
+                for i, record in result.items()
+                for key, value in record.items()
+            )
+
+        for i, key, value in assertion_iterator:
+            assert value == data[key][i]
+            assert type(value) is expected_types[key][i]
+
+    @pytest.mark.parametrize("orient", ["dict", "list", "series", "records", "index"])
+    def test_to_dict_index_false_error(self, orient):
+        # GH#46398
+        df = DataFrame({"col1": [1, 2], "col2": [3, 4]}, index=["row1", "row2"])
+        msg = "'index=False' is only valid when 'orient' is 'split' or 'tight'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_dict(orient=orient, index=False)
+
+    @pytest.mark.parametrize(
+        "orient, expected",
+        [
+            ("split", {"columns": ["col1", "col2"], "data": [[1, 3], [2, 4]]}),
+            (
+                "tight",
+                {
+                    "columns": ["col1", "col2"],
+                    "data": [[1, 3], [2, 4]],
+                    "column_names": [None],
+                },
+            ),
+        ],
+    )
+    def test_to_dict_index_false(self, orient, expected):
+        # GH#46398
+        df = DataFrame({"col1": [1, 2], "col2": [3, 4]}, index=["row1", "row2"])
+        result = df.to_dict(orient=orient, index=False)
+        tm.assert_dict_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "orient, expected",
+        [
+            ("dict", {"a": {0: 1, 1: None}}),
+            ("list", {"a": [1, None]}),
+            ("split", {"index": [0, 1], "columns": ["a"], "data": [[1], [None]]}),
+            (
+                "tight",
+                {
+                    "index": [0, 1],
+                    "columns": ["a"],
+                    "data": [[1], [None]],
+                    "index_names": [None],
+                    "column_names": [None],
+                },
+            ),
+            ("records", [{"a": 1}, {"a": None}]),
+            ("index", {0: {"a": 1}, 1: {"a": None}}),
+        ],
+    )
+    def test_to_dict_na_to_none(self, orient, expected):
+        # GH#50795
+        df = DataFrame({"a": [1, NA]}, dtype="Int64")
+        result = df.to_dict(orient=orient)
+        assert result == expected
+
+    def test_to_dict_masked_native_python(self):
+        # GH#34665
+        df = DataFrame({"a": Series([1, 2], dtype="Int64"), "B": 1})
+        result = df.to_dict(orient="records")
+        assert isinstance(result[0]["a"], int)
+
+        df = DataFrame({"a": Series([1, NA], dtype="Int64"), "B": 1})
+        result = df.to_dict(orient="records")
+        assert isinstance(result[0]["a"], int)
+
+    def test_to_dict_tight_no_warning_with_duplicate_column(self):
+        # GH#58281
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], columns=["A", "A"])
+        with tm.assert_produces_warning(None):
+            result = df.to_dict(orient="tight")
+        expected = {
+            "index": [0, 1, 2],
+            "columns": ["A", "A"],
+            "data": [[1, 2], [3, 4], [5, 6]],
+            "index_names": [None],
+            "column_names": [None],
+        }
+        assert result == expected
+
+
+@pytest.mark.parametrize(
+    "val", [Timestamp(2020, 1, 1), Timedelta(1), Period("2020"), Interval(1, 2)]
+)
+def test_to_dict_list_pd_scalars(val):
+    # GH 54824
+    df = DataFrame({"a": [val]})
+    result = df.to_dict(orient="list")
+    expected = {"a": [val]}
+    assert result == expected
diff --git a/pandas/pandas/tests/frame/methods/test_to_dict_of_blocks.py b/pandas/pandas/tests/frame/methods/test_to_dict_of_blocks.py
new file mode 100644
index 0000000000000..4f621b4643b70
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_dict_of_blocks.py
@@ -0,0 +1,57 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+import pandas._testing as tm
+from pandas.core.arrays import NumpyExtensionArray
+
+
+class TestToDictOfBlocks:
+    def test_no_copy_blocks(self, float_frame):
+        # GH#9607
+        df = DataFrame(float_frame, copy=True)
+        column = df.columns[0]
+
+        _last_df = None
+        # use the copy=False, change a column
+        blocks = df._to_dict_of_blocks()
+        for _df in blocks.values():
+            _last_df = _df
+            if column in _df:
+                _df.loc[:, column] = _df[column] + 1
+        assert _last_df is not None and not _last_df[column].equals(df[column])
+
+
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
+def test_to_dict_of_blocks_item_cache():
+    # Calling to_dict_of_blocks should not poison item_cache
+    df = DataFrame({"a": [1, 2, 3, 4], "b": ["a", "b", "c", "d"]})
+    df["c"] = NumpyExtensionArray(np.array([1, 2, None, 3], dtype=object))
+    mgr = df._mgr
+    assert len(mgr.blocks) == 3  # i.e. not consolidated
+
+    ser = df["b"]  # populations item_cache["b"]
+
+    df._to_dict_of_blocks()
+
+    with pytest.raises(ValueError, match="read-only"):
+        ser.values[0] = "foo"
+
+
+def test_set_change_dtype_slice():
+    # GH#8850
+    cols = MultiIndex.from_tuples([("1st", "a"), ("2nd", "b"), ("3rd", "c")])
+    df = DataFrame([[1.0, 2, 3], [4.0, 5, 6]], columns=cols)
+    df["2nd"] = df["2nd"] * 2.0
+
+    blocks = df._to_dict_of_blocks()
+    assert sorted(blocks.keys()) == ["float64", "int64"]
+    tm.assert_frame_equal(
+        blocks["float64"], DataFrame([[1.0, 4.0], [4.0, 10.0]], columns=cols[:2])
+    )
+    tm.assert_frame_equal(blocks["int64"], DataFrame([[3], [6]], columns=cols[2:]))
diff --git a/pandas/pandas/tests/frame/methods/test_to_numpy.py b/pandas/pandas/tests/frame/methods/test_to_numpy.py
new file mode 100644
index 0000000000000..36088cceb13f1
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_numpy.py
@@ -0,0 +1,43 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestToNumpy:
+    def test_to_numpy(self):
+        df = DataFrame({"A": [1, 2], "B": [3, 4.5]})
+        expected = np.array([[1, 3], [2, 4.5]])
+        result = df.to_numpy()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_to_numpy_dtype(self):
+        df = DataFrame({"A": [1, 2], "B": [3, 4.5]})
+        expected = np.array([[1, 3], [2, 4]], dtype="int64")
+        result = df.to_numpy(dtype="int64")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_to_numpy_copy(self):
+        arr = np.random.default_rng(2).standard_normal((4, 3))
+        df = DataFrame(arr)
+        assert df.values.base is not arr
+        assert df.to_numpy(copy=False).base is df.values.base
+        assert df.to_numpy(copy=True).base is not arr
+
+        # we still don't want a copy when na_value=np.nan is passed,
+        #  and that can be respected because we are already numpy-float
+        assert df.to_numpy(copy=False).base is df.values.base
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_to_numpy_mixed_dtype_to_str(self):
+        # https://github.com/pandas-dev/pandas/issues/35455
+        df = DataFrame([[Timestamp("2020-01-01 00:00:00"), 100.0]])
+        result = df.to_numpy(dtype=str)
+        expected = np.array([["2020-01-01 00:00:00", "100.0"]], dtype=str)
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_to_period.py b/pandas/pandas/tests/frame/methods/test_to_period.py
new file mode 100644
index 0000000000000..6a3e6b8c0e059
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_period.py
@@ -0,0 +1,89 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestToPeriod:
+    def test_to_period(self, frame_or_series):
+        K = 5
+
+        dr = date_range("1/1/2000", "1/1/2001", freq="D")
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(dr), K)),
+            index=dr,
+            columns=["A", "B", "C", "D", "E"],
+        )
+        obj["mix"] = "a"
+        obj = tm.get_obj(obj, frame_or_series)
+
+        pts = obj.to_period()
+        exp = obj.copy()
+        exp.index = period_range("1/1/2000", "1/1/2001")
+        tm.assert_equal(pts, exp)
+
+        pts = obj.to_period("M")
+        exp.index = exp.index.asfreq("M")
+        tm.assert_equal(pts, exp)
+
+    def test_to_period_without_freq(self, frame_or_series):
+        # GH#7606 without freq
+        idx = DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03", "2011-01-04"])
+        exp_idx = PeriodIndex(
+            ["2011-01-01", "2011-01-02", "2011-01-03", "2011-01-04"], freq="D"
+        )
+
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), index=idx, columns=idx
+        )
+        obj = tm.get_obj(obj, frame_or_series)
+        expected = obj.copy()
+        expected.index = exp_idx
+        tm.assert_equal(obj.to_period(), expected)
+
+        if frame_or_series is DataFrame:
+            expected = obj.copy()
+            expected.columns = exp_idx
+            tm.assert_frame_equal(obj.to_period(axis=1), expected)
+
+    def test_to_period_columns(self):
+        dr = date_range("1/1/2000", "1/1/2001")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(dr), 5)), index=dr)
+        df["mix"] = "a"
+
+        df = df.T
+        pts = df.to_period(axis=1)
+        exp = df.copy()
+        exp.columns = period_range("1/1/2000", "1/1/2001")
+        tm.assert_frame_equal(pts, exp)
+
+        pts = df.to_period("M", axis=1)
+        tm.assert_index_equal(pts.columns, exp.columns.asfreq("M"))
+
+    def test_to_period_invalid_axis(self):
+        dr = date_range("1/1/2000", "1/1/2001")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(dr), 5)), index=dr)
+        df["mix"] = "a"
+
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.to_period(axis=2)
+
+    def test_to_period_raises(self, index, frame_or_series):
+        # https://github.com/pandas-dev/pandas/issues/33327
+        obj = Series(index=index, dtype=object)
+        if frame_or_series is DataFrame:
+            obj = obj.to_frame()
+
+        if not isinstance(index, DatetimeIndex):
+            msg = f"unsupported Type {type(index).__name__}"
+            with pytest.raises(TypeError, match=msg):
+                obj.to_period()
diff --git a/pandas/pandas/tests/frame/methods/test_to_records.py b/pandas/pandas/tests/frame/methods/test_to_records.py
new file mode 100644
index 0000000000000..fab90b112fa94
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_records.py
@@ -0,0 +1,523 @@
+from collections import abc
+import email
+from email.parser import Parser
+
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalDtype,
+    DataFrame,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameToRecords:
+    def test_to_records_timeseries(self):
+        index = date_range("1/1/2000", periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)),
+            index=index,
+            columns=["a", "b", "c"],
+        )
+
+        result = df.to_records()
+        assert result["index"].dtype == "M8[ns]"
+
+        result = df.to_records(index=False)
+
+    def test_to_records_dt64(self):
+        df = DataFrame(
+            [["one", "two", "three"], ["four", "five", "six"]],
+            index=date_range("2012-01-01", "2012-01-02"),
+        )
+
+        expected = df.index.values[0]
+        result = df.to_records()["index"][0]
+        assert expected == result
+
+    def test_to_records_dt64tz_column(self):
+        # GH#32535 dont less tz in to_records
+        df = DataFrame({"A": date_range("2012-01-01", "2012-01-02", tz="US/Eastern")})
+
+        result = df.to_records()
+
+        assert result.dtype["A"] == object
+        val = result[0][1]
+        assert isinstance(val, Timestamp)
+        assert val == df.loc[0, "A"]
+
+    def test_to_records_with_multindex(self):
+        # GH#3189
+        index = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        data = np.zeros((8, 4))
+        df = DataFrame(data, index=index)
+        r = df.to_records(index=True)["level_0"]
+        assert "bar" in r
+        assert "one" not in r
+
+    def test_to_records_with_Mapping_type(self):
+        abc.Mapping.register(email.message.Message)
+
+        headers = Parser().parsestr(
+            "From: <user@example.com>\n"
+            "To: <someone_else@example.com>\n"
+            "Subject: Test message\n"
+            "\n"
+            "Body would go here\n"
+        )
+
+        frame = DataFrame.from_records([headers])
+        all(x in frame for x in ["Type", "Subject", "From"])
+
+    def test_to_records_floats(self):
+        df = DataFrame(np.random.default_rng(2).random((10, 10)))
+        df.to_records()
+
+    def test_to_records_index_name(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 3)))
+        df.index.name = "X"
+        rs = df.to_records()
+        assert "X" in rs.dtype.fields
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 3)))
+        rs = df.to_records()
+        assert "index" in rs.dtype.fields
+
+        df.index = MultiIndex.from_tuples([("a", "x"), ("a", "y"), ("b", "z")])
+        df.index.names = ["A", None]
+        result = df.to_records()
+        expected = np.rec.fromarrays(
+            [np.array(["a", "a", "b"]), np.array(["x", "y", "z"])]
+            + [np.asarray(df.iloc[:, i]) for i in range(3)],
+            dtype={
+                "names": ["A", "level_1", "0", "1", "2"],
+                "formats": [
+                    "O",
+                    "O",
+                    f"{tm.ENDIAN}f8",
+                    f"{tm.ENDIAN}f8",
+                    f"{tm.ENDIAN}f8",
+                ],
+            },
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_to_records_with_unicode_index(self):
+        # GH#13172
+        # unicode_literals conflict with to_records
+        result = DataFrame([{"a": "x", "b": "y"}]).set_index("a").to_records()
+        expected = np.rec.array([("x", "y")], dtype=[("a", "O"), ("b", "O")])
+        tm.assert_almost_equal(result, expected)
+
+    def test_to_records_index_dtype(self):
+        # GH 47263: consistent data types for Index and MultiIndex
+        df = DataFrame(
+            {
+                1: date_range("2022-01-01", periods=2),
+                2: date_range("2022-01-01", periods=2),
+                3: date_range("2022-01-01", periods=2),
+            }
+        )
+
+        expected = np.rec.array(
+            [
+                ("2022-01-01", "2022-01-01", "2022-01-01"),
+                ("2022-01-02", "2022-01-02", "2022-01-02"),
+            ],
+            dtype=[
+                ("1", f"{tm.ENDIAN}M8[ns]"),
+                ("2", f"{tm.ENDIAN}M8[ns]"),
+                ("3", f"{tm.ENDIAN}M8[ns]"),
+            ],
+        )
+
+        result = df.to_records(index=False)
+        tm.assert_almost_equal(result, expected)
+
+        result = df.set_index(1).to_records(index=True)
+        tm.assert_almost_equal(result, expected)
+
+        result = df.set_index([1, 2]).to_records(index=True)
+        tm.assert_almost_equal(result, expected)
+
+    def test_to_records_with_unicode_column_names(self):
+        # xref issue: https://github.com/numpy/numpy/issues/2407
+        # Issue GH#11879. to_records used to raise an exception when used
+        # with column names containing non-ascii characters in Python 2
+        result = DataFrame(data={"accented_name_é": [1.0]}).to_records()
+
+        # Note that numpy allows for unicode field names but dtypes need
+        # to be specified using dictionary instead of list of tuples.
+        expected = np.rec.array(
+            [(0, 1.0)],
+            dtype={"names": ["index", "accented_name_é"], "formats": ["=i8", "=f8"]},
+        )
+        tm.assert_almost_equal(result, expected)
+
+    def test_to_records_with_categorical(self):
+        # GH#8626
+
+        # dict creation
+        df = DataFrame({"A": list("abc")}, dtype="category")
+        expected = Series(list("abc"), dtype="category", name="A")
+        tm.assert_series_equal(df["A"], expected)
+
+        # list-like creation
+        df = DataFrame(list("abc"), dtype="category")
+        expected = Series(list("abc"), dtype="category", name=0)
+        tm.assert_series_equal(df[0], expected)
+
+        # to record array
+        # this coerces
+        result = df.to_records()
+        expected = np.rec.array(
+            [(0, "a"), (1, "b"), (2, "c")], dtype=[("index", "=i8"), ("0", "O")]
+        )
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "kwargs,expected",
+        [
+            # No dtypes --> default to array dtypes.
+            (
+                {},
+                np.rec.array(
+                    [(0, 1, 0.2, "a"), (1, 2, 1.5, "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Should have no effect in this case.
+            (
+                {"index": True},
+                np.rec.array(
+                    [(0, 1, 0.2, "a"), (1, 2, 1.5, "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Column dtype applied across the board. Index unaffected.
+            (
+                {"column_dtypes": f"{tm.ENDIAN}U4"},
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", f"{tm.ENDIAN}U4"),
+                        ("B", f"{tm.ENDIAN}U4"),
+                        ("C", f"{tm.ENDIAN}U4"),
+                    ],
+                ),
+            ),
+            # Index dtype applied across the board. Columns unaffected.
+            (
+                {"index_dtypes": f"{tm.ENDIAN}U1"},
+                np.rec.array(
+                    [("0", 1, 0.2, "a"), ("1", 2, 1.5, "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}U1"),
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Pass in a type instance.
+            (
+                {"column_dtypes": str},
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", f"{tm.ENDIAN}U"),
+                        ("B", f"{tm.ENDIAN}U"),
+                        ("C", f"{tm.ENDIAN}U"),
+                    ],
+                ),
+            ),
+            # Pass in a dtype instance.
+            (
+                {"column_dtypes": np.dtype(np.str_)},
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", f"{tm.ENDIAN}U"),
+                        ("B", f"{tm.ENDIAN}U"),
+                        ("C", f"{tm.ENDIAN}U"),
+                    ],
+                ),
+            ),
+            # Pass in a dictionary (name-only).
+            (
+                {
+                    "column_dtypes": {
+                        "A": np.int8,
+                        "B": np.float32,
+                        "C": f"{tm.ENDIAN}U2",
+                    }
+                },
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", "i1"),
+                        ("B", f"{tm.ENDIAN}f4"),
+                        ("C", f"{tm.ENDIAN}U2"),
+                    ],
+                ),
+            ),
+            # Pass in a dictionary (indices-only).
+            (
+                {"index_dtypes": {0: "int16"}},
+                np.rec.array(
+                    [(0, 1, 0.2, "a"), (1, 2, 1.5, "bc")],
+                    dtype=[
+                        ("index", "i2"),
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Ignore index mappings if index is not True.
+            (
+                {"index": False, "index_dtypes": f"{tm.ENDIAN}U2"},
+                np.rec.array(
+                    [(1, 0.2, "a"), (2, 1.5, "bc")],
+                    dtype=[
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Non-existent names / indices in mapping should not error.
+            (
+                {"index_dtypes": {0: "int16", "not-there": "float32"}},
+                np.rec.array(
+                    [(0, 1, 0.2, "a"), (1, 2, 1.5, "bc")],
+                    dtype=[
+                        ("index", "i2"),
+                        ("A", f"{tm.ENDIAN}i8"),
+                        ("B", f"{tm.ENDIAN}f8"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Names / indices not in mapping default to array dtype.
+            (
+                {"column_dtypes": {"A": np.int8, "B": np.float32}},
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", "i1"),
+                        ("B", f"{tm.ENDIAN}f4"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Names / indices not in dtype mapping default to array dtype.
+            (
+                {"column_dtypes": {"A": np.dtype("int8"), "B": np.dtype("float32")}},
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}i8"),
+                        ("A", "i1"),
+                        ("B", f"{tm.ENDIAN}f4"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Mixture of everything.
+            (
+                {
+                    "column_dtypes": {"A": np.int8, "B": np.float32},
+                    "index_dtypes": f"{tm.ENDIAN}U2",
+                },
+                np.rec.array(
+                    [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}U2"),
+                        ("A", "i1"),
+                        ("B", f"{tm.ENDIAN}f4"),
+                        ("C", "O"),
+                    ],
+                ),
+            ),
+            # Invalid dype values.
+            (
+                {"index": False, "column_dtypes": []},
+                (ValueError, "Invalid dtype \\[\\] specified for column A"),
+            ),
+            (
+                {"index": False, "column_dtypes": {"A": "int32", "B": 5}},
+                (ValueError, "Invalid dtype 5 specified for column B"),
+            ),
+            # Numpy can't handle EA types, so check error is raised
+            (
+                {
+                    "index": False,
+                    "column_dtypes": {"A": "int32", "B": CategoricalDtype(["a", "b"])},
+                },
+                (ValueError, "Invalid dtype category specified for column B"),
+            ),
+            # Check that bad types raise
+            (
+                {"index": False, "column_dtypes": {"A": "int32", "B": "foo"}},
+                (TypeError, "data type [\"']foo[\"'] not understood"),
+            ),
+        ],
+    )
+    def test_to_records_dtype(self, kwargs, expected):
+        # see GH#18146
+        df = DataFrame({"A": [1, 2], "B": [0.2, 1.5], "C": ["a", "bc"]})
+
+        if not isinstance(expected, np.rec.recarray):
+            with pytest.raises(expected[0], match=expected[1]):
+                df.to_records(**kwargs)
+        else:
+            result = df.to_records(**kwargs)
+            tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "df,kwargs,expected",
+        [
+            # MultiIndex in the index.
+            (
+                DataFrame(
+                    [[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=list("abc")
+                ).set_index(["a", "b"]),
+                {"column_dtypes": "float64", "index_dtypes": {0: "int32", 1: "int8"}},
+                np.rec.array(
+                    [(1, 2, 3.0), (4, 5, 6.0), (7, 8, 9.0)],
+                    dtype=[
+                        ("a", f"{tm.ENDIAN}i4"),
+                        ("b", "i1"),
+                        ("c", f"{tm.ENDIAN}f8"),
+                    ],
+                ),
+            ),
+            # MultiIndex in the columns.
+            (
+                DataFrame(
+                    [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+                    columns=MultiIndex.from_tuples(
+                        [("a", "d"), ("b", "e"), ("c", "f")]
+                    ),
+                ),
+                {
+                    "column_dtypes": {0: f"{tm.ENDIAN}U1", 2: "float32"},
+                    "index_dtypes": "float32",
+                },
+                np.rec.array(
+                    [(0.0, "1", 2, 3.0), (1.0, "4", 5, 6.0), (2.0, "7", 8, 9.0)],
+                    dtype=[
+                        ("index", f"{tm.ENDIAN}f4"),
+                        ("('a', 'd')", f"{tm.ENDIAN}U1"),
+                        ("('b', 'e')", f"{tm.ENDIAN}i8"),
+                        ("('c', 'f')", f"{tm.ENDIAN}f4"),
+                    ],
+                ),
+            ),
+            # MultiIndex in both the columns and index.
+            (
+                DataFrame(
+                    [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+                    columns=MultiIndex.from_tuples(
+                        [("a", "d"), ("b", "e"), ("c", "f")], names=list("ab")
+                    ),
+                    index=MultiIndex.from_tuples(
+                        [("d", -4), ("d", -5), ("f", -6)], names=list("cd")
+                    ),
+                ),
+                {
+                    "column_dtypes": "float64",
+                    "index_dtypes": {0: f"{tm.ENDIAN}U2", 1: "int8"},
+                },
+                np.rec.array(
+                    [
+                        ("d", -4, 1.0, 2.0, 3.0),
+                        ("d", -5, 4.0, 5.0, 6.0),
+                        ("f", -6, 7, 8, 9.0),
+                    ],
+                    dtype=[
+                        ("c", f"{tm.ENDIAN}U2"),
+                        ("d", "i1"),
+                        ("('a', 'd')", f"{tm.ENDIAN}f8"),
+                        ("('b', 'e')", f"{tm.ENDIAN}f8"),
+                        ("('c', 'f')", f"{tm.ENDIAN}f8"),
+                    ],
+                ),
+            ),
+        ],
+    )
+    def test_to_records_dtype_mi(self, df, kwargs, expected):
+        # see GH#18146
+        result = df.to_records(**kwargs)
+        tm.assert_almost_equal(result, expected)
+
+    def test_to_records_dict_like(self):
+        # see GH#18146
+        class DictLike:
+            def __init__(self, **kwargs) -> None:
+                self.d = kwargs.copy()
+
+            def __getitem__(self, key):
+                return self.d.__getitem__(key)
+
+            def __contains__(self, key) -> bool:
+                return key in self.d
+
+            def keys(self):
+                return self.d.keys()
+
+        df = DataFrame({"A": [1, 2], "B": [0.2, 1.5], "C": ["a", "bc"]})
+
+        dtype_mappings = {
+            "column_dtypes": DictLike(A=np.int8, B=np.float32),
+            "index_dtypes": f"{tm.ENDIAN}U2",
+        }
+
+        result = df.to_records(**dtype_mappings)
+        expected = np.rec.array(
+            [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
+            dtype=[
+                ("index", f"{tm.ENDIAN}U2"),
+                ("A", "i1"),
+                ("B", f"{tm.ENDIAN}f4"),
+                ("C", "O"),
+            ],
+        )
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["UTC", "GMT", "US/Eastern"])
+    def test_to_records_datetimeindex_with_tz(self, tz):
+        # GH#13937
+        dr = date_range("2016-01-01", periods=10, freq="s", tz=tz)
+
+        df = DataFrame({"datetime": dr}, index=dr)
+
+        expected = df.to_records()
+        result = df.tz_convert("UTC").to_records()
+
+        # both converted to UTC, so they are equal
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_to_timestamp.py b/pandas/pandas/tests/frame/methods/test_to_timestamp.py
new file mode 100644
index 0000000000000..0e7e1d595d6be
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_to_timestamp.py
@@ -0,0 +1,154 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+def _get_with_delta(delta, freq="YE-DEC"):
+    return date_range(
+        to_datetime("1/1/2001") + delta,
+        to_datetime("12/31/2009") + delta,
+        freq=freq,
+    )
+
+
+class TestToTimestamp:
+    def test_to_timestamp(self, frame_or_series):
+        K = 5
+        index = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), K)),
+            index=index,
+            columns=["A", "B", "C", "D", "E"],
+        )
+        obj["mix"] = "a"
+        obj = tm.get_obj(obj, frame_or_series)
+
+        exp_index = date_range("1/1/2001", end="12/31/2009", freq="YE-DEC")
+        exp_index = exp_index + Timedelta(1, "D") - Timedelta(1, "ns")
+        result = obj.to_timestamp("D", "end")
+        tm.assert_index_equal(result.index, exp_index)
+        tm.assert_numpy_array_equal(result.values, obj.values)
+        if frame_or_series is Series:
+            assert result.name == "A"
+
+        exp_index = date_range("1/1/2001", end="1/1/2009", freq="YS-JAN")
+        result = obj.to_timestamp("D", "start")
+        tm.assert_index_equal(result.index, exp_index)
+
+        result = obj.to_timestamp(how="start")
+        tm.assert_index_equal(result.index, exp_index)
+
+        delta = timedelta(hours=23)
+        result = obj.to_timestamp("H", "end")
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "h") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.index, exp_index)
+
+        delta = timedelta(hours=23, minutes=59)
+        result = obj.to_timestamp("T", "end")
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "m") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.index, exp_index)
+
+        result = obj.to_timestamp("S", "end")
+        delta = timedelta(hours=23, minutes=59, seconds=59)
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "s") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.index, exp_index)
+
+    def test_to_timestamp_columns(self):
+        K = 5
+        index = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), K)),
+            index=index,
+            columns=["A", "B", "C", "D", "E"],
+        )
+        df["mix"] = "a"
+
+        # columns
+        df = df.T
+
+        exp_index = date_range("1/1/2001", end="12/31/2009", freq="YE-DEC")
+        exp_index = exp_index + Timedelta(1, "D") - Timedelta(1, "ns")
+        result = df.to_timestamp("D", "end", axis=1)
+        tm.assert_index_equal(result.columns, exp_index)
+        tm.assert_numpy_array_equal(result.values, df.values)
+
+        exp_index = date_range("1/1/2001", end="1/1/2009", freq="YS-JAN")
+        result = df.to_timestamp("D", "start", axis=1)
+        tm.assert_index_equal(result.columns, exp_index)
+
+        delta = timedelta(hours=23)
+        result = df.to_timestamp("H", "end", axis=1)
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "h") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.columns, exp_index)
+
+        delta = timedelta(hours=23, minutes=59)
+        result = df.to_timestamp("min", "end", axis=1)
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "m") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.columns, exp_index)
+
+        result = df.to_timestamp("S", "end", axis=1)
+        delta = timedelta(hours=23, minutes=59, seconds=59)
+        exp_index = _get_with_delta(delta)
+        exp_index = exp_index + Timedelta(1, "s") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.columns, exp_index)
+
+        result1 = df.to_timestamp("5min", axis=1)
+        result2 = df.to_timestamp("min", axis=1)
+        expected = date_range("2001-01-01", "2009-01-01", freq="YS")
+        assert isinstance(result1.columns, DatetimeIndex)
+        assert isinstance(result2.columns, DatetimeIndex)
+        tm.assert_numpy_array_equal(result1.columns.asi8, expected.asi8)
+        tm.assert_numpy_array_equal(result2.columns.asi8, expected.asi8)
+        # PeriodIndex.to_timestamp always use 'infer'
+        assert result1.columns.freqstr == "YS-JAN"
+        assert result2.columns.freqstr == "YS-JAN"
+
+    def test_to_timestamp_invalid_axis(self):
+        index = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 5)), index=index
+        )
+
+        # invalid axis
+        with pytest.raises(ValueError, match="axis"):
+            obj.to_timestamp(axis=2)
+
+    def test_to_timestamp_hourly(self, frame_or_series):
+        index = period_range(freq="h", start="1/1/2001", end="1/2/2001")
+        obj = Series(1, index=index, name="foo")
+        if frame_or_series is not Series:
+            obj = obj.to_frame()
+
+        exp_index = date_range("1/1/2001 00:59:59", end="1/2/2001 00:59:59", freq="h")
+        result = obj.to_timestamp(how="end")
+        exp_index = exp_index + Timedelta(1, "s") - Timedelta(1, "ns")
+        tm.assert_index_equal(result.index, exp_index)
+        if frame_or_series is Series:
+            assert result.name == "foo"
+
+    def test_to_timestamp_raises(self, index, frame_or_series):
+        # GH#33327
+        obj = frame_or_series(index=index, dtype=object)
+
+        if not isinstance(index, PeriodIndex):
+            msg = f"unsupported Type {type(index).__name__}"
+            with pytest.raises(TypeError, match=msg):
+                obj.to_timestamp()
diff --git a/pandas/pandas/tests/frame/methods/test_transpose.py b/pandas/pandas/tests/frame/methods/test_transpose.py
new file mode 100644
index 0000000000000..1b7b30ac40363
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_transpose.py
@@ -0,0 +1,197 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    Series,
+    Timestamp,
+    bdate_range,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTranspose:
+    def test_transpose_td64_intervals(self):
+        # GH#44917
+        tdi = timedelta_range("0 Days", "3 Days")
+        ii = IntervalIndex.from_breaks(tdi)
+        ii = ii.insert(-1, np.nan)
+        df = DataFrame(ii)
+
+        result = df.T
+        result.columns = Index(list(range(len(ii))))
+        expected = DataFrame({i: ii[i : i + 1] for i in range(len(ii))})
+        tm.assert_frame_equal(result, expected)
+
+    def test_transpose_empty_preserves_datetimeindex(self):
+        # GH#41382
+        dti = DatetimeIndex([], dtype="M8[ns]")
+        df = DataFrame(index=dti)
+
+        expected = DatetimeIndex([], dtype="datetime64[ns]", freq=None)
+
+        result1 = df.T.sum().index
+        result2 = df.sum(axis=1).index
+
+        tm.assert_index_equal(result1, expected)
+        tm.assert_index_equal(result2, expected)
+
+    def test_transpose_tzaware_1col_single_tz(self):
+        # GH#26825
+        dti = date_range("2016-04-05 04:30", periods=3, tz="UTC")
+
+        df = DataFrame(dti)
+        assert (df.dtypes == dti.dtype).all()
+        res = df.T
+        assert (res.dtypes == dti.dtype).all()
+
+    def test_transpose_tzaware_2col_single_tz(self):
+        # GH#26825
+        dti = date_range("2016-04-05 04:30", periods=3, tz="UTC")
+
+        df3 = DataFrame({"A": dti, "B": dti})
+        assert (df3.dtypes == dti.dtype).all()
+        res3 = df3.T
+        assert (res3.dtypes == dti.dtype).all()
+
+    def test_transpose_tzaware_2col_mixed_tz(self):
+        # GH#26825
+        dti = date_range("2016-04-05 04:30", periods=3, tz="UTC")
+        dti2 = dti.tz_convert("US/Pacific")
+
+        df4 = DataFrame({"A": dti, "B": dti2})
+        assert (df4.dtypes == [dti.dtype, dti2.dtype]).all()
+        assert (df4.T.dtypes == object).all()
+        tm.assert_frame_equal(df4.T.T, df4.astype(object))
+
+    @pytest.mark.parametrize("tz", [None, "America/New_York"])
+    def test_transpose_preserves_dtindex_equality_with_dst(self, tz):
+        # GH#19970
+        idx = date_range("20161101", "20161130", freq="4h", tz=tz)
+        df = DataFrame({"a": range(len(idx)), "b": range(len(idx))}, index=idx)
+        result = df.T == df.T
+        expected = DataFrame(True, index=list("ab"), columns=idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_transpose_object_to_tzaware_mixed_tz(self):
+        # GH#26825
+        dti = date_range("2016-04-05 04:30", periods=3, tz="UTC")
+        dti2 = dti.tz_convert("US/Pacific")
+
+        # mixed all-tzaware dtypes
+        df2 = DataFrame([dti, dti2])
+        assert (df2.dtypes == object).all()
+        res2 = df2.T
+        assert (res2.dtypes == object).all()
+
+    def test_transpose_uint64(self):
+        df = DataFrame(
+            {"A": np.arange(3), "B": [2**63, 2**63 + 5, 2**63 + 10]},
+            dtype=np.uint64,
+        )
+        result = df.T
+        expected = DataFrame(df.values.T)
+        expected.index = ["A", "B"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_transpose_float(self, float_frame):
+        frame = float_frame
+        dft = frame.T
+        for idx, series in dft.items():
+            for col, value in series.items():
+                if np.isnan(value):
+                    assert np.isnan(frame[col][idx])
+                else:
+                    assert value == frame[col][idx]
+
+    def test_transpose_mixed(self):
+        # mixed type
+        mixed = DataFrame(
+            {
+                "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+                "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+                "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+                "D": bdate_range("1/1/2009", periods=5),
+            },
+            index=Index(["a", "b", "c", "d", "e"], dtype=object),
+        )
+
+        mixed_T = mixed.T
+        for col, s in mixed_T.items():
+            assert s.dtype == np.object_
+
+    def test_transpose_get_view(self, float_frame):
+        dft = float_frame.T
+        dft.iloc[:, 5:10] = 5
+        assert (float_frame.values[5:10] != 5).all()
+
+    def test_transpose_get_view_dt64tzget_view(self):
+        dti = date_range("2016-01-01", periods=6, tz="US/Pacific")
+        arr = dti._data.reshape(3, 2)
+        df = DataFrame(arr)
+        assert df._mgr.nblocks == 1
+
+        result = df.T
+        assert result._mgr.nblocks == 1
+
+        rtrip = result._mgr.blocks[0].values
+        assert np.shares_memory(df._mgr.blocks[0].values._ndarray, rtrip._ndarray)
+
+    def test_transpose_not_inferring_dt(self):
+        # GH#51546
+        df = DataFrame(
+            {
+                "a": [Timestamp("2019-12-31"), Timestamp("2019-12-31")],
+            },
+            dtype=object,
+        )
+        result = df.T
+        expected = DataFrame(
+            [[Timestamp("2019-12-31"), Timestamp("2019-12-31")]],
+            index=["a"],
+            dtype=object,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_transpose_not_inferring_dt_mixed_blocks(self):
+        # GH#51546
+        df = DataFrame(
+            {
+                "a": Series(
+                    [Timestamp("2019-12-31"), Timestamp("2019-12-31")], dtype=object
+                ),
+                "b": [Timestamp("2019-12-31"), Timestamp("2019-12-31")],
+            }
+        )
+        result = df.T
+        expected = DataFrame(
+            [
+                [Timestamp("2019-12-31"), Timestamp("2019-12-31")],
+                [Timestamp("2019-12-31"), Timestamp("2019-12-31")],
+            ],
+            index=["a", "b"],
+            dtype=object,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype1", ["Int64", "Float64"])
+    @pytest.mark.parametrize("dtype2", ["Int64", "Float64"])
+    def test_transpose(self, dtype1, dtype2):
+        # GH#57315 - transpose should have F contiguous blocks
+        df = DataFrame(
+            {
+                "a": pd.array([1, 1, 2], dtype=dtype1),
+                "b": pd.array([3, 4, 5], dtype=dtype2),
+            }
+        )
+        result = df.T
+        for blk in result._mgr.blocks:
+            # When dtypes are unequal, we get NumPy object array
+            data = blk.values._data if dtype1 == dtype2 else blk.values
+            assert data.flags["F_CONTIGUOUS"]
diff --git a/pandas/pandas/tests/frame/methods/test_truncate.py b/pandas/pandas/tests/frame/methods/test_truncate.py
new file mode 100644
index 0000000000000..f28f811148c5d
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_truncate.py
@@ -0,0 +1,154 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameTruncate:
+    def test_truncate(self, datetime_frame, frame_or_series):
+        ts = datetime_frame[::3]
+        ts = tm.get_obj(ts, frame_or_series)
+
+        start, end = datetime_frame.index[3], datetime_frame.index[6]
+
+        start_missing = datetime_frame.index[2]
+        end_missing = datetime_frame.index[7]
+
+        # neither specified
+        truncated = ts.truncate()
+        tm.assert_equal(truncated, ts)
+
+        # both specified
+        expected = ts[1:3]
+
+        truncated = ts.truncate(start, end)
+        tm.assert_equal(truncated, expected)
+
+        truncated = ts.truncate(start_missing, end_missing)
+        tm.assert_equal(truncated, expected)
+
+        # start specified
+        expected = ts[1:]
+
+        truncated = ts.truncate(before=start)
+        tm.assert_equal(truncated, expected)
+
+        truncated = ts.truncate(before=start_missing)
+        tm.assert_equal(truncated, expected)
+
+        # end specified
+        expected = ts[:3]
+
+        truncated = ts.truncate(after=end)
+        tm.assert_equal(truncated, expected)
+
+        truncated = ts.truncate(after=end_missing)
+        tm.assert_equal(truncated, expected)
+
+        # corner case, empty series/frame returned
+        truncated = ts.truncate(after=ts.index[0] - ts.index.freq)
+        assert len(truncated) == 0
+
+        truncated = ts.truncate(before=ts.index[-1] + ts.index.freq)
+        assert len(truncated) == 0
+
+        msg = "Truncate: 2000-01-06 00:00:00 must be after 2000-01-11 00:00:00"
+        with pytest.raises(ValueError, match=msg):
+            ts.truncate(
+                before=ts.index[-1] - ts.index.freq, after=ts.index[0] + ts.index.freq
+            )
+
+    def test_truncate_nonsortedindex(self, frame_or_series):
+        # GH#17935
+
+        obj = DataFrame({"A": ["a", "b", "c", "d", "e"]}, index=[5, 3, 2, 9, 0])
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = "truncate requires a sorted index"
+        with pytest.raises(ValueError, match=msg):
+            obj.truncate(before=3, after=9)
+
+    def test_sort_values_nonsortedindex(self):
+        rng = date_range("2011-01-01", "2012-01-01", freq="W")
+        ts = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(len(rng)),
+                "B": np.random.default_rng(2).standard_normal(len(rng)),
+            },
+            index=rng,
+        )
+
+        decreasing = ts.sort_values("A", ascending=False)
+
+        msg = "truncate requires a sorted index"
+        with pytest.raises(ValueError, match=msg):
+            decreasing.truncate(before="2011-11", after="2011-12")
+
+    def test_truncate_nonsortedindex_axis1(self):
+        # GH#17935
+
+        df = DataFrame(
+            {
+                3: np.random.default_rng(2).standard_normal(5),
+                20: np.random.default_rng(2).standard_normal(5),
+                2: np.random.default_rng(2).standard_normal(5),
+                0: np.random.default_rng(2).standard_normal(5),
+            },
+            columns=[3, 20, 2, 0],
+        )
+        msg = "truncate requires a sorted index"
+        with pytest.raises(ValueError, match=msg):
+            df.truncate(before=2, after=20, axis=1)
+
+    @pytest.mark.parametrize(
+        "before, after, indices",
+        [(1, 2, [2, 1]), (None, 2, [2, 1, 0]), (1, None, [3, 2, 1])],
+    )
+    @pytest.mark.parametrize("dtyp", [*tm.ALL_REAL_NUMPY_DTYPES, "datetime64[ns]"])
+    def test_truncate_decreasing_index(
+        self, before, after, indices, dtyp, frame_or_series
+    ):
+        # https://github.com/pandas-dev/pandas/issues/33756
+        idx = Index([3, 2, 1, 0], dtype=dtyp)
+        if isinstance(idx, DatetimeIndex):
+            before = pd.Timestamp(before) if before is not None else None
+            after = pd.Timestamp(after) if after is not None else None
+            indices = [pd.Timestamp(i) for i in indices]
+        values = frame_or_series(range(len(idx)), index=idx)
+        result = values.truncate(before=before, after=after)
+        expected = values.loc[indices]
+        tm.assert_equal(result, expected)
+
+    def test_truncate_multiindex(self, frame_or_series):
+        # GH 34564
+        mi = pd.MultiIndex.from_product([[1, 2, 3, 4], ["A", "B"]], names=["L1", "L2"])
+        s1 = DataFrame(range(mi.shape[0]), index=mi, columns=["col"])
+        s1 = tm.get_obj(s1, frame_or_series)
+
+        result = s1.truncate(before=2, after=3)
+
+        df = DataFrame.from_dict(
+            {"L1": [2, 2, 3, 3], "L2": ["A", "B", "A", "B"], "col": [2, 3, 4, 5]}
+        )
+        expected = df.set_index(["L1", "L2"])
+        expected = tm.get_obj(expected, frame_or_series)
+
+        tm.assert_equal(result, expected)
+
+    def test_truncate_index_only_one_unique_value(self, frame_or_series):
+        # GH 42365
+        obj = Series(0, index=date_range("2021-06-30", "2021-06-30")).repeat(5)
+        if frame_or_series is DataFrame:
+            obj = obj.to_frame(name="a")
+
+        truncated = obj.truncate("2021-06-28", "2021-07-01")
+
+        tm.assert_equal(truncated, obj)
diff --git a/pandas/pandas/tests/frame/methods/test_tz_convert.py b/pandas/pandas/tests/frame/methods/test_tz_convert.py
new file mode 100644
index 0000000000000..5ee4021102f22
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_tz_convert.py
@@ -0,0 +1,140 @@
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestTZConvert:
+    def test_tz_convert(self, frame_or_series):
+        rng = date_range(
+            "1/1/2011", periods=200, freq="D", tz=zoneinfo.ZoneInfo("US/Eastern")
+        )
+
+        obj = DataFrame({"a": 1}, index=rng)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        berlin = zoneinfo.ZoneInfo("Europe/Berlin")
+        result = obj.tz_convert(berlin)
+        expected = DataFrame({"a": 1}, rng.tz_convert(berlin))
+        expected = tm.get_obj(expected, frame_or_series)
+
+        assert result.index.tz.key == "Europe/Berlin"
+        tm.assert_equal(result, expected)
+
+    def test_tz_convert_axis1(self):
+        rng = date_range(
+            "1/1/2011", periods=200, freq="D", tz=zoneinfo.ZoneInfo("US/Eastern")
+        )
+
+        obj = DataFrame({"a": 1}, index=rng)
+
+        obj = obj.T
+        berlin = zoneinfo.ZoneInfo("Europe/Berlin")
+        result = obj.tz_convert(berlin, axis=1)
+        assert result.columns.tz.key == "Europe/Berlin"
+
+        expected = DataFrame({"a": 1}, rng.tz_convert(berlin))
+
+        tm.assert_equal(result, expected.T)
+
+    def test_tz_convert_naive(self, frame_or_series):
+        # can't convert tz-naive
+        rng = date_range("1/1/2011", periods=200, freq="D")
+        ts = Series(1, index=rng)
+        ts = frame_or_series(ts)
+
+        with pytest.raises(TypeError, match="Cannot convert tz-naive"):
+            ts.tz_convert("US/Eastern")
+
+    @pytest.mark.parametrize("fn", ["tz_localize", "tz_convert"])
+    def test_tz_convert_and_localize(self, fn):
+        l0 = date_range("20140701", periods=5, freq="D")
+        l1 = date_range("20140701", periods=5, freq="D")
+
+        int_idx = Index(range(5))
+
+        if fn == "tz_convert":
+            l0 = l0.tz_localize("UTC")
+            l1 = l1.tz_localize("UTC")
+
+        for idx in [l0, l1]:
+            l0_expected = getattr(idx, fn)("US/Pacific")
+            l1_expected = getattr(idx, fn)("US/Pacific")
+
+            df1 = DataFrame(np.ones(5), index=l0)
+            df1 = getattr(df1, fn)("US/Pacific")
+            tm.assert_index_equal(df1.index, l0_expected)
+
+            # MultiIndex
+            # GH7846
+            df2 = DataFrame(np.ones(5), MultiIndex.from_arrays([l0, l1]))
+
+            # freq is not preserved in MultiIndex construction
+            l1_expected = l1_expected._with_freq(None)
+            l0_expected = l0_expected._with_freq(None)
+            l1 = l1._with_freq(None)
+            l0 = l0._with_freq(None)
+
+            df3 = getattr(df2, fn)("US/Pacific", level=0)
+            assert not df3.index.levels[0].equals(l0)
+            tm.assert_index_equal(df3.index.levels[0], l0_expected)
+            tm.assert_index_equal(df3.index.levels[1], l1)
+            assert not df3.index.levels[1].equals(l1_expected)
+
+            df3 = getattr(df2, fn)("US/Pacific", level=1)
+            tm.assert_index_equal(df3.index.levels[0], l0)
+            assert not df3.index.levels[0].equals(l0_expected)
+            tm.assert_index_equal(df3.index.levels[1], l1_expected)
+            assert not df3.index.levels[1].equals(l1)
+
+            df4 = DataFrame(np.ones(5), MultiIndex.from_arrays([int_idx, l0]))
+
+            # TODO: untested
+            getattr(df4, fn)("US/Pacific", level=1)
+
+            tm.assert_index_equal(df3.index.levels[0], l0)
+            assert not df3.index.levels[0].equals(l0_expected)
+            tm.assert_index_equal(df3.index.levels[1], l1_expected)
+            assert not df3.index.levels[1].equals(l1)
+
+    @pytest.mark.parametrize("fn", ["tz_localize", "tz_convert"])
+    def test_tz_convert_and_localize_bad_input(self, fn):
+        int_idx = Index(range(5))
+        l0 = date_range("20140701", periods=5, freq="D")
+        # Not DatetimeIndex / PeriodIndex
+        df = DataFrame(index=int_idx)
+        with pytest.raises(TypeError, match="DatetimeIndex"):
+            getattr(df, fn)("US/Pacific")
+
+        # Not DatetimeIndex / PeriodIndex
+        df = DataFrame(np.ones(5), MultiIndex.from_arrays([int_idx, l0]))
+        with pytest.raises(TypeError, match="DatetimeIndex"):
+            getattr(df, fn)("US/Pacific", level=0)
+
+        # Invalid level
+        df = DataFrame(index=l0)
+        with pytest.raises(ValueError, match="not valid"):
+            getattr(df, fn)("US/Pacific", level=1)
+
+    def test_tz_convert_copy_inplace_mutate(self, frame_or_series):
+        # GH#6326
+        obj = frame_or_series(
+            np.arange(0, 5),
+            index=date_range("20131027", periods=5, freq="h", tz="Europe/Berlin"),
+        )
+        orig = obj.copy()
+        result = obj.tz_convert("UTC")
+        expected = frame_or_series(np.arange(0, 5), index=obj.index.tz_convert("UTC"))
+        tm.assert_equal(result, expected)
+        tm.assert_equal(obj, orig)
+        assert result.index is not obj.index
+        assert result is not obj
diff --git a/pandas/pandas/tests/frame/methods/test_tz_localize.py b/pandas/pandas/tests/frame/methods/test_tz_localize.py
new file mode 100644
index 0000000000000..c0f5a90a9d2ee
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_tz_localize.py
@@ -0,0 +1,67 @@
+from datetime import timezone
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestTZLocalize:
+    # See also:
+    # test_tz_convert_and_localize in test_tz_convert
+
+    def test_tz_localize(self, frame_or_series):
+        rng = date_range("1/1/2011", periods=100, freq="h")
+
+        obj = DataFrame({"a": 1}, index=rng)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        result = obj.tz_localize("utc")
+        expected = DataFrame({"a": 1}, rng.tz_localize("UTC"))
+        expected = tm.get_obj(expected, frame_or_series)
+
+        assert result.index.tz is timezone.utc
+        tm.assert_equal(result, expected)
+
+    def test_tz_localize_axis1(self):
+        rng = date_range("1/1/2011", periods=100, freq="h")
+
+        df = DataFrame({"a": 1}, index=rng)
+
+        df = df.T
+        result = df.tz_localize("utc", axis=1)
+        assert result.columns.tz is timezone.utc
+
+        expected = DataFrame({"a": 1}, rng.tz_localize("UTC"))
+
+        tm.assert_frame_equal(result, expected.T)
+
+    def test_tz_localize_naive(self, frame_or_series):
+        # Can't localize if already tz-aware
+        rng = date_range("1/1/2011", periods=100, freq="h", tz="utc")
+        ts = Series(1, index=rng)
+        ts = frame_or_series(ts)
+
+        with pytest.raises(TypeError, match="Already tz-aware"):
+            ts.tz_localize("US/Eastern")
+
+    def test_tz_localize_copy_inplace_mutate(self, frame_or_series):
+        # GH#6326
+        obj = frame_or_series(
+            np.arange(0, 5), index=date_range("20131027", periods=5, freq="1h", tz=None)
+        )
+        orig = obj.copy()
+        result = obj.tz_localize("UTC")
+        expected = frame_or_series(
+            np.arange(0, 5),
+            index=date_range("20131027", periods=5, freq="1h", tz="UTC"),
+        )
+        tm.assert_equal(result, expected)
+        tm.assert_equal(obj, orig)
+        assert result.index is not obj.index
+        assert result is not obj
diff --git a/pandas/pandas/tests/frame/methods/test_update.py b/pandas/pandas/tests/frame/methods/test_update.py
new file mode 100644
index 0000000000000..ea63b2264d4f6
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_update.py
@@ -0,0 +1,229 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameUpdate:
+    def test_update_nan(self):
+        # #15593 #15617
+        # test 1
+        df1 = DataFrame({"A": [1.0, 2, 3], "B": date_range("2000", periods=3)})
+        df2 = DataFrame({"A": [None, 2, 3]})
+        expected = df1.copy()
+        df1.update(df2, overwrite=False)
+
+        tm.assert_frame_equal(df1, expected)
+
+        # test 2
+        df1 = DataFrame({"A": [1.0, None, 3], "B": date_range("2000", periods=3)})
+        df2 = DataFrame({"A": [None, 2, 3]})
+        expected = DataFrame({"A": [1.0, 2, 3], "B": date_range("2000", periods=3)})
+        df1.update(df2, overwrite=False)
+
+        tm.assert_frame_equal(df1, expected)
+
+    def test_update(self):
+        df = DataFrame(
+            [[1.5, np.nan, 3.0], [1.5, np.nan, 3.0], [1.5, np.nan, 3], [1.5, np.nan, 3]]
+        )
+
+        other = DataFrame([[3.6, 2.0, np.nan], [np.nan, np.nan, 7]], index=[1, 3])
+
+        df.update(other)
+
+        expected = DataFrame(
+            [[1.5, np.nan, 3], [3.6, 2, 3], [1.5, np.nan, 3], [1.5, np.nan, 7.0]]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_dtypes(self):
+        # gh 3016
+        df = DataFrame(
+            [[1.0, 2.0, False, True], [4.0, 5.0, True, False]],
+            columns=["A", "B", "bool1", "bool2"],
+        )
+
+        other = DataFrame([[45, 45]], index=[0], columns=["A", "B"])
+        df.update(other)
+
+        expected = DataFrame(
+            [[45.0, 45.0, False, True], [4.0, 5.0, True, False]],
+            columns=["A", "B", "bool1", "bool2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_nooverwrite(self):
+        df = DataFrame(
+            [[1.5, np.nan, 3.0], [1.5, np.nan, 3.0], [1.5, np.nan, 3], [1.5, np.nan, 3]]
+        )
+
+        other = DataFrame([[3.6, 2.0, np.nan], [np.nan, np.nan, 7]], index=[1, 3])
+
+        df.update(other, overwrite=False)
+
+        expected = DataFrame(
+            [[1.5, np.nan, 3], [1.5, 2, 3], [1.5, np.nan, 3], [1.5, np.nan, 3.0]]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_filtered(self):
+        df = DataFrame(
+            [[1.5, np.nan, 3.0], [1.5, np.nan, 3.0], [1.5, np.nan, 3], [1.5, np.nan, 3]]
+        )
+
+        other = DataFrame([[3.6, 2.0, np.nan], [np.nan, np.nan, 7]], index=[1, 3])
+
+        df.update(other, filter_func=lambda x: x > 2)
+
+        expected = DataFrame(
+            [[1.5, np.nan, 3], [1.5, np.nan, 3], [1.5, np.nan, 3], [1.5, np.nan, 7.0]]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "bad_kwarg, exception, msg",
+        [
+            # errors must be 'ignore' or 'raise'
+            ({"errors": "something"}, ValueError, "The parameter errors must.*"),
+            ({"join": "inner"}, NotImplementedError, "Only left join is supported"),
+        ],
+    )
+    def test_update_raise_bad_parameter(self, bad_kwarg, exception, msg):
+        df = DataFrame([[1.5, 1, 3.0]])
+        with pytest.raises(exception, match=msg):
+            df.update(df, **bad_kwarg)
+
+    def test_update_raise_on_overlap(self):
+        df = DataFrame(
+            [[1.5, 1, 3.0], [1.5, np.nan, 3.0], [1.5, np.nan, 3], [1.5, np.nan, 3]]
+        )
+
+        other = DataFrame([[2.0, np.nan], [np.nan, 7]], index=[1, 3], columns=[1, 2])
+        with pytest.raises(ValueError, match="Data overlaps"):
+            df.update(other, errors="raise")
+
+    def test_update_from_non_df(self):
+        d = {"a": Series([1, 2, 3, 4]), "b": Series([5, 6, 7, 8])}
+        df = DataFrame(d)
+
+        d["a"] = Series([5, 6, 7, 8])
+        df.update(d)
+
+        expected = DataFrame(d)
+
+        tm.assert_frame_equal(df, expected)
+
+        d = {"a": [1, 2, 3, 4], "b": [5, 6, 7, 8]}
+        df = DataFrame(d)
+
+        d["a"] = [5, 6, 7, 8]
+        df.update(d)
+
+        expected = DataFrame(d)
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_datetime_tz(self):
+        # GH 25807
+        result = DataFrame([pd.Timestamp("2019", tz="UTC")])
+        with tm.assert_produces_warning(None):
+            result.update(result)
+        expected = DataFrame([pd.Timestamp("2019", tz="UTC")])
+        tm.assert_frame_equal(result, expected)
+
+    def test_update_datetime_tz_in_place(self):
+        # https://github.com/pandas-dev/pandas/issues/56227
+        result = DataFrame([pd.Timestamp("2019", tz="UTC")])
+        orig = result.copy()
+        view = result[:]
+        result.update(result + pd.Timedelta(days=1))
+        expected = DataFrame([pd.Timestamp("2019-01-02", tz="UTC")])
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(view, orig)
+
+    def test_update_with_different_dtype(self):
+        # GH#3217
+        df = DataFrame({"a": [1, 3], "b": [np.nan, 2]})
+        df["c"] = np.nan
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.update({"c": Series(["foo"], index=[0])})
+
+    def test_update_modify_view(self, using_infer_string):
+        # GH#47188
+        df = DataFrame({"A": ["1", np.nan], "B": ["100", np.nan]})
+        df2 = DataFrame({"A": ["a", "x"], "B": ["100", "200"]})
+        df2_orig = df2.copy()
+        result_view = df2[:]
+        df2.update(df)
+        expected = DataFrame({"A": ["1", "x"], "B": ["100", "200"]})
+        tm.assert_frame_equal(df2, expected)
+        tm.assert_frame_equal(result_view, df2_orig)
+
+    def test_update_dt_column_with_NaT_create_column(self):
+        # GH#16713
+        df = DataFrame({"A": [1, None], "B": [pd.NaT, pd.to_datetime("2016-01-01")]})
+        df2 = DataFrame({"A": [2, 3]})
+        df.update(df2, overwrite=False)
+        expected = DataFrame(
+            {"A": [1.0, 3.0], "B": [pd.NaT, pd.to_datetime("2016-01-01")]}
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "value_df, value_other, dtype",
+        [
+            (True, False, bool),
+            (1, 2, int),
+            (1.0, 2.0, float),
+            (1.0 + 1j, 2.0 + 2j, complex),
+            (np.uint64(1), np.uint(2), np.dtype("ubyte")),
+            (np.uint64(1), np.uint(2), np.dtype("intc")),
+            ("a", "b", pd.StringDtype()),
+            (
+                pd.to_timedelta("1 ms"),
+                pd.to_timedelta("2 ms"),
+                np.dtype("timedelta64[ns]"),
+            ),
+            (
+                np.datetime64("2000-01-01T00:00:00"),
+                np.datetime64("2000-01-02T00:00:00"),
+                np.dtype("datetime64[ns]"),
+            ),
+        ],
+    )
+    def test_update_preserve_dtype(self, value_df, value_other, dtype):
+        # GH#55509
+        df = DataFrame({"a": [value_df] * 2}, index=[1, 2], dtype=dtype)
+        other = DataFrame({"a": [value_other]}, index=[1], dtype=dtype)
+        expected = DataFrame({"a": [value_other, value_df]}, index=[1, 2], dtype=dtype)
+        df.update(other)
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_raises_on_duplicate_argument_index(self):
+        # GH#55509
+        df = DataFrame({"a": [1, 1]}, index=[1, 2])
+        other = DataFrame({"a": [2, 3]}, index=[1, 1])
+        with pytest.raises(ValueError, match="duplicate index"):
+            df.update(other)
+
+    def test_update_raises_without_intersection(self):
+        # GH#55509
+        df = DataFrame({"a": [1]}, index=[1])
+        other = DataFrame({"a": [2]}, index=[2])
+        with pytest.raises(ValueError, match="no intersection"):
+            df.update(other)
+
+    def test_update_on_duplicate_frame_unique_argument_index(self):
+        # GH#55509
+        df = DataFrame({"a": [1, 1, 1]}, index=[1, 1, 2], dtype=np.dtype("intc"))
+        other = DataFrame({"a": [2, 3]}, index=[1, 2], dtype=np.dtype("intc"))
+        expected = DataFrame({"a": [2, 2, 3]}, index=[1, 1, 2], dtype=np.dtype("intc"))
+        df.update(other)
+        tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_value_counts.py b/pandas/pandas/tests/frame/methods/test_value_counts.py
new file mode 100644
index 0000000000000..43db234267f21
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_value_counts.py
@@ -0,0 +1,205 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_data_frame_value_counts_unsorted():
+    df = pd.DataFrame(
+        {"num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        index=["falcon", "dog", "cat", "ant"],
+    )
+
+    result = df.value_counts(sort=False)
+    expected = pd.Series(
+        data=[1, 2, 1],
+        index=pd.MultiIndex.from_arrays(
+            [(2, 4, 6), (2, 0, 0)], names=["num_legs", "num_wings"]
+        ),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_ascending():
+    df = pd.DataFrame(
+        {"num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        index=["falcon", "dog", "cat", "ant"],
+    )
+
+    result = df.value_counts(ascending=True)
+    expected = pd.Series(
+        data=[1, 1, 2],
+        index=pd.MultiIndex.from_arrays(
+            [(2, 6, 4), (2, 0, 0)], names=["num_legs", "num_wings"]
+        ),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_default():
+    df = pd.DataFrame(
+        {"num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        index=["falcon", "dog", "cat", "ant"],
+    )
+
+    result = df.value_counts()
+    expected = pd.Series(
+        data=[2, 1, 1],
+        index=pd.MultiIndex.from_arrays(
+            [(4, 2, 6), (0, 2, 0)], names=["num_legs", "num_wings"]
+        ),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_normalize():
+    df = pd.DataFrame(
+        {"num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        index=["falcon", "dog", "cat", "ant"],
+    )
+
+    result = df.value_counts(normalize=True)
+    expected = pd.Series(
+        data=[0.5, 0.25, 0.25],
+        index=pd.MultiIndex.from_arrays(
+            [(4, 2, 6), (0, 2, 0)], names=["num_legs", "num_wings"]
+        ),
+        name="proportion",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_single_col_default():
+    df = pd.DataFrame({"num_legs": [2, 4, 4, 6]})
+
+    result = df.value_counts()
+    expected = pd.Series(
+        data=[2, 1, 1],
+        index=pd.MultiIndex.from_arrays([[4, 2, 6]], names=["num_legs"]),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_empty():
+    df_no_cols = pd.DataFrame()
+
+    result = df_no_cols.value_counts()
+    expected = pd.Series(
+        [], dtype=np.int64, name="count", index=np.array([], dtype=np.intp)
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_empty_normalize():
+    df_no_cols = pd.DataFrame()
+
+    result = df_no_cols.value_counts(normalize=True)
+    expected = pd.Series(
+        [], dtype=np.float64, name="proportion", index=np.array([], dtype=np.intp)
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_dropna_true(nulls_fixture):
+    # GH 41334
+    df = pd.DataFrame(
+        {
+            "first_name": ["John", "Anne", "John", "Beth"],
+            "middle_name": ["Smith", nulls_fixture, nulls_fixture, "Louise"],
+        },
+    )
+    result = df.value_counts()
+    expected = pd.Series(
+        data=[1, 1],
+        index=pd.MultiIndex.from_arrays(
+            [("John", "Beth"), ("Smith", "Louise")], names=["first_name", "middle_name"]
+        ),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_data_frame_value_counts_dropna_false(nulls_fixture):
+    # GH 41334
+    df = pd.DataFrame(
+        {
+            "first_name": ["John", "Anne", "John", "Beth"],
+            "middle_name": ["Smith", nulls_fixture, nulls_fixture, "Louise"],
+        },
+    )
+
+    result = df.value_counts(dropna=False)
+    expected = pd.Series(
+        data=[1, 1, 1, 1],
+        index=pd.MultiIndex(
+            levels=[
+                pd.Index(["Anne", "Beth", "John"]),
+                pd.Index(["Louise", "Smith", np.nan]),
+            ],
+            codes=[[2, 0, 2, 1], [1, 2, 2, 0]],
+            names=["first_name", "middle_name"],
+        ),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("columns", (["first_name", "middle_name"], [0, 1]))
+def test_data_frame_value_counts_subset(nulls_fixture, columns):
+    # GH 50829
+    df = pd.DataFrame(
+        {
+            columns[0]: ["John", "Anne", "John", "Beth"],
+            columns[1]: ["Smith", nulls_fixture, nulls_fixture, "Louise"],
+        },
+    )
+    result = df.value_counts(columns[0])
+    expected = pd.Series(
+        data=[2, 1, 1],
+        index=pd.Index(["John", "Anne", "Beth"], name=columns[0]),
+        name="count",
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_categorical_future_warning():
+    # GH#54775
+    df = pd.DataFrame({"a": [1, 2, 3]}, dtype="category")
+    result = df.value_counts()
+    expected = pd.Series(
+        1,
+        index=pd.MultiIndex.from_arrays(
+            [pd.Index([1, 2, 3], name="a", dtype="category")]
+        ),
+        name="count",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_with_missing_category():
+    # GH-54836
+    df = pd.DataFrame({"a": pd.Categorical([1, 2, 4], categories=[1, 2, 3, 4])})
+    result = df.value_counts()
+    expected = pd.Series(
+        [1, 1, 1, 0],
+        index=pd.MultiIndex.from_arrays(
+            [pd.CategoricalIndex([1, 2, 4, 3], categories=[1, 2, 3, 4], name="a")]
+        ),
+        name="count",
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/methods/test_values.py b/pandas/pandas/tests/frame/methods/test_values.py
new file mode 100644
index 0000000000000..2de2053bb705f
--- /dev/null
+++ b/pandas/pandas/tests/frame/methods/test_values.py
@@ -0,0 +1,263 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameValues:
+    def test_values(self, float_frame):
+        with pytest.raises(ValueError, match="read-only"):
+            float_frame.values[:, 0] = 5.0
+        assert (float_frame.values[:, 0] != 5).all()
+
+    def test_more_values(self, float_string_frame):
+        values = float_string_frame.values
+        assert values.shape[1] == len(float_string_frame.columns)
+
+    def test_values_mixed_dtypes(self, float_frame, float_string_frame):
+        frame = float_frame
+        arr = frame.values
+
+        frame_cols = frame.columns
+        for i, row in enumerate(arr):
+            for j, value in enumerate(row):
+                col = frame_cols[j]
+                if np.isnan(value):
+                    assert np.isnan(frame[col].iloc[i])
+                else:
+                    assert value == frame[col].iloc[i]
+
+        # mixed type
+        arr = float_string_frame[["foo", "A"]].values
+        assert arr[0, 0] == "bar"
+
+        df = DataFrame({"complex": [1j, 2j, 3j], "real": [1, 2, 3]})
+        arr = df.values
+        assert arr[0, 0] == 1j
+
+    def test_values_duplicates(self):
+        df = DataFrame(
+            [[1, 2, "a", "b"], [1, 2, "a", "b"]], columns=["one", "one", "two", "two"]
+        )
+
+        result = df.values
+        expected = np.array([[1, 2, "a", "b"], [1, 2, "a", "b"]], dtype=object)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_values_with_duplicate_columns(self):
+        df = DataFrame([[1, 2.5], [3, 4.5]], index=[1, 2], columns=["x", "x"])
+        result = df.values
+        expected = np.array([[1, 2.5], [3, 4.5]])
+        assert (result == expected).all().all()
+
+    @pytest.mark.parametrize("constructor", [date_range, period_range])
+    def test_values_casts_datetimelike_to_object(self, constructor):
+        series = Series(constructor("2000-01-01", periods=10, freq="D"))
+
+        expected = series.astype("object")
+
+        df = DataFrame(
+            {"a": series, "b": np.random.default_rng(2).standard_normal(len(series))}
+        )
+
+        result = df.values.squeeze()
+        assert (result[:, 0] == expected.values).all()
+
+        df = DataFrame({"a": series, "b": ["foo"] * len(series)})
+
+        result = df.values.squeeze()
+        assert (result[:, 0] == expected.values).all()
+
+    def test_frame_values_with_tz(self):
+        tz = "US/Central"
+        df = DataFrame({"A": date_range("2000", periods=4, tz=tz)})
+        result = df.values
+        expected = np.array(
+            [
+                [Timestamp("2000-01-01", tz=tz)],
+                [Timestamp("2000-01-02", tz=tz)],
+                [Timestamp("2000-01-03", tz=tz)],
+                [Timestamp("2000-01-04", tz=tz)],
+            ]
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+        # two columns, homogeneous
+
+        df["B"] = df["A"]
+        result = df.values
+        expected = np.concatenate([expected, expected], axis=1)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # three columns, heterogeneous
+        est = "US/Eastern"
+        df["C"] = df["A"].dt.tz_convert(est)
+
+        new = np.array(
+            [
+                [Timestamp("2000-01-01T01:00:00", tz=est)],
+                [Timestamp("2000-01-02T01:00:00", tz=est)],
+                [Timestamp("2000-01-03T01:00:00", tz=est)],
+                [Timestamp("2000-01-04T01:00:00", tz=est)],
+            ]
+        )
+        expected = np.concatenate([expected, new], axis=1)
+        result = df.values
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_interleave_with_tzaware(self, timezone_frame):
+        # interleave with object
+        result = timezone_frame.assign(D="foo").values
+        expected = np.array(
+            [
+                [
+                    Timestamp("2013-01-01 00:00:00"),
+                    Timestamp("2013-01-02 00:00:00"),
+                    Timestamp("2013-01-03 00:00:00"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00-0500", tz="US/Eastern"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00+0100", tz="CET"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00+0100", tz="CET"),
+                ],
+                ["foo", "foo", "foo"],
+            ],
+            dtype=object,
+        ).T
+        tm.assert_numpy_array_equal(result, expected)
+
+        # interleave with only datetime64[ns]
+        result = timezone_frame.values
+        expected = np.array(
+            [
+                [
+                    Timestamp("2013-01-01 00:00:00"),
+                    Timestamp("2013-01-02 00:00:00"),
+                    Timestamp("2013-01-03 00:00:00"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00-0500", tz="US/Eastern"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00+0100", tz="CET"),
+                    NaT,
+                    Timestamp("2013-01-03 00:00:00+0100", tz="CET"),
+                ],
+            ],
+            dtype=object,
+        ).T
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_values_interleave_non_unique_cols(self):
+        df = DataFrame(
+            [[Timestamp("20130101"), 3.5], [Timestamp("20130102"), 4.5]],
+            columns=["x", "x"],
+            index=[1, 2],
+        )
+
+        df_unique = df.copy()
+        df_unique.columns = ["x", "y"]
+        assert df_unique.values.shape == df.values.shape
+        tm.assert_numpy_array_equal(df_unique.values[0], df.values[0])
+        tm.assert_numpy_array_equal(df_unique.values[1], df.values[1])
+
+    def test_values_numeric_cols(self, float_frame):
+        float_frame["foo"] = "bar"
+
+        values = float_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+    def test_values_lcd(self, mixed_float_frame, mixed_int_frame):
+        # mixed lcd
+        values = mixed_float_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+        values = mixed_float_frame[["A", "B", "C"]].values
+        assert values.dtype == np.float32
+
+        values = mixed_float_frame[["C"]].values
+        assert values.dtype == np.float16
+
+        # GH#10364
+        # B uint64 forces float because there are other signed int types
+        values = mixed_int_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+        values = mixed_int_frame[["A", "D"]].values
+        assert values.dtype == np.int64
+
+        # B uint64 forces float because there are other signed int types
+        values = mixed_int_frame[["A", "B", "C"]].values
+        assert values.dtype == np.float64
+
+        # as B and C are both unsigned, no forcing to float is needed
+        values = mixed_int_frame[["B", "C"]].values
+        assert values.dtype == np.uint64
+
+        values = mixed_int_frame[["A", "C"]].values
+        assert values.dtype == np.int32
+
+        values = mixed_int_frame[["C", "D"]].values
+        assert values.dtype == np.int64
+
+        values = mixed_int_frame[["A"]].values
+        assert values.dtype == np.int32
+
+        values = mixed_int_frame[["C"]].values
+        assert values.dtype == np.uint8
+
+
+class TestPrivateValues:
+    def test_private_values_dt64tz(self):
+        dta = date_range("2000", periods=4, tz="US/Central")._data.reshape(-1, 1)
+
+        df = DataFrame(dta, columns=["A"])
+        tm.assert_equal(df._values, dta)
+
+        assert not np.shares_memory(df._values._ndarray, dta._ndarray)
+
+        # TimedeltaArray
+        tda = dta - dta
+        df2 = df - df
+        tm.assert_equal(df2._values, tda)
+
+    def test_private_values_dt64tz_multicol(self):
+        dta = date_range("2000", periods=8, tz="US/Central")._data.reshape(-1, 2)
+
+        df = DataFrame(dta, columns=["A", "B"])
+        tm.assert_equal(df._values, dta)
+
+        assert not np.shares_memory(df._values._ndarray, dta._ndarray)
+
+        # TimedeltaArray
+        tda = dta - dta
+        df2 = df - df
+        tm.assert_equal(df2._values, tda)
+
+    def test_private_values_dt64_multiblock(self):
+        dta = date_range("2000", periods=8)._data
+
+        df = DataFrame({"A": dta[:4]}, copy=False)
+        df["B"] = dta[4:]
+
+        assert len(df._mgr.blocks) == 2
+
+        result = df._values
+        expected = dta.reshape(2, 4).T
+        tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_alter_axes.py b/pandas/pandas/tests/frame/test_alter_axes.py
new file mode 100644
index 0000000000000..b4c16b94fcf8b
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_alter_axes.py
@@ -0,0 +1,31 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestDataFrameAlterAxes:
+    # Tests for setting index/columns attributes directly (i.e. __setattr__)
+
+    def test_set_axis_setattr_index(self):
+        # GH 6785
+        # set the index manually
+
+        df = DataFrame([{"ts": datetime(2014, 4, 1, tzinfo=timezone.utc), "foo": 1}])
+        expected = df.set_index("ts")
+        df.index = df["ts"]
+        df.pop("ts")
+        tm.assert_frame_equal(df, expected)
+
+    # Renaming
+
+    def test_assign_columns(self, float_frame):
+        float_frame["hi"] = "there"
+
+        df = float_frame.copy()
+        df.columns = ["foo", "bar", "baz", "quux", "foo2"]
+        tm.assert_series_equal(float_frame["C"], df["baz"], check_names=False)
+        tm.assert_series_equal(float_frame["hi"], df["foo2"], check_names=False)
diff --git a/pandas/pandas/tests/frame/test_api.py b/pandas/pandas/tests/frame/test_api.py
new file mode 100644
index 0000000000000..2b0bf1b0576f9
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_api.py
@@ -0,0 +1,380 @@
+from copy import deepcopy
+import inspect
+import pydoc
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+from pandas._config.config import option_context
+
+from pandas.compat import HAS_PYARROW
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameMisc:
+    def test_getitem_pop_assign_name(self, float_frame):
+        s = float_frame["A"]
+        assert s.name == "A"
+
+        s = float_frame.pop("A")
+        assert s.name == "A"
+
+        s = float_frame.loc[:, "B"]
+        assert s.name == "B"
+
+        s2 = s.loc[:]
+        assert s2.name == "B"
+
+    def test_get_axis(self, float_frame):
+        f = float_frame
+        assert f._get_axis_number(0) == 0
+        assert f._get_axis_number(1) == 1
+        assert f._get_axis_number("index") == 0
+        assert f._get_axis_number("rows") == 0
+        assert f._get_axis_number("columns") == 1
+
+        assert f._get_axis_name(0) == "index"
+        assert f._get_axis_name(1) == "columns"
+        assert f._get_axis_name("index") == "index"
+        assert f._get_axis_name("rows") == "index"
+        assert f._get_axis_name("columns") == "columns"
+
+        assert f._get_axis(0) is f.index
+        assert f._get_axis(1) is f.columns
+
+        with pytest.raises(ValueError, match="No axis named"):
+            f._get_axis_number(2)
+
+        with pytest.raises(ValueError, match="No axis.*foo"):
+            f._get_axis_name("foo")
+
+        with pytest.raises(ValueError, match="No axis.*None"):
+            f._get_axis_name(None)
+
+        with pytest.raises(ValueError, match="No axis named"):
+            f._get_axis_number(None)
+
+    def test_column_contains_raises(self, float_frame):
+        with pytest.raises(TypeError, match="unhashable type: 'Index'"):
+            float_frame.columns in float_frame
+
+    def test_tab_completion(self):
+        # DataFrame whose columns are identifiers shall have them in __dir__.
+        df = DataFrame([list("abcd"), list("efgh")], columns=list("ABCD"))
+        for key in list("ABCD"):
+            assert key in dir(df)
+        assert isinstance(df.__getitem__("A"), Series)
+
+        # DataFrame whose first-level columns are identifiers shall have
+        # them in __dir__.
+        df = DataFrame(
+            [list("abcd"), list("efgh")],
+            columns=pd.MultiIndex.from_tuples(list(zip("ABCD", "EFGH"))),
+        )
+        for key in list("ABCD"):
+            assert key in dir(df)
+        for key in list("EFGH"):
+            assert key not in dir(df)
+        assert isinstance(df.__getitem__("A"), DataFrame)
+
+    def test_display_max_dir_items(self):
+        # display.max_dir_items increases the number of columns that are in __dir__.
+        columns = ["a" + str(i) for i in range(420)]
+        values = [range(420), range(420)]
+        df = DataFrame(values, columns=columns)
+
+        # The default value for display.max_dir_items is 100
+        assert "a99" in dir(df)
+        assert "a100" not in dir(df)
+
+        with option_context("display.max_dir_items", 300):
+            df = DataFrame(values, columns=columns)
+            assert "a299" in dir(df)
+            assert "a300" not in dir(df)
+
+        with option_context("display.max_dir_items", None):
+            df = DataFrame(values, columns=columns)
+            assert "a419" in dir(df)
+
+    def test_not_hashable(self):
+        empty_frame = DataFrame()
+
+        df = DataFrame([1])
+        msg = "unhashable type: 'DataFrame'"
+        with pytest.raises(TypeError, match=msg):
+            hash(df)
+        with pytest.raises(TypeError, match=msg):
+            hash(empty_frame)
+
+    @pytest.mark.xfail(
+        using_string_dtype() and HAS_PYARROW, reason="surrogates not allowed"
+    )
+    def test_column_name_contains_unicode_surrogate(self):
+        # GH 25509
+        colname = "\ud83d"
+        df = DataFrame({colname: []})
+        # this should not crash
+        assert colname not in dir(df)
+        assert df.columns[0] == colname
+
+    def test_new_empty_index(self):
+        df1 = DataFrame(np.random.default_rng(2).standard_normal((0, 3)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((0, 3)))
+        df1.index.name = "foo"
+        assert df2.index.name is None
+
+    def test_get_agg_axis(self, float_frame):
+        cols = float_frame._get_agg_axis(0)
+        assert cols is float_frame.columns
+
+        idx = float_frame._get_agg_axis(1)
+        assert idx is float_frame.index
+
+        msg = r"Axis must be 0 or 1 \(got 2\)"
+        with pytest.raises(ValueError, match=msg):
+            float_frame._get_agg_axis(2)
+
+    def test_empty(self, float_frame, float_string_frame):
+        empty_frame = DataFrame()
+        assert empty_frame.empty
+
+        assert not float_frame.empty
+        assert not float_string_frame.empty
+
+        # corner case
+        df = DataFrame({"A": [1.0, 2.0, 3.0], "B": ["a", "b", "c"]}, index=np.arange(3))
+        del df["A"]
+        assert not df.empty
+
+    def test_len(self, float_frame):
+        assert len(float_frame) == len(float_frame.index)
+
+        # single block corner case
+        arr = float_frame[["A", "B"]].values
+        expected = float_frame.reindex(columns=["A", "B"]).values
+        tm.assert_almost_equal(arr, expected)
+
+    def test_axis_aliases(self, float_frame):
+        f = float_frame
+
+        # reg name
+        expected = f.sum(axis=0)
+        result = f.sum(axis="index")
+        tm.assert_series_equal(result, expected)
+
+        expected = f.sum(axis=1)
+        result = f.sum(axis="columns")
+        tm.assert_series_equal(result, expected)
+
+    def test_class_axis(self):
+        # GH 18147
+        # no exception and no empty docstring
+        assert pydoc.getdoc(DataFrame.index)
+        assert pydoc.getdoc(DataFrame.columns)
+
+    def test_series_put_names(self, float_string_frame):
+        series = float_string_frame._series
+        for k, v in series.items():
+            assert v.name == k
+
+    def test_empty_nonzero(self):
+        df = DataFrame([1, 2, 3])
+        assert not df.empty
+        df = DataFrame(index=[1], columns=[1])
+        assert not df.empty
+        df = DataFrame(index=["a", "b"], columns=["c", "d"]).dropna()
+        assert df.empty
+        assert df.T.empty
+
+    @pytest.mark.parametrize(
+        "df",
+        [
+            DataFrame(),
+            DataFrame(index=[1]),
+            DataFrame(columns=[1]),
+            DataFrame({1: []}),
+        ],
+    )
+    def test_empty_like(self, df):
+        assert df.empty
+        assert df.T.empty
+
+    def test_with_datetimelikes(self):
+        df = DataFrame(
+            {
+                "A": date_range("20130101", periods=10),
+                "B": timedelta_range("1 day", periods=10),
+            }
+        )
+        t = df.T
+
+        result = t.dtypes.value_counts()
+        expected = Series({np.dtype("object"): 10}, name="count")
+        tm.assert_series_equal(result, expected)
+
+    def test_deepcopy(self, float_frame):
+        cp = deepcopy(float_frame)
+        cp.loc[0, "A"] = 10
+        assert not float_frame.equals(cp)
+
+    def test_inplace_return_self(self):
+        # GH 1893
+
+        data = DataFrame(
+            {"a": ["foo", "bar", "baz", "qux"], "b": [0, 0, 1, 1], "c": [1, 2, 3, 4]}
+        )
+
+        def _check_f(base, f):
+            result = f(base)
+            assert result is None
+
+        # -----DataFrame-----
+
+        # set_index
+        f = lambda x: x.set_index("a", inplace=True)
+        _check_f(data.copy(), f)
+
+        # reset_index
+        f = lambda x: x.reset_index(inplace=True)
+        _check_f(data.set_index("a"), f)
+
+        # drop_duplicates
+        f = lambda x: x.drop_duplicates(inplace=True)
+        _check_f(data.copy(), f)
+
+        # sort
+        f = lambda x: x.sort_values("b", inplace=True)
+        _check_f(data.copy(), f)
+
+        # sort_index
+        f = lambda x: x.sort_index(inplace=True)
+        _check_f(data.copy(), f)
+
+        # fillna
+        f = lambda x: x.fillna(0, inplace=True)
+        _check_f(data.copy(), f)
+
+        # replace
+        f = lambda x: x.replace(1, 0, inplace=True)
+        _check_f(data.copy(), f)
+
+        # rename
+        f = lambda x: x.rename({1: "foo"}, inplace=True)
+        _check_f(data.copy(), f)
+
+        # -----Series-----
+        d = data.copy()["c"]
+
+        # reset_index
+        f = lambda x: x.reset_index(inplace=True, drop=True)
+        _check_f(data.set_index("a")["c"], f)
+
+        # fillna
+        f = lambda x: x.fillna(0, inplace=True)
+        _check_f(d.copy(), f)
+
+        # replace
+        f = lambda x: x.replace(1, 0, inplace=True)
+        _check_f(d.copy(), f)
+
+        # rename
+        f = lambda x: x.rename({1: "foo"}, inplace=True)
+        _check_f(d.copy(), f)
+
+    def test_tab_complete_warning(self, ip, frame_or_series):
+        # GH 16409
+        pytest.importorskip("IPython", minversion="6.0.0")
+        from IPython.core.completer import provisionalcompleter
+
+        if frame_or_series is DataFrame:
+            code = "from pandas import DataFrame; obj = DataFrame()"
+        else:
+            code = "from pandas import Series; obj = Series(dtype=object)"
+
+        ip.run_cell(code)
+        # GH 31324 newer jedi version raises Deprecation warning;
+        #  appears resolved 2021-02-02
+        with tm.assert_produces_warning(None, raise_on_extra_warnings=False):
+            with provisionalcompleter("ignore"):
+                list(ip.Completer.completions("obj.", 1))
+
+    def test_attrs(self):
+        df = DataFrame({"A": [2, 3]})
+        assert df.attrs == {}
+        df.attrs["version"] = 1
+
+        result = df.rename(columns=str)
+        assert result.attrs == {"version": 1}
+
+    def test_attrs_deepcopy(self):
+        df = DataFrame({"A": [2, 3]})
+        assert df.attrs == {}
+        df.attrs["tags"] = {"spam", "ham"}
+
+        result = df.rename(columns=str)
+        assert result.attrs == df.attrs
+        assert result.attrs["tags"] is not df.attrs["tags"]
+
+    @pytest.mark.parametrize("allows_duplicate_labels", [True, False, None])
+    def test_set_flags(
+        self,
+        allows_duplicate_labels,
+        frame_or_series,
+    ):
+        obj = DataFrame({"A": [1, 2]})
+        key = (0, 0)
+        if frame_or_series is Series:
+            obj = obj["A"]
+            key = 0
+
+        result = obj.set_flags(allows_duplicate_labels=allows_duplicate_labels)
+
+        if allows_duplicate_labels is None:
+            # We don't update when it's not provided
+            assert result.flags.allows_duplicate_labels is True
+        else:
+            assert result.flags.allows_duplicate_labels is allows_duplicate_labels
+
+        # We made a copy
+        assert obj is not result
+
+        # We didn't mutate obj
+        assert obj.flags.allows_duplicate_labels is True
+
+        # But we didn't copy data
+        if frame_or_series is Series:
+            assert np.may_share_memory(obj.values, result.values)
+        else:
+            assert np.may_share_memory(obj["A"].values, result["A"].values)
+
+        result.iloc[key] = 0
+        assert obj.iloc[key] == 1
+
+        # Now we do copy.
+        result = obj.set_flags(allows_duplicate_labels=allows_duplicate_labels)
+        result.iloc[key] = 10
+        assert obj.iloc[key] == 1
+
+    def test_constructor_expanddim(self):
+        # GH#33628 accessing _constructor_expanddim should not raise NotImplementedError
+        # GH38782 pandas has no container higher than DataFrame (two-dim), so
+        # DataFrame._constructor_expand_dim, doesn't make sense, so is removed.
+        df = DataFrame()
+
+        msg = "'DataFrame' object has no attribute '_constructor_expanddim'"
+        with pytest.raises(AttributeError, match=msg):
+            df._constructor_expanddim(np.arange(27).reshape(3, 3, 3))
+
+    def test_inspect_getmembers(self):
+        # GH38740
+        df = DataFrame()
+        inspect.getmembers(df)
diff --git a/pandas/pandas/tests/frame/test_arithmetic.py b/pandas/pandas/tests/frame/test_arithmetic.py
new file mode 100644
index 0000000000000..8239de3f39c20
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_arithmetic.py
@@ -0,0 +1,2164 @@
+from collections import deque
+from datetime import (
+    datetime,
+    timezone,
+)
+from enum import Enum
+import functools
+import operator
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import HAS_PYARROW
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.computation import expressions as expr
+from pandas.tests.frame.common import (
+    _check_mixed_float,
+    _check_mixed_int,
+)
+
+
+@pytest.fixture
+def simple_frame():
+    """
+    Fixture for simple 3x3 DataFrame
+
+    Columns are ['one', 'two', 'three'], index is ['a', 'b', 'c'].
+
+       one  two  three
+    a  1.0  2.0    3.0
+    b  4.0  5.0    6.0
+    c  7.0  8.0    9.0
+    """
+    arr = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0]])
+
+    return DataFrame(arr, columns=["one", "two", "three"], index=["a", "b", "c"])
+
+
+@pytest.fixture(autouse=True, params=[0, 100], ids=["numexpr", "python"])
+def switch_numexpr_min_elements(request, monkeypatch):
+    with monkeypatch.context() as m:
+        m.setattr(expr, "_MIN_ELEMENTS", request.param)
+        yield request.param
+
+
+class DummyElement:
+    def __init__(self, value, dtype) -> None:
+        self.value = value
+        self.dtype = np.dtype(dtype)
+
+    def __array__(self, dtype=None, copy=None):
+        return np.array(self.value, dtype=self.dtype)
+
+    def __str__(self) -> str:
+        return f"DummyElement({self.value}, {self.dtype})"
+
+    def __repr__(self) -> str:
+        return str(self)
+
+    def astype(self, dtype, copy=False):
+        self.dtype = dtype
+        return self
+
+    def view(self, dtype):
+        return type(self)(self.value.view(dtype), dtype)
+
+    def any(self, axis=None):
+        return bool(self.value)
+
+
+# -------------------------------------------------------------------
+# Comparisons
+
+
+class TestFrameComparisons:
+    # Specifically _not_ flex-comparisons
+
+    def test_comparison_with_categorical_dtype(self):
+        # GH#12564
+
+        df = DataFrame({"A": ["foo", "bar", "baz"]})
+        exp = DataFrame({"A": [True, False, False]})
+
+        res = df == "foo"
+        tm.assert_frame_equal(res, exp)
+
+        # casting to categorical shouldn't affect the result
+        df["A"] = df["A"].astype("category")
+
+        res = df == "foo"
+        tm.assert_frame_equal(res, exp)
+
+    def test_frame_in_list(self):
+        # GH#12689 this should raise at the DataFrame level, not blocks
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)), columns=list("ABCD")
+        )
+        msg = "The truth value of a DataFrame is ambiguous"
+        with pytest.raises(ValueError, match=msg):
+            df in [None]
+
+    @pytest.mark.parametrize(
+        "arg, arg2",
+        [
+            [
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": pd.date_range("20010101", periods=10),
+                },
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": np.random.default_rng(2).integers(10, size=10),
+                },
+            ],
+            [
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": np.random.default_rng(2).integers(10, size=10),
+                },
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": pd.date_range("20010101", periods=10),
+                },
+            ],
+            [
+                {
+                    "a": pd.date_range("20010101", periods=10),
+                    "b": pd.date_range("20010101", periods=10),
+                },
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": np.random.default_rng(2).integers(10, size=10),
+                },
+            ],
+            [
+                {
+                    "a": np.random.default_rng(2).integers(10, size=10),
+                    "b": pd.date_range("20010101", periods=10),
+                },
+                {
+                    "a": pd.date_range("20010101", periods=10),
+                    "b": pd.date_range("20010101", periods=10),
+                },
+            ],
+        ],
+    )
+    def test_comparison_invalid(self, arg, arg2):
+        # GH4968
+        # invalid date/int comparisons
+        x = DataFrame(arg)
+        y = DataFrame(arg2)
+        # we expect the result to match Series comparisons for
+        # == and !=, inequalities should raise
+        result = x == y
+        expected = DataFrame(
+            {col: x[col] == y[col] for col in x.columns},
+            index=x.index,
+            columns=x.columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = x != y
+        expected = DataFrame(
+            {col: x[col] != y[col] for col in x.columns},
+            index=x.index,
+            columns=x.columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+        msgs = [
+            r"Invalid comparison between dtype=datetime64\[ns\] and ndarray",
+            "invalid type promotion",
+            (
+                # npdev 1.20.0
+                r"The DTypes <class 'numpy.dtype\[.*\]'> and "
+                r"<class 'numpy.dtype\[.*\]'> do not have a common DType."
+            ),
+        ]
+        msg = "|".join(msgs)
+        with pytest.raises(TypeError, match=msg):
+            x >= y
+        with pytest.raises(TypeError, match=msg):
+            x > y
+        with pytest.raises(TypeError, match=msg):
+            x < y
+        with pytest.raises(TypeError, match=msg):
+            x <= y
+
+    @pytest.mark.parametrize(
+        "left, right",
+        [
+            ("gt", "lt"),
+            ("lt", "gt"),
+            ("ge", "le"),
+            ("le", "ge"),
+            ("eq", "eq"),
+            ("ne", "ne"),
+        ],
+    )
+    def test_timestamp_compare(self, left, right):
+        # make sure we can compare Timestamps on the right AND left hand side
+        # GH#4982
+        df = DataFrame(
+            {
+                "dates1": pd.date_range("20010101", periods=10),
+                "dates2": pd.date_range("20010102", periods=10),
+                "intcol": np.random.default_rng(2).integers(1000000000, size=10),
+                "floatcol": np.random.default_rng(2).standard_normal(10),
+                "stringcol": [chr(100 + i) for i in range(10)],
+            }
+        )
+        df.loc[np.random.default_rng(2).random(len(df)) > 0.5, "dates2"] = pd.NaT
+        left_f = getattr(operator, left)
+        right_f = getattr(operator, right)
+
+        # no nats
+        if left in ["eq", "ne"]:
+            expected = left_f(df, pd.Timestamp("20010109"))
+            result = right_f(pd.Timestamp("20010109"), df)
+            tm.assert_frame_equal(result, expected)
+        else:
+            msg = (
+                "'(<|>)=?' not supported between "
+                "instances of 'numpy.ndarray' and 'Timestamp'"
+            )
+            with pytest.raises(TypeError, match=msg):
+                left_f(df, pd.Timestamp("20010109"))
+            with pytest.raises(TypeError, match=msg):
+                right_f(pd.Timestamp("20010109"), df)
+        # nats
+        if left in ["eq", "ne"]:
+            expected = left_f(df, pd.Timestamp("nat"))
+            result = right_f(pd.Timestamp("nat"), df)
+            tm.assert_frame_equal(result, expected)
+        else:
+            msg = (
+                "'(<|>)=?' not supported between "
+                "instances of 'numpy.ndarray' and 'NaTType'"
+            )
+            with pytest.raises(TypeError, match=msg):
+                left_f(df, pd.Timestamp("nat"))
+            with pytest.raises(TypeError, match=msg):
+                right_f(pd.Timestamp("nat"), df)
+
+    def test_mixed_comparison(self):
+        # GH#13128, GH#22163 != datetime64 vs non-dt64 should be False,
+        # not raise TypeError
+        # (this appears to be fixed before GH#22163, not sure when)
+        df = DataFrame([["1989-08-01", 1], ["1989-08-01", 2]])
+        other = DataFrame([["a", "b"], ["c", "d"]])
+
+        result = df == other
+        assert not result.any().any()
+
+        result = df != other
+        assert result.all().all()
+
+    def test_df_boolean_comparison_error(self):
+        # GH#4576, GH#22880
+        # comparing DataFrame against list/tuple with len(obj) matching
+        #  len(df.columns) is supported as of GH#22800
+        df = DataFrame(np.arange(6).reshape((3, 2)))
+
+        expected = DataFrame([[False, False], [True, False], [False, False]])
+
+        result = df == (2, 2)
+        tm.assert_frame_equal(result, expected)
+
+        result = df == [2, 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_float_none_comparison(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 3)),
+            index=range(8),
+            columns=["A", "B", "C"],
+        )
+
+        result = df.__eq__(None)
+        assert not result.any().any()
+
+    def test_df_string_comparison(self):
+        df = DataFrame([{"a": 1, "b": "foo"}, {"a": 2, "b": "bar"}])
+        mask_a = df.a > 1
+        tm.assert_frame_equal(df[mask_a], df.loc[1:1, :])
+        tm.assert_frame_equal(df[-mask_a], df.loc[0:0, :])
+
+        mask_b = df.b == "foo"
+        tm.assert_frame_equal(df[mask_b], df.loc[0:0, :])
+        tm.assert_frame_equal(df[-mask_b], df.loc[1:1, :])
+
+
+class TestFrameFlexComparisons:
+    # TODO: test_bool_flex_frame needs a better name
+    def test_bool_flex_frame(self, comparison_op):
+        data = np.random.default_rng(2).standard_normal((5, 3))
+        other_data = np.random.default_rng(2).standard_normal((5, 3))
+        df = DataFrame(data)
+        other = DataFrame(other_data)
+        ndim_5 = np.ones(df.shape + (1, 3))
+
+        # DataFrame
+        assert df.eq(df).values.all()
+        assert not df.ne(df).values.any()
+        f = getattr(df, comparison_op.__name__)
+        o = comparison_op
+        # No NAs
+        tm.assert_frame_equal(f(other), o(df, other))
+        # Unaligned
+        part_o = other.loc[3:, 1:].copy()
+        rs = f(part_o)
+        xp = o(df, part_o.reindex(index=df.index, columns=df.columns))
+        tm.assert_frame_equal(rs, xp)
+        # ndarray
+        tm.assert_frame_equal(f(other.values), o(df, other.values))
+        # scalar
+        tm.assert_frame_equal(f(0), o(df, 0))
+        # NAs
+        msg = "Unable to coerce to Series/DataFrame"
+        tm.assert_frame_equal(f(np.nan), o(df, np.nan))
+        with pytest.raises(ValueError, match=msg):
+            f(ndim_5)
+
+    @pytest.mark.parametrize("box", [np.array, Series])
+    def test_bool_flex_series(self, box):
+        # Series
+        # list/tuple
+        data = np.random.default_rng(2).standard_normal((5, 3))
+        df = DataFrame(data)
+        idx_ser = box(np.random.default_rng(2).standard_normal(5))
+        col_ser = box(np.random.default_rng(2).standard_normal(3))
+
+        idx_eq = df.eq(idx_ser, axis=0)
+        col_eq = df.eq(col_ser)
+        idx_ne = df.ne(idx_ser, axis=0)
+        col_ne = df.ne(col_ser)
+        tm.assert_frame_equal(col_eq, df == Series(col_ser))
+        tm.assert_frame_equal(col_eq, -col_ne)
+        tm.assert_frame_equal(idx_eq, -idx_ne)
+        tm.assert_frame_equal(idx_eq, df.T.eq(idx_ser).T)
+        tm.assert_frame_equal(col_eq, df.eq(list(col_ser)))
+        tm.assert_frame_equal(idx_eq, df.eq(Series(idx_ser), axis=0))
+        tm.assert_frame_equal(idx_eq, df.eq(list(idx_ser), axis=0))
+
+        idx_gt = df.gt(idx_ser, axis=0)
+        col_gt = df.gt(col_ser)
+        idx_le = df.le(idx_ser, axis=0)
+        col_le = df.le(col_ser)
+
+        tm.assert_frame_equal(col_gt, df > Series(col_ser))
+        tm.assert_frame_equal(col_gt, -col_le)
+        tm.assert_frame_equal(idx_gt, -idx_le)
+        tm.assert_frame_equal(idx_gt, df.T.gt(idx_ser).T)
+
+        idx_ge = df.ge(idx_ser, axis=0)
+        col_ge = df.ge(col_ser)
+        idx_lt = df.lt(idx_ser, axis=0)
+        col_lt = df.lt(col_ser)
+        tm.assert_frame_equal(col_ge, df >= Series(col_ser))
+        tm.assert_frame_equal(col_ge, -col_lt)
+        tm.assert_frame_equal(idx_ge, -idx_lt)
+        tm.assert_frame_equal(idx_ge, df.T.ge(idx_ser).T)
+
+        idx_ser = Series(np.random.default_rng(2).standard_normal(5))
+        col_ser = Series(np.random.default_rng(2).standard_normal(3))
+
+    def test_bool_flex_frame_na(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        # NA
+        df.loc[0, 0] = np.nan
+        rs = df.eq(df)
+        assert not rs.loc[0, 0]
+        rs = df.ne(df)
+        assert rs.loc[0, 0]
+        rs = df.gt(df)
+        assert not rs.loc[0, 0]
+        rs = df.lt(df)
+        assert not rs.loc[0, 0]
+        rs = df.ge(df)
+        assert not rs.loc[0, 0]
+        rs = df.le(df)
+        assert not rs.loc[0, 0]
+
+    def test_bool_flex_frame_complex_dtype(self):
+        # complex
+        arr = np.array([np.nan, 1, 6, np.nan])
+        arr2 = np.array([2j, np.nan, 7, None])
+        df = DataFrame({"a": arr})
+        df2 = DataFrame({"a": arr2})
+
+        msg = "|".join(
+            [
+                "'>' not supported between instances of '.*' and 'complex'",
+                r"unorderable types: .*complex\(\)",  # PY35
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            # inequalities are not well-defined for complex numbers
+            df.gt(df2)
+        with pytest.raises(TypeError, match=msg):
+            # regression test that we get the same behavior for Series
+            df["a"].gt(df2["a"])
+        with pytest.raises(TypeError, match=msg):
+            # Check that we match numpy behavior here
+            df.values > df2.values
+
+        rs = df.ne(df2)
+        assert rs.values.all()
+
+        arr3 = np.array([2j, np.nan, None])
+        df3 = DataFrame({"a": arr3})
+
+        with pytest.raises(TypeError, match=msg):
+            # inequalities are not well-defined for complex numbers
+            df3.gt(2j)
+        with pytest.raises(TypeError, match=msg):
+            # regression test that we get the same behavior for Series
+            df3["a"].gt(2j)
+        with pytest.raises(TypeError, match=msg):
+            # Check that we match numpy behavior here
+            df3.values > 2j
+
+    def test_bool_flex_frame_object_dtype(self):
+        # corner, dtype=object
+        df1 = DataFrame({"col": ["foo", np.nan, "bar"]}, dtype=object)
+        df2 = DataFrame({"col": ["foo", datetime.now(), "bar"]}, dtype=object)
+        result = df1.ne(df2)
+        exp = DataFrame({"col": [False, True, False]})
+        tm.assert_frame_equal(result, exp)
+
+    def test_flex_comparison_nat(self):
+        # GH 15697, GH 22163 df.eq(pd.NaT) should behave like df == pd.NaT,
+        # and _definitely_ not be NaN
+        df = DataFrame([pd.NaT])
+
+        result = df == pd.NaT
+        # result.iloc[0, 0] is a np.bool_ object
+        assert result.iloc[0, 0].item() is False
+
+        result = df.eq(pd.NaT)
+        assert result.iloc[0, 0].item() is False
+
+        result = df != pd.NaT
+        assert result.iloc[0, 0].item() is True
+
+        result = df.ne(pd.NaT)
+        assert result.iloc[0, 0].item() is True
+
+    def test_df_flex_cmp_constant_return_types(self, comparison_op):
+        # GH 15077, non-empty DataFrame
+        df = DataFrame({"x": [1, 2, 3], "y": [1.0, 2.0, 3.0]})
+        const = 2
+
+        result = getattr(df, comparison_op.__name__)(const).dtypes.value_counts()
+        tm.assert_series_equal(
+            result, Series([2], index=[np.dtype(bool)], name="count")
+        )
+
+    def test_df_flex_cmp_constant_return_types_empty(self, comparison_op):
+        # GH 15077 empty DataFrame
+        df = DataFrame({"x": [1, 2, 3], "y": [1.0, 2.0, 3.0]})
+        const = 2
+
+        empty = df.iloc[:0]
+        result = getattr(empty, comparison_op.__name__)(const).dtypes.value_counts()
+        tm.assert_series_equal(
+            result, Series([2], index=[np.dtype(bool)], name="count")
+        )
+
+    def test_df_flex_cmp_ea_dtype_with_ndarray_series(self):
+        ii = pd.IntervalIndex.from_breaks([1, 2, 3])
+        df = DataFrame({"A": ii, "B": ii})
+
+        ser = Series([0, 0])
+        res = df.eq(ser, axis=0)
+
+        expected = DataFrame({"A": [False, False], "B": [False, False]})
+        tm.assert_frame_equal(res, expected)
+
+        ser2 = Series([1, 2], index=["A", "B"])
+        res2 = df.eq(ser2, axis=1)
+        tm.assert_frame_equal(res2, expected)
+
+
+# -------------------------------------------------------------------
+# Arithmetic
+
+
+class TestFrameFlexArithmetic:
+    def test_floordiv_axis0(self):
+        # make sure we df.floordiv(ser, axis=0) matches column-wise result
+        arr = np.arange(3)
+        ser = Series(arr)
+        df = DataFrame({"A": ser, "B": ser})
+
+        result = df.floordiv(ser, axis=0)
+
+        expected = DataFrame({col: df[col] // ser for col in df.columns})
+
+        tm.assert_frame_equal(result, expected)
+
+        result2 = df.floordiv(ser.values, axis=0)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_df_add_td64_columnwise(self):
+        # GH 22534 Check that column-wise addition broadcasts correctly
+        dti = pd.date_range("2016-01-01", periods=10)
+        tdi = pd.timedelta_range("1", periods=10)
+        tser = Series(tdi)
+        df = DataFrame({0: dti, 1: tdi})
+
+        result = df.add(tser, axis=0)
+        expected = DataFrame({0: dti + tdi, 1: tdi + tdi})
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_add_flex_filled_mixed_dtypes(self):
+        # GH 19611
+        dti = pd.date_range("2016-01-01", periods=3)
+        ser = Series(["1 Day", "NaT", "2 Days"], dtype="timedelta64[ns]")
+        df = DataFrame({"A": dti, "B": ser})
+        other = DataFrame({"A": ser, "B": ser})
+        fill = pd.Timedelta(days=1).to_timedelta64()
+        result = df.add(other, fill_value=fill)
+
+        expected = DataFrame(
+            {
+                "A": Series(
+                    ["2016-01-02", "2016-01-03", "2016-01-05"], dtype="datetime64[ns]"
+                ),
+                "B": ser * 2,
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_arith_flex_frame(
+        self, all_arithmetic_operators, float_frame, mixed_float_frame
+    ):
+        # one instance of parametrized fixture
+        op = all_arithmetic_operators
+
+        def f(x, y):
+            # r-versions not in operator-stdlib; get op without "r" and invert
+            if op.startswith("__r"):
+                return getattr(operator, op.replace("__r", "__"))(y, x)
+            return getattr(operator, op)(x, y)
+
+        result = getattr(float_frame, op)(2 * float_frame)
+        expected = f(float_frame, 2 * float_frame)
+        tm.assert_frame_equal(result, expected)
+
+        # vs mix float
+        result = getattr(mixed_float_frame, op)(2 * mixed_float_frame)
+        expected = f(mixed_float_frame, 2 * mixed_float_frame)
+        tm.assert_frame_equal(result, expected)
+        _check_mixed_float(result, dtype={"C": None})
+
+    @pytest.mark.parametrize("op", ["__add__", "__sub__", "__mul__"])
+    def test_arith_flex_frame_mixed(
+        self,
+        op,
+        int_frame,
+        mixed_int_frame,
+        mixed_float_frame,
+        switch_numexpr_min_elements,
+    ):
+        f = getattr(operator, op)
+
+        # vs mix int
+        result = getattr(mixed_int_frame, op)(2 + mixed_int_frame)
+        expected = f(mixed_int_frame, 2 + mixed_int_frame)
+
+        # no overflow in the uint
+        dtype = None
+        if op in ["__sub__"]:
+            dtype = {"B": "uint64", "C": None}
+        elif op in ["__add__", "__mul__"]:
+            dtype = {"C": None}
+        if expr.USE_NUMEXPR and switch_numexpr_min_elements == 0:
+            # when using numexpr, the casting rules are slightly different:
+            # in the `2 + mixed_int_frame` operation, int32 column becomes
+            # and int64 column (not preserving dtype in operation with Python
+            # scalar), and then the int32/int64 combo results in int64 result
+            dtype["A"] = (2 + mixed_int_frame)["A"].dtype
+        tm.assert_frame_equal(result, expected)
+        _check_mixed_int(result, dtype=dtype)
+
+        # vs mix float
+        result = getattr(mixed_float_frame, op)(2 * mixed_float_frame)
+        expected = f(mixed_float_frame, 2 * mixed_float_frame)
+        tm.assert_frame_equal(result, expected)
+        _check_mixed_float(result, dtype={"C": None})
+
+        # vs plain int
+        result = getattr(int_frame, op)(2 * int_frame)
+        expected = f(int_frame, 2 * int_frame)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dim", range(3, 6))
+    def test_arith_flex_frame_raise(self, all_arithmetic_operators, float_frame, dim):
+        # one instance of parametrized fixture
+        op = all_arithmetic_operators
+
+        # Check that arrays with dim >= 3 raise
+        arr = np.ones((1,) * dim)
+        msg = "Unable to coerce to Series/DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            getattr(float_frame, op)(arr)
+
+    def test_arith_flex_frame_corner(self, float_frame):
+        const_add = float_frame.add(1)
+        tm.assert_frame_equal(const_add, float_frame + 1)
+
+        # corner cases
+        result = float_frame.add(float_frame[:0])
+        expected = float_frame.sort_index() * np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = float_frame[:0].add(float_frame)
+        expected = float_frame.sort_index() * np.nan
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(NotImplementedError, match="fill_value"):
+            float_frame.add(float_frame.iloc[0], fill_value=3)
+
+        with pytest.raises(NotImplementedError, match="fill_value"):
+            float_frame.add(float_frame.iloc[0], axis="index", fill_value=3)
+
+    @pytest.mark.parametrize("op", ["add", "sub", "mul", "mod"])
+    def test_arith_flex_series_ops(self, simple_frame, op):
+        # after arithmetic refactor, add truediv here
+        df = simple_frame
+
+        row = df.xs("a")
+        col = df["two"]
+        f = getattr(df, op)
+        op = getattr(operator, op)
+        tm.assert_frame_equal(f(row), op(df, row))
+        tm.assert_frame_equal(f(col, axis=0), op(df.T, col).T)
+
+    def test_arith_flex_series(self, simple_frame):
+        df = simple_frame
+
+        row = df.xs("a")
+        col = df["two"]
+        # special case for some reason
+        tm.assert_frame_equal(df.add(row, axis=None), df + row)
+
+        # cases which will be refactored after big arithmetic refactor
+        tm.assert_frame_equal(df.div(row), df / row)
+        tm.assert_frame_equal(df.div(col, axis=0), (df.T / col).T)
+
+    def test_arith_flex_series_broadcasting(self, any_real_numpy_dtype):
+        # broadcasting issue in GH 7325
+        df = DataFrame(np.arange(3 * 2).reshape((3, 2)), dtype=any_real_numpy_dtype)
+        expected = DataFrame([[np.nan, np.inf], [1.0, 1.5], [1.0, 1.25]])
+        if any_real_numpy_dtype == "float32":
+            expected = expected.astype(any_real_numpy_dtype)
+        result = df.div(df[0], axis="index")
+        tm.assert_frame_equal(result, expected)
+
+    def test_arith_flex_zero_len_raises(self):
+        # GH 19522 passing fill_value to frame flex arith methods should
+        # raise even in the zero-length special cases
+        ser_len0 = Series([], dtype=object)
+        df_len0 = DataFrame(columns=["A", "B"])
+        df = DataFrame([[1, 2], [3, 4]], columns=["A", "B"])
+
+        with pytest.raises(NotImplementedError, match="fill_value"):
+            df.add(ser_len0, fill_value="E")
+
+        with pytest.raises(NotImplementedError, match="fill_value"):
+            df_len0.sub(df["A"], axis=None, fill_value=3)
+
+    def test_flex_add_scalar_fill_value(self):
+        # GH#12723
+        dat = np.array([0, 1, np.nan, 3, 4, 5], dtype="float")
+        df = DataFrame({"foo": dat}, index=range(6))
+
+        exp = df.fillna(0).add(2)
+        res = df.add(2, fill_value=0)
+        tm.assert_frame_equal(res, exp)
+
+    def test_sub_alignment_with_duplicate_index(self):
+        # GH#5185 dup aligning operations should work
+        df1 = DataFrame([1, 2, 3, 4, 5], index=[1, 2, 1, 2, 3])
+        df2 = DataFrame([1, 2, 3], index=[1, 2, 3])
+        expected = DataFrame([0, 2, 0, 2, 2], index=[1, 1, 2, 2, 3])
+        result = df1.sub(df2)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("op", ["__add__", "__mul__", "__sub__", "__truediv__"])
+    def test_arithmetic_with_duplicate_columns(self, op):
+        # operations
+        df = DataFrame({"A": np.arange(10), "B": np.random.default_rng(2).random(10)})
+        expected = getattr(df, op)(df)
+        expected.columns = ["A", "A"]
+        df.columns = ["A", "A"]
+        result = getattr(df, op)(df)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("level", [0, None])
+    def test_broadcast_multiindex(self, level):
+        # GH34388
+        df1 = DataFrame({"A": [0, 1, 2], "B": [1, 2, 3]})
+        df1.columns = df1.columns.set_names("L1")
+
+        df2 = DataFrame({("A", "C"): [0, 0, 0], ("A", "D"): [0, 0, 0]})
+        df2.columns = df2.columns.set_names(["L1", "L2"])
+
+        result = df1.add(df2, level=level)
+        expected = DataFrame({("A", "C"): [0, 1, 2], ("A", "D"): [0, 1, 2]})
+        expected.columns = expected.columns.set_names(["L1", "L2"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_multiindex_operations(self):
+        # GH 43321
+        df = DataFrame(
+            {2010: [1, 2, 3], 2020: [3, 4, 5]},
+            index=MultiIndex.from_product(
+                [["a"], ["b"], [0, 1, 2]], names=["scen", "mod", "id"]
+            ),
+        )
+
+        series = Series(
+            [0.4],
+            index=MultiIndex.from_product([["b"], ["a"]], names=["mod", "scen"]),
+        )
+
+        expected = DataFrame(
+            {2010: [1.4, 2.4, 3.4], 2020: [3.4, 4.4, 5.4]},
+            index=MultiIndex.from_product(
+                [["a"], ["b"], [0, 1, 2]], names=["scen", "mod", "id"]
+            ),
+        )
+        result = df.add(series, axis=0)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_multiindex_operations_series_index_to_frame_index(self):
+        # GH 43321
+        df = DataFrame(
+            {2010: [1], 2020: [3]},
+            index=MultiIndex.from_product([["a"], ["b"]], names=["scen", "mod"]),
+        )
+
+        series = Series(
+            [10.0, 20.0, 30.0],
+            index=MultiIndex.from_product(
+                [["a"], ["b"], [0, 1, 2]], names=["scen", "mod", "id"]
+            ),
+        )
+
+        expected = DataFrame(
+            {2010: [11.0, 21, 31.0], 2020: [13.0, 23.0, 33.0]},
+            index=MultiIndex.from_product(
+                [["a"], ["b"], [0, 1, 2]], names=["scen", "mod", "id"]
+            ),
+        )
+        result = df.add(series, axis=0)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_multiindex_operations_no_align(self):
+        df = DataFrame(
+            {2010: [1, 2, 3], 2020: [3, 4, 5]},
+            index=MultiIndex.from_product(
+                [["a"], ["b"], [0, 1, 2]], names=["scen", "mod", "id"]
+            ),
+        )
+
+        series = Series(
+            [0.4],
+            index=MultiIndex.from_product([["c"], ["a"]], names=["mod", "scen"]),
+        )
+
+        expected = DataFrame(
+            {2010: np.nan, 2020: np.nan},
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", "b", 0),
+                    ("a", "b", 1),
+                    ("a", "b", 2),
+                    ("a", "c", np.nan),
+                ],
+                names=["scen", "mod", "id"],
+            ),
+        )
+        result = df.add(series, axis=0)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_multiindex_operations_part_align(self):
+        df = DataFrame(
+            {2010: [1, 2, 3], 2020: [3, 4, 5]},
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", "b", 0),
+                    ("a", "b", 1),
+                    ("a", "c", 2),
+                ],
+                names=["scen", "mod", "id"],
+            ),
+        )
+
+        series = Series(
+            [0.4],
+            index=MultiIndex.from_product([["b"], ["a"]], names=["mod", "scen"]),
+        )
+
+        expected = DataFrame(
+            {2010: [1.4, 2.4, np.nan], 2020: [3.4, 4.4, np.nan]},
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", "b", 0),
+                    ("a", "b", 1),
+                    ("a", "c", 2),
+                ],
+                names=["scen", "mod", "id"],
+            ),
+        )
+        result = df.add(series, axis=0)
+
+        tm.assert_frame_equal(result, expected)
+
+
+class TestFrameArithmetic:
+    def test_td64_op_nat_casting(self):
+        # Make sure we don't accidentally treat timedelta64(NaT) as datetime64
+        #  when calling dispatch_to_series in DataFrame arithmetic
+        ser = Series(["NaT", "NaT"], dtype="timedelta64[ns]")
+        df = DataFrame([[1, 2], [3, 4]])
+
+        result = df * ser
+        expected = DataFrame({0: ser, 1: ser})
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_add_2d_array_rowlike_broadcasts(self):
+        # GH#23000
+        arr = np.arange(6).reshape(3, 2)
+        df = DataFrame(arr, columns=[True, False], index=["A", "B", "C"])
+
+        rowlike = arr[[1], :]  # shape --> (1, ncols)
+        assert rowlike.shape == (1, df.shape[1])
+
+        expected = DataFrame(
+            [[2, 4], [4, 6], [6, 8]],
+            columns=df.columns,
+            index=df.index,
+            # specify dtype explicitly to avoid failing
+            # on 32bit builds
+            dtype=arr.dtype,
+        )
+        result = df + rowlike
+        tm.assert_frame_equal(result, expected)
+        result = rowlike + df
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_add_2d_array_collike_broadcasts(self):
+        # GH#23000
+        arr = np.arange(6).reshape(3, 2)
+        df = DataFrame(arr, columns=[True, False], index=["A", "B", "C"])
+
+        collike = arr[:, [1]]  # shape --> (nrows, 1)
+        assert collike.shape == (df.shape[0], 1)
+
+        expected = DataFrame(
+            [[1, 2], [5, 6], [9, 10]],
+            columns=df.columns,
+            index=df.index,
+            # specify dtype explicitly to avoid failing
+            # on 32bit builds
+            dtype=arr.dtype,
+        )
+        result = df + collike
+        tm.assert_frame_equal(result, expected)
+        result = collike + df
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_arith_2d_array_rowlike_broadcasts(
+        self, request, all_arithmetic_operators
+    ):
+        # GH#23000
+        opname = all_arithmetic_operators
+        arr = np.arange(6).reshape(3, 2)
+        df = DataFrame(arr, columns=[True, False], index=["A", "B", "C"])
+
+        rowlike = arr[[1], :]  # shape --> (1, ncols)
+        assert rowlike.shape == (1, df.shape[1])
+
+        exvals = [
+            getattr(df.loc["A"], opname)(rowlike.squeeze()),
+            getattr(df.loc["B"], opname)(rowlike.squeeze()),
+            getattr(df.loc["C"], opname)(rowlike.squeeze()),
+        ]
+
+        expected = DataFrame(exvals, columns=df.columns, index=df.index)
+
+        result = getattr(df, opname)(rowlike)
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_arith_2d_array_collike_broadcasts(
+        self, request, all_arithmetic_operators
+    ):
+        # GH#23000
+        opname = all_arithmetic_operators
+        arr = np.arange(6).reshape(3, 2)
+        df = DataFrame(arr, columns=[True, False], index=["A", "B", "C"])
+
+        collike = arr[:, [1]]  # shape --> (nrows, 1)
+        assert collike.shape == (df.shape[0], 1)
+
+        exvals = {
+            True: getattr(df[True], opname)(collike.squeeze()),
+            False: getattr(df[False], opname)(collike.squeeze()),
+        }
+
+        dtype = None
+        if opname in ["__rmod__", "__rfloordiv__"]:
+            # Series ops may return mixed int/float dtypes in cases where
+            #   DataFrame op will return all-float.  So we upcast `expected`
+            dtype = np.common_type(*(x.values for x in exvals.values()))
+
+        expected = DataFrame(exvals, columns=df.columns, index=df.index, dtype=dtype)
+
+        result = getattr(df, opname)(collike)
+        tm.assert_frame_equal(result, expected)
+
+    def test_df_bool_mul_int(self):
+        # GH 22047, GH 22163 multiplication by 1 should result in int dtype,
+        # not object dtype
+        df = DataFrame([[False, True], [False, False]])
+        result = df * 1
+
+        # On appveyor this comes back as np.int32 instead of np.int64,
+        # so we check dtype.kind instead of just dtype
+        kinds = result.dtypes.apply(lambda x: x.kind)
+        assert (kinds == "i").all()
+
+        result = 1 * df
+        kinds = result.dtypes.apply(lambda x: x.kind)
+        assert (kinds == "i").all()
+
+    def test_arith_mixed(self):
+        left = DataFrame({"A": ["a", "b", "c"], "B": [1, 2, 3]})
+
+        result = left + left
+        expected = DataFrame({"A": ["aa", "bb", "cc"], "B": [2, 4, 6]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("col", ["A", "B"])
+    def test_arith_getitem_commute(self, all_arithmetic_functions, col):
+        df = DataFrame({"A": [1.1, 3.3], "B": [2.5, -3.9]})
+        result = all_arithmetic_functions(df, 1)[col]
+        expected = all_arithmetic_functions(df[col], 1)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values", [[1, 2], (1, 2), np.array([1, 2]), range(1, 3), deque([1, 2])]
+    )
+    def test_arith_alignment_non_pandas_object(self, values):
+        # GH#17901
+        df = DataFrame({"A": [1, 1], "B": [1, 1]})
+        expected = DataFrame({"A": [2, 2], "B": [3, 3]})
+        result = df + values
+        tm.assert_frame_equal(result, expected)
+
+    def test_arith_non_pandas_object(self):
+        df = DataFrame(
+            np.arange(1, 10, dtype="f8").reshape(3, 3),
+            columns=["one", "two", "three"],
+            index=["a", "b", "c"],
+        )
+
+        val1 = df.xs("a").values
+        added = DataFrame(df.values + val1, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(df + val1, added)
+
+        added = DataFrame((df.values.T + val1).T, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(df.add(val1, axis=0), added)
+
+        val2 = list(df["two"])
+
+        added = DataFrame(df.values + val2, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(df + val2, added)
+
+        added = DataFrame((df.values.T + val2).T, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(df.add(val2, axis="index"), added)
+
+        val3 = np.random.default_rng(2).random(df.shape)
+        added = DataFrame(df.values + val3, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(df.add(val3), added)
+
+    def test_operations_with_interval_categories_index(self, all_arithmetic_operators):
+        # GH#27415
+        op = all_arithmetic_operators
+        ind = pd.CategoricalIndex(pd.interval_range(start=0.0, end=2.0))
+        data = [1, 2]
+        df = DataFrame([data], columns=ind)
+        num = 10
+        result = getattr(df, op)(num)
+        expected = DataFrame([[getattr(n, op)(num) for n in data]], columns=ind)
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_with_frame_reindex(self):
+        # GH#31623
+        df = DataFrame(
+            {
+                "foo": [pd.Timestamp("2019"), pd.Timestamp("2020")],
+                "bar": [pd.Timestamp("2018"), pd.Timestamp("2021")],
+            },
+            columns=["foo", "bar"],
+            dtype="M8[ns]",
+        )
+        df2 = df[["foo"]]
+
+        result = df - df2
+
+        expected = DataFrame(
+            {"foo": [pd.Timedelta(0), pd.Timedelta(0)], "bar": [np.nan, np.nan]},
+            columns=["bar", "foo"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "value, dtype",
+        [
+            (1, "i8"),
+            (1.0, "f8"),
+            (2**63, "f8"),
+            (1j, "complex128"),
+            (2**63, "complex128"),
+            (True, "bool"),
+            (np.timedelta64(20, "ns"), "<m8[ns]"),
+            (np.datetime64(20, "ns"), "<M8[ns]"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "op",
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.truediv,
+            operator.mod,
+            operator.pow,
+        ],
+        ids=lambda x: x.__name__,
+    )
+    def test_binop_other(self, op, value, dtype, switch_numexpr_min_elements):
+        skip = {
+            (operator.truediv, "bool"),
+            (operator.pow, "bool"),
+            (operator.add, "bool"),
+            (operator.mul, "bool"),
+        }
+
+        elem = DummyElement(value, dtype)
+        df = DataFrame({"A": [elem.value, elem.value]}, dtype=elem.dtype)
+
+        invalid = {
+            (operator.pow, "<M8[ns]"),
+            (operator.mod, "<M8[ns]"),
+            (operator.truediv, "<M8[ns]"),
+            (operator.mul, "<M8[ns]"),
+            (operator.add, "<M8[ns]"),
+            (operator.pow, "<m8[ns]"),
+            (operator.mul, "<m8[ns]"),
+            (operator.sub, "bool"),
+            (operator.mod, "complex128"),
+        }
+
+        if (op, dtype) in invalid:
+            warn = None
+            if (dtype == "<M8[ns]" and op == operator.add) or (
+                dtype == "<m8[ns]" and op == operator.mul
+            ):
+                msg = None
+            elif dtype == "complex128":
+                msg = "ufunc 'remainder' not supported for the input types"
+            elif op is operator.sub:
+                msg = "numpy boolean subtract, the `-` operator, is "
+                if (
+                    dtype == "bool"
+                    and expr.USE_NUMEXPR
+                    and switch_numexpr_min_elements == 0
+                ):
+                    warn = UserWarning
+            else:
+                msg = (
+                    f"cannot perform __{op.__name__}__ with this "
+                    "index type: (DatetimeArray|TimedeltaArray)"
+                )
+
+            with pytest.raises(TypeError, match=msg):
+                with tm.assert_produces_warning(warn, match="evaluating in Python"):
+                    op(df, elem.value)
+
+        elif (op, dtype) in skip:
+            if op in [operator.add, operator.mul]:
+                if expr.USE_NUMEXPR and switch_numexpr_min_elements == 0:
+                    warn = UserWarning
+                else:
+                    warn = None
+                with tm.assert_produces_warning(warn, match="evaluating in Python"):
+                    op(df, elem.value)
+
+            else:
+                msg = "operator '.*' not implemented for .* dtypes"
+                with pytest.raises(NotImplementedError, match=msg):
+                    op(df, elem.value)
+
+        else:
+            with tm.assert_produces_warning(None):
+                result = op(df, elem.value).dtypes
+                expected = op(df, value).dtypes
+            tm.assert_series_equal(result, expected)
+
+    def test_arithmetic_midx_cols_different_dtypes(self):
+        # GH#49769
+        midx = MultiIndex.from_arrays([Series([1, 2]), Series([3, 4])])
+        midx2 = MultiIndex.from_arrays([Series([1, 2], dtype="Int8"), Series([3, 4])])
+        left = DataFrame([[1, 2], [3, 4]], columns=midx)
+        right = DataFrame([[1, 2], [3, 4]], columns=midx2)
+        result = left - right
+        expected = DataFrame([[0, 0], [0, 0]], columns=midx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_arithmetic_midx_cols_different_dtypes_different_order(self):
+        # GH#49769
+        midx = MultiIndex.from_arrays([Series([1, 2]), Series([3, 4])])
+        midx2 = MultiIndex.from_arrays([Series([2, 1], dtype="Int8"), Series([4, 3])])
+        left = DataFrame([[1, 2], [3, 4]], columns=midx)
+        right = DataFrame([[1, 2], [3, 4]], columns=midx2)
+        result = left - right
+        expected = DataFrame([[-1, 1], [-1, 1]], columns=midx)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_frame_with_zero_len_series_corner_cases():
+    # GH#28600
+    # easy all-float case
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal(6).reshape(3, 2), columns=["A", "B"]
+    )
+    ser = Series(dtype=np.float64)
+
+    result = df + ser
+    expected = DataFrame(df.values * np.nan, columns=df.columns)
+    tm.assert_frame_equal(result, expected)
+
+    with pytest.raises(ValueError, match="not aligned"):
+        # Automatic alignment for comparisons deprecated GH#36795, enforced 2.0
+        df == ser
+
+    # non-float case should not raise TypeError on comparison
+    df2 = DataFrame(df.values.view("M8[ns]"), columns=df.columns)
+    with pytest.raises(ValueError, match="not aligned"):
+        # Automatic alignment for comparisons deprecated
+        df2 == ser
+
+
+def test_zero_len_frame_with_series_corner_cases():
+    # GH#28600
+    df = DataFrame(columns=["A", "B"], dtype=np.float64)
+    ser = Series([1, 2], index=["A", "B"])
+
+    result = df + ser
+    expected = df
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_single_columns_object_sum_axis_1():
+    # GH 13758
+    data = {
+        "One": Series(["A", 1.2, np.nan]),
+    }
+    df = DataFrame(data)
+    result = df.sum(axis=1)
+    expected = Series(["A", 1.2, 0])
+    tm.assert_series_equal(result, expected)
+
+
+# -------------------------------------------------------------------
+# Unsorted
+#  These arithmetic tests were previously in other files, eventually
+#  should be parametrized and put into tests.arithmetic
+
+
+class TestFrameArithmeticUnsorted:
+    def test_frame_add_tz_mismatch_converts_to_utc(self):
+        rng = pd.date_range("1/1/2011", periods=10, freq="h", tz="US/Eastern")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(len(rng)), index=rng, columns=["a"]
+        )
+
+        df_moscow = df.tz_convert("Europe/Moscow")
+        result = df + df_moscow
+        assert result.index.tz is timezone.utc
+
+        result = df_moscow + df
+        assert result.index.tz is timezone.utc
+
+    def test_align_frame(self):
+        rng = pd.period_range("1/1/2000", "1/1/2010", freq="Y")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 3)), index=rng
+        )
+
+        result = ts + ts[::2]
+        expected = ts + ts
+        expected.iloc[1::2] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        half = ts[::2]
+        result = ts + half.take(np.random.default_rng(2).permutation(len(half)))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "op", [operator.add, operator.sub, operator.mul, operator.truediv]
+    )
+    def test_operators_none_as_na(self, op):
+        df = DataFrame(
+            {"col1": [2, 5.0, 123, None], "col2": [1, 2, 3, 4]}, dtype=object
+        )
+
+        # since filling converts dtypes from object, changed expected to be
+        # object
+
+        filled = df.fillna(np.nan)
+        result = op(df, 3)
+        expected = op(filled, 3).astype(object)
+        expected[pd.isna(expected)] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = op(df, df)
+        expected = op(filled, filled).astype(object)
+        expected[pd.isna(expected)] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = op(df, df.fillna(7))
+        tm.assert_frame_equal(result, expected)
+
+        result = op(df.fillna(7), df)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("op,res", [("__eq__", False), ("__ne__", True)])
+    # TODO: not sure what's correct here.
+    @pytest.mark.filterwarnings("ignore:elementwise:FutureWarning")
+    def test_logical_typeerror_with_non_valid(self, op, res, float_frame):
+        # we are comparing floats vs a string
+        result = getattr(float_frame, op)("foo")
+        assert bool(result.all().all()) is res
+
+    @pytest.mark.parametrize("op", ["add", "sub", "mul", "div", "truediv"])
+    def test_binary_ops_align(self, op):
+        # test aligning binary ops
+
+        # GH 6681
+        index = MultiIndex.from_product(
+            [list("abc"), ["one", "two", "three"], [1, 2, 3]],
+            names=["first", "second", "third"],
+        )
+
+        df = DataFrame(
+            np.arange(27 * 3).reshape(27, 3),
+            index=index,
+            columns=["value1", "value2", "value3"],
+        ).sort_index()
+
+        idx = pd.IndexSlice
+        opa = getattr(operator, op, None)
+        if opa is None:
+            return
+
+        x = Series([1.0, 10.0, 100.0], [1, 2, 3])
+        result = getattr(df, op)(x, level="third", axis=0)
+
+        expected = pd.concat(
+            [opa(df.loc[idx[:, :, i], :], v) for i, v in x.items()]
+        ).sort_index()
+        tm.assert_frame_equal(result, expected)
+
+        x = Series([1.0, 10.0], ["two", "three"])
+        result = getattr(df, op)(x, level="second", axis=0)
+
+        expected = (
+            pd.concat([opa(df.loc[idx[:, i], :], v) for i, v in x.items()])
+            .reindex_like(df)
+            .sort_index()
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_binary_ops_align_series_dataframe(self):
+        # GH9463 (alignment level of dataframe with series)
+
+        midx = MultiIndex.from_product([["A", "B"], ["a", "b"]])
+        df = DataFrame(np.ones((2, 4), dtype="int64"), columns=midx)
+        s = Series({"a": 1, "b": 2})
+
+        df2 = df.copy()
+        df2.columns.names = ["lvl0", "lvl1"]
+        s2 = s.copy()
+        s2.index.name = "lvl1"
+
+        # different cases of integer/string level names:
+        res1 = df.mul(s, axis=1, level=1)
+        res2 = df.mul(s2, axis=1, level=1)
+        res3 = df2.mul(s, axis=1, level=1)
+        res4 = df2.mul(s2, axis=1, level=1)
+        res5 = df2.mul(s, axis=1, level="lvl1")
+        res6 = df2.mul(s2, axis=1, level="lvl1")
+
+        exp = DataFrame(
+            np.array([[1, 2, 1, 2], [1, 2, 1, 2]], dtype="int64"), columns=midx
+        )
+
+        for res in [res1, res2]:
+            tm.assert_frame_equal(res, exp)
+
+        exp.columns.names = ["lvl0", "lvl1"]
+        for res in [res3, res4, res5, res6]:
+            tm.assert_frame_equal(res, exp)
+
+    def test_add_with_dti_mismatched_tzs(self):
+        base = pd.DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"], tz="UTC")
+        idx1 = base.tz_convert("Asia/Tokyo")[:2]
+        idx2 = base.tz_convert("US/Eastern")[1:]
+
+        df1 = DataFrame({"A": [1, 2]}, index=idx1)
+        df2 = DataFrame({"A": [1, 1]}, index=idx2)
+        exp = DataFrame({"A": [np.nan, 3, np.nan]}, index=base)
+        tm.assert_frame_equal(df1 + df2, exp)
+
+    def test_combineFrame(self, float_frame, mixed_float_frame, mixed_int_frame):
+        frame_copy = float_frame.reindex(float_frame.index[::2])
+
+        del frame_copy["D"]
+        # adding NAs to first 5 values of column "C"
+        frame_copy.loc[: frame_copy.index[4], "C"] = np.nan
+
+        added = float_frame + frame_copy
+
+        indexer = added["A"].dropna().index
+        exp = (float_frame["A"] * 2).copy()
+
+        tm.assert_series_equal(added["A"].dropna(), exp.loc[indexer])
+
+        exp.loc[~exp.index.isin(indexer)] = np.nan
+        tm.assert_series_equal(added["A"], exp.loc[added["A"].index])
+
+        assert np.isnan(added["C"].reindex(frame_copy.index)[:5]).all()
+
+        # assert(False)
+
+        assert np.isnan(added["D"]).all()
+
+        self_added = float_frame + float_frame
+        tm.assert_index_equal(self_added.index, float_frame.index)
+
+        added_rev = frame_copy + float_frame
+        assert np.isnan(added["D"]).all()
+        assert np.isnan(added_rev["D"]).all()
+
+        # corner cases
+
+        # empty
+        plus_empty = float_frame + DataFrame()
+        assert np.isnan(plus_empty.values).all()
+
+        empty_plus = DataFrame() + float_frame
+        assert np.isnan(empty_plus.values).all()
+
+        empty_empty = DataFrame() + DataFrame()
+        assert empty_empty.empty
+
+        # out of order
+        reverse = float_frame.reindex(columns=float_frame.columns[::-1])
+
+        tm.assert_frame_equal(reverse + float_frame, float_frame * 2)
+
+        # mix vs float64, upcast
+        added = float_frame + mixed_float_frame
+        _check_mixed_float(added, dtype="float64")
+        added = mixed_float_frame + float_frame
+        _check_mixed_float(added, dtype="float64")
+
+        # mix vs mix
+        added = mixed_float_frame + mixed_float_frame
+        _check_mixed_float(added, dtype={"C": None})
+
+        # with int
+        added = float_frame + mixed_int_frame
+        _check_mixed_float(added, dtype="float64")
+
+    def test_combine_series(self, float_frame, mixed_float_frame, mixed_int_frame):
+        # Series
+        series = float_frame.xs(float_frame.index[0])
+
+        added = float_frame + series
+
+        for key, s in added.items():
+            tm.assert_series_equal(s, float_frame[key] + series[key])
+
+        larger_series = series.to_dict()
+        larger_series["E"] = 1
+        larger_series = Series(larger_series)
+        larger_added = float_frame + larger_series
+
+        for key, s in float_frame.items():
+            tm.assert_series_equal(larger_added[key], s + series[key])
+        assert "E" in larger_added
+        assert np.isnan(larger_added["E"]).all()
+
+        # no upcast needed
+        added = mixed_float_frame + series
+        assert np.all(added.dtypes == series.dtype)
+
+        # vs mix (upcast) as needed
+        added = mixed_float_frame + series.astype("float32")
+        _check_mixed_float(added, dtype={"C": None})
+        added = mixed_float_frame + series.astype("float16")
+        _check_mixed_float(added, dtype={"C": None})
+
+        # these used to raise with numexpr as we are adding an int64 to an
+        #  uint64....weird vs int
+        added = mixed_int_frame + (100 * series).astype("int64")
+        _check_mixed_int(
+            added, dtype={"A": "int64", "B": "float64", "C": "int64", "D": "int64"}
+        )
+        added = mixed_int_frame + (100 * series).astype("int32")
+        _check_mixed_int(
+            added, dtype={"A": "int32", "B": "float64", "C": "int32", "D": "int64"}
+        )
+
+    def test_combine_timeseries(self, datetime_frame):
+        # TimeSeries
+        ts = datetime_frame["A"]
+
+        # 10890
+        # we no longer allow auto timeseries broadcasting
+        # and require explicit broadcasting
+        added = datetime_frame.add(ts, axis="index")
+
+        for key, col in datetime_frame.items():
+            result = col + ts
+            tm.assert_series_equal(added[key], result, check_names=False)
+            assert added[key].name == key
+            if col.name == ts.name:
+                assert result.name == "A"
+            else:
+                assert result.name is None
+
+        smaller_frame = datetime_frame[:-5]
+        smaller_added = smaller_frame.add(ts, axis="index")
+
+        tm.assert_index_equal(smaller_added.index, datetime_frame.index)
+
+        smaller_ts = ts[:-5]
+        smaller_added2 = datetime_frame.add(smaller_ts, axis="index")
+        tm.assert_frame_equal(smaller_added, smaller_added2)
+
+        # length 0, result is all-nan
+        result = datetime_frame.add(ts[:0], axis="index")
+        expected = DataFrame(
+            np.nan, index=datetime_frame.index, columns=datetime_frame.columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Frame is all-nan
+        result = datetime_frame[:0].add(ts, axis="index")
+        expected = DataFrame(
+            np.nan, index=datetime_frame.index, columns=datetime_frame.columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # empty but with non-empty index
+        frame = datetime_frame[:1].reindex(columns=[])
+        result = frame.mul(ts, axis="index")
+        assert len(result) == len(ts)
+
+    def test_combineFunc(self, float_frame, mixed_float_frame):
+        result = float_frame * 2
+        tm.assert_numpy_array_equal(result.values, float_frame.values * 2)
+
+        # vs mix
+        result = mixed_float_frame * 2
+        for c, s in result.items():
+            tm.assert_numpy_array_equal(s.values, mixed_float_frame[c].values * 2)
+        _check_mixed_float(result, dtype={"C": None})
+
+        result = DataFrame() * 2
+        assert result.index.equals(DataFrame().index)
+        assert len(result.columns) == 0
+
+    @pytest.mark.parametrize(
+        "func",
+        [operator.eq, operator.ne, operator.lt, operator.gt, operator.ge, operator.le],
+    )
+    def test_comparisons(self, simple_frame, float_frame, func):
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=pd.date_range("2000-01-01", periods=30, freq="B"),
+        )
+        df2 = df1.copy()
+
+        row = simple_frame.xs("a")
+        ndim_5 = np.ones(df1.shape + (1, 1, 1))
+
+        result = func(df1, df2)
+        tm.assert_numpy_array_equal(result.values, func(df1.values, df2.values))
+
+        msg = (
+            "Unable to coerce to Series/DataFrame, "
+            "dimension must be <= 2: (30, 4, 1, 1, 1)"
+        )
+        with pytest.raises(ValueError, match=re.escape(msg)):
+            func(df1, ndim_5)
+
+        result2 = func(simple_frame, row)
+        tm.assert_numpy_array_equal(
+            result2.values, func(simple_frame.values, row.values)
+        )
+
+        result3 = func(float_frame, 0)
+        tm.assert_numpy_array_equal(result3.values, func(float_frame.values, 0))
+
+        msg = (
+            r"Can only compare identically-labeled \(both index and columns\) "
+            "DataFrame objects"
+        )
+        with pytest.raises(ValueError, match=msg):
+            func(simple_frame, simple_frame[:2])
+
+    def test_strings_to_numbers_comparisons_raises(self, compare_operators_no_eq_ne):
+        # GH 11565
+        df = DataFrame(
+            {x: {"x": "foo", "y": "bar", "z": "baz"} for x in ["a", "b", "c"]}
+        )
+
+        f = getattr(operator, compare_operators_no_eq_ne)
+        msg = "|".join(
+            [
+                "'[<>]=?' not supported between instances of 'str' and 'int'",
+                "Invalid comparison between dtype=str and int",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            f(df, 0)
+
+    def test_comparison_protected_from_errstate(self):
+        missing_df = DataFrame(
+            np.ones((10, 4), dtype=np.float64),
+            columns=Index(list("ABCD"), dtype=object),
+        )
+        missing_df.loc[missing_df.index[0], "A"] = np.nan
+        with np.errstate(invalid="ignore"):
+            expected = missing_df.values < 0
+        with np.errstate(invalid="raise"):
+            result = (missing_df < 0).values
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_boolean_comparison(self):
+        # GH 4576
+        # boolean comparisons with a tuple/list give unexpected results
+        df = DataFrame(np.arange(6).reshape((3, 2)))
+        b = np.array([2, 2])
+        b_r = np.atleast_2d([2, 2])
+        b_c = b_r.T
+        lst = [2, 2, 2]
+        tup = tuple(lst)
+
+        # gt
+        expected = DataFrame([[False, False], [False, True], [True, True]])
+        result = df > b
+        tm.assert_frame_equal(result, expected)
+
+        result = df.values > b
+        tm.assert_numpy_array_equal(result, expected.values)
+
+        msg1d = "Unable to coerce to Series, length must be 2: given 3"
+        msg2d = "Unable to coerce to DataFrame, shape must be"
+        msg2db = "operands could not be broadcast together with shapes"
+        with pytest.raises(ValueError, match=msg1d):
+            # wrong shape
+            df > lst
+
+        with pytest.raises(ValueError, match=msg1d):
+            # wrong shape
+            df > tup
+
+        # broadcasts like ndarray (GH#23000)
+        result = df > b_r
+        tm.assert_frame_equal(result, expected)
+
+        result = df.values > b_r
+        tm.assert_numpy_array_equal(result, expected.values)
+
+        with pytest.raises(ValueError, match=msg2d):
+            df > b_c
+
+        with pytest.raises(ValueError, match=msg2db):
+            df.values > b_c
+
+        # ==
+        expected = DataFrame([[False, False], [True, False], [False, False]])
+        result = df == b
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(ValueError, match=msg1d):
+            df == lst
+
+        with pytest.raises(ValueError, match=msg1d):
+            df == tup
+
+        # broadcasts like ndarray (GH#23000)
+        result = df == b_r
+        tm.assert_frame_equal(result, expected)
+
+        result = df.values == b_r
+        tm.assert_numpy_array_equal(result, expected.values)
+
+        with pytest.raises(ValueError, match=msg2d):
+            df == b_c
+
+        assert df.values.shape != b_c.shape
+
+        # with alignment
+        df = DataFrame(
+            np.arange(6).reshape((3, 2)), columns=list("AB"), index=list("abc")
+        )
+        expected.index = df.index
+        expected.columns = df.columns
+
+        with pytest.raises(ValueError, match=msg1d):
+            df == lst
+
+        with pytest.raises(ValueError, match=msg1d):
+            df == tup
+
+    def test_inplace_ops_alignment(self):
+        # inplace ops / ops alignment
+        # GH 8511
+
+        columns = list("abcdefg")
+        X_orig = DataFrame(
+            np.arange(10 * len(columns)).reshape(-1, len(columns)),
+            columns=columns,
+            index=range(10),
+        )
+        Z = 100 * X_orig.iloc[:, 1:-1].copy()
+        block1 = list("bedcf")
+        subs = list("bcdef")
+
+        # add
+        X = X_orig.copy()
+        result1 = (X[block1] + Z).reindex(columns=subs)
+
+        X[block1] += Z
+        result2 = X.reindex(columns=subs)
+
+        X = X_orig.copy()
+        result3 = (X[block1] + Z[block1]).reindex(columns=subs)
+
+        X[block1] += Z[block1]
+        result4 = X.reindex(columns=subs)
+
+        tm.assert_frame_equal(result1, result2)
+        tm.assert_frame_equal(result1, result3)
+        tm.assert_frame_equal(result1, result4)
+
+        # sub
+        X = X_orig.copy()
+        result1 = (X[block1] - Z).reindex(columns=subs)
+
+        X[block1] -= Z
+        result2 = X.reindex(columns=subs)
+
+        X = X_orig.copy()
+        result3 = (X[block1] - Z[block1]).reindex(columns=subs)
+
+        X[block1] -= Z[block1]
+        result4 = X.reindex(columns=subs)
+
+        tm.assert_frame_equal(result1, result2)
+        tm.assert_frame_equal(result1, result3)
+        tm.assert_frame_equal(result1, result4)
+
+    def test_inplace_ops_identity(self):
+        # GH 5104
+        # make sure that we are actually changing the object
+        s_orig = Series([1, 2, 3])
+        df_orig = DataFrame(
+            np.random.default_rng(2).integers(0, 5, size=10).reshape(-1, 5)
+        )
+
+        # no dtype change
+        s = s_orig.copy()
+        s2 = s
+        s += 1
+        tm.assert_series_equal(s, s2)
+        tm.assert_series_equal(s_orig + 1, s)
+        assert s is s2
+        assert s._mgr is s2._mgr
+
+        df = df_orig.copy()
+        df2 = df
+        df += 1
+        tm.assert_frame_equal(df, df2)
+        tm.assert_frame_equal(df_orig + 1, df)
+        assert df is df2
+        assert df._mgr is df2._mgr
+
+        # dtype change
+        s = s_orig.copy()
+        s2 = s
+        s += 1.5
+        tm.assert_series_equal(s, s2)
+        tm.assert_series_equal(s_orig + 1.5, s)
+
+        df = df_orig.copy()
+        df2 = df
+        df += 1.5
+        tm.assert_frame_equal(df, df2)
+        tm.assert_frame_equal(df_orig + 1.5, df)
+        assert df is df2
+        assert df._mgr is df2._mgr
+
+        # mixed dtype
+        arr = np.random.default_rng(2).integers(0, 10, size=5)
+        df_orig = DataFrame({"A": arr.copy(), "B": "foo"})
+        df = df_orig.copy()
+        df2 = df
+        df["A"] += 1
+        expected = DataFrame({"A": arr.copy() + 1, "B": "foo"})
+        tm.assert_frame_equal(df, expected)
+        tm.assert_frame_equal(df2, expected)
+        assert df._mgr is df2._mgr
+
+        df = df_orig.copy()
+        df2 = df
+        df["A"] += 1.5
+        expected = DataFrame({"A": arr.copy() + 1.5, "B": "foo"})
+        tm.assert_frame_equal(df, expected)
+        tm.assert_frame_equal(df2, expected)
+        assert df._mgr is df2._mgr
+
+    @pytest.mark.parametrize(
+        "op",
+        [
+            "add",
+            "and",
+            pytest.param(
+                "div",
+                marks=pytest.mark.xfail(
+                    raises=AttributeError, reason="__idiv__ not implemented"
+                ),
+            ),
+            "floordiv",
+            "mod",
+            "mul",
+            "or",
+            "pow",
+            "sub",
+            "truediv",
+            "xor",
+        ],
+    )
+    def test_inplace_ops_identity2(self, op):
+        df = DataFrame({"a": [1.0, 2.0, 3.0], "b": [1, 2, 3]})
+
+        operand = 2
+        if op in ("and", "or", "xor"):
+            # cannot use floats for boolean ops
+            df["a"] = [True, False, True]
+
+        df_copy = df.copy()
+        iop = f"__i{op}__"
+        op = f"__{op}__"
+
+        # no id change and value is correct
+        getattr(df, iop)(operand)
+        expected = getattr(df_copy, op)(operand)
+        tm.assert_frame_equal(df, expected)
+        expected = id(df)
+        assert id(df) == expected
+
+    @pytest.mark.parametrize(
+        "val",
+        [
+            [1, 2, 3],
+            (1, 2, 3),
+            np.array([1, 2, 3], dtype=np.int64),
+            range(1, 4),
+        ],
+    )
+    def test_alignment_non_pandas(self, val):
+        index = ["A", "B", "C"]
+        columns = ["X", "Y", "Z"]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=index,
+            columns=columns,
+        )
+
+        align = DataFrame._align_for_op
+
+        expected = DataFrame({"X": val, "Y": val, "Z": val}, index=df.index)
+        tm.assert_frame_equal(align(df, val, axis=0)[1], expected)
+
+        expected = DataFrame(
+            {"X": [1, 1, 1], "Y": [2, 2, 2], "Z": [3, 3, 3]}, index=df.index
+        )
+        tm.assert_frame_equal(align(df, val, axis=1)[1], expected)
+
+    @pytest.mark.parametrize("val", [[1, 2], (1, 2), np.array([1, 2]), range(1, 3)])
+    def test_alignment_non_pandas_length_mismatch(self, val):
+        index = ["A", "B", "C"]
+        columns = ["X", "Y", "Z"]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=index,
+            columns=columns,
+        )
+
+        align = DataFrame._align_for_op
+        # length mismatch
+        msg = "Unable to coerce to Series, length must be 3: given 2"
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=0)
+
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=1)
+
+    def test_alignment_non_pandas_index_columns(self):
+        index = ["A", "B", "C"]
+        columns = ["X", "Y", "Z"]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=index,
+            columns=columns,
+        )
+
+        align = DataFrame._align_for_op
+        val = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
+        tm.assert_frame_equal(
+            align(df, val, axis=0)[1],
+            DataFrame(val, index=df.index, columns=df.columns),
+        )
+        tm.assert_frame_equal(
+            align(df, val, axis=1)[1],
+            DataFrame(val, index=df.index, columns=df.columns),
+        )
+
+        # shape mismatch
+        msg = "Unable to coerce to DataFrame, shape must be"
+        val = np.array([[1, 2, 3], [4, 5, 6]])
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=0)
+
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=1)
+
+        val = np.zeros((3, 3, 3))
+        msg = re.escape(
+            "Unable to coerce to Series/DataFrame, dimension must be <= 2: (3, 3, 3)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=0)
+        with pytest.raises(ValueError, match=msg):
+            align(df, val, axis=1)
+
+    def test_no_warning(self, all_arithmetic_operators):
+        df = DataFrame({"A": [0.0, 0.0], "B": [0.0, None]})
+        b = df["B"]
+        with tm.assert_produces_warning(None):
+            getattr(df, all_arithmetic_operators)(b)
+
+    def test_dunder_methods_binary(self, all_arithmetic_operators):
+        # GH#??? frame.__foo__ should only accept one argument
+        df = DataFrame({"A": [0.0, 0.0], "B": [0.0, None]})
+        b = df["B"]
+        with pytest.raises(TypeError, match="takes 2 positional arguments"):
+            getattr(df, all_arithmetic_operators)(b, 0)
+
+    def test_align_int_fill_bug(self):
+        # GH#910
+        X = np.arange(10 * 10, dtype="float64").reshape(10, 10)
+        Y = np.ones((10, 1), dtype=int)
+
+        df1 = DataFrame(X)
+        df1["0.X"] = Y.squeeze()
+
+        df2 = df1.astype(float)
+
+        result = df1 - df1.mean()
+        expected = df2 - df2.mean()
+        tm.assert_frame_equal(result, expected)
+
+
+def test_pow_with_realignment():
+    # GH#32685 pow has special semantics for operating with null values
+    left = DataFrame({"A": [0, 1, 2]})
+    right = DataFrame(index=[0, 1, 2])
+
+    result = left**right
+    expected = DataFrame({"A": [np.nan, 1.0, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dataframe_series_extension_dtypes():
+    # https://github.com/pandas-dev/pandas/issues/34311
+    df = DataFrame(
+        np.random.default_rng(2).integers(0, 100, (10, 3)), columns=["a", "b", "c"]
+    )
+    ser = Series([1, 2, 3], index=["a", "b", "c"])
+
+    expected = df.to_numpy("int64") + ser.to_numpy("int64").reshape(-1, 3)
+    expected = DataFrame(expected, columns=df.columns, dtype="Int64")
+
+    df_ea = df.astype("Int64")
+    result = df_ea + ser
+    tm.assert_frame_equal(result, expected)
+    result = df_ea + ser.astype("Int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dataframe_blockwise_slicelike():
+    # GH#34367
+    arr = np.random.default_rng(2).integers(0, 1000, (100, 10))
+    df1 = DataFrame(arr)
+    # Explicit cast to float to avoid implicit cast when setting nan
+    df2 = df1.copy().astype({1: "float", 3: "float", 7: "float"})
+    df2.iloc[0, [1, 3, 7]] = np.nan
+
+    # Explicit cast to float to avoid implicit cast when setting nan
+    df3 = df1.copy().astype({5: "float"})
+    df3.iloc[0, [5]] = np.nan
+
+    # Explicit cast to float to avoid implicit cast when setting nan
+    df4 = df1.copy().astype({2: "float", 3: "float", 4: "float"})
+    df4.iloc[0, np.arange(2, 5)] = np.nan
+    # Explicit cast to float to avoid implicit cast when setting nan
+    df5 = df1.copy().astype({4: "float", 5: "float", 6: "float"})
+    df5.iloc[0, np.arange(4, 7)] = np.nan
+
+    for left, right in [(df1, df2), (df2, df3), (df4, df5)]:
+        res = left + right
+
+        expected = DataFrame({i: left[i] + right[i] for i in left.columns})
+        tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "df, col_dtype",
+    [
+        (DataFrame([[1.0, 2.0], [4.0, 5.0]], columns=list("ab")), "float64"),
+        (
+            DataFrame([[1.0, "b"], [4.0, "b"]], columns=list("ab")).astype(
+                {"b": object}
+            ),
+            "object",
+        ),
+    ],
+)
+def test_dataframe_operation_with_non_numeric_types(df, col_dtype):
+    # GH #22663
+    expected = DataFrame([[0.0, np.nan], [3.0, np.nan]], columns=list("ab"))
+    expected = expected.astype({"b": col_dtype})
+    result = df + Series([-1.0], index=list("a"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_arith_reindex_with_duplicates():
+    # https://github.com/pandas-dev/pandas/issues/35194
+    df1 = DataFrame(data=[[0]], columns=["second"])
+    df2 = DataFrame(data=[[0, 0, 0]], columns=["first", "second", "second"])
+    result = df1 + df2
+    expected = DataFrame([[np.nan, 0, 0]], columns=["first", "second", "second"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("to_add", [[Series([1, 1])], [Series([1, 1]), Series([1, 1])]])
+def test_arith_list_of_arraylike_raise(to_add):
+    # GH 36702. Raise when trying to add list of array-like to DataFrame
+    df = DataFrame({"x": [1, 2], "y": [1, 2]})
+
+    msg = f"Unable to coerce list of {type(to_add[0])} to Series/DataFrame"
+    with pytest.raises(ValueError, match=msg):
+        df + to_add
+    with pytest.raises(ValueError, match=msg):
+        to_add + df
+
+
+def test_inplace_arithmetic_series_update():
+    # https://github.com/pandas-dev/pandas/issues/36373
+    df = DataFrame({"A": [1, 2, 3]})
+    df_orig = df.copy()
+    series = df["A"]
+    vals = series._values
+
+    series += 1
+    assert series._values is not vals
+    tm.assert_frame_equal(df, df_orig)
+
+
+def test_arithmetic_multiindex_align():
+    """
+    Regression test for: https://github.com/pandas-dev/pandas/issues/33765
+    """
+    df1 = DataFrame(
+        [[1]],
+        index=["a"],
+        columns=MultiIndex.from_product([[0], [1]], names=["a", "b"]),
+    )
+    df2 = DataFrame([[1]], index=["a"], columns=Index([0], name="a"))
+    expected = DataFrame(
+        [[0]],
+        index=["a"],
+        columns=MultiIndex.from_product([[0], [1]], names=["a", "b"]),
+    )
+    result = df1 - df2
+    tm.assert_frame_equal(result, expected)
+
+
+def test_arithmetic_multiindex_column_align():
+    # GH#60498
+    df1 = DataFrame(
+        data=100,
+        columns=MultiIndex.from_product(
+            [["1A", "1B"], ["2A", "2B"]], names=["Lev1", "Lev2"]
+        ),
+        index=["C1", "C2"],
+    )
+    df2 = DataFrame(
+        data=np.array([[0.1, 0.25], [0.2, 0.45]]),
+        columns=MultiIndex.from_product([["1A", "1B"]], names=["Lev1"]),
+        index=["C1", "C2"],
+    )
+    expected = DataFrame(
+        data=np.array([[10.0, 10.0, 25.0, 25.0], [20.0, 20.0, 45.0, 45.0]]),
+        columns=MultiIndex.from_product(
+            [["1A", "1B"], ["2A", "2B"]], names=["Lev1", "Lev2"]
+        ),
+        index=["C1", "C2"],
+    )
+    result = df1 * df2
+    tm.assert_frame_equal(result, expected)
+
+
+def test_arithmetic_multiindex_column_align_with_fillvalue():
+    # GH#60903
+    df1 = DataFrame(
+        data=[[1.0, 2.0]],
+        columns=MultiIndex.from_tuples([("A", "one"), ("A", "two")]),
+    )
+    df2 = DataFrame(
+        data=[[3.0, 4.0]],
+        columns=MultiIndex.from_tuples([("B", "one"), ("B", "two")]),
+    )
+    expected = DataFrame(
+        data=[[1.0, 2.0, 3.0, 4.0]],
+        columns=MultiIndex.from_tuples(
+            [("A", "one"), ("A", "two"), ("B", "one"), ("B", "two")]
+        ),
+    )
+    result = df1.add(df2, fill_value=0)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_bool_frame_mult_float():
+    # GH 18549
+    df = DataFrame(True, list("ab"), list("cd"))
+    result = df * 1.0
+    expected = DataFrame(np.ones((2, 2)), list("ab"), list("cd"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_sub_nullable_int(any_int_ea_dtype):
+    # GH 32822
+    series1 = Series([1, 2, None], dtype=any_int_ea_dtype)
+    series2 = Series([1, 2, 3], dtype=any_int_ea_dtype)
+    expected = DataFrame([0, 0, None], dtype=any_int_ea_dtype)
+    result = series1.to_frame() - series2.to_frame()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+)
+def test_frame_op_subclass_nonclass_constructor():
+    # GH#43201 subclass._constructor is a function, not the subclass itself
+
+    class SubclassedSeries(Series):
+        @property
+        def _constructor(self):
+            return SubclassedSeries
+
+        @property
+        def _constructor_expanddim(self):
+            return SubclassedDataFrame
+
+    class SubclassedDataFrame(DataFrame):
+        _metadata = ["my_extra_data"]
+
+        def __init__(self, my_extra_data, *args, **kwargs) -> None:
+            self.my_extra_data = my_extra_data
+            super().__init__(*args, **kwargs)
+
+        @property
+        def _constructor(self):
+            return functools.partial(type(self), self.my_extra_data)
+
+        @property
+        def _constructor_sliced(self):
+            return SubclassedSeries
+
+    sdf = SubclassedDataFrame("some_data", {"A": [1, 2, 3], "B": [4, 5, 6]})
+    result = sdf * 2
+    expected = SubclassedDataFrame("some_data", {"A": [2, 4, 6], "B": [8, 10, 12]})
+    tm.assert_frame_equal(result, expected)
+
+    result = sdf + sdf
+    tm.assert_frame_equal(result, expected)
+
+
+def test_enum_column_equality():
+    Cols = Enum("Cols", "col1 col2")
+
+    q1 = DataFrame({Cols.col1: [1, 2, 3]})
+    q2 = DataFrame({Cols.col1: [1, 2, 3]})
+
+    result = q1[Cols.col1] == q2[Cols.col1]
+    expected = Series([True, True, True], name=Cols.col1)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_mixed_col_index_dtype(using_infer_string):
+    # GH 47382
+    df1 = DataFrame(columns=list("abc"), data=1.0, index=[0])
+    df2 = DataFrame(columns=list("abc"), data=0.0, index=[0])
+    df1.columns = df2.columns.astype("string")
+    result = df1 + df2
+    expected = DataFrame(columns=list("abc"), data=1.0, index=[0])
+    if using_infer_string:
+        # df2.columns.dtype will be "str" instead of object,
+        #  so the aligned result will be "string", not object
+        if HAS_PYARROW:
+            dtype = "string[pyarrow]"
+        else:
+            dtype = "string"
+        expected.columns = expected.columns.astype(dtype)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_arrow_interface.py b/pandas/pandas/tests/frame/test_arrow_interface.py
new file mode 100644
index 0000000000000..b36b6b5ffe0cc
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_arrow_interface.py
@@ -0,0 +1,47 @@
+import ctypes
+
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+
+pa = pytest.importorskip("pyarrow")
+
+
+@td.skip_if_no("pyarrow", min_version="14.0")
+def test_dataframe_arrow_interface(using_infer_string):
+    df = pd.DataFrame({"a": [1, 2, 3], "b": ["a", "b", "c"]})
+
+    capsule = df.__arrow_c_stream__()
+    assert (
+        ctypes.pythonapi.PyCapsule_IsValid(
+            ctypes.py_object(capsule), b"arrow_array_stream"
+        )
+        == 1
+    )
+
+    table = pa.table(df)
+    string_type = pa.large_string() if using_infer_string else pa.string()
+    expected = pa.table({"a": [1, 2, 3], "b": pa.array(["a", "b", "c"], string_type)})
+    assert table.equals(expected)
+
+    schema = pa.schema([("a", pa.int8()), ("b", pa.string())])
+    table = pa.table(df, schema=schema)
+    expected = expected.cast(schema)
+    assert table.equals(expected)
+
+
+@td.skip_if_no("pyarrow", min_version="15.0")
+def test_dataframe_to_arrow(using_infer_string):
+    df = pd.DataFrame({"a": [1, 2, 3], "b": ["a", "b", "c"]})
+
+    table = pa.RecordBatchReader.from_stream(df).read_all()
+    string_type = pa.large_string() if using_infer_string else pa.string()
+    expected = pa.table({"a": [1, 2, 3], "b": pa.array(["a", "b", "c"], string_type)})
+    assert table.equals(expected)
+
+    schema = pa.schema([("a", pa.int8()), ("b", pa.string())])
+    table = pa.RecordBatchReader.from_stream(df, schema=schema).read_all()
+    expected = expected.cast(schema)
+    assert table.equals(expected)
diff --git a/pandas/pandas/tests/frame/test_block_internals.py b/pandas/pandas/tests/frame/test_block_internals.py
new file mode 100644
index 0000000000000..6fdbfac8f4e0a
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_block_internals.py
@@ -0,0 +1,410 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+import itertools
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.core.internals.blocks import NumpyBlock
+
+# Segregated collection of methods that require the BlockManager internal data
+# structure
+
+
+class TestDataFrameBlockInternals:
+    def test_setitem_invalidates_datetime_index_freq(self):
+        # GH#24096 altering a datetime64tz column inplace invalidates the
+        #  `freq` attribute on the underlying DatetimeIndex
+
+        dti = date_range("20130101", periods=3, tz="US/Eastern")
+        ts = dti[1]
+
+        df = DataFrame({"B": dti})
+        assert df["B"]._values.freq is None
+
+        df.iloc[1, 0] = pd.NaT
+        assert df["B"]._values.freq is None
+
+        # check that the DatetimeIndex was not altered in place
+        assert dti.freq == "D"
+        assert dti[1] == ts
+
+    def test_cast_internals(self, float_frame):
+        msg = "Passing a BlockManager to DataFrame"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            casted = DataFrame(float_frame._mgr, dtype=int)
+        expected = DataFrame(float_frame._series, dtype=int)
+        tm.assert_frame_equal(casted, expected)
+
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            casted = DataFrame(float_frame._mgr, dtype=np.int32)
+        expected = DataFrame(float_frame._series, dtype=np.int32)
+        tm.assert_frame_equal(casted, expected)
+
+    def test_consolidate(self, float_frame):
+        float_frame["E"] = 7.0
+        consolidated = float_frame._consolidate()
+        assert len(consolidated._mgr.blocks) == 1
+
+        # Ensure copy, do I want this?
+        recons = consolidated._consolidate()
+        assert recons is not consolidated
+        tm.assert_frame_equal(recons, consolidated)
+
+        float_frame["F"] = 8.0
+        assert len(float_frame._mgr.blocks) == 3
+
+        return_value = float_frame._consolidate_inplace()
+        assert return_value is None
+        assert len(float_frame._mgr.blocks) == 1
+
+    def test_consolidate_inplace(self, float_frame):
+        # triggers in-place consolidation
+        for letter in range(ord("A"), ord("Z")):
+            float_frame[chr(letter)] = chr(letter)
+
+    def test_modify_values(self, float_frame):
+        with pytest.raises(ValueError, match="read-only"):
+            float_frame.values[5] = 5
+        assert (float_frame.values[5] != 5).all()
+
+    def test_boolean_set_uncons(self, float_frame):
+        float_frame["E"] = 7.0
+
+        expected = float_frame.values.copy()
+        expected[expected > 1] = 2
+
+        float_frame[float_frame > 1] = 2
+        tm.assert_almost_equal(expected, float_frame.values)
+
+    def test_constructor_with_convert(self):
+        # this is actually mostly a test of lib.maybe_convert_objects
+        # #2845
+        df = DataFrame({"A": [2**63 - 1]})
+        result = df["A"]
+        expected = Series(np.asarray([2**63 - 1], np.int64), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [2**63]})
+        result = df["A"]
+        expected = Series(np.asarray([2**63], np.uint64), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [datetime(2005, 1, 1), True]})
+        result = df["A"]
+        expected = Series(
+            np.asarray([datetime(2005, 1, 1), True], np.object_), name="A"
+        )
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [None, 1]})
+        result = df["A"]
+        expected = Series(np.asarray([np.nan, 1], np.float64), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0, 2]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0, 2], np.float64), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0 + 2.0j, 3]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0 + 2.0j, 3], np.complex128), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0 + 2.0j, 3.0]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0 + 2.0j, 3.0], np.complex128), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0 + 2.0j, True]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0 + 2.0j, True], np.object_), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0, None]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0, np.nan], np.float64), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [1.0 + 2.0j, None]})
+        result = df["A"]
+        expected = Series(np.asarray([1.0 + 2.0j, np.nan], np.complex128), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [2.0, 1, True, None]})
+        result = df["A"]
+        expected = Series(np.asarray([2.0, 1, True, None], np.object_), name="A")
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"A": [2.0, 1, datetime(2006, 1, 1), None]})
+        result = df["A"]
+        expected = Series(
+            np.asarray([2.0, 1, datetime(2006, 1, 1), None], np.object_), name="A"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_construction_with_mixed(self, float_string_frame, using_infer_string):
+        # mixed-type frames
+        float_string_frame["datetime"] = datetime.now()
+        float_string_frame["timedelta"] = timedelta(days=1, seconds=1)
+        assert float_string_frame["datetime"].dtype == "M8[us]"
+        assert float_string_frame["timedelta"].dtype == "m8[us]"
+        result = float_string_frame.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 4
+            + [
+                np.dtype("object")
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan),
+                np.dtype("datetime64[us]"),
+                np.dtype("timedelta64[us]"),
+            ],
+            index=list("ABCD") + ["foo", "datetime", "timedelta"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_construction_with_conversions(self):
+        # convert from a numpy array of non-ns timedelta64; as of 2.0 this does
+        #  *not* convert
+        arr = np.array([1, 2, 3], dtype="timedelta64[s]")
+        df = DataFrame({"A": arr})
+        expected = DataFrame(
+            {"A": pd.timedelta_range("00:00:01", periods=3, freq="s")}, index=range(3)
+        )
+        tm.assert_numpy_array_equal(df["A"].to_numpy(), arr)
+
+        expected = DataFrame(
+            {
+                "dt1": Timestamp("20130101"),
+                "dt2": date_range("20130101", periods=3).astype("M8[s]"),
+                # 'dt3' : date_range('20130101 00:00:01',periods=3,freq='s'),
+                # FIXME: don't leave commented-out
+            },
+            index=range(3),
+        )
+        assert expected.dtypes["dt1"] == "M8[s]"
+        assert expected.dtypes["dt2"] == "M8[s]"
+
+        dt1 = np.datetime64("2013-01-01")
+        dt2 = np.array(
+            ["2013-01-01", "2013-01-02", "2013-01-03"], dtype="datetime64[D]"
+        )
+        df = DataFrame({"dt1": dt1, "dt2": dt2})
+
+        # df['dt3'] = np.array(['2013-01-01 00:00:01','2013-01-01
+        # 00:00:02','2013-01-01 00:00:03'],dtype='datetime64[s]')
+        # FIXME: don't leave commented-out
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_compound_dtypes(self):
+        # GH 5191
+        # compound dtypes should raise not-implementederror
+
+        def f(dtype):
+            data = list(itertools.repeat((datetime(2001, 1, 1), "aa", 20), 9))
+            return DataFrame(data=data, columns=["A", "B", "C"], dtype=dtype)
+
+        msg = "compound dtypes are not implemented in the DataFrame constructor"
+        with pytest.raises(NotImplementedError, match=msg):
+            f([("A", "datetime64[h]"), ("B", "str"), ("C", "int32")])
+
+        # pre-2.0 these used to work (though results may be unexpected)
+        with pytest.raises(TypeError, match="argument must be"):
+            f("int64")
+        with pytest.raises(TypeError, match="argument must be"):
+            f("float64")
+
+        # 10822
+        msg = "^Unknown datetime string format, unable to parse: aa$"
+        with pytest.raises(ValueError, match=msg):
+            f("M8[ns]")
+
+    def test_pickle_float_string_frame(self, float_string_frame):
+        unpickled = tm.round_trip_pickle(float_string_frame)
+        tm.assert_frame_equal(float_string_frame, unpickled)
+
+        # buglet
+        float_string_frame._mgr.ndim
+
+    def test_pickle_empty(self):
+        empty_frame = DataFrame()
+        unpickled = tm.round_trip_pickle(empty_frame)
+        repr(unpickled)
+
+    def test_pickle_empty_tz_frame(self, timezone_frame):
+        unpickled = tm.round_trip_pickle(timezone_frame)
+        tm.assert_frame_equal(timezone_frame, unpickled)
+
+    def test_consolidate_datetime64(self):
+        # numpy vstack bug
+
+        df = DataFrame(
+            {
+                "starting": pd.to_datetime(
+                    [
+                        "2012-06-21 00:00",
+                        "2012-06-23 07:00",
+                        "2012-06-23 16:30",
+                        "2012-06-25 08:00",
+                        "2012-06-26 12:00",
+                    ]
+                ),
+                "ending": pd.to_datetime(
+                    [
+                        "2012-06-23 07:00",
+                        "2012-06-23 16:30",
+                        "2012-06-25 08:00",
+                        "2012-06-26 12:00",
+                        "2012-06-27 08:00",
+                    ]
+                ),
+                "measure": [77, 65, 77, 0, 77],
+            }
+        )
+
+        ser_starting = df.starting
+        ser_starting.index = ser_starting.values
+        ser_starting = ser_starting.tz_localize("US/Eastern")
+        ser_starting = ser_starting.tz_convert("UTC")
+        ser_starting.index.name = "starting"
+
+        ser_ending = df.ending
+        ser_ending.index = ser_ending.values
+        ser_ending = ser_ending.tz_localize("US/Eastern")
+        ser_ending = ser_ending.tz_convert("UTC")
+        ser_ending.index.name = "ending"
+
+        df.starting = ser_starting.index
+        df.ending = ser_ending.index
+
+        tm.assert_index_equal(pd.DatetimeIndex(df.starting), ser_starting.index)
+        tm.assert_index_equal(pd.DatetimeIndex(df.ending), ser_ending.index)
+
+    def test_is_mixed_type(self, float_frame, float_string_frame):
+        assert not float_frame._is_mixed_type
+        assert float_string_frame._is_mixed_type
+
+    def test_stale_cached_series_bug_473(self):
+        # this is chained, but ok
+        with option_context("chained_assignment", None):
+            Y = DataFrame(
+                np.random.default_rng(2).random((4, 4)),
+                index=("a", "b", "c", "d"),
+                columns=("e", "f", "g", "h"),
+            )
+            repr(Y)
+            Y["e"] = Y["e"].astype("object")
+            with tm.raises_chained_assignment_error():
+                Y["g"]["c"] = np.nan
+            repr(Y)
+            Y.sum()
+            Y["g"].sum()
+            assert not pd.isna(Y["g"]["c"])
+
+    def test_strange_column_corruption_issue(self, performance_warning):
+        # TODO(wesm): Unclear how exactly this is related to internal matters
+        df = DataFrame(index=[0, 1])
+        df[0] = np.nan
+        wasCol = {}
+
+        with tm.assert_produces_warning(
+            performance_warning, raise_on_extra_warnings=False
+        ):
+            for i, dt in enumerate(df.index):
+                for col in range(100, 200):
+                    if col not in wasCol:
+                        wasCol[col] = 1
+                        df[col] = np.nan
+                    df.loc[dt, col] = i
+
+        myid = 100
+
+        first = len(df.loc[pd.isna(df[myid]), [myid]])
+        second = len(df.loc[pd.isna(df[myid]), [myid]])
+        assert first == second == 0
+
+    def test_constructor_no_pandas_array(self):
+        # Ensure that NumpyExtensionArray isn't allowed inside Series
+        # See https://github.com/pandas-dev/pandas/issues/23995 for more.
+        arr = Series([1, 2, 3]).array
+        result = DataFrame({"A": arr})
+        expected = DataFrame({"A": [1, 2, 3]})
+        tm.assert_frame_equal(result, expected)
+        assert isinstance(result._mgr.blocks[0], NumpyBlock)
+        assert result._mgr.blocks[0].is_numeric
+
+    def test_add_column_with_pandas_array(self):
+        # GH 26390
+        df = DataFrame({"a": [1, 2, 3, 4], "b": ["a", "b", "c", "d"]})
+        df["c"] = pd.arrays.NumpyExtensionArray(np.array([1, 2, None, 3], dtype=object))
+        df2 = DataFrame(
+            {
+                "a": [1, 2, 3, 4],
+                "b": ["a", "b", "c", "d"],
+                "c": pd.arrays.NumpyExtensionArray(
+                    np.array([1, 2, None, 3], dtype=object)
+                ),
+            }
+        )
+        assert type(df["c"]._mgr.blocks[0]) == NumpyBlock
+        assert df["c"]._mgr.blocks[0].is_object
+        assert type(df2["c"]._mgr.blocks[0]) == NumpyBlock
+        assert df2["c"]._mgr.blocks[0].is_object
+        tm.assert_frame_equal(df, df2)
+
+
+def test_update_inplace_sets_valid_block_values():
+    # https://github.com/pandas-dev/pandas/issues/33457
+    df = DataFrame({"a": Series([1, 2, None], dtype="category")})
+
+    # inplace update of a single column
+    with tm.raises_chained_assignment_error():
+        df["a"].fillna(1, inplace=True)
+
+    # check we haven't put a Series into any block.values
+    assert isinstance(df._mgr.blocks[0].values, Categorical)
+
+
+def test_nonconsolidated_item_cache_take():
+    # https://github.com/pandas-dev/pandas/issues/35521
+
+    # create non-consolidated dataframe with object dtype columns
+    df = DataFrame(
+        {
+            "col1": Series(["a"], dtype=object),
+        }
+    )
+    df["col2"] = Series([0], dtype=object)
+    assert not df._mgr.is_consolidated()
+
+    # access column (item cache)
+    df["col1"] == "A"
+    # take operation
+    # (regression was that this consolidated but didn't reset item cache,
+    # resulting in an invalid cache and the .at operation not working properly)
+    df[df["col2"] == 0]
+
+    # now setting value should update actual dataframe
+    df.at[0, "col1"] = "A"
+
+    expected = DataFrame({"col1": ["A"], "col2": [0]}, dtype=object)
+    tm.assert_frame_equal(df, expected)
+    assert df.at[0, "col1"] == "A"
diff --git a/pandas/pandas/tests/frame/test_constructors.py b/pandas/pandas/tests/frame/test_constructors.py
new file mode 100644
index 0000000000000..037a2ae294bb2
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_constructors.py
@@ -0,0 +1,3340 @@
+import array
+from collections import (
+    OrderedDict,
+    abc,
+    defaultdict,
+    namedtuple,
+)
+from collections.abc import Iterator
+from dataclasses import make_dataclass
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+import functools
+import re
+import zoneinfo
+
+import numpy as np
+from numpy import ma
+from numpy.ma import mrecords
+import pytest
+
+from pandas._libs import lib
+from pandas.compat.numpy import np_version_gt2
+from pandas.errors import IntCastingNaNError
+
+from pandas.core.dtypes.common import is_integer_dtype
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    IntervalDtype,
+    NumpyEADtype,
+    PeriodDtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Interval,
+    MultiIndex,
+    Period,
+    RangeIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    cut,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+from pandas.arrays import (
+    DatetimeArray,
+    IntervalArray,
+    PeriodArray,
+    SparseArray,
+    TimedeltaArray,
+)
+
+MIXED_FLOAT_DTYPES = ["float16", "float32", "float64"]
+MIXED_INT_DTYPES = [
+    "uint8",
+    "uint16",
+    "uint32",
+    "uint64",
+    "int8",
+    "int16",
+    "int32",
+    "int64",
+]
+
+
+class TestDataFrameConstructors:
+    def test_constructor_from_ndarray_with_str_dtype(self):
+        # If we don't ravel/reshape around ensure_str_array, we end up
+        #  with an array of strings each of which is e.g. "[0 1 2]"
+        arr = np.arange(12).reshape(4, 3)
+        df = DataFrame(arr, dtype=str)
+        expected = DataFrame(arr.astype(str), dtype="str")
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_from_2d_datetimearray(self):
+        dti = date_range("2016-01-01", periods=6, tz="US/Pacific")
+        dta = dti._data.reshape(3, 2)
+
+        df = DataFrame(dta)
+        expected = DataFrame({0: dta[:, 0], 1: dta[:, 1]})
+        tm.assert_frame_equal(df, expected)
+        # GH#44724 big performance hit if we de-consolidate
+        assert len(df._mgr.blocks) == 1
+
+    def test_constructor_dict_with_tzaware_scalar(self):
+        # GH#42505
+        dt = Timestamp("2019-11-03 01:00:00-0700").tz_convert("America/Los_Angeles")
+        dt = dt.as_unit("ns")
+
+        df = DataFrame({"dt": dt}, index=[0])
+        expected = DataFrame({"dt": [dt]})
+        tm.assert_frame_equal(df, expected, check_index_type=False)
+
+        # Non-homogeneous
+        df = DataFrame({"dt": dt, "value": [1]})
+        expected = DataFrame({"dt": [dt], "value": [1]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_construct_ndarray_with_nas_and_int_dtype(self):
+        # GH#26919 match Series by not casting np.nan to meaningless int
+        arr = np.array([[1, np.nan], [2, 3]])
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            DataFrame(arr, dtype="i8")
+
+        # check this matches Series behavior
+        with pytest.raises(IntCastingNaNError, match=msg):
+            Series(arr[0], dtype="i8", name=0)
+
+    def test_construct_from_list_of_datetimes(self):
+        df = DataFrame([datetime.now(), datetime.now()])
+        assert df[0].dtype == np.dtype("M8[us]")
+
+    def test_constructor_from_tzaware_datetimeindex(self):
+        # don't cast a DatetimeIndex WITH a tz, leave as object
+        # GH#6032
+        naive = DatetimeIndex(["2013-1-1 13:00", "2013-1-2 14:00"], name="B")
+        idx = naive.tz_localize("US/Pacific")
+
+        expected = Series(np.array(idx.tolist(), dtype="object"), name="B")
+        assert expected.dtype == idx.dtype
+
+        # convert index to series
+        result = Series(idx)
+        tm.assert_series_equal(result, expected)
+
+    def test_columns_with_leading_underscore_work_with_to_dict(self):
+        col_underscore = "_b"
+        df = DataFrame({"a": [1, 2], col_underscore: [3, 4]})
+        d = df.to_dict(orient="records")
+
+        ref_d = [{"a": 1, col_underscore: 3}, {"a": 2, col_underscore: 4}]
+
+        assert ref_d == d
+
+    def test_columns_with_leading_number_and_underscore_work_with_to_dict(self):
+        col_with_num = "1_b"
+        df = DataFrame({"a": [1, 2], col_with_num: [3, 4]})
+        d = df.to_dict(orient="records")
+
+        ref_d = [{"a": 1, col_with_num: 3}, {"a": 2, col_with_num: 4}]
+
+        assert ref_d == d
+
+    def test_array_of_dt64_nat_with_td64dtype_raises(self, frame_or_series):
+        # GH#39462
+        nat = np.datetime64("NaT", "ns")
+        arr = np.array([nat], dtype=object)
+        if frame_or_series is DataFrame:
+            arr = arr.reshape(1, 1)
+
+        msg = "Invalid type for timedelta scalar: <class 'numpy.datetime64'>"
+        with pytest.raises(TypeError, match=msg):
+            frame_or_series(arr, dtype="m8[ns]")
+
+    @pytest.mark.parametrize("kind", ["m", "M"])
+    def test_datetimelike_values_with_object_dtype(self, kind, frame_or_series):
+        # with dtype=object, we should cast dt64 values to Timestamps, not pydatetimes
+        if kind == "M":
+            dtype = "M8[ns]"
+            scalar_type = Timestamp
+        else:
+            dtype = "m8[ns]"
+            scalar_type = Timedelta
+
+        arr = np.arange(6, dtype="i8").view(dtype).reshape(3, 2)
+        if frame_or_series is Series:
+            arr = arr[:, 0]
+
+        obj = frame_or_series(arr, dtype=object)
+        assert obj._mgr.blocks[0].values.dtype == object
+        assert isinstance(obj._mgr.blocks[0].values.ravel()[0], scalar_type)
+
+        # go through a different path in internals.construction
+        obj = frame_or_series(frame_or_series(arr), dtype=object)
+        assert obj._mgr.blocks[0].values.dtype == object
+        assert isinstance(obj._mgr.blocks[0].values.ravel()[0], scalar_type)
+
+        obj = frame_or_series(frame_or_series(arr), dtype=NumpyEADtype(object))
+        assert obj._mgr.blocks[0].values.dtype == object
+        assert isinstance(obj._mgr.blocks[0].values.ravel()[0], scalar_type)
+
+        if frame_or_series is DataFrame:
+            # other paths through internals.construction
+            sers = [Series(x) for x in arr]
+            obj = frame_or_series(sers, dtype=object)
+            assert obj._mgr.blocks[0].values.dtype == object
+            assert isinstance(obj._mgr.blocks[0].values.ravel()[0], scalar_type)
+
+    def test_series_with_name_not_matching_column(self):
+        # GH#9232
+        x = Series(range(5), name=1)
+        y = Series(range(5), name=0)
+
+        result = DataFrame(x, columns=[0])
+        expected = DataFrame([], columns=[0])
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(y, columns=[1])
+        expected = DataFrame([], columns=[1])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "constructor",
+        [
+            lambda: DataFrame(),
+            lambda: DataFrame(None),
+            lambda: DataFrame(()),
+            lambda: DataFrame([]),
+            lambda: DataFrame(_ for _ in []),
+            lambda: DataFrame(range(0)),
+            lambda: DataFrame(data=None),
+            lambda: DataFrame(data=()),
+            lambda: DataFrame(data=[]),
+            lambda: DataFrame(data=(_ for _ in [])),
+            lambda: DataFrame(data=range(0)),
+        ],
+    )
+    def test_empty_constructor(self, constructor):
+        expected = DataFrame()
+        result = constructor()
+        assert len(result.index) == 0
+        assert len(result.columns) == 0
+        tm.assert_frame_equal(result, expected)
+
+    def test_empty_constructor_object_index(self):
+        expected = DataFrame(index=RangeIndex(0), columns=RangeIndex(0))
+        result = DataFrame({})
+        assert len(result.index) == 0
+        assert len(result.columns) == 0
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    @pytest.mark.parametrize(
+        "emptylike,expected_index,expected_columns",
+        [
+            ([[]], RangeIndex(1), RangeIndex(0)),
+            ([[], []], RangeIndex(2), RangeIndex(0)),
+            ([(_ for _ in [])], RangeIndex(1), RangeIndex(0)),
+        ],
+    )
+    def test_emptylike_constructor(self, emptylike, expected_index, expected_columns):
+        expected = DataFrame(index=expected_index, columns=expected_columns)
+        result = DataFrame(emptylike)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_mixed(self, float_string_frame, using_infer_string):
+        dtype = "str" if using_infer_string else np.object_
+        assert float_string_frame["foo"].dtype == dtype
+
+    def test_constructor_cast_failure(self):
+        # as of 2.0, we raise if we can't respect "dtype", previously we
+        #  silently ignored
+        msg = "could not convert string to float"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"a": ["a", "b", "c"]}, dtype=np.float64)
+
+        # GH 3010, constructing with odd arrays
+        df = DataFrame(np.ones((4, 2)))
+
+        # this is ok
+        df["foo"] = np.ones((4, 2)).tolist()
+
+        # this is not ok
+        msg = "Expected a 1D array, got an array with shape \\(4, 2\\)"
+        with pytest.raises(ValueError, match=msg):
+            df["test"] = np.ones((4, 2))
+
+        # this is ok
+        df["foo2"] = np.ones((4, 2)).tolist()
+
+    def test_constructor_dtype_copy(self):
+        orig_df = DataFrame({"col1": [1.0], "col2": [2.0], "col3": [3.0]})
+
+        new_df = DataFrame(orig_df, dtype=float, copy=True)
+
+        new_df["col1"] = 200.0
+        assert orig_df["col1"][0] == 1.0
+
+    def test_constructor_dtype_nocast_view_dataframe(self):
+        df = DataFrame([[1, 2]])
+        should_be_view = DataFrame(df, dtype=df[0].dtype)
+        should_be_view.iloc[0, 0] = 99
+        assert df.values[0, 0] == 1
+
+    def test_constructor_dtype_nocast_view_2d_array(self):
+        df = DataFrame([[1, 2], [3, 4]], dtype="int64")
+        df2 = DataFrame(df.values, dtype=df[0].dtype)
+        assert df2._mgr.blocks[0].values.flags.c_contiguous
+
+    def test_1d_object_array_does_not_copy(self, using_infer_string):
+        # https://github.com/pandas-dev/pandas/issues/39272
+        arr = np.array(["a", "b"], dtype="object")
+        df = DataFrame(arr, copy=False)
+        if using_infer_string:
+            if df[0].dtype.storage == "pyarrow":
+                # object dtype strings are converted to arrow memory,
+                # no numpy arrays to compare
+                pass
+            else:
+                assert np.shares_memory(df[0].to_numpy(), arr)
+        else:
+            assert np.shares_memory(df.values, arr)
+
+        df = DataFrame(arr, dtype=object, copy=False)
+        assert np.shares_memory(df.values, arr)
+
+    def test_2d_object_array_does_not_copy(self, using_infer_string):
+        # https://github.com/pandas-dev/pandas/issues/39272
+        arr = np.array([["a", "b"], ["c", "d"]], dtype="object")
+        df = DataFrame(arr, copy=False)
+        if using_infer_string:
+            if df[0].dtype.storage == "pyarrow":
+                # object dtype strings are converted to arrow memory,
+                # no numpy arrays to compare
+                pass
+            else:
+                assert np.shares_memory(df[0].to_numpy(), arr)
+        else:
+            assert np.shares_memory(df.values, arr)
+
+        df = DataFrame(arr, dtype=object, copy=False)
+        assert np.shares_memory(df.values, arr)
+
+    def test_constructor_dtype_list_data(self):
+        df = DataFrame([[1, "2"], [None, "a"]], dtype=object)
+        assert df.loc[1, 0] is None
+        assert df.loc[0, 1] == "2"
+
+    def test_constructor_list_of_2d_raises(self):
+        # https://github.com/pandas-dev/pandas/issues/32289
+        a = DataFrame()
+        b = np.empty((0, 0))
+        with pytest.raises(ValueError, match=r"shape=\(1, 0, 0\)"):
+            DataFrame([a])
+
+        with pytest.raises(ValueError, match=r"shape=\(1, 0, 0\)"):
+            DataFrame([b])
+
+        a = DataFrame({"A": [1, 2]})
+        with pytest.raises(ValueError, match=r"shape=\(2, 2, 1\)"):
+            DataFrame([a, a])
+
+    @pytest.mark.parametrize(
+        "typ, ad",
+        [
+            # mixed floating and integer coexist in the same frame
+            ["float", {}],
+            # add lots of types
+            ["float", {"A": 1, "B": "foo", "C": "bar"}],
+            # GH 622
+            ["int", {}],
+        ],
+    )
+    def test_constructor_mixed_dtypes(self, typ, ad):
+        if typ == "int":
+            dtypes = MIXED_INT_DTYPES
+            arrays = [
+                np.array(np.random.default_rng(2).random(10), dtype=d) for d in dtypes
+            ]
+        elif typ == "float":
+            dtypes = MIXED_FLOAT_DTYPES
+            arrays = [
+                np.array(np.random.default_rng(2).integers(10, size=10), dtype=d)
+                for d in dtypes
+            ]
+
+        for d, a in zip(dtypes, arrays):
+            assert a.dtype == d
+        ad.update(dict(zip(dtypes, arrays)))
+        df = DataFrame(ad)
+
+        dtypes = MIXED_FLOAT_DTYPES + MIXED_INT_DTYPES
+        for d in dtypes:
+            if d in df:
+                assert df.dtypes[d] == d
+
+    def test_constructor_complex_dtypes(self):
+        # GH10952
+        a = np.random.default_rng(2).random(10).astype(np.complex64)
+        b = np.random.default_rng(2).random(10).astype(np.complex128)
+
+        df = DataFrame({"a": a, "b": b})
+        assert a.dtype == df.a.dtype
+        assert b.dtype == df.b.dtype
+
+    def test_constructor_dtype_str_na_values(self, string_dtype):
+        # https://github.com/pandas-dev/pandas/issues/21083
+        df = DataFrame({"A": ["x", None]}, dtype=string_dtype)
+        result = df.isna()
+        expected = DataFrame({"A": [False, True]})
+        tm.assert_frame_equal(result, expected)
+        assert df.iloc[1, 0] is None
+
+        df = DataFrame({"A": ["x", np.nan]}, dtype=string_dtype)
+        assert np.isnan(df.iloc[1, 0])
+
+    def test_constructor_rec(self, float_frame):
+        rec = float_frame.to_records(index=False)
+        rec.dtype.names = list(rec.dtype.names)[::-1]
+
+        index = float_frame.index
+
+        df = DataFrame(rec)
+        tm.assert_index_equal(df.columns, Index(rec.dtype.names))
+
+        df2 = DataFrame(rec, index=index)
+        tm.assert_index_equal(df2.columns, Index(rec.dtype.names))
+        tm.assert_index_equal(df2.index, index)
+
+        # case with columns != the ones we would infer from the data
+        rng = np.arange(len(rec))[::-1]
+        df3 = DataFrame(rec, index=rng, columns=["C", "B"])
+        expected = DataFrame(rec, index=rng).reindex(columns=["C", "B"])
+        tm.assert_frame_equal(df3, expected)
+
+    def test_constructor_bool(self):
+        df = DataFrame({0: np.ones(10, dtype=bool), 1: np.zeros(10, dtype=bool)})
+        assert df.values.dtype == np.bool_
+
+    def test_constructor_overflow_int64(self):
+        # see gh-14881
+        values = np.array([2**64 - i for i in range(1, 10)], dtype=np.uint64)
+
+        result = DataFrame({"a": values})
+        assert result["a"].dtype == np.uint64
+
+        # see gh-2355
+        data_scores = [
+            (6311132704823138710, 273),
+            (2685045978526272070, 23),
+            (8921811264899370420, 45),
+            (17019687244989530680, 270),
+            (9930107427299601010, 273),
+        ]
+        dtype = [("uid", "u8"), ("score", "u8")]
+        data = np.zeros((len(data_scores),), dtype=dtype)
+        data[:] = data_scores
+        df_crawls = DataFrame(data)
+        assert df_crawls["uid"].dtype == np.uint64
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            np.array([2**64], dtype=object),
+            np.array([2**65]),
+            [2**64 + 1],
+            np.array([-(2**63) - 4], dtype=object),
+            np.array([-(2**64) - 1]),
+            [-(2**65) - 2],
+        ],
+    )
+    def test_constructor_int_overflow(self, values):
+        # see gh-18584
+        value = values[0]
+        result = DataFrame(values)
+
+        assert result[0].dtype == object
+        assert result[0][0] == value
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            np.array([1], dtype=np.uint16),
+            np.array([1], dtype=np.uint32),
+            np.array([1], dtype=np.uint64),
+            [np.uint16(1)],
+            [np.uint32(1)],
+            [np.uint64(1)],
+        ],
+    )
+    def test_constructor_numpy_uints(self, values):
+        # GH#47294
+        value = values[0]
+        result = DataFrame(values)
+
+        assert result[0].dtype == value.dtype
+        assert result[0][0] == value
+
+    def test_constructor_ordereddict(self):
+        nitems = 100
+        nums = list(range(nitems))
+        np.random.default_rng(2).shuffle(nums)
+        expected = [f"A{i:d}" for i in nums]
+        df = DataFrame(OrderedDict(zip(expected, [[0]] * nitems)))
+        assert expected == list(df.columns)
+
+    def test_constructor_dict(self):
+        datetime_series = Series(
+            np.arange(30, dtype=np.float64), index=date_range("2020-01-01", periods=30)
+        )
+        # test expects index shifted by 5
+        datetime_series_short = datetime_series[5:]
+
+        frame = DataFrame({"col1": datetime_series, "col2": datetime_series_short})
+
+        # col2 is padded with NaN
+        assert len(datetime_series) == 30
+        assert len(datetime_series_short) == 25
+
+        tm.assert_series_equal(frame["col1"], datetime_series.rename("col1"))
+
+        exp = Series(
+            np.concatenate([[np.nan] * 5, datetime_series_short.values]),
+            index=datetime_series.index,
+            name="col2",
+        )
+        tm.assert_series_equal(exp, frame["col2"])
+
+        frame = DataFrame(
+            {"col1": datetime_series, "col2": datetime_series_short},
+            columns=["col2", "col3", "col4"],
+        )
+
+        assert len(frame) == len(datetime_series_short)
+        assert "col1" not in frame
+        assert isna(frame["col3"]).all()
+
+        # Corner cases
+        assert len(DataFrame()) == 0
+
+        # mix dict and array, wrong size - no spec for which error should raise
+        # first
+        msg = "Mixing dicts with non-Series may lead to ambiguous ordering."
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"A": {"a": "a", "b": "b"}, "B": ["a", "b", "c"]})
+
+    def test_constructor_dict_length1(self):
+        # Length-one dict micro-optimization
+        frame = DataFrame({"A": {"1": 1, "2": 2}})
+        tm.assert_index_equal(frame.index, Index(["1", "2"]))
+
+    def test_constructor_dict_with_index(self):
+        # empty dict plus index
+        idx = Index([0, 1, 2])
+        frame = DataFrame({}, index=idx)
+        assert frame.index is idx
+
+    def test_constructor_dict_with_index_and_columns(self):
+        # empty dict with index and columns
+        idx = Index([0, 1, 2])
+        frame = DataFrame({}, index=idx, columns=idx)
+        assert frame.index is idx
+        assert frame.columns is idx
+        assert len(frame._series) == 3
+
+    def test_constructor_dict_of_empty_lists(self):
+        # with dict of empty list and Series
+        frame = DataFrame({"A": [], "B": []}, columns=["A", "B"])
+        tm.assert_index_equal(frame.index, RangeIndex(0), exact=True)
+
+    def test_constructor_dict_with_none(self):
+        # GH 14381
+        # Dict with None value
+        frame_none = DataFrame({"a": None}, index=[0])
+        frame_none_list = DataFrame({"a": [None]}, index=[0])
+        assert frame_none._get_value(0, "a") is None
+        assert frame_none_list._get_value(0, "a") is None
+        tm.assert_frame_equal(frame_none, frame_none_list)
+
+    def test_constructor_dict_errors(self):
+        # GH10856
+        # dict with scalar values should raise error, even if columns passed
+        msg = "If using all scalar values, you must pass an index"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"a": 0.7})
+
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"a": 0.7}, columns=["a"])
+
+    @pytest.mark.parametrize("scalar", [2, np.nan, None, "D"])
+    def test_constructor_invalid_items_unused(self, scalar):
+        # No error if invalid (scalar) value is in fact not used:
+        result = DataFrame({"a": scalar}, columns=["b"])
+        expected = DataFrame(columns=["b"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("value", [4, np.nan, None, float("nan")])
+    def test_constructor_dict_nan_key(self, value):
+        # GH 18455
+        cols = [1, value, 3]
+        idx = ["a", value]
+        values = [[0, 3], [1, 4], [2, 5]]
+        data = {cols[c]: Series(values[c], index=idx) for c in range(3)}
+        result = DataFrame(data).sort_values(1).sort_values("a", axis=1)
+        expected = DataFrame(
+            np.arange(6, dtype="int64").reshape(2, 3), index=idx, columns=cols
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(data, index=idx).sort_values("a", axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(data, index=idx, columns=cols)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("value", [np.nan, None, float("nan")])
+    def test_constructor_dict_nan_tuple_key(self, value):
+        # GH 18455
+        cols = Index([(11, 21), (value, 22), (13, value)])
+        idx = Index([("a", value), (value, 2)])
+        values = [[0, 3], [1, 4], [2, 5]]
+        data = {cols[c]: Series(values[c], index=idx) for c in range(3)}
+        result = DataFrame(data).sort_values((11, 21)).sort_values(("a", value), axis=1)
+        expected = DataFrame(
+            np.arange(6, dtype="int64").reshape(2, 3), index=idx, columns=cols
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(data, index=idx).sort_values(("a", value), axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        result = DataFrame(data, index=idx, columns=cols)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_dict_order_insertion(self):
+        datetime_series = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        datetime_series_short = datetime_series[:5]
+
+        # GH19018
+        # initialization ordering: by insertion order if python>= 3.6
+        d = {"b": datetime_series_short, "a": datetime_series}
+        frame = DataFrame(data=d)
+        expected = DataFrame(data=d, columns=list("ba"))
+        tm.assert_frame_equal(frame, expected)
+
+    def test_constructor_dict_nan_key_and_columns(self):
+        # GH 16894
+        result = DataFrame({np.nan: [1, 2], 2: [2, 3]}, columns=[np.nan, 2])
+        expected = DataFrame([[1, 2], [2, 3]], columns=[np.nan, 2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_multi_index(self):
+        # GH 4078
+        # construction error with mi and all-nan frame
+        tuples = [(2, 3), (3, 3), (3, 3)]
+        mi = MultiIndex.from_tuples(tuples)
+        df = DataFrame(index=mi, columns=mi)
+        assert isna(df).values.ravel().all()
+
+        tuples = [(3, 3), (2, 3), (3, 3)]
+        mi = MultiIndex.from_tuples(tuples)
+        df = DataFrame(index=mi, columns=mi)
+        assert isna(df).values.ravel().all()
+
+    def test_constructor_2d_index(self):
+        # GH 25416
+        # handling of 2d index in construction
+        df = DataFrame([[1]], columns=[[1]], index=[1, 2])
+        expected = DataFrame(
+            [1, 1],
+            index=Index([1, 2], dtype="int64"),
+            columns=MultiIndex(levels=[[1]], codes=[[0]]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame([[1]], columns=[[1]], index=[[1, 2]])
+        expected = DataFrame(
+            [1, 1],
+            index=MultiIndex(levels=[[1, 2]], codes=[[0, 1]]),
+            columns=MultiIndex(levels=[[1]], codes=[[0]]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_error_msgs(self):
+        msg = "Empty data passed with indices specified."
+        # passing an empty array with columns specified.
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(np.empty(0), index=[1])
+
+        msg = "Mixing dicts with non-Series may lead to ambiguous ordering."
+        # mix dict and array, wrong size
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"A": {"a": "a", "b": "b"}, "B": ["a", "b", "c"]})
+
+        # wrong size ndarray, GH 3105
+        msg = r"Shape of passed values is \(4, 3\), indices imply \(3, 3\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(
+                np.arange(12).reshape((4, 3)),
+                columns=["foo", "bar", "baz"],
+                index=date_range("2000-01-01", periods=3),
+            )
+
+        arr = np.array([[4, 5, 6]])
+        msg = r"Shape of passed values is \(1, 3\), indices imply \(1, 4\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(index=[0], columns=range(4), data=arr)
+
+        arr = np.array([4, 5, 6])
+        msg = r"Shape of passed values is \(3, 1\), indices imply \(1, 4\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(index=[0], columns=range(4), data=arr)
+
+        # higher dim raise exception
+        with pytest.raises(ValueError, match="Must pass 2-d input"):
+            DataFrame(np.zeros((3, 3, 3)), columns=["A", "B", "C"], index=[1])
+
+        # wrong size axis labels
+        msg = r"Shape of passed values is \(2, 3\), indices imply \(1, 3\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(
+                np.random.default_rng(2).random((2, 3)),
+                columns=["A", "B", "C"],
+                index=[1],
+            )
+
+        msg = r"Shape of passed values is \(2, 3\), indices imply \(2, 2\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(
+                np.random.default_rng(2).random((2, 3)),
+                columns=["A", "B"],
+                index=[1, 2],
+            )
+
+        # gh-26429
+        msg = "2 columns passed, passed data had 10 columns"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame((range(10), range(10, 20)), columns=("ones", "twos"))
+
+        msg = "If using all scalar values, you must pass an index"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"a": False, "b": True})
+
+    def test_constructor_subclass_dict(self, dict_subclass):
+        # Test for passing dict subclass to constructor
+        data = {
+            "col1": dict_subclass((x, 10.0 * x) for x in range(10)),
+            "col2": dict_subclass((x, 20.0 * x) for x in range(10)),
+        }
+        df = DataFrame(data)
+        refdf = DataFrame({col: dict(val.items()) for col, val in data.items()})
+        tm.assert_frame_equal(refdf, df)
+
+        data = dict_subclass(data.items())
+        df = DataFrame(data)
+        tm.assert_frame_equal(refdf, df)
+
+    def test_constructor_defaultdict(self, float_frame):
+        # try with defaultdict
+        data = {}
+        float_frame.loc[: float_frame.index[10], "B"] = np.nan
+
+        for k, v in float_frame.items():
+            dct = defaultdict(dict)
+            dct.update(v.to_dict())
+            data[k] = dct
+        frame = DataFrame(data)
+        expected = frame.reindex(index=float_frame.index)
+        tm.assert_frame_equal(float_frame, expected)
+
+    def test_constructor_dict_block(self):
+        expected = np.array([[4.0, 3.0, 2.0, 1.0]])
+        df = DataFrame(
+            {"d": [4.0], "c": [3.0], "b": [2.0], "a": [1.0]},
+            columns=["d", "c", "b", "a"],
+        )
+        tm.assert_numpy_array_equal(df.values, expected)
+
+    def test_constructor_dict_cast(self, using_infer_string):
+        # cast float tests
+        test_data = {"A": {"1": 1, "2": 2}, "B": {"1": "1", "2": "2", "3": "3"}}
+        frame = DataFrame(test_data, dtype=float)
+        assert len(frame) == 3
+        assert frame["B"].dtype == np.float64
+        assert frame["A"].dtype == np.float64
+
+        frame = DataFrame(test_data)
+        assert len(frame) == 3
+        assert frame["B"].dtype == np.object_ if not using_infer_string else "str"
+        assert frame["A"].dtype == np.float64
+
+    def test_constructor_dict_cast2(self):
+        # can't cast to float
+        test_data = {
+            "A": dict(zip(range(20), [f"word_{i}" for i in range(20)])),
+            "B": dict(zip(range(15), np.random.default_rng(2).standard_normal(15))),
+        }
+        with pytest.raises(ValueError, match="could not convert string"):
+            DataFrame(test_data, dtype=float)
+
+    def test_constructor_dict_dont_upcast(self):
+        d = {"Col1": {"Row1": "A String", "Row2": np.nan}}
+        df = DataFrame(d)
+        assert isinstance(df["Col1"]["Row2"], float)
+
+    def test_constructor_dict_dont_upcast2(self):
+        dm = DataFrame([[1, 2], ["a", "b"]], index=[1, 2], columns=[1, 2])
+        assert isinstance(dm[1][1], int)
+
+    def test_constructor_dict_of_tuples(self):
+        # GH #1491
+        data = {"a": (1, 2, 3), "b": (4, 5, 6)}
+
+        result = DataFrame(data)
+        expected = DataFrame({k: list(v) for k, v in data.items()})
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_constructor_dict_of_ranges(self):
+        # GH 26356
+        data = {"a": range(3), "b": range(3, 6)}
+
+        result = DataFrame(data)
+        expected = DataFrame({"a": [0, 1, 2], "b": [3, 4, 5]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_dict_of_iterators(self):
+        # GH 26349
+        data = {"a": iter(range(3)), "b": reversed(range(3))}
+
+        result = DataFrame(data)
+        expected = DataFrame({"a": [0, 1, 2], "b": [2, 1, 0]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_dict_of_generators(self):
+        # GH 26349
+        data = {"a": (i for i in (range(3))), "b": (i for i in reversed(range(3)))}
+        result = DataFrame(data)
+        expected = DataFrame({"a": [0, 1, 2], "b": [2, 1, 0]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_dict_multiindex(self):
+        d = {
+            ("a", "a"): {("i", "i"): 0, ("i", "j"): 1, ("j", "i"): 2},
+            ("b", "a"): {("i", "i"): 6, ("i", "j"): 5, ("j", "i"): 4},
+            ("b", "c"): {("i", "i"): 7, ("i", "j"): 8, ("j", "i"): 9},
+        }
+        _d = sorted(d.items())
+        df = DataFrame(d)
+        expected = DataFrame(
+            [x[1] for x in _d], index=MultiIndex.from_tuples([x[0] for x in _d])
+        ).T
+        expected.index = MultiIndex.from_tuples(expected.index)
+        tm.assert_frame_equal(
+            df,
+            expected,
+        )
+
+        d["z"] = {"y": 123.0, ("i", "i"): 111, ("i", "j"): 111, ("j", "i"): 111}
+        _d.insert(0, ("z", d["z"]))
+        expected = DataFrame(
+            [x[1] for x in _d], index=Index([x[0] for x in _d], tupleize_cols=False)
+        ).T
+        expected.index = Index(expected.index, tupleize_cols=False)
+        df = DataFrame(d)
+        df = df.reindex(columns=expected.columns, index=expected.index)
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_dict_datetime64_index(self):
+        # GH 10160
+        dates_as_str = ["1984-02-19", "1988-11-06", "1989-12-03", "1990-03-15"]
+
+        def create_data(constructor):
+            return {i: {constructor(s): 2 * i} for i, s in enumerate(dates_as_str)}
+
+        data_datetime64 = create_data(np.datetime64)
+        data_datetime = create_data(lambda x: datetime.strptime(x, "%Y-%m-%d"))
+        data_Timestamp = create_data(Timestamp)
+
+        expected = DataFrame(
+            [
+                [0, None, None, None],
+                [None, 2, None, None],
+                [None, None, 4, None],
+                [None, None, None, 6],
+            ],
+            index=[Timestamp(dt) for dt in dates_as_str],
+        )
+
+        result_datetime64 = DataFrame(data_datetime64)
+        result_datetime = DataFrame(data_datetime)
+        assert result_datetime.index.unit == "us"
+        result_datetime.index = result_datetime.index.as_unit("s")
+        result_Timestamp = DataFrame(data_Timestamp)
+        tm.assert_frame_equal(result_datetime64, expected)
+        tm.assert_frame_equal(result_datetime, expected)
+        tm.assert_frame_equal(result_Timestamp, expected)
+
+    @pytest.mark.parametrize(
+        "klass,name",
+        [
+            (lambda x: np.timedelta64(x, "D"), "timedelta64"),
+            (lambda x: timedelta(days=x), "pytimedelta"),
+            (lambda x: Timedelta(x, "D"), "Timedelta[ns]"),
+            (lambda x: Timedelta(x, "D").as_unit("s"), "Timedelta[s]"),
+        ],
+    )
+    def test_constructor_dict_timedelta64_index(self, klass, name):
+        # GH 10160
+        td_as_int = [1, 2, 3, 4]
+
+        data = {i: {klass(s): 2 * i} for i, s in enumerate(td_as_int)}
+
+        expected = DataFrame(
+            [
+                {0: 0, 1: None, 2: None, 3: None},
+                {0: None, 1: 2, 2: None, 3: None},
+                {0: None, 1: None, 2: 4, 3: None},
+                {0: None, 1: None, 2: None, 3: 6},
+            ],
+            index=[Timedelta(td, "D") for td in td_as_int],
+        )
+
+        result = DataFrame(data)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_period_dict(self):
+        # PeriodIndex
+        a = pd.PeriodIndex(["2012-01", "NaT", "2012-04"], freq="M")
+        b = pd.PeriodIndex(["2012-02-01", "2012-03-01", "NaT"], freq="D")
+        df = DataFrame({"a": a, "b": b})
+        assert df["a"].dtype == a.dtype
+        assert df["b"].dtype == b.dtype
+
+        # list of periods
+        df = DataFrame({"a": a.astype(object).tolist(), "b": b.astype(object).tolist()})
+        assert df["a"].dtype == a.dtype
+        assert df["b"].dtype == b.dtype
+
+    def test_constructor_dict_extension_scalar(self, ea_scalar_and_dtype):
+        ea_scalar, ea_dtype = ea_scalar_and_dtype
+        df = DataFrame({"a": ea_scalar}, index=[0])
+        assert df["a"].dtype == ea_dtype
+
+        expected = DataFrame(index=[0], columns=["a"], data=ea_scalar)
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "data,dtype",
+        [
+            (Period("2020-01"), PeriodDtype("M")),
+            (Interval(left=0, right=5), IntervalDtype("int64", "right")),
+            (
+                Timestamp("2011-01-01", tz="US/Eastern"),
+                DatetimeTZDtype(unit="s", tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_constructor_extension_scalar_data(self, data, dtype):
+        # GH 34832
+        df = DataFrame(index=range(2), columns=["a", "b"], data=data)
+
+        assert df["a"].dtype == dtype
+        assert df["b"].dtype == dtype
+
+        arr = pd.array([data] * 2, dtype=dtype)
+        expected = DataFrame({"a": arr, "b": arr})
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_nested_dict_frame_constructor(self):
+        rng = pd.period_range("1/1/2000", periods=5)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)), columns=rng)
+
+        data = {}
+        for col in df.columns:
+            for row in df.index:
+                data.setdefault(col, {})[row] = df._get_value(row, col)
+
+        result = DataFrame(data, columns=rng)
+        tm.assert_frame_equal(result, df)
+
+        data = {}
+        for col in df.columns:
+            for row in df.index:
+                data.setdefault(row, {})[col] = df._get_value(row, col)
+
+        result = DataFrame(data, index=rng).T
+        tm.assert_frame_equal(result, df)
+
+    def _check_basic_constructor(self, empty):
+        # mat: 2d matrix with shape (3, 2) to input. empty - makes sized
+        # objects
+        mat = empty((2, 3), dtype=float)
+        # 2-D input
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+
+        assert len(frame.index) == 2
+        assert len(frame.columns) == 3
+
+        # 1-D input
+        frame = DataFrame(empty((3,)), columns=["A"], index=[1, 2, 3])
+        assert len(frame.index) == 3
+        assert len(frame.columns) == 1
+
+        if empty is not np.ones:
+            msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+            with pytest.raises(IntCastingNaNError, match=msg):
+                DataFrame(mat, columns=["A", "B", "C"], index=[1, 2], dtype=np.int64)
+            return
+        else:
+            frame = DataFrame(
+                mat, columns=["A", "B", "C"], index=[1, 2], dtype=np.int64
+            )
+            assert frame.values.dtype == np.int64
+
+        # wrong size axis labels
+        msg = r"Shape of passed values is \(2, 3\), indices imply \(1, 3\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(mat, columns=["A", "B", "C"], index=[1])
+        msg = r"Shape of passed values is \(2, 3\), indices imply \(2, 2\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(mat, columns=["A", "B"], index=[1, 2])
+
+        # higher dim raise exception
+        with pytest.raises(ValueError, match="Must pass 2-d input"):
+            DataFrame(empty((3, 3, 3)), columns=["A", "B", "C"], index=[1])
+
+        # automatic labeling
+        frame = DataFrame(mat)
+        tm.assert_index_equal(frame.index, Index(range(2)), exact=True)
+        tm.assert_index_equal(frame.columns, Index(range(3)), exact=True)
+
+        frame = DataFrame(mat, index=[1, 2])
+        tm.assert_index_equal(frame.columns, Index(range(3)), exact=True)
+
+        frame = DataFrame(mat, columns=["A", "B", "C"])
+        tm.assert_index_equal(frame.index, Index(range(2)), exact=True)
+
+        # 0-length axis
+        frame = DataFrame(empty((0, 3)))
+        assert len(frame.index) == 0
+
+        frame = DataFrame(empty((3, 0)))
+        assert len(frame.columns) == 0
+
+    def test_constructor_ndarray(self):
+        self._check_basic_constructor(np.ones)
+
+        frame = DataFrame(["foo", "bar"], index=[0, 1], columns=["A"])
+        assert len(frame) == 2
+
+    def test_constructor_maskedarray(self):
+        self._check_basic_constructor(ma.masked_all)
+
+        # Check non-masked values
+        mat = ma.masked_all((2, 3), dtype=float)
+        mat[0, 0] = 1.0
+        mat[1, 2] = 2.0
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+        assert 1.0 == frame["A"][1]
+        assert 2.0 == frame["C"][2]
+
+        # what is this even checking??
+        mat = ma.masked_all((2, 3), dtype=float)
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+        assert np.all(~np.asarray(frame == frame))
+
+    @pytest.mark.filterwarnings(
+        "ignore:elementwise comparison failed:DeprecationWarning"
+    )
+    def test_constructor_maskedarray_nonfloat(self):
+        # masked int promoted to float
+        mat = ma.masked_all((2, 3), dtype=int)
+        # 2-D input
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+
+        assert len(frame.index) == 2
+        assert len(frame.columns) == 3
+        assert np.all(~np.asarray(frame == frame))
+
+        # cast type
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2], dtype=np.float64)
+        assert frame.values.dtype == np.float64
+
+        # Check non-masked values
+        mat2 = ma.copy(mat)
+        mat2[0, 0] = 1
+        mat2[1, 2] = 2
+        frame = DataFrame(mat2, columns=["A", "B", "C"], index=[1, 2])
+        assert 1 == frame["A"][1]
+        assert 2 == frame["C"][2]
+
+        # masked np.datetime64 stays (use NaT as null)
+        mat = ma.masked_all((2, 3), dtype="M8[ns]")
+        # 2-D input
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+
+        assert len(frame.index) == 2
+        assert len(frame.columns) == 3
+        assert isna(frame).values.all()
+
+        # cast type
+        msg = r"datetime64\[ns\] values and dtype=int64 is not supported"
+        with pytest.raises(TypeError, match=msg):
+            DataFrame(mat, columns=["A", "B", "C"], index=[1, 2], dtype=np.int64)
+
+        # Check non-masked values
+        mat2 = ma.copy(mat)
+        mat2[0, 0] = 1
+        mat2[1, 2] = 2
+        frame = DataFrame(mat2, columns=["A", "B", "C"], index=[1, 2])
+        assert 1 == frame["A"].astype("i8")[1]
+        assert 2 == frame["C"].astype("i8")[2]
+
+        # masked bool promoted to object
+        mat = ma.masked_all((2, 3), dtype=bool)
+        # 2-D input
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2])
+
+        assert len(frame.index) == 2
+        assert len(frame.columns) == 3
+        assert np.all(~np.asarray(frame == frame))
+
+        # cast type
+        frame = DataFrame(mat, columns=["A", "B", "C"], index=[1, 2], dtype=object)
+        assert frame.values.dtype == object
+
+        # Check non-masked values
+        mat2 = ma.copy(mat)
+        mat2[0, 0] = True
+        mat2[1, 2] = False
+        frame = DataFrame(mat2, columns=["A", "B", "C"], index=[1, 2])
+        assert frame["A"][1] is True
+        assert frame["C"][2] is False
+
+    def test_constructor_maskedarray_hardened(self):
+        # Check numpy masked arrays with hard masks -- from GH24574
+        mat_hard = ma.masked_all((2, 2), dtype=float).harden_mask()
+        result = DataFrame(mat_hard, columns=["A", "B"], index=[1, 2])
+        expected = DataFrame(
+            {"A": [np.nan, np.nan], "B": [np.nan, np.nan]},
+            columns=["A", "B"],
+            index=[1, 2],
+            dtype=float,
+        )
+        tm.assert_frame_equal(result, expected)
+        # Check case where mask is hard but no data are masked
+        mat_hard = ma.ones((2, 2), dtype=float).harden_mask()
+        result = DataFrame(mat_hard, columns=["A", "B"], index=[1, 2])
+        expected = DataFrame(
+            {"A": [1.0, 1.0], "B": [1.0, 1.0]},
+            columns=["A", "B"],
+            index=[1, 2],
+            dtype=float,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_maskedrecarray_dtype(self):
+        # Ensure constructor honors dtype
+        data = np.ma.array(
+            np.ma.zeros(5, dtype=[("date", "<f8"), ("price", "<f8")]), mask=[False] * 5
+        )
+        data = data.view(mrecords.mrecarray)
+        with pytest.raises(TypeError, match=r"Pass \{name: data\[name\]"):
+            # Support for MaskedRecords deprecated GH#40363
+            DataFrame(data, dtype=int)
+
+    def test_constructor_corner_shape(self):
+        df = DataFrame(index=[])
+        assert df.values.shape == (0, 0)
+
+    @pytest.mark.parametrize(
+        "data, index, columns, dtype, expected",
+        [
+            (None, list(range(10)), ["a", "b"], object, np.object_),
+            (None, None, ["a", "b"], "int64", np.dtype("int64")),
+            (None, list(range(10)), ["a", "b"], int, np.dtype("float64")),
+            ({}, None, ["foo", "bar"], None, np.object_),
+            ({"b": 1}, list(range(10)), list("abc"), int, np.dtype("float64")),
+        ],
+    )
+    def test_constructor_dtype(self, data, index, columns, dtype, expected):
+        df = DataFrame(data, index, columns, dtype)
+        assert df.values.dtype == expected
+
+    @pytest.mark.parametrize(
+        "data,input_dtype,expected_dtype",
+        (
+            ([True, False, None], "boolean", pd.BooleanDtype),
+            ([1.0, 2.0, None], "Float64", pd.Float64Dtype),
+            ([1, 2, None], "Int64", pd.Int64Dtype),
+            (["a", "b", "c"], "string", pd.StringDtype),
+        ),
+    )
+    def test_constructor_dtype_nullable_extension_arrays(
+        self, data, input_dtype, expected_dtype
+    ):
+        df = DataFrame({"a": data}, dtype=input_dtype)
+        assert df["a"].dtype == expected_dtype()
+
+    def test_constructor_scalar_inference(self, using_infer_string):
+        data = {"int": 1, "bool": True, "float": 3.0, "complex": 4j, "object": "foo"}
+        df = DataFrame(data, index=np.arange(10))
+
+        assert df["int"].dtype == np.int64
+        assert df["bool"].dtype == np.bool_
+        assert df["float"].dtype == np.float64
+        assert df["complex"].dtype == np.complex128
+        assert df["object"].dtype == np.object_ if not using_infer_string else "str"
+
+    def test_constructor_arrays_and_scalars(self):
+        df = DataFrame({"a": np.random.default_rng(2).standard_normal(10), "b": True})
+        exp = DataFrame({"a": df["a"].values, "b": [True] * 10})
+
+        tm.assert_frame_equal(df, exp)
+        with pytest.raises(ValueError, match="must pass an index"):
+            DataFrame({"a": False, "b": True})
+
+    def test_constructor_DataFrame(self, float_frame):
+        df = DataFrame(float_frame)
+        tm.assert_frame_equal(df, float_frame)
+
+        df_casted = DataFrame(float_frame, dtype=np.int64)
+        assert df_casted.values.dtype == np.int64
+
+    def test_constructor_empty_dataframe(self):
+        # GH 20624
+        actual = DataFrame(DataFrame(), dtype="object")
+        expected = DataFrame([], dtype="object")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_constructor_more(self, float_frame):
+        # used to be in test_matrix.py
+        arr = np.random.default_rng(2).standard_normal(10)
+        dm = DataFrame(arr, columns=["A"], index=np.arange(10))
+        assert dm.values.ndim == 2
+
+        arr = np.random.default_rng(2).standard_normal(0)
+        dm = DataFrame(arr)
+        assert dm.values.ndim == 2
+        assert dm.values.ndim == 2
+
+        # no data specified
+        dm = DataFrame(columns=["A", "B"], index=np.arange(10))
+        assert dm.values.shape == (10, 2)
+
+        dm = DataFrame(columns=["A", "B"])
+        assert dm.values.shape == (0, 2)
+
+        dm = DataFrame(index=np.arange(10))
+        assert dm.values.shape == (10, 0)
+
+        # can't cast
+        mat = np.array(["foo", "bar"], dtype=object).reshape(2, 1)
+        msg = "could not convert string to float: 'foo'"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(mat, index=[0, 1], columns=[0], dtype=float)
+
+        dm = DataFrame(DataFrame(float_frame._series))
+        tm.assert_frame_equal(dm, float_frame)
+
+        # int cast
+        dm = DataFrame(
+            {"A": np.ones(10, dtype=int), "B": np.ones(10, dtype=np.float64)},
+            index=np.arange(10),
+        )
+
+        assert len(dm.columns) == 2
+        assert dm.values.dtype == np.float64
+
+    def test_constructor_empty_list(self):
+        df = DataFrame([], index=[])
+        expected = DataFrame(index=[])
+        tm.assert_frame_equal(df, expected)
+
+        # GH 9939
+        df = DataFrame([], columns=["A", "B"])
+        expected = DataFrame({}, columns=["A", "B"])
+        tm.assert_frame_equal(df, expected)
+
+        # Empty generator: list(empty_gen()) == []
+        def empty_gen():
+            yield from ()
+
+        df = DataFrame(empty_gen(), columns=["A", "B"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_list_of_lists(self, using_infer_string):
+        # GH #484
+        df = DataFrame(data=[[1, "a"], [2, "b"]], columns=["num", "str"])
+        assert is_integer_dtype(df["num"])
+        assert df["str"].dtype == np.object_ if not using_infer_string else "str"
+
+        # GH 4851
+        # list of 0-dim ndarrays
+        expected = DataFrame(np.arange(10))
+        data = [np.array(x) for x in range(10)]
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_pandasarray_matches_nested_ndarray(self):
+        # GH#43986
+        ser = Series([1, 2])
+
+        arr = np.array([None, None], dtype=object)
+        arr[0] = ser
+        arr[1] = ser * 2
+
+        df = DataFrame(arr)
+        expected = DataFrame(pd.array(arr))
+        tm.assert_frame_equal(df, expected)
+        assert df.shape == (2, 1)
+        tm.assert_numpy_array_equal(df[0].values, arr)
+
+    def test_constructor_list_like_data_nested_list_column(self):
+        # GH 32173
+        arrays = [list("abcd"), list("cdef")]
+        result = DataFrame([[1, 2, 3, 4], [4, 5, 6, 7]], columns=arrays)
+
+        mi = MultiIndex.from_arrays(arrays)
+        expected = DataFrame([[1, 2, 3, 4], [4, 5, 6, 7]], columns=mi)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_wrong_length_nested_list_column(self):
+        # GH 32173
+        arrays = [list("abc"), list("cde")]
+
+        msg = "3 columns passed, passed data had 4"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame([[1, 2, 3, 4], [4, 5, 6, 7]], columns=arrays)
+
+    def test_constructor_unequal_length_nested_list_column(self):
+        # GH 32173
+        arrays = [list("abcd"), list("cde")]
+
+        # exception raised inside MultiIndex constructor
+        msg = "all arrays must be same length"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame([[1, 2, 3, 4], [4, 5, 6, 7]], columns=arrays)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            [[Timestamp("2021-01-01")]],
+            [{"x": Timestamp("2021-01-01")}],
+            {"x": [Timestamp("2021-01-01")]},
+            {"x": Timestamp("2021-01-01")},
+        ],
+    )
+    def test_constructor_one_element_data_list(self, data):
+        # GH#42810
+        result = DataFrame(data, index=range(3), columns=["x"])
+        expected = DataFrame({"x": [Timestamp("2021-01-01")] * 3})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_sequence_like(self):
+        # GH 3783
+        # collections.Sequence like
+
+        class DummyContainer(abc.Sequence):
+            def __init__(self, lst) -> None:
+                self._lst = lst
+
+            def __getitem__(self, n):
+                return self._lst.__getitem__(n)
+
+            def __len__(self) -> int:
+                return self._lst.__len__()
+
+        lst_containers = [DummyContainer([1, "a"]), DummyContainer([2, "b"])]
+        columns = ["num", "str"]
+        result = DataFrame(lst_containers, columns=columns)
+        expected = DataFrame([[1, "a"], [2, "b"]], columns=columns)
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_constructor_stdlib_array(self):
+        # GH 4297
+        # support Array
+        result = DataFrame({"A": array.array("i", range(10))})
+        expected = DataFrame({"A": list(range(10))})
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+        expected = DataFrame([list(range(10)), list(range(10))])
+        result = DataFrame([array.array("i", range(10)), array.array("i", range(10))])
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_constructor_range(self):
+        # GH26342
+        result = DataFrame(range(10))
+        expected = DataFrame(list(range(10)))
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_ranges(self):
+        result = DataFrame([range(10), range(10)])
+        expected = DataFrame([list(range(10)), list(range(10))])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_iterable(self):
+        # GH 21987
+        class Iter:
+            def __iter__(self) -> Iterator:
+                for i in range(10):
+                    yield [1, 2, 3]
+
+        expected = DataFrame([[1, 2, 3]] * 10)
+        result = DataFrame(Iter())
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_iterator(self):
+        result = DataFrame(iter(range(10)))
+        expected = DataFrame(list(range(10)))
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_iterators(self):
+        result = DataFrame([iter(range(10)), iter(range(10))])
+        expected = DataFrame([list(range(10)), list(range(10))])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_generator(self):
+        # related #2305
+
+        gen1 = (i for i in range(10))
+        gen2 = (i for i in range(10))
+
+        expected = DataFrame([list(range(10)), list(range(10))])
+        result = DataFrame([gen1, gen2])
+        tm.assert_frame_equal(result, expected)
+
+        gen = ([i, "a"] for i in range(10))
+        result = DataFrame(gen)
+        expected = DataFrame({0: range(10), 1: "a"})
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_constructor_list_of_dicts(self):
+        result = DataFrame([{}])
+        expected = DataFrame(index=RangeIndex(1), columns=[])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_ordered_dict_nested_preserve_order(self):
+        # see gh-18166
+        nested1 = OrderedDict([("b", 1), ("a", 2)])
+        nested2 = OrderedDict([("b", 2), ("a", 5)])
+        data = OrderedDict([("col2", nested1), ("col1", nested2)])
+        result = DataFrame(data)
+        data = {"col2": [1, 2], "col1": [2, 5]}
+        expected = DataFrame(data=data, index=["b", "a"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dict_type", [dict, OrderedDict])
+    def test_constructor_ordered_dict_preserve_order(self, dict_type):
+        # see gh-13304
+        expected = DataFrame([[2, 1]], columns=["b", "a"])
+
+        data = dict_type()
+        data["b"] = [2]
+        data["a"] = [1]
+
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+        data = dict_type()
+        data["b"] = 2
+        data["a"] = 1
+
+        result = DataFrame([data])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dict_type", [dict, OrderedDict])
+    def test_constructor_ordered_dict_conflicting_orders(self, dict_type):
+        # the first dict element sets the ordering for the DataFrame,
+        # even if there are conflicting orders from subsequent ones
+        row_one = dict_type()
+        row_one["b"] = 2
+        row_one["a"] = 1
+
+        row_two = dict_type()
+        row_two["a"] = 1
+        row_two["b"] = 2
+
+        row_three = {"b": 2, "a": 1}
+
+        expected = DataFrame([[2, 1], [2, 1]], columns=["b", "a"])
+        result = DataFrame([row_one, row_two])
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([[2, 1], [2, 1], [2, 1]], columns=["b", "a"])
+        result = DataFrame([row_one, row_two, row_three])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_series_aligned_index(self):
+        series = [Series(i, index=["b", "a", "c"], name=str(i)) for i in range(3)]
+        result = DataFrame(series)
+        expected = DataFrame(
+            {"b": [0, 1, 2], "a": [0, 1, 2], "c": [0, 1, 2]},
+            columns=["b", "a", "c"],
+            index=["0", "1", "2"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_derived_dicts(self):
+        class CustomDict(dict):
+            pass
+
+        d = {"a": 1.5, "b": 3}
+
+        data_custom = [CustomDict(d)]
+        data = [d]
+
+        result_custom = DataFrame(data_custom)
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, result_custom)
+
+    def test_constructor_ragged(self):
+        data = {
+            "A": np.random.default_rng(2).standard_normal(10),
+            "B": np.random.default_rng(2).standard_normal(8),
+        }
+        with pytest.raises(ValueError, match="All arrays must be of the same length"):
+            DataFrame(data)
+
+    def test_constructor_scalar(self):
+        idx = Index(range(3))
+        df = DataFrame({"a": 0}, index=idx)
+        expected = DataFrame({"a": [0, 0, 0]}, index=idx)
+        tm.assert_frame_equal(df, expected, check_dtype=False)
+
+    def test_constructor_Series_copy_bug(self, float_frame):
+        df = DataFrame(float_frame["A"], index=float_frame.index, columns=["A"])
+        df.copy()
+
+    def test_constructor_mixed_dict_and_Series(self):
+        data = {}
+        data["A"] = {"foo": 1, "bar": 2, "baz": 3}
+        data["B"] = Series([4, 3, 2, 1], index=["bar", "qux", "baz", "foo"])
+
+        result = DataFrame(data)
+        assert result.index.is_monotonic_increasing
+
+        # ordering ambiguous, raise exception
+        with pytest.raises(ValueError, match="ambiguous ordering"):
+            DataFrame({"A": ["a", "b"], "B": {"a": "a", "b": "b"}})
+
+        # this is OK though
+        result = DataFrame({"A": ["a", "b"], "B": Series(["a", "b"], index=["a", "b"])})
+        expected = DataFrame({"A": ["a", "b"], "B": ["a", "b"]}, index=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_mixed_type_rows(self):
+        # Issue 25075
+        data = [[1, 2], (3, 4)]
+        result = DataFrame(data)
+        expected = DataFrame([[1, 2], [3, 4]])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tuples,lists",
+        [
+            ((), []),
+            (((),), [[]]),
+            (((), ()), [(), ()]),
+            (((), ()), [[], []]),
+            (([], []), [[], []]),
+            (([1], [2]), [[1], [2]]),  # GH 32776
+            (([1, 2, 3], [4, 5, 6]), [[1, 2, 3], [4, 5, 6]]),
+        ],
+    )
+    def test_constructor_tuple(self, tuples, lists):
+        # GH 25691
+        result = DataFrame(tuples)
+        expected = DataFrame(lists)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_tuples(self):
+        result = DataFrame({"A": [(1, 2), (3, 4)]})
+        expected = DataFrame({"A": Series([(1, 2), (3, 4)])})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_namedtuples(self):
+        # GH11181
+        named_tuple = namedtuple("Pandas", list("ab"))
+        tuples = [named_tuple(1, 3), named_tuple(2, 4)]
+        expected = DataFrame({"a": [1, 2], "b": [3, 4]})
+        result = DataFrame(tuples)
+        tm.assert_frame_equal(result, expected)
+
+        # with columns
+        expected = DataFrame({"y": [1, 2], "z": [3, 4]})
+        result = DataFrame(tuples, columns=["y", "z"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_dataclasses(self):
+        # GH21910
+        Point = make_dataclass("Point", [("x", int), ("y", int)])
+
+        data = [Point(0, 3), Point(1, 3)]
+        expected = DataFrame({"x": [0, 1], "y": [3, 3]})
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_dataclasses_with_varying_types(self):
+        # GH21910
+        # varying types
+        Point = make_dataclass("Point", [("x", int), ("y", int)])
+        HLine = make_dataclass("HLine", [("x0", int), ("x1", int), ("y", int)])
+
+        data = [Point(0, 3), HLine(1, 3, 3)]
+
+        expected = DataFrame(
+            {"x": [0, np.nan], "y": [3, 3], "x0": [np.nan, 1], "x1": [np.nan, 3]}
+        )
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_of_dataclasses_error_thrown(self):
+        # GH21910
+        Point = make_dataclass("Point", [("x", int), ("y", int)])
+
+        # expect TypeError
+        msg = "asdict() should be called on dataclass instances"
+        with pytest.raises(TypeError, match=re.escape(msg)):
+            DataFrame([Point(0, 0), {"x": 1, "y": 0}])
+
+    def test_constructor_list_of_dict_order(self):
+        # GH10056
+        data = [
+            {"First": 1, "Second": 4, "Third": 7, "Fourth": 10},
+            {"Second": 5, "First": 2, "Fourth": 11, "Third": 8},
+            {"Second": 6, "First": 3, "Fourth": 12, "Third": 9, "YYY": 14, "XXX": 13},
+        ]
+        expected = DataFrame(
+            {
+                "First": [1, 2, 3],
+                "Second": [4, 5, 6],
+                "Third": [7, 8, 9],
+                "Fourth": [10, 11, 12],
+                "YYY": [None, None, 14],
+                "XXX": [None, None, 13],
+            }
+        )
+        result = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_Series_named(self):
+        a = Series([1, 2, 3], index=["a", "b", "c"], name="x")
+        df = DataFrame(a)
+        assert df.columns[0] == "x"
+        tm.assert_index_equal(df.index, a.index)
+
+        # ndarray like
+        arr = np.random.default_rng(2).standard_normal(10)
+        s = Series(arr, name="x")
+        df = DataFrame(s)
+        expected = DataFrame({"x": s})
+        tm.assert_frame_equal(df, expected)
+
+        s = Series(arr, index=range(3, 13))
+        df = DataFrame(s)
+        expected = DataFrame({0: s})
+        tm.assert_frame_equal(df, expected, check_column_type=False)
+
+        msg = r"Shape of passed values is \(10, 1\), indices imply \(10, 2\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(s, columns=[1, 2])
+
+        # #2234
+        a = Series([], name="x", dtype=object)
+        df = DataFrame(a)
+        assert df.columns[0] == "x"
+
+        # series with name and w/o
+        s1 = Series(arr, name="x")
+        df = DataFrame([s1, arr]).T
+        expected = DataFrame({"x": s1, "Unnamed 0": arr}, columns=["x", "Unnamed 0"])
+        tm.assert_frame_equal(df, expected)
+
+        # this is a bit non-intuitive here; the series collapse down to arrays
+        df = DataFrame([arr, s1]).T
+        expected = DataFrame({1: s1, 0: arr}, columns=range(2))
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_Series_named_and_columns(self):
+        # GH 9232 validation
+
+        s0 = Series(range(5), name=0)
+        s1 = Series(range(5), name=1)
+
+        # matching name and column gives standard frame
+        tm.assert_frame_equal(DataFrame(s0, columns=[0]), s0.to_frame())
+        tm.assert_frame_equal(DataFrame(s1, columns=[1]), s1.to_frame())
+
+        # non-matching produces empty frame
+        assert DataFrame(s0, columns=[1]).empty
+        assert DataFrame(s1, columns=[0]).empty
+
+    def test_constructor_Series_differently_indexed(self):
+        # name
+        s1 = Series([1, 2, 3], index=["a", "b", "c"], name="x")
+
+        # no name
+        s2 = Series([1, 2, 3], index=["a", "b", "c"])
+
+        other_index = Index(["a", "b"])
+
+        df1 = DataFrame(s1, index=other_index)
+        exp1 = DataFrame(s1.reindex(other_index))
+        assert df1.columns[0] == "x"
+        tm.assert_frame_equal(df1, exp1)
+
+        df2 = DataFrame(s2, index=other_index)
+        exp2 = DataFrame(s2.reindex(other_index))
+        assert df2.columns[0] == 0
+        tm.assert_index_equal(df2.index, other_index)
+        tm.assert_frame_equal(df2, exp2)
+
+    @pytest.mark.parametrize(
+        "name_in1,name_in2,name_in3,name_out",
+        [
+            ("idx", "idx", "idx", "idx"),
+            ("idx", "idx", None, None),
+            ("idx", None, None, None),
+            ("idx1", "idx2", None, None),
+            ("idx1", "idx1", "idx2", None),
+            ("idx1", "idx2", "idx3", None),
+            (None, None, None, None),
+        ],
+    )
+    def test_constructor_index_names(self, name_in1, name_in2, name_in3, name_out):
+        # GH13475
+        indices = [
+            Index(["a", "b", "c"], name=name_in1),
+            Index(["b", "c", "d"], name=name_in2),
+            Index(["c", "d", "e"], name=name_in3),
+        ]
+        series = {
+            c: Series([0, 1, 2], index=i) for i, c in zip(indices, ["x", "y", "z"])
+        }
+        result = DataFrame(series)
+
+        exp_ind = Index(["a", "b", "c", "d", "e"], name=name_out)
+        expected = DataFrame(
+            {
+                "x": [0, 1, 2, np.nan, np.nan],
+                "y": [np.nan, 0, 1, 2, np.nan],
+                "z": [np.nan, np.nan, 0, 1, 2],
+            },
+            index=exp_ind,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_manager_resize(self, float_frame):
+        index = list(float_frame.index[:5])
+        columns = list(float_frame.columns[:3])
+
+        msg = "Passing a BlockManager to DataFrame"
+        with tm.assert_produces_warning(
+            DeprecationWarning, match=msg, check_stacklevel=False
+        ):
+            result = DataFrame(float_frame._mgr, index=index, columns=columns)
+        tm.assert_index_equal(result.index, Index(index))
+        tm.assert_index_equal(result.columns, Index(columns))
+
+    def test_constructor_mix_series_nonseries(self, float_frame):
+        df = DataFrame(
+            {"A": float_frame["A"], "B": list(float_frame["B"])}, columns=["A", "B"]
+        )
+        tm.assert_frame_equal(df, float_frame.loc[:, ["A", "B"]])
+
+        msg = "does not match index length"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"A": float_frame["A"], "B": list(float_frame["B"])[:-2]})
+
+    def test_constructor_miscast_na_int_dtype(self):
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+
+        with pytest.raises(IntCastingNaNError, match=msg):
+            DataFrame([[np.nan, 1], [1, 0]], dtype=np.int64)
+
+    def test_constructor_column_duplicates(self):
+        # it works! #2079
+        df = DataFrame([[8, 5]], columns=["a", "a"])
+        edf = DataFrame([[8, 5]])
+        edf.columns = ["a", "a"]
+
+        tm.assert_frame_equal(df, edf)
+
+        idf = DataFrame.from_records([(8, 5)], columns=["a", "a"])
+
+        tm.assert_frame_equal(idf, edf)
+
+    def test_constructor_empty_with_string_dtype(self, using_infer_string):
+        # GH 9428
+        expected = DataFrame(index=[0, 1], columns=[0, 1], dtype=object)
+        expected_str = DataFrame(
+            index=[0, 1], columns=[0, 1], dtype=pd.StringDtype(na_value=np.nan)
+        )
+
+        df = DataFrame(index=[0, 1], columns=[0, 1], dtype=str)
+        if using_infer_string:
+            tm.assert_frame_equal(df, expected_str)
+        else:
+            tm.assert_frame_equal(df, expected)
+        df = DataFrame(index=[0, 1], columns=[0, 1], dtype=np.str_)
+        tm.assert_frame_equal(df, expected)
+        df = DataFrame(index=[0, 1], columns=[0, 1], dtype="U5")
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_empty_with_string_extension(self, nullable_string_dtype):
+        # GH 34915
+        expected = DataFrame(columns=["c1"], dtype=nullable_string_dtype)
+        df = DataFrame(columns=["c1"], dtype=nullable_string_dtype)
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_single_value(self):
+        # expecting single value upcasting here
+        df = DataFrame(0.0, index=[1, 2, 3], columns=["a", "b", "c"])
+        tm.assert_frame_equal(
+            df, DataFrame(np.zeros(df.shape).astype("float64"), df.index, df.columns)
+        )
+
+        df = DataFrame(0, index=[1, 2, 3], columns=["a", "b", "c"])
+        tm.assert_frame_equal(
+            df, DataFrame(np.zeros(df.shape).astype("int64"), df.index, df.columns)
+        )
+
+        df = DataFrame("a", index=[1, 2], columns=["a", "c"])
+        tm.assert_frame_equal(
+            df,
+            DataFrame(
+                np.array([["a", "a"], ["a", "a"]], dtype=object),
+                index=[1, 2],
+                columns=["a", "c"],
+            ),
+        )
+
+        msg = "DataFrame constructor not properly called!"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame("a", [1, 2])
+        with pytest.raises(ValueError, match=msg):
+            DataFrame("a", columns=["a", "c"])
+
+        msg = "incompatible data and dtype"
+        with pytest.raises(TypeError, match=msg):
+            DataFrame("a", [1, 2], ["a", "c"], float)
+
+    def test_constructor_with_datetimes(self, using_infer_string):
+        intname = np.dtype(int).name
+        floatname = np.dtype(np.float64).name
+        objectname = np.dtype(np.object_).name
+
+        # single item
+        df = DataFrame(
+            {
+                "A": 1,
+                "B": "foo",
+                "C": "bar",
+                "D": Timestamp("20010101"),
+                "E": datetime(2001, 1, 2, 0, 0),
+            },
+            index=np.arange(10),
+        )
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("int64")]
+            + [
+                np.dtype(objectname)
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan)
+            ]
+            * 2
+            + [np.dtype("M8[s]"), np.dtype("M8[us]")],
+            index=list("ABCDE"),
+        )
+        tm.assert_series_equal(result, expected)
+
+        # check with ndarray construction ndim==0 (e.g. we are passing a ndim 0
+        # ndarray with a dtype specified)
+        df = DataFrame(
+            {
+                "a": 1.0,
+                "b": 2,
+                "c": "foo",
+                floatname: np.array(1.0, dtype=floatname),
+                intname: np.array(1, dtype=intname),
+            },
+            index=np.arange(10),
+        )
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")]
+            + [np.dtype("int64")]
+            + [
+                np.dtype("object")
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan)
+            ]
+            + [np.dtype("float64")]
+            + [np.dtype(intname)],
+            index=["a", "b", "c", floatname, intname],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # check with ndarray construction ndim>0
+        df = DataFrame(
+            {
+                "a": 1.0,
+                "b": 2,
+                "c": "foo",
+                floatname: np.array([1.0] * 10, dtype=floatname),
+                intname: np.array([1] * 10, dtype=intname),
+            },
+            index=np.arange(10),
+        )
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("float64")]
+            + [np.dtype("int64")]
+            + [
+                np.dtype("object")
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan)
+            ]
+            + [np.dtype("float64")]
+            + [np.dtype(intname)],
+            index=["a", "b", "c", floatname, intname],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_with_datetimes1(self):
+        # GH 2809
+        ind = date_range(start="2000-01-01", freq="D", periods=10)
+        datetimes = [ts.to_pydatetime() for ts in ind]
+        datetime_s = Series(datetimes)
+        assert datetime_s.dtype == "M8[us]"
+
+    def test_constructor_with_datetimes2(self):
+        # GH 2810
+        ind = date_range(start="2000-01-01", freq="D", periods=10)
+        datetimes = [ts.to_pydatetime() for ts in ind]
+        dates = [ts.date() for ts in ind]
+        df = DataFrame(datetimes, columns=["datetimes"])
+        df["dates"] = dates
+        result = df.dtypes
+        expected = Series(
+            [np.dtype("datetime64[us]"), np.dtype("object")],
+            index=["datetimes", "dates"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_with_datetimes3(self):
+        # GH 7594
+        # don't coerce tz-aware
+        dt = datetime(2012, 1, 1, tzinfo=zoneinfo.ZoneInfo("US/Eastern"))
+
+        df = DataFrame({"End Date": dt}, index=[0])
+        assert df.iat[0, 0] == dt
+        tm.assert_series_equal(
+            df.dtypes, Series({"End Date": "datetime64[us, US/Eastern]"}, dtype=object)
+        )
+
+        df = DataFrame([{"End Date": dt}])
+        assert df.iat[0, 0] == dt
+        tm.assert_series_equal(
+            df.dtypes, Series({"End Date": "datetime64[us, US/Eastern]"}, dtype=object)
+        )
+
+    def test_constructor_with_datetimes4(self):
+        # tz-aware (UTC and other tz's)
+        # GH 8411
+        dr = date_range("20130101", periods=3)
+        df = DataFrame({"value": dr})
+        assert df.iat[0, 0].tz is None
+        dr = date_range("20130101", periods=3, tz="UTC")
+        df = DataFrame({"value": dr})
+        assert str(df.iat[0, 0].tz) == "UTC"
+        dr = date_range("20130101", periods=3, tz="US/Eastern")
+        df = DataFrame({"value": dr})
+        assert str(df.iat[0, 0].tz) == "US/Eastern"
+
+    def test_constructor_with_datetimes5(self):
+        # GH 7822
+        # preserver an index with a tz on dict construction
+        i = date_range("1/1/2011", periods=5, freq="10s", tz="US/Eastern")
+
+        expected = DataFrame({"a": i.to_series().reset_index(drop=True)})
+        df = DataFrame()
+        df["a"] = i
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame({"a": i})
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_with_datetimes6(self):
+        # multiples
+        i = date_range("1/1/2011", periods=5, freq="10s", tz="US/Eastern")
+        i_no_tz = date_range("1/1/2011", periods=5, freq="10s")
+        df = DataFrame({"a": i, "b": i_no_tz})
+        expected = DataFrame({"a": i.to_series().reset_index(drop=True), "b": i_no_tz})
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            np.array([None, None, None, None, datetime.now(), None]),
+            np.array([None, None, datetime.now(), None]),
+            [[np.datetime64("NaT")], [None]],
+            [[np.datetime64("NaT")], [pd.NaT]],
+            [[None], [np.datetime64("NaT")]],
+            [[None], [pd.NaT]],
+            [[pd.NaT], [np.datetime64("NaT")]],
+            [[pd.NaT], [None]],
+        ],
+    )
+    def test_constructor_datetimes_with_nulls(self, arr):
+        # gh-15869, GH#11220
+        result = DataFrame(arr).dtypes
+        unit = "ns"
+        if isinstance(arr, np.ndarray):
+            # inferred from a pydatetime object
+            unit = "us"
+        elif not any(isinstance(x, np.datetime64) for y in arr for x in y):
+            # TODO: this condition is not clear about why we have different behavior
+            unit = "s"
+        expected = Series([np.dtype(f"datetime64[{unit}]")])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("order", ["K", "A", "C", "F"])
+    @pytest.mark.parametrize(
+        "unit",
+        ["M", "D", "h", "m", "s", "ms", "us", "ns"],
+    )
+    def test_constructor_datetimes_non_ns(self, order, unit):
+        dtype = f"datetime64[{unit}]"
+        na = np.array(
+            [
+                ["2015-01-01", "2015-01-02", "2015-01-03"],
+                ["2017-01-01", "2017-01-02", "2017-02-03"],
+            ],
+            dtype=dtype,
+            order=order,
+        )
+        df = DataFrame(na)
+        expected = DataFrame(na.astype("M8[ns]"))
+        if unit in ["M", "D", "h", "m"]:
+            with pytest.raises(TypeError, match="Cannot cast"):
+                expected.astype(dtype)
+
+            # instead the constructor casts to the closest supported reso, i.e. "s"
+            expected = expected.astype("datetime64[s]")
+        else:
+            expected = expected.astype(dtype=dtype)
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("order", ["K", "A", "C", "F"])
+    @pytest.mark.parametrize(
+        "unit",
+        [
+            "D",
+            "h",
+            "m",
+            "s",
+            "ms",
+            "us",
+            "ns",
+        ],
+    )
+    def test_constructor_timedelta_non_ns(self, order, unit):
+        dtype = f"timedelta64[{unit}]"
+        na = np.array(
+            [
+                [np.timedelta64(1, "D"), np.timedelta64(2, "D")],
+                [np.timedelta64(4, "D"), np.timedelta64(5, "D")],
+            ],
+            dtype=dtype,
+            order=order,
+        )
+        df = DataFrame(na)
+        if unit in ["D", "h", "m"]:
+            # we get the nearest supported unit, i.e. "s"
+            exp_unit = "s"
+        else:
+            exp_unit = unit
+        exp_dtype = np.dtype(f"m8[{exp_unit}]")
+        expected = DataFrame(
+            [
+                [Timedelta(1, "D"), Timedelta(2, "D")],
+                [Timedelta(4, "D"), Timedelta(5, "D")],
+            ],
+            dtype=exp_dtype,
+        )
+        # TODO(2.0): ideally we should get the same 'expected' without passing
+        #  dtype=exp_dtype.
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_for_list_with_dtypes(self, using_infer_string):
+        # test list of lists/ndarrays
+        df = DataFrame([np.arange(5) for x in range(5)])
+        result = df.dtypes
+        expected = Series([np.dtype("int")] * 5)
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame([np.array(np.arange(5), dtype="int32") for x in range(5)])
+        result = df.dtypes
+        expected = Series([np.dtype("int32")] * 5)
+        tm.assert_series_equal(result, expected)
+
+        # overflow issue? (we always expected int64 upcasting here)
+        df = DataFrame({"a": [2**31, 2**31 + 1]})
+        assert df.dtypes.iloc[0] == np.dtype("int64")
+
+        # GH #2751 (construction with no index specified), make sure we cast to
+        # platform values
+        df = DataFrame([1, 2])
+        assert df.dtypes.iloc[0] == np.dtype("int64")
+
+        df = DataFrame([1.0, 2.0])
+        assert df.dtypes.iloc[0] == np.dtype("float64")
+
+        df = DataFrame({"a": [1, 2]})
+        assert df.dtypes.iloc[0] == np.dtype("int64")
+
+        df = DataFrame({"a": [1.0, 2.0]})
+        assert df.dtypes.iloc[0] == np.dtype("float64")
+
+        df = DataFrame({"a": 1}, index=range(3))
+        assert df.dtypes.iloc[0] == np.dtype("int64")
+
+        df = DataFrame({"a": 1.0}, index=range(3))
+        assert df.dtypes.iloc[0] == np.dtype("float64")
+
+        # with object list
+        df = DataFrame(
+            {
+                "a": [1, 2, 4, 7],
+                "b": [1.2, 2.3, 5.1, 6.3],
+                "c": list("abcd"),
+                "d": [datetime(2000, 1, 1) for i in range(4)],
+                "e": [1.0, 2, 4.0, 7],
+            }
+        )
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("int64"),
+                np.dtype("float64"),
+                np.dtype("object")
+                if not using_infer_string
+                else pd.StringDtype(na_value=np.nan),
+                np.dtype("datetime64[us]"),
+                np.dtype("float64"),
+            ],
+            index=list("abcde"),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_frame_copy(self, float_frame):
+        cop = DataFrame(float_frame, copy=True)
+        cop["A"] = 5
+        assert (cop["A"] == 5).all()
+        assert not (float_frame["A"] == 5).all()
+
+    def test_constructor_frame_shallow_copy(self, float_frame):
+        # constructing a DataFrame from DataFrame with copy=False should still
+        # give a "shallow" copy (share data, not attributes)
+        # https://github.com/pandas-dev/pandas/issues/49523
+        orig = float_frame.copy()
+        cop = DataFrame(float_frame)
+        assert cop._mgr is not float_frame._mgr
+        # Overwriting index of copy doesn't change original
+        cop.index = np.arange(len(cop))
+        tm.assert_frame_equal(float_frame, orig)
+
+    def test_constructor_ndarray_copy(self, float_frame):
+        arr = float_frame.values.copy()
+        df = DataFrame(arr)
+
+        arr[5] = 5
+        assert not (df.values[5] == 5).all()
+        df = DataFrame(arr, copy=True)
+        arr[6] = 6
+        assert not (df.values[6] == 6).all()
+
+    def test_constructor_series_copy(self, float_frame):
+        series = float_frame._series
+
+        df = DataFrame({"A": series["A"]}, copy=True)
+        # TODO can be replaced with `df.loc[:, "A"] = 5` after deprecation about
+        # inplace mutation is enforced
+        df.loc[df.index[0] : df.index[-1], "A"] = 5
+
+        assert not (series["A"] == 5).all()
+
+    @pytest.mark.parametrize(
+        "df",
+        [
+            DataFrame([[1, 2, 3], [4, 5, 6]], index=[1, np.nan]),
+            DataFrame([[1, 2, 3], [4, 5, 6]], columns=[1.1, 2.2, np.nan]),
+            DataFrame([[0, 1, 2, 3], [4, 5, 6, 7]], columns=[np.nan, 1.1, 2.2, np.nan]),
+            DataFrame(
+                [[0.0, 1, 2, 3.0], [4, 5, 6, 7]], columns=[np.nan, 1.1, 2.2, np.nan]
+            ),
+            DataFrame([[0.0, 1, 2, 3.0], [4, 5, 6, 7]], columns=[np.nan, 1, 2, 2]),
+        ],
+    )
+    def test_constructor_with_nas(self, df):
+        # GH 5016
+        # na's in indices
+        # GH 21428 (non-unique columns)
+
+        for i in range(len(df.columns)):
+            df.iloc[:, i]
+
+        indexer = np.arange(len(df.columns))[isna(df.columns)]
+
+        # No NaN found -> error
+        if len(indexer) == 0:
+            with pytest.raises(KeyError, match="^nan$"):
+                df.loc[:, np.nan]
+        # single nan should result in Series
+        elif len(indexer) == 1:
+            tm.assert_series_equal(df.iloc[:, indexer[0]], df.loc[:, np.nan])
+        # multiple nans should result in DataFrame
+        else:
+            tm.assert_frame_equal(df.iloc[:, indexer], df.loc[:, np.nan])
+
+    def test_constructor_lists_to_object_dtype(self):
+        # from #1074
+        d = DataFrame({"a": [np.nan, False]})
+        assert d["a"].dtype == np.object_
+        assert not d["a"][1]
+
+    def test_constructor_ndarray_categorical_dtype(self):
+        cat = Categorical(["A", "B", "C"])
+        arr = np.array(cat).reshape(-1, 1)
+        arr = np.broadcast_to(arr, (3, 4))
+
+        result = DataFrame(arr, dtype=cat.dtype)
+
+        expected = DataFrame({0: cat, 1: cat, 2: cat, 3: cat})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_categorical(self):
+        # GH8626
+
+        # dict creation
+        df = DataFrame({"A": list("abc")}, dtype="category")
+        expected = Series(list("abc"), dtype="category", name="A")
+        tm.assert_series_equal(df["A"], expected)
+
+        # to_frame
+        s = Series(list("abc"), dtype="category")
+        result = s.to_frame()
+        expected = Series(list("abc"), dtype="category", name=0)
+        tm.assert_series_equal(result[0], expected)
+        result = s.to_frame(name="foo")
+        expected = Series(list("abc"), dtype="category", name="foo")
+        tm.assert_series_equal(result["foo"], expected)
+
+        # list-like creation
+        df = DataFrame(list("abc"), dtype="category")
+        expected = Series(list("abc"), dtype="category", name=0)
+        tm.assert_series_equal(df[0], expected)
+
+    def test_construct_from_1item_list_of_categorical(self):
+        # pre-2.0 this behaved as DataFrame({0: cat}), in 2.0 we remove
+        #  Categorical special case
+        # ndim != 1
+        cat = Categorical(list("abc"))
+        df = DataFrame([cat])
+        expected = DataFrame([cat.astype(object)])
+        tm.assert_frame_equal(df, expected)
+
+    def test_construct_from_list_of_categoricals(self):
+        # pre-2.0 this behaved as DataFrame({0: cat}), in 2.0 we remove
+        #  Categorical special case
+
+        df = DataFrame([Categorical(list("abc")), Categorical(list("abd"))])
+        expected = DataFrame([["a", "b", "c"], ["a", "b", "d"]])
+        tm.assert_frame_equal(df, expected)
+
+    def test_from_nested_listlike_mixed_types(self):
+        # pre-2.0 this behaved as DataFrame({0: cat}), in 2.0 we remove
+        #  Categorical special case
+        # mixed
+        df = DataFrame([Categorical(list("abc")), list("def")])
+        expected = DataFrame([["a", "b", "c"], ["d", "e", "f"]])
+        tm.assert_frame_equal(df, expected)
+
+    def test_construct_from_listlikes_mismatched_lengths(self):
+        df = DataFrame([Categorical(list("abc")), Categorical(list("abdefg"))])
+        expected = DataFrame([list("abc"), list("abdefg")])
+        tm.assert_frame_equal(df, expected)
+
+    def test_constructor_categorical_series(self):
+        items = [1, 2, 3, 1]
+        exp = Series(items).astype("category")
+        res = Series(items, dtype="category")
+        tm.assert_series_equal(res, exp)
+
+        items = ["a", "b", "c", "a"]
+        exp = Series(items).astype("category")
+        res = Series(items, dtype="category")
+        tm.assert_series_equal(res, exp)
+
+        # insert into frame with different index
+        # GH 8076
+        index = date_range("20000101", periods=3)
+        expected = Series(
+            Categorical(values=[np.nan, np.nan, np.nan], categories=["a", "b", "c"])
+        )
+        expected.index = index
+
+        expected = DataFrame({"x": expected})
+        df = DataFrame({"x": Series(["a", "b", "c"], dtype="category")}, index=index)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        tm.ALL_NUMERIC_DTYPES
+        + tm.DATETIME64_DTYPES
+        + tm.TIMEDELTA64_DTYPES
+        + tm.BOOL_DTYPES,
+    )
+    def test_check_dtype_empty_numeric_column(self, dtype):
+        # GH24386: Ensure dtypes are set correctly for an empty DataFrame.
+        # Empty DataFrame is generated via dictionary data with non-overlapping columns.
+        data = DataFrame({"a": [1, 2]}, columns=["b"], dtype=dtype)
+
+        assert data.b.dtype == dtype
+
+    @pytest.mark.parametrize(
+        "dtype", tm.STRING_DTYPES + tm.BYTES_DTYPES + tm.OBJECT_DTYPES
+    )
+    def test_check_dtype_empty_string_column(self, dtype):
+        # GH24386: Ensure dtypes are set correctly for an empty DataFrame.
+        # Empty DataFrame is generated via dictionary data with non-overlapping columns.
+        data = DataFrame({"a": [1, 2]}, columns=["b"], dtype=dtype)
+        assert data.b.dtype.name == "object"
+
+    def test_to_frame_with_falsey_names(self):
+        # GH 16114
+        result = Series(name=0, dtype=object).to_frame().dtypes
+        expected = Series({0: object})
+        tm.assert_series_equal(result, expected)
+
+        result = DataFrame(Series(name=0, dtype=object)).dtypes
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.arm_slow
+    @pytest.mark.parametrize("dtype", [None, "uint8", "category"])
+    def test_constructor_range_dtype(self, dtype):
+        expected = DataFrame({"A": [0, 1, 2, 3, 4]}, dtype=dtype or "int64")
+
+        # GH 26342
+        result = DataFrame(range(5), columns=["A"], dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        # GH 16804
+        result = DataFrame({"A": range(5)}, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_from_list_subclass(self):
+        # GH21226
+        class List(list):
+            pass
+
+        expected = DataFrame([[1, 2, 3], [4, 5, 6]])
+        result = DataFrame(List([List([1, 2, 3]), List([4, 5, 6])]))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "extension_arr",
+        [
+            Categorical(list("aabbc")),
+            SparseArray([1, np.nan, np.nan, np.nan]),
+            IntervalArray([Interval(0, 1), Interval(1, 5)]),
+            PeriodArray(pd.period_range(start="1/1/2017", end="1/1/2018", freq="M")),
+        ],
+    )
+    def test_constructor_with_extension_array(self, extension_arr):
+        # GH11363
+        expected = DataFrame(Series(extension_arr))
+        result = DataFrame(extension_arr)
+        tm.assert_frame_equal(result, expected)
+
+    def test_datetime_date_tuple_columns_from_dict(self):
+        # GH 10863
+        v = date.today()
+        tup = v, v
+        result = DataFrame({tup: Series(range(3), index=range(3))}, columns=[tup])
+        expected = DataFrame([0, 1, 2], columns=Index(Series([tup])))
+        tm.assert_frame_equal(result, expected)
+
+    def test_construct_with_two_categoricalindex_series(self):
+        # GH 14600
+        s1 = Series([39, 6, 4], index=CategoricalIndex(["female", "male", "unknown"]))
+        s2 = Series(
+            [2, 152, 2, 242, 150],
+            index=CategoricalIndex(["f", "female", "m", "male", "unknown"]),
+        )
+        result = DataFrame([s1, s2])
+        expected = DataFrame(
+            np.array([[39, 6, 4, np.nan, np.nan], [152.0, 242.0, 150.0, 2.0, 2.0]]),
+            columns=["female", "male", "unknown", "f", "m"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_constructor_series_nonexact_categoricalindex(self):
+        # GH 42424
+        ser = Series(range(100))
+        ser1 = cut(ser, 10).value_counts().head(5)
+        ser2 = cut(ser, 10).value_counts().tail(5)
+        result = DataFrame({"1": ser1, "2": ser2})
+        index = CategoricalIndex(
+            [
+                Interval(-0.099, 9.9, closed="right"),
+                Interval(9.9, 19.8, closed="right"),
+                Interval(19.8, 29.7, closed="right"),
+                Interval(29.7, 39.6, closed="right"),
+                Interval(39.6, 49.5, closed="right"),
+                Interval(49.5, 59.4, closed="right"),
+                Interval(59.4, 69.3, closed="right"),
+                Interval(69.3, 79.2, closed="right"),
+                Interval(79.2, 89.1, closed="right"),
+                Interval(89.1, 99, closed="right"),
+            ],
+            ordered=True,
+        )
+        expected = DataFrame(
+            {"1": [10] * 5 + [np.nan] * 5, "2": [np.nan] * 5 + [10] * 5}, index=index
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_from_M8_structured(self):
+        dates = [(datetime(2012, 9, 9, 0, 0), datetime(2012, 9, 8, 15, 10))]
+        arr = np.array(dates, dtype=[("Date", "M8[us]"), ("Forecasting", "M8[us]")])
+        df = DataFrame(arr)
+
+        assert df["Date"][0] == dates[0][0]
+        assert df["Forecasting"][0] == dates[0][1]
+
+        s = Series(arr["Date"])
+        assert isinstance(s[0], Timestamp)
+        assert s[0] == dates[0][0]
+
+    def test_from_datetime_subclass(self):
+        # GH21142 Verify whether Datetime subclasses are also of dtype datetime
+        class DatetimeSubclass(datetime):
+            pass
+
+        data = DataFrame({"datetime": [DatetimeSubclass(2020, 1, 1, 1, 1)]})
+        assert data.datetime.dtype == "datetime64[us]"
+
+    def test_with_mismatched_index_length_raises(self):
+        # GH#33437
+        dti = date_range("2016-01-01", periods=3, tz="US/Pacific")
+        msg = "Shape of passed values|Passed arrays should have the same length"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(dti, index=range(4))
+
+    def test_frame_ctor_datetime64_column(self):
+        rng = date_range("1/1/2000 00:00:00", "1/1/2000 1:59:50", freq="10s")
+        dates = np.asarray(rng)
+
+        df = DataFrame(
+            {"A": np.random.default_rng(2).standard_normal(len(rng)), "B": dates}
+        )
+        assert np.issubdtype(df["B"].dtype, np.dtype("M8[ns]"))
+
+    def test_dataframe_constructor_infer_multiindex(self):
+        index_lists = [["a", "a", "b", "b"], ["x", "y", "x", "y"]]
+
+        multi = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=[np.array(x) for x in index_lists],
+        )
+        assert isinstance(multi.index, MultiIndex)
+        assert not isinstance(multi.columns, MultiIndex)
+
+        multi = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), columns=index_lists
+        )
+        assert isinstance(multi.columns, MultiIndex)
+
+    @pytest.mark.parametrize(
+        "input_vals",
+        [
+            ([1, 2]),
+            (["1", "2"]),
+            (list(date_range("1/1/2011", periods=2, freq="h"))),
+            (list(date_range("1/1/2011", periods=2, freq="h", tz="US/Eastern"))),
+            ([Interval(left=0, right=5)]),
+        ],
+    )
+    def test_constructor_list_str(self, input_vals, string_dtype):
+        # GH#16605
+        # Ensure that data elements are converted to strings when
+        # dtype is str, 'str', or 'U'
+
+        result = DataFrame({"A": input_vals}, dtype=string_dtype)
+        expected = DataFrame({"A": input_vals}).astype({"A": string_dtype})
+        tm.assert_frame_equal(result, expected)
+
+    def test_constructor_list_str_na(self, string_dtype):
+        result = DataFrame({"A": [1.0, 2.0, None]}, dtype=string_dtype)
+        expected = DataFrame({"A": ["1.0", "2.0", None]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("copy", [False, True])
+    def test_dict_nocopy(
+        self,
+        copy,
+        any_numeric_ea_dtype,
+        any_numpy_dtype,
+    ):
+        a = np.array([1, 2], dtype=any_numpy_dtype)
+        b = np.array([3, 4], dtype=any_numpy_dtype)
+        if b.dtype.kind in ["S", "U"]:
+            # These get cast, making the checks below more cumbersome
+            pytest.skip(f"{b.dtype} get cast, making the checks below more cumbersome")
+
+        c = pd.array([1, 2], dtype=any_numeric_ea_dtype)
+        c_orig = c.copy()
+        df = DataFrame({"a": a, "b": b, "c": c}, copy=copy)
+
+        def get_base(obj):
+            if isinstance(obj, np.ndarray):
+                return obj.base
+            elif isinstance(obj.dtype, np.dtype):
+                # i.e. DatetimeArray, TimedeltaArray
+                return obj._ndarray.base
+            else:
+                raise TypeError
+
+        def check_views(c_only: bool = False):
+            # Check that the underlying data behind df["c"] is still `c`
+            #  after setting with iloc.  Since we don't know which entry in
+            #  df._mgr.blocks corresponds to df["c"], we just check that exactly
+            #  one of these arrays is `c`.  GH#38939
+            assert sum(x.values is c for x in df._mgr.blocks) == 1
+            if c_only:
+                # If we ever stop consolidating in setitem_with_indexer,
+                #  this will become unnecessary.
+                return
+
+            assert (
+                sum(
+                    get_base(x.values) is a
+                    for x in df._mgr.blocks
+                    if isinstance(x.values.dtype, np.dtype)
+                )
+                == 1
+            )
+            assert (
+                sum(
+                    get_base(x.values) is b
+                    for x in df._mgr.blocks
+                    if isinstance(x.values.dtype, np.dtype)
+                )
+                == 1
+            )
+
+        if not copy:
+            # constructor preserves views
+            check_views()
+
+        # TODO: most of the rest of this test belongs in indexing tests
+        should_raise = not lib.is_np_dtype(df.dtypes.iloc[0], "fciuO")
+        if should_raise:
+            with pytest.raises(TypeError, match="Invalid value"):
+                df.iloc[0, 0] = 0
+                df.iloc[0, 1] = 0
+            return
+        else:
+            df.iloc[0, 0] = 0
+            df.iloc[0, 1] = 0
+        if not copy:
+            check_views(True)
+
+        # FIXME(GH#35417): until GH#35417, iloc.setitem into EA values does not preserve
+        #  view, so we have to check in the other direction
+        df.iloc[:, 2] = pd.array([45, 46], dtype=c.dtype)
+        assert df.dtypes.iloc[2] == c.dtype
+        if copy:
+            if a.dtype.kind == "M":
+                assert a[0] == a.dtype.type(1, "ns")
+                assert b[0] == b.dtype.type(3, "ns")
+            else:
+                assert a[0] == a.dtype.type(1)
+                assert b[0] == b.dtype.type(3)
+            # FIXME(GH#35417): enable after GH#35417
+            assert c[0] == c_orig[0]  # i.e. df.iloc[0, 2]=45 did *not* update c
+
+    def test_construct_from_dict_ea_series(self):
+        # GH#53744 - default of copy=True should also apply for Series with
+        # extension dtype
+        ser = Series([1, 2, 3], dtype="Int64")
+        df = DataFrame({"a": ser})
+        assert not np.shares_memory(ser.values._data, df["a"].values._data)
+
+    def test_from_series_with_name_with_columns(self):
+        # GH 7893
+        result = DataFrame(Series(1, name="foo"), columns=["bar"])
+        expected = DataFrame(columns=["bar"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_list_columns(self):
+        # GH 14467
+        result = DataFrame(
+            [[1, 2, 3], [4, 5, 6]], columns=[["A", "A", "A"], ["a", "b", "c"]]
+        )
+        expected = DataFrame(
+            [[1, 2, 3], [4, 5, 6]],
+            columns=MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("A", "c")]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_2d_object_array_of_periods_or_intervals(self):
+        # Period analogue to GH#26825
+        pi = pd.period_range("2016-04-05", periods=3)
+        data = pi._data.astype(object).reshape(1, -1)
+        df = DataFrame(data)
+        assert df.shape == (1, 3)
+        assert (df.dtypes == pi.dtype).all()
+        assert (df == pi).all().all()
+
+        ii = pd.IntervalIndex.from_breaks([3, 4, 5, 6])
+        data2 = ii._data.astype(object).reshape(1, -1)
+        df2 = DataFrame(data2)
+        assert df2.shape == (1, 3)
+        assert (df2.dtypes == ii.dtype).all()
+        assert (df2 == ii).all().all()
+
+        # mixed
+        data3 = np.r_[data, data2, data, data2].T
+        df3 = DataFrame(data3)
+        expected = DataFrame({0: pi, 1: ii, 2: pi, 3: ii})
+        tm.assert_frame_equal(df3, expected)
+
+    @pytest.mark.parametrize(
+        "col_a, col_b",
+        [
+            ([[1], [2]], np.array([[1], [2]])),
+            (np.array([[1], [2]]), [[1], [2]]),
+            (np.array([[1], [2]]), np.array([[1], [2]])),
+        ],
+    )
+    def test_error_from_2darray(self, col_a, col_b):
+        msg = "Per-column arrays must each be 1-dimensional"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({"a": col_a, "b": col_b})
+
+    def test_from_dict_with_missing_copy_false(self):
+        # GH#45369 filled columns should not be views of one another
+        df = DataFrame(index=[1, 2, 3], columns=["a", "b", "c"], copy=False)
+        assert not np.shares_memory(df["a"]._values, df["b"]._values)
+
+        df.iloc[0, 0] = 0
+        expected = DataFrame(
+            {
+                "a": [0, np.nan, np.nan],
+                "b": [np.nan, np.nan, np.nan],
+                "c": [np.nan, np.nan, np.nan],
+            },
+            index=[1, 2, 3],
+            dtype=object,
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_construction_empty_array_multi_column_raises(self):
+        # GH#46822
+        msg = r"Shape of passed values is \(0, 1\), indices imply \(0, 2\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(data=np.array([]), columns=["a", "b"])
+
+    def test_construct_with_strings_and_none(self):
+        # GH#32218
+        df = DataFrame(["1", "2", None], columns=["a"], dtype="str")
+        expected = DataFrame({"a": ["1", "2", None]}, dtype="str")
+        tm.assert_frame_equal(df, expected)
+
+    def test_frame_string_inference(self):
+        # GH#54430
+        dtype = pd.StringDtype(na_value=np.nan)
+        expected = DataFrame(
+            {"a": ["a", "b"]}, dtype=dtype, columns=Index(["a"], dtype=dtype)
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame({"a": ["a", "b"]})
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame(
+            {"a": ["a", "b"]},
+            dtype=dtype,
+            columns=Index(["a"], dtype=dtype),
+            index=Index(["x", "y"], dtype=dtype),
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame({"a": ["a", "b"]}, index=["x", "y"])
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame(
+            {"a": ["a", 1]}, dtype="object", columns=Index(["a"], dtype=dtype)
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame({"a": ["a", 1]})
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame(
+            {"a": ["a", "b"]}, dtype="object", columns=Index(["a"], dtype=dtype)
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame({"a": ["a", "b"]}, dtype="object")
+        tm.assert_frame_equal(df, expected)
+
+    def test_frame_string_inference_array_string_dtype(self):
+        # GH#54496
+        dtype = pd.StringDtype(na_value=np.nan)
+        expected = DataFrame(
+            {"a": ["a", "b"]}, dtype=dtype, columns=Index(["a"], dtype=dtype)
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame({"a": np.array(["a", "b"])})
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame({0: ["a", "b"], 1: ["c", "d"]}, dtype=dtype)
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame(np.array([["a", "c"], ["b", "d"]]))
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame(
+            {"a": ["a", "b"], "b": ["c", "d"]},
+            dtype=dtype,
+            columns=Index(["a", "b"], dtype=dtype),
+        )
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame(np.array([["a", "c"], ["b", "d"]]), columns=["a", "b"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_frame_string_inference_block_dim(self):
+        # GH#55363
+        with pd.option_context("future.infer_string", True):
+            df = DataFrame(np.array([["hello", "goodbye"], ["hello", "Hello"]]))
+        assert df._mgr.blocks[0].ndim == 2
+
+    @pytest.mark.parametrize("klass", [Series, Index])
+    def test_inference_on_pandas_objects(self, klass):
+        # GH#56012
+        obj = klass([Timestamp("2019-12-31")], dtype=object)
+        result = DataFrame(obj, columns=["a"])
+        assert result.dtypes.iloc[0] == np.object_
+
+        result = DataFrame({"a": obj})
+        assert result.dtypes.iloc[0] == np.object_
+
+    def test_dict_keys_returns_rangeindex(self):
+        result = DataFrame({0: [1], 1: [2]}).columns
+        expected = RangeIndex(2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    @pytest.mark.parametrize(
+        "cons", [Series, Index, DatetimeIndex, DataFrame, pd.array, pd.to_datetime]
+    )
+    def test_construction_datetime_resolution_inference(self, cons):
+        ts = Timestamp(2999, 1, 1)
+        ts2 = ts.tz_localize("US/Pacific")
+
+        obj = cons([ts])
+        res_dtype = tm.get_dtype(obj)
+        assert res_dtype == "M8[us]", res_dtype
+
+        obj2 = cons([ts2])
+        res_dtype2 = tm.get_dtype(obj2)
+        assert res_dtype2 == "M8[us, US/Pacific]", res_dtype2
+
+    def test_construction_nan_value_timedelta64_dtype(self):
+        # GH#60064
+        result = DataFrame([None, 1], dtype="timedelta64[ns]")
+        expected = DataFrame(
+            ["NaT", "0 days 00:00:00.000000001"], dtype="timedelta64[ns]"
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameConstructorIndexInference:
+    def test_frame_from_dict_of_series_overlapping_monthly_period_indexes(self):
+        rng1 = pd.period_range("1/1/1999", "1/1/2012", freq="M")
+        s1 = Series(np.random.default_rng(2).standard_normal(len(rng1)), rng1)
+
+        rng2 = pd.period_range("1/1/1980", "12/1/2001", freq="M")
+        s2 = Series(np.random.default_rng(2).standard_normal(len(rng2)), rng2)
+        df = DataFrame({"s1": s1, "s2": s2})
+
+        exp = pd.period_range("1/1/1980", "1/1/2012", freq="M")
+        tm.assert_index_equal(df.index, exp)
+
+    def test_frame_from_dict_with_mixed_tzaware_indexes(self):
+        # GH#44091
+        dti = date_range("2016-01-01", periods=3)
+
+        ser1 = Series(range(3), index=dti)
+        ser2 = Series(range(3), index=dti.tz_localize("UTC"))
+        ser3 = Series(range(3), index=dti.tz_localize("US/Central"))
+        ser4 = Series(range(3))
+
+        # no tz-naive, but we do have mixed tzs and a non-DTI
+        df1 = DataFrame({"A": ser2, "B": ser3, "C": ser4})
+        exp_index = Index(
+            list(ser2.index) + list(ser3.index) + list(ser4.index), dtype=object
+        )
+        tm.assert_index_equal(df1.index, exp_index)
+
+        df2 = DataFrame({"A": ser2, "C": ser4, "B": ser3})
+        exp_index3 = Index(
+            list(ser2.index) + list(ser4.index) + list(ser3.index), dtype=object
+        )
+        tm.assert_index_equal(df2.index, exp_index3)
+
+        df3 = DataFrame({"B": ser3, "A": ser2, "C": ser4})
+        exp_index3 = Index(
+            list(ser3.index) + list(ser2.index) + list(ser4.index), dtype=object
+        )
+        tm.assert_index_equal(df3.index, exp_index3)
+
+        df4 = DataFrame({"C": ser4, "B": ser3, "A": ser2})
+        exp_index4 = Index(
+            list(ser4.index) + list(ser3.index) + list(ser2.index), dtype=object
+        )
+        tm.assert_index_equal(df4.index, exp_index4)
+
+        # TODO: not clear if these raising is desired (no extant tests),
+        #  but this is de facto behavior 2021-12-22
+        msg = "Cannot join tz-naive with tz-aware DatetimeIndex"
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"A": ser2, "B": ser3, "C": ser4, "D": ser1})
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"A": ser2, "B": ser3, "D": ser1})
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"D": ser1, "A": ser2, "B": ser3})
+
+    @pytest.mark.parametrize(
+        "key_val, col_vals, col_type",
+        [
+            ["3", ["3", "4"], "utf8"],
+            [3, [3, 4], "int8"],
+        ],
+    )
+    def test_dict_data_arrow_column_expansion(self, key_val, col_vals, col_type):
+        # GH 53617
+        pa = pytest.importorskip("pyarrow")
+        cols = pd.arrays.ArrowExtensionArray(
+            pa.array(col_vals, type=pa.dictionary(pa.int8(), getattr(pa, col_type)()))
+        )
+        result = DataFrame({key_val: [1, 2]}, columns=cols)
+        expected = DataFrame([[1, np.nan], [2, np.nan]], columns=cols)
+        expected.isetitem(1, expected.iloc[:, 1].astype(object))
+        tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameConstructorWithDtypeCoercion:
+    def test_floating_values_integer_dtype(self):
+        # GH#40110 make DataFrame behavior with arraylike floating data and
+        #  inty dtype match Series behavior
+
+        arr = np.random.default_rng(2).standard_normal((10, 5))
+
+        # GH#49599 in 2.0 we raise instead of either
+        #  a) silently ignoring dtype and returningfloat (the old Series behavior) or
+        #  b) rounding (the old DataFrame behavior)
+        msg = "Trying to coerce float values to integers"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(arr, dtype="i8")
+
+        df = DataFrame(arr.round(), dtype="i8")
+        assert (df.dtypes == "i8").all()
+
+        # with NaNs, we go through a different path with a different warning
+        arr[0, 0] = np.nan
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            DataFrame(arr, dtype="i8")
+        with pytest.raises(IntCastingNaNError, match=msg):
+            Series(arr[0], dtype="i8")
+        # The future (raising) behavior matches what we would get via astype:
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            DataFrame(arr).astype("i8")
+        with pytest.raises(IntCastingNaNError, match=msg):
+            Series(arr[0]).astype("i8")
+
+
+class TestDataFrameConstructorWithDatetimeTZ:
+    @pytest.mark.parametrize("tz", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_construction_preserves_tzaware_dtypes(self, tz):
+        # after GH#7822
+        # these retain the timezones on dict construction
+        dr = date_range("2011/1/1", "2012/1/1", freq="W-FRI")
+        dr_tz = dr.tz_localize(tz)
+        df = DataFrame({"A": "foo", "B": dr_tz}, index=dr)
+        tz_expected = DatetimeTZDtype("ns", dr_tz.tzinfo)
+        assert df["B"].dtype == tz_expected
+
+        # GH#2810 (with timezones)
+        datetimes_naive = [ts.to_pydatetime() for ts in dr]
+        datetimes_with_tz = [ts.to_pydatetime() for ts in dr_tz]
+        df = DataFrame({"dr": dr})
+        df["dr_tz"] = dr_tz
+        df["datetimes_naive"] = datetimes_naive
+        df["datetimes_with_tz"] = datetimes_with_tz
+        result = df.dtypes
+        expected = Series(
+            [
+                np.dtype("datetime64[ns]"),
+                DatetimeTZDtype(tz=tz),
+                np.dtype("datetime64[us]"),
+                DatetimeTZDtype(tz=tz, unit="us"),
+            ],
+            index=["dr", "dr_tz", "datetimes_naive", "datetimes_with_tz"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("pydt", [True, False])
+    def test_constructor_data_aware_dtype_naive(self, tz_aware_fixture, pydt):
+        # GH#25843, GH#41555, GH#33401
+        tz = tz_aware_fixture
+        ts = Timestamp("2019", tz=tz)
+        if pydt:
+            ts = ts.to_pydatetime()
+
+        msg = (
+            "Cannot convert timezone-aware data to timezone-naive dtype. "
+            r"Use pd.Series\(values\).dt.tz_localize\(None\) instead."
+        )
+        with pytest.raises(ValueError, match=msg):
+            DataFrame({0: [ts]}, dtype="datetime64[ns]")
+
+        msg2 = "Cannot unbox tzaware Timestamp to tznaive dtype"
+        with pytest.raises(TypeError, match=msg2):
+            DataFrame({0: ts}, index=[0], dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            DataFrame([ts], dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(np.array([ts], dtype=object), dtype="datetime64[ns]")
+
+        with pytest.raises(TypeError, match=msg2):
+            DataFrame(ts, index=[0], columns=[0], dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            DataFrame([Series([ts])], dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            DataFrame([[ts]], columns=[0], dtype="datetime64[ns]")
+
+    def test_from_dict(self):
+        # 8260
+        # support datetime64 with tz
+
+        idx = Index(date_range("20130101", periods=3, tz="US/Eastern"), name="foo")
+        dr = date_range("20130110", periods=3)
+
+        # construction
+        df = DataFrame({"A": idx, "B": dr})
+        assert df["A"].dtype, "M8[ns, US/Eastern"
+        assert df["A"].name == "A"
+        tm.assert_series_equal(df["A"], Series(idx, name="A"))
+        tm.assert_series_equal(df["B"], Series(dr, name="B"))
+
+    def test_from_index(self):
+        # from index
+        idx2 = date_range("20130101", periods=3, tz="US/Eastern", name="foo")
+        df2 = DataFrame(idx2)
+        tm.assert_series_equal(df2["foo"], Series(idx2, name="foo"))
+        df2 = DataFrame(Series(idx2))
+        tm.assert_series_equal(df2["foo"], Series(idx2, name="foo"))
+
+        idx2 = date_range("20130101", periods=3, tz="US/Eastern")
+        df2 = DataFrame(idx2)
+        tm.assert_series_equal(df2[0], Series(idx2, name=0))
+        df2 = DataFrame(Series(idx2))
+        tm.assert_series_equal(df2[0], Series(idx2, name=0))
+
+    def test_frame_dict_constructor_datetime64_1680(self):
+        dr = date_range("1/1/2012", periods=10)
+        s = Series(dr, index=dr)
+
+        # it works!
+        DataFrame({"a": "foo", "b": s}, index=dr)
+        DataFrame({"a": "foo", "b": s.values}, index=dr)
+
+    def test_frame_datetime64_mixed_index_ctor_1681(self):
+        dr = date_range("2011/1/1", "2012/1/1", freq="W-FRI")
+        ts = Series(dr)
+
+        # it works!
+        d = DataFrame({"A": "foo", "B": ts}, index=dr)
+        assert d["B"].isna().all()
+
+    def test_frame_timeseries_column(self):
+        # GH19157
+        dr = date_range(
+            start="20130101T10:00:00", periods=3, freq="min", tz="US/Eastern"
+        )
+        result = DataFrame(dr, columns=["timestamps"])
+        expected = DataFrame(
+            {
+                "timestamps": [
+                    Timestamp("20130101T10:00:00", tz="US/Eastern"),
+                    Timestamp("20130101T10:01:00", tz="US/Eastern"),
+                    Timestamp("20130101T10:02:00", tz="US/Eastern"),
+                ]
+            },
+            dtype="M8[ns, US/Eastern]",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_dict_construction(self):
+        # GH22227
+        columns = ["Nevada", "Ohio"]
+        pop = {
+            "Nevada": {2001: 2.4, 2002: 2.9},
+            "Ohio": {2000: 1.5, 2001: 1.7, 2002: 3.6},
+        }
+        result = DataFrame(pop, index=[2001, 2002, 2003], columns=columns)
+        expected = DataFrame(
+            [(2.4, 1.7), (2.9, 3.6), (np.nan, np.nan)],
+            columns=columns,
+            index=Index([2001, 2002, 2003]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_tzaware_object_array(self):
+        # GH#26825 2D object array of tzaware timestamps should not raise
+        dti = date_range("2016-04-05 04:30", periods=3, tz="UTC")
+        data = dti._data.astype(object).reshape(1, -1)
+        df = DataFrame(data)
+        assert df.shape == (1, 3)
+        assert (df.dtypes == dti.dtype).all()
+        assert (df == dti).all().all()
+
+    def test_from_tzaware_mixed_object_array(self):
+        # GH#26825
+        arr = np.array(
+            [
+                [
+                    Timestamp("2013-01-01 00:00:00"),
+                    Timestamp("2013-01-02 00:00:00"),
+                    Timestamp("2013-01-03 00:00:00"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern"),
+                    pd.NaT,
+                    Timestamp("2013-01-03 00:00:00-0500", tz="US/Eastern"),
+                ],
+                [
+                    Timestamp("2013-01-01 00:00:00+0100", tz="CET"),
+                    pd.NaT,
+                    Timestamp("2013-01-03 00:00:00+0100", tz="CET"),
+                ],
+            ],
+            dtype=object,
+        ).T
+        res = DataFrame(arr, columns=["A", "B", "C"])
+
+        expected_dtypes = [
+            "datetime64[s]",
+            "datetime64[s, US/Eastern]",
+            "datetime64[s, CET]",
+        ]
+        assert (res.dtypes == expected_dtypes).all()
+
+    def test_from_2d_ndarray_with_dtype(self):
+        # GH#12513
+        array_dim2 = np.arange(10).reshape((5, 2))
+        df = DataFrame(array_dim2, dtype="datetime64[ns, UTC]")
+
+        expected = DataFrame(array_dim2).astype("datetime64[ns, UTC]")
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("typ", [set, frozenset])
+    def test_construction_from_set_raises(self, typ):
+        # https://github.com/pandas-dev/pandas/issues/32582
+        values = typ({1, 2, 3})
+        msg = f"'{typ.__name__}' type is unordered"
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"a": values})
+
+        with pytest.raises(TypeError, match=msg):
+            Series(values)
+
+    def test_construction_from_ndarray_datetimelike(self):
+        # ensure the underlying arrays are properly wrapped as EA when
+        # constructed from 2D ndarray
+        arr = np.arange(0, 12, dtype="datetime64[ns]").reshape(4, 3)
+        df = DataFrame(arr)
+        assert all(isinstance(block.values, DatetimeArray) for block in df._mgr.blocks)
+
+    def test_construction_from_ndarray_with_eadtype_mismatched_columns(self):
+        arr = np.random.default_rng(2).standard_normal((10, 2))
+        dtype = pd.array([2.0]).dtype
+        msg = r"len\(arrays\) must match len\(columns\)"
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(arr, columns=["foo"], dtype=dtype)
+
+        arr2 = pd.array([2.0, 3.0, 4.0])
+        with pytest.raises(ValueError, match=msg):
+            DataFrame(arr2, columns=["foo", "bar"])
+
+    def test_columns_indexes_raise_on_sets(self):
+        # GH 47215
+        data = [[1, 2, 3], [4, 5, 6]]
+        with pytest.raises(ValueError, match="index cannot be a set"):
+            DataFrame(data, index={"a", "b"})
+        with pytest.raises(ValueError, match="columns cannot be a set"):
+            DataFrame(data, columns={"a", "b", "c"})
+
+    def test_from_dict_with_columns_na_scalar(self):
+        result = DataFrame({"a": pd.NaT}, columns=["a"], index=range(2))
+        expected = DataFrame({"a": Series([pd.NaT, pd.NaT])})
+        tm.assert_frame_equal(result, expected)
+
+    # TODO: make this not cast to object in pandas 3.0
+    @pytest.mark.skipif(
+        not np_version_gt2, reason="StringDType only available in numpy 2 and above"
+    )
+    @pytest.mark.parametrize(
+        "data",
+        [
+            {"a": ["a", "b", "c"], "b": [1.0, 2.0, 3.0], "c": ["d", "e", "f"]},
+        ],
+    )
+    def test_np_string_array_object_cast(self, data):
+        from numpy.dtypes import StringDType
+
+        data["a"] = np.array(data["a"], dtype=StringDType())
+        res = DataFrame(data)
+        assert res["a"].dtype == np.object_
+        assert (res["a"] == data["a"]).all()
+
+
+def get1(obj):  # TODO: make a helper in tm?
+    if isinstance(obj, Series):
+        return obj.iloc[0]
+    else:
+        return obj.iloc[0, 0]
+
+
+class TestFromScalar:
+    @pytest.fixture(params=[list, dict, None])
+    def box(self, request):
+        return request.param
+
+    @pytest.fixture
+    def constructor(self, frame_or_series, box):
+        extra = {"index": range(2)}
+        if frame_or_series is DataFrame:
+            extra["columns"] = ["A"]
+
+        if box is None:
+            return functools.partial(frame_or_series, **extra)
+
+        elif box is dict:
+            if frame_or_series is Series:
+                return lambda x, **kwargs: frame_or_series(
+                    {0: x, 1: x}, **extra, **kwargs
+                )
+            else:
+                return lambda x, **kwargs: frame_or_series({"A": x}, **extra, **kwargs)
+        elif frame_or_series is Series:
+            return lambda x, **kwargs: frame_or_series([x, x], **extra, **kwargs)
+        else:
+            return lambda x, **kwargs: frame_or_series({"A": [x, x]}, **extra, **kwargs)
+
+    @pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+    def test_from_nat_scalar(self, dtype, constructor):
+        obj = constructor(pd.NaT, dtype=dtype)
+        assert np.all(obj.dtypes == dtype)
+        assert np.all(obj.isna())
+
+    def test_from_timedelta_scalar_preserves_nanos(self, constructor):
+        td = Timedelta(1)
+
+        obj = constructor(td, dtype="m8[ns]")
+        assert get1(obj) == td
+
+    def test_from_timestamp_scalar_preserves_nanos(self, constructor, fixed_now_ts):
+        ts = fixed_now_ts + Timedelta(1)
+
+        obj = constructor(ts, dtype="M8[ns]")
+        assert get1(obj) == ts
+
+    def test_from_timedelta64_scalar_object(self, constructor):
+        td = Timedelta(1)
+        td64 = td.to_timedelta64()
+
+        obj = constructor(td64, dtype=object)
+        assert isinstance(get1(obj), np.timedelta64)
+
+    @pytest.mark.parametrize("cls", [np.datetime64, np.timedelta64])
+    def test_from_scalar_datetimelike_mismatched(self, constructor, cls):
+        scalar = cls("NaT", "ns")
+        dtype = {np.datetime64: "m8[ns]", np.timedelta64: "M8[ns]"}[cls]
+
+        if cls is np.datetime64:
+            msg1 = "Invalid type for timedelta scalar: <class 'numpy.datetime64'>"
+        else:
+            msg1 = "<class 'numpy.timedelta64'> is not convertible to datetime"
+        msg = "|".join(["Cannot cast", msg1])
+
+        with pytest.raises(TypeError, match=msg):
+            constructor(scalar, dtype=dtype)
+
+        scalar = cls(4, "ns")
+        with pytest.raises(TypeError, match=msg):
+            constructor(scalar, dtype=dtype)
+
+    @pytest.mark.parametrize("cls", [datetime, np.datetime64])
+    def test_from_out_of_bounds_ns_datetime(
+        self, constructor, cls, request, box, frame_or_series
+    ):
+        # scalar that won't fit in nanosecond dt64, but will fit in microsecond
+        scalar = datetime(9999, 1, 1)
+        exp_dtype = "M8[us]"  # pydatetime objects default to this reso
+
+        if cls is np.datetime64:
+            scalar = np.datetime64(scalar, "D")
+            exp_dtype = "M8[s]"  # closest reso to input
+        result = constructor(scalar)
+
+        item = get1(result)
+        dtype = tm.get_dtype(result)
+
+        assert type(item) is Timestamp
+        assert item.asm8.dtype == exp_dtype
+        assert dtype == exp_dtype
+
+    @pytest.mark.skip_ubsan
+    def test_out_of_s_bounds_datetime64(self, constructor):
+        scalar = np.datetime64(np.iinfo(np.int64).max, "D")
+        result = constructor(scalar)
+        item = get1(result)
+        assert type(item) is np.datetime64
+        dtype = tm.get_dtype(result)
+        assert dtype == object
+
+    @pytest.mark.parametrize("cls", [timedelta, np.timedelta64])
+    def test_from_out_of_bounds_ns_timedelta(
+        self, constructor, cls, request, box, frame_or_series
+    ):
+        # scalar that won't fit in nanosecond td64, but will fit in microsecond
+        if box is list or (frame_or_series is Series and box is dict):
+            mark = pytest.mark.xfail(
+                reason="TimedeltaArray constructor has been updated to cast td64 "
+                "to non-nano, but TimedeltaArray._from_sequence has not",
+                strict=True,
+            )
+            request.applymarker(mark)
+
+        scalar = datetime(9999, 1, 1) - datetime(1970, 1, 1)
+        exp_dtype = "m8[us]"  # smallest reso that fits
+        if cls is np.timedelta64:
+            scalar = np.timedelta64(scalar, "D")
+            exp_dtype = "m8[s]"  # closest reso to input
+        result = constructor(scalar)
+
+        item = get1(result)
+        dtype = tm.get_dtype(result)
+
+        assert type(item) is Timedelta
+        assert item.asm8.dtype == exp_dtype
+        assert dtype == exp_dtype
+
+    @pytest.mark.skip_ubsan
+    @pytest.mark.parametrize("cls", [np.datetime64, np.timedelta64])
+    def test_out_of_s_bounds_timedelta64(self, constructor, cls):
+        scalar = cls(np.iinfo(np.int64).max, "D")
+        result = constructor(scalar)
+        item = get1(result)
+        assert type(item) is cls
+        dtype = tm.get_dtype(result)
+        assert dtype == object
+
+    def test_tzaware_data_tznaive_dtype(self, constructor, box, frame_or_series):
+        tz = "US/Eastern"
+        ts = Timestamp("2019", tz=tz)
+
+        if box is None or (frame_or_series is DataFrame and box is dict):
+            msg = "Cannot unbox tzaware Timestamp to tznaive dtype"
+            err = TypeError
+        else:
+            msg = (
+                "Cannot convert timezone-aware data to timezone-naive dtype. "
+                r"Use pd.Series\(values\).dt.tz_localize\(None\) instead."
+            )
+            err = ValueError
+
+        with pytest.raises(err, match=msg):
+            constructor(ts, dtype="M8[ns]")
+
+
+# TODO: better location for this test?
+class TestAllowNonNano:
+    # Until 2.0, we do not preserve non-nano dt64/td64 when passed as ndarray,
+    #  but do preserve it when passed as DTA/TDA
+
+    @pytest.fixture(params=[True, False])
+    def as_td(self, request):
+        return request.param
+
+    @pytest.fixture
+    def arr(self, as_td):
+        values = np.arange(5).astype(np.int64).view("M8[s]")
+        if as_td:
+            values = values - values[0]
+            return TimedeltaArray._simple_new(values, dtype=values.dtype)
+        else:
+            return DatetimeArray._simple_new(values, dtype=values.dtype)
+
+    def test_index_allow_non_nano(self, arr):
+        idx = Index(arr)
+        assert idx.dtype == arr.dtype
+
+    def test_dti_tdi_allow_non_nano(self, arr, as_td):
+        if as_td:
+            idx = pd.TimedeltaIndex(arr)
+        else:
+            idx = DatetimeIndex(arr)
+        assert idx.dtype == arr.dtype
+
+    def test_series_allow_non_nano(self, arr):
+        ser = Series(arr)
+        assert ser.dtype == arr.dtype
+
+    def test_frame_allow_non_nano(self, arr):
+        df = DataFrame(arr)
+        assert df.dtypes[0] == arr.dtype
+
+    def test_frame_from_dict_allow_non_nano(self, arr):
+        df = DataFrame({0: arr})
+        assert df.dtypes[0] == arr.dtype
diff --git a/pandas/pandas/tests/frame/test_cumulative.py b/pandas/pandas/tests/frame/test_cumulative.py
new file mode 100644
index 0000000000000..ab217e1b1332a
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_cumulative.py
@@ -0,0 +1,107 @@
+"""
+Tests for DataFrame cumulative operations
+
+See also
+--------
+tests.series.test_cumulative
+"""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameCumulativeOps:
+    # ---------------------------------------------------------------------
+    # Cumulative Operations - cumsum, cummax, ...
+
+    def test_cumulative_ops_smoke(self):
+        # it works
+        df = DataFrame({"A": np.arange(20)}, index=np.arange(20))
+        df.cummax()
+        df.cummin()
+        df.cumsum()
+
+        dm = DataFrame(np.arange(20).reshape(4, 5), index=range(4), columns=range(5))
+        # TODO(wesm): do something with this?
+        dm.cumsum()
+
+    def test_cumprod_smoke(self, datetime_frame):
+        datetime_frame.iloc[5:10, 0] = np.nan
+        datetime_frame.iloc[10:15, 1] = np.nan
+        datetime_frame.iloc[15:, 2] = np.nan
+
+        # ints
+        df = datetime_frame.fillna(0).astype(int)
+        df.cumprod(0)
+        df.cumprod(1)
+
+        # ints32
+        df = datetime_frame.fillna(0).astype(np.int32)
+        df.cumprod(0)
+        df.cumprod(1)
+
+    def test_cumulative_ops_match_series_apply(
+        self, datetime_frame, all_numeric_accumulations
+    ):
+        datetime_frame.iloc[5:10, 0] = np.nan
+        datetime_frame.iloc[10:15, 1] = np.nan
+        datetime_frame.iloc[15:, 2] = np.nan
+
+        # axis = 0
+        result = getattr(datetime_frame, all_numeric_accumulations)()
+        expected = datetime_frame.apply(getattr(Series, all_numeric_accumulations))
+        tm.assert_frame_equal(result, expected)
+
+        # axis = 1
+        result = getattr(datetime_frame, all_numeric_accumulations)(axis=1)
+        expected = datetime_frame.apply(
+            getattr(Series, all_numeric_accumulations), axis=1
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # fix issue TODO: GH ref?
+        assert np.shape(result) == np.shape(datetime_frame)
+
+    def test_cumsum_preserve_dtypes(self):
+        # GH#19296 dont incorrectly upcast to object
+        df = DataFrame({"A": [1, 2, 3], "B": [1, 2, 3.0], "C": [True, False, False]})
+
+        result = df.cumsum()
+
+        expected = DataFrame(
+            {
+                "A": Series([1, 3, 6], dtype=np.int64),
+                "B": Series([1, 3, 6], dtype=np.float64),
+                "C": df["C"].cumsum(),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["cumsum", "cumprod", "cummin", "cummax"])
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_numeric_only_flag(self, method, axis):
+        df = DataFrame(
+            {
+                "int": [1, 2, 3],
+                "bool": [True, False, False],
+                "string": ["a", "b", "c"],
+                "float": [1.0, 3.5, 4.0],
+                "datetime": [
+                    Timestamp(2018, 1, 1),
+                    Timestamp(2019, 1, 1),
+                    Timestamp(2020, 1, 1),
+                ],
+            }
+        )
+        df_numeric_only = df.drop(["string", "datetime"], axis=1)
+
+        result = getattr(df, method)(axis=axis, numeric_only=True)
+        expected = getattr(df_numeric_only, method)(axis)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_iteration.py b/pandas/pandas/tests/frame/test_iteration.py
new file mode 100644
index 0000000000000..a1c23ff05f3e1
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_iteration.py
@@ -0,0 +1,160 @@
+import datetime
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestIteration:
+    def test_keys(self, float_frame):
+        assert float_frame.keys() is float_frame.columns
+
+    def test_iteritems(self):
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["a", "a", "b"])
+        for k, v in df.items():
+            assert isinstance(v, DataFrame._constructor_sliced)
+
+    def test_items(self):
+        # GH#17213, GH#13918
+        cols = ["a", "b", "c"]
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=cols)
+        for c, (k, v) in zip(cols, df.items()):
+            assert c == k
+            assert isinstance(v, Series)
+            assert (df[k] == v).all()
+
+    def test_items_names(self, float_string_frame):
+        for k, v in float_string_frame.items():
+            assert v.name == k
+
+    def test_iter(self, float_frame):
+        assert list(float_frame) == list(float_frame.columns)
+
+    def test_iterrows(self, float_frame, float_string_frame):
+        for k, v in float_frame.iterrows():
+            exp = float_frame.loc[k]
+            tm.assert_series_equal(v, exp)
+
+        for k, v in float_string_frame.iterrows():
+            exp = float_string_frame.loc[k]
+            tm.assert_series_equal(v, exp)
+
+    def test_iterrows_iso8601(self):
+        # GH#19671
+        s = DataFrame(
+            {
+                "non_iso8601": ["M1701", "M1802", "M1903", "M2004"],
+                "iso8601": date_range("2000-01-01", periods=4, freq="ME"),
+            }
+        )
+        for k, v in s.iterrows():
+            exp = s.loc[k]
+            tm.assert_series_equal(v, exp)
+
+    def test_iterrows_corner(self):
+        # GH#12222
+        df = DataFrame(
+            {
+                "a": [datetime.datetime(2015, 1, 1)],
+                "b": [None],
+                "c": [None],
+                "d": [""],
+                "e": [[]],
+                "f": [set()],
+                "g": [{}],
+            }
+        )
+        expected = Series(
+            [datetime.datetime(2015, 1, 1), None, None, "", [], set(), {}],
+            index=list("abcdefg"),
+            name=0,
+            dtype="object",
+        )
+        _, result = next(df.iterrows())
+        tm.assert_series_equal(result, expected)
+
+    def test_itertuples(self, float_frame):
+        for i, tup in enumerate(float_frame.itertuples()):
+            ser = DataFrame._constructor_sliced(tup[1:])
+            ser.name = tup[0]
+            expected = float_frame.iloc[i, :].reset_index(drop=True)
+            tm.assert_series_equal(ser, expected)
+
+    def test_itertuples_index_false(self):
+        df = DataFrame(
+            {"floats": np.random.default_rng(2).standard_normal(5), "ints": range(5)},
+            columns=["floats", "ints"],
+        )
+
+        for tup in df.itertuples(index=False):
+            assert isinstance(tup[1], int)
+
+    def test_itertuples_duplicate_cols(self):
+        df = DataFrame(data={"a": [1, 2, 3], "b": [4, 5, 6]})
+        dfaa = df[["a", "a"]]
+
+        assert list(dfaa.itertuples()) == [(0, 1, 1), (1, 2, 2), (2, 3, 3)]
+
+        # repr with int on 32-bit/windows
+        if not (is_platform_windows() or not IS64):
+            assert (
+                repr(list(df.itertuples(name=None)))
+                == "[(0, 1, 4), (1, 2, 5), (2, 3, 6)]"
+            )
+
+    def test_itertuples_tuple_name(self):
+        df = DataFrame(data={"a": [1, 2, 3], "b": [4, 5, 6]})
+        tup = next(df.itertuples(name="TestName"))
+        assert tup._fields == ("Index", "a", "b")
+        assert (tup.Index, tup.a, tup.b) == tup
+        assert type(tup).__name__ == "TestName"
+
+    def test_itertuples_disallowed_col_labels(self):
+        df = DataFrame(data={"def": [1, 2, 3], "return": [4, 5, 6]})
+        tup2 = next(df.itertuples(name="TestName"))
+        assert tup2 == (0, 1, 4)
+        assert tup2._fields == ("Index", "_1", "_2")
+
+    @pytest.mark.parametrize("limit", [254, 255, 1024])
+    @pytest.mark.parametrize("index", [True, False])
+    def test_itertuples_py2_3_field_limit_namedtuple(self, limit, index):
+        # GH#28282
+        df = DataFrame([{f"foo_{i}": f"bar_{i}" for i in range(limit)}])
+        result = next(df.itertuples(index=index))
+        assert isinstance(result, tuple)
+        assert hasattr(result, "_fields")
+
+    def test_sequence_like_with_categorical(self):
+        # GH#7839
+        # make sure can iterate
+        df = DataFrame(
+            {"id": [1, 2, 3, 4, 5, 6], "raw_grade": ["a", "b", "b", "a", "a", "e"]}
+        )
+        df["grade"] = Categorical(df["raw_grade"])
+
+        # basic sequencing testing
+        result = list(df.grade.values)
+        expected = np.array(df.grade.values).tolist()
+        tm.assert_almost_equal(result, expected)
+
+        # iteration
+        for t in df.itertuples(index=False):
+            str(t)
+
+        for row, s in df.iterrows():
+            str(s)
+
+        for c, col in df.items():
+            str(col)
diff --git a/pandas/pandas/tests/frame/test_logical_ops.py b/pandas/pandas/tests/frame/test_logical_ops.py
new file mode 100644
index 0000000000000..fb43578744eb2
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_logical_ops.py
@@ -0,0 +1,211 @@
+import operator
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Interval,
+    Series,
+    isnull,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameLogicalOperators:
+    # &, |, ^
+
+    @pytest.mark.parametrize(
+        "left, right, op, expected",
+        [
+            (
+                [True, False, np.nan],
+                [True, False, True],
+                operator.and_,
+                [True, False, False],
+            ),
+            (
+                [True, False, True],
+                [True, False, np.nan],
+                operator.and_,
+                [True, False, False],
+            ),
+            (
+                [True, False, np.nan],
+                [True, False, True],
+                operator.or_,
+                [True, False, False],
+            ),
+            (
+                [True, False, True],
+                [True, False, np.nan],
+                operator.or_,
+                [True, False, True],
+            ),
+        ],
+    )
+    def test_logical_operators_nans(self, left, right, op, expected, frame_or_series):
+        # GH#13896
+        result = op(frame_or_series(left), frame_or_series(right))
+        expected = frame_or_series(expected)
+
+        tm.assert_equal(result, expected)
+
+    def test_logical_ops_empty_frame(self):
+        # GH#5808
+        # empty frames, non-mixed dtype
+        df = DataFrame(index=[1])
+
+        result = df & df
+        tm.assert_frame_equal(result, df)
+
+        result = df | df
+        tm.assert_frame_equal(result, df)
+
+        df2 = DataFrame(index=[1, 2])
+        result = df & df2
+        tm.assert_frame_equal(result, df2)
+
+        dfa = DataFrame(index=[1], columns=["A"])
+
+        result = dfa & dfa
+        expected = DataFrame(False, index=[1], columns=["A"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_logical_ops_bool_frame(self):
+        # GH#5808
+        df1a_bool = DataFrame(True, index=[1], columns=["A"])
+
+        result = df1a_bool & df1a_bool
+        tm.assert_frame_equal(result, df1a_bool)
+
+        result = df1a_bool | df1a_bool
+        tm.assert_frame_equal(result, df1a_bool)
+
+    def test_logical_ops_int_frame(self):
+        # GH#5808
+        df1a_int = DataFrame(1, index=[1], columns=["A"])
+        df1a_bool = DataFrame(True, index=[1], columns=["A"])
+
+        result = df1a_int | df1a_bool
+        tm.assert_frame_equal(result, df1a_bool)
+
+        # Check that this matches Series behavior
+        res_ser = df1a_int["A"] | df1a_bool["A"]
+        tm.assert_series_equal(res_ser, df1a_bool["A"])
+
+    def test_logical_ops_invalid(self, using_infer_string):
+        # GH#5808
+
+        df1 = DataFrame(1.0, index=[1], columns=["A"])
+        df2 = DataFrame(True, index=[1], columns=["A"])
+        msg = re.escape("unsupported operand type(s) for |: 'float' and 'bool'")
+        with pytest.raises(TypeError, match=msg):
+            df1 | df2
+
+        df1 = DataFrame("foo", index=[1], columns=["A"])
+        df2 = DataFrame(True, index=[1], columns=["A"])
+        if using_infer_string and df1["A"].dtype.storage == "pyarrow":
+            msg = "operation 'or_' not supported for dtype 'str'"
+        else:
+            msg = re.escape("unsupported operand type(s) for |: 'str' and 'bool'")
+        with pytest.raises(TypeError, match=msg):
+            df1 | df2
+
+    def test_logical_operators(self):
+        def _check_bin_op(op):
+            result = op(df1, df2)
+            expected = DataFrame(
+                op(df1.values, df2.values), index=df1.index, columns=df1.columns
+            )
+            assert result.values.dtype == np.bool_
+            tm.assert_frame_equal(result, expected)
+
+        def _check_unary_op(op):
+            result = op(df1)
+            expected = DataFrame(op(df1.values), index=df1.index, columns=df1.columns)
+            assert result.values.dtype == np.bool_
+            tm.assert_frame_equal(result, expected)
+
+        df1 = {
+            "a": {"a": True, "b": False, "c": False, "d": True, "e": True},
+            "b": {"a": False, "b": True, "c": False, "d": False, "e": False},
+            "c": {"a": False, "b": False, "c": True, "d": False, "e": False},
+            "d": {"a": True, "b": False, "c": False, "d": True, "e": True},
+            "e": {"a": True, "b": False, "c": False, "d": True, "e": True},
+        }
+
+        df2 = {
+            "a": {"a": True, "b": False, "c": True, "d": False, "e": False},
+            "b": {"a": False, "b": True, "c": False, "d": False, "e": False},
+            "c": {"a": True, "b": False, "c": True, "d": False, "e": False},
+            "d": {"a": False, "b": False, "c": False, "d": True, "e": False},
+            "e": {"a": False, "b": False, "c": False, "d": False, "e": True},
+        }
+
+        df1 = DataFrame(df1)
+        df2 = DataFrame(df2)
+
+        _check_bin_op(operator.and_)
+        _check_bin_op(operator.or_)
+        _check_bin_op(operator.xor)
+
+        _check_unary_op(operator.inv)  # TODO: belongs elsewhere
+
+    def test_logical_with_nas(self):
+        d = DataFrame({"a": [np.nan, False], "b": [True, True]})
+
+        # GH4947
+        # bool comparisons should return bool
+        result = d["a"] | d["b"]
+        expected = Series([False, True])
+        tm.assert_series_equal(result, expected)
+
+        # GH4604, automatic casting here
+        result = d["a"].fillna(False) | d["b"]
+        expected = Series([True, True])
+        tm.assert_series_equal(result, expected)
+        result = d["a"].fillna(False) | d["b"]
+        expected = Series([True, True])
+        tm.assert_series_equal(result, expected)
+
+    def test_logical_ops_categorical_columns(self):
+        # GH#38367
+        intervals = [Interval(1, 2), Interval(3, 4)]
+        data = DataFrame(
+            [[1, np.nan], [2, np.nan]],
+            columns=CategoricalIndex(
+                intervals, categories=intervals + [Interval(5, 6)]
+            ),
+        )
+        mask = DataFrame(
+            [[False, False], [False, False]], columns=data.columns, dtype=bool
+        )
+        result = mask | isnull(data)
+        expected = DataFrame(
+            [[False, True], [False, True]],
+            columns=CategoricalIndex(
+                intervals, categories=intervals + [Interval(5, 6)]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_int_dtype_different_index_not_bool(self):
+        # GH 52500
+        df1 = DataFrame([1, 2, 3], index=[10, 11, 23], columns=["a"])
+        df2 = DataFrame([10, 20, 30], index=[11, 10, 23], columns=["a"])
+        result = np.bitwise_xor(df1, df2)
+        expected = DataFrame([21, 8, 29], index=[10, 11, 23], columns=["a"])
+        tm.assert_frame_equal(result, expected)
+
+        result = df1 ^ df2
+        tm.assert_frame_equal(result, expected)
+
+    def test_different_dtypes_different_index_raises(self):
+        # GH 52538
+        df1 = DataFrame([1, 2], index=["a", "b"])
+        df2 = DataFrame([3, 4], index=["b", "c"])
+        with pytest.raises(TypeError, match="unsupported operand type"):
+            df1 & df2
diff --git a/pandas/pandas/tests/frame/test_nonunique_indexes.py b/pandas/pandas/tests/frame/test_nonunique_indexes.py
new file mode 100644
index 0000000000000..1e9aa2325e880
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_nonunique_indexes.py
@@ -0,0 +1,336 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameNonuniqueIndexes:
+    def test_setattr_columns_vs_construct_with_columns(self):
+        # assignment
+        # GH 3687
+        arr = np.random.default_rng(2).standard_normal((3, 2))
+        idx = list(range(2))
+        df = DataFrame(arr, columns=["A", "A"])
+        df.columns = idx
+        expected = DataFrame(arr, columns=idx)
+        tm.assert_frame_equal(df, expected)
+
+    def test_setattr_columns_vs_construct_with_columns_datetimeindx(self):
+        idx = date_range("20130101", periods=4, freq="QE-NOV")
+        df = DataFrame(
+            [[1, 1, 1, 5], [1, 1, 2, 5], [2, 1, 3, 5]], columns=["a", "a", "a", "a"]
+        )
+        df.columns = idx
+        expected = DataFrame([[1, 1, 1, 5], [1, 1, 2, 5], [2, 1, 3, 5]], columns=idx)
+        tm.assert_frame_equal(df, expected)
+
+    def test_insert_with_duplicate_columns(self):
+        # insert
+        df = DataFrame(
+            [[1, 1, 1, 5], [1, 1, 2, 5], [2, 1, 3, 5]],
+            columns=["foo", "bar", "foo", "hello"],
+        )
+        df["string"] = "bah"
+        expected = DataFrame(
+            [[1, 1, 1, 5, "bah"], [1, 1, 2, 5, "bah"], [2, 1, 3, 5, "bah"]],
+            columns=["foo", "bar", "foo", "hello", "string"],
+        )
+        tm.assert_frame_equal(df, expected)
+        with pytest.raises(ValueError, match="Length of value"):
+            df.insert(0, "AnotherColumn", range(len(df.index) - 1))
+
+        # insert same dtype
+        df["foo2"] = 3
+        expected = DataFrame(
+            [[1, 1, 1, 5, "bah", 3], [1, 1, 2, 5, "bah", 3], [2, 1, 3, 5, "bah", 3]],
+            columns=["foo", "bar", "foo", "hello", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # set (non-dup)
+        df["foo2"] = 4
+        expected = DataFrame(
+            [[1, 1, 1, 5, "bah", 4], [1, 1, 2, 5, "bah", 4], [2, 1, 3, 5, "bah", 4]],
+            columns=["foo", "bar", "foo", "hello", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+        df["foo2"] = 3
+
+        # delete (non dup)
+        del df["bar"]
+        expected = DataFrame(
+            [[1, 1, 5, "bah", 3], [1, 2, 5, "bah", 3], [2, 3, 5, "bah", 3]],
+            columns=["foo", "foo", "hello", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # try to delete again (its not consolidated)
+        del df["hello"]
+        expected = DataFrame(
+            [[1, 1, "bah", 3], [1, 2, "bah", 3], [2, 3, "bah", 3]],
+            columns=["foo", "foo", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # consolidate
+        df = df._consolidate()
+        expected = DataFrame(
+            [[1, 1, "bah", 3], [1, 2, "bah", 3], [2, 3, "bah", 3]],
+            columns=["foo", "foo", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # insert
+        df.insert(2, "new_col", 5.0)
+        expected = DataFrame(
+            [[1, 1, 5.0, "bah", 3], [1, 2, 5.0, "bah", 3], [2, 3, 5.0, "bah", 3]],
+            columns=["foo", "foo", "new_col", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # insert a dup
+        with pytest.raises(ValueError, match="cannot insert"):
+            df.insert(2, "new_col", 4.0)
+
+        df.insert(2, "new_col", 4.0, allow_duplicates=True)
+        expected = DataFrame(
+            [
+                [1, 1, 4.0, 5.0, "bah", 3],
+                [1, 2, 4.0, 5.0, "bah", 3],
+                [2, 3, 4.0, 5.0, "bah", 3],
+            ],
+            columns=["foo", "foo", "new_col", "new_col", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # delete (dup)
+        del df["foo"]
+        expected = DataFrame(
+            [[4.0, 5.0, "bah", 3], [4.0, 5.0, "bah", 3], [4.0, 5.0, "bah", 3]],
+            columns=["new_col", "new_col", "string", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_dup_across_dtypes(self):
+        # dup across dtypes
+        df = DataFrame(
+            [[1, 1, 1.0, 5], [1, 1, 2.0, 5], [2, 1, 3.0, 5]],
+            columns=["foo", "bar", "foo", "hello"],
+        )
+
+        df["foo2"] = 7.0
+        expected = DataFrame(
+            [[1, 1, 1.0, 5, 7.0], [1, 1, 2.0, 5, 7.0], [2, 1, 3.0, 5, 7.0]],
+            columns=["foo", "bar", "foo", "hello", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        result = df["foo"]
+        expected = DataFrame([[1, 1.0], [1, 2.0], [2, 3.0]], columns=["foo", "foo"])
+        tm.assert_frame_equal(result, expected)
+
+        # multiple replacements
+        df["foo"] = "string"
+        expected = DataFrame(
+            [
+                ["string", 1, "string", 5, 7.0],
+                ["string", 1, "string", 5, 7.0],
+                ["string", 1, "string", 5, 7.0],
+            ],
+            columns=["foo", "bar", "foo", "hello", "foo2"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+        del df["foo"]
+        expected = DataFrame(
+            [[1, 5, 7.0], [1, 5, 7.0], [1, 5, 7.0]], columns=["bar", "hello", "foo2"]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_column_dups_indexes(self):
+        # check column dups with index equal and not equal to df's index
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=["a", "b", "c", "d", "e"],
+            columns=["A", "B", "A"],
+        )
+        for index in [df.index, pd.Index(list("edcba"))]:
+            this_df = df.copy()
+            expected_ser = Series(index.values, index=this_df.index)
+            expected_df = DataFrame(
+                {"A": expected_ser, "B": this_df["B"]},
+                columns=["A", "B", "A"],
+            )
+            this_df["A"] = index
+            tm.assert_frame_equal(this_df, expected_df)
+
+    def test_changing_dtypes_with_duplicate_columns(self):
+        # multiple assignments that change dtypes
+        # the location indexer is a slice
+        # GH 6120
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=["that", "that"]
+        )
+        expected = DataFrame(1.0, index=range(5), columns=["that", "that"])
+
+        df["that"] = 1.0
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(
+            np.random.default_rng(2).random((5, 2)), columns=["that", "that"]
+        )
+        expected = DataFrame(1, index=range(5), columns=["that", "that"])
+
+        df["that"] = 1
+        tm.assert_frame_equal(df, expected)
+
+    def test_dup_columns_comparisons(self):
+        # equality
+        df1 = DataFrame([[1, 2], [2, np.nan], [3, 4], [4, 4]], columns=["A", "B"])
+        df2 = DataFrame([[0, 1], [2, 4], [2, np.nan], [4, 5]], columns=["A", "A"])
+
+        # not-comparing like-labelled
+        msg = (
+            r"Can only compare identically-labeled \(both index and columns\) "
+            "DataFrame objects"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df1 == df2
+
+        df1r = df1.reindex_like(df2)
+        result = df1r == df2
+        expected = DataFrame(
+            [[False, True], [True, False], [False, False], [True, False]],
+            columns=["A", "A"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_mixed_column_selection(self):
+        # mixed column selection
+        # GH 5639
+        dfbool = DataFrame(
+            {
+                "one": Series([True, True, False], index=["a", "b", "c"]),
+                "two": Series([False, False, True, False], index=["a", "b", "c", "d"]),
+                "three": Series([False, True, True, True], index=["a", "b", "c", "d"]),
+            }
+        )
+        expected = pd.concat([dfbool["one"], dfbool["three"], dfbool["one"]], axis=1)
+        result = dfbool[["one", "three", "one"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_multi_axis_dups(self):
+        # multi-axis dups
+        # GH 6121
+        df = DataFrame(
+            np.arange(25.0).reshape(5, 5),
+            index=["a", "b", "c", "d", "e"],
+            columns=["A", "B", "C", "D", "E"],
+        )
+        z = df[["A", "C", "A"]].copy()
+        expected = z.loc[["a", "c", "a"]]
+
+        df = DataFrame(
+            np.arange(25.0).reshape(5, 5),
+            index=["a", "b", "c", "d", "e"],
+            columns=["A", "B", "C", "D", "E"],
+        )
+        z = df[["A", "C", "A"]]
+        result = z.loc[["a", "c", "a"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_columns_with_dups(self):
+        # GH 3468 related
+
+        # basic
+        df = DataFrame([[1, 2]], columns=["a", "a"])
+        df.columns = ["a", "a.1"]
+        expected = DataFrame([[1, 2]], columns=["a", "a.1"])
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame([[1, 2, 3]], columns=["b", "a", "a"])
+        df.columns = ["b", "a", "a.1"]
+        expected = DataFrame([[1, 2, 3]], columns=["b", "a", "a.1"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_columns_with_dup_index(self):
+        # with a dup index
+        df = DataFrame([[1, 2]], columns=["a", "a"])
+        df.columns = ["b", "b"]
+        expected = DataFrame([[1, 2]], columns=["b", "b"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_multi_dtype(self):
+        # multi-dtype
+        df = DataFrame(
+            [[1, 2, 1.0, 2.0, 3.0, "foo", "bar"]],
+            columns=["a", "a", "b", "b", "d", "c", "c"],
+        )
+        df.columns = list("ABCDEFG")
+        expected = DataFrame(
+            [[1, 2, 1.0, 2.0, 3.0, "foo", "bar"]], columns=list("ABCDEFG")
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_multi_dtype2(self):
+        df = DataFrame([[1, 2, "foo", "bar"]], columns=["a", "a", "a", "a"])
+        df.columns = ["a", "a.1", "a.2", "a.3"]
+        expected = DataFrame([[1, 2, "foo", "bar"]], columns=["a", "a.1", "a.2", "a.3"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_dups_across_blocks(self):
+        # dups across blocks
+        df_float = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), dtype="float64"
+        )
+        df_int = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)).astype("int64")
+        )
+        df_bool = DataFrame(True, index=df_float.index, columns=df_float.columns)
+        df_object = DataFrame("foo", index=df_float.index, columns=df_float.columns)
+        df_dt = DataFrame(
+            pd.Timestamp("20010101"), index=df_float.index, columns=df_float.columns
+        )
+        df = pd.concat([df_float, df_int, df_bool, df_object, df_dt], axis=1)
+
+        assert len(df._mgr.blknos) == len(df.columns)
+        assert len(df._mgr.blklocs) == len(df.columns)
+
+        # testing iloc
+        for i in range(len(df.columns)):
+            df.iloc[:, i]
+
+    def test_dup_columns_across_dtype(self):
+        # dup columns across dtype GH 2079/2194
+        vals = [[1, -1, 2.0], [2, -2, 3.0]]
+        rs = DataFrame(vals, columns=["A", "A", "B"])
+        xp = DataFrame(vals)
+        xp.columns = ["A", "A", "B"]
+        tm.assert_frame_equal(rs, xp)
+
+    def test_set_value_by_index(self):
+        # See gh-12344
+        warn = None
+        msg = "will attempt to set the values inplace"
+
+        df = DataFrame(np.arange(9).reshape(3, 3).T)
+        df.columns = list("AAA")
+        expected = df.iloc[:, 2].copy()
+
+        with tm.assert_produces_warning(warn, match=msg):
+            df.iloc[:, 0] = 3
+        tm.assert_series_equal(df.iloc[:, 2], expected)
+
+        df = DataFrame(np.arange(9).reshape(3, 3).T)
+        df.columns = [2, float(2), str(2)]
+        expected = df.iloc[:, 1].copy()
+
+        with tm.assert_produces_warning(warn, match=msg):
+            df.iloc[:, 0] = 3
+        tm.assert_series_equal(df.iloc[:, 1], expected)
diff --git a/pandas/pandas/tests/frame/test_npfuncs.py b/pandas/pandas/tests/frame/test_npfuncs.py
new file mode 100644
index 0000000000000..e9a241202d156
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_npfuncs.py
@@ -0,0 +1,84 @@
+"""
+Tests for np.foo applied to DataFrame, not necessarily ufuncs.
+"""
+
+import numpy as np
+
+from pandas import (
+    Categorical,
+    DataFrame,
+)
+import pandas._testing as tm
+
+
+class TestAsArray:
+    def test_asarray_homogeneous(self):
+        df = DataFrame({"A": Categorical([1, 2]), "B": Categorical([1, 2])})
+        result = np.asarray(df)
+        # may change from object in the future
+        expected = np.array([[1, 1], [2, 2]], dtype="object")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_np_sqrt(self, float_frame):
+        with np.errstate(all="ignore"):
+            result = np.sqrt(float_frame)
+        assert isinstance(result, type(float_frame))
+        assert result.index.is_(float_frame.index)
+        assert result.columns.is_(float_frame.columns)
+
+        tm.assert_frame_equal(result, float_frame.apply(np.sqrt))
+
+    def test_sum_axis_behavior(self):
+        # GH#52042 df.sum(axis=None) now reduces over both axes, which gets
+        #  called when we do np.sum(df)
+
+        arr = np.random.default_rng(2).standard_normal((4, 3))
+        df = DataFrame(arr)
+
+        res = np.sum(df)
+        expected = df.to_numpy().sum(axis=None)
+        assert res == expected
+
+    def test_np_ravel(self):
+        # GH26247
+        arr = np.array(
+            [
+                [0.11197053, 0.44361564, -0.92589452],
+                [0.05883648, -0.00948922, -0.26469934],
+            ]
+        )
+
+        result = np.ravel([DataFrame(batch.reshape(1, 3)) for batch in arr])
+        expected = np.array(
+            [
+                0.11197053,
+                0.44361564,
+                -0.92589452,
+                0.05883648,
+                -0.00948922,
+                -0.26469934,
+            ]
+        )
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.ravel(DataFrame(arr[0].reshape(1, 3), columns=["x1", "x2", "x3"]))
+        expected = np.array([0.11197053, 0.44361564, -0.92589452])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.ravel(
+            [
+                DataFrame(batch.reshape(1, 3), columns=["x1", "x2", "x3"])
+                for batch in arr
+            ]
+        )
+        expected = np.array(
+            [
+                0.11197053,
+                0.44361564,
+                -0.92589452,
+                0.05883648,
+                -0.00948922,
+                -0.26469934,
+            ]
+        )
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_query_eval.py b/pandas/pandas/tests/frame/test_query_eval.py
new file mode 100644
index 0000000000000..375b9b00a4988
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_query_eval.py
@@ -0,0 +1,1611 @@
+import operator
+from tokenize import TokenError
+
+import numpy as np
+import pytest
+
+from pandas.errors import (
+    NumExprClobberingError,
+    UndefinedVariableError,
+)
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.computation.check import NUMEXPR_INSTALLED
+
+
+@pytest.fixture(params=["python", "pandas"], ids=lambda x: x)
+def parser(request):
+    return request.param
+
+
+@pytest.fixture(
+    params=["python", pytest.param("numexpr", marks=td.skip_if_no("numexpr"))],
+    ids=lambda x: x,
+)
+def engine(request):
+    return request.param
+
+
+def skip_if_no_pandas_parser(parser):
+    if parser != "pandas":
+        pytest.skip(f"cannot evaluate with parser={parser}")
+
+
+class TestCompat:
+    @pytest.fixture
+    def df(self):
+        return DataFrame({"A": [1, 2, 3]})
+
+    @pytest.fixture
+    def expected1(self, df):
+        return df[df.A > 0]
+
+    @pytest.fixture
+    def expected2(self, df):
+        return df.A + 1
+
+    def test_query_default(self, df, expected1, expected2):
+        # GH 12749
+        # this should always work, whether NUMEXPR_INSTALLED or not
+        result = df.query("A>0")
+        tm.assert_frame_equal(result, expected1)
+        result = df.eval("A+1")
+        tm.assert_series_equal(result, expected2)
+
+    def test_query_None(self, df, expected1, expected2):
+        result = df.query("A>0", engine=None)
+        tm.assert_frame_equal(result, expected1)
+        result = df.eval("A+1", engine=None)
+        tm.assert_series_equal(result, expected2)
+
+    def test_query_python(self, df, expected1, expected2):
+        result = df.query("A>0", engine="python")
+        tm.assert_frame_equal(result, expected1)
+        result = df.eval("A+1", engine="python")
+        tm.assert_series_equal(result, expected2)
+
+    def test_query_numexpr(self, df, expected1, expected2):
+        if NUMEXPR_INSTALLED:
+            result = df.query("A>0", engine="numexpr")
+            tm.assert_frame_equal(result, expected1)
+            result = df.eval("A+1", engine="numexpr")
+            tm.assert_series_equal(result, expected2)
+        else:
+            msg = (
+                r"'numexpr' is not installed or an unsupported version. "
+                r"Cannot use engine='numexpr' for query/eval if 'numexpr' is "
+                r"not installed"
+            )
+            with pytest.raises(ImportError, match=msg):
+                df.query("A>0", engine="numexpr")
+            with pytest.raises(ImportError, match=msg):
+                df.eval("A+1", engine="numexpr")
+
+
+class TestDataFrameEval:
+    # smaller hits python, larger hits numexpr
+    @pytest.mark.parametrize("n", [4, 4000])
+    @pytest.mark.parametrize(
+        "op_str,op,rop",
+        [
+            ("+", "__add__", "__radd__"),
+            ("-", "__sub__", "__rsub__"),
+            ("*", "__mul__", "__rmul__"),
+            ("/", "__truediv__", "__rtruediv__"),
+        ],
+    )
+    def test_ops(self, op_str, op, rop, n):
+        # tst ops and reversed ops in evaluation
+        # GH7198
+
+        df = DataFrame(1, index=range(n), columns=list("abcd"))
+        df.iloc[0] = 2
+        m = df.mean()
+
+        base = DataFrame(  # noqa: F841
+            np.tile(m.values, n).reshape(n, -1), columns=list("abcd")
+        )
+
+        expected = eval(f"base {op_str} df")
+
+        # ops as strings
+        result = eval(f"m {op_str} df")
+        tm.assert_frame_equal(result, expected)
+
+        # these are commutative
+        if op in ["+", "*"]:
+            result = getattr(df, op)(m)
+            tm.assert_frame_equal(result, expected)
+
+        # these are not
+        elif op in ["-", "/"]:
+            result = getattr(df, rop)(m)
+            tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_sub_numexpr_path(self):
+        # GH7192: Note we need a large number of rows to ensure this
+        #  goes through the numexpr path
+        df = DataFrame({"A": np.random.default_rng(2).standard_normal(25000)})
+        df.iloc[0:5] = np.nan
+        expected = 1 - np.isnan(df.iloc[0:25])
+        result = (1 - np.isnan(df)).iloc[0:25]
+        tm.assert_frame_equal(result, expected)
+
+    def test_query_non_str(self):
+        # GH 11485
+        df = DataFrame({"A": [1, 2, 3], "B": ["a", "b", "b"]})
+
+        msg = "expr must be a string to be evaluated"
+        with pytest.raises(ValueError, match=msg):
+            df.query(lambda x: x.B == "b")
+
+        with pytest.raises(ValueError, match=msg):
+            df.query(111)
+
+    def test_query_empty_string(self):
+        # GH 13139
+        df = DataFrame({"A": [1, 2, 3]})
+
+        msg = "expr cannot be an empty string"
+        with pytest.raises(ValueError, match=msg):
+            df.query("")
+
+    def test_query_duplicate_column_name(self, engine, parser):
+        df = DataFrame(
+            {
+                "A": range(3),
+                "B": range(3),
+                "C": range(3)
+            }
+        ).rename(columns={"B": "A"})
+
+        res = df.query('C == 1', engine=engine, parser=parser)
+
+        expect = DataFrame(
+            [[1, 1, 1]],
+            columns=["A", "A", "C"],
+            index=[1]
+        )
+
+        tm.assert_frame_equal(res, expect)
+
+    def test_eval_resolvers_as_list(self):
+        # GH 14095
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)), columns=list("ab")
+        )
+        dict1 = {"a": 1}
+        dict2 = {"b": 2}
+        assert df.eval("a + b", resolvers=[dict1, dict2]) == dict1["a"] + dict2["b"]
+        assert pd.eval("a + b", resolvers=[dict1, dict2]) == dict1["a"] + dict2["b"]
+
+    def test_eval_resolvers_combined(self):
+        # GH 34966
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)), columns=list("ab")
+        )
+        dict1 = {"c": 2}
+
+        # Both input and default index/column resolvers should be usable
+        result = df.eval("a + b * c", resolvers=[dict1])
+
+        expected = df["a"] + df["b"] * dict1["c"]
+        tm.assert_series_equal(result, expected)
+
+    def test_eval_object_dtype_binop(self):
+        # GH#24883
+        df = DataFrame({"a1": ["Y", "N"]})
+        res = df.eval("c = ((a1 == 'Y') & True)")
+        expected = DataFrame({"a1": ["Y", "N"], "c": [True, False]})
+        tm.assert_frame_equal(res, expected)
+
+    def test_using_numpy(self, engine, parser):
+        # GH 58041
+        skip_if_no_pandas_parser(parser)
+        df = Series([0.2, 1.5, 2.8], name="a").to_frame()
+        res = df.eval("@np.floor(a)", engine=engine, parser=parser)
+        expected = np.floor(df["a"])
+        tm.assert_series_equal(expected, res)
+
+    def test_eval_simple(self, engine, parser):
+        df = Series([0.2, 1.5, 2.8], name="a").to_frame()
+        res = df.eval("a", engine=engine, parser=parser)
+        expected = df["a"]
+        tm.assert_series_equal(expected, res)
+
+    def test_extension_array_eval(self, engine, parser, request):
+        # GH#58748
+        if engine == "numexpr":
+            mark = pytest.mark.xfail(
+                reason="numexpr does not support extension array dtypes"
+            )
+            request.applymarker(mark)
+        df = DataFrame({"a": pd.array([1, 2, 3]), "b": pd.array([4, 5, 6])})
+        result = df.eval("a / b", engine=engine, parser=parser)
+        expected = Series(pd.array([0.25, 0.40, 0.50]))
+        tm.assert_series_equal(result, expected)
+
+    def test_complex_eval(self, engine, parser):
+        # GH#21374
+        df = DataFrame({"a": [1 + 2j], "b": [1 + 1j]})
+        result = df.eval("a/b", engine=engine, parser=parser)
+        expected = Series([1.5 + 0.5j])
+        tm.assert_series_equal(result, expected)
+
+
+class TestDataFrameQueryWithMultiIndex:
+    def test_query_with_named_multiindex(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        a = np.random.default_rng(2).choice(["red", "green"], size=10)
+        b = np.random.default_rng(2).choice(["eggs", "ham"], size=10)
+        index = MultiIndex.from_arrays([a, b], names=["color", "food"])
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)), index=index)
+        ind = Series(
+            df.index.get_level_values("color").values, index=index, name="color"
+        )
+
+        # equality
+        res1 = df.query('color == "red"', parser=parser, engine=engine)
+        res2 = df.query('"red" == color', parser=parser, engine=engine)
+        exp = df[ind == "red"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # inequality
+        res1 = df.query('color != "red"', parser=parser, engine=engine)
+        res2 = df.query('"red" != color', parser=parser, engine=engine)
+        exp = df[ind != "red"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # list equality (really just set membership)
+        res1 = df.query('color == ["red"]', parser=parser, engine=engine)
+        res2 = df.query('["red"] == color', parser=parser, engine=engine)
+        exp = df[ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('color != ["red"]', parser=parser, engine=engine)
+        res2 = df.query('["red"] != color', parser=parser, engine=engine)
+        exp = df[~ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # in/not in ops
+        res1 = df.query('["red"] in color', parser=parser, engine=engine)
+        res2 = df.query('"red" in color', parser=parser, engine=engine)
+        exp = df[ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('["red"] not in color', parser=parser, engine=engine)
+        res2 = df.query('"red" not in color', parser=parser, engine=engine)
+        exp = df[~ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+    def test_query_with_unnamed_multiindex(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        a = np.random.default_rng(2).choice(["red", "green"], size=10)
+        b = np.random.default_rng(2).choice(["eggs", "ham"], size=10)
+        index = MultiIndex.from_arrays([a, b])
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)), index=index)
+        ind = Series(df.index.get_level_values(0).values, index=index)
+
+        res1 = df.query('ilevel_0 == "red"', parser=parser, engine=engine)
+        res2 = df.query('"red" == ilevel_0', parser=parser, engine=engine)
+        exp = df[ind == "red"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # inequality
+        res1 = df.query('ilevel_0 != "red"', parser=parser, engine=engine)
+        res2 = df.query('"red" != ilevel_0', parser=parser, engine=engine)
+        exp = df[ind != "red"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # list equality (really just set membership)
+        res1 = df.query('ilevel_0 == ["red"]', parser=parser, engine=engine)
+        res2 = df.query('["red"] == ilevel_0', parser=parser, engine=engine)
+        exp = df[ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('ilevel_0 != ["red"]', parser=parser, engine=engine)
+        res2 = df.query('["red"] != ilevel_0', parser=parser, engine=engine)
+        exp = df[~ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # in/not in ops
+        res1 = df.query('["red"] in ilevel_0', parser=parser, engine=engine)
+        res2 = df.query('"red" in ilevel_0', parser=parser, engine=engine)
+        exp = df[ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('["red"] not in ilevel_0', parser=parser, engine=engine)
+        res2 = df.query('"red" not in ilevel_0', parser=parser, engine=engine)
+        exp = df[~ind.isin(["red"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # ## LEVEL 1
+        ind = Series(df.index.get_level_values(1).values, index=index)
+        res1 = df.query('ilevel_1 == "eggs"', parser=parser, engine=engine)
+        res2 = df.query('"eggs" == ilevel_1', parser=parser, engine=engine)
+        exp = df[ind == "eggs"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # inequality
+        res1 = df.query('ilevel_1 != "eggs"', parser=parser, engine=engine)
+        res2 = df.query('"eggs" != ilevel_1', parser=parser, engine=engine)
+        exp = df[ind != "eggs"]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # list equality (really just set membership)
+        res1 = df.query('ilevel_1 == ["eggs"]', parser=parser, engine=engine)
+        res2 = df.query('["eggs"] == ilevel_1', parser=parser, engine=engine)
+        exp = df[ind.isin(["eggs"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('ilevel_1 != ["eggs"]', parser=parser, engine=engine)
+        res2 = df.query('["eggs"] != ilevel_1', parser=parser, engine=engine)
+        exp = df[~ind.isin(["eggs"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        # in/not in ops
+        res1 = df.query('["eggs"] in ilevel_1', parser=parser, engine=engine)
+        res2 = df.query('"eggs" in ilevel_1', parser=parser, engine=engine)
+        exp = df[ind.isin(["eggs"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+        res1 = df.query('["eggs"] not in ilevel_1', parser=parser, engine=engine)
+        res2 = df.query('"eggs" not in ilevel_1', parser=parser, engine=engine)
+        exp = df[~ind.isin(["eggs"])]
+        tm.assert_frame_equal(res1, exp)
+        tm.assert_frame_equal(res2, exp)
+
+    def test_query_with_partially_named_multiindex(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        a = np.random.default_rng(2).choice(["red", "green"], size=10)
+        b = np.arange(10)
+        index = MultiIndex.from_arrays([a, b])
+        index.names = [None, "rating"]
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)), index=index)
+        res = df.query("rating == 1", parser=parser, engine=engine)
+        ind = Series(
+            df.index.get_level_values("rating").values, index=index, name="rating"
+        )
+        exp = df[ind == 1]
+        tm.assert_frame_equal(res, exp)
+
+        res = df.query("rating != 1", parser=parser, engine=engine)
+        ind = Series(
+            df.index.get_level_values("rating").values, index=index, name="rating"
+        )
+        exp = df[ind != 1]
+        tm.assert_frame_equal(res, exp)
+
+        res = df.query('ilevel_0 == "red"', parser=parser, engine=engine)
+        ind = Series(df.index.get_level_values(0).values, index=index)
+        exp = df[ind == "red"]
+        tm.assert_frame_equal(res, exp)
+
+        res = df.query('ilevel_0 != "red"', parser=parser, engine=engine)
+        ind = Series(df.index.get_level_values(0).values, index=index)
+        exp = df[ind != "red"]
+        tm.assert_frame_equal(res, exp)
+
+    def test_query_multiindex_get_index_resolvers(self):
+        df = DataFrame(
+            np.ones((10, 3)),
+            index=MultiIndex.from_arrays(
+                [range(10) for _ in range(2)], names=["spam", "eggs"]
+            ),
+        )
+        resolvers = df._get_index_resolvers()
+
+        def to_series(mi, level):
+            level_values = mi.get_level_values(level)
+            s = level_values.to_series()
+            s.index = mi
+            return s
+
+        col_series = df.columns.to_series()
+        expected = {
+            "index": df.index,
+            "columns": col_series,
+            "spam": to_series(df.index, "spam"),
+            "eggs": to_series(df.index, "eggs"),
+            "clevel_0": col_series,
+        }
+        for k, v in resolvers.items():
+            if isinstance(v, Index):
+                assert v.is_(expected[k])
+            elif isinstance(v, Series):
+                tm.assert_series_equal(v, expected[k])
+            else:
+                raise AssertionError("object must be a Series or Index")
+
+
+@td.skip_if_no("numexpr")
+class TestDataFrameQueryNumExprPandas:
+    @pytest.fixture
+    def engine(self):
+        return "numexpr"
+
+    @pytest.fixture
+    def parser(self):
+        return "pandas"
+
+    def test_date_query_with_attribute_access(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=5)
+        df["dates2"] = date_range("1/1/2013", periods=5)
+        df["dates3"] = date_range("1/1/2014", periods=5)
+        res = df.query(
+            "@df.dates1 < 20130101 < @df.dates3", engine=engine, parser=parser
+        )
+        expec = df[(df.dates1 < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_query_no_attribute_access(self, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=5)
+        df["dates2"] = date_range("1/1/2013", periods=5)
+        df["dates3"] = date_range("1/1/2014", periods=5)
+        res = df.query("dates1 < 20130101 < dates3", engine=engine, parser=parser)
+        expec = df[(df.dates1 < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_query_with_NaT(self, engine, parser):
+        n = 10
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates2"] = date_range("1/1/2013", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates1"] = pd.NaT
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates3"] = pd.NaT
+        res = df.query("dates1 < 20130101 < dates3", engine=engine, parser=parser)
+        expec = df[(df.dates1 < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query(self, engine, parser):
+        n = 10
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        res = df.query("index < 20130101 < dates3", engine=engine, parser=parser)
+        expec = df[(df.index < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query_with_NaT(self, engine, parser):
+        n = 10
+        # Cast to object to avoid implicit cast when setting entry to pd.NaT below
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3))).astype(
+            {0: object}
+        )
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        df.iloc[0, 0] = pd.NaT
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        res = df.query("index < 20130101 < dates3", engine=engine, parser=parser)
+        expec = df[(df.index < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query_with_NaT_duplicates(self, engine, parser):
+        n = 10
+        d = {}
+        d["dates1"] = date_range("1/1/2012", periods=n)
+        d["dates3"] = date_range("1/1/2014", periods=n)
+        df = DataFrame(d)
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates1"] = pd.NaT
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        res = df.query("dates1 < 20130101 < dates3", engine=engine, parser=parser)
+        expec = df[(df.index.to_series() < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_query_with_non_date(self, engine, parser):
+        n = 10
+        df = DataFrame(
+            {"dates": date_range("1/1/2012", periods=n), "nondate": np.arange(n)}
+        )
+
+        result = df.query("dates == nondate", parser=parser, engine=engine)
+        assert len(result) == 0
+
+        result = df.query("dates != nondate", parser=parser, engine=engine)
+        tm.assert_frame_equal(result, df)
+
+        msg = r"Invalid comparison between dtype=datetime64\[ns\] and ndarray"
+        for op in ["<", ">", "<=", ">="]:
+            with pytest.raises(TypeError, match=msg):
+                df.query(f"dates {op} nondate", parser=parser, engine=engine)
+
+    def test_query_syntax_error(self, engine, parser):
+        df = DataFrame({"i": range(10), "+": range(3, 13), "r": range(4, 14)})
+        msg = "invalid syntax"
+        with pytest.raises(SyntaxError, match=msg):
+            df.query("i - +", engine=engine, parser=parser)
+
+    def test_query_scope(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 2)), columns=list("ab")
+        )
+
+        a, b = 1, 2  # noqa: F841
+        res = df.query("a > b", engine=engine, parser=parser)
+        expected = df[df.a > df.b]
+        tm.assert_frame_equal(res, expected)
+
+        res = df.query("@a > b", engine=engine, parser=parser)
+        expected = df[a > df.b]
+        tm.assert_frame_equal(res, expected)
+
+        # no local variable c
+        with pytest.raises(
+            UndefinedVariableError, match="local variable 'c' is not defined"
+        ):
+            df.query("@a > b > @c", engine=engine, parser=parser)
+
+        # no column named 'c'
+        with pytest.raises(UndefinedVariableError, match="name 'c' is not defined"):
+            df.query("@a > b > c", engine=engine, parser=parser)
+
+    def test_query_doesnt_pickup_local(self, engine, parser):
+        n = m = 10
+        df = DataFrame(
+            np.random.default_rng(2).integers(m, size=(n, 3)), columns=list("abc")
+        )
+
+        # we don't pick up the local 'sin'
+        with pytest.raises(UndefinedVariableError, match="name 'sin' is not defined"):
+            df.query("sin > 5", engine=engine, parser=parser)
+
+    def test_query_builtin(self, engine, parser):
+        n = m = 10
+        df = DataFrame(
+            np.random.default_rng(2).integers(m, size=(n, 3)), columns=list("abc")
+        )
+
+        df.index.name = "sin"
+        msg = "Variables in expression.+"
+        with pytest.raises(NumExprClobberingError, match=msg):
+            df.query("sin > 5", engine=engine, parser=parser)
+
+    def test_query(self, engine, parser):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=["a", "b", "c"]
+        )
+
+        tm.assert_frame_equal(
+            df.query("a < b", engine=engine, parser=parser), df[df.a < df.b]
+        )
+        tm.assert_frame_equal(
+            df.query("a + b > b * c", engine=engine, parser=parser),
+            df[df.a + df.b > df.b * df.c],
+        )
+
+    def test_query_index_with_name(self, engine, parser):
+        df = DataFrame(
+            np.random.default_rng(2).integers(10, size=(10, 3)),
+            index=Index(range(10), name="blob"),
+            columns=["a", "b", "c"],
+        )
+        res = df.query("(blob < 5) & (a < b)", engine=engine, parser=parser)
+        expec = df[(df.index < 5) & (df.a < df.b)]
+        tm.assert_frame_equal(res, expec)
+
+        res = df.query("blob < b", engine=engine, parser=parser)
+        expec = df[df.index < df.b]
+
+        tm.assert_frame_equal(res, expec)
+
+    def test_query_index_without_name(self, engine, parser):
+        df = DataFrame(
+            np.random.default_rng(2).integers(10, size=(10, 3)),
+            index=range(10),
+            columns=["a", "b", "c"],
+        )
+
+        # "index" should refer to the index
+        res = df.query("index < b", engine=engine, parser=parser)
+        expec = df[df.index < df.b]
+        tm.assert_frame_equal(res, expec)
+
+        # test against a scalar
+        res = df.query("index < 5", engine=engine, parser=parser)
+        expec = df[df.index < 5]
+        tm.assert_frame_equal(res, expec)
+
+    def test_nested_scope(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        expected = df[(df > 0) & (df2 > 0)]
+
+        result = df.query("(@df > 0) & (@df2 > 0)", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.eval("df[df > 0 and df2 > 0]", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.eval(
+            "df[df > 0 and df2 > 0 and df[df > 0] > 0]", engine=engine, parser=parser
+        )
+        expected = df[(df > 0) & (df2 > 0) & (df[df > 0] > 0)]
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.eval("df[(df>0) & (df2>0)]", engine=engine, parser=parser)
+        expected = df.query("(@df>0) & (@df2>0)", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_raises_on_local_self_reference(self, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+
+        # can't reference ourself b/c we're a local so @ is necessary
+        with pytest.raises(UndefinedVariableError, match="name 'df' is not defined"):
+            df.query("df > 0", engine=engine, parser=parser)
+
+    def test_local_syntax(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 10)),
+            columns=list("abcdefghij"),
+        )
+        b = 1
+        expect = df[df.a < b]
+        result = df.query("a < @b", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expect)
+
+        expect = df[df.a < df.b]
+        result = df.query("a < b", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expect)
+
+    def test_chained_cmp_and_in(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+        cols = list("abc")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((100, len(cols))), columns=cols
+        )
+        res = df.query(
+            "a < b < c and a not in b not in c", engine=engine, parser=parser
+        )
+        ind = (df.a < df.b) & (df.b < df.c) & ~df.b.isin(df.a) & ~df.c.isin(df.b)
+        expec = df[ind]
+        tm.assert_frame_equal(res, expec)
+
+    def test_local_variable_with_in(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+        a = Series(np.random.default_rng(2).integers(3, size=15), name="a")
+        b = Series(np.random.default_rng(2).integers(10, size=15), name="b")
+        df = DataFrame({"a": a, "b": b})
+
+        expected = df.loc[(df.b - 1).isin(a)]
+        result = df.query("b - 1 in a", engine=engine, parser=parser)
+        tm.assert_frame_equal(expected, result)
+
+        b = Series(np.random.default_rng(2).integers(10, size=15), name="b")
+        expected = df.loc[(b - 1).isin(a)]
+        result = df.query("@b - 1 in a", engine=engine, parser=parser)
+        tm.assert_frame_equal(expected, result)
+
+    def test_at_inside_string(self, engine, parser):
+        skip_if_no_pandas_parser(parser)
+        c = 1  # noqa: F841
+        df = DataFrame({"a": ["a", "a", "b", "b", "@c", "@c"]})
+        result = df.query('a == "@c"', engine=engine, parser=parser)
+        expected = df[df.a == "@c"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_query_undefined_local(self):
+        engine, parser = self.engine, self.parser
+        skip_if_no_pandas_parser(parser)
+
+        df = DataFrame(np.random.default_rng(2).random((10, 2)), columns=list("ab"))
+        with pytest.raises(
+            UndefinedVariableError, match="local variable 'c' is not defined"
+        ):
+            df.query("a == @c", engine=engine, parser=parser)
+
+    def test_index_resolvers_come_after_columns_with_the_same_name(
+        self, engine, parser
+    ):
+        n = 1  # noqa: F841
+        a = np.r_[20:101:20]
+
+        df = DataFrame(
+            {"index": a, "b": np.random.default_rng(2).standard_normal(a.size)}
+        )
+        df.index.name = "index"
+        result = df.query("index > 5", engine=engine, parser=parser)
+        expected = df[df["index"] > 5]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {"index": a, "b": np.random.default_rng(2).standard_normal(a.size)}
+        )
+        result = df.query("ilevel_0 > 5", engine=engine, parser=parser)
+        expected = df.loc[df.index[df.index > 5]]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({"a": a, "b": np.random.default_rng(2).standard_normal(a.size)})
+        df.index.name = "a"
+        result = df.query("a > 5", engine=engine, parser=parser)
+        expected = df[df.a > 5]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.query("index > 5", engine=engine, parser=parser)
+        expected = df.loc[df.index[df.index > 5]]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("op, f", [["==", operator.eq], ["!=", operator.ne]])
+    def test_inf(self, op, f, engine, parser):
+        n = 10
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).random(n),
+                "b": np.random.default_rng(2).random(n),
+            }
+        )
+        df.loc[::2, 0] = np.inf
+        q = f"a {op} inf"
+        expected = df[f(df.a, np.inf)]
+        result = df.query(q, engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+    def test_check_tz_aware_index_query(self, tz_aware_fixture):
+        # https://github.com/pandas-dev/pandas/issues/29463
+        tz = tz_aware_fixture
+        df_index = date_range(
+            start="2019-01-01", freq="1D", periods=10, tz=tz, name="time"
+        )
+        expected = DataFrame(index=df_index)
+        df = DataFrame(index=df_index)
+        result = df.query('"2018-01-03 00:00:00+00" < time')
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(df_index)
+        result = df.reset_index().query('"2018-01-03 00:00:00+00" < time')
+        tm.assert_frame_equal(result, expected)
+
+    def test_method_calls_in_query(self, engine, parser):
+        # https://github.com/pandas-dev/pandas/issues/22435
+        n = 10
+        df = DataFrame(
+            {
+                "a": 2 * np.random.default_rng(2).random(n),
+                "b": np.random.default_rng(2).random(n),
+            }
+        )
+        expected = df[df["a"].astype("int") == 0]
+        result = df.query("a.astype('int') == 0", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "a": np.where(
+                    np.random.default_rng(2).random(n) < 0.5,
+                    np.nan,
+                    np.random.default_rng(2).standard_normal(n),
+                ),
+                "b": np.random.default_rng(2).standard_normal(n),
+            }
+        )
+        expected = df[df["a"].notnull()]
+        result = df.query("a.notnull()", engine=engine, parser=parser)
+        tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("numexpr")
+class TestDataFrameQueryNumExprPython(TestDataFrameQueryNumExprPandas):
+    @pytest.fixture
+    def engine(self):
+        return "numexpr"
+
+    @pytest.fixture
+    def parser(self):
+        return "python"
+
+    def test_date_query_no_attribute_access(self, engine, parser):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=5)
+        df["dates2"] = date_range("1/1/2013", periods=5)
+        df["dates3"] = date_range("1/1/2014", periods=5)
+        res = df.query(
+            "(dates1 < 20130101) & (20130101 < dates3)", engine=engine, parser=parser
+        )
+        expec = df[(df.dates1 < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_query_with_NaT(self, engine, parser):
+        n = 10
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates2"] = date_range("1/1/2013", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates1"] = pd.NaT
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates3"] = pd.NaT
+        res = df.query(
+            "(dates1 < 20130101) & (20130101 < dates3)", engine=engine, parser=parser
+        )
+        expec = df[(df.dates1 < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query(self, engine, parser):
+        n = 10
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        res = df.query(
+            "(index < 20130101) & (20130101 < dates3)", engine=engine, parser=parser
+        )
+        expec = df[(df.index < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query_with_NaT(self, engine, parser):
+        n = 10
+        # Cast to object to avoid implicit cast when setting entry to pd.NaT below
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3))).astype(
+            {0: object}
+        )
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        df.iloc[0, 0] = pd.NaT
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        res = df.query(
+            "(index < 20130101) & (20130101 < dates3)", engine=engine, parser=parser
+        )
+        expec = df[(df.index < "20130101") & ("20130101" < df.dates3)]
+        tm.assert_frame_equal(res, expec)
+
+    def test_date_index_query_with_NaT_duplicates(self, engine, parser):
+        n = 10
+        df = DataFrame(np.random.default_rng(2).standard_normal((n, 3)))
+        df["dates1"] = date_range("1/1/2012", periods=n)
+        df["dates3"] = date_range("1/1/2014", periods=n)
+        df.loc[np.random.default_rng(2).random(n) > 0.5, "dates1"] = pd.NaT
+        return_value = df.set_index("dates1", inplace=True, drop=True)
+        assert return_value is None
+        msg = r"'BoolOp' nodes are not implemented"
+        with pytest.raises(NotImplementedError, match=msg):
+            df.query("index < 20130101 < dates3", engine=engine, parser=parser)
+
+    def test_nested_scope(self, engine, parser):
+        # smoke test
+        x = 1  # noqa: F841
+        result = pd.eval("x + 1", engine=engine, parser=parser)
+        assert result == 2
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+
+        # don't have the pandas parser
+        msg = r"The '@' prefix is only supported by the pandas parser"
+        with pytest.raises(SyntaxError, match=msg):
+            df.query("(@df>0) & (@df2>0)", engine=engine, parser=parser)
+
+        with pytest.raises(UndefinedVariableError, match="name 'df' is not defined"):
+            df.query("(df>0) & (df2>0)", engine=engine, parser=parser)
+
+        expected = df[(df > 0) & (df2 > 0)]
+        result = pd.eval("df[(df > 0) & (df2 > 0)]", engine=engine, parser=parser)
+        tm.assert_frame_equal(expected, result)
+
+        expected = df[(df > 0) & (df2 > 0) & (df[df > 0] > 0)]
+        result = pd.eval(
+            "df[(df > 0) & (df2 > 0) & (df[df > 0] > 0)]", engine=engine, parser=parser
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_query_numexpr_with_min_and_max_columns(self):
+        df = DataFrame({"min": [1, 2, 3], "max": [4, 5, 6]})
+        regex_to_match = (
+            r"Variables in expression \"\(min\) == \(1\)\" "
+            r"overlap with builtins: \('min'\)"
+        )
+        with pytest.raises(NumExprClobberingError, match=regex_to_match):
+            df.query("min == 1")
+
+        regex_to_match = (
+            r"Variables in expression \"\(max\) == \(1\)\" "
+            r"overlap with builtins: \('max'\)"
+        )
+        with pytest.raises(NumExprClobberingError, match=regex_to_match):
+            df.query("max == 1")
+
+
+class TestDataFrameQueryPythonPandas(TestDataFrameQueryNumExprPandas):
+    @pytest.fixture
+    def engine(self):
+        return "python"
+
+    @pytest.fixture
+    def parser(self):
+        return "pandas"
+
+    def test_query_builtin(self, engine, parser):
+        n = m = 10
+        df = DataFrame(
+            np.random.default_rng(2).integers(m, size=(n, 3)), columns=list("abc")
+        )
+
+        df.index.name = "sin"
+        expected = df[df.index > 5]
+        result = df.query("sin > 5", engine=engine, parser=parser)
+        tm.assert_frame_equal(expected, result)
+
+
+class TestDataFrameQueryPythonPython(TestDataFrameQueryNumExprPython):
+    @pytest.fixture
+    def engine(self):
+        return "python"
+
+    @pytest.fixture
+    def parser(self):
+        return "python"
+
+    def test_query_builtin(self, engine, parser):
+        n = m = 10
+        df = DataFrame(
+            np.random.default_rng(2).integers(m, size=(n, 3)), columns=list("abc")
+        )
+
+        df.index.name = "sin"
+        expected = df[df.index > 5]
+        result = df.query("sin > 5", engine=engine, parser=parser)
+        tm.assert_frame_equal(expected, result)
+
+
+class TestDataFrameQueryStrings:
+    def test_str_query_method(self, parser, engine):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 1)), columns=["b"])
+        df["strings"] = Series(list("aabbccddee"))
+        expect = df[df.strings == "a"]
+
+        if parser != "pandas":
+            col = "strings"
+            lst = '"a"'
+
+            lhs = [col] * 2 + [lst] * 2
+            rhs = lhs[::-1]
+
+            eq, ne = "==", "!="
+            ops = 2 * ([eq] + [ne])
+            msg = r"'(Not)?In' nodes are not implemented"
+
+            for lh, op_, rh in zip(lhs, ops, rhs):
+                ex = f"{lh} {op_} {rh}"
+                with pytest.raises(NotImplementedError, match=msg):
+                    df.query(
+                        ex,
+                        engine=engine,
+                        parser=parser,
+                        local_dict={"strings": df.strings},
+                    )
+        else:
+            res = df.query('"a" == strings', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+            res = df.query('strings == "a"', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+            tm.assert_frame_equal(res, df[df.strings.isin(["a"])])
+
+            expect = df[df.strings != "a"]
+            res = df.query('strings != "a"', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+            res = df.query('"a" != strings', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+            tm.assert_frame_equal(res, df[~df.strings.isin(["a"])])
+
+    def test_str_list_query_method(self, parser, engine):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 1)), columns=["b"])
+        df["strings"] = Series(list("aabbccddee"))
+        expect = df[df.strings.isin(["a", "b"])]
+
+        if parser != "pandas":
+            col = "strings"
+            lst = '["a", "b"]'
+
+            lhs = [col] * 2 + [lst] * 2
+            rhs = lhs[::-1]
+
+            eq, ne = "==", "!="
+            ops = 2 * ([eq] + [ne])
+            msg = r"'(Not)?In' nodes are not implemented"
+
+            for lh, ops_, rh in zip(lhs, ops, rhs):
+                ex = f"{lh} {ops_} {rh}"
+                with pytest.raises(NotImplementedError, match=msg):
+                    df.query(ex, engine=engine, parser=parser)
+        else:
+            res = df.query('strings == ["a", "b"]', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+            res = df.query('["a", "b"] == strings', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+            expect = df[~df.strings.isin(["a", "b"])]
+
+            res = df.query('strings != ["a", "b"]', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+            res = df.query('["a", "b"] != strings', engine=engine, parser=parser)
+            tm.assert_frame_equal(res, expect)
+
+    def test_query_with_string_columns(self, parser, engine):
+        df = DataFrame(
+            {
+                "a": list("aaaabbbbcccc"),
+                "b": list("aabbccddeeff"),
+                "c": np.random.default_rng(2).integers(5, size=12),
+                "d": np.random.default_rng(2).integers(9, size=12),
+            }
+        )
+        if parser == "pandas":
+            res = df.query("a in b", parser=parser, engine=engine)
+            expec = df[df.a.isin(df.b)]
+            tm.assert_frame_equal(res, expec)
+
+            res = df.query("a in b and c < d", parser=parser, engine=engine)
+            expec = df[df.a.isin(df.b) & (df.c < df.d)]
+            tm.assert_frame_equal(res, expec)
+        else:
+            msg = r"'(Not)?In' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                df.query("a in b", parser=parser, engine=engine)
+
+            msg = r"'BoolOp' nodes are not implemented"
+            with pytest.raises(NotImplementedError, match=msg):
+                df.query("a in b and c < d", parser=parser, engine=engine)
+
+    def test_object_array_eq_ne(self, parser, engine):
+        df = DataFrame(
+            {
+                "a": list("aaaabbbbcccc"),
+                "b": list("aabbccddeeff"),
+                "c": np.random.default_rng(2).integers(5, size=12),
+                "d": np.random.default_rng(2).integers(9, size=12),
+            }
+        )
+        res = df.query("a == b", parser=parser, engine=engine)
+        exp = df[df.a == df.b]
+        tm.assert_frame_equal(res, exp)
+
+        res = df.query("a != b", parser=parser, engine=engine)
+        exp = df[df.a != df.b]
+        tm.assert_frame_equal(res, exp)
+
+    def test_query_with_nested_strings(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        events = [
+            f"page {n} {act}" for n in range(1, 4) for act in ["load", "exit"]
+        ] * 2
+        stamps1 = date_range("2014-01-01 0:00:01", freq="30s", periods=6)
+        stamps2 = date_range("2014-02-01 1:00:01", freq="30s", periods=6)
+        df = DataFrame(
+            {
+                "id": np.arange(1, 7).repeat(2),
+                "event": events,
+                "timestamp": stamps1.append(stamps2),
+            }
+        )
+
+        expected = df[df.event == '"page 1 load"']
+        res = df.query("""'"page 1 load"' in event""", parser=parser, engine=engine)
+        tm.assert_frame_equal(expected, res)
+
+    def test_query_with_nested_special_character(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        df = DataFrame({"a": ["a", "b", "test & test"], "b": [1, 2, 3]})
+        res = df.query('a == "test & test"', parser=parser, engine=engine)
+        expec = df[df.a == "test & test"]
+        tm.assert_frame_equal(res, expec)
+
+    @pytest.mark.parametrize(
+        "op, func",
+        [
+            ["<", operator.lt],
+            [">", operator.gt],
+            ["<=", operator.le],
+            [">=", operator.ge],
+        ],
+    )
+    def test_query_lex_compare_strings(
+        self, parser, engine, op, func
+    ):
+        a = Series(np.random.default_rng(2).choice(list("abcde"), 20))
+        b = Series(np.arange(a.size))
+        df = DataFrame({"X": a, "Y": b})
+
+        res = df.query(f'X {op} "d"', engine=engine, parser=parser)
+        expected = df[func(df.X, "d")]
+        tm.assert_frame_equal(res, expected)
+
+    def test_query_single_element_booleans(self, parser, engine):
+        columns = "bid", "bidsize", "ask", "asksize"
+        data = np.random.default_rng(2).integers(2, size=(1, len(columns))).astype(bool)
+        df = DataFrame(data, columns=columns)
+        res = df.query("bid & ask", engine=engine, parser=parser)
+        expected = df[df.bid & df.ask]
+        tm.assert_frame_equal(res, expected)
+
+    def test_query_string_scalar_variable(self, parser, engine):
+        skip_if_no_pandas_parser(parser)
+        df = DataFrame(
+            {
+                "Symbol": ["BUD US", "BUD US", "IBM US", "IBM US"],
+                "Price": [109.70, 109.72, 183.30, 183.35],
+            }
+        )
+        e = df[df.Symbol == "BUD US"]
+        symb = "BUD US"  # noqa: F841
+        r = df.query("Symbol == @symb", parser=parser, engine=engine)
+        tm.assert_frame_equal(e, r)
+
+    @pytest.mark.parametrize(
+        "in_list",
+        [
+            [None, "asdf", "ghjk"],
+            ["asdf", None, "ghjk"],
+            ["asdf", "ghjk", None],
+            [None, None, "asdf"],
+            ["asdf", None, None],
+            [None, None, None],
+        ],
+    )
+    def test_query_string_null_elements(self, in_list):
+        # GITHUB ISSUE #31516
+        parser = "pandas"
+        engine = "python"
+        expected = {i: value for i, value in enumerate(in_list) if value == "asdf"}
+
+        df_expected = DataFrame({"a": expected}, dtype="string")
+        df_expected.index = df_expected.index.astype("int64")
+        df = DataFrame({"a": in_list}, dtype="string")
+        df.index = Index(list(df.index), dtype=df.index.dtype)
+        res1 = df.query("a == 'asdf'", parser=parser, engine=engine)
+        res2 = df[df["a"] == "asdf"]
+        res3 = df.query("a <= 'asdf'", parser=parser, engine=engine)
+        tm.assert_frame_equal(res1, df_expected)
+        tm.assert_frame_equal(res1, res2)
+        tm.assert_frame_equal(res1, res3)
+        tm.assert_frame_equal(res2, res3)
+
+
+class TestDataFrameEvalWithFrame:
+    @pytest.fixture
+    def frame(self):
+        return DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=list("abc")
+        )
+
+    def test_simple_expr(self, frame, parser, engine):
+        res = frame.eval("a + b", engine=engine, parser=parser)
+        expect = frame.a + frame.b
+        tm.assert_series_equal(res, expect)
+
+    def test_bool_arith_expr(self, frame, parser, engine):
+        res = frame.eval("a[a < 1] + b", engine=engine, parser=parser)
+        expect = frame.a[frame.a < 1] + frame.b
+        tm.assert_series_equal(res, expect)
+
+    @pytest.mark.parametrize("op", ["+", "-", "*", "/"])
+    def test_invalid_type_for_operator_raises(self, parser, engine, op):
+        df = DataFrame({"a": [1, 2], "b": ["c", "d"]})
+        msg = r"unsupported operand type\(s\) for .+: '.+' and '.+'|Cannot"
+
+        with pytest.raises(TypeError, match=msg):
+            df.eval(f"a {op} b", engine=engine, parser=parser)
+
+
+class TestDataFrameQueryBacktickQuoting:
+    @pytest.fixture
+    def df(self):
+        """
+        Yields a dataframe with strings that may or may not need escaping
+        by backticks. The last two columns cannot be escaped by backticks
+        and should raise a ValueError.
+        """
+        return DataFrame(
+            {
+                "A": [1, 2, 3],
+                "B B": [3, 2, 1],
+                "C C": [4, 5, 6],
+                "C  C": [7, 4, 3],
+                "C_C": [8, 9, 10],
+                "D_D D": [11, 1, 101],
+                "E.E": [6, 3, 5],
+                "F-F": [8, 1, 10],
+                "1e1": [2, 4, 8],
+                "def": [10, 11, 2],
+                "A (x)": [4, 1, 3],
+                "B(x)": [1, 1, 5],
+                "B (x)": [2, 7, 4],
+                "  &^ :!€$?(} >    <++*''  ": [2, 5, 6],
+                "": [10, 11, 1],
+                " A": [4, 7, 9],
+                "  ": [1, 2, 1],
+                "it's": [6, 3, 1],
+                "that's": [9, 1, 8],
+                "☺": [8, 7, 6],
+                "xy （z）": [1, 2, 3],  # noqa: RUF001
+                "xy （z\\uff09": [4, 5, 6],  # noqa: RUF001
+                "foo#bar": [2, 4, 5],
+                1: [5, 7, 9],
+            }
+        )
+
+    def test_single_backtick_variable_query(self, df):
+        res = df.query("1 < `B B`")
+        expect = df[1 < df["B B"]]
+        tm.assert_frame_equal(res, expect)
+
+    def test_two_backtick_variables_query(self, df):
+        res = df.query("1 < `B B` and 4 < `C C`")
+        expect = df[(1 < df["B B"]) & (4 < df["C C"])]
+        tm.assert_frame_equal(res, expect)
+
+    def test_single_backtick_variable_expr(self, df):
+        res = df.eval("A + `B B`")
+        expect = df["A"] + df["B B"]
+        tm.assert_series_equal(res, expect)
+
+    def test_two_backtick_variables_expr(self, df):
+        res = df.eval("`B B` + `C C`")
+        expect = df["B B"] + df["C C"]
+        tm.assert_series_equal(res, expect)
+
+    def test_already_underscore_variable(self, df):
+        res = df.eval("`C_C` + A")
+        expect = df["C_C"] + df["A"]
+        tm.assert_series_equal(res, expect)
+
+    def test_same_name_but_underscores(self, df):
+        res = df.eval("C_C + `C C`")
+        expect = df["C_C"] + df["C C"]
+        tm.assert_series_equal(res, expect)
+
+    def test_mixed_underscores_and_spaces(self, df):
+        res = df.eval("A + `D_D D`")
+        expect = df["A"] + df["D_D D"]
+        tm.assert_series_equal(res, expect)
+
+    def test_backtick_quote_name_with_no_spaces(self, df):
+        res = df.eval("A + `C_C`")
+        expect = df["A"] + df["C_C"]
+        tm.assert_series_equal(res, expect)
+
+    def test_special_characters(self, df):
+        res = df.eval("`E.E` + `F-F` - A")
+        expect = df["E.E"] + df["F-F"] - df["A"]
+        tm.assert_series_equal(res, expect)
+
+    def test_start_with_digit(self, df):
+        res = df.eval("A + `1e1`")
+        expect = df["A"] + df["1e1"]
+        tm.assert_series_equal(res, expect)
+
+    def test_keyword(self, df):
+        res = df.eval("A + `def`")
+        expect = df["A"] + df["def"]
+        tm.assert_series_equal(res, expect)
+
+    def test_unneeded_quoting(self, df):
+        res = df.query("`A` > 2")
+        expect = df[df["A"] > 2]
+        tm.assert_frame_equal(res, expect)
+
+    def test_parenthesis(self, df):
+        res = df.query("`A (x)` > 2")
+        expect = df[df["A (x)"] > 2]
+        tm.assert_frame_equal(res, expect)
+
+    def test_empty_string(self, df):
+        res = df.query("`` > 5")
+        expect = df[df[""] > 5]
+        tm.assert_frame_equal(res, expect)
+
+    def test_multiple_spaces(self, df):
+        res = df.query("`C  C` > 5")
+        expect = df[df["C  C"] > 5]
+        tm.assert_frame_equal(res, expect)
+
+    def test_start_with_spaces(self, df):
+        res = df.eval("` A` + `  `")
+        expect = df[" A"] + df["  "]
+        tm.assert_series_equal(res, expect)
+
+    def test_lots_of_operators_string(self, df):
+        res = df.query("`  &^ :!€$?(} >    <++*''  ` > 4")
+        expect = df[df["  &^ :!€$?(} >    <++*''  "] > 4]
+        tm.assert_frame_equal(res, expect)
+
+    def test_missing_attribute(self, df):
+        message = "module 'pandas' has no attribute 'thing'"
+        with pytest.raises(AttributeError, match=message):
+            df.eval("@pd.thing")
+
+    def test_quote(self, df):
+        res = df.query("`it's` > `that's`")
+        expect = df[df["it's"] > df["that's"]]
+        tm.assert_frame_equal(res, expect)
+
+    def test_character_outside_range_smiley(self, df):
+        res = df.query("`☺` > 4")
+        expect = df[df["☺"] > 4]
+        tm.assert_frame_equal(res, expect)
+
+    def test_character_outside_range_2_byte_parens(self, df):
+        # GH 49633
+        res = df.query("`xy （z）` == 2")  # noqa: RUF001
+        expect = df[df["xy （z）"] == 2]  # noqa: RUF001
+        tm.assert_frame_equal(res, expect)
+
+    def test_character_outside_range_and_actual_backslash(self, df):
+        # GH 49633
+        res = df.query("`xy （z\\uff09` == 2")  # noqa: RUF001
+        expect = df[df["xy \uff08z\\uff09"] == 2]
+        tm.assert_frame_equal(res, expect)
+
+    def test_hashtag(self, df):
+        res = df.query("`foo#bar` > 4")
+        expect = df[df["foo#bar"] > 4]
+        tm.assert_frame_equal(res, expect)
+
+    def test_expr_with_column_name_with_hashtag_character(self):
+        # GH 59285
+        df = DataFrame((1, 2, 3), columns=["a#"])
+        result = df.query("`a#` < 2")
+        expected = df[df["a#"] < 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_comment(self):
+        # GH 59285
+        df = DataFrame((1, 2, 3), columns=["a#"])
+        result = df.query("`a#` < 2  # This is a comment")
+        expected = df[df["a#"] < 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_column_name_with_backtick_and_hash(self):
+        # GH 59285
+        df = DataFrame((1, 2, 3), columns=["a`#b"])
+        result = df.query("`a``#b` < 2")
+        expected = df[df["a`#b"] < 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_column_name_with_backtick(self):
+        # GH 59285
+        df = DataFrame({"a`b": (1, 2, 3), "ab": (4, 5, 6)})
+        result = df.query("`a``b` < 2")  # noqa
+        # Note: Formatting checks may wrongly consider the above ``inline code``.
+        expected = df[df["a`b"] < 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_string_with_backticks(self):
+        # GH 59285
+        df = DataFrame(("`", "`````", "``````````"), columns=["#backticks"])
+        result = df.query("'```' < `#backticks`")
+        expected = df["```" < df["#backticks"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_string_with_backticked_substring_same_as_column_name(self):
+        # GH 59285
+        df = DataFrame(("`", "`````", "``````````"), columns=["#backticks"])
+        result = df.query("'`#backticks`' < `#backticks`")
+        expected = df["`#backticks`" < df["#backticks"]]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "col1,col2,expr",
+        [
+            ("it's", "that's", "`it's` < `that's`"),
+            ('it"s', 'that"s', '`it"s` < `that"s`'),
+            ("it's", 'that\'s "nice"', "`it's` < `that's \"nice\"`"),
+            ("it's", "that's #cool", "`it's` < `that's #cool` # This is a comment"),
+        ],
+    )
+    def test_expr_with_column_names_with_special_characters(self, col1, col2, expr):
+        # GH 59285
+        df = DataFrame(
+            [
+                {col1: 1, col2: 2},
+                {col1: 3, col2: 4},
+                {col1: -1, col2: -2},
+                {col1: -3, col2: -4},
+            ]
+        )
+        result = df.query(expr)
+        expected = df[df[col1] < df[col2]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_no_backticks(self):
+        # GH 59285
+        df = DataFrame(("aaa", "vvv", "zzz"), columns=["column_name"])
+        result = df.query("'value' < column_name")
+        expected = df["value" < df["column_name"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_no_quotes_and_backtick_is_unmatched(self):
+        # GH 59285
+        df = DataFrame((1, 5, 10), columns=["column-name"])
+        with pytest.raises((SyntaxError, TokenError), match="invalid syntax"):
+            df.query("5 < `column-name")
+
+    def test_expr_with_no_quotes_and_backtick_is_matched(self):
+        # GH 59285
+        df = DataFrame((1, 5, 10), columns=["column-name"])
+        result = df.query("5 < `column-name`")
+        expected = df[5 < df["column-name"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_backtick_opened_before_quote_and_backtick_is_unmatched(self):
+        # GH 59285
+        df = DataFrame((1, 5, 10), columns=["It's"])
+        with pytest.raises(
+            (SyntaxError, TokenError), match="unterminated string literal"
+        ):
+            df.query("5 < `It's")
+
+    def test_expr_with_backtick_opened_before_quote_and_backtick_is_matched(self):
+        # GH 59285
+        df = DataFrame((1, 5, 10), columns=["It's"])
+        result = df.query("5 < `It's`")
+        expected = df[5 < df["It's"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_quote_opened_before_backtick_and_quote_is_unmatched(self):
+        # GH 59285
+        df = DataFrame(("aaa", "vvv", "zzz"), columns=["column-name"])
+        with pytest.raises(
+            (SyntaxError, TokenError), match="unterminated string literal"
+        ):
+            df.query("`column-name` < 'It`s that\\'s \"quote\" #hash")
+
+    def test_expr_with_quote_opened_before_backtick_and_quote_is_matched_at_end(self):
+        # GH 59285
+        df = DataFrame(("aaa", "vvv", "zzz"), columns=["column-name"])
+        result = df.query("`column-name` < 'It`s that\\'s \"quote\" #hash'")
+        expected = df[df["column-name"] < 'It`s that\'s "quote" #hash']
+        tm.assert_frame_equal(result, expected)
+
+    def test_expr_with_quote_opened_before_backtick_and_quote_is_matched_in_mid(self):
+        # GH 59285
+        df = DataFrame(("aaa", "vvv", "zzz"), columns=["column-name"])
+        result = df.query("'It`s that\\'s \"quote\" #hash' < `column-name`")
+        expected = df['It`s that\'s "quote" #hash' < df["column-name"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_call_non_named_expression(self, df):
+        """
+        Only attributes and variables ('named functions') can be called.
+        .__call__() is not an allowed attribute because that would allow
+        calling anything.
+        https://github.com/pandas-dev/pandas/pull/32460
+        """
+
+        def func(*_):
+            return 1
+
+        funcs = [func]  # noqa: F841
+
+        df.eval("@func()")
+
+        with pytest.raises(TypeError, match="Only named functions are supported"):
+            df.eval("@funcs[0]()")
+
+        with pytest.raises(TypeError, match="Only named functions are supported"):
+            df.eval("@funcs[0].__call__()")
+
+    def test_ea_dtypes(self, any_numeric_ea_and_arrow_dtype):
+        # GH#29618
+        df = DataFrame(
+            [[1, 2], [3, 4]], columns=["a", "b"], dtype=any_numeric_ea_and_arrow_dtype
+        )
+        warning = RuntimeWarning if NUMEXPR_INSTALLED else None
+        with tm.assert_produces_warning(warning):
+            result = df.eval("c = b - a")
+        expected = DataFrame(
+            [[1, 2, 1], [3, 4, 1]],
+            columns=["a", "b", "c"],
+            dtype=any_numeric_ea_and_arrow_dtype,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_ea_dtypes_and_scalar(self):
+        # GH#29618
+        df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"], dtype="Float64")
+        warning = RuntimeWarning if NUMEXPR_INSTALLED else None
+        with tm.assert_produces_warning(warning):
+            result = df.eval("c = b - 1")
+        expected = DataFrame(
+            [[1, 2, 1], [3, 4, 3]], columns=["a", "b", "c"], dtype="Float64"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_ea_dtypes_and_scalar_operation(self, any_numeric_ea_and_arrow_dtype):
+        # GH#29618
+        df = DataFrame(
+            [[1, 2], [3, 4]], columns=["a", "b"], dtype=any_numeric_ea_and_arrow_dtype
+        )
+        result = df.eval("c = 2 - 1")
+        expected = DataFrame(
+            {
+                "a": Series([1, 3], dtype=any_numeric_ea_and_arrow_dtype),
+                "b": Series([2, 4], dtype=any_numeric_ea_and_arrow_dtype),
+                "c": Series([1, 1], dtype=result["c"].dtype),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["int64", "Int64", "int64[pyarrow]"])
+    def test_query_ea_dtypes(self, dtype):
+        if dtype == "int64[pyarrow]":
+            pytest.importorskip("pyarrow")
+        # GH#50261
+        df = DataFrame({"a": [1, 2]}, dtype=dtype)
+        ref = {2}  # noqa: F841
+        warning = RuntimeWarning if dtype == "Int64" and NUMEXPR_INSTALLED else None
+        with tm.assert_produces_warning(warning):
+            result = df.query("a in @ref")
+        expected = DataFrame({"a": [2]}, index=range(1, 2), dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("engine", ["python", "numexpr"])
+    @pytest.mark.parametrize("dtype", ["int64", "Int64", "int64[pyarrow]"])
+    def test_query_ea_equality_comparison(self, dtype, engine):
+        # GH#50261
+        warning = RuntimeWarning if engine == "numexpr" else None
+        if engine == "numexpr" and not NUMEXPR_INSTALLED:
+            pytest.skip("numexpr not installed")
+        if dtype == "int64[pyarrow]":
+            pytest.importorskip("pyarrow")
+        df = DataFrame(
+            {"A": Series([1, 1, 2], dtype="Int64"), "B": Series([1, 2, 2], dtype=dtype)}
+        )
+        with tm.assert_produces_warning(warning):
+            result = df.query("A == B", engine=engine)
+        expected = DataFrame(
+            {
+                "A": Series([1, 2], dtype="Int64", index=range(0, 4, 2)),
+                "B": Series([1, 2], dtype=dtype, index=range(0, 4, 2)),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_all_nat_in_object(self):
+        # GH#57068
+        now = pd.Timestamp.now("UTC")  # noqa: F841
+        df = DataFrame({"a": pd.to_datetime([None, None], utc=True)}, dtype=object)
+        result = df.query("a > @now")
+        expected = DataFrame({"a": []}, dtype=object)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_reductions.py b/pandas/pandas/tests/frame/test_reductions.py
new file mode 100644
index 0000000000000..64e686d25faa7
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_reductions.py
@@ -0,0 +1,2131 @@
+from datetime import timedelta
+from decimal import Decimal
+import re
+
+from dateutil.tz import tzlocal
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+from pandas.compat.numpy import np_version_gt2
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+    notna,
+    to_datetime,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.core import (
+    algorithms,
+    nanops,
+)
+
+is_windows_np2_or_is32 = (is_platform_windows() and not np_version_gt2) or not IS64
+is_windows_or_is32 = is_platform_windows() or not IS64
+
+
+def make_skipna_wrapper(alternative, skipna_alternative=None):
+    """
+    Create a function for calling on an array.
+
+    Parameters
+    ----------
+    alternative : function
+        The function to be called on the array with no NaNs.
+        Only used when 'skipna_alternative' is None.
+    skipna_alternative : function
+        The function to be called on the original array
+
+    Returns
+    -------
+    function
+    """
+    if skipna_alternative:
+
+        def skipna_wrapper(x):
+            return skipna_alternative(x.values)
+
+    else:
+
+        def skipna_wrapper(x):
+            nona = x.dropna()
+            if len(nona) == 0:
+                return np.nan
+            return alternative(nona)
+
+    return skipna_wrapper
+
+
+def assert_stat_op_calc(
+    opname,
+    alternative,
+    frame,
+    has_skipna=True,
+    check_dtype=True,
+    check_dates=False,
+    rtol=1e-5,
+    atol=1e-8,
+    skipna_alternative=None,
+):
+    """
+    Check that operator opname works as advertised on frame
+
+    Parameters
+    ----------
+    opname : str
+        Name of the operator to test on frame
+    alternative : function
+        Function that opname is tested against; i.e. "frame.opname()" should
+        equal "alternative(frame)".
+    frame : DataFrame
+        The object that the tests are executed on
+    has_skipna : bool, default True
+        Whether the method "opname" has the kwarg "skip_na"
+    check_dtype : bool, default True
+        Whether the dtypes of the result of "frame.opname()" and
+        "alternative(frame)" should be checked.
+    check_dates : bool, default false
+        Whether opname should be tested on a Datetime Series
+    rtol : float, default 1e-5
+        Relative tolerance.
+    atol : float, default 1e-8
+        Absolute tolerance.
+    skipna_alternative : function, default None
+        NaN-safe version of alternative
+    """
+    f = getattr(frame, opname)
+
+    if check_dates:
+        df = DataFrame({"b": date_range("1/1/2001", periods=2)})
+        with tm.assert_produces_warning(None):
+            result = getattr(df, opname)()
+        assert isinstance(result, Series)
+
+        df["a"] = range(len(df))
+        with tm.assert_produces_warning(None):
+            result = getattr(df, opname)()
+        assert isinstance(result, Series)
+        assert len(result)
+
+    if has_skipna:
+
+        def wrapper(x):
+            return alternative(x.values)
+
+        skipna_wrapper = make_skipna_wrapper(alternative, skipna_alternative)
+        result0 = f(axis=0, skipna=False)
+        result1 = f(axis=1, skipna=False)
+        tm.assert_series_equal(
+            result0, frame.apply(wrapper), check_dtype=check_dtype, rtol=rtol, atol=atol
+        )
+        tm.assert_series_equal(
+            result1,
+            frame.apply(wrapper, axis=1),
+            rtol=rtol,
+            atol=atol,
+        )
+    else:
+        skipna_wrapper = alternative
+
+    result0 = f(axis=0)
+    result1 = f(axis=1)
+    tm.assert_series_equal(
+        result0,
+        frame.apply(skipna_wrapper),
+        check_dtype=check_dtype,
+        rtol=rtol,
+        atol=atol,
+    )
+
+    if opname in ["sum", "prod"]:
+        expected = frame.apply(skipna_wrapper, axis=1)
+        tm.assert_series_equal(
+            result1, expected, check_dtype=False, rtol=rtol, atol=atol
+        )
+
+    # check dtypes
+    if check_dtype:
+        lcd_dtype = frame.values.dtype
+        assert lcd_dtype == result0.dtype
+        assert lcd_dtype == result1.dtype
+
+    # bad axis
+    with pytest.raises(ValueError, match="No axis named 2"):
+        f(axis=2)
+
+    # all NA case
+    if has_skipna:
+        all_na = frame * np.nan
+        r0 = getattr(all_na, opname)(axis=0)
+        r1 = getattr(all_na, opname)(axis=1)
+        if opname in ["sum", "prod"]:
+            unit = 1 if opname == "prod" else 0  # result for empty sum/prod
+            expected = Series(unit, index=r0.index, dtype=r0.dtype)
+            tm.assert_series_equal(r0, expected)
+            expected = Series(unit, index=r1.index, dtype=r1.dtype)
+            tm.assert_series_equal(r1, expected)
+
+
+@pytest.fixture
+def bool_frame_with_na():
+    """
+    Fixture for DataFrame of booleans with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D']; some entries are missing
+    """
+    df = DataFrame(
+        np.concatenate(
+            [np.ones((15, 4), dtype=bool), np.zeros((15, 4), dtype=bool)], axis=0
+        ),
+        index=Index([f"foo_{i}" for i in range(30)], dtype=object),
+        columns=Index(list("ABCD"), dtype=object),
+        dtype=object,
+    )
+    # set some NAs
+    df.iloc[5:10] = np.nan
+    df.iloc[15:20, -2:] = np.nan
+    return df
+
+
+@pytest.fixture
+def float_frame_with_na():
+    """
+    Fixture for DataFrame of floats with index of unique strings
+
+    Columns are ['A', 'B', 'C', 'D']; some entries are missing
+    """
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        index=Index([f"foo_{i}" for i in range(30)], dtype=object),
+        columns=Index(list("ABCD"), dtype=object),
+    )
+    # set some NAs
+    df.iloc[5:10] = np.nan
+    df.iloc[15:20, -2:] = np.nan
+    return df
+
+
+class TestDataFrameAnalytics:
+    # ---------------------------------------------------------------------
+    # Reductions
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.parametrize(
+        "opname",
+        [
+            "count",
+            "sum",
+            "mean",
+            "product",
+            "median",
+            "min",
+            "max",
+            "nunique",
+            "var",
+            "std",
+            "sem",
+            pytest.param("skew", marks=td.skip_if_no("scipy")),
+            pytest.param("kurt", marks=td.skip_if_no("scipy")),
+        ],
+    )
+    def test_stat_op_api_float_string_frame(self, float_string_frame, axis, opname):
+        if (opname in ("sum", "min", "max") and axis == 0) or opname in (
+            "count",
+            "nunique",
+        ):
+            getattr(float_string_frame, opname)(axis=axis)
+        else:
+            if opname in ["var", "std", "sem", "skew", "kurt"]:
+                msg = "could not convert string to float: 'bar'"
+            elif opname == "product":
+                if axis == 1:
+                    msg = "can't multiply sequence by non-int of type 'float'"
+                else:
+                    msg = "can't multiply sequence by non-int of type 'str'"
+            elif opname == "sum":
+                msg = r"unsupported operand type\(s\) for \+: 'float' and 'str'"
+            elif opname == "mean":
+                if axis == 0:
+                    # different message on different builds
+                    msg = "|".join(
+                        [
+                            r"Could not convert \['.*'\] to numeric",
+                            "Could not convert string '(bar){30}' to numeric",
+                        ]
+                    )
+                else:
+                    msg = r"unsupported operand type\(s\) for \+: 'float' and 'str'"
+            elif opname in ["min", "max"]:
+                msg = "'[><]=' not supported between instances of 'float' and 'str'"
+            elif opname == "median":
+                msg = re.compile(
+                    r"Cannot convert \[.*\] to numeric|does not support|Cannot perform",
+                    flags=re.S,
+                )
+            if not isinstance(msg, re.Pattern):
+                msg = msg + "|does not support|Cannot perform reduction"
+            with pytest.raises(TypeError, match=msg):
+                getattr(float_string_frame, opname)(axis=axis)
+        if opname != "nunique":
+            getattr(float_string_frame, opname)(axis=axis, numeric_only=True)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.parametrize(
+        "opname",
+        [
+            "count",
+            "sum",
+            "mean",
+            "product",
+            "median",
+            "min",
+            "max",
+            "var",
+            "std",
+            "sem",
+            pytest.param("skew", marks=td.skip_if_no("scipy")),
+            pytest.param("kurt", marks=td.skip_if_no("scipy")),
+        ],
+    )
+    def test_stat_op_api_float_frame(self, float_frame, axis, opname):
+        getattr(float_frame, opname)(axis=axis, numeric_only=False)
+
+    def test_stat_op_calc(self, float_frame_with_na, mixed_float_frame):
+        def count(s):
+            return notna(s).sum()
+
+        def nunique(s):
+            return len(algorithms.unique1d(s.dropna()))
+
+        def var(x):
+            return np.var(x, ddof=1)
+
+        def std(x):
+            return np.std(x, ddof=1)
+
+        def sem(x):
+            return np.std(x, ddof=1) / np.sqrt(len(x))
+
+        assert_stat_op_calc(
+            "nunique",
+            nunique,
+            float_frame_with_na,
+            has_skipna=False,
+            check_dtype=False,
+            check_dates=True,
+        )
+
+        # GH#32571: rol needed for flaky CI builds
+        # mixed types (with upcasting happening)
+        assert_stat_op_calc(
+            "sum",
+            np.sum,
+            mixed_float_frame.astype("float32"),
+            check_dtype=False,
+            rtol=1e-3,
+        )
+
+        assert_stat_op_calc(
+            "sum", np.sum, float_frame_with_na, skipna_alternative=np.nansum
+        )
+        assert_stat_op_calc("mean", np.mean, float_frame_with_na, check_dates=True)
+        assert_stat_op_calc(
+            "product", np.prod, float_frame_with_na, skipna_alternative=np.nanprod
+        )
+
+        assert_stat_op_calc("var", var, float_frame_with_na)
+        assert_stat_op_calc("std", std, float_frame_with_na)
+        assert_stat_op_calc("sem", sem, float_frame_with_na)
+
+        assert_stat_op_calc(
+            "count",
+            count,
+            float_frame_with_na,
+            has_skipna=False,
+            check_dtype=False,
+            check_dates=True,
+        )
+
+    def test_stat_op_calc_skew_kurtosis(self, float_frame_with_na):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        def skewness(x):
+            if len(x) < 3:
+                return np.nan
+            return sp_stats.skew(x, bias=False)
+
+        def kurt(x):
+            if len(x) < 4:
+                return np.nan
+            return sp_stats.kurtosis(x, bias=False)
+
+        assert_stat_op_calc("skew", skewness, float_frame_with_na)
+        assert_stat_op_calc("kurt", kurt, float_frame_with_na)
+
+    def test_median(self, float_frame_with_na, int_frame):
+        def wrapper(x):
+            if isna(x).any():
+                return np.nan
+            return np.median(x)
+
+        assert_stat_op_calc("median", wrapper, float_frame_with_na, check_dates=True)
+        assert_stat_op_calc(
+            "median", wrapper, int_frame, check_dtype=False, check_dates=True
+        )
+
+    @pytest.mark.parametrize(
+        "method", ["sum", "mean", "prod", "var", "std", "skew", "min", "max"]
+    )
+    @pytest.mark.parametrize(
+        "df",
+        [
+            DataFrame(
+                {
+                    "a": [
+                        -0.00049987540199591344,
+                        -0.0016467257772919831,
+                        0.00067695870775883013,
+                    ],
+                    "b": [-0, -0, 0.0],
+                    "c": [
+                        0.00031111847529610595,
+                        0.0014902627951905339,
+                        -0.00094099200035979691,
+                    ],
+                },
+                index=["foo", "bar", "baz"],
+                dtype="O",
+            ),
+            DataFrame({0: [np.nan, 2], 1: [np.nan, 3], 2: [np.nan, 4]}, dtype=object),
+        ],
+    )
+    @pytest.mark.filterwarnings("ignore:Mismatched null-like values:FutureWarning")
+    def test_stat_operators_attempt_obj_array(self, method, df, axis):
+        # GH#676
+        assert df.values.dtype == np.object_
+        result = getattr(df, method)(axis=axis)
+        expected = getattr(df.astype("f8"), method)(axis=axis).astype(object)
+        if axis in [1, "columns"] and method in ["min", "max"]:
+            expected[expected.isna()] = None
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("op", ["mean", "std", "var", "skew", "kurt", "sem"])
+    def test_mixed_ops(self, op):
+        # GH#16116
+        df = DataFrame(
+            {
+                "int": [1, 2, 3, 4],
+                "float": [1.0, 2.0, 3.0, 4.0],
+                "str": ["a", "b", "c", "d"],
+            }
+        )
+        msg = "|".join(
+            [
+                "Could not convert",
+                "could not convert",
+                "can't multiply sequence by non-int",
+                "does not support",
+                "Cannot perform",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            getattr(df, op)()
+
+        with pd.option_context("use_bottleneck", False):
+            with pytest.raises(TypeError, match=msg):
+                getattr(df, op)()
+
+    def test_reduce_mixed_frame(self):
+        # GH 6806
+        df = DataFrame(
+            {
+                "bool_data": [True, True, False, False, False],
+                "int_data": [10, 20, 30, 40, 50],
+                "string_data": ["a", "b", "c", "d", "e"],
+            }
+        )
+        df.reindex(columns=["bool_data", "int_data", "string_data"])
+        test = df.sum(axis=0)
+        tm.assert_numpy_array_equal(
+            test.values, np.array([2, 150, "abcde"], dtype=object)
+        )
+        alt = df.T.sum(axis=1)
+        tm.assert_series_equal(test, alt)
+
+    def test_nunique(self):
+        df = DataFrame({"A": [1, 1, 1], "B": [1, 2, 3], "C": [1, np.nan, 3]})
+        tm.assert_series_equal(df.nunique(), Series({"A": 1, "B": 3, "C": 2}))
+        tm.assert_series_equal(
+            df.nunique(dropna=False), Series({"A": 1, "B": 3, "C": 3})
+        )
+        tm.assert_series_equal(df.nunique(axis=1), Series([1, 2, 2]))
+        tm.assert_series_equal(df.nunique(axis=1, dropna=False), Series([1, 3, 2]))
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_mean_mixed_datetime_numeric(self, tz):
+        # https://github.com/pandas-dev/pandas/issues/24752
+        df = DataFrame({"A": [1, 1], "B": [Timestamp("2000", tz=tz)] * 2})
+        result = df.mean()
+        expected = Series([1.0, Timestamp("2000", tz=tz)], index=["A", "B"])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_mean_includes_datetimes(self, tz):
+        # https://github.com/pandas-dev/pandas/issues/24752
+        # Behavior in 0.24.0rc1 was buggy.
+        # As of 2.0 with numeric_only=None we do *not* drop datetime columns
+        df = DataFrame({"A": [Timestamp("2000", tz=tz)] * 2})
+        result = df.mean()
+
+        expected = Series([Timestamp("2000", tz=tz)], index=["A"])
+        tm.assert_series_equal(result, expected)
+
+    def test_mean_mixed_string_decimal(self):
+        # GH 11670
+        # possible bug when calculating mean of DataFrame?
+
+        d = [
+            {"A": 2, "B": None, "C": Decimal("628.00")},
+            {"A": 1, "B": None, "C": Decimal("383.00")},
+            {"A": 3, "B": None, "C": Decimal("651.00")},
+            {"A": 2, "B": None, "C": Decimal("575.00")},
+            {"A": 4, "B": None, "C": Decimal("1114.00")},
+            {"A": 1, "B": "TEST", "C": Decimal("241.00")},
+            {"A": 2, "B": None, "C": Decimal("572.00")},
+            {"A": 4, "B": None, "C": Decimal("609.00")},
+            {"A": 3, "B": None, "C": Decimal("820.00")},
+            {"A": 5, "B": None, "C": Decimal("1223.00")},
+        ]
+
+        df = DataFrame(d)
+
+        with pytest.raises(
+            TypeError, match="unsupported operand type|does not support|Cannot perform"
+        ):
+            df.mean()
+        result = df[["A", "C"]].mean()
+        expected = Series([2.7, 681.6], index=["A", "C"], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    def test_var_std(self, datetime_frame):
+        result = datetime_frame.std(ddof=4)
+        expected = datetime_frame.apply(lambda x: x.std(ddof=4))
+        tm.assert_almost_equal(result, expected)
+
+        result = datetime_frame.var(ddof=4)
+        expected = datetime_frame.apply(lambda x: x.var(ddof=4))
+        tm.assert_almost_equal(result, expected)
+
+        arr = np.repeat(np.random.default_rng(2).random((1, 1000)), 1000, 0)
+        result = nanops.nanvar(arr, axis=0)
+        assert not (result < 0).any()
+
+        with pd.option_context("use_bottleneck", False):
+            result = nanops.nanvar(arr, axis=0)
+            assert not (result < 0).any()
+
+    @pytest.mark.parametrize("meth", ["sem", "var", "std"])
+    def test_numeric_only_flag(self, meth):
+        # GH 9201
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            columns=["foo", "bar", "baz"],
+        )
+        # Cast to object to avoid implicit cast when setting entry to "100" below
+        df1 = df1.astype({"foo": object})
+        # set one entry to a number in str format
+        df1.loc[0, "foo"] = "100"
+
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            columns=["foo", "bar", "baz"],
+        )
+        # Cast to object to avoid implicit cast when setting entry to "a" below
+        df2 = df2.astype({"foo": object})
+        # set one entry to a non-number str
+        df2.loc[0, "foo"] = "a"
+
+        result = getattr(df1, meth)(axis=1, numeric_only=True)
+        expected = getattr(df1[["bar", "baz"]], meth)(axis=1)
+        tm.assert_series_equal(expected, result)
+
+        result = getattr(df2, meth)(axis=1, numeric_only=True)
+        expected = getattr(df2[["bar", "baz"]], meth)(axis=1)
+        tm.assert_series_equal(expected, result)
+
+        # df1 has all numbers, df2 has a letter inside
+        msg = r"unsupported operand type\(s\) for -: 'float' and 'str'"
+        with pytest.raises(TypeError, match=msg):
+            getattr(df1, meth)(axis=1, numeric_only=False)
+        msg = "could not convert string to float: 'a'"
+        with pytest.raises(TypeError, match=msg):
+            getattr(df2, meth)(axis=1, numeric_only=False)
+
+    def test_sem(self, datetime_frame):
+        result = datetime_frame.sem(ddof=4)
+        expected = datetime_frame.apply(lambda x: x.std(ddof=4) / np.sqrt(len(x)))
+        tm.assert_almost_equal(result, expected)
+
+        arr = np.repeat(np.random.default_rng(2).random((1, 1000)), 1000, 0)
+        result = nanops.nansem(arr, axis=0)
+        assert not (result < 0).any()
+
+        with pd.option_context("use_bottleneck", False):
+            result = nanops.nansem(arr, axis=0)
+            assert not (result < 0).any()
+
+    @pytest.mark.parametrize(
+        "dropna, expected",
+        [
+            (
+                True,
+                {
+                    "A": [12],
+                    "B": [10.0],
+                    "C": [1.0],
+                    "D": ["a"],
+                    "E": Categorical(["a"], categories=["a"]),
+                    "F": DatetimeIndex(["2000-01-02"], dtype="M8[ns]"),
+                    "G": to_timedelta(["1 days"]),
+                },
+            ),
+            (
+                False,
+                {
+                    "A": [12],
+                    "B": [10.0],
+                    "C": [np.nan],
+                    "D": Series([np.nan], dtype="str"),
+                    "E": Categorical([np.nan], categories=["a"]),
+                    "F": DatetimeIndex([pd.NaT], dtype="M8[ns]"),
+                    "G": to_timedelta([pd.NaT]),
+                },
+            ),
+            (
+                True,
+                {
+                    "H": [8, 9, np.nan, np.nan],
+                    "I": [8, 9, np.nan, np.nan],
+                    "J": [1, np.nan, np.nan, np.nan],
+                    "K": Categorical(["a", np.nan, np.nan, np.nan], categories=["a"]),
+                    "L": DatetimeIndex(
+                        ["2000-01-02", "NaT", "NaT", "NaT"], dtype="M8[ns]"
+                    ),
+                    "M": to_timedelta(["1 days", "nan", "nan", "nan"]),
+                    "N": [0, 1, 2, 3],
+                },
+            ),
+            (
+                False,
+                {
+                    "H": [8, 9, np.nan, np.nan],
+                    "I": [8, 9, np.nan, np.nan],
+                    "J": [1, np.nan, np.nan, np.nan],
+                    "K": Categorical([np.nan, "a", np.nan, np.nan], categories=["a"]),
+                    "L": DatetimeIndex(
+                        ["NaT", "2000-01-02", "NaT", "NaT"], dtype="M8[ns]"
+                    ),
+                    "M": to_timedelta(["nan", "1 days", "nan", "nan"]),
+                    "N": [0, 1, 2, 3],
+                },
+            ),
+        ],
+    )
+    def test_mode_dropna(self, dropna, expected):
+        df = DataFrame(
+            {
+                "A": [12, 12, 19, 11],
+                "B": [10, 10, np.nan, 3],
+                "C": [1, np.nan, np.nan, np.nan],
+                "D": Series([np.nan, np.nan, "a", np.nan], dtype="str"),
+                "E": Categorical([np.nan, np.nan, "a", np.nan]),
+                "F": DatetimeIndex(["NaT", "2000-01-02", "NaT", "NaT"], dtype="M8[ns]"),
+                "G": to_timedelta(["1 days", "nan", "nan", "nan"]),
+                "H": [8, 8, 9, 9],
+                "I": [9, 9, 8, 8],
+                "J": [1, 1, np.nan, np.nan],
+                "K": Categorical(["a", np.nan, "a", np.nan]),
+                "L": DatetimeIndex(
+                    ["2000-01-02", "2000-01-02", "NaT", "NaT"], dtype="M8[ns]"
+                ),
+                "M": to_timedelta(["1 days", "nan", "1 days", "nan"]),
+                "N": np.arange(4, dtype="int64"),
+            }
+        )
+
+        result = df[sorted(expected.keys())].mode(dropna=dropna)
+        expected = DataFrame(expected)
+        tm.assert_frame_equal(result, expected)
+
+    def test_mode_sort_with_na(self, using_infer_string):
+        df = DataFrame({"A": [np.nan, np.nan, "a", "a"]})
+        expected = DataFrame({"A": ["a", np.nan]})
+        result = df.mode(dropna=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_mode_empty_df(self):
+        df = DataFrame([], columns=["a", "b"])
+        expected = df.copy()
+        result = df.mode()
+        tm.assert_frame_equal(result, expected)
+
+    def test_operators_timedelta64(self):
+        df = DataFrame(
+            {
+                "A": date_range("2012-1-1", periods=3, freq="D"),
+                "B": date_range("2012-1-2", periods=3, freq="D"),
+                "C": Timestamp("20120101") - timedelta(minutes=5, seconds=5),
+            }
+        )
+
+        diffs = DataFrame({"A": df["A"] - df["C"], "B": df["A"] - df["B"]})
+
+        # min
+        result = diffs.min()
+        assert result.iloc[0] == diffs.loc[0, "A"]
+        assert result.iloc[1] == diffs.loc[0, "B"]
+
+        result = diffs.min(axis=1)
+        assert (result == diffs.loc[0, "B"]).all()
+
+        # max
+        result = diffs.max()
+        assert result.iloc[0] == diffs.loc[2, "A"]
+        assert result.iloc[1] == diffs.loc[2, "B"]
+
+        result = diffs.max(axis=1)
+        assert (result == diffs["A"]).all()
+
+        # abs
+        result = diffs.abs()
+        result2 = abs(diffs)
+        expected = DataFrame({"A": df["A"] - df["C"], "B": df["B"] - df["A"]})
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        # mixed frame
+        mixed = diffs.copy()
+        mixed["C"] = "foo"
+        mixed["D"] = 1
+        mixed["E"] = 1.0
+        mixed["F"] = Timestamp("20130101")
+
+        # results in an object array
+        result = mixed.min()
+        expected = Series(
+            [
+                pd.Timedelta(timedelta(seconds=5 * 60 + 5)),
+                pd.Timedelta(timedelta(days=-1)),
+                "foo",
+                1,
+                1.0,
+                Timestamp("20130101"),
+            ],
+            index=mixed.columns,
+        )
+        tm.assert_series_equal(result, expected)
+
+        # excludes non-numeric
+        result = mixed.min(axis=1, numeric_only=True)
+        expected = Series([1, 1, 1.0])
+        tm.assert_series_equal(result, expected)
+
+        # works when only those columns are selected
+        result = mixed[["A", "B"]].min(axis=1)
+        expected = Series([timedelta(days=-1)] * 3)
+        tm.assert_series_equal(result, expected)
+
+        result = mixed[["A", "B"]].min()
+        expected = Series(
+            [timedelta(seconds=5 * 60 + 5), timedelta(days=-1)], index=["A", "B"]
+        )
+        tm.assert_series_equal(result, expected)
+
+        # GH 3106
+        df = DataFrame(
+            {
+                "time": date_range("20130102", periods=5),
+                "time2": date_range("20130105", periods=5),
+            }
+        )
+        df["off1"] = df["time2"] - df["time"]
+        assert df["off1"].dtype == "timedelta64[ns]"
+
+        df["off2"] = df["time"] - df["time2"]
+        df._consolidate_inplace()
+        assert df["off1"].dtype == "timedelta64[ns]"
+        assert df["off2"].dtype == "timedelta64[ns]"
+
+    def test_std_timedelta64_skipna_false(self):
+        # GH#37392
+        tdi = pd.timedelta_range("1 Day", periods=10)
+        df = DataFrame({"A": tdi, "B": tdi}, copy=True)
+        df.iloc[-2, -1] = pd.NaT
+
+        result = df.std(skipna=False)
+        expected = Series(
+            [df["A"].std(), pd.NaT], index=["A", "B"], dtype="timedelta64[ns]"
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = df.std(axis=1, skipna=False)
+        expected = Series([pd.Timedelta(0)] * 8 + [pd.NaT, pd.Timedelta(0)])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values", [["2022-01-01", "2022-01-02", pd.NaT, "2022-01-03"], 4 * [pd.NaT]]
+    )
+    def test_std_datetime64_with_nat(self, values, skipna, request, unit):
+        # GH#51335
+        dti = to_datetime(values).as_unit(unit)
+        df = DataFrame({"a": dti})
+        result = df.std(skipna=skipna)
+        if not skipna or all(value is pd.NaT for value in values):
+            expected = Series({"a": pd.NaT}, dtype=f"timedelta64[{unit}]")
+        else:
+            # 86400000000000ns == 1 day
+            expected = Series({"a": 86400000000000}, dtype=f"timedelta64[{unit}]")
+        tm.assert_series_equal(result, expected)
+
+    def test_sum_corner(self):
+        empty_frame = DataFrame()
+
+        axis0 = empty_frame.sum(axis=0)
+        axis1 = empty_frame.sum(axis=1)
+        assert isinstance(axis0, Series)
+        assert isinstance(axis1, Series)
+        assert len(axis0) == 0
+        assert len(axis1) == 0
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            RangeIndex(0),
+            DatetimeIndex([]),
+            Index([], dtype=np.int64),
+            Index([], dtype=np.float64),
+            DatetimeIndex([], freq="ME"),
+            PeriodIndex([], freq="D"),
+        ],
+    )
+    def test_axis_1_empty(self, all_reductions, index):
+        df = DataFrame(columns=["a"], index=index)
+        result = getattr(df, all_reductions)(axis=1)
+        if all_reductions in ("any", "all"):
+            expected_dtype = "bool"
+        elif all_reductions == "count":
+            expected_dtype = "int64"
+        else:
+            expected_dtype = "object"
+        expected = Series([], index=index, dtype=expected_dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("method, unit", [("sum", 0), ("prod", 1)])
+    @pytest.mark.parametrize("numeric_only", [None, True, False])
+    def test_sum_prod_nanops(self, method, unit, numeric_only):
+        idx = ["a", "b", "c"]
+        df = DataFrame({"a": [unit, unit], "b": [unit, np.nan], "c": [np.nan, np.nan]})
+        # The default
+        result = getattr(df, method)(numeric_only=numeric_only)
+        expected = Series([unit, unit, unit], index=idx, dtype="float64")
+        tm.assert_series_equal(result, expected)
+
+        # min_count=1
+        result = getattr(df, method)(numeric_only=numeric_only, min_count=1)
+        expected = Series([unit, unit, np.nan], index=idx)
+        tm.assert_series_equal(result, expected)
+
+        # min_count=0
+        result = getattr(df, method)(numeric_only=numeric_only, min_count=0)
+        expected = Series([unit, unit, unit], index=idx, dtype="float64")
+        tm.assert_series_equal(result, expected)
+
+        result = getattr(df.iloc[1:], method)(numeric_only=numeric_only, min_count=1)
+        expected = Series([unit, np.nan, np.nan], index=idx)
+        tm.assert_series_equal(result, expected)
+
+        # min_count > 1
+        df = DataFrame({"A": [unit] * 10, "B": [unit] * 5 + [np.nan] * 5})
+        result = getattr(df, method)(numeric_only=numeric_only, min_count=5)
+        expected = Series(result, index=["A", "B"])
+        tm.assert_series_equal(result, expected)
+
+        result = getattr(df, method)(numeric_only=numeric_only, min_count=6)
+        expected = Series(result, index=["A", "B"])
+        tm.assert_series_equal(result, expected)
+
+    def test_sum_nanops_timedelta(self):
+        # prod isn't defined on timedeltas
+        idx = ["a", "b", "c"]
+        df = DataFrame({"a": [0, 0], "b": [0, np.nan], "c": [np.nan, np.nan]})
+
+        df2 = df.apply(to_timedelta)
+
+        # 0 by default
+        result = df2.sum()
+        expected = Series([0, 0, 0], dtype="m8[ns]", index=idx)
+        tm.assert_series_equal(result, expected)
+
+        # min_count=0
+        result = df2.sum(min_count=0)
+        tm.assert_series_equal(result, expected)
+
+        # min_count=1
+        result = df2.sum(min_count=1)
+        expected = Series([0, 0, np.nan], dtype="m8[ns]", index=idx)
+        tm.assert_series_equal(result, expected)
+
+    def test_sum_nanops_min_count(self):
+        # https://github.com/pandas-dev/pandas/issues/39738
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+        result = df.sum(min_count=10)
+        expected = Series([np.nan, np.nan], index=["x", "y"])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("float_type", ["float16", "float32", "float64"])
+    @pytest.mark.parametrize(
+        "kwargs, expected_result",
+        [
+            ({"axis": 1, "min_count": 2}, [3.2, 5.3, np.nan]),
+            ({"axis": 1, "min_count": 3}, [np.nan, np.nan, np.nan]),
+            ({"axis": 1, "skipna": False}, [3.2, 5.3, np.nan]),
+        ],
+    )
+    def test_sum_nanops_dtype_min_count(self, float_type, kwargs, expected_result):
+        # GH#46947
+        df = DataFrame({"a": [1.0, 2.3, 4.4], "b": [2.2, 3, np.nan]}, dtype=float_type)
+        result = df.sum(**kwargs)
+        expected = Series(expected_result).astype(float_type)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("float_type", ["float16", "float32", "float64"])
+    @pytest.mark.parametrize(
+        "kwargs, expected_result",
+        [
+            ({"axis": 1, "min_count": 2}, [2.0, 4.0, np.nan]),
+            ({"axis": 1, "min_count": 3}, [np.nan, np.nan, np.nan]),
+            ({"axis": 1, "skipna": False}, [2.0, 4.0, np.nan]),
+        ],
+    )
+    def test_prod_nanops_dtype_min_count(self, float_type, kwargs, expected_result):
+        # GH#46947
+        df = DataFrame(
+            {"a": [1.0, 2.0, 4.4], "b": [2.0, 2.0, np.nan]}, dtype=float_type
+        )
+        result = df.prod(**kwargs)
+        expected = Series(expected_result).astype(float_type)
+        tm.assert_series_equal(result, expected)
+
+    def test_sum_object(self, float_frame):
+        values = float_frame.values.astype(int)
+        frame = DataFrame(values, index=float_frame.index, columns=float_frame.columns)
+        deltas = frame * timedelta(1)
+        deltas.sum()
+
+    def test_sum_bool(self, float_frame):
+        # ensure this works, bug report
+        bools = np.isnan(float_frame)
+        bools.sum(axis=1)
+        bools.sum(axis=0)
+
+    def test_sum_mixed_datetime(self):
+        # GH#30886
+        df = DataFrame({"A": date_range("2000", periods=4), "B": [1, 2, 3, 4]}).reindex(
+            [2, 3, 4]
+        )
+        with pytest.raises(TypeError, match="does not support operation 'sum'"):
+            df.sum()
+
+    def test_mean_corner(self, float_frame, float_string_frame):
+        # unit test when have object data
+        msg = "Could not convert|does not support|Cannot perform"
+        with pytest.raises(TypeError, match=msg):
+            float_string_frame.mean(axis=0)
+
+        # xs sum mixed type, just want to know it works...
+        with pytest.raises(TypeError, match="unsupported operand type"):
+            float_string_frame.mean(axis=1)
+
+        # take mean of boolean column
+        float_frame["bool"] = float_frame["A"] > 0
+        means = float_frame.mean(axis=0)
+        assert means["bool"] == float_frame["bool"].values.mean()
+
+    def test_mean_datetimelike(self):
+        # GH#24757 check that datetimelike are excluded by default, handled
+        #  correctly with numeric_only=True
+        #  As of 2.0, datetimelike are *not* excluded with numeric_only=None
+
+        df = DataFrame(
+            {
+                "A": np.arange(3),
+                "B": date_range("2016-01-01", periods=3),
+                "C": pd.timedelta_range("1D", periods=3),
+                "D": pd.period_range("2016", periods=3, freq="Y"),
+            }
+        )
+        result = df.mean(numeric_only=True)
+        expected = Series({"A": 1.0})
+        tm.assert_series_equal(result, expected)
+
+        with pytest.raises(TypeError, match="mean is not implemented for PeriodArray"):
+            df.mean()
+
+    def test_mean_datetimelike_numeric_only_false(self):
+        df = DataFrame(
+            {
+                "A": np.arange(3),
+                "B": date_range("2016-01-01", periods=3),
+                "C": pd.timedelta_range("1D", periods=3),
+            }
+        )
+
+        # datetime(tz) and timedelta work
+        result = df.mean(numeric_only=False)
+        expected = Series({"A": 1, "B": df.loc[1, "B"], "C": df.loc[1, "C"]})
+        tm.assert_series_equal(result, expected)
+
+        # mean of period is not allowed
+        df["D"] = pd.period_range("2016", periods=3, freq="Y")
+
+        with pytest.raises(TypeError, match="mean is not implemented for Period"):
+            df.mean(numeric_only=False)
+
+    def test_mean_extensionarray_numeric_only_true(self):
+        # https://github.com/pandas-dev/pandas/issues/33256
+        arr = np.random.default_rng(2).integers(1000, size=(10, 5))
+        df = DataFrame(arr, dtype="Int64")
+        result = df.mean(numeric_only=True)
+        expected = DataFrame(arr).mean().astype("Float64")
+        tm.assert_series_equal(result, expected)
+
+    def test_stats_mixed_type(self, float_string_frame):
+        with pytest.raises(TypeError, match="could not convert"):
+            float_string_frame.std(axis=1)
+        with pytest.raises(TypeError, match="could not convert"):
+            float_string_frame.var(axis=1)
+        with pytest.raises(TypeError, match="unsupported operand type"):
+            float_string_frame.mean(axis=1)
+        with pytest.raises(TypeError, match="could not convert"):
+            float_string_frame.skew(axis=1)
+
+    def test_sum_bools(self):
+        df = DataFrame(index=range(1), columns=range(10))
+        bools = isna(df)
+        assert bools.sum(axis=1)[0] == 10
+
+    # ----------------------------------------------------------------------
+    # Index of max / min
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_idxmin(self, float_frame, int_frame, skipna, axis):
+        frame = float_frame
+        frame.iloc[5:10] = np.nan
+        frame.iloc[15:20, -2:] = np.nan
+        for df in [frame, int_frame]:
+            if (not skipna or axis == 1) and df is not int_frame:
+                if skipna:
+                    msg = "Encountered all NA values"
+                else:
+                    msg = "Encountered an NA value"
+                with pytest.raises(ValueError, match=msg):
+                    df.idxmin(axis=axis, skipna=skipna)
+                with pytest.raises(ValueError, match=msg):
+                    df.idxmin(axis=axis, skipna=skipna)
+            else:
+                result = df.idxmin(axis=axis, skipna=skipna)
+                expected = df.apply(Series.idxmin, axis=axis, skipna=skipna)
+                expected = expected.astype(df.index.dtype)
+                tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_idxmin_empty(self, index, skipna, axis):
+        # GH53265
+        if axis == 0:
+            frame = DataFrame(index=index)
+        else:
+            frame = DataFrame(columns=index)
+
+        result = frame.idxmin(axis=axis, skipna=skipna)
+        expected = Series(dtype=index.dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("numeric_only", [True, False])
+    def test_idxmin_numeric_only(self, numeric_only):
+        df = DataFrame({"a": [2, 3, 1], "b": [2, 1, 1], "c": list("xyx")})
+        result = df.idxmin(numeric_only=numeric_only)
+        if numeric_only:
+            expected = Series([2, 1], index=["a", "b"])
+        else:
+            expected = Series([2, 1, 0], index=["a", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+    def test_idxmin_axis_2(self, float_frame):
+        frame = float_frame
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            frame.idxmin(axis=2)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_idxmax(self, float_frame, int_frame, skipna, axis):
+        frame = float_frame
+        frame.iloc[5:10] = np.nan
+        frame.iloc[15:20, -2:] = np.nan
+        for df in [frame, int_frame]:
+            if (skipna is False or axis == 1) and df is frame:
+                if skipna:
+                    msg = "Encountered all NA values"
+                else:
+                    msg = "Encountered an NA value"
+                with pytest.raises(ValueError, match=msg):
+                    df.idxmax(axis=axis, skipna=skipna)
+                return
+
+            result = df.idxmax(axis=axis, skipna=skipna)
+            expected = df.apply(Series.idxmax, axis=axis, skipna=skipna)
+            expected = expected.astype(df.index.dtype)
+            tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_idxmax_empty(self, index, skipna, axis):
+        # GH53265
+        if axis == 0:
+            frame = DataFrame(index=index)
+        else:
+            frame = DataFrame(columns=index)
+
+        result = frame.idxmax(axis=axis, skipna=skipna)
+        expected = Series(dtype=index.dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("numeric_only", [True, False])
+    def test_idxmax_numeric_only(self, numeric_only):
+        df = DataFrame({"a": [2, 3, 1], "b": [2, 1, 1], "c": list("xyx")})
+        result = df.idxmax(numeric_only=numeric_only)
+        if numeric_only:
+            expected = Series([1, 0], index=["a", "b"])
+        else:
+            expected = Series([1, 0, 1], index=["a", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+    def test_idxmax_arrow_types(self):
+        # GH#55368
+        pytest.importorskip("pyarrow")
+
+        df = DataFrame({"a": [2, 3, 1], "b": [2, 1, 1]}, dtype="int64[pyarrow]")
+        result = df.idxmax()
+        expected = Series([1, 0], index=["a", "b"])
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin()
+        expected = Series([2, 1], index=["a", "b"])
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"a": ["b", "c", "a"]}, dtype="string[pyarrow]")
+        result = df.idxmax(numeric_only=False)
+        expected = Series([1], index=["a"])
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin(numeric_only=False)
+        expected = Series([2], index=["a"])
+        tm.assert_series_equal(result, expected)
+
+    def test_idxmax_axis_2(self, float_frame):
+        frame = float_frame
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            frame.idxmax(axis=2)
+
+    def test_idxmax_mixed_dtype(self):
+        # don't cast to object, which would raise in nanops
+        dti = date_range("2016-01-01", periods=3)
+        df = DataFrame({1: [0, 2, 1], 2: range(3)[::-1], 3: dti})
+
+        result = df.idxmax()
+        expected = Series([1, 0, 2], index=range(1, 4))
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin()
+        expected = Series([0, 2, 0], index=range(1, 4))
+        tm.assert_series_equal(result, expected)
+
+        # with NaTs
+        df.loc[0, 3] = pd.NaT
+        result = df.idxmax()
+        expected = Series([1, 0, 2], index=range(1, 4))
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin()
+        expected = Series([0, 2, 1], index=range(1, 4))
+        tm.assert_series_equal(result, expected)
+
+        # with multi-column dt64 block
+        df[4] = dti[::-1]
+        df._consolidate_inplace()
+
+        result = df.idxmax()
+        expected = Series([1, 0, 2, 0], index=range(1, 5))
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin()
+        expected = Series([0, 2, 1, 2], index=range(1, 5))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "op, expected_value",
+        [("idxmax", [0, 4]), ("idxmin", [0, 5])],
+    )
+    def test_idxmax_idxmin_convert_dtypes(self, op, expected_value):
+        # GH 40346
+        df = DataFrame(
+            {
+                "ID": [100, 100, 100, 200, 200, 200],
+                "value": [0, 0, 0, 1, 2, 0],
+            },
+            dtype="Int64",
+        )
+        df = df.groupby("ID")
+
+        result = getattr(df, op)()
+        expected = DataFrame(
+            {"value": expected_value},
+            index=Index([100, 200], name="ID", dtype="Int64"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_idxmax_dt64_multicolumn_axis1(self):
+        dti = date_range("2016-01-01", periods=3)
+        df = DataFrame({3: dti, 4: dti[::-1]}, copy=True)
+        df.iloc[0, 0] = pd.NaT
+
+        df._consolidate_inplace()
+
+        result = df.idxmax(axis=1)
+        expected = Series([4, 3, 3])
+        tm.assert_series_equal(result, expected)
+
+        result = df.idxmin(axis=1)
+        expected = Series([4, 3, 4])
+        tm.assert_series_equal(result, expected)
+
+    # ----------------------------------------------------------------------
+    # Logical reductions
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    @pytest.mark.parametrize("bool_only", [False, True])
+    def test_any_all_mixed_float(
+        self, all_boolean_reductions, axis, bool_only, float_string_frame
+    ):
+        # make sure op works on mixed-type frame
+        mixed = float_string_frame
+        mixed["_bool_"] = np.random.default_rng(2).standard_normal(len(mixed)) > 0.5
+
+        getattr(mixed, all_boolean_reductions)(axis=axis, bool_only=bool_only)
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_any_all_bool_with_na(
+        self, all_boolean_reductions, axis, bool_frame_with_na
+    ):
+        getattr(bool_frame_with_na, all_boolean_reductions)(axis=axis, bool_only=False)
+
+    def test_any_all_bool_frame(self, all_boolean_reductions, bool_frame_with_na):
+        # GH#12863: numpy gives back non-boolean data for object type
+        # so fill NaNs to compare with pandas behavior
+        frame = bool_frame_with_na.fillna(True)
+        alternative = getattr(np, all_boolean_reductions)
+        f = getattr(frame, all_boolean_reductions)
+
+        def skipna_wrapper(x):
+            nona = x.dropna().values
+            return alternative(nona)
+
+        def wrapper(x):
+            return alternative(x.values)
+
+        result0 = f(axis=0, skipna=False)
+        result1 = f(axis=1, skipna=False)
+
+        tm.assert_series_equal(result0, frame.apply(wrapper))
+        tm.assert_series_equal(result1, frame.apply(wrapper, axis=1))
+
+        result0 = f(axis=0)
+        result1 = f(axis=1)
+
+        tm.assert_series_equal(result0, frame.apply(skipna_wrapper))
+        tm.assert_series_equal(
+            result1, frame.apply(skipna_wrapper, axis=1), check_dtype=False
+        )
+
+        # bad axis
+        with pytest.raises(ValueError, match="No axis named 2"):
+            f(axis=2)
+
+        # all NA case
+        all_na = frame * np.nan
+        r0 = getattr(all_na, all_boolean_reductions)(axis=0)
+        r1 = getattr(all_na, all_boolean_reductions)(axis=1)
+        if all_boolean_reductions == "any":
+            assert not r0.any()
+            assert not r1.any()
+        else:
+            assert r0.all()
+            assert r1.all()
+
+    def test_any_all_extra(self):
+        df = DataFrame(
+            {
+                "A": [True, False, False],
+                "B": [True, True, False],
+                "C": [True, True, True],
+            },
+            index=["a", "b", "c"],
+        )
+        result = df[["A", "B"]].any(axis=1)
+        expected = Series([True, True, False], index=["a", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+        result = df[["A", "B"]].any(axis=1, bool_only=True)
+        tm.assert_series_equal(result, expected)
+
+        result = df.all(axis=1)
+        expected = Series([True, False, False], index=["a", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+        result = df.all(axis=1, bool_only=True)
+        tm.assert_series_equal(result, expected)
+
+        # Axis is None
+        result = df.all(axis=None).item()
+        assert result is False
+
+        result = df.any(axis=None).item()
+        assert result is True
+
+        result = df[["C"]].all(axis=None).item()
+        assert result is True
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_any_all_object_dtype(self, axis, all_boolean_reductions, skipna):
+        # GH#35450
+        df = DataFrame(
+            data=[
+                [1, np.nan, np.nan, True],
+                [np.nan, 2, np.nan, True],
+                [np.nan, np.nan, np.nan, True],
+                [np.nan, np.nan, "5", np.nan],
+            ]
+        )
+        result = getattr(df, all_boolean_reductions)(axis=axis, skipna=skipna)
+        expected = Series([True, True, True, True])
+        tm.assert_series_equal(result, expected)
+
+    def test_any_datetime(self):
+        # GH 23070
+        float_data = [1, np.nan, 3, np.nan]
+        datetime_data = [
+            Timestamp("1960-02-15"),
+            Timestamp("1960-02-16"),
+            pd.NaT,
+            pd.NaT,
+        ]
+        df = DataFrame({"A": float_data, "B": datetime_data})
+
+        msg = "datetime64 type does not support operation 'any'"
+        with pytest.raises(TypeError, match=msg):
+            df.any(axis=1)
+
+    def test_any_all_bool_only(self):
+        # GH 25101
+        df = DataFrame(
+            {"col1": [1, 2, 3], "col2": [4, 5, 6], "col3": [None, None, None]},
+            columns=Index(["col1", "col2", "col3"], dtype=object),
+        )
+
+        result = df.all(bool_only=True)
+        expected = Series(dtype=np.bool_, index=[])
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "col1": [1, 2, 3],
+                "col2": [4, 5, 6],
+                "col3": [None, None, None],
+                "col4": [False, False, True],
+            }
+        )
+
+        result = df.all(bool_only=True)
+        expected = Series({"col4": False})
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "func, data, expected",
+        [
+            (np.any, {}, False),
+            (np.all, {}, True),
+            (np.any, {"A": []}, False),
+            (np.all, {"A": []}, True),
+            (np.any, {"A": [False, False]}, False),
+            (np.all, {"A": [False, False]}, False),
+            (np.any, {"A": [True, False]}, True),
+            (np.all, {"A": [True, False]}, False),
+            (np.any, {"A": [True, True]}, True),
+            (np.all, {"A": [True, True]}, True),
+            (np.any, {"A": [False], "B": [False]}, False),
+            (np.all, {"A": [False], "B": [False]}, False),
+            (np.any, {"A": [False, False], "B": [False, True]}, True),
+            (np.all, {"A": [False, False], "B": [False, True]}, False),
+            # other types
+            (np.all, {"A": Series([0.0, 1.0], dtype="float")}, False),
+            (np.any, {"A": Series([0.0, 1.0], dtype="float")}, True),
+            (np.all, {"A": Series([0, 1], dtype=int)}, False),
+            (np.any, {"A": Series([0, 1], dtype=int)}, True),
+            pytest.param(np.all, {"A": Series([0, 1], dtype="M8[ns]")}, False),
+            pytest.param(np.all, {"A": Series([0, 1], dtype="M8[ns, UTC]")}, False),
+            pytest.param(np.any, {"A": Series([0, 1], dtype="M8[ns]")}, True),
+            pytest.param(np.any, {"A": Series([0, 1], dtype="M8[ns, UTC]")}, True),
+            pytest.param(np.all, {"A": Series([1, 2], dtype="M8[ns]")}, True),
+            pytest.param(np.all, {"A": Series([1, 2], dtype="M8[ns, UTC]")}, True),
+            pytest.param(np.any, {"A": Series([1, 2], dtype="M8[ns]")}, True),
+            pytest.param(np.any, {"A": Series([1, 2], dtype="M8[ns, UTC]")}, True),
+            pytest.param(np.all, {"A": Series([0, 1], dtype="m8[ns]")}, False),
+            pytest.param(np.any, {"A": Series([0, 1], dtype="m8[ns]")}, True),
+            pytest.param(np.all, {"A": Series([1, 2], dtype="m8[ns]")}, True),
+            pytest.param(np.any, {"A": Series([1, 2], dtype="m8[ns]")}, True),
+            # np.all on Categorical raises, so the reduction drops the
+            #  column, so all is being done on an empty Series, so is True
+            (np.all, {"A": Series([0, 1], dtype="category")}, True),
+            (np.any, {"A": Series([0, 1], dtype="category")}, False),
+            (np.all, {"A": Series([1, 2], dtype="category")}, True),
+            (np.any, {"A": Series([1, 2], dtype="category")}, False),
+            # Mix GH#21484
+            pytest.param(
+                np.all,
+                {
+                    "A": Series([10, 20], dtype="M8[ns]"),
+                    "B": Series([10, 20], dtype="m8[ns]"),
+                },
+                True,
+            ),
+        ],
+    )
+    def test_any_all_np_func(self, func, data, expected):
+        # GH 19976
+        data = DataFrame(data)
+
+        if any(isinstance(x, CategoricalDtype) for x in data.dtypes):
+            with pytest.raises(
+                TypeError, match=".* dtype category does not support operation"
+            ):
+                func(data)
+
+            # method version
+            with pytest.raises(
+                TypeError, match=".* dtype category does not support operation"
+            ):
+                getattr(DataFrame(data), func.__name__)(axis=None)
+        if data.dtypes.apply(lambda x: x.kind == "M").any():
+            # GH#34479
+            msg = "datetime64 type does not support operation '(any|all)'"
+            with pytest.raises(TypeError, match=msg):
+                func(data)
+
+            # method version
+            with pytest.raises(TypeError, match=msg):
+                getattr(DataFrame(data), func.__name__)(axis=None)
+
+        elif data.dtypes.apply(lambda x: x != "category").any():
+            result = func(data)
+            assert isinstance(result, np.bool_)
+            assert result.item() is expected
+
+            # method version
+            result = getattr(DataFrame(data), func.__name__)(axis=None)
+            assert isinstance(result, np.bool_)
+            assert result.item() is expected
+
+    def test_any_all_object(self):
+        # GH 19976
+        result = np.all(DataFrame(columns=["a", "b"])).item()
+        assert result is True
+
+        result = np.any(DataFrame(columns=["a", "b"])).item()
+        assert result is False
+
+    def test_any_all_object_bool_only(self):
+        df = DataFrame({"A": ["foo", 2], "B": [True, False]}).astype(object)
+        df._consolidate_inplace()
+        df["C"] = Series([True, True])
+
+        # Categorical of bools is _not_ considered booly
+        df["D"] = df["C"].astype("category")
+
+        # The underlying bug is in DataFrame._get_bool_data, so we check
+        #  that while we're here
+        res = df._get_bool_data()
+        expected = df[["C"]]
+        tm.assert_frame_equal(res, expected)
+
+        res = df.all(bool_only=True, axis=0)
+        expected = Series([True], index=["C"])
+        tm.assert_series_equal(res, expected)
+
+        # operating on a subset of columns should not produce a _larger_ Series
+        res = df[["B", "C"]].all(bool_only=True, axis=0)
+        tm.assert_series_equal(res, expected)
+
+        assert df.all(bool_only=True, axis=None)
+
+        res = df.any(bool_only=True, axis=0)
+        expected = Series([True], index=["C"])
+        tm.assert_series_equal(res, expected)
+
+        # operating on a subset of columns should not produce a _larger_ Series
+        res = df[["C"]].any(bool_only=True, axis=0)
+        tm.assert_series_equal(res, expected)
+
+        assert df.any(bool_only=True, axis=None)
+
+    # ---------------------------------------------------------------------
+    # Unsorted
+
+    def test_series_broadcasting(self):
+        # smoke test for numpy warnings
+        # GH 16378, GH 16306
+        df = DataFrame([1.0, 1.0, 1.0])
+        df_nan = DataFrame({"A": [np.nan, 2.0, np.nan]})
+        s = Series([1, 1, 1])
+        s_nan = Series([np.nan, np.nan, 1])
+
+        with tm.assert_produces_warning(None):
+            df_nan.clip(lower=s, axis=0)
+            for op in ["lt", "le", "gt", "ge", "eq", "ne"]:
+                getattr(df, op)(s_nan, axis=0)
+
+
+class TestDataFrameReductions:
+    def test_min_max_dt64_with_NaT(self):
+        # Both NaT and Timestamp are in DataFrame.
+        df = DataFrame({"foo": [pd.NaT, pd.NaT, Timestamp("2012-05-01")]})
+
+        res = df.min()
+        exp = Series([Timestamp("2012-05-01")], index=["foo"])
+        tm.assert_series_equal(res, exp)
+
+        res = df.max()
+        exp = Series([Timestamp("2012-05-01")], index=["foo"])
+        tm.assert_series_equal(res, exp)
+
+        # GH12941, only NaTs are in DataFrame.
+        df = DataFrame({"foo": [pd.NaT, pd.NaT]})
+
+        res = df.min()
+        exp = Series([pd.NaT], index=["foo"])
+        tm.assert_series_equal(res, exp)
+
+        res = df.max()
+        exp = Series([pd.NaT], index=["foo"])
+        tm.assert_series_equal(res, exp)
+
+    def test_min_max_dt64_with_NaT_precision(self):
+        # GH#60646 Make sure the reduction doesn't cast input timestamps to
+        # float and lose precision.
+        df = DataFrame(
+            {"foo": [pd.NaT, pd.NaT, Timestamp("2012-05-01 09:20:00.123456789")]},
+            dtype="datetime64[ns]",
+        )
+
+        res = df.min(axis=1)
+        exp = df.foo.rename(None)
+        tm.assert_series_equal(res, exp)
+
+        res = df.max(axis=1)
+        exp = df.foo.rename(None)
+        tm.assert_series_equal(res, exp)
+
+    def test_min_max_td64_with_NaT_precision(self):
+        # GH#60646 Make sure the reduction doesn't cast input timedeltas to
+        # float and lose precision.
+        df = DataFrame(
+            {
+                "foo": [
+                    pd.NaT,
+                    pd.NaT,
+                    to_timedelta("10000 days 06:05:01.123456789"),
+                ],
+            },
+            dtype="timedelta64[ns]",
+        )
+
+        res = df.min(axis=1)
+        exp = df.foo.rename(None)
+        tm.assert_series_equal(res, exp)
+
+        res = df.max(axis=1)
+        exp = df.foo.rename(None)
+        tm.assert_series_equal(res, exp)
+
+    def test_min_max_dt64_with_NaT_skipna_false(self, request, tz_naive_fixture):
+        # GH#36907
+        tz = tz_naive_fixture
+        if isinstance(tz, tzlocal) and is_platform_windows():
+            pytest.skip(
+                "GH#37659 OSError raised within tzlocal bc Windows "
+                "chokes in times before 1970-01-01"
+            )
+
+        df = DataFrame(
+            {
+                "a": [
+                    Timestamp("2020-01-01 08:00:00", tz=tz),
+                    Timestamp("1920-02-01 09:00:00", tz=tz),
+                ],
+                "b": [Timestamp("2020-02-01 08:00:00", tz=tz), pd.NaT],
+            }
+        )
+        res = df.min(axis=1, skipna=False)
+        expected = Series([df.loc[0, "a"], pd.NaT])
+        assert expected.dtype == df["a"].dtype
+
+        tm.assert_series_equal(res, expected)
+
+        res = df.max(axis=1, skipna=False)
+        expected = Series([df.loc[0, "b"], pd.NaT])
+        assert expected.dtype == df["a"].dtype
+
+        tm.assert_series_equal(res, expected)
+
+    def test_min_max_dt64_api_consistency_with_NaT(self):
+        # Calling the following sum functions returned an error for dataframes but
+        # returned NaT for series. These tests check that the API is consistent in
+        # min/max calls on empty Series/DataFrames. See GH:33704 for more
+        # information
+        df = DataFrame({"x": to_datetime([])})
+        expected_dt_series = Series(to_datetime([]))
+        # check axis 0
+        assert (df.min(axis=0).x is pd.NaT) == (expected_dt_series.min() is pd.NaT)
+        assert (df.max(axis=0).x is pd.NaT) == (expected_dt_series.max() is pd.NaT)
+
+        # check axis 1
+        tm.assert_series_equal(df.min(axis=1), expected_dt_series)
+        tm.assert_series_equal(df.max(axis=1), expected_dt_series)
+
+    def test_min_max_dt64_api_consistency_empty_df(self):
+        # check DataFrame/Series api consistency when calling min/max on an empty
+        # DataFrame/Series.
+        df = DataFrame({"x": []})
+        expected_float_series = Series([], dtype=float)
+        # check axis 0
+        assert np.isnan(df.min(axis=0).x) == np.isnan(expected_float_series.min())
+        assert np.isnan(df.max(axis=0).x) == np.isnan(expected_float_series.max())
+        # check axis 1
+        tm.assert_series_equal(df.min(axis=1), expected_float_series)
+        tm.assert_series_equal(df.min(axis=1), expected_float_series)
+
+    @pytest.mark.parametrize(
+        "initial",
+        ["2018-10-08 13:36:45+00:00", "2018-10-08 13:36:45+03:00"],  # Non-UTC timezone
+    )
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_preserve_timezone(self, initial: str, method):
+        # GH 28552
+        initial_dt = to_datetime(initial)
+        expected = Series([initial_dt])
+        df = DataFrame([expected])
+        result = getattr(df, method)(axis=1)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_minmax_tzaware_skipna_axis_1(self, method, skipna):
+        # GH#51242
+        val = to_datetime("1900-01-01", utc=True)
+        df = DataFrame(
+            {"a": Series([pd.NaT, pd.NaT, val]), "b": Series([pd.NaT, val, val])}
+        )
+        op = getattr(df, method)
+        result = op(axis=1, skipna=skipna)
+        if skipna:
+            expected = Series([pd.NaT, val, val])
+        else:
+            expected = Series([pd.NaT, pd.NaT, val])
+        tm.assert_series_equal(result, expected)
+
+    def test_frame_any_with_timedelta(self):
+        # GH#17667
+        df = DataFrame(
+            {
+                "a": Series([0, 0]),
+                "t": Series([to_timedelta(0, "s"), to_timedelta(1, "ms")]),
+            }
+        )
+
+        result = df.any(axis=0)
+        expected = Series(data=[False, True], index=["a", "t"])
+        tm.assert_series_equal(result, expected)
+
+        result = df.any(axis=1)
+        expected = Series(data=[False, True])
+        tm.assert_series_equal(result, expected)
+
+    def test_reductions_skipna_none_raises(
+        self, request, frame_or_series, all_reductions
+    ):
+        if all_reductions == "count":
+            request.applymarker(
+                pytest.mark.xfail(reason="Count does not accept skipna")
+            )
+        obj = frame_or_series([1, 2, 3])
+        msg = 'For argument "skipna" expected type bool, received type NoneType.'
+        with pytest.raises(ValueError, match=msg):
+            getattr(obj, all_reductions)(skipna=None)
+
+    def test_reduction_timestamp_smallest_unit(self):
+        # GH#52524
+        df = DataFrame(
+            {
+                "a": Series([Timestamp("2019-12-31")], dtype="datetime64[s]"),
+                "b": Series(
+                    [Timestamp("2019-12-31 00:00:00.123")], dtype="datetime64[ms]"
+                ),
+            }
+        )
+        result = df.max()
+        expected = Series(
+            [Timestamp("2019-12-31"), Timestamp("2019-12-31 00:00:00.123")],
+            dtype="datetime64[ms]",
+            index=["a", "b"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_reduction_timedelta_smallest_unit(self):
+        # GH#52524
+        df = DataFrame(
+            {
+                "a": Series([pd.Timedelta("1 days")], dtype="timedelta64[s]"),
+                "b": Series([pd.Timedelta("1 days")], dtype="timedelta64[ms]"),
+            }
+        )
+        result = df.max()
+        expected = Series(
+            [pd.Timedelta("1 days"), pd.Timedelta("1 days")],
+            dtype="timedelta64[ms]",
+            index=["a", "b"],
+        )
+        tm.assert_series_equal(result, expected)
+
+
+class TestNuisanceColumns:
+    def test_any_all_categorical_dtype_nuisance_column(self, all_boolean_reductions):
+        # GH#36076 DataFrame should match Series behavior
+        ser = Series([0, 1], dtype="category", name="A")
+        df = ser.to_frame()
+
+        # Double-check the Series behavior is to raise
+        with pytest.raises(TypeError, match="does not support operation"):
+            getattr(ser, all_boolean_reductions)()
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            getattr(np, all_boolean_reductions)(ser)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            getattr(df, all_boolean_reductions)(bool_only=False)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            getattr(df, all_boolean_reductions)(bool_only=None)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            getattr(np, all_boolean_reductions)(df, axis=0)
+
+    def test_median_categorical_dtype_nuisance_column(self):
+        # GH#21020 DataFrame.median should match Series.median
+        df = DataFrame({"A": Categorical([1, 2, 2, 2, 3])})
+        ser = df["A"]
+
+        # Double-check the Series behavior is to raise
+        with pytest.raises(TypeError, match="does not support operation"):
+            ser.median()
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            df.median(numeric_only=False)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            df.median()
+
+        # same thing, but with an additional non-categorical column
+        df["B"] = df["A"].astype(int)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            df.median(numeric_only=False)
+
+        with pytest.raises(TypeError, match="does not support operation"):
+            df.median()
+
+        # TODO: np.median(df, axis=0) gives np.array([2.0, 2.0]) instead
+        #  of expected.values
+
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_min_max_categorical_dtype_non_ordered_nuisance_column(self, method):
+        # GH#28949 DataFrame.min should behave like Series.min
+        cat = Categorical(["a", "b", "c", "b"], ordered=False)
+        ser = Series(cat)
+        df = ser.to_frame("A")
+
+        # Double-check the Series behavior
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(ser, method)()
+
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(np, method)(ser)
+
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(df, method)(numeric_only=False)
+
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(df, method)()
+
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(np, method)(df, axis=0)
+
+        # same thing, but with an additional non-categorical column
+        df["B"] = df["A"].astype(object)
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(df, method)()
+
+        with pytest.raises(TypeError, match="is not ordered for operation"):
+            getattr(np, method)(df, axis=0)
+
+
+class TestEmptyDataFrameReductions:
+    @pytest.mark.parametrize(
+        "opname, dtype, exp_value, exp_dtype",
+        [
+            ("sum", np.int8, 0, np.int64),
+            ("prod", np.int8, 1, np.int_),
+            ("sum", np.int64, 0, np.int64),
+            ("prod", np.int64, 1, np.int64),
+            ("sum", np.uint8, 0, np.uint64),
+            ("prod", np.uint8, 1, np.uint),
+            ("sum", np.uint64, 0, np.uint64),
+            ("prod", np.uint64, 1, np.uint64),
+            ("sum", np.float32, 0, np.float32),
+            ("prod", np.float32, 1, np.float32),
+            ("sum", np.float64, 0, np.float64),
+        ],
+    )
+    def test_df_empty_min_count_0(self, opname, dtype, exp_value, exp_dtype):
+        df = DataFrame({0: [], 1: []}, dtype=dtype)
+        result = getattr(df, opname)(min_count=0)
+
+        expected = Series([exp_value, exp_value], dtype=exp_dtype, index=range(2))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "opname, dtype, exp_dtype",
+        [
+            ("sum", np.int8, np.float64),
+            ("prod", np.int8, np.float64),
+            ("sum", np.int64, np.float64),
+            ("prod", np.int64, np.float64),
+            ("sum", np.uint8, np.float64),
+            ("prod", np.uint8, np.float64),
+            ("sum", np.uint64, np.float64),
+            ("prod", np.uint64, np.float64),
+            ("sum", np.float32, np.float32),
+            ("prod", np.float32, np.float32),
+            ("sum", np.float64, np.float64),
+        ],
+    )
+    def test_df_empty_min_count_1(self, opname, dtype, exp_dtype):
+        df = DataFrame({0: [], 1: []}, dtype=dtype)
+        result = getattr(df, opname)(min_count=1)
+
+        expected = Series([np.nan, np.nan], dtype=exp_dtype, index=Index([0, 1]))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "opname, dtype, exp_value, exp_dtype",
+        [
+            ("sum", "Int8", 0, ("Int32" if is_windows_np2_or_is32 else "Int64")),
+            ("prod", "Int8", 1, ("Int32" if is_windows_np2_or_is32 else "Int64")),
+            ("sum", "Int64", 0, "Int64"),
+            ("prod", "Int64", 1, "Int64"),
+            ("sum", "UInt8", 0, ("UInt32" if is_windows_np2_or_is32 else "UInt64")),
+            ("prod", "UInt8", 1, ("UInt32" if is_windows_np2_or_is32 else "UInt64")),
+            ("sum", "UInt64", 0, "UInt64"),
+            ("prod", "UInt64", 1, "UInt64"),
+            ("sum", "Float32", 0, "Float32"),
+            ("prod", "Float32", 1, "Float32"),
+            ("sum", "Float64", 0, "Float64"),
+        ],
+    )
+    def test_df_empty_nullable_min_count_0(self, opname, dtype, exp_value, exp_dtype):
+        df = DataFrame({0: [], 1: []}, dtype=dtype)
+        result = getattr(df, opname)(min_count=0)
+
+        expected = Series([exp_value, exp_value], dtype=exp_dtype, index=Index([0, 1]))
+        tm.assert_series_equal(result, expected)
+
+    # TODO: why does min_count=1 impact the resulting Windows dtype
+    # differently than min_count=0?
+    @pytest.mark.parametrize(
+        "opname, dtype, exp_dtype",
+        [
+            ("sum", "Int8", ("Int32" if is_windows_or_is32 else "Int64")),
+            ("prod", "Int8", ("Int32" if is_windows_or_is32 else "Int64")),
+            ("sum", "Int64", "Int64"),
+            ("prod", "Int64", "Int64"),
+            ("sum", "UInt8", ("UInt32" if is_windows_or_is32 else "UInt64")),
+            ("prod", "UInt8", ("UInt32" if is_windows_or_is32 else "UInt64")),
+            ("sum", "UInt64", "UInt64"),
+            ("prod", "UInt64", "UInt64"),
+            ("sum", "Float32", "Float32"),
+            ("prod", "Float32", "Float32"),
+            ("sum", "Float64", "Float64"),
+        ],
+    )
+    def test_df_empty_nullable_min_count_1(self, opname, dtype, exp_dtype):
+        df = DataFrame({0: [], 1: []}, dtype=dtype)
+        result = getattr(df, opname)(min_count=1)
+
+        expected = Series([pd.NA, pd.NA], dtype=exp_dtype, index=Index([0, 1]))
+        tm.assert_series_equal(result, expected)
+
+
+def test_sum_timedelta64_skipna_false():
+    # GH#17235
+    arr = np.arange(8).astype(np.int64).view("m8[s]").reshape(4, 2)
+    arr[-1, -1] = "Nat"
+
+    df = DataFrame(arr)
+    assert (df.dtypes == arr.dtype).all()
+
+    result = df.sum(skipna=False)
+    expected = Series([pd.Timedelta(seconds=12), pd.NaT], dtype="m8[s]")
+    tm.assert_series_equal(result, expected)
+
+    result = df.sum(axis=0, skipna=False)
+    tm.assert_series_equal(result, expected)
+
+    result = df.sum(axis=1, skipna=False)
+    expected = Series(
+        [
+            pd.Timedelta(seconds=1),
+            pd.Timedelta(seconds=5),
+            pd.Timedelta(seconds=9),
+            pd.NaT,
+        ],
+        dtype="m8[s]",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_mixed_frame_with_integer_sum():
+    # https://github.com/pandas-dev/pandas/issues/34520
+    df = DataFrame([["a", 1]], columns=list("ab"))
+    df = df.astype({"b": "Int64"})
+    result = df.sum()
+    expected = Series(["a", 1], index=["a", "b"])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("numeric_only", [True, False, None])
+@pytest.mark.parametrize("method", ["min", "max"])
+def test_minmax_extensionarray(method, numeric_only):
+    # https://github.com/pandas-dev/pandas/issues/32651
+    int64_info = np.iinfo("int64")
+    ser = Series([int64_info.max, None, int64_info.min], dtype=pd.Int64Dtype())
+    df = DataFrame({"Int64": ser})
+    result = getattr(df, method)(numeric_only=numeric_only)
+    expected = Series(
+        [getattr(int64_info, method)],
+        dtype="Int64",
+        index=Index(["Int64"]),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ts_value", [Timestamp("2000-01-01"), pd.NaT])
+def test_frame_mixed_numeric_object_with_timestamp(ts_value):
+    # GH 13912
+    df = DataFrame({"a": [1], "b": [1.1], "c": ["foo"], "d": [ts_value]})
+    with pytest.raises(TypeError, match="does not support operation|Cannot perform"):
+        df.sum()
+
+
+def test_prod_sum_min_count_mixed_object():
+    # https://github.com/pandas-dev/pandas/issues/41074
+    df = DataFrame([1, "a", True])
+
+    result = df.prod(axis=0, min_count=1, numeric_only=False)
+    expected = Series(["a"], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    msg = re.escape("unsupported operand type(s) for +: 'int' and 'str'")
+    with pytest.raises(TypeError, match=msg):
+        df.sum(axis=0, min_count=1, numeric_only=False)
+
+
+@pytest.mark.parametrize("method", ["min", "max", "mean", "median", "skew", "kurt"])
+@pytest.mark.parametrize("numeric_only", [True, False])
+@pytest.mark.parametrize("dtype", ["float64", "Float64"])
+def test_reduction_axis_none_returns_scalar(method, numeric_only, dtype):
+    # GH#21597 As of 2.0, axis=None reduces over all axes.
+
+    df = DataFrame(np.random.default_rng(2).standard_normal((4, 4)), dtype=dtype)
+
+    result = getattr(df, method)(axis=None, numeric_only=numeric_only)
+    np_arr = df.to_numpy(dtype=np.float64)
+    if method in {"skew", "kurt"}:
+        comp_mod = pytest.importorskip("scipy.stats")
+        if method == "kurt":
+            method = "kurtosis"
+        expected = getattr(comp_mod, method)(np_arr, bias=False, axis=None)
+        tm.assert_almost_equal(result, expected)
+    else:
+        expected = getattr(np, method)(np_arr, axis=None)
+        assert result == expected
+
+
+@pytest.mark.parametrize(
+    "kernel",
+    [
+        "corr",
+        "corrwith",
+        "cov",
+        "idxmax",
+        "idxmin",
+        "kurt",
+        "max",
+        "mean",
+        "median",
+        "min",
+        "prod",
+        "quantile",
+        "sem",
+        "skew",
+        "std",
+        "sum",
+        "var",
+    ],
+)
+def test_fails_on_non_numeric(kernel):
+    # GH#46852
+    df = DataFrame({"a": [1, 2, 3], "b": object})
+    args = (df,) if kernel == "corrwith" else ()
+    msg = "|".join(
+        [
+            "not allowed for this dtype",
+            "argument must be a string or a number",
+            "not supported between instances of",
+            "unsupported operand type",
+            "argument must be a string or a real number",
+        ]
+    )
+    if kernel == "median":
+        # slightly different message on different builds
+        msg1 = (
+            r"Cannot convert \[\[<class 'object'> <class 'object'> "
+            r"<class 'object'>\]\] to numeric"
+        )
+        msg2 = (
+            r"Cannot convert \[<class 'object'> <class 'object'> "
+            r"<class 'object'>\] to numeric"
+        )
+        msg = "|".join([msg1, msg2])
+    with pytest.raises(TypeError, match=msg):
+        getattr(df, kernel)(*args)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        "all",
+        "any",
+        "count",
+        "idxmax",
+        "idxmin",
+        "kurt",
+        "kurtosis",
+        "max",
+        "mean",
+        "median",
+        "min",
+        "nunique",
+        "prod",
+        "product",
+        "sem",
+        "skew",
+        "std",
+        "sum",
+        "var",
+    ],
+)
+@pytest.mark.parametrize("min_count", [0, 2])
+def test_numeric_ea_axis_1(method, skipna, min_count, any_numeric_ea_dtype):
+    # GH 54341
+    df = DataFrame(
+        {
+            "a": Series([0, 1, 2, 3], dtype=any_numeric_ea_dtype),
+            "b": Series([0, 1, pd.NA, 3], dtype=any_numeric_ea_dtype),
+        },
+    )
+    expected_df = DataFrame(
+        {
+            "a": [0.0, 1.0, 2.0, 3.0],
+            "b": [0.0, 1.0, np.nan, 3.0],
+        },
+    )
+    if method in ("count", "nunique"):
+        expected_dtype = "int64"
+    elif method in ("all", "any"):
+        expected_dtype = "boolean"
+    elif method in (
+        "kurt",
+        "kurtosis",
+        "mean",
+        "median",
+        "sem",
+        "skew",
+        "std",
+        "var",
+    ) and not any_numeric_ea_dtype.startswith("Float"):
+        expected_dtype = "Float64"
+    else:
+        expected_dtype = any_numeric_ea_dtype
+
+    kwargs = {}
+    if method not in ("count", "nunique", "quantile"):
+        kwargs["skipna"] = skipna
+    if method in ("prod", "product", "sum"):
+        kwargs["min_count"] = min_count
+
+    if not skipna and method in ("idxmax", "idxmin"):
+        # GH#57745 - EAs use groupby for axis=1 which still needs a proper deprecation.
+        msg = f"The behavior of DataFrame.{method} with all-NA values"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            getattr(df, method)(axis=1, **kwargs)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            getattr(expected_df, method)(axis=1, **kwargs)
+        return
+    result = getattr(df, method)(axis=1, **kwargs)
+    expected = getattr(expected_df, method)(axis=1, **kwargs)
+    if method not in ("idxmax", "idxmin"):
+        expected = expected.astype(expected_dtype)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_repr.py b/pandas/pandas/tests/frame/test_repr.py
new file mode 100644
index 0000000000000..73628424725e5
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_repr.py
@@ -0,0 +1,498 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    Timestamp,
+    date_range,
+    option_context,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameRepr:
+    def test_repr_should_return_str(self):
+        # https://docs.python.org/3/reference/datamodel.html#object.__repr__
+        # "...The return value must be a string object."
+
+        # (str on py2.x, str (unicode) on py3)
+
+        data = [8, 5, 3, 5]
+        index1 = ["\u03c3", "\u03c4", "\u03c5", "\u03c6"]
+        cols = ["\u03c8"]
+        df = DataFrame(data, columns=cols, index=index1)
+        assert type(df.__repr__()) is str
+
+        ser = df[cols[0]]
+        assert type(ser.__repr__()) is str
+
+    def test_repr_bytes_61_lines(self):
+        # GH#12857
+        lets = list("ACDEFGHIJKLMNOP")
+        words = np.random.default_rng(2).choice(lets, (1000, 50))
+        df = DataFrame(words).astype("U1")
+        assert (df.dtypes == object).all()
+
+        # smoke tests; at one point this raised with 61 but not 60
+        repr(df)
+        repr(df.iloc[:60, :])
+        repr(df.iloc[:61, :])
+
+    def test_repr_unicode_level_names(self, frame_or_series):
+        index = MultiIndex.from_tuples([(0, 0), (1, 1)], names=["\u0394", "i1"])
+
+        obj = DataFrame(np.random.default_rng(2).standard_normal((2, 4)), index=index)
+        obj = tm.get_obj(obj, frame_or_series)
+        repr(obj)
+
+    def test_assign_index_sequences(self):
+        # GH#2200
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]}).set_index(
+            ["a", "b"]
+        )
+        index = list(df.index)
+        index[0] = ("faz", "boo")
+        df.index = index
+        repr(df)
+
+        # this travels an improper code path
+        index[0] = ["faz", "boo"]
+        df.index = index
+        repr(df)
+
+    def test_repr_with_mi_nat(self):
+        df = DataFrame({"X": [1, 2]}, index=[[NaT, Timestamp("20130101")], ["a", "b"]])
+        result = repr(df)
+        expected = "              X\nNaT        a  1\n2013-01-01 b  2"
+        assert result == expected
+
+    def test_repr_with_different_nulls(self):
+        # GH45263
+        df = DataFrame([1, 2, 3, 4], [True, None, np.nan, NaT])
+        result = repr(df)
+        expected = """      0
+True  1
+None  2
+NaN   3
+NaT   4"""
+        assert result == expected
+
+    def test_repr_with_different_nulls_cols(self):
+        # GH45263
+        d = {np.nan: [1, 2], None: [3, 4], NaT: [6, 7], True: [8, 9]}
+        df = DataFrame(data=d)
+        result = repr(df)
+        expected = """   NaN  None  NaT  True
+0    1     3    6     8
+1    2     4    7     9"""
+        assert result == expected
+
+    def test_multiindex_na_repr(self):
+        # only an issue with long columns
+        df3 = DataFrame(
+            {
+                "A" * 30: {("A", "A0006000", "nuit"): "A0006000"},
+                "B" * 30: {("A", "A0006000", "nuit"): np.nan},
+                "C" * 30: {("A", "A0006000", "nuit"): np.nan},
+                "D" * 30: {("A", "A0006000", "nuit"): np.nan},
+                "E" * 30: {("A", "A0006000", "nuit"): "A"},
+                "F" * 30: {("A", "A0006000", "nuit"): np.nan},
+            }
+        )
+
+        idf = df3.set_index(["A" * 30, "C" * 30])
+        repr(idf)
+
+    def test_repr_name_coincide(self):
+        index = MultiIndex.from_tuples(
+            [("a", 0, "foo"), ("b", 1, "bar")], names=["a", "b", "c"]
+        )
+
+        df = DataFrame({"value": [0, 1]}, index=index)
+
+        lines = repr(df).split("\n")
+        assert lines[2].startswith("a 0 foo")
+
+    def test_repr_to_string(
+        self,
+        multiindex_year_month_day_dataframe_random_data,
+        multiindex_dataframe_random_data,
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+        frame = multiindex_dataframe_random_data
+
+        repr(frame)
+        repr(ymd)
+        repr(frame.T)
+        repr(ymd.T)
+
+        buf = StringIO()
+        frame.to_string(buf=buf)
+        ymd.to_string(buf=buf)
+        frame.T.to_string(buf=buf)
+        ymd.T.to_string(buf=buf)
+
+    def test_repr_empty(self):
+        # empty
+        repr(DataFrame())
+
+        # empty with index
+        frame = DataFrame(index=np.arange(1000))
+        repr(frame)
+
+    def test_repr_mixed(self, float_string_frame):
+        # mixed
+        repr(float_string_frame)
+
+    @pytest.mark.slow
+    def test_repr_mixed_big(self):
+        # big mixed
+        biggie = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(200),
+                "B": [str(i) for i in range(200)],
+            },
+            index=range(200),
+        )
+        biggie.loc[:20, "A"] = np.nan
+        biggie.loc[:20, "B"] = np.nan
+
+        repr(biggie)
+
+    def test_repr(self):
+        # columns but no index
+        no_index = DataFrame(columns=[0, 1, 3])
+        repr(no_index)
+
+        df = DataFrame(["a\n\r\tb"], columns=["a\n\r\td"], index=["a\n\r\tf"])
+        assert "\t" not in repr(df)
+        assert "\r" not in repr(df)
+        assert "a\n" not in repr(df)
+
+    def test_repr_dimensions(self):
+        df = DataFrame([[1, 2], [3, 4]])
+        with option_context("display.show_dimensions", True):
+            assert "2 rows x 2 columns" in repr(df)
+
+        with option_context("display.show_dimensions", False):
+            assert "2 rows x 2 columns" not in repr(df)
+
+        with option_context("display.show_dimensions", "truncate"):
+            assert "2 rows x 2 columns" not in repr(df)
+
+    @pytest.mark.slow
+    def test_repr_big(self):
+        # big one
+        biggie = DataFrame(np.zeros((200, 4)), columns=range(4), index=range(200))
+        repr(biggie)
+
+    def test_repr_unsortable(self):
+        # columns are not sortable
+
+        unsortable = DataFrame(
+            {
+                "foo": [1] * 50,
+                datetime.today(): [1] * 50,
+                "bar": ["bar"] * 50,
+                datetime.today() + timedelta(1): ["bar"] * 50,
+            },
+            index=np.arange(50),
+        )
+        repr(unsortable)
+
+    def test_repr_float_frame_options(self, float_frame):
+        repr(float_frame)
+
+        with option_context("display.precision", 3):
+            repr(float_frame)
+
+        with option_context("display.max_rows", 10, "display.max_columns", 2):
+            repr(float_frame)
+
+        with option_context("display.max_rows", 1000, "display.max_columns", 1000):
+            repr(float_frame)
+
+    def test_repr_unicode(self):
+        uval = "\u03c3\u03c3\u03c3\u03c3"
+
+        df = DataFrame({"A": [uval, uval]})
+
+        result = repr(df)
+        ex_top = "      A"
+        assert result.split("\n")[0].rstrip() == ex_top
+
+        df = DataFrame({"A": [uval, uval]})
+        result = repr(df)
+        assert result.split("\n")[0].rstrip() == ex_top
+
+    def test_unicode_string_with_unicode(self):
+        df = DataFrame({"A": ["\u05d0"]})
+        str(df)
+
+    def test_repr_unicode_columns(self):
+        df = DataFrame({"\u05d0": [1, 2, 3], "\u05d1": [4, 5, 6], "c": [7, 8, 9]})
+        repr(df.columns)  # should not raise UnicodeDecodeError
+
+    def test_str_to_bytes_raises(self):
+        # GH 26447
+        df = DataFrame({"A": ["abc"]})
+        msg = "^'str' object cannot be interpreted as an integer$"
+        with pytest.raises(TypeError, match=msg):
+            bytes(df)
+
+    def test_very_wide_repr(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 20)),
+            columns=np.array(["a" * 10] * 20, dtype=object),
+        )
+        repr(df)
+
+    def test_repr_column_name_unicode_truncation_bug(self):
+        # #1906
+        df = DataFrame(
+            {
+                "Id": [7117434],
+                "StringCol": (
+                    "Is it possible to modify drop plot code"
+                    "so that the output graph is displayed "
+                    "in iphone simulator, Is it possible to "
+                    "modify drop plot code so that the "
+                    "output graph is \xe2\x80\xa8displayed "
+                    "in iphone simulator.Now we are adding "
+                    "the CSV file externally. I want to Call "
+                    "the File through the code.."
+                ),
+            }
+        )
+
+        with option_context("display.max_columns", 20):
+            assert "StringCol" in repr(df)
+
+    def test_latex_repr(self):
+        pytest.importorskip("jinja2")
+        expected = r"""\begin{tabular}{llll}
+\toprule
+ & 0 & 1 & 2 \\
+\midrule
+0 & $\alpha$ & b & c \\
+1 & 1 & 2 & 3 \\
+\bottomrule
+\end{tabular}
+"""
+        with option_context(
+            "styler.format.escape", None, "styler.render.repr", "latex"
+        ):
+            df = DataFrame([[r"$\alpha$", "b", "c"], [1, 2, 3]])
+            result = df._repr_latex_()
+            assert result == expected
+
+        # GH 12182
+        assert df._repr_latex_() is None
+
+    def test_repr_with_datetimeindex(self):
+        df = DataFrame({"A": [1, 2, 3]}, index=date_range("2000", periods=3))
+        result = repr(df)
+        expected = "            A\n2000-01-01  1\n2000-01-02  2\n2000-01-03  3"
+        assert result == expected
+
+    def test_repr_with_intervalindex(self):
+        # https://github.com/pandas-dev/pandas/pull/24134/files
+        df = DataFrame(
+            {"A": [1, 2, 3, 4]}, index=IntervalIndex.from_breaks([0, 1, 2, 3, 4])
+        )
+        result = repr(df)
+        expected = "        A\n(0, 1]  1\n(1, 2]  2\n(2, 3]  3\n(3, 4]  4"
+        assert result == expected
+
+    def test_repr_with_categorical_index(self):
+        df = DataFrame({"A": [1, 2, 3]}, index=CategoricalIndex(["a", "b", "c"]))
+        result = repr(df)
+        expected = "   A\na  1\nb  2\nc  3"
+        assert result == expected
+
+    def test_repr_categorical_dates_periods(self):
+        # normal DataFrame
+        dt = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        p = period_range("2011-01", freq="M", periods=5)
+        df = DataFrame({"dt": dt, "p": p})
+        exp = """                         dt        p
+0 2011-01-01 09:00:00-05:00  2011-01
+1 2011-01-01 10:00:00-05:00  2011-02
+2 2011-01-01 11:00:00-05:00  2011-03
+3 2011-01-01 12:00:00-05:00  2011-04
+4 2011-01-01 13:00:00-05:00  2011-05"""
+
+        assert repr(df) == exp
+
+        df2 = DataFrame({"dt": Categorical(dt), "p": Categorical(p)})
+        assert repr(df2) == exp
+
+    @pytest.mark.parametrize("arg", [np.datetime64, np.timedelta64])
+    @pytest.mark.parametrize(
+        "box, expected",
+        [[Series, "0    NaT\ndtype: object"], [DataFrame, "     0\n0  NaT"]],
+    )
+    def test_repr_np_nat_with_object(self, arg, box, expected):
+        # GH 25445
+        result = repr(box([arg("NaT")], dtype=object))
+        assert result == expected
+
+    def test_frame_datetime64_pre1900_repr(self):
+        df = DataFrame({"year": date_range("1/1/1700", periods=50, freq="YE-DEC")})
+        # it works!
+        repr(df)
+
+    def test_frame_to_string_with_periodindex(self):
+        index = PeriodIndex(["2011-1", "2011-2", "2011-3"], freq="M")
+        frame = DataFrame(np.random.default_rng(2).standard_normal((3, 4)), index=index)
+
+        # it works!
+        frame.to_string()
+
+    def test_to_string_ea_na_in_multiindex(self):
+        # GH#47986
+        df = DataFrame(
+            {"a": [1, 2]},
+            index=MultiIndex.from_arrays([Series([NA, 1], dtype="Int64")]),
+        )
+
+        result = df.to_string()
+        expected = """      a
+<NA>  1
+1     2"""
+        assert result == expected
+
+    def test_datetime64tz_slice_non_truncate(self):
+        # GH 30263
+        df = DataFrame({"x": date_range("2019", periods=10, tz="UTC")})
+        expected = repr(df)
+        df = df.iloc[:, :5]
+        result = repr(df)
+        assert result == expected
+
+    def test_to_records_no_typeerror_in_repr(self):
+        # GH 48526
+        df = DataFrame([["a", "b"], ["c", "d"], ["e", "f"]], columns=["left", "right"])
+        df["record"] = df[["left", "right"]].to_records()
+        expected = """  left right     record
+0    a     b  [0, a, b]
+1    c     d  [1, c, d]
+2    e     f  [2, e, f]"""
+        result = repr(df)
+        assert result == expected
+
+    def test_to_records_with_na_record_value(self):
+        # GH 48526
+        df = DataFrame(
+            [["a", np.nan], ["c", "d"], ["e", "f"]], columns=["left", "right"]
+        )
+        df["record"] = df[["left", "right"]].to_records()
+        expected = """  left right       record
+0    a   NaN  [0, a, nan]
+1    c     d    [1, c, d]
+2    e     f    [2, e, f]"""
+        result = repr(df)
+        assert result == expected
+
+    def test_to_records_with_na_record(self):
+        # GH 48526
+        df = DataFrame(
+            [["a", "b"], [np.nan, np.nan], ["e", "f"]], columns=[np.nan, "right"]
+        )
+        df["record"] = df[[np.nan, "right"]].to_records()
+        expected = """   NaN right         record
+0    a     b      [0, a, b]
+1  NaN   NaN  [1, nan, nan]
+2    e     f      [2, e, f]"""
+        result = repr(df)
+        assert result == expected
+
+    def test_to_records_with_inf_record(self):
+        # GH 48526
+        expected = """   NaN  inf         record
+0  inf    b    [0, inf, b]
+1  NaN  NaN  [1, nan, nan]
+2    e    f      [2, e, f]"""
+        df = DataFrame(
+            [[np.inf, "b"], [np.nan, np.nan], ["e", "f"]],
+            columns=[np.nan, np.inf],
+        )
+        df["record"] = df[[np.nan, np.inf]].to_records()
+        result = repr(df)
+        assert result == expected
+
+    def test_masked_ea_with_formatter(self):
+        # GH#39336
+        df = DataFrame(
+            {
+                "a": Series([0.123456789, 1.123456789], dtype="Float64"),
+                "b": Series([1, 2], dtype="Int64"),
+            }
+        )
+        result = df.to_string(formatters=["{:.2f}".format, "{:.2f}".format])
+        expected = """      a     b
+0  0.12  1.00
+1  1.12  2.00"""
+        assert result == expected
+
+    def test_repr_ea_columns(self, any_string_dtype):
+        # GH#54797
+        pytest.importorskip("pyarrow")
+        df = DataFrame({"long_column_name": [1, 2, 3], "col2": [4, 5, 6]})
+        df.columns = df.columns.astype(any_string_dtype)
+        expected = """   long_column_name  col2
+0                 1     4
+1                 2     5
+2                 3     6"""
+        assert repr(df) == expected
+
+
+@pytest.mark.parametrize(
+    "data,output",
+    [
+        ([2, complex("nan"), 1], [" 2.0+0.0j", " NaN+0.0j", " 1.0+0.0j"]),
+        ([2, complex("nan"), -1], [" 2.0+0.0j", " NaN+0.0j", "-1.0+0.0j"]),
+        ([-2, complex("nan"), -1], ["-2.0+0.0j", " NaN+0.0j", "-1.0+0.0j"]),
+        ([-1.23j, complex("nan"), -1], ["-0.00-1.23j", "  NaN+0.00j", "-1.00+0.00j"]),
+        ([1.23j, complex("nan"), 1.23], [" 0.00+1.23j", "  NaN+0.00j", " 1.23+0.00j"]),
+        (
+            [-1.23j, complex(np.nan, np.nan), 1],
+            ["-0.00-1.23j", "  NaN+ NaNj", " 1.00+0.00j"],
+        ),
+        (
+            [-1.23j, complex(1.2, np.nan), 1],
+            ["-0.00-1.23j", " 1.20+ NaNj", " 1.00+0.00j"],
+        ),
+        (
+            [-1.23j, complex(np.nan, -1.2), 1],
+            ["-0.00-1.23j", "  NaN-1.20j", " 1.00+0.00j"],
+        ),
+    ],
+)
+@pytest.mark.parametrize("as_frame", [True, False])
+def test_repr_with_complex_nans(data, output, as_frame):
+    # GH#53762, GH#53841
+    obj = Series(np.array(data))
+    if as_frame:
+        obj = obj.to_frame(name="val")
+        reprs = [f"{i} {val}" for i, val in enumerate(output)]
+        expected = f"{'val': >{len(reprs[0])}}\n" + "\n".join(reprs)
+    else:
+        reprs = [f"{i}   {val}" for i, val in enumerate(output)]
+        expected = "\n".join(reprs) + "\ndtype: complex128"
+    assert str(obj) == expected, f"\n{obj!s}\n\n{expected}"
diff --git a/pandas/pandas/tests/frame/test_stack_unstack.py b/pandas/pandas/tests/frame/test_stack_unstack.py
new file mode 100644
index 0000000000000..22fdfd3a01408
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_stack_unstack.py
@@ -0,0 +1,2730 @@
+from datetime import datetime
+import itertools
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    Series,
+    Timedelta,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.reshape import reshape as reshape_lib
+
+
+@pytest.fixture(params=[True, False])
+def future_stack(request):
+    return request.param
+
+
+class TestDataFrameReshape:
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_unstack(self, float_frame, future_stack):
+        df = float_frame.copy()
+        df[:] = np.arange(np.prod(df.shape)).reshape(df.shape)
+
+        stacked = df.stack(future_stack=future_stack)
+        stacked_df = DataFrame({"foo": stacked, "bar": stacked})
+
+        unstacked = stacked.unstack()
+        unstacked_df = stacked_df.unstack()
+
+        tm.assert_frame_equal(unstacked, df)
+        tm.assert_frame_equal(unstacked_df["bar"], df)
+
+        unstacked_cols = stacked.unstack(0)
+        unstacked_cols_df = stacked_df.unstack(0)
+        tm.assert_frame_equal(unstacked_cols.T, df)
+        tm.assert_frame_equal(unstacked_cols_df["bar"].T, df)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_mixed_level(self, future_stack):
+        # GH 18310
+        levels = [range(3), [3, "a", "b"], [1, 2]]
+
+        # flat columns:
+        df = DataFrame(1, index=levels[0], columns=levels[1])
+        result = df.stack(future_stack=future_stack)
+        expected = Series(1, index=MultiIndex.from_product(levels[:2]))
+        tm.assert_series_equal(result, expected)
+
+        # MultiIndex columns:
+        df = DataFrame(1, index=levels[0], columns=MultiIndex.from_product(levels[1:]))
+        result = df.stack(1, future_stack=future_stack)
+        expected = DataFrame(
+            1, index=MultiIndex.from_product([levels[0], levels[2]]), columns=levels[1]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # as above, but used labels in level are actually of homogeneous type
+        result = df[["a", "b"]].stack(1, future_stack=future_stack)
+        expected = expected[["a", "b"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_not_consolidated(self):
+        # Gh#34708
+        df = DataFrame({"x": [1, 2, np.nan], "y": [3.0, 4, np.nan]})
+        df2 = df[["x"]]
+        df2["y"] = df["y"]
+        assert len(df2._mgr.blocks) == 2
+
+        res = df2.unstack()
+        expected = df.unstack()
+        tm.assert_series_equal(res, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_fill(self, future_stack):
+        # GH #9746: fill_value keyword argument for Series
+        # and DataFrame unstack
+
+        # From a series
+        data = Series([1, 2, 4, 5], dtype=np.int16)
+        data.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        result = data.unstack(fill_value=-1)
+        expected = DataFrame(
+            {"a": [1, -1, 5], "b": [2, 4, -1]}, index=["x", "y", "z"], dtype=np.int16
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # From a series with incorrect data type for fill_value
+        result = data.unstack(fill_value=0.5)
+        expected = DataFrame(
+            {"a": [1, 0.5, 5], "b": [2, 4, 0.5]}, index=["x", "y", "z"], dtype=float
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # GH #13971: fill_value when unstacking multiple levels:
+        df = DataFrame(
+            {"x": ["a", "a", "b"], "y": ["j", "k", "j"], "z": [0, 1, 2], "w": [0, 1, 2]}
+        ).set_index(["x", "y", "z"])
+        unstacked = df.unstack(["x", "y"], fill_value=0)
+        key = ("w", "b", "j")
+        expected = unstacked[key]
+        result = Series([0, 0, 2], index=unstacked.index, name=key)
+        tm.assert_series_equal(result, expected)
+
+        stacked = unstacked.stack(["x", "y"], future_stack=future_stack)
+        stacked.index = stacked.index.reorder_levels(df.index.names)
+        # Workaround for GH #17886 (unnecessarily casts to float):
+        stacked = stacked.astype(np.int64)
+        result = stacked.loc[df.index]
+        tm.assert_frame_equal(result, df)
+
+        # From a series
+        s = df["w"]
+        result = s.unstack(["x", "y"], fill_value=0)
+        expected = unstacked["w"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_fill_frame(self):
+        # From a dataframe
+        rows = [[1, 2], [3, 4], [5, 6], [7, 8]]
+        df = DataFrame(rows, columns=list("AB"), dtype=np.int32)
+        df.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        result = df.unstack(fill_value=-1)
+
+        rows = [[1, 3, 2, 4], [-1, 5, -1, 6], [7, -1, 8, -1]]
+        expected = DataFrame(rows, index=list("xyz"), dtype=np.int32)
+        expected.columns = MultiIndex.from_tuples(
+            [("A", "a"), ("A", "b"), ("B", "a"), ("B", "b")]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # From a mixed type dataframe
+        df["A"] = df["A"].astype(np.int16)
+        df["B"] = df["B"].astype(np.float64)
+
+        result = df.unstack(fill_value=-1)
+        expected["A"] = expected["A"].astype(np.int16)
+        expected["B"] = expected["B"].astype(np.float64)
+        tm.assert_frame_equal(result, expected)
+
+        # From a dataframe with incorrect data type for fill_value
+        result = df.unstack(fill_value=0.5)
+
+        rows = [[1, 3, 2, 4], [0.5, 5, 0.5, 6], [7, 0.5, 8, 0.5]]
+        expected = DataFrame(rows, index=list("xyz"), dtype=float)
+        expected.columns = MultiIndex.from_tuples(
+            [("A", "a"), ("A", "b"), ("B", "a"), ("B", "b")]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_fill_frame_datetime(self):
+        # Test unstacking with date times
+        dv = date_range("2012-01-01", periods=4).values
+        data = Series(dv)
+        data.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        result = data.unstack()
+        expected = DataFrame(
+            {"a": [dv[0], pd.NaT, dv[3]], "b": [dv[1], dv[2], pd.NaT]},
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = data.unstack(fill_value=dv[0])
+        expected = DataFrame(
+            {"a": [dv[0], dv[0], dv[3]], "b": [dv[1], dv[2], dv[0]]},
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_fill_frame_timedelta(self):
+        # Test unstacking with time deltas
+        td = [Timedelta(days=i) for i in range(4)]
+        data = Series(td)
+        data.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        result = data.unstack()
+        expected = DataFrame(
+            {"a": [td[0], pd.NaT, td[3]], "b": [td[1], td[2], pd.NaT]},
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = data.unstack(fill_value=td[1])
+        expected = DataFrame(
+            {"a": [td[0], td[1], td[3]], "b": [td[1], td[2], td[1]]},
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_fill_frame_period(self):
+        # Test unstacking with period
+        periods = [
+            Period("2012-01"),
+            Period("2012-02"),
+            Period("2012-03"),
+            Period("2012-04"),
+        ]
+        data = Series(periods)
+        data.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        result = data.unstack()
+        expected = DataFrame(
+            {"a": [periods[0], None, periods[3]], "b": [periods[1], periods[2], None]},
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = data.unstack(fill_value=periods[1])
+        expected = DataFrame(
+            {
+                "a": [periods[0], periods[1], periods[3]],
+                "b": [periods[1], periods[2], periods[1]],
+            },
+            index=["x", "y", "z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_fill_frame_categorical(self):
+        # Test unstacking with categorical
+        data = Series(["a", "b", "c", "a"], dtype="category")
+        data.index = MultiIndex.from_tuples(
+            [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+        )
+
+        # By default missing values will be NaN
+        result = data.unstack()
+        expected = DataFrame(
+            {
+                "a": pd.Categorical(list("axa"), categories=list("abc")),
+                "b": pd.Categorical(list("bcx"), categories=list("abc")),
+            },
+            index=list("xyz"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Fill with non-category results in a ValueError
+        msg = r"Cannot setitem on a Categorical with a new category \(d\)"
+        with pytest.raises(TypeError, match=msg):
+            data.unstack(fill_value="d")
+
+        # Fill with category value replaces missing values as expected
+        result = data.unstack(fill_value="c")
+        expected = DataFrame(
+            {
+                "a": pd.Categorical(list("aca"), categories=list("abc")),
+                "b": pd.Categorical(list("bcc"), categories=list("abc")),
+            },
+            index=list("xyz"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_tuplename_in_multiindex(self):
+        # GH 19966
+        idx = MultiIndex.from_product(
+            [["a", "b", "c"], [1, 2, 3]], names=[("A", "a"), ("B", "b")]
+        )
+        df = DataFrame({"d": [1] * 9, "e": [2] * 9}, index=idx)
+        result = df.unstack(("A", "a"))
+
+        expected = DataFrame(
+            [[1, 1, 1, 2, 2, 2], [1, 1, 1, 2, 2, 2], [1, 1, 1, 2, 2, 2]],
+            columns=MultiIndex.from_tuples(
+                [
+                    ("d", "a"),
+                    ("d", "b"),
+                    ("d", "c"),
+                    ("e", "a"),
+                    ("e", "b"),
+                    ("e", "c"),
+                ],
+                names=[None, ("A", "a")],
+            ),
+            index=Index([1, 2, 3], name=("B", "b")),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "unstack_idx, expected_values, expected_index, expected_columns",
+        [
+            (
+                ("A", "a"),
+                [[1, 1, 2, 2], [1, 1, 2, 2], [1, 1, 2, 2], [1, 1, 2, 2]],
+                MultiIndex.from_tuples(
+                    [(1, 3), (1, 4), (2, 3), (2, 4)], names=["B", "C"]
+                ),
+                MultiIndex.from_tuples(
+                    [("d", "a"), ("d", "b"), ("e", "a"), ("e", "b")],
+                    names=[None, ("A", "a")],
+                ),
+            ),
+            (
+                (("A", "a"), "B"),
+                [[1, 1, 1, 1, 2, 2, 2, 2], [1, 1, 1, 1, 2, 2, 2, 2]],
+                Index([3, 4], name="C"),
+                MultiIndex.from_tuples(
+                    [
+                        ("d", "a", 1),
+                        ("d", "a", 2),
+                        ("d", "b", 1),
+                        ("d", "b", 2),
+                        ("e", "a", 1),
+                        ("e", "a", 2),
+                        ("e", "b", 1),
+                        ("e", "b", 2),
+                    ],
+                    names=[None, ("A", "a"), "B"],
+                ),
+            ),
+        ],
+    )
+    def test_unstack_mixed_type_name_in_multiindex(
+        self, unstack_idx, expected_values, expected_index, expected_columns
+    ):
+        # GH 19966
+        idx = MultiIndex.from_product(
+            [["a", "b"], [1, 2], [3, 4]], names=[("A", "a"), "B", "C"]
+        )
+        df = DataFrame({"d": [1] * 8, "e": [2] * 8}, index=idx)
+        result = df.unstack(unstack_idx)
+
+        expected = DataFrame(
+            expected_values, columns=expected_columns, index=expected_index
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_preserve_dtypes(self):
+        # Checks fix for #11847
+        df = DataFrame(
+            {
+                "state": ["IL", "MI", "NC"],
+                "index": ["a", "b", "c"],
+                "some_categories": Series(["a", "b", "c"]).astype("category"),
+                "A": np.random.default_rng(2).random(3),
+                "B": 1,
+                "C": "foo",
+                "D": pd.Timestamp("20010102"),
+                "E": Series([1.0, 50.0, 100.0]).astype("float32"),
+                "F": Series([3.0, 4.0, 5.0]).astype("float64"),
+                "G": False,
+                "H": Series([1, 200, 923442]).astype("int8"),
+            }
+        )
+
+        def unstack_and_compare(df, column_name):
+            unstacked1 = df.unstack([column_name])
+            unstacked2 = df.unstack(column_name)
+            tm.assert_frame_equal(unstacked1, unstacked2)
+
+        df1 = df.set_index(["state", "index"])
+        unstack_and_compare(df1, "index")
+
+        df1 = df.set_index(["state", "some_categories"])
+        unstack_and_compare(df1, "some_categories")
+
+        df1 = df.set_index(["F", "C"])
+        unstack_and_compare(df1, "F")
+
+        df1 = df.set_index(["G", "B", "state"])
+        unstack_and_compare(df1, "B")
+
+        df1 = df.set_index(["E", "A"])
+        unstack_and_compare(df1, "E")
+
+        df1 = df.set_index(["state", "index"])
+        s = df1["A"]
+        unstack_and_compare(s, "index")
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_ints(self, future_stack):
+        columns = MultiIndex.from_tuples(list(itertools.product(range(3), repeat=3)))
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 27)), columns=columns
+        )
+
+        tm.assert_frame_equal(
+            df.stack(level=[1, 2], future_stack=future_stack),
+            df.stack(level=1, future_stack=future_stack).stack(
+                level=1, future_stack=future_stack
+            ),
+        )
+        tm.assert_frame_equal(
+            df.stack(level=[-2, -1], future_stack=future_stack),
+            df.stack(level=1, future_stack=future_stack).stack(
+                level=1, future_stack=future_stack
+            ),
+        )
+
+        df_named = df.copy()
+        return_value = df_named.columns.set_names(range(3), inplace=True)
+        assert return_value is None
+
+        tm.assert_frame_equal(
+            df_named.stack(level=[1, 2], future_stack=future_stack),
+            df_named.stack(level=1, future_stack=future_stack).stack(
+                level=1, future_stack=future_stack
+            ),
+        )
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_mixed_levels(self, future_stack):
+        columns = MultiIndex.from_tuples(
+            [
+                ("A", "cat", "long"),
+                ("B", "cat", "long"),
+                ("A", "dog", "short"),
+                ("B", "dog", "short"),
+            ],
+            names=["exp", "animal", "hair_length"],
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), columns=columns
+        )
+
+        animal_hair_stacked = df.stack(
+            level=["animal", "hair_length"], future_stack=future_stack
+        )
+        exp_hair_stacked = df.stack(
+            level=["exp", "hair_length"], future_stack=future_stack
+        )
+
+        # GH #8584: Need to check that stacking works when a number
+        # is passed that is both a level name and in the range of
+        # the level numbers
+        df2 = df.copy()
+        df2.columns.names = ["exp", "animal", 1]
+        tm.assert_frame_equal(
+            df2.stack(level=["animal", 1], future_stack=future_stack),
+            animal_hair_stacked,
+            check_names=False,
+        )
+        tm.assert_frame_equal(
+            df2.stack(level=["exp", 1], future_stack=future_stack),
+            exp_hair_stacked,
+            check_names=False,
+        )
+
+        # When mixed types are passed and the ints are not level
+        # names, raise
+        msg = (
+            "level should contain all level names or all level numbers, not "
+            "a mixture of the two"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df2.stack(level=["animal", 0], future_stack=future_stack)
+
+        # GH #8584: Having 0 in the level names could raise a
+        # strange error about lexsort depth
+        df3 = df.copy()
+        df3.columns.names = ["exp", "animal", 0]
+        tm.assert_frame_equal(
+            df3.stack(level=["animal", 0], future_stack=future_stack),
+            animal_hair_stacked,
+            check_names=False,
+        )
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_int_level_names(self, future_stack):
+        columns = MultiIndex.from_tuples(
+            [
+                ("A", "cat", "long"),
+                ("B", "cat", "long"),
+                ("A", "dog", "short"),
+                ("B", "dog", "short"),
+            ],
+            names=["exp", "animal", "hair_length"],
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), columns=columns
+        )
+
+        exp_animal_stacked = df.stack(
+            level=["exp", "animal"], future_stack=future_stack
+        )
+        animal_hair_stacked = df.stack(
+            level=["animal", "hair_length"], future_stack=future_stack
+        )
+        exp_hair_stacked = df.stack(
+            level=["exp", "hair_length"], future_stack=future_stack
+        )
+
+        df2 = df.copy()
+        df2.columns.names = [0, 1, 2]
+        tm.assert_frame_equal(
+            df2.stack(level=[1, 2], future_stack=future_stack),
+            animal_hair_stacked,
+            check_names=False,
+        )
+        tm.assert_frame_equal(
+            df2.stack(level=[0, 1], future_stack=future_stack),
+            exp_animal_stacked,
+            check_names=False,
+        )
+        tm.assert_frame_equal(
+            df2.stack(level=[0, 2], future_stack=future_stack),
+            exp_hair_stacked,
+            check_names=False,
+        )
+
+        # Out-of-order int column names
+        df3 = df.copy()
+        df3.columns.names = [2, 0, 1]
+        tm.assert_frame_equal(
+            df3.stack(level=[0, 1], future_stack=future_stack),
+            animal_hair_stacked,
+            check_names=False,
+        )
+        tm.assert_frame_equal(
+            df3.stack(level=[2, 0], future_stack=future_stack),
+            exp_animal_stacked,
+            check_names=False,
+        )
+        tm.assert_frame_equal(
+            df3.stack(level=[2, 1], future_stack=future_stack),
+            exp_hair_stacked,
+            check_names=False,
+        )
+
+    def test_unstack_bool(self):
+        df = DataFrame(
+            [False, False],
+            index=MultiIndex.from_arrays([["a", "b"], ["c", "l"]]),
+            columns=["col"],
+        )
+        rs = df.unstack()
+        xp = DataFrame(
+            np.array([[False, np.nan], [np.nan, False]], dtype=object),
+            index=["a", "b"],
+            columns=MultiIndex.from_arrays([["col", "col"], ["c", "l"]]),
+        )
+        tm.assert_frame_equal(rs, xp)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_level_binding(self, future_stack):
+        # GH9856
+        mi = MultiIndex(
+            levels=[["foo", "bar"], ["one", "two"], ["a", "b"]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1], [1, 0, 1, 0]],
+            names=["first", "second", "third"],
+        )
+        s = Series(0, index=mi)
+        result = s.unstack([1, 2]).stack(0, future_stack=future_stack)
+
+        expected_mi = MultiIndex(
+            levels=[["foo", "bar"], ["one", "two"]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+            names=["first", "second"],
+        )
+
+        expected = DataFrame(
+            np.array(
+                [[0, np.nan], [np.nan, 0], [0, np.nan], [np.nan, 0]], dtype=np.float64
+            ),
+            index=expected_mi,
+            columns=Index(["b", "a"], name="third"),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_to_series(self, float_frame):
+        # check reversibility
+        data = float_frame.unstack()
+
+        assert isinstance(data, Series)
+        undo = data.unstack().T
+        tm.assert_frame_equal(undo, float_frame)
+
+        # check NA handling
+        data = DataFrame({"x": [1, 2, np.nan], "y": [3.0, 4, np.nan]})
+        data.index = Index(["a", "b", "c"])
+        result = data.unstack()
+
+        midx = MultiIndex(
+            levels=[["x", "y"], ["a", "b", "c"]],
+            codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]],
+        )
+        expected = Series([1, 2, np.nan, 3, 4, np.nan], index=midx)
+
+        tm.assert_series_equal(result, expected)
+
+        # check composability of unstack
+        old_data = data.copy()
+        for _ in range(4):
+            data = data.unstack()
+        tm.assert_frame_equal(old_data, data)
+
+    def test_unstack_dtypes(self, using_infer_string):
+        # GH 2929
+        rows = [[1, 1, 3, 4], [1, 2, 3, 4], [2, 1, 3, 4], [2, 2, 3, 4]]
+
+        df = DataFrame(rows, columns=list("ABCD"))
+        result = df.dtypes
+        expected = Series([np.dtype("int64")] * 4, index=list("ABCD"))
+        tm.assert_series_equal(result, expected)
+
+        # single dtype
+        df2 = df.set_index(["A", "B"])
+        df3 = df2.unstack("B")
+        result = df3.dtypes
+        expected = Series(
+            [np.dtype("int64")] * 4,
+            index=MultiIndex.from_arrays(
+                [["C", "C", "D", "D"], [1, 2, 1, 2]], names=(None, "B")
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+
+        # mixed
+        df2 = df.set_index(["A", "B"])
+        df2["C"] = 3.0
+        df3 = df2.unstack("B")
+        result = df3.dtypes
+        expected = Series(
+            [np.dtype("float64")] * 2 + [np.dtype("int64")] * 2,
+            index=MultiIndex.from_arrays(
+                [["C", "C", "D", "D"], [1, 2, 1, 2]], names=(None, "B")
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+        df2["D"] = "foo"
+        df3 = df2.unstack("B")
+        result = df3.dtypes
+        dtype = (
+            pd.StringDtype(na_value=np.nan)
+            if using_infer_string
+            else np.dtype("object")
+        )
+        expected = Series(
+            [np.dtype("float64")] * 2 + [dtype] * 2,
+            index=MultiIndex.from_arrays(
+                [["C", "C", "D", "D"], [1, 2, 1, 2]], names=(None, "B")
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "c, d",
+        (
+            (np.zeros(5), np.zeros(5)),
+            (np.arange(5, dtype="f8"), np.arange(5, 10, dtype="f8")),
+        ),
+    )
+    def test_unstack_dtypes_mixed_date(self, c, d):
+        # GH7405
+        df = DataFrame(
+            {
+                "A": ["a"] * 5,
+                "C": c,
+                "D": d,
+                "B": date_range("2012-01-01", periods=5),
+            }
+        )
+
+        right = df.iloc[:3].copy(deep=True)
+
+        df = df.set_index(["A", "B"])
+        df["D"] = df["D"].astype("int64")
+
+        left = df.iloc[:3].unstack(0)
+        right = right.set_index(["A", "B"]).unstack(0)
+        right[("D", "a")] = right[("D", "a")].astype("int64")
+
+        assert left.shape == (3, 2)
+        tm.assert_frame_equal(left, right)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_non_unique_index_names(self, future_stack):
+        idx = MultiIndex.from_tuples([("a", "b"), ("c", "d")], names=["c1", "c1"])
+        df = DataFrame([1, 2], index=idx)
+        msg = "The name c1 occurs multiple times, use a level number"
+        with pytest.raises(ValueError, match=msg):
+            df.unstack("c1")
+
+        with pytest.raises(ValueError, match=msg):
+            df.T.stack("c1", future_stack=future_stack)
+
+    def test_unstack_unused_levels(self):
+        # GH 17845: unused codes in index make unstack() cast int to float
+        idx = MultiIndex.from_product([["a"], ["A", "B", "C", "D"]])[:-1]
+        df = DataFrame([[1, 0]] * 3, index=idx)
+
+        result = df.unstack()
+        exp_col = MultiIndex.from_product([range(2), ["A", "B", "C"]])
+        expected = DataFrame([[1, 1, 1, 0, 0, 0]], index=["a"], columns=exp_col)
+        tm.assert_frame_equal(result, expected)
+        assert (result.columns.levels[1] == idx.levels[1]).all()
+
+        # Unused items on both levels
+        levels = [range(3), range(4)]
+        codes = [[0, 0, 1, 1], [0, 2, 0, 2]]
+        idx = MultiIndex(levels, codes)
+        block = np.arange(4).reshape(2, 2)
+        df = DataFrame(np.concatenate([block, block + 4]), index=idx)
+        result = df.unstack()
+        expected = DataFrame(
+            np.concatenate([block * 2, block * 2 + 1], axis=1), columns=idx
+        )
+        tm.assert_frame_equal(result, expected)
+        assert (result.columns.levels[1] == idx.levels[1]).all()
+
+    @pytest.mark.parametrize(
+        "level, idces, col_level, idx_level",
+        (
+            (0, [13, 16, 6, 9, 2, 5, 8, 11], [np.nan, "a", 2], [np.nan, 5, 1]),
+            (1, [8, 11, 1, 4, 12, 15, 13, 16], [np.nan, 5, 1], [np.nan, "a", 2]),
+        ),
+    )
+    def test_unstack_unused_levels_mixed_with_nan(
+        self, level, idces, col_level, idx_level
+    ):
+        # With mixed dtype and NaN
+        levels = [["a", 2, "c"], [1, 3, 5, 7]]
+        codes = [[0, -1, 1, 1], [0, 2, -1, 2]]
+        idx = MultiIndex(levels, codes)
+        data = np.arange(8)
+        df = DataFrame(data.reshape(4, 2), index=idx)
+
+        result = df.unstack(level=level)
+        exp_data = np.zeros(18) * np.nan
+        exp_data[idces] = data
+        cols = MultiIndex.from_product([range(2), col_level])
+        expected = DataFrame(exp_data.reshape(3, 6), index=idx_level, columns=cols)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("cols", [["A", "C"], slice(None)])
+    def test_unstack_unused_level(self, cols):
+        # GH 18562 : unused codes on the unstacked level
+        df = DataFrame([[2010, "a", "I"], [2011, "b", "II"]], columns=["A", "B", "C"])
+
+        ind = df.set_index(["A", "B", "C"], drop=False)
+        selection = ind.loc[(slice(None), slice(None), "I"), cols]
+        result = selection.unstack()
+
+        expected = ind.iloc[[0]][cols]
+        expected.columns = MultiIndex.from_product(
+            [expected.columns, ["I"]], names=[None, "C"]
+        )
+        expected.index = expected.index.droplevel("C")
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_long_index(self):
+        # PH 32624: Error when using a lot of indices to unstack.
+        # The error occurred only, if a lot of indices are used.
+        df = DataFrame(
+            [[1]],
+            columns=MultiIndex.from_tuples([[0]], names=["c1"]),
+            index=MultiIndex.from_tuples(
+                [[0, 0, 1, 0, 0, 0, 1]],
+                names=["i1", "i2", "i3", "i4", "i5", "i6", "i7"],
+            ),
+        )
+        result = df.unstack(["i2", "i3", "i4", "i5", "i6", "i7"])
+        expected = DataFrame(
+            [[1]],
+            columns=MultiIndex.from_tuples(
+                [[0, 0, 1, 0, 0, 0, 1]],
+                names=["c1", "i2", "i3", "i4", "i5", "i6", "i7"],
+            ),
+            index=Index([0], name="i1"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_multi_level_cols(self):
+        # PH 24729: Unstack a df with multi level columns
+        df = DataFrame(
+            [[0.0, 0.0], [0.0, 0.0]],
+            columns=MultiIndex.from_tuples(
+                [["B", "C"], ["B", "D"]], names=["c1", "c2"]
+            ),
+            index=MultiIndex.from_tuples(
+                [[10, 20, 30], [10, 20, 40]], names=["i1", "i2", "i3"]
+            ),
+        )
+        assert df.unstack(["i2", "i1"]).columns.names[-2:] == ["i2", "i1"]
+
+    def test_unstack_multi_level_rows_and_cols(self):
+        # PH 28306: Unstack df with multi level cols and rows
+        df = DataFrame(
+            [[1, 2], [3, 4], [-1, -2], [-3, -4]],
+            columns=MultiIndex.from_tuples([["a", "b", "c"], ["d", "e", "f"]]),
+            index=MultiIndex.from_tuples(
+                [
+                    ["m1", "P3", 222],
+                    ["m1", "A5", 111],
+                    ["m2", "P3", 222],
+                    ["m2", "A5", 111],
+                ],
+                names=["i1", "i2", "i3"],
+            ),
+        )
+        result = df.unstack(["i3", "i2"])
+        expected = df.unstack(["i3"]).unstack(["i2"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("idx", [("jim", "joe"), ("joe", "jim")])
+    @pytest.mark.parametrize("lev", list(range(2)))
+    def test_unstack_nan_index1(self, idx, lev):
+        # GH7466
+        def cast(val):
+            val_str = "" if val != val else val
+            return f"{val_str:1}"
+
+        df = DataFrame(
+            {
+                "jim": ["a", "b", np.nan, "d"],
+                "joe": ["w", "x", "y", "z"],
+                "jolie": ["a.w", "b.x", " .y", "d.z"],
+            }
+        )
+
+        left = df.set_index(["jim", "joe"]).unstack()["jolie"]
+        right = df.set_index(["joe", "jim"]).unstack()["jolie"].T
+        tm.assert_frame_equal(left, right)
+
+        mi = df.set_index(list(idx))
+        udf = mi.unstack(level=lev)
+        assert udf.notna().values.sum() == len(df)
+        mk_list = lambda a: list(a) if isinstance(a, tuple) else [a]
+        rows, cols = udf["jolie"].notna().values.nonzero()
+        for i, j in zip(rows, cols):
+            left = sorted(udf["jolie"].iloc[i, j].split("."))
+            right = mk_list(udf["jolie"].index[i]) + mk_list(udf["jolie"].columns[j])
+            right = sorted(map(cast, right))
+            assert left == right
+
+    @pytest.mark.parametrize("idx", itertools.permutations(["1st", "2nd", "3rd"]))
+    @pytest.mark.parametrize("lev", list(range(3)))
+    @pytest.mark.parametrize("col", ["4th", "5th"])
+    def test_unstack_nan_index_repeats(self, idx, lev, col):
+        def cast(val):
+            val_str = "" if val != val else val
+            return f"{val_str:1}"
+
+        df = DataFrame(
+            {
+                "1st": ["d"] * 3
+                + [np.nan] * 5
+                + ["a"] * 2
+                + ["c"] * 3
+                + ["e"] * 2
+                + ["b"] * 5,
+                "2nd": ["y"] * 2
+                + ["w"] * 3
+                + [np.nan] * 3
+                + ["z"] * 4
+                + [np.nan] * 3
+                + ["x"] * 3
+                + [np.nan] * 2,
+                "3rd": [
+                    67,
+                    39,
+                    53,
+                    72,
+                    57,
+                    80,
+                    31,
+                    18,
+                    11,
+                    30,
+                    59,
+                    50,
+                    62,
+                    59,
+                    76,
+                    52,
+                    14,
+                    53,
+                    60,
+                    51,
+                ],
+            }
+        )
+
+        df["4th"], df["5th"] = (
+            df.apply(lambda r: ".".join(map(cast, r)), axis=1),
+            df.apply(lambda r: ".".join(map(cast, r.iloc[::-1])), axis=1),
+        )
+
+        mi = df.set_index(list(idx))
+        udf = mi.unstack(level=lev)
+        assert udf.notna().values.sum() == 2 * len(df)
+        mk_list = lambda a: list(a) if isinstance(a, tuple) else [a]
+        rows, cols = udf[col].notna().values.nonzero()
+        for i, j in zip(rows, cols):
+            left = sorted(udf[col].iloc[i, j].split("."))
+            right = mk_list(udf[col].index[i]) + mk_list(udf[col].columns[j])
+            right = sorted(map(cast, right))
+            assert left == right
+
+    def test_unstack_nan_index2(self):
+        # GH7403
+        df = DataFrame({"A": list("aaaabbbb"), "B": range(8), "C": range(8)})
+        # Explicit cast to avoid implicit cast when setting to np.nan
+        df = df.astype({"B": "float"})
+        df.iloc[3, 1] = np.nan
+        left = df.set_index(["A", "B"]).unstack(0)
+
+        vals = [
+            [3, 0, 1, 2, np.nan, np.nan, np.nan, np.nan],
+            [np.nan, np.nan, np.nan, np.nan, 4, 5, 6, 7],
+        ]
+        vals = list(map(list, zip(*vals)))
+        idx = Index([np.nan, 0, 1, 2, 4, 5, 6, 7], name="B")
+        cols = MultiIndex(
+            levels=[["C"], ["a", "b"]], codes=[[0, 0], [0, 1]], names=[None, "A"]
+        )
+
+        right = DataFrame(vals, columns=cols, index=idx)
+        tm.assert_frame_equal(left, right)
+
+        df = DataFrame({"A": list("aaaabbbb"), "B": list(range(4)) * 2, "C": range(8)})
+        # Explicit cast to avoid implicit cast when setting to np.nan
+        df = df.astype({"B": "float"})
+        df.iloc[2, 1] = np.nan
+        left = df.set_index(["A", "B"]).unstack(0)
+
+        vals = [[2, np.nan], [0, 4], [1, 5], [np.nan, 6], [3, 7]]
+        cols = MultiIndex(
+            levels=[["C"], ["a", "b"]], codes=[[0, 0], [0, 1]], names=[None, "A"]
+        )
+        idx = Index([np.nan, 0, 1, 2, 3], name="B")
+        right = DataFrame(vals, columns=cols, index=idx)
+        tm.assert_frame_equal(left, right)
+
+        df = DataFrame({"A": list("aaaabbbb"), "B": list(range(4)) * 2, "C": range(8)})
+        # Explicit cast to avoid implicit cast when setting to np.nan
+        df = df.astype({"B": "float"})
+        df.iloc[3, 1] = np.nan
+        left = df.set_index(["A", "B"]).unstack(0)
+
+        vals = [[3, np.nan], [0, 4], [1, 5], [2, 6], [np.nan, 7]]
+        cols = MultiIndex(
+            levels=[["C"], ["a", "b"]], codes=[[0, 0], [0, 1]], names=[None, "A"]
+        )
+        idx = Index([np.nan, 0, 1, 2, 3], name="B")
+        right = DataFrame(vals, columns=cols, index=idx)
+        tm.assert_frame_equal(left, right)
+
+    def test_unstack_nan_index3(self):
+        # GH7401
+        df = DataFrame(
+            {
+                "A": list("aaaaabbbbb"),
+                "B": (date_range("2012-01-01", periods=5).tolist() * 2),
+                "C": np.arange(10),
+            }
+        )
+
+        df.iloc[3, 1] = np.nan
+        left = df.set_index(["A", "B"]).unstack()
+
+        vals = np.array([[3, 0, 1, 2, np.nan, 4], [np.nan, 5, 6, 7, 8, 9]])
+        idx = Index(["a", "b"], name="A")
+        cols = MultiIndex(
+            levels=[["C"], date_range("2012-01-01", periods=5)],
+            codes=[[0, 0, 0, 0, 0, 0], [-1, 0, 1, 2, 3, 4]],
+            names=[None, "B"],
+        )
+
+        right = DataFrame(vals, columns=cols, index=idx)
+        tm.assert_frame_equal(left, right)
+
+    def test_unstack_nan_index4(self):
+        # GH4862
+        vals = [
+            ["Hg", np.nan, np.nan, 680585148],
+            ["U", 0.0, np.nan, 680585148],
+            ["Pb", 7.07e-06, np.nan, 680585148],
+            ["Sn", 2.3614e-05, 0.0133, 680607017],
+            ["Ag", 0.0, 0.0133, 680607017],
+            ["Hg", -0.00015, 0.0133, 680607017],
+        ]
+        df = DataFrame(
+            vals,
+            columns=["agent", "change", "dosage", "s_id"],
+            index=[17263, 17264, 17265, 17266, 17267, 17268],
+        )
+
+        left = df.copy().set_index(["s_id", "dosage", "agent"]).unstack()
+
+        vals = [
+            [np.nan, np.nan, 7.07e-06, np.nan, 0.0],
+            [0.0, -0.00015, np.nan, 2.3614e-05, np.nan],
+        ]
+
+        idx = MultiIndex(
+            levels=[[680585148, 680607017], [0.0133]],
+            codes=[[0, 1], [-1, 0]],
+            names=["s_id", "dosage"],
+        )
+
+        cols = MultiIndex(
+            levels=[["change"], ["Ag", "Hg", "Pb", "Sn", "U"]],
+            codes=[[0, 0, 0, 0, 0], [0, 1, 2, 3, 4]],
+            names=[None, "agent"],
+        )
+
+        right = DataFrame(vals, columns=cols, index=idx)
+        tm.assert_frame_equal(left, right)
+
+        left = df.loc[17264:].copy().set_index(["s_id", "dosage", "agent"])
+        tm.assert_frame_equal(left.unstack(), right)
+
+    def test_unstack_nan_index5(self):
+        # GH9497 - multiple unstack with nulls
+        df = DataFrame(
+            {
+                "1st": [1, 2, 1, 2, 1, 2],
+                "2nd": date_range("2014-02-01", periods=6, freq="D"),
+                "jim": 100 + np.arange(6),
+                "joe": (np.random.default_rng(2).standard_normal(6) * 10).round(2),
+            }
+        )
+
+        df["3rd"] = df["2nd"] - pd.Timestamp("2014-02-02")
+        df.loc[1, "2nd"] = df.loc[3, "2nd"] = np.nan
+        df.loc[1, "3rd"] = df.loc[4, "3rd"] = np.nan
+
+        left = df.set_index(["1st", "2nd", "3rd"]).unstack(["2nd", "3rd"])
+        assert left.notna().values.sum() == 2 * len(df)
+
+        for col in ["jim", "joe"]:
+            for _, r in df.iterrows():
+                key = r["1st"], (col, r["2nd"], r["3rd"])
+                assert r[col] == left.loc[key]
+
+    def test_stack_datetime_column_multiIndex(self, future_stack):
+        # GH 8039
+        t = datetime(2014, 1, 1)
+        df = DataFrame([1, 2, 3, 4], columns=MultiIndex.from_tuples([(t, "A", "B")]))
+        warn = None if future_stack else FutureWarning
+        msg = "The previous implementation of stack is deprecated"
+        with tm.assert_produces_warning(warn, match=msg):
+            result = df.stack(future_stack=future_stack)
+
+        eidx = MultiIndex.from_product([range(4), ("B",)])
+        ecols = MultiIndex.from_tuples([(t, "A")])
+        expected = DataFrame([1, 2, 3, 4], index=eidx, columns=ecols)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "multiindex_columns",
+        [
+            [0, 1, 2, 3, 4],
+            [0, 1, 2, 3],
+            [0, 1, 2, 4],
+            [0, 1, 2],
+            [1, 2, 3],
+            [2, 3, 4],
+            [0, 1],
+            [0, 2],
+            [0, 3],
+            [0],
+            [2],
+            [4],
+            [4, 3, 2, 1, 0],
+            [3, 2, 1, 0],
+            [4, 2, 1, 0],
+            [2, 1, 0],
+            [3, 2, 1],
+            [4, 3, 2],
+            [1, 0],
+            [2, 0],
+            [3, 0],
+        ],
+    )
+    @pytest.mark.parametrize("level", (-1, 0, 1, [0, 1], [1, 0]))
+    def test_stack_partial_multiIndex(self, multiindex_columns, level, future_stack):
+        # GH 8844
+        dropna = False if not future_stack else lib.no_default
+        full_multiindex = MultiIndex.from_tuples(
+            [("B", "x"), ("B", "z"), ("A", "y"), ("C", "x"), ("C", "u")],
+            names=["Upper", "Lower"],
+        )
+        multiindex = full_multiindex[multiindex_columns]
+        df = DataFrame(
+            np.arange(3 * len(multiindex)).reshape(3, len(multiindex)),
+            columns=multiindex,
+        )
+        result = df.stack(level=level, dropna=dropna, future_stack=future_stack)
+
+        if isinstance(level, int) and not future_stack:
+            # Stacking a single level should not make any all-NaN rows,
+            # so df.stack(level=level, dropna=False) should be the same
+            # as df.stack(level=level, dropna=True).
+            expected = df.stack(level=level, dropna=True, future_stack=future_stack)
+            if isinstance(expected, Series):
+                tm.assert_series_equal(result, expected)
+            else:
+                tm.assert_frame_equal(result, expected)
+
+        df.columns = MultiIndex.from_tuples(
+            df.columns.to_numpy(), names=df.columns.names
+        )
+        expected = df.stack(level=level, dropna=dropna, future_stack=future_stack)
+        if isinstance(expected, Series):
+            tm.assert_series_equal(result, expected)
+        else:
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_full_multiIndex(self, future_stack):
+        # GH 8844
+        full_multiindex = MultiIndex.from_tuples(
+            [("B", "x"), ("B", "z"), ("A", "y"), ("C", "x"), ("C", "u")],
+            names=["Upper", "Lower"],
+        )
+        df = DataFrame(np.arange(6).reshape(2, 3), columns=full_multiindex[[0, 1, 3]])
+        dropna = False if not future_stack else lib.no_default
+        result = df.stack(dropna=dropna, future_stack=future_stack)
+        expected = DataFrame(
+            [[0, 2], [1, np.nan], [3, 5], [4, np.nan]],
+            index=MultiIndex(
+                levels=[range(2), ["u", "x", "y", "z"]],
+                codes=[[0, 0, 1, 1], [1, 3, 1, 3]],
+                names=[None, "Lower"],
+            ),
+            columns=Index(["B", "C"], name="Upper"),
+        )
+        expected["B"] = expected["B"].astype(df.dtypes.iloc[0])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize("ordered", [False, True])
+    def test_stack_preserve_categorical_dtype(self, ordered, future_stack):
+        # GH13854
+        cidx = pd.CategoricalIndex(list("yxz"), categories=list("xyz"), ordered=ordered)
+        df = DataFrame([[10, 11, 12]], columns=cidx)
+        result = df.stack(future_stack=future_stack)
+
+        # `MultiIndex.from_product` preserves categorical dtype -
+        # it's tested elsewhere.
+        midx = MultiIndex.from_product([df.index, cidx])
+        expected = Series([10, 11, 12], index=midx)
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize("ordered", [False, True])
+    @pytest.mark.parametrize(
+        "labels,data",
+        [
+            (list("xyz"), [10, 11, 12, 13, 14, 15]),
+            (list("zyx"), [14, 15, 12, 13, 10, 11]),
+        ],
+    )
+    def test_stack_multi_preserve_categorical_dtype(
+        self, ordered, labels, data, future_stack
+    ):
+        # GH-36991
+        cidx = pd.CategoricalIndex(labels, categories=sorted(labels), ordered=ordered)
+        cidx2 = pd.CategoricalIndex(["u", "v"], ordered=ordered)
+        midx = MultiIndex.from_product([cidx, cidx2])
+        df = DataFrame([sorted(data)], columns=midx)
+        result = df.stack([0, 1], future_stack=future_stack)
+
+        labels = labels if future_stack else sorted(labels)
+        s_cidx = pd.CategoricalIndex(labels, ordered=ordered)
+        expected_data = sorted(data) if future_stack else data
+        expected = Series(
+            expected_data, index=MultiIndex.from_product([range(1), s_cidx, cidx2])
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_preserve_categorical_dtype_values(self, future_stack):
+        # GH-23077
+        cat = pd.Categorical(["a", "a", "b", "c"])
+        df = DataFrame({"A": cat, "B": cat})
+        result = df.stack(future_stack=future_stack)
+        index = MultiIndex.from_product([range(4), ["A", "B"]])
+        expected = Series(
+            pd.Categorical(["a", "a", "a", "a", "b", "b", "c", "c"]), index=index
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "index",
+        [
+            [0, 0, 1, 1],
+            [0, 0, 2, 3],
+            [0, 1, 2, 3],
+        ],
+    )
+    def test_stack_multi_columns_non_unique_index(self, index, future_stack):
+        # GH-28301
+        columns = MultiIndex.from_product([[1, 2], ["a", "b"]])
+        df = DataFrame(index=index, columns=columns).fillna(1)
+        stacked = df.stack(future_stack=future_stack)
+        new_index = MultiIndex.from_tuples(stacked.index.to_numpy())
+        expected = DataFrame(
+            stacked.to_numpy(), index=new_index, columns=stacked.columns
+        )
+        tm.assert_frame_equal(stacked, expected)
+        stacked_codes = np.asarray(stacked.index.codes)
+        expected_codes = np.asarray(new_index.codes)
+        tm.assert_numpy_array_equal(stacked_codes, expected_codes)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "vals1, vals2, dtype1, dtype2, expected_dtype",
+        [
+            ([1, 2], [3.0, 4.0], "Int64", "Float64", "Float64"),
+            ([1, 2], ["foo", "bar"], "Int64", "string", "object"),
+        ],
+    )
+    def test_stack_multi_columns_mixed_extension_types(
+        self, vals1, vals2, dtype1, dtype2, expected_dtype, future_stack
+    ):
+        # GH45740
+        df = DataFrame(
+            {
+                ("A", 1): Series(vals1, dtype=dtype1),
+                ("A", 2): Series(vals2, dtype=dtype2),
+            }
+        )
+        result = df.stack(future_stack=future_stack)
+        expected = (
+            df.astype(object).stack(future_stack=future_stack).astype(expected_dtype)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("level", [0, 1])
+    def test_unstack_mixed_extension_types(self, level):
+        index = MultiIndex.from_tuples([("A", 0), ("A", 1), ("B", 1)], names=["a", "b"])
+        df = DataFrame(
+            {
+                "A": pd.array([0, 1, None], dtype="Int64"),
+                "B": pd.Categorical(["a", "a", "b"]),
+            },
+            index=index,
+        )
+
+        result = df.unstack(level=level)
+        expected = df.astype(object).unstack(level=level)
+        if level == 0:
+            expected[("A", "B")] = expected[("A", "B")].fillna(pd.NA)
+        else:
+            expected[("A", 0)] = expected[("A", 0)].fillna(pd.NA)
+
+        expected_dtypes = Series(
+            [df.A.dtype] * 2 + [df.B.dtype] * 2, index=result.columns
+        )
+        tm.assert_series_equal(result.dtypes, expected_dtypes)
+        tm.assert_frame_equal(result.astype(object), expected)
+
+    @pytest.mark.parametrize("level", [0, "baz"])
+    def test_unstack_swaplevel_sortlevel(self, level):
+        # GH 20994
+        mi = MultiIndex.from_product([range(1), ["d", "c"]], names=["bar", "baz"])
+        df = DataFrame([[0, 2], [1, 3]], index=mi, columns=["B", "A"])
+        df.columns.name = "foo"
+
+        expected = DataFrame(
+            [[3, 1, 2, 0]],
+            columns=MultiIndex.from_tuples(
+                [("c", "A"), ("c", "B"), ("d", "A"), ("d", "B")], names=["baz", "foo"]
+            ),
+        )
+        expected.index.name = "bar"
+
+        result = df.unstack().swaplevel(axis=1).sort_index(axis=1, level=level)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["float64", "Float64"])
+def test_unstack_sort_false(frame_or_series, dtype):
+    # GH 15105
+    index = MultiIndex.from_tuples(
+        [("two", "z", "b"), ("two", "y", "a"), ("one", "z", "b"), ("one", "y", "a")]
+    )
+    obj = frame_or_series(np.arange(1.0, 5.0), index=index, dtype=dtype)
+
+    result = obj.unstack(level=0, sort=False)
+
+    if frame_or_series is DataFrame:
+        expected_columns = MultiIndex.from_tuples([(0, "two"), (0, "one")])
+    else:
+        expected_columns = ["two", "one"]
+    expected = DataFrame(
+        [[1.0, 3.0], [2.0, 4.0]],
+        index=MultiIndex.from_tuples([("z", "b"), ("y", "a")]),
+        columns=expected_columns,
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = obj.unstack(level=-1, sort=False)
+
+    if frame_or_series is DataFrame:
+        expected_columns = MultiIndex(
+            levels=[range(1), ["b", "a"]], codes=[[0, 0], [0, 1]]
+        )
+    else:
+        expected_columns = ["b", "a"]
+    expected = DataFrame(
+        [[1.0, np.nan], [np.nan, 2.0], [3.0, np.nan], [np.nan, 4.0]],
+        columns=expected_columns,
+        index=MultiIndex.from_tuples(
+            [("two", "z"), ("two", "y"), ("one", "z"), ("one", "y")]
+        ),
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = obj.unstack(level=[1, 2], sort=False)
+
+    if frame_or_series is DataFrame:
+        expected_columns = MultiIndex(
+            levels=[range(1), ["z", "y"], ["b", "a"]], codes=[[0, 0], [0, 1], [0, 1]]
+        )
+    else:
+        expected_columns = MultiIndex.from_tuples([("z", "b"), ("y", "a")])
+    expected = DataFrame(
+        [[1.0, 2.0], [3.0, 4.0]],
+        index=["two", "one"],
+        columns=expected_columns,
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_fill_frame_object():
+    # GH12815 Test unstacking with object.
+    data = Series(["a", "b", "c", "a"], dtype="object")
+    data.index = MultiIndex.from_tuples(
+        [("x", "a"), ("x", "b"), ("y", "b"), ("z", "a")]
+    )
+
+    # By default missing values will be NaN
+    result = data.unstack()
+    expected = DataFrame(
+        {"a": ["a", np.nan, "a"], "b": ["b", "c", np.nan]},
+        index=list("xyz"),
+        dtype=object,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # Fill with any value replaces missing values as expected
+    result = data.unstack(fill_value="d")
+    expected = DataFrame(
+        {"a": ["a", "d", "a"], "b": ["b", "c", "d"]}, index=list("xyz"), dtype=object
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_timezone_aware_values():
+    # GH 18338
+    df = DataFrame(
+        {
+            "timestamp": [pd.Timestamp("2017-08-27 01:00:00.709949+0000", tz="UTC")],
+            "a": ["a"],
+            "b": ["b"],
+            "c": ["c"],
+        },
+        columns=["timestamp", "a", "b", "c"],
+    )
+    result = df.set_index(["a", "b"]).unstack()
+    expected = DataFrame(
+        [[pd.Timestamp("2017-08-27 01:00:00.709949+0000", tz="UTC"), "c"]],
+        index=Index(["a"], name="a"),
+        columns=MultiIndex(
+            levels=[["timestamp", "c"], ["b"]],
+            codes=[[0, 1], [0, 0]],
+            names=[None, "b"],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+def test_stack_timezone_aware_values(future_stack):
+    # GH 19420
+    ts = date_range(freq="D", start="20180101", end="20180103", tz="America/New_York")
+    df = DataFrame({"A": ts}, index=["a", "b", "c"])
+    result = df.stack(future_stack=future_stack)
+    expected = Series(
+        ts,
+        index=MultiIndex(levels=[["a", "b", "c"], ["A"]], codes=[[0, 1, 2], [0, 0, 0]]),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+@pytest.mark.parametrize("dropna", [True, False, lib.no_default])
+def test_stack_empty_frame(dropna, future_stack):
+    # GH 36113
+    levels = [pd.RangeIndex(0), pd.RangeIndex(0)]
+    expected = Series(dtype=np.float64, index=MultiIndex(levels=levels, codes=[[], []]))
+    if future_stack and dropna is not lib.no_default:
+        with pytest.raises(ValueError, match="dropna must be unspecified"):
+            DataFrame(dtype=np.float64).stack(dropna=dropna, future_stack=future_stack)
+    else:
+        result = DataFrame(dtype=np.float64).stack(
+            dropna=dropna, future_stack=future_stack
+        )
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+@pytest.mark.parametrize("dropna", [True, False, lib.no_default])
+def test_stack_empty_level(dropna, future_stack, int_frame):
+    # GH 60740
+    if future_stack and dropna is not lib.no_default:
+        with pytest.raises(ValueError, match="dropna must be unspecified"):
+            DataFrame(dtype=np.int64).stack(dropna=dropna, future_stack=future_stack)
+    else:
+        expected = int_frame
+        result = int_frame.copy().stack(
+            level=[], dropna=dropna, future_stack=future_stack
+        )
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame()
+        result = DataFrame().stack(level=[], dropna=dropna, future_stack=future_stack)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+@pytest.mark.parametrize("dropna", [True, False, lib.no_default])
+@pytest.mark.parametrize("fill_value", [None, 0])
+def test_stack_unstack_empty_frame(dropna, fill_value, future_stack):
+    # GH 36113
+    if future_stack and dropna is not lib.no_default:
+        with pytest.raises(ValueError, match="dropna must be unspecified"):
+            DataFrame(dtype=np.int64).stack(
+                dropna=dropna, future_stack=future_stack
+            ).unstack(fill_value=fill_value)
+    else:
+        result = (
+            DataFrame(dtype=np.int64)
+            .stack(dropna=dropna, future_stack=future_stack)
+            .unstack(fill_value=fill_value)
+        )
+        expected = DataFrame(dtype=np.int64)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_single_index_series():
+    # GH 36113
+    msg = r"index must be a MultiIndex to unstack.*"
+    with pytest.raises(ValueError, match=msg):
+        Series(dtype=np.int64).unstack()
+
+
+def test_unstacking_multi_index_df():
+    # see gh-30740
+    df = DataFrame(
+        {
+            "name": ["Alice", "Bob"],
+            "score": [9.5, 8],
+            "employed": [False, True],
+            "kids": [0, 0],
+            "gender": ["female", "male"],
+        }
+    )
+    df = df.set_index(["name", "employed", "kids", "gender"])
+    df = df.unstack(["gender"], fill_value=0)
+    expected = df.unstack("employed", fill_value=0).unstack("kids", fill_value=0)
+    result = df.unstack(["employed", "kids"], fill_value=0)
+    expected = DataFrame(
+        [[9.5, 0.0, 0.0, 0.0], [0.0, 0.0, 0.0, 8.0]],
+        index=Index(["Alice", "Bob"], name="name"),
+        columns=MultiIndex.from_tuples(
+            [
+                ("score", "female", False, 0),
+                ("score", "female", True, 0),
+                ("score", "male", False, 0),
+                ("score", "male", True, 0),
+            ],
+            names=[None, "gender", "employed", "kids"],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+def test_stack_positional_level_duplicate_column_names(future_stack):
+    # https://github.com/pandas-dev/pandas/issues/36353
+    columns = MultiIndex.from_product([("x", "y"), ("y", "z")], names=["a", "a"])
+    df = DataFrame([[1, 1, 1, 1]], columns=columns)
+    result = df.stack(0, future_stack=future_stack)
+
+    new_columns = Index(["y", "z"], name="a")
+    new_index = MultiIndex(
+        levels=[range(1), ["x", "y"]], codes=[[0, 0], [0, 1]], names=[None, "a"]
+    )
+    expected = DataFrame([[1, 1], [1, 1]], index=new_index, columns=new_columns)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_non_slice_like_blocks():
+    # Case where the mgr_locs of a DataFrame's underlying blocks are not slice-like
+
+    mi = MultiIndex.from_product([range(5), ["A", "B", "C"]])
+    df = DataFrame(
+        {
+            0: np.random.default_rng(2).standard_normal(15),
+            1: np.random.default_rng(2).standard_normal(15).astype(np.int64),
+            2: np.random.default_rng(2).standard_normal(15),
+            3: np.random.default_rng(2).standard_normal(15),
+        },
+        index=mi,
+    )
+    assert any(not x.mgr_locs.is_slice_like for x in df._mgr.blocks)
+
+    res = df.unstack()
+
+    expected = pd.concat([df[n].unstack() for n in range(4)], keys=range(4), axis=1)
+    tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+def test_stack_sort_false(future_stack):
+    # GH 15105
+    data = [[1, 2, 3.0, 4.0], [2, 3, 4.0, 5.0], [3, 4, np.nan, np.nan]]
+    df = DataFrame(
+        data,
+        columns=MultiIndex(
+            levels=[["B", "A"], ["x", "y"]], codes=[[0, 0, 1, 1], [0, 1, 0, 1]]
+        ),
+    )
+    kwargs = {} if future_stack else {"sort": False}
+    result = df.stack(level=0, future_stack=future_stack, **kwargs)
+    if future_stack:
+        expected = DataFrame(
+            {
+                "x": [1.0, 3.0, 2.0, 4.0, 3.0, np.nan],
+                "y": [2.0, 4.0, 3.0, 5.0, 4.0, np.nan],
+            },
+            index=MultiIndex.from_arrays(
+                [[0, 0, 1, 1, 2, 2], ["B", "A", "B", "A", "B", "A"]]
+            ),
+        )
+    else:
+        expected = DataFrame(
+            {"x": [1.0, 3.0, 2.0, 4.0, 3.0], "y": [2.0, 4.0, 3.0, 5.0, 4.0]},
+            index=MultiIndex.from_arrays([[0, 0, 1, 1, 2], ["B", "A", "B", "A", "B"]]),
+        )
+    tm.assert_frame_equal(result, expected)
+
+    # Codes sorted in this call
+    df = DataFrame(
+        data,
+        columns=MultiIndex.from_arrays([["B", "B", "A", "A"], ["x", "y", "x", "y"]]),
+    )
+    kwargs = {} if future_stack else {"sort": False}
+    result = df.stack(level=0, future_stack=future_stack, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+def test_stack_sort_false_multi_level(future_stack):
+    # GH 15105
+    idx = MultiIndex.from_tuples([("weight", "kg"), ("height", "m")])
+    df = DataFrame([[1.0, 2.0], [3.0, 4.0]], index=["cat", "dog"], columns=idx)
+    kwargs = {} if future_stack else {"sort": False}
+    result = df.stack([0, 1], future_stack=future_stack, **kwargs)
+    expected_index = MultiIndex.from_tuples(
+        [
+            ("cat", "weight", "kg"),
+            ("cat", "height", "m"),
+            ("dog", "weight", "kg"),
+            ("dog", "height", "m"),
+        ]
+    )
+    expected = Series([1.0, 2.0, 3.0, 4.0], index=expected_index)
+    tm.assert_series_equal(result, expected)
+
+
+class TestStackUnstackMultiLevel:
+    def test_unstack(self, multiindex_year_month_day_dataframe_random_data):
+        # just check that it works for now
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        unstacked = ymd.unstack()
+        unstacked.unstack()
+
+        # test that ints work
+        ymd.astype(int).unstack()
+
+        # test that int32 work
+        ymd.astype(np.int32).unstack()
+
+    @pytest.mark.parametrize(
+        "result_rows,result_columns,index_product,expected_row",
+        [
+            (
+                [[1, 1, None, None, 30.0, None], [2, 2, None, None, 30.0, None]],
+                ["ix1", "ix2", "col1", "col2", "col3", "col4"],
+                2,
+                [None, None, 30.0, None],
+            ),
+            (
+                [[1, 1, None, None, 30.0], [2, 2, None, None, 30.0]],
+                ["ix1", "ix2", "col1", "col2", "col3"],
+                2,
+                [None, None, 30.0],
+            ),
+            (
+                [[1, 1, None, None, 30.0], [2, None, None, None, 30.0]],
+                ["ix1", "ix2", "col1", "col2", "col3"],
+                None,
+                [None, None, 30.0],
+            ),
+        ],
+    )
+    def test_unstack_partial(
+        self, result_rows, result_columns, index_product, expected_row
+    ):
+        # check for regressions on this issue:
+        # https://github.com/pandas-dev/pandas/issues/19351
+        # make sure DataFrame.unstack() works when its run on a subset of the DataFrame
+        # and the Index levels contain values that are not present in the subset
+        result = DataFrame(result_rows, columns=result_columns).set_index(
+            ["ix1", "ix2"]
+        )
+        result = result.iloc[1:2].unstack("ix2")
+        expected = DataFrame(
+            [expected_row],
+            columns=MultiIndex.from_product(
+                [result_columns[2:], [index_product]], names=[None, "ix2"]
+            ),
+            index=Index([2], name="ix1"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_multiple_no_empty_columns(self):
+        index = MultiIndex.from_tuples(
+            [(0, "foo", 0), (0, "bar", 0), (1, "baz", 1), (1, "qux", 1)]
+        )
+
+        s = Series(np.random.default_rng(2).standard_normal(4), index=index)
+
+        unstacked = s.unstack([1, 2])
+        expected = unstacked.dropna(axis=1, how="all")
+        tm.assert_frame_equal(unstacked, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack(self, multiindex_year_month_day_dataframe_random_data, future_stack):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        # regular roundtrip
+        unstacked = ymd.unstack()
+        restacked = unstacked.stack(future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        tm.assert_frame_equal(restacked, ymd)
+
+        unlexsorted = ymd.sort_index(level=2)
+
+        unstacked = unlexsorted.unstack(2)
+        restacked = unstacked.stack(future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        tm.assert_frame_equal(restacked.sort_index(level=0), ymd)
+
+        unlexsorted = unlexsorted[::-1]
+        unstacked = unlexsorted.unstack(1)
+        restacked = unstacked.stack(future_stack=future_stack).swaplevel(1, 2)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        tm.assert_frame_equal(restacked.sort_index(level=0), ymd)
+
+        unlexsorted = unlexsorted.swaplevel(0, 1)
+        unstacked = unlexsorted.unstack(0).swaplevel(0, 1, axis=1)
+        restacked = unstacked.stack(0, future_stack=future_stack).swaplevel(1, 2)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        tm.assert_frame_equal(restacked.sort_index(level=0), ymd)
+
+        # columns unsorted
+        unstacked = ymd.unstack()
+        restacked = unstacked.stack(future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        tm.assert_frame_equal(restacked, ymd)
+
+        # more than 2 levels in the columns
+        unstacked = ymd.unstack(1).unstack(1)
+
+        result = unstacked.stack(1, future_stack=future_stack)
+        expected = ymd.unstack()
+        tm.assert_frame_equal(result, expected)
+
+        result = unstacked.stack(2, future_stack=future_stack)
+        expected = ymd.unstack(1)
+        tm.assert_frame_equal(result, expected)
+
+        result = unstacked.stack(0, future_stack=future_stack)
+        expected = ymd.stack(future_stack=future_stack).unstack(1).unstack(1)
+        tm.assert_frame_equal(result, expected)
+
+        # not all levels present in each echelon
+        unstacked = ymd.unstack(2).loc[:, ::3]
+        stacked = unstacked.stack(future_stack=future_stack).stack(
+            future_stack=future_stack
+        )
+        ymd_stacked = ymd.stack(future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            stacked = stacked.dropna(how="all")
+            ymd_stacked = ymd_stacked.dropna(how="all")
+        tm.assert_series_equal(stacked, ymd_stacked.reindex(stacked.index))
+
+        # stack with negative number
+        result = ymd.unstack(0).stack(-2, future_stack=future_stack)
+        expected = ymd.unstack(0).stack(0, future_stack=future_stack)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "idx, exp_idx",
+        [
+            [
+                list("abab"),
+                MultiIndex(
+                    levels=[["a", "b"], ["1st", "2nd"]],
+                    codes=[np.tile(np.arange(2).repeat(3), 2), np.tile([0, 1, 0], 4)],
+                ),
+            ],
+            [
+                MultiIndex.from_tuples((("a", 2), ("b", 1), ("a", 1), ("b", 2))),
+                MultiIndex(
+                    levels=[["a", "b"], [1, 2], ["1st", "2nd"]],
+                    codes=[
+                        np.tile(np.arange(2).repeat(3), 2),
+                        np.repeat([1, 0, 1], [3, 6, 3]),
+                        np.tile([0, 1, 0], 4),
+                    ],
+                ),
+            ],
+        ],
+    )
+    def test_stack_duplicate_index(self, idx, exp_idx, future_stack):
+        # GH10417
+        df = DataFrame(
+            np.arange(12).reshape(4, 3),
+            index=idx,
+            columns=["1st", "2nd", "1st"],
+        )
+        if future_stack:
+            msg = "Columns with duplicate values are not supported in stack"
+            with pytest.raises(ValueError, match=msg):
+                df.stack(future_stack=future_stack)
+        else:
+            result = df.stack(future_stack=future_stack)
+            expected = Series(np.arange(12), index=exp_idx)
+            tm.assert_series_equal(result, expected)
+            assert result.index.is_unique is False
+            li, ri = result.index, expected.index
+            tm.assert_index_equal(li, ri)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_odd_failure(self, future_stack):
+        mi = MultiIndex.from_arrays(
+            [
+                ["Fri"] * 4 + ["Sat"] * 2 + ["Sun"] * 2 + ["Thu"] * 3,
+                ["Dinner"] * 2 + ["Lunch"] * 2 + ["Dinner"] * 5 + ["Lunch"] * 2,
+                ["No", "Yes"] * 4 + ["No", "No", "Yes"],
+            ],
+            names=["day", "time", "smoker"],
+        )
+        df = DataFrame(
+            {
+                "sum": np.arange(11, dtype="float64"),
+                "len": np.arange(11, dtype="float64"),
+            },
+            index=mi,
+        )
+        # it works, #2100
+        result = df.unstack(2)
+
+        recons = result.stack(future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            recons = recons.dropna(how="all")
+        tm.assert_frame_equal(recons, df)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_mixed_dtype(self, multiindex_dataframe_random_data, future_stack):
+        frame = multiindex_dataframe_random_data
+
+        df = frame.T
+        df["foo", "four"] = "foo"
+        df = df.sort_index(level=1, axis=1)
+
+        stacked = df.stack(future_stack=future_stack)
+        result = df["foo"].stack(future_stack=future_stack).sort_index()
+        tm.assert_series_equal(stacked["foo"], result, check_names=False)
+        assert result.name is None
+        assert stacked["bar"].dtype == np.float64
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_bug(self, future_stack):
+        df = DataFrame(
+            {
+                "state": ["naive", "naive", "naive", "active", "active", "active"],
+                "exp": ["a", "b", "b", "b", "a", "a"],
+                "barcode": [1, 2, 3, 4, 1, 3],
+                "v": ["hi", "hi", "bye", "bye", "bye", "peace"],
+                "extra": np.arange(6.0),
+            }
+        )
+
+        result = df.groupby(["state", "exp", "barcode", "v"]).apply(len)
+        unstacked = result.unstack()
+        restacked = unstacked.stack(future_stack=future_stack)
+        tm.assert_series_equal(restacked, result.reindex(restacked.index).astype(float))
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_unstack_preserve_names(
+        self, multiindex_dataframe_random_data, future_stack
+    ):
+        frame = multiindex_dataframe_random_data
+
+        unstacked = frame.unstack()
+        assert unstacked.index.name == "first"
+        assert unstacked.columns.names == ["exp", "second"]
+
+        restacked = unstacked.stack(future_stack=future_stack)
+        assert restacked.index.names == frame.index.names
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize("method", ["stack", "unstack"])
+    def test_stack_unstack_wrong_level_name(
+        self, method, multiindex_dataframe_random_data, future_stack
+    ):
+        # GH 18303 - wrong level name should raise
+        frame = multiindex_dataframe_random_data
+
+        # A DataFrame with flat axes:
+        df = frame.loc["foo"]
+
+        kwargs = {"future_stack": future_stack} if method == "stack" else {}
+        with pytest.raises(KeyError, match="does not match index name"):
+            getattr(df, method)("mistake", **kwargs)
+
+        if method == "unstack":
+            # Same on a Series:
+            s = df.iloc[:, 0]
+            with pytest.raises(KeyError, match="does not match index name"):
+                getattr(s, method)("mistake", **kwargs)
+
+    def test_unstack_level_name(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        result = frame.unstack("second")
+        expected = frame.unstack(level=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_level_name(self, multiindex_dataframe_random_data, future_stack):
+        frame = multiindex_dataframe_random_data
+
+        unstacked = frame.unstack("second")
+        result = unstacked.stack("exp", future_stack=future_stack)
+        expected = frame.unstack().stack(0, future_stack=future_stack)
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.stack("exp", future_stack=future_stack)
+        expected = frame.stack(future_stack=future_stack)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_unstack_multiple(
+        self, multiindex_year_month_day_dataframe_random_data, future_stack
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        unstacked = ymd.unstack(["year", "month"])
+        expected = ymd.unstack("year").unstack("month")
+        tm.assert_frame_equal(unstacked, expected)
+        assert unstacked.columns.names == expected.columns.names
+
+        # series
+        s = ymd["A"]
+        s_unstacked = s.unstack(["year", "month"])
+        tm.assert_frame_equal(s_unstacked, expected["A"])
+
+        restacked = unstacked.stack(["year", "month"], future_stack=future_stack)
+        if future_stack:
+            # NA values in unstacked persist to restacked in version 3
+            restacked = restacked.dropna(how="all")
+        restacked = restacked.swaplevel(0, 1).swaplevel(1, 2)
+        restacked = restacked.sort_index(level=0)
+
+        tm.assert_frame_equal(restacked, ymd)
+        assert restacked.index.names == ymd.index.names
+
+        # GH #451
+        unstacked = ymd.unstack([1, 2])
+        expected = ymd.unstack(1).unstack(1).dropna(axis=1, how="all")
+        tm.assert_frame_equal(unstacked, expected)
+
+        unstacked = ymd.unstack([2, 1])
+        expected = ymd.unstack(2).unstack(1).dropna(axis=1, how="all")
+        tm.assert_frame_equal(unstacked, expected.loc[:, unstacked.columns])
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_names_and_numbers(
+        self, multiindex_year_month_day_dataframe_random_data, future_stack
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        unstacked = ymd.unstack(["year", "month"])
+
+        # Can't use mixture of names and numbers to stack
+        with pytest.raises(ValueError, match="level should contain"):
+            unstacked.stack([0, "month"], future_stack=future_stack)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_multiple_out_of_bounds(
+        self, multiindex_year_month_day_dataframe_random_data, future_stack
+    ):
+        # nlevels == 3
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        unstacked = ymd.unstack(["year", "month"])
+
+        with pytest.raises(IndexError, match="Too many levels"):
+            unstacked.stack([2, 3], future_stack=future_stack)
+        with pytest.raises(IndexError, match="not a valid level number"):
+            unstacked.stack([-4, -3], future_stack=future_stack)
+
+    def test_unstack_period_series(self):
+        # GH4342
+        idx1 = pd.PeriodIndex(
+            ["2013-01", "2013-01", "2013-02", "2013-02", "2013-03", "2013-03"],
+            freq="M",
+            name="period",
+        )
+        idx2 = Index(["A", "B"] * 3, name="str")
+        value = [1, 2, 3, 4, 5, 6]
+
+        idx = MultiIndex.from_arrays([idx1, idx2])
+        s = Series(value, index=idx)
+
+        result1 = s.unstack()
+        result2 = s.unstack(level=1)
+        result3 = s.unstack(level=0)
+
+        e_idx = pd.PeriodIndex(
+            ["2013-01", "2013-02", "2013-03"], freq="M", name="period"
+        )
+        expected = DataFrame(
+            {"A": [1, 3, 5], "B": [2, 4, 6]}, index=e_idx, columns=["A", "B"]
+        )
+        expected.columns.name = "str"
+
+        tm.assert_frame_equal(result1, expected)
+        tm.assert_frame_equal(result2, expected)
+        tm.assert_frame_equal(result3, expected.T)
+
+        idx1 = pd.PeriodIndex(
+            ["2013-01", "2013-01", "2013-02", "2013-02", "2013-03", "2013-03"],
+            freq="M",
+            name="period1",
+        )
+
+        idx2 = pd.PeriodIndex(
+            ["2013-12", "2013-11", "2013-10", "2013-09", "2013-08", "2013-07"],
+            freq="M",
+            name="period2",
+        )
+        idx = MultiIndex.from_arrays([idx1, idx2])
+        s = Series(value, index=idx)
+
+        result1 = s.unstack()
+        result2 = s.unstack(level=1)
+        result3 = s.unstack(level=0)
+
+        e_idx = pd.PeriodIndex(
+            ["2013-01", "2013-02", "2013-03"], freq="M", name="period1"
+        )
+        e_cols = pd.PeriodIndex(
+            ["2013-07", "2013-08", "2013-09", "2013-10", "2013-11", "2013-12"],
+            freq="M",
+            name="period2",
+        )
+        expected = DataFrame(
+            [
+                [np.nan, np.nan, np.nan, np.nan, 2, 1],
+                [np.nan, np.nan, 4, 3, np.nan, np.nan],
+                [6, 5, np.nan, np.nan, np.nan, np.nan],
+            ],
+            index=e_idx,
+            columns=e_cols,
+        )
+
+        tm.assert_frame_equal(result1, expected)
+        tm.assert_frame_equal(result2, expected)
+        tm.assert_frame_equal(result3, expected.T)
+
+    def test_unstack_period_frame(self):
+        # GH4342
+        idx1 = pd.PeriodIndex(
+            ["2014-01", "2014-02", "2014-02", "2014-02", "2014-01", "2014-01"],
+            freq="M",
+            name="period1",
+        )
+        idx2 = pd.PeriodIndex(
+            ["2013-12", "2013-12", "2014-02", "2013-10", "2013-10", "2014-02"],
+            freq="M",
+            name="period2",
+        )
+        value = {"A": [1, 2, 3, 4, 5, 6], "B": [6, 5, 4, 3, 2, 1]}
+        idx = MultiIndex.from_arrays([idx1, idx2])
+        df = DataFrame(value, index=idx)
+
+        result1 = df.unstack()
+        result2 = df.unstack(level=1)
+        result3 = df.unstack(level=0)
+
+        e_1 = pd.PeriodIndex(["2014-01", "2014-02"], freq="M", name="period1")
+        e_2 = pd.PeriodIndex(
+            ["2013-10", "2013-12", "2014-02", "2013-10", "2013-12", "2014-02"],
+            freq="M",
+            name="period2",
+        )
+        e_cols = MultiIndex.from_arrays(["A A A B B B".split(), e_2])
+        expected = DataFrame(
+            [[5, 1, 6, 2, 6, 1], [4, 2, 3, 3, 5, 4]], index=e_1, columns=e_cols
+        )
+
+        tm.assert_frame_equal(result1, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        e_1 = pd.PeriodIndex(
+            ["2014-01", "2014-02", "2014-01", "2014-02"], freq="M", name="period1"
+        )
+        e_2 = pd.PeriodIndex(
+            ["2013-10", "2013-12", "2014-02"], freq="M", name="period2"
+        )
+        e_cols = MultiIndex.from_arrays(["A A B B".split(), e_1])
+        expected = DataFrame(
+            [[5, 4, 2, 3], [1, 2, 6, 5], [6, 3, 1, 4]], index=e_2, columns=e_cols
+        )
+
+        tm.assert_frame_equal(result3, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_multiple_bug(self, future_stack, using_infer_string):
+        # bug when some uniques are not present in the data GH#3170
+        id_col = ([1] * 3) + ([2] * 3)
+        name = (["a"] * 3) + (["b"] * 3)
+        date = pd.to_datetime(["2013-01-03", "2013-01-04", "2013-01-05"] * 2)
+        var1 = np.random.default_rng(2).integers(0, 100, 6)
+        df = DataFrame({"ID": id_col, "NAME": name, "DATE": date, "VAR1": var1})
+
+        multi = df.set_index(["DATE", "ID"])
+        multi.columns.name = "Params"
+        unst = multi.unstack("ID")
+        msg = re.escape("agg function failed [how->mean,dtype->")
+        if using_infer_string:
+            msg = "dtype 'str' does not support operation 'mean'"
+        with pytest.raises(TypeError, match=msg):
+            unst.resample("W-THU").mean()
+        down = unst.resample("W-THU").mean(numeric_only=True)
+        rs = down.stack("ID", future_stack=future_stack)
+        xp = (
+            unst.loc[:, ["VAR1"]]
+            .resample("W-THU")
+            .mean()
+            .stack("ID", future_stack=future_stack)
+        )
+        xp.columns.name = "Params"
+        tm.assert_frame_equal(rs, xp)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_dropna(self, future_stack):
+        # GH#3997
+        df = DataFrame({"A": ["a1", "a2"], "B": ["b1", "b2"], "C": [1, 1]})
+        df = df.set_index(["A", "B"])
+
+        dropna = False if not future_stack else lib.no_default
+        stacked = df.unstack().stack(dropna=dropna, future_stack=future_stack)
+        assert len(stacked) > len(stacked.dropna())
+
+        if future_stack:
+            with pytest.raises(ValueError, match="dropna must be unspecified"):
+                df.unstack().stack(dropna=True, future_stack=future_stack)
+        else:
+            stacked = df.unstack().stack(dropna=True, future_stack=future_stack)
+            tm.assert_frame_equal(stacked, stacked.dropna())
+
+    def test_unstack_multiple_hierarchical(self, future_stack):
+        df = DataFrame(
+            index=[
+                [0, 0, 0, 0, 1, 1, 1, 1],
+                [0, 0, 1, 1, 0, 0, 1, 1],
+                [0, 1, 0, 1, 0, 1, 0, 1],
+            ],
+            columns=[[0, 0, 1, 1], [0, 1, 0, 1]],
+        )
+
+        df.index.names = ["a", "b", "c"]
+        df.columns.names = ["d", "e"]
+
+        # it works!
+        df.unstack(["b", "c"])
+
+    def test_unstack_sparse_keyspace(self):
+        # memory problems with naive impl GH#2278
+        # Generate Long File & Test Pivot
+        NUM_ROWS = 1000
+
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).integers(100, size=NUM_ROWS),
+                "B": np.random.default_rng(3).integers(300, size=NUM_ROWS),
+                "C": np.random.default_rng(4).integers(-7, 7, size=NUM_ROWS),
+                "D": np.random.default_rng(5).integers(-19, 19, size=NUM_ROWS),
+                "E": np.random.default_rng(6).integers(3000, size=NUM_ROWS),
+                "F": np.random.default_rng(7).standard_normal(NUM_ROWS),
+            }
+        )
+
+        idf = df.set_index(["A", "B", "C", "D", "E"])
+
+        # it works! is sufficient
+        idf.unstack("E")
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_unobserved_keys(self, future_stack):
+        # related to GH#2278 refactoring
+        levels = [[0, 1], [0, 1, 2, 3]]
+        codes = [[0, 0, 1, 1], [0, 2, 0, 2]]
+
+        index = MultiIndex(levels, codes)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 2)), index=index)
+
+        result = df.unstack()
+        assert len(result.columns) == 4
+
+        recons = result.stack(future_stack=future_stack)
+        tm.assert_frame_equal(recons, df)
+
+    @pytest.mark.slow
+    def test_unstack_number_of_levels_larger_than_int32(
+        self, performance_warning, monkeypatch
+    ):
+        # GH#20601
+        # GH 26314: Change ValueError to PerformanceWarning
+
+        class MockUnstacker(reshape_lib._Unstacker):
+            def __init__(self, *args, **kwargs) -> None:
+                # __init__ will raise the warning
+                super().__init__(*args, **kwargs)
+                raise Exception("Don't compute final result.")
+
+        with monkeypatch.context() as m:
+            m.setattr(reshape_lib, "_Unstacker", MockUnstacker)
+            df = DataFrame(
+                np.zeros((2**16, 2)),
+                index=[np.arange(2**16), np.arange(2**16)],
+            )
+            msg = "The following operation may generate"
+            with tm.assert_produces_warning(performance_warning, match=msg):
+                with pytest.raises(Exception, match="Don't compute final result."):
+                    df.unstack()
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    @pytest.mark.parametrize(
+        "levels",
+        itertools.chain.from_iterable(
+            itertools.product(itertools.permutations([0, 1, 2], width), repeat=2)
+            for width in [2, 3]
+        ),
+    )
+    @pytest.mark.parametrize("stack_lev", range(2))
+    def test_stack_order_with_unsorted_levels(
+        self, levels, stack_lev, sort, future_stack
+    ):
+        # GH#16323
+        # deep check for 1-row case
+        columns = MultiIndex(levels=levels, codes=[[0, 0, 1, 1], [0, 1, 0, 1]])
+        df = DataFrame(columns=columns, data=[range(4)])
+        kwargs = {} if future_stack else {"sort": sort}
+        df_stacked = df.stack(stack_lev, future_stack=future_stack, **kwargs)
+        for row in df.index:
+            for col in df.columns:
+                expected = df.loc[row, col]
+                result_row = row, col[stack_lev]
+                result_col = col[1 - stack_lev]
+                result = df_stacked.loc[result_row, result_col]
+                assert result == expected
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_order_with_unsorted_levels_multi_row(self, future_stack):
+        # GH#16323
+
+        # check multi-row case
+        mi = MultiIndex(
+            levels=[["A", "C", "B"], ["B", "A", "C"]],
+            codes=[np.repeat(range(3), 3), np.tile(range(3), 3)],
+        )
+        df = DataFrame(
+            columns=mi, index=range(5), data=np.arange(5 * len(mi)).reshape(5, -1)
+        )
+        assert all(
+            df.loc[row, col]
+            == df.stack(0, future_stack=future_stack).loc[(row, col[0]), col[1]]
+            for row in df.index
+            for col in df.columns
+        )
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_order_with_unsorted_levels_multi_row_2(self, future_stack):
+        # GH#53636
+        levels = ((0, 1), (1, 0))
+        stack_lev = 1
+        columns = MultiIndex(levels=levels, codes=[[0, 0, 1, 1], [0, 1, 0, 1]])
+        df = DataFrame(columns=columns, data=[range(4)], index=[1, 0, 2, 3])
+        kwargs = {} if future_stack else {"sort": True}
+        result = df.stack(stack_lev, future_stack=future_stack, **kwargs)
+        expected_index = MultiIndex(
+            levels=[[0, 1, 2, 3], [0, 1]],
+            codes=[[1, 1, 0, 0, 2, 2, 3, 3], [1, 0, 1, 0, 1, 0, 1, 0]],
+        )
+        expected = DataFrame(
+            {
+                0: [0, 1, 0, 1, 0, 1, 0, 1],
+                1: [2, 3, 2, 3, 2, 3, 2, 3],
+            },
+            index=expected_index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_unstack_unordered_multiindex(self, future_stack):
+        # GH# 18265
+        values = np.arange(5)
+        data = np.vstack(
+            [
+                [f"b{x}" for x in values],  # b0, b1, ..
+                [f"a{x}" for x in values],  # a0, a1, ..
+            ]
+        )
+        df = DataFrame(data.T, columns=["b", "a"])
+        df.columns.name = "first"
+        second_level_dict = {"x": df}
+        multi_level_df = pd.concat(second_level_dict, axis=1)
+        multi_level_df.columns.names = ["second", "first"]
+        df = multi_level_df.reindex(sorted(multi_level_df.columns), axis=1)
+        result = df.stack(["first", "second"], future_stack=future_stack).unstack(
+            ["first", "second"]
+        )
+        expected = DataFrame(
+            [["a0", "b0"], ["a1", "b1"], ["a2", "b2"], ["a3", "b3"], ["a4", "b4"]],
+            index=range(5),
+            columns=MultiIndex.from_tuples(
+                [("a", "x"), ("b", "x")], names=["first", "second"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_preserve_types(
+        self, multiindex_year_month_day_dataframe_random_data, using_infer_string
+    ):
+        # GH#403
+        ymd = multiindex_year_month_day_dataframe_random_data
+        ymd["E"] = "foo"
+        ymd["F"] = 2
+
+        unstacked = ymd.unstack("month")
+        assert unstacked["A", 1].dtype == np.float64
+        assert (
+            unstacked["E", 1].dtype == np.object_
+            if not using_infer_string
+            else "string"
+        )
+        assert unstacked["F", 1].dtype == np.float64
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_unstack_group_index_overflow(self, future_stack):
+        codes = np.tile(np.arange(500), 2)
+        level = np.arange(500)
+
+        index = MultiIndex(
+            levels=[level] * 8 + [[0, 1]],
+            codes=[codes] * 8 + [np.arange(2).repeat(500)],
+        )
+
+        s = Series(np.arange(1000), index=index)
+        result = s.unstack()
+        assert result.shape == (500, 2)
+
+        # test roundtrip
+        stacked = result.stack(future_stack=future_stack)
+        tm.assert_series_equal(s, stacked.reindex(s.index))
+
+        # put it at beginning
+        index = MultiIndex(
+            levels=[[0, 1]] + [level] * 8,
+            codes=[np.arange(2).repeat(500)] + [codes] * 8,
+        )
+
+        s = Series(np.arange(1000), index=index)
+        result = s.unstack(0)
+        assert result.shape == (500, 2)
+
+        # put it in middle
+        index = MultiIndex(
+            levels=[level] * 4 + [[0, 1]] + [level] * 4,
+            codes=([codes] * 4 + [np.arange(2).repeat(500)] + [codes] * 4),
+        )
+
+        s = Series(np.arange(1000), index=index)
+        result = s.unstack(4)
+        assert result.shape == (500, 2)
+
+    def test_unstack_with_missing_int_cast_to_float(self):
+        # https://github.com/pandas-dev/pandas/issues/37115
+        df = DataFrame(
+            {
+                "a": ["A", "A", "B"],
+                "b": ["ca", "cb", "cb"],
+                "v": [10] * 3,
+            }
+        ).set_index(["a", "b"])
+
+        # add another int column to get 2 blocks
+        df["is_"] = 1
+        assert len(df._mgr.blocks) == 2
+
+        result = df.unstack("b")
+        result[("is_", "ca")] = result[("is_", "ca")].fillna(0)
+
+        expected = DataFrame(
+            [[10.0, 10.0, 1.0, 1.0], [np.nan, 10.0, 0.0, 1.0]],
+            index=Index(["A", "B"], name="a"),
+            columns=MultiIndex.from_tuples(
+                [("v", "ca"), ("v", "cb"), ("is_", "ca"), ("is_", "cb")],
+                names=[None, "b"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_with_level_has_nan(self):
+        # GH 37510
+        df1 = DataFrame(
+            {
+                "L1": [1, 2, 3, 4],
+                "L2": [3, 4, 1, 2],
+                "L3": [1, 1, 1, 1],
+                "x": [1, 2, 3, 4],
+            }
+        )
+        df1 = df1.set_index(["L1", "L2", "L3"])
+        new_levels = ["n1", "n2", "n3", None]
+        df1.index = df1.index.set_levels(levels=new_levels, level="L1")
+        df1.index = df1.index.set_levels(levels=new_levels, level="L2")
+
+        result = df1.unstack("L3")[("x", 1)].sort_index().index
+        expected = MultiIndex(
+            levels=[["n1", "n2", "n3", None], ["n1", "n2", "n3", None]],
+            codes=[[0, 1, 2, 3], [2, 3, 0, 1]],
+            names=["L1", "L2"],
+        )
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_nan_in_multiindex_columns(self, future_stack):
+        # GH#39481
+        df = DataFrame(
+            np.zeros([1, 5]),
+            columns=MultiIndex.from_tuples(
+                [
+                    (0, None, None),
+                    (0, 2, 0),
+                    (0, 2, 1),
+                    (0, 3, 0),
+                    (0, 3, 1),
+                ],
+            ),
+        )
+        result = df.stack(2, future_stack=future_stack)
+        if future_stack:
+            index = MultiIndex(levels=[[0], [0.0, 1.0]], codes=[[0, 0, 0], [-1, 0, 1]])
+            columns = MultiIndex(levels=[[0], [2, 3]], codes=[[0, 0, 0], [-1, 0, 1]])
+        else:
+            index = Index([(0, None), (0, 0), (0, 1)])
+            columns = Index([(0, None), (0, 2), (0, 3)])
+        expected = DataFrame(
+            [[0.0, np.nan, np.nan], [np.nan, 0.0, 0.0], [np.nan, 0.0, 0.0]],
+            index=index,
+            columns=columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_multi_level_stack_categorical(self, future_stack):
+        # GH 15239
+        midx = MultiIndex.from_arrays(
+            [
+                ["A"] * 2 + ["B"] * 2,
+                pd.Categorical(list("abab")),
+                pd.Categorical(list("ccdd")),
+            ]
+        )
+        df = DataFrame(np.arange(8).reshape(2, 4), columns=midx)
+        result = df.stack([1, 2], future_stack=future_stack)
+        if future_stack:
+            expected = DataFrame(
+                [
+                    [0, np.nan],
+                    [1, np.nan],
+                    [np.nan, 2],
+                    [np.nan, 3],
+                    [4, np.nan],
+                    [5, np.nan],
+                    [np.nan, 6],
+                    [np.nan, 7],
+                ],
+                columns=["A", "B"],
+                index=MultiIndex.from_arrays(
+                    [
+                        [0] * 4 + [1] * 4,
+                        pd.Categorical(list("abababab")),
+                        pd.Categorical(list("ccddccdd")),
+                    ]
+                ),
+            )
+        else:
+            expected = DataFrame(
+                [
+                    [0, np.nan],
+                    [np.nan, 2],
+                    [1, np.nan],
+                    [np.nan, 3],
+                    [4, np.nan],
+                    [np.nan, 6],
+                    [5, np.nan],
+                    [np.nan, 7],
+                ],
+                columns=["A", "B"],
+                index=MultiIndex.from_arrays(
+                    [
+                        [0] * 4 + [1] * 4,
+                        pd.Categorical(list("aabbaabb")),
+                        pd.Categorical(list("cdcdcdcd")),
+                    ]
+                ),
+            )
+        tm.assert_frame_equal(result, expected, check_index_type=False)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_nan_level(self, future_stack):
+        # GH 9406
+        df_nan = DataFrame(
+            np.arange(4).reshape(2, 2),
+            columns=MultiIndex.from_tuples(
+                [("A", np.nan), ("B", "b")], names=["Upper", "Lower"]
+            ),
+            index=Index([0, 1], name="Num"),
+            dtype=np.float64,
+        )
+        result = df_nan.stack(future_stack=future_stack)
+        if future_stack:
+            index = MultiIndex(
+                levels=[[0, 1], [np.nan, "b"]],
+                codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+                names=["Num", "Lower"],
+            )
+        else:
+            index = MultiIndex.from_tuples(
+                [(0, np.nan), (0, "b"), (1, np.nan), (1, "b")], names=["Num", "Lower"]
+            )
+        expected = DataFrame(
+            [[0.0, np.nan], [np.nan, 1], [2.0, np.nan], [np.nan, 3.0]],
+            columns=Index(["A", "B"], name="Upper"),
+            index=index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_categorical_columns(self):
+        # GH 14018
+        idx = MultiIndex.from_product([["A"], [0, 1]])
+        df = DataFrame({"cat": pd.Categorical(["a", "b"])}, index=idx)
+        result = df.unstack()
+        expected = DataFrame(
+            {
+                0: pd.Categorical(["a"], categories=["a", "b"]),
+                1: pd.Categorical(["b"], categories=["a", "b"]),
+            },
+            index=["A"],
+        )
+        expected.columns = MultiIndex.from_tuples([("cat", 0), ("cat", 1)])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_unsorted(self, future_stack):
+        # GH 16925
+        PAE = ["ITA", "FRA"]
+        VAR = ["A1", "A2"]
+        TYP = ["CRT", "DBT", "NET"]
+        MI = MultiIndex.from_product([PAE, VAR, TYP], names=["PAE", "VAR", "TYP"])
+
+        V = list(range(len(MI)))
+        DF = DataFrame(data=V, index=MI, columns=["VALUE"])
+
+        DF = DF.unstack(["VAR", "TYP"])
+        DF.columns = DF.columns.droplevel(0)
+        DF.loc[:, ("A0", "NET")] = 9999
+
+        result = DF.stack(["VAR", "TYP"], future_stack=future_stack).sort_index()
+        expected = (
+            DF.sort_index(axis=1)
+            .stack(["VAR", "TYP"], future_stack=future_stack)
+            .sort_index()
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:The previous implementation of stack is deprecated"
+    )
+    def test_stack_nullable_dtype(self, future_stack):
+        # GH#43561
+        columns = MultiIndex.from_product(
+            [["54511", "54515"], ["r", "t_mean"]], names=["station", "element"]
+        )
+        index = Index([1, 2, 3], name="time")
+
+        arr = np.array([[50, 226, 10, 215], [10, 215, 9, 220], [305, 232, 111, 220]])
+        df = DataFrame(arr, columns=columns, index=index, dtype=pd.Int64Dtype())
+
+        result = df.stack("station", future_stack=future_stack)
+
+        expected = (
+            df.astype(np.int64)
+            .stack("station", future_stack=future_stack)
+            .astype(pd.Int64Dtype())
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # non-homogeneous case
+        df[df.columns[0]] = df[df.columns[0]].astype(pd.Float64Dtype())
+        result = df.stack("station", future_stack=future_stack)
+
+        expected = DataFrame(
+            {
+                "r": pd.array(
+                    [50.0, 10.0, 10.0, 9.0, 305.0, 111.0], dtype=pd.Float64Dtype()
+                ),
+                "t_mean": pd.array(
+                    [226, 215, 215, 220, 232, 220], dtype=pd.Int64Dtype()
+                ),
+            },
+            index=MultiIndex.from_product([index, columns.levels[0]]),
+        )
+        expected.columns.name = "element"
+        tm.assert_frame_equal(result, expected)
+
+    def test_unstack_mixed_level_names(self):
+        # GH#48763
+        arrays = [["a", "a"], [1, 2], ["red", "blue"]]
+        idx = MultiIndex.from_arrays(arrays, names=("x", 0, "y"))
+        df = DataFrame({"m": [1, 2]}, index=idx)
+        result = df.unstack("x")
+        expected = DataFrame(
+            [[1], [2]],
+            columns=MultiIndex.from_tuples([("m", "a")], names=[None, "x"]),
+            index=MultiIndex.from_tuples([(1, "red"), (2, "blue")], names=[0, "y"]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:The previous implementation of stack is deprecated")
+def test_stack_tuple_columns(future_stack):
+    # GH#54948 - test stack when the input has a non-MultiIndex with tuples
+    df = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=[("a", 1), ("a", 2), ("b", 1)]
+    )
+    result = df.stack(future_stack=future_stack)
+    expected = Series(
+        [1, 2, 3, 4, 5, 6, 7, 8, 9],
+        index=MultiIndex(
+            levels=[range(3), [("a", 1), ("a", 2), ("b", 1)]],
+            codes=[[0, 0, 0, 1, 1, 1, 2, 2, 2], [0, 1, 2, 0, 1, 2, 0, 1, 2]],
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype, na_value",
+    [
+        ("float64", np.nan),
+        ("Float64", np.nan),
+        ("Float64", pd.NA),
+        ("Int64", pd.NA),
+    ],
+)
+@pytest.mark.parametrize("test_multiindex", [True, False])
+def test_stack_preserves_na(dtype, na_value, test_multiindex):
+    # GH#56573
+    if test_multiindex:
+        index = MultiIndex.from_arrays(2 * [Index([na_value], dtype=dtype)])
+    else:
+        index = Index([na_value], dtype=dtype)
+    df = DataFrame({"a": [1]}, index=index)
+    result = df.stack()
+
+    if test_multiindex:
+        expected_index = MultiIndex.from_arrays(
+            [
+                Index([na_value], dtype=dtype),
+                Index([na_value], dtype=dtype),
+                Index(["a"]),
+            ]
+        )
+    else:
+        expected_index = MultiIndex.from_arrays(
+            [
+                Index([na_value], dtype=dtype),
+                Index(["a"]),
+            ]
+        )
+    expected = Series(1, index=expected_index)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/frame/test_subclass.py b/pandas/pandas/tests/frame/test_subclass.py
new file mode 100644
index 0000000000000..cbd563a03b908
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_subclass.py
@@ -0,0 +1,817 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+)
+
+
+class TestDataFrameSubclassing:
+    def test_no_warning_on_mgr(self):
+        # GH#57032
+        df = tm.SubclassedDataFrame(
+            {"X": [1, 2, 3], "Y": [1, 2, 3]}, index=["a", "b", "c"]
+        )
+        with tm.assert_produces_warning(None):
+            # df.isna() goes through _constructor_from_mgr, which we want to
+            #  *not* pass a Manager do __init__
+            df.isna()
+            df["X"].isna()
+
+    def test_frame_subclassing_and_slicing(self):
+        # Subclass frame and ensure it returns the right class on slicing it
+        # In reference to PR 9632
+
+        class CustomSeries(Series):
+            @property
+            def _constructor(self):
+                return CustomSeries
+
+            def custom_series_function(self):
+                return "OK"
+
+        class CustomDataFrame(DataFrame):
+            """
+            Subclasses pandas DF, fills DF with simulation results, adds some
+            custom plotting functions.
+            """
+
+            def __init__(self, *args, **kw) -> None:
+                super().__init__(*args, **kw)
+
+            @property
+            def _constructor(self):
+                return CustomDataFrame
+
+            _constructor_sliced = CustomSeries
+
+            def custom_frame_function(self):
+                return "OK"
+
+        data = {"col1": range(10), "col2": range(10)}
+        cdf = CustomDataFrame(data)
+
+        # Did we get back our own DF class?
+        assert isinstance(cdf, CustomDataFrame)
+
+        # Do we get back our own Series class after selecting a column?
+        cdf_series = cdf.col1
+        assert isinstance(cdf_series, CustomSeries)
+        assert cdf_series.custom_series_function() == "OK"
+
+        # Do we get back our own DF class after slicing row-wise?
+        cdf_rows = cdf[1:5]
+        assert isinstance(cdf_rows, CustomDataFrame)
+        assert cdf_rows.custom_frame_function() == "OK"
+
+        # Make sure sliced part of multi-index frame is custom class
+        mcol = MultiIndex.from_tuples([("A", "A"), ("A", "B")])
+        cdf_multi = CustomDataFrame([[0, 1], [2, 3]], columns=mcol)
+        assert isinstance(cdf_multi["A"], CustomDataFrame)
+
+        mcol = MultiIndex.from_tuples([("A", ""), ("B", "")])
+        cdf_multi2 = CustomDataFrame([[0, 1], [2, 3]], columns=mcol)
+        assert isinstance(cdf_multi2["A"], CustomSeries)
+
+    def test_dataframe_metadata(self):
+        df = tm.SubclassedDataFrame(
+            {"X": [1, 2, 3], "Y": [1, 2, 3]}, index=["a", "b", "c"]
+        )
+        df.testattr = "XXX"
+
+        assert df.testattr == "XXX"
+        assert df[["X"]].testattr == "XXX"
+        assert df.loc[["a", "b"], :].testattr == "XXX"
+        assert df.iloc[[0, 1], :].testattr == "XXX"
+
+        # see gh-9776
+        assert df.iloc[0:1, :].testattr == "XXX"
+
+        # see gh-10553
+        unpickled = tm.round_trip_pickle(df)
+        tm.assert_frame_equal(df, unpickled)
+        assert df._metadata == unpickled._metadata
+        assert df.testattr == unpickled.testattr
+
+    def test_indexing_sliced(self):
+        # GH 11559
+        df = tm.SubclassedDataFrame(
+            {"X": [1, 2, 3], "Y": [4, 5, 6], "Z": [7, 8, 9]}, index=["a", "b", "c"]
+        )
+        res = df.loc[:, "X"]
+        exp = tm.SubclassedSeries([1, 2, 3], index=list("abc"), name="X")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+        res = df.iloc[:, 1]
+        exp = tm.SubclassedSeries([4, 5, 6], index=list("abc"), name="Y")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+        res = df.loc[:, "Z"]
+        exp = tm.SubclassedSeries([7, 8, 9], index=list("abc"), name="Z")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+        res = df.loc["a", :]
+        exp = tm.SubclassedSeries([1, 4, 7], index=list("XYZ"), name="a")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+        res = df.iloc[1, :]
+        exp = tm.SubclassedSeries([2, 5, 8], index=list("XYZ"), name="b")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+        res = df.loc["c", :]
+        exp = tm.SubclassedSeries([3, 6, 9], index=list("XYZ"), name="c")
+        tm.assert_series_equal(res, exp)
+        assert isinstance(res, tm.SubclassedSeries)
+
+    def test_subclass_attr_err_propagation(self):
+        # GH 11808
+        class A(DataFrame):
+            @property
+            def nonexistence(self):
+                return self.i_dont_exist
+
+        with pytest.raises(AttributeError, match=".*i_dont_exist.*"):
+            A().nonexistence
+
+    def test_subclass_align(self):
+        # GH 12983
+        df1 = tm.SubclassedDataFrame(
+            {"a": [1, 3, 5], "b": [1, 3, 5]}, index=list("ACE")
+        )
+        df2 = tm.SubclassedDataFrame(
+            {"c": [1, 2, 4], "d": [1, 2, 4]}, index=list("ABD")
+        )
+
+        res1, res2 = df1.align(df2, axis=0)
+        exp1 = tm.SubclassedDataFrame(
+            {"a": [1, np.nan, 3, np.nan, 5], "b": [1, np.nan, 3, np.nan, 5]},
+            index=list("ABCDE"),
+        )
+        exp2 = tm.SubclassedDataFrame(
+            {"c": [1, 2, np.nan, 4, np.nan], "d": [1, 2, np.nan, 4, np.nan]},
+            index=list("ABCDE"),
+        )
+        assert isinstance(res1, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(res1, exp1)
+        assert isinstance(res2, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(res2, exp2)
+
+        res1, res2 = df1.a.align(df2.c)
+        assert isinstance(res1, tm.SubclassedSeries)
+        tm.assert_series_equal(res1, exp1.a)
+        assert isinstance(res2, tm.SubclassedSeries)
+        tm.assert_series_equal(res2, exp2.c)
+
+    def test_subclass_align_combinations(self):
+        # GH 12983
+        df = tm.SubclassedDataFrame({"a": [1, 3, 5], "b": [1, 3, 5]}, index=list("ACE"))
+        s = tm.SubclassedSeries([1, 2, 4], index=list("ABD"), name="x")
+
+        # frame + series
+        res1, res2 = df.align(s, axis=0)
+        exp1 = tm.SubclassedDataFrame(
+            {"a": [1, np.nan, 3, np.nan, 5], "b": [1, np.nan, 3, np.nan, 5]},
+            index=list("ABCDE"),
+        )
+        # name is lost when
+        exp2 = tm.SubclassedSeries(
+            [1, 2, np.nan, 4, np.nan], index=list("ABCDE"), name="x"
+        )
+
+        assert isinstance(res1, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(res1, exp1)
+        assert isinstance(res2, tm.SubclassedSeries)
+        tm.assert_series_equal(res2, exp2)
+
+        # series + frame
+        res1, res2 = s.align(df)
+        assert isinstance(res1, tm.SubclassedSeries)
+        tm.assert_series_equal(res1, exp2)
+        assert isinstance(res2, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(res2, exp1)
+
+    def test_subclass_iterrows(self):
+        # GH 13977
+        df = tm.SubclassedDataFrame({"a": [1]})
+        for i, row in df.iterrows():
+            assert isinstance(row, tm.SubclassedSeries)
+            tm.assert_series_equal(row, df.loc[i])
+
+    def test_subclass_stack(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+            index=["a", "b", "c"],
+            columns=["X", "Y", "Z"],
+        )
+
+        res = df.stack()
+        exp = tm.SubclassedSeries(
+            [1, 2, 3, 4, 5, 6, 7, 8, 9], index=[list("aaabbbccc"), list("XYZXYZXYZ")]
+        )
+
+        tm.assert_series_equal(res, exp)
+
+    def test_subclass_stack_multi(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [[10, 11, 12, 13], [20, 21, 22, 23], [30, 31, 32, 33], [40, 41, 42, 43]],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AABB"), list("cdcd"))), names=["aaa", "ccc"]
+            ),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWXX"), list("yzyz"))), names=["www", "yyy"]
+            ),
+        )
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10, 12],
+                [11, 13],
+                [20, 22],
+                [21, 23],
+                [30, 32],
+                [31, 33],
+                [40, 42],
+                [41, 43],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AAAABBBB"), list("ccddccdd"), list("yzyzyzyz"))),
+                names=["aaa", "ccc", "yyy"],
+            ),
+            columns=Index(["W", "X"], name="www"),
+        )
+
+        res = df.stack()
+        tm.assert_frame_equal(res, exp)
+
+        res = df.stack("yyy")
+        tm.assert_frame_equal(res, exp)
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10, 11],
+                [12, 13],
+                [20, 21],
+                [22, 23],
+                [30, 31],
+                [32, 33],
+                [40, 41],
+                [42, 43],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AAAABBBB"), list("ccddccdd"), list("WXWXWXWX"))),
+                names=["aaa", "ccc", "www"],
+            ),
+            columns=Index(["y", "z"], name="yyy"),
+        )
+
+        res = df.stack("www")
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_stack_multi_mixed(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [
+                [10, 11, 12.0, 13.0],
+                [20, 21, 22.0, 23.0],
+                [30, 31, 32.0, 33.0],
+                [40, 41, 42.0, 43.0],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AABB"), list("cdcd"))), names=["aaa", "ccc"]
+            ),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWXX"), list("yzyz"))), names=["www", "yyy"]
+            ),
+        )
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10, 12.0],
+                [11, 13.0],
+                [20, 22.0],
+                [21, 23.0],
+                [30, 32.0],
+                [31, 33.0],
+                [40, 42.0],
+                [41, 43.0],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AAAABBBB"), list("ccddccdd"), list("yzyzyzyz"))),
+                names=["aaa", "ccc", "yyy"],
+            ),
+            columns=Index(["W", "X"], name="www"),
+        )
+
+        res = df.stack()
+        tm.assert_frame_equal(res, exp)
+
+        res = df.stack("yyy")
+        tm.assert_frame_equal(res, exp)
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10.0, 11.0],
+                [12.0, 13.0],
+                [20.0, 21.0],
+                [22.0, 23.0],
+                [30.0, 31.0],
+                [32.0, 33.0],
+                [40.0, 41.0],
+                [42.0, 43.0],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AAAABBBB"), list("ccddccdd"), list("WXWXWXWX"))),
+                names=["aaa", "ccc", "www"],
+            ),
+            columns=Index(["y", "z"], name="yyy"),
+        )
+
+        res = df.stack("www")
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_unstack(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+            index=["a", "b", "c"],
+            columns=["X", "Y", "Z"],
+        )
+
+        res = df.unstack()
+        exp = tm.SubclassedSeries(
+            [1, 4, 7, 2, 5, 8, 3, 6, 9], index=[list("XXXYYYZZZ"), list("abcabcabc")]
+        )
+
+        tm.assert_series_equal(res, exp)
+
+    def test_subclass_unstack_multi(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [[10, 11, 12, 13], [20, 21, 22, 23], [30, 31, 32, 33], [40, 41, 42, 43]],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AABB"), list("cdcd"))), names=["aaa", "ccc"]
+            ),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWXX"), list("yzyz"))), names=["www", "yyy"]
+            ),
+        )
+
+        exp = tm.SubclassedDataFrame(
+            [[10, 20, 11, 21, 12, 22, 13, 23], [30, 40, 31, 41, 32, 42, 33, 43]],
+            index=Index(["A", "B"], name="aaa"),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWWWXXXX"), list("yyzzyyzz"), list("cdcdcdcd"))),
+                names=["www", "yyy", "ccc"],
+            ),
+        )
+
+        res = df.unstack()
+        tm.assert_frame_equal(res, exp)
+
+        res = df.unstack("ccc")
+        tm.assert_frame_equal(res, exp)
+
+        exp = tm.SubclassedDataFrame(
+            [[10, 30, 11, 31, 12, 32, 13, 33], [20, 40, 21, 41, 22, 42, 23, 43]],
+            index=Index(["c", "d"], name="ccc"),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWWWXXXX"), list("yyzzyyzz"), list("ABABABAB"))),
+                names=["www", "yyy", "aaa"],
+            ),
+        )
+
+        res = df.unstack("aaa")
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_unstack_multi_mixed(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            [
+                [10, 11, 12.0, 13.0],
+                [20, 21, 22.0, 23.0],
+                [30, 31, 32.0, 33.0],
+                [40, 41, 42.0, 43.0],
+            ],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AABB"), list("cdcd"))), names=["aaa", "ccc"]
+            ),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWXX"), list("yzyz"))), names=["www", "yyy"]
+            ),
+        )
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10, 20, 11, 21, 12.0, 22.0, 13.0, 23.0],
+                [30, 40, 31, 41, 32.0, 42.0, 33.0, 43.0],
+            ],
+            index=Index(["A", "B"], name="aaa"),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWWWXXXX"), list("yyzzyyzz"), list("cdcdcdcd"))),
+                names=["www", "yyy", "ccc"],
+            ),
+        )
+
+        res = df.unstack()
+        tm.assert_frame_equal(res, exp)
+
+        res = df.unstack("ccc")
+        tm.assert_frame_equal(res, exp)
+
+        exp = tm.SubclassedDataFrame(
+            [
+                [10, 30, 11, 31, 12.0, 32.0, 13.0, 33.0],
+                [20, 40, 21, 41, 22.0, 42.0, 23.0, 43.0],
+            ],
+            index=Index(["c", "d"], name="ccc"),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWWWXXXX"), list("yyzzyyzz"), list("ABABABAB"))),
+                names=["www", "yyy", "aaa"],
+            ),
+        )
+
+        res = df.unstack("aaa")
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_pivot(self):
+        # GH 15564
+        df = tm.SubclassedDataFrame(
+            {
+                "index": ["A", "B", "C", "C", "B", "A"],
+                "columns": ["One", "One", "One", "Two", "Two", "Two"],
+                "values": [1.0, 2.0, 3.0, 3.0, 2.0, 1.0],
+            }
+        )
+
+        pivoted = df.pivot(index="index", columns="columns", values="values")
+
+        expected = tm.SubclassedDataFrame(
+            {
+                "One": {"A": 1.0, "B": 2.0, "C": 3.0},
+                "Two": {"A": 1.0, "B": 2.0, "C": 3.0},
+            }
+        )
+
+        expected.index.name, expected.columns.name = "index", "columns"
+
+        tm.assert_frame_equal(pivoted, expected)
+
+    def test_subclassed_melt(self):
+        # GH 15564
+        cheese = tm.SubclassedDataFrame(
+            {
+                "first": ["John", "Mary"],
+                "last": ["Doe", "Bo"],
+                "height": [5.5, 6.0],
+                "weight": [130, 150],
+            }
+        )
+
+        melted = pd.melt(cheese, id_vars=["first", "last"])
+
+        expected = tm.SubclassedDataFrame(
+            [
+                ["John", "Doe", "height", 5.5],
+                ["Mary", "Bo", "height", 6.0],
+                ["John", "Doe", "weight", 130],
+                ["Mary", "Bo", "weight", 150],
+            ],
+            columns=["first", "last", "variable", "value"],
+        )
+
+        tm.assert_frame_equal(melted, expected)
+
+    def test_subclassed_wide_to_long(self):
+        # GH 9762
+
+        x = np.random.default_rng(2).standard_normal(3)
+        df = tm.SubclassedDataFrame(
+            {
+                "A1970": {0: "a", 1: "b", 2: "c"},
+                "A1980": {0: "d", 1: "e", 2: "f"},
+                "B1970": {0: 2.5, 1: 1.2, 2: 0.7},
+                "B1980": {0: 3.2, 1: 1.3, 2: 0.1},
+                "X": dict(zip(range(3), x)),
+            }
+        )
+
+        df["id"] = df.index
+        exp_data = {
+            "X": x.tolist() + x.tolist(),
+            "A": ["a", "b", "c", "d", "e", "f"],
+            "B": [2.5, 1.2, 0.7, 3.2, 1.3, 0.1],
+            "year": [1970, 1970, 1970, 1980, 1980, 1980],
+            "id": [0, 1, 2, 0, 1, 2],
+        }
+        expected = tm.SubclassedDataFrame(exp_data)
+        expected = expected.set_index(["id", "year"])[["X", "A", "B"]]
+        long_frame = pd.wide_to_long(df, ["A", "B"], i="id", j="year")
+
+        tm.assert_frame_equal(long_frame, expected)
+
+    def test_subclassed_apply(self):
+        # GH 19822
+
+        def check_row_subclass(row):
+            assert isinstance(row, tm.SubclassedSeries)
+
+        def stretch(row):
+            if row["variable"] == "height":
+                row["value"] += 0.5
+            return row
+
+        df = tm.SubclassedDataFrame(
+            [
+                ["John", "Doe", "height", 5.5],
+                ["Mary", "Bo", "height", 6.0],
+                ["John", "Doe", "weight", 130],
+                ["Mary", "Bo", "weight", 150],
+            ],
+            columns=["first", "last", "variable", "value"],
+        )
+
+        df.apply(lambda x: check_row_subclass(x))
+        df.apply(lambda x: check_row_subclass(x), axis=1)
+
+        expected = tm.SubclassedDataFrame(
+            [
+                ["John", "Doe", "height", 6.0],
+                ["Mary", "Bo", "height", 6.5],
+                ["John", "Doe", "weight", 130],
+                ["Mary", "Bo", "weight", 150],
+            ],
+            columns=["first", "last", "variable", "value"],
+        )
+
+        result = df.apply(lambda x: stretch(x), axis=1)
+        assert isinstance(result, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(result, expected)
+
+        expected = tm.SubclassedDataFrame([[1, 2, 3], [1, 2, 3], [1, 2, 3], [1, 2, 3]])
+
+        result = df.apply(lambda x: tm.SubclassedSeries([1, 2, 3]), axis=1)
+        assert isinstance(result, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(result, expected)
+
+        result = df.apply(lambda x: [1, 2, 3], axis=1, result_type="expand")
+        assert isinstance(result, tm.SubclassedDataFrame)
+        tm.assert_frame_equal(result, expected)
+
+        expected = tm.SubclassedSeries([[1, 2, 3], [1, 2, 3], [1, 2, 3], [1, 2, 3]])
+
+        result = df.apply(lambda x: [1, 2, 3], axis=1)
+        assert not isinstance(result, tm.SubclassedDataFrame)
+        tm.assert_series_equal(result, expected)
+
+    def test_subclassed_reductions(self, all_reductions):
+        # GH 25596
+
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = getattr(df, all_reductions)()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_subclassed_count(self):
+        df = tm.SubclassedDataFrame(
+            {
+                "Person": ["John", "Myla", "Lewis", "John", "Myla"],
+                "Age": [24.0, np.nan, 21.0, 33, 26],
+                "Single": [False, True, True, True, False],
+            }
+        )
+        result = df.count()
+        assert isinstance(result, tm.SubclassedSeries)
+
+        df = tm.SubclassedDataFrame({"A": [1, 0, 3], "B": [0, 5, 6], "C": [7, 8, 0]})
+        result = df.count()
+        assert isinstance(result, tm.SubclassedSeries)
+
+        df = tm.SubclassedDataFrame(
+            [[10, 11, 12, 13], [20, 21, 22, 23], [30, 31, 32, 33], [40, 41, 42, 43]],
+            index=MultiIndex.from_tuples(
+                list(zip(list("AABB"), list("cdcd"))), names=["aaa", "ccc"]
+            ),
+            columns=MultiIndex.from_tuples(
+                list(zip(list("WWXX"), list("yzyz"))), names=["www", "yyy"]
+            ),
+        )
+        result = df.count()
+        assert isinstance(result, tm.SubclassedSeries)
+
+        df = tm.SubclassedDataFrame()
+        result = df.count()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_isin(self):
+        df = tm.SubclassedDataFrame(
+            {"num_legs": [2, 4], "num_wings": [2, 0]}, index=["falcon", "dog"]
+        )
+        result = df.isin([0, 2])
+        assert isinstance(result, tm.SubclassedDataFrame)
+
+    def test_duplicated(self):
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = df.duplicated()
+        assert isinstance(result, tm.SubclassedSeries)
+
+        df = tm.SubclassedDataFrame()
+        result = df.duplicated()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    @pytest.mark.parametrize("idx_method", ["idxmax", "idxmin"])
+    def test_idx(self, idx_method):
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = getattr(df, idx_method)()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_dot(self):
+        df = tm.SubclassedDataFrame([[0, 1, -2, -1], [1, 1, 1, 1]])
+        s = tm.SubclassedSeries([1, 1, 2, 1])
+        result = df.dot(s)
+        assert isinstance(result, tm.SubclassedSeries)
+
+        df = tm.SubclassedDataFrame([[0, 1, -2, -1], [1, 1, 1, 1]])
+        s = tm.SubclassedDataFrame([1, 1, 2, 1])
+        result = df.dot(s)
+        assert isinstance(result, tm.SubclassedDataFrame)
+
+    def test_memory_usage(self):
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = df.memory_usage()
+        assert isinstance(result, tm.SubclassedSeries)
+
+        result = df.memory_usage(index=False)
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_corrwith(self):
+        pytest.importorskip("scipy")
+        index = ["a", "b", "c", "d", "e"]
+        columns = ["one", "two", "three", "four"]
+        df1 = tm.SubclassedDataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            index=index,
+            columns=columns,
+        )
+        df2 = tm.SubclassedDataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=index[:4],
+            columns=columns,
+        )
+        correls = df1.corrwith(df2, axis=1, drop=True, method="kendall")
+
+        assert isinstance(correls, (tm.SubclassedSeries))
+
+    def test_asof(self):
+        N = 3
+        rng = pd.date_range("1/1/1990", periods=N, freq="53s")
+        df = tm.SubclassedDataFrame(
+            {
+                "A": [np.nan, np.nan, np.nan],
+                "B": [np.nan, np.nan, np.nan],
+                "C": [np.nan, np.nan, np.nan],
+            },
+            index=rng,
+        )
+
+        result = df.asof(rng[-2:])
+        assert isinstance(result, tm.SubclassedDataFrame)
+
+        result = df.asof(rng[-2])
+        assert isinstance(result, tm.SubclassedSeries)
+
+        result = df.asof("1989-12-31")
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_idxmin_preserves_subclass(self):
+        # GH 28330
+
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = df.idxmin()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_idxmax_preserves_subclass(self):
+        # GH 28330
+
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = df.idxmax()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_convert_dtypes_preserves_subclass(self):
+        # GH 43668
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+        result = df.convert_dtypes()
+        assert isinstance(result, tm.SubclassedDataFrame)
+
+    def test_convert_dtypes_preserves_subclass_with_constructor(self):
+        class SubclassedDataFrame(DataFrame):
+            @property
+            def _constructor(self):
+                return SubclassedDataFrame
+
+        df = SubclassedDataFrame({"a": [1, 2, 3]})
+        result = df.convert_dtypes()
+        assert isinstance(result, SubclassedDataFrame)
+
+    def test_astype_preserves_subclass(self):
+        # GH#40810
+        df = tm.SubclassedDataFrame({"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]})
+
+        result = df.astype({"A": np.int64, "B": np.int32, "C": np.float64})
+        assert isinstance(result, tm.SubclassedDataFrame)
+
+    def test_equals_subclass(self):
+        # https://github.com/pandas-dev/pandas/pull/34402
+        # allow subclass in both directions
+        df1 = DataFrame({"a": [1, 2, 3]})
+        df2 = tm.SubclassedDataFrame({"a": [1, 2, 3]})
+        assert df1.equals(df2)
+        assert df2.equals(df1)
+
+
+class MySubclassWithMetadata(DataFrame):
+    _metadata = ["my_metadata"]
+
+    def __init__(self, *args, **kwargs) -> None:
+        super().__init__(*args, **kwargs)
+
+        my_metadata = kwargs.pop("my_metadata", None)
+        if args and isinstance(args[0], MySubclassWithMetadata):
+            my_metadata = args[0].my_metadata  # type: ignore[has-type]
+        self.my_metadata = my_metadata
+
+    @property
+    def _constructor(self):
+        return MySubclassWithMetadata
+
+
+def test_constructor_with_metadata():
+    # https://github.com/pandas-dev/pandas/pull/54922
+    # https://github.com/pandas-dev/pandas/issues/55120
+    df = MySubclassWithMetadata(
+        np.random.default_rng(2).random((5, 3)), columns=["A", "B", "C"]
+    )
+    subset = df[["A", "B"]]
+    assert isinstance(subset, MySubclassWithMetadata)
+
+
+def test_constructor_with_metadata_from_records():
+    # GH#57008
+    df = MySubclassWithMetadata.from_records([{"a": 1, "b": 2}])
+    assert df.my_metadata is None
+    assert type(df) is MySubclassWithMetadata
+
+
+class SimpleDataFrameSubClass(DataFrame):
+    """A subclass of DataFrame that does not define a constructor."""
+
+
+class SimpleSeriesSubClass(Series):
+    """A subclass of Series that does not define a constructor."""
+
+
+class TestSubclassWithoutConstructor:
+    def test_copy_df(self):
+        expected = DataFrame({"a": [1, 2, 3]})
+        result = SimpleDataFrameSubClass(expected).copy()
+
+        assert (
+            type(result) is DataFrame
+        )  # assert_frame_equal only checks isinstance(lhs, type(rhs))
+        tm.assert_frame_equal(result, expected)
+
+    def test_copy_series(self):
+        expected = Series([1, 2, 3])
+        result = SimpleSeriesSubClass(expected).copy()
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_to_frame(self):
+        orig = Series([1, 2, 3])
+        expected = orig.to_frame()
+        result = SimpleSeriesSubClass(orig).to_frame()
+
+        assert (
+            type(result) is DataFrame
+        )  # assert_frame_equal only checks isinstance(lhs, type(rhs))
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby(self):
+        df = SimpleDataFrameSubClass(DataFrame({"a": [1, 2, 3]}))
+
+        for _, v in df.groupby("a"):
+            assert type(v) is DataFrame
diff --git a/pandas/pandas/tests/frame/test_ufunc.py b/pandas/pandas/tests/frame/test_ufunc.py
new file mode 100644
index 0000000000000..092e65dd4b431
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_ufunc.py
@@ -0,0 +1,312 @@
+from functools import partial
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.types import is_extension_array_dtype
+
+dtypes = [
+    "int64",
+    "Int64",
+    {"A": "int64", "B": "Int64"},
+]
+
+
+@pytest.mark.parametrize("dtype", dtypes)
+def test_unary_unary(dtype):
+    # unary input, unary output
+    values = np.array([[-1, -1], [1, 1]], dtype="int64")
+    df = pd.DataFrame(values, columns=["A", "B"], index=["a", "b"]).astype(dtype=dtype)
+    result = np.positive(df)
+    expected = pd.DataFrame(
+        np.positive(values), index=df.index, columns=df.columns
+    ).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", dtypes)
+def test_unary_binary(request, dtype):
+    # unary input, binary output
+    if is_extension_array_dtype(dtype) or isinstance(dtype, dict):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Extension / mixed with multiple outputs not implemented."
+            )
+        )
+
+    values = np.array([[-1, -1], [1, 1]], dtype="int64")
+    df = pd.DataFrame(values, columns=["A", "B"], index=["a", "b"]).astype(dtype=dtype)
+    result_pandas = np.modf(df)
+    assert isinstance(result_pandas, tuple)
+    assert len(result_pandas) == 2
+    expected_numpy = np.modf(values)
+
+    for result, b in zip(result_pandas, expected_numpy):
+        expected = pd.DataFrame(b, index=df.index, columns=df.columns)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", dtypes)
+def test_binary_input_dispatch_binop(dtype):
+    # binop ufuncs are dispatched to our dunder methods.
+    values = np.array([[-1, -1], [1, 1]], dtype="int64")
+    df = pd.DataFrame(values, columns=["A", "B"], index=["a", "b"]).astype(dtype=dtype)
+    result = np.add(df, df)
+    expected = pd.DataFrame(
+        np.add(values, values), index=df.index, columns=df.columns
+    ).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func,arg,expected",
+    [
+        (np.add, 1, [2, 3, 4, 5]),
+        (
+            partial(np.add, where=[[False, True], [True, False]]),  # type: ignore[misc]
+            np.array([[1, 1], [1, 1]]),
+            [0, 3, 4, 0],
+        ),
+        (np.power, np.array([[1, 1], [2, 2]]), [1, 2, 9, 16]),
+        (np.subtract, 2, [-1, 0, 1, 2]),
+        (
+            partial(np.negative, where=np.array([[False, True], [True, False]])),  # type: ignore[misc]
+            None,
+            [0, -2, -3, 0],
+        ),
+    ],
+)
+def test_ufunc_passes_args(func, arg, expected):
+    # GH#40662
+    arr = np.array([[1, 2], [3, 4]])
+    df = pd.DataFrame(arr)
+    result_inplace = np.zeros_like(arr)
+    # 1-argument ufunc
+    if arg is None:
+        result = func(df, out=result_inplace)
+    else:
+        result = func(df, arg, out=result_inplace)
+
+    expected = np.array(expected).reshape(2, 2)
+    tm.assert_numpy_array_equal(result_inplace, expected)
+
+    expected = pd.DataFrame(expected)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype_a", dtypes)
+@pytest.mark.parametrize("dtype_b", dtypes)
+def test_binary_input_aligns_columns(request, dtype_a, dtype_b):
+    if (
+        is_extension_array_dtype(dtype_a)
+        or isinstance(dtype_a, dict)
+        or is_extension_array_dtype(dtype_b)
+        or isinstance(dtype_b, dict)
+    ):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Extension / mixed with multiple inputs not implemented."
+            )
+        )
+
+    df1 = pd.DataFrame({"A": [1, 2], "B": [3, 4]}).astype(dtype_a)
+
+    if isinstance(dtype_a, dict) and isinstance(dtype_b, dict):
+        dtype_b = dtype_b.copy()
+        dtype_b["C"] = dtype_b.pop("B")
+    df2 = pd.DataFrame({"A": [1, 2], "C": [3, 4]}).astype(dtype_b)
+    # As of 2.0, align first before applying the ufunc
+    result = np.heaviside(df1, df2)
+    expected = np.heaviside(
+        np.array([[1, 3, np.nan], [2, 4, np.nan]]),
+        np.array([[1, np.nan, 3], [2, np.nan, 4]]),
+    )
+    expected = pd.DataFrame(expected, index=[0, 1], columns=["A", "B", "C"])
+    tm.assert_frame_equal(result, expected)
+
+    result = np.heaviside(df1, df2.values)
+    expected = pd.DataFrame([[1.0, 1.0], [1.0, 1.0]], columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", dtypes)
+def test_binary_input_aligns_index(request, dtype):
+    if is_extension_array_dtype(dtype) or isinstance(dtype, dict):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Extension / mixed with multiple inputs not implemented."
+            )
+        )
+    df1 = pd.DataFrame({"A": [1, 2], "B": [3, 4]}, index=["a", "b"]).astype(dtype)
+    df2 = pd.DataFrame({"A": [1, 2], "B": [3, 4]}, index=["a", "c"]).astype(dtype)
+    result = np.heaviside(df1, df2)
+    expected = np.heaviside(
+        np.array([[1, 3], [3, 4], [np.nan, np.nan]]),
+        np.array([[1, 3], [np.nan, np.nan], [3, 4]]),
+    )
+    # TODO(FloatArray): this will be Float64Dtype.
+    expected = pd.DataFrame(expected, index=["a", "b", "c"], columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+    result = np.heaviside(df1, df2.values)
+    expected = pd.DataFrame(
+        [[1.0, 1.0], [1.0, 1.0]], columns=["A", "B"], index=["a", "b"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_binary_frame_series_raises():
+    # We don't currently implement
+    df = pd.DataFrame({"A": [1, 2]})
+    with pytest.raises(NotImplementedError, match="logaddexp"):
+        np.logaddexp(df, df["A"])
+
+    with pytest.raises(NotImplementedError, match="logaddexp"):
+        np.logaddexp(df["A"], df)
+
+
+def test_unary_accumulate_axis():
+    # https://github.com/pandas-dev/pandas/issues/39259
+    df = pd.DataFrame({"a": [1, 3, 2, 4]})
+    result = np.maximum.accumulate(df)
+    expected = pd.DataFrame({"a": [1, 3, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+
+    df = pd.DataFrame({"a": [1, 3, 2, 4], "b": [0.1, 4.0, 3.0, 2.0]})
+    result = np.maximum.accumulate(df)
+    # in theory could preserve int dtype for default axis=0
+    expected = pd.DataFrame({"a": [1.0, 3.0, 3.0, 4.0], "b": [0.1, 4.0, 4.0, 4.0]})
+    tm.assert_frame_equal(result, expected)
+
+    result = np.maximum.accumulate(df, axis=0)
+    tm.assert_frame_equal(result, expected)
+
+    result = np.maximum.accumulate(df, axis=1)
+    expected = pd.DataFrame({"a": [1.0, 3.0, 2.0, 4.0], "b": [1.0, 4.0, 3.0, 4.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_outer_disallowed():
+    df = pd.DataFrame({"A": [1, 2]})
+    with pytest.raises(NotImplementedError, match=""):
+        # deprecation enforced in 2.0
+        np.subtract.outer(df, df)
+
+
+def test_alignment_deprecation_enforced():
+    # Enforced in 2.0
+    # https://github.com/pandas-dev/pandas/issues/39184
+    df1 = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df2 = pd.DataFrame({"b": [1, 2, 3], "c": [4, 5, 6]})
+    s1 = pd.Series([1, 2], index=["a", "b"])
+    s2 = pd.Series([1, 2], index=["b", "c"])
+
+    # binary dataframe / dataframe
+    expected = pd.DataFrame({"a": [2, 4, 6], "b": [8, 10, 12]})
+
+    with tm.assert_produces_warning(None):
+        # aligned -> no warning!
+        result = np.add(df1, df1)
+    tm.assert_frame_equal(result, expected)
+
+    result = np.add(df1, df2.values)
+    tm.assert_frame_equal(result, expected)
+
+    result = np.add(df1, df2)
+    expected = pd.DataFrame({"a": [np.nan] * 3, "b": [5, 7, 9], "c": [np.nan] * 3})
+    tm.assert_frame_equal(result, expected)
+
+    result = np.add(df1.values, df2)
+    expected = pd.DataFrame({"b": [2, 4, 6], "c": [8, 10, 12]})
+    tm.assert_frame_equal(result, expected)
+
+    # binary dataframe / series
+    expected = pd.DataFrame({"a": [2, 3, 4], "b": [6, 7, 8]})
+
+    with tm.assert_produces_warning(None):
+        # aligned -> no warning!
+        result = np.add(df1, s1)
+    tm.assert_frame_equal(result, expected)
+
+    result = np.add(df1, s2.values)
+    tm.assert_frame_equal(result, expected)
+
+    expected = pd.DataFrame(
+        {"a": [np.nan] * 3, "b": [5.0, 6.0, 7.0], "c": [np.nan] * 3}
+    )
+    result = np.add(df1, s2)
+    tm.assert_frame_equal(result, expected)
+
+    msg = "Cannot apply ufunc <ufunc 'add'> to mixed DataFrame and Series inputs."
+    with pytest.raises(NotImplementedError, match=msg):
+        np.add(s2, df1)
+
+
+@pytest.mark.single_cpu
+def test_alignment_deprecation_many_inputs_enforced():
+    # Enforced in 2.0
+    # https://github.com/pandas-dev/pandas/issues/39184
+    # test that the deprecation also works with > 2 inputs -> using a numba
+    # written ufunc for this because numpy itself doesn't have such ufuncs
+    numba = pytest.importorskip("numba")
+
+    @numba.vectorize([numba.float64(numba.float64, numba.float64, numba.float64)])
+    def my_ufunc(x, y, z):
+        return x + y + z
+
+    df1 = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    df2 = pd.DataFrame({"b": [1, 2, 3], "c": [4, 5, 6]})
+    df3 = pd.DataFrame({"a": [1, 2, 3], "c": [4, 5, 6]})
+
+    result = my_ufunc(df1, df2, df3)
+    expected = pd.DataFrame(np.full((3, 3), np.nan), columns=["a", "b", "c"])
+    tm.assert_frame_equal(result, expected)
+
+    # all aligned -> no warning
+    with tm.assert_produces_warning(None):
+        result = my_ufunc(df1, df1, df1)
+    expected = pd.DataFrame([[3.0, 12.0], [6.0, 15.0], [9.0, 18.0]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+    # mixed frame / arrays
+    msg = (
+        r"operands could not be broadcast together with shapes \(3,3\) \(3,3\) \(3,2\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        my_ufunc(df1, df2, df3.values)
+
+    # single frame -> no warning
+    with tm.assert_produces_warning(None):
+        result = my_ufunc(df1, df2.values, df3.values)
+    tm.assert_frame_equal(result, expected)
+
+    # takes indices of first frame
+    msg = (
+        r"operands could not be broadcast together with shapes \(3,2\) \(3,3\) \(3,3\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        my_ufunc(df1.values, df2, df3)
+
+
+def test_array_ufuncs_for_many_arguments():
+    # GH39853
+    def add3(x, y, z):
+        return x + y + z
+
+    ufunc = np.frompyfunc(add3, 3, 1)
+    df = pd.DataFrame([[1, 2], [3, 4]])
+
+    result = ufunc(df, df, 1)
+    expected = pd.DataFrame([[3, 5], [7, 9]], dtype=object)
+    tm.assert_frame_equal(result, expected)
+
+    ser = pd.Series([1, 2])
+    msg = (
+        "Cannot apply ufunc <ufunc 'add3 (vectorized)'> "
+        "to mixed DataFrame and Series inputs."
+    )
+    with pytest.raises(NotImplementedError, match=re.escape(msg)):
+        ufunc(df, df, ser)
diff --git a/pandas/pandas/tests/frame/test_unary.py b/pandas/pandas/tests/frame/test_unary.py
new file mode 100644
index 0000000000000..652f52bd226af
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_unary.py
@@ -0,0 +1,185 @@
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gte1p25
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestDataFrameUnaryOperators:
+    # __pos__, __neg__, __invert__
+
+    @pytest.mark.parametrize(
+        "df_data,expected_data",
+        [
+            ([-1, 1], [1, -1]),
+            ([False, True], [True, False]),
+            (pd.to_timedelta([-1, 1]), pd.to_timedelta([1, -1])),
+        ],
+    )
+    def test_neg_numeric(self, df_data, expected_data):
+        df = pd.DataFrame({"a": df_data})
+        expected = pd.DataFrame({"a": expected_data})
+        tm.assert_frame_equal(-df, expected)
+        tm.assert_series_equal(-df["a"], expected["a"])
+
+    @pytest.mark.parametrize(
+        "df, expected",
+        [
+            (np.array([1, 2], dtype=object), np.array([-1, -2], dtype=object)),
+            ([Decimal("1.0"), Decimal("2.0")], [Decimal("-1.0"), Decimal("-2.0")]),
+        ],
+    )
+    def test_neg_object(self, df, expected):
+        # GH#21380
+        df = pd.DataFrame({"a": df})
+        expected = pd.DataFrame({"a": expected})
+        tm.assert_frame_equal(-df, expected)
+        tm.assert_series_equal(-df["a"], expected["a"])
+
+    @pytest.mark.parametrize(
+        "df_data",
+        [
+            ["a", "b"],
+            pd.to_datetime(["2017-01-22", "1970-01-01"]),
+        ],
+    )
+    def test_neg_raises(self, df_data, using_infer_string):
+        df = pd.DataFrame({"a": df_data})
+        msg = (
+            "bad operand type for unary -: 'str'|"
+            r"bad operand type for unary -: 'DatetimeArray'|"
+            "unary '-' not supported for dtype"
+        )
+        with pytest.raises(TypeError, match=msg):
+            (-df)
+        with pytest.raises(TypeError, match=msg):
+            (-df["a"])
+
+    def test_invert(self, float_frame):
+        df = float_frame
+
+        tm.assert_frame_equal(-(df < 0), ~(df < 0))
+
+    def test_invert_mixed(self):
+        shape = (10, 5)
+        df = pd.concat(
+            [
+                pd.DataFrame(np.zeros(shape, dtype="bool")),
+                pd.DataFrame(np.zeros(shape, dtype=int)),
+            ],
+            axis=1,
+            ignore_index=True,
+        )
+        result = ~df
+        expected = pd.concat(
+            [
+                pd.DataFrame(np.ones(shape, dtype="bool")),
+                pd.DataFrame(-np.ones(shape, dtype=int)),
+            ],
+            axis=1,
+            ignore_index=True,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_invert_empty_not_input(self):
+        # GH#51032
+        df = pd.DataFrame()
+        result = ~df
+        tm.assert_frame_equal(df, result)
+        assert df is not result
+
+    @pytest.mark.parametrize(
+        "df_data",
+        [
+            [-1, 1],
+            [False, True],
+            pd.to_timedelta([-1, 1]),
+        ],
+    )
+    def test_pos_numeric(self, df_data):
+        # GH#16073
+        df = pd.DataFrame({"a": df_data})
+        tm.assert_frame_equal(+df, df)
+        tm.assert_series_equal(+df["a"], df["a"])
+
+    @pytest.mark.parametrize(
+        "df_data",
+        [
+            np.array([-1, 2], dtype=object),
+            [Decimal("-1.0"), Decimal("2.0")],
+        ],
+    )
+    def test_pos_object(self, df_data):
+        # GH#21380
+        df = pd.DataFrame({"a": df_data})
+        tm.assert_frame_equal(+df, df)
+        tm.assert_series_equal(+df["a"], df["a"])
+
+    @pytest.mark.filterwarnings("ignore:Applying:DeprecationWarning")
+    def test_pos_object_raises(self):
+        # GH#21380
+        df = pd.DataFrame({"a": ["a", "b"]})
+        if np_version_gte1p25:
+            with pytest.raises(
+                TypeError, match=r"^bad operand type for unary \+: \'str\'$"
+            ):
+                tm.assert_frame_equal(+df, df)
+        else:
+            tm.assert_series_equal(+df["a"], df["a"])
+
+    def test_pos_raises(self):
+        df = pd.DataFrame({"a": pd.to_datetime(["2017-01-22", "1970-01-01"])})
+        msg = r"bad operand type for unary \+: 'DatetimeArray'"
+        with pytest.raises(TypeError, match=msg):
+            (+df)
+        with pytest.raises(TypeError, match=msg):
+            (+df["a"])
+
+    def test_unary_nullable(self):
+        df = pd.DataFrame(
+            {
+                "a": pd.array([1, -2, 3, pd.NA], dtype="Int64"),
+                "b": pd.array([4.0, -5.0, 6.0, pd.NA], dtype="Float32"),
+                "c": pd.array([True, False, False, pd.NA], dtype="boolean"),
+                # include numpy bool to make sure bool-vs-boolean behavior
+                #  is consistent in non-NA locations
+                "d": np.array([True, False, False, True]),
+            }
+        )
+
+        result = +df
+        res_ufunc = np.positive(df)
+        expected = df
+        # TODO: assert that we have copies?
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(res_ufunc, expected)
+
+        result = -df
+        res_ufunc = np.negative(df)
+        expected = pd.DataFrame(
+            {
+                "a": pd.array([-1, 2, -3, pd.NA], dtype="Int64"),
+                "b": pd.array([-4.0, 5.0, -6.0, pd.NA], dtype="Float32"),
+                "c": pd.array([False, True, True, pd.NA], dtype="boolean"),
+                "d": np.array([False, True, True, False]),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(res_ufunc, expected)
+
+        result = abs(df)
+        res_ufunc = np.abs(df)
+        expected = pd.DataFrame(
+            {
+                "a": pd.array([1, 2, 3, pd.NA], dtype="Int64"),
+                "b": pd.array([4.0, 5.0, 6.0, pd.NA], dtype="Float32"),
+                "c": pd.array([True, False, False, pd.NA], dtype="boolean"),
+                "d": np.array([True, False, False, True]),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(res_ufunc, expected)
diff --git a/pandas/pandas/tests/frame/test_validate.py b/pandas/pandas/tests/frame/test_validate.py
new file mode 100644
index 0000000000000..fdeecba29a617
--- /dev/null
+++ b/pandas/pandas/tests/frame/test_validate.py
@@ -0,0 +1,37 @@
+import pytest
+
+from pandas.core.frame import DataFrame
+
+
+class TestDataFrameValidate:
+    """Tests for error handling related to data types of method arguments."""
+
+    @pytest.mark.parametrize(
+        "func",
+        [
+            "query",
+            "eval",
+            "set_index",
+            "reset_index",
+            "dropna",
+            "drop_duplicates",
+            "sort_values",
+        ],
+    )
+    @pytest.mark.parametrize("inplace", [1, "True", [1, 2, 3], 5.0])
+    def test_validate_bool_args(self, func, inplace):
+        dataframe = DataFrame({"a": [1, 2], "b": [3, 4]})
+        msg = 'For argument "inplace" expected type bool'
+        kwargs = {"inplace": inplace}
+
+        if func == "query":
+            kwargs["expr"] = "a > b"
+        elif func == "eval":
+            kwargs["expr"] = "a + b"
+        elif func == "set_index":
+            kwargs["keys"] = ["a"]
+        elif func == "sort_values":
+            kwargs["by"] = ["a"]
+
+        with pytest.raises(ValueError, match=msg):
+            getattr(dataframe, func)(**kwargs)
diff --git a/pandas/pandas/tests/generic/__init__.py b/pandas/pandas/tests/generic/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/generic/test_duplicate_labels.py b/pandas/pandas/tests/generic/test_duplicate_labels.py
new file mode 100644
index 0000000000000..cfa3cabbc1747
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_duplicate_labels.py
@@ -0,0 +1,393 @@
+"""Tests dealing with the NDFrame.allows_duplicates."""
+
+import operator
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+not_implemented = pytest.mark.xfail(reason="Not implemented.")
+
+# ----------------------------------------------------------------------------
+# Preservation
+
+
+class TestPreserves:
+    @pytest.mark.parametrize(
+        "cls, data",
+        [
+            (pd.Series, np.array([])),
+            (pd.Series, [1, 2]),
+            (pd.DataFrame, {}),
+            (pd.DataFrame, {"A": [1, 2]}),
+        ],
+    )
+    def test_construction_ok(self, cls, data):
+        result = cls(data)
+        assert result.flags.allows_duplicate_labels is True
+
+        result = cls(data).set_flags(allows_duplicate_labels=False)
+        assert result.flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize(
+        "func",
+        [
+            operator.itemgetter(["a"]),
+            operator.methodcaller("add", 1),
+            operator.methodcaller("rename", str.upper),
+            operator.methodcaller("rename", "name"),
+            operator.methodcaller("abs"),
+            np.abs,
+        ],
+    )
+    def test_preserved_series(self, func):
+        s = pd.Series([0, 1], index=["a", "b"]).set_flags(allows_duplicate_labels=False)
+        assert func(s).flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize("index", [["a", "b", "c"], ["a", "b"]])
+    # TODO: frame
+    @not_implemented
+    def test_align(self, index):
+        other = pd.Series(0, index=index)
+        s = pd.Series([0, 1], index=["a", "b"]).set_flags(allows_duplicate_labels=False)
+        a, b = s.align(other)
+        assert a.flags.allows_duplicate_labels is False
+        assert b.flags.allows_duplicate_labels is False
+
+    def test_preserved_frame(self):
+        df = pd.DataFrame({"A": [1, 2], "B": [3, 4]}, index=["a", "b"]).set_flags(
+            allows_duplicate_labels=False
+        )
+        assert df.loc[["a"]].flags.allows_duplicate_labels is False
+        assert df.loc[:, ["A", "B"]].flags.allows_duplicate_labels is False
+
+    def test_to_frame(self):
+        ser = pd.Series(dtype=float).set_flags(allows_duplicate_labels=False)
+        assert ser.to_frame().flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize("func", ["add", "sub"])
+    @pytest.mark.parametrize("frame", [False, True])
+    @pytest.mark.parametrize("other", [1, pd.Series([1, 2], name="A")])
+    def test_binops(self, func, other, frame):
+        df = pd.Series([1, 2], name="A", index=["a", "b"]).set_flags(
+            allows_duplicate_labels=False
+        )
+        if frame:
+            df = df.to_frame()
+        if isinstance(other, pd.Series) and frame:
+            other = other.to_frame()
+        func = operator.methodcaller(func, other)
+        assert df.flags.allows_duplicate_labels is False
+        assert func(df).flags.allows_duplicate_labels is False
+
+    def test_preserve_getitem(self):
+        df = pd.DataFrame({"A": [1, 2]}).set_flags(allows_duplicate_labels=False)
+        assert df[["A"]].flags.allows_duplicate_labels is False
+        assert df["A"].flags.allows_duplicate_labels is False
+        assert df.loc[0].flags.allows_duplicate_labels is False
+        assert df.loc[[0]].flags.allows_duplicate_labels is False
+        assert df.loc[0, ["A"]].flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize(
+        "objs, kwargs",
+        [
+            # Series
+            (
+                [
+                    pd.Series(1, index=["a", "b"]),
+                    pd.Series(2, index=["c", "d"]),
+                ],
+                {},
+            ),
+            (
+                [
+                    pd.Series(1, index=["a", "b"]),
+                    pd.Series(2, index=["a", "b"]),
+                ],
+                {"ignore_index": True},
+            ),
+            (
+                [
+                    pd.Series(1, index=["a", "b"]),
+                    pd.Series(2, index=["a", "b"]),
+                ],
+                {"axis": 1},
+            ),
+            # Frame
+            (
+                [
+                    pd.DataFrame({"A": [1, 2]}, index=["a", "b"]),
+                    pd.DataFrame({"A": [1, 2]}, index=["c", "d"]),
+                ],
+                {},
+            ),
+            (
+                [
+                    pd.DataFrame({"A": [1, 2]}, index=["a", "b"]),
+                    pd.DataFrame({"A": [1, 2]}, index=["a", "b"]),
+                ],
+                {"ignore_index": True},
+            ),
+            (
+                [
+                    pd.DataFrame({"A": [1, 2]}, index=["a", "b"]),
+                    pd.DataFrame({"B": [1, 2]}, index=["a", "b"]),
+                ],
+                {"axis": 1},
+            ),
+            # Series / Frame
+            (
+                [
+                    pd.DataFrame({"A": [1, 2]}, index=["a", "b"]),
+                    pd.Series([1, 2], index=["a", "b"], name="B"),
+                ],
+                {"axis": 1},
+            ),
+        ],
+    )
+    def test_concat(self, objs, kwargs):
+        objs = [x.set_flags(allows_duplicate_labels=False) for x in objs]
+        result = pd.concat(objs, **kwargs)
+        assert result.flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize(
+        "left, right, expected",
+        [
+            # false false false
+            pytest.param(
+                pd.DataFrame({"A": [0, 1]}, index=["a", "b"]).set_flags(
+                    allows_duplicate_labels=False
+                ),
+                pd.DataFrame({"B": [0, 1]}, index=["a", "d"]).set_flags(
+                    allows_duplicate_labels=False
+                ),
+                False,
+                marks=not_implemented,
+            ),
+            # false true false
+            pytest.param(
+                pd.DataFrame({"A": [0, 1]}, index=["a", "b"]).set_flags(
+                    allows_duplicate_labels=False
+                ),
+                pd.DataFrame({"B": [0, 1]}, index=["a", "d"]),
+                False,
+                marks=not_implemented,
+            ),
+            # true true true
+            (
+                pd.DataFrame({"A": [0, 1]}, index=["a", "b"]),
+                pd.DataFrame({"B": [0, 1]}, index=["a", "d"]),
+                True,
+            ),
+        ],
+    )
+    def test_merge(self, left, right, expected):
+        result = pd.merge(left, right, left_index=True, right_index=True)
+        assert result.flags.allows_duplicate_labels is expected
+
+    @not_implemented
+    def test_groupby(self):
+        # XXX: This is under tested
+        # TODO:
+        #  - apply
+        #  - transform
+        #  - Should passing a grouper that disallows duplicates propagate?
+        df = pd.DataFrame({"A": [1, 2, 3]}).set_flags(allows_duplicate_labels=False)
+        result = df.groupby([0, 0, 1]).agg("count")
+        assert result.flags.allows_duplicate_labels is False
+
+    @pytest.mark.parametrize("frame", [True, False])
+    @not_implemented
+    def test_window(self, frame):
+        df = pd.Series(
+            1,
+            index=pd.date_range("2000", periods=12),
+            name="A",
+            allows_duplicate_labels=False,
+        )
+        if frame:
+            df = df.to_frame()
+        assert df.rolling(3).mean().flags.allows_duplicate_labels is False
+        assert df.ewm(3).mean().flags.allows_duplicate_labels is False
+        assert df.expanding(3).mean().flags.allows_duplicate_labels is False
+
+
+# ----------------------------------------------------------------------------
+# Raises
+
+
+class TestRaises:
+    @pytest.mark.parametrize(
+        "cls, axes",
+        [
+            (pd.Series, {"index": ["a", "a"], "dtype": float}),
+            (pd.DataFrame, {"index": ["a", "a"]}),
+            (pd.DataFrame, {"index": ["a", "a"], "columns": ["b", "b"]}),
+            (pd.DataFrame, {"columns": ["b", "b"]}),
+        ],
+    )
+    def test_set_flags_with_duplicates(self, cls, axes):
+        result = cls(**axes)
+        assert result.flags.allows_duplicate_labels is True
+
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            cls(**axes).set_flags(allows_duplicate_labels=False)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            pd.Series(index=[0, 0], dtype=float),
+            pd.DataFrame(index=[0, 0]),
+            pd.DataFrame(columns=[0, 0]),
+        ],
+    )
+    def test_setting_allows_duplicate_labels_raises(self, data):
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            data.flags.allows_duplicate_labels = False
+
+        assert data.flags.allows_duplicate_labels is True
+
+    def test_series_raises(self):
+        a = pd.Series(0, index=["a", "b"])
+        b = pd.Series([0, 1], index=["a", "b"]).set_flags(allows_duplicate_labels=False)
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            pd.concat([a, b])
+
+    @pytest.mark.parametrize(
+        "getter, target",
+        [
+            (operator.itemgetter(["A", "A"]), None),
+            # loc
+            (operator.itemgetter(["a", "a"]), "loc"),
+            pytest.param(operator.itemgetter(("a", ["A", "A"])), "loc"),
+            (operator.itemgetter((["a", "a"], "A")), "loc"),
+            # iloc
+            (operator.itemgetter([0, 0]), "iloc"),
+            pytest.param(operator.itemgetter((0, [0, 0])), "iloc"),
+            pytest.param(operator.itemgetter(([0, 0], 0)), "iloc"),
+        ],
+    )
+    def test_getitem_raises(self, getter, target):
+        df = pd.DataFrame({"A": [1, 2], "B": [3, 4]}, index=["a", "b"]).set_flags(
+            allows_duplicate_labels=False
+        )
+        if target:
+            # df, df.loc, or df.iloc
+            target = getattr(df, target)
+        else:
+            target = df
+
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            getter(target)
+
+    def test_concat_raises(self):
+        objs = [
+            pd.Series(1, index=[0, 1], name="a"),
+            pd.Series(2, index=[0, 1], name="a"),
+        ]
+        objs = [x.set_flags(allows_duplicate_labels=False) for x in objs]
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            pd.concat(objs, axis=1)
+
+    @not_implemented
+    def test_merge_raises(self):
+        a = pd.DataFrame({"A": [0, 1, 2]}, index=["a", "b", "c"]).set_flags(
+            allows_duplicate_labels=False
+        )
+        b = pd.DataFrame({"B": [0, 1, 2]}, index=["a", "b", "b"])
+        msg = "Index has duplicates."
+        with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+            pd.merge(a, b, left_index=True, right_index=True)
+
+
+@pytest.mark.parametrize(
+    "idx",
+    [
+        pd.Index([1, 1]),
+        pd.Index(["a", "a"]),
+        pd.Index([1.1, 1.1]),
+        pd.PeriodIndex([pd.Period("2000", "D")] * 2),
+        pd.DatetimeIndex([pd.Timestamp("2000")] * 2),
+        pd.TimedeltaIndex([pd.Timedelta("1D")] * 2),
+        pd.CategoricalIndex(["a", "a"]),
+        pd.IntervalIndex([pd.Interval(0, 1)] * 2),
+        pd.MultiIndex.from_tuples([("a", 1), ("a", 1)]),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def test_raises_basic(idx):
+    msg = "Index has duplicates."
+    with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+        pd.Series(1, index=idx).set_flags(allows_duplicate_labels=False)
+
+    with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+        pd.DataFrame({"A": [1, 1]}, index=idx).set_flags(allows_duplicate_labels=False)
+
+    with pytest.raises(pd.errors.DuplicateLabelError, match=msg):
+        pd.DataFrame([[1, 2]], columns=idx).set_flags(allows_duplicate_labels=False)
+
+
+def test_format_duplicate_labels_message():
+    idx = pd.Index(["a", "b", "a", "b", "c"])
+    result = idx._format_duplicate_message()
+    expected = pd.DataFrame(
+        {"positions": [[0, 2], [1, 3]]}, index=pd.Index(["a", "b"], name="label")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_format_duplicate_labels_message_multi():
+    idx = pd.MultiIndex.from_product([["A"], ["a", "b", "a", "b", "c"]])
+    result = idx._format_duplicate_message()
+    expected = pd.DataFrame(
+        {"positions": [[0, 2], [1, 3]]},
+        index=pd.MultiIndex.from_product([["A"], ["a", "b"]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dataframe_insert_raises():
+    df = pd.DataFrame({"A": [1, 2]}).set_flags(allows_duplicate_labels=False)
+    msg = "Cannot specify"
+    with pytest.raises(ValueError, match=msg):
+        df.insert(0, "A", [3, 4], allow_duplicates=True)
+
+
+@pytest.mark.parametrize(
+    "method, frame_only",
+    [
+        (operator.methodcaller("set_index", "A", inplace=True), True),
+        (operator.methodcaller("reset_index", inplace=True), True),
+        (operator.methodcaller("rename", lambda x: x, inplace=True), False),
+    ],
+)
+def test_inplace_raises(method, frame_only):
+    df = pd.DataFrame({"A": [0, 0], "B": [1, 2]}).set_flags(
+        allows_duplicate_labels=False
+    )
+    s = df["A"]
+    s.flags.allows_duplicate_labels = False
+    msg = "Cannot specify"
+
+    with pytest.raises(ValueError, match=msg):
+        method(df)
+    if not frame_only:
+        with pytest.raises(ValueError, match=msg):
+            method(s)
+
+
+def test_pickle():
+    a = pd.Series([1, 2]).set_flags(allows_duplicate_labels=False)
+    b = tm.round_trip_pickle(a)
+    tm.assert_series_equal(a, b)
+
+    a = pd.DataFrame({"A": []}).set_flags(allows_duplicate_labels=False)
+    b = tm.round_trip_pickle(a)
+    tm.assert_frame_equal(a, b)
diff --git a/pandas/pandas/tests/generic/test_finalize.py b/pandas/pandas/tests/generic/test_finalize.py
new file mode 100644
index 0000000000000..a88090b00499d
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_finalize.py
@@ -0,0 +1,712 @@
+"""
+An exhaustive list of pandas methods exercising NDFrame.__finalize__.
+"""
+
+import operator
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+
+# TODO:
+# * Binary methods (mul, div, etc.)
+# * Binary outputs (align, etc.)
+# * top-level methods (concat, merge, get_dummies, etc.)
+# * window
+# * cumulative reductions
+
+not_implemented_mark = pytest.mark.xfail(reason="not implemented")
+
+mi = pd.MultiIndex.from_product([["a", "b"], [0, 1]], names=["A", "B"])
+
+frame_data = ({"A": [1]},)
+frame_mi_data = ({"A": [1, 2, 3, 4]}, mi)
+
+
+# Tuple of
+# - Callable: Constructor (Series, DataFrame)
+# - Tuple: Constructor args
+# - Callable: pass the constructed value with attrs set to this.
+
+_all_methods = [
+    (pd.Series, ([0],), operator.methodcaller("take", [])),
+    (pd.Series, ([0],), operator.methodcaller("__getitem__", [True])),
+    (pd.Series, ([0],), operator.methodcaller("repeat", 2)),
+    (pd.Series, ([0],), operator.methodcaller("reset_index")),
+    (pd.Series, ([0],), operator.methodcaller("reset_index", drop=True)),
+    (pd.Series, ([0],), operator.methodcaller("to_frame")),
+    (pd.Series, ([0, 0],), operator.methodcaller("drop_duplicates")),
+    (pd.Series, ([0, 0],), operator.methodcaller("duplicated")),
+    (pd.Series, ([0, 0],), operator.methodcaller("round")),
+    (pd.Series, ([0, 0],), operator.methodcaller("rename", lambda x: x + 1)),
+    (pd.Series, ([0, 0],), operator.methodcaller("rename", "name")),
+    (pd.Series, ([0, 0],), operator.methodcaller("set_axis", ["a", "b"])),
+    (pd.Series, ([0, 0],), operator.methodcaller("reindex", [1, 0])),
+    (pd.Series, ([0, 0],), operator.methodcaller("drop", [0])),
+    (pd.Series, (pd.array([0, pd.NA]),), operator.methodcaller("fillna", 0)),
+    (pd.Series, ([0, 0],), operator.methodcaller("replace", {0: 1})),
+    (pd.Series, ([0, 0],), operator.methodcaller("shift")),
+    (pd.Series, ([0, 0],), operator.methodcaller("isin", [0, 1])),
+    (pd.Series, ([0, 0],), operator.methodcaller("between", 0, 2)),
+    (pd.Series, ([0, 0],), operator.methodcaller("isna")),
+    (pd.Series, ([0, 0],), operator.methodcaller("isnull")),
+    (pd.Series, ([0, 0],), operator.methodcaller("notna")),
+    (pd.Series, ([0, 0],), operator.methodcaller("notnull")),
+    (pd.Series, ([1],), operator.methodcaller("add", pd.Series([1]))),
+    # TODO: mul, div, etc.
+    (
+        pd.Series,
+        ([0], pd.period_range("2000", periods=1)),
+        operator.methodcaller("to_timestamp"),
+    ),
+    (
+        pd.Series,
+        ([0], pd.date_range("2000", periods=1)),
+        operator.methodcaller("to_period"),
+    ),
+    pytest.param(
+        (
+            pd.DataFrame,
+            frame_data,
+            operator.methodcaller("dot", pd.DataFrame(index=["A"])),
+        ),
+        marks=pytest.mark.xfail(reason="Implement binary finalize"),
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("transpose")),
+    (pd.DataFrame, frame_data, operator.methodcaller("__getitem__", "A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("__getitem__", ["A"])),
+    (pd.DataFrame, frame_data, operator.methodcaller("__getitem__", np.array([True]))),
+    (pd.DataFrame, ({("A", "a"): [1]},), operator.methodcaller("__getitem__", ["A"])),
+    (pd.DataFrame, frame_data, operator.methodcaller("query", "A == 1")),
+    (pd.DataFrame, frame_data, operator.methodcaller("eval", "A + 1", engine="python")),
+    (pd.DataFrame, frame_data, operator.methodcaller("select_dtypes", include="int")),
+    (pd.DataFrame, frame_data, operator.methodcaller("assign", b=1)),
+    (pd.DataFrame, frame_data, operator.methodcaller("set_axis", ["A"])),
+    (pd.DataFrame, frame_data, operator.methodcaller("reindex", [0, 1])),
+    (pd.DataFrame, frame_data, operator.methodcaller("drop", columns=["A"])),
+    (pd.DataFrame, frame_data, operator.methodcaller("drop", index=[0])),
+    (pd.DataFrame, frame_data, operator.methodcaller("rename", columns={"A": "a"})),
+    (pd.DataFrame, frame_data, operator.methodcaller("rename", index=lambda x: x)),
+    (pd.DataFrame, frame_data, operator.methodcaller("fillna", "A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("set_index", "A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("reset_index")),
+    (pd.DataFrame, frame_data, operator.methodcaller("isna")),
+    (pd.DataFrame, frame_data, operator.methodcaller("isnull")),
+    (pd.DataFrame, frame_data, operator.methodcaller("notna")),
+    (pd.DataFrame, frame_data, operator.methodcaller("notnull")),
+    (pd.DataFrame, frame_data, operator.methodcaller("dropna")),
+    (pd.DataFrame, frame_data, operator.methodcaller("drop_duplicates")),
+    (pd.DataFrame, frame_data, operator.methodcaller("duplicated")),
+    (pd.DataFrame, frame_data, operator.methodcaller("sort_values", by="A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("sort_index")),
+    (pd.DataFrame, frame_data, operator.methodcaller("nlargest", 1, "A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("nsmallest", 1, "A")),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("swaplevel")),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("add", pd.DataFrame(*frame_data)),
+    ),
+    # TODO: div, mul, etc.
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("combine", pd.DataFrame(*frame_data), operator.add),
+    ),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("combine_first", pd.DataFrame(*frame_data)),
+    ),
+    pytest.param(
+        (
+            pd.DataFrame,
+            frame_data,
+            operator.methodcaller("update", pd.DataFrame(*frame_data)),
+        ),
+        marks=not_implemented_mark,
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("pivot", columns="A")),
+    (
+        pd.DataFrame,
+        ({"A": [1], "B": [1]},),
+        operator.methodcaller("pivot_table", columns="A"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1], "B": [1]},),
+        operator.methodcaller("pivot_table", columns="A", aggfunc=["mean", "sum"]),
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("stack")),
+    (pd.DataFrame, frame_data, operator.methodcaller("explode", "A")),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("unstack")),
+    (
+        pd.DataFrame,
+        ({"A": ["a", "b", "c"], "B": [1, 3, 5], "C": [2, 4, 6]},),
+        operator.methodcaller("melt", id_vars=["A"], value_vars=["B"]),
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("map", lambda x: x)),
+    pytest.param(
+        (
+            pd.DataFrame,
+            frame_data,
+            operator.methodcaller("merge", pd.DataFrame({"A": [1]})),
+        ),
+        marks=not_implemented_mark,
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("round", 2)),
+    (pd.DataFrame, frame_data, operator.methodcaller("corr")),
+    pytest.param(
+        (pd.DataFrame, frame_data, operator.methodcaller("cov")),
+        marks=[
+            pytest.mark.filterwarnings("ignore::RuntimeWarning"),
+        ],
+    ),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("corrwith", pd.DataFrame(*frame_data)),
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("count")),
+    (pd.DataFrame, frame_data, operator.methodcaller("nunique")),
+    (pd.DataFrame, frame_data, operator.methodcaller("idxmin")),
+    (pd.DataFrame, frame_data, operator.methodcaller("idxmax")),
+    (pd.DataFrame, frame_data, operator.methodcaller("mode")),
+    (pd.Series, [0], operator.methodcaller("mode")),
+    (pd.DataFrame, frame_data, operator.methodcaller("median")),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("quantile", numeric_only=True),
+    ),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("quantile", q=[0.25, 0.75], numeric_only=True),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [pd.Timedelta(days=1), pd.Timedelta(days=2)]},),
+        operator.methodcaller("quantile", numeric_only=False),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [np.datetime64("2022-01-01"), np.datetime64("2022-01-02")]},),
+        operator.methodcaller("quantile", numeric_only=True),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1]}, [pd.Period("2000", "D")]),
+        operator.methodcaller("to_timestamp"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1]}, [pd.Timestamp("2000")]),
+        operator.methodcaller("to_period", freq="D"),
+    ),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("isin", [1])),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("isin", pd.Series([1]))),
+    (
+        pd.DataFrame,
+        frame_mi_data,
+        operator.methodcaller("isin", pd.DataFrame({"A": [1]})),
+    ),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("droplevel", "A")),
+    (pd.DataFrame, frame_data, operator.methodcaller("pop", "A")),
+    # Squeeze on columns, otherwise we'll end up with a scalar
+    (pd.DataFrame, frame_data, operator.methodcaller("squeeze", axis="columns")),
+    (pd.Series, ([1, 2],), operator.methodcaller("squeeze")),
+    (pd.Series, ([1, 2],), operator.methodcaller("rename_axis", index="a")),
+    (pd.DataFrame, frame_data, operator.methodcaller("rename_axis", columns="a")),
+    # Unary ops
+    (pd.DataFrame, frame_data, operator.neg),
+    (pd.Series, [1], operator.neg),
+    (pd.DataFrame, frame_data, operator.pos),
+    (pd.Series, [1], operator.pos),
+    (pd.DataFrame, frame_data, operator.inv),
+    (pd.Series, [1], operator.inv),
+    (pd.DataFrame, frame_data, abs),
+    (pd.Series, [1], abs),
+    (pd.DataFrame, frame_data, round),
+    (pd.Series, [1], round),
+    (pd.DataFrame, frame_data, operator.methodcaller("take", [0, 0])),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("xs", "a")),
+    (pd.Series, (1, mi), operator.methodcaller("xs", "a")),
+    (pd.DataFrame, frame_data, operator.methodcaller("get", "A")),
+    (
+        pd.DataFrame,
+        frame_data,
+        operator.methodcaller("reindex_like", pd.DataFrame({"A": [1, 2, 3]})),
+    ),
+    (
+        pd.Series,
+        frame_data,
+        operator.methodcaller("reindex_like", pd.Series([0, 1, 2])),
+    ),
+    (pd.DataFrame, frame_data, operator.methodcaller("add_prefix", "_")),
+    (pd.DataFrame, frame_data, operator.methodcaller("add_suffix", "_")),
+    (pd.Series, (1, ["a", "b"]), operator.methodcaller("add_prefix", "_")),
+    (pd.Series, (1, ["a", "b"]), operator.methodcaller("add_suffix", "_")),
+    (pd.Series, ([3, 2],), operator.methodcaller("sort_values")),
+    (pd.Series, ([1] * 10,), operator.methodcaller("head")),
+    (pd.DataFrame, ({"A": [1] * 10},), operator.methodcaller("head")),
+    (pd.Series, ([1] * 10,), operator.methodcaller("tail")),
+    (pd.DataFrame, ({"A": [1] * 10},), operator.methodcaller("tail")),
+    (pd.Series, ([1, 2],), operator.methodcaller("sample", n=2, replace=True)),
+    (pd.DataFrame, (frame_data,), operator.methodcaller("sample", n=2, replace=True)),
+    (pd.Series, ([1, 2],), operator.methodcaller("astype", float)),
+    (pd.DataFrame, frame_data, operator.methodcaller("astype", float)),
+    (pd.Series, ([1, 2],), operator.methodcaller("copy")),
+    (pd.DataFrame, frame_data, operator.methodcaller("copy")),
+    (pd.Series, ([1, 2], None, object), operator.methodcaller("infer_objects")),
+    (
+        pd.DataFrame,
+        ({"A": np.array([1, 2], dtype=object)},),
+        operator.methodcaller("infer_objects"),
+    ),
+    (pd.Series, ([1, 2],), operator.methodcaller("convert_dtypes")),
+    (pd.DataFrame, frame_data, operator.methodcaller("convert_dtypes")),
+    (pd.Series, ([1, None, 3],), operator.methodcaller("interpolate")),
+    (pd.DataFrame, ({"A": [1, None, 3]},), operator.methodcaller("interpolate")),
+    (pd.Series, ([1, 2],), operator.methodcaller("clip", lower=1)),
+    (pd.DataFrame, frame_data, operator.methodcaller("clip", lower=1)),
+    (
+        pd.Series,
+        (1, pd.date_range("2000", periods=4)),
+        operator.methodcaller("asfreq", "h"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1, 1, 1, 1]}, pd.date_range("2000", periods=4)),
+        operator.methodcaller("asfreq", "h"),
+    ),
+    (
+        pd.Series,
+        (1, pd.date_range("2000", periods=4)),
+        operator.methodcaller("at_time", "12:00"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1, 1, 1, 1]}, pd.date_range("2000", periods=4)),
+        operator.methodcaller("at_time", "12:00"),
+    ),
+    (
+        pd.Series,
+        (1, pd.date_range("2000", periods=4)),
+        operator.methodcaller("between_time", "12:00", "13:00"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1, 1, 1, 1]}, pd.date_range("2000", periods=4)),
+        operator.methodcaller("between_time", "12:00", "13:00"),
+    ),
+    (pd.Series, ([1, 2],), operator.methodcaller("rank")),
+    (pd.DataFrame, frame_data, operator.methodcaller("rank")),
+    (pd.Series, ([1, 2],), operator.methodcaller("where", np.array([True, False]))),
+    (pd.DataFrame, frame_data, operator.methodcaller("where", np.array([[True]]))),
+    (pd.Series, ([1, 2],), operator.methodcaller("mask", np.array([True, False]))),
+    (pd.DataFrame, frame_data, operator.methodcaller("mask", np.array([[True]]))),
+    (pd.Series, ([1, 2],), operator.methodcaller("truncate", before=0)),
+    (pd.DataFrame, frame_data, operator.methodcaller("truncate", before=0)),
+    (
+        pd.Series,
+        (1, pd.date_range("2000", periods=4, tz="UTC")),
+        operator.methodcaller("tz_convert", "CET"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1, 1, 1, 1]}, pd.date_range("2000", periods=4, tz="UTC")),
+        operator.methodcaller("tz_convert", "CET"),
+    ),
+    (
+        pd.Series,
+        (1, pd.date_range("2000", periods=4)),
+        operator.methodcaller("tz_localize", "CET"),
+    ),
+    (
+        pd.DataFrame,
+        ({"A": [1, 1, 1, 1]}, pd.date_range("2000", periods=4)),
+        operator.methodcaller("tz_localize", "CET"),
+    ),
+    (pd.Series, ([1, 2],), operator.methodcaller("describe")),
+    (pd.DataFrame, frame_data, operator.methodcaller("describe")),
+    (pd.Series, ([1, 2],), operator.methodcaller("pct_change")),
+    (pd.DataFrame, frame_data, operator.methodcaller("pct_change")),
+    (pd.Series, ([1],), operator.methodcaller("transform", lambda x: x - x.min())),
+    (
+        pd.DataFrame,
+        frame_mi_data,
+        operator.methodcaller("transform", lambda x: x - x.min()),
+    ),
+    (pd.Series, ([1],), operator.methodcaller("apply", lambda x: x)),
+    (pd.DataFrame, frame_mi_data, operator.methodcaller("apply", lambda x: x)),
+    # Cumulative reductions
+    (pd.Series, ([1],), operator.methodcaller("cumsum")),
+    (pd.DataFrame, frame_data, operator.methodcaller("cumsum")),
+    (pd.Series, ([1],), operator.methodcaller("cummin")),
+    (pd.DataFrame, frame_data, operator.methodcaller("cummin")),
+    (pd.Series, ([1],), operator.methodcaller("cummax")),
+    (pd.DataFrame, frame_data, operator.methodcaller("cummax")),
+    (pd.Series, ([1],), operator.methodcaller("cumprod")),
+    (pd.DataFrame, frame_data, operator.methodcaller("cumprod")),
+    # Reductions
+    (pd.DataFrame, frame_data, operator.methodcaller("any")),
+    (pd.DataFrame, frame_data, operator.methodcaller("all")),
+    (pd.DataFrame, frame_data, operator.methodcaller("min")),
+    (pd.DataFrame, frame_data, operator.methodcaller("max")),
+    (pd.DataFrame, frame_data, operator.methodcaller("sum")),
+    (pd.DataFrame, frame_data, operator.methodcaller("std")),
+    (pd.DataFrame, frame_data, operator.methodcaller("mean")),
+    (pd.DataFrame, frame_data, operator.methodcaller("prod")),
+    (pd.DataFrame, frame_data, operator.methodcaller("sem")),
+    (pd.DataFrame, frame_data, operator.methodcaller("skew")),
+    (pd.DataFrame, frame_data, operator.methodcaller("kurt")),
+]
+
+
+def idfn(x):
+    xpr = re.compile(r"'(.*)?'")
+    m = xpr.search(str(x))
+    if m:
+        return m.group(1)
+    else:
+        return str(x)
+
+
+@pytest.mark.parametrize("ndframe_method", _all_methods, ids=lambda x: idfn(x[-1]))
+def test_finalize_called(ndframe_method):
+    cls, init_args, method = ndframe_method
+    ndframe = cls(*init_args)
+
+    ndframe.attrs = {"a": 1}
+    result = method(ndframe)
+
+    assert result.attrs == {"a": 1}
+
+
+@not_implemented_mark
+def test_finalize_called_eval_numexpr():
+    pytest.importorskip("numexpr")
+    df = pd.DataFrame({"A": [1, 2]})
+    df.attrs["A"] = 1
+    result = df.eval("A + 1", engine="numexpr")
+    assert result.attrs == {"A": 1}
+
+
+# ----------------------------------------------------------------------------
+# Binary operations
+
+
+@pytest.mark.parametrize("annotate", ["left", "right", "both"])
+@pytest.mark.parametrize(
+    "args",
+    [
+        (1, pd.Series([1])),
+        (1, pd.DataFrame({"A": [1]})),
+        (pd.Series([1]), 1),
+        (pd.DataFrame({"A": [1]}), 1),
+        (pd.Series([1]), pd.Series([1])),
+        (pd.DataFrame({"A": [1]}), pd.DataFrame({"A": [1]})),
+        (pd.Series([1]), pd.DataFrame({"A": [1]})),
+        (pd.DataFrame({"A": [1]}), pd.Series([1])),
+    ],
+    ids=lambda x: f"({type(x[0]).__name__},{type(x[1]).__name__})",
+)
+def test_binops(request, args, annotate, all_binary_operators):
+    # This generates 624 tests... Is that needed?
+    left, right = args
+    if isinstance(left, (pd.DataFrame, pd.Series)):
+        left.attrs = {}
+    if isinstance(right, (pd.DataFrame, pd.Series)):
+        right.attrs = {}
+
+    if annotate == "left" and isinstance(left, int):
+        pytest.skip("left is an int and doesn't support .attrs")
+    if annotate == "right" and isinstance(right, int):
+        pytest.skip("right is an int and doesn't support .attrs")
+
+    if not (isinstance(left, int) or isinstance(right, int)) and annotate != "both":
+        if not all_binary_operators.__name__.startswith("r"):
+            if annotate == "right" and isinstance(left, type(right)):
+                request.applymarker(
+                    pytest.mark.xfail(
+                        reason=f"{all_binary_operators} doesn't work when right has "
+                        f"attrs and both are {type(left)}"
+                    )
+                )
+            if not isinstance(left, type(right)):
+                if annotate == "left" and isinstance(left, pd.Series):
+                    request.applymarker(
+                        pytest.mark.xfail(
+                            reason=f"{all_binary_operators} doesn't work when the "
+                            "objects are different Series has attrs"
+                        )
+                    )
+                elif annotate == "right" and isinstance(right, pd.Series):
+                    request.applymarker(
+                        pytest.mark.xfail(
+                            reason=f"{all_binary_operators} doesn't work when the "
+                            "objects are different Series has attrs"
+                        )
+                    )
+        else:
+            if annotate == "left" and isinstance(left, type(right)):
+                request.applymarker(
+                    pytest.mark.xfail(
+                        reason=f"{all_binary_operators} doesn't work when left has "
+                        f"attrs and both are {type(left)}"
+                    )
+                )
+            if not isinstance(left, type(right)):
+                if annotate == "right" and isinstance(right, pd.Series):
+                    request.applymarker(
+                        pytest.mark.xfail(
+                            reason=f"{all_binary_operators} doesn't work when the "
+                            "objects are different Series has attrs"
+                        )
+                    )
+                elif annotate == "left" and isinstance(left, pd.Series):
+                    request.applymarker(
+                        pytest.mark.xfail(
+                            reason=f"{all_binary_operators} doesn't work when the "
+                            "objects are different Series has attrs"
+                        )
+                    )
+    if annotate in {"left", "both"} and not isinstance(left, int):
+        left.attrs = {"a": 1}
+    if annotate in {"right", "both"} and not isinstance(right, int):
+        right.attrs = {"a": 1}
+
+    is_cmp = all_binary_operators in [
+        operator.eq,
+        operator.ne,
+        operator.gt,
+        operator.ge,
+        operator.lt,
+        operator.le,
+    ]
+    if is_cmp and isinstance(left, pd.DataFrame) and isinstance(right, pd.Series):
+        # in 2.0 silent alignment on comparisons was removed xref GH#28759
+        left, right = left.align(right, axis=1)
+    elif is_cmp and isinstance(left, pd.Series) and isinstance(right, pd.DataFrame):
+        right, left = right.align(left, axis=1)
+
+    result = all_binary_operators(left, right)
+    assert result.attrs == {"a": 1}
+
+
+# ----------------------------------------------------------------------------
+# Accessors
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        operator.methodcaller("capitalize"),
+        operator.methodcaller("casefold"),
+        operator.methodcaller("cat", ["a"]),
+        operator.methodcaller("contains", "a"),
+        operator.methodcaller("count", "a"),
+        operator.methodcaller("encode", "utf-8"),
+        operator.methodcaller("endswith", "a"),
+        operator.methodcaller("extract", r"(\w)(\d)"),
+        operator.methodcaller("extract", r"(\w)(\d)", expand=False),
+        operator.methodcaller("find", "a"),
+        operator.methodcaller("findall", "a"),
+        operator.methodcaller("get", 0),
+        operator.methodcaller("index", "a"),
+        operator.methodcaller("len"),
+        operator.methodcaller("ljust", 4),
+        operator.methodcaller("lower"),
+        operator.methodcaller("lstrip"),
+        operator.methodcaller("match", r"\w"),
+        operator.methodcaller("normalize", "NFC"),
+        operator.methodcaller("pad", 4),
+        operator.methodcaller("partition", "a"),
+        operator.methodcaller("repeat", 2),
+        operator.methodcaller("replace", "a", "b"),
+        operator.methodcaller("rfind", "a"),
+        operator.methodcaller("rindex", "a"),
+        operator.methodcaller("rjust", 4),
+        operator.methodcaller("rpartition", "a"),
+        operator.methodcaller("rstrip"),
+        operator.methodcaller("slice", 4),
+        operator.methodcaller("slice_replace", 1, repl="a"),
+        operator.methodcaller("startswith", "a"),
+        operator.methodcaller("strip"),
+        operator.methodcaller("swapcase"),
+        operator.methodcaller("translate", {"a": "b"}),
+        operator.methodcaller("upper"),
+        operator.methodcaller("wrap", 4),
+        operator.methodcaller("zfill", 4),
+        operator.methodcaller("isalnum"),
+        operator.methodcaller("isalpha"),
+        operator.methodcaller("isdigit"),
+        operator.methodcaller("isspace"),
+        operator.methodcaller("islower"),
+        operator.methodcaller("isupper"),
+        operator.methodcaller("istitle"),
+        operator.methodcaller("isnumeric"),
+        operator.methodcaller("isdecimal"),
+        operator.methodcaller("get_dummies"),
+    ],
+    ids=idfn,
+)
+def test_string_method(method):
+    s = pd.Series(["a1"])
+    s.attrs = {"a": 1}
+    result = method(s.str)
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        operator.methodcaller("to_period"),
+        operator.methodcaller("tz_localize", "CET"),
+        operator.methodcaller("normalize"),
+        operator.methodcaller("strftime", "%Y"),
+        operator.methodcaller("round", "h"),
+        operator.methodcaller("floor", "h"),
+        operator.methodcaller("ceil", "h"),
+        operator.methodcaller("month_name"),
+        operator.methodcaller("day_name"),
+    ],
+    ids=idfn,
+)
+def test_datetime_method(method):
+    s = pd.Series(pd.date_range("2000", periods=4))
+    s.attrs = {"a": 1}
+    result = method(s.dt)
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "attr",
+    [
+        "date",
+        "time",
+        "timetz",
+        "year",
+        "month",
+        "day",
+        "hour",
+        "minute",
+        "second",
+        "microsecond",
+        "nanosecond",
+        "dayofweek",
+        "day_of_week",
+        "dayofyear",
+        "day_of_year",
+        "quarter",
+        "is_month_start",
+        "is_month_end",
+        "is_quarter_start",
+        "is_quarter_end",
+        "is_year_start",
+        "is_year_end",
+        "is_leap_year",
+        "daysinmonth",
+        "days_in_month",
+    ],
+)
+def test_datetime_property(attr):
+    s = pd.Series(pd.date_range("2000", periods=4))
+    s.attrs = {"a": 1}
+    result = getattr(s.dt, attr)
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "attr", ["days", "seconds", "microseconds", "nanoseconds", "components"]
+)
+def test_timedelta_property(attr):
+    s = pd.Series(pd.timedelta_range("2000", periods=4))
+    s.attrs = {"a": 1}
+    result = getattr(s.dt, attr)
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize("method", [operator.methodcaller("total_seconds")])
+def test_timedelta_methods(method):
+    s = pd.Series(pd.timedelta_range("2000", periods=4))
+    s.attrs = {"a": 1}
+    result = method(s.dt)
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        operator.methodcaller("add_categories", ["c"]),
+        operator.methodcaller("as_ordered"),
+        operator.methodcaller("as_unordered"),
+        lambda x: x.codes,
+        operator.methodcaller("remove_categories", "a"),
+        operator.methodcaller("remove_unused_categories"),
+        operator.methodcaller("rename_categories", {"a": "A", "b": "B"}),
+        operator.methodcaller("reorder_categories", ["b", "a"]),
+        operator.methodcaller("set_categories", ["A", "B"]),
+    ],
+)
+@not_implemented_mark
+def test_categorical_accessor(method):
+    s = pd.Series(["a", "b"], dtype="category")
+    s.attrs = {"a": 1}
+    result = method(s.cat)
+    assert result.attrs == {"a": 1}
+
+
+# ----------------------------------------------------------------------------
+# Groupby
+
+
+@pytest.mark.parametrize(
+    "obj", [pd.Series([0, 0]), pd.DataFrame({"A": [0, 1], "B": [1, 2]})]
+)
+@pytest.mark.parametrize(
+    "method",
+    [
+        operator.methodcaller("sum"),
+        lambda x: x.apply(lambda y: y),
+        lambda x: x.agg("sum"),
+        lambda x: x.agg("mean"),
+        lambda x: x.agg("median"),
+    ],
+)
+def test_groupby_finalize(obj, method):
+    obj.attrs = {"a": 1}
+    result = method(obj.groupby([0, 0], group_keys=False))
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "obj", [pd.Series([0, 0]), pd.DataFrame({"A": [0, 1], "B": [1, 2]})]
+)
+@pytest.mark.parametrize(
+    "method",
+    [
+        lambda x: x.agg(["sum", "count"]),
+        lambda x: x.agg("std"),
+        lambda x: x.agg("var"),
+        lambda x: x.agg("sem"),
+        lambda x: x.agg("size"),
+        lambda x: x.agg("ohlc"),
+    ],
+)
+@not_implemented_mark
+def test_groupby_finalize_not_implemented(obj, method):
+    obj.attrs = {"a": 1}
+    result = method(obj.groupby([0, 0]))
+    assert result.attrs == {"a": 1}
+
+
+def test_finalize_frame_series_name():
+    # https://github.com/pandas-dev/pandas/pull/37186/files#r506978889
+    # ensure we don't copy the column `name` to the Series.
+    df = pd.DataFrame({"name": [1, 2]})
+    result = pd.Series([1, 2]).__finalize__(df)
+    assert result.name is None
diff --git a/pandas/pandas/tests/generic/test_frame.py b/pandas/pandas/tests/generic/test_frame.py
new file mode 100644
index 0000000000000..1d0f491529b56
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_frame.py
@@ -0,0 +1,192 @@
+from copy import deepcopy
+from operator import methodcaller
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDataFrame:
+    @pytest.mark.parametrize("func", ["_set_axis_name", "rename_axis"])
+    def test_set_axis_name(self, func):
+        df = DataFrame([[1, 2], [3, 4]])
+
+        result = methodcaller(func, "foo")(df)
+        assert df.index.name is None
+        assert result.index.name == "foo"
+
+        result = methodcaller(func, "cols", axis=1)(df)
+        assert df.columns.name is None
+        assert result.columns.name == "cols"
+
+    @pytest.mark.parametrize("func", ["_set_axis_name", "rename_axis"])
+    def test_set_axis_name_mi(self, func):
+        df = DataFrame(
+            np.empty((3, 3)),
+            index=MultiIndex.from_tuples([("A", x) for x in list("aBc")]),
+            columns=MultiIndex.from_tuples([("C", x) for x in list("xyz")]),
+        )
+
+        level_names = ["L1", "L2"]
+
+        result = methodcaller(func, level_names)(df)
+        assert result.index.names == level_names
+        assert result.columns.names == [None, None]
+
+        result = methodcaller(func, level_names, axis=1)(df)
+        assert result.columns.names == ["L1", "L2"]
+        assert result.index.names == [None, None]
+
+    def test_nonzero_single_element(self):
+        df = DataFrame([[False, False]])
+        msg_err = "The truth value of a DataFrame is ambiguous"
+        with pytest.raises(ValueError, match=msg_err):
+            bool(df)
+
+    def test_metadata_propagation_indiv_groupby(self):
+        # groupby
+        df = DataFrame(
+            {
+                "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+                "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+                "C": np.random.default_rng(2).standard_normal(8),
+                "D": np.random.default_rng(2).standard_normal(8),
+            }
+        )
+        result = df.groupby("A").sum()
+        tm.assert_metadata_equivalent(df, result)
+
+    def test_metadata_propagation_indiv_resample(self):
+        # resample
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 2)),
+            index=date_range("20130101", periods=1000, freq="s"),
+        )
+        result = df.resample("1min")
+        tm.assert_metadata_equivalent(df, result)
+
+    def test_metadata_propagation_indiv(self, monkeypatch):
+        # merging with override
+        # GH 6923
+
+        def finalize(self, other, method=None, **kwargs):
+            for name in self._metadata:
+                if method == "merge":
+                    left, right = other.left, other.right
+                    value = getattr(left, name, "") + "|" + getattr(right, name, "")
+                    object.__setattr__(self, name, value)
+                elif method == "concat":
+                    value = "+".join(
+                        [getattr(o, name) for o in other.objs if getattr(o, name, None)]
+                    )
+                    object.__setattr__(self, name, value)
+                else:
+                    object.__setattr__(self, name, getattr(other, name, ""))
+
+            return self
+
+        with monkeypatch.context() as m:
+            m.setattr(DataFrame, "_metadata", ["filename"])
+            m.setattr(DataFrame, "__finalize__", finalize)
+
+            df1 = DataFrame(
+                np.random.default_rng(2).integers(0, 4, (3, 2)), columns=["a", "b"]
+            )
+            df2 = DataFrame(
+                np.random.default_rng(2).integers(0, 4, (3, 2)), columns=["c", "d"]
+            )
+            DataFrame._metadata = ["filename"]
+            df1.filename = "fname1.csv"
+            df2.filename = "fname2.csv"
+
+            result = df1.merge(df2, left_on=["a"], right_on=["c"], how="inner")
+            assert result.filename == "fname1.csv|fname2.csv"
+
+            # concat
+            # GH#6927
+            df1 = DataFrame(
+                np.random.default_rng(2).integers(0, 4, (3, 2)), columns=list("ab")
+            )
+            df1.filename = "foo"
+
+            result = pd.concat([df1, df1])
+            assert result.filename == "foo+foo"
+
+    def test_set_attribute(self):
+        # Test for consistent setattr behavior when an attribute and a column
+        # have the same name (Issue #8994)
+        df = DataFrame({"x": [1, 2, 3]})
+
+        df.y = 2
+        df["y"] = [2, 4, 6]
+        df.y = 5
+
+        assert df.y == 5
+        tm.assert_series_equal(df["y"], Series([2, 4, 6], name="y"))
+
+    def test_deepcopy_empty(self):
+        # This test covers empty frame copying with non-empty column sets
+        # as reported in issue GH15370
+        empty_frame = DataFrame(data=[], index=[], columns=["A"])
+        empty_frame_copy = deepcopy(empty_frame)
+
+        tm.assert_frame_equal(empty_frame_copy, empty_frame)
+
+
+# formerly in Generic but only test DataFrame
+class TestDataFrame2:
+    @pytest.mark.parametrize("value", [1, "True", [1, 2, 3], 5.0])
+    def test_validate_bool_args(self, value):
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+        msg = 'For argument "inplace" expected type bool, received type'
+        with pytest.raises(ValueError, match=msg):
+            df.copy().rename_axis(mapper={"a": "x", "b": "y"}, axis=1, inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy().drop("a", axis=1, inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy().fillna(value=0, inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy().replace(to_replace=1, value=7, inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy().interpolate(inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy()._where(cond=df.a > 2, inplace=value)
+
+        with pytest.raises(ValueError, match=msg):
+            df.copy().mask(cond=df.a > 2, inplace=value)
+
+    def test_unexpected_keyword(self):
+        # GH8597
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=["jim", "joe"]
+        )
+        ca = pd.Categorical([0, 0, 2, 2, 3, np.nan])
+        ts = df["joe"].copy()
+        ts[2] = np.nan
+
+        msg = "unexpected keyword"
+        with pytest.raises(TypeError, match=msg):
+            df.drop("joe", axis=1, in_place=True)
+
+        with pytest.raises(TypeError, match=msg):
+            df.reindex([1, 0], inplace=True)
+
+        with pytest.raises(TypeError, match=msg):
+            ca.fillna(0, inplace=True)
+
+        with pytest.raises(TypeError, match=msg):
+            ts.fillna(0, in_place=True)
diff --git a/pandas/pandas/tests/generic/test_generic.py b/pandas/pandas/tests/generic/test_generic.py
new file mode 100644
index 0000000000000..b591b1b1092d4
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_generic.py
@@ -0,0 +1,489 @@
+from copy import (
+    copy,
+    deepcopy,
+)
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_scalar
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+# ----------------------------------------------------------------------
+# Generic types test cases
+
+
+def construct(box, shape, value=None, dtype=None, **kwargs):
+    """
+    construct an object for the given shape
+    if value is specified use that if its a scalar
+    if value is an array, repeat it as needed
+    """
+    if isinstance(shape, int):
+        shape = tuple([shape] * box._AXIS_LEN)
+    if value is not None:
+        if is_scalar(value):
+            if value == "empty":
+                arr = None
+                dtype = np.float64
+
+                # remove the info axis
+                kwargs.pop(box._info_axis_name, None)
+            else:
+                arr = np.empty(shape, dtype=dtype)
+                arr.fill(value)
+        else:
+            fshape = np.prod(shape)
+            arr = value.ravel()
+            new_shape = fshape / arr.shape[0]
+            if fshape % arr.shape[0] != 0:
+                raise Exception("invalid value passed in construct")
+
+            arr = np.repeat(arr, new_shape).reshape(shape)
+    else:
+        arr = np.random.default_rng(2).standard_normal(shape)
+    return box(arr, dtype=dtype, **kwargs)
+
+
+class TestGeneric:
+    @pytest.mark.parametrize(
+        "func",
+        [
+            str.lower,
+            {x: x.lower() for x in list("ABCD")},
+            Series({x: x.lower() for x in list("ABCD")}),
+        ],
+    )
+    def test_rename(self, frame_or_series, func):
+        # single axis
+        idx = list("ABCD")
+
+        for axis in frame_or_series._AXIS_ORDERS:
+            kwargs = {axis: idx}
+            obj = construct(frame_or_series, 4, **kwargs)
+
+            # rename a single axis
+            result = obj.rename(**{axis: func})
+            expected = obj.copy()
+            setattr(expected, axis, list("abcd"))
+            tm.assert_equal(result, expected)
+
+    def test_get_numeric_data(self, frame_or_series):
+        n = 4
+        kwargs = {
+            frame_or_series._get_axis_name(i): list(range(n))
+            for i in range(frame_or_series._AXIS_LEN)
+        }
+
+        # get the numeric data
+        o = construct(frame_or_series, n, **kwargs)
+        result = o._get_numeric_data()
+        tm.assert_equal(result, o)
+
+        # non-inclusion
+        result = o._get_bool_data()
+        expected = construct(frame_or_series, n, value="empty", **kwargs)
+        if isinstance(o, DataFrame):
+            # preserve columns dtype
+            expected.columns = o.columns[:0]
+        tm.assert_equal(result, expected)
+
+        # get the bool data
+        arr = np.array([True, True, False, True])
+        o = construct(frame_or_series, n, value=arr, **kwargs)
+        result = o._get_numeric_data()
+        tm.assert_equal(result, o)
+
+    def test_get_bool_data_empty_preserve_index(self):
+        expected = Series([], dtype="bool")
+        result = expected._get_bool_data()
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_nonzero(self, frame_or_series):
+        # GH 4633
+        # look at the boolean/nonzero behavior for objects
+        obj = construct(frame_or_series, shape=4)
+        msg = f"The truth value of a {frame_or_series.__name__} is ambiguous"
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 0)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 1)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj)
+
+        obj = construct(frame_or_series, shape=4, value=1)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 0)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 1)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj)
+
+        obj = construct(frame_or_series, shape=4, value=np.nan)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 0)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj == 1)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj)
+
+        # empty
+        obj = construct(frame_or_series, shape=0)
+        with pytest.raises(ValueError, match=msg):
+            bool(obj)
+
+        # invalid behaviors
+
+        obj1 = construct(frame_or_series, shape=4, value=1)
+        obj2 = construct(frame_or_series, shape=4, value=1)
+
+        with pytest.raises(ValueError, match=msg):
+            if obj1:
+                pass
+
+        with pytest.raises(ValueError, match=msg):
+            obj1 and obj2
+        with pytest.raises(ValueError, match=msg):
+            obj1 or obj2
+        with pytest.raises(ValueError, match=msg):
+            not obj1
+
+    def test_frame_or_series_compound_dtypes(self, frame_or_series):
+        # see gh-5191
+        # Compound dtypes should raise NotImplementedError.
+
+        def f(dtype):
+            return construct(frame_or_series, shape=3, value=1, dtype=dtype)
+
+        msg = (
+            "compound dtypes are not implemented "
+            f"in the {frame_or_series.__name__} constructor"
+        )
+
+        with pytest.raises(NotImplementedError, match=msg):
+            f([("A", "datetime64[h]"), ("B", "str"), ("C", "int32")])
+
+        # these work (though results may be unexpected)
+        f("int64")
+        f("float64")
+        f("M8[ns]")
+
+    def test_metadata_propagation(self, frame_or_series):
+        # check that the metadata matches up on the resulting ops
+
+        o = construct(frame_or_series, shape=3)
+        o.name = "foo"
+        o2 = construct(frame_or_series, shape=3)
+        o2.name = "bar"
+
+        # ----------
+        # preserving
+        # ----------
+
+        # simple ops with scalars
+        for op in ["__add__", "__sub__", "__truediv__", "__mul__"]:
+            result = getattr(o, op)(1)
+            tm.assert_metadata_equivalent(o, result)
+
+        # ops with like
+        for op in ["__add__", "__sub__", "__truediv__", "__mul__"]:
+            result = getattr(o, op)(o)
+            tm.assert_metadata_equivalent(o, result)
+
+        # simple boolean
+        for op in ["__eq__", "__le__", "__ge__"]:
+            v1 = getattr(o, op)(o)
+            tm.assert_metadata_equivalent(o, v1)
+            tm.assert_metadata_equivalent(o, v1 & v1)
+            tm.assert_metadata_equivalent(o, v1 | v1)
+
+        # combine_first
+        result = o.combine_first(o2)
+        tm.assert_metadata_equivalent(o, result)
+
+        # ---------------------------
+        # non-preserving (by default)
+        # ---------------------------
+
+        # add non-like
+        result = o + o2
+        tm.assert_metadata_equivalent(result)
+
+        # simple boolean
+        for op in ["__eq__", "__le__", "__ge__"]:
+            # this is a name matching op
+            v1 = getattr(o, op)(o)
+            v2 = getattr(o, op)(o2)
+            tm.assert_metadata_equivalent(v2)
+            tm.assert_metadata_equivalent(v1 & v2)
+            tm.assert_metadata_equivalent(v1 | v2)
+
+    def test_size_compat(self, frame_or_series):
+        # GH8846
+        # size property should be defined
+
+        o = construct(frame_or_series, shape=10)
+        assert o.size == np.prod(o.shape)
+        assert o.size == 10 ** len(o.axes)
+
+    def test_split_compat(self, frame_or_series):
+        # xref GH8846
+        o = construct(frame_or_series, shape=10)
+        assert len(np.array_split(o, 5)) == 5
+        assert len(np.array_split(o, 2)) == 2
+
+    # See gh-12301
+    def test_stat_unexpected_keyword(self, frame_or_series):
+        obj = construct(frame_or_series, 5)
+        starwars = "Star Wars"
+        errmsg = "unexpected keyword"
+
+        with pytest.raises(TypeError, match=errmsg):
+            obj.max(epic=starwars)  # stat_function
+        with pytest.raises(TypeError, match=errmsg):
+            obj.var(epic=starwars)  # stat_function_ddof
+        with pytest.raises(TypeError, match=errmsg):
+            obj.sum(epic=starwars)  # cum_function
+        with pytest.raises(TypeError, match=errmsg):
+            obj.any(epic=starwars)  # logical_function
+
+    @pytest.mark.parametrize("func", ["sum", "cumsum", "any", "var"])
+    def test_api_compat(self, func, frame_or_series):
+        # GH 12021
+        # compat for __name__, __qualname__
+
+        obj = construct(frame_or_series, 5)
+        f = getattr(obj, func)
+        assert f.__name__ == func
+        assert f.__qualname__.endswith(func)
+
+    def test_stat_non_defaults_args(self, frame_or_series):
+        obj = construct(frame_or_series, 5)
+        out = np.array([0])
+        errmsg = "the 'out' parameter is not supported"
+
+        with pytest.raises(ValueError, match=errmsg):
+            obj.max(out=out)  # stat_function
+        with pytest.raises(ValueError, match=errmsg):
+            obj.var(out=out)  # stat_function_ddof
+        with pytest.raises(ValueError, match=errmsg):
+            obj.sum(out=out)  # cum_function
+        with pytest.raises(ValueError, match=errmsg):
+            obj.any(out=out)  # logical_function
+
+    def test_truncate_out_of_bounds(self, frame_or_series):
+        # GH11382
+
+        # small
+        shape = [2000] + ([1] * (frame_or_series._AXIS_LEN - 1))
+        small = construct(frame_or_series, shape, dtype="int8", value=1)
+        tm.assert_equal(small.truncate(), small)
+        tm.assert_equal(small.truncate(before=0, after=3e3), small)
+        tm.assert_equal(small.truncate(before=-1, after=2e3), small)
+
+        # big
+        shape = [2_000_000] + ([1] * (frame_or_series._AXIS_LEN - 1))
+        big = construct(frame_or_series, shape, dtype="int8", value=1)
+        tm.assert_equal(big.truncate(), big)
+        tm.assert_equal(big.truncate(before=0, after=3e6), big)
+        tm.assert_equal(big.truncate(before=-1, after=2e6), big)
+
+    @pytest.mark.parametrize(
+        "func",
+        [copy, deepcopy, lambda x: x.copy(deep=False), lambda x: x.copy(deep=True)],
+    )
+    @pytest.mark.parametrize("shape", [0, 1, 2])
+    def test_copy_and_deepcopy(self, frame_or_series, shape, func):
+        # GH 15444
+        obj = construct(frame_or_series, shape)
+        obj_copy = func(obj)
+        assert obj_copy is not obj
+        tm.assert_equal(obj_copy, obj)
+
+
+class TestNDFrame:
+    # tests that don't fit elsewhere
+
+    @pytest.mark.parametrize(
+        "ser",
+        [
+            Series(range(10), dtype=np.float64),
+            Series([str(i) for i in range(10)], dtype=object),
+        ],
+    )
+    def test_squeeze_series_noop(self, ser):
+        # noop
+        tm.assert_series_equal(ser.squeeze(), ser)
+
+    def test_squeeze_frame_noop(self):
+        # noop
+        df = DataFrame(np.eye(2))
+        tm.assert_frame_equal(df.squeeze(), df)
+
+    def test_squeeze_frame_reindex(self):
+        # squeezing
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        ).reindex(columns=["A"])
+        tm.assert_series_equal(df.squeeze(), df["A"])
+
+    def test_squeeze_0_len_dim(self):
+        # don't fail with 0 length dimensions GH11229 & GH8999
+        empty_series = Series([], name="five", dtype=np.float64)
+        empty_frame = DataFrame([empty_series])
+        tm.assert_series_equal(empty_series, empty_series.squeeze())
+        tm.assert_series_equal(empty_series, empty_frame.squeeze())
+
+    def test_squeeze_axis(self):
+        # axis argument
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=1, freq="B"),
+        ).iloc[:, :1]
+        assert df.shape == (1, 1)
+        tm.assert_series_equal(df.squeeze(axis=0), df.iloc[0])
+        tm.assert_series_equal(df.squeeze(axis="index"), df.iloc[0])
+        tm.assert_series_equal(df.squeeze(axis=1), df.iloc[:, 0])
+        tm.assert_series_equal(df.squeeze(axis="columns"), df.iloc[:, 0])
+        assert df.squeeze() == df.iloc[0, 0]
+        msg = "No axis named 2 for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.squeeze(axis=2)
+        msg = "No axis named x for object type DataFrame"
+        with pytest.raises(ValueError, match=msg):
+            df.squeeze(axis="x")
+
+    def test_squeeze_axis_len_3(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=3, freq="B"),
+        )
+        tm.assert_frame_equal(df.squeeze(axis=0), df)
+
+    def test_numpy_squeeze(self):
+        s = Series(range(2), dtype=np.float64)
+        tm.assert_series_equal(np.squeeze(s), s)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        ).reindex(columns=["A"])
+        tm.assert_series_equal(np.squeeze(df), df["A"])
+
+    @pytest.mark.parametrize(
+        "ser",
+        [
+            Series(range(10), dtype=np.float64),
+            Series([str(i) for i in range(10)], dtype=object),
+        ],
+    )
+    def test_transpose_series(self, ser):
+        # calls implementation in pandas/core/base.py
+        tm.assert_series_equal(ser.transpose(), ser)
+
+    def test_transpose_frame(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        tm.assert_frame_equal(df.transpose().transpose(), df)
+
+    def test_numpy_transpose(self, frame_or_series):
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        obj = tm.get_obj(obj, frame_or_series)
+
+        if frame_or_series is Series:
+            # 1D -> np.transpose is no-op
+            tm.assert_series_equal(np.transpose(obj), obj)
+
+        # round-trip preserved
+        tm.assert_equal(np.transpose(np.transpose(obj)), obj)
+
+        msg = "the 'axes' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.transpose(obj, axes=1)
+
+    @pytest.mark.parametrize(
+        "ser",
+        [
+            Series(range(10), dtype=np.float64),
+            Series([str(i) for i in range(10)], dtype=object),
+        ],
+    )
+    def test_take_series(self, ser):
+        indices = [1, 5, -2, 6, 3, -1]
+        out = ser.take(indices)
+        expected = Series(
+            data=ser.values.take(indices),
+            index=ser.index.take(indices),
+            dtype=ser.dtype,
+        )
+        tm.assert_series_equal(out, expected)
+
+    def test_take_frame(self):
+        indices = [1, 5, -2, 6, 3, -1]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        out = df.take(indices)
+        expected = DataFrame(
+            data=df.values.take(indices, axis=0),
+            index=df.index.take(indices),
+            columns=df.columns,
+        )
+        tm.assert_frame_equal(out, expected)
+
+    def test_take_invalid_kwargs(self, frame_or_series):
+        indices = [-3, 2, 0, 1]
+
+        obj = DataFrame(range(5))
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = r"take\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            obj.take(indices, foo=2)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            obj.take(indices, out=indices)
+
+        msg = "the 'mode' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            obj.take(indices, mode="clip")
+
+    def test_axis_classmethods(self, frame_or_series):
+        box = frame_or_series
+        obj = box(dtype=object)
+        values = box._AXIS_TO_AXIS_NUMBER.keys()
+        for v in values:
+            assert obj._get_axis_number(v) == box._get_axis_number(v)
+            assert obj._get_axis_name(v) == box._get_axis_name(v)
+            assert obj._get_block_manager_axis(v) == box._get_block_manager_axis(v)
+
+    def test_flags_identity(self, frame_or_series):
+        obj = Series([1, 2])
+        if frame_or_series is DataFrame:
+            obj = obj.to_frame()
+
+        assert obj.flags is obj.flags
+        obj2 = obj.copy()
+        assert obj2.flags is not obj.flags
diff --git a/pandas/pandas/tests/generic/test_label_or_level_utils.py b/pandas/pandas/tests/generic/test_label_or_level_utils.py
new file mode 100644
index 0000000000000..80c24c647bc11
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_label_or_level_utils.py
@@ -0,0 +1,329 @@
+import pytest
+
+from pandas.core.dtypes.missing import array_equivalent
+
+import pandas as pd
+
+
+# Fixtures
+# ========
+@pytest.fixture
+def df():
+    """DataFrame with columns 'L1', 'L2', and 'L3'"""
+    return pd.DataFrame({"L1": [1, 2, 3], "L2": [11, 12, 13], "L3": ["A", "B", "C"]})
+
+
+@pytest.fixture(params=[[], ["L1"], ["L1", "L2"], ["L1", "L2", "L3"]])
+def df_levels(request, df):
+    """DataFrame with columns or index levels 'L1', 'L2', and 'L3'"""
+    levels = request.param
+
+    if levels:
+        df = df.set_index(levels)
+
+    return df
+
+
+@pytest.fixture
+def df_ambig(df):
+    """DataFrame with levels 'L1' and 'L2' and labels 'L1' and 'L3'"""
+    df = df.set_index(["L1", "L2"])
+
+    df["L1"] = df["L3"]
+
+    return df
+
+
+# Test is label/level reference
+# =============================
+def get_labels_levels(df_levels):
+    expected_labels = list(df_levels.columns)
+    expected_levels = [name for name in df_levels.index.names if name is not None]
+    return expected_labels, expected_levels
+
+
+def assert_label_reference(frame, labels, axis):
+    for label in labels:
+        assert frame._is_label_reference(label, axis=axis)
+        assert not frame._is_level_reference(label, axis=axis)
+        assert frame._is_label_or_level_reference(label, axis=axis)
+
+
+def assert_level_reference(frame, levels, axis):
+    for level in levels:
+        assert frame._is_level_reference(level, axis=axis)
+        assert not frame._is_label_reference(level, axis=axis)
+        assert frame._is_label_or_level_reference(level, axis=axis)
+
+
+# DataFrame
+# ---------
+def test_is_level_or_label_reference_df_simple(df_levels, axis):
+    axis = df_levels._get_axis_number(axis)
+    # Compute expected labels and levels
+    expected_labels, expected_levels = get_labels_levels(df_levels)
+
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_levels = df_levels.T
+
+    # Perform checks
+    assert_level_reference(df_levels, expected_levels, axis=axis)
+    assert_label_reference(df_levels, expected_labels, axis=axis)
+
+
+def test_is_level_reference_df_ambig(df_ambig, axis):
+    axis = df_ambig._get_axis_number(axis)
+
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_ambig = df_ambig.T
+
+    # df has both an on-axis level and off-axis label named L1
+    # Therefore L1 should reference the label, not the level
+    assert_label_reference(df_ambig, ["L1"], axis=axis)
+
+    # df has an on-axis level named L2 and it is not ambiguous
+    # Therefore L2 is an level reference
+    assert_level_reference(df_ambig, ["L2"], axis=axis)
+
+    # df has a column named L3 and it not an level reference
+    assert_label_reference(df_ambig, ["L3"], axis=axis)
+
+
+# Series
+# ------
+def test_is_level_reference_series_simple_axis0(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+    assert_level_reference(s, ["L1"], axis=0)
+    assert not s._is_level_reference("L2")
+
+    # Make series with L1 and L2 as index
+    s = df.set_index(["L1", "L2"]).L3
+    assert_level_reference(s, ["L1", "L2"], axis=0)
+    assert not s._is_level_reference("L3")
+
+
+def test_is_level_reference_series_axis1_error(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+
+    with pytest.raises(ValueError, match="No axis named 1"):
+        s._is_level_reference("L1", axis=1)
+
+
+# Test _check_label_or_level_ambiguity_df
+# =======================================
+
+
+# DataFrame
+# ---------
+def test_check_label_or_level_ambiguity_df(df_ambig, axis):
+    axis = df_ambig._get_axis_number(axis)
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_ambig = df_ambig.T
+        msg = "'L1' is both a column level and an index label"
+
+    else:
+        msg = "'L1' is both an index level and a column label"
+    # df_ambig has both an on-axis level and off-axis label named L1
+    # Therefore, L1 is ambiguous.
+    with pytest.raises(ValueError, match=msg):
+        df_ambig._check_label_or_level_ambiguity("L1", axis=axis)
+
+    # df_ambig has an on-axis level named L2,, and it is not ambiguous.
+    df_ambig._check_label_or_level_ambiguity("L2", axis=axis)
+
+    # df_ambig has an off-axis label named L3, and it is not ambiguous
+    assert not df_ambig._check_label_or_level_ambiguity("L3", axis=axis)
+
+
+# Series
+# ------
+def test_check_label_or_level_ambiguity_series(df):
+    # A series has no columns and therefore references are never ambiguous
+
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+    s._check_label_or_level_ambiguity("L1", axis=0)
+    s._check_label_or_level_ambiguity("L2", axis=0)
+
+    # Make series with L1 and L2 as index
+    s = df.set_index(["L1", "L2"]).L3
+    s._check_label_or_level_ambiguity("L1", axis=0)
+    s._check_label_or_level_ambiguity("L2", axis=0)
+    s._check_label_or_level_ambiguity("L3", axis=0)
+
+
+def test_check_label_or_level_ambiguity_series_axis1_error(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+
+    with pytest.raises(ValueError, match="No axis named 1"):
+        s._check_label_or_level_ambiguity("L1", axis=1)
+
+
+# Test _get_label_or_level_values
+# ===============================
+def assert_label_values(frame, labels, axis):
+    axis = frame._get_axis_number(axis)
+    for label in labels:
+        if axis == 0:
+            expected = frame[label]._values
+        else:
+            expected = frame.loc[label]._values
+
+        result = frame._get_label_or_level_values(label, axis=axis)
+        assert array_equivalent(expected, result)
+
+
+def assert_level_values(frame, levels, axis):
+    axis = frame._get_axis_number(axis)
+    for level in levels:
+        if axis == 0:
+            expected = frame.index.get_level_values(level=level)._values
+        else:
+            expected = frame.columns.get_level_values(level=level)._values
+
+        result = frame._get_label_or_level_values(level, axis=axis)
+        assert array_equivalent(expected, result)
+
+
+# DataFrame
+# ---------
+def test_get_label_or_level_values_df_simple(df_levels, axis):
+    # Compute expected labels and levels
+    expected_labels, expected_levels = get_labels_levels(df_levels)
+
+    axis = df_levels._get_axis_number(axis)
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_levels = df_levels.T
+
+    # Perform checks
+    assert_label_values(df_levels, expected_labels, axis=axis)
+    assert_level_values(df_levels, expected_levels, axis=axis)
+
+
+def test_get_label_or_level_values_df_ambig(df_ambig, axis):
+    axis = df_ambig._get_axis_number(axis)
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_ambig = df_ambig.T
+
+    # df has an on-axis level named L2, and it is not ambiguous.
+    assert_level_values(df_ambig, ["L2"], axis=axis)
+
+    # df has an off-axis label named L3, and it is not ambiguous.
+    assert_label_values(df_ambig, ["L3"], axis=axis)
+
+
+def test_get_label_or_level_values_df_duplabels(df, axis):
+    df = df.set_index(["L1"])
+    df_duplabels = pd.concat([df, df["L2"]], axis=1)
+    axis = df_duplabels._get_axis_number(axis)
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_duplabels = df_duplabels.T
+
+    # df has unambiguous level 'L1'
+    assert_level_values(df_duplabels, ["L1"], axis=axis)
+
+    # df has unique label 'L3'
+    assert_label_values(df_duplabels, ["L3"], axis=axis)
+
+    # df has duplicate labels 'L2'
+    if axis == 0:
+        expected_msg = "The column label 'L2' is not unique"
+    else:
+        expected_msg = "The index label 'L2' is not unique"
+
+    with pytest.raises(ValueError, match=expected_msg):
+        assert_label_values(df_duplabels, ["L2"], axis=axis)
+
+
+# Series
+# ------
+def test_get_label_or_level_values_series_axis0(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+    assert_level_values(s, ["L1"], axis=0)
+
+    # Make series with L1 and L2 as index
+    s = df.set_index(["L1", "L2"]).L3
+    assert_level_values(s, ["L1", "L2"], axis=0)
+
+
+def test_get_label_or_level_values_series_axis1_error(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+
+    with pytest.raises(ValueError, match="No axis named 1"):
+        s._get_label_or_level_values("L1", axis=1)
+
+
+# Test _drop_labels_or_levels
+# ===========================
+def assert_labels_dropped(frame, labels, axis):
+    axis = frame._get_axis_number(axis)
+    for label in labels:
+        df_dropped = frame._drop_labels_or_levels(label, axis=axis)
+
+        if axis == 0:
+            assert label in frame.columns
+            assert label not in df_dropped.columns
+        else:
+            assert label in frame.index
+            assert label not in df_dropped.index
+
+
+def assert_levels_dropped(frame, levels, axis):
+    axis = frame._get_axis_number(axis)
+    for level in levels:
+        df_dropped = frame._drop_labels_or_levels(level, axis=axis)
+
+        if axis == 0:
+            assert level in frame.index.names
+            assert level not in df_dropped.index.names
+        else:
+            assert level in frame.columns.names
+            assert level not in df_dropped.columns.names
+
+
+# DataFrame
+# ---------
+def test_drop_labels_or_levels_df(df_levels, axis):
+    # Compute expected labels and levels
+    expected_labels, expected_levels = get_labels_levels(df_levels)
+
+    axis = df_levels._get_axis_number(axis)
+    # Transpose frame if axis == 1
+    if axis == 1:
+        df_levels = df_levels.T
+
+    # Perform checks
+    assert_labels_dropped(df_levels, expected_labels, axis=axis)
+    assert_levels_dropped(df_levels, expected_levels, axis=axis)
+
+    with pytest.raises(ValueError, match="not valid labels or levels"):
+        df_levels._drop_labels_or_levels("L4", axis=axis)
+
+
+# Series
+# ------
+def test_drop_labels_or_levels_series(df):
+    # Make series with L1 as index
+    s = df.set_index("L1").L2
+    assert_levels_dropped(s, ["L1"], axis=0)
+
+    with pytest.raises(ValueError, match="not valid labels or levels"):
+        s._drop_labels_or_levels("L4", axis=0)
+
+    # Make series with L1 and L2 as index
+    s = df.set_index(["L1", "L2"]).L3
+    assert_levels_dropped(s, ["L1", "L2"], axis=0)
+
+    with pytest.raises(ValueError, match="not valid labels or levels"):
+        s._drop_labels_or_levels("L4", axis=0)
diff --git a/pandas/pandas/tests/generic/test_series.py b/pandas/pandas/tests/generic/test_series.py
new file mode 100644
index 0000000000000..7dcdcd96cce51
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_series.py
@@ -0,0 +1,119 @@
+from operator import methodcaller
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeries:
+    @pytest.mark.parametrize("func", ["rename_axis", "_set_axis_name"])
+    def test_set_axis_name_mi(self, func):
+        ser = Series(
+            [11, 21, 31],
+            index=MultiIndex.from_tuples(
+                [("A", x) for x in ["a", "B", "c"]], names=["l1", "l2"]
+            ),
+        )
+
+        result = methodcaller(func, ["L1", "L2"])(ser)
+        assert ser.index.name is None
+        assert ser.index.names == ["l1", "l2"]
+        assert result.index.name is None
+        assert result.index.names, ["L1", "L2"]
+
+    def test_set_axis_name_raises(self):
+        ser = Series([1])
+        msg = "No axis named 1 for object type Series"
+        with pytest.raises(ValueError, match=msg):
+            ser._set_axis_name(name="a", axis=1)
+
+    def test_get_bool_data_preserve_dtype(self):
+        ser = Series([True, False, True])
+        result = ser._get_bool_data()
+        tm.assert_series_equal(result, ser)
+
+    @pytest.mark.parametrize("data", [np.nan, pd.NaT, True, False])
+    def test_nonzero_single_element_raise_1(self, data):
+        # single item nan to raise
+        series = Series([data])
+
+        msg = "The truth value of a Series is ambiguous"
+        with pytest.raises(ValueError, match=msg):
+            bool(series)
+
+    @pytest.mark.parametrize("data", [(True, True), (False, False)])
+    def test_nonzero_multiple_element_raise(self, data):
+        # multiple bool are still an error
+        msg_err = "The truth value of a Series is ambiguous"
+        series = Series([data])
+        with pytest.raises(ValueError, match=msg_err):
+            bool(series)
+
+    @pytest.mark.parametrize("data", [1, 0, "a", 0.0])
+    def test_nonbool_single_element_raise(self, data):
+        # single non-bool are an error
+        msg_err1 = "The truth value of a Series is ambiguous"
+        series = Series([data])
+        with pytest.raises(ValueError, match=msg_err1):
+            bool(series)
+
+    def test_metadata_propagation_indiv_resample(self):
+        # resample
+        ts = Series(
+            np.random.default_rng(2).random(1000),
+            index=date_range("20130101", periods=1000, freq="s"),
+            name="foo",
+        )
+        result = ts.resample("1min").mean()
+        tm.assert_metadata_equivalent(ts, result)
+
+        result = ts.resample("1min").min()
+        tm.assert_metadata_equivalent(ts, result)
+
+        result = ts.resample("1min").apply(lambda x: x.sum())
+        tm.assert_metadata_equivalent(ts, result)
+
+    def test_metadata_propagation_indiv(self, monkeypatch):
+        # check that the metadata matches up on the resulting ops
+
+        ser = Series(range(3), range(3))
+        ser.name = "foo"
+        ser2 = Series(range(3), range(3))
+        ser2.name = "bar"
+
+        result = ser.T
+        tm.assert_metadata_equivalent(ser, result)
+
+        def finalize(self, other, method=None, **kwargs):
+            for name in self._metadata:
+                if method == "concat" and name == "filename":
+                    value = "+".join(
+                        [
+                            getattr(obj, name)
+                            for obj in other.objs
+                            if getattr(obj, name, None)
+                        ]
+                    )
+                    object.__setattr__(self, name, value)
+                else:
+                    object.__setattr__(self, name, getattr(other, name, None))
+
+            return self
+
+        with monkeypatch.context() as m:
+            m.setattr(Series, "_metadata", ["name", "filename"])
+            m.setattr(Series, "__finalize__", finalize)
+
+            ser.filename = "foo"
+            ser2.filename = "bar"
+
+            result = pd.concat([ser, ser2])
+            assert result.filename == "foo+bar"
+            assert result.name is None
diff --git a/pandas/pandas/tests/generic/test_to_xarray.py b/pandas/pandas/tests/generic/test_to_xarray.py
new file mode 100644
index 0000000000000..9fe9bca8abdc9
--- /dev/null
+++ b/pandas/pandas/tests/generic/test_to_xarray.py
@@ -0,0 +1,130 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+pytest.importorskip("xarray")
+
+
+class TestDataFrameToXArray:
+    @pytest.fixture
+    def df(self):
+        return DataFrame(
+            {
+                "a": list("abcd"),
+                "b": list(range(1, 5)),
+                "c": np.arange(3, 7).astype("u1"),
+                "d": np.arange(4.0, 8.0, dtype="float64"),
+                "e": [True, False, True, False],
+                "f": Categorical(list("abcd")),
+                "g": date_range("20130101", periods=4),
+                "h": date_range("20130101", periods=4, tz="US/Eastern"),
+            }
+        )
+
+    def test_to_xarray_index_types(self, index_flat, df, using_infer_string):
+        index = index_flat
+        # MultiIndex is tested in test_to_xarray_with_multiindex
+        if len(index) == 0:
+            pytest.skip("Test doesn't make sense for empty index")
+
+        from xarray import Dataset
+
+        df.index = index[:4]
+        df.index.name = "foo"
+        df.columns.name = "bar"
+        result = df.to_xarray()
+        assert result.sizes["foo"] == 4
+        assert len(result.coords) == 1
+        assert len(result.data_vars) == 8
+        tm.assert_almost_equal(list(result.coords.keys()), ["foo"])
+        assert isinstance(result, Dataset)
+
+        # idempotency
+        # datetimes w/tz are preserved
+        # column names are lost
+        expected = df.copy()
+        expected["f"] = expected["f"].astype(
+            object if not using_infer_string else "str"
+        )
+        expected.columns.name = None
+        tm.assert_frame_equal(result.to_dataframe(), expected)
+
+    def test_to_xarray_empty(self, df):
+        from xarray import Dataset
+
+        df.index.name = "foo"
+        result = df[0:0].to_xarray()
+        assert result.sizes["foo"] == 0
+        assert isinstance(result, Dataset)
+
+    def test_to_xarray_with_multiindex(self, df, using_infer_string):
+        from xarray import Dataset
+
+        # MultiIndex
+        df.index = MultiIndex.from_product([["a"], range(4)], names=["one", "two"])
+        result = df.to_xarray()
+        assert result.sizes["one"] == 1
+        assert result.sizes["two"] == 4
+        assert len(result.coords) == 2
+        assert len(result.data_vars) == 8
+        tm.assert_almost_equal(list(result.coords.keys()), ["one", "two"])
+        assert isinstance(result, Dataset)
+
+        result = result.to_dataframe()
+        expected = df.copy()
+        expected["f"] = expected["f"].astype(
+            object if not using_infer_string else "str"
+        )
+        expected.columns.name = None
+        tm.assert_frame_equal(result, expected)
+
+
+class TestSeriesToXArray:
+    def test_to_xarray_index_types(self, index_flat):
+        index = index_flat
+        # MultiIndex is tested in test_to_xarray_with_multiindex
+
+        from xarray import DataArray
+
+        ser = Series(range(len(index)), index=index, dtype="int64")
+        ser.index.name = "foo"
+        result = ser.to_xarray()
+        repr(result)
+        assert len(result) == len(index)
+        assert len(result.coords) == 1
+        tm.assert_almost_equal(list(result.coords.keys()), ["foo"])
+        assert isinstance(result, DataArray)
+
+        # idempotency
+        tm.assert_series_equal(result.to_series(), ser)
+
+    def test_to_xarray_empty(self):
+        from xarray import DataArray
+
+        ser = Series([], dtype=object)
+        ser.index.name = "foo"
+        result = ser.to_xarray()
+        assert len(result) == 0
+        assert len(result.coords) == 1
+        tm.assert_almost_equal(list(result.coords.keys()), ["foo"])
+        assert isinstance(result, DataArray)
+
+    def test_to_xarray_with_multiindex(self):
+        from xarray import DataArray
+
+        mi = MultiIndex.from_product([["a", "b"], range(3)], names=["one", "two"])
+        ser = Series(range(6), dtype="int64", index=mi)
+        result = ser.to_xarray()
+        assert len(result) == 2
+        tm.assert_almost_equal(list(result.coords.keys()), ["one", "two"])
+        assert isinstance(result, DataArray)
+        res = result.to_series()
+        tm.assert_series_equal(res, ser)
diff --git a/pandas/pandas/tests/groupby/__init__.py b/pandas/pandas/tests/groupby/__init__.py
new file mode 100644
index 0000000000000..79046cd7ed415
--- /dev/null
+++ b/pandas/pandas/tests/groupby/__init__.py
@@ -0,0 +1,25 @@
+def get_groupby_method_args(name, obj):
+    """
+    Get required arguments for a groupby method.
+
+    When parametrizing a test over groupby methods (e.g. "sum", "mean"),
+    it is often the case that arguments are required for certain methods.
+
+    Parameters
+    ----------
+    name: str
+        Name of the method.
+    obj: Series or DataFrame
+        pandas object that is being grouped.
+
+    Returns
+    -------
+    A tuple of required arguments for the method.
+    """
+    if name in ("nth", "take"):
+        return (0,)
+    if name == "quantile":
+        return (0.5,)
+    if name == "corrwith":
+        return (obj,)
+    return ()
diff --git a/pandas/pandas/tests/groupby/aggregate/__init__.py b/pandas/pandas/tests/groupby/aggregate/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/groupby/aggregate/test_aggregate.py b/pandas/pandas/tests/groupby/aggregate/test_aggregate.py
new file mode 100644
index 0000000000000..b7e6e55739c17
--- /dev/null
+++ b/pandas/pandas/tests/groupby/aggregate/test_aggregate.py
@@ -0,0 +1,1809 @@
+"""
+test .agg behavior / note that .apply is tested generally in test_groupby.py
+"""
+
+import datetime
+import functools
+from functools import partial
+
+import numpy as np
+import pytest
+
+from pandas.errors import SpecificationError
+
+from pandas.core.dtypes.common import is_integer_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.groupby.grouper import Grouping
+
+
+def test_groupby_agg_no_extra_calls():
+    # GH#31760
+    df = DataFrame({"key": ["a", "b", "c", "c"], "value": [1, 2, 3, 4]})
+    gb = df.groupby("key")["value"]
+
+    def dummy_func(x):
+        assert len(x) != 0
+        return x.sum()
+
+    gb.agg(dummy_func)
+
+
+def test_agg_regression1(tsframe):
+    grouped = tsframe.groupby([lambda x: x.year, lambda x: x.month])
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_must_agg(df):
+    grouped = df.groupby("A")["C"]
+
+    msg = "Must produce aggregated value"
+    with pytest.raises(Exception, match=msg):
+        grouped.agg(lambda x: x.describe())
+    with pytest.raises(Exception, match=msg):
+        grouped.agg(lambda x: x.index[:2])
+
+
+def test_agg_ser_multi_key(df):
+    f = lambda x: x.sum()
+    results = df.C.groupby([df.A, df.B]).aggregate(f)
+    expected = df.groupby(["A", "B"]).sum()["C"]
+    tm.assert_series_equal(results, expected)
+
+
+def test_agg_with_missing_values():
+    # GH#58810
+    missing_df = DataFrame(
+        {
+            "nan": [np.nan, np.nan, np.nan, np.nan],
+            "na": [pd.NA, pd.NA, pd.NA, pd.NA],
+            "nat": [pd.NaT, pd.NaT, pd.NaT, pd.NaT],
+            "none": [None, None, None, None],
+            "values": [1, 2, 3, 4],
+        }
+    )
+
+    result = missing_df.agg(x=("nan", "min"), y=("na", "min"), z=("values", "sum"))
+
+    expected = DataFrame(
+        {
+            "nan": [np.nan, np.nan, np.nan],
+            "na": [np.nan, np.nan, np.nan],
+            "values": [np.nan, np.nan, 10.0],
+        },
+        index=["x", "y", "z"],
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_mixed_dtype():
+    # GH 6212
+    expected = DataFrame(
+        {
+            "v1": [5, 5, 7, np.nan, 3, 3, 4, 1],
+            "v2": [55, 55, 77, np.nan, 33, 33, 44, 11],
+        },
+        index=MultiIndex.from_tuples(
+            [
+                (1, 95),
+                (1, 99),
+                (2, 95),
+                (2, 99),
+                ("big", "damp"),
+                ("blue", "dry"),
+                ("red", "red"),
+                ("red", "wet"),
+            ],
+            names=["by1", "by2"],
+        ),
+    )
+
+    df = DataFrame(
+        {
+            "v1": [1, 3, 5, 7, 8, 3, 5, np.nan, 4, 5, 7, 9],
+            "v2": [11, 33, 55, 77, 88, 33, 55, np.nan, 44, 55, 77, 99],
+            "by1": ["red", "blue", 1, 2, np.nan, "big", 1, 2, "red", 1, np.nan, 12],
+            "by2": [
+                "wet",
+                "dry",
+                99,
+                95,
+                np.nan,
+                "damp",
+                95,
+                99,
+                "red",
+                99,
+                np.nan,
+                np.nan,
+            ],
+        }
+    )
+
+    g = df.groupby(["by1", "by2"])
+    result = g[["v1", "v2"]].mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_apply_corner(ts, tsframe):
+    # nothing to group, all NA
+    grouped = ts.groupby(ts * np.nan, group_keys=False)
+    assert ts.dtype == np.float64
+
+    # groupby float64 values results in a float64 Index
+    exp = Series([], dtype=np.float64, index=Index([], dtype=np.float64))
+    tm.assert_series_equal(grouped.sum(), exp)
+    tm.assert_series_equal(grouped.agg("sum"), exp)
+    tm.assert_series_equal(grouped.apply("sum"), exp, check_index_type=False)
+
+    # DataFrame
+    grouped = tsframe.groupby(tsframe["A"] * np.nan, group_keys=False)
+    exp_df = DataFrame(
+        columns=tsframe.columns,
+        dtype=float,
+        index=Index([], name="A", dtype=np.float64),
+    )
+    tm.assert_frame_equal(grouped.sum(), exp_df)
+    tm.assert_frame_equal(grouped.agg("sum"), exp_df)
+
+    res = grouped.apply(np.sum, axis=0)
+    exp_df = exp_df.reset_index(drop=True)
+    tm.assert_frame_equal(res, exp_df)
+
+
+def test_with_na_groups(any_real_numpy_dtype):
+    index = Index(np.arange(10))
+    values = Series(np.ones(10), index, dtype=any_real_numpy_dtype)
+    labels = Series(
+        [np.nan, "foo", "bar", "bar", np.nan, np.nan, "bar", "bar", np.nan, "foo"],
+        index=index,
+    )
+
+    # this SHOULD be an int
+    grouped = values.groupby(labels)
+    agged = grouped.agg(len)
+    expected = Series([4, 2], index=["bar", "foo"])
+
+    tm.assert_series_equal(agged, expected, check_dtype=False)
+
+    # assert issubclass(agged.dtype.type, np.integer)
+
+    # explicitly return a float from my function
+    def f(x):
+        return float(len(x))
+
+    agged = grouped.agg(f)
+    expected = Series([4.0, 2.0], index=["bar", "foo"])
+
+    tm.assert_series_equal(agged, expected)
+
+
+def test_agg_grouping_is_list_tuple(ts):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=pd.date_range("2000-01-01", periods=30, freq="B"),
+    )
+
+    grouped = df.groupby(lambda x: x.year)
+    grouper = grouped._grouper.groupings[0].grouping_vector
+    grouped._grouper.groupings[0] = Grouping(ts.index, list(grouper))
+
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+    grouped._grouper.groupings[0] = Grouping(ts.index, tuple(grouper))
+
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_python_multiindex(multiindex_dataframe_random_data):
+    grouped = multiindex_dataframe_random_data.groupby(["A", "B"])
+
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "groupbyfunc", [lambda x: x.weekday(), [lambda x: x.month, lambda x: x.weekday()]]
+)
+def test_aggregate_str_func(tsframe, groupbyfunc):
+    grouped = tsframe.groupby(groupbyfunc)
+
+    # single series
+    result = grouped["A"].agg("std")
+    expected = grouped["A"].std()
+    tm.assert_series_equal(result, expected)
+
+    # group frame by function name
+    result = grouped.aggregate("var")
+    expected = grouped.var()
+    tm.assert_frame_equal(result, expected)
+
+    # group frame by function dict
+    result = grouped.agg({"A": "var", "B": "std", "C": "mean", "D": "sem"})
+    expected = DataFrame(
+        {
+            "A": grouped["A"].var(),
+            "B": grouped["B"].std(),
+            "C": grouped["C"].mean(),
+            "D": grouped["D"].sem(),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_std_masked_dtype(any_numeric_ea_dtype):
+    # GH#35516
+    df = DataFrame(
+        {
+            "a": [2, 1, 1, 1, 2, 2, 1],
+            "b": Series([pd.NA, 1, 2, 1, 1, 1, 2], dtype="Float64"),
+        }
+    )
+    result = df.groupby("a").std()
+    expected = DataFrame(
+        {"b": [0.57735, 0]}, index=Index([1, 2], name="a"), dtype="Float64"
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_str_with_kwarg_axis_1_raises(df, reduction_func):
+    gb = df.groupby(level=0)
+    msg = f"Operation {reduction_func} does not support axis=1"
+    with pytest.raises(ValueError, match=msg):
+        gb.agg(reduction_func, axis=1)
+
+
+def test_aggregate_item_by_item(df):
+    grouped = df.groupby("A")
+
+    aggfun_0 = lambda ser: ser.size
+    result = grouped.agg(aggfun_0)
+    foosum = (df.A == "foo").sum()
+    barsum = (df.A == "bar").sum()
+    K = len(result.columns)
+
+    # GH5782
+    exp = Series(np.array([foosum] * K), index=list("BCD"), name="foo")
+    tm.assert_series_equal(result.xs("foo"), exp)
+
+    exp = Series(np.array([barsum] * K), index=list("BCD"), name="bar")
+    tm.assert_almost_equal(result.xs("bar"), exp)
+
+    def aggfun_1(ser):
+        return ser.size
+
+    result = DataFrame().groupby(df.A).agg(aggfun_1)
+    assert isinstance(result, DataFrame)
+    assert len(result) == 0
+
+
+def test_wrap_agg_out(three_group):
+    grouped = three_group.groupby(["A", "B"])
+
+    def func(ser):
+        if ser.dtype == object or ser.dtype == "string":
+            raise TypeError("Test error message")
+        return ser.sum()
+
+    with pytest.raises(TypeError, match="Test error message"):
+        grouped.aggregate(func)
+    result = grouped[["D", "E", "F"]].aggregate(func)
+    exp_grouped = three_group.loc[:, ["A", "B", "D", "E", "F"]]
+    expected = exp_grouped.groupby(["A", "B"]).aggregate(func)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_multiple_functions_maintain_order(df):
+    # GH #610
+    funcs = [("mean", np.mean), ("max", np.max), ("min", np.min)]
+    result = df.groupby("A")["C"].agg(funcs)
+    exp_cols = Index(["mean", "max", "min"])
+
+    tm.assert_index_equal(result.columns, exp_cols)
+
+
+def test_series_index_name(df):
+    grouped = df.loc[:, ["C"]].groupby(df["A"])
+    result = grouped.agg(lambda x: x.mean())
+    assert result.index.name == "A"
+
+
+def test_agg_multiple_functions_same_name():
+    # GH 30880
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((1000, 3)),
+        index=pd.date_range("1/1/2012", freq="s", periods=1000),
+        columns=["A", "B", "C"],
+    )
+    result = df.resample("3min").agg(
+        {"A": [partial(np.quantile, q=0.9999), partial(np.quantile, q=0.1111)]}
+    )
+    expected_index = pd.date_range("1/1/2012", freq="3min", periods=6)
+    expected_columns = MultiIndex.from_tuples([("A", "quantile"), ("A", "quantile")])
+    expected_values = np.array(
+        [df.resample("3min").A.quantile(q=q).values for q in [0.9999, 0.1111]]
+    ).T
+    expected = DataFrame(
+        expected_values, columns=expected_columns, index=expected_index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_multiple_functions_same_name_with_ohlc_present():
+    # GH 30880
+    # ohlc expands dimensions, so different test to the above is required.
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((1000, 3)),
+        index=pd.date_range("1/1/2012", freq="s", periods=1000, name="dti"),
+        columns=Index(["A", "B", "C"], name="alpha"),
+    )
+    result = df.resample("3min").agg(
+        {"A": ["ohlc", partial(np.quantile, q=0.9999), partial(np.quantile, q=0.1111)]}
+    )
+    expected_index = pd.date_range("1/1/2012", freq="3min", periods=6, name="dti")
+    expected_columns = MultiIndex.from_tuples(
+        [
+            ("A", "ohlc", "open"),
+            ("A", "ohlc", "high"),
+            ("A", "ohlc", "low"),
+            ("A", "ohlc", "close"),
+            ("A", "quantile", "A"),
+            ("A", "quantile", "A"),
+        ],
+        names=["alpha", None, None],
+    )
+    non_ohlc_expected_values = np.array(
+        [df.resample("3min").A.quantile(q=q).values for q in [0.9999, 0.1111]]
+    ).T
+    expected_values = np.hstack(
+        [df.resample("3min").A.ohlc(), non_ohlc_expected_values]
+    )
+    expected = DataFrame(
+        expected_values, columns=expected_columns, index=expected_index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multiple_functions_tuples_and_non_tuples(df):
+    # #1359
+    # Columns B and C would cause partial failure
+    df = df.drop(columns=["B", "C"])
+
+    funcs = [("foo", "mean"), "std"]
+    ex_funcs = [("foo", "mean"), ("std", "std")]
+
+    result = df.groupby("A")["D"].agg(funcs)
+    expected = df.groupby("A")["D"].agg(ex_funcs)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("A").agg(funcs)
+    expected = df.groupby("A").agg(ex_funcs)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_more_flexible_frame_multi_function(df):
+    grouped = df.groupby("A")
+
+    exmean = grouped.agg({"C": "mean", "D": "mean"})
+    exstd = grouped.agg({"C": "std", "D": "std"})
+
+    expected = concat([exmean, exstd], keys=["mean", "std"], axis=1)
+    expected = expected.swaplevel(0, 1, axis=1).sort_index(level=0, axis=1)
+
+    d = {"C": ["mean", "std"], "D": ["mean", "std"]}
+    result = grouped.aggregate(d)
+
+    tm.assert_frame_equal(result, expected)
+
+    # be careful
+    result = grouped.aggregate({"C": "mean", "D": ["mean", "std"]})
+    expected = grouped.aggregate({"C": "mean", "D": ["mean", "std"]})
+    tm.assert_frame_equal(result, expected)
+
+    def numpymean(x):
+        return np.mean(x)
+
+    def numpystd(x):
+        return np.std(x, ddof=1)
+
+    # this uses column selection & renaming
+    msg = r"nested renamer is not supported"
+    d = {"C": "mean", "D": {"foo": "mean", "bar": "std"}}
+    with pytest.raises(SpecificationError, match=msg):
+        grouped.aggregate(d)
+
+    # But without renaming, these functions are OK
+    d = {"C": ["mean"], "D": [numpymean, numpystd]}
+    grouped.aggregate(d)
+
+
+def test_multi_function_flexible_mix(df):
+    # GH #1268
+    grouped = df.groupby("A")
+
+    # Expected
+    d = {"C": {"foo": "mean", "bar": "std"}, "D": {"sum": "sum"}}
+    # this uses column selection & renaming
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        grouped.aggregate(d)
+
+    # Test 1
+    d = {"C": {"foo": "mean", "bar": "std"}, "D": "sum"}
+    # this uses column selection & renaming
+    with pytest.raises(SpecificationError, match=msg):
+        grouped.aggregate(d)
+
+    # Test 2
+    d = {"C": {"foo": "mean", "bar": "std"}, "D": "sum"}
+    # this uses column selection & renaming
+    with pytest.raises(SpecificationError, match=msg):
+        grouped.aggregate(d)
+
+
+def test_groupby_agg_coercing_bools():
+    # issue 14873
+    dat = DataFrame({"a": [1, 1, 2, 2], "b": [0, 1, 2, 3], "c": [None, None, 1, 1]})
+    gp = dat.groupby("a")
+
+    index = Index([1, 2], name="a")
+
+    result = gp["b"].aggregate(lambda x: (x != 0).all())
+    expected = Series([False, True], index=index, name="b")
+    tm.assert_series_equal(result, expected)
+
+    result = gp["c"].aggregate(lambda x: x.isnull().all())
+    expected = Series([True, False], index=index, name="c")
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_agg_dict_with_getitem():
+    # issue 25471
+    dat = DataFrame({"A": ["A", "A", "B", "B", "B"], "B": [1, 2, 1, 1, 2]})
+    result = dat.groupby("A")[["B"]].agg({"B": "sum"})
+
+    expected = DataFrame({"B": [3, 4]}, index=["A", "B"]).rename_axis("A", axis=0)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_agg_dict_dup_columns():
+    # GH#55006
+    df = DataFrame(
+        [[1, 2, 3, 4], [1, 3, 4, 5], [2, 4, 5, 6]],
+        columns=["a", "b", "c", "c"],
+    )
+    gb = df.groupby("a")
+    result = gb.agg({"b": "sum"})
+    expected = DataFrame({"b": [5, 4]}, index=Index([1, 2], name="a"))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op",
+    [
+        lambda x: x.sum(),
+        lambda x: x.cumsum(),
+        lambda x: x.transform("sum"),
+        lambda x: x.transform("cumsum"),
+        lambda x: x.agg("sum"),
+        lambda x: x.agg("cumsum"),
+    ],
+)
+def test_bool_agg_dtype(op):
+    # GH 7001
+    # Bool sum aggregations result in int
+    df = DataFrame({"a": [1, 1], "b": [False, True]})
+    s = df.set_index("a")["b"]
+
+    result = op(df.groupby("a"))["b"].dtype
+    assert is_integer_dtype(result)
+
+    result = op(s.groupby("a")).dtype
+    assert is_integer_dtype(result)
+
+
+@pytest.mark.parametrize(
+    "keys, agg_index",
+    [
+        (["a"], Index([1], name="a")),
+        (["a", "b"], MultiIndex([[1], [2]], [[0], [0]], names=["a", "b"])),
+    ],
+)
+@pytest.mark.parametrize(
+    "input_dtype", ["bool", "int32", "int64", "float32", "float64"]
+)
+@pytest.mark.parametrize(
+    "result_dtype", ["bool", "int32", "int64", "float32", "float64"]
+)
+@pytest.mark.parametrize("method", ["apply", "aggregate", "transform"])
+def test_callable_result_dtype_frame(
+    keys, agg_index, input_dtype, result_dtype, method
+):
+    # GH 21240
+    df = DataFrame({"a": [1], "b": [2], "c": [True]})
+    df["c"] = df["c"].astype(input_dtype)
+    op = getattr(df.groupby(keys)[["c"]], method)
+    result = op(lambda x: x.astype(result_dtype).iloc[0])
+    expected_index = pd.RangeIndex(0, 1) if method == "transform" else agg_index
+    expected = DataFrame({"c": [df["c"].iloc[0]]}, index=expected_index).astype(
+        result_dtype
+    )
+    if method == "apply":
+        expected.columns.names = [0]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keys, agg_index",
+    [
+        (["a"], Index([1], name="a")),
+        (["a", "b"], MultiIndex([[1], [2]], [[0], [0]], names=["a", "b"])),
+    ],
+)
+@pytest.mark.parametrize("input", [True, 1, 1.0])
+@pytest.mark.parametrize("dtype", [bool, int, float])
+@pytest.mark.parametrize("method", ["apply", "aggregate", "transform"])
+def test_callable_result_dtype_series(keys, agg_index, input, dtype, method):
+    # GH 21240
+    df = DataFrame({"a": [1], "b": [2], "c": [input]})
+    op = getattr(df.groupby(keys)["c"], method)
+    result = op(lambda x: x.astype(dtype).iloc[0])
+    expected_index = pd.RangeIndex(0, 1) if method == "transform" else agg_index
+    expected = Series([df["c"].iloc[0]], index=expected_index, name="c").astype(dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_order_aggregate_multiple_funcs():
+    # GH 25692
+    df = DataFrame({"A": [1, 1, 2, 2], "B": [1, 2, 3, 4]})
+
+    res = df.groupby("A").agg(["sum", "max", "mean", "ohlc", "min"])
+    result = res.columns.levels[1]
+
+    expected = Index(["sum", "max", "mean", "ohlc", "min"])
+
+    tm.assert_index_equal(result, expected)
+
+
+def test_ohlc_ea_dtypes(any_numeric_ea_dtype):
+    # GH#37493
+    df = DataFrame(
+        {"a": [1, 1, 2, 3, 4, 4], "b": [22, 11, pd.NA, 10, 20, pd.NA]},
+        dtype=any_numeric_ea_dtype,
+    )
+    gb = df.groupby("a")
+    result = gb.ohlc()
+    expected = DataFrame(
+        [[22, 22, 11, 11], [pd.NA] * 4, [10] * 4, [20] * 4],
+        columns=MultiIndex.from_product([["b"], ["open", "high", "low", "close"]]),
+        index=Index([1, 2, 3, 4], dtype=any_numeric_ea_dtype, name="a"),
+        dtype=any_numeric_ea_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    gb2 = df.groupby("a", as_index=False)
+    result2 = gb2.ohlc()
+    expected2 = expected.reset_index()
+    tm.assert_frame_equal(result2, expected2)
+
+
+@pytest.mark.parametrize("dtype", [np.int64, np.uint64])
+@pytest.mark.parametrize("how", ["first", "last", "min", "max", "mean", "median"])
+def test_uint64_type_handling(dtype, how):
+    # GH 26310
+    df = DataFrame({"x": 6903052872240755750, "y": [1, 2]})
+    expected = df.groupby("y").agg({"x": how})
+    df.x = df.x.astype(dtype)
+    result = df.groupby("y").agg({"x": how})
+    if how not in ("mean", "median"):
+        # mean and median always result in floats
+        result.x = result.x.astype(np.int64)
+    tm.assert_frame_equal(result, expected, check_exact=True)
+
+
+def test_func_duplicates_raises():
+    # GH28426
+    msg = "Function names"
+    df = DataFrame({"A": [0, 0, 1, 1], "B": [1, 2, 3, 4]})
+    with pytest.raises(SpecificationError, match=msg):
+        df.groupby("A").agg(["min", "min"])
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        pd.CategoricalIndex(list("abc")),
+        pd.interval_range(0, 3),
+        pd.period_range("2020", periods=3, freq="D"),
+        MultiIndex.from_tuples([("a", 0), ("a", 1), ("b", 0)]),
+    ],
+)
+def test_agg_index_has_complex_internals(index):
+    # GH 31223
+    df = DataFrame({"group": [1, 1, 2], "value": [0, 1, 0]}, index=index)
+    result = df.groupby("group").agg({"value": Series.nunique})
+    expected = DataFrame({"group": [1, 2], "value": [2, 1]}).set_index("group")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_split_block():
+    # https://github.com/pandas-dev/pandas/issues/31522
+    df = DataFrame(
+        {
+            "key1": ["a", "a", "b", "b", "a"],
+            "key2": ["one", "two", "one", "two", "one"],
+            "key3": ["three", "three", "three", "six", "six"],
+        }
+    )
+    result = df.groupby("key1").min()
+    expected = DataFrame(
+        {"key2": ["one", "one"], "key3": ["six", "six"]},
+        index=Index(["a", "b"], name="key1"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_split_object_part_datetime():
+    # https://github.com/pandas-dev/pandas/pull/31616
+    df = DataFrame(
+        {
+            "A": pd.date_range("2000", periods=4),
+            "B": ["a", "b", "c", "d"],
+            "C": [1, 2, 3, 4],
+            "D": ["b", "c", "d", "e"],
+            "E": pd.date_range("2000", periods=4),
+            "F": [1, 2, 3, 4],
+        }
+    ).astype(object)
+    result = df.groupby([0, 0, 0, 0]).min()
+    expected = DataFrame(
+        {
+            "A": [pd.Timestamp("2000")],
+            "B": ["a"],
+            "C": [1],
+            "D": ["b"],
+            "E": [pd.Timestamp("2000")],
+            "F": [1],
+        },
+        index=np.array([0]),
+        dtype=object,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+class TestNamedAggregationSeries:
+    def test_series_named_agg(self):
+        df = Series([1, 2, 3, 4])
+        gr = df.groupby([0, 0, 1, 1])
+        result = gr.agg(a="sum", b="min")
+        expected = DataFrame(
+            {"a": [3, 7], "b": [1, 3]}, columns=["a", "b"], index=np.array([0, 1])
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = gr.agg(b="min", a="sum")
+        expected = expected[["b", "a"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_no_args_raises(self):
+        gr = Series([1, 2]).groupby([0, 1])
+        with pytest.raises(TypeError, match="Must provide"):
+            gr.agg()
+
+        # but we do allow this
+        result = gr.agg([])
+        expected = DataFrame(columns=[])
+        tm.assert_frame_equal(result, expected)
+
+    def test_series_named_agg_duplicates_no_raises(self):
+        # GH28426
+        gr = Series([1, 2, 3]).groupby([0, 0, 1])
+        grouped = gr.agg(a="sum", b="sum")
+        expected = DataFrame({"a": [3, 3], "b": [3, 3]}, index=np.array([0, 1]))
+        tm.assert_frame_equal(expected, grouped)
+
+    def test_mangled(self):
+        gr = Series([1, 2, 3]).groupby([0, 0, 1])
+        result = gr.agg(a=lambda x: 0, b=lambda x: 1)
+        expected = DataFrame({"a": [0, 0], "b": [1, 1]}, index=np.array([0, 1]))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "inp",
+        [
+            pd.NamedAgg(column="anything", aggfunc="min"),
+            ("anything", "min"),
+            ["anything", "min"],
+        ],
+    )
+    def test_named_agg_nametuple(self, inp):
+        # GH34422
+        s = Series([1, 1, 2, 2, 3, 3, 4, 5])
+        msg = f"func is expected but received {type(inp).__name__}"
+        with pytest.raises(TypeError, match=msg):
+            s.groupby(s.values).agg(a=inp)
+
+
+class TestNamedAggregationDataFrame:
+    def test_agg_relabel(self):
+        df = DataFrame(
+            {"group": ["a", "a", "b", "b"], "A": [0, 1, 2, 3], "B": [5, 6, 7, 8]}
+        )
+        result = df.groupby("group").agg(a_max=("A", "max"), b_max=("B", "max"))
+        expected = DataFrame(
+            {"a_max": [1, 3], "b_max": [6, 8]},
+            index=Index(["a", "b"], name="group"),
+            columns=["a_max", "b_max"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # order invariance
+        p98 = functools.partial(np.percentile, q=98)
+        result = df.groupby("group").agg(
+            b_min=("B", "min"),
+            a_min=("A", "min"),
+            a_mean=("A", "mean"),
+            a_max=("A", "max"),
+            b_max=("B", "max"),
+            a_98=("A", p98),
+        )
+        expected = DataFrame(
+            {
+                "b_min": [5, 7],
+                "a_min": [0, 2],
+                "a_mean": [0.5, 2.5],
+                "a_max": [1, 3],
+                "b_max": [6, 8],
+                "a_98": [0.98, 2.98],
+            },
+            index=Index(["a", "b"], name="group"),
+            columns=["b_min", "a_min", "a_mean", "a_max", "b_max", "a_98"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_agg_relabel_non_identifier(self):
+        df = DataFrame(
+            {"group": ["a", "a", "b", "b"], "A": [0, 1, 2, 3], "B": [5, 6, 7, 8]}
+        )
+
+        result = df.groupby("group").agg(**{"my col": ("A", "max")})
+        expected = DataFrame({"my col": [1, 3]}, index=Index(["a", "b"], name="group"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_duplicate_no_raises(self):
+        # GH 28426, if use same input function on same column,
+        # no error should raise
+        df = DataFrame({"A": [0, 0, 1, 1], "B": [1, 2, 3, 4]})
+
+        grouped = df.groupby("A").agg(a=("B", "min"), b=("B", "min"))
+        expected = DataFrame({"a": [1, 3], "b": [1, 3]}, index=Index([0, 1], name="A"))
+        tm.assert_frame_equal(grouped, expected)
+
+        quant50 = functools.partial(np.percentile, q=50)
+        quant70 = functools.partial(np.percentile, q=70)
+        quant50.__name__ = "quant50"
+        quant70.__name__ = "quant70"
+
+        test = DataFrame({"col1": ["a", "a", "b", "b", "b"], "col2": [1, 2, 3, 4, 5]})
+
+        grouped = test.groupby("col1").agg(
+            quantile_50=("col2", quant50), quantile_70=("col2", quant70)
+        )
+        expected = DataFrame(
+            {"quantile_50": [1.5, 4.0], "quantile_70": [1.7, 4.4]},
+            index=Index(["a", "b"], name="col1"),
+        )
+        tm.assert_frame_equal(grouped, expected)
+
+    def test_agg_relabel_with_level(self):
+        df = DataFrame(
+            {"A": [0, 0, 1, 1], "B": [1, 2, 3, 4]},
+            index=MultiIndex.from_product([["A", "B"], ["a", "b"]]),
+        )
+        result = df.groupby(level=0).agg(
+            aa=("A", "max"), bb=("A", "min"), cc=("B", "mean")
+        )
+        expected = DataFrame(
+            {"aa": [0, 1], "bb": [0, 1], "cc": [1.5, 3.5]}, index=["A", "B"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_agg_relabel_other_raises(self):
+        df = DataFrame({"A": [0, 0, 1], "B": [1, 2, 3]})
+        grouped = df.groupby("A")
+        match = "Must provide"
+        with pytest.raises(TypeError, match=match):
+            grouped.agg(foo=1)
+
+        with pytest.raises(TypeError, match=match):
+            grouped.agg()
+
+        with pytest.raises(TypeError, match=match):
+            grouped.agg(a=("B", "max"), b=(1, 2, 3))
+
+    def test_missing_raises(self):
+        df = DataFrame({"A": [0, 1], "B": [1, 2]})
+        msg = r"Label\(s\) \['C'\] do not exist"
+        with pytest.raises(KeyError, match=msg):
+            df.groupby("A").agg(c=("C", "sum"))
+
+    def test_agg_namedtuple(self):
+        df = DataFrame({"A": [0, 1], "B": [1, 2]})
+        result = df.groupby("A").agg(
+            b=pd.NamedAgg("B", "sum"), c=pd.NamedAgg(column="B", aggfunc="count")
+        )
+        expected = df.groupby("A").agg(b=("B", "sum"), c=("B", "count"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_mangled(self):
+        df = DataFrame({"A": [0, 1], "B": [1, 2], "C": [3, 4]})
+        result = df.groupby("A").agg(b=("B", lambda x: 0), c=("C", lambda x: 1))
+        expected = DataFrame({"b": [0, 0], "c": [1, 1]}, index=Index([0, 1], name="A"))
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg_col1, agg_col2, agg_col3, agg_result1, agg_result2, agg_result3",
+    [
+        (
+            (("y", "A"), "max"),
+            (("y", "A"), np.mean),
+            (("y", "B"), "mean"),
+            [1, 3],
+            [0.5, 2.5],
+            [5.5, 7.5],
+        ),
+        (
+            (("y", "A"), lambda x: max(x)),
+            (("y", "A"), lambda x: 1),
+            (("y", "B"), np.mean),
+            [1, 3],
+            [1, 1],
+            [5.5, 7.5],
+        ),
+        (
+            pd.NamedAgg(("y", "A"), "max"),
+            pd.NamedAgg(("y", "B"), np.mean),
+            pd.NamedAgg(("y", "A"), lambda x: 1),
+            [1, 3],
+            [5.5, 7.5],
+            [1, 1],
+        ),
+    ],
+)
+def test_agg_relabel_multiindex_column(
+    agg_col1, agg_col2, agg_col3, agg_result1, agg_result2, agg_result3
+):
+    # GH 29422, add tests for multiindex column cases
+    df = DataFrame(
+        {"group": ["a", "a", "b", "b"], "A": [0, 1, 2, 3], "B": [5, 6, 7, 8]}
+    )
+    df.columns = MultiIndex.from_tuples([("x", "group"), ("y", "A"), ("y", "B")])
+    idx = Index(["a", "b"], name=("x", "group"))
+
+    result = df.groupby(("x", "group")).agg(a_max=(("y", "A"), "max"))
+    expected = DataFrame({"a_max": [1, 3]}, index=idx)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(("x", "group")).agg(
+        col_1=agg_col1, col_2=agg_col2, col_3=agg_col3
+    )
+    expected = DataFrame(
+        {"col_1": agg_result1, "col_2": agg_result2, "col_3": agg_result3}, index=idx
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_relabel_multiindex_raises_not_exist():
+    # GH 29422, add test for raises scenario when aggregate column does not exist
+    df = DataFrame(
+        {"group": ["a", "a", "b", "b"], "A": [0, 1, 2, 3], "B": [5, 6, 7, 8]}
+    )
+    df.columns = MultiIndex.from_tuples([("x", "group"), ("y", "A"), ("y", "B")])
+
+    with pytest.raises(KeyError, match="do not exist"):
+        df.groupby(("x", "group")).agg(a=(("Y", "a"), "max"))
+
+
+def test_agg_relabel_multiindex_duplicates():
+    # GH29422, add test for raises scenario when getting duplicates
+    # GH28426, after this change, duplicates should also work if the relabelling is
+    # different
+    df = DataFrame(
+        {"group": ["a", "a", "b", "b"], "A": [0, 1, 2, 3], "B": [5, 6, 7, 8]}
+    )
+    df.columns = MultiIndex.from_tuples([("x", "group"), ("y", "A"), ("y", "B")])
+
+    result = df.groupby(("x", "group")).agg(
+        a=(("y", "A"), "min"), b=(("y", "A"), "min")
+    )
+    idx = Index(["a", "b"], name=("x", "group"))
+    expected = DataFrame({"a": [0, 2], "b": [0, 2]}, index=idx)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kwargs", [{"c": ["min"]}, {"b": [], "c": ["min"]}])
+def test_groupby_aggregate_empty_key(kwargs):
+    # GH: 32580
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 2, 3], "c": [1, 2, 4]})
+    result = df.groupby("a").agg(kwargs)
+    expected = DataFrame(
+        [1, 4],
+        index=Index([1, 2], dtype="int64", name="a"),
+        columns=MultiIndex.from_tuples([["c", "min"]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregate_empty_key_empty_return():
+    # GH: 32580 Check if everything works, when return is empty
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 2, 3], "c": [1, 2, 4]})
+    result = df.groupby("a").agg({"b": []})
+    expected = DataFrame(columns=MultiIndex(levels=[["b"], []], codes=[[], []]))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregate_empty_with_multiindex_frame():
+    # GH 39178
+    df = DataFrame(columns=["a", "b", "c"])
+    result = df.groupby(["a", "b"], group_keys=False).agg(d=("c", list))
+    expected = DataFrame(
+        columns=["d"], index=MultiIndex([[], []], [[], []], names=["a", "b"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_grouby_agg_loses_results_with_as_index_false_relabel():
+    # GH 32240: When the aggregate function relabels column names and
+    # as_index=False is specified, the results are dropped.
+
+    df = DataFrame(
+        {"key": ["x", "y", "z", "x", "y", "z"], "val": [1.0, 0.8, 2.0, 3.0, 3.6, 0.75]}
+    )
+
+    grouped = df.groupby("key", as_index=False)
+    result = grouped.agg(min_val=pd.NamedAgg(column="val", aggfunc="min"))
+    expected = DataFrame({"key": ["x", "y", "z"], "min_val": [1.0, 0.8, 0.75]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_grouby_agg_loses_results_with_as_index_false_relabel_multiindex():
+    # GH 32240: When the aggregate function relabels column names and
+    # as_index=False is specified, the results are dropped. Check if
+    # multiindex is returned in the right order
+
+    df = DataFrame(
+        {
+            "key": ["x", "y", "x", "y", "x", "x"],
+            "key1": ["a", "b", "c", "b", "a", "c"],
+            "val": [1.0, 0.8, 2.0, 3.0, 3.6, 0.75],
+        }
+    )
+
+    grouped = df.groupby(["key", "key1"], as_index=False)
+    result = grouped.agg(min_val=pd.NamedAgg(column="val", aggfunc="min"))
+    expected = DataFrame(
+        {"key": ["x", "x", "y"], "key1": ["a", "c", "b"], "min_val": [1.0, 0.75, 0.8]}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_as_index_agg(df):
+    grouped = df.groupby("A", as_index=False)
+
+    # single-key
+
+    result = grouped[["C", "D"]].agg("mean")
+    expected = grouped.mean(numeric_only=True)
+    tm.assert_frame_equal(result, expected)
+
+    result2 = grouped.agg({"C": "mean", "D": "sum"})
+    expected2 = grouped.mean(numeric_only=True)
+    expected2["D"] = grouped.sum()["D"]
+    tm.assert_frame_equal(result2, expected2)
+
+    grouped = df.groupby("A", as_index=True)
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        grouped["C"].agg({"Q": "sum"})
+
+    # multi-key
+
+    grouped = df.groupby(["A", "B"], as_index=False)
+
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+    result2 = grouped.agg({"C": "mean", "D": "sum"})
+    expected2 = grouped.mean()
+    expected2["D"] = grouped.sum()["D"]
+    tm.assert_frame_equal(result2, expected2)
+
+    expected3 = grouped["C"].sum()
+    expected3 = DataFrame(expected3).rename(columns={"C": "Q"})
+    msg = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        grouped["C"].agg({"Q": "sum"})
+
+    # GH7115 & GH8112 & GH8582
+    df = DataFrame(
+        np.random.default_rng(2).integers(0, 100, (50, 3)),
+        columns=["jim", "joe", "jolie"],
+    )
+    ts = Series(np.random.default_rng(2).integers(5, 10, 50), name="jim")
+
+    gr = df.groupby(ts)
+    gr.nth(0)  # invokes set_selection_from_grouper internally
+
+    for attr in ["mean", "max", "count", "idxmax", "cumsum", "all"]:
+        gr = df.groupby(ts, as_index=False)
+        left = getattr(gr, attr)()
+
+        gr = df.groupby(ts.values, as_index=True)
+        right = getattr(gr, attr)().reset_index(drop=True)
+
+        tm.assert_frame_equal(left, right)
+
+
+@pytest.mark.parametrize(
+    "func", [lambda s: s.mean(), lambda s: np.mean(s), lambda s: np.nanmean(s)]
+)
+def test_multiindex_custom_func(func):
+    # GH 31777
+    data = [[1, 4, 2], [5, 7, 1]]
+    df = DataFrame(
+        data,
+        columns=MultiIndex.from_arrays(
+            [[1, 1, 2], [3, 4, 3]], names=["Sisko", "Janeway"]
+        ),
+    )
+    result = df.groupby(np.array([0, 1])).agg(func)
+    expected_dict = {
+        (1, 3): {0: 1.0, 1: 5.0},
+        (1, 4): {0: 4.0, 1: 7.0},
+        (2, 3): {0: 2.0, 1: 1.0},
+    }
+    expected = DataFrame(expected_dict, index=np.array([0, 1]), columns=df.columns)
+    tm.assert_frame_equal(result, expected)
+
+
+def myfunc(s):
+    return np.percentile(s, q=0.90)
+
+
+@pytest.mark.parametrize("func", [lambda s: np.percentile(s, q=0.90), myfunc])
+def test_lambda_named_agg(func):
+    # see gh-28467
+    animals = DataFrame(
+        {
+            "kind": ["cat", "dog", "cat", "dog"],
+            "height": [9.1, 6.0, 9.5, 34.0],
+            "weight": [7.9, 7.5, 9.9, 198.0],
+        }
+    )
+
+    result = animals.groupby("kind").agg(
+        mean_height=("height", "mean"), perc90=("height", func)
+    )
+    expected = DataFrame(
+        [[9.3, 9.1036], [20.0, 6.252]],
+        columns=["mean_height", "perc90"],
+        index=Index(["cat", "dog"], name="kind"),
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_aggregate_mixed_types():
+    # GH 16916
+    df = DataFrame(
+        data=np.array([0] * 9).reshape(3, 3), columns=list("XYZ"), index=list("abc")
+    )
+    df["grouping"] = ["group 1", "group 1", 2]
+    result = df.groupby("grouping").aggregate(lambda x: x.tolist())
+    expected_data = [[[0], [0], [0]], [[0, 0], [0, 0], [0, 0]]]
+    expected = DataFrame(
+        expected_data,
+        index=Index([2, "group 1"], dtype="object", name="grouping"),
+        columns=Index(["X", "Y", "Z"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.xfail(reason="Not implemented;see GH 31256")
+def test_aggregate_udf_na_extension_type():
+    # https://github.com/pandas-dev/pandas/pull/31359
+    # This is currently failing to cast back to Int64Dtype.
+    # The presence of the NA causes two problems
+    # 1. NA is not an instance of Int64Dtype.type (numpy.int64)
+    # 2. The presence of an NA forces object type, so the non-NA values is
+    #    a Python int rather than a NumPy int64. Python ints aren't
+    #    instances of numpy.int64.
+    def aggfunc(x):
+        if all(x > 2):
+            return 1
+        else:
+            return pd.NA
+
+    df = DataFrame({"A": pd.array([1, 2, 3])})
+    result = df.groupby([1, 1, 2]).agg(aggfunc)
+    expected = DataFrame({"A": pd.array([1, pd.NA], dtype="Int64")}, index=[1, 2])
+    tm.assert_frame_equal(result, expected)
+
+
+class TestLambdaMangling:
+    def test_basic(self):
+        df = DataFrame({"A": [0, 0, 1, 1], "B": [1, 2, 3, 4]})
+        result = df.groupby("A").agg({"B": [lambda x: 0, lambda x: 1]})
+
+        expected = DataFrame(
+            {("B", "<lambda_0>"): [0, 0], ("B", "<lambda_1>"): [1, 1]},
+            index=Index([0, 1], name="A"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_mangle_series_groupby(self):
+        gr = Series([1, 2, 3, 4]).groupby([0, 0, 1, 1])
+        result = gr.agg([lambda x: 0, lambda x: 1])
+        exp_data = {"<lambda_0>": [0, 0], "<lambda_1>": [1, 1]}
+        expected = DataFrame(exp_data, index=np.array([0, 1]))
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.xfail(reason="GH-26611. kwargs for multi-agg.")
+    def test_with_kwargs(self):
+        f1 = lambda x, y, b=1: x.sum() + y + b
+        f2 = lambda x, y, b=2: x.sum() + y * b
+        result = Series([1, 2]).groupby([0, 0]).agg([f1, f2], 0)
+        expected = DataFrame({"<lambda_0>": [4], "<lambda_1>": [6]})
+        tm.assert_frame_equal(result, expected)
+
+        result = Series([1, 2]).groupby([0, 0]).agg([f1, f2], 0, b=10)
+        expected = DataFrame({"<lambda_0>": [13], "<lambda_1>": [30]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_agg_with_one_lambda(self):
+        # GH 25719, write tests for DataFrameGroupby.agg with only one lambda
+        df = DataFrame(
+            {
+                "kind": ["cat", "dog", "cat", "dog"],
+                "height": [9.1, 6.0, 9.5, 34.0],
+                "weight": [7.9, 7.5, 9.9, 198.0],
+            }
+        )
+
+        columns = ["height_sqr_min", "height_max", "weight_max"]
+        expected = DataFrame(
+            {
+                "height_sqr_min": [82.81, 36.00],
+                "height_max": [9.5, 34.0],
+                "weight_max": [9.9, 198.0],
+            },
+            index=Index(["cat", "dog"], name="kind"),
+            columns=columns,
+        )
+
+        # check pd.NameAgg case
+        result1 = df.groupby(by="kind").agg(
+            height_sqr_min=pd.NamedAgg(column="height", aggfunc=lambda x: np.min(x**2)),
+            height_max=pd.NamedAgg(column="height", aggfunc="max"),
+            weight_max=pd.NamedAgg(column="weight", aggfunc="max"),
+        )
+        tm.assert_frame_equal(result1, expected)
+
+        # check agg(key=(col, aggfunc)) case
+        result2 = df.groupby(by="kind").agg(
+            height_sqr_min=("height", lambda x: np.min(x**2)),
+            height_max=("height", "max"),
+            weight_max=("weight", "max"),
+        )
+        tm.assert_frame_equal(result2, expected)
+
+    def test_agg_multiple_lambda(self):
+        # GH25719, test for DataFrameGroupby.agg with multiple lambdas
+        # with mixed aggfunc
+        df = DataFrame(
+            {
+                "kind": ["cat", "dog", "cat", "dog"],
+                "height": [9.1, 6.0, 9.5, 34.0],
+                "weight": [7.9, 7.5, 9.9, 198.0],
+            }
+        )
+        columns = [
+            "height_sqr_min",
+            "height_max",
+            "weight_max",
+            "height_max_2",
+            "weight_min",
+        ]
+        expected = DataFrame(
+            {
+                "height_sqr_min": [82.81, 36.00],
+                "height_max": [9.5, 34.0],
+                "weight_max": [9.9, 198.0],
+                "height_max_2": [9.5, 34.0],
+                "weight_min": [7.9, 7.5],
+            },
+            index=Index(["cat", "dog"], name="kind"),
+            columns=columns,
+        )
+
+        # check agg(key=(col, aggfunc)) case
+        result1 = df.groupby(by="kind").agg(
+            height_sqr_min=("height", lambda x: np.min(x**2)),
+            height_max=("height", "max"),
+            weight_max=("weight", "max"),
+            height_max_2=("height", lambda x: np.max(x)),
+            weight_min=("weight", lambda x: np.min(x)),
+        )
+        tm.assert_frame_equal(result1, expected)
+
+        # check pd.NamedAgg case
+        result2 = df.groupby(by="kind").agg(
+            height_sqr_min=pd.NamedAgg(column="height", aggfunc=lambda x: np.min(x**2)),
+            height_max=pd.NamedAgg(column="height", aggfunc="max"),
+            weight_max=pd.NamedAgg(column="weight", aggfunc="max"),
+            height_max_2=pd.NamedAgg(column="height", aggfunc=lambda x: np.max(x)),
+            weight_min=pd.NamedAgg(column="weight", aggfunc=lambda x: np.min(x)),
+        )
+        tm.assert_frame_equal(result2, expected)
+
+
+def test_pass_args_kwargs_duplicate_columns(tsframe, as_index):
+    # go through _aggregate_frame with self.axis == 0 and duplicate columns
+    tsframe.columns = ["A", "B", "A", "C"]
+    gb = tsframe.groupby(lambda x: x.month, as_index=as_index)
+
+    res = gb.agg(np.percentile, 80, axis=0)
+
+    ex_data = {
+        1: tsframe[tsframe.index.month == 1].quantile(0.8),
+        2: tsframe[tsframe.index.month == 2].quantile(0.8),
+    }
+    expected = DataFrame(ex_data).T
+    if not as_index:
+        expected.insert(0, "index", [1, 2])
+        expected.index = Index(range(2))
+
+    tm.assert_frame_equal(res, expected)
+
+
+def test_groupby_get_by_index():
+    # GH 33439
+    df = DataFrame({"A": ["S", "W", "W"], "B": [1.0, 1.0, 2.0]})
+    res = df.groupby("A").agg({"B": lambda x: x.get(x.index[-1])})
+    expected = DataFrame({"A": ["S", "W"], "B": [1.0, 2.0]}).set_index("A")
+    tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "grp_col_dict, exp_data",
+    [
+        ({"nr": "min", "cat_ord": "min"}, {"nr": [1, 5], "cat_ord": ["a", "c"]}),
+        ({"cat_ord": "min"}, {"cat_ord": ["a", "c"]}),
+        ({"nr": "min"}, {"nr": [1, 5]}),
+    ],
+)
+def test_groupby_single_agg_cat_cols(grp_col_dict, exp_data):
+    # test single aggregations on ordered categorical cols GHGH27800
+
+    # create the result dataframe
+    input_df = DataFrame(
+        {
+            "nr": [1, 2, 3, 4, 5, 6, 7, 8],
+            "cat_ord": list("aabbccdd"),
+            "cat": list("aaaabbbb"),
+        }
+    )
+
+    input_df = input_df.astype({"cat": "category", "cat_ord": "category"})
+    input_df["cat_ord"] = input_df["cat_ord"].cat.as_ordered()
+    result_df = input_df.groupby("cat", observed=False).agg(grp_col_dict)
+
+    # create expected dataframe
+    cat_index = pd.CategoricalIndex(
+        ["a", "b"], categories=["a", "b"], ordered=False, name="cat", dtype="category"
+    )
+
+    expected_df = DataFrame(data=exp_data, index=cat_index)
+
+    if "cat_ord" in expected_df:
+        # ordered categorical columns should be preserved
+        dtype = input_df["cat_ord"].dtype
+        expected_df["cat_ord"] = expected_df["cat_ord"].astype(dtype)
+
+    tm.assert_frame_equal(result_df, expected_df)
+
+
+@pytest.mark.parametrize(
+    "grp_col_dict, exp_data",
+    [
+        ({"nr": ["min", "max"], "cat_ord": "min"}, [(1, 4, "a"), (5, 8, "c")]),
+        ({"nr": "min", "cat_ord": ["min", "max"]}, [(1, "a", "b"), (5, "c", "d")]),
+        ({"cat_ord": ["min", "max"]}, [("a", "b"), ("c", "d")]),
+    ],
+)
+def test_groupby_combined_aggs_cat_cols(grp_col_dict, exp_data):
+    # test combined aggregations on ordered categorical cols GH27800
+
+    # create the result dataframe
+    input_df = DataFrame(
+        {
+            "nr": [1, 2, 3, 4, 5, 6, 7, 8],
+            "cat_ord": list("aabbccdd"),
+            "cat": list("aaaabbbb"),
+        }
+    )
+
+    input_df = input_df.astype({"cat": "category", "cat_ord": "category"})
+    input_df["cat_ord"] = input_df["cat_ord"].cat.as_ordered()
+    result_df = input_df.groupby("cat", observed=False).agg(grp_col_dict)
+
+    # create expected dataframe
+    cat_index = pd.CategoricalIndex(
+        ["a", "b"], categories=["a", "b"], ordered=False, name="cat", dtype="category"
+    )
+
+    # unpack the grp_col_dict to create the multi-index tuple
+    # this tuple will be used to create the expected dataframe index
+    multi_index_list = []
+    for k, v in grp_col_dict.items():
+        if isinstance(v, list):
+            multi_index_list.extend([k, value] for value in v)
+        else:
+            multi_index_list.append([k, v])
+    multi_index = MultiIndex.from_tuples(tuple(multi_index_list))
+
+    expected_df = DataFrame(data=exp_data, columns=multi_index, index=cat_index)
+    for col in expected_df.columns:
+        if isinstance(col, tuple) and "cat_ord" in col:
+            # ordered categorical should be preserved
+            expected_df[col] = expected_df[col].astype(input_df["cat_ord"].dtype)
+
+    tm.assert_frame_equal(result_df, expected_df)
+
+
+def test_nonagg_agg():
+    # GH 35490 - Single/Multiple agg of non-agg function give same results
+    # TODO: agg should raise for functions that don't aggregate
+    df = DataFrame({"a": [1, 1, 2, 2], "b": [1, 2, 2, 1]})
+    g = df.groupby("a")
+
+    result = g.agg(["cumsum"])
+    result.columns = result.columns.droplevel(-1)
+    expected = g.agg("cumsum")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_aggregate_datetime_objects():
+    # https://github.com/pandas-dev/pandas/issues/36003
+    # ensure we don't raise an error but keep object dtype for out-of-bounds
+    # datetimes
+    df = DataFrame(
+        {
+            "A": ["X", "Y"],
+            "B": [
+                datetime.datetime(2005, 1, 1, 10, 30, 23, 540000),
+                datetime.datetime(3005, 1, 1, 10, 30, 23, 540000),
+            ],
+        }
+    )
+    result = df.groupby("A").B.max()
+    expected = df.set_index("A")["B"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_index_object_dtype():
+    # GH 40014
+    df = DataFrame({"c0": ["x", "x", "x"], "c1": ["x", "x", "y"], "p": [0, 1, 2]})
+    df.index = df.index.astype("O")
+    grouped = df.groupby(["c0", "c1"])
+    res = grouped.p.agg(lambda x: all(x > 0))
+    # Check that providing a user-defined function in agg()
+    # produces the correct index shape when using an object-typed index.
+    expected_index = MultiIndex.from_tuples(
+        [("x", "x"), ("x", "y")], names=("c0", "c1")
+    )
+    expected = Series([False, True], index=expected_index, name="p")
+    tm.assert_series_equal(res, expected)
+
+
+def test_timeseries_groupby_agg():
+    # GH#43290
+
+    def func(ser):
+        if ser.isna().all():
+            return None
+        return np.sum(ser)
+
+    df = DataFrame([1.0], index=[pd.Timestamp("2018-01-16 00:00:00+00:00")])
+    res = df.groupby(lambda x: 1).agg(func)
+
+    expected = DataFrame([[1.0]], index=[1])
+    tm.assert_frame_equal(res, expected)
+
+
+def test_groupby_agg_precision(any_real_numeric_dtype):
+    if any_real_numeric_dtype in tm.ALL_INT_NUMPY_DTYPES:
+        max_value = np.iinfo(any_real_numeric_dtype).max
+    if any_real_numeric_dtype in tm.FLOAT_NUMPY_DTYPES:
+        max_value = np.finfo(any_real_numeric_dtype).max
+    if any_real_numeric_dtype in tm.FLOAT_EA_DTYPES:
+        max_value = np.finfo(any_real_numeric_dtype.lower()).max
+    if any_real_numeric_dtype in tm.ALL_INT_EA_DTYPES:
+        max_value = np.iinfo(any_real_numeric_dtype.lower()).max
+
+    df = DataFrame(
+        {
+            "key1": ["a"],
+            "key2": ["b"],
+            "key3": pd.array([max_value], dtype=any_real_numeric_dtype),
+        }
+    )
+    arrays = [["a"], ["b"]]
+    index = MultiIndex.from_arrays(arrays, names=("key1", "key2"))
+
+    expected = DataFrame(
+        {"key3": pd.array([max_value], dtype=any_real_numeric_dtype)}, index=index
+    )
+    result = df.groupby(["key1", "key2"]).agg(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregate_directory(reduction_func):
+    # GH#32793
+    if reduction_func in ["corrwith", "nth"]:
+        return None
+
+    obj = DataFrame([[0, 1], [0, np.nan]])
+
+    result_reduced_series = obj.groupby(0).agg(reduction_func)
+    result_reduced_frame = obj.groupby(0).agg({1: reduction_func})
+
+    if reduction_func in ["size", "ngroup"]:
+        # names are different: None / 1
+        tm.assert_series_equal(
+            result_reduced_series, result_reduced_frame[1], check_names=False
+        )
+    else:
+        tm.assert_frame_equal(result_reduced_series, result_reduced_frame)
+        tm.assert_series_equal(
+            result_reduced_series.dtypes, result_reduced_frame.dtypes
+        )
+
+
+def test_group_mean_timedelta_nat():
+    # GH43132
+    data = Series(["1 day", "3 days", "NaT"], dtype="timedelta64[ns]")
+    expected = Series(["2 days"], dtype="timedelta64[ns]", index=np.array([0]))
+
+    result = data.groupby([0, 0, 0]).mean()
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "input_data, expected_output",
+    [
+        (  # no timezone
+            ["2021-01-01T00:00", "NaT", "2021-01-01T02:00"],
+            ["2021-01-01T01:00"],
+        ),
+        (  # timezone
+            ["2021-01-01T00:00-0100", "NaT", "2021-01-01T02:00-0100"],
+            ["2021-01-01T01:00-0100"],
+        ),
+    ],
+)
+def test_group_mean_datetime64_nat(input_data, expected_output):
+    # GH43132
+    data = to_datetime(Series(input_data))
+    expected = to_datetime(Series(expected_output, index=np.array([0])))
+
+    result = data.groupby([0, 0, 0]).mean()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, output", [("mean", [8 + 18j, 10 + 22j]), ("sum", [40 + 90j, 50 + 110j])]
+)
+def test_groupby_complex(func, output):
+    # GH#43701
+    data = Series(np.arange(20).reshape(10, 2).dot([1, 2j]))
+    result = data.groupby(data.index % 2).agg(func)
+    expected = Series(output)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max", "var"])
+def test_groupby_complex_raises(func):
+    # GH#43701
+    data = Series(np.arange(20).reshape(10, 2).dot([1, 2j]))
+    msg = "No matching signature found"
+    with pytest.raises(TypeError, match=msg):
+        data.groupby(data.index % 2).agg(func)
+
+
+@pytest.mark.parametrize(
+    "test, constant",
+    [
+        ([[20, "A"], [20, "B"], [10, "C"]], {0: [10, 20], 1: ["C", ["A", "B"]]}),
+        ([[20, "A"], [20, "B"], [30, "C"]], {0: [20, 30], 1: [["A", "B"], "C"]}),
+        ([["a", 1], ["a", 1], ["b", 2], ["b", 3]], {0: ["a", "b"], 1: [1, [2, 3]]}),
+        pytest.param(
+            [["a", 1], ["a", 2], ["b", 3], ["b", 3]],
+            {0: ["a", "b"], 1: [[1, 2], 3]},
+            marks=pytest.mark.xfail,
+        ),
+    ],
+)
+def test_agg_of_mode_list(test, constant):
+    # GH#25581
+    df1 = DataFrame(test)
+    result = df1.groupby(0).agg(Series.mode)
+    # Mode usually only returns 1 value, but can return a list in the case of a tie.
+
+    expected = DataFrame(constant)
+    expected = expected.set_index(0)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dataframe_groupy_agg_list_like_func_with_args():
+    # GH#50624
+    df = DataFrame({"x": [1, 2, 3], "y": ["a", "b", "c"]})
+    gb = df.groupby("y")
+
+    def foo1(x, a=1, c=0):
+        return x.sum() + a + c
+
+    def foo2(x, b=2, c=0):
+        return x.sum() + b + c
+
+    msg = r"foo1\(\) got an unexpected keyword argument 'b'"
+    with pytest.raises(TypeError, match=msg):
+        gb.agg([foo1, foo2], 3, b=3, c=4)
+
+    result = gb.agg([foo1, foo2], 3, c=4)
+    expected = DataFrame(
+        [[8, 8], [9, 9], [10, 10]],
+        index=Index(["a", "b", "c"], name="y"),
+        columns=MultiIndex.from_tuples([("x", "foo1"), ("x", "foo2")]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_series_groupy_agg_list_like_func_with_args():
+    # GH#50624
+    s = Series([1, 2, 3])
+    sgb = s.groupby(s)
+
+    def foo1(x, a=1, c=0):
+        return x.sum() + a + c
+
+    def foo2(x, b=2, c=0):
+        return x.sum() + b + c
+
+    msg = r"foo1\(\) got an unexpected keyword argument 'b'"
+    with pytest.raises(TypeError, match=msg):
+        sgb.agg([foo1, foo2], 3, b=3, c=4)
+
+    result = sgb.agg([foo1, foo2], 3, c=4)
+    expected = DataFrame(
+        [[8, 8], [9, 9], [10, 10]], index=Index([1, 2, 3]), columns=["foo1", "foo2"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_groupings_selection():
+    # GH#51186 - a selected grouping should be in the output of agg
+    df = DataFrame({"a": [1, 1, 2], "b": [3, 3, 4], "c": [5, 6, 7]})
+    gb = df.groupby(["a", "b"])
+    selected_gb = gb[["b", "c"]]
+    result = selected_gb.agg(lambda x: x.sum())
+    index = MultiIndex(
+        levels=[[1, 2], [3, 4]], codes=[[0, 1], [0, 1]], names=["a", "b"]
+    )
+    expected = DataFrame({"b": [6, 4], "c": [11, 7]}, index=index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_multiple_with_as_index_false_subset_to_a_single_column():
+    # GH#50724
+    df = DataFrame({"a": [1, 1, 2], "b": [3, 4, 5]})
+    gb = df.groupby("a", as_index=False)["b"]
+    result = gb.agg(["sum", "mean"])
+    expected = DataFrame({"a": [1, 2], "sum": [7, 5], "mean": [3.5, 5.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_with_as_index_false_with_list():
+    # GH#52849
+    df = DataFrame({"a1": [0, 0, 1], "a2": [2, 3, 3], "b": [4, 5, 6]})
+    gb = df.groupby(by=["a1", "a2"], as_index=False)
+    result = gb.agg(["sum"])
+
+    expected = DataFrame(
+        data=[[0, 2, 4], [0, 3, 5], [1, 3, 6]],
+        columns=MultiIndex.from_tuples([("a1", ""), ("a2", ""), ("b", "sum")]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_agg_extension_timedelta_cumsum_with_named_aggregation():
+    # GH#41720
+    expected = DataFrame(
+        {
+            "td": {
+                0: pd.Timedelta("0 days 01:00:00"),
+                1: pd.Timedelta("0 days 01:15:00"),
+                2: pd.Timedelta("0 days 01:15:00"),
+            }
+        }
+    )
+    df = DataFrame(
+        {
+            "td": Series(
+                ["0 days 01:00:00", "0 days 00:15:00", "0 days 01:15:00"],
+                dtype="timedelta64[ns]",
+            ),
+            "grps": ["a", "a", "b"],
+        }
+    )
+    gb = df.groupby("grps")
+    result = gb.agg(td=("td", "cumsum"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_empty_group():
+    # https://github.com/pandas-dev/pandas/issues/18869
+    def func(x):
+        if len(x) == 0:
+            raise ValueError("length must not be 0")
+        return len(x)
+
+    df = DataFrame(
+        {"A": pd.Categorical(["a", "a"], categories=["a", "b", "c"]), "B": [1, 1]}
+    )
+    msg = "length must not be 0"
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("A", observed=False).agg(func)
+
+
+def test_groupby_aggregation_duplicate_columns_single_dict_value():
+    # GH#55041
+    df = DataFrame(
+        [[1, 2, 3, 4], [1, 3, 4, 5], [2, 4, 5, 6]],
+        columns=["a", "b", "c", "c"],
+    )
+    gb = df.groupby("a")
+    result = gb.agg({"c": "sum"})
+
+    expected = DataFrame(
+        [[7, 9], [5, 6]], columns=["c", "c"], index=Index([1, 2], name="a")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_duplicate_columns_multiple_dict_values():
+    # GH#55041
+    df = DataFrame(
+        [[1, 2, 3, 4], [1, 3, 4, 5], [2, 4, 5, 6]],
+        columns=["a", "b", "c", "c"],
+    )
+    gb = df.groupby("a")
+    result = gb.agg({"c": ["sum", "min", "max", "min"]})
+
+    expected = DataFrame(
+        [[7, 3, 4, 3, 9, 4, 5, 4], [5, 5, 5, 5, 6, 6, 6, 6]],
+        columns=MultiIndex(
+            levels=[["c"], ["sum", "min", "max"]],
+            codes=[[0, 0, 0, 0, 0, 0, 0, 0], [0, 1, 2, 1, 0, 1, 2, 1]],
+        ),
+        index=Index([1, 2], name="a"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_duplicate_columns_some_empty_result():
+    # GH#55041
+    df = DataFrame(
+        [
+            [1, 9843, 43, 54, 7867],
+            [2, 940, 9, -34, 44],
+            [1, -34, -546, -549358, 0],
+            [2, 244, -33, -100, 44],
+        ],
+        columns=["a", "b", "b", "c", "c"],
+    )
+    gb = df.groupby("a")
+    result = gb.agg({"b": [], "c": ["var"]})
+
+    expected = DataFrame(
+        [[1.509268e11, 30944844.5], [2.178000e03, 0.0]],
+        columns=MultiIndex(levels=[["c"], ["var"]], codes=[[0, 0], [0, 0]]),
+        index=Index([1, 2], name="a"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_multi_index_duplicate_columns():
+    # GH#55041
+    df = DataFrame(
+        [
+            [1, -9843, 43, 54, 7867],
+            [2, 940, 9, -34, 44],
+            [1, -34, 546, -549358, 0],
+            [2, 244, -33, -100, 44],
+        ],
+        columns=MultiIndex(
+            levels=[["level1.1", "level1.2"], ["level2.1", "level2.2"]],
+            codes=[[0, 0, 0, 1, 1], [0, 1, 1, 0, 1]],
+        ),
+        index=MultiIndex(
+            levels=[["level1.1", "level1.2"], ["level2.1", "level2.2"]],
+            codes=[[0, 0, 0, 1], [0, 1, 1, 0]],
+        ),
+    )
+    gb = df.groupby(level=0)
+    result = gb.agg({("level1.1", "level2.2"): "min"})
+
+    expected = DataFrame(
+        [[-9843, 9], [244, -33]],
+        columns=MultiIndex(levels=[["level1.1"], ["level2.2"]], codes=[[0, 0], [0, 0]]),
+        index=Index(["level1.1", "level1.2"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_func_list_multi_index_duplicate_columns():
+    # GH#55041
+    df = DataFrame(
+        [
+            [1, -9843, 43, 54, 7867],
+            [2, 940, 9, -34, 44],
+            [1, -34, 546, -549358, 0],
+            [2, 244, -33, -100, 44],
+        ],
+        columns=MultiIndex(
+            levels=[["level1.1", "level1.2"], ["level2.1", "level2.2"]],
+            codes=[[0, 0, 0, 1, 1], [0, 1, 1, 0, 1]],
+        ),
+        index=MultiIndex(
+            levels=[["level1.1", "level1.2"], ["level2.1", "level2.2"]],
+            codes=[[0, 0, 0, 1], [0, 1, 1, 0]],
+        ),
+    )
+    gb = df.groupby(level=0)
+    result = gb.agg({("level1.1", "level2.2"): ["min", "max"]})
+
+    expected = DataFrame(
+        [[-9843, 940, 9, 546], [244, 244, -33, -33]],
+        columns=MultiIndex(
+            levels=[["level1.1"], ["level2.2"], ["min", "max"]],
+            codes=[[0, 0, 0, 0], [0, 0, 0, 0], [0, 1, 0, 1]],
+        ),
+        index=Index(["level1.1", "level1.2"]),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/aggregate/test_cython.py b/pandas/pandas/tests/groupby/aggregate/test_cython.py
new file mode 100644
index 0000000000000..a706ea795a0e2
--- /dev/null
+++ b/pandas/pandas/tests/groupby/aggregate/test_cython.py
@@ -0,0 +1,415 @@
+"""
+test cython .agg behavior
+"""
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_integer_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    Timedelta,
+    Timestamp,
+    bdate_range,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "op_name",
+    [
+        "count",
+        "sum",
+        "std",
+        "var",
+        "sem",
+        "mean",
+        pytest.param(
+            "median",
+            # ignore mean of empty slice
+            # and all-NaN
+            marks=[pytest.mark.filterwarnings("ignore::RuntimeWarning")],
+        ),
+        "prod",
+        "min",
+        "max",
+    ],
+)
+def test_cythonized_aggers(op_name):
+    data = {
+        "A": [0, 0, 0, 0, 1, 1, 1, 1, 1, 1.0, np.nan, np.nan],
+        "B": ["A", "B"] * 6,
+        "C": np.random.default_rng(2).standard_normal(12),
+    }
+    df = DataFrame(data)
+    df.loc[2:10:2, "C"] = np.nan
+
+    op = lambda x: getattr(x, op_name)()
+
+    # single column
+    grouped = df.drop(["B"], axis=1).groupby("A")
+    exp = {cat: op(group["C"]) for cat, group in grouped}
+    exp = DataFrame({"C": exp})
+    exp.index.name = "A"
+    result = op(grouped)
+    tm.assert_frame_equal(result, exp)
+
+    # multiple columns
+    grouped = df.groupby(["A", "B"])
+    expd = {}
+    for (cat1, cat2), group in grouped:
+        expd.setdefault(cat1, {})[cat2] = op(group["C"])
+    exp = DataFrame(expd).T.stack()
+    exp.index.names = ["A", "B"]
+    exp.name = "C"
+
+    result = op(grouped)["C"]
+    if op_name in ["sum", "prod"]:
+        tm.assert_series_equal(result, exp)
+
+
+def test_cython_agg_boolean():
+    frame = DataFrame(
+        {
+            "a": np.random.default_rng(2).integers(0, 5, 50),
+            "b": np.random.default_rng(2).integers(0, 2, 50).astype("bool"),
+        }
+    )
+    result = frame.groupby("a")["b"].mean()
+    # GH#53425
+    expected = frame.groupby("a")["b"].agg(np.mean)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_cython_agg_nothing_to_agg():
+    frame = DataFrame(
+        {"a": np.random.default_rng(2).integers(0, 5, 50), "b": ["foo", "bar"] * 25}
+    )
+
+    msg = "Cannot use numeric_only=True with SeriesGroupBy.mean and non-numeric dtypes"
+    with pytest.raises(TypeError, match=msg):
+        frame.groupby("a")["b"].mean(numeric_only=True)
+
+    frame = DataFrame(
+        {"a": np.random.default_rng(2).integers(0, 5, 50), "b": ["foo", "bar"] * 25}
+    )
+
+    result = frame[["b"]].groupby(frame["a"]).mean(numeric_only=True)
+    expected = DataFrame(
+        [],
+        index=frame["a"].sort_values().drop_duplicates(),
+        columns=Index([], dtype="str"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cython_agg_nothing_to_agg_with_dates():
+    frame = DataFrame(
+        {
+            "a": np.random.default_rng(2).integers(0, 5, 50),
+            "b": ["foo", "bar"] * 25,
+            "dates": pd.date_range("now", periods=50, freq="min"),
+        }
+    )
+    msg = "Cannot use numeric_only=True with SeriesGroupBy.mean and non-numeric dtypes"
+    with pytest.raises(TypeError, match=msg):
+        frame.groupby("b").dates.mean(numeric_only=True)
+
+
+def test_cython_agg_return_dict():
+    # GH 16741
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+
+    ts = df.groupby("A")["B"].agg(lambda x: x.value_counts().to_dict())
+    expected = Series(
+        [{"two": 1, "one": 1, "three": 1}, {"two": 2, "one": 2, "three": 1}],
+        index=Index(["bar", "foo"], name="A"),
+        name="B",
+    )
+    tm.assert_series_equal(ts, expected)
+
+
+def test_cython_fail_agg():
+    dr = bdate_range("1/1/2000", periods=50)
+    ts = Series(["A", "B", "C", "D", "E"] * 10, dtype=object, index=dr)
+
+    grouped = ts.groupby(lambda x: x.month)
+    summed = grouped.sum()
+    expected = grouped.agg(np.sum).astype(object)
+    tm.assert_series_equal(summed, expected)
+
+
+@pytest.mark.parametrize(
+    "op, targop",
+    [
+        ("mean", np.mean),
+        ("median", np.median),
+        ("var", np.var),
+        ("sum", np.sum),
+        ("prod", np.prod),
+        ("min", np.min),
+        ("max", np.max),
+        ("first", lambda x: x.iloc[0]),
+        ("last", lambda x: x.iloc[-1]),
+    ],
+)
+def test__cython_agg_general(op, targop):
+    df = DataFrame(np.random.default_rng(2).standard_normal(1000))
+    labels = np.random.default_rng(2).integers(0, 50, size=1000).astype(float)
+    kwargs = {"ddof": 1} if op == "var" else {}
+    if op not in ["first", "last"]:
+        kwargs["axis"] = 0
+
+    result = df.groupby(labels)._cython_agg_general(op, alt=None, numeric_only=True)
+    expected = df.groupby(labels).agg(targop, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op, targop",
+    [
+        ("mean", np.mean),
+        ("median", lambda x: np.median(x) if len(x) > 0 else np.nan),
+        ("var", lambda x: np.var(x, ddof=1)),
+        ("min", np.min),
+        ("max", np.max),
+    ],
+)
+def test_cython_agg_empty_buckets(op, targop, observed):
+    df = DataFrame([11, 12, 13])
+    grps = range(0, 55, 5)
+
+    # calling _cython_agg_general directly, instead of via the user API
+    # which sets different values for min_count, so do that here.
+    g = df.groupby(pd.cut(df[0], grps), observed=observed)
+    result = g._cython_agg_general(op, alt=None, numeric_only=True)
+
+    g = df.groupby(pd.cut(df[0], grps), observed=observed)
+    expected = g.agg(lambda x: targop(x))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cython_agg_empty_buckets_nanops(observed):
+    # GH-18869 can't call nanops on empty groups, so hardcode expected
+    # for these
+    df = DataFrame([11, 12, 13], columns=["a"])
+    grps = np.arange(0, 25, 5, dtype=int)
+    # add / sum
+    result = df.groupby(pd.cut(df["a"], grps), observed=observed)._cython_agg_general(
+        "sum", alt=None, numeric_only=True
+    )
+    intervals = pd.interval_range(0, 20, freq=5)
+    expected = DataFrame(
+        {"a": [0, 0, 36, 0]},
+        index=pd.CategoricalIndex(intervals, name="a", ordered=True),
+    )
+    if observed:
+        expected = expected[expected.a != 0]
+
+    tm.assert_frame_equal(result, expected)
+
+    # prod
+    result = df.groupby(pd.cut(df["a"], grps), observed=observed)._cython_agg_general(
+        "prod", alt=None, numeric_only=True
+    )
+    expected = DataFrame(
+        {"a": [1, 1, 1716, 1]},
+        index=pd.CategoricalIndex(intervals, name="a", ordered=True),
+    )
+    if observed:
+        expected = expected[expected.a != 1]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", ["first", "last", "max", "min"])
+@pytest.mark.parametrize(
+    "data", [Timestamp("2016-10-14 21:00:44.557"), Timedelta("17088 days 21:00:44.557")]
+)
+def test_cython_with_timestamp_and_nat(op, data):
+    # https://github.com/pandas-dev/pandas/issues/19526
+    df = DataFrame({"a": [0, 1], "b": [data, NaT]})
+    index = Index([0, 1], name="a")
+
+    # We will group by a and test the cython aggregations
+    expected = DataFrame({"b": [data, NaT]}, index=index)
+
+    result = df.groupby("a").aggregate(op)
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize(
+    "agg",
+    [
+        "min",
+        "max",
+        "count",
+        "sum",
+        "prod",
+        "var",
+        "mean",
+        "median",
+        "ohlc",
+        "cumprod",
+        "cumsum",
+        "shift",
+        "any",
+        "all",
+        "quantile",
+        "first",
+        "last",
+        "rank",
+        "cummin",
+        "cummax",
+    ],
+)
+def test_read_only_buffer_source_agg(agg):
+    # https://github.com/pandas-dev/pandas/issues/36014
+    df = DataFrame(
+        {
+            "sepal_length": [5.1, 4.9, 4.7, 4.6, 5.0],
+            "species": ["setosa", "setosa", "setosa", "setosa", "setosa"],
+        }
+    )
+    df._mgr.blocks[0].values.flags.writeable = False
+
+    result = df.groupby(["species"]).agg({"sepal_length": agg})
+    expected = df.copy().groupby(["species"]).agg({"sepal_length": agg})
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op_name",
+    [
+        "count",
+        "sum",
+        "std",
+        "var",
+        "sem",
+        "mean",
+        "median",
+        "prod",
+        "min",
+        "max",
+    ],
+)
+def test_cython_agg_nullable_int(op_name):
+    # ensure that the cython-based aggregations don't fail for nullable dtype
+    # (eg https://github.com/pandas-dev/pandas/issues/37415)
+    df = DataFrame(
+        {
+            "A": ["A", "B"] * 5,
+            "B": pd.array([1, 2, 3, 4, 5, 6, 7, 8, 9, pd.NA], dtype="Int64"),
+        }
+    )
+    result = getattr(df.groupby("A")["B"], op_name)()
+    df2 = df.assign(B=df["B"].astype("float64"))
+    expected = getattr(df2.groupby("A")["B"], op_name)()
+    if op_name in ("mean", "median"):
+        convert_integer = False
+    else:
+        convert_integer = True
+    expected = expected.convert_dtypes(convert_integer=convert_integer)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["Int64", "Float64", "boolean"])
+def test_count_masked_returns_masked_dtype(dtype):
+    df = DataFrame(
+        {
+            "A": [1, 1],
+            "B": pd.array([1, pd.NA], dtype=dtype),
+            "C": pd.array([1, 1], dtype=dtype),
+        }
+    )
+    result = df.groupby("A").count()
+    expected = DataFrame(
+        [[1, 2]], index=Index([1], name="A"), columns=["B", "C"], dtype="Int64"
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("with_na", [True, False])
+@pytest.mark.parametrize(
+    "op_name, action",
+    [
+        # ("count", "always_int"),
+        ("sum", "large_int"),
+        # ("std", "always_float"),
+        ("var", "always_float"),
+        # ("sem", "always_float"),
+        ("mean", "always_float"),
+        ("median", "always_float"),
+        ("prod", "large_int"),
+        ("min", "preserve"),
+        ("max", "preserve"),
+        ("first", "preserve"),
+        ("last", "preserve"),
+    ],
+)
+@pytest.mark.parametrize(
+    "data",
+    [
+        pd.array([1, 2, 3, 4], dtype="Int64"),
+        pd.array([1, 2, 3, 4], dtype="Int8"),
+        pd.array([0.1, 0.2, 0.3, 0.4], dtype="Float32"),
+        pd.array([0.1, 0.2, 0.3, 0.4], dtype="Float64"),
+        pd.array([True, True, False, False], dtype="boolean"),
+    ],
+)
+def test_cython_agg_EA_known_dtypes(data, op_name, action, with_na):
+    if with_na:
+        data[3] = pd.NA
+
+    df = DataFrame({"key": ["a", "a", "b", "b"], "col": data})
+    grouped = df.groupby("key")
+
+    if action == "always_int":
+        # always Int64
+        expected_dtype = pd.Int64Dtype()
+    elif action == "large_int":
+        # for any int/bool use Int64, for float preserve dtype
+        if is_float_dtype(data.dtype):
+            expected_dtype = data.dtype
+        elif is_integer_dtype(data.dtype):
+            # match the numpy dtype we'd get with the non-nullable analogue
+            expected_dtype = data.dtype
+        else:
+            expected_dtype = pd.Int64Dtype()
+    elif action == "always_float":
+        # for any int/bool use Float64, for float preserve dtype
+        if is_float_dtype(data.dtype):
+            expected_dtype = data.dtype
+        else:
+            expected_dtype = pd.Float64Dtype()
+    elif action == "preserve":
+        expected_dtype = data.dtype
+
+    result = getattr(grouped, op_name)()
+    assert result["col"].dtype == expected_dtype
+
+    result = grouped.aggregate(op_name)
+    assert result["col"].dtype == expected_dtype
+
+    result = getattr(grouped["col"], op_name)()
+    assert result.dtype == expected_dtype
+
+    result = grouped["col"].aggregate(op_name)
+    assert result.dtype == expected_dtype
diff --git a/pandas/pandas/tests/groupby/aggregate/test_numba.py b/pandas/pandas/tests/groupby/aggregate/test_numba.py
new file mode 100644
index 0000000000000..afddc90fdd055
--- /dev/null
+++ b/pandas/pandas/tests/groupby/aggregate/test_numba.py
@@ -0,0 +1,441 @@
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+from pandas.errors import NumbaUtilError
+
+from pandas import (
+    DataFrame,
+    Index,
+    NamedAgg,
+    Series,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [pytest.mark.single_cpu]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+def test_correct_function_signature():
+    pytest.importorskip("numba")
+
+    def incorrect_function(x):
+        return sum(x) * 2.7
+
+    data = DataFrame(
+        {"key": ["a", "a", "b", "b", "a"], "data": [1.0, 2.0, 3.0, 4.0, 5.0]},
+        columns=["key", "data"],
+    )
+    with pytest.raises(NumbaUtilError, match="The first 2"):
+        data.groupby("key").agg(incorrect_function, engine="numba")
+
+    with pytest.raises(NumbaUtilError, match="The first 2"):
+        data.groupby("key")["data"].agg(incorrect_function, engine="numba")
+
+
+def test_check_nopython_kwargs():
+    pytest.importorskip("numba")
+
+    def incorrect_function(values, index, *, a):
+        return sum(values) * 2.7 + a
+
+    def correct_function(values, index, a):
+        return sum(values) * 2.7 + a
+
+    data = DataFrame(
+        {"key": ["a", "a", "b", "b", "a"], "data": [1.0, 2.0, 3.0, 4.0, 5.0]},
+        columns=["key", "data"],
+    )
+    expected = data.groupby("key").sum() * 2.7
+
+    # py signature binding
+    with pytest.raises(
+        TypeError, match="missing a required (keyword-only argument|argument): 'a'"
+    ):
+        data.groupby("key").agg(incorrect_function, engine="numba", b=1)
+    with pytest.raises(TypeError, match="missing a required argument: 'a'"):
+        data.groupby("key").agg(correct_function, engine="numba", b=1)
+
+    with pytest.raises(
+        TypeError, match="missing a required (keyword-only argument|argument): 'a'"
+    ):
+        data.groupby("key")["data"].agg(incorrect_function, engine="numba", b=1)
+    with pytest.raises(TypeError, match="missing a required argument: 'a'"):
+        data.groupby("key")["data"].agg(correct_function, engine="numba", b=1)
+
+    # numba signature check after binding
+    with pytest.raises(NumbaUtilError, match="numba does not support"):
+        data.groupby("key").agg(incorrect_function, engine="numba", a=1)
+    actual = data.groupby("key").agg(correct_function, engine="numba", a=1)
+    tm.assert_frame_equal(expected + 1, actual)
+
+    with pytest.raises(NumbaUtilError, match="numba does not support"):
+        data.groupby("key")["data"].agg(incorrect_function, engine="numba", a=1)
+    actual = data.groupby("key")["data"].agg(correct_function, engine="numba", a=1)
+    tm.assert_series_equal(expected["data"] + 1, actual)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+@pytest.mark.parametrize("jit", [True, False])
+def test_numba_vs_cython(jit, frame_or_series, nogil, parallel, nopython, as_index):
+    pytest.importorskip("numba")
+
+    def func_numba(values, index):
+        return np.mean(values) * 2.7
+
+    if jit:
+        # Test accepted jitted functions
+        import numba
+
+        func_numba = numba.jit(func_numba)
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+    grouped = data.groupby(0, as_index=as_index)
+    if frame_or_series is Series:
+        grouped = grouped[1]
+
+    result = grouped.agg(func_numba, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.agg(lambda x: np.mean(x) * 2.7, engine="cython")
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+@pytest.mark.parametrize("jit", [True, False])
+def test_cache(jit, frame_or_series, nogil, parallel, nopython):
+    # Test that the functions are cached correctly if we switch functions
+    pytest.importorskip("numba")
+
+    def func_1(values, index):
+        return np.mean(values) - 3.4
+
+    def func_2(values, index):
+        return np.mean(values) * 2.7
+
+    if jit:
+        import numba
+
+        func_1 = numba.jit(func_1)
+        func_2 = numba.jit(func_2)
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+    grouped = data.groupby(0)
+    if frame_or_series is Series:
+        grouped = grouped[1]
+
+    result = grouped.agg(func_1, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.agg(lambda x: np.mean(x) - 3.4, engine="cython")
+    tm.assert_equal(result, expected)
+
+    # Add func_2 to the cache
+    result = grouped.agg(func_2, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.agg(lambda x: np.mean(x) * 2.7, engine="cython")
+    tm.assert_equal(result, expected)
+
+    # Retest func_1 which should use the cache
+    result = grouped.agg(func_1, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.agg(lambda x: np.mean(x) - 3.4, engine="cython")
+    tm.assert_equal(result, expected)
+
+
+def test_use_global_config():
+    pytest.importorskip("numba")
+
+    def func_1(values, index):
+        return np.mean(values) - 3.4
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    grouped = data.groupby(0)
+    expected = grouped.agg(func_1, engine="numba")
+    with option_context("compute.use_numba", True):
+        result = grouped.agg(func_1, engine=None)
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize(
+    "agg_kwargs",
+    [
+        {"func": ["min", "max"]},
+        {"func": "min"},
+        {"func": {1: ["min", "max"], 2: "sum"}},
+        {"bmin": NamedAgg(column=1, aggfunc="min")},
+    ],
+)
+def test_multifunc_numba_vs_cython_frame(agg_kwargs):
+    pytest.importorskip("numba")
+    data = DataFrame(
+        {
+            0: ["a", "a", "b", "b", "a"],
+            1: [1.0, 2.0, 3.0, 4.0, 5.0],
+            2: [1, 2, 3, 4, 5],
+        },
+        columns=[0, 1, 2],
+    )
+    grouped = data.groupby(0)
+    result = grouped.agg(**agg_kwargs, engine="numba")
+    expected = grouped.agg(**agg_kwargs, engine="cython")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["sum", "mean", "var", "std", "min", "max"])
+def test_multifunc_numba_vs_cython_frame_noskipna(func):
+    pytest.importorskip("numba")
+    data = DataFrame(
+        {
+            0: ["a", "a", "b", "b", "a"],
+            1: [1.0, np.nan, 3.0, 4.0, 5.0],
+            2: [1, 2, 3, 4, 5],
+        },
+        columns=[0, 1, 2],
+    )
+    grouped = data.groupby(0)
+    result = grouped.agg(func, skipna=False, engine="numba")
+    expected = grouped.agg(func, skipna=False, engine="cython")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg_kwargs,expected_func",
+    [
+        ({"func": lambda values, index: values.sum()}, "sum"),
+        # FIXME
+        pytest.param(
+            {
+                "func": [
+                    lambda values, index: values.sum(),
+                    lambda values, index: values.min(),
+                ]
+            },
+            ["sum", "min"],
+            marks=pytest.mark.xfail(
+                reason="This doesn't work yet! Fails in nopython pipeline!"
+            ),
+        ),
+    ],
+)
+def test_multifunc_numba_udf_frame(agg_kwargs, expected_func):
+    pytest.importorskip("numba")
+    data = DataFrame(
+        {
+            0: ["a", "a", "b", "b", "a"],
+            1: [1.0, 2.0, 3.0, 4.0, 5.0],
+            2: [1, 2, 3, 4, 5],
+        },
+        columns=[0, 1, 2],
+    )
+    grouped = data.groupby(0)
+    result = grouped.agg(**agg_kwargs, engine="numba")
+    expected = grouped.agg(expected_func, engine="cython")
+    # check_dtype can be removed if GH 44952 is addressed
+    # Currently, UDFs still always return float64 while reductions can preserve dtype
+    tm.assert_frame_equal(result, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize(
+    "agg_kwargs",
+    [{"func": ["min", "max"]}, {"func": "min"}, {"min_val": "min", "max_val": "max"}],
+)
+def test_multifunc_numba_vs_cython_series(agg_kwargs):
+    pytest.importorskip("numba")
+    labels = ["a", "a", "b", "b", "a"]
+    data = Series([1.0, 2.0, 3.0, 4.0, 5.0])
+    grouped = data.groupby(labels)
+    agg_kwargs["engine"] = "numba"
+    result = grouped.agg(**agg_kwargs)
+    agg_kwargs["engine"] = "cython"
+    expected = grouped.agg(**agg_kwargs)
+    if isinstance(expected, DataFrame):
+        tm.assert_frame_equal(result, expected)
+    else:
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.single_cpu
+@pytest.mark.parametrize(
+    "data,agg_kwargs",
+    [
+        (Series([1.0, 2.0, 3.0, 4.0, 5.0]), {"func": ["min", "max"]}),
+        (Series([1.0, 2.0, 3.0, 4.0, 5.0]), {"func": "min"}),
+        (
+            DataFrame(
+                {1: [1.0, 2.0, 3.0, 4.0, 5.0], 2: [1, 2, 3, 4, 5]}, columns=[1, 2]
+            ),
+            {"func": ["min", "max"]},
+        ),
+        (
+            DataFrame(
+                {1: [1.0, 2.0, 3.0, 4.0, 5.0], 2: [1, 2, 3, 4, 5]}, columns=[1, 2]
+            ),
+            {"func": "min"},
+        ),
+        (
+            DataFrame(
+                {1: [1.0, 2.0, 3.0, 4.0, 5.0], 2: [1, 2, 3, 4, 5]}, columns=[1, 2]
+            ),
+            {"func": {1: ["min", "max"], 2: "sum"}},
+        ),
+        (
+            DataFrame(
+                {1: [1.0, 2.0, 3.0, 4.0, 5.0], 2: [1, 2, 3, 4, 5]}, columns=[1, 2]
+            ),
+            {"min_col": NamedAgg(column=1, aggfunc="min")},
+        ),
+    ],
+)
+def test_multifunc_numba_kwarg_propagation(data, agg_kwargs):
+    pytest.importorskip("numba")
+    labels = ["a", "a", "b", "b", "a"]
+    grouped = data.groupby(labels)
+    result = grouped.agg(**agg_kwargs, engine="numba", engine_kwargs={"parallel": True})
+    expected = grouped.agg(**agg_kwargs, engine="numba")
+    if isinstance(expected, DataFrame):
+        tm.assert_frame_equal(result, expected)
+    else:
+        tm.assert_series_equal(result, expected)
+
+
+def test_args_not_cached():
+    # GH 41647
+    pytest.importorskip("numba")
+
+    def sum_last(values, index, n):
+        return values[-n:].sum()
+
+    df = DataFrame({"id": [0, 0, 1, 1], "x": [1, 1, 1, 1]})
+    grouped_x = df.groupby("id")["x"]
+    result = grouped_x.agg(sum_last, 1, engine="numba")
+    expected = Series([1.0] * 2, name="x", index=Index([0, 1], name="id"))
+    tm.assert_series_equal(result, expected)
+
+    result = grouped_x.agg(sum_last, 2, engine="numba")
+    expected = Series([2.0] * 2, name="x", index=Index([0, 1], name="id"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_index_data_correctly_passed():
+    # GH 43133
+    pytest.importorskip("numba")
+
+    def f(values, index):
+        return np.mean(index)
+
+    df = DataFrame({"group": ["A", "A", "B"], "v": [4, 5, 6]}, index=[-1, -2, -3])
+    result = df.groupby("group").aggregate(f, engine="numba")
+    expected = DataFrame(
+        [-1.5, -3.0], columns=["v"], index=Index(["A", "B"], name="group")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_engine_kwargs_not_cached():
+    # If the user passes a different set of engine_kwargs don't return the same
+    # jitted function
+    pytest.importorskip("numba")
+    nogil = True
+    parallel = False
+    nopython = True
+
+    def func_kwargs(values, index):
+        return nogil + parallel + nopython
+
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    df = DataFrame({"value": [0, 0, 0]})
+    result = df.groupby(level=0).aggregate(
+        func_kwargs, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame({"value": [2.0, 2.0, 2.0]})
+    tm.assert_frame_equal(result, expected)
+
+    nogil = False
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    result = df.groupby(level=0).aggregate(
+        func_kwargs, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame({"value": [1.0, 1.0, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore")
+def test_multiindex_one_key(nogil, parallel, nopython):
+    pytest.importorskip("numba")
+
+    def numba_func(values, index):
+        return 1
+
+    df = DataFrame([{"A": 1, "B": 2, "C": 3}]).set_index(["A", "B"])
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    result = df.groupby("A").agg(
+        numba_func, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame([1.0], index=Index([1], name="A"), columns=["C"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multiindex_multi_key_not_supported(nogil, parallel, nopython):
+    pytest.importorskip("numba")
+
+    def numba_func(values, index):
+        return 1
+
+    df = DataFrame([{"A": 1, "B": 2, "C": 3}]).set_index(["A", "B"])
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    with pytest.raises(NotImplementedError, match="more than 1 grouping labels"):
+        df.groupby(["A", "B"]).agg(
+            numba_func, engine="numba", engine_kwargs=engine_kwargs
+        )
+
+
+def test_multilabel_numba_vs_cython(numba_supported_reductions):
+    pytest.importorskip("numba")
+    reduction, kwargs = numba_supported_reductions
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+    gb = df.groupby(["A", "B"])
+    res_agg = gb.agg(reduction, engine="numba", **kwargs)
+    expected_agg = gb.agg(reduction, engine="cython", **kwargs)
+    tm.assert_frame_equal(res_agg, expected_agg)
+    # Test that calling the aggregation directly also works
+    direct_res = getattr(gb, reduction)(engine="numba", **kwargs)
+    direct_expected = getattr(gb, reduction)(engine="cython", **kwargs)
+    tm.assert_frame_equal(direct_res, direct_expected)
+
+
+def test_multilabel_udf_numba_vs_cython():
+    pytest.importorskip("numba")
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+    gb = df.groupby(["A", "B"])
+    result = gb.agg(lambda values, index: values.min(), engine="numba")
+    expected = gb.agg(lambda x: x.min(), engine="cython")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/aggregate/test_other.py b/pandas/pandas/tests/groupby/aggregate/test_other.py
new file mode 100644
index 0000000000000..1c016143d50c3
--- /dev/null
+++ b/pandas/pandas/tests/groupby/aggregate/test_other.py
@@ -0,0 +1,665 @@
+"""
+test all other .agg behavior
+"""
+
+import datetime as dt
+from functools import partial
+
+import numpy as np
+import pytest
+
+from pandas.errors import SpecificationError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+from pandas.io.formats.printing import pprint_thing
+
+
+def test_agg_partial_failure_raises():
+    # GH#43741
+
+    df = DataFrame(
+        {
+            "data1": np.random.default_rng(2).standard_normal(5),
+            "data2": np.random.default_rng(2).standard_normal(5),
+            "key1": ["a", "a", "b", "b", "a"],
+            "key2": ["one", "two", "one", "two", "one"],
+        }
+    )
+    grouped = df.groupby("key1")
+
+    def peak_to_peak(arr):
+        return arr.max() - arr.min()
+
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        grouped.agg([peak_to_peak])
+
+    with pytest.raises(TypeError, match="unsupported operand type"):
+        grouped.agg(peak_to_peak)
+
+
+def test_agg_datetimes_mixed():
+    data = [[1, "2012-01-01", 1.0], [2, "2012-01-02", 2.0], [3, None, 3.0]]
+
+    df1 = DataFrame(
+        {
+            "key": [x[0] for x in data],
+            "date": [x[1] for x in data],
+            "value": [x[2] for x in data],
+        }
+    )
+
+    data = [
+        [
+            row[0],
+            (dt.datetime.strptime(row[1], "%Y-%m-%d").date() if row[1] else None),
+            row[2],
+        ]
+        for row in data
+    ]
+
+    df2 = DataFrame(
+        {
+            "key": [x[0] for x in data],
+            "date": [x[1] for x in data],
+            "value": [x[2] for x in data],
+        }
+    )
+
+    df1["weights"] = df1["value"] / df1["value"].sum()
+    gb1 = df1.groupby("date").aggregate("sum")
+
+    df2["weights"] = df1["value"] / df1["value"].sum()
+    gb2 = df2.groupby("date").aggregate("sum")
+
+    assert len(gb1) == len(gb2)
+
+
+def test_agg_period_index():
+    prng = period_range("2012-1-1", freq="M", periods=3)
+    df = DataFrame(np.random.default_rng(2).standard_normal((3, 2)), index=prng)
+    rs = df.groupby(level=0).sum()
+    assert isinstance(rs.index, PeriodIndex)
+
+    # GH 3579
+    index = period_range(start="1999-01", periods=5, freq="M")
+    s1 = Series(np.random.default_rng(2).random(len(index)), index=index)
+    s2 = Series(np.random.default_rng(2).random(len(index)), index=index)
+    df = DataFrame.from_dict({"s1": s1, "s2": s2})
+    grouped = df.groupby(df.index.month)
+    list(grouped)
+
+
+def test_agg_dict_parameter_cast_result_dtypes():
+    # GH 12821
+
+    df = DataFrame(
+        {
+            "class": ["A", "A", "B", "B", "C", "C", "D", "D"],
+            "time": date_range("1/1/2011", periods=8, freq="h"),
+        }
+    )
+    df.loc[[0, 1, 2, 5], "time"] = None
+
+    # test for `first` function
+    exp = df.loc[[0, 3, 4, 6]].set_index("class")
+    grouped = df.groupby("class")
+    tm.assert_frame_equal(grouped.first(), exp)
+    tm.assert_frame_equal(grouped.agg("first"), exp)
+    tm.assert_frame_equal(grouped.agg({"time": "first"}), exp)
+    tm.assert_series_equal(grouped.time.first(), exp["time"])
+    tm.assert_series_equal(grouped.time.agg("first"), exp["time"])
+
+    # test for `last` function
+    exp = df.loc[[0, 3, 4, 7]].set_index("class")
+    grouped = df.groupby("class")
+    tm.assert_frame_equal(grouped.last(), exp)
+    tm.assert_frame_equal(grouped.agg("last"), exp)
+    tm.assert_frame_equal(grouped.agg({"time": "last"}), exp)
+    tm.assert_series_equal(grouped.time.last(), exp["time"])
+    tm.assert_series_equal(grouped.time.agg("last"), exp["time"])
+
+    # count
+    exp = Series([2, 2, 2, 2], index=Index(list("ABCD"), name="class"), name="time")
+    tm.assert_series_equal(grouped.time.agg(len), exp)
+    tm.assert_series_equal(grouped.time.size(), exp)
+
+    exp = Series([0, 1, 1, 2], index=Index(list("ABCD"), name="class"), name="time")
+    tm.assert_series_equal(grouped.time.count(), exp)
+
+
+def test_agg_cast_results_dtypes():
+    # similar to GH12821
+    # xref #11444
+    u = [dt.datetime(2015, x + 1, 1) for x in range(12)]
+    v = list("aaabbbbbbccd")
+    df = DataFrame({"X": v, "Y": u})
+
+    result = df.groupby("X")["Y"].agg(len)
+    expected = df.groupby("X")["Y"].count()
+    tm.assert_series_equal(result, expected)
+
+
+def test_aggregate_float64_no_int64():
+    # see gh-11199
+    df = DataFrame({"a": [1, 2, 3, 4, 5], "b": [1, 2, 2, 4, 5], "c": [1, 2, 3, 4, 5]})
+
+    expected = DataFrame({"a": [1, 2.5, 4, 5]}, index=[1, 2, 4, 5])
+    expected.index.name = "b"
+
+    result = df.groupby("b")[["a"]].mean()
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame({"a": [1, 2.5, 4, 5], "c": [1, 2.5, 4, 5]}, index=[1, 2, 4, 5])
+    expected.index.name = "b"
+
+    result = df.groupby("b")[["a", "c"]].mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_aggregate_api_consistency():
+    # GH 9052
+    # make sure that the aggregates via dict
+    # are consistent
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": np.random.default_rng(2).standard_normal(8) + 1.0,
+            "D": np.arange(8),
+        }
+    )
+
+    grouped = df.groupby(["A", "B"])
+    c_mean = grouped["C"].mean()
+    c_sum = grouped["C"].sum()
+    d_mean = grouped["D"].mean()
+    d_sum = grouped["D"].sum()
+
+    result = grouped["D"].agg(["sum", "mean"])
+    expected = pd.concat([d_sum, d_mean], axis=1)
+    expected.columns = ["sum", "mean"]
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+    result = grouped.agg(["sum", "mean"])
+    expected = pd.concat([c_sum, c_mean, d_sum, d_mean], axis=1)
+    expected.columns = MultiIndex.from_product([["C", "D"], ["sum", "mean"]])
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+    result = grouped[["D", "C"]].agg(["sum", "mean"])
+    expected = pd.concat([d_sum, d_mean, c_sum, c_mean], axis=1)
+    expected.columns = MultiIndex.from_product([["D", "C"], ["sum", "mean"]])
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+    result = grouped.agg({"C": "mean", "D": "sum"})
+    expected = pd.concat([d_sum, c_mean], axis=1)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+    result = grouped.agg({"C": ["mean", "sum"], "D": ["mean", "sum"]})
+    expected = pd.concat([c_mean, c_sum, d_mean, d_sum], axis=1)
+    expected.columns = MultiIndex.from_product([["C", "D"], ["mean", "sum"]])
+
+    msg = r"Label\(s\) \['r', 'r2'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        grouped[["D", "C"]].agg({"r": "sum", "r2": "mean"})
+
+
+def test_agg_dict_renaming_deprecation():
+    # 15931
+    df = DataFrame({"A": [1, 1, 1, 2, 2], "B": range(5), "C": range(5)})
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        df.groupby("A").agg(
+            {"B": {"foo": ["sum", "max"]}, "C": {"bar": ["count", "min"]}}
+        )
+
+    msg = r"Label\(s\) \['ma'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        df.groupby("A")[["B", "C"]].agg({"ma": "max"})
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        df.groupby("A").B.agg({"foo": "count"})
+
+
+def test_agg_compat():
+    # GH 12334
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": np.random.default_rng(2).standard_normal(8) + 1.0,
+            "D": np.arange(8),
+        }
+    )
+
+    g = df.groupby(["A", "B"])
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        g["D"].agg({"C": ["sum", "std"]})
+
+    with pytest.raises(SpecificationError, match=msg):
+        g["D"].agg({"C": "sum", "D": "std"})
+
+
+def test_agg_nested_dicts():
+    # API change for disallowing these types of nested dicts
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": np.random.default_rng(2).standard_normal(8) + 1.0,
+            "D": np.arange(8),
+        }
+    )
+
+    g = df.groupby(["A", "B"])
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        g.aggregate({"r1": {"C": ["mean", "sum"]}, "r2": {"D": ["mean", "sum"]}})
+
+    with pytest.raises(SpecificationError, match=msg):
+        g.agg({"C": {"ra": ["mean", "std"]}, "D": {"rb": ["mean", "std"]}})
+
+    # same name as the original column
+    # GH9052
+    with pytest.raises(SpecificationError, match=msg):
+        g["D"].agg({"result1": np.sum, "result2": np.mean})
+
+    with pytest.raises(SpecificationError, match=msg):
+        g["D"].agg({"D": np.sum, "result2": np.mean})
+
+
+def test_agg_item_by_item_raise_typeerror():
+    df = DataFrame(np.random.default_rng(2).integers(10, size=(20, 10)))
+
+    def raiseException(df):
+        pprint_thing("----------------------------------------")
+        pprint_thing(df.to_string())
+        raise TypeError("test")
+
+    with pytest.raises(TypeError, match="test"):
+        df.groupby(0).agg(raiseException)
+
+
+def test_series_agg_multikey():
+    ts = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+    grouped = ts.groupby([lambda x: x.year, lambda x: x.month])
+
+    result = grouped.agg("sum")
+    expected = grouped.sum()
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_agg_multi_pure_python():
+    data = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    def bad(x):
+        if isinstance(x.values, np.ndarray):
+            assert len(x.values.base) > 0
+        return "foo"
+
+    result = data.groupby(["A", "B"]).agg(bad)
+    expected = data.groupby(["A", "B"]).agg(lambda x: "foo")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_consistency():
+    # agg with ([]) and () not consistent
+    # GH 6715
+    def P1(a):
+        return np.percentile(a.dropna(), q=1)
+
+    df = DataFrame(
+        {
+            "col1": [1, 2, 3, 4],
+            "col2": [10, 25, 26, 31],
+            "date": [
+                dt.date(2013, 2, 10),
+                dt.date(2013, 2, 10),
+                dt.date(2013, 2, 11),
+                dt.date(2013, 2, 11),
+            ],
+        }
+    )
+
+    g = df.groupby("date")
+
+    expected = g.agg([P1])
+    expected.columns = expected.columns.levels[0]
+
+    result = g.agg(P1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_callables():
+    # GH 7929
+    df = DataFrame({"foo": [1, 2], "bar": [3, 4]}).astype(np.int64)
+
+    class fn_class:
+        def __call__(self, x):
+            return sum(x)
+
+    equiv_callables = [
+        sum,
+        np.sum,
+        lambda x: sum(x),
+        lambda x: x.sum(),
+        partial(sum),
+        fn_class(),
+    ]
+
+    expected = df.groupby("foo").agg("sum")
+    for ecall in equiv_callables:
+        result = df.groupby("foo").agg(ecall)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_agg_over_numpy_arrays():
+    # GH 3788
+    df = DataFrame(
+        [
+            [1, np.array([10, 20, 30])],
+            [1, np.array([40, 50, 60])],
+            [2, np.array([20, 30, 40])],
+        ],
+        columns=["category", "arraydata"],
+    )
+    gb = df.groupby("category")
+
+    expected_data = [[np.array([50, 70, 90])], [np.array([20, 30, 40])]]
+    expected_index = Index([1, 2], name="category")
+    expected_column = ["arraydata"]
+    expected = DataFrame(expected_data, index=expected_index, columns=expected_column)
+
+    alt = gb.sum(numeric_only=False)
+    tm.assert_frame_equal(alt, expected)
+
+    result = gb.agg("sum", numeric_only=False)
+    tm.assert_frame_equal(result, expected)
+
+    # FIXME: the original version of this test called `gb.agg(sum)`
+    #  and that raises TypeError if `numeric_only=False` is passed
+
+
+@pytest.mark.parametrize("as_period", [True, False])
+def test_agg_tzaware_non_datetime_result(as_period):
+    # discussed in GH#29589, fixed in GH#29641, operating on tzaware values
+    #  with function that is not dtype-preserving
+    dti = date_range("2012-01-01", periods=4, tz="UTC")
+    if as_period:
+        dti = dti.tz_localize(None).to_period("D")
+
+    df = DataFrame({"a": [0, 0, 1, 1], "b": dti})
+    gb = df.groupby("a")
+
+    # Case that _does_ preserve the dtype
+    result = gb["b"].agg(lambda x: x.iloc[0])
+    expected = Series(dti[::2], name="b")
+    expected.index.name = "a"
+    tm.assert_series_equal(result, expected)
+
+    # Cases that do _not_ preserve the dtype
+    result = gb["b"].agg(lambda x: x.iloc[0].year)
+    expected = Series([2012, 2012], name="b")
+    expected.index.name = "a"
+    tm.assert_series_equal(result, expected)
+
+    result = gb["b"].agg(lambda x: x.iloc[-1] - x.iloc[0])
+    expected = Series([pd.Timedelta(days=1), pd.Timedelta(days=1)], name="b")
+    expected.index.name = "a"
+    if as_period:
+        expected = Series([pd.offsets.Day(1), pd.offsets.Day(1)], name="b")
+        expected.index.name = "a"
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_timezone_round_trip():
+    # GH 15426
+    ts = pd.Timestamp("2016-01-01 12:00:00", tz="US/Pacific")
+    df = DataFrame({"a": 1, "b": [ts + dt.timedelta(minutes=nn) for nn in range(10)]})
+
+    result1 = df.groupby("a")["b"].agg("min").iloc[0]
+    result2 = df.groupby("a")["b"].agg(lambda x: np.min(x)).iloc[0]
+    result3 = df.groupby("a")["b"].min().iloc[0]
+
+    assert result1 == ts
+    assert result2 == ts
+    assert result3 == ts
+
+    dates = [
+        pd.Timestamp(f"2016-01-0{i:d} 12:00:00", tz="US/Pacific") for i in range(1, 5)
+    ]
+    df = DataFrame({"A": ["a", "b"] * 2, "B": dates})
+    grouped = df.groupby("A")
+
+    ts = df["B"].iloc[0]
+    assert ts == grouped.nth(0)["B"].iloc[0]
+    assert ts == grouped.head(1)["B"].iloc[0]
+    assert ts == grouped.first()["B"].iloc[0]
+
+    # GH#27110 applying iloc should return a DataFrame
+    assert ts == grouped.apply(lambda x: x.iloc[0])["B"].iloc[0]
+
+    ts = df["B"].iloc[2]
+    assert ts == grouped.last()["B"].iloc[0]
+
+    # GH#27110 applying iloc should return a DataFrame
+    assert ts == grouped.apply(lambda x: x.iloc[-1])["B"].iloc[0]
+
+
+def test_sum_uint64_overflow():
+    # see gh-14758
+    # Convert to uint64 and don't overflow
+    df = DataFrame([[1, 2], [3, 4], [5, 6]], dtype=object)
+    df = df + 9223372036854775807
+
+    index = Index(
+        [9223372036854775808, 9223372036854775810, 9223372036854775812], dtype=np.uint64
+    )
+    expected = DataFrame(
+        {1: [9223372036854775809, 9223372036854775811, 9223372036854775813]},
+        index=index,
+        dtype=object,
+    )
+
+    expected.index.name = 0
+    result = df.groupby(0).sum(numeric_only=False)
+    tm.assert_frame_equal(result, expected)
+
+    # out column is non-numeric, so with numeric_only=True it is dropped
+    result2 = df.groupby(0).sum(numeric_only=True)
+    expected2 = expected[[]]
+    tm.assert_frame_equal(result2, expected2)
+
+
+@pytest.mark.parametrize(
+    "structure, cast_as",
+    [
+        (tuple, tuple),
+        (list, list),
+        (lambda x: tuple(x), tuple),
+        (lambda x: list(x), list),
+    ],
+)
+def test_agg_structs_dataframe(structure, cast_as):
+    df = DataFrame(
+        {"A": [1, 1, 1, 3, 3, 3], "B": [1, 1, 1, 4, 4, 4], "C": [1, 1, 1, 3, 4, 4]}
+    )
+
+    result = df.groupby(["A", "B"]).aggregate(structure)
+    expected = DataFrame(
+        {"C": {(1, 1): cast_as([1, 1, 1]), (3, 4): cast_as([3, 4, 4])}}
+    )
+    expected.index.names = ["A", "B"]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "structure, cast_as",
+    [
+        (tuple, tuple),
+        (list, list),
+        (lambda x: tuple(x), tuple),
+        (lambda x: list(x), list),
+    ],
+)
+def test_agg_structs_series(structure, cast_as):
+    # Issue #18079
+    df = DataFrame(
+        {"A": [1, 1, 1, 3, 3, 3], "B": [1, 1, 1, 4, 4, 4], "C": [1, 1, 1, 3, 4, 4]}
+    )
+
+    result = df.groupby("A")["C"].aggregate(structure)
+    expected = Series([cast_as([1, 1, 1]), cast_as([3, 4, 4])], index=[1, 3], name="C")
+    expected.index.name = "A"
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_category_nansum(observed):
+    categories = ["a", "b", "c"]
+    df = DataFrame(
+        {"A": pd.Categorical(["a", "a", "b"], categories=categories), "B": [1, 2, 3]}
+    )
+    result = df.groupby("A", observed=observed).B.agg(np.nansum)
+    expected = Series(
+        [3, 3, 0],
+        index=pd.CategoricalIndex(["a", "b", "c"], categories=categories, name="A"),
+        name="B",
+    )
+    if observed:
+        expected = expected[expected != 0]
+    tm.assert_series_equal(result, expected)
+
+
+def test_agg_list_like_func():
+    # GH 18473
+    df = DataFrame({"A": [str(x) for x in range(3)], "B": [str(x) for x in range(3)]})
+    grouped = df.groupby("A", as_index=False, sort=False)
+    result = grouped.agg({"B": lambda x: list(x)})
+    expected = DataFrame(
+        {"A": [str(x) for x in range(3)], "B": [[str(x)] for x in range(3)]}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_lambda_with_timezone():
+    # GH 23683
+    df = DataFrame(
+        {
+            "tag": [1, 1],
+            "date": [
+                pd.Timestamp("2018-01-01", tz="UTC"),
+                pd.Timestamp("2018-01-02", tz="UTC"),
+            ],
+        }
+    )
+    result = df.groupby("tag").agg({"date": lambda e: e.head(1)})
+    expected = DataFrame(
+        [pd.Timestamp("2018-01-01", tz="UTC")],
+        index=Index([1], name="tag"),
+        columns=["date"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "err_cls",
+    [
+        NotImplementedError,
+        RuntimeError,
+        KeyError,
+        IndexError,
+        OSError,
+        ValueError,
+        ArithmeticError,
+        AttributeError,
+    ],
+)
+def test_groupby_agg_err_catching(err_cls):
+    # make sure we suppress anything other than TypeError or AssertionError
+    #  in _python_agg_general
+
+    # Use a non-standard EA to make sure we don't go down ndarray paths
+    from pandas.tests.extension.decimal.array import (
+        DecimalArray,
+        make_data,
+        to_decimal,
+    )
+
+    data = make_data()[:5]
+    df = DataFrame(
+        {"id1": [0, 0, 0, 1, 1], "id2": [0, 1, 0, 1, 1], "decimals": DecimalArray(data)}
+    )
+
+    expected = Series(to_decimal([data[0], data[3]]))
+
+    def weird_func(x):
+        # weird function that raise something other than TypeError or IndexError
+        #  in _python_agg_general
+        if len(x) == 0:
+            raise err_cls
+        return x.iloc[0]
+
+    result = df["decimals"].groupby(df["id1"]).agg(weird_func)
+    tm.assert_series_equal(result, expected, check_names=False)
diff --git a/pandas/pandas/tests/groupby/conftest.py b/pandas/pandas/tests/groupby/conftest.py
new file mode 100644
index 0000000000000..2745f7c2b8d0f
--- /dev/null
+++ b/pandas/pandas/tests/groupby/conftest.py
@@ -0,0 +1,166 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+)
+from pandas.core.groupby.base import (
+    reduction_kernels,
+    transformation_kernels,
+)
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+
+
+@pytest.fixture
+def ts():
+    return Series(
+        np.random.default_rng(2).standard_normal(30),
+        index=date_range("2000-01-01", periods=30, freq="B"),
+    )
+
+
+@pytest.fixture
+def tsframe():
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=30, freq="B"),
+    )
+
+
+@pytest.fixture
+def three_group():
+    return DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+
+@pytest.fixture
+def slice_test_df():
+    data = [
+        [0, "a", "a0_at_0"],
+        [1, "b", "b0_at_1"],
+        [2, "a", "a1_at_2"],
+        [3, "b", "b1_at_3"],
+        [4, "c", "c0_at_4"],
+        [5, "a", "a2_at_5"],
+        [6, "a", "a3_at_6"],
+        [7, "a", "a4_at_7"],
+    ]
+    df = DataFrame(data, columns=["Index", "Group", "Value"])
+    return df.set_index("Index")
+
+
+@pytest.fixture
+def slice_test_grouped(slice_test_df):
+    return slice_test_df.groupby("Group", as_index=False)
+
+
+@pytest.fixture(params=sorted(reduction_kernels))
+def reduction_func(request):
+    """
+    yields the string names of all groupby reduction functions, one at a time.
+    """
+    return request.param
+
+
+@pytest.fixture(params=sorted(transformation_kernels))
+def transformation_func(request):
+    """yields the string names of all groupby transformation functions."""
+    return request.param
+
+
+@pytest.fixture(params=sorted(reduction_kernels) + sorted(transformation_kernels))
+def groupby_func(request):
+    """yields both aggregation and transformation functions."""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ("mean", {}),
+        ("var", {"ddof": 1}),
+        ("var", {"ddof": 0}),
+        ("std", {"ddof": 1}),
+        ("std", {"ddof": 0}),
+        ("sum", {}),
+        ("min", {}),
+        ("max", {}),
+        ("sum", {"min_count": 2}),
+        ("min", {"min_count": 2}),
+        ("max", {"min_count": 2}),
+    ],
+    ids=[
+        "mean",
+        "var_1",
+        "var_0",
+        "std_1",
+        "std_0",
+        "sum",
+        "min",
+        "max",
+        "sum-min_count",
+        "min-min_count",
+        "max-min_count",
+    ],
+)
+def numba_supported_reductions(request):
+    """reductions supported with engine='numba'"""
+    return request.param
diff --git a/pandas/pandas/tests/groupby/methods/__init__.py b/pandas/pandas/tests/groupby/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/groupby/methods/test_describe.py b/pandas/pandas/tests/groupby/methods/test_describe.py
new file mode 100644
index 0000000000000..6c4b913574d9e
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_describe.py
@@ -0,0 +1,271 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_apply_describe_bug(multiindex_dataframe_random_data):
+    grouped = multiindex_dataframe_random_data.groupby(level="first")
+    grouped.describe()  # it works!
+
+
+def test_series_describe_multikey():
+    ts = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+    grouped = ts.groupby([lambda x: x.year, lambda x: x.month])
+    result = grouped.describe()
+    tm.assert_series_equal(result["mean"], grouped.mean(), check_names=False)
+    tm.assert_series_equal(result["std"], grouped.std(), check_names=False)
+    tm.assert_series_equal(result["min"], grouped.min(), check_names=False)
+
+
+def test_series_describe_single():
+    ts = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+    grouped = ts.groupby(lambda x: x.month)
+    result = grouped.apply(lambda x: x.describe())
+    expected = grouped.describe().stack()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", ["key1", ["key1", "key2"]])
+def test_series_describe_as_index(as_index, keys):
+    # GH#49256
+    df = DataFrame(
+        {
+            "key1": ["one", "two", "two", "three", "two"],
+            "key2": ["one", "two", "two", "three", "two"],
+            "foo2": [1, 2, 4, 4, 6],
+        }
+    )
+    gb = df.groupby(keys, as_index=as_index)["foo2"]
+    result = gb.describe()
+    expected = DataFrame(
+        {
+            "key1": ["one", "three", "two"],
+            "count": [1.0, 1.0, 3.0],
+            "mean": [1.0, 4.0, 4.0],
+            "std": [np.nan, np.nan, 2.0],
+            "min": [1.0, 4.0, 2.0],
+            "25%": [1.0, 4.0, 3.0],
+            "50%": [1.0, 4.0, 4.0],
+            "75%": [1.0, 4.0, 5.0],
+            "max": [1.0, 4.0, 6.0],
+        }
+    )
+    if len(keys) == 2:
+        expected.insert(1, "key2", expected["key1"])
+    if as_index:
+        expected = expected.set_index(keys)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_describe_multikey(tsframe):
+    grouped = tsframe.groupby([lambda x: x.year, lambda x: x.month])
+    result = grouped.describe()
+    desc_groups = []
+    for col in tsframe:
+        group = grouped[col].describe()
+        # GH 17464 - Remove duplicate MultiIndex levels
+        group_col = MultiIndex(
+            levels=[Index([col], dtype=tsframe.columns.dtype), group.columns],
+            codes=[[0] * len(group.columns), range(len(group.columns))],
+        )
+        group = DataFrame(group.values, columns=group_col, index=group.index)
+        desc_groups.append(group)
+    expected = pd.concat(desc_groups, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_describe_tupleindex():
+    # GH 14848 - regression from 0.19.0 to 0.19.1
+    name = "k"
+    df = DataFrame(
+        {
+            "x": [1, 2, 3, 4, 5] * 3,
+            name: [(0, 0, 1), (0, 1, 0), (1, 0, 0)] * 5,
+        }
+    )
+    result = df.groupby(name).describe()
+    expected = DataFrame(
+        [[5.0, 3.0, 1.581139, 1.0, 2.0, 3.0, 4.0, 5.0]] * 3,
+        index=Index([(0, 0, 1), (0, 1, 0), (1, 0, 0)], tupleize_cols=False, name=name),
+        columns=MultiIndex.from_arrays(
+            [["x"] * 8, ["count", "mean", "std", "min", "25%", "50%", "75%", "max"]]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_describe_unstacked_format():
+    # GH 4792
+    prices = {
+        Timestamp("2011-01-06 10:59:05", tz=None): 24990,
+        Timestamp("2011-01-06 12:43:33", tz=None): 25499,
+        Timestamp("2011-01-06 12:54:09", tz=None): 25499,
+    }
+    volumes = {
+        Timestamp("2011-01-06 10:59:05", tz=None): 1500000000,
+        Timestamp("2011-01-06 12:43:33", tz=None): 5000000000,
+        Timestamp("2011-01-06 12:54:09", tz=None): 100000000,
+    }
+    df = DataFrame({"PRICE": prices, "VOLUME": volumes})
+    result = df.groupby("PRICE").VOLUME.describe()
+    data = [
+        df[df.PRICE == 24990].VOLUME.describe().values.tolist(),
+        df[df.PRICE == 25499].VOLUME.describe().values.tolist(),
+    ]
+    expected = DataFrame(
+        data,
+        index=Index([24990, 25499], name="PRICE"),
+        columns=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:"
+    "indexing past lexsort depth may impact performance:"
+    "pandas.errors.PerformanceWarning"
+)
+@pytest.mark.parametrize("keys", [["a1"], ["a1", "a2"]])
+def test_describe_with_duplicate_output_column_names(as_index, keys):
+    # GH 35314
+    df = DataFrame(
+        {
+            "a1": [99, 99, 99, 88, 88, 88],
+            "a2": [99, 99, 99, 88, 88, 88],
+            "b": [1, 2, 3, 4, 5, 6],
+            "c": [10, 20, 30, 40, 50, 60],
+        },
+        columns=["a1", "a2", "b", "b"],
+        copy=False,
+    )
+    if keys == ["a1"]:
+        df = df.drop(columns="a2")
+
+    expected = (
+        DataFrame.from_records(
+            [
+                ("b", "count", 3.0, 3.0),
+                ("b", "mean", 5.0, 2.0),
+                ("b", "std", 1.0, 1.0),
+                ("b", "min", 4.0, 1.0),
+                ("b", "25%", 4.5, 1.5),
+                ("b", "50%", 5.0, 2.0),
+                ("b", "75%", 5.5, 2.5),
+                ("b", "max", 6.0, 3.0),
+                ("b", "count", 3.0, 3.0),
+                ("b", "mean", 5.0, 2.0),
+                ("b", "std", 1.0, 1.0),
+                ("b", "min", 4.0, 1.0),
+                ("b", "25%", 4.5, 1.5),
+                ("b", "50%", 5.0, 2.0),
+                ("b", "75%", 5.5, 2.5),
+                ("b", "max", 6.0, 3.0),
+            ],
+        )
+        .set_index([0, 1])
+        .T
+    )
+    expected.columns.names = [None, None]
+    if len(keys) == 2:
+        expected.index = MultiIndex(
+            levels=[[88, 99], [88, 99]], codes=[[0, 1], [0, 1]], names=["a1", "a2"]
+        )
+    else:
+        expected.index = Index([88, 99], name="a1")
+
+    if not as_index:
+        expected = expected.reset_index()
+
+    result = df.groupby(keys, as_index=as_index).describe()
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_describe_duplicate_columns():
+    # GH#50806
+    df = DataFrame([[0, 1, 2, 3]])
+    df.columns = [0, 1, 2, 0]
+    gb = df.groupby(df[1])
+    result = gb.describe(percentiles=[])
+
+    columns = ["count", "mean", "std", "min", "50%", "max"]
+    frames = [
+        DataFrame([[1.0, val, np.nan, val, val, val]], index=[1], columns=columns)
+        for val in (0.0, 2.0, 3.0)
+    ]
+    expected = pd.concat(frames, axis=1)
+    expected.columns = MultiIndex(
+        levels=[[0, 2], columns],
+        codes=[6 * [0] + 6 * [1] + 6 * [0], 3 * list(range(6))],
+    )
+    expected.index.names = [1]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_describe_non_cython_paths():
+    # GH#5610 non-cython calls should not include the grouper
+    # Tests for code not expected to go through cython paths.
+    df = DataFrame(
+        [[1, 2, "foo"], [1, np.nan, "bar"], [3, np.nan, "baz"]],
+        columns=["A", "B", "C"],
+    )
+    gb = df.groupby("A")
+    expected_index = Index([1, 3], name="A")
+    expected_col = MultiIndex(
+        levels=[["B"], ["count", "mean", "std", "min", "25%", "50%", "75%", "max"]],
+        codes=[[0] * 8, list(range(8))],
+    )
+    expected = DataFrame(
+        [
+            [1.0, 2.0, np.nan, 2.0, 2.0, 2.0, 2.0, 2.0],
+            [0.0, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+        ],
+        index=expected_index,
+        columns=expected_col,
+    )
+    result = gb.describe()
+    tm.assert_frame_equal(result, expected)
+
+    gni = df.groupby("A", as_index=False)
+    expected = expected.reset_index()
+    result = gni.describe()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [int, float, object])
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"percentiles": [0.10, 0.20, 0.30], "include": "all", "exclude": None},
+        {"percentiles": [0.10, 0.20, 0.30], "include": None, "exclude": ["int"]},
+        {"percentiles": [0.10, 0.20, 0.30], "include": ["int"], "exclude": None},
+    ],
+)
+def test_groupby_empty_dataset(dtype, kwargs):
+    # GH#41575
+    df = DataFrame([[1, 2, 3]], columns=["A", "B", "C"], dtype=dtype)
+    df["B"] = df["B"].astype(int)
+    df["C"] = df["C"].astype(float)
+
+    result = df.iloc[:0].groupby("A").describe(**kwargs)
+    expected = df.groupby("A").describe(**kwargs).reset_index(drop=True).iloc[:0]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.iloc[:0].groupby("A").B.describe(**kwargs)
+    expected = df.groupby("A").B.describe(**kwargs).reset_index(drop=True).iloc[:0]
+    expected.index = Index([], dtype=df.columns.dtype)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_groupby_shift_diff.py b/pandas/pandas/tests/groupby/methods/test_groupby_shift_diff.py
new file mode 100644
index 0000000000000..1256046d81949
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_groupby_shift_diff.py
@@ -0,0 +1,250 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    NaT,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_group_shift_with_null_key():
+    # This test is designed to replicate the segfault in issue #13813.
+    n_rows = 1200
+
+    # Generate a moderately large dataframe with occasional missing
+    # values in column `B`, and then group by [`A`, `B`]. This should
+    # force `-1` in `labels` array of `g._grouper.group_info` exactly
+    # at those places, where the group-by key is partially missing.
+    df = DataFrame(
+        [(i % 12, i % 3 if i % 3 else np.nan, i) for i in range(n_rows)],
+        dtype=float,
+        columns=["A", "B", "Z"],
+        index=None,
+    )
+    g = df.groupby(["A", "B"])
+
+    expected = DataFrame(
+        [(i + 12 if i % 3 and i < n_rows - 12 else np.nan) for i in range(n_rows)],
+        dtype=float,
+        columns=["Z"],
+        index=None,
+    )
+    result = g.shift(-1)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_group_shift_with_fill_value():
+    # GH #24128
+    n_rows = 24
+    df = DataFrame(
+        [(i % 12, i % 3, i) for i in range(n_rows)],
+        dtype=float,
+        columns=["A", "B", "Z"],
+        index=None,
+    )
+    g = df.groupby(["A", "B"])
+
+    expected = DataFrame(
+        [(i + 12 if i < n_rows - 12 else 0) for i in range(n_rows)],
+        dtype=float,
+        columns=["Z"],
+        index=None,
+    )
+    result = g.shift(-1, fill_value=0)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_group_shift_lose_timezone():
+    # GH 30134
+    now_dt = Timestamp.now("UTC").as_unit("ns")
+    df = DataFrame({"a": [1, 1], "date": now_dt})
+    result = df.groupby("a").shift(0).iloc[0]
+    expected = Series({"date": now_dt}, name=result.name)
+    tm.assert_series_equal(result, expected)
+
+
+def test_group_diff_real_series(any_real_numpy_dtype):
+    df = DataFrame(
+        {"a": [1, 2, 3, 3, 2], "b": [1, 2, 3, 4, 5]},
+        dtype=any_real_numpy_dtype,
+    )
+    result = df.groupby("a")["b"].diff()
+    exp_dtype = "float"
+    if any_real_numpy_dtype in ["int8", "int16", "float32"]:
+        exp_dtype = "float32"
+    expected = Series([np.nan, np.nan, np.nan, 1.0, 3.0], dtype=exp_dtype, name="b")
+    tm.assert_series_equal(result, expected)
+
+
+def test_group_diff_real_frame(any_real_numpy_dtype):
+    df = DataFrame(
+        {
+            "a": [1, 2, 3, 3, 2],
+            "b": [1, 2, 3, 4, 5],
+            "c": [1, 2, 3, 4, 6],
+        },
+        dtype=any_real_numpy_dtype,
+    )
+    result = df.groupby("a").diff()
+    exp_dtype = "float"
+    if any_real_numpy_dtype in ["int8", "int16", "float32"]:
+        exp_dtype = "float32"
+    expected = DataFrame(
+        {
+            "b": [np.nan, np.nan, np.nan, 1.0, 3.0],
+            "c": [np.nan, np.nan, np.nan, 1.0, 4.0],
+        },
+        dtype=exp_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        [
+            Timestamp("2013-01-01"),
+            Timestamp("2013-01-02"),
+            Timestamp("2013-01-03"),
+        ],
+        [Timedelta("5 days"), Timedelta("6 days"), Timedelta("7 days")],
+    ],
+)
+def test_group_diff_datetimelike(data, unit):
+    df = DataFrame({"a": [1, 2, 2], "b": data})
+    df["b"] = df["b"].dt.as_unit(unit)
+    result = df.groupby("a")["b"].diff()
+    expected = Series([NaT, NaT, Timedelta("1 days")], name="b").dt.as_unit(unit)
+    tm.assert_series_equal(result, expected)
+
+
+def test_group_diff_bool():
+    df = DataFrame({"a": [1, 2, 3, 3, 2], "b": [True, True, False, False, True]})
+    result = df.groupby("a")["b"].diff()
+    expected = Series([np.nan, np.nan, np.nan, False, False], name="b")
+    tm.assert_series_equal(result, expected)
+
+
+def test_group_diff_object_raises(object_dtype):
+    df = DataFrame(
+        {"a": ["foo", "bar", "bar"], "b": ["baz", "foo", "foo"]}, dtype=object_dtype
+    )
+    with pytest.raises(TypeError, match=r"unsupported operand type\(s\) for -"):
+        df.groupby("a")["b"].diff()
+
+
+def test_empty_shift_with_fill():
+    # GH 41264, single-index check
+    df = DataFrame(columns=["a", "b", "c"])
+    shifted = df.groupby(["a"]).shift(1)
+    shifted_with_fill = df.groupby(["a"]).shift(1, fill_value=0)
+    tm.assert_frame_equal(shifted, shifted_with_fill)
+    tm.assert_index_equal(shifted.index, shifted_with_fill.index)
+
+
+def test_multindex_empty_shift_with_fill():
+    # GH 41264, multi-index check
+    df = DataFrame(columns=["a", "b", "c"])
+    shifted = df.groupby(["a", "b"]).shift(1)
+    shifted_with_fill = df.groupby(["a", "b"]).shift(1, fill_value=0)
+    tm.assert_frame_equal(shifted, shifted_with_fill)
+    tm.assert_index_equal(shifted.index, shifted_with_fill.index)
+
+
+def test_shift_periods_freq():
+    # GH 54093
+    data = {"a": [1, 2, 3, 4, 5, 6], "b": [0, 0, 0, 1, 1, 1]}
+    df = DataFrame(data, index=date_range(start="20100101", periods=6))
+    result = df.groupby(df.index).shift(periods=-2, freq="D")
+    expected = DataFrame(data, index=date_range(start="2009-12-30", periods=6))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_shift_disallow_freq_and_fill_value():
+    # GH 53832
+    data = {"a": [1, 2, 3, 4, 5, 6], "b": [0, 0, 0, 1, 1, 1]}
+    df = DataFrame(data, index=date_range(start="20100101", periods=6))
+    msg = "Passing a 'freq' together with a 'fill_value'"
+    with pytest.raises(ValueError, match=msg):
+        df.groupby(df.index).shift(periods=-2, freq="D", fill_value="1")
+
+
+def test_shift_disallow_suffix_if_periods_is_int():
+    # GH#44424
+    data = {"a": [1, 2, 3, 4, 5, 6], "b": [0, 0, 0, 1, 1, 1]}
+    df = DataFrame(data)
+    msg = "Cannot specify `suffix` if `periods` is an int."
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("b").shift(1, suffix="fails")
+
+
+def test_group_shift_with_multiple_periods():
+    # GH#44424
+    df = DataFrame({"a": [1, 2, 3, 3, 2], "b": [True, True, False, False, True]})
+
+    shifted_df = df.groupby("b")[["a"]].shift([0, 1])
+    expected_df = DataFrame(
+        {"a_0": [1, 2, 3, 3, 2], "a_1": [np.nan, 1.0, np.nan, 3.0, 2.0]}
+    )
+    tm.assert_frame_equal(shifted_df, expected_df)
+
+    # series
+    shifted_series = df.groupby("b")["a"].shift([0, 1])
+    tm.assert_frame_equal(shifted_series, expected_df)
+
+
+def test_group_shift_with_multiple_periods_and_freq():
+    # GH#44424
+    df = DataFrame(
+        {"a": [1, 2, 3, 4, 5], "b": [True, True, False, False, True]},
+        index=date_range("1/1/2000", periods=5, freq="h"),
+    )
+    shifted_df = df.groupby("b")[["a"]].shift(
+        [0, 1],
+        freq="h",
+    )
+    expected_df = DataFrame(
+        {
+            "a_0": [1.0, 2.0, 3.0, 4.0, 5.0, np.nan],
+            "a_1": [
+                np.nan,
+                1.0,
+                2.0,
+                3.0,
+                4.0,
+                5.0,
+            ],
+        },
+        index=date_range("1/1/2000", periods=6, freq="h"),
+    )
+    tm.assert_frame_equal(shifted_df, expected_df)
+
+
+def test_group_shift_with_multiple_periods_and_fill_value():
+    # GH#44424
+    df = DataFrame(
+        {"a": [1, 2, 3, 4, 5], "b": [True, True, False, False, True]},
+    )
+    shifted_df = df.groupby("b")[["a"]].shift([0, 1], fill_value=-1)
+    expected_df = DataFrame(
+        {"a_0": [1, 2, 3, 4, 5], "a_1": [-1, 1, -1, 3, 2]},
+    )
+    tm.assert_frame_equal(shifted_df, expected_df)
+
+
+def test_group_shift_with_multiple_periods_and_both_fill_and_freq_deprecated():
+    # GH#44424
+    df = DataFrame(
+        {"a": [1, 2, 3, 4, 5], "b": [True, True, False, False, True]},
+        index=date_range("1/1/2000", periods=5, freq="h"),
+    )
+    msg = "Passing a 'freq' together with a 'fill_value'"
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("b")[["a"]].shift([1, 2], fill_value=1, freq="h")
diff --git a/pandas/pandas/tests/groupby/methods/test_is_monotonic.py b/pandas/pandas/tests/groupby/methods/test_is_monotonic.py
new file mode 100644
index 0000000000000..3428fc90f6e51
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_is_monotonic.py
@@ -0,0 +1,78 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "in_vals, out_vals",
+    [
+        # Basics: strictly increasing (T), strictly decreasing (F),
+        # abs val increasing (F), non-strictly increasing (T)
+        ([1, 2, 5, 3, 2, 0, 4, 5, -6, 1, 1], [True, False, False, True]),
+        # Test with inf vals
+        (
+            [1, 2.1, np.inf, 3, 2, np.inf, -np.inf, 5, 11, 1, -np.inf],
+            [True, False, True, False],
+        ),
+        # Test with nan vals; should always be False
+        (
+            [1, 2, np.nan, 3, 2, np.nan, np.nan, 5, -np.inf, 1, np.nan],
+            [False, False, False, False],
+        ),
+    ],
+)
+def test_is_monotonic_increasing(in_vals, out_vals):
+    # GH 17015
+    source_dict = {
+        "A": ["1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11"],
+        "B": ["a", "a", "a", "b", "b", "b", "c", "c", "c", "d", "d"],
+        "C": in_vals,
+    }
+    df = DataFrame(source_dict)
+    result = df.groupby("B").C.is_monotonic_increasing
+    index = Index(list("abcd"), name="B")
+    expected = Series(index=index, data=out_vals, name="C")
+    tm.assert_series_equal(result, expected)
+
+    # Also check result equal to manually taking x.is_monotonic_increasing.
+    expected = df.groupby(["B"]).C.apply(lambda x: x.is_monotonic_increasing)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "in_vals, out_vals",
+    [
+        # Basics: strictly decreasing (T), strictly increasing (F),
+        # abs val decreasing (F), non-strictly increasing (T)
+        ([10, 9, 7, 3, 4, 5, -3, 2, 0, 1, 1], [True, False, False, True]),
+        # Test with inf vals
+        (
+            [np.inf, 1, -np.inf, np.inf, 2, -3, -np.inf, 5, -3, -np.inf, -np.inf],
+            [True, True, False, True],
+        ),
+        # Test with nan vals; should always be False
+        (
+            [1, 2, np.nan, 3, 2, np.nan, np.nan, 5, -np.inf, 1, np.nan],
+            [False, False, False, False],
+        ),
+    ],
+)
+def test_is_monotonic_decreasing(in_vals, out_vals):
+    # GH 17015
+    source_dict = {
+        "A": ["1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11"],
+        "B": ["a", "a", "a", "b", "b", "b", "c", "c", "c", "d", "d"],
+        "C": in_vals,
+    }
+
+    df = DataFrame(source_dict)
+    result = df.groupby("B").C.is_monotonic_decreasing
+    index = Index(list("abcd"), name="B")
+    expected = Series(index=index, data=out_vals, name="C")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_kurt.py b/pandas/pandas/tests/groupby/methods/test_kurt.py
new file mode 100644
index 0000000000000..21b7c50c3c5aa
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_kurt.py
@@ -0,0 +1,90 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_groupby_kurt_equivalence():
+    # GH#40139
+    # Test that that groupby kurt method (which uses libgroupby.group_kurt)
+    #  matches the results of operating group-by-group (which uses nanops.nankurt)
+    nrows = 1000
+    ngroups = 3
+    ncols = 2
+    nan_frac = 0.05
+
+    arr = np.random.default_rng(2).standard_normal((nrows, ncols))
+    arr[np.random.default_rng(2).random(nrows) < nan_frac] = np.nan
+
+    df = pd.DataFrame(arr)
+    grps = np.random.default_rng(2).integers(0, ngroups, size=nrows)
+    gb = df.groupby(grps)
+
+    result = gb.kurt()
+
+    grpwise = [grp.kurt().to_frame(i).T for i, grp in gb]
+    expected = pd.concat(grpwise, axis=0)
+    expected.index = expected.index.astype("int64")  # 32bit builds
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        "Float64",
+    ],
+)
+def test_groupby_kurt_arrow_float64(dtype):
+    # GH#40139
+    # Test groupby.kurt() with float64[pyarrow] and Float64 dtypes
+    df = pd.DataFrame(
+        {
+            "x": [1.0, np.nan, 3.2, 4.8, 2.3, 1.9, 8.9],
+            "y": [1.6, 3.3, 3.2, 6.8, 1.3, 2.9, 9.0],
+        },
+        dtype=dtype,
+    )
+    gb = df.groupby(by=lambda x: 0)
+
+    result = gb.kurt()
+    expected = pd.DataFrame({"x": [2.1644713], "y": [0.1513969]}, dtype=dtype)
+    tm.assert_almost_equal(result, expected)
+
+
+def test_groupby_kurt_noskipna():
+    # GH#40139
+    # Test groupby.kurt() with skipna = False
+    df = pd.DataFrame(
+        {
+            "x": [1.0, np.nan, 3.2, 4.8, 2.3, 1.9, 8.9],
+            "y": [1.6, 3.3, 3.2, 6.8, 1.3, 2.9, 9.0],
+        }
+    )
+    gb = df.groupby(by=lambda x: 0)
+
+    result = gb.kurt(skipna=False)
+    expected = pd.DataFrame({"x": [np.nan], "y": [0.1513969]})
+    tm.assert_almost_equal(result, expected)
+
+
+def test_groupby_kurt_all_ones():
+    # GH#40139
+    # Test groupby.kurt() with constant values
+    df = pd.DataFrame(
+        {
+            "x": [1.0] * 10,
+        }
+    )
+    gb = df.groupby(by=lambda x: 0)
+
+    result = gb.kurt(skipna=False)
+    expected = pd.DataFrame(
+        {
+            "x": [0.0],  # Same behavior as pd.DataFrame.kurt()
+        }
+    )
+    tm.assert_almost_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_nlargest_nsmallest.py b/pandas/pandas/tests/groupby/methods/test_nlargest_nsmallest.py
new file mode 100644
index 0000000000000..1225c325c4c23
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_nlargest_nsmallest.py
@@ -0,0 +1,114 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_nlargest():
+    a = Series([1, 3, 5, 7, 2, 9, 0, 4, 6, 10])
+    b = Series(list("a" * 5 + "b" * 5))
+    gb = a.groupby(b)
+    r = gb.nlargest(3)
+    e = Series(
+        [7, 5, 3, 10, 9, 6],
+        index=MultiIndex.from_arrays([list("aaabbb"), [3, 2, 1, 9, 5, 8]]),
+    )
+    tm.assert_series_equal(r, e)
+
+    a = Series([1, 1, 3, 2, 0, 3, 3, 2, 1, 0])
+    gb = a.groupby(b)
+    e = Series(
+        [3, 2, 1, 3, 3, 2],
+        index=MultiIndex.from_arrays([list("aaabbb"), [2, 3, 1, 6, 5, 7]]),
+    )
+    tm.assert_series_equal(gb.nlargest(3, keep="last"), e)
+
+
+def test_nlargest_mi_grouper():
+    # see gh-21411
+    npr = np.random.default_rng(2)
+
+    dts = date_range("20180101", periods=10)
+    iterables = [dts, ["one", "two"]]
+
+    idx = MultiIndex.from_product(iterables, names=["first", "second"])
+    s = Series(npr.standard_normal(20), index=idx)
+
+    result = s.groupby("first").nlargest(1)
+
+    exp_idx = MultiIndex.from_tuples(
+        [
+            (dts[0], dts[0], "one"),
+            (dts[1], dts[1], "one"),
+            (dts[2], dts[2], "one"),
+            (dts[3], dts[3], "two"),
+            (dts[4], dts[4], "one"),
+            (dts[5], dts[5], "one"),
+            (dts[6], dts[6], "one"),
+            (dts[7], dts[7], "one"),
+            (dts[8], dts[8], "one"),
+            (dts[9], dts[9], "one"),
+        ],
+        names=["first", "first", "second"],
+    )
+
+    exp_values = [
+        0.18905338179353307,
+        -0.41306354339189344,
+        1.799707382720902,
+        0.7738065867276614,
+        0.28121066979764925,
+        0.9775674511260357,
+        -0.3288239040579627,
+        0.45495807124085547,
+        0.5452887139646817,
+        0.12682784711186987,
+    ]
+
+    expected = Series(exp_values, index=exp_idx)
+    tm.assert_series_equal(result, expected, check_exact=False, rtol=1e-3)
+
+
+def test_nsmallest():
+    a = Series([1, 3, 5, 7, 2, 9, 0, 4, 6, 10])
+    b = Series(list("a" * 5 + "b" * 5))
+    gb = a.groupby(b)
+    r = gb.nsmallest(3)
+    e = Series(
+        [1, 2, 3, 0, 4, 6],
+        index=MultiIndex.from_arrays([list("aaabbb"), [0, 4, 1, 6, 7, 8]]),
+    )
+    tm.assert_series_equal(r, e)
+
+    a = Series([1, 1, 3, 2, 0, 3, 3, 2, 1, 0])
+    gb = a.groupby(b)
+    e = Series(
+        [0, 1, 1, 0, 1, 2],
+        index=MultiIndex.from_arrays([list("aaabbb"), [4, 1, 0, 9, 8, 7]]),
+    )
+    tm.assert_series_equal(gb.nsmallest(3, keep="last"), e)
+
+
+@pytest.mark.parametrize(
+    "data, groups",
+    [([0, 1, 2, 3], [0, 0, 1, 1]), ([0], [0])],
+)
+@pytest.mark.parametrize("dtype", [None, *tm.ALL_INT_NUMPY_DTYPES])
+def test_nlargest_and_smallest_noop(data, groups, dtype, nselect_method):
+    # GH 15272, GH 16345, GH 29129
+    # Test nlargest/smallest when it results in a noop,
+    # i.e. input is sorted and group size <= n
+    if dtype is not None:
+        data = np.array(data, dtype=dtype)
+    if nselect_method == "nlargest":
+        data = list(reversed(data))
+    ser = Series(data, name="a")
+    result = getattr(ser.groupby(groups), nselect_method)(n=2)
+    expidx = np.array(groups, dtype=int) if isinstance(groups, list) else groups
+    expected = Series(data, index=MultiIndex.from_arrays([expidx, ser.index]), name="a")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_nth.py b/pandas/pandas/tests/groupby/methods/test_nth.py
new file mode 100644
index 0000000000000..d9c8706ec9202
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_nth.py
@@ -0,0 +1,848 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    isna,
+)
+import pandas._testing as tm
+
+
+def test_first_last_nth(df):
+    # tests for first / last / nth
+    grouped = df.groupby("A")
+    first = grouped.first()
+    expected = df.loc[[1, 0], ["B", "C", "D"]]
+    expected.index = Index(["bar", "foo"], name="A")
+    expected = expected.sort_index()
+    tm.assert_frame_equal(first, expected)
+
+    nth = grouped.nth(0)
+    expected = df.loc[[0, 1]]
+    tm.assert_frame_equal(nth, expected)
+
+    last = grouped.last()
+    expected = df.loc[[5, 7], ["B", "C", "D"]]
+    expected.index = Index(["bar", "foo"], name="A")
+    tm.assert_frame_equal(last, expected)
+
+    nth = grouped.nth(-1)
+    expected = df.iloc[[5, 7]]
+    tm.assert_frame_equal(nth, expected)
+
+    nth = grouped.nth(1)
+    expected = df.iloc[[2, 3]]
+    tm.assert_frame_equal(nth, expected)
+
+    # it works!
+    grouped["B"].first()
+    grouped["B"].last()
+    grouped["B"].nth(0)
+
+    df = df.copy()
+    df.loc[df["A"] == "foo", "B"] = np.nan
+    grouped = df.groupby("A")
+    assert isna(grouped["B"].first()["foo"])
+    assert isna(grouped["B"].last()["foo"])
+    assert isna(grouped["B"].nth(0).iloc[0])
+
+    # v0.14.0 whatsnew
+    df = DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=["A", "B"])
+    g = df.groupby("A")
+    result = g.first()
+    expected = df.iloc[[1, 2]].set_index("A")
+    tm.assert_frame_equal(result, expected)
+
+    expected = df.iloc[[1, 2]]
+    result = g.nth(0, dropna="any")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["first", "last"])
+def test_first_last_with_na_object(method, nulls_fixture):
+    # https://github.com/pandas-dev/pandas/issues/32123
+    groups = DataFrame({"a": [1, 1, 2, 2], "b": [1, 2, 3, nulls_fixture]}).groupby("a")
+    result = getattr(groups, method)()
+
+    if method == "first":
+        values = [1, 3]
+    else:
+        values = [2, 3]
+
+    values = np.array(values, dtype=result["b"].dtype)
+    idx = Index([1, 2], name="a")
+    expected = DataFrame({"b": values}, index=idx)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index", [0, -1])
+def test_nth_with_na_object(index, nulls_fixture):
+    # https://github.com/pandas-dev/pandas/issues/32123
+    df = DataFrame({"a": [1, 1, 2, 2], "b": [1, 2, 3, nulls_fixture]})
+    groups = df.groupby("a")
+    result = groups.nth(index)
+    expected = df.iloc[[0, 2]] if index == 0 else df.iloc[[1, 3]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["first", "last"])
+def test_first_last_with_None(method):
+    # https://github.com/pandas-dev/pandas/issues/32800
+    # None should be preserved as object dtype
+    df = DataFrame.from_dict({"id": ["a"], "value": [None]})
+    groups = df.groupby("id", as_index=False)
+    result = getattr(groups, method)()
+
+    tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("method", ["first", "last"])
+@pytest.mark.parametrize(
+    "df, expected",
+    [
+        (
+            DataFrame({"id": "a", "value": [None, "foo", np.nan]}),
+            DataFrame({"value": ["foo"]}, index=Index(["a"], name="id")),
+        ),
+        (
+            DataFrame({"id": "a", "value": [np.nan]}, dtype=object),
+            DataFrame({"value": [None]}, index=Index(["a"], name="id")),
+        ),
+    ],
+)
+def test_first_last_with_None_expanded(method, df, expected):
+    # GH 32800, 38286
+    result = getattr(df.groupby("id"), method)()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_first_last_nth_dtypes():
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.array(np.random.default_rng(2).standard_normal(8), dtype="float32"),
+        }
+    )
+    df["E"] = True
+    df["F"] = 1
+
+    # tests for first / last / nth
+    grouped = df.groupby("A")
+    first = grouped.first()
+    expected = df.loc[[1, 0], ["B", "C", "D", "E", "F"]]
+    expected.index = Index(["bar", "foo"], name="A")
+    expected = expected.sort_index()
+    tm.assert_frame_equal(first, expected)
+
+    last = grouped.last()
+    expected = df.loc[[5, 7], ["B", "C", "D", "E", "F"]]
+    expected.index = Index(["bar", "foo"], name="A")
+    expected = expected.sort_index()
+    tm.assert_frame_equal(last, expected)
+
+    nth = grouped.nth(1)
+    expected = df.iloc[[2, 3]]
+    tm.assert_frame_equal(nth, expected)
+
+
+def test_first_last_nth_dtypes2():
+    # GH 2763, first/last shifting dtypes
+    idx = list(range(10))
+    idx.append(9)
+    ser = Series(data=range(11), index=idx, name="IntCol")
+    assert ser.dtype == "int64"
+    f = ser.groupby(level=0).first()
+    assert f.dtype == "int64"
+
+
+def test_first_last_nth_nan_dtype():
+    # GH 33591
+    df = DataFrame({"data": ["A"], "nans": Series([None], dtype=object)})
+    grouped = df.groupby("data")
+
+    expected = df.set_index("data").nans
+    tm.assert_series_equal(grouped.nans.first(), expected)
+    tm.assert_series_equal(grouped.nans.last(), expected)
+
+    expected = df.nans
+    tm.assert_series_equal(grouped.nans.nth(-1), expected)
+    tm.assert_series_equal(grouped.nans.nth(0), expected)
+
+
+def test_first_strings_timestamps():
+    # GH 11244
+    test = DataFrame(
+        {
+            Timestamp("2012-01-01 00:00:00"): ["a", "b"],
+            Timestamp("2012-01-02 00:00:00"): ["c", "d"],
+            "name": ["e", "e"],
+            "aaaa": ["f", "g"],
+        }
+    )
+    result = test.groupby("name").first()
+    expected = DataFrame(
+        [["a", "c", "f"]],
+        columns=Index([Timestamp("2012-01-01"), Timestamp("2012-01-02"), "aaaa"]),
+        index=Index(["e"], name="name"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nth():
+    df = DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=["A", "B"])
+    gb = df.groupby("A")
+
+    tm.assert_frame_equal(gb.nth(0), df.iloc[[0, 2]])
+    tm.assert_frame_equal(gb.nth(1), df.iloc[[1]])
+    tm.assert_frame_equal(gb.nth(2), df.loc[[]])
+    tm.assert_frame_equal(gb.nth(-1), df.iloc[[1, 2]])
+    tm.assert_frame_equal(gb.nth(-2), df.iloc[[0]])
+    tm.assert_frame_equal(gb.nth(-3), df.loc[[]])
+    tm.assert_series_equal(gb.B.nth(0), df.B.iloc[[0, 2]])
+    tm.assert_series_equal(gb.B.nth(1), df.B.iloc[[1]])
+    tm.assert_frame_equal(gb[["B"]].nth(0), df[["B"]].iloc[[0, 2]])
+
+    tm.assert_frame_equal(gb.nth(0, dropna="any"), df.iloc[[1, 2]])
+    tm.assert_frame_equal(gb.nth(-1, dropna="any"), df.iloc[[1, 2]])
+
+    tm.assert_frame_equal(gb.nth(7, dropna="any"), df.iloc[:0])
+    tm.assert_frame_equal(gb.nth(2, dropna="any"), df.iloc[:0])
+
+
+def test_nth2():
+    # out of bounds, regression from 0.13.1
+    # GH 6621
+    df = DataFrame(
+        {
+            "color": {0: "green", 1: "green", 2: "red", 3: "red", 4: "red"},
+            "food": {0: "ham", 1: "eggs", 2: "eggs", 3: "ham", 4: "pork"},
+            "two": {
+                0: 1.5456590000000001,
+                1: -0.070345000000000005,
+                2: -2.4004539999999999,
+                3: 0.46206000000000003,
+                4: 0.52350799999999997,
+            },
+            "one": {
+                0: 0.56573799999999996,
+                1: -0.9742360000000001,
+                2: 1.033801,
+                3: -0.78543499999999999,
+                4: 0.70422799999999997,
+            },
+        }
+    ).set_index(["color", "food"])
+
+    result = df.groupby(level=0, as_index=False).nth(2)
+    expected = df.iloc[[-1]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(level=0, as_index=False).nth(3)
+    expected = df.loc[[]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nth3():
+    # GH 7559
+    # from the vbench
+    df = DataFrame(np.random.default_rng(2).integers(1, 10, (100, 2)), dtype="int64")
+    ser = df[1]
+    gb = df[0]
+    expected = ser.groupby(gb).first()
+    expected2 = ser.groupby(gb).apply(lambda x: x.iloc[0])
+    tm.assert_series_equal(expected2, expected, check_names=False)
+    assert expected.name == 1
+    assert expected2.name == 1
+
+    # validate first
+    v = ser[gb == 1].iloc[0]
+    assert expected.iloc[0] == v
+    assert expected2.iloc[0] == v
+
+    with pytest.raises(ValueError, match="For a DataFrame"):
+        ser.groupby(gb, sort=False).nth(0, dropna=True)
+
+
+def test_nth4():
+    # doc example
+    df = DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=["A", "B"])
+    gb = df.groupby("A")
+    result = gb.B.nth(0, dropna="all")
+    expected = df.B.iloc[[1, 2]]
+    tm.assert_series_equal(result, expected)
+
+
+def test_nth5():
+    # test multiple nth values
+    df = DataFrame([[1, np.nan], [1, 3], [1, 4], [5, 6], [5, 7]], columns=["A", "B"])
+    gb = df.groupby("A")
+
+    tm.assert_frame_equal(gb.nth(0), df.iloc[[0, 3]])
+    tm.assert_frame_equal(gb.nth([0]), df.iloc[[0, 3]])
+    tm.assert_frame_equal(gb.nth([0, 1]), df.iloc[[0, 1, 3, 4]])
+    tm.assert_frame_equal(gb.nth([0, -1]), df.iloc[[0, 2, 3, 4]])
+    tm.assert_frame_equal(gb.nth([0, 1, 2]), df.iloc[[0, 1, 2, 3, 4]])
+    tm.assert_frame_equal(gb.nth([0, 1, -1]), df.iloc[[0, 1, 2, 3, 4]])
+    tm.assert_frame_equal(gb.nth([2]), df.iloc[[2]])
+    tm.assert_frame_equal(gb.nth([3, 4]), df.loc[[]])
+
+
+def test_nth_bdays(unit):
+    business_dates = pd.date_range(
+        start="4/1/2014", end="6/30/2014", freq="B", unit=unit
+    )
+    df = DataFrame(1, index=business_dates, columns=["a", "b"])
+    # get the first, fourth and last two business days for each month
+    key = [df.index.year, df.index.month]
+    result = df.groupby(key, as_index=False).nth([0, 3, -2, -1])
+    expected_dates = pd.to_datetime(
+        [
+            "2014/4/1",
+            "2014/4/4",
+            "2014/4/29",
+            "2014/4/30",
+            "2014/5/1",
+            "2014/5/6",
+            "2014/5/29",
+            "2014/5/30",
+            "2014/6/2",
+            "2014/6/5",
+            "2014/6/27",
+            "2014/6/30",
+        ]
+    ).as_unit(unit)
+    expected = DataFrame(1, columns=["a", "b"], index=expected_dates)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nth_multi_grouper(three_group):
+    # PR 9090, related to issue 8979
+    # test nth on multiple groupers
+    grouped = three_group.groupby(["A", "B"])
+    result = grouped.nth(0)
+    expected = three_group.iloc[[0, 3, 4, 7]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected_first, expected_last",
+    [
+        (
+            {
+                "id": ["A"],
+                "time": Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                "foo": [1],
+            },
+            {
+                "id": ["A"],
+                "time": Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                "foo": [1],
+            },
+            {
+                "id": ["A"],
+                "time": Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                "foo": [1],
+            },
+        ),
+        (
+            {
+                "id": ["A", "B", "A"],
+                "time": [
+                    Timestamp("2012-01-01 13:00:00", tz="America/New_York"),
+                    Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                    Timestamp("2012-03-01 12:00:00", tz="Europe/London"),
+                ],
+                "foo": [1, 2, 3],
+            },
+            {
+                "id": ["A", "B"],
+                "time": [
+                    Timestamp("2012-01-01 13:00:00", tz="America/New_York"),
+                    Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                ],
+                "foo": [1, 2],
+            },
+            {
+                "id": ["A", "B"],
+                "time": [
+                    Timestamp("2012-03-01 12:00:00", tz="Europe/London"),
+                    Timestamp("2012-02-01 14:00:00", tz="US/Central"),
+                ],
+                "foo": [3, 2],
+            },
+        ),
+    ],
+)
+def test_first_last_tz(data, expected_first, expected_last):
+    # GH15884
+    # Test that the timezone is retained when calling first
+    # or last on groupby with as_index=False
+
+    df = DataFrame(data)
+
+    result = df.groupby("id", as_index=False).first()
+    expected = DataFrame(expected_first)
+    cols = ["id", "time", "foo"]
+    tm.assert_frame_equal(result[cols], expected[cols])
+
+    result = df.groupby("id", as_index=False)["time"].first()
+    tm.assert_frame_equal(result, expected[["id", "time"]])
+
+    result = df.groupby("id", as_index=False).last()
+    expected = DataFrame(expected_last)
+    cols = ["id", "time", "foo"]
+    tm.assert_frame_equal(result[cols], expected[cols])
+
+    result = df.groupby("id", as_index=False)["time"].last()
+    tm.assert_frame_equal(result, expected[["id", "time"]])
+
+
+@pytest.mark.parametrize(
+    "method, ts, alpha",
+    [
+        ["first", Timestamp("2013-01-01", tz="US/Eastern"), "a"],
+        ["last", Timestamp("2013-01-02", tz="US/Eastern"), "b"],
+    ],
+)
+def test_first_last_tz_multi_column(method, ts, alpha, unit):
+    # GH 21603
+    category_string = Series(list("abc")).astype("category")
+    dti = pd.date_range("20130101", periods=3, tz="US/Eastern", unit=unit)
+    df = DataFrame(
+        {
+            "group": [1, 1, 2],
+            "category_string": category_string,
+            "datetimetz": dti,
+        }
+    )
+    result = getattr(df.groupby("group"), method)()
+    expected = DataFrame(
+        {
+            "category_string": pd.Categorical(
+                [alpha, "c"], dtype=category_string.dtype
+            ),
+            "datetimetz": [ts, Timestamp("2013-01-03", tz="US/Eastern")],
+        },
+        index=Index([1, 2], name="group"),
+    )
+    expected["datetimetz"] = expected["datetimetz"].dt.as_unit(unit)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        pd.array([True, False], dtype="boolean"),
+        pd.array([1, 2], dtype="Int64"),
+        pd.to_datetime(["2020-01-01", "2020-02-01"]),
+        pd.to_timedelta([1, 2], unit="D"),
+    ],
+)
+@pytest.mark.parametrize("function", ["first", "last", "min", "max"])
+def test_first_last_extension_array_keeps_dtype(values, function):
+    # https://github.com/pandas-dev/pandas/issues/33071
+    # https://github.com/pandas-dev/pandas/issues/32194
+    df = DataFrame({"a": [1, 2], "b": values})
+    grouped = df.groupby("a")
+    idx = Index([1, 2], name="a")
+    expected_series = Series(values, name="b", index=idx)
+    expected_frame = DataFrame({"b": values}, index=idx)
+
+    result_series = getattr(grouped["b"], function)()
+    tm.assert_series_equal(result_series, expected_series)
+
+    result_frame = grouped.agg({"b": function})
+    tm.assert_frame_equal(result_frame, expected_frame)
+
+
+def test_nth_multi_index_as_expected():
+    # PR 9090, related to issue 8979
+    # test nth on MultiIndex
+    three_group = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+        }
+    )
+    grouped = three_group.groupby(["A", "B"])
+    result = grouped.nth(0)
+    expected = three_group.iloc[[0, 3, 4, 7]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op, n, expected_rows",
+    [
+        ("head", -1, [0]),
+        ("head", 0, []),
+        ("head", 1, [0, 2]),
+        ("head", 7, [0, 1, 2]),
+        ("tail", -1, [1]),
+        ("tail", 0, []),
+        ("tail", 1, [1, 2]),
+        ("tail", 7, [0, 1, 2]),
+    ],
+)
+@pytest.mark.parametrize("columns", [None, [], ["A"], ["B"], ["A", "B"]])
+def test_groupby_head_tail(op, n, expected_rows, columns, as_index):
+    df = DataFrame([[1, 2], [1, 4], [5, 6]], columns=["A", "B"])
+    g = df.groupby("A", as_index=as_index)
+    expected = df.iloc[expected_rows]
+    if columns is not None:
+        g = g[columns]
+        expected = expected[columns]
+    result = getattr(g, op)(n)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_group_selection_cache():
+    # GH 12839 nth, head, and tail should return same result consistently
+    df = DataFrame([[1, 2], [1, 4], [5, 6]], columns=["A", "B"])
+    expected = df.iloc[[0, 2]]
+
+    g = df.groupby("A")
+    result1 = g.head(n=2)
+    result2 = g.nth(0)
+    tm.assert_frame_equal(result1, df)
+    tm.assert_frame_equal(result2, expected)
+
+    g = df.groupby("A")
+    result1 = g.tail(n=2)
+    result2 = g.nth(0)
+    tm.assert_frame_equal(result1, df)
+    tm.assert_frame_equal(result2, expected)
+
+    g = df.groupby("A")
+    result1 = g.nth(0)
+    result2 = g.head(n=2)
+    tm.assert_frame_equal(result1, expected)
+    tm.assert_frame_equal(result2, df)
+
+    g = df.groupby("A")
+    result1 = g.nth(0)
+    result2 = g.tail(n=2)
+    tm.assert_frame_equal(result1, expected)
+    tm.assert_frame_equal(result2, df)
+
+
+def test_nth_empty():
+    # GH 16064
+    df = DataFrame(index=[0], columns=["a", "b", "c"])
+    result = df.groupby("a").nth(10)
+    expected = df.iloc[:0]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(["a", "b"]).nth(10)
+    expected = df.iloc[:0]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nth_column_order():
+    # GH 20760
+    # Check that nth preserves column order
+    df = DataFrame(
+        [[1, "b", 100], [1, "a", 50], [1, "a", np.nan], [2, "c", 200], [2, "d", 150]],
+        columns=["A", "C", "B"],
+    )
+    result = df.groupby("A").nth(0)
+    expected = df.iloc[[0, 3]]
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("A").nth(-1, dropna="any")
+    expected = df.iloc[[1, 4]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dropna", [None, "any", "all"])
+def test_nth_nan_in_grouper(dropna):
+    # GH 26011
+    df = DataFrame(
+        {
+            "a": [np.nan, "a", np.nan, "b", np.nan],
+            "b": [0, 2, 4, 6, 8],
+            "c": [1, 3, 5, 7, 9],
+        }
+    )
+    result = df.groupby("a").nth(0, dropna=dropna)
+    expected = df.iloc[[1, 3]]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dropna", [None, "any", "all"])
+def test_nth_nan_in_grouper_series(dropna):
+    # GH 26454
+    df = DataFrame(
+        {
+            "a": [np.nan, "a", np.nan, "b", np.nan],
+            "b": [0, 2, 4, 6, 8],
+        }
+    )
+    result = df.groupby("a")["b"].nth(0, dropna=dropna)
+    expected = df["b"].iloc[[1, 3]]
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_first_categorical_and_datetime_data_nat():
+    # GH 20520
+    df = DataFrame(
+        {
+            "group": ["first", "first", "second", "third", "third"],
+            "time": 5 * [np.datetime64("NaT")],
+            "categories": Series(["a", "b", "c", "a", "b"], dtype="category"),
+        }
+    )
+    result = df.groupby("group").first()
+    expected = DataFrame(
+        {
+            "time": 3 * [np.datetime64("NaT")],
+            "categories": Series(["a", "c", "a"]).astype(
+                pd.CategoricalDtype(["a", "b", "c"])
+            ),
+        }
+    )
+    expected.index = Index(["first", "second", "third"], name="group")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_first_multi_key_groupby_categorical():
+    # GH 22512
+    df = DataFrame(
+        {
+            "A": [1, 1, 1, 2, 2],
+            "B": [100, 100, 200, 100, 100],
+            "C": ["apple", "orange", "mango", "mango", "orange"],
+            "D": ["jupiter", "mercury", "mars", "venus", "venus"],
+        }
+    )
+    df = df.astype({"D": "category"})
+    result = df.groupby(by=["A", "B"]).first()
+    expected = DataFrame(
+        {
+            "C": ["apple", "mango", "mango"],
+            "D": Series(["jupiter", "mars", "venus"]).astype(
+                pd.CategoricalDtype(["jupiter", "mars", "mercury", "venus"])
+            ),
+        }
+    )
+    expected.index = MultiIndex.from_tuples(
+        [(1, 100), (1, 200), (2, 100)], names=["A", "B"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["first", "last", "nth"])
+def test_groupby_last_first_nth_with_none(method, nulls_fixture):
+    # GH29645
+    expected = Series(["y"], dtype=object)
+    data = Series(
+        [nulls_fixture, nulls_fixture, nulls_fixture, "y", nulls_fixture],
+        index=[0, 0, 0, 0, 0],
+        dtype=object,
+    ).groupby(level=0)
+
+    if method == "nth":
+        result = getattr(data, method)(3)
+    else:
+        result = getattr(data, method)()
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg, expected_rows",
+    [
+        [slice(None, 3, 2), [0, 1, 4, 5]],
+        [slice(None, -2), [0, 2, 5]],
+        [[slice(None, 2), slice(-2, None)], [0, 1, 2, 3, 4, 6, 7]],
+        [[0, 1, slice(-2, None)], [0, 1, 2, 3, 4, 6, 7]],
+    ],
+)
+def test_slice(slice_test_df, slice_test_grouped, arg, expected_rows):
+    # Test slices     GH #42947
+
+    result = slice_test_grouped.nth[arg]
+    equivalent = slice_test_grouped.nth(arg)
+    expected = slice_test_df.iloc[expected_rows]
+
+    tm.assert_frame_equal(result, expected)
+    tm.assert_frame_equal(equivalent, expected)
+
+
+def test_nth_indexed(slice_test_df, slice_test_grouped):
+    # Test index notation     GH #44688
+
+    result = slice_test_grouped.nth[0, 1, -2:]
+    equivalent = slice_test_grouped.nth([0, 1, slice(-2, None)])
+    expected = slice_test_df.iloc[[0, 1, 2, 3, 4, 6, 7]]
+
+    tm.assert_frame_equal(result, expected)
+    tm.assert_frame_equal(equivalent, expected)
+
+
+def test_invalid_argument(slice_test_grouped):
+    # Test for error on invalid argument
+
+    with pytest.raises(TypeError, match="Invalid index"):
+        slice_test_grouped.nth(3.14)
+
+
+def test_negative_step(slice_test_grouped):
+    # Test for error on negative slice step
+
+    with pytest.raises(ValueError, match="Invalid step"):
+        slice_test_grouped.nth(slice(None, None, -1))
+
+
+def test_np_ints(slice_test_df, slice_test_grouped):
+    # Test np ints work
+
+    result = slice_test_grouped.nth(np.array([0, 1]))
+    expected = slice_test_df.iloc[[0, 1, 2, 3, 4]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_nth_interval():
+    # GH#24205
+    idx_result = MultiIndex(
+        [
+            pd.CategoricalIndex([pd.Interval(0, 1), pd.Interval(1, 2)]),
+            pd.CategoricalIndex([pd.Interval(0, 10), pd.Interval(10, 20)]),
+        ],
+        [[0, 0, 0, 1, 1], [0, 1, 1, 0, -1]],
+    )
+    df_result = DataFrame({"col": range(len(idx_result))}, index=idx_result)
+    result = df_result.groupby(level=[0, 1], observed=False).nth(0)
+    val_expected = [0, 1, 3]
+    idx_expected = MultiIndex(
+        [
+            pd.CategoricalIndex([pd.Interval(0, 1), pd.Interval(1, 2)]),
+            pd.CategoricalIndex([pd.Interval(0, 10), pd.Interval(10, 20)]),
+        ],
+        [[0, 0, 1], [0, 1, 0]],
+    )
+    expected = DataFrame(val_expected, index=idx_expected, columns=["col"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:invalid value encountered in remainder:RuntimeWarning"
+)
+def test_head_tail_dropna_true():
+    # GH#45089
+    df = DataFrame(
+        [["a", "z"], ["b", np.nan], ["c", np.nan], ["c", np.nan]], columns=["X", "Y"]
+    )
+    expected = DataFrame([["a", "z"]], columns=["X", "Y"])
+
+    result = df.groupby(["X", "Y"]).head(n=1)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(["X", "Y"]).tail(n=1)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(["X", "Y"]).nth(n=0)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_head_tail_dropna_false():
+    # GH#45089
+    df = DataFrame([["a", "z"], ["b", np.nan], ["c", np.nan]], columns=["X", "Y"])
+    expected = DataFrame([["a", "z"], ["b", np.nan], ["c", np.nan]], columns=["X", "Y"])
+
+    result = df.groupby(["X", "Y"], dropna=False).head(n=1)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(["X", "Y"], dropna=False).tail(n=1)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(["X", "Y"], dropna=False).nth(n=0)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("selection", ("b", ["b"], ["b", "c"]))
+@pytest.mark.parametrize("dropna", ["any", "all", None])
+def test_nth_after_selection(selection, dropna):
+    # GH#11038, GH#53518
+    df = DataFrame(
+        {
+            "a": [1, 1, 2],
+            "b": [np.nan, 3, 4],
+            "c": [5, 6, 7],
+        }
+    )
+    gb = df.groupby("a")[selection]
+    result = gb.nth(0, dropna=dropna)
+    if dropna == "any" or (dropna == "all" and selection != ["b", "c"]):
+        locs = [1, 2]
+    else:
+        locs = [0, 2]
+    expected = df.loc[locs, selection]
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        (
+            Timestamp("2011-01-15 12:50:28.502376"),
+            Timestamp("2011-01-20 12:50:28.593448"),
+        ),
+        (24650000000000001, 24650000000000002),
+    ],
+)
+def test_groupby_nth_int_like_precision(data):
+    # GH#6620, GH#9311
+    df = DataFrame({"a": [1, 1], "b": data})
+
+    grouped = df.groupby("a")
+    result = grouped.nth(0)
+    expected = DataFrame({"a": 1, "b": [data[0]]})
+
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_quantile.py b/pandas/pandas/tests/groupby/methods/test_quantile.py
new file mode 100644
index 0000000000000..28cb25b515ed2
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_quantile.py
@@ -0,0 +1,458 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "interpolation", ["linear", "lower", "higher", "nearest", "midpoint"]
+)
+@pytest.mark.parametrize(
+    "a_vals,b_vals",
+    [
+        # Ints
+        ([1, 2, 3, 4, 5], [5, 4, 3, 2, 1]),
+        ([1, 2, 3, 4], [4, 3, 2, 1]),
+        ([1, 2, 3, 4, 5], [4, 3, 2, 1]),
+        # Floats
+        ([1.0, 2.0, 3.0, 4.0, 5.0], [5.0, 4.0, 3.0, 2.0, 1.0]),
+        # Missing data
+        ([1.0, np.nan, 3.0, np.nan, 5.0], [5.0, np.nan, 3.0, np.nan, 1.0]),
+        ([np.nan, 4.0, np.nan, 2.0, np.nan], [np.nan, 4.0, np.nan, 2.0, np.nan]),
+        # Timestamps
+        (
+            pd.date_range("1/1/18", freq="D", periods=5),
+            pd.date_range("1/1/18", freq="D", periods=5)[::-1],
+        ),
+        (
+            pd.date_range("1/1/18", freq="D", periods=5).as_unit("s"),
+            pd.date_range("1/1/18", freq="D", periods=5)[::-1].as_unit("s"),
+        ),
+        # All NA
+        ([np.nan] * 5, [np.nan] * 5),
+    ],
+)
+@pytest.mark.parametrize("q", [0, 0.25, 0.5, 0.75, 1])
+def test_quantile(interpolation, a_vals, b_vals, q):
+    all_vals = pd.concat([pd.Series(a_vals), pd.Series(b_vals)])
+
+    a_expected = pd.Series(a_vals).quantile(q, interpolation=interpolation)
+    b_expected = pd.Series(b_vals).quantile(q, interpolation=interpolation)
+
+    df = DataFrame({"key": ["a"] * len(a_vals) + ["b"] * len(b_vals), "val": all_vals})
+
+    expected = DataFrame(
+        [a_expected, b_expected], columns=["val"], index=Index(["a", "b"], name="key")
+    )
+    if all_vals.dtype.kind == "M" and expected.dtypes.values[0].kind == "M":
+        # TODO(non-nano): this should be unnecessary once array_to_datetime
+        #  correctly infers non-nano from Timestamp.unit
+        expected = expected.astype(all_vals.dtype)
+    result = df.groupby("key").quantile(q, interpolation=interpolation)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_quantile_array():
+    # https://github.com/pandas-dev/pandas/issues/27526
+    df = DataFrame({"A": [0, 1, 2, 3, 4]})
+    key = np.array([0, 0, 1, 1, 1], dtype=np.int64)
+    result = df.groupby(key).quantile([0.25])
+
+    index = pd.MultiIndex.from_product([[0, 1], [0.25]])
+    expected = DataFrame({"A": [0.25, 2.50]}, index=index)
+    tm.assert_frame_equal(result, expected)
+
+    df = DataFrame({"A": [0, 1, 2, 3], "B": [4, 5, 6, 7]})
+    index = pd.MultiIndex.from_product([[0, 1], [0.25, 0.75]])
+
+    key = np.array([0, 0, 1, 1], dtype=np.int64)
+    result = df.groupby(key).quantile([0.25, 0.75])
+    expected = DataFrame(
+        {"A": [0.25, 0.75, 2.25, 2.75], "B": [4.25, 4.75, 6.25, 6.75]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_quantile_array2():
+    # https://github.com/pandas-dev/pandas/pull/28085#issuecomment-524066959
+    arr = np.random.default_rng(2).integers(0, 5, size=(10, 3), dtype=np.int64)
+    df = DataFrame(arr, columns=list("ABC"))
+    result = df.groupby("A").quantile([0.3, 0.7])
+    expected = DataFrame(
+        {
+            "B": [2.0, 2.0, 2.3, 2.7, 0.3, 0.7, 3.2, 4.0, 0.3, 0.7],
+            "C": [1.0, 1.0, 1.9, 3.0999999999999996, 0.3, 0.7, 2.6, 3.0, 1.2, 2.8],
+        },
+        index=pd.MultiIndex.from_product(
+            [[0, 1, 2, 3, 4], [0.3, 0.7]], names=["A", None]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_quantile_array_no_sort():
+    df = DataFrame({"A": [0, 1, 2], "B": [3, 4, 5]})
+    key = np.array([1, 0, 1], dtype=np.int64)
+    result = df.groupby(key, sort=False).quantile([0.25, 0.5, 0.75])
+    expected = DataFrame(
+        {"A": [0.5, 1.0, 1.5, 1.0, 1.0, 1.0], "B": [3.5, 4.0, 4.5, 4.0, 4.0, 4.0]},
+        index=pd.MultiIndex.from_product([[1, 0], [0.25, 0.5, 0.75]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(key, sort=False).quantile([0.75, 0.25])
+    expected = DataFrame(
+        {"A": [1.5, 0.5, 1.0, 1.0], "B": [4.5, 3.5, 4.0, 4.0]},
+        index=pd.MultiIndex.from_product([[1, 0], [0.75, 0.25]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_quantile_array_multiple_levels():
+    df = DataFrame(
+        {"A": [0, 1, 2], "B": [3, 4, 5], "c": ["a", "a", "a"], "d": ["a", "a", "b"]}
+    )
+    result = df.groupby(["c", "d"]).quantile([0.25, 0.75])
+    index = pd.MultiIndex.from_tuples(
+        [("a", "a", 0.25), ("a", "a", 0.75), ("a", "b", 0.25), ("a", "b", 0.75)],
+        names=["c", "d", None],
+    )
+    expected = DataFrame(
+        {"A": [0.25, 0.75, 2.0, 2.0], "B": [3.25, 3.75, 5.0, 5.0]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("frame_size", [(2, 3), (100, 10)])
+@pytest.mark.parametrize("groupby", [[0], [0, 1]])
+@pytest.mark.parametrize("q", [[0.5, 0.6]])
+def test_groupby_quantile_with_arraylike_q_and_int_columns(frame_size, groupby, q):
+    # GH30289
+    nrow, ncol = frame_size
+    df = DataFrame(np.array([ncol * [_ % 4] for _ in range(nrow)]), columns=range(ncol))
+
+    idx_levels = [np.arange(min(nrow, 4))] * len(groupby) + [q]
+    idx_codes = [[x for x in range(min(nrow, 4)) for _ in q]] * len(groupby) + [
+        list(range(len(q))) * min(nrow, 4)
+    ]
+    expected_index = pd.MultiIndex(
+        levels=idx_levels, codes=idx_codes, names=groupby + [None]
+    )
+    expected_values = [
+        [float(x)] * (ncol - len(groupby)) for x in range(min(nrow, 4)) for _ in q
+    ]
+    expected_columns = [x for x in range(ncol) if x not in groupby]
+    expected = DataFrame(
+        expected_values, index=expected_index, columns=expected_columns
+    )
+    result = df.groupby(groupby).quantile(q)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_quantile_raises():
+    df = DataFrame([["foo", "a"], ["foo", "b"], ["foo", "c"]], columns=["key", "val"])
+
+    msg = "dtype '(object|str)' does not support operation 'quantile'"
+    with pytest.raises(TypeError, match=msg):
+        df.groupby("key").quantile()
+
+
+def test_quantile_out_of_bounds_q_raises():
+    # https://github.com/pandas-dev/pandas/issues/27470
+    df = DataFrame({"a": [0, 0, 0, 1, 1, 1], "b": range(6)})
+    g = df.groupby([0, 0, 0, 1, 1, 1])
+    with pytest.raises(ValueError, match="Got '50.0' instead"):
+        g.quantile(50)
+
+    with pytest.raises(ValueError, match="Got '-1.0' instead"):
+        g.quantile(-1)
+
+
+def test_quantile_missing_group_values_no_segfaults():
+    # GH 28662
+    data = np.array([1.0, np.nan, 1.0])
+    df = DataFrame({"key": data, "val": range(3)})
+
+    # Random segfaults; would have been guaranteed in loop
+    grp = df.groupby("key")
+    for _ in range(100):
+        grp.quantile()
+
+
+@pytest.mark.parametrize(
+    "key, val, expected_key, expected_val",
+    [
+        ([1.0, np.nan, 3.0, np.nan], range(4), [1.0, 3.0], [0.0, 2.0]),
+        ([1.0, np.nan, 2.0, 2.0], range(4), [1.0, 2.0], [0.0, 2.5]),
+        (["a", "b", "b", np.nan], range(4), ["a", "b"], [0, 1.5]),
+        ([0], [42], [0], [42.0]),
+        ([], [], np.array([], dtype="float64"), np.array([], dtype="float64")),
+    ],
+)
+def test_quantile_missing_group_values_correct_results(
+    key, val, expected_key, expected_val
+):
+    # GH 28662, GH 33200, GH 33569
+    df = DataFrame({"key": key, "val": val})
+
+    expected = DataFrame(
+        expected_val, index=Index(expected_key, name="key"), columns=["val"]
+    )
+
+    grp = df.groupby("key")
+
+    result = grp.quantile(0.5)
+    tm.assert_frame_equal(result, expected)
+
+    result = grp.quantile()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        pd.array([1, 0, None] * 2, dtype="Int64"),
+        pd.array([True, False, None] * 2, dtype="boolean"),
+    ],
+)
+@pytest.mark.parametrize("q", [0.5, [0.0, 0.5, 1.0]])
+def test_groupby_quantile_nullable_array(values, q):
+    # https://github.com/pandas-dev/pandas/issues/33136
+    df = DataFrame({"a": ["x"] * 3 + ["y"] * 3, "b": values})
+    result = df.groupby("a")["b"].quantile(q)
+
+    if isinstance(q, list):
+        idx = pd.MultiIndex.from_product((["x", "y"], q), names=["a", None])
+        true_quantiles = [0.0, 0.5, 1.0]
+    else:
+        idx = Index(["x", "y"], name="a")
+        true_quantiles = [0.5]
+
+    expected = pd.Series(true_quantiles * 2, index=idx, name="b", dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("q", [0.5, [0.0, 0.5, 1.0]])
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_groupby_quantile_raises_on_invalid_dtype(q, numeric_only):
+    df = DataFrame({"a": [1], "b": [2.0], "c": ["x"]})
+    if numeric_only:
+        result = df.groupby("a").quantile(q, numeric_only=numeric_only)
+        expected = df.groupby("a")[["b"]].quantile(q)
+        tm.assert_frame_equal(result, expected)
+    else:
+        msg = "dtype '.*' does not support operation 'quantile'"
+        with pytest.raises(TypeError, match=msg):
+            df.groupby("a").quantile(q, numeric_only=numeric_only)
+
+
+def test_groupby_quantile_NA_float(any_float_dtype):
+    # GH#42849
+    df = DataFrame({"x": [1, 1], "y": [0.2, np.nan]}, dtype=any_float_dtype)
+    result = df.groupby("x")["y"].quantile(0.5)
+    exp_index = Index([1.0], dtype=any_float_dtype, name="x")
+
+    if any_float_dtype in ["Float32", "Float64"]:
+        expected_dtype = any_float_dtype
+    else:
+        expected_dtype = None
+
+    expected = pd.Series([0.2], dtype=expected_dtype, index=exp_index, name="y")
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby("x")["y"].quantile([0.5, 0.75])
+    expected = pd.Series(
+        [0.2] * 2,
+        index=pd.MultiIndex.from_product((exp_index, [0.5, 0.75]), names=["x", None]),
+        name="y",
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_quantile_NA_int(any_int_ea_dtype):
+    # GH#42849
+    df = DataFrame({"x": [1, 1], "y": [2, 5]}, dtype=any_int_ea_dtype)
+    result = df.groupby("x")["y"].quantile(0.5)
+    expected = pd.Series(
+        [3.5],
+        dtype="Float64",
+        index=Index([1], name="x", dtype=any_int_ea_dtype),
+        name="y",
+    )
+    tm.assert_series_equal(expected, result)
+
+    result = df.groupby("x").quantile(0.5)
+    expected = DataFrame(
+        {"y": 3.5}, dtype="Float64", index=Index([1], name="x", dtype=any_int_ea_dtype)
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "interpolation, val1, val2", [("lower", 2, 2), ("higher", 2, 3), ("nearest", 2, 2)]
+)
+def test_groupby_quantile_all_na_group_masked(
+    interpolation, val1, val2, any_numeric_ea_dtype
+):
+    # GH#37493
+    df = DataFrame(
+        {"a": [1, 1, 1, 2], "b": [1, 2, 3, pd.NA]}, dtype=any_numeric_ea_dtype
+    )
+    result = df.groupby("a").quantile(q=[0.5, 0.7], interpolation=interpolation)
+    expected = DataFrame(
+        {"b": [val1, val2, pd.NA, pd.NA]},
+        dtype=any_numeric_ea_dtype,
+        index=pd.MultiIndex.from_arrays(
+            [pd.Series([1, 1, 2, 2], dtype=any_numeric_ea_dtype), [0.5, 0.7, 0.5, 0.7]],
+            names=["a", None],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("interpolation", ["midpoint", "linear"])
+def test_groupby_quantile_all_na_group_masked_interp(
+    interpolation, any_numeric_ea_dtype
+):
+    # GH#37493
+    df = DataFrame(
+        {"a": [1, 1, 1, 2], "b": [1, 2, 3, pd.NA]}, dtype=any_numeric_ea_dtype
+    )
+    result = df.groupby("a").quantile(q=[0.5, 0.75], interpolation=interpolation)
+
+    if any_numeric_ea_dtype == "Float32":
+        expected_dtype = any_numeric_ea_dtype
+    else:
+        expected_dtype = "Float64"
+
+    expected = DataFrame(
+        {"b": [2.0, 2.5, pd.NA, pd.NA]},
+        dtype=expected_dtype,
+        index=pd.MultiIndex.from_arrays(
+            [
+                pd.Series([1, 1, 2, 2], dtype=any_numeric_ea_dtype),
+                [0.5, 0.75, 0.5, 0.75],
+            ],
+            names=["a", None],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["Float64", "Float32"])
+def test_groupby_quantile_allNA_column(dtype):
+    # GH#42849
+    df = DataFrame({"x": [1, 1], "y": [pd.NA] * 2}, dtype=dtype)
+    result = df.groupby("x")["y"].quantile(0.5)
+    expected = pd.Series(
+        [np.nan], dtype=dtype, index=Index([1.0], dtype=dtype), name="y"
+    )
+    expected.index.name = "x"
+    tm.assert_series_equal(expected, result)
+
+
+def test_groupby_timedelta_quantile():
+    # GH: 29485
+    df = DataFrame(
+        {"value": pd.to_timedelta(np.arange(4), unit="s"), "group": [1, 1, 2, 2]}
+    )
+    result = df.groupby("group").quantile(0.99)
+    expected = DataFrame(
+        {
+            "value": [
+                pd.Timedelta("0 days 00:00:00.990000"),
+                pd.Timedelta("0 days 00:00:02.990000"),
+            ]
+        },
+        index=Index([1, 2], name="group"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_timestamp_groupby_quantile(unit):
+    # GH 33168
+    dti = pd.date_range(
+        start="2020-04-19 00:00:00", freq="1min", periods=100, tz="UTC", unit=unit
+    ).floor("1h")
+    df = DataFrame(
+        {
+            "timestamp": dti,
+            "category": list(range(1, 101)),
+            "value": list(range(101, 201)),
+        }
+    )
+
+    result = df.groupby("timestamp").quantile([0.2, 0.8])
+
+    mi = pd.MultiIndex.from_product([dti[::99], [0.2, 0.8]], names=("timestamp", None))
+    expected = DataFrame(
+        [
+            {"category": 12.8, "value": 112.8},
+            {"category": 48.2, "value": 148.2},
+            {"category": 68.8, "value": 168.8},
+            {"category": 92.2, "value": 192.2},
+        ],
+        index=mi,
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_quantile_dt64tz_period():
+    # GH#51373
+    dti = pd.date_range("2016-01-01", periods=1000)
+    df = pd.Series(dti).to_frame().copy()
+    df[1] = dti.tz_localize("US/Pacific")
+    df[2] = dti.to_period("D")
+    df[3] = dti - dti[0]
+    df.iloc[-1] = pd.NaT
+
+    by = np.tile(np.arange(5), 200)
+    gb = df.groupby(by)
+
+    result = gb.quantile(0.5)
+
+    # Check that we match the group-by-group result
+    exp = {i: df.iloc[i::5].quantile(0.5) for i in range(5)}
+    expected = DataFrame(exp).T.infer_objects()
+    expected.index = expected.index.astype(int)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_quantile_nonmulti_levels_order():
+    # Non-regression test for GH #53009
+    ind = pd.MultiIndex.from_tuples(
+        [
+            (0, "a", "B"),
+            (0, "a", "A"),
+            (0, "b", "B"),
+            (0, "b", "A"),
+            (1, "a", "B"),
+            (1, "a", "A"),
+            (1, "b", "B"),
+            (1, "b", "A"),
+        ],
+        names=["sample", "cat0", "cat1"],
+    )
+    ser = pd.Series(range(8), index=ind)
+    result = ser.groupby(level="cat1", sort=False).quantile([0.2, 0.8])
+
+    qind = pd.MultiIndex.from_tuples(
+        [("B", 0.2), ("B", 0.8), ("A", 0.2), ("A", 0.8)], names=["cat1", None]
+    )
+    expected = pd.Series([1.2, 4.8, 2.2, 5.8], index=qind)
+
+    tm.assert_series_equal(result, expected)
+
+    # We need to check that index levels are not sorted
+    expected_levels = pd.core.indexes.frozen.FrozenList([["B", "A"], [0.2, 0.8]])
+    tm.assert_equal(result.index.levels, expected_levels)
diff --git a/pandas/pandas/tests/groupby/methods/test_rank.py b/pandas/pandas/tests/groupby/methods/test_rank.py
new file mode 100644
index 0000000000000..1fb878cbc1111
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_rank.py
@@ -0,0 +1,643 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    NaT,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+
+def test_rank_unordered_categorical_typeerror():
+    # GH#51034 should be TypeError, not NotImplementedError
+    cat = pd.Categorical([], ordered=False)
+    ser = Series(cat)
+    df = ser.to_frame()
+
+    msg = "Cannot perform rank with non-ordered Categorical"
+
+    gb = ser.groupby(cat, observed=False)
+    with pytest.raises(TypeError, match=msg):
+        gb.rank()
+
+    gb2 = df.groupby(cat, observed=False)
+    with pytest.raises(TypeError, match=msg):
+        gb2.rank()
+
+
+def test_rank_apply():
+    lev1 = np.array(["a" * 10] * 100, dtype=object)
+    lev2 = np.array(["b" * 10] * 130, dtype=object)
+    lab1 = np.random.default_rng(2).integers(0, 100, size=500, dtype=int)
+    lab2 = np.random.default_rng(2).integers(0, 130, size=500, dtype=int)
+
+    df = DataFrame(
+        {
+            "value": np.random.default_rng(2).standard_normal(500),
+            "key1": lev1.take(lab1),
+            "key2": lev2.take(lab2),
+        }
+    )
+
+    result = df.groupby(["key1", "key2"]).value.rank()
+
+    expected = [piece.value.rank() for key, piece in df.groupby(["key1", "key2"])]
+    expected = concat(expected, axis=0)
+    expected = expected.reindex(result.index)
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby(["key1", "key2"]).value.rank(pct=True)
+
+    expected = [
+        piece.value.rank(pct=True) for key, piece in df.groupby(["key1", "key2"])
+    ]
+    expected = concat(expected, axis=0)
+    expected = expected.reindex(result.index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("grps", [["qux"], ["qux", "quux"]])
+@pytest.mark.parametrize(
+    "vals",
+    [
+        np.array([2, 2, 8, 2, 6], dtype=dtype)
+        for dtype in ["i8", "i4", "i2", "i1", "u8", "u4", "u2", "u1", "f8", "f4", "f2"]
+    ]
+    + [
+        [
+            pd.Timestamp("2018-01-02"),
+            pd.Timestamp("2018-01-02"),
+            pd.Timestamp("2018-01-08"),
+            pd.Timestamp("2018-01-02"),
+            pd.Timestamp("2018-01-06"),
+        ],
+        [
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            pd.Timestamp("2018-01-08", tz="US/Pacific"),
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            pd.Timestamp("2018-01-06", tz="US/Pacific"),
+        ],
+        [
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-08") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-06") - pd.Timestamp(0),
+        ],
+        [
+            pd.Timestamp("2018-01-02").to_period("D"),
+            pd.Timestamp("2018-01-02").to_period("D"),
+            pd.Timestamp("2018-01-08").to_period("D"),
+            pd.Timestamp("2018-01-02").to_period("D"),
+            pd.Timestamp("2018-01-06").to_period("D"),
+        ],
+    ],
+    ids=lambda x: type(x[0]),
+)
+@pytest.mark.parametrize(
+    "ties_method,ascending,pct,exp",
+    [
+        ("average", True, False, [2.0, 2.0, 5.0, 2.0, 4.0]),
+        ("average", True, True, [0.4, 0.4, 1.0, 0.4, 0.8]),
+        ("average", False, False, [4.0, 4.0, 1.0, 4.0, 2.0]),
+        ("average", False, True, [0.8, 0.8, 0.2, 0.8, 0.4]),
+        ("min", True, False, [1.0, 1.0, 5.0, 1.0, 4.0]),
+        ("min", True, True, [0.2, 0.2, 1.0, 0.2, 0.8]),
+        ("min", False, False, [3.0, 3.0, 1.0, 3.0, 2.0]),
+        ("min", False, True, [0.6, 0.6, 0.2, 0.6, 0.4]),
+        ("max", True, False, [3.0, 3.0, 5.0, 3.0, 4.0]),
+        ("max", True, True, [0.6, 0.6, 1.0, 0.6, 0.8]),
+        ("max", False, False, [5.0, 5.0, 1.0, 5.0, 2.0]),
+        ("max", False, True, [1.0, 1.0, 0.2, 1.0, 0.4]),
+        ("first", True, False, [1.0, 2.0, 5.0, 3.0, 4.0]),
+        ("first", True, True, [0.2, 0.4, 1.0, 0.6, 0.8]),
+        ("first", False, False, [3.0, 4.0, 1.0, 5.0, 2.0]),
+        ("first", False, True, [0.6, 0.8, 0.2, 1.0, 0.4]),
+        ("dense", True, False, [1.0, 1.0, 3.0, 1.0, 2.0]),
+        ("dense", True, True, [1.0 / 3.0, 1.0 / 3.0, 3.0 / 3.0, 1.0 / 3.0, 2.0 / 3.0]),
+        ("dense", False, False, [3.0, 3.0, 1.0, 3.0, 2.0]),
+        ("dense", False, True, [3.0 / 3.0, 3.0 / 3.0, 1.0 / 3.0, 3.0 / 3.0, 2.0 / 3.0]),
+    ],
+)
+def test_rank_args(grps, vals, ties_method, ascending, pct, exp):
+    key = np.repeat(grps, len(vals))
+
+    orig_vals = vals
+    vals = list(vals) * len(grps)
+    if isinstance(orig_vals, np.ndarray):
+        vals = np.array(vals, dtype=orig_vals.dtype)
+
+    df = DataFrame({"key": key, "val": vals})
+    result = df.groupby("key").rank(method=ties_method, ascending=ascending, pct=pct)
+
+    exp_df = DataFrame(exp * len(grps), columns=["val"])
+    tm.assert_frame_equal(result, exp_df)
+
+
+@pytest.mark.parametrize("grps", [["qux"], ["qux", "quux"]])
+@pytest.mark.parametrize(
+    "vals", [[-np.inf, -np.inf, np.nan, 1.0, np.nan, np.inf, np.inf]]
+)
+@pytest.mark.parametrize(
+    "ties_method,ascending,na_option,exp",
+    [
+        ("average", True, "keep", [1.5, 1.5, np.nan, 3, np.nan, 4.5, 4.5]),
+        ("average", True, "top", [3.5, 3.5, 1.5, 5.0, 1.5, 6.5, 6.5]),
+        ("average", True, "bottom", [1.5, 1.5, 6.5, 3.0, 6.5, 4.5, 4.5]),
+        ("average", False, "keep", [4.5, 4.5, np.nan, 3, np.nan, 1.5, 1.5]),
+        ("average", False, "top", [6.5, 6.5, 1.5, 5.0, 1.5, 3.5, 3.5]),
+        ("average", False, "bottom", [4.5, 4.5, 6.5, 3.0, 6.5, 1.5, 1.5]),
+        ("min", True, "keep", [1.0, 1.0, np.nan, 3.0, np.nan, 4.0, 4.0]),
+        ("min", True, "top", [3.0, 3.0, 1.0, 5.0, 1.0, 6.0, 6.0]),
+        ("min", True, "bottom", [1.0, 1.0, 6.0, 3.0, 6.0, 4.0, 4.0]),
+        ("min", False, "keep", [4.0, 4.0, np.nan, 3.0, np.nan, 1.0, 1.0]),
+        ("min", False, "top", [6.0, 6.0, 1.0, 5.0, 1.0, 3.0, 3.0]),
+        ("min", False, "bottom", [4.0, 4.0, 6.0, 3.0, 6.0, 1.0, 1.0]),
+        ("max", True, "keep", [2.0, 2.0, np.nan, 3.0, np.nan, 5.0, 5.0]),
+        ("max", True, "top", [4.0, 4.0, 2.0, 5.0, 2.0, 7.0, 7.0]),
+        ("max", True, "bottom", [2.0, 2.0, 7.0, 3.0, 7.0, 5.0, 5.0]),
+        ("max", False, "keep", [5.0, 5.0, np.nan, 3.0, np.nan, 2.0, 2.0]),
+        ("max", False, "top", [7.0, 7.0, 2.0, 5.0, 2.0, 4.0, 4.0]),
+        ("max", False, "bottom", [5.0, 5.0, 7.0, 3.0, 7.0, 2.0, 2.0]),
+        ("first", True, "keep", [1.0, 2.0, np.nan, 3.0, np.nan, 4.0, 5.0]),
+        ("first", True, "top", [3.0, 4.0, 1.0, 5.0, 2.0, 6.0, 7.0]),
+        ("first", True, "bottom", [1.0, 2.0, 6.0, 3.0, 7.0, 4.0, 5.0]),
+        ("first", False, "keep", [4.0, 5.0, np.nan, 3.0, np.nan, 1.0, 2.0]),
+        ("first", False, "top", [6.0, 7.0, 1.0, 5.0, 2.0, 3.0, 4.0]),
+        ("first", False, "bottom", [4.0, 5.0, 6.0, 3.0, 7.0, 1.0, 2.0]),
+        ("dense", True, "keep", [1.0, 1.0, np.nan, 2.0, np.nan, 3.0, 3.0]),
+        ("dense", True, "top", [2.0, 2.0, 1.0, 3.0, 1.0, 4.0, 4.0]),
+        ("dense", True, "bottom", [1.0, 1.0, 4.0, 2.0, 4.0, 3.0, 3.0]),
+        ("dense", False, "keep", [3.0, 3.0, np.nan, 2.0, np.nan, 1.0, 1.0]),
+        ("dense", False, "top", [4.0, 4.0, 1.0, 3.0, 1.0, 2.0, 2.0]),
+        ("dense", False, "bottom", [3.0, 3.0, 4.0, 2.0, 4.0, 1.0, 1.0]),
+    ],
+)
+def test_infs_n_nans(grps, vals, ties_method, ascending, na_option, exp):
+    # GH 20561
+    key = np.repeat(grps, len(vals))
+    vals = vals * len(grps)
+    df = DataFrame({"key": key, "val": vals})
+    result = df.groupby("key").rank(
+        method=ties_method, ascending=ascending, na_option=na_option
+    )
+    exp_df = DataFrame(exp * len(grps), columns=["val"])
+    tm.assert_frame_equal(result, exp_df)
+
+
+@pytest.mark.parametrize("grps", [["qux"], ["qux", "quux"]])
+@pytest.mark.parametrize(
+    "vals",
+    [
+        np.array([2, 2, np.nan, 8, 2, 6, np.nan, np.nan], dtype=dtype)
+        for dtype in ["f8", "f4", "f2"]
+    ]
+    + [
+        [
+            pd.Timestamp("2018-01-02"),
+            pd.Timestamp("2018-01-02"),
+            np.nan,
+            pd.Timestamp("2018-01-08"),
+            pd.Timestamp("2018-01-02"),
+            pd.Timestamp("2018-01-06"),
+            np.nan,
+            np.nan,
+        ],
+        [
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            np.nan,
+            pd.Timestamp("2018-01-08", tz="US/Pacific"),
+            pd.Timestamp("2018-01-02", tz="US/Pacific"),
+            pd.Timestamp("2018-01-06", tz="US/Pacific"),
+            np.nan,
+            np.nan,
+        ],
+        [
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            np.nan,
+            pd.Timestamp("2018-01-08") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-02") - pd.Timestamp(0),
+            pd.Timestamp("2018-01-06") - pd.Timestamp(0),
+            np.nan,
+            np.nan,
+        ],
+        [
+            pd.Timestamp("2018-01-02").to_period("D"),
+            pd.Timestamp("2018-01-02").to_period("D"),
+            np.nan,
+            pd.Timestamp("2018-01-08").to_period("D"),
+            pd.Timestamp("2018-01-02").to_period("D"),
+            pd.Timestamp("2018-01-06").to_period("D"),
+            np.nan,
+            np.nan,
+        ],
+    ],
+    ids=lambda x: type(x[0]),
+)
+@pytest.mark.parametrize(
+    "ties_method,ascending,na_option,pct,exp",
+    [
+        (
+            "average",
+            True,
+            "keep",
+            False,
+            [2.0, 2.0, np.nan, 5.0, 2.0, 4.0, np.nan, np.nan],
+        ),
+        (
+            "average",
+            True,
+            "keep",
+            True,
+            [0.4, 0.4, np.nan, 1.0, 0.4, 0.8, np.nan, np.nan],
+        ),
+        (
+            "average",
+            False,
+            "keep",
+            False,
+            [4.0, 4.0, np.nan, 1.0, 4.0, 2.0, np.nan, np.nan],
+        ),
+        (
+            "average",
+            False,
+            "keep",
+            True,
+            [0.8, 0.8, np.nan, 0.2, 0.8, 0.4, np.nan, np.nan],
+        ),
+        ("min", True, "keep", False, [1.0, 1.0, np.nan, 5.0, 1.0, 4.0, np.nan, np.nan]),
+        ("min", True, "keep", True, [0.2, 0.2, np.nan, 1.0, 0.2, 0.8, np.nan, np.nan]),
+        (
+            "min",
+            False,
+            "keep",
+            False,
+            [3.0, 3.0, np.nan, 1.0, 3.0, 2.0, np.nan, np.nan],
+        ),
+        ("min", False, "keep", True, [0.6, 0.6, np.nan, 0.2, 0.6, 0.4, np.nan, np.nan]),
+        ("max", True, "keep", False, [3.0, 3.0, np.nan, 5.0, 3.0, 4.0, np.nan, np.nan]),
+        ("max", True, "keep", True, [0.6, 0.6, np.nan, 1.0, 0.6, 0.8, np.nan, np.nan]),
+        (
+            "max",
+            False,
+            "keep",
+            False,
+            [5.0, 5.0, np.nan, 1.0, 5.0, 2.0, np.nan, np.nan],
+        ),
+        ("max", False, "keep", True, [1.0, 1.0, np.nan, 0.2, 1.0, 0.4, np.nan, np.nan]),
+        (
+            "first",
+            True,
+            "keep",
+            False,
+            [1.0, 2.0, np.nan, 5.0, 3.0, 4.0, np.nan, np.nan],
+        ),
+        (
+            "first",
+            True,
+            "keep",
+            True,
+            [0.2, 0.4, np.nan, 1.0, 0.6, 0.8, np.nan, np.nan],
+        ),
+        (
+            "first",
+            False,
+            "keep",
+            False,
+            [3.0, 4.0, np.nan, 1.0, 5.0, 2.0, np.nan, np.nan],
+        ),
+        (
+            "first",
+            False,
+            "keep",
+            True,
+            [0.6, 0.8, np.nan, 0.2, 1.0, 0.4, np.nan, np.nan],
+        ),
+        (
+            "dense",
+            True,
+            "keep",
+            False,
+            [1.0, 1.0, np.nan, 3.0, 1.0, 2.0, np.nan, np.nan],
+        ),
+        (
+            "dense",
+            True,
+            "keep",
+            True,
+            [
+                1.0 / 3.0,
+                1.0 / 3.0,
+                np.nan,
+                3.0 / 3.0,
+                1.0 / 3.0,
+                2.0 / 3.0,
+                np.nan,
+                np.nan,
+            ],
+        ),
+        (
+            "dense",
+            False,
+            "keep",
+            False,
+            [3.0, 3.0, np.nan, 1.0, 3.0, 2.0, np.nan, np.nan],
+        ),
+        (
+            "dense",
+            False,
+            "keep",
+            True,
+            [
+                3.0 / 3.0,
+                3.0 / 3.0,
+                np.nan,
+                1.0 / 3.0,
+                3.0 / 3.0,
+                2.0 / 3.0,
+                np.nan,
+                np.nan,
+            ],
+        ),
+        ("average", True, "bottom", False, [2.0, 2.0, 7.0, 5.0, 2.0, 4.0, 7.0, 7.0]),
+        (
+            "average",
+            True,
+            "bottom",
+            True,
+            [0.25, 0.25, 0.875, 0.625, 0.25, 0.5, 0.875, 0.875],
+        ),
+        ("average", False, "bottom", False, [4.0, 4.0, 7.0, 1.0, 4.0, 2.0, 7.0, 7.0]),
+        (
+            "average",
+            False,
+            "bottom",
+            True,
+            [0.5, 0.5, 0.875, 0.125, 0.5, 0.25, 0.875, 0.875],
+        ),
+        ("min", True, "bottom", False, [1.0, 1.0, 6.0, 5.0, 1.0, 4.0, 6.0, 6.0]),
+        (
+            "min",
+            True,
+            "bottom",
+            True,
+            [0.125, 0.125, 0.75, 0.625, 0.125, 0.5, 0.75, 0.75],
+        ),
+        ("min", False, "bottom", False, [3.0, 3.0, 6.0, 1.0, 3.0, 2.0, 6.0, 6.0]),
+        (
+            "min",
+            False,
+            "bottom",
+            True,
+            [0.375, 0.375, 0.75, 0.125, 0.375, 0.25, 0.75, 0.75],
+        ),
+        ("max", True, "bottom", False, [3.0, 3.0, 8.0, 5.0, 3.0, 4.0, 8.0, 8.0]),
+        ("max", True, "bottom", True, [0.375, 0.375, 1.0, 0.625, 0.375, 0.5, 1.0, 1.0]),
+        ("max", False, "bottom", False, [5.0, 5.0, 8.0, 1.0, 5.0, 2.0, 8.0, 8.0]),
+        (
+            "max",
+            False,
+            "bottom",
+            True,
+            [0.625, 0.625, 1.0, 0.125, 0.625, 0.25, 1.0, 1.0],
+        ),
+        ("first", True, "bottom", False, [1.0, 2.0, 6.0, 5.0, 3.0, 4.0, 7.0, 8.0]),
+        (
+            "first",
+            True,
+            "bottom",
+            True,
+            [0.125, 0.25, 0.75, 0.625, 0.375, 0.5, 0.875, 1.0],
+        ),
+        ("first", False, "bottom", False, [3.0, 4.0, 6.0, 1.0, 5.0, 2.0, 7.0, 8.0]),
+        (
+            "first",
+            False,
+            "bottom",
+            True,
+            [0.375, 0.5, 0.75, 0.125, 0.625, 0.25, 0.875, 1.0],
+        ),
+        ("dense", True, "bottom", False, [1.0, 1.0, 4.0, 3.0, 1.0, 2.0, 4.0, 4.0]),
+        ("dense", True, "bottom", True, [0.25, 0.25, 1.0, 0.75, 0.25, 0.5, 1.0, 1.0]),
+        ("dense", False, "bottom", False, [3.0, 3.0, 4.0, 1.0, 3.0, 2.0, 4.0, 4.0]),
+        ("dense", False, "bottom", True, [0.75, 0.75, 1.0, 0.25, 0.75, 0.5, 1.0, 1.0]),
+    ],
+)
+def test_rank_args_missing(grps, vals, ties_method, ascending, na_option, pct, exp):
+    key = np.repeat(grps, len(vals))
+
+    orig_vals = vals
+    vals = list(vals) * len(grps)
+    if isinstance(orig_vals, np.ndarray):
+        vals = np.array(vals, dtype=orig_vals.dtype)
+
+    df = DataFrame({"key": key, "val": vals})
+    result = df.groupby("key").rank(
+        method=ties_method, ascending=ascending, na_option=na_option, pct=pct
+    )
+
+    exp_df = DataFrame(exp * len(grps), columns=["val"])
+    tm.assert_frame_equal(result, exp_df)
+
+
+@pytest.mark.parametrize(
+    "pct,exp", [(False, [3.0, 3.0, 3.0, 3.0, 3.0]), (True, [0.6, 0.6, 0.6, 0.6, 0.6])]
+)
+def test_rank_resets_each_group(pct, exp):
+    df = DataFrame(
+        {"key": ["a", "a", "a", "a", "a", "b", "b", "b", "b", "b"], "val": [1] * 10}
+    )
+    result = df.groupby("key").rank(pct=pct)
+    exp_df = DataFrame(exp * 2, columns=["val"])
+    tm.assert_frame_equal(result, exp_df)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int64", "int32", "uint64", "uint32", "float64", "float32"]
+)
+@pytest.mark.parametrize("upper", [True, False])
+def test_rank_avg_even_vals(dtype, upper):
+    if upper:
+        # use IntegerDtype/FloatingDtype
+        dtype = dtype[0].upper() + dtype[1:]
+        dtype = dtype.replace("Ui", "UI")
+    df = DataFrame({"key": ["a"] * 4, "val": [1] * 4})
+    df["val"] = df["val"].astype(dtype)
+    assert df["val"].dtype == dtype
+
+    result = df.groupby("key").rank()
+    exp_df = DataFrame([2.5, 2.5, 2.5, 2.5], columns=["val"])
+    if upper:
+        exp_df = exp_df.astype("Float64")
+    tm.assert_frame_equal(result, exp_df)
+
+
+@pytest.mark.parametrize("na_option", ["keep", "top", "bottom"])
+@pytest.mark.parametrize("pct", [True, False])
+@pytest.mark.parametrize(
+    "vals", [["bar", "bar", "foo", "bar", "baz"], ["bar", np.nan, "foo", np.nan, "baz"]]
+)
+def test_rank_object_dtype(rank_method, ascending, na_option, pct, vals):
+    df = DataFrame({"key": ["foo"] * 5, "val": vals})
+    mask = df["val"].isna()
+
+    gb = df.groupby("key")
+    res = gb.rank(method=rank_method, ascending=ascending, na_option=na_option, pct=pct)
+
+    # construct our expected by using numeric values with the same ordering
+    if mask.any():
+        df2 = DataFrame({"key": ["foo"] * 5, "val": [0, np.nan, 2, np.nan, 1]})
+    else:
+        df2 = DataFrame({"key": ["foo"] * 5, "val": [0, 0, 2, 0, 1]})
+
+    gb2 = df2.groupby("key")
+    alt = gb2.rank(
+        method=rank_method, ascending=ascending, na_option=na_option, pct=pct
+    )
+
+    tm.assert_frame_equal(res, alt)
+
+
+@pytest.mark.parametrize("na_option", [True, "bad", 1])
+@pytest.mark.parametrize("pct", [True, False])
+@pytest.mark.parametrize(
+    "vals",
+    [
+        ["bar", "bar", "foo", "bar", "baz"],
+        ["bar", np.nan, "foo", np.nan, "baz"],
+        [1, np.nan, 2, np.nan, 3],
+    ],
+)
+def test_rank_naoption_raises(rank_method, ascending, na_option, pct, vals):
+    df = DataFrame({"key": ["foo"] * 5, "val": vals})
+    msg = "na_option must be one of 'keep', 'top', or 'bottom'"
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("key").rank(
+            method=rank_method, ascending=ascending, na_option=na_option, pct=pct
+        )
+
+
+def test_rank_empty_group():
+    # see gh-22519
+    column = "A"
+    df = DataFrame({"A": [0, 1, 0], "B": [1.0, np.nan, 2.0]})
+
+    result = df.groupby(column).B.rank(pct=True)
+    expected = Series([0.5, np.nan, 1.0], name="B")
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby(column).rank(pct=True)
+    expected = DataFrame({"B": [0.5, np.nan, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "input_key,input_value,output_value",
+    [
+        ([1, 2], [1, 1], [1.0, 1.0]),
+        ([1, 1, 2, 2], [1, 2, 1, 2], [0.5, 1.0, 0.5, 1.0]),
+        ([1, 1, 2, 2], [1, 2, 1, np.nan], [0.5, 1.0, 1.0, np.nan]),
+        ([1, 1, 2], [1, 2, np.nan], [0.5, 1.0, np.nan]),
+    ],
+)
+def test_rank_zero_div(input_key, input_value, output_value):
+    # GH 23666
+    df = DataFrame({"A": input_key, "B": input_value})
+
+    result = df.groupby("A").rank(method="dense", pct=True)
+    expected = DataFrame({"B": output_value})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rank_min_int():
+    # GH-32859
+    df = DataFrame(
+        {
+            "grp": [1, 1, 2],
+            "int_col": [
+                np.iinfo(np.int64).min,
+                np.iinfo(np.int64).max,
+                np.iinfo(np.int64).min,
+            ],
+            "datetimelike": [NaT, datetime(2001, 1, 1), NaT],
+        }
+    )
+
+    result = df.groupby("grp").rank()
+    expected = DataFrame(
+        {"int_col": [1.0, 2.0, 1.0], "datetimelike": [np.nan, 1.0, np.nan]}
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("use_nan", [True, False])
+def test_rank_pct_equal_values_on_group_transition(use_nan):
+    # GH#40518
+    fill_value = np.nan if use_nan else 3
+    df = DataFrame(
+        [
+            [-1, 1],
+            [-1, 2],
+            [1, fill_value],
+            [-1, fill_value],
+        ],
+        columns=["group", "val"],
+    )
+    result = df.groupby(["group"])["val"].rank(
+        method="dense",
+        pct=True,
+    )
+    if use_nan:
+        expected = Series([0.5, 1, np.nan, np.nan], name="val")
+    else:
+        expected = Series([1 / 3, 2 / 3, 1, 1], name="val")
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_non_unique_index():
+    # GH 16577
+    df = DataFrame(
+        {"A": [1.0, 2.0, 3.0, np.nan], "value": 1.0},
+        index=[pd.Timestamp("20170101", tz="US/Eastern")] * 4,
+    )
+    result = df.groupby([df.index, "A"]).value.rank(ascending=True, pct=True)
+    expected = Series(
+        [1.0, 1.0, 1.0, np.nan],
+        index=[pd.Timestamp("20170101", tz="US/Eastern")] * 4,
+        name="value",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_rank_categorical():
+    cat = pd.Categorical(["a", "a", "b", np.nan, "c", "b"], ordered=True)
+    cat2 = pd.Categorical([1, 2, 3, np.nan, 4, 5], ordered=True)
+
+    df = DataFrame({"col1": [0, 1, 0, 1, 0, 1], "col2": cat, "col3": cat2})
+
+    gb = df.groupby("col1")
+
+    res = gb.rank()
+
+    expected = df.astype(object).groupby("col1").rank()
+    tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize("na_option", ["top", "bottom"])
+def test_groupby_op_with_nullables(na_option):
+    # GH 54206
+    df = DataFrame({"x": [None]}, dtype="Float64")
+    result = df.groupby("x", dropna=False)["x"].rank(method="min", na_option=na_option)
+    expected = Series([1.0], dtype="Float64", name=result.name)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_sample.py b/pandas/pandas/tests/groupby/methods/test_sample.py
new file mode 100644
index 0000000000000..4dd474741740d
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_sample.py
@@ -0,0 +1,154 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("n, frac", [(2, None), (None, 0.2)])
+def test_groupby_sample_balanced_groups_shape(n, frac):
+    values = [1] * 10 + [2] * 10
+    df = DataFrame({"a": values, "b": values})
+
+    result = df.groupby("a").sample(n=n, frac=frac)
+    values = [1] * 2 + [2] * 2
+    expected = DataFrame({"a": values, "b": values}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(n=n, frac=frac)
+    expected = Series(values, name="b", index=result.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_sample_unbalanced_groups_shape():
+    values = [1] * 10 + [2] * 20
+    df = DataFrame({"a": values, "b": values})
+
+    result = df.groupby("a").sample(n=5)
+    values = [1] * 5 + [2] * 5
+    expected = DataFrame({"a": values, "b": values}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(n=5)
+    expected = Series(values, name="b", index=result.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_sample_index_value_spans_groups():
+    values = [1] * 3 + [2] * 3
+    df = DataFrame({"a": values, "b": values}, index=[1, 2, 2, 2, 2, 2])
+
+    result = df.groupby("a").sample(n=2)
+    values = [1] * 2 + [2] * 2
+    expected = DataFrame({"a": values, "b": values}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(n=2)
+    expected = Series(values, name="b", index=result.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_sample_n_and_frac_raises():
+    df = DataFrame({"a": [1, 2], "b": [1, 2]})
+    msg = "Please enter a value for `frac` OR `n`, not both"
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a").sample(n=1, frac=1.0)
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a")["b"].sample(n=1, frac=1.0)
+
+
+def test_groupby_sample_frac_gt_one_without_replacement_raises():
+    df = DataFrame({"a": [1, 2], "b": [1, 2]})
+    msg = "Replace has to be set to `True` when upsampling the population `frac` > 1."
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a").sample(frac=1.5, replace=False)
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a")["b"].sample(frac=1.5, replace=False)
+
+
+@pytest.mark.parametrize("n", [-1, 1.5])
+def test_groupby_sample_invalid_n_raises(n):
+    df = DataFrame({"a": [1, 2], "b": [1, 2]})
+
+    if n < 0:
+        msg = "A negative number of rows requested. Please provide `n` >= 0."
+    else:
+        msg = "Only integers accepted as `n` values"
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a").sample(n=n)
+
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("a")["b"].sample(n=n)
+
+
+def test_groupby_sample_oversample():
+    values = [1] * 10 + [2] * 10
+    df = DataFrame({"a": values, "b": values})
+
+    result = df.groupby("a").sample(frac=2.0, replace=True)
+    values = [1] * 20 + [2] * 20
+    expected = DataFrame({"a": values, "b": values}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(frac=2.0, replace=True)
+    expected = Series(values, name="b", index=result.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_sample_without_n_or_frac():
+    values = [1] * 10 + [2] * 10
+    df = DataFrame({"a": values, "b": values})
+
+    result = df.groupby("a").sample(n=None, frac=None)
+    expected = DataFrame({"a": [1, 2], "b": [1, 2]}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(n=None, frac=None)
+    expected = Series([1, 2], name="b", index=result.index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index, expected_index",
+    [(["w", "x", "y", "z"], ["w", "w", "y", "y"]), ([3, 4, 5, 6], [3, 3, 5, 5])],
+)
+def test_groupby_sample_with_weights(index, expected_index):
+    # GH 39927 - tests for integer index needed
+    values = [1] * 2 + [2] * 2
+    df = DataFrame({"a": values, "b": values}, index=Index(index))
+
+    result = df.groupby("a").sample(n=2, replace=True, weights=[1, 0, 1, 0])
+    expected = DataFrame({"a": values, "b": values}, index=Index(expected_index))
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a")["b"].sample(n=2, replace=True, weights=[1, 0, 1, 0])
+    expected = Series(values, name="b", index=Index(expected_index))
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_sample_with_selections():
+    # GH 39928
+    values = [1] * 10 + [2] * 10
+    df = DataFrame({"a": values, "b": values, "c": values})
+
+    result = df.groupby("a")[["b", "c"]].sample(n=None, frac=None)
+    expected = DataFrame({"b": [1, 2], "c": [1, 2]}, index=result.index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_sample_with_empty_inputs():
+    # GH48459
+    df = DataFrame({"a": [], "b": []})
+    groupby_df = df.groupby("a")
+
+    result = groupby_df.sample()
+    expected = df
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_size.py b/pandas/pandas/tests/groupby/methods/test_size.py
new file mode 100644
index 0000000000000..6664563bd2272
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_size.py
@@ -0,0 +1,90 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    PeriodIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("by", ["A", "B", ["A", "B"]])
+def test_size(df, by):
+    grouped = df.groupby(by=by)
+    result = grouped.size()
+    for key, group in grouped:
+        assert result[key] == len(group)
+
+
+@pytest.mark.parametrize("by", ["A", "B", ["A", "B"]])
+def test_size_sort(sort, by):
+    df = DataFrame(np.random.default_rng(2).choice(20, (1000, 3)), columns=list("ABC"))
+    left = df.groupby(by=by, sort=sort).size()
+    right = df.groupby(by=by, sort=sort)["C"].apply(lambda a: a.shape[0])
+    tm.assert_series_equal(left, right, check_names=False)
+
+
+def test_size_series_dataframe():
+    # https://github.com/pandas-dev/pandas/issues/11699
+    df = DataFrame(columns=["A", "B"])
+    out = Series(dtype="int64", index=Index([], name="A"))
+    tm.assert_series_equal(df.groupby("A").size(), out)
+
+
+def test_size_groupby_all_null():
+    # https://github.com/pandas-dev/pandas/issues/23050
+    # Assert no 'Value Error : Length of passed values is 2, index implies 0'
+    df = DataFrame({"A": [None, None]})  # all-null groups
+    result = df.groupby("A").size()
+    expected = Series(dtype="int64", index=Index([], name="A"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_size_period_index():
+    # https://github.com/pandas-dev/pandas/issues/34010
+    ser = Series([1], index=PeriodIndex(["2000"], name="A", freq="D"))
+    grp = ser.groupby(level="A")
+    result = grp.size()
+    tm.assert_series_equal(result, ser)
+
+
+def test_size_on_categorical(as_index):
+    df = DataFrame([[1, 1], [2, 2]], columns=["A", "B"])
+    df["A"] = df["A"].astype("category")
+    result = df.groupby(["A", "B"], as_index=as_index, observed=False).size()
+
+    expected = DataFrame(
+        [[1, 1, 1], [1, 2, 0], [2, 1, 0], [2, 2, 1]], columns=["A", "B", "size"]
+    )
+    expected["A"] = expected["A"].astype("category")
+    if as_index:
+        expected = expected.set_index(["A", "B"])["size"].rename(None)
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["Int64", "Float64", "boolean"])
+def test_size_series_masked_type_returns_Int64(dtype):
+    # GH 54132
+    ser = Series([1, 1, 1], index=["a", "a", "b"], dtype=dtype)
+    result = ser.groupby(level=0).size()
+    expected = Series([2, 1], dtype="Int64", index=["a", "b"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_size_strings(any_string_dtype, using_infer_string):
+    # GH#55627
+    dtype = any_string_dtype
+    df = DataFrame({"a": ["a", "a", "b"], "b": "a"}, dtype=dtype)
+    result = df.groupby("a")["b"].size()
+    exp_dtype = "Int64" if dtype == "string[pyarrow]" else "int64"
+    exp_index_dtype = "str" if using_infer_string and dtype == "object" else dtype
+    expected = Series(
+        [2, 1],
+        index=Index(["a", "b"], name="a", dtype=exp_index_dtype),
+        name="b",
+        dtype=exp_dtype,
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_skew.py b/pandas/pandas/tests/groupby/methods/test_skew.py
new file mode 100644
index 0000000000000..563da89b6ab24
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_skew.py
@@ -0,0 +1,27 @@
+import numpy as np
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_groupby_skew_equivalence():
+    # Test that that groupby skew method (which uses libgroupby.group_skew)
+    #  matches the results of operating group-by-group (which uses nanops.nanskew)
+    nrows = 1000
+    ngroups = 3
+    ncols = 2
+    nan_frac = 0.05
+
+    arr = np.random.default_rng(2).standard_normal((nrows, ncols))
+    arr[np.random.default_rng(2).random(nrows) < nan_frac] = np.nan
+
+    df = pd.DataFrame(arr)
+    grps = np.random.default_rng(2).integers(0, ngroups, size=nrows)
+    gb = df.groupby(grps)
+
+    result = gb.skew()
+
+    grpwise = [grp.skew().to_frame(i).T for i, grp in gb]
+    expected = pd.concat(grpwise, axis=0)
+    expected.index = expected.index.astype(result.index.dtype)  # 32bit builds
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/methods/test_value_counts.py b/pandas/pandas/tests/groupby/methods/test_value_counts.py
new file mode 100644
index 0000000000000..1050f8154572a
--- /dev/null
+++ b/pandas/pandas/tests/groupby/methods/test_value_counts.py
@@ -0,0 +1,1240 @@
+"""
+these are systematically testing all of the args to value_counts
+with different size combinations. This is to ensure stability of the sorting
+and proper parameter handling
+"""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    Grouper,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+def tests_value_counts_index_names_category_column():
+    # GH44324 Missing name of index category column
+    df = DataFrame(
+        {
+            "gender": ["female"],
+            "country": ["US"],
+        }
+    )
+    df["gender"] = df["gender"].astype("category")
+    result = df.groupby("country")["gender"].value_counts()
+
+    # Construct expected, very specific multiindex
+    df_mi_expected = DataFrame([["US", "female"]], columns=["country", "gender"])
+    df_mi_expected["gender"] = df_mi_expected["gender"].astype("category")
+    mi_expected = MultiIndex.from_frame(df_mi_expected)
+    expected = Series([1], index=mi_expected, name="count")
+
+    tm.assert_series_equal(result, expected)
+
+
+def seed_df(seed_nans, n, m):
+    days = date_range("2015-08-24", periods=10)
+
+    frame = DataFrame(
+        {
+            "1st": np.random.default_rng(2).choice(list("abcd"), n),
+            "2nd": np.random.default_rng(2).choice(days, n),
+            "3rd": np.random.default_rng(2).integers(1, m + 1, n),
+        }
+    )
+
+    if seed_nans:
+        # Explicitly cast to float to avoid implicit cast when setting nan
+        frame["3rd"] = frame["3rd"].astype("float")
+        frame.loc[1::11, "1st"] = np.nan
+        frame.loc[3::17, "2nd"] = np.nan
+        frame.loc[7::19, "3rd"] = np.nan
+        frame.loc[8::19, "3rd"] = np.nan
+        frame.loc[9::19, "3rd"] = np.nan
+
+    return frame
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize("seed_nans", [True, False])
+@pytest.mark.parametrize("num_rows", [10, 50])
+@pytest.mark.parametrize("max_int", [5, 20])
+@pytest.mark.parametrize("keys", ["1st", "2nd", ["1st", "2nd"]], ids=repr)
+@pytest.mark.parametrize("bins", [None, [0, 5]], ids=repr)
+@pytest.mark.parametrize("isort", [True, False])
+@pytest.mark.parametrize("normalize, name", [(True, "proportion"), (False, "count")])
+def test_series_groupby_value_counts(
+    seed_nans,
+    num_rows,
+    max_int,
+    keys,
+    bins,
+    isort,
+    normalize,
+    name,
+    sort,
+    ascending,
+    dropna,
+):
+    df = seed_df(seed_nans, num_rows, max_int)
+
+    def rebuild_index(df):
+        arr = list(map(df.index.get_level_values, range(df.index.nlevels)))
+        df.index = MultiIndex.from_arrays(arr, names=df.index.names)
+        return df
+
+    kwargs = {
+        "normalize": normalize,
+        "sort": sort,
+        "ascending": ascending,
+        "dropna": dropna,
+        "bins": bins,
+    }
+
+    gr = df.groupby(keys, sort=isort)
+    left = gr["3rd"].value_counts(**kwargs)
+
+    gr = df.groupby(keys, sort=isort)
+    right = gr["3rd"].apply(Series.value_counts, **kwargs)
+    right.index.names = right.index.names[:-1] + ["3rd"]
+    # https://github.com/pandas-dev/pandas/issues/49909
+    right = right.rename(name)
+
+    # have to sort on index because of unstable sort on values
+    left, right = map(rebuild_index, (left, right))  # xref GH9212
+    tm.assert_series_equal(left.sort_index(), right.sort_index())
+
+
+@pytest.mark.parametrize("utc", [True, False])
+def test_series_groupby_value_counts_with_grouper(utc):
+    # GH28479
+    df = DataFrame(
+        {
+            "Timestamp": [
+                1565083561,
+                1565083561 + 86400,
+                1565083561 + 86500,
+                1565083561 + 86400 * 2,
+                1565083561 + 86400 * 3,
+                1565083561 + 86500 * 3,
+                1565083561 + 86400 * 4,
+            ],
+            "Food": ["apple", "apple", "banana", "banana", "orange", "orange", "pear"],
+        }
+    ).drop([3])
+
+    df["Datetime"] = to_datetime(df["Timestamp"], utc=utc, unit="s")
+    dfg = df.groupby(Grouper(freq="1D", key="Datetime"))
+
+    # have to sort on index because of unstable sort on values xref GH9212
+    result = dfg["Food"].value_counts().sort_index()
+    expected = dfg["Food"].apply(Series.value_counts).sort_index()
+    expected.index.names = result.index.names
+    # https://github.com/pandas-dev/pandas/issues/49909
+    expected = expected.rename("count")
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("columns", [["A", "B"], ["A", "B", "C"]])
+def test_series_groupby_value_counts_empty(columns):
+    # GH39172
+    df = DataFrame(columns=columns)
+    dfg = df.groupby(columns[:-1])
+
+    result = dfg[columns[-1]].value_counts()
+    expected = Series([], dtype=result.dtype, name="count")
+    expected.index = MultiIndex.from_arrays([[]] * len(columns), names=columns)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("columns", [["A", "B"], ["A", "B", "C"]])
+def test_series_groupby_value_counts_one_row(columns):
+    # GH42618
+    df = DataFrame(data=[range(len(columns))], columns=columns)
+    dfg = df.groupby(columns[:-1])
+
+    result = dfg[columns[-1]].value_counts()
+    expected = df.value_counts()
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_groupby_value_counts_on_categorical():
+    # GH38672
+
+    s = Series(Categorical(["a"], categories=["a", "b"]))
+    result = s.groupby([0]).value_counts()
+
+    expected = Series(
+        data=[1, 0],
+        index=MultiIndex.from_arrays(
+            [
+                np.array([0, 0]),
+                CategoricalIndex(
+                    ["a", "b"], categories=["a", "b"], ordered=False, dtype="category"
+                ),
+            ]
+        ),
+        name="count",
+    )
+
+    # Expected:
+    # 0  a    1
+    #    b    0
+    # dtype: int64
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_groupby_value_counts_no_sort():
+    # GH#50482
+    df = DataFrame(
+        {
+            "gender": ["male", "male", "female", "male", "female", "male"],
+            "education": ["low", "medium", "high", "low", "high", "low"],
+            "country": ["US", "FR", "US", "FR", "FR", "FR"],
+        }
+    )
+    gb = df.groupby(["country", "gender"], sort=False)["education"]
+    result = gb.value_counts(sort=False)
+    index = MultiIndex(
+        levels=[["US", "FR"], ["male", "female"], ["low", "medium", "high"]],
+        codes=[[0, 1, 0, 1, 1], [0, 0, 1, 0, 1], [0, 1, 2, 0, 2]],
+        names=["country", "gender", "education"],
+    )
+    expected = Series([1, 1, 1, 2, 1], index=index, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.fixture
+def education_df():
+    return DataFrame(
+        {
+            "gender": ["male", "male", "female", "male", "female", "male"],
+            "education": ["low", "medium", "high", "low", "high", "low"],
+            "country": ["US", "FR", "US", "FR", "FR", "FR"],
+        }
+    )
+
+
+def test_bad_subset(education_df):
+    gp = education_df.groupby("country")
+    with pytest.raises(ValueError, match="subset"):
+        gp.value_counts(subset=["country"])
+
+
+def test_basic(education_df, request):
+    # gh43564
+    if Version(np.__version__) >= Version("1.25"):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+    result = education_df.groupby("country")[["gender", "education"]].value_counts(
+        normalize=True
+    )
+    expected = Series(
+        data=[0.5, 0.25, 0.25, 0.5, 0.5],
+        index=MultiIndex.from_tuples(
+            [
+                ("FR", "male", "low"),
+                ("FR", "male", "medium"),
+                ("FR", "female", "high"),
+                ("US", "male", "low"),
+                ("US", "female", "high"),
+            ],
+            names=["country", "gender", "education"],
+        ),
+        name="proportion",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def _frame_value_counts(df, keys, normalize, sort, ascending):
+    return df[keys].value_counts(normalize=normalize, sort=sort, ascending=ascending)
+
+
+@pytest.mark.parametrize("groupby", ["column", "array", "function"])
+@pytest.mark.parametrize("normalize, name", [(True, "proportion"), (False, "count")])
+@pytest.mark.parametrize(
+    "sort, ascending",
+    [
+        (False, None),
+        (True, True),
+        (True, False),
+    ],
+)
+@pytest.mark.parametrize("frame", [True, False])
+def test_against_frame_and_seriesgroupby(
+    education_df,
+    groupby,
+    normalize,
+    name,
+    sort,
+    ascending,
+    as_index,
+    frame,
+    request,
+    using_infer_string,
+):
+    # test all parameters:
+    # - Use column, array or function as by= parameter
+    # - Whether or not to normalize
+    # - Whether or not to sort and how
+    # - Whether or not to use the groupby as an index
+    # - 3-way compare against:
+    #   - apply with :meth:`~DataFrame.value_counts`
+    #   - `~SeriesGroupBy.value_counts`
+    if Version(np.__version__) >= Version("1.25") and frame and sort and normalize:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+    by = {
+        "column": "country",
+        "array": education_df["country"].values,
+        "function": lambda x: education_df["country"][x] == "US",
+    }[groupby]
+
+    gp = education_df.groupby(by=by, as_index=as_index)
+    result = gp[["gender", "education"]].value_counts(
+        normalize=normalize, sort=sort, ascending=ascending
+    )
+    if frame:
+        # compare against apply with DataFrame value_counts
+        expected = gp.apply(
+            _frame_value_counts, ["gender", "education"], normalize, sort, ascending
+        )
+
+        if as_index:
+            tm.assert_series_equal(result, expected)
+        else:
+            name = "proportion" if normalize else "count"
+            expected = expected.reset_index().rename({0: name}, axis=1)
+            if groupby in ["array", "function"] and (not as_index and frame):
+                expected.insert(loc=0, column="level_0", value=result["level_0"])
+            else:
+                expected.insert(loc=0, column="country", value=result["country"])
+            tm.assert_frame_equal(result, expected)
+    else:
+        # compare against SeriesGroupBy value_counts
+        education_df["both"] = education_df["gender"] + "-" + education_df["education"]
+        expected = gp["both"].value_counts(
+            normalize=normalize, sort=sort, ascending=ascending
+        )
+        expected.name = name
+        if as_index:
+            index_frame = expected.index.to_frame(index=False)
+            index_frame["gender"] = index_frame["both"].str.split("-").str.get(0)
+            index_frame["education"] = index_frame["both"].str.split("-").str.get(1)
+            del index_frame["both"]
+            index_frame2 = index_frame.rename({0: None}, axis=1)
+            expected.index = MultiIndex.from_frame(index_frame2)
+
+            if index_frame2.columns.isna()[0]:
+                # with using_infer_string, the columns in index_frame as string
+                #  dtype, which makes the rename({0: None}) above use np.nan
+                #  instead of None, so we need to set None more explicitly.
+                expected.index.names = [None] + expected.index.names[1:]
+            tm.assert_series_equal(result, expected)
+        else:
+            expected.insert(1, "gender", expected["both"].str.split("-").str.get(0))
+            expected.insert(2, "education", expected["both"].str.split("-").str.get(1))
+            if using_infer_string:
+                expected = expected.astype({"gender": "str", "education": "str"})
+            del expected["both"]
+            tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("normalize", [True, False])
+@pytest.mark.parametrize(
+    "sort, ascending, expected_rows, expected_count, expected_group_size",
+    [
+        (False, None, [0, 1, 2, 3, 4], [1, 1, 1, 2, 1], [1, 3, 1, 3, 1]),
+        (True, False, [3, 0, 1, 2, 4], [2, 1, 1, 1, 1], [3, 1, 3, 1, 1]),
+        (True, True, [0, 1, 2, 4, 3], [1, 1, 1, 1, 2], [1, 3, 1, 1, 3]),
+    ],
+)
+def test_compound(
+    education_df,
+    normalize,
+    sort,
+    ascending,
+    expected_rows,
+    expected_count,
+    expected_group_size,
+    any_string_dtype,
+    using_infer_string,
+):
+    dtype = any_string_dtype
+    education_df = education_df.astype(dtype)
+    education_df.columns = education_df.columns.astype(dtype)
+    # Multiple groupby keys and as_index=False
+    gp = education_df.groupby(["country", "gender"], as_index=False, sort=False)
+    result = gp["education"].value_counts(
+        normalize=normalize, sort=sort, ascending=ascending
+    )
+    expected = DataFrame()
+    for column in ["country", "gender", "education"]:
+        expected[column] = [education_df[column][row] for row in expected_rows]
+        expected = expected.astype(dtype)
+        expected.columns = expected.columns.astype(dtype)
+    if normalize:
+        expected["proportion"] = expected_count
+        expected["proportion"] /= expected_group_size
+        if dtype == "string[pyarrow]":
+            # TODO(nullable) also string[python] should return nullable dtypes
+            expected["proportion"] = expected["proportion"].convert_dtypes()
+    else:
+        expected["count"] = expected_count
+        if dtype == "string[pyarrow]":
+            expected["count"] = expected["count"].convert_dtypes()
+    if using_infer_string and dtype == object:
+        expected = expected.astype(
+            {"country": "str", "gender": "str", "education": "str"}
+        )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "sort, ascending, normalize, name, expected_data, expected_index",
+    [
+        (False, None, False, "count", [1, 2, 1], [(1, 1, 1), (2, 4, 6), (2, 0, 0)]),
+        (True, True, False, "count", [1, 1, 2], [(1, 1, 1), (2, 6, 4), (2, 0, 0)]),
+        (True, False, False, "count", [2, 1, 1], [(1, 1, 1), (4, 2, 6), (0, 2, 0)]),
+        (
+            True,
+            False,
+            True,
+            "proportion",
+            [0.5, 0.25, 0.25],
+            [(1, 1, 1), (4, 2, 6), (0, 2, 0)],
+        ),
+    ],
+)
+def test_data_frame_value_counts(
+    sort, ascending, normalize, name, expected_data, expected_index
+):
+    # 3-way compare with :meth:`~DataFrame.value_counts`
+    # Tests from frame/methods/test_value_counts.py
+    animals_df = DataFrame(
+        {"key": [1, 1, 1, 1], "num_legs": [2, 4, 4, 6], "num_wings": [2, 0, 0, 0]},
+        index=["falcon", "dog", "cat", "ant"],
+    )
+    result_frame = animals_df.value_counts(
+        sort=sort, ascending=ascending, normalize=normalize
+    )
+    expected = Series(
+        data=expected_data,
+        index=MultiIndex.from_arrays(
+            expected_index, names=["key", "num_legs", "num_wings"]
+        ),
+        name=name,
+    )
+    tm.assert_series_equal(result_frame, expected)
+
+    result_frame_groupby = animals_df.groupby("key").value_counts(
+        sort=sort, ascending=ascending, normalize=normalize
+    )
+
+    tm.assert_series_equal(result_frame_groupby, expected)
+
+
+@pytest.mark.parametrize(
+    "group_dropna, count_dropna, expected_rows, expected_values",
+    [
+        (
+            False,
+            False,
+            [0, 1, 3, 5, 6, 7, 8, 2, 4],
+            [0.5, 0.5, 1.0, 0.25, 0.25, 0.25, 0.25, 1.0, 1.0],
+        ),
+        (False, True, [0, 1, 3, 5, 2, 4], [0.5, 0.5, 1.0, 1.0, 1.0, 1.0]),
+        (True, False, [0, 1, 5, 6, 7, 8], [0.5, 0.5, 0.25, 0.25, 0.25, 0.25]),
+        (True, True, [0, 1, 5], [0.5, 0.5, 1.0]),
+    ],
+)
+def test_dropna_combinations(
+    group_dropna, count_dropna, expected_rows, expected_values, request
+):
+    if Version(np.__version__) >= Version("1.25") and not group_dropna:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+    nulls_df = DataFrame(
+        {
+            "A": [1, 1, np.nan, 4, np.nan, 6, 6, 6, 6],
+            "B": [1, 1, 3, np.nan, np.nan, 6, 6, 6, 6],
+            "C": [1, 2, 3, 4, 5, 6, np.nan, 8, np.nan],
+            "D": [1, 2, 3, 4, 5, 6, 7, np.nan, np.nan],
+        }
+    )
+    gp = nulls_df.groupby(["A", "B"], dropna=group_dropna)
+    result = gp.value_counts(normalize=True, sort=True, dropna=count_dropna)
+    columns = DataFrame()
+    for column in nulls_df.columns:
+        columns[column] = [nulls_df[column][row] for row in expected_rows]
+    index = MultiIndex.from_frame(columns)
+    expected = Series(data=expected_values, index=index, name="proportion")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, expected_data, expected_index",
+    [
+        (
+            True,
+            [1, 1],
+            MultiIndex.from_arrays(
+                [(1, 1), ("John", "Beth"), ("Smith", "Louise")],
+                names=["key", "first_name", "middle_name"],
+            ),
+        ),
+        (
+            False,
+            [1, 1, 1, 1],
+            MultiIndex(
+                levels=[
+                    Index([1]),
+                    Index(["Anne", "Beth", "John"]),
+                    Index(["Louise", "Smith", np.nan]),
+                ],
+                codes=[[0, 0, 0, 0], [2, 0, 2, 1], [1, 2, 2, 0]],
+                names=["key", "first_name", "middle_name"],
+            ),
+        ),
+    ],
+)
+@pytest.mark.parametrize("normalize, name", [(False, "count"), (True, "proportion")])
+def test_data_frame_value_counts_dropna(
+    nulls_fixture, dropna, normalize, name, expected_data, expected_index
+):
+    # GH 41334
+    # 3-way compare with :meth:`~DataFrame.value_counts`
+    # Tests with nulls from frame/methods/test_value_counts.py
+    names_with_nulls_df = DataFrame(
+        {
+            "key": [1, 1, 1, 1],
+            "first_name": ["John", "Anne", "John", "Beth"],
+            "middle_name": ["Smith", nulls_fixture, nulls_fixture, "Louise"],
+        },
+    )
+    result_frame = names_with_nulls_df.value_counts(dropna=dropna, normalize=normalize)
+    expected = Series(
+        data=expected_data,
+        index=expected_index,
+        name=name,
+    )
+    if normalize:
+        expected /= float(len(expected_data))
+
+    tm.assert_series_equal(result_frame, expected)
+
+    result_frame_groupby = names_with_nulls_df.groupby("key").value_counts(
+        dropna=dropna, normalize=normalize
+    )
+
+    tm.assert_series_equal(result_frame_groupby, expected)
+
+
+@pytest.mark.parametrize("observed", [False, True])
+@pytest.mark.parametrize(
+    "normalize, name, expected_data",
+    [
+        (
+            False,
+            "count",
+            np.array([2, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0], dtype=np.int64),
+        ),
+        (
+            True,
+            "proportion",
+            np.array([0.5, 0.25, 0.25, 0.0, 0.0, 0.0, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0]),
+        ),
+    ],
+)
+def test_categorical_single_grouper_with_only_observed_categories(
+    education_df, as_index, observed, normalize, name, expected_data, request
+):
+    # Test single categorical grouper with only observed grouping categories
+    # when non-groupers are also categorical
+    if Version(np.__version__) >= Version("1.25"):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+
+    gp = education_df.astype("category").groupby(
+        "country", as_index=as_index, observed=observed
+    )
+    result = gp.value_counts(normalize=normalize)
+
+    expected_index = MultiIndex.from_tuples(
+        [
+            ("FR", "male", "low"),
+            ("FR", "male", "medium"),
+            ("FR", "female", "high"),
+            ("FR", "male", "high"),
+            ("FR", "female", "low"),
+            ("FR", "female", "medium"),
+            ("US", "male", "low"),
+            ("US", "female", "high"),
+            ("US", "male", "medium"),
+            ("US", "male", "high"),
+            ("US", "female", "low"),
+            ("US", "female", "medium"),
+        ],
+        names=["country", "gender", "education"],
+    )
+
+    expected_series = Series(
+        data=expected_data,
+        index=expected_index,
+        name=name,
+    )
+    for i in range(3):
+        expected_series.index = expected_series.index.set_levels(
+            CategoricalIndex(expected_series.index.levels[i]), level=i
+        )
+
+    if as_index:
+        tm.assert_series_equal(result, expected_series)
+    else:
+        expected = expected_series.reset_index(
+            name="proportion" if normalize else "count"
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def assert_categorical_single_grouper(
+    education_df, as_index, observed, expected_index, normalize, name, expected_data
+):
+    # Test single categorical grouper when non-groupers are also categorical
+    education_df = education_df.copy().astype("category")
+
+    # Add non-observed grouping categories
+    education_df["country"] = education_df["country"].cat.add_categories(["ASIA"])
+
+    gp = education_df.groupby("country", as_index=as_index, observed=observed)
+    result = gp.value_counts(normalize=normalize)
+
+    expected_series = Series(
+        data=expected_data,
+        index=MultiIndex.from_tuples(
+            expected_index,
+            names=["country", "gender", "education"],
+        ),
+        name=name,
+    )
+    for i in range(3):
+        index_level = CategoricalIndex(expected_series.index.levels[i])
+        if i == 0:
+            index_level = index_level.set_categories(
+                education_df["country"].cat.categories
+            )
+        expected_series.index = expected_series.index.set_levels(index_level, level=i)
+
+    if as_index:
+        tm.assert_series_equal(result, expected_series)
+    else:
+        expected = expected_series.reset_index(name=name)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "normalize, name, expected_data",
+    [
+        (
+            False,
+            "count",
+            np.array([2, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0], dtype=np.int64),
+        ),
+        (
+            True,
+            "proportion",
+            np.array([0.5, 0.25, 0.25, 0.0, 0.0, 0.0, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0]),
+        ),
+    ],
+)
+def test_categorical_single_grouper_observed_true(
+    education_df, as_index, normalize, name, expected_data, request
+):
+    # GH#46357
+
+    if Version(np.__version__) >= Version("1.25"):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+
+    expected_index = [
+        ("FR", "male", "low"),
+        ("FR", "male", "medium"),
+        ("FR", "female", "high"),
+        ("FR", "male", "high"),
+        ("FR", "female", "low"),
+        ("FR", "female", "medium"),
+        ("US", "male", "low"),
+        ("US", "female", "high"),
+        ("US", "male", "medium"),
+        ("US", "male", "high"),
+        ("US", "female", "low"),
+        ("US", "female", "medium"),
+    ]
+
+    assert_categorical_single_grouper(
+        education_df=education_df,
+        as_index=as_index,
+        observed=True,
+        expected_index=expected_index,
+        normalize=normalize,
+        name=name,
+        expected_data=expected_data,
+    )
+
+
+@pytest.mark.parametrize(
+    "normalize, name, expected_data",
+    [
+        (
+            False,
+            "count",
+            np.array(
+                [2, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], dtype=np.int64
+            ),
+        ),
+        (
+            True,
+            "proportion",
+            np.array(
+                [
+                    0.5,
+                    0.25,
+                    0.25,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.5,
+                    0.5,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                ]
+            ),
+        ),
+    ],
+)
+def test_categorical_single_grouper_observed_false(
+    education_df, as_index, normalize, name, expected_data, request
+):
+    # GH#46357
+
+    if Version(np.__version__) >= Version("1.25"):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+
+    expected_index = [
+        ("FR", "male", "low"),
+        ("FR", "male", "medium"),
+        ("FR", "female", "high"),
+        ("FR", "male", "high"),
+        ("FR", "female", "low"),
+        ("FR", "female", "medium"),
+        ("US", "male", "low"),
+        ("US", "female", "high"),
+        ("US", "male", "medium"),
+        ("US", "male", "high"),
+        ("US", "female", "low"),
+        ("US", "female", "medium"),
+        ("ASIA", "male", "low"),
+        ("ASIA", "male", "medium"),
+        ("ASIA", "male", "high"),
+        ("ASIA", "female", "low"),
+        ("ASIA", "female", "medium"),
+        ("ASIA", "female", "high"),
+    ]
+
+    assert_categorical_single_grouper(
+        education_df=education_df,
+        as_index=as_index,
+        observed=False,
+        expected_index=expected_index,
+        normalize=normalize,
+        name=name,
+        expected_data=expected_data,
+    )
+
+
+@pytest.mark.parametrize(
+    "observed, expected_index",
+    [
+        (
+            False,
+            [
+                ("FR", "high", "female"),
+                ("FR", "high", "male"),
+                ("FR", "low", "male"),
+                ("FR", "low", "female"),
+                ("FR", "medium", "male"),
+                ("FR", "medium", "female"),
+                ("US", "high", "female"),
+                ("US", "high", "male"),
+                ("US", "low", "male"),
+                ("US", "low", "female"),
+                ("US", "medium", "male"),
+                ("US", "medium", "female"),
+            ],
+        ),
+        (
+            True,
+            [
+                ("FR", "high", "female"),
+                ("FR", "low", "male"),
+                ("FR", "medium", "male"),
+                ("US", "high", "female"),
+                ("US", "low", "male"),
+            ],
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    "normalize, name, expected_data",
+    [
+        (
+            False,
+            "count",
+            np.array([1, 0, 2, 0, 1, 0, 1, 0, 1, 0, 0, 0], dtype=np.int64),
+        ),
+        (
+            True,
+            "proportion",
+            # NaN values corresponds to non-observed groups
+            np.array([1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 0.0, 0.0]),
+        ),
+    ],
+)
+def test_categorical_multiple_groupers(
+    education_df, as_index, observed, expected_index, normalize, name, expected_data
+):
+    # GH#46357
+
+    # Test multiple categorical groupers when non-groupers are non-categorical
+    education_df = education_df.copy()
+    education_df["country"] = education_df["country"].astype("category")
+    education_df["education"] = education_df["education"].astype("category")
+
+    gp = education_df.groupby(
+        ["country", "education"], as_index=as_index, observed=observed
+    )
+    result = gp.value_counts(normalize=normalize)
+
+    expected_series = Series(
+        data=expected_data[expected_data > 0.0] if observed else expected_data,
+        index=MultiIndex.from_tuples(
+            expected_index,
+            names=["country", "education", "gender"],
+        ),
+        name=name,
+    )
+    for i in range(2):
+        expected_series.index = expected_series.index.set_levels(
+            CategoricalIndex(expected_series.index.levels[i]), level=i
+        )
+
+    if as_index:
+        tm.assert_series_equal(result, expected_series)
+    else:
+        expected = expected_series.reset_index(
+            name="proportion" if normalize else "count"
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("observed", [False, True])
+@pytest.mark.parametrize(
+    "normalize, name, expected_data",
+    [
+        (
+            False,
+            "count",
+            np.array([2, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0], dtype=np.int64),
+        ),
+        (
+            True,
+            "proportion",
+            # NaN values corresponds to non-observed groups
+            np.array([0.5, 0.25, 0.25, 0.0, 0.0, 0.0, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0]),
+        ),
+    ],
+)
+def test_categorical_non_groupers(
+    education_df, as_index, observed, normalize, name, expected_data, request
+):
+    # GH#46357 Test non-observed categories are included in the result,
+    # regardless of `observed`
+
+    if Version(np.__version__) >= Version("1.25"):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "pandas default unstable sorting of duplicates"
+                    "issue with numpy>=1.25 with AVX instructions"
+                ),
+                strict=False,
+            )
+        )
+
+    education_df = education_df.copy()
+    education_df["gender"] = education_df["gender"].astype("category")
+    education_df["education"] = education_df["education"].astype("category")
+
+    gp = education_df.groupby("country", as_index=as_index, observed=observed)
+    result = gp.value_counts(normalize=normalize)
+
+    expected_index = [
+        ("FR", "male", "low"),
+        ("FR", "male", "medium"),
+        ("FR", "female", "high"),
+        ("FR", "male", "high"),
+        ("FR", "female", "low"),
+        ("FR", "female", "medium"),
+        ("US", "male", "low"),
+        ("US", "female", "high"),
+        ("US", "male", "medium"),
+        ("US", "male", "high"),
+        ("US", "female", "low"),
+        ("US", "female", "medium"),
+    ]
+    expected_series = Series(
+        data=expected_data,
+        index=MultiIndex.from_tuples(
+            expected_index,
+            names=["country", "gender", "education"],
+        ),
+        name=name,
+    )
+    for i in range(1, 3):
+        expected_series.index = expected_series.index.set_levels(
+            CategoricalIndex(expected_series.index.levels[i]), level=i
+        )
+
+    if as_index:
+        tm.assert_series_equal(result, expected_series)
+    else:
+        expected = expected_series.reset_index(
+            name="proportion" if normalize else "count"
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "normalize, expected_label, expected_values",
+    [
+        (False, "count", [1, 1, 1]),
+        (True, "proportion", [0.5, 0.5, 1.0]),
+    ],
+)
+def test_mixed_groupings(normalize, expected_label, expected_values):
+    # Test multiple groupings
+    df = DataFrame({"A": [1, 2, 1], "B": [1, 2, 3]})
+    gp = df.groupby([[4, 5, 4], "A", lambda i: 7 if i == 1 else 8], as_index=False)
+    result = gp.value_counts(sort=True, normalize=normalize)
+    expected = DataFrame(
+        {
+            "level_0": np.array([4, 4, 5], dtype=int),
+            "A": [1, 1, 2],
+            "level_2": [8, 8, 7],
+            "B": [1, 3, 2],
+            expected_label: expected_values,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "test, columns, expected_names",
+    [
+        ("repeat", list("abbde"), ["a", None, "d", "b", "b", "e"]),
+        ("level", list("abcd") + ["level_1"], ["a", None, "d", "b", "c", "level_1"]),
+    ],
+)
+def test_column_label_duplicates(test, columns, expected_names, as_index):
+    # GH 44992
+    # Test for duplicate input column labels and generated duplicate labels
+    df = DataFrame([[1, 3, 5, 7, 9], [2, 4, 6, 8, 10]], columns=columns)
+    expected_data = [(1, 0, 7, 3, 5, 9), (2, 1, 8, 4, 6, 10)]
+    keys = ["a", np.array([0, 1], dtype=np.int64), "d"]
+    result = df.groupby(keys, as_index=as_index).value_counts()
+    if as_index:
+        expected = Series(
+            data=(1, 1),
+            index=MultiIndex.from_tuples(
+                expected_data,
+                names=expected_names,
+            ),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected)
+    else:
+        expected_data = [list(row) + [1] for row in expected_data]
+        expected_columns = list(expected_names)
+        expected_columns[1] = "level_1"
+        expected_columns.append("count")
+        expected = DataFrame(expected_data, columns=expected_columns)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "normalize, expected_label",
+    [
+        (False, "count"),
+        (True, "proportion"),
+    ],
+)
+def test_result_label_duplicates(normalize, expected_label):
+    # Test for result column label duplicating an input column label
+    gb = DataFrame([[1, 2, 3]], columns=["a", "b", expected_label]).groupby(
+        "a", as_index=False
+    )
+    msg = f"Column label '{expected_label}' is duplicate of result column"
+    with pytest.raises(ValueError, match=msg):
+        gb.value_counts(normalize=normalize)
+
+
+def test_ambiguous_grouping():
+    # Test that groupby is not confused by groupings length equal to row count
+    df = DataFrame({"a": [1, 1]})
+    gb = df.groupby(np.array([1, 1], dtype=np.int64))
+    result = gb.value_counts()
+    expected = Series(
+        [2], index=MultiIndex.from_tuples([[1, 1]], names=[None, "a"]), name="count"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_subset_overlaps_gb_key_raises():
+    # GH 46383
+    df = DataFrame({"c1": ["a", "b", "c"], "c2": ["x", "y", "y"]}, index=[0, 1, 1])
+    msg = "Keys {'c1'} in subset cannot be in the groupby column keys."
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("c1").value_counts(subset=["c1"])
+
+
+def test_subset_doesnt_exist_in_frame():
+    # GH 46383
+    df = DataFrame({"c1": ["a", "b", "c"], "c2": ["x", "y", "y"]}, index=[0, 1, 1])
+    msg = "Keys {'c3'} in subset do not exist in the DataFrame."
+    with pytest.raises(ValueError, match=msg):
+        df.groupby("c1").value_counts(subset=["c3"])
+
+
+def test_subset():
+    # GH 46383
+    df = DataFrame({"c1": ["a", "b", "c"], "c2": ["x", "y", "y"]}, index=[0, 1, 1])
+    result = df.groupby(level=0).value_counts(subset=["c2"])
+    expected = Series(
+        [1, 2],
+        index=MultiIndex.from_arrays([[0, 1], ["x", "y"]], names=[None, "c2"]),
+        name="count",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_subset_duplicate_columns():
+    # GH 46383
+    df = DataFrame(
+        [["a", "x", "x"], ["b", "y", "y"], ["b", "y", "y"]],
+        index=[0, 1, 1],
+        columns=["c1", "c2", "c2"],
+    )
+    result = df.groupby(level=0).value_counts(subset=["c2"])
+    expected = Series(
+        [1, 2],
+        index=MultiIndex.from_arrays(
+            [[0, 1], ["x", "y"], ["x", "y"]], names=[None, "c2", "c2"]
+        ),
+        name="count",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("utc", [True, False])
+def test_value_counts_time_grouper(utc, unit):
+    # GH#50486
+    df = DataFrame(
+        {
+            "Timestamp": [
+                1565083561,
+                1565083561 + 86400,
+                1565083561 + 86500,
+                1565083561 + 86400 * 2,
+                1565083561 + 86400 * 3,
+                1565083561 + 86500 * 3,
+                1565083561 + 86400 * 4,
+            ],
+            "Food": ["apple", "apple", "banana", "banana", "orange", "orange", "pear"],
+        }
+    ).drop([3])
+
+    df["Datetime"] = to_datetime(df["Timestamp"], utc=utc, unit="s").dt.as_unit(unit)
+    gb = df.groupby(Grouper(freq="1D", key="Datetime"))
+    result = gb.value_counts()
+    dates = to_datetime(
+        ["2019-08-06", "2019-08-07", "2019-08-09", "2019-08-10"], utc=utc
+    ).as_unit(unit)
+    timestamps = df["Timestamp"].unique()
+    index = MultiIndex(
+        levels=[dates, timestamps, ["apple", "banana", "orange", "pear"]],
+        codes=[[0, 1, 1, 2, 2, 3], range(6), [0, 0, 1, 2, 2, 3]],
+        names=["Datetime", "Timestamp", "Food"],
+    )
+    expected = Series(1, index=index, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_value_counts_integer_columns():
+    # GH#55627
+    df = DataFrame({1: ["a", "a", "a"], 2: ["a", "a", "d"], 3: ["a", "b", "c"]})
+    gp = df.groupby([1, 2], as_index=False, sort=False)
+    result = gp[3].value_counts()
+    expected = DataFrame(
+        {1: ["a", "a", "a"], 2: ["a", "a", "d"], 3: ["a", "b", "c"], "count": 1}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("vc_sort", [True, False])
+@pytest.mark.parametrize("normalize", [True, False])
+def test_value_counts_sort(sort, vc_sort, normalize):
+    # GH#55951
+    df = DataFrame({"a": [2, 1, 1, 1], 0: [3, 4, 3, 3]})
+    gb = df.groupby("a", sort=sort)
+    result = gb.value_counts(sort=vc_sort, normalize=normalize)
+
+    if normalize:
+        values = [2 / 3, 1 / 3, 1.0]
+    else:
+        values = [2, 1, 1]
+    index = MultiIndex(
+        levels=[[1, 2], [3, 4]], codes=[[0, 0, 1], [0, 1, 0]], names=["a", 0]
+    )
+    expected = Series(values, index=index, name="proportion" if normalize else "count")
+    if sort and vc_sort:
+        taker = [0, 1, 2]
+    elif sort and not vc_sort:
+        taker = [1, 0, 2]
+    elif not sort and vc_sort:
+        taker = [0, 2, 1]
+    else:
+        taker = [2, 1, 0]
+    expected = expected.take(taker)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("vc_sort", [True, False])
+@pytest.mark.parametrize("normalize", [True, False])
+def test_value_counts_sort_categorical(sort, vc_sort, normalize):
+    # GH#55951
+    df = DataFrame({"a": [2, 1, 1, 1], 0: [3, 4, 3, 3]}, dtype="category")
+    gb = df.groupby("a", sort=sort, observed=True)
+    result = gb.value_counts(sort=vc_sort, normalize=normalize)
+
+    if normalize:
+        values = [2 / 3, 1 / 3, 1.0, 0.0]
+    else:
+        values = [2, 1, 1, 0]
+    name = "proportion" if normalize else "count"
+    expected = DataFrame(
+        {
+            "a": Categorical([1, 1, 2, 2]),
+            0: Categorical([3, 4, 3, 4]),
+            name: values,
+        }
+    ).set_index(["a", 0])[name]
+    if sort and vc_sort:
+        taker = [0, 1, 2, 3]
+    elif sort and not vc_sort:
+        taker = [0, 1, 2, 3]
+    elif not sort and vc_sort:
+        taker = [0, 2, 1, 3]
+    else:
+        taker = [2, 1, 0, 3]
+    expected = expected.take(taker)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("groupby_sort", [True, False])
+def test_value_counts_all_na(sort, dropna, groupby_sort):
+    # GH#59989
+    df = DataFrame({"a": [2, 1, 1], "b": np.nan})
+    gb = df.groupby("a", sort=groupby_sort)
+    result = gb.value_counts(sort=sort, dropna=dropna)
+
+    kwargs = {"levels": [[1, 2], [np.nan]], "names": ["a", "b"]}
+    if dropna:
+        data = []
+        index = MultiIndex(codes=[[], []], **kwargs)
+    elif not groupby_sort and not sort:
+        data = [1, 2]
+        index = MultiIndex(codes=[[1, 0], [0, 0]], **kwargs)
+    else:
+        data = [2, 1]
+        index = MultiIndex(codes=[[0, 1], [0, 0]], **kwargs)
+    expected = Series(data, index=index, dtype="int64", name="count")
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_all_methods.py b/pandas/pandas/tests/groupby/test_all_methods.py
new file mode 100644
index 0000000000000..4625c5c27a803
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_all_methods.py
@@ -0,0 +1,84 @@
+"""
+Tests that apply to all groupby operation methods.
+
+The only tests that should appear here are those that use the `groupby_func` fixture.
+Even if it does use that fixture, prefer a more specific test file if it available
+such as:
+
+ - test_categorical
+ - test_groupby_dropna
+ - test_groupby_subclass
+ - test_raises
+"""
+
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+def test_multiindex_group_all_columns_when_empty(groupby_func):
+    # GH 32464
+    df = DataFrame({"a": [], "b": [], "c": []}).set_index(["a", "b", "c"])
+    gb = df.groupby(["a", "b", "c"], group_keys=True)
+    method = getattr(gb, groupby_func)
+    args = get_groupby_method_args(groupby_func, df)
+    if groupby_func == "corrwith":
+        warn = FutureWarning
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        warn_msg = ""
+    with tm.assert_produces_warning(warn, match=warn_msg):
+        result = method(*args).index
+    expected = df.index
+    tm.assert_index_equal(result, expected)
+
+
+def test_duplicate_columns(request, groupby_func, as_index):
+    # GH#50806
+    if groupby_func == "corrwith":
+        msg = "GH#50845 - corrwith fails when there are duplicate columns"
+        request.applymarker(pytest.mark.xfail(reason=msg))
+    df = DataFrame([[1, 3, 6], [1, 4, 7], [2, 5, 8]], columns=list("abb"))
+    args = get_groupby_method_args(groupby_func, df)
+    gb = df.groupby("a", as_index=as_index)
+    result = getattr(gb, groupby_func)(*args)
+
+    expected_df = df.set_axis(["a", "b", "c"], axis=1)
+    expected_args = get_groupby_method_args(groupby_func, expected_df)
+    expected_gb = expected_df.groupby("a", as_index=as_index)
+    expected = getattr(expected_gb, groupby_func)(*expected_args)
+    if groupby_func not in ("size", "ngroup", "cumcount"):
+        expected = expected.rename(columns={"c": "b"})
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "idx",
+    [
+        pd.Index(["a", "a"], name="foo"),
+        pd.MultiIndex.from_tuples((("a", "a"), ("a", "a")), names=["foo", "bar"]),
+    ],
+)
+def test_dup_labels_output_shape(groupby_func, idx):
+    if groupby_func in {"size", "ngroup", "cumcount"}:
+        pytest.skip(f"Not applicable for {groupby_func}")
+
+    df = DataFrame([[1, 1]], columns=idx)
+    grp_by = df.groupby([0])
+
+    args = get_groupby_method_args(groupby_func, df)
+    if groupby_func == "corrwith":
+        warn = FutureWarning
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        warn_msg = ""
+    with tm.assert_produces_warning(warn, match=warn_msg):
+        result = getattr(grp_by, groupby_func)(*args)
+
+    assert result.shape == (1, 2)
+    tm.assert_index_equal(result.columns, idx)
diff --git a/pandas/pandas/tests/groupby/test_api.py b/pandas/pandas/tests/groupby/test_api.py
new file mode 100644
index 0000000000000..215e627abb018
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_api.py
@@ -0,0 +1,274 @@
+"""
+Tests of the groupby API, including internal consistency and with other pandas objects.
+
+Tests in this file should only check the existence, names, and arguments of groupby
+methods. It should not test the results of any groupby operation.
+"""
+
+import inspect
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+from pandas.core.groupby.base import (
+    groupby_other_methods,
+    reduction_kernels,
+    transformation_kernels,
+)
+from pandas.core.groupby.generic import (
+    DataFrameGroupBy,
+    SeriesGroupBy,
+)
+
+
+def test_tab_completion(multiindex_dataframe_random_data):
+    grp = multiindex_dataframe_random_data.groupby(level="second")
+    results = {v for v in dir(grp) if not v.startswith("_")}
+    expected = {
+        "A",
+        "B",
+        "C",
+        "agg",
+        "aggregate",
+        "apply",
+        "boxplot",
+        "filter",
+        "first",
+        "get_group",
+        "groups",
+        "hist",
+        "indices",
+        "last",
+        "max",
+        "mean",
+        "median",
+        "min",
+        "ngroups",
+        "nth",
+        "ohlc",
+        "plot",
+        "prod",
+        "size",
+        "std",
+        "sum",
+        "transform",
+        "var",
+        "sem",
+        "count",
+        "nunique",
+        "head",
+        "describe",
+        "cummax",
+        "quantile",
+        "rank",
+        "cumprod",
+        "tail",
+        "resample",
+        "cummin",
+        "cumsum",
+        "cumcount",
+        "ngroup",
+        "all",
+        "shift",
+        "skew",
+        "kurt",
+        "take",
+        "pct_change",
+        "any",
+        "corr",
+        "corrwith",
+        "cov",
+        "ndim",
+        "diff",
+        "idxmax",
+        "idxmin",
+        "ffill",
+        "bfill",
+        "rolling",
+        "expanding",
+        "pipe",
+        "sample",
+        "ewm",
+        "value_counts",
+    }
+    assert results == expected
+
+
+def test_all_methods_categorized(multiindex_dataframe_random_data):
+    grp = multiindex_dataframe_random_data.groupby(
+        multiindex_dataframe_random_data.iloc[:, 0]
+    )
+    names = {_ for _ in dir(grp) if not _.startswith("_")} - set(
+        multiindex_dataframe_random_data.columns
+    )
+    new_names = set(names)
+    new_names -= reduction_kernels
+    new_names -= transformation_kernels
+    new_names -= groupby_other_methods
+
+    assert not reduction_kernels & transformation_kernels
+    assert not reduction_kernels & groupby_other_methods
+    assert not transformation_kernels & groupby_other_methods
+
+    # new public method?
+    if new_names:
+        msg = f"""
+There are uncategorized methods defined on the Grouper class:
+{new_names}.
+
+Was a new method recently added?
+
+Every public method On Grouper must appear in exactly one the
+following three lists defined in pandas.core.groupby.base:
+- `reduction_kernels`
+- `transformation_kernels`
+- `groupby_other_methods`
+see the comments in pandas/core/groupby/base.py for guidance on
+how to fix this test.
+        """
+        raise AssertionError(msg)
+
+    # removed a public method?
+    all_categorized = reduction_kernels | transformation_kernels | groupby_other_methods
+    if names != all_categorized:
+        msg = f"""
+Some methods which are supposed to be on the Grouper class
+are missing:
+{all_categorized - names}.
+
+They're still defined in one of the lists that live in pandas/core/groupby/base.py.
+If you removed a method, you should update them
+"""
+        raise AssertionError(msg)
+
+
+def test_frame_consistency(groupby_func):
+    # GH#48028
+    if groupby_func in ("first", "last"):
+        msg = "first and last don't exist for DataFrame anymore"
+        pytest.skip(reason=msg)
+
+    if groupby_func in ("cumcount", "ngroup"):
+        assert not hasattr(DataFrame, groupby_func)
+        return
+
+    frame_method = getattr(DataFrame, groupby_func)
+    gb_method = getattr(DataFrameGroupBy, groupby_func)
+    result = set(inspect.signature(gb_method).parameters)
+    if groupby_func == "size":
+        # "size" is a method on GroupBy but property on DataFrame:
+        expected = {"self"}
+    else:
+        expected = set(inspect.signature(frame_method).parameters)
+
+    # Exclude certain arguments from result and expected depending on the operation
+    # Some of these may be purposeful inconsistencies between the APIs
+    exclude_expected, exclude_result = set(), set()
+    if groupby_func in ("any", "all"):
+        exclude_expected = {"kwargs", "bool_only", "axis"}
+    elif groupby_func in ("count",):
+        exclude_expected = {"numeric_only", "axis"}
+    elif groupby_func in ("nunique",):
+        exclude_expected = {"axis"}
+    elif groupby_func in ("max", "min"):
+        exclude_expected = {"axis", "kwargs"}
+        exclude_result = {"min_count", "engine", "engine_kwargs"}
+    elif groupby_func in ("sum", "mean", "std", "var"):
+        exclude_expected = {"axis", "kwargs"}
+        exclude_result = {"engine", "engine_kwargs"}
+    elif groupby_func in ("median", "prod", "sem"):
+        exclude_expected = {"axis", "kwargs"}
+    elif groupby_func in ("bfill", "ffill"):
+        exclude_expected = {"inplace", "axis", "limit_area"}
+    elif groupby_func in ("cummax", "cummin"):
+        exclude_expected = {"axis", "skipna", "args"}
+    elif groupby_func in ("cumprod", "cumsum"):
+        exclude_expected = {"axis", "skipna"}
+    elif groupby_func in ("pct_change",):
+        exclude_expected = {"kwargs"}
+    elif groupby_func in ("rank",):
+        exclude_expected = {"numeric_only"}
+    elif groupby_func in ("quantile",):
+        exclude_expected = {"method", "axis"}
+    elif groupby_func in ["corrwith"]:
+        exclude_expected = {"min_periods"}
+    if groupby_func not in ["pct_change", "size"]:
+        exclude_expected |= {"axis"}
+
+    # Ensure excluded arguments are actually in the signatures
+    assert result & exclude_result == exclude_result
+    assert expected & exclude_expected == exclude_expected
+
+    result -= exclude_result
+    expected -= exclude_expected
+    assert result == expected
+
+
+def test_series_consistency(request, groupby_func):
+    # GH#48028
+    if groupby_func in ("first", "last"):
+        msg = "first and last don't exist for Series anymore"
+        pytest.skip(msg)
+
+    if groupby_func in ("cumcount", "corrwith", "ngroup"):
+        assert not hasattr(Series, groupby_func)
+        return
+
+    series_method = getattr(Series, groupby_func)
+    gb_method = getattr(SeriesGroupBy, groupby_func)
+    result = set(inspect.signature(gb_method).parameters)
+    if groupby_func == "size":
+        # "size" is a method on GroupBy but property on Series
+        expected = {"self"}
+    else:
+        expected = set(inspect.signature(series_method).parameters)
+
+    # Exclude certain arguments from result and expected depending on the operation
+    # Some of these may be purposeful inconsistencies between the APIs
+    exclude_expected, exclude_result = set(), set()
+    if groupby_func in ("any", "all"):
+        exclude_expected = {"kwargs", "bool_only", "axis"}
+    elif groupby_func in ("max", "min"):
+        exclude_expected = {"axis", "kwargs"}
+        exclude_result = {"min_count", "engine", "engine_kwargs"}
+    elif groupby_func in ("sum", "mean", "std", "var"):
+        exclude_expected = {"axis", "kwargs"}
+        exclude_result = {"engine", "engine_kwargs"}
+    elif groupby_func in ("median", "prod", "sem"):
+        exclude_expected = {"axis", "kwargs"}
+    elif groupby_func in ("bfill", "ffill"):
+        exclude_expected = {"inplace", "axis", "limit_area"}
+    elif groupby_func in ("cummax", "cummin"):
+        exclude_expected = {"skipna", "args"}
+        exclude_result = {"numeric_only"}
+    elif groupby_func in ("cumprod", "cumsum"):
+        exclude_expected = {"skipna"}
+        exclude_result = {"numeric_only"}
+    elif groupby_func in ("pct_change",):
+        exclude_expected = {"kwargs"}
+    elif groupby_func in ("rank",):
+        exclude_expected = {"numeric_only"}
+    elif groupby_func in ("idxmin", "idxmax"):
+        exclude_expected = {"args", "kwargs"}
+    elif groupby_func in ("quantile",):
+        exclude_result = {"numeric_only"}
+    if groupby_func not in [
+        "diff",
+        "pct_change",
+        "count",
+        "nunique",
+        "quantile",
+        "size",
+    ]:
+        exclude_expected |= {"axis"}
+
+    # Ensure excluded arguments are actually in the signatures
+    assert result & exclude_result == exclude_result
+    assert expected & exclude_expected == exclude_expected
+
+    result -= exclude_result
+    expected -= exclude_expected
+    assert result == expected
diff --git a/pandas/pandas/tests/groupby/test_apply.py b/pandas/pandas/tests/groupby/test_apply.py
new file mode 100644
index 0000000000000..5bf16ee9ad0b8
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_apply.py
@@ -0,0 +1,1516 @@
+from datetime import (
+    date,
+    datetime,
+)
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    bdate_range,
+)
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+def test_apply_func_that_appends_group_to_list_without_copy():
+    # GH: 17718
+
+    df = DataFrame(1, index=list(range(10)) * 10, columns=[0]).reset_index()
+    groups = []
+
+    def store(group):
+        groups.append(group)
+
+    df.groupby("index").apply(store)
+    expected_value = DataFrame({0: [1] * 10}, index=pd.RangeIndex(0, 100, 10))
+    expected_value.columns = expected_value.columns.astype(object)
+
+    tm.assert_frame_equal(groups[0], expected_value)
+
+
+def test_apply_index_date(using_infer_string):
+    # GH 5788
+    ts = [
+        "2011-05-16 00:00",
+        "2011-05-16 01:00",
+        "2011-05-16 02:00",
+        "2011-05-16 03:00",
+        "2011-05-17 02:00",
+        "2011-05-17 03:00",
+        "2011-05-17 04:00",
+        "2011-05-17 05:00",
+        "2011-05-18 02:00",
+        "2011-05-18 03:00",
+        "2011-05-18 04:00",
+        "2011-05-18 05:00",
+    ]
+    df = DataFrame(
+        {
+            "value": [
+                1.40893,
+                1.40760,
+                1.40750,
+                1.40649,
+                1.40893,
+                1.40760,
+                1.40750,
+                1.40649,
+                1.40893,
+                1.40760,
+                1.40750,
+                1.40649,
+            ],
+        },
+        index=Index(pd.to_datetime(ts), name="date_time"),
+    )
+    expected = df.groupby(df.index.date).idxmax()
+    result = df.groupby(df.index.date).apply(lambda x: x.idxmax())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_index_date_object():
+    # GH 5789
+    # don't auto coerce dates
+    ts = [
+        "2011-05-16 00:00",
+        "2011-05-16 01:00",
+        "2011-05-16 02:00",
+        "2011-05-16 03:00",
+        "2011-05-17 02:00",
+        "2011-05-17 03:00",
+        "2011-05-17 04:00",
+        "2011-05-17 05:00",
+        "2011-05-18 02:00",
+        "2011-05-18 03:00",
+        "2011-05-18 04:00",
+        "2011-05-18 05:00",
+    ]
+    df = DataFrame([row.split() for row in ts], columns=["date", "time"])
+    df["value"] = [
+        1.40893,
+        1.40760,
+        1.40750,
+        1.40649,
+        1.40893,
+        1.40760,
+        1.40750,
+        1.40649,
+        1.40893,
+        1.40760,
+        1.40750,
+        1.40649,
+    ]
+    exp_idx = Index(["2011-05-16", "2011-05-17", "2011-05-18"], name="date")
+    expected = Series(["00:00", "02:00", "02:00"], index=exp_idx)
+    result = df.groupby("date").apply(lambda x: x["time"][x["value"].idxmax()])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df, group_names",
+    [
+        (DataFrame({"a": [1, 1, 1, 2, 3], "b": ["a", "a", "a", "b", "c"]}), [1, 2, 3]),
+        (DataFrame({"a": [0, 0, 1, 1], "b": [0, 1, 0, 1]}), [0, 1]),
+        (DataFrame({"a": [1]}), [1]),
+        (DataFrame({"a": [1, 1, 1, 2, 2, 1, 1, 2], "b": range(8)}), [1, 2]),
+        (DataFrame({"a": [1, 2, 3, 1, 2, 3], "two": [4, 5, 6, 7, 8, 9]}), [1, 2, 3]),
+        (
+            DataFrame(
+                {
+                    "a": list("aaabbbcccc"),
+                    "B": [3, 4, 3, 6, 5, 2, 1, 9, 5, 4],
+                    "C": [4, 0, 2, 2, 2, 7, 8, 6, 2, 8],
+                }
+            ),
+            ["a", "b", "c"],
+        ),
+        (DataFrame([[1, 2, 3], [2, 2, 3]], columns=["a", "b", "c"]), [1, 2]),
+    ],
+    ids=[
+        "GH2936",
+        "GH7739 & GH10519",
+        "GH10519",
+        "GH2656",
+        "GH12155",
+        "GH20084",
+        "GH21417",
+    ],
+)
+def test_group_apply_once_per_group(df, group_names):
+    # GH2936, GH7739, GH10519, GH2656, GH12155, GH20084, GH21417
+
+    # This test should ensure that a function is only evaluated
+    # once per group. Previously the function has been evaluated twice
+    # on the first group to check if the Cython index slider is safe to use
+    # This test ensures that the side effect (append to list) is only triggered
+    # once per group
+
+    names = []
+    # cannot parameterize over the functions since they need external
+    # `names` to detect side effects
+
+    def f_copy(group):
+        # this takes the fast apply path
+        names.append(group.name)
+        return group.copy()
+
+    def f_nocopy(group):
+        # this takes the slow apply path
+        names.append(group.name)
+        return group
+
+    def f_scalar(group):
+        # GH7739, GH2656
+        names.append(group.name)
+        return 0
+
+    def f_none(group):
+        # GH10519, GH12155, GH21417
+        names.append(group.name)
+
+    def f_constant_df(group):
+        # GH2936, GH20084
+        names.append(group.name)
+        return DataFrame({"a": [1], "b": [1]})
+
+    for func in [f_copy, f_nocopy, f_scalar, f_none, f_constant_df]:
+        del names[:]
+
+        df.groupby("a").apply(func)
+        assert names == group_names
+
+
+def test_group_apply_once_per_group2(capsys):
+    # GH: 31111
+    # groupby-apply need to execute len(set(group_by_columns)) times
+
+    expected = 2  # Number of times `apply` should call a function for the current test
+
+    df = DataFrame(
+        {
+            "group_by_column": [0, 0, 0, 0, 1, 1, 1, 1],
+            "test_column": ["0", "2", "4", "6", "8", "10", "12", "14"],
+        },
+        index=["0", "2", "4", "6", "8", "10", "12", "14"],
+    )
+
+    df.groupby("group_by_column", group_keys=False).apply(
+        lambda df: print("function_called")
+    )
+
+    result = capsys.readouterr().out.count("function_called")
+    # If `groupby` behaves unexpectedly, this test will break
+    assert result == expected
+
+
+def test_apply_fast_slow_identical():
+    # GH 31613
+
+    df = DataFrame({"A": [0, 0, 1], "b": range(3)})
+
+    # For simple index structures we check for fast/slow apply using
+    # an identity check on in/output
+    def slow(group):
+        return group
+
+    def fast(group):
+        return group.copy()
+
+    fast_df = df.groupby("A", group_keys=False).apply(fast)
+    slow_df = df.groupby("A", group_keys=False).apply(slow)
+    tm.assert_frame_equal(fast_df, slow_df)
+
+
+def test_apply_fast_slow_identical_index():
+    # GH#44803
+    df = DataFrame(
+        {
+            "name": ["Alice", "Bob", "Carl"],
+            "age": [20, 21, 20],
+        }
+    ).set_index("name")
+
+    grp_by_same_value = df.groupby(["age"], group_keys=False).apply(lambda group: group)
+    grp_by_copy = df.groupby(["age"], group_keys=False).apply(
+        lambda group: group.copy()
+    )
+    tm.assert_frame_equal(grp_by_same_value, grp_by_copy)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x: x,
+        lambda x: x[:],
+        lambda x: x.copy(deep=False),
+        lambda x: x.copy(deep=True),
+    ],
+)
+def test_groupby_apply_identity_maybecopy_index_identical(func):
+    # GH 14927
+    # Whether the function returns a copy of the input data or not should not
+    # have an impact on the index structure of the result since this is not
+    # transparent to the user
+
+    df = DataFrame({"g": [1, 2, 2, 2], "a": [1, 2, 3, 4], "b": [5, 6, 7, 8]})
+    result = df.groupby("g", group_keys=False).apply(func)
+    tm.assert_frame_equal(result, df[["a", "b"]])
+
+
+def test_apply_with_mixed_dtype():
+    # GH3480, apply with mixed dtype on axis=1 breaks in 0.11
+    df = DataFrame(
+        {
+            "foo1": np.random.default_rng(2).standard_normal(6),
+            "foo2": ["one", "two", "two", "three", "one", "two"],
+        }
+    )
+    result = df.apply(lambda x: x, axis=1)
+    expected = df
+    tm.assert_frame_equal(result, expected)
+
+    # GH 3610 incorrect dtype conversion with as_index=False
+    df = DataFrame({"c1": [1, 2, 6, 6, 8]})
+    df["c2"] = df.c1 / 2.0
+    result1 = df.groupby("c2").mean().reset_index()
+    result2 = df.groupby("c2", as_index=False).mean()
+    tm.assert_frame_equal(result1, result2)
+
+
+def test_groupby_as_index_apply(as_index):
+    # GH #4648 and #3417
+    df = DataFrame(
+        {
+            "item_id": ["b", "b", "a", "c", "a", "b"],
+            "user_id": [1, 2, 1, 1, 3, 1],
+            "time": range(6),
+        }
+    )
+    gb = df.groupby("user_id", as_index=as_index)
+
+    expected = DataFrame(
+        {
+            "item_id": ["b", "b", "a", "a"],
+            "user_id": [1, 2, 1, 3],
+            "time": [0, 1, 2, 4],
+        },
+        index=[0, 1, 2, 4],
+    )
+    result = gb.head(2)
+    tm.assert_frame_equal(result, expected)
+
+    # apply doesn't maintain the original ordering
+    # changed in GH5610 as the as_index=False returns a MI here
+    if as_index:
+        tp = [(1, 0), (1, 2), (2, 1), (3, 4)]
+        index = MultiIndex.from_tuples(tp, names=["user_id", None])
+    else:
+        index = Index([0, 2, 1, 4])
+    expected = DataFrame(
+        {
+            "item_id": list("baba"),
+            "time": [0, 2, 1, 4],
+        },
+        index=index,
+    )
+    result = gb.apply(lambda x: x.head(2))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_as_index_apply_str():
+    ind = Index(list("abcde"))
+    df = DataFrame([[1, 2], [2, 3], [1, 4], [1, 5], [2, 6]], index=ind)
+    res = df.groupby(0, as_index=False, group_keys=False).apply(lambda x: x).index
+    tm.assert_index_equal(res, ind)
+
+
+def test_apply_concat_preserve_names(three_group):
+    grouped = three_group.groupby(["A", "B"])
+
+    def desc(group):
+        result = group.describe()
+        result.index.name = "stat"
+        return result
+
+    def desc2(group):
+        result = group.describe()
+        result.index.name = "stat"
+        result = result[: len(group)]
+        # weirdo
+        return result
+
+    def desc3(group):
+        result = group.describe()
+
+        # names are different
+        result.index.name = f"stat_{len(group):d}"
+
+        result = result[: len(group)]
+        # weirdo
+        return result
+
+    result = grouped.apply(desc)
+    assert result.index.names == ("A", "B", "stat")
+
+    result2 = grouped.apply(desc2)
+    assert result2.index.names == ("A", "B", "stat")
+
+    result3 = grouped.apply(desc3)
+    assert result3.index.names == ("A", "B", None)
+
+
+def test_apply_series_to_frame():
+    def f(piece):
+        with np.errstate(invalid="ignore"):
+            logged = np.log(piece)
+        return DataFrame(
+            {"value": piece, "demeaned": piece - piece.mean(), "logged": logged}
+        )
+
+    dr = bdate_range("1/1/2000", periods=10)
+    ts = Series(np.random.default_rng(2).standard_normal(10), index=dr)
+
+    grouped = ts.groupby(lambda x: x.month, group_keys=False)
+    result = grouped.apply(f)
+
+    assert isinstance(result, DataFrame)
+    assert not hasattr(result, "name")  # GH49907
+    tm.assert_index_equal(result.index, ts.index)
+
+
+def test_apply_series_yield_constant(df):
+    result = df.groupby(["A", "B"])["C"].apply(len)
+    assert result.index.names[:2] == ("A", "B")
+
+
+def test_apply_frame_yield_constant(df):
+    # GH13568
+    result = df.groupby(["A", "B"]).apply(len)
+    assert isinstance(result, Series)
+    assert result.name is None
+
+    result = df.groupby(["A", "B"])[["C", "D"]].apply(len)
+    assert isinstance(result, Series)
+    assert result.name is None
+
+
+def test_apply_frame_to_series(df):
+    grouped = df.groupby(["A", "B"])
+    result = grouped.apply(len)
+    expected = grouped.count()["C"]
+    tm.assert_index_equal(result.index, expected.index)
+    tm.assert_numpy_array_equal(result.values, expected.values)
+
+
+def test_apply_frame_not_as_index_column_name(df):
+    # GH 35964 - path within _wrap_applied_output not hit by a test
+    grouped = df.groupby(["A", "B"], as_index=False)
+    result = grouped.apply(len)
+    expected = grouped.count().rename(columns={"C": np.nan}).drop(columns="D")
+    # TODO(GH#34306): Use assert_frame_equal when column name is not np.nan
+    tm.assert_index_equal(result.index, expected.index)
+    tm.assert_numpy_array_equal(result.values, expected.values)
+
+
+def test_apply_frame_concat_series():
+    def trans(group):
+        return group.groupby("B")["C"].sum().sort_values().iloc[:2]
+
+    def trans2(group):
+        grouped = group.groupby(df.reindex(group.index)["B"])
+        return grouped.sum().sort_values().iloc[:2]
+
+    df = DataFrame(
+        {
+            "A": np.random.default_rng(2).integers(0, 5, 1000),
+            "B": np.random.default_rng(2).integers(0, 5, 1000),
+            "C": np.random.default_rng(2).standard_normal(1000),
+        }
+    )
+
+    result = df.groupby("A").apply(trans)
+    exp = df.groupby("A")["C"].apply(trans2)
+    tm.assert_series_equal(result, exp, check_names=False)
+    assert result.name == "C"
+
+
+def test_apply_transform(ts):
+    grouped = ts.groupby(lambda x: x.month, group_keys=False)
+    result = grouped.apply(lambda x: x * 2)
+    expected = grouped.transform(lambda x: x * 2)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_multikey_corner(tsframe):
+    grouped = tsframe.groupby([lambda x: x.year, lambda x: x.month])
+
+    def f(group):
+        return group.sort_values("A")[-5:]
+
+    result = grouped.apply(f)
+    for key, group in grouped:
+        tm.assert_frame_equal(result.loc[key], f(group))
+
+
+@pytest.mark.parametrize("group_keys", [True, False])
+def test_apply_chunk_view(group_keys):
+    # Low level tinkering could be unsafe, make sure not
+    df = DataFrame({"key": [1, 1, 1, 2, 2, 2, 3, 3, 3], "value": range(9)})
+
+    result = df.groupby("key", group_keys=group_keys).apply(lambda x: x.iloc[:2])
+    expected = df[["value"]].take([0, 1, 3, 4, 6, 7])
+    if group_keys:
+        expected.index = MultiIndex.from_arrays(
+            [[1, 1, 2, 2, 3, 3], expected.index], names=["key", None]
+        )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_no_name_column_conflict():
+    df = DataFrame(
+        {
+            "name": [1, 1, 1, 1, 1, 1, 2, 2, 2, 2],
+            "name2": [0, 0, 0, 1, 1, 1, 0, 0, 1, 1],
+            "value": range(9, -1, -1),
+        }
+    )
+
+    # it works! #2605
+    grouped = df.groupby(["name", "name2"])
+    grouped.apply(lambda x: x.sort_values("value", inplace=True))
+
+
+def test_apply_typecast_fail():
+    df = DataFrame(
+        {
+            "d": [1.0, 1.0, 1.0, 2.0, 2.0, 2.0],
+            "c": np.tile(["a", "b", "c"], 2),
+            "v": np.arange(1.0, 7.0),
+        }
+    )
+
+    def f(group):
+        v = group["v"]
+        group["v2"] = (v - v.min()) / (v.max() - v.min())
+        return group
+
+    result = df.groupby("d", group_keys=False).apply(f)
+
+    expected = df[["c", "v"]]
+    expected["v2"] = np.tile([0.0, 0.5, 1], 2)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_multiindex_fail():
+    index = MultiIndex.from_arrays([[0, 0, 0, 1, 1, 1], [1, 2, 3, 1, 2, 3]])
+    df = DataFrame(
+        {
+            "d": [1.0, 1.0, 1.0, 2.0, 2.0, 2.0],
+            "c": np.tile(["a", "b", "c"], 2),
+            "v": np.arange(1.0, 7.0),
+        },
+        index=index,
+    )
+
+    def f(group):
+        v = group["v"]
+        group["v2"] = (v - v.min()) / (v.max() - v.min())
+        return group
+
+    result = df.groupby("d", group_keys=False).apply(f)
+
+    expected = df[["c", "v"]]
+    expected["v2"] = np.tile([0.0, 0.5, 1], 2)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_corner(tsframe):
+    result = tsframe.groupby(lambda x: x.year, group_keys=False).apply(lambda x: x * 2)
+    expected = tsframe * 2
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_without_copy():
+    # GH 5545
+    # returning a non-copy in an applied function fails
+
+    data = DataFrame(
+        {
+            "id_field": [100, 100, 200, 300],
+            "category": ["a", "b", "c", "c"],
+            "value": [1, 2, 3, 4],
+        }
+    )
+
+    def filt1(x):
+        if x.shape[0] == 1:
+            return x.copy()
+        else:
+            return x[x.category == "c"]
+
+    def filt2(x):
+        if x.shape[0] == 1:
+            return x
+        else:
+            return x[x.category == "c"]
+
+    expected = data.groupby("id_field").apply(filt1)
+    result = data.groupby("id_field").apply(filt2)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("test_series", [True, False])
+def test_apply_with_duplicated_non_sorted_axis(test_series):
+    # GH 30667
+    df = DataFrame(
+        [["x", "p"], ["x", "p"], ["x", "o"]], columns=["X", "Y"], index=[1, 2, 2]
+    )
+    if test_series:
+        ser = df.set_index("Y")["X"]
+        result = ser.groupby(level=0, group_keys=False).apply(lambda x: x)
+        expected = ser
+        tm.assert_series_equal(result, expected)
+    else:
+        result = df.groupby("Y", group_keys=False).apply(lambda x: x)
+        expected = df[["X"]]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_apply_reindex_values():
+    # GH: 26209
+    # reindexing from a single column of a groupby object with duplicate indices caused
+    # a ValueError (cannot reindex from duplicate axis) in 0.24.2, the problem was
+    # solved in #30679
+    values = [1, 2, 3, 4]
+    indices = [1, 1, 2, 2]
+    df = DataFrame({"group": ["Group1", "Group2"] * 2, "value": values}, index=indices)
+    expected = Series(values, index=indices, name="value")
+
+    def reindex_helper(x):
+        return x.reindex(np.arange(x.index.min(), x.index.max() + 1))
+
+    # the following group by raised a ValueError
+    result = df.groupby("group", group_keys=False).value.apply(reindex_helper)
+    tm.assert_series_equal(expected, result)
+
+
+def test_apply_corner_cases():
+    # #535, can't use sliding iterator
+
+    N = 10
+    labels = np.random.default_rng(2).integers(0, 100, size=N)
+    df = DataFrame(
+        {
+            "key": labels,
+            "value1": np.random.default_rng(2).standard_normal(N),
+            "value2": ["foo", "bar", "baz", "qux", "a"] * (N // 5),
+        }
+    )
+
+    grouped = df.groupby("key", group_keys=False)
+
+    def f(g):
+        g["value3"] = g["value1"] * 2
+        return g
+
+    result = grouped.apply(f)
+    assert "value3" in result
+
+
+def test_apply_numeric_coercion_when_datetime():
+    # In the past, group-by/apply operations have been over-eager
+    # in converting dtypes to numeric, in the presence of datetime
+    # columns.  Various GH issues were filed, the reproductions
+    # for which are here.
+
+    # GH 15670
+    df = DataFrame(
+        {"Number": [1, 2], "Date": ["2017-03-02"] * 2, "Str": ["foo", "inf"]}
+    )
+    expected = df.groupby(["Number"]).apply(lambda x: x.iloc[0])
+    df.Date = pd.to_datetime(df.Date)
+    result = df.groupby(["Number"]).apply(lambda x: x.iloc[0])
+    tm.assert_series_equal(result["Str"], expected["Str"])
+
+
+def test_apply_numeric_coercion_when_datetime_getitem():
+    # GH 15421
+    df = DataFrame(
+        {"A": [10, 20, 30], "B": ["foo", "3", "4"], "T": [pd.Timestamp("12:31:22")] * 3}
+    )
+
+    def get_B(g):
+        return g.iloc[0][["B"]]
+
+    result = df.groupby("A").apply(get_B)["B"]
+    expected = df.B
+    expected.index = df.A
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_numeric_coercion_when_datetime_with_nat():
+    # GH 14423
+    def predictions(tool):
+        out = Series(index=["p1", "p2", "useTime"], dtype=object)
+        if "step1" in list(tool.State):
+            out["p1"] = str(tool[tool.State == "step1"].Machine.values[0])
+        if "step2" in list(tool.State):
+            out["p2"] = str(tool[tool.State == "step2"].Machine.values[0])
+            out["useTime"] = str(tool[tool.State == "step2"].oTime.values[0])
+        return out
+
+    df1 = DataFrame(
+        {
+            "Key": ["B", "B", "A", "A"],
+            "State": ["step1", "step2", "step1", "step2"],
+            "oTime": ["", "2016-09-19 05:24:33", "", "2016-09-19 23:59:04"],
+            "Machine": ["23", "36L", "36R", "36R"],
+        }
+    )
+    df2 = df1.copy()
+    df2.oTime = pd.to_datetime(df2.oTime)
+    expected = df1.groupby("Key").apply(predictions).p1
+    result = df2.groupby("Key").apply(predictions).p1
+    tm.assert_series_equal(expected, result)
+
+
+def test_apply_aggregating_timedelta_and_datetime():
+    # Regression test for GH 15562
+    # The following groupby caused ValueErrors and IndexErrors pre 0.20.0
+
+    df = DataFrame(
+        {
+            "clientid": ["A", "B", "C"],
+            "datetime": [np.datetime64("2017-02-01 00:00:00")] * 3,
+        }
+    )
+    df["time_delta_zero"] = df.datetime - df.datetime
+    result = df.groupby("clientid").apply(
+        lambda ddf: Series(
+            {"clientid_age": ddf.time_delta_zero.min(), "date": ddf.datetime.min()}
+        )
+    )
+    expected = DataFrame(
+        {
+            "clientid": ["A", "B", "C"],
+            "clientid_age": [np.timedelta64(0, "D")] * 3,
+            "date": [np.datetime64("2017-02-01 00:00:00")] * 3,
+        }
+    ).set_index("clientid")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_groupby_datetimeindex():
+    # GH 26182
+    # groupby apply failed on dataframe with DatetimeIndex
+
+    data = [["A", 10], ["B", 20], ["B", 30], ["C", 40], ["C", 50]]
+    df = DataFrame(
+        data, columns=["Name", "Value"], index=pd.date_range("2020-09-01", "2020-09-05")
+    )
+
+    result = df.groupby("Name").sum()
+
+    expected = DataFrame({"Name": ["A", "B", "C"], "Value": [10, 50, 90]})
+    expected.set_index("Name", inplace=True)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_time_field_bug():
+    # Test a fix for the following error related to GH issue 11324 When
+    # non-key fields in a group-by dataframe contained time-based fields
+    # that were not returned by the apply function, an exception would be
+    # raised.
+
+    df = DataFrame({"a": 1, "b": [datetime.now() for nn in range(10)]})
+
+    def func_with_no_date(batch):
+        return Series({"c": 2})
+
+    def func_with_date(batch):
+        return Series({"b": datetime(2015, 1, 1), "c": 2})
+
+    dfg_no_conversion = df.groupby(by=["a"]).apply(func_with_no_date)
+    dfg_no_conversion_expected = DataFrame({"c": 2}, index=[1])
+    dfg_no_conversion_expected.index.name = "a"
+
+    dfg_conversion = df.groupby(by=["a"]).apply(func_with_date)
+    dfg_conversion_expected = DataFrame(
+        {"b": pd.Timestamp(2015, 1, 1), "c": 2}, index=[1]
+    )
+    dfg_conversion_expected.index.name = "a"
+
+    tm.assert_frame_equal(dfg_no_conversion, dfg_no_conversion_expected)
+    tm.assert_frame_equal(dfg_conversion, dfg_conversion_expected)
+
+
+def test_gb_apply_list_of_unequal_len_arrays():
+    # GH1738
+    df = DataFrame(
+        {
+            "group1": ["a", "a", "a", "b", "b", "b", "a", "a", "a", "b", "b", "b"],
+            "group2": ["c", "c", "d", "d", "d", "e", "c", "c", "d", "d", "d", "e"],
+            "weight": [1.1, 2, 3, 4, 5, 6, 2, 4, 6, 8, 1, 2],
+            "value": [7.1, 8, 9, 10, 11, 12, 8, 7, 6, 5, 4, 3],
+        }
+    )
+    df = df.set_index(["group1", "group2"])
+    df_grouped = df.groupby(level=["group1", "group2"], sort=True)
+
+    def noddy(value, weight):
+        out = np.array(value * weight).repeat(3)
+        return out
+
+    # the kernel function returns arrays of unequal length
+    # pandas sniffs the first one, sees it's an array and not
+    # a list, and assumed the rest are of equal length
+    # and so tries a vstack
+
+    # don't die
+    df_grouped.apply(lambda x: noddy(x.value, x.weight))
+
+
+def test_groupby_apply_all_none():
+    # Tests to make sure no errors if apply function returns all None
+    # values. Issue 9684.
+    test_df = DataFrame({"groups": [0, 0, 1, 1], "random_vars": [8, 7, 4, 5]})
+
+    def test_func(x):
+        pass
+
+    result = test_df.groupby("groups").apply(test_func)
+    expected = DataFrame(columns=["random_vars"], dtype="int64")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "in_data, out_idx, out_data",
+    [
+        [
+            {"groups": [1, 1, 1, 2], "vars": [0, 1, 2, 3]},
+            [[1, 1], [0, 2]],
+            {"vars": [0, 2]},
+        ],
+        [
+            {"groups": [1, 2, 2, 2], "vars": [0, 1, 2, 3]},
+            [[2, 2], [1, 3]],
+            {"vars": [1, 3]},
+        ],
+    ],
+)
+def test_groupby_apply_none_first(in_data, out_idx, out_data):
+    # GH 12824. Tests if apply returns None first.
+    test_df1 = DataFrame(in_data)
+
+    def test_func(x):
+        if x.shape[0] < 2:
+            return None
+        return x.iloc[[0, -1]]
+
+    result1 = test_df1.groupby("groups").apply(test_func)
+    index1 = MultiIndex.from_arrays(out_idx, names=["groups", None])
+    expected1 = DataFrame(out_data, index=index1)
+    tm.assert_frame_equal(result1, expected1)
+
+
+def test_groupby_apply_return_empty_chunk():
+    # GH 22221: apply filter which returns some empty groups
+    df = DataFrame({"value": [0, 1], "group": ["filled", "empty"]})
+    groups = df.groupby("group")
+    result = groups.apply(lambda group: group[group.value != 1]["value"])
+    expected = Series(
+        [0],
+        name="value",
+        index=MultiIndex.from_product(
+            [["empty", "filled"], [0]], names=["group", None]
+        ).drop("empty"),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("meth", ["apply", "transform"])
+def test_apply_with_mixed_types(meth):
+    # gh-20949
+    df = DataFrame({"A": "a a b".split(), "B": [1, 2, 3], "C": [4, 6, 5]})
+    g = df.groupby("A", group_keys=False)
+
+    result = getattr(g, meth)(lambda x: x / x.sum())
+    expected = DataFrame({"B": [1 / 3.0, 2 / 3.0, 1], "C": [0.4, 0.6, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_func_returns_object():
+    # GH 28652
+    df = DataFrame({"a": [1, 2]}, index=Index([1, 2]))
+    result = df.groupby("a").apply(lambda g: g.index)
+    expected = Series([Index([1]), Index([2])], index=Index([1, 2], name="a"))
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "group_column_dtlike",
+    [datetime.today(), datetime.today().date(), datetime.today().time()],
+)
+def test_apply_datetime_issue(group_column_dtlike):
+    # GH-28247
+    # groupby-apply throws an error if one of the columns in the DataFrame
+    #   is a datetime object and the column labels are different from
+    #   standard int values in range(len(num_columns))
+
+    df = DataFrame({"a": ["foo"], "b": [group_column_dtlike]})
+    result = df.groupby("a").apply(lambda x: Series(["spam"], index=[42]))
+
+    expected = DataFrame(["spam"], Index(["foo"], dtype="str", name="a"), columns=[42])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_series_return_dataframe_groups():
+    # GH 10078
+    tdf = DataFrame(
+        {
+            "day": {
+                0: pd.Timestamp("2015-02-24 00:00:00"),
+                1: pd.Timestamp("2015-02-24 00:00:00"),
+                2: pd.Timestamp("2015-02-24 00:00:00"),
+                3: pd.Timestamp("2015-02-24 00:00:00"),
+                4: pd.Timestamp("2015-02-24 00:00:00"),
+            },
+            "userAgent": {
+                0: "some UA string",
+                1: "some UA string",
+                2: "some UA string",
+                3: "another UA string",
+                4: "some UA string",
+            },
+            "userId": {
+                0: "17661101",
+                1: "17661101",
+                2: "17661101",
+                3: "17661101",
+                4: "17661101",
+            },
+        }
+    )
+
+    def most_common_values(df):
+        return Series({c: s.value_counts().index[0] for c, s in df.items()})
+
+    result = tdf.groupby("day").apply(most_common_values)["userId"]
+    expected = Series(
+        ["17661101"], index=pd.DatetimeIndex(["2015-02-24"], name="day"), name="userId"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("category", [False, True])
+def test_apply_multi_level_name(category):
+    # https://github.com/pandas-dev/pandas/issues/31068
+    b = [1, 2] * 5
+    if category:
+        b = pd.Categorical(b, categories=[1, 2, 3])
+        expected_index = pd.CategoricalIndex([1, 2, 3], categories=[1, 2, 3], name="B")
+        expected_values = [20, 25, 0]
+    else:
+        expected_index = Index([1, 2], name="B")
+        expected_values = [20, 25]
+    expected = DataFrame(
+        {"C": expected_values, "D": expected_values}, index=expected_index
+    )
+
+    df = DataFrame(
+        {"A": np.arange(10), "B": b, "C": list(range(10)), "D": list(range(10))}
+    ).set_index(["A", "B"])
+    result = df.groupby("B", observed=False).apply(lambda x: x.sum())
+    tm.assert_frame_equal(result, expected)
+    assert df.index.names == ["A", "B"]
+
+
+def test_groupby_apply_datetime_result_dtypes(using_infer_string):
+    # GH 14849
+    data = DataFrame.from_records(
+        [
+            (pd.Timestamp(2016, 1, 1), "red", "dark", 1, "8"),
+            (pd.Timestamp(2015, 1, 1), "green", "stormy", 2, "9"),
+            (pd.Timestamp(2014, 1, 1), "blue", "bright", 3, "10"),
+            (pd.Timestamp(2013, 1, 1), "blue", "calm", 4, "potato"),
+        ],
+        columns=["observation", "color", "mood", "intensity", "score"],
+    )
+    result = data.groupby("color").apply(lambda g: g.iloc[0]).dtypes
+    dtype = pd.StringDtype(na_value=np.nan) if using_infer_string else object
+    expected = Series(
+        [np.dtype("datetime64[us]"), dtype, np.int64, dtype],
+        index=["observation", "mood", "intensity", "score"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        pd.CategoricalIndex(list("abc")),
+        pd.interval_range(0, 3),
+        pd.period_range("2020", periods=3, freq="D"),
+        MultiIndex.from_tuples([("a", 0), ("a", 1), ("b", 0)]),
+    ],
+)
+def test_apply_index_has_complex_internals(index):
+    # GH 31248
+    df = DataFrame({"group": [1, 1, 2], "value": [0, 1, 0]}, index=index)
+    result = df.groupby("group", group_keys=False).apply(lambda x: x)
+    tm.assert_frame_equal(result, df[["value"]])
+
+
+@pytest.mark.parametrize(
+    "function, expected_values",
+    [
+        (lambda x: x.index.to_list(), [[0, 1], [2, 3]]),
+        (lambda x: set(x.index.to_list()), [{0, 1}, {2, 3}]),
+        (lambda x: tuple(x.index.to_list()), [(0, 1), (2, 3)]),
+        (
+            lambda x: dict(enumerate(x.index.to_list())),
+            [{0: 0, 1: 1}, {0: 2, 1: 3}],
+        ),
+        (
+            lambda x: [{n: i} for (n, i) in enumerate(x.index.to_list())],
+            [[{0: 0}, {1: 1}], [{0: 2}, {1: 3}]],
+        ),
+    ],
+)
+def test_apply_function_returns_non_pandas_non_scalar(function, expected_values):
+    # GH 31441
+    df = DataFrame(["A", "A", "B", "B"], columns=["groups"])
+    result = df.groupby("groups").apply(function)
+    expected = Series(expected_values, index=Index(["A", "B"], name="groups"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_function_returns_numpy_array():
+    # GH 31605
+    def fct(group):
+        return group["B"].values.flatten()
+
+    df = DataFrame({"A": ["a", "a", "b", "none"], "B": [1, 2, 3, np.nan]})
+
+    result = df.groupby("A").apply(fct)
+    expected = Series(
+        [[1.0, 2.0], [3.0], [np.nan]], index=Index(["a", "b", "none"], name="A")
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("function", [lambda gr: gr.index, lambda gr: gr.index + 1 - 1])
+def test_apply_function_index_return(function):
+    # GH: 22541
+    df = DataFrame([1, 2, 2, 2, 1, 2, 3, 1, 3, 1], columns=["id"])
+    result = df.groupby("id").apply(function)
+    expected = Series(
+        [Index([0, 4, 7, 9]), Index([1, 2, 3, 5]), Index([6, 8])],
+        index=Index([1, 2, 3], name="id"),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_function_with_indexing_return_column():
+    # GH#7002, GH#41480, GH#49256
+    df = DataFrame(
+        {
+            "foo1": ["one", "two", "two", "three", "one", "two"],
+            "foo2": [1, 2, 4, 4, 5, 6],
+        }
+    )
+    result = df.groupby("foo1", as_index=False).apply(lambda x: x.mean())
+    expected = DataFrame(
+        {
+            "foo1": ["one", "three", "two"],
+            "foo2": [3.0, 4.0, 4.0],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "udf",
+    [lambda x: x.copy(), lambda x: x.copy().rename(lambda y: y + 1)],
+)
+@pytest.mark.parametrize("group_keys", [True, False])
+def test_apply_result_type(group_keys, udf):
+    # https://github.com/pandas-dev/pandas/issues/34809
+    # We'd like to control whether the group keys end up in the index
+    # regardless of whether the UDF happens to be a transform.
+    df = DataFrame({"A": ["a", "b"], "B": [1, 2]})
+    df_result = df.groupby("A", group_keys=group_keys).apply(udf)
+    series_result = df.B.groupby(df.A, group_keys=group_keys).apply(udf)
+
+    if group_keys:
+        assert df_result.index.nlevels == 2
+        assert series_result.index.nlevels == 2
+    else:
+        assert df_result.index.nlevels == 1
+        assert series_result.index.nlevels == 1
+
+
+def test_result_order_group_keys_false():
+    # GH 34998
+    # apply result order should not depend on whether index is the same or just equal
+    df = DataFrame({"A": [2, 1, 2], "B": [1, 2, 3]})
+    result = df.groupby("A", group_keys=False).apply(lambda x: x)
+    expected = df.groupby("A", group_keys=False).apply(lambda x: x.copy())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_with_timezones_aware():
+    # GH: 27212
+    dates = ["2001-01-01"] * 2 + ["2001-01-02"] * 2 + ["2001-01-03"] * 2
+    index_no_tz = pd.DatetimeIndex(dates)
+    index_tz = pd.DatetimeIndex(dates, tz="UTC")
+    df1 = DataFrame({"x": list(range(2)) * 3, "y": range(6), "t": index_no_tz})
+    df2 = DataFrame({"x": list(range(2)) * 3, "y": range(6), "t": index_tz})
+
+    result1 = df1.groupby("x", group_keys=False).apply(lambda df: df[["y"]].copy())
+    result2 = df2.groupby("x", group_keys=False).apply(lambda df: df[["y"]].copy())
+
+    tm.assert_frame_equal(result1, result2)
+
+
+def test_apply_is_unchanged_when_other_methods_are_called_first(reduction_func):
+    # GH #34656
+    # GH #34271
+    df = DataFrame(
+        {
+            "a": [99, 99, 99, 88, 88, 88],
+            "b": [1, 2, 3, 4, 5, 6],
+            "c": [10, 20, 30, 40, 50, 60],
+        }
+    )
+
+    expected = DataFrame(
+        {"b": [15, 6], "c": [150, 60]},
+        index=Index([88, 99], name="a"),
+    )
+
+    # Check output when no other methods are called before .apply()
+    grp = df.groupby(by="a")
+    result = grp.apply(np.sum, axis=0)
+    tm.assert_frame_equal(result, expected)
+
+    # Check output when another method is called before .apply()
+    grp = df.groupby(by="a")
+    args = get_groupby_method_args(reduction_func, df)
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        msg = ""
+    with tm.assert_produces_warning(warn, match=msg):
+        _ = getattr(grp, reduction_func)(*args)
+    result = grp.apply(np.sum, axis=0)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_with_date_in_multiindex_does_not_convert_to_timestamp():
+    # GH 29617
+
+    df = DataFrame(
+        {
+            "A": ["a", "a", "a", "b"],
+            "B": [
+                date(2020, 1, 10),
+                date(2020, 1, 10),
+                date(2020, 2, 10),
+                date(2020, 2, 10),
+            ],
+            "C": [1, 2, 3, 4],
+        },
+        index=Index([100, 101, 102, 103], name="idx"),
+    )
+
+    grp = df.groupby(["A", "B"])
+    result = grp.apply(lambda x: x.head(1))
+
+    expected = df.iloc[[0, 2, 3]]
+    expected = expected.reset_index()
+    expected.index = MultiIndex.from_frame(expected[["A", "B", "idx"]])
+    expected = expected.drop(columns=["A", "B", "idx"])
+
+    tm.assert_frame_equal(result, expected)
+    for val in result.index.levels[1]:
+        assert type(val) is date
+
+
+def test_apply_dropna_with_indexed_same(dropna):
+    # GH 38227
+    # GH#43205
+    df = DataFrame(
+        {
+            "col": [1, 2, 3, 4, 5],
+            "group": ["a", np.nan, np.nan, "b", "b"],
+        },
+        index=list("xxyxz"),
+    )
+    result = df.groupby("group", dropna=dropna, group_keys=False).apply(lambda x: x)
+    expected = df.dropna()[["col"]] if dropna else df[["col"]].iloc[[0, 3, 1, 2, 4]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "as_index, expected",
+    [
+        [
+            False,
+            DataFrame(
+                [[1, 1, 1], [2, 2, 1]], columns=Index(["a", "b", None], dtype=object)
+            ),
+        ],
+        [
+            True,
+            Series(
+                [1, 1], index=MultiIndex.from_tuples([(1, 1), (2, 2)], names=["a", "b"])
+            ),
+        ],
+    ],
+)
+def test_apply_as_index_constant_lambda(as_index, expected):
+    # GH 13217
+    df = DataFrame({"a": [1, 1, 2, 2], "b": [1, 1, 2, 2], "c": [1, 1, 1, 1]})
+    result = df.groupby(["a", "b"], as_index=as_index).apply(lambda x: 1)
+    tm.assert_equal(result, expected)
+
+
+def test_sort_index_groups():
+    # GH 20420
+    df = DataFrame(
+        {"A": [1, 2, 3, 4, 5], "B": [6, 7, 8, 9, 0], "C": [1, 1, 1, 2, 2]},
+        index=range(5),
+    )
+    result = df.groupby("C").apply(lambda x: x.A.sort_index())
+    expected = Series(
+        range(1, 6),
+        index=MultiIndex.from_tuples(
+            [(1, 0), (1, 1), (1, 2), (2, 3), (2, 4)], names=["C", None]
+        ),
+        name="A",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_positional_slice_groups_datetimelike():
+    # GH 21651
+    expected = DataFrame(
+        {
+            "date": pd.date_range("2010-01-01", freq="12h", periods=5),
+            "vals": range(5),
+            "let": list("abcde"),
+        }
+    )
+    result = expected.groupby(
+        [expected.let, expected.date.dt.date], group_keys=False
+    ).apply(lambda x: x.iloc[0:])
+    tm.assert_frame_equal(result, expected[["date", "vals"]])
+
+
+def test_groupby_apply_shape_cache_safety():
+    # GH#42702 this fails if we cache_readonly Block.shape
+    df = DataFrame({"A": ["a", "a", "b"], "B": [1, 2, 3], "C": [4, 6, 5]})
+    gb = df.groupby("A")
+    result = gb[["B", "C"]].apply(lambda x: x.astype(float).max() - x.min())
+
+    expected = DataFrame(
+        {"B": [1.0, 0.0], "C": [2.0, 0.0]}, index=Index(["a", "b"], name="A")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_apply_to_series_name():
+    # GH52444
+    df = DataFrame.from_dict(
+        {
+            "a": ["a", "b", "a", "b"],
+            "b1": ["aa", "ac", "ac", "ad"],
+            "b2": ["aa", "aa", "aa", "ac"],
+        }
+    )
+    grp = df.groupby("a")[["b1", "b2"]]
+    result = grp.apply(lambda x: x.unstack().value_counts())
+
+    expected_idx = MultiIndex.from_arrays(
+        arrays=[["a", "a", "b", "b", "b"], ["aa", "ac", "ac", "ad", "aa"]],
+        names=["a", None],
+    )
+    expected = Series([3, 1, 2, 1, 1], index=expected_idx, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_na(dropna):
+    # GH#28984
+    df = DataFrame(
+        {"grp": [1, 1, 2, 2], "y": [1, 0, 2, 5], "z": [1, 2, np.nan, np.nan]}
+    )
+    dfgrp = df.groupby("grp", dropna=dropna)
+    result = dfgrp.apply(lambda grp_df: grp_df.nlargest(1, "z"))
+    expected = dfgrp.apply(lambda x: x.sort_values("z", ascending=False).head(1))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_empty_string_nan_coerce_bug():
+    # GH#24903
+    result = (
+        DataFrame(
+            {
+                "a": [1, 1, 2, 2],
+                "b": ["", "", "", ""],
+                "c": pd.to_datetime([1, 2, 3, 4], unit="s"),
+            }
+        )
+        .groupby(["a", "b"])
+        .apply(lambda df: df.iloc[-1])
+    )
+    expected = DataFrame(
+        [[pd.to_datetime(2, unit="s")], [pd.to_datetime(4, unit="s")]],
+        columns=["c"],
+        index=MultiIndex.from_tuples([(1, ""), (2, "")], names=["a", "b"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_values", [[1, 2, 3], [1.0, 2.0, 3.0]])
+def test_apply_index_key_error_bug(index_values):
+    # GH 44310
+    result = DataFrame(
+        {
+            "a": ["aa", "a2", "a3"],
+            "b": [1, 2, 3],
+        },
+        index=Index(index_values),
+    )
+    expected = DataFrame(
+        {
+            "b_mean": [2.0, 3.0, 1.0],
+        },
+        index=Index(["a2", "a3", "aa"], name="a"),
+    )
+    result = result.groupby("a").apply(
+        lambda df: Series([df["b"].mean()], index=["b_mean"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg,idx",
+    [
+        [
+            [
+                1,
+                2,
+                3,
+            ],
+            [
+                0.1,
+                0.3,
+                0.2,
+            ],
+        ],
+        [
+            [
+                1,
+                2,
+                3,
+            ],
+            [
+                0.1,
+                0.2,
+                0.3,
+            ],
+        ],
+        [
+            [
+                1,
+                4,
+                3,
+            ],
+            [
+                0.1,
+                0.4,
+                0.2,
+            ],
+        ],
+    ],
+)
+def test_apply_nonmonotonic_float_index(arg, idx):
+    # GH 34455
+    df = DataFrame({"grp": arg, "col": arg}, index=idx)
+    result = df.groupby("grp", group_keys=False).apply(lambda x: x)
+    expected = df[["col"]]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("args, kwargs", [([True], {}), ([], {"numeric_only": True})])
+def test_apply_str_with_args(df, args, kwargs):
+    # GH#46479
+    gb = df.groupby("A")
+    result = gb.apply("sum", *args, **kwargs)
+    expected = gb.sum(numeric_only=True)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("name", ["some_name", None])
+def test_result_name_when_one_group(name):
+    # GH 46369
+    ser = Series([1, 2], name=name)
+    result = ser.groupby(["a", "a"], group_keys=False).apply(lambda x: x)
+    expected = Series([1, 2], name=name)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, op",
+    [
+        ("apply", lambda gb: gb.values[-1]),
+        ("apply", lambda gb: gb["b"].iloc[0]),
+        ("agg", "skew"),
+        ("agg", "kurt"),
+        ("agg", "prod"),
+        ("agg", "sum"),
+    ],
+)
+def test_empty_df(method, op):
+    # GH 47985
+    empty_df = DataFrame({"a": [], "b": []})
+    gb = empty_df.groupby("a", group_keys=True)
+    group = gb.b
+
+    result = getattr(group, method)(op)
+    expected = Series(
+        [], name="b", dtype="float64", index=Index([], dtype="float64", name="a")
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_include_groups():
+    # GH#7155
+    df = DataFrame({"a": [1, 1, 2], "b": [3, 4, 5]})
+    gb = df.groupby("a")
+    with pytest.raises(ValueError, match="include_groups=True is no longer allowed"):
+        gb.apply(lambda x: x.sum(), include_groups=True)
+
+
+@pytest.mark.parametrize("func, value", [(max, 2), (min, 1), (sum, 3)])
+def test_builtins_apply(func, value):
+    # GH#8155, GH#53974
+    # Builtins act as e.g. sum(group), which sums the column labels of group
+    df = DataFrame({0: [1, 1, 2], 1: [3, 4, 5], 2: [3, 4, 5]})
+    gb = df.groupby(0)
+    result = gb.apply(func)
+
+    expected = Series([value, value], index=Index([1, 2], name=0))
+    tm.assert_series_equal(result, expected)
+
+
+def test_inconsistent_return_type():
+    # GH5592
+    # inconsistent return type
+    df = DataFrame(
+        {
+            "A": ["Tiger", "Tiger", "Tiger", "Lamb", "Lamb", "Pony", "Pony"],
+            "B": Series(np.arange(7), dtype="int64"),
+            "C": pd.date_range("20130101", periods=7),
+        }
+    )
+
+    def f_0(grp):
+        return grp.iloc[0]
+
+    expected = df.groupby("A").first()[["B"]]
+    result = df.groupby("A").apply(f_0)[["B"]]
+    tm.assert_frame_equal(result, expected)
+
+    def f_1(grp):
+        if grp.name == "Tiger":
+            return None
+        return grp.iloc[0]
+
+    result = df.groupby("A").apply(f_1)[["B"]]
+    e = expected.copy()
+    e.loc["Tiger"] = np.nan
+    tm.assert_frame_equal(result, e)
+
+    def f_2(grp):
+        if grp.name == "Pony":
+            return None
+        return grp.iloc[0]
+
+    result = df.groupby("A").apply(f_2)[["B"]]
+    e = expected.copy()
+    e.loc["Pony"] = np.nan
+    tm.assert_frame_equal(result, e)
+
+    # 5592 revisited, with datetimes
+    def f_3(grp):
+        if grp.name == "Pony":
+            return None
+        return grp.iloc[0]
+
+    result = df.groupby("A").apply(f_3)[["C"]]
+    e = df.groupby("A").first()[["C"]]
+    e.loc["Pony"] = pd.NaT
+    tm.assert_frame_equal(result, e)
+
+    # scalar outputs
+    def f_4(grp):
+        if grp.name == "Pony":
+            return None
+        return grp.iloc[0].loc["C"]
+
+    result = df.groupby("A").apply(f_4)
+    e = df.groupby("A").first()["C"].copy()
+    e.loc["Pony"] = np.nan
+    e.name = None
+    tm.assert_series_equal(result, e)
+
+
+def test_nonreducer_nonstransform():
+    # GH3380, GH60619
+    # Was originally testing mutating in a UDF; now kept as an example
+    # of using apply with a nonreducer and nontransformer.
+    df = DataFrame(
+        {
+            "cat1": ["a"] * 8 + ["b"] * 6,
+            "cat2": ["c"] * 2
+            + ["d"] * 2
+            + ["e"] * 2
+            + ["f"] * 2
+            + ["c"] * 2
+            + ["d"] * 2
+            + ["e"] * 2,
+            "val": np.random.default_rng(2).integers(100, size=14),
+        }
+    )
+
+    def f(x):
+        x = x.copy()
+        x["rank"] = x.val.rank(method="min")
+        return x.groupby("cat2")["rank"].min()
+
+    expected = DataFrame(
+        {
+            "cat1": list("aaaabbb"),
+            "cat2": list("cdefcde"),
+            "rank": [3.0, 2.0, 5.0, 1.0, 2.0, 4.0, 1.0],
+        }
+    ).set_index(["cat1", "cat2"])["rank"]
+    result = df.groupby("cat1").apply(f)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_bin_groupby.py b/pandas/pandas/tests/groupby/test_bin_groupby.py
new file mode 100644
index 0000000000000..07d52308e308a
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_bin_groupby.py
@@ -0,0 +1,67 @@
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def assert_block_lengths(x):
+    assert len(x) == len(x._mgr.blocks[0].mgr_locs)
+    return 0
+
+
+def cumsum_max(x):
+    x.cumsum().max()
+    return 0
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        cumsum_max,
+        assert_block_lengths,
+    ],
+)
+def test_mgr_locs_updated(func):
+    # https://github.com/pandas-dev/pandas/issues/31802
+    # Some operations may require creating new blocks, which requires
+    # valid mgr_locs
+    df = pd.DataFrame({"A": ["a", "a", "a"], "B": ["a", "b", "b"], "C": [1, 1, 1]})
+    result = df.groupby(["A", "B"]).agg(func)
+    expected = pd.DataFrame(
+        {"C": [0, 0]},
+        index=pd.MultiIndex.from_product([["a"], ["a", "b"]], names=["A", "B"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "binner,closed,expected",
+    [
+        (
+            [0, 3, 6, 9],
+            "left",
+            [2, 5, 6],
+        ),
+        (
+            [0, 3, 6, 9],
+            "right",
+            [3, 6, 6],
+        ),
+        ([0, 3, 6], "left", [2, 5]),
+        (
+            [0, 3, 6],
+            "right",
+            [3, 6],
+        ),
+    ],
+)
+def test_generate_bins(binner, closed, expected):
+    values = np.array([1, 2, 3, 4, 5, 6], dtype=np.int64)
+    result = lib.generate_bins_dt64(
+        values, np.array(binner, dtype=np.int64), closed=closed
+    )
+    expected = np.array(expected, dtype=np.int64)
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_categorical.py b/pandas/pandas/tests/groupby/test_categorical.py
new file mode 100644
index 0000000000000..e49be8c00b426
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_categorical.py
@@ -0,0 +1,2122 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    qcut,
+)
+import pandas._testing as tm
+from pandas.api.typing import SeriesGroupBy
+from pandas.tests.groupby import get_groupby_method_args
+
+
+def cartesian_product_for_groupers(result, args, names, fill_value=np.nan):
+    """Reindex to a cartesian production for the groupers,
+    preserving the nature (Categorical) of each grouper
+    """
+
+    def f(a):
+        if isinstance(a, (CategoricalIndex, Categorical)):
+            categories = a.categories
+            a = Categorical.from_codes(
+                np.arange(len(categories)), categories=categories, ordered=a.ordered
+            )
+        return a
+
+    index = MultiIndex.from_product(map(f, args), names=names)
+    return result.reindex(index, fill_value=fill_value).sort_index()
+
+
+_results_for_groupbys_with_missing_categories = {
+    # This maps the builtin groupby functions to their expected outputs for
+    # missing categories when they are called on a categorical grouper with
+    # observed=False. Some functions are expected to return NaN, some zero.
+    # These expected values can be used across several tests (i.e. they are
+    # the same for SeriesGroupBy and DataFrameGroupBy) but they should only be
+    # hardcoded in one place.
+    "all": True,
+    "any": False,
+    "count": 0,
+    "corrwith": np.nan,
+    "first": np.nan,
+    "idxmax": np.nan,
+    "idxmin": np.nan,
+    "last": np.nan,
+    "max": np.nan,
+    "mean": np.nan,
+    "median": np.nan,
+    "min": np.nan,
+    "nth": np.nan,
+    "nunique": 0,
+    "prod": 1,
+    "quantile": np.nan,
+    "sem": np.nan,
+    "size": 0,
+    "skew": np.nan,
+    "kurt": np.nan,
+    "std": np.nan,
+    "sum": 0,
+    "var": np.nan,
+}
+
+
+def test_apply_use_categorical_name(df):
+    cats = qcut(df.C, 4)
+
+    def get_stats(group):
+        return {
+            "min": group.min(),
+            "max": group.max(),
+            "count": group.count(),
+            "mean": group.mean(),
+        }
+
+    result = df.groupby(cats, observed=False).D.apply(get_stats)
+    assert result.index.names[0] == "C"
+
+
+def test_basic():
+    cats = Categorical(
+        ["a", "a", "a", "b", "b", "b", "c", "c", "c"],
+        categories=["a", "b", "c", "d"],
+        ordered=True,
+    )
+    data = DataFrame({"a": [1, 1, 1, 2, 2, 2, 3, 4, 5], "b": cats})
+
+    exp_index = CategoricalIndex(list("abcd"), name="b", ordered=True)
+    expected = DataFrame({"a": [1, 2, 4, np.nan]}, index=exp_index)
+    result = data.groupby("b", observed=False).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_basic_single_grouper():
+    cat1 = Categorical(["a", "a", "b", "b"], categories=["a", "b", "z"], ordered=True)
+    cat2 = Categorical(["c", "d", "c", "d"], categories=["c", "d", "y"], ordered=True)
+    df = DataFrame({"A": cat1, "B": cat2, "values": [1, 2, 3, 4]})
+
+    gb = df.groupby("A", observed=False)
+    exp_idx = CategoricalIndex(["a", "b", "z"], name="A", ordered=True)
+    expected = DataFrame({"values": Series([3, 7, 0], index=exp_idx)})
+    result = gb.sum(numeric_only=True)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_basic_string(using_infer_string):
+    # GH 8623
+    x = DataFrame(
+        [[1, "John P. Doe"], [2, "Jane Dove"], [1, "John P. Doe"]],
+        columns=["person_id", "person_name"],
+    )
+    x["person_name"] = Categorical(x.person_name)
+
+    g = x.groupby(["person_id"], observed=False)
+    result = g.transform(lambda x: x)
+    tm.assert_frame_equal(result, x[["person_name"]])
+
+    result = x.drop_duplicates("person_name")
+    expected = x.iloc[[0, 1]]
+    tm.assert_frame_equal(result, expected)
+
+    def f(x):
+        return x.drop_duplicates("person_name").iloc[0]
+
+    result = g.apply(f)
+    expected = x[["person_name"]].iloc[[0, 1]]
+    expected.index = Index([1, 2], name="person_id")
+    dtype = "str" if using_infer_string else object
+    expected["person_name"] = expected["person_name"].astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_basic_monotonic():
+    # GH 9921
+    df = DataFrame({"a": [5, 15, 25]})
+    c = pd.cut(df.a, bins=[0, 10, 20, 30, 40])
+
+    result = df.a.groupby(c, observed=False).transform(sum)
+    tm.assert_series_equal(result, df["a"])
+
+    tm.assert_series_equal(
+        df.a.groupby(c, observed=False).transform(lambda xs: np.sum(xs)), df["a"]
+    )
+    result = df.groupby(c, observed=False).transform(sum)
+    expected = df[["a"]]
+    tm.assert_frame_equal(result, expected)
+
+    gbc = df.groupby(c, observed=False)
+    result = gbc.transform(lambda xs: np.max(xs, axis=0))
+    tm.assert_frame_equal(result, df[["a"]])
+
+    result2 = gbc.transform(lambda xs: np.max(xs, axis=0))
+    result3 = gbc.transform(max)
+    result4 = gbc.transform(np.maximum.reduce)
+    result5 = gbc.transform(lambda xs: np.maximum.reduce(xs))
+    tm.assert_frame_equal(result2, df[["a"]], check_dtype=False)
+    tm.assert_frame_equal(result3, df[["a"]], check_dtype=False)
+    tm.assert_frame_equal(result4, df[["a"]])
+    tm.assert_frame_equal(result5, df[["a"]])
+
+    # Filter
+    tm.assert_series_equal(df.a.groupby(c, observed=False).filter(np.all), df["a"])
+    tm.assert_frame_equal(df.groupby(c, observed=False).filter(np.all), df)
+
+
+def test_basic_non_monotonic():
+    df = DataFrame({"a": [5, 15, 25, -5]})
+    c = pd.cut(df.a, bins=[-10, 0, 10, 20, 30, 40])
+
+    result = df.a.groupby(c, observed=False).transform(sum)
+    tm.assert_series_equal(result, df["a"])
+
+    tm.assert_series_equal(
+        df.a.groupby(c, observed=False).transform(lambda xs: np.sum(xs)), df["a"]
+    )
+    result = df.groupby(c, observed=False).transform(sum)
+    expected = df[["a"]]
+    tm.assert_frame_equal(result, expected)
+
+    tm.assert_frame_equal(
+        df.groupby(c, observed=False).transform(lambda xs: np.sum(xs)), df[["a"]]
+    )
+
+
+def test_basic_cut_grouping():
+    # GH 9603
+    df = DataFrame({"a": [1, 0, 0, 0]})
+    c = pd.cut(df.a, [0, 1, 2, 3, 4], labels=Categorical(list("abcd")))
+    result = df.groupby(c, observed=False).apply(len)
+
+    exp_index = CategoricalIndex(c.values.categories, ordered=c.values.ordered)
+    expected = Series([1, 0, 0, 0], index=exp_index)
+    expected.index.name = "a"
+    tm.assert_series_equal(result, expected)
+
+
+def test_more_basic():
+    levels = ["foo", "bar", "baz", "qux"]
+    codes = np.random.default_rng(2).integers(0, 4, size=10)
+
+    cats = Categorical.from_codes(codes, levels, ordered=True)
+
+    data = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+
+    result = data.groupby(cats, observed=False).mean()
+
+    expected = data.groupby(np.asarray(cats), observed=False).mean()
+    exp_idx = CategoricalIndex(levels, categories=cats.categories, ordered=True)
+    expected = expected.reindex(exp_idx)
+
+    tm.assert_frame_equal(result, expected)
+
+    grouped = data.groupby(cats, observed=False)
+    desc_result = grouped.describe()
+
+    idx = cats.codes.argsort()
+    ord_labels = np.asarray(cats).take(idx)
+    ord_data = data.take(idx)
+
+    exp_cats = Categorical(
+        ord_labels, ordered=True, categories=["foo", "bar", "baz", "qux"]
+    )
+    expected = ord_data.groupby(exp_cats, sort=False, observed=False).describe()
+    tm.assert_frame_equal(desc_result, expected)
+
+    # GH 10460
+    expc = Categorical.from_codes(np.arange(4).repeat(8), levels, ordered=True)
+    exp = CategoricalIndex(expc)
+    tm.assert_index_equal(desc_result.stack().index.get_level_values(0), exp)
+    exp = Index(["count", "mean", "std", "min", "25%", "50%", "75%", "max"] * 4)
+    tm.assert_index_equal(desc_result.stack().index.get_level_values(1), exp)
+
+
+def test_level_get_group(observed):
+    # GH15155
+    df = DataFrame(
+        data=np.arange(2, 22, 2),
+        index=MultiIndex(
+            levels=[CategoricalIndex(["a", "b"]), range(10)],
+            codes=[[0] * 5 + [1] * 5, range(10)],
+            names=["Index1", "Index2"],
+        ),
+    )
+    g = df.groupby(level=["Index1"], observed=observed)
+
+    # expected should equal test.loc[["a"]]
+    # GH15166
+    expected = DataFrame(
+        data=np.arange(2, 12, 2),
+        index=MultiIndex(
+            levels=[CategoricalIndex(["a", "b"]), range(5)],
+            codes=[[0] * 5, range(5)],
+            names=["Index1", "Index2"],
+        ),
+    )
+    result = g.get_group(("a",))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sorting_with_different_categoricals():
+    # GH 24271
+    df = DataFrame(
+        {
+            "group": ["A"] * 6 + ["B"] * 6,
+            "dose": ["high", "med", "low"] * 4,
+            "outcomes": np.arange(12.0),
+        }
+    )
+
+    df.dose = Categorical(df.dose, categories=["low", "med", "high"], ordered=True)
+
+    result = df.groupby("group")["dose"].value_counts()
+    result = result.sort_index(level=0, sort_remaining=True)
+    index = ["low", "med", "high", "low", "med", "high"]
+    index = Categorical(index, categories=["low", "med", "high"], ordered=True)
+    index = [["A", "A", "A", "B", "B", "B"], CategoricalIndex(index)]
+    index = MultiIndex.from_arrays(index, names=["group", "dose"])
+    expected = Series([2] * 6, index=index, name="count")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_apply(ordered):
+    # GH 10138
+
+    dense = Categorical(list("abc"), ordered=ordered)
+
+    # 'b' is in the categories but not in the list
+    missing = Categorical(list("aaa"), categories=["a", "b"], ordered=ordered)
+    values = np.arange(len(dense))
+    df = DataFrame({"missing": missing, "dense": dense, "values": values})
+    grouped = df.groupby(["missing", "dense"], observed=True)
+
+    # missing category 'b' should still exist in the output index
+    idx = MultiIndex.from_arrays([missing, dense], names=["missing", "dense"])
+    expected = DataFrame([0, 1, 2.0], index=idx, columns=["values"])
+
+    result = grouped.apply(lambda x: np.mean(x, axis=0))
+    tm.assert_frame_equal(result, expected)
+
+    result = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+    result = grouped.agg(np.mean)
+    tm.assert_frame_equal(result, expected)
+
+    # but for transform we should still get back the original index
+    idx = MultiIndex.from_arrays([missing, dense], names=["missing", "dense"])
+    expected = Series(1, index=idx)
+    result = grouped.apply(lambda x: 1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_observed(request, using_infer_string, observed):
+    # multiple groupers, don't re-expand the output space
+    # of the grouper
+    # gh-14942 (implement)
+    # gh-10132 (back-compat)
+    # gh-8138 (back-compat)
+    # gh-8869
+
+    if using_infer_string and not observed:
+        # TODO(infer_string) this fails with filling the string column with 0
+        request.applymarker(pytest.mark.xfail(reason="TODO(infer_string)"))
+
+    cat1 = Categorical(["a", "a", "b", "b"], categories=["a", "b", "z"], ordered=True)
+    cat2 = Categorical(["c", "d", "c", "d"], categories=["c", "d", "y"], ordered=True)
+    df = DataFrame({"A": cat1, "B": cat2, "values": [1, 2, 3, 4]})
+    df["C"] = ["foo", "bar"] * 2
+
+    # multiple groupers with a non-cat
+    gb = df.groupby(["A", "B", "C"], observed=observed)
+    exp_index = MultiIndex.from_arrays(
+        [cat1, cat2, ["foo", "bar"] * 2], names=["A", "B", "C"]
+    )
+    expected = DataFrame({"values": Series([1, 2, 3, 4], index=exp_index)}).sort_index()
+    result = gb.sum()
+    if not observed:
+        expected = cartesian_product_for_groupers(
+            expected, [cat1, cat2, ["foo", "bar"]], list("ABC"), fill_value=0
+        )
+
+    tm.assert_frame_equal(result, expected)
+
+    gb = df.groupby(["A", "B"], observed=observed)
+    exp_index = MultiIndex.from_arrays([cat1, cat2], names=["A", "B"])
+    expected = DataFrame(
+        {"values": [1, 2, 3, 4], "C": ["foo", "bar", "foo", "bar"]}, index=exp_index
+    )
+    result = gb.sum()
+    if not observed:
+        expected = cartesian_product_for_groupers(
+            expected, [cat1, cat2], list("AB"), fill_value=0
+        )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_observed_single_column(observed):
+    # https://github.com/pandas-dev/pandas/issues/8138
+    d = {
+        "cat": Categorical(
+            ["a", "b", "a", "b"], categories=["a", "b", "c"], ordered=True
+        ),
+        "ints": [1, 1, 2, 2],
+        "val": [10, 20, 30, 40],
+    }
+    df = DataFrame(d)
+
+    groups_single_key = df.groupby("cat", observed=observed)
+    result = groups_single_key.mean()
+
+    exp_index = CategoricalIndex(
+        list("ab"), name="cat", categories=list("abc"), ordered=True
+    )
+    expected = DataFrame({"ints": [1.5, 1.5], "val": [20.0, 30]}, index=exp_index)
+    if not observed:
+        index = CategoricalIndex(
+            list("abc"), name="cat", categories=list("abc"), ordered=True
+        )
+        expected = expected.reindex(index)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_observed_two_columns(observed):
+    # https://github.com/pandas-dev/pandas/issues/8138
+    d = {
+        "cat": Categorical(
+            ["a", "b", "a", "b"], categories=["a", "b", "c"], ordered=True
+        ),
+        "ints": [1, 1, 2, 2],
+        "val": [10, 20, 30, 40],
+    }
+    df = DataFrame(d)
+    groups_double_key = df.groupby(["cat", "ints"], observed=observed)
+    result = groups_double_key.agg("mean")
+    expected = DataFrame(
+        {
+            "val": [10.0, 30.0, 20.0, 40.0],
+            "cat": Categorical(
+                ["a", "a", "b", "b"], categories=["a", "b", "c"], ordered=True
+            ),
+            "ints": [1, 2, 1, 2],
+        }
+    ).set_index(["cat", "ints"])
+    if not observed:
+        expected = cartesian_product_for_groupers(
+            expected, [df.cat.values, [1, 2]], ["cat", "ints"]
+        )
+
+    tm.assert_frame_equal(result, expected)
+
+    # GH 10132
+    for key in [("a", 1), ("b", 2), ("b", 1), ("a", 2)]:
+        c, i = key
+        result = groups_double_key.get_group(key)
+        expected = df[(df.cat == c) & (df.ints == i)]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_observed_with_as_index(observed):
+    # gh-8869
+    # with as_index
+    d = {
+        "foo": [10, 8, 4, 8, 4, 1, 1],
+        "bar": [10, 20, 30, 40, 50, 60, 70],
+        "baz": ["d", "c", "e", "a", "a", "d", "c"],
+    }
+    df = DataFrame(d)
+    cat = pd.cut(df["foo"], np.linspace(0, 10, 3))
+    df["range"] = cat
+    groups = df.groupby(["range", "baz"], as_index=False, observed=observed)
+    result = groups.agg("mean")
+
+    groups2 = df.groupby(["range", "baz"], as_index=True, observed=observed)
+    expected = groups2.agg("mean").reset_index()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_observed_codes_remap(observed):
+    d = {"C1": [3, 3, 4, 5], "C2": [1, 2, 3, 4], "C3": [10, 100, 200, 34]}
+    df = DataFrame(d)
+    values = pd.cut(df["C1"], [1, 2, 3, 6])
+    values.name = "cat"
+    groups_double_key = df.groupby([values, "C2"], observed=observed)
+
+    idx = MultiIndex.from_arrays([values, [1, 2, 3, 4]], names=["cat", "C2"])
+    expected = DataFrame(
+        {"C1": [3.0, 3.0, 4.0, 5.0], "C3": [10.0, 100.0, 200.0, 34.0]}, index=idx
+    )
+    if not observed:
+        expected = cartesian_product_for_groupers(
+            expected, [values.values, [1, 2, 3, 4]], ["cat", "C2"]
+        )
+
+    result = groups_double_key.agg("mean")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_observed_perf():
+    # we create a cartesian product, so this is
+    # non-performant if we don't use observed values
+    # gh-14942
+    df = DataFrame(
+        {
+            "cat": np.random.default_rng(2).integers(0, 255, size=30000),
+            "int_id": np.random.default_rng(2).integers(0, 255, size=30000),
+            "other_id": np.random.default_rng(2).integers(0, 10000, size=30000),
+            "foo": 0,
+        }
+    )
+    df["cat"] = df.cat.astype(str).astype("category")
+
+    grouped = df.groupby(["cat", "int_id", "other_id"], observed=True)
+    result = grouped.count()
+    assert result.index.levels[0].nunique() == df.cat.nunique()
+    assert result.index.levels[1].nunique() == df.int_id.nunique()
+    assert result.index.levels[2].nunique() == df.other_id.nunique()
+
+
+def test_observed_groups(observed):
+    # gh-20583
+    # test that we have the appropriate groups
+
+    cat = Categorical(["a", "c", "a"], categories=["a", "b", "c"])
+    df = DataFrame({"cat": cat, "vals": [1, 2, 3]})
+    g = df.groupby("cat", observed=observed)
+
+    result = g.groups
+    if observed:
+        expected = {"a": Index([0, 2], dtype="int64"), "c": Index([1], dtype="int64")}
+    else:
+        expected = {
+            "a": Index([0, 2], dtype="int64"),
+            "b": Index([], dtype="int64"),
+            "c": Index([1], dtype="int64"),
+        }
+
+    tm.assert_dict_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keys, expected_values, expected_index_levels",
+    [
+        ("a", [15, 9, 0], CategoricalIndex([1, 2, 3], name="a")),
+        (
+            ["a", "b"],
+            [7, 8, 0, 0, 0, 9, 0, 0, 0],
+            [CategoricalIndex([1, 2, 3], name="a"), Index([4, 5, 6])],
+        ),
+        (
+            ["a", "a2"],
+            [15, 0, 0, 0, 9, 0, 0, 0, 0],
+            [
+                CategoricalIndex([1, 2, 3], name="a"),
+                CategoricalIndex([1, 2, 3], name="a"),
+            ],
+        ),
+    ],
+)
+@pytest.mark.parametrize("test_series", [True, False])
+def test_unobserved_in_index(keys, expected_values, expected_index_levels, test_series):
+    # GH#49354 - ensure unobserved cats occur when grouping by index levels
+    df = DataFrame(
+        {
+            "a": Categorical([1, 1, 2], categories=[1, 2, 3]),
+            "a2": Categorical([1, 1, 2], categories=[1, 2, 3]),
+            "b": [4, 5, 6],
+            "c": [7, 8, 9],
+        }
+    ).set_index(["a", "a2"])
+    if "b" not in keys:
+        # Only keep b when it is used for grouping for consistent columns in the result
+        df = df.drop(columns="b")
+
+    gb = df.groupby(keys, observed=False)
+    if test_series:
+        gb = gb["c"]
+    result = gb.sum()
+
+    if len(keys) == 1:
+        index = expected_index_levels
+    else:
+        codes = [[0, 0, 0, 1, 1, 1, 2, 2, 2], 3 * [0, 1, 2]]
+        index = MultiIndex(
+            expected_index_levels,
+            codes=codes,
+            names=keys,
+        )
+    expected = DataFrame({"c": expected_values}, index=index)
+    if test_series:
+        expected = expected["c"]
+    tm.assert_equal(result, expected)
+
+
+def test_observed_groups_with_nan(observed):
+    # GH 24740
+    df = DataFrame(
+        {
+            "cat": Categorical(["a", np.nan, "a"], categories=["a", "b", "d"]),
+            "vals": [1, 2, 3],
+        }
+    )
+    g = df.groupby("cat", observed=observed)
+    result = g.groups
+    if observed:
+        expected = {"a": Index([0, 2], dtype="int64")}
+    else:
+        expected = {
+            "a": Index([0, 2], dtype="int64"),
+            "b": Index([], dtype="int64"),
+            "d": Index([], dtype="int64"),
+        }
+    tm.assert_dict_equal(result, expected)
+
+
+def test_observed_nth():
+    # GH 26385
+    cat = Categorical(["a", np.nan, np.nan], categories=["a", "b", "c"])
+    ser = Series([1, 2, 3])
+    df = DataFrame({"cat": cat, "ser": ser})
+
+    result = df.groupby("cat", observed=False)["ser"].nth(0)
+    expected = df["ser"].iloc[[0]]
+    tm.assert_series_equal(result, expected)
+
+
+def test_dataframe_categorical_with_nan(observed):
+    # GH 21151
+    s1 = Categorical([np.nan, "a", np.nan, "a"], categories=["a", "b", "c"])
+    s2 = Series([1, 2, 3, 4])
+    df = DataFrame({"s1": s1, "s2": s2})
+    result = df.groupby("s1", observed=observed).first().reset_index()
+    if observed:
+        expected = DataFrame(
+            {"s1": Categorical(["a"], categories=["a", "b", "c"]), "s2": [2]}
+        )
+    else:
+        expected = DataFrame(
+            {
+                "s1": Categorical(["a", "b", "c"], categories=["a", "b", "c"]),
+                "s2": [2, np.nan, np.nan],
+            }
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_dataframe_categorical_ordered_observed_sort(ordered, observed, sort):
+    # GH 25871: Fix groupby sorting on ordered Categoricals
+    # GH 25167: Groupby with observed=True doesn't sort
+
+    # Build a dataframe with cat having one unobserved category ('missing'),
+    # and a Series with identical values
+    label = Categorical(
+        ["d", "a", "b", "a", "d", "b"],
+        categories=["a", "b", "missing", "d"],
+        ordered=ordered,
+    )
+    val = Series(["d", "a", "b", "a", "d", "b"])
+    df = DataFrame({"label": label, "val": val})
+
+    # aggregate on the Categorical
+    result = df.groupby("label", observed=observed, sort=sort)["val"].aggregate("first")
+
+    # If ordering works, we expect index labels equal to aggregation results,
+    # except for 'observed=False': label 'missing' has aggregation None
+    label = Series(result.index.array, dtype="object")
+    aggr = Series(result.array)
+    if not observed:
+        aggr[aggr.isna()] = "missing"
+    if not all(label == aggr):
+        msg = (
+            "Labels and aggregation results not consistently sorted\n"
+            f"for (ordered={ordered}, observed={observed}, sort={sort})\n"
+            f"Result:\n{result}"
+        )
+        pytest.fail(msg)
+
+
+def test_datetime():
+    # GH9049: ensure backward compatibility
+    levels = pd.date_range("2014-01-01", periods=4)
+    codes = np.random.default_rng(2).integers(0, 4, size=10)
+
+    cats = Categorical.from_codes(codes, levels, ordered=True)
+
+    data = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+    result = data.groupby(cats, observed=False).mean()
+
+    expected = data.groupby(np.asarray(cats), observed=False).mean()
+    expected = expected.reindex(levels)
+    expected.index = CategoricalIndex(
+        expected.index, categories=expected.index, ordered=True
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+    grouped = data.groupby(cats, observed=False)
+    desc_result = grouped.describe()
+
+    idx = cats.codes.argsort()
+    ord_labels = cats.take(idx)
+    ord_data = data.take(idx)
+    expected = ord_data.groupby(ord_labels, observed=False).describe()
+    tm.assert_frame_equal(desc_result, expected)
+    tm.assert_index_equal(desc_result.index, expected.index)
+    tm.assert_index_equal(
+        desc_result.index.get_level_values(0), expected.index.get_level_values(0)
+    )
+
+    # GH 10460
+    expc = Categorical.from_codes(np.arange(4).repeat(8), levels, ordered=True)
+    exp = CategoricalIndex(expc)
+    tm.assert_index_equal((desc_result.stack().index.get_level_values(0)), exp)
+    exp = Index(["count", "mean", "std", "min", "25%", "50%", "75%", "max"] * 4)
+    tm.assert_index_equal((desc_result.stack().index.get_level_values(1)), exp)
+
+
+def test_categorical_index():
+    s = np.random.default_rng(2)
+    levels = ["foo", "bar", "baz", "qux"]
+    codes = s.integers(0, 4, size=20)
+    cats = Categorical.from_codes(codes, levels, ordered=True)
+    df = DataFrame(np.repeat(np.arange(20), 4).reshape(-1, 4), columns=list("abcd"))
+    df["cats"] = cats
+
+    # with a cat index
+    result = df.set_index("cats").groupby(level=0, observed=False).sum()
+    expected = df[list("abcd")].groupby(cats.codes, observed=False).sum()
+    expected.index = CategoricalIndex(
+        Categorical.from_codes([0, 1, 2, 3], levels, ordered=True), name="cats"
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # with a cat column, should produce a cat index
+    result = df.groupby("cats", observed=False).sum()
+    expected = df[list("abcd")].groupby(cats.codes, observed=False).sum()
+    expected.index = CategoricalIndex(
+        Categorical.from_codes([0, 1, 2, 3], levels, ordered=True), name="cats"
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_describe_categorical_columns():
+    # GH 11558
+    cats = CategoricalIndex(
+        ["qux", "foo", "baz", "bar"],
+        categories=["foo", "bar", "baz", "qux"],
+        ordered=True,
+    )
+    df = DataFrame(np.random.default_rng(2).standard_normal((20, 4)), columns=cats)
+    result = df.groupby([1, 2, 3, 4] * 5).describe()
+
+    tm.assert_index_equal(result.stack().columns, cats)
+    tm.assert_categorical_equal(result.stack().columns.values, cats.values)
+
+
+def test_unstack_categorical():
+    # GH11558 (example is taken from the original issue)
+    df = DataFrame(
+        {"a": range(10), "medium": ["A", "B"] * 5, "artist": list("XYXXY") * 2}
+    )
+    df["medium"] = df["medium"].astype("category")
+
+    gcat = df.groupby(["artist", "medium"], observed=False)["a"].count().unstack()
+    result = gcat.describe()
+
+    exp_columns = CategoricalIndex(["A", "B"], ordered=False, name="medium")
+    tm.assert_index_equal(result.columns, exp_columns)
+    tm.assert_categorical_equal(result.columns.values, exp_columns.values)
+
+    result = gcat["A"] + gcat["B"]
+    expected = Series([6, 4], index=Index(["X", "Y"], name="artist"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_bins_unequal_len():
+    # GH3011
+    series = Series([np.nan, np.nan, 1, 1, 2, 2, 3, 3, 4, 4])
+    bins = pd.cut(series.dropna().values, 4)
+
+    # len(bins) != len(series) here
+    with pytest.raises(ValueError, match="Grouper and axis must be same length"):
+        series.groupby(bins).mean()
+
+
+@pytest.mark.parametrize(
+    ["series", "data"],
+    [
+        # Group a series with length and index equal to those of the grouper.
+        (Series(range(4)), {"A": [0, 3], "B": [1, 2]}),
+        # Group a series with length equal to that of the grouper and index unequal to
+        # that of the grouper.
+        (Series(range(4)).rename(lambda idx: idx + 1), {"A": [2], "B": [0, 1]}),
+        # GH44179: Group a series with length unequal to that of the grouper.
+        (Series(range(7)), {"A": [0, 3], "B": [1, 2]}),
+    ],
+)
+def test_categorical_series(series, data):
+    # Group the given series by a series with categorical data type such that group A
+    # takes indices 0 and 3 and group B indices 1 and 2, obtaining the values mapped in
+    # the given data.
+    groupby = series.groupby(Series(list("ABBA"), dtype="category"), observed=False)
+    result = groupby.aggregate(list)
+    expected = Series(data, index=CategoricalIndex(data.keys()))
+    tm.assert_series_equal(result, expected)
+
+
+def test_as_index():
+    # GH13204
+    df = DataFrame(
+        {
+            "cat": Categorical([1, 2, 2], [1, 2, 3]),
+            "A": [10, 11, 11],
+            "B": [101, 102, 103],
+        }
+    )
+    result = df.groupby(["cat", "A"], as_index=False, observed=True).sum()
+    expected = DataFrame(
+        {
+            "cat": Categorical([1, 2], categories=df.cat.cat.categories),
+            "A": [10, 11],
+            "B": [101, 205],
+        },
+        columns=["cat", "A", "B"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # function grouper
+    f = lambda r: df.loc[r, "A"]
+    result = df.groupby(["cat", f], as_index=False, observed=True).sum()
+    expected = DataFrame(
+        {
+            "cat": Categorical([1, 2], categories=df.cat.cat.categories),
+            "level_1": [10, 11],
+            "A": [10, 22],
+            "B": [101, 205],
+        },
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # another not in-axis grouper (conflicting names in index)
+    s = Series(["a", "b", "b"], name="cat")
+    result = df.groupby(["cat", s], as_index=False, observed=True).sum()
+    expected = DataFrame(
+        {
+            "cat": ["a", "b"],
+            "A": [10, 22],
+            "B": [101, 205],
+        },
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # is original index dropped?
+    group_columns = ["cat", "A"]
+    expected = DataFrame(
+        {
+            "cat": Categorical([1, 2], categories=df.cat.cat.categories),
+            "A": [10, 11],
+            "B": [101, 205],
+        },
+        columns=["cat", "A", "B"],
+    )
+
+    for name in [None, "X", "B"]:
+        df.index = Index(list("abc"), name=name)
+        result = df.groupby(group_columns, as_index=False, observed=True).sum()
+
+        tm.assert_frame_equal(result, expected)
+
+
+def test_preserve_categories():
+    # GH-13179
+    categories = list("abc")
+
+    # ordered=True
+    df = DataFrame({"A": Categorical(list("ba"), categories=categories, ordered=True)})
+    sort_index = CategoricalIndex(categories, categories, ordered=True, name="A")
+    nosort_index = CategoricalIndex(list("bac"), categories, ordered=True, name="A")
+    tm.assert_index_equal(
+        df.groupby("A", sort=True, observed=False).first().index, sort_index
+    )
+    # GH#42482 - don't sort result when sort=False, even when ordered=True
+    tm.assert_index_equal(
+        df.groupby("A", sort=False, observed=False).first().index, nosort_index
+    )
+
+
+def test_preserve_categories_ordered_false():
+    # GH-13179
+    categories = list("abc")
+    df = DataFrame({"A": Categorical(list("ba"), categories=categories, ordered=False)})
+    sort_index = CategoricalIndex(categories, categories, ordered=False, name="A")
+    # GH#48749 - don't change order of categories
+    # GH#42482 - don't sort result when sort=False, even when ordered=True
+    nosort_index = CategoricalIndex(list("bac"), list("abc"), ordered=False, name="A")
+    tm.assert_index_equal(
+        df.groupby("A", sort=True, observed=False).first().index, sort_index
+    )
+    tm.assert_index_equal(
+        df.groupby("A", sort=False, observed=False).first().index, nosort_index
+    )
+
+
+@pytest.mark.parametrize("col", ["C1", "C2"])
+def test_preserve_categorical_dtype(col):
+    # GH13743, GH13854
+    df = DataFrame(
+        {
+            "A": [1, 2, 1, 1, 2],
+            "B": [10, 16, 22, 28, 34],
+            "C1": Categorical(list("abaab"), categories=list("bac"), ordered=False),
+            "C2": Categorical(list("abaab"), categories=list("bac"), ordered=True),
+        }
+    )
+    # single grouper
+    exp_full = DataFrame(
+        {
+            "A": [2.0, 1.0, np.nan],
+            "B": [25.0, 20.0, np.nan],
+            "C1": Categorical(list("bac"), categories=list("bac"), ordered=False),
+            "C2": Categorical(list("bac"), categories=list("bac"), ordered=True),
+        }
+    )
+    result1 = df.groupby(by=col, as_index=False, observed=False).mean(numeric_only=True)
+    result2 = (
+        df.groupby(by=col, as_index=True, observed=False)
+        .mean(numeric_only=True)
+        .reset_index()
+    )
+    expected = exp_full.reindex(columns=result1.columns)
+    tm.assert_frame_equal(result1, expected)
+    tm.assert_frame_equal(result2, expected)
+
+
+@pytest.mark.parametrize(
+    "func, values",
+    [
+        ("first", ["second", "first"]),
+        ("last", ["fourth", "third"]),
+        ("min", ["fourth", "first"]),
+        ("max", ["second", "third"]),
+    ],
+)
+def test_preserve_on_ordered_ops(func, values):
+    # gh-18502
+    # preserve the categoricals on ops
+    c = Categorical(["first", "second", "third", "fourth"], ordered=True)
+    df = DataFrame({"payload": [-1, -2, -1, -2], "col": c})
+    g = df.groupby("payload")
+    result = getattr(g, func)()
+    expected = DataFrame(
+        {"payload": [-2, -1], "col": Series(values, dtype=c.dtype)}
+    ).set_index("payload")
+    tm.assert_frame_equal(result, expected)
+
+    # we should also preserve categorical for SeriesGroupBy
+    sgb = df.groupby("payload")["col"]
+    result = getattr(sgb, func)()
+    expected = expected["col"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_categorical_no_compress():
+    data = Series(np.random.default_rng(2).standard_normal(9))
+
+    codes = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2])
+    cats = Categorical.from_codes(codes, [0, 1, 2], ordered=True)
+
+    result = data.groupby(cats, observed=False).mean()
+    exp = data.groupby(codes, observed=False).mean()
+
+    exp.index = CategoricalIndex(
+        exp.index, categories=cats.categories, ordered=cats.ordered
+    )
+    tm.assert_series_equal(result, exp)
+
+    codes = np.array([0, 0, 0, 1, 1, 1, 3, 3, 3])
+    cats = Categorical.from_codes(codes, [0, 1, 2, 3], ordered=True)
+
+    result = data.groupby(cats, observed=False).mean()
+    exp = data.groupby(codes, observed=False).mean().reindex(cats.categories)
+    exp.index = CategoricalIndex(
+        exp.index, categories=cats.categories, ordered=cats.ordered
+    )
+    tm.assert_series_equal(result, exp)
+
+
+def test_categorical_no_compress_string():
+    cats = Categorical(
+        ["a", "a", "a", "b", "b", "b", "c", "c", "c"],
+        categories=["a", "b", "c", "d"],
+        ordered=True,
+    )
+    data = DataFrame({"a": [1, 1, 1, 2, 2, 2, 3, 4, 5], "b": cats})
+
+    result = data.groupby("b", observed=False).mean()
+    result = result["a"].values
+    exp = np.array([1, 2, 4, np.nan])
+    tm.assert_numpy_array_equal(result, exp)
+
+
+def test_groupby_empty_with_category():
+    # GH-9614
+    # test fix for when group by on None resulted in
+    # coercion of dtype categorical -> float
+    df = DataFrame({"A": [None] * 3, "B": Categorical(["train", "train", "test"])})
+    result = df.groupby("A").first()["B"]
+    expected = Series(
+        Categorical([], categories=["test", "train"]),
+        index=Series([], dtype="object", name="A"),
+        name="B",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_sort():
+    # https://stackoverflow.com/questions/23814368/sorting-pandas-
+    #        categorical-labels-after-groupby
+    # This should result in a properly sorted Series so that the plot
+    # has a sorted x axis
+    # self.cat.groupby(['value_group'])['value_group'].count().plot(kind='bar')
+
+    df = DataFrame({"value": np.random.default_rng(2).integers(0, 10000, 10)})
+    labels = [f"{i} - {i + 499}" for i in range(0, 10000, 500)]
+    cat_labels = Categorical(labels, labels)
+
+    df = df.sort_values(by=["value"], ascending=True)
+    df["value_group"] = pd.cut(
+        df.value, range(0, 10500, 500), right=False, labels=cat_labels
+    )
+
+    res = df.groupby(["value_group"], observed=False)["value_group"].count()
+    exp = res[sorted(res.index, key=lambda x: float(x.split()[0]))]
+    exp.index = CategoricalIndex(exp.index, name=exp.index.name)
+    tm.assert_series_equal(res, exp)
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_sort2(sort, ordered):
+    # dataframe groupby sort was being ignored # GH 8868
+    # GH#48749 - don't change order of categories
+    # GH#42482 - don't sort result when sort=False, even when ordered=True
+    df = DataFrame(
+        [
+            ["(7.5, 10]", 10, 10],
+            ["(7.5, 10]", 8, 20],
+            ["(2.5, 5]", 5, 30],
+            ["(5, 7.5]", 6, 40],
+            ["(2.5, 5]", 4, 50],
+            ["(0, 2.5]", 1, 60],
+            ["(5, 7.5]", 7, 70],
+        ],
+        columns=["range", "foo", "bar"],
+    )
+    df["range"] = Categorical(df["range"], ordered=ordered)
+    result = df.groupby("range", sort=sort, observed=False).first()
+
+    if sort:
+        data_values = [[1, 60], [5, 30], [6, 40], [10, 10]]
+        index_values = ["(0, 2.5]", "(2.5, 5]", "(5, 7.5]", "(7.5, 10]"]
+    else:
+        data_values = [[10, 10], [5, 30], [6, 40], [1, 60]]
+        index_values = ["(7.5, 10]", "(2.5, 5]", "(5, 7.5]", "(0, 2.5]"]
+    expected = DataFrame(
+        data_values,
+        columns=["foo", "bar"],
+        index=CategoricalIndex(index_values, name="range", ordered=ordered),
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_sort_datetimelike(sort, ordered):
+    # GH10505
+    # GH#42482 - don't sort result when sort=False, even when ordered=True
+
+    # use same data as test_groupby_sort_categorical, which category is
+    # corresponding to datetime.month
+    df = DataFrame(
+        {
+            "dt": [
+                datetime(2011, 7, 1),
+                datetime(2011, 7, 1),
+                datetime(2011, 2, 1),
+                datetime(2011, 5, 1),
+                datetime(2011, 2, 1),
+                datetime(2011, 1, 1),
+                datetime(2011, 5, 1),
+            ],
+            "foo": [10, 8, 5, 6, 4, 1, 7],
+            "bar": [10, 20, 30, 40, 50, 60, 70],
+        },
+        columns=["dt", "foo", "bar"],
+    )
+
+    # ordered=True
+    df["dt"] = Categorical(df["dt"], ordered=ordered)
+    if sort:
+        data_values = [[1, 60], [5, 30], [6, 40], [10, 10]]
+        index_values = [
+            datetime(2011, 1, 1),
+            datetime(2011, 2, 1),
+            datetime(2011, 5, 1),
+            datetime(2011, 7, 1),
+        ]
+    else:
+        data_values = [[10, 10], [5, 30], [6, 40], [1, 60]]
+        index_values = [
+            datetime(2011, 7, 1),
+            datetime(2011, 2, 1),
+            datetime(2011, 5, 1),
+            datetime(2011, 1, 1),
+        ]
+    expected = DataFrame(
+        data_values,
+        columns=["foo", "bar"],
+        index=CategoricalIndex(index_values, name="dt", ordered=ordered),
+    )
+    result = df.groupby("dt", sort=sort, observed=False).first()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_empty_sum():
+    # https://github.com/pandas-dev/pandas/issues/18678
+    df = DataFrame(
+        {"A": Categorical(["a", "a", "b"], categories=["a", "b", "c"]), "B": [1, 2, 1]}
+    )
+    expected_idx = CategoricalIndex(["a", "b", "c"], name="A")
+
+    # 0 by default
+    result = df.groupby("A", observed=False).B.sum()
+    expected = Series([3, 1, 0], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+    # min_count=0
+    result = df.groupby("A", observed=False).B.sum(min_count=0)
+    expected = Series([3, 1, 0], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+    # min_count=1
+    result = df.groupby("A", observed=False).B.sum(min_count=1)
+    expected = Series([3, 1, np.nan], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+    # min_count>1
+    result = df.groupby("A", observed=False).B.sum(min_count=2)
+    expected = Series([3, np.nan, np.nan], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+
+def test_empty_prod():
+    # https://github.com/pandas-dev/pandas/issues/18678
+    df = DataFrame(
+        {"A": Categorical(["a", "a", "b"], categories=["a", "b", "c"]), "B": [1, 2, 1]}
+    )
+
+    expected_idx = CategoricalIndex(["a", "b", "c"], name="A")
+
+    # 1 by default
+    result = df.groupby("A", observed=False).B.prod()
+    expected = Series([2, 1, 1], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+    # min_count=0
+    result = df.groupby("A", observed=False).B.prod(min_count=0)
+    expected = Series([2, 1, 1], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+    # min_count=1
+    result = df.groupby("A", observed=False).B.prod(min_count=1)
+    expected = Series([2, 1, np.nan], expected_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_multiindex_categorical_datetime():
+    # https://github.com/pandas-dev/pandas/issues/21390
+
+    df = DataFrame(
+        {
+            "key1": Categorical(list("abcbabcba")),
+            "key2": Categorical(
+                list(pd.date_range("2018-06-01 00", freq="1min", periods=3)) * 3
+            ),
+            "values": np.arange(9),
+        }
+    )
+    result = df.groupby(["key1", "key2"], observed=False).mean()
+
+    idx = MultiIndex.from_product(
+        [
+            Categorical(["a", "b", "c"]),
+            Categorical(pd.date_range("2018-06-01 00", freq="1min", periods=3)),
+        ],
+        names=["key1", "key2"],
+    )
+    expected = DataFrame({"values": [0, 4, 8, 3, 4, 5, 6, np.nan, 2]}, index=idx)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "as_index, expected",
+    [
+        (
+            True,
+            Series(
+                index=MultiIndex.from_arrays(
+                    [Series([1, 1, 2], dtype="category"), [1, 2, 2]], names=["a", "b"]
+                ),
+                data=[1, 2, 3],
+                name="x",
+            ),
+        ),
+        (
+            False,
+            DataFrame(
+                {
+                    "a": Series([1, 1, 2], dtype="category"),
+                    "b": [1, 2, 2],
+                    "x": [1, 2, 3],
+                }
+            ),
+        ),
+    ],
+)
+def test_groupby_agg_observed_true_single_column(as_index, expected):
+    # GH-23970
+    df = DataFrame(
+        {"a": Series([1, 1, 2], dtype="category"), "b": [1, 2, 2], "x": [1, 2, 3]}
+    )
+
+    result = df.groupby(["a", "b"], as_index=as_index, observed=True)["x"].sum()
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("fill_value", [None, np.nan, pd.NaT])
+def test_shift(fill_value):
+    ct = Categorical(
+        ["a", "b", "c", "d"], categories=["a", "b", "c", "d"], ordered=False
+    )
+    expected = Categorical(
+        [None, "a", "b", "c"], categories=["a", "b", "c", "d"], ordered=False
+    )
+    res = ct.shift(1, fill_value=fill_value)
+    tm.assert_equal(res, expected)
+
+
+@pytest.fixture
+def df_cat(df):
+    """
+    DataFrame with multiple categorical columns and a column of integers.
+    Shortened so as not to contain all possible combinations of categories.
+    Useful for testing `observed` kwarg functionality on GroupBy objects.
+
+    Parameters
+    ----------
+    df: DataFrame
+        Non-categorical, longer DataFrame from another fixture, used to derive
+        this one
+
+    Returns
+    -------
+    df_cat: DataFrame
+    """
+    df_cat = df.copy()[:4]  # leave out some groups
+    df_cat["A"] = df_cat["A"].astype("category")
+    df_cat["B"] = df_cat["B"].astype("category")
+    df_cat["C"] = Series([1, 2, 3, 4])
+    df_cat = df_cat.drop(["D"], axis=1)
+    return df_cat
+
+
+@pytest.mark.parametrize("operation", ["agg", "apply"])
+def test_seriesgroupby_observed_true(df_cat, operation):
+    # GH#24880
+    # GH#49223 - order of results was wrong when grouping by index levels
+    lev_a = Index(["bar", "bar", "foo", "foo"], dtype=df_cat["A"].dtype, name="A")
+    lev_b = Index(["one", "three", "one", "two"], dtype=df_cat["B"].dtype, name="B")
+    index = MultiIndex.from_arrays([lev_a, lev_b])
+    expected = Series(data=[2, 4, 1, 3], index=index, name="C").sort_index()
+
+    grouped = df_cat.groupby(["A", "B"], observed=True)["C"]
+    result = getattr(grouped, operation)(sum)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("operation", ["agg", "apply"])
+@pytest.mark.parametrize("observed", [False, None])
+def test_seriesgroupby_observed_false_or_none(df_cat, observed, operation):
+    # GH 24880
+    # GH#49223 - order of results was wrong when grouping by index levels
+    index, _ = MultiIndex.from_product(
+        [
+            CategoricalIndex(["bar", "foo"], ordered=False),
+            CategoricalIndex(["one", "three", "two"], ordered=False),
+        ],
+        names=["A", "B"],
+    ).sortlevel()
+
+    expected = Series(data=[2, 4, 0, 1, 0, 3], index=index, name="C")
+    grouped = df_cat.groupby(["A", "B"], observed=observed)["C"]
+    result = getattr(grouped, operation)(sum)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "observed, index, data",
+    [
+        (
+            True,
+            MultiIndex.from_arrays(
+                [
+                    Index(["bar"] * 4 + ["foo"] * 4, dtype="category", name="A"),
+                    Index(
+                        ["one", "one", "three", "three", "one", "one", "two", "two"],
+                        dtype="category",
+                        name="B",
+                    ),
+                    Index(["min", "max"] * 4),
+                ]
+            ),
+            [2, 2, 4, 4, 1, 1, 3, 3],
+        ),
+        (
+            False,
+            MultiIndex.from_product(
+                [
+                    CategoricalIndex(["bar", "foo"], ordered=False),
+                    CategoricalIndex(["one", "three", "two"], ordered=False),
+                    Index(["min", "max"]),
+                ],
+                names=["A", "B", None],
+            ),
+            [2, 2, 4, 4, np.nan, np.nan, 1, 1, np.nan, np.nan, 3, 3],
+        ),
+        (
+            None,
+            MultiIndex.from_product(
+                [
+                    CategoricalIndex(["bar", "foo"], ordered=False),
+                    CategoricalIndex(["one", "three", "two"], ordered=False),
+                    Index(["min", "max"]),
+                ],
+                names=["A", "B", None],
+            ),
+            [2, 2, 4, 4, np.nan, np.nan, 1, 1, np.nan, np.nan, 3, 3],
+        ),
+    ],
+)
+def test_seriesgroupby_observed_apply_dict(df_cat, observed, index, data):
+    # GH 24880
+    expected = Series(data=data, index=index, name="C")
+    result = df_cat.groupby(["A", "B"], observed=observed)["C"].apply(
+        lambda x: {"min": x.min(), "max": x.max()}
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_categorical_series_dataframe_consistent(df_cat):
+    # GH 20416
+    expected = df_cat.groupby(["A", "B"], observed=False)["C"].mean()
+    result = df_cat.groupby(["A", "B"], observed=False).mean()["C"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_cat_preserves_structure(observed, ordered):
+    # GH 28787
+    df = DataFrame(
+        {"Name": Categorical(["Bob", "Greg"], ordered=ordered), "Item": [1, 2]},
+        columns=["Name", "Item"],
+    )
+    expected = df.copy()
+
+    result = (
+        df.groupby("Name", observed=observed)
+        .agg(DataFrame.sum, skipna=True)
+        .reset_index()
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_get_nonexistent_category():
+    # Accessing a Category that is not in the dataframe
+    df = DataFrame({"var": ["a", "a", "b", "b"], "val": range(4)})
+    with pytest.raises(KeyError, match="'vau'"):
+        df.groupby("var").apply(lambda rows: DataFrame({"val": [rows.iloc[-1]["vau"]]}))
+
+
+def test_series_groupby_on_2_categoricals_unobserved(reduction_func, observed):
+    # GH 17605
+    if reduction_func == "ngroup":
+        pytest.skip("ngroup is not truly a reduction")
+
+    df = DataFrame(
+        {
+            "cat_1": Categorical(list("AABB"), categories=list("ABCD")),
+            "cat_2": Categorical(list("AB") * 2, categories=list("ABCD")),
+            "value": [0.1] * 4,
+        }
+    )
+    args = get_groupby_method_args(reduction_func, df)
+
+    expected_length = 4 if observed else 16
+
+    series_groupby = df.groupby(["cat_1", "cat_2"], observed=observed)["value"]
+
+    if reduction_func == "corrwith":
+        # TODO: implemented SeriesGroupBy.corrwith. See GH 32293
+        assert not hasattr(series_groupby, reduction_func)
+        return
+
+    agg = getattr(series_groupby, reduction_func)
+
+    if not observed and reduction_func in ["idxmin", "idxmax"]:
+        # idxmin and idxmax are designed to fail on empty inputs
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            agg(*args)
+        return
+
+    result = agg(*args)
+
+    assert len(result) == expected_length
+
+
+def test_series_groupby_on_2_categoricals_unobserved_zeroes_or_nans(
+    reduction_func, request
+):
+    # GH 17605
+    # Tests whether the unobserved categories in the result contain 0 or NaN
+
+    if reduction_func == "ngroup":
+        pytest.skip("ngroup is not truly a reduction")
+
+    if reduction_func == "corrwith":  # GH 32293
+        mark = pytest.mark.xfail(
+            reason="TODO: implemented SeriesGroupBy.corrwith. See GH 32293"
+        )
+        request.applymarker(mark)
+
+    df = DataFrame(
+        {
+            "cat_1": Categorical(list("AABB"), categories=list("ABC")),
+            "cat_2": Categorical(list("AB") * 2, categories=list("ABC")),
+            "value": [0.1] * 4,
+        }
+    )
+    unobserved = [tuple("AC"), tuple("BC"), tuple("CA"), tuple("CB"), tuple("CC")]
+    args = get_groupby_method_args(reduction_func, df)
+
+    series_groupby = df.groupby(["cat_1", "cat_2"], observed=False)["value"]
+    agg = getattr(series_groupby, reduction_func)
+
+    if reduction_func in ["idxmin", "idxmax"]:
+        # idxmin and idxmax are designed to fail on empty inputs
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            agg(*args)
+        return
+
+    result = agg(*args)
+
+    missing_fillin = _results_for_groupbys_with_missing_categories[reduction_func]
+
+    for idx in unobserved:
+        val = result.loc[idx]
+        assert (pd.isna(missing_fillin) and pd.isna(val)) or (val == missing_fillin)
+
+    # If we expect unobserved values to be zero, we also expect the dtype to be int.
+    # Except for .sum(). If the observed categories sum to dtype=float (i.e. their
+    # sums have decimals), then the zeros for the missing categories should also be
+    # floats.
+    if missing_fillin == 0:
+        if reduction_func in ["count", "nunique", "size"]:
+            assert np.issubdtype(result.dtype, np.integer)
+        else:
+            assert reduction_func in ["sum", "any"]
+
+
+def test_dataframe_groupby_on_2_categoricals_when_observed_is_true(reduction_func):
+    # GH 23865
+    # GH 27075
+    # Ensure that df.groupby, when 'by' is two Categorical variables,
+    # does not return the categories that are not in df when observed=True
+    if reduction_func == "ngroup":
+        pytest.skip("ngroup does not return the Categories on the index")
+
+    df = DataFrame(
+        {
+            "cat_1": Categorical(list("AABB"), categories=list("ABC")),
+            "cat_2": Categorical(list("1111"), categories=list("12")),
+            "value": [0.1, 0.1, 0.1, 0.1],
+        }
+    )
+    unobserved_cats = [("A", "2"), ("B", "2"), ("C", "1"), ("C", "2")]
+
+    df_grp = df.groupby(["cat_1", "cat_2"], observed=True)
+
+    args = get_groupby_method_args(reduction_func, df)
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        warn_msg = ""
+    with tm.assert_produces_warning(warn, match=warn_msg):
+        res = getattr(df_grp, reduction_func)(*args)
+
+    for cat in unobserved_cats:
+        assert cat not in res.index
+
+
+@pytest.mark.parametrize("observed", [False, None])
+def test_dataframe_groupby_on_2_categoricals_when_observed_is_false(
+    reduction_func, observed
+):
+    # GH 23865
+    # GH 27075
+    # Ensure that df.groupby, when 'by' is two Categorical variables,
+    # returns the categories that are not in df when observed=False/None
+
+    if reduction_func == "ngroup":
+        pytest.skip("ngroup does not return the Categories on the index")
+
+    df = DataFrame(
+        {
+            "cat_1": Categorical(list("AABB"), categories=list("ABC")),
+            "cat_2": Categorical(list("1111"), categories=list("12")),
+            "value": [0.1, 0.1, 0.1, 0.1],
+        }
+    )
+    unobserved_cats = [("A", "2"), ("B", "2"), ("C", "1"), ("C", "2")]
+
+    df_grp = df.groupby(["cat_1", "cat_2"], observed=observed)
+
+    args = get_groupby_method_args(reduction_func, df)
+
+    if not observed and reduction_func in ["idxmin", "idxmax"]:
+        # idxmin and idxmax are designed to fail on empty inputs
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            getattr(df_grp, reduction_func)(*args)
+        return
+
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        warn_msg = ""
+    with tm.assert_produces_warning(warn, match=warn_msg):
+        res = getattr(df_grp, reduction_func)(*args)
+
+    expected = _results_for_groupbys_with_missing_categories[reduction_func]
+
+    if expected is np.nan:
+        assert res.loc[unobserved_cats].isnull().all().all()
+    else:
+        assert (res.loc[unobserved_cats] == expected).all().all()
+
+
+def test_series_groupby_categorical_aggregation_getitem():
+    # GH 8870
+    d = {"foo": [10, 8, 4, 1], "bar": [10, 20, 30, 40], "baz": ["d", "c", "d", "c"]}
+    df = DataFrame(d)
+    cat = pd.cut(df["foo"], np.linspace(0, 20, 5))
+    df["range"] = cat
+    groups = df.groupby(["range", "baz"], as_index=True, sort=True, observed=False)
+    result = groups["foo"].agg("mean")
+    expected = groups.agg("mean")["foo"]
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, expected_values",
+    [(Series.nunique, [1, 1, 2]), (Series.count, [1, 2, 2])],
+)
+def test_groupby_agg_categorical_columns(func, expected_values):
+    # 31256
+    df = DataFrame(
+        {
+            "id": [0, 1, 2, 3, 4],
+            "groups": [0, 1, 1, 2, 2],
+            "value": Categorical([0, 0, 0, 0, 1]),
+        }
+    ).set_index("id")
+    result = df.groupby("groups").agg(func)
+
+    expected = DataFrame(
+        {"value": expected_values}, index=Index([0, 1, 2], name="groups")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_agg_non_numeric():
+    df = DataFrame({"A": Categorical(["a", "a", "b"], categories=["a", "b", "c"])})
+    expected = DataFrame({"A": [2, 1]}, index=np.array([1, 2]))
+
+    result = df.groupby([1, 2, 1]).agg(Series.nunique)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby([1, 2, 1]).nunique()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["first", "last"])
+def test_groupby_first_returned_categorical_instead_of_dataframe(func):
+    # GH 28641: groupby drops index, when grouping over categorical column with
+    # first/last. Renamed Categorical instead of DataFrame previously.
+    df = DataFrame({"A": [1997], "B": Series(["b"], dtype="category").cat.as_ordered()})
+    df_grouped = df.groupby("A")["B"]
+    result = getattr(df_grouped, func)()
+
+    # ordered categorical dtype should be preserved
+    expected = Series(
+        ["b"], index=Index([1997], name="A"), name="B", dtype=df["B"].dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_read_only_category_no_sort():
+    # GH33410
+    cats = np.array([1, 2])
+    cats.flags.writeable = False
+    df = DataFrame(
+        {"a": [1, 3, 5, 7], "b": Categorical([1, 1, 2, 2], categories=Index(cats))}
+    )
+    expected = DataFrame(data={"a": [2.0, 6.0]}, index=CategoricalIndex(cats, name="b"))
+    result = df.groupby("b", sort=False, observed=False).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sorted_missing_category_values():
+    # GH 28597
+    df = DataFrame(
+        {
+            "foo": [
+                "small",
+                "large",
+                "large",
+                "large",
+                "medium",
+                "large",
+                "large",
+                "medium",
+            ],
+            "bar": ["C", "A", "A", "C", "A", "C", "A", "C"],
+        }
+    )
+    df["foo"] = (
+        df["foo"]
+        .astype("category")
+        .cat.set_categories(["tiny", "small", "medium", "large"], ordered=True)
+    )
+
+    expected = DataFrame(
+        {
+            "tiny": {"A": 0, "C": 0},
+            "small": {"A": 0, "C": 1},
+            "medium": {"A": 1, "C": 1},
+            "large": {"A": 3, "C": 2},
+        }
+    )
+    expected = expected.rename_axis("bar", axis="index")
+    expected.columns = CategoricalIndex(
+        ["tiny", "small", "medium", "large"],
+        categories=["tiny", "small", "medium", "large"],
+        ordered=True,
+        name="foo",
+        dtype="category",
+    )
+
+    result = df.groupby(["bar", "foo"], observed=False).size().unstack()
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_agg_cython_category_not_implemented_fallback():
+    # https://github.com/pandas-dev/pandas/issues/31450
+    df = DataFrame({"col_num": [1, 1, 2, 3]})
+    df["col_cat"] = df["col_num"].astype("category")
+
+    result = df.groupby("col_num").col_cat.first()
+
+    # ordered categorical dtype should definitely be preserved;
+    #  this is unordered, so is less-clear case (if anything, it should raise)
+    expected = Series(
+        [1, 2, 3],
+        index=Index([1, 2, 3], name="col_num"),
+        name="col_cat",
+        dtype=df["col_cat"].dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby("col_num").agg({"col_cat": "first"})
+    expected = expected.to_frame()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_aggregate_categorical_with_isnan():
+    # GH 29837
+    df = DataFrame(
+        {
+            "A": [1, 1, 1, 1],
+            "B": [1, 2, 1, 2],
+            "numerical_col": [0.1, 0.2, np.nan, 0.3],
+            "object_col": ["foo", "bar", "foo", "fee"],
+            "categorical_col": ["foo", "bar", "foo", "fee"],
+        }
+    )
+
+    df = df.astype({"categorical_col": "category"})
+
+    result = df.groupby(["A", "B"]).agg(lambda df: df.isna().sum())
+    index = MultiIndex.from_arrays([[1, 1], [1, 2]], names=("A", "B"))
+    expected = DataFrame(
+        data={
+            "numerical_col": [1, 0],
+            "object_col": [0, 0],
+            "categorical_col": [0, 0],
+        },
+        index=index,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_transform():
+    # GH 29037
+    df = DataFrame(
+        {
+            "package_id": [1, 1, 1, 2, 2, 3],
+            "status": [
+                "Waiting",
+                "OnTheWay",
+                "Delivered",
+                "Waiting",
+                "OnTheWay",
+                "Waiting",
+            ],
+        }
+    )
+
+    delivery_status_type = pd.CategoricalDtype(
+        categories=["Waiting", "OnTheWay", "Delivered"], ordered=True
+    )
+    df["status"] = df["status"].astype(delivery_status_type)
+    df["last_status"] = df.groupby("package_id")["status"].transform(max)
+    result = df.copy()
+
+    expected = DataFrame(
+        {
+            "package_id": [1, 1, 1, 2, 2, 3],
+            "status": [
+                "Waiting",
+                "OnTheWay",
+                "Delivered",
+                "Waiting",
+                "OnTheWay",
+                "Waiting",
+            ],
+            "last_status": [
+                "Waiting",
+                "Waiting",
+                "Waiting",
+                "Waiting",
+                "Waiting",
+                "Waiting",
+            ],
+        }
+    )
+
+    expected["status"] = expected["status"].astype(delivery_status_type)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["first", "last"])
+def test_series_groupby_first_on_categorical_col_grouped_on_2_categoricals(
+    func: str, observed: bool
+):
+    # GH 34951
+    cat = Categorical([0, 0, 1, 1])
+    val = [0, 1, 1, 0]
+    df = DataFrame({"a": cat, "b": cat, "c": val})
+
+    cat2 = Categorical([0, 1])
+    idx = MultiIndex.from_product([cat2, cat2], names=["a", "b"])
+    expected_dict = {
+        "first": Series([0, np.nan, np.nan, 1], idx, name="c"),
+        "last": Series([1, np.nan, np.nan, 0], idx, name="c"),
+    }
+
+    expected = expected_dict[func]
+    if observed:
+        expected = expected.dropna().astype(np.int64)
+
+    srs_grp = df.groupby(["a", "b"], observed=observed)["c"]
+    result = getattr(srs_grp, func)()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["first", "last"])
+def test_df_groupby_first_on_categorical_col_grouped_on_2_categoricals(
+    func: str, observed: bool
+):
+    # GH 34951
+    cat = Categorical([0, 0, 1, 1])
+    val = [0, 1, 1, 0]
+    df = DataFrame({"a": cat, "b": cat, "c": val})
+
+    cat2 = Categorical([0, 1])
+    idx = MultiIndex.from_product([cat2, cat2], names=["a", "b"])
+    expected_dict = {
+        "first": Series([0, np.nan, np.nan, 1], idx, name="c"),
+        "last": Series([1, np.nan, np.nan, 0], idx, name="c"),
+    }
+
+    expected = expected_dict[func].to_frame()
+    if observed:
+        expected = expected.dropna().astype(np.int64)
+
+    df_grp = df.groupby(["a", "b"], observed=observed)
+    result = getattr(df_grp, func)()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_categorical_indices_unused_categories():
+    # GH#38642
+    df = DataFrame(
+        {
+            "key": Categorical(["b", "b", "a"], categories=["a", "b", "c"]),
+            "col": range(3),
+        }
+    )
+    grouped = df.groupby("key", sort=False, observed=False)
+    result = grouped.indices
+    expected = {
+        "b": np.array([0, 1], dtype="intp"),
+        "a": np.array([2], dtype="intp"),
+        "c": np.array([], dtype="intp"),
+    }
+    assert result.keys() == expected.keys()
+    for key in result.keys():
+        tm.assert_numpy_array_equal(result[key], expected[key])
+
+
+@pytest.mark.parametrize("func", ["first", "last"])
+def test_groupby_last_first_preserve_categoricaldtype(func):
+    # GH#33090
+    df = DataFrame({"a": [1, 2, 3]})
+    df["b"] = df["a"].astype("category")
+    result = getattr(df.groupby("a")["b"], func)()
+    expected = Series(
+        Categorical([1, 2, 3]), name="b", index=Index([1, 2, 3], name="a")
+    )
+    tm.assert_series_equal(expected, result)
+
+
+def test_groupby_categorical_observed_nunique():
+    # GH#45128
+    df = DataFrame({"a": [1, 2], "b": [1, 2], "c": [10, 11]})
+    df = df.astype(dtype={"a": "category", "b": "category"})
+    result = df.groupby(["a", "b"], observed=True).nunique()["c"]
+    expected = Series(
+        [1, 1],
+        index=MultiIndex.from_arrays(
+            [CategoricalIndex([1, 2], name="a"), CategoricalIndex([1, 2], name="b")]
+        ),
+        name="c",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_categorical_aggregate_functions():
+    # GH#37275
+    dtype = pd.CategoricalDtype(categories=["small", "big"], ordered=True)
+    df = DataFrame(
+        [[1, "small"], [1, "big"], [2, "small"]], columns=["grp", "description"]
+    ).astype({"description": dtype})
+
+    result = df.groupby("grp")["description"].max()
+    expected = Series(
+        ["big", "small"],
+        index=Index([1, 2], name="grp"),
+        name="description",
+        dtype=pd.CategoricalDtype(categories=["small", "big"], ordered=True),
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_categorical_dropna(observed, dropna):
+    # GH#48645 - dropna should have no impact on the result when there are no NA values
+    cat = Categorical([1, 2], categories=[1, 2, 3])
+    df = DataFrame({"x": Categorical([1, 2], categories=[1, 2, 3]), "y": [3, 4]})
+    gb = df.groupby("x", observed=observed, dropna=dropna)
+    result = gb.sum()
+
+    if observed:
+        expected = DataFrame({"y": [3, 4]}, index=cat)
+    else:
+        index = CategoricalIndex([1, 2, 3], [1, 2, 3])
+        expected = DataFrame({"y": [3, 4, 0]}, index=index)
+    expected.index.name = "x"
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["range", "single", "multi"])
+@pytest.mark.parametrize("ordered", [True, False])
+def test_category_order_reducer(
+    request, as_index, sort, observed, reduction_func, index_kind, ordered
+):
+    # GH#48749
+    if reduction_func == "corrwith" and not as_index and index_kind != "single":
+        msg = "GH#49950 - corrwith with as_index=False may not have grouping column"
+        request.applymarker(pytest.mark.xfail(reason=msg))
+    elif index_kind != "range" and not as_index:
+        pytest.skip(reason="Result doesn't have categories, nothing to test")
+    df = DataFrame(
+        {
+            "a": Categorical([2, 1, 2, 3], categories=[1, 4, 3, 2], ordered=ordered),
+            "b": range(4),
+        }
+    )
+    if index_kind == "range":
+        keys = ["a"]
+    elif index_kind == "single":
+        keys = ["a"]
+        df = df.set_index(keys)
+    elif index_kind == "multi":
+        keys = ["a", "a2"]
+        df["a2"] = df["a"]
+        df = df.set_index(keys)
+    args = get_groupby_method_args(reduction_func, df)
+    gb = df.groupby(keys, as_index=as_index, sort=sort, observed=observed)
+
+    if not observed and reduction_func in ["idxmin", "idxmax"]:
+        # idxmin and idxmax are designed to fail on empty inputs
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            getattr(gb, reduction_func)(*args)
+        return
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        warn_msg = ""
+    with tm.assert_produces_warning(warn, match=warn_msg):
+        op_result = getattr(gb, reduction_func)(*args)
+    if as_index:
+        result = op_result.index.get_level_values("a").categories
+    else:
+        result = op_result["a"].cat.categories
+    expected = Index([1, 4, 3, 2])
+    tm.assert_index_equal(result, expected)
+
+    if index_kind == "multi":
+        result = op_result.index.get_level_values("a2").categories
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["single", "multi"])
+@pytest.mark.parametrize("ordered", [True, False])
+def test_category_order_transformer(
+    as_index, sort, observed, transformation_func, index_kind, ordered
+):
+    # GH#48749
+    df = DataFrame(
+        {
+            "a": Categorical([2, 1, 2, 3], categories=[1, 4, 3, 2], ordered=ordered),
+            "b": range(4),
+        }
+    )
+    if index_kind == "single":
+        keys = ["a"]
+        df = df.set_index(keys)
+    elif index_kind == "multi":
+        keys = ["a", "a2"]
+        df["a2"] = df["a"]
+        df = df.set_index(keys)
+    args = get_groupby_method_args(transformation_func, df)
+    gb = df.groupby(keys, as_index=as_index, sort=sort, observed=observed)
+    op_result = getattr(gb, transformation_func)(*args)
+    result = op_result.index.get_level_values("a").categories
+    expected = Index([1, 4, 3, 2])
+    tm.assert_index_equal(result, expected)
+
+    if index_kind == "multi":
+        result = op_result.index.get_level_values("a2").categories
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["range", "single", "multi"])
+@pytest.mark.parametrize("method", ["head", "tail"])
+@pytest.mark.parametrize("ordered", [True, False])
+def test_category_order_head_tail(
+    as_index, sort, observed, method, index_kind, ordered
+):
+    # GH#48749
+    df = DataFrame(
+        {
+            "a": Categorical([2, 1, 2, 3], categories=[1, 4, 3, 2], ordered=ordered),
+            "b": range(4),
+        }
+    )
+    if index_kind == "range":
+        keys = ["a"]
+    elif index_kind == "single":
+        keys = ["a"]
+        df = df.set_index(keys)
+    elif index_kind == "multi":
+        keys = ["a", "a2"]
+        df["a2"] = df["a"]
+        df = df.set_index(keys)
+    gb = df.groupby(keys, as_index=as_index, sort=sort, observed=observed)
+    op_result = getattr(gb, method)()
+    if index_kind == "range":
+        result = op_result["a"].cat.categories
+    else:
+        result = op_result.index.get_level_values("a").categories
+    expected = Index([1, 4, 3, 2])
+    tm.assert_index_equal(result, expected)
+
+    if index_kind == "multi":
+        result = op_result.index.get_level_values("a2").categories
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["range", "single", "multi"])
+@pytest.mark.parametrize("method", ["apply", "agg", "transform"])
+@pytest.mark.parametrize("ordered", [True, False])
+def test_category_order_apply(as_index, sort, observed, method, index_kind, ordered):
+    # GH#48749
+    if (method == "transform" and index_kind == "range") or (
+        not as_index and index_kind != "range"
+    ):
+        pytest.skip("No categories in result, nothing to test")
+    df = DataFrame(
+        {
+            "a": Categorical([2, 1, 2, 3], categories=[1, 4, 3, 2], ordered=ordered),
+            "b": range(4),
+        }
+    )
+    if index_kind == "range":
+        keys = ["a"]
+    elif index_kind == "single":
+        keys = ["a"]
+        df = df.set_index(keys)
+    elif index_kind == "multi":
+        keys = ["a", "a2"]
+        df["a2"] = df["a"]
+        df = df.set_index(keys)
+    gb = df.groupby(keys, as_index=as_index, sort=sort, observed=observed)
+    op_result = getattr(gb, method)(lambda x: x.sum(numeric_only=True))
+    if (method == "transform" or not as_index) and index_kind == "range":
+        result = op_result["a"].cat.categories
+    else:
+        result = op_result.index.get_level_values("a").categories
+    expected = Index([1, 4, 3, 2])
+    tm.assert_index_equal(result, expected)
+
+    if index_kind == "multi":
+        result = op_result.index.get_level_values("a2").categories
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["range", "single", "multi"])
+def test_many_categories(as_index, sort, index_kind, ordered):
+    # GH#48749 - Test when the grouper has many categories
+    if index_kind != "range" and not as_index:
+        pytest.skip(reason="Result doesn't have categories, nothing to test")
+    categories = np.arange(9999, -1, -1)
+    grouper = Categorical([2, 1, 2, 3], categories=categories, ordered=ordered)
+    df = DataFrame({"a": grouper, "b": range(4)})
+    if index_kind == "range":
+        keys = ["a"]
+    elif index_kind == "single":
+        keys = ["a"]
+        df = df.set_index(keys)
+    elif index_kind == "multi":
+        keys = ["a", "a2"]
+        df["a2"] = df["a"]
+        df = df.set_index(keys)
+    gb = df.groupby(keys, as_index=as_index, sort=sort, observed=True)
+    result = gb.sum()
+
+    # Test is setup so that data and index are the same values
+    data = [3, 2, 1] if sort else [2, 1, 3]
+
+    index = CategoricalIndex(
+        data, categories=grouper.categories, ordered=ordered, name="a"
+    )
+    if as_index:
+        expected = DataFrame({"b": data})
+        if index_kind == "multi":
+            expected.index = MultiIndex.from_frame(DataFrame({"a": index, "a2": index}))
+        else:
+            expected.index = index
+    elif index_kind == "multi":
+        expected = DataFrame({"a": Series(index), "a2": Series(index), "b": data})
+    else:
+        expected = DataFrame({"a": Series(index), "b": data})
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("test_series", [True, False])
+@pytest.mark.parametrize("keys", [["a1"], ["a1", "a2"]])
+def test_agg_list(request, as_index, observed, reduction_func, test_series, keys):
+    # GH#52760
+    if test_series and reduction_func == "corrwith":
+        assert not hasattr(SeriesGroupBy, "corrwith")
+        pytest.skip("corrwith not implemented for SeriesGroupBy")
+    elif reduction_func == "corrwith":
+        msg = "GH#32293: attempts to call SeriesGroupBy.corrwith"
+        request.applymarker(pytest.mark.xfail(reason=msg))
+
+    df = DataFrame({"a1": [0, 0, 1], "a2": [2, 3, 3], "b": [4, 5, 6]})
+    df = df.astype({"a1": "category", "a2": "category"})
+    if "a2" not in keys:
+        df = df.drop(columns="a2")
+    gb = df.groupby(by=keys, as_index=as_index, observed=observed)
+    if test_series:
+        gb = gb["b"]
+    args = get_groupby_method_args(reduction_func, df)
+
+    if not observed and reduction_func in ["idxmin", "idxmax"] and keys == ["a1", "a2"]:
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            gb.agg([reduction_func], *args)
+        return
+
+    result = gb.agg([reduction_func], *args)
+    expected = getattr(gb, reduction_func)(*args)
+
+    if as_index and (test_series or reduction_func == "size"):
+        expected = expected.to_frame(reduction_func)
+    if not test_series:
+        expected.columns = MultiIndex.from_tuples(
+            [(ind, "") for ind in expected.columns[:-1]] + [("b", reduction_func)]
+        )
+    elif not as_index:
+        expected.columns = keys + [reduction_func]
+
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_counting.py b/pandas/pandas/tests/groupby/test_counting.py
new file mode 100644
index 0000000000000..679f7eb7f7f11
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_counting.py
@@ -0,0 +1,394 @@
+from itertools import product
+from string import ascii_lowercase
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestCounting:
+    def test_cumcount(self):
+        df = DataFrame([["a"], ["a"], ["a"], ["b"], ["a"]], columns=["A"])
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0, 1, 2, 0, 3])
+
+        tm.assert_series_equal(expected, g.cumcount())
+        tm.assert_series_equal(expected, sg.cumcount())
+
+    def test_cumcount_empty(self):
+        ge = DataFrame().groupby(level=0)
+        se = Series(dtype=object).groupby(level=0)
+
+        # edge case, as this is usually considered float
+        e = Series(dtype="int64")
+
+        tm.assert_series_equal(e, ge.cumcount())
+        tm.assert_series_equal(e, se.cumcount())
+
+    def test_cumcount_dupe_index(self):
+        df = DataFrame(
+            [["a"], ["a"], ["a"], ["b"], ["a"]], columns=["A"], index=[0] * 5
+        )
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0, 1, 2, 0, 3], index=[0] * 5)
+
+        tm.assert_series_equal(expected, g.cumcount())
+        tm.assert_series_equal(expected, sg.cumcount())
+
+    def test_cumcount_mi(self):
+        mi = MultiIndex.from_tuples([[0, 1], [1, 2], [2, 2], [2, 2], [1, 0]])
+        df = DataFrame([["a"], ["a"], ["a"], ["b"], ["a"]], columns=["A"], index=mi)
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0, 1, 2, 0, 3], index=mi)
+
+        tm.assert_series_equal(expected, g.cumcount())
+        tm.assert_series_equal(expected, sg.cumcount())
+
+    def test_cumcount_groupby_not_col(self):
+        df = DataFrame(
+            [["a"], ["a"], ["a"], ["b"], ["a"]], columns=["A"], index=[0] * 5
+        )
+        g = df.groupby([0, 0, 0, 1, 0])
+        sg = g.A
+
+        expected = Series([0, 1, 2, 0, 3], index=[0] * 5)
+
+        tm.assert_series_equal(expected, g.cumcount())
+        tm.assert_series_equal(expected, sg.cumcount())
+
+    def test_ngroup(self):
+        df = DataFrame({"A": list("aaaba")})
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0, 0, 0, 1, 0])
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_distinct(self):
+        df = DataFrame({"A": list("abcde")})
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series(range(5), dtype="int64")
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_one_group(self):
+        df = DataFrame({"A": [0] * 5})
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0] * 5)
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_empty(self):
+        ge = DataFrame().groupby(level=0)
+        se = Series(dtype=object).groupby(level=0)
+
+        # edge case, as this is usually considered float
+        e = Series(dtype="int64")
+
+        tm.assert_series_equal(e, ge.ngroup())
+        tm.assert_series_equal(e, se.ngroup())
+
+    def test_ngroup_series_matches_frame(self):
+        df = DataFrame({"A": list("aaaba")})
+        s = Series(list("aaaba"))
+
+        tm.assert_series_equal(df.groupby(s).ngroup(), s.groupby(s).ngroup())
+
+    def test_ngroup_dupe_index(self):
+        df = DataFrame({"A": list("aaaba")}, index=[0] * 5)
+        g = df.groupby("A")
+        sg = g.A
+
+        expected = Series([0, 0, 0, 1, 0], index=[0] * 5)
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_mi(self):
+        mi = MultiIndex.from_tuples([[0, 1], [1, 2], [2, 2], [2, 2], [1, 0]])
+        df = DataFrame({"A": list("aaaba")}, index=mi)
+        g = df.groupby("A")
+        sg = g.A
+        expected = Series([0, 0, 0, 1, 0], index=mi)
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_groupby_not_col(self):
+        df = DataFrame({"A": list("aaaba")}, index=[0] * 5)
+        g = df.groupby([0, 0, 0, 1, 0])
+        sg = g.A
+
+        expected = Series([0, 0, 0, 1, 0], index=[0] * 5)
+
+        tm.assert_series_equal(expected, g.ngroup())
+        tm.assert_series_equal(expected, sg.ngroup())
+
+    def test_ngroup_descending(self):
+        df = DataFrame(["a", "a", "b", "a", "b"], columns=["A"])
+        g = df.groupby(["A"])
+
+        ascending = Series([0, 0, 1, 0, 1])
+        descending = Series([1, 1, 0, 1, 0])
+
+        tm.assert_series_equal(descending, (g.ngroups - 1) - ascending)
+        tm.assert_series_equal(ascending, g.ngroup(ascending=True))
+        tm.assert_series_equal(descending, g.ngroup(ascending=False))
+
+    def test_ngroup_matches_cumcount(self):
+        # verify one manually-worked out case works
+        df = DataFrame(
+            [["a", "x"], ["a", "y"], ["b", "x"], ["a", "x"], ["b", "y"]],
+            columns=["A", "X"],
+        )
+        g = df.groupby(["A", "X"])
+        g_ngroup = g.ngroup()
+        g_cumcount = g.cumcount()
+        expected_ngroup = Series([0, 1, 2, 0, 3])
+        expected_cumcount = Series([0, 0, 0, 1, 0])
+
+        tm.assert_series_equal(g_ngroup, expected_ngroup)
+        tm.assert_series_equal(g_cumcount, expected_cumcount)
+
+    def test_ngroup_cumcount_pair(self):
+        # brute force comparison for all small series
+        for p in product(range(3), repeat=4):
+            df = DataFrame({"a": p})
+            g = df.groupby(["a"])
+
+            order = sorted(set(p))
+            ngroupd = [order.index(val) for val in p]
+            cumcounted = [p[:i].count(val) for i, val in enumerate(p)]
+
+            tm.assert_series_equal(g.ngroup(), Series(ngroupd))
+            tm.assert_series_equal(g.cumcount(), Series(cumcounted))
+
+    def test_ngroup_respects_groupby_order(self, sort):
+        df = DataFrame({"a": np.random.default_rng(2).choice(list("abcdef"), 100)})
+        g = df.groupby("a", sort=sort)
+        df["group_id"] = -1
+        df["group_index"] = -1
+
+        for i, (_, group) in enumerate(g):
+            df.loc[group.index, "group_id"] = i
+            for j, ind in enumerate(group.index):
+                df.loc[ind, "group_index"] = j
+
+        tm.assert_series_equal(Series(df["group_id"].values), g.ngroup())
+        tm.assert_series_equal(Series(df["group_index"].values), g.cumcount())
+
+    @pytest.mark.parametrize(
+        "datetimelike",
+        [
+            [Timestamp(f"2016-05-{i:02d} 20:09:25+00:00") for i in range(1, 4)],
+            [Timestamp(f"2016-05-{i:02d} 20:09:25") for i in range(1, 4)],
+            [Timestamp(f"2016-05-{i:02d} 20:09:25", tz="UTC") for i in range(1, 4)],
+            [Timedelta(x, unit="h") for x in range(1, 4)],
+            [Period(freq="2W", year=2017, month=x) for x in range(1, 4)],
+        ],
+    )
+    def test_count_with_datetimelike(self, datetimelike):
+        # test for #13393, where DataframeGroupBy.count() fails
+        # when counting a datetimelike column.
+
+        df = DataFrame({"x": ["a", "a", "b"], "y": datetimelike})
+        res = df.groupby("x").count()
+        expected = DataFrame({"y": [2, 1]}, index=["a", "b"])
+        expected.index.name = "x"
+        tm.assert_frame_equal(expected, res)
+
+    def test_count_with_only_nans_in_first_group(self):
+        # GH21956
+        df = DataFrame({"A": [np.nan, np.nan], "B": ["a", "b"], "C": [1, 2]})
+        result = df.groupby(["A", "B"]).C.count()
+        mi = MultiIndex(levels=[[], ["a", "b"]], codes=[[], []], names=["A", "B"])
+        expected = Series([], index=mi, dtype=np.int64, name="C")
+        tm.assert_series_equal(result, expected, check_index_type=False)
+
+    def test_count_groupby_column_with_nan_in_groupby_column(self):
+        # https://github.com/pandas-dev/pandas/issues/32841
+        df = DataFrame({"A": [1, 1, 1, 1, 1], "B": [5, 4, np.nan, 3, 0]})
+        res = df.groupby(["B"]).count()
+        expected = DataFrame(
+            index=Index([0.0, 3.0, 4.0, 5.0], name="B"), data={"A": [1, 1, 1, 1]}
+        )
+        tm.assert_frame_equal(expected, res)
+
+    def test_groupby_count_dateparseerror(self):
+        dr = date_range(start="1/1/2012", freq="5min", periods=10)
+
+        # BAD Example, datetimes first
+        ser = Series(np.arange(10), index=[dr, np.arange(10)])
+        grouped = ser.groupby(lambda x: x[1] % 2 == 0)
+        result = grouped.count()
+
+        ser = Series(np.arange(10), index=[np.arange(10), dr])
+        grouped = ser.groupby(lambda x: x[0] % 2 == 0)
+        expected = grouped.count()
+
+        tm.assert_series_equal(result, expected)
+
+
+def test_groupby_timedelta_cython_count():
+    df = DataFrame(
+        {"g": list("ab" * 2), "delta": np.arange(4).astype("timedelta64[ns]")}
+    )
+    expected = Series([2, 2], index=Index(["a", "b"], name="g"), name="delta")
+    result = df.groupby("g").delta.count()
+    tm.assert_series_equal(expected, result)
+
+
+def test_count():
+    n = 1 << 15
+    dr = date_range("2015-08-30", periods=n // 10, freq="min")
+
+    df = DataFrame(
+        {
+            "1st": np.random.default_rng(2).choice(list(ascii_lowercase), n),
+            "2nd": np.random.default_rng(2).integers(0, 5, n),
+            "3rd": np.random.default_rng(2).standard_normal(n).round(3),
+            "4th": np.random.default_rng(2).integers(-10, 10, n),
+            "5th": np.random.default_rng(2).choice(dr, n),
+            "6th": np.random.default_rng(2).standard_normal(n).round(3),
+            "7th": np.random.default_rng(2).standard_normal(n).round(3),
+            "8th": np.random.default_rng(2).choice(dr, n)
+            - np.random.default_rng(2).choice(dr, 1),
+            "9th": np.random.default_rng(2).choice(list(ascii_lowercase), n),
+        }
+    )
+
+    for col in df.columns.drop(["1st", "2nd", "4th"]):
+        df.loc[np.random.default_rng(2).choice(n, n // 10), col] = np.nan
+
+    df["9th"] = df["9th"].astype("category")
+
+    for key in ["1st", "2nd", ["1st", "2nd"]]:
+        left = df.groupby(key).count()
+        right = df.groupby(key).apply(DataFrame.count)
+        tm.assert_frame_equal(left, right)
+
+
+def test_count_non_nulls():
+    # GH#5610
+    # count counts non-nulls
+    df = DataFrame(
+        [[1, 2, "foo"], [1, np.nan, "bar"], [3, np.nan, np.nan]],
+        columns=["A", "B", "C"],
+    )
+
+    count_as = df.groupby("A").count()
+    count_not_as = df.groupby("A", as_index=False).count()
+
+    expected = DataFrame([[1, 2], [0, 0]], columns=["B", "C"], index=[1, 3])
+    expected.index.name = "A"
+    tm.assert_frame_equal(count_not_as, expected.reset_index())
+    tm.assert_frame_equal(count_as, expected)
+
+    count_B = df.groupby("A")["B"].count()
+    tm.assert_series_equal(count_B, expected["B"])
+
+
+def test_count_object():
+    df = DataFrame({"a": ["a"] * 3 + ["b"] * 3, "c": [2] * 3 + [3] * 3})
+    result = df.groupby("c").a.count()
+    expected = Series([3, 3], index=Index([2, 3], name="c"), name="a")
+    tm.assert_series_equal(result, expected)
+
+
+def test_count_object_nan():
+    df = DataFrame({"a": ["a", np.nan, np.nan] + ["b"] * 3, "c": [2] * 3 + [3] * 3})
+    result = df.groupby("c").a.count()
+    expected = Series([1, 3], index=Index([2, 3], name="c"), name="a")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("typ", ["object", "float32"])
+def test_count_cross_type(typ):
+    # GH8169
+    # Set float64 dtype to avoid upcast when setting nan below
+    vals = np.hstack(
+        (
+            np.random.default_rng(2).integers(0, 5, (10, 2)),
+            np.random.default_rng(2).integers(0, 2, (10, 2)),
+        )
+    ).astype("float64")
+
+    df = DataFrame(vals, columns=["a", "b", "c", "d"])
+    df[df == 2] = np.nan
+    expected = df.groupby(["c", "d"]).count()
+
+    df["a"] = df["a"].astype(typ)
+    df["b"] = df["b"].astype(typ)
+    result = df.groupby(["c", "d"]).count()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_lower_int_prec_count():
+    df = DataFrame(
+        {
+            "a": np.array([0, 1, 2, 100], np.int8),
+            "b": np.array([1, 2, 3, 6], np.uint32),
+            "c": np.array([4, 5, 6, 8], np.int16),
+            "grp": list("ab" * 2),
+        }
+    )
+    result = df.groupby("grp").count()
+    expected = DataFrame(
+        {"a": [2, 2], "b": [2, 2], "c": [2, 2]}, index=Index(list("ab"), name="grp")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_count_uses_size_on_exception():
+    class RaisingObjectException(Exception):
+        pass
+
+    class RaisingObject:
+        def __init__(self, msg="I will raise inside Cython") -> None:
+            super().__init__()
+            self.msg = msg
+
+        def __eq__(self, other):
+            # gets called in Cython to check that raising calls the method
+            raise RaisingObjectException(self.msg)
+
+    df = DataFrame({"a": [RaisingObject() for _ in range(4)], "grp": list("ab" * 2)})
+    result = df.groupby("grp").count()
+    expected = DataFrame({"a": [2, 2]}, index=Index(list("ab"), name="grp"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_count_arrow_string_array(any_string_dtype):
+    # GH#54751
+    pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {"a": [1, 2, 3], "b": Series(["a", "b", "a"], dtype=any_string_dtype)}
+    )
+    result = df.groupby("a").count()
+    expected = DataFrame({"b": 1}, index=Index([1, 2, 3], name="a"))
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_cumulative.py b/pandas/pandas/tests/groupby/test_cumulative.py
new file mode 100644
index 0000000000000..b0a0414c1feb2
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_cumulative.py
@@ -0,0 +1,332 @@
+import numpy as np
+import pytest
+
+from pandas.errors import UnsupportedFunctionCall
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(
+    params=[np.int32, np.int64, np.float32, np.float64, "Int64", "Float64"],
+    ids=["np.int32", "np.int64", "np.float32", "np.float64", "Int64", "Float64"],
+)
+def dtypes_for_minmax(request):
+    """
+    Fixture of dtypes with min and max values used for testing
+    cummin and cummax
+    """
+    dtype = request.param
+
+    np_type = dtype
+    if dtype == "Int64":
+        np_type = np.int64
+    elif dtype == "Float64":
+        np_type = np.float64
+
+    min_val = (
+        np.iinfo(np_type).min
+        if np.dtype(np_type).kind == "i"
+        else np.finfo(np_type).min
+    )
+    max_val = (
+        np.iinfo(np_type).max
+        if np.dtype(np_type).kind == "i"
+        else np.finfo(np_type).max
+    )
+
+    return (dtype, min_val, max_val)
+
+
+def test_groupby_cumprod():
+    # GH 4095
+    df = DataFrame({"key": ["b"] * 10, "value": 2})
+
+    actual = df.groupby("key")["value"].cumprod()
+    expected = df.groupby("key", group_keys=False)["value"].apply(lambda x: x.cumprod())
+    expected.name = "value"
+    tm.assert_series_equal(actual, expected)
+
+    df = DataFrame({"key": ["b"] * 100, "value": 2})
+    df["value"] = df["value"].astype(float)
+    actual = df.groupby("key")["value"].cumprod()
+    expected = df.groupby("key", group_keys=False)["value"].apply(lambda x: x.cumprod())
+    expected.name = "value"
+    tm.assert_series_equal(actual, expected)
+
+
+@pytest.mark.skip_ubsan
+def test_groupby_cumprod_overflow():
+    # GH#37493 if we overflow we return garbage consistent with numpy
+    df = DataFrame({"key": ["b"] * 4, "value": 100_000})
+    actual = df.groupby("key")["value"].cumprod()
+    expected = Series(
+        [100_000, 10_000_000_000, 1_000_000_000_000_000, 7766279631452241920],
+        name="value",
+    )
+    tm.assert_series_equal(actual, expected)
+
+    numpy_result = df.groupby("key", group_keys=False)["value"].apply(
+        lambda x: x.cumprod()
+    )
+    numpy_result.name = "value"
+    tm.assert_series_equal(actual, numpy_result)
+
+
+def test_groupby_cumprod_nan_influences_other_columns():
+    # GH#48064
+    df = DataFrame(
+        {
+            "a": 1,
+            "b": [1, np.nan, 2],
+            "c": [1, 2, 3.0],
+        }
+    )
+    result = df.groupby("a").cumprod(numeric_only=True, skipna=False)
+    expected = DataFrame({"b": [1, np.nan, np.nan], "c": [1, 2, 6.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummin(dtypes_for_minmax):
+    dtype = dtypes_for_minmax[0]
+
+    # GH 15048
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    expected_mins = [3, 3, 3, 2, 2, 2, 2, 1]
+
+    df = base_df.astype(dtype)
+    expected = DataFrame({"B": expected_mins}).astype(dtype)
+    result = df.groupby("A").cummin()
+    tm.assert_frame_equal(result, expected)
+    result = df.groupby("A", group_keys=False).B.apply(lambda x: x.cummin()).to_frame()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummin_min_value_for_dtype(dtypes_for_minmax):
+    dtype = dtypes_for_minmax[0]
+    min_val = dtypes_for_minmax[1]
+
+    # GH 15048
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    expected_mins = [3, 3, 3, 2, 2, 2, 2, 1]
+    expected = DataFrame({"B": expected_mins}).astype(dtype)
+    df = base_df.astype(dtype)
+    df.loc[[2, 6], "B"] = min_val
+    df.loc[[1, 5], "B"] = min_val + 1
+    expected.loc[[2, 3, 6, 7], "B"] = min_val
+    expected.loc[[1, 5], "B"] = min_val + 1  # should not be rounded to min_val
+    result = df.groupby("A").cummin()
+    tm.assert_frame_equal(result, expected, check_exact=True)
+    expected = (
+        df.groupby("A", group_keys=False).B.apply(lambda x: x.cummin()).to_frame()
+    )
+    tm.assert_frame_equal(result, expected, check_exact=True)
+
+
+def test_cummin_nan_in_some_values(dtypes_for_minmax):
+    # Explicit cast to float to avoid implicit cast when setting nan
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    base_df = base_df.astype({"B": "float"})
+    base_df.loc[[0, 2, 4, 6], "B"] = np.nan
+    expected = DataFrame({"B": [np.nan, 4, np.nan, 2, np.nan, 3, np.nan, 1]})
+    result = base_df.groupby("A").cummin()
+    tm.assert_frame_equal(result, expected)
+    expected = (
+        base_df.groupby("A", group_keys=False).B.apply(lambda x: x.cummin()).to_frame()
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummin_datetime():
+    # GH 15561
+    df = DataFrame({"a": [1], "b": pd.to_datetime(["2001"])})
+    expected = Series(pd.to_datetime("2001"), index=[0], name="b")
+
+    result = df.groupby("a")["b"].cummin()
+    tm.assert_series_equal(expected, result)
+
+
+def test_cummin_getattr_series():
+    # GH 15635
+    df = DataFrame({"a": [1, 2, 1], "b": [1, 2, 2]})
+    result = df.groupby("a").b.cummin()
+    expected = Series([1, 2, 1], name="b")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["cummin", "cummax"])
+@pytest.mark.parametrize("dtype", ["UInt64", "Int64", "Float64", "float", "boolean"])
+def test_cummin_max_all_nan_column(method, dtype):
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [np.nan] * 8})
+    base_df["B"] = base_df["B"].astype(dtype)
+    grouped = base_df.groupby("A")
+
+    expected = DataFrame({"B": [np.nan] * 8}, dtype=dtype)
+    result = getattr(grouped, method)()
+    tm.assert_frame_equal(expected, result)
+
+    result = getattr(grouped["B"], method)().to_frame()
+    tm.assert_frame_equal(expected, result)
+
+
+def test_cummax(dtypes_for_minmax):
+    dtype = dtypes_for_minmax[0]
+
+    # GH 15048
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    expected_maxs = [3, 4, 4, 4, 2, 3, 3, 3]
+
+    df = base_df.astype(dtype)
+
+    expected = DataFrame({"B": expected_maxs}).astype(dtype)
+    result = df.groupby("A").cummax()
+    tm.assert_frame_equal(result, expected)
+    result = df.groupby("A", group_keys=False).B.apply(lambda x: x.cummax()).to_frame()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummax_min_value_for_dtype(dtypes_for_minmax):
+    dtype = dtypes_for_minmax[0]
+    max_val = dtypes_for_minmax[2]
+
+    # GH 15048
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    expected_maxs = [3, 4, 4, 4, 2, 3, 3, 3]
+
+    df = base_df.astype(dtype)
+    df.loc[[2, 6], "B"] = max_val
+    expected = DataFrame({"B": expected_maxs}).astype(dtype)
+    expected.loc[[2, 3, 6, 7], "B"] = max_val
+    result = df.groupby("A").cummax()
+    tm.assert_frame_equal(result, expected)
+    expected = (
+        df.groupby("A", group_keys=False).B.apply(lambda x: x.cummax()).to_frame()
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummax_nan_in_some_values(dtypes_for_minmax):
+    # Test nan in some values
+    # Explicit cast to float to avoid implicit cast when setting nan
+    base_df = DataFrame({"A": [1, 1, 1, 1, 2, 2, 2, 2], "B": [3, 4, 3, 2, 2, 3, 2, 1]})
+    base_df = base_df.astype({"B": "float"})
+    base_df.loc[[0, 2, 4, 6], "B"] = np.nan
+    expected = DataFrame({"B": [np.nan, 4, np.nan, 4, np.nan, 3, np.nan, 3]})
+    result = base_df.groupby("A").cummax()
+    tm.assert_frame_equal(result, expected)
+    expected = (
+        base_df.groupby("A", group_keys=False).B.apply(lambda x: x.cummax()).to_frame()
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cummax_datetime():
+    # GH 15561
+    df = DataFrame({"a": [1], "b": pd.to_datetime(["2001"])})
+    expected = Series(pd.to_datetime("2001"), index=[0], name="b")
+
+    result = df.groupby("a")["b"].cummax()
+    tm.assert_series_equal(expected, result)
+
+
+def test_cummax_getattr_series():
+    # GH 15635
+    df = DataFrame({"a": [1, 2, 1], "b": [2, 1, 1]})
+    result = df.groupby("a").b.cummax()
+    expected = Series([2, 1, 2], name="b")
+    tm.assert_series_equal(result, expected)
+
+
+def test_cummax_i8_at_implementation_bound():
+    # the minimum value used to be treated as NPY_NAT+1 instead of NPY_NAT
+    #  for int64 dtype GH#46382
+    ser = Series([pd.NaT._value + n for n in range(5)])
+    df = DataFrame({"A": 1, "B": ser, "C": ser._values.view("M8[ns]")})
+    gb = df.groupby("A")
+
+    res = gb.cummax()
+    exp = df[["B", "C"]]
+    tm.assert_frame_equal(res, exp)
+
+
+@pytest.mark.parametrize("method", ["cummin", "cummax"])
+@pytest.mark.parametrize("dtype", ["float", "Int64", "Float64"])
+@pytest.mark.parametrize(
+    "groups,expected_data",
+    [
+        ([1, 1, 1], [1, None, None]),
+        ([1, 2, 3], [1, None, 2]),
+        ([1, 3, 3], [1, None, None]),
+    ],
+)
+def test_cummin_max_skipna(method, dtype, groups, expected_data):
+    # GH-34047
+    df = DataFrame({"a": Series([1, None, 2], dtype=dtype)})
+    orig = df.copy()
+    gb = df.groupby(groups)["a"]
+
+    result = getattr(gb, method)(skipna=False)
+    expected = Series(expected_data, dtype=dtype, name="a")
+
+    # check we didn't accidentally alter df
+    tm.assert_frame_equal(df, orig)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["cummin", "cummax"])
+def test_cummin_max_skipna_multiple_cols(method):
+    # Ensure missing value in "a" doesn't cause "b" to be nan-filled
+    df = DataFrame({"a": [np.nan, 2.0, 2.0], "b": [2.0, 2.0, 2.0]})
+    gb = df.groupby([1, 1, 1])[["a", "b"]]
+
+    result = getattr(gb, method)(skipna=False)
+    expected = DataFrame({"a": [np.nan, np.nan, np.nan], "b": [2.0, 2.0, 2.0]})
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["cumprod", "cumsum"])
+def test_numpy_compat(func):
+    # see gh-12811
+    df = DataFrame({"A": [1, 2, 1], "B": [1, 2, 3]})
+    g = df.groupby("A")
+
+    msg = "numpy operations are not valid with groupby"
+
+    with pytest.raises(UnsupportedFunctionCall, match=msg):
+        getattr(g, func)(1, 2, 3)
+    with pytest.raises(UnsupportedFunctionCall, match=msg):
+        getattr(g, func)(foo=1)
+
+
+@td.skip_if_32bit
+@pytest.mark.parametrize("method", ["cummin", "cummax"])
+@pytest.mark.parametrize(
+    "dtype,val", [("UInt64", np.iinfo("uint64").max), ("Int64", 2**53 + 1)]
+)
+def test_nullable_int_not_cast_as_float(method, dtype, val):
+    data = [val, pd.NA]
+    df = DataFrame({"grp": [1, 1], "b": data}, dtype=dtype)
+    grouped = df.groupby("grp")
+
+    result = grouped.transform(method)
+    expected = DataFrame({"b": data}, dtype=dtype)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cython_api2(as_index):
+    # this takes the fast apply path
+
+    # cumsum (GH5614)
+    # GH 5755 - cumsum is a transformer and should ignore as_index
+    df = DataFrame([[1, 2, np.nan], [1, np.nan, 9], [3, 4, 9]], columns=["A", "B", "C"])
+    expected = DataFrame([[2, np.nan], [np.nan, 9], [4, 9]], columns=["B", "C"])
+    result = df.groupby("A", as_index=as_index).cumsum()
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_filters.py b/pandas/pandas/tests/groupby/test_filters.py
new file mode 100644
index 0000000000000..4fe3aac629513
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_filters.py
@@ -0,0 +1,608 @@
+from string import ascii_lowercase
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+def test_filter_series():
+    s = Series([1, 3, 20, 5, 22, 24, 7])
+    expected_odd = Series([1, 3, 5, 7], index=[0, 1, 3, 6])
+    expected_even = Series([20, 22, 24], index=[2, 4, 5])
+    grouper = s.apply(lambda x: x % 2)
+    grouped = s.groupby(grouper)
+    tm.assert_series_equal(grouped.filter(lambda x: x.mean() < 10), expected_odd)
+    tm.assert_series_equal(grouped.filter(lambda x: x.mean() > 10), expected_even)
+    # Test dropna=False.
+    tm.assert_series_equal(
+        grouped.filter(lambda x: x.mean() < 10, dropna=False),
+        expected_odd.reindex(s.index),
+    )
+    tm.assert_series_equal(
+        grouped.filter(lambda x: x.mean() > 10, dropna=False),
+        expected_even.reindex(s.index),
+    )
+
+
+def test_filter_single_column_df():
+    df = DataFrame([1, 3, 20, 5, 22, 24, 7])
+    expected_odd = DataFrame([1, 3, 5, 7], index=[0, 1, 3, 6])
+    expected_even = DataFrame([20, 22, 24], index=[2, 4, 5])
+    grouper = df[0].apply(lambda x: x % 2)
+    grouped = df.groupby(grouper)
+    tm.assert_frame_equal(grouped.filter(lambda x: x.mean() < 10), expected_odd)
+    tm.assert_frame_equal(grouped.filter(lambda x: x.mean() > 10), expected_even)
+    # Test dropna=False.
+    tm.assert_frame_equal(
+        grouped.filter(lambda x: x.mean() < 10, dropna=False),
+        expected_odd.reindex(df.index),
+    )
+    tm.assert_frame_equal(
+        grouped.filter(lambda x: x.mean() > 10, dropna=False),
+        expected_even.reindex(df.index),
+    )
+
+
+def test_filter_multi_column_df():
+    df = DataFrame({"A": [1, 12, 12, 1], "B": [1, 1, 1, 1]})
+    grouper = df["A"].apply(lambda x: x % 2)
+    grouped = df.groupby(grouper)
+    expected = DataFrame({"A": [12, 12], "B": [1, 1]}, index=[1, 2])
+    tm.assert_frame_equal(
+        grouped.filter(lambda x: x["A"].sum() - x["B"].sum() > 10), expected
+    )
+
+
+def test_filter_mixed_df():
+    df = DataFrame({"A": [1, 12, 12, 1], "B": "a b c d".split()})
+    grouper = df["A"].apply(lambda x: x % 2)
+    grouped = df.groupby(grouper)
+    expected = DataFrame({"A": [12, 12], "B": ["b", "c"]}, index=[1, 2])
+    tm.assert_frame_equal(grouped.filter(lambda x: x["A"].sum() > 10), expected)
+
+
+def test_filter_out_all_groups():
+    s = Series([1, 3, 20, 5, 22, 24, 7])
+    grouper = s.apply(lambda x: x % 2)
+    grouped = s.groupby(grouper)
+    tm.assert_series_equal(grouped.filter(lambda x: x.mean() > 1000), s[[]])
+    df = DataFrame({"A": [1, 12, 12, 1], "B": "a b c d".split()})
+    grouper = df["A"].apply(lambda x: x % 2)
+    grouped = df.groupby(grouper)
+    tm.assert_frame_equal(grouped.filter(lambda x: x["A"].sum() > 1000), df.loc[[]])
+
+
+def test_filter_out_no_groups():
+    s = Series([1, 3, 20, 5, 22, 24, 7])
+    grouper = s.apply(lambda x: x % 2)
+    grouped = s.groupby(grouper)
+    filtered = grouped.filter(lambda x: x.mean() > 0)
+    tm.assert_series_equal(filtered, s)
+
+
+def test_filter_out_no_groups_dataframe():
+    df = DataFrame({"A": [1, 12, 12, 1], "B": "a b c d".split()})
+    grouper = df["A"].apply(lambda x: x % 2)
+    grouped = df.groupby(grouper)
+    filtered = grouped.filter(lambda x: x["A"].mean() > 0)
+    tm.assert_frame_equal(filtered, df)
+
+
+def test_filter_out_all_groups_in_df():
+    # GH12768
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 2, 0]})
+    res = df.groupby("a")
+    res = res.filter(lambda x: x["b"].sum() > 5, dropna=False)
+    expected = DataFrame({"a": [np.nan] * 3, "b": [np.nan] * 3})
+    tm.assert_frame_equal(expected, res)
+
+
+def test_filter_out_all_groups_in_df_dropna_true():
+    # GH12768
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 2, 0]})
+    res = df.groupby("a")
+    res = res.filter(lambda x: x["b"].sum() > 5, dropna=True)
+    expected = DataFrame({"a": [], "b": []}, dtype="int64")
+    tm.assert_frame_equal(expected, res)
+
+
+def test_filter_condition_raises():
+    def raise_if_sum_is_zero(x):
+        if x.sum() == 0:
+            raise ValueError
+        return x.sum() > 0
+
+    s = Series([-1, 0, 1, 2])
+    grouper = s.apply(lambda x: x % 2)
+    grouped = s.groupby(grouper)
+    msg = "the filter must return a boolean result"
+    with pytest.raises(TypeError, match=msg):
+        grouped.filter(raise_if_sum_is_zero)
+
+
+def test_filter_bad_shapes():
+    df = DataFrame({"A": np.arange(8), "B": list("aabbbbcc"), "C": np.arange(8)})
+    s = df["B"]
+    g_df = df.groupby("B")
+    g_s = s.groupby(s)
+
+    f = lambda x: x
+    msg = "filter function returned a DataFrame, but expected a scalar bool"
+    with pytest.raises(TypeError, match=msg):
+        g_df.filter(f)
+    msg = "the filter must return a boolean result"
+    with pytest.raises(TypeError, match=msg):
+        g_s.filter(f)
+
+    f = lambda x: x == 1
+    msg = "filter function returned a DataFrame, but expected a scalar bool"
+    with pytest.raises(TypeError, match=msg):
+        g_df.filter(f)
+    msg = "the filter must return a boolean result"
+    with pytest.raises(TypeError, match=msg):
+        g_s.filter(f)
+
+    f = lambda x: np.outer(x, x)
+    msg = "can't multiply sequence by non-int of type 'str'"
+    with pytest.raises(TypeError, match=msg):
+        g_df.filter(f)
+    msg = "the filter must return a boolean result"
+    with pytest.raises(TypeError, match=msg):
+        g_s.filter(f)
+
+
+def test_filter_nan_is_false():
+    df = DataFrame({"A": np.arange(8), "B": list("aabbbbcc"), "C": np.arange(8)})
+    s = df["B"]
+    g_df = df.groupby(df["B"])
+    g_s = s.groupby(s)
+
+    f = lambda x: np.nan
+    tm.assert_frame_equal(g_df.filter(f), df.loc[[]])
+    tm.assert_series_equal(g_s.filter(f), s[[]])
+
+
+def test_filter_pdna_is_false():
+    # in particular, dont raise in filter trying to call bool(pd.NA)
+    df = DataFrame({"A": np.arange(8), "B": list("aabbbbcc"), "C": np.arange(8)})
+    ser = df["B"]
+    g_df = df.groupby(df["B"])
+    g_s = ser.groupby(ser)
+
+    func = lambda x: pd.NA
+    res = g_df.filter(func)
+    tm.assert_frame_equal(res, df.loc[[]])
+    res = g_s.filter(func)
+    tm.assert_series_equal(res, ser[[]])
+
+
+def test_filter_against_workaround_ints():
+    # Series of ints
+    s = Series(np.random.default_rng(2).integers(0, 100, 10))
+    grouper = s.apply(lambda x: np.round(x, -1))
+    grouped = s.groupby(grouper)
+    f = lambda x: x.mean() > 10
+
+    old_way = s[grouped.transform(f).astype("bool")]
+    new_way = grouped.filter(f)
+    tm.assert_series_equal(new_way.sort_values(), old_way.sort_values())
+
+
+def test_filter_against_workaround_floats():
+    # Series of floats
+    s = 100 * Series(np.random.default_rng(2).random(10))
+    grouper = s.apply(lambda x: np.round(x, -1))
+    grouped = s.groupby(grouper)
+    f = lambda x: x.mean() > 10
+    old_way = s[grouped.transform(f).astype("bool")]
+    new_way = grouped.filter(f)
+    tm.assert_series_equal(new_way.sort_values(), old_way.sort_values())
+
+
+def test_filter_against_workaround_dataframe():
+    # Set up DataFrame of ints, floats, strings.
+    letters = np.array(list(ascii_lowercase))
+    N = 10
+    random_letters = letters.take(
+        np.random.default_rng(2).integers(0, 26, N, dtype=int)
+    )
+    df = DataFrame(
+        {
+            "ints": Series(np.random.default_rng(2).integers(0, 10, N)),
+            "floats": N / 10 * Series(np.random.default_rng(2).random(N)),
+            "letters": Series(random_letters),
+        }
+    )
+
+    # Group by ints; filter on floats.
+    grouped = df.groupby("ints")
+    old_way = df[grouped.floats.transform(lambda x: x.mean() > N / 2).astype("bool")]
+    new_way = grouped.filter(lambda x: x["floats"].mean() > N / 2)
+    tm.assert_frame_equal(new_way, old_way)
+
+    # Group by floats (rounded); filter on strings.
+    grouper = df.floats.apply(lambda x: np.round(x, -1))
+    grouped = df.groupby(grouper)
+    old_way = df[grouped.letters.transform(lambda x: len(x) < N / 2).astype("bool")]
+    new_way = grouped.filter(lambda x: len(x.letters) < N / 2)
+    tm.assert_frame_equal(new_way, old_way)
+
+    # Group by strings; filter on ints.
+    grouped = df.groupby("letters")
+    old_way = df[grouped.ints.transform(lambda x: x.mean() > N / 2).astype("bool")]
+    new_way = grouped.filter(lambda x: x["ints"].mean() > N / 2)
+    tm.assert_frame_equal(new_way, old_way)
+
+
+def test_filter_using_len():
+    # GH 4447
+    df = DataFrame({"A": np.arange(8), "B": list("aabbbbcc"), "C": np.arange(8)})
+    grouped = df.groupby("B")
+    actual = grouped.filter(lambda x: len(x) > 2)
+    expected = DataFrame(
+        {"A": np.arange(2, 6), "B": list("bbbb"), "C": np.arange(2, 6)},
+        index=range(2, 6),
+    )
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped.filter(lambda x: len(x) > 4)
+    expected = df.loc[[]]
+    tm.assert_frame_equal(actual, expected)
+
+
+def test_filter_using_len_series():
+    # GH 4447
+    s = Series(list("aabbbbcc"), name="B")
+    grouped = s.groupby(s)
+    actual = grouped.filter(lambda x: len(x) > 2)
+    expected = Series(4 * ["b"], index=range(2, 6), name="B")
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped.filter(lambda x: len(x) > 4)
+    expected = s[[]]
+    tm.assert_series_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "index", [range(8), range(7, -1, -1), [0, 2, 1, 3, 4, 6, 5, 7]]
+)
+def test_filter_maintains_ordering(index):
+    # GH 4621
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    s = df["pid"]
+    grouped = df.groupby("tag")
+    actual = grouped.filter(lambda x: len(x) > 1)
+    expected = df.iloc[[1, 2, 4, 7]]
+    tm.assert_frame_equal(actual, expected)
+
+    grouped = s.groupby(df["tag"])
+    actual = grouped.filter(lambda x: len(x) > 1)
+    expected = s.iloc[[1, 2, 4, 7]]
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_multiple_timestamp():
+    # GH 10114
+    df = DataFrame(
+        {
+            "A": np.arange(5, dtype="int64"),
+            "B": ["foo", "bar", "foo", "bar", "bar"],
+            "C": Timestamp("20130101"),
+        }
+    )
+
+    grouped = df.groupby(["B", "C"])
+
+    result = grouped["A"].filter(lambda x: True)
+    tm.assert_series_equal(df["A"], result)
+
+    result = grouped["A"].transform(len)
+    expected = Series([2, 3, 2, 3, 3], name="A")
+    tm.assert_series_equal(result, expected)
+
+    result = grouped.filter(lambda x: True)
+    tm.assert_frame_equal(df, result)
+
+    result = grouped.transform("sum")
+    expected = DataFrame({"A": [2, 8, 2, 8, 8]})
+    tm.assert_frame_equal(result, expected)
+
+    result = grouped.transform(len)
+    expected = DataFrame({"A": [2, 3, 2, 3, 3]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_filter_and_transform_with_non_unique_int_index():
+    # GH4620
+    index = [1, 1, 1, 2, 1, 1, 0, 1]
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    grouped_df = df.groupby("tag")
+    ser = df["pid"]
+    grouped_ser = ser.groupby(df["tag"])
+    expected_indexes = [1, 2, 4, 7]
+
+    # Filter DataFrame
+    actual = grouped_df.filter(lambda x: len(x) > 1)
+    expected = df.iloc[expected_indexes]
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped_df.filter(lambda x: len(x) > 1, dropna=False)
+    # Cast to avoid upcast when setting nan below
+    expected = df.copy().astype("float64")
+    expected.iloc[[0, 3, 5, 6]] = np.nan
+    tm.assert_frame_equal(actual, expected)
+
+    # Filter Series
+    actual = grouped_ser.filter(lambda x: len(x) > 1)
+    expected = ser.take(expected_indexes)
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped_ser.filter(lambda x: len(x) > 1, dropna=False)
+    expected = Series([np.nan, 1, 1, np.nan, 2, np.nan, np.nan, 3], index, name="pid")
+    # ^ made manually because this can get confusing!
+    tm.assert_series_equal(actual, expected)
+
+    # Transform Series
+    actual = grouped_ser.transform(len)
+    expected = Series([1, 2, 2, 1, 2, 1, 1, 2], index, name="pid")
+    tm.assert_series_equal(actual, expected)
+
+    # Transform (a column from) DataFrameGroupBy
+    actual = grouped_df.pid.transform(len)
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_and_transform_with_multiple_non_unique_int_index():
+    # GH4620
+    index = [1, 1, 1, 2, 0, 0, 0, 1]
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    grouped_df = df.groupby("tag")
+    ser = df["pid"]
+    grouped_ser = ser.groupby(df["tag"])
+    expected_indexes = [1, 2, 4, 7]
+
+    # Filter DataFrame
+    actual = grouped_df.filter(lambda x: len(x) > 1)
+    expected = df.iloc[expected_indexes]
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped_df.filter(lambda x: len(x) > 1, dropna=False)
+    # Cast to avoid upcast when setting nan below
+    expected = df.copy().astype("float64")
+    expected.iloc[[0, 3, 5, 6]] = np.nan
+    tm.assert_frame_equal(actual, expected)
+
+    # Filter Series
+    actual = grouped_ser.filter(lambda x: len(x) > 1)
+    expected = ser.take(expected_indexes)
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped_ser.filter(lambda x: len(x) > 1, dropna=False)
+    expected = Series([np.nan, 1, 1, np.nan, 2, np.nan, np.nan, 3], index, name="pid")
+    # ^ made manually because this can get confusing!
+    tm.assert_series_equal(actual, expected)
+
+    # Transform Series
+    actual = grouped_ser.transform(len)
+    expected = Series([1, 2, 2, 1, 2, 1, 1, 2], index, name="pid")
+    tm.assert_series_equal(actual, expected)
+
+    # Transform (a column from) DataFrameGroupBy
+    actual = grouped_df.pid.transform(len)
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_and_transform_with_non_unique_float_index():
+    # GH4620
+    index = np.array([1, 1, 1, 2, 1, 1, 0, 1], dtype=float)
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    grouped_df = df.groupby("tag")
+    ser = df["pid"]
+    grouped_ser = ser.groupby(df["tag"])
+    expected_indexes = [1, 2, 4, 7]
+
+    # Filter DataFrame
+    actual = grouped_df.filter(lambda x: len(x) > 1)
+    expected = df.iloc[expected_indexes]
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped_df.filter(lambda x: len(x) > 1, dropna=False)
+    # Cast to avoid upcast when setting nan below
+    expected = df.copy().astype("float64")
+    expected.iloc[[0, 3, 5, 6]] = np.nan
+    tm.assert_frame_equal(actual, expected)
+
+    # Filter Series
+    actual = grouped_ser.filter(lambda x: len(x) > 1)
+    expected = ser.take(expected_indexes)
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped_ser.filter(lambda x: len(x) > 1, dropna=False)
+    expected = Series([np.nan, 1, 1, np.nan, 2, np.nan, np.nan, 3], index, name="pid")
+    # ^ made manually because this can get confusing!
+    tm.assert_series_equal(actual, expected)
+
+    # Transform Series
+    actual = grouped_ser.transform(len)
+    expected = Series([1, 2, 2, 1, 2, 1, 1, 2], index, name="pid")
+    tm.assert_series_equal(actual, expected)
+
+    # Transform (a column from) DataFrameGroupBy
+    actual = grouped_df.pid.transform(len)
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_and_transform_with_non_unique_timestamp_index():
+    # GH4620
+    t0 = Timestamp("2013-09-30 00:05:00")
+    t1 = Timestamp("2013-10-30 00:05:00")
+    t2 = Timestamp("2013-11-30 00:05:00")
+    index = [t1, t1, t1, t2, t1, t1, t0, t1]
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    grouped_df = df.groupby("tag")
+    ser = df["pid"]
+    grouped_ser = ser.groupby(df["tag"])
+    expected_indexes = [1, 2, 4, 7]
+
+    # Filter DataFrame
+    actual = grouped_df.filter(lambda x: len(x) > 1)
+    expected = df.iloc[expected_indexes]
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped_df.filter(lambda x: len(x) > 1, dropna=False)
+    # Cast to avoid upcast when setting nan below
+    expected = df.copy().astype("float64")
+    expected.iloc[[0, 3, 5, 6]] = np.nan
+    tm.assert_frame_equal(actual, expected)
+
+    # Filter Series
+    actual = grouped_ser.filter(lambda x: len(x) > 1)
+    expected = ser.take(expected_indexes)
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped_ser.filter(lambda x: len(x) > 1, dropna=False)
+    expected = Series([np.nan, 1, 1, np.nan, 2, np.nan, np.nan, 3], index, name="pid")
+    # ^ made manually because this can get confusing!
+    tm.assert_series_equal(actual, expected)
+
+    # Transform Series
+    actual = grouped_ser.transform(len)
+    expected = Series([1, 2, 2, 1, 2, 1, 1, 2], index, name="pid")
+    tm.assert_series_equal(actual, expected)
+
+    # Transform (a column from) DataFrameGroupBy
+    actual = grouped_df.pid.transform(len)
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_and_transform_with_non_unique_string_index():
+    # GH4620
+    index = list("bbbcbbab")
+    df = DataFrame(
+        {"pid": [1, 1, 1, 2, 2, 3, 3, 3], "tag": [23, 45, 62, 24, 45, 34, 25, 62]},
+        index=index,
+    )
+    grouped_df = df.groupby("tag")
+    ser = df["pid"]
+    grouped_ser = ser.groupby(df["tag"])
+    expected_indexes = [1, 2, 4, 7]
+
+    # Filter DataFrame
+    actual = grouped_df.filter(lambda x: len(x) > 1)
+    expected = df.iloc[expected_indexes]
+    tm.assert_frame_equal(actual, expected)
+
+    actual = grouped_df.filter(lambda x: len(x) > 1, dropna=False)
+    # Cast to avoid upcast when setting nan below
+    expected = df.copy().astype("float64")
+    expected.iloc[[0, 3, 5, 6]] = np.nan
+    tm.assert_frame_equal(actual, expected)
+
+    # Filter Series
+    actual = grouped_ser.filter(lambda x: len(x) > 1)
+    expected = ser.take(expected_indexes)
+    tm.assert_series_equal(actual, expected)
+
+    actual = grouped_ser.filter(lambda x: len(x) > 1, dropna=False)
+    expected = Series([np.nan, 1, 1, np.nan, 2, np.nan, np.nan, 3], index, name="pid")
+    # ^ made manually because this can get confusing!
+    tm.assert_series_equal(actual, expected)
+
+    # Transform Series
+    actual = grouped_ser.transform(len)
+    expected = Series([1, 2, 2, 1, 2, 1, 1, 2], index, name="pid")
+    tm.assert_series_equal(actual, expected)
+
+    # Transform (a column from) DataFrameGroupBy
+    actual = grouped_df.pid.transform(len)
+    tm.assert_series_equal(actual, expected)
+
+
+def test_filter_has_access_to_grouped_cols():
+    df = DataFrame([[1, 2], [1, 3], [5, 6]], columns=["A", "B"])
+    g = df.groupby("A")
+    # previously didn't have access to col A #????
+    filt = g.filter(lambda x: x["A"].sum() == 2)
+    tm.assert_frame_equal(filt, df.iloc[[0, 1]])
+
+
+def test_filter_enforces_scalarness():
+    df = DataFrame(
+        [
+            ["best", "a", "x"],
+            ["worst", "b", "y"],
+            ["best", "c", "x"],
+            ["best", "d", "y"],
+            ["worst", "d", "y"],
+            ["worst", "d", "y"],
+            ["best", "d", "z"],
+        ],
+        columns=["a", "b", "c"],
+    )
+    with pytest.raises(TypeError, match="filter function returned a.*"):
+        df.groupby("c").filter(lambda g: g["a"] == "best")
+
+
+def test_filter_non_bool_raises():
+    df = DataFrame(
+        [
+            ["best", "a", 1],
+            ["worst", "b", 1],
+            ["best", "c", 1],
+            ["best", "d", 1],
+            ["worst", "d", 1],
+            ["worst", "d", 1],
+            ["best", "d", 1],
+        ],
+        columns=["a", "b", "c"],
+    )
+    with pytest.raises(TypeError, match="filter function returned a.*"):
+        df.groupby("a").filter(lambda g: g.c.mean())
+
+
+def test_filter_dropna_with_empty_groups():
+    # GH 10780
+    data = Series(np.random.default_rng(2).random(9), index=np.repeat([1, 2, 3], 3))
+    grouped = data.groupby(level=0)
+    result_false = grouped.filter(lambda x: x.mean() > 1, dropna=False)
+    expected_false = Series([np.nan] * 9, index=np.repeat([1, 2, 3], 3))
+    tm.assert_series_equal(result_false, expected_false)
+
+    result_true = grouped.filter(lambda x: x.mean() > 1, dropna=True)
+    expected_true = Series(index=pd.Index([], dtype=int), dtype=np.float64)
+    tm.assert_series_equal(result_true, expected_true)
+
+
+def test_filter_consistent_result_before_after_agg_func():
+    # GH 17091
+    df = DataFrame({"data": range(6), "key": list("ABCABC")})
+    grouper = df.groupby("key")
+    result = grouper.filter(lambda x: True)
+    expected = DataFrame({"data": range(6), "key": list("ABCABC")})
+    tm.assert_frame_equal(result, expected)
+
+    grouper.sum()
+    result = grouper.filter(lambda x: True)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_groupby.py b/pandas/pandas/tests/groupby/test_groupby.py
new file mode 100644
index 0000000000000..e08ddf750b00c
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_groupby.py
@@ -0,0 +1,2980 @@
+from datetime import datetime
+import decimal
+from decimal import Decimal
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import SpecificationError
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Grouper,
+    Index,
+    Interval,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.arrays import BooleanArray
+import pandas.core.common as com
+
+pytestmark = pytest.mark.filterwarnings("ignore:Mean of empty slice:RuntimeWarning")
+
+
+def test_repr():
+    # GH18203
+    result = repr(Grouper(key="A", level="B"))
+    expected = "Grouper(key='A', level='B', sort=False, dropna=True)"
+    assert result == expected
+
+
+def test_groupby_nonobject_dtype(multiindex_dataframe_random_data):
+    key = multiindex_dataframe_random_data.index.codes[0]
+    grouped = multiindex_dataframe_random_data.groupby(key)
+    result = grouped.sum()
+
+    expected = multiindex_dataframe_random_data.groupby(key.astype("O")).sum()
+    assert result.index.dtype == np.int8
+    assert expected.index.dtype == np.int64
+    tm.assert_frame_equal(result, expected, check_index_type=False)
+
+
+def test_groupby_nonobject_dtype_mixed():
+    # GH 3911, mixed frame non-conversion
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.array(np.random.default_rng(2).standard_normal(8), dtype="float32"),
+        }
+    )
+    df["value"] = range(len(df))
+
+    def max_value(group):
+        return group.loc[group["value"].idxmax()]
+
+    applied = df.groupby("A").apply(max_value)
+    result = applied.dtypes
+    expected = df.drop(columns="A").dtypes
+    tm.assert_series_equal(result, expected)
+
+
+def test_pass_args_kwargs(ts):
+    def f(x, q=None, axis=0):
+        return np.percentile(x, q, axis=axis)
+
+    g = lambda x: np.percentile(x, 80, axis=0)
+
+    # Series
+    ts_grouped = ts.groupby(lambda x: x.month)
+    agg_result = ts_grouped.agg(np.percentile, 80, axis=0)
+    apply_result = ts_grouped.apply(np.percentile, 80, axis=0)
+    trans_result = ts_grouped.transform(np.percentile, 80, axis=0)
+
+    agg_expected = ts_grouped.quantile(0.8)
+    trans_expected = ts_grouped.transform(g)
+
+    tm.assert_series_equal(apply_result, agg_expected)
+    tm.assert_series_equal(agg_result, agg_expected)
+    tm.assert_series_equal(trans_result, trans_expected)
+
+    agg_result = ts_grouped.agg(f, q=80)
+    apply_result = ts_grouped.apply(f, q=80)
+    trans_result = ts_grouped.transform(f, q=80)
+    tm.assert_series_equal(agg_result, agg_expected)
+    tm.assert_series_equal(apply_result, agg_expected)
+    tm.assert_series_equal(trans_result, trans_expected)
+
+
+def test_pass_args_kwargs_dataframe(tsframe, as_index):
+    def f(x, q=None, axis=0):
+        return np.percentile(x, q, axis=axis)
+
+    df_grouped = tsframe.groupby(lambda x: x.month, as_index=as_index)
+    agg_result = df_grouped.agg(np.percentile, 80, axis=0)
+    apply_result = df_grouped.apply(DataFrame.quantile, 0.8)
+    expected = df_grouped.quantile(0.8)
+    tm.assert_frame_equal(apply_result, expected, check_names=False)
+    tm.assert_frame_equal(agg_result, expected)
+
+    apply_result = df_grouped.apply(DataFrame.quantile, [0.4, 0.8])
+    expected_seq = df_grouped.quantile([0.4, 0.8])
+    if not as_index:
+        # apply treats the op as a transform; .quantile knows it's a reduction
+        apply_result.index = range(4)
+        apply_result.insert(loc=0, column="level_0", value=[1, 1, 2, 2])
+        apply_result.insert(loc=1, column="level_1", value=[0.4, 0.8, 0.4, 0.8])
+    tm.assert_frame_equal(apply_result, expected_seq, check_names=False)
+
+    agg_result = df_grouped.agg(f, q=80)
+    apply_result = df_grouped.apply(DataFrame.quantile, q=0.8)
+    tm.assert_frame_equal(agg_result, expected)
+    tm.assert_frame_equal(apply_result, expected, check_names=False)
+
+
+def test_len():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    grouped = df.groupby([lambda x: x.year, lambda x: x.month, lambda x: x.day])
+    assert len(grouped) == len(df)
+
+    grouped = df.groupby([lambda x: x.year, lambda x: x.month])
+    expected = len({(x.year, x.month) for x in df.index})
+    assert len(grouped) == expected
+
+
+def test_len_nan_group():
+    # issue 11016
+    df = DataFrame({"a": [np.nan] * 3, "b": [1, 2, 3]})
+    assert len(df.groupby("a")) == 0
+    assert len(df.groupby("b")) == 3
+    assert len(df.groupby(["a", "b"])) == 0
+
+
+def test_groupby_timedelta_median():
+    # issue 57926
+    expected = Series(data=Timedelta("1D"), index=["foo"])
+    df = DataFrame({"label": ["foo", "foo"], "timedelta": [pd.NaT, Timedelta("1D")]})
+    gb = df.groupby("label")["timedelta"]
+    actual = gb.median()
+    tm.assert_series_equal(actual, expected, check_names=False)
+
+
+@pytest.mark.parametrize("keys", [["a"], ["a", "b"]])
+def test_len_categorical(dropna, observed, keys):
+    # GH#57595
+    df = DataFrame(
+        {
+            "a": Categorical([1, 1, 2, np.nan], categories=[1, 2, 3]),
+            "b": Categorical([1, 1, 2, np.nan], categories=[1, 2, 3]),
+            "c": 1,
+        }
+    )
+    gb = df.groupby(keys, observed=observed, dropna=dropna)
+    result = len(gb)
+    if observed and dropna:
+        expected = 2
+    elif observed and not dropna:
+        expected = 3
+    elif len(keys) == 1:
+        expected = 3 if dropna else 4
+    else:
+        expected = 9 if dropna else 16
+    assert result == expected, f"{result} vs {expected}"
+
+
+def test_basic_regression():
+    # regression
+    result = Series([1.0 * x for x in list(range(1, 10)) * 10])
+
+    data = np.random.default_rng(2).random(1100) * 10.0
+    groupings = Series(data)
+
+    grouped = result.groupby(groupings)
+    grouped.mean()
+
+
+def test_indices_concatenation_order():
+    # GH 2808
+
+    def f1(x):
+        y = x[(x.b % 2) == 1] ** 2
+        if y.empty:
+            multiindex = MultiIndex(levels=[[]] * 2, codes=[[]] * 2, names=["b", "c"])
+            res = DataFrame(columns=["a"], index=multiindex)
+            return res
+        else:
+            y = y.set_index(["b", "c"])
+            return y
+
+    def f2(x):
+        y = x[(x.b % 2) == 1] ** 2
+        if y.empty:
+            return DataFrame()
+        else:
+            y = y.set_index(["b", "c"])
+            return y
+
+    def f3(x):
+        y = x[(x.b % 2) == 1] ** 2
+        if y.empty:
+            multiindex = MultiIndex(
+                levels=[[]] * 2, codes=[[]] * 2, names=["foo", "bar"]
+            )
+            res = DataFrame(columns=["a", "b"], index=multiindex)
+            return res
+        else:
+            return y
+
+    df = DataFrame({"a": [1, 2, 2, 2], "b": range(4), "c": range(5, 9)})
+
+    df2 = DataFrame({"a": [3, 2, 2, 2], "b": range(4), "c": range(5, 9)})
+
+    # correct result
+    result1 = df.groupby("a").apply(f1)
+    result2 = df2.groupby("a").apply(f1)
+    tm.assert_frame_equal(result1, result2)
+
+    # should fail (not the same number of levels)
+    msg = "Cannot concat indices that do not have the same number of levels"
+    with pytest.raises(AssertionError, match=msg):
+        df.groupby("a").apply(f2)
+    with pytest.raises(AssertionError, match=msg):
+        df2.groupby("a").apply(f2)
+
+    # should fail (incorrect shape)
+    with pytest.raises(AssertionError, match=msg):
+        df.groupby("a").apply(f3)
+    with pytest.raises(AssertionError, match=msg):
+        df2.groupby("a").apply(f3)
+
+
+def test_attr_wrapper(ts):
+    grouped = ts.groupby(lambda x: x.weekday())
+
+    result = grouped.std()
+    expected = grouped.agg(lambda x: np.std(x, ddof=1))
+    tm.assert_series_equal(result, expected)
+
+    # this is pretty cool
+    result = grouped.describe()
+    expected = {name: gp.describe() for name, gp in grouped}
+    expected = DataFrame(expected).T
+    tm.assert_frame_equal(result, expected)
+
+    # get attribute
+    result = grouped.dtype
+    expected = grouped.agg(lambda x: x.dtype)
+    tm.assert_series_equal(result, expected)
+
+    # make sure raises error
+    msg = "'SeriesGroupBy' object has no attribute 'foo'"
+    with pytest.raises(AttributeError, match=msg):
+        grouped.foo
+
+
+def test_frame_groupby(tsframe):
+    grouped = tsframe.groupby(lambda x: x.weekday())
+
+    # aggregate
+    aggregated = grouped.aggregate("mean")
+    assert len(aggregated) == 5
+    assert len(aggregated.columns) == 4
+
+    # by string
+    tscopy = tsframe.copy()
+    tscopy["weekday"] = [x.weekday() for x in tscopy.index]
+    stragged = tscopy.groupby("weekday").aggregate("mean")
+    tm.assert_frame_equal(stragged, aggregated, check_names=False)
+
+    # transform
+    grouped = tsframe.head(30).groupby(lambda x: x.weekday())
+    transformed = grouped.transform(lambda x: x - x.mean())
+    assert len(transformed) == 30
+    assert len(transformed.columns) == 4
+
+    # transform propagate
+    transformed = grouped.transform(lambda x: x.mean())
+    for name, group in grouped:
+        mean = group.mean()
+        for idx in group.index:
+            tm.assert_series_equal(transformed.xs(idx), mean, check_names=False)
+
+    # iterate
+    for weekday, group in grouped:
+        assert group.index[0].weekday() == weekday
+
+    # groups / group_indices
+    groups = grouped.groups
+    indices = grouped.indices
+
+    for k, v in groups.items():
+        samething = tsframe.index.take(indices[k])
+        assert (samething == v).all()
+
+
+def test_frame_set_name_single(df):
+    grouped = df.groupby("A")
+
+    result = grouped.mean(numeric_only=True)
+    assert result.index.name == "A"
+
+    result = df.groupby("A", as_index=False).mean(numeric_only=True)
+    assert result.index.name != "A"
+
+    result = grouped[["C", "D"]].agg("mean")
+    assert result.index.name == "A"
+
+    result = grouped.agg({"C": "mean", "D": "std"})
+    assert result.index.name == "A"
+
+    result = grouped["C"].mean()
+    assert result.index.name == "A"
+    result = grouped["C"].agg("mean")
+    assert result.index.name == "A"
+    result = grouped["C"].agg(["mean", "std"])
+    assert result.index.name == "A"
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        grouped["C"].agg({"foo": "mean", "bar": "std"})
+
+
+def test_multi_func(df):
+    col1 = df["A"]
+    col2 = df["B"]
+
+    grouped = df.groupby([col1.get, col2.get])
+    agged = grouped.mean(numeric_only=True)
+    expected = df.groupby(["A", "B"]).mean()
+
+    # TODO groupby get drops names
+    tm.assert_frame_equal(
+        agged.loc[:, ["C", "D"]], expected.loc[:, ["C", "D"]], check_names=False
+    )
+
+    # some "groups" with no data
+    df = DataFrame(
+        {
+            "v1": np.random.default_rng(2).standard_normal(6),
+            "v2": np.random.default_rng(2).standard_normal(6),
+            "k1": np.array(["b", "b", "b", "a", "a", "a"]),
+            "k2": np.array(["1", "1", "1", "2", "2", "2"]),
+        },
+        index=["one", "two", "three", "four", "five", "six"],
+    )
+    # only verify that it works for now
+    grouped = df.groupby(["k1", "k2"])
+    grouped.agg("sum")
+
+
+def test_multi_key_multiple_functions(df):
+    grouped = df.groupby(["A", "B"])["C"]
+
+    agged = grouped.agg(["mean", "std"])
+    expected = DataFrame({"mean": grouped.agg("mean"), "std": grouped.agg("std")})
+    tm.assert_frame_equal(agged, expected)
+
+
+def test_frame_multi_key_function_list():
+    data = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    grouped = data.groupby(["A", "B"])
+    funcs = ["mean", "std"]
+    agged = grouped.agg(funcs)
+    expected = pd.concat(
+        [grouped["D"].agg(funcs), grouped["E"].agg(funcs), grouped["F"].agg(funcs)],
+        keys=["D", "E", "F"],
+        axis=1,
+    )
+    assert isinstance(agged.index, MultiIndex)
+    assert isinstance(expected.index, MultiIndex)
+    tm.assert_frame_equal(agged, expected)
+
+
+def test_frame_multi_key_function_list_partial_failure(using_infer_string):
+    data = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    grouped = data.groupby(["A", "B"])
+    funcs = ["mean", "std"]
+    msg = re.escape("agg function failed [how->mean,dtype->")
+    if using_infer_string:
+        msg = "dtype 'str' does not support operation 'mean'"
+    with pytest.raises(TypeError, match=msg):
+        grouped.agg(funcs)
+
+
+@pytest.mark.parametrize("op", [lambda x: x.sum(), lambda x: x.mean()])
+def test_groupby_multiple_columns(df, op):
+    data = df
+    grouped = data.groupby(["A", "B"])
+
+    result1 = op(grouped)
+
+    keys = []
+    values = []
+    for n1, gp1 in data.groupby("A"):
+        for n2, gp2 in gp1.groupby("B"):
+            keys.append((n1, n2))
+            values.append(op(gp2.loc[:, ["C", "D"]]))
+
+    mi = MultiIndex.from_tuples(keys, names=["A", "B"])
+    expected = pd.concat(values, axis=1).T
+    expected.index = mi
+
+    # a little bit crude
+    for col in ["C", "D"]:
+        result_col = op(grouped[col])
+        pivoted = result1[col]
+        exp = expected[col]
+        tm.assert_series_equal(result_col, exp)
+        tm.assert_series_equal(pivoted, exp)
+
+    # test single series works the same
+    result = data["C"].groupby([data["A"], data["B"]]).mean()
+    expected = data.groupby(["A", "B"]).mean()["C"]
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_as_index_select_column():
+    # GH 5764
+    df = DataFrame([[1, 2], [1, 4], [5, 6]], columns=["A", "B"])
+    result = df.groupby("A", as_index=False)["B"].get_group(1)
+    expected = Series([2, 4], name="B")
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby("A", as_index=False, group_keys=True)["B"].apply(
+        lambda x: x.cumsum()
+    )
+    expected = Series([2, 6, 6], name="B", index=range(3))
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_as_index_select_column_sum_empty_df():
+    # GH 35246
+    df = DataFrame(columns=Index(["A", "B", "C"], name="alpha"))
+    left = df.groupby(by="A", as_index=False)["B"].sum(numeric_only=False)
+
+    expected = DataFrame(columns=df.columns[:2], index=range(0))
+    # GH#50744 - Columns after selection shouldn't retain names
+    expected.columns.names = [None]
+    tm.assert_frame_equal(left, expected)
+
+
+def test_ops_not_as_index(reduction_func):
+    # GH 10355, 21090
+    # Using as_index=False should not modify grouped column
+
+    if reduction_func in ("corrwith", "nth", "ngroup"):
+        pytest.skip(f"GH 5755: Test not applicable for {reduction_func}")
+
+    df = DataFrame(
+        np.random.default_rng(2).integers(0, 5, size=(100, 2)), columns=["a", "b"]
+    )
+    expected = getattr(df.groupby("a"), reduction_func)()
+    if reduction_func == "size":
+        expected = expected.rename("size")
+    expected = expected.reset_index()
+
+    if reduction_func != "size":
+        # 32 bit compat -> groupby preserves dtype whereas reset_index casts to int64
+        expected["a"] = expected["a"].astype(df["a"].dtype)
+
+    g = df.groupby("a", as_index=False)
+
+    result = getattr(g, reduction_func)()
+    tm.assert_frame_equal(result, expected)
+
+    result = g.agg(reduction_func)
+    tm.assert_frame_equal(result, expected)
+
+    result = getattr(g["b"], reduction_func)()
+    tm.assert_frame_equal(result, expected)
+
+    result = g["b"].agg(reduction_func)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_as_index_series_return_frame(df):
+    grouped = df.groupby("A", as_index=False)
+    grouped2 = df.groupby(["A", "B"], as_index=False)
+
+    result = grouped["C"].agg("sum")
+    expected = grouped.agg("sum").loc[:, ["A", "C"]]
+    assert isinstance(result, DataFrame)
+    tm.assert_frame_equal(result, expected)
+
+    result2 = grouped2["C"].agg("sum")
+    expected2 = grouped2.agg("sum").loc[:, ["A", "B", "C"]]
+    assert isinstance(result2, DataFrame)
+    tm.assert_frame_equal(result2, expected2)
+
+    result = grouped["C"].sum()
+    expected = grouped.sum().loc[:, ["A", "C"]]
+    assert isinstance(result, DataFrame)
+    tm.assert_frame_equal(result, expected)
+
+    result2 = grouped2["C"].sum()
+    expected2 = grouped2.sum().loc[:, ["A", "B", "C"]]
+    assert isinstance(result2, DataFrame)
+    tm.assert_frame_equal(result2, expected2)
+
+
+def test_as_index_series_column_slice_raises(df):
+    # GH15072
+    grouped = df.groupby("A", as_index=False)
+    msg = r"Column\(s\) C already selected"
+
+    with pytest.raises(IndexError, match=msg):
+        grouped["C"].__getitem__("D")
+
+
+def test_groupby_as_index_cython(df):
+    data = df
+
+    # single-key
+    grouped = data.groupby("A", as_index=False)
+    result = grouped.mean(numeric_only=True)
+    expected = data.groupby(["A"]).mean(numeric_only=True)
+    expected.insert(0, "A", expected.index)
+    expected.index = RangeIndex(len(expected))
+    tm.assert_frame_equal(result, expected)
+
+    # multi-key
+    grouped = data.groupby(["A", "B"], as_index=False)
+    result = grouped.mean()
+    expected = data.groupby(["A", "B"]).mean()
+
+    arrays = list(zip(*expected.index.values))
+    expected.insert(0, "A", arrays[0])
+    expected.insert(1, "B", arrays[1])
+    expected.index = RangeIndex(len(expected))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_as_index_series_scalar(df):
+    grouped = df.groupby(["A", "B"], as_index=False)
+
+    # GH #421
+
+    result = grouped["C"].agg(len)
+    expected = grouped.agg(len).loc[:, ["A", "B", "C"]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_multiple_key():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    grouped = df.groupby([lambda x: x.year, lambda x: x.month, lambda x: x.day])
+    agged = grouped.sum()
+    tm.assert_almost_equal(df.values, agged.values)
+
+
+def test_groupby_multi_corner(df):
+    # test that having an all-NA column doesn't mess you up
+    df = df.copy()
+    df["bad"] = np.nan
+    agged = df.groupby(["A", "B"]).mean()
+
+    expected = df.groupby(["A", "B"]).mean()
+    expected["bad"] = np.nan
+
+    tm.assert_frame_equal(agged, expected)
+
+
+def test_raises_on_nuisance(df, using_infer_string):
+    grouped = df.groupby("A")
+    msg = re.escape("agg function failed [how->mean,dtype->")
+    if using_infer_string:
+        msg = "dtype 'str' does not support operation 'mean'"
+    with pytest.raises(TypeError, match=msg):
+        grouped.agg("mean")
+    with pytest.raises(TypeError, match=msg):
+        grouped.mean()
+
+    df = df.loc[:, ["A", "C", "D"]]
+    df["E"] = datetime.now()
+    grouped = df.groupby("A")
+    msg = "datetime64 type does not support operation 'sum'"
+    with pytest.raises(TypeError, match=msg):
+        grouped.agg("sum")
+    with pytest.raises(TypeError, match=msg):
+        grouped.sum()
+
+
+@pytest.mark.parametrize(
+    "agg_function",
+    ["max", "min"],
+)
+def test_keep_nuisance_agg(df, agg_function):
+    # GH 38815
+    grouped = df.groupby("A")
+    result = getattr(grouped, agg_function)()
+    expected = result.copy()
+    expected.loc["bar", "B"] = getattr(df.loc[df["A"] == "bar", "B"], agg_function)()
+    expected.loc["foo", "B"] = getattr(df.loc[df["A"] == "foo", "B"], agg_function)()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg_function",
+    ["sum", "mean", "prod", "std", "var", "sem", "median"],
+)
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_omit_nuisance_agg(df, agg_function, numeric_only, using_infer_string):
+    # GH 38774, GH 38815
+    grouped = df.groupby("A")
+
+    no_drop_nuisance = ("var", "std", "sem", "mean", "prod", "median")
+    if agg_function in no_drop_nuisance and not numeric_only:
+        # Added numeric_only as part of GH#46560; these do not drop nuisance
+        # columns when numeric_only is False
+        if using_infer_string:
+            msg = f"dtype 'str' does not support operation '{agg_function}'"
+            klass = TypeError
+        elif agg_function in ("std", "sem"):
+            klass = ValueError
+            msg = "could not convert string to float: 'one'"
+        else:
+            klass = TypeError
+            msg = re.escape(f"agg function failed [how->{agg_function},dtype->")
+        with pytest.raises(klass, match=msg):
+            getattr(grouped, agg_function)(numeric_only=numeric_only)
+    else:
+        result = getattr(grouped, agg_function)(numeric_only=numeric_only)
+        if not numeric_only and agg_function == "sum":
+            # sum is successful on column B
+            columns = ["A", "B", "C", "D"]
+        else:
+            columns = ["A", "C", "D"]
+        expected = getattr(df.loc[:, columns].groupby("A"), agg_function)(
+            numeric_only=numeric_only
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_raise_on_nuisance_python_single(df, using_infer_string):
+    # GH 38815
+    grouped = df.groupby("A")
+
+    err = ValueError
+    msg = "could not convert"
+    if using_infer_string:
+        err = TypeError
+        msg = "dtype 'str' does not support operation 'skew'"
+    with pytest.raises(err, match=msg):
+        grouped.skew()
+
+
+def test_raise_on_nuisance_python_multiple(three_group, using_infer_string):
+    grouped = three_group.groupby(["A", "B"])
+    msg = re.escape("agg function failed [how->mean,dtype->")
+    if using_infer_string:
+        msg = "dtype 'str' does not support operation 'mean'"
+    with pytest.raises(TypeError, match=msg):
+        grouped.agg("mean")
+    with pytest.raises(TypeError, match=msg):
+        grouped.mean()
+
+
+def test_empty_groups_corner(multiindex_dataframe_random_data):
+    # handle empty groups
+    df = DataFrame(
+        {
+            "k1": np.array(["b", "b", "b", "a", "a", "a"]),
+            "k2": np.array(["1", "1", "1", "2", "2", "2"]),
+            "k3": ["foo", "bar"] * 3,
+            "v1": np.random.default_rng(2).standard_normal(6),
+            "v2": np.random.default_rng(2).standard_normal(6),
+        }
+    )
+
+    grouped = df.groupby(["k1", "k2"])
+    result = grouped[["v1", "v2"]].agg("mean")
+    expected = grouped.mean(numeric_only=True)
+    tm.assert_frame_equal(result, expected)
+
+    grouped = multiindex_dataframe_random_data[3:5].groupby(level=0)
+    agged = grouped.apply(lambda x: x.mean())
+    agged_A = grouped["A"].apply("mean")
+    tm.assert_series_equal(agged["A"], agged_A)
+    assert agged.index.name == "first"
+
+
+def test_nonsense_func():
+    df = DataFrame([0])
+    msg = r"unsupported operand type\(s\) for \+: 'int' and 'str'"
+    with pytest.raises(TypeError, match=msg):
+        df.groupby(lambda x: x + "foo")
+
+
+def test_wrap_aggregated_output_multindex(
+    multiindex_dataframe_random_data, using_infer_string
+):
+    df = multiindex_dataframe_random_data.T
+    df["baz", "two"] = "peekaboo"
+
+    keys = [np.array([0, 0, 1]), np.array([0, 0, 1])]
+    msg = re.escape("agg function failed [how->mean,dtype->")
+    if using_infer_string:
+        msg = "dtype 'str' does not support operation 'mean'"
+    with pytest.raises(TypeError, match=msg):
+        df.groupby(keys).agg("mean")
+    agged = df.drop(columns=("baz", "two")).groupby(keys).agg("mean")
+    assert isinstance(agged.columns, MultiIndex)
+
+    def aggfun(ser):
+        if ser.name == ("foo", "one"):
+            raise TypeError("Test error message")
+        return ser.sum()
+
+    with pytest.raises(TypeError, match="Test error message"):
+        df.groupby(keys).aggregate(aggfun)
+
+
+def test_groupby_level_apply(multiindex_dataframe_random_data):
+    result = multiindex_dataframe_random_data.groupby(level=0).count()
+    assert result.index.name == "first"
+    result = multiindex_dataframe_random_data.groupby(level=1).count()
+    assert result.index.name == "second"
+
+    result = multiindex_dataframe_random_data["A"].groupby(level=0).count()
+    assert result.index.name == "first"
+
+
+def test_groupby_level_mapper(multiindex_dataframe_random_data):
+    deleveled = multiindex_dataframe_random_data.reset_index()
+
+    mapper0 = {"foo": 0, "bar": 0, "baz": 1, "qux": 1}
+    mapper1 = {"one": 0, "two": 0, "three": 1}
+
+    result0 = multiindex_dataframe_random_data.groupby(mapper0, level=0).sum()
+    result1 = multiindex_dataframe_random_data.groupby(mapper1, level=1).sum()
+
+    mapped_level0 = np.array(
+        [mapper0.get(x) for x in deleveled["first"]], dtype=np.int64
+    )
+    mapped_level1 = np.array(
+        [mapper1.get(x) for x in deleveled["second"]], dtype=np.int64
+    )
+    expected0 = multiindex_dataframe_random_data.groupby(mapped_level0).sum()
+    expected1 = multiindex_dataframe_random_data.groupby(mapped_level1).sum()
+    expected0.index.name, expected1.index.name = "first", "second"
+
+    tm.assert_frame_equal(result0, expected0)
+    tm.assert_frame_equal(result1, expected1)
+
+
+def test_groupby_level_nonmulti():
+    # GH 1313, GH 13901
+    s = Series([1, 2, 3, 10, 4, 5, 20, 6], Index([1, 2, 3, 1, 4, 5, 2, 6], name="foo"))
+    expected = Series([11, 22, 3, 4, 5, 6], Index(list(range(1, 7)), name="foo"))
+
+    result = s.groupby(level=0).sum()
+    tm.assert_series_equal(result, expected)
+    result = s.groupby(level=[0]).sum()
+    tm.assert_series_equal(result, expected)
+    result = s.groupby(level=-1).sum()
+    tm.assert_series_equal(result, expected)
+    result = s.groupby(level=[-1]).sum()
+    tm.assert_series_equal(result, expected)
+
+    msg = "level > 0 or level < -1 only valid with MultiIndex"
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=1)
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=-2)
+    msg = "No group keys passed!"
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=[])
+    msg = "multiple levels only valid with MultiIndex"
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=[0, 0])
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=[0, 1])
+    msg = "level > 0 or level < -1 only valid with MultiIndex"
+    with pytest.raises(ValueError, match=msg):
+        s.groupby(level=[1])
+
+
+def test_groupby_complex():
+    # GH 12902
+    a = Series(data=np.arange(4) * (1 + 2j), index=[0, 0, 1, 1])
+    expected = Series((1 + 2j, 5 + 10j), index=Index([0, 1]))
+
+    result = a.groupby(level=0).sum()
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_complex_mean():
+    # GH 26475
+    df = DataFrame(
+        [
+            {"a": 2, "b": 1 + 2j},
+            {"a": 1, "b": 1 + 1j},
+            {"a": 1, "b": 1 + 2j},
+        ]
+    )
+    result = df.groupby("b").mean()
+    expected = DataFrame(
+        [[1.0], [1.5]],
+        index=Index([(1 + 1j), (1 + 2j)], name="b"),
+        columns=Index(["a"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_complex_numbers():
+    # GH 17927
+    df = DataFrame(
+        [
+            {"a": 1, "b": 1 + 1j},
+            {"a": 1, "b": 1 + 2j},
+            {"a": 4, "b": 1},
+        ]
+    )
+    expected = DataFrame(
+        np.array([1, 1, 1], dtype=np.int64),
+        index=Index([(1 + 1j), (1 + 2j), (1 + 0j)], name="b"),
+        columns=Index(["a"]),
+    )
+    result = df.groupby("b", sort=False).count()
+    tm.assert_frame_equal(result, expected)
+
+    # Sorted by the magnitude of the complex numbers
+    expected.index = Index([(1 + 0j), (1 + 1j), (1 + 2j)], name="b")
+    result = df.groupby("b", sort=True).count()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_series_indexed_differently():
+    s1 = Series(
+        [5.0, -9.0, 4.0, 100.0, -5.0, 55.0, 6.7],
+        index=Index(["a", "b", "c", "d", "e", "f", "g"]),
+    )
+    s2 = Series(
+        [1.0, 1.0, 4.0, 5.0, 5.0, 7.0], index=Index(["a", "b", "d", "f", "g", "h"])
+    )
+
+    grouped = s1.groupby(s2)
+    agged = grouped.mean()
+    exp = s1.groupby(s2.reindex(s1.index).get).mean()
+    tm.assert_series_equal(agged, exp)
+
+
+def test_groupby_with_hier_columns():
+    tuples = list(
+        zip(
+            *[
+                ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+                ["one", "two", "one", "two", "one", "two", "one", "two"],
+            ]
+        )
+    )
+    index = MultiIndex.from_tuples(tuples)
+    columns = MultiIndex.from_tuples(
+        [("A", "cat"), ("B", "dog"), ("B", "cat"), ("A", "dog")]
+    )
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((8, 4)), index=index, columns=columns
+    )
+
+    result = df.groupby(level=0).mean()
+    tm.assert_index_equal(result.columns, columns)
+
+    result = df.groupby(level=0).agg("mean")
+    tm.assert_index_equal(result.columns, columns)
+
+    result = df.groupby(level=0).apply(lambda x: x.mean())
+    tm.assert_index_equal(result.columns, columns)
+
+    # add a nuisance column
+    sorted_columns, _ = columns.sortlevel(0)
+    df["A", "foo"] = "bar"
+    result = df.groupby(level=0).mean(numeric_only=True)
+    tm.assert_index_equal(result.columns, df.columns[:-1])
+
+
+def test_grouping_ndarray(df):
+    grouped = df.groupby(df["A"].values)
+    grouped2 = df.groupby(df["A"].rename(None))
+
+    result = grouped.sum()
+    expected = grouped2.sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_wrong_multi_labels():
+    index = Index([0, 1, 2, 3, 4], name="index")
+    data = DataFrame(
+        {
+            "foo": ["foo1", "foo1", "foo2", "foo1", "foo3"],
+            "bar": ["bar1", "bar2", "bar2", "bar1", "bar1"],
+            "baz": ["baz1", "baz1", "baz1", "baz2", "baz2"],
+            "spam": ["spam2", "spam3", "spam2", "spam1", "spam1"],
+            "data": [20, 30, 40, 50, 60],
+        },
+        index=index,
+    )
+
+    grouped = data.groupby(["foo", "bar", "baz", "spam"])
+
+    result = grouped.agg("mean")
+    expected = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_series_with_name(df):
+    result = df.groupby(df["A"]).mean(numeric_only=True)
+    result2 = df.groupby(df["A"], as_index=False).mean(numeric_only=True)
+    assert result.index.name == "A"
+    assert "A" in result2
+
+    result = df.groupby([df["A"], df["B"]]).mean()
+    result2 = df.groupby([df["A"], df["B"]], as_index=False).mean()
+    assert result.index.names == ("A", "B")
+    assert "A" in result2
+    assert "B" in result2
+
+
+def test_seriesgroupby_name_attr(df):
+    # GH 6265
+    result = df.groupby("A")["C"]
+    assert result.count().name == "C"
+    assert result.mean().name == "C"
+
+    testFunc = lambda x: np.sum(x) * 2
+    assert result.agg(testFunc).name == "C"
+
+
+def test_consistency_name():
+    # GH 12363
+
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "two", "two", "two", "one", "two"],
+            "C": np.random.default_rng(2).standard_normal(8) + 1.0,
+            "D": np.arange(8),
+        }
+    )
+
+    expected = df.groupby(["A"]).B.count()
+    result = df.B.groupby(df.A).count()
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_name_propagation(df):
+    # GH 6124
+    def summarize(df, name=None):
+        return Series({"count": 1, "mean": 2, "omissions": 3}, name=name)
+
+    def summarize_random_name(df):
+        # Provide a different name for each Series.  In this case, groupby
+        # should not attempt to propagate the Series name since they are
+        # inconsistent.
+        return Series({"count": 1, "mean": 2, "omissions": 3}, name=df.iloc[0]["C"])
+
+    metrics = df.groupby("A").apply(summarize)
+    assert metrics.columns.name is None
+    metrics = df.groupby("A").apply(summarize, "metrics")
+    assert metrics.columns.name == "metrics"
+    metrics = df.groupby("A").apply(summarize_random_name)
+    assert metrics.columns.name is None
+
+
+def test_groupby_nonstring_columns():
+    df = DataFrame([np.arange(10) for x in range(10)])
+    grouped = df.groupby(0)
+    result = grouped.mean()
+    expected = df.groupby(df[0]).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_mixed_type_columns():
+    # GH 13432, unorderable types in py3
+    df = DataFrame([[0, 1, 2]], columns=["A", "B", 0])
+    expected = DataFrame([[1, 2]], columns=["B", 0], index=Index([0], name="A"))
+
+    result = df.groupby("A").first()
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("A").sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_cython_grouper_series_bug_noncontig():
+    arr = np.empty((100, 100))
+    arr.fill(np.nan)
+    obj = Series(arr[:, 0])
+    inds = np.tile(range(10), 10)
+
+    result = obj.groupby(inds).agg(Series.median)
+    assert result.isna().all()
+
+
+def test_series_grouper_noncontig_index():
+    index = Index(["a" * 10] * 100)
+
+    values = Series(np.random.default_rng(2).standard_normal(50), index=index[::2])
+    labels = np.random.default_rng(2).integers(0, 5, 50)
+
+    # it works!
+    grouped = values.groupby(labels)
+
+    # accessing the index elements causes segfault
+    f = lambda x: len(set(map(id, x.index)))
+    grouped.agg(f)
+
+
+def test_convert_objects_leave_decimal_alone():
+    s = Series(range(5))
+    labels = np.array(["a", "b", "c", "d", "e"], dtype="O")
+
+    def convert_fast(x):
+        return Decimal(str(x.mean()))
+
+    def convert_force_pure(x):
+        # base will be length 0
+        assert len(x.values.base) > 0
+        return Decimal(str(x.mean()))
+
+    grouped = s.groupby(labels)
+
+    result = grouped.agg(convert_fast)
+    assert result.dtype == np.object_
+    assert isinstance(result.iloc[0], Decimal)
+
+    result = grouped.agg(convert_force_pure)
+    assert result.dtype == np.object_
+    assert isinstance(result.iloc[0], Decimal)
+
+
+def test_groupby_dtype_inference_empty():
+    # GH 6733
+    df = DataFrame({"x": [], "range": np.arange(0, dtype="int64")})
+    assert df["x"].dtype == np.float64
+
+    result = df.groupby("x").first()
+    exp_index = Index([], name="x", dtype=np.float64)
+    expected = DataFrame({"range": Series([], index=exp_index, dtype="int64")})
+    tm.assert_frame_equal(result, expected, by_blocks=True)
+
+
+def test_groupby_unit64_float_conversion():
+    # GH: 30859 groupby converts unit64 to floats sometimes
+    df = DataFrame({"first": [1], "second": [1], "value": [16148277970000000000]})
+    result = df.groupby(["first", "second"])["value"].max()
+    expected = Series(
+        [16148277970000000000],
+        MultiIndex.from_product([[1], [1]], names=["first", "second"]),
+        name="value",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_list_infer_array_like(df):
+    result = df.groupby(list(df["A"])).mean(numeric_only=True)
+    expected = df.groupby(df["A"]).mean(numeric_only=True)
+    tm.assert_frame_equal(result, expected, check_names=False)
+
+    with pytest.raises(KeyError, match=r"^'foo'$"):
+        df.groupby(list(df["A"][:-1]))
+
+    # pathological case of ambiguity
+    df = DataFrame(
+        {
+            "foo": [0, 1],
+            "bar": [3, 4],
+            "val": np.random.default_rng(2).standard_normal(2),
+        }
+    )
+
+    result = df.groupby(["foo", "bar"]).mean()
+    expected = df.groupby([df["foo"], df["bar"]]).mean()[["val"]]
+
+
+def test_groupby_keys_same_size_as_index():
+    # GH 11185
+    freq = "s"
+    index = date_range(
+        start=Timestamp("2015-09-29T11:34:44-0700"), periods=2, freq=freq
+    )
+    df = DataFrame([["A", 10], ["B", 15]], columns=["metric", "values"], index=index)
+    result = df.groupby([Grouper(level=0, freq=freq), "metric"]).mean()
+    expected = df.set_index([df.index, "metric"]).astype(float)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_one_row():
+    # GH 11741
+    msg = r"^'Z'$"
+    df1 = DataFrame(
+        np.random.default_rng(2).standard_normal((1, 4)), columns=list("ABCD")
+    )
+    with pytest.raises(KeyError, match=msg):
+        df1.groupby("Z")
+    df2 = DataFrame(
+        np.random.default_rng(2).standard_normal((2, 4)), columns=list("ABCD")
+    )
+    with pytest.raises(KeyError, match=msg):
+        df2.groupby("Z")
+
+
+def test_groupby_nat_exclude():
+    # GH 6992
+    df = DataFrame(
+        {
+            "values": np.random.default_rng(2).standard_normal(8),
+            "dt": [
+                np.nan,
+                Timestamp("2013-01-01"),
+                np.nan,
+                Timestamp("2013-02-01"),
+                np.nan,
+                Timestamp("2013-02-01"),
+                np.nan,
+                Timestamp("2013-01-01"),
+            ],
+            "str": [np.nan, "a", np.nan, "a", np.nan, "a", np.nan, "b"],
+        }
+    )
+    grouped = df.groupby("dt")
+
+    expected = [
+        RangeIndex(start=1, stop=13, step=6),
+        RangeIndex(start=3, stop=7, step=2),
+    ]
+    keys = sorted(grouped.groups.keys())
+    assert len(keys) == 2
+    for k, e in zip(keys, expected):
+        # grouped.groups keys are np.datetime64 with system tz
+        # not to be affected by tz, only compare values
+        tm.assert_index_equal(grouped.groups[k], e)
+
+    # confirm obj is not filtered
+    tm.assert_frame_equal(grouped._grouper.groupings[0].obj, df)
+    assert grouped.ngroups == 2
+
+    expected = {
+        Timestamp("2013-01-01 00:00:00"): np.array([1, 7], dtype=np.intp),
+        Timestamp("2013-02-01 00:00:00"): np.array([3, 5], dtype=np.intp),
+    }
+
+    for k in grouped.indices:
+        tm.assert_numpy_array_equal(grouped.indices[k], expected[k])
+
+    tm.assert_frame_equal(grouped.get_group(Timestamp("2013-01-01")), df.iloc[[1, 7]])
+    tm.assert_frame_equal(grouped.get_group(Timestamp("2013-02-01")), df.iloc[[3, 5]])
+
+    with pytest.raises(KeyError, match=r"^NaT$"):
+        grouped.get_group(pd.NaT)
+
+    nan_df = DataFrame(
+        {"nan": [np.nan, np.nan, np.nan], "nat": [pd.NaT, pd.NaT, pd.NaT]}
+    )
+    assert nan_df["nan"].dtype == "float64"
+    assert nan_df["nat"].dtype == "datetime64[s]"
+
+    for key in ["nan", "nat"]:
+        grouped = nan_df.groupby(key)
+        assert grouped.groups == {}
+        assert grouped.ngroups == 0
+        assert grouped.indices == {}
+        with pytest.raises(KeyError, match=r"^nan$"):
+            grouped.get_group(np.nan)
+        with pytest.raises(KeyError, match=r"^NaT$"):
+            grouped.get_group(pd.NaT)
+
+
+def test_groupby_two_group_keys_all_nan():
+    # GH #36842: Grouping over two group keys shouldn't raise an error
+    df = DataFrame({"a": [np.nan, np.nan], "b": [np.nan, np.nan], "c": [1, 2]})
+    result = df.groupby(["a", "b"]).indices
+    assert result == {}
+
+
+def test_groupby_2d_malformed():
+    d = DataFrame(index=range(2))
+    d["group"] = ["g1", "g2"]
+    d["zeros"] = [0, 0]
+    d["ones"] = [1, 1]
+    d["label"] = ["l1", "l2"]
+    tmp = d.groupby(["group"]).mean(numeric_only=True)
+    res_values = np.array([[0.0, 1.0], [0.0, 1.0]])
+    tm.assert_index_equal(tmp.columns, Index(["zeros", "ones"]))
+    tm.assert_numpy_array_equal(tmp.values, res_values)
+
+
+def test_int32_overflow():
+    B = np.concatenate((np.arange(10000), np.arange(10000), np.arange(5000)))
+    A = np.arange(25000)
+    df = DataFrame(
+        {
+            "A": A,
+            "B": B,
+            "C": A,
+            "D": B,
+            "E": np.random.default_rng(2).standard_normal(25000),
+        }
+    )
+
+    left = df.groupby(["A", "B", "C", "D"]).sum()
+    right = df.groupby(["D", "C", "B", "A"]).sum()
+    assert len(left) == len(right)
+
+
+def test_groupby_sort_multi():
+    df = DataFrame(
+        {
+            "a": ["foo", "bar", "baz"],
+            "b": [3, 2, 1],
+            "c": [0, 1, 2],
+            "d": np.random.default_rng(2).standard_normal(3),
+        }
+    )
+
+    tups = [tuple(row) for row in df[["a", "b", "c"]].values]
+    tups = com.asarray_tuplesafe(tups)
+    result = df.groupby(["a", "b", "c"], sort=True).sum()
+    tm.assert_numpy_array_equal(result.index.values, tups[[1, 2, 0]])
+
+    tups = [tuple(row) for row in df[["c", "a", "b"]].values]
+    tups = com.asarray_tuplesafe(tups)
+    result = df.groupby(["c", "a", "b"], sort=True).sum()
+    tm.assert_numpy_array_equal(result.index.values, tups)
+
+    tups = [tuple(x) for x in df[["b", "c", "a"]].values]
+    tups = com.asarray_tuplesafe(tups)
+    result = df.groupby(["b", "c", "a"], sort=True).sum()
+    tm.assert_numpy_array_equal(result.index.values, tups[[2, 1, 0]])
+
+    df = DataFrame(
+        {
+            "a": [0, 1, 2, 0, 1, 2],
+            "b": [0, 0, 0, 1, 1, 1],
+            "d": np.random.default_rng(2).standard_normal(6),
+        }
+    )
+    grouped = df.groupby(["a", "b"])["d"]
+    result = grouped.sum()
+
+    def _check_groupby(df, result, keys, field, f=lambda x: x.sum()):
+        tups = [tuple(row) for row in df[keys].values]
+        tups = com.asarray_tuplesafe(tups)
+        expected = f(df.groupby(tups)[field])
+        for k, v in expected.items():
+            assert result[k] == v
+
+    _check_groupby(df, result, ["a", "b"], "d")
+
+
+def test_dont_clobber_name_column():
+    df = DataFrame(
+        {"key": ["a", "a", "a", "b", "b", "b"], "name": ["foo", "bar", "baz"] * 2}
+    )
+
+    result = df.groupby("key", group_keys=False).apply(lambda x: x)
+    tm.assert_frame_equal(result, df[["name"]])
+
+
+def test_skip_group_keys():
+    tsf = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    grouped = tsf.groupby(lambda x: x.month, group_keys=False)
+    result = grouped.apply(lambda x: x.sort_values(by="A")[:3])
+
+    pieces = [group.sort_values(by="A")[:3] for key, group in grouped]
+
+    expected = pd.concat(pieces)
+    tm.assert_frame_equal(result, expected)
+
+    grouped = tsf["A"].groupby(lambda x: x.month, group_keys=False)
+    result = grouped.apply(lambda x: x.sort_values()[:3])
+
+    pieces = [group.sort_values()[:3] for key, group in grouped]
+
+    expected = pd.concat(pieces)
+    tm.assert_series_equal(result, expected)
+
+
+def test_no_nonsense_name(float_frame):
+    # GH #995
+    s = float_frame["C"].copy()
+    s.name = None
+
+    result = s.groupby(float_frame["A"]).agg("sum")
+    assert result.name is None
+
+
+def test_multifunc_sum_bug():
+    # GH #1065
+    x = DataFrame(np.arange(9).reshape(3, 3))
+    x["test"] = 0
+    x["fl"] = [1.3, 1.5, 1.6]
+
+    grouped = x.groupby("test")
+    result = grouped.agg({"fl": "sum", 2: "size"})
+    assert result["fl"].dtype == np.float64
+
+
+def test_handle_dict_return_value(df):
+    def f(group):
+        return {"max": group.max(), "min": group.min()}
+
+    def g(group):
+        return Series({"max": group.max(), "min": group.min()})
+
+    result = df.groupby("A")["C"].apply(f)
+    expected = df.groupby("A")["C"].apply(g)
+
+    assert isinstance(result, Series)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("grouper", ["A", ["A", "B"]])
+def test_set_group_name(df, grouper):
+    def f(group):
+        assert group.name is not None
+        return group
+
+    def freduce(group):
+        assert group.name is not None
+        return group.sum()
+
+    def freducex(x):
+        return freduce(x)
+
+    grouped = df.groupby(grouper, group_keys=False)
+
+    # make sure all these work
+    grouped.apply(f)
+    grouped.aggregate(freduce)
+    grouped.aggregate({"C": freduce, "D": freduce})
+    grouped.transform(f)
+
+    grouped["C"].apply(f)
+    grouped["C"].aggregate(freduce)
+    grouped["C"].aggregate([freduce, freducex])
+    grouped["C"].transform(f)
+
+
+def test_group_name_available_in_inference_pass():
+    # gh-15062
+    df = DataFrame({"a": [0, 0, 1, 1, 2, 2], "b": np.arange(6)})
+
+    names = []
+
+    def f(group):
+        names.append(group.name)
+        return group.copy()
+
+    df.groupby("a", sort=False, group_keys=False).apply(f)
+    expected_names = [0, 1, 2]
+    assert names == expected_names
+
+
+def test_no_dummy_key_names(df):
+    # see gh-1291
+    result = df.groupby(df["A"].values).sum()
+    assert result.index.name is None
+
+    result2 = df.groupby([df["A"].values, df["B"].values]).sum()
+    assert result2.index.names == (None, None)
+
+
+def test_groupby_sort_multiindex_series():
+    # series multiindex groupby sort argument was not being passed through
+    # _compress_group_index
+    # GH 9444
+    index = MultiIndex(
+        levels=[[1, 2], [1, 2]],
+        codes=[[0, 0, 0, 0, 1, 1], [1, 1, 0, 0, 0, 0]],
+        names=["a", "b"],
+    )
+    mseries = Series([0, 1, 2, 3, 4, 5], index=index)
+    index = MultiIndex(
+        levels=[[1, 2], [1, 2]], codes=[[0, 0, 1], [1, 0, 0]], names=["a", "b"]
+    )
+    mseries_result = Series([0, 2, 4], index=index)
+
+    result = mseries.groupby(level=["a", "b"], sort=False).first()
+    tm.assert_series_equal(result, mseries_result)
+    result = mseries.groupby(level=["a", "b"], sort=True).first()
+    tm.assert_series_equal(result, mseries_result.sort_index())
+
+
+def test_groupby_reindex_inside_function():
+    periods = 1000
+    ind = date_range(start="2012/1/1", freq="5min", periods=periods)
+    df = DataFrame({"high": np.arange(periods), "low": np.arange(periods)}, index=ind)
+
+    def agg_before(func, fix=False):
+        """
+        Run an aggregate func on the subset of data.
+        """
+
+        def _func(data):
+            d = data.loc[data.index.map(lambda x: x.hour < 11)].dropna()
+            if fix:
+                data[data.index[0]]
+            if len(d) == 0:
+                return None
+            return func(d)
+
+        return _func
+
+    grouped = df.groupby(lambda x: datetime(x.year, x.month, x.day))
+    closure_bad = grouped.agg({"high": agg_before(np.max)})
+    closure_good = grouped.agg({"high": agg_before(np.max, True)})
+
+    tm.assert_frame_equal(closure_bad, closure_good)
+
+
+def test_groupby_multiindex_missing_pair():
+    # GH9049
+    df = DataFrame(
+        {
+            "group1": ["a", "a", "a", "b"],
+            "group2": ["c", "c", "d", "c"],
+            "value": [1, 1, 1, 5],
+        }
+    )
+    df = df.set_index(["group1", "group2"])
+    df_grouped = df.groupby(level=["group1", "group2"], sort=True)
+
+    res = df_grouped.agg("sum")
+    idx = MultiIndex.from_tuples(
+        [("a", "c"), ("a", "d"), ("b", "c")], names=["group1", "group2"]
+    )
+    exp = DataFrame([[2], [1], [5]], index=idx, columns=["value"])
+
+    tm.assert_frame_equal(res, exp)
+
+
+def test_groupby_multiindex_not_lexsorted(performance_warning):
+    # GH 11640
+
+    # define the lexsorted version
+    lexsorted_mi = MultiIndex.from_tuples(
+        [("a", ""), ("b1", "c1"), ("b2", "c2")], names=["b", "c"]
+    )
+    lexsorted_df = DataFrame([[1, 3, 4]], columns=lexsorted_mi)
+    assert lexsorted_df.columns._is_lexsorted()
+
+    # define the non-lexsorted version
+    not_lexsorted_df = DataFrame(
+        columns=["a", "b", "c", "d"], data=[[1, "b1", "c1", 3], [1, "b2", "c2", 4]]
+    )
+    not_lexsorted_df = not_lexsorted_df.pivot_table(
+        index="a", columns=["b", "c"], values="d"
+    )
+    not_lexsorted_df = not_lexsorted_df.reset_index()
+    assert not not_lexsorted_df.columns._is_lexsorted()
+
+    expected = lexsorted_df.groupby("a").mean()
+    with tm.assert_produces_warning(performance_warning):
+        result = not_lexsorted_df.groupby("a").mean()
+    tm.assert_frame_equal(expected, result)
+
+    # a transforming function should work regardless of sort
+    # GH 14776
+    df = DataFrame(
+        {"x": ["a", "a", "b", "a"], "y": [1, 1, 2, 2], "z": [1, 2, 3, 4]}
+    ).set_index(["x", "y"])
+    assert not df.index._is_lexsorted()
+
+    for level in [0, 1, [0, 1]]:
+        for sort in [False, True]:
+            result = df.groupby(level=level, sort=sort, group_keys=False).apply(
+                DataFrame.drop_duplicates
+            )
+            expected = df
+            tm.assert_frame_equal(expected, result)
+
+            result = (
+                df.sort_index()
+                .groupby(level=level, sort=sort, group_keys=False)
+                .apply(DataFrame.drop_duplicates)
+            )
+            expected = df.sort_index()
+            tm.assert_frame_equal(expected, result)
+
+
+def test_index_label_overlaps_location():
+    # checking we don't have any label/location confusion in the
+    # wake of GH5375
+    df = DataFrame(list("ABCDE"), index=[2, 0, 2, 1, 1])
+    g = df.groupby(list("ababb"))
+    actual = g.filter(lambda x: len(x) > 2)
+    expected = df.iloc[[1, 3, 4]]
+    tm.assert_frame_equal(actual, expected)
+
+    ser = df[0]
+    g = ser.groupby(list("ababb"))
+    actual = g.filter(lambda x: len(x) > 2)
+    expected = ser.take([1, 3, 4])
+    tm.assert_series_equal(actual, expected)
+
+    #  and again, with a generic Index of floats
+    df.index = df.index.astype(float)
+    g = df.groupby(list("ababb"))
+    actual = g.filter(lambda x: len(x) > 2)
+    expected = df.iloc[[1, 3, 4]]
+    tm.assert_frame_equal(actual, expected)
+
+    ser = df[0]
+    g = ser.groupby(list("ababb"))
+    actual = g.filter(lambda x: len(x) > 2)
+    expected = ser.take([1, 3, 4])
+    tm.assert_series_equal(actual, expected)
+
+
+def test_transform_doesnt_clobber_ints():
+    # GH 7972
+    n = 6
+    x = np.arange(n)
+    df = DataFrame({"a": x // 2, "b": 2.0 * x, "c": 3.0 * x})
+    df2 = DataFrame({"a": x // 2 * 1.0, "b": 2.0 * x, "c": 3.0 * x})
+
+    gb = df.groupby("a")
+    result = gb.transform("mean")
+
+    gb2 = df2.groupby("a")
+    expected = gb2.transform("mean")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "sort_column",
+    ["ints", "floats", "strings", ["ints", "floats"], ["ints", "strings"]],
+)
+@pytest.mark.parametrize(
+    "group_column", ["int_groups", "string_groups", ["int_groups", "string_groups"]]
+)
+def test_groupby_preserves_sort(sort_column, group_column):
+    # Test to ensure that groupby always preserves sort order of original
+    # object. Issue #8588 and #9651
+
+    df = DataFrame(
+        {
+            "int_groups": [3, 1, 0, 1, 0, 3, 3, 3],
+            "string_groups": ["z", "a", "z", "a", "a", "g", "g", "g"],
+            "ints": [8, 7, 4, 5, 2, 9, 1, 1],
+            "floats": [2.3, 5.3, 6.2, -2.4, 2.2, 1.1, 1.1, 5],
+            "strings": ["z", "d", "a", "e", "word", "word2", "42", "47"],
+        }
+    )
+
+    # Try sorting on different types and with different group types
+
+    df = df.sort_values(by=sort_column)
+    g = df.groupby(group_column)
+
+    def test_sort(x):
+        tm.assert_frame_equal(x, x.sort_values(by=sort_column))
+
+    g.apply(test_sort)
+
+
+def test_pivot_table_values_key_error():
+    # This test is designed to replicate the error in issue #14938
+    df = DataFrame(
+        {
+            "eventDate": date_range(datetime.today(), periods=20, freq="ME").tolist(),
+            "thename": range(20),
+        }
+    )
+
+    df["year"] = df.set_index("eventDate").index.year
+    df["month"] = df.set_index("eventDate").index.month
+
+    with pytest.raises(KeyError, match="'badname'"):
+        df.reset_index().pivot_table(
+            index="year", columns="month", values="badname", aggfunc="count"
+        )
+
+
+@pytest.mark.parametrize("columns", ["C", ["C"]])
+@pytest.mark.parametrize("keys", [["A"], ["A", "B"]])
+@pytest.mark.parametrize(
+    "values",
+    [
+        [True],
+        [0],
+        [0.0],
+        ["a"],
+        Categorical([0]),
+        [to_datetime(0)],
+        date_range(0, 1, 1, tz="US/Eastern"),
+        pd.period_range("2016-01-01", periods=3, freq="D"),
+        pd.array([0], dtype="Int64"),
+        pd.array([0], dtype="Float64"),
+        pd.array([False], dtype="boolean"),
+    ],
+    ids=[
+        "bool",
+        "int",
+        "float",
+        "str",
+        "cat",
+        "dt64",
+        "dt64tz",
+        "period",
+        "Int64",
+        "Float64",
+        "boolean",
+    ],
+)
+@pytest.mark.parametrize("method", ["attr", "agg", "apply"])
+@pytest.mark.parametrize(
+    "op", ["idxmax", "idxmin", "min", "max", "sum", "prod", "skew", "kurt"]
+)
+def test_empty_groupby(columns, keys, values, method, op, dropna, using_infer_string):
+    # GH8093 & GH26411
+    override_dtype = None
+
+    if isinstance(values, BooleanArray) and op in ["sum", "prod"]:
+        # We expect to get Int64 back for these
+        override_dtype = "Int64"
+
+    if isinstance(values[0], bool) and op in ("prod", "sum"):
+        # sum/product of bools is an integer
+        override_dtype = "int64"
+
+    df = DataFrame({"A": values, "B": values, "C": values}, columns=list("ABC"))
+
+    if hasattr(values, "dtype"):
+        # check that we did the construction right
+        assert (df.dtypes == values.dtype).all()
+
+    df = df.iloc[:0]
+
+    gb = df.groupby(keys, group_keys=False, dropna=dropna, observed=False)[columns]
+
+    def get_result(**kwargs):
+        if method == "attr":
+            return getattr(gb, op)(**kwargs)
+        else:
+            return getattr(gb, method)(op, **kwargs)
+
+    def get_categorical_invalid_expected():
+        # Categorical is special without 'observed=True', we get an NaN entry
+        #  corresponding to the unobserved group. If we passed observed=True
+        #  to groupby, expected would just be 'df.set_index(keys)[columns]'
+        #  as below
+        lev = Categorical([0], dtype=values.dtype)
+        if len(keys) != 1:
+            idx = MultiIndex.from_product([lev, lev], names=keys)
+        else:
+            # all columns are dropped, but we end up with one row
+            # Categorical is special without 'observed=True'
+            idx = Index(lev, name=keys[0])
+
+        if using_infer_string:
+            columns = Index([], dtype="str")
+        else:
+            columns = []
+        expected = DataFrame([], columns=columns, index=idx)
+        return expected
+
+    is_per = isinstance(df.dtypes.iloc[0], pd.PeriodDtype)
+    is_dt64 = df.dtypes.iloc[0].kind == "M"
+    is_cat = isinstance(values, Categorical)
+    is_str = isinstance(df.dtypes.iloc[0], pd.StringDtype)
+
+    if (
+        isinstance(values, Categorical)
+        and not values.ordered
+        and op in ["min", "max", "idxmin", "idxmax"]
+    ):
+        if op in ["min", "max"]:
+            msg = f"Cannot perform {op} with non-ordered Categorical"
+            klass = TypeError
+        else:
+            msg = f"Can't get {op} of an empty group due to unobserved categories"
+            klass = ValueError
+        with pytest.raises(klass, match=msg):
+            get_result()
+
+        if op in ["min", "max", "idxmin", "idxmax"] and isinstance(columns, list):
+            # i.e. DataframeGroupBy, not SeriesGroupBy
+            result = get_result(numeric_only=True)
+            expected = get_categorical_invalid_expected()
+            tm.assert_equal(result, expected)
+        return
+
+    if op in ["prod", "sum", "skew", "kurt"]:
+        # ops that require more than just ordered-ness
+        if is_dt64 or is_cat or is_per or (is_str and op != "sum"):
+            # GH#41291
+            # datetime64 -> prod and sum are invalid
+            if is_dt64:
+                msg = "datetime64 type does not support"
+            elif is_per:
+                msg = "Period type does not support"
+            elif is_str:
+                msg = f"dtype 'str' does not support operation '{op}'"
+            else:
+                msg = "category type does not support"
+            if op in ["skew", "kurt"]:
+                msg = "|".join([msg, f"does not support operation '{op}'"])
+            with pytest.raises(TypeError, match=msg):
+                get_result()
+
+            if not isinstance(columns, list):
+                # i.e. SeriesGroupBy
+                return
+            elif op in ["skew", "kurt"]:
+                # TODO: test the numeric_only=True case
+                return
+            else:
+                # i.e. op in ["prod", "sum"]:
+                # i.e. DataFrameGroupBy
+                # ops that require more than just ordered-ness
+                # GH#41291
+                result = get_result(numeric_only=True)
+
+                # with numeric_only=True, these are dropped, and we get
+                # an empty DataFrame back
+                expected = df.set_index(keys)[[]]
+                if is_cat:
+                    expected = get_categorical_invalid_expected()
+                tm.assert_equal(result, expected)
+                return
+
+    result = get_result()
+    expected = df.set_index(keys)[columns]
+    if op in ["idxmax", "idxmin"]:
+        expected = expected.astype(df.index.dtype)
+    if override_dtype is not None:
+        expected = expected.astype(override_dtype)
+    if len(keys) == 1:
+        expected.index.name = keys[0]
+    tm.assert_equal(result, expected)
+
+
+def test_empty_groupby_apply_nonunique_columns():
+    # GH#44417
+    df = DataFrame(np.random.default_rng(2).standard_normal((0, 4)))
+    df[3] = df[3].astype(np.int64)
+    df.columns = [0, 1, 2, 0]
+    gb = df.groupby(df[1], group_keys=False)
+    res = gb.apply(lambda x: x)
+    assert (res.dtypes == df.drop(columns=1).dtypes).all()
+
+
+def test_tuple_as_grouping():
+    # https://github.com/pandas-dev/pandas/issues/18314
+    df = DataFrame(
+        {
+            ("a", "b"): [1, 1, 1, 1],
+            "a": [2, 2, 2, 2],
+            "b": [2, 2, 2, 2],
+            "c": [1, 1, 1, 1],
+        }
+    )
+
+    with pytest.raises(KeyError, match=r"('a', 'b')"):
+        df[["a", "b", "c"]].groupby(("a", "b"))
+
+    result = df.groupby(("a", "b"))["c"].sum()
+    expected = Series([4], name="c", index=Index([1], name=("a", "b")))
+    tm.assert_series_equal(result, expected)
+
+
+def test_tuple_correct_keyerror():
+    # https://github.com/pandas-dev/pandas/issues/18798
+    df = DataFrame(1, index=range(3), columns=MultiIndex.from_product([[1, 2], [3, 4]]))
+    with pytest.raises(KeyError, match=r"^\(7, 8\)$"):
+        df.groupby((7, 8)).mean()
+
+
+def test_groupby_agg_ohlc_non_first():
+    # GH 21716
+    df = DataFrame(
+        [[1], [1]],
+        columns=Index(["foo"], name="mycols"),
+        index=date_range("2018-01-01", periods=2, freq="D", name="dti"),
+    )
+
+    expected = DataFrame(
+        [[1, 1, 1, 1, 1], [1, 1, 1, 1, 1]],
+        columns=MultiIndex.from_tuples(
+            (
+                ("foo", "sum", "foo"),
+                ("foo", "ohlc", "open"),
+                ("foo", "ohlc", "high"),
+                ("foo", "ohlc", "low"),
+                ("foo", "ohlc", "close"),
+            ),
+            names=["mycols", None, None],
+        ),
+        index=date_range("2018-01-01", periods=2, freq="D", name="dti"),
+    )
+
+    result = df.groupby(Grouper(freq="D")).agg(["sum", "ohlc"])
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_multiindex_nat():
+    # GH 9236
+    values = [
+        (pd.NaT, "a"),
+        (datetime(2012, 1, 2), "a"),
+        (datetime(2012, 1, 2), "b"),
+        (datetime(2012, 1, 3), "a"),
+    ]
+    mi = MultiIndex.from_tuples(values, names=["date", None])
+    ser = Series([3, 2, 2.5, 4], index=mi)
+
+    result = ser.groupby(level=1).mean()
+    expected = Series([3.0, 2.5], index=["a", "b"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_empty_list_raises():
+    # GH 5289
+    values = zip(range(10), range(10))
+    df = DataFrame(values, columns=["apple", "b"])
+    msg = "Grouper and axis must be same length"
+    with pytest.raises(ValueError, match=msg):
+        df.groupby([[]])
+
+
+def test_groupby_multiindex_series_keys_len_equal_group_axis():
+    # GH 25704
+    index_array = [["x", "x"], ["a", "b"], ["k", "k"]]
+    index_names = ["first", "second", "third"]
+    ri = MultiIndex.from_arrays(index_array, names=index_names)
+    s = Series(data=[1, 2], index=ri)
+    result = s.groupby(["first", "third"]).sum()
+
+    index_array = [["x"], ["k"]]
+    index_names = ["first", "third"]
+    ei = MultiIndex.from_arrays(index_array, names=index_names)
+    expected = Series([3], index=ei)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_groups_in_BaseGrouper():
+    # GH 26326
+    # Test if DataFrame grouped with a pandas.Grouper has correct groups
+    mi = MultiIndex.from_product([["A", "B"], ["C", "D"]], names=["alpha", "beta"])
+    df = DataFrame({"foo": [1, 2, 1, 2], "bar": [1, 2, 3, 4]}, index=mi)
+    result = df.groupby([Grouper(level="alpha"), "beta"])
+    expected = df.groupby(["alpha", "beta"])
+    assert result.groups == expected.groups
+
+    result = df.groupby(["beta", Grouper(level="alpha")])
+    expected = df.groupby(["beta", "alpha"])
+    assert result.groups == expected.groups
+
+
+def test_groups_sort_dropna(sort, dropna):
+    # GH#56966, GH#56851
+    df = DataFrame([[2.0, 1.0], [np.nan, 4.0], [0.0, 3.0]])
+    keys = [(2.0, 1.0), (np.nan, 4.0), (0.0, 3.0)]
+    values = [
+        RangeIndex(0, 1),
+        RangeIndex(1, 2),
+        RangeIndex(2, 3),
+    ]
+    if sort:
+        taker = [2, 0] if dropna else [2, 0, 1]
+    else:
+        taker = [0, 2] if dropna else [0, 1, 2]
+    expected = {keys[idx]: values[idx] for idx in taker}
+
+    gb = df.groupby([0, 1], sort=sort, dropna=dropna)
+    result = gb.groups
+
+    for result_key, expected_key in zip(result.keys(), expected.keys()):
+        # Compare as NumPy arrays to handle np.nan
+        result_key = np.array(result_key)
+        expected_key = np.array(expected_key)
+        tm.assert_numpy_array_equal(result_key, expected_key)
+    for result_value, expected_value in zip(result.values(), expected.values()):
+        tm.assert_index_equal(result_value, expected_value)
+
+
+@pytest.mark.parametrize(
+    "op, expected",
+    [
+        (
+            "shift",
+            {
+                "time": [
+                    None,
+                    None,
+                    Timestamp("2019-01-01 12:00:00"),
+                    Timestamp("2019-01-01 12:30:00"),
+                    None,
+                    None,
+                ]
+            },
+        ),
+        (
+            "bfill",
+            {
+                "time": [
+                    Timestamp("2019-01-01 12:00:00"),
+                    Timestamp("2019-01-01 12:30:00"),
+                    Timestamp("2019-01-01 14:00:00"),
+                    Timestamp("2019-01-01 14:30:00"),
+                    Timestamp("2019-01-01 14:00:00"),
+                    Timestamp("2019-01-01 14:30:00"),
+                ]
+            },
+        ),
+        (
+            "ffill",
+            {
+                "time": [
+                    Timestamp("2019-01-01 12:00:00"),
+                    Timestamp("2019-01-01 12:30:00"),
+                    Timestamp("2019-01-01 12:00:00"),
+                    Timestamp("2019-01-01 12:30:00"),
+                    Timestamp("2019-01-01 14:00:00"),
+                    Timestamp("2019-01-01 14:30:00"),
+                ]
+            },
+        ),
+    ],
+)
+def test_shift_bfill_ffill_tz(tz_naive_fixture, op, expected):
+    # GH19995, GH27992: Check that timezone does not drop in shift, bfill, and ffill
+    tz = tz_naive_fixture
+    data = {
+        "id": ["A", "B", "A", "B", "A", "B"],
+        "time": [
+            Timestamp("2019-01-01 12:00:00"),
+            Timestamp("2019-01-01 12:30:00"),
+            None,
+            None,
+            Timestamp("2019-01-01 14:00:00"),
+            Timestamp("2019-01-01 14:30:00"),
+        ],
+    }
+    df = DataFrame(data).assign(time=lambda x: x.time.dt.tz_localize(tz))
+
+    grouped = df.groupby("id")
+    result = getattr(grouped, op)()
+    expected = DataFrame(expected).assign(time=lambda x: x.time.dt.tz_localize(tz))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_only_none_group():
+    # see GH21624
+    # this was crashing with "ValueError: Length of passed values is 1, index implies 0"
+    df = DataFrame({"g": [None], "x": 1})
+    actual = df.groupby("g")["x"].transform("sum")
+    expected = Series([np.nan], name="x")
+
+    tm.assert_series_equal(actual, expected)
+
+
+def test_groupby_duplicate_index():
+    # GH#29189 the groupby call here used to raise
+    ser = Series([2, 5, 6, 8], index=[2.0, 4.0, 4.0, 5.0])
+    gb = ser.groupby(level=0)
+
+    result = gb.mean()
+    expected = Series([2, 5.5, 8], index=[2.0, 4.0, 5.0])
+    tm.assert_series_equal(result, expected)
+
+
+def test_group_on_empty_multiindex(transformation_func, request):
+    # GH 47787
+    # With one row, those are transforms so the schema should be the same
+    df = DataFrame(
+        data=[[1, Timestamp("today"), 3, 4]],
+        columns=["col_1", "col_2", "col_3", "col_4"],
+    )
+    df["col_3"] = df["col_3"].astype(int)
+    df["col_4"] = df["col_4"].astype(int)
+    df = df.set_index(["col_1", "col_2"])
+    result = df.iloc[:0].groupby(["col_1"]).transform(transformation_func)
+    expected = df.groupby(["col_1"]).transform(transformation_func).iloc[:0]
+    if transformation_func in ("diff", "shift"):
+        expected = expected.astype(int)
+    tm.assert_equal(result, expected)
+
+    result = df["col_3"].iloc[:0].groupby(["col_1"]).transform(transformation_func)
+    expected = df["col_3"].groupby(["col_1"]).transform(transformation_func).iloc[:0]
+    if transformation_func in ("diff", "shift"):
+        expected = expected.astype(int)
+    tm.assert_equal(result, expected)
+
+
+def test_groupby_crash_on_nunique():
+    # Fix following 30253
+    dti = date_range("2016-01-01", periods=2, name="foo")
+    df = DataFrame({("A", "B"): [1, 2], ("A", "C"): [1, 3], ("D", "B"): [0, 0]})
+    df.columns.names = ("bar", "baz")
+    df.index = dti
+
+    df = df.T
+    gb = df.groupby(level=0)
+    result = gb.nunique()
+
+    expected = DataFrame({"A": [1, 2], "D": [1, 1]}, index=dti)
+    expected.columns.name = "bar"
+    expected = expected.T
+
+    tm.assert_frame_equal(result, expected)
+
+    # same thing, but empty columns
+    gb2 = df[[]].groupby(level=0)
+    exp = expected[[]]
+
+    res = gb2.nunique()
+    tm.assert_frame_equal(res, exp)
+
+
+def test_groupby_list_level():
+    # GH 9790
+    expected = DataFrame(np.arange(0, 9).reshape(3, 3), dtype=float)
+    result = expected.groupby(level=[0]).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "max_seq_items, expected",
+    [
+        (5, "{0: [0], 1: [1], 2: [2], 3: [3], 4: [4]}"),
+        (4, "{0: [0], 1: [1], 2: [2], 3: [3], ...}"),
+        (1, "{0: [0], ...}"),
+    ],
+)
+def test_groups_repr_truncates(max_seq_items, expected):
+    # GH 1135
+    df = DataFrame(np.random.default_rng(2).standard_normal((5, 1)))
+    df["a"] = df.index
+
+    with pd.option_context("display.max_seq_items", max_seq_items):
+        result = df.groupby("a").groups.__repr__()
+        assert result == expected
+
+        result = df.groupby(np.array(df.a)).groups.__repr__()
+        assert result == expected
+
+
+def test_group_on_two_row_multiindex_returns_one_tuple_key():
+    # GH 18451
+    df = DataFrame([{"a": 1, "b": 2, "c": 99}, {"a": 1, "b": 2, "c": 88}])
+    df = df.set_index(["a", "b"])
+
+    grp = df.groupby(["a", "b"])
+    result = grp.indices
+    expected = {(1, 2): np.array([0, 1], dtype=np.int64)}
+
+    assert len(result) == 1
+    key = (1, 2)
+    assert (result[key] == expected[key]).all()
+
+
+@pytest.mark.parametrize(
+    "klass, attr, value",
+    [
+        (DataFrame, "level", "a"),
+        (DataFrame, "as_index", False),
+        (DataFrame, "sort", False),
+        (DataFrame, "group_keys", False),
+        (DataFrame, "observed", True),
+        (DataFrame, "dropna", False),
+        (Series, "level", "a"),
+        (Series, "as_index", False),
+        (Series, "sort", False),
+        (Series, "group_keys", False),
+        (Series, "observed", True),
+        (Series, "dropna", False),
+    ],
+)
+def test_subsetting_columns_keeps_attrs(klass, attr, value):
+    # GH 9959 - When subsetting columns, don't drop attributes
+    df = DataFrame({"a": [1], "b": [2], "c": [3]})
+    if attr != "axis":
+        df = df.set_index("a")
+
+    expected = df.groupby("a", **{attr: value})
+    result = expected[["b"]] if klass is DataFrame else expected["b"]
+    assert getattr(result, attr) == getattr(expected, attr)
+
+
+@pytest.mark.parametrize("func", ["sum", "any", "shift"])
+def test_groupby_column_index_name_lost(func):
+    # GH: 29764 groupby loses index sometimes
+    expected = Index(["a"], name="idx")
+    df = DataFrame([[1]], columns=expected)
+    df_grouped = df.groupby([1])
+    result = getattr(df_grouped, func)().columns
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "infer_string",
+    [
+        False,
+        pytest.param(True, marks=td.skip_if_no("pyarrow")),
+    ],
+)
+def test_groupby_duplicate_columns(infer_string):
+    # GH: 31735
+    if infer_string:
+        pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {"A": ["f", "e", "g", "h"], "B": ["a", "b", "c", "d"], "C": [1, 2, 3, 4]}
+    ).astype(object)
+    df.columns = ["A", "B", "B"]
+    with pd.option_context("future.infer_string", infer_string):
+        result = df.groupby([0, 0, 0, 0]).min()
+    expected = DataFrame(
+        [["e", "a", 1]], index=np.array([0]), columns=["A", "B", "B"], dtype=object
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_series_with_tuple_name():
+    # GH 37755
+    ser = Series([1, 2, 3, 4], index=[1, 1, 2, 2], name=("a", "a"))
+    ser.index.name = ("b", "b")
+    result = ser.groupby(level=0).last()
+    expected = Series([2, 4], index=[1, 2], name=("a", "a"))
+    expected.index.name = ("b", "b")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, values", [("sum", [97.0, 98.0]), ("mean", [24.25, 24.5])]
+)
+def test_groupby_numerical_stability_sum_mean(func, values):
+    # GH#38778
+    data = [1e16, 1e16, 97, 98, -5e15, -5e15, -5e15, -5e15]
+    df = DataFrame({"group": [1, 2] * 4, "a": data, "b": data})
+    result = getattr(df.groupby("group"), func)()
+    expected = DataFrame({"a": values, "b": values}, index=Index([1, 2], name="group"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_numerical_stability_cumsum():
+    # GH#38934
+    data = [1e16, 1e16, 97, 98, -5e15, -5e15, -5e15, -5e15]
+    df = DataFrame({"group": [1, 2] * 4, "a": data, "b": data})
+    result = df.groupby("group").cumsum()
+    exp_data = (
+        [1e16] * 2 + [1e16 + 96, 1e16 + 98] + [5e15 + 97, 5e15 + 98] + [97.0, 98.0]
+    )
+    expected = DataFrame({"a": exp_data, "b": exp_data})
+    tm.assert_frame_equal(result, expected, check_exact=True)
+
+
+def test_groupby_cumsum_skipna_false():
+    # GH#46216 don't propagate np.nan above the diagonal
+    arr = np.random.default_rng(2).standard_normal((5, 5))
+    df = DataFrame(arr)
+    for i in range(5):
+        df.iloc[i, i] = np.nan
+
+    df["A"] = 1
+    gb = df.groupby("A")
+
+    res = gb.cumsum(skipna=False)
+
+    expected = df[[0, 1, 2, 3, 4]].cumsum(skipna=False)
+    tm.assert_frame_equal(res, expected)
+
+
+def test_groupby_cumsum_timedelta64():
+    # GH#46216 don't ignore is_datetimelike in libgroupby.group_cumsum
+    dti = date_range("2016-01-01", periods=5)
+    ser = Series(dti) - dti[0]
+    ser[2] = pd.NaT
+
+    df = DataFrame({"A": 1, "B": ser})
+    gb = df.groupby("A")
+
+    res = gb.cumsum(numeric_only=False, skipna=True)
+    exp = DataFrame({"B": [ser[0], ser[1], pd.NaT, ser[4], ser[4] * 2]})
+    tm.assert_frame_equal(res, exp)
+
+    res = gb.cumsum(numeric_only=False, skipna=False)
+    exp = DataFrame({"B": [ser[0], ser[1], pd.NaT, pd.NaT, pd.NaT]})
+    tm.assert_frame_equal(res, exp)
+
+
+def test_groupby_mean_duplicate_index(rand_series_with_duplicate_datetimeindex):
+    dups = rand_series_with_duplicate_datetimeindex
+    result = dups.groupby(level=0).mean()
+    expected = dups.groupby(dups.index).mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_all_nan_groups_drop():
+    # GH 15036
+    s = Series([1, 2, 3], [np.nan, np.nan, np.nan])
+    result = s.groupby(s.index).sum()
+    expected = Series([], index=Index([], dtype=np.float64), dtype=np.int64)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_groupby_empty_multi_column(as_index, numeric_only):
+    # GH 15106 & GH 41998
+    df = DataFrame(data=[], columns=["A", "B", "C"])
+    gb = df.groupby(["A", "B"], as_index=as_index)
+    result = gb.sum(numeric_only=numeric_only)
+    if as_index:
+        index = MultiIndex([[], []], [[], []], names=["A", "B"])
+        columns = ["C"] if not numeric_only else Index([], dtype="str")
+    else:
+        index = RangeIndex(0)
+        columns = ["A", "B", "C"] if not numeric_only else ["A", "B"]
+    expected = DataFrame([], columns=columns, index=index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_non_numeric_dtype():
+    # GH #43108
+    df = DataFrame(
+        [["M", [1]], ["M", [1]], ["W", [10]], ["W", [20]]], columns=["MW", "v"]
+    )
+
+    expected = DataFrame(
+        {
+            "v": [[1, 1], [10, 20]],
+        },
+        index=Index(["M", "W"], name="MW"),
+    )
+
+    gb = df.groupby(by=["MW"])
+    result = gb.sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_multi_non_numeric_dtype():
+    # GH #42395
+    df = DataFrame(
+        {
+            "x": [1, 0, 1, 1, 0],
+            "y": [Timedelta(i, "days") for i in range(1, 6)],
+            "z": [Timedelta(i * 10, "days") for i in range(1, 6)],
+        }
+    )
+
+    expected = DataFrame(
+        {
+            "y": [Timedelta(i, "days") for i in range(7, 9)],
+            "z": [Timedelta(i * 10, "days") for i in range(7, 9)],
+        },
+        index=Index([0, 1], dtype="int64", name="x"),
+    )
+
+    gb = df.groupby(by=["x"])
+    result = gb.sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_aggregation_numeric_with_non_numeric_dtype():
+    # GH #43108
+    df = DataFrame(
+        {
+            "x": [1, 0, 1, 1, 0],
+            "y": [Timedelta(i, "days") for i in range(1, 6)],
+            "z": list(range(1, 6)),
+        }
+    )
+
+    expected = DataFrame(
+        {"y": [Timedelta(7, "days"), Timedelta(8, "days")], "z": [7, 8]},
+        index=Index([0, 1], dtype="int64", name="x"),
+    )
+
+    gb = df.groupby(by=["x"])
+    result = gb.sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_filtered_df_std():
+    # GH 16174
+    dicts = [
+        {"filter_col": False, "groupby_col": True, "bool_col": True, "float_col": 10.5},
+        {"filter_col": True, "groupby_col": True, "bool_col": True, "float_col": 20.5},
+        {"filter_col": True, "groupby_col": True, "bool_col": True, "float_col": 30.5},
+    ]
+    df = DataFrame(dicts)
+
+    df_filter = df[df["filter_col"] == True]  # noqa: E712
+    dfgb = df_filter.groupby("groupby_col")
+    result = dfgb.std()
+    expected = DataFrame(
+        [[0.0, 0.0, 7.071068]],
+        columns=["filter_col", "bool_col", "float_col"],
+        index=Index([True], name="groupby_col"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_datetime_categorical_multikey_groupby_indices():
+    # GH 26859
+    df = DataFrame(
+        {
+            "a": Series(list("abc")),
+            "b": Series(
+                to_datetime(["2018-01-01", "2018-02-01", "2018-03-01"]),
+                dtype="category",
+            ),
+            "c": Categorical.from_codes([-1, 0, 1], categories=[0, 1]),
+        }
+    )
+    result = df.groupby(["a", "b"], observed=False).indices
+    expected = {
+        ("a", Timestamp("2018-01-01 00:00:00")): np.array([0]),
+        ("b", Timestamp("2018-02-01 00:00:00")): np.array([1]),
+        ("c", Timestamp("2018-03-01 00:00:00")): np.array([2]),
+    }
+    assert result == expected
+
+
+def test_rolling_wrong_param_min_period():
+    # GH34037
+    name_l = ["Alice"] * 5 + ["Bob"] * 5
+    val_l = [np.nan, np.nan, 1, 2, 3] + [np.nan, 1, 2, 3, 4]
+    test_df = DataFrame([name_l, val_l]).T
+    test_df.columns = ["name", "val"]
+
+    result_error_msg = (
+        r"^[a-zA-Z._]*\(\) got an unexpected keyword argument 'min_period'"
+    )
+    with pytest.raises(TypeError, match=result_error_msg):
+        test_df.groupby("name")["val"].rolling(window=2, min_period=1).sum()
+
+
+def test_by_column_values_with_same_starting_value(any_string_dtype):
+    # GH29635
+    df = DataFrame(
+        {
+            "Name": ["Thomas", "Thomas", "Thomas John"],
+            "Credit": [1200, 1300, 900],
+            "Mood": Series(["sad", "happy", "happy"], dtype=any_string_dtype),
+        }
+    )
+    aggregate_details = {"Mood": Series.mode, "Credit": "sum"}
+
+    result = df.groupby(["Name"]).agg(aggregate_details)
+    expected_result = DataFrame(
+        {
+            "Mood": [["happy", "sad"], "happy"],
+            "Credit": [2500, 900],
+            "Name": ["Thomas", "Thomas John"],
+        }
+    ).set_index("Name")
+
+    tm.assert_frame_equal(result, expected_result)
+
+
+def test_groupby_none_in_first_mi_level():
+    # GH#47348
+    arr = [[None, 1, 0, 1], [2, 3, 2, 3]]
+    ser = Series(1, index=MultiIndex.from_arrays(arr, names=["a", "b"]))
+    result = ser.groupby(level=[0, 1]).sum()
+    expected = Series(
+        [1, 2], MultiIndex.from_tuples([(0.0, 2), (1.0, 3)], names=["a", "b"])
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_none_column_name(using_infer_string):
+    # GH#47348
+    df = DataFrame({None: [1, 1, 2, 2], "b": [1, 1, 2, 3], "c": [4, 5, 6, 7]})
+    by = [np.nan] if using_infer_string else [None]
+    gb = df.groupby(by=by)
+    result = gb.sum()
+    expected = DataFrame({"b": [2, 5], "c": [9, 13]}, index=Index([1, 2], name=by[0]))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("selection", [None, "a", ["a"]])
+def test_single_element_list_grouping(selection):
+    # GH#42795, GH#53500
+    df = DataFrame({"a": [1, 2], "b": [np.nan, 5], "c": [np.nan, 2]}, index=["x", "y"])
+    grouped = df.groupby(["a"]) if selection is None else df.groupby(["a"])[selection]
+    result = [key for key, _ in grouped]
+
+    expected = [(1,), (2,)]
+    assert result == expected
+
+
+def test_groupby_string_dtype():
+    # GH 40148
+    df = DataFrame({"str_col": ["a", "b", "c", "a"], "num_col": [1, 2, 3, 2]})
+    df["str_col"] = df["str_col"].astype("string")
+    expected = DataFrame(
+        {
+            "str_col": [
+                "a",
+                "b",
+                "c",
+            ],
+            "num_col": [1.5, 2.0, 3.0],
+        }
+    )
+    expected["str_col"] = expected["str_col"].astype("string")
+    grouped = df.groupby("str_col", as_index=False)
+    result = grouped.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "level_arg, multiindex", [([0], False), ((0,), False), ([0], True), ((0,), True)]
+)
+def test_single_element_listlike_level_grouping(level_arg, multiindex):
+    # GH 51583
+    df = DataFrame({"a": [1, 2], "b": [3, 4], "c": [5, 6]}, index=["x", "y"])
+    if multiindex:
+        df = df.set_index(["a", "b"])
+    result = [key for key, _ in df.groupby(level=level_arg)]
+    expected = [(1,), (2,)] if multiindex else [("x",), ("y",)]
+    assert result == expected
+
+
+@pytest.mark.parametrize("func", ["sum", "cumsum", "cumprod", "prod"])
+def test_groupby_avoid_casting_to_float(func):
+    # GH#37493
+    val = 922337203685477580
+    df = DataFrame({"a": 1, "b": [val]})
+    result = getattr(df.groupby("a"), func)() - val
+    expected = DataFrame({"b": [0]}, index=Index([1], name="a"))
+    if func in ["cumsum", "cumprod"]:
+        expected = expected.reset_index(drop=True)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func, val", [("sum", 3), ("prod", 2)])
+def test_groupby_sum_support_mask(any_numeric_ea_dtype, func, val):
+    # GH#37493
+    df = DataFrame({"a": 1, "b": [1, 2, pd.NA]}, dtype=any_numeric_ea_dtype)
+    result = getattr(df.groupby("a"), func)()
+    expected = DataFrame(
+        {"b": [val]},
+        index=Index([1], name="a", dtype=any_numeric_ea_dtype),
+        dtype=any_numeric_ea_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("val, dtype", [(111, "int"), (222, "uint")])
+def test_groupby_overflow(val, dtype):
+    # GH#37493
+    df = DataFrame({"a": 1, "b": [val, val]}, dtype=f"{dtype}8")
+    result = df.groupby("a").sum()
+    expected = DataFrame(
+        {"b": [val * 2]},
+        index=Index([1], name="a", dtype=f"{dtype}8"),
+        dtype=f"{dtype}64",
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a").cumsum()
+    expected = DataFrame({"b": [val, val * 2]}, dtype=f"{dtype}64")
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("a").prod()
+    expected = DataFrame(
+        {"b": [val * val]},
+        index=Index([1], name="a", dtype=f"{dtype}8"),
+        dtype=f"{dtype}64",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("skipna, val", [(True, 3), (False, pd.NA)])
+def test_groupby_cumsum_mask(any_numeric_ea_dtype, skipna, val):
+    # GH#37493
+    df = DataFrame({"a": 1, "b": [1, pd.NA, 2]}, dtype=any_numeric_ea_dtype)
+    result = df.groupby("a").cumsum(skipna=skipna)
+    expected = DataFrame(
+        {"b": [1, pd.NA, val]},
+        dtype=any_numeric_ea_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "val_in, index, val_out",
+    [
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0],
+            ["foo", "foo", "bar", "baz", "blah"],
+            [3.0, 4.0, 5.0, 3.0],
+        ),
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0],
+            ["foo", "foo", "bar", "baz", "blah", "blah"],
+            [3.0, 4.0, 11.0, 3.0],
+        ),
+    ],
+)
+def test_groupby_index_name_in_index_content(val_in, index, val_out):
+    # GH 48567
+    series = Series(data=val_in, name="values", index=Index(index, name="blah"))
+    result = series.groupby("blah").sum()
+    expected = Series(
+        data=val_out,
+        name="values",
+        index=Index(["bar", "baz", "blah", "foo"], name="blah"),
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = series.to_frame().groupby("blah").sum()
+    expected = expected.to_frame()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("n", [1, 10, 32, 100, 1000])
+def test_sum_of_booleans(n):
+    # GH 50347
+    df = DataFrame({"groupby_col": 1, "bool": [True] * n})
+    df["bool"] = df["bool"].eq(True)
+    result = df.groupby("groupby_col").sum()
+    expected = DataFrame({"bool": [n]}, index=Index([1], name="groupby_col"))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:invalid value encountered in remainder:RuntimeWarning"
+)
+@pytest.mark.parametrize("method", ["head", "tail", "nth", "first", "last"])
+def test_groupby_method_drop_na(method):
+    # GH 21755
+    df = DataFrame({"A": ["a", np.nan, "b", np.nan, "c"], "B": range(5)})
+
+    if method == "nth":
+        result = getattr(df.groupby("A"), method)(n=0)
+    else:
+        result = getattr(df.groupby("A"), method)()
+
+    if method in ["first", "last"]:
+        expected = DataFrame({"B": [0, 2, 4]}).set_index(
+            Series(["a", "b", "c"], name="A")
+        )
+    else:
+        expected = DataFrame(
+            {"A": ["a", "b", "c"], "B": [0, 2, 4]}, index=range(0, 6, 2)
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_reduce_period():
+    # GH#51040
+    pi = pd.period_range("2016-01-01", periods=100, freq="D")
+    grps = list(range(10)) * 10
+    ser = pi.to_series()
+    gb = ser.groupby(grps)
+
+    with pytest.raises(TypeError, match="Period type does not support sum operations"):
+        gb.sum()
+    with pytest.raises(
+        TypeError, match="Period type does not support cumsum operations"
+    ):
+        gb.cumsum()
+    with pytest.raises(TypeError, match="Period type does not support prod operations"):
+        gb.prod()
+    with pytest.raises(
+        TypeError, match="Period type does not support cumprod operations"
+    ):
+        gb.cumprod()
+
+    res = gb.max()
+    expected = ser[-10:]
+    expected.index = Index(range(10), dtype=int)
+    tm.assert_series_equal(res, expected)
+
+    res = gb.min()
+    expected = ser[:10]
+    expected.index = Index(range(10), dtype=int)
+    tm.assert_series_equal(res, expected)
+
+
+def test_obj_with_exclusions_duplicate_columns():
+    # GH#50806
+    df = DataFrame([[0, 1, 2, 3]])
+    df.columns = [0, 1, 2, 0]
+    gb = df.groupby(df[1])
+    result = gb._obj_with_exclusions
+    expected = df.take([0, 2, 3], axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_groupby_numeric_only_std_no_result(numeric_only):
+    # GH 51080
+    dicts_non_numeric = [{"a": "foo", "b": "bar"}, {"a": "car", "b": "dar"}]
+    df = DataFrame(dicts_non_numeric, dtype=object)
+    dfgb = df.groupby("a", as_index=False, sort=False)
+
+    if numeric_only:
+        result = dfgb.std(numeric_only=True)
+        expected_df = DataFrame(["foo", "car"], columns=["a"])
+        tm.assert_frame_equal(result, expected_df)
+    else:
+        with pytest.raises(
+            ValueError, match="could not convert string to float: 'bar'"
+        ):
+            dfgb.std(numeric_only=numeric_only)
+
+
+def test_grouping_with_categorical_interval_columns():
+    # GH#34164
+    df = DataFrame({"x": [0.1, 0.2, 0.3, -0.4, 0.5], "w": ["a", "b", "a", "c", "a"]})
+    qq = pd.qcut(df["x"], q=np.linspace(0, 1, 5))
+    result = df.groupby([qq, "w"], observed=False)["x"].agg("mean")
+    categorical_index_level_1 = Categorical(
+        [
+            Interval(-0.401, 0.1, closed="right"),
+            Interval(0.1, 0.2, closed="right"),
+            Interval(0.2, 0.3, closed="right"),
+            Interval(0.3, 0.5, closed="right"),
+        ],
+        ordered=True,
+    )
+    index_level_2 = ["a", "b", "c"]
+    mi = MultiIndex.from_product(
+        [categorical_index_level_1, index_level_2], names=["x", "w"]
+    )
+    expected = Series(
+        np.array(
+            [
+                0.1,
+                np.nan,
+                -0.4,
+                np.nan,
+                0.2,
+                np.nan,
+                0.3,
+                np.nan,
+                np.nan,
+                0.5,
+                np.nan,
+                np.nan,
+            ]
+        ),
+        index=mi,
+        name="x",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("bug_var", [1, "a"])
+def test_groupby_sum_on_nan_should_return_nan(bug_var):
+    # GH 24196
+    df = DataFrame({"A": [bug_var, bug_var, bug_var, np.nan]})
+    if isinstance(bug_var, str):
+        df = df.astype(object)
+    dfgb = df.groupby(lambda x: x)
+    result = dfgb.sum(min_count=1)
+
+    expected_df = DataFrame(
+        [bug_var, bug_var, bug_var, None], columns=["A"], dtype=df["A"].dtype
+    )
+    tm.assert_frame_equal(result, expected_df)
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        "count",
+        "corr",
+        "cummax",
+        "cummin",
+        "cumprod",
+        "describe",
+        "rank",
+        "quantile",
+        "diff",
+        "shift",
+        "all",
+        "any",
+        "idxmin",
+        "idxmax",
+        "ffill",
+        "bfill",
+        "pct_change",
+    ],
+)
+def test_groupby_selection_with_methods(df, method):
+    # some methods which require DatetimeIndex
+    rng = date_range("2014", periods=len(df))
+    df.index = rng
+
+    g = df.groupby(["A"])[["C"]]
+    g_exp = df[["C"]].groupby(df["A"])
+    # TODO check groupby with > 1 col ?
+
+    res = getattr(g, method)()
+    exp = getattr(g_exp, method)()
+
+    # should always be frames!
+    tm.assert_frame_equal(res, exp)
+
+
+def test_groupby_selection_other_methods(df):
+    # some methods which require DatetimeIndex
+    rng = date_range("2014", periods=len(df))
+    df.columns.name = "foo"
+    df.index = rng
+
+    g = df.groupby(["A"])[["C"]]
+    g_exp = df[["C"]].groupby(df["A"])
+
+    # methods which aren't just .foo()
+    tm.assert_frame_equal(g.apply(lambda x: x.sum()), g_exp.apply(lambda x: x.sum()))
+
+    tm.assert_frame_equal(g.resample("D").mean(), g_exp.resample("D").mean())
+    tm.assert_frame_equal(g.resample("D").ohlc(), g_exp.resample("D").ohlc())
+
+    tm.assert_frame_equal(
+        g.filter(lambda x: len(x) == 3), g_exp.filter(lambda x: len(x) == 3)
+    )
+
+
+def test_groupby_with_Time_Grouper(unit):
+    idx2 = to_datetime(
+        [
+            "2016-08-31 22:08:12.000",
+            "2016-08-31 22:09:12.200",
+            "2016-08-31 22:20:12.400",
+        ]
+    ).as_unit(unit)
+
+    test_data = DataFrame(
+        {"quant": [1.0, 1.0, 3.0], "quant2": [1.0, 1.0, 3.0], "time2": idx2}
+    )
+
+    time2 = date_range("2016-08-31 22:08:00", periods=13, freq="1min", unit=unit)
+    expected_output = DataFrame(
+        {
+            "time2": time2,
+            "quant": [1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1],
+            "quant2": [1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1],
+        }
+    )
+
+    gb = test_data.groupby(Grouper(key="time2", freq="1min"))
+    result = gb.count().reset_index()
+
+    tm.assert_frame_equal(result, expected_output)
+
+
+def test_groupby_series_with_datetimeindex_month_name():
+    # GH 48509
+    s = Series([0, 1, 0], index=date_range("2022-01-01", periods=3), name="jan")
+    result = s.groupby(s).count()
+    expected = Series([2, 1], name="jan")
+    expected.index.name = "jan"
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("test_series", [True, False])
+@pytest.mark.parametrize(
+    "kwarg, value, name, warn",
+    [
+        ("by", "a", 1, None),
+        ("by", ["a"], (1,), None),
+        ("level", 0, 1, None),
+        ("level", [0], (1,), None),
+    ],
+)
+def test_get_group_len_1_list_likes(test_series, kwarg, value, name, warn):
+    # GH#25971
+    obj = DataFrame({"b": [3, 4, 5]}, index=Index([1, 1, 2], name="a"))
+    if test_series:
+        obj = obj["b"]
+    gb = obj.groupby(**{kwarg: value})
+    result = gb.get_group(name)
+    if test_series:
+        expected = Series([3, 4], index=Index([1, 1], name="a"), name="b")
+    else:
+        expected = DataFrame({"b": [3, 4]}, index=Index([1, 1], name="a"))
+    tm.assert_equal(result, expected)
+
+
+def test_groupby_ngroup_with_nan():
+    # GH#50100
+    df = DataFrame({"a": Categorical([np.nan]), "b": [1]})
+    result = df.groupby(["a", "b"], dropna=False, observed=False).ngroup()
+    expected = Series([0])
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_ffill_with_duplicated_index():
+    # GH#43412
+    df = DataFrame({"a": [1, 2, 3, 4, np.nan, np.nan]}, index=[0, 1, 2, 0, 1, 2])
+
+    result = df.groupby(level=0).ffill()
+    expected = DataFrame({"a": [1, 2, 3, 4, 2, 3]}, index=[0, 1, 2, 0, 1, 2])
+    tm.assert_frame_equal(result, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize("test_series", [True, False])
+def test_decimal_na_sort(test_series):
+    # GH#54847
+    # We catch both TypeError and decimal.InvalidOperation exceptions in safe_sort.
+    # If this next assert raises, we can just catch TypeError
+    assert not isinstance(decimal.InvalidOperation, TypeError)
+    df = DataFrame(
+        {
+            "key": [Decimal(1), Decimal(1), None, None],
+            "value": [Decimal(2), Decimal(3), Decimal(4), Decimal(5)],
+        }
+    )
+    gb = df.groupby("key", dropna=False)
+    if test_series:
+        gb = gb["value"]
+    result = gb._grouper.result_index
+    expected = Index([Decimal(1), None], name="key")
+    tm.assert_index_equal(result, expected)
+
+
+def test_groupby_dropna_with_nunique_unique():
+    # GH#42016
+    df = [[1, 1, 1, "A"], [1, None, 1, "A"], [1, None, 2, "A"], [1, None, 3, "A"]]
+    df_dropna = DataFrame(df, columns=["a", "b", "c", "partner"])
+    result = df_dropna.groupby(["a", "b", "c"], dropna=False).agg(
+        {"partner": ["nunique", "unique"]}
+    )
+
+    index = MultiIndex.from_tuples(
+        [(1, 1.0, 1), (1, np.nan, 1), (1, np.nan, 2), (1, np.nan, 3)],
+        names=["a", "b", "c"],
+    )
+    columns = MultiIndex.from_tuples([("partner", "nunique"), ("partner", "unique")])
+    expected = DataFrame(
+        [(1, ["A"]), (1, ["A"]), (1, ["A"]), (1, ["A"])], index=index, columns=columns
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_agg_namedagg_with_duplicate_columns():
+    # GH#58446
+    df = DataFrame(
+        {
+            "col1": [2, 1, 1, 0, 2, 0],
+            "col2": [4, 5, 36, 7, 4, 5],
+            "col3": [3.1, 8.0, 12, 10, 4, 1.1],
+            "col4": [17, 3, 16, 15, 5, 6],
+            "col5": [-1, 3, -1, 3, -2, -1],
+        }
+    )
+
+    result = df.groupby(by=["col1", "col1", "col2"], as_index=False).agg(
+        new_col=pd.NamedAgg(column="col1", aggfunc="min"),
+        new_col1=pd.NamedAgg(column="col1", aggfunc="max"),
+        new_col2=pd.NamedAgg(column="col2", aggfunc="count"),
+    )
+
+    expected = DataFrame(
+        {
+            "col1": [0, 0, 1, 1, 2],
+            "col2": [5, 7, 5, 36, 4],
+            "new_col": [0, 0, 1, 1, 2],
+            "new_col1": [0, 0, 1, 1, 2],
+            "new_col2": [1, 1, 1, 1, 2],
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_multi_index_codes():
+    # GH#54347
+    df = DataFrame(
+        {"A": [1, 2, 3, 4], "B": [1, float("nan"), 2, float("nan")], "C": [2, 4, 6, 8]}
+    )
+    df_grouped = df.groupby(["A", "B"], dropna=False).sum()
+
+    index = df_grouped.index
+    tm.assert_index_equal(index, MultiIndex.from_frame(index.to_frame()))
diff --git a/pandas/pandas/tests/groupby/test_groupby_dropna.py b/pandas/pandas/tests/groupby/test_groupby_dropna.py
new file mode 100644
index 0000000000000..8c4ab42b7be7a
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_groupby_dropna.py
@@ -0,0 +1,684 @@
+import numpy as np
+import pytest
+
+from pandas.compat.pyarrow import pa_version_under10p1
+
+from pandas.core.dtypes.missing import na_value_for_dtype
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+@pytest.mark.parametrize(
+    "dropna, tuples, outputs",
+    [
+        (
+            True,
+            [["A", "B"], ["B", "A"]],
+            {"c": [13.0, 123.23], "d": [13.0, 123.0], "e": [13.0, 1.0]},
+        ),
+        (
+            False,
+            [["A", "B"], ["A", np.nan], ["B", "A"]],
+            {
+                "c": [13.0, 12.3, 123.23],
+                "d": [13.0, 233.0, 123.0],
+                "e": [13.0, 12.0, 1.0],
+            },
+        ),
+    ],
+)
+def test_groupby_dropna_multi_index_dataframe_nan_in_one_group(
+    dropna, tuples, outputs, nulls_fixture
+):
+    # GH 3729 this is to test that NA is in one group
+    df_list = [
+        ["A", "B", 12, 12, 12],
+        ["A", nulls_fixture, 12.3, 233.0, 12],
+        ["B", "A", 123.23, 123, 1],
+        ["A", "B", 1, 1, 1.0],
+    ]
+    df = pd.DataFrame(df_list, columns=["a", "b", "c", "d", "e"])
+    grouped = df.groupby(["a", "b"], dropna=dropna).sum()
+
+    mi = pd.MultiIndex.from_tuples(tuples, names=list("ab"))
+
+    # Since right now, by default MI will drop NA from levels when we create MI
+    # via `from_*`, so we need to add NA for level manually afterwards.
+    if not dropna:
+        mi = mi.set_levels(["A", "B", np.nan], level="b")
+    expected = pd.DataFrame(outputs, index=mi)
+
+    tm.assert_frame_equal(grouped, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, tuples, outputs",
+    [
+        (
+            True,
+            [["A", "B"], ["B", "A"]],
+            {"c": [12.0, 123.23], "d": [12.0, 123.0], "e": [12.0, 1.0]},
+        ),
+        (
+            False,
+            [["A", "B"], ["A", np.nan], ["B", "A"], [np.nan, "B"]],
+            {
+                "c": [12.0, 13.3, 123.23, 1.0],
+                "d": [12.0, 234.0, 123.0, 1.0],
+                "e": [12.0, 13.0, 1.0, 1.0],
+            },
+        ),
+    ],
+)
+def test_groupby_dropna_multi_index_dataframe_nan_in_two_groups(
+    dropna, tuples, outputs, nulls_fixture, nulls_fixture2
+):
+    # GH 3729 this is to test that NA in different groups with different representations
+    df_list = [
+        ["A", "B", 12, 12, 12],
+        ["A", nulls_fixture, 12.3, 233.0, 12],
+        ["B", "A", 123.23, 123, 1],
+        [nulls_fixture2, "B", 1, 1, 1.0],
+        ["A", nulls_fixture2, 1, 1, 1.0],
+    ]
+    df = pd.DataFrame(df_list, columns=["a", "b", "c", "d", "e"])
+    grouped = df.groupby(["a", "b"], dropna=dropna).sum()
+
+    mi = pd.MultiIndex.from_tuples(tuples, names=list("ab"))
+
+    # Since right now, by default MI will drop NA from levels when we create MI
+    # via `from_*`, so we need to add NA for level manually afterwards.
+    if not dropna:
+        mi = mi.set_levels([["A", "B", np.nan], ["A", "B", np.nan]])
+    expected = pd.DataFrame(outputs, index=mi)
+
+    tm.assert_frame_equal(grouped, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, idx, outputs",
+    [
+        (True, ["A", "B"], {"b": [123.23, 13.0], "c": [123.0, 13.0], "d": [1.0, 13.0]}),
+        (
+            False,
+            ["A", "B", np.nan],
+            {
+                "b": [123.23, 13.0, 12.3],
+                "c": [123.0, 13.0, 233.0],
+                "d": [1.0, 13.0, 12.0],
+            },
+        ),
+    ],
+)
+def test_groupby_dropna_normal_index_dataframe(dropna, idx, outputs):
+    # GH 3729
+    df_list = [
+        ["B", 12, 12, 12],
+        [None, 12.3, 233.0, 12],
+        ["A", 123.23, 123, 1],
+        ["B", 1, 1, 1.0],
+    ]
+    df = pd.DataFrame(df_list, columns=["a", "b", "c", "d"])
+    grouped = df.groupby("a", dropna=dropna).sum()
+
+    expected = pd.DataFrame(outputs, index=pd.Index(idx, name="a"))
+
+    tm.assert_frame_equal(grouped, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, idx, expected",
+    [
+        (True, ["a", "a", "b", np.nan], pd.Series([3, 3], index=["a", "b"])),
+        (
+            False,
+            ["a", "a", "b", np.nan],
+            pd.Series([3, 3, 3], index=["a", "b", np.nan]),
+        ),
+    ],
+)
+def test_groupby_dropna_series_level(dropna, idx, expected):
+    ser = pd.Series([1, 2, 3, 3], index=idx)
+
+    result = ser.groupby(level=0, dropna=dropna).sum()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, expected",
+    [
+        (True, pd.Series([210.0, 350.0], index=["a", "b"], name="Max Speed")),
+        (
+            False,
+            pd.Series([210.0, 350.0, 20.0], index=["a", "b", np.nan], name="Max Speed"),
+        ),
+    ],
+)
+def test_groupby_dropna_series_by(dropna, expected):
+    ser = pd.Series(
+        [390.0, 350.0, 30.0, 20.0],
+        index=["Falcon", "Falcon", "Parrot", "Parrot"],
+        name="Max Speed",
+    )
+
+    result = ser.groupby(["a", "b", "a", np.nan], dropna=dropna).mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_grouper_dropna_propagation(dropna):
+    # GH 36604
+    df = pd.DataFrame({"A": [0, 0, 1, None], "B": [1, 2, 3, None]})
+    gb = df.groupby("A", dropna=dropna)
+    assert gb._grouper.dropna == dropna
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        pd.RangeIndex(0, 4),
+        list("abcd"),
+        pd.MultiIndex.from_product([(1, 2), ("R", "B")], names=["num", "col"]),
+    ],
+)
+def test_groupby_dataframe_slice_then_transform(dropna, index):
+    # GH35014 & GH35612
+    expected_data = {"B": [2, 2, 1, np.nan if dropna else 1]}
+
+    df = pd.DataFrame({"A": [0, 0, 1, None], "B": [1, 2, 3, None]}, index=index)
+    gb = df.groupby("A", dropna=dropna)
+
+    result = gb.transform(len)
+    expected = pd.DataFrame(expected_data, index=index)
+    tm.assert_frame_equal(result, expected)
+
+    result = gb[["B"]].transform(len)
+    expected = pd.DataFrame(expected_data, index=index)
+    tm.assert_frame_equal(result, expected)
+
+    result = gb["B"].transform(len)
+    expected = pd.Series(expected_data["B"], index=index, name="B")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, tuples, outputs",
+    [
+        (
+            True,
+            [["A", "B"], ["B", "A"]],
+            {"c": [13.0, 123.23], "d": [12.0, 123.0], "e": [1.0, 1.0]},
+        ),
+        (
+            False,
+            [["A", "B"], ["A", np.nan], ["B", "A"]],
+            {
+                "c": [13.0, 12.3, 123.23],
+                "d": [12.0, 233.0, 123.0],
+                "e": [1.0, 12.0, 1.0],
+            },
+        ),
+    ],
+)
+def test_groupby_dropna_multi_index_dataframe_agg(dropna, tuples, outputs):
+    # GH 3729
+    df_list = [
+        ["A", "B", 12, 12, 12],
+        ["A", None, 12.3, 233.0, 12],
+        ["B", "A", 123.23, 123, 1],
+        ["A", "B", 1, 1, 1.0],
+    ]
+    df = pd.DataFrame(df_list, columns=["a", "b", "c", "d", "e"])
+    agg_dict = {"c": "sum", "d": "max", "e": "min"}
+    grouped = df.groupby(["a", "b"], dropna=dropna).agg(agg_dict)
+
+    mi = pd.MultiIndex.from_tuples(tuples, names=list("ab"))
+
+    # Since right now, by default MI will drop NA from levels when we create MI
+    # via `from_*`, so we need to add NA for level manually afterwards.
+    if not dropna:
+        mi = mi.set_levels(["A", "B", np.nan], level="b")
+    expected = pd.DataFrame(outputs, index=mi)
+
+    tm.assert_frame_equal(grouped, expected)
+
+
+@pytest.mark.arm_slow
+@pytest.mark.parametrize(
+    "datetime1, datetime2",
+    [
+        (pd.Timestamp("2020-01-01"), pd.Timestamp("2020-02-01")),
+        (pd.Timedelta("-2 days"), pd.Timedelta("-1 days")),
+        (pd.Period("2020-01-01"), pd.Period("2020-02-01")),
+    ],
+)
+@pytest.mark.parametrize("dropna, values", [(True, [12, 3]), (False, [12, 3, 6])])
+def test_groupby_dropna_datetime_like_data(
+    dropna, values, datetime1, datetime2, unique_nulls_fixture, unique_nulls_fixture2
+):
+    # 3729
+    df = pd.DataFrame(
+        {
+            "values": [1, 2, 3, 4, 5, 6],
+            "dt": [
+                datetime1,
+                unique_nulls_fixture,
+                datetime2,
+                unique_nulls_fixture2,
+                datetime1,
+                datetime1,
+            ],
+        }
+    )
+
+    if dropna:
+        indexes = [datetime1, datetime2]
+    else:
+        indexes = [datetime1, datetime2, np.nan]
+
+    grouped = df.groupby("dt", dropna=dropna).agg({"values": "sum"})
+    expected = pd.DataFrame({"values": values}, index=pd.Index(indexes, name="dt"))
+
+    tm.assert_frame_equal(grouped, expected)
+
+
+@pytest.mark.parametrize(
+    "dropna, data, selected_data, levels",
+    [
+        pytest.param(
+            False,
+            {"groups": ["a", "a", "b", np.nan], "values": [10, 10, 20, 30]},
+            {"values": [0, 1, 0, 0]},
+            ["a", "b", np.nan],
+            id="dropna_false_has_nan",
+        ),
+        pytest.param(
+            True,
+            {"groups": ["a", "a", "b", np.nan], "values": [10, 10, 20, 30]},
+            {"values": [0, 1, 0]},
+            None,
+            id="dropna_true_has_nan",
+        ),
+        pytest.param(
+            # no nan in "groups"; dropna=True|False should be same.
+            False,
+            {"groups": ["a", "a", "b", "c"], "values": [10, 10, 20, 30]},
+            {"values": [0, 1, 0, 0]},
+            None,
+            id="dropna_false_no_nan",
+        ),
+        pytest.param(
+            # no nan in "groups"; dropna=True|False should be same.
+            True,
+            {"groups": ["a", "a", "b", "c"], "values": [10, 10, 20, 30]},
+            {"values": [0, 1, 0, 0]},
+            None,
+            id="dropna_true_no_nan",
+        ),
+    ],
+)
+def test_groupby_apply_with_dropna_for_multi_index(dropna, data, selected_data, levels):
+    # GH 35889
+
+    df = pd.DataFrame(data)
+    gb = df.groupby("groups", dropna=dropna)
+    result = gb.apply(lambda grp: pd.DataFrame({"values": range(len(grp))}))
+
+    mi_tuples = tuple(zip(data["groups"], selected_data["values"]))
+    mi = pd.MultiIndex.from_tuples(mi_tuples, names=["groups", None])
+    # Since right now, by default MI will drop NA from levels when we create MI
+    # via `from_*`, so we need to add NA for level manually afterwards.
+    if not dropna and levels:
+        mi = mi.set_levels(levels, level="groups")
+
+    expected = pd.DataFrame(selected_data, index=mi)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("input_index", [None, ["a"], ["a", "b"]])
+@pytest.mark.parametrize("keys", [["a"], ["a", "b"]])
+@pytest.mark.parametrize("series", [True, False])
+def test_groupby_dropna_with_multiindex_input(input_index, keys, series):
+    # GH#46783
+    obj = pd.DataFrame(
+        {
+            "a": [1, np.nan],
+            "b": [1, 1],
+            "c": [2, 3],
+        }
+    )
+
+    expected = obj.set_index(keys)
+    if series:
+        expected = expected["c"]
+    elif input_index == ["a", "b"] and keys == ["a"]:
+        # Column b should not be aggregated
+        expected = expected[["c"]]
+
+    if input_index is not None:
+        obj = obj.set_index(input_index)
+    gb = obj.groupby(keys, dropna=False)
+    if series:
+        gb = gb["c"]
+    result = gb.sum()
+
+    tm.assert_equal(result, expected)
+
+
+def test_groupby_nan_included():
+    # GH 35646
+    data = {"group": ["g1", np.nan, "g1", "g2", np.nan], "B": [0, 1, 2, 3, 4]}
+    df = pd.DataFrame(data)
+    grouped = df.groupby("group", dropna=False)
+    result = grouped.indices
+    dtype = np.intp
+    expected = {
+        "g1": np.array([0, 2], dtype=dtype),
+        "g2": np.array([3], dtype=dtype),
+        np.nan: np.array([1, 4], dtype=dtype),
+    }
+    for result_values, expected_values in zip(result.values(), expected.values()):
+        tm.assert_numpy_array_equal(result_values, expected_values)
+    assert np.isnan(list(result.keys())[2])
+    assert list(result.keys())[0:2] == ["g1", "g2"]
+
+
+def test_groupby_drop_nan_with_multi_index():
+    # GH 39895
+    df = pd.DataFrame([[np.nan, 0, 1]], columns=["a", "b", "c"])
+    df = df.set_index(["a", "b"])
+    result = df.groupby(["a", "b"], dropna=False).first()
+    expected = df
+    tm.assert_frame_equal(result, expected)
+
+
+# sequence_index enumerates all strings made up of x, y, z of length 4
+@pytest.mark.parametrize("sequence_index", range(3**4))
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        None,
+        "UInt8",
+        "Int8",
+        "UInt16",
+        "Int16",
+        "UInt32",
+        "Int32",
+        "UInt64",
+        "Int64",
+        "Float32",
+        "Float64",
+        "category",
+        "string",
+        pytest.param(
+            "string[pyarrow]",
+            marks=pytest.mark.skipif(
+                pa_version_under10p1, reason="pyarrow is not installed"
+            ),
+        ),
+        "datetime64[ns]",
+        "period[D]",
+        "Sparse[float]",
+    ],
+)
+@pytest.mark.parametrize("test_series", [True, False])
+def test_no_sort_keep_na(sequence_index, dtype, test_series, as_index):
+    # GH#46584, GH#48794
+
+    # Convert sequence_index into a string sequence, e.g. 5 becomes "xxyz"
+    # This sequence is used for the grouper.
+    sequence = "".join(
+        [{0: "x", 1: "y", 2: "z"}[sequence_index // (3**k) % 3] for k in range(4)]
+    )
+
+    # Unique values to use for grouper, depends on dtype
+    if dtype in ("string", "string[pyarrow]"):
+        uniques = {"x": "x", "y": "y", "z": pd.NA}
+    elif dtype in ("datetime64[ns]", "period[D]"):
+        uniques = {"x": "2016-01-01", "y": "2017-01-01", "z": pd.NA}
+    else:
+        uniques = {"x": 1, "y": 2, "z": np.nan}
+
+    df = pd.DataFrame(
+        {
+            "key": pd.Series([uniques[label] for label in sequence], dtype=dtype),
+            "a": [0, 1, 2, 3],
+        }
+    )
+    gb = df.groupby("key", dropna=False, sort=False, as_index=as_index, observed=False)
+    if test_series:
+        gb = gb["a"]
+    result = gb.sum()
+
+    # Manually compute the groupby sum, use the labels "x", "y", and "z" to avoid
+    # issues with hashing np.nan
+    summed = {}
+    for idx, label in enumerate(sequence):
+        summed[label] = summed.get(label, 0) + idx
+    if dtype == "category":
+        index = pd.CategoricalIndex(
+            [uniques[e] for e in summed],
+            df["key"].cat.categories,
+            name="key",
+        )
+    elif isinstance(dtype, str) and dtype.startswith("Sparse"):
+        index = pd.Index(
+            pd.array([uniques[label] for label in summed], dtype=dtype), name="key"
+        )
+    else:
+        index = pd.Index([uniques[label] for label in summed], dtype=dtype, name="key")
+    expected = pd.Series(summed.values(), index=index, name="a", dtype=None)
+    if not test_series:
+        expected = expected.to_frame()
+    if not as_index:
+        expected = expected.reset_index()
+        if dtype is not None and dtype.startswith("Sparse"):
+            expected["key"] = expected["key"].astype(dtype)
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("test_series", [True, False])
+@pytest.mark.parametrize("dtype", [object, None])
+def test_null_is_null_for_dtype(
+    sort, dtype, nulls_fixture, nulls_fixture2, test_series
+):
+    # GH#48506 - groups should always result in using the null for the dtype
+    df = pd.DataFrame({"a": [1, 2]})
+    groups = pd.Series([nulls_fixture, nulls_fixture2], dtype=dtype)
+    obj = df["a"] if test_series else df
+    gb = obj.groupby(groups, dropna=False, sort=sort)
+    result = gb.sum()
+    index = pd.Index([na_value_for_dtype(groups.dtype)])
+    expected = pd.DataFrame({"a": [3]}, index=index)
+    if test_series:
+        tm.assert_series_equal(result, expected["a"])
+    else:
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_kind", ["range", "single", "multi"])
+def test_categorical_reducers(reduction_func, observed, sort, as_index, index_kind):
+    # Ensure there is at least one null value by appending to the end
+    values = np.append(np.random.default_rng(2).choice([1, 2, None], size=19), None)
+    df = pd.DataFrame(
+        {"x": pd.Categorical(values, categories=[1, 2, 3]), "y": range(20)}
+    )
+
+    # Strategy: Compare to dropna=True by filling null values with a new code
+    df_filled = df.copy()
+    df_filled["x"] = pd.Categorical(values, categories=[1, 2, 3, 4]).fillna(4)
+
+    if index_kind == "range":
+        keys = ["x"]
+    elif index_kind == "single":
+        keys = ["x"]
+        df = df.set_index("x")
+        df_filled = df_filled.set_index("x")
+    else:
+        keys = ["x", "x2"]
+        df["x2"] = df["x"]
+        df = df.set_index(["x", "x2"])
+        df_filled["x2"] = df_filled["x"]
+        df_filled = df_filled.set_index(["x", "x2"])
+    args = get_groupby_method_args(reduction_func, df)
+    args_filled = get_groupby_method_args(reduction_func, df_filled)
+    if reduction_func == "corrwith" and index_kind == "range":
+        # Don't include the grouping columns so we can call reset_index
+        args = (args[0].drop(columns=keys),)
+        args_filled = (args_filled[0].drop(columns=keys),)
+
+    gb_keepna = df.groupby(
+        keys, dropna=False, observed=observed, sort=sort, as_index=as_index
+    )
+
+    if not observed and reduction_func in ["idxmin", "idxmax"]:
+        with pytest.raises(
+            ValueError, match="empty group due to unobserved categories"
+        ):
+            getattr(gb_keepna, reduction_func)(*args)
+        return
+
+    gb_filled = df_filled.groupby(keys, observed=observed, sort=sort, as_index=True)
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        msg = ""
+    with tm.assert_produces_warning(warn, match=msg):
+        expected = getattr(gb_filled, reduction_func)(*args_filled).reset_index()
+    expected["x"] = expected["x"].cat.remove_categories([4])
+    if index_kind == "multi":
+        expected["x2"] = expected["x2"].cat.remove_categories([4])
+    if as_index:
+        if index_kind == "multi":
+            expected = expected.set_index(["x", "x2"])
+        else:
+            expected = expected.set_index("x")
+    if reduction_func in ("idxmax", "idxmin") and index_kind != "range":
+        # expected was computed with a RangeIndex; need to translate to index values
+        values = expected["y"].values.tolist()
+        if index_kind == "single":
+            values = [np.nan if e == 4 else e for e in values]
+            expected["y"] = pd.Categorical(values, categories=[1, 2, 3])
+        else:
+            values = [(np.nan, np.nan) if e == (4, 4) else e for e in values]
+            expected["y"] = values
+    if reduction_func == "size":
+        # size, unlike other methods, has the desired behavior in GH#49519
+        expected = expected.rename(columns={0: "size"})
+        if as_index:
+            expected = expected["size"].rename(None)
+
+    if reduction_func == "corrwith":
+        warn = FutureWarning
+        msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        msg = ""
+    with tm.assert_produces_warning(warn, match=msg):
+        result = getattr(gb_keepna, reduction_func)(*args)
+
+    # size will return a Series, others are DataFrame
+    tm.assert_equal(result, expected)
+
+
+def test_categorical_transformers(transformation_func, observed, sort, as_index):
+    # GH#36327
+    values = np.append(np.random.default_rng(2).choice([1, 2, None], size=19), None)
+    df = pd.DataFrame(
+        {"x": pd.Categorical(values, categories=[1, 2, 3]), "y": range(20)}
+    )
+    args = get_groupby_method_args(transformation_func, df)
+
+    # Compute result for null group
+    null_group_values = df[df["x"].isnull()]["y"]
+    if transformation_func == "cumcount":
+        null_group_data = list(range(len(null_group_values)))
+    elif transformation_func == "ngroup":
+        if sort:
+            if observed:
+                na_group = df["x"].nunique(dropna=False) - 1
+            else:
+                # TODO: Should this be 3?
+                na_group = df["x"].nunique(dropna=False) - 1
+        else:
+            na_group = df.iloc[: null_group_values.index[0]]["x"].nunique()
+        null_group_data = len(null_group_values) * [na_group]
+    else:
+        null_group_data = getattr(null_group_values, transformation_func)(*args)
+    null_group_result = pd.DataFrame({"y": null_group_data})
+
+    gb_keepna = df.groupby(
+        "x", dropna=False, observed=observed, sort=sort, as_index=as_index
+    )
+    gb_dropna = df.groupby("x", dropna=True, observed=observed, sort=sort)
+
+    result = getattr(gb_keepna, transformation_func)(*args)
+    expected = getattr(gb_dropna, transformation_func)(*args)
+
+    for iloc, value in zip(
+        df[df["x"].isnull()].index.tolist(), null_group_result.values.ravel()
+    ):
+        if expected.ndim == 1:
+            expected.iloc[iloc] = value
+        else:
+            expected.iloc[iloc, 0] = value
+    if transformation_func == "ngroup":
+        expected[df["x"].notnull() & expected.ge(na_group)] += 1
+    if transformation_func not in ("rank", "diff", "pct_change", "shift"):
+        expected = expected.astype("int64")
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["head", "tail"])
+def test_categorical_head_tail(method, observed, sort, as_index):
+    # GH#36327
+    values = np.random.default_rng(2).choice([1, 2, None], 30)
+    df = pd.DataFrame(
+        {"x": pd.Categorical(values, categories=[1, 2, 3]), "y": range(len(values))}
+    )
+    gb = df.groupby("x", dropna=False, observed=observed, sort=sort, as_index=as_index)
+    result = getattr(gb, method)()
+
+    if method == "tail":
+        values = values[::-1]
+    # Take the top 5 values from each group
+    mask = (
+        ((values == 1) & ((values == 1).cumsum() <= 5))
+        | ((values == 2) & ((values == 2).cumsum() <= 5))
+        # flake8 doesn't like the vectorized check for None, thinks we should use `is`
+        | ((values == None) & ((values == None).cumsum() <= 5))  # noqa: E711
+    )
+    if method == "tail":
+        mask = mask[::-1]
+    expected = df[mask]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_agg():
+    # GH#36327
+    values = np.random.default_rng(2).choice([1, 2, None], 30)
+    df = pd.DataFrame(
+        {"x": pd.Categorical(values, categories=[1, 2, 3]), "y": range(len(values))}
+    )
+    gb = df.groupby("x", dropna=False, observed=False)
+    result = gb.agg(lambda x: x.sum())
+    expected = gb.sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_transform():
+    # GH#36327
+    values = np.random.default_rng(2).choice([1, 2, None], 30)
+    df = pd.DataFrame(
+        {"x": pd.Categorical(values, categories=[1, 2, 3]), "y": range(len(values))}
+    )
+    gb = df.groupby("x", dropna=False, observed=False)
+    result = gb.transform(lambda x: x.sum())
+    expected = gb.transform("sum")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_groupby_subclass.py b/pandas/pandas/tests/groupby/test_groupby_subclass.py
new file mode 100644
index 0000000000000..3ee9c9ea0c7fd
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_groupby_subclass.py
@@ -0,0 +1,121 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+)
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        tm.SubclassedDataFrame({"A": np.arange(0, 10)}),
+        tm.SubclassedSeries(np.arange(0, 10), name="A"),
+    ],
+)
+def test_groupby_preserves_subclass(obj, groupby_func):
+    # GH28330 -- preserve subclass through groupby operations
+
+    if isinstance(obj, Series) and groupby_func in {"corrwith"}:
+        pytest.skip(f"Not applicable for Series and {groupby_func}")
+
+    grouped = obj.groupby(np.arange(0, 10))
+
+    # Groups should preserve subclass type
+    assert isinstance(grouped.get_group(0), type(obj))
+
+    args = get_groupby_method_args(groupby_func, obj)
+
+    warn = FutureWarning if groupby_func == "corrwith" else None
+    msg = f"{type(grouped).__name__}.corrwith is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        result1 = getattr(grouped, groupby_func)(*args)
+    with tm.assert_produces_warning(warn, match=msg):
+        result2 = grouped.agg(groupby_func, *args)
+
+    # Reduction or transformation kernels should preserve type
+    slices = {"ngroup", "cumcount", "size"}
+    if isinstance(obj, DataFrame) and groupby_func in slices:
+        assert isinstance(result1, tm.SubclassedSeries)
+    else:
+        assert isinstance(result1, type(obj))
+
+    # Confirm .agg() groupby operations return same results
+    if isinstance(result1, DataFrame):
+        tm.assert_frame_equal(result1, result2)
+    else:
+        tm.assert_series_equal(result1, result2)
+
+
+def test_groupby_preserves_metadata():
+    # GH-37343
+    custom_df = tm.SubclassedDataFrame({"a": [1, 2, 3], "b": [1, 1, 2], "c": [7, 8, 9]})
+    assert "testattr" in custom_df._metadata
+    custom_df.testattr = "hello"
+    for _, group_df in custom_df.groupby("c"):
+        assert group_df.testattr == "hello"
+
+    # GH-45314
+    def func(group):
+        assert isinstance(group, tm.SubclassedDataFrame)
+        assert hasattr(group, "testattr")
+        assert group.testattr == "hello"
+        return group.testattr
+
+    result = custom_df.groupby("c").apply(func)
+    expected = tm.SubclassedSeries(["hello"] * 3, index=Index([7, 8, 9], name="c"))
+    tm.assert_series_equal(result, expected)
+
+    result = custom_df.groupby("c").apply(func)
+    tm.assert_series_equal(result, expected)
+
+    # https://github.com/pandas-dev/pandas/pull/56761
+    result = custom_df.groupby("c")[["a", "b"]].apply(func)
+    tm.assert_series_equal(result, expected)
+
+    def func2(group):
+        assert isinstance(group, tm.SubclassedSeries)
+        assert hasattr(group, "testattr")
+        return group.testattr
+
+    custom_series = tm.SubclassedSeries([1, 2, 3])
+    custom_series.testattr = "hello"
+    result = custom_series.groupby(custom_df["c"]).apply(func2)
+    tm.assert_series_equal(result, expected)
+    result = custom_series.groupby(custom_df["c"]).agg(func2)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("obj", [DataFrame, tm.SubclassedDataFrame])
+def test_groupby_resample_preserves_subclass(obj):
+    # GH28330 -- preserve subclass through groupby.resample()
+
+    df = obj(
+        {
+            "Buyer": Series("Carl Carl Carl Carl Joe Carl".split(), dtype=object),
+            "Quantity": [18, 3, 5, 1, 9, 3],
+            "Date": [
+                datetime(2013, 9, 1, 13, 0),
+                datetime(2013, 9, 1, 13, 5),
+                datetime(2013, 10, 1, 20, 0),
+                datetime(2013, 10, 3, 10, 0),
+                datetime(2013, 12, 2, 12, 0),
+                datetime(2013, 9, 2, 14, 0),
+            ],
+        }
+    )
+    df = df.set_index("Date")
+
+    # Confirm groupby.resample() preserves dataframe type
+    result = df.groupby("Buyer").resample("5D").sum()
+    assert isinstance(result, obj)
diff --git a/pandas/pandas/tests/groupby/test_grouping.py b/pandas/pandas/tests/groupby/test_grouping.py
new file mode 100644
index 0000000000000..53e9c53efebf7
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_grouping.py
@@ -0,0 +1,1197 @@
+"""
+test where we are determining what we are grouping, or getting groups
+"""
+
+from datetime import (
+    date,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.errors import SpecificationError
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Grouper,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core.groupby.grouper import Grouping
+
+# selection
+# --------------------------------
+
+
+class TestSelection:
+    def test_select_bad_cols(self):
+        df = DataFrame([[1, 2]], columns=["A", "B"])
+        g = df.groupby("A")
+        with pytest.raises(KeyError, match="\"Columns not found: 'C'\""):
+            g[["C"]]
+
+        with pytest.raises(KeyError, match="^[^A]+$"):
+            # A should not be referenced as a bad column...
+            # will have to rethink regex if you change message!
+            g[["A", "C"]]
+
+    def test_groupby_duplicated_column_errormsg(self):
+        # GH7511
+        df = DataFrame(
+            columns=["A", "B", "A", "C"], data=[range(4), range(2, 6), range(0, 8, 2)]
+        )
+
+        msg = "Grouper for 'A' not 1-dimensional"
+        with pytest.raises(ValueError, match=msg):
+            df.groupby("A")
+        with pytest.raises(ValueError, match=msg):
+            df.groupby(["A", "B"])
+
+        grouped = df.groupby("B")
+        c = grouped.count()
+        assert c.columns.nlevels == 1
+        assert c.columns.size == 3
+
+    def test_column_select_via_attr(self, df):
+        result = df.groupby("A").C.sum()
+        expected = df.groupby("A")["C"].sum()
+        tm.assert_series_equal(result, expected)
+
+        df["mean"] = 1.5
+        result = df.groupby("A").mean(numeric_only=True)
+        expected = df.groupby("A")[["C", "D", "mean"]].agg("mean")
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_list_of_columns(self):
+        df = DataFrame(
+            {
+                "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+                "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+                "C": np.random.default_rng(2).standard_normal(8),
+                "D": np.random.default_rng(2).standard_normal(8),
+                "E": np.random.default_rng(2).standard_normal(8),
+            }
+        )
+
+        result = df.groupby("A")[["C", "D"]].mean()
+        result2 = df.groupby("A")[df.columns[2:4]].mean()
+
+        expected = df.loc[:, ["A", "C", "D"]].groupby("A").mean()
+
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_getitem_numeric_column_names(self):
+        # GH #13731
+        df = DataFrame(
+            {
+                0: list("abcd") * 2,
+                2: np.random.default_rng(2).standard_normal(8),
+                4: np.random.default_rng(2).standard_normal(8),
+                6: np.random.default_rng(2).standard_normal(8),
+            }
+        )
+        result = df.groupby(0)[df.columns[1:3]].mean()
+        result2 = df.groupby(0)[[2, 4]].mean()
+
+        expected = df.loc[:, [0, 2, 4]].groupby(0).mean()
+
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        # per GH 23566 enforced deprecation raises a ValueError
+        with pytest.raises(ValueError, match="Cannot subset columns with a tuple"):
+            df.groupby(0)[2, 4].mean()
+
+    def test_getitem_single_tuple_of_columns_raises(self, df):
+        # per GH 23566 enforced deprecation raises a ValueError
+        with pytest.raises(ValueError, match="Cannot subset columns with a tuple"):
+            df.groupby("A")["C", "D"].mean()
+
+    def test_getitem_single_column(self):
+        df = DataFrame(
+            {
+                "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+                "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+                "C": np.random.default_rng(2).standard_normal(8),
+                "D": np.random.default_rng(2).standard_normal(8),
+                "E": np.random.default_rng(2).standard_normal(8),
+            }
+        )
+
+        result = df.groupby("A")["C"].mean()
+
+        as_frame = df.loc[:, ["A", "C"]].groupby("A").mean()
+        as_series = as_frame.iloc[:, 0]
+        expected = as_series
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "func", [lambda x: x.sum(), lambda x: x.agg(lambda y: y.sum())]
+    )
+    def test_getitem_from_grouper(self, func):
+        # GH 50383
+        df = DataFrame({"a": [1, 1, 2], "b": 3, "c": 4, "d": 5})
+        gb = df.groupby(["a", "b"])[["a", "c"]]
+
+        idx = MultiIndex.from_tuples([(1, 3), (2, 3)], names=["a", "b"])
+        expected = DataFrame({"a": [2, 2], "c": [8, 4]}, index=idx)
+        result = func(gb)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_indices_grouped_by_tuple_with_lambda(self):
+        # GH 36158
+        df = DataFrame(
+            {
+                "Tuples": (
+                    (x, y)
+                    for x in [0, 1]
+                    for y in np.random.default_rng(2).integers(3, 5, 5)
+                )
+            }
+        )
+
+        gb = df.groupby("Tuples")
+        gb_lambda = df.groupby(lambda x: df.iloc[x, 0])
+
+        expected = gb.indices
+        result = gb_lambda.indices
+
+        tm.assert_dict_equal(result, expected)
+
+
+# grouping
+# --------------------------------
+
+
+class TestGrouping:
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcde")),
+            Index(np.arange(5)),
+            Index(np.arange(5, dtype=float)),
+            date_range("2020-01-01", periods=5),
+            period_range("2020-01-01", periods=5),
+        ],
+    )
+    def test_grouper_index_types(self, index):
+        # related GH5375
+        # groupby misbehaving when using a Floatlike index
+        df = DataFrame(np.arange(10).reshape(5, 2), columns=list("AB"), index=index)
+
+        df.groupby(list("abcde"), group_keys=False).apply(lambda x: x)
+
+        df.index = df.index[::-1]
+        df.groupby(list("abcde"), group_keys=False).apply(lambda x: x)
+
+    def test_grouper_multilevel_freq(self):
+        # GH 7885
+        # with level and freq specified in a Grouper
+        d0 = date.today() - timedelta(days=14)
+        dates = date_range(d0, date.today())
+        date_index = MultiIndex.from_product([dates, dates], names=["foo", "bar"])
+        df = DataFrame(np.random.default_rng(2).integers(0, 100, 225), index=date_index)
+
+        # Check string level
+        expected = (
+            df.reset_index()
+            .groupby([Grouper(key="foo", freq="W"), Grouper(key="bar", freq="W")])
+            .sum()
+        )
+        # reset index changes columns dtype to object
+        expected.columns = Index([0], dtype="int64")
+
+        result = df.groupby(
+            [Grouper(level="foo", freq="W"), Grouper(level="bar", freq="W")]
+        ).sum()
+        tm.assert_frame_equal(result, expected)
+
+        # Check integer level
+        result = df.groupby(
+            [Grouper(level=0, freq="W"), Grouper(level=1, freq="W")]
+        ).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_creation_bug(self):
+        # GH 8795
+        df = DataFrame({"A": [0, 0, 1, 1, 2, 2], "B": [1, 2, 3, 4, 5, 6]})
+        g = df.groupby("A")
+        expected = g.sum()
+
+        g = df.groupby(Grouper(key="A"))
+        result = g.sum()
+        tm.assert_frame_equal(result, expected)
+
+        result = g.apply(lambda x: x.sum())
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_creation_bug2(self):
+        # GH14334
+        # Grouper(key=...) may be passed in a list
+        df = DataFrame(
+            {"A": [0, 0, 0, 1, 1, 1], "B": [1, 1, 2, 2, 3, 3], "C": [1, 2, 3, 4, 5, 6]}
+        )
+        # Group by single column
+        expected = df.groupby("A").sum()
+        g = df.groupby([Grouper(key="A")])
+        result = g.sum()
+        tm.assert_frame_equal(result, expected)
+
+        # Group by two columns
+        # using a combination of strings and Grouper objects
+        expected = df.groupby(["A", "B"]).sum()
+
+        # Group with two Grouper objects
+        g = df.groupby([Grouper(key="A"), Grouper(key="B")])
+        result = g.sum()
+        tm.assert_frame_equal(result, expected)
+
+        # Group with a string and a Grouper object
+        g = df.groupby(["A", Grouper(key="B")])
+        result = g.sum()
+        tm.assert_frame_equal(result, expected)
+
+        # Group with a Grouper object and a string
+        g = df.groupby([Grouper(key="A"), "B"])
+        result = g.sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_creation_bug3(self, unit):
+        # GH8866
+        dti = date_range("20130101", periods=2, unit=unit)
+        mi = MultiIndex.from_product(
+            [list("ab"), range(2), dti],
+            names=["one", "two", "three"],
+        )
+        ser = Series(
+            np.arange(8, dtype="int64"),
+            index=mi,
+        )
+        result = ser.groupby(Grouper(level="three", freq="ME")).sum()
+        exp_dti = pd.DatetimeIndex(
+            [Timestamp("2013-01-31")], freq="ME", name="three"
+        ).as_unit(unit)
+        expected = Series(
+            [28],
+            index=exp_dti,
+        )
+        tm.assert_series_equal(result, expected)
+
+        # just specifying a level breaks
+        result = ser.groupby(Grouper(level="one")).sum()
+        expected = ser.groupby(level="one").sum()
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("func", [False, True])
+    def test_grouper_returning_tuples(self, func):
+        # GH 22257 , both with dict and with callable
+        df = DataFrame({"X": ["A", "B", "A", "B"], "Y": [1, 4, 3, 2]})
+        mapping = dict(zip(range(4), [("C", 5), ("D", 6)] * 2))
+
+        if func:
+            gb = df.groupby(by=lambda idx: mapping[idx], sort=False)
+        else:
+            gb = df.groupby(by=mapping, sort=False)
+
+        name, expected = next(iter(gb))
+        assert name == ("C", 5)
+        result = gb.get_group(name)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_column_and_index(self):
+        # GH 14327
+
+        # Grouping a multi-index frame by a column and an index level should
+        # be equivalent to resetting the index and grouping by two columns
+        idx = MultiIndex.from_tuples(
+            [("a", 1), ("a", 2), ("a", 3), ("b", 1), ("b", 2), ("b", 3)]
+        )
+        idx.names = ["outer", "inner"]
+        df_multi = DataFrame(
+            {"A": np.arange(6), "B": ["one", "one", "two", "two", "one", "one"]},
+            index=idx,
+        )
+        result = df_multi.groupby(["B", Grouper(level="inner")]).mean(numeric_only=True)
+        expected = (
+            df_multi.reset_index().groupby(["B", "inner"]).mean(numeric_only=True)
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Test the reverse grouping order
+        result = df_multi.groupby([Grouper(level="inner"), "B"]).mean(numeric_only=True)
+        expected = (
+            df_multi.reset_index().groupby(["inner", "B"]).mean(numeric_only=True)
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Grouping a single-index frame by a column and the index should
+        # be equivalent to resetting the index and grouping by two columns
+        df_single = df_multi.reset_index("outer")
+        result = df_single.groupby(["B", Grouper(level="inner")]).mean(
+            numeric_only=True
+        )
+        expected = (
+            df_single.reset_index().groupby(["B", "inner"]).mean(numeric_only=True)
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # Test the reverse grouping order
+        result = df_single.groupby([Grouper(level="inner"), "B"]).mean(
+            numeric_only=True
+        )
+        expected = (
+            df_single.reset_index().groupby(["inner", "B"]).mean(numeric_only=True)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_levels_and_columns(self):
+        # GH9344, GH9049
+        idx_names = ["x", "y"]
+        idx = MultiIndex.from_tuples([(1, 1), (1, 2), (3, 4), (5, 6)], names=idx_names)
+        df = DataFrame(np.arange(12).reshape(-1, 3), index=idx)
+
+        by_levels = df.groupby(level=idx_names).mean()
+        # reset_index changes columns dtype to object
+        by_columns = df.reset_index().groupby(idx_names).mean()
+
+        # without casting, by_columns.columns is object-dtype
+        by_columns.columns = by_columns.columns.astype(np.int64)
+        tm.assert_frame_equal(by_levels, by_columns)
+
+    def test_groupby_categorical_index_and_columns(self, observed):
+        # GH18432, adapted for GH25871
+        columns = ["A", "B", "A", "B"]
+        categories = ["B", "A"]
+        data = np.array(
+            [[1, 2, 1, 2], [1, 2, 1, 2], [1, 2, 1, 2], [1, 2, 1, 2], [1, 2, 1, 2]], int
+        )
+        cat_columns = CategoricalIndex(columns, categories=categories, ordered=True)
+        expected_data = np.array([[4, 2], [4, 2], [4, 2], [4, 2], [4, 2]], int)
+        expected_columns = CategoricalIndex(
+            categories, categories=categories, ordered=True
+        )
+
+        # test transposed version
+        df = DataFrame(data.T, index=cat_columns)
+        result = df.groupby(level=0, observed=observed).sum()
+        expected = DataFrame(data=expected_data.T, index=expected_columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_getting_correct_binner(self):
+        # GH 10063
+        # using a non-time-based grouper and a time-based grouper
+        # and specifying levels
+        df = DataFrame(
+            {"A": 1},
+            index=MultiIndex.from_product(
+                [list("ab"), date_range("20130101", periods=80)], names=["one", "two"]
+            ),
+        )
+        result = df.groupby(
+            [Grouper(level="one"), Grouper(level="two", freq="ME")]
+        ).sum()
+        expected = DataFrame(
+            {"A": [31, 28, 21, 31, 28, 21]},
+            index=MultiIndex.from_product(
+                [list("ab"), date_range("20130101", freq="ME", periods=3)],
+                names=["one", "two"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_iter(self, df):
+        gb = df.groupby("A")
+        grouper = gb._grouper
+        result = sorted(grouper)
+        expected = ["bar", "foo"]
+        assert result == expected
+
+    def test_empty_groups(self, df):
+        # see gh-1048
+        with pytest.raises(ValueError, match="No group keys passed!"):
+            df.groupby([])
+
+    def test_groupby_grouper(self, df):
+        grouped = df.groupby("A")
+        grouper = grouped._grouper
+        result = df.groupby(grouper).mean(numeric_only=True)
+        expected = grouped.mean(numeric_only=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_dict_mapping(self):
+        # GH #679
+        s = Series({"T1": 5})
+        result = s.groupby({"T1": "T2"}).agg("sum")
+        expected = s.groupby(["T2"]).agg("sum")
+        tm.assert_series_equal(result, expected)
+
+        s = Series([1.0, 2.0, 3.0, 4.0], index=list("abcd"))
+        mapping = {"a": 0, "b": 0, "c": 1, "d": 1}
+
+        result = s.groupby(mapping).mean()
+        result2 = s.groupby(mapping).agg("mean")
+        exp_key = np.array([0, 0, 1, 1], dtype=np.int64)
+        expected = s.groupby(exp_key).mean()
+        expected2 = s.groupby(exp_key).mean()
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result, result2)
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            [0, 1, 2, 3],
+            ["a", "b", "c", "d"],
+            [Timestamp(2021, 7, 28 + i) for i in range(4)],
+        ],
+    )
+    def test_groupby_series_named_with_tuple(self, frame_or_series, index):
+        # GH 42731
+        obj = frame_or_series([1, 2, 3, 4], index=index)
+        groups = Series([1, 0, 1, 0], index=index, name=("a", "a"))
+        result = obj.groupby(groups).last()
+        expected = frame_or_series([4, 3])
+        expected.index.name = ("a", "a")
+        tm.assert_equal(result, expected)
+
+    def test_groupby_grouper_f_sanity_checked(self):
+        dates = date_range("01-Jan-2013", periods=12, freq="MS")
+        ts = Series(np.random.default_rng(2).standard_normal(12), index=dates)
+
+        # GH51979
+        # simple check that the passed function doesn't operates on the whole index
+        msg = "'Timestamp' object is not subscriptable"
+        with pytest.raises(TypeError, match=msg):
+            ts.groupby(lambda key: key[0:6])
+
+        result = ts.groupby(lambda x: x).sum()
+        expected = ts.groupby(ts.index).sum()
+        expected.index.freq = None
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_with_datetime_key(self):
+        # GH 51158
+        df = DataFrame(
+            {
+                "id": ["a", "b"] * 3,
+                "b": date_range("2000-01-01", "2000-01-03", freq="9h"),
+            }
+        )
+        grouper = Grouper(key="b", freq="D")
+        gb = df.groupby([grouper, "id"])
+
+        # test number of groups
+        expected = {
+            (Timestamp("2000-01-01"), "a"): [0, 2],
+            (Timestamp("2000-01-01"), "b"): [1],
+            (Timestamp("2000-01-02"), "a"): [4],
+            (Timestamp("2000-01-02"), "b"): [3, 5],
+        }
+        tm.assert_dict_equal(gb.groups, expected)
+
+        # test number of group keys
+        assert len(gb.groups.keys()) == 4
+
+    def test_grouping_error_on_multidim_input(self, df):
+        msg = "Grouper for '<class 'pandas.DataFrame'>' not 1-dimensional"
+        with pytest.raises(ValueError, match=msg):
+            Grouping(df.index, df[["A", "A"]])
+
+    def test_multiindex_negative_level(self, multiindex_dataframe_random_data):
+        # GH 13901
+        result = multiindex_dataframe_random_data.groupby(level=-1).sum()
+        expected = multiindex_dataframe_random_data.groupby(level="second").sum()
+        tm.assert_frame_equal(result, expected)
+
+        result = multiindex_dataframe_random_data.groupby(level=-2).sum()
+        expected = multiindex_dataframe_random_data.groupby(level="first").sum()
+        tm.assert_frame_equal(result, expected)
+
+        result = multiindex_dataframe_random_data.groupby(level=[-2, -1]).sum()
+        expected = multiindex_dataframe_random_data.sort_index()
+        tm.assert_frame_equal(result, expected)
+
+        result = multiindex_dataframe_random_data.groupby(level=[-1, "first"]).sum()
+        expected = multiindex_dataframe_random_data.groupby(
+            level=["second", "first"]
+        ).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_agg_with_dict_raises(self, df):
+        df.columns = np.arange(len(df.columns))
+        msg = "nested renamer is not supported"
+        with pytest.raises(SpecificationError, match=msg):
+            df.groupby(1, as_index=False)[2].agg({"Q": np.mean})
+
+    def test_multiindex_columns_empty_level(self):
+        lst = [["count", "values"], ["to filter", ""]]
+        midx = MultiIndex.from_tuples(lst)
+
+        df = DataFrame([[1, "A"]], columns=midx)
+
+        msg = "`groups` by one element list returns scalar is deprecated"
+        grouped = df.groupby("to filter").groups
+        assert grouped["A"] == [0]
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            grouped = df.groupby([("to filter", "")]).groups
+        assert grouped["A"] == [0]
+
+        df = DataFrame([[1, "A"], [2, "B"]], columns=midx)
+
+        expected = df.groupby("to filter").groups
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = df.groupby([("to filter", "")]).groups
+        assert result == expected
+
+        df = DataFrame([[1, "A"], [2, "A"]], columns=midx)
+
+        expected = df.groupby("to filter").groups
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = df.groupby([("to filter", "")]).groups
+        tm.assert_dict_equal(result, expected)
+
+    def test_groupby_multiindex_tuple(self):
+        # GH 17979, GH#59179
+        df = DataFrame(
+            [[1, 2, 3, 4], [3, 4, 5, 6], [1, 4, 2, 3]],
+            columns=MultiIndex.from_arrays([["a", "b", "b", "c"], [1, 1, 2, 2]]),
+        )
+
+        msg = "`groups` by one element list returns scalar is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            expected = df.groupby([("b", 1)]).groups
+        result = df.groupby(("b", 1)).groups
+        tm.assert_dict_equal(expected, result)
+
+        df2 = DataFrame(
+            df.values,
+            columns=MultiIndex.from_arrays(
+                [["a", "b", "b", "c"], ["d", "d", "e", "e"]]
+            ),
+        )
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            expected = df2.groupby([("b", "d")]).groups
+        result = df.groupby(("b", 1)).groups
+        tm.assert_dict_equal(expected, result)
+
+        df3 = DataFrame(df.values, columns=[("a", "d"), ("b", "d"), ("b", "e"), "c"])
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            expected = df3.groupby([("b", "d")]).groups
+        result = df.groupby(("b", 1)).groups
+        tm.assert_dict_equal(expected, result)
+
+    def test_groupby_multiindex_partial_indexing_equivalence(self):
+        # GH 17977, GH#59179
+        df = DataFrame(
+            [[1, 2, 3, 4], [3, 4, 5, 6], [1, 4, 2, 3]],
+            columns=MultiIndex.from_arrays([["a", "b", "b", "c"], [1, 1, 2, 2]]),
+        )
+
+        expected_mean = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].mean()
+        result_mean = df.groupby([("a", 1)])["b"].mean()
+        tm.assert_frame_equal(expected_mean, result_mean)
+
+        expected_sum = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].sum()
+        result_sum = df.groupby([("a", 1)])["b"].sum()
+        tm.assert_frame_equal(expected_sum, result_sum)
+
+        expected_count = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].count()
+        result_count = df.groupby([("a", 1)])["b"].count()
+        tm.assert_frame_equal(expected_count, result_count)
+
+        expected_min = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].min()
+        result_min = df.groupby([("a", 1)])["b"].min()
+        tm.assert_frame_equal(expected_min, result_min)
+
+        expected_max = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].max()
+        result_max = df.groupby([("a", 1)])["b"].max()
+        tm.assert_frame_equal(expected_max, result_max)
+
+        msg = "`groups` by one element list returns scalar is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            expected_groups = df.groupby([("a", 1)])[[("b", 1), ("b", 2)]].groups
+            result_groups = df.groupby([("a", 1)])["b"].groups
+        tm.assert_dict_equal(expected_groups, result_groups)
+
+    def test_groupby_level(self, sort, multiindex_dataframe_random_data, df):
+        # GH 17537
+        frame = multiindex_dataframe_random_data
+        deleveled = frame.reset_index()
+
+        result0 = frame.groupby(level=0, sort=sort).sum()
+        result1 = frame.groupby(level=1, sort=sort).sum()
+
+        expected0 = frame.groupby(deleveled["first"].values, sort=sort).sum()
+        expected1 = frame.groupby(deleveled["second"].values, sort=sort).sum()
+
+        expected0.index.name = "first"
+        expected1.index.name = "second"
+
+        assert result0.index.name == "first"
+        assert result1.index.name == "second"
+
+        tm.assert_frame_equal(result0, expected0)
+        tm.assert_frame_equal(result1, expected1)
+        assert result0.index.name == frame.index.names[0]
+        assert result1.index.name == frame.index.names[1]
+
+        # groupby level name
+        result0 = frame.groupby(level="first", sort=sort).sum()
+        result1 = frame.groupby(level="second", sort=sort).sum()
+        tm.assert_frame_equal(result0, expected0)
+        tm.assert_frame_equal(result1, expected1)
+
+        # raise exception for non-MultiIndex
+        msg = "level > 0 or level < -1 only valid with MultiIndex"
+        with pytest.raises(ValueError, match=msg):
+            df.groupby(level=1)
+
+    def test_groupby_level_index_names(self):
+        # GH4014 this used to raise ValueError since 'exp'>1 (in py2)
+        df = DataFrame({"exp": ["A"] * 3 + ["B"] * 3, "var1": range(6)}).set_index(
+            "exp"
+        )
+        df.groupby(level="exp")
+        msg = "level name foo is not the name of the index"
+        with pytest.raises(ValueError, match=msg):
+            df.groupby(level="foo")
+
+    def test_groupby_level_with_nas(self, sort):
+        # GH 17537
+        index = MultiIndex(
+            levels=[[1, 0], [0, 1, 2, 3]],
+            codes=[[1, 1, 1, 1, 0, 0, 0, 0], [0, 1, 2, 3, 0, 1, 2, 3]],
+        )
+
+        # factorizing doesn't confuse things
+        s = Series(np.arange(8.0), index=index)
+        result = s.groupby(level=0, sort=sort).sum()
+        expected = Series([6.0, 22.0], index=[0, 1])
+        tm.assert_series_equal(result, expected)
+
+        index = MultiIndex(
+            levels=[[1, 0], [0, 1, 2, 3]],
+            codes=[[1, 1, 1, 1, -1, 0, 0, 0], [0, 1, 2, 3, 0, 1, 2, 3]],
+        )
+
+        # factorizing doesn't confuse things
+        s = Series(np.arange(8.0), index=index)
+        result = s.groupby(level=0, sort=sort).sum()
+        expected = Series([6.0, 18.0], index=[0.0, 1.0])
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_args(self, multiindex_dataframe_random_data):
+        # PR8618 and issue 8015
+        frame = multiindex_dataframe_random_data
+
+        msg = "You have to supply one of 'by' and 'level'"
+        with pytest.raises(TypeError, match=msg):
+            frame.groupby()
+
+        msg = "You have to supply one of 'by' and 'level'"
+        with pytest.raises(TypeError, match=msg):
+            frame.groupby(by=None, level=None)
+
+    @pytest.mark.parametrize(
+        "sort,labels",
+        [
+            [True, [2, 2, 2, 0, 0, 1, 1, 3, 3, 3]],
+            [False, [0, 0, 0, 1, 1, 2, 2, 3, 3, 3]],
+        ],
+    )
+    def test_level_preserve_order(self, sort, labels, multiindex_dataframe_random_data):
+        # GH 17537
+        grouped = multiindex_dataframe_random_data.groupby(level=0, sort=sort)
+        exp_labels = np.array(labels, np.intp)
+        tm.assert_almost_equal(grouped._grouper.ids, exp_labels)
+
+    def test_grouping_labels(self, multiindex_dataframe_random_data):
+        grouped = multiindex_dataframe_random_data.groupby(
+            multiindex_dataframe_random_data.index.get_level_values(0)
+        )
+        exp_labels = np.array([2, 2, 2, 0, 0, 1, 1, 3, 3, 3], dtype=np.intp)
+        tm.assert_almost_equal(grouped._grouper.codes[0], exp_labels)
+
+    def test_list_grouper_with_nat(self):
+        # GH 14715, GH#59179
+        df = DataFrame({"date": date_range("1/1/2011", periods=365, freq="D")})
+        df.iloc[-1] = pd.NaT
+        grouper = Grouper(key="date", freq="YS")
+        msg = "`groups` by one element list returns scalar is deprecated"
+
+        # Grouper in a list grouping
+        gb = df.groupby([grouper])
+        expected = {Timestamp("2011-01-01"): Index(list(range(364)))}
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = gb.groups
+        tm.assert_dict_equal(result, expected)
+
+        # Test case without a list
+        result = df.groupby(grouper)
+        expected = {Timestamp("2011-01-01"): 365}
+        tm.assert_dict_equal(result.groups, expected)
+
+    @pytest.mark.parametrize(
+        "func,expected",
+        [
+            (
+                "transform",
+                Series(name=2, dtype=np.float64),
+            ),
+            (
+                "agg",
+                Series(
+                    name=2, dtype=np.float64, index=Index([], dtype=np.float64, name=1)
+                ),
+            ),
+            (
+                "apply",
+                Series(
+                    name=2, dtype=np.float64, index=Index([], dtype=np.float64, name=1)
+                ),
+            ),
+        ],
+    )
+    def test_evaluate_with_empty_groups(self, func, expected):
+        # 26208
+        # test transform'ing empty groups
+        # (not testing other agg fns, because they return
+        # different index objects.
+        df = DataFrame({1: [], 2: []})
+        g = df.groupby(1, group_keys=True)
+        result = getattr(g[2], func)(lambda x: x)
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_apply_empty_with_group_keys_false(self):
+        # 60471
+        # test apply'ing empty groups with group_keys False
+        # (not testing other agg fns, because they return
+        # different index objects.
+        df = DataFrame({"A": [], "B": [], "C": []})
+        g = df.groupby("A", group_keys=False)
+        result = g.apply(lambda x: x / x.sum())
+        expected = DataFrame({"B": [], "C": []}, index=None)
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_empty(self):
+        # https://github.com/pandas-dev/pandas/issues/27190
+        s = Series([], name="name", dtype="float64")
+        gr = s.groupby([])
+
+        result = gr.mean()
+        expected = s.set_axis(Index([], dtype=np.intp))
+        tm.assert_series_equal(result, expected)
+
+        # check group properties
+        assert len(gr._grouper.groupings) == 1
+        tm.assert_numpy_array_equal(
+            gr._grouper.ids, np.array([], dtype=np.dtype(np.intp))
+        )
+
+        assert gr._grouper.ngroups == 0
+
+        # check name
+        gb = s.groupby(s)
+        grouper = gb._grouper
+        result = grouper.names
+        expected = ["name"]
+        assert result == expected
+
+    def test_groupby_level_index_value_all_na(self):
+        # issue 20519
+        df = DataFrame(
+            [["x", np.nan, 10], [None, np.nan, 20]], columns=["A", "B", "C"]
+        ).set_index(["A", "B"])
+        result = df.groupby(level=["A", "B"]).sum()
+        expected = DataFrame(
+            data=[],
+            index=MultiIndex(
+                levels=[Index(["x"], dtype="str"), Index([], dtype="float64")],
+                codes=[[], []],
+                names=["A", "B"],
+            ),
+            columns=["C"],
+            dtype="int64",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_multiindex_level_empty(self):
+        # https://github.com/pandas-dev/pandas/issues/31670
+        df = DataFrame(
+            [[123, "a", 1.0], [123, "b", 2.0]], columns=["id", "category", "value"]
+        )
+        df = df.set_index(["id", "category"])
+        empty = df[df.value < 0]
+        result = empty.groupby("id").sum()
+        expected = DataFrame(
+            dtype="float64",
+            columns=["value"],
+            index=Index([], dtype=np.int64, name="id"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_tuple_keys_handle_multiindex(self):
+        # https://github.com/pandas-dev/pandas/issues/21340
+        df = DataFrame(
+            {
+                "num1": [0, 8, 9, 4, 3, 3, 5, 9, 3, 6],
+                "num2": [3, 8, 6, 4, 9, 2, 1, 7, 0, 9],
+                "num3": [6, 5, 7, 8, 5, 1, 1, 10, 7, 8],
+                "category_tuple": [
+                    (0, 1),
+                    (0, 1),
+                    (0, 1),
+                    (0, 4),
+                    (2, 3),
+                    (2, 3),
+                    (2, 3),
+                    (2, 3),
+                    (5,),
+                    (6,),
+                ],
+                "category_string": list("aaabbbbcde"),
+            }
+        )
+        expected = df.sort_values(by=["category_tuple", "num1"])
+        result = df.groupby("category_tuple").apply(lambda x: x.sort_values(by="num1"))
+        expected = expected[result.columns]
+        tm.assert_frame_equal(result.reset_index(drop=True), expected)
+
+
+# get_group
+# --------------------------------
+
+
+class TestGetGroup:
+    def test_get_group(self):
+        # GH 5267
+        # be datelike friendly
+        df = DataFrame(
+            {
+                "DATE": pd.to_datetime(
+                    [
+                        "10-Oct-2013",
+                        "10-Oct-2013",
+                        "10-Oct-2013",
+                        "11-Oct-2013",
+                        "11-Oct-2013",
+                        "11-Oct-2013",
+                    ]
+                ),
+                "label": ["foo", "foo", "bar", "foo", "foo", "bar"],
+                "VAL": [1, 2, 3, 4, 5, 6],
+            }
+        )
+
+        g = df.groupby("DATE")
+        key = next(iter(g.groups))
+        result1 = g.get_group(key)
+        result2 = g.get_group(Timestamp(key).to_pydatetime())
+        result3 = g.get_group(str(Timestamp(key)))
+        tm.assert_frame_equal(result1, result2)
+        tm.assert_frame_equal(result1, result3)
+
+        g = df.groupby(["DATE", "label"])
+
+        key = next(iter(g.groups))
+        result1 = g.get_group(key)
+        result2 = g.get_group((Timestamp(key[0]).to_pydatetime(), key[1]))
+        result3 = g.get_group((str(Timestamp(key[0])), key[1]))
+        tm.assert_frame_equal(result1, result2)
+        tm.assert_frame_equal(result1, result3)
+
+        # must pass a same-length tuple with multiple keys
+        msg = "must supply a tuple to get_group with multiple grouping keys"
+        with pytest.raises(ValueError, match=msg):
+            g.get_group("foo")
+        with pytest.raises(ValueError, match=msg):
+            g.get_group("foo")
+        msg = "must supply a same-length tuple to get_group with multiple grouping keys"
+        with pytest.raises(ValueError, match=msg):
+            g.get_group(("foo", "bar", "baz"))
+
+    def test_get_group_empty_bins(self, observed):
+        d = DataFrame([3, 1, 7, 6])
+        bins = [0, 5, 10, 15]
+        g = d.groupby(pd.cut(d[0], bins), observed=observed)
+
+        # TODO: should prob allow a str of Interval work as well
+        # IOW '(0, 5]'
+        result = g.get_group(pd.Interval(0, 5))
+        expected = DataFrame([3, 1], index=[0, 1])
+        tm.assert_frame_equal(result, expected)
+
+        msg = r"Interval\(10, 15, closed='right'\)"
+        with pytest.raises(KeyError, match=msg):
+            g.get_group(pd.Interval(10, 15))
+
+    def test_get_group_grouped_by_tuple(self):
+        # GH 8121
+        df = DataFrame([[(1,), (1, 2), (1,), (1, 2)]], index=["ids"]).T
+        gr = df.groupby("ids")
+        expected = DataFrame({"ids": [(1,), (1,)]}, index=[0, 2])
+        result = gr.get_group((1,))
+        tm.assert_frame_equal(result, expected)
+
+        dt = pd.to_datetime(["2010-01-01", "2010-01-02", "2010-01-01", "2010-01-02"])
+        df = DataFrame({"ids": [(x,) for x in dt]})
+        gr = df.groupby("ids")
+        result = gr.get_group(("2010-01-01",))
+        expected = DataFrame({"ids": [(dt[0],), (dt[0],)]}, index=[0, 2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_group_grouped_by_tuple_with_lambda(self):
+        # GH 36158
+        df = DataFrame(
+            {
+                "Tuples": (
+                    (x, y)
+                    for x in [0, 1]
+                    for y in np.random.default_rng(2).integers(3, 5, 5)
+                )
+            }
+        )
+
+        gb = df.groupby("Tuples")
+        gb_lambda = df.groupby(lambda x: df.iloc[x, 0])
+
+        expected = gb.get_group(next(iter(gb.groups.keys())))
+        result = gb_lambda.get_group(next(iter(gb_lambda.groups.keys())))
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_with_empty(self):
+        index = pd.DatetimeIndex(())
+        data = ()
+        series = Series(data, index, dtype=object)
+        grouper = Grouper(freq="D")
+        grouped = series.groupby(grouper)
+        assert next(iter(grouped), None) is None
+
+    def test_groupby_with_single_column(self):
+        df = DataFrame({"a": list("abssbab")})
+        tm.assert_frame_equal(df.groupby("a").get_group("a"), df.iloc[[0, 5]])
+        # GH 13530
+        exp = DataFrame(
+            index=Index(["a", "b", "s"], name="a"), columns=Index([], dtype="str")
+        )
+        tm.assert_frame_equal(df.groupby("a").count(), exp)
+        tm.assert_frame_equal(df.groupby("a").sum(), exp)
+
+        exp = df.iloc[[3, 4, 5]]
+        tm.assert_frame_equal(df.groupby("a").nth(1), exp)
+
+    def test_gb_key_len_equal_axis_len(self):
+        # GH16843
+        # test ensures that index and column keys are recognized correctly
+        # when number of keys equals axis length of groupby
+        df = DataFrame(
+            [["foo", "bar", "B", 1], ["foo", "bar", "B", 2], ["foo", "baz", "C", 3]],
+            columns=["first", "second", "third", "one"],
+        )
+        df = df.set_index(["first", "second"])
+        df = df.groupby(["first", "second", "third"]).size()
+        assert df.loc[("foo", "bar", "B")] == 2
+        assert df.loc[("foo", "baz", "C")] == 1
+
+
+# groups & iteration
+# --------------------------------
+
+
+class TestIteration:
+    def test_groups(self, df):
+        grouped = df.groupby(["A"])
+        msg = "`groups` by one element list returns scalar is deprecated"
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            groups = grouped.groups
+            assert groups is grouped.groups  # caching works
+
+        for k, v in groups.items():
+            assert (df.loc[v]["A"] == k).all()
+
+        grouped = df.groupby(["A", "B"])
+        groups = grouped.groups
+        assert groups is grouped.groups  # caching works
+
+        for k, v in groups.items():
+            assert (df.loc[v]["A"] == k[0]).all()
+            assert (df.loc[v]["B"] == k[1]).all()
+
+    def test_grouping_is_iterable(self, tsframe):
+        # this code path isn't used anywhere else
+        # not sure it's useful
+        grouped = tsframe.groupby([lambda x: x.weekday(), lambda x: x.year])
+
+        # test it works
+        for g in grouped._grouper.groupings[0]:
+            pass
+
+    def test_multi_iter(self):
+        s = Series(np.arange(6))
+        k1 = np.array(["a", "a", "a", "b", "b", "b"])
+        k2 = np.array(["1", "2", "1", "2", "1", "2"])
+
+        grouped = s.groupby([k1, k2])
+
+        iterated = list(grouped)
+        expected = [
+            ("a", "1", s[[0, 2]]),
+            ("a", "2", s[[1]]),
+            ("b", "1", s[[4]]),
+            ("b", "2", s[[3, 5]]),
+        ]
+        for i, ((one, two), three) in enumerate(iterated):
+            e1, e2, e3 = expected[i]
+            assert e1 == one
+            assert e2 == two
+            tm.assert_series_equal(three, e3)
+
+    def test_multi_iter_frame(self, three_group):
+        k1 = np.array(["b", "b", "b", "a", "a", "a"])
+        k2 = np.array(["1", "2", "1", "2", "1", "2"])
+        df = DataFrame(
+            {
+                "v1": np.random.default_rng(2).standard_normal(6),
+                "v2": np.random.default_rng(2).standard_normal(6),
+                "k1": k1,
+                "k2": k2,
+            },
+            index=["one", "two", "three", "four", "five", "six"],
+        )
+
+        grouped = df.groupby(["k1", "k2"])
+
+        # things get sorted!
+        iterated = list(grouped)
+        idx = df.index
+        expected = [
+            ("a", "1", df.loc[idx[[4]]]),
+            ("a", "2", df.loc[idx[[3, 5]]]),
+            ("b", "1", df.loc[idx[[0, 2]]]),
+            ("b", "2", df.loc[idx[[1]]]),
+        ]
+        for i, ((one, two), three) in enumerate(iterated):
+            e1, e2, e3 = expected[i]
+            assert e1 == one
+            assert e2 == two
+            tm.assert_frame_equal(three, e3)
+
+        # don't iterate through groups with no data
+        df["k1"] = np.array(["b", "b", "b", "a", "a", "a"])
+        df["k2"] = np.array(["1", "1", "1", "2", "2", "2"])
+        grouped = df.groupby(["k1", "k2"])
+        # calling `dict` on a DataFrameGroupBy leads to a TypeError,
+        # we need to use a dictionary comprehension here
+        groups = {key: gp for key, gp in grouped}  # noqa: C416
+        assert len(groups) == 2
+
+    def test_dictify(self, df):
+        dict(iter(df.groupby("A")))
+        dict(iter(df.groupby(["A", "B"])))
+        dict(iter(df["C"].groupby(df["A"])))
+        dict(iter(df["C"].groupby([df["A"], df["B"]])))
+        dict(iter(df.groupby("A")["C"]))
+        dict(iter(df.groupby(["A", "B"])["C"]))
+
+    def test_groupby_with_small_elem(self):
+        # GH 8542
+        # length=2
+        df = DataFrame(
+            {"event": ["start", "start"], "change": [1234, 5678]},
+            index=pd.DatetimeIndex(["2014-09-10", "2013-10-10"]),
+        )
+        grouped = df.groupby([Grouper(freq="ME"), "event"])
+        assert len(grouped.groups) == 2
+        assert grouped.ngroups == 2
+        assert (Timestamp("2014-09-30"), "start") in grouped.groups
+        assert (Timestamp("2013-10-31"), "start") in grouped.groups
+
+        res = grouped.get_group((Timestamp("2014-09-30"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[0], :])
+        res = grouped.get_group((Timestamp("2013-10-31"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[1], :])
+
+        df = DataFrame(
+            {"event": ["start", "start", "start"], "change": [1234, 5678, 9123]},
+            index=pd.DatetimeIndex(["2014-09-10", "2013-10-10", "2014-09-15"]),
+        )
+        grouped = df.groupby([Grouper(freq="ME"), "event"])
+        assert len(grouped.groups) == 2
+        assert grouped.ngroups == 2
+        assert (Timestamp("2014-09-30"), "start") in grouped.groups
+        assert (Timestamp("2013-10-31"), "start") in grouped.groups
+
+        res = grouped.get_group((Timestamp("2014-09-30"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[0, 2], :])
+        res = grouped.get_group((Timestamp("2013-10-31"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[1], :])
+
+        # length=3
+        df = DataFrame(
+            {"event": ["start", "start", "start"], "change": [1234, 5678, 9123]},
+            index=pd.DatetimeIndex(["2014-09-10", "2013-10-10", "2014-08-05"]),
+        )
+        grouped = df.groupby([Grouper(freq="ME"), "event"])
+        assert len(grouped.groups) == 3
+        assert grouped.ngroups == 3
+        assert (Timestamp("2014-09-30"), "start") in grouped.groups
+        assert (Timestamp("2013-10-31"), "start") in grouped.groups
+        assert (Timestamp("2014-08-31"), "start") in grouped.groups
+
+        res = grouped.get_group((Timestamp("2014-09-30"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[0], :])
+        res = grouped.get_group((Timestamp("2013-10-31"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[1], :])
+        res = grouped.get_group((Timestamp("2014-08-31"), "start"))
+        tm.assert_frame_equal(res, df.iloc[[2], :])
+
+    def test_grouping_string_repr(self):
+        # GH 13394
+        mi = MultiIndex.from_arrays([list("AAB"), list("aba")])
+        df = DataFrame([[1, 2, 3]], columns=mi)
+        gr = df.groupby(df[("A", "a")])
+
+        result = gr._grouper.groupings[0].__repr__()
+        expected = "Grouping(('A', 'a'))"
+        assert result == expected
+
+
+def test_grouping_by_key_is_in_axis():
+    # GH#50413 - Groupers specified by key are in-axis
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 1, 2], "c": [3, 4, 5]}).set_index("a")
+    gb = df.groupby([Grouper(level="a"), Grouper(key="b")], as_index=False)
+    assert not gb._grouper.groupings[0].in_axis
+    assert gb._grouper.groupings[1].in_axis
+
+    result = gb.sum()
+    expected = DataFrame({"a": [1, 2], "b": [1, 2], "c": [7, 5]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_any_with_timedelta():
+    # GH#59712
+    df = DataFrame({"value": [pd.Timedelta(1), pd.NaT]})
+
+    result = df.groupby(np.array([0, 1], dtype=np.int64))["value"].any()
+
+    expected = Series({0: True, 1: False}, name="value", dtype=bool)
+    expected.index = expected.index.astype(np.int64)
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_index_as_string.py b/pandas/pandas/tests/groupby/test_index_as_string.py
new file mode 100644
index 0000000000000..743db7e70b14b
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_index_as_string.py
@@ -0,0 +1,72 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "key_strs,groupers",
+    [
+        ("inner", pd.Grouper(level="inner")),  # Index name
+        (["inner"], [pd.Grouper(level="inner")]),  # List of index name
+        (["B", "inner"], ["B", pd.Grouper(level="inner")]),  # Column and index
+        (["inner", "B"], [pd.Grouper(level="inner"), "B"]),  # Index and column
+    ],
+)
+@pytest.mark.parametrize("levels", [["inner"], ["inner", "outer"]])
+def test_grouper_index_level_as_string(levels, key_strs, groupers):
+    frame = pd.DataFrame(
+        {
+            "outer": ["a", "a", "a", "b", "b", "b"],
+            "inner": [1, 2, 3, 1, 2, 3],
+            "A": np.arange(6),
+            "B": ["one", "one", "two", "two", "one", "one"],
+        }
+    )
+    frame = frame.set_index(levels)
+    if "B" not in key_strs or "outer" in frame.columns:
+        result = frame.groupby(key_strs).mean(numeric_only=True)
+        expected = frame.groupby(groupers).mean(numeric_only=True)
+    else:
+        result = frame.groupby(key_strs).mean()
+        expected = frame.groupby(groupers).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "levels",
+    [
+        "inner",
+        "outer",
+        "B",
+        ["inner"],
+        ["outer"],
+        ["B"],
+        ["inner", "outer"],
+        ["outer", "inner"],
+        ["inner", "outer", "B"],
+        ["B", "outer", "inner"],
+    ],
+)
+def test_grouper_index_level_as_string_series(levels):
+    # Compute expected result
+    df = pd.DataFrame(
+        {
+            "outer": ["a", "a", "a", "b", "b", "b"],
+            "inner": [1, 2, 3, 1, 2, 3],
+            "A": np.arange(6),
+            "B": ["one", "one", "two", "two", "one", "one"],
+        }
+    )
+    series = df.set_index(["outer", "inner", "B"])["A"]
+    if isinstance(levels, list):
+        groupers = [pd.Grouper(level=lv) for lv in levels]
+    else:
+        groupers = pd.Grouper(level=levels)
+
+    expected = series.groupby(groupers).mean()
+
+    # Compute and check result
+    result = series.groupby(levels).mean()
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_indexing.py b/pandas/pandas/tests/groupby/test_indexing.py
new file mode 100644
index 0000000000000..a3d3f509e186a
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_indexing.py
@@ -0,0 +1,310 @@
+# Test GroupBy._positional_selector positional grouped indexing GH#42864
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "arg, expected_rows",
+    [
+        [0, [0, 1, 4]],
+        [2, [5]],
+        [5, []],
+        [-1, [3, 4, 7]],
+        [-2, [1, 6]],
+        [-6, []],
+    ],
+)
+def test_int(slice_test_df, slice_test_grouped, arg, expected_rows):
+    # Test single integer
+    result = slice_test_grouped._positional_selector[arg]
+    expected = slice_test_df.iloc[expected_rows]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_slice(slice_test_df, slice_test_grouped):
+    # Test single slice
+    result = slice_test_grouped._positional_selector[0:3:2]
+    expected = slice_test_df.iloc[[0, 1, 4, 5]]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg, expected_rows",
+    [
+        [[0, 2], [0, 1, 4, 5]],
+        [[0, 2, -1], [0, 1, 3, 4, 5, 7]],
+        [range(0, 3, 2), [0, 1, 4, 5]],
+        [{0, 2}, [0, 1, 4, 5]],
+    ],
+    ids=[
+        "list",
+        "negative",
+        "range",
+        "set",
+    ],
+)
+def test_list(slice_test_df, slice_test_grouped, arg, expected_rows):
+    # Test lists of integers and integer valued iterables
+    result = slice_test_grouped._positional_selector[arg]
+    expected = slice_test_df.iloc[expected_rows]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ints(slice_test_df, slice_test_grouped):
+    # Test tuple of ints
+    result = slice_test_grouped._positional_selector[0, 2, -1]
+    expected = slice_test_df.iloc[[0, 1, 3, 4, 5, 7]]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_slices(slice_test_df, slice_test_grouped):
+    # Test tuple of slices
+    result = slice_test_grouped._positional_selector[:2, -2:]
+    expected = slice_test_df.iloc[[0, 1, 2, 3, 4, 6, 7]]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_mix(slice_test_df, slice_test_grouped):
+    # Test mixed tuple of ints and slices
+    result = slice_test_grouped._positional_selector[0, 1, -2:]
+    expected = slice_test_df.iloc[[0, 1, 2, 3, 4, 6, 7]]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg, expected_rows",
+    [
+        [0, [0, 1, 4]],
+        [[0, 2, -1], [0, 1, 3, 4, 5, 7]],
+        [(slice(None, 2), slice(-2, None)), [0, 1, 2, 3, 4, 6, 7]],
+    ],
+)
+def test_as_index(slice_test_df, arg, expected_rows):
+    # Test the default as_index behaviour
+    result = slice_test_df.groupby("Group", sort=False)._positional_selector[arg]
+    expected = slice_test_df.iloc[expected_rows]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_doc_examples():
+    # Test the examples in the documentation
+    df = pd.DataFrame(
+        [["a", 1], ["a", 2], ["a", 3], ["b", 4], ["b", 5]], columns=["A", "B"]
+    )
+
+    grouped = df.groupby("A", as_index=False)
+
+    result = grouped._positional_selector[1:2]
+    expected = pd.DataFrame([["a", 2], ["b", 5]], columns=["A", "B"], index=[1, 4])
+
+    tm.assert_frame_equal(result, expected)
+
+    result = grouped._positional_selector[1, -1]
+    expected = pd.DataFrame(
+        [["a", 2], ["a", 3], ["b", 5]], columns=["A", "B"], index=[1, 2, 4]
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multiindex():
+    # Test the multiindex mentioned as the use-case in the documentation
+
+    def _make_df_from_data(data):
+        rows = {}
+        for date in data:
+            for level in data[date]:
+                rows[(date, level[0])] = {"A": level[1], "B": level[2]}
+
+        df = pd.DataFrame.from_dict(rows, orient="index")
+        df.index.names = ("Date", "Item")
+        return df
+
+    rng = np.random.default_rng(2)
+    ndates = 100
+    nitems = 20
+    dates = pd.date_range("20130101", periods=ndates, freq="D")
+    items = [f"item {i}" for i in range(nitems)]
+
+    multiindex_data = {}
+    for date in dates:
+        nitems_for_date = nitems - rng.integers(0, 12)
+        levels = [
+            (item, rng.integers(0, 10000) / 100, rng.integers(0, 10000) / 100)
+            for item in items[:nitems_for_date]
+        ]
+        levels.sort(key=lambda x: x[1])
+        multiindex_data[date] = levels
+
+    df = _make_df_from_data(multiindex_data)
+    result = df.groupby("Date", as_index=False).nth(slice(3, -3))
+
+    sliced = {date: values[3:-3] for date, values in multiindex_data.items()}
+    expected = _make_df_from_data(sliced)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("arg", [1, 5, 30, 1000, -1, -5, -30, -1000])
+@pytest.mark.parametrize("method", ["head", "tail"])
+@pytest.mark.parametrize("simulated", [True, False])
+def test_against_head_and_tail(arg, method, simulated):
+    # Test gives the same results as grouped head and tail
+    n_groups = 100
+    n_rows_per_group = 30
+
+    data = {
+        "group": [
+            f"group {g}" for j in range(n_rows_per_group) for g in range(n_groups)
+        ],
+        "value": [
+            f"group {g} row {j}"
+            for j in range(n_rows_per_group)
+            for g in range(n_groups)
+        ],
+    }
+    df = pd.DataFrame(data)
+    grouped = df.groupby("group", as_index=False)
+    size = arg if arg >= 0 else n_rows_per_group + arg
+
+    if method == "head":
+        result = grouped._positional_selector[:arg]
+
+        if simulated:
+            indices = [
+                j * n_groups + i
+                for j in range(size)
+                for i in range(n_groups)
+                if j * n_groups + i < n_groups * n_rows_per_group
+            ]
+            expected = df.iloc[indices]
+
+        else:
+            expected = grouped.head(arg)
+
+    else:
+        result = grouped._positional_selector[-arg:]
+
+        if simulated:
+            indices = [
+                (n_rows_per_group + j - size) * n_groups + i
+                for j in range(size)
+                for i in range(n_groups)
+                if (n_rows_per_group + j - size) * n_groups + i >= 0
+            ]
+            expected = df.iloc[indices]
+
+        else:
+            expected = grouped.tail(arg)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("start", [None, 0, 1, 10, -1, -10])
+@pytest.mark.parametrize("stop", [None, 0, 1, 10, -1, -10])
+@pytest.mark.parametrize("step", [None, 1, 5])
+def test_against_df_iloc(start, stop, step):
+    # Test that a single group gives the same results as DataFrame.iloc
+    n_rows = 30
+
+    data = {
+        "group": ["group 0"] * n_rows,
+        "value": list(range(n_rows)),
+    }
+    df = pd.DataFrame(data)
+    grouped = df.groupby("group", as_index=False)
+
+    result = grouped._positional_selector[start:stop:step]
+    expected = df.iloc[start:stop:step]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_series():
+    # Test grouped Series
+    ser = pd.Series([1, 2, 3, 4, 5], index=["a", "a", "a", "b", "b"])
+    grouped = ser.groupby(level=0)
+    result = grouped._positional_selector[1:2]
+    expected = pd.Series([2, 5], index=["a", "b"])
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("step", [1, 2, 3, 4, 5])
+def test_step(step):
+    # Test slice with various step values
+    data = [["x", f"x{i}"] for i in range(5)]
+    data += [["y", f"y{i}"] for i in range(4)]
+    data += [["z", f"z{i}"] for i in range(3)]
+    df = pd.DataFrame(data, columns=["A", "B"])
+
+    grouped = df.groupby("A", as_index=False)
+
+    result = grouped._positional_selector[::step]
+
+    data = [["x", f"x{i}"] for i in range(0, 5, step)]
+    data += [["y", f"y{i}"] for i in range(0, 4, step)]
+    data += [["z", f"z{i}"] for i in range(0, 3, step)]
+
+    index = [0 + i for i in range(0, 5, step)]
+    index += [5 + i for i in range(0, 4, step)]
+    index += [9 + i for i in range(0, 3, step)]
+
+    expected = pd.DataFrame(data, columns=["A", "B"], index=index)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_columns_on_iter():
+    # GitHub issue #44821
+    df = pd.DataFrame({k: range(10) for k in "ABC"})
+
+    # Group-by and select columns
+    cols = ["A", "B"]
+    for _, dg in df.groupby(df.A < 4)[cols]:
+        tm.assert_index_equal(dg.columns, pd.Index(cols))
+        assert "C" not in dg.columns
+
+
+@pytest.mark.parametrize("func", [list, pd.Index, pd.Series, np.array])
+def test_groupby_duplicated_columns(func):
+    # GH#44924
+    df = pd.DataFrame(
+        {
+            "A": [1, 2],
+            "B": [3, 3],
+            "C": ["G", "G"],
+        }
+    )
+    result = df.groupby("C")[func(["A", "B", "A"])].mean()
+    expected = pd.DataFrame(
+        [[1.5, 3.0, 1.5]], columns=["A", "B", "A"], index=pd.Index(["G"], name="C")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_get_nonexisting_groups():
+    # GH#32492
+    df = pd.DataFrame(
+        data={
+            "A": ["a1", "a2", None],
+            "B": ["b1", "b2", "b1"],
+            "val": [1, 2, 3],
+        }
+    )
+    grps = df.groupby(by=["A", "B"])
+
+    msg = "('a2', 'b1')"
+    with pytest.raises(KeyError, match=msg):
+        grps.get_group(("a2", "b1"))
diff --git a/pandas/pandas/tests/groupby/test_libgroupby.py b/pandas/pandas/tests/groupby/test_libgroupby.py
new file mode 100644
index 0000000000000..35b8fa93b8e03
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_libgroupby.py
@@ -0,0 +1,331 @@
+import numpy as np
+import pytest
+
+from pandas._libs import groupby as libgroupby
+from pandas._libs.groupby import (
+    group_cumprod,
+    group_cumsum,
+    group_mean,
+    group_sum,
+    group_var,
+)
+
+from pandas.core.dtypes.common import ensure_platform_int
+
+from pandas import isna
+import pandas._testing as tm
+
+
+class GroupVarTestMixin:
+    def test_group_var_generic_1d(self):
+        prng = np.random.default_rng(2)
+
+        out = (np.nan * np.ones((5, 1))).astype(self.dtype)
+        counts = np.zeros(5, dtype="int64")
+        values = 10 * prng.random((15, 1)).astype(self.dtype)
+        labels = np.tile(np.arange(5), (3,)).astype("intp")
+
+        expected_out = (
+            np.squeeze(values).reshape((5, 3), order="F").std(axis=1, ddof=1) ** 2
+        )[:, np.newaxis]
+        expected_counts = counts + 3
+
+        self.algo(out, counts, values, labels)
+        assert np.allclose(out, expected_out, self.rtol)
+        tm.assert_numpy_array_equal(counts, expected_counts)
+
+    def test_group_var_generic_1d_flat_labels(self):
+        prng = np.random.default_rng(2)
+
+        out = (np.nan * np.ones((1, 1))).astype(self.dtype)
+        counts = np.zeros(1, dtype="int64")
+        values = 10 * prng.random((5, 1)).astype(self.dtype)
+        labels = np.zeros(5, dtype="intp")
+
+        expected_out = np.array([[values.std(ddof=1) ** 2]])
+        expected_counts = counts + 5
+
+        self.algo(out, counts, values, labels)
+
+        assert np.allclose(out, expected_out, self.rtol)
+        tm.assert_numpy_array_equal(counts, expected_counts)
+
+    def test_group_var_generic_2d_all_finite(self):
+        prng = np.random.default_rng(2)
+
+        out = (np.nan * np.ones((5, 2))).astype(self.dtype)
+        counts = np.zeros(5, dtype="int64")
+        values = 10 * prng.random((10, 2)).astype(self.dtype)
+        labels = np.tile(np.arange(5), (2,)).astype("intp")
+
+        expected_out = np.std(values.reshape(2, 5, 2), ddof=1, axis=0) ** 2
+        expected_counts = counts + 2
+
+        self.algo(out, counts, values, labels)
+        assert np.allclose(out, expected_out, self.rtol)
+        tm.assert_numpy_array_equal(counts, expected_counts)
+
+    def test_group_var_generic_2d_some_nan(self):
+        prng = np.random.default_rng(2)
+
+        out = (np.nan * np.ones((5, 2))).astype(self.dtype)
+        counts = np.zeros(5, dtype="int64")
+        values = 10 * prng.random((10, 2)).astype(self.dtype)
+        values[:, 1] = np.nan
+        labels = np.tile(np.arange(5), (2,)).astype("intp")
+
+        expected_out = np.vstack(
+            [
+                values[:, 0].reshape(5, 2, order="F").std(ddof=1, axis=1) ** 2,
+                np.nan * np.ones(5),
+            ]
+        ).T.astype(self.dtype)
+        expected_counts = counts + 2
+
+        self.algo(out, counts, values, labels)
+        tm.assert_almost_equal(out, expected_out, rtol=0.5e-06)
+        tm.assert_numpy_array_equal(counts, expected_counts)
+
+    def test_group_var_constant(self):
+        # Regression test from GH 10448.
+
+        out = np.array([[np.nan]], dtype=self.dtype)
+        counts = np.array([0], dtype="int64")
+        values = 0.832845131556193 * np.ones((3, 1), dtype=self.dtype)
+        labels = np.zeros(3, dtype="intp")
+
+        self.algo(out, counts, values, labels)
+
+        assert counts[0] == 3
+        assert out[0, 0] >= 0
+        tm.assert_almost_equal(out[0, 0], 0.0)
+
+
+class TestGroupVarFloat64(GroupVarTestMixin):
+    __test__ = True
+
+    algo = staticmethod(group_var)
+    dtype = np.float64
+    rtol = 1e-5
+
+    def test_group_var_large_inputs(self):
+        prng = np.random.default_rng(2)
+
+        out = np.array([[np.nan]], dtype=self.dtype)
+        counts = np.array([0], dtype="int64")
+        values = (prng.random(10**6) + 10**12).astype(self.dtype)
+        values.shape = (10**6, 1)
+        labels = np.zeros(10**6, dtype="intp")
+
+        self.algo(out, counts, values, labels)
+
+        assert counts[0] == 10**6
+        tm.assert_almost_equal(out[0, 0], 1.0 / 12, rtol=0.5e-3)
+
+
+class TestGroupVarFloat32(GroupVarTestMixin):
+    __test__ = True
+
+    algo = staticmethod(group_var)
+    dtype = np.float32
+    rtol = 1e-2
+
+
+@pytest.mark.parametrize("dtype", ["float32", "float64"])
+def test_group_ohlc(dtype):
+    obj = np.array(np.random.default_rng(2).standard_normal(20), dtype=dtype)
+
+    bins = np.array([6, 12, 20])
+    out = np.zeros((3, 4), dtype)
+    counts = np.zeros(len(out), dtype=np.int64)
+    labels = ensure_platform_int(np.repeat(np.arange(3), np.diff(np.r_[0, bins])))
+
+    func = libgroupby.group_ohlc
+    func(out, counts, obj[:, None], labels)
+
+    def _ohlc(group):
+        if isna(group).all():
+            return np.repeat(np.nan, 4)
+        return [group[0], group.max(), group.min(), group[-1]]
+
+    expected = np.array([_ohlc(obj[:6]), _ohlc(obj[6:12]), _ohlc(obj[12:])])
+
+    tm.assert_almost_equal(out, expected)
+    tm.assert_numpy_array_equal(counts, np.array([6, 6, 8], dtype=np.int64))
+
+    obj[:6] = np.nan
+    func(out, counts, obj[:, None], labels)
+    expected[0] = np.nan
+    tm.assert_almost_equal(out, expected)
+
+
+def _check_cython_group_transform_cumulative(pd_op, np_op, dtype):
+    """
+    Check a group transform that executes a cumulative function.
+
+    Parameters
+    ----------
+    pd_op : callable
+        The pandas cumulative function.
+    np_op : callable
+        The analogous one in NumPy.
+    dtype : type
+        The specified dtype of the data.
+    """
+    is_datetimelike = False
+
+    data = np.array([[1], [2], [3], [4]], dtype=dtype)
+    answer = np.zeros_like(data)
+
+    labels = np.array([0, 0, 0, 0], dtype=np.intp)
+    ngroups = 1
+    pd_op(answer, data, labels, ngroups, is_datetimelike)
+
+    tm.assert_numpy_array_equal(np_op(data), answer[:, 0], check_dtype=False)
+
+
+@pytest.mark.parametrize("np_dtype", ["int64", "uint64", "float32", "float64"])
+def test_cython_group_transform_cumsum(np_dtype):
+    # see gh-4095
+    dtype = np.dtype(np_dtype).type
+    pd_op, np_op = group_cumsum, np.cumsum
+    _check_cython_group_transform_cumulative(pd_op, np_op, dtype)
+
+
+def test_cython_group_transform_cumprod():
+    # see gh-4095
+    dtype = np.float64
+    pd_op, np_op = group_cumprod, np.cumprod
+    _check_cython_group_transform_cumulative(pd_op, np_op, dtype)
+
+
+def test_cython_group_transform_algos():
+    # see gh-4095
+    is_datetimelike = False
+
+    # with nans
+    labels = np.array([0, 0, 0, 0, 0], dtype=np.intp)
+    ngroups = 1
+
+    data = np.array([[1], [2], [3], [np.nan], [4]], dtype="float64")
+    actual = np.zeros_like(data)
+    actual.fill(np.nan)
+    group_cumprod(actual, data, labels, ngroups, is_datetimelike)
+    expected = np.array([1, 2, 6, np.nan, 24], dtype="float64")
+    tm.assert_numpy_array_equal(actual[:, 0], expected)
+
+    actual = np.zeros_like(data)
+    actual.fill(np.nan)
+    group_cumsum(actual, data, labels, ngroups, is_datetimelike)
+    expected = np.array([1, 3, 6, np.nan, 10], dtype="float64")
+    tm.assert_numpy_array_equal(actual[:, 0], expected)
+
+    # timedelta
+    is_datetimelike = True
+    data = np.array([np.timedelta64(1, "ns")] * 5, dtype="m8[ns]")[:, None]
+    actual = np.zeros_like(data, dtype="int64")
+    group_cumsum(actual, data.view("int64"), labels, ngroups, is_datetimelike)
+    expected = np.array(
+        [
+            np.timedelta64(1, "ns"),
+            np.timedelta64(2, "ns"),
+            np.timedelta64(3, "ns"),
+            np.timedelta64(4, "ns"),
+            np.timedelta64(5, "ns"),
+        ]
+    )
+    tm.assert_numpy_array_equal(actual[:, 0].view("m8[ns]"), expected)
+
+
+def test_cython_group_mean_datetimelike():
+    actual = np.zeros(shape=(1, 1), dtype="float64")
+    counts = np.array([0], dtype="int64")
+    data = (
+        np.array(
+            [np.timedelta64(2, "ns"), np.timedelta64(4, "ns"), np.timedelta64("NaT")],
+            dtype="m8[ns]",
+        )[:, None]
+        .view("int64")
+        .astype("float64")
+    )
+    labels = np.zeros(len(data), dtype=np.intp)
+
+    group_mean(actual, counts, data, labels, is_datetimelike=True)
+
+    tm.assert_numpy_array_equal(actual[:, 0], np.array([3], dtype="float64"))
+
+
+def test_cython_group_mean_wrong_min_count():
+    actual = np.zeros(shape=(1, 1), dtype="float64")
+    counts = np.zeros(1, dtype="int64")
+    data = np.zeros(1, dtype="float64")[:, None]
+    labels = np.zeros(1, dtype=np.intp)
+
+    with pytest.raises(AssertionError, match="min_count"):
+        group_mean(actual, counts, data, labels, is_datetimelike=True, min_count=0)
+
+
+def test_cython_group_mean_not_datetimelike_but_has_NaT_values():
+    actual = np.zeros(shape=(1, 1), dtype="float64")
+    counts = np.array([0], dtype="int64")
+    data = (
+        np.array(
+            [np.timedelta64("NaT"), np.timedelta64("NaT")],
+            dtype="m8[ns]",
+        )[:, None]
+        .view("int64")
+        .astype("float64")
+    )
+    labels = np.zeros(len(data), dtype=np.intp)
+
+    group_mean(actual, counts, data, labels, is_datetimelike=False)
+
+    tm.assert_numpy_array_equal(
+        actual[:, 0], np.array(np.divide(np.add(data[0], data[1]), 2), dtype="float64")
+    )
+
+
+def test_cython_group_mean_Inf_at_begining_and_end():
+    # GH 50367
+    actual = np.array([[np.nan, np.nan], [np.nan, np.nan]], dtype="float64")
+    counts = np.array([0, 0], dtype="int64")
+    data = np.array(
+        [[np.inf, 1.0], [1.0, 2.0], [2.0, 3.0], [3.0, 4.0], [4.0, 5.0], [5, np.inf]],
+        dtype="float64",
+    )
+    labels = np.array([0, 1, 0, 1, 0, 1], dtype=np.intp)
+
+    group_mean(actual, counts, data, labels, is_datetimelike=False)
+
+    expected = np.array([[np.inf, 3], [3, np.inf]], dtype="float64")
+
+    tm.assert_numpy_array_equal(
+        actual,
+        expected,
+    )
+
+
+@pytest.mark.parametrize(
+    "values, out",
+    [
+        ([[np.inf], [np.inf], [np.inf]], [[np.inf], [np.inf]]),
+        ([[np.inf], [np.inf], [-np.inf]], [[np.inf], [np.nan]]),
+        ([[np.inf], [-np.inf], [np.inf]], [[np.inf], [np.nan]]),
+        ([[np.inf], [-np.inf], [-np.inf]], [[np.inf], [-np.inf]]),
+    ],
+)
+def test_cython_group_sum_Inf_at_begining_and_end(values, out):
+    # GH #53606
+    actual = np.array([[np.nan], [np.nan]], dtype="float64")
+    counts = np.array([0, 0], dtype="int64")
+    data = np.array(values, dtype="float64")
+    labels = np.array([0, 1, 1], dtype=np.intp)
+
+    group_sum(actual, counts, data, labels, None, is_datetimelike=False)
+
+    expected = np.array(out, dtype="float64")
+
+    tm.assert_numpy_array_equal(
+        actual,
+        expected,
+    )
diff --git a/pandas/pandas/tests/groupby/test_missing.py b/pandas/pandas/tests/groupby/test_missing.py
new file mode 100644
index 0000000000000..2b590c50371e9
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_missing.py
@@ -0,0 +1,89 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill"])
+def test_groupby_column_index_name_lost_fill_funcs(func):
+    # GH: 29764 groupby loses index sometimes
+    df = DataFrame(
+        [[1, 1.0, -1.0], [1, np.nan, np.nan], [1, 2.0, -2.0]],
+        columns=Index(["type", "a", "b"], name="idx"),
+    )
+    df_grouped = df.groupby(["type"])[["a", "b"]]
+    result = getattr(df_grouped, func)().columns
+    expected = Index(["a", "b"], name="idx")
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill"])
+def test_groupby_fill_duplicate_column_names(func):
+    # GH: 25610 ValueError with duplicate column names
+    df1 = DataFrame({"field1": [1, 3, 4], "field2": [1, 3, 4]})
+    df2 = DataFrame({"field1": [1, np.nan, 4]})
+    df_grouped = pd.concat([df1, df2], axis=1).groupby(by=["field2"])
+    expected = DataFrame(
+        [[1, 1.0], [3, np.nan], [4, 4.0]], columns=["field1", "field1"]
+    )
+    result = getattr(df_grouped, func)()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["ffill", "bfill"])
+@pytest.mark.parametrize("has_nan_group", [True, False])
+def test_ffill_handles_nan_groups(dropna, method, has_nan_group):
+    # GH 34725
+
+    df_without_nan_rows = DataFrame([(1, 0.1), (2, 0.2)])
+
+    ridx = [-1, 0, -1, -1, 1, -1]
+    df = df_without_nan_rows.reindex(ridx).reset_index(drop=True)
+
+    group_b = np.nan if has_nan_group else "b"
+    df["group_col"] = pd.Series(["a"] * 3 + [group_b] * 3)
+
+    grouped = df.groupby(by="group_col", dropna=dropna)
+    result = getattr(grouped, method)(limit=None)
+
+    expected_rows = {
+        ("ffill", True, True): [-1, 0, 0, -1, -1, -1],
+        ("ffill", True, False): [-1, 0, 0, -1, 1, 1],
+        ("ffill", False, True): [-1, 0, 0, -1, 1, 1],
+        ("ffill", False, False): [-1, 0, 0, -1, 1, 1],
+        ("bfill", True, True): [0, 0, -1, -1, -1, -1],
+        ("bfill", True, False): [0, 0, -1, 1, 1, -1],
+        ("bfill", False, True): [0, 0, -1, 1, 1, -1],
+        ("bfill", False, False): [0, 0, -1, 1, 1, -1],
+    }
+
+    ridx = expected_rows.get((method, dropna, has_nan_group))
+    expected = df_without_nan_rows.reindex(ridx).reset_index(drop=True)
+    # columns are a 'take' on df.columns, which are object dtype
+    expected.columns = expected.columns.astype(object)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("min_count, value", [(2, np.nan), (-1, 1.0)])
+@pytest.mark.parametrize("func", ["first", "last", "max", "min"])
+def test_min_count(func, min_count, value):
+    # GH#37821
+    df = DataFrame({"a": [1] * 3, "b": [1, np.nan, np.nan], "c": [np.nan] * 3})
+    result = getattr(df.groupby("a"), func)(min_count=min_count)
+    expected = DataFrame({"b": [value], "c": [np.nan]}, index=Index([1], name="a"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_indices_with_missing():
+    # GH 9304
+    df = DataFrame({"a": [1, 1, np.nan], "b": [2, 3, 4], "c": [5, 6, 7]})
+    g = df.groupby(["a", "b"])
+    result = g.indices
+    expected = {(1.0, 2): np.array([0]), (1.0, 3): np.array([1])}
+    assert result == expected
diff --git a/pandas/pandas/tests/groupby/test_numba.py b/pandas/pandas/tests/groupby/test_numba.py
new file mode 100644
index 0000000000000..082319d8479f0
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_numba.py
@@ -0,0 +1,82 @@
+import pytest
+
+from pandas.compat import is_platform_arm
+
+from pandas import (
+    DataFrame,
+    Series,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [pytest.mark.single_cpu]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+class TestEngine:
+    def test_cython_vs_numba_frame(
+        self, sort, nogil, parallel, nopython, numba_supported_reductions
+    ):
+        func, kwargs = numba_supported_reductions
+        df = DataFrame({"a": [3, 2, 3, 2], "b": range(4), "c": range(1, 5)})
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        gb = df.groupby("a", sort=sort)
+        result = getattr(gb, func)(
+            engine="numba", engine_kwargs=engine_kwargs, **kwargs
+        )
+        expected = getattr(gb, func)(**kwargs)
+        tm.assert_frame_equal(result, expected)
+
+    def test_cython_vs_numba_getitem(
+        self, sort, nogil, parallel, nopython, numba_supported_reductions
+    ):
+        func, kwargs = numba_supported_reductions
+        df = DataFrame({"a": [3, 2, 3, 2], "b": range(4), "c": range(1, 5)})
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        gb = df.groupby("a", sort=sort)["c"]
+        result = getattr(gb, func)(
+            engine="numba", engine_kwargs=engine_kwargs, **kwargs
+        )
+        expected = getattr(gb, func)(**kwargs)
+        tm.assert_series_equal(result, expected)
+
+    def test_cython_vs_numba_series(
+        self, sort, nogil, parallel, nopython, numba_supported_reductions
+    ):
+        func, kwargs = numba_supported_reductions
+        ser = Series(range(3), index=[1, 2, 1], name="foo")
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        gb = ser.groupby(level=0, sort=sort)
+        result = getattr(gb, func)(
+            engine="numba", engine_kwargs=engine_kwargs, **kwargs
+        )
+        expected = getattr(gb, func)(**kwargs)
+        tm.assert_series_equal(result, expected)
+
+    def test_as_index_false_unsupported(self, numba_supported_reductions):
+        func, kwargs = numba_supported_reductions
+        df = DataFrame({"a": [3, 2, 3, 2], "b": range(4), "c": range(1, 5)})
+        gb = df.groupby("a", as_index=False)
+        with pytest.raises(NotImplementedError, match="as_index=False"):
+            getattr(gb, func)(engine="numba", **kwargs)
+
+    def test_no_engine_doesnt_raise(self):
+        # GH55520
+        df = DataFrame({"a": [3, 2, 3, 2], "b": range(4), "c": range(1, 5)})
+        gb = df.groupby("a")
+        # Make sure behavior of functions w/out engine argument don't raise
+        # when the global use_numba option is set
+        with option_context("compute.use_numba", True):
+            res = gb.agg({"b": "first"})
+        expected = gb.agg({"b": "first"})
+        tm.assert_frame_equal(res, expected)
diff --git a/pandas/pandas/tests/groupby/test_numeric_only.py b/pandas/pandas/tests/groupby/test_numeric_only.py
new file mode 100644
index 0000000000000..99a88a5d8fe7c
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_numeric_only.py
@@ -0,0 +1,444 @@
+import re
+
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+class TestNumericOnly:
+    # make sure that we are passing thru kwargs to our agg functions
+
+    @pytest.fixture
+    def df(self):
+        # GH3668
+        # GH5724
+        df = DataFrame(
+            {
+                "group": [1, 1, 2],
+                "int": [1, 2, 3],
+                "float": [4.0, 5.0, 6.0],
+                "string": Series(["a", "b", "c"], dtype="str"),
+                "object": Series(["a", "b", "c"], dtype=object),
+                "category_string": Series(list("abc")).astype("category"),
+                "category_int": [7, 8, 9],
+                "datetime": date_range("20130101", periods=3),
+                "datetimetz": date_range("20130101", periods=3, tz="US/Eastern"),
+                "timedelta": pd.timedelta_range("1 s", periods=3, freq="s"),
+            },
+            columns=[
+                "group",
+                "int",
+                "float",
+                "string",
+                "object",
+                "category_string",
+                "category_int",
+                "datetime",
+                "datetimetz",
+                "timedelta",
+            ],
+        )
+        return df
+
+    @pytest.mark.parametrize("method", ["mean", "median"])
+    def test_averages(self, df, method):
+        # mean / median
+        expected_columns_numeric = Index(["int", "float", "category_int"])
+
+        gb = df.groupby("group")
+        expected = DataFrame(
+            {
+                "category_int": [7.5, 9],
+                "float": [4.5, 6.0],
+                "timedelta": [pd.Timedelta("1.5s"), pd.Timedelta("3s")],
+                "int": [1.5, 3],
+                "datetime": [
+                    Timestamp("2013-01-01 12:00:00"),
+                    Timestamp("2013-01-03 00:00:00"),
+                ],
+                "datetimetz": [
+                    Timestamp("2013-01-01 12:00:00", tz="US/Eastern"),
+                    Timestamp("2013-01-03 00:00:00", tz="US/Eastern"),
+                ],
+            },
+            index=Index([1, 2], name="group"),
+            columns=[
+                "int",
+                "float",
+                "category_int",
+            ],
+        )
+
+        result = getattr(gb, method)(numeric_only=True)
+        tm.assert_frame_equal(result.reindex_like(expected), expected)
+
+        expected_columns = expected.columns
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    @pytest.mark.parametrize("method", ["min", "max"])
+    def test_extrema(self, df, method):
+        # TODO: min, max *should* handle
+        # categorical (ordered) dtype
+
+        expected_columns = Index(
+            [
+                "int",
+                "float",
+                "string",
+                "category_int",
+                "datetime",
+                "datetimetz",
+                "timedelta",
+            ]
+        )
+        expected_columns_numeric = expected_columns
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    @pytest.mark.parametrize("method", ["first", "last"])
+    def test_first_last(self, df, method):
+        expected_columns = Index(
+            [
+                "int",
+                "float",
+                "string",
+                "object",
+                "category_string",
+                "category_int",
+                "datetime",
+                "datetimetz",
+                "timedelta",
+            ]
+        )
+        expected_columns_numeric = expected_columns
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    @pytest.mark.parametrize("method", ["sum", "cumsum"])
+    def test_sum_cumsum(self, df, method):
+        expected_columns_numeric = Index(["int", "float", "category_int"])
+        expected_columns = Index(
+            ["int", "float", "string", "category_int", "timedelta"]
+        )
+        if method == "cumsum":
+            # cumsum loses string
+            expected_columns = Index(["int", "float", "category_int", "timedelta"])
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    @pytest.mark.parametrize("method", ["prod", "cumprod"])
+    def test_prod_cumprod(self, df, method):
+        expected_columns = Index(["int", "float", "category_int"])
+        expected_columns_numeric = expected_columns
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    @pytest.mark.parametrize("method", ["cummin", "cummax"])
+    def test_cummin_cummax(self, df, method):
+        # like min, max, but don't include strings
+        expected_columns = Index(
+            ["int", "float", "category_int", "datetime", "datetimetz", "timedelta"]
+        )
+
+        # GH#15561: numeric_only=False set by default like min/max
+        expected_columns_numeric = expected_columns
+
+        self._check(df, method, expected_columns, expected_columns_numeric)
+
+    def _check(self, df, method, expected_columns, expected_columns_numeric):
+        gb = df.groupby("group")
+
+        # object dtypes for transformations are not implemented in Cython and
+        # have no Python fallback
+        exception = (
+            (NotImplementedError, TypeError) if method.startswith("cum") else TypeError
+        )
+
+        if method in ("min", "max", "cummin", "cummax", "cumsum", "cumprod"):
+            # The methods default to numeric_only=False and raise TypeError
+            msg = "|".join(
+                [
+                    "Categorical is not ordered",
+                    f"Cannot perform {method} with non-ordered Categorical",
+                    re.escape(f"agg function failed [how->{method},dtype->object]"),
+                    # cumsum/cummin/cummax/cumprod
+                    "function is not implemented for this dtype",
+                    f"dtype 'str' does not support operation '{method}'",
+                ]
+            )
+            with pytest.raises(exception, match=msg):
+                getattr(gb, method)()
+        elif method in ("sum", "mean", "median", "prod"):
+            msg = "|".join(
+                [
+                    "category type does not support sum operations",
+                    re.escape(f"agg function failed [how->{method},dtype->object]"),
+                    re.escape(f"agg function failed [how->{method},dtype->string]"),
+                    f"dtype 'str' does not support operation '{method}'",
+                ]
+            )
+            with pytest.raises(exception, match=msg):
+                getattr(gb, method)()
+        else:
+            result = getattr(gb, method)()
+            tm.assert_index_equal(result.columns, expected_columns_numeric)
+
+        if method not in ("first", "last"):
+            msg = "|".join(
+                [
+                    "Categorical is not ordered",
+                    "category type does not support",
+                    "function is not implemented for this dtype",
+                    f"Cannot perform {method} with non-ordered Categorical",
+                    re.escape(f"agg function failed [how->{method},dtype->object]"),
+                    re.escape(f"agg function failed [how->{method},dtype->string]"),
+                    f"dtype 'str' does not support operation '{method}'",
+                ]
+            )
+            with pytest.raises(exception, match=msg):
+                getattr(gb, method)(numeric_only=False)
+        else:
+            result = getattr(gb, method)(numeric_only=False)
+            tm.assert_index_equal(result.columns, expected_columns)
+
+
+@pytest.mark.parametrize(
+    "kernel, has_arg",
+    [
+        ("all", False),
+        ("any", False),
+        ("bfill", False),
+        ("corr", True),
+        ("corrwith", True),
+        ("cov", True),
+        ("cummax", True),
+        ("cummin", True),
+        ("cumprod", True),
+        ("cumsum", True),
+        ("diff", False),
+        ("ffill", False),
+        ("first", True),
+        ("idxmax", True),
+        ("idxmin", True),
+        ("last", True),
+        ("max", True),
+        ("mean", True),
+        ("median", True),
+        ("min", True),
+        ("nth", False),
+        ("nunique", False),
+        ("pct_change", False),
+        ("prod", True),
+        ("quantile", True),
+        ("sem", True),
+        ("skew", True),
+        ("kurt", True),
+        ("std", True),
+        ("sum", True),
+        ("var", True),
+    ],
+)
+@pytest.mark.parametrize("numeric_only", [True, False, lib.no_default])
+@pytest.mark.parametrize("keys", [["a1"], ["a1", "a2"]])
+def test_numeric_only(kernel, has_arg, numeric_only, keys):
+    # GH#46072
+    # drops_nuisance: Whether the op drops nuisance columns even when numeric_only=False
+    # has_arg: Whether the op has a numeric_only arg
+    df = DataFrame({"a1": [1, 1], "a2": [2, 2], "a3": [5, 6], "b": 2 * [object]})
+
+    args = get_groupby_method_args(kernel, df)
+    kwargs = {} if numeric_only is lib.no_default else {"numeric_only": numeric_only}
+
+    gb = df.groupby(keys)
+    method = getattr(gb, kernel)
+    if has_arg and numeric_only is True:
+        # Cases where b does not appear in the result
+        if kernel == "corrwith":
+            warn = FutureWarning
+            msg = "DataFrameGroupBy.corrwith is deprecated"
+        else:
+            warn = None
+            msg = ""
+        with tm.assert_produces_warning(warn, match=msg):
+            result = method(*args, **kwargs)
+        assert "b" not in result.columns
+    elif (
+        # kernels that work on any dtype and have numeric_only arg
+        kernel in ("first", "last")
+        or (
+            # kernels that work on any dtype and don't have numeric_only arg
+            kernel in ("any", "all", "bfill", "ffill", "nth", "nunique")
+            and numeric_only is lib.no_default
+        )
+    ):
+        result = method(*args, **kwargs)
+        assert "b" in result.columns
+    elif has_arg:
+        assert numeric_only is not True
+        # kernels that are successful on any dtype were above; this will fail
+
+        # object dtypes for transformations are not implemented in Cython and
+        # have no Python fallback
+        exception = NotImplementedError if kernel.startswith("cum") else TypeError
+
+        msg = "|".join(
+            [
+                "not allowed for this dtype",
+                "cannot be performed against 'object' dtypes",
+                "must be a string or a real number",
+                "unsupported operand type",
+                "function is not implemented for this dtype",
+                re.escape(f"agg function failed [how->{kernel},dtype->object]"),
+            ]
+        )
+        if kernel == "quantile":
+            msg = "dtype 'object' does not support operation 'quantile'"
+        elif kernel == "idxmin":
+            msg = "'<' not supported between instances of 'type' and 'type'"
+        elif kernel == "idxmax":
+            msg = "'>' not supported between instances of 'type' and 'type'"
+        with pytest.raises(exception, match=msg):
+            if kernel == "corrwith":
+                warn = FutureWarning
+                msg = "DataFrameGroupBy.corrwith is deprecated"
+            else:
+                warn = None
+                msg = ""
+            with tm.assert_produces_warning(warn, match=msg):
+                method(*args, **kwargs)
+    elif not has_arg and numeric_only is not lib.no_default:
+        with pytest.raises(
+            TypeError, match="got an unexpected keyword argument 'numeric_only'"
+        ):
+            method(*args, **kwargs)
+    else:
+        assert kernel in ("diff", "pct_change")
+        assert numeric_only is lib.no_default
+        # Doesn't have numeric_only argument and fails on nuisance columns
+        with pytest.raises(TypeError, match=r"unsupported operand type"):
+            method(*args, **kwargs)
+
+
+@pytest.mark.parametrize("dtype", [bool, int, float, object])
+def test_deprecate_numeric_only_series(dtype, groupby_func, request):
+    # GH#46560
+    grouper = [0, 0, 1]
+
+    ser = Series([1, 0, 0], dtype=dtype)
+    gb = ser.groupby(grouper)
+
+    if groupby_func == "corrwith":
+        # corrwith is not implemented on SeriesGroupBy
+        assert not hasattr(gb, groupby_func)
+        return
+
+    method = getattr(gb, groupby_func)
+
+    expected_ser = Series([1, 0, 0])
+    expected_gb = expected_ser.groupby(grouper)
+    expected_method = getattr(expected_gb, groupby_func)
+
+    args = get_groupby_method_args(groupby_func, ser)
+
+    fails_on_numeric_object = (
+        "corr",
+        "cov",
+        "cummax",
+        "cummin",
+        "cumprod",
+        "cumsum",
+        "quantile",
+    )
+    # ops that give an object result on object input
+    obj_result = (
+        "first",
+        "last",
+        "nth",
+        "bfill",
+        "ffill",
+        "shift",
+        "sum",
+        "diff",
+        "pct_change",
+        "var",
+        "mean",
+        "median",
+        "min",
+        "max",
+        "prod",
+        "skew",
+        "kurt",
+    )
+
+    # Test default behavior; kernels that fail may be enabled in the future but kernels
+    # that succeed should not be allowed to fail (without deprecation, at least)
+    if groupby_func in fails_on_numeric_object and dtype is object:
+        if groupby_func == "quantile":
+            msg = "dtype 'object' does not support operation 'quantile'"
+        else:
+            msg = "is not supported for object dtype"
+        with pytest.raises(TypeError, match=msg):
+            method(*args)
+    elif dtype is object:
+        result = method(*args)
+        expected = expected_method(*args)
+        if groupby_func in obj_result:
+            expected = expected.astype(object)
+        tm.assert_series_equal(result, expected)
+
+    has_numeric_only = (
+        "first",
+        "last",
+        "max",
+        "mean",
+        "median",
+        "min",
+        "prod",
+        "quantile",
+        "sem",
+        "skew",
+        "kurt",
+        "std",
+        "sum",
+        "var",
+        "cummax",
+        "cummin",
+        "cumprod",
+        "cumsum",
+    )
+    if groupby_func not in has_numeric_only:
+        msg = "got an unexpected keyword argument 'numeric_only'"
+        with pytest.raises(TypeError, match=msg):
+            method(*args, numeric_only=True)
+    elif dtype is object:
+        msg = "|".join(
+            [
+                "SeriesGroupBy.sem called with numeric_only=True and dtype object",
+                "Series.skew does not allow numeric_only=True with non-numeric",
+                "cum(sum|prod|min|max) is not supported for object dtype",
+                r"Cannot use numeric_only=True with SeriesGroupBy\..* and non-numeric",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            method(*args, numeric_only=True)
+    elif dtype == bool and groupby_func == "quantile":
+        msg = "Cannot use quantile with bool dtype"
+        with pytest.raises(TypeError, match=msg):
+            # GH#51424
+            method(*args, numeric_only=False)
+    else:
+        result = method(*args, numeric_only=True)
+        expected = method(*args, numeric_only=False)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_pipe.py b/pandas/pandas/tests/groupby/test_pipe.py
new file mode 100644
index 0000000000000..ee59a93695bcf
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_pipe.py
@@ -0,0 +1,80 @@
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+def test_pipe():
+    # Test the pipe method of DataFrameGroupBy.
+    # Issue #17871
+
+    random_state = np.random.default_rng(2)
+
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": random_state.standard_normal(8),
+            "C": random_state.standard_normal(8),
+        }
+    )
+
+    def f(dfgb):
+        return dfgb.B.max() - dfgb.C.min().min()
+
+    def square(srs):
+        return srs**2
+
+    # Note that the transformations are
+    # GroupBy -> Series
+    # Series -> Series
+    # This then chains the GroupBy.pipe and the
+    # NDFrame.pipe methods
+    result = df.groupby("A").pipe(f).pipe(square)
+
+    index = Index(["bar", "foo"], name="A")
+    expected = pd.Series([3.749306591013693, 6.717707873081384], name="B", index=index)
+
+    tm.assert_series_equal(expected, result)
+
+
+def test_pipe_args():
+    # Test passing args to the pipe method of DataFrameGroupBy.
+    # Issue #17871
+
+    df = DataFrame(
+        {
+            "group": ["A", "A", "B", "B", "C"],
+            "x": [1.0, 2.0, 3.0, 2.0, 5.0],
+            "y": [10.0, 100.0, 1000.0, -100.0, -1000.0],
+        }
+    )
+
+    def f(dfgb, arg1):
+        filtered = dfgb.filter(lambda grp: grp.y.mean() > arg1, dropna=False)
+        return filtered.groupby("group")
+
+    def g(dfgb, arg2):
+        return dfgb.sum() / dfgb.sum().sum() + arg2
+
+    def h(df, arg3):
+        return df.x + df.y - arg3
+
+    result = df.groupby("group").pipe(f, 0).pipe(g, 10).pipe(h, 100)
+
+    # Assert the results here
+    index = Index(["A", "B"], name="group")
+    expected = pd.Series([-79.5160891089, -78.4839108911], index=index)
+
+    tm.assert_series_equal(result, expected)
+
+    # test SeriesGroupby.pipe
+    ser = pd.Series([1, 1, 2, 2, 3, 3])
+    result = ser.groupby(ser).pipe(lambda grp: grp.sum() * grp.count())
+
+    expected = pd.Series([4, 8, 12], index=Index([1, 2, 3], dtype=np.int64))
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_raises.py b/pandas/pandas/tests/groupby/test_raises.py
new file mode 100644
index 0000000000000..864b9e5d55991
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_raises.py
@@ -0,0 +1,730 @@
+# Only tests that raise an error and have no better location should go here.
+# Tests for specific groupby methods should go in their respective
+# test file.
+
+import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Grouper,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+@pytest.fixture(
+    params=[
+        "a",
+        ["a"],
+        ["a", "b"],
+        Grouper(key="a"),
+        lambda x: x % 2,
+        [0, 0, 0, 1, 2, 2, 2, 3, 3],
+        np.array([0, 0, 0, 1, 2, 2, 2, 3, 3]),
+        dict(zip(range(9), [0, 0, 0, 1, 2, 2, 2, 3, 3])),
+        Series([1, 1, 1, 1, 1, 2, 2, 2, 2]),
+        [Series([1, 1, 1, 1, 1, 2, 2, 2, 2]), Series([3, 3, 4, 4, 4, 4, 4, 3, 3])],
+    ]
+)
+def by(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def groupby_series(request):
+    return request.param
+
+
+@pytest.fixture
+def df_with_string_col():
+    df = DataFrame(
+        {
+            "a": [1, 1, 1, 1, 1, 2, 2, 2, 2],
+            "b": [3, 3, 4, 4, 4, 4, 4, 3, 3],
+            "c": range(9),
+            "d": list("xyzwtyuio"),
+        }
+    )
+    return df
+
+
+@pytest.fixture
+def df_with_datetime_col():
+    df = DataFrame(
+        {
+            "a": [1, 1, 1, 1, 1, 2, 2, 2, 2],
+            "b": [3, 3, 4, 4, 4, 4, 4, 3, 3],
+            "c": range(9),
+            "d": datetime.datetime(2005, 1, 1, 10, 30, 23, 540000),
+        }
+    )
+    return df
+
+
+@pytest.fixture
+def df_with_cat_col():
+    df = DataFrame(
+        {
+            "a": [1, 1, 1, 1, 1, 2, 2, 2, 2],
+            "b": [3, 3, 4, 4, 4, 4, 4, 3, 3],
+            "c": range(9),
+            "d": Categorical(
+                ["a", "a", "a", "a", "b", "b", "b", "b", "c"],
+                categories=["a", "b", "c", "d"],
+                ordered=True,
+            ),
+        }
+    )
+    return df
+
+
+def _call_and_check(klass, msg, how, gb, groupby_func, args, warn_msg=""):
+    warn_klass = None if warn_msg == "" else FutureWarning
+    with tm.assert_produces_warning(warn_klass, match=warn_msg, check_stacklevel=False):
+        if klass is None:
+            if how == "method":
+                getattr(gb, groupby_func)(*args)
+            elif how == "agg":
+                gb.agg(groupby_func, *args)
+            else:
+                gb.transform(groupby_func, *args)
+        else:
+            with pytest.raises(klass, match=msg):
+                if how == "method":
+                    getattr(gb, groupby_func)(*args)
+                elif how == "agg":
+                    gb.agg(groupby_func, *args)
+                else:
+                    gb.transform(groupby_func, *args)
+
+
+@pytest.mark.parametrize("how", ["method", "agg", "transform"])
+def test_groupby_raises_string(
+    how, by, groupby_series, groupby_func, df_with_string_col, using_infer_string
+):
+    df = df_with_string_col
+    args = get_groupby_method_args(groupby_func, df)
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+        if groupby_func == "corrwith":
+            assert not hasattr(gb, "corrwith")
+            return
+
+    klass, msg = {
+        "all": (None, ""),
+        "any": (None, ""),
+        "bfill": (None, ""),
+        "corrwith": (TypeError, "Could not convert"),
+        "count": (None, ""),
+        "cumcount": (None, ""),
+        "cummax": (
+            (NotImplementedError, TypeError),
+            "(function|cummax) is not (implemented|supported) for (this|object) dtype",
+        ),
+        "cummin": (
+            (NotImplementedError, TypeError),
+            "(function|cummin) is not (implemented|supported) for (this|object) dtype",
+        ),
+        "cumprod": (
+            (NotImplementedError, TypeError),
+            "(function|cumprod) is not (implemented|supported) for (this|object) dtype",
+        ),
+        "cumsum": (
+            (NotImplementedError, TypeError),
+            "(function|cumsum) is not (implemented|supported) for (this|object) dtype",
+        ),
+        "diff": (TypeError, "unsupported operand type"),
+        "ffill": (None, ""),
+        "first": (None, ""),
+        "idxmax": (None, ""),
+        "idxmin": (None, ""),
+        "last": (None, ""),
+        "max": (None, ""),
+        "mean": (
+            TypeError,
+            re.escape("agg function failed [how->mean,dtype->object]"),
+        ),
+        "median": (
+            TypeError,
+            re.escape("agg function failed [how->median,dtype->object]"),
+        ),
+        "min": (None, ""),
+        "ngroup": (None, ""),
+        "nunique": (None, ""),
+        "pct_change": (TypeError, "unsupported operand type"),
+        "prod": (
+            TypeError,
+            re.escape("agg function failed [how->prod,dtype->object]"),
+        ),
+        "quantile": (TypeError, "dtype 'object' does not support operation 'quantile'"),
+        "rank": (None, ""),
+        "sem": (ValueError, "could not convert string to float"),
+        "shift": (None, ""),
+        "size": (None, ""),
+        "skew": (ValueError, "could not convert string to float"),
+        "kurt": (ValueError, "could not convert string to float"),
+        "std": (ValueError, "could not convert string to float"),
+        "sum": (None, ""),
+        "var": (
+            TypeError,
+            re.escape("agg function failed [how->var,dtype->"),
+        ),
+    }[groupby_func]
+
+    if using_infer_string:
+        if groupby_func in [
+            "prod",
+            "mean",
+            "median",
+            "cumsum",
+            "cumprod",
+            "std",
+            "sem",
+            "var",
+            "skew",
+            "kurt",
+            "quantile",
+        ]:
+            msg = f"dtype 'str' does not support operation '{groupby_func}'"
+            if groupby_func in ["sem", "std", "skew", "kurt"]:
+                # The object-dtype raises ValueError when trying to convert to numeric.
+                klass = TypeError
+        elif groupby_func == "pct_change" and df["d"].dtype.storage == "pyarrow":
+            # This doesn't go through EA._groupby_op so the message isn't controlled
+            #  there.
+            msg = "operation 'truediv' not supported for dtype 'str' with dtype 'str'"
+        elif groupby_func == "diff" and df["d"].dtype.storage == "pyarrow":
+            # This doesn't go through EA._groupby_op so the message isn't controlled
+            #  there.
+            msg = "operation 'sub' not supported for dtype 'str' with dtype 'str'"
+
+        elif groupby_func in ["cummin", "cummax"]:
+            msg = msg.replace("object", "str")
+        elif groupby_func == "corrwith":
+            msg = "Cannot perform reduction 'mean' with string dtype"
+
+    if groupby_func == "corrwith":
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn_msg = ""
+    _call_and_check(klass, msg, how, gb, groupby_func, args, warn_msg)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+def test_groupby_raises_string_udf(how, by, groupby_series, df_with_string_col):
+    df = df_with_string_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    def func(x):
+        raise TypeError("Test error message")
+
+    with pytest.raises(TypeError, match="Test error message"):
+        getattr(gb, how)(func)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+@pytest.mark.parametrize("groupby_func_np", [np.sum, np.mean])
+def test_groupby_raises_string_np(
+    how,
+    by,
+    groupby_series,
+    groupby_func_np,
+    df_with_string_col,
+    using_infer_string,
+):
+    # GH#50749
+    df = df_with_string_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    klass, msg = {
+        np.sum: (None, ""),
+        np.mean: (
+            TypeError,
+            "Could not convert string .* to numeric|"
+            "Cannot perform reduction 'mean' with string dtype",
+        ),
+    }[groupby_func_np]
+
+    if using_infer_string:
+        if groupby_func_np is np.mean:
+            klass = TypeError
+        msg = f"Cannot perform reduction '{groupby_func_np.__name__}' with string dtype"
+
+    _call_and_check(klass, msg, how, gb, groupby_func_np, ())
+
+
+@pytest.mark.parametrize("how", ["method", "agg", "transform"])
+def test_groupby_raises_datetime(
+    how, by, groupby_series, groupby_func, df_with_datetime_col
+):
+    df = df_with_datetime_col
+    args = get_groupby_method_args(groupby_func, df)
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+        if groupby_func == "corrwith":
+            assert not hasattr(gb, "corrwith")
+            return
+
+    klass, msg = {
+        "all": (TypeError, "'all' with datetime64 dtypes is no longer supported"),
+        "any": (TypeError, "'any' with datetime64 dtypes is no longer supported"),
+        "bfill": (None, ""),
+        "corrwith": (TypeError, "cannot perform __mul__ with this index type"),
+        "count": (None, ""),
+        "cumcount": (None, ""),
+        "cummax": (None, ""),
+        "cummin": (None, ""),
+        "cumprod": (TypeError, "datetime64 type does not support operation 'cumprod'"),
+        "cumsum": (TypeError, "datetime64 type does not support operation 'cumsum'"),
+        "diff": (None, ""),
+        "ffill": (None, ""),
+        "first": (None, ""),
+        "idxmax": (None, ""),
+        "idxmin": (None, ""),
+        "last": (None, ""),
+        "max": (None, ""),
+        "mean": (None, ""),
+        "median": (None, ""),
+        "min": (None, ""),
+        "ngroup": (None, ""),
+        "nunique": (None, ""),
+        "pct_change": (TypeError, "cannot perform __truediv__ with this index type"),
+        "prod": (TypeError, "datetime64 type does not support operation 'prod'"),
+        "quantile": (None, ""),
+        "rank": (None, ""),
+        "sem": (None, ""),
+        "shift": (None, ""),
+        "size": (None, ""),
+        "skew": (
+            TypeError,
+            "|".join(
+                [
+                    r"dtype datetime64\[ns\] does not support operation",
+                    "datetime64 type does not support operation 'skew'",
+                ]
+            ),
+        ),
+        "kurt": (
+            TypeError,
+            "|".join(
+                [
+                    r"dtype datetime64\[ns\] does not support operation",
+                    "datetime64 type does not support operation 'kurt'",
+                ]
+            ),
+        ),
+        "std": (None, ""),
+        "sum": (TypeError, "datetime64 type does not support operation 'sum"),
+        "var": (TypeError, "datetime64 type does not support operation 'var'"),
+    }[groupby_func]
+
+    if groupby_func == "corrwith":
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn_msg = ""
+    _call_and_check(klass, msg, how, gb, groupby_func, args, warn_msg=warn_msg)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+def test_groupby_raises_datetime_udf(how, by, groupby_series, df_with_datetime_col):
+    df = df_with_datetime_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    def func(x):
+        raise TypeError("Test error message")
+
+    with pytest.raises(TypeError, match="Test error message"):
+        getattr(gb, how)(func)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+@pytest.mark.parametrize("groupby_func_np", [np.sum, np.mean])
+def test_groupby_raises_datetime_np(
+    how, by, groupby_series, groupby_func_np, df_with_datetime_col
+):
+    # GH#50749
+    df = df_with_datetime_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    klass, msg = {
+        np.sum: (
+            TypeError,
+            re.escape("datetime64[us] does not support operation 'sum'"),
+        ),
+        np.mean: (None, ""),
+    }[groupby_func_np]
+    _call_and_check(klass, msg, how, gb, groupby_func_np, ())
+
+
+@pytest.mark.parametrize("func", ["prod", "cumprod", "skew", "kurt", "var"])
+def test_groupby_raises_timedelta(func):
+    df = DataFrame(
+        {
+            "a": [1, 1, 1, 1, 1, 2, 2, 2, 2],
+            "b": [3, 3, 4, 4, 4, 4, 4, 3, 3],
+            "c": range(9),
+            "d": datetime.timedelta(days=1),
+        }
+    )
+    gb = df.groupby(by="a")
+
+    _call_and_check(
+        TypeError,
+        "timedelta64 type does not support .* operations",
+        "method",
+        gb,
+        func,
+        [],
+    )
+
+
+@pytest.mark.parametrize("how", ["method", "agg", "transform"])
+def test_groupby_raises_category(
+    how, by, groupby_series, groupby_func, df_with_cat_col
+):
+    # GH#50749
+    df = df_with_cat_col
+    args = get_groupby_method_args(groupby_func, df)
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+        if groupby_func == "corrwith":
+            assert not hasattr(gb, "corrwith")
+            return
+
+    klass, msg = {
+        "all": (None, ""),
+        "any": (None, ""),
+        "bfill": (None, ""),
+        "corrwith": (
+            TypeError,
+            r"unsupported operand type\(s\) for \*: 'Categorical' and 'int'",
+        ),
+        "count": (None, ""),
+        "cumcount": (None, ""),
+        "cummax": (
+            (NotImplementedError, TypeError),
+            "(category type does not support cummax operations|"
+            "category dtype not supported|"
+            "cummax is not supported for category dtype)",
+        ),
+        "cummin": (
+            (NotImplementedError, TypeError),
+            "(category type does not support cummin operations|"
+            "category dtype not supported|"
+            "cummin is not supported for category dtype)",
+        ),
+        "cumprod": (
+            (NotImplementedError, TypeError),
+            "(category type does not support cumprod operations|"
+            "category dtype not supported|"
+            "cumprod is not supported for category dtype)",
+        ),
+        "cumsum": (
+            (NotImplementedError, TypeError),
+            "(category type does not support cumsum operations|"
+            "category dtype not supported|"
+            "cumsum is not supported for category dtype)",
+        ),
+        "diff": (
+            TypeError,
+            r"unsupported operand type\(s\) for -: 'Categorical' and 'Categorical'",
+        ),
+        "ffill": (None, ""),
+        "first": (None, ""),
+        "idxmax": (None, ""),
+        "idxmin": (None, ""),
+        "last": (None, ""),
+        "max": (None, ""),
+        "mean": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'mean'",
+                    "category dtype does not support aggregation 'mean'",
+                ]
+            ),
+        ),
+        "median": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'median'",
+                    "category dtype does not support aggregation 'median'",
+                ]
+            ),
+        ),
+        "min": (None, ""),
+        "ngroup": (None, ""),
+        "nunique": (None, ""),
+        "pct_change": (
+            TypeError,
+            r"unsupported operand type\(s\) for /: 'Categorical' and 'Categorical'",
+        ),
+        "prod": (TypeError, "category type does not support prod operations"),
+        "quantile": (TypeError, "No matching signature found"),
+        "rank": (None, ""),
+        "sem": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'sem'",
+                    "category dtype does not support aggregation 'sem'",
+                ]
+            ),
+        ),
+        "shift": (None, ""),
+        "size": (None, ""),
+        "skew": (
+            TypeError,
+            "|".join(
+                [
+                    "dtype category does not support operation 'skew'",
+                    "category type does not support skew operations",
+                ]
+            ),
+        ),
+        "kurt": (
+            TypeError,
+            "|".join(
+                [
+                    "dtype category does not support operation 'kurt'",
+                    "category type does not support kurt operations",
+                ]
+            ),
+        ),
+        "std": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'std'",
+                    "category dtype does not support aggregation 'std'",
+                ]
+            ),
+        ),
+        "sum": (TypeError, "category type does not support sum operations"),
+        "var": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'var'",
+                    "category dtype does not support aggregation 'var'",
+                ]
+            ),
+        ),
+    }[groupby_func]
+
+    if groupby_func == "corrwith":
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn_msg = ""
+    _call_and_check(klass, msg, how, gb, groupby_func, args, warn_msg)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+def test_groupby_raises_category_udf(how, by, groupby_series, df_with_cat_col):
+    # GH#50749
+    df = df_with_cat_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    def func(x):
+        raise TypeError("Test error message")
+
+    with pytest.raises(TypeError, match="Test error message"):
+        getattr(gb, how)(func)
+
+
+@pytest.mark.parametrize("how", ["agg", "transform"])
+@pytest.mark.parametrize("groupby_func_np", [np.sum, np.mean])
+def test_groupby_raises_category_np(
+    how, by, groupby_series, groupby_func_np, df_with_cat_col
+):
+    # GH#50749
+    df = df_with_cat_col
+    gb = df.groupby(by=by)
+
+    if groupby_series:
+        gb = gb["d"]
+
+    klass, msg = {
+        np.sum: (TypeError, "dtype category does not support operation 'sum'"),
+        np.mean: (
+            TypeError,
+            "dtype category does not support operation 'mean'",
+        ),
+    }[groupby_func_np]
+    _call_and_check(klass, msg, how, gb, groupby_func_np, ())
+
+
+@pytest.mark.filterwarnings(
+    "ignore:`groups` by one element list returns scalar is deprecated"
+)
+@pytest.mark.parametrize("how", ["method", "agg", "transform"])
+def test_groupby_raises_category_on_category(
+    how,
+    by,
+    groupby_series,
+    groupby_func,
+    observed,
+    df_with_cat_col,
+):
+    # GH#50749
+    df = df_with_cat_col
+    df["a"] = Categorical(
+        ["a", "a", "a", "a", "b", "b", "b", "b", "c"],
+        categories=["a", "b", "c", "d"],
+        ordered=True,
+    )
+    args = get_groupby_method_args(groupby_func, df)
+    gb = df.groupby(by=by, observed=observed)
+
+    if groupby_series:
+        gb = gb["d"]
+
+        if groupby_func == "corrwith":
+            assert not hasattr(gb, "corrwith")
+            return
+
+    empty_groups = not observed and any(group.empty for group in gb.groups.values())
+    if how == "transform":
+        # empty groups will be ignored
+        empty_groups = False
+
+    klass, msg = {
+        "all": (None, ""),
+        "any": (None, ""),
+        "bfill": (None, ""),
+        "corrwith": (
+            TypeError,
+            r"unsupported operand type\(s\) for \*: 'Categorical' and 'int'",
+        ),
+        "count": (None, ""),
+        "cumcount": (None, ""),
+        "cummax": (
+            (NotImplementedError, TypeError),
+            "(cummax is not supported for category dtype|"
+            "category dtype not supported|"
+            "category type does not support cummax operations)",
+        ),
+        "cummin": (
+            (NotImplementedError, TypeError),
+            "(cummin is not supported for category dtype|"
+            "category dtype not supported|"
+            "category type does not support cummin operations)",
+        ),
+        "cumprod": (
+            (NotImplementedError, TypeError),
+            "(cumprod is not supported for category dtype|"
+            "category dtype not supported|"
+            "category type does not support cumprod operations)",
+        ),
+        "cumsum": (
+            (NotImplementedError, TypeError),
+            "(cumsum is not supported for category dtype|"
+            "category dtype not supported|"
+            "category type does not support cumsum operations)",
+        ),
+        "diff": (TypeError, "unsupported operand type"),
+        "ffill": (None, ""),
+        "first": (None, ""),
+        "idxmax": (ValueError, "empty group due to unobserved categories")
+        if empty_groups
+        else (None, ""),
+        "idxmin": (ValueError, "empty group due to unobserved categories")
+        if empty_groups
+        else (None, ""),
+        "last": (None, ""),
+        "max": (None, ""),
+        "mean": (TypeError, "category dtype does not support aggregation 'mean'"),
+        "median": (TypeError, "category dtype does not support aggregation 'median'"),
+        "min": (None, ""),
+        "ngroup": (None, ""),
+        "nunique": (None, ""),
+        "pct_change": (TypeError, "unsupported operand type"),
+        "prod": (TypeError, "category type does not support prod operations"),
+        "quantile": (TypeError, ""),
+        "rank": (None, ""),
+        "sem": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'sem'",
+                    "category dtype does not support aggregation 'sem'",
+                ]
+            ),
+        ),
+        "shift": (None, ""),
+        "size": (None, ""),
+        "skew": (
+            TypeError,
+            "|".join(
+                [
+                    "category type does not support skew operations",
+                    "dtype category does not support operation 'skew'",
+                ]
+            ),
+        ),
+        "kurt": (
+            TypeError,
+            "|".join(
+                [
+                    "category type does not support kurt operations",
+                    "dtype category does not support operation 'kurt'",
+                ]
+            ),
+        ),
+        "std": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'std'",
+                    "category dtype does not support aggregation 'std'",
+                ]
+            ),
+        ),
+        "sum": (TypeError, "category type does not support sum operations"),
+        "var": (
+            TypeError,
+            "|".join(
+                [
+                    "'Categorical' .* does not support operation 'var'",
+                    "category dtype does not support aggregation 'var'",
+                ]
+            ),
+        ),
+    }[groupby_func]
+
+    if groupby_func == "corrwith":
+        warn_msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn_msg = ""
+    _call_and_check(klass, msg, how, gb, groupby_func, args, warn_msg)
diff --git a/pandas/pandas/tests/groupby/test_reductions.py b/pandas/pandas/tests/groupby/test_reductions.py
new file mode 100644
index 0000000000000..ea876cfdf4933
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_reductions.py
@@ -0,0 +1,1427 @@
+import builtins
+import datetime as dt
+from string import ascii_lowercase
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+
+from pandas.core.dtypes.common import pandas_dtype
+from pandas.core.dtypes.missing import na_value_for_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+from pandas.util import _test_decorators as td
+
+
+@pytest.mark.parametrize("dtype", ["int64", "int32", "float64", "float32"])
+def test_basic_aggregations(dtype):
+    data = Series(np.arange(9) // 3, index=np.arange(9), dtype=dtype)
+
+    index = np.arange(9)
+    np.random.default_rng(2).shuffle(index)
+    data = data.reindex(index)
+
+    grouped = data.groupby(lambda x: x // 3, group_keys=False)
+
+    for k, v in grouped:
+        assert len(v) == 3
+
+    agged = grouped.aggregate(np.mean)
+    assert agged[1] == 1
+
+    expected = grouped.agg(np.mean)
+    tm.assert_series_equal(agged, expected)  # shorthand
+    tm.assert_series_equal(agged, grouped.mean())
+    result = grouped.sum()
+    expected = grouped.agg(np.sum)
+    if dtype == "int32":
+        # NumPy's sum returns int64
+        expected = expected.astype("int32")
+    tm.assert_series_equal(result, expected)
+
+    expected = grouped.apply(lambda x: x * x.sum())
+    transformed = grouped.transform(lambda x: x * x.sum())
+    assert transformed[7] == 12
+    tm.assert_series_equal(transformed, expected)
+
+    value_grouped = data.groupby(data)
+    result = value_grouped.aggregate(np.mean)
+    tm.assert_series_equal(result, agged, check_index_type=False)
+
+    # complex agg
+    agged = grouped.aggregate([np.mean, np.std])
+
+    msg = r"nested renamer is not supported"
+    with pytest.raises(pd.errors.SpecificationError, match=msg):
+        grouped.aggregate({"one": np.mean, "two": np.std})
+
+    # corner cases
+    msg = "Must produce aggregated value"
+    # exception raised is type Exception
+    with pytest.raises(Exception, match=msg):
+        grouped.aggregate(lambda x: x * 2)
+
+
+@pytest.mark.parametrize(
+    "vals",
+    [
+        ["foo", "bar", "baz"],
+        ["foo", "", ""],
+        ["", "", ""],
+        [1, 2, 3],
+        [1, 0, 0],
+        [0, 0, 0],
+        [1.0, 2.0, 3.0],
+        [1.0, 0.0, 0.0],
+        [0.0, 0.0, 0.0],
+        [True, True, True],
+        [True, False, False],
+        [False, False, False],
+        [np.nan, np.nan, np.nan],
+    ],
+)
+def test_groupby_bool_aggs(skipna, all_boolean_reductions, vals):
+    df = DataFrame({"key": ["a"] * 3 + ["b"] * 3, "val": vals * 2})
+
+    # Figure out expectation using Python builtin
+    exp = getattr(builtins, all_boolean_reductions)(vals)
+
+    # edge case for missing data with skipna and 'any'
+    if skipna and all(isna(vals)) and all_boolean_reductions == "any":
+        exp = False
+
+    expected = DataFrame(
+        [exp] * 2, columns=["val"], index=pd.Index(["a", "b"], name="key")
+    )
+    result = getattr(df.groupby("key"), all_boolean_reductions)(skipna=skipna)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_any():
+    df = DataFrame(
+        [[1, 2, "foo"], [1, np.nan, "bar"], [3, np.nan, "baz"]],
+        columns=["A", "B", "C"],
+    )
+    expected = DataFrame(
+        [[True, True], [False, True]], columns=["B", "C"], index=[1, 3]
+    )
+    expected.index.name = "A"
+    result = df.groupby("A").any()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_bool_aggs_dup_column_labels(all_boolean_reductions):
+    # GH#21668
+    df = DataFrame([[True, True]], columns=["a", "a"])
+    grp_by = df.groupby([0])
+    result = getattr(grp_by, all_boolean_reductions)()
+
+    expected = df.set_axis(np.array([0]))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        [False, False, False],
+        [True, True, True],
+        [pd.NA, pd.NA, pd.NA],
+        [False, pd.NA, False],
+        [True, pd.NA, True],
+        [True, pd.NA, False],
+    ],
+)
+def test_masked_kleene_logic(all_boolean_reductions, skipna, data):
+    # GH#37506
+    ser = Series(data, dtype="boolean")
+
+    # The result should match aggregating on the whole series. Correctness
+    # there is verified in test_reductions.py::test_any_all_boolean_kleene_logic
+    expected_data = getattr(ser, all_boolean_reductions)(skipna=skipna)
+    expected = Series(expected_data, index=np.array([0]), dtype="boolean")
+
+    result = ser.groupby([0, 0, 0]).agg(all_boolean_reductions, skipna=skipna)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype1,dtype2,exp_col1,exp_col2",
+    [
+        (
+            "float",
+            "Float64",
+            np.array([True], dtype=bool),
+            pd.array([pd.NA], dtype="boolean"),
+        ),
+        (
+            "Int64",
+            "float",
+            pd.array([pd.NA], dtype="boolean"),
+            np.array([True], dtype=bool),
+        ),
+        (
+            "Int64",
+            "Int64",
+            pd.array([pd.NA], dtype="boolean"),
+            pd.array([pd.NA], dtype="boolean"),
+        ),
+        (
+            "Float64",
+            "boolean",
+            pd.array([pd.NA], dtype="boolean"),
+            pd.array([pd.NA], dtype="boolean"),
+        ),
+    ],
+)
+def test_masked_mixed_types(dtype1, dtype2, exp_col1, exp_col2):
+    # GH#37506
+    data = [1.0, np.nan]
+    df = DataFrame(
+        {"col1": pd.array(data, dtype=dtype1), "col2": pd.array(data, dtype=dtype2)}
+    )
+    result = df.groupby([1, 1]).agg("all", skipna=False)
+
+    expected = DataFrame({"col1": exp_col1, "col2": exp_col2}, index=np.array([1]))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["Int64", "Float64", "boolean"])
+def test_masked_bool_aggs_skipna(
+    all_boolean_reductions, dtype, skipna, frame_or_series
+):
+    # GH#40585
+    obj = frame_or_series([pd.NA, 1], dtype=dtype)
+    expected_res = True
+    if not skipna and all_boolean_reductions == "all":
+        expected_res = pd.NA
+    expected = frame_or_series([expected_res], index=np.array([1]), dtype="boolean")
+
+    result = obj.groupby([1, 1]).agg(all_boolean_reductions, skipna=skipna)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "bool_agg_func,data,expected_res",
+    [
+        ("any", [pd.NA, np.nan], False),
+        ("any", [pd.NA, 1, np.nan], True),
+        ("all", [pd.NA, pd.NaT], True),
+        ("all", [pd.NA, False, pd.NaT], False),
+    ],
+)
+def test_object_type_missing_vals(bool_agg_func, data, expected_res, frame_or_series):
+    # GH#37501
+    obj = frame_or_series(data, dtype=object)
+    result = obj.groupby([1] * len(data)).agg(bool_agg_func)
+    expected = frame_or_series([expected_res], index=np.array([1]), dtype="bool")
+    tm.assert_equal(result, expected)
+
+
+def test_object_NA_raises_with_skipna_false(all_boolean_reductions):
+    # GH#37501
+    ser = Series([pd.NA], dtype=object)
+    with pytest.raises(TypeError, match="boolean value of NA is ambiguous"):
+        ser.groupby([1]).agg(all_boolean_reductions, skipna=False)
+
+
+def test_empty(frame_or_series, all_boolean_reductions):
+    # GH 45231
+    kwargs = {"columns": ["a"]} if frame_or_series is DataFrame else {"name": "a"}
+    obj = frame_or_series(**kwargs, dtype=object)
+    result = getattr(obj.groupby(obj.index), all_boolean_reductions)()
+    expected = frame_or_series(**kwargs, dtype=bool)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["idxmin", "idxmax"])
+def test_idxmin_idxmax_extremes(how, any_real_numpy_dtype):
+    # GH#57040
+    if any_real_numpy_dtype is int or any_real_numpy_dtype is float:
+        # No need to test
+        return
+    info = np.iinfo if "int" in any_real_numpy_dtype else np.finfo
+    min_value = info(any_real_numpy_dtype).min
+    max_value = info(any_real_numpy_dtype).max
+    df = DataFrame(
+        {"a": [2, 1, 1, 2], "b": [min_value, max_value, max_value, min_value]},
+        dtype=any_real_numpy_dtype,
+    )
+    gb = df.groupby("a")
+    result = getattr(gb, how)()
+    expected = DataFrame(
+        {"b": [1, 0]}, index=pd.Index([1, 2], name="a", dtype=any_real_numpy_dtype)
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["idxmin", "idxmax"])
+def test_idxmin_idxmax_extremes_skipna(skipna, how, float_numpy_dtype):
+    # GH#57040
+    min_value = np.finfo(float_numpy_dtype).min
+    max_value = np.finfo(float_numpy_dtype).max
+    df = DataFrame(
+        {
+            "a": Series(np.repeat(range(1, 6), repeats=2), dtype="intp"),
+            "b": Series(
+                [
+                    np.nan,
+                    min_value,
+                    np.nan,
+                    max_value,
+                    min_value,
+                    np.nan,
+                    max_value,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                ],
+                dtype=float_numpy_dtype,
+            ),
+        },
+    )
+    gb = df.groupby("a")
+
+    if not skipna:
+        msg = f"DataFrameGroupBy.{how} with skipna=False"
+        with pytest.raises(ValueError, match=msg):
+            getattr(gb, how)(skipna=skipna)
+        return
+    result = getattr(gb, how)(skipna=skipna)
+    expected = DataFrame(
+        {"b": [1, 3, 4, 6, np.nan]}, index=pd.Index(range(1, 6), name="a", dtype="intp")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, values",
+    [
+        ("idxmin", {"c_int": [0, 2], "c_float": [1, 3], "c_date": [1, 2]}),
+        ("idxmax", {"c_int": [1, 3], "c_float": [0, 2], "c_date": [0, 3]}),
+    ],
+)
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_idxmin_idxmax_returns_int_types(func, values, numeric_only):
+    # GH 25444
+    df = DataFrame(
+        {
+            "name": ["A", "A", "B", "B"],
+            "c_int": [1, 2, 3, 4],
+            "c_float": [4.02, 3.03, 2.04, 1.05],
+            "c_date": ["2019", "2018", "2016", "2017"],
+        }
+    )
+    df["c_date"] = pd.to_datetime(df["c_date"])
+    df["c_date_tz"] = df["c_date"].dt.tz_localize("US/Pacific")
+    df["c_timedelta"] = df["c_date"] - df["c_date"].iloc[0]
+    df["c_period"] = df["c_date"].dt.to_period("W")
+    df["c_Integer"] = df["c_int"].astype("Int64")
+    df["c_Floating"] = df["c_float"].astype("Float64")
+
+    result = getattr(df.groupby("name"), func)(numeric_only=numeric_only)
+
+    expected = DataFrame(values, index=pd.Index(["A", "B"], name="name"))
+    if numeric_only:
+        expected = expected.drop(columns=["c_date"])
+    else:
+        expected["c_date_tz"] = expected["c_date"]
+        expected["c_timedelta"] = expected["c_date"]
+        expected["c_period"] = expected["c_date"]
+    expected["c_Integer"] = expected["c_int"]
+    expected["c_Floating"] = expected["c_float"]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        (
+            Timestamp("2011-01-15 12:50:28.502376"),
+            Timestamp("2011-01-20 12:50:28.593448"),
+        ),
+        (24650000000000001, 24650000000000002),
+    ],
+)
+@pytest.mark.parametrize("method", ["count", "min", "max", "first", "last"])
+def test_groupby_non_arithmetic_agg_int_like_precision(method, data):
+    # GH#6620, GH#9311
+    df = DataFrame({"a": [1, 1], "b": data})
+
+    grouped = df.groupby("a")
+    result = getattr(grouped, method)()
+    if method == "count":
+        expected_value = 2
+    elif method == "first":
+        expected_value = data[0]
+    elif method == "last":
+        expected_value = data[1]
+    else:
+        expected_value = getattr(df["b"], method)()
+    expected = DataFrame({"b": [expected_value]}, index=pd.Index([1], name="a"))
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["first", "last"])
+def test_first_last_skipna(any_real_nullable_dtype, sort, skipna, how):
+    # GH#57019
+    na_value = na_value_for_dtype(pandas_dtype(any_real_nullable_dtype))
+    df = DataFrame(
+        {
+            "a": [2, 1, 1, 2, 3, 3],
+            "b": [na_value, 3.0, na_value, 4.0, np.nan, np.nan],
+            "c": [na_value, 3.0, na_value, 4.0, np.nan, np.nan],
+        },
+        dtype=any_real_nullable_dtype,
+    )
+    gb = df.groupby("a", sort=sort)
+    method = getattr(gb, how)
+    result = method(skipna=skipna)
+
+    ilocs = {
+        ("first", True): [3, 1, 4],
+        ("first", False): [0, 1, 4],
+        ("last", True): [3, 1, 5],
+        ("last", False): [3, 2, 5],
+    }[how, skipna]
+    expected = df.iloc[ilocs].set_index("a")
+    if sort:
+        expected = expected.sort_index()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_mean_no_overflow():
+    # Regression test for (#22487)
+    df = DataFrame(
+        {
+            "user": ["A", "A", "A", "A", "A"],
+            "connections": [4970, 4749, 4719, 4704, 18446744073699999744],
+        }
+    )
+    assert df.groupby("user")["connections"].mean()["A"] == 3689348814740003840
+
+
+def test_mean_on_timedelta():
+    # GH 17382
+    df = DataFrame({"time": pd.to_timedelta(range(10)), "cat": ["A", "B"] * 5})
+    result = df.groupby("cat")["time"].mean()
+    expected = Series(
+        pd.to_timedelta([4, 5]), name="time", index=pd.Index(["A", "B"], name="cat")
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values, dtype, result_dtype",
+    [
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "float64", "float64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "Float64", "Float64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "Int64", "Float64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "timedelta64[ns]", "timedelta64[ns]"),
+        (
+            pd.to_datetime(
+                [
+                    "2019-05-09",
+                    pd.NaT,
+                    "2019-05-11",
+                    "2019-05-12",
+                    "2019-05-13",
+                    "2019-05-14",
+                    "2019-05-15",
+                    "2019-05-16",
+                    "2019-05-17",
+                    "2019-05-18",
+                ]
+            ),
+            "datetime64[ns]",
+            "datetime64[ns]",
+        ),
+    ],
+)
+def test_mean_skipna(values, dtype, result_dtype, skipna):
+    # GH#15675
+    df = DataFrame(
+        {
+            "val": values,
+            "cat": ["A", "B"] * 5,
+        }
+    ).astype({"val": dtype})
+    # We need to recast the expected values to the result_dtype because
+    # Series.mean() changes the dtype to float64/object depending on the input dtype
+    expected = (
+        df.groupby("cat")["val"]
+        .apply(lambda x: x.mean(skipna=skipna))
+        .astype(result_dtype)
+    )
+    result = df.groupby("cat")["val"].mean(skipna=skipna)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values, dtype",
+    [
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "float64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "Float64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "Int64"),
+        ([0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9], "timedelta64[ns]"),
+    ],
+)
+def test_sum_skipna(values, dtype, skipna):
+    # GH#15675
+    df = DataFrame(
+        {
+            "val": values,
+            "cat": ["A", "B"] * 5,
+        }
+    ).astype({"val": dtype})
+    # We need to recast the expected values to the original dtype because
+    # Series.sum() changes the dtype
+    expected = (
+        df.groupby("cat")["val"].apply(lambda x: x.sum(skipna=skipna)).astype(dtype)
+    )
+    result = df.groupby("cat")["val"].sum(skipna=skipna)
+    tm.assert_series_equal(result, expected)
+
+
+def test_sum_skipna_object(skipna):
+    # GH#15675
+    df = DataFrame(
+        {
+            "val": ["a", "b", np.nan, "d", "e", "f", "g", "h", "i", "j"],
+            "cat": ["A", "B"] * 5,
+        }
+    ).astype({"val": object})
+    if skipna:
+        expected = Series(
+            ["aegi", "bdfhj"], index=pd.Index(["A", "B"], name="cat"), name="val"
+        ).astype(object)
+    else:
+        expected = Series(
+            [np.nan, "bdfhj"], index=pd.Index(["A", "B"], name="cat"), name="val"
+        ).astype(object)
+    result = df.groupby("cat")["val"].sum(skipna=skipna)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, values, dtype, result_dtype",
+    [
+        ("prod", [0, 1, 3, np.nan, 4, 5, 6, 7, -8, 9], "float64", "float64"),
+        ("prod", [0, -1, 3, 4, 5, np.nan, 6, 7, 8, 9], "Float64", "Float64"),
+        ("prod", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "Int64", "Int64"),
+        ("prod", [np.nan] * 10, "float64", "float64"),
+        ("prod", [np.nan] * 10, "Float64", "Float64"),
+        ("prod", [np.nan] * 10, "Int64", "Int64"),
+        ("var", [0, -1, 3, 4, np.nan, 5, 6, 7, 8, 9], "float64", "float64"),
+        ("var", [0, 1, 3, -4, 5, 6, 7, -8, 9, np.nan], "Float64", "Float64"),
+        ("var", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "Int64", "Float64"),
+        ("var", [np.nan] * 10, "float64", "float64"),
+        ("var", [np.nan] * 10, "Float64", "Float64"),
+        ("var", [np.nan] * 10, "Int64", "Float64"),
+        ("std", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "float64", "float64"),
+        ("std", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "Float64", "Float64"),
+        ("std", [0, 1, 3, -4, 5, 6, 7, -8, 9, np.nan], "Int64", "Float64"),
+        ("std", [np.nan] * 10, "float64", "float64"),
+        ("std", [np.nan] * 10, "Float64", "Float64"),
+        ("std", [np.nan] * 10, "Int64", "Float64"),
+        ("sem", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "float64", "float64"),
+        ("sem", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "Float64", "Float64"),
+        ("sem", [0, -1, 3, 4, 5, -6, 7, 8, 9, np.nan], "Int64", "Float64"),
+        ("sem", [np.nan] * 10, "float64", "float64"),
+        ("sem", [np.nan] * 10, "Float64", "Float64"),
+        ("sem", [np.nan] * 10, "Int64", "Float64"),
+        ("min", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "float64", "float64"),
+        ("min", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "Float64", "Float64"),
+        ("min", [0, -1, 3, 4, 5, -6, 7, 8, 9, np.nan], "Int64", "Int64"),
+        (
+            "min",
+            [0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9],
+            "timedelta64[ns]",
+            "timedelta64[ns]",
+        ),
+        (
+            "min",
+            pd.to_datetime(
+                [
+                    "2019-05-09",
+                    pd.NaT,
+                    "2019-05-11",
+                    "2019-05-12",
+                    "2019-05-13",
+                    "2019-05-14",
+                    "2019-05-15",
+                    "2019-05-16",
+                    "2019-05-17",
+                    "2019-05-18",
+                ]
+            ),
+            "datetime64[ns]",
+            "datetime64[ns]",
+        ),
+        ("min", [np.nan] * 10, "float64", "float64"),
+        ("min", [np.nan] * 10, "Float64", "Float64"),
+        ("min", [np.nan] * 10, "Int64", "Int64"),
+        ("max", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "float64", "float64"),
+        ("max", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "Float64", "Float64"),
+        ("max", [0, -1, 3, 4, 5, -6, 7, 8, 9, np.nan], "Int64", "Int64"),
+        (
+            "max",
+            [0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9],
+            "timedelta64[ns]",
+            "timedelta64[ns]",
+        ),
+        (
+            "max",
+            pd.to_datetime(
+                [
+                    "2019-05-09",
+                    pd.NaT,
+                    "2019-05-11",
+                    "2019-05-12",
+                    "2019-05-13",
+                    "2019-05-14",
+                    "2019-05-15",
+                    "2019-05-16",
+                    "2019-05-17",
+                    "2019-05-18",
+                ]
+            ),
+            "datetime64[ns]",
+            "datetime64[ns]",
+        ),
+        ("max", [np.nan] * 10, "float64", "float64"),
+        ("max", [np.nan] * 10, "Float64", "Float64"),
+        ("max", [np.nan] * 10, "Int64", "Int64"),
+        ("median", [0, -1, 3, 4, 5, -6, 7, np.nan, 8, 9], "float64", "float64"),
+        ("median", [0, 1, 3, -4, 5, 6, 7, -8, np.nan, 9], "Float64", "Float64"),
+        ("median", [0, -1, 3, 4, 5, -6, 7, 8, 9, np.nan], "Int64", "Float64"),
+        (
+            "median",
+            [0, 1, np.nan, 3, 4, 5, 6, 7, 8, 9],
+            "timedelta64[ns]",
+            "timedelta64[ns]",
+        ),
+        (
+            "median",
+            pd.to_datetime(
+                [
+                    "2019-05-09",
+                    pd.NaT,
+                    "2019-05-11",
+                    "2019-05-12",
+                    "2019-05-13",
+                    "2019-05-14",
+                    "2019-05-15",
+                    "2019-05-16",
+                    "2019-05-17",
+                    "2019-05-18",
+                ]
+            ),
+            "datetime64[ns]",
+            "datetime64[ns]",
+        ),
+        ("median", [np.nan] * 10, "float64", "float64"),
+        ("median", [np.nan] * 10, "Float64", "Float64"),
+        ("median", [np.nan] * 10, "Int64", "Float64"),
+    ],
+)
+def test_multifunc_skipna(func, values, dtype, result_dtype, skipna):
+    # GH#15675
+    df = DataFrame(
+        {
+            "val": values,
+            "cat": ["A", "B"] * 5,
+        }
+    ).astype({"val": dtype})
+    # We need to recast the expected values to the result_dtype as some operations
+    # change the dtype
+    expected = (
+        df.groupby("cat")["val"]
+        .apply(lambda x: getattr(x, func)(skipna=skipna))
+        .astype(result_dtype)
+    )
+    result = getattr(df.groupby("cat")["val"], func)(skipna=skipna)
+    tm.assert_series_equal(result, expected)
+
+
+def test_cython_median():
+    arr = np.random.default_rng(2).standard_normal(1000)
+    arr[::2] = np.nan
+    df = DataFrame(arr)
+
+    labels = np.random.default_rng(2).integers(0, 50, size=1000).astype(float)
+    labels[::17] = np.nan
+
+    result = df.groupby(labels).median()
+    exp = df.groupby(labels).agg(np.nanmedian)
+    tm.assert_frame_equal(result, exp)
+
+    df = DataFrame(np.random.default_rng(2).standard_normal((1000, 5)))
+    rs = df.groupby(labels).agg(np.median)
+    xp = df.groupby(labels).median()
+    tm.assert_frame_equal(rs, xp)
+
+
+def test_median_empty_bins(observed):
+    df = DataFrame(np.random.default_rng(2).integers(0, 44, 500))
+
+    grps = range(0, 55, 5)
+    bins = pd.cut(df[0], grps)
+
+    result = df.groupby(bins, observed=observed).median()
+    expected = df.groupby(bins, observed=observed).agg(lambda x: x.median())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_max_min_non_numeric():
+    # #2700
+    aa = DataFrame({"nn": [11, 11, 22, 22], "ii": [1, 2, 3, 4], "ss": 4 * ["mama"]})
+
+    result = aa.groupby("nn").max()
+    assert "ss" in result
+
+    result = aa.groupby("nn").max(numeric_only=False)
+    assert "ss" in result
+
+    result = aa.groupby("nn").min()
+    assert "ss" in result
+
+    result = aa.groupby("nn").min(numeric_only=False)
+    assert "ss" in result
+
+
+def test_max_min_object_multiple_columns(using_infer_string):
+    # GH#41111 case where the aggregation is valid for some columns but not
+    # others; we split object blocks column-wise, consistent with
+    # DataFrame._reduce
+
+    df = DataFrame(
+        {
+            "A": [1, 1, 2, 2, 3],
+            "B": [1, "foo", 2, "bar", False],
+            "C": ["a", "b", "c", "d", "e"],
+        }
+    )
+    df._consolidate_inplace()  # should already be consolidate, but double-check
+    assert len(df._mgr.blocks) == 3 if using_infer_string else 2
+
+    gb = df.groupby("A")
+
+    result = gb[["C"]].max()
+    # "max" is valid for column "C" but not for "B"
+    ei = pd.Index([1, 2, 3], name="A")
+    expected = DataFrame({"C": ["b", "d", "e"]}, index=ei)
+    tm.assert_frame_equal(result, expected)
+
+    result = gb[["C"]].min()
+    # "min" is valid for column "C" but not for "B"
+    ei = pd.Index([1, 2, 3], name="A")
+    expected = DataFrame({"C": ["a", "c", "e"]}, index=ei)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_min_date_with_nans():
+    # GH26321
+    dates = pd.to_datetime(
+        Series(["2019-05-09", "2019-05-09", "2019-05-09"]), format="%Y-%m-%d"
+    ).dt.date
+    df = DataFrame({"a": [np.nan, "1", np.nan], "b": [0, 1, 1], "c": dates})
+
+    result = df.groupby("b", as_index=False)["c"].min()["c"]
+    expected = pd.to_datetime(
+        Series(["2019-05-09", "2019-05-09"], name="c"), format="%Y-%m-%d"
+    ).dt.date
+    tm.assert_series_equal(result, expected)
+
+    result = df.groupby("b")["c"].min()
+    expected.index.name = "b"
+    tm.assert_series_equal(result, expected)
+
+
+def test_max_inat():
+    # GH#40767 dont interpret iNaT as NaN
+    ser = Series([1, iNaT])
+    key = np.array([1, 1], dtype=np.int64)
+    gb = ser.groupby(key)
+
+    result = gb.max(min_count=2)
+    expected = Series({1: 1}, dtype=np.int64)
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+    result = gb.min(min_count=2)
+    expected = Series({1: iNaT}, dtype=np.int64)
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+    # not enough entries -> gets masked to NaN
+    result = gb.min(min_count=3)
+    expected = Series({1: np.nan})
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+
+def test_max_inat_not_all_na():
+    # GH#40767 dont interpret iNaT as NaN
+
+    # make sure we dont round iNaT+1 to iNaT
+    ser = Series([1, iNaT, 2, iNaT + 1])
+    gb = ser.groupby([1, 2, 3, 3])
+    result = gb.min(min_count=2)
+
+    # Note: in converting to float64, the iNaT + 1 maps to iNaT, i.e. is lossy
+    expected = Series({1: np.nan, 2: np.nan, 3: iNaT + 1})
+    expected.index = expected.index.astype(int)
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_groupby_aggregate_period_column(func):
+    # GH 31471
+    groups = [1, 2]
+    periods = pd.period_range("2020", periods=2, freq="Y")
+    df = DataFrame({"a": groups, "b": periods})
+
+    result = getattr(df.groupby("a")["b"], func)()
+    idx = pd.Index([1, 2], name="a")
+    expected = Series(periods, index=idx, name="b")
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_groupby_aggregate_period_frame(func):
+    # GH 31471
+    groups = [1, 2]
+    periods = pd.period_range("2020", periods=2, freq="Y")
+    df = DataFrame({"a": groups, "b": periods})
+
+    result = getattr(df.groupby("a"), func)()
+    idx = pd.Index([1, 2], name="a")
+    expected = DataFrame({"b": periods}, index=idx)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_aggregate_numeric_object_dtype():
+    # https://github.com/pandas-dev/pandas/issues/39329
+    # simplified case: multiple object columns where one is all-NaN
+    # -> gets split as the all-NaN is inferred as float
+    df = DataFrame(
+        {"key": ["A", "A", "B", "B"], "col1": list("abcd"), "col2": [np.nan] * 4},
+    ).astype(object)
+    result = df.groupby("key").min()
+    expected = (
+        DataFrame(
+            {"key": ["A", "B"], "col1": ["a", "c"], "col2": [np.nan, np.nan]},
+        )
+        .set_index("key")
+        .astype(object)
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # same but with numbers
+    df = DataFrame(
+        {"key": ["A", "A", "B", "B"], "col1": list("abcd"), "col2": range(4)},
+    ).astype(object)
+    result = df.groupby("key").min()
+    expected = (
+        DataFrame({"key": ["A", "B"], "col1": ["a", "c"], "col2": [0, 2]})
+        .set_index("key")
+        .astype(object)
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_aggregate_categorical_lost_index(func: str):
+    # GH: 28641 groupby drops index, when grouping over categorical column with min/max
+    ds = Series(["b"], dtype="category").cat.as_ordered()
+    df = DataFrame({"A": [1997], "B": ds})
+    result = df.groupby("A").agg({"B": func})
+    expected = DataFrame({"B": ["b"]}, index=pd.Index([1997], name="A"))
+
+    # ordered categorical dtype should be preserved
+    expected["B"] = expected["B"].astype(ds.dtype)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["Int64", "Int32", "Float64", "Float32", "boolean"])
+def test_groupby_min_max_nullable(dtype):
+    if dtype == "Int64":
+        # GH#41743 avoid precision loss
+        ts = 1618556707013635762
+    elif dtype == "boolean":
+        ts = 0
+    else:
+        ts = 4.0
+
+    df = DataFrame({"id": [2, 2], "ts": [ts, ts + 1]})
+    df["ts"] = df["ts"].astype(dtype)
+
+    gb = df.groupby("id")
+
+    result = gb.min()
+    expected = df.iloc[:1].set_index("id")
+    tm.assert_frame_equal(result, expected)
+
+    res_max = gb.max()
+    expected_max = df.iloc[1:].set_index("id")
+    tm.assert_frame_equal(res_max, expected_max)
+
+    result2 = gb.min(min_count=3)
+    expected2 = DataFrame({"ts": [pd.NA]}, index=expected.index, dtype=dtype)
+    tm.assert_frame_equal(result2, expected2)
+
+    res_max2 = gb.max(min_count=3)
+    tm.assert_frame_equal(res_max2, expected2)
+
+    # Case with NA values
+    df2 = DataFrame({"id": [2, 2, 2], "ts": [ts, pd.NA, ts + 1]})
+    df2["ts"] = df2["ts"].astype(dtype)
+    gb2 = df2.groupby("id")
+
+    result3 = gb2.min()
+    tm.assert_frame_equal(result3, expected)
+
+    res_max3 = gb2.max()
+    tm.assert_frame_equal(res_max3, expected_max)
+
+    result4 = gb2.min(min_count=100)
+    tm.assert_frame_equal(result4, expected2)
+
+    res_max4 = gb2.max(min_count=100)
+    tm.assert_frame_equal(res_max4, expected2)
+
+
+def test_min_max_nullable_uint64_empty_group():
+    # don't raise NotImplementedError from libgroupby
+    cat = pd.Categorical([0] * 10, categories=[0, 1])
+    df = DataFrame({"A": cat, "B": pd.array(np.arange(10, dtype=np.uint64))})
+    gb = df.groupby("A", observed=False)
+
+    res = gb.min()
+
+    idx = pd.CategoricalIndex([0, 1], dtype=cat.dtype, name="A")
+    expected = DataFrame({"B": pd.array([0, pd.NA], dtype="UInt64")}, index=idx)
+    tm.assert_frame_equal(res, expected)
+
+    res = gb.max()
+    expected.iloc[0, 0] = 9
+    tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize("func", ["first", "last", "min", "max"])
+def test_groupby_min_max_categorical(func):
+    # GH: 52151
+    df = DataFrame(
+        {
+            "col1": pd.Categorical(["A"], categories=list("AB"), ordered=True),
+            "col2": pd.Categorical([1], categories=[1, 2], ordered=True),
+            "value": 0.1,
+        }
+    )
+    result = getattr(df.groupby("col1", observed=False), func)()
+
+    idx = pd.CategoricalIndex(data=["A", "B"], name="col1", ordered=True)
+    expected = DataFrame(
+        {
+            "col2": pd.Categorical([1, None], categories=[1, 2], ordered=True),
+            "value": [0.1, None],
+        },
+        index=idx,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_min_empty_string_dtype(func, string_dtype_no_object):
+    # GH#55619
+    dtype = string_dtype_no_object
+    df = DataFrame({"a": ["a"], "b": "a", "c": "a"}, dtype=dtype).iloc[:0]
+    result = getattr(df.groupby("a"), func)()
+    expected = DataFrame(
+        columns=["b", "c"], dtype=dtype, index=pd.Index([], dtype=dtype, name="a")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_max_nan_bug():
+    df = DataFrame(
+        {
+            "Unnamed: 0": ["-04-23", "-05-06", "-05-07"],
+            "Date": [
+                "2013-04-23 00:00:00",
+                "2013-05-06 00:00:00",
+                "2013-05-07 00:00:00",
+            ],
+            "app": Series([np.nan, np.nan, "OE"]),
+            "File": ["log080001.log", "log.log", "xlsx"],
+        }
+    )
+    gb = df.groupby("Date")
+    r = gb[["File"]].max()
+    e = gb["File"].max().to_frame()
+    tm.assert_frame_equal(r, e)
+    assert not r["File"].isna().any()
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize("with_nan", [True, False])
+@pytest.mark.parametrize("keys", [["joe"], ["joe", "jim"]])
+def test_series_groupby_nunique(sort, dropna, as_index, with_nan, keys):
+    n = 100
+    m = 10
+    days = date_range("2015-08-23", periods=10)
+    df = DataFrame(
+        {
+            "jim": np.random.default_rng(2).choice(list(ascii_lowercase), n),
+            "joe": np.random.default_rng(2).choice(days, n),
+            "julie": np.random.default_rng(2).integers(0, m, n),
+        }
+    )
+    if with_nan:
+        df = df.astype({"julie": float})  # Explicit cast to avoid implicit cast below
+        df.loc[1::17, "jim"] = None
+        df.loc[3::37, "joe"] = None
+        df.loc[7::19, "julie"] = None
+        df.loc[8::19, "julie"] = None
+        df.loc[9::19, "julie"] = None
+    original_df = df.copy()
+    gr = df.groupby(keys, as_index=as_index, sort=sort)
+    left = gr["julie"].nunique(dropna=dropna)
+
+    gr = df.groupby(keys, as_index=as_index, sort=sort)
+    right = gr["julie"].apply(Series.nunique, dropna=dropna)
+    if not as_index:
+        right = right.reset_index(drop=True)
+
+    if as_index:
+        tm.assert_series_equal(left, right, check_names=False)
+    else:
+        tm.assert_frame_equal(left, right, check_names=False)
+    tm.assert_frame_equal(df, original_df)
+
+
+def test_nunique():
+    df = DataFrame({"A": list("abbacc"), "B": list("abxacc"), "C": list("abbacx")})
+
+    expected = DataFrame({"A": list("abc"), "B": [1, 2, 1], "C": [1, 1, 2]})
+    result = df.groupby("A", as_index=False).nunique()
+    tm.assert_frame_equal(result, expected)
+
+    # as_index
+    expected.index = list("abc")
+    expected.index.name = "A"
+    expected = expected.drop(columns="A")
+    result = df.groupby("A").nunique()
+    tm.assert_frame_equal(result, expected)
+
+    # with na
+    result = df.replace({"x": None}).groupby("A").nunique(dropna=False)
+    tm.assert_frame_equal(result, expected)
+
+    # dropna
+    expected = DataFrame({"B": [1] * 3, "C": [1] * 3}, index=list("abc"))
+    expected.index.name = "A"
+    result = df.replace({"x": None}).groupby("A").nunique()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nunique_with_object():
+    # GH 11077
+    data = DataFrame(
+        [
+            [100, 1, "Alice"],
+            [200, 2, "Bob"],
+            [300, 3, "Charlie"],
+            [-400, 4, "Dan"],
+            [500, 5, "Edith"],
+        ],
+        columns=["amount", "id", "name"],
+    )
+
+    result = data.groupby(["id", "amount"])["name"].nunique()
+    index = MultiIndex.from_arrays([data.id, data.amount])
+    expected = Series([1] * 5, name="name", index=index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_nunique_with_empty_series():
+    # GH 12553
+    data = Series(name="name", dtype=object)
+    result = data.groupby(level=0).nunique()
+    expected = Series(name="name", dtype="int64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_nunique_with_timegrouper():
+    # GH 13453
+    test = DataFrame(
+        {
+            "time": [
+                Timestamp("2016-06-28 09:35:35"),
+                Timestamp("2016-06-28 16:09:30"),
+                Timestamp("2016-06-28 16:46:28"),
+            ],
+            "data": ["1", "2", "3"],
+        }
+    ).set_index("time")
+    result = test.groupby(pd.Grouper(freq="h"))["data"].nunique()
+    expected = test.groupby(pd.Grouper(freq="h"))["data"].apply(Series.nunique)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "key, data, dropna, expected",
+    [
+        (
+            ["x", "x", "x"],
+            [Timestamp("2019-01-01"), pd.NaT, Timestamp("2019-01-01")],
+            True,
+            Series([1], index=pd.Index(["x"], name="key"), name="data"),
+        ),
+        (
+            ["x", "x", "x"],
+            [dt.date(2019, 1, 1), pd.NaT, dt.date(2019, 1, 1)],
+            True,
+            Series([1], index=pd.Index(["x"], name="key"), name="data"),
+        ),
+        (
+            ["x", "x", "x", "y", "y"],
+            [
+                dt.date(2019, 1, 1),
+                pd.NaT,
+                dt.date(2019, 1, 1),
+                pd.NaT,
+                dt.date(2019, 1, 1),
+            ],
+            False,
+            Series([2, 2], index=pd.Index(["x", "y"], name="key"), name="data"),
+        ),
+        (
+            ["x", "x", "x", "x", "y"],
+            [
+                dt.date(2019, 1, 1),
+                pd.NaT,
+                dt.date(2019, 1, 1),
+                pd.NaT,
+                dt.date(2019, 1, 1),
+            ],
+            False,
+            Series([2, 1], index=pd.Index(["x", "y"], name="key"), name="data"),
+        ),
+    ],
+)
+def test_nunique_with_NaT(key, data, dropna, expected):
+    # GH 27951
+    df = DataFrame({"key": key, "data": data})
+    result = df.groupby(["key"])["data"].nunique(dropna=dropna)
+    tm.assert_series_equal(result, expected)
+
+
+def test_nunique_preserves_column_level_names():
+    # GH 23222
+    test = DataFrame([1, 2, 2], columns=pd.Index(["A"], name="level_0"))
+    result = test.groupby([0, 0, 0]).nunique()
+    expected = DataFrame([2], index=np.array([0]), columns=test.columns)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nunique_transform_with_datetime():
+    # GH 35109 - transform with nunique on datetimes results in integers
+    df = DataFrame(date_range("2008-12-31", "2009-01-02"), columns=["date"])
+    result = df.groupby([0, 0, 1])["date"].transform("nunique")
+    expected = Series([2, 2, 1], name="date")
+    tm.assert_series_equal(result, expected)
+
+
+def test_empty_categorical(observed):
+    # GH#21334
+    cat = Series([1]).astype("category")
+    ser = cat[:0]
+    gb = ser.groupby(ser, observed=observed)
+    result = gb.nunique()
+    if observed:
+        expected = Series([], index=cat[:0], dtype="int64")
+    else:
+        expected = Series([0], index=cat, dtype="int64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_intercept_builtin_sum():
+    s = Series([1.0, 2.0, np.nan, 3.0])
+    grouped = s.groupby([0, 1, 2, 2])
+
+    # GH#53425
+    result = grouped.agg(builtins.sum)
+    # GH#53425
+    result2 = grouped.apply(builtins.sum)
+    expected = Series([1.0, 2.0, np.nan], index=np.array([0, 1, 2]))
+    tm.assert_series_equal(result, expected)
+    tm.assert_series_equal(result2, expected)
+
+
+@pytest.mark.parametrize("min_count", [0, 10])
+def test_groupby_sum_mincount_boolean(min_count):
+    b = True
+    a = False
+    na = np.nan
+    dfg = pd.array([b, b, na, na, a, a, b], dtype="boolean")
+
+    df = DataFrame({"A": [1, 1, 2, 2, 3, 3, 1], "B": dfg})
+    result = df.groupby("A").sum(min_count=min_count)
+    if min_count == 0:
+        expected = DataFrame(
+            {"B": pd.array([3, 0, 0], dtype="Int64")},
+            index=pd.Index([1, 2, 3], name="A"),
+        )
+        tm.assert_frame_equal(result, expected)
+    else:
+        expected = DataFrame(
+            {"B": pd.array([pd.NA] * 3, dtype="Int64")},
+            index=pd.Index([1, 2, 3], name="A"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_sum_below_mincount_nullable_integer():
+    # https://github.com/pandas-dev/pandas/issues/32861
+    df = DataFrame({"a": [0, 1, 2], "b": [0, 1, 2], "c": [0, 1, 2]}, dtype="Int64")
+    grouped = df.groupby("a")
+    idx = pd.Index([0, 1, 2], name="a", dtype="Int64")
+
+    result = grouped["b"].sum(min_count=2)
+    expected = Series([pd.NA] * 3, dtype="Int64", index=idx, name="b")
+    tm.assert_series_equal(result, expected)
+
+    result = grouped.sum(min_count=2)
+    expected = DataFrame({"b": [pd.NA] * 3, "c": [pd.NA] * 3}, dtype="Int64", index=idx)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_sum_timedelta_with_nat():
+    # GH#42659
+    df = DataFrame(
+        {
+            "a": [1, 1, 2, 2],
+            "b": [pd.Timedelta("1D"), pd.Timedelta("2D"), pd.Timedelta("3D"), pd.NaT],
+        }
+    )
+    td3 = pd.Timedelta(days=3)
+
+    gb = df.groupby("a")
+
+    res = gb.sum()
+    expected = DataFrame({"b": [td3, td3]}, index=pd.Index([1, 2], name="a"))
+    tm.assert_frame_equal(res, expected)
+
+    res = gb["b"].sum()
+    tm.assert_series_equal(res, expected["b"])
+
+    res = gb["b"].sum(min_count=2)
+    expected = Series([td3, pd.NaT], dtype="m8[ns]", name="b", index=expected.index)
+    tm.assert_series_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["int8", "int16", "int32", "int64", "float32", "float64", "uint64"]
+)
+@pytest.mark.parametrize(
+    "method,data",
+    [
+        ("first", {"df": [{"a": 1, "b": 1}, {"a": 2, "b": 3}]}),
+        ("last", {"df": [{"a": 1, "b": 2}, {"a": 2, "b": 4}]}),
+        ("min", {"df": [{"a": 1, "b": 1}, {"a": 2, "b": 3}]}),
+        ("max", {"df": [{"a": 1, "b": 2}, {"a": 2, "b": 4}]}),
+        ("count", {"df": [{"a": 1, "b": 2}, {"a": 2, "b": 2}], "out_type": "int64"}),
+    ],
+)
+def test_groupby_non_arithmetic_agg_types(dtype, method, data):
+    # GH9311, GH6620
+    df = DataFrame(
+        [{"a": 1, "b": 1}, {"a": 1, "b": 2}, {"a": 2, "b": 3}, {"a": 2, "b": 4}]
+    )
+
+    df["b"] = df.b.astype(dtype)
+
+    if "args" not in data:
+        data["args"] = []
+
+    if "out_type" in data:
+        out_type = data["out_type"]
+    else:
+        out_type = dtype
+
+    exp = data["df"]
+    df_out = DataFrame(exp)
+
+    df_out["b"] = df_out.b.astype(out_type)
+    df_out.set_index("a", inplace=True)
+
+    grpd = df.groupby("a")
+    t = getattr(grpd, method)(*data["args"])
+    tm.assert_frame_equal(t, df_out)
+
+
+def scipy_sem(*args, **kwargs):
+    from scipy.stats import sem
+
+    return sem(*args, ddof=1, **kwargs)
+
+
+@pytest.mark.parametrize(
+    "op,targop",
+    [
+        ("mean", np.mean),
+        ("median", np.median),
+        ("std", np.std),
+        ("var", np.var),
+        ("sum", np.sum),
+        ("prod", np.prod),
+        ("min", np.min),
+        ("max", np.max),
+        ("first", lambda x: x.iloc[0]),
+        ("last", lambda x: x.iloc[-1]),
+        ("count", np.size),
+        pytest.param("sem", scipy_sem, marks=td.skip_if_no("scipy")),
+    ],
+)
+def test_ops_general(op, targop):
+    df = DataFrame(np.random.default_rng(2).standard_normal(1000))
+    labels = np.random.default_rng(2).integers(0, 50, size=1000).astype(float)
+
+    result = getattr(df.groupby(labels), op)()
+    kwargs = {"ddof": 1, "axis": 0} if op in ["std", "var"] else {}
+    expected = df.groupby(labels).agg(targop, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values",
+    [
+        {
+            "a": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+            "b": [1, pd.NA, 2, 1, pd.NA, 2, 1, pd.NA, 2],
+        },
+        {"a": [1, 1, 2, 2, 3, 3], "b": [1, 2, 1, 2, 1, 2]},
+    ],
+)
+@pytest.mark.parametrize("function", ["mean", "median", "var"])
+def test_apply_to_nullable_integer_returns_float(values, function):
+    # https://github.com/pandas-dev/pandas/issues/32219
+    output = 0.5 if function == "var" else 1.5
+    arr = np.array([output] * 3, dtype=float)
+    idx = pd.Index([1, 2, 3], name="a", dtype="Int64")
+    expected = DataFrame({"b": arr}, index=idx).astype("Float64")
+
+    groups = DataFrame(values, dtype="Int64").groupby("a")
+
+    result = getattr(groups, function)()
+    tm.assert_frame_equal(result, expected)
+
+    result = groups.agg(function)
+    tm.assert_frame_equal(result, expected)
+
+    result = groups.agg([function])
+    expected.columns = MultiIndex.from_tuples([("b", function)])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op",
+    [
+        "sum",
+        "prod",
+        "min",
+        "max",
+        "median",
+        "mean",
+        "skew",
+        "kurt",
+        "std",
+        "var",
+        "sem",
+    ],
+)
+def test_regression_allowlist_methods(op, skipna, sort):
+    # GH6944
+    # GH 17537
+    # explicitly test the allowlist methods
+    frame = DataFrame([0])
+
+    grouped = frame.groupby(level=0, sort=sort)
+
+    if op in ["skew", "kurt", "sum", "mean"]:
+        # skew, kurt, sum, mean have skipna
+        result = getattr(grouped, op)(skipna=skipna)
+        expected = frame.groupby(level=0).apply(lambda h: getattr(h, op)(skipna=skipna))
+        if sort:
+            expected = expected.sort_index()
+        tm.assert_frame_equal(result, expected)
+    else:
+        result = getattr(grouped, op)()
+        expected = frame.groupby(level=0).apply(lambda h: getattr(h, op)())
+        if sort:
+            expected = expected.sort_index()
+        tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_prod_with_int64_dtype():
+    # GH#46573
+    data = [
+        [1, 11],
+        [1, 41],
+        [1, 17],
+        [1, 37],
+        [1, 7],
+        [1, 29],
+        [1, 31],
+        [1, 2],
+        [1, 3],
+        [1, 43],
+        [1, 5],
+        [1, 47],
+        [1, 19],
+        [1, 88],
+    ]
+    df = DataFrame(data, columns=["A", "B"], dtype="int64")
+    result = df.groupby(["A"]).prod().reset_index()
+    expected = DataFrame({"A": [1], "B": [180970905912331920]}, dtype="int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_std_datetimelike():
+    # GH#48481
+    tdi = pd.timedelta_range("1 Day", periods=10000)
+    ser = Series(tdi)
+    ser[::5] *= 2  # get different std for different groups
+
+    df = ser.to_frame("A").copy()
+
+    df["B"] = ser + Timestamp(0)
+    df["C"] = ser + Timestamp(0, tz="UTC")
+    df.iloc[-1] = pd.NaT  # last group includes NaTs
+
+    gb = df.groupby(list(range(5)) * 2000)
+
+    result = gb.std()
+
+    # Note: this does not _exactly_ match what we would get if we did
+    # [gb.get_group(i).std() for i in gb.groups]
+    #  but it _does_ match the floating point error we get doing the
+    #  same operation on int64 data xref GH#51332
+    td1 = pd.Timedelta("2887 days 11:21:02.326710176")
+    td4 = pd.Timedelta("2886 days 00:42:34.664668096")
+    exp_ser = Series([td1 * 2, td1, td1, td1, td4], index=np.arange(5))
+    expected = DataFrame({"A": exp_ser, "B": exp_ser, "C": exp_ser})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/test_timegrouper.py b/pandas/pandas/tests/groupby/test_timegrouper.py
new file mode 100644
index 0000000000000..550efe9187fe8
--- /dev/null
+++ b/pandas/pandas/tests/groupby/test_timegrouper.py
@@ -0,0 +1,960 @@
+"""
+test with the TimeGrouper / grouping with datetimes
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+from pandas.core.groupby.grouper import Grouper
+from pandas.core.groupby.ops import BinGrouper
+
+
+@pytest.fixture
+def frame_for_truncated_bingrouper():
+    """
+    DataFrame used by groupby_with_truncated_bingrouper, made into
+    a separate fixture for easier reuse in
+    test_groupby_apply_timegrouper_with_nat_apply_squeeze
+    """
+    df = DataFrame(
+        {
+            "Quantity": [18, 3, 5, 1, 9, 3],
+            "Date": [
+                Timestamp(2013, 9, 1, 13, 0),
+                Timestamp(2013, 9, 1, 13, 5),
+                Timestamp(2013, 10, 1, 20, 0),
+                Timestamp(2013, 10, 3, 10, 0),
+                pd.NaT,
+                Timestamp(2013, 9, 2, 14, 0),
+            ],
+        }
+    )
+    return df
+
+
+@pytest.fixture
+def groupby_with_truncated_bingrouper(frame_for_truncated_bingrouper):
+    """
+    GroupBy object such that gb._grouper is a BinGrouper and
+    len(gb._grouper.result_index) < len(gb._grouper.group_keys_seq)
+
+    Aggregations on this groupby should have
+
+        dti = date_range("2013-09-01", "2013-10-01", freq="5D", name="Date")
+
+    As either the index or an index level.
+    """
+    df = frame_for_truncated_bingrouper
+
+    tdg = Grouper(key="Date", freq="5D")
+    gb = df.groupby(tdg)
+
+    # check we're testing the case we're interested in
+    assert len(gb._grouper.result_index) != len(gb._grouper.codes)
+
+    return gb
+
+
+class TestGroupBy:
+    # TODO(infer_string) resample sum introduces 0's
+    # https://github.com/pandas-dev/pandas/issues/60229
+    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
+    def test_groupby_with_timegrouper(self):
+        # GH 4161
+        # TimeGrouper requires a sorted index
+        # also verifies that the resultant index has the correct name
+        df_original = DataFrame(
+            {
+                "Buyer": "Carl Carl Carl Carl Joe Carl".split(),
+                "Quantity": [18, 3, 5, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 9, 1, 13, 0),
+                    datetime(2013, 9, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 3, 10, 0),
+                    datetime(2013, 12, 2, 12, 0),
+                    datetime(2013, 9, 2, 14, 0),
+                ],
+            }
+        )
+
+        # GH 6908 change target column's order
+        df_reordered = df_original.sort_values(by="Quantity")
+
+        for df in [df_original, df_reordered]:
+            df = df.set_index(["Date"])
+
+            exp_dti = date_range(
+                "20130901",
+                "20131205",
+                freq="5D",
+                name="Date",
+                inclusive="left",
+                unit=df.index.unit,
+            )
+            expected = DataFrame(
+                {"Buyer": 0, "Quantity": 0},
+                index=exp_dti,
+            )
+            # Cast to object to avoid implicit cast when setting entry to "CarlCarlCarl"
+            expected = expected.astype({"Buyer": object})
+            expected.iloc[0, 0] = "CarlCarlCarl"
+            expected.iloc[6, 0] = "CarlCarl"
+            expected.iloc[18, 0] = "Joe"
+            expected.iloc[[0, 6, 18], 1] = np.array([24, 6, 9], dtype="int64")
+
+            result1 = df.resample("5D").sum()
+            tm.assert_frame_equal(result1, expected)
+
+            df_sorted = df.sort_index()
+            result2 = df_sorted.groupby(Grouper(freq="5D")).sum()
+            tm.assert_frame_equal(result2, expected)
+
+            result3 = df.groupby(Grouper(freq="5D")).sum()
+            tm.assert_frame_equal(result3, expected)
+
+    @pytest.mark.parametrize("should_sort", [True, False])
+    def test_groupby_with_timegrouper_methods(self, should_sort):
+        # GH 3881
+        # make sure API of timegrouper conforms
+
+        df = DataFrame(
+            {
+                "Branch": "A A A A A B".split(),
+                "Buyer": "Carl Mark Carl Joe Joe Carl".split(),
+                "Quantity": [1, 3, 5, 8, 9, 3],
+                "Date": [
+                    datetime(2013, 1, 1, 13, 0),
+                    datetime(2013, 1, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 12, 2, 12, 0),
+                    datetime(2013, 12, 2, 14, 0),
+                ],
+            }
+        )
+
+        if should_sort:
+            df = df.sort_values(by="Quantity", ascending=False)
+
+        df = df.set_index("Date", drop=False)
+        g = df.groupby(Grouper(freq="6ME"))
+        assert g.group_keys
+
+        assert isinstance(g._grouper, BinGrouper)
+        groups = g.groups
+        assert isinstance(groups, dict)
+        assert len(groups) == 3
+
+    def test_timegrouper_with_reg_groups(self):
+        # GH 3794
+        # allow combination of timegrouper/reg groups
+
+        df_original = DataFrame(
+            {
+                "Branch": "A A A A A A A B".split(),
+                "Buyer": "Carl Mark Carl Carl Joe Joe Joe Carl".split(),
+                "Quantity": [1, 3, 5, 1, 8, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 1, 1, 13, 0),
+                    datetime(2013, 1, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 12, 2, 12, 0),
+                    datetime(2013, 12, 2, 14, 0),
+                ],
+            }
+        ).set_index("Date")
+
+        df_sorted = df_original.sort_values(by="Quantity", ascending=False)
+
+        for df in [df_original, df_sorted]:
+            expected = DataFrame(
+                {
+                    "Buyer": "Carl Joe Mark".split(),
+                    "Quantity": [10, 18, 3],
+                    "Date": [
+                        datetime(2013, 12, 31, 0, 0),
+                        datetime(2013, 12, 31, 0, 0),
+                        datetime(2013, 12, 31, 0, 0),
+                    ],
+                }
+            ).set_index(["Date", "Buyer"])
+
+            msg = "The default value of numeric_only"
+            result = df.groupby([Grouper(freq="YE"), "Buyer"]).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+            expected = DataFrame(
+                {
+                    "Buyer": "Carl Mark Carl Joe".split(),
+                    "Quantity": [1, 3, 9, 18],
+                    "Date": [
+                        datetime(2013, 1, 1, 0, 0),
+                        datetime(2013, 1, 1, 0, 0),
+                        datetime(2013, 7, 1, 0, 0),
+                        datetime(2013, 7, 1, 0, 0),
+                    ],
+                }
+            ).set_index(["Date", "Buyer"])
+            result = df.groupby([Grouper(freq="6MS"), "Buyer"]).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+        df_original = DataFrame(
+            {
+                "Branch": "A A A A A A A B".split(),
+                "Buyer": "Carl Mark Carl Carl Joe Joe Joe Carl".split(),
+                "Quantity": [1, 3, 5, 1, 8, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 10, 1, 13, 0),
+                    datetime(2013, 10, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 10, 2, 12, 0),
+                    datetime(2013, 10, 2, 14, 0),
+                ],
+            }
+        ).set_index("Date")
+
+        df_sorted = df_original.sort_values(by="Quantity", ascending=False)
+        for df in [df_original, df_sorted]:
+            expected = DataFrame(
+                {
+                    "Buyer": "Carl Joe Mark Carl Joe".split(),
+                    "Quantity": [6, 8, 3, 4, 10],
+                    "Date": [
+                        datetime(2013, 10, 1, 0, 0),
+                        datetime(2013, 10, 1, 0, 0),
+                        datetime(2013, 10, 1, 0, 0),
+                        datetime(2013, 10, 2, 0, 0),
+                        datetime(2013, 10, 2, 0, 0),
+                    ],
+                }
+            ).set_index(["Date", "Buyer"])
+
+            result = df.groupby([Grouper(freq="1D"), "Buyer"]).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+            result = df.groupby([Grouper(freq="1ME"), "Buyer"]).sum(numeric_only=True)
+            expected = DataFrame(
+                {
+                    "Buyer": "Carl Joe Mark".split(),
+                    "Quantity": [10, 18, 3],
+                    "Date": [
+                        datetime(2013, 10, 31, 0, 0),
+                        datetime(2013, 10, 31, 0, 0),
+                        datetime(2013, 10, 31, 0, 0),
+                    ],
+                }
+            ).set_index(["Date", "Buyer"])
+            tm.assert_frame_equal(result, expected)
+
+            # passing the name
+            df = df.reset_index()
+            result = df.groupby([Grouper(freq="1ME", key="Date"), "Buyer"]).sum(
+                numeric_only=True
+            )
+            tm.assert_frame_equal(result, expected)
+
+            with pytest.raises(KeyError, match="'The grouper name foo is not found'"):
+                df.groupby([Grouper(freq="1ME", key="foo"), "Buyer"]).sum()
+
+            # passing the level
+            df = df.set_index("Date")
+            result = df.groupby([Grouper(freq="1ME", level="Date"), "Buyer"]).sum(
+                numeric_only=True
+            )
+            tm.assert_frame_equal(result, expected)
+            result = df.groupby([Grouper(freq="1ME", level=0), "Buyer"]).sum(
+                numeric_only=True
+            )
+            tm.assert_frame_equal(result, expected)
+
+            with pytest.raises(ValueError, match="The level foo is not valid"):
+                df.groupby([Grouper(freq="1ME", level="foo"), "Buyer"]).sum()
+
+            # multi names
+            df = df.copy()
+            df["Date"] = df.index + offsets.MonthEnd(2)
+            result = df.groupby([Grouper(freq="1ME", key="Date"), "Buyer"]).sum(
+                numeric_only=True
+            )
+            expected = DataFrame(
+                {
+                    "Buyer": "Carl Joe Mark".split(),
+                    "Quantity": [10, 18, 3],
+                    "Date": [
+                        datetime(2013, 11, 30, 0, 0),
+                        datetime(2013, 11, 30, 0, 0),
+                        datetime(2013, 11, 30, 0, 0),
+                    ],
+                }
+            ).set_index(["Date", "Buyer"])
+            tm.assert_frame_equal(result, expected)
+
+            # error as we have both a level and a name!
+            msg = "The Grouper cannot specify both a key and a level!"
+            with pytest.raises(ValueError, match=msg):
+                df.groupby(
+                    [Grouper(freq="1ME", key="Date", level="Date"), "Buyer"]
+                ).sum()
+
+            # single groupers
+            expected = DataFrame(
+                [[31]],
+                columns=["Quantity"],
+                index=DatetimeIndex(
+                    [datetime(2013, 10, 31, 0, 0)], freq=offsets.MonthEnd(), name="Date"
+                ),
+            )
+            result = df.groupby(Grouper(freq="1ME")).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+            result = df.groupby([Grouper(freq="1ME")]).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+            expected.index = expected.index.shift(1)
+            assert expected.index.freq == offsets.MonthEnd()
+            result = df.groupby(Grouper(freq="1ME", key="Date")).sum(numeric_only=True)
+            tm.assert_frame_equal(result, expected)
+
+            result = df.groupby([Grouper(freq="1ME", key="Date")]).sum(
+                numeric_only=True
+            )
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["D", "ME", "YE", "QE-APR"])
+    def test_timegrouper_with_reg_groups_freq(self, freq):
+        # GH 6764 multiple grouping with/without sort
+        df = DataFrame(
+            {
+                "date": pd.to_datetime(
+                    [
+                        "20121002",
+                        "20121007",
+                        "20130130",
+                        "20130202",
+                        "20130305",
+                        "20121002",
+                        "20121207",
+                        "20130130",
+                        "20130202",
+                        "20130305",
+                        "20130202",
+                        "20130305",
+                    ]
+                ),
+                "user_id": [1, 1, 1, 1, 1, 3, 3, 3, 5, 5, 5, 5],
+                "whole_cost": [
+                    1790,
+                    364,
+                    280,
+                    259,
+                    201,
+                    623,
+                    90,
+                    312,
+                    359,
+                    301,
+                    359,
+                    801,
+                ],
+                "cost1": [12, 15, 10, 24, 39, 1, 0, 90, 45, 34, 1, 12],
+            }
+        ).set_index("date")
+
+        expected = (
+            df.groupby("user_id")["whole_cost"]
+            .resample(freq)
+            .sum(min_count=1)  # XXX
+            .dropna()
+            .reorder_levels(["date", "user_id"])
+            .sort_index()
+            .astype("int64")
+        )
+        expected.name = "whole_cost"
+
+        result1 = (
+            df.sort_index().groupby([Grouper(freq=freq), "user_id"])["whole_cost"].sum()
+        )
+        tm.assert_series_equal(result1, expected)
+
+        result2 = df.groupby([Grouper(freq=freq), "user_id"])["whole_cost"].sum()
+        tm.assert_series_equal(result2, expected)
+
+    def test_timegrouper_get_group(self):
+        # GH 6914
+
+        df_original = DataFrame(
+            {
+                "Buyer": "Carl Joe Joe Carl Joe Carl".split(),
+                "Quantity": [18, 3, 5, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 9, 1, 13, 0),
+                    datetime(2013, 9, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 3, 10, 0),
+                    datetime(2013, 12, 2, 12, 0),
+                    datetime(2013, 9, 2, 14, 0),
+                ],
+            }
+        )
+        df_reordered = df_original.sort_values(by="Quantity")
+
+        # single grouping
+        expected_list = [
+            df_original.iloc[[0, 1, 5]],
+            df_original.iloc[[2, 3]],
+            df_original.iloc[[4]],
+        ]
+        dt_list = ["2013-09-30", "2013-10-31", "2013-12-31"]
+
+        for df in [df_original, df_reordered]:
+            grouped = df.groupby(Grouper(freq="ME", key="Date"))
+            for t, expected in zip(dt_list, expected_list):
+                dt = Timestamp(t)
+                result = grouped.get_group(dt)
+                tm.assert_frame_equal(result, expected)
+
+        # multiple grouping
+        expected_list = [
+            df_original.iloc[[1]],
+            df_original.iloc[[3]],
+            df_original.iloc[[4]],
+        ]
+        g_list = [("Joe", "2013-09-30"), ("Carl", "2013-10-31"), ("Joe", "2013-12-31")]
+
+        for df in [df_original, df_reordered]:
+            grouped = df.groupby(["Buyer", Grouper(freq="ME", key="Date")])
+            for (b, t), expected in zip(g_list, expected_list):
+                dt = Timestamp(t)
+                result = grouped.get_group((b, dt))
+                tm.assert_frame_equal(result, expected)
+
+        # with index
+        df_original = df_original.set_index("Date")
+        df_reordered = df_original.sort_values(by="Quantity")
+
+        expected_list = [
+            df_original.iloc[[0, 1, 5]],
+            df_original.iloc[[2, 3]],
+            df_original.iloc[[4]],
+        ]
+
+        for df in [df_original, df_reordered]:
+            grouped = df.groupby(Grouper(freq="ME"))
+            for t, expected in zip(dt_list, expected_list):
+                dt = Timestamp(t)
+                result = grouped.get_group(dt)
+                tm.assert_frame_equal(result, expected)
+
+    def test_timegrouper_apply_return_type_series(self):
+        # Using `apply` with the `TimeGrouper` should give the
+        # same return type as an `apply` with a `Grouper`.
+        # Issue #11742
+        df = DataFrame({"date": ["10/10/2000", "11/10/2000"], "value": [10, 13]})
+        df_dt = df.copy()
+        df_dt["date"] = pd.to_datetime(df_dt["date"])
+
+        def sumfunc_series(x):
+            return Series([x["value"].sum()], ("sum",))
+
+        expected = df.groupby(Grouper(key="date")).apply(sumfunc_series)
+        result = df_dt.groupby(Grouper(freq="ME", key="date")).apply(sumfunc_series)
+        tm.assert_frame_equal(
+            result.reset_index(drop=True), expected.reset_index(drop=True)
+        )
+
+    def test_timegrouper_apply_return_type_value(self):
+        # Using `apply` with the `TimeGrouper` should give the
+        # same return type as an `apply` with a `Grouper`.
+        # Issue #11742
+        df = DataFrame({"date": ["10/10/2000", "11/10/2000"], "value": [10, 13]})
+        df_dt = df.copy()
+        df_dt["date"] = pd.to_datetime(df_dt["date"])
+
+        def sumfunc_value(x):
+            return x.value.sum()
+
+        expected = df.groupby(Grouper(key="date")).apply(sumfunc_value)
+        result = df_dt.groupby(Grouper(freq="ME", key="date")).apply(sumfunc_value)
+        tm.assert_series_equal(
+            result.reset_index(drop=True), expected.reset_index(drop=True)
+        )
+
+    def test_groupby_groups_datetimeindex(self):
+        # GH#1430
+        periods = 1000
+        ind = date_range(start="2012/1/1", freq="5min", periods=periods)
+        df = DataFrame(
+            {"high": np.arange(periods), "low": np.arange(periods)}, index=ind
+        )
+        grouped = df.groupby(lambda x: datetime(x.year, x.month, x.day))
+
+        # it works!
+        groups = grouped.groups
+        assert isinstance(next(iter(groups.keys())), datetime)
+
+    def test_groupby_groups_datetimeindex2(self):
+        # GH#11442
+        index = date_range("2015/01/01", periods=5, name="date")
+        df = DataFrame({"A": [5, 6, 7, 8, 9], "B": [1, 2, 3, 4, 5]}, index=index)
+        result = df.groupby(level="date").groups
+        dates = ["2015-01-05", "2015-01-04", "2015-01-03", "2015-01-02", "2015-01-01"]
+        expected = {
+            Timestamp(date): DatetimeIndex([date], name="date") for date in dates
+        }
+        tm.assert_dict_equal(result, expected)
+
+        grouped = df.groupby(level="date")
+        for date in dates:
+            result = grouped.get_group(date)
+            data = [[df.loc[date, "A"], df.loc[date, "B"]]]
+            expected_index = DatetimeIndex(
+                [date], name="date", freq="D", dtype=index.dtype
+            )
+            expected = DataFrame(data, columns=list("AB"), index=expected_index)
+            tm.assert_frame_equal(result, expected)
+
+    def test_groupby_groups_datetimeindex_tz(self):
+        # GH 3950
+        dates = [
+            "2011-07-19 07:00:00",
+            "2011-07-19 08:00:00",
+            "2011-07-19 09:00:00",
+            "2011-07-19 07:00:00",
+            "2011-07-19 08:00:00",
+            "2011-07-19 09:00:00",
+        ]
+        df = DataFrame(
+            {
+                "label": ["a", "a", "a", "b", "b", "b"],
+                "datetime": dates,
+                "value1": np.arange(6, dtype="int64"),
+                "value2": [1, 2] * 3,
+            }
+        )
+        df["datetime"] = df["datetime"].apply(lambda d: Timestamp(d, tz="US/Pacific"))
+
+        exp_idx1 = DatetimeIndex(
+            [
+                "2011-07-19 07:00:00",
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+                "2011-07-19 09:00:00",
+            ],
+            tz="US/Pacific",
+            name="datetime",
+        )
+        exp_idx2 = Index(["a", "b"] * 3, name="label")
+        exp_idx = MultiIndex.from_arrays([exp_idx1, exp_idx2])
+        expected = DataFrame(
+            {"value1": [0, 3, 1, 4, 2, 5], "value2": [1, 2, 2, 1, 1, 2]},
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        result = df.groupby(["datetime", "label"]).sum()
+        tm.assert_frame_equal(result, expected)
+
+        # by level
+        didx = DatetimeIndex(dates, tz="Asia/Tokyo")
+        df = DataFrame(
+            {"value1": np.arange(6, dtype="int64"), "value2": [1, 2, 3, 1, 2, 3]},
+            index=didx,
+        )
+
+        exp_idx = DatetimeIndex(
+            ["2011-07-19 07:00:00", "2011-07-19 08:00:00", "2011-07-19 09:00:00"],
+            tz="Asia/Tokyo",
+        )
+        expected = DataFrame(
+            {"value1": [3, 5, 7], "value2": [2, 4, 6]},
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        result = df.groupby(level=0).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_datetime64_handling_groupby(self):
+        # it works!
+        df = DataFrame(
+            [(3, np.datetime64("2012-07-03")), (3, np.datetime64("2012-07-04"))],
+            columns=["a", "date"],
+        )
+        result = df.groupby("a").first()
+        assert result["date"][3] == Timestamp("2012-07-03")
+
+    def test_groupby_multi_timezone(self):
+        # combining multiple / different timezones yields UTC
+        df = DataFrame(
+            {
+                "value": range(5),
+                "date": [
+                    "2000-01-28 16:47:00",
+                    "2000-01-29 16:48:00",
+                    "2000-01-30 16:49:00",
+                    "2000-01-31 16:50:00",
+                    "2000-01-01 16:50:00",
+                ],
+                "tz": [
+                    "America/Chicago",
+                    "America/Chicago",
+                    "America/Los_Angeles",
+                    "America/Chicago",
+                    "America/New_York",
+                ],
+            }
+        )
+
+        result = df.groupby("tz", group_keys=False).date.apply(
+            lambda x: pd.to_datetime(x).dt.tz_localize(x.name)
+        )
+
+        expected = Series(
+            [
+                Timestamp("2000-01-28 16:47:00-0600", tz="America/Chicago"),
+                Timestamp("2000-01-29 16:48:00-0600", tz="America/Chicago"),
+                Timestamp("2000-01-30 16:49:00-0800", tz="America/Los_Angeles"),
+                Timestamp("2000-01-31 16:50:00-0600", tz="America/Chicago"),
+                Timestamp("2000-01-01 16:50:00-0500", tz="America/New_York"),
+            ],
+            name="date",
+            dtype=object,
+        )
+        tm.assert_series_equal(result, expected)
+
+        tz = "America/Chicago"
+        res_values = df.groupby("tz").date.get_group(tz)
+        result = pd.to_datetime(res_values).dt.tz_localize(tz)
+        exp_values = Series(
+            ["2000-01-28 16:47:00", "2000-01-29 16:48:00", "2000-01-31 16:50:00"],
+            index=[0, 1, 3],
+            name="date",
+        )
+        expected = pd.to_datetime(exp_values).dt.tz_localize(tz)
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_groups_periods(self):
+        dates = [
+            "2011-07-19 07:00:00",
+            "2011-07-19 08:00:00",
+            "2011-07-19 09:00:00",
+            "2011-07-19 07:00:00",
+            "2011-07-19 08:00:00",
+            "2011-07-19 09:00:00",
+        ]
+        df = DataFrame(
+            {
+                "label": ["a", "a", "a", "b", "b", "b"],
+                "period": [pd.Period(d, freq="h") for d in dates],
+                "value1": np.arange(6, dtype="int64"),
+                "value2": [1, 2] * 3,
+            }
+        )
+
+        exp_idx1 = pd.PeriodIndex(
+            [
+                "2011-07-19 07:00:00",
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+                "2011-07-19 09:00:00",
+            ],
+            freq="h",
+            name="period",
+        )
+        exp_idx2 = Index(["a", "b"] * 3, name="label")
+        exp_idx = MultiIndex.from_arrays([exp_idx1, exp_idx2])
+        expected = DataFrame(
+            {"value1": [0, 3, 1, 4, 2, 5], "value2": [1, 2, 2, 1, 1, 2]},
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        result = df.groupby(["period", "label"]).sum()
+        tm.assert_frame_equal(result, expected)
+
+        # by level
+        didx = pd.PeriodIndex(dates, freq="h")
+        df = DataFrame(
+            {"value1": np.arange(6, dtype="int64"), "value2": [1, 2, 3, 1, 2, 3]},
+            index=didx,
+        )
+
+        exp_idx = pd.PeriodIndex(
+            ["2011-07-19 07:00:00", "2011-07-19 08:00:00", "2011-07-19 09:00:00"],
+            freq="h",
+        )
+        expected = DataFrame(
+            {"value1": [3, 5, 7], "value2": [2, 4, 6]},
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        result = df.groupby(level=0).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_first_datetime64(self):
+        df = DataFrame([(1, 1351036800000000000), (2, 1351036800000000000)])
+        df[1] = df[1].astype("M8[ns]")
+
+        assert issubclass(df[1].dtype.type, np.datetime64)
+
+        result = df.groupby(level=0).first()
+        got_dt = result[1].dtype
+        assert issubclass(got_dt.type, np.datetime64)
+
+        result = df[1].groupby(level=0).first()
+        got_dt = result.dtype
+        assert issubclass(got_dt.type, np.datetime64)
+
+    def test_groupby_max_datetime64(self):
+        # GH 5869
+        # datetimelike dtype conversion from int
+        df = DataFrame({"A": Timestamp("20130101"), "B": np.arange(5)})
+        # TODO: can we retain second reso in .apply here?
+        expected = df.groupby("A")["A"].apply(lambda x: x.max()).astype("M8[s]")
+        result = df.groupby("A")["A"].max()
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_datetime64_32_bit(self):
+        # GH 6410 / numpy 4328
+        # 32-bit under 1.9-dev indexing issue
+
+        df = DataFrame({"A": range(2), "B": [Timestamp("2000-01-1")] * 2})
+        result = df.groupby("A")["B"].transform("min")
+        expected = Series([Timestamp("2000-01-1")] * 2, name="B")
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_with_timezone_selection(self):
+        # GH 11616
+        # Test that column selection returns output in correct timezone.
+
+        df = DataFrame(
+            {
+                "factor": np.random.default_rng(2).integers(0, 3, size=60),
+                "time": date_range("01/01/2000 00:00", periods=60, freq="s", tz="UTC"),
+            }
+        )
+        df1 = df.groupby("factor").max()["time"]
+        df2 = df.groupby("factor")["time"].max()
+        tm.assert_series_equal(df1, df2)
+
+    def test_timezone_info(self):
+        # see gh-11682: Timezone info lost when broadcasting
+        # scalar datetime to DataFrame
+        utc = timezone.utc
+        df = DataFrame({"a": [1], "b": [datetime.now(utc)]})
+        assert df["b"][0].tzinfo == utc
+        df = DataFrame({"a": [1, 2, 3]})
+        df["b"] = datetime.now(utc)
+        assert df["b"][0].tzinfo == utc
+
+    def test_datetime_count(self):
+        df = DataFrame(
+            {"a": [1, 2, 3] * 2, "dates": date_range("now", periods=6, freq="min")}
+        )
+        result = df.groupby("a").dates.count()
+        expected = Series([2, 2, 2], index=Index([1, 2, 3], name="a"), name="dates")
+        tm.assert_series_equal(result, expected)
+
+    def test_first_last_max_min_on_time_data(self):
+        # GH 10295
+        # Verify that NaT is not in the result of max, min, first and last on
+        # Dataframe with datetime or timedelta values.
+        df_test = DataFrame(
+            {
+                "dt": [
+                    np.nan,
+                    "2015-07-24 10:10",
+                    "2015-07-25 11:11",
+                    "2015-07-23 12:12",
+                    np.nan,
+                ],
+                "td": [
+                    np.nan,
+                    timedelta(days=1),
+                    timedelta(days=2),
+                    timedelta(days=3),
+                    np.nan,
+                ],
+            }
+        )
+        df_test.dt = pd.to_datetime(df_test.dt)
+        df_test["group"] = "A"
+        df_ref = df_test[df_test.dt.notna()]
+
+        grouped_test = df_test.groupby("group")
+        grouped_ref = df_ref.groupby("group")
+
+        tm.assert_frame_equal(grouped_ref.max(), grouped_test.max())
+        tm.assert_frame_equal(grouped_ref.min(), grouped_test.min())
+        tm.assert_frame_equal(grouped_ref.first(), grouped_test.first())
+        tm.assert_frame_equal(grouped_ref.last(), grouped_test.last())
+
+    def test_nunique_with_timegrouper_and_nat(self):
+        # GH 17575
+        test = DataFrame(
+            {
+                "time": [
+                    Timestamp("2016-06-28 09:35:35"),
+                    pd.NaT,
+                    Timestamp("2016-06-28 16:46:28"),
+                ],
+                "data": ["1", "2", "3"],
+            }
+        )
+
+        grouper = Grouper(key="time", freq="h")
+        result = test.groupby(grouper)["data"].nunique()
+        expected = test[test.time.notnull()].groupby(grouper)["data"].nunique()
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+    def test_scalar_call_versus_list_call(self):
+        # Issue: 17530
+        data_frame = {
+            "location": ["shanghai", "beijing", "shanghai"],
+            "time": Series(
+                ["2017-08-09 13:32:23", "2017-08-11 23:23:15", "2017-08-11 22:23:15"],
+                dtype="datetime64[ns]",
+            ),
+            "value": [1, 2, 3],
+        }
+        data_frame = DataFrame(data_frame).set_index("time")
+        grouper = Grouper(freq="D")
+
+        grouped = data_frame.groupby(grouper)
+        result = grouped.count()
+        grouped = data_frame.groupby([grouper])
+        expected = grouped.count()
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_grouper_period_index(self):
+        # GH 32108
+        periods = 2
+        index = pd.period_range(
+            start="2018-01", periods=periods, freq="M", name="Month"
+        )
+        period_series = Series(range(periods), index=index)
+        result = period_series.groupby(period_series.index.month).sum()
+
+        expected = Series(
+            range(periods), index=Index(range(1, periods + 1), name=index.name)
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_apply_timegrouper_with_nat_dict_returns(
+        self, groupby_with_truncated_bingrouper
+    ):
+        # GH#43500 case where gb._grouper.result_index and gb._grouper.group_keys_seq
+        #  have different lengths that goes through the `isinstance(values[0], dict)`
+        #  path
+        gb = groupby_with_truncated_bingrouper
+
+        res = gb["Quantity"].apply(lambda x: {"foo": len(x)})
+
+        df = gb.obj
+        unit = df["Date"]._values.unit
+        dti = date_range("2013-09-01", "2013-10-01", freq="5D", name="Date", unit=unit)
+        mi = MultiIndex.from_arrays([dti, ["foo"] * len(dti)])
+        expected = Series([3, 0, 0, 0, 0, 0, 2], index=mi, name="Quantity")
+        tm.assert_series_equal(res, expected)
+
+    def test_groupby_apply_timegrouper_with_nat_scalar_returns(
+        self, groupby_with_truncated_bingrouper
+    ):
+        # GH#43500 Previously raised ValueError bc used index with incorrect
+        #  length in wrap_applied_result
+        gb = groupby_with_truncated_bingrouper
+
+        res = gb["Quantity"].apply(lambda x: x.iloc[0] if len(x) else np.nan)
+
+        df = gb.obj
+        unit = df["Date"]._values.unit
+        dti = date_range("2013-09-01", "2013-10-01", freq="5D", name="Date", unit=unit)
+        expected = Series(
+            [18, np.nan, np.nan, np.nan, np.nan, np.nan, 5],
+            index=dti._with_freq(None),
+            name="Quantity",
+        )
+
+        tm.assert_series_equal(res, expected)
+
+    def test_groupby_apply_timegrouper_with_nat_apply_squeeze(
+        self, frame_for_truncated_bingrouper
+    ):
+        df = frame_for_truncated_bingrouper
+
+        # We need to create a GroupBy object with only one non-NaT group,
+        #  so use a huge freq so that all non-NaT dates will be grouped together
+        tdg = Grouper(key="Date", freq="100YE")
+        gb = df.groupby(tdg)
+
+        # check that we will go through the singular_series path
+        #  in _wrap_applied_output_series
+        assert gb.ngroups == 1
+        assert gb._selected_obj.index.nlevels == 1
+
+        # function that returns a Series
+        res = gb.apply(lambda x: x["Quantity"] * 2)
+
+        dti = Index([Timestamp("2013-12-31")], dtype=df["Date"].dtype, name="Date")
+        expected = DataFrame(
+            [[36, 6, 6, 10, 2]],
+            index=dti,
+            columns=Index([0, 1, 5, 2, 3], name="Quantity"),
+        )
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.single_cpu
+    def test_groupby_agg_numba_timegrouper_with_nat(
+        self, groupby_with_truncated_bingrouper
+    ):
+        pytest.importorskip("numba")
+
+        # See discussion in GH#43487
+        gb = groupby_with_truncated_bingrouper
+
+        result = gb["Quantity"].aggregate(
+            lambda values, index: np.nanmean(values), engine="numba"
+        )
+
+        expected = gb["Quantity"].aggregate("mean")
+        tm.assert_series_equal(result, expected)
+
+        result_df = gb[["Quantity"]].aggregate(
+            lambda values, index: np.nanmean(values), engine="numba"
+        )
+        expected_df = gb[["Quantity"]].aggregate("mean")
+        tm.assert_frame_equal(result_df, expected_df)
diff --git a/pandas/pandas/tests/groupby/transform/__init__.py b/pandas/pandas/tests/groupby/transform/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/groupby/transform/test_numba.py b/pandas/pandas/tests/groupby/transform/test_numba.py
new file mode 100644
index 0000000000000..e19b7592f75b3
--- /dev/null
+++ b/pandas/pandas/tests/groupby/transform/test_numba.py
@@ -0,0 +1,331 @@
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+from pandas.errors import NumbaUtilError
+
+from pandas import (
+    DataFrame,
+    Series,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [pytest.mark.single_cpu]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+def test_correct_function_signature():
+    pytest.importorskip("numba")
+
+    def incorrect_function(x):
+        return x + 1
+
+    data = DataFrame(
+        {"key": ["a", "a", "b", "b", "a"], "data": [1.0, 2.0, 3.0, 4.0, 5.0]},
+        columns=["key", "data"],
+    )
+    with pytest.raises(NumbaUtilError, match="The first 2"):
+        data.groupby("key").transform(incorrect_function, engine="numba")
+
+    with pytest.raises(NumbaUtilError, match="The first 2"):
+        data.groupby("key")["data"].transform(incorrect_function, engine="numba")
+
+
+def test_check_nopython_kwargs():
+    pytest.importorskip("numba")
+
+    def incorrect_function(values, index, *, a):
+        return values + a
+
+    def correct_function(values, index, a):
+        return values + a
+
+    data = DataFrame(
+        {"key": ["a", "a", "b", "b", "a"], "data": [1.0, 2.0, 3.0, 4.0, 5.0]},
+        columns=["key", "data"],
+    )
+    # py signature binding
+    with pytest.raises(
+        TypeError, match="missing a required (keyword-only argument|argument): 'a'"
+    ):
+        data.groupby("key").transform(incorrect_function, engine="numba", b=1)
+    with pytest.raises(TypeError, match="missing a required argument: 'a'"):
+        data.groupby("key").transform(correct_function, engine="numba", b=1)
+
+    with pytest.raises(
+        TypeError, match="missing a required (keyword-only argument|argument): 'a'"
+    ):
+        data.groupby("key")["data"].transform(incorrect_function, engine="numba", b=1)
+    with pytest.raises(TypeError, match="missing a required argument: 'a'"):
+        data.groupby("key")["data"].transform(correct_function, engine="numba", b=1)
+
+    # numba signature check after binding
+    with pytest.raises(NumbaUtilError, match="numba does not support"):
+        data.groupby("key").transform(incorrect_function, engine="numba", a=1)
+    actual = data.groupby("key").transform(correct_function, engine="numba", a=1)
+    tm.assert_frame_equal(data[["data"]] + 1, actual)
+
+    with pytest.raises(NumbaUtilError, match="numba does not support"):
+        data.groupby("key")["data"].transform(incorrect_function, engine="numba", a=1)
+    actual = data.groupby("key")["data"].transform(
+        correct_function, engine="numba", a=1
+    )
+    tm.assert_series_equal(data["data"] + 1, actual)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+@pytest.mark.parametrize("jit", [True, False])
+def test_numba_vs_cython(jit, frame_or_series, nogil, parallel, nopython, as_index):
+    pytest.importorskip("numba")
+
+    def func(values, index):
+        return values + 1
+
+    if jit:
+        # Test accepted jitted functions
+        import numba
+
+        func = numba.jit(func)
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+    grouped = data.groupby(0, as_index=as_index)
+    if frame_or_series is Series:
+        grouped = grouped[1]
+
+    result = grouped.transform(func, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.transform(lambda x: x + 1, engine="cython")
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+@pytest.mark.parametrize("jit", [True, False])
+def test_cache(jit, frame_or_series, nogil, parallel, nopython):
+    # Test that the functions are cached correctly if we switch functions
+    pytest.importorskip("numba")
+
+    def func_1(values, index):
+        return values + 1
+
+    def func_2(values, index):
+        return values * 5
+
+    if jit:
+        import numba
+
+        func_1 = numba.jit(func_1)
+        func_2 = numba.jit(func_2)
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+    grouped = data.groupby(0)
+    if frame_or_series is Series:
+        grouped = grouped[1]
+
+    result = grouped.transform(func_1, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.transform(lambda x: x + 1, engine="cython")
+    tm.assert_equal(result, expected)
+
+    result = grouped.transform(func_2, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.transform(lambda x: x * 5, engine="cython")
+    tm.assert_equal(result, expected)
+
+    # Retest func_1 which should use the cache
+    result = grouped.transform(func_1, engine="numba", engine_kwargs=engine_kwargs)
+    expected = grouped.transform(lambda x: x + 1, engine="cython")
+    tm.assert_equal(result, expected)
+
+
+def test_use_global_config():
+    pytest.importorskip("numba")
+
+    def func_1(values, index):
+        return values + 1
+
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    grouped = data.groupby(0)
+    expected = grouped.transform(func_1, engine="numba")
+    with option_context("compute.use_numba", True):
+        result = grouped.transform(func_1, engine=None)
+    tm.assert_frame_equal(expected, result)
+
+
+# TODO: Test more than just reductions (e.g. actually test transformations once we have
+@pytest.mark.parametrize(
+    "agg_func", [["min", "max"], "min", {"B": ["min", "max"], "C": "sum"}]
+)
+def test_string_cython_vs_numba(agg_func, numba_supported_reductions):
+    pytest.importorskip("numba")
+    agg_func, kwargs = numba_supported_reductions
+    data = DataFrame(
+        {0: ["a", "a", "b", "b", "a"], 1: [1.0, 2.0, 3.0, 4.0, 5.0]}, columns=[0, 1]
+    )
+    grouped = data.groupby(0)
+
+    result = grouped.transform(agg_func, engine="numba", **kwargs)
+    expected = grouped.transform(agg_func, engine="cython", **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+    result = grouped[1].transform(agg_func, engine="numba", **kwargs)
+    expected = grouped[1].transform(agg_func, engine="cython", **kwargs)
+    tm.assert_series_equal(result, expected)
+
+
+def test_args_not_cached():
+    # GH 41647
+    pytest.importorskip("numba")
+
+    def sum_last(values, index, n):
+        return values[-n:].sum()
+
+    df = DataFrame({"id": [0, 0, 1, 1], "x": [1, 1, 1, 1]})
+    grouped_x = df.groupby("id")["x"]
+    result = grouped_x.transform(sum_last, 1, engine="numba")
+    expected = Series([1.0] * 4, name="x")
+    tm.assert_series_equal(result, expected)
+
+    result = grouped_x.transform(sum_last, 2, engine="numba")
+    expected = Series([2.0] * 4, name="x")
+    tm.assert_series_equal(result, expected)
+
+
+def test_index_data_correctly_passed():
+    # GH 43133
+    pytest.importorskip("numba")
+
+    def f(values, index):
+        return index - 1
+
+    df = DataFrame({"group": ["A", "A", "B"], "v": [4, 5, 6]}, index=[-1, -2, -3])
+    result = df.groupby("group").transform(f, engine="numba")
+    expected = DataFrame([-2.0, -3.0, -4.0], columns=["v"], index=[-1, -2, -3])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_index_order_consistency_preserved():
+    # GH 57069
+    pytest.importorskip("numba")
+
+    def f(values, index):
+        return values
+
+    df = DataFrame(
+        {"vals": [0.0, 1.0, 2.0, 3.0], "group": [0, 1, 0, 1]}, index=range(3, -1, -1)
+    )
+    result = df.groupby("group")["vals"].transform(f, engine="numba")
+    expected = Series([0.0, 1.0, 2.0, 3.0], index=range(3, -1, -1), name="vals")
+    tm.assert_series_equal(result, expected)
+
+
+def test_engine_kwargs_not_cached():
+    # If the user passes a different set of engine_kwargs don't return the same
+    # jitted function
+    pytest.importorskip("numba")
+    nogil = True
+    parallel = False
+    nopython = True
+
+    def func_kwargs(values, index):
+        return nogil + parallel + nopython
+
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    df = DataFrame({"value": [0, 0, 0]})
+    result = df.groupby(level=0).transform(
+        func_kwargs, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame({"value": [2.0, 2.0, 2.0]})
+    tm.assert_frame_equal(result, expected)
+
+    nogil = False
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    result = df.groupby(level=0).transform(
+        func_kwargs, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame({"value": [1.0, 1.0, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore")
+def test_multiindex_one_key(nogil, parallel, nopython):
+    pytest.importorskip("numba")
+
+    def numba_func(values, index):
+        return 1
+
+    df = DataFrame([{"A": 1, "B": 2, "C": 3}]).set_index(["A", "B"])
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    result = df.groupby("A").transform(
+        numba_func, engine="numba", engine_kwargs=engine_kwargs
+    )
+    expected = DataFrame([{"A": 1, "B": 2, "C": 1.0}]).set_index(["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multiindex_multi_key_not_supported(nogil, parallel, nopython):
+    pytest.importorskip("numba")
+
+    def numba_func(values, index):
+        return 1
+
+    df = DataFrame([{"A": 1, "B": 2, "C": 3}]).set_index(["A", "B"])
+    engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+    with pytest.raises(NotImplementedError, match="more than 1 grouping labels"):
+        df.groupby(["A", "B"]).transform(
+            numba_func, engine="numba", engine_kwargs=engine_kwargs
+        )
+
+
+def test_multilabel_numba_vs_cython(numba_supported_reductions):
+    pytest.importorskip("numba")
+    reduction, kwargs = numba_supported_reductions
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+    gb = df.groupby(["A", "B"])
+    res_agg = gb.transform(reduction, engine="numba", **kwargs)
+    expected_agg = gb.transform(reduction, engine="cython", **kwargs)
+    tm.assert_frame_equal(res_agg, expected_agg)
+
+
+def test_multilabel_udf_numba_vs_cython():
+    pytest.importorskip("numba")
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+            "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+            "C": np.random.default_rng(2).standard_normal(8),
+            "D": np.random.default_rng(2).standard_normal(8),
+        }
+    )
+    gb = df.groupby(["A", "B"])
+    result = gb.transform(
+        lambda values, index: (values - values.min()) / (values.max() - values.min()),
+        engine="numba",
+    )
+    expected = gb.transform(
+        lambda x: (x - x.min()) / (x.max() - x.min()), engine="cython"
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/groupby/transform/test_transform.py b/pandas/pandas/tests/groupby/transform/test_transform.py
new file mode 100644
index 0000000000000..fecd20fd6cece
--- /dev/null
+++ b/pandas/pandas/tests/groupby/transform/test_transform.py
@@ -0,0 +1,1596 @@
+"""test with the .transform"""
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.common import ensure_platform_int
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.groupby import get_groupby_method_args
+
+
+def assert_fp_equal(a, b):
+    assert (np.abs(a - b) < 1e-12).all()
+
+
+def test_transform():
+    data = Series(np.arange(9) // 3, index=np.arange(9))
+
+    index = np.arange(9)
+    np.random.default_rng(2).shuffle(index)
+    data = data.reindex(index)
+
+    grouped = data.groupby(lambda x: x // 3)
+
+    transformed = grouped.transform(lambda x: x * x.sum())
+    assert transformed[7] == 12
+
+    # GH 8046
+    # make sure that we preserve the input order
+
+    df = DataFrame(
+        np.arange(6, dtype="int64").reshape(3, 2), columns=["a", "b"], index=[0, 2, 1]
+    )
+    key = [0, 0, 1]
+    expected = (
+        df.sort_index()
+        .groupby(key)
+        .transform(lambda x: x - x.mean())
+        .groupby(key)
+        .mean()
+    )
+    result = df.groupby(key).transform(lambda x: x - x.mean()).groupby(key).mean()
+    tm.assert_frame_equal(result, expected)
+
+    def demean(arr):
+        return arr - arr.mean(axis=0)
+
+    people = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 5)),
+        columns=["a", "b", "c", "d", "e"],
+        index=["Joe", "Steve", "Wes", "Jim", "Travis"],
+    )
+    key = ["one", "two", "one", "two", "one"]
+    result = people.groupby(key).transform(demean).groupby(key).mean()
+    expected = people.groupby(key, group_keys=False).apply(demean).groupby(key).mean()
+    tm.assert_frame_equal(result, expected)
+
+    # GH 8430
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((50, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=50, freq="B"),
+    )
+    g = df.groupby(pd.Grouper(freq="ME"))
+    g.transform(lambda x: x - 1)
+
+    # GH 9700
+    df = DataFrame({"a": range(5, 10), "b": range(5)})
+    result = df.groupby("a").transform(max)
+    expected = DataFrame({"b": range(5)})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_fast():
+    df = DataFrame(
+        {
+            "id": np.arange(10) / 3,
+            "val": np.random.default_rng(2).standard_normal(10),
+        }
+    )
+
+    grp = df.groupby("id")["val"]
+
+    values = np.repeat(grp.mean().values, ensure_platform_int(grp.count().values))
+    expected = Series(values, index=df.index, name="val")
+
+    result = grp.transform(np.mean)
+    tm.assert_series_equal(result, expected)
+
+    result = grp.transform("mean")
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_fast2():
+    # GH 12737
+    df = DataFrame(
+        {
+            "grouping": [0, 1, 1, 3],
+            "f": [1.1, 2.1, 3.1, 4.5],
+            "d": date_range("2014-1-1", "2014-1-4"),
+            "i": [1, 2, 3, 4],
+        },
+        columns=["grouping", "f", "i", "d"],
+    )
+    result = df.groupby("grouping").transform("first")
+
+    dates = Index(
+        [
+            Timestamp("2014-1-1"),
+            Timestamp("2014-1-2"),
+            Timestamp("2014-1-2"),
+            Timestamp("2014-1-4"),
+        ],
+        dtype="M8[ns]",
+    )
+    expected = DataFrame(
+        {"f": [1.1, 2.1, 2.1, 4.5], "d": dates, "i": [1, 2, 2, 4]},
+        columns=["f", "i", "d"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # selection
+    result = df.groupby("grouping")[["f", "i"]].transform("first")
+    expected = expected[["f", "i"]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_fast3():
+    # dup columns
+    df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["g", "a", "a"])
+    result = df.groupby("g").transform("first")
+    expected = df.drop("g", axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_broadcast(tsframe, ts):
+    grouped = ts.groupby(lambda x: x.month)
+    result = grouped.transform(np.mean)
+
+    tm.assert_index_equal(result.index, ts.index)
+    for _, gp in grouped:
+        assert_fp_equal(result.reindex(gp.index), gp.mean())
+
+    grouped = tsframe.groupby(lambda x: x.month)
+    result = grouped.transform(np.mean)
+    tm.assert_index_equal(result.index, tsframe.index)
+    for _, gp in grouped:
+        agged = gp.mean(axis=0)
+        res = result.reindex(gp.index)
+        for col in tsframe:
+            assert_fp_equal(res[col], agged[col])
+
+
+def test_transform_axis_ts(tsframe):
+    # make sure that we are setting the axes correctly
+    # in the presence of a non-monotonic indexer
+    # GH12713
+
+    base = tsframe.iloc[0:5]
+    r = len(base.index)
+    c = len(base.columns)
+    tso = DataFrame(
+        np.random.default_rng(2).standard_normal((r, c)),
+        index=base.index,
+        columns=base.columns,
+        dtype="float64",
+    )
+    # monotonic
+    ts = tso
+    grouped = ts.groupby(lambda x: x.weekday(), group_keys=False)
+    result = ts - grouped.transform("mean")
+    expected = grouped.apply(lambda x: x - x.mean(axis=0))
+    tm.assert_frame_equal(result, expected)
+
+    # non-monotonic
+    ts = tso.iloc[[1, 0] + list(range(2, len(base)))]
+    grouped = ts.groupby(lambda x: x.weekday(), group_keys=False)
+    result = ts - grouped.transform("mean")
+    expected = grouped.apply(lambda x: x - x.mean(axis=0))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_dtype():
+    # GH 9807
+    # Check transform dtype output is preserved
+    df = DataFrame([[1, 3], [2, 3]])
+    result = df.groupby(1).transform("mean")
+    expected = DataFrame([[1.5], [1.5]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_bug():
+    # GH 5712
+    # transforming on a datetime column
+    df = DataFrame({"A": Timestamp("20130101"), "B": np.arange(5)})
+    result = df.groupby("A")["B"].transform(lambda x: x.rank(ascending=False))
+    expected = Series(np.arange(5, 0, step=-1), name="B", dtype="float64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_numeric_to_boolean():
+    # GH 16875
+    # inconsistency in transforming boolean values
+    expected = Series([True, True], name="A")
+
+    df = DataFrame({"A": [1.1, 2.2], "B": [1, 2]})
+    result = df.groupby("B").A.transform(lambda x: True)
+    tm.assert_series_equal(result, expected)
+
+    df = DataFrame({"A": [1, 2], "B": [1, 2]})
+    result = df.groupby("B").A.transform(lambda x: True)
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_datetime_to_timedelta():
+    # GH 15429
+    # transforming a datetime to timedelta
+    df = DataFrame({"A": Timestamp("20130101"), "B": np.arange(5)})
+    expected = Series(
+        Timestamp("20130101") - Timestamp("20130101"), index=range(5), name="A"
+    )
+
+    # this does date math without changing result type in transform
+    base_time = df["A"][0]
+    result = (
+        df.groupby("A")["A"].transform(lambda x: x.max() - x.min() + base_time)
+        - base_time
+    )
+    tm.assert_series_equal(result, expected)
+
+    # this does date math and causes the transform to return timedelta
+    result = df.groupby("A")["A"].transform(lambda x: x.max() - x.min())
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_datetime_to_numeric():
+    # GH 10972
+    # convert dt to float
+    df = DataFrame({"a": 1, "b": date_range("2015-01-01", periods=2, freq="D")})
+    result = df.groupby("a").b.transform(
+        lambda x: x.dt.dayofweek - x.dt.dayofweek.mean()
+    )
+
+    expected = Series([-0.5, 0.5], name="b")
+    tm.assert_series_equal(result, expected)
+
+    # convert dt to int
+    df = DataFrame({"a": 1, "b": date_range("2015-01-01", periods=2, freq="D")})
+    result = df.groupby("a").b.transform(
+        lambda x: x.dt.dayofweek - x.dt.dayofweek.min()
+    )
+
+    expected = Series([0, 1], dtype=np.int32, name="b")
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_casting():
+    # 13046
+    times = [
+        "13:43:27",
+        "14:26:19",
+        "14:29:01",
+        "18:39:34",
+        "18:40:18",
+        "18:44:30",
+        "18:46:00",
+        "18:52:15",
+        "18:59:59",
+        "19:17:48",
+        "19:21:38",
+    ]
+    df = DataFrame(
+        {
+            "A": [f"B-{i}" for i in range(11)],
+            "ID3": np.take(
+                ["a", "b", "c", "d", "e"], [0, 1, 2, 1, 3, 1, 1, 1, 4, 1, 1]
+            ),
+            "DATETIME": pd.to_datetime([f"2014-10-08 {time}" for time in times]),
+        },
+        index=pd.RangeIndex(11, name="idx"),
+    )
+
+    result = df.groupby("ID3")["DATETIME"].transform(lambda x: x.diff())
+    assert lib.is_np_dtype(result.dtype, "m")
+
+    result = df[["ID3", "DATETIME"]].groupby("ID3").transform(lambda x: x.diff())
+    assert lib.is_np_dtype(result.DATETIME.dtype, "m")
+
+
+def test_transform_multiple(ts):
+    grouped = ts.groupby([lambda x: x.year, lambda x: x.month])
+    grouped.transform(lambda x: x * 2)
+    grouped.transform(np.mean)
+
+
+def test_dispatch_transform(tsframe):
+    df = tsframe[::5].reindex(tsframe.index)
+
+    grouped = df.groupby(lambda x: x.month)
+
+    filled = grouped.ffill()
+    fillit = lambda x: x.ffill()
+    expected = df.groupby(lambda x: x.month).transform(fillit)
+    tm.assert_frame_equal(filled, expected)
+
+
+def test_transform_transformation_func(transformation_func):
+    # GH 30918
+    df = DataFrame(
+        {
+            "A": ["foo", "foo", "foo", "foo", "bar", "bar", "baz"],
+            "B": [1, 2, np.nan, 3, 3, np.nan, 4],
+        },
+        index=date_range("2020-01-01", "2020-01-07"),
+    )
+    if transformation_func == "cumcount":
+        test_op = lambda x: x.transform("cumcount")
+        mock_op = lambda x: Series(range(len(x)), x.index)
+    elif transformation_func == "ngroup":
+        test_op = lambda x: x.transform("ngroup")
+        counter = -1
+
+        def mock_op(x):
+            nonlocal counter
+            counter += 1
+            return Series(counter, index=x.index)
+
+    else:
+        test_op = lambda x: x.transform(transformation_func)
+        mock_op = lambda x: getattr(x, transformation_func)()
+
+    result = test_op(df.groupby("A"))
+
+    # pass the group in same order as iterating `for ... in df.groupby(...)`
+    # but reorder to match df's index since this is a transform
+    groups = [df[["B"]].iloc[4:6], df[["B"]].iloc[6:], df[["B"]].iloc[:4]]
+    expected = concat([mock_op(g) for g in groups]).sort_index()
+    # sort_index does not preserve the freq
+    expected = expected.set_axis(df.index)
+
+    if transformation_func in ("cumcount", "ngroup"):
+        tm.assert_series_equal(result, expected)
+    else:
+        tm.assert_frame_equal(result, expected)
+
+
+def test_transform_select_columns(df):
+    f = lambda x: x.mean()
+    result = df.groupby("A")[["C", "D"]].transform(f)
+
+    selection = df[["C", "D"]]
+    expected = selection.groupby(df["A"]).transform(f)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_nuisance_raises(df, using_infer_string):
+    # case that goes through _transform_item_by_item
+
+    df.columns = ["A", "B", "B", "D"]
+
+    # this also tests orderings in transform between
+    # series/frame to make sure it's consistent
+    grouped = df.groupby("A")
+
+    gbc = grouped["B"]
+    msg = "Could not convert"
+    if using_infer_string:
+        msg = "Cannot perform reduction 'mean' with string dtype"
+    with pytest.raises(TypeError, match=msg):
+        gbc.transform(lambda x: np.mean(x))
+
+    with pytest.raises(TypeError, match=msg):
+        df.groupby("A").transform(lambda x: np.mean(x))
+
+
+def test_transform_function_aliases(df):
+    result = df.groupby("A").transform("mean", numeric_only=True)
+    expected = df.groupby("A")[["C", "D"]].transform(np.mean)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby("A")["C"].transform("mean")
+    expected = df.groupby("A")["C"].transform(np.mean)
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_fast_transform_date():
+    # GH 13191
+    df = DataFrame(
+        {"grouping": [np.nan, 1, 1, 3], "d": date_range("2014-1-1", "2014-1-4")}
+    )
+    result = df.groupby("grouping")["d"].transform("first")
+    dates = [
+        pd.NaT,
+        Timestamp("2014-1-2"),
+        Timestamp("2014-1-2"),
+        Timestamp("2014-1-4"),
+    ]
+    expected = Series(dates, name="d", dtype="M8[ns]")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", [lambda x: np.nansum(x), sum])
+def test_transform_length(func):
+    # GH 9697
+    df = DataFrame({"col1": [1, 1, 2, 2], "col2": [1, 2, 3, np.nan]})
+    if func is sum:
+        expected = Series([3.0, 3.0, np.nan, np.nan])
+    else:
+        expected = Series([3.0] * 4)
+
+    results = [
+        df.groupby("col1").transform(func)["col2"],
+        df.groupby("col1")["col2"].transform(func),
+    ]
+    for result in results:
+        tm.assert_series_equal(result, expected, check_names=False)
+
+
+def test_transform_coercion():
+    # 14457
+    # when we are transforming be sure to not coerce
+    # via assignment
+    df = DataFrame({"A": ["a", "a", "b", "b"], "B": [0, 1, 3, 4]})
+    g = df.groupby("A")
+
+    expected = g.transform(np.mean)
+    result = g.transform(lambda x: np.mean(x, axis=0))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_transform_with_int(using_infer_string):
+    # GH 3740, make sure that we might upcast on item-by-item transform
+
+    # floats
+    df = DataFrame(
+        {
+            "A": [1, 1, 1, 2, 2, 2],
+            "B": Series(1, dtype="float64"),
+            "C": Series([1, 2, 3, 1, 2, 3], dtype="float64"),
+            "D": "foo",
+        }
+    )
+    with np.errstate(all="ignore"):
+        result = df.groupby("A")[["B", "C"]].transform(
+            lambda x: (x - x.mean()) / x.std()
+        )
+    expected = DataFrame(
+        {"B": np.nan, "C": Series([-1, 0, 1, -1, 0, 1], dtype="float64")}
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # int case
+    df = DataFrame(
+        {
+            "A": [1, 1, 1, 2, 2, 2],
+            "B": 1,
+            "C": [1, 2, 3, 1, 2, 3],
+            "D": "foo",
+        }
+    )
+    msg = "Could not convert"
+    if using_infer_string:
+        msg = "Cannot perform reduction 'mean' with string dtype"
+    with np.errstate(all="ignore"):
+        with pytest.raises(TypeError, match=msg):
+            df.groupby("A").transform(lambda x: (x - x.mean()) / x.std())
+        result = df.groupby("A")[["B", "C"]].transform(
+            lambda x: (x - x.mean()) / x.std()
+        )
+    expected = DataFrame({"B": np.nan, "C": [-1.0, 0.0, 1.0, -1.0, 0.0, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+    # int that needs float conversion
+    s = Series([2, 3, 4, 10, 5, -1])
+    df = DataFrame({"A": [1, 1, 1, 2, 2, 2], "B": 1, "C": s, "D": "foo"})
+    with np.errstate(all="ignore"):
+        with pytest.raises(TypeError, match=msg):
+            df.groupby("A").transform(lambda x: (x - x.mean()) / x.std())
+        result = df.groupby("A")[["B", "C"]].transform(
+            lambda x: (x - x.mean()) / x.std()
+        )
+
+    s1 = s.iloc[0:3]
+    s1 = (s1 - s1.mean()) / s1.std()
+    s2 = s.iloc[3:6]
+    s2 = (s2 - s2.mean()) / s2.std()
+    expected = DataFrame({"B": np.nan, "C": concat([s1, s2])})
+    tm.assert_frame_equal(result, expected)
+
+    # int doesn't get downcasted
+    result = df.groupby("A")[["B", "C"]].transform(lambda x: x * 2 / 2)
+    expected = DataFrame({"B": 1.0, "C": [2.0, 3.0, 4.0, 10.0, 5.0, -1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_transform_with_nan_group():
+    # GH 9941
+    df = DataFrame({"a": range(10), "b": [1, 1, 2, 3, np.nan, 4, 4, 5, 5, 5]})
+    result = df.groupby(df.b)["a"].transform(max)
+    expected = Series([1.0, 1.0, 2.0, 3.0, np.nan, 6.0, 6.0, 9.0, 9.0, 9.0], name="a")
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_mixed_type():
+    index = MultiIndex.from_arrays([[0, 0, 0, 1, 1, 1], [1, 2, 3, 1, 2, 3]])
+    df = DataFrame(
+        {
+            "d": [1.0, 1.0, 1.0, 2.0, 2.0, 2.0],
+            "c": np.tile(["a", "b", "c"], 2),
+            "v": np.arange(1.0, 7.0),
+        },
+        index=index,
+    )
+
+    def f(group):
+        group["g"] = group["d"] * 2
+        return group[:1]
+
+    grouped = df.groupby("c")
+    result = grouped.apply(f)
+
+    assert result["d"].dtype == np.float64
+
+    # this is by definition a mutating operation!
+    for key, group in grouped:
+        res = f(group.drop(columns="c"))
+        tm.assert_frame_equal(res, result.loc[key])
+
+
+@pytest.mark.parametrize(
+    "op, args, targop",
+    [
+        ("cumprod", (), lambda x: x.cumprod()),
+        ("cumsum", (), lambda x: x.cumsum()),
+        ("shift", (-1,), lambda x: x.shift(-1)),
+        ("shift", (1,), lambda x: x.shift()),
+    ],
+)
+def test_cython_transform_series(op, args, targop):
+    # GH 4095
+    s = Series(np.random.default_rng(2).standard_normal(1000))
+    s_missing = s.copy()
+    s_missing.iloc[2:10] = np.nan
+    labels = np.random.default_rng(2).integers(0, 50, size=1000).astype(float)
+
+    # series
+    for data in [s, s_missing]:
+        # print(data.head())
+        expected = data.groupby(labels).transform(targop)
+
+        tm.assert_series_equal(expected, data.groupby(labels).transform(op, *args))
+        tm.assert_series_equal(expected, getattr(data.groupby(labels), op)(*args))
+
+
+@pytest.mark.parametrize("op", ["cumprod", "cumsum"])
+@pytest.mark.parametrize(
+    "input, exp",
+    [
+        # When everything is NaN
+        ({"key": ["b"] * 10, "value": np.nan}, Series([np.nan] * 10, name="value")),
+        # When there is a single NaN
+        (
+            {"key": ["b"] * 10 + ["a"] * 2, "value": [3] * 3 + [np.nan] + [3] * 8},
+            {
+                ("cumprod", False): [3.0, 9.0, 27.0] + [np.nan] * 7 + [3.0, 9.0],
+                ("cumprod", True): [
+                    3.0,
+                    9.0,
+                    27.0,
+                    np.nan,
+                    81.0,
+                    243.0,
+                    729.0,
+                    2187.0,
+                    6561.0,
+                    19683.0,
+                    3.0,
+                    9.0,
+                ],
+                ("cumsum", False): [3.0, 6.0, 9.0] + [np.nan] * 7 + [3.0, 6.0],
+                ("cumsum", True): [
+                    3.0,
+                    6.0,
+                    9.0,
+                    np.nan,
+                    12.0,
+                    15.0,
+                    18.0,
+                    21.0,
+                    24.0,
+                    27.0,
+                    3.0,
+                    6.0,
+                ],
+            },
+        ),
+    ],
+)
+def test_groupby_cum_skipna(op, skipna, input, exp):
+    df = DataFrame(input)
+    result = df.groupby("key")["value"].transform(op, skipna=skipna)
+    if isinstance(exp, dict):
+        expected = exp[(op, skipna)]
+    else:
+        expected = exp
+    expected = Series(expected, name="value")
+    tm.assert_series_equal(expected, result)
+
+
+@pytest.fixture
+def frame():
+    floating = Series(np.random.default_rng(2).standard_normal(10))
+    floating_missing = floating.copy()
+    floating_missing.iloc[2:7] = np.nan
+    strings = list("abcde") * 2
+    strings_missing = strings[:]
+    strings_missing[5] = np.nan
+
+    df = DataFrame(
+        {
+            "float": floating,
+            "float_missing": floating_missing,
+            "int": [1, 1, 1, 1, 2] * 2,
+            "datetime": date_range("1990-1-1", periods=10),
+            "timedelta": pd.timedelta_range(1, freq="s", periods=10),
+            "string": strings,
+            "string_missing": strings_missing,
+            "cat": Categorical(strings),
+        },
+    )
+    return df
+
+
+@pytest.fixture
+def frame_mi(frame):
+    frame.index = MultiIndex.from_product([range(5), range(2)])
+    return frame
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize(
+    "op, args, targop",
+    [
+        ("cumprod", (), lambda x: x.cumprod()),
+        ("cumsum", (), lambda x: x.cumsum()),
+        ("shift", (-1,), lambda x: x.shift(-1)),
+        ("shift", (1,), lambda x: x.shift()),
+    ],
+)
+@pytest.mark.parametrize("df_fix", ["frame", "frame_mi"])
+@pytest.mark.parametrize(
+    "gb_target",
+    [
+        {"by": np.random.default_rng(2).integers(0, 50, size=10).astype(float)},
+        {"level": 0},
+        {"by": "string"},
+        pytest.param({"by": "string_missing"}, marks=pytest.mark.xfail),
+        {"by": ["int", "string"]},
+    ],
+)
+def test_cython_transform_frame(request, op, args, targop, df_fix, gb_target):
+    df = request.getfixturevalue(df_fix)
+    gb = df.groupby(group_keys=False, **gb_target)
+
+    if op != "shift" and "int" not in gb_target:
+        # numeric apply fastpath promotes dtype so have
+        # to apply separately and concat
+        i = gb[["int"]].apply(targop)
+        f = gb[["float", "float_missing"]].apply(targop)
+        expected = concat([f, i], axis=1)
+    else:
+        expected = gb.apply(targop)
+
+    expected = expected.sort_index(axis=1)
+    if op == "shift":
+        expected["string_missing"] = expected["string_missing"].fillna(np.nan)
+        by = gb_target.get("by")
+        if not isinstance(by, (str, list)) or (by != "string" and "string" not in by):
+            expected["string"] = expected["string"].fillna(np.nan)
+
+    result = gb[expected.columns].transform(op, *args).sort_index(axis=1)
+    tm.assert_frame_equal(result, expected)
+    result = getattr(gb[expected.columns], op)(*args).sort_index(axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize(
+    "op, args, targop",
+    [
+        ("cumprod", (), lambda x: x.cumprod()),
+        ("cumsum", (), lambda x: x.cumsum()),
+        ("shift", (-1,), lambda x: x.shift(-1)),
+        ("shift", (1,), lambda x: x.shift()),
+    ],
+)
+@pytest.mark.parametrize("df_fix", ["frame", "frame_mi"])
+@pytest.mark.parametrize(
+    "gb_target",
+    [
+        {"by": np.random.default_rng(2).integers(0, 50, size=10).astype(float)},
+        {"level": 0},
+        {"by": "string"},
+        # TODO: create xfail condition given other params
+        # {"by": 'string_missing'},
+        {"by": ["int", "string"]},
+    ],
+)
+@pytest.mark.parametrize(
+    "column",
+    [
+        "float",
+        "float_missing",
+        "int",
+        "datetime",
+        "timedelta",
+        "string",
+        "string_missing",
+    ],
+)
+def test_cython_transform_frame_column(
+    request, op, args, targop, df_fix, gb_target, column
+):
+    df = request.getfixturevalue(df_fix)
+    gb = df.groupby(group_keys=False, **gb_target)
+    c = column
+    if (
+        c not in ["float", "int", "float_missing"]
+        and op != "shift"
+        and not (c == "timedelta" and op == "cumsum")
+    ):
+        msg = "|".join(
+            [
+                "does not support .* operations",
+                "does not support operation",
+                ".* is not supported for object dtype",
+                "is not implemented for this dtype",
+                ".* is not supported for str dtype",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            gb[c].transform(op)
+        with pytest.raises(TypeError, match=msg):
+            getattr(gb[c], op)()
+    else:
+        expected = gb[c].apply(targop)
+        expected.name = c
+        if c in ["string_missing", "string"]:
+            expected = expected.fillna(np.nan)
+
+        res = gb[c].transform(op, *args)
+        tm.assert_series_equal(expected, res)
+        res2 = getattr(gb[c], op)(*args)
+        tm.assert_series_equal(expected, res2)
+
+
+@pytest.mark.parametrize(
+    "cols,expected",
+    [
+        ("a", Series([1, 1, 1], name="a")),
+        (
+            ["a", "c"],
+            DataFrame({"a": [1, 1, 1], "c": [1, 1, 1]}),
+        ),
+    ],
+)
+@pytest.mark.parametrize("agg_func", ["count", "rank", "size"])
+def test_transform_numeric_ret(cols, expected, agg_func):
+    # GH#19200 and GH#27469
+    df = DataFrame(
+        {"a": date_range("2018-01-01", periods=3), "b": range(3), "c": range(7, 10)}
+    )
+    result = df.groupby("b")[cols].transform(agg_func)
+
+    if agg_func == "rank":
+        expected = expected.astype("float")
+    elif agg_func == "size" and cols == ["a", "c"]:
+        # transform("size") returns a Series
+        expected = expected["a"].rename(None)
+    tm.assert_equal(result, expected)
+
+
+def test_transform_ffill():
+    # GH 24211
+    data = [["a", 0.0], ["a", float("nan")], ["b", 1.0], ["b", float("nan")]]
+    df = DataFrame(data, columns=["key", "values"])
+    result = df.groupby("key").transform("ffill")
+    expected = DataFrame({"values": [0.0, 0.0, 1.0, 1.0]})
+    tm.assert_frame_equal(result, expected)
+    result = df.groupby("key")["values"].transform("ffill")
+    expected = Series([0.0, 0.0, 1.0, 1.0], name="values")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("mix_groupings", [True, False])
+@pytest.mark.parametrize("as_series", [True, False])
+@pytest.mark.parametrize("val1,val2", [("foo", "bar"), (1, 2), (1.0, 2.0)])
+@pytest.mark.parametrize(
+    "fill_method,limit,exp_vals",
+    [
+        (
+            "ffill",
+            None,
+            [np.nan, np.nan, "val1", "val1", "val1", "val2", "val2", "val2"],
+        ),
+        ("ffill", 1, [np.nan, np.nan, "val1", "val1", np.nan, "val2", "val2", np.nan]),
+        (
+            "bfill",
+            None,
+            ["val1", "val1", "val1", "val2", "val2", "val2", np.nan, np.nan],
+        ),
+        ("bfill", 1, [np.nan, "val1", "val1", np.nan, "val2", "val2", np.nan, np.nan]),
+    ],
+)
+def test_group_fill_methods(
+    mix_groupings, as_series, val1, val2, fill_method, limit, exp_vals
+):
+    vals = [np.nan, np.nan, val1, np.nan, np.nan, val2, np.nan, np.nan]
+    _exp_vals = list(exp_vals)
+    # Overwrite placeholder values
+    for index, exp_val in enumerate(_exp_vals):
+        if exp_val == "val1":
+            _exp_vals[index] = val1
+        elif exp_val == "val2":
+            _exp_vals[index] = val2
+
+    # Need to modify values and expectations depending on the
+    # Series / DataFrame that we ultimately want to generate
+    if mix_groupings:  # ['a', 'b', 'a, 'b', ...]
+        keys = ["a", "b"] * len(vals)
+
+        def interweave(list_obj):
+            temp = []
+            for x in list_obj:
+                temp.extend([x, x])
+
+            return temp
+
+        _exp_vals = interweave(_exp_vals)
+        vals = interweave(vals)
+    else:  # ['a', 'a', 'a', ... 'b', 'b', 'b']
+        keys = ["a"] * len(vals) + ["b"] * len(vals)
+        _exp_vals = _exp_vals * 2
+        vals = vals * 2
+
+    df = DataFrame({"key": keys, "val": vals})
+    if as_series:
+        result = getattr(df.groupby("key")["val"], fill_method)(limit=limit)
+        exp = Series(_exp_vals, name="val")
+        tm.assert_series_equal(result, exp)
+    else:
+        result = getattr(df.groupby("key"), fill_method)(limit=limit)
+        exp = DataFrame({"val": _exp_vals})
+        tm.assert_frame_equal(result, exp)
+
+
+@pytest.mark.parametrize("fill_method", ["ffill", "bfill"])
+def test_pad_stable_sorting(fill_method):
+    # GH 21207
+    x = [0] * 20
+    y = [np.nan] * 10 + [1] * 10
+
+    if fill_method == "bfill":
+        y = y[::-1]
+
+    df = DataFrame({"x": x, "y": y})
+    expected = df.drop("x", axis=1)
+
+    result = getattr(df.groupby("x"), fill_method)()
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "freq",
+    [
+        None,
+        pytest.param(
+            "D",
+            marks=pytest.mark.xfail(
+                reason="GH#23918 before method uses freq in vectorized approach"
+            ),
+        ),
+    ],
+)
+@pytest.mark.parametrize("periods", [1, -1])
+def test_pct_change(frame_or_series, freq, periods):
+    # GH 21200, 21621, 30463
+    vals = [3, np.nan, np.nan, np.nan, 1, 2, 4, 10, np.nan, 4]
+    keys = ["a", "b"]
+    key_v = np.repeat(keys, len(vals))
+    df = DataFrame({"key": key_v, "vals": vals * 2})
+
+    df_g = df
+    grp = df_g.groupby(df.key)
+
+    expected = grp["vals"].obj / grp["vals"].shift(periods) - 1
+
+    gb = df.groupby("key")
+
+    if frame_or_series is Series:
+        gb = gb["vals"]
+    else:
+        expected = expected.to_frame("vals")
+
+    result = gb.pct_change(periods=periods, freq=freq)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, expected_status",
+    [
+        ("ffill", ["shrt", "shrt", "lng", np.nan, "shrt", "ntrl", "ntrl"]),
+        ("bfill", ["shrt", "lng", "lng", "shrt", "shrt", "ntrl", np.nan]),
+    ],
+)
+def test_ffill_bfill_non_unique_multilevel(func, expected_status):
+    # GH 19437
+    date = pd.to_datetime(
+        [
+            "2018-01-01",
+            "2018-01-01",
+            "2018-01-01",
+            "2018-01-01",
+            "2018-01-02",
+            "2018-01-01",
+            "2018-01-02",
+        ]
+    )
+    symbol = ["MSFT", "MSFT", "MSFT", "AAPL", "AAPL", "TSLA", "TSLA"]
+    status = ["shrt", np.nan, "lng", np.nan, "shrt", "ntrl", np.nan]
+
+    df = DataFrame({"date": date, "symbol": symbol, "status": status})
+    df = df.set_index(["date", "symbol"])
+    result = getattr(df.groupby("symbol")["status"], func)()
+
+    index = MultiIndex.from_tuples(
+        tuples=list(zip(*[date, symbol])), names=["date", "symbol"]
+    )
+    expected = Series(expected_status, index=index, name="status")
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", [np.any, np.all])
+def test_any_all_np_func(func):
+    # GH 20653
+    df = DataFrame(
+        [["foo", True], [np.nan, True], ["foo", True]], columns=["key", "val"]
+    )
+
+    exp = Series([True, np.nan, True], name="val")
+
+    res = df.groupby("key")["val"].transform(func)
+    tm.assert_series_equal(res, exp)
+
+
+def test_groupby_transform_rename():
+    # https://github.com/pandas-dev/pandas/issues/23461
+    def demean_rename(x):
+        result = x - x.mean()
+
+        if isinstance(x, Series):
+            return result
+
+        result = result.rename(columns={c: f"{c}_demeaned" for c in result.columns})
+
+        return result
+
+    df = DataFrame({"group": list("ababa"), "value": [1, 1, 1, 2, 2]})
+    expected = DataFrame({"value": [-1.0 / 3, -0.5, -1.0 / 3, 0.5, 2.0 / 3]})
+
+    result = df.groupby("group").transform(demean_rename)
+    tm.assert_frame_equal(result, expected)
+    result_single = df.groupby("group").value.transform(demean_rename)
+    tm.assert_series_equal(result_single, expected["value"])
+
+
+@pytest.mark.parametrize("func", [min, max, np.min, np.max, "first", "last"])
+def test_groupby_transform_timezone_column(func):
+    # GH 24198
+    ts = pd.to_datetime("now", utc=True).tz_convert("Asia/Singapore")
+    result = DataFrame({"end_time": [ts], "id": [1]})
+    result["max_end_time"] = result.groupby("id").end_time.transform(func)
+    expected = DataFrame([[ts, 1, ts]], columns=["end_time", "id", "max_end_time"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, values",
+    [
+        ("idxmin", ["1/1/2011"] * 2 + ["1/3/2011"] * 7 + ["1/10/2011"]),
+        ("idxmax", ["1/2/2011"] * 2 + ["1/9/2011"] * 7 + ["1/10/2011"]),
+    ],
+)
+def test_groupby_transform_with_datetimes(func, values):
+    # GH 15306
+    dates = date_range("1/1/2011", periods=10, freq="D")
+
+    stocks = DataFrame({"price": np.arange(10.0)}, index=dates)
+    stocks["week_id"] = dates.isocalendar().week
+
+    result = stocks.groupby(stocks["week_id"])["price"].transform(func)
+
+    expected = Series(
+        data=pd.to_datetime(values).as_unit("ns"), index=dates, name="price"
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_transform_dtype():
+    # GH 22243
+    df = DataFrame({"a": [1], "val": [1.35]})
+
+    result = df["val"].transform(lambda x: x.map(lambda y: f"+{y}"))
+    expected1 = Series(["+1.35"], name="val")
+    tm.assert_series_equal(result, expected1)
+
+    result = df.groupby("a")["val"].transform(lambda x: x.map(lambda y: f"+{y}"))
+    tm.assert_series_equal(result, expected1)
+
+    result = df.groupby("a")["val"].transform(lambda x: x.map(lambda y: f"+({y})"))
+    expected2 = Series(["+(1.35)"], name="val")
+    tm.assert_series_equal(result, expected2)
+
+    df["val"] = df["val"].astype(object)
+    result = df.groupby("a")["val"].transform(lambda x: x.map(lambda y: f"+{y}"))
+    tm.assert_series_equal(result, expected1)
+
+
+def test_transform_absent_categories(all_numeric_accumulations):
+    # GH 16771
+    # cython transforms with more groups than rows
+    x_vals = [1]
+    x_cats = range(2)
+    y = [1]
+    df = DataFrame({"x": Categorical(x_vals, x_cats), "y": y})
+    result = getattr(df.y.groupby(df.x, observed=False), all_numeric_accumulations)()
+    expected = df.y
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["ffill", "bfill", "shift"])
+@pytest.mark.parametrize("key, val", [("level", 0), ("by", Series([0]))])
+def test_ffill_not_in_axis(func, key, val):
+    # GH 21521
+    df = DataFrame([[np.nan]])
+    result = getattr(df.groupby(**{key: val}), func)()
+    expected = df
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_invalid_name_raises():
+    # GH#27486
+    df = DataFrame({"a": [0, 1, 1, 2]})
+    g = df.groupby(["a", "b", "b", "c"])
+    with pytest.raises(ValueError, match="not a valid function name"):
+        g.transform("some_arbitrary_name")
+
+    # method exists on the object, but is not a valid transformation/agg
+    assert hasattr(g, "aggregate")  # make sure the method exists
+    with pytest.raises(ValueError, match="not a valid function name"):
+        g.transform("aggregate")
+
+    # Test SeriesGroupBy
+    g = df["a"].groupby(["a", "b", "b", "c"])
+    with pytest.raises(ValueError, match="not a valid function name"):
+        g.transform("some_arbitrary_name")
+
+
+def test_transform_agg_by_name(request, reduction_func, frame_or_series):
+    func = reduction_func
+
+    obj = DataFrame(
+        {"a": [0, 0, 0, 0, 1, 1, 1, 1], "b": range(8)},
+        index=["A", "B", "C", "D", "E", "F", "G", "H"],
+    )
+    if frame_or_series is Series:
+        obj = obj["a"]
+
+    g = obj.groupby(np.repeat([0, 1], 4))
+
+    if func == "corrwith" and isinstance(obj, Series):  # GH#32293
+        # TODO: implement SeriesGroupBy.corrwith
+        assert not hasattr(g, func)
+        return
+
+    args = get_groupby_method_args(reduction_func, obj)
+    if func == "corrwith":
+        warn = FutureWarning
+        msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        msg = ""
+    with tm.assert_produces_warning(warn, match=msg):
+        result = g.transform(func, *args)
+
+    # this is the *definition* of a transformation
+    tm.assert_index_equal(result.index, obj.index)
+
+    if func not in ("ngroup", "size") and obj.ndim == 2:
+        # size/ngroup return a Series, unlike other transforms
+        tm.assert_index_equal(result.columns, obj.columns)
+
+    # verify that values were broadcasted across each group
+    assert len(set(DataFrame(result).iloc[-4:, -1])) == 1
+
+
+def test_transform_lambda_with_datetimetz():
+    # GH 27496
+    df = DataFrame(
+        {
+            "time": [
+                Timestamp("2010-07-15 03:14:45"),
+                Timestamp("2010-11-19 18:47:06"),
+            ],
+            "timezone": ["Etc/GMT+4", "US/Eastern"],
+        }
+    )
+    result = df.groupby(["timezone"])["time"].transform(
+        lambda x: x.dt.tz_localize(x.name)
+    )
+    expected = Series(
+        [
+            Timestamp("2010-07-15 03:14:45", tz="Etc/GMT+4"),
+            Timestamp("2010-11-19 18:47:06", tz="US/Eastern"),
+        ],
+        name="time",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_fastpath_raises():
+    # GH#29631 case where fastpath defined in groupby.generic _choose_path
+    #  raises, but slow_path does not
+
+    df = DataFrame({"A": [1, 1, 2, 2], "B": [1, -1, 1, 2]})
+    gb = df.groupby("A")
+
+    def func(grp):
+        # we want a function such that func(frame) fails but func.apply(frame)
+        #  works
+        if grp.ndim == 2:
+            # Ensure that fast_path fails
+            raise NotImplementedError("Don't cross the streams")
+        return grp * 2
+
+    # Check that the fastpath raises, see _transform_general
+    obj = gb._obj_with_exclusions
+    gen = gb._grouper.get_iterator(obj)
+    fast_path, slow_path = gb._define_paths(func)
+    _, group = next(gen)
+
+    with pytest.raises(NotImplementedError, match="Don't cross the streams"):
+        fast_path(group)
+
+    result = gb.transform(func)
+
+    expected = DataFrame([2, -2, 2, 4], columns=["B"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_lambda_indexing():
+    # GH 7883
+    df = DataFrame(
+        {
+            "A": ["foo", "bar", "foo", "bar", "foo", "flux", "foo", "flux"],
+            "B": ["one", "one", "two", "three", "two", "six", "five", "three"],
+            "C": range(8),
+            "D": range(8),
+            "E": range(8),
+        }
+    )
+    df = df.set_index(["A", "B"])
+    df = df.sort_index()
+    result = df.groupby(level="A").transform(lambda x: x.iloc[-1])
+    expected = DataFrame(
+        {
+            "C": [3, 3, 7, 7, 4, 4, 4, 4],
+            "D": [3, 3, 7, 7, 4, 4, 4, 4],
+            "E": [3, 3, 7, 7, 4, 4, 4, 4],
+        },
+        index=MultiIndex.from_tuples(
+            [
+                ("bar", "one"),
+                ("bar", "three"),
+                ("flux", "six"),
+                ("flux", "three"),
+                ("foo", "five"),
+                ("foo", "one"),
+                ("foo", "two"),
+                ("foo", "two"),
+            ],
+            names=["A", "B"],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_and_not_categorical_key(observed):
+    # Checks that groupby-transform, when grouping by both a categorical
+    # and a non-categorical key, doesn't try to expand the output to include
+    # non-observed categories but instead matches the input shape.
+    # GH 32494
+    df_with_categorical = DataFrame(
+        {
+            "A": Categorical(["a", "b", "a"], categories=["a", "b", "c"]),
+            "B": [1, 2, 3],
+            "C": ["a", "b", "a"],
+        }
+    )
+    df_without_categorical = DataFrame(
+        {"A": ["a", "b", "a"], "B": [1, 2, 3], "C": ["a", "b", "a"]}
+    )
+
+    # DataFrame case
+    result = df_with_categorical.groupby(["A", "C"], observed=observed).transform("sum")
+    expected = df_without_categorical.groupby(["A", "C"]).transform("sum")
+    tm.assert_frame_equal(result, expected)
+    expected_explicit = DataFrame({"B": [4, 2, 4]})
+    tm.assert_frame_equal(result, expected_explicit)
+
+    # Series case
+    gb = df_with_categorical.groupby(["A", "C"], observed=observed)
+    gbp = gb["B"]
+    result = gbp.transform("sum")
+    expected = df_without_categorical.groupby(["A", "C"])["B"].transform("sum")
+    tm.assert_series_equal(result, expected)
+    expected_explicit = Series([4, 2, 4], name="B")
+    tm.assert_series_equal(result, expected_explicit)
+
+
+def test_string_rank_grouping():
+    # GH 19354
+    df = DataFrame({"A": [1, 1, 2], "B": [1, 2, 3]})
+    result = df.groupby("A").transform("rank")
+    expected = DataFrame({"B": [1.0, 2.0, 1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_cumcount():
+    # GH 27472
+    df = DataFrame({"a": [0, 0, 0, 1, 1, 1], "b": range(6)})
+    grp = df.groupby(np.repeat([0, 1], 3))
+
+    result = grp.cumcount()
+    expected = Series([0, 1, 2, 0, 1, 2])
+    tm.assert_series_equal(result, expected)
+
+    result = grp.transform("cumcount")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", [["A1"], ["A1", "A2"]])
+def test_null_group_lambda_self(sort, dropna, keys):
+    # GH 17093
+    size = 50
+    nulls1 = np.random.default_rng(2).choice([False, True], size)
+    nulls2 = np.random.default_rng(2).choice([False, True], size)
+    # Whether a group contains a null value or not
+    nulls_grouper = nulls1 if len(keys) == 1 else nulls1 | nulls2
+
+    a1 = np.random.default_rng(2).integers(0, 5, size=size).astype(float)
+    a1[nulls1] = np.nan
+    a2 = np.random.default_rng(2).integers(0, 5, size=size).astype(float)
+    a2[nulls2] = np.nan
+    values = np.random.default_rng(2).integers(0, 5, size=a1.shape)
+    df = DataFrame({"A1": a1, "A2": a2, "B": values})
+
+    expected_values = values
+    if dropna and nulls_grouper.any():
+        expected_values = expected_values.astype(float)
+        expected_values[nulls_grouper] = np.nan
+    expected = DataFrame(expected_values, columns=["B"])
+
+    gb = df.groupby(keys, dropna=dropna, sort=sort)
+    result = gb[["B"]].transform(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_null_group_str_reducer(request, dropna, reduction_func):
+    # GH 17093
+    if reduction_func == "corrwith":
+        msg = "incorrectly raises"
+        request.applymarker(pytest.mark.xfail(reason=msg))
+
+    index = [1, 2, 3, 4]  # test transform preserves non-standard index
+    df = DataFrame({"A": [1, 1, np.nan, np.nan], "B": [1, 2, 2, 3]}, index=index)
+    gb = df.groupby("A", dropna=dropna)
+
+    args = get_groupby_method_args(reduction_func, df)
+
+    # Manually handle reducers that don't fit the generic pattern
+    # Set expected with dropna=False, then replace if necessary
+    if reduction_func == "first":
+        expected = DataFrame({"B": [1, 1, 2, 2]}, index=index)
+    elif reduction_func == "last":
+        expected = DataFrame({"B": [2, 2, 3, 3]}, index=index)
+    elif reduction_func == "nth":
+        expected = DataFrame({"B": [1, 1, 2, 2]}, index=index)
+    elif reduction_func == "size":
+        expected = Series([2, 2, 2, 2], index=index)
+    elif reduction_func == "corrwith":
+        expected = DataFrame({"B": [1.0, 1.0, 1.0, 1.0]}, index=index)
+    else:
+        expected_gb = df.groupby("A", dropna=False)
+        buffer = []
+        for idx, group in expected_gb:
+            res = getattr(group["B"], reduction_func)()
+            buffer.append(Series(res, index=group.index))
+        expected = concat(buffer).to_frame("B")
+    if dropna:
+        dtype = object if reduction_func in ("any", "all") else float
+        expected = expected.astype(dtype)
+        if expected.ndim == 2:
+            expected.iloc[[2, 3], 0] = np.nan
+        else:
+            expected.iloc[[2, 3]] = np.nan
+
+    result = gb.transform(reduction_func, *args)
+    tm.assert_equal(result, expected)
+
+
+def test_null_group_str_transformer(dropna, transformation_func):
+    # GH 17093
+    df = DataFrame({"A": [1, 1, np.nan], "B": [1, 2, 2]}, index=[1, 2, 3])
+    args = get_groupby_method_args(transformation_func, df)
+    gb = df.groupby("A", dropna=dropna)
+
+    buffer = []
+    for k, (idx, group) in enumerate(gb):
+        if transformation_func == "cumcount":
+            # DataFrame has no cumcount method
+            res = DataFrame({"B": range(len(group))}, index=group.index)
+        elif transformation_func == "ngroup":
+            res = DataFrame(len(group) * [k], index=group.index, columns=["B"])
+        else:
+            res = getattr(group[["B"]], transformation_func)(*args)
+        buffer.append(res)
+    if dropna:
+        dtype = object if transformation_func in ("any", "all") else None
+        buffer.append(DataFrame([[np.nan]], index=[3], dtype=dtype, columns=["B"]))
+    expected = concat(buffer)
+
+    if transformation_func in ("cumcount", "ngroup"):
+        # ngroup/cumcount always returns a Series as it counts the groups, not values
+        expected = expected["B"].rename(None)
+
+    result = gb.transform(transformation_func, *args)
+    tm.assert_equal(result, expected)
+
+
+def test_null_group_str_reducer_series(request, dropna, reduction_func):
+    # GH 17093
+    index = [1, 2, 3, 4]  # test transform preserves non-standard index
+    ser = Series([1, 2, 2, 3], index=index)
+    gb = ser.groupby([1, 1, np.nan, np.nan], dropna=dropna)
+
+    if reduction_func == "corrwith":
+        # corrwith not implemented for SeriesGroupBy
+        assert not hasattr(gb, reduction_func)
+        return
+
+    args = get_groupby_method_args(reduction_func, ser)
+
+    # Manually handle reducers that don't fit the generic pattern
+    # Set expected with dropna=False, then replace if necessary
+    if reduction_func == "first":
+        expected = Series([1, 1, 2, 2], index=index)
+    elif reduction_func == "last":
+        expected = Series([2, 2, 3, 3], index=index)
+    elif reduction_func == "nth":
+        expected = Series([1, 1, 2, 2], index=index)
+    elif reduction_func == "size":
+        expected = Series([2, 2, 2, 2], index=index)
+    elif reduction_func == "corrwith":
+        expected = Series([1, 1, 2, 2], index=index)
+    else:
+        expected_gb = ser.groupby([1, 1, np.nan, np.nan], dropna=False)
+        buffer = []
+        for idx, group in expected_gb:
+            res = getattr(group, reduction_func)()
+            buffer.append(Series(res, index=group.index))
+        expected = concat(buffer)
+    if dropna:
+        dtype = object if reduction_func in ("any", "all") else float
+        expected = expected.astype(dtype)
+        expected.iloc[[2, 3]] = np.nan
+
+    result = gb.transform(reduction_func, *args)
+    tm.assert_series_equal(result, expected)
+
+
+def test_null_group_str_transformer_series(dropna, transformation_func):
+    # GH 17093
+    ser = Series([1, 2, 2], index=[1, 2, 3])
+    args = get_groupby_method_args(transformation_func, ser)
+    gb = ser.groupby([1, 1, np.nan], dropna=dropna)
+
+    buffer = []
+    for k, (idx, group) in enumerate(gb):
+        if transformation_func == "cumcount":
+            # Series has no cumcount method
+            res = Series(range(len(group)), index=group.index)
+        elif transformation_func == "ngroup":
+            res = Series(k, index=group.index)
+        else:
+            res = getattr(group, transformation_func)(*args)
+        buffer.append(res)
+    if dropna:
+        dtype = object if transformation_func in ("any", "all") else None
+        buffer.append(Series([np.nan], index=[3], dtype=dtype))
+    expected = concat(buffer)
+    result = gb.transform(transformation_func, *args)
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, expected_values",
+    [
+        (Series.sort_values, [5, 4, 3, 2, 1]),
+        (lambda x: x.head(1), [5.0, np.nan, 3, 2, np.nan]),
+    ],
+)
+@pytest.mark.parametrize("keys", [["a1"], ["a1", "a2"]])
+@pytest.mark.parametrize("keys_in_index", [True, False])
+def test_transform_aligns(func, frame_or_series, expected_values, keys, keys_in_index):
+    # GH#45648 - transform should align with the input's index
+    df = DataFrame({"a1": [1, 1, 3, 2, 2], "b": [5, 4, 3, 2, 1]})
+    if "a2" in keys:
+        df["a2"] = df["a1"]
+    if keys_in_index:
+        df = df.set_index(keys, append=True)
+
+    gb = df.groupby(keys)
+    if frame_or_series is Series:
+        gb = gb["b"]
+
+    result = gb.transform(func)
+    expected = DataFrame({"b": expected_values}, index=df.index)
+    if frame_or_series is Series:
+        expected = expected["b"]
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", ["A", ["A", "B"]])
+def test_as_index_no_change(keys, df, groupby_func):
+    # GH#49834 - as_index should have no impact on DataFrameGroupBy.transform
+    if keys == "A":
+        # Column B is string dtype; will fail on some ops
+        df = df.drop(columns="B")
+    args = get_groupby_method_args(groupby_func, df)
+    gb_as_index_true = df.groupby(keys, as_index=True)
+    gb_as_index_false = df.groupby(keys, as_index=False)
+    if groupby_func == "corrwith":
+        warn = FutureWarning
+        msg = "DataFrameGroupBy.corrwith is deprecated"
+    else:
+        warn = None
+        msg = ""
+    with tm.assert_produces_warning(warn, match=msg):
+        result = gb_as_index_true.transform(groupby_func, *args)
+    with tm.assert_produces_warning(warn, match=msg):
+        expected = gb_as_index_false.transform(groupby_func, *args)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["idxmax", "idxmin"])
+@pytest.mark.parametrize("numeric_only", [True, False])
+def test_idxmin_idxmax_transform_args(how, skipna, numeric_only):
+    # GH#55268 - ensure *args are passed through when calling transform
+    df = DataFrame({"a": [1, 1, 1, 2], "b": [3.0, 4.0, np.nan, 6.0], "c": list("abcd")})
+    gb = df.groupby("a")
+    if skipna:
+        result = gb.transform(how, skipna, numeric_only)
+        expected = gb.transform(how, skipna=skipna, numeric_only=numeric_only)
+        tm.assert_frame_equal(result, expected)
+    else:
+        msg = f"DataFrameGroupBy.{how} with skipna=False encountered an NA value"
+        with pytest.raises(ValueError, match=msg):
+            gb.transform(how, skipna, numeric_only)
+
+
+def test_transform_sum_one_column_no_matching_labels():
+    df = DataFrame({"X": [1.0]})
+    series = Series(["Y"])
+    result = df.groupby(series, as_index=False).transform("sum")
+    expected = DataFrame({"X": [1.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_sum_no_matching_labels():
+    df = DataFrame({"X": [1.0, -93204, 4935]})
+    series = Series(["A", "B", "C"])
+
+    result = df.groupby(series, as_index=False).transform("sum")
+    expected = DataFrame({"X": [1.0, -93204, 4935]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_sum_one_column_with_matching_labels():
+    df = DataFrame({"X": [1.0, -93204, 4935]})
+    series = Series(["A", "B", "A"])
+
+    result = df.groupby(series, as_index=False).transform("sum")
+    expected = DataFrame({"X": [4936.0, -93204, 4936.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_sum_one_column_with_missing_labels():
+    df = DataFrame({"X": [1.0, -93204, 4935]})
+    series = Series(["A", "C"])
+
+    result = df.groupby(series, as_index=False).transform("sum")
+    expected = DataFrame({"X": [1.0, -93204, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_transform_sum_one_column_with_matching_labels_and_missing_labels():
+    df = DataFrame({"X": [1.0, -93204, 4935]})
+    series = Series(["A", "A"])
+
+    result = df.groupby(series, as_index=False).transform("sum")
+    expected = DataFrame({"X": [-93203.0, -93203.0, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int32", "float32"])
+def test_min_one_unobserved_category_no_type_coercion(dtype):
+    # GH#58084
+    df = DataFrame({"A": Categorical([1, 1, 2], categories=[1, 2, 3]), "B": [3, 4, 5]})
+    df["B"] = df["B"].astype(dtype)
+    gb = df.groupby("A", observed=False)
+    result = gb.transform("min")
+
+    expected = DataFrame({"B": [3, 3, 5]}, dtype=dtype)
+    tm.assert_frame_equal(expected, result)
+
+
+def test_min_all_empty_data_no_type_coercion():
+    # GH#58084
+    df = DataFrame(
+        {
+            "X": Categorical(
+                [],
+                categories=[1, "randomcat", 100],
+            ),
+            "Y": [],
+        }
+    )
+    df["Y"] = df["Y"].astype("int32")
+
+    gb = df.groupby("X", observed=False)
+    result = gb.transform("min")
+
+    expected = DataFrame({"Y": []}, dtype="int32")
+    tm.assert_frame_equal(expected, result)
+
+
+def test_min_one_dim_no_type_coercion():
+    # GH#58084
+    df = DataFrame({"Y": [9435, -5465765, 5055, 0, 954960]})
+    df["Y"] = df["Y"].astype("int32")
+    categories = Categorical([1, 2, 2, 5, 1], categories=[1, 2, 3, 4, 5])
+
+    gb = df.groupby(categories, observed=False)
+    result = gb.transform("min")
+
+    expected = DataFrame({"Y": [9435, -5465765, -5465765, 0, 9435]}, dtype="int32")
+    tm.assert_frame_equal(expected, result)
+
+
+def test_nan_in_cumsum_group_label():
+    # GH#58811
+    df = DataFrame({"A": [1, None], "B": [2, 3]}, dtype="Int16")
+    gb = df.groupby("A")["B"]
+    result = gb.cumsum()
+    expected = Series([2, None], dtype="Int16", name="B")
+    tm.assert_series_equal(expected, result)
diff --git a/pandas/pandas/tests/indexes/__init__.py b/pandas/pandas/tests/indexes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/base_class/__init__.py b/pandas/pandas/tests/indexes/base_class/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/base_class/test_constructors.py b/pandas/pandas/tests/indexes/base_class/test_constructors.py
new file mode 100644
index 0000000000000..0896b97e8a40e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_constructors.py
@@ -0,0 +1,71 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestIndexConstructor:
+    # Tests for the Index constructor, specifically for cases that do
+    #  not return a subclass
+
+    @pytest.mark.parametrize("value", [1, np.int64(1)])
+    def test_constructor_corner(self, value):
+        # corner case
+        msg = (
+            r"Index\(\.\.\.\) must be called with a collection of some "
+            f"kind, {value} was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            Index(value)
+
+    @pytest.mark.parametrize("index_vals", [[("A", 1), "B"], ["B", ("A", 1)]])
+    def test_construction_list_mixed_tuples(self, index_vals):
+        # see gh-10697: if we are constructing from a mixed list of tuples,
+        # make sure that we are independent of the sorting order.
+        index = Index(index_vals)
+        assert isinstance(index, Index)
+        assert not isinstance(index, MultiIndex)
+
+    def test_constructor_cast(self):
+        msg = "could not convert string to float"
+        with pytest.raises(ValueError, match=msg):
+            Index(["a", "b", "c"], dtype=float)
+
+    @pytest.mark.parametrize("tuple_list", [[()], [(), ()]])
+    def test_construct_empty_tuples(self, tuple_list):
+        # GH #45608
+        result = Index(tuple_list)
+        expected = MultiIndex.from_tuples(tuple_list)
+
+        tm.assert_index_equal(result, expected)
+
+    def test_index_string_inference(self):
+        # GH#54430
+        expected = Index(["a", "b"], dtype=pd.StringDtype(na_value=np.nan))
+        with pd.option_context("future.infer_string", True):
+            ser = Index(["a", "b"])
+        tm.assert_index_equal(ser, expected)
+
+        expected = Index(["a", 1], dtype="object")
+        with pd.option_context("future.infer_string", True):
+            ser = Index(["a", 1])
+        tm.assert_index_equal(ser, expected)
+
+    @pytest.mark.parametrize("klass", [Series, Index])
+    def test_inference_on_pandas_objects(self, klass):
+        # GH#56012
+        obj = klass([pd.Timestamp("2019-12-31")], dtype=object)
+        result = Index(obj)
+        assert result.dtype == np.object_
+
+    def test_constructor_not_read_only(self):
+        # GH#57130
+        ser = Series([1, 2], dtype=object)
+        idx = Index(ser)
+        assert idx._values.flags.writeable
diff --git a/pandas/pandas/tests/indexes/base_class/test_formats.py b/pandas/pandas/tests/indexes/base_class/test_formats.py
new file mode 100644
index 0000000000000..260b4203a4f04
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_formats.py
@@ -0,0 +1,149 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+import pandas._config.config as cf
+
+from pandas import Index
+import pandas._testing as tm
+
+
+class TestIndexRendering:
+    def test_repr_is_valid_construction_code(self):
+        # for the case of Index, where the repr is traditional rather than
+        # stylized
+        idx = Index(["a", "b"])
+        res = eval(repr(idx))
+        tm.assert_index_equal(res, idx)
+
+    @pytest.mark.xfail(using_string_dtype(), reason="repr different")
+    @pytest.mark.parametrize(
+        "index,expected",
+        [
+            # ASCII
+            # short
+            (
+                Index(["a", "bb", "ccc"]),
+                """Index(['a', 'bb', 'ccc'], dtype='object')""",
+            ),
+            # multiple lines
+            (
+                Index(["a", "bb", "ccc"] * 10),
+                "Index(['a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', "
+                "'bb', 'ccc', 'a', 'bb', 'ccc',\n"
+                "       'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', "
+                "'bb', 'ccc', 'a', 'bb', 'ccc',\n"
+                "       'a', 'bb', 'ccc', 'a', 'bb', 'ccc'],\n"
+                "      dtype='object')",
+            ),
+            # truncated
+            (
+                Index(["a", "bb", "ccc"] * 100),
+                "Index(['a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a',\n"
+                "       ...\n"
+                "       'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc'],\n"
+                "      dtype='object', length=300)",
+            ),
+            # Non-ASCII
+            # short
+            (
+                Index(["あ", "いい", "ううう"]),
+                """Index(['あ', 'いい', 'ううう'], dtype='object')""",
+            ),
+            # multiple lines
+            (
+                Index(["あ", "いい", "ううう"] * 10),
+                (
+                    "Index(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', "
+                    "'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', "
+                    "'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう'],\n"
+                    "      dtype='object')"
+                ),
+            ),
+            # truncated
+            (
+                Index(["あ", "いい", "ううう"] * 100),
+                (
+                    "Index(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', "
+                    "'あ', 'いい', 'ううう', 'あ',\n"
+                    "       ...\n"
+                    "       'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう', 'あ', 'いい', 'ううう'],\n"
+                    "      dtype='object', length=300)"
+                ),
+            ),
+        ],
+    )
+    def test_string_index_repr(self, index, expected):
+        result = repr(index)
+        assert result == expected
+
+    @pytest.mark.xfail(using_string_dtype(), reason="repr different")
+    @pytest.mark.parametrize(
+        "index,expected",
+        [
+            # short
+            (
+                Index(["あ", "いい", "ううう"]),
+                ("Index(['あ', 'いい', 'ううう'], dtype='object')"),
+            ),
+            # multiple lines
+            (
+                Index(["あ", "いい", "ううう"] * 10),
+                (
+                    "Index(['あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ', 'いい', 'ううう'],\n"
+                    "      dtype='object')"
+                    ""
+                ),
+            ),
+            # truncated
+            (
+                Index(["あ", "いい", "ううう"] * 100),
+                (
+                    "Index(['あ', 'いい', 'ううう', 'あ', 'いい', "
+                    "'ううう', 'あ', 'いい', 'ううう',\n"
+                    "       'あ',\n"
+                    "       ...\n"
+                    "       'ううう', 'あ', 'いい', 'ううう', 'あ', "
+                    "'いい', 'ううう', 'あ', 'いい',\n"
+                    "       'ううう'],\n"
+                    "      dtype='object', length=300)"
+                ),
+            ),
+        ],
+    )
+    def test_string_index_repr_with_unicode_option(self, index, expected):
+        # Enable Unicode option -----------------------------------------
+        with cf.option_context("display.unicode.east_asian_width", True):
+            result = repr(index)
+            assert result == expected
+
+    def test_repr_summary(self):
+        with cf.option_context("display.max_seq_items", 10):
+            result = repr(Index(np.arange(1000)))
+            assert len(result) < 200
+            assert "..." in result
+
+    def test_summary_bug(self):
+        # GH#3869
+        ind = Index(["{other}%s", "~:{range}:0"], name="A")
+        result = ind._summary()
+        # shouldn't be formatted accidentally.
+        assert "~:{range}:0" in result
+        assert "{other}%s" in result
+
+    def test_index_repr_bool_nan(self):
+        # GH32146
+        arr = Index([True, False, np.nan], dtype=object)
+        exp2 = repr(arr)
+        out2 = "Index([True, False, nan], dtype='object')"
+        assert out2 == exp2
diff --git a/pandas/pandas/tests/indexes/base_class/test_indexing.py b/pandas/pandas/tests/indexes/base_class/test_indexing.py
new file mode 100644
index 0000000000000..2988fa7d1baa1
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_indexing.py
@@ -0,0 +1,104 @@
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+
+import pandas as pd
+from pandas import (
+    Index,
+    NaT,
+)
+import pandas._testing as tm
+
+
+class TestGetSliceBounds:
+    @pytest.mark.parametrize("side, expected", [("left", 4), ("right", 5)])
+    def test_get_slice_bounds_within(self, side, expected):
+        index = Index(list("abcdef"))
+        result = index.get_slice_bound("e", side=side)
+        assert result == expected
+
+    @pytest.mark.parametrize("side", ["left", "right"])
+    @pytest.mark.parametrize(
+        "data, bound, expected", [(list("abcdef"), "x", 6), (list("bcdefg"), "a", 0)]
+    )
+    def test_get_slice_bounds_outside(self, side, expected, data, bound):
+        index = Index(data)
+        result = index.get_slice_bound(bound, side=side)
+        assert result == expected
+
+    def test_get_slice_bounds_invalid_side(self):
+        with pytest.raises(ValueError, match="Invalid value for side kwarg"):
+            Index([]).get_slice_bound("a", side="middle")
+
+
+class TestGetIndexerNonUnique:
+    def test_get_indexer_non_unique_dtype_mismatch(self):
+        # GH#25459
+        indexes, missing = Index(["A", "B"]).get_indexer_non_unique(Index([0]))
+        tm.assert_numpy_array_equal(np.array([-1], dtype=np.intp), indexes)
+        tm.assert_numpy_array_equal(np.array([0], dtype=np.intp), missing)
+
+    @pytest.mark.parametrize(
+        "idx_values,idx_non_unique",
+        [
+            ([np.nan, 100, 200, 100], [np.nan, 100]),
+            ([np.nan, 100.0, 200.0, 100.0], [np.nan, 100.0]),
+        ],
+    )
+    def test_get_indexer_non_unique_int_index(self, idx_values, idx_non_unique):
+        indexes, missing = Index(idx_values).get_indexer_non_unique(Index([np.nan]))
+        tm.assert_numpy_array_equal(np.array([0], dtype=np.intp), indexes)
+        tm.assert_numpy_array_equal(np.array([], dtype=np.intp), missing)
+
+        indexes, missing = Index(idx_values).get_indexer_non_unique(
+            Index(idx_non_unique)
+        )
+        tm.assert_numpy_array_equal(np.array([0, 1, 3], dtype=np.intp), indexes)
+        tm.assert_numpy_array_equal(np.array([], dtype=np.intp), missing)
+
+
+class TestGetLoc:
+    @pytest.mark.slow  # to_flat_index takes a while
+    def test_get_loc_tuple_monotonic_above_size_cutoff(self, monkeypatch):
+        # Go through the libindex path for which using
+        # _bin_search vs ndarray.searchsorted makes a difference
+
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", 100)
+            lev = list("ABCD")
+            dti = pd.date_range("2016-01-01", periods=10)
+
+            mi = pd.MultiIndex.from_product([lev, range(5), dti])
+            oidx = mi.to_flat_index()
+
+            loc = len(oidx) // 2
+            tup = oidx[loc]
+
+            res = oidx.get_loc(tup)
+        assert res == loc
+
+    def test_get_loc_nan_object_dtype_nonmonotonic_nonunique(self):
+        # case that goes through _maybe_get_bool_indexer
+        idx = Index(["foo", np.nan, None, "foo", 1.0, None], dtype=object)
+
+        # we dont raise KeyError on nan
+        res = idx.get_loc(np.nan)
+        assert res == 1
+
+        # we only match on None, not on np.nan
+        res = idx.get_loc(None)
+        expected = np.array([False, False, True, False, False, True])
+        tm.assert_numpy_array_equal(res, expected)
+
+        # we don't match at all on mismatched NA
+        with pytest.raises(KeyError, match="NaT"):
+            idx.get_loc(NaT)
+
+
+def test_getitem_boolean_ea_indexer():
+    # GH#45806
+    ser = pd.Series([True, False, pd.NA], dtype="boolean")
+    result = ser.index[ser]
+    expected = Index([0])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/base_class/test_pickle.py b/pandas/pandas/tests/indexes/base_class/test_pickle.py
new file mode 100644
index 0000000000000..c670921decb78
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_pickle.py
@@ -0,0 +1,11 @@
+from pandas import Index
+import pandas._testing as tm
+
+
+def test_pickle_preserves_object_dtype():
+    # GH#43188, GH#43155 don't infer numeric dtype
+    index = Index([1, 2, 3], dtype=object)
+
+    result = tm.round_trip_pickle(index)
+    assert result.dtype == object
+    tm.assert_index_equal(index, result)
diff --git a/pandas/pandas/tests/indexes/base_class/test_reshape.py b/pandas/pandas/tests/indexes/base_class/test_reshape.py
new file mode 100644
index 0000000000000..d4932712de1bb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_reshape.py
@@ -0,0 +1,95 @@
+"""
+Tests for ndarray-like method on the base Index class
+"""
+
+import numpy as np
+import pytest
+
+from pandas import Index
+import pandas._testing as tm
+
+
+class TestReshape:
+    def test_repeat(self):
+        repeats = 2
+        index = Index([1, 2, 3])
+        expected = Index([1, 1, 2, 2, 3, 3])
+
+        result = index.repeat(repeats)
+        tm.assert_index_equal(result, expected)
+
+    def test_insert(self):
+        # GH 7256
+        # validate neg/pos inserts
+        result = Index(["b", "c", "d"])
+
+        # test 0th element
+        tm.assert_index_equal(Index(["a", "b", "c", "d"]), result.insert(0, "a"))
+
+        # test Nth element that follows Python list behavior
+        tm.assert_index_equal(Index(["b", "c", "e", "d"]), result.insert(-1, "e"))
+
+        # test loc +/- neq (0, -1)
+        tm.assert_index_equal(result.insert(1, "z"), result.insert(-2, "z"))
+
+        # test empty
+        null_index = Index([])
+        tm.assert_index_equal(Index(["a"]), null_index.insert(0, "a"))
+
+    def test_insert_missing(self, nulls_fixture, using_infer_string):
+        # GH#22295
+        # test there is no mangling of NA values
+        expected = Index(["a", nulls_fixture, "b", "c"], dtype=object)
+        result = Index(list("abc"), dtype=object).insert(
+            1, Index([nulls_fixture], dtype=object)
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "val", [(1, 2), np.datetime64("2019-12-31"), np.timedelta64(1, "D")]
+    )
+    @pytest.mark.parametrize("loc", [-1, 2])
+    def test_insert_datetime_into_object(self, loc, val):
+        # GH#44509
+        idx = Index(["1", "2", "3"])
+        result = idx.insert(loc, val)
+        expected = Index(["1", "2", val, "3"])
+        tm.assert_index_equal(result, expected)
+        assert type(expected[2]) is type(val)
+
+    def test_insert_none_into_string_numpy(self, string_dtype_no_object):
+        # GH#55365
+        index = Index(["a", "b", "c"], dtype=string_dtype_no_object)
+        result = index.insert(-1, None)
+        expected = Index(["a", "b", None, "c"], dtype=string_dtype_no_object)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "pos,expected",
+        [
+            (0, Index(["b", "c", "d"], name="index")),
+            (-1, Index(["a", "b", "c"], name="index")),
+        ],
+    )
+    def test_delete(self, pos, expected):
+        index = Index(["a", "b", "c", "d"], name="index")
+        result = index.delete(pos)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+
+    def test_delete_raises(self):
+        index = Index(["a", "b", "c", "d"], name="index")
+        msg = "index 5 is out of bounds for axis 0 with size 4"
+        with pytest.raises(IndexError, match=msg):
+            index.delete(5)
+
+    def test_append_multiple(self):
+        index = Index(["a", "b", "c", "d", "e", "f"])
+
+        foos = [index[:2], index[2:4], index[4:]]
+        result = foos[0].append(foos[1:])
+        tm.assert_index_equal(result, index)
+
+        # empty
+        result = index.append([])
+        tm.assert_index_equal(result, index)
diff --git a/pandas/pandas/tests/indexes/base_class/test_setops.py b/pandas/pandas/tests/indexes/base_class/test_setops.py
new file mode 100644
index 0000000000000..d57df82b2358c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_setops.py
@@ -0,0 +1,266 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.algorithms import safe_sort
+
+
+def equal_contents(arr1, arr2) -> bool:
+    """
+    Checks if the set of unique elements of arr1 and arr2 are equivalent.
+    """
+    return frozenset(arr1) == frozenset(arr2)
+
+
+class TestIndexSetOps:
+    @pytest.mark.parametrize(
+        "method", ["union", "intersection", "difference", "symmetric_difference"]
+    )
+    def test_setops_sort_validation(self, method):
+        idx1 = Index(["a", "b"])
+        idx2 = Index(["b", "c"])
+
+        with pytest.raises(ValueError, match="The 'sort' keyword only takes"):
+            getattr(idx1, method)(idx2, sort=2)
+
+        # sort=True is supported as of GH#??
+        getattr(idx1, method)(idx2, sort=True)
+
+    def test_setops_preserve_object_dtype(self):
+        idx = Index([1, 2, 3], dtype=object)
+        result = idx.intersection(idx[1:])
+        expected = idx[1:]
+        tm.assert_index_equal(result, expected)
+
+        # if other is not monotonic increasing, intersection goes through
+        #  a different route
+        result = idx.intersection(idx[1:][::-1])
+        tm.assert_index_equal(result, expected)
+
+        result = idx._union(idx[1:], sort=None)
+        expected = idx
+        tm.assert_numpy_array_equal(result, expected.values)
+
+        result = idx.union(idx[1:], sort=None)
+        tm.assert_index_equal(result, expected)
+
+        # if other is not monotonic increasing, _union goes through
+        #  a different route
+        result = idx._union(idx[1:][::-1], sort=None)
+        tm.assert_numpy_array_equal(result, expected.values)
+
+        result = idx.union(idx[1:][::-1], sort=None)
+        tm.assert_index_equal(result, expected)
+
+    def test_union_base(self):
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[3:]
+        second = index[:5]
+
+        result = first.union(second)
+
+        expected = Index([0, 1, 2, "a", "b", "c"])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [np.array, Series, list])
+    def test_union_different_type_base(self, klass):
+        # GH 10149
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[3:]
+        second = index[:5]
+
+        result = first.union(klass(second.values))
+
+        assert equal_contents(result, index)
+
+    def test_union_sort_other_incomparable(self):
+        # https://github.com/pandas-dev/pandas/issues/24959
+        idx = Index([1, pd.Timestamp("2000")])
+        # default (sort=None)
+        with tm.assert_produces_warning(RuntimeWarning, match="not supported between"):
+            result = idx.union(idx[:1])
+
+        tm.assert_index_equal(result, idx)
+
+        # sort=None
+        with tm.assert_produces_warning(RuntimeWarning, match="not supported between"):
+            result = idx.union(idx[:1], sort=None)
+        tm.assert_index_equal(result, idx)
+
+        # sort=False
+        result = idx.union(idx[:1], sort=False)
+        tm.assert_index_equal(result, idx)
+
+    def test_union_sort_other_incomparable_true(self):
+        idx = Index([1, pd.Timestamp("2000")])
+        with pytest.raises(TypeError, match=".*"):
+            idx.union(idx[:1], sort=True)
+
+    def test_intersection_equal_sort_true(self):
+        idx = Index(["c", "a", "b"])
+        sorted_ = Index(["a", "b", "c"])
+        tm.assert_index_equal(idx.intersection(idx, sort=True), sorted_)
+
+    def test_intersection_base(self, sort):
+        # (same results for py2 and py3 but sortedness not tested elsewhere)
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[:5]
+        second = index[:3]
+
+        expected = Index([0, 1, "a"]) if sort is None else Index([0, "a", 1])
+        result = first.intersection(second, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [np.array, Series, list])
+    def test_intersection_different_type_base(self, klass, sort):
+        # GH 10149
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[:5]
+        second = index[:3]
+
+        result = first.intersection(klass(second.values), sort=sort)
+        assert equal_contents(result, second)
+
+    def test_intersection_nosort(self):
+        result = Index(["c", "b", "a"]).intersection(["b", "a"])
+        expected = Index(["b", "a"])
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_equal_sort(self):
+        idx = Index(["c", "a", "b"])
+        tm.assert_index_equal(idx.intersection(idx, sort=False), idx)
+        tm.assert_index_equal(idx.intersection(idx, sort=None), idx)
+
+    def test_intersection_str_dates(self, sort):
+        dt_dates = [datetime(2012, 2, 9), datetime(2012, 2, 22)]
+
+        i1 = Index(dt_dates, dtype=object)
+        i2 = Index(["aa"], dtype=object)
+        result = i2.intersection(i1, sort=sort)
+
+        assert len(result) == 0
+
+    @pytest.mark.parametrize(
+        "index2,expected_arr",
+        [(["B", "D"], ["B"]), (["B", "D", "A"], ["A", "B"])],
+    )
+    def test_intersection_non_monotonic_non_unique(self, index2, expected_arr, sort):
+        # non-monotonic non-unique
+        index1 = Index(["A", "B", "A", "C"])
+        expected = Index(expected_arr)
+        result = index1.intersection(Index(index2), sort=sort)
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+    def test_difference_base(self, sort):
+        # (same results for py2 and py3 but sortedness not tested elsewhere)
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[:4]
+        second = index[3:]
+
+        result = first.difference(second, sort)
+        expected = Index([0, "a", 1])
+        if sort is None:
+            expected = Index(safe_sort(expected))
+        tm.assert_index_equal(result, expected)
+
+    def test_symmetric_difference(self):
+        # (same results for py2 and py3 but sortedness not tested elsewhere)
+        index = Index([0, "a", 1, "b", 2, "c"])
+        first = index[:4]
+        second = index[3:]
+
+        result = first.symmetric_difference(second)
+        expected = Index([0, 1, 2, "a", "c"])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "method,expected,sort",
+        [
+            (
+                "intersection",
+                np.array(
+                    [(1, "A"), (2, "A"), (1, "B"), (2, "B")],
+                    dtype=[("num", int), ("let", "S1")],
+                ),
+                False,
+            ),
+            (
+                "intersection",
+                np.array(
+                    [(1, "A"), (1, "B"), (2, "A"), (2, "B")],
+                    dtype=[("num", int), ("let", "S1")],
+                ),
+                None,
+            ),
+            (
+                "union",
+                np.array(
+                    [(1, "A"), (1, "B"), (1, "C"), (2, "A"), (2, "B"), (2, "C")],
+                    dtype=[("num", int), ("let", "S1")],
+                ),
+                None,
+            ),
+        ],
+    )
+    def test_tuple_union_bug(self, method, expected, sort):
+        index1 = Index(
+            np.array(
+                [(1, "A"), (2, "A"), (1, "B"), (2, "B")],
+                dtype=[("num", int), ("let", "S1")],
+            )
+        )
+        index2 = Index(
+            np.array(
+                [(1, "A"), (2, "A"), (1, "B"), (2, "B"), (1, "C"), (2, "C")],
+                dtype=[("num", int), ("let", "S1")],
+            )
+        )
+
+        result = getattr(index1, method)(index2, sort=sort)
+        assert result.ndim == 1
+
+        expected = Index(expected)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("first_list", [["b", "a"], []])
+    @pytest.mark.parametrize("second_list", [["a", "b"], []])
+    @pytest.mark.parametrize(
+        "first_name, second_name, expected_name",
+        [("A", "B", None), (None, "B", None), ("A", None, None)],
+    )
+    def test_union_name_preservation(
+        self, first_list, second_list, first_name, second_name, expected_name, sort
+    ):
+        first = Index(first_list, name=first_name)
+        second = Index(second_list, name=second_name)
+        union = first.union(second, sort=sort)
+
+        vals = set(first_list).union(second_list)
+
+        if sort is None and len(first_list) > 0 and len(second_list) > 0:
+            expected = Index(sorted(vals), name=expected_name)
+            tm.assert_index_equal(union, expected)
+        else:
+            expected = Index(vals, name=expected_name)
+            tm.assert_index_equal(union.sort_values(), expected.sort_values())
+
+    @pytest.mark.parametrize(
+        "diff_type, expected",
+        [["difference", [1, "B"]], ["symmetric_difference", [1, 2, "B", "C"]]],
+    )
+    def test_difference_object_type(self, diff_type, expected):
+        # GH 13432
+        idx1 = Index([0, 1, "A", "B"])
+        idx2 = Index([0, 2, "A", "C"])
+        result = getattr(idx1, diff_type)(idx2)
+        expected = Index(expected)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/base_class/test_where.py b/pandas/pandas/tests/indexes/base_class/test_where.py
new file mode 100644
index 0000000000000..0c8969735e14e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/base_class/test_where.py
@@ -0,0 +1,13 @@
+import numpy as np
+
+from pandas import Index
+import pandas._testing as tm
+
+
+class TestWhere:
+    def test_where_intlike_str_doesnt_cast_ints(self):
+        idx = Index(range(3))
+        mask = np.array([True, False, True])
+        res = idx.where(mask, "2")
+        expected = Index([0, "2", 2])
+        tm.assert_index_equal(res, expected)
diff --git a/pandas/pandas/tests/indexes/categorical/__init__.py b/pandas/pandas/tests/indexes/categorical/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/categorical/test_append.py b/pandas/pandas/tests/indexes/categorical/test_append.py
new file mode 100644
index 0000000000000..b48c3219f5111
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_append.py
@@ -0,0 +1,62 @@
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    Index,
+)
+import pandas._testing as tm
+
+
+class TestAppend:
+    @pytest.fixture
+    def ci(self):
+        categories = list("cab")
+        return CategoricalIndex(list("aabbca"), categories=categories, ordered=False)
+
+    def test_append(self, ci):
+        # append cats with the same categories
+        result = ci[:3].append(ci[3:])
+        tm.assert_index_equal(result, ci, exact=True)
+
+        foos = [ci[:1], ci[1:3], ci[3:]]
+        result = foos[0].append(foos[1:])
+        tm.assert_index_equal(result, ci, exact=True)
+
+    def test_append_empty(self, ci):
+        # empty
+        result = ci.append([])
+        tm.assert_index_equal(result, ci, exact=True)
+
+    def test_append_mismatched_categories(self, ci):
+        # appending with different categories or reordered is not ok
+        msg = "all inputs must be Index"
+        with pytest.raises(TypeError, match=msg):
+            ci.append(ci.values.set_categories(list("abcd")))
+        with pytest.raises(TypeError, match=msg):
+            ci.append(ci.values.reorder_categories(list("abc")))
+
+    def test_append_category_objects(self, ci):
+        # with objects
+        result = ci.append(Index(["c", "a"]))
+        expected = CategoricalIndex(list("aabbcaca"), categories=ci.categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_append_non_categories(self, ci):
+        # invalid objects -> cast to object via concat_compat
+        result = ci.append(Index(["a", "d"]))
+        expected = Index(["a", "a", "b", "b", "c", "a", "a", "d"])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_append_object(self, ci):
+        # GH#14298 - if base object is not categorical -> coerce to object
+        result = Index(["c", "a"]).append(ci)
+        expected = Index(list("caaabbca"))
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_append_to_another(self):
+        # hits Index._concat
+        fst = Index(["a", "b"])
+        snd = CategoricalIndex(["d", "e"])
+        result = fst.append(snd)
+        expected = Index(["a", "b", "d", "e"])
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/categorical/test_astype.py b/pandas/pandas/tests/indexes/categorical/test_astype.py
new file mode 100644
index 0000000000000..a17627b7515b2
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_astype.py
@@ -0,0 +1,90 @@
+from datetime import date
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    Index,
+    IntervalIndex,
+)
+import pandas._testing as tm
+
+
+class TestAstype:
+    def test_astype(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+
+        result = ci.astype(object)
+        tm.assert_index_equal(result, Index(np.array(ci), dtype=object))
+
+        # this IS equal, but not the same class
+        assert result.equals(ci)
+        assert isinstance(result, Index)
+        assert not isinstance(result, CategoricalIndex)
+
+        # interval
+        ii = IntervalIndex.from_arrays(left=[-0.001, 2.0], right=[2, 4], closed="right")
+
+        ci = CategoricalIndex(
+            Categorical.from_codes([0, 1, -1], categories=ii, ordered=True)
+        )
+
+        result = ci.astype("interval")
+        expected = ii.take([0, 1, -1], allow_fill=True, fill_value=np.nan)
+        tm.assert_index_equal(result, expected)
+
+        result = IntervalIndex(result.values)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("name", [None, "foo"])
+    @pytest.mark.parametrize("dtype_ordered", [True, False])
+    @pytest.mark.parametrize("index_ordered", [True, False])
+    def test_astype_category(self, name, dtype_ordered, index_ordered):
+        # GH#18630
+        index = CategoricalIndex(
+            list("aabbca"), categories=list("cab"), ordered=index_ordered
+        )
+        if name:
+            index = index.rename(name)
+
+        # standard categories
+        dtype = CategoricalDtype(ordered=dtype_ordered)
+        result = index.astype(dtype)
+        expected = CategoricalIndex(
+            index.tolist(),
+            name=name,
+            categories=index.categories,
+            ordered=dtype_ordered,
+        )
+        tm.assert_index_equal(result, expected)
+
+        # non-standard categories
+        dtype = CategoricalDtype(index.unique().tolist()[:-1], dtype_ordered)
+        result = index.astype(dtype)
+        expected = CategoricalIndex(index.tolist(), name=name, dtype=dtype)
+        tm.assert_index_equal(result, expected)
+
+        if dtype_ordered is False:
+            # dtype='category' can't specify ordered, so only test once
+            result = index.astype("category")
+            expected = index
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [True, False])
+    def test_categorical_date_roundtrip(self, box):
+        # astype to categorical and back should preserve date objects
+        v = date.today()
+
+        obj = Index([v, v])
+        assert obj.dtype == object
+        if box:
+            obj = obj.array
+
+        cat = obj.astype("category")
+
+        rtrip = cat.astype(object)
+        assert rtrip.dtype == object
+        assert type(rtrip[0]) is date
diff --git a/pandas/pandas/tests/indexes/categorical/test_category.py b/pandas/pandas/tests/indexes/categorical/test_category.py
new file mode 100644
index 0000000000000..d9c9fdc62b0bc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_category.py
@@ -0,0 +1,404 @@
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import index as libindex
+from pandas._libs.arrays import NDArrayBacked
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+)
+import pandas._testing as tm
+from pandas.core.indexes.api import (
+    CategoricalIndex,
+    Index,
+)
+
+
+class TestCategoricalIndex:
+    @pytest.fixture
+    def simple_index(self) -> CategoricalIndex:
+        """
+        Fixture that provides a CategoricalIndex.
+        """
+        return CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+
+    def test_can_hold_identifiers(self):
+        idx = CategoricalIndex(list("aabbca"), categories=None, ordered=False)
+        key = idx[0]
+        assert idx._can_hold_identifiers_and_holds_name(key) is True
+
+    def test_insert(self, simple_index):
+        ci = simple_index
+        categories = ci.categories
+
+        # test 0th element
+        result = ci.insert(0, "a")
+        expected = CategoricalIndex(list("aaabbca"), categories=categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # test Nth element that follows Python list behavior
+        result = ci.insert(-1, "a")
+        expected = CategoricalIndex(list("aabbcaa"), categories=categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # test empty
+        result = CategoricalIndex([], categories=categories).insert(0, "a")
+        expected = CategoricalIndex(["a"], categories=categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # invalid -> cast to object
+        expected = ci.astype(object).insert(0, "d")
+        result = ci.insert(0, "d").astype(object)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # GH 18295 (test missing)
+        expected = CategoricalIndex(["a", np.nan, "a", "b", "c", "b"])
+        for na in (np.nan, pd.NaT, None):
+            result = CategoricalIndex(list("aabcb")).insert(1, na)
+            tm.assert_index_equal(result, expected)
+
+    def test_insert_na_mismatched_dtype(self):
+        ci = CategoricalIndex([0, 1, 1])
+        result = ci.insert(0, pd.NaT)
+        expected = Index([pd.NaT, 0, 1, 1], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_delete(self, simple_index):
+        ci = simple_index
+        categories = ci.categories
+
+        result = ci.delete(0)
+        expected = CategoricalIndex(list("abbca"), categories=categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = ci.delete(-1)
+        expected = CategoricalIndex(list("aabbc"), categories=categories)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        with tm.external_error_raised((IndexError, ValueError)):
+            # Either depending on NumPy version
+            ci.delete(10)
+
+    @pytest.mark.parametrize(
+        "data, non_lexsorted_data",
+        [[[1, 2, 3], [9, 0, 1, 2, 3]], [list("abc"), list("fabcd")]],
+    )
+    def test_is_monotonic(self, data, non_lexsorted_data):
+        c = CategoricalIndex(data)
+        assert c.is_monotonic_increasing is True
+        assert c.is_monotonic_decreasing is False
+
+        c = CategoricalIndex(data, ordered=True)
+        assert c.is_monotonic_increasing is True
+        assert c.is_monotonic_decreasing is False
+
+        c = CategoricalIndex(data, categories=reversed(data))
+        assert c.is_monotonic_increasing is False
+        assert c.is_monotonic_decreasing is True
+
+        c = CategoricalIndex(data, categories=reversed(data), ordered=True)
+        assert c.is_monotonic_increasing is False
+        assert c.is_monotonic_decreasing is True
+
+        # test when data is neither monotonic increasing nor decreasing
+        reordered_data = [data[0], data[2], data[1]]
+        c = CategoricalIndex(reordered_data, categories=reversed(data))
+        assert c.is_monotonic_increasing is False
+        assert c.is_monotonic_decreasing is False
+
+        # non lexsorted categories
+        categories = non_lexsorted_data
+
+        c = CategoricalIndex(categories[:2], categories=categories)
+        assert c.is_monotonic_increasing is True
+        assert c.is_monotonic_decreasing is False
+
+        c = CategoricalIndex(categories[1:3], categories=categories)
+        assert c.is_monotonic_increasing is True
+        assert c.is_monotonic_decreasing is False
+
+    def test_has_duplicates(self):
+        idx = CategoricalIndex([0, 0, 0], name="foo")
+        assert idx.is_unique is False
+        assert idx.has_duplicates is True
+
+        idx = CategoricalIndex([0, 1], categories=[2, 3], name="foo")
+        assert idx.is_unique is False
+        assert idx.has_duplicates is True
+
+        idx = CategoricalIndex([0, 1, 2, 3], categories=[1, 2, 3], name="foo")
+        assert idx.is_unique is True
+        assert idx.has_duplicates is False
+
+    @pytest.mark.parametrize(
+        "data, categories, expected",
+        [
+            (
+                [1, 1, 1],
+                [1, 2, 3],
+                {
+                    "first": np.array([False, True, True]),
+                    "last": np.array([True, True, False]),
+                    False: np.array([True, True, True]),
+                },
+            ),
+            (
+                [1, 1, 1],
+                list("abc"),
+                {
+                    "first": np.array([False, True, True]),
+                    "last": np.array([True, True, False]),
+                    False: np.array([True, True, True]),
+                },
+            ),
+            (
+                [2, "a", "b"],
+                list("abc"),
+                {
+                    "first": np.zeros(shape=(3), dtype=np.bool_),
+                    "last": np.zeros(shape=(3), dtype=np.bool_),
+                    False: np.zeros(shape=(3), dtype=np.bool_),
+                },
+            ),
+            (
+                list("abb"),
+                list("abc"),
+                {
+                    "first": np.array([False, False, True]),
+                    "last": np.array([False, True, False]),
+                    False: np.array([False, True, True]),
+                },
+            ),
+        ],
+    )
+    def test_drop_duplicates(self, data, categories, expected):
+        idx = CategoricalIndex(data, categories=categories, name="foo")
+        for keep, e in expected.items():
+            tm.assert_numpy_array_equal(idx.duplicated(keep=keep), e)
+            e = idx[~e]
+            result = idx.drop_duplicates(keep=keep)
+            tm.assert_index_equal(result, e)
+
+    @pytest.mark.parametrize(
+        "data, categories, expected_data",
+        [
+            ([1, 1, 1], [1, 2, 3], [1]),
+            ([1, 1, 1], list("abc"), [np.nan]),
+            ([1, 2, "a"], [1, 2, 3], [1, 2, np.nan]),
+            ([2, "a", "b"], list("abc"), [np.nan, "a", "b"]),
+        ],
+    )
+    def test_unique(self, data, categories, expected_data, ordered):
+        dtype = CategoricalDtype(categories, ordered=ordered)
+
+        idx = CategoricalIndex(data, dtype=dtype)
+        expected = CategoricalIndex(expected_data, dtype=dtype)
+        tm.assert_index_equal(idx.unique(), expected)
+
+    @pytest.mark.xfail(using_string_dtype(), reason="repr doesn't roundtrip")
+    def test_repr_roundtrip(self):
+        ci = CategoricalIndex(["a", "b"], categories=["a", "b"], ordered=True)
+        str(ci)
+        tm.assert_index_equal(eval(repr(ci)), ci, exact=True)
+
+        # formatting
+        str(ci)
+
+        # long format
+        # this is not reprable
+        ci = CategoricalIndex(np.random.default_rng(2).integers(0, 5, size=100))
+        str(ci)
+
+    def test_isin(self):
+        ci = CategoricalIndex(list("aabca") + [np.nan], categories=["c", "a", "b"])
+        tm.assert_numpy_array_equal(
+            ci.isin(["c"]), np.array([False, False, False, True, False, False])
+        )
+        tm.assert_numpy_array_equal(
+            ci.isin(["c", "a", "b"]), np.array([True] * 5 + [False])
+        )
+        tm.assert_numpy_array_equal(
+            ci.isin(["c", "a", "b", np.nan]), np.array([True] * 6)
+        )
+
+        # mismatched categorical -> coerced to ndarray so doesn't matter
+        result = ci.isin(ci.set_categories(list("abcdefghi")))
+        expected = np.array([True] * 6)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = ci.isin(ci.set_categories(list("defghi")))
+        expected = np.array([False] * 5 + [True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_isin_overlapping_intervals(self):
+        # GH 34974
+        idx = pd.IntervalIndex([pd.Interval(0, 2), pd.Interval(0, 1)])
+        result = CategoricalIndex(idx).isin(idx)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_identical(self):
+        ci1 = CategoricalIndex(["a", "b"], categories=["a", "b"], ordered=True)
+        ci2 = CategoricalIndex(["a", "b"], categories=["a", "b", "c"], ordered=True)
+        assert ci1.identical(ci1)
+        assert ci1.identical(ci1.copy())
+        assert not ci1.identical(ci2)
+
+    def test_ensure_copied_data(self):
+        # gh-12309: Check the "copy" argument of each
+        # Index.__new__ is honored.
+        #
+        # Must be tested separately from other indexes because
+        # self.values is not an ndarray.
+        index = CategoricalIndex(list("ab") * 5)
+
+        result = CategoricalIndex(index.values, copy=True)
+        tm.assert_index_equal(index, result)
+        assert not np.shares_memory(result._data._codes, index._data._codes)
+
+        result = CategoricalIndex(index.values, copy=False)
+        assert result._data._codes is index._data._codes
+
+
+class TestCategoricalIndex2:
+    def test_view_i8(self):
+        # GH#25464
+        ci = CategoricalIndex(list("ab") * 50)
+        msg = "When changing to a larger dtype, its size must be a divisor"
+        with pytest.raises(ValueError, match=msg):
+            ci.view("i8")
+        with pytest.raises(ValueError, match=msg):
+            ci._data.view("i8")
+
+        ci = ci[:-4]  # length divisible by 8
+
+        res = ci.view("i8")
+        expected = ci._data.codes.view("i8")
+        tm.assert_numpy_array_equal(res, expected)
+
+        cat = ci._data
+        tm.assert_numpy_array_equal(cat.view("i8"), expected)
+
+    @pytest.mark.parametrize(
+        "dtype, engine_type",
+        [
+            (np.int8, libindex.Int8Engine),
+            (np.int16, libindex.Int16Engine),
+            (np.int32, libindex.Int32Engine),
+            (np.int64, libindex.Int64Engine),
+        ],
+    )
+    def test_engine_type(self, dtype, engine_type):
+        if dtype != np.int64:
+            # num. of uniques required to push CategoricalIndex.codes to a
+            # dtype (128 categories required for .codes dtype to be int16 etc.)
+            num_uniques = {np.int8: 1, np.int16: 128, np.int32: 32768}[dtype]
+            ci = CategoricalIndex(range(num_uniques))
+        else:
+            # having 2**32 - 2**31 categories would be very memory-intensive,
+            # so we cheat a bit with the dtype
+            ci = CategoricalIndex(range(32768))  # == 2**16 - 2**(16 - 1)
+            arr = ci.values._ndarray.astype("int64")
+            NDArrayBacked.__init__(ci._data, arr, ci.dtype)
+        assert np.issubdtype(ci.codes.dtype, dtype)
+        assert isinstance(ci._engine, engine_type)
+
+    @pytest.mark.parametrize(
+        "func,op_name",
+        [
+            (lambda idx: idx - idx, "__sub__"),
+            (lambda idx: idx + idx, "__add__"),
+            (lambda idx: idx - ["a", "b"], "__sub__"),
+            (lambda idx: idx + ["a", "b"], "__add__"),
+            (lambda idx: ["a", "b"] - idx, "__rsub__"),
+            (lambda idx: ["a", "b"] + idx, "__radd__"),
+        ],
+    )
+    def test_disallow_addsub_ops(self, func, op_name):
+        # GH 10039
+        # set ops (+/-) raise TypeError
+        idx = Index(Categorical(["a", "b"]))
+        cat_or_list = "'(Categorical|list)' and '(Categorical|list)'"
+        msg = "|".join(
+            [
+                f"cannot perform {op_name} with this index type: CategoricalIndex",
+                "can only concatenate list",
+                rf"unsupported operand type\(s\) for [\+-]: {cat_or_list}",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            func(idx)
+
+    def test_method_delegation(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("cabdef"))
+        result = ci.set_categories(list("cab"))
+        tm.assert_index_equal(
+            result, CategoricalIndex(list("aabbca"), categories=list("cab"))
+        )
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"))
+        result = ci.rename_categories(list("efg"))
+        tm.assert_index_equal(
+            result, CategoricalIndex(list("ffggef"), categories=list("efg"))
+        )
+
+        # GH18862 (let rename_categories take callables)
+        result = ci.rename_categories(lambda x: x.upper())
+        tm.assert_index_equal(
+            result, CategoricalIndex(list("AABBCA"), categories=list("CAB"))
+        )
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"))
+        result = ci.add_categories(["d"])
+        tm.assert_index_equal(
+            result, CategoricalIndex(list("aabbca"), categories=list("cabd"))
+        )
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"))
+        result = ci.remove_categories(["c"])
+        tm.assert_index_equal(
+            result,
+            CategoricalIndex(list("aabb") + [np.nan] + ["a"], categories=list("ab")),
+        )
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cabdef"))
+        result = ci.as_unordered()
+        tm.assert_index_equal(result, ci)
+
+        ci = CategoricalIndex(list("aabbca"), categories=list("cabdef"))
+        result = ci.as_ordered()
+        tm.assert_index_equal(
+            result,
+            CategoricalIndex(list("aabbca"), categories=list("cabdef"), ordered=True),
+        )
+
+        # invalid
+        msg = "cannot use inplace with CategoricalIndex"
+        with pytest.raises(ValueError, match=msg):
+            ci.set_categories(list("cab"), inplace=True)
+
+    def test_remove_maintains_order(self):
+        ci = CategoricalIndex(list("abcdda"), categories=list("abcd"))
+        result = ci.reorder_categories(["d", "c", "b", "a"], ordered=True)
+        tm.assert_index_equal(
+            result,
+            CategoricalIndex(list("abcdda"), categories=list("dcba"), ordered=True),
+        )
+        result = result.remove_categories(["c"])
+        tm.assert_index_equal(
+            result,
+            CategoricalIndex(
+                ["a", "b", np.nan, "d", "d", "a"], categories=list("dba"), ordered=True
+            ),
+        )
+
+
+def test_contains_rangeindex_categories_no_engine():
+    ci = CategoricalIndex(range(3))
+    assert 2 in ci
+    assert 5 not in ci
+    assert "_engine" not in ci._cache
diff --git a/pandas/pandas/tests/indexes/categorical/test_constructors.py b/pandas/pandas/tests/indexes/categorical/test_constructors.py
new file mode 100644
index 0000000000000..f0c5307fc5c64
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_constructors.py
@@ -0,0 +1,142 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    Index,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalIndexConstructors:
+    def test_construction_disallows_scalar(self):
+        msg = "must be called with a collection of some kind"
+        with pytest.raises(TypeError, match=msg):
+            CategoricalIndex(data=1, categories=list("abcd"), ordered=False)
+        with pytest.raises(TypeError, match=msg):
+            CategoricalIndex(categories=list("abcd"), ordered=False)
+
+    def test_construction(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("abcd"), ordered=False)
+        categories = ci.categories
+
+        result = Index(ci)
+        tm.assert_index_equal(result, ci, exact=True)
+        assert not result.ordered
+
+        result = Index(ci.values)
+        tm.assert_index_equal(result, ci, exact=True)
+        assert not result.ordered
+
+        # empty
+        result = CategoricalIndex([], categories=categories)
+        tm.assert_index_equal(result.categories, Index(categories))
+        tm.assert_numpy_array_equal(result.codes, np.array([], dtype="int8"))
+        assert not result.ordered
+
+        # passing categories
+        result = CategoricalIndex(list("aabbca"), categories=categories)
+        tm.assert_index_equal(result.categories, Index(categories))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, 2, 0], dtype="int8")
+        )
+
+        c = Categorical(list("aabbca"))
+        result = CategoricalIndex(c)
+        tm.assert_index_equal(result.categories, Index(list("abc")))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, 2, 0], dtype="int8")
+        )
+        assert not result.ordered
+
+        result = CategoricalIndex(c, categories=categories)
+        tm.assert_index_equal(result.categories, Index(categories))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, 2, 0], dtype="int8")
+        )
+        assert not result.ordered
+
+        ci = CategoricalIndex(c, categories=list("abcd"))
+        result = CategoricalIndex(ci)
+        tm.assert_index_equal(result.categories, Index(categories))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, 2, 0], dtype="int8")
+        )
+        assert not result.ordered
+
+        result = CategoricalIndex(ci, categories=list("ab"))
+        tm.assert_index_equal(result.categories, Index(list("ab")))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, -1, 0], dtype="int8")
+        )
+        assert not result.ordered
+
+        result = CategoricalIndex(ci, categories=list("ab"), ordered=True)
+        tm.assert_index_equal(result.categories, Index(list("ab")))
+        tm.assert_numpy_array_equal(
+            result.codes, np.array([0, 0, 1, 1, -1, 0], dtype="int8")
+        )
+        assert result.ordered
+
+        result = CategoricalIndex(ci, categories=list("ab"), ordered=True)
+        expected = CategoricalIndex(
+            ci, categories=list("ab"), ordered=True, dtype="category"
+        )
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # turn me to an Index
+        result = Index(np.array(ci))
+        assert isinstance(result, Index)
+        assert not isinstance(result, CategoricalIndex)
+
+    def test_construction_with_dtype(self):
+        # specify dtype
+        ci = CategoricalIndex(list("aabbca"), categories=list("abc"), ordered=False)
+
+        result = Index(np.array(ci), dtype="category")
+        tm.assert_index_equal(result, ci, exact=True)
+
+        result = Index(np.array(ci).tolist(), dtype="category")
+        tm.assert_index_equal(result, ci, exact=True)
+
+        # these are generally only equal when the categories are reordered
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+
+        result = Index(np.array(ci), dtype="category").reorder_categories(ci.categories)
+        tm.assert_index_equal(result, ci, exact=True)
+
+        # make sure indexes are handled
+        idx = Index(range(3))
+        expected = CategoricalIndex([0, 1, 2], categories=idx, ordered=True)
+        result = CategoricalIndex(idx, categories=idx, ordered=True)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_construction_empty_with_bool_categories(self):
+        # see GH#22702
+        cat = CategoricalIndex([], categories=[True, False])
+        categories = sorted(cat.categories.tolist())
+        assert categories == [False, True]
+
+    def test_construction_with_categorical_dtype(self):
+        # construction with CategoricalDtype
+        # GH#18109
+        data, cats, ordered = "a a b b".split(), "c b a".split(), True
+        dtype = CategoricalDtype(categories=cats, ordered=ordered)
+
+        result = CategoricalIndex(data, dtype=dtype)
+        expected = CategoricalIndex(data, categories=cats, ordered=ordered)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # GH#19032
+        result = Index(data, dtype=dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # error when combining categories/ordered and dtype kwargs
+        msg = "Cannot specify `categories` or `ordered` together with `dtype`."
+        with pytest.raises(ValueError, match=msg):
+            CategoricalIndex(data, categories=cats, dtype=dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            CategoricalIndex(data, ordered=ordered, dtype=dtype)
diff --git a/pandas/pandas/tests/indexes/categorical/test_equals.py b/pandas/pandas/tests/indexes/categorical/test_equals.py
new file mode 100644
index 0000000000000..a8353f301a3c3
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_equals.py
@@ -0,0 +1,96 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    Index,
+    MultiIndex,
+)
+
+
+class TestEquals:
+    def test_equals_categorical(self):
+        ci1 = CategoricalIndex(["a", "b"], categories=["a", "b"], ordered=True)
+        ci2 = CategoricalIndex(["a", "b"], categories=["a", "b", "c"], ordered=True)
+
+        assert ci1.equals(ci1)
+        assert not ci1.equals(ci2)
+        assert ci1.equals(ci1.astype(object))
+        assert ci1.astype(object).equals(ci1)
+
+        assert (ci1 == ci1).all()
+        assert not (ci1 != ci1).all()
+        assert not (ci1 > ci1).all()
+        assert not (ci1 < ci1).all()
+        assert (ci1 <= ci1).all()
+        assert (ci1 >= ci1).all()
+
+        assert not (ci1 == 1).all()
+        assert (ci1 == Index(["a", "b"])).all()
+        assert (ci1 == ci1.values).all()
+
+        # invalid comparisons
+        with pytest.raises(ValueError, match="Lengths must match"):
+            ci1 == Index(["a", "b", "c"])
+
+        msg = "Categoricals can only be compared if 'categories' are the same"
+        with pytest.raises(TypeError, match=msg):
+            ci1 == ci2
+        with pytest.raises(TypeError, match=msg):
+            ci1 == Categorical(ci1.values, ordered=False)
+        with pytest.raises(TypeError, match=msg):
+            ci1 == Categorical(ci1.values, categories=list("abc"))
+
+        # tests
+        # make sure that we are testing for category inclusion properly
+        ci = CategoricalIndex(list("aabca"), categories=["c", "a", "b"])
+        assert not ci.equals(list("aabca"))
+        # Same categories, but different order
+        # Unordered
+        assert ci.equals(CategoricalIndex(list("aabca")))
+        # Ordered
+        assert not ci.equals(CategoricalIndex(list("aabca"), ordered=True))
+        assert ci.equals(ci.copy())
+
+        ci = CategoricalIndex(list("aabca") + [np.nan], categories=["c", "a", "b"])
+        assert not ci.equals(list("aabca"))
+        assert not ci.equals(CategoricalIndex(list("aabca")))
+        assert ci.equals(ci.copy())
+
+        ci = CategoricalIndex(list("aabca") + [np.nan], categories=["c", "a", "b"])
+        assert not ci.equals(list("aabca") + [np.nan])
+        assert ci.equals(CategoricalIndex(list("aabca") + [np.nan]))
+        assert not ci.equals(CategoricalIndex(list("aabca") + [np.nan], ordered=True))
+        assert ci.equals(ci.copy())
+
+    def test_equals_categorical_unordered(self):
+        # https://github.com/pandas-dev/pandas/issues/16603
+        a = CategoricalIndex(["A"], categories=["A", "B"])
+        b = CategoricalIndex(["A"], categories=["B", "A"])
+        c = CategoricalIndex(["C"], categories=["B", "A"])
+        assert a.equals(b)
+        assert not a.equals(c)
+        assert not b.equals(c)
+
+    def test_equals_non_category(self):
+        # GH#37667 Case where other contains a value not among ci's
+        #  categories ("D") and also contains np.nan
+        ci = CategoricalIndex(["A", "B", np.nan, np.nan])
+        other = Index(["A", "B", "D", np.nan])
+
+        assert not ci.equals(other)
+
+    def test_equals_multiindex(self):
+        # dont raise NotImplementedError when calling is_dtype_compat
+
+        mi = MultiIndex.from_arrays([["A", "B", "C", "D"], range(4)])
+        ci = mi.to_flat_index().astype("category")
+
+        assert not ci.equals(mi)
+
+    def test_equals_string_dtype(self, any_string_dtype):
+        # GH#55364
+        idx = CategoricalIndex(list("abc"), name="B")
+        other = Index(["a", "b", "c"], name="B", dtype=any_string_dtype)
+        assert idx.equals(other)
diff --git a/pandas/pandas/tests/indexes/categorical/test_fillna.py b/pandas/pandas/tests/indexes/categorical/test_fillna.py
new file mode 100644
index 0000000000000..09de578f3c649
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_fillna.py
@@ -0,0 +1,54 @@
+import numpy as np
+import pytest
+
+from pandas import CategoricalIndex
+import pandas._testing as tm
+
+
+class TestFillNA:
+    def test_fillna_categorical(self):
+        # GH#11343
+        idx = CategoricalIndex([1.0, np.nan, 3.0, 1.0], name="x")
+        # fill by value in categories
+        exp = CategoricalIndex([1.0, 1.0, 3.0, 1.0], name="x")
+        tm.assert_index_equal(idx.fillna(1.0), exp)
+
+        cat = idx._data
+
+        # fill by value not in categories raises TypeError on EA, casts on CI
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            cat.fillna(2.0)
+
+        result = idx.fillna(2.0)
+        expected = idx.astype(object).fillna(2.0)
+        tm.assert_index_equal(result, expected)
+
+    def test_fillna_copies_with_no_nas(self):
+        # Nothing to fill, should still get a copy for the Categorical method,
+        #  but OK to get a view on CategoricalIndex method
+        ci = CategoricalIndex([0, 1, 1])
+        result = ci.fillna(0)
+        assert result is not ci
+        assert tm.shares_memory(result, ci)
+
+        # But at the EA level we always get a copy.
+        cat = ci._data
+        result = cat.fillna(0)
+        assert result._ndarray is not cat._ndarray
+        assert result._ndarray.base is None
+        assert not tm.shares_memory(result, cat)
+
+    def test_fillna_validates_with_no_nas(self):
+        # We validate the fill value even if fillna is a no-op
+        ci = CategoricalIndex([2, 3, 3])
+        cat = ci._data
+
+        msg = "Cannot setitem on a Categorical with a new category"
+        res = ci.fillna(False)
+        # nothing to fill, so we dont cast
+        tm.assert_index_equal(res, ci)
+
+        # Same check directly on the Categorical
+        with pytest.raises(TypeError, match=msg):
+            cat.fillna(False)
diff --git a/pandas/pandas/tests/indexes/categorical/test_formats.py b/pandas/pandas/tests/indexes/categorical/test_formats.py
new file mode 100644
index 0000000000000..b1361b3e8106e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_formats.py
@@ -0,0 +1,112 @@
+"""
+Tests for CategoricalIndex.__repr__ and related methods.
+"""
+
+import pytest
+
+from pandas._config import using_string_dtype
+import pandas._config.config as cf
+
+from pandas import CategoricalIndex
+
+
+class TestCategoricalIndexRepr:
+    @pytest.mark.xfail(using_string_dtype(), reason="repr different")
+    def test_string_categorical_index_repr(self):
+        # short
+        idx = CategoricalIndex(["a", "bb", "ccc"])
+        expected = """CategoricalIndex(['a', 'bb', 'ccc'], categories=['a', 'bb', 'ccc'], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(idx) == expected
+
+        # multiple lines
+        idx = CategoricalIndex(["a", "bb", "ccc"] * 10)
+        expected = """CategoricalIndex(['a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a',
+                  'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb',
+                  'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc'],
+                 categories=['a', 'bb', 'ccc'], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # truncated
+        idx = CategoricalIndex(["a", "bb", "ccc"] * 100)
+        expected = """CategoricalIndex(['a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a',
+                  ...
+                  'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc', 'a', 'bb', 'ccc'],
+                 categories=['a', 'bb', 'ccc'], ordered=False, dtype='category', length=300)"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # larger categories
+        idx = CategoricalIndex(list("abcdefghijklmmo"))
+        expected = """CategoricalIndex(['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l',
+                  'm', 'm', 'o'],
+                 categories=['a', 'b', 'c', 'd', ..., 'k', 'l', 'm', 'o'], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # short
+        idx = CategoricalIndex(["あ", "いい", "ううう"])
+        expected = """CategoricalIndex(['あ', 'いい', 'ううう'], categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category')"""  # noqa: E501
+        assert repr(idx) == expected
+
+        # multiple lines
+        idx = CategoricalIndex(["あ", "いい", "ううう"] * 10)
+        expected = """CategoricalIndex(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ',
+                  'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい',
+                  'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう'],
+                 categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # truncated
+        idx = CategoricalIndex(["あ", "いい", "ううう"] * 100)
+        expected = """CategoricalIndex(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ',
+                  ...
+                  'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう'],
+                 categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category', length=300)"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # larger categories
+        idx = CategoricalIndex(list("あいうえおかきくけこさしすせそ"))
+        expected = """CategoricalIndex(['あ', 'い', 'う', 'え', 'お', 'か', 'き', 'く', 'け', 'こ', 'さ', 'し',
+                  'す', 'せ', 'そ'],
+                 categories=['あ', 'い', 'う', 'え', ..., 'し', 'す', 'せ', 'そ'], ordered=False, dtype='category')"""  # noqa: E501
+
+        assert repr(idx) == expected
+
+        # Enable Unicode option -----------------------------------------
+        with cf.option_context("display.unicode.east_asian_width", True):
+            # short
+            idx = CategoricalIndex(["あ", "いい", "ううう"])
+            expected = """CategoricalIndex(['あ', 'いい', 'ううう'], categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category')"""  # noqa: E501
+            assert repr(idx) == expected
+
+            # multiple lines
+            idx = CategoricalIndex(["あ", "いい", "ううう"] * 10)
+            expected = """CategoricalIndex(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい',
+                  'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう',
+                  'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい',
+                  'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう'],
+                 categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category')"""  # noqa: E501
+
+            assert repr(idx) == expected
+
+            # truncated
+            idx = CategoricalIndex(["あ", "いい", "ううう"] * 100)
+            expected = """CategoricalIndex(['あ', 'いい', 'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい',
+                  'ううう', 'あ',
+                  ...
+                  'ううう', 'あ', 'いい', 'ううう', 'あ', 'いい', 'ううう',
+                  'あ', 'いい', 'ううう'],
+                 categories=['あ', 'いい', 'ううう'], ordered=False, dtype='category', length=300)"""  # noqa: E501
+
+            assert repr(idx) == expected
+
+            # larger categories
+            idx = CategoricalIndex(list("あいうえおかきくけこさしすせそ"))
+            expected = """CategoricalIndex(['あ', 'い', 'う', 'え', 'お', 'か', 'き', 'く', 'け', 'こ',
+                  'さ', 'し', 'す', 'せ', 'そ'],
+                 categories=['あ', 'い', 'う', 'え', ..., 'し', 'す', 'せ', 'そ'], ordered=False, dtype='category')"""  # noqa: E501
+
+            assert repr(idx) == expected
diff --git a/pandas/pandas/tests/indexes/categorical/test_indexing.py b/pandas/pandas/tests/indexes/categorical/test_indexing.py
new file mode 100644
index 0000000000000..25232075a07d9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_indexing.py
@@ -0,0 +1,418 @@
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    Index,
+    IntervalIndex,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestTake:
+    def test_take_fill_value(self):
+        # GH 12631
+
+        # numeric category
+        idx = CategoricalIndex([1, 2, 3], name="xxx")
+        result = idx.take(np.array([1, 0, -1]))
+        expected = CategoricalIndex([2, 1, 3], name="xxx")
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = CategoricalIndex([2, 1, np.nan], categories=[1, 2, 3], name="xxx")
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = CategoricalIndex([2, 1, 3], name="xxx")
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        # object category
+        idx = CategoricalIndex(
+            list("CBA"), categories=list("ABC"), ordered=True, name="xxx"
+        )
+        result = idx.take(np.array([1, 0, -1]))
+        expected = CategoricalIndex(
+            list("BCA"), categories=list("ABC"), ordered=True, name="xxx"
+        )
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = CategoricalIndex(
+            ["B", "C", np.nan], categories=list("ABC"), ordered=True, name="xxx"
+        )
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = CategoricalIndex(
+            list("BCA"), categories=list("ABC"), ordered=True, name="xxx"
+        )
+        tm.assert_index_equal(result, expected)
+        tm.assert_categorical_equal(result.values, expected.values)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+    def test_take_fill_value_datetime(self):
+        # datetime category
+        idx = pd.DatetimeIndex(["2011-01-01", "2011-02-01", "2011-03-01"], name="xxx")
+        idx = CategoricalIndex(idx)
+        result = idx.take(np.array([1, 0, -1]))
+        expected = pd.DatetimeIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx"
+        )
+        expected = CategoricalIndex(expected)
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = pd.DatetimeIndex(["2011-02-01", "2011-01-01", "NaT"], name="xxx")
+        exp_cats = pd.DatetimeIndex(["2011-01-01", "2011-02-01", "2011-03-01"])
+        expected = CategoricalIndex(expected, categories=exp_cats)
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = pd.DatetimeIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx"
+        )
+        expected = CategoricalIndex(expected)
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+    def test_take_invalid_kwargs(self):
+        idx = CategoricalIndex([1, 2, 3], name="foo")
+        indices = [1, 0, -1]
+
+        msg = r"take\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            idx.take(indices, foo=2)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, out=indices)
+
+        msg = "the 'mode' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, mode="clip")
+
+
+class TestGetLoc:
+    def test_get_loc(self):
+        # GH 12531
+        cidx1 = CategoricalIndex(list("abcde"), categories=list("edabc"))
+        idx1 = Index(list("abcde"))
+        assert cidx1.get_loc("a") == idx1.get_loc("a")
+        assert cidx1.get_loc("e") == idx1.get_loc("e")
+
+        for i in [cidx1, idx1]:
+            with pytest.raises(KeyError, match="'NOT-EXIST'"):
+                i.get_loc("NOT-EXIST")
+
+        # non-unique
+        cidx2 = CategoricalIndex(list("aacded"), categories=list("edabc"))
+        idx2 = Index(list("aacded"))
+
+        # results in bool array
+        res = cidx2.get_loc("d")
+        tm.assert_numpy_array_equal(res, idx2.get_loc("d"))
+        tm.assert_numpy_array_equal(
+            res, np.array([False, False, False, True, False, True])
+        )
+        # unique element results in scalar
+        res = cidx2.get_loc("e")
+        assert res == idx2.get_loc("e")
+        assert res == 4
+
+        for i in [cidx2, idx2]:
+            with pytest.raises(KeyError, match="'NOT-EXIST'"):
+                i.get_loc("NOT-EXIST")
+
+        # non-unique, sliceable
+        cidx3 = CategoricalIndex(list("aabbb"), categories=list("abc"))
+        idx3 = Index(list("aabbb"))
+
+        # results in slice
+        res = cidx3.get_loc("a")
+        assert res == idx3.get_loc("a")
+        assert res == slice(0, 2, None)
+
+        res = cidx3.get_loc("b")
+        assert res == idx3.get_loc("b")
+        assert res == slice(2, 5, None)
+
+        for i in [cidx3, idx3]:
+            with pytest.raises(KeyError, match="'c'"):
+                i.get_loc("c")
+
+    def test_get_loc_unique(self):
+        cidx = CategoricalIndex(list("abc"))
+        result = cidx.get_loc("b")
+        assert result == 1
+
+    def test_get_loc_monotonic_nonunique(self):
+        cidx = CategoricalIndex(list("abbc"))
+        result = cidx.get_loc("b")
+        expected = slice(1, 3, None)
+        assert result == expected
+
+    def test_get_loc_nonmonotonic_nonunique(self):
+        cidx = CategoricalIndex(list("abcb"))
+        result = cidx.get_loc("b")
+        expected = np.array([False, True, False, True], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_loc_nan(self):
+        # GH#41933
+        ci = CategoricalIndex(["A", "B", np.nan])
+        res = ci.get_loc(np.nan)
+
+        assert res == 2
+
+
+class TestGetIndexer:
+    def test_get_indexer_base(self):
+        # Determined by cat ordering.
+        idx = CategoricalIndex(list("cab"), categories=list("cab"))
+        expected = np.arange(len(idx), dtype=np.intp)
+
+        actual = idx.get_indexer(idx)
+        tm.assert_numpy_array_equal(expected, actual)
+
+        with pytest.raises(ValueError, match="Invalid fill method"):
+            idx.get_indexer(idx, method="invalid")
+
+    def test_get_indexer_requires_unique(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+        oidx = Index(np.array(ci))
+
+        msg = "Reindexing only valid with uniquely valued Index objects"
+
+        for n in [1, 2, 5, len(ci)]:
+            finder = oidx[np.random.default_rng(2).integers(0, len(ci), size=n)]
+
+            with pytest.raises(InvalidIndexError, match=msg):
+                ci.get_indexer(finder)
+
+        # see gh-17323
+        #
+        # Even when indexer is equal to the
+        # members in the index, we should
+        # respect duplicates instead of taking
+        # the fast-track path.
+        for finder in [list("aabbca"), list("aababca")]:
+            with pytest.raises(InvalidIndexError, match=msg):
+                ci.get_indexer(finder)
+
+    def test_get_indexer_non_unique(self):
+        idx1 = CategoricalIndex(list("aabcde"), categories=list("edabc"))
+        idx2 = CategoricalIndex(list("abf"))
+
+        for indexer in [idx2, list("abf"), Index(list("abf"))]:
+            msg = "Reindexing only valid with uniquely valued Index objects"
+            with pytest.raises(InvalidIndexError, match=msg):
+                idx1.get_indexer(indexer)
+
+            r1, _ = idx1.get_indexer_non_unique(indexer)
+            expected = np.array([0, 1, 2, -1], dtype=np.intp)
+            tm.assert_almost_equal(r1, expected)
+
+    def test_get_indexer_method(self):
+        idx1 = CategoricalIndex(list("aabcde"), categories=list("edabc"))
+        idx2 = CategoricalIndex(list("abf"))
+
+        msg = "method pad not yet implemented for CategoricalIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            idx2.get_indexer(idx1, method="pad")
+        msg = "method backfill not yet implemented for CategoricalIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            idx2.get_indexer(idx1, method="backfill")
+
+        msg = "method nearest not yet implemented for CategoricalIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            idx2.get_indexer(idx1, method="nearest")
+
+    def test_get_indexer_array(self):
+        arr = np.array(
+            [Timestamp("1999-12-31 00:00:00"), Timestamp("2000-12-31 00:00:00")],
+            dtype=object,
+        )
+        cats = [Timestamp("1999-12-31 00:00:00"), Timestamp("2000-12-31 00:00:00")]
+        ci = CategoricalIndex(cats, categories=cats, ordered=False, dtype="category")
+        result = ci.get_indexer(arr)
+        expected = np.array([0, 1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_same_categories_same_order(self):
+        ci = CategoricalIndex(["a", "b"], categories=["a", "b"])
+
+        result = ci.get_indexer(CategoricalIndex(["b", "b"], categories=["a", "b"]))
+        expected = np.array([1, 1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_same_categories_different_order(self):
+        # https://github.com/pandas-dev/pandas/issues/19551
+        ci = CategoricalIndex(["a", "b"], categories=["a", "b"])
+
+        result = ci.get_indexer(CategoricalIndex(["b", "b"], categories=["b", "a"]))
+        expected = np.array([1, 1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_nans_in_index_and_target(self):
+        # GH 45361
+        ci = CategoricalIndex([1, 2, np.nan, 3])
+        other1 = [2, 3, 4, np.nan]
+        res1 = ci.get_indexer(other1)
+        expected1 = np.array([1, 3, -1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(res1, expected1)
+        other2 = [1, 4, 2, 3]
+        res2 = ci.get_indexer(other2)
+        expected2 = np.array([0, -1, 1, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(res2, expected2)
+
+
+class TestWhere:
+    def test_where(self, listlike_box):
+        klass = listlike_box
+
+        i = CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+        cond = [True] * len(i)
+        expected = i
+        result = i.where(klass(cond))
+        tm.assert_index_equal(result, expected)
+
+        cond = [False] + [True] * (len(i) - 1)
+        expected = CategoricalIndex([np.nan] + i[1:].tolist(), categories=i.categories)
+        result = i.where(klass(cond))
+        tm.assert_index_equal(result, expected)
+
+    def test_where_non_categories(self):
+        ci = CategoricalIndex(["a", "b", "c", "d"])
+        mask = np.array([True, False, True, False])
+
+        result = ci.where(mask, 2)
+        expected = Index(["a", 2, "c", 2], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            # Test the Categorical method directly
+            ci._data._where(mask, 2)
+
+
+class TestContains:
+    def test_contains(self):
+        ci = CategoricalIndex(list("aabbca"), categories=list("cabdef"), ordered=False)
+
+        assert "a" in ci
+        assert "z" not in ci
+        assert "e" not in ci
+        assert np.nan not in ci
+
+        # assert codes NOT in index
+        assert 0 not in ci
+        assert 1 not in ci
+
+    def test_contains_nan(self):
+        ci = CategoricalIndex(list("aabbca") + [np.nan], categories=list("cabdef"))
+        assert np.nan in ci
+
+    @pytest.mark.parametrize("unwrap", [True, False])
+    def test_contains_na_dtype(self, unwrap):
+        dti = pd.date_range("2016-01-01", periods=100).insert(0, pd.NaT)
+        pi = dti.to_period("D")
+        tdi = dti - dti[-1]
+        ci = CategoricalIndex(dti)
+
+        obj = ci
+        if unwrap:
+            obj = ci._data
+
+        assert np.nan in obj
+        assert None in obj
+        assert pd.NaT in obj
+        assert np.datetime64("NaT") in obj
+        assert np.timedelta64("NaT") not in obj
+
+        obj2 = CategoricalIndex(tdi)
+        if unwrap:
+            obj2 = obj2._data
+
+        assert np.nan in obj2
+        assert None in obj2
+        assert pd.NaT in obj2
+        assert np.datetime64("NaT") not in obj2
+        assert np.timedelta64("NaT") in obj2
+
+        obj3 = CategoricalIndex(pi)
+        if unwrap:
+            obj3 = obj3._data
+
+        assert np.nan in obj3
+        assert None in obj3
+        assert pd.NaT in obj3
+        assert np.datetime64("NaT") not in obj3
+        assert np.timedelta64("NaT") not in obj3
+
+    @pytest.mark.parametrize(
+        "item, expected",
+        [
+            (pd.Interval(0, 1), True),
+            (1.5, True),
+            (pd.Interval(0.5, 1.5), False),
+            ("a", False),
+            (Timestamp(1), False),
+            (pd.Timedelta(1), False),
+        ],
+        ids=str,
+    )
+    def test_contains_interval(self, item, expected):
+        # GH 23705
+        ci = CategoricalIndex(IntervalIndex.from_breaks(range(3)))
+        result = item in ci
+        assert result is expected
+
+    def test_contains_list(self):
+        # GH#21729
+        idx = CategoricalIndex([1, 2, 3])
+
+        assert "a" not in idx
+
+        with pytest.raises(TypeError, match="unhashable type"):
+            ["a"] in idx
+
+        with pytest.raises(TypeError, match="unhashable type"):
+            ["a", "b"] in idx
diff --git a/pandas/pandas/tests/indexes/categorical/test_map.py b/pandas/pandas/tests/indexes/categorical/test_map.py
new file mode 100644
index 0000000000000..baf836594dfb5
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_map.py
@@ -0,0 +1,144 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "data, categories",
+    [
+        (list("abcbca"), list("cab")),
+        (pd.interval_range(0, 3).repeat(3), pd.interval_range(0, 3)),
+    ],
+    ids=["string", "interval"],
+)
+def test_map_str(data, categories, ordered):
+    # GH 31202 - override base class since we want to maintain categorical/ordered
+    index = CategoricalIndex(data, categories=categories, ordered=ordered)
+    result = index.map(str)
+    expected = CategoricalIndex(
+        map(str, data), categories=map(str, categories), ordered=ordered
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_map():
+    ci = CategoricalIndex(list("ABABC"), categories=list("CBA"), ordered=True)
+    result = ci.map(lambda x: x.lower())
+    exp = CategoricalIndex(list("ababc"), categories=list("cba"), ordered=True)
+    tm.assert_index_equal(result, exp)
+
+    ci = CategoricalIndex(
+        list("ABABC"), categories=list("BAC"), ordered=False, name="XXX"
+    )
+    result = ci.map(lambda x: x.lower())
+    exp = CategoricalIndex(
+        list("ababc"), categories=list("bac"), ordered=False, name="XXX"
+    )
+    tm.assert_index_equal(result, exp)
+
+    # GH 12766: Return an index not an array
+    tm.assert_index_equal(
+        ci.map(lambda x: 1), Index(np.array([1] * 5, dtype=np.int64), name="XXX")
+    )
+
+    # change categories dtype
+    ci = CategoricalIndex(list("ABABC"), categories=list("BAC"), ordered=False)
+
+    def f(x):
+        return {"A": 10, "B": 20, "C": 30}.get(x)
+
+    result = ci.map(f)
+    exp = CategoricalIndex([10, 20, 10, 20, 30], categories=[20, 10, 30], ordered=False)
+    tm.assert_index_equal(result, exp)
+
+    result = ci.map(Series([10, 20, 30], index=["A", "B", "C"]))
+    tm.assert_index_equal(result, exp)
+
+    result = ci.map({"A": 10, "B": 20, "C": 30})
+    tm.assert_index_equal(result, exp)
+
+
+def test_map_with_categorical_series():
+    # GH 12756
+    a = Index([1, 2, 3, 4])
+    b = Series(["even", "odd", "even", "odd"], dtype="category")
+    c = Series(["even", "odd", "even", "odd"])
+
+    exp = CategoricalIndex(["odd", "even", "odd", np.nan])
+    tm.assert_index_equal(a.map(b), exp)
+    exp = Index(["odd", "even", "odd", np.nan])
+    tm.assert_index_equal(a.map(c), exp)
+
+
+@pytest.mark.parametrize(
+    ("data", "f", "expected"),
+    (
+        ([1, 1, np.nan], pd.isna, CategoricalIndex([False, False, np.nan])),
+        ([1, 2, np.nan], pd.isna, Index([False, False, np.nan])),
+        ([1, 1, np.nan], {1: False}, CategoricalIndex([False, False, np.nan])),
+        ([1, 2, np.nan], {1: False, 2: False}, Index([False, False, np.nan])),
+        (
+            [1, 1, np.nan],
+            Series([False, False]),
+            CategoricalIndex([False, False, np.nan]),
+        ),
+        (
+            [1, 2, np.nan],
+            Series([False, False, False]),
+            Index([False, False, np.nan]),
+        ),
+    ),
+)
+def test_map_with_nan_ignore(data, f, expected):  # GH 24241
+    values = CategoricalIndex(data)
+    result = values.map(f, na_action="ignore")
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("data", "f", "expected"),
+    (
+        ([1, 1, np.nan], pd.isna, Index([False, False, True])),
+        ([1, 2, np.nan], pd.isna, Index([False, False, True])),
+        ([1, 1, np.nan], {1: False}, CategoricalIndex([False, False, np.nan])),
+        ([1, 2, np.nan], {1: False, 2: False}, Index([False, False, np.nan])),
+        (
+            [1, 1, np.nan],
+            Series([False, False]),
+            CategoricalIndex([False, False, np.nan]),
+        ),
+        (
+            [1, 2, np.nan],
+            Series([False, False, False]),
+            Index([False, False, np.nan]),
+        ),
+    ),
+)
+def test_map_with_nan_none(data, f, expected):  # GH 24241
+    values = CategoricalIndex(data)
+    result = values.map(f, na_action=None)
+    tm.assert_index_equal(result, expected)
+
+
+def test_map_with_dict_or_series():
+    orig_values = ["a", "B", 1, "a"]
+    new_values = ["one", 2, 3.0, "one"]
+    cur_index = CategoricalIndex(orig_values, name="XXX")
+    expected = CategoricalIndex(new_values, name="XXX", categories=[3.0, 2, "one"])
+
+    mapper = Series(new_values[:-1], index=orig_values[:-1])
+    result = cur_index.map(mapper)
+    # Order of categories in result can be different
+    tm.assert_index_equal(result, expected)
+
+    mapper = dict(zip(orig_values[:-1], new_values[:-1]))
+    result = cur_index.map(mapper)
+    # Order of categories in result can be different
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/categorical/test_reindex.py b/pandas/pandas/tests/indexes/categorical/test_reindex.py
new file mode 100644
index 0000000000000..5b1f2b9fb159a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_reindex.py
@@ -0,0 +1,78 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    Index,
+    Interval,
+)
+import pandas._testing as tm
+
+
+class TestReindex:
+    def test_reindex_list_non_unique(self):
+        # GH#11586
+        msg = "cannot reindex on an axis with duplicate labels"
+        ci = CategoricalIndex(["a", "b", "c", "a"])
+        with pytest.raises(ValueError, match=msg):
+            ci.reindex(["a", "c"])
+
+    def test_reindex_categorical_non_unique(self):
+        msg = "cannot reindex on an axis with duplicate labels"
+        ci = CategoricalIndex(["a", "b", "c", "a"])
+        with pytest.raises(ValueError, match=msg):
+            ci.reindex(Categorical(["a", "c"]))
+
+    def test_reindex_list_non_unique_unused_category(self):
+        msg = "cannot reindex on an axis with duplicate labels"
+        ci = CategoricalIndex(["a", "b", "c", "a"], categories=["a", "b", "c", "d"])
+        with pytest.raises(ValueError, match=msg):
+            ci.reindex(["a", "c"])
+
+    def test_reindex_categorical_non_unique_unused_category(self):
+        msg = "cannot reindex on an axis with duplicate labels"
+        ci = CategoricalIndex(["a", "b", "c", "a"], categories=["a", "b", "c", "d"])
+        with pytest.raises(ValueError, match=msg):
+            ci.reindex(Categorical(["a", "c"]))
+
+    def test_reindex_duplicate_target(self):
+        # See GH25459
+        cat = CategoricalIndex(["a", "b", "c"], categories=["a", "b", "c", "d"])
+        res, indexer = cat.reindex(["a", "c", "c"])
+        exp = Index(["a", "c", "c"])
+        tm.assert_index_equal(res, exp, exact=True)
+        tm.assert_numpy_array_equal(indexer, np.array([0, 2, 2], dtype=np.intp))
+
+        res, indexer = cat.reindex(
+            CategoricalIndex(["a", "c", "c"], categories=["a", "b", "c", "d"])
+        )
+        exp = CategoricalIndex(["a", "c", "c"], categories=["a", "b", "c", "d"])
+        tm.assert_index_equal(res, exp, exact=True)
+        tm.assert_numpy_array_equal(indexer, np.array([0, 2, 2], dtype=np.intp))
+
+    def test_reindex_empty_index(self):
+        # See GH16770
+        c = CategoricalIndex([])
+        res, indexer = c.reindex(["a", "b"])
+        tm.assert_index_equal(res, Index(["a", "b"]), exact=True)
+        tm.assert_numpy_array_equal(indexer, np.array([-1, -1], dtype=np.intp))
+
+    def test_reindex_categorical_added_category(self):
+        # GH 42424
+        ci = CategoricalIndex(
+            [Interval(0, 1, closed="right"), Interval(1, 2, closed="right")],
+            ordered=True,
+        )
+        ci_add = CategoricalIndex(
+            [
+                Interval(0, 1, closed="right"),
+                Interval(1, 2, closed="right"),
+                Interval(2, 3, closed="right"),
+                Interval(3, 4, closed="right"),
+            ],
+            ordered=True,
+        )
+        result, _ = ci.reindex(ci_add)
+        expected = ci_add
+        tm.assert_index_equal(expected, result)
diff --git a/pandas/pandas/tests/indexes/categorical/test_setops.py b/pandas/pandas/tests/indexes/categorical/test_setops.py
new file mode 100644
index 0000000000000..2e87b90efd54c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/categorical/test_setops.py
@@ -0,0 +1,18 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    Index,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("na_value", [None, np.nan])
+def test_difference_with_na(na_value):
+    # GH 57318
+    ci = CategoricalIndex(["a", "b", "c", None])
+    other = Index(["c", na_value])
+    result = ci.difference(other)
+    expected = CategoricalIndex(["a", "b"], categories=["a", "b", "c"])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/conftest.py b/pandas/pandas/tests/indexes/conftest.py
new file mode 100644
index 0000000000000..61f6be000cee8
--- /dev/null
+++ b/pandas/pandas/tests/indexes/conftest.py
@@ -0,0 +1,40 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    array,
+)
+
+
+@pytest.fixture(params=[None, False])
+def sort(request):
+    """
+    Valid values for the 'sort' parameter used in the Index
+    setops methods (intersection, union, etc.)
+
+    Caution:
+        Don't confuse this one with the "sort" fixture used
+        for concat. That one has parameters [True, False].
+
+        We can't combine them as sort=True is not permitted
+        in the Index setops methods.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["D", "3D", "-3D", "h", "2h", "-2h", "min", "2min", "s", "-3s"])
+def freq_sample(request):
+    """
+    Valid values for 'freq' parameter used to create date_range and
+    timedelta_range..
+    """
+    return request.param
+
+
+@pytest.fixture(params=[list, tuple, np.array, array, Series])
+def listlike_box(request):
+    """
+    Types that may be passed as the indexer to searchsorted.
+    """
+    return request.param
diff --git a/pandas/pandas/tests/indexes/datetimelike_/__init__.py b/pandas/pandas/tests/indexes/datetimelike_/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_drop_duplicates.py b/pandas/pandas/tests/indexes/datetimelike_/test_drop_duplicates.py
new file mode 100644
index 0000000000000..c2d76c0bcc8bd
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_drop_duplicates.py
@@ -0,0 +1,101 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class DropDuplicates:
+    def test_drop_duplicates_metadata(self, idx):
+        # GH#10115
+        result = idx.drop_duplicates()
+        tm.assert_index_equal(idx, result)
+        assert idx.freq == result.freq
+
+        idx_dup = idx.append(idx)
+        result = idx_dup.drop_duplicates()
+
+        expected = idx
+        if not isinstance(idx, PeriodIndex):
+            # freq is reset except for PeriodIndex
+            assert idx_dup.freq is None
+            assert result.freq is None
+            expected = idx._with_freq(None)
+        else:
+            assert result.freq == expected.freq
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "keep, expected, index",
+        [
+            (
+                "first",
+                np.concatenate(([False] * 10, [True] * 5)),
+                np.arange(0, 10, dtype=np.int64),
+            ),
+            (
+                "last",
+                np.concatenate(([True] * 5, [False] * 10)),
+                np.arange(5, 15, dtype=np.int64),
+            ),
+            (
+                False,
+                np.concatenate(([True] * 5, [False] * 5, [True] * 5)),
+                np.arange(5, 10, dtype=np.int64),
+            ),
+        ],
+    )
+    def test_drop_duplicates(self, keep, expected, index, idx):
+        # to check Index/Series compat
+        idx = idx.append(idx[:5])
+
+        tm.assert_numpy_array_equal(idx.duplicated(keep=keep), expected)
+        expected = idx[~expected]
+
+        result = idx.drop_duplicates(keep=keep)
+        tm.assert_index_equal(result, expected)
+
+        result = Series(idx).drop_duplicates(keep=keep)
+        expected = Series(expected, index=index)
+        tm.assert_series_equal(result, expected)
+
+
+class TestDropDuplicatesPeriodIndex(DropDuplicates):
+    @pytest.fixture(params=["D", "3D", "h", "2h", "min", "2min", "s", "3s"])
+    def freq(self, request):
+        """
+        Fixture to test for different frequencies for PeriodIndex.
+        """
+        return request.param
+
+    @pytest.fixture
+    def idx(self, freq):
+        """
+        Fixture to get PeriodIndex for 10 periods for different frequencies.
+        """
+        return period_range("2011-01-01", periods=10, freq=freq, name="idx")
+
+
+class TestDropDuplicatesDatetimeIndex(DropDuplicates):
+    @pytest.fixture
+    def idx(self, freq_sample):
+        """
+        Fixture to get DatetimeIndex for 10 periods for different frequencies.
+        """
+        return date_range("2011-01-01", freq=freq_sample, periods=10, name="idx")
+
+
+class TestDropDuplicatesTimedeltaIndex(DropDuplicates):
+    @pytest.fixture
+    def idx(self, freq_sample):
+        """
+        Fixture to get TimedeltaIndex for 10 periods for different frequencies.
+        """
+        return timedelta_range("1 day", periods=10, freq=freq_sample, name="idx")
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_equals.py b/pandas/pandas/tests/indexes/datetimelike_/test_equals.py
new file mode 100644
index 0000000000000..df9182b2dd1c2
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_equals.py
@@ -0,0 +1,185 @@
+"""
+Tests shared for DatetimeIndex/TimedeltaIndex/PeriodIndex
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    PeriodIndex,
+    TimedeltaIndex,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class EqualsTests:
+    def test_not_equals_numeric(self, index):
+        assert not index.equals(Index(index.asi8))
+        assert not index.equals(Index(index.asi8.astype("u8")))
+        assert not index.equals(Index(index.asi8).astype("f8"))
+
+    def test_equals(self, index):
+        assert index.equals(index)
+        assert index.equals(index.astype(object))
+        assert index.equals(CategoricalIndex(index))
+        assert index.equals(CategoricalIndex(index.astype(object)))
+
+    def test_not_equals_non_arraylike(self, index):
+        assert not index.equals(list(index))
+
+    def test_not_equals_strings(self, index):
+        other = Index([str(x) for x in index], dtype=object)
+        assert not index.equals(other)
+        assert not index.equals(CategoricalIndex(other))
+
+    def test_not_equals_misc_strs(self, index):
+        other = Index(list("abc"))
+        assert not index.equals(other)
+
+
+class TestPeriodIndexEquals(EqualsTests):
+    @pytest.fixture
+    def index(self):
+        """Fixture for creating a PeriodIndex for use in equality tests."""
+        return period_range("2013-01-01", periods=5, freq="D")
+
+    # TODO: de-duplicate with other test_equals2 methods
+    @pytest.mark.parametrize("freq", ["D", "M"])
+    def test_equals2(self, freq):
+        # GH#13107
+        idx = PeriodIndex(["2011-01-01", "2011-01-02", "NaT"], freq=freq)
+        assert idx.equals(idx)
+        assert idx.equals(idx.copy())
+        assert idx.equals(idx.astype(object))
+        assert idx.astype(object).equals(idx)
+        assert idx.astype(object).equals(idx.astype(object))
+        assert not idx.equals(list(idx))
+        assert not idx.equals(pd.Series(idx))
+
+        idx2 = PeriodIndex(["2011-01-01", "2011-01-02", "NaT"], freq="h")
+        assert not idx.equals(idx2)
+        assert not idx.equals(idx2.copy())
+        assert not idx.equals(idx2.astype(object))
+        assert not idx.astype(object).equals(idx2)
+        assert not idx.equals(list(idx2))
+        assert not idx.equals(pd.Series(idx2))
+
+        # same internal, different tz
+        idx3 = PeriodIndex._simple_new(
+            idx._values._simple_new(idx._values.asi8, dtype=pd.PeriodDtype("h"))
+        )
+        tm.assert_numpy_array_equal(idx.asi8, idx3.asi8)
+        assert not idx.equals(idx3)
+        assert not idx.equals(idx3.copy())
+        assert not idx.equals(idx3.astype(object))
+        assert not idx.astype(object).equals(idx3)
+        assert not idx.equals(list(idx3))
+        assert not idx.equals(pd.Series(idx3))
+
+
+class TestDatetimeIndexEquals(EqualsTests):
+    @pytest.fixture
+    def index(self):
+        """Fixture for creating a DatetimeIndex for use in equality tests."""
+        return date_range("2013-01-01", periods=5)
+
+    def test_equals2(self):
+        # GH#13107
+        idx = DatetimeIndex(["2011-01-01", "2011-01-02", "NaT"])
+        assert idx.equals(idx)
+        assert idx.equals(idx.copy())
+        assert idx.equals(idx.astype(object))
+        assert idx.astype(object).equals(idx)
+        assert idx.astype(object).equals(idx.astype(object))
+        assert not idx.equals(list(idx))
+        assert not idx.equals(pd.Series(idx))
+
+        idx2 = DatetimeIndex(["2011-01-01", "2011-01-02", "NaT"], tz="US/Pacific")
+        assert not idx.equals(idx2)
+        assert not idx.equals(idx2.copy())
+        assert not idx.equals(idx2.astype(object))
+        assert not idx.astype(object).equals(idx2)
+        assert not idx.equals(list(idx2))
+        assert not idx.equals(pd.Series(idx2))
+
+        # same internal, different tz
+        idx3 = DatetimeIndex(idx.asi8, tz="US/Pacific")
+        tm.assert_numpy_array_equal(idx.asi8, idx3.asi8)
+        assert not idx.equals(idx3)
+        assert not idx.equals(idx3.copy())
+        assert not idx.equals(idx3.astype(object))
+        assert not idx.astype(object).equals(idx3)
+        assert not idx.equals(list(idx3))
+        assert not idx.equals(pd.Series(idx3))
+
+        # check that we do not raise when comparing with OutOfBounds objects
+        oob = Index([datetime(2500, 1, 1)] * 3, dtype=object)
+        assert not idx.equals(oob)
+        assert not idx2.equals(oob)
+        assert not idx3.equals(oob)
+
+        # check that we do not raise when comparing with OutOfBounds dt64
+        oob2 = oob.map(np.datetime64)
+        assert not idx.equals(oob2)
+        assert not idx2.equals(oob2)
+        assert not idx3.equals(oob2)
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_not_equals_bday(self, freq):
+        rng = date_range("2009-01-01", "2010-01-01", freq=freq)
+        assert not rng.equals(list(rng))
+
+
+class TestTimedeltaIndexEquals(EqualsTests):
+    @pytest.fixture
+    def index(self):
+        """Fixture for creating a TimedeltaIndex for use in equality tests."""
+        return timedelta_range("1 day", periods=10)
+
+    def test_equals2(self):
+        # GH#13107
+        idx = TimedeltaIndex(["1 days", "2 days", "NaT"])
+        assert idx.equals(idx)
+        assert idx.equals(idx.copy())
+        assert idx.equals(idx.astype(object))
+        assert idx.astype(object).equals(idx)
+        assert idx.astype(object).equals(idx.astype(object))
+        assert not idx.equals(list(idx))
+        assert not idx.equals(pd.Series(idx))
+
+        idx2 = TimedeltaIndex(["2 days", "1 days", "NaT"])
+        assert not idx.equals(idx2)
+        assert not idx.equals(idx2.copy())
+        assert not idx.equals(idx2.astype(object))
+        assert not idx.astype(object).equals(idx2)
+        assert not idx.astype(object).equals(idx2.astype(object))
+        assert not idx.equals(list(idx2))
+        assert not idx.equals(pd.Series(idx2))
+
+        # Check that we dont raise OverflowError on comparisons outside the
+        #  implementation range GH#28532
+        oob = Index([timedelta(days=10**6)] * 3, dtype=object)
+        assert not idx.equals(oob)
+        assert not idx2.equals(oob)
+
+        oob2 = Index([np.timedelta64(x) for x in oob], dtype=object)
+        assert (oob == oob2).all()
+        assert not idx.equals(oob2)
+        assert not idx2.equals(oob2)
+
+        oob3 = oob.map(np.timedelta64)
+        assert (oob3 == oob).all()
+        assert not idx.equals(oob3)
+        assert not idx2.equals(oob3)
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_indexing.py b/pandas/pandas/tests/indexes/datetimelike_/test_indexing.py
new file mode 100644
index 0000000000000..7b2c81aaf17de
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_indexing.py
@@ -0,0 +1,45 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+)
+import pandas._testing as tm
+
+dtlike_dtypes = [
+    np.dtype("timedelta64[ns]"),
+    np.dtype("datetime64[ns]"),
+    pd.DatetimeTZDtype("ns", "Asia/Tokyo"),
+    pd.PeriodDtype("ns"),
+]
+
+
+@pytest.mark.parametrize("ldtype", dtlike_dtypes)
+@pytest.mark.parametrize("rdtype", dtlike_dtypes)
+def test_get_indexer_non_unique_wrong_dtype(ldtype, rdtype):
+    vals = np.tile(3600 * 10**9 * np.arange(3, dtype=np.int64), 2)
+
+    def construct(dtype):
+        if dtype is dtlike_dtypes[-1]:
+            # PeriodArray will try to cast ints to strings
+            return DatetimeIndex(vals).astype(dtype)
+        return Index(vals, dtype=dtype)
+
+    left = construct(ldtype)
+    right = construct(rdtype)
+
+    result = left.get_indexer_non_unique(right)
+
+    if ldtype is rdtype:
+        ex1 = np.array([0, 3, 1, 4, 2, 5] * 2, dtype=np.intp)
+        ex2 = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(result[0], ex1)
+        tm.assert_numpy_array_equal(result[1], ex2)
+
+    else:
+        no_matches = np.array([-1] * 6, dtype=np.intp)
+        missing = np.arange(6, dtype=np.intp)
+        tm.assert_numpy_array_equal(result[0], no_matches)
+        tm.assert_numpy_array_equal(result[1], missing)
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_is_monotonic.py b/pandas/pandas/tests/indexes/datetimelike_/test_is_monotonic.py
new file mode 100644
index 0000000000000..b0e42e660b751
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_is_monotonic.py
@@ -0,0 +1,46 @@
+from pandas import (
+    Index,
+    NaT,
+    date_range,
+)
+
+
+def test_is_monotonic_with_nat():
+    # GH#31437
+    # PeriodIndex.is_monotonic_increasing should behave analogously to DatetimeIndex,
+    #  in particular never be monotonic when we have NaT
+    dti = date_range("2016-01-01", periods=3)
+    pi = dti.to_period("D")
+    tdi = Index(dti.view("timedelta64[ns]"))
+
+    for obj in [pi, pi._engine, dti, dti._engine, tdi, tdi._engine]:
+        if isinstance(obj, Index):
+            # i.e. not Engines
+            assert obj.is_monotonic_increasing
+        assert obj.is_monotonic_increasing
+        assert not obj.is_monotonic_decreasing
+        assert obj.is_unique
+
+    dti1 = dti.insert(0, NaT)
+    pi1 = dti1.to_period("D")
+    tdi1 = Index(dti1.view("timedelta64[ns]"))
+
+    for obj in [pi1, pi1._engine, dti1, dti1._engine, tdi1, tdi1._engine]:
+        if isinstance(obj, Index):
+            # i.e. not Engines
+            assert not obj.is_monotonic_increasing
+        assert not obj.is_monotonic_increasing
+        assert not obj.is_monotonic_decreasing
+        assert obj.is_unique
+
+    dti2 = dti.insert(3, NaT)
+    pi2 = dti2.to_period("h")
+    tdi2 = Index(dti2.view("timedelta64[ns]"))
+
+    for obj in [pi2, pi2._engine, dti2, dti2._engine, tdi2, tdi2._engine]:
+        if isinstance(obj, Index):
+            # i.e. not Engines
+            assert not obj.is_monotonic_increasing
+        assert not obj.is_monotonic_increasing
+        assert not obj.is_monotonic_decreasing
+        assert obj.is_unique
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_nat.py b/pandas/pandas/tests/indexes/datetimelike_/test_nat.py
new file mode 100644
index 0000000000000..3dd0ce1cbd637
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_nat.py
@@ -0,0 +1,42 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    PeriodIndex,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "index_without_na",
+    [
+        TimedeltaIndex(["1 days", "2 days"]),
+        PeriodIndex(["2011-01-01", "2011-01-02"], freq="D"),
+        DatetimeIndex(["2011-01-01", "2011-01-02"]),
+        DatetimeIndex(["2011-01-01", "2011-01-02"], tz="UTC"),
+    ],
+)
+def test_nat(index_without_na):
+    empty_index = index_without_na[:0]
+
+    index_with_na = index_without_na.copy(deep=True)
+    index_with_na._data[1] = NaT
+
+    assert empty_index._na_value is NaT
+    assert index_with_na._na_value is NaT
+    assert index_without_na._na_value is NaT
+
+    idx = index_without_na
+    assert idx._can_hold_na
+
+    tm.assert_numpy_array_equal(idx._isnan, np.array([False, False]))
+    assert idx.hasnans is False
+
+    idx = index_with_na
+    assert idx._can_hold_na
+
+    tm.assert_numpy_array_equal(idx._isnan, np.array([False, True]))
+    assert idx.hasnans is True
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_sort_values.py b/pandas/pandas/tests/indexes/datetimelike_/test_sort_values.py
new file mode 100644
index 0000000000000..bfef0faebeebf
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_sort_values.py
@@ -0,0 +1,311 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    NaT,
+    PeriodIndex,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+def check_freq_ascending(ordered, orig, ascending):
+    """
+    Check the expected freq on a PeriodIndex/DatetimeIndex/TimedeltaIndex
+    when the original index is generated (or generate-able) with
+    period_range/date_range/timedelta_range.
+    """
+    if isinstance(ordered, PeriodIndex):
+        assert ordered.freq == orig.freq
+    elif isinstance(ordered, (DatetimeIndex, TimedeltaIndex)):
+        if ascending:
+            assert ordered.freq.n == orig.freq.n
+        else:
+            assert ordered.freq.n == -1 * orig.freq.n
+
+
+def check_freq_nonmonotonic(ordered, orig):
+    """
+    Check the expected freq on a PeriodIndex/DatetimeIndex/TimedeltaIndex
+    when the original index is _not_ generated (or generate-able) with
+    period_range/date_range//timedelta_range.
+    """
+    if isinstance(ordered, PeriodIndex):
+        assert ordered.freq == orig.freq
+    elif isinstance(ordered, (DatetimeIndex, TimedeltaIndex)):
+        assert ordered.freq is None
+
+
+class TestSortValues:
+    @pytest.fixture(params=[DatetimeIndex, TimedeltaIndex, PeriodIndex])
+    def non_monotonic_idx(self, request):
+        if request.param is DatetimeIndex:
+            return DatetimeIndex(["2000-01-04", "2000-01-01", "2000-01-02"])
+        elif request.param is PeriodIndex:
+            dti = DatetimeIndex(["2000-01-04", "2000-01-01", "2000-01-02"])
+            return dti.to_period("D")
+        else:
+            return TimedeltaIndex(
+                ["1 day 00:00:05", "1 day 00:00:01", "1 day 00:00:02"]
+            )
+
+    def test_argmin_argmax(self, non_monotonic_idx):
+        assert non_monotonic_idx.argmin() == 1
+        assert non_monotonic_idx.argmax() == 0
+
+    def test_sort_values(self, non_monotonic_idx):
+        idx = non_monotonic_idx
+        ordered = idx.sort_values()
+        assert ordered.is_monotonic_increasing
+        ordered = idx.sort_values(ascending=False)
+        assert ordered[::-1].is_monotonic_increasing
+
+        ordered, dexer = idx.sort_values(return_indexer=True)
+        assert ordered.is_monotonic_increasing
+        tm.assert_numpy_array_equal(dexer, np.array([1, 2, 0], dtype=np.intp))
+
+        ordered, dexer = idx.sort_values(return_indexer=True, ascending=False)
+        assert ordered[::-1].is_monotonic_increasing
+        tm.assert_numpy_array_equal(dexer, np.array([0, 2, 1], dtype=np.intp))
+
+    def check_sort_values_with_freq(self, idx):
+        ordered = idx.sort_values()
+        tm.assert_index_equal(ordered, idx)
+        check_freq_ascending(ordered, idx, True)
+
+        ordered = idx.sort_values(ascending=False)
+        expected = idx[::-1]
+        tm.assert_index_equal(ordered, expected)
+        check_freq_ascending(ordered, idx, False)
+
+        ordered, indexer = idx.sort_values(return_indexer=True)
+        tm.assert_index_equal(ordered, idx)
+        tm.assert_numpy_array_equal(indexer, np.array([0, 1, 2], dtype=np.intp))
+        check_freq_ascending(ordered, idx, True)
+
+        ordered, indexer = idx.sort_values(return_indexer=True, ascending=False)
+        expected = idx[::-1]
+        tm.assert_index_equal(ordered, expected)
+        tm.assert_numpy_array_equal(indexer, np.array([2, 1, 0], dtype=np.intp))
+        check_freq_ascending(ordered, idx, False)
+
+    @pytest.mark.parametrize("freq", ["D", "h"])
+    def test_sort_values_with_freq_timedeltaindex(self, freq):
+        # GH#10295
+        idx = timedelta_range(start=f"1{freq}", periods=3, freq=freq).rename("idx")
+
+        self.check_sort_values_with_freq(idx)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            DatetimeIndex(
+                ["2011-01-01", "2011-01-02", "2011-01-03"], freq="D", name="idx"
+            ),
+            DatetimeIndex(
+                ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
+                freq="h",
+                name="tzidx",
+                tz="Asia/Tokyo",
+            ),
+        ],
+    )
+    def test_sort_values_with_freq_datetimeindex(self, idx):
+        self.check_sort_values_with_freq(idx)
+
+    @pytest.mark.parametrize("freq", ["D", "2D", "4D"])
+    def test_sort_values_with_freq_periodindex(self, freq):
+        # here with_freq refers to being period_range-like
+        idx = PeriodIndex(
+            ["2011-01-01", "2011-01-02", "2011-01-03"], freq=freq, name="idx"
+        )
+        self.check_sort_values_with_freq(idx)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            PeriodIndex(["2011", "2012", "2013"], name="pidx", freq="Y"),
+            Index([2011, 2012, 2013], name="idx"),  # for compatibility check
+        ],
+    )
+    def test_sort_values_with_freq_periodindex2(self, idx):
+        # here with_freq indicates this is period_range-like
+        self.check_sort_values_with_freq(idx)
+
+    def check_sort_values_without_freq(self, idx, expected):
+        ordered = idx.sort_values(na_position="first")
+        tm.assert_index_equal(ordered, expected)
+        check_freq_nonmonotonic(ordered, idx)
+
+        if not idx.isna().any():
+            ordered = idx.sort_values()
+            tm.assert_index_equal(ordered, expected)
+            check_freq_nonmonotonic(ordered, idx)
+
+        ordered = idx.sort_values(ascending=False)
+        tm.assert_index_equal(ordered, expected[::-1])
+        check_freq_nonmonotonic(ordered, idx)
+
+        ordered, indexer = idx.sort_values(return_indexer=True, na_position="first")
+        tm.assert_index_equal(ordered, expected)
+
+        exp = np.array([0, 4, 3, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, exp)
+        check_freq_nonmonotonic(ordered, idx)
+
+        if not idx.isna().any():
+            ordered, indexer = idx.sort_values(return_indexer=True)
+            tm.assert_index_equal(ordered, expected)
+
+            exp = np.array([0, 4, 3, 1, 2], dtype=np.intp)
+            tm.assert_numpy_array_equal(indexer, exp)
+            check_freq_nonmonotonic(ordered, idx)
+
+        ordered, indexer = idx.sort_values(return_indexer=True, ascending=False)
+        tm.assert_index_equal(ordered, expected[::-1])
+
+        exp = np.array([2, 1, 3, 0, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, exp)
+        check_freq_nonmonotonic(ordered, idx)
+
+    def test_sort_values_without_freq_timedeltaindex(self):
+        # GH#10295
+
+        idx = TimedeltaIndex(
+            ["1 hour", "3 hour", "5 hour", "2 hour ", "1 hour"], name="idx1"
+        )
+        expected = TimedeltaIndex(
+            ["1 hour", "1 hour", "2 hour", "3 hour", "5 hour"], name="idx1"
+        )
+        self.check_sort_values_without_freq(idx, expected)
+
+    @pytest.mark.parametrize(
+        "index_dates,expected_dates",
+        [
+            (
+                ["2011-01-01", "2011-01-03", "2011-01-05", "2011-01-02", "2011-01-01"],
+                ["2011-01-01", "2011-01-01", "2011-01-02", "2011-01-03", "2011-01-05"],
+            ),
+            (
+                [NaT, "2011-01-03", "2011-01-05", "2011-01-02", NaT],
+                [NaT, NaT, "2011-01-02", "2011-01-03", "2011-01-05"],
+            ),
+        ],
+    )
+    def test_sort_values_without_freq_datetimeindex(
+        self, index_dates, expected_dates, tz_naive_fixture
+    ):
+        tz = tz_naive_fixture
+
+        # without freq
+        idx = DatetimeIndex(index_dates, tz=tz, name="idx")
+        expected = DatetimeIndex(expected_dates, tz=tz, name="idx")
+
+        self.check_sort_values_without_freq(idx, expected)
+
+    @pytest.mark.parametrize(
+        "idx,expected",
+        [
+            (
+                PeriodIndex(
+                    [
+                        "2011-01-01",
+                        "2011-01-03",
+                        "2011-01-05",
+                        "2011-01-02",
+                        "2011-01-01",
+                    ],
+                    freq="D",
+                    name="idx1",
+                ),
+                PeriodIndex(
+                    [
+                        "2011-01-01",
+                        "2011-01-01",
+                        "2011-01-02",
+                        "2011-01-03",
+                        "2011-01-05",
+                    ],
+                    freq="D",
+                    name="idx1",
+                ),
+            ),
+            (
+                PeriodIndex(
+                    [
+                        "2011-01-01",
+                        "2011-01-03",
+                        "2011-01-05",
+                        "2011-01-02",
+                        "2011-01-01",
+                    ],
+                    freq="D",
+                    name="idx2",
+                ),
+                PeriodIndex(
+                    [
+                        "2011-01-01",
+                        "2011-01-01",
+                        "2011-01-02",
+                        "2011-01-03",
+                        "2011-01-05",
+                    ],
+                    freq="D",
+                    name="idx2",
+                ),
+            ),
+            (
+                PeriodIndex(
+                    [NaT, "2011-01-03", "2011-01-05", "2011-01-02", NaT],
+                    freq="D",
+                    name="idx3",
+                ),
+                PeriodIndex(
+                    [NaT, NaT, "2011-01-02", "2011-01-03", "2011-01-05"],
+                    freq="D",
+                    name="idx3",
+                ),
+            ),
+            (
+                PeriodIndex(
+                    ["2011", "2013", "2015", "2012", "2011"], name="pidx", freq="Y"
+                ),
+                PeriodIndex(
+                    ["2011", "2011", "2012", "2013", "2015"], name="pidx", freq="Y"
+                ),
+            ),
+            (
+                # For compatibility check
+                Index([2011, 2013, 2015, 2012, 2011], name="idx"),
+                Index([2011, 2011, 2012, 2013, 2015], name="idx"),
+            ),
+        ],
+    )
+    def test_sort_values_without_freq_periodindex(self, idx, expected):
+        # here without_freq means not generateable by period_range
+        self.check_sort_values_without_freq(idx, expected)
+
+    def test_sort_values_without_freq_periodindex_nat(self):
+        # doesn't quite fit into check_sort_values_without_freq
+        idx = PeriodIndex(["2011", "2013", "NaT", "2011"], name="pidx", freq="D")
+        expected = PeriodIndex(["NaT", "2011", "2011", "2013"], name="pidx", freq="D")
+
+        ordered = idx.sort_values(na_position="first")
+        tm.assert_index_equal(ordered, expected)
+        check_freq_nonmonotonic(ordered, idx)
+
+        ordered = idx.sort_values(ascending=False)
+        tm.assert_index_equal(ordered, expected[::-1])
+        check_freq_nonmonotonic(ordered, idx)
+
+
+def test_order_stability_compat():
+    # GH#35922. sort_values is stable both for normal and datetime-like Index
+    pidx = PeriodIndex(["2011", "2013", "2015", "2012", "2011"], name="pidx", freq="Y")
+    iidx = Index([2011, 2013, 2015, 2012, 2011], name="idx")
+    ordered1, indexer1 = pidx.sort_values(return_indexer=True, ascending=False)
+    ordered2, indexer2 = iidx.sort_values(return_indexer=True, ascending=False)
+    tm.assert_numpy_array_equal(indexer1, indexer2)
diff --git a/pandas/pandas/tests/indexes/datetimelike_/test_value_counts.py b/pandas/pandas/tests/indexes/datetimelike_/test_value_counts.py
new file mode 100644
index 0000000000000..069e354a364c9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimelike_/test_value_counts.py
@@ -0,0 +1,103 @@
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    TimedeltaIndex,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestValueCounts:
+    # GH#7735
+
+    def test_value_counts_unique_datetimeindex(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        orig = date_range("2011-01-01 09:00", freq="h", periods=10, tz=tz)
+        self._check_value_counts_with_repeats(orig)
+
+    def test_value_counts_unique_timedeltaindex(self):
+        orig = timedelta_range("1 days 09:00:00", freq="h", periods=10)
+        self._check_value_counts_with_repeats(orig)
+
+    def test_value_counts_unique_periodindex(self):
+        orig = period_range("2011-01-01 09:00", freq="h", periods=10)
+        self._check_value_counts_with_repeats(orig)
+
+    def _check_value_counts_with_repeats(self, orig):
+        # create repeated values, 'n'th element is repeated by n+1 times
+        idx = type(orig)(
+            np.repeat(orig._values, range(1, len(orig) + 1)), dtype=orig.dtype
+        )
+
+        exp_idx = orig[::-1]
+        if not isinstance(exp_idx, PeriodIndex):
+            exp_idx = exp_idx._with_freq(None)
+        expected = Series(range(10, 0, -1), index=exp_idx, dtype="int64", name="count")
+
+        for obj in [idx, Series(idx)]:
+            tm.assert_series_equal(obj.value_counts(), expected)
+
+        tm.assert_index_equal(idx.unique(), orig)
+
+    def test_value_counts_unique_datetimeindex2(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        idx = DatetimeIndex(
+            [
+                "2013-01-01 09:00",
+                "2013-01-01 09:00",
+                "2013-01-01 09:00",
+                "2013-01-01 08:00",
+                "2013-01-01 08:00",
+                NaT,
+            ],
+            tz=tz,
+        )
+        self._check_value_counts_dropna(idx)
+
+    def test_value_counts_unique_timedeltaindex2(self):
+        idx = TimedeltaIndex(
+            [
+                "1 days 09:00:00",
+                "1 days 09:00:00",
+                "1 days 09:00:00",
+                "1 days 08:00:00",
+                "1 days 08:00:00",
+                NaT,
+            ]
+        )
+        self._check_value_counts_dropna(idx)
+
+    def test_value_counts_unique_periodindex2(self):
+        idx = PeriodIndex(
+            [
+                "2013-01-01 09:00",
+                "2013-01-01 09:00",
+                "2013-01-01 09:00",
+                "2013-01-01 08:00",
+                "2013-01-01 08:00",
+                NaT,
+            ],
+            freq="h",
+        )
+        self._check_value_counts_dropna(idx)
+
+    def _check_value_counts_dropna(self, idx):
+        exp_idx = idx[[2, 3]]
+        expected = Series([3, 2], index=exp_idx, name="count")
+
+        for obj in [idx, Series(idx)]:
+            tm.assert_series_equal(obj.value_counts(), expected)
+
+        exp_idx = idx[[2, 3, -1]]
+        expected = Series([3, 2, 1], index=exp_idx, name="count")
+
+        for obj in [idx, Series(idx)]:
+            tm.assert_series_equal(obj.value_counts(dropna=False), expected)
+
+        tm.assert_index_equal(idx.unique(), exp_idx)
diff --git a/pandas/pandas/tests/indexes/datetimes/__init__.py b/pandas/pandas/tests/indexes/datetimes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/__init__.py b/pandas/pandas/tests/indexes/datetimes/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_asof.py b/pandas/pandas/tests/indexes/datetimes/methods/test_asof.py
new file mode 100644
index 0000000000000..dc92f533087bc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_asof.py
@@ -0,0 +1,30 @@
+from datetime import timedelta
+
+from pandas import (
+    Index,
+    Timestamp,
+    date_range,
+    isna,
+)
+
+
+class TestAsOf:
+    def test_asof_partial(self):
+        index = date_range("2010-01-01", periods=2, freq="ME")
+        expected = Timestamp("2010-02-28")
+        result = index.asof("2010-02")
+        assert result == expected
+        assert not isinstance(result, Index)
+
+    def test_asof(self):
+        index = date_range("2020-01-01", periods=10)
+
+        dt = index[0]
+        assert index.asof(dt) == dt
+        assert isna(index.asof(dt - timedelta(1)))
+
+        dt = index[-1]
+        assert index.asof(dt + timedelta(1)) == dt
+
+        dt = index[0].to_pydatetime()
+        assert isinstance(index.asof(dt), Timestamp)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_astype.py b/pandas/pandas/tests/indexes/datetimes/methods/test_astype.py
new file mode 100644
index 0000000000000..62be8903da206
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_astype.py
@@ -0,0 +1,339 @@
+from datetime import datetime
+
+import dateutil
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    NaT,
+    PeriodIndex,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndex:
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_astype_asobject_around_dst_transition(self, tzstr):
+        # GH#1345
+
+        # dates around a dst transition
+        rng = date_range("2/13/2010", "5/6/2010", tz=tzstr)
+
+        objs = rng.astype(object)
+        for i, x in enumerate(objs):
+            exval = rng[i]
+            assert x == exval
+            assert x.tzinfo == exval.tzinfo
+
+        objs = rng.astype(object)
+        for i, x in enumerate(objs):
+            exval = rng[i]
+            assert x == exval
+            assert x.tzinfo == exval.tzinfo
+
+    def test_astype(self):
+        # GH 13149, GH 13209
+        idx = DatetimeIndex(
+            ["2016-05-16", "NaT", NaT, np.nan], dtype="M8[ns]", name="idx"
+        )
+
+        result = idx.astype(object)
+        expected = Index(
+            [Timestamp("2016-05-16")] + [NaT] * 3, dtype=object, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.astype(np.int64)
+        expected = Index(
+            [1463356800000000000] + [-9223372036854775808] * 3,
+            dtype=np.int64,
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_astype2(self):
+        rng = date_range("1/1/2000", periods=10, name="idx")
+        result = rng.astype("i8")
+        tm.assert_index_equal(result, Index(rng.asi8, name="idx"))
+        tm.assert_numpy_array_equal(result.values, rng.asi8)
+
+    def test_astype_uint(self):
+        arr = date_range("2000", periods=2, name="idx")
+
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint64")
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint32")
+
+    def test_astype_with_tz(self):
+        # with tz
+        rng = date_range("1/1/2000", periods=10, tz="US/Eastern")
+        msg = "Cannot use .astype to convert from timezone-aware"
+        with pytest.raises(TypeError, match=msg):
+            # deprecated
+            rng.astype("datetime64[ns]")
+        with pytest.raises(TypeError, match=msg):
+            # check DatetimeArray while we're here deprecated
+            rng._data.astype("datetime64[ns]")
+
+    def test_astype_tzaware_to_tzaware(self):
+        # GH 18951: tz-aware to tz-aware
+        idx = date_range("20170101", periods=4, tz="US/Pacific")
+        result = idx.astype("datetime64[ns, US/Eastern]")
+        expected = date_range("20170101 03:00:00", periods=4, tz="US/Eastern")
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+    def test_astype_tznaive_to_tzaware(self):
+        # GH 18951: tz-naive to tz-aware
+        idx = date_range("20170101", periods=4)
+        idx = idx._with_freq(None)  # tz_localize does not preserve freq
+        msg = "Cannot use .astype to convert from timezone-naive"
+        with pytest.raises(TypeError, match=msg):
+            # dt64->dt64tz deprecated
+            idx.astype("datetime64[ns, US/Eastern]")
+        with pytest.raises(TypeError, match=msg):
+            # dt64->dt64tz deprecated
+            idx._data.astype("datetime64[ns, US/Eastern]")
+
+    def test_astype_str_nat(self, using_infer_string):
+        # GH 13149, GH 13209
+        # verify that we are returning NaT as a string (and not unicode)
+
+        idx = DatetimeIndex(["2016-05-16", "NaT", NaT, np.nan])
+        result = idx.astype(str)
+        if using_infer_string:
+            expected = Index(["2016-05-16", None, None, None], dtype="str")
+        else:
+            expected = Index(["2016-05-16", "NaT", "NaT", "NaT"], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_astype_str(self):
+        # test astype string - #10442
+        dti = date_range("2012-01-01", periods=4, name="test_name")
+        result = dti.astype(str)
+        expected = Index(
+            ["2012-01-01", "2012-01-02", "2012-01-03", "2012-01-04"],
+            name="test_name",
+            dtype="str",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_astype_str_tz_and_name(self):
+        # test astype string with tz and name
+        dti = date_range("2012-01-01", periods=3, name="test_name", tz="US/Eastern")
+        result = dti.astype(str)
+        expected = Index(
+            [
+                "2012-01-01 00:00:00-05:00",
+                "2012-01-02 00:00:00-05:00",
+                "2012-01-03 00:00:00-05:00",
+            ],
+            name="test_name",
+            dtype="str",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_astype_str_freq_and_name(self):
+        # test astype string with freqH and name
+        dti = date_range("1/1/2011", periods=3, freq="h", name="test_name")
+        result = dti.astype(str)
+        expected = Index(
+            ["2011-01-01 00:00:00", "2011-01-01 01:00:00", "2011-01-01 02:00:00"],
+            name="test_name",
+            dtype="str",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_astype_str_freq_and_tz(self):
+        # test astype string with freqH and timezone
+        dti = date_range(
+            "3/6/2012 00:00", periods=2, freq="h", tz="Europe/London", name="test_name"
+        )
+        result = dti.astype(str)
+        expected = Index(
+            ["2012-03-06 00:00:00+00:00", "2012-03-06 01:00:00+00:00"],
+            dtype="str",
+            name="test_name",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_astype_datetime64(self):
+        # GH 13149, GH 13209
+        idx = DatetimeIndex(
+            ["2016-05-16", "NaT", NaT, np.nan], dtype="M8[ns]", name="idx"
+        )
+
+        result = idx.astype("datetime64[ns]")
+        tm.assert_index_equal(result, idx)
+        assert result is not idx
+
+        result = idx.astype("datetime64[ns]", copy=False)
+        tm.assert_index_equal(result, idx)
+        assert result is idx
+
+        idx_tz = DatetimeIndex(["2016-05-16", "NaT", NaT, np.nan], tz="EST", name="idx")
+        msg = "Cannot use .astype to convert from timezone-aware"
+        with pytest.raises(TypeError, match=msg):
+            # dt64tz->dt64 deprecated
+            result = idx_tz.astype("datetime64[ns]")
+
+    def test_astype_object(self):
+        rng = date_range("1/1/2000", periods=20)
+
+        casted = rng.astype("O")
+        exp_values = list(rng)
+
+        tm.assert_index_equal(casted, Index(exp_values, dtype=np.object_))
+        assert casted.tolist() == exp_values
+
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo"])
+    def test_astype_object_tz(self, tz):
+        idx = date_range(start="2013-01-01", periods=4, freq="ME", name="idx", tz=tz)
+        expected_list = [
+            Timestamp("2013-01-31", tz=tz),
+            Timestamp("2013-02-28", tz=tz),
+            Timestamp("2013-03-31", tz=tz),
+            Timestamp("2013-04-30", tz=tz),
+        ]
+        expected = Index(expected_list, dtype=object, name="idx")
+        result = idx.astype(object)
+        tm.assert_index_equal(result, expected)
+        assert idx.tolist() == expected_list
+
+    def test_astype_object_with_nat(self):
+        idx = DatetimeIndex(
+            [datetime(2013, 1, 1), datetime(2013, 1, 2), NaT, datetime(2013, 1, 4)],
+            name="idx",
+        )
+        expected_list = [
+            Timestamp("2013-01-01"),
+            Timestamp("2013-01-02"),
+            NaT,
+            Timestamp("2013-01-04"),
+        ]
+        expected = Index(expected_list, dtype=object, name="idx")
+        result = idx.astype(object)
+        tm.assert_index_equal(result, expected)
+        assert idx.tolist() == expected_list
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [float, "timedelta64", "timedelta64[ns]", "datetime64", "datetime64[D]"],
+    )
+    def test_astype_raises(self, dtype):
+        # GH 13149, GH 13209
+        idx = DatetimeIndex(["2016-05-16", "NaT", NaT, np.nan])
+        msg = "Cannot cast DatetimeIndex to dtype"
+        if dtype == "datetime64":
+            msg = "Casting to unit-less dtype 'datetime64' is not supported"
+        with pytest.raises(TypeError, match=msg):
+            idx.astype(dtype)
+
+    def test_index_convert_to_datetime_array(self):
+        def _check_rng(rng):
+            converted = rng.to_pydatetime()
+            assert isinstance(converted, np.ndarray)
+            for x, stamp in zip(converted, rng):
+                assert isinstance(x, datetime)
+                assert x == stamp.to_pydatetime()
+                assert x.tzinfo == stamp.tzinfo
+
+        rng = date_range("20090415", "20090519")
+        rng_eastern = date_range("20090415", "20090519", tz="US/Eastern")
+        rng_utc = date_range("20090415", "20090519", tz="utc")
+
+        _check_rng(rng)
+        _check_rng(rng_eastern)
+        _check_rng(rng_utc)
+
+    def test_index_convert_to_datetime_array_explicit_pytz(self):
+        pytz = pytest.importorskip("pytz")
+
+        def _check_rng(rng):
+            converted = rng.to_pydatetime()
+            assert isinstance(converted, np.ndarray)
+            for x, stamp in zip(converted, rng):
+                assert isinstance(x, datetime)
+                assert x == stamp.to_pydatetime()
+                assert x.tzinfo == stamp.tzinfo
+
+        rng = date_range("20090415", "20090519")
+        rng_eastern = date_range("20090415", "20090519", tz=pytz.timezone("US/Eastern"))
+        rng_utc = date_range("20090415", "20090519", tz=pytz.utc)
+
+        _check_rng(rng)
+        _check_rng(rng_eastern)
+        _check_rng(rng_utc)
+
+    def test_index_convert_to_datetime_array_dateutil(self):
+        def _check_rng(rng):
+            converted = rng.to_pydatetime()
+            assert isinstance(converted, np.ndarray)
+            for x, stamp in zip(converted, rng):
+                assert isinstance(x, datetime)
+                assert x == stamp.to_pydatetime()
+                assert x.tzinfo == stamp.tzinfo
+
+        rng = date_range("20090415", "20090519")
+        rng_eastern = date_range("20090415", "20090519", tz="dateutil/US/Eastern")
+        rng_utc = date_range("20090415", "20090519", tz=dateutil.tz.tzutc())
+
+        _check_rng(rng)
+        _check_rng(rng_eastern)
+        _check_rng(rng_utc)
+
+    @pytest.mark.parametrize(
+        "tz, dtype",
+        [["US/Pacific", "datetime64[ns, US/Pacific]"], [None, "datetime64[ns]"]],
+    )
+    def test_integer_index_astype_datetime(self, tz, dtype):
+        # GH 20997, 20964, 24559
+        val = [Timestamp("2018-01-01", tz=tz).as_unit("ns")._value]
+        result = Index(val, name="idx").astype(dtype)
+        expected = DatetimeIndex(["2018-01-01"], tz=tz, name="idx").as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_astype_period(self):
+        idx = DatetimeIndex([NaT, "2011-01-01", "2011-02-01"], name="idx")
+
+        res = idx.astype("period[M]")
+        exp = PeriodIndex(["NaT", "2011-01", "2011-02"], freq="M", name="idx")
+        tm.assert_index_equal(res, exp)
+
+        res = idx.astype("period[3M]")
+        exp = PeriodIndex(["NaT", "2011-01", "2011-02"], freq="3M", name="idx")
+        tm.assert_index_equal(res, exp)
+
+
+class TestAstype:
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_astype_category(self, tz):
+        obj = date_range("2000", periods=2, tz=tz, name="idx")
+        result = obj.astype("category")
+        dti = DatetimeIndex(["2000-01-01", "2000-01-02"], tz=tz).as_unit("ns")
+        expected = pd.CategoricalIndex(
+            dti,
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype("category")
+        expected = expected.values
+        tm.assert_categorical_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_astype_array_fallback(self, tz):
+        obj = date_range("2000", periods=2, tz=tz, name="idx")
+        result = obj.astype(bool)
+        expected = Index(np.array([True, True]), name="idx")
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype(bool)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_delete.py b/pandas/pandas/tests/indexes/datetimes/methods/test_delete.py
new file mode 100644
index 0000000000000..2341499977f22
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_delete.py
@@ -0,0 +1,141 @@
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDelete:
+    def test_delete(self, unit):
+        idx = date_range(
+            start="2000-01-01", periods=5, freq="ME", name="idx", unit=unit
+        )
+
+        # preserve freq
+        expected_0 = date_range(
+            start="2000-02-01", periods=4, freq="ME", name="idx", unit=unit
+        )
+        expected_4 = date_range(
+            start="2000-01-01", periods=4, freq="ME", name="idx", unit=unit
+        )
+
+        # reset freq to None
+        expected_1 = DatetimeIndex(
+            ["2000-01-31", "2000-03-31", "2000-04-30", "2000-05-31"],
+            freq=None,
+            name="idx",
+        ).as_unit(unit)
+
+        cases = {
+            0: expected_0,
+            -5: expected_0,
+            -1: expected_4,
+            4: expected_4,
+            1: expected_1,
+        }
+        for n, expected in cases.items():
+            result = idx.delete(n)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+        with pytest.raises((IndexError, ValueError), match="out of bounds"):
+            # either depending on numpy version
+            idx.delete(5)
+
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo", "US/Pacific"])
+    def test_delete2(self, tz):
+        idx = date_range(
+            start="2000-01-01 09:00", periods=10, freq="h", name="idx", tz=tz
+        )
+
+        expected = date_range(
+            start="2000-01-01 10:00", periods=9, freq="h", name="idx", tz=tz
+        )
+        result = idx.delete(0)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freqstr == "h"
+        assert result.tz == expected.tz
+
+        expected = date_range(
+            start="2000-01-01 09:00", periods=9, freq="h", name="idx", tz=tz
+        )
+        result = idx.delete(-1)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freqstr == "h"
+        assert result.tz == expected.tz
+
+    def test_delete_slice(self, unit):
+        idx = date_range(
+            start="2000-01-01", periods=10, freq="D", name="idx", unit=unit
+        )
+
+        # preserve freq
+        expected_0_2 = date_range(
+            start="2000-01-04", periods=7, freq="D", name="idx", unit=unit
+        )
+        expected_7_9 = date_range(
+            start="2000-01-01", periods=7, freq="D", name="idx", unit=unit
+        )
+
+        # reset freq to None
+        expected_3_5 = DatetimeIndex(
+            [
+                "2000-01-01",
+                "2000-01-02",
+                "2000-01-03",
+                "2000-01-07",
+                "2000-01-08",
+                "2000-01-09",
+                "2000-01-10",
+            ],
+            freq=None,
+            name="idx",
+        ).as_unit(unit)
+
+        cases = {
+            (0, 1, 2): expected_0_2,
+            (7, 8, 9): expected_7_9,
+            (3, 4, 5): expected_3_5,
+        }
+        for n, expected in cases.items():
+            result = idx.delete(n)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+            result = idx.delete(slice(n[0], n[-1] + 1))
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+    # TODO: belongs in Series.drop tests?
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo", "US/Pacific"])
+    def test_delete_slice2(self, tz, unit):
+        dti = date_range(
+            "2000-01-01 09:00", periods=10, freq="h", name="idx", tz=tz, unit=unit
+        )
+        ts = Series(
+            1,
+            index=dti,
+        )
+        # preserve freq
+        result = ts.drop(ts.index[:5]).index
+        expected = dti[5:]
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freq == expected.freq
+        assert result.tz == expected.tz
+
+        # reset freq to None
+        result = ts.drop(ts.index[[1, 3, 5, 7, 9]]).index
+        expected = dti[::2]._with_freq(None)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freq == expected.freq
+        assert result.tz == expected.tz
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_factorize.py b/pandas/pandas/tests/indexes/datetimes/methods/test_factorize.py
new file mode 100644
index 0000000000000..41ecf9ee6b823
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_factorize.py
@@ -0,0 +1,125 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    date_range,
+    factorize,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexFactorize:
+    def test_factorize(self):
+        idx1 = DatetimeIndex(
+            ["2014-01", "2014-01", "2014-02", "2014-02", "2014-03", "2014-03"]
+        )
+
+        exp_arr = np.array([0, 0, 1, 1, 2, 2], dtype=np.intp)
+        exp_idx = DatetimeIndex(["2014-01", "2014-02", "2014-03"])
+
+        arr, idx = idx1.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+        arr, idx = idx1.factorize(sort=True)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+        # tz must be preserved
+        idx1 = idx1.tz_localize("Asia/Tokyo")
+        exp_idx = exp_idx.tz_localize("Asia/Tokyo")
+
+        arr, idx = idx1.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+        idx2 = DatetimeIndex(
+            ["2014-03", "2014-03", "2014-02", "2014-01", "2014-03", "2014-01"]
+        )
+
+        exp_arr = np.array([2, 2, 1, 0, 2, 0], dtype=np.intp)
+        exp_idx = DatetimeIndex(["2014-01", "2014-02", "2014-03"])
+        arr, idx = idx2.factorize(sort=True)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+        exp_arr = np.array([0, 0, 1, 2, 0, 2], dtype=np.intp)
+        exp_idx = DatetimeIndex(["2014-03", "2014-02", "2014-01"])
+        arr, idx = idx2.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+    def test_factorize_preserves_freq(self):
+        # GH#38120 freq should be preserved
+        idx3 = date_range("2000-01", periods=4, freq="ME", tz="Asia/Tokyo")
+        exp_arr = np.array([0, 1, 2, 3], dtype=np.intp)
+
+        arr, idx = idx3.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, idx3)
+        assert idx.freq == idx3.freq
+
+        arr, idx = factorize(idx3)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, idx3)
+        assert idx.freq == idx3.freq
+
+    def test_factorize_tz(self, tz_naive_fixture, index_or_series):
+        tz = tz_naive_fixture
+        # GH#13750
+        base = date_range("2016-11-05", freq="h", periods=100, tz=tz)
+        idx = base.repeat(5)
+
+        exp_arr = np.arange(100, dtype=np.intp).repeat(5)
+
+        obj = index_or_series(idx)
+
+        arr, res = obj.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        expected = base._with_freq(None)
+        tm.assert_index_equal(res, expected)
+        assert res.freq == expected.freq
+
+    def test_factorize_dst(self, index_or_series):
+        # GH#13750
+        idx = date_range("2016-11-06", freq="h", periods=12, tz="US/Eastern")
+        obj = index_or_series(idx)
+
+        arr, res = obj.factorize()
+        tm.assert_numpy_array_equal(arr, np.arange(12, dtype=np.intp))
+        tm.assert_index_equal(res, idx)
+        if index_or_series is Index:
+            assert res.freq == idx.freq
+
+        idx = date_range("2016-06-13", freq="h", periods=12, tz="US/Eastern")
+        obj = index_or_series(idx)
+
+        arr, res = obj.factorize()
+        tm.assert_numpy_array_equal(arr, np.arange(12, dtype=np.intp))
+        tm.assert_index_equal(res, idx)
+        if index_or_series is Index:
+            assert res.freq == idx.freq
+
+    @pytest.mark.parametrize("sort", [True, False])
+    def test_factorize_no_freq_non_nano(self, tz_naive_fixture, sort):
+        # GH#51978 case that does not go through the fastpath based on
+        #  non-None freq
+        tz = tz_naive_fixture
+        idx = date_range("2016-11-06", freq="h", periods=5, tz=tz)[[0, 4, 1, 3, 2]]
+        exp_codes, exp_uniques = idx.factorize(sort=sort)
+
+        res_codes, res_uniques = idx.as_unit("s").factorize(sort=sort)
+
+        tm.assert_numpy_array_equal(res_codes, exp_codes)
+        tm.assert_index_equal(res_uniques, exp_uniques.as_unit("s"))
+
+        res_codes, res_uniques = idx.as_unit("s").to_series().factorize(sort=sort)
+        tm.assert_numpy_array_equal(res_codes, exp_codes)
+        tm.assert_index_equal(res_uniques, exp_uniques.as_unit("s"))
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_fillna.py b/pandas/pandas/tests/indexes/datetimes/methods/test_fillna.py
new file mode 100644
index 0000000000000..5fbe60bb0c50f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_fillna.py
@@ -0,0 +1,62 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestDatetimeIndexFillNA:
+    @pytest.mark.parametrize("tz", ["US/Eastern", "Asia/Tokyo"])
+    def test_fillna_datetime64(self, tz):
+        # GH 11343
+        idx = pd.DatetimeIndex(["2011-01-01 09:00", pd.NaT, "2011-01-01 11:00"])
+
+        exp = pd.DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"]
+        )
+        tm.assert_index_equal(idx.fillna(pd.Timestamp("2011-01-01 10:00")), exp)
+
+        # tz mismatch
+        exp = pd.Index(
+            [
+                pd.Timestamp("2011-01-01 09:00"),
+                pd.Timestamp("2011-01-01 10:00", tz=tz),
+                pd.Timestamp("2011-01-01 11:00"),
+            ],
+            dtype=object,
+        )
+        tm.assert_index_equal(idx.fillna(pd.Timestamp("2011-01-01 10:00", tz=tz)), exp)
+
+        # object
+        exp = pd.Index(
+            [pd.Timestamp("2011-01-01 09:00"), "x", pd.Timestamp("2011-01-01 11:00")],
+            dtype=object,
+        )
+        tm.assert_index_equal(idx.fillna("x"), exp)
+
+        idx = pd.DatetimeIndex(["2011-01-01 09:00", pd.NaT, "2011-01-01 11:00"], tz=tz)
+
+        exp = pd.DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"], tz=tz
+        )
+        tm.assert_index_equal(idx.fillna(pd.Timestamp("2011-01-01 10:00", tz=tz)), exp)
+
+        exp = pd.Index(
+            [
+                pd.Timestamp("2011-01-01 09:00", tz=tz),
+                pd.Timestamp("2011-01-01 10:00"),
+                pd.Timestamp("2011-01-01 11:00", tz=tz),
+            ],
+            dtype=object,
+        )
+        tm.assert_index_equal(idx.fillna(pd.Timestamp("2011-01-01 10:00")), exp)
+
+        # object
+        exp = pd.Index(
+            [
+                pd.Timestamp("2011-01-01 09:00", tz=tz),
+                "x",
+                pd.Timestamp("2011-01-01 11:00", tz=tz),
+            ],
+            dtype=object,
+        )
+        tm.assert_index_equal(idx.fillna("x"), exp)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_insert.py b/pandas/pandas/tests/indexes/datetimes/methods/test_insert.py
new file mode 100644
index 0000000000000..4a5b7bcc1a86f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_insert.py
@@ -0,0 +1,275 @@
+from datetime import datetime
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DatetimeIndex,
+    Index,
+    NaT,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestInsert:
+    @pytest.mark.parametrize("null", [None, np.nan, np.datetime64("NaT"), NaT, NA])
+    @pytest.mark.parametrize("tz", [None, "UTC", "US/Eastern"])
+    def test_insert_nat(self, tz, null):
+        # GH#16537, GH#18295 (test missing)
+
+        idx = DatetimeIndex(["2017-01-01"], tz=tz)
+        expected = DatetimeIndex(["NaT", "2017-01-01"], tz=tz)
+        if tz is not None and isinstance(null, np.datetime64):
+            expected = Index([null, idx[0]], dtype=object)
+
+        res = idx.insert(0, null)
+        tm.assert_index_equal(res, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC", "US/Eastern"])
+    def test_insert_invalid_na(self, tz):
+        idx = DatetimeIndex(["2017-01-01"], tz=tz)
+
+        item = np.timedelta64("NaT")
+        result = idx.insert(0, item)
+        expected = Index([item] + list(idx), dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_empty_preserves_freq(self, tz_naive_fixture):
+        # GH#33573
+        tz = tz_naive_fixture
+        dti = DatetimeIndex([], tz=tz, freq="D")
+        item = Timestamp("2017-04-05").tz_localize(tz)
+
+        result = dti.insert(0, item)
+        assert result.freq == dti.freq
+
+        # But not when we insert an item that doesn't conform to freq
+        dti = DatetimeIndex([], tz=tz, freq="W-THU")
+        result = dti.insert(0, item)
+        assert result.freq is None
+
+    def test_insert(self, unit):
+        idx = DatetimeIndex(
+            ["2000-01-04", "2000-01-01", "2000-01-02"], name="idx"
+        ).as_unit(unit)
+
+        result = idx.insert(2, datetime(2000, 1, 5))
+        exp = DatetimeIndex(
+            ["2000-01-04", "2000-01-01", "2000-01-05", "2000-01-02"], name="idx"
+        ).as_unit(unit)
+        tm.assert_index_equal(result, exp)
+
+        # insertion of non-datetime should coerce to object index
+        result = idx.insert(1, "inserted")
+        expected = Index(
+            [
+                datetime(2000, 1, 4),
+                "inserted",
+                datetime(2000, 1, 1),
+                datetime(2000, 1, 2),
+            ],
+            name="idx",
+        )
+        assert not isinstance(result, DatetimeIndex)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+
+    def test_insert2(self, unit):
+        idx = date_range("1/1/2000", periods=3, freq="ME", name="idx", unit=unit)
+
+        # preserve freq
+        expected_0 = DatetimeIndex(
+            ["1999-12-31", "2000-01-31", "2000-02-29", "2000-03-31"],
+            name="idx",
+            freq="ME",
+        ).as_unit(unit)
+        expected_3 = DatetimeIndex(
+            ["2000-01-31", "2000-02-29", "2000-03-31", "2000-04-30"],
+            name="idx",
+            freq="ME",
+        ).as_unit(unit)
+
+        # reset freq to None
+        expected_1_nofreq = DatetimeIndex(
+            ["2000-01-31", "2000-01-31", "2000-02-29", "2000-03-31"],
+            name="idx",
+            freq=None,
+        ).as_unit(unit)
+        expected_3_nofreq = DatetimeIndex(
+            ["2000-01-31", "2000-02-29", "2000-03-31", "2000-01-02"],
+            name="idx",
+            freq=None,
+        ).as_unit(unit)
+
+        cases = [
+            (0, datetime(1999, 12, 31), expected_0),
+            (-3, datetime(1999, 12, 31), expected_0),
+            (3, datetime(2000, 4, 30), expected_3),
+            (1, datetime(2000, 1, 31), expected_1_nofreq),
+            (3, datetime(2000, 1, 2), expected_3_nofreq),
+        ]
+
+        for n, d, expected in cases:
+            result = idx.insert(n, d)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+    def test_insert3(self, unit):
+        idx = date_range("1/1/2000", periods=3, freq="ME", name="idx", unit=unit)
+
+        # reset freq to None
+        result = idx.insert(3, datetime(2000, 1, 2))
+        expected = DatetimeIndex(
+            ["2000-01-31", "2000-02-29", "2000-03-31", "2000-01-02"],
+            name="idx",
+            freq=None,
+        ).as_unit(unit)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freq is None
+
+    @pytest.mark.parametrize("tz", ["US/Pacific", "Asia/Singapore"])
+    @pytest.mark.parametrize(
+        "to_ts",
+        [lambda x: x, lambda x: x.to_pydatetime()],
+        ids=["Timestamp", "datetime"],
+    )
+    def test_insert4(self, unit, tz, to_ts):
+        idx = date_range(
+            "1/1/2000 09:00", periods=6, freq="h", tz=tz, name="idx", unit=unit
+        )
+        # preserve freq
+        expected = date_range(
+            "1/1/2000 09:00", periods=7, freq="h", tz=tz, name="idx", unit=unit
+        )
+        tz = zoneinfo.ZoneInfo(tz)
+        d = to_ts(Timestamp("2000-01-01 15:00", tz=tz))
+        result = idx.insert(6, d)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freq == expected.freq
+        assert result.tz == expected.tz
+
+    @pytest.mark.parametrize("tz", ["US/Pacific", "Asia/Singapore"])
+    @pytest.mark.parametrize(
+        "to_ts",
+        [lambda x: x, lambda x: x.to_pydatetime()],
+        ids=["Timestamp", "datetime"],
+    )
+    def test_insert4_no_freq(self, unit, tz, to_ts):
+        idx = date_range(
+            "1/1/2000 09:00", periods=6, freq="h", tz=tz, name="idx", unit=unit
+        )
+        expected = DatetimeIndex(
+            [
+                "2000-01-01 09:00",
+                "2000-01-01 10:00",
+                "2000-01-01 11:00",
+                "2000-01-01 12:00",
+                "2000-01-01 13:00",
+                "2000-01-01 14:00",
+                "2000-01-01 10:00",
+            ],
+            name="idx",
+            tz=tz,
+            freq=None,
+        ).as_unit(unit)
+        # reset freq to None
+        d = to_ts(Timestamp("2000-01-01 10:00", tz=tz))
+        result = idx.insert(6, d)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.tz == expected.tz
+        assert result.freq is None
+
+    # TODO: also changes DataFrame.__setitem__ with expansion
+    def test_insert_mismatched_tzawareness(self):
+        # see GH#7299
+        idx = date_range("1/1/2000", periods=3, freq="D", tz="Asia/Tokyo", name="idx")
+
+        # mismatched tz-awareness
+        item = Timestamp("2000-01-04")
+        result = idx.insert(3, item)
+        expected = Index(
+            list(idx[:3]) + [item] + list(idx[3:]), dtype=object, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        # mismatched tz-awareness
+        item = datetime(2000, 1, 4)
+        result = idx.insert(3, item)
+        expected = Index(
+            list(idx[:3]) + [item] + list(idx[3:]), dtype=object, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+    # TODO: also changes DataFrame.__setitem__ with expansion
+    def test_insert_mismatched_tz(self):
+        # see GH#7299
+        # pre-2.0 with mismatched tzs we would cast to object
+        idx = date_range("1/1/2000", periods=3, freq="D", tz="Asia/Tokyo", name="idx")
+
+        # mismatched tz -> cast to object (could reasonably cast to same tz or UTC)
+        item = Timestamp("2000-01-04", tz="US/Eastern")
+        result = idx.insert(3, item)
+        expected = Index(
+            list(idx[:3]) + [item.tz_convert(idx.tz)] + list(idx[3:]),
+            name="idx",
+        )
+        assert expected.dtype == idx.dtype
+        tm.assert_index_equal(result, expected)
+
+        item = datetime(2000, 1, 4, tzinfo=zoneinfo.ZoneInfo("US/Eastern"))
+        result = idx.insert(3, item)
+        expected = Index(
+            list(idx[:3]) + [item.astimezone(idx.tzinfo)] + list(idx[3:]),
+            name="idx",
+        )
+        assert expected.dtype == idx.dtype
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "item", [0, np.int64(0), np.float64(0), np.array(0), np.timedelta64(456)]
+    )
+    def test_insert_mismatched_types_raises(self, tz_aware_fixture, item):
+        # GH#33703 dont cast these to dt64
+        tz = tz_aware_fixture
+        dti = date_range("2019-11-04", periods=9, freq="-1D", name=9, tz=tz)
+
+        result = dti.insert(1, item)
+
+        if isinstance(item, np.ndarray):
+            assert item.item() == 0
+            expected = Index([dti[0], 0] + list(dti[1:]), dtype=object, name=9)
+        else:
+            expected = Index([dti[0], item] + list(dti[1:]), dtype=object, name=9)
+
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_castable_str(self, tz_aware_fixture):
+        # GH#33703
+        tz = tz_aware_fixture
+        dti = date_range("2019-11-04", periods=3, freq="-1D", name=9, tz=tz)
+
+        value = "2019-11-05"
+        result = dti.insert(0, value)
+
+        ts = Timestamp(value).tz_localize(tz)
+        expected = DatetimeIndex([ts] + list(dti), dtype=dti.dtype, name=9)
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_non_castable_str(self, tz_aware_fixture):
+        # GH#33703
+        tz = tz_aware_fixture
+        dti = date_range("2019-11-04", periods=3, freq="-1D", name=9, tz=tz)
+
+        value = "foo"
+        result = dti.insert(0, value)
+
+        expected = Index(["foo"] + list(dti), dtype=object, name=9)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_isocalendar.py b/pandas/pandas/tests/indexes/datetimes/methods/test_isocalendar.py
new file mode 100644
index 0000000000000..97f1003e0f43f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_isocalendar.py
@@ -0,0 +1,28 @@
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_isocalendar_returns_correct_values_close_to_new_year_with_tz():
+    # GH#6538: Check that DatetimeIndex and its TimeStamp elements
+    # return the same weekofyear accessor close to new year w/ tz
+    dates = ["2013/12/29", "2013/12/30", "2013/12/31"]
+    dates = DatetimeIndex(dates, tz="Europe/Brussels")
+    result = dates.isocalendar()
+    expected_data_frame = DataFrame(
+        [[2013, 52, 7], [2014, 1, 1], [2014, 1, 2]],
+        columns=["year", "week", "day"],
+        index=dates,
+        dtype="UInt32",
+    )
+    tm.assert_frame_equal(result, expected_data_frame)
+
+
+def test_dti_timestamp_isocalendar_fields():
+    idx = date_range("2020-01-01", periods=10)
+    expected = tuple(idx.isocalendar().iloc[-1].to_list())
+    result = idx[-1].isocalendar()
+    assert result == expected
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_map.py b/pandas/pandas/tests/indexes/datetimes/methods/test_map.py
new file mode 100644
index 0000000000000..f35f07bd32068
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_map.py
@@ -0,0 +1,47 @@
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Period,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestMap:
+    def test_map(self):
+        rng = date_range("1/1/2000", periods=10)
+
+        f = lambda x: x.strftime("%Y%m%d")
+        result = rng.map(f)
+        exp = Index([f(x) for x in rng])
+        tm.assert_index_equal(result, exp)
+
+    def test_map_fallthrough(self, capsys):
+        # GH#22067, check we don't get warnings about silently ignored errors
+        dti = date_range("2017-01-01", "2018-01-01", freq="B")
+
+        dti.map(lambda x: Period(year=x.year, month=x.month, freq="M"))
+
+        captured = capsys.readouterr()
+        assert captured.err == ""
+
+    def test_map_bug_1677(self):
+        index = DatetimeIndex(["2012-04-25 09:30:00.393000"])
+        f = index.asof
+
+        result = index.map(f)
+        expected = Index([f(index[0])])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("name", [None, "name"])
+    def test_index_map(self, name):
+        # see GH#20990
+        count = 6
+        index = date_range("2018-01-01", periods=count, freq="ME", name=name).map(
+            lambda x: (x.year, x.month)
+        )
+        exp_index = MultiIndex.from_product(((2018,), range(1, 7)), names=[name, name])
+        tm.assert_index_equal(index, exp_index)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_normalize.py b/pandas/pandas/tests/indexes/datetimes/methods/test_normalize.py
new file mode 100644
index 0000000000000..0ce28d60b53b9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_normalize.py
@@ -0,0 +1,99 @@
+from dateutil.tz import tzlocal
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestNormalize:
+    def test_normalize(self):
+        rng = date_range("1/1/2000 9:30", periods=10, freq="D")
+
+        result = rng.normalize()
+        expected = date_range("1/1/2000", periods=10, freq="D")
+        tm.assert_index_equal(result, expected)
+
+        arr_ns = np.array([1380585623454345752, 1380585612343234312]).astype(
+            "datetime64[ns]"
+        )
+        rng_ns = DatetimeIndex(arr_ns)
+        rng_ns_normalized = rng_ns.normalize()
+
+        arr_ns = np.array([1380585600000000000, 1380585600000000000]).astype(
+            "datetime64[ns]"
+        )
+        expected = DatetimeIndex(arr_ns)
+        tm.assert_index_equal(rng_ns_normalized, expected)
+
+        assert result.is_normalized
+        assert not rng.is_normalized
+
+    def test_normalize_nat(self):
+        dti = DatetimeIndex([NaT, Timestamp("2018-01-01 01:00:00")])
+        result = dti.normalize()
+        expected = DatetimeIndex([NaT, Timestamp("2018-01-01")])
+        tm.assert_index_equal(result, expected)
+
+    def test_normalize_tz(self):
+        rng = date_range("1/1/2000 9:30", periods=10, freq="D", tz="US/Eastern")
+
+        result = rng.normalize()  # does not preserve freq
+        expected = date_range("1/1/2000", periods=10, freq="D", tz="US/Eastern")
+        tm.assert_index_equal(result, expected._with_freq(None))
+
+        assert result.is_normalized
+        assert not rng.is_normalized
+
+        rng = date_range("1/1/2000 9:30", periods=10, freq="D", tz="UTC")
+
+        result = rng.normalize()
+        expected = date_range("1/1/2000", periods=10, freq="D", tz="UTC")
+        tm.assert_index_equal(result, expected)
+
+        assert result.is_normalized
+        assert not rng.is_normalized
+
+        rng = date_range("1/1/2000 9:30", periods=10, freq="D", tz=tzlocal())
+        result = rng.normalize()  # does not preserve freq
+        expected = date_range("1/1/2000", periods=10, freq="D", tz=tzlocal())
+        tm.assert_index_equal(result, expected._with_freq(None))
+
+        assert result.is_normalized
+        assert not rng.is_normalized
+
+    @td.skip_if_windows
+    @pytest.mark.skipif(
+        WASM, reason="tzset is available only on Unix-like systems, not WASM"
+    )
+    @pytest.mark.parametrize(
+        "timezone",
+        [
+            "US/Pacific",
+            "US/Eastern",
+            "UTC",
+            "Asia/Kolkata",
+            "Asia/Shanghai",
+            "Australia/Canberra",
+        ],
+    )
+    def test_normalize_tz_local(self, timezone):
+        # GH#13459
+        with tm.set_timezone(timezone):
+            rng = date_range("1/1/2000 9:30", periods=10, freq="D", tz=tzlocal())
+
+            result = rng.normalize()
+            expected = date_range("1/1/2000", periods=10, freq="D", tz=tzlocal())
+            expected = expected._with_freq(None)
+            tm.assert_index_equal(result, expected)
+
+            assert result.is_normalized
+            assert not rng.is_normalized
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_repeat.py b/pandas/pandas/tests/indexes/datetimes/methods/test_repeat.py
new file mode 100644
index 0000000000000..92501755f8c5b
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_repeat.py
@@ -0,0 +1,83 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestRepeat:
+    def test_repeat_range(self, tz_naive_fixture):
+        rng = date_range("1/1/2000", "1/1/2001")
+
+        result = rng.repeat(5)
+        assert result.freq is None
+        assert len(result) == 5 * len(rng)
+
+    def test_repeat_range2(self, tz_naive_fixture, unit):
+        tz = tz_naive_fixture
+        index = date_range("2001-01-01", periods=2, freq="D", tz=tz, unit=unit)
+        exp = DatetimeIndex(
+            ["2001-01-01", "2001-01-01", "2001-01-02", "2001-01-02"], tz=tz
+        ).as_unit(unit)
+        for res in [index.repeat(2), np.repeat(index, 2)]:
+            tm.assert_index_equal(res, exp)
+            assert res.freq is None
+
+    def test_repeat_range3(self, tz_naive_fixture, unit):
+        tz = tz_naive_fixture
+        index = date_range("2001-01-01", periods=2, freq="2D", tz=tz, unit=unit)
+        exp = DatetimeIndex(
+            ["2001-01-01", "2001-01-01", "2001-01-03", "2001-01-03"], tz=tz
+        ).as_unit(unit)
+        for res in [index.repeat(2), np.repeat(index, 2)]:
+            tm.assert_index_equal(res, exp)
+            assert res.freq is None
+
+    def test_repeat_range4(self, tz_naive_fixture, unit):
+        tz = tz_naive_fixture
+        index = DatetimeIndex(["2001-01-01", "NaT", "2003-01-01"], tz=tz).as_unit(unit)
+        exp = DatetimeIndex(
+            [
+                "2001-01-01",
+                "2001-01-01",
+                "2001-01-01",
+                "NaT",
+                "NaT",
+                "NaT",
+                "2003-01-01",
+                "2003-01-01",
+                "2003-01-01",
+            ],
+            tz=tz,
+        ).as_unit(unit)
+        for res in [index.repeat(3), np.repeat(index, 3)]:
+            tm.assert_index_equal(res, exp)
+            assert res.freq is None
+
+    def test_repeat(self, tz_naive_fixture, unit):
+        tz = tz_naive_fixture
+        reps = 2
+        msg = "the 'axis' parameter is not supported"
+
+        rng = date_range(start="2016-01-01", periods=2, freq="30Min", tz=tz, unit=unit)
+
+        expected_rng = DatetimeIndex(
+            [
+                Timestamp("2016-01-01 00:00:00", tz=tz),
+                Timestamp("2016-01-01 00:00:00", tz=tz),
+                Timestamp("2016-01-01 00:30:00", tz=tz),
+                Timestamp("2016-01-01 00:30:00", tz=tz),
+            ]
+        ).as_unit(unit)
+
+        res = rng.repeat(reps)
+        tm.assert_index_equal(res, expected_rng)
+        assert res.freq is None
+
+        tm.assert_index_equal(np.repeat(rng, reps), expected_rng)
+        with pytest.raises(ValueError, match=msg):
+            np.repeat(rng, reps, axis=1)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_resolution.py b/pandas/pandas/tests/indexes/datetimes/methods/test_resolution.py
new file mode 100644
index 0000000000000..42c3ab0617b7c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_resolution.py
@@ -0,0 +1,35 @@
+from dateutil.tz import tzlocal
+import pytest
+
+from pandas.compat import (
+    IS64,
+    WASM,
+)
+
+from pandas import date_range
+
+
+@pytest.mark.parametrize(
+    "freq,expected",
+    [
+        ("YE", "day"),
+        ("QE", "day"),
+        ("ME", "day"),
+        ("D", "day"),
+        ("h", "hour"),
+        ("min", "minute"),
+        ("s", "second"),
+        ("ms", "millisecond"),
+        ("us", "microsecond"),
+    ],
+)
+@pytest.mark.skipif(WASM, reason="OverflowError received on WASM")
+def test_dti_resolution(request, tz_naive_fixture, freq, expected):
+    tz = tz_naive_fixture
+    if freq == "YE" and ((not IS64) or WASM) and isinstance(tz, tzlocal):
+        request.applymarker(
+            pytest.mark.xfail(reason="OverflowError inside tzlocal past 2038")
+        )
+
+    idx = date_range(start="2013-04-01", periods=30, freq=freq, tz=tz)
+    assert idx.resolution == expected
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_round.py b/pandas/pandas/tests/indexes/datetimes/methods/test_round.py
new file mode 100644
index 0000000000000..b023542ba0a4c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_round.py
@@ -0,0 +1,221 @@
+import pytest
+
+from pandas._libs.tslibs import to_offset
+from pandas._libs.tslibs.offsets import INVALID_FREQ_ERR_MSG
+
+from pandas import (
+    DatetimeIndex,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexRound:
+    def test_round_daily(self):
+        dti = date_range("20130101 09:10:11", periods=5)
+        result = dti.round("D")
+        expected = date_range("20130101", periods=5)
+        tm.assert_index_equal(result, expected)
+
+        dti = dti.tz_localize("UTC").tz_convert("US/Eastern")
+        result = dti.round("D")
+        expected = date_range("20130101", periods=5).tz_localize("US/Eastern")
+        tm.assert_index_equal(result, expected)
+
+        result = dti.round("s")
+        tm.assert_index_equal(result, dti)
+
+    @pytest.mark.parametrize(
+        "freq, error_msg",
+        [
+            ("YE", "<YearEnd: month=12> is a non-fixed frequency"),
+            ("ME", "<MonthEnd> is a non-fixed frequency"),
+            ("foobar", "Invalid frequency: foobar"),
+        ],
+    )
+    def test_round_invalid(self, freq, error_msg):
+        dti = date_range("20130101 09:10:11", periods=5)
+        dti = dti.tz_localize("UTC").tz_convert("US/Eastern")
+        with pytest.raises(ValueError, match=error_msg):
+            dti.round(freq)
+
+    def test_round(self, tz_naive_fixture, unit):
+        tz = tz_naive_fixture
+        rng = date_range(start="2016-01-01", periods=5, freq="30Min", tz=tz, unit=unit)
+        elt = rng[1]
+
+        expected_rng = DatetimeIndex(
+            [
+                Timestamp("2016-01-01 00:00:00", tz=tz),
+                Timestamp("2016-01-01 00:00:00", tz=tz),
+                Timestamp("2016-01-01 01:00:00", tz=tz),
+                Timestamp("2016-01-01 02:00:00", tz=tz),
+                Timestamp("2016-01-01 02:00:00", tz=tz),
+            ]
+        ).as_unit(unit)
+        expected_elt = expected_rng[1]
+
+        result = rng.round(freq="h")
+        tm.assert_index_equal(result, expected_rng)
+        assert elt.round(freq="h") == expected_elt
+
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            rng.round(freq="foo")
+        with pytest.raises(ValueError, match=msg):
+            elt.round(freq="foo")
+
+        msg = "<MonthEnd> is a non-fixed frequency"
+        with pytest.raises(ValueError, match=msg):
+            rng.round(freq="ME")
+        with pytest.raises(ValueError, match=msg):
+            elt.round(freq="ME")
+
+    def test_round2(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        # GH#14440 & GH#15578
+        index = DatetimeIndex(["2016-10-17 12:00:00.0015"], tz=tz).as_unit("ns")
+        result = index.round("ms")
+        expected = DatetimeIndex(["2016-10-17 12:00:00.002000"], tz=tz).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+        for freq in ["us", "ns"]:
+            tm.assert_index_equal(index, index.round(freq))
+
+    def test_round3(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        index = DatetimeIndex(["2016-10-17 12:00:00.00149"], tz=tz).as_unit("ns")
+        result = index.round("ms")
+        expected = DatetimeIndex(["2016-10-17 12:00:00.001000"], tz=tz).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_round4(self, tz_naive_fixture):
+        index = DatetimeIndex(["2016-10-17 12:00:00.001501031"], dtype="M8[ns]")
+        result = index.round("10ns")
+        expected = DatetimeIndex(["2016-10-17 12:00:00.001501030"], dtype="M8[ns]")
+        tm.assert_index_equal(result, expected)
+
+        ts = "2016-10-17 12:00:00.001501031"
+        dti = DatetimeIndex([ts], dtype="M8[ns]")
+        with tm.assert_produces_warning(False):
+            dti.round("1010ns")
+
+    def test_no_rounding_occurs(self, tz_naive_fixture):
+        # GH 21262
+        tz = tz_naive_fixture
+        rng = date_range(start="2016-01-01", periods=5, freq="2Min", tz=tz)
+
+        expected_rng = DatetimeIndex(
+            [
+                Timestamp("2016-01-01 00:00:00", tz=tz),
+                Timestamp("2016-01-01 00:02:00", tz=tz),
+                Timestamp("2016-01-01 00:04:00", tz=tz),
+                Timestamp("2016-01-01 00:06:00", tz=tz),
+                Timestamp("2016-01-01 00:08:00", tz=tz),
+            ]
+        ).as_unit("ns")
+
+        result = rng.round(freq="2min")
+        tm.assert_index_equal(result, expected_rng)
+
+    @pytest.mark.parametrize(
+        "test_input, rounder, freq, expected",
+        [
+            (["2117-01-01 00:00:45"], "floor", "15s", ["2117-01-01 00:00:45"]),
+            (["2117-01-01 00:00:45"], "ceil", "15s", ["2117-01-01 00:00:45"]),
+            (
+                ["2117-01-01 00:00:45.000000012"],
+                "floor",
+                "10ns",
+                ["2117-01-01 00:00:45.000000010"],
+            ),
+            (
+                ["1823-01-01 00:00:01.000000012"],
+                "ceil",
+                "10ns",
+                ["1823-01-01 00:00:01.000000020"],
+            ),
+            (["1823-01-01 00:00:01"], "floor", "1s", ["1823-01-01 00:00:01"]),
+            (["1823-01-01 00:00:01"], "ceil", "1s", ["1823-01-01 00:00:01"]),
+            (["2018-01-01 00:15:00"], "ceil", "15min", ["2018-01-01 00:15:00"]),
+            (["2018-01-01 00:15:00"], "floor", "15min", ["2018-01-01 00:15:00"]),
+            (["1823-01-01 03:00:00"], "ceil", "3h", ["1823-01-01 03:00:00"]),
+            (["1823-01-01 03:00:00"], "floor", "3h", ["1823-01-01 03:00:00"]),
+            (
+                ("NaT", "1823-01-01 00:00:01"),
+                "floor",
+                "1s",
+                ("NaT", "1823-01-01 00:00:01"),
+            ),
+            (
+                ("NaT", "1823-01-01 00:00:01"),
+                "ceil",
+                "1s",
+                ("NaT", "1823-01-01 00:00:01"),
+            ),
+        ],
+    )
+    def test_ceil_floor_edge(self, test_input, rounder, freq, expected):
+        dt = DatetimeIndex(list(test_input))
+        func = getattr(dt, rounder)
+        result = func(freq)
+        expected = DatetimeIndex(list(expected))
+        assert expected.equals(result)
+
+    @pytest.mark.parametrize(
+        "start, index_freq, periods",
+        [("2018-01-01", "12h", 25), ("2018-01-01 0:0:0.124999", "1ns", 1000)],
+    )
+    @pytest.mark.parametrize(
+        "round_freq",
+        [
+            "2ns",
+            "3ns",
+            "4ns",
+            "5ns",
+            "6ns",
+            "7ns",
+            "250ns",
+            "500ns",
+            "750ns",
+            "1us",
+            "19us",
+            "250us",
+            "500us",
+            "750us",
+            "1s",
+            "2s",
+            "3s",
+            "12h",
+            "1D",
+        ],
+    )
+    def test_round_int64(self, start, index_freq, periods, round_freq):
+        dt = date_range(start=start, freq=index_freq, periods=periods)
+        unit = to_offset(round_freq).nanos
+
+        # test floor
+        result = dt.floor(round_freq)
+        diff = dt.asi8 - result.asi8
+        mod = result.asi8 % unit
+        assert (mod == 0).all(), f"floor not a {round_freq} multiple"
+        assert (0 <= diff).all() and (diff < unit).all(), "floor error"
+
+        # test ceil
+        result = dt.ceil(round_freq)
+        diff = result.asi8 - dt.asi8
+        mod = result.asi8 % unit
+        assert (mod == 0).all(), f"ceil not a {round_freq} multiple"
+        assert (0 <= diff).all() and (diff < unit).all(), "ceil error"
+
+        # test round
+        result = dt.round(round_freq)
+        diff = abs(result.asi8 - dt.asi8)
+        mod = result.asi8 % unit
+        assert (mod == 0).all(), f"round not a {round_freq} multiple"
+        assert (diff <= unit // 2).all(), "round error"
+        if unit % 2 == 0:
+            assert (result.asi8[diff == unit // 2] % 2 == 0).all(), (
+                "round half to even error"
+            )
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_shift.py b/pandas/pandas/tests/indexes/datetimes/methods/test_shift.py
new file mode 100644
index 0000000000000..a202627550cd2
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_shift.py
@@ -0,0 +1,181 @@
+from datetime import datetime
+import zoneinfo
+
+import pytest
+
+from pandas.errors import NullFrequencyError
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexShift:
+    # -------------------------------------------------------------
+    # DatetimeIndex.shift is used in integer addition
+
+    def test_dti_shift_tzaware(self, tz_naive_fixture, unit):
+        # GH#9903
+        tz = tz_naive_fixture
+        idx = DatetimeIndex([], name="xxx", tz=tz).as_unit(unit)
+        tm.assert_index_equal(idx.shift(0, freq="h"), idx)
+        tm.assert_index_equal(idx.shift(3, freq="h"), idx)
+
+        idx = DatetimeIndex(
+            ["2011-01-01 10:00", "2011-01-01 11:00", "2011-01-01 12:00"],
+            name="xxx",
+            tz=tz,
+            freq="h",
+        ).as_unit(unit)
+        tm.assert_index_equal(idx.shift(0, freq="h"), idx)
+        exp = DatetimeIndex(
+            ["2011-01-01 13:00", "2011-01-01 14:00", "2011-01-01 15:00"],
+            name="xxx",
+            tz=tz,
+            freq="h",
+        ).as_unit(unit)
+        tm.assert_index_equal(idx.shift(3, freq="h"), exp)
+        exp = DatetimeIndex(
+            ["2011-01-01 07:00", "2011-01-01 08:00", "2011-01-01 09:00"],
+            name="xxx",
+            tz=tz,
+            freq="h",
+        ).as_unit(unit)
+        tm.assert_index_equal(idx.shift(-3, freq="h"), exp)
+
+    def test_dti_shift_freqs(self, unit):
+        # test shift for DatetimeIndex and non DatetimeIndex
+        # GH#8083
+        drange = date_range("20130101", periods=5, unit=unit)
+        result = drange.shift(1)
+        expected = DatetimeIndex(
+            ["2013-01-02", "2013-01-03", "2013-01-04", "2013-01-05", "2013-01-06"],
+            dtype=f"M8[{unit}]",
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = drange.shift(-1)
+        expected = DatetimeIndex(
+            ["2012-12-31", "2013-01-01", "2013-01-02", "2013-01-03", "2013-01-04"],
+            dtype=f"M8[{unit}]",
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = drange.shift(3, freq="2D")
+        expected = DatetimeIndex(
+            ["2013-01-07", "2013-01-08", "2013-01-09", "2013-01-10", "2013-01-11"],
+            dtype=f"M8[{unit}]",
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_shift_int(self, unit):
+        rng = date_range("1/1/2000", periods=20, unit=unit)
+
+        result = rng + 5 * rng.freq
+        expected = rng.shift(5)
+        tm.assert_index_equal(result, expected)
+
+        result = rng - 5 * rng.freq
+        expected = rng.shift(-5)
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_shift_no_freq(self, unit):
+        # GH#19147
+        dti = DatetimeIndex(["2011-01-01 10:00", "2011-01-01"], freq=None).as_unit(unit)
+        with pytest.raises(NullFrequencyError, match="Cannot shift with no freq"):
+            dti.shift(2)
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_shift_localized(self, tzstr, unit):
+        dr = date_range("2011/1/1", "2012/1/1", freq="W-FRI", unit=unit)
+        dr_tz = dr.tz_localize(tzstr)
+
+        result = dr_tz.shift(1, "10min")
+        assert result.tz == dr_tz.tz
+
+    def test_dti_shift_across_dst(self, unit):
+        # GH 8616
+        idx = date_range(
+            "2013-11-03", tz="America/Chicago", periods=7, freq="h", unit=unit
+        )
+        ser = Series(index=idx[:-1], dtype=object)
+        result = ser.shift(freq="h")
+        expected = Series(index=idx[1:], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "shift, result_time",
+        [
+            [0, "2014-11-14 00:00:00"],
+            [-1, "2014-11-13 23:00:00"],
+            [1, "2014-11-14 01:00:00"],
+        ],
+    )
+    def test_dti_shift_near_midnight(self, shift, result_time, unit):
+        # GH 8616
+        tz = zoneinfo.ZoneInfo("US/Eastern")
+        dt_est = datetime(2014, 11, 14, 0, tzinfo=tz)
+        idx = DatetimeIndex([dt_est]).as_unit(unit)
+        ser = Series(data=[1], index=idx)
+        result = ser.shift(shift, freq="h")
+        exp_index = DatetimeIndex([result_time], tz=tz).as_unit(unit)
+        expected = Series(1, index=exp_index)
+        tm.assert_series_equal(result, expected)
+
+    def test_shift_periods(self, unit):
+        # GH#22458 : argument 'n' was deprecated in favor of 'periods'
+        idx = date_range(
+            start=datetime(2009, 1, 1), end=datetime(2010, 1, 1), periods=3, unit=unit
+        )
+        tm.assert_index_equal(idx.shift(periods=0), idx)
+        tm.assert_index_equal(idx.shift(0), idx)
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_shift_bday(self, freq, unit):
+        rng = date_range(
+            datetime(2009, 1, 1), datetime(2010, 1, 1), freq=freq, unit=unit
+        )
+        shifted = rng.shift(5)
+        assert shifted[0] == rng[5]
+        assert shifted.freq == rng.freq
+
+        shifted = rng.shift(-5)
+        assert shifted[5] == rng[0]
+        assert shifted.freq == rng.freq
+
+        shifted = rng.shift(0)
+        assert shifted[0] == rng[0]
+        assert shifted.freq == rng.freq
+
+    def test_shift_bmonth(self, performance_warning, unit):
+        rng = date_range(
+            datetime(2009, 1, 1),
+            datetime(2010, 1, 1),
+            freq=pd.offsets.BMonthEnd(),
+            unit=unit,
+        )
+        shifted = rng.shift(1, freq=pd.offsets.BDay())
+        assert shifted[0] == rng[0] + pd.offsets.BDay()
+
+        rng = date_range(
+            datetime(2009, 1, 1),
+            datetime(2010, 1, 1),
+            freq=pd.offsets.BMonthEnd(),
+            unit=unit,
+        )
+        with tm.assert_produces_warning(performance_warning):
+            shifted = rng.shift(1, freq=pd.offsets.CDay())
+            assert shifted[0] == rng[0] + pd.offsets.CDay()
+
+    def test_shift_empty(self, unit):
+        # GH#14811
+        dti = date_range(start="2016-10-21", end="2016-10-21", freq="BME", unit=unit)
+        result = dti.shift(1)
+        tm.assert_index_equal(result, dti)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_snap.py b/pandas/pandas/tests/indexes/datetimes/methods/test_snap.py
new file mode 100644
index 0000000000000..a3c06ac6257cf
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_snap.py
@@ -0,0 +1,52 @@
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.filterwarnings("ignore:Period with BDay freq:FutureWarning")
+@pytest.mark.parametrize("tz", [None, "Asia/Shanghai", "Europe/Berlin"])
+@pytest.mark.parametrize("name", [None, "my_dti"])
+def test_dti_snap(name, tz, unit):
+    dti = DatetimeIndex(
+        [
+            "1/1/2002",
+            "1/2/2002",
+            "1/3/2002",
+            "1/4/2002",
+            "1/5/2002",
+            "1/6/2002",
+            "1/7/2002",
+        ],
+        name=name,
+        tz=tz,
+        freq="D",
+    )
+    dti = dti.as_unit(unit)
+
+    result = dti.snap(freq="W-MON")
+    msg = "'w-mon' is deprecated and will be removed in a future version."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        expected = date_range("12/31/2001", "1/7/2002", name=name, tz=tz, freq="w-mon")
+    expected = expected.repeat([3, 4])
+    expected = expected.as_unit(unit)
+    tm.assert_index_equal(result, expected)
+    assert result.tz == expected.tz
+    assert result.freq is None
+    assert expected.freq is None
+
+    result = dti.snap(freq="B")
+
+    msg = "'b' is deprecated and will be removed in a future version."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        expected = date_range("1/1/2002", "1/7/2002", name=name, tz=tz, freq="b")
+    expected = expected.repeat([1, 1, 1, 2, 2])
+    expected = expected.as_unit(unit)
+    tm.assert_index_equal(result, expected)
+    assert result.tz == expected.tz
+    assert result.freq is None
+    assert expected.freq is None
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_to_frame.py b/pandas/pandas/tests/indexes/datetimes/methods/test_to_frame.py
new file mode 100644
index 0000000000000..c829109d4e06c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_to_frame.py
@@ -0,0 +1,28 @@
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestToFrame:
+    def test_to_frame_datetime_tz(self):
+        # GH#25809
+        idx = date_range(start="2019-01-01", end="2019-01-30", freq="D", tz="UTC")
+        result = idx.to_frame()
+        expected = DataFrame(idx, index=idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_frame_respects_none_name(self):
+        # GH#44212 if we explicitly pass name=None, then that should be respected,
+        #  not changed to 0
+        # GH-45448 this is first deprecated to only change in the future
+        idx = date_range(start="2019-01-01", end="2019-01-30", freq="D", tz="UTC")
+        result = idx.to_frame(name=None)
+        exp_idx = Index([None], dtype=object)
+        tm.assert_index_equal(exp_idx, result.columns)
+
+        result = idx.rename("foo").to_frame(name=None)
+        exp_idx = Index([None], dtype=object)
+        tm.assert_index_equal(exp_idx, result.columns)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_to_julian_date.py b/pandas/pandas/tests/indexes/datetimes/methods/test_to_julian_date.py
new file mode 100644
index 0000000000000..fc1f0595c21c5
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_to_julian_date.py
@@ -0,0 +1,45 @@
+import numpy as np
+
+from pandas import (
+    Index,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDateTimeIndexToJulianDate:
+    def test_1700(self):
+        dr = date_range(start=Timestamp("1710-10-01"), periods=5, freq="D")
+        r1 = Index([x.to_julian_date() for x in dr])
+        r2 = dr.to_julian_date()
+        assert isinstance(r2, Index) and r2.dtype == np.float64
+        tm.assert_index_equal(r1, r2)
+
+    def test_2000(self):
+        dr = date_range(start=Timestamp("2000-02-27"), periods=5, freq="D")
+        r1 = Index([x.to_julian_date() for x in dr])
+        r2 = dr.to_julian_date()
+        assert isinstance(r2, Index) and r2.dtype == np.float64
+        tm.assert_index_equal(r1, r2)
+
+    def test_hour(self):
+        dr = date_range(start=Timestamp("2000-02-27"), periods=5, freq="h")
+        r1 = Index([x.to_julian_date() for x in dr])
+        r2 = dr.to_julian_date()
+        assert isinstance(r2, Index) and r2.dtype == np.float64
+        tm.assert_index_equal(r1, r2)
+
+    def test_minute(self):
+        dr = date_range(start=Timestamp("2000-02-27"), periods=5, freq="min")
+        r1 = Index([x.to_julian_date() for x in dr])
+        r2 = dr.to_julian_date()
+        assert isinstance(r2, Index) and r2.dtype == np.float64
+        tm.assert_index_equal(r1, r2)
+
+    def test_second(self):
+        dr = date_range(start=Timestamp("2000-02-27"), periods=5, freq="s")
+        r1 = Index([x.to_julian_date() for x in dr])
+        r2 = dr.to_julian_date()
+        assert isinstance(r2, Index) and r2.dtype == np.float64
+        tm.assert_index_equal(r1, r2)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_to_period.py b/pandas/pandas/tests/indexes/datetimes/methods/test_to_period.py
new file mode 100644
index 0000000000000..cd4a142dd5b30
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_to_period.py
@@ -0,0 +1,220 @@
+from datetime import timezone
+
+import dateutil.tz
+from dateutil.tz import tzlocal
+import pytest
+
+from pandas._libs.tslibs.ccalendar import MONTHS
+from pandas._libs.tslibs.offsets import MonthEnd
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+
+from pandas import (
+    DatetimeIndex,
+    Period,
+    PeriodIndex,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestToPeriod:
+    def test_dti_to_period(self):
+        dti = date_range(start="1/1/2005", end="12/1/2005", freq="ME")
+        pi1 = dti.to_period()
+        pi2 = dti.to_period(freq="D")
+        pi3 = dti.to_period(freq="3D")
+
+        assert pi1[0] == Period("Jan 2005", freq="M")
+        assert pi2[0] == Period("1/31/2005", freq="D")
+        assert pi3[0] == Period("1/31/2005", freq="3D")
+
+        assert pi1[-1] == Period("Nov 2005", freq="M")
+        assert pi2[-1] == Period("11/30/2005", freq="D")
+        assert pi3[-1], Period("11/30/2005", freq="3D")
+
+        tm.assert_index_equal(pi1, period_range("1/1/2005", "11/1/2005", freq="M"))
+        tm.assert_index_equal(
+            pi2, period_range("1/1/2005", "11/1/2005", freq="M").asfreq("D")
+        )
+        tm.assert_index_equal(
+            pi3, period_range("1/1/2005", "11/1/2005", freq="M").asfreq("3D")
+        )
+
+    @pytest.mark.parametrize("month", MONTHS)
+    def test_to_period_quarterly(self, month):
+        # make sure we can make the round trip
+        freq = f"Q-{month}"
+        rng = period_range("1989Q3", "1991Q3", freq=freq)
+        stamps = rng.to_timestamp()
+        result = stamps.to_period(freq)
+        tm.assert_index_equal(rng, result)
+
+    @pytest.mark.parametrize("off", ["BQE", "QS", "BQS"])
+    def test_to_period_quarterlyish(self, off):
+        rng = date_range("01-Jan-2012", periods=8, freq=off)
+        prng = rng.to_period()
+        assert prng.freq == "QE-DEC"
+
+    @pytest.mark.parametrize("off", ["BYE", "YS", "BYS"])
+    def test_to_period_annualish(self, off):
+        rng = date_range("01-Jan-2012", periods=8, freq=off)
+        prng = rng.to_period()
+        assert prng.freq == "YE-DEC"
+
+    def test_to_period_monthish(self):
+        offsets = ["MS", "BME"]
+        for off in offsets:
+            rng = date_range("01-Jan-2012", periods=8, freq=off)
+            prng = rng.to_period()
+            assert prng.freqstr == "M"
+
+        rng = date_range("01-Jan-2012", periods=8, freq="ME")
+        prng = rng.to_period()
+        assert prng.freqstr == "M"
+
+        with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+            date_range("01-Jan-2012", periods=8, freq="EOM")
+
+    @pytest.mark.parametrize(
+        "freq_offset, freq_period",
+        [
+            ("2ME", "2M"),
+            (MonthEnd(2), MonthEnd(2)),
+        ],
+    )
+    def test_dti_to_period_2monthish(self, freq_offset, freq_period):
+        dti = date_range("2020-01-01", periods=3, freq=freq_offset)
+        pi = dti.to_period()
+
+        tm.assert_index_equal(pi, period_range("2020-01", "2020-05", freq=freq_period))
+
+    @pytest.mark.parametrize(
+        "freq", ["2ME", "1me", "2QE", "2QE-SEP", "1YE", "ye", "2YE-MAR"]
+    )
+    def test_to_period_frequency_M_Q_Y_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        rng = date_range("01-Jan-2012", periods=8, freq="ME")
+        with pytest.raises(ValueError, match=msg):
+            rng.to_period(freq)
+
+    def test_to_period_infer(self):
+        # https://github.com/pandas-dev/pandas/issues/33358
+        rng = date_range(
+            start="2019-12-22 06:40:00+00:00",
+            end="2019-12-22 08:45:00+00:00",
+            freq="5min",
+        )
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            pi1 = rng.to_period("5min")
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            pi2 = rng.to_period()
+
+        tm.assert_index_equal(pi1, pi2)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_period_dt64_round_trip(self):
+        dti = date_range("1/1/2000", "1/7/2002", freq="B")
+        pi = dti.to_period()
+        tm.assert_index_equal(pi.to_timestamp(), dti)
+
+        dti = date_range("1/1/2000", "1/7/2002", freq="B")
+        pi = dti.to_period(freq="h")
+        tm.assert_index_equal(pi.to_timestamp(), dti)
+
+    def test_to_period_millisecond(self):
+        index = DatetimeIndex(
+            [
+                Timestamp("2007-01-01 10:11:12.123456Z"),
+                Timestamp("2007-01-01 10:11:13.789123Z"),
+            ]
+        )
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            period = index.to_period(freq="ms")
+        assert 2 == len(period)
+        assert period[0] == Period("2007-01-01 10:11:12.123Z", "ms")
+        assert period[1] == Period("2007-01-01 10:11:13.789Z", "ms")
+
+    def test_to_period_microsecond(self):
+        index = DatetimeIndex(
+            [
+                Timestamp("2007-01-01 10:11:12.123456Z"),
+                Timestamp("2007-01-01 10:11:13.789123Z"),
+            ]
+        )
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            period = index.to_period(freq="us")
+        assert 2 == len(period)
+        assert period[0] == Period("2007-01-01 10:11:12.123456Z", "us")
+        assert period[1] == Period("2007-01-01 10:11:13.789123Z", "us")
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "US/Eastern",
+            timezone.utc,
+            tzlocal(),
+            "dateutil/US/Eastern",
+            dateutil.tz.tzutc(),
+        ],
+    )
+    def test_to_period_tz(self, tz):
+        ts = date_range("1/1/2000", "2/1/2000", tz=tz)
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            result = ts.to_period()[0]
+            expected = ts[0].to_period(ts.freq)
+
+        assert result == expected
+
+        expected = date_range("1/1/2000", "2/1/2000").to_period()
+
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            result = ts.to_period(ts.freq)
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["Etc/GMT-1", "Etc/GMT+1"])
+    def test_to_period_tz_utc_offset_consistency(self, tz):
+        # GH#22905
+        ts = date_range("1/1/2000", "2/1/2000", tz="Etc/GMT-1")
+        with tm.assert_produces_warning(UserWarning, match="drop timezone info"):
+            result = ts.to_period()[0]
+            expected = ts[0].to_period(ts.freq)
+            assert result == expected
+
+    def test_to_period_nofreq(self):
+        idx = DatetimeIndex(["2000-01-01", "2000-01-02", "2000-01-04"])
+        msg = "You must pass a freq argument as current index has none."
+        with pytest.raises(ValueError, match=msg):
+            idx.to_period()
+
+        idx = DatetimeIndex(["2000-01-01", "2000-01-02", "2000-01-03"], freq="infer")
+        assert idx.freqstr == "D"
+        expected = PeriodIndex(["2000-01-01", "2000-01-02", "2000-01-03"], freq="D")
+        tm.assert_index_equal(idx.to_period(), expected)
+
+        # GH#7606
+        idx = DatetimeIndex(["2000-01-01", "2000-01-02", "2000-01-03"])
+        assert idx.freqstr is None
+        tm.assert_index_equal(idx.to_period(), expected)
+
+    @pytest.mark.parametrize("freq", ["2BME", "SME-15", "2BMS"])
+    def test_to_period_offsets_not_supported(self, freq):
+        # GH#56243
+        msg = "|".join(
+            [
+                f"Invalid frequency: {freq}",
+                f"{freq} is not supported as period frequency",
+            ]
+        )
+
+        ts = date_range("1/1/2012", periods=4, freq=freq)
+        with pytest.raises(ValueError, match=msg):
+            ts.to_period()
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_to_pydatetime.py b/pandas/pandas/tests/indexes/datetimes/methods/test_to_pydatetime.py
new file mode 100644
index 0000000000000..fe97ff0cca8eb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_to_pydatetime.py
@@ -0,0 +1,51 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import dateutil.parser
+import dateutil.tz
+from dateutil.tz import tzlocal
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.tests.indexes.datetimes.test_timezones import FixedOffset
+
+fixed_off = FixedOffset(-420, "-07:00")
+
+
+class TestToPyDatetime:
+    def test_dti_to_pydatetime(self):
+        dt = dateutil.parser.parse("2012-06-13T01:39:00Z")
+        dt = dt.replace(tzinfo=tzlocal())
+
+        arr = np.array([dt], dtype=object)
+
+        result = to_datetime(arr, utc=True)
+        assert result.tz is timezone.utc
+
+        rng = date_range("2012-11-03 03:00", "2012-11-05 03:00", tz=tzlocal())
+        arr = rng.to_pydatetime()
+        result = to_datetime(arr, utc=True)
+        assert result.tz is timezone.utc
+
+    def test_dti_to_pydatetime_fizedtz(self):
+        dates = np.array(
+            [
+                datetime(2000, 1, 1, tzinfo=fixed_off),
+                datetime(2000, 1, 2, tzinfo=fixed_off),
+                datetime(2000, 1, 3, tzinfo=fixed_off),
+            ]
+        )
+        dti = DatetimeIndex(dates)
+
+        result = dti.to_pydatetime()
+        tm.assert_numpy_array_equal(dates, result)
+
+        result = dti._mpl_repr()
+        tm.assert_numpy_array_equal(dates, result)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_to_series.py b/pandas/pandas/tests/indexes/datetimes/methods/test_to_series.py
new file mode 100644
index 0000000000000..cd67775b7a5fc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_to_series.py
@@ -0,0 +1,18 @@
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestToSeries:
+    def test_to_series(self):
+        naive = DatetimeIndex(["2013-1-1 13:00", "2013-1-2 14:00"], name="B")
+        idx = naive.tz_localize("US/Pacific")
+
+        expected = Series(np.array(idx.tolist(), dtype="object"), name="B")
+        result = idx.to_series(index=range(2))
+        assert expected.dtype == idx.dtype
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_tz_convert.py b/pandas/pandas/tests/indexes/datetimes/methods/test_tz_convert.py
new file mode 100644
index 0000000000000..9eabb742b93a4
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_tz_convert.py
@@ -0,0 +1,285 @@
+from datetime import datetime
+
+import dateutil.tz
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import timezones
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    NaT,
+    Timestamp,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+
+
+class TestTZConvert:
+    def test_tz_convert_nat(self):
+        # GH#5546
+        dates = [NaT]
+        idx = DatetimeIndex(dates)
+        idx = idx.tz_localize("US/Pacific")
+        tm.assert_index_equal(idx, DatetimeIndex(dates, tz="US/Pacific"))
+        idx = idx.tz_convert("US/Eastern")
+        tm.assert_index_equal(idx, DatetimeIndex(dates, tz="US/Eastern"))
+        idx = idx.tz_convert("UTC")
+        tm.assert_index_equal(idx, DatetimeIndex(dates, tz="UTC"))
+
+        dates = ["2010-12-01 00:00", "2010-12-02 00:00", NaT]
+        idx = DatetimeIndex(dates)
+        idx = idx.tz_localize("US/Pacific")
+        tm.assert_index_equal(idx, DatetimeIndex(dates, tz="US/Pacific"))
+        idx = idx.tz_convert("US/Eastern")
+        expected = ["2010-12-01 03:00", "2010-12-02 03:00", NaT]
+        tm.assert_index_equal(idx, DatetimeIndex(expected, tz="US/Eastern"))
+
+        idx = idx + offsets.Hour(5)
+        expected = ["2010-12-01 08:00", "2010-12-02 08:00", NaT]
+        tm.assert_index_equal(idx, DatetimeIndex(expected, tz="US/Eastern"))
+        idx = idx.tz_convert("US/Pacific")
+        expected = ["2010-12-01 05:00", "2010-12-02 05:00", NaT]
+        tm.assert_index_equal(idx, DatetimeIndex(expected, tz="US/Pacific"))
+
+        idx = idx + np.timedelta64(3, "h")
+        expected = ["2010-12-01 08:00", "2010-12-02 08:00", NaT]
+        tm.assert_index_equal(idx, DatetimeIndex(expected, tz="US/Pacific"))
+
+        idx = idx.tz_convert("US/Eastern")
+        expected = ["2010-12-01 11:00", "2010-12-02 11:00", NaT]
+        tm.assert_index_equal(idx, DatetimeIndex(expected, tz="US/Eastern"))
+
+    @pytest.mark.parametrize("prefix", ["", "dateutil/"])
+    def test_dti_tz_convert_compat_timestamp(self, prefix):
+        strdates = ["1/1/2012", "3/1/2012", "4/1/2012"]
+        idx = DatetimeIndex(strdates, tz=prefix + "US/Eastern")
+
+        conv = idx[0].tz_convert(prefix + "US/Pacific")
+        expected = idx.tz_convert(prefix + "US/Pacific")[0]
+
+        assert conv == expected
+
+    def test_dti_tz_convert_hour_overflow_dst(self):
+        # Regression test for GH#13306
+
+        # sorted case US/Eastern -> UTC
+        ts = ["2008-05-12 09:50:00", "2008-12-12 09:50:35", "2009-05-12 09:50:32"]
+        tt = DatetimeIndex(ts).tz_localize("US/Eastern")
+        ut = tt.tz_convert("UTC")
+        expected = Index([13, 14, 13], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # sorted case UTC -> US/Eastern
+        ts = ["2008-05-12 13:50:00", "2008-12-12 14:50:35", "2009-05-12 13:50:32"]
+        tt = DatetimeIndex(ts).tz_localize("UTC")
+        ut = tt.tz_convert("US/Eastern")
+        expected = Index([9, 9, 9], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # unsorted case US/Eastern -> UTC
+        ts = ["2008-05-12 09:50:00", "2008-12-12 09:50:35", "2008-05-12 09:50:32"]
+        tt = DatetimeIndex(ts).tz_localize("US/Eastern")
+        ut = tt.tz_convert("UTC")
+        expected = Index([13, 14, 13], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # unsorted case UTC -> US/Eastern
+        ts = ["2008-05-12 13:50:00", "2008-12-12 14:50:35", "2008-05-12 13:50:32"]
+        tt = DatetimeIndex(ts).tz_localize("UTC")
+        ut = tt.tz_convert("US/Eastern")
+        expected = Index([9, 9, 9], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_tz_convert_hour_overflow_dst_timestamps(self, tz):
+        # Regression test for GH#13306
+
+        # sorted case US/Eastern -> UTC
+        ts = [
+            Timestamp("2008-05-12 09:50:00", tz=tz),
+            Timestamp("2008-12-12 09:50:35", tz=tz),
+            Timestamp("2009-05-12 09:50:32", tz=tz),
+        ]
+        tt = DatetimeIndex(ts)
+        ut = tt.tz_convert("UTC")
+        expected = Index([13, 14, 13], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # sorted case UTC -> US/Eastern
+        ts = [
+            Timestamp("2008-05-12 13:50:00", tz="UTC"),
+            Timestamp("2008-12-12 14:50:35", tz="UTC"),
+            Timestamp("2009-05-12 13:50:32", tz="UTC"),
+        ]
+        tt = DatetimeIndex(ts)
+        ut = tt.tz_convert("US/Eastern")
+        expected = Index([9, 9, 9], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # unsorted case US/Eastern -> UTC
+        ts = [
+            Timestamp("2008-05-12 09:50:00", tz=tz),
+            Timestamp("2008-12-12 09:50:35", tz=tz),
+            Timestamp("2008-05-12 09:50:32", tz=tz),
+        ]
+        tt = DatetimeIndex(ts)
+        ut = tt.tz_convert("UTC")
+        expected = Index([13, 14, 13], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+        # unsorted case UTC -> US/Eastern
+        ts = [
+            Timestamp("2008-05-12 13:50:00", tz="UTC"),
+            Timestamp("2008-12-12 14:50:35", tz="UTC"),
+            Timestamp("2008-05-12 13:50:32", tz="UTC"),
+        ]
+        tt = DatetimeIndex(ts)
+        ut = tt.tz_convert("US/Eastern")
+        expected = Index([9, 9, 9], dtype=np.int32)
+        tm.assert_index_equal(ut.hour, expected)
+
+    @pytest.mark.parametrize("freq, n", [("h", 1), ("min", 60), ("s", 3600)])
+    def test_dti_tz_convert_trans_pos_plus_1__bug(self, freq, n):
+        # Regression test for tslib.tz_convert(vals, tz1, tz2).
+        # See GH#4496 for details.
+        idx = date_range(datetime(2011, 3, 26, 23), datetime(2011, 3, 27, 1), freq=freq)
+        idx = idx.tz_localize("UTC")
+        idx = idx.tz_convert("Europe/Moscow")
+
+        expected = np.repeat(np.array([3, 4, 5]), np.array([n, n, 1]))
+        tm.assert_index_equal(idx.hour, Index(expected, dtype=np.int32))
+
+    def test_dti_tz_convert_dst(self):
+        for freq, n in [("h", 1), ("min", 60), ("s", 3600)]:
+            # Start DST
+            idx = date_range(
+                "2014-03-08 23:00", "2014-03-09 09:00", freq=freq, tz="UTC"
+            )
+            idx = idx.tz_convert("US/Eastern")
+            expected = np.repeat(
+                np.array([18, 19, 20, 21, 22, 23, 0, 1, 3, 4, 5]),
+                np.array([n, n, n, n, n, n, n, n, n, n, 1]),
+            )
+            tm.assert_index_equal(idx.hour, Index(expected, dtype=np.int32))
+
+            idx = date_range(
+                "2014-03-08 18:00", "2014-03-09 05:00", freq=freq, tz="US/Eastern"
+            )
+            idx = idx.tz_convert("UTC")
+            expected = np.repeat(
+                np.array([23, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9]),
+                np.array([n, n, n, n, n, n, n, n, n, n, 1]),
+            )
+            tm.assert_index_equal(idx.hour, Index(expected, dtype=np.int32))
+
+            # End DST
+            idx = date_range(
+                "2014-11-01 23:00", "2014-11-02 09:00", freq=freq, tz="UTC"
+            )
+            idx = idx.tz_convert("US/Eastern")
+            expected = np.repeat(
+                np.array([19, 20, 21, 22, 23, 0, 1, 1, 2, 3, 4]),
+                np.array([n, n, n, n, n, n, n, n, n, n, 1]),
+            )
+            tm.assert_index_equal(idx.hour, Index(expected, dtype=np.int32))
+
+            idx = date_range(
+                "2014-11-01 18:00", "2014-11-02 05:00", freq=freq, tz="US/Eastern"
+            )
+            idx = idx.tz_convert("UTC")
+            expected = np.repeat(
+                np.array([22, 23, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]),
+                np.array([n, n, n, n, n, n, n, n, n, n, n, n, 1]),
+            )
+            tm.assert_index_equal(idx.hour, Index(expected, dtype=np.int32))
+
+        # daily
+        # Start DST
+        idx = date_range("2014-03-08 00:00", "2014-03-09 00:00", freq="D", tz="UTC")
+        idx = idx.tz_convert("US/Eastern")
+        tm.assert_index_equal(idx.hour, Index([19, 19], dtype=np.int32))
+
+        idx = date_range(
+            "2014-03-08 00:00", "2014-03-09 00:00", freq="D", tz="US/Eastern"
+        )
+        idx = idx.tz_convert("UTC")
+        tm.assert_index_equal(idx.hour, Index([5, 5], dtype=np.int32))
+
+        # End DST
+        idx = date_range("2014-11-01 00:00", "2014-11-02 00:00", freq="D", tz="UTC")
+        idx = idx.tz_convert("US/Eastern")
+        tm.assert_index_equal(idx.hour, Index([20, 20], dtype=np.int32))
+
+        idx = date_range(
+            "2014-11-01 00:00", "2014-11-02 000:00", freq="D", tz="US/Eastern"
+        )
+        idx = idx.tz_convert("UTC")
+        tm.assert_index_equal(idx.hour, Index([4, 4], dtype=np.int32))
+
+    def test_tz_convert_roundtrip(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        idx1 = date_range(start="2014-01-01", end="2014-12-31", freq="ME", tz="UTC")
+        exp1 = date_range(start="2014-01-01", end="2014-12-31", freq="ME")
+
+        idx2 = date_range(start="2014-01-01", end="2014-12-31", freq="D", tz="UTC")
+        exp2 = date_range(start="2014-01-01", end="2014-12-31", freq="D")
+
+        idx3 = date_range(start="2014-01-01", end="2014-03-01", freq="h", tz="UTC")
+        exp3 = date_range(start="2014-01-01", end="2014-03-01", freq="h")
+
+        idx4 = date_range(start="2014-08-01", end="2014-10-31", freq="min", tz="UTC")
+        exp4 = date_range(start="2014-08-01", end="2014-10-31", freq="min")
+
+        for idx, expected in [(idx1, exp1), (idx2, exp2), (idx3, exp3), (idx4, exp4)]:
+            converted = idx.tz_convert(tz)
+            reset = converted.tz_convert(None)
+            tm.assert_index_equal(reset, expected)
+            assert reset.tzinfo is None
+            expected = converted.tz_convert("UTC").tz_localize(None)
+            expected = expected._with_freq("infer")
+            tm.assert_index_equal(reset, expected)
+
+    def test_dti_tz_convert_tzlocal(self):
+        # GH#13583
+        # tz_convert doesn't affect to internal
+        dti = date_range(start="2001-01-01", end="2001-03-01", tz="UTC")
+        dti2 = dti.tz_convert(dateutil.tz.tzlocal())
+        tm.assert_numpy_array_equal(dti2.asi8, dti.asi8)
+
+        dti = date_range(start="2001-01-01", end="2001-03-01", tz=dateutil.tz.tzlocal())
+        dti2 = dti.tz_convert(None)
+        tm.assert_numpy_array_equal(dti2.asi8, dti.asi8)
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "US/Eastern",
+            "dateutil/US/Eastern",
+            "pytz/US/Eastern",
+            gettz("US/Eastern"),
+        ],
+    )
+    def test_dti_tz_convert_utc_to_local_no_modify(self, tz):
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        rng = date_range("3/11/2012", "3/12/2012", freq="h", tz="utc")
+        rng_eastern = rng.tz_convert(tz)
+
+        # Values are unmodified
+        tm.assert_numpy_array_equal(rng.asi8, rng_eastern.asi8)
+
+        assert timezones.tz_compare(rng_eastern.tz, timezones.maybe_get_tz(tz))
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_tz_convert_unsorted(self, tzstr):
+        dr = date_range("2012-03-09", freq="h", periods=100, tz="utc")
+        dr = dr.tz_convert(tzstr)
+
+        result = dr[::-1].hour
+        exp = dr.hour[::-1]
+        tm.assert_almost_equal(result, exp)
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_tz_localize.py b/pandas/pandas/tests/indexes/datetimes/methods/test_tz_localize.py
new file mode 100644
index 0000000000000..78a79ac7d1546
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_tz_localize.py
@@ -0,0 +1,378 @@
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+from zoneinfo import ZoneInfo
+
+import dateutil.tz
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Timestamp,
+    bdate_range,
+    date_range,
+    offsets,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(params=["pytz/US/Eastern", gettz("US/Eastern"), ZoneInfo("US/Eastern")])
+def tz(request):
+    if isinstance(request.param, str) and request.param.startswith("pytz/"):
+        pytz = pytest.importorskip("pytz")
+        return pytz.timezone(request.param.removeprefix("pytz/"))
+    return request.param
+
+
+class TestTZLocalize:
+    def test_tz_localize_invalidates_freq(self):
+        # we only preserve freq in unambiguous cases
+
+        # if localized to US/Eastern, this crosses a DST transition
+        dti = date_range("2014-03-08 23:00", "2014-03-09 09:00", freq="h")
+        assert dti.freq == "h"
+
+        result = dti.tz_localize(None)  # no-op
+        assert result.freq == "h"
+
+        result = dti.tz_localize("UTC")  # unambiguous freq preservation
+        assert result.freq == "h"
+
+        result = dti.tz_localize("US/Eastern", nonexistent="shift_forward")
+        assert result.freq is None
+        assert result.inferred_freq is None  # i.e. we are not _too_ strict here
+
+        # Case where we _can_ keep freq because we're length==1
+        dti2 = dti[:1]
+        result = dti2.tz_localize("US/Eastern")
+        assert result.freq == "h"
+
+    def test_tz_localize_utc_copies(self, utc_fixture):
+        # GH#46460
+        times = ["2015-03-08 01:00", "2015-03-08 02:00", "2015-03-08 03:00"]
+        index = DatetimeIndex(times)
+
+        res = index.tz_localize(utc_fixture)
+        assert not tm.shares_memory(res, index)
+
+        res2 = index._data.tz_localize(utc_fixture)
+        assert not tm.shares_memory(index._data, res2)
+
+    def test_dti_tz_localize_nonexistent_raise_coerce(self):
+        # GH#13057
+        times = ["2015-03-08 01:00", "2015-03-08 02:00", "2015-03-08 03:00"]
+        index = DatetimeIndex(times)
+        tz = "US/Eastern"
+        with pytest.raises(ValueError, match="|".join(times)):
+            index.tz_localize(tz=tz)
+
+        with pytest.raises(ValueError, match="|".join(times)):
+            index.tz_localize(tz=tz, nonexistent="raise")
+
+        result = index.tz_localize(tz=tz, nonexistent="NaT")
+        test_times = ["2015-03-08 01:00-05:00", "NaT", "2015-03-08 03:00-04:00"]
+        dti = to_datetime(test_times, utc=True)
+        expected = dti.tz_convert("US/Eastern")
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_tz_localize_ambiguous_infer(self, tz):
+        # November 6, 2011, fall back, repeat 2 AM hour
+        # With no repeated hours, we cannot infer the transition
+        dr = date_range(datetime(2011, 11, 6, 0), periods=5, freq=offsets.Hour())
+        with pytest.raises(ValueError, match="Cannot infer dst time"):
+            dr.tz_localize(tz)
+
+    def test_dti_tz_localize_ambiguous_infer2(self, tz, unit):
+        # With repeated hours, we can infer the transition
+        dr = date_range(
+            datetime(2011, 11, 6, 0), periods=5, freq=offsets.Hour(), tz=tz, unit=unit
+        )
+        times = [
+            "11/06/2011 00:00",
+            "11/06/2011 01:00",
+            "11/06/2011 01:00",
+            "11/06/2011 02:00",
+            "11/06/2011 03:00",
+        ]
+        di = DatetimeIndex(times).as_unit(unit)
+        result = di.tz_localize(tz, ambiguous="infer")
+        expected = dr._with_freq(None)
+        tm.assert_index_equal(result, expected)
+        result2 = DatetimeIndex(times, tz=tz, ambiguous="infer").as_unit(unit)
+        tm.assert_index_equal(result2, expected)
+
+    def test_dti_tz_localize_ambiguous_infer3(self, tz):
+        # When there is no dst transition, nothing special happens
+        dr = date_range(datetime(2011, 6, 1, 0), periods=10, freq=offsets.Hour())
+        localized = dr.tz_localize(tz)
+        localized_infer = dr.tz_localize(tz, ambiguous="infer")
+        tm.assert_index_equal(localized, localized_infer)
+
+    def test_dti_tz_localize_ambiguous_times(self, tz):
+        # March 13, 2011, spring forward, skip from 2 AM to 3 AM
+        dr = date_range(datetime(2011, 3, 13, 1, 30), periods=3, freq=offsets.Hour())
+        with pytest.raises(ValueError, match="2011-03-13 02:30:00"):
+            dr.tz_localize(tz)
+
+        # after dst transition, it works
+        dr = date_range(
+            datetime(2011, 3, 13, 3, 30), periods=3, freq=offsets.Hour(), tz=tz
+        )
+
+        # November 6, 2011, fall back, repeat 2 AM hour
+        dr = date_range(datetime(2011, 11, 6, 1, 30), periods=3, freq=offsets.Hour())
+        with pytest.raises(ValueError, match="Cannot infer dst time"):
+            dr.tz_localize(tz)
+
+        # UTC is OK
+        dr = date_range(
+            datetime(2011, 3, 13), periods=48, freq=offsets.Minute(30), tz=timezone.utc
+        )
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_tz_localize_pass_dates_to_utc(self, tzstr):
+        strdates = ["1/1/2012", "3/1/2012", "4/1/2012"]
+
+        idx = DatetimeIndex(strdates)
+        conv = idx.tz_localize(tzstr)
+
+        fromdates = DatetimeIndex(strdates, tz=tzstr)
+
+        assert conv.tz == fromdates.tz
+        tm.assert_numpy_array_equal(conv.values, fromdates.values)
+
+    @pytest.mark.parametrize("prefix", ["", "dateutil/"])
+    def test_dti_tz_localize(self, prefix):
+        tzstr = prefix + "US/Eastern"
+        dti = date_range(start="1/1/2005", end="1/1/2005 0:00:30.256", freq="ms")
+        dti2 = dti.tz_localize(tzstr)
+
+        dti_utc = date_range(
+            start="1/1/2005 05:00", end="1/1/2005 5:00:30.256", freq="ms", tz="utc"
+        )
+
+        tm.assert_numpy_array_equal(dti2.values, dti_utc.values)
+
+        dti3 = dti2.tz_convert(prefix + "US/Pacific")
+        tm.assert_numpy_array_equal(dti3.values, dti_utc.values)
+
+        dti = date_range(start="11/6/2011 1:59", end="11/6/2011 2:00", freq="ms")
+        with pytest.raises(ValueError, match="Cannot infer dst time"):
+            dti.tz_localize(tzstr)
+
+        dti = date_range(start="3/13/2011 1:59", end="3/13/2011 2:00", freq="ms")
+        with pytest.raises(ValueError, match="2011-03-13 02:00:00"):
+            dti.tz_localize(tzstr)
+
+    def test_dti_tz_localize_utc_conversion(self, tz):
+        # Localizing to time zone should:
+        #  1) check for DST ambiguities
+        #  2) convert to UTC
+
+        rng = date_range("3/10/2012", "3/11/2012", freq="30min")
+
+        converted = rng.tz_localize(tz)
+        expected_naive = rng + offsets.Hour(5)
+        tm.assert_numpy_array_equal(converted.asi8, expected_naive.asi8)
+
+        # DST ambiguity, this should fail
+        rng = date_range("3/11/2012", "3/12/2012", freq="30min")
+        # Is this really how it should fail??
+        with pytest.raises(ValueError, match="2012-03-11 02:00:00"):
+            rng.tz_localize(tz)
+
+    def test_dti_tz_localize_roundtrip(self, tz_aware_fixture):
+        # note: this tz tests that a tz-naive index can be localized
+        # and de-localized successfully, when there are no DST transitions
+        # in the range.
+        idx = date_range(start="2014-06-01", end="2014-08-30", freq="15min")
+        tz = tz_aware_fixture
+        localized = idx.tz_localize(tz)
+        # can't localize a tz-aware object
+        with pytest.raises(
+            TypeError, match="Already tz-aware, use tz_convert to convert"
+        ):
+            localized.tz_localize(tz)
+        reset = localized.tz_localize(None)
+        assert reset.tzinfo is None
+        expected = idx._with_freq(None)
+        tm.assert_index_equal(reset, expected)
+
+    def test_dti_tz_localize_naive(self):
+        rng = date_range("1/1/2011", periods=100, freq="h")
+
+        conv = rng.tz_localize("US/Pacific")
+        exp = date_range("1/1/2011", periods=100, freq="h", tz="US/Pacific")
+
+        tm.assert_index_equal(conv, exp._with_freq(None))
+
+    def test_dti_tz_localize_tzlocal(self):
+        # GH#13583
+        offset = dateutil.tz.tzlocal().utcoffset(datetime(2011, 1, 1))
+        offset = int(offset.total_seconds() * 1000000000)
+
+        dti = date_range(start="2001-01-01", end="2001-03-01")
+        dti2 = dti.tz_localize(dateutil.tz.tzlocal())
+        tm.assert_numpy_array_equal(dti2.asi8 + offset, dti.asi8)
+
+        dti = date_range(start="2001-01-01", end="2001-03-01", tz=dateutil.tz.tzlocal())
+        dti2 = dti.tz_localize(None)
+        tm.assert_numpy_array_equal(dti2.asi8 - offset, dti.asi8)
+
+    def test_dti_tz_localize_ambiguous_nat(self, tz):
+        times = [
+            "11/06/2011 00:00",
+            "11/06/2011 01:00",
+            "11/06/2011 01:00",
+            "11/06/2011 02:00",
+            "11/06/2011 03:00",
+        ]
+        di = DatetimeIndex(times)
+        localized = di.tz_localize(tz, ambiguous="NaT")
+
+        times = [
+            "11/06/2011 00:00",
+            np.nan,
+            np.nan,
+            "11/06/2011 02:00",
+            "11/06/2011 03:00",
+        ]
+        di_test = DatetimeIndex(times, tz="US/Eastern")
+
+        # left dtype is datetime64[ns, US/Eastern]
+        # right is datetime64[ns, tzfile('/usr/share/zoneinfo/US/Eastern')]
+        tm.assert_numpy_array_equal(di_test.values, localized.values)
+
+    def test_dti_tz_localize_ambiguous_flags(self, tz, unit):
+        # November 6, 2011, fall back, repeat 2 AM hour
+
+        # Pass in flags to determine right dst transition
+        dr = date_range(
+            datetime(2011, 11, 6, 0), periods=5, freq=offsets.Hour(), tz=tz, unit=unit
+        )
+        times = [
+            "11/06/2011 00:00",
+            "11/06/2011 01:00",
+            "11/06/2011 01:00",
+            "11/06/2011 02:00",
+            "11/06/2011 03:00",
+        ]
+
+        # Test tz_localize
+        di = DatetimeIndex(times).as_unit(unit)
+        is_dst = [1, 1, 0, 0, 0]
+        localized = di.tz_localize(tz, ambiguous=is_dst)
+        expected = dr._with_freq(None)
+        tm.assert_index_equal(expected, localized)
+
+        result = DatetimeIndex(times, tz=tz, ambiguous=is_dst).as_unit(unit)
+        tm.assert_index_equal(result, expected)
+
+        localized = di.tz_localize(tz, ambiguous=np.array(is_dst))
+        tm.assert_index_equal(dr, localized)
+
+        localized = di.tz_localize(tz, ambiguous=np.array(is_dst).astype("bool"))
+        tm.assert_index_equal(dr, localized)
+
+        # Test constructor
+        localized = DatetimeIndex(times, tz=tz, ambiguous=is_dst).as_unit(unit)
+        tm.assert_index_equal(dr, localized)
+
+        # Test duplicate times where inferring the dst fails
+        times += times
+        di = DatetimeIndex(times).as_unit(unit)
+
+        # When the sizes are incompatible, make sure error is raised
+        msg = "Length of ambiguous bool-array must be the same size as vals"
+        with pytest.raises(Exception, match=msg):
+            di.tz_localize(tz, ambiguous=is_dst)
+
+        # When sizes are compatible and there are repeats ('infer' won't work)
+        is_dst = np.hstack((is_dst, is_dst))
+        localized = di.tz_localize(tz, ambiguous=is_dst)
+        dr = dr.append(dr)
+        tm.assert_index_equal(dr, localized)
+
+    def test_dti_tz_localize_ambiguous_flags2(self, tz):
+        # When there is no dst transition, nothing special happens
+        dr = date_range(datetime(2011, 6, 1, 0), periods=10, freq=offsets.Hour())
+        is_dst = np.array([1] * 10)
+        localized = dr.tz_localize(tz)
+        localized_is_dst = dr.tz_localize(tz, ambiguous=is_dst)
+        tm.assert_index_equal(localized, localized_is_dst)
+
+    def test_dti_tz_localize_bdate_range(self):
+        dr = bdate_range("1/1/2009", "1/1/2010")
+        dr_utc = bdate_range("1/1/2009", "1/1/2010", tz=timezone.utc)
+        localized = dr.tz_localize(timezone.utc)
+        tm.assert_index_equal(dr_utc, localized)
+
+    @pytest.mark.parametrize(
+        "start_ts, tz, end_ts, shift",
+        [
+            ["2015-03-29 02:20:00", "Europe/Warsaw", "2015-03-29 03:00:00", "forward"],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 01:59:59.999999999",
+                "backward",
+            ],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 03:20:00",
+                timedelta(hours=1),
+            ],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 01:20:00",
+                timedelta(hours=-1),
+            ],
+            ["2018-03-11 02:33:00", "US/Pacific", "2018-03-11 03:00:00", "forward"],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 01:59:59.999999999",
+                "backward",
+            ],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 03:33:00",
+                timedelta(hours=1),
+            ],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 01:33:00",
+                timedelta(hours=-1),
+            ],
+        ],
+    )
+    @pytest.mark.parametrize("tz_type", ["", "dateutil/"])
+    def test_dti_tz_localize_nonexistent_shift(
+        self, start_ts, tz, end_ts, shift, tz_type, unit
+    ):
+        # GH#8917
+        tz = tz_type + tz
+        if isinstance(shift, str):
+            shift = "shift_" + shift
+        dti = DatetimeIndex([Timestamp(start_ts)]).as_unit(unit)
+        result = dti.tz_localize(tz, nonexistent=shift)
+        expected = DatetimeIndex([Timestamp(end_ts)]).tz_localize(tz).as_unit(unit)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("offset", [-1, 1])
+    def test_dti_tz_localize_nonexistent_shift_invalid(self, offset, warsaw):
+        # GH#8917
+        tz = warsaw
+        dti = DatetimeIndex([Timestamp("2015-03-29 02:20:00")])
+        msg = "The provided timedelta will relocalize on a nonexistent time"
+        with pytest.raises(ValueError, match=msg):
+            dti.tz_localize(tz, nonexistent=timedelta(seconds=offset))
diff --git a/pandas/pandas/tests/indexes/datetimes/methods/test_unique.py b/pandas/pandas/tests/indexes/datetimes/methods/test_unique.py
new file mode 100644
index 0000000000000..3c419b23c749a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/methods/test_unique.py
@@ -0,0 +1,77 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+def test_unique(tz_naive_fixture):
+    idx = DatetimeIndex(["2017"] * 2, tz=tz_naive_fixture)
+    expected = idx[:1]
+
+    result = idx.unique()
+    tm.assert_index_equal(result, expected)
+    # GH#21737
+    # Ensure the underlying data is consistent
+    assert result[0] == expected[0]
+
+
+def test_index_unique(rand_series_with_duplicate_datetimeindex):
+    dups = rand_series_with_duplicate_datetimeindex
+    index = dups.index
+
+    uniques = index.unique()
+    expected = DatetimeIndex(
+        [
+            datetime(2000, 1, 2),
+            datetime(2000, 1, 3),
+            datetime(2000, 1, 4),
+            datetime(2000, 1, 5),
+        ],
+        dtype=index.dtype,
+    )
+    assert uniques.dtype == index.dtype  # sanity
+    tm.assert_index_equal(uniques, expected)
+    assert index.nunique() == 4
+
+    # GH#2563
+    assert isinstance(uniques, DatetimeIndex)
+
+    dups_local = index.tz_localize("US/Eastern")
+    dups_local.name = "foo"
+    result = dups_local.unique()
+    expected = DatetimeIndex(expected, name="foo")
+    expected = expected.tz_localize("US/Eastern")
+    assert result.tz is not None
+    assert result.name == "foo"
+    tm.assert_index_equal(result, expected)
+
+
+def test_index_unique2():
+    # NaT, note this is excluded
+    arr = [1370745748 + t for t in range(20)] + [NaT._value]
+    idx = DatetimeIndex(arr * 3)
+    tm.assert_index_equal(idx.unique(), DatetimeIndex(arr))
+    assert idx.nunique() == 20
+    assert idx.nunique(dropna=False) == 21
+
+
+def test_index_unique3():
+    arr = [
+        Timestamp("2013-06-09 02:42:28") + timedelta(seconds=t) for t in range(20)
+    ] + [NaT]
+    idx = DatetimeIndex(arr * 3)
+    tm.assert_index_equal(idx.unique(), DatetimeIndex(arr))
+    assert idx.nunique() == 20
+    assert idx.nunique(dropna=False) == 21
+
+
+def test_is_unique_monotonic(rand_series_with_duplicate_datetimeindex):
+    index = rand_series_with_duplicate_datetimeindex.index
+    assert not index.is_unique
diff --git a/pandas/pandas/tests/indexes/datetimes/test_arithmetic.py b/pandas/pandas/tests/indexes/datetimes/test_arithmetic.py
new file mode 100644
index 0000000000000..bac849301d1f7
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_arithmetic.py
@@ -0,0 +1,56 @@
+# Arithmetic tests specific to DatetimeIndex are generally about `freq`
+#  retention or inference.  Other arithmetic tests belong in
+#  tests/arithmetic/test_datetime64.py
+import pytest
+
+from pandas import (
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexArithmetic:
+    def test_add_timedelta_preserves_freq(self):
+        # GH#37295 should hold for any DTI with freq=None or Tick freq
+        tz = "Canada/Eastern"
+        dti = date_range(
+            start=Timestamp("2019-03-26 00:00:00-0400", tz=tz),
+            end=Timestamp("2020-10-17 00:00:00-0400", tz=tz),
+            freq="D",
+        )
+        result = dti + Timedelta(days=1)
+        assert result.freq == dti.freq
+
+    def test_sub_datetime_preserves_freq(self, tz_naive_fixture):
+        # GH#48818
+        dti = date_range("2016-01-01", periods=12, tz=tz_naive_fixture)
+
+        res = dti - dti[0]
+        expected = timedelta_range("0 Days", "11 Days")
+        tm.assert_index_equal(res, expected)
+        assert res.freq == expected.freq
+
+    @pytest.mark.xfail(
+        reason="The inherited freq is incorrect bc dti.freq is incorrect "
+        "https://github.com/pandas-dev/pandas/pull/48818/files#r982793461"
+    )
+    def test_sub_datetime_preserves_freq_across_dst(self):
+        # GH#48818
+        ts = Timestamp("2016-03-11", tz="US/Pacific")
+        dti = date_range(ts, periods=4)
+
+        res = dti - dti[0]
+        expected = TimedeltaIndex(
+            [
+                Timedelta(days=0),
+                Timedelta(days=1),
+                Timedelta(days=2),
+                Timedelta(days=2, hours=23),
+            ]
+        )
+        tm.assert_index_equal(res, expected)
+        assert res.freq == expected.freq
diff --git a/pandas/pandas/tests/indexes/datetimes/test_constructors.py b/pandas/pandas/tests/indexes/datetimes/test_constructors.py
new file mode 100644
index 0000000000000..c418b2a18008b
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_constructors.py
@@ -0,0 +1,1206 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+from functools import partial
+from operator import attrgetter
+import zoneinfo
+
+import dateutil
+import dateutil.tz
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    astype_overflowsafe,
+    timezones,
+)
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    Timestamp,
+    date_range,
+    offsets,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.arrays import period_array
+
+
+class TestDatetimeIndex:
+    def test_from_dt64_unsupported_unit(self):
+        # GH#49292
+        val = np.datetime64(1, "D")
+        result = DatetimeIndex([val], tz="US/Pacific")
+
+        expected = DatetimeIndex([val.astype("M8[s]")], tz="US/Pacific")
+        tm.assert_index_equal(result, expected)
+
+    def test_explicit_tz_none(self):
+        # GH#48659
+        dti = date_range("2016-01-01", periods=10, tz="UTC")
+
+        msg = "Passed data is timezone-aware, incompatible with 'tz=None'"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(dti, tz=None)
+
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(np.array(dti), tz=None)
+
+        msg = "Cannot pass both a timezone-aware dtype and tz=None"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex([], dtype="M8[ns, UTC]", tz=None)
+
+    def test_freq_validation_with_nat(self):
+        # GH#11587 make sure we get a useful error message when generate_range
+        #  raises
+        msg = (
+            "Inferred frequency None from passed values does not conform "
+            "to passed frequency D"
+        )
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex([pd.NaT, Timestamp("2011-01-01")], freq="D")
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex([pd.NaT, Timestamp("2011-01-01")._value], freq="D")
+
+    # TODO: better place for tests shared by DTI/TDI?
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("2016-01-01", periods=5, tz="US/Pacific"),
+            pd.timedelta_range("1 Day", periods=5),
+        ],
+    )
+    def test_shallow_copy_inherits_array_freq(self, index):
+        # If we pass a DTA/TDA to shallow_copy and dont specify a freq,
+        #  we should inherit the array's freq, not our own.
+        array = index._data
+
+        arr = array[[0, 3, 2, 4, 1]]
+        assert arr.freq is None
+
+        result = index._shallow_copy(arr)
+        assert result.freq is None
+
+    def test_categorical_preserves_tz(self):
+        # GH#18664 retain tz when going DTI-->Categorical-->DTI
+        dti = DatetimeIndex(
+            [pd.NaT, "2015-01-01", "1999-04-06 15:14:13", "2015-01-01"], tz="US/Eastern"
+        )
+
+        for dtobj in [dti, dti._data]:
+            # works for DatetimeIndex or DatetimeArray
+
+            ci = pd.CategoricalIndex(dtobj)
+            carr = pd.Categorical(dtobj)
+            cser = pd.Series(ci)
+
+            for obj in [ci, carr, cser]:
+                result = DatetimeIndex(obj)
+                tm.assert_index_equal(result, dti)
+
+    def test_dti_with_period_data_raises(self):
+        # GH#23675
+        data = pd.PeriodIndex(["2016Q1", "2016Q2"], freq="Q")
+
+        with pytest.raises(TypeError, match="PeriodDtype data is invalid"):
+            DatetimeIndex(data)
+
+        with pytest.raises(TypeError, match="PeriodDtype data is invalid"):
+            to_datetime(data)
+
+        with pytest.raises(TypeError, match="PeriodDtype data is invalid"):
+            DatetimeIndex(period_array(data))
+
+        with pytest.raises(TypeError, match="PeriodDtype data is invalid"):
+            to_datetime(period_array(data))
+
+    def test_dti_with_timedelta64_data_raises(self):
+        # GH#23675 deprecated, enforced in GH#29794
+        data = np.array([0], dtype="m8[ns]")
+        msg = r"timedelta64\[ns\] cannot be converted to datetime64"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeIndex(data)
+
+        with pytest.raises(TypeError, match=msg):
+            to_datetime(data)
+
+        with pytest.raises(TypeError, match=msg):
+            DatetimeIndex(pd.TimedeltaIndex(data))
+
+        with pytest.raises(TypeError, match=msg):
+            to_datetime(pd.TimedeltaIndex(data))
+
+    def test_constructor_from_sparse_array(self):
+        # https://github.com/pandas-dev/pandas/issues/35843
+        values = [
+            Timestamp("2012-05-01T01:00:00.000000"),
+            Timestamp("2016-05-01T01:00:00.000000"),
+        ]
+        arr = pd.arrays.SparseArray(values)
+        result = Index(arr)
+        assert type(result) is Index
+        assert result.dtype == arr.dtype
+
+    def test_construction_caching(self):
+        df = pd.DataFrame(
+            {
+                "dt": date_range("20130101", periods=3),
+                "dttz": date_range(
+                    "20130101", periods=3, tz=zoneinfo.ZoneInfo("US/Eastern")
+                ),
+                "dt_with_null": [
+                    Timestamp("20130101"),
+                    pd.NaT,
+                    Timestamp("20130103"),
+                ],
+                "dtns": date_range("20130101", periods=3, freq="ns"),
+            }
+        )
+        assert df.dttz.dtype.tz.key == "US/Eastern"
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [{"tz": "dtype.tz"}, {"dtype": "dtype"}, {"dtype": "dtype", "tz": "dtype.tz"}],
+    )
+    def test_construction_with_alt(self, kwargs, tz_aware_fixture):
+        tz = tz_aware_fixture
+        i = date_range("20130101", periods=5, freq="h", tz=tz)
+        kwargs = {key: attrgetter(val)(i) for key, val in kwargs.items()}
+        result = DatetimeIndex(i, **kwargs)
+        tm.assert_index_equal(i, result)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [{"tz": "dtype.tz"}, {"dtype": "dtype"}, {"dtype": "dtype", "tz": "dtype.tz"}],
+    )
+    def test_construction_with_alt_tz_localize(self, kwargs, tz_aware_fixture):
+        tz = tz_aware_fixture
+        i = date_range("20130101", periods=5, freq="h", tz=tz)
+        i = i._with_freq(None)
+        kwargs = {key: attrgetter(val)(i) for key, val in kwargs.items()}
+
+        if "tz" in kwargs:
+            result = DatetimeIndex(i.asi8, tz="UTC").tz_convert(kwargs["tz"])
+
+            expected = DatetimeIndex(i, **kwargs)
+            tm.assert_index_equal(result, expected)
+
+        # localize into the provided tz
+        i2 = DatetimeIndex(i.tz_localize(None).asi8, tz="UTC")
+        expected = i.tz_localize(None).tz_localize("UTC")
+        tm.assert_index_equal(i2, expected)
+
+        # incompat tz/dtype
+        msg = "cannot supply both a tz and a dtype with a tz"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(
+                i.tz_localize(None).asi8,
+                dtype=i.dtype,
+                tz=zoneinfo.ZoneInfo("US/Hawaii"),
+            )
+
+    def test_construction_index_with_mixed_timezones(self):
+        # gh-11488: no tz results in DatetimeIndex
+        result = Index([Timestamp("2011-01-01"), Timestamp("2011-01-02")], name="idx")
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01"), Timestamp("2011-01-02")], name="idx"
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is None
+
+        # same tz results in DatetimeIndex
+        result = Index(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="Asia/Tokyo"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00"), Timestamp("2011-01-02 10:00")],
+            tz="Asia/Tokyo",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is not None
+        assert result.tz == exp.tz
+
+        # same tz results in DatetimeIndex (DST)
+        result = Index(
+            [
+                Timestamp("2011-01-01 10:00", tz="US/Eastern"),
+                Timestamp("2011-08-01 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00"), Timestamp("2011-08-01 10:00")],
+            tz="US/Eastern",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is not None
+        assert result.tz == exp.tz
+
+        # Different tz results in Index(dtype=object)
+        result = Index(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = Index(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            dtype="object",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert not isinstance(result, DatetimeIndex)
+
+        result = Index(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = Index(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            dtype="object",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert not isinstance(result, DatetimeIndex)
+
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(["2013-11-02 22:00-05:00", "2013-11-03 22:00-06:00"])
+
+        # length = 1
+        result = Index([Timestamp("2011-01-01")], name="idx")
+        exp = DatetimeIndex([Timestamp("2011-01-01")], name="idx")
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is None
+
+        # length = 1 with tz
+        result = Index([Timestamp("2011-01-01 10:00", tz="Asia/Tokyo")], name="idx")
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00")], tz="Asia/Tokyo", name="idx"
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is not None
+        assert result.tz == exp.tz
+
+    def test_construction_index_with_mixed_timezones_with_NaT(self):
+        # see gh-11488
+        result = Index(
+            [pd.NaT, Timestamp("2011-01-01"), pd.NaT, Timestamp("2011-01-02")],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [pd.NaT, Timestamp("2011-01-01"), pd.NaT, Timestamp("2011-01-02")],
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is None
+
+        # Same tz results in DatetimeIndex
+        result = Index(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00", tz="Asia/Tokyo"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00"),
+            ],
+            tz="Asia/Tokyo",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is not None
+        assert result.tz == exp.tz
+
+        # same tz results in DatetimeIndex (DST)
+        result = Index(
+            [
+                Timestamp("2011-01-01 10:00", tz="US/Eastern"),
+                pd.NaT,
+                Timestamp("2011-08-01 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00"), pd.NaT, Timestamp("2011-08-01 10:00")],
+            tz="US/Eastern",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is not None
+        assert result.tz == exp.tz
+
+        # different tz results in Index(dtype=object)
+        result = Index(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = Index(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            dtype="object",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert not isinstance(result, DatetimeIndex)
+
+        result = Index(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = Index(
+            [
+                pd.NaT,
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                pd.NaT,
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            dtype="object",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert not isinstance(result, DatetimeIndex)
+
+        # all NaT
+        result = Index([pd.NaT, pd.NaT], name="idx")
+        exp = DatetimeIndex([pd.NaT, pd.NaT], name="idx")
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is None
+
+    def test_construction_dti_with_mixed_timezones(self):
+        # GH 11488 (not changed, added explicit tests)
+
+        # no tz results in DatetimeIndex
+        result = DatetimeIndex(
+            [Timestamp("2011-01-01"), Timestamp("2011-01-02")], name="idx"
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01"), Timestamp("2011-01-02")], name="idx"
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+
+        # same tz results in DatetimeIndex
+        result = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="Asia/Tokyo"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00"), Timestamp("2011-01-02 10:00")],
+            tz="Asia/Tokyo",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+
+        # same tz results in DatetimeIndex (DST)
+        result = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="US/Eastern"),
+                Timestamp("2011-08-01 10:00", tz="US/Eastern"),
+            ],
+            name="idx",
+        )
+        exp = DatetimeIndex(
+            [Timestamp("2011-01-01 10:00"), Timestamp("2011-08-01 10:00")],
+            tz="US/Eastern",
+            name="idx",
+        )
+        tm.assert_index_equal(result, exp, exact=True)
+        assert isinstance(result, DatetimeIndex)
+
+        # tz mismatch affecting to tz-aware raises TypeError/ValueError
+
+        msg = "cannot be converted to datetime64"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(
+                [
+                    Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                    Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+                ],
+                name="idx",
+            )
+
+        # pre-2.0 this raised bc of awareness mismatch. in 2.0 with a tz#
+        #  specified we behave as if this was called pointwise, so
+        #  the naive Timestamp is treated as a wall time.
+        dti = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            tz="Asia/Tokyo",
+            name="idx",
+        )
+        expected = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern").tz_convert("Asia/Tokyo"),
+            ],
+            tz="Asia/Tokyo",
+            name="idx",
+        )
+        tm.assert_index_equal(dti, expected)
+
+        # pre-2.0 mixed-tz scalars raised even if a tz/dtype was specified.
+        #  as of 2.0 we successfully return the requested tz/dtype
+        dti = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            tz="US/Eastern",
+            name="idx",
+        )
+        expected = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo").tz_convert("US/Eastern"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            tz="US/Eastern",
+            name="idx",
+        )
+        tm.assert_index_equal(dti, expected)
+
+        # same thing but pass dtype instead of tz
+        dti = DatetimeIndex(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-02 10:00", tz="US/Eastern"),
+            ],
+            dtype="M8[s, US/Eastern]",
+            name="idx",
+        )
+        tm.assert_index_equal(dti, expected)
+
+    def test_construction_base_constructor(self):
+        arr = [Timestamp("2011-01-01"), pd.NaT, Timestamp("2011-01-03")]
+        tm.assert_index_equal(Index(arr), DatetimeIndex(arr))
+        tm.assert_index_equal(Index(np.array(arr)), DatetimeIndex(np.array(arr)))
+
+        arr = [np.nan, pd.NaT, Timestamp("2011-01-03")]
+        tm.assert_index_equal(Index(arr), DatetimeIndex(arr))
+        tm.assert_index_equal(Index(np.array(arr)), DatetimeIndex(np.array(arr)))
+
+    def test_construction_outofbounds(self):
+        # GH 13663
+        dates = [
+            datetime(3000, 1, 1),
+            datetime(4000, 1, 1),
+            datetime(5000, 1, 1),
+            datetime(6000, 1, 1),
+        ]
+        exp = Index(dates, dtype="M8[us]")
+        res = Index(dates)
+        tm.assert_index_equal(res, exp)
+
+        DatetimeIndex(dates)
+
+    @pytest.mark.parametrize("data", [["1400-01-01"], [datetime(1400, 1, 1)]])
+    def test_dti_date_out_of_range(self, data):
+        # GH#1475
+        DatetimeIndex(data)
+
+    def test_construction_with_ndarray(self):
+        # GH 5152
+        dates = [datetime(2013, 10, 7), datetime(2013, 10, 8), datetime(2013, 10, 9)]
+        data = DatetimeIndex(dates, freq=offsets.BDay()).values
+        result = DatetimeIndex(data, freq=offsets.BDay())
+        expected = DatetimeIndex(
+            ["2013-10-07", "2013-10-08", "2013-10-09"], dtype="M8[us]", freq="B"
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_integer_values_and_tz_interpreted_as_utc(self):
+        # GH-24559
+        val = np.datetime64("2000-01-01 00:00:00", "ns")
+        values = np.array([val.view("i8")])
+
+        result = DatetimeIndex(values).tz_localize("US/Central")
+
+        expected = DatetimeIndex(["2000-01-01T00:00:00"], dtype="M8[ns, US/Central]")
+        tm.assert_index_equal(result, expected)
+
+        # but UTC is *not* deprecated.
+        with tm.assert_produces_warning(None):
+            result = DatetimeIndex(values, tz="UTC")
+        expected = DatetimeIndex(["2000-01-01T00:00:00"], dtype="M8[ns, UTC]")
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_coverage(self):
+        msg = r"DatetimeIndex\(\.\.\.\) must be called with a collection"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeIndex("1/1/2000")
+
+        # generator expression
+        gen = (datetime(2000, 1, 1) + timedelta(i) for i in range(10))
+        result = DatetimeIndex(gen)
+        expected = DatetimeIndex(
+            [datetime(2000, 1, 1) + timedelta(i) for i in range(10)]
+        )
+        tm.assert_index_equal(result, expected)
+
+        # NumPy string array
+        strings = np.array(["2000-01-01", "2000-01-02", "2000-01-03"])
+        result = DatetimeIndex(strings)
+        expected = DatetimeIndex(strings.astype("O"))
+        tm.assert_index_equal(result, expected)
+
+        from_ints = DatetimeIndex(expected.as_unit("ns").asi8).as_unit("s")
+        tm.assert_index_equal(from_ints, expected)
+
+        # string with NaT
+        strings = np.array(["2000-01-01", "2000-01-02", "NaT"])
+        result = DatetimeIndex(strings)
+        expected = DatetimeIndex(strings.astype("O"))
+        tm.assert_index_equal(result, expected)
+
+        from_ints = DatetimeIndex(expected.as_unit("ns").asi8).as_unit("s")
+        tm.assert_index_equal(from_ints, expected)
+
+        # non-conforming
+        msg = (
+            "Inferred frequency None from passed values does not conform "
+            "to passed frequency D"
+        )
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(["2000-01-01", "2000-01-02", "2000-01-04"], freq="D")
+
+    @pytest.mark.parametrize("freq", ["YS", "W-SUN"])
+    def test_constructor_datetime64_tzformat(self, freq):
+        # see GH#6572: ISO 8601 format results in stdlib timezone object
+        idx = date_range(
+            "2013-01-01T00:00:00-05:00", "2016-01-01T23:59:59-05:00", freq=freq
+        )
+        expected = date_range(
+            "2013-01-01T00:00:00",
+            "2016-01-01T23:59:59",
+            freq=freq,
+            tz=timezone(timedelta(minutes=-300)),
+        )
+        tm.assert_index_equal(idx, expected)
+        # Unable to use `US/Eastern` because of DST
+        expected_i8 = date_range(
+            "2013-01-01T00:00:00", "2016-01-01T23:59:59", freq=freq, tz="America/Lima"
+        )
+        tm.assert_numpy_array_equal(idx.asi8, expected_i8.asi8)
+
+        idx = date_range(
+            "2013-01-01T00:00:00+09:00", "2016-01-01T23:59:59+09:00", freq=freq
+        )
+        expected = date_range(
+            "2013-01-01T00:00:00",
+            "2016-01-01T23:59:59",
+            freq=freq,
+            tz=timezone(timedelta(minutes=540)),
+        )
+        tm.assert_index_equal(idx, expected)
+        expected_i8 = date_range(
+            "2013-01-01T00:00:00", "2016-01-01T23:59:59", freq=freq, tz="Asia/Tokyo"
+        )
+        tm.assert_numpy_array_equal(idx.asi8, expected_i8.asi8)
+
+        # Non ISO 8601 format results in dateutil.tz.tzoffset
+        idx = date_range("2013/1/1 0:00:00-5:00", "2016/1/1 23:59:59-5:00", freq=freq)
+        expected = date_range(
+            "2013-01-01T00:00:00",
+            "2016-01-01T23:59:59",
+            freq=freq,
+            tz=timezone(timedelta(minutes=-300)),
+        )
+        tm.assert_index_equal(idx, expected)
+        # Unable to use `US/Eastern` because of DST
+        expected_i8 = date_range(
+            "2013-01-01T00:00:00", "2016-01-01T23:59:59", freq=freq, tz="America/Lima"
+        )
+        tm.assert_numpy_array_equal(idx.asi8, expected_i8.asi8)
+
+        idx = date_range("2013/1/1 0:00:00+9:00", "2016/1/1 23:59:59+09:00", freq=freq)
+        expected = date_range(
+            "2013-01-01T00:00:00",
+            "2016-01-01T23:59:59",
+            freq=freq,
+            tz=timezone(timedelta(minutes=540)),
+        )
+        tm.assert_index_equal(idx, expected)
+        expected_i8 = date_range(
+            "2013-01-01T00:00:00", "2016-01-01T23:59:59", freq=freq, tz="Asia/Tokyo"
+        )
+        tm.assert_numpy_array_equal(idx.asi8, expected_i8.asi8)
+
+    def test_constructor_dtype(self):
+        # passing a dtype with a tz should localize
+        idx = DatetimeIndex(
+            ["2013-01-01", "2013-01-02"], dtype="datetime64[ns, US/Eastern]"
+        )
+        expected = (
+            DatetimeIndex(["2013-01-01", "2013-01-02"])
+            .as_unit("ns")
+            .tz_localize("US/Eastern")
+        )
+        tm.assert_index_equal(idx, expected)
+
+        idx = DatetimeIndex(["2013-01-01", "2013-01-02"], tz="US/Eastern").as_unit("ns")
+        tm.assert_index_equal(idx, expected)
+
+    def test_constructor_dtype_tz_mismatch_raises(self):
+        # if we already have a tz and its not the same, then raise
+        idx = DatetimeIndex(
+            ["2013-01-01", "2013-01-02"], dtype="datetime64[ns, US/Eastern]"
+        )
+
+        msg = (
+            "cannot supply both a tz and a timezone-naive dtype "
+            r"\(i\.e\. datetime64\[ns\]\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(idx, dtype="datetime64[ns]")
+
+        # this is effectively trying to convert tz's
+        msg = "data is already tz-aware US/Eastern, unable to set specified tz: CET"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeIndex(idx, dtype="datetime64[ns, CET]")
+        msg = "cannot supply both a tz and a dtype with a tz"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(idx, tz="CET", dtype="datetime64[ns, US/Eastern]")
+
+        result = DatetimeIndex(idx, dtype="datetime64[ns, US/Eastern]")
+        tm.assert_index_equal(idx, result)
+
+    @pytest.mark.parametrize("dtype", [object, np.int32, np.int64])
+    def test_constructor_invalid_dtype_raises(self, dtype):
+        # GH 23986
+        msg = "Unexpected value for 'dtype'"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex([1, 2], dtype=dtype)
+
+    def test_000constructor_resolution(self):
+        # 2252
+        t1 = Timestamp((1352934390 * 1000000000) + 1000000 + 1000 + 1)
+        idx = DatetimeIndex([t1])
+
+        assert idx.nanosecond[0] == t1.nanosecond
+
+    def test_disallow_setting_tz(self):
+        # GH 3746
+        dti = DatetimeIndex(["2010"], tz="UTC")
+        msg = "Cannot directly set timezone"
+        with pytest.raises(AttributeError, match=msg):
+            dti.tz = zoneinfo.ZoneInfo("US/Pacific")
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            None,
+            "America/Los_Angeles",
+            zoneinfo.ZoneInfo("America/Los_Angeles"),
+            Timestamp("2000", tz="America/Los_Angeles").tz,
+        ],
+    )
+    def test_constructor_start_end_with_tz(self, tz):
+        # GH 18595
+        start = Timestamp("2013-01-01 06:00:00", tz="America/Los_Angeles")
+        end = Timestamp("2013-01-02 06:00:00", tz="America/Los_Angeles")
+        result = date_range(freq="D", start=start, end=end, tz=tz)
+        expected = DatetimeIndex(
+            ["2013-01-01 06:00:00", "2013-01-02 06:00:00"],
+            dtype="M8[ns, America/Los_Angeles]",
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+        # Especially assert that the timezone is consistent for zoneinfo
+        assert zoneinfo.ZoneInfo("America/Los_Angeles") is result.tz
+
+    @pytest.mark.parametrize("tz", ["US/Pacific", "US/Eastern", "Asia/Tokyo"])
+    def test_constructor_with_non_normalized_pytz(self, tz):
+        # GH 18595
+        pytz = pytest.importorskip("pytz")
+        tz_in = pytz.timezone(tz)
+        non_norm_tz = Timestamp("2010", tz=tz_in).tz
+        result = DatetimeIndex(["2010"], tz=non_norm_tz)
+        assert pytz.timezone(tz) is result.tz
+
+    def test_constructor_timestamp_near_dst(self):
+        # GH 20854
+        ts = [
+            Timestamp("2016-10-30 03:00:00+0300", tz="Europe/Helsinki"),
+            Timestamp("2016-10-30 03:00:00+0200", tz="Europe/Helsinki"),
+        ]
+        result = DatetimeIndex(ts).as_unit("ns")
+        expected = DatetimeIndex(
+            [ts[0].to_pydatetime(), ts[1].to_pydatetime()]
+        ).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [Index, DatetimeIndex])
+    @pytest.mark.parametrize("box", [np.array, partial(np.array, dtype=object), list])
+    @pytest.mark.parametrize(
+        "tz, dtype",
+        [("US/Pacific", "datetime64[ns, US/Pacific]"), (None, "datetime64[ns]")],
+    )
+    def test_constructor_with_int_tz(self, klass, box, tz, dtype):
+        # GH 20997, 20964
+        ts = Timestamp("2018-01-01", tz=tz).as_unit("ns")
+        result = klass(box([ts._value]), dtype=dtype)
+        expected = klass([ts])
+        assert result == expected
+
+    def test_construction_int_rountrip(self, tz_naive_fixture):
+        # GH 12619, GH#24559
+        tz = tz_naive_fixture
+
+        result = 1293858000000000000
+        expected = DatetimeIndex([result], tz=tz).asi8[0]
+        assert result == expected
+
+    def test_construction_from_replaced_timestamps_with_dst(self):
+        # GH 18785
+        index = date_range(
+            Timestamp(2000, 12, 31),
+            Timestamp(2005, 12, 31),
+            freq="YE-DEC",
+            tz="Australia/Melbourne",
+        )
+        result = DatetimeIndex([x.replace(month=6, day=1) for x in index])
+        expected = DatetimeIndex(
+            [
+                "2000-06-01 00:00:00",
+                "2001-06-01 00:00:00",
+                "2002-06-01 00:00:00",
+                "2003-06-01 00:00:00",
+                "2004-06-01 00:00:00",
+                "2005-06-01 00:00:00",
+            ],
+            tz="Australia/Melbourne",
+        ).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_construction_with_tz_and_tz_aware_dti(self):
+        # GH 23579
+        dti = date_range("2016-01-01", periods=3, tz="US/Central")
+        msg = "data is already tz-aware US/Central, unable to set specified tz"
+        with pytest.raises(TypeError, match=msg):
+            DatetimeIndex(dti, tz="Asia/Tokyo")
+
+    def test_construction_with_nat_and_tzlocal(self):
+        tz = dateutil.tz.tzlocal()
+        result = DatetimeIndex(["2018", "NaT"], tz=tz).as_unit("ns")
+        expected = DatetimeIndex([Timestamp("2018", tz=tz), pd.NaT]).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_with_ambiguous_keyword_arg(self):
+        # GH 35297
+
+        expected = DatetimeIndex(
+            ["2020-11-01 01:00:00", "2020-11-02 01:00:00"],
+            dtype="datetime64[ns, America/New_York]",
+            freq="D",
+            ambiguous=False,
+        )
+
+        # ambiguous keyword in start
+        timezone = "America/New_York"
+        start = Timestamp(year=2020, month=11, day=1, hour=1).tz_localize(
+            timezone, ambiguous=False
+        )
+        result = date_range(start=start, periods=2, ambiguous=False)
+        tm.assert_index_equal(result, expected)
+
+        # ambiguous keyword in end
+        timezone = "America/New_York"
+        end = Timestamp(year=2020, month=11, day=2, hour=1).tz_localize(
+            timezone, ambiguous=False
+        )
+        result = date_range(end=end, periods=2, ambiguous=False)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_with_nonexistent_keyword_arg(self, warsaw):
+        # GH 35297
+        timezone = warsaw
+
+        # nonexistent keyword in start
+        start = Timestamp("2015-03-29 02:30:00").tz_localize(
+            timezone, nonexistent="shift_forward"
+        )
+        result = date_range(start=start, periods=2, freq="h")
+        expected = DatetimeIndex(
+            [
+                Timestamp("2015-03-29 03:00:00+02:00", tz=timezone),
+                Timestamp("2015-03-29 04:00:00+02:00", tz=timezone),
+            ]
+        ).as_unit("ns")
+
+        tm.assert_index_equal(result, expected)
+
+        # nonexistent keyword in end
+        end = start
+        result = date_range(end=end, periods=2, freq="h")
+        expected = DatetimeIndex(
+            [
+                Timestamp("2015-03-29 01:00:00+01:00", tz=timezone),
+                Timestamp("2015-03-29 03:00:00+02:00", tz=timezone),
+            ]
+        ).as_unit("ns")
+
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_no_precision_raises(self):
+        # GH-24753, GH-24739
+
+        msg = "with no precision is not allowed"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(["2000"], dtype="datetime64")
+
+        msg = "The 'datetime64' dtype has no unit. Please pass in"
+        with pytest.raises(ValueError, match=msg):
+            Index(["2000"], dtype="datetime64")
+
+    def test_constructor_wrong_precision_raises(self):
+        dti = DatetimeIndex(["2000"], dtype="datetime64[us]")
+        assert dti.dtype == "M8[us]"
+        assert dti[0] == Timestamp(2000, 1, 1)
+
+    def test_index_constructor_with_numpy_object_array_and_timestamp_tz_with_nan(self):
+        # GH 27011
+        result = Index(np.array([Timestamp("2019", tz="UTC"), np.nan], dtype=object))
+        expected = DatetimeIndex([Timestamp("2019", tz="UTC"), pd.NaT])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tz", [zoneinfo.ZoneInfo("US/Eastern"), gettz("US/Eastern")]
+    )
+    def test_dti_from_tzaware_datetime(self, tz):
+        d = [datetime(2012, 8, 19, tzinfo=tz)]
+
+        index = DatetimeIndex(d)
+        assert timezones.tz_compare(index.tz, tz)
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_tz_constructors(self, tzstr):
+        """Test different DatetimeIndex constructions with timezone
+        Follow-up of GH#4229
+        """
+        arr = ["11/10/2005 08:00:00", "11/10/2005 09:00:00"]
+
+        idx1 = to_datetime(arr).tz_localize(tzstr)
+        idx2 = date_range(
+            start="2005-11-10 08:00:00", freq="h", periods=2, tz=tzstr, unit="s"
+        )
+        idx2 = idx2._with_freq(None)  # the others all have freq=None
+        idx3 = DatetimeIndex(arr, tz=tzstr).as_unit("s")
+        idx4 = DatetimeIndex(np.array(arr), tz=tzstr).as_unit("s")
+
+        tm.assert_index_equal(idx1, idx2)
+        tm.assert_index_equal(idx1, idx3)
+        tm.assert_index_equal(idx1, idx4)
+
+    def test_dti_construction_idempotent(self, unit):
+        rng = date_range(
+            "03/12/2012 00:00", periods=10, freq="W-FRI", tz="US/Eastern", unit=unit
+        )
+        rng2 = DatetimeIndex(data=rng, tz="US/Eastern")
+        tm.assert_index_equal(rng, rng2)
+
+    @pytest.mark.parametrize("prefix", ["", "dateutil/"])
+    def test_dti_constructor_static_tzinfo(self, prefix):
+        # it works!
+        index = DatetimeIndex([datetime(2012, 1, 1)], tz=prefix + "EST")
+        index.hour
+        index[0]
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_convert_datetime_list(self, tzstr):
+        dr = date_range("2012-06-02", periods=10, tz=tzstr, name="foo")
+        dr2 = DatetimeIndex(list(dr), name="foo", freq="D")
+        tm.assert_index_equal(dr, dr2)
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "pytz/US/Eastern",
+            gettz("US/Eastern"),
+        ],
+    )
+    @pytest.mark.parametrize("use_str", [True, False])
+    @pytest.mark.parametrize("box_cls", [Timestamp, DatetimeIndex])
+    def test_dti_ambiguous_matches_timestamp(self, tz, use_str, box_cls, request):
+        # GH#47471 check that we get the same raising behavior in the DTI
+        # constructor and Timestamp constructor
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        dtstr = "2013-11-03 01:59:59.999999"
+        item = dtstr
+        if not use_str:
+            item = Timestamp(dtstr).to_pydatetime()
+        if box_cls is not Timestamp:
+            item = [item]
+
+        if not use_str and isinstance(tz, dateutil.tz.tzfile):
+            # FIXME: The Timestamp constructor here behaves differently than all
+            #  the other cases bc with dateutil/zoneinfo tzinfos we implicitly
+            #  get fold=0. Having this raise is not important, but having the
+            #  behavior be consistent across cases is.
+            mark = pytest.mark.xfail(reason="We implicitly get fold=0.")
+            request.applymarker(mark)
+
+        with pytest.raises(ValueError, match=dtstr):
+            box_cls(item, tz=tz)
+
+    @pytest.mark.parametrize("tz", [None, "UTC", "US/Pacific"])
+    def test_dti_constructor_with_non_nano_dtype(self, tz):
+        # GH#55756, GH#54620
+        ts = Timestamp("2999-01-01")
+        dtype = "M8[us]"
+        if tz is not None:
+            dtype = f"M8[us, {tz}]"
+        vals = [ts, "2999-01-02 03:04:05.678910", 2500]
+        result = DatetimeIndex(vals, dtype=dtype)
+        # The 2500 is interpreted as microseconds, consistent with what
+        #  we would get if we created DatetimeIndexes from vals[:2] and vals[2:]
+        #  and concated the results.
+        pointwise = [
+            vals[0].tz_localize(tz),
+            Timestamp(vals[1], tz=tz),
+            to_datetime(vals[2], unit="us", utc=True).tz_convert(tz),
+        ]
+        exp_vals = [x.as_unit("us").asm8 for x in pointwise]
+        exp_arr = np.array(exp_vals, dtype="M8[us]")
+        expected = DatetimeIndex(exp_arr, dtype="M8[us]")
+        if tz is not None:
+            expected = expected.tz_localize("UTC").tz_convert(tz)
+        tm.assert_index_equal(result, expected)
+
+        result2 = DatetimeIndex(np.array(vals, dtype=object), dtype=dtype)
+        tm.assert_index_equal(result2, expected)
+
+    def test_dti_constructor_with_non_nano_now_today(self, request):
+        # GH#55756
+        now = Timestamp.now()
+        today = Timestamp.today()
+        result = DatetimeIndex(["now", "today"], dtype="M8[s]")
+        assert result.dtype == "M8[s]"
+
+        diff0 = result[0] - now.as_unit("s")
+        diff1 = result[1] - today.as_unit("s")
+        assert diff1 >= pd.Timedelta(0), f"The difference is {diff0}"
+        assert diff0 >= pd.Timedelta(0), f"The difference is {diff0}"
+
+        # result may not exactly match [now, today] so we'll test it up to a tolerance.
+        #  (it *may* match exactly due to rounding)
+        # GH 57535
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="result may not exactly match [now, today]", strict=False
+            )
+        )
+        tolerance = pd.Timedelta(seconds=1)
+        assert diff0 < tolerance, f"The difference is {diff0}"
+        assert diff1 < tolerance, f"The difference is {diff0}"
+
+    def test_dti_constructor_object_float_matches_float_dtype(self):
+        # GH#55780
+        arr = np.array([0, np.nan], dtype=np.float64)
+        arr2 = arr.astype(object)
+
+        dti1 = DatetimeIndex(arr, tz="CET")
+        dti2 = DatetimeIndex(arr2, tz="CET")
+        tm.assert_index_equal(dti1, dti2)
+
+    @pytest.mark.parametrize("dtype", ["M8[us]", "M8[us, US/Pacific]"])
+    def test_dti_constructor_with_dtype_object_int_matches_int_dtype(self, dtype):
+        # Going through the object path should match the non-object path
+
+        vals1 = np.arange(5, dtype="i8") * 1000
+        vals1[0] = pd.NaT.value
+
+        vals2 = vals1.astype(np.float64)
+        vals2[0] = np.nan
+
+        vals3 = vals1.astype(object)
+        # change lib.infer_dtype(vals3) from "integer" so we go through
+        #  array_to_datetime in _sequence_to_dt64
+        vals3[0] = pd.NaT
+
+        vals4 = vals2.astype(object)
+
+        res1 = DatetimeIndex(vals1, dtype=dtype)
+        res2 = DatetimeIndex(vals2, dtype=dtype)
+        res3 = DatetimeIndex(vals3, dtype=dtype)
+        res4 = DatetimeIndex(vals4, dtype=dtype)
+
+        expected = DatetimeIndex(vals1.view("M8[us]"))
+        if res1.tz is not None:
+            expected = expected.tz_localize("UTC").tz_convert(res1.tz)
+        tm.assert_index_equal(res1, expected)
+        tm.assert_index_equal(res2, expected)
+        tm.assert_index_equal(res3, expected)
+        tm.assert_index_equal(res4, expected)
+
+
+class TestTimeSeries:
+    def test_dti_constructor_preserve_dti_freq(self):
+        rng = date_range("1/1/2000", "1/2/2000", freq="5min")
+
+        rng2 = DatetimeIndex(rng)
+        assert rng.freq == rng2.freq
+
+    def test_explicit_none_freq(self):
+        # Explicitly passing freq=None is respected
+        rng = date_range("1/1/2000", "1/2/2000", freq="5min")
+
+        result = DatetimeIndex(rng, freq=None)
+        assert result.freq is None
+
+        result = DatetimeIndex(rng._data, freq=None)
+        assert result.freq is None
+
+    def test_dti_constructor_small_int(self, any_int_numpy_dtype):
+        # see gh-13721
+        exp = DatetimeIndex(
+            [
+                "1970-01-01 00:00:00.00000000",
+                "1970-01-01 00:00:00.00000001",
+                "1970-01-01 00:00:00.00000002",
+            ]
+        )
+
+        arr = np.array([0, 10, 20], dtype=any_int_numpy_dtype)
+        tm.assert_index_equal(DatetimeIndex(arr), exp)
+
+    def test_ctor_str_intraday(self):
+        rng = DatetimeIndex(["1-1-2000 00:00:01"])
+        assert rng[0].second == 1
+
+    def test_index_cast_datetime64_other_units(self):
+        arr = np.arange(0, 100, 10, dtype=np.int64).view("M8[D]")
+        idx = Index(arr)
+
+        assert (idx.values == astype_overflowsafe(arr, dtype=np.dtype("M8[ns]"))).all()
+
+    def test_constructor_int64_nocopy(self):
+        # GH#1624
+        arr = np.arange(1000, dtype=np.int64)
+        index = DatetimeIndex(arr)
+
+        arr[50:100] = -1
+        assert (index.asi8[50:100] == -1).all()
+
+        arr = np.arange(1000, dtype=np.int64)
+        index = DatetimeIndex(arr, copy=True)
+
+        arr[50:100] = -1
+        assert (index.asi8[50:100] != -1).all()
+
+    @pytest.mark.parametrize(
+        "freq",
+        ["ME", "QE", "YE", "D", "B", "bh", "min", "s", "ms", "us", "h", "ns", "C"],
+    )
+    def test_from_freq_recreate_from_data(self, freq):
+        org = date_range(start="2001/02/01 09:00", freq=freq, periods=1)
+        idx = DatetimeIndex(org, freq=freq)
+        tm.assert_index_equal(idx, org)
+
+        org = date_range(
+            start="2001/02/01 09:00", freq=freq, tz="US/Pacific", periods=1
+        )
+        idx = DatetimeIndex(org, freq=freq, tz="US/Pacific")
+        tm.assert_index_equal(idx, org)
+
+    def test_datetimeindex_constructor_misc(self):
+        arr = ["1/1/2005", "1/2/2005", "Jn 3, 2005", "2005-01-04"]
+        msg = r"(\(')?Unknown datetime string format(:', 'Jn 3, 2005'\))?"
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(arr)
+
+        arr = ["1/1/2005", "1/2/2005", "1/3/2005", "2005-01-04"]
+        idx1 = DatetimeIndex(arr)
+
+        arr = [datetime(2005, 1, 1), "1/2/2005", "1/3/2005", "2005-01-04"]
+        idx2 = DatetimeIndex(arr)
+
+        arr = [Timestamp(datetime(2005, 1, 1)), "1/2/2005", "1/3/2005", "2005-01-04"]
+        idx3 = DatetimeIndex(arr)
+
+        arr = np.array(["1/1/2005", "1/2/2005", "1/3/2005", "2005-01-04"], dtype="O")
+        idx4 = DatetimeIndex(arr)
+
+        idx5 = DatetimeIndex(["12/05/2007", "25/01/2008"], dayfirst=True)
+        idx6 = DatetimeIndex(
+            ["2007/05/12", "2008/01/25"], dayfirst=False, yearfirst=True
+        )
+        tm.assert_index_equal(idx5, idx6)
+
+        for other in [idx2, idx3, idx4]:
+            assert (idx1.values == other.values).all()
+
+    def test_dti_constructor_object_dtype_dayfirst_yearfirst_with_tz(self):
+        # GH#55813
+        val = "5/10/16"
+
+        dfirst = Timestamp(2016, 10, 5, tz="US/Pacific")
+        yfirst = Timestamp(2005, 10, 16, tz="US/Pacific")
+
+        result1 = DatetimeIndex([val], tz="US/Pacific", dayfirst=True)
+        expected1 = DatetimeIndex([dfirst]).as_unit("s")
+        tm.assert_index_equal(result1, expected1)
+
+        result2 = DatetimeIndex([val], tz="US/Pacific", yearfirst=True)
+        expected2 = DatetimeIndex([yfirst]).as_unit("s")
+        tm.assert_index_equal(result2, expected2)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_date_range.py b/pandas/pandas/tests/indexes/datetimes/test_date_range.py
new file mode 100644
index 0000000000000..e09883e95ecec
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_date_range.py
@@ -0,0 +1,1738 @@
+"""
+test date_range, bdate_range construction from the convenience range functions
+"""
+
+from datetime import (
+    datetime,
+    time,
+    timedelta,
+)
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import timezones
+from pandas._libs.tslibs.offsets import (
+    BDay,
+    CDay,
+    DateOffset,
+    MonthEnd,
+    prefix_mapping,
+)
+from pandas.errors import OutOfBoundsDatetime
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    bdate_range,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+from pandas.core.arrays.datetimes import _generate_range as generate_range
+from pandas.tests.indexes.datetimes.test_timezones import (
+    FixedOffset,
+    fixed_off_no_name,
+)
+
+from pandas.tseries.holiday import USFederalHolidayCalendar
+
+START, END = datetime(2009, 1, 1), datetime(2010, 1, 1)
+
+
+def _get_expected_range(
+    begin_to_match,
+    end_to_match,
+    both_range,
+    inclusive_endpoints,
+):
+    """Helper to get expected range from a both inclusive range"""
+    left_match = begin_to_match == both_range[0]
+    right_match = end_to_match == both_range[-1]
+
+    if inclusive_endpoints == "left" and right_match:
+        expected_range = both_range[:-1]
+    elif inclusive_endpoints == "right" and left_match:
+        expected_range = both_range[1:]
+    elif inclusive_endpoints == "neither" and left_match and right_match:
+        expected_range = both_range[1:-1]
+    elif inclusive_endpoints == "neither" and right_match:
+        expected_range = both_range[:-1]
+    elif inclusive_endpoints == "neither" and left_match:
+        expected_range = both_range[1:]
+    elif inclusive_endpoints == "both":
+        expected_range = both_range[:]
+    else:
+        expected_range = both_range[:]
+
+    return expected_range
+
+
+class TestTimestampEquivDateRange:
+    # Older tests in TestTimeSeries constructed their `stamp` objects
+    # using `date_range` instead of the `Timestamp` constructor.
+    # TestTimestampEquivDateRange checks that these are equivalent in the
+    # pertinent cases.
+
+    def test_date_range_timestamp_equiv(self):
+        rng = date_range("20090415", "20090519", tz="US/Eastern")
+        stamp = rng[0]
+
+        ts = Timestamp("20090415", tz="US/Eastern")
+        assert ts == stamp
+
+    def test_date_range_timestamp_equiv_dateutil(self):
+        rng = date_range("20090415", "20090519", tz="dateutil/US/Eastern")
+        stamp = rng[0]
+
+        ts = Timestamp("20090415", tz="dateutil/US/Eastern")
+        assert ts == stamp
+
+    def test_date_range_timestamp_equiv_explicit_pytz(self):
+        pytz = pytest.importorskip("pytz")
+        rng = date_range("20090415", "20090519", tz=pytz.timezone("US/Eastern"))
+        stamp = rng[0]
+
+        ts = Timestamp("20090415", tz=pytz.timezone("US/Eastern"))
+        assert ts == stamp
+
+    @td.skip_if_windows
+    def test_date_range_timestamp_equiv_explicit_dateutil(self):
+        from pandas._libs.tslibs.timezones import dateutil_gettz as gettz
+
+        rng = date_range("20090415", "20090519", tz=gettz("US/Eastern"))
+        stamp = rng[0]
+
+        ts = Timestamp("20090415", tz=gettz("US/Eastern"))
+        assert ts == stamp
+
+    def test_date_range_timestamp_equiv_from_datetime_instance(self):
+        datetime_instance = datetime(2014, 3, 4)
+        # build a timestamp with a frequency, since then it supports
+        # addition/subtraction of integers
+        timestamp_instance = date_range(datetime_instance, periods=1, freq="D")[0]
+
+        ts = Timestamp(datetime_instance)
+        assert ts == timestamp_instance
+
+    def test_date_range_timestamp_equiv_preserve_frequency(self):
+        timestamp_instance = date_range("2014-03-05", periods=1, freq="D")[0]
+        ts = Timestamp("2014-03-05")
+
+        assert timestamp_instance == ts
+
+
+class TestDateRanges:
+    def test_date_range_name(self):
+        idx = date_range(start="2000-01-01", periods=1, freq="YE", name="TEST")
+        assert idx.name == "TEST"
+
+    def test_date_range_invalid_periods(self):
+        msg = "periods must be an integer, got foo"
+        with pytest.raises(TypeError, match=msg):
+            date_range(start="1/1/2000", periods="foo", freq="D")
+
+    def test_date_range_fractional_period(self):
+        msg = "periods must be an integer"
+        with pytest.raises(TypeError, match=msg):
+            date_range("1/1/2000", periods=10.5)
+
+    @pytest.mark.parametrize("freq", ["2M", "1m", "2SM", "2BQ", "1bq", "2BY"])
+    def test_date_range_frequency_M_SM_BQ_BY_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range("1/1/2000", periods=4, freq=freq)
+
+    def test_date_range_tuple_freq_raises(self):
+        # GH#34703
+        edate = datetime(2000, 1, 1)
+        with pytest.raises(TypeError, match="pass as a string instead"):
+            date_range(end=edate, freq=("D", 5), periods=20)
+
+    @pytest.mark.parametrize("freq", ["ns", "us", "ms", "min", "s", "h", "D"])
+    def test_date_range_edges(self, freq):
+        # GH#13672
+        td = Timedelta(f"1{freq}")
+        ts = Timestamp("1970-01-01")
+
+        idx = date_range(
+            start=ts + td,
+            end=ts + 4 * td,
+            freq=freq,
+        )
+        exp = DatetimeIndex(
+            [ts + n * td for n in range(1, 5)],
+            dtype="M8[ns]",
+            freq=freq,
+        )
+        tm.assert_index_equal(idx, exp)
+
+        # start after end
+        idx = date_range(
+            start=ts + 4 * td,
+            end=ts + td,
+            freq=freq,
+        )
+        exp = DatetimeIndex([], dtype="M8[ns]", freq=freq)
+        tm.assert_index_equal(idx, exp)
+
+        # start matches end
+        idx = date_range(
+            start=ts + td,
+            end=ts + td,
+            freq=freq,
+        )
+        exp = DatetimeIndex([ts + td], dtype="M8[ns]", freq=freq)
+        tm.assert_index_equal(idx, exp)
+
+    def test_date_range_near_implementation_bound(self):
+        # GH#???
+        freq = Timedelta(1)
+
+        with pytest.raises(OutOfBoundsDatetime, match="Cannot generate range with"):
+            date_range(end=Timestamp.min, periods=2, freq=freq)
+
+    def test_date_range_nat(self):
+        # GH#11587
+        msg = "Neither `start` nor `end` can be NaT"
+        with pytest.raises(ValueError, match=msg):
+            date_range(start="2016-01-01", end=pd.NaT, freq="D")
+        with pytest.raises(ValueError, match=msg):
+            date_range(start=pd.NaT, end="2016-01-01", freq="D")
+
+    def test_date_range_multiplication_overflow(self):
+        # GH#24255
+        # check that overflows in calculating `addend = periods * stride`
+        #  are caught
+        with tm.assert_produces_warning(None):
+            # we should _not_ be seeing a overflow RuntimeWarning
+            dti = date_range(start="1677-09-22", periods=213503, freq="D")
+
+        assert dti[0] == Timestamp("1677-09-22")
+        assert len(dti) == 213503
+
+        msg = "Cannot generate range with"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range("1969-05-04", periods=200000000, freq="30000D")
+
+    def test_date_range_unsigned_overflow_handling(self):
+        # GH#24255
+        # case where `addend = periods * stride` overflows int64 bounds
+        #  but not uint64 bounds
+        dti = date_range(start="1677-09-22", end="2262-04-11", freq="D")
+
+        dti2 = date_range(start=dti[0], periods=len(dti), freq="D")
+        assert dti2.equals(dti)
+
+        dti3 = date_range(end=dti[-1], periods=len(dti), freq="D")
+        assert dti3.equals(dti)
+
+    def test_date_range_int64_overflow_non_recoverable(self):
+        # GH#24255
+        # case with start later than 1970-01-01, overflow int64 but not uint64
+        msg = "Cannot generate range with"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range(start="1970-02-01", periods=106752 * 24, freq="h")
+
+        # case with end before 1970-01-01, overflow int64 but not uint64
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range(end="1969-11-14", periods=106752 * 24, freq="h")
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "s_ts, e_ts", [("2262-02-23", "1969-11-14"), ("1970-02-01", "1677-10-22")]
+    )
+    def test_date_range_int64_overflow_stride_endpoint_different_signs(
+        self, s_ts, e_ts
+    ):
+        # cases where stride * periods overflow int64 and stride/endpoint
+        #  have different signs
+        start = Timestamp(s_ts)
+        end = Timestamp(e_ts)
+
+        expected = date_range(start=start, end=end, freq="-1h")
+        assert expected[0] == start
+        assert expected[-1] == end
+
+        dti = date_range(end=end, periods=len(expected), freq="-1h")
+        tm.assert_index_equal(dti, expected)
+
+    def test_date_range_out_of_bounds(self):
+        # GH#14187
+        msg = "Cannot generate range"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range("2016-01-01", periods=100000, freq="D")
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range(end="1763-10-12", periods=100000, freq="D")
+
+    def test_date_range_gen_error(self):
+        rng = date_range("1/1/2000 00:00", "1/1/2000 00:18", freq="5min")
+        assert len(rng) == 4
+
+    def test_date_range_normalize(self):
+        snap = datetime.today()
+        n = 50
+
+        rng = date_range(snap, periods=n, normalize=False, freq="2D")
+
+        offset = timedelta(2)
+        expected = DatetimeIndex(
+            [snap + i * offset for i in range(n)], dtype="M8[ns]", freq=offset
+        )
+
+        tm.assert_index_equal(rng, expected)
+
+        rng = date_range("1/1/2000 08:15", periods=n, normalize=False, freq="B")
+        the_time = time(8, 15)
+        for val in rng:
+            assert val.time() == the_time
+
+    def test_date_range_ambiguous_arguments(self):
+        # #2538
+        start = datetime(2011, 1, 1, 5, 3, 40)
+        end = datetime(2011, 1, 1, 8, 9, 40)
+
+        msg = (
+            "Of the four parameters: start, end, periods, and "
+            "freq, exactly three must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            date_range(start, end, periods=10, freq="s")
+
+    def test_date_range_convenience_periods(self, unit):
+        # GH 20808
+        result = date_range("2018-04-24", "2018-04-27", periods=3, unit=unit)
+        expected = DatetimeIndex(
+            ["2018-04-24 00:00:00", "2018-04-25 12:00:00", "2018-04-27 00:00:00"],
+            dtype=f"M8[{unit}]",
+            freq=None,
+        )
+
+        tm.assert_index_equal(result, expected)
+
+        # Test if spacing remains linear if tz changes to dst in range
+        result = date_range(
+            "2018-04-01 01:00:00",
+            "2018-04-01 04:00:00",
+            tz="Australia/Sydney",
+            periods=3,
+            unit=unit,
+        )
+        expected = DatetimeIndex(
+            [
+                Timestamp("2018-04-01 01:00:00+1100", tz="Australia/Sydney"),
+                Timestamp("2018-04-01 02:00:00+1000", tz="Australia/Sydney"),
+                Timestamp("2018-04-01 04:00:00+1000", tz="Australia/Sydney"),
+            ]
+        ).as_unit(unit)
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_index_comparison(self):
+        rng = date_range("2011-01-01", periods=3, tz="US/Eastern")
+        df = Series(rng).to_frame()
+        arr = np.array([rng.to_list()]).T
+        arr2 = np.array([rng]).T
+
+        with pytest.raises(ValueError, match="Unable to coerce to Series"):
+            rng == df
+
+        with pytest.raises(ValueError, match="Unable to coerce to Series"):
+            df == rng
+
+        expected = DataFrame([True, True, True])
+
+        results = df == arr2
+        tm.assert_frame_equal(results, expected)
+
+        expected = Series([True, True, True], name=0)
+
+        results = df[0] == arr2[:, 0]
+        tm.assert_series_equal(results, expected)
+
+        expected = np.array(
+            [[True, False, False], [False, True, False], [False, False, True]]
+        )
+        results = rng == arr
+        tm.assert_numpy_array_equal(results, expected)
+
+    @pytest.mark.parametrize(
+        "start,end,result_tz",
+        [
+            ["20180101", "20180103", "US/Eastern"],
+            [datetime(2018, 1, 1), datetime(2018, 1, 3), "US/Eastern"],
+            [Timestamp("20180101"), Timestamp("20180103"), "US/Eastern"],
+            [
+                Timestamp("20180101", tz="US/Eastern"),
+                Timestamp("20180103", tz="US/Eastern"),
+                "US/Eastern",
+            ],
+            [
+                Timestamp("20180101", tz="US/Eastern"),
+                Timestamp("20180103", tz="US/Eastern"),
+                None,
+            ],
+        ],
+    )
+    def test_date_range_linspacing_tz(self, start, end, result_tz):
+        # GH 20983
+        result = date_range(start, end, periods=3, tz=result_tz)
+        expected = date_range("20180101", periods=3, freq="D", tz="US/Eastern")
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_timedelta(self):
+        start = "2020-01-01"
+        end = "2020-01-11"
+        rng1 = date_range(start, end, freq="3D")
+        rng2 = date_range(start, end, freq=timedelta(days=3))
+        tm.assert_index_equal(rng1, rng2)
+
+    def test_range_misspecified(self):
+        # GH #1095
+        msg = (
+            "Of the four parameters: start, end, periods, and "
+            "freq, exactly three must be specified"
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(start="1/1/2000")
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(end="1/1/2000")
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(periods=10)
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(start="1/1/2000", freq="h")
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(end="1/1/2000", freq="h")
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(periods=10, freq="h")
+
+        with pytest.raises(ValueError, match=msg):
+            date_range()
+
+    def test_compat_replace(self):
+        # https://github.com/statsmodels/statsmodels/issues/3349
+        # replace should take ints/longs for compat
+        result = date_range(Timestamp("1960-04-01 00:00:00"), periods=76, freq="QS-JAN")
+        assert len(result) == 76
+
+    def test_catch_infinite_loop(self):
+        offset = offsets.DateOffset(minute=5)
+        # blow up, don't loop forever
+        msg = "Offset <DateOffset: minute=5> did not increment date"
+        with pytest.raises(ValueError, match=msg):
+            date_range(datetime(2011, 11, 11), datetime(2011, 11, 12), freq=offset)
+
+    def test_construct_over_dst(self, unit):
+        # GH 20854
+        pre_dst = Timestamp("2010-11-07 01:00:00").tz_localize(
+            "US/Pacific", ambiguous=True
+        )
+        pst_dst = Timestamp("2010-11-07 01:00:00").tz_localize(
+            "US/Pacific", ambiguous=False
+        )
+        expect_data = [
+            Timestamp("2010-11-07 00:00:00", tz="US/Pacific"),
+            pre_dst,
+            pst_dst,
+        ]
+        expected = DatetimeIndex(expect_data, freq="h").as_unit(unit)
+        result = date_range(
+            start="2010-11-7", periods=3, freq="h", tz="US/Pacific", unit=unit
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_construct_with_different_start_end_string_format(self, unit):
+        # GH 12064
+        result = date_range(
+            "2013-01-01 00:00:00+09:00",
+            "2013/01/01 02:00:00+09:00",
+            freq="h",
+            unit=unit,
+        )
+        expected = DatetimeIndex(
+            [
+                Timestamp("2013-01-01 00:00:00+09:00"),
+                Timestamp("2013-01-01 01:00:00+09:00"),
+                Timestamp("2013-01-01 02:00:00+09:00"),
+            ],
+            freq="h",
+        ).as_unit(unit)
+        tm.assert_index_equal(result, expected)
+
+    def test_error_with_zero_monthends(self):
+        msg = r"Offset <0 \* MonthEnds> did not increment date"
+        with pytest.raises(ValueError, match=msg):
+            date_range("1/1/2000", "1/1/2001", freq=MonthEnd(0))
+
+    def test_range_bug(self, unit):
+        # GH #770
+        offset = DateOffset(months=3)
+        result = date_range("2011-1-1", "2012-1-31", freq=offset, unit=unit)
+
+        start = datetime(2011, 1, 1)
+        expected = DatetimeIndex(
+            [start + i * offset for i in range(5)], dtype=f"M8[{unit}]", freq=offset
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_range_tz_pytz(self):
+        # see gh-2906
+        pytz = pytest.importorskip("pytz")
+        tz = pytz.timezone("US/Eastern")
+        start = tz.localize(datetime(2011, 1, 1))
+        end = tz.localize(datetime(2011, 1, 3))
+
+        dr = date_range(start=start, periods=3)
+        assert dr.tz.zone == tz.zone
+        assert dr[0] == start
+        assert dr[2] == end
+
+        dr = date_range(end=end, periods=3)
+        assert dr.tz.zone == tz.zone
+        assert dr[0] == start
+        assert dr[2] == end
+
+        dr = date_range(start=start, end=end)
+        assert dr.tz.zone == tz.zone
+        assert dr[0] == start
+        assert dr[2] == end
+
+    @pytest.mark.parametrize(
+        "start, end",
+        [
+            [datetime(2014, 3, 6), datetime(2014, 3, 12)],
+            [datetime(2013, 11, 1), datetime(2013, 11, 6)],
+        ],
+    )
+    def test_range_tz_dst_straddle_pytz(self, start, end):
+        pytz = pytest.importorskip("pytz")
+        tz = pytz.timezone("US/Eastern")
+        start = Timestamp(start, tz=tz)
+        end = Timestamp(end, tz=tz)
+        dr = date_range(start, end, freq="D")
+        assert dr[0] == start
+        assert dr[-1] == end
+        assert np.all(dr.hour == 0)
+
+        dr = date_range(start, end, freq="D", tz=tz)
+        assert dr[0] == start
+        assert dr[-1] == end
+        assert np.all(dr.hour == 0)
+
+        dr = date_range(
+            start.replace(tzinfo=None),
+            end.replace(tzinfo=None),
+            freq="D",
+            tz=tz,
+        )
+        assert dr[0] == start
+        assert dr[-1] == end
+        assert np.all(dr.hour == 0)
+
+    def test_range_tz_dateutil(self):
+        # see gh-2906
+
+        # Use maybe_get_tz to fix filename in tz under dateutil.
+        from pandas._libs.tslibs.timezones import maybe_get_tz
+
+        tz = lambda x: maybe_get_tz("dateutil/" + x)
+
+        start = datetime(2011, 1, 1, tzinfo=tz("US/Eastern"))
+        end = datetime(2011, 1, 3, tzinfo=tz("US/Eastern"))
+
+        dr = date_range(start=start, periods=3)
+        assert dr.tz == tz("US/Eastern")
+        assert dr[0] == start
+        assert dr[2] == end
+
+        dr = date_range(end=end, periods=3)
+        assert dr.tz == tz("US/Eastern")
+        assert dr[0] == start
+        assert dr[2] == end
+
+        dr = date_range(start=start, end=end)
+        assert dr.tz == tz("US/Eastern")
+        assert dr[0] == start
+        assert dr[2] == end
+
+    @pytest.mark.parametrize("freq", ["1D", "3D", "2ME", "7W", "3h", "YE"])
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_range_closed(self, freq, tz, inclusive_endpoints_fixture):
+        # GH#12409, GH#12684
+
+        begin = Timestamp("2011/1/1", tz=tz)
+        end = Timestamp("2014/1/1", tz=tz)
+
+        result_range = date_range(
+            begin, end, inclusive=inclusive_endpoints_fixture, freq=freq
+        )
+        both_range = date_range(begin, end, inclusive="both", freq=freq)
+        expected_range = _get_expected_range(
+            begin, end, both_range, inclusive_endpoints_fixture
+        )
+
+        tm.assert_index_equal(expected_range, result_range)
+
+    @pytest.mark.parametrize("freq", ["1D", "3D", "2ME", "7W", "3h", "YE"])
+    def test_range_with_tz_closed_with_tz_aware_start_end(
+        self, freq, inclusive_endpoints_fixture
+    ):
+        begin = Timestamp("2011/1/1")
+        end = Timestamp("2014/1/1")
+        begintz = Timestamp("2011/1/1", tz="US/Eastern")
+        endtz = Timestamp("2014/1/1", tz="US/Eastern")
+
+        result_range = date_range(
+            begin,
+            end,
+            inclusive=inclusive_endpoints_fixture,
+            freq=freq,
+            tz="US/Eastern",
+        )
+        both_range = date_range(
+            begin, end, inclusive="both", freq=freq, tz="US/Eastern"
+        )
+        expected_range = _get_expected_range(
+            begintz,
+            endtz,
+            both_range,
+            inclusive_endpoints_fixture,
+        )
+
+        tm.assert_index_equal(expected_range, result_range)
+
+    def test_range_closed_boundary(self, inclusive_endpoints_fixture):
+        # GH#11804
+        right_boundary = date_range(
+            "2015-09-12",
+            "2015-12-01",
+            freq="QS-MAR",
+            inclusive=inclusive_endpoints_fixture,
+        )
+        left_boundary = date_range(
+            "2015-09-01",
+            "2015-09-12",
+            freq="QS-MAR",
+            inclusive=inclusive_endpoints_fixture,
+        )
+        both_boundary = date_range(
+            "2015-09-01",
+            "2015-12-01",
+            freq="QS-MAR",
+            inclusive=inclusive_endpoints_fixture,
+        )
+        neither_boundary = date_range(
+            "2015-09-11",
+            "2015-09-12",
+            freq="QS-MAR",
+            inclusive=inclusive_endpoints_fixture,
+        )
+
+        expected_right = both_boundary
+        expected_left = both_boundary
+        expected_both = both_boundary
+
+        if inclusive_endpoints_fixture == "right":
+            expected_left = both_boundary[1:]
+        elif inclusive_endpoints_fixture == "left":
+            expected_right = both_boundary[:-1]
+        elif inclusive_endpoints_fixture == "both":
+            expected_right = both_boundary[1:]
+            expected_left = both_boundary[:-1]
+
+        expected_neither = both_boundary[1:-1]
+
+        tm.assert_index_equal(right_boundary, expected_right)
+        tm.assert_index_equal(left_boundary, expected_left)
+        tm.assert_index_equal(both_boundary, expected_both)
+        tm.assert_index_equal(neither_boundary, expected_neither)
+
+    def test_date_range_years_only(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        # GH#6961
+        rng1 = date_range("2014", "2015", freq="ME", tz=tz)
+        expected1 = date_range("2014-01-31", "2014-12-31", freq="ME", tz=tz)
+        tm.assert_index_equal(rng1, expected1)
+
+        rng2 = date_range("2014", "2015", freq="MS", tz=tz)
+        expected2 = date_range("2014-01-01", "2015-01-01", freq="MS", tz=tz)
+        tm.assert_index_equal(rng2, expected2)
+
+        rng3 = date_range("2014", "2020", freq="YE", tz=tz)
+        expected3 = date_range("2014-12-31", "2019-12-31", freq="YE", tz=tz)
+        tm.assert_index_equal(rng3, expected3)
+
+        rng4 = date_range("2014", "2020", freq="YS", tz=tz)
+        expected4 = date_range("2014-01-01", "2020-01-01", freq="YS", tz=tz)
+        tm.assert_index_equal(rng4, expected4)
+
+    def test_freq_divides_end_in_nanos(self):
+        # GH 10885
+        result_1 = date_range("2005-01-12 10:00", "2005-01-12 16:00", freq="345min")
+        result_2 = date_range("2005-01-13 10:00", "2005-01-13 16:00", freq="345min")
+        expected_1 = DatetimeIndex(
+            ["2005-01-12 10:00:00", "2005-01-12 15:45:00"],
+            dtype="datetime64[ns]",
+            freq="345min",
+            tz=None,
+        )
+        expected_2 = DatetimeIndex(
+            ["2005-01-13 10:00:00", "2005-01-13 15:45:00"],
+            dtype="datetime64[ns]",
+            freq="345min",
+            tz=None,
+        )
+        tm.assert_index_equal(result_1, expected_1)
+        tm.assert_index_equal(result_2, expected_2)
+
+    def test_cached_range_bug(self):
+        rng = date_range("2010-09-01 05:00:00", periods=50, freq=DateOffset(hours=6))
+        assert len(rng) == 50
+        assert rng[0] == datetime(2010, 9, 1, 5)
+
+    def test_timezone_comparison_bug(self):
+        # smoke test
+        start = Timestamp("20130220 10:00", tz="US/Eastern")
+        result = date_range(start, periods=2, tz="US/Eastern")
+        assert len(result) == 2
+
+    def test_timezone_comparison_assert(self):
+        start = Timestamp("20130220 10:00", tz="US/Eastern")
+        msg = "Inferred time zone not equal to passed time zone"
+        with pytest.raises(AssertionError, match=msg):
+            date_range(start, periods=2, tz="Europe/Berlin")
+
+    def test_negative_non_tick_frequency_descending_dates(self, tz_aware_fixture):
+        # GH 23270
+        tz = tz_aware_fixture
+        result = date_range(start="2011-06-01", end="2011-01-01", freq="-1MS", tz=tz)
+        expected = date_range(end="2011-06-01", start="2011-01-01", freq="1MS", tz=tz)[
+            ::-1
+        ]
+        tm.assert_index_equal(result, expected)
+
+    def test_range_where_start_equal_end(self, inclusive_endpoints_fixture):
+        # GH 43394
+        start = "2021-09-02"
+        end = "2021-09-02"
+        result = date_range(
+            start=start, end=end, freq="D", inclusive=inclusive_endpoints_fixture
+        )
+
+        both_range = date_range(start=start, end=end, freq="D", inclusive="both")
+        if inclusive_endpoints_fixture == "neither":
+            expected = both_range[1:-1]
+        elif inclusive_endpoints_fixture in ("left", "right", "both"):
+            expected = both_range[:]
+
+        tm.assert_index_equal(result, expected)
+
+    def test_freq_dateoffset_with_relateivedelta_nanos(self):
+        # GH 46877
+        freq = DateOffset(hours=10, days=57, nanoseconds=3)
+        result = date_range(end="1970-01-01 00:00:00", periods=10, freq=freq, name="a")
+        expected = DatetimeIndex(
+            [
+                "1968-08-02T05:59:59.999999973",
+                "1968-09-28T15:59:59.999999976",
+                "1968-11-25T01:59:59.999999979",
+                "1969-01-21T11:59:59.999999982",
+                "1969-03-19T21:59:59.999999985",
+                "1969-05-16T07:59:59.999999988",
+                "1969-07-12T17:59:59.999999991",
+                "1969-09-08T03:59:59.999999994",
+                "1969-11-04T13:59:59.999999997",
+                "1970-01-01T00:00:00.000000000",
+            ],
+            name="a",
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["2T", "2L", "1l", "1U", "2N", "2n"])
+    def test_frequency_H_T_S_L_U_N_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+        with pytest.raises(ValueError, match=msg):
+            date_range("1/1/2000", periods=2, freq=freq)
+
+    @pytest.mark.parametrize(
+        "freq_depr", ["m", "bm", "CBM", "SM", "BQ", "q-feb", "y-may", "Y-MAY"]
+    )
+    def test_frequency_raises(self, freq_depr):
+        msg = f"Invalid frequency: {freq_depr}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range("1/1/2000", periods=2, freq=freq_depr)
+
+    def test_date_range_bday(self):
+        sdate = datetime(1999, 12, 25)
+        idx = date_range(start=sdate, freq="1B", periods=20)
+        assert len(idx) == 20
+        assert idx[0] == sdate + 0 * offsets.BDay()
+        assert idx.freq == "B"
+
+    @pytest.mark.parametrize("freq", ["200A", "2A-MAY"])
+    def test_frequency_A_raises(self, freq):
+        freq_msg = re.split("[0-9]*", freq, maxsplit=1)[1]
+        msg = f"Invalid frequency: {freq_msg}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range("1/1/2000", periods=2, freq=freq)
+
+    @pytest.mark.parametrize(
+        "freq,freq_depr",
+        [
+            ("2W", "2w"),
+            ("2W-WED", "2w-wed"),
+            ("2B", "2b"),
+            ("2D", "2d"),
+            ("2C", "2c"),
+        ],
+    )
+    def test_date_range_depr_lowercase_frequency(self, freq, freq_depr):
+        # GH#58998
+        depr_msg = f"'{freq_depr[1:]}' is deprecated and will be removed "
+        "in a future version."
+
+        expected = date_range("1/1/2000", periods=4, freq=freq)
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            result = date_range("1/1/2000", periods=4, freq=freq_depr)
+        tm.assert_index_equal(result, expected)
+
+
+class TestDateRangeTZ:
+    """Tests for date_range with timezones"""
+
+    def test_hongkong_tz_convert(self):
+        # GH#1673 smoke test
+        dr = date_range("2012-01-01", "2012-01-10", freq="D", tz="Hongkong")
+
+        # it works!
+        dr.hour
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_date_range_span_dst_transition(self, tzstr):
+        # GH#1778
+
+        # Standard -> Daylight Savings Time
+        dr = date_range("03/06/2012 00:00", periods=200, freq="W-FRI", tz="US/Eastern")
+
+        assert (dr.hour == 0).all()
+
+        dr = date_range("2012-11-02", periods=10, tz=tzstr)
+        result = dr.hour
+        expected = pd.Index([0] * 10, dtype="int32")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_date_range_timezone_str_argument(self, tzstr):
+        tz = timezones.maybe_get_tz(tzstr)
+        result = date_range("1/1/2000", periods=10, tz=tzstr)
+        expected = date_range("1/1/2000", periods=10, tz=tz)
+
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_with_fixed_tz(self):
+        off = FixedOffset(420, "+07:00")
+        start = datetime(2012, 3, 11, 5, 0, 0, tzinfo=off)
+        end = datetime(2012, 6, 11, 5, 0, 0, tzinfo=off)
+        rng = date_range(start=start, end=end)
+        assert off == rng.tz
+
+        rng2 = date_range(start, periods=len(rng), tz=off)
+        tm.assert_index_equal(rng, rng2)
+
+        rng3 = date_range("3/11/2012 05:00:00+07:00", "6/11/2012 05:00:00+07:00")
+        assert (rng.values == rng3.values).all()
+
+    def test_date_range_with_fixedoffset_noname(self):
+        off = fixed_off_no_name
+        start = datetime(2012, 3, 11, 5, 0, 0, tzinfo=off)
+        end = datetime(2012, 6, 11, 5, 0, 0, tzinfo=off)
+        rng = date_range(start=start, end=end)
+        assert off == rng.tz
+
+        idx = pd.Index([start, end])
+        assert off == idx.tz
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_date_range_with_tz(self, tzstr):
+        stamp = Timestamp("3/11/2012 05:00", tz=tzstr)
+        assert stamp.hour == 5
+
+        rng = date_range("3/11/2012 04:00", periods=10, freq="h", tz=tzstr)
+
+        assert stamp == rng[1]
+
+    @pytest.mark.parametrize("tz", ["Europe/London", "dateutil/Europe/London"])
+    def test_date_range_ambiguous_endpoint(self, tz):
+        # construction with an ambiguous end-point
+        # GH#11626
+
+        with pytest.raises(ValueError, match="Cannot infer dst time"):
+            date_range(
+                "2013-10-26 23:00", "2013-10-27 01:00", tz="Europe/London", freq="h"
+            )
+
+        times = date_range(
+            "2013-10-26 23:00", "2013-10-27 01:00", freq="h", tz=tz, ambiguous="infer"
+        )
+        assert times[0] == Timestamp("2013-10-26 23:00", tz=tz)
+        assert times[-1] == Timestamp("2013-10-27 01:00:00+0000", tz=tz)
+
+    @pytest.mark.parametrize(
+        "tz, option, expected",
+        [
+            ["US/Pacific", "shift_forward", "2019-03-10 03:00"],
+            ["dateutil/US/Pacific", "shift_forward", "2019-03-10 03:00"],
+            ["US/Pacific", "shift_backward", "2019-03-10 01:00"],
+            ["dateutil/US/Pacific", "shift_backward", "2019-03-10 01:00"],
+            ["US/Pacific", timedelta(hours=1), "2019-03-10 03:00"],
+        ],
+    )
+    def test_date_range_nonexistent_endpoint(self, tz, option, expected):
+        # construction with an nonexistent end-point
+
+        with pytest.raises(ValueError, match="2019-03-10 02:00:00"):
+            date_range(
+                "2019-03-10 00:00", "2019-03-10 02:00", tz="US/Pacific", freq="h"
+            )
+
+        times = date_range(
+            "2019-03-10 00:00", "2019-03-10 02:00", freq="h", tz=tz, nonexistent=option
+        )
+        assert times[-1] == Timestamp(expected, tz=tz)
+
+
+class TestGenRangeGeneration:
+    @pytest.mark.parametrize(
+        "freqstr,offset",
+        [
+            ("B", BDay()),
+            ("C", CDay()),
+        ],
+    )
+    def test_generate(self, freqstr, offset):
+        rng1 = list(generate_range(START, END, periods=None, offset=offset, unit="ns"))
+        rng2 = list(generate_range(START, END, periods=None, offset=freqstr, unit="ns"))
+        assert rng1 == rng2
+
+    def test_1(self):
+        rng = list(
+            generate_range(
+                start=datetime(2009, 3, 25),
+                end=None,
+                periods=2,
+                offset=BDay(),
+                unit="ns",
+            )
+        )
+        expected = [datetime(2009, 3, 25), datetime(2009, 3, 26)]
+        assert rng == expected
+
+    def test_2(self):
+        rng = list(
+            generate_range(
+                start=datetime(2008, 1, 1),
+                end=datetime(2008, 1, 3),
+                periods=None,
+                offset=BDay(),
+                unit="ns",
+            )
+        )
+        expected = [datetime(2008, 1, 1), datetime(2008, 1, 2), datetime(2008, 1, 3)]
+        assert rng == expected
+
+    def test_3(self):
+        rng = list(
+            generate_range(
+                start=datetime(2008, 1, 5),
+                end=datetime(2008, 1, 6),
+                periods=None,
+                offset=BDay(),
+                unit="ns",
+            )
+        )
+        expected = []
+        assert rng == expected
+
+    def test_precision_finer_than_offset(self):
+        # GH#9907
+        result1 = date_range(
+            start="2015-04-15 00:00:03", end="2016-04-22 00:00:00", freq="QE"
+        )
+        result2 = date_range(
+            start="2015-04-15 00:00:03", end="2015-06-22 00:00:04", freq="W"
+        )
+        expected1_list = [
+            "2015-06-30 00:00:03",
+            "2015-09-30 00:00:03",
+            "2015-12-31 00:00:03",
+            "2016-03-31 00:00:03",
+        ]
+        expected2_list = [
+            "2015-04-19 00:00:03",
+            "2015-04-26 00:00:03",
+            "2015-05-03 00:00:03",
+            "2015-05-10 00:00:03",
+            "2015-05-17 00:00:03",
+            "2015-05-24 00:00:03",
+            "2015-05-31 00:00:03",
+            "2015-06-07 00:00:03",
+            "2015-06-14 00:00:03",
+            "2015-06-21 00:00:03",
+        ]
+        expected1 = DatetimeIndex(
+            expected1_list, dtype="datetime64[ns]", freq="QE-DEC", tz=None
+        )
+        expected2 = DatetimeIndex(
+            expected2_list, dtype="datetime64[ns]", freq="W-SUN", tz=None
+        )
+        tm.assert_index_equal(result1, expected1)
+        tm.assert_index_equal(result2, expected2)
+
+    dt1, dt2 = "2017-01-01", "2017-01-01"
+    tz1, tz2 = "US/Eastern", "Europe/London"
+
+    @pytest.mark.parametrize(
+        "start,end",
+        [
+            (Timestamp(dt1, tz=tz1), Timestamp(dt2)),
+            (Timestamp(dt1), Timestamp(dt2, tz=tz2)),
+            (Timestamp(dt1, tz=tz1), Timestamp(dt2, tz=tz2)),
+            (Timestamp(dt1, tz=tz2), Timestamp(dt2, tz=tz1)),
+        ],
+    )
+    def test_mismatching_tz_raises_err(self, start, end):
+        # issue 18488
+        msg = "Start and end cannot both be tz-aware with different timezones"
+        with pytest.raises(TypeError, match=msg):
+            date_range(start, end)
+        with pytest.raises(TypeError, match=msg):
+            date_range(start, end, freq=BDay())
+
+
+class TestBusinessDateRange:
+    def test_constructor(self):
+        bdate_range(START, END, freq=BDay())
+        bdate_range(START, periods=20, freq=BDay())
+        bdate_range(end=START, periods=20, freq=BDay())
+
+        msg = "periods must be an integer, got B"
+        with pytest.raises(TypeError, match=msg):
+            date_range("2011-1-1", "2012-1-1", "B")
+
+        with pytest.raises(TypeError, match=msg):
+            bdate_range("2011-1-1", "2012-1-1", "B")
+
+        msg = "freq must be specified for bdate_range; use date_range instead"
+        with pytest.raises(TypeError, match=msg):
+            bdate_range(START, END, periods=10, freq=None)
+
+    def test_misc(self):
+        end = datetime(2009, 5, 13)
+        dr = bdate_range(end=end, periods=20)
+        firstDate = end - 19 * BDay()
+
+        assert len(dr) == 20
+        assert dr[0] == firstDate
+        assert dr[-1] == end
+
+    def test_date_parse_failure(self):
+        badly_formed_date = "2007/100/1"
+
+        msg = "Unknown datetime string format, unable to parse: 2007/100/1"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(badly_formed_date)
+
+        with pytest.raises(ValueError, match=msg):
+            bdate_range(start=badly_formed_date, periods=10)
+
+        with pytest.raises(ValueError, match=msg):
+            bdate_range(end=badly_formed_date, periods=10)
+
+        with pytest.raises(ValueError, match=msg):
+            bdate_range(badly_formed_date, badly_formed_date)
+
+    def test_daterange_bug_456(self):
+        # GH #456
+        rng1 = bdate_range("12/5/2011", "12/5/2011")
+        rng2 = bdate_range("12/2/2011", "12/5/2011")
+        assert rng2._data.freq == BDay()
+
+        result = rng1.union(rng2)
+        assert isinstance(result, DatetimeIndex)
+
+    def test_bdays_and_open_boundaries(self, inclusive_endpoints_fixture):
+        # GH 6673
+        start = "2018-07-21"  # Saturday
+        end = "2018-07-29"  # Sunday
+        result = date_range(start, end, freq="B", inclusive=inclusive_endpoints_fixture)
+
+        bday_start = "2018-07-23"  # Monday
+        bday_end = "2018-07-27"  # Friday
+        expected = date_range(bday_start, bday_end, freq="D")
+        tm.assert_index_equal(result, expected)
+        # Note: we do _not_ expect the freqs to match here
+
+    def test_bday_near_overflow(self):
+        # GH#24252 avoid doing unnecessary addition that _would_ overflow
+        start = Timestamp.max.floor("D").to_pydatetime()
+        rng = date_range(start, end=None, periods=1, freq="B")
+        expected = DatetimeIndex([start], freq="B").as_unit("ns")
+        tm.assert_index_equal(rng, expected)
+
+    def test_bday_overflow_error(self):
+        # GH#24252 check that we get OutOfBoundsDatetime and not OverflowError
+        msg = "Out of bounds nanosecond timestamp"
+        start = Timestamp.max.floor("D").to_pydatetime()
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            date_range(start, periods=2, freq="B")
+
+
+class TestCustomDateRange:
+    def test_constructor(self):
+        bdate_range(START, END, freq=CDay())
+        bdate_range(START, periods=20, freq=CDay())
+        bdate_range(end=START, periods=20, freq=CDay())
+
+        msg = "periods must be an integer, got C"
+        with pytest.raises(TypeError, match=msg):
+            date_range("2011-1-1", "2012-1-1", "C")
+
+        with pytest.raises(TypeError, match=msg):
+            bdate_range("2011-1-1", "2012-1-1", "C")
+
+    def test_misc(self):
+        end = datetime(2009, 5, 13)
+        dr = bdate_range(end=end, periods=20, freq="C")
+        firstDate = end - 19 * CDay()
+
+        assert len(dr) == 20
+        assert dr[0] == firstDate
+        assert dr[-1] == end
+
+    def test_daterange_bug_456(self):
+        # GH #456
+        rng1 = bdate_range("12/5/2011", "12/5/2011", freq="C")
+        rng2 = bdate_range("12/2/2011", "12/5/2011", freq="C")
+        assert rng2._data.freq == CDay()
+
+        result = rng1.union(rng2)
+        assert isinstance(result, DatetimeIndex)
+
+    def test_cdaterange(self, unit):
+        result = bdate_range("2013-05-01", periods=3, freq="C", unit=unit)
+        expected = DatetimeIndex(
+            ["2013-05-01", "2013-05-02", "2013-05-03"], dtype=f"M8[{unit}]", freq="C"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+    def test_cdaterange_weekmask(self, unit):
+        result = bdate_range(
+            "2013-05-01", periods=3, freq="C", weekmask="Sun Mon Tue Wed Thu", unit=unit
+        )
+        expected = DatetimeIndex(
+            ["2013-05-01", "2013-05-02", "2013-05-05"],
+            dtype=f"M8[{unit}]",
+            freq=result.freq,
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        # raise with non-custom freq
+        msg = (
+            "a custom frequency string is required when holidays or "
+            "weekmask are passed, got frequency B"
+        )
+        with pytest.raises(ValueError, match=msg):
+            bdate_range("2013-05-01", periods=3, weekmask="Sun Mon Tue Wed Thu")
+
+    def test_cdaterange_holidays(self, unit):
+        result = bdate_range(
+            "2013-05-01", periods=3, freq="C", holidays=["2013-05-01"], unit=unit
+        )
+        expected = DatetimeIndex(
+            ["2013-05-02", "2013-05-03", "2013-05-06"],
+            dtype=f"M8[{unit}]",
+            freq=result.freq,
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        # raise with non-custom freq
+        msg = (
+            "a custom frequency string is required when holidays or "
+            "weekmask are passed, got frequency B"
+        )
+        with pytest.raises(ValueError, match=msg):
+            bdate_range("2013-05-01", periods=3, holidays=["2013-05-01"])
+
+    def test_cdaterange_weekmask_and_holidays(self, unit):
+        result = bdate_range(
+            "2013-05-01",
+            periods=3,
+            freq="C",
+            weekmask="Sun Mon Tue Wed Thu",
+            holidays=["2013-05-01"],
+            unit=unit,
+        )
+        expected = DatetimeIndex(
+            ["2013-05-02", "2013-05-05", "2013-05-06"],
+            dtype=f"M8[{unit}]",
+            freq=result.freq,
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+    def test_cdaterange_holidays_weekmask_requires_freqstr(self):
+        # raise with non-custom freq
+        msg = (
+            "a custom frequency string is required when holidays or "
+            "weekmask are passed, got frequency B"
+        )
+        with pytest.raises(ValueError, match=msg):
+            bdate_range(
+                "2013-05-01",
+                periods=3,
+                weekmask="Sun Mon Tue Wed Thu",
+                holidays=["2013-05-01"],
+            )
+
+    @pytest.mark.parametrize(
+        "freq", [freq for freq in prefix_mapping if freq.startswith("C")]
+    )
+    def test_all_custom_freq(self, freq):
+        # should not raise
+        bdate_range(
+            START, END, freq=freq, weekmask="Mon Wed Fri", holidays=["2009-03-14"]
+        )
+
+        bad_freq = freq + "FOO"
+        msg = f"invalid custom frequency string: {bad_freq}"
+        with pytest.raises(ValueError, match=msg):
+            bdate_range(START, END, freq=bad_freq)
+
+    @pytest.mark.parametrize(
+        "start_end",
+        [
+            ("2018-01-01T00:00:01.000Z", "2018-01-03T00:00:01.000Z"),
+            ("2018-01-01T00:00:00.010Z", "2018-01-03T00:00:00.010Z"),
+            ("2001-01-01T00:00:00.010Z", "2001-01-03T00:00:00.010Z"),
+        ],
+    )
+    def test_range_with_millisecond_resolution(self, start_end):
+        # https://github.com/pandas-dev/pandas/issues/24110
+        start, end = start_end
+        result = date_range(start=start, end=end, periods=2, inclusive="left")
+        expected = DatetimeIndex([start], dtype="M8[ns, UTC]")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "start,period,expected",
+        [
+            ("2022-07-23 00:00:00+02:00", 1, ["2022-07-25 00:00:00+02:00"]),
+            ("2022-07-22 00:00:00+02:00", 1, ["2022-07-22 00:00:00+02:00"]),
+            (
+                "2022-07-22 00:00:00+02:00",
+                2,
+                ["2022-07-22 00:00:00+02:00", "2022-07-25 00:00:00+02:00"],
+            ),
+        ],
+    )
+    def test_range_with_timezone_and_custombusinessday(self, start, period, expected):
+        # GH49441
+        result = date_range(start=start, periods=period, freq="C")
+        expected = DatetimeIndex(expected).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_data_range_custombusinessday_partial_time(self, unit):
+        # GH#57456
+        offset = offsets.CustomBusinessDay(weekmask="Sun Mon Tue")
+        start = datetime(2024, 2, 6, 23)
+        # end datetime is partial and not in the offset
+        end = datetime(2024, 2, 14, 14)
+        result = date_range(start, end, freq=offset, unit=unit)
+        expected = DatetimeIndex(
+            [
+                "2024-02-06 23:00:00",
+                "2024-02-11 23:00:00",
+                "2024-02-12 23:00:00",
+                "2024-02-13 23:00:00",
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_index_equal(result, expected)
+
+
+class TestDateRangeNonNano:
+    def test_date_range_reso_validation(self):
+        msg = "'unit' must be one of 's', 'ms', 'us', 'ns'"
+        with pytest.raises(ValueError, match=msg):
+            date_range("2016-01-01", "2016-03-04", periods=3, unit="h")
+
+    def test_date_range_freq_higher_than_reso(self):
+        # freq being higher-resolution than reso is a problem
+        msg = "Use a lower freq or a higher unit instead"
+        with pytest.raises(ValueError, match=msg):
+            #    # TODO give a more useful or informative message?
+            date_range("2016-01-01", "2016-01-02", freq="ns", unit="ms")
+
+    def test_date_range_freq_matches_reso(self):
+        # GH#49106 matching reso is OK
+        dti = date_range("2016-01-01", "2016-01-01 00:00:01", freq="ms", unit="ms")
+        rng = np.arange(1_451_606_400_000, 1_451_606_401_001, dtype=np.int64)
+        expected = DatetimeIndex(rng.view("M8[ms]"), freq="ms")
+        tm.assert_index_equal(dti, expected)
+
+        dti = date_range("2016-01-01", "2016-01-01 00:00:01", freq="us", unit="us")
+        rng = np.arange(1_451_606_400_000_000, 1_451_606_401_000_001, dtype=np.int64)
+        expected = DatetimeIndex(rng.view("M8[us]"), freq="us")
+        tm.assert_index_equal(dti, expected)
+
+        dti = date_range("2016-01-01", "2016-01-01 00:00:00.001", freq="ns", unit="ns")
+        rng = np.arange(
+            1_451_606_400_000_000_000, 1_451_606_400_001_000_001, dtype=np.int64
+        )
+        expected = DatetimeIndex(rng.view("M8[ns]"), freq="ns")
+        tm.assert_index_equal(dti, expected)
+
+    def test_date_range_freq_lower_than_endpoints(self):
+        start = Timestamp("2022-10-19 11:50:44.719781")
+        end = Timestamp("2022-10-19 11:50:47.066458")
+
+        # start and end cannot be cast to "s" unit without lossy rounding,
+        #  so we do not allow this in date_range
+        with pytest.raises(ValueError, match="Cannot losslessly convert units"):
+            date_range(start, end, periods=3, unit="s")
+
+        # but we can losslessly cast to "us"
+        dti = date_range(start, end, periods=2, unit="us")
+        rng = np.array(
+            [start.as_unit("us")._value, end.as_unit("us")._value], dtype=np.int64
+        )
+        expected = DatetimeIndex(rng.view("M8[us]"))
+        tm.assert_index_equal(dti, expected)
+
+    def test_date_range_non_nano(self):
+        start = np.datetime64("1066-10-14")  # Battle of Hastings
+        end = np.datetime64("2305-07-13")  # Jean-Luc Picard's birthday
+
+        dti = date_range(start, end, freq="D", unit="s")
+        assert dti.freq == "D"
+        assert dti.dtype == "M8[s]"
+
+        exp = np.arange(
+            start.astype("M8[s]").view("i8"),
+            (end + 1).astype("M8[s]").view("i8"),
+            24 * 3600,
+        ).view("M8[s]")
+
+        tm.assert_numpy_array_equal(dti.to_numpy(), exp)
+
+
+class TestDateRangeNonTickFreq:
+    # Tests revolving around less-common (non-Tick) `freq` keywords.
+
+    def test_date_range_custom_business_month_begin(self, unit):
+        hcal = USFederalHolidayCalendar()
+        freq = offsets.CBMonthBegin(calendar=hcal)
+        dti = date_range(start="20120101", end="20130101", freq=freq, unit=unit)
+        assert all(freq.is_on_offset(x) for x in dti)
+
+        expected = DatetimeIndex(
+            [
+                "2012-01-03",
+                "2012-02-01",
+                "2012-03-01",
+                "2012-04-02",
+                "2012-05-01",
+                "2012-06-01",
+                "2012-07-02",
+                "2012-08-01",
+                "2012-09-04",
+                "2012-10-01",
+                "2012-11-01",
+                "2012-12-03",
+            ],
+            dtype=f"M8[{unit}]",
+            freq=freq,
+        )
+        tm.assert_index_equal(dti, expected)
+
+    def test_date_range_custom_business_month_end(self, unit):
+        hcal = USFederalHolidayCalendar()
+        freq = offsets.CBMonthEnd(calendar=hcal)
+        dti = date_range(start="20120101", end="20130101", freq=freq, unit=unit)
+        assert all(freq.is_on_offset(x) for x in dti)
+
+        expected = DatetimeIndex(
+            [
+                "2012-01-31",
+                "2012-02-29",
+                "2012-03-30",
+                "2012-04-30",
+                "2012-05-31",
+                "2012-06-29",
+                "2012-07-31",
+                "2012-08-31",
+                "2012-09-28",
+                "2012-10-31",
+                "2012-11-30",
+                "2012-12-31",
+            ],
+            dtype=f"M8[{unit}]",
+            freq=freq,
+        )
+        tm.assert_index_equal(dti, expected)
+
+    def test_date_range_with_custom_holidays(self, unit):
+        # GH#30593
+        freq = offsets.CustomBusinessHour(start="15:00", holidays=["2020-11-26"])
+        result = date_range(start="2020-11-25 15:00", periods=4, freq=freq, unit=unit)
+        expected = DatetimeIndex(
+            [
+                "2020-11-25 15:00:00",
+                "2020-11-25 16:00:00",
+                "2020-11-27 15:00:00",
+                "2020-11-27 16:00:00",
+            ],
+            dtype=f"M8[{unit}]",
+            freq=freq,
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_businesshour(self, unit):
+        idx = DatetimeIndex(
+            [
+                "2014-07-04 09:00",
+                "2014-07-04 10:00",
+                "2014-07-04 11:00",
+                "2014-07-04 12:00",
+                "2014-07-04 13:00",
+                "2014-07-04 14:00",
+                "2014-07-04 15:00",
+                "2014-07-04 16:00",
+            ],
+            dtype=f"M8[{unit}]",
+            freq="bh",
+        )
+        rng = date_range("2014-07-04 09:00", "2014-07-04 16:00", freq="bh", unit=unit)
+        tm.assert_index_equal(idx, rng)
+
+        idx = DatetimeIndex(
+            ["2014-07-04 16:00", "2014-07-07 09:00"], dtype=f"M8[{unit}]", freq="bh"
+        )
+        rng = date_range("2014-07-04 16:00", "2014-07-07 09:00", freq="bh", unit=unit)
+        tm.assert_index_equal(idx, rng)
+
+        idx = DatetimeIndex(
+            [
+                "2014-07-04 09:00",
+                "2014-07-04 10:00",
+                "2014-07-04 11:00",
+                "2014-07-04 12:00",
+                "2014-07-04 13:00",
+                "2014-07-04 14:00",
+                "2014-07-04 15:00",
+                "2014-07-04 16:00",
+                "2014-07-07 09:00",
+                "2014-07-07 10:00",
+                "2014-07-07 11:00",
+                "2014-07-07 12:00",
+                "2014-07-07 13:00",
+                "2014-07-07 14:00",
+                "2014-07-07 15:00",
+                "2014-07-07 16:00",
+                "2014-07-08 09:00",
+                "2014-07-08 10:00",
+                "2014-07-08 11:00",
+                "2014-07-08 12:00",
+                "2014-07-08 13:00",
+                "2014-07-08 14:00",
+                "2014-07-08 15:00",
+                "2014-07-08 16:00",
+            ],
+            dtype=f"M8[{unit}]",
+            freq="bh",
+        )
+        rng = date_range("2014-07-04 09:00", "2014-07-08 16:00", freq="bh", unit=unit)
+        tm.assert_index_equal(idx, rng)
+
+    def test_date_range_business_hour2(self, unit):
+        idx1 = date_range(
+            start="2014-07-04 15:00", end="2014-07-08 10:00", freq="bh", unit=unit
+        )
+        idx2 = date_range(start="2014-07-04 15:00", periods=12, freq="bh", unit=unit)
+        idx3 = date_range(end="2014-07-08 10:00", periods=12, freq="bh", unit=unit)
+        expected = DatetimeIndex(
+            [
+                "2014-07-04 15:00",
+                "2014-07-04 16:00",
+                "2014-07-07 09:00",
+                "2014-07-07 10:00",
+                "2014-07-07 11:00",
+                "2014-07-07 12:00",
+                "2014-07-07 13:00",
+                "2014-07-07 14:00",
+                "2014-07-07 15:00",
+                "2014-07-07 16:00",
+                "2014-07-08 09:00",
+                "2014-07-08 10:00",
+            ],
+            dtype=f"M8[{unit}]",
+            freq="bh",
+        )
+        tm.assert_index_equal(idx1, expected)
+        tm.assert_index_equal(idx2, expected)
+        tm.assert_index_equal(idx3, expected)
+
+        idx4 = date_range(
+            start="2014-07-04 15:45", end="2014-07-08 10:45", freq="bh", unit=unit
+        )
+        idx5 = date_range(start="2014-07-04 15:45", periods=12, freq="bh", unit=unit)
+        idx6 = date_range(end="2014-07-08 10:45", periods=12, freq="bh", unit=unit)
+
+        expected2 = expected + Timedelta(minutes=45).as_unit(unit)
+        expected2.freq = "bh"
+        tm.assert_index_equal(idx4, expected2)
+        tm.assert_index_equal(idx5, expected2)
+        tm.assert_index_equal(idx6, expected2)
+
+    def test_date_range_business_hour_short(self, unit):
+        # GH#49835
+        idx4 = date_range(start="2014-07-01 10:00", freq="bh", periods=1, unit=unit)
+        expected4 = DatetimeIndex(["2014-07-01 10:00"], dtype=f"M8[{unit}]", freq="bh")
+        tm.assert_index_equal(idx4, expected4)
+
+    def test_date_range_year_start(self, unit):
+        # see GH#9313
+        rng = date_range("1/1/2013", "7/1/2017", freq="YS", unit=unit)
+        exp = DatetimeIndex(
+            ["2013-01-01", "2014-01-01", "2015-01-01", "2016-01-01", "2017-01-01"],
+            dtype=f"M8[{unit}]",
+            freq="YS",
+        )
+        tm.assert_index_equal(rng, exp)
+
+    def test_date_range_year_end(self, unit):
+        # see GH#9313
+        rng = date_range("1/1/2013", "7/1/2017", freq="YE", unit=unit)
+        exp = DatetimeIndex(
+            ["2013-12-31", "2014-12-31", "2015-12-31", "2016-12-31"],
+            dtype=f"M8[{unit}]",
+            freq="YE",
+        )
+        tm.assert_index_equal(rng, exp)
+
+    def test_date_range_negative_freq_year_end(self, unit):
+        # GH#11018
+        rng = date_range("2011-12-31", freq="-2YE", periods=3, unit=unit)
+        exp = DatetimeIndex(
+            ["2011-12-31", "2009-12-31", "2007-12-31"], dtype=f"M8[{unit}]", freq="-2YE"
+        )
+        tm.assert_index_equal(rng, exp)
+        assert rng.freq == "-2YE"
+
+    def test_date_range_business_year_end_year(self, unit):
+        # see GH#9313
+        rng = date_range("1/1/2013", "7/1/2017", freq="BYE", unit=unit)
+        exp = DatetimeIndex(
+            ["2013-12-31", "2014-12-31", "2015-12-31", "2016-12-30"],
+            dtype=f"M8[{unit}]",
+            freq="BYE",
+        )
+        tm.assert_index_equal(rng, exp)
+
+    def test_date_range_bms(self, unit):
+        # GH#1645
+        result = date_range("1/1/2000", periods=10, freq="BMS", unit=unit)
+
+        expected = DatetimeIndex(
+            [
+                "2000-01-03",
+                "2000-02-01",
+                "2000-03-01",
+                "2000-04-03",
+                "2000-05-01",
+                "2000-06-01",
+                "2000-07-03",
+                "2000-08-01",
+                "2000-09-01",
+                "2000-10-02",
+            ],
+            dtype=f"M8[{unit}]",
+            freq="BMS",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_semi_month_begin(self, unit):
+        dates = [
+            datetime(2007, 12, 15),
+            datetime(2008, 1, 1),
+            datetime(2008, 1, 15),
+            datetime(2008, 2, 1),
+            datetime(2008, 2, 15),
+            datetime(2008, 3, 1),
+            datetime(2008, 3, 15),
+            datetime(2008, 4, 1),
+            datetime(2008, 4, 15),
+            datetime(2008, 5, 1),
+            datetime(2008, 5, 15),
+            datetime(2008, 6, 1),
+            datetime(2008, 6, 15),
+            datetime(2008, 7, 1),
+            datetime(2008, 7, 15),
+            datetime(2008, 8, 1),
+            datetime(2008, 8, 15),
+            datetime(2008, 9, 1),
+            datetime(2008, 9, 15),
+            datetime(2008, 10, 1),
+            datetime(2008, 10, 15),
+            datetime(2008, 11, 1),
+            datetime(2008, 11, 15),
+            datetime(2008, 12, 1),
+            datetime(2008, 12, 15),
+        ]
+        # ensure generating a range with DatetimeIndex gives same result
+        result = date_range(start=dates[0], end=dates[-1], freq="SMS", unit=unit)
+        exp = DatetimeIndex(dates, dtype=f"M8[{unit}]", freq="SMS")
+        tm.assert_index_equal(result, exp)
+
+    def test_date_range_semi_month_end(self, unit):
+        dates = [
+            datetime(2007, 12, 31),
+            datetime(2008, 1, 15),
+            datetime(2008, 1, 31),
+            datetime(2008, 2, 15),
+            datetime(2008, 2, 29),
+            datetime(2008, 3, 15),
+            datetime(2008, 3, 31),
+            datetime(2008, 4, 15),
+            datetime(2008, 4, 30),
+            datetime(2008, 5, 15),
+            datetime(2008, 5, 31),
+            datetime(2008, 6, 15),
+            datetime(2008, 6, 30),
+            datetime(2008, 7, 15),
+            datetime(2008, 7, 31),
+            datetime(2008, 8, 15),
+            datetime(2008, 8, 31),
+            datetime(2008, 9, 15),
+            datetime(2008, 9, 30),
+            datetime(2008, 10, 15),
+            datetime(2008, 10, 31),
+            datetime(2008, 11, 15),
+            datetime(2008, 11, 30),
+            datetime(2008, 12, 15),
+            datetime(2008, 12, 31),
+        ]
+        # ensure generating a range with DatetimeIndex gives same result
+        result = date_range(start=dates[0], end=dates[-1], freq="SME", unit=unit)
+        exp = DatetimeIndex(dates, dtype=f"M8[{unit}]", freq="SME")
+        tm.assert_index_equal(result, exp)
+
+    def test_date_range_week_of_month(self, unit):
+        # GH#20517
+        # Note the start here is not on_offset for this freq
+        result = date_range(start="20110101", periods=1, freq="WOM-1MON", unit=unit)
+        expected = DatetimeIndex(["2011-01-03"], dtype=f"M8[{unit}]", freq="WOM-1MON")
+        tm.assert_index_equal(result, expected)
+
+        result2 = date_range(start="20110101", periods=2, freq="WOM-1MON", unit=unit)
+        expected2 = DatetimeIndex(
+            ["2011-01-03", "2011-02-07"], dtype=f"M8[{unit}]", freq="WOM-1MON"
+        )
+        tm.assert_index_equal(result2, expected2)
+
+    def test_date_range_week_of_month2(self, unit):
+        # GH#5115, GH#5348
+        result = date_range("2013-1-1", periods=4, freq="WOM-1SAT", unit=unit)
+        expected = DatetimeIndex(
+            ["2013-01-05", "2013-02-02", "2013-03-02", "2013-04-06"],
+            dtype=f"M8[{unit}]",
+            freq="WOM-1SAT",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_negative_freq_month_end(self, unit):
+        # GH#11018
+        rng = date_range("2011-01-31", freq="-2ME", periods=3, unit=unit)
+        exp = DatetimeIndex(
+            ["2011-01-31", "2010-11-30", "2010-09-30"], dtype=f"M8[{unit}]", freq="-2ME"
+        )
+        tm.assert_index_equal(rng, exp)
+        assert rng.freq == "-2ME"
+
+    def test_date_range_fy5253(self, unit):
+        freq = offsets.FY5253(startingMonth=1, weekday=3, variation="nearest")
+        dti = date_range(
+            start="2013-01-01",
+            periods=2,
+            freq=freq,
+            unit=unit,
+        )
+        expected = DatetimeIndex(
+            ["2013-01-31", "2014-01-30"], dtype=f"M8[{unit}]", freq=freq
+        )
+
+        tm.assert_index_equal(dti, expected)
+
+    @pytest.mark.parametrize(
+        "freqstr,offset",
+        [
+            ("QS", offsets.QuarterBegin(startingMonth=1)),
+            ("BQE", offsets.BQuarterEnd(startingMonth=12)),
+            ("W-SUN", offsets.Week(weekday=6)),
+        ],
+    )
+    def test_date_range_freqstr_matches_offset(self, freqstr, offset):
+        sdate = datetime(1999, 12, 25)
+        edate = datetime(2000, 1, 1)
+
+        idx1 = date_range(start=sdate, end=edate, freq=freqstr)
+        idx2 = date_range(start=sdate, end=edate, freq=offset)
+        assert len(idx1) == len(idx2)
+        assert idx1.freq == idx2.freq
+
+    def test_date_range_partial_day_year_end(self, unit):
+        # GH#56134
+        rng = date_range(
+            start="2021-12-31 00:00:01",
+            end="2023-10-31 00:00:00",
+            freq="YE",
+            unit=unit,
+        )
+        exp = DatetimeIndex(
+            ["2021-12-31 00:00:01", "2022-12-31 00:00:01"],
+            dtype=f"M8[{unit}]",
+            freq="YE",
+        )
+        tm.assert_index_equal(rng, exp)
+
+    def test_date_range_negative_freq_year_end_inbounds(self, unit):
+        # GH#56147
+        rng = date_range(
+            start="2023-10-31 00:00:00",
+            end="2021-10-31 00:00:00",
+            freq="-1YE",
+            unit=unit,
+        )
+        exp = DatetimeIndex(
+            ["2022-12-31 00:00:00", "2021-12-31 00:00:00"],
+            dtype=f"M8[{unit}]",
+            freq="-1YE",
+        )
+        tm.assert_index_equal(rng, exp)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_datetime.py b/pandas/pandas/tests/indexes/datetimes/test_datetime.py
new file mode 100644
index 0000000000000..04334a1d8d0c8
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_datetime.py
@@ -0,0 +1,155 @@
+import datetime as dt
+from datetime import date
+
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_long
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Timestamp,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndex:
+    def test_is_(self):
+        dti = date_range(start="1/1/2005", end="12/1/2005", freq="ME")
+        assert dti.is_(dti)
+        assert dti.is_(dti.view())
+        assert not dti.is_(dti.copy())
+
+    def test_time_overflow_for_32bit_machines(self):
+        # GH8943.  On some machines NumPy defaults to np.int32 (for example,
+        # 32-bit Linux machines).  In the function _generate_regular_range
+        # found in tseries/index.py, `periods` gets multiplied by `strides`
+        # (which has value 1e9) and since the max value for np.int32 is ~2e9,
+        # and since those machines won't promote np.int32 to np.int64, we get
+        # overflow.
+        periods = np_long(1000)
+
+        idx1 = date_range(start="2000", periods=periods, freq="s")
+        assert len(idx1) == periods
+
+        idx2 = date_range(end="2000", periods=periods, freq="s")
+        assert len(idx2) == periods
+
+    def test_nat(self):
+        assert DatetimeIndex([np.nan])[0] is pd.NaT
+
+    def test_week_of_month_frequency(self):
+        # GH 5348: "ValueError: Could not evaluate WOM-1SUN" shouldn't raise
+        d1 = date(2002, 9, 1)
+        d2 = date(2013, 10, 27)
+        d3 = date(2012, 9, 30)
+        idx1 = DatetimeIndex([d1, d2])
+        idx2 = DatetimeIndex([d3])
+        result_append = idx1.append(idx2)
+        expected = DatetimeIndex([d1, d2, d3])
+        tm.assert_index_equal(result_append, expected)
+        result_union = idx1.union(idx2)
+        expected = DatetimeIndex([d1, d3, d2])
+        tm.assert_index_equal(result_union, expected)
+
+    def test_append_nondatetimeindex(self):
+        rng = date_range("1/1/2000", periods=10)
+        idx = Index(["a", "b", "c", "d"])
+
+        result = rng.append(idx)
+        assert isinstance(result[0], Timestamp)
+
+    def test_misc_coverage(self):
+        rng = date_range("1/1/2000", periods=5)
+        result = rng.groupby(rng.day)
+        assert isinstance(next(iter(result.values()))[0], Timestamp)
+
+    # TODO: belongs in frame groupby tests?
+    def test_groupby_function_tuple_1677(self):
+        df = DataFrame(
+            np.random.default_rng(2).random(100),
+            index=date_range("1/1/2000", periods=100),
+        )
+        monthly_group = df.groupby(lambda x: (x.year, x.month))
+
+        result = monthly_group.mean()
+        assert isinstance(result.index[0], tuple)
+
+    def assert_index_parameters(self, index):
+        assert index.freq == "40960ns"
+        assert index.inferred_freq == "40960ns"
+
+    def test_ns_index(self):
+        nsamples = 400
+        ns = int(1e9 / 24414)
+        dtstart = np.datetime64("2012-09-20T00:00:00")
+
+        dt = dtstart + np.arange(nsamples) * np.timedelta64(ns, "ns")
+        freq = ns * offsets.Nano()
+        index = DatetimeIndex(dt, freq=freq, name="time")
+        self.assert_index_parameters(index)
+
+        new_index = date_range(start=index[0], end=index[-1], freq=index.freq)
+        self.assert_index_parameters(new_index)
+
+    def test_asarray_tz_naive(self):
+        # This shouldn't produce a warning.
+        idx = date_range("2000", periods=2)
+        # M8[ns] by default
+        result = np.asarray(idx)
+
+        expected = np.array(["2000-01-01", "2000-01-02"], dtype="M8[ns]")
+        tm.assert_numpy_array_equal(result, expected)
+
+        # optionally, object
+        result = np.asarray(idx, dtype=object)
+
+        expected = np.array([Timestamp("2000-01-01"), Timestamp("2000-01-02")])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_asarray_tz_aware(self):
+        tz = "US/Central"
+        idx = date_range("2000", periods=2, tz=tz)
+        expected = np.array(["2000-01-01T06", "2000-01-02T06"], dtype="M8[ns]")
+        result = np.asarray(idx, dtype="datetime64[ns]")
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        # Old behavior with no warning
+        result = np.asarray(idx, dtype="M8[ns]")
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        # Future behavior with no warning
+        expected = np.array(
+            [Timestamp("2000-01-01", tz=tz), Timestamp("2000-01-02", tz=tz)]
+        )
+        result = np.asarray(idx, dtype=object)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["2H", "2BH", "2S"])
+    def test_CBH_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(dt.datetime(2022, 12, 11), dt.datetime(2022, 12, 13), freq=freq)
+
+    @pytest.mark.parametrize("freq", ["2BM", "1bm", "2BQ", "1BQ-MAR", "2BY-JUN", "1by"])
+    def test_BM_BQ_BY_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(start="2016-02-21", end="2016-08-21", freq=freq)
+
+    @pytest.mark.parametrize("freq", ["2BA-MAR", "1BAS-MAY", "2AS-AUG"])
+    def test_BA_BAS_raises(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            date_range(start="2016-02-21", end="2016-08-21", freq=freq)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_formats.py b/pandas/pandas/tests/indexes/datetimes/test_formats.py
new file mode 100644
index 0000000000000..f4e0a63043335
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_formats.py
@@ -0,0 +1,281 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import dateutil.tz
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_get_values_for_csv():
+    index = pd.date_range(freq="1D", periods=3, start="2017-01-01")
+
+    # First, with no arguments.
+    expected = np.array(["2017-01-01", "2017-01-02", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv()
+    tm.assert_numpy_array_equal(result, expected)
+
+    # No NaN values, so na_rep has no effect
+    result = index._get_values_for_csv(na_rep="pandas")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # Make sure date formatting works
+    expected = np.array(["01-2017-01", "01-2017-02", "01-2017-03"], dtype=object)
+
+    result = index._get_values_for_csv(date_format="%m-%Y-%d")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # NULL object handling should work
+    index = DatetimeIndex(["2017-01-01", NaT, "2017-01-03"])
+    expected = np.array(["2017-01-01", "NaT", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv(na_rep="NaT")
+    tm.assert_numpy_array_equal(result, expected)
+
+    expected = np.array(["2017-01-01", "pandas", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv(na_rep="pandas")
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = index._get_values_for_csv(na_rep="NaT", date_format="%Y-%m-%d %H:%M:%S.%f")
+    expected = np.array(
+        ["2017-01-01 00:00:00.000000", "NaT", "2017-01-03 00:00:00.000000"],
+        dtype=object,
+    )
+    tm.assert_numpy_array_equal(result, expected)
+
+    # invalid format
+    result = index._get_values_for_csv(na_rep="NaT", date_format="foo")
+    expected = np.array(["foo", "NaT", "foo"], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+class TestDatetimeIndexRendering:
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_with_timezone_repr(self, tzstr):
+        rng = pd.date_range("4/13/2010", "5/6/2010")
+
+        rng_eastern = rng.tz_localize(tzstr)
+
+        rng_repr = repr(rng_eastern)
+        assert "2010-04-13 00:00:00" in rng_repr
+
+    def test_dti_repr_dates(self):
+        text = str(pd.to_datetime([datetime(2013, 1, 1), datetime(2014, 1, 1)]))
+        assert "['2013-01-01'," in text
+        assert ", '2014-01-01']" in text
+
+    def test_dti_repr_mixed(self):
+        text = str(
+            pd.to_datetime(
+                [datetime(2013, 1, 1), datetime(2014, 1, 1, 12), datetime(2014, 1, 1)]
+            )
+        )
+        assert "'2013-01-01 00:00:00'," in text
+        assert "'2014-01-01 00:00:00']" in text
+
+    def test_dti_repr_short(self):
+        dr = pd.date_range(start="1/1/2012", periods=1)
+        repr(dr)
+
+        dr = pd.date_range(start="1/1/2012", periods=2)
+        repr(dr)
+
+        dr = pd.date_range(start="1/1/2012", periods=3)
+        repr(dr)
+
+    @pytest.mark.parametrize(
+        "dates, freq, expected_repr",
+        [
+            (
+                ["2012-01-01 00:00:00"],
+                "60min",
+                (
+                    "DatetimeIndex(['2012-01-01 00:00:00'], "
+                    "dtype='datetime64[ns]', freq='60min')"
+                ),
+            ),
+            (
+                ["2012-01-01 00:00:00", "2012-01-01 01:00:00"],
+                "60min",
+                "DatetimeIndex(['2012-01-01 00:00:00', '2012-01-01 01:00:00'], "
+                "dtype='datetime64[ns]', freq='60min')",
+            ),
+            (
+                ["2012-01-01"],
+                "24h",
+                "DatetimeIndex(['2012-01-01'], dtype='datetime64[ns]', freq='24h')",
+            ),
+        ],
+    )
+    def test_dti_repr_time_midnight(self, dates, freq, expected_repr, unit):
+        # GH53634
+        dti = DatetimeIndex(dates, freq).as_unit(unit)
+        actual_repr = repr(dti)
+        assert actual_repr == expected_repr.replace("[ns]", f"[{unit}]")
+
+    def test_dti_representation(self, unit):
+        idxs = []
+        idxs.append(DatetimeIndex([], freq="D"))
+        idxs.append(DatetimeIndex(["2011-01-01"], freq="D"))
+        idxs.append(DatetimeIndex(["2011-01-01", "2011-01-02"], freq="D"))
+        idxs.append(DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D"))
+        idxs.append(
+            DatetimeIndex(
+                ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
+                freq="h",
+                tz="Asia/Tokyo",
+            )
+        )
+        idxs.append(
+            DatetimeIndex(
+                ["2011-01-01 09:00", "2011-01-01 10:00", NaT], tz="US/Eastern"
+            )
+        )
+        idxs.append(
+            DatetimeIndex(["2011-01-01 09:00", "2011-01-01 10:00", NaT], tz="UTC")
+        )
+
+        exp = []
+        exp.append("DatetimeIndex([], dtype='datetime64[ns]', freq='D')")
+        exp.append("DatetimeIndex(['2011-01-01'], dtype='datetime64[ns]', freq='D')")
+        exp.append(
+            "DatetimeIndex(['2011-01-01', '2011-01-02'], "
+            "dtype='datetime64[ns]', freq='D')"
+        )
+        exp.append(
+            "DatetimeIndex(['2011-01-01', '2011-01-02', '2011-01-03'], "
+            "dtype='datetime64[ns]', freq='D')"
+        )
+        exp.append(
+            "DatetimeIndex(['2011-01-01 09:00:00+09:00', "
+            "'2011-01-01 10:00:00+09:00', '2011-01-01 11:00:00+09:00']"
+            ", dtype='datetime64[ns, Asia/Tokyo]', freq='h')"
+        )
+        exp.append(
+            "DatetimeIndex(['2011-01-01 09:00:00-05:00', "
+            "'2011-01-01 10:00:00-05:00', 'NaT'], "
+            "dtype='datetime64[ns, US/Eastern]', freq=None)"
+        )
+        exp.append(
+            "DatetimeIndex(['2011-01-01 09:00:00+00:00', "
+            "'2011-01-01 10:00:00+00:00', 'NaT'], "
+            "dtype='datetime64[ns, UTC]', freq=None)"
+            ""
+        )
+
+        with pd.option_context("display.width", 300):
+            for index, expected in zip(idxs, exp):
+                index = index.as_unit(unit)
+                expected = expected.replace("[ns", f"[{unit}")
+                result = repr(index)
+                assert result == expected
+                result = str(index)
+                assert result == expected
+
+    # TODO: this is a Series.__repr__ test
+    def test_dti_representation_to_series(self, unit):
+        idx1 = DatetimeIndex([], freq="D")
+        idx2 = DatetimeIndex(["2011-01-01"], freq="D")
+        idx3 = DatetimeIndex(["2011-01-01", "2011-01-02"], freq="D")
+        idx4 = DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        idx5 = DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
+            freq="h",
+            tz="Asia/Tokyo",
+        )
+        idx6 = DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", NaT], tz="US/Eastern"
+        )
+        idx7 = DatetimeIndex(["2011-01-01 09:00", "2011-01-02 10:15"])
+
+        exp1 = """Series([], dtype: datetime64[ns])"""
+
+        exp2 = "0   2011-01-01\ndtype: datetime64[ns]"
+
+        exp3 = "0   2011-01-01\n1   2011-01-02\ndtype: datetime64[ns]"
+
+        exp4 = "0   2011-01-01\n1   2011-01-02\n2   2011-01-03\ndtype: datetime64[ns]"
+
+        exp5 = (
+            "0   2011-01-01 09:00:00+09:00\n"
+            "1   2011-01-01 10:00:00+09:00\n"
+            "2   2011-01-01 11:00:00+09:00\n"
+            "dtype: datetime64[ns, Asia/Tokyo]"
+        )
+
+        exp6 = (
+            "0   2011-01-01 09:00:00-05:00\n"
+            "1   2011-01-01 10:00:00-05:00\n"
+            "2                         NaT\n"
+            "dtype: datetime64[ns, US/Eastern]"
+        )
+
+        exp7 = "0   2011-01-01 09:00:00\n1   2011-01-02 10:15:00\ndtype: datetime64[ns]"
+
+        with pd.option_context("display.width", 300):
+            for idx, expected in zip(
+                [idx1, idx2, idx3, idx4, idx5, idx6, idx7],
+                [exp1, exp2, exp3, exp4, exp5, exp6, exp7],
+            ):
+                ser = Series(idx.as_unit(unit))
+                result = repr(ser)
+                assert result == expected.replace("[ns", f"[{unit}")
+
+    def test_dti_summary(self):
+        # GH#9116
+        idx1 = DatetimeIndex([], freq="D")
+        idx2 = DatetimeIndex(["2011-01-01"], freq="D")
+        idx3 = DatetimeIndex(["2011-01-01", "2011-01-02"], freq="D")
+        idx4 = DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        idx5 = DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
+            freq="h",
+            tz="Asia/Tokyo",
+        )
+        idx6 = DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", NaT], tz="US/Eastern"
+        )
+
+        exp1 = "DatetimeIndex: 0 entries\nFreq: D"
+
+        exp2 = "DatetimeIndex: 1 entries, 2011-01-01 to 2011-01-01\nFreq: D"
+
+        exp3 = "DatetimeIndex: 2 entries, 2011-01-01 to 2011-01-02\nFreq: D"
+
+        exp4 = "DatetimeIndex: 3 entries, 2011-01-01 to 2011-01-03\nFreq: D"
+
+        exp5 = (
+            "DatetimeIndex: 3 entries, 2011-01-01 09:00:00+09:00 "
+            "to 2011-01-01 11:00:00+09:00\n"
+            "Freq: h"
+        )
+
+        exp6 = """DatetimeIndex: 3 entries, 2011-01-01 09:00:00-05:00 to NaT"""
+
+        for idx, expected in zip(
+            [idx1, idx2, idx3, idx4, idx5, idx6], [exp1, exp2, exp3, exp4, exp5, exp6]
+        ):
+            result = idx._summary()
+            assert result == expected
+
+    @pytest.mark.parametrize("tz", [None, timezone.utc, dateutil.tz.tzutc()])
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_dti_business_repr_etc_smoke(self, tz, freq):
+        # only really care that it works
+        dti = pd.bdate_range(
+            datetime(2009, 1, 1), datetime(2010, 1, 1), tz=tz, freq=freq
+        )
+        repr(dti)
+        dti._summary()
+        dti[2:2]._summary()
diff --git a/pandas/pandas/tests/indexes/datetimes/test_freq_attr.py b/pandas/pandas/tests/indexes/datetimes/test_freq_attr.py
new file mode 100644
index 0000000000000..5cddf56cd1c73
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_freq_attr.py
@@ -0,0 +1,61 @@
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    date_range,
+)
+
+from pandas.tseries.offsets import (
+    BDay,
+    DateOffset,
+    Day,
+    Hour,
+)
+
+
+class TestFreq:
+    def test_freq_setter_errors(self):
+        # GH#20678
+        idx = DatetimeIndex(["20180101", "20180103", "20180105"])
+
+        # setting with an incompatible freq
+        msg = (
+            "Inferred frequency 2D from passed values does not conform to "
+            "passed frequency 5D"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx._data.freq = "5D"
+
+        # setting with non-freq string
+        with pytest.raises(ValueError, match="Invalid frequency"):
+            idx._data.freq = "foo"
+
+    @pytest.mark.parametrize("values", [["20180101", "20180103", "20180105"], []])
+    @pytest.mark.parametrize("freq", ["2D", Day(2), "2B", BDay(2), "48h", Hour(48)])
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_freq_setter(self, values, freq, tz):
+        # GH#20678
+        idx = DatetimeIndex(values, tz=tz)
+
+        # can set to an offset, converting from string if necessary
+        idx._data.freq = freq
+        assert idx.freq == freq
+        assert isinstance(idx.freq, DateOffset)
+
+        # can reset to None
+        idx._data.freq = None
+        assert idx.freq is None
+
+    def test_freq_view_safe(self):
+        # Setting the freq for one DatetimeIndex shouldn't alter the freq
+        #  for another that views the same data
+
+        dti = date_range("2016-01-01", periods=5)
+        dta = dti._data
+
+        dti2 = DatetimeIndex(dta)._with_freq(None)
+        assert dti2.freq is None
+
+        # Original was not altered
+        assert dti.freq == "D"
+        assert dta.freq == "D"
diff --git a/pandas/pandas/tests/indexes/datetimes/test_indexing.py b/pandas/pandas/tests/indexes/datetimes/test_indexing.py
new file mode 100644
index 0000000000000..c44345273466c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_indexing.py
@@ -0,0 +1,715 @@
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+from pandas.compat.numpy import np_long
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    Timestamp,
+    bdate_range,
+    date_range,
+    notna,
+)
+import pandas._testing as tm
+
+from pandas.tseries.frequencies import to_offset
+
+START, END = datetime(2009, 1, 1), datetime(2010, 1, 1)
+
+
+class TestGetItem:
+    def test_getitem_slice_keeps_name(self):
+        # GH4226
+        st = Timestamp("2013-07-01 00:00:00", tz="America/Los_Angeles")
+        et = Timestamp("2013-07-02 00:00:00", tz="America/Los_Angeles")
+        dr = date_range(st, et, freq="h", name="timebucket")
+        assert dr[1:].name == dr.name
+
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo"])
+    def test_getitem(self, tz):
+        idx = date_range("2011-01-01", "2011-01-31", freq="D", tz=tz, name="idx")
+
+        result = idx[0]
+        assert result == Timestamp("2011-01-01", tz=idx.tz)
+
+        result = idx[0:5]
+        expected = date_range(
+            "2011-01-01", "2011-01-05", freq="D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx[0:10:2]
+        expected = date_range(
+            "2011-01-01", "2011-01-09", freq="2D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx[-20:-5:3]
+        expected = date_range(
+            "2011-01-12", "2011-01-24", freq="3D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx[4::-1]
+        expected = DatetimeIndex(
+            ["2011-01-05", "2011-01-04", "2011-01-03", "2011-01-02", "2011-01-01"],
+            dtype=idx.dtype,
+            freq="-1D",
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_dti_business_getitem(self, freq):
+        rng = bdate_range(START, END, freq=freq)
+        smaller = rng[:5]
+        exp = DatetimeIndex(rng.view(np.ndarray)[:5], freq=freq)
+        tm.assert_index_equal(smaller, exp)
+        assert smaller.freq == exp.freq
+        assert smaller.freq == rng.freq
+
+        sliced = rng[::5]
+        assert sliced.freq == to_offset(freq) * 5
+
+        fancy_indexed = rng[[4, 3, 2, 1, 0]]
+        assert len(fancy_indexed) == 5
+        assert isinstance(fancy_indexed, DatetimeIndex)
+        assert fancy_indexed.freq is None
+
+        # 32-bit vs. 64-bit platforms
+        assert rng[4] == rng[np_long(4)]
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_dti_business_getitem_matplotlib_hackaround(self, freq):
+        rng = bdate_range(START, END, freq=freq)
+        with pytest.raises(ValueError, match="Multi-dimensional indexing"):
+            # GH#30588 multi-dimensional indexing deprecated
+            rng[:, None]
+
+    def test_getitem_int_list(self):
+        dti = date_range(start="1/1/2005", end="12/1/2005", freq="ME")
+        dti2 = dti[[1, 3, 5]]
+
+        v1 = dti2[0]
+        v2 = dti2[1]
+        v3 = dti2[2]
+
+        assert v1 == Timestamp("2/28/2005")
+        assert v2 == Timestamp("4/30/2005")
+        assert v3 == Timestamp("6/30/2005")
+
+        # getitem with non-slice drops freq
+        assert dti2.freq is None
+
+
+class TestWhere:
+    def test_where_doesnt_retain_freq(self):
+        dti = date_range("20130101", periods=3, freq="D", name="idx")
+        cond = [True, True, False]
+        expected = DatetimeIndex([dti[0], dti[1], dti[0]], freq=None, name="idx")
+
+        result = dti.where(cond, dti[::-1])
+        tm.assert_index_equal(result, expected)
+
+    def test_where_other(self):
+        # other is ndarray or Index
+        i = date_range("20130101", periods=3, tz="US/Eastern")
+
+        for arr in [np.nan, pd.NaT]:
+            result = i.where(notna(i), other=arr)
+            expected = i
+            tm.assert_index_equal(result, expected)
+
+        i2 = i.copy()
+        i2 = Index([pd.NaT, pd.NaT] + i[2:].tolist())
+        result = i.where(notna(i2), i2)
+        tm.assert_index_equal(result, i2)
+
+        i2 = i.copy()
+        i2 = Index([pd.NaT, pd.NaT] + i[2:].tolist())
+        result = i.where(notna(i2), i2._values)
+        tm.assert_index_equal(result, i2)
+
+    def test_where_invalid_dtypes(self):
+        dti = date_range("20130101", periods=3, tz="US/Eastern")
+
+        tail = dti[2:].tolist()
+        i2 = Index([pd.NaT, pd.NaT] + tail)
+
+        mask = notna(i2)
+
+        # passing tz-naive ndarray to tzaware DTI
+        result = dti.where(mask, i2.values)
+        expected = Index([pd.NaT.asm8, pd.NaT.asm8] + tail, dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        # passing tz-aware DTI to tznaive DTI
+        naive = dti.tz_localize(None)
+        result = naive.where(mask, i2)
+        expected = Index([i2[0], i2[1]] + naive[2:].tolist(), dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        pi = i2.tz_localize(None).to_period("D")
+        result = dti.where(mask, pi)
+        expected = Index([pi[0], pi[1]] + tail, dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        tda = i2.asi8.view("timedelta64[ns]")
+        result = dti.where(mask, tda)
+        expected = Index([tda[0], tda[1]] + tail, dtype=object)
+        assert isinstance(expected[0], np.timedelta64)
+        tm.assert_index_equal(result, expected)
+
+        result = dti.where(mask, i2.asi8)
+        expected = Index([pd.NaT._value, pd.NaT._value] + tail, dtype=object)
+        assert isinstance(expected[0], int)
+        tm.assert_index_equal(result, expected)
+
+        # non-matching scalar
+        td = pd.Timedelta(days=4)
+        result = dti.where(mask, td)
+        expected = Index([td, td] + tail, dtype=object)
+        assert expected[0] is td
+        tm.assert_index_equal(result, expected)
+
+    def test_where_mismatched_nat(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        dti = date_range("2013-01-01", periods=3, tz=tz)
+        cond = np.array([True, False, True])
+
+        tdnat = np.timedelta64("NaT", "ns")
+        expected = Index([dti[0], tdnat, dti[2]], dtype=object)
+        assert expected[1] is tdnat
+
+        result = dti.where(cond, tdnat)
+        tm.assert_index_equal(result, expected)
+
+    def test_where_tz(self):
+        i = date_range("20130101", periods=3, tz="US/Eastern")
+        result = i.where(notna(i))
+        expected = i
+        tm.assert_index_equal(result, expected)
+
+        i2 = i.copy()
+        i2 = Index([pd.NaT, pd.NaT] + i[2:].tolist())
+        result = i.where(notna(i2))
+        expected = i2
+        tm.assert_index_equal(result, expected)
+
+
+class TestTake:
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_take_dont_lose_meta(self, tzstr):
+        rng = date_range("1/1/2000", periods=20, tz=tzstr)
+
+        result = rng.take(range(5))
+        assert result.tz == rng.tz
+        assert result.freq == rng.freq
+
+    def test_take_nan_first_datetime(self):
+        index = DatetimeIndex([pd.NaT, Timestamp("20130101"), Timestamp("20130102")])
+        result = index.take([-1, 0, 1])
+        expected = DatetimeIndex([index[-1], index[0], index[1]])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo"])
+    def test_take(self, tz):
+        # GH#10295
+        idx = date_range("2011-01-01", "2011-01-31", freq="D", name="idx", tz=tz)
+
+        result = idx.take([0])
+        assert result == Timestamp("2011-01-01", tz=idx.tz)
+
+        result = idx.take([0, 1, 2])
+        expected = date_range(
+            "2011-01-01", "2011-01-03", freq="D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx.take([0, 2, 4])
+        expected = date_range(
+            "2011-01-01", "2011-01-05", freq="2D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx.take([7, 4, 1])
+        expected = date_range(
+            "2011-01-08", "2011-01-02", freq="-3D", tz=idx.tz, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+        result = idx.take([3, 2, 5])
+        expected = DatetimeIndex(
+            ["2011-01-04", "2011-01-03", "2011-01-06"],
+            dtype=idx.dtype,
+            freq=None,
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq is None
+
+        result = idx.take([-3, 2, 5])
+        expected = DatetimeIndex(
+            ["2011-01-29", "2011-01-03", "2011-01-06"],
+            dtype=idx.dtype,
+            freq=None,
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.freq is None
+
+    def test_take_invalid_kwargs(self):
+        idx = date_range("2011-01-01", "2011-01-31", freq="D", name="idx")
+        indices = [1, 6, 5, 9, 10, 13, 15, 3]
+
+        msg = r"take\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            idx.take(indices, foo=2)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, out=indices)
+
+        msg = "the 'mode' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, mode="clip")
+
+    # TODO: This method came from test_datetime; de-dup with version above
+    @pytest.mark.parametrize("tz", [None, "US/Eastern", "Asia/Tokyo"])
+    def test_take2(self, tz):
+        dates = [
+            datetime(2010, 1, 1, 14),
+            datetime(2010, 1, 1, 15),
+            datetime(2010, 1, 1, 17),
+            datetime(2010, 1, 1, 21),
+        ]
+
+        idx = date_range(
+            start="2010-01-01 09:00",
+            end="2010-02-01 09:00",
+            freq="h",
+            tz=tz,
+            name="idx",
+        )
+        expected = DatetimeIndex(dates, freq=None, name="idx", dtype=idx.dtype)
+
+        taken1 = idx.take([5, 6, 8, 12])
+        taken2 = idx[[5, 6, 8, 12]]
+
+        for taken in [taken1, taken2]:
+            tm.assert_index_equal(taken, expected)
+            assert isinstance(taken, DatetimeIndex)
+            assert taken.freq is None
+            assert taken.tz == expected.tz
+            assert taken.name == expected.name
+
+    def test_take_fill_value(self):
+        # GH#12631
+        idx = DatetimeIndex(["2011-01-01", "2011-02-01", "2011-03-01"], name="xxx")
+        result = idx.take(np.array([1, 0, -1]))
+        expected = DatetimeIndex(["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = DatetimeIndex(["2011-02-01", "2011-01-01", "NaT"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = DatetimeIndex(["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "out of bounds"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+    def test_take_fill_value_with_timezone(self):
+        idx = DatetimeIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01"], name="xxx", tz="US/Eastern"
+        )
+        result = idx.take(np.array([1, 0, -1]))
+        expected = DatetimeIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx", tz="US/Eastern"
+        )
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = DatetimeIndex(
+            ["2011-02-01", "2011-01-01", "NaT"], name="xxx", tz="US/Eastern"
+        )
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = DatetimeIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx", tz="US/Eastern"
+        )
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "out of bounds"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+
+class TestGetLoc:
+    def test_get_loc_key_unit_mismatch(self):
+        idx = date_range("2000-01-01", periods=3)
+        key = idx[1].as_unit("ms")
+        loc = idx.get_loc(key)
+        assert loc == 1
+        assert key in idx
+
+    def test_get_loc_key_unit_mismatch_not_castable(self):
+        dta = date_range("2000-01-01", periods=3)._data.astype("M8[s]")
+        dti = DatetimeIndex(dta)
+        key = dta[0].as_unit("ns") + pd.Timedelta(1)
+
+        with pytest.raises(
+            KeyError, match=r"Timestamp\('2000-01-01 00:00:00.000000001'\)"
+        ):
+            dti.get_loc(key)
+
+        assert key not in dti
+
+    def test_get_loc_time_obj(self):
+        # time indexing
+        idx = date_range("2000-01-01", periods=24, freq="h")
+
+        result = idx.get_loc(time(12))
+        expected = np.array([12])
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+        result = idx.get_loc(time(12, 30))
+        expected = np.array([])
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize("offset", [-10, 10])
+    def test_get_loc_time_obj2(self, monkeypatch, offset):
+        # GH#8667
+        size_cutoff = 50
+        n = size_cutoff + offset
+        key = time(15, 11, 30)
+        start = key.hour * 3600 + key.minute * 60 + key.second
+        step = 24 * 3600
+
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+            idx = date_range("2014-11-26", periods=n, freq="s")
+            ts = pd.Series(np.random.default_rng(2).standard_normal(n), index=idx)
+            locs = np.arange(start, n, step, dtype=np.intp)
+
+            result = ts.index.get_loc(key)
+            tm.assert_numpy_array_equal(result, locs)
+            tm.assert_series_equal(ts[key], ts.iloc[locs])
+
+            left, right = ts.copy(), ts.copy()
+            left[key] *= -10
+            right.iloc[locs] *= -10
+            tm.assert_series_equal(left, right)
+
+    def test_get_loc_time_nat(self):
+        # GH#35114
+        # Case where key's total microseconds happens to match iNaT % 1e6 // 1000
+        tic = time(minute=12, second=43, microsecond=145224)
+        dti = DatetimeIndex([pd.NaT])
+
+        loc = dti.get_loc(tic)
+        expected = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(loc, expected)
+
+    def test_get_loc_nat(self):
+        # GH#20464
+        index = DatetimeIndex(["1/3/2000", "NaT"])
+        assert index.get_loc(pd.NaT) == 1
+
+        assert index.get_loc(None) == 1
+
+        assert index.get_loc(np.nan) == 1
+
+        assert index.get_loc(pd.NA) == 1
+
+        assert index.get_loc(np.datetime64("NaT")) == 1
+
+        with pytest.raises(KeyError, match="NaT"):
+            index.get_loc(np.timedelta64("NaT"))
+
+    @pytest.mark.parametrize("key", [pd.Timedelta(0), pd.Timedelta(1), timedelta(0)])
+    def test_get_loc_timedelta_invalid_key(self, key):
+        # GH#20464
+        dti = date_range("1970-01-01", periods=10)
+        msg = "Cannot index DatetimeIndex with [Tt]imedelta"
+        with pytest.raises(TypeError, match=msg):
+            dti.get_loc(key)
+
+    def test_get_loc_reasonable_key_error(self):
+        # GH#1062
+        index = DatetimeIndex(["1/3/2000"])
+        with pytest.raises(KeyError, match="2000"):
+            index.get_loc("1/1/2000")
+
+    def test_get_loc_year_str(self):
+        rng = date_range("1/1/2000", "1/1/2010")
+
+        result = rng.get_loc("2009")
+        expected = slice(3288, 3653)
+        assert result == expected
+
+
+class TestContains:
+    def test_dti_contains_with_duplicates(self):
+        d = datetime(2011, 12, 5, 20, 30)
+        ix = DatetimeIndex([d, d])
+        assert d in ix
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            [0, 1, 0],
+            [0, 0, -1],
+            [0, -1, -1],
+            ["2015", "2015", "2016"],
+            ["2015", "2015", "2014"],
+        ],
+    )
+    def test_contains_nonunique(self, vals):
+        # GH#9512
+        idx = DatetimeIndex(vals)
+        assert idx[0] in idx
+
+
+class TestGetIndexer:
+    def test_get_indexer_date_objs(self):
+        rng = date_range("1/1/2000", periods=20)
+
+        result = rng.get_indexer(rng.map(lambda x: x.date()))
+        expected = rng.get_indexer(rng)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer(self):
+        idx = date_range("2000-01-01", periods=3)
+        exp = np.array([0, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(idx.get_indexer(idx), exp)
+
+        target = idx[0] + pd.to_timedelta(["-1 hour", "12 hours", "1 day 1 hour"])
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "pad"), np.array([-1, 0, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "backfill"), np.array([0, 1, 2], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest"), np.array([0, 1, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest", tolerance=pd.Timedelta("1 hour")),
+            np.array([0, -1, 1], dtype=np.intp),
+        )
+        tol_raw = [
+            pd.Timedelta("1 hour"),
+            pd.Timedelta("1 hour"),
+            pd.Timedelta("1 hour").to_timedelta64(),
+        ]
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(
+                target, "nearest", tolerance=[np.timedelta64(x) for x in tol_raw]
+            ),
+            np.array([0, -1, 1], dtype=np.intp),
+        )
+        tol_bad = [
+            pd.Timedelta("2 hour").to_timedelta64(),
+            pd.Timedelta("1 hour").to_timedelta64(),
+            "foo",
+        ]
+        msg = "Could not convert 'foo' to NumPy timedelta"
+        with pytest.raises(ValueError, match=msg):
+            idx.get_indexer(target, "nearest", tolerance=tol_bad)
+        with pytest.raises(ValueError, match="abbreviation w/o a number"):
+            idx.get_indexer(idx[[0]], method="nearest", tolerance="foo")
+
+    @pytest.mark.parametrize(
+        "target",
+        [
+            [date(2020, 1, 1), Timestamp("2020-01-02")],
+            [Timestamp("2020-01-01"), date(2020, 1, 2)],
+        ],
+    )
+    def test_get_indexer_mixed_dtypes(self, target):
+        # https://github.com/pandas-dev/pandas/issues/33741
+        values = DatetimeIndex([Timestamp("2020-01-01"), Timestamp("2020-01-02")])
+        result = values.get_indexer(target)
+        expected = np.array([0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "target, positions",
+        [
+            ([date(9999, 1, 1), Timestamp("2020-01-01")], [-1, 0]),
+            ([Timestamp("2020-01-01"), date(9999, 1, 1)], [0, -1]),
+            ([date(9999, 1, 1), date(9999, 1, 1)], [-1, -1]),
+        ],
+    )
+    def test_get_indexer_out_of_bounds_date(self, target, positions):
+        values = DatetimeIndex([Timestamp("2020-01-01"), Timestamp("2020-01-02")])
+
+        result = values.get_indexer(target)
+        expected = np.array(positions, dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_pad_requires_monotonicity(self):
+        rng = date_range("1/1/2000", "3/1/2000", freq="B")
+
+        # neither monotonic increasing or decreasing
+        rng2 = rng[[1, 0, 2]]
+
+        msg = "index must be monotonic increasing or decreasing"
+        with pytest.raises(ValueError, match=msg):
+            rng2.get_indexer(rng, method="pad")
+
+
+class TestMaybeCastSliceBound:
+    def test_maybe_cast_slice_bounds_empty(self):
+        # GH#14354
+        empty_idx = date_range(freq="1h", periods=0, end="2015")
+
+        right = empty_idx._maybe_cast_slice_bound("2015-01-02", "right")
+        exp = Timestamp("2015-01-02 23:59:59.999999999")
+        assert right == exp
+
+        left = empty_idx._maybe_cast_slice_bound("2015-01-02", "left")
+        exp = Timestamp("2015-01-02 00:00:00")
+        assert left == exp
+
+    def test_maybe_cast_slice_duplicate_monotonic(self):
+        # https://github.com/pandas-dev/pandas/issues/16515
+        idx = DatetimeIndex(["2017", "2017"])
+        result = idx._maybe_cast_slice_bound("2017-01-01", "left")
+        expected = Timestamp("2017-01-01")
+        assert result == expected
+
+
+class TestGetSliceBounds:
+    @pytest.mark.parametrize("box", [date, datetime, Timestamp])
+    @pytest.mark.parametrize("side, expected", [("left", 4), ("right", 5)])
+    def test_get_slice_bounds_datetime_within(
+        self, box, side, expected, tz_aware_fixture
+    ):
+        # GH 35690
+        tz = tz_aware_fixture
+        index = bdate_range("2000-01-03", "2000-02-11").tz_localize(tz)
+        key = box(year=2000, month=1, day=7)
+
+        if tz is not None:
+            with pytest.raises(TypeError, match="Cannot compare tz-naive"):
+                # GH#36148 we require tzawareness-compat as of 2.0
+                index.get_slice_bound(key, side=side)
+        else:
+            result = index.get_slice_bound(key, side=side)
+            assert result == expected
+
+    @pytest.mark.parametrize("box", [datetime, Timestamp])
+    @pytest.mark.parametrize("side", ["left", "right"])
+    @pytest.mark.parametrize("year, expected", [(1999, 0), (2020, 30)])
+    def test_get_slice_bounds_datetime_outside(
+        self, box, side, year, expected, tz_aware_fixture
+    ):
+        # GH 35690
+        tz = tz_aware_fixture
+        index = bdate_range("2000-01-03", "2000-02-11").tz_localize(tz)
+        key = box(year=year, month=1, day=7)
+
+        if tz is not None:
+            with pytest.raises(TypeError, match="Cannot compare tz-naive"):
+                # GH#36148 we require tzawareness-compat as of 2.0
+                index.get_slice_bound(key, side=side)
+        else:
+            result = index.get_slice_bound(key, side=side)
+            assert result == expected
+
+    @pytest.mark.parametrize("box", [datetime, Timestamp])
+    def test_slice_datetime_locs(self, box, tz_aware_fixture):
+        # GH 34077
+        tz = tz_aware_fixture
+        index = DatetimeIndex(["2010-01-01", "2010-01-03"]).tz_localize(tz)
+        key = box(2010, 1, 1)
+
+        if tz is not None:
+            with pytest.raises(TypeError, match="Cannot compare tz-naive"):
+                # GH#36148 we require tzawareness-compat as of 2.0
+                index.slice_locs(key, box(2010, 1, 2))
+        else:
+            result = index.slice_locs(key, box(2010, 1, 2))
+            expected = (0, 1)
+            assert result == expected
+
+
+class TestIndexerBetweenTime:
+    def test_indexer_between_time(self):
+        # GH#11818
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        msg = r"Cannot convert arg \[datetime\.datetime\(2010, 1, 2, 1, 0\)\] to a time"
+        with pytest.raises(ValueError, match=msg):
+            rng.indexer_between_time(datetime(2010, 1, 2, 1), datetime(2010, 1, 2, 5))
+
+    @pytest.mark.parametrize("unit", ["us", "ms", "s"])
+    def test_indexer_between_time_non_nano(self, unit):
+        # For simple cases like this, the non-nano indexer_between_time
+        #  should match the nano result
+
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        arr_nano = rng._data._ndarray
+
+        arr = arr_nano.astype(f"M8[{unit}]")
+
+        dta = type(rng._data)._simple_new(arr, dtype=arr.dtype)
+        dti = DatetimeIndex(dta)
+        assert dti.dtype == arr.dtype
+
+        tic = time(1, 25)
+        toc = time(2, 29)
+
+        result = dti.indexer_between_time(tic, toc)
+        expected = rng.indexer_between_time(tic, toc)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # case with non-zero micros in arguments
+        tic = time(1, 25, 0, 45678)
+        toc = time(2, 29, 0, 1234)
+
+        result = dti.indexer_between_time(tic, toc)
+        expected = rng.indexer_between_time(tic, toc)
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_iter.py b/pandas/pandas/tests/indexes/datetimes/test_iter.py
new file mode 100644
index 0000000000000..dbd233eed908f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_iter.py
@@ -0,0 +1,76 @@
+import dateutil.tz
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    date_range,
+    to_datetime,
+)
+from pandas.core.arrays import datetimes
+
+
+class TestDatetimeIndexIteration:
+    @pytest.mark.parametrize(
+        "tz", [None, "UTC", "US/Central", dateutil.tz.tzoffset(None, -28800)]
+    )
+    def test_iteration_preserves_nanoseconds(self, tz):
+        # GH#19603
+        index = DatetimeIndex(
+            ["2018-02-08 15:00:00.168456358", "2018-02-08 15:00:00.168456359"], tz=tz
+        )
+        for i, ts in enumerate(index):
+            assert ts == index[i]
+
+    def test_iter_readonly(self):
+        # GH#28055 ints_to_pydatetime with readonly array
+        arr = np.array([np.datetime64("2012-02-15T12:00:00.000000000")])
+        arr.setflags(write=False)
+        dti = to_datetime(arr)
+        list(dti)
+
+    def test_iteration_preserves_tz(self):
+        # see GH#8890
+        index = date_range("2012-01-01", periods=3, freq="h", tz="US/Eastern")
+
+        for i, ts in enumerate(index):
+            result = ts
+            expected = index[i]
+            assert result == expected
+
+    def test_iteration_preserves_tz2(self):
+        index = date_range(
+            "2012-01-01", periods=3, freq="h", tz=dateutil.tz.tzoffset(None, -28800)
+        )
+
+        for i, ts in enumerate(index):
+            result = ts
+            expected = index[i]
+            assert result._repr_base == expected._repr_base
+            assert result == expected
+
+    def test_iteration_preserves_tz3(self):
+        # GH#9100
+        index = DatetimeIndex(
+            ["2014-12-01 03:32:39.987000-08:00", "2014-12-01 04:12:34.987000-08:00"]
+        )
+        for i, ts in enumerate(index):
+            result = ts
+            expected = index[i]
+            assert result._repr_base == expected._repr_base
+            assert result == expected
+
+    @pytest.mark.parametrize("offset", [-5, -1, 0, 1])
+    def test_iteration_over_chunksize(self, offset, monkeypatch):
+        # GH#21012
+        chunksize = 5
+        index = date_range(
+            "2000-01-01 00:00:00", periods=chunksize - offset, freq="min"
+        )
+        num = 0
+        with monkeypatch.context() as m:
+            m.setattr(datetimes, "_ITER_CHUNKSIZE", chunksize)
+            for stamp in index:
+                assert index[num] == stamp
+                num += 1
+        assert num == len(index)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_join.py b/pandas/pandas/tests/indexes/datetimes/test_join.py
new file mode 100644
index 0000000000000..abf6809d67f9c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_join.py
@@ -0,0 +1,153 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Timestamp,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import (
+    BDay,
+    BMonthEnd,
+)
+
+
+class TestJoin:
+    def test_does_not_convert_mixed_integer(self):
+        df = DataFrame(np.ones((3, 2)), columns=date_range("2020-01-01", periods=2))
+        cols = df.columns.join(df.index, how="outer")
+        joined = cols.join(df.columns)
+        assert cols.dtype == np.dtype("O")
+        assert cols.dtype == joined.dtype
+        tm.assert_numpy_array_equal(cols.values, joined.values)
+
+    def test_join_self(self, join_type):
+        index = date_range("1/1/2000", periods=10)
+        joined = index.join(index, how=join_type)
+        assert index is joined
+
+    def test_join_with_period_index(self, join_type):
+        df = DataFrame(
+            np.ones((10, 2)),
+            index=date_range("2020-01-01", periods=10),
+            columns=period_range("2020-01-01", periods=2),
+        )
+        s = df.iloc[:5, 0]
+
+        expected = df.columns.astype("O").join(s.index, how=join_type)
+        result = df.columns.join(s.index, how=join_type)
+        tm.assert_index_equal(expected, result)
+
+    def test_join_object_index(self):
+        rng = date_range("1/1/2000", periods=10)
+        idx = Index(["a", "b", "c", "d"])
+
+        result = rng.join(idx, how="outer")
+        assert isinstance(result[0], Timestamp)
+
+    def test_join_utc_convert(self, join_type):
+        rng = date_range("1/1/2011", periods=100, freq="h", tz="utc")
+
+        left = rng.tz_convert("US/Eastern")
+        right = rng.tz_convert("Europe/Berlin")
+
+        result = left.join(left[:-5], how=join_type)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz == left.tz
+
+        result = left.join(right[:-5], how=join_type)
+        assert isinstance(result, DatetimeIndex)
+        assert result.tz is timezone.utc
+
+    def test_datetimeindex_union_join_empty(self, sort, using_infer_string):
+        dti = date_range(start="1/1/2001", end="2/1/2001", freq="D")
+        empty = Index([])
+
+        result = dti.union(empty, sort=sort)
+        if using_infer_string:
+            assert isinstance(result, DatetimeIndex)
+            tm.assert_index_equal(result, dti)
+        else:
+            expected = dti.astype("O")
+            tm.assert_index_equal(result, expected)
+
+        result = dti.join(empty)
+        assert isinstance(result, DatetimeIndex)
+        tm.assert_index_equal(result, dti)
+
+    def test_join_nonunique(self):
+        idx1 = to_datetime(["2012-11-06 16:00:11.477563", "2012-11-06 16:00:11.477563"])
+        idx2 = to_datetime(["2012-11-06 15:11:09.006507", "2012-11-06 15:11:09.006507"])
+        rs = idx1.join(idx2, how="outer")
+        assert rs.is_monotonic_increasing
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_outer_join(self, freq):
+        # should just behave as union
+        start, end = datetime(2009, 1, 1), datetime(2010, 1, 1)
+        rng = date_range(start=start, end=end, freq=freq)
+
+        # overlapping
+        left = rng[:10]
+        right = rng[5:10]
+
+        the_join = left.join(right, how="outer")
+        assert isinstance(the_join, DatetimeIndex)
+
+        # non-overlapping, gap in middle
+        left = rng[:5]
+        right = rng[10:]
+
+        the_join = left.join(right, how="outer")
+        assert isinstance(the_join, DatetimeIndex)
+        assert the_join.freq is None
+
+        # non-overlapping, no gap
+        left = rng[:5]
+        right = rng[5:10]
+
+        the_join = left.join(right, how="outer")
+        assert isinstance(the_join, DatetimeIndex)
+
+        # overlapping, but different offset
+        other = date_range(start, end, freq=BMonthEnd())
+
+        the_join = rng.join(other, how="outer")
+        assert isinstance(the_join, DatetimeIndex)
+        assert the_join.freq is None
+
+    def test_naive_aware_conflicts(self):
+        start, end = datetime(2009, 1, 1), datetime(2010, 1, 1)
+        naive = date_range(start, end, freq=BDay(), tz=None)
+        aware = date_range(start, end, freq=BDay(), tz="Asia/Hong_Kong")
+
+        msg = "tz-naive.*tz-aware"
+        with pytest.raises(TypeError, match=msg):
+            naive.join(aware)
+
+        with pytest.raises(TypeError, match=msg):
+            aware.join(naive)
+
+    @pytest.mark.parametrize("tz", [None, "US/Pacific"])
+    def test_join_preserves_freq(self, tz):
+        # GH#32157
+        dti = date_range("2016-01-01", periods=10, tz=tz)
+        result = dti[:5].join(dti[5:], how="outer")
+        assert result.freq == dti.freq
+        tm.assert_index_equal(result, dti)
+
+        result = dti[:5].join(dti[6:], how="outer")
+        assert result.freq is None
+        expected = dti.delete(5)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_npfuncs.py b/pandas/pandas/tests/indexes/datetimes/test_npfuncs.py
new file mode 100644
index 0000000000000..6c3e44c2a5db1
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_npfuncs.py
@@ -0,0 +1,13 @@
+import numpy as np
+
+from pandas import date_range
+import pandas._testing as tm
+
+
+class TestSplit:
+    def test_split_non_utc(self):
+        # GH#14042
+        indices = date_range("2016-01-01 00:00:00+0200", freq="s", periods=10)
+        result = np.split(indices, indices_or_sections=[])[0]
+        expected = indices._with_freq(None)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_ops.py b/pandas/pandas/tests/indexes/datetimes/test_ops.py
new file mode 100644
index 0000000000000..bac9548b932c1
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_ops.py
@@ -0,0 +1,56 @@
+from datetime import datetime
+
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    bdate_range,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexOps:
+    def test_infer_freq(self, freq_sample):
+        # GH 11018
+        idx = date_range("2011-01-01 09:00:00", freq=freq_sample, periods=10)
+        result = DatetimeIndex(idx.asi8, freq="infer")
+        tm.assert_index_equal(idx, result)
+        assert result.freq == freq_sample
+
+
+@pytest.mark.parametrize("freq", ["B", "C"])
+class TestBusinessDatetimeIndex:
+    @pytest.fixture
+    def rng(self, freq):
+        START, END = datetime(2009, 1, 1), datetime(2010, 1, 1)
+        return bdate_range(START, END, freq=freq)
+
+    def test_comparison(self, rng):
+        d = rng[10]
+
+        comp = rng > d
+        assert comp[11]
+        assert not comp[9]
+
+    def test_copy(self, rng):
+        cp = rng.copy()
+        tm.assert_index_equal(cp, rng)
+
+    def test_identical(self, rng):
+        t1 = rng.copy()
+        t2 = rng.copy()
+        assert t1.identical(t2)
+
+        # name
+        t1 = t1.rename("foo")
+        assert t1.equals(t2)
+        assert not t1.identical(t2)
+        t2 = t2.rename("foo")
+        assert t1.identical(t2)
+
+        # freq
+        t2v = Index(t2.values)
+        assert t1.equals(t2v)
+        assert not t1.identical(t2v)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_partial_slicing.py b/pandas/pandas/tests/indexes/datetimes/test_partial_slicing.py
new file mode 100644
index 0000000000000..94175a56f1c4a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_partial_slicing.py
@@ -0,0 +1,466 @@
+"""test partial slicing on Series/Frame"""
+
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSlicing:
+    def test_string_index_series_name_converted(self):
+        # GH#1644
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            index=date_range("1/1/2000", periods=10),
+        )
+
+        result = df.loc["1/3/2000"]
+        assert result.name == df.index[2]
+
+        result = df.T["1/3/2000"]
+        assert result.name == df.index[2]
+
+    def test_stringified_slice_with_tz(self):
+        # GH#2658
+        start = "2013-01-07"
+        idx = date_range(start=start, freq="1D", periods=10, tz="US/Eastern")
+        df = DataFrame(np.arange(10), index=idx)
+        df["2013-01-14 23:44:34.437768-05:00":]  # no exception here
+
+    def test_return_type_doesnt_depend_on_monotonicity(self):
+        # GH#24892 we get Series back regardless of whether our DTI is monotonic
+        dti = date_range(start="2015-5-13 23:59:00", freq="min", periods=3)
+        ser = Series(range(3), index=dti)
+
+        # non-monotonic index
+        ser2 = Series(range(3), index=[dti[1], dti[0], dti[2]])
+
+        # key with resolution strictly lower than "min"
+        key = "2015-5-14 00"
+
+        # monotonic increasing index
+        result = ser.loc[key]
+        expected = ser.iloc[1:]
+        tm.assert_series_equal(result, expected)
+
+        # monotonic decreasing index
+        result = ser.iloc[::-1].loc[key]
+        expected = ser.iloc[::-1][:-1]
+        tm.assert_series_equal(result, expected)
+
+        # non-monotonic index
+        result2 = ser2.loc[key]
+        expected2 = ser2.iloc[::2]
+        tm.assert_series_equal(result2, expected2)
+
+    def test_return_type_doesnt_depend_on_monotonicity_higher_reso(self):
+        # GH#24892 we get Series back regardless of whether our DTI is monotonic
+        dti = date_range(start="2015-5-13 23:59:00", freq="min", periods=3)
+        ser = Series(range(3), index=dti)
+
+        # non-monotonic index
+        ser2 = Series(range(3), index=[dti[1], dti[0], dti[2]])
+
+        # key with resolution strictly *higher) than "min"
+        key = "2015-5-14 00:00:00"
+
+        # monotonic increasing index
+        result = ser.loc[key]
+        assert result == 1
+
+        # monotonic decreasing index
+        result = ser.iloc[::-1].loc[key]
+        assert result == 1
+
+        # non-monotonic index
+        result2 = ser2.loc[key]
+        assert result2 == 0
+
+    def test_monotone_DTI_indexing_bug(self):
+        # GH 19362
+        # Testing accessing the first element in a monotonic descending
+        # partial string indexing.
+
+        df = DataFrame(list(range(5)))
+        date_list = [
+            "2018-01-02",
+            "2017-02-10",
+            "2016-03-10",
+            "2015-03-15",
+            "2014-03-16",
+        ]
+        date_index = DatetimeIndex(date_list)
+        df["date"] = date_index
+        expected = DataFrame({0: list(range(5)), "date": date_index})
+        tm.assert_frame_equal(df, expected)
+
+        # We get a slice because df.index's resolution is hourly and we
+        #  are slicing with a daily-resolution string.  If both were daily,
+        #  we would get a single item back
+        dti = date_range("20170101 01:00:00", periods=3)
+        df = DataFrame({"A": [1, 2, 3]}, index=dti[::-1])
+
+        expected = DataFrame({"A": 1}, index=dti[-1:][::-1])
+        result = df.loc["2017-01-03"]
+        tm.assert_frame_equal(result, expected)
+
+        result2 = df.iloc[::-1].loc["2017-01-03"]
+        expected2 = expected.iloc[::-1]
+        tm.assert_frame_equal(result2, expected2)
+
+    def test_slice_year(self):
+        dti = date_range(freq="B", start=datetime(2005, 1, 1), periods=500)
+
+        s = Series(np.arange(len(dti)), index=dti)
+        result = s["2005"]
+        expected = s[s.index.year == 2005]
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame(np.random.default_rng(2).random((len(dti), 5)), index=dti)
+        result = df.loc["2005"]
+        expected = df[df.index.year == 2005]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "partial_dtime",
+        [
+            "2019",
+            "2019Q4",
+            "Dec 2019",
+            "2019-12-31",
+            "2019-12-31 23",
+            "2019-12-31 23:59",
+        ],
+    )
+    def test_slice_end_of_period_resolution(self, partial_dtime):
+        # GH#31064
+        dti = date_range("2019-12-31 23:59:55.999999999", periods=10, freq="s")
+
+        ser = Series(range(10), index=dti)
+        result = ser[partial_dtime]
+        expected = ser.iloc[:5]
+        tm.assert_series_equal(result, expected)
+
+    def test_slice_quarter(self):
+        dti = date_range(freq="D", start=datetime(2000, 6, 1), periods=500)
+
+        s = Series(np.arange(len(dti)), index=dti)
+        assert len(s["2001Q1"]) == 90
+
+        df = DataFrame(np.random.default_rng(2).random((len(dti), 5)), index=dti)
+        assert len(df.loc["1Q01"]) == 90
+
+    def test_slice_month(self):
+        dti = date_range(freq="D", start=datetime(2005, 1, 1), periods=500)
+        s = Series(np.arange(len(dti)), index=dti)
+        assert len(s["2005-11"]) == 30
+
+        df = DataFrame(np.random.default_rng(2).random((len(dti), 5)), index=dti)
+        assert len(df.loc["2005-11"]) == 30
+
+        tm.assert_series_equal(s["2005-11"], s["11-2005"])
+
+    def test_partial_slice(self):
+        rng = date_range(freq="D", start=datetime(2005, 1, 1), periods=500)
+        s = Series(np.arange(len(rng)), index=rng)
+
+        result = s["2005-05":"2006-02"]
+        expected = s["20050501":"20060228"]
+        tm.assert_series_equal(result, expected)
+
+        result = s["2005-05":]
+        expected = s["20050501":]
+        tm.assert_series_equal(result, expected)
+
+        result = s[:"2006-02"]
+        expected = s[:"20060228"]
+        tm.assert_series_equal(result, expected)
+
+        result = s["2005-1-1"]
+        assert result == s.iloc[0]
+
+        with pytest.raises(KeyError, match=r"^'2004-12-31'$"):
+            s["2004-12-31"]
+
+    def test_partial_slice_daily(self):
+        rng = date_range(freq="h", start=datetime(2005, 1, 31), periods=500)
+        s = Series(np.arange(len(rng)), index=rng)
+
+        result = s["2005-1-31"]
+        tm.assert_series_equal(result, s.iloc[:24])
+
+        with pytest.raises(KeyError, match=r"^'2004-12-31 00'$"):
+            s["2004-12-31 00"]
+
+    def test_partial_slice_hourly(self):
+        rng = date_range(freq="min", start=datetime(2005, 1, 1, 20, 0, 0), periods=500)
+        s = Series(np.arange(len(rng)), index=rng)
+
+        result = s["2005-1-1"]
+        tm.assert_series_equal(result, s.iloc[: 60 * 4])
+
+        result = s["2005-1-1 20"]
+        tm.assert_series_equal(result, s.iloc[:60])
+
+        assert s["2005-1-1 20:00"] == s.iloc[0]
+        with pytest.raises(KeyError, match=r"^'2004-12-31 00:15'$"):
+            s["2004-12-31 00:15"]
+
+    def test_partial_slice_minutely(self):
+        rng = date_range(freq="s", start=datetime(2005, 1, 1, 23, 59, 0), periods=500)
+        s = Series(np.arange(len(rng)), index=rng)
+
+        result = s["2005-1-1 23:59"]
+        tm.assert_series_equal(result, s.iloc[:60])
+
+        result = s["2005-1-1"]
+        tm.assert_series_equal(result, s.iloc[:60])
+
+        assert s[Timestamp("2005-1-1 23:59:00")] == s.iloc[0]
+        with pytest.raises(KeyError, match=r"^'2004-12-31 00:00:00'$"):
+            s["2004-12-31 00:00:00"]
+
+    def test_partial_slice_second_precision(self):
+        rng = date_range(
+            start=datetime(2005, 1, 1, 0, 0, 59, microsecond=999990),
+            periods=20,
+            freq="us",
+        )
+        s = Series(np.arange(20), rng)
+
+        tm.assert_series_equal(s["2005-1-1 00:00"], s.iloc[:10])
+        tm.assert_series_equal(s["2005-1-1 00:00:59"], s.iloc[:10])
+
+        tm.assert_series_equal(s["2005-1-1 00:01"], s.iloc[10:])
+        tm.assert_series_equal(s["2005-1-1 00:01:00"], s.iloc[10:])
+
+        assert s[Timestamp("2005-1-1 00:00:59.999990")] == s.iloc[0]
+        with pytest.raises(KeyError, match="2005-1-1 00:00:00"):
+            s["2005-1-1 00:00:00"]
+
+    def test_partial_slicing_dataframe(self):
+        # GH14856
+        # Test various combinations of string slicing resolution vs.
+        # index resolution
+        # - If string resolution is less precise than index resolution,
+        # string is considered a slice
+        # - If string resolution is equal to or more precise than index
+        # resolution, string is considered an exact match
+        formats = [
+            "%Y",
+            "%Y-%m",
+            "%Y-%m-%d",
+            "%Y-%m-%d %H",
+            "%Y-%m-%d %H:%M",
+            "%Y-%m-%d %H:%M:%S",
+        ]
+        resolutions = ["year", "month", "day", "hour", "minute", "second"]
+        for rnum, resolution in enumerate(resolutions[2:], 2):
+            # we check only 'day', 'hour', 'minute' and 'second'
+            unit = Timedelta("1 " + resolution)
+            middate = datetime(2012, 1, 1, 0, 0, 0)
+            index = DatetimeIndex([middate - unit, middate, middate + unit])
+            values = [1, 2, 3]
+            df = DataFrame({"a": values}, index, dtype=np.int64)
+            assert df.index.resolution == resolution
+
+            # Timestamp with the same resolution as index
+            # Should be exact match for Series (return scalar)
+            # and raise KeyError for Frame
+            for timestamp, expected in zip(index, values):
+                ts_string = timestamp.strftime(formats[rnum])
+                # make ts_string as precise as index
+                result = df["a"][ts_string]
+                assert isinstance(result, np.int64)
+                assert result == expected
+                msg = rf"^'{ts_string}'$"
+                with pytest.raises(KeyError, match=msg):
+                    df[ts_string]
+
+            # Timestamp with resolution less precise than index
+            for fmt in formats[:rnum]:
+                for element, theslice in [[0, slice(None, 1)], [1, slice(1, None)]]:
+                    ts_string = index[element].strftime(fmt)
+
+                    # Series should return slice
+                    result = df["a"][ts_string]
+                    expected = df["a"][theslice]
+                    tm.assert_series_equal(result, expected)
+
+                    # pre-2.0 df[ts_string] was overloaded to interpret this
+                    #  as slicing along index
+                    with pytest.raises(KeyError, match=ts_string):
+                        df[ts_string]
+
+            # Timestamp with resolution more precise than index
+            # Compatible with existing key
+            # Should return scalar for Series
+            # and raise KeyError for Frame
+            for fmt in formats[rnum + 1 :]:
+                ts_string = index[1].strftime(fmt)
+                result = df["a"][ts_string]
+                assert isinstance(result, np.int64)
+                assert result == 2
+                msg = rf"^'{ts_string}'$"
+                with pytest.raises(KeyError, match=msg):
+                    df[ts_string]
+
+            # Not compatible with existing key
+            # Should raise KeyError
+            for fmt, res in list(zip(formats, resolutions))[rnum + 1 :]:
+                ts = index[1] + Timedelta("1 " + res)
+                ts_string = ts.strftime(fmt)
+                msg = rf"^'{ts_string}'$"
+                with pytest.raises(KeyError, match=msg):
+                    df["a"][ts_string]
+                with pytest.raises(KeyError, match=msg):
+                    df[ts_string]
+
+    def test_partial_slicing_with_multiindex(self):
+        # GH 4758
+        # partial string indexing with a multi-index buggy
+        df = DataFrame(
+            {
+                "ACCOUNT": ["ACCT1", "ACCT1", "ACCT1", "ACCT2"],
+                "TICKER": ["ABC", "MNP", "XYZ", "XYZ"],
+                "val": [1, 2, 3, 4],
+            },
+            index=date_range("2013-06-19 09:30:00", periods=4, freq="5min"),
+        )
+        df_multi = df.set_index(["ACCOUNT", "TICKER"], append=True)
+
+        expected = DataFrame(
+            [[1]], index=Index(["ABC"], name="TICKER"), columns=["val"]
+        )
+        result = df_multi.loc[("2013-06-19 09:30:00", "ACCT1")]
+        tm.assert_frame_equal(result, expected)
+
+        expected = df_multi.loc[
+            (Timestamp("2013-06-19 09:30:00", tz=None), "ACCT1", "ABC")
+        ]
+        result = df_multi.loc[("2013-06-19 09:30:00", "ACCT1", "ABC")]
+        tm.assert_series_equal(result, expected)
+
+        # partial string indexing on first level, scalar indexing on the other two
+        result = df_multi.loc[("2013-06-19", "ACCT1", "ABC")]
+        expected = df_multi.iloc[:1].droplevel([1, 2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_partial_slicing_with_multiindex_series(self):
+        # GH 4294
+        # partial slice on a series mi
+        ser = Series(
+            range(250),
+            index=MultiIndex.from_product(
+                [date_range("2000-1-1", periods=50), range(5)]
+            ),
+        )
+
+        s2 = ser[:-1].copy()
+        expected = s2["2000-1-4"]
+        result = s2[Timestamp("2000-1-4")]
+        tm.assert_series_equal(result, expected)
+
+        result = ser[Timestamp("2000-1-4")]
+        expected = ser["2000-1-4"]
+        tm.assert_series_equal(result, expected)
+
+        df2 = DataFrame(ser)
+        expected = df2.xs("2000-1-4")
+        result = df2.loc[Timestamp("2000-1-4")]
+        tm.assert_frame_equal(result, expected)
+
+    def test_partial_slice_requires_monotonicity(self):
+        # Disallowed since 2.0 (GH 37819)
+        ser = Series(np.arange(10), date_range("2014-01-01", periods=10))
+
+        nonmonotonic = ser.iloc[[3, 5, 4]]
+        timestamp = Timestamp("2014-01-10")
+        with pytest.raises(
+            KeyError, match="Value based partial slicing on non-monotonic"
+        ):
+            nonmonotonic["2014-01-10":]
+
+        with pytest.raises(KeyError, match=r"Timestamp\('2014-01-10 00:00:00'\)"):
+            nonmonotonic[timestamp:]
+
+        with pytest.raises(
+            KeyError, match="Value based partial slicing on non-monotonic"
+        ):
+            nonmonotonic.loc["2014-01-10":]
+
+        with pytest.raises(KeyError, match=r"Timestamp\('2014-01-10 00:00:00'\)"):
+            nonmonotonic.loc[timestamp:]
+
+    def test_loc_datetime_length_one(self):
+        # GH16071
+        df = DataFrame(
+            columns=["1"],
+            index=date_range("2016-10-01T00:00:00", "2016-10-01T23:59:59"),
+        )
+        result = df.loc[datetime(2016, 10, 1) :]
+        tm.assert_frame_equal(result, df)
+
+        result = df.loc["2016-10-01T00:00:00":]
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize(
+        "start",
+        [
+            "2018-12-02 21:50:00+00:00",
+            Timestamp("2018-12-02 21:50:00+00:00"),
+            Timestamp("2018-12-02 21:50:00+00:00").to_pydatetime(),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "end",
+        [
+            "2018-12-02 21:52:00+00:00",
+            Timestamp("2018-12-02 21:52:00+00:00"),
+            Timestamp("2018-12-02 21:52:00+00:00").to_pydatetime(),
+        ],
+    )
+    def test_getitem_with_datestring_with_UTC_offset(self, start, end):
+        # GH 24076
+        idx = date_range(
+            start="2018-12-02 14:50:00-07:00",
+            end="2018-12-02 14:50:00-07:00",
+            freq="1min",
+        )
+        df = DataFrame(1, index=idx, columns=["A"])
+        result = df[start:end]
+        expected = df.iloc[0:3, :]
+        tm.assert_frame_equal(result, expected)
+
+        # GH 16785
+        start = str(start)
+        end = str(end)
+        with pytest.raises(ValueError, match="Both dates must"):
+            df[start : end[:-4] + "1:00"]
+
+        df = df.tz_localize(None)
+        with pytest.raises(ValueError, match="The index must be timezone"):
+            df[start:end]
+
+    def test_slice_reduce_to_series(self):
+        # GH 27516
+        df = DataFrame(
+            {"A": range(24)}, index=date_range("2000", periods=24, freq="ME")
+        )
+        expected = Series(
+            range(12), index=date_range("2000", periods=12, freq="ME"), name="A"
+        )
+        result = df.loc["2000", "A"]
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_pickle.py b/pandas/pandas/tests/indexes/datetimes/test_pickle.py
new file mode 100644
index 0000000000000..922b4a18119f4
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_pickle.py
@@ -0,0 +1,45 @@
+import pytest
+
+from pandas import (
+    NaT,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+class TestPickle:
+    def test_pickle(self):
+        # GH#4606
+        idx = to_datetime(["2013-01-01", NaT, "2014-01-06"])
+        idx_p = tm.round_trip_pickle(idx)
+        assert idx_p[0] == idx[0]
+        assert idx_p[1] is NaT
+        assert idx_p[2] == idx[2]
+
+    def test_pickle_dont_infer_freq(self):
+        # GH#11002
+        # don't infer freq
+        idx = date_range("1750-1-1", "2050-1-1", freq="7D")
+        idx_p = tm.round_trip_pickle(idx)
+        tm.assert_index_equal(idx, idx_p)
+
+    def test_pickle_after_set_freq(self):
+        dti = date_range("20130101", periods=3, tz="US/Eastern", name="foo")
+        dti = dti._with_freq(None)
+
+        res = tm.round_trip_pickle(dti)
+        tm.assert_index_equal(res, dti)
+
+    def test_roundtrip_pickle_with_tz(self):
+        # GH#8367
+        # round-trip of timezone
+        index = date_range("20130101", periods=3, tz="US/Eastern", name="foo")
+        unpickled = tm.round_trip_pickle(index)
+        tm.assert_index_equal(index, unpickled)
+
+    @pytest.mark.parametrize("freq", ["B", "C"])
+    def test_pickle_unpickle(self, freq):
+        rng = date_range("2009-01-01", "2010-01-01", freq=freq)
+        unpickled = tm.round_trip_pickle(rng)
+        assert unpickled.freq == freq
diff --git a/pandas/pandas/tests/indexes/datetimes/test_reindex.py b/pandas/pandas/tests/indexes/datetimes/test_reindex.py
new file mode 100644
index 0000000000000..e4911aa3c4a29
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_reindex.py
@@ -0,0 +1,56 @@
+from datetime import timedelta
+
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndexReindex:
+    def test_reindex_preserves_tz_if_target_is_empty_list_or_array(self):
+        # GH#7774
+        index = date_range("2013-01-01", periods=3, tz="US/Eastern")
+        assert str(index.reindex([])[0].tz) == "US/Eastern"
+        assert str(index.reindex(np.array([]))[0].tz) == "US/Eastern"
+
+    def test_reindex_with_same_tz_nearest(self):
+        # GH#32740
+        rng_a = date_range("2010-01-01", "2010-01-02", periods=24, tz="utc")
+        rng_b = date_range("2010-01-01", "2010-01-02", periods=23, tz="utc")
+        result1, result2 = rng_a.reindex(
+            rng_b, method="nearest", tolerance=timedelta(seconds=20)
+        )
+        expected_list1 = [
+            "2010-01-01 00:00:00",
+            "2010-01-01 01:05:27.272727272",
+            "2010-01-01 02:10:54.545454545",
+            "2010-01-01 03:16:21.818181818",
+            "2010-01-01 04:21:49.090909090",
+            "2010-01-01 05:27:16.363636363",
+            "2010-01-01 06:32:43.636363636",
+            "2010-01-01 07:38:10.909090909",
+            "2010-01-01 08:43:38.181818181",
+            "2010-01-01 09:49:05.454545454",
+            "2010-01-01 10:54:32.727272727",
+            "2010-01-01 12:00:00",
+            "2010-01-01 13:05:27.272727272",
+            "2010-01-01 14:10:54.545454545",
+            "2010-01-01 15:16:21.818181818",
+            "2010-01-01 16:21:49.090909090",
+            "2010-01-01 17:27:16.363636363",
+            "2010-01-01 18:32:43.636363636",
+            "2010-01-01 19:38:10.909090909",
+            "2010-01-01 20:43:38.181818181",
+            "2010-01-01 21:49:05.454545454",
+            "2010-01-01 22:54:32.727272727",
+            "2010-01-02 00:00:00",
+        ]
+        expected1 = DatetimeIndex(
+            expected_list1, dtype="datetime64[ns, UTC]", freq=None
+        )
+        expected2 = np.array([0] + [-1] * 21 + [23], dtype=np.dtype("intp"))
+        tm.assert_index_equal(result1, expected1)
+        tm.assert_numpy_array_equal(result2, expected2)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_scalar_compat.py b/pandas/pandas/tests/indexes/datetimes/test_scalar_compat.py
new file mode 100644
index 0000000000000..eb472b099fb1f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_scalar_compat.py
@@ -0,0 +1,451 @@
+"""
+Tests for DatetimeIndex methods behaving like their Timestamp counterparts
+"""
+
+import calendar
+from datetime import (
+    date,
+    datetime,
+    time,
+)
+import locale
+import unicodedata
+
+from hypothesis import given
+import hypothesis.strategies as st
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import timezones
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    NaT,
+    Timestamp,
+    date_range,
+    offsets,
+)
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+
+
+class TestDatetimeIndexOps:
+    def test_dti_no_millisecond_field(self):
+        msg = "type object 'DatetimeIndex' has no attribute 'millisecond'"
+        with pytest.raises(AttributeError, match=msg):
+            DatetimeIndex.millisecond
+
+        msg = "'DatetimeIndex' object has no attribute 'millisecond'"
+        with pytest.raises(AttributeError, match=msg):
+            DatetimeIndex([]).millisecond
+
+    def test_dti_time(self):
+        rng = date_range("1/1/2000", freq="12min", periods=10)
+        result = Index(rng).time
+        expected = [t.time() for t in rng]
+        assert (result == expected).all()
+
+    def test_dti_date(self):
+        rng = date_range("1/1/2000", freq="12h", periods=10)
+        result = Index(rng).date
+        expected = [t.date() for t in rng]
+        assert (result == expected).all()
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [None, "datetime64[ns, CET]", "datetime64[ns, EST]", "datetime64[ns, UTC]"],
+    )
+    def test_dti_date2(self, dtype):
+        # Regression test for GH#21230
+        expected = np.array([date(2018, 6, 4), NaT])
+
+        index = DatetimeIndex(["2018-06-04 10:00:00", NaT], dtype=dtype)
+        result = index.date
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [None, "datetime64[ns, CET]", "datetime64[ns, EST]", "datetime64[ns, UTC]"],
+    )
+    def test_dti_time2(self, dtype):
+        # Regression test for GH#21267
+        expected = np.array([time(10, 20, 30), NaT])
+
+        index = DatetimeIndex(["2018-06-04 10:20:30", NaT], dtype=dtype)
+        result = index.time
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_dti_timetz(self, tz_naive_fixture):
+        # GH#21358
+        tz = timezones.maybe_get_tz(tz_naive_fixture)
+
+        expected = np.array([time(10, 20, 30, tzinfo=tz), NaT])
+
+        index = DatetimeIndex(["2018-06-04 10:20:30", NaT], tz=tz)
+        result = index.timetz
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "field",
+        [
+            "dayofweek",
+            "day_of_week",
+            "dayofyear",
+            "day_of_year",
+            "quarter",
+            "days_in_month",
+            "is_month_start",
+            "is_month_end",
+            "is_quarter_start",
+            "is_quarter_end",
+            "is_year_start",
+            "is_year_end",
+        ],
+    )
+    def test_dti_timestamp_fields(self, field):
+        # extra fields from DatetimeIndex like quarter and week
+        idx = date_range("2020-01-01", periods=10)
+        expected = getattr(idx, field)[-1]
+
+        result = getattr(Timestamp(idx[-1]), field)
+        assert result == expected
+
+    def test_dti_nanosecond(self):
+        dti = DatetimeIndex(np.arange(10))
+        expected = Index(np.arange(10, dtype=np.int32))
+
+        tm.assert_index_equal(dti.nanosecond, expected)
+
+    @pytest.mark.parametrize("prefix", ["", "dateutil/"])
+    def test_dti_hour_tzaware(self, prefix):
+        strdates = ["1/1/2012", "3/1/2012", "4/1/2012"]
+        rng = DatetimeIndex(strdates, tz=prefix + "US/Eastern")
+        assert (rng.hour == 0).all()
+
+        # a more unusual time zone, GH#1946
+        dr = date_range(
+            "2011-10-02 00:00", freq="h", periods=10, tz=prefix + "America/Atikokan"
+        )
+
+        expected = Index(np.arange(10, dtype=np.int32))
+        tm.assert_index_equal(dr.hour, expected)
+
+    # GH#12806
+    # error: Unsupported operand types for + ("List[None]" and "List[str]")
+    @pytest.mark.parametrize(
+        "time_locale",
+        [None] + tm.get_locales(),  # type: ignore[operator]
+    )
+    def test_day_name_month_name(self, time_locale):
+        # Test Monday -> Sunday and January -> December, in that sequence
+        if time_locale is None:
+            # If the time_locale is None, day-name and month_name should
+            # return the english attributes
+            expected_days = [
+                "Monday",
+                "Tuesday",
+                "Wednesday",
+                "Thursday",
+                "Friday",
+                "Saturday",
+                "Sunday",
+            ]
+            expected_months = [
+                "January",
+                "February",
+                "March",
+                "April",
+                "May",
+                "June",
+                "July",
+                "August",
+                "September",
+                "October",
+                "November",
+                "December",
+            ]
+        else:
+            with tm.set_locale(time_locale, locale.LC_TIME):
+                expected_days = calendar.day_name[:]
+                expected_months = calendar.month_name[1:]
+
+        # GH#11128
+        dti = date_range(freq="D", start=datetime(1998, 1, 1), periods=365)
+        english_days = [
+            "Monday",
+            "Tuesday",
+            "Wednesday",
+            "Thursday",
+            "Friday",
+            "Saturday",
+            "Sunday",
+        ]
+        for day, name, eng_name in zip(range(4, 11), expected_days, english_days):
+            name = name.capitalize()
+            assert dti.day_name(locale=time_locale)[day] == name
+            assert dti.day_name(locale=None)[day] == eng_name
+            ts = Timestamp(datetime(2016, 4, day))
+            assert ts.day_name(locale=time_locale) == name
+        dti = dti.append(DatetimeIndex([NaT]))
+        assert np.isnan(dti.day_name(locale=time_locale)[-1])
+        ts = Timestamp(NaT)
+        assert np.isnan(ts.day_name(locale=time_locale))
+
+        # GH#12805
+        dti = date_range(freq="ME", start="2012", end="2013")
+        result = dti.month_name(locale=time_locale)
+        expected = Index([month.capitalize() for month in expected_months])
+
+        # work around different normalization schemes GH#22342
+        result = result.str.normalize("NFD")
+        expected = expected.str.normalize("NFD")
+
+        tm.assert_index_equal(result, expected)
+
+        for item, expected in zip(dti, expected_months):
+            result = item.month_name(locale=time_locale)
+            expected = expected.capitalize()
+
+            result = unicodedata.normalize("NFD", result)
+            expected = unicodedata.normalize("NFD", result)
+
+            assert result == expected
+        dti = dti.append(DatetimeIndex([NaT]))
+        assert np.isnan(dti.month_name(locale=time_locale)[-1])
+
+    def test_dti_week(self):
+        # GH#6538: Check that DatetimeIndex and its TimeStamp elements
+        # return the same weekofyear accessor close to new year w/ tz
+        dates = ["2013/12/29", "2013/12/30", "2013/12/31"]
+        dates = DatetimeIndex(dates, tz="Europe/Brussels")
+        expected = [52, 1, 1]
+        assert dates.isocalendar().week.tolist() == expected
+        assert [d.weekofyear for d in dates] == expected
+
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_dti_fields(self, tz):
+        # GH#13303
+        dti = date_range(freq="D", start=datetime(1998, 1, 1), periods=365, tz=tz)
+        assert dti.year[0] == 1998
+        assert dti.month[0] == 1
+        assert dti.day[0] == 1
+        assert dti.hour[0] == 0
+        assert dti.minute[0] == 0
+        assert dti.second[0] == 0
+        assert dti.microsecond[0] == 0
+        assert dti.dayofweek[0] == 3
+
+        assert dti.dayofyear[0] == 1
+        assert dti.dayofyear[120] == 121
+
+        assert dti.isocalendar().week.iloc[0] == 1
+        assert dti.isocalendar().week.iloc[120] == 18
+
+        assert dti.quarter[0] == 1
+        assert dti.quarter[120] == 2
+
+        assert dti.days_in_month[0] == 31
+        assert dti.days_in_month[90] == 30
+
+        assert dti.is_month_start[0]
+        assert not dti.is_month_start[1]
+        assert dti.is_month_start[31]
+        assert dti.is_quarter_start[0]
+        assert dti.is_quarter_start[90]
+        assert dti.is_year_start[0]
+        assert not dti.is_year_start[364]
+        assert not dti.is_month_end[0]
+        assert dti.is_month_end[30]
+        assert not dti.is_month_end[31]
+        assert dti.is_month_end[364]
+        assert not dti.is_quarter_end[0]
+        assert not dti.is_quarter_end[30]
+        assert dti.is_quarter_end[89]
+        assert dti.is_quarter_end[364]
+        assert not dti.is_year_end[0]
+        assert dti.is_year_end[364]
+
+        assert len(dti.year) == 365
+        assert len(dti.month) == 365
+        assert len(dti.day) == 365
+        assert len(dti.hour) == 365
+        assert len(dti.minute) == 365
+        assert len(dti.second) == 365
+        assert len(dti.microsecond) == 365
+        assert len(dti.dayofweek) == 365
+        assert len(dti.dayofyear) == 365
+        assert len(dti.isocalendar()) == 365
+        assert len(dti.quarter) == 365
+        assert len(dti.is_month_start) == 365
+        assert len(dti.is_month_end) == 365
+        assert len(dti.is_quarter_start) == 365
+        assert len(dti.is_quarter_end) == 365
+        assert len(dti.is_year_start) == 365
+        assert len(dti.is_year_end) == 365
+
+        dti.name = "name"
+
+        # non boolean accessors -> return Index
+        for accessor in DatetimeArray._field_ops:
+            res = getattr(dti, accessor)
+            assert len(res) == 365
+            assert isinstance(res, Index)
+            assert res.name == "name"
+
+        # boolean accessors -> return array
+        for accessor in DatetimeArray._bool_ops:
+            res = getattr(dti, accessor)
+            assert len(res) == 365
+            assert isinstance(res, np.ndarray)
+
+        # test boolean indexing
+        res = dti[dti.is_quarter_start]
+        exp = dti[[0, 90, 181, 273]]
+        tm.assert_index_equal(res, exp)
+        res = dti[dti.is_leap_year]
+        exp = DatetimeIndex([], freq="D", tz=dti.tz, name="name").as_unit("ns")
+        tm.assert_index_equal(res, exp)
+
+    def test_dti_is_year_quarter_start(self):
+        dti = date_range(freq="BQE-FEB", start=datetime(1998, 1, 1), periods=4)
+
+        assert sum(dti.is_quarter_start) == 0
+        assert sum(dti.is_quarter_end) == 4
+        assert sum(dti.is_year_start) == 0
+        assert sum(dti.is_year_end) == 1
+
+    def test_dti_is_month_start(self):
+        dti = DatetimeIndex(["2000-01-01", "2000-01-02", "2000-01-03"])
+
+        assert dti.is_month_start[0] == 1
+
+    def test_dti_is_month_start_custom(self):
+        # Ensure is_start/end accessors throw ValueError for CustomBusinessDay,
+        bday_egypt = offsets.CustomBusinessDay(weekmask="Sun Mon Tue Wed Thu")
+        dti = date_range(datetime(2013, 4, 30), periods=5, freq=bday_egypt)
+        msg = "Custom business days is not supported by is_month_start"
+        with pytest.raises(ValueError, match=msg):
+            dti.is_month_start
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "MS", 3, np.array([False, True, False])),
+            ("2017-12-01", "QS", 3, np.array([True, False, False])),
+            ("2017-12-01", "YS", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_year_start(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_year_start
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "ME", 3, np.array([True, False, False])),
+            ("2017-12-01", "QE", 3, np.array([True, False, False])),
+            ("2017-12-01", "YE", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_year_end(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_year_end
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "MS", 3, np.array([False, True, False])),
+            ("2017-12-01", "QS", 3, np.array([True, True, True])),
+            ("2017-12-01", "YS", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_quarter_start(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_quarter_start
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "ME", 3, np.array([True, False, False])),
+            ("2017-12-01", "QE", 3, np.array([True, True, True])),
+            ("2017-12-01", "YE", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_quarter_end(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_quarter_end
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "MS", 3, np.array([True, True, True])),
+            ("2017-12-01", "QS", 3, np.array([True, True, True])),
+            ("2017-12-01", "YS", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_month_start(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_month_start
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, periods, expected_values",
+        [
+            ("2017-12-01", "ME", 3, np.array([True, True, True])),
+            ("2017-12-01", "QE", 3, np.array([True, True, True])),
+            ("2017-12-01", "YE", 3, np.array([True, True, True])),
+        ],
+    )
+    def test_dti_dr_is_month_end(self, timestamp, freq, periods, expected_values):
+        # GH57377
+        result = date_range(timestamp, freq=freq, periods=periods).is_month_end
+        tm.assert_numpy_array_equal(result, expected_values)
+
+    def test_dti_is_year_quarter_start_doubledigit_freq(self):
+        # GH#58523
+        dr = date_range("2017-01-01", periods=2, freq="10YS")
+        assert all(dr.is_year_start)
+
+        dr = date_range("2017-01-01", periods=2, freq="10QS")
+        assert all(dr.is_quarter_start)
+
+    def test_dti_is_year_start_freq_custom_business_day_with_digit(self):
+        # GH#58664
+        dr = date_range("2020-01-01", periods=2, freq="2C")
+        msg = "Custom business days is not supported by is_year_start"
+        with pytest.raises(ValueError, match=msg):
+            dr.is_year_start
+
+    @pytest.mark.parametrize("freq", ["3BMS", offsets.BusinessMonthBegin(3)])
+    def test_dti_is_year_quarter_start_freq_business_month_begin(self, freq):
+        # GH#58729
+        dr = date_range("2020-01-01", periods=5, freq=freq)
+        result = [x.is_year_start for x in dr]
+        assert result == [True, False, False, False, True]
+
+        dr = date_range("2020-01-01", periods=4, freq=freq)
+        result = [x.is_quarter_start for x in dr]
+        assert all(dr.is_quarter_start)
+
+
+@given(
+    dt=st.datetimes(min_value=datetime(1960, 1, 1), max_value=datetime(1980, 1, 1)),
+    n=st.integers(min_value=1, max_value=10),
+    freq=st.sampled_from(["MS", "QS", "YS"]),
+)
+@pytest.mark.slow
+def test_against_scalar_parametric(freq, dt, n):
+    # https://github.com/pandas-dev/pandas/issues/49606
+    freq = f"{n}{freq}"
+    d = date_range(dt, periods=3, freq=freq)
+    result = list(d.is_year_start)
+    expected = [x.is_year_start for x in d]
+    assert result == expected
diff --git a/pandas/pandas/tests/indexes/datetimes/test_setops.py b/pandas/pandas/tests/indexes/datetimes/test_setops.py
new file mode 100644
index 0000000000000..7ef6efad0ff6f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_setops.py
@@ -0,0 +1,696 @@
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Series,
+    Timestamp,
+    bdate_range,
+    date_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import (
+    BMonthEnd,
+    Minute,
+    MonthEnd,
+)
+
+START, END = datetime(2009, 1, 1), datetime(2010, 1, 1)
+
+
+class TestDatetimeIndexSetOps:
+    tz = [
+        None,
+        "UTC",
+        "Asia/Tokyo",
+        "US/Eastern",
+        "dateutil/Asia/Singapore",
+        "dateutil/US/Pacific",
+    ]
+
+    # TODO: moved from test_datetimelike; dedup with version below
+    def test_union2(self, sort):
+        everything = date_range("2020-01-01", periods=10)
+        first = everything[:5]
+        second = everything[5:]
+        union = first.union(second, sort=sort)
+        tm.assert_index_equal(union, everything)
+
+    @pytest.mark.parametrize("box", [np.array, Series, list])
+    def test_union3(self, sort, box):
+        everything = date_range("2020-01-01", periods=10)
+        first = everything[:5]
+        second = everything[5:]
+
+        # GH 10149 support listlike inputs other than Index objects
+        expected = first.union(second, sort=sort)
+        case = box(second.values)
+        result = first.union(case, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", tz)
+    def test_union(self, tz, sort):
+        rng1 = date_range("1/1/2000", freq="D", periods=5, tz=tz)
+        other1 = date_range("1/6/2000", freq="D", periods=5, tz=tz)
+        expected1 = date_range("1/1/2000", freq="D", periods=10, tz=tz)
+        expected1_notsorted = DatetimeIndex(list(other1) + list(rng1))
+
+        rng2 = date_range("1/1/2000", freq="D", periods=5, tz=tz)
+        other2 = date_range("1/4/2000", freq="D", periods=5, tz=tz)
+        expected2 = date_range("1/1/2000", freq="D", periods=8, tz=tz)
+        expected2_notsorted = DatetimeIndex(list(other2) + list(rng2[:3]))
+
+        rng3 = date_range("1/1/2000", freq="D", periods=5, tz=tz)
+        other3 = DatetimeIndex([], tz=tz).as_unit("ns")
+        expected3 = date_range("1/1/2000", freq="D", periods=5, tz=tz)
+        expected3_notsorted = rng3
+
+        for rng, other, exp, exp_notsorted in [
+            (rng1, other1, expected1, expected1_notsorted),
+            (rng2, other2, expected2, expected2_notsorted),
+            (rng3, other3, expected3, expected3_notsorted),
+        ]:
+            result_union = rng.union(other, sort=sort)
+            tm.assert_index_equal(result_union, exp)
+
+            result_union = other.union(rng, sort=sort)
+            if sort is None:
+                tm.assert_index_equal(result_union, exp)
+            else:
+                tm.assert_index_equal(result_union, exp_notsorted)
+
+    def test_union_coverage(self, sort):
+        idx = DatetimeIndex(["2000-01-03", "2000-01-01", "2000-01-02"])
+        ordered = DatetimeIndex(idx.sort_values(), freq="infer")
+        result = ordered.union(idx, sort=sort)
+        tm.assert_index_equal(result, ordered)
+
+        result = ordered[:0].union(ordered, sort=sort)
+        tm.assert_index_equal(result, ordered)
+        assert result.freq == ordered.freq
+
+    def test_union_bug_1730(self, sort):
+        rng_a = date_range("1/1/2012", periods=4, freq="3h")
+        rng_b = date_range("1/1/2012", periods=4, freq="4h")
+
+        result = rng_a.union(rng_b, sort=sort)
+        exp = list(rng_a) + list(rng_b[1:])
+        if sort is None:
+            exp = DatetimeIndex(sorted(exp))
+        else:
+            exp = DatetimeIndex(exp)
+        tm.assert_index_equal(result, exp)
+
+    def test_union_bug_1745(self, sort):
+        left = DatetimeIndex(["2012-05-11 15:19:49.695000"])
+        right = DatetimeIndex(
+            [
+                "2012-05-29 13:04:21.322000",
+                "2012-05-11 15:27:24.873000",
+                "2012-05-11 15:31:05.350000",
+            ]
+        )
+
+        result = left.union(right, sort=sort)
+        exp = DatetimeIndex(
+            [
+                "2012-05-11 15:19:49.695000",
+                "2012-05-29 13:04:21.322000",
+                "2012-05-11 15:27:24.873000",
+                "2012-05-11 15:31:05.350000",
+            ]
+        )
+        if sort is None:
+            exp = exp.sort_values()
+        tm.assert_index_equal(result, exp)
+
+    def test_union_bug_4564(self, sort):
+        from pandas import DateOffset
+
+        left = date_range("2013-01-01", "2013-02-01")
+        right = left + DateOffset(minutes=15)
+
+        result = left.union(right, sort=sort)
+        exp = list(left) + list(right)
+        if sort is None:
+            exp = DatetimeIndex(sorted(exp))
+        else:
+            exp = DatetimeIndex(exp)
+        tm.assert_index_equal(result, exp)
+
+    def test_union_freq_both_none(self, sort):
+        # GH11086
+        expected = bdate_range("20150101", periods=10)
+        expected._data.freq = None
+
+        result = expected.union(expected, sort=sort)
+        tm.assert_index_equal(result, expected)
+        assert result.freq is None
+
+    def test_union_freq_infer(self):
+        # When taking the union of two DatetimeIndexes, we infer
+        #  a freq even if the arguments don't have freq.  This matches
+        #  TimedeltaIndex behavior.
+        dti = date_range("2016-01-01", periods=5)
+        left = dti[[0, 1, 3, 4]]
+        right = dti[[2, 3, 1]]
+
+        assert left.freq is None
+        assert right.freq is None
+
+        result = left.union(right)
+        tm.assert_index_equal(result, dti)
+        assert result.freq == "D"
+
+    def test_union_dataframe_index(self):
+        rng1 = date_range("1/1/1999", "1/1/2012", freq="MS")
+        s1 = Series(np.random.default_rng(2).standard_normal(len(rng1)), rng1)
+
+        rng2 = date_range("1/1/1980", "12/1/2001", freq="MS")
+        s2 = Series(np.random.default_rng(2).standard_normal(len(rng2)), rng2)
+        df = DataFrame({"s1": s1, "s2": s2})
+
+        exp = date_range("1/1/1980", "1/1/2012", freq="MS")
+        tm.assert_index_equal(df.index, exp)
+
+    def test_union_with_DatetimeIndex(self, sort):
+        i1 = Index(np.arange(0, 20, 2, dtype=np.int64))
+        i2 = date_range(start="2012-01-03 00:00:00", periods=10, freq="D")
+        # Works
+        i1.union(i2, sort=sort)
+        # Fails with "AttributeError: can't set attribute"
+        i2.union(i1, sort=sort)
+
+    def test_union_same_timezone_different_units(self):
+        # GH 55238
+        idx1 = date_range("2000-01-01", periods=3, tz="UTC").as_unit("ms")
+        idx2 = date_range("2000-01-01", periods=3, tz="UTC").as_unit("us")
+        result = idx1.union(idx2)
+        expected = date_range("2000-01-01", periods=3, tz="UTC").as_unit("us")
+        tm.assert_index_equal(result, expected)
+
+    # TODO: moved from test_datetimelike; de-duplicate with version below
+    def test_intersection2(self):
+        first = date_range("2020-01-01", periods=10)
+        second = first[5:]
+        intersect = first.intersection(second)
+        tm.assert_index_equal(intersect, second)
+
+        # GH 10149
+        cases = [klass(second.values) for klass in [np.array, Series, list]]
+        for case in cases:
+            result = first.intersection(case)
+            tm.assert_index_equal(result, second)
+
+        third = Index(["a", "b", "c"])
+        result = first.intersection(third)
+        expected = Index([], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tz", [None, "Asia/Tokyo", "US/Eastern", "dateutil/US/Pacific"]
+    )
+    def test_intersection(self, tz, sort):
+        # GH 4690 (with tz)
+        base = date_range("6/1/2000", "6/30/2000", freq="D", name="idx")
+
+        # if target has the same name, it is preserved
+        rng2 = date_range("5/15/2000", "6/20/2000", freq="D", name="idx")
+        expected2 = date_range("6/1/2000", "6/20/2000", freq="D", name="idx")
+
+        # if target name is different, it will be reset
+        rng3 = date_range("5/15/2000", "6/20/2000", freq="D", name="other")
+        expected3 = date_range("6/1/2000", "6/20/2000", freq="D", name=None)
+
+        rng4 = date_range("7/1/2000", "7/31/2000", freq="D", name="idx")
+        expected4 = DatetimeIndex([], freq="D", name="idx", dtype="M8[ns]")
+
+        for rng, expected in [
+            (rng2, expected2),
+            (rng3, expected3),
+            (rng4, expected4),
+        ]:
+            result = base.intersection(rng)
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+        # non-monotonic
+        base = DatetimeIndex(
+            ["2011-01-05", "2011-01-04", "2011-01-02", "2011-01-03"], tz=tz, name="idx"
+        ).as_unit("ns")
+
+        rng2 = DatetimeIndex(
+            ["2011-01-04", "2011-01-02", "2011-02-02", "2011-02-03"], tz=tz, name="idx"
+        ).as_unit("ns")
+        expected2 = DatetimeIndex(
+            ["2011-01-04", "2011-01-02"], tz=tz, name="idx"
+        ).as_unit("ns")
+
+        rng3 = DatetimeIndex(
+            ["2011-01-04", "2011-01-02", "2011-02-02", "2011-02-03"],
+            tz=tz,
+            name="other",
+        ).as_unit("ns")
+        expected3 = DatetimeIndex(
+            ["2011-01-04", "2011-01-02"], tz=tz, name=None
+        ).as_unit("ns")
+
+        # GH 7880
+        rng4 = date_range("7/1/2000", "7/31/2000", freq="D", tz=tz, name="idx")
+        expected4 = DatetimeIndex([], tz=tz, name="idx").as_unit("ns")
+        assert expected4.freq is None
+
+        for rng, expected in [
+            (rng2, expected2),
+            (rng3, expected3),
+            (rng4, expected4),
+        ]:
+            result = base.intersection(rng, sort=sort)
+            if sort is None:
+                expected = expected.sort_values()
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+    # parametrize over both anchored and non-anchored freqs, as they
+    #  have different code paths
+    @pytest.mark.parametrize("freq", ["min", "B"])
+    def test_intersection_empty(self, tz_aware_fixture, freq):
+        # empty same freq GH2129
+        tz = tz_aware_fixture
+        rng = date_range("6/1/2000", "6/15/2000", freq=freq, tz=tz)
+        result = rng[0:0].intersection(rng)
+        assert len(result) == 0
+        assert result.freq == rng.freq
+
+        result = rng.intersection(rng[0:0])
+        assert len(result) == 0
+        assert result.freq == rng.freq
+
+        # no overlap GH#33604
+        check_freq = freq != "min"  # We don't preserve freq on non-anchored offsets
+        result = rng[:3].intersection(rng[-3:])
+        tm.assert_index_equal(result, rng[:0])
+        if check_freq:
+            # We don't preserve freq on non-anchored offsets
+            assert result.freq == rng.freq
+
+        # swapped left and right
+        result = rng[-3:].intersection(rng[:3])
+        tm.assert_index_equal(result, rng[:0])
+        if check_freq:
+            # We don't preserve freq on non-anchored offsets
+            assert result.freq == rng.freq
+
+    def test_intersection_bug_1708(self):
+        from pandas import DateOffset
+
+        index_1 = date_range("1/1/2012", periods=4, freq="12h")
+        index_2 = index_1 + DateOffset(hours=1)
+
+        result = index_1.intersection(index_2)
+        assert len(result) == 0
+
+    @pytest.mark.parametrize("tz", tz)
+    def test_difference(self, tz, sort):
+        rng_dates = ["1/2/2000", "1/3/2000", "1/1/2000", "1/4/2000", "1/5/2000"]
+
+        rng1 = DatetimeIndex(rng_dates, tz=tz)
+        other1 = date_range("1/6/2000", freq="D", periods=5, tz=tz)
+        expected1 = DatetimeIndex(rng_dates, tz=tz)
+
+        rng2 = DatetimeIndex(rng_dates, tz=tz)
+        other2 = date_range("1/4/2000", freq="D", periods=5, tz=tz)
+        expected2 = DatetimeIndex(rng_dates[:3], tz=tz)
+
+        rng3 = DatetimeIndex(rng_dates, tz=tz)
+        other3 = DatetimeIndex([], tz=tz)
+        expected3 = DatetimeIndex(rng_dates, tz=tz)
+
+        for rng, other, expected in [
+            (rng1, other1, expected1),
+            (rng2, other2, expected2),
+            (rng3, other3, expected3),
+        ]:
+            result_diff = rng.difference(other, sort)
+            if sort is None and len(other):
+                # We dont sort (yet?) when empty GH#24959
+                expected = expected.sort_values()
+            tm.assert_index_equal(result_diff, expected)
+
+    def test_difference_freq(self, sort):
+        # GH14323: difference of DatetimeIndex should not preserve frequency
+
+        index = date_range("20160920", "20160925", freq="D")
+        other = date_range("20160921", "20160924", freq="D")
+        expected = DatetimeIndex(["20160920", "20160925"], dtype="M8[ns]", freq=None)
+        idx_diff = index.difference(other, sort)
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+        # preserve frequency when the difference is a contiguous
+        # subset of the original range
+        other = date_range("20160922", "20160925", freq="D")
+        idx_diff = index.difference(other, sort)
+        expected = DatetimeIndex(["20160920", "20160921"], dtype="M8[ns]", freq="D")
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+    def test_datetimeindex_diff(self, sort):
+        dti1 = date_range(freq="QE-JAN", start=datetime(1997, 12, 31), periods=100)
+        dti2 = date_range(freq="QE-JAN", start=datetime(1997, 12, 31), periods=98)
+        assert len(dti1.difference(dti2, sort)) == 2
+
+    @pytest.mark.parametrize("tz", [None, "Asia/Tokyo", "US/Eastern"])
+    def test_setops_preserve_freq(self, tz):
+        rng = date_range("1/1/2000", "1/1/2002", name="idx", tz=tz)
+
+        result = rng[:50].union(rng[50:100])
+        assert result.name == rng.name
+        assert result.freq == rng.freq
+        assert result.tz == rng.tz
+
+        result = rng[:50].union(rng[30:100])
+        assert result.name == rng.name
+        assert result.freq == rng.freq
+        assert result.tz == rng.tz
+
+        result = rng[:50].union(rng[60:100])
+        assert result.name == rng.name
+        assert result.freq is None
+        assert result.tz == rng.tz
+
+        result = rng[:50].intersection(rng[25:75])
+        assert result.name == rng.name
+        assert result.freqstr == "D"
+        assert result.tz == rng.tz
+
+        nofreq = DatetimeIndex(list(rng[25:75]), name="other")
+        result = rng[:50].union(nofreq)
+        assert result.name is None
+        assert result.freq == rng.freq
+        assert result.tz == rng.tz
+
+        result = rng[:50].intersection(nofreq)
+        assert result.name is None
+        assert result.freq == rng.freq
+        assert result.tz == rng.tz
+
+    def test_intersection_non_tick_no_fastpath(self):
+        # GH#42104
+        dti = DatetimeIndex(
+            [
+                "2018-12-31",
+                "2019-03-31",
+                "2019-06-30",
+                "2019-09-30",
+                "2019-12-31",
+                "2020-03-31",
+            ],
+            freq="QE-DEC",
+        )
+        result = dti[::2].intersection(dti[1::2])
+        expected = dti[:0]
+        tm.assert_index_equal(result, expected)
+
+    def test_dti_intersection(self):
+        rng = date_range("1/1/2011", periods=100, freq="h", tz="utc")
+
+        left = rng[10:90][::-1]
+        right = rng[20:80][::-1]
+
+        assert left.tz == rng.tz
+        result = left.intersection(right)
+        assert result.tz == left.tz
+
+    # Note: not difference, as there is no symmetry requirement there
+    @pytest.mark.parametrize("setop", ["union", "intersection", "symmetric_difference"])
+    def test_dti_setop_aware(self, setop):
+        # non-overlapping
+        # GH#39328 as of 2.0 we cast these to UTC instead of object
+        rng = date_range("2012-11-15 00:00:00", periods=6, freq="h", tz="US/Central")
+
+        rng2 = date_range("2012-11-15 12:00:00", periods=6, freq="h", tz="US/Eastern")
+
+        result = getattr(rng, setop)(rng2)
+
+        left = rng.tz_convert("UTC")
+        right = rng2.tz_convert("UTC")
+        expected = getattr(left, setop)(right)
+        tm.assert_index_equal(result, expected)
+        assert result.tz == left.tz
+        if len(result):
+            assert result[0].tz is timezone.utc
+            assert result[-1].tz is timezone.utc
+
+    def test_dti_union_mixed(self):
+        # GH#21671
+        rng = DatetimeIndex([Timestamp("2011-01-01"), pd.NaT])
+        rng2 = DatetimeIndex(["2012-01-01", "2012-01-02"], tz="Asia/Tokyo")
+        result = rng.union(rng2)
+        expected = Index(
+            [
+                Timestamp("2011-01-01"),
+                pd.NaT,
+                Timestamp("2012-01-01", tz="Asia/Tokyo"),
+                Timestamp("2012-01-02", tz="Asia/Tokyo"),
+            ],
+            dtype=object,
+        )
+        tm.assert_index_equal(result, expected)
+
+
+class TestBusinessDatetimeIndex:
+    def test_union(self, sort):
+        rng = bdate_range(START, END)
+        # overlapping
+        left = rng[:10]
+        right = rng[5:10]
+
+        the_union = left.union(right, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+        # non-overlapping, gap in middle
+        left = rng[:5]
+        right = rng[10:]
+
+        the_union = left.union(right, sort=sort)
+        assert isinstance(the_union, Index)
+
+        # non-overlapping, no gap
+        left = rng[:5]
+        right = rng[5:10]
+
+        the_union = left.union(right, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+        # order does not matter
+        if sort is None:
+            tm.assert_index_equal(right.union(left, sort=sort), the_union)
+        else:
+            expected = DatetimeIndex(list(right) + list(left))
+            tm.assert_index_equal(right.union(left, sort=sort), expected)
+
+        # overlapping, but different offset
+        rng = date_range(START, END, freq=BMonthEnd())
+
+        the_union = rng.union(rng, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+    def test_union_not_cacheable(self, sort):
+        rng = date_range("1/1/2000", periods=50, freq=Minute())
+        rng1 = rng[10:]
+        rng2 = rng[:25]
+        the_union = rng1.union(rng2, sort=sort)
+        if sort is None:
+            tm.assert_index_equal(the_union, rng)
+        else:
+            expected = DatetimeIndex(list(rng[10:]) + list(rng[:10]))
+            tm.assert_index_equal(the_union, expected)
+
+        rng1 = rng[10:]
+        rng2 = rng[15:35]
+        the_union = rng1.union(rng2, sort=sort)
+        expected = rng[10:]
+        tm.assert_index_equal(the_union, expected)
+
+    def test_intersection(self):
+        rng = date_range("1/1/2000", periods=50, freq=Minute())
+        rng1 = rng[10:]
+        rng2 = rng[:25]
+        the_int = rng1.intersection(rng2)
+        expected = rng[10:25]
+        tm.assert_index_equal(the_int, expected)
+        assert isinstance(the_int, DatetimeIndex)
+        assert the_int.freq == rng.freq
+
+        the_int = rng1.intersection(rng2)
+        tm.assert_index_equal(the_int, expected)
+
+        # non-overlapping
+        the_int = rng[:10].intersection(rng[10:])
+        expected = DatetimeIndex([]).as_unit("ns")
+        tm.assert_index_equal(the_int, expected)
+
+    def test_intersection_bug(self):
+        # GH #771
+        a = bdate_range("11/30/2011", "12/31/2011")
+        b = bdate_range("12/10/2011", "12/20/2011")
+        result = a.intersection(b)
+        tm.assert_index_equal(result, b)
+        assert result.freq == b.freq
+
+    def test_intersection_list(self):
+        # GH#35876
+        # values is not an Index -> no name -> retain "a"
+        values = [Timestamp("2020-01-01"), Timestamp("2020-02-01")]
+        idx = DatetimeIndex(values, name="a")
+        res = idx.intersection(values)
+        tm.assert_index_equal(res, idx)
+
+    def test_month_range_union_tz_pytz(self, sort):
+        pytz = pytest.importorskip("pytz")
+        tz = pytz.timezone("US/Eastern")
+
+        early_start = datetime(2011, 1, 1)
+        early_end = datetime(2011, 3, 1)
+
+        late_start = datetime(2011, 3, 1)
+        late_end = datetime(2011, 5, 1)
+
+        early_dr = date_range(start=early_start, end=early_end, tz=tz, freq=MonthEnd())
+        late_dr = date_range(start=late_start, end=late_end, tz=tz, freq=MonthEnd())
+
+        early_dr.union(late_dr, sort=sort)
+
+    @td.skip_if_windows
+    def test_month_range_union_tz_dateutil(self, sort):
+        from pandas._libs.tslibs.timezones import dateutil_gettz
+
+        tz = dateutil_gettz("US/Eastern")
+
+        early_start = datetime(2011, 1, 1)
+        early_end = datetime(2011, 3, 1)
+
+        late_start = datetime(2011, 3, 1)
+        late_end = datetime(2011, 5, 1)
+
+        early_dr = date_range(start=early_start, end=early_end, tz=tz, freq=MonthEnd())
+        late_dr = date_range(start=late_start, end=late_end, tz=tz, freq=MonthEnd())
+
+        early_dr.union(late_dr, sort=sort)
+
+    @pytest.mark.parametrize("sort", [False, None])
+    def test_intersection_duplicates(self, sort):
+        # GH#38196
+        idx1 = Index(
+            [
+                Timestamp("2019-12-13"),
+                Timestamp("2019-12-12"),
+                Timestamp("2019-12-12"),
+            ]
+        )
+        result = idx1.intersection(idx1, sort=sort)
+        expected = Index([Timestamp("2019-12-13"), Timestamp("2019-12-12")])
+        tm.assert_index_equal(result, expected)
+
+
+class TestCustomDatetimeIndex:
+    def test_union(self, sort):
+        # overlapping
+        rng = bdate_range(START, END, freq="C")
+        left = rng[:10]
+        right = rng[5:10]
+
+        the_union = left.union(right, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+        # non-overlapping, gap in middle
+        left = rng[:5]
+        right = rng[10:]
+
+        the_union = left.union(right, sort)
+        assert isinstance(the_union, Index)
+
+        # non-overlapping, no gap
+        left = rng[:5]
+        right = rng[5:10]
+
+        the_union = left.union(right, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+        # order does not matter
+        if sort is None:
+            tm.assert_index_equal(right.union(left, sort=sort), the_union)
+
+        # overlapping, but different offset
+        rng = date_range(START, END, freq=BMonthEnd())
+
+        the_union = rng.union(rng, sort=sort)
+        assert isinstance(the_union, DatetimeIndex)
+
+    def test_intersection_bug(self):
+        # GH #771
+        a = bdate_range("11/30/2011", "12/31/2011", freq="C")
+        b = bdate_range("12/10/2011", "12/20/2011", freq="C")
+        result = a.intersection(b)
+        tm.assert_index_equal(result, b)
+        assert result.freq == b.freq
+
+    @pytest.mark.parametrize(
+        "tz", [None, "UTC", "Europe/Berlin", timezone(timedelta(hours=-1))]
+    )
+    def test_intersection_dst_transition(self, tz):
+        # GH 46702: Europe/Berlin has DST transition
+        idx1 = date_range("2020-03-27", periods=5, freq="D", tz=tz)
+        idx2 = date_range("2020-03-30", periods=5, freq="D", tz=tz)
+        result = idx1.intersection(idx2)
+        expected = date_range("2020-03-30", periods=2, freq="D", tz=tz)
+        tm.assert_index_equal(result, expected)
+
+        # GH#45863 same problem for union
+        index1 = date_range("2021-10-28", periods=3, freq="D", tz="Europe/London")
+        index2 = date_range("2021-10-30", periods=4, freq="D", tz="Europe/London")
+        result = index1.union(index2)
+        expected = date_range("2021-10-28", periods=6, freq="D", tz="Europe/London")
+        tm.assert_index_equal(result, expected)
+
+
+def test_union_non_nano_rangelike():
+    # GH 59036
+    l1 = DatetimeIndex(
+        ["2024-05-11", "2024-05-12"], dtype="datetime64[us]", name="Date", freq="D"
+    )
+    l2 = DatetimeIndex(["2024-05-13"], dtype="datetime64[us]", name="Date", freq="D")
+    result = l1.union(l2)
+    expected = DatetimeIndex(
+        ["2024-05-11", "2024-05-12", "2024-05-13"],
+        dtype="datetime64[us]",
+        name="Date",
+        freq="D",
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_intersection_non_nano_rangelike():
+    # GH 59271
+    l1 = date_range("2024-01-01", "2024-01-03", unit="s")
+    l2 = date_range("2024-01-02", "2024-01-04", unit="s")
+    result = l1.intersection(l2)
+    expected = DatetimeIndex(
+        ["2024-01-02", "2024-01-03"],
+        dtype="datetime64[s]",
+        freq="D",
+    )
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/datetimes/test_timezones.py b/pandas/pandas/tests/indexes/datetimes/test_timezones.py
new file mode 100644
index 0000000000000..8d9340818b511
--- /dev/null
+++ b/pandas/pandas/tests/indexes/datetimes/test_timezones.py
@@ -0,0 +1,256 @@
+"""
+Tests for DatetimeIndex timezone-related methods
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+    tzinfo,
+)
+import zoneinfo
+
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    conversion,
+    timezones,
+)
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Timestamp,
+    bdate_range,
+    date_range,
+    isna,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+class FixedOffset(tzinfo):
+    """Fixed offset in minutes east from UTC."""
+
+    def __init__(self, offset, name) -> None:
+        self.__offset = timedelta(minutes=offset)
+        self.__name = name
+
+    def utcoffset(self, dt):
+        return self.__offset
+
+    def tzname(self, dt):
+        return self.__name
+
+    def dst(self, dt):
+        return timedelta(0)
+
+
+fixed_off_no_name = FixedOffset(-330, None)
+
+
+class TestDatetimeIndexTimezones:
+    # -------------------------------------------------------------
+    # Unsorted
+
+    def test_dti_drop_dont_lose_tz(self):
+        # GH#2621
+        ind = date_range("2012-12-01", periods=10, tz="utc")
+        ind = ind.drop(ind[-1])
+
+        assert ind.tz is not None
+
+    def test_dti_tz_conversion_freq(self, tz_naive_fixture):
+        # GH25241
+        t3 = DatetimeIndex(["2019-01-01 10:00"], freq="h")
+        assert t3.tz_localize(tz=tz_naive_fixture).freq == t3.freq
+        t4 = DatetimeIndex(["2019-01-02 12:00"], tz="UTC", freq="min")
+        assert t4.tz_convert(tz="UTC").freq == t4.freq
+
+    def test_drop_dst_boundary(self):
+        # see gh-18031
+        tz = "Europe/Brussels"
+        freq = "15min"
+
+        start = Timestamp("201710290100", tz=tz)
+        end = Timestamp("201710290300", tz=tz)
+        index = date_range(start=start, end=end, freq=freq)
+
+        expected = DatetimeIndex(
+            [
+                "201710290115",
+                "201710290130",
+                "201710290145",
+                "201710290200",
+                "201710290215",
+                "201710290230",
+                "201710290245",
+                "201710290200",
+                "201710290215",
+                "201710290230",
+                "201710290245",
+                "201710290300",
+            ],
+            dtype="M8[ns, Europe/Brussels]",
+            freq=freq,
+            ambiguous=[
+                True,
+                True,
+                True,
+                True,
+                True,
+                True,
+                True,
+                False,
+                False,
+                False,
+                False,
+                False,
+            ],
+        )
+        result = index.drop(index[0])
+        tm.assert_index_equal(result, expected)
+
+    def test_date_range_localize(self, unit):
+        rng = date_range(
+            "3/11/2012 03:00", periods=15, freq="h", tz="US/Eastern", unit=unit
+        )
+        rng2 = DatetimeIndex(
+            ["3/11/2012 03:00", "3/11/2012 04:00"], dtype=f"M8[{unit}, US/Eastern]"
+        )
+        rng3 = date_range("3/11/2012 03:00", periods=15, freq="h", unit=unit)
+        rng3 = rng3.tz_localize("US/Eastern")
+
+        tm.assert_index_equal(rng._with_freq(None), rng3)
+
+        # DST transition time
+        val = rng[0]
+        exp = Timestamp("3/11/2012 03:00", tz="US/Eastern")
+
+        assert val.hour == 3
+        assert exp.hour == 3
+        assert val == exp  # same UTC value
+        tm.assert_index_equal(rng[:2], rng2)
+
+    def test_date_range_localize2(self, unit):
+        # Right before the DST transition
+        rng = date_range(
+            "3/11/2012 00:00", periods=2, freq="h", tz="US/Eastern", unit=unit
+        )
+        rng2 = DatetimeIndex(
+            ["3/11/2012 00:00", "3/11/2012 01:00"],
+            dtype=f"M8[{unit}, US/Eastern]",
+            freq="h",
+        )
+        tm.assert_index_equal(rng, rng2)
+        exp = Timestamp("3/11/2012 00:00", tz="US/Eastern")
+        assert exp.hour == 0
+        assert rng[0] == exp
+        exp = Timestamp("3/11/2012 01:00", tz="US/Eastern")
+        assert exp.hour == 1
+        assert rng[1] == exp
+
+        rng = date_range(
+            "3/11/2012 00:00", periods=10, freq="h", tz="US/Eastern", unit=unit
+        )
+        assert rng[2].hour == 3
+
+    def test_timestamp_equality_different_timezones(self):
+        utc_range = date_range("1/1/2000", periods=20, tz="UTC")
+        eastern_range = utc_range.tz_convert("US/Eastern")
+        berlin_range = utc_range.tz_convert("Europe/Berlin")
+
+        for a, b, c in zip(utc_range, eastern_range, berlin_range):
+            assert a == b
+            assert b == c
+            assert a == c
+
+        assert (utc_range == eastern_range).all()
+        assert (utc_range == berlin_range).all()
+        assert (berlin_range == eastern_range).all()
+
+    def test_dti_equals_with_tz(self):
+        left = date_range("1/1/2011", periods=100, freq="h", tz="utc")
+        right = date_range("1/1/2011", periods=100, freq="h", tz="US/Eastern")
+
+        assert not left.equals(right)
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_dti_tz_nat(self, tzstr):
+        idx = DatetimeIndex([Timestamp("2013-1-1", tz=tzstr), pd.NaT])
+
+        assert isna(idx[1])
+        assert idx[0].tzinfo is not None
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_utc_box_timestamp_and_localize(self, tzstr):
+        tz = timezones.maybe_get_tz(tzstr)
+
+        rng = date_range("3/11/2012", "3/12/2012", freq="h", tz="utc")
+        rng_eastern = rng.tz_convert(tzstr)
+
+        expected = rng[-1].astimezone(tz)
+
+        stamp = rng_eastern[-1]
+        assert stamp == expected
+        assert stamp.tzinfo == expected.tzinfo
+
+        # right tzinfo
+        rng = date_range("3/13/2012", "3/14/2012", freq="h", tz="utc")
+        rng_eastern = rng.tz_convert(tzstr)
+        if "dateutil" in tzstr:
+            assert "EDT" in repr(rng_eastern[0].tzinfo) or "tzfile" in repr(
+                rng_eastern[0].tzinfo
+            )
+
+    @pytest.mark.parametrize(
+        "tz", [zoneinfo.ZoneInfo("US/Central"), gettz("US/Central")]
+    )
+    def test_with_tz(self, tz):
+        # just want it to work
+        start = datetime(2011, 3, 12, tzinfo=timezone.utc)
+        dr = bdate_range(start, periods=50, freq=pd.offsets.Hour())
+        assert dr.tz is timezone.utc
+
+        # DateRange with naive datetimes
+        dr = bdate_range("1/1/2005", "1/1/2009", tz=timezone.utc)
+        dr = bdate_range("1/1/2005", "1/1/2009", tz=tz)
+
+        # normalized
+        central = dr.tz_convert(tz)
+        assert central.tz is tz
+        naive = central[0].to_pydatetime().replace(tzinfo=None)
+        comp = conversion.localize_pydatetime(naive, tz).tzinfo
+        assert central[0].tz is comp
+
+        # compare vs a localized tz
+        naive = dr[0].to_pydatetime().replace(tzinfo=None)
+        comp = conversion.localize_pydatetime(naive, tz).tzinfo
+        assert central[0].tz is comp
+
+        # datetimes with tzinfo set
+        dr = bdate_range(
+            datetime(2005, 1, 1, tzinfo=timezone.utc),
+            datetime(2009, 1, 1, tzinfo=timezone.utc),
+        )
+        msg = "Start and end cannot both be tz-aware with different timezones"
+        with pytest.raises(Exception, match=msg):
+            bdate_range(datetime(2005, 1, 1, tzinfo=timezone.utc), "1/1/2009", tz=tz)
+
+    @pytest.mark.parametrize(
+        "tz", [zoneinfo.ZoneInfo("US/Eastern"), gettz("US/Eastern")]
+    )
+    def test_dti_convert_tz_aware_datetime_datetime(self, tz):
+        # GH#1581
+        dates = [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)]
+
+        dates_aware = [conversion.localize_pydatetime(x, tz) for x in dates]
+        result = DatetimeIndex(dates_aware).as_unit("ns")
+        assert timezones.tz_compare(result.tz, tz)
+
+        converted = to_datetime(dates_aware, utc=True).as_unit("ns")
+        ex_vals = np.array([Timestamp(x).as_unit("ns")._value for x in dates_aware])
+        tm.assert_numpy_array_equal(converted.asi8, ex_vals)
+        assert converted.tz is timezone.utc
diff --git a/pandas/pandas/tests/indexes/interval/__init__.py b/pandas/pandas/tests/indexes/interval/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/interval/test_astype.py b/pandas/pandas/tests/indexes/interval/test_astype.py
new file mode 100644
index 0000000000000..dde5f38074efb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_astype.py
@@ -0,0 +1,254 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    IntervalDtype,
+)
+
+from pandas import (
+    CategoricalIndex,
+    Index,
+    IntervalIndex,
+    NaT,
+    Timedelta,
+    Timestamp,
+    interval_range,
+)
+import pandas._testing as tm
+
+
+class AstypeTests:
+    """Tests common to IntervalIndex with any subtype"""
+
+    def test_astype_idempotent(self, index):
+        result = index.astype("interval")
+        tm.assert_index_equal(result, index)
+
+        result = index.astype(index.dtype)
+        tm.assert_index_equal(result, index)
+
+    def test_astype_object(self, index):
+        result = index.astype(object)
+        expected = Index(index.values, dtype="object")
+        tm.assert_index_equal(result, expected)
+        assert not result.equals(index)
+
+    def test_astype_category(self, index):
+        result = index.astype("category")
+        expected = CategoricalIndex(index.values)
+        tm.assert_index_equal(result, expected)
+
+        result = index.astype(CategoricalDtype())
+        tm.assert_index_equal(result, expected)
+
+        # non-default params
+        categories = index.dropna().unique().values[:-1]
+        dtype = CategoricalDtype(categories=categories, ordered=True)
+        result = index.astype(dtype)
+        expected = CategoricalIndex(index.values, categories=categories, ordered=True)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "int64",
+            "uint64",
+            "float64",
+            "complex128",
+            "period[M]",
+            "timedelta64",
+            "timedelta64[ns]",
+            "datetime64",
+            "datetime64[ns]",
+            "datetime64[ns, US/Eastern]",
+        ],
+    )
+    def test_astype_cannot_cast(self, index, dtype):
+        msg = "Cannot cast IntervalIndex to dtype"
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+    def test_astype_invalid_dtype(self, index):
+        msg = "data type [\"']fake_dtype[\"'] not understood"
+        with pytest.raises(TypeError, match=msg):
+            index.astype("fake_dtype")
+
+
+class TestIntSubtype(AstypeTests):
+    """Tests specific to IntervalIndex with integer-like subtype"""
+
+    indexes = [
+        IntervalIndex.from_breaks(np.arange(-10, 11, dtype="int64")),
+        IntervalIndex.from_breaks(np.arange(100, dtype="uint64"), closed="left"),
+    ]
+
+    @pytest.fixture(params=indexes)
+    def index(self, request):
+        return request.param
+
+    @pytest.mark.parametrize(
+        "subtype", ["float64", "datetime64[ns]", "timedelta64[ns]"]
+    )
+    def test_subtype_conversion(self, index, subtype):
+        dtype = IntervalDtype(subtype, index.closed)
+        result = index.astype(dtype)
+        expected = IntervalIndex.from_arrays(
+            index.left.astype(subtype), index.right.astype(subtype), closed=index.closed
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "subtype_start, subtype_end", [("int64", "uint64"), ("uint64", "int64")]
+    )
+    def test_subtype_integer(self, subtype_start, subtype_end):
+        index = IntervalIndex.from_breaks(np.arange(100, dtype=subtype_start))
+        dtype = IntervalDtype(subtype_end, index.closed)
+        result = index.astype(dtype)
+        expected = IntervalIndex.from_arrays(
+            index.left.astype(subtype_end),
+            index.right.astype(subtype_end),
+            closed=index.closed,
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.xfail(reason="GH#15832")
+    def test_subtype_integer_errors(self):
+        # int64 -> uint64 fails with negative values
+        index = interval_range(-10, 10)
+        dtype = IntervalDtype("uint64", "right")
+
+        # Until we decide what the exception message _should_ be, we
+        #  assert something that it should _not_ be.
+        #  We should _not_ be getting a message suggesting that the -10
+        #  has been wrapped around to a large-positive integer
+        msg = "^(?!(left side of interval must be <= right side))"
+        with pytest.raises(ValueError, match=msg):
+            index.astype(dtype)
+
+
+class TestFloatSubtype(AstypeTests):
+    """Tests specific to IntervalIndex with float subtype"""
+
+    indexes = [
+        interval_range(-10.0, 10.0, closed="neither"),
+        IntervalIndex.from_arrays(
+            [-1.5, np.nan, 0.0, 0.0, 1.5], [-0.5, np.nan, 1.0, 1.0, 3.0], closed="both"
+        ),
+    ]
+
+    @pytest.fixture(params=indexes)
+    def index(self, request):
+        return request.param
+
+    @pytest.mark.parametrize("subtype", ["int64", "uint64"])
+    def test_subtype_integer(self, subtype):
+        index = interval_range(0.0, 10.0)
+        dtype = IntervalDtype(subtype, "right")
+        result = index.astype(dtype)
+        expected = IntervalIndex.from_arrays(
+            index.left.astype(subtype), index.right.astype(subtype), closed=index.closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # raises with NA
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(ValueError, match=msg):
+            index.insert(0, np.nan).astype(dtype)
+
+    @pytest.mark.parametrize("subtype", ["int64", "uint64"])
+    def test_subtype_integer_with_non_integer_borders(self, subtype):
+        index = interval_range(0.0, 3.0, freq=0.25)
+        dtype = IntervalDtype(subtype, "right")
+        result = index.astype(dtype)
+        expected = IntervalIndex.from_arrays(
+            index.left.astype(subtype), index.right.astype(subtype), closed=index.closed
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_subtype_integer_errors(self):
+        # float64 -> uint64 fails with negative values
+        index = interval_range(-10.0, 10.0)
+        dtype = IntervalDtype("uint64", "right")
+        msg = re.escape(
+            "Cannot convert interval[float64, right] to interval[uint64, right]; "
+            "subtypes are incompatible"
+        )
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+    @pytest.mark.parametrize("subtype", ["datetime64[ns]", "timedelta64[ns]"])
+    def test_subtype_datetimelike(self, index, subtype):
+        dtype = IntervalDtype(subtype, "right")
+        msg = "Cannot convert .* to .*; subtypes are incompatible"
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_astype_category(self, index):
+        super().test_astype_category(index)
+
+
+class TestDatetimelikeSubtype(AstypeTests):
+    """Tests specific to IntervalIndex with datetime-like subtype"""
+
+    indexes = [
+        interval_range(Timestamp("2018-01-01"), periods=10, closed="neither"),
+        interval_range(Timestamp("2018-01-01"), periods=10).insert(2, NaT),
+        interval_range(Timestamp("2018-01-01", tz="US/Eastern"), periods=10),
+        interval_range(Timedelta("0 days"), periods=10, closed="both"),
+        interval_range(Timedelta("0 days"), periods=10).insert(2, NaT),
+    ]
+
+    @pytest.fixture(params=indexes)
+    def index(self, request):
+        return request.param
+
+    @pytest.mark.parametrize("subtype", ["int64", "uint64"])
+    def test_subtype_integer(self, index, subtype):
+        dtype = IntervalDtype(subtype, "right")
+
+        if subtype != "int64":
+            msg = (
+                r"Cannot convert interval\[(timedelta64|datetime64)\[ns.*\], .*\] "
+                r"to interval\[uint64, .*\]"
+            )
+            with pytest.raises(TypeError, match=msg):
+                index.astype(dtype)
+            return
+
+        result = index.astype(dtype)
+        new_left = index.left.astype(subtype)
+        new_right = index.right.astype(subtype)
+
+        expected = IntervalIndex.from_arrays(new_left, new_right, closed=index.closed)
+        tm.assert_index_equal(result, expected)
+
+    def test_subtype_float(self, index):
+        dtype = IntervalDtype("float64", "right")
+        msg = "Cannot convert .* to .*; subtypes are incompatible"
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+    def test_subtype_datetimelike(self):
+        # datetime -> timedelta raises
+        dtype = IntervalDtype("timedelta64[ns]", "right")
+        msg = "Cannot convert .* to .*; subtypes are incompatible"
+
+        index = interval_range(Timestamp("2018-01-01"), periods=10)
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+        index = interval_range(Timestamp("2018-01-01", tz="CET"), periods=10)
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
+
+        # timedelta -> datetime raises
+        dtype = IntervalDtype("datetime64[ns]", "right")
+        index = interval_range(Timedelta("0 days"), periods=10)
+        with pytest.raises(TypeError, match=msg):
+            index.astype(dtype)
diff --git a/pandas/pandas/tests/indexes/interval/test_constructors.py b/pandas/pandas/tests/indexes/interval/test_constructors.py
new file mode 100644
index 0000000000000..90423149658ab
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_constructors.py
@@ -0,0 +1,527 @@
+from functools import partial
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import is_unsigned_integer_dtype
+from pandas.core.dtypes.dtypes import IntervalDtype
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    date_range,
+    notna,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+import pandas.core.common as com
+
+
+class ConstructorTests:
+    """
+    Common tests for all variations of IntervalIndex construction. Input data
+    to be supplied in breaks format, then converted by the subclass method
+    get_kwargs_from_breaks to the expected format.
+    """
+
+    @pytest.mark.parametrize(
+        "breaks_and_expected_subtype",
+        [
+            ([3, 14, 15, 92, 653], np.int64),
+            (np.arange(10, dtype="int64"), np.int64),
+            (Index(np.arange(-10, 11, dtype=np.int64)), np.int64),
+            (Index(np.arange(10, 31, dtype=np.uint64)), np.uint64),
+            (Index(np.arange(20, 30, 0.5), dtype=np.float64), np.float64),
+            (date_range("20180101", periods=10), "M8[ns]"),
+            (
+                date_range("20180101", periods=10, tz="US/Eastern"),
+                "datetime64[ns, US/Eastern]",
+            ),
+            (timedelta_range("1 day", periods=10), "m8[ns]"),
+        ],
+    )
+    @pytest.mark.parametrize("name", [None, "foo"])
+    def test_constructor(self, constructor, breaks_and_expected_subtype, closed, name):
+        breaks, expected_subtype = breaks_and_expected_subtype
+
+        result_kwargs = self.get_kwargs_from_breaks(breaks, closed)
+
+        result = constructor(closed=closed, name=name, **result_kwargs)
+
+        assert result.closed == closed
+        assert result.name == name
+        assert result.dtype.subtype == expected_subtype
+        tm.assert_index_equal(result.left, Index(breaks[:-1], dtype=expected_subtype))
+        tm.assert_index_equal(result.right, Index(breaks[1:], dtype=expected_subtype))
+
+    @pytest.mark.parametrize(
+        "breaks, subtype",
+        [
+            (Index([0, 1, 2, 3, 4], dtype=np.int64), "float64"),
+            (Index([0, 1, 2, 3, 4], dtype=np.int64), "datetime64[ns]"),
+            (Index([0, 1, 2, 3, 4], dtype=np.int64), "timedelta64[ns]"),
+            (Index([0, 1, 2, 3, 4], dtype=np.float64), "int64"),
+            (date_range("2017-01-01", periods=5), "int64"),
+            (timedelta_range("1 day", periods=5), "int64"),
+        ],
+    )
+    def test_constructor_dtype(self, constructor, breaks, subtype):
+        # GH 19262: conversion via dtype parameter
+        expected_kwargs = self.get_kwargs_from_breaks(breaks.astype(subtype))
+        expected = constructor(**expected_kwargs)
+
+        result_kwargs = self.get_kwargs_from_breaks(breaks)
+        iv_dtype = IntervalDtype(subtype, "right")
+        for dtype in (iv_dtype, str(iv_dtype)):
+            result = constructor(dtype=dtype, **result_kwargs)
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            Index([0, 1, 2, 3, 4], dtype=np.int64),
+            Index([0, 1, 2, 3, 4], dtype=np.uint64),
+            Index([0, 1, 2, 3, 4], dtype=np.float64),
+            date_range("2017-01-01", periods=5),
+            timedelta_range("1 day", periods=5),
+        ],
+    )
+    def test_constructor_pass_closed(self, constructor, breaks):
+        # not passing closed to IntervalDtype, but to IntervalArray constructor
+        iv_dtype = IntervalDtype(breaks.dtype)
+
+        result_kwargs = self.get_kwargs_from_breaks(breaks)
+
+        for dtype in (iv_dtype, str(iv_dtype)):
+            with tm.assert_produces_warning(None):
+                result = constructor(dtype=dtype, closed="left", **result_kwargs)
+            assert result.dtype.closed == "left"
+
+    @pytest.mark.parametrize("breaks", [[np.nan] * 2, [np.nan] * 4, [np.nan] * 50])
+    def test_constructor_nan(self, constructor, breaks, closed):
+        # GH 18421
+        result_kwargs = self.get_kwargs_from_breaks(breaks)
+        result = constructor(closed=closed, **result_kwargs)
+
+        expected_subtype = np.float64
+        expected_values = np.array(breaks[:-1], dtype=object)
+
+        assert result.closed == closed
+        assert result.dtype.subtype == expected_subtype
+        tm.assert_numpy_array_equal(np.array(result), expected_values)
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            [],
+            np.array([], dtype="int64"),
+            np.array([], dtype="uint64"),
+            np.array([], dtype="float64"),
+            np.array([], dtype="datetime64[ns]"),
+            np.array([], dtype="timedelta64[ns]"),
+        ],
+    )
+    def test_constructor_empty(self, constructor, breaks, closed):
+        # GH 18421
+        result_kwargs = self.get_kwargs_from_breaks(breaks)
+        result = constructor(closed=closed, **result_kwargs)
+
+        expected_values = np.array([], dtype=object)
+        expected_subtype = getattr(breaks, "dtype", np.int64)
+
+        assert result.empty
+        assert result.closed == closed
+        assert result.dtype.subtype == expected_subtype
+        tm.assert_numpy_array_equal(np.array(result), expected_values)
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            tuple("0123456789"),
+            list("abcdefghij"),
+            np.array(list("abcdefghij"), dtype=object),
+            np.array(list("abcdefghij"), dtype="<U1"),
+        ],
+    )
+    def test_constructor_string(self, constructor, breaks):
+        # GH 19016
+        msg = (
+            "category, object, and string subtypes are not supported for IntervalIndex"
+        )
+        with pytest.raises(TypeError, match=msg):
+            constructor(**self.get_kwargs_from_breaks(breaks))
+
+    @pytest.mark.parametrize("cat_constructor", [Categorical, CategoricalIndex])
+    def test_constructor_categorical_valid(self, constructor, cat_constructor):
+        # GH 21243/21253
+
+        breaks = np.arange(10, dtype="int64")
+        expected = IntervalIndex.from_breaks(breaks)
+
+        cat_breaks = cat_constructor(breaks)
+        result_kwargs = self.get_kwargs_from_breaks(cat_breaks)
+        result = constructor(**result_kwargs)
+        tm.assert_index_equal(result, expected)
+
+    def test_generic_errors(self, constructor):
+        # filler input data to be used when supplying invalid kwargs
+        filler = self.get_kwargs_from_breaks(range(10))
+
+        # invalid closed
+        msg = "closed must be one of 'right', 'left', 'both', 'neither'"
+        with pytest.raises(ValueError, match=msg):
+            constructor(closed="invalid", **filler)
+
+        # unsupported dtype
+        msg = "dtype must be an IntervalDtype, got int64"
+        with pytest.raises(TypeError, match=msg):
+            constructor(dtype="int64", **filler)
+
+        # invalid dtype
+        msg = "data type [\"']invalid[\"'] not understood"
+        with pytest.raises(TypeError, match=msg):
+            constructor(dtype="invalid", **filler)
+
+        # no point in nesting periods in an IntervalIndex
+        periods = period_range("2000-01-01", periods=10)
+        periods_kwargs = self.get_kwargs_from_breaks(periods)
+        msg = "Period dtypes are not supported, use a PeriodIndex instead"
+        with pytest.raises(ValueError, match=msg):
+            constructor(**periods_kwargs)
+
+        # decreasing values
+        decreasing_kwargs = self.get_kwargs_from_breaks(range(10, -1, -1))
+        msg = "left side of interval must be <= right side"
+        with pytest.raises(ValueError, match=msg):
+            constructor(**decreasing_kwargs)
+
+
+class TestFromArrays(ConstructorTests):
+    """Tests specific to IntervalIndex.from_arrays"""
+
+    @pytest.fixture
+    def constructor(self):
+        """Fixture for IntervalIndex.from_arrays constructor"""
+        return IntervalIndex.from_arrays
+
+    def get_kwargs_from_breaks(self, breaks, closed="right"):
+        """
+        converts intervals in breaks format to a dictionary of kwargs to
+        specific to the format expected by IntervalIndex.from_arrays
+        """
+        return {"left": breaks[:-1], "right": breaks[1:]}
+
+    def test_constructor_errors(self):
+        # GH 19016: categorical data
+        data = Categorical(list("01234abcde"), ordered=True)
+        msg = (
+            "category, object, and string subtypes are not supported for IntervalIndex"
+        )
+        with pytest.raises(TypeError, match=msg):
+            IntervalIndex.from_arrays(data[:-1], data[1:])
+
+        # unequal length
+        left = [0, 1, 2]
+        right = [2, 3]
+        msg = "left and right must have the same length"
+        with pytest.raises(ValueError, match=msg):
+            IntervalIndex.from_arrays(left, right)
+
+    @pytest.mark.parametrize(
+        "left_subtype, right_subtype", [(np.int64, np.float64), (np.float64, np.int64)]
+    )
+    def test_mixed_float_int(self, left_subtype, right_subtype):
+        """mixed int/float left/right results in float for both sides"""
+        left = np.arange(9, dtype=left_subtype)
+        right = np.arange(1, 10, dtype=right_subtype)
+        result = IntervalIndex.from_arrays(left, right)
+
+        expected_left = Index(left, dtype=np.float64)
+        expected_right = Index(right, dtype=np.float64)
+        expected_subtype = np.float64
+
+        tm.assert_index_equal(result.left, expected_left)
+        tm.assert_index_equal(result.right, expected_right)
+        assert result.dtype.subtype == expected_subtype
+
+    @pytest.mark.parametrize("interval_cls", [IntervalArray, IntervalIndex])
+    def test_from_arrays_mismatched_datetimelike_resos(self, interval_cls):
+        # GH#55714
+        left = date_range("2016-01-01", periods=3, unit="s")
+        right = date_range("2017-01-01", periods=3, unit="ms")
+        result = interval_cls.from_arrays(left, right)
+        expected = interval_cls.from_arrays(left.as_unit("ms"), right)
+        tm.assert_equal(result, expected)
+
+        # td64
+        left2 = left - left[0]
+        right2 = right - left[0]
+        result2 = interval_cls.from_arrays(left2, right2)
+        expected2 = interval_cls.from_arrays(left2.as_unit("ms"), right2)
+        tm.assert_equal(result2, expected2)
+
+        # dt64tz
+        left3 = left.tz_localize("UTC")
+        right3 = right.tz_localize("UTC")
+        result3 = interval_cls.from_arrays(left3, right3)
+        expected3 = interval_cls.from_arrays(left3.as_unit("ms"), right3)
+        tm.assert_equal(result3, expected3)
+
+
+class TestFromBreaks(ConstructorTests):
+    """Tests specific to IntervalIndex.from_breaks"""
+
+    @pytest.fixture
+    def constructor(self):
+        """Fixture for IntervalIndex.from_breaks constructor"""
+        return IntervalIndex.from_breaks
+
+    def get_kwargs_from_breaks(self, breaks, closed="right"):
+        """
+        converts intervals in breaks format to a dictionary of kwargs to
+        specific to the format expected by IntervalIndex.from_breaks
+        """
+        return {"breaks": breaks}
+
+    def test_constructor_errors(self):
+        # GH 19016: categorical data
+        data = Categorical(list("01234abcde"), ordered=True)
+        msg = (
+            "category, object, and string subtypes are not supported for IntervalIndex"
+        )
+        with pytest.raises(TypeError, match=msg):
+            IntervalIndex.from_breaks(data)
+
+    def test_length_one(self):
+        """breaks of length one produce an empty IntervalIndex"""
+        breaks = [0]
+        result = IntervalIndex.from_breaks(breaks)
+        expected = IntervalIndex.from_breaks([])
+        tm.assert_index_equal(result, expected)
+
+    def test_left_right_dont_share_data(self):
+        # GH#36310
+        breaks = np.arange(5)
+        result = IntervalIndex.from_breaks(breaks)._data
+        assert result._left.base is None or result._left.base is not result._right.base
+
+
+class TestFromTuples(ConstructorTests):
+    """Tests specific to IntervalIndex.from_tuples"""
+
+    @pytest.fixture
+    def constructor(self):
+        """Fixture for IntervalIndex.from_tuples constructor"""
+        return IntervalIndex.from_tuples
+
+    def get_kwargs_from_breaks(self, breaks, closed="right"):
+        """
+        converts intervals in breaks format to a dictionary of kwargs to
+        specific to the format expected by IntervalIndex.from_tuples
+        """
+        if is_unsigned_integer_dtype(breaks):
+            pytest.skip(f"{breaks.dtype} not relevant IntervalIndex.from_tuples tests")
+
+        if len(breaks) == 0:
+            return {"data": breaks}
+
+        tuples = list(zip(breaks[:-1], breaks[1:]))
+        if isinstance(breaks, (list, tuple)):
+            return {"data": tuples}
+        elif isinstance(getattr(breaks, "dtype", None), CategoricalDtype):
+            return {"data": breaks._constructor(tuples)}
+        return {"data": com.asarray_tuplesafe(tuples)}
+
+    def test_constructor_errors(self):
+        # non-tuple
+        tuples = [(0, 1), 2, (3, 4)]
+        msg = "IntervalIndex.from_tuples received an invalid item, 2"
+        with pytest.raises(TypeError, match=msg.format(t=tuples)):
+            IntervalIndex.from_tuples(tuples)
+
+        # too few/many items
+        tuples = [(0, 1), (2,), (3, 4)]
+        msg = "IntervalIndex.from_tuples requires tuples of length 2, got {t}"
+        with pytest.raises(ValueError, match=msg.format(t=tuples)):
+            IntervalIndex.from_tuples(tuples)
+
+        tuples = [(0, 1), (2, 3, 4), (5, 6)]
+        with pytest.raises(ValueError, match=msg.format(t=tuples)):
+            IntervalIndex.from_tuples(tuples)
+
+    def test_na_tuples(self):
+        # tuple (NA, NA) evaluates the same as NA as an element
+        na_tuple = [(0, 1), (np.nan, np.nan), (2, 3)]
+        idx_na_tuple = IntervalIndex.from_tuples(na_tuple)
+        idx_na_element = IntervalIndex.from_tuples([(0, 1), np.nan, (2, 3)])
+        tm.assert_index_equal(idx_na_tuple, idx_na_element)
+
+
+class TestClassConstructors(ConstructorTests):
+    """Tests specific to the IntervalIndex/Index constructors"""
+
+    @pytest.fixture
+    def constructor(self):
+        """Fixture for IntervalIndex class constructor"""
+        return IntervalIndex
+
+    def get_kwargs_from_breaks(self, breaks, closed="right"):
+        """
+        converts intervals in breaks format to a dictionary of kwargs to
+        specific to the format expected by the IntervalIndex/Index constructors
+        """
+        if is_unsigned_integer_dtype(breaks):
+            pytest.skip(f"{breaks.dtype} not relevant for class constructor tests")
+
+        if len(breaks) == 0:
+            return {"data": breaks}
+
+        ivs = [
+            Interval(left, right, closed) if notna(left) else left
+            for left, right in zip(breaks[:-1], breaks[1:])
+        ]
+
+        if isinstance(breaks, list):
+            return {"data": ivs}
+        elif isinstance(getattr(breaks, "dtype", None), CategoricalDtype):
+            return {"data": breaks._constructor(ivs)}
+        return {"data": np.array(ivs, dtype=object)}
+
+    def test_generic_errors(self, constructor):
+        """
+        override the base class implementation since errors are handled
+        differently; checks unnecessary since caught at the Interval level
+        """
+
+    def test_constructor_string(self):
+        # GH23013
+        # When forming the interval from breaks,
+        # the interval of strings is already forbidden.
+        pass
+
+    @pytest.mark.parametrize(
+        "klass",
+        [IntervalIndex, partial(Index, dtype="interval")],
+        ids=["IntervalIndex", "Index"],
+    )
+    def test_constructor_errors(self, klass):
+        # mismatched closed within intervals with no constructor override
+        ivs = [Interval(0, 1, closed="right"), Interval(2, 3, closed="left")]
+        msg = "intervals must all be closed on the same side"
+        with pytest.raises(ValueError, match=msg):
+            klass(ivs)
+
+        # scalar
+        msg = (
+            r"(IntervalIndex|Index)\(...\) must be called with a collection of "
+            "some kind, 5 was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            klass(5)
+
+        # not an interval; dtype depends on 32bit/windows builds
+        msg = "type <class 'numpy.int(32|64)'> with value 0 is not an interval"
+        with pytest.raises(TypeError, match=msg):
+            klass([0, 1])
+
+    @pytest.mark.parametrize(
+        "data, closed",
+        [
+            ([], "both"),
+            ([np.nan, np.nan], "neither"),
+            (
+                [Interval(0, 3, closed="neither"), Interval(2, 5, closed="neither")],
+                "left",
+            ),
+            (
+                [Interval(0, 3, closed="left"), Interval(2, 5, closed="right")],
+                "neither",
+            ),
+            (IntervalIndex.from_breaks(range(5), closed="both"), "right"),
+        ],
+    )
+    def test_override_inferred_closed(self, constructor, data, closed):
+        # GH 19370
+        if isinstance(data, IntervalIndex):
+            tuples = data.to_tuples()
+        else:
+            tuples = [(iv.left, iv.right) if notna(iv) else iv for iv in data]
+        expected = IntervalIndex.from_tuples(tuples, closed=closed)
+        result = constructor(data, closed=closed)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values_constructor", [list, np.array, IntervalIndex, IntervalArray]
+    )
+    def test_index_object_dtype(self, values_constructor):
+        # Index(intervals, dtype=object) is an Index (not an IntervalIndex)
+        intervals = [Interval(0, 1), Interval(1, 2), Interval(2, 3)]
+        values = values_constructor(intervals)
+        result = Index(values, dtype=object)
+
+        assert type(result) is Index
+        tm.assert_numpy_array_equal(result.values, np.array(values))
+
+    def test_index_mixed_closed(self):
+        # GH27172
+        intervals = [
+            Interval(0, 1, closed="left"),
+            Interval(1, 2, closed="right"),
+            Interval(2, 3, closed="neither"),
+            Interval(3, 4, closed="both"),
+        ]
+        result = Index(intervals)
+        expected = Index(intervals, dtype=object)
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("timezone", ["UTC", "US/Pacific", "GMT"])
+def test_interval_index_subtype(timezone, inclusive_endpoints_fixture):
+    # GH#46999
+    dates = date_range("2022", periods=3, tz=timezone)
+    dtype = f"interval[datetime64[ns, {timezone}], {inclusive_endpoints_fixture}]"
+    result = IntervalIndex.from_arrays(
+        ["2022-01-01", "2022-01-02"],
+        ["2022-01-02", "2022-01-03"],
+        closed=inclusive_endpoints_fixture,
+        dtype=dtype,
+    )
+    expected = IntervalIndex.from_arrays(
+        dates[:-1], dates[1:], closed=inclusive_endpoints_fixture
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_dtype_closed_mismatch():
+    # GH#38394 closed specified in both dtype and IntervalIndex constructor
+
+    dtype = IntervalDtype(np.int64, "left")
+
+    msg = "closed keyword does not match dtype.closed"
+    with pytest.raises(ValueError, match=msg):
+        IntervalIndex([], dtype=dtype, closed="neither")
+
+    with pytest.raises(ValueError, match=msg):
+        IntervalArray([], dtype=dtype, closed="neither")
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    ["Float64", pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow"))],
+)
+def test_ea_dtype(dtype):
+    # GH#56765
+    bins = [(0.0, 0.4), (0.4, 0.6)]
+    interval_dtype = IntervalDtype(subtype=dtype, closed="left")
+    result = IntervalIndex.from_tuples(bins, closed="left", dtype=interval_dtype)
+    assert result.dtype == interval_dtype
+    expected = IntervalIndex.from_tuples(bins, closed="left").astype(interval_dtype)
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/interval/test_equals.py b/pandas/pandas/tests/indexes/interval/test_equals.py
new file mode 100644
index 0000000000000..87e2348e5fdb3
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_equals.py
@@ -0,0 +1,36 @@
+import numpy as np
+
+from pandas import (
+    IntervalIndex,
+    date_range,
+)
+
+
+class TestEquals:
+    def test_equals(self, closed):
+        expected = IntervalIndex.from_breaks(np.arange(5), closed=closed)
+        assert expected.equals(expected)
+        assert expected.equals(expected.copy())
+
+        assert not expected.equals(expected.astype(object))
+        assert not expected.equals(np.array(expected))
+        assert not expected.equals(list(expected))
+
+        assert not expected.equals([1, 2])
+        assert not expected.equals(np.array([1, 2]))
+        assert not expected.equals(date_range("20130101", periods=2))
+
+        expected_name1 = IntervalIndex.from_breaks(
+            np.arange(5), closed=closed, name="foo"
+        )
+        expected_name2 = IntervalIndex.from_breaks(
+            np.arange(5), closed=closed, name="bar"
+        )
+        assert expected.equals(expected_name1)
+        assert expected_name1.equals(expected_name2)
+
+        for other_closed in {"left", "right", "both", "neither"} - {closed}:
+            expected_other_closed = IntervalIndex.from_breaks(
+                np.arange(5), closed=other_closed
+            )
+            assert not expected.equals(expected_other_closed)
diff --git a/pandas/pandas/tests/indexes/interval/test_formats.py b/pandas/pandas/tests/indexes/interval/test_formats.py
new file mode 100644
index 0000000000000..d45d894c485c9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_formats.py
@@ -0,0 +1,114 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestIntervalIndexRendering:
+    # TODO: this is a test for DataFrame/Series, not IntervalIndex
+    @pytest.mark.parametrize(
+        "constructor,expected",
+        [
+            (
+                Series,
+                ("(0.0, 1.0]    a\nNaN           b\n(2.0, 3.0]    c\ndtype: object"),
+            ),
+            (DataFrame, ("            0\n(0.0, 1.0]  a\nNaN         b\n(2.0, 3.0]  c")),
+        ],
+    )
+    def test_repr_missing(self, constructor, expected, using_infer_string, request):
+        # GH 25984
+        if using_infer_string and constructor is Series:
+            request.applymarker(pytest.mark.xfail(reason="repr different"))
+        index = IntervalIndex.from_tuples([(0, 1), np.nan, (2, 3)])
+        obj = constructor(list("abc"), index=index)
+        result = repr(obj)
+        assert result == expected
+
+    def test_repr_floats(self):
+        # GH 32553
+
+        markers = Series(
+            [1, 2],
+            index=IntervalIndex(
+                [
+                    Interval(left, right)
+                    for left, right in zip(
+                        Index([329.973, 345.137], dtype="float64"),
+                        Index([345.137, 360.191], dtype="float64"),
+                    )
+                ]
+            ),
+        )
+        result = str(markers)
+        expected = "(329.973, 345.137]    1\n(345.137, 360.191]    2\ndtype: int64"
+        assert result == expected
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    @pytest.mark.parametrize(
+        "tuples, closed, expected_data",
+        [
+            ([(0, 1), (1, 2), (2, 3)], "left", ["[0, 1)", "[1, 2)", "[2, 3)"]),
+            (
+                [(0.5, 1.0), np.nan, (2.0, 3.0)],
+                "right",
+                ["(0.5, 1.0]", "NaN", "(2.0, 3.0]"],
+            ),
+            (
+                [
+                    (Timestamp("20180101"), Timestamp("20180102")),
+                    np.nan,
+                    ((Timestamp("20180102"), Timestamp("20180103"))),
+                ],
+                "both",
+                [
+                    "[2018-01-01 00:00:00, 2018-01-02 00:00:00]",
+                    "NaN",
+                    "[2018-01-02 00:00:00, 2018-01-03 00:00:00]",
+                ],
+            ),
+            (
+                [
+                    (Timedelta("0 days"), Timedelta("1 days")),
+                    (Timedelta("1 days"), Timedelta("2 days")),
+                    np.nan,
+                ],
+                "neither",
+                [
+                    "(0 days 00:00:00, 1 days 00:00:00)",
+                    "(1 days 00:00:00, 2 days 00:00:00)",
+                    "NaN",
+                ],
+            ),
+        ],
+    )
+    def test_get_values_for_csv(self, tuples, closed, expected_data):
+        # GH 28210
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        result = index._get_values_for_csv(na_rep="NaN")
+        expected = np.array(expected_data)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_timestamp_with_timezone(self, unit):
+        # GH 55035
+        left = DatetimeIndex(["2020-01-01"], dtype=f"M8[{unit}, UTC]")
+        right = DatetimeIndex(["2020-01-02"], dtype=f"M8[{unit}, UTC]")
+        index = IntervalIndex.from_arrays(left, right)
+        result = repr(index)
+        expected = (
+            "IntervalIndex([(2020-01-01 00:00:00+00:00, 2020-01-02 00:00:00+00:00]], "
+            f"dtype='interval[datetime64[{unit}, UTC], right]')"
+        )
+        assert result == expected
diff --git a/pandas/pandas/tests/indexes/interval/test_indexing.py b/pandas/pandas/tests/indexes/interval/test_indexing.py
new file mode 100644
index 0000000000000..5783a16e81d37
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_indexing.py
@@ -0,0 +1,673 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+from pandas import (
+    NA,
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    Timedelta,
+    Timestamp,
+    array,
+    date_range,
+    interval_range,
+    isna,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestGetItem:
+    def test_getitem(self, closed):
+        idx = IntervalIndex.from_arrays((0, 1, np.nan), (1, 2, np.nan), closed=closed)
+        assert idx[0] == Interval(0.0, 1.0, closed=closed)
+        assert idx[1] == Interval(1.0, 2.0, closed=closed)
+        assert isna(idx[2])
+
+        result = idx[0:1]
+        expected = IntervalIndex.from_arrays((0.0,), (1.0,), closed=closed)
+        tm.assert_index_equal(result, expected)
+
+        result = idx[0:2]
+        expected = IntervalIndex.from_arrays((0.0, 1), (1.0, 2.0), closed=closed)
+        tm.assert_index_equal(result, expected)
+
+        result = idx[1:3]
+        expected = IntervalIndex.from_arrays(
+            (1.0, np.nan), (2.0, np.nan), closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_getitem_2d_deprecated(self):
+        # GH#30588 multi-dim indexing is deprecated, but raising is also acceptable
+        idx = IntervalIndex.from_breaks(range(11), closed="right")
+        with pytest.raises(ValueError, match="multi-dimensional indexing not allowed"):
+            idx[:, None]
+        with pytest.raises(ValueError, match="multi-dimensional indexing not allowed"):
+            # GH#44051
+            idx[True]
+        with pytest.raises(ValueError, match="multi-dimensional indexing not allowed"):
+            # GH#44051
+            idx[False]
+
+
+class TestWhere:
+    def test_where(self, listlike_box):
+        klass = listlike_box
+
+        idx = IntervalIndex.from_breaks(range(11), closed="right")
+        cond = [True] * len(idx)
+        expected = idx
+        result = expected.where(klass(cond))
+        tm.assert_index_equal(result, expected)
+
+        cond = [False] + [True] * len(idx[1:])
+        expected = IntervalIndex([np.nan] + idx[1:].tolist())
+        result = idx.where(klass(cond))
+        tm.assert_index_equal(result, expected)
+
+
+class TestTake:
+    def test_take(self, closed):
+        index = IntervalIndex.from_breaks(range(11), closed=closed)
+
+        result = index.take(range(10))
+        tm.assert_index_equal(result, index)
+
+        result = index.take([0, 0, 1])
+        expected = IntervalIndex.from_arrays([0, 0, 1], [1, 1, 2], closed=closed)
+        tm.assert_index_equal(result, expected)
+
+
+class TestGetLoc:
+    @pytest.mark.parametrize("side", ["right", "left", "both", "neither"])
+    def test_get_loc_interval(self, closed, side):
+        idx = IntervalIndex.from_tuples([(0, 1), (2, 3)], closed=closed)
+
+        for bound in [[0, 1], [1, 2], [2, 3], [3, 4], [0, 2], [2.5, 3], [-1, 4]]:
+            # if get_loc is supplied an interval, it should only search
+            # for exact matches, not overlaps or covers, else KeyError.
+            msg = re.escape(f"Interval({bound[0]}, {bound[1]}, closed='{side}')")
+            if closed == side:
+                if bound == [0, 1]:
+                    assert idx.get_loc(Interval(0, 1, closed=side)) == 0
+                elif bound == [2, 3]:
+                    assert idx.get_loc(Interval(2, 3, closed=side)) == 1
+                else:
+                    with pytest.raises(KeyError, match=msg):
+                        idx.get_loc(Interval(*bound, closed=side))
+            else:
+                with pytest.raises(KeyError, match=msg):
+                    idx.get_loc(Interval(*bound, closed=side))
+
+    @pytest.mark.parametrize("scalar", [-0.5, 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5])
+    def test_get_loc_scalar(self, closed, scalar):
+        # correct = {side: {query: answer}}.
+        # If query is not in the dict, that query should raise a KeyError
+        correct = {
+            "right": {0.5: 0, 1: 0, 2.5: 1, 3: 1},
+            "left": {0: 0, 0.5: 0, 2: 1, 2.5: 1},
+            "both": {0: 0, 0.5: 0, 1: 0, 2: 1, 2.5: 1, 3: 1},
+            "neither": {0.5: 0, 2.5: 1},
+        }
+
+        idx = IntervalIndex.from_tuples([(0, 1), (2, 3)], closed=closed)
+
+        # if get_loc is supplied a scalar, it should return the index of
+        # the interval which contains the scalar, or KeyError.
+        if scalar in correct[closed].keys():
+            assert idx.get_loc(scalar) == correct[closed][scalar]
+        else:
+            with pytest.raises(KeyError, match=str(scalar)):
+                idx.get_loc(scalar)
+
+    @pytest.mark.parametrize("scalar", [-1, 0, 0.5, 3, 4.5, 5, 6])
+    def test_get_loc_length_one_scalar(self, scalar, closed):
+        # GH 20921
+        index = IntervalIndex.from_tuples([(0, 5)], closed=closed)
+        if scalar in index[0]:
+            result = index.get_loc(scalar)
+            assert result == 0
+        else:
+            with pytest.raises(KeyError, match=str(scalar)):
+                index.get_loc(scalar)
+
+    @pytest.mark.parametrize("left, right", [(0, 5), (-1, 4), (-1, 6), (6, 7)])
+    def test_get_loc_length_one_interval(self, left, right, closed, other_closed):
+        # GH 20921
+        index = IntervalIndex.from_tuples([(0, 5)], closed=closed)
+        interval = Interval(left, right, closed=other_closed)
+        if interval == index[0]:
+            result = index.get_loc(interval)
+            assert result == 0
+        else:
+            with pytest.raises(
+                KeyError,
+                match=re.escape(f"Interval({left}, {right}, closed='{other_closed}')"),
+            ):
+                index.get_loc(interval)
+
+    # Make consistent with test_interval_new.py (see #16316, #16386)
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            date_range("20180101", periods=4),
+            date_range("20180101", periods=4, tz="US/Eastern"),
+            timedelta_range("0 days", periods=4),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def test_get_loc_datetimelike_nonoverlapping(self, breaks):
+        # GH 20636
+        # nonoverlapping = IntervalIndex method and no i8 conversion
+        index = IntervalIndex.from_breaks(breaks)
+
+        value = index[0].mid
+        result = index.get_loc(value)
+        expected = 0
+        assert result == expected
+
+        interval = Interval(index[0].left, index[0].right)
+        result = index.get_loc(interval)
+        expected = 0
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "arrays",
+        [
+            (date_range("20180101", periods=4), date_range("20180103", periods=4)),
+            (
+                date_range("20180101", periods=4, tz="US/Eastern"),
+                date_range("20180103", periods=4, tz="US/Eastern"),
+            ),
+            (
+                timedelta_range("0 days", periods=4),
+                timedelta_range("2 days", periods=4),
+            ),
+        ],
+        ids=lambda x: str(x[0].dtype),
+    )
+    def test_get_loc_datetimelike_overlapping(self, arrays):
+        # GH 20636
+        index = IntervalIndex.from_arrays(*arrays)
+
+        value = index[0].mid + Timedelta("12 hours")
+        result = index.get_loc(value)
+        expected = slice(0, 2, None)
+        assert result == expected
+
+        interval = Interval(index[0].left, index[0].right)
+        result = index.get_loc(interval)
+        expected = 0
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            date_range("2018-01-04", periods=4, freq="-1D"),
+            date_range("2018-01-04", periods=4, freq="-1D", tz="US/Eastern"),
+            timedelta_range("3 days", periods=4, freq="-1D"),
+            np.arange(3.0, -1.0, -1.0),
+            np.arange(3, -1, -1),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def test_get_loc_decreasing(self, values):
+        # GH 25860
+        index = IntervalIndex.from_arrays(values[1:], values[:-1])
+        result = index.get_loc(index[0])
+        expected = 0
+        assert result == expected
+
+    @pytest.mark.parametrize("key", [[5], (2, 3)])
+    def test_get_loc_non_scalar_errors(self, key):
+        # GH 31117
+        idx = IntervalIndex.from_tuples([(1, 3), (2, 4), (3, 5), (7, 10), (3, 10)])
+
+        msg = str(key)
+        with pytest.raises(InvalidIndexError, match=msg):
+            idx.get_loc(key)
+
+    def test_get_indexer_with_nans(self):
+        # GH#41831
+        index = IntervalIndex([np.nan, Interval(1, 2), np.nan])
+
+        expected = np.array([True, False, True])
+        for key in [None, np.nan, NA]:
+            assert key in index
+            result = index.get_loc(key)
+            tm.assert_numpy_array_equal(result, expected)
+
+        for key in [NaT, np.timedelta64("NaT", "ns"), np.datetime64("NaT", "ns")]:
+            with pytest.raises(KeyError, match=str(key)):
+                index.get_loc(key)
+
+
+class TestGetIndexer:
+    @pytest.mark.parametrize(
+        "query, expected",
+        [
+            ([Interval(2, 4, closed="right")], [1]),
+            ([Interval(2, 4, closed="left")], [-1]),
+            ([Interval(2, 4, closed="both")], [-1]),
+            ([Interval(2, 4, closed="neither")], [-1]),
+            ([Interval(1, 4, closed="right")], [-1]),
+            ([Interval(0, 4, closed="right")], [-1]),
+            ([Interval(0.5, 1.5, closed="right")], [-1]),
+            ([Interval(2, 4, closed="right"), Interval(0, 1, closed="right")], [1, -1]),
+            ([Interval(2, 4, closed="right"), Interval(2, 4, closed="right")], [1, 1]),
+            ([Interval(5, 7, closed="right"), Interval(2, 4, closed="right")], [2, 1]),
+            ([Interval(2, 4, closed="right"), Interval(2, 4, closed="left")], [1, -1]),
+        ],
+    )
+    def test_get_indexer_with_interval(self, query, expected):
+        tuples = [(0, 2), (2, 4), (5, 7)]
+        index = IntervalIndex.from_tuples(tuples, closed="right")
+
+        result = index.get_indexer(query)
+        expected = np.array(expected, dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "query, expected",
+        [
+            ([-0.5], [-1]),
+            ([0], [-1]),
+            ([0.5], [0]),
+            ([1], [0]),
+            ([1.5], [1]),
+            ([2], [1]),
+            ([2.5], [-1]),
+            ([3], [-1]),
+            ([3.5], [2]),
+            ([4], [2]),
+            ([4.5], [-1]),
+            ([1, 2], [0, 1]),
+            ([1, 2, 3], [0, 1, -1]),
+            ([1, 2, 3, 4], [0, 1, -1, 2]),
+            ([1, 2, 3, 4, 2], [0, 1, -1, 2, 1]),
+        ],
+    )
+    def test_get_indexer_with_int_and_float(self, query, expected):
+        tuples = [(0, 1), (1, 2), (3, 4)]
+        index = IntervalIndex.from_tuples(tuples, closed="right")
+
+        result = index.get_indexer(query)
+        expected = np.array(expected, dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("item", [[3], np.arange(0.5, 5, 0.5)])
+    def test_get_indexer_length_one(self, item, closed):
+        # GH 17284
+        index = IntervalIndex.from_tuples([(0, 5)], closed=closed)
+        result = index.get_indexer(item)
+        expected = np.array([0] * len(item), dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("size", [1, 5])
+    def test_get_indexer_length_one_interval(self, size, closed):
+        # GH 17284
+        index = IntervalIndex.from_tuples([(0, 5)], closed=closed)
+        result = index.get_indexer([Interval(0, 5, closed)] * size)
+        expected = np.array([0] * size, dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "target",
+        [
+            IntervalIndex.from_tuples([(7, 8), (1, 2), (3, 4), (0, 1)]),
+            IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4), np.nan]),
+            IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4)], closed="both"),
+            [-1, 0, 0.5, 1, 2, 2.5, np.nan],
+            ["foo", "foo", "bar", "baz"],
+        ],
+    )
+    def test_get_indexer_categorical(self, target, ordered):
+        # GH 30063: categorical and non-categorical results should be consistent
+        index = IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4)])
+        categorical_target = CategoricalIndex(target, ordered=ordered)
+
+        result = index.get_indexer(categorical_target)
+        expected = index.get_indexer(target)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_get_indexer_categorical_with_nans(self):
+        # GH#41934 nans in both index and in target
+        ii = IntervalIndex.from_breaks(range(5))
+        ii2 = ii.append(IntervalIndex([np.nan]))
+        ci2 = CategoricalIndex(ii2)
+
+        result = ii2.get_indexer(ci2)
+        expected = np.arange(5, dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # not-all-matches
+        result = ii2[1:].get_indexer(ci2[::-1])
+        expected = np.array([3, 2, 1, 0, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # non-unique target, non-unique nans
+        result = ii2.get_indexer(ci2.append(ci2))
+        expected = np.array([0, 1, 2, 3, 4, 0, 1, 2, 3, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_datetime(self):
+        ii = IntervalIndex.from_breaks(date_range("2018-01-01", periods=4))
+        # TODO: with mismatched resolution get_indexer currently raises;
+        #  this should probably coerce?
+        target = DatetimeIndex(["2018-01-02"], dtype="M8[ns]")
+        result = ii.get_indexer(target)
+        expected = np.array([0], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = ii.get_indexer(target.astype(str))
+        tm.assert_numpy_array_equal(result, expected)
+
+        # https://github.com/pandas-dev/pandas/issues/47772
+        result = ii.get_indexer(target.asi8)
+        expected = np.array([-1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tuples, closed",
+        [
+            ([(0, 2), (1, 3), (3, 4)], "neither"),
+            ([(0, 5), (1, 4), (6, 7)], "left"),
+            ([(0, 1), (0, 1), (1, 2)], "right"),
+            ([(0, 1), (2, 3), (3, 4)], "both"),
+        ],
+    )
+    def test_get_indexer_errors(self, tuples, closed):
+        # IntervalIndex needs non-overlapping for uniqueness when querying
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+
+        msg = (
+            "cannot handle overlapping indices; use "
+            "IntervalIndex.get_indexer_non_unique"
+        )
+        with pytest.raises(InvalidIndexError, match=msg):
+            index.get_indexer([0, 2])
+
+    @pytest.mark.parametrize(
+        "query, expected",
+        [
+            ([-0.5], ([-1], [0])),
+            ([0], ([0], [])),
+            ([0.5], ([0], [])),
+            ([1], ([0, 1], [])),
+            ([1.5], ([0, 1], [])),
+            ([2], ([0, 1, 2], [])),
+            ([2.5], ([1, 2], [])),
+            ([3], ([2], [])),
+            ([3.5], ([2], [])),
+            ([4], ([-1], [0])),
+            ([4.5], ([-1], [0])),
+            ([1, 2], ([0, 1, 0, 1, 2], [])),
+            ([1, 2, 3], ([0, 1, 0, 1, 2, 2], [])),
+            ([1, 2, 3, 4], ([0, 1, 0, 1, 2, 2, -1], [3])),
+            ([1, 2, 3, 4, 2], ([0, 1, 0, 1, 2, 2, -1, 0, 1, 2], [3])),
+        ],
+    )
+    def test_get_indexer_non_unique_with_int_and_float(self, query, expected):
+        tuples = [(0, 2.5), (1, 3), (2, 4)]
+        index = IntervalIndex.from_tuples(tuples, closed="left")
+
+        result_indexer, result_missing = index.get_indexer_non_unique(query)
+        expected_indexer = np.array(expected[0], dtype="intp")
+        expected_missing = np.array(expected[1], dtype="intp")
+
+        tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+        tm.assert_numpy_array_equal(result_missing, expected_missing)
+
+        # TODO we may also want to test get_indexer for the case when
+        # the intervals are duplicated, decreasing, non-monotonic, etc..
+
+    def test_get_indexer_non_monotonic(self):
+        # GH 16410
+        idx1 = IntervalIndex.from_tuples([(2, 3), (4, 5), (0, 1)])
+        idx2 = IntervalIndex.from_tuples([(0, 1), (2, 3), (6, 7), (8, 9)])
+        result = idx1.get_indexer(idx2)
+        expected = np.array([2, 0, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx1.get_indexer(idx1[1:])
+        expected = np.array([1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_with_nans(self):
+        # GH#41831
+        index = IntervalIndex([np.nan, np.nan])
+        other = IntervalIndex([np.nan])
+
+        assert not index._index_as_unique
+
+        result = index.get_indexer_for(other)
+        expected = np.array([0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_index_non_unique_non_monotonic(self):
+        # GH#44084 (root cause)
+        index = IntervalIndex.from_tuples(
+            [(0.0, 1.0), (1.0, 2.0), (0.0, 1.0), (1.0, 2.0)]
+        )
+
+        result, _ = index.get_indexer_non_unique([Interval(1.0, 2.0)])
+        expected = np.array([1, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_multiindex_with_intervals(self):
+        # GH#44084 (MultiIndex case as reported)
+        interval_index = IntervalIndex.from_tuples(
+            [(2.0, 3.0), (0.0, 1.0), (1.0, 2.0)], name="interval"
+        )
+        foo_index = Index([1, 2, 3], name="foo")
+
+        multi_index = MultiIndex.from_product([foo_index, interval_index])
+
+        result = multi_index.get_level_values("interval").get_indexer_for(
+            [Interval(0.0, 1.0)]
+        )
+        expected = np.array([1, 4, 7], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [IntervalIndex, array, list])
+    def test_get_indexer_interval_index(self, box):
+        # GH#30178
+        rng = period_range("2022-07-01", freq="D", periods=3)
+        idx = box(interval_range(Timestamp("2022-07-01"), freq="3D", periods=3))
+
+        actual = rng.get_indexer(idx)
+        expected = np.array([-1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(actual, expected)
+
+    def test_get_indexer_read_only(self):
+        idx = interval_range(start=0, end=5)
+        arr = np.array([1, 2])
+        arr.flags.writeable = False
+        result = idx.get_indexer(arr)
+        expected = np.array([0, 1])
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+        result = idx.get_indexer_non_unique(arr)[0]
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+
+class TestSliceLocs:
+    def test_slice_locs_with_interval(self):
+        # increasing monotonically
+        index = IntervalIndex.from_tuples([(0, 2), (1, 3), (2, 4)])
+
+        assert index.slice_locs(start=Interval(0, 2), end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(start=Interval(0, 2)) == (0, 3)
+        assert index.slice_locs(end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(end=Interval(0, 2)) == (0, 1)
+        assert index.slice_locs(start=Interval(2, 4), end=Interval(0, 2)) == (2, 1)
+
+        # decreasing monotonically
+        index = IntervalIndex.from_tuples([(2, 4), (1, 3), (0, 2)])
+
+        assert index.slice_locs(start=Interval(0, 2), end=Interval(2, 4)) == (2, 1)
+        assert index.slice_locs(start=Interval(0, 2)) == (2, 3)
+        assert index.slice_locs(end=Interval(2, 4)) == (0, 1)
+        assert index.slice_locs(end=Interval(0, 2)) == (0, 3)
+        assert index.slice_locs(start=Interval(2, 4), end=Interval(0, 2)) == (0, 3)
+
+        # sorted duplicates
+        index = IntervalIndex.from_tuples([(0, 2), (0, 2), (2, 4)])
+
+        assert index.slice_locs(start=Interval(0, 2), end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(start=Interval(0, 2)) == (0, 3)
+        assert index.slice_locs(end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(end=Interval(0, 2)) == (0, 2)
+        assert index.slice_locs(start=Interval(2, 4), end=Interval(0, 2)) == (2, 2)
+
+        # unsorted duplicates
+        index = IntervalIndex.from_tuples([(0, 2), (2, 4), (0, 2)])
+
+        with pytest.raises(
+            KeyError,
+            match=re.escape(
+                '"Cannot get left slice bound for non-unique label: '
+                "Interval(0, 2, closed='right')\""
+            ),
+        ):
+            index.slice_locs(start=Interval(0, 2), end=Interval(2, 4))
+
+        with pytest.raises(
+            KeyError,
+            match=re.escape(
+                '"Cannot get left slice bound for non-unique label: '
+                "Interval(0, 2, closed='right')\""
+            ),
+        ):
+            index.slice_locs(start=Interval(0, 2))
+
+        assert index.slice_locs(end=Interval(2, 4)) == (0, 2)
+
+        with pytest.raises(
+            KeyError,
+            match=re.escape(
+                '"Cannot get right slice bound for non-unique label: '
+                "Interval(0, 2, closed='right')\""
+            ),
+        ):
+            index.slice_locs(end=Interval(0, 2))
+
+        with pytest.raises(
+            KeyError,
+            match=re.escape(
+                '"Cannot get right slice bound for non-unique label: '
+                "Interval(0, 2, closed='right')\""
+            ),
+        ):
+            index.slice_locs(start=Interval(2, 4), end=Interval(0, 2))
+
+        # another unsorted duplicates
+        index = IntervalIndex.from_tuples([(0, 2), (0, 2), (2, 4), (1, 3)])
+
+        assert index.slice_locs(start=Interval(0, 2), end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(start=Interval(0, 2)) == (0, 4)
+        assert index.slice_locs(end=Interval(2, 4)) == (0, 3)
+        assert index.slice_locs(end=Interval(0, 2)) == (0, 2)
+        assert index.slice_locs(start=Interval(2, 4), end=Interval(0, 2)) == (2, 2)
+
+    def test_slice_locs_with_ints_and_floats_succeeds(self):
+        # increasing non-overlapping
+        index = IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4)])
+
+        assert index.slice_locs(0, 1) == (0, 1)
+        assert index.slice_locs(0, 2) == (0, 2)
+        assert index.slice_locs(0, 3) == (0, 2)
+        assert index.slice_locs(3, 1) == (2, 1)
+        assert index.slice_locs(3, 4) == (2, 3)
+        assert index.slice_locs(0, 4) == (0, 3)
+
+        # decreasing non-overlapping
+        index = IntervalIndex.from_tuples([(3, 4), (1, 2), (0, 1)])
+        assert index.slice_locs(0, 1) == (3, 3)
+        assert index.slice_locs(0, 2) == (3, 2)
+        assert index.slice_locs(0, 3) == (3, 1)
+        assert index.slice_locs(3, 1) == (1, 3)
+        assert index.slice_locs(3, 4) == (1, 1)
+        assert index.slice_locs(0, 4) == (3, 1)
+
+    @pytest.mark.parametrize("query", [[0, 1], [0, 2], [0, 3], [0, 4]])
+    @pytest.mark.parametrize(
+        "tuples",
+        [
+            [(0, 2), (1, 3), (2, 4)],
+            [(2, 4), (1, 3), (0, 2)],
+            [(0, 2), (0, 2), (2, 4)],
+            [(0, 2), (2, 4), (0, 2)],
+            [(0, 2), (0, 2), (2, 4), (1, 3)],
+        ],
+    )
+    def test_slice_locs_with_ints_and_floats_errors(self, tuples, query):
+        start, stop = query
+        index = IntervalIndex.from_tuples(tuples)
+        with pytest.raises(
+            KeyError,
+            match=(
+                "'can only get slices from an IntervalIndex if bounds are "
+                "non-overlapping and all monotonic increasing or decreasing'"
+            ),
+        ):
+            index.slice_locs(start, stop)
+
+
+class TestPutmask:
+    @pytest.mark.parametrize("tz", ["US/Pacific", None])
+    def test_putmask_dt64(self, tz):
+        # GH#37968
+        dti = date_range("2016-01-01", periods=9, tz=tz)
+        idx = IntervalIndex.from_breaks(dti)
+        mask = np.zeros(idx.shape, dtype=bool)
+        mask[0:3] = True
+
+        result = idx.putmask(mask, idx[-1])
+        expected = IntervalIndex([idx[-1]] * 3 + list(idx[3:]))
+        tm.assert_index_equal(result, expected)
+
+    def test_putmask_td64(self):
+        # GH#37968
+        dti = date_range("2016-01-01", periods=9)
+        tdi = dti - dti[0]
+        idx = IntervalIndex.from_breaks(tdi)
+        mask = np.zeros(idx.shape, dtype=bool)
+        mask[0:3] = True
+
+        result = idx.putmask(mask, idx[-1])
+        expected = IntervalIndex([idx[-1]] * 3 + list(idx[3:]))
+        tm.assert_index_equal(result, expected)
+
+
+class TestContains:
+    # .__contains__, not .contains
+
+    def test_contains_dunder(self):
+        index = IntervalIndex.from_arrays([0, 1], [1, 2], closed="right")
+
+        # __contains__ requires perfect matches to intervals.
+        assert 0 not in index
+        assert 1 not in index
+        assert 2 not in index
+
+        assert Interval(0, 1, closed="right") in index
+        assert Interval(0, 2, closed="right") not in index
+        assert Interval(0, 0.5, closed="right") not in index
+        assert Interval(3, 5, closed="right") not in index
+        assert Interval(-1, 0, closed="left") not in index
+        assert Interval(0, 1, closed="left") not in index
+        assert Interval(0, 1, closed="both") not in index
diff --git a/pandas/pandas/tests/indexes/interval/test_interval.py b/pandas/pandas/tests/indexes/interval/test_interval.py
new file mode 100644
index 0000000000000..006a06e529971
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_interval.py
@@ -0,0 +1,910 @@
+from itertools import permutations
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    Interval,
+    IntervalIndex,
+    Timedelta,
+    Timestamp,
+    date_range,
+    interval_range,
+    isna,
+    notna,
+    timedelta_range,
+)
+import pandas._testing as tm
+import pandas.core.common as com
+
+
+class TestIntervalIndex:
+    index = IntervalIndex.from_arrays([0, 1], [1, 2])
+
+    def create_index(self, closed="right"):
+        return IntervalIndex.from_breaks(range(11), closed=closed)
+
+    def create_index_with_nan(self, closed="right"):
+        mask = [True, False] + [True] * 8
+        return IntervalIndex.from_arrays(
+            np.where(mask, np.arange(10), np.nan),
+            np.where(mask, np.arange(1, 11), np.nan),
+            closed=closed,
+        )
+
+    def test_properties(self, closed):
+        index = self.create_index(closed=closed)
+        assert len(index) == 10
+        assert index.size == 10
+        assert index.shape == (10,)
+
+        tm.assert_index_equal(index.left, Index(np.arange(10, dtype=np.int64)))
+        tm.assert_index_equal(index.right, Index(np.arange(1, 11, dtype=np.int64)))
+        tm.assert_index_equal(index.mid, Index(np.arange(0.5, 10.5, dtype=np.float64)))
+
+        assert index.closed == closed
+
+        ivs = [
+            Interval(left, right, closed)
+            for left, right in zip(range(10), range(1, 11))
+        ]
+        expected = np.array(ivs, dtype=object)
+        tm.assert_numpy_array_equal(np.asarray(index), expected)
+
+        # with nans
+        index = self.create_index_with_nan(closed=closed)
+        assert len(index) == 10
+        assert index.size == 10
+        assert index.shape == (10,)
+
+        expected_left = Index([0, np.nan, 2, 3, 4, 5, 6, 7, 8, 9])
+        expected_right = expected_left + 1
+        expected_mid = expected_left + 0.5
+        tm.assert_index_equal(index.left, expected_left)
+        tm.assert_index_equal(index.right, expected_right)
+        tm.assert_index_equal(index.mid, expected_mid)
+
+        assert index.closed == closed
+
+        ivs = [
+            Interval(left, right, closed) if notna(left) else np.nan
+            for left, right in zip(expected_left, expected_right)
+        ]
+        expected = np.array(ivs, dtype=object)
+        tm.assert_numpy_array_equal(np.asarray(index), expected)
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            [1, 1, 2, 5, 15, 53, 217, 1014, 5335, 31240, 201608],
+            [-np.inf, -100, -10, 0.5, 1, 1.5, 3.8, 101, 202, np.inf],
+            date_range("2017-01-01", "2017-01-04"),
+            pytest.param(
+                date_range("2017-01-01", "2017-01-04", unit="s"),
+                marks=pytest.mark.xfail(reason="mismatched result unit"),
+            ),
+            pd.to_timedelta(["1ns", "2ms", "3s", "4min", "5h", "6D"]),
+        ],
+    )
+    def test_length(self, closed, breaks):
+        # GH 18789
+        index = IntervalIndex.from_breaks(breaks, closed=closed)
+        result = index.length
+        expected = Index(iv.length for iv in index)
+        tm.assert_index_equal(result, expected)
+
+        # with NA
+        index = index.insert(1, np.nan)
+        result = index.length
+        expected = Index(iv.length if notna(iv) else iv for iv in index)
+        tm.assert_index_equal(result, expected)
+
+    def test_with_nans(self, closed):
+        index = self.create_index(closed=closed)
+        assert index.hasnans is False
+
+        result = index.isna()
+        expected = np.zeros(len(index), dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = index.notna()
+        expected = np.ones(len(index), dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        index = self.create_index_with_nan(closed=closed)
+        assert index.hasnans is True
+
+        result = index.isna()
+        expected = np.array([False, True] + [False] * (len(index) - 2))
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = index.notna()
+        expected = np.array([True, False] + [True] * (len(index) - 2))
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_copy(self, closed):
+        expected = self.create_index(closed=closed)
+
+        result = expected.copy()
+        assert result.equals(expected)
+
+        result = expected.copy(deep=True)
+        assert result.equals(expected)
+        assert result.left is not expected.left
+
+    def test_ensure_copied_data(self, closed):
+        # exercise the copy flag in the constructor
+
+        # not copying
+        index = self.create_index(closed=closed)
+        result = IntervalIndex(index, copy=False)
+        tm.assert_numpy_array_equal(
+            index.left.values, result.left.values, check_same="same"
+        )
+        tm.assert_numpy_array_equal(
+            index.right.values, result.right.values, check_same="same"
+        )
+
+        # by-definition make a copy
+        result = IntervalIndex(np.array(index), copy=False)
+        tm.assert_numpy_array_equal(
+            index.left.values, result.left.values, check_same="copy"
+        )
+        tm.assert_numpy_array_equal(
+            index.right.values, result.right.values, check_same="copy"
+        )
+
+    def test_delete(self, closed):
+        breaks = np.arange(1, 11, dtype=np.int64)
+        expected = IntervalIndex.from_breaks(breaks, closed=closed)
+        result = self.create_index(closed=closed).delete(0)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            interval_range(0, periods=10, closed="neither"),
+            interval_range(1.7, periods=8, freq=2.5, closed="both"),
+            interval_range(Timestamp("20170101"), periods=12, closed="left"),
+            interval_range(Timedelta("1 day"), periods=6, closed="right"),
+        ],
+    )
+    def test_insert(self, data):
+        item = data[0]
+        idx_item = IntervalIndex([item])
+
+        # start
+        expected = idx_item.append(data)
+        result = data.insert(0, item)
+        tm.assert_index_equal(result, expected)
+
+        # end
+        expected = data.append(idx_item)
+        result = data.insert(len(data), item)
+        tm.assert_index_equal(result, expected)
+
+        # mid
+        expected = data[:3].append(idx_item).append(data[3:])
+        result = data.insert(3, item)
+        tm.assert_index_equal(result, expected)
+
+        # invalid type
+        res = data.insert(1, "foo")
+        expected = data.astype(object).insert(1, "foo")
+        tm.assert_index_equal(res, expected)
+
+        msg = "can only insert Interval objects and NA into an IntervalArray"
+        with pytest.raises(TypeError, match=msg):
+            data._data.insert(1, "foo")
+
+        # invalid closed
+        msg = "'value.closed' is 'left', expected 'right'."
+        for closed in {"left", "right", "both", "neither"} - {item.closed}:
+            msg = f"'value.closed' is '{closed}', expected '{item.closed}'."
+            bad_item = Interval(item.left, item.right, closed=closed)
+            res = data.insert(1, bad_item)
+            expected = data.astype(object).insert(1, bad_item)
+            tm.assert_index_equal(res, expected)
+            with pytest.raises(ValueError, match=msg):
+                data._data.insert(1, bad_item)
+
+        # GH 18295 (test missing)
+        na_idx = IntervalIndex([np.nan], closed=data.closed)
+        for na in [np.nan, None, pd.NA]:
+            expected = data[:1].append(na_idx).append(data[1:])
+            result = data.insert(1, na)
+            tm.assert_index_equal(result, expected)
+
+        if data.left.dtype.kind not in ["m", "M"]:
+            # trying to insert pd.NaT into a numeric-dtyped Index should cast
+            expected = data.astype(object).insert(1, pd.NaT)
+
+            msg = "can only insert Interval objects and NA into an IntervalArray"
+            with pytest.raises(TypeError, match=msg):
+                data._data.insert(1, pd.NaT)
+
+        result = data.insert(1, pd.NaT)
+        tm.assert_index_equal(result, expected)
+
+    def test_is_unique_interval(self, closed):
+        """
+        Interval specific tests for is_unique in addition to base class tests
+        """
+        # unique overlapping - distinct endpoints
+        idx = IntervalIndex.from_tuples([(0, 1), (0.5, 1.5)], closed=closed)
+        assert idx.is_unique is True
+
+        # unique overlapping - shared endpoints
+        idx = IntervalIndex.from_tuples([(1, 2), (1, 3), (2, 3)], closed=closed)
+        assert idx.is_unique is True
+
+        # unique nested
+        idx = IntervalIndex.from_tuples([(-1, 1), (-2, 2)], closed=closed)
+        assert idx.is_unique is True
+
+        # unique NaN
+        idx = IntervalIndex.from_tuples([(np.nan, np.nan)], closed=closed)
+        assert idx.is_unique is True
+
+        # non-unique NaN
+        idx = IntervalIndex.from_tuples(
+            [(np.nan, np.nan), (np.nan, np.nan)], closed=closed
+        )
+        assert idx.is_unique is False
+
+    def test_monotonic(self, closed):
+        # increasing non-overlapping
+        idx = IntervalIndex.from_tuples([(0, 1), (2, 3), (4, 5)], closed=closed)
+        assert idx.is_monotonic_increasing is True
+        assert idx._is_strictly_monotonic_increasing is True
+        assert idx.is_monotonic_decreasing is False
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # decreasing non-overlapping
+        idx = IntervalIndex.from_tuples([(4, 5), (2, 3), (1, 2)], closed=closed)
+        assert idx.is_monotonic_increasing is False
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is True
+        assert idx._is_strictly_monotonic_decreasing is True
+
+        # unordered non-overlapping
+        idx = IntervalIndex.from_tuples([(0, 1), (4, 5), (2, 3)], closed=closed)
+        assert idx.is_monotonic_increasing is False
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is False
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # increasing overlapping
+        idx = IntervalIndex.from_tuples([(0, 2), (0.5, 2.5), (1, 3)], closed=closed)
+        assert idx.is_monotonic_increasing is True
+        assert idx._is_strictly_monotonic_increasing is True
+        assert idx.is_monotonic_decreasing is False
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # decreasing overlapping
+        idx = IntervalIndex.from_tuples([(1, 3), (0.5, 2.5), (0, 2)], closed=closed)
+        assert idx.is_monotonic_increasing is False
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is True
+        assert idx._is_strictly_monotonic_decreasing is True
+
+        # unordered overlapping
+        idx = IntervalIndex.from_tuples([(0.5, 2.5), (0, 2), (1, 3)], closed=closed)
+        assert idx.is_monotonic_increasing is False
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is False
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # increasing overlapping shared endpoints
+        idx = IntervalIndex.from_tuples([(1, 2), (1, 3), (2, 3)], closed=closed)
+        assert idx.is_monotonic_increasing is True
+        assert idx._is_strictly_monotonic_increasing is True
+        assert idx.is_monotonic_decreasing is False
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # decreasing overlapping shared endpoints
+        idx = IntervalIndex.from_tuples([(2, 3), (1, 3), (1, 2)], closed=closed)
+        assert idx.is_monotonic_increasing is False
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is True
+        assert idx._is_strictly_monotonic_decreasing is True
+
+        # stationary
+        idx = IntervalIndex.from_tuples([(0, 1), (0, 1)], closed=closed)
+        assert idx.is_monotonic_increasing is True
+        assert idx._is_strictly_monotonic_increasing is False
+        assert idx.is_monotonic_decreasing is True
+        assert idx._is_strictly_monotonic_decreasing is False
+
+        # empty
+        idx = IntervalIndex([], closed=closed)
+        assert idx.is_monotonic_increasing is True
+        assert idx._is_strictly_monotonic_increasing is True
+        assert idx.is_monotonic_decreasing is True
+        assert idx._is_strictly_monotonic_decreasing is True
+
+    def test_is_monotonic_with_nans(self):
+        # GH#41831
+        index = IntervalIndex([np.nan, np.nan])
+
+        assert not index.is_monotonic_increasing
+        assert not index._is_strictly_monotonic_increasing
+        assert not index.is_monotonic_increasing
+        assert not index._is_strictly_monotonic_decreasing
+        assert not index.is_monotonic_decreasing
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [
+            date_range("20180101", periods=4),
+            date_range("20180101", periods=4, tz="US/Eastern"),
+            timedelta_range("0 days", periods=4),
+        ],
+        ids=lambda x: str(x.dtype),
+    )
+    def test_maybe_convert_i8(self, breaks):
+        # GH 20636
+        index = IntervalIndex.from_breaks(breaks)
+
+        # intervalindex
+        result = index._maybe_convert_i8(index)
+        expected = IntervalIndex.from_breaks(breaks.asi8)
+        tm.assert_index_equal(result, expected)
+
+        # interval
+        interval = Interval(breaks[0], breaks[1])
+        result = index._maybe_convert_i8(interval)
+        expected = Interval(breaks[0]._value, breaks[1]._value)
+        assert result == expected
+
+        # datetimelike index
+        result = index._maybe_convert_i8(breaks)
+        expected = Index(breaks.asi8)
+        tm.assert_index_equal(result, expected)
+
+        # datetimelike scalar
+        result = index._maybe_convert_i8(breaks[0])
+        expected = breaks[0]._value
+        assert result == expected
+
+        # list-like of datetimelike scalars
+        result = index._maybe_convert_i8(list(breaks))
+        expected = Index(breaks.asi8)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "breaks",
+        [date_range("2018-01-01", periods=5), timedelta_range("0 days", periods=5)],
+    )
+    def test_maybe_convert_i8_nat(self, breaks):
+        # GH 20636
+        index = IntervalIndex.from_breaks(breaks)
+
+        to_convert = breaks._constructor([pd.NaT] * 3).as_unit("ns")
+        expected = Index([np.nan] * 3, dtype=np.float64)
+        result = index._maybe_convert_i8(to_convert)
+        tm.assert_index_equal(result, expected)
+
+        to_convert = to_convert.insert(0, breaks[0])
+        expected = expected.insert(0, float(breaks[0]._value))
+        result = index._maybe_convert_i8(to_convert)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "make_key",
+        [lambda breaks: breaks, list],
+        ids=["lambda", "list"],
+    )
+    def test_maybe_convert_i8_numeric(self, make_key, any_real_numpy_dtype):
+        # GH 20636
+        breaks = np.arange(5, dtype=any_real_numpy_dtype)
+        index = IntervalIndex.from_breaks(breaks)
+        key = make_key(breaks)
+
+        result = index._maybe_convert_i8(key)
+        kind = breaks.dtype.kind
+        expected_dtype = {"i": np.int64, "u": np.uint64, "f": np.float64}[kind]
+        expected = Index(key, dtype=expected_dtype)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "make_key",
+        [
+            IntervalIndex.from_breaks,
+            lambda breaks: Interval(breaks[0], breaks[1]),
+            lambda breaks: breaks[0],
+        ],
+        ids=["IntervalIndex", "Interval", "scalar"],
+    )
+    def test_maybe_convert_i8_numeric_identical(self, make_key, any_real_numpy_dtype):
+        # GH 20636
+        breaks = np.arange(5, dtype=any_real_numpy_dtype)
+        index = IntervalIndex.from_breaks(breaks)
+        key = make_key(breaks)
+
+        # test if _maybe_convert_i8 won't change key if an Interval or IntervalIndex
+        result = index._maybe_convert_i8(key)
+        assert result is key
+
+    @pytest.mark.parametrize(
+        "breaks1, breaks2",
+        permutations(
+            [
+                date_range("20180101", periods=4),
+                date_range("20180101", periods=4, tz="US/Eastern"),
+                timedelta_range("0 days", periods=4),
+            ],
+            2,
+        ),
+        ids=lambda x: str(x.dtype),
+    )
+    @pytest.mark.parametrize(
+        "make_key",
+        [
+            IntervalIndex.from_breaks,
+            lambda breaks: Interval(breaks[0], breaks[1]),
+            lambda breaks: breaks,
+            lambda breaks: breaks[0],
+            list,
+        ],
+        ids=["IntervalIndex", "Interval", "Index", "scalar", "list"],
+    )
+    def test_maybe_convert_i8_errors(self, breaks1, breaks2, make_key):
+        # GH 20636
+        index = IntervalIndex.from_breaks(breaks1)
+        key = make_key(breaks2)
+
+        msg = (
+            f"Cannot index an IntervalIndex of subtype {breaks1.dtype} with "
+            f"values of dtype {breaks2.dtype}"
+        )
+        msg = re.escape(msg)
+        with pytest.raises(ValueError, match=msg):
+            index._maybe_convert_i8(key)
+
+    def test_contains_method(self):
+        # can select values that are IN the range of a value
+        i = IntervalIndex.from_arrays([0, 1], [1, 2])
+
+        expected = np.array([False, False], dtype="bool")
+        actual = i.contains(0)
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = i.contains(3)
+        tm.assert_numpy_array_equal(actual, expected)
+
+        expected = np.array([True, False], dtype="bool")
+        actual = i.contains(0.5)
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = i.contains(1)
+        tm.assert_numpy_array_equal(actual, expected)
+
+        # __contains__ not implemented for "interval in interval", follow
+        # that for the contains method for now
+        with pytest.raises(
+            NotImplementedError, match="contains not implemented for two"
+        ):
+            i.contains(Interval(0, 1))
+
+    def test_dropna(self, closed):
+        expected = IntervalIndex.from_tuples([(0.0, 1.0), (1.0, 2.0)], closed=closed)
+
+        ii = IntervalIndex.from_tuples([(0, 1), (1, 2), np.nan], closed=closed)
+        result = ii.dropna()
+        tm.assert_index_equal(result, expected)
+
+        ii = IntervalIndex.from_arrays([0, 1, np.nan], [1, 2, np.nan], closed=closed)
+        result = ii.dropna()
+        tm.assert_index_equal(result, expected)
+
+    def test_non_contiguous(self, closed):
+        index = IntervalIndex.from_tuples([(0, 1), (2, 3)], closed=closed)
+        target = [0.5, 1.5, 2.5]
+        actual = index.get_indexer(target)
+        expected = np.array([0, -1, 1], dtype="intp")
+        tm.assert_numpy_array_equal(actual, expected)
+
+        assert 1.5 not in index
+
+    def test_isin(self, closed):
+        index = self.create_index(closed=closed)
+
+        expected = np.array([True] + [False] * (len(index) - 1))
+        result = index.isin(index[:1])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = index.isin([index[0]])
+        tm.assert_numpy_array_equal(result, expected)
+
+        other = IntervalIndex.from_breaks(np.arange(-2, 10), closed=closed)
+        expected = np.array([True] * (len(index) - 1) + [False])
+        result = index.isin(other)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = index.isin(other.tolist())
+        tm.assert_numpy_array_equal(result, expected)
+
+        for other_closed in ["right", "left", "both", "neither"]:
+            other = self.create_index(closed=other_closed)
+            expected = np.repeat(closed == other_closed, len(index))
+            result = index.isin(other)
+            tm.assert_numpy_array_equal(result, expected)
+
+            result = index.isin(other.tolist())
+            tm.assert_numpy_array_equal(result, expected)
+
+    def test_comparison(self):
+        actual = Interval(0, 1) < self.index
+        expected = np.array([False, True])
+        tm.assert_numpy_array_equal(actual, expected)
+
+        actual = Interval(0.5, 1.5) < self.index
+        expected = np.array([False, True])
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = self.index > Interval(0.5, 1.5)
+        tm.assert_numpy_array_equal(actual, expected)
+
+        actual = self.index == self.index
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = self.index <= self.index
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = self.index >= self.index
+        tm.assert_numpy_array_equal(actual, expected)
+
+        actual = self.index < self.index
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(actual, expected)
+        actual = self.index > self.index
+        tm.assert_numpy_array_equal(actual, expected)
+
+        actual = self.index == IntervalIndex.from_breaks([0, 1, 2], "left")
+        tm.assert_numpy_array_equal(actual, expected)
+
+        actual = self.index == self.index.values
+        tm.assert_numpy_array_equal(actual, np.array([True, True]))
+        actual = self.index.values == self.index
+        tm.assert_numpy_array_equal(actual, np.array([True, True]))
+        actual = self.index <= self.index.values
+        tm.assert_numpy_array_equal(actual, np.array([True, True]))
+        actual = self.index != self.index.values
+        tm.assert_numpy_array_equal(actual, np.array([False, False]))
+        actual = self.index > self.index.values
+        tm.assert_numpy_array_equal(actual, np.array([False, False]))
+        actual = self.index.values > self.index
+        tm.assert_numpy_array_equal(actual, np.array([False, False]))
+
+        # invalid comparisons
+        actual = self.index == 0
+        tm.assert_numpy_array_equal(actual, np.array([False, False]))
+        actual = self.index == self.index.left
+        tm.assert_numpy_array_equal(actual, np.array([False, False]))
+
+        msg = "|".join(
+            [
+                "not supported between instances of 'int' and '.*.Interval'",
+                r"Invalid comparison between dtype=interval\[int64, right\] and ",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            self.index > 0
+        with pytest.raises(TypeError, match=msg):
+            self.index <= 0
+        with pytest.raises(TypeError, match=msg):
+            self.index > np.arange(2)
+
+        msg = "Lengths must match to compare"
+        with pytest.raises(ValueError, match=msg):
+            self.index > np.arange(3)
+
+    def test_missing_values(self, closed):
+        idx = Index(
+            [np.nan, Interval(0, 1, closed=closed), Interval(1, 2, closed=closed)]
+        )
+        idx2 = IntervalIndex.from_arrays([np.nan, 0, 1], [np.nan, 1, 2], closed=closed)
+        assert idx.equals(idx2)
+
+        msg = (
+            "missing values must be missing in the same location both left "
+            "and right sides"
+        )
+        with pytest.raises(ValueError, match=msg):
+            IntervalIndex.from_arrays(
+                [np.nan, 0, 1], np.array([0, 1, 2]), closed=closed
+            )
+
+        tm.assert_numpy_array_equal(isna(idx), np.array([True, False, False]))
+
+    def test_sort_values(self, closed):
+        index = self.create_index(closed=closed)
+
+        result = index.sort_values()
+        tm.assert_index_equal(result, index)
+
+        result = index.sort_values(ascending=False)
+        tm.assert_index_equal(result, index[::-1])
+
+        # with nan
+        index = IntervalIndex([Interval(1, 2), np.nan, Interval(0, 1)])
+
+        result = index.sort_values()
+        expected = IntervalIndex([Interval(0, 1), Interval(1, 2), np.nan])
+        tm.assert_index_equal(result, expected)
+
+        result = index.sort_values(ascending=False, na_position="first")
+        expected = IntervalIndex([np.nan, Interval(1, 2), Interval(0, 1)])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_datetime(self, tz):
+        start = Timestamp("2000-01-01", tz=tz)
+        dates = date_range(start=start, periods=10)
+        index = IntervalIndex.from_breaks(dates)
+
+        # test mid
+        start = Timestamp("2000-01-01T12:00", tz=tz)
+        expected = date_range(start=start, periods=9)
+        tm.assert_index_equal(index.mid, expected)
+
+        # __contains__ doesn't check individual points
+        assert Timestamp("2000-01-01", tz=tz) not in index
+        assert Timestamp("2000-01-01T12", tz=tz) not in index
+        assert Timestamp("2000-01-02", tz=tz) not in index
+        iv_true = Interval(
+            Timestamp("2000-01-02", tz=tz), Timestamp("2000-01-03", tz=tz)
+        )
+        iv_false = Interval(
+            Timestamp("1999-12-31", tz=tz), Timestamp("2000-01-01", tz=tz)
+        )
+        assert iv_true in index
+        assert iv_false not in index
+
+        # .contains does check individual points
+        assert not index.contains(Timestamp("2000-01-01", tz=tz)).any()
+        assert index.contains(Timestamp("2000-01-01T12", tz=tz)).any()
+        assert index.contains(Timestamp("2000-01-02", tz=tz)).any()
+
+        # test get_indexer
+        start = Timestamp("1999-12-31T12:00", tz=tz)
+        target = date_range(start=start, periods=7, freq="12h")
+        actual = index.get_indexer(target)
+        expected = np.array([-1, -1, 0, 0, 1, 1, 2], dtype="intp")
+        tm.assert_numpy_array_equal(actual, expected)
+
+        start = Timestamp("2000-01-08T18:00", tz=tz)
+        target = date_range(start=start, periods=7, freq="6h")
+        actual = index.get_indexer(target)
+        expected = np.array([7, 7, 8, 8, 8, 8, -1], dtype="intp")
+        tm.assert_numpy_array_equal(actual, expected)
+
+    def test_append(self, closed):
+        index1 = IntervalIndex.from_arrays([0, 1], [1, 2], closed=closed)
+        index2 = IntervalIndex.from_arrays([1, 2], [2, 3], closed=closed)
+
+        result = index1.append(index2)
+        expected = IntervalIndex.from_arrays([0, 1, 1, 2], [1, 2, 2, 3], closed=closed)
+        tm.assert_index_equal(result, expected)
+
+        result = index1.append([index1, index2])
+        expected = IntervalIndex.from_arrays(
+            [0, 1, 0, 1, 1, 2], [1, 2, 1, 2, 2, 3], closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        for other_closed in {"left", "right", "both", "neither"} - {closed}:
+            index_other_closed = IntervalIndex.from_arrays(
+                [0, 1], [1, 2], closed=other_closed
+            )
+            result = index1.append(index_other_closed)
+            expected = index1.astype(object).append(index_other_closed.astype(object))
+            tm.assert_index_equal(result, expected)
+
+    def test_is_non_overlapping_monotonic(self, closed):
+        # Should be True in all cases
+        tpls = [(0, 1), (2, 3), (4, 5), (6, 7)]
+        idx = IntervalIndex.from_tuples(tpls, closed=closed)
+        assert idx.is_non_overlapping_monotonic is True
+
+        idx = IntervalIndex.from_tuples(tpls[::-1], closed=closed)
+        assert idx.is_non_overlapping_monotonic is True
+
+        # Should be False in all cases (overlapping)
+        tpls = [(0, 2), (1, 3), (4, 5), (6, 7)]
+        idx = IntervalIndex.from_tuples(tpls, closed=closed)
+        assert idx.is_non_overlapping_monotonic is False
+
+        idx = IntervalIndex.from_tuples(tpls[::-1], closed=closed)
+        assert idx.is_non_overlapping_monotonic is False
+
+        # Should be False in all cases (non-monotonic)
+        tpls = [(0, 1), (2, 3), (6, 7), (4, 5)]
+        idx = IntervalIndex.from_tuples(tpls, closed=closed)
+        assert idx.is_non_overlapping_monotonic is False
+
+        idx = IntervalIndex.from_tuples(tpls[::-1], closed=closed)
+        assert idx.is_non_overlapping_monotonic is False
+
+        # Should be False for closed='both', otherwise True (GH16560)
+        if closed == "both":
+            idx = IntervalIndex.from_breaks(range(4), closed=closed)
+            assert idx.is_non_overlapping_monotonic is False
+        else:
+            idx = IntervalIndex.from_breaks(range(4), closed=closed)
+            assert idx.is_non_overlapping_monotonic is True
+
+    @pytest.mark.parametrize(
+        "start, shift, na_value",
+        [
+            (0, 1, np.nan),
+            (Timestamp("2018-01-01"), Timedelta("1 day"), pd.NaT),
+            (Timedelta("0 days"), Timedelta("1 day"), pd.NaT),
+        ],
+    )
+    def test_is_overlapping(self, start, shift, na_value, closed):
+        # GH 23309
+        # see test_interval_tree.py for extensive tests; interface tests here
+
+        # non-overlapping
+        tuples = [(start + n * shift, start + (n + 1) * shift) for n in (0, 2, 4)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        assert index.is_overlapping is False
+
+        # non-overlapping with NA
+        tuples = [(na_value, na_value)] + tuples + [(na_value, na_value)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        assert index.is_overlapping is False
+
+        # overlapping
+        tuples = [(start + n * shift, start + (n + 2) * shift) for n in range(3)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        assert index.is_overlapping is True
+
+        # overlapping with NA
+        tuples = [(na_value, na_value)] + tuples + [(na_value, na_value)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        assert index.is_overlapping is True
+
+        # common endpoints
+        tuples = [(start + n * shift, start + (n + 1) * shift) for n in range(3)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        result = index.is_overlapping
+        expected = closed == "both"
+        assert result is expected
+
+        # common endpoints with NA
+        tuples = [(na_value, na_value)] + tuples + [(na_value, na_value)]
+        index = IntervalIndex.from_tuples(tuples, closed=closed)
+        result = index.is_overlapping
+        assert result is expected
+
+        # intervals with duplicate left values
+        a = [10, 15, 20, 25, 30, 35, 40, 45, 45, 50, 55, 60, 65, 70, 75, 80, 85]
+        b = [15, 20, 25, 30, 35, 40, 45, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90]
+        index = IntervalIndex.from_arrays(a, b, closed="right")
+        result = index.is_overlapping
+        assert result is False
+
+    @pytest.mark.parametrize(
+        "tuples",
+        [
+            zip(range(10), range(1, 11)),
+            zip(
+                date_range("20170101", periods=10),
+                date_range("20170101", periods=10),
+            ),
+            zip(
+                timedelta_range("0 days", periods=10),
+                timedelta_range("1 day", periods=10),
+            ),
+        ],
+    )
+    def test_to_tuples(self, tuples):
+        # GH 18756
+        tuples = list(tuples)
+        idx = IntervalIndex.from_tuples(tuples)
+        result = idx.to_tuples()
+        expected = Index(com.asarray_tuplesafe(tuples))
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tuples",
+        [
+            list(zip(range(10), range(1, 11))) + [np.nan],
+            list(
+                zip(
+                    date_range("20170101", periods=10),
+                    date_range("20170101", periods=10),
+                )
+            )
+            + [np.nan],
+            list(
+                zip(
+                    timedelta_range("0 days", periods=10),
+                    timedelta_range("1 day", periods=10),
+                )
+            )
+            + [np.nan],
+        ],
+    )
+    @pytest.mark.parametrize("na_tuple", [True, False])
+    def test_to_tuples_na(self, tuples, na_tuple):
+        # GH 18756
+        idx = IntervalIndex.from_tuples(tuples)
+        result = idx.to_tuples(na_tuple=na_tuple)
+
+        # check the non-NA portion
+        expected_notna = Index(com.asarray_tuplesafe(tuples[:-1]))
+        result_notna = result[:-1]
+        tm.assert_index_equal(result_notna, expected_notna)
+
+        # check the NA portion
+        result_na = result[-1]
+        if na_tuple:
+            assert isinstance(result_na, tuple)
+            assert len(result_na) == 2
+            assert all(isna(x) for x in result_na)
+        else:
+            assert isna(result_na)
+
+    def test_nbytes(self):
+        # GH 19209
+        left = np.arange(0, 4, dtype="i8")
+        right = np.arange(1, 5, dtype="i8")
+
+        result = IntervalIndex.from_arrays(left, right).nbytes
+        expected = 64  # 4 * 8 * 2
+        assert result == expected
+
+    @pytest.mark.parametrize("name", [None, "foo"])
+    def test_set_closed(self, name, closed, other_closed):
+        # GH 21670
+        index = interval_range(0, 5, closed=closed, name=name)
+        result = index.set_closed(other_closed)
+        expected = interval_range(0, 5, closed=other_closed, name=name)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("bad_closed", ["foo", 10, "LEFT", True, False])
+    def test_set_closed_errors(self, bad_closed):
+        # GH 21670
+        index = interval_range(0, 5)
+        msg = f"invalid option for 'closed': {bad_closed}"
+        with pytest.raises(ValueError, match=msg):
+            index.set_closed(bad_closed)
+
+    def test_is_all_dates(self):
+        # GH 23576
+        year_2017 = Interval(
+            Timestamp("2017-01-01 00:00:00"), Timestamp("2018-01-01 00:00:00")
+        )
+        year_2017_index = IntervalIndex([year_2017])
+        assert not year_2017_index._is_all_dates
+
+
+def test_dir():
+    # GH#27571 dir(interval_index) should not raise
+    index = IntervalIndex.from_arrays([0, 1], [1, 2])
+    result = dir(index)
+    assert "str" not in result
+
+
+def test_searchsorted_different_argument_classes(listlike_box):
+    # https://github.com/pandas-dev/pandas/issues/32762
+    values = IntervalIndex([Interval(0, 1), Interval(1, 2)])
+    result = values.searchsorted(listlike_box(values))
+    expected = np.array([0, 1], dtype=result.dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = values._data.searchsorted(listlike_box(values))
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg", [[1, 2], ["a", "b"], [Timestamp("2020-01-01", tz="Europe/London")] * 2]
+)
+def test_searchsorted_invalid_argument(arg):
+    values = IntervalIndex([Interval(0, 1), Interval(1, 2)])
+    msg = "'<' not supported between instances of 'pandas._libs.interval.Interval' and "
+    with pytest.raises(TypeError, match=msg):
+        values.searchsorted(arg)
diff --git a/pandas/pandas/tests/indexes/interval/test_interval_range.py b/pandas/pandas/tests/indexes/interval/test_interval_range.py
new file mode 100644
index 0000000000000..5252b85ad8d0e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_interval_range.py
@@ -0,0 +1,382 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_integer
+
+from pandas import (
+    DateOffset,
+    Interval,
+    IntervalIndex,
+    Timedelta,
+    Timestamp,
+    date_range,
+    interval_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import Day
+
+
+@pytest.fixture(params=[None, "foo"])
+def name(request):
+    return request.param
+
+
+class TestIntervalRange:
+    @pytest.mark.parametrize("freq, periods", [(1, 100), (2.5, 40), (5, 20), (25, 4)])
+    def test_constructor_numeric(self, closed, name, freq, periods):
+        start, end = 0, 100
+        breaks = np.arange(101, step=freq)
+        expected = IntervalIndex.from_breaks(breaks, name=name, closed=closed)
+
+        # defined from start/end/freq
+        result = interval_range(
+            start=start, end=end, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from start/periods/freq
+        result = interval_range(
+            start=start, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from end/periods/freq
+        result = interval_range(
+            end=end, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # GH 20976: linspace behavior defined from start/end/periods
+        result = interval_range(
+            start=start, end=end, periods=periods, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    @pytest.mark.parametrize(
+        "freq, periods", [("D", 364), ("2D", 182), ("22D18h", 16), ("ME", 11)]
+    )
+    def test_constructor_timestamp(self, closed, name, freq, periods, tz):
+        start, end = Timestamp("20180101", tz=tz), Timestamp("20181231", tz=tz)
+        breaks = date_range(start=start, end=end, freq=freq)
+        expected = IntervalIndex.from_breaks(breaks, name=name, closed=closed)
+
+        # defined from start/end/freq
+        result = interval_range(
+            start=start, end=end, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from start/periods/freq
+        result = interval_range(
+            start=start, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from end/periods/freq
+        result = interval_range(
+            end=end, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # GH 20976: linspace behavior defined from start/end/periods
+        if not breaks.freq.n == 1 and tz is None:
+            result = interval_range(
+                start=start, end=end, periods=periods, name=name, closed=closed
+            )
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "freq, periods", [("D", 100), ("2D12h", 40), ("5D", 20), ("25D", 4)]
+    )
+    def test_constructor_timedelta(self, closed, name, freq, periods):
+        start, end = Timedelta("0 days"), Timedelta("100 days")
+        breaks = timedelta_range(start=start, end=end, freq=freq)
+        expected = IntervalIndex.from_breaks(breaks, name=name, closed=closed)
+
+        # defined from start/end/freq
+        result = interval_range(
+            start=start, end=end, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from start/periods/freq
+        result = interval_range(
+            start=start, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # defined from end/periods/freq
+        result = interval_range(
+            end=end, periods=periods, freq=freq, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+        # GH 20976: linspace behavior defined from start/end/periods
+        result = interval_range(
+            start=start, end=end, periods=periods, name=name, closed=closed
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "start, end, freq, expected_endpoint",
+        [
+            (0, 10, 3, 9),
+            (0, 10, 1.5, 9),
+            (0.5, 10, 3, 9.5),
+            (Timedelta("0D"), Timedelta("10D"), "2D4h", Timedelta("8D16h")),
+            (
+                Timestamp("2018-01-01"),
+                Timestamp("2018-02-09"),
+                "MS",
+                Timestamp("2018-02-01"),
+            ),
+            (
+                Timestamp("2018-01-01", tz="US/Eastern"),
+                Timestamp("2018-01-20", tz="US/Eastern"),
+                "5D12h",
+                Timestamp("2018-01-17 12:00:00", tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_early_truncation(self, start, end, freq, expected_endpoint):
+        # index truncates early if freq causes end to be skipped
+        result = interval_range(start=start, end=end, freq=freq)
+        result_endpoint = result.right[-1]
+        assert result_endpoint == expected_endpoint
+
+    @pytest.mark.parametrize(
+        "start, end, freq",
+        [(0.5, None, None), (None, 4.5, None), (0.5, None, 1.5), (None, 6.5, 1.5)],
+    )
+    def test_no_invalid_float_truncation(self, start, end, freq):
+        # GH 21161
+        if freq is None:
+            breaks = [0.5, 1.5, 2.5, 3.5, 4.5]
+        else:
+            breaks = [0.5, 2.0, 3.5, 5.0, 6.5]
+        expected = IntervalIndex.from_breaks(breaks)
+
+        result = interval_range(start=start, end=end, periods=4, freq=freq)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "start, mid, end",
+        [
+            (
+                Timestamp("2018-03-10", tz="US/Eastern"),
+                Timestamp("2018-03-10 23:30:00", tz="US/Eastern"),
+                Timestamp("2018-03-12", tz="US/Eastern"),
+            ),
+            (
+                Timestamp("2018-11-03", tz="US/Eastern"),
+                Timestamp("2018-11-04 00:30:00", tz="US/Eastern"),
+                Timestamp("2018-11-05", tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_linspace_dst_transition(self, start, mid, end):
+        # GH 20976: linspace behavior defined from start/end/periods
+        # accounts for the hour gained/lost during DST transition
+        start = start.as_unit("ns")
+        mid = mid.as_unit("ns")
+        end = end.as_unit("ns")
+        result = interval_range(start=start, end=end, periods=2)
+        expected = IntervalIndex.from_breaks([start, mid, end])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", [2, 2.0])
+    @pytest.mark.parametrize("end", [10, 10.0])
+    @pytest.mark.parametrize("start", [0, 0.0])
+    def test_float_subtype(self, start, end, freq):
+        # Has float subtype if any of start/end/freq are float, even if all
+        # resulting endpoints can safely be upcast to integers
+
+        # defined from start/end/freq
+        index = interval_range(start=start, end=end, freq=freq)
+        result = index.dtype.subtype
+        expected = "int64" if is_integer(start + end + freq) else "float64"
+        assert result == expected
+
+        # defined from start/periods/freq
+        index = interval_range(start=start, periods=5, freq=freq)
+        result = index.dtype.subtype
+        expected = "int64" if is_integer(start + freq) else "float64"
+        assert result == expected
+
+        # defined from end/periods/freq
+        index = interval_range(end=end, periods=5, freq=freq)
+        result = index.dtype.subtype
+        expected = "int64" if is_integer(end + freq) else "float64"
+        assert result == expected
+
+        # GH 20976: linspace behavior defined from start/end/periods
+        index = interval_range(start=start, end=end, periods=5)
+        result = index.dtype.subtype
+        expected = "int64" if is_integer(start + end) else "float64"
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "start, end, expected",
+        [
+            (np.int8(1), np.int8(10), np.dtype("int8")),
+            (np.int8(1), np.float16(10), np.dtype("float64")),
+            (np.float32(1), np.float32(10), np.dtype("float32")),
+            (1, 10, np.dtype("int64")),
+            (1, 10.0, np.dtype("float64")),
+        ],
+    )
+    def test_interval_dtype(self, start, end, expected):
+        result = interval_range(start=start, end=end).dtype.subtype
+        assert result == expected
+
+    def test_interval_range_fractional_period(self):
+        # float value for periods
+        msg = "periods must be an integer, got 10.5"
+        ts = Timestamp("2024-03-25")
+        with pytest.raises(TypeError, match=msg):
+            interval_range(ts, periods=10.5)
+
+    def test_constructor_coverage(self):
+        # equivalent timestamp-like start/end
+        start, end = Timestamp("2017-01-01"), Timestamp("2017-01-15")
+        expected = interval_range(start=start, end=end)
+
+        result = interval_range(start=start.to_pydatetime(), end=end.to_pydatetime())
+        tm.assert_index_equal(result, expected)
+
+        result = interval_range(start=start.asm8, end=end.asm8)
+        tm.assert_index_equal(result, expected)
+
+        # equivalent freq with timestamp
+        equiv_freq = [
+            "D",
+            Day(),
+            Timedelta(days=1),
+            timedelta(days=1),
+            DateOffset(days=1),
+        ]
+        for freq in equiv_freq:
+            result = interval_range(start=start, end=end, freq=freq)
+            tm.assert_index_equal(result, expected)
+
+        # equivalent timedelta-like start/end
+        start, end = Timedelta(days=1), Timedelta(days=10)
+        expected = interval_range(start=start, end=end)
+
+        result = interval_range(start=start.to_pytimedelta(), end=end.to_pytimedelta())
+        tm.assert_index_equal(result, expected)
+
+        result = interval_range(start=start.asm8, end=end.asm8)
+        tm.assert_index_equal(result, expected)
+
+        # equivalent freq with timedelta
+        equiv_freq = ["D", Day(), Timedelta(days=1), timedelta(days=1)]
+        for freq in equiv_freq:
+            result = interval_range(start=start, end=end, freq=freq)
+            tm.assert_index_equal(result, expected)
+
+    def test_errors(self):
+        # not enough params
+        msg = (
+            "Of the four parameters: start, end, periods, and freq, "
+            "exactly three must be specified"
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range(start=0)
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range(end=5)
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range(periods=2)
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range()
+
+        # too many params
+        with pytest.raises(ValueError, match=msg):
+            interval_range(start=0, end=5, periods=6, freq=1.5)
+
+        # mixed units
+        msg = "start, end, freq need to be type compatible"
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=0, end=Timestamp("20130101"), freq=2)
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=0, end=Timedelta("1 day"), freq=2)
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=0, end=10, freq="D")
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=Timestamp("20130101"), end=10, freq="D")
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(
+                start=Timestamp("20130101"), end=Timedelta("1 day"), freq="D"
+            )
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(
+                start=Timestamp("20130101"), end=Timestamp("20130110"), freq=2
+            )
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=Timedelta("1 day"), end=10, freq="D")
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(
+                start=Timedelta("1 day"), end=Timestamp("20130110"), freq="D"
+            )
+
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=Timedelta("1 day"), end=Timedelta("10 days"), freq=2)
+
+        # invalid periods
+        msg = "periods must be an integer, got foo"
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=0, periods="foo")
+
+        # invalid start
+        msg = "start must be numeric or datetime-like, got foo"
+        with pytest.raises(ValueError, match=msg):
+            interval_range(start="foo", periods=10)
+
+        # invalid end
+        msg = r"end must be numeric or datetime-like, got \(0, 1\]"
+        with pytest.raises(ValueError, match=msg):
+            interval_range(end=Interval(0, 1), periods=10)
+
+        # invalid freq for datetime-like
+        msg = "freq must be numeric or convertible to DateOffset, got foo"
+        with pytest.raises(ValueError, match=msg):
+            interval_range(start=0, end=10, freq="foo")
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range(start=Timestamp("20130101"), periods=10, freq="foo")
+
+        with pytest.raises(ValueError, match=msg):
+            interval_range(end=Timedelta("1 day"), periods=10, freq="foo")
+
+        # mixed tz
+        start = Timestamp("2017-01-01", tz="US/Eastern")
+        end = Timestamp("2017-01-07", tz="US/Pacific")
+        msg = "Start and end cannot both be tz-aware with different timezones"
+        with pytest.raises(TypeError, match=msg):
+            interval_range(start=start, end=end)
+
+    def test_float_freq(self):
+        # GH 54477
+        result = interval_range(0, 1, freq=0.1)
+        expected = IntervalIndex.from_breaks([0 + 0.1 * n for n in range(11)])
+        tm.assert_index_equal(result, expected)
+
+        result = interval_range(0, 1, freq=0.6)
+        expected = IntervalIndex.from_breaks([0, 0.6])
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/interval/test_interval_tree.py b/pandas/pandas/tests/indexes/interval/test_interval_tree.py
new file mode 100644
index 0000000000000..df9c3b390f660
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_interval_tree.py
@@ -0,0 +1,208 @@
+from itertools import permutations
+
+import numpy as np
+import pytest
+
+from pandas._libs.interval import IntervalTree
+from pandas.compat import (
+    IS64,
+    WASM,
+)
+
+import pandas._testing as tm
+
+
+def skipif_32bit(param):
+    """
+    Skip parameters in a parametrize on 32bit systems. Specifically used
+    here to skip leaf_size parameters related to GH 23440.
+    """
+    marks = pytest.mark.skipif(not IS64, reason="GH 23440: int type mismatch on 32bit")
+    return pytest.param(param, marks=marks)
+
+
+@pytest.fixture(params=[skipif_32bit(1), skipif_32bit(2), 10])
+def leaf_size(request):
+    """
+    Fixture to specify IntervalTree leaf_size parameter; to be used with the
+    tree fixture.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        np.arange(5, dtype="int64"),
+        np.arange(5, dtype="uint64"),
+        np.arange(5, dtype="float64"),
+        np.array([0, 1, 2, 3, 4, np.nan], dtype="float64"),
+    ]
+)
+def tree(request, leaf_size):
+    left = request.param
+    return IntervalTree(left, left + 2, leaf_size=leaf_size)
+
+
+class TestIntervalTree:
+    def test_get_indexer(self, tree):
+        result = tree.get_indexer(np.array([1.0, 5.5, 6.5]))
+        expected = np.array([0, 4, -1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+        with pytest.raises(
+            KeyError, match="'indexer does not intersect a unique set of intervals'"
+        ):
+            tree.get_indexer(np.array([3.0]))
+
+    @pytest.mark.parametrize(
+        "dtype, target_value, target_dtype",
+        [("int64", 2**63 + 1, "uint64"), ("uint64", -1, "int64")],
+    )
+    def test_get_indexer_overflow(self, dtype, target_value, target_dtype):
+        left, right = np.array([0, 1], dtype=dtype), np.array([1, 2], dtype=dtype)
+        tree = IntervalTree(left, right)
+
+        result = tree.get_indexer(np.array([target_value], dtype=target_dtype))
+        expected = np.array([-1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_non_unique(self, tree):
+        indexer, missing = tree.get_indexer_non_unique(np.array([1.0, 2.0, 6.5]))
+
+        result = indexer[:1]
+        expected = np.array([0], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.sort(indexer[1:3])
+        expected = np.array([0, 1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.sort(indexer[3:])
+        expected = np.array([-1], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = missing
+        expected = np.array([2], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, target_value, target_dtype",
+        [("int64", 2**63 + 1, "uint64"), ("uint64", -1, "int64")],
+    )
+    def test_get_indexer_non_unique_overflow(self, dtype, target_value, target_dtype):
+        left, right = np.array([0, 2], dtype=dtype), np.array([1, 3], dtype=dtype)
+        tree = IntervalTree(left, right)
+        target = np.array([target_value], dtype=target_dtype)
+
+        result_indexer, result_missing = tree.get_indexer_non_unique(target)
+        expected_indexer = np.array([-1], dtype="intp")
+        tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+        expected_missing = np.array([0], dtype="intp")
+        tm.assert_numpy_array_equal(result_missing, expected_missing)
+
+    @pytest.mark.parametrize("dtype", ["int64", "float64", "uint64"])
+    def test_duplicates(self, dtype):
+        left = np.array([0, 0, 0], dtype=dtype)
+        tree = IntervalTree(left, left + 1)
+
+        with pytest.raises(
+            KeyError, match="'indexer does not intersect a unique set of intervals'"
+        ):
+            tree.get_indexer(np.array([0.5]))
+
+        indexer, missing = tree.get_indexer_non_unique(np.array([0.5]))
+        result = np.sort(indexer)
+        expected = np.array([0, 1, 2], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = missing
+        expected = np.array([], dtype="intp")
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "leaf_size", [skipif_32bit(1), skipif_32bit(10), skipif_32bit(100), 10000]
+    )
+    def test_get_indexer_closed(self, closed, leaf_size):
+        x = np.arange(1000, dtype="float64")
+        found = x.astype("intp")
+        not_found = (-1 * np.ones(1000)).astype("intp")
+
+        tree = IntervalTree(x, x + 0.5, closed=closed, leaf_size=leaf_size)
+        tm.assert_numpy_array_equal(found, tree.get_indexer(x + 0.25))
+
+        expected = found if tree.closed_left else not_found
+        tm.assert_numpy_array_equal(expected, tree.get_indexer(x + 0.0))
+
+        expected = found if tree.closed_right else not_found
+        tm.assert_numpy_array_equal(expected, tree.get_indexer(x + 0.5))
+
+    @pytest.mark.parametrize(
+        "left, right, expected",
+        [
+            (np.array([0, 1, 4], dtype="int64"), np.array([2, 3, 5]), True),
+            (np.array([0, 1, 2], dtype="int64"), np.array([5, 4, 3]), True),
+            (np.array([0, 1, np.nan]), np.array([5, 4, np.nan]), True),
+            (np.array([0, 2, 4], dtype="int64"), np.array([1, 3, 5]), False),
+            (np.array([0, 2, np.nan]), np.array([1, 3, np.nan]), False),
+        ],
+    )
+    @pytest.mark.parametrize("order", (list(x) for x in permutations(range(3))))
+    def test_is_overlapping(self, closed, order, left, right, expected):
+        # GH 23309
+        tree = IntervalTree(left[order], right[order], closed=closed)
+        result = tree.is_overlapping
+        assert result is expected
+
+    @pytest.mark.parametrize("order", (list(x) for x in permutations(range(3))))
+    def test_is_overlapping_endpoints(self, closed, order):
+        """shared endpoints are marked as overlapping"""
+        # GH 23309
+        left, right = np.arange(3, dtype="int64"), np.arange(1, 4)
+        tree = IntervalTree(left[order], right[order], closed=closed)
+        result = tree.is_overlapping
+        expected = closed == "both"
+        assert result is expected
+
+    @pytest.mark.parametrize(
+        "left, right",
+        [
+            (np.array([], dtype="int64"), np.array([], dtype="int64")),
+            (np.array([0], dtype="int64"), np.array([1], dtype="int64")),
+            (np.array([np.nan]), np.array([np.nan])),
+            (np.array([np.nan] * 3), np.array([np.nan] * 3)),
+        ],
+    )
+    def test_is_overlapping_trivial(self, closed, left, right):
+        # GH 23309
+        tree = IntervalTree(left, right, closed=closed)
+        assert tree.is_overlapping is False
+
+    @pytest.mark.skipif(not IS64, reason="GH 23440")
+    def test_construction_overflow(self):
+        # GH 25485
+        left, right = np.arange(101, dtype="int64"), [np.iinfo(np.int64).max] * 101
+        tree = IntervalTree(left, right)
+
+        # pivot should be average of left/right medians
+        result = tree.root.pivot
+        expected = (50 + np.iinfo(np.int64).max) / 2
+        assert result == expected
+
+    @pytest.mark.xfail(WASM, reason="GH 23440")
+    @pytest.mark.parametrize(
+        "left, right, expected",
+        [
+            ([-np.inf, 1.0], [1.0, 2.0], 0.0),
+            ([-np.inf, -2.0], [-2.0, -1.0], -2.0),
+            ([-2.0, -1.0], [-1.0, np.inf], 0.0),
+            ([1.0, 2.0], [2.0, np.inf], 2.0),
+        ],
+    )
+    def test_inf_bound_infinite_recursion(self, left, right, expected):
+        # GH 46658
+
+        tree = IntervalTree(left * 101, right * 101)
+
+        result = tree.root.pivot
+        assert result == expected
diff --git a/pandas/pandas/tests/indexes/interval/test_join.py b/pandas/pandas/tests/indexes/interval/test_join.py
new file mode 100644
index 0000000000000..2f42c530a6686
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_join.py
@@ -0,0 +1,44 @@
+import pytest
+
+from pandas import (
+    IntervalIndex,
+    MultiIndex,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def range_index():
+    return RangeIndex(3, name="range_index")
+
+
+@pytest.fixture
+def interval_index():
+    return IntervalIndex.from_tuples(
+        [(0.0, 1.0), (1.0, 2.0), (1.5, 2.5)], name="interval_index"
+    )
+
+
+def test_join_overlapping_in_mi_to_same_intervalindex(range_index, interval_index):
+    #  GH-45661
+    multi_index = MultiIndex.from_product([interval_index, range_index])
+    result = multi_index.join(interval_index)
+
+    tm.assert_index_equal(result, multi_index)
+
+
+def test_join_overlapping_to_multiindex_with_same_interval(range_index, interval_index):
+    #  GH-45661
+    multi_index = MultiIndex.from_product([interval_index, range_index])
+    result = interval_index.join(multi_index)
+
+    tm.assert_index_equal(result, multi_index)
+
+
+def test_join_overlapping_interval_to_another_intervalindex(interval_index):
+    #  GH-45661
+    flipped_interval_index = interval_index[::-1]
+    result = interval_index.join(flipped_interval_index)
+
+    tm.assert_index_equal(result, interval_index)
diff --git a/pandas/pandas/tests/indexes/interval/test_pickle.py b/pandas/pandas/tests/indexes/interval/test_pickle.py
new file mode 100644
index 0000000000000..c52106e5f0786
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_pickle.py
@@ -0,0 +1,10 @@
+from pandas import IntervalIndex
+import pandas._testing as tm
+
+
+class TestPickle:
+    def test_pickle_round_trip_closed(self, closed):
+        # https://github.com/pandas-dev/pandas/issues/35658
+        idx = IntervalIndex.from_tuples([(1, 2), (2, 3)], closed=closed)
+        result = tm.round_trip_pickle(idx)
+        tm.assert_index_equal(result, idx)
diff --git a/pandas/pandas/tests/indexes/interval/test_setops.py b/pandas/pandas/tests/indexes/interval/test_setops.py
new file mode 100644
index 0000000000000..1b0816a9405cb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/interval/test_setops.py
@@ -0,0 +1,208 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    IntervalIndex,
+    Timestamp,
+    interval_range,
+)
+import pandas._testing as tm
+
+
+def monotonic_index(start, end, dtype="int64", closed="right"):
+    return IntervalIndex.from_breaks(np.arange(start, end, dtype=dtype), closed=closed)
+
+
+def empty_index(dtype="int64", closed="right"):
+    return IntervalIndex(np.array([], dtype=dtype), closed=closed)
+
+
+class TestIntervalIndex:
+    def test_union(self, closed, sort):
+        index = monotonic_index(0, 11, closed=closed)
+        other = monotonic_index(5, 13, closed=closed)
+
+        expected = monotonic_index(0, 13, closed=closed)
+        result = index[::-1].union(other, sort=sort)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        result = other[::-1].union(index, sort=sort)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        tm.assert_index_equal(index.union(index, sort=sort), index)
+        tm.assert_index_equal(index.union(index[:1], sort=sort), index)
+
+    def test_union_empty_result(self, closed, sort):
+        # GH 19101: empty result, same dtype
+        index = empty_index(dtype="int64", closed=closed)
+        result = index.union(index, sort=sort)
+        tm.assert_index_equal(result, index)
+
+        # GH 19101: empty result, different numeric dtypes -> common dtype is f8
+        other = empty_index(dtype="float64", closed=closed)
+        result = index.union(other, sort=sort)
+        expected = other
+        tm.assert_index_equal(result, expected)
+
+        other = index.union(index, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        other = empty_index(dtype="uint64", closed=closed)
+        result = index.union(other, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection(self, closed, sort):
+        index = monotonic_index(0, 11, closed=closed)
+        other = monotonic_index(5, 13, closed=closed)
+
+        expected = monotonic_index(5, 11, closed=closed)
+        result = index[::-1].intersection(other, sort=sort)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        result = other[::-1].intersection(index, sort=sort)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        tm.assert_index_equal(index.intersection(index, sort=sort), index)
+
+        # GH 26225: nested intervals
+        index = IntervalIndex.from_tuples([(1, 2), (1, 3), (1, 4), (0, 2)])
+        other = IntervalIndex.from_tuples([(1, 2), (1, 3)])
+        expected = IntervalIndex.from_tuples([(1, 2), (1, 3)])
+        result = index.intersection(other)
+        tm.assert_index_equal(result, expected)
+
+        # GH 26225
+        index = IntervalIndex.from_tuples([(0, 3), (0, 2)])
+        other = IntervalIndex.from_tuples([(0, 2), (1, 3)])
+        expected = IntervalIndex.from_tuples([(0, 2)])
+        result = index.intersection(other)
+        tm.assert_index_equal(result, expected)
+
+        # GH 26225: duplicate nan element
+        index = IntervalIndex([np.nan, np.nan])
+        other = IntervalIndex([np.nan])
+        expected = IntervalIndex([np.nan])
+        result = index.intersection(other)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_empty_result(self, closed, sort):
+        index = monotonic_index(0, 11, closed=closed)
+
+        # GH 19101: empty result, same dtype
+        other = monotonic_index(300, 314, closed=closed)
+        expected = empty_index(dtype="int64", closed=closed)
+        result = index.intersection(other, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        # GH 19101: empty result, different numeric dtypes -> common dtype is float64
+        other = monotonic_index(300, 314, dtype="float64", closed=closed)
+        result = index.intersection(other, sort=sort)
+        expected = other[:0]
+        tm.assert_index_equal(result, expected)
+
+        other = monotonic_index(300, 314, dtype="uint64", closed=closed)
+        result = index.intersection(other, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_duplicates(self):
+        # GH#38743
+        index = IntervalIndex.from_tuples([(1, 2), (1, 2), (2, 3), (3, 4)])
+        other = IntervalIndex.from_tuples([(1, 2), (2, 3)])
+        expected = IntervalIndex.from_tuples([(1, 2), (2, 3)])
+        result = index.intersection(other)
+        tm.assert_index_equal(result, expected)
+
+    def test_difference(self, closed, sort):
+        index = IntervalIndex.from_arrays([1, 0, 3, 2], [1, 2, 3, 4], closed=closed)
+        result = index.difference(index[:1], sort=sort)
+        expected = index[1:]
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+        # GH 19101: empty result, same dtype
+        result = index.difference(index, sort=sort)
+        expected = empty_index(dtype="int64", closed=closed)
+        tm.assert_index_equal(result, expected)
+
+        # GH 19101: empty result, different dtypes
+        other = IntervalIndex.from_arrays(
+            index.left.astype("float64"), index.right, closed=closed
+        )
+        result = index.difference(other, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_symmetric_difference(self, closed, sort):
+        index = monotonic_index(0, 11, closed=closed)
+        result = index[1:].symmetric_difference(index[:-1], sort=sort)
+        expected = IntervalIndex([index[0], index[-1]])
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        # GH 19101: empty result, same dtype
+        result = index.symmetric_difference(index, sort=sort)
+        expected = empty_index(dtype="int64", closed=closed)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+
+        # GH 19101: empty result, different dtypes
+        other = IntervalIndex.from_arrays(
+            index.left.astype("float64"), index.right, closed=closed
+        )
+        result = index.symmetric_difference(other, sort=sort)
+        expected = empty_index(dtype="float64", closed=closed)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:'<' not supported between:RuntimeWarning")
+    @pytest.mark.parametrize(
+        "op_name", ["union", "intersection", "difference", "symmetric_difference"]
+    )
+    def test_set_incompatible_types(self, closed, op_name, sort):
+        index = monotonic_index(0, 11, closed=closed)
+        set_op = getattr(index, op_name)
+
+        # TODO: standardize return type of non-union setops type(self vs other)
+        # non-IntervalIndex
+        if op_name == "difference":
+            expected = index
+        else:
+            expected = getattr(index.astype("O"), op_name)(Index([1, 2, 3]))
+        result = set_op(Index([1, 2, 3]), sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        # mixed closed -> cast to object
+        for other_closed in {"right", "left", "both", "neither"} - {closed}:
+            other = monotonic_index(0, 11, closed=other_closed)
+            expected = getattr(index.astype(object), op_name)(other, sort=sort)
+            if op_name == "difference":
+                expected = index
+            result = set_op(other, sort=sort)
+            tm.assert_index_equal(result, expected)
+
+        # GH 19016: incompatible dtypes -> cast to object
+        other = interval_range(Timestamp("20180101"), periods=9, closed=closed)
+        expected = getattr(index.astype(object), op_name)(other, sort=sort)
+        if op_name == "difference":
+            expected = index
+        result = set_op(other, sort=sort)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/__init__.py b/pandas/pandas/tests/indexes/multi/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/multi/conftest.py b/pandas/pandas/tests/indexes/multi/conftest.py
new file mode 100644
index 0000000000000..15062aee56e3a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/conftest.py
@@ -0,0 +1,27 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    MultiIndex,
+)
+
+
+# Note: identical the "multi" entry in the top-level "index" fixture
+@pytest.fixture
+def idx():
+    # a MultiIndex used to test the general functionality of the
+    # general functionality of this object
+    major_axis = Index(["foo", "bar", "baz", "qux"])
+    minor_axis = Index(["one", "two"])
+
+    major_codes = np.array([0, 0, 1, 2, 3, 3])
+    minor_codes = np.array([0, 1, 0, 1, 0, 1])
+    index_names = ["first", "second"]
+    mi = MultiIndex(
+        levels=[major_axis, minor_axis],
+        codes=[major_codes, minor_codes],
+        names=index_names,
+        verify_integrity=False,
+    )
+    return mi
diff --git a/pandas/pandas/tests/indexes/multi/test_analytics.py b/pandas/pandas/tests/indexes/multi/test_analytics.py
new file mode 100644
index 0000000000000..87f1439db5fc8
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_analytics.py
@@ -0,0 +1,263 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+def test_infer_objects(idx):
+    with pytest.raises(NotImplementedError, match="to_frame"):
+        idx.infer_objects()
+
+
+def test_shift(idx):
+    # GH8083 test the base class for shift
+    msg = (
+        "This method is only implemented for DatetimeIndex, PeriodIndex and "
+        "TimedeltaIndex; Got type MultiIndex"
+    )
+    with pytest.raises(NotImplementedError, match=msg):
+        idx.shift(1)
+    with pytest.raises(NotImplementedError, match=msg):
+        idx.shift(1, 2)
+
+
+def test_groupby(idx):
+    groups = idx.groupby(np.array([1, 1, 1, 2, 2, 2]))
+    labels = idx.tolist()
+    exp = {1: labels[:3], 2: labels[3:]}
+    tm.assert_dict_equal(groups, exp)
+
+    # GH5620
+    groups = idx.groupby(idx)
+    exp = {key: [key] for key in idx}
+    tm.assert_dict_equal(groups, exp)
+
+
+def test_truncate_multiindex():
+    # GH 34564 for MultiIndex level names check
+    major_axis = Index(list(range(4)))
+    minor_axis = Index(list(range(2)))
+
+    major_codes = np.array([0, 0, 1, 2, 3, 3])
+    minor_codes = np.array([0, 1, 0, 1, 0, 1])
+
+    index = MultiIndex(
+        levels=[major_axis, minor_axis],
+        codes=[major_codes, minor_codes],
+        names=["L1", "L2"],
+    )
+
+    result = index.truncate(before=1)
+    assert "foo" not in result.levels[0]
+    assert 1 in result.levels[0]
+    assert index.names == result.names
+
+    result = index.truncate(after=1)
+    assert 2 not in result.levels[0]
+    assert 1 in result.levels[0]
+    assert index.names == result.names
+
+    result = index.truncate(before=1, after=2)
+    assert len(result.levels[0]) == 2
+    assert index.names == result.names
+
+    msg = "after < before"
+    with pytest.raises(ValueError, match=msg):
+        index.truncate(3, 1)
+
+
+# TODO: reshape
+
+
+def test_reorder_levels(idx):
+    # this blows up
+    with pytest.raises(IndexError, match="^Too many levels"):
+        idx.reorder_levels([2, 1, 0])
+
+
+def test_numpy_repeat():
+    reps = 2
+    numbers = [1, 2, 3]
+    names = np.array(["foo", "bar"])
+
+    m = MultiIndex.from_product([numbers, names], names=names)
+    expected = MultiIndex.from_product([numbers, names.repeat(reps)], names=names)
+    tm.assert_index_equal(np.repeat(m, reps), expected)
+
+    msg = "the 'axis' parameter is not supported"
+    with pytest.raises(ValueError, match=msg):
+        np.repeat(m, reps, axis=1)
+
+
+def test_append_mixed_dtypes():
+    # GH 13660
+    dti = date_range("2011-01-01", freq="ME", periods=3)
+    dti_tz = date_range("2011-01-01", freq="ME", periods=3, tz="US/Eastern")
+    pi = period_range("2011-01", freq="M", periods=3)
+
+    mi = MultiIndex.from_arrays(
+        [[1, 2, 3], [1.1, np.nan, 3.3], ["a", "b", "c"], dti, dti_tz, pi]
+    )
+    assert mi.nlevels == 6
+
+    res = mi.append(mi)
+    exp = MultiIndex.from_arrays(
+        [
+            [1, 2, 3, 1, 2, 3],
+            [1.1, np.nan, 3.3, 1.1, np.nan, 3.3],
+            ["a", "b", "c", "a", "b", "c"],
+            dti.append(dti),
+            dti_tz.append(dti_tz),
+            pi.append(pi),
+        ]
+    )
+    tm.assert_index_equal(res, exp)
+
+    other = MultiIndex.from_arrays(
+        [
+            ["x", "y", "z"],
+            ["x", "y", "z"],
+            ["x", "y", "z"],
+            ["x", "y", "z"],
+            ["x", "y", "z"],
+            ["x", "y", "z"],
+        ]
+    )
+
+    res = mi.append(other)
+    exp = MultiIndex.from_arrays(
+        [
+            [1, 2, 3, "x", "y", "z"],
+            [1.1, np.nan, 3.3, "x", "y", "z"],
+            ["a", "b", "c", "x", "y", "z"],
+            dti.append(Index(["x", "y", "z"])),
+            dti_tz.append(Index(["x", "y", "z"])),
+            pi.append(Index(["x", "y", "z"])),
+        ]
+    )
+    tm.assert_index_equal(res, exp)
+
+
+def test_iter(idx):
+    result = list(idx)
+    expected = [
+        ("foo", "one"),
+        ("foo", "two"),
+        ("bar", "one"),
+        ("baz", "two"),
+        ("qux", "one"),
+        ("qux", "two"),
+    ]
+    assert result == expected
+
+
+def test_sub(idx):
+    first = idx
+
+    # - now raises (previously was set op difference)
+    msg = "cannot perform __sub__ with this index type: MultiIndex"
+    with pytest.raises(TypeError, match=msg):
+        first - idx[-3:]
+    with pytest.raises(TypeError, match=msg):
+        idx[-3:] - first
+    with pytest.raises(TypeError, match=msg):
+        idx[-3:] - first.tolist()
+    msg = "cannot perform __rsub__ with this index type: MultiIndex"
+    with pytest.raises(TypeError, match=msg):
+        first.tolist() - idx[-3:]
+
+
+def test_map(idx):
+    # callable
+    index = idx
+
+    result = index.map(lambda x: x)
+    tm.assert_index_equal(result, index)
+
+
+@pytest.mark.parametrize(
+    "mapper",
+    [
+        lambda values, idx: {i: e for e, i in zip(values, idx)},
+        lambda values, idx: pd.Series(values, idx),
+    ],
+)
+def test_map_dictlike(idx, mapper):
+    identity = mapper(idx.values, idx)
+
+    # we don't infer to uint64 dtype for a dict
+    if idx.dtype == np.uint64 and isinstance(identity, dict):
+        expected = idx.astype("int64")
+    else:
+        expected = idx
+
+    result = idx.map(identity)
+    tm.assert_index_equal(result, expected)
+
+    # empty mappable
+    expected = Index([np.nan] * len(idx))
+    result = idx.map(mapper(expected, idx))
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        np.exp,
+        np.exp2,
+        np.expm1,
+        np.log,
+        np.log2,
+        np.log10,
+        np.log1p,
+        np.sqrt,
+        np.sin,
+        np.cos,
+        np.tan,
+        np.arcsin,
+        np.arccos,
+        np.arctan,
+        np.sinh,
+        np.cosh,
+        np.tanh,
+        np.arcsinh,
+        np.arccosh,
+        np.arctanh,
+        np.deg2rad,
+        np.rad2deg,
+    ],
+    ids=lambda func: func.__name__,
+)
+def test_numpy_ufuncs(idx, func):
+    # test ufuncs of numpy. see:
+    # https://numpy.org/doc/stable/reference/ufuncs.html
+
+    expected_exception = TypeError
+    msg = (
+        "loop of ufunc does not support argument 0 of type tuple which "
+        f"has no callable {func.__name__} method"
+    )
+    with pytest.raises(expected_exception, match=msg):
+        func(idx)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [np.isfinite, np.isinf, np.isnan, np.signbit],
+    ids=lambda func: func.__name__,
+)
+def test_numpy_type_funcs(idx, func):
+    msg = (
+        f"ufunc '{func.__name__}' not supported for the input types, and the inputs "
+        "could not be safely coerced to any supported types according to "
+        "the casting rule ''safe''"
+    )
+    with pytest.raises(TypeError, match=msg):
+        func(idx)
diff --git a/pandas/pandas/tests/indexes/multi/test_astype.py b/pandas/pandas/tests/indexes/multi/test_astype.py
new file mode 100644
index 0000000000000..29908537fbe59
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_astype.py
@@ -0,0 +1,30 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas._testing as tm
+
+
+def test_astype(idx):
+    expected = idx.copy()
+    actual = idx.astype("O")
+    tm.assert_copy(actual.levels, expected.levels)
+    tm.assert_copy(actual.codes, expected.codes)
+    assert actual.names == list(expected.names)
+
+    with pytest.raises(TypeError, match="^Setting.*dtype.*object"):
+        idx.astype(np.dtype(int))
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_astype_category(idx, ordered):
+    # GH 18630
+    msg = "> 1 ndim Categorical are not supported at this time"
+    with pytest.raises(NotImplementedError, match=msg):
+        idx.astype(CategoricalDtype(ordered=ordered))
+
+    if ordered is False:
+        # dtype='category' defaults to ordered=False, so only test once
+        with pytest.raises(NotImplementedError, match=msg):
+            idx.astype("category")
diff --git a/pandas/pandas/tests/indexes/multi/test_compat.py b/pandas/pandas/tests/indexes/multi/test_compat.py
new file mode 100644
index 0000000000000..27a8c6e9b7158
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_compat.py
@@ -0,0 +1,122 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import MultiIndex
+import pandas._testing as tm
+
+
+def test_numeric_compat(idx):
+    with pytest.raises(TypeError, match="cannot perform __mul__"):
+        idx * 1
+
+    with pytest.raises(TypeError, match="cannot perform __rmul__"):
+        1 * idx
+
+    div_err = "cannot perform __truediv__"
+    with pytest.raises(TypeError, match=div_err):
+        idx / 1
+
+    div_err = div_err.replace(" __", " __r")
+    with pytest.raises(TypeError, match=div_err):
+        1 / idx
+
+    with pytest.raises(TypeError, match="cannot perform __floordiv__"):
+        idx // 1
+
+    with pytest.raises(TypeError, match="cannot perform __rfloordiv__"):
+        1 // idx
+
+
+@pytest.mark.parametrize("method", ["all", "any", "__invert__"])
+def test_logical_compat(idx, method):
+    msg = f"cannot perform {method}"
+
+    with pytest.raises(TypeError, match=msg):
+        getattr(idx, method)()
+
+
+def test_inplace_mutation_resets_values():
+    levels = [["a", "b", "c"], [4]]
+    levels2 = [[1, 2, 3], ["a"]]
+    codes = [[0, 1, 0, 2, 2, 0], [0, 0, 0, 0, 0, 0]]
+
+    mi1 = MultiIndex(levels=levels, codes=codes)
+    mi2 = MultiIndex(levels=levels2, codes=codes)
+
+    # instantiating MultiIndex should not access/cache _.values
+    assert "_values" not in mi1._cache
+    assert "_values" not in mi2._cache
+
+    vals = mi1.values.copy()
+    vals2 = mi2.values.copy()
+
+    # accessing .values should cache ._values
+    assert mi1._values is mi1._cache["_values"]
+    assert mi1.values is mi1._cache["_values"]
+    assert isinstance(mi1._cache["_values"], np.ndarray)
+
+    # Make sure level setting works
+    new_vals = mi1.set_levels(levels2).values
+    tm.assert_almost_equal(vals2, new_vals)
+
+    #  Doesn't drop _values from _cache [implementation detail]
+    tm.assert_almost_equal(mi1._cache["_values"], vals)
+
+    # ...and values is still same too
+    tm.assert_almost_equal(mi1.values, vals)
+
+    # Make sure label setting works too
+    codes2 = [[0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0]]
+    exp_values = np.empty((6,), dtype=object)
+    exp_values[:] = [(1, "a")] * 6
+
+    # Must be 1d array of tuples
+    assert exp_values.shape == (6,)
+
+    new_mi = mi2.set_codes(codes2)
+    assert "_values" not in new_mi._cache
+    new_values = new_mi.values
+    assert "_values" in new_mi._cache
+
+    # Shouldn't change cache
+    tm.assert_almost_equal(mi2._cache["_values"], vals2)
+
+    # Should have correct values
+    tm.assert_almost_equal(exp_values, new_values)
+
+
+def test_boxable_categorical_values():
+    cat = pd.Categorical(pd.date_range("2012-01-01", periods=3, freq="h"))
+    result = MultiIndex.from_product([["a", "b", "c"], cat]).values
+    expected = pd.Series(
+        [
+            ("a", pd.Timestamp("2012-01-01 00:00:00")),
+            ("a", pd.Timestamp("2012-01-01 01:00:00")),
+            ("a", pd.Timestamp("2012-01-01 02:00:00")),
+            ("b", pd.Timestamp("2012-01-01 00:00:00")),
+            ("b", pd.Timestamp("2012-01-01 01:00:00")),
+            ("b", pd.Timestamp("2012-01-01 02:00:00")),
+            ("c", pd.Timestamp("2012-01-01 00:00:00")),
+            ("c", pd.Timestamp("2012-01-01 01:00:00")),
+            ("c", pd.Timestamp("2012-01-01 02:00:00")),
+        ]
+    ).values
+    tm.assert_numpy_array_equal(result, expected)
+    result = pd.DataFrame({"a": ["a", "b", "c"], "b": cat, "c": np.array(cat)}).values
+    expected = pd.DataFrame(
+        {
+            "a": ["a", "b", "c"],
+            "b": [
+                pd.Timestamp("2012-01-01 00:00:00"),
+                pd.Timestamp("2012-01-01 01:00:00"),
+                pd.Timestamp("2012-01-01 02:00:00"),
+            ],
+            "c": [
+                pd.Timestamp("2012-01-01 00:00:00"),
+                pd.Timestamp("2012-01-01 01:00:00"),
+                pd.Timestamp("2012-01-01 02:00:00"),
+            ],
+        }
+    ).values
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_constructors.py b/pandas/pandas/tests/indexes/multi/test_constructors.py
new file mode 100644
index 0000000000000..b2867d4ac8e68
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_constructors.py
@@ -0,0 +1,859 @@
+from datetime import (
+    date,
+    datetime,
+)
+import itertools
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_constructor_single_level():
+    result = MultiIndex(
+        levels=[["foo", "bar", "baz", "qux"]], codes=[[0, 1, 2, 3]], names=["first"]
+    )
+    assert isinstance(result, MultiIndex)
+    expected = Index(["foo", "bar", "baz", "qux"], name="first")
+    tm.assert_index_equal(result.levels[0], expected)
+    assert result.names == ["first"]
+
+
+def test_constructor_no_levels():
+    msg = "non-zero number of levels/codes"
+    with pytest.raises(ValueError, match=msg):
+        MultiIndex(levels=[], codes=[])
+
+    msg = "Must pass both levels and codes"
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex(levels=[])
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex(codes=[])
+
+
+def test_constructor_nonhashable_names():
+    # GH 20527
+    levels = [[1, 2], ["one", "two"]]
+    codes = [[0, 0, 1, 1], [0, 1, 0, 1]]
+    names = (["foo"], ["bar"])
+    msg = r"MultiIndex\.name must be a hashable type"
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex(levels=levels, codes=codes, names=names)
+
+    # With .rename()
+    mi = MultiIndex(
+        levels=[[1, 2], ["one", "two"]],
+        codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+        names=("foo", "bar"),
+    )
+    renamed = [["fooo"], ["barr"]]
+    with pytest.raises(TypeError, match=msg):
+        mi.rename(names=renamed)
+
+    # With .set_names()
+    with pytest.raises(TypeError, match=msg):
+        mi.set_names(names=renamed)
+
+
+def test_constructor_mismatched_codes_levels(idx):
+    codes = [np.array([1]), np.array([2]), np.array([3])]
+    levels = ["a"]
+
+    msg = "Length of levels and codes must be the same"
+    with pytest.raises(ValueError, match=msg):
+        MultiIndex(levels=levels, codes=codes)
+
+    length_error = (
+        r"On level 0, code max \(3\) >= length of level \(1\)\. "
+        "NOTE: this index is in an inconsistent state"
+    )
+    label_error = r"Unequal code lengths: \[4, 2\]"
+    code_value_error = r"On level 0, code value \(-2\) < -1"
+
+    # important to check that it's looking at the right thing.
+    with pytest.raises(ValueError, match=length_error):
+        MultiIndex(levels=[["a"], ["b"]], codes=[[0, 1, 2, 3], [0, 3, 4, 1]])
+
+    with pytest.raises(ValueError, match=label_error):
+        MultiIndex(levels=[["a"], ["b"]], codes=[[0, 0, 0, 0], [0, 0]])
+
+    # external API
+    with pytest.raises(ValueError, match=length_error):
+        idx.copy().set_levels([["a"], ["b"]])
+
+    with pytest.raises(ValueError, match=label_error):
+        idx.copy().set_codes([[0, 0, 0, 0], [0, 0]])
+
+    # test set_codes with verify_integrity=False
+    # the setting should not raise any value error
+    idx.copy().set_codes(codes=[[0, 0, 0, 0], [0, 0]], verify_integrity=False)
+
+    # code value smaller than -1
+    with pytest.raises(ValueError, match=code_value_error):
+        MultiIndex(levels=[["a"], ["b"]], codes=[[0, -2], [0, 0]])
+
+
+def test_na_levels():
+    # GH26408
+    # test if codes are re-assigned value -1 for levels
+    # with missing values (NaN, NaT, None)
+    result = MultiIndex(
+        levels=[[np.nan, None, pd.NaT, 128, 2]], codes=[[0, -1, 1, 2, 3, 4]]
+    )
+    expected = MultiIndex(
+        levels=[[np.nan, None, pd.NaT, 128, 2]], codes=[[-1, -1, -1, -1, 3, 4]]
+    )
+    tm.assert_index_equal(result, expected)
+
+    result = MultiIndex(
+        levels=[[np.nan, "s", pd.NaT, 128, None]], codes=[[0, -1, 1, 2, 3, 4]]
+    )
+    expected = MultiIndex(
+        levels=[[np.nan, "s", pd.NaT, 128, None]], codes=[[-1, -1, 1, -1, 3, -1]]
+    )
+    tm.assert_index_equal(result, expected)
+
+    # verify set_levels and set_codes
+    result = MultiIndex(
+        levels=[[1, 2, 3, 4, 5]], codes=[[0, -1, 1, 2, 3, 4]]
+    ).set_levels([[np.nan, "s", pd.NaT, 128, None]])
+    tm.assert_index_equal(result, expected)
+
+    result = MultiIndex(
+        levels=[[np.nan, "s", pd.NaT, 128, None]], codes=[[1, 2, 2, 2, 2, 2]]
+    ).set_codes([[0, -1, 1, 2, 3, 4]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_copy_in_constructor():
+    levels = np.array(["a", "b", "c"])
+    codes = np.array([1, 1, 2, 0, 0, 1, 1])
+    val = codes[0]
+    mi = MultiIndex(levels=[levels, levels], codes=[codes, codes], copy=True)
+    assert mi.codes[0][0] == val
+    codes[0] = 15
+    assert mi.codes[0][0] == val
+    val = levels[0]
+    levels[0] = "PANDA"
+    assert mi.levels[0][0] == val
+
+
+# ----------------------------------------------------------------------------
+# from_arrays
+# ----------------------------------------------------------------------------
+def test_from_arrays(idx):
+    arrays = [
+        np.asarray(lev).take(level_codes)
+        for lev, level_codes in zip(idx.levels, idx.codes)
+    ]
+
+    # list of arrays as input
+    result = MultiIndex.from_arrays(arrays, names=idx.names)
+    tm.assert_index_equal(result, idx)
+
+    # infer correctly
+    result = MultiIndex.from_arrays([[pd.NaT, Timestamp("20130101")], ["a", "b"]])
+    assert result.levels[0].equals(Index([Timestamp("20130101")]))
+    assert result.levels[1].equals(Index(["a", "b"]))
+
+
+def test_from_arrays_iterator(idx):
+    # GH 18434
+    arrays = [
+        np.asarray(lev).take(level_codes)
+        for lev, level_codes in zip(idx.levels, idx.codes)
+    ]
+
+    # iterator as input
+    result = MultiIndex.from_arrays(iter(arrays), names=idx.names)
+    tm.assert_index_equal(result, idx)
+
+    # invalid iterator input
+    msg = "Input must be a list / sequence of array-likes."
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_arrays(0)
+
+
+def test_from_arrays_tuples(idx):
+    arrays = tuple(
+        tuple(np.asarray(lev).take(level_codes))
+        for lev, level_codes in zip(idx.levels, idx.codes)
+    )
+
+    # tuple of tuples as input
+    result = MultiIndex.from_arrays(arrays, names=idx.names)
+    tm.assert_index_equal(result, idx)
+
+
+@pytest.mark.parametrize(
+    ("idx1", "idx2"),
+    [
+        (
+            pd.period_range("2011-01-01", freq="D", periods=3),
+            pd.period_range("2015-01-01", freq="h", periods=3),
+        ),
+        (
+            date_range("2015-01-01 10:00", freq="D", periods=3, tz="US/Eastern"),
+            date_range("2015-01-01 10:00", freq="h", periods=3, tz="Asia/Tokyo"),
+        ),
+        (
+            pd.timedelta_range("1 days", freq="D", periods=3),
+            pd.timedelta_range("2 hours", freq="h", periods=3),
+        ),
+    ],
+)
+def test_from_arrays_index_series_period_datetimetz_and_timedelta(idx1, idx2):
+    result = MultiIndex.from_arrays([idx1, idx2])
+    tm.assert_index_equal(result.get_level_values(0), idx1)
+    tm.assert_index_equal(result.get_level_values(1), idx2)
+
+    result2 = MultiIndex.from_arrays([Series(idx1), Series(idx2)])
+    tm.assert_index_equal(result2.get_level_values(0), idx1)
+    tm.assert_index_equal(result2.get_level_values(1), idx2)
+
+    tm.assert_index_equal(result, result2)
+
+
+def test_from_arrays_index_datetimelike_mixed():
+    idx1 = date_range("2015-01-01 10:00", freq="D", periods=3, tz="US/Eastern")
+    idx2 = date_range("2015-01-01 10:00", freq="h", periods=3)
+    idx3 = pd.timedelta_range("1 days", freq="D", periods=3)
+    idx4 = pd.period_range("2011-01-01", freq="D", periods=3)
+
+    result = MultiIndex.from_arrays([idx1, idx2, idx3, idx4])
+    tm.assert_index_equal(result.get_level_values(0), idx1)
+    tm.assert_index_equal(result.get_level_values(1), idx2)
+    tm.assert_index_equal(result.get_level_values(2), idx3)
+    tm.assert_index_equal(result.get_level_values(3), idx4)
+
+    result2 = MultiIndex.from_arrays(
+        [Series(idx1), Series(idx2), Series(idx3), Series(idx4)]
+    )
+    tm.assert_index_equal(result2.get_level_values(0), idx1)
+    tm.assert_index_equal(result2.get_level_values(1), idx2)
+    tm.assert_index_equal(result2.get_level_values(2), idx3)
+    tm.assert_index_equal(result2.get_level_values(3), idx4)
+
+    tm.assert_index_equal(result, result2)
+
+
+def test_from_arrays_index_series_categorical():
+    # GH13743
+    idx1 = pd.CategoricalIndex(list("abcaab"), categories=list("bac"), ordered=False)
+    idx2 = pd.CategoricalIndex(list("abcaab"), categories=list("bac"), ordered=True)
+
+    result = MultiIndex.from_arrays([idx1, idx2])
+    tm.assert_index_equal(result.get_level_values(0), idx1)
+    tm.assert_index_equal(result.get_level_values(1), idx2)
+
+    result2 = MultiIndex.from_arrays([Series(idx1), Series(idx2)])
+    tm.assert_index_equal(result2.get_level_values(0), idx1)
+    tm.assert_index_equal(result2.get_level_values(1), idx2)
+
+    result3 = MultiIndex.from_arrays([idx1.values, idx2.values])
+    tm.assert_index_equal(result3.get_level_values(0), idx1)
+    tm.assert_index_equal(result3.get_level_values(1), idx2)
+
+
+def test_from_arrays_empty():
+    # 0 levels
+    msg = "Must pass non-zero number of levels/codes"
+    with pytest.raises(ValueError, match=msg):
+        MultiIndex.from_arrays(arrays=[])
+
+    # 1 level
+    result = MultiIndex.from_arrays(arrays=[[]], names=["A"])
+    assert isinstance(result, MultiIndex)
+    expected = Index([], name="A")
+    tm.assert_index_equal(result.levels[0], expected)
+    assert result.names == ["A"]
+
+    # N levels
+    for N in [2, 3]:
+        arrays = [[]] * N
+        names = list("ABC")[:N]
+        result = MultiIndex.from_arrays(arrays=arrays, names=names)
+        expected = MultiIndex(levels=[[]] * N, codes=[[]] * N, names=names)
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "invalid_sequence_of_arrays",
+    [
+        1,
+        [1],
+        [1, 2],
+        [[1], 2],
+        [1, [2]],
+        "a",
+        ["a"],
+        ["a", "b"],
+        [["a"], "b"],
+        (1,),
+        (1, 2),
+        ([1], 2),
+        (1, [2]),
+        ("a",),
+        ("a", "b"),
+        (["a"], "b"),
+        [(1,), 2],
+        [1, (2,)],
+        [("a",), "b"],
+        ((1,), 2),
+        (1, (2,)),
+        (("a",), "b"),
+    ],
+)
+def test_from_arrays_invalid_input(invalid_sequence_of_arrays):
+    msg = "Input must be a list / sequence of array-likes"
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_arrays(arrays=invalid_sequence_of_arrays)
+
+
+@pytest.mark.parametrize(
+    "idx1, idx2", [([1, 2, 3], ["a", "b"]), ([], ["a", "b"]), ([1, 2, 3], [])]
+)
+def test_from_arrays_different_lengths(idx1, idx2):
+    # see gh-13599
+    msg = "^all arrays must be same length$"
+    with pytest.raises(ValueError, match=msg):
+        MultiIndex.from_arrays([idx1, idx2])
+
+
+def test_from_arrays_respects_none_names():
+    # GH27292
+    a = Series([1, 2, 3], name="foo")
+    b = Series(["a", "b", "c"], name="bar")
+
+    result = MultiIndex.from_arrays([a, b], names=None)
+    expected = MultiIndex(
+        levels=[[1, 2, 3], ["a", "b", "c"]], codes=[[0, 1, 2], [0, 1, 2]], names=None
+    )
+
+    tm.assert_index_equal(result, expected)
+
+
+# ----------------------------------------------------------------------------
+# from_tuples
+# ----------------------------------------------------------------------------
+def test_from_tuples():
+    msg = "Cannot infer number of levels from empty list"
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_tuples([])
+
+    expected = MultiIndex(
+        levels=[[1, 3], [2, 4]], codes=[[0, 1], [0, 1]], names=["a", "b"]
+    )
+
+    # input tuples
+    result = MultiIndex.from_tuples(((1, 2), (3, 4)), names=["a", "b"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_from_tuples_iterator():
+    # GH 18434
+    # input iterator for tuples
+    expected = MultiIndex(
+        levels=[[1, 3], [2, 4]], codes=[[0, 1], [0, 1]], names=["a", "b"]
+    )
+
+    result = MultiIndex.from_tuples(zip([1, 3], [2, 4]), names=["a", "b"])
+    tm.assert_index_equal(result, expected)
+
+    # input non-iterables
+    msg = "Input must be a list / sequence of tuple-likes."
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_tuples(0)
+
+
+def test_from_tuples_empty():
+    # GH 16777
+    result = MultiIndex.from_tuples([], names=["a", "b"])
+    expected = MultiIndex.from_arrays(arrays=[[], []], names=["a", "b"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_from_tuples_index_values(idx):
+    result = MultiIndex.from_tuples(idx)
+    assert (result.values == idx.values).all()
+
+
+def test_tuples_with_name_string():
+    # GH 15110 and GH 14848
+
+    li = [(0, 0, 1), (0, 1, 0), (1, 0, 0)]
+    msg = "Names should be list-like for a MultiIndex"
+    with pytest.raises(ValueError, match=msg):
+        Index(li, name="abc")
+    with pytest.raises(ValueError, match=msg):
+        Index(li, name="a")
+
+
+def test_from_tuples_with_tuple_label():
+    # GH 15457
+    expected = pd.DataFrame(
+        [[2, 1, 2], [4, (1, 2), 3]], columns=["a", "b", "c"]
+    ).set_index(["a", "b"])
+    idx = MultiIndex.from_tuples([(2, 1), (4, (1, 2))], names=("a", "b"))
+    result = pd.DataFrame([2, 3], columns=["c"], index=idx)
+    tm.assert_frame_equal(expected, result)
+
+
+# ----------------------------------------------------------------------------
+# from_product
+# ----------------------------------------------------------------------------
+def test_from_product_empty_zero_levels():
+    # 0 levels
+    msg = "Must pass non-zero number of levels/codes"
+    with pytest.raises(ValueError, match=msg):
+        MultiIndex.from_product([])
+
+
+def test_from_product_empty_one_level():
+    result = MultiIndex.from_product([[]], names=["A"])
+    expected = Index([], name="A")
+    tm.assert_index_equal(result.levels[0], expected)
+    assert result.names == ["A"]
+
+
+@pytest.mark.parametrize(
+    "first, second", [([], []), (["foo", "bar", "baz"], []), ([], ["a", "b", "c"])]
+)
+def test_from_product_empty_two_levels(first, second):
+    names = ["A", "B"]
+    result = MultiIndex.from_product([first, second], names=names)
+    expected = MultiIndex(levels=[first, second], codes=[[], []], names=names)
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("N", list(range(4)))
+def test_from_product_empty_three_levels(N):
+    # GH12258
+    names = ["A", "B", "C"]
+    lvl2 = list(range(N))
+    result = MultiIndex.from_product([[], lvl2, []], names=names)
+    expected = MultiIndex(levels=[[], lvl2, []], codes=[[], [], []], names=names)
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "invalid_input", [1, [1], [1, 2], [[1], 2], "a", ["a"], ["a", "b"], [["a"], "b"]]
+)
+def test_from_product_invalid_input(invalid_input):
+    msg = r"Input must be a list / sequence of iterables|Input must be list-like"
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_product(iterables=invalid_input)
+
+
+def test_from_product_datetimeindex():
+    dt_index = date_range("2000-01-01", periods=2)
+    mi = MultiIndex.from_product([[1, 2], dt_index])
+    etalon = construct_1d_object_array_from_listlike(
+        [
+            (1, Timestamp("2000-01-01")),
+            (1, Timestamp("2000-01-02")),
+            (2, Timestamp("2000-01-01")),
+            (2, Timestamp("2000-01-02")),
+        ]
+    )
+    tm.assert_numpy_array_equal(mi.values, etalon)
+
+
+def test_from_product_rangeindex():
+    # RangeIndex is preserved by factorize, so preserved in levels
+    rng = Index(range(5))
+    other = ["a", "b"]
+    mi = MultiIndex.from_product([rng, other])
+    tm.assert_index_equal(mi._levels[0], rng, exact=True)
+
+
+@pytest.mark.parametrize("ordered", [False, True])
+@pytest.mark.parametrize("f", [lambda x: x, lambda x: Series(x), lambda x: x.values])
+def test_from_product_index_series_categorical(ordered, f):
+    # GH13743
+    first = ["foo", "bar"]
+
+    idx = pd.CategoricalIndex(list("abcaab"), categories=list("bac"), ordered=ordered)
+    expected = pd.CategoricalIndex(
+        list("abcaab") + list("abcaab"), categories=list("bac"), ordered=ordered
+    )
+
+    result = MultiIndex.from_product([first, f(idx)])
+    tm.assert_index_equal(result.get_level_values(1), expected)
+
+
+def test_from_product():
+    first = ["foo", "bar", "buz"]
+    second = ["a", "b", "c"]
+    names = ["first", "second"]
+    result = MultiIndex.from_product([first, second], names=names)
+
+    tuples = [
+        ("foo", "a"),
+        ("foo", "b"),
+        ("foo", "c"),
+        ("bar", "a"),
+        ("bar", "b"),
+        ("bar", "c"),
+        ("buz", "a"),
+        ("buz", "b"),
+        ("buz", "c"),
+    ]
+    expected = MultiIndex.from_tuples(tuples, names=names)
+
+    tm.assert_index_equal(result, expected)
+
+
+def test_from_product_iterator():
+    # GH 18434
+    first = ["foo", "bar", "buz"]
+    second = ["a", "b", "c"]
+    names = ["first", "second"]
+    tuples = [
+        ("foo", "a"),
+        ("foo", "b"),
+        ("foo", "c"),
+        ("bar", "a"),
+        ("bar", "b"),
+        ("bar", "c"),
+        ("buz", "a"),
+        ("buz", "b"),
+        ("buz", "c"),
+    ]
+    expected = MultiIndex.from_tuples(tuples, names=names)
+
+    # iterator as input
+    result = MultiIndex.from_product(iter([first, second]), names=names)
+    tm.assert_index_equal(result, expected)
+
+    # Invalid non-iterable input
+    msg = "Input must be a list / sequence of iterables."
+    with pytest.raises(TypeError, match=msg):
+        MultiIndex.from_product(0)
+
+
+@pytest.mark.parametrize(
+    "a, b, expected_names",
+    [
+        (
+            Series([1, 2, 3], name="foo"),
+            Series(["a", "b"], name="bar"),
+            ["foo", "bar"],
+        ),
+        (Series([1, 2, 3], name="foo"), ["a", "b"], ["foo", None]),
+        ([1, 2, 3], ["a", "b"], None),
+    ],
+)
+def test_from_product_infer_names(a, b, expected_names):
+    # GH27292
+    result = MultiIndex.from_product([a, b])
+    expected = MultiIndex(
+        levels=[[1, 2, 3], ["a", "b"]],
+        codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+        names=expected_names,
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_from_product_respects_none_names():
+    # GH27292
+    a = Series([1, 2, 3], name="foo")
+    b = Series(["a", "b"], name="bar")
+
+    result = MultiIndex.from_product([a, b], names=None)
+    expected = MultiIndex(
+        levels=[[1, 2, 3], ["a", "b"]],
+        codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+        names=None,
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_from_product_readonly():
+    # GH#15286 passing read-only array to from_product
+    a = np.array(range(3))
+    b = ["a", "b"]
+    expected = MultiIndex.from_product([a, b])
+
+    a.setflags(write=False)
+    result = MultiIndex.from_product([a, b])
+    tm.assert_index_equal(result, expected)
+
+
+def test_create_index_existing_name(idx):
+    # GH11193, when an existing index is passed, and a new name is not
+    # specified, the new index should inherit the previous object name
+    index = idx
+    index.names = ["foo", "bar"]
+    result = Index(index)
+    expected = Index(
+        Index(
+            [
+                ("foo", "one"),
+                ("foo", "two"),
+                ("bar", "one"),
+                ("baz", "two"),
+                ("qux", "one"),
+                ("qux", "two"),
+            ],
+            dtype="object",
+        )
+    )
+    tm.assert_index_equal(result, expected)
+
+    result = Index(index, name="A")
+    expected = Index(
+        Index(
+            [
+                ("foo", "one"),
+                ("foo", "two"),
+                ("bar", "one"),
+                ("baz", "two"),
+                ("qux", "one"),
+                ("qux", "two"),
+            ],
+            dtype="object",
+        ),
+        name="A",
+    )
+    tm.assert_index_equal(result, expected)
+
+
+# ----------------------------------------------------------------------------
+# from_frame
+# ----------------------------------------------------------------------------
+def test_from_frame():
+    # GH 22420
+    df = pd.DataFrame(
+        [["a", "a"], ["a", "b"], ["b", "a"], ["b", "b"]], columns=["L1", "L2"]
+    )
+    expected = MultiIndex.from_tuples(
+        [("a", "a"), ("a", "b"), ("b", "a"), ("b", "b")], names=["L1", "L2"]
+    )
+    result = MultiIndex.from_frame(df)
+    tm.assert_index_equal(expected, result)
+
+
+def test_from_frame_missing_values_multiIndex():
+    # GH 39984
+    pa = pytest.importorskip("pyarrow")
+
+    df = pd.DataFrame(
+        {
+            "a": Series([1, 2, None], dtype="Int64"),
+            "b": pd.Float64Dtype().__from_arrow__(pa.array([0.2, np.nan, None])),
+        }
+    )
+    multi_indexed = MultiIndex.from_frame(df)
+    expected = MultiIndex.from_arrays(
+        [
+            Series([1, 2, None]).astype("Int64"),
+            pd.Float64Dtype().__from_arrow__(pa.array([0.2, np.nan, None])),
+        ],
+        names=["a", "b"],
+    )
+    tm.assert_index_equal(multi_indexed, expected)
+
+
+@pytest.mark.parametrize(
+    "non_frame",
+    [
+        Series([1, 2, 3, 4]),
+        [1, 2, 3, 4],
+        [[1, 2], [3, 4], [5, 6]],
+        Index([1, 2, 3, 4]),
+        np.array([[1, 2], [3, 4], [5, 6]]),
+        27,
+    ],
+)
+def test_from_frame_error(non_frame):
+    # GH 22420
+    with pytest.raises(TypeError, match="Input must be a DataFrame"):
+        MultiIndex.from_frame(non_frame)
+
+
+def test_from_frame_dtype_fidelity():
+    # GH 22420
+    df = pd.DataFrame(
+        {
+            "dates": date_range("19910905", periods=6, tz="US/Eastern"),
+            "a": [1, 1, 1, 2, 2, 2],
+            "b": pd.Categorical(["a", "a", "b", "b", "c", "c"], ordered=True),
+            "c": ["x", "x", "y", "z", "x", "y"],
+        }
+    )
+    original_dtypes = df.dtypes.to_dict()
+
+    expected_mi = MultiIndex.from_arrays(
+        [
+            date_range("19910905", periods=6, tz="US/Eastern"),
+            [1, 1, 1, 2, 2, 2],
+            pd.Categorical(["a", "a", "b", "b", "c", "c"], ordered=True),
+            ["x", "x", "y", "z", "x", "y"],
+        ],
+        names=["dates", "a", "b", "c"],
+    )
+    mi = MultiIndex.from_frame(df)
+    mi_dtypes = {name: mi.levels[i].dtype for i, name in enumerate(mi.names)}
+
+    tm.assert_index_equal(expected_mi, mi)
+    assert original_dtypes == mi_dtypes
+
+
+@pytest.mark.parametrize(
+    "names_in,names_out", [(None, [("L1", "x"), ("L2", "y")]), (["x", "y"], ["x", "y"])]
+)
+def test_from_frame_valid_names(names_in, names_out):
+    # GH 22420
+    df = pd.DataFrame(
+        [["a", "a"], ["a", "b"], ["b", "a"], ["b", "b"]],
+        columns=MultiIndex.from_tuples([("L1", "x"), ("L2", "y")]),
+    )
+    mi = MultiIndex.from_frame(df, names=names_in)
+    assert mi.names == names_out
+
+
+@pytest.mark.parametrize(
+    "names,expected_error_msg",
+    [
+        ("bad_input", "Names should be list-like for a MultiIndex"),
+        (["a", "b", "c"], "Length of names must match number of levels in MultiIndex"),
+    ],
+)
+def test_from_frame_invalid_names(names, expected_error_msg):
+    # GH 22420
+    df = pd.DataFrame(
+        [["a", "a"], ["a", "b"], ["b", "a"], ["b", "b"]],
+        columns=MultiIndex.from_tuples([("L1", "x"), ("L2", "y")]),
+    )
+    with pytest.raises(ValueError, match=expected_error_msg):
+        MultiIndex.from_frame(df, names=names)
+
+
+def test_index_equal_empty_iterable():
+    # #16844
+    a = MultiIndex(levels=[[], []], codes=[[], []], names=["a", "b"])
+    b = MultiIndex.from_arrays(arrays=[[], []], names=["a", "b"])
+    tm.assert_index_equal(a, b)
+
+
+def test_raise_invalid_sortorder():
+    # Test that the MultiIndex constructor raise when a incorrect sortorder is given
+    # GH#28518
+
+    levels = [[0, 1], [0, 1, 2]]
+
+    # Correct sortorder
+    MultiIndex(
+        levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]], sortorder=2
+    )
+
+    with pytest.raises(ValueError, match=r".* sortorder 2 with lexsort_depth 1.*"):
+        MultiIndex(
+            levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 2, 1]], sortorder=2
+        )
+
+    with pytest.raises(ValueError, match=r".* sortorder 1 with lexsort_depth 0.*"):
+        MultiIndex(
+            levels=levels, codes=[[0, 0, 1, 0, 1, 1], [0, 1, 0, 2, 2, 1]], sortorder=1
+        )
+
+
+def test_datetimeindex():
+    idx1 = pd.DatetimeIndex(
+        ["2013-04-01 9:00", "2013-04-02 9:00", "2013-04-03 9:00"] * 2, tz="Asia/Tokyo"
+    )
+    idx2 = date_range("2010/01/01", periods=6, freq="ME", tz="US/Eastern")
+    idx = MultiIndex.from_arrays([idx1, idx2])
+
+    expected1 = pd.DatetimeIndex(
+        ["2013-04-01 9:00", "2013-04-02 9:00", "2013-04-03 9:00"], tz="Asia/Tokyo"
+    )
+
+    tm.assert_index_equal(idx.levels[0], expected1)
+    tm.assert_index_equal(idx.levels[1], idx2)
+
+    # from datetime combos
+    # GH 7888
+    date1 = np.datetime64("today")
+    date2 = datetime.today()
+    date3 = Timestamp.today()
+
+    for d1, d2 in itertools.product([date1, date2, date3], [date1, date2, date3]):
+        index = MultiIndex.from_product([[d1], [d2]])
+        assert isinstance(index.levels[0], pd.DatetimeIndex)
+        assert isinstance(index.levels[1], pd.DatetimeIndex)
+
+    # but NOT date objects, matching Index behavior
+    date4 = date.today()
+    index = MultiIndex.from_product([[date4], [date2]])
+    assert not isinstance(index.levels[0], pd.DatetimeIndex)
+    assert isinstance(index.levels[1], pd.DatetimeIndex)
+
+
+def test_constructor_with_tz():
+    index = pd.DatetimeIndex(
+        ["2013/01/01 09:00", "2013/01/02 09:00"], name="dt1", tz="US/Pacific"
+    )
+    columns = pd.DatetimeIndex(
+        ["2014/01/01 09:00", "2014/01/02 09:00"], name="dt2", tz="Asia/Tokyo"
+    )
+
+    result = MultiIndex.from_arrays([index, columns])
+
+    assert result.names == ["dt1", "dt2"]
+    tm.assert_index_equal(result.levels[0], index)
+    tm.assert_index_equal(result.levels[1], columns)
+
+    result = MultiIndex.from_arrays([Series(index), Series(columns)])
+
+    assert result.names == ["dt1", "dt2"]
+    tm.assert_index_equal(result.levels[0], index)
+    tm.assert_index_equal(result.levels[1], columns)
+
+
+def test_multiindex_inference_consistency():
+    # check that inference behavior matches the base class
+
+    v = date.today()
+
+    arr = [v, v]
+
+    idx = Index(arr)
+    assert idx.dtype == object
+
+    mi = MultiIndex.from_arrays([arr])
+    lev = mi.levels[0]
+    assert lev.dtype == object
+
+    mi = MultiIndex.from_product([arr])
+    lev = mi.levels[0]
+    assert lev.dtype == object
+
+    mi = MultiIndex.from_tuples([(x,) for x in arr])
+    lev = mi.levels[0]
+    assert lev.dtype == object
+
+
+def test_dtype_representation(using_infer_string):
+    # GH#46900
+    pmidx = MultiIndex.from_arrays([[1], ["a"]], names=[("a", "b"), ("c", "d")])
+    result = pmidx.dtypes
+    exp = "object" if not using_infer_string else pd.StringDtype(na_value=np.nan)
+    expected = Series(
+        ["int64", exp],
+        index=MultiIndex.from_tuples([("a", "b"), ("c", "d")]),
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_conversion.py b/pandas/pandas/tests/indexes/multi/test_conversion.py
new file mode 100644
index 0000000000000..347d6b206e3b9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_conversion.py
@@ -0,0 +1,208 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gt2
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+def test_to_numpy(idx):
+    result = idx.to_numpy()
+    exp = idx.values
+    tm.assert_numpy_array_equal(result, exp)
+
+
+def test_array_interface(idx):
+    # https://github.com/pandas-dev/pandas/pull/60046
+    result = np.asarray(idx)
+    expected = np.empty((6,), dtype=object)
+    expected[:] = [
+        ("foo", "one"),
+        ("foo", "two"),
+        ("bar", "one"),
+        ("baz", "two"),
+        ("qux", "one"),
+        ("qux", "two"),
+    ]
+    tm.assert_numpy_array_equal(result, expected)
+
+    # it always gives a copy by default, but the values are cached, so results
+    # are still sharing memory
+    result_copy1 = np.asarray(idx)
+    result_copy2 = np.asarray(idx)
+    assert np.may_share_memory(result_copy1, result_copy2)
+
+    # with explicit copy=True, then it is an actual copy
+    result_copy1 = np.array(idx, copy=True)
+    result_copy2 = np.array(idx, copy=True)
+    assert not np.may_share_memory(result_copy1, result_copy2)
+
+    if not np_version_gt2:
+        # copy=False semantics are only supported in NumPy>=2.
+        return
+
+    # for MultiIndex, copy=False is never allowed
+    with pytest.raises(ValueError, match="Unable to avoid copy while creating"):
+        np.array(idx, copy=False)
+
+
+def test_to_frame():
+    tuples = [(1, "one"), (1, "two"), (2, "one"), (2, "two")]
+
+    index = MultiIndex.from_tuples(tuples)
+    result = index.to_frame(index=False)
+    expected = DataFrame(tuples)
+    tm.assert_frame_equal(result, expected)
+
+    result = index.to_frame()
+    expected.index = index
+    tm.assert_frame_equal(result, expected)
+
+    tuples = [(1, "one"), (1, "two"), (2, "one"), (2, "two")]
+    index = MultiIndex.from_tuples(tuples, names=["first", "second"])
+    result = index.to_frame(index=False)
+    expected = DataFrame(tuples)
+    expected.columns = ["first", "second"]
+    tm.assert_frame_equal(result, expected)
+
+    result = index.to_frame()
+    expected.index = index
+    tm.assert_frame_equal(result, expected)
+
+    # See GH-22580
+    index = MultiIndex.from_tuples(tuples)
+    result = index.to_frame(index=False, name=["first", "second"])
+    expected = DataFrame(tuples)
+    expected.columns = ["first", "second"]
+    tm.assert_frame_equal(result, expected)
+
+    result = index.to_frame(name=["first", "second"])
+    expected.index = index
+    expected.columns = ["first", "second"]
+    tm.assert_frame_equal(result, expected)
+
+    msg = "'name' must be a list / sequence of column names."
+    with pytest.raises(TypeError, match=msg):
+        index.to_frame(name="first")
+
+    msg = "'name' should have same length as number of levels on index."
+    with pytest.raises(ValueError, match=msg):
+        index.to_frame(name=["first"])
+
+    # Tests for datetime index
+    index = MultiIndex.from_product([range(5), pd.date_range("20130101", periods=3)])
+    result = index.to_frame(index=False)
+    expected = DataFrame(
+        {
+            0: np.repeat(np.arange(5, dtype="int64"), 3),
+            1: np.tile(pd.date_range("20130101", periods=3), 5),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = index.to_frame()
+    expected.index = index
+    tm.assert_frame_equal(result, expected)
+
+    # See GH-22580
+    result = index.to_frame(index=False, name=["first", "second"])
+    expected = DataFrame(
+        {
+            "first": np.repeat(np.arange(5, dtype="int64"), 3),
+            "second": np.tile(pd.date_range("20130101", periods=3), 5),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = index.to_frame(name=["first", "second"])
+    expected.index = index
+    tm.assert_frame_equal(result, expected)
+
+
+def test_to_frame_dtype_fidelity():
+    # GH 22420
+    mi = MultiIndex.from_arrays(
+        [
+            pd.date_range("19910905", periods=6, tz="US/Eastern"),
+            [1, 1, 1, 2, 2, 2],
+            pd.Categorical(["a", "a", "b", "b", "c", "c"], ordered=True),
+            ["x", "x", "y", "z", "x", "y"],
+        ],
+        names=["dates", "a", "b", "c"],
+    )
+    original_dtypes = {name: mi.levels[i].dtype for i, name in enumerate(mi.names)}
+
+    expected_df = DataFrame(
+        {
+            "dates": pd.date_range("19910905", periods=6, tz="US/Eastern"),
+            "a": [1, 1, 1, 2, 2, 2],
+            "b": pd.Categorical(["a", "a", "b", "b", "c", "c"], ordered=True),
+            "c": ["x", "x", "y", "z", "x", "y"],
+        }
+    )
+    df = mi.to_frame(index=False)
+    df_dtypes = df.dtypes.to_dict()
+
+    tm.assert_frame_equal(df, expected_df)
+    assert original_dtypes == df_dtypes
+
+
+def test_to_frame_resulting_column_order():
+    # GH 22420
+    expected = ["z", 0, "a"]
+    mi = MultiIndex.from_arrays(
+        [["a", "b", "c"], ["x", "y", "z"], ["q", "w", "e"]], names=expected
+    )
+    result = mi.to_frame().columns.tolist()
+    assert result == expected
+
+
+def test_to_frame_duplicate_labels():
+    # GH 45245
+    data = [(1, 2), (3, 4)]
+    names = ["a", "a"]
+    index = MultiIndex.from_tuples(data, names=names)
+    with pytest.raises(ValueError, match="Cannot create duplicate column labels"):
+        index.to_frame()
+
+    result = index.to_frame(allow_duplicates=True)
+    expected = DataFrame(data, index=index, columns=names)
+    tm.assert_frame_equal(result, expected)
+
+    names = [None, 0]
+    index = MultiIndex.from_tuples(data, names=names)
+    with pytest.raises(ValueError, match="Cannot create duplicate column labels"):
+        index.to_frame()
+
+    result = index.to_frame(allow_duplicates=True)
+    expected = DataFrame(data, index=index, columns=[0, 0])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_to_frame_column_rangeindex():
+    mi = MultiIndex.from_arrays([[1, 2], ["a", "b"]])
+    result = mi.to_frame().columns
+    expected = RangeIndex(2)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_to_flat_index(idx):
+    expected = pd.Index(
+        (
+            ("foo", "one"),
+            ("foo", "two"),
+            ("bar", "one"),
+            ("baz", "two"),
+            ("qux", "one"),
+            ("qux", "two"),
+        ),
+        tupleize_cols=False,
+    )
+    result = idx.to_flat_index()
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_copy.py b/pandas/pandas/tests/indexes/multi/test_copy.py
new file mode 100644
index 0000000000000..2e09a580f9528
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_copy.py
@@ -0,0 +1,96 @@
+from copy import (
+    copy,
+    deepcopy,
+)
+
+import pytest
+
+from pandas import MultiIndex
+import pandas._testing as tm
+
+
+def assert_multiindex_copied(copy, original):
+    # Levels should be (at least, shallow copied)
+    tm.assert_copy(copy.levels, original.levels)
+    tm.assert_almost_equal(copy.codes, original.codes)
+
+    # Labels doesn't matter which way copied
+    tm.assert_almost_equal(copy.codes, original.codes)
+    assert copy.codes is not original.codes
+
+    # Names doesn't matter which way copied
+    assert copy.names == original.names
+    assert copy.names is not original.names
+
+    # Sort order should be copied
+    assert copy.sortorder == original.sortorder
+
+
+def test_copy(idx):
+    i_copy = idx.copy()
+
+    assert_multiindex_copied(i_copy, idx)
+
+
+def test_shallow_copy(idx):
+    i_copy = idx._view()
+
+    assert_multiindex_copied(i_copy, idx)
+
+
+def test_view(idx):
+    i_view = idx.view()
+    assert_multiindex_copied(i_view, idx)
+
+
+@pytest.mark.parametrize("func", [copy, deepcopy])
+def test_copy_and_deepcopy(func):
+    idx = MultiIndex(
+        levels=[["foo", "bar"], ["fizz", "buzz"]],
+        codes=[[0, 0, 0, 1], [0, 0, 1, 1]],
+        names=["first", "second"],
+    )
+    idx_copy = func(idx)
+    assert idx_copy is not idx
+    assert idx_copy.equals(idx)
+
+
+@pytest.mark.parametrize("deep", [True, False])
+def test_copy_method(deep):
+    idx = MultiIndex(
+        levels=[["foo", "bar"], ["fizz", "buzz"]],
+        codes=[[0, 0, 0, 1], [0, 0, 1, 1]],
+        names=["first", "second"],
+    )
+    idx_copy = idx.copy(deep=deep)
+    assert idx_copy.equals(idx)
+
+
+@pytest.mark.parametrize("deep", [True, False])
+@pytest.mark.parametrize(
+    "kwarg, value",
+    [
+        ("names", ["third", "fourth"]),
+    ],
+)
+def test_copy_method_kwargs(deep, kwarg, value):
+    # gh-12309: Check that the "name" argument as well other kwargs are honored
+    idx = MultiIndex(
+        levels=[["foo", "bar"], ["fizz", "buzz"]],
+        codes=[[0, 0, 0, 1], [0, 0, 1, 1]],
+        names=["first", "second"],
+    )
+    idx_copy = idx.copy(**{kwarg: value, "deep": deep})
+    assert getattr(idx_copy, kwarg) == value
+
+
+def test_copy_deep_false_retains_id():
+    # GH#47878
+    idx = MultiIndex(
+        levels=[["foo", "bar"], ["fizz", "buzz"]],
+        codes=[[0, 0, 0, 1], [0, 0, 1, 1]],
+        names=["first", "second"],
+    )
+
+    res = idx.copy(deep=False)
+    assert res._id is idx._id
diff --git a/pandas/pandas/tests/indexes/multi/test_drop.py b/pandas/pandas/tests/indexes/multi/test_drop.py
new file mode 100644
index 0000000000000..b83680be6a5ce
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_drop.py
@@ -0,0 +1,188 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+def test_drop(idx):
+    dropped = idx.drop([("foo", "two"), ("qux", "one")])
+
+    index = MultiIndex.from_tuples([("foo", "two"), ("qux", "one")])
+    dropped2 = idx.drop(index)
+
+    expected = idx[[0, 2, 3, 5]]
+    tm.assert_index_equal(dropped, expected)
+    tm.assert_index_equal(dropped2, expected)
+
+    dropped = idx.drop(["bar"])
+    expected = idx[[0, 1, 3, 4, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    dropped = idx.drop("foo")
+    expected = idx[[2, 3, 4, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    index = MultiIndex.from_tuples([("bar", "two")])
+    with pytest.raises(KeyError, match=r"^\('bar', 'two'\)$"):
+        idx.drop([("bar", "two")])
+    with pytest.raises(KeyError, match=r"^\('bar', 'two'\)$"):
+        idx.drop(index)
+    with pytest.raises(KeyError, match=r"^'two'$"):
+        idx.drop(["foo", "two"])
+
+    # partially correct argument
+    mixed_index = MultiIndex.from_tuples([("qux", "one"), ("bar", "two")])
+    with pytest.raises(KeyError, match=r"^\('bar', 'two'\)$"):
+        idx.drop(mixed_index)
+
+    # error='ignore'
+    dropped = idx.drop(index, errors="ignore")
+    expected = idx[[0, 1, 2, 3, 4, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    dropped = idx.drop(mixed_index, errors="ignore")
+    expected = idx[[0, 1, 2, 3, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    dropped = idx.drop(["foo", "two"], errors="ignore")
+    expected = idx[[2, 3, 4, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    # mixed partial / full drop
+    dropped = idx.drop(["foo", ("qux", "one")])
+    expected = idx[[2, 3, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+    # mixed partial / full drop / error='ignore'
+    mixed_index = ["foo", ("qux", "one"), "two"]
+    with pytest.raises(KeyError, match=r"^'two'$"):
+        idx.drop(mixed_index)
+    dropped = idx.drop(mixed_index, errors="ignore")
+    expected = idx[[2, 3, 5]]
+    tm.assert_index_equal(dropped, expected)
+
+
+def test_droplevel_with_names(idx):
+    index = idx[idx.get_loc("foo")]
+    dropped = index.droplevel(0)
+    assert dropped.name == "second"
+
+    index = MultiIndex(
+        levels=[Index(range(4)), Index(range(4)), Index(range(4))],
+        codes=[
+            np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+            np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+            np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+        ],
+        names=["one", "two", "three"],
+    )
+    dropped = index.droplevel(0)
+    assert dropped.names == ("two", "three")
+
+    dropped = index.droplevel("two")
+    expected = index.droplevel(1)
+    assert dropped.equals(expected)
+
+
+def test_droplevel_list():
+    index = MultiIndex(
+        levels=[Index(range(4)), Index(range(4)), Index(range(4))],
+        codes=[
+            np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+            np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+            np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+        ],
+        names=["one", "two", "three"],
+    )
+
+    dropped = index[:2].droplevel(["three", "one"])
+    expected = index[:2].droplevel(2).droplevel(0)
+    assert dropped.equals(expected)
+
+    dropped = index[:2].droplevel([])
+    expected = index[:2]
+    assert dropped.equals(expected)
+
+    msg = (
+        "Cannot remove 3 levels from an index with 3 levels: "
+        "at least one level must be left"
+    )
+    with pytest.raises(ValueError, match=msg):
+        index[:2].droplevel(["one", "two", "three"])
+
+    with pytest.raises(KeyError, match="'Level four not found'"):
+        index[:2].droplevel(["one", "four"])
+
+
+def test_drop_not_lexsorted(performance_warning):
+    # GH 12078
+
+    # define the lexsorted version of the multi-index
+    tuples = [("a", ""), ("b1", "c1"), ("b2", "c2")]
+    lexsorted_mi = MultiIndex.from_tuples(tuples, names=["b", "c"])
+    assert lexsorted_mi._is_lexsorted()
+
+    # and the not-lexsorted version
+    df = pd.DataFrame(
+        columns=["a", "b", "c", "d"], data=[[1, "b1", "c1", 3], [1, "b2", "c2", 4]]
+    )
+    df = df.pivot_table(index="a", columns=["b", "c"], values="d")
+    df = df.reset_index()
+    not_lexsorted_mi = df.columns
+    assert not not_lexsorted_mi._is_lexsorted()
+
+    # compare the results
+    tm.assert_index_equal(lexsorted_mi, not_lexsorted_mi)
+    with tm.assert_produces_warning(performance_warning):
+        tm.assert_index_equal(lexsorted_mi.drop("a"), not_lexsorted_mi.drop("a"))
+
+
+def test_drop_with_nan_in_index(nulls_fixture):
+    # GH#18853
+    mi = MultiIndex.from_tuples([("blah", nulls_fixture)], names=["name", "date"])
+    msg = r"labels \[Timestamp\('2001-01-01 00:00:00'\)\] not found in level"
+    with pytest.raises(KeyError, match=msg):
+        mi.drop(pd.Timestamp("2001"), level="date")
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+def test_drop_with_non_monotonic_duplicates():
+    # GH#33494
+    mi = MultiIndex.from_tuples([(1, 2), (2, 3), (1, 2)])
+    result = mi.drop((1, 2))
+    expected = MultiIndex.from_tuples([(2, 3)])
+    tm.assert_index_equal(result, expected)
+
+
+def test_single_level_drop_partially_missing_elements():
+    # GH 37820
+
+    mi = MultiIndex.from_tuples([(1, 2), (2, 2), (3, 2)])
+    msg = r"labels \[4\] not found in level"
+    with pytest.raises(KeyError, match=msg):
+        mi.drop(4, level=0)
+    with pytest.raises(KeyError, match=msg):
+        mi.drop([1, 4], level=0)
+    msg = r"labels \[nan\] not found in level"
+    with pytest.raises(KeyError, match=msg):
+        mi.drop([np.nan], level=0)
+    with pytest.raises(KeyError, match=msg):
+        mi.drop([np.nan, 1, 2, 3], level=0)
+
+    mi = MultiIndex.from_tuples([(np.nan, 1), (1, 2)])
+    msg = r"labels \['a'\] not found in level"
+    with pytest.raises(KeyError, match=msg):
+        mi.drop([np.nan, 1, "a"], level=0)
+
+
+def test_droplevel_multiindex_one_level():
+    # GH#37208
+    index = MultiIndex.from_tuples([(2,)], names=("b",))
+    result = index.droplevel([])
+    expected = Index([2], name="b")
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_duplicates.py b/pandas/pandas/tests/indexes/multi/test_duplicates.py
new file mode 100644
index 0000000000000..1bbeedac3fb10
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_duplicates.py
@@ -0,0 +1,357 @@
+from itertools import product
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    hashtable,
+    index as libindex,
+)
+
+from pandas import (
+    NA,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def idx_dup():
+    # compare tests/indexes/multi/conftest.py
+    major_axis = Index(["foo", "bar", "baz", "qux"])
+    minor_axis = Index(["one", "two"])
+
+    major_codes = np.array([0, 0, 1, 0, 1, 1])
+    minor_codes = np.array([0, 1, 0, 1, 0, 1])
+    index_names = ["first", "second"]
+    mi = MultiIndex(
+        levels=[major_axis, minor_axis],
+        codes=[major_codes, minor_codes],
+        names=index_names,
+        verify_integrity=False,
+    )
+    return mi
+
+
+@pytest.mark.parametrize("names", [None, ["first", "second"]])
+def test_unique(names):
+    mi = MultiIndex.from_arrays([[1, 2, 1, 2], [1, 1, 1, 2]], names=names)
+
+    res = mi.unique()
+    exp = MultiIndex.from_arrays([[1, 2, 2], [1, 1, 2]], names=mi.names)
+    tm.assert_index_equal(res, exp)
+
+    mi = MultiIndex.from_arrays([list("aaaa"), list("abab")], names=names)
+    res = mi.unique()
+    exp = MultiIndex.from_arrays([list("aa"), list("ab")], names=mi.names)
+    tm.assert_index_equal(res, exp)
+
+    mi = MultiIndex.from_arrays([list("aaaa"), list("aaaa")], names=names)
+    res = mi.unique()
+    exp = MultiIndex.from_arrays([["a"], ["a"]], names=mi.names)
+    tm.assert_index_equal(res, exp)
+
+    # GH #20568 - empty MI
+    mi = MultiIndex.from_arrays([[], []], names=names)
+    res = mi.unique()
+    tm.assert_index_equal(mi, res)
+
+
+def test_unique_datetimelike():
+    idx1 = DatetimeIndex(
+        ["2015-01-01", "2015-01-01", "2015-01-01", "2015-01-01", "NaT", "NaT"]
+    )
+    idx2 = DatetimeIndex(
+        ["2015-01-01", "2015-01-01", "2015-01-02", "2015-01-02", "NaT", "2015-01-01"],
+        tz="Asia/Tokyo",
+    )
+    result = MultiIndex.from_arrays([idx1, idx2]).unique()
+
+    eidx1 = DatetimeIndex(["2015-01-01", "2015-01-01", "NaT", "NaT"])
+    eidx2 = DatetimeIndex(
+        ["2015-01-01", "2015-01-02", "NaT", "2015-01-01"], tz="Asia/Tokyo"
+    )
+    exp = MultiIndex.from_arrays([eidx1, eidx2])
+    tm.assert_index_equal(result, exp)
+
+
+@pytest.mark.parametrize("level", [0, "first", 1, "second"])
+def test_unique_level(idx, level):
+    # GH #17896 - with level= argument
+    result = idx.unique(level=level)
+    expected = idx.get_level_values(level).unique()
+    tm.assert_index_equal(result, expected)
+
+    # With already unique level
+    mi = MultiIndex.from_arrays([[1, 3, 2, 4], [1, 3, 2, 5]], names=["first", "second"])
+    result = mi.unique(level=level)
+    expected = mi.get_level_values(level)
+    tm.assert_index_equal(result, expected)
+
+    # With empty MI
+    mi = MultiIndex.from_arrays([[], []], names=["first", "second"])
+    result = mi.unique(level=level)
+    expected = mi.get_level_values(level)
+    tm.assert_index_equal(result, expected)
+
+
+def test_duplicate_multiindex_codes():
+    # GH 17464
+    # Make sure that a MultiIndex with duplicate levels throws a ValueError
+    msg = r"Level values must be unique: \[[A', ]+\] on level 0"
+    with pytest.raises(ValueError, match=msg):
+        mi = MultiIndex([["A"] * 10, range(10)], [[0] * 10, range(10)])
+
+    # And that using set_levels with duplicate levels fails
+    mi = MultiIndex.from_arrays([["A", "A", "B", "B", "B"], [1, 2, 1, 2, 3]])
+    msg = r"Level values must be unique: \[[AB', ]+\] on level 0"
+    with pytest.raises(ValueError, match=msg):
+        mi.set_levels([["A", "B", "A", "A", "B"], [2, 1, 3, -2, 5]])
+
+
+@pytest.mark.parametrize("names", [["a", "b", "a"], [1, 1, 2], [1, "a", 1]])
+def test_duplicate_level_names(names):
+    # GH18872, GH19029
+    mi = MultiIndex.from_product([[0, 1]] * 3, names=names)
+    assert mi.names == names
+
+    # With .rename()
+    mi = MultiIndex.from_product([[0, 1]] * 3)
+    mi = mi.rename(names)
+    assert mi.names == names
+
+    # With .rename(., level=)
+    mi.rename(names[1], level=1, inplace=True)
+    mi = mi.rename([names[0], names[2]], level=[0, 2])
+    assert mi.names == names
+
+
+def test_duplicate_meta_data():
+    # GH 10115
+    mi = MultiIndex(
+        levels=[[0, 1], [0, 1, 2]], codes=[[0, 0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 0, 1, 2]]
+    )
+
+    for idx in [
+        mi,
+        mi.set_names([None, None]),
+        mi.set_names([None, "Num"]),
+        mi.set_names(["Upper", "Num"]),
+    ]:
+        assert idx.has_duplicates
+        assert idx.drop_duplicates().names == idx.names
+
+
+def test_has_duplicates(idx, idx_dup):
+    # see fixtures
+    assert idx.is_unique is True
+    assert idx.has_duplicates is False
+    assert idx_dup.is_unique is False
+    assert idx_dup.has_duplicates is True
+
+    mi = MultiIndex(
+        levels=[[0, 1], [0, 1, 2]], codes=[[0, 0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 0, 1, 2]]
+    )
+    assert mi.is_unique is False
+    assert mi.has_duplicates is True
+
+    # single instance of NaN
+    mi_nan = MultiIndex(
+        levels=[["a", "b"], [0, 1]], codes=[[-1, 0, 0, 1, 1], [-1, 0, 1, 0, 1]]
+    )
+    assert mi_nan.is_unique is True
+    assert mi_nan.has_duplicates is False
+
+    # multiple instances of NaN
+    mi_nan_dup = MultiIndex(
+        levels=[["a", "b"], [0, 1]], codes=[[-1, -1, 0, 0, 1, 1], [-1, -1, 0, 1, 0, 1]]
+    )
+    assert mi_nan_dup.is_unique is False
+    assert mi_nan_dup.has_duplicates is True
+
+
+def test_has_duplicates_from_tuples():
+    # GH 9075
+    t = [
+        ("x", "out", "z", 5, "y", "in", "z", 169),
+        ("x", "out", "z", 7, "y", "in", "z", 119),
+        ("x", "out", "z", 9, "y", "in", "z", 135),
+        ("x", "out", "z", 13, "y", "in", "z", 145),
+        ("x", "out", "z", 14, "y", "in", "z", 158),
+        ("x", "out", "z", 16, "y", "in", "z", 122),
+        ("x", "out", "z", 17, "y", "in", "z", 160),
+        ("x", "out", "z", 18, "y", "in", "z", 180),
+        ("x", "out", "z", 20, "y", "in", "z", 143),
+        ("x", "out", "z", 21, "y", "in", "z", 128),
+        ("x", "out", "z", 22, "y", "in", "z", 129),
+        ("x", "out", "z", 25, "y", "in", "z", 111),
+        ("x", "out", "z", 28, "y", "in", "z", 114),
+        ("x", "out", "z", 29, "y", "in", "z", 121),
+        ("x", "out", "z", 31, "y", "in", "z", 126),
+        ("x", "out", "z", 32, "y", "in", "z", 155),
+        ("x", "out", "z", 33, "y", "in", "z", 123),
+        ("x", "out", "z", 12, "y", "in", "z", 144),
+    ]
+
+    mi = MultiIndex.from_tuples(t)
+    assert not mi.has_duplicates
+
+
+@pytest.mark.parametrize("nlevels", [4, 8])
+@pytest.mark.parametrize("with_nulls", [True, False])
+def test_has_duplicates_overflow(nlevels, with_nulls):
+    # handle int64 overflow if possible
+    # no overflow with 4
+    # overflow possible with 8
+    codes = np.tile(np.arange(500), 2)
+    level = np.arange(500)
+
+    if with_nulls:  # inject some null values
+        codes[500] = -1  # common nan value
+        codes = [codes.copy() for i in range(nlevels)]
+        for i in range(nlevels):
+            codes[i][500 + i - nlevels // 2] = -1
+
+        codes += [np.array([-1, 1]).repeat(500)]
+    else:
+        codes = [codes] * nlevels + [np.arange(2).repeat(500)]
+
+    levels = [level] * nlevels + [[0, 1]]
+
+    # no dups
+    mi = MultiIndex(levels=levels, codes=codes)
+    assert not mi.has_duplicates
+
+    # with a dup
+    if with_nulls:
+
+        def f(a):
+            return np.insert(a, 1000, a[0])
+
+        codes = list(map(f, codes))
+        mi = MultiIndex(levels=levels, codes=codes)
+    else:
+        values = mi.values.tolist()
+        mi = MultiIndex.from_tuples(values + [values[0]])
+
+    assert mi.has_duplicates
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", [False, False, False, True, True, False]),
+        ("last", [False, True, True, False, False, False]),
+        (False, [False, True, True, True, True, False]),
+    ],
+)
+def test_duplicated(idx_dup, keep, expected):
+    result = idx_dup.duplicated(keep=keep)
+    expected = np.array(expected)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.arm_slow
+def test_duplicated_hashtable_impl(keep, monkeypatch):
+    # GH 9125
+    n, k = 6, 10
+    levels = [np.arange(n), [str(i) for i in range(n)], 1000 + np.arange(n)]
+    codes = [np.random.default_rng(2).choice(n, k * n) for _ in levels]
+    with monkeypatch.context() as m:
+        m.setattr(libindex, "_SIZE_CUTOFF", 50)
+        mi = MultiIndex(levels=levels, codes=codes)
+
+        result = mi.duplicated(keep=keep)
+        expected = hashtable.duplicated(mi.values, keep=keep)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [101, 102])
+def test_duplicated_with_nan(val):
+    # GH5873
+    mi = MultiIndex.from_arrays([[101, val], [3.5, np.nan]])
+    assert not mi.has_duplicates
+
+    tm.assert_numpy_array_equal(mi.duplicated(), np.zeros(2, dtype="bool"))
+
+
+@pytest.mark.parametrize("n", range(1, 6))
+@pytest.mark.parametrize("m", range(1, 5))
+def test_duplicated_with_nan_multi_shape(n, m):
+    # GH5873
+    # all possible unique combinations, including nan
+    codes = product(range(-1, n), range(-1, m))
+    mi = MultiIndex(
+        levels=[list("abcde")[:n], list("WXYZ")[:m]],
+        codes=np.random.default_rng(2).permutation(list(codes)).T,
+    )
+    assert len(mi) == (n + 1) * (m + 1)
+    assert not mi.has_duplicates
+
+    tm.assert_numpy_array_equal(mi.duplicated(), np.zeros(len(mi), dtype="bool"))
+
+
+def test_duplicated_drop_duplicates():
+    # GH#4060
+    idx = MultiIndex.from_arrays(([1, 2, 3, 1, 2, 3], [1, 1, 1, 1, 2, 2]))
+
+    expected = np.array([False, False, False, True, False, False], dtype=bool)
+    duplicated = idx.duplicated()
+    tm.assert_numpy_array_equal(duplicated, expected)
+    assert duplicated.dtype == bool
+    expected = MultiIndex.from_arrays(([1, 2, 3, 2, 3], [1, 1, 1, 2, 2]))
+    tm.assert_index_equal(idx.drop_duplicates(), expected)
+
+    expected = np.array([True, False, False, False, False, False])
+    duplicated = idx.duplicated(keep="last")
+    tm.assert_numpy_array_equal(duplicated, expected)
+    assert duplicated.dtype == bool
+    expected = MultiIndex.from_arrays(([2, 3, 1, 2, 3], [1, 1, 1, 2, 2]))
+    tm.assert_index_equal(idx.drop_duplicates(keep="last"), expected)
+
+    expected = np.array([True, False, False, True, False, False])
+    duplicated = idx.duplicated(keep=False)
+    tm.assert_numpy_array_equal(duplicated, expected)
+    assert duplicated.dtype == bool
+    expected = MultiIndex.from_arrays(([2, 3, 2, 3], [1, 1, 2, 2]))
+    tm.assert_index_equal(idx.drop_duplicates(keep=False), expected)
+
+
+def test_duplicated_series_complex_numbers(complex_dtype):
+    # GH 17927
+    expected = Series(
+        [False, False, False, True, False, False, False, True, False, True],
+        dtype=bool,
+    )
+    result = Series(
+        [
+            np.nan + np.nan * 1j,
+            0,
+            1j,
+            1j,
+            1,
+            1 + 1j,
+            1 + 2j,
+            1 + 1j,
+            np.nan,
+            np.nan + np.nan * 1j,
+        ],
+        dtype=complex_dtype,
+    ).duplicated()
+    tm.assert_series_equal(result, expected)
+
+
+def test_midx_unique_ea_dtype():
+    # GH#48335
+    vals_a = Series([1, 2, NA, NA], dtype="Int64")
+    vals_b = np.array([1, 2, 3, 3])
+    midx = MultiIndex.from_arrays([vals_a, vals_b], names=["a", "b"])
+    result = midx.unique()
+
+    exp_vals_a = Series([1, 2, NA], dtype="Int64")
+    exp_vals_b = np.array([1, 2, 3])
+    expected = MultiIndex.from_arrays([exp_vals_a, exp_vals_b], names=["a", "b"])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_equivalence.py b/pandas/pandas/tests/indexes/multi/test_equivalence.py
new file mode 100644
index 0000000000000..9babbd5b8d56d
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_equivalence.py
@@ -0,0 +1,284 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_any_real_numeric_dtype
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_equals(idx):
+    assert idx.equals(idx)
+    assert idx.equals(idx.copy())
+    assert idx.equals(idx.astype(object))
+    assert idx.equals(idx.to_flat_index())
+    assert idx.equals(idx.to_flat_index().astype("category"))
+
+    assert not idx.equals(list(idx))
+    assert not idx.equals(np.array(idx))
+
+    same_values = Index(idx, dtype=object)
+    assert idx.equals(same_values)
+    assert same_values.equals(idx)
+
+    if idx.nlevels == 1:
+        # do not test MultiIndex
+        assert not idx.equals(Series(idx))
+
+
+def test_equals_op(idx):
+    # GH9947, GH10637
+    index_a = idx
+
+    n = len(index_a)
+    index_b = index_a[0:-1]
+    index_c = index_a[0:-1].append(index_a[-2:-1])
+    index_d = index_a[0:1]
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == index_b
+    expected1 = np.array([True] * n)
+    expected2 = np.array([True] * (n - 1) + [False])
+    tm.assert_numpy_array_equal(index_a == index_a, expected1)
+    tm.assert_numpy_array_equal(index_a == index_c, expected2)
+
+    # test comparisons with numpy arrays
+    array_a = np.array(index_a)
+    array_b = np.array(index_a[0:-1])
+    array_c = np.array(index_a[0:-1].append(index_a[-2:-1]))
+    array_d = np.array(index_a[0:1])
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == array_b
+    tm.assert_numpy_array_equal(index_a == array_a, expected1)
+    tm.assert_numpy_array_equal(index_a == array_c, expected2)
+
+    # test comparisons with Series
+    series_a = Series(array_a)
+    series_b = Series(array_b)
+    series_c = Series(array_c)
+    series_d = Series(array_d)
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == series_b
+
+    tm.assert_numpy_array_equal(index_a == series_a, expected1)
+    tm.assert_numpy_array_equal(index_a == series_c, expected2)
+
+    # cases where length is 1 for one of them
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == index_d
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == series_d
+    with pytest.raises(ValueError, match="Lengths must match"):
+        index_a == array_d
+    msg = "Can only compare identically-labeled Series objects"
+    with pytest.raises(ValueError, match=msg):
+        series_a == series_d
+    with pytest.raises(ValueError, match="Lengths must match"):
+        series_a == array_d
+
+    # comparing with a scalar should broadcast; note that we are excluding
+    # MultiIndex because in this case each item in the index is a tuple of
+    # length 2, and therefore is considered an array of length 2 in the
+    # comparison instead of a scalar
+    if not isinstance(index_a, MultiIndex):
+        expected3 = np.array([False] * (len(index_a) - 2) + [True, False])
+        # assuming the 2nd to last item is unique in the data
+        item = index_a[-2]
+        tm.assert_numpy_array_equal(index_a == item, expected3)
+        tm.assert_series_equal(series_a == item, Series(expected3))
+
+
+def test_compare_tuple():
+    # GH#21517
+    mi = MultiIndex.from_product([[1, 2]] * 2)
+
+    all_false = np.array([False, False, False, False])
+
+    result = mi == mi[0]
+    expected = np.array([True, False, False, False])
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = mi != mi[0]
+    tm.assert_numpy_array_equal(result, ~expected)
+
+    result = mi < mi[0]
+    tm.assert_numpy_array_equal(result, all_false)
+
+    result = mi <= mi[0]
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = mi > mi[0]
+    tm.assert_numpy_array_equal(result, ~expected)
+
+    result = mi >= mi[0]
+    tm.assert_numpy_array_equal(result, ~all_false)
+
+
+def test_compare_tuple_strs():
+    # GH#34180
+
+    mi = MultiIndex.from_tuples([("a", "b"), ("b", "c"), ("c", "a")])
+
+    result = mi == ("c", "a")
+    expected = np.array([False, False, True])
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = mi == ("c",)
+    expected = np.array([False, False, False])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_equals_multi(idx):
+    assert idx.equals(idx)
+    assert not idx.equals(idx.values)
+    assert idx.equals(Index(idx.values))
+
+    assert idx.equal_levels(idx)
+    assert not idx.equals(idx[:-1])
+    assert not idx.equals(idx[-1])
+
+    # different number of levels
+    index = MultiIndex(
+        levels=[Index(list(range(4))), Index(list(range(4))), Index(list(range(4)))],
+        codes=[
+            np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+            np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+            np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+        ],
+    )
+
+    index2 = MultiIndex(levels=index.levels[:-1], codes=index.codes[:-1])
+    assert not index.equals(index2)
+    assert not index.equal_levels(index2)
+
+    # levels are different
+    major_axis = Index(list(range(4)))
+    minor_axis = Index(list(range(2)))
+
+    major_codes = np.array([0, 0, 1, 2, 2, 3])
+    minor_codes = np.array([0, 1, 0, 0, 1, 0])
+
+    index = MultiIndex(
+        levels=[major_axis, minor_axis], codes=[major_codes, minor_codes]
+    )
+    assert not idx.equals(index)
+    assert not idx.equal_levels(index)
+
+    # some of the labels are different
+    major_axis = Index(["foo", "bar", "baz", "qux"])
+    minor_axis = Index(["one", "two"])
+
+    major_codes = np.array([0, 0, 2, 2, 3, 3])
+    minor_codes = np.array([0, 1, 0, 1, 0, 1])
+
+    index = MultiIndex(
+        levels=[major_axis, minor_axis], codes=[major_codes, minor_codes]
+    )
+    assert not idx.equals(index)
+
+
+def test_identical(idx):
+    mi = idx.copy()
+    mi2 = idx.copy()
+    assert mi.identical(mi2)
+
+    mi = mi.set_names(["new1", "new2"])
+    assert mi.equals(mi2)
+    assert not mi.identical(mi2)
+
+    mi2 = mi2.set_names(["new1", "new2"])
+    assert mi.identical(mi2)
+
+    mi4 = Index(mi.tolist(), tupleize_cols=False)
+    assert not mi.identical(mi4)
+    assert mi.equals(mi4)
+
+
+def test_equals_operator(idx):
+    # GH9785
+    assert (idx == idx).all()
+
+
+def test_equals_missing_values():
+    # make sure take is not using -1
+    i = MultiIndex.from_tuples([(0, pd.NaT), (0, pd.Timestamp("20130101"))])
+    result = i[0:1].equals(i[0])
+    assert not result
+    result = i[1:2].equals(i[1])
+    assert not result
+
+
+def test_equals_missing_values_differently_sorted():
+    # GH#38439
+    mi1 = MultiIndex.from_tuples([(81.0, np.nan), (np.nan, np.nan)])
+    mi2 = MultiIndex.from_tuples([(np.nan, np.nan), (81.0, np.nan)])
+    assert not mi1.equals(mi2)
+
+    mi2 = MultiIndex.from_tuples([(81.0, np.nan), (np.nan, np.nan)])
+    assert mi1.equals(mi2)
+
+
+def test_is_():
+    mi = MultiIndex.from_tuples(zip(range(10), range(10)))
+    assert mi.is_(mi)
+    assert mi.is_(mi.view())
+    assert mi.is_(mi.view().view().view().view())
+    mi2 = mi.view()
+    # names are metadata, they don't change id
+    mi2.names = ["A", "B"]
+    assert mi2.is_(mi)
+    assert mi.is_(mi2)
+
+    assert not mi.is_(mi.set_names(["C", "D"]))
+    # levels are inherent properties, they change identity
+    mi3 = mi2.set_levels([list(range(10)), list(range(10))])
+    assert not mi3.is_(mi2)
+    # shouldn't change
+    assert mi2.is_(mi)
+    mi4 = mi3.view()
+
+    # GH 17464 - Remove duplicate MultiIndex levels
+    mi4 = mi4.set_levels([list(range(10)), list(range(10))])
+    assert not mi4.is_(mi3)
+    mi5 = mi.view()
+    mi5 = mi5.set_levels(mi5.levels)
+    assert not mi5.is_(mi)
+
+
+def test_is_all_dates(idx):
+    assert not idx._is_all_dates
+
+
+def test_is_numeric(idx):
+    # MultiIndex is never numeric
+    assert not is_any_real_numeric_dtype(idx)
+
+
+def test_multiindex_compare():
+    # GH 21149
+    # Ensure comparison operations for MultiIndex with nlevels == 1
+    # behave consistently with those for MultiIndex with nlevels > 1
+
+    midx = MultiIndex.from_product([[0, 1]])
+
+    # Equality self-test: MultiIndex object vs self
+    expected = Series([True, True])
+    result = Series(midx == midx)
+    tm.assert_series_equal(result, expected)
+
+    # Greater than comparison: MultiIndex object vs self
+    expected = Series([False, False])
+    result = Series(midx > midx)
+    tm.assert_series_equal(result, expected)
+
+
+def test_equals_ea_int_regular_int():
+    # GH#46026
+    mi1 = MultiIndex.from_arrays([Index([1, 2], dtype="Int64"), [3, 4]])
+    mi2 = MultiIndex.from_arrays([[1, 2], [3, 4]])
+    assert not mi1.equals(mi2)
+    assert not mi2.equals(mi1)
diff --git a/pandas/pandas/tests/indexes/multi/test_formats.py b/pandas/pandas/tests/indexes/multi/test_formats.py
new file mode 100644
index 0000000000000..6ea42349bd04a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_formats.py
@@ -0,0 +1,207 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+)
+
+
+def test_repr_with_unicode_data():
+    with pd.option_context("display.encoding", "UTF-8"):
+        d = {"a": ["\u05d0", 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]}
+        index = pd.DataFrame(d).set_index(["a", "b"]).index
+        assert "\\" not in repr(index)  # we don't want unicode-escaped
+
+
+def test_repr_roundtrip_raises():
+    mi = MultiIndex.from_product([list("ab"), range(3)], names=["first", "second"])
+    msg = "Must pass both levels and codes"
+    with pytest.raises(TypeError, match=msg):
+        eval(repr(mi))
+
+
+def test_unicode_string_with_unicode():
+    d = {"a": ["\u05d0", 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]}
+    idx = pd.DataFrame(d).set_index(["a", "b"]).index
+    str(idx)
+
+
+def test_repr_max_seq_item_setting(idx):
+    # GH10182
+    idx = idx.repeat(50)
+    with pd.option_context("display.max_seq_items", None):
+        repr(idx)
+        assert "..." not in str(idx)
+
+
+class TestRepr:
+    def test_unicode_repr_issues(self):
+        levels = [Index(["a/\u03c3", "b/\u03c3", "c/\u03c3"]), Index([0, 1])]
+        codes = [np.arange(3).repeat(2), np.tile(np.arange(2), 3)]
+        index = MultiIndex(levels=levels, codes=codes)
+
+        repr(index.levels)
+        repr(index.get_level_values(1))
+
+    def test_repr_max_seq_items_equal_to_n(self, idx):
+        # display.max_seq_items == n
+        with pd.option_context("display.max_seq_items", 6):
+            result = idx.__repr__()
+            expected = """\
+MultiIndex([('foo', 'one'),
+            ('foo', 'two'),
+            ('bar', 'one'),
+            ('baz', 'two'),
+            ('qux', 'one'),
+            ('qux', 'two')],
+           names=['first', 'second'])"""
+            assert result == expected
+
+    def test_repr(self, idx):
+        result = idx[:1].__repr__()
+        expected = """\
+MultiIndex([('foo', 'one')],
+           names=['first', 'second'])"""
+        assert result == expected
+
+        result = idx.__repr__()
+        expected = """\
+MultiIndex([('foo', 'one'),
+            ('foo', 'two'),
+            ('bar', 'one'),
+            ('baz', 'two'),
+            ('qux', 'one'),
+            ('qux', 'two')],
+           names=['first', 'second'])"""
+        assert result == expected
+
+        with pd.option_context("display.max_seq_items", 5):
+            result = idx.__repr__()
+            expected = """\
+MultiIndex([('foo', 'one'),
+            ('foo', 'two'),
+            ...
+            ('qux', 'one'),
+            ('qux', 'two')],
+           names=['first', 'second'], length=6)"""
+            assert result == expected
+
+        # display.max_seq_items == 1
+        with pd.option_context("display.max_seq_items", 1):
+            result = idx.__repr__()
+            expected = """\
+MultiIndex([...
+            ('qux', 'two')],
+           names=['first', ...], length=6)"""
+            assert result == expected
+
+    def test_rjust(self):
+        n = 1000
+        ci = pd.CategoricalIndex(list("a" * n) + (["abc"] * n))
+        dti = pd.date_range("2000-01-01", freq="s", periods=n * 2)
+        mi = MultiIndex.from_arrays([ci, ci.codes + 9, dti], names=["a", "b", "dti"])
+        result = mi[:1].__repr__()
+        expected = """\
+MultiIndex([('a', 9, '2000-01-01 00:00:00')],
+           names=['a', 'b', 'dti'])"""
+        assert result == expected
+
+        result = mi[::500].__repr__()
+        expected = """\
+MultiIndex([(  'a',  9, '2000-01-01 00:00:00'),
+            (  'a',  9, '2000-01-01 00:08:20'),
+            ('abc', 10, '2000-01-01 00:16:40'),
+            ('abc', 10, '2000-01-01 00:25:00')],
+           names=['a', 'b', 'dti'])"""
+        assert result == expected
+
+        result = mi.__repr__()
+        expected = """\
+MultiIndex([(  'a',  9, '2000-01-01 00:00:00'),
+            (  'a',  9, '2000-01-01 00:00:01'),
+            (  'a',  9, '2000-01-01 00:00:02'),
+            (  'a',  9, '2000-01-01 00:00:03'),
+            (  'a',  9, '2000-01-01 00:00:04'),
+            (  'a',  9, '2000-01-01 00:00:05'),
+            (  'a',  9, '2000-01-01 00:00:06'),
+            (  'a',  9, '2000-01-01 00:00:07'),
+            (  'a',  9, '2000-01-01 00:00:08'),
+            (  'a',  9, '2000-01-01 00:00:09'),
+            ...
+            ('abc', 10, '2000-01-01 00:33:10'),
+            ('abc', 10, '2000-01-01 00:33:11'),
+            ('abc', 10, '2000-01-01 00:33:12'),
+            ('abc', 10, '2000-01-01 00:33:13'),
+            ('abc', 10, '2000-01-01 00:33:14'),
+            ('abc', 10, '2000-01-01 00:33:15'),
+            ('abc', 10, '2000-01-01 00:33:16'),
+            ('abc', 10, '2000-01-01 00:33:17'),
+            ('abc', 10, '2000-01-01 00:33:18'),
+            ('abc', 10, '2000-01-01 00:33:19')],
+           names=['a', 'b', 'dti'], length=2000)"""
+        assert result == expected
+
+    def test_tuple_width(self):
+        n = 1000
+        ci = pd.CategoricalIndex(list("a" * n) + (["abc"] * n))
+        dti = pd.date_range("2000-01-01", freq="s", periods=n * 2)
+        levels = [ci, ci.codes + 9, dti, dti, dti]
+        names = ["a", "b", "dti_1", "dti_2", "dti_3"]
+        mi = MultiIndex.from_arrays(levels, names=names)
+        result = mi[:1].__repr__()
+        expected = """MultiIndex([('a', 9, '2000-01-01 00:00:00', '2000-01-01 00:00:00', ...)],
+           names=['a', 'b', 'dti_1', 'dti_2', 'dti_3'])"""  # noqa: E501
+        assert result == expected
+
+        result = mi[:10].__repr__()
+        expected = """\
+MultiIndex([('a', 9, '2000-01-01 00:00:00', '2000-01-01 00:00:00', ...),
+            ('a', 9, '2000-01-01 00:00:01', '2000-01-01 00:00:01', ...),
+            ('a', 9, '2000-01-01 00:00:02', '2000-01-01 00:00:02', ...),
+            ('a', 9, '2000-01-01 00:00:03', '2000-01-01 00:00:03', ...),
+            ('a', 9, '2000-01-01 00:00:04', '2000-01-01 00:00:04', ...),
+            ('a', 9, '2000-01-01 00:00:05', '2000-01-01 00:00:05', ...),
+            ('a', 9, '2000-01-01 00:00:06', '2000-01-01 00:00:06', ...),
+            ('a', 9, '2000-01-01 00:00:07', '2000-01-01 00:00:07', ...),
+            ('a', 9, '2000-01-01 00:00:08', '2000-01-01 00:00:08', ...),
+            ('a', 9, '2000-01-01 00:00:09', '2000-01-01 00:00:09', ...)],
+           names=['a', 'b', 'dti_1', 'dti_2', 'dti_3'])"""
+        assert result == expected
+
+        result = mi.__repr__()
+        expected = """\
+MultiIndex([(  'a',  9, '2000-01-01 00:00:00', '2000-01-01 00:00:00', ...),
+            (  'a',  9, '2000-01-01 00:00:01', '2000-01-01 00:00:01', ...),
+            (  'a',  9, '2000-01-01 00:00:02', '2000-01-01 00:00:02', ...),
+            (  'a',  9, '2000-01-01 00:00:03', '2000-01-01 00:00:03', ...),
+            (  'a',  9, '2000-01-01 00:00:04', '2000-01-01 00:00:04', ...),
+            (  'a',  9, '2000-01-01 00:00:05', '2000-01-01 00:00:05', ...),
+            (  'a',  9, '2000-01-01 00:00:06', '2000-01-01 00:00:06', ...),
+            (  'a',  9, '2000-01-01 00:00:07', '2000-01-01 00:00:07', ...),
+            (  'a',  9, '2000-01-01 00:00:08', '2000-01-01 00:00:08', ...),
+            (  'a',  9, '2000-01-01 00:00:09', '2000-01-01 00:00:09', ...),
+            ...
+            ('abc', 10, '2000-01-01 00:33:10', '2000-01-01 00:33:10', ...),
+            ('abc', 10, '2000-01-01 00:33:11', '2000-01-01 00:33:11', ...),
+            ('abc', 10, '2000-01-01 00:33:12', '2000-01-01 00:33:12', ...),
+            ('abc', 10, '2000-01-01 00:33:13', '2000-01-01 00:33:13', ...),
+            ('abc', 10, '2000-01-01 00:33:14', '2000-01-01 00:33:14', ...),
+            ('abc', 10, '2000-01-01 00:33:15', '2000-01-01 00:33:15', ...),
+            ('abc', 10, '2000-01-01 00:33:16', '2000-01-01 00:33:16', ...),
+            ('abc', 10, '2000-01-01 00:33:17', '2000-01-01 00:33:17', ...),
+            ('abc', 10, '2000-01-01 00:33:18', '2000-01-01 00:33:18', ...),
+            ('abc', 10, '2000-01-01 00:33:19', '2000-01-01 00:33:19', ...)],
+           names=['a', 'b', 'dti_1', 'dti_2', 'dti_3'], length=2000)"""
+        assert result == expected
+
+    def test_multiindex_long_element(self):
+        # Non-regression test towards GH#52960
+        data = MultiIndex.from_tuples([("c" * 62,)])
+
+        expected = (
+            "MultiIndex([('cccccccccccccccccccccccccccccccccccccccc"
+            "cccccccccccccccccccccc',)],\n           )"
+        )
+        assert str(data) == expected
diff --git a/pandas/pandas/tests/indexes/multi/test_get_level_values.py b/pandas/pandas/tests/indexes/multi/test_get_level_values.py
new file mode 100644
index 0000000000000..4db74a716c514
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_get_level_values.py
@@ -0,0 +1,133 @@
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    Index,
+    MultiIndex,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestGetLevelValues:
+    def test_get_level_values_box_datetime64(self):
+        dates = date_range("1/1/2000", periods=4)
+        levels = [dates, [0, 1]]
+        codes = [[0, 0, 1, 1, 2, 2, 3, 3], [0, 1, 0, 1, 0, 1, 0, 1]]
+
+        index = MultiIndex(levels=levels, codes=codes)
+
+        assert isinstance(index.get_level_values(0)[0], Timestamp)
+
+
+def test_get_level_values(idx):
+    result = idx.get_level_values(0)
+    expected = Index(["foo", "foo", "bar", "baz", "qux", "qux"], name="first")
+    tm.assert_index_equal(result, expected)
+    assert result.name == "first"
+
+    result = idx.get_level_values("first")
+    expected = idx.get_level_values(0)
+    tm.assert_index_equal(result, expected)
+
+    # GH 10460
+    index = MultiIndex(
+        levels=[CategoricalIndex(["A", "B"]), CategoricalIndex([1, 2, 3])],
+        codes=[np.array([0, 0, 0, 1, 1, 1]), np.array([0, 1, 2, 0, 1, 2])],
+    )
+
+    exp = CategoricalIndex(["A", "A", "A", "B", "B", "B"])
+    tm.assert_index_equal(index.get_level_values(0), exp)
+    exp = CategoricalIndex([1, 2, 3, 1, 2, 3])
+    tm.assert_index_equal(index.get_level_values(1), exp)
+
+
+def test_get_level_values_all_na():
+    # GH#17924 when level entirely consists of nan
+    arrays = [[np.nan, np.nan, np.nan], ["a", np.nan, 1]]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(0)
+    expected = Index([np.nan, np.nan, np.nan], dtype=np.float64)
+    tm.assert_index_equal(result, expected)
+
+    result = index.get_level_values(1)
+    expected = Index(["a", np.nan, 1], dtype=object)
+    tm.assert_index_equal(result, expected)
+
+
+def test_get_level_values_int_with_na():
+    # GH#17924
+    arrays = [["a", "b", "b"], [1, np.nan, 2]]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(1)
+    expected = Index([1, np.nan, 2])
+    tm.assert_index_equal(result, expected)
+
+    arrays = [["a", "b", "b"], [np.nan, np.nan, 2]]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(1)
+    expected = Index([np.nan, np.nan, 2])
+    tm.assert_index_equal(result, expected)
+
+
+def test_get_level_values_na():
+    arrays = [[np.nan, np.nan, np.nan], ["a", np.nan, 1]]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(0)
+    expected = Index([np.nan, np.nan, np.nan])
+    tm.assert_index_equal(result, expected)
+
+    result = index.get_level_values(1)
+    expected = Index(["a", np.nan, 1])
+    tm.assert_index_equal(result, expected)
+
+    arrays = [["a", "b", "b"], pd.DatetimeIndex([0, 1, pd.NaT])]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(1)
+    expected = pd.DatetimeIndex([0, 1, pd.NaT])
+    tm.assert_index_equal(result, expected)
+
+    arrays = [[], []]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.get_level_values(0)
+    expected = Index([], dtype=object)
+    tm.assert_index_equal(result, expected)
+
+
+def test_get_level_values_when_periods():
+    # GH33131. See also discussion in GH32669.
+    # This test can probably be removed when PeriodIndex._engine is removed.
+    from pandas import (
+        Period,
+        PeriodIndex,
+    )
+
+    idx = MultiIndex.from_arrays(
+        [PeriodIndex([Period("2019Q1"), Period("2019Q2")], name="b")]
+    )
+    idx2 = MultiIndex.from_arrays(
+        [idx._get_level_values(level) for level in range(idx.nlevels)]
+    )
+    assert all(x.is_monotonic_increasing for x in idx2.levels)
+
+
+def test_values_loses_freq_of_underlying_index():
+    # GH#49054
+    idx = pd.DatetimeIndex(date_range("20200101", periods=3, freq="BME"))
+    expected = idx.copy(deep=True)
+    idx2 = Index([1, 2, 3])
+    midx = MultiIndex(levels=[idx, idx2], codes=[[0, 1, 2], [0, 1, 2]])
+    midx.values
+    assert idx.freq is not None
+    tm.assert_index_equal(idx, expected)
+
+
+def test_get_level_values_gets_frequency_correctly():
+    # GH#57949 GH#58327
+    datetime_index = date_range(start=pd.to_datetime("1/1/2018"), periods=4, freq="YS")
+    other_index = ["A"]
+    multi_index = MultiIndex.from_product([datetime_index, other_index])
+
+    assert multi_index.get_level_values(0).freq == datetime_index.freq
diff --git a/pandas/pandas/tests/indexes/multi/test_get_set.py b/pandas/pandas/tests/indexes/multi/test_get_set.py
new file mode 100644
index 0000000000000..7f292aacc39ca
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_get_set.py
@@ -0,0 +1,382 @@
+import numpy as np
+import pytest
+
+from pandas.compat import PY311
+
+from pandas.core.dtypes.dtypes import DatetimeTZDtype
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+def assert_matching(actual, expected, check_dtype=False):
+    # avoid specifying internal representation
+    # as much as possible
+    assert len(actual) == len(expected)
+    for act, exp in zip(actual, expected):
+        act = np.asarray(act)
+        exp = np.asarray(exp)
+        tm.assert_numpy_array_equal(act, exp, check_dtype=check_dtype)
+
+
+def test_get_level_number_integer(idx):
+    idx.names = [1, 0]
+    assert idx._get_level_number(1) == 0
+    assert idx._get_level_number(0) == 1
+    msg = "Too many levels: Index has only 2 levels, not 3"
+    with pytest.raises(IndexError, match=msg):
+        idx._get_level_number(2)
+    with pytest.raises(KeyError, match="Level fourth not found"):
+        idx._get_level_number("fourth")
+
+
+def test_get_dtypes(using_infer_string):
+    # Test MultiIndex.dtypes (# Gh37062)
+    idx_multitype = MultiIndex.from_product(
+        [[1, 2, 3], ["a", "b", "c"], pd.date_range("20200101", periods=2, tz="UTC")],
+        names=["int", "string", "dt"],
+    )
+
+    exp = "object" if not using_infer_string else pd.StringDtype(na_value=np.nan)
+    expected = pd.Series(
+        {
+            "int": np.dtype("int64"),
+            "string": exp,
+            "dt": DatetimeTZDtype(tz="utc"),
+        }
+    )
+    tm.assert_series_equal(expected, idx_multitype.dtypes)
+
+
+def test_get_dtypes_no_level_name(using_infer_string):
+    # Test MultiIndex.dtypes (# GH38580 )
+    idx_multitype = MultiIndex.from_product(
+        [
+            [1, 2, 3],
+            ["a", "b", "c"],
+            pd.date_range("20200101", periods=2, tz="UTC"),
+        ],
+    )
+    exp = "object" if not using_infer_string else pd.StringDtype(na_value=np.nan)
+    expected = pd.Series(
+        {
+            "level_0": np.dtype("int64"),
+            "level_1": exp,
+            "level_2": DatetimeTZDtype(tz="utc"),
+        }
+    )
+    tm.assert_series_equal(expected, idx_multitype.dtypes)
+
+
+def test_get_dtypes_duplicate_level_names(using_infer_string):
+    # Test MultiIndex.dtypes with non-unique level names (# GH45174)
+    result = MultiIndex.from_product(
+        [
+            [1, 2, 3],
+            ["a", "b", "c"],
+            pd.date_range("20200101", periods=2, tz="UTC"),
+        ],
+        names=["A", "A", "A"],
+    ).dtypes
+    exp = "object" if not using_infer_string else pd.StringDtype(na_value=np.nan)
+    expected = pd.Series(
+        [np.dtype("int64"), exp, DatetimeTZDtype(tz="utc")],
+        index=["A", "A", "A"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_get_level_number_out_of_bounds(multiindex_dataframe_random_data):
+    frame = multiindex_dataframe_random_data
+
+    with pytest.raises(IndexError, match="Too many levels"):
+        frame.index._get_level_number(2)
+    with pytest.raises(IndexError, match="not a valid level number"):
+        frame.index._get_level_number(-3)
+
+
+def test_set_name_methods(idx):
+    # so long as these are synonyms, we don't need to test set_names
+    index_names = ["first", "second"]
+    assert idx.rename == idx.set_names
+    new_names = [name + "SUFFIX" for name in index_names]
+    ind = idx.set_names(new_names)
+    assert idx.names == index_names
+    assert ind.names == new_names
+    msg = "Length of names must match number of levels in MultiIndex"
+    with pytest.raises(ValueError, match=msg):
+        ind.set_names(new_names + new_names)
+    new_names2 = [name + "SUFFIX2" for name in new_names]
+    res = ind.set_names(new_names2, inplace=True)
+    assert res is None
+    assert ind.names == new_names2
+
+    # set names for specific level (# GH7792)
+    ind = idx.set_names(new_names[0], level=0)
+    assert idx.names == index_names
+    assert ind.names == [new_names[0], index_names[1]]
+
+    res = ind.set_names(new_names2[0], level=0, inplace=True)
+    assert res is None
+    assert ind.names == [new_names2[0], index_names[1]]
+
+    # set names for multiple levels
+    ind = idx.set_names(new_names, level=[0, 1])
+    assert idx.names == index_names
+    assert ind.names == new_names
+
+    res = ind.set_names(new_names2, level=[0, 1], inplace=True)
+    assert res is None
+    assert ind.names == new_names2
+
+
+def test_set_levels_codes_directly(idx):
+    # setting levels/codes directly raises AttributeError
+
+    levels = idx.levels
+    new_levels = [[lev + "a" for lev in level] for level in levels]
+
+    codes = idx.codes
+    major_codes, minor_codes = codes
+    major_codes = [(x + 1) % 3 for x in major_codes]
+    minor_codes = [(x + 1) % 1 for x in minor_codes]
+    new_codes = [major_codes, minor_codes]
+
+    msg = "Can't set attribute"
+    with pytest.raises(AttributeError, match=msg):
+        idx.levels = new_levels
+
+    msg = (
+        "property 'codes' of 'MultiIndex' object has no setter"
+        if PY311
+        else "can't set attribute"
+    )
+    with pytest.raises(AttributeError, match=msg):
+        idx.codes = new_codes
+
+
+def test_set_levels(idx):
+    # side note - you probably wouldn't want to use levels and codes
+    # directly like this - but it is possible.
+    levels = idx.levels
+    new_levels = [[lev + "a" for lev in level] for level in levels]
+
+    # level changing [w/o mutation]
+    ind2 = idx.set_levels(new_levels)
+    assert_matching(ind2.levels, new_levels)
+    assert_matching(idx.levels, levels)
+
+    # level changing specific level [w/o mutation]
+    ind2 = idx.set_levels(new_levels[0], level=0)
+    assert_matching(ind2.levels, [new_levels[0], levels[1]])
+    assert_matching(idx.levels, levels)
+
+    ind2 = idx.set_levels(new_levels[1], level=1)
+    assert_matching(ind2.levels, [levels[0], new_levels[1]])
+    assert_matching(idx.levels, levels)
+
+    # level changing multiple levels [w/o mutation]
+    ind2 = idx.set_levels(new_levels, level=[0, 1])
+    assert_matching(ind2.levels, new_levels)
+    assert_matching(idx.levels, levels)
+
+    # illegal level changing should not change levels
+    # GH 13754
+    original_index = idx.copy()
+    with pytest.raises(ValueError, match="^On"):
+        idx.set_levels(["c"], level=0)
+    assert_matching(idx.levels, original_index.levels, check_dtype=True)
+
+    with pytest.raises(ValueError, match="^On"):
+        idx.set_codes([0, 1, 2, 3, 4, 5], level=0)
+    assert_matching(idx.codes, original_index.codes, check_dtype=True)
+
+    with pytest.raises(TypeError, match="^Levels"):
+        idx.set_levels("c", level=0)
+    assert_matching(idx.levels, original_index.levels, check_dtype=True)
+
+    with pytest.raises(TypeError, match="^Codes"):
+        idx.set_codes(1, level=0)
+    assert_matching(idx.codes, original_index.codes, check_dtype=True)
+
+
+def test_set_codes(idx):
+    # side note - you probably wouldn't want to use levels and codes
+    # directly like this - but it is possible.
+    codes = idx.codes
+    major_codes, minor_codes = codes
+    major_codes = [(x + 1) % 3 for x in major_codes]
+    minor_codes = [(x + 1) % 1 for x in minor_codes]
+    new_codes = [major_codes, minor_codes]
+
+    # changing codes w/o mutation
+    ind2 = idx.set_codes(new_codes)
+    assert_matching(ind2.codes, new_codes)
+    assert_matching(idx.codes, codes)
+
+    # codes changing specific level w/o mutation
+    ind2 = idx.set_codes(new_codes[0], level=0)
+    assert_matching(ind2.codes, [new_codes[0], codes[1]])
+    assert_matching(idx.codes, codes)
+
+    ind2 = idx.set_codes(new_codes[1], level=1)
+    assert_matching(ind2.codes, [codes[0], new_codes[1]])
+    assert_matching(idx.codes, codes)
+
+    # codes changing multiple levels w/o mutation
+    ind2 = idx.set_codes(new_codes, level=[0, 1])
+    assert_matching(ind2.codes, new_codes)
+    assert_matching(idx.codes, codes)
+
+    # label changing for levels of different magnitude of categories
+    ind = MultiIndex.from_tuples([(0, i) for i in range(130)])
+    new_codes = range(129, -1, -1)
+    expected = MultiIndex.from_tuples([(0, i) for i in new_codes])
+
+    # [w/o mutation]
+    result = ind.set_codes(codes=new_codes, level=1)
+    assert result.equals(expected)
+
+
+def test_set_levels_codes_names_bad_input(idx):
+    levels, codes = idx.levels, idx.codes
+    names = idx.names
+
+    with pytest.raises(ValueError, match="Length of levels"):
+        idx.set_levels([levels[0]])
+
+    with pytest.raises(ValueError, match="Length of codes"):
+        idx.set_codes([codes[0]])
+
+    with pytest.raises(ValueError, match="Length of names"):
+        idx.set_names([names[0]])
+
+    # shouldn't scalar data error, instead should demand list-like
+    with pytest.raises(TypeError, match="list of lists-like"):
+        idx.set_levels(levels[0])
+
+    # shouldn't scalar data error, instead should demand list-like
+    with pytest.raises(TypeError, match="list of lists-like"):
+        idx.set_codes(codes[0])
+
+    # shouldn't scalar data error, instead should demand list-like
+    with pytest.raises(TypeError, match="list-like"):
+        idx.set_names(names[0])
+
+    # should have equal lengths
+    with pytest.raises(TypeError, match="list of lists-like"):
+        idx.set_levels(levels[0], level=[0, 1])
+
+    with pytest.raises(TypeError, match="list-like"):
+        idx.set_levels(levels, level=0)
+
+    # should have equal lengths
+    with pytest.raises(TypeError, match="list of lists-like"):
+        idx.set_codes(codes[0], level=[0, 1])
+
+    with pytest.raises(TypeError, match="list-like"):
+        idx.set_codes(codes, level=0)
+
+    # should have equal lengths
+    with pytest.raises(ValueError, match="Length of names"):
+        idx.set_names(names[0], level=[0, 1])
+
+    with pytest.raises(TypeError, match="Names must be a"):
+        idx.set_names(names, level=0)
+
+
+@pytest.mark.parametrize("inplace", [True, False])
+def test_set_names_with_nlevel_1(inplace):
+    # GH 21149
+    # Ensure that .set_names for MultiIndex with
+    # nlevels == 1 does not raise any errors
+    expected = MultiIndex(levels=[[0, 1]], codes=[[0, 1]], names=["first"])
+    m = MultiIndex.from_product([[0, 1]])
+    result = m.set_names("first", level=0, inplace=inplace)
+
+    if inplace:
+        result = m
+
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_set_levels_categorical(ordered):
+    # GH13854
+    index = MultiIndex.from_arrays([list("xyzx"), [0, 1, 2, 3]])
+
+    cidx = CategoricalIndex(list("bac"), ordered=ordered)
+    result = index.set_levels(cidx, level=0)
+    expected = MultiIndex(levels=[cidx, [0, 1, 2, 3]], codes=index.codes)
+    tm.assert_index_equal(result, expected)
+
+    result_lvl = result.get_level_values(0)
+    expected_lvl = CategoricalIndex(
+        list("bacb"), categories=cidx.categories, ordered=cidx.ordered
+    )
+    tm.assert_index_equal(result_lvl, expected_lvl)
+
+
+def test_set_value_keeps_names():
+    # motivating example from #3742
+    lev1 = ["hans", "hans", "hans", "grethe", "grethe", "grethe"]
+    lev2 = ["1", "2", "3"] * 2
+    idx = MultiIndex.from_arrays([lev1, lev2], names=["Name", "Number"])
+    df = pd.DataFrame(
+        np.random.default_rng(2).standard_normal((6, 4)),
+        columns=["one", "two", "three", "four"],
+        index=idx,
+    )
+    df = df.sort_index()
+    assert df.index.names == ("Name", "Number")
+    df.at[("grethe", "4"), "one"] = 99.34
+    assert df.index.names == ("Name", "Number")
+
+
+def test_set_levels_with_iterable():
+    # GH23273
+    sizes = [1, 2, 3]
+    colors = ["black"] * 3
+    index = MultiIndex.from_arrays([sizes, colors], names=["size", "color"])
+
+    result = index.set_levels(map(int, ["3", "2", "1"]), level="size")
+
+    expected_sizes = [3, 2, 1]
+    expected = MultiIndex.from_arrays([expected_sizes, colors], names=["size", "color"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_set_empty_level():
+    # GH#48636
+    midx = MultiIndex.from_arrays([[]], names=["A"])
+    result = midx.set_levels(pd.DatetimeIndex([]), level=0)
+    expected = MultiIndex.from_arrays([pd.DatetimeIndex([])], names=["A"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_set_levels_pos_args_removal():
+    # https://github.com/pandas-dev/pandas/issues/41485
+    idx = MultiIndex.from_tuples(
+        [
+            (1, "one"),
+            (3, "one"),
+        ],
+        names=["foo", "bar"],
+    )
+    with pytest.raises(TypeError, match="positional arguments"):
+        idx.set_levels(["a", "b", "c"], 0)
+
+    with pytest.raises(TypeError, match="positional arguments"):
+        idx.set_codes([[0, 1], [1, 0]], 0)
+
+
+def test_set_levels_categorical_keep_dtype():
+    # GH#52125
+    midx = MultiIndex.from_arrays([[5, 6]])
+    result = midx.set_levels(levels=pd.Categorical([1, 2]), level=0)
+    expected = MultiIndex.from_arrays([pd.Categorical([1, 2])])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_indexing.py b/pandas/pandas/tests/indexes/multi/test_indexing.py
new file mode 100644
index 0000000000000..f098690be2afa
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_indexing.py
@@ -0,0 +1,1031 @@
+from collections import namedtuple
+from datetime import timedelta
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+from pandas.errors import InvalidIndexError
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSliceLocs:
+    def test_slice_locs_partial(self, idx):
+        sorted_idx, _ = idx.sortlevel(0)
+
+        result = sorted_idx.slice_locs(("foo", "two"), ("qux", "one"))
+        assert result == (1, 5)
+
+        result = sorted_idx.slice_locs(None, ("qux", "one"))
+        assert result == (0, 5)
+
+        result = sorted_idx.slice_locs(("foo", "two"), None)
+        assert result == (1, len(sorted_idx))
+
+        result = sorted_idx.slice_locs("bar", "baz")
+        assert result == (2, 4)
+
+    def test_slice_locs(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((50, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=50, freq="B"),
+        )
+        stacked = df.stack()
+        idx = stacked.index
+
+        slob = slice(*idx.slice_locs(df.index[5], df.index[15]))
+        sliced = stacked[slob]
+        expected = df[5:16].stack()
+        tm.assert_almost_equal(sliced.values, expected.values)
+
+        slob = slice(
+            *idx.slice_locs(
+                df.index[5] + timedelta(seconds=30),
+                df.index[15] - timedelta(seconds=30),
+            )
+        )
+        sliced = stacked[slob]
+        expected = df[6:15].stack()
+        tm.assert_almost_equal(sliced.values, expected.values)
+
+    def test_slice_locs_with_type_mismatch(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        stacked = df.stack()
+        idx = stacked.index
+        with pytest.raises(TypeError, match="^Level type mismatch"):
+            idx.slice_locs((1, 3))
+        with pytest.raises(TypeError, match="^Level type mismatch"):
+            idx.slice_locs(df.index[5] + timedelta(seconds=30), (5, 2))
+        df = DataFrame(
+            np.ones((5, 5)),
+            index=Index([f"i-{i}" for i in range(5)], name="a"),
+            columns=Index([f"i-{i}" for i in range(5)], name="a"),
+        )
+        stacked = df.stack()
+        idx = stacked.index
+        with pytest.raises(TypeError, match="^Level type mismatch"):
+            idx.slice_locs(timedelta(seconds=30))
+        # TODO: Try creating a UnicodeDecodeError in exception message
+        with pytest.raises(TypeError, match="^Level type mismatch"):
+            idx.slice_locs(df.index[1], (16, "a"))
+
+    def test_slice_locs_not_sorted(self):
+        index = MultiIndex(
+            levels=[Index(np.arange(4)), Index(np.arange(4)), Index(np.arange(4))],
+            codes=[
+                np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+                np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+                np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+            ],
+        )
+        msg = "[Kk]ey length.*greater than MultiIndex lexsort depth"
+        with pytest.raises(KeyError, match=msg):
+            index.slice_locs((1, 0, 1), (2, 1, 0))
+
+        # works
+        sorted_index, _ = index.sortlevel(0)
+        # should there be a test case here???
+        sorted_index.slice_locs((1, 0, 1), (2, 1, 0))
+
+    def test_slice_locs_not_contained(self):
+        # some searchsorted action
+
+        index = MultiIndex(
+            levels=[[0, 2, 4, 6], [0, 2, 4]],
+            codes=[[0, 0, 0, 1, 1, 2, 3, 3, 3], [0, 1, 2, 1, 2, 2, 0, 1, 2]],
+        )
+
+        result = index.slice_locs((1, 0), (5, 2))
+        assert result == (3, 6)
+
+        result = index.slice_locs(1, 5)
+        assert result == (3, 6)
+
+        result = index.slice_locs((2, 2), (5, 2))
+        assert result == (3, 6)
+
+        result = index.slice_locs(2, 5)
+        assert result == (3, 6)
+
+        result = index.slice_locs((1, 0), (6, 3))
+        assert result == (3, 8)
+
+        result = index.slice_locs(-1, 10)
+        assert result == (0, len(index))
+
+    @pytest.mark.parametrize(
+        "index_arr,expected,start_idx,end_idx",
+        [
+            ([[np.nan, "a", "b"], ["c", "d", "e"]], (0, 3), np.nan, None),
+            ([[np.nan, "a", "b"], ["c", "d", "e"]], (0, 3), np.nan, "b"),
+            ([[np.nan, "a", "b"], ["c", "d", "e"]], (0, 3), np.nan, ("b", "e")),
+            ([["a", "b", "c"], ["d", np.nan, "e"]], (1, 3), ("b", np.nan), None),
+            ([["a", "b", "c"], ["d", np.nan, "e"]], (1, 3), ("b", np.nan), "c"),
+            ([["a", "b", "c"], ["d", np.nan, "e"]], (1, 3), ("b", np.nan), ("c", "e")),
+        ],
+    )
+    def test_slice_locs_with_missing_value(
+        self, index_arr, expected, start_idx, end_idx
+    ):
+        # issue 19132
+        idx = MultiIndex.from_arrays(index_arr)
+        result = idx.slice_locs(start=start_idx, end=end_idx)
+        assert result == expected
+
+
+class TestPutmask:
+    def test_putmask_with_wrong_mask(self, idx):
+        # GH18368
+
+        msg = "putmask: mask and data must be the same size"
+        with pytest.raises(ValueError, match=msg):
+            idx.putmask(np.ones(len(idx) + 1, np.bool_), 1)
+
+        with pytest.raises(ValueError, match=msg):
+            idx.putmask(np.ones(len(idx) - 1, np.bool_), 1)
+
+        with pytest.raises(ValueError, match=msg):
+            idx.putmask("foo", 1)
+
+    def test_putmask_multiindex_other(self):
+        # GH#43212 `value` is also a MultiIndex
+
+        left = MultiIndex.from_tuples([(np.nan, 6), (np.nan, 6), ("a", 4)])
+        right = MultiIndex.from_tuples([("a", 1), ("a", 1), ("d", 1)])
+        mask = np.array([True, True, False])
+
+        result = left.putmask(mask, right)
+
+        expected = MultiIndex.from_tuples([right[0], right[1], left[2]])
+        tm.assert_index_equal(result, expected)
+
+    def test_putmask_keep_dtype(self, any_numeric_ea_dtype):
+        # GH#49830
+        midx = MultiIndex.from_arrays(
+            [pd.Series([1, 2, 3], dtype=any_numeric_ea_dtype), [10, 11, 12]]
+        )
+        midx2 = MultiIndex.from_arrays(
+            [pd.Series([5, 6, 7], dtype=any_numeric_ea_dtype), [-1, -2, -3]]
+        )
+        result = midx.putmask([True, False, False], midx2)
+        expected = MultiIndex.from_arrays(
+            [pd.Series([5, 2, 3], dtype=any_numeric_ea_dtype), [-1, 11, 12]]
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_putmask_keep_dtype_shorter_value(self, any_numeric_ea_dtype):
+        # GH#49830
+        midx = MultiIndex.from_arrays(
+            [pd.Series([1, 2, 3], dtype=any_numeric_ea_dtype), [10, 11, 12]]
+        )
+        midx2 = MultiIndex.from_arrays(
+            [pd.Series([5], dtype=any_numeric_ea_dtype), [-1]]
+        )
+        result = midx.putmask([True, False, False], midx2)
+        expected = MultiIndex.from_arrays(
+            [pd.Series([5, 2, 3], dtype=any_numeric_ea_dtype), [-1, 11, 12]]
+        )
+        tm.assert_index_equal(result, expected)
+
+
+class TestGetIndexer:
+    def test_get_indexer(self):
+        major_axis = Index(np.arange(4))
+        minor_axis = Index(np.arange(2))
+
+        major_codes = np.array([0, 0, 1, 2, 2, 3, 3], dtype=np.intp)
+        minor_codes = np.array([0, 1, 0, 0, 1, 0, 1], dtype=np.intp)
+
+        index = MultiIndex(
+            levels=[major_axis, minor_axis], codes=[major_codes, minor_codes]
+        )
+        idx1 = index[:5]
+        idx2 = index[[1, 3, 5]]
+
+        r1 = idx1.get_indexer(idx2)
+        tm.assert_almost_equal(r1, np.array([1, 3, -1], dtype=np.intp))
+
+        r1 = idx2.get_indexer(idx1, method="pad")
+        e1 = np.array([-1, 0, 0, 1, 1], dtype=np.intp)
+        tm.assert_almost_equal(r1, e1)
+
+        r2 = idx2.get_indexer(idx1[::-1], method="pad")
+        tm.assert_almost_equal(r2, e1[::-1])
+
+        rffill1 = idx2.get_indexer(idx1, method="ffill")
+        tm.assert_almost_equal(r1, rffill1)
+
+        r1 = idx2.get_indexer(idx1, method="backfill")
+        e1 = np.array([0, 0, 1, 1, 2], dtype=np.intp)
+        tm.assert_almost_equal(r1, e1)
+
+        r2 = idx2.get_indexer(idx1[::-1], method="backfill")
+        tm.assert_almost_equal(r2, e1[::-1])
+
+        rbfill1 = idx2.get_indexer(idx1, method="bfill")
+        tm.assert_almost_equal(r1, rbfill1)
+
+        # pass non-MultiIndex
+        r1 = idx1.get_indexer(idx2.values)
+        rexp1 = idx1.get_indexer(idx2)
+        tm.assert_almost_equal(r1, rexp1)
+
+        r1 = idx1.get_indexer([1, 2, 3])
+        assert (r1 == [-1, -1, -1]).all()
+
+        # create index with duplicates
+        idx1 = Index(list(range(10)) + list(range(10)))
+        idx2 = Index(list(range(20)))
+
+        msg = "Reindexing only valid with uniquely valued Index objects"
+        with pytest.raises(InvalidIndexError, match=msg):
+            idx1.get_indexer(idx2)
+
+    def test_get_indexer_nearest(self):
+        midx = MultiIndex.from_tuples([("a", 1), ("b", 2)])
+        msg = (
+            "method='nearest' not implemented yet for MultiIndex; see GitHub issue 9365"
+        )
+        with pytest.raises(NotImplementedError, match=msg):
+            midx.get_indexer(["a"], method="nearest")
+        msg = "tolerance not implemented yet for MultiIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            midx.get_indexer(["a"], method="pad", tolerance=2)
+
+    def test_get_indexer_categorical_time(self):
+        # https://github.com/pandas-dev/pandas/issues/21390
+        midx = MultiIndex.from_product(
+            [
+                Categorical(["a", "b", "c"]),
+                Categorical(date_range("2012-01-01", periods=3, freq="h")),
+            ]
+        )
+        result = midx.get_indexer(midx)
+        tm.assert_numpy_array_equal(result, np.arange(9, dtype=np.intp))
+
+    @pytest.mark.parametrize(
+        "index_arr,labels,expected",
+        [
+            (
+                [[1, np.nan, 2], [3, 4, 5]],
+                [1, np.nan, 2],
+                np.array([-1, -1, -1], dtype=np.intp),
+            ),
+            ([[1, np.nan, 2], [3, 4, 5]], [(np.nan, 4)], np.array([1], dtype=np.intp)),
+            ([[1, 2, 3], [np.nan, 4, 5]], [(1, np.nan)], np.array([0], dtype=np.intp)),
+            (
+                [[1, 2, 3], [np.nan, 4, 5]],
+                [np.nan, 4, 5],
+                np.array([-1, -1, -1], dtype=np.intp),
+            ),
+        ],
+    )
+    def test_get_indexer_with_missing_value(self, index_arr, labels, expected):
+        # issue 19132
+        idx = MultiIndex.from_arrays(index_arr)
+        result = idx.get_indexer(labels)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_methods(self):
+        # https://github.com/pandas-dev/pandas/issues/29896
+        # test getting an indexer for another index with different methods
+        # confirms that getting an indexer without a filling method, getting an
+        # indexer and backfilling, and getting an indexer and padding all behave
+        # correctly in the case where all of the target values fall in between
+        # several levels in the MultiIndex into which they are getting an indexer
+        #
+        # visually, the MultiIndexes used in this test are:
+        # mult_idx_1:
+        #  0: -1 0
+        #  1:    2
+        #  2:    3
+        #  3:    4
+        #  4:  0 0
+        #  5:    2
+        #  6:    3
+        #  7:    4
+        #  8:  1 0
+        #  9:    2
+        # 10:    3
+        # 11:    4
+        #
+        # mult_idx_2:
+        #  0: 0 1
+        #  1:   3
+        #  2:   4
+        mult_idx_1 = MultiIndex.from_product([[-1, 0, 1], [0, 2, 3, 4]])
+        mult_idx_2 = MultiIndex.from_product([[0], [1, 3, 4]])
+
+        indexer = mult_idx_1.get_indexer(mult_idx_2)
+        expected = np.array([-1, 6, 7], dtype=indexer.dtype)
+        tm.assert_almost_equal(expected, indexer)
+
+        backfill_indexer = mult_idx_1.get_indexer(mult_idx_2, method="backfill")
+        expected = np.array([5, 6, 7], dtype=backfill_indexer.dtype)
+        tm.assert_almost_equal(expected, backfill_indexer)
+
+        # ensure the legacy "bfill" option functions identically to "backfill"
+        backfill_indexer = mult_idx_1.get_indexer(mult_idx_2, method="bfill")
+        expected = np.array([5, 6, 7], dtype=backfill_indexer.dtype)
+        tm.assert_almost_equal(expected, backfill_indexer)
+
+        pad_indexer = mult_idx_1.get_indexer(mult_idx_2, method="pad")
+        expected = np.array([4, 6, 7], dtype=pad_indexer.dtype)
+        tm.assert_almost_equal(expected, pad_indexer)
+
+        # ensure the legacy "ffill" option functions identically to "pad"
+        pad_indexer = mult_idx_1.get_indexer(mult_idx_2, method="ffill")
+        expected = np.array([4, 6, 7], dtype=pad_indexer.dtype)
+        tm.assert_almost_equal(expected, pad_indexer)
+
+    @pytest.mark.parametrize("method", ["pad", "ffill", "backfill", "bfill", "nearest"])
+    def test_get_indexer_methods_raise_for_non_monotonic(self, method):
+        # 53452
+        mi = MultiIndex.from_arrays([[0, 4, 2], [0, 4, 2]])
+        if method == "nearest":
+            err = NotImplementedError
+            msg = "not implemented yet for MultiIndex"
+        else:
+            err = ValueError
+            msg = "index must be monotonic increasing or decreasing"
+        with pytest.raises(err, match=msg):
+            mi.get_indexer([(1, 1)], method=method)
+
+    def test_get_indexer_three_or_more_levels(self):
+        # https://github.com/pandas-dev/pandas/issues/29896
+        # tests get_indexer() on MultiIndexes with 3+ levels
+        # visually, these are
+        # mult_idx_1:
+        #  0: 1 2 5
+        #  1:     7
+        #  2:   4 5
+        #  3:     7
+        #  4:   6 5
+        #  5:     7
+        #  6: 3 2 5
+        #  7:     7
+        #  8:   4 5
+        #  9:     7
+        # 10:   6 5
+        # 11:     7
+        #
+        # mult_idx_2:
+        #  0: 1 1 8
+        #  1: 1 5 9
+        #  2: 1 6 7
+        #  3: 2 1 6
+        #  4: 2 7 6
+        #  5: 2 7 8
+        #  6: 3 6 8
+        mult_idx_1 = MultiIndex.from_product([[1, 3], [2, 4, 6], [5, 7]])
+        mult_idx_2 = MultiIndex.from_tuples(
+            [
+                (1, 1, 8),
+                (1, 5, 9),
+                (1, 6, 7),
+                (2, 1, 6),
+                (2, 7, 7),
+                (2, 7, 8),
+                (3, 6, 8),
+            ]
+        )
+        # sanity check
+        assert mult_idx_1.is_monotonic_increasing
+        assert mult_idx_1.is_unique
+        assert mult_idx_2.is_monotonic_increasing
+        assert mult_idx_2.is_unique
+
+        # show the relationships between the two
+        assert mult_idx_2[0] < mult_idx_1[0]
+        assert mult_idx_1[3] < mult_idx_2[1] < mult_idx_1[4]
+        assert mult_idx_1[5] == mult_idx_2[2]
+        assert mult_idx_1[5] < mult_idx_2[3] < mult_idx_1[6]
+        assert mult_idx_1[5] < mult_idx_2[4] < mult_idx_1[6]
+        assert mult_idx_1[5] < mult_idx_2[5] < mult_idx_1[6]
+        assert mult_idx_1[-1] < mult_idx_2[6]
+
+        indexer_no_fill = mult_idx_1.get_indexer(mult_idx_2)
+        expected = np.array([-1, -1, 5, -1, -1, -1, -1], dtype=indexer_no_fill.dtype)
+        tm.assert_almost_equal(expected, indexer_no_fill)
+
+        # test with backfilling
+        indexer_backfilled = mult_idx_1.get_indexer(mult_idx_2, method="backfill")
+        expected = np.array([0, 4, 5, 6, 6, 6, -1], dtype=indexer_backfilled.dtype)
+        tm.assert_almost_equal(expected, indexer_backfilled)
+
+        # now, the same thing, but forward-filled (aka "padded")
+        indexer_padded = mult_idx_1.get_indexer(mult_idx_2, method="pad")
+        expected = np.array([-1, 3, 5, 5, 5, 5, 11], dtype=indexer_padded.dtype)
+        tm.assert_almost_equal(expected, indexer_padded)
+
+        # now, do the indexing in the other direction
+        assert mult_idx_2[0] < mult_idx_1[0] < mult_idx_2[1]
+        assert mult_idx_2[0] < mult_idx_1[1] < mult_idx_2[1]
+        assert mult_idx_2[0] < mult_idx_1[2] < mult_idx_2[1]
+        assert mult_idx_2[0] < mult_idx_1[3] < mult_idx_2[1]
+        assert mult_idx_2[1] < mult_idx_1[4] < mult_idx_2[2]
+        assert mult_idx_2[2] == mult_idx_1[5]
+        assert mult_idx_2[5] < mult_idx_1[6] < mult_idx_2[6]
+        assert mult_idx_2[5] < mult_idx_1[7] < mult_idx_2[6]
+        assert mult_idx_2[5] < mult_idx_1[8] < mult_idx_2[6]
+        assert mult_idx_2[5] < mult_idx_1[9] < mult_idx_2[6]
+        assert mult_idx_2[5] < mult_idx_1[10] < mult_idx_2[6]
+        assert mult_idx_2[5] < mult_idx_1[11] < mult_idx_2[6]
+
+        indexer = mult_idx_2.get_indexer(mult_idx_1)
+        expected = np.array(
+            [-1, -1, -1, -1, -1, 2, -1, -1, -1, -1, -1, -1], dtype=indexer.dtype
+        )
+        tm.assert_almost_equal(expected, indexer)
+
+        backfill_indexer = mult_idx_2.get_indexer(mult_idx_1, method="bfill")
+        expected = np.array(
+            [1, 1, 1, 1, 2, 2, 6, 6, 6, 6, 6, 6], dtype=backfill_indexer.dtype
+        )
+        tm.assert_almost_equal(expected, backfill_indexer)
+
+        pad_indexer = mult_idx_2.get_indexer(mult_idx_1, method="pad")
+        expected = np.array(
+            [0, 0, 0, 0, 1, 2, 5, 5, 5, 5, 5, 5], dtype=pad_indexer.dtype
+        )
+        tm.assert_almost_equal(expected, pad_indexer)
+
+    def test_get_indexer_crossing_levels(self):
+        # https://github.com/pandas-dev/pandas/issues/29896
+        # tests a corner case with get_indexer() with MultiIndexes where, when we
+        # need to "carry" across levels, proper tuple ordering is respected
+        #
+        # the MultiIndexes used in this test, visually, are:
+        # mult_idx_1:
+        #  0: 1 1 1 1
+        #  1:       2
+        #  2:     2 1
+        #  3:       2
+        #  4: 1 2 1 1
+        #  5:       2
+        #  6:     2 1
+        #  7:       2
+        #  8: 2 1 1 1
+        #  9:       2
+        # 10:     2 1
+        # 11:       2
+        # 12: 2 2 1 1
+        # 13:       2
+        # 14:     2 1
+        # 15:       2
+        #
+        # mult_idx_2:
+        #  0: 1 3 2 2
+        #  1: 2 3 2 2
+        mult_idx_1 = MultiIndex.from_product([[1, 2]] * 4)
+        mult_idx_2 = MultiIndex.from_tuples([(1, 3, 2, 2), (2, 3, 2, 2)])
+
+        # show the tuple orderings, which get_indexer() should respect
+        assert mult_idx_1[7] < mult_idx_2[0] < mult_idx_1[8]
+        assert mult_idx_1[-1] < mult_idx_2[1]
+
+        indexer = mult_idx_1.get_indexer(mult_idx_2)
+        expected = np.array([-1, -1], dtype=indexer.dtype)
+        tm.assert_almost_equal(expected, indexer)
+
+        backfill_indexer = mult_idx_1.get_indexer(mult_idx_2, method="bfill")
+        expected = np.array([8, -1], dtype=backfill_indexer.dtype)
+        tm.assert_almost_equal(expected, backfill_indexer)
+
+        pad_indexer = mult_idx_1.get_indexer(mult_idx_2, method="ffill")
+        expected = np.array([7, 15], dtype=pad_indexer.dtype)
+        tm.assert_almost_equal(expected, pad_indexer)
+
+    def test_get_indexer_kwarg_validation(self):
+        # GH#41918
+        mi = MultiIndex.from_product([range(3), ["A", "B"]])
+
+        msg = "limit argument only valid if doing pad, backfill or nearest"
+        with pytest.raises(ValueError, match=msg):
+            mi.get_indexer(mi[:-1], limit=4)
+
+        msg = "tolerance argument only valid if doing pad, backfill or nearest"
+        with pytest.raises(ValueError, match=msg):
+            mi.get_indexer(mi[:-1], tolerance="piano")
+
+    def test_get_indexer_nan(self):
+        # GH#37222
+        idx1 = MultiIndex.from_product([["A"], [1.0, 2.0]], names=["id1", "id2"])
+        idx2 = MultiIndex.from_product([["A"], [np.nan, 2.0]], names=["id1", "id2"])
+        expected = np.array([-1, 1])
+        result = idx2.get_indexer(idx1)
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+        result = idx1.get_indexer(idx2)
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+
+def test_getitem(idx):
+    # scalar
+    assert idx[2] == ("bar", "one")
+
+    # slice
+    result = idx[2:5]
+    expected = idx[[2, 3, 4]]
+    assert result.equals(expected)
+
+    # boolean
+    result = idx[[True, False, True, False, True, True]]
+    result2 = idx[np.array([True, False, True, False, True, True])]
+    expected = idx[[0, 2, 4, 5]]
+    assert result.equals(expected)
+    assert result2.equals(expected)
+
+
+def test_getitem_group_select(idx):
+    sorted_idx, _ = idx.sortlevel(0)
+    assert sorted_idx.get_loc("baz") == slice(3, 4)
+    assert sorted_idx.get_loc("foo") == slice(0, 2)
+
+
+@pytest.mark.parametrize("box", [list, Index])
+def test_getitem_bool_index_all(box):
+    # GH#22533
+    ind1 = box([True] * 5)
+    idx = MultiIndex.from_tuples([(10, 1), (20, 2), (30, 3), (40, 4), (50, 5)])
+    tm.assert_index_equal(idx[ind1], idx)
+
+    ind2 = box([True, False, True, False, False])
+    expected = MultiIndex.from_tuples([(10, 1), (30, 3)])
+    tm.assert_index_equal(idx[ind2], expected)
+
+
+@pytest.mark.parametrize("box", [list, Index])
+def test_getitem_bool_index_single(box):
+    # GH#22533
+    ind1 = box([True])
+    idx = MultiIndex.from_tuples([(10, 1)])
+    tm.assert_index_equal(idx[ind1], idx)
+
+    ind2 = box([False])
+    expected = MultiIndex(
+        levels=[np.array([], dtype=np.int64), np.array([], dtype=np.int64)],
+        codes=[[], []],
+    )
+    tm.assert_index_equal(idx[ind2], expected)
+
+
+class TestGetLoc:
+    def test_get_loc(self, idx):
+        assert idx.get_loc(("foo", "two")) == 1
+        assert idx.get_loc(("baz", "two")) == 3
+        with pytest.raises(KeyError, match=r"^\('bar', 'two'\)$"):
+            idx.get_loc(("bar", "two"))
+        with pytest.raises(KeyError, match=r"^'quux'$"):
+            idx.get_loc("quux")
+
+        # 3 levels
+        index = MultiIndex(
+            levels=[Index(np.arange(4)), Index(np.arange(4)), Index(np.arange(4))],
+            codes=[
+                np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+                np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+                np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+            ],
+        )
+        with pytest.raises(KeyError, match=r"^\(1, 1\)$"):
+            index.get_loc((1, 1))
+        assert index.get_loc((2, 0)) == slice(3, 5)
+
+    def test_get_loc_duplicates(self):
+        index = Index([2, 2, 2, 2])
+        result = index.get_loc(2)
+        expected = slice(0, 4)
+        assert result == expected
+
+        index = Index(["c", "a", "a", "b", "b"])
+        rs = index.get_loc("c")
+        xp = 0
+        assert rs == xp
+
+        with pytest.raises(KeyError, match="2"):
+            index.get_loc(2)
+
+    def test_get_loc_level(self):
+        index = MultiIndex(
+            levels=[Index(np.arange(4)), Index(np.arange(4)), Index(np.arange(4))],
+            codes=[
+                np.array([0, 0, 1, 2, 2, 2, 3, 3]),
+                np.array([0, 1, 0, 0, 0, 1, 0, 1]),
+                np.array([1, 0, 1, 1, 0, 0, 1, 0]),
+            ],
+        )
+        loc, new_index = index.get_loc_level((0, 1))
+        expected = slice(1, 2)
+        exp_index = index[expected].droplevel(0).droplevel(0)
+        assert loc == expected
+        assert new_index.equals(exp_index)
+
+        loc, new_index = index.get_loc_level((0, 1, 0))
+        expected = 1
+        assert loc == expected
+        assert new_index is None
+
+        with pytest.raises(KeyError, match=r"^\(2, 2\)$"):
+            index.get_loc_level((2, 2))
+        # GH 22221: unused label
+        with pytest.raises(KeyError, match=r"^2$"):
+            index.drop(2).get_loc_level(2)
+        # Unused label on unsorted level:
+        with pytest.raises(KeyError, match=r"^2$"):
+            index.drop(1, level=2).get_loc_level(2, level=2)
+
+        index = MultiIndex(
+            levels=[[2000], list(range(4))],
+            codes=[np.array([0, 0, 0, 0]), np.array([0, 1, 2, 3])],
+        )
+        result, new_index = index.get_loc_level((2000, slice(None, None)))
+        expected = slice(None, None)
+        assert result == expected
+        assert new_index.equals(index.droplevel(0))
+
+    @pytest.mark.parametrize("dtype1", [int, float, bool, str])
+    @pytest.mark.parametrize("dtype2", [int, float, bool, str])
+    def test_get_loc_multiple_dtypes(self, dtype1, dtype2):
+        # GH 18520
+        levels = [np.array([0, 1]).astype(dtype1), np.array([0, 1]).astype(dtype2)]
+        idx = MultiIndex.from_product(levels)
+        assert idx.get_loc(idx[2]) == 2
+
+    @pytest.mark.parametrize("level", [0, 1])
+    @pytest.mark.parametrize("dtypes", [[int, float], [float, int]])
+    def test_get_loc_implicit_cast(self, level, dtypes):
+        # GH 18818, GH 15994 : as flat index, cast int to float and vice-versa
+        levels = [["a", "b"], ["c", "d"]]
+        key = ["b", "d"]
+        lev_dtype, key_dtype = dtypes
+        levels[level] = np.array([0, 1], dtype=lev_dtype)
+        key[level] = key_dtype(1)
+        idx = MultiIndex.from_product(levels)
+        assert idx.get_loc(tuple(key)) == 3
+
+    @pytest.mark.parametrize("dtype", [bool, object])
+    def test_get_loc_cast_bool(self, dtype):
+        # GH 19086 : int is casted to bool, but not vice-versa (for object dtype)
+        #  With bool dtype, we don't cast in either direction.
+        levels = [Index([False, True], dtype=dtype), np.arange(2, dtype="int64")]
+        idx = MultiIndex.from_product(levels)
+
+        if dtype is bool:
+            with pytest.raises(KeyError, match=r"^\(0, 1\)$"):
+                assert idx.get_loc((0, 1)) == 1
+            with pytest.raises(KeyError, match=r"^\(1, 0\)$"):
+                assert idx.get_loc((1, 0)) == 2
+        else:
+            # We use python object comparisons, which treat 0 == False and 1 == True
+            assert idx.get_loc((0, 1)) == 1
+            assert idx.get_loc((1, 0)) == 2
+
+        with pytest.raises(KeyError, match=r"^\(False, True\)$"):
+            idx.get_loc((False, True))
+        with pytest.raises(KeyError, match=r"^\(True, False\)$"):
+            idx.get_loc((True, False))
+
+    @pytest.mark.parametrize("level", [0, 1])
+    def test_get_loc_nan(self, level, nulls_fixture):
+        # GH 18485 : NaN in MultiIndex
+        levels = [["a", "b"], ["c", "d"]]
+        key = ["b", "d"]
+        levels[level] = np.array([0, nulls_fixture], dtype=type(nulls_fixture))
+        key[level] = nulls_fixture
+        idx = MultiIndex.from_product(levels)
+        assert idx.get_loc(tuple(key)) == 3
+
+    def test_get_loc_missing_nan(self):
+        # GH 8569
+        idx = MultiIndex.from_arrays([[1.0, 2.0], [3.0, 4.0]])
+        assert isinstance(idx.get_loc(1), slice)
+        with pytest.raises(KeyError, match=r"^3$"):
+            idx.get_loc(3)
+        with pytest.raises(KeyError, match=r"^nan$"):
+            idx.get_loc(np.nan)
+        with pytest.raises(InvalidIndexError, match=r"\[nan\]"):
+            # listlike/non-hashable raises TypeError
+            idx.get_loc([np.nan])
+
+    def test_get_loc_with_values_including_missing_values(self):
+        # issue 19132
+        idx = MultiIndex.from_product([[np.nan, 1]] * 2)
+        expected = slice(0, 2, None)
+        assert idx.get_loc(np.nan) == expected
+
+        idx = MultiIndex.from_arrays([[np.nan, 1, 2, np.nan]])
+        expected = np.array([True, False, False, True])
+        tm.assert_numpy_array_equal(idx.get_loc(np.nan), expected)
+
+        idx = MultiIndex.from_product([[np.nan, 1]] * 3)
+        expected = slice(2, 4, None)
+        assert idx.get_loc((np.nan, 1)) == expected
+
+    def test_get_loc_duplicates2(self):
+        # TODO: de-duplicate with test_get_loc_duplicates above?
+        index = MultiIndex(
+            levels=[["D", "B", "C"], [0, 26, 27, 37, 57, 67, 75, 82]],
+            codes=[[0, 0, 0, 1, 2, 2, 2, 2, 2, 2], [1, 3, 4, 6, 0, 2, 2, 3, 5, 7]],
+            names=["tag", "day"],
+        )
+
+        assert index.get_loc("D") == slice(0, 3)
+
+    def test_get_loc_past_lexsort_depth(self, performance_warning):
+        # GH#30053
+        idx = MultiIndex(
+            levels=[["a"], [0, 7], [1]],
+            codes=[[0, 0], [1, 0], [0, 0]],
+            names=["x", "y", "z"],
+            sortorder=0,
+        )
+        key = ("a", 7)
+
+        with tm.assert_produces_warning(performance_warning):
+            # PerformanceWarning: indexing past lexsort depth may impact performance
+            result = idx.get_loc(key)
+
+        assert result == slice(0, 1, None)
+
+    def test_multiindex_get_loc_list_raises(self):
+        # GH#35878
+        idx = MultiIndex.from_tuples([("a", 1), ("b", 2)])
+        msg = r"\[\]"
+        with pytest.raises(InvalidIndexError, match=msg):
+            idx.get_loc([])
+
+    def test_get_loc_nested_tuple_raises_keyerror(self):
+        # raise KeyError, not TypeError
+        mi = MultiIndex.from_product([range(3), range(4), range(5), range(6)])
+        key = ((2, 3, 4), "foo")
+
+        with pytest.raises(KeyError, match=re.escape(str(key))):
+            mi.get_loc(key)
+
+
+class TestWhere:
+    def test_where(self):
+        i = MultiIndex.from_tuples([("A", 1), ("A", 2)])
+
+        msg = r"\.where is not supported for MultiIndex operations"
+        with pytest.raises(NotImplementedError, match=msg):
+            i.where(True)
+
+    def test_where_array_like(self, listlike_box):
+        mi = MultiIndex.from_tuples([("A", 1), ("A", 2)])
+        cond = [False, True]
+        msg = r"\.where is not supported for MultiIndex operations"
+        with pytest.raises(NotImplementedError, match=msg):
+            mi.where(listlike_box(cond))
+
+
+class TestContains:
+    def test_contains_top_level(self):
+        midx = MultiIndex.from_product([["A", "B"], [1, 2]])
+        assert "A" in midx
+        assert "A" not in midx._engine
+
+    def test_contains_with_nat(self):
+        # MI with a NaT
+        mi = MultiIndex(
+            levels=[["C"], date_range("2012-01-01", periods=5)],
+            codes=[[0, 0, 0, 0, 0, 0], [-1, 0, 1, 2, 3, 4]],
+            names=[None, "B"],
+        )
+        assert ("C", pd.Timestamp("2012-01-01")) in mi
+        for val in mi.values:
+            assert val in mi
+
+    def test_contains(self, idx):
+        assert ("foo", "two") in idx
+        assert ("bar", "two") not in idx
+        assert None not in idx
+
+    def test_contains_with_missing_value(self):
+        # GH#19132
+        idx = MultiIndex.from_arrays([[1, np.nan, 2]])
+        assert np.nan in idx
+
+        idx = MultiIndex.from_arrays([[1, 2], [np.nan, 3]])
+        assert np.nan not in idx
+        assert (1, np.nan) in idx
+
+    def test_multiindex_contains_dropped(self):
+        # GH#19027
+        # test that dropped MultiIndex levels are not in the MultiIndex
+        # despite continuing to be in the MultiIndex's levels
+        idx = MultiIndex.from_product([[1, 2], [3, 4]])
+        assert 2 in idx
+        idx = idx.drop(2)
+
+        # drop implementation keeps 2 in the levels
+        assert 2 in idx.levels[0]
+        # but it should no longer be in the index itself
+        assert 2 not in idx
+
+        # also applies to strings
+        idx = MultiIndex.from_product([["a", "b"], ["c", "d"]])
+        assert "a" in idx
+        idx = idx.drop("a")
+        assert "a" in idx.levels[0]
+        assert "a" not in idx
+
+    def test_contains_td64_level(self):
+        # GH#24570
+        tx = pd.timedelta_range("09:30:00", "16:00:00", freq="30 min")
+        idx = MultiIndex.from_arrays([tx, np.arange(len(tx))])
+        assert tx[0] in idx
+        assert "element_not_exit" not in idx
+        assert "0 day 09:30:00" in idx
+
+    def test_large_mi_contains(self, monkeypatch):
+        # GH#10645
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", 10)
+            result = MultiIndex.from_arrays([range(10), range(10)])
+            assert (10, 0) not in result
+
+
+def test_timestamp_multiindex_indexer():
+    # https://github.com/pandas-dev/pandas/issues/26944
+    idx = MultiIndex.from_product(
+        [
+            date_range("2019-01-01T00:15:33", periods=100, freq="h", name="date"),
+            ["x"],
+            [3],
+        ]
+    )
+    df = DataFrame({"foo": np.arange(len(idx))}, idx)
+    result = df.loc[pd.IndexSlice["2019-1-2":, "x", :], "foo"]
+    qidx = MultiIndex.from_product(
+        [
+            date_range(
+                start="2019-01-02T00:15:33",
+                end="2019-01-05T03:15:33",
+                freq="h",
+                name="date",
+            ),
+            ["x"],
+            [3],
+        ]
+    )
+    should_be = pd.Series(data=np.arange(24, len(qidx) + 24), index=qidx, name="foo")
+    tm.assert_series_equal(result, should_be)
+
+
+@pytest.mark.parametrize(
+    "index_arr,expected,target,algo",
+    [
+        ([[np.nan, "a", "b"], ["c", "d", "e"]], 0, np.nan, "left"),
+        ([[np.nan, "a", "b"], ["c", "d", "e"]], 1, (np.nan, "c"), "right"),
+        ([["a", "b", "c"], ["d", np.nan, "d"]], 1, ("b", np.nan), "left"),
+    ],
+)
+def test_get_slice_bound_with_missing_value(index_arr, expected, target, algo):
+    # issue 19132
+    idx = MultiIndex.from_arrays(index_arr)
+    result = idx.get_slice_bound(target, side=algo)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "index_arr,expected,start_idx,end_idx",
+    [
+        ([[np.nan, 1, 2], [3, 4, 5]], slice(0, 2, None), np.nan, 1),
+        ([[np.nan, 1, 2], [3, 4, 5]], slice(0, 3, None), np.nan, (2, 5)),
+        ([[1, 2, 3], [4, np.nan, 5]], slice(1, 3, None), (2, np.nan), 3),
+        ([[1, 2, 3], [4, np.nan, 5]], slice(1, 3, None), (2, np.nan), (3, 5)),
+    ],
+)
+def test_slice_indexer_with_missing_value(index_arr, expected, start_idx, end_idx):
+    # issue 19132
+    idx = MultiIndex.from_arrays(index_arr)
+    result = idx.slice_indexer(start=start_idx, end=end_idx)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "N, expected_dtype",
+    [
+        (1, "uint8"),  # 2*4*N = 8
+        (2, "uint16"),  # 2*4*N = 16
+        (4, "uint32"),  # 2*4*N = 32
+        (8, "uint64"),  # 2*4*N = 64
+        (10, "object"),  # 2*4*N = 80
+    ],
+)
+def test_pyint_engine(N, expected_dtype):
+    # GH#18519 : when combinations of codes cannot be represented in 64
+    # bits, the index underlying the MultiIndex engine works with Python
+    # integers, rather than uint64.
+    keys = [
+        tuple(arr)
+        for arr in [
+            [0] * 4 * N,
+            [1] * 4 * N,
+            [np.nan] * N + [0] * 3 * N,
+            [0] * N + [1] * 3 * N,
+            [np.nan] * N + [1] * 2 * N + [0] * N,
+        ]
+    ]
+    # Each level contains 3 elements (NaN, 0, 1), and it's represented
+    # in 2 bits to store 4 possible values (0=notfound, 1=NaN, 2=0, 3=1), for
+    # a total of 2*N*4 = 80 > 64 bits where N=10 and the number of levels is N*4.
+    # If we were using a 64 bit engine and truncating the first levels, the
+    # fourth and fifth keys would collide; if truncating the last levels, the
+    # fifth and sixth; if rotating bits rather than shifting, the third and fifth.
+
+    index = MultiIndex.from_tuples(keys)
+    assert index._engine.values.dtype == expected_dtype
+
+    for idx, key_value in enumerate(keys):
+        assert index.get_loc(key_value) == idx
+
+        expected = np.arange(idx + 1, dtype=np.intp)
+        result = index.get_indexer([keys[i] for i in expected])
+        tm.assert_numpy_array_equal(result, expected)
+
+    # With missing key:
+    idces = range(len(keys))
+    expected = np.array([-1] + list(idces), dtype=np.intp)
+    missing = tuple([0, 1, 0, 1] * N)
+    result = index.get_indexer([missing] + [keys[i] for i in idces])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keys,expected",
+    [
+        ((slice(None), [5, 4]), [1, 0]),
+        ((slice(None), [4, 5]), [0, 1]),
+        (([True, False, True], [4, 6]), [0, 2]),
+        (([True, False, True], [6, 4]), [0, 2]),
+        ((2, [4, 5]), [0, 1]),
+        ((2, [5, 4]), [1, 0]),
+        (([2], [4, 5]), [0, 1]),
+        (([2], [5, 4]), [1, 0]),
+    ],
+)
+def test_get_locs_reordering(keys, expected):
+    # GH48384
+    idx = MultiIndex.from_arrays(
+        [
+            [2, 2, 1],
+            [4, 5, 6],
+        ]
+    )
+    result = idx.get_locs(keys)
+    expected = np.array(expected, dtype=np.intp)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_get_indexer_for_multiindex_with_nans(nulls_fixture):
+    # GH37222
+    idx1 = MultiIndex.from_product([["A"], [1.0, 2.0]], names=["id1", "id2"])
+    idx2 = MultiIndex.from_product([["A"], [nulls_fixture, 2.0]], names=["id1", "id2"])
+
+    result = idx2.get_indexer(idx1)
+    expected = np.array([-1, 1], dtype=np.intp)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = idx1.get_indexer(idx2)
+    expected = np.array([-1, 1], dtype=np.intp)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_get_loc_namedtuple_behaves_like_tuple():
+    # GH57922
+    NamedIndex = namedtuple("NamedIndex", ("a", "b"))
+    multi_idx = MultiIndex.from_tuples(
+        [NamedIndex("i1", "i2"), NamedIndex("i3", "i4"), NamedIndex("i5", "i6")]
+    )
+    for idx in (multi_idx, multi_idx.to_flat_index()):
+        assert idx.get_loc(NamedIndex("i1", "i2")) == 0
+        assert idx.get_loc(NamedIndex("i3", "i4")) == 1
+        assert idx.get_loc(NamedIndex("i5", "i6")) == 2
+        assert idx.get_loc(("i1", "i2")) == 0
+        assert idx.get_loc(("i3", "i4")) == 1
+        assert idx.get_loc(("i5", "i6")) == 2
+    multi_idx = MultiIndex.from_tuples([("i1", "i2"), ("i3", "i4"), ("i5", "i6")])
+    for idx in (multi_idx, multi_idx.to_flat_index()):
+        assert idx.get_loc(NamedIndex("i1", "i2")) == 0
+        assert idx.get_loc(NamedIndex("i3", "i4")) == 1
+        assert idx.get_loc(NamedIndex("i5", "i6")) == 2
+        assert idx.get_loc(("i1", "i2")) == 0
+        assert idx.get_loc(("i3", "i4")) == 1
+        assert idx.get_loc(("i5", "i6")) == 2
diff --git a/pandas/pandas/tests/indexes/multi/test_integrity.py b/pandas/pandas/tests/indexes/multi/test_integrity.py
new file mode 100644
index 0000000000000..d570e911bf584
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_integrity.py
@@ -0,0 +1,289 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+
+from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
+
+import pandas as pd
+from pandas import (
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+def test_labels_dtypes():
+    # GH 8456
+    i = MultiIndex.from_tuples([("A", 1), ("A", 2)])
+    assert i.codes[0].dtype == "int8"
+    assert i.codes[1].dtype == "int8"
+
+    i = MultiIndex.from_product([["a"], range(40)])
+    assert i.codes[1].dtype == "int8"
+    i = MultiIndex.from_product([["a"], range(400)])
+    assert i.codes[1].dtype == "int16"
+    i = MultiIndex.from_product([["a"], range(40000)])
+    assert i.codes[1].dtype == "int32"
+
+    i = MultiIndex.from_product([["a"], range(1000)])
+    assert (i.codes[0] >= 0).all()
+    assert (i.codes[1] >= 0).all()
+
+
+def test_values_boxed():
+    tuples = [
+        (1, pd.Timestamp("2000-01-01")),
+        (2, pd.NaT),
+        (3, pd.Timestamp("2000-01-03")),
+        (1, pd.Timestamp("2000-01-04")),
+        (2, pd.Timestamp("2000-01-02")),
+        (3, pd.Timestamp("2000-01-03")),
+    ]
+    result = MultiIndex.from_tuples(tuples)
+    expected = construct_1d_object_array_from_listlike(tuples)
+    tm.assert_numpy_array_equal(result.values, expected)
+    # Check that code branches for boxed values produce identical results
+    tm.assert_numpy_array_equal(result.values[:4], result[:4].values)
+
+
+def test_values_multiindex_datetimeindex():
+    # Test to ensure we hit the boxing / nobox part of MI.values
+    ints = np.arange(10**18, 10**18 + 5)
+    naive = pd.DatetimeIndex(ints)
+
+    aware = pd.DatetimeIndex(ints, tz="US/Central")
+
+    idx = MultiIndex.from_arrays([naive, aware])
+    result = idx.values
+
+    outer = pd.DatetimeIndex([x[0] for x in result])
+    tm.assert_index_equal(outer, naive)
+
+    inner = pd.DatetimeIndex([x[1] for x in result])
+    tm.assert_index_equal(inner, aware)
+
+    # n_lev > n_lab
+    result = idx[:2].values
+
+    outer = pd.DatetimeIndex([x[0] for x in result])
+    tm.assert_index_equal(outer, naive[:2])
+
+    inner = pd.DatetimeIndex([x[1] for x in result])
+    tm.assert_index_equal(inner, aware[:2])
+
+
+def test_values_multiindex_periodindex():
+    # Test to ensure we hit the boxing / nobox part of MI.values
+    ints = np.arange(2007, 2012)
+    pidx = pd.PeriodIndex(ints, freq="D")
+
+    idx = MultiIndex.from_arrays([ints, pidx])
+    result = idx.values
+
+    outer = Index([x[0] for x in result])
+    tm.assert_index_equal(outer, Index(ints, dtype=np.int64))
+
+    inner = pd.PeriodIndex([x[1] for x in result])
+    tm.assert_index_equal(inner, pidx)
+
+    # n_lev > n_lab
+    result = idx[:2].values
+
+    outer = Index([x[0] for x in result])
+    tm.assert_index_equal(outer, Index(ints[:2], dtype=np.int64))
+
+    inner = pd.PeriodIndex([x[1] for x in result])
+    tm.assert_index_equal(inner, pidx[:2])
+
+
+def test_consistency():
+    # need to construct an overflow
+    major_axis = list(range(70000))
+    minor_axis = list(range(10))
+
+    major_codes = np.arange(70000)
+    minor_codes = np.repeat(range(10), 7000)
+
+    # the fact that is works means it's consistent
+    index = MultiIndex(
+        levels=[major_axis, minor_axis], codes=[major_codes, minor_codes]
+    )
+
+    # inconsistent
+    major_codes = np.array([0, 0, 1, 1, 1, 2, 2, 3, 3])
+    minor_codes = np.array([0, 1, 0, 1, 1, 0, 1, 0, 1])
+    index = MultiIndex(
+        levels=[major_axis, minor_axis], codes=[major_codes, minor_codes]
+    )
+
+    assert index.is_unique is False
+
+
+@pytest.mark.slow
+def test_hash_collisions(monkeypatch):
+    # non-smoke test that we don't get hash collisions
+    size_cutoff = 50
+    with monkeypatch.context() as m:
+        m.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+        index = MultiIndex.from_product(
+            [np.arange(8), np.arange(8)], names=["one", "two"]
+        )
+        result = index.get_indexer(index.values)
+        tm.assert_numpy_array_equal(result, np.arange(len(index), dtype="intp"))
+
+        for i in [0, 1, len(index) - 2, len(index) - 1]:
+            result = index.get_loc(index[i])
+            assert result == i
+
+
+def test_dims():
+    pass
+
+
+def test_take_invalid_kwargs():
+    vals = [["A", "B"], [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-01-02")]]
+    idx = MultiIndex.from_product(vals, names=["str", "dt"])
+    indices = [1, 2]
+
+    msg = r"take\(\) got an unexpected keyword argument 'foo'"
+    with pytest.raises(TypeError, match=msg):
+        idx.take(indices, foo=2)
+
+    msg = "the 'out' parameter is not supported"
+    with pytest.raises(ValueError, match=msg):
+        idx.take(indices, out=indices)
+
+    msg = "the 'mode' parameter is not supported"
+    with pytest.raises(ValueError, match=msg):
+        idx.take(indices, mode="clip")
+
+
+def test_isna_behavior(idx):
+    # should not segfault GH5123
+    # NOTE: if MI representation changes, may make sense to allow
+    # isna(MI)
+    msg = "isna is not defined for MultiIndex"
+    with pytest.raises(NotImplementedError, match=msg):
+        pd.isna(idx)
+
+
+def test_large_multiindex_error(monkeypatch):
+    # GH12527
+    size_cutoff = 50
+    with monkeypatch.context() as m:
+        m.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+        df_below_cutoff = pd.DataFrame(
+            1,
+            index=MultiIndex.from_product([[1, 2], range(size_cutoff - 1)]),
+            columns=["dest"],
+        )
+        with pytest.raises(KeyError, match=r"^\(-1, 0\)$"):
+            df_below_cutoff.loc[(-1, 0), "dest"]
+        with pytest.raises(KeyError, match=r"^\(3, 0\)$"):
+            df_below_cutoff.loc[(3, 0), "dest"]
+        df_above_cutoff = pd.DataFrame(
+            1,
+            index=MultiIndex.from_product([[1, 2], range(size_cutoff + 1)]),
+            columns=["dest"],
+        )
+        with pytest.raises(KeyError, match=r"^\(-1, 0\)$"):
+            df_above_cutoff.loc[(-1, 0), "dest"]
+        with pytest.raises(KeyError, match=r"^\(3, 0\)$"):
+            df_above_cutoff.loc[(3, 0), "dest"]
+
+
+def test_mi_hashtable_populated_attribute_error(monkeypatch):
+    # GH 18165
+    monkeypatch.setattr(libindex, "_SIZE_CUTOFF", 50)
+    r = range(50)
+    df = pd.DataFrame({"a": r, "b": r}, index=MultiIndex.from_arrays([r, r]))
+
+    msg = "'Series' object has no attribute 'foo'"
+    with pytest.raises(AttributeError, match=msg):
+        df["a"].foo()
+
+
+def test_can_hold_identifiers(idx):
+    key = idx[0]
+    assert idx._can_hold_identifiers_and_holds_name(key) is True
+
+
+def test_metadata_immutable(idx):
+    levels, codes = idx.levels, idx.codes
+    # shouldn't be able to set at either the top level or base level
+    mutable_regex = re.compile("does not support mutable operations")
+    with pytest.raises(TypeError, match=mutable_regex):
+        levels[0] = levels[0]
+    with pytest.raises(TypeError, match=mutable_regex):
+        levels[0][0] = levels[0][0]
+    # ditto for labels
+    with pytest.raises(TypeError, match=mutable_regex):
+        codes[0] = codes[0]
+    with pytest.raises(ValueError, match="assignment destination is read-only"):
+        codes[0][0] = codes[0][0]
+    # and for names
+    names = idx.names
+    with pytest.raises(TypeError, match=mutable_regex):
+        names[0] = names[0]
+
+
+def test_level_setting_resets_attributes():
+    ind = MultiIndex.from_arrays([["A", "A", "B", "B", "B"], [1, 2, 1, 2, 3]])
+    assert ind.is_monotonic_increasing
+    ind = ind.set_levels([["A", "B"], [1, 3, 2]])
+    # if this fails, probably didn't reset the cache correctly.
+    assert not ind.is_monotonic_increasing
+
+
+def test_rangeindex_fallback_coercion_bug():
+    # GH 12893
+    df1 = pd.DataFrame(np.arange(100).reshape((10, 10)))
+    df2 = pd.DataFrame(np.arange(100).reshape((10, 10)))
+    df = pd.concat(
+        {"df1": df1.stack(), "df2": df2.stack()},
+        axis=1,
+    )
+    df.index.names = ["fizz", "buzz"]
+
+    expected = pd.DataFrame(
+        {"df2": np.arange(100), "df1": np.arange(100)},
+        index=MultiIndex.from_product([range(10), range(10)], names=["fizz", "buzz"]),
+    )
+    tm.assert_frame_equal(df, expected, check_like=True)
+
+    result = df.index.get_level_values("fizz")
+    expected = Index(np.arange(10, dtype=np.int64), name="fizz").repeat(10)
+    tm.assert_index_equal(result, expected)
+
+    result = df.index.get_level_values("buzz")
+    expected = Index(np.tile(np.arange(10, dtype=np.int64), 10), name="buzz")
+    tm.assert_index_equal(result, expected)
+
+
+def test_memory_usage(idx):
+    result = idx.memory_usage()
+    if len(idx):
+        idx.get_loc(idx[0])
+        result2 = idx.memory_usage()
+        result3 = idx.memory_usage(deep=True)
+
+        # RangeIndex, IntervalIndex
+        # don't have engines
+        if not isinstance(idx, (RangeIndex, IntervalIndex)):
+            assert result2 > result
+
+        if idx.inferred_type == "object":
+            assert result3 > result2
+
+    else:
+        # we report 0 for no-length
+        assert result == 0
+
+
+def test_nlevels(idx):
+    assert idx.nlevels == 2
diff --git a/pandas/pandas/tests/indexes/multi/test_isin.py b/pandas/pandas/tests/indexes/multi/test_isin.py
new file mode 100644
index 0000000000000..92ac2468d5993
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_isin.py
@@ -0,0 +1,104 @@
+import numpy as np
+import pytest
+
+from pandas import MultiIndex
+import pandas._testing as tm
+
+
+def test_isin_nan():
+    idx = MultiIndex.from_arrays([["foo", "bar"], [1.0, np.nan]])
+    tm.assert_numpy_array_equal(idx.isin([("bar", np.nan)]), np.array([False, True]))
+    tm.assert_numpy_array_equal(
+        idx.isin([("bar", float("nan"))]), np.array([False, True])
+    )
+
+
+def test_isin_missing(nulls_fixture):
+    # GH48905
+    mi1 = MultiIndex.from_tuples([(1, nulls_fixture)])
+    mi2 = MultiIndex.from_tuples([(1, 1), (1, 2)])
+    result = mi2.isin(mi1)
+    expected = np.array([False, False])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_isin():
+    values = [("foo", 2), ("bar", 3), ("quux", 4)]
+
+    idx = MultiIndex.from_arrays([["qux", "baz", "foo", "bar"], np.arange(4)])
+    result = idx.isin(values)
+    expected = np.array([False, False, True, True])
+    tm.assert_numpy_array_equal(result, expected)
+
+    # empty, return dtype bool
+    idx = MultiIndex.from_arrays([[], []])
+    result = idx.isin(values)
+    assert len(result) == 0
+    assert result.dtype == np.bool_
+
+
+def test_isin_level_kwarg():
+    idx = MultiIndex.from_arrays([["qux", "baz", "foo", "bar"], np.arange(4)])
+
+    vals_0 = ["foo", "bar", "quux"]
+    vals_1 = [2, 3, 10]
+
+    expected = np.array([False, False, True, True])
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_0, level=0))
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_0, level=-2))
+
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_1, level=1))
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_1, level=-1))
+
+    msg = "Too many levels: Index has only 2 levels, not 6"
+    with pytest.raises(IndexError, match=msg):
+        idx.isin(vals_0, level=5)
+    msg = "Too many levels: Index has only 2 levels, -5 is not a valid level number"
+    with pytest.raises(IndexError, match=msg):
+        idx.isin(vals_0, level=-5)
+
+    with pytest.raises(KeyError, match=r"'Level 1\.0 not found'"):
+        idx.isin(vals_0, level=1.0)
+    with pytest.raises(KeyError, match=r"'Level -1\.0 not found'"):
+        idx.isin(vals_1, level=-1.0)
+    with pytest.raises(KeyError, match="'Level A not found'"):
+        idx.isin(vals_1, level="A")
+
+    idx.names = ["A", "B"]
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_0, level="A"))
+    tm.assert_numpy_array_equal(expected, idx.isin(vals_1, level="B"))
+
+    with pytest.raises(KeyError, match="'Level C not found'"):
+        idx.isin(vals_1, level="C")
+
+
+@pytest.mark.parametrize(
+    "labels,expected,level",
+    [
+        ([("b", np.nan)], [False, False, True], None),
+        ([np.nan, "a"], [True, True, False], 0),
+        (["d", np.nan], [False, True, True], 1),
+    ],
+)
+def test_isin_multi_index_with_missing_value(labels, expected, level):
+    # GH 19132
+    midx = MultiIndex.from_arrays([[np.nan, "a", "b"], ["c", "d", np.nan]])
+    result = midx.isin(labels, level=level)
+    expected = np.array(expected)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_isin_empty():
+    # GH#51599
+    midx = MultiIndex.from_arrays([[1, 2], [3, 4]])
+    result = midx.isin([])
+    expected = np.array([False, False])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_isin_generator():
+    # GH#52568
+    midx = MultiIndex.from_tuples([(1, 2)])
+    result = midx.isin(x for x in [(1, 2)])
+    expected = np.array([True])
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_join.py b/pandas/pandas/tests/indexes/multi/test_join.py
new file mode 100644
index 0000000000000..2be6bba475af7
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_join.py
@@ -0,0 +1,270 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Interval,
+    MultiIndex,
+    Series,
+    StringDtype,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("other", [["three", "one", "two"], ["one"], ["one", "three"]])
+def test_join_level(idx, other, join_type):
+    other = Index(other)
+    join_index, lidx, ridx = other.join(
+        idx, how=join_type, level="second", return_indexers=True
+    )
+
+    exp_level = other.join(idx.levels[1], how=join_type)
+    assert join_index.levels[0].equals(idx.levels[0])
+    assert join_index.levels[1].equals(exp_level)
+
+    # pare down levels
+    mask = np.array([x[1] in exp_level for x in idx], dtype=bool)
+    exp_values = idx.values[mask]
+    tm.assert_numpy_array_equal(join_index.values, exp_values)
+
+    if join_type in ("outer", "inner"):
+        join_index2, ridx2, lidx2 = idx.join(
+            other, how=join_type, level="second", return_indexers=True
+        )
+
+        assert join_index.equals(join_index2)
+        tm.assert_numpy_array_equal(lidx, lidx2)
+        if ridx is None:
+            assert ridx == ridx2
+        else:
+            tm.assert_numpy_array_equal(ridx, ridx2)
+        tm.assert_numpy_array_equal(join_index2.values, exp_values)
+
+
+def test_join_level_corner_case(idx):
+    # some corner cases
+    index = Index(["three", "one", "two"])
+    result = index.join(idx, level="second")
+    assert isinstance(result, MultiIndex)
+
+    with pytest.raises(TypeError, match="Join.*MultiIndex.*ambiguous"):
+        idx.join(idx, level=1)
+
+
+def test_join_self(idx, join_type):
+    result = idx.join(idx, how=join_type)
+    expected = idx
+    if join_type == "outer":
+        expected = expected.sort_values()
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_multi():
+    # GH 10665
+    midx = MultiIndex.from_product([np.arange(4), np.arange(4)], names=["a", "b"])
+    idx = Index([1, 2, 5], name="b")
+
+    # inner
+    jidx, lidx, ridx = midx.join(idx, how="inner", return_indexers=True)
+    exp_idx = MultiIndex.from_product([np.arange(4), [1, 2]], names=["a", "b"])
+    exp_lidx = np.array([1, 2, 5, 6, 9, 10, 13, 14], dtype=np.intp)
+    exp_ridx = np.array([0, 1, 0, 1, 0, 1, 0, 1], dtype=np.intp)
+    tm.assert_index_equal(jidx, exp_idx)
+    tm.assert_numpy_array_equal(lidx, exp_lidx)
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+    # flip
+    jidx, ridx, lidx = idx.join(midx, how="inner", return_indexers=True)
+    tm.assert_index_equal(jidx, exp_idx)
+    tm.assert_numpy_array_equal(lidx, exp_lidx)
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+
+    # keep MultiIndex
+    jidx, lidx, ridx = midx.join(idx, how="left", return_indexers=True)
+    exp_ridx = np.array(
+        [-1, 0, 1, -1, -1, 0, 1, -1, -1, 0, 1, -1, -1, 0, 1, -1], dtype=np.intp
+    )
+    tm.assert_index_equal(jidx, midx)
+    assert lidx is None
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+    # flip
+    jidx, ridx, lidx = idx.join(midx, how="right", return_indexers=True)
+    tm.assert_index_equal(jidx, midx)
+    assert lidx is None
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+
+
+def test_join_multi_wrong_order():
+    # GH 25760
+    # GH 28956
+
+    midx1 = MultiIndex.from_product([[1, 2], [3, 4]], names=["a", "b"])
+    midx2 = MultiIndex.from_product([[1, 2], [3, 4]], names=["b", "a"])
+
+    join_idx, lidx, ridx = midx1.join(midx2, return_indexers=True)
+
+    exp_ridx = np.array([-1, -1, -1, -1], dtype=np.intp)
+
+    tm.assert_index_equal(midx1, join_idx)
+    assert lidx is None
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+
+
+def test_join_multi_return_indexers():
+    # GH 34074
+
+    midx1 = MultiIndex.from_product([[1, 2], [3, 4], [5, 6]], names=["a", "b", "c"])
+    midx2 = MultiIndex.from_product([[1, 2], [3, 4]], names=["a", "b"])
+
+    result = midx1.join(midx2, return_indexers=False)
+    tm.assert_index_equal(result, midx1)
+
+
+def test_join_overlapping_interval_level():
+    # GH 44096
+    idx_1 = MultiIndex.from_tuples(
+        [
+            (1, Interval(0.0, 1.0)),
+            (1, Interval(1.0, 2.0)),
+            (1, Interval(2.0, 5.0)),
+            (2, Interval(0.0, 1.0)),
+            (2, Interval(1.0, 3.0)),  # interval limit is here at 3.0, not at 2.0
+            (2, Interval(3.0, 5.0)),
+        ],
+        names=["num", "interval"],
+    )
+
+    idx_2 = MultiIndex.from_tuples(
+        [
+            (1, Interval(2.0, 5.0)),
+            (1, Interval(0.0, 1.0)),
+            (1, Interval(1.0, 2.0)),
+            (2, Interval(3.0, 5.0)),
+            (2, Interval(0.0, 1.0)),
+            (2, Interval(1.0, 3.0)),
+        ],
+        names=["num", "interval"],
+    )
+
+    expected = MultiIndex.from_tuples(
+        [
+            (1, Interval(0.0, 1.0)),
+            (1, Interval(1.0, 2.0)),
+            (1, Interval(2.0, 5.0)),
+            (2, Interval(0.0, 1.0)),
+            (2, Interval(1.0, 3.0)),
+            (2, Interval(3.0, 5.0)),
+        ],
+        names=["num", "interval"],
+    )
+    result = idx_1.join(idx_2, how="outer")
+
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_midx_ea():
+    # GH#49277
+    midx = MultiIndex.from_arrays(
+        [Series([1, 1, 3], dtype="Int64"), Series([1, 2, 3], dtype="Int64")],
+        names=["a", "b"],
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series([1], dtype="Int64"), Series([3], dtype="Int64")], names=["a", "c"]
+    )
+    result = midx.join(midx2, how="inner")
+    expected = MultiIndex.from_arrays(
+        [
+            Series([1, 1], dtype="Int64"),
+            Series([1, 2], dtype="Int64"),
+            Series([3, 3], dtype="Int64"),
+        ],
+        names=["a", "b", "c"],
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_midx_string():
+    # GH#49277
+    midx = MultiIndex.from_arrays(
+        [
+            Series(["a", "a", "c"], dtype=StringDtype()),
+            Series(["a", "b", "c"], dtype=StringDtype()),
+        ],
+        names=["a", "b"],
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series(["a"], dtype=StringDtype()), Series(["c"], dtype=StringDtype())],
+        names=["a", "c"],
+    )
+    result = midx.join(midx2, how="inner")
+    expected = MultiIndex.from_arrays(
+        [
+            Series(["a", "a"], dtype=StringDtype()),
+            Series(["a", "b"], dtype=StringDtype()),
+            Series(["c", "c"], dtype=StringDtype()),
+        ],
+        names=["a", "b", "c"],
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_multi_with_nan():
+    # GH29252
+    df1 = DataFrame(
+        data={"col1": [1.1, 1.2]},
+        index=MultiIndex.from_product([["A"], [1.0, 2.0]], names=["id1", "id2"]),
+    )
+    df2 = DataFrame(
+        data={"col2": [2.1, 2.2]},
+        index=MultiIndex.from_product([["A"], [np.nan, 2.0]], names=["id1", "id2"]),
+    )
+    result = df1.join(df2)
+    expected = DataFrame(
+        data={"col1": [1.1, 1.2], "col2": [np.nan, 2.2]},
+        index=MultiIndex.from_product([["A"], [1.0, 2.0]], names=["id1", "id2"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [0, 5])
+def test_join_dtypes(any_numeric_ea_dtype, val):
+    # GH#49830
+    midx = MultiIndex.from_arrays([Series([1, 2], dtype=any_numeric_ea_dtype), [3, 4]])
+    midx2 = MultiIndex.from_arrays(
+        [Series([1, val, val], dtype=any_numeric_ea_dtype), [3, 4, 4]]
+    )
+    result = midx.join(midx2, how="outer")
+    expected = MultiIndex.from_arrays(
+        [Series([val, val, 1, 2], dtype=any_numeric_ea_dtype), [4, 4, 3, 4]]
+    ).sort_values()
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_dtypes_all_nan(any_numeric_ea_dtype):
+    # GH#49830
+    midx = MultiIndex.from_arrays(
+        [Series([1, 2], dtype=any_numeric_ea_dtype), [np.nan, np.nan]]
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series([1, 0, 0], dtype=any_numeric_ea_dtype), [np.nan, np.nan, np.nan]]
+    )
+    result = midx.join(midx2, how="outer")
+    expected = MultiIndex.from_arrays(
+        [
+            Series([0, 0, 1, 2], dtype=any_numeric_ea_dtype),
+            [np.nan, np.nan, np.nan, np.nan],
+        ]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_join_index_levels():
+    # GH#53093
+    midx = MultiIndex.from_tuples([("a", "2019-02-01"), ("a", "2019-02-01")])
+    midx2 = MultiIndex.from_tuples([("a", "2019-01-31")])
+    result = midx.join(midx2, how="outer")
+    expected = MultiIndex.from_tuples(
+        [("a", "2019-01-31"), ("a", "2019-02-01"), ("a", "2019-02-01")]
+    )
+    tm.assert_index_equal(result.levels[1], expected.levels[1])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_lexsort.py b/pandas/pandas/tests/indexes/multi/test_lexsort.py
new file mode 100644
index 0000000000000..fc16a4197a3a4
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_lexsort.py
@@ -0,0 +1,46 @@
+from pandas import MultiIndex
+
+
+class TestIsLexsorted:
+    def test_is_lexsorted(self):
+        levels = [[0, 1], [0, 1, 2]]
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]]
+        )
+        assert index._is_lexsorted()
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 2, 1]]
+        )
+        assert not index._is_lexsorted()
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 1, 0, 1, 1], [0, 1, 0, 2, 2, 1]]
+        )
+        assert not index._is_lexsorted()
+        assert index._lexsort_depth == 0
+
+
+class TestLexsortDepth:
+    def test_lexsort_depth(self):
+        # Test that lexsort_depth return the correct sortorder
+        # when it was given to the MultiIndex const.
+        # GH#28518
+
+        levels = [[0, 1], [0, 1, 2]]
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]], sortorder=2
+        )
+        assert index._lexsort_depth == 2
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 2, 1]], sortorder=1
+        )
+        assert index._lexsort_depth == 1
+
+        index = MultiIndex(
+            levels=levels, codes=[[0, 0, 1, 0, 1, 1], [0, 1, 0, 2, 2, 1]], sortorder=0
+        )
+        assert index._lexsort_depth == 0
diff --git a/pandas/pandas/tests/indexes/multi/test_missing.py b/pandas/pandas/tests/indexes/multi/test_missing.py
new file mode 100644
index 0000000000000..41cfa093ae53c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_missing.py
@@ -0,0 +1,111 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import MultiIndex
+import pandas._testing as tm
+
+
+def test_fillna(idx):
+    # GH 11343
+    msg = "fillna is not defined for MultiIndex"
+    with pytest.raises(NotImplementedError, match=msg):
+        idx.fillna(idx[0])
+
+
+def test_dropna():
+    # GH 6194
+    idx = MultiIndex.from_arrays(
+        [
+            [1, np.nan, 3, np.nan, 5],
+            [1, 2, np.nan, np.nan, 5],
+            ["a", "b", "c", np.nan, "e"],
+        ]
+    )
+
+    exp = MultiIndex.from_arrays([[1, 5], [1, 5], ["a", "e"]])
+    tm.assert_index_equal(idx.dropna(), exp)
+    tm.assert_index_equal(idx.dropna(how="any"), exp)
+
+    exp = MultiIndex.from_arrays(
+        [[1, np.nan, 3, 5], [1, 2, np.nan, 5], ["a", "b", "c", "e"]]
+    )
+    tm.assert_index_equal(idx.dropna(how="all"), exp)
+
+    msg = "invalid how option: xxx"
+    with pytest.raises(ValueError, match=msg):
+        idx.dropna(how="xxx")
+
+    # GH26408
+    # test if missing values are dropped for multiindex constructed
+    # from codes and values
+    idx = MultiIndex(
+        levels=[[np.nan, None, pd.NaT, "128", 2], [np.nan, None, pd.NaT, "128", 2]],
+        codes=[[0, -1, 1, 2, 3, 4], [0, -1, 3, 3, 3, 4]],
+    )
+    expected = MultiIndex.from_arrays([["128", 2], ["128", 2]])
+    tm.assert_index_equal(idx.dropna(), expected)
+    tm.assert_index_equal(idx.dropna(how="any"), expected)
+
+    expected = MultiIndex.from_arrays(
+        [[np.nan, np.nan, "128", 2], ["128", "128", "128", 2]]
+    )
+    tm.assert_index_equal(idx.dropna(how="all"), expected)
+
+
+def test_nulls(idx):
+    # this is really a smoke test for the methods
+    # as these are adequately tested for function elsewhere
+
+    msg = "isna is not defined for MultiIndex"
+    with pytest.raises(NotImplementedError, match=msg):
+        idx.isna()
+
+
+@pytest.mark.xfail(reason="isna is not defined for MultiIndex")
+def test_hasnans_isnans(idx):
+    # GH 11343, added tests for hasnans / isnans
+    index = idx.copy()
+
+    # cases in indices doesn't include NaN
+    expected = np.array([False] * len(index), dtype=bool)
+    tm.assert_numpy_array_equal(index._isnan, expected)
+    assert index.hasnans is False
+
+    index = idx.copy()
+    values = index.values
+    values[1] = np.nan
+
+    index = type(idx)(values)
+
+    expected = np.array([False] * len(index), dtype=bool)
+    expected[1] = True
+    tm.assert_numpy_array_equal(index._isnan, expected)
+    assert index.hasnans is True
+
+
+def test_nan_stays_float():
+    # GH 7031
+    idx0 = MultiIndex(levels=[["A", "B"], []], codes=[[1, 0], [-1, -1]], names=[0, 1])
+    idx1 = MultiIndex(levels=[["C"], ["D"]], codes=[[0], [0]], names=[0, 1])
+    idxm = idx0.join(idx1, how="outer")
+    assert pd.isna(idx0.get_level_values(1)).all()
+    # the following failed in 0.14.1
+    assert pd.isna(idxm.get_level_values(1)[:-1]).all()
+
+    df0 = pd.DataFrame([[1, 2]], index=idx0)
+    df1 = pd.DataFrame([[3, 4]], index=idx1)
+    dfm = df0 - df1
+    assert pd.isna(df0.index.get_level_values(1)).all()
+    # the following failed in 0.14.1
+    assert pd.isna(dfm.index.get_level_values(1)[:-1]).all()
+
+
+def test_tuples_have_na():
+    index = MultiIndex(
+        levels=[[1, 0], [0, 1, 2, 3]],
+        codes=[[1, 1, 1, 1, -1, 0, 0, 0], [0, 1, 2, 3, 0, 1, 2, 3]],
+    )
+
+    assert pd.isna(index[4][0])
+    assert pd.isna(index.values[4][0])
diff --git a/pandas/pandas/tests/indexes/multi/test_monotonic.py b/pandas/pandas/tests/indexes/multi/test_monotonic.py
new file mode 100644
index 0000000000000..2b0b3f7cb36d7
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_monotonic.py
@@ -0,0 +1,188 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    MultiIndex,
+)
+
+
+def test_is_monotonic_increasing_lexsorted(lexsorted_two_level_string_multiindex):
+    # string ordering
+    mi = lexsorted_two_level_string_multiindex
+    assert mi.is_monotonic_increasing is False
+    assert Index(mi.values).is_monotonic_increasing is False
+    assert mi._is_strictly_monotonic_increasing is False
+    assert Index(mi.values)._is_strictly_monotonic_increasing is False
+
+
+def test_is_monotonic_increasing():
+    i = MultiIndex.from_product([np.arange(10), np.arange(10)], names=["one", "two"])
+    assert i.is_monotonic_increasing is True
+    assert i._is_strictly_monotonic_increasing is True
+    assert Index(i.values).is_monotonic_increasing is True
+    assert i._is_strictly_monotonic_increasing is True
+
+    i = MultiIndex.from_product(
+        [np.arange(10, 0, -1), np.arange(10)], names=["one", "two"]
+    )
+    assert i.is_monotonic_increasing is False
+    assert i._is_strictly_monotonic_increasing is False
+    assert Index(i.values).is_monotonic_increasing is False
+    assert Index(i.values)._is_strictly_monotonic_increasing is False
+
+    i = MultiIndex.from_product(
+        [np.arange(10), np.arange(10, 0, -1)], names=["one", "two"]
+    )
+    assert i.is_monotonic_increasing is False
+    assert i._is_strictly_monotonic_increasing is False
+    assert Index(i.values).is_monotonic_increasing is False
+    assert Index(i.values)._is_strictly_monotonic_increasing is False
+
+    i = MultiIndex.from_product([[1.0, np.nan, 2.0], ["a", "b", "c"]])
+    assert i.is_monotonic_increasing is False
+    assert i._is_strictly_monotonic_increasing is False
+    assert Index(i.values).is_monotonic_increasing is False
+    assert Index(i.values)._is_strictly_monotonic_increasing is False
+
+    i = MultiIndex(
+        levels=[["bar", "baz", "foo", "qux"], ["mom", "next", "zenith"]],
+        codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+        names=["first", "second"],
+    )
+    assert i.is_monotonic_increasing is True
+    assert Index(i.values).is_monotonic_increasing is True
+    assert i._is_strictly_monotonic_increasing is True
+    assert Index(i.values)._is_strictly_monotonic_increasing is True
+
+    # mixed levels, hits the TypeError
+    i = MultiIndex(
+        levels=[
+            [1, 2, 3, 4],
+            [
+                "gb00b03mlx29",
+                "lu0197800237",
+                "nl0000289783",
+                "nl0000289965",
+                "nl0000301109",
+            ],
+        ],
+        codes=[[0, 1, 1, 2, 2, 2, 3], [4, 2, 0, 0, 1, 3, -1]],
+        names=["household_id", "asset_id"],
+    )
+
+    assert i.is_monotonic_increasing is False
+    assert i._is_strictly_monotonic_increasing is False
+
+    # empty
+    i = MultiIndex.from_arrays([[], []])
+    assert i.is_monotonic_increasing is True
+    assert Index(i.values).is_monotonic_increasing is True
+    assert i._is_strictly_monotonic_increasing is True
+    assert Index(i.values)._is_strictly_monotonic_increasing is True
+
+
+def test_is_monotonic_decreasing():
+    i = MultiIndex.from_product(
+        [np.arange(9, -1, -1), np.arange(9, -1, -1)], names=["one", "two"]
+    )
+    assert i.is_monotonic_decreasing is True
+    assert i._is_strictly_monotonic_decreasing is True
+    assert Index(i.values).is_monotonic_decreasing is True
+    assert i._is_strictly_monotonic_decreasing is True
+
+    i = MultiIndex.from_product(
+        [np.arange(10), np.arange(10, 0, -1)], names=["one", "two"]
+    )
+    assert i.is_monotonic_decreasing is False
+    assert i._is_strictly_monotonic_decreasing is False
+    assert Index(i.values).is_monotonic_decreasing is False
+    assert Index(i.values)._is_strictly_monotonic_decreasing is False
+
+    i = MultiIndex.from_product(
+        [np.arange(10, 0, -1), np.arange(10)], names=["one", "two"]
+    )
+    assert i.is_monotonic_decreasing is False
+    assert i._is_strictly_monotonic_decreasing is False
+    assert Index(i.values).is_monotonic_decreasing is False
+    assert Index(i.values)._is_strictly_monotonic_decreasing is False
+
+    i = MultiIndex.from_product([[2.0, np.nan, 1.0], ["c", "b", "a"]])
+    assert i.is_monotonic_decreasing is False
+    assert i._is_strictly_monotonic_decreasing is False
+    assert Index(i.values).is_monotonic_decreasing is False
+    assert Index(i.values)._is_strictly_monotonic_decreasing is False
+
+    # string ordering
+    i = MultiIndex(
+        levels=[["qux", "foo", "baz", "bar"], ["three", "two", "one"]],
+        codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+        names=["first", "second"],
+    )
+    assert i.is_monotonic_decreasing is False
+    assert Index(i.values).is_monotonic_decreasing is False
+    assert i._is_strictly_monotonic_decreasing is False
+    assert Index(i.values)._is_strictly_monotonic_decreasing is False
+
+    i = MultiIndex(
+        levels=[["qux", "foo", "baz", "bar"], ["zenith", "next", "mom"]],
+        codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+        names=["first", "second"],
+    )
+    assert i.is_monotonic_decreasing is True
+    assert Index(i.values).is_monotonic_decreasing is True
+    assert i._is_strictly_monotonic_decreasing is True
+    assert Index(i.values)._is_strictly_monotonic_decreasing is True
+
+    # mixed levels, hits the TypeError
+    i = MultiIndex(
+        levels=[
+            [4, 3, 2, 1],
+            [
+                "nl0000301109",
+                "nl0000289965",
+                "nl0000289783",
+                "lu0197800237",
+                "gb00b03mlx29",
+            ],
+        ],
+        codes=[[0, 1, 1, 2, 2, 2, 3], [4, 2, 0, 0, 1, 3, -1]],
+        names=["household_id", "asset_id"],
+    )
+
+    assert i.is_monotonic_decreasing is False
+    assert i._is_strictly_monotonic_decreasing is False
+
+    # empty
+    i = MultiIndex.from_arrays([[], []])
+    assert i.is_monotonic_decreasing is True
+    assert Index(i.values).is_monotonic_decreasing is True
+    assert i._is_strictly_monotonic_decreasing is True
+    assert Index(i.values)._is_strictly_monotonic_decreasing is True
+
+
+def test_is_strictly_monotonic_increasing():
+    idx = MultiIndex(
+        levels=[["bar", "baz"], ["mom", "next"]], codes=[[0, 0, 1, 1], [0, 0, 0, 1]]
+    )
+    assert idx.is_monotonic_increasing is True
+    assert idx._is_strictly_monotonic_increasing is False
+
+
+def test_is_strictly_monotonic_decreasing():
+    idx = MultiIndex(
+        levels=[["baz", "bar"], ["next", "mom"]], codes=[[0, 0, 1, 1], [0, 0, 0, 1]]
+    )
+    assert idx.is_monotonic_decreasing is True
+    assert idx._is_strictly_monotonic_decreasing is False
+
+
+@pytest.mark.parametrize("attr", ["is_monotonic_increasing", "is_monotonic_decreasing"])
+@pytest.mark.parametrize(
+    "values",
+    [[(np.nan,), (1,), (2,)], [(1,), (np.nan,), (2,)], [(1,), (2,), (np.nan,)]],
+)
+def test_is_monotonic_with_nans(values, attr):
+    # GH: 37220
+    idx = MultiIndex.from_tuples(values, names=["test"])
+    assert getattr(idx, attr) is False
diff --git a/pandas/pandas/tests/indexes/multi/test_names.py b/pandas/pandas/tests/indexes/multi/test_names.py
new file mode 100644
index 0000000000000..45f19b4d70fb9
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_names.py
@@ -0,0 +1,201 @@
+import pytest
+
+import pandas as pd
+from pandas import MultiIndex
+import pandas._testing as tm
+
+
+def check_level_names(index, names):
+    assert [level.name for level in index.levels] == list(names)
+
+
+def test_slice_keep_name():
+    x = MultiIndex.from_tuples([("a", "b"), (1, 2), ("c", "d")], names=["x", "y"])
+    assert x[1:].names == x.names
+
+
+def test_index_name_retained():
+    # GH9857
+    result = pd.DataFrame({"x": [1, 2, 6], "y": [2, 2, 8], "z": [-5, 0, 5]})
+    result = result.set_index("z")
+    result.loc[10] = [9, 10]
+    df_expected = pd.DataFrame(
+        {"x": [1, 2, 6, 9], "y": [2, 2, 8, 10], "z": [-5, 0, 5, 10]}
+    )
+    df_expected = df_expected.set_index("z")
+    tm.assert_frame_equal(result, df_expected)
+
+
+def test_changing_names(idx):
+    assert [level.name for level in idx.levels] == ["first", "second"]
+
+    view = idx.view()
+    copy = idx.copy()
+    shallow_copy = idx._view()
+
+    # changing names should not change level names on object
+    new_names = [name + "a" for name in idx.names]
+    idx.names = new_names
+    check_level_names(idx, ["firsta", "seconda"])
+
+    # and not on copies
+    check_level_names(view, ["first", "second"])
+    check_level_names(copy, ["first", "second"])
+    check_level_names(shallow_copy, ["first", "second"])
+
+    # and copies shouldn't change original
+    shallow_copy.names = [name + "c" for name in shallow_copy.names]
+    check_level_names(idx, ["firsta", "seconda"])
+
+
+def test_take_preserve_name(idx):
+    taken = idx.take([3, 0, 1])
+    assert taken.names == idx.names
+
+
+def test_copy_names():
+    # Check that adding a "names" parameter to the copy is honored
+    # GH14302
+    multi_idx = MultiIndex.from_tuples([(1, 2), (3, 4)], names=["MyName1", "MyName2"])
+    multi_idx1 = multi_idx.copy()
+
+    assert multi_idx.equals(multi_idx1)
+    assert multi_idx.names == ["MyName1", "MyName2"]
+    assert multi_idx1.names == ["MyName1", "MyName2"]
+
+    multi_idx2 = multi_idx.copy(names=["NewName1", "NewName2"])
+
+    assert multi_idx.equals(multi_idx2)
+    assert multi_idx.names == ["MyName1", "MyName2"]
+    assert multi_idx2.names == ["NewName1", "NewName2"]
+
+    multi_idx3 = multi_idx.copy(name=["NewName1", "NewName2"])
+
+    assert multi_idx.equals(multi_idx3)
+    assert multi_idx.names == ["MyName1", "MyName2"]
+    assert multi_idx3.names == ["NewName1", "NewName2"]
+
+    # gh-35592
+    with pytest.raises(ValueError, match="Length of new names must be 2, got 1"):
+        multi_idx.copy(names=["mario"])
+
+    with pytest.raises(TypeError, match="MultiIndex.name must be a hashable type"):
+        multi_idx.copy(names=[["mario"], ["luigi"]])
+
+
+def test_names(idx):
+    # names are assigned in setup
+    assert idx.names == ["first", "second"]
+    level_names = [level.name for level in idx.levels]
+    assert level_names == idx.names
+
+    # setting bad names on existing
+    index = idx
+    with pytest.raises(ValueError, match="^Length of names"):
+        setattr(index, "names", list(index.names) + ["third"])
+    with pytest.raises(ValueError, match="^Length of names"):
+        setattr(index, "names", [])
+
+    # initializing with bad names (should always be equivalent)
+    major_axis, minor_axis = idx.levels
+    major_codes, minor_codes = idx.codes
+    with pytest.raises(ValueError, match="^Length of names"):
+        MultiIndex(
+            levels=[major_axis, minor_axis],
+            codes=[major_codes, minor_codes],
+            names=["first"],
+        )
+    with pytest.raises(ValueError, match="^Length of names"):
+        MultiIndex(
+            levels=[major_axis, minor_axis],
+            codes=[major_codes, minor_codes],
+            names=["first", "second", "third"],
+        )
+
+    # names are assigned on index, but not transferred to the levels
+    index.names = ["a", "b"]
+    level_names = [level.name for level in index.levels]
+    assert level_names == ["a", "b"]
+
+
+def test_duplicate_level_names_access_raises(idx):
+    # GH19029
+    idx.names = ["foo", "foo"]
+    with pytest.raises(ValueError, match="name foo occurs multiple times"):
+        idx._get_level_number("foo")
+
+
+def test_get_names_from_levels():
+    idx = MultiIndex.from_product([["a"], [1, 2]], names=["a", "b"])
+
+    assert idx.levels[0].name == "a"
+    assert idx.levels[1].name == "b"
+
+
+def test_setting_names_from_levels_raises():
+    idx = MultiIndex.from_product([["a"], [1, 2]], names=["a", "b"])
+    with pytest.raises(RuntimeError, match="set_names"):
+        idx.levels[0].name = "foo"
+
+    with pytest.raises(RuntimeError, match="set_names"):
+        idx.levels[1].name = "foo"
+
+    new = pd.Series(1, index=idx.levels[0])
+    with pytest.raises(RuntimeError, match="set_names"):
+        new.index.name = "bar"
+
+    assert pd.Index._no_setting_name is False
+    assert pd.RangeIndex._no_setting_name is False
+
+
+@pytest.mark.parametrize("func", ["rename", "set_names"])
+@pytest.mark.parametrize(
+    "rename_dict, exp_names",
+    [
+        ({"x": "z"}, ["z", "y", "z"]),
+        ({"x": "z", "y": "x"}, ["z", "x", "z"]),
+        ({"y": "z"}, ["x", "z", "x"]),
+        ({}, ["x", "y", "x"]),
+        ({"z": "a"}, ["x", "y", "x"]),
+        ({"y": "z", "a": "b"}, ["x", "z", "x"]),
+    ],
+)
+def test_name_mi_with_dict_like_duplicate_names(func, rename_dict, exp_names):
+    # GH#20421
+    mi = MultiIndex.from_arrays([[1, 2], [3, 4], [5, 6]], names=["x", "y", "x"])
+    result = getattr(mi, func)(rename_dict)
+    expected = MultiIndex.from_arrays([[1, 2], [3, 4], [5, 6]], names=exp_names)
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["rename", "set_names"])
+@pytest.mark.parametrize(
+    "rename_dict, exp_names",
+    [
+        ({"x": "z"}, ["z", "y"]),
+        ({"x": "z", "y": "x"}, ["z", "x"]),
+        ({"a": "z"}, ["x", "y"]),
+        ({}, ["x", "y"]),
+    ],
+)
+def test_name_mi_with_dict_like(func, rename_dict, exp_names):
+    # GH#20421
+    mi = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["x", "y"])
+    result = getattr(mi, func)(rename_dict)
+    expected = MultiIndex.from_arrays([[1, 2], [3, 4]], names=exp_names)
+    tm.assert_index_equal(result, expected)
+
+
+def test_index_name_with_dict_like_raising():
+    # GH#20421
+    ix = pd.Index([1, 2])
+    msg = "Can only pass dict-like as `names` for MultiIndex."
+    with pytest.raises(TypeError, match=msg):
+        ix.set_names({"x": "z"})
+
+
+def test_multiindex_name_and_level_raising():
+    # GH#20421
+    mi = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["x", "y"])
+    with pytest.raises(TypeError, match="Can not pass level for dictlike `names`."):
+        mi.set_names(names={"x": "z"}, level={"x": "z"})
diff --git a/pandas/pandas/tests/indexes/multi/test_partial_indexing.py b/pandas/pandas/tests/indexes/multi/test_partial_indexing.py
new file mode 100644
index 0000000000000..64cc1fa621b31
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_partial_indexing.py
@@ -0,0 +1,148 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    IndexSlice,
+    MultiIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df():
+    #                        c1
+    # 2016-01-01 00:00:00 a   0
+    #                     b   1
+    #                     c   2
+    # 2016-01-01 12:00:00 a   3
+    #                     b   4
+    #                     c   5
+    # 2016-01-02 00:00:00 a   6
+    #                     b   7
+    #                     c   8
+    # 2016-01-02 12:00:00 a   9
+    #                     b  10
+    #                     c  11
+    # 2016-01-03 00:00:00 a  12
+    #                     b  13
+    #                     c  14
+    dr = date_range("2016-01-01", "2016-01-03", freq="12h")
+    abc = ["a", "b", "c"]
+    mi = MultiIndex.from_product([dr, abc])
+    frame = DataFrame({"c1": range(15)}, index=mi)
+    return frame
+
+
+def test_partial_string_matching_single_index(df):
+    # partial string matching on a single index
+    for df_swap in [df.swaplevel(), df.swaplevel(0), df.swaplevel(0, 1)]:
+        df_swap = df_swap.sort_index()
+        just_a = df_swap.loc["a"]
+        result = just_a.loc["2016-01-01"]
+        expected = df.loc[IndexSlice[:, "a"], :].iloc[0:2]
+        expected.index = expected.index.droplevel(1)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_get_loc_partial_timestamp_multiindex(df):
+    mi = df.index
+    key = ("2016-01-01", "a")
+    loc = mi.get_loc(key)
+
+    expected = np.zeros(len(mi), dtype=bool)
+    expected[[0, 3]] = True
+    tm.assert_numpy_array_equal(loc, expected)
+
+    key2 = ("2016-01-02", "a")
+    loc2 = mi.get_loc(key2)
+    expected2 = np.zeros(len(mi), dtype=bool)
+    expected2[[6, 9]] = True
+    tm.assert_numpy_array_equal(loc2, expected2)
+
+    key3 = ("2016-01", "a")
+    loc3 = mi.get_loc(key3)
+    expected3 = np.zeros(len(mi), dtype=bool)
+    expected3[mi.get_level_values(1).get_loc("a")] = True
+    tm.assert_numpy_array_equal(loc3, expected3)
+
+    key4 = ("2016", "a")
+    loc4 = mi.get_loc(key4)
+    expected4 = expected3
+    tm.assert_numpy_array_equal(loc4, expected4)
+
+    # non-monotonic
+    taker = np.arange(len(mi), dtype=np.intp)
+    taker[::2] = taker[::-2]
+    mi2 = mi.take(taker)
+    loc5 = mi2.get_loc(key)
+    expected5 = np.zeros(len(mi2), dtype=bool)
+    expected5[[3, 14]] = True
+    tm.assert_numpy_array_equal(loc5, expected5)
+
+
+def test_partial_string_timestamp_multiindex(df):
+    # GH10331
+    df_swap = df.swaplevel(0, 1).sort_index()
+    SLC = IndexSlice
+
+    # indexing with IndexSlice
+    result = df.loc[SLC["2016-01-01":"2016-02-01", :], :]
+    expected = df
+    tm.assert_frame_equal(result, expected)
+
+    # match on secondary index
+    result = df_swap.loc[SLC[:, "2016-01-01":"2016-01-01"], :]
+    expected = df_swap.iloc[[0, 1, 5, 6, 10, 11]]
+    tm.assert_frame_equal(result, expected)
+
+    # partial string match on year only
+    result = df.loc["2016"]
+    expected = df
+    tm.assert_frame_equal(result, expected)
+
+    # partial string match on date
+    result = df.loc["2016-01-01"]
+    expected = df.iloc[0:6]
+    tm.assert_frame_equal(result, expected)
+
+    # partial string match on date and hour, from middle
+    result = df.loc["2016-01-02 12"]
+    # hourly resolution, same as index.levels[0], so we are _not_ slicing on
+    #  that level, so that level gets dropped
+    expected = df.iloc[9:12].droplevel(0)
+    tm.assert_frame_equal(result, expected)
+
+    # partial string match on secondary index
+    result = df_swap.loc[SLC[:, "2016-01-02"], :]
+    expected = df_swap.iloc[[2, 3, 7, 8, 12, 13]]
+    tm.assert_frame_equal(result, expected)
+
+    # tuple selector with partial string match on date
+    # "2016-01-01" has daily resolution, so _is_ a slice on the first level.
+    result = df.loc[("2016-01-01", "a"), :]
+    expected = df.iloc[[0, 3]]
+    expected = df.iloc[[0, 3]].droplevel(1)
+    tm.assert_frame_equal(result, expected)
+
+    # Slicing date on first level should break (of course) bc the DTI is the
+    #  second level on df_swap
+    with pytest.raises(KeyError, match="'2016-01-01'"):
+        df_swap.loc["2016-01-01"]
+
+
+def test_partial_string_timestamp_multiindex_str_key_raises(df):
+    # Even though this syntax works on a single index, this is somewhat
+    # ambiguous and we don't want to extend this behavior forward to work
+    # in multi-indexes. This would amount to selecting a scalar from a
+    # column.
+    with pytest.raises(KeyError, match="'2016-01-01'"):
+        df["2016-01-01"]
+
+
+def test_partial_string_timestamp_multiindex_daily_resolution(df):
+    # GH12685 (partial string with daily resolution or below)
+    result = df.loc[IndexSlice["2013-03":"2013-03", :], :]
+    expected = df.iloc[118:180]
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_pickle.py b/pandas/pandas/tests/indexes/multi/test_pickle.py
new file mode 100644
index 0000000000000..1d8b721404421
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_pickle.py
@@ -0,0 +1,10 @@
+import pytest
+
+from pandas import MultiIndex
+
+
+def test_pickle_compat_construction():
+    # this is testing for pickle compat
+    # need an object to create with
+    with pytest.raises(TypeError, match="Must pass both levels and codes"):
+        MultiIndex()
diff --git a/pandas/pandas/tests/indexes/multi/test_reindex.py b/pandas/pandas/tests/indexes/multi/test_reindex.py
new file mode 100644
index 0000000000000..d1b4fe8b98760
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_reindex.py
@@ -0,0 +1,174 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+def test_reindex(idx):
+    result, indexer = idx.reindex(list(idx[:4]))
+    assert isinstance(result, MultiIndex)
+    assert result.names == ["first", "second"]
+    assert [level.name for level in result.levels] == ["first", "second"]
+
+    result, indexer = idx.reindex(list(idx))
+    assert isinstance(result, MultiIndex)
+    assert indexer is None
+    assert result.names == ["first", "second"]
+    assert [level.name for level in result.levels] == ["first", "second"]
+
+
+def test_reindex_level(idx):
+    index = Index(["one"])
+
+    target, indexer = idx.reindex(index, level="second")
+    target2, indexer2 = index.reindex(idx, level="second")
+
+    exp_index = idx.join(index, level="second", how="right")
+    exp_index2 = idx.join(index, level="second", how="left")
+
+    assert target.equals(exp_index)
+    exp_indexer = np.array([0, 2, 4])
+    tm.assert_numpy_array_equal(indexer, exp_indexer, check_dtype=False)
+
+    assert target2.equals(exp_index2)
+    exp_indexer2 = np.array([0, -1, 0, -1, 0, -1])
+    tm.assert_numpy_array_equal(indexer2, exp_indexer2, check_dtype=False)
+
+    with pytest.raises(TypeError, match="Fill method not supported"):
+        idx.reindex(idx, method="pad", level="second")
+
+
+def test_reindex_preserves_names_when_target_is_list_or_ndarray(idx):
+    # GH6552
+    idx = idx.copy()
+    target = idx.copy()
+    idx.names = target.names = [None, None]
+
+    other_dtype = MultiIndex.from_product([[1, 2], [3, 4]])
+
+    # list & ndarray cases
+    assert idx.reindex([])[0].names == [None, None]
+    assert idx.reindex(np.array([]))[0].names == [None, None]
+    assert idx.reindex(target.tolist())[0].names == [None, None]
+    assert idx.reindex(target.values)[0].names == [None, None]
+    assert idx.reindex(other_dtype.tolist())[0].names == [None, None]
+    assert idx.reindex(other_dtype.values)[0].names == [None, None]
+
+    idx.names = ["foo", "bar"]
+    assert idx.reindex([])[0].names == ["foo", "bar"]
+    assert idx.reindex(np.array([]))[0].names == ["foo", "bar"]
+    assert idx.reindex(target.tolist())[0].names == ["foo", "bar"]
+    assert idx.reindex(target.values)[0].names == ["foo", "bar"]
+    assert idx.reindex(other_dtype.tolist())[0].names == ["foo", "bar"]
+    assert idx.reindex(other_dtype.values)[0].names == ["foo", "bar"]
+
+
+def test_reindex_lvl_preserves_names_when_target_is_list_or_array():
+    # GH7774
+    idx = MultiIndex.from_product([[0, 1], ["a", "b"]], names=["foo", "bar"])
+    assert idx.reindex([], level=0)[0].names == ["foo", "bar"]
+    assert idx.reindex([], level=1)[0].names == ["foo", "bar"]
+
+
+def test_reindex_lvl_preserves_type_if_target_is_empty_list_or_array(
+    using_infer_string,
+):
+    # GH7774
+    idx = MultiIndex.from_product([[0, 1], ["a", "b"]])
+    assert idx.reindex([], level=0)[0].levels[0].dtype.type == np.int64
+    exp = np.object_ if not using_infer_string else str
+    assert idx.reindex([], level=1)[0].levels[1].dtype.type == exp
+
+    # case with EA levels
+    cat = pd.Categorical(["foo", "bar"])
+    dti = pd.date_range("2016-01-01", periods=2, tz="US/Pacific")
+    mi = MultiIndex.from_product([cat, dti])
+    assert mi.reindex([], level=0)[0].levels[0].dtype == cat.dtype
+    assert mi.reindex([], level=1)[0].levels[1].dtype == dti.dtype
+
+
+def test_reindex_base(idx):
+    expected = np.arange(idx.size, dtype=np.intp)
+
+    actual = idx.get_indexer(idx)
+    tm.assert_numpy_array_equal(expected, actual)
+
+    with pytest.raises(ValueError, match="Invalid fill method"):
+        idx.get_indexer(idx, method="invalid")
+
+
+def test_reindex_non_unique():
+    idx = MultiIndex.from_tuples([(0, 0), (1, 1), (1, 1), (2, 2)])
+    a = pd.Series(np.arange(4), index=idx)
+    new_idx = MultiIndex.from_tuples([(0, 0), (1, 1), (2, 2)])
+
+    msg = "cannot handle a non-unique multi-index!"
+    with pytest.raises(ValueError, match=msg):
+        a.reindex(new_idx)
+
+
+@pytest.mark.parametrize("values", [[["a"], ["x"]], [[], []]])
+def test_reindex_empty_with_level(values):
+    # GH41170
+    idx = MultiIndex.from_arrays(values)
+    result, result_indexer = idx.reindex(np.array(["b"]), level=0)
+    expected = MultiIndex(levels=[["b"], values[1]], codes=[[], []])
+    expected_indexer = np.array([], dtype=result_indexer.dtype)
+    tm.assert_index_equal(result, expected)
+    tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+
+def test_reindex_not_all_tuples():
+    keys = [("i", "i"), ("i", "j"), ("j", "i"), "j"]
+    mi = MultiIndex.from_tuples(keys[:-1])
+    idx = Index(keys)
+    res, indexer = mi.reindex(idx)
+
+    tm.assert_index_equal(res, idx)
+    expected = np.array([0, 1, 2, -1], dtype=np.intp)
+    tm.assert_numpy_array_equal(indexer, expected)
+
+
+def test_reindex_limit_arg_with_multiindex():
+    # GH21247
+
+    idx = MultiIndex.from_tuples([(3, "A"), (4, "A"), (4, "B")])
+
+    df = pd.Series([0.02, 0.01, 0.012], index=idx)
+
+    new_idx = MultiIndex.from_tuples(
+        [
+            (3, "A"),
+            (3, "B"),
+            (4, "A"),
+            (4, "B"),
+            (4, "C"),
+            (5, "B"),
+            (5, "C"),
+            (6, "B"),
+            (6, "C"),
+        ]
+    )
+
+    with pytest.raises(
+        ValueError,
+        match="limit argument only valid if doing pad, backfill or nearest reindexing",
+    ):
+        df.reindex(new_idx, fill_value=0, limit=1)
+
+
+def test_reindex_with_none_in_nested_multiindex():
+    # GH42883
+    index = MultiIndex.from_tuples([(("a", None), 1), (("b", None), 2)])
+    index2 = MultiIndex.from_tuples([(("b", None), 2), (("a", None), 1)])
+    df1_dtype = pd.DataFrame([1, 2], index=index)
+    df2_dtype = pd.DataFrame([2, 1], index=index2)
+
+    result = df1_dtype.reindex_like(df2_dtype)
+    expected = df2_dtype
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_reshape.py b/pandas/pandas/tests/indexes/multi/test_reshape.py
new file mode 100644
index 0000000000000..cc3dadc6bb61c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_reshape.py
@@ -0,0 +1,224 @@
+from datetime import datetime
+import zoneinfo
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+def test_insert(idx):
+    # key contained in all levels
+    new_index = idx.insert(0, ("bar", "two"))
+    assert new_index.equal_levels(idx)
+    assert new_index[0] == ("bar", "two")
+
+    # key not contained in all levels
+    new_index = idx.insert(0, ("abc", "three"))
+
+    exp0 = Index(list(idx.levels[0]) + ["abc"], name="first")
+    tm.assert_index_equal(new_index.levels[0], exp0)
+    assert new_index.names == ["first", "second"]
+
+    exp1 = Index(list(idx.levels[1]) + ["three"], name="second")
+    tm.assert_index_equal(new_index.levels[1], exp1)
+    assert new_index[0] == ("abc", "three")
+
+    # key wrong length
+    msg = "Item must have length equal to number of levels"
+    with pytest.raises(ValueError, match=msg):
+        idx.insert(0, ("foo2",))
+
+    left = pd.DataFrame([["a", "b", 0], ["b", "d", 1]], columns=["1st", "2nd", "3rd"])
+    left.set_index(["1st", "2nd"], inplace=True)
+    ts = left["3rd"].copy(deep=True)
+
+    left.loc[("b", "x"), "3rd"] = 2
+    left.loc[("b", "a"), "3rd"] = -1
+    left.loc[("b", "b"), "3rd"] = 3
+    left.loc[("a", "x"), "3rd"] = 4
+    left.loc[("a", "w"), "3rd"] = 5
+    left.loc[("a", "a"), "3rd"] = 6
+
+    ts.loc[("b", "x")] = 2
+    ts.loc["b", "a"] = -1
+    ts.loc[("b", "b")] = 3
+    ts.loc["a", "x"] = 4
+    ts.loc[("a", "w")] = 5
+    ts.loc["a", "a"] = 6
+
+    right = pd.DataFrame(
+        [
+            ["a", "b", 0],
+            ["b", "d", 1],
+            ["b", "x", 2],
+            ["b", "a", -1],
+            ["b", "b", 3],
+            ["a", "x", 4],
+            ["a", "w", 5],
+            ["a", "a", 6],
+        ],
+        columns=["1st", "2nd", "3rd"],
+    )
+    right.set_index(["1st", "2nd"], inplace=True)
+    # FIXME data types changes to float because
+    # of intermediate nan insertion;
+    tm.assert_frame_equal(left, right, check_dtype=False)
+    tm.assert_series_equal(ts, right["3rd"])
+
+
+def test_insert2():
+    # GH9250
+    idx = (
+        [("test1", i) for i in range(5)]
+        + [("test2", i) for i in range(6)]
+        + [("test", 17), ("test", 18)]
+    )
+
+    left = pd.Series(np.linspace(0, 10, 11), MultiIndex.from_tuples(idx[:-2]))
+
+    left.loc[("test", 17)] = 11
+    left.loc[("test", 18)] = 12
+
+    right = pd.Series(np.linspace(0, 12, 13), MultiIndex.from_tuples(idx))
+
+    tm.assert_series_equal(left, right)
+
+
+def test_append(idx):
+    result = idx[:3].append(idx[3:])
+    assert result.equals(idx)
+
+    foos = [idx[:1], idx[1:3], idx[3:]]
+    result = foos[0].append(foos[1:])
+    assert result.equals(idx)
+
+    # empty
+    result = idx.append([])
+    assert result.equals(idx)
+
+
+def test_append_index():
+    idx1 = Index([1.1, 1.2, 1.3])
+    idx2 = pd.date_range("2011-01-01", freq="D", periods=3, tz="Asia/Tokyo")
+    idx3 = Index(["A", "B", "C"])
+
+    midx_lv2 = MultiIndex.from_arrays([idx1, idx2])
+    midx_lv3 = MultiIndex.from_arrays([idx1, idx2, idx3])
+
+    result = idx1.append(midx_lv2)
+
+    # see gh-7112
+    tz = zoneinfo.ZoneInfo("Asia/Tokyo")
+    expected_tuples = [
+        (1.1, datetime(2011, 1, 1, tzinfo=tz)),
+        (1.2, datetime(2011, 1, 2, tzinfo=tz)),
+        (1.3, datetime(2011, 1, 3, tzinfo=tz)),
+    ]
+    expected = Index([1.1, 1.2, 1.3] + expected_tuples)
+    tm.assert_index_equal(result, expected)
+
+    result = midx_lv2.append(idx1)
+    expected = Index(expected_tuples + [1.1, 1.2, 1.3])
+    tm.assert_index_equal(result, expected)
+
+    result = midx_lv2.append(midx_lv2)
+    expected = MultiIndex.from_arrays([idx1.append(idx1), idx2.append(idx2)])
+    tm.assert_index_equal(result, expected)
+
+    result = midx_lv2.append(midx_lv3)
+    tm.assert_index_equal(result, expected)
+
+    result = midx_lv3.append(midx_lv2)
+    expected = Index._simple_new(
+        np.array(
+            [
+                (1.1, datetime(2011, 1, 1, tzinfo=tz), "A"),
+                (1.2, datetime(2011, 1, 2, tzinfo=tz), "B"),
+                (1.3, datetime(2011, 1, 3, tzinfo=tz), "C"),
+            ]
+            + expected_tuples,
+            dtype=object,
+        ),
+        None,
+    )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("name, exp", [("b", "b"), ("c", None)])
+def test_append_names_match(name, exp):
+    # GH#48288
+    midx = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["a", "b"])
+    midx2 = MultiIndex.from_arrays([[3], [5]], names=["a", name])
+    result = midx.append(midx2)
+    expected = MultiIndex.from_arrays([[1, 2, 3], [3, 4, 5]], names=["a", exp])
+    tm.assert_index_equal(result, expected)
+
+
+def test_append_names_dont_match():
+    # GH#48288
+    midx = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["a", "b"])
+    midx2 = MultiIndex.from_arrays([[3], [5]], names=["x", "y"])
+    result = midx.append(midx2)
+    expected = MultiIndex.from_arrays([[1, 2, 3], [3, 4, 5]], names=None)
+    tm.assert_index_equal(result, expected)
+
+
+def test_append_overlapping_interval_levels():
+    # GH 54934
+    ivl1 = pd.IntervalIndex.from_breaks([0.0, 1.0, 2.0])
+    ivl2 = pd.IntervalIndex.from_breaks([0.5, 1.5, 2.5])
+    mi1 = MultiIndex.from_product([ivl1, ivl1])
+    mi2 = MultiIndex.from_product([ivl2, ivl2])
+    result = mi1.append(mi2)
+    expected = MultiIndex.from_tuples(
+        [
+            (pd.Interval(0.0, 1.0), pd.Interval(0.0, 1.0)),
+            (pd.Interval(0.0, 1.0), pd.Interval(1.0, 2.0)),
+            (pd.Interval(1.0, 2.0), pd.Interval(0.0, 1.0)),
+            (pd.Interval(1.0, 2.0), pd.Interval(1.0, 2.0)),
+            (pd.Interval(0.5, 1.5), pd.Interval(0.5, 1.5)),
+            (pd.Interval(0.5, 1.5), pd.Interval(1.5, 2.5)),
+            (pd.Interval(1.5, 2.5), pd.Interval(0.5, 1.5)),
+            (pd.Interval(1.5, 2.5), pd.Interval(1.5, 2.5)),
+        ]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_repeat():
+    reps = 2
+    numbers = [1, 2, 3]
+    names = np.array(["foo", "bar"])
+
+    m = MultiIndex.from_product([numbers, names], names=names)
+    expected = MultiIndex.from_product([numbers, names.repeat(reps)], names=names)
+    tm.assert_index_equal(m.repeat(reps), expected)
+
+
+def test_insert_base(idx):
+    result = idx[1:4]
+
+    # test 0th element
+    assert idx[0:4].equals(result.insert(0, idx[0]))
+
+
+def test_delete_base(idx):
+    expected = idx[1:]
+    result = idx.delete(0)
+    assert result.equals(expected)
+    assert result.name == expected.name
+
+    expected = idx[:-1]
+    result = idx.delete(-1)
+    assert result.equals(expected)
+    assert result.name == expected.name
+
+    msg = "index 6 is out of bounds for axis 0 with size 6"
+    with pytest.raises(IndexError, match=msg):
+        idx.delete(len(idx))
diff --git a/pandas/pandas/tests/indexes/multi/test_setops.py b/pandas/pandas/tests/indexes/multi/test_setops.py
new file mode 100644
index 0000000000000..f7544cf62e5fa
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_setops.py
@@ -0,0 +1,766 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.api.types import (
+    is_float_dtype,
+    is_unsigned_integer_dtype,
+)
+
+
+@pytest.mark.parametrize("case", [0.5, "xxx"])
+@pytest.mark.parametrize(
+    "method", ["intersection", "union", "difference", "symmetric_difference"]
+)
+def test_set_ops_error_cases(idx, case, sort, method):
+    # non-iterable input
+    msg = "Input must be Index or array-like"
+    with pytest.raises(TypeError, match=msg):
+        getattr(idx, method)(case, sort=sort)
+
+
+@pytest.mark.parametrize("klass", [MultiIndex, np.array, Series, list])
+def test_intersection_base(idx, sort, klass):
+    first = idx[2::-1]  # first 3 elements reversed
+    second = idx[:5]
+
+    if klass is not MultiIndex:
+        second = klass(second.values)
+
+    intersect = first.intersection(second, sort=sort)
+    if sort is None:
+        expected = first.sort_values()
+    else:
+        expected = first
+    tm.assert_index_equal(intersect, expected)
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        first.intersection([1, 2, 3], sort=sort)
+
+
+@pytest.mark.arm_slow
+@pytest.mark.parametrize("klass", [MultiIndex, np.array, Series, list])
+def test_union_base(idx, sort, klass):
+    first = idx[::-1]
+    second = idx[:5]
+
+    if klass is not MultiIndex:
+        second = klass(second.values)
+
+    union = first.union(second, sort=sort)
+    if sort is None:
+        expected = first.sort_values()
+    else:
+        expected = first
+    tm.assert_index_equal(union, expected)
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        first.union([1, 2, 3], sort=sort)
+
+
+def test_difference_base(idx, sort):
+    second = idx[4:]
+    answer = idx[:4]
+    result = idx.difference(second, sort=sort)
+
+    if sort is None:
+        answer = answer.sort_values()
+
+    assert result.equals(answer)
+    tm.assert_index_equal(result, answer)
+
+    # GH 10149
+    cases = [klass(second.values) for klass in [np.array, Series, list]]
+    for case in cases:
+        result = idx.difference(case, sort=sort)
+        tm.assert_index_equal(result, answer)
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        idx.difference([1, 2, 3], sort=sort)
+
+
+def test_symmetric_difference(idx, sort):
+    first = idx[1:]
+    second = idx[:-1]
+    answer = idx[[-1, 0]]
+    result = first.symmetric_difference(second, sort=sort)
+
+    if sort is None:
+        answer = answer.sort_values()
+
+    tm.assert_index_equal(result, answer)
+
+    # GH 10149
+    cases = [klass(second.values) for klass in [np.array, Series, list]]
+    for case in cases:
+        result = first.symmetric_difference(case, sort=sort)
+        tm.assert_index_equal(result, answer)
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        first.symmetric_difference([1, 2, 3], sort=sort)
+
+
+def test_multiindex_symmetric_difference():
+    # GH 13490
+    idx = MultiIndex.from_product([["a", "b"], ["A", "B"]], names=["a", "b"])
+    result = idx.symmetric_difference(idx)
+    assert result.names == idx.names
+
+    idx2 = idx.copy().rename(["A", "B"])
+    result = idx.symmetric_difference(idx2)
+    assert result.names == [None, None]
+
+
+def test_empty(idx):
+    # GH 15270
+    assert not idx.empty
+    assert idx[:0].empty
+
+
+def test_difference(idx, sort):
+    first = idx
+    result = first.difference(idx[-3:], sort=sort)
+    vals = idx[:-3].values
+
+    if sort is None:
+        vals = sorted(vals)
+
+    expected = MultiIndex.from_tuples(vals, sortorder=0, names=idx.names)
+
+    assert isinstance(result, MultiIndex)
+    assert result.equals(expected)
+    assert result.names == idx.names
+    tm.assert_index_equal(result, expected)
+
+    # empty difference: reflexive
+    result = idx.difference(idx, sort=sort)
+    expected = idx[:0]
+    assert result.equals(expected)
+    assert result.names == idx.names
+
+    # empty difference: superset
+    result = idx[-3:].difference(idx, sort=sort)
+    expected = idx[:0]
+    assert result.equals(expected)
+    assert result.names == idx.names
+
+    # empty difference: degenerate
+    result = idx[:0].difference(idx, sort=sort)
+    expected = idx[:0]
+    assert result.equals(expected)
+    assert result.names == idx.names
+
+    # names not the same
+    chunklet = idx[-3:]
+    chunklet.names = ["foo", "baz"]
+    result = first.difference(chunklet, sort=sort)
+    assert result.names == (None, None)
+
+    # empty, but non-equal
+    result = idx.difference(idx.sortlevel(1)[0], sort=sort)
+    assert len(result) == 0
+
+    # raise Exception called with non-MultiIndex
+    result = first.difference(first.values, sort=sort)
+    assert result.equals(first[:0])
+
+    # name from empty array
+    result = first.difference([], sort=sort)
+    assert first.equals(result)
+    assert first.names == result.names
+
+    # name from non-empty array
+    result = first.difference([("foo", "one")], sort=sort)
+    expected = MultiIndex.from_tuples(
+        [("bar", "one"), ("baz", "two"), ("foo", "two"), ("qux", "one"), ("qux", "two")]
+    )
+    expected.names = first.names
+    assert first.names == result.names
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        first.difference([1, 2, 3, 4, 5], sort=sort)
+
+
+def test_difference_sort_special():
+    # GH-24959
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+    # sort=None, the default
+    result = idx.difference([])
+    tm.assert_index_equal(result, idx)
+
+
+def test_difference_sort_special_true():
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+    result = idx.difference([], sort=True)
+    expected = MultiIndex.from_product([[0, 1], ["a", "b"]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_difference_sort_incomparable():
+    # GH-24959
+    idx = MultiIndex.from_product([[1, pd.Timestamp("2000"), 2], ["a", "b"]])
+
+    other = MultiIndex.from_product([[3, pd.Timestamp("2000"), 4], ["c", "d"]])
+    # sort=None, the default
+    msg = "sort order is undefined for incomparable objects"
+    with tm.assert_produces_warning(RuntimeWarning, match=msg):
+        result = idx.difference(other)
+    tm.assert_index_equal(result, idx)
+
+    # sort=False
+    result = idx.difference(other, sort=False)
+    tm.assert_index_equal(result, idx)
+
+
+def test_difference_sort_incomparable_true():
+    idx = MultiIndex.from_product([[1, pd.Timestamp("2000"), 2], ["a", "b"]])
+    other = MultiIndex.from_product([[3, pd.Timestamp("2000"), 4], ["c", "d"]])
+
+    # TODO: this is raising in constructing a Categorical when calling
+    #  algos.safe_sort. Should we catch and re-raise with a better message?
+    msg = "'values' is not ordered, please explicitly specify the categories order "
+    with pytest.raises(TypeError, match=msg):
+        idx.difference(other, sort=True)
+
+
+def test_union(idx, sort):
+    piece1 = idx[:5][::-1]
+    piece2 = idx[3:]
+
+    the_union = piece1.union(piece2, sort=sort)
+
+    if sort in (None, False):
+        tm.assert_index_equal(the_union.sort_values(), idx.sort_values())
+    else:
+        tm.assert_index_equal(the_union, idx)
+
+    # corner case, pass self or empty thing:
+    the_union = idx.union(idx, sort=sort)
+    tm.assert_index_equal(the_union, idx)
+
+    the_union = idx.union(idx[:0], sort=sort)
+    tm.assert_index_equal(the_union, idx)
+
+    tuples = idx.values
+    result = idx[:4].union(tuples[4:], sort=sort)
+    if sort is None:
+        tm.assert_index_equal(result.sort_values(), idx.sort_values())
+    else:
+        assert result.equals(idx)
+
+
+def test_union_with_regular_index(idx, using_infer_string):
+    other = Index(["A", "B", "C"])
+
+    result = other.union(idx)
+    assert ("foo", "one") in result
+    assert "B" in result
+
+    if using_infer_string:
+        with pytest.raises(NotImplementedError, match="Can only union"):
+            idx.union(other)
+    else:
+        msg = "The values in the array are unorderable"
+        with tm.assert_produces_warning(RuntimeWarning, match=msg):
+            result2 = idx.union(other)
+        # This is more consistent now, if sorting fails then we don't sort at all
+        # in the MultiIndex case.
+        assert not result.equals(result2)
+
+
+def test_intersection(idx, sort):
+    piece1 = idx[:5][::-1]
+    piece2 = idx[3:]
+
+    the_int = piece1.intersection(piece2, sort=sort)
+
+    if sort in (None, True):
+        tm.assert_index_equal(the_int, idx[3:5])
+    else:
+        tm.assert_index_equal(the_int.sort_values(), idx[3:5])
+
+    # corner case, pass self
+    the_int = idx.intersection(idx, sort=sort)
+    tm.assert_index_equal(the_int, idx)
+
+    # empty intersection: disjoint
+    empty = idx[:2].intersection(idx[2:], sort=sort)
+    expected = idx[:0]
+    assert empty.equals(expected)
+
+    tuples = idx.values
+    result = idx.intersection(tuples)
+    assert result.equals(idx)
+
+
+@pytest.mark.parametrize(
+    "method", ["intersection", "union", "difference", "symmetric_difference"]
+)
+def test_setop_with_categorical(idx, sort, method):
+    other = idx.to_flat_index().astype("category")
+    res_names = [None] * idx.nlevels
+
+    result = getattr(idx, method)(other, sort=sort)
+    expected = getattr(idx, method)(idx, sort=sort).rename(res_names)
+    tm.assert_index_equal(result, expected)
+
+    result = getattr(idx, method)(other[:5], sort=sort)
+    expected = getattr(idx, method)(idx[:5], sort=sort).rename(res_names)
+    tm.assert_index_equal(result, expected)
+
+
+def test_intersection_non_object(idx, sort):
+    other = Index(range(3), name="foo")
+
+    result = idx.intersection(other, sort=sort)
+    expected = MultiIndex(levels=idx.levels, codes=[[]] * idx.nlevels, names=None)
+    tm.assert_index_equal(result, expected, exact=True)
+
+    # if we pass a length-0 ndarray (i.e. no name, we retain our idx.name)
+    result = idx.intersection(np.asarray(other)[:0], sort=sort)
+    expected = MultiIndex(levels=idx.levels, codes=[[]] * idx.nlevels, names=idx.names)
+    tm.assert_index_equal(result, expected, exact=True)
+
+    msg = "other must be a MultiIndex or a list of tuples"
+    with pytest.raises(TypeError, match=msg):
+        # With non-zero length non-index, we try and fail to convert to tuples
+        idx.intersection(np.asarray(other), sort=sort)
+
+
+def test_intersect_equal_sort():
+    # GH-24959
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+    tm.assert_index_equal(idx.intersection(idx, sort=False), idx)
+    tm.assert_index_equal(idx.intersection(idx, sort=None), idx)
+
+
+def test_intersect_equal_sort_true():
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+    expected = MultiIndex.from_product([[0, 1], ["a", "b"]])
+    result = idx.intersection(idx, sort=True)
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("slice_", [slice(None), slice(0)])
+def test_union_sort_other_empty(slice_):
+    # https://github.com/pandas-dev/pandas/issues/24959
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+
+    # default, sort=None
+    other = idx[slice_]
+    tm.assert_index_equal(idx.union(other), idx)
+    tm.assert_index_equal(other.union(idx), idx)
+
+    # sort=False
+    tm.assert_index_equal(idx.union(other, sort=False), idx)
+
+
+def test_union_sort_other_empty_sort():
+    idx = MultiIndex.from_product([[1, 0], ["a", "b"]])
+    other = idx[:0]
+    result = idx.union(other, sort=True)
+    expected = MultiIndex.from_product([[0, 1], ["a", "b"]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_sort_other_incomparable():
+    # https://github.com/pandas-dev/pandas/issues/24959
+    idx = MultiIndex.from_product([[1, pd.Timestamp("2000")], ["a", "b"]])
+
+    # default, sort=None
+    with tm.assert_produces_warning(RuntimeWarning, match="are unorderable"):
+        result = idx.union(idx[:1])
+    tm.assert_index_equal(result, idx)
+
+    # sort=False
+    result = idx.union(idx[:1], sort=False)
+    tm.assert_index_equal(result, idx)
+
+
+def test_union_sort_other_incomparable_sort():
+    idx = MultiIndex.from_product([[1, pd.Timestamp("2000")], ["a", "b"]])
+    msg = "'<' not supported between instances of 'Timestamp' and 'int'"
+    with pytest.raises(TypeError, match=msg):
+        idx.union(idx[:1], sort=True)
+
+
+def test_union_non_object_dtype_raises():
+    # GH#32646 raise NotImplementedError instead of less-informative error
+    mi = MultiIndex.from_product([["a", "b"], [1, 2]])
+
+    idx = mi.levels[1]
+
+    msg = "Can only union MultiIndex with MultiIndex or Index of tuples"
+    with pytest.raises(NotImplementedError, match=msg):
+        mi.union(idx)
+
+
+def test_union_empty_self_different_names():
+    # GH#38423
+    mi = MultiIndex.from_arrays([[]])
+    mi2 = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["a", "b"])
+    result = mi.union(mi2)
+    expected = MultiIndex.from_arrays([[1, 2], [3, 4]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_multiindex_empty_rangeindex():
+    # GH#41234
+    mi = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["a", "b"])
+    ri = pd.RangeIndex(0)
+
+    result_left = mi.union(ri)
+    tm.assert_index_equal(mi, result_left, check_names=False)
+
+    result_right = ri.union(mi)
+    tm.assert_index_equal(mi, result_right, check_names=False)
+
+
+@pytest.mark.parametrize(
+    "method", ["union", "intersection", "difference", "symmetric_difference"]
+)
+def test_setops_sort_validation(method):
+    idx1 = MultiIndex.from_product([["a", "b"], [1, 2]])
+    idx2 = MultiIndex.from_product([["b", "c"], [1, 2]])
+
+    with pytest.raises(ValueError, match="The 'sort' keyword only takes"):
+        getattr(idx1, method)(idx2, sort=2)
+
+    # sort=True is supported as of GH#?
+    getattr(idx1, method)(idx2, sort=True)
+
+
+@pytest.mark.parametrize("val", [pd.NA, 100])
+def test_difference_keep_ea_dtypes(any_numeric_ea_dtype, val):
+    # GH#48606
+    midx = MultiIndex.from_arrays(
+        [Series([1, 2], dtype=any_numeric_ea_dtype), [2, 1]], names=["a", None]
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series([1, 2, val], dtype=any_numeric_ea_dtype), [1, 1, 3]]
+    )
+    result = midx.difference(midx2)
+    expected = MultiIndex.from_arrays([Series([1], dtype=any_numeric_ea_dtype), [2]])
+    tm.assert_index_equal(result, expected)
+
+    result = midx.difference(midx.sort_values(ascending=False))
+    expected = MultiIndex.from_arrays(
+        [Series([], dtype=any_numeric_ea_dtype), Series([], dtype=np.int64)],
+        names=["a", None],
+    )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [pd.NA, 5])
+def test_symmetric_difference_keeping_ea_dtype(any_numeric_ea_dtype, val):
+    # GH#48607
+    midx = MultiIndex.from_arrays(
+        [Series([1, 2], dtype=any_numeric_ea_dtype), [2, 1]], names=["a", None]
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series([1, 2, val], dtype=any_numeric_ea_dtype), [1, 1, 3]]
+    )
+    result = midx.symmetric_difference(midx2)
+    expected = MultiIndex.from_arrays(
+        [Series([1, 1, val], dtype=any_numeric_ea_dtype), [1, 2, 3]]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("tuples", "exp_tuples"),
+    [
+        ([("val1", "test1")], [("val1", "test1")]),
+        ([("val1", "test1"), ("val1", "test1")], [("val1", "test1")]),
+        (
+            [("val2", "test2"), ("val1", "test1")],
+            [("val2", "test2"), ("val1", "test1")],
+        ),
+    ],
+)
+def test_intersect_with_duplicates(tuples, exp_tuples):
+    # GH#36915
+    left = MultiIndex.from_tuples(tuples, names=["first", "second"])
+    right = MultiIndex.from_tuples(
+        [("val1", "test1"), ("val1", "test1"), ("val2", "test2")],
+        names=["first", "second"],
+    )
+    result = left.intersection(right)
+    expected = MultiIndex.from_tuples(exp_tuples, names=["first", "second"])
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, names, expected",
+    [
+        ((1,), None, [None, None]),
+        ((1,), ["a"], [None, None]),
+        ((1,), ["b"], [None, None]),
+        ((1, 2), ["c", "d"], [None, None]),
+        ((1, 2), ["b", "a"], [None, None]),
+        ((1, 2, 3), ["a", "b", "c"], [None, None]),
+        ((1, 2), ["a", "c"], ["a", None]),
+        ((1, 2), ["c", "b"], [None, "b"]),
+        ((1, 2), ["a", "b"], ["a", "b"]),
+        ((1, 2), [None, "b"], [None, "b"]),
+    ],
+)
+def test_maybe_match_names(data, names, expected):
+    # GH#38323
+    mi = MultiIndex.from_tuples([], names=["a", "b"])
+    mi2 = MultiIndex.from_tuples([data], names=names)
+    result = mi._maybe_match_names(mi2)
+    assert result == expected
+
+
+def test_intersection_equal_different_names():
+    # GH#30302
+    mi1 = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["c", "b"])
+    mi2 = MultiIndex.from_arrays([[1, 2], [3, 4]], names=["a", "b"])
+
+    result = mi1.intersection(mi2)
+    expected = MultiIndex.from_arrays([[1, 2], [3, 4]], names=[None, "b"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_intersection_different_names():
+    # GH#38323
+    mi = MultiIndex.from_arrays([[1], [3]], names=["c", "b"])
+    mi2 = MultiIndex.from_arrays([[1], [3]])
+    result = mi.intersection(mi2)
+    tm.assert_index_equal(result, mi2)
+
+
+def test_intersection_with_missing_values_on_both_sides(nulls_fixture):
+    # GH#38623
+    mi1 = MultiIndex.from_arrays([[3, nulls_fixture, 4, nulls_fixture], [1, 2, 4, 2]])
+    mi2 = MultiIndex.from_arrays([[3, nulls_fixture, 3], [1, 2, 4]])
+    result = mi1.intersection(mi2)
+    expected = MultiIndex.from_arrays([[3, nulls_fixture], [1, 2]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_with_missing_values_on_both_sides(nulls_fixture):
+    # GH#38623
+    mi1 = MultiIndex.from_arrays([[1, nulls_fixture]])
+    mi2 = MultiIndex.from_arrays([[1, nulls_fixture, 3]])
+    result = mi1.union(mi2)
+    expected = MultiIndex.from_arrays([[1, 3, nulls_fixture]])
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["float64", "Float64"])
+@pytest.mark.parametrize("sort", [None, False])
+def test_union_nan_got_duplicated(dtype, sort):
+    # GH#38977, GH#49010
+    mi1 = MultiIndex.from_arrays([pd.array([1.0, np.nan], dtype=dtype), [2, 3]])
+    mi2 = MultiIndex.from_arrays([pd.array([1.0, np.nan, 3.0], dtype=dtype), [2, 3, 4]])
+    result = mi1.union(mi2, sort=sort)
+    if sort is None:
+        expected = MultiIndex.from_arrays(
+            [pd.array([1.0, 3.0, np.nan], dtype=dtype), [2, 4, 3]]
+        )
+    else:
+        expected = mi2
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [4, 1])
+def test_union_keep_ea_dtype(any_numeric_ea_dtype, val):
+    # GH#48505
+
+    arr1 = Series([val, 2], dtype=any_numeric_ea_dtype)
+    arr2 = Series([2, 1], dtype=any_numeric_ea_dtype)
+    midx = MultiIndex.from_arrays([arr1, [1, 2]], names=["a", None])
+    midx2 = MultiIndex.from_arrays([arr2, [2, 1]])
+    result = midx.union(midx2)
+    if val == 4:
+        expected = MultiIndex.from_arrays(
+            [Series([1, 2, 4], dtype=any_numeric_ea_dtype), [1, 2, 1]]
+        )
+    else:
+        expected = MultiIndex.from_arrays(
+            [Series([1, 2], dtype=any_numeric_ea_dtype), [1, 2]]
+        )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("dupe_val", [3, pd.NA])
+def test_union_with_duplicates_keep_ea_dtype(dupe_val, any_numeric_ea_dtype):
+    # GH48900
+    mi1 = MultiIndex.from_arrays(
+        [
+            Series([1, dupe_val, 2], dtype=any_numeric_ea_dtype),
+            Series([1, dupe_val, 2], dtype=any_numeric_ea_dtype),
+        ]
+    )
+    mi2 = MultiIndex.from_arrays(
+        [
+            Series([2, dupe_val, dupe_val], dtype=any_numeric_ea_dtype),
+            Series([2, dupe_val, dupe_val], dtype=any_numeric_ea_dtype),
+        ]
+    )
+    result = mi1.union(mi2)
+    expected = MultiIndex.from_arrays(
+        [
+            Series([1, 2, dupe_val, dupe_val], dtype=any_numeric_ea_dtype),
+            Series([1, 2, dupe_val, dupe_val], dtype=any_numeric_ea_dtype),
+        ]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_union_duplicates(index, request):
+    # GH#38977
+    if index.empty or isinstance(index, (IntervalIndex, CategoricalIndex)):
+        pytest.skip(f"No duplicates in an empty {type(index).__name__}")
+
+    values = index.unique().values.tolist()
+    mi1 = MultiIndex.from_arrays([values, [1] * len(values)])
+    mi2 = MultiIndex.from_arrays([[values[0]] + values, [1] * (len(values) + 1)])
+    result = mi2.union(mi1)
+    expected = mi2.sort_values()
+    tm.assert_index_equal(result, expected)
+
+    if (
+        is_unsigned_integer_dtype(mi2.levels[0])
+        and (mi2.get_level_values(0) < 2**63).all()
+    ):
+        # GH#47294 - union uses lib.fast_zip, converting data to Python integers
+        # and loses type information. Result is then unsigned only when values are
+        # sufficiently large to require unsigned dtype. This happens only if other
+        # has dups or one of both have missing values
+        expected = expected.set_levels(
+            [expected.levels[0].astype(np.int64), expected.levels[1]]
+        )
+    elif is_float_dtype(mi2.levels[0]):
+        # mi2 has duplicates witch is a different path than above, Fix that path
+        # to use correct float dtype?
+        expected = expected.set_levels(
+            [expected.levels[0].astype(float), expected.levels[1]]
+        )
+
+    result = mi1.union(mi2)
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_keep_dtype_precision(any_real_numeric_dtype):
+    # GH#48498
+    arr1 = Series([4, 1, 1], dtype=any_real_numeric_dtype)
+    arr2 = Series([1, 4], dtype=any_real_numeric_dtype)
+    midx = MultiIndex.from_arrays([arr1, [2, 1, 1]], names=["a", None])
+    midx2 = MultiIndex.from_arrays([arr2, [1, 2]], names=["a", None])
+
+    result = midx.union(midx2)
+    expected = MultiIndex.from_arrays(
+        ([Series([1, 1, 4], dtype=any_real_numeric_dtype), [1, 1, 2]]),
+        names=["a", None],
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_keep_ea_dtype_with_na(any_numeric_ea_dtype):
+    # GH#48498
+    arr1 = Series([4, pd.NA], dtype=any_numeric_ea_dtype)
+    arr2 = Series([1, pd.NA], dtype=any_numeric_ea_dtype)
+    midx = MultiIndex.from_arrays([arr1, [2, 1]], names=["a", None])
+    midx2 = MultiIndex.from_arrays([arr2, [1, 2]])
+    result = midx.union(midx2)
+    expected = MultiIndex.from_arrays(
+        [Series([1, 4, pd.NA, pd.NA], dtype=any_numeric_ea_dtype), [1, 2, 1, 2]]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "levels1, levels2, codes1, codes2, names",
+    [
+        (
+            [["a", "b", "c"], [0, ""]],
+            [["c", "d", "b"], [""]],
+            [[0, 1, 2], [1, 1, 1]],
+            [[0, 1, 2], [0, 0, 0]],
+            ["name1", "name2"],
+        ),
+    ],
+)
+def test_intersection_lexsort_depth(levels1, levels2, codes1, codes2, names):
+    # GH#25169
+    mi1 = MultiIndex(levels=levels1, codes=codes1, names=names)
+    mi2 = MultiIndex(levels=levels2, codes=codes2, names=names)
+    mi_int = mi1.intersection(mi2)
+    assert mi_int._lexsort_depth == 2
+
+
+@pytest.mark.parametrize(
+    "a",
+    [pd.Categorical(["a", "b"], categories=["a", "b"]), ["a", "b"]],
+)
+@pytest.mark.parametrize("b_ordered", [True, False])
+def test_intersection_with_non_lex_sorted_categories(a, b_ordered):
+    # GH#49974
+    other = ["1", "2"]
+    b = pd.Categorical(["a", "b"], categories=["b", "a"], ordered=b_ordered)
+    df1 = DataFrame({"x": a, "y": other})
+    df2 = DataFrame({"x": b, "y": other})
+
+    expected = MultiIndex.from_arrays([a, other], names=["x", "y"])
+
+    res1 = MultiIndex.from_frame(df1).intersection(
+        MultiIndex.from_frame(df2.sort_values(["x", "y"]))
+    )
+    res2 = MultiIndex.from_frame(df1).intersection(MultiIndex.from_frame(df2))
+    res3 = MultiIndex.from_frame(df1.sort_values(["x", "y"])).intersection(
+        MultiIndex.from_frame(df2)
+    )
+    res4 = MultiIndex.from_frame(df1.sort_values(["x", "y"])).intersection(
+        MultiIndex.from_frame(df2.sort_values(["x", "y"]))
+    )
+
+    tm.assert_index_equal(res1, expected)
+    tm.assert_index_equal(res2, expected)
+    tm.assert_index_equal(res3, expected)
+    tm.assert_index_equal(res4, expected)
+
+
+@pytest.mark.parametrize("val", [pd.NA, 100])
+def test_intersection_keep_ea_dtypes(val, any_numeric_ea_dtype):
+    # GH#48604
+    midx = MultiIndex.from_arrays(
+        [Series([1, 2], dtype=any_numeric_ea_dtype), [2, 1]], names=["a", None]
+    )
+    midx2 = MultiIndex.from_arrays(
+        [Series([1, 2, val], dtype=any_numeric_ea_dtype), [1, 1, 3]]
+    )
+    result = midx.intersection(midx2)
+    expected = MultiIndex.from_arrays([Series([2], dtype=any_numeric_ea_dtype), [1]])
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_with_na_when_constructing_dataframe():
+    # GH43222
+    series1 = Series(
+        (1,),
+        index=MultiIndex.from_arrays(
+            [Series([None], dtype="str"), Series([None], dtype="str")]
+        ),
+    )
+    series2 = Series((10, 20), index=MultiIndex.from_tuples(((None, None), ("a", "b"))))
+    result = DataFrame([series1, series2])
+    expected = DataFrame({(np.nan, np.nan): [1.0, 10.0], ("a", "b"): [np.nan, 20.0]})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_sorting.py b/pandas/pandas/tests/indexes/multi/test_sorting.py
new file mode 100644
index 0000000000000..3d21ee8a57716
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_sorting.py
@@ -0,0 +1,346 @@
+import numpy as np
+import pytest
+
+from pandas.errors import UnsortedIndexError
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.indexes.frozen import FrozenList
+
+
+def test_sortlevel(idx):
+    tuples = list(idx)
+    np.random.default_rng(2).shuffle(tuples)
+
+    index = MultiIndex.from_tuples(tuples)
+
+    sorted_idx, _ = index.sortlevel(0)
+    expected = MultiIndex.from_tuples(sorted(tuples))
+    assert sorted_idx.equals(expected)
+
+    sorted_idx, _ = index.sortlevel(0, ascending=False)
+    assert sorted_idx.equals(expected[::-1])
+
+    sorted_idx, _ = index.sortlevel(1)
+    by1 = sorted(tuples, key=lambda x: (x[1], x[0]))
+    expected = MultiIndex.from_tuples(by1)
+    assert sorted_idx.equals(expected)
+
+    sorted_idx, _ = index.sortlevel(1, ascending=False)
+    assert sorted_idx.equals(expected[::-1])
+
+
+def test_sortlevel_not_sort_remaining():
+    mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+    sorted_idx, _ = mi.sortlevel("A", sort_remaining=False)
+    assert sorted_idx.equals(mi)
+
+
+def test_sortlevel_deterministic():
+    tuples = [
+        ("bar", "one"),
+        ("foo", "two"),
+        ("qux", "two"),
+        ("foo", "one"),
+        ("baz", "two"),
+        ("qux", "one"),
+    ]
+
+    index = MultiIndex.from_tuples(tuples)
+
+    sorted_idx, _ = index.sortlevel(0)
+    expected = MultiIndex.from_tuples(sorted(tuples))
+    assert sorted_idx.equals(expected)
+
+    sorted_idx, _ = index.sortlevel(0, ascending=False)
+    assert sorted_idx.equals(expected[::-1])
+
+    sorted_idx, _ = index.sortlevel(1)
+    by1 = sorted(tuples, key=lambda x: (x[1], x[0]))
+    expected = MultiIndex.from_tuples(by1)
+    assert sorted_idx.equals(expected)
+
+    sorted_idx, _ = index.sortlevel(1, ascending=False)
+    assert sorted_idx.equals(expected[::-1])
+
+
+def test_sortlevel_na_position():
+    # GH#51612
+    midx = MultiIndex.from_tuples([(1, np.nan), (1, 1)])
+    result = midx.sortlevel(level=[0, 1], na_position="last")[0]
+    expected = MultiIndex.from_tuples([(1, 1), (1, np.nan)])
+    tm.assert_index_equal(result, expected)
+
+
+def test_numpy_argsort(idx):
+    result = np.argsort(idx)
+    expected = idx.argsort()
+    tm.assert_numpy_array_equal(result, expected)
+
+    # these are the only two types that perform
+    # pandas compatibility input validation - the
+    # rest already perform separate (or no) such
+    # validation via their 'values' attribute as
+    # defined in pandas.core.indexes/base.py - they
+    # cannot be changed at the moment due to
+    # backwards compatibility concerns
+    if isinstance(type(idx), (CategoricalIndex, RangeIndex)):
+        msg = "the 'axis' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argsort(idx, axis=1)
+
+        msg = "the 'kind' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argsort(idx, kind="mergesort")
+
+        msg = "the 'order' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argsort(idx, order=("a", "b"))
+
+
+def test_unsortedindex():
+    # GH 11897
+    mi = MultiIndex.from_tuples(
+        [("z", "a"), ("x", "a"), ("y", "b"), ("x", "b"), ("y", "a"), ("z", "b")],
+        names=["one", "two"],
+    )
+    df = DataFrame([[i, 10 * i] for i in range(6)], index=mi, columns=["one", "two"])
+
+    # GH 16734: not sorted, but no real slicing
+    result = df.loc(axis=0)["z", "a"]
+    expected = df.iloc[0]
+    tm.assert_series_equal(result, expected)
+
+    msg = (
+        "MultiIndex slicing requires the index to be lexsorted: "
+        r"slicing on levels \[1\], lexsort depth 0"
+    )
+    with pytest.raises(UnsortedIndexError, match=msg):
+        df.loc(axis=0)["z", slice("a")]
+    df.sort_index(inplace=True)
+    assert len(df.loc(axis=0)["z", :]) == 2
+
+    with pytest.raises(KeyError, match="'q'"):
+        df.loc(axis=0)["q", :]
+
+
+def test_unsortedindex_doc_examples(performance_warning):
+    # https://pandas.pydata.org/pandas-docs/stable/advanced.html#sorting-a-multiindex
+    dfm = DataFrame(
+        {
+            "jim": [0, 0, 1, 1],
+            "joe": ["x", "x", "z", "y"],
+            "jolie": np.random.default_rng(2).random(4),
+        }
+    )
+
+    dfm = dfm.set_index(["jim", "joe"])
+    with tm.assert_produces_warning(performance_warning):
+        dfm.loc[(1, "z")]
+
+    msg = r"Key length \(2\) was greater than MultiIndex lexsort depth \(1\)"
+    with pytest.raises(UnsortedIndexError, match=msg):
+        dfm.loc[(0, "y") : (1, "z")]
+
+    assert not dfm.index._is_lexsorted()
+    assert dfm.index._lexsort_depth == 1
+
+    # sort it
+    dfm = dfm.sort_index()
+    dfm.loc[(1, "z")]
+    dfm.loc[(0, "y") : (1, "z")]
+
+    assert dfm.index._is_lexsorted()
+    assert dfm.index._lexsort_depth == 2
+
+
+def test_reconstruct_sort():
+    # starts off lexsorted & monotonic
+    mi = MultiIndex.from_arrays([["A", "A", "B", "B", "B"], [1, 2, 1, 2, 3]])
+    assert mi.is_monotonic_increasing
+    recons = mi._sort_levels_monotonic()
+    assert recons.is_monotonic_increasing
+    assert mi is recons
+
+    assert mi.equals(recons)
+    assert Index(mi.values).equals(Index(recons.values))
+
+    # cannot convert to lexsorted
+    mi = MultiIndex.from_tuples(
+        [("z", "a"), ("x", "a"), ("y", "b"), ("x", "b"), ("y", "a"), ("z", "b")],
+        names=["one", "two"],
+    )
+    assert not mi.is_monotonic_increasing
+    recons = mi._sort_levels_monotonic()
+    assert not recons.is_monotonic_increasing
+    assert mi.equals(recons)
+    assert Index(mi.values).equals(Index(recons.values))
+
+    # cannot convert to lexsorted
+    mi = MultiIndex(
+        levels=[["b", "d", "a"], [1, 2, 3]],
+        codes=[[0, 1, 0, 2], [2, 0, 0, 1]],
+        names=["col1", "col2"],
+    )
+    assert not mi.is_monotonic_increasing
+    recons = mi._sort_levels_monotonic()
+    assert not recons.is_monotonic_increasing
+    assert mi.equals(recons)
+    assert Index(mi.values).equals(Index(recons.values))
+
+
+def test_reconstruct_remove_unused():
+    # xref to GH 2770
+    df = DataFrame(
+        [["deleteMe", 1, 9], ["keepMe", 2, 9], ["keepMeToo", 3, 9]],
+        columns=["first", "second", "third"],
+    )
+    df2 = df.set_index(["first", "second"], drop=False)
+    df2 = df2[df2["first"] != "deleteMe"]
+
+    # removed levels are there
+    expected = MultiIndex(
+        levels=[["deleteMe", "keepMe", "keepMeToo"], [1, 2, 3]],
+        codes=[[1, 2], [1, 2]],
+        names=["first", "second"],
+    )
+    result = df2.index
+    tm.assert_index_equal(result, expected)
+
+    expected = MultiIndex(
+        levels=[["keepMe", "keepMeToo"], [2, 3]],
+        codes=[[0, 1], [0, 1]],
+        names=["first", "second"],
+    )
+    result = df2.index.remove_unused_levels()
+    tm.assert_index_equal(result, expected)
+
+    # idempotent
+    result2 = result.remove_unused_levels()
+    tm.assert_index_equal(result2, expected)
+    assert result2.is_(result)
+
+
+@pytest.mark.parametrize(
+    "first_type,second_type", [("int64", "int64"), ("datetime64[D]", "str")]
+)
+def test_remove_unused_levels_large(first_type, second_type):
+    # GH16556
+
+    # because tests should be deterministic (and this test in particular
+    # checks that levels are removed, which is not the case for every
+    # random input):
+    rng = np.random.default_rng(10)  # seed is arbitrary value that works
+
+    size = 1 << 16
+    df = DataFrame(
+        {
+            "first": rng.integers(0, 1 << 13, size).astype(first_type),
+            "second": rng.integers(0, 1 << 10, size).astype(second_type),
+            "third": rng.random(size),
+        }
+    )
+    df = df.groupby(["first", "second"]).sum()
+    df = df[df.third < 0.1]
+
+    result = df.index.remove_unused_levels()
+    assert len(result.levels[0]) < len(df.index.levels[0])
+    assert len(result.levels[1]) < len(df.index.levels[1])
+    assert result.equals(df.index)
+
+    expected = df.reset_index().set_index(["first", "second"]).index
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("level0", [["a", "d", "b"], ["a", "d", "b", "unused"]])
+@pytest.mark.parametrize(
+    "level1", [["w", "x", "y", "z"], ["w", "x", "y", "z", "unused"]]
+)
+def test_remove_unused_nan(level0, level1):
+    # GH 18417
+    mi = MultiIndex(levels=[level0, level1], codes=[[0, 2, -1, 1, -1], [0, 1, 2, 3, 2]])
+
+    result = mi.remove_unused_levels()
+    tm.assert_index_equal(result, mi)
+    for level in 0, 1:
+        assert "unused" not in result.levels[level]
+
+
+def test_argsort(idx):
+    result = idx.argsort()
+    expected = idx.values.argsort()
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_remove_unused_levels_with_nan():
+    # GH 37510
+    idx = Index([(1, np.nan), (3, 4)]).rename(["id1", "id2"])
+    idx = idx.set_levels(["a", np.nan], level="id1")
+    idx = idx.remove_unused_levels()
+    result = idx.levels
+    expected = FrozenList([["a", np.nan], [4]])
+    assert str(result) == str(expected)
+
+
+def test_sort_values_nan():
+    # GH48495, GH48626
+    midx = MultiIndex(levels=[["A", "B", "C"], ["D"]], codes=[[1, 0, 2], [-1, -1, 0]])
+    result = midx.sort_values()
+    expected = MultiIndex(
+        levels=[["A", "B", "C"], ["D"]], codes=[[0, 1, 2], [-1, -1, 0]]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_sort_values_incomparable():
+    # GH48495
+    mi = MultiIndex.from_arrays(
+        [
+            [1, Timestamp("2000-01-01")],
+            [3, 4],
+        ]
+    )
+    match = "'<' not supported between instances of 'Timestamp' and 'int'"
+    with pytest.raises(TypeError, match=match):
+        mi.sort_values()
+
+
+@pytest.mark.parametrize("na_position", ["first", "last"])
+@pytest.mark.parametrize("dtype", ["float64", "Int64", "Float64"])
+def test_sort_values_with_na_na_position(dtype, na_position):
+    # 51612
+    arrays = [
+        Series([1, 1, 2], dtype=dtype),
+        Series([1, None, 3], dtype=dtype),
+    ]
+    index = MultiIndex.from_arrays(arrays)
+    result = index.sort_values(na_position=na_position)
+    if na_position == "first":
+        arrays = [
+            Series([1, 1, 2], dtype=dtype),
+            Series([None, 1, 3], dtype=dtype),
+        ]
+    else:
+        arrays = [
+            Series([1, 1, 2], dtype=dtype),
+            Series([1, None, 3], dtype=dtype),
+        ]
+    expected = MultiIndex.from_arrays(arrays)
+    tm.assert_index_equal(result, expected)
+
+
+def test_sort_unnecessary_warning():
+    # GH#55386
+    midx = MultiIndex.from_tuples([(1.5, 2), (3.5, 3), (0, 1)])
+    midx = midx.set_levels([2.5, np.nan, 1], level=0)
+    result = midx.sort_values()
+    expected = MultiIndex.from_tuples([(1, 3), (2.5, 1), (np.nan, 2)])
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/multi/test_take.py b/pandas/pandas/tests/indexes/multi/test_take.py
new file mode 100644
index 0000000000000..543cba25c373b
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_take.py
@@ -0,0 +1,78 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_take(idx):
+    indexer = [4, 3, 0, 2]
+    result = idx.take(indexer)
+    expected = idx[indexer]
+    assert result.equals(expected)
+
+    # GH 10791
+    msg = "'MultiIndex' object has no attribute 'freq'"
+    with pytest.raises(AttributeError, match=msg):
+        idx.freq
+
+
+def test_take_invalid_kwargs(idx):
+    indices = [1, 2]
+
+    msg = r"take\(\) got an unexpected keyword argument 'foo'"
+    with pytest.raises(TypeError, match=msg):
+        idx.take(indices, foo=2)
+
+    msg = "the 'out' parameter is not supported"
+    with pytest.raises(ValueError, match=msg):
+        idx.take(indices, out=indices)
+
+    msg = "the 'mode' parameter is not supported"
+    with pytest.raises(ValueError, match=msg):
+        idx.take(indices, mode="clip")
+
+
+def test_take_fill_value():
+    # GH 12631
+    vals = [["A", "B"], [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-01-02")]]
+    idx = pd.MultiIndex.from_product(vals, names=["str", "dt"])
+
+    result = idx.take(np.array([1, 0, -1]))
+    exp_vals = [
+        ("A", pd.Timestamp("2011-01-02")),
+        ("A", pd.Timestamp("2011-01-01")),
+        ("B", pd.Timestamp("2011-01-02")),
+    ]
+    expected = pd.MultiIndex.from_tuples(exp_vals, names=["str", "dt"])
+    tm.assert_index_equal(result, expected)
+
+    # fill_value
+    result = idx.take(np.array([1, 0, -1]), fill_value=True)
+    exp_vals = [
+        ("A", pd.Timestamp("2011-01-02")),
+        ("A", pd.Timestamp("2011-01-01")),
+        (np.nan, pd.NaT),
+    ]
+    expected = pd.MultiIndex.from_tuples(exp_vals, names=["str", "dt"])
+    tm.assert_index_equal(result, expected)
+
+    # allow_fill=False
+    result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+    exp_vals = [
+        ("A", pd.Timestamp("2011-01-02")),
+        ("A", pd.Timestamp("2011-01-01")),
+        ("B", pd.Timestamp("2011-01-02")),
+    ]
+    expected = pd.MultiIndex.from_tuples(exp_vals, names=["str", "dt"])
+    tm.assert_index_equal(result, expected)
+
+    msg = "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+    with pytest.raises(ValueError, match=msg):
+        idx.take(np.array([1, 0, -2]), fill_value=True)
+    with pytest.raises(ValueError, match=msg):
+        idx.take(np.array([1, 0, -5]), fill_value=True)
+
+    msg = "index -5 is out of bounds for( axis 0 with)? size 4"
+    with pytest.raises(IndexError, match=msg):
+        idx.take(np.array([1, -5]))
diff --git a/pandas/pandas/tests/indexes/multi/test_util.py b/pandas/pandas/tests/indexes/multi/test_util.py
new file mode 100644
index 0000000000000..68792ce53f04e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/multi/test_util.py
@@ -0,0 +1,63 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.indexes.multi import cartesian_product
+
+
+class TestCartesianProduct:
+    def test_simple(self):
+        x, y = list("ABC"), [1, 22]
+        result1, result2 = cartesian_product([x, y])
+        expected1 = np.array(["A", "A", "B", "B", "C", "C"])
+        expected2 = np.array([1, 22, 1, 22, 1, 22])
+        tm.assert_numpy_array_equal(result1, expected1)
+        tm.assert_numpy_array_equal(result2, expected2)
+
+    def test_datetimeindex(self):
+        # regression test for GitHub issue #6439
+        # make sure that the ordering on datetimeindex is consistent
+        x = date_range("2000-01-01", periods=2)
+        result1, result2 = (Index(y).day for y in cartesian_product([x, x]))
+        expected1 = Index([1, 1, 2, 2], dtype=np.int32)
+        expected2 = Index([1, 2, 1, 2], dtype=np.int32)
+        tm.assert_index_equal(result1, expected1)
+        tm.assert_index_equal(result2, expected2)
+
+    @pytest.mark.parametrize("x, y", [[[], []], [[0, 1], []], [[], ["a", "b", "c"]]])
+    def test_empty(self, x, y):
+        # product of empty factors
+        expected1 = np.array([], dtype=np.asarray(x).dtype)
+        expected2 = np.array([], dtype=np.asarray(y).dtype)
+        result1, result2 = cartesian_product([x, y])
+        tm.assert_numpy_array_equal(result1, expected1)
+        tm.assert_numpy_array_equal(result2, expected2)
+
+    def test_empty_input(self):
+        # empty product (empty input):
+        result = cartesian_product([])
+        expected = []
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "X", [1, [1], [1, 2], [[1], 2], "a", ["a"], ["a", "b"], [["a"], "b"]]
+    )
+    def test_invalid_input(self, X):
+        msg = "Input must be a list-like of list-likes"
+
+        with pytest.raises(TypeError, match=msg):
+            cartesian_product(X=X)
+
+    def test_exceed_product_space(self):
+        # GH31355: raise useful error when produce space is too large
+        msg = "Product space too large to allocate arrays!"
+
+        dims = [np.arange(0, 22, dtype=np.int16) for i in range(12)] + [
+            (np.arange(15128, dtype=np.int16)),
+        ]
+        with pytest.raises(ValueError, match=msg):
+            cartesian_product(X=dims)
diff --git a/pandas/pandas/tests/indexes/numeric/__init__.py b/pandas/pandas/tests/indexes/numeric/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/numeric/test_astype.py b/pandas/pandas/tests/indexes/numeric/test_astype.py
new file mode 100644
index 0000000000000..1c2df6008de5d
--- /dev/null
+++ b/pandas/pandas/tests/indexes/numeric/test_astype.py
@@ -0,0 +1,95 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    to_datetime,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+
+class TestAstype:
+    def test_astype_float64_to_uint64(self):
+        # GH#45309 used to incorrectly return Index with int64 dtype
+        idx = Index([0.0, 5.0, 10.0, 15.0, 20.0], dtype=np.float64)
+        result = idx.astype("u8")
+        expected = Index([0, 5, 10, 15, 20], dtype=np.uint64)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        idx_with_negatives = idx - 10
+        with pytest.raises(ValueError, match="losslessly"):
+            idx_with_negatives.astype(np.uint64)
+
+    def test_astype_float64_to_object(self):
+        float_index = Index([0.0, 2.5, 5.0, 7.5, 10.0], dtype=np.float64)
+        result = float_index.astype(object)
+        assert result.equals(float_index)
+        assert float_index.equals(result)
+        assert isinstance(result, Index) and result.dtype == object
+
+    def test_astype_float64_mixed_to_object(self):
+        # mixed int-float
+        idx = Index([1.5, 2, 3, 4, 5], dtype=np.float64)
+        idx.name = "foo"
+        result = idx.astype(object)
+        assert result.equals(idx)
+        assert idx.equals(result)
+        assert isinstance(result, Index) and result.dtype == object
+
+    @pytest.mark.parametrize("dtype", ["int16", "int32", "int64"])
+    def test_astype_float64_to_int_dtype(self, dtype):
+        # GH#12881
+        # a float astype int
+        idx = Index([0, 1, 2], dtype=np.float64)
+        result = idx.astype(dtype)
+        expected = Index([0, 1, 2], dtype=dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        idx = Index([0, 1.1, 2], dtype=np.float64)
+        result = idx.astype(dtype)
+        expected = Index([0, 1, 2], dtype=dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    @pytest.mark.parametrize("dtype", ["float32", "float64"])
+    def test_astype_float64_to_float_dtype(self, dtype):
+        # GH#12881
+        # a float astype int
+        idx = Index([0, 1, 2], dtype=np.float64)
+        result = idx.astype(dtype)
+        assert isinstance(result, Index) and result.dtype == dtype
+
+    @pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+    def test_astype_float_to_datetimelike(self, dtype):
+        # GH#49660 pre-2.0 Index.astype from floating to M8/m8/Period raised,
+        #  inconsistent with Series.astype
+        idx = Index([0, 1.1, 2], dtype=np.float64)
+
+        result = idx.astype(dtype)
+        if dtype[0] == "M":
+            expected = to_datetime(idx.values)
+        else:
+            expected = to_timedelta(idx.values)
+        tm.assert_index_equal(result, expected)
+
+        # check that we match Series behavior
+        result = idx.to_series().set_axis(range(3)).astype(dtype)
+        expected = expected.to_series().set_axis(range(3))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [int, "int16", "int32", "int64"])
+    @pytest.mark.parametrize("non_finite", [np.inf, np.nan])
+    def test_cannot_cast_inf_to_int(self, non_finite, dtype):
+        # GH#13149
+        idx = Index([1, 2, non_finite], dtype=np.float64)
+
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(ValueError, match=msg):
+            idx.astype(dtype)
+
+    def test_astype_from_object(self):
+        index = Index([1.0, np.nan, 0.2], dtype="object")
+        result = index.astype(float)
+        expected = Index([1.0, np.nan, 0.2], dtype=np.float64)
+        assert result.dtype == expected.dtype
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/numeric/test_indexing.py b/pandas/pandas/tests/indexes/numeric/test_indexing.py
new file mode 100644
index 0000000000000..3c1b98d57b2a0
--- /dev/null
+++ b/pandas/pandas/tests/indexes/numeric/test_indexing.py
@@ -0,0 +1,607 @@
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+from pandas import (
+    NA,
+    Index,
+    RangeIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    ArrowExtensionArray,
+    FloatingArray,
+)
+
+
+class TestGetLoc:
+    def test_get_loc(self):
+        index = Index([0, 1, 2])
+        assert index.get_loc(1) == 1
+
+    def test_get_loc_raises_bad_label(self):
+        index = Index([0, 1, 2])
+        with pytest.raises(InvalidIndexError, match=r"\[1, 2\]"):
+            index.get_loc([1, 2])
+
+    def test_get_loc_float64(self):
+        idx = Index([0.0, 1.0, 2.0], dtype=np.float64)
+
+        with pytest.raises(KeyError, match="^'foo'$"):
+            idx.get_loc("foo")
+        with pytest.raises(KeyError, match=r"^1\.5$"):
+            idx.get_loc(1.5)
+        with pytest.raises(KeyError, match="^True$"):
+            idx.get_loc(True)
+        with pytest.raises(KeyError, match="^False$"):
+            idx.get_loc(False)
+
+    def test_get_loc_na(self):
+        idx = Index([np.nan, 1, 2], dtype=np.float64)
+        assert idx.get_loc(1) == 1
+        assert idx.get_loc(np.nan) == 0
+
+        idx = Index([np.nan, 1, np.nan], dtype=np.float64)
+        assert idx.get_loc(1) == 1
+
+        # representable by slice [0:2:2]
+        msg = "'Cannot get left slice bound for non-unique label: nan'"
+        with pytest.raises(KeyError, match=msg):
+            idx.slice_locs(np.nan)
+        # not representable by slice
+        idx = Index([np.nan, 1, np.nan, np.nan], dtype=np.float64)
+        assert idx.get_loc(1) == 1
+        msg = "'Cannot get left slice bound for non-unique label: nan"
+        with pytest.raises(KeyError, match=msg):
+            idx.slice_locs(np.nan)
+
+    def test_get_loc_missing_nan(self):
+        # GH#8569
+        idx = Index([1, 2], dtype=np.float64)
+        assert idx.get_loc(1) == 0
+        with pytest.raises(KeyError, match=r"^3$"):
+            idx.get_loc(3)
+        with pytest.raises(KeyError, match="^nan$"):
+            idx.get_loc(np.nan)
+        with pytest.raises(InvalidIndexError, match=r"\[nan\]"):
+            # listlike/non-hashable raises TypeError
+            idx.get_loc([np.nan])
+
+    @pytest.mark.parametrize("vals", [[1], [1.0], [Timestamp("2019-12-31")], ["test"]])
+    def test_get_loc_float_index_nan_with_method(self, vals):
+        # GH#39382
+        idx = Index(vals)
+        with pytest.raises(KeyError, match="nan"):
+            idx.get_loc(np.nan)
+
+    @pytest.mark.parametrize("dtype", ["f8", "i8", "u8"])
+    def test_get_loc_numericindex_none_raises(self, dtype):
+        # case that goes through searchsorted and key is non-comparable to values
+        arr = np.arange(10**7, dtype=dtype)
+        idx = Index(arr)
+        with pytest.raises(KeyError, match="None"):
+            idx.get_loc(None)
+
+    def test_get_loc_overflows(self):
+        # unique but non-monotonic goes through IndexEngine.mapping.get_item
+        idx = Index([0, 2, 1])
+
+        val = np.iinfo(np.int64).max + 1
+
+        with pytest.raises(KeyError, match=str(val)):
+            idx.get_loc(val)
+        with pytest.raises(KeyError, match=str(val)):
+            idx._engine.get_loc(val)
+
+
+class TestGetIndexer:
+    def test_get_indexer(self):
+        index1 = Index([1, 2, 3, 4, 5])
+        index2 = Index([2, 4, 6])
+
+        r1 = index1.get_indexer(index2)
+        e1 = np.array([1, 3, -1], dtype=np.intp)
+        tm.assert_almost_equal(r1, e1)
+
+    @pytest.mark.parametrize("reverse", [True, False])
+    @pytest.mark.parametrize(
+        "expected,method",
+        [
+            ([-1, 0, 0, 1, 1], "pad"),
+            ([-1, 0, 0, 1, 1], "ffill"),
+            ([0, 0, 1, 1, 2], "backfill"),
+            ([0, 0, 1, 1, 2], "bfill"),
+        ],
+    )
+    def test_get_indexer_methods(self, reverse, expected, method):
+        index1 = Index([1, 2, 3, 4, 5])
+        index2 = Index([2, 4, 6])
+        expected = np.array(expected, dtype=np.intp)
+        if reverse:
+            index1 = index1[::-1]
+            expected = expected[::-1]
+
+        result = index2.get_indexer(index1, method=method)
+        tm.assert_almost_equal(result, expected)
+
+    def test_get_indexer_invalid(self):
+        # GH10411
+        index = Index(np.arange(10))
+
+        with pytest.raises(ValueError, match="tolerance argument"):
+            index.get_indexer([1, 0], tolerance=1)
+
+        with pytest.raises(ValueError, match="limit argument"):
+            index.get_indexer([1, 0], limit=1)
+
+    @pytest.mark.parametrize(
+        "method, tolerance, indexer, expected",
+        [
+            ("pad", None, [0, 5, 9], [0, 5, 9]),
+            ("backfill", None, [0, 5, 9], [0, 5, 9]),
+            ("nearest", None, [0, 5, 9], [0, 5, 9]),
+            ("pad", 0, [0, 5, 9], [0, 5, 9]),
+            ("backfill", 0, [0, 5, 9], [0, 5, 9]),
+            ("nearest", 0, [0, 5, 9], [0, 5, 9]),
+            ("pad", None, [0.2, 1.8, 8.5], [0, 1, 8]),
+            ("backfill", None, [0.2, 1.8, 8.5], [1, 2, 9]),
+            ("nearest", None, [0.2, 1.8, 8.5], [0, 2, 9]),
+            ("pad", 1, [0.2, 1.8, 8.5], [0, 1, 8]),
+            ("backfill", 1, [0.2, 1.8, 8.5], [1, 2, 9]),
+            ("nearest", 1, [0.2, 1.8, 8.5], [0, 2, 9]),
+            ("pad", 0.2, [0.2, 1.8, 8.5], [0, -1, -1]),
+            ("backfill", 0.2, [0.2, 1.8, 8.5], [-1, 2, -1]),
+            ("nearest", 0.2, [0.2, 1.8, 8.5], [0, 2, -1]),
+        ],
+    )
+    def test_get_indexer_nearest(self, method, tolerance, indexer, expected):
+        index = Index(np.arange(10))
+
+        actual = index.get_indexer(indexer, method=method, tolerance=tolerance)
+        tm.assert_numpy_array_equal(actual, np.array(expected, dtype=np.intp))
+
+    @pytest.mark.parametrize("listtype", [list, tuple, Series, np.array])
+    @pytest.mark.parametrize(
+        "tolerance, expected",
+        [
+            [[0.3, 0.3, 0.1], [0, 2, -1]],
+            [[0.2, 0.1, 0.1], [0, -1, -1]],
+            [[0.1, 0.5, 0.5], [-1, 2, 9]],
+        ],
+    )
+    def test_get_indexer_nearest_listlike_tolerance(
+        self, tolerance, expected, listtype
+    ):
+        index = Index(np.arange(10))
+
+        actual = index.get_indexer(
+            [0.2, 1.8, 8.5], method="nearest", tolerance=listtype(tolerance)
+        )
+        tm.assert_numpy_array_equal(actual, np.array(expected, dtype=np.intp))
+
+    def test_get_indexer_nearest_error(self):
+        index = Index(np.arange(10))
+        with pytest.raises(ValueError, match="limit argument"):
+            index.get_indexer([1, 0], method="nearest", limit=1)
+
+        with pytest.raises(ValueError, match="tolerance size must match"):
+            index.get_indexer([1, 0], method="nearest", tolerance=[1, 2, 3])
+
+    @pytest.mark.parametrize(
+        "method,expected",
+        [("pad", [8, 7, 0]), ("backfill", [9, 8, 1]), ("nearest", [9, 7, 0])],
+    )
+    def test_get_indexer_nearest_decreasing(self, method, expected):
+        index = Index(np.arange(10))[::-1]
+
+        actual = index.get_indexer([0, 5, 9], method=method)
+        tm.assert_numpy_array_equal(actual, np.array([9, 4, 0], dtype=np.intp))
+
+        actual = index.get_indexer([0.2, 1.8, 8.5], method=method)
+        tm.assert_numpy_array_equal(actual, np.array(expected, dtype=np.intp))
+
+    @pytest.mark.parametrize("idx_dtype", ["int64", "float64", "uint64", "range"])
+    @pytest.mark.parametrize("method", ["get_indexer", "get_indexer_non_unique"])
+    def test_get_indexer_numeric_index_boolean_target(self, method, idx_dtype):
+        # GH 16877
+
+        if idx_dtype == "range":
+            numeric_index = RangeIndex(4)
+        else:
+            numeric_index = Index(np.arange(4, dtype=idx_dtype))
+
+        other = Index([True, False, True])
+
+        result = getattr(numeric_index, method)(other)
+        expected = np.array([-1, -1, -1], dtype=np.intp)
+        if method == "get_indexer":
+            tm.assert_numpy_array_equal(result, expected)
+        else:
+            missing = np.arange(3, dtype=np.intp)
+            tm.assert_numpy_array_equal(result[0], expected)
+            tm.assert_numpy_array_equal(result[1], missing)
+
+    @pytest.mark.parametrize("method", ["pad", "backfill", "nearest"])
+    def test_get_indexer_with_method_numeric_vs_bool(self, method):
+        left = Index([1, 2, 3])
+        right = Index([True, False])
+
+        with pytest.raises(TypeError, match="Cannot compare"):
+            left.get_indexer(right, method=method)
+
+        with pytest.raises(TypeError, match="Cannot compare"):
+            right.get_indexer(left, method=method)
+
+    def test_get_indexer_numeric_vs_bool(self):
+        left = Index([1, 2, 3])
+        right = Index([True, False])
+
+        res = left.get_indexer(right)
+        expected = -1 * np.ones(len(right), dtype=np.intp)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = right.get_indexer(left)
+        expected = -1 * np.ones(len(left), dtype=np.intp)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = left.get_indexer_non_unique(right)[0]
+        expected = -1 * np.ones(len(right), dtype=np.intp)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = right.get_indexer_non_unique(left)[0]
+        expected = -1 * np.ones(len(left), dtype=np.intp)
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_get_indexer_float64(self):
+        idx = Index([0.0, 1.0, 2.0], dtype=np.float64)
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(idx), np.array([0, 1, 2], dtype=np.intp)
+        )
+
+        target = [-0.1, 0.5, 1.1]
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "pad"), np.array([-1, 0, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "backfill"), np.array([0, 1, 2], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest"), np.array([0, 1, 1], dtype=np.intp)
+        )
+
+    def test_get_indexer_nan(self):
+        # GH#7820
+        result = Index([1, 2, np.nan], dtype=np.float64).get_indexer([np.nan])
+        expected = np.array([2], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_int64(self):
+        index = Index(range(0, 20, 2), dtype=np.int64)
+        target = Index(np.arange(10), dtype=np.int64)
+        indexer = index.get_indexer(target)
+        expected = np.array([0, -1, 1, -1, 2, -1, 3, -1, 4, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+        target = Index(np.arange(10), dtype=np.int64)
+        indexer = index.get_indexer(target, method="pad")
+        expected = np.array([0, 0, 1, 1, 2, 2, 3, 3, 4, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+        target = Index(np.arange(10), dtype=np.int64)
+        indexer = index.get_indexer(target, method="backfill")
+        expected = np.array([0, 1, 1, 2, 2, 3, 3, 4, 4, 5], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+    def test_get_indexer_uint64(self):
+        index_large = Index(
+            [2**63, 2**63 + 10, 2**63 + 15, 2**63 + 20, 2**63 + 25],
+            dtype=np.uint64,
+        )
+        target = Index(np.arange(10).astype("uint64") * 5 + 2**63)
+        indexer = index_large.get_indexer(target)
+        expected = np.array([0, -1, 1, 2, 3, 4, -1, -1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+        target = Index(np.arange(10).astype("uint64") * 5 + 2**63)
+        indexer = index_large.get_indexer(target, method="pad")
+        expected = np.array([0, 0, 1, 2, 3, 4, 4, 4, 4, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+        target = Index(np.arange(10).astype("uint64") * 5 + 2**63)
+        indexer = index_large.get_indexer(target, method="backfill")
+        expected = np.array([0, 1, 1, 2, 3, 4, -1, -1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+    @pytest.mark.parametrize("val, val2", [(4, 5), (4, 4), (4, NA), (NA, NA)])
+    def test_get_loc_masked(self, val, val2, any_numeric_ea_and_arrow_dtype):
+        # GH#39133
+        idx = Index([1, 2, 3, val, val2], dtype=any_numeric_ea_and_arrow_dtype)
+        result = idx.get_loc(2)
+        assert result == 1
+
+        with pytest.raises(KeyError, match="9"):
+            idx.get_loc(9)
+
+    def test_get_loc_masked_na(self, any_numeric_ea_and_arrow_dtype):
+        # GH#39133
+        idx = Index([1, 2, NA], dtype=any_numeric_ea_and_arrow_dtype)
+        result = idx.get_loc(NA)
+        assert result == 2
+
+        idx = Index([1, 2, NA, NA], dtype=any_numeric_ea_and_arrow_dtype)
+        result = idx.get_loc(NA)
+        tm.assert_numpy_array_equal(result, np.array([False, False, True, True]))
+
+        idx = Index([1, 2, 3], dtype=any_numeric_ea_and_arrow_dtype)
+        with pytest.raises(KeyError, match="NA"):
+            idx.get_loc(NA)
+
+    def test_get_loc_masked_na_and_nan(self):
+        # GH#39133
+        idx = Index(
+            FloatingArray(
+                np.array([1, 2, 1, np.nan]), mask=np.array([False, False, True, False])
+            )
+        )
+        result = idx.get_loc(NA)
+        assert result == 2
+        result = idx.get_loc(np.nan)
+        assert result == 3
+
+        idx = Index(
+            FloatingArray(np.array([1, 2, 1.0]), mask=np.array([False, False, True]))
+        )
+        result = idx.get_loc(NA)
+        assert result == 2
+        with pytest.raises(KeyError, match="nan"):
+            idx.get_loc(np.nan)
+
+        idx = Index(
+            FloatingArray(
+                np.array([1, 2, np.nan]), mask=np.array([False, False, False])
+            )
+        )
+        result = idx.get_loc(np.nan)
+        assert result == 2
+        with pytest.raises(KeyError, match="NA"):
+            idx.get_loc(NA)
+
+    @pytest.mark.parametrize("val", [4, 2])
+    def test_get_indexer_masked_na(self, any_numeric_ea_and_arrow_dtype, val):
+        # GH#39133
+        idx = Index([1, 2, NA, 3, val], dtype=any_numeric_ea_and_arrow_dtype)
+        result = idx.get_indexer_for([1, NA, 5])
+        expected = np.array([0, 2, -1])
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize("dtype", ["boolean", "bool[pyarrow]"])
+    def test_get_indexer_masked_na_boolean(self, dtype):
+        # GH#39133
+        if dtype == "bool[pyarrow]":
+            pytest.importorskip("pyarrow")
+        idx = Index([True, False, NA], dtype=dtype)
+        result = idx.get_loc(False)
+        assert result == 1
+        result = idx.get_loc(NA)
+        assert result == 2
+
+    def test_get_indexer_arrow_dictionary_target(self):
+        pa = pytest.importorskip("pyarrow")
+        target = Index(
+            ArrowExtensionArray(
+                pa.array([1, 2], type=pa.dictionary(pa.int8(), pa.int8()))
+            )
+        )
+        idx = Index([1])
+
+        result = idx.get_indexer(target)
+        expected = np.array([0, -1], dtype=np.int64)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result_1, result_2 = idx.get_indexer_non_unique(target)
+        expected_1, expected_2 = (
+            np.array([0, -1], dtype=np.int64),
+            np.array([1], dtype=np.int64),
+        )
+        tm.assert_numpy_array_equal(result_1, expected_1)
+        tm.assert_numpy_array_equal(result_2, expected_2)
+
+
+class TestWhere:
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(np.arange(5, dtype="float64")),
+            Index(range(0, 20, 2), dtype=np.int64),
+            Index(np.arange(5, dtype="uint64")),
+        ],
+    )
+    def test_where(self, listlike_box, index):
+        cond = [True] * len(index)
+        expected = index
+        result = index.where(listlike_box(cond))
+
+        cond = [False] + [True] * (len(index) - 1)
+        expected = Index([index._na_value] + index[1:].tolist(), dtype=np.float64)
+        result = index.where(listlike_box(cond))
+        tm.assert_index_equal(result, expected)
+
+    def test_where_uint64(self):
+        idx = Index([0, 6, 2], dtype=np.uint64)
+        mask = np.array([False, True, False])
+        other = np.array([1], dtype=np.int64)
+
+        expected = Index([1, 6, 1], dtype=np.uint64)
+
+        result = idx.where(mask, other)
+        tm.assert_index_equal(result, expected)
+
+        result = idx.putmask(~mask, other)
+        tm.assert_index_equal(result, expected)
+
+    def test_where_infers_type_instead_of_trying_to_convert_string_to_float(self):
+        # GH 32413
+        index = Index([1, np.nan])
+        cond = index.notna()
+        other = Index(["a", "b"], dtype="string")
+
+        expected = Index([1.0, "b"])
+        result = index.where(cond, other)
+
+        tm.assert_index_equal(result, expected)
+
+
+class TestTake:
+    @pytest.mark.parametrize("idx_dtype", [np.float64, np.int64, np.uint64])
+    def test_take_preserve_name(self, idx_dtype):
+        index = Index([1, 2, 3, 4], dtype=idx_dtype, name="foo")
+        taken = index.take([3, 0, 1])
+        assert index.name == taken.name
+
+    def test_take_fill_value_float64(self):
+        # GH 12631
+        idx = Index([1.0, 2.0, 3.0], name="xxx", dtype=np.float64)
+        result = idx.take(np.array([1, 0, -1]))
+        expected = Index([2.0, 1.0, 3.0], dtype=np.float64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = Index([2.0, 1.0, np.nan], dtype=np.float64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = Index([2.0, 1.0, 3.0], dtype=np.float64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64])
+    def test_take_fill_value_ints(self, dtype):
+        # see gh-12631
+        idx = Index([1, 2, 3], dtype=dtype, name="xxx")
+        result = idx.take(np.array([1, 0, -1]))
+        expected = Index([2, 1, 3], dtype=dtype, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        name = type(idx).__name__
+        msg = f"Unable to fill values because {name} cannot contain NA"
+
+        # fill_value=True
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -1]), fill_value=True)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = Index([2, 1, 3], dtype=dtype, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+
+class TestContains:
+    @pytest.mark.parametrize("dtype", [np.float64, np.int64, np.uint64])
+    def test_contains_none(self, dtype):
+        # GH#35788 should return False, not raise TypeError
+        index = Index([0, 1, 2, 3, 4], dtype=dtype)
+        assert None not in index
+
+    def test_contains_float64_nans(self):
+        index = Index([1.0, 2.0, np.nan], dtype=np.float64)
+        assert np.nan in index
+
+    def test_contains_float64_not_nans(self):
+        index = Index([1.0, 2.0, np.nan], dtype=np.float64)
+        assert 1.0 in index
+
+
+class TestSliceLocs:
+    @pytest.mark.parametrize("dtype", [int, float])
+    def test_slice_locs(self, dtype):
+        index = Index(np.array([0, 1, 2, 5, 6, 7, 9, 10], dtype=dtype))
+        n = len(index)
+
+        assert index.slice_locs(start=2) == (2, n)
+        assert index.slice_locs(start=3) == (3, n)
+        assert index.slice_locs(3, 8) == (3, 6)
+        assert index.slice_locs(5, 10) == (3, n)
+        assert index.slice_locs(end=8) == (0, 6)
+        assert index.slice_locs(end=9) == (0, 7)
+
+        # reversed
+        index2 = index[::-1]
+        assert index2.slice_locs(8, 2) == (2, 6)
+        assert index2.slice_locs(7, 3) == (2, 5)
+
+    @pytest.mark.parametrize("dtype", [int, float])
+    def test_slice_locs_float_locs(self, dtype):
+        index = Index(np.array([0, 1, 2, 5, 6, 7, 9, 10], dtype=dtype))
+        n = len(index)
+        assert index.slice_locs(5.0, 10.0) == (3, n)
+        assert index.slice_locs(4.5, 10.5) == (3, 8)
+
+        index2 = index[::-1]
+        assert index2.slice_locs(8.5, 1.5) == (2, 6)
+        assert index2.slice_locs(10.5, -1) == (0, n)
+
+    @pytest.mark.parametrize("dtype", [int, float])
+    def test_slice_locs_dup_numeric(self, dtype):
+        index = Index(np.array([10, 12, 12, 14], dtype=dtype))
+        assert index.slice_locs(12, 12) == (1, 3)
+        assert index.slice_locs(11, 13) == (1, 3)
+
+        index2 = index[::-1]
+        assert index2.slice_locs(12, 12) == (1, 3)
+        assert index2.slice_locs(13, 11) == (1, 3)
+
+    def test_slice_locs_na(self):
+        index = Index([np.nan, 1, 2])
+        assert index.slice_locs(1) == (1, 3)
+        assert index.slice_locs(np.nan) == (0, 3)
+
+        index = Index([0, np.nan, np.nan, 1, 2])
+        assert index.slice_locs(np.nan) == (1, 5)
+
+    def test_slice_locs_na_raises(self):
+        index = Index([np.nan, 1, 2])
+        with pytest.raises(KeyError, match=""):
+            index.slice_locs(start=1.5)
+
+        with pytest.raises(KeyError, match=""):
+            index.slice_locs(end=1.5)
+
+
+class TestGetSliceBounds:
+    @pytest.mark.parametrize("side, expected", [("left", 4), ("right", 5)])
+    def test_get_slice_bounds_within(self, side, expected):
+        index = Index(range(6))
+        result = index.get_slice_bound(4, side=side)
+        assert result == expected
+
+    @pytest.mark.parametrize("side", ["left", "right"])
+    @pytest.mark.parametrize("bound, expected", [(-1, 0), (10, 6)])
+    def test_get_slice_bounds_outside(self, side, expected, bound):
+        index = Index(range(6))
+        result = index.get_slice_bound(bound, side=side)
+        assert result == expected
diff --git a/pandas/pandas/tests/indexes/numeric/test_join.py b/pandas/pandas/tests/indexes/numeric/test_join.py
new file mode 100644
index 0000000000000..4737af987a04e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/numeric/test_join.py
@@ -0,0 +1,368 @@
+import numpy as np
+import pytest
+
+import pandas._testing as tm
+from pandas.core.indexes.api import Index
+
+
+class TestJoinInt64Index:
+    def test_join_non_unique(self):
+        left = Index([4, 4, 3, 3])
+
+        joined, lidx, ridx = left.join(left, return_indexers=True)
+
+        exp_joined = Index([4, 4, 4, 4, 3, 3, 3, 3])
+        tm.assert_index_equal(joined, exp_joined)
+
+        exp_lidx = np.array([0, 0, 1, 1, 2, 2, 3, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(lidx, exp_lidx)
+
+        exp_ridx = np.array([0, 1, 0, 1, 2, 3, 2, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(ridx, exp_ridx)
+
+    def test_join_inner(self):
+        index = Index(range(0, 20, 2), dtype=np.int64, name="lhs")
+        other = Index([7, 12, 25, 1, 2, 5], dtype=np.int64, name="rhs")
+        other_mono = Index([1, 2, 5, 7, 12, 25], dtype=np.int64, name="rhs")
+
+        # not monotonic
+        res, lidx, ridx = index.join(other, how="inner", return_indexers=True)
+
+        eres = Index([2, 12], dtype=np.int64, name="lhs")
+        elidx = np.array([1, 6], dtype=np.intp)
+        eridx = np.array([4, 1], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index.join(other_mono, how="inner", return_indexers=True)
+
+        res2 = index.intersection(other_mono).set_names(["lhs"])
+        tm.assert_index_equal(res, res2)
+
+        elidx = np.array([1, 6], dtype=np.intp)
+        eridx = np.array([1, 4], dtype=np.intp)
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_left(self):
+        index = Index(range(0, 20, 2), dtype=np.int64, name="lhs")
+        other = Index([7, 12, 25, 1, 2, 5], dtype=np.int64, name="rhs")
+        other_mono = Index([1, 2, 5, 7, 12, 25], dtype=np.int64, name="rhs")
+
+        # not monotonic
+        res, lidx, ridx = index.join(other, how="left", return_indexers=True)
+        eres = index
+        eridx = np.array([-1, 4, -1, -1, -1, -1, 1, -1, -1, -1], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index.join(other_mono, how="left", return_indexers=True)
+        eridx = np.array([-1, 1, -1, -1, -1, -1, 4, -1, -1, -1], dtype=np.intp)
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # non-unique
+        idx = Index([1, 1, 2, 5], name="rhs")
+        idx2 = Index([1, 2, 5, 7, 9], name="lhs")
+        res, lidx, ridx = idx2.join(idx, how="left", return_indexers=True)
+        eres = Index([1, 1, 2, 5, 7, 9], name="lhs")  # 1 is in idx2, so it should be x2
+        eridx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+        elidx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_right(self):
+        index = Index(range(0, 20, 2), dtype=np.int64, name="lhs")
+        other = Index([7, 12, 25, 1, 2, 5], dtype=np.int64, name="rhs")
+        other_mono = Index([1, 2, 5, 7, 12, 25], dtype=np.int64, name="rhs")
+
+        # not monotonic
+        res, lidx, ridx = index.join(other, how="right", return_indexers=True)
+        eres = other
+        elidx = np.array([-1, 6, -1, -1, 1, -1], dtype=np.intp)
+
+        assert isinstance(other, Index) and other.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        assert ridx is None
+
+        # monotonic
+        res, lidx, ridx = index.join(other_mono, how="right", return_indexers=True)
+        eres = other_mono
+        elidx = np.array([-1, 1, -1, -1, 6, -1], dtype=np.intp)
+        assert isinstance(other, Index) and other.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        assert ridx is None
+
+        # non-unique
+        idx = Index([1, 1, 2, 5], name="lhs")
+        idx2 = Index([1, 2, 5, 7, 9], name="rhs")
+        res, lidx, ridx = idx.join(idx2, how="right", return_indexers=True)
+        eres = Index([1, 1, 2, 5, 7, 9], name="rhs")  # 1 is in idx2, so it should be x2
+        elidx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+        eridx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_non_int_index(self):
+        index = Index(range(0, 20, 2), dtype=np.int64)
+        other = Index([3, 6, 7, 8, 10], dtype=object)
+
+        outer = index.join(other, how="outer")
+        outer2 = other.join(index, how="outer")
+        expected = Index([0, 2, 3, 4, 6, 7, 8, 10, 12, 14, 16, 18])
+        tm.assert_index_equal(outer, outer2)
+        tm.assert_index_equal(outer, expected)
+
+        inner = index.join(other, how="inner")
+        inner2 = other.join(index, how="inner")
+        expected = Index([6, 8, 10])
+        tm.assert_index_equal(inner, inner2)
+        tm.assert_index_equal(inner, expected)
+
+        left = index.join(other, how="left")
+        tm.assert_index_equal(left, index.astype(object))
+
+        left2 = other.join(index, how="left")
+        tm.assert_index_equal(left2, other)
+
+        right = index.join(other, how="right")
+        tm.assert_index_equal(right, other)
+
+        right2 = other.join(index, how="right")
+        tm.assert_index_equal(right2, index.astype(object))
+
+    def test_join_outer(self):
+        index = Index(range(0, 20, 2), dtype=np.int64)
+        other = Index([7, 12, 25, 1, 2, 5], dtype=np.int64)
+        other_mono = Index([1, 2, 5, 7, 12, 25], dtype=np.int64)
+
+        # not monotonic
+        # guarantee of sortedness
+        res, lidx, ridx = index.join(other, how="outer", return_indexers=True)
+        noidx_res = index.join(other, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        eres = Index([0, 1, 2, 4, 5, 6, 7, 8, 10, 12, 14, 16, 18, 25], dtype=np.int64)
+        elidx = np.array([0, -1, 1, 2, -1, 3, -1, 4, 5, 6, 7, 8, 9, -1], dtype=np.intp)
+        eridx = np.array(
+            [-1, 3, 4, -1, 5, -1, 0, -1, -1, 1, -1, -1, -1, 2], dtype=np.intp
+        )
+
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index.join(other_mono, how="outer", return_indexers=True)
+        noidx_res = index.join(other_mono, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        elidx = np.array([0, -1, 1, 2, -1, 3, -1, 4, 5, 6, 7, 8, 9, -1], dtype=np.intp)
+        eridx = np.array(
+            [-1, 0, 1, -1, 2, -1, 3, -1, -1, 4, -1, -1, -1, 5], dtype=np.intp
+        )
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+
+class TestJoinUInt64Index:
+    @pytest.fixture
+    def index_large(self):
+        # large values used in TestUInt64Index where no compat needed with int64/float64
+        large = [2**63, 2**63 + 10, 2**63 + 15, 2**63 + 20, 2**63 + 25]
+        return Index(large, dtype=np.uint64)
+
+    def test_join_inner(self, index_large):
+        other = Index(2**63 + np.array([7, 12, 25, 1, 2, 10], dtype="uint64"))
+        other_mono = Index(2**63 + np.array([1, 2, 7, 10, 12, 25], dtype="uint64"))
+
+        # not monotonic
+        res, lidx, ridx = index_large.join(other, how="inner", return_indexers=True)
+
+        # no guarantee of sortedness, so sort for comparison purposes
+        ind = res.argsort()
+        res = res.take(ind)
+        lidx = lidx.take(ind)
+        ridx = ridx.take(ind)
+
+        eres = Index(2**63 + np.array([10, 25], dtype="uint64"))
+        elidx = np.array([1, 4], dtype=np.intp)
+        eridx = np.array([5, 2], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index_large.join(
+            other_mono, how="inner", return_indexers=True
+        )
+
+        res2 = index_large.intersection(other_mono)
+        tm.assert_index_equal(res, res2)
+
+        elidx = np.array([1, 4], dtype=np.intp)
+        eridx = np.array([3, 5], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_left(self, index_large):
+        other = Index(2**63 + np.array([7, 12, 25, 1, 2, 10], dtype="uint64"))
+        other_mono = Index(2**63 + np.array([1, 2, 7, 10, 12, 25], dtype="uint64"))
+
+        # not monotonic
+        res, lidx, ridx = index_large.join(other, how="left", return_indexers=True)
+        eres = index_large
+        eridx = np.array([-1, 5, -1, -1, 2], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index_large.join(other_mono, how="left", return_indexers=True)
+        eridx = np.array([-1, 3, -1, -1, 5], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # non-unique
+        idx = Index(2**63 + np.array([1, 1, 2, 5], dtype="uint64"))
+        idx2 = Index(2**63 + np.array([1, 2, 5, 7, 9], dtype="uint64"))
+        res, lidx, ridx = idx2.join(idx, how="left", return_indexers=True)
+
+        # 1 is in idx2, so it should be x2
+        eres = Index(2**63 + np.array([1, 1, 2, 5, 7, 9], dtype="uint64"))
+        eridx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+        elidx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_right(self, index_large):
+        other = Index(2**63 + np.array([7, 12, 25, 1, 2, 10], dtype="uint64"))
+        other_mono = Index(2**63 + np.array([1, 2, 7, 10, 12, 25], dtype="uint64"))
+
+        # not monotonic
+        res, lidx, ridx = index_large.join(other, how="right", return_indexers=True)
+        eres = other
+        elidx = np.array([-1, -1, 4, -1, -1, 1], dtype=np.intp)
+
+        tm.assert_numpy_array_equal(lidx, elidx)
+        assert isinstance(other, Index) and other.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        assert ridx is None
+
+        # monotonic
+        res, lidx, ridx = index_large.join(
+            other_mono, how="right", return_indexers=True
+        )
+        eres = other_mono
+        elidx = np.array([-1, -1, -1, 1, -1, 4], dtype=np.intp)
+
+        assert isinstance(other, Index) and other.dtype == np.uint64
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_index_equal(res, eres)
+        assert ridx is None
+
+        # non-unique
+        idx = Index(2**63 + np.array([1, 1, 2, 5], dtype="uint64"))
+        idx2 = Index(2**63 + np.array([1, 2, 5, 7, 9], dtype="uint64"))
+        res, lidx, ridx = idx.join(idx2, how="right", return_indexers=True)
+
+        # 1 is in idx2, so it should be x2
+        eres = Index(2**63 + np.array([1, 1, 2, 5, 7, 9], dtype="uint64"))
+        elidx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+        eridx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_non_int_index(self, index_large):
+        other = Index(2**63 + np.array([1, 5, 7, 10, 20], dtype="uint64"), dtype=object)
+
+        outer = index_large.join(other, how="outer")
+        outer2 = other.join(index_large, how="outer")
+        expected = Index(2**63 + np.array([0, 1, 5, 7, 10, 15, 20, 25], dtype="uint64"))
+        tm.assert_index_equal(outer, outer2)
+        tm.assert_index_equal(outer, expected)
+
+        inner = index_large.join(other, how="inner")
+        inner2 = other.join(index_large, how="inner")
+        expected = Index(2**63 + np.array([10, 20], dtype="uint64"))
+        tm.assert_index_equal(inner, inner2)
+        tm.assert_index_equal(inner, expected)
+
+        left = index_large.join(other, how="left")
+        tm.assert_index_equal(left, index_large.astype(object))
+
+        left2 = other.join(index_large, how="left")
+        tm.assert_index_equal(left2, other)
+
+        right = index_large.join(other, how="right")
+        tm.assert_index_equal(right, other)
+
+        right2 = other.join(index_large, how="right")
+        tm.assert_index_equal(right2, index_large.astype(object))
+
+    def test_join_outer(self, index_large):
+        other = Index(2**63 + np.array([7, 12, 25, 1, 2, 10], dtype="uint64"))
+        other_mono = Index(2**63 + np.array([1, 2, 7, 10, 12, 25], dtype="uint64"))
+
+        # not monotonic
+        # guarantee of sortedness
+        res, lidx, ridx = index_large.join(other, how="outer", return_indexers=True)
+        noidx_res = index_large.join(other, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        eres = Index(2**63 + np.array([0, 1, 2, 7, 10, 12, 15, 20, 25], dtype="uint64"))
+        elidx = np.array([0, -1, -1, -1, 1, -1, 2, 3, 4], dtype=np.intp)
+        eridx = np.array([-1, 3, 4, 0, 5, 1, -1, -1, 2], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # monotonic
+        res, lidx, ridx = index_large.join(
+            other_mono, how="outer", return_indexers=True
+        )
+        noidx_res = index_large.join(other_mono, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        elidx = np.array([0, -1, -1, -1, 1, -1, 2, 3, 4], dtype=np.intp)
+        eridx = np.array([-1, 0, 1, 2, 3, 4, -1, -1, 5], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.uint64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
diff --git a/pandas/pandas/tests/indexes/numeric/test_numeric.py b/pandas/pandas/tests/indexes/numeric/test_numeric.py
new file mode 100644
index 0000000000000..8dde5febe810d
--- /dev/null
+++ b/pandas/pandas/tests/indexes/numeric/test_numeric.py
@@ -0,0 +1,550 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestFloatNumericIndex:
+    @pytest.fixture(params=[np.float64, np.float32])
+    def dtype(self, request):
+        return request.param
+
+    @pytest.fixture
+    def mixed_index(self, dtype):
+        return Index([1.5, 2, 3, 4, 5], dtype=dtype)
+
+    @pytest.fixture
+    def float_index(self, dtype):
+        return Index([0.0, 2.5, 5.0, 7.5, 10.0], dtype=dtype)
+
+    @pytest.mark.parametrize(
+        "index_data",
+        [
+            [1.5, 2, 3, 4, 5],
+            [0.0, 2.5, 5.0, 7.5, 10.0],
+            [5, 4, 3, 2, 1.5],
+            [10.0, 7.5, 5.0, 2.5, 0.0],
+        ],
+        ids=["mixed", "float", "mixed_dec", "float_dec"],
+    )
+    def test_repr_roundtrip(self, index_data, dtype):
+        index = Index(index_data, dtype=dtype)
+        tm.assert_index_equal(eval(repr(index)), index, exact=True)
+
+    def check_coerce(self, a, b, is_float_index=True):
+        assert a.equals(b)
+        tm.assert_index_equal(a, b, exact=False)
+        if is_float_index:
+            assert isinstance(b, Index)
+        else:
+            assert type(b) is Index
+
+    def test_constructor_from_list_no_dtype(self):
+        index = Index([1.5, 2.5, 3.5])
+        assert index.dtype == np.float64
+
+    def test_constructor(self, dtype):
+        index_cls = Index
+
+        # explicit construction
+        index = index_cls([1, 2, 3, 4, 5], dtype=dtype)
+
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        expected = np.array([1, 2, 3, 4, 5], dtype=dtype)
+        tm.assert_numpy_array_equal(index.values, expected)
+
+        index = index_cls(np.array([1, 2, 3, 4, 5]), dtype=dtype)
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        index = index_cls([1.0, 2, 3, 4, 5], dtype=dtype)
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        index = index_cls(np.array([1.0, 2, 3, 4, 5]), dtype=dtype)
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        index = index_cls([1.0, 2, 3, 4, 5], dtype=dtype)
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        index = index_cls(np.array([1.0, 2, 3, 4, 5]), dtype=dtype)
+        assert isinstance(index, index_cls)
+        assert index.dtype == dtype
+
+        # nan handling
+        result = index_cls([np.nan, np.nan], dtype=dtype)
+        assert pd.isna(result.values).all()
+
+        result = index_cls(np.array([np.nan]), dtype=dtype)
+        assert pd.isna(result.values).all()
+
+    def test_constructor_invalid(self):
+        index_cls = Index
+        cls_name = index_cls.__name__
+        # invalid
+        msg = (
+            rf"{cls_name}\(\.\.\.\) must be called with a collection of "
+            r"some kind, 0\.0 was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            index_cls(0.0)
+
+    def test_constructor_coerce(self, mixed_index, float_index):
+        self.check_coerce(mixed_index, Index([1.5, 2, 3, 4, 5]))
+        self.check_coerce(float_index, Index(np.arange(5) * 2.5))
+
+        result = Index(np.array(np.arange(5) * 2.5, dtype=object))
+        assert result.dtype == object  # as of 2.0 to match Series
+        self.check_coerce(float_index, result.astype("float64"))
+
+    def test_constructor_explicit(self, mixed_index, float_index):
+        # these don't auto convert
+        self.check_coerce(
+            float_index, Index((np.arange(5) * 2.5), dtype=object), is_float_index=False
+        )
+        self.check_coerce(
+            mixed_index, Index([1.5, 2, 3, 4, 5], dtype=object), is_float_index=False
+        )
+
+    def test_type_coercion_fail(self, any_int_numpy_dtype):
+        # see gh-15832
+        msg = "Trying to coerce float values to integers"
+        with pytest.raises(ValueError, match=msg):
+            Index([1, 2, 3.5], dtype=any_int_numpy_dtype)
+
+    def test_equals_numeric(self):
+        index_cls = Index
+
+        idx = index_cls([1.0, 2.0])
+        assert idx.equals(idx)
+        assert idx.identical(idx)
+
+        idx2 = index_cls([1.0, 2.0])
+        assert idx.equals(idx2)
+
+        idx = index_cls([1.0, np.nan])
+        assert idx.equals(idx)
+        assert idx.identical(idx)
+
+        idx2 = index_cls([1.0, np.nan])
+        assert idx.equals(idx2)
+
+    @pytest.mark.parametrize(
+        "other",
+        (
+            Index([1, 2], dtype=np.int64),
+            Index([1.0, 2.0], dtype=object),
+            Index([1, 2], dtype=object),
+        ),
+    )
+    def test_equals_numeric_other_index_type(self, other):
+        idx = Index([1.0, 2.0])
+        assert idx.equals(other)
+        assert other.equals(idx)
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            pd.date_range("2016-01-01", periods=3),
+            pd.timedelta_range("1 Day", periods=3),
+        ],
+    )
+    def test_lookups_datetimelike_values(self, vals, dtype):
+        # If we have datetime64 or timedelta64 values, make sure they are
+        #  wrapped correctly  GH#31163
+        ser = Series(vals, index=range(3, 6))
+        ser.index = ser.index.astype(dtype)
+
+        expected = vals[1]
+
+        result = ser[4.0]
+        assert isinstance(result, type(expected)) and result == expected
+        result = ser[4]
+        assert isinstance(result, type(expected)) and result == expected
+
+        result = ser.loc[4.0]
+        assert isinstance(result, type(expected)) and result == expected
+        result = ser.loc[4]
+        assert isinstance(result, type(expected)) and result == expected
+
+        result = ser.at[4.0]
+        assert isinstance(result, type(expected)) and result == expected
+        # GH#31329 .at[4] should cast to 4.0, matching .loc behavior
+        result = ser.at[4]
+        assert isinstance(result, type(expected)) and result == expected
+
+        result = ser.iloc[1]
+        assert isinstance(result, type(expected)) and result == expected
+
+        result = ser.iat[1]
+        assert isinstance(result, type(expected)) and result == expected
+
+    def test_doesnt_contain_all_the_things(self):
+        idx = Index([np.nan])
+        assert not idx.isin([0]).item()
+        assert not idx.isin([1]).item()
+        assert idx.isin([np.nan]).item()
+
+    def test_nan_multiple_containment(self):
+        index_cls = Index
+
+        idx = index_cls([1.0, np.nan])
+        tm.assert_numpy_array_equal(idx.isin([1.0]), np.array([True, False]))
+        tm.assert_numpy_array_equal(idx.isin([2.0, np.pi]), np.array([False, False]))
+        tm.assert_numpy_array_equal(idx.isin([np.nan]), np.array([False, True]))
+        tm.assert_numpy_array_equal(idx.isin([1.0, np.nan]), np.array([True, True]))
+        idx = index_cls([1.0, 2.0])
+        tm.assert_numpy_array_equal(idx.isin([np.nan]), np.array([False, False]))
+
+    def test_fillna_float64(self):
+        index_cls = Index
+        # GH 11343
+        idx = Index([1.0, np.nan, 3.0], dtype=float, name="x")
+        # can't downcast
+        exp = Index([1.0, 0.1, 3.0], name="x")
+        tm.assert_index_equal(idx.fillna(0.1), exp, exact=True)
+
+        # downcast
+        exp = index_cls([1.0, 2.0, 3.0], name="x")
+        tm.assert_index_equal(idx.fillna(2), exp)
+
+        # object
+        exp = Index([1.0, "obj", 3.0], name="x")
+        tm.assert_index_equal(idx.fillna("obj"), exp, exact=True)
+
+    def test_logical_compat(self, dtype):
+        idx = Index(np.arange(5, dtype=dtype))
+        assert idx.all() == idx.values.all()
+        assert idx.any() == idx.values.any()
+
+        assert idx.all() == idx.to_series().all()
+        assert idx.any() == idx.to_series().any()
+
+
+class TestNumericInt:
+    @pytest.fixture
+    def dtype(self, any_int_numpy_dtype):
+        return np.dtype(any_int_numpy_dtype)
+
+    @pytest.fixture
+    def simple_index(self, dtype):
+        return Index(range(0, 20, 2), dtype=dtype)
+
+    def test_is_monotonic(self):
+        index_cls = Index
+
+        index = index_cls([1, 2, 3, 4])
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index._is_strictly_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is False
+        assert index._is_strictly_monotonic_decreasing is False
+
+        index = index_cls([4, 3, 2, 1])
+        assert index.is_monotonic_increasing is False
+        assert index._is_strictly_monotonic_increasing is False
+        assert index._is_strictly_monotonic_decreasing is True
+
+        index = index_cls([1])
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_increasing is True
+        assert index._is_strictly_monotonic_decreasing is True
+
+    def test_is_strictly_monotonic(self):
+        index_cls = Index
+
+        index = index_cls([1, 1, 2, 3])
+        assert index.is_monotonic_increasing is True
+        assert index._is_strictly_monotonic_increasing is False
+
+        index = index_cls([3, 2, 1, 1])
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_decreasing is False
+
+        index = index_cls([1, 1])
+        assert index.is_monotonic_increasing
+        assert index.is_monotonic_decreasing
+        assert not index._is_strictly_monotonic_increasing
+        assert not index._is_strictly_monotonic_decreasing
+
+    def test_logical_compat(self, simple_index):
+        idx = simple_index
+        assert idx.all() == idx.values.all()
+        assert idx.any() == idx.values.any()
+
+    def test_identical(self, simple_index, dtype):
+        index = simple_index
+
+        idx = Index(index.copy())
+        assert idx.identical(index)
+
+        same_values_different_type = Index(idx, dtype=object)
+        assert not idx.identical(same_values_different_type)
+
+        idx = index.astype(dtype=object)
+        idx = idx.rename("foo")
+        same_values = Index(idx, dtype=object)
+        assert same_values.identical(idx)
+
+        assert not idx.identical(index)
+        assert Index(same_values, name="foo", dtype=object).identical(idx)
+
+        assert not index.astype(dtype=object).identical(index.astype(dtype=dtype))
+
+    def test_cant_or_shouldnt_cast(self, dtype):
+        msg = r"invalid literal for int\(\) with base 10: 'foo'"
+
+        # can't
+        data = ["foo", "bar", "baz"]
+        with pytest.raises(ValueError, match=msg):
+            Index(data, dtype=dtype)
+
+    def test_view_index(self, simple_index):
+        index = simple_index
+        msg = (
+            "Cannot change data-type for array of references.|"
+            "Cannot change data-type for object array.|"
+        )
+        with pytest.raises(TypeError, match=msg):
+            index.view(Index)
+
+    def test_prevent_casting(self, simple_index):
+        index = simple_index
+        result = index.astype("O")
+        assert result.dtype == np.object_
+
+
+class TestIntNumericIndex:
+    @pytest.fixture(params=[np.int64, np.int32, np.int16, np.int8])
+    def dtype(self, request):
+        return request.param
+
+    def test_constructor_from_list_no_dtype(self):
+        index = Index([1, 2, 3])
+        assert index.dtype == np.int64
+
+    def test_constructor(self, dtype):
+        index_cls = Index
+
+        # scalar raise Exception
+        msg = (
+            rf"{index_cls.__name__}\(\.\.\.\) must be called with a collection of some "
+            "kind, 5 was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            index_cls(5)
+
+        # copy
+        # pass list, coerce fine
+        index = index_cls([-5, 0, 1, 2], dtype=dtype)
+        arr = index.values.copy()
+        new_index = index_cls(arr, copy=True)
+        tm.assert_index_equal(new_index, index, exact=True)
+        val = int(arr[0]) + 3000
+
+        # this should not change index
+        if dtype != np.int8:
+            # NEP 50 won't allow assignment that would overflow
+            arr[0] = val
+            assert new_index[0] != val
+
+        if dtype == np.int64:
+            # pass list, coerce fine
+            index = index_cls([-5, 0, 1, 2], dtype=dtype)
+            expected = Index([-5, 0, 1, 2], dtype=dtype)
+            tm.assert_index_equal(index, expected)
+
+            # from iterable
+            index = index_cls(iter([-5, 0, 1, 2]), dtype=dtype)
+            expected = index_cls([-5, 0, 1, 2], dtype=dtype)
+            tm.assert_index_equal(index, expected, exact=True)
+
+            # interpret list-like
+            expected = index_cls([5, 0], dtype=dtype)
+            for cls in [Index, index_cls]:
+                for idx in [
+                    cls([5, 0], dtype=dtype),
+                    cls(np.array([5, 0]), dtype=dtype),
+                    cls(Series([5, 0]), dtype=dtype),
+                ]:
+                    tm.assert_index_equal(idx, expected)
+
+    def test_constructor_corner(self, dtype):
+        index_cls = Index
+
+        arr = np.array([1, 2, 3, 4], dtype=object)
+
+        index = index_cls(arr, dtype=dtype)
+        assert index.values.dtype == index.dtype
+        if dtype == np.int64:
+            without_dtype = Index(arr)
+            # as of 2.0 we do not infer a dtype when we get an object-dtype
+            #  ndarray of numbers, matching Series behavior
+            assert without_dtype.dtype == object
+
+            tm.assert_index_equal(index, without_dtype.astype(np.int64))
+
+        # preventing casting
+        arr = np.array([1, "2", 3, "4"], dtype=object)
+        msg = "Trying to coerce object values to integers"
+        with pytest.raises(ValueError, match=msg):
+            index_cls(arr, dtype=dtype)
+
+    def test_constructor_coercion_signed_to_unsigned(
+        self,
+        any_unsigned_int_numpy_dtype,
+    ):
+        # see gh-15832
+        msg = "|".join(
+            [
+                "Trying to coerce negative values to unsigned integers",
+                "The elements provided in the data cannot all be casted",
+            ]
+        )
+        with pytest.raises(OverflowError, match=msg):
+            Index([-1], dtype=any_unsigned_int_numpy_dtype)
+
+    def test_constructor_np_signed(self, any_signed_int_numpy_dtype):
+        # GH#47475
+        scalar = np.dtype(any_signed_int_numpy_dtype).type(1)
+        result = Index([scalar])
+        expected = Index([1], dtype=any_signed_int_numpy_dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_constructor_np_unsigned(self, any_unsigned_int_numpy_dtype):
+        # GH#47475
+        scalar = np.dtype(any_unsigned_int_numpy_dtype).type(1)
+        result = Index([scalar])
+        expected = Index([1], dtype=any_unsigned_int_numpy_dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_coerce_list(self):
+        # coerce things
+        arr = Index([1, 2, 3, 4])
+        assert isinstance(arr, Index)
+
+        # but not if explicit dtype passed
+        arr = Index([1, 2, 3, 4], dtype=object)
+        assert type(arr) is Index
+
+
+class TestFloat16Index:
+    # float 16 indexes not supported
+    # GH 49535
+    def test_constructor(self):
+        index_cls = Index
+        dtype = np.float16
+
+        msg = "float16 indexes are not supported"
+
+        # explicit construction
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls([1, 2, 3, 4, 5], dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls(np.array([1, 2, 3, 4, 5]), dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls([1.0, 2, 3, 4, 5], dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls(np.array([1.0, 2, 3, 4, 5]), dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls([1.0, 2, 3, 4, 5], dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls(np.array([1.0, 2, 3, 4, 5]), dtype=dtype)
+
+        # nan handling
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls([np.nan, np.nan], dtype=dtype)
+
+        with pytest.raises(NotImplementedError, match=msg):
+            index_cls(np.array([np.nan]), dtype=dtype)
+
+
+@pytest.mark.parametrize(
+    "box",
+    [list, lambda x: np.array(x, dtype=object), lambda x: Index(x, dtype=object)],
+)
+def test_uint_index_does_not_convert_to_float64(box):
+    # https://github.com/pandas-dev/pandas/issues/28279
+    # https://github.com/pandas-dev/pandas/issues/28023
+    series = Series(
+        [0, 1, 2, 3, 4, 5],
+        index=[
+            7606741985629028552,
+            17876870360202815256,
+            17876870360202815256,
+            13106359306506049338,
+            8991270399732411471,
+            8991270399732411472,
+        ],
+    )
+
+    result = series.loc[box([7606741985629028552, 17876870360202815256])]
+
+    expected = Index(
+        [7606741985629028552, 17876870360202815256, 17876870360202815256],
+        dtype="uint64",
+    )
+    tm.assert_index_equal(result.index, expected)
+
+    tm.assert_equal(result, series.iloc[:3])
+
+
+def test_float64_index_equals():
+    # https://github.com/pandas-dev/pandas/issues/35217
+    float_index = Index([1.0, 2, 3])
+    string_index = Index(["1", "2", "3"])
+
+    result = float_index.equals(string_index)
+    assert result is False
+
+    result = string_index.equals(float_index)
+    assert result is False
+
+
+def test_map_dtype_inference_unsigned_to_signed():
+    # GH#44609 cases where we don't retain dtype
+    idx = Index([1, 2, 3], dtype=np.uint64)
+    result = idx.map(lambda x: -x)
+    expected = Index([-1, -2, -3], dtype=np.int64)
+    tm.assert_index_equal(result, expected)
+
+
+def test_map_dtype_inference_overflows():
+    # GH#44609 case where we have to upcast
+    idx = Index(np.array([1, 2, 3], dtype=np.int8))
+    result = idx.map(lambda x: x * 1000)
+    # TODO: we could plausibly try to infer down to int16 here
+    expected = Index([1000, 2000, 3000], dtype=np.int64)
+    tm.assert_index_equal(result, expected)
+
+
+def test_view_to_datetimelike():
+    # GH#55710
+    idx = Index([1, 2, 3])
+    res = idx.view("m8[s]")
+    expected = pd.TimedeltaIndex(idx.values.view("m8[s]"))
+    tm.assert_index_equal(res, expected)
+
+    res2 = idx.view("m8[D]")
+    expected2 = idx.values.view("m8[D]")
+    tm.assert_numpy_array_equal(res2, expected2)
+
+    res3 = idx.view("M8[h]")
+    expected3 = idx.values.view("M8[h]")
+    tm.assert_numpy_array_equal(res3, expected3)
diff --git a/pandas/pandas/tests/indexes/numeric/test_setops.py b/pandas/pandas/tests/indexes/numeric/test_setops.py
new file mode 100644
index 0000000000000..5d3981dbf93d0
--- /dev/null
+++ b/pandas/pandas/tests/indexes/numeric/test_setops.py
@@ -0,0 +1,173 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+import pandas._testing as tm
+from pandas.core.indexes.api import (
+    Index,
+    RangeIndex,
+)
+
+
+@pytest.fixture
+def index_large():
+    # large values used in TestUInt64Index where no compat needed with int64/float64
+    large = [2**63, 2**63 + 10, 2**63 + 15, 2**63 + 20, 2**63 + 25]
+    return Index(large, dtype=np.uint64)
+
+
+class TestSetOps:
+    @pytest.mark.parametrize("dtype", ["f8", "u8", "i8"])
+    def test_union_non_numeric(self, dtype):
+        # corner case, non-numeric
+        index = Index(np.arange(5, dtype=dtype), dtype=dtype)
+        assert index.dtype == dtype
+
+        other = Index([datetime.now() + timedelta(i) for i in range(4)], dtype=object)
+        result = index.union(other)
+        expected = Index(np.concatenate((index, other)))
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index)
+        expected = Index(np.concatenate((other, index)))
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection(self):
+        index = Index(range(5), dtype=np.int64)
+
+        other = Index([1, 2, 3, 4, 5])
+        result = index.intersection(other)
+        expected = Index(range(1, 5))
+        tm.assert_index_equal(result, expected)
+
+        result = other.intersection(index)
+        expected = Index(
+            np.sort(np.asarray(np.intersect1d(index.values, other.values)))
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["int64", "uint64"])
+    def test_int_float_union_dtype(self, dtype):
+        # https://github.com/pandas-dev/pandas/issues/26778
+        # [u]int | float -> float
+        index = Index([0, 2, 3], dtype=dtype)
+        other = Index([0.5, 1.5], dtype=np.float64)
+        expected = Index([0.0, 0.5, 1.5, 2.0, 3.0], dtype=np.float64)
+        result = index.union(other)
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index)
+        tm.assert_index_equal(result, expected)
+
+    def test_range_float_union_dtype(self):
+        # https://github.com/pandas-dev/pandas/issues/26778
+        index = RangeIndex(start=0, stop=3)
+        other = Index([0.5, 1.5], dtype=np.float64)
+        result = index.union(other)
+        expected = Index([0.0, 0.5, 1, 1.5, 2.0], dtype=np.float64)
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index)
+        tm.assert_index_equal(result, expected)
+
+    def test_range_uint64_union_dtype(self):
+        # https://github.com/pandas-dev/pandas/issues/26778
+        index = RangeIndex(start=0, stop=3)
+        other = Index([0, 10], dtype=np.uint64)
+        result = index.union(other)
+        expected = Index([0, 1, 2, 10], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index)
+        tm.assert_index_equal(result, expected)
+
+    def test_float64_index_difference(self):
+        # https://github.com/pandas-dev/pandas/issues/35217
+        float_index = Index([1.0, 2, 3])
+        string_index = Index(["1", "2", "3"])
+
+        result = float_index.difference(string_index)
+        tm.assert_index_equal(result, float_index)
+
+        result = string_index.difference(float_index)
+        tm.assert_index_equal(result, string_index)
+
+    def test_intersection_uint64_outside_int64_range(self):
+        index_large = Index(
+            [2**63, 2**63 + 10, 2**63 + 15, 2**63 + 20, 2**63 + 25],
+            dtype=np.uint64,
+        )
+        other = Index([2**63, 2**63 + 5, 2**63 + 10, 2**63 + 15, 2**63 + 20])
+        result = index_large.intersection(other)
+        expected = Index(np.sort(np.intersect1d(index_large.values, other.values)))
+        tm.assert_index_equal(result, expected)
+
+        result = other.intersection(index_large)
+        expected = Index(
+            np.sort(np.asarray(np.intersect1d(index_large.values, other.values)))
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index2_name,keeps_name",
+        [
+            ("index", True),
+            ("other", False),
+        ],
+    )
+    def test_intersection_monotonic(self, index2_name, keeps_name, sort):
+        index2 = Index([4, 7, 6, 5, 3], name=index2_name)
+        index1 = Index([5, 3, 2, 4, 1], name="index")
+        expected = Index([5, 3, 4])
+
+        if keeps_name:
+            expected.name = "index"
+
+        result = index1.intersection(index2, sort=sort)
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+    def test_symmetric_difference(self, sort):
+        # smoke
+        index1 = Index([5, 2, 3, 4], name="index1")
+        index2 = Index([2, 3, 4, 1])
+        result = index1.symmetric_difference(index2, sort=sort)
+        expected = Index([5, 1])
+        if sort is not None:
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result, expected.sort_values())
+        assert result.name is None
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+
+class TestSetOpsSort:
+    @pytest.mark.parametrize("slice_", [slice(None), slice(0)])
+    def test_union_sort_other_special(self, slice_):
+        # https://github.com/pandas-dev/pandas/issues/24959
+
+        idx = Index([1, 0, 2])
+        # default, sort=None
+        other = idx[slice_]
+        tm.assert_index_equal(idx.union(other), idx)
+        tm.assert_index_equal(other.union(idx), idx)
+
+        # sort=False
+        tm.assert_index_equal(idx.union(other, sort=False), idx)
+
+    @pytest.mark.parametrize("slice_", [slice(None), slice(0)])
+    def test_union_sort_special_true(self, slice_):
+        idx = Index([1, 0, 2])
+        # default, sort=None
+        other = idx[slice_]
+
+        result = idx.union(other, sort=True)
+        expected = Index([0, 1, 2])
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/object/__init__.py b/pandas/pandas/tests/indexes/object/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/object/test_astype.py b/pandas/pandas/tests/indexes/object/test_astype.py
new file mode 100644
index 0000000000000..7e0de138aacfb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/object/test_astype.py
@@ -0,0 +1,15 @@
+import pytest
+
+from pandas import (
+    Index,
+    NaT,
+)
+
+
+def test_astype_invalid_nas_to_tdt64_raises():
+    # GH#45722 don't cast np.datetime64 NaTs to timedelta64 NaT
+    idx = Index([NaT.asm8] * 2, dtype=object)
+
+    msg = r"Invalid type for timedelta scalar: <class 'numpy.datetime64'>"
+    with pytest.raises(TypeError, match=msg):
+        idx.astype("m8[ns]")
diff --git a/pandas/pandas/tests/indexes/object/test_indexing.py b/pandas/pandas/tests/indexes/object/test_indexing.py
new file mode 100644
index 0000000000000..2c5968314e5cf
--- /dev/null
+++ b/pandas/pandas/tests/indexes/object/test_indexing.py
@@ -0,0 +1,159 @@
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas._libs.missing import is_matching_na
+
+from pandas import Index
+import pandas._testing as tm
+
+
+class TestGetIndexer:
+    @pytest.mark.parametrize(
+        "method,expected",
+        [
+            ("pad", [-1, 0, 1, 1]),
+            ("backfill", [0, 0, 1, -1]),
+        ],
+    )
+    def test_get_indexer_strings(self, method, expected):
+        expected = np.array(expected, dtype=np.intp)
+        index = Index(["b", "c"], dtype=object)
+        actual = index.get_indexer(["a", "b", "c", "d"], method=method)
+
+        tm.assert_numpy_array_equal(actual, expected)
+
+    def test_get_indexer_strings_raises(self):
+        index = Index(["b", "c"], dtype=object)
+
+        msg = "|".join(
+            [
+                "operation 'sub' not supported for dtype 'str'",
+                r"unsupported operand type\(s\) for -: 'str' and 'str'",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(["a", "b", "c", "d"], method="nearest")
+
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(["a", "b", "c", "d"], method="pad", tolerance=2)
+
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(
+                ["a", "b", "c", "d"], method="pad", tolerance=[2, 2, 2, 2]
+            )
+
+    def test_get_indexer_with_NA_values(
+        self, unique_nulls_fixture, unique_nulls_fixture2
+    ):
+        # GH#22332
+        # check pairwise, that no pair of na values
+        # is mangled
+        if unique_nulls_fixture is unique_nulls_fixture2:
+            return  # skip it, values are not unique
+        arr = np.array([unique_nulls_fixture, unique_nulls_fixture2], dtype=object)
+        index = Index(arr, dtype=object)
+        result = index.get_indexer(
+            Index(
+                [unique_nulls_fixture, unique_nulls_fixture2, "Unknown"], dtype=object
+            )
+        )
+        expected = np.array([0, 1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_infer_string_missing_values(self):
+        # ensure the passed list is not cast to string but to object so that
+        # the None value is matched in the index
+        # https://github.com/pandas-dev/pandas/issues/55834
+        idx = Index(["a", "b", None], dtype="object")
+        result = idx.get_indexer([None, "x"])
+        expected = np.array([2, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestGetIndexerNonUnique:
+    def test_get_indexer_non_unique_nas(self, nulls_fixture):
+        # even though this isn't non-unique, this should still work
+        index = Index(["a", "b", nulls_fixture], dtype=object)
+        indexer, missing = index.get_indexer_non_unique([nulls_fixture])
+
+        expected_indexer = np.array([2], dtype=np.intp)
+        expected_missing = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected_indexer)
+        tm.assert_numpy_array_equal(missing, expected_missing)
+
+        # actually non-unique
+        index = Index(["a", nulls_fixture, "b", nulls_fixture], dtype=object)
+        indexer, missing = index.get_indexer_non_unique([nulls_fixture])
+
+        expected_indexer = np.array([1, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected_indexer)
+        tm.assert_numpy_array_equal(missing, expected_missing)
+
+        # matching-but-not-identical nans
+        if is_matching_na(nulls_fixture, float("NaN")):
+            index = Index(["a", float("NaN"), "b", float("NaN")], dtype=object)
+            match_but_not_identical = True
+        elif is_matching_na(nulls_fixture, Decimal("NaN")):
+            index = Index(["a", Decimal("NaN"), "b", Decimal("NaN")], dtype=object)
+            match_but_not_identical = True
+        else:
+            match_but_not_identical = False
+
+        if match_but_not_identical:
+            indexer, missing = index.get_indexer_non_unique([nulls_fixture])
+
+            expected_indexer = np.array([1, 3], dtype=np.intp)
+            tm.assert_numpy_array_equal(indexer, expected_indexer)
+            tm.assert_numpy_array_equal(missing, expected_missing)
+
+    @pytest.mark.filterwarnings("ignore:elementwise comp:DeprecationWarning")
+    def test_get_indexer_non_unique_np_nats(self, np_nat_fixture, np_nat_fixture2):
+        expected_missing = np.array([], dtype=np.intp)
+        # matching-but-not-identical nats
+        if is_matching_na(np_nat_fixture, np_nat_fixture2):
+            # ensure nats are different objects
+            index = Index(
+                np.array(
+                    ["2021-10-02", np_nat_fixture.copy(), np_nat_fixture2.copy()],
+                    dtype=object,
+                ),
+                dtype=object,
+            )
+            # pass as index to prevent target from being casted to DatetimeIndex
+            indexer, missing = index.get_indexer_non_unique(
+                Index([np_nat_fixture], dtype=object)
+            )
+            expected_indexer = np.array([1, 2], dtype=np.intp)
+            tm.assert_numpy_array_equal(indexer, expected_indexer)
+            tm.assert_numpy_array_equal(missing, expected_missing)
+        # dt64nat vs td64nat
+        else:
+            try:
+                np_nat_fixture == np_nat_fixture2
+            except (TypeError, OverflowError):
+                # Numpy will raise on uncomparable types, like
+                # np.datetime64('NaT', 'Y') and np.datetime64('NaT', 'ps')
+                # https://github.com/numpy/numpy/issues/22762
+                return
+            index = Index(
+                np.array(
+                    [
+                        "2021-10-02",
+                        np_nat_fixture,
+                        np_nat_fixture2,
+                        np_nat_fixture,
+                        np_nat_fixture2,
+                    ],
+                    dtype=object,
+                ),
+                dtype=object,
+            )
+            # pass as index to prevent target from being casted to DatetimeIndex
+            indexer, missing = index.get_indexer_non_unique(
+                Index([np_nat_fixture], dtype=object)
+            )
+            expected_indexer = np.array([1, 3], dtype=np.intp)
+            tm.assert_numpy_array_equal(indexer, expected_indexer)
+            tm.assert_numpy_array_equal(missing, expected_missing)
diff --git a/pandas/pandas/tests/indexes/period/__init__.py b/pandas/pandas/tests/indexes/period/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/period/methods/__init__.py b/pandas/pandas/tests/indexes/period/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/period/methods/test_asfreq.py b/pandas/pandas/tests/indexes/period/methods/test_asfreq.py
new file mode 100644
index 0000000000000..8fca53c28a036
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_asfreq.py
@@ -0,0 +1,182 @@
+import re
+
+import pytest
+
+from pandas import (
+    PeriodIndex,
+    Series,
+    period_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+class TestPeriodIndex:
+    def test_asfreq(self):
+        pi1 = period_range(freq="Y", start="1/1/2001", end="1/1/2001")
+        pi2 = period_range(freq="Q", start="1/1/2001", end="1/1/2001")
+        pi3 = period_range(freq="M", start="1/1/2001", end="1/1/2001")
+        pi4 = period_range(freq="D", start="1/1/2001", end="1/1/2001")
+        pi5 = period_range(freq="h", start="1/1/2001", end="1/1/2001 00:00")
+        pi6 = period_range(freq="Min", start="1/1/2001", end="1/1/2001 00:00")
+        pi7 = period_range(freq="s", start="1/1/2001", end="1/1/2001 00:00:00")
+
+        assert pi1.asfreq("Q", "s") == pi2
+        assert pi1.asfreq("Q", "s") == pi2
+        assert pi1.asfreq("M", "start") == pi3
+        assert pi1.asfreq("D", "StarT") == pi4
+        assert pi1.asfreq("h", "beGIN") == pi5
+        assert pi1.asfreq("Min", "s") == pi6
+        assert pi1.asfreq("s", "s") == pi7
+
+        assert pi2.asfreq("Y", "s") == pi1
+        assert pi2.asfreq("M", "s") == pi3
+        assert pi2.asfreq("D", "s") == pi4
+        assert pi2.asfreq("h", "s") == pi5
+        assert pi2.asfreq("Min", "s") == pi6
+        assert pi2.asfreq("s", "s") == pi7
+
+        assert pi3.asfreq("Y", "s") == pi1
+        assert pi3.asfreq("Q", "s") == pi2
+        assert pi3.asfreq("D", "s") == pi4
+        assert pi3.asfreq("h", "s") == pi5
+        assert pi3.asfreq("Min", "s") == pi6
+        assert pi3.asfreq("s", "s") == pi7
+
+        assert pi4.asfreq("Y", "s") == pi1
+        assert pi4.asfreq("Q", "s") == pi2
+        assert pi4.asfreq("M", "s") == pi3
+        assert pi4.asfreq("h", "s") == pi5
+        assert pi4.asfreq("Min", "s") == pi6
+        assert pi4.asfreq("s", "s") == pi7
+
+        assert pi5.asfreq("Y", "s") == pi1
+        assert pi5.asfreq("Q", "s") == pi2
+        assert pi5.asfreq("M", "s") == pi3
+        assert pi5.asfreq("D", "s") == pi4
+        assert pi5.asfreq("Min", "s") == pi6
+        assert pi5.asfreq("s", "s") == pi7
+
+        assert pi6.asfreq("Y", "s") == pi1
+        assert pi6.asfreq("Q", "s") == pi2
+        assert pi6.asfreq("M", "s") == pi3
+        assert pi6.asfreq("D", "s") == pi4
+        assert pi6.asfreq("h", "s") == pi5
+        assert pi6.asfreq("s", "s") == pi7
+
+        assert pi7.asfreq("Y", "s") == pi1
+        assert pi7.asfreq("Q", "s") == pi2
+        assert pi7.asfreq("M", "s") == pi3
+        assert pi7.asfreq("D", "s") == pi4
+        assert pi7.asfreq("h", "s") == pi5
+        assert pi7.asfreq("Min", "s") == pi6
+
+        msg = "How must be one of S or E"
+        with pytest.raises(ValueError, match=msg):
+            pi7.asfreq("T", "foo")
+        result1 = pi1.asfreq("3M")
+        result2 = pi1.asfreq("M")
+        expected = period_range(freq="M", start="2001-12", end="2001-12")
+        tm.assert_numpy_array_equal(result1.asi8, expected.asi8)
+        assert result1.freqstr == "3M"
+        tm.assert_numpy_array_equal(result2.asi8, expected.asi8)
+        assert result2.freqstr == "M"
+
+    def test_asfreq_nat(self):
+        idx = PeriodIndex(["2011-01", "2011-02", "NaT", "2011-04"], freq="M")
+        result = idx.asfreq(freq="Q")
+        expected = PeriodIndex(["2011Q1", "2011Q1", "NaT", "2011Q2"], freq="Q")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["D", "3D"])
+    def test_asfreq_mult_pi(self, freq):
+        pi = PeriodIndex(["2001-01", "2001-02", "NaT", "2001-03"], freq="2M")
+
+        result = pi.asfreq(freq)
+        exp = PeriodIndex(["2001-02-28", "2001-03-31", "NaT", "2001-04-30"], freq=freq)
+        tm.assert_index_equal(result, exp)
+        assert result.freq == exp.freq
+
+        result = pi.asfreq(freq, how="S")
+        exp = PeriodIndex(["2001-01-01", "2001-02-01", "NaT", "2001-03-01"], freq=freq)
+        tm.assert_index_equal(result, exp)
+        assert result.freq == exp.freq
+
+    def test_asfreq_combined_pi(self):
+        pi = PeriodIndex(["2001-01-01 00:00", "2001-01-02 02:00", "NaT"], freq="h")
+        exp = PeriodIndex(["2001-01-01 00:00", "2001-01-02 02:00", "NaT"], freq="25h")
+        for freq, how in zip(["1D1h", "1h1D"], ["S", "E"]):
+            result = pi.asfreq(freq, how=how)
+            tm.assert_index_equal(result, exp)
+            assert result.freq == exp.freq
+
+        for freq in ["1D1h", "1h1D"]:
+            pi = PeriodIndex(["2001-01-01 00:00", "2001-01-02 02:00", "NaT"], freq=freq)
+            result = pi.asfreq("h")
+            exp = PeriodIndex(["2001-01-02 00:00", "2001-01-03 02:00", "NaT"], freq="h")
+            tm.assert_index_equal(result, exp)
+            assert result.freq == exp.freq
+
+            pi = PeriodIndex(["2001-01-01 00:00", "2001-01-02 02:00", "NaT"], freq=freq)
+            result = pi.asfreq("h", how="S")
+            exp = PeriodIndex(["2001-01-01 00:00", "2001-01-02 02:00", "NaT"], freq="h")
+            tm.assert_index_equal(result, exp)
+            assert result.freq == exp.freq
+
+    def test_astype_asfreq(self):
+        pi1 = PeriodIndex(["2011-01-01", "2011-02-01", "2011-03-01"], freq="D")
+        exp = PeriodIndex(["2011-01", "2011-02", "2011-03"], freq="M")
+        tm.assert_index_equal(pi1.asfreq("M"), exp)
+        tm.assert_index_equal(pi1.astype("period[M]"), exp)
+
+        exp = PeriodIndex(["2011-01", "2011-02", "2011-03"], freq="3M")
+        tm.assert_index_equal(pi1.asfreq("3M"), exp)
+        tm.assert_index_equal(pi1.astype("period[3M]"), exp)
+
+    def test_asfreq_with_different_n(self):
+        ser = Series([1, 2], index=PeriodIndex(["2020-01", "2020-03"], freq="2M"))
+        result = ser.asfreq("M")
+
+        excepted = Series([1, 2], index=PeriodIndex(["2020-02", "2020-04"], freq="M"))
+        tm.assert_series_equal(result, excepted)
+
+    @pytest.mark.parametrize(
+        "freq",
+        [
+            "2BMS",
+            "2YS-MAR",
+            "2bh",
+            offsets.MonthBegin(2),
+            offsets.BusinessMonthEnd(2),
+        ],
+    )
+    def test_pi_asfreq_not_supported_frequency(self, freq):
+        # GH#55785, GH#56945
+        msg = "|".join(
+            [
+                f"Invalid frequency: {freq}",
+                re.escape(f"{freq} is not supported as period frequency"),
+                "bh is not supported as period frequency",
+            ]
+        )
+
+        pi = PeriodIndex(["2020-01-01", "2021-01-01"], freq="M")
+        with pytest.raises(ValueError, match=msg):
+            pi.asfreq(freq=freq)
+
+    @pytest.mark.parametrize(
+        "freq",
+        [
+            "2BME",
+            "2YE-MAR",
+            "2QE",
+        ],
+    )
+    def test_pi_asfreq_invalid_frequency(self, freq):
+        # GH#55785
+        msg = f"Invalid frequency: {freq}"
+
+        pi = PeriodIndex(["2020-01-01", "2021-01-01"], freq="M")
+        with pytest.raises(ValueError, match=msg):
+            pi.asfreq(freq=freq)
diff --git a/pandas/pandas/tests/indexes/period/methods/test_astype.py b/pandas/pandas/tests/indexes/period/methods/test_astype.py
new file mode 100644
index 0000000000000..af3c2667f51b4
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_astype.py
@@ -0,0 +1,156 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    NaT,
+    Period,
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodIndexAsType:
+    @pytest.mark.parametrize("dtype", [float, "timedelta64", "timedelta64[ns]"])
+    def test_astype_raises(self, dtype):
+        # GH#13149, GH#13209
+        idx = PeriodIndex(["2016-05-16", "NaT", NaT, np.nan], freq="D")
+        msg = "Cannot cast PeriodIndex to dtype"
+        with pytest.raises(TypeError, match=msg):
+            idx.astype(dtype)
+
+    def test_astype_conversion(self, using_infer_string):
+        # GH#13149, GH#13209
+        idx = PeriodIndex(["2016-05-16", "NaT", NaT, np.nan], freq="D", name="idx")
+
+        result = idx.astype(object)
+        expected = Index(
+            [Period("2016-05-16", freq="D")] + [Period(NaT, freq="D")] * 3,
+            dtype="object",
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.astype(np.int64)
+        expected = Index(
+            [16937] + [-9223372036854775808] * 3, dtype=np.int64, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.astype(str)
+        if using_infer_string:
+            expected = Index(
+                [str(x) if x is not NaT else None for x in idx], name="idx", dtype="str"
+            )
+        else:
+            expected = Index([str(x) for x in idx], name="idx", dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        idx = period_range("1990", "2009", freq="Y", name="idx")
+        result = idx.astype("i8")
+        tm.assert_index_equal(result, Index(idx.asi8, name="idx"))
+        tm.assert_numpy_array_equal(result.values, idx.asi8)
+
+    def test_astype_uint(self):
+        arr = period_range("2000", periods=2, name="idx")
+
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint64")
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint32")
+
+    def test_astype_object(self):
+        idx = PeriodIndex([], freq="M")
+
+        exp = np.array([], dtype=object)
+        tm.assert_numpy_array_equal(idx.astype(object).values, exp)
+        tm.assert_numpy_array_equal(idx._mpl_repr(), exp)
+
+        idx = PeriodIndex(["2011-01", NaT], freq="M")
+
+        exp = np.array([Period("2011-01", freq="M"), NaT], dtype=object)
+        tm.assert_numpy_array_equal(idx.astype(object).values, exp)
+        tm.assert_numpy_array_equal(idx._mpl_repr(), exp)
+
+        exp = np.array([Period("2011-01-01", freq="D"), NaT], dtype=object)
+        idx = PeriodIndex(["2011-01-01", NaT], freq="D")
+        tm.assert_numpy_array_equal(idx.astype(object).values, exp)
+        tm.assert_numpy_array_equal(idx._mpl_repr(), exp)
+
+    # TODO: de-duplicate this version (from test_ops) with the one above
+    # (from test_period)
+    def test_astype_object2(self):
+        idx = period_range(start="2013-01-01", periods=4, freq="M", name="idx")
+        expected_list = [
+            Period("2013-01-31", freq="M"),
+            Period("2013-02-28", freq="M"),
+            Period("2013-03-31", freq="M"),
+            Period("2013-04-30", freq="M"),
+        ]
+        expected = Index(expected_list, dtype=object, name="idx")
+        result = idx.astype(object)
+        assert isinstance(result, Index)
+        assert result.dtype == object
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert idx.tolist() == expected_list
+
+        idx = PeriodIndex(
+            ["2013-01-01", "2013-01-02", "NaT", "2013-01-04"], freq="D", name="idx"
+        )
+        expected_list = [
+            Period("2013-01-01", freq="D"),
+            Period("2013-01-02", freq="D"),
+            Period("NaT", freq="D"),
+            Period("2013-01-04", freq="D"),
+        ]
+        expected = Index(expected_list, dtype=object, name="idx")
+        result = idx.astype(object)
+        assert isinstance(result, Index)
+        assert result.dtype == object
+        tm.assert_index_equal(result, expected)
+        for i in [0, 1, 3]:
+            assert result[i] == expected[i]
+        assert result[2] is NaT
+        assert result.name == expected.name
+
+        result_list = idx.tolist()
+        for i in [0, 1, 3]:
+            assert result_list[i] == expected_list[i]
+        assert result_list[2] is NaT
+
+    def test_astype_category(self):
+        obj = period_range("2000", periods=2, name="idx")
+        result = obj.astype("category")
+        expected = CategoricalIndex(
+            [Period("2000-01-01", freq="D"), Period("2000-01-02", freq="D")], name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype("category")
+        expected = expected.values
+        tm.assert_categorical_equal(result, expected)
+
+    def test_astype_array_fallback(self):
+        obj = period_range("2000", periods=2, name="idx")
+        result = obj.astype(bool)
+        expected = Index(np.array([True, True]), name="idx")
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype(bool)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_period_astype_to_timestamp(self, unit):
+        # GH#55958
+        pi = PeriodIndex(["2011-01", "2011-02", "2011-03"], freq="M")
+
+        exp = DatetimeIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01"], tz="US/Eastern"
+        ).as_unit(unit)
+        res = pi.astype(f"datetime64[{unit}, US/Eastern]")
+        tm.assert_index_equal(res, exp)
+        assert res.freq == exp.freq
diff --git a/pandas/pandas/tests/indexes/period/methods/test_factorize.py b/pandas/pandas/tests/indexes/period/methods/test_factorize.py
new file mode 100644
index 0000000000000..1239eae6091b8
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_factorize.py
@@ -0,0 +1,41 @@
+import numpy as np
+
+from pandas import PeriodIndex
+import pandas._testing as tm
+
+
+class TestFactorize:
+    def test_factorize_period(self):
+        idx1 = PeriodIndex(
+            ["2014-01", "2014-01", "2014-02", "2014-02", "2014-03", "2014-03"],
+            freq="M",
+        )
+
+        exp_arr = np.array([0, 0, 1, 1, 2, 2], dtype=np.intp)
+        exp_idx = PeriodIndex(["2014-01", "2014-02", "2014-03"], freq="M")
+
+        arr, idx = idx1.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+
+        arr, idx = idx1.factorize(sort=True)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+
+    def test_factorize_period_nonmonotonic(self):
+        idx2 = PeriodIndex(
+            ["2014-03", "2014-03", "2014-02", "2014-01", "2014-03", "2014-01"],
+            freq="M",
+        )
+        exp_idx = PeriodIndex(["2014-01", "2014-02", "2014-03"], freq="M")
+
+        exp_arr = np.array([2, 2, 1, 0, 2, 0], dtype=np.intp)
+        arr, idx = idx2.factorize(sort=True)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+
+        exp_arr = np.array([0, 0, 1, 2, 0, 2], dtype=np.intp)
+        exp_idx = PeriodIndex(["2014-03", "2014-02", "2014-01"], freq="M")
+        arr, idx = idx2.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
diff --git a/pandas/pandas/tests/indexes/period/methods/test_fillna.py b/pandas/pandas/tests/indexes/period/methods/test_fillna.py
new file mode 100644
index 0000000000000..ed6b4686a06de
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_fillna.py
@@ -0,0 +1,41 @@
+from pandas import (
+    Index,
+    NaT,
+    Period,
+    PeriodIndex,
+)
+import pandas._testing as tm
+
+
+class TestFillNA:
+    def test_fillna_period(self):
+        # GH#11343
+        idx = PeriodIndex(["2011-01-01 09:00", NaT, "2011-01-01 11:00"], freq="h")
+
+        exp = PeriodIndex(
+            ["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"], freq="h"
+        )
+        result = idx.fillna(Period("2011-01-01 10:00", freq="h"))
+        tm.assert_index_equal(result, exp)
+
+        exp = Index(
+            [
+                Period("2011-01-01 09:00", freq="h"),
+                "x",
+                Period("2011-01-01 11:00", freq="h"),
+            ],
+            dtype=object,
+        )
+        result = idx.fillna("x")
+        tm.assert_index_equal(result, exp)
+
+        exp = Index(
+            [
+                Period("2011-01-01 09:00", freq="h"),
+                Period("2011-01-01", freq="D"),
+                Period("2011-01-01 11:00", freq="h"),
+            ],
+            dtype=object,
+        )
+        result = idx.fillna(Period("2011-01-01", freq="D"))
+        tm.assert_index_equal(result, exp)
diff --git a/pandas/pandas/tests/indexes/period/methods/test_insert.py b/pandas/pandas/tests/indexes/period/methods/test_insert.py
new file mode 100644
index 0000000000000..32bbe09d92567
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_insert.py
@@ -0,0 +1,18 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NaT,
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestInsert:
+    @pytest.mark.parametrize("na", [np.nan, NaT, None])
+    def test_insert(self, na):
+        # GH#18295 (test missing)
+        expected = PeriodIndex(["2017Q1", NaT, "2017Q2", "2017Q3", "2017Q4"], freq="Q")
+        result = period_range("2017Q1", periods=4, freq="Q").insert(1, na)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/period/methods/test_is_full.py b/pandas/pandas/tests/indexes/period/methods/test_is_full.py
new file mode 100644
index 0000000000000..b4105bedbe21d
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_is_full.py
@@ -0,0 +1,23 @@
+import pytest
+
+from pandas import PeriodIndex
+
+
+def test_is_full():
+    index = PeriodIndex([2005, 2007, 2009], freq="Y")
+    assert not index.is_full
+
+    index = PeriodIndex([2005, 2006, 2007], freq="Y")
+    assert index.is_full
+
+    index = PeriodIndex([2005, 2005, 2007], freq="Y")
+    assert not index.is_full
+
+    index = PeriodIndex([2005, 2005, 2006], freq="Y")
+    assert index.is_full
+
+    index = PeriodIndex([2006, 2005, 2005], freq="Y")
+    with pytest.raises(ValueError, match="Index is not monotonic"):
+        index.is_full
+
+    assert index[:0].is_full
diff --git a/pandas/pandas/tests/indexes/period/methods/test_repeat.py b/pandas/pandas/tests/indexes/period/methods/test_repeat.py
new file mode 100644
index 0000000000000..fc344b06420d1
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_repeat.py
@@ -0,0 +1,26 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestRepeat:
+    @pytest.mark.parametrize("use_numpy", [True, False])
+    @pytest.mark.parametrize(
+        "index",
+        [
+            period_range("2000-01-01", periods=3, freq="D"),
+            period_range("2001-01-01", periods=3, freq="2D"),
+            PeriodIndex(["2001-01", "NaT", "2003-01"], freq="M"),
+        ],
+    )
+    def test_repeat_freqstr(self, index, use_numpy):
+        # GH#10183
+        expected = PeriodIndex([per for per in index for _ in range(3)])
+        result = np.repeat(index, 3) if use_numpy else index.repeat(3)
+        tm.assert_index_equal(result, expected)
+        assert result.freqstr == index.freqstr
diff --git a/pandas/pandas/tests/indexes/period/methods/test_shift.py b/pandas/pandas/tests/indexes/period/methods/test_shift.py
new file mode 100644
index 0000000000000..fca3e3a559e1f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_shift.py
@@ -0,0 +1,122 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodIndexShift:
+    # ---------------------------------------------------------------
+    # PeriodIndex.shift is used by __add__ and __sub__
+
+    def test_pi_shift_ndarray(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="M", name="idx"
+        )
+        result = idx.shift(np.array([1, 2, 3, 4]))
+        expected = PeriodIndex(
+            ["2011-02", "2011-04", "NaT", "2011-08"], freq="M", name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.shift(np.array([1, -2, 3, -4]))
+        expected = PeriodIndex(
+            ["2011-02", "2010-12", "NaT", "2010-12"], freq="M", name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_shift(self):
+        pi1 = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="Y", start="1/1/2002", end="12/1/2010")
+
+        tm.assert_index_equal(pi1.shift(0), pi1)
+
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(1), pi2)
+
+        pi1 = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="Y", start="1/1/2000", end="12/1/2008")
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(-1), pi2)
+
+        pi1 = period_range(freq="M", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="M", start="2/1/2001", end="1/1/2010")
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(1), pi2)
+
+        pi1 = period_range(freq="M", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="M", start="12/1/2000", end="11/1/2009")
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(-1), pi2)
+
+        pi1 = period_range(freq="D", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="D", start="1/2/2001", end="12/2/2009")
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(1), pi2)
+
+        pi1 = period_range(freq="D", start="1/1/2001", end="12/1/2009")
+        pi2 = period_range(freq="D", start="12/31/2000", end="11/30/2009")
+        assert len(pi1) == len(pi2)
+        tm.assert_index_equal(pi1.shift(-1), pi2)
+
+    def test_shift_corner_cases(self):
+        # GH#9903
+        idx = PeriodIndex([], name="xxx", freq="h")
+
+        msg = "`freq` argument is not supported for PeriodIndex.shift"
+        with pytest.raises(TypeError, match=msg):
+            # period shift doesn't accept freq
+            idx.shift(1, freq="h")
+
+        tm.assert_index_equal(idx.shift(0), idx)
+        tm.assert_index_equal(idx.shift(3), idx)
+
+        idx = PeriodIndex(
+            ["2011-01-01 10:00", "2011-01-01 11:00", "2011-01-01 12:00"],
+            name="xxx",
+            freq="h",
+        )
+        tm.assert_index_equal(idx.shift(0), idx)
+        exp = PeriodIndex(
+            ["2011-01-01 13:00", "2011-01-01 14:00", "2011-01-01 15:00"],
+            name="xxx",
+            freq="h",
+        )
+        tm.assert_index_equal(idx.shift(3), exp)
+        exp = PeriodIndex(
+            ["2011-01-01 07:00", "2011-01-01 08:00", "2011-01-01 09:00"],
+            name="xxx",
+            freq="h",
+        )
+        tm.assert_index_equal(idx.shift(-3), exp)
+
+    def test_shift_nat(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2011-04"], freq="M", name="idx"
+        )
+        result = idx.shift(1)
+        expected = PeriodIndex(
+            ["2011-02", "2011-03", "NaT", "2011-05"], freq="M", name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+
+    def test_shift_gh8083(self):
+        # test shift for PeriodIndex
+        # GH#8083
+        drange = period_range("20130101", periods=5, freq="D")
+        result = drange.shift(1)
+        expected = PeriodIndex(
+            ["2013-01-02", "2013-01-03", "2013-01-04", "2013-01-05", "2013-01-06"],
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_shift_periods(self):
+        # GH #22458 : argument 'n' was deprecated in favor of 'periods'
+        idx = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        tm.assert_index_equal(idx.shift(periods=0), idx)
+        tm.assert_index_equal(idx.shift(0), idx)
diff --git a/pandas/pandas/tests/indexes/period/methods/test_to_timestamp.py b/pandas/pandas/tests/indexes/period/methods/test_to_timestamp.py
new file mode 100644
index 0000000000000..4fe429ce71ee4
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/methods/test_to_timestamp.py
@@ -0,0 +1,149 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    PeriodIndex,
+    Timedelta,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestToTimestamp:
+    def test_to_timestamp_non_contiguous(self):
+        # GH#44100
+        dti = date_range("2021-10-18", periods=9, freq="D")
+        pi = dti.to_period()
+
+        result = pi[::2].to_timestamp()
+        expected = dti[::2]
+        tm.assert_index_equal(result, expected)
+
+        result = pi._data[::2].to_timestamp()
+        expected = dti._data[::2]
+        # TODO: can we get the freq to round-trip?
+        tm.assert_datetime_array_equal(result, expected, check_freq=False)
+
+        result = pi[::-1].to_timestamp()
+        expected = dti[::-1]
+        tm.assert_index_equal(result, expected)
+
+        result = pi._data[::-1].to_timestamp()
+        expected = dti._data[::-1]
+        tm.assert_datetime_array_equal(result, expected, check_freq=False)
+
+        result = pi[::2][::-1].to_timestamp()
+        expected = dti[::2][::-1]
+        tm.assert_index_equal(result, expected)
+
+        result = pi._data[::2][::-1].to_timestamp()
+        expected = dti._data[::2][::-1]
+        tm.assert_datetime_array_equal(result, expected, check_freq=False)
+
+    def test_to_timestamp_freq(self):
+        idx = period_range("2017", periods=12, freq="Y-DEC")
+        result = idx.to_timestamp()
+        expected = date_range("2017", periods=12, freq="YS-JAN")
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timestamp_pi_nat(self):
+        # GH#7228
+        index = PeriodIndex(["NaT", "2011-01", "2011-02"], freq="M", name="idx")
+
+        result = index.to_timestamp("D")
+        expected = DatetimeIndex(
+            [NaT, datetime(2011, 1, 1), datetime(2011, 2, 1)],
+            dtype="M8[ns]",
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+        assert result.name == "idx"
+
+        result2 = result.to_period(freq="M")
+        tm.assert_index_equal(result2, index)
+        assert result2.name == "idx"
+
+        result3 = result.to_period(freq="3M")
+        exp = PeriodIndex(["NaT", "2011-01", "2011-02"], freq="3M", name="idx")
+        tm.assert_index_equal(result3, exp)
+        assert result3.freqstr == "3M"
+
+        msg = "Frequency must be positive, because it represents span: -2Y"
+        with pytest.raises(ValueError, match=msg):
+            result.to_period(freq="-2Y")
+
+    def test_to_timestamp_preserve_name(self):
+        index = period_range(freq="Y", start="1/1/2001", end="12/1/2009", name="foo")
+        assert index.name == "foo"
+
+        conv = index.to_timestamp("D")
+        assert conv.name == "foo"
+
+    def test_to_timestamp_quarterly_bug(self):
+        years = np.arange(1960, 2000).repeat(4)
+        quarters = np.tile(list(range(1, 5)), 40)
+
+        pindex = PeriodIndex.from_fields(year=years, quarter=quarters)
+
+        stamps = pindex.to_timestamp("D", "end")
+        expected = DatetimeIndex([x.to_timestamp("D", "end") for x in pindex])
+        tm.assert_index_equal(stamps, expected)
+        assert stamps.freq == expected.freq
+
+    def test_to_timestamp_pi_mult(self):
+        idx = PeriodIndex(["2011-01", "NaT", "2011-02"], freq="2M", name="idx")
+
+        result = idx.to_timestamp()
+        expected = DatetimeIndex(
+            ["2011-01-01", "NaT", "2011-02-01"], dtype="M8[ns]", name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.to_timestamp(how="E")
+        expected = DatetimeIndex(
+            ["2011-02-28", "NaT", "2011-03-31"], dtype="M8[ns]", name="idx"
+        )
+        expected = expected + Timedelta(1, "D") - Timedelta(1, "ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timestamp_pi_combined(self):
+        idx = period_range(start="2011", periods=2, freq="1D1h", name="idx")
+
+        result = idx.to_timestamp()
+        expected = DatetimeIndex(
+            ["2011-01-01 00:00", "2011-01-02 01:00"], dtype="M8[ns]", name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.to_timestamp(how="E")
+        expected = DatetimeIndex(
+            ["2011-01-02 00:59:59", "2011-01-03 01:59:59"], name="idx", dtype="M8[ns]"
+        )
+        expected = expected + Timedelta(1, "s") - Timedelta(1, "ns")
+        tm.assert_index_equal(result, expected)
+
+        result = idx.to_timestamp(how="E", freq="h")
+        expected = DatetimeIndex(
+            ["2011-01-02 00:00", "2011-01-03 01:00"], dtype="M8[ns]", name="idx"
+        )
+        expected = expected + Timedelta(1, "h") - Timedelta(1, "ns")
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timestamp_1703(self):
+        index = period_range("1/1/2012", periods=4, freq="D")
+
+        result = index.to_timestamp()
+        assert result[0] == Timestamp("1/1/2012")
+
+
+def test_ms_to_timestamp_error_message():
+    # https://github.com/pandas-dev/pandas/issues/58974#issuecomment-2164265446
+    ix = period_range("2000", periods=3, freq="M")
+    with pytest.raises(ValueError, match="for Period, please use 'M' instead of 'MS'"):
+        ix.to_timestamp("MS")
diff --git a/pandas/pandas/tests/indexes/period/test_constructors.py b/pandas/pandas/tests/indexes/period/test_constructors.py
new file mode 100644
index 0000000000000..be07a71b283fd
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_constructors.py
@@ -0,0 +1,677 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.period import IncompatibleFrequency
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+from pandas import (
+    Index,
+    NaT,
+    Period,
+    PeriodIndex,
+    Series,
+    date_range,
+    offsets,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import PeriodArray
+
+
+class TestPeriodIndexDisallowedFreqs:
+    @pytest.mark.parametrize(
+        "freq,freq_depr",
+        [
+            ("2M", "2ME"),
+            ("2Q-MAR", "2QE-MAR"),
+            ("2Y-FEB", "2YE-FEB"),
+            ("2M", "2me"),
+            ("2Q-MAR", "2qe-MAR"),
+            ("2Y-FEB", "2yE-feb"),
+        ],
+    )
+    def test_period_index_offsets_frequency_error_message(self, freq, freq_depr):
+        # GH#52064
+        msg = f"Invalid frequency: {freq_depr}"
+
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(["2020-01-01", "2020-01-02"], freq=freq_depr)
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(start="2020-01-01", end="2020-01-02", freq=freq_depr)
+
+    @pytest.mark.parametrize(
+        "freq",
+        ["2SME", "2sme", "2BYE", "2Bye", "2CBME"],
+    )
+    def test_period_index_frequency_invalid_freq(self, freq):
+        # GH#9586
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            period_range("2020-01", "2020-05", freq=freq)
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(["2020-01", "2020-05"], freq=freq)
+
+    @pytest.mark.parametrize("freq", ["2BQE-SEP", "2BYE-MAR", "2BME"])
+    def test_period_index_from_datetime_index_invalid_freq(self, freq):
+        # GH#56899
+        msg = f"Invalid frequency: {freq}"
+
+        rng = date_range("01-Jan-2012", periods=8, freq=freq)
+        with pytest.raises(ValueError, match=msg):
+            rng.to_period()
+
+    @pytest.mark.parametrize("freq_depr", ["2T", "1l", "2U", "n"])
+    def test_period_index_T_L_U_N_raises(self, freq_depr):
+        # GH#9586
+        msg = f"Invalid frequency: {freq_depr}"
+
+        with pytest.raises(ValueError, match=msg):
+            period_range("2020-01", "2020-05", freq=freq_depr)
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(["2020-01", "2020-05"], freq=freq_depr)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    @pytest.mark.filterwarnings("ignore:Period with BDay freq:FutureWarning")
+    @pytest.mark.parametrize(
+        "freq,freq_depr",
+        [("2W", "2w"), ("2W-FRI", "2w-fri"), ("2D", "2d"), ("2B", "2b")],
+    )
+    def test_period_index_depr_lowercase_frequency(self, freq, freq_depr):
+        # GH#58998
+        msg = (
+            f"'{freq_depr[1:]}' is deprecated and will be removed in a future version."
+        )
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = PeriodIndex(["2020-01-01", "2020-01-02"], freq=freq_depr)
+
+        expected = PeriodIndex(["2020-01-01", "2020-01-02"], freq=freq)
+        tm.assert_index_equal(result, expected)
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = period_range(start="2020-01-01", end="2020-01-02", freq=freq_depr)
+
+        expected = period_range(start="2020-01-01", end="2020-01-02", freq=freq)
+        tm.assert_index_equal(result, expected)
+
+
+class TestPeriodIndex:
+    def test_from_ordinals(self):
+        Period(ordinal=-1000, freq="Y")
+        Period(ordinal=0, freq="Y")
+
+        idx1 = PeriodIndex.from_ordinals(ordinals=[-1, 0, 1], freq="Y")
+        idx2 = PeriodIndex.from_ordinals(ordinals=np.array([-1, 0, 1]), freq="Y")
+        tm.assert_index_equal(idx1, idx2)
+
+    def test_construction_base_constructor(self):
+        # GH 13664
+        arr = [Period("2011-01", freq="M"), NaT, Period("2011-03", freq="M")]
+        tm.assert_index_equal(Index(arr), PeriodIndex(arr))
+        tm.assert_index_equal(Index(np.array(arr)), PeriodIndex(np.array(arr)))
+
+        arr = [np.nan, NaT, Period("2011-03", freq="M")]
+        tm.assert_index_equal(Index(arr), PeriodIndex(arr))
+        tm.assert_index_equal(Index(np.array(arr)), PeriodIndex(np.array(arr)))
+
+        arr = [Period("2011-01", freq="M"), NaT, Period("2011-03", freq="D")]
+        tm.assert_index_equal(Index(arr), Index(arr, dtype=object))
+
+        tm.assert_index_equal(Index(np.array(arr)), Index(np.array(arr), dtype=object))
+
+    def test_base_constructor_with_period_dtype(self):
+        dtype = PeriodDtype("D")
+        values = ["2011-01-01", "2012-03-04", "2014-05-01"]
+        result = Index(values, dtype=dtype)
+
+        expected = PeriodIndex(values, dtype=dtype)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values_constructor", [list, np.array, PeriodIndex, PeriodArray._from_sequence]
+    )
+    def test_index_object_dtype(self, values_constructor):
+        # Index(periods, dtype=object) is an Index (not an PeriodIndex)
+        periods = [
+            Period("2011-01", freq="M"),
+            NaT,
+            Period("2011-03", freq="M"),
+        ]
+        values = values_constructor(periods)
+        result = Index(values, dtype=object)
+
+        assert type(result) is Index
+        tm.assert_numpy_array_equal(result.values, np.array(values))
+
+    def test_constructor_use_start_freq(self):
+        # GH #1118
+        msg1 = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg1):
+            p = Period("4/2/2012", freq="B")
+        msg2 = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg2):
+            expected = period_range(start="4/2/2012", periods=10, freq="B")
+
+        with tm.assert_produces_warning(FutureWarning, match=msg2):
+            index = period_range(start=p, periods=10)
+        tm.assert_index_equal(index, expected)
+
+    def test_constructor_field_arrays(self):
+        # GH #1264
+
+        years = np.arange(1990, 2010).repeat(4)[2:-2]
+        quarters = np.tile(np.arange(1, 5), 20)[2:-2]
+
+        index = PeriodIndex.from_fields(year=years, quarter=quarters, freq="Q-DEC")
+        expected = period_range("1990Q3", "2009Q2", freq="Q-DEC")
+        tm.assert_index_equal(index, expected)
+
+        index2 = PeriodIndex.from_fields(year=years, quarter=quarters, freq="2Q-DEC")
+        tm.assert_numpy_array_equal(index.asi8, index2.asi8)
+
+        index = PeriodIndex.from_fields(year=years, quarter=quarters)
+        tm.assert_index_equal(index, expected)
+
+        years = [2007, 2007, 2007]
+        months = [1, 2]
+
+        msg = "Mismatched Period array lengths"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex.from_fields(year=years, month=months, freq="M")
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex.from_fields(year=years, month=months, freq="2M")
+
+        years = [2007, 2007, 2007]
+        months = [1, 2, 3]
+        idx = PeriodIndex.from_fields(year=years, month=months, freq="M")
+        exp = period_range("2007-01", periods=3, freq="M")
+        tm.assert_index_equal(idx, exp)
+
+    def test_constructor_nano(self):
+        idx = period_range(
+            start=Period(ordinal=1, freq="ns"),
+            end=Period(ordinal=4, freq="ns"),
+            freq="ns",
+        )
+        exp = PeriodIndex(
+            [
+                Period(ordinal=1, freq="ns"),
+                Period(ordinal=2, freq="ns"),
+                Period(ordinal=3, freq="ns"),
+                Period(ordinal=4, freq="ns"),
+            ],
+            freq="ns",
+        )
+        tm.assert_index_equal(idx, exp)
+
+    def test_constructor_arrays_negative_year(self):
+        years = np.arange(1960, 2000, dtype=np.int64).repeat(4)
+        quarters = np.tile(np.array([1, 2, 3, 4], dtype=np.int64), 40)
+        pindex = PeriodIndex.from_fields(year=years, quarter=quarters)
+
+        tm.assert_index_equal(pindex.year, Index(years))
+        tm.assert_index_equal(pindex.quarter, Index(quarters))
+
+    def test_constructor_invalid_quarters(self):
+        msg = "Quarter must be 1 <= q <= 4"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex.from_fields(
+                year=range(2000, 2004), quarter=list(range(4)), freq="Q-DEC"
+            )
+
+    def test_period_range_fractional_period(self):
+        msg = "periods must be an integer, got 10.5"
+        with pytest.raises(TypeError, match=msg):
+            period_range("2007-01", periods=10.5, freq="M")
+
+    def test_constructor_with_without_freq(self):
+        # GH53687
+        start = Period("2002-01-01 00:00", freq="30min")
+        exp = period_range(start=start, periods=5, freq=start.freq)
+        result = period_range(start=start, periods=5)
+        tm.assert_index_equal(exp, result)
+
+    def test_constructor_fromarraylike(self):
+        idx = period_range("2007-01", periods=20, freq="M")
+
+        # values is an array of Period, thus can retrieve freq
+        tm.assert_index_equal(PeriodIndex(idx.values), idx)
+        tm.assert_index_equal(PeriodIndex(list(idx.values)), idx)
+
+        msg = "freq not specified and cannot be inferred"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(idx.asi8)
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(list(idx.asi8))
+
+        msg = "'Period' object is not iterable"
+        with pytest.raises(TypeError, match=msg):
+            PeriodIndex(data=Period("2007", freq="Y"))
+
+        result = PeriodIndex(iter(idx))
+        tm.assert_index_equal(result, idx)
+
+        result = PeriodIndex(idx)
+        tm.assert_index_equal(result, idx)
+
+        result = PeriodIndex(idx, freq="M")
+        tm.assert_index_equal(result, idx)
+
+        result = PeriodIndex(idx, freq=offsets.MonthEnd())
+        tm.assert_index_equal(result, idx)
+        assert result.freq == "ME"
+
+        result = PeriodIndex(idx, freq="2M")
+        tm.assert_index_equal(result, idx.asfreq("2M"))
+        assert result.freq == "2ME"
+
+        result = PeriodIndex(idx, freq=offsets.MonthEnd(2))
+        tm.assert_index_equal(result, idx.asfreq("2M"))
+        assert result.freq == "2ME"
+
+        result = PeriodIndex(idx, freq="D")
+        exp = idx.asfreq("D", "e")
+        tm.assert_index_equal(result, exp)
+
+    def test_constructor_datetime64arr(self):
+        vals = np.arange(100000, 100000 + 10000, 100, dtype=np.int64)
+        vals = vals.view(np.dtype("M8[us]"))
+
+        pi = PeriodIndex(vals, freq="D")
+
+        expected = PeriodIndex(vals.astype("M8[ns]"), freq="D")
+        tm.assert_index_equal(pi, expected)
+
+    @pytest.mark.parametrize("box", [None, "series", "index"])
+    def test_constructor_datetime64arr_ok(self, box):
+        # https://github.com/pandas-dev/pandas/issues/23438
+        data = date_range("2017", periods=4, freq="ME")
+        if box is None:
+            data = data._values
+        elif box == "series":
+            data = Series(data)
+
+        result = PeriodIndex(data, freq="D")
+        expected = PeriodIndex(
+            ["2017-01-31", "2017-02-28", "2017-03-31", "2017-04-30"], freq="D"
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_dtype(self):
+        # passing a dtype with a tz should localize
+        idx = PeriodIndex(["2013-01", "2013-03"], dtype="period[M]")
+        exp = PeriodIndex(["2013-01", "2013-03"], freq="M")
+        tm.assert_index_equal(idx, exp)
+        assert idx.dtype == "period[M]"
+
+        idx = PeriodIndex(["2013-01-05", "2013-03-05"], dtype="period[3D]")
+        exp = PeriodIndex(["2013-01-05", "2013-03-05"], freq="3D")
+        tm.assert_index_equal(idx, exp)
+        assert idx.dtype == "period[3D]"
+
+        # if we already have a freq and its not the same, then asfreq
+        # (not changed)
+        idx = PeriodIndex(["2013-01-01", "2013-01-02"], freq="D")
+
+        res = PeriodIndex(idx, dtype="period[M]")
+        exp = PeriodIndex(["2013-01", "2013-01"], freq="M")
+        tm.assert_index_equal(res, exp)
+        assert res.dtype == "period[M]"
+
+        res = PeriodIndex(idx, freq="M")
+        tm.assert_index_equal(res, exp)
+        assert res.dtype == "period[M]"
+
+        msg = "specified freq and dtype are different"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            PeriodIndex(["2011-01"], freq="M", dtype="period[D]")
+
+    def test_constructor_empty(self):
+        idx = PeriodIndex([], freq="M")
+        assert isinstance(idx, PeriodIndex)
+        assert len(idx) == 0
+        assert idx.freq == "ME"
+
+        with pytest.raises(ValueError, match="freq not specified"):
+            PeriodIndex([])
+
+    def test_constructor_pi_nat(self):
+        idx = PeriodIndex(
+            [Period("2011-01", freq="M"), NaT, Period("2011-01", freq="M")]
+        )
+        exp = PeriodIndex(["2011-01", "NaT", "2011-01"], freq="M")
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex(
+            np.array([Period("2011-01", freq="M"), NaT, Period("2011-01", freq="M")])
+        )
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex(
+            [NaT, NaT, Period("2011-01", freq="M"), Period("2011-01", freq="M")]
+        )
+        exp = PeriodIndex(["NaT", "NaT", "2011-01", "2011-01"], freq="M")
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex(
+            np.array(
+                [NaT, NaT, Period("2011-01", freq="M"), Period("2011-01", freq="M")]
+            )
+        )
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex([NaT, NaT, "2011-01", "2011-01"], freq="M")
+        tm.assert_index_equal(idx, exp)
+
+        with pytest.raises(ValueError, match="freq not specified"):
+            PeriodIndex([NaT, NaT])
+
+        with pytest.raises(ValueError, match="freq not specified"):
+            PeriodIndex(np.array([NaT, NaT]))
+
+        with pytest.raises(ValueError, match="freq not specified"):
+            PeriodIndex(["NaT", "NaT"])
+
+        with pytest.raises(ValueError, match="freq not specified"):
+            PeriodIndex(np.array(["NaT", "NaT"]))
+
+    def test_constructor_incompat_freq(self):
+        msg = "Input has different freq=D from PeriodIndex\\(freq=M\\)"
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            PeriodIndex([Period("2011-01", freq="M"), NaT, Period("2011-01", freq="D")])
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            PeriodIndex(
+                np.array(
+                    [Period("2011-01", freq="M"), NaT, Period("2011-01", freq="D")]
+                )
+            )
+
+        # first element is NaT
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            PeriodIndex([NaT, Period("2011-01", freq="M"), Period("2011-01", freq="D")])
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            PeriodIndex(
+                np.array(
+                    [NaT, Period("2011-01", freq="M"), Period("2011-01", freq="D")]
+                )
+            )
+
+    def test_constructor_mixed(self):
+        idx = PeriodIndex(["2011-01", NaT, Period("2011-01", freq="M")])
+        exp = PeriodIndex(["2011-01", "NaT", "2011-01"], freq="M")
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex(["NaT", NaT, Period("2011-01", freq="M")])
+        exp = PeriodIndex(["NaT", "NaT", "2011-01"], freq="M")
+        tm.assert_index_equal(idx, exp)
+
+        idx = PeriodIndex([Period("2011-01-01", freq="D"), NaT, "2012-01-01"])
+        exp = PeriodIndex(["2011-01-01", "NaT", "2012-01-01"], freq="D")
+        tm.assert_index_equal(idx, exp)
+
+    @pytest.mark.parametrize("floats", [[1.1, 2.1], np.array([1.1, 2.1])])
+    def test_constructor_floats(self, floats):
+        msg = "PeriodIndex does not allow floating point in construction"
+        with pytest.raises(TypeError, match=msg):
+            PeriodIndex(floats)
+
+    def test_constructor_year_and_quarter(self):
+        year = Series([2001, 2002, 2003])
+        quarter = year - 2000
+        idx = PeriodIndex.from_fields(year=year, quarter=quarter)
+        strs = [f"{t[0]:d}Q{t[1]:d}" for t in zip(quarter, year)]
+        lops = list(map(Period, strs))
+        p = PeriodIndex(lops)
+        tm.assert_index_equal(p, idx)
+
+    def test_constructor_freq_mult(self):
+        # GH #7811
+        pidx = period_range(start="2014-01", freq="2M", periods=4)
+        expected = PeriodIndex(["2014-01", "2014-03", "2014-05", "2014-07"], freq="2M")
+        tm.assert_index_equal(pidx, expected)
+
+        pidx = period_range(start="2014-01-02", end="2014-01-15", freq="3D")
+        expected = PeriodIndex(
+            ["2014-01-02", "2014-01-05", "2014-01-08", "2014-01-11", "2014-01-14"],
+            freq="3D",
+        )
+        tm.assert_index_equal(pidx, expected)
+
+        pidx = period_range(end="2014-01-01 17:00", freq="4h", periods=3)
+        expected = PeriodIndex(
+            ["2014-01-01 09:00", "2014-01-01 13:00", "2014-01-01 17:00"], freq="4h"
+        )
+        tm.assert_index_equal(pidx, expected)
+
+        msg = "Frequency must be positive, because it represents span: -1M"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(["2011-01"], freq="-1M")
+
+        msg = "Frequency must be positive, because it represents span: 0M"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(["2011-01"], freq="0M")
+
+        msg = "Frequency must be positive, because it represents span: 0M"
+        with pytest.raises(ValueError, match=msg):
+            period_range("2011-01", periods=3, freq="0M")
+
+    @pytest.mark.parametrize(
+        "freq_offset, freq_period",
+        [
+            ("YE", "Y"),
+            ("ME", "M"),
+            ("D", "D"),
+            ("min", "min"),
+            ("s", "s"),
+        ],
+    )
+    @pytest.mark.parametrize("mult", [1, 2, 3, 4, 5])
+    def test_constructor_freq_mult_dti_compat(self, mult, freq_offset, freq_period):
+        freqstr_offset = str(mult) + freq_offset
+        freqstr_period = str(mult) + freq_period
+        pidx = period_range(start="2014-04-01", freq=freqstr_period, periods=10)
+        expected = date_range(
+            start="2014-04-01", freq=freqstr_offset, periods=10
+        ).to_period(freqstr_period)
+        tm.assert_index_equal(pidx, expected)
+
+    @pytest.mark.parametrize("mult", [1, 2, 3, 4, 5])
+    def test_constructor_freq_mult_dti_compat_month(self, mult):
+        pidx = period_range(start="2014-04-01", freq=f"{mult}M", periods=10)
+        expected = date_range(
+            start="2014-04-01", freq=f"{mult}ME", periods=10
+        ).to_period(f"{mult}M")
+        tm.assert_index_equal(pidx, expected)
+
+    def test_constructor_freq_combined(self):
+        for freq in ["1D1h", "1h1D"]:
+            pidx = PeriodIndex(["2016-01-01", "2016-01-02"], freq=freq)
+            expected = PeriodIndex(["2016-01-01 00:00", "2016-01-02 00:00"], freq="25h")
+        for freq in ["1D1h", "1h1D"]:
+            pidx = period_range(start="2016-01-01", periods=2, freq=freq)
+            expected = PeriodIndex(["2016-01-01 00:00", "2016-01-02 01:00"], freq="25h")
+            tm.assert_index_equal(pidx, expected)
+
+    def test_period_range_length(self):
+        pi = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        assert len(pi) == 9
+
+        pi = period_range(freq="Q", start="1/1/2001", end="12/1/2009")
+        assert len(pi) == 4 * 9
+
+        pi = period_range(freq="M", start="1/1/2001", end="12/1/2009")
+        assert len(pi) == 12 * 9
+
+        pi = period_range(freq="D", start="1/1/2001", end="12/31/2009")
+        assert len(pi) == 365 * 9 + 2
+
+        msg = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            pi = period_range(freq="B", start="1/1/2001", end="12/31/2009")
+        assert len(pi) == 261 * 9
+
+        pi = period_range(freq="h", start="1/1/2001", end="12/31/2001 23:00")
+        assert len(pi) == 365 * 24
+
+        pi = period_range(freq="Min", start="1/1/2001", end="1/1/2001 23:59")
+        assert len(pi) == 24 * 60
+
+        pi = period_range(freq="s", start="1/1/2001", end="1/1/2001 23:59:59")
+        assert len(pi) == 24 * 60 * 60
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            start = Period("02-Apr-2005", "B")
+            i1 = period_range(start=start, periods=20)
+        assert len(i1) == 20
+        assert i1.freq == start.freq
+        assert i1[0] == start
+
+        end_intv = Period("2006-12-31", "W")
+        i1 = period_range(end=end_intv, periods=10)
+        assert len(i1) == 10
+        assert i1.freq == end_intv.freq
+        assert i1[-1] == end_intv
+
+        msg = "'w' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            end_intv = Period("2006-12-31", "1w")
+        i2 = period_range(end=end_intv, periods=10)
+        assert len(i1) == len(i2)
+        assert (i1 == i2).all()
+        assert i1.freq == i2.freq
+
+    def test_infer_freq_from_first_element(self):
+        msg = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            start = Period("02-Apr-2005", "B")
+            end_intv = Period("2005-05-01", "B")
+            period_range(start=start, end=end_intv)
+
+            # infer freq from first element
+            i2 = PeriodIndex([end_intv, Period("2005-05-05", "B")])
+        assert len(i2) == 2
+        assert i2[0] == end_intv
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            i2 = PeriodIndex(np.array([end_intv, Period("2005-05-05", "B")]))
+        assert len(i2) == 2
+        assert i2[0] == end_intv
+
+    def test_mixed_freq_raises(self):
+        # Mixed freq should fail
+        msg = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            end_intv = Period("2005-05-01", "B")
+
+        vals = [end_intv, Period("2006-12-31", "W")]
+        msg = r"Input has different freq=W-SUN from PeriodIndex\(freq=B\)"
+        depr_msg = r"PeriodDtype\[B\] is deprecated"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+                PeriodIndex(vals)
+        vals = np.array(vals)
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+                PeriodIndex(vals)
+
+    @pytest.mark.parametrize(
+        "freq", ["M", "Q", "Y", "D", "B", "min", "s", "ms", "us", "ns", "h"]
+    )
+    @pytest.mark.filterwarnings(
+        r"ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_recreate_from_data(self, freq):
+        org = period_range(start="2001/04/01", freq=freq, periods=1)
+        idx = PeriodIndex(org.values, freq=freq)
+        tm.assert_index_equal(idx, org)
+
+    def test_map_with_string_constructor(self):
+        raw = [2005, 2007, 2009]
+        index = PeriodIndex(raw, freq="Y")
+
+        expected = Index([str(num) for num in raw])
+        res = index.map(str)
+
+        # should return an Index
+        assert isinstance(res, Index)
+
+        # preserve element types
+        assert all(isinstance(resi, str) for resi in res)
+
+        # lastly, values should compare equal
+        tm.assert_index_equal(res, expected)
+
+
+class TestSimpleNew:
+    def test_constructor_simple_new(self):
+        idx = period_range("2007-01", name="p", periods=2, freq="M")
+
+        with pytest.raises(AssertionError, match="<class .*PeriodIndex'>"):
+            idx._simple_new(idx, name="p")
+
+        result = idx._simple_new(idx._data, name="p")
+        tm.assert_index_equal(result, idx)
+
+        msg = "Should be numpy array of type i8"
+        with pytest.raises(AssertionError, match=msg):
+            # Need ndarray, not int64 Index
+            type(idx._data)._simple_new(Index(idx.asi8), dtype=idx.dtype)
+
+        arr = type(idx._data)._simple_new(idx.asi8, dtype=idx.dtype)
+        result = idx._simple_new(arr, name="p")
+        tm.assert_index_equal(result, idx)
+
+    def test_constructor_simple_new_empty(self):
+        # GH13079
+        idx = PeriodIndex([], freq="M", name="p")
+        with pytest.raises(AssertionError, match="<class .*PeriodIndex'>"):
+            idx._simple_new(idx, name="p")
+
+        result = idx._simple_new(idx._data, name="p")
+        tm.assert_index_equal(result, idx)
+
+    @pytest.mark.parametrize("floats", [[1.1, 2.1], np.array([1.1, 2.1])])
+    def test_period_index_simple_new_disallows_floats(self, floats):
+        with pytest.raises(AssertionError, match="<class "):
+            PeriodIndex._simple_new(floats)
+
+
+class TestShallowCopy:
+    def test_shallow_copy_empty(self):
+        # GH#13067
+        idx = PeriodIndex([], freq="M")
+        result = idx._view()
+        expected = idx
+
+        tm.assert_index_equal(result, expected)
+
+    def test_shallow_copy_disallow_i8(self):
+        # GH#24391
+        pi = period_range("2018-01-01", periods=3, freq="2D")
+        with pytest.raises(AssertionError, match="ndarray"):
+            pi._shallow_copy(pi.asi8)
+
+    def test_shallow_copy_requires_disallow_period_index(self):
+        pi = period_range("2018-01-01", periods=3, freq="2D")
+        with pytest.raises(AssertionError, match="PeriodIndex"):
+            pi._shallow_copy(pi)
+
+
+class TestSeriesPeriod:
+    def test_constructor_cant_cast_period(self):
+        msg = "Cannot cast PeriodIndex to dtype float64"
+        with pytest.raises(TypeError, match=msg):
+            Series(period_range("2000-01-01", periods=10, freq="D"), dtype=float)
+
+    def test_constructor_cast_object(self):
+        pi = period_range("1/1/2000", periods=10)
+        ser = Series(pi, dtype=PeriodDtype("D"))
+        exp = Series(pi)
+        tm.assert_series_equal(ser, exp)
diff --git a/pandas/pandas/tests/indexes/period/test_formats.py b/pandas/pandas/tests/indexes/period/test_formats.py
new file mode 100644
index 0000000000000..dc95e19523842
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_formats.py
@@ -0,0 +1,193 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    PeriodIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_get_values_for_csv():
+    index = PeriodIndex(["2017-01-01", "2017-01-02", "2017-01-03"], freq="D")
+
+    # First, with no arguments.
+    expected = np.array(["2017-01-01", "2017-01-02", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv()
+    tm.assert_numpy_array_equal(result, expected)
+
+    # No NaN values, so na_rep has no effect
+    result = index._get_values_for_csv(na_rep="pandas")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # Make sure date formatting works
+    expected = np.array(["01-2017-01", "01-2017-02", "01-2017-03"], dtype=object)
+
+    result = index._get_values_for_csv(date_format="%m-%Y-%d")
+    tm.assert_numpy_array_equal(result, expected)
+
+    # NULL object handling should work
+    index = PeriodIndex(["2017-01-01", pd.NaT, "2017-01-03"], freq="D")
+    expected = np.array(["2017-01-01", "NaT", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv(na_rep="NaT")
+    tm.assert_numpy_array_equal(result, expected)
+
+    expected = np.array(["2017-01-01", "pandas", "2017-01-03"], dtype=object)
+
+    result = index._get_values_for_csv(na_rep="pandas")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+class TestPeriodIndexRendering:
+    @pytest.mark.parametrize("method", ["__repr__", "__str__"])
+    def test_representation(self, method):
+        # GH#7601
+        idx1 = PeriodIndex([], freq="D")
+        idx2 = PeriodIndex(["2011-01-01"], freq="D")
+        idx3 = PeriodIndex(["2011-01-01", "2011-01-02"], freq="D")
+        idx4 = PeriodIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        idx5 = PeriodIndex(["2011", "2012", "2013"], freq="Y")
+        idx6 = PeriodIndex(["2011-01-01 09:00", "2012-02-01 10:00", "NaT"], freq="h")
+        idx7 = pd.period_range("2013Q1", periods=1, freq="Q")
+        idx8 = pd.period_range("2013Q1", periods=2, freq="Q")
+        idx9 = pd.period_range("2013Q1", periods=3, freq="Q")
+        idx10 = PeriodIndex(["2011-01-01", "2011-02-01"], freq="3D")
+
+        exp1 = "PeriodIndex([], dtype='period[D]')"
+
+        exp2 = "PeriodIndex(['2011-01-01'], dtype='period[D]')"
+
+        exp3 = "PeriodIndex(['2011-01-01', '2011-01-02'], dtype='period[D]')"
+
+        exp4 = (
+            "PeriodIndex(['2011-01-01', '2011-01-02', '2011-01-03'], dtype='period[D]')"
+        )
+
+        exp5 = "PeriodIndex(['2011', '2012', '2013'], dtype='period[Y-DEC]')"
+
+        exp6 = (
+            "PeriodIndex(['2011-01-01 09:00', '2012-02-01 10:00', 'NaT'], "
+            "dtype='period[h]')"
+        )
+
+        exp7 = "PeriodIndex(['2013Q1'], dtype='period[Q-DEC]')"
+
+        exp8 = "PeriodIndex(['2013Q1', '2013Q2'], dtype='period[Q-DEC]')"
+
+        exp9 = "PeriodIndex(['2013Q1', '2013Q2', '2013Q3'], dtype='period[Q-DEC]')"
+
+        exp10 = "PeriodIndex(['2011-01-01', '2011-02-01'], dtype='period[3D]')"
+
+        for idx, expected in zip(
+            [idx1, idx2, idx3, idx4, idx5, idx6, idx7, idx8, idx9, idx10],
+            [exp1, exp2, exp3, exp4, exp5, exp6, exp7, exp8, exp9, exp10],
+        ):
+            result = getattr(idx, method)()
+            assert result == expected
+
+    # TODO: These are Series.__repr__ tests
+    def test_representation_to_series(self):
+        # GH#10971
+        idx1 = PeriodIndex([], freq="D")
+        idx2 = PeriodIndex(["2011-01-01"], freq="D")
+        idx3 = PeriodIndex(["2011-01-01", "2011-01-02"], freq="D")
+        idx4 = PeriodIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        idx5 = PeriodIndex(["2011", "2012", "2013"], freq="Y")
+        idx6 = PeriodIndex(["2011-01-01 09:00", "2012-02-01 10:00", "NaT"], freq="h")
+
+        idx7 = pd.period_range("2013Q1", periods=1, freq="Q")
+        idx8 = pd.period_range("2013Q1", periods=2, freq="Q")
+        idx9 = pd.period_range("2013Q1", periods=3, freq="Q")
+
+        exp1 = """Series([], dtype: period[D])"""
+
+        exp2 = """0    2011-01-01
+dtype: period[D]"""
+
+        exp3 = """0    2011-01-01
+1    2011-01-02
+dtype: period[D]"""
+
+        exp4 = """0    2011-01-01
+1    2011-01-02
+2    2011-01-03
+dtype: period[D]"""
+
+        exp5 = """0    2011
+1    2012
+2    2013
+dtype: period[Y-DEC]"""
+
+        exp6 = """0    2011-01-01 09:00
+1    2012-02-01 10:00
+2                 NaT
+dtype: period[h]"""
+
+        exp7 = """0    2013Q1
+dtype: period[Q-DEC]"""
+
+        exp8 = """0    2013Q1
+1    2013Q2
+dtype: period[Q-DEC]"""
+
+        exp9 = """0    2013Q1
+1    2013Q2
+2    2013Q3
+dtype: period[Q-DEC]"""
+
+        for idx, expected in zip(
+            [idx1, idx2, idx3, idx4, idx5, idx6, idx7, idx8, idx9],
+            [exp1, exp2, exp3, exp4, exp5, exp6, exp7, exp8, exp9],
+        ):
+            result = repr(Series(idx))
+            assert result == expected
+
+    def test_summary(self):
+        # GH#9116
+        idx1 = PeriodIndex([], freq="D")
+        idx2 = PeriodIndex(["2011-01-01"], freq="D")
+        idx3 = PeriodIndex(["2011-01-01", "2011-01-02"], freq="D")
+        idx4 = PeriodIndex(["2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        idx5 = PeriodIndex(["2011", "2012", "2013"], freq="Y")
+        idx6 = PeriodIndex(["2011-01-01 09:00", "2012-02-01 10:00", "NaT"], freq="h")
+
+        idx7 = pd.period_range("2013Q1", periods=1, freq="Q")
+        idx8 = pd.period_range("2013Q1", periods=2, freq="Q")
+        idx9 = pd.period_range("2013Q1", periods=3, freq="Q")
+
+        exp1 = """PeriodIndex: 0 entries
+Freq: D"""
+
+        exp2 = """PeriodIndex: 1 entries, 2011-01-01 to 2011-01-01
+Freq: D"""
+
+        exp3 = """PeriodIndex: 2 entries, 2011-01-01 to 2011-01-02
+Freq: D"""
+
+        exp4 = """PeriodIndex: 3 entries, 2011-01-01 to 2011-01-03
+Freq: D"""
+
+        exp5 = """PeriodIndex: 3 entries, 2011 to 2013
+Freq: Y-DEC"""
+
+        exp6 = """PeriodIndex: 3 entries, 2011-01-01 09:00 to NaT
+Freq: h"""
+
+        exp7 = """PeriodIndex: 1 entries, 2013Q1 to 2013Q1
+Freq: Q-DEC"""
+
+        exp8 = """PeriodIndex: 2 entries, 2013Q1 to 2013Q2
+Freq: Q-DEC"""
+
+        exp9 = """PeriodIndex: 3 entries, 2013Q1 to 2013Q3
+Freq: Q-DEC"""
+
+        for idx, expected in zip(
+            [idx1, idx2, idx3, idx4, idx5, idx6, idx7, idx8, idx9],
+            [exp1, exp2, exp3, exp4, exp5, exp6, exp7, exp8, exp9],
+        ):
+            result = idx._summary()
+            assert result == expected
diff --git a/pandas/pandas/tests/indexes/period/test_freq_attr.py b/pandas/pandas/tests/indexes/period/test_freq_attr.py
new file mode 100644
index 0000000000000..e1ecffa4982bd
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_freq_attr.py
@@ -0,0 +1,28 @@
+import pytest
+
+from pandas.compat import PY311
+
+from pandas import (
+    offsets,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestFreq:
+    def test_freq_setter_deprecated(self):
+        # GH#20678
+        idx = period_range("2018Q1", periods=4, freq="Q")
+
+        # no warning for getter
+        with tm.assert_produces_warning(None):
+            idx.freq
+
+        # warning for setter
+        msg = (
+            "property 'freq' of 'PeriodArray' object has no setter"
+            if PY311
+            else "can't set attribute"
+        )
+        with pytest.raises(AttributeError, match=msg):
+            idx.freq = offsets.Day()
diff --git a/pandas/pandas/tests/indexes/period/test_indexing.py b/pandas/pandas/tests/indexes/period/test_indexing.py
new file mode 100644
index 0000000000000..00e8262ddfa4c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_indexing.py
@@ -0,0 +1,814 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import period as libperiod
+from pandas.errors import InvalidIndexError
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    Period,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    date_range,
+    notna,
+    period_range,
+)
+import pandas._testing as tm
+
+dti4 = date_range("2016-01-01", periods=4)
+dti = dti4[:-1]
+rng = pd.Index(range(3))
+
+
+@pytest.fixture(
+    params=[
+        dti,
+        dti.tz_localize("UTC"),
+        dti.to_period("W"),
+        dti - dti[0],
+        rng,
+        pd.Index([1, 2, 3]),
+        pd.Index([2.0, 3.0, 4.0]),
+        pd.Index([4, 5, 6], dtype="u8"),
+        pd.IntervalIndex.from_breaks(dti4),
+    ]
+)
+def non_comparable_idx(request):
+    # All have length 3
+    return request.param
+
+
+class TestGetItem:
+    def test_getitem_slice_keeps_name(self):
+        idx = period_range("20010101", periods=10, freq="D", name="bob")
+        assert idx.name == idx[1:].name
+
+    def test_getitem(self):
+        idx1 = period_range("2011-01-01", "2011-01-31", freq="D", name="idx")
+
+        for idx in [idx1]:
+            result = idx[0]
+            assert result == Period("2011-01-01", freq="D")
+
+            result = idx[-1]
+            assert result == Period("2011-01-31", freq="D")
+
+            result = idx[0:5]
+            expected = period_range("2011-01-01", "2011-01-05", freq="D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx[0:10:2]
+            expected = PeriodIndex(
+                ["2011-01-01", "2011-01-03", "2011-01-05", "2011-01-07", "2011-01-09"],
+                freq="D",
+                name="idx",
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx[-20:-5:3]
+            expected = PeriodIndex(
+                ["2011-01-12", "2011-01-15", "2011-01-18", "2011-01-21", "2011-01-24"],
+                freq="D",
+                name="idx",
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx[4::-1]
+            expected = PeriodIndex(
+                ["2011-01-05", "2011-01-04", "2011-01-03", "2011-01-02", "2011-01-01"],
+                freq="D",
+                name="idx",
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+    def test_getitem_index(self):
+        idx = period_range("2007-01", periods=10, freq="M", name="x")
+
+        result = idx[[1, 3, 5]]
+        exp = PeriodIndex(["2007-02", "2007-04", "2007-06"], freq="M", name="x")
+        tm.assert_index_equal(result, exp)
+
+        result = idx[[True, True, False, False, False, True, True, False, False, False]]
+        exp = PeriodIndex(
+            ["2007-01", "2007-02", "2007-06", "2007-07"], freq="M", name="x"
+        )
+        tm.assert_index_equal(result, exp)
+
+    def test_getitem_partial(self):
+        rng = period_range("2007-01", periods=50, freq="M")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+
+        with pytest.raises(KeyError, match=r"^'2006'$"):
+            ts["2006"]
+
+        result = ts["2008"]
+        assert (result.index.year == 2008).all()
+
+        result = ts["2008":"2009"]
+        assert len(result) == 24
+
+        result = ts["2008-1":"2009-12"]
+        assert len(result) == 24
+
+        result = ts["2008Q1":"2009Q4"]
+        assert len(result) == 24
+
+        result = ts[:"2009"]
+        assert len(result) == 36
+
+        result = ts["2009":]
+        assert len(result) == 50 - 24
+
+        exp = result
+        result = ts[24:]
+        tm.assert_series_equal(exp, result)
+
+        ts = pd.concat([ts[10:], ts[10:]])
+        msg = "left slice bound for non-unique label: '2008'"
+        with pytest.raises(KeyError, match=msg):
+            ts[slice("2008", "2009")]
+
+    def test_getitem_datetime(self):
+        rng = period_range(start="2012-01-01", periods=10, freq="W-MON")
+        ts = Series(range(len(rng)), index=rng)
+
+        dt1 = datetime(2011, 10, 2)
+        dt4 = datetime(2012, 4, 20)
+
+        rs = ts[dt1:dt4]
+        tm.assert_series_equal(rs, ts)
+
+    def test_getitem_nat(self):
+        idx = PeriodIndex(["2011-01", "NaT", "2011-02"], freq="M")
+        assert idx[0] == Period("2011-01", freq="M")
+        assert idx[1] is NaT
+
+        s = Series([0, 1, 2], index=idx)
+        assert s[NaT] == 1
+
+        s = Series(idx, index=idx)
+        assert s[Period("2011-01", freq="M")] == Period("2011-01", freq="M")
+        assert s[NaT] is NaT
+
+    def test_getitem_list_periods(self):
+        # GH 7710
+        rng = period_range(start="2012-01-01", periods=10, freq="D")
+        ts = Series(range(len(rng)), index=rng)
+        exp = ts.iloc[[1]]
+        tm.assert_series_equal(ts[[Period("2012-01-02", freq="D")]], exp)
+
+    @pytest.mark.arm_slow
+    def test_getitem_seconds(self):
+        # GH#6716
+        didx = date_range(start="2013/01/01 09:00:00", freq="s", periods=4000)
+        pidx = period_range(start="2013/01/01 09:00:00", freq="s", periods=4000)
+
+        for idx in [didx, pidx]:
+            # getitem against index should raise ValueError
+            values = [
+                "2014",
+                "2013/02",
+                "2013/01/02",
+                "2013/02/01 9h",
+                "2013/02/01 09:00",
+            ]
+            for val in values:
+                # GH7116
+                # these show deprecations as we are trying
+                # to slice with non-integer indexers
+                with pytest.raises(IndexError, match="only integers, slices"):
+                    idx[val]
+
+            ser = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+            tm.assert_series_equal(ser["2013/01/01 10:00"], ser[3600:3660])
+            tm.assert_series_equal(ser["2013/01/01 9h"], ser[:3600])
+            for d in ["2013/01/01", "2013/01", "2013"]:
+                tm.assert_series_equal(ser[d], ser)
+
+    @pytest.mark.parametrize(
+        "idx_range",
+        [
+            date_range,
+            period_range,
+        ],
+    )
+    def test_getitem_day(self, idx_range):
+        # GH#6716
+        # Confirm DatetimeIndex and PeriodIndex works identically
+        # getitem against index should raise ValueError
+        idx = idx_range(start="2013/01/01", freq="D", periods=400)
+        values = [
+            "2014",
+            "2013/02",
+            "2013/01/02",
+            "2013/02/01 9h",
+            "2013/02/01 09:00",
+        ]
+        for val in values:
+            # GH7116
+            # these show deprecations as we are trying
+            # to slice with non-integer indexers
+            with pytest.raises(IndexError, match="only integers, slices"):
+                idx[val]
+
+        ser = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+        tm.assert_series_equal(ser["2013/01"], ser[0:31])
+        tm.assert_series_equal(ser["2013/02"], ser[31:59])
+        tm.assert_series_equal(ser["2014"], ser[365:])
+
+        invalid = ["2013/02/01 9h", "2013/02/01 09:00"]
+        for val in invalid:
+            with pytest.raises(KeyError, match=val):
+                ser[val]
+
+
+class TestGetLoc:
+    def test_get_loc_msg(self):
+        idx = period_range("2000-1-1", freq="Y", periods=10)
+        bad_period = Period("2012", "Y")
+        with pytest.raises(KeyError, match=r"^Period\('2012', 'Y-DEC'\)$"):
+            idx.get_loc(bad_period)
+
+        try:
+            idx.get_loc(bad_period)
+        except KeyError as inst:
+            assert inst.args[0] == bad_period
+
+    def test_get_loc_nat(self):
+        didx = DatetimeIndex(["2011-01-01", "NaT", "2011-01-03"])
+        pidx = PeriodIndex(["2011-01-01", "NaT", "2011-01-03"], freq="M")
+
+        # check DatetimeIndex compat
+        for idx in [didx, pidx]:
+            assert idx.get_loc(NaT) == 1
+            assert idx.get_loc(None) == 1
+            assert idx.get_loc(float("nan")) == 1
+            assert idx.get_loc(np.nan) == 1
+
+    def test_get_loc(self):
+        # GH 17717
+        p0 = Period("2017-09-01")
+        p1 = Period("2017-09-02")
+        p2 = Period("2017-09-03")
+
+        # get the location of p1/p2 from
+        # monotonic increasing PeriodIndex with non-duplicate
+        idx0 = PeriodIndex([p0, p1, p2])
+        expected_idx1_p1 = 1
+        expected_idx1_p2 = 2
+
+        assert idx0.get_loc(p1) == expected_idx1_p1
+        assert idx0.get_loc(str(p1)) == expected_idx1_p1
+        assert idx0.get_loc(p2) == expected_idx1_p2
+        assert idx0.get_loc(str(p2)) == expected_idx1_p2
+
+        msg = "Cannot interpret 'foo' as period"
+        with pytest.raises(KeyError, match=msg):
+            idx0.get_loc("foo")
+        with pytest.raises(KeyError, match=r"^1\.1$"):
+            idx0.get_loc(1.1)
+
+        with pytest.raises(InvalidIndexError, match=re.escape(str(idx0))):
+            idx0.get_loc(idx0)
+
+        # get the location of p1/p2 from
+        # monotonic increasing PeriodIndex with duplicate
+        idx1 = PeriodIndex([p1, p1, p2])
+        expected_idx1_p1 = slice(0, 2)
+        expected_idx1_p2 = 2
+
+        assert idx1.get_loc(p1) == expected_idx1_p1
+        assert idx1.get_loc(str(p1)) == expected_idx1_p1
+        assert idx1.get_loc(p2) == expected_idx1_p2
+        assert idx1.get_loc(str(p2)) == expected_idx1_p2
+
+        msg = "Cannot interpret 'foo' as period"
+        with pytest.raises(KeyError, match=msg):
+            idx1.get_loc("foo")
+
+        with pytest.raises(KeyError, match=r"^1\.1$"):
+            idx1.get_loc(1.1)
+
+        with pytest.raises(InvalidIndexError, match=re.escape(str(idx1))):
+            idx1.get_loc(idx1)
+
+        # get the location of p1/p2 from
+        # non-monotonic increasing/decreasing PeriodIndex with duplicate
+        idx2 = PeriodIndex([p2, p1, p2])
+        expected_idx2_p1 = 1
+        expected_idx2_p2 = np.array([True, False, True])
+
+        assert idx2.get_loc(p1) == expected_idx2_p1
+        assert idx2.get_loc(str(p1)) == expected_idx2_p1
+        tm.assert_numpy_array_equal(idx2.get_loc(p2), expected_idx2_p2)
+        tm.assert_numpy_array_equal(idx2.get_loc(str(p2)), expected_idx2_p2)
+
+    def test_get_loc_integer(self):
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+        with pytest.raises(KeyError, match="16801"):
+            pi.get_loc(16801)
+
+        pi2 = dti.to_period("Y")  # duplicates, ordinals are all 46
+        with pytest.raises(KeyError, match="46"):
+            pi2.get_loc(46)
+
+    def test_get_loc_invalid_string_raises_keyerror(self):
+        # GH#34240
+        pi = period_range("2000", periods=3, name="A")
+        with pytest.raises(KeyError, match="A"):
+            pi.get_loc("A")
+
+        ser = Series([1, 2, 3], index=pi)
+        with pytest.raises(KeyError, match="A"):
+            ser.loc["A"]
+
+        with pytest.raises(KeyError, match="A"):
+            ser["A"]
+
+        assert "A" not in ser
+        assert "A" not in pi
+
+    def test_get_loc_mismatched_freq(self):
+        # see also test_get_indexer_mismatched_dtype testing we get analogous
+        # behavior for get_loc
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+        pi2 = dti.to_period("W")
+        pi3 = pi.view(pi2.dtype)  # i.e. matching i8 representations
+
+        with pytest.raises(KeyError, match="W-SUN"):
+            pi.get_loc(pi2[0])
+
+        with pytest.raises(KeyError, match="W-SUN"):
+            # even though we have matching i8 values
+            pi.get_loc(pi3[0])
+
+
+class TestGetIndexer:
+    def test_get_indexer(self):
+        # GH 17717
+        p1 = Period("2017-09-01")
+        p2 = Period("2017-09-04")
+        p3 = Period("2017-09-07")
+
+        tp0 = Period("2017-08-31")
+        tp1 = Period("2017-09-02")
+        tp2 = Period("2017-09-05")
+        tp3 = Period("2017-09-09")
+
+        idx = PeriodIndex([p1, p2, p3])
+
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(idx), np.array([0, 1, 2], dtype=np.intp)
+        )
+
+        target = PeriodIndex([tp0, tp1, tp2, tp3])
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "pad"), np.array([-1, 0, 1, 2], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "backfill"), np.array([0, 1, 2, -1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest"), np.array([0, 0, 1, 2], dtype=np.intp)
+        )
+
+        res = idx.get_indexer(target, "nearest", tolerance=Timedelta("1 day"))
+        tm.assert_numpy_array_equal(res, np.array([0, 0, 1, -1], dtype=np.intp))
+
+    def test_get_indexer_mismatched_dtype(self):
+        # Check that we return all -1s and do not raise or cast incorrectly
+
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+        pi2 = dti.to_period("W")
+
+        expected = np.array([-1, -1, -1], dtype=np.intp)
+
+        result = pi.get_indexer(dti)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # This should work in both directions
+        result = dti.get_indexer(pi)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = pi.get_indexer(pi2)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # We expect the same from get_indexer_non_unique
+        result = pi.get_indexer_non_unique(dti)[0]
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = dti.get_indexer_non_unique(pi)[0]
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = pi.get_indexer_non_unique(pi2)[0]
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_indexer_mismatched_dtype_different_length(self, non_comparable_idx):
+        # without method we aren't checking inequalities, so get all-missing
+        #  but do not raise
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+
+        other = non_comparable_idx
+
+        res = pi[:-1].get_indexer(other)
+        expected = -np.ones(other.shape, dtype=np.intp)
+        tm.assert_numpy_array_equal(res, expected)
+
+    @pytest.mark.parametrize("method", ["pad", "backfill", "nearest"])
+    def test_get_indexer_mismatched_dtype_with_method(self, non_comparable_idx, method):
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+
+        other = non_comparable_idx
+
+        msg = re.escape(f"Cannot compare dtypes {pi.dtype} and {other.dtype}")
+        with pytest.raises(TypeError, match=msg):
+            pi.get_indexer(other, method=method)
+
+        for dtype in ["object", "category"]:
+            other2 = other.astype(dtype)
+            if dtype == "object" and isinstance(other, PeriodIndex):
+                continue
+            # Two different error message patterns depending on dtypes
+            msg = "|".join(
+                [
+                    re.escape(msg)
+                    for msg in (
+                        f"Cannot compare dtypes {pi.dtype} and {other.dtype}",
+                        " not supported between instances of ",
+                    )
+                ]
+            )
+            with pytest.raises(TypeError, match=msg):
+                pi.get_indexer(other2, method=method)
+
+    def test_get_indexer_non_unique(self):
+        # GH 17717
+        p1 = Period("2017-09-02")
+        p2 = Period("2017-09-03")
+        p3 = Period("2017-09-04")
+        p4 = Period("2017-09-05")
+
+        idx1 = PeriodIndex([p1, p2, p1])
+        idx2 = PeriodIndex([p2, p1, p3, p4])
+
+        result = idx1.get_indexer_non_unique(idx2)
+        expected_indexer = np.array([1, 0, 2, -1, -1], dtype=np.intp)
+        expected_missing = np.array([2, 3], dtype=np.intp)
+
+        tm.assert_numpy_array_equal(result[0], expected_indexer)
+        tm.assert_numpy_array_equal(result[1], expected_missing)
+
+    # TODO: This method came from test_period; de-dup with version above
+    def test_get_indexer2(self):
+        idx = period_range("2000-01-01", periods=3).asfreq("h", how="start")
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(idx), np.array([0, 1, 2], dtype=np.intp)
+        )
+
+        target = PeriodIndex(
+            ["1999-12-31T23", "2000-01-01T12", "2000-01-02T01"], freq="h"
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "pad"), np.array([-1, 0, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "backfill"), np.array([0, 1, 2], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest"), np.array([0, 1, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest", tolerance="1 hour"),
+            np.array([0, -1, 1], dtype=np.intp),
+        )
+
+        msg = "Input has different freq=None from PeriodArray\\(freq=h\\)"
+        with pytest.raises(ValueError, match=msg):
+            idx.get_indexer(target, "nearest", tolerance="1 minute")
+
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest", tolerance="1 day"),
+            np.array([0, 1, 1], dtype=np.intp),
+        )
+        tol_raw = [
+            Timedelta("1 hour"),
+            Timedelta("1 hour"),
+            np.timedelta64(1, "D"),
+        ]
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(
+                target, "nearest", tolerance=[np.timedelta64(x) for x in tol_raw]
+            ),
+            np.array([0, -1, 1], dtype=np.intp),
+        )
+        tol_bad = [
+            Timedelta("2 hour").to_timedelta64(),
+            Timedelta("1 hour").to_timedelta64(),
+            np.timedelta64(1, "M"),
+        ]
+        with pytest.raises(
+            libperiod.IncompatibleFrequency, match="Input has different freq=None from"
+        ):
+            idx.get_indexer(target, "nearest", tolerance=tol_bad)
+
+
+class TestWhere:
+    def test_where(self, listlike_box):
+        i = period_range("20130101", periods=5, freq="D")
+        cond = [True] * len(i)
+        expected = i
+        result = i.where(listlike_box(cond))
+        tm.assert_index_equal(result, expected)
+
+        cond = [False] + [True] * (len(i) - 1)
+        expected = PeriodIndex([NaT] + i[1:].tolist(), freq="D")
+        result = i.where(listlike_box(cond))
+        tm.assert_index_equal(result, expected)
+
+    def test_where_other(self):
+        i = period_range("20130101", periods=5, freq="D")
+        for arr in [np.nan, NaT]:
+            result = i.where(notna(i), other=arr)
+            expected = i
+            tm.assert_index_equal(result, expected)
+
+        i2 = i.copy()
+        i2 = PeriodIndex([NaT, NaT] + i[2:].tolist(), freq="D")
+        result = i.where(notna(i2), i2)
+        tm.assert_index_equal(result, i2)
+
+        i2 = i.copy()
+        i2 = PeriodIndex([NaT, NaT] + i[2:].tolist(), freq="D")
+        result = i.where(notna(i2), i2.values)
+        tm.assert_index_equal(result, i2)
+
+    def test_where_invalid_dtypes(self):
+        pi = period_range("20130101", periods=5, freq="D")
+
+        tail = pi[2:].tolist()
+        i2 = PeriodIndex([NaT, NaT] + tail, freq="D")
+        mask = notna(i2)
+
+        result = pi.where(mask, i2.asi8)
+        expected = pd.Index([NaT._value, NaT._value] + tail, dtype=object)
+        assert isinstance(expected[0], int)
+        tm.assert_index_equal(result, expected)
+
+        tdi = i2.asi8.view("timedelta64[ns]")
+        expected = pd.Index([tdi[0], tdi[1]] + tail, dtype=object)
+        assert isinstance(expected[0], np.timedelta64)
+        result = pi.where(mask, tdi)
+        tm.assert_index_equal(result, expected)
+
+        dti = i2.to_timestamp("s")
+        expected = pd.Index([dti[0], dti[1]] + tail, dtype=object)
+        assert expected[0] is NaT
+        result = pi.where(mask, dti)
+        tm.assert_index_equal(result, expected)
+
+        td = Timedelta(days=4)
+        expected = pd.Index([td, td] + tail, dtype=object)
+        assert expected[0] == td
+        result = pi.where(mask, td)
+        tm.assert_index_equal(result, expected)
+
+    def test_where_mismatched_nat(self):
+        pi = period_range("20130101", periods=5, freq="D")
+        cond = np.array([True, False, True, True, False])
+
+        tdnat = np.timedelta64("NaT", "ns")
+        expected = pd.Index([pi[0], tdnat, pi[2], pi[3], tdnat], dtype=object)
+        assert expected[1] is tdnat
+        result = pi.where(cond, tdnat)
+        tm.assert_index_equal(result, expected)
+
+
+class TestTake:
+    def test_take(self):
+        # GH#10295
+        idx1 = period_range("2011-01-01", "2011-01-31", freq="D", name="idx")
+
+        for idx in [idx1]:
+            result = idx.take([0])
+            assert result == Period("2011-01-01", freq="D")
+
+            result = idx.take([5])
+            assert result == Period("2011-01-06", freq="D")
+
+            result = idx.take([0, 1, 2])
+            expected = period_range("2011-01-01", "2011-01-03", freq="D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == "D"
+            assert result.freq == expected.freq
+
+            result = idx.take([0, 2, 4])
+            expected = PeriodIndex(
+                ["2011-01-01", "2011-01-03", "2011-01-05"], freq="D", name="idx"
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx.take([7, 4, 1])
+            expected = PeriodIndex(
+                ["2011-01-08", "2011-01-05", "2011-01-02"], freq="D", name="idx"
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx.take([3, 2, 5])
+            expected = PeriodIndex(
+                ["2011-01-04", "2011-01-03", "2011-01-06"], freq="D", name="idx"
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+            result = idx.take([-3, 2, 5])
+            expected = PeriodIndex(
+                ["2011-01-29", "2011-01-03", "2011-01-06"], freq="D", name="idx"
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+            assert result.freq == "D"
+
+    def test_take_misc(self):
+        index = period_range(start="1/1/10", end="12/31/12", freq="D", name="idx")
+        expected = PeriodIndex(
+            [
+                datetime(2010, 1, 6),
+                datetime(2010, 1, 7),
+                datetime(2010, 1, 9),
+                datetime(2010, 1, 13),
+            ],
+            freq="D",
+            name="idx",
+        )
+
+        taken1 = index.take([5, 6, 8, 12])
+        taken2 = index[[5, 6, 8, 12]]
+
+        for taken in [taken1, taken2]:
+            tm.assert_index_equal(taken, expected)
+            assert isinstance(taken, PeriodIndex)
+            assert taken.freq == index.freq
+            assert taken.name == expected.name
+
+    def test_take_fill_value(self):
+        # GH#12631
+        idx = PeriodIndex(
+            ["2011-01-01", "2011-02-01", "2011-03-01"], name="xxx", freq="D"
+        )
+        result = idx.take(np.array([1, 0, -1]))
+        expected = PeriodIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx", freq="D"
+        )
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = PeriodIndex(
+            ["2011-02-01", "2011-01-01", "NaT"], name="xxx", freq="D"
+        )
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = PeriodIndex(
+            ["2011-02-01", "2011-01-01", "2011-03-01"], name="xxx", freq="D"
+        )
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for( axis 0 with)? size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+
+class TestGetValue:
+    @pytest.mark.parametrize("freq", ["h", "D"])
+    def test_get_value_datetime_hourly(self, freq):
+        # get_loc and get_value should treat datetime objects symmetrically
+        # TODO: this test used to test get_value, which is removed in 2.0.
+        #  should this test be moved somewhere, or is what's left redundant?
+        dti = date_range("2016-01-01", periods=3, freq="MS")
+        pi = dti.to_period(freq)
+        ser = Series(range(7, 10), index=pi)
+
+        ts = dti[0]
+
+        assert pi.get_loc(ts) == 0
+        assert ser[ts] == 7
+        assert ser.loc[ts] == 7
+
+        ts2 = ts + Timedelta(hours=3)
+        if freq == "h":
+            with pytest.raises(KeyError, match="2016-01-01 03:00"):
+                pi.get_loc(ts2)
+            with pytest.raises(KeyError, match="2016-01-01 03:00"):
+                ser[ts2]
+            with pytest.raises(KeyError, match="2016-01-01 03:00"):
+                ser.loc[ts2]
+        else:
+            assert pi.get_loc(ts2) == 0
+            assert ser[ts2] == 7
+            assert ser.loc[ts2] == 7
+
+
+class TestContains:
+    def test_contains(self):
+        # GH 17717
+        p0 = Period("2017-09-01")
+        p1 = Period("2017-09-02")
+        p2 = Period("2017-09-03")
+        p3 = Period("2017-09-04")
+
+        ps0 = [p0, p1, p2]
+        idx0 = PeriodIndex(ps0)
+
+        for p in ps0:
+            assert p in idx0
+            assert str(p) in idx0
+
+        # GH#31172
+        # Higher-resolution period-like are _not_ considered as contained
+        key = "2017-09-01 00:00:01"
+        assert key not in idx0
+        with pytest.raises(KeyError, match=key):
+            idx0.get_loc(key)
+
+        assert "2017-09" in idx0
+
+        assert p3 not in idx0
+
+    def test_contains_freq_mismatch(self):
+        rng = period_range("2007-01", freq="M", periods=10)
+
+        assert Period("2007-01", freq="M") in rng
+        assert Period("2007-01", freq="D") not in rng
+        assert Period("2007-01", freq="2M") not in rng
+
+    def test_contains_nat(self):
+        # see gh-13582
+        idx = period_range("2007-01", freq="M", periods=10)
+        assert NaT not in idx
+        assert None not in idx
+        assert float("nan") not in idx
+        assert np.nan not in idx
+
+        idx = PeriodIndex(["2011-01", "NaT", "2011-02"], freq="M")
+        assert NaT in idx
+        assert None in idx
+        assert float("nan") in idx
+        assert np.nan in idx
+
+
+class TestAsOfLocs:
+    def test_asof_locs_mismatched_type(self):
+        dti = date_range("2016-01-01", periods=3)
+        pi = dti.to_period("D")
+        pi2 = dti.to_period("h")
+
+        mask = np.array([0, 1, 0], dtype=bool)
+
+        msg = "must be DatetimeIndex or PeriodIndex"
+        with pytest.raises(TypeError, match=msg):
+            pi.asof_locs(pd.Index(pi.asi8, dtype=np.int64), mask)
+
+        with pytest.raises(TypeError, match=msg):
+            pi.asof_locs(pd.Index(pi.asi8, dtype=np.float64), mask)
+
+        with pytest.raises(TypeError, match=msg):
+            # TimedeltaIndex
+            pi.asof_locs(dti - dti, mask)
+
+        msg = "Input has different freq=h"
+        with pytest.raises(libperiod.IncompatibleFrequency, match=msg):
+            pi.asof_locs(pi2, mask)
diff --git a/pandas/pandas/tests/indexes/period/test_join.py b/pandas/pandas/tests/indexes/period/test_join.py
new file mode 100644
index 0000000000000..3e659c1a63266
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_join.py
@@ -0,0 +1,58 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import IncompatibleFrequency
+
+from pandas import (
+    DataFrame,
+    Index,
+    PeriodIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestJoin:
+    def test_join_outer_indexer(self):
+        pi = period_range("1/1/2000", "1/20/2000", freq="D")
+
+        result = pi._outer_indexer(pi)
+        tm.assert_extension_array_equal(result[0], pi._values)
+        tm.assert_numpy_array_equal(result[1], np.arange(len(pi), dtype=np.intp))
+        tm.assert_numpy_array_equal(result[2], np.arange(len(pi), dtype=np.intp))
+
+    def test_joins(self, join_type):
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+
+        joined = index.join(index[:-5], how=join_type)
+
+        assert isinstance(joined, PeriodIndex)
+        assert joined.freq == index.freq
+
+    def test_join_self(self, join_type):
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+
+        res = index.join(index, how=join_type)
+        assert index is res
+
+    def test_join_does_not_recur(self):
+        df = DataFrame(
+            np.ones((3, 2)),
+            index=date_range("2020-01-01", periods=3),
+            columns=period_range("2020-01-01", periods=2),
+        )
+        ser = df.iloc[:2, 0]
+
+        res = ser.index.join(df.columns, how="outer")
+        expected = Index(
+            [ser.index[0], ser.index[1], df.columns[0], df.columns[1]], object
+        )
+        tm.assert_index_equal(res, expected)
+
+    def test_join_mismatched_freq_raises(self):
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+        index3 = period_range("1/1/2000", "1/20/2000", freq="2D")
+        msg = r".*Input has different freq=2D from Period\(freq=D\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            index.join(index3)
diff --git a/pandas/pandas/tests/indexes/period/test_monotonic.py b/pandas/pandas/tests/indexes/period/test_monotonic.py
new file mode 100644
index 0000000000000..15cb8f71cdcf3
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_monotonic.py
@@ -0,0 +1,42 @@
+from pandas import (
+    Period,
+    PeriodIndex,
+)
+
+
+def test_is_monotonic_increasing():
+    # GH#17717
+    p0 = Period("2017-09-01")
+    p1 = Period("2017-09-02")
+    p2 = Period("2017-09-03")
+
+    idx_inc0 = PeriodIndex([p0, p1, p2])
+    idx_inc1 = PeriodIndex([p0, p1, p1])
+    idx_dec0 = PeriodIndex([p2, p1, p0])
+    idx_dec1 = PeriodIndex([p2, p1, p1])
+    idx = PeriodIndex([p1, p2, p0])
+
+    assert idx_inc0.is_monotonic_increasing is True
+    assert idx_inc1.is_monotonic_increasing is True
+    assert idx_dec0.is_monotonic_increasing is False
+    assert idx_dec1.is_monotonic_increasing is False
+    assert idx.is_monotonic_increasing is False
+
+
+def test_is_monotonic_decreasing():
+    # GH#17717
+    p0 = Period("2017-09-01")
+    p1 = Period("2017-09-02")
+    p2 = Period("2017-09-03")
+
+    idx_inc0 = PeriodIndex([p0, p1, p2])
+    idx_inc1 = PeriodIndex([p0, p1, p1])
+    idx_dec0 = PeriodIndex([p2, p1, p0])
+    idx_dec1 = PeriodIndex([p2, p1, p1])
+    idx = PeriodIndex([p1, p2, p0])
+
+    assert idx_inc0.is_monotonic_decreasing is False
+    assert idx_inc1.is_monotonic_decreasing is False
+    assert idx_dec0.is_monotonic_decreasing is True
+    assert idx_dec1.is_monotonic_decreasing is True
+    assert idx.is_monotonic_decreasing is False
diff --git a/pandas/pandas/tests/indexes/period/test_partial_slicing.py b/pandas/pandas/tests/indexes/period/test_partial_slicing.py
new file mode 100644
index 0000000000000..8d173d850583f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_partial_slicing.py
@@ -0,0 +1,192 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    PeriodIndex,
+    Series,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodIndex:
+    def test_getitem_periodindex_duplicates_string_slice(self):
+        # monotonic
+        idx = PeriodIndex([2000, 2007, 2007, 2009, 2009], freq="Y-JUN")
+        ts = Series(np.random.default_rng(2).standard_normal(len(idx)), index=idx)
+        original = ts.copy()
+
+        result = ts["2007"]
+        expected = ts[1:3]
+        tm.assert_series_equal(result, expected)
+        result[:] = 1
+        tm.assert_series_equal(ts, original)
+
+        # not monotonic
+        idx = PeriodIndex([2000, 2007, 2007, 2009, 2007], freq="Y-JUN")
+        ts = Series(np.random.default_rng(2).standard_normal(len(idx)), index=idx)
+
+        result = ts["2007"]
+        expected = ts[idx == "2007"]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_periodindex_quarter_string(self):
+        pi = PeriodIndex(["2Q05", "3Q05", "4Q05", "1Q06", "2Q06"], freq="Q")
+        ser = Series(np.random.default_rng(2).random(len(pi)), index=pi).cumsum()
+        # Todo: fix these accessors!
+        assert ser["05Q4"] == ser.iloc[2]
+
+    def test_pindex_slice_index(self):
+        pi = period_range(start="1/1/10", end="12/31/12", freq="M")
+        s = Series(np.random.default_rng(2).random(len(pi)), index=pi)
+        res = s["2010"]
+        exp = s[0:12]
+        tm.assert_series_equal(res, exp)
+        res = s["2011"]
+        exp = s[12:24]
+        tm.assert_series_equal(res, exp)
+
+    @pytest.mark.parametrize("make_range", [date_range, period_range])
+    def test_range_slice_day(self, make_range):
+        # GH#6716
+        idx = make_range(start="2013/01/01", freq="D", periods=400)
+
+        msg = "slice indices must be integers or None or have an __index__ method"
+        # slices against index should raise IndexError
+        values = [
+            "2014",
+            "2013/02",
+            "2013/01/02",
+            "2013/02/01 9H",
+            "2013/02/01 09:00",
+        ]
+        for v in values:
+            with pytest.raises(TypeError, match=msg):
+                idx[v:]
+
+        s = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+
+        tm.assert_series_equal(s["2013/01/02":], s[1:])
+        tm.assert_series_equal(s["2013/01/02":"2013/01/05"], s[1:5])
+        tm.assert_series_equal(s["2013/02":], s[31:])
+        tm.assert_series_equal(s["2014":], s[365:])
+
+        invalid = ["2013/02/01 9H", "2013/02/01 09:00"]
+        for v in invalid:
+            with pytest.raises(TypeError, match=msg):
+                idx[v:]
+
+    @pytest.mark.parametrize("make_range", [date_range, period_range])
+    def test_range_slice_seconds(self, make_range):
+        # GH#6716
+        idx = make_range(start="2013/01/01 09:00:00", freq="s", periods=4000)
+        msg = "slice indices must be integers or None or have an __index__ method"
+
+        # slices against index should raise IndexError
+        values = [
+            "2014",
+            "2013/02",
+            "2013/01/02",
+            "2013/02/01 9H",
+            "2013/02/01 09:00",
+        ]
+        for v in values:
+            with pytest.raises(TypeError, match=msg):
+                idx[v:]
+
+        s = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+
+        tm.assert_series_equal(s["2013/01/01 09:05":"2013/01/01 09:10"], s[300:660])
+        tm.assert_series_equal(s["2013/01/01 10:00":"2013/01/01 10:05"], s[3600:3960])
+        tm.assert_series_equal(s["2013/01/01 10H":], s[3600:])
+        tm.assert_series_equal(s[:"2013/01/01 09:30"], s[:1860])
+        for d in ["2013/01/01", "2013/01", "2013"]:
+            tm.assert_series_equal(s[d:], s)
+
+    @pytest.mark.parametrize("make_range", [date_range, period_range])
+    def test_range_slice_outofbounds(self, make_range):
+        # GH#5407
+        idx = make_range(start="2013/10/01", freq="D", periods=10)
+
+        df = DataFrame({"units": [100 + i for i in range(10)]}, index=idx)
+        empty = DataFrame(index=idx[:0], columns=["units"])
+        empty["units"] = empty["units"].astype("int64")
+
+        tm.assert_frame_equal(df["2013/09/01":"2013/09/30"], empty)
+        tm.assert_frame_equal(df["2013/09/30":"2013/10/02"], df.iloc[:2])
+        tm.assert_frame_equal(df["2013/10/01":"2013/10/02"], df.iloc[:2])
+        tm.assert_frame_equal(df["2013/10/02":"2013/09/30"], empty)
+        tm.assert_frame_equal(df["2013/10/15":"2013/10/17"], empty)
+        tm.assert_frame_equal(df["2013-06":"2013-09"], empty)
+        tm.assert_frame_equal(df["2013-11":"2013-12"], empty)
+
+    @pytest.mark.parametrize("make_range", [date_range, period_range])
+    def test_maybe_cast_slice_bound(self, make_range, frame_or_series):
+        idx = make_range(start="2013/10/01", freq="D", periods=10)
+
+        obj = DataFrame({"units": [100 + i for i in range(10)]}, index=idx)
+        obj = tm.get_obj(obj, frame_or_series)
+
+        msg = (
+            f"cannot do slice indexing on {type(idx).__name__} with "
+            r"these indexers \[foo\] of type str"
+        )
+
+        # Check the lower-level calls are raising where expected.
+        with pytest.raises(TypeError, match=msg):
+            idx._maybe_cast_slice_bound("foo", "left")
+        with pytest.raises(TypeError, match=msg):
+            idx.get_slice_bound("foo", "left")
+
+        with pytest.raises(TypeError, match=msg):
+            obj["2013/09/30":"foo"]
+        with pytest.raises(TypeError, match=msg):
+            obj["foo":"2013/09/30"]
+        with pytest.raises(TypeError, match=msg):
+            obj.loc["2013/09/30":"foo"]
+        with pytest.raises(TypeError, match=msg):
+            obj.loc["foo":"2013/09/30"]
+
+    def test_partial_slice_doesnt_require_monotonicity(self):
+        # See also: DatetimeIndex test ofm the same name
+        dti = date_range("2014-01-01", periods=30, freq="30D")
+        pi = dti.to_period("D")
+
+        ser_montonic = Series(np.arange(30), index=pi)
+
+        shuffler = list(range(0, 30, 2)) + list(range(1, 31, 2))
+        ser = ser_montonic.iloc[shuffler]
+        nidx = ser.index
+
+        # Manually identified locations of year==2014
+        indexer_2014 = np.array(
+            [0, 1, 2, 3, 4, 5, 6, 15, 16, 17, 18, 19, 20], dtype=np.intp
+        )
+        assert (nidx[indexer_2014].year == 2014).all()
+        assert not (nidx[~indexer_2014].year == 2014).any()
+
+        result = nidx.get_loc("2014")
+        tm.assert_numpy_array_equal(result, indexer_2014)
+
+        expected = ser.iloc[indexer_2014]
+        result = ser.loc["2014"]
+        tm.assert_series_equal(result, expected)
+
+        result = ser["2014"]
+        tm.assert_series_equal(result, expected)
+
+        # Manually identified locations where ser.index is within Mat 2015
+        indexer_may2015 = np.array([23], dtype=np.intp)
+        assert nidx[23].year == 2015 and nidx[23].month == 5
+
+        result = nidx.get_loc("May 2015")
+        tm.assert_numpy_array_equal(result, indexer_may2015)
+
+        expected = ser.iloc[indexer_may2015]
+        result = ser.loc["May 2015"]
+        tm.assert_series_equal(result, expected)
+
+        result = ser["May 2015"]
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/period/test_period.py b/pandas/pandas/tests/indexes/period/test_period.py
new file mode 100644
index 0000000000000..77b8e76894647
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_period.py
@@ -0,0 +1,231 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    NaT,
+    Period,
+    PeriodIndex,
+    Series,
+    date_range,
+    offsets,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodIndex:
+    def test_view_asi8(self):
+        idx = PeriodIndex([], freq="M")
+
+        exp = np.array([], dtype=np.int64)
+        tm.assert_numpy_array_equal(idx.view("i8"), exp)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+        idx = PeriodIndex(["2011-01", NaT], freq="M")
+
+        exp = np.array([492, -9223372036854775808], dtype=np.int64)
+        tm.assert_numpy_array_equal(idx.view("i8"), exp)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+        exp = np.array([14975, -9223372036854775808], dtype=np.int64)
+        idx = PeriodIndex(["2011-01-01", NaT], freq="D")
+        tm.assert_numpy_array_equal(idx.view("i8"), exp)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+    def test_values(self):
+        idx = PeriodIndex([], freq="M")
+
+        exp = np.array([], dtype=object)
+        tm.assert_numpy_array_equal(idx.values, exp)
+        tm.assert_numpy_array_equal(idx.to_numpy(), exp)
+
+        exp = np.array([], dtype=np.int64)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+        idx = PeriodIndex(["2011-01", NaT], freq="M")
+
+        exp = np.array([Period("2011-01", freq="M"), NaT], dtype=object)
+        tm.assert_numpy_array_equal(idx.values, exp)
+        tm.assert_numpy_array_equal(idx.to_numpy(), exp)
+        exp = np.array([492, -9223372036854775808], dtype=np.int64)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+        idx = PeriodIndex(["2011-01-01", NaT], freq="D")
+
+        exp = np.array([Period("2011-01-01", freq="D"), NaT], dtype=object)
+        tm.assert_numpy_array_equal(idx.values, exp)
+        tm.assert_numpy_array_equal(idx.to_numpy(), exp)
+        exp = np.array([14975, -9223372036854775808], dtype=np.int64)
+        tm.assert_numpy_array_equal(idx.asi8, exp)
+
+    @pytest.mark.parametrize(
+        "field",
+        [
+            "year",
+            "month",
+            "day",
+            "hour",
+            "minute",
+            "second",
+            "weekofyear",
+            "week",
+            "dayofweek",
+            "day_of_week",
+            "dayofyear",
+            "day_of_year",
+            "quarter",
+            "qyear",
+            "days_in_month",
+        ],
+    )
+    @pytest.mark.parametrize(
+        "periodindex",
+        [
+            period_range(freq="Y", start="1/1/2001", end="12/1/2005"),
+            period_range(freq="Q", start="1/1/2001", end="12/1/2002"),
+            period_range(freq="M", start="1/1/2001", end="1/1/2002"),
+            period_range(freq="D", start="12/1/2001", end="6/1/2001"),
+            period_range(freq="h", start="12/31/2001", end="1/1/2002 23:00"),
+            period_range(freq="Min", start="12/31/2001", end="1/1/2002 00:20"),
+            period_range(
+                freq="s", start="12/31/2001 00:00:00", end="12/31/2001 00:05:00"
+            ),
+            period_range(end=Period("2006-12-31", "W"), periods=10),
+        ],
+    )
+    def test_fields(self, periodindex, field):
+        periods = list(periodindex)
+        ser = Series(periodindex)
+
+        field_idx = getattr(periodindex, field)
+        assert len(periodindex) == len(field_idx)
+        for x, val in zip(periods, field_idx):
+            assert getattr(x, field) == val
+
+        if len(ser) == 0:
+            return
+
+        field_s = getattr(ser.dt, field)
+        assert len(periodindex) == len(field_s)
+        for x, val in zip(periods, field_s):
+            assert getattr(x, field) == val
+
+    def test_is_(self):
+        create_index = lambda: period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        index = create_index()
+        assert index.is_(index)
+        assert not index.is_(create_index())
+        assert index.is_(index.view())
+        assert index.is_(index.view().view().view().view().view())
+        assert index.view().is_(index)
+        ind2 = index.view()
+        index.name = "Apple"
+        assert ind2.is_(index)
+        assert not index.is_(index[:])
+        assert not index.is_(index.asfreq("M"))
+        assert not index.is_(index.asfreq("Y"))
+
+        assert not index.is_(index - 2)
+        assert not index.is_(index - 0)
+
+    def test_index_unique(self):
+        idx = PeriodIndex([2000, 2007, 2007, 2009, 2009], freq="Y-JUN")
+        expected = PeriodIndex([2000, 2007, 2009], freq="Y-JUN")
+        tm.assert_index_equal(idx.unique(), expected)
+        assert idx.nunique() == 3
+
+    def test_pindex_fieldaccessor_nat(self):
+        idx = PeriodIndex(
+            ["2011-01", "2011-02", "NaT", "2012-03", "2012-04"], freq="D", name="name"
+        )
+
+        exp = Index([2011, 2011, -1, 2012, 2012], dtype=np.int64, name="name")
+        tm.assert_index_equal(idx.year, exp)
+        exp = Index([1, 2, -1, 3, 4], dtype=np.int64, name="name")
+        tm.assert_index_equal(idx.month, exp)
+
+    def test_pindex_multiples(self):
+        expected = PeriodIndex(
+            ["2011-01", "2011-03", "2011-05", "2011-07", "2011-09", "2011-11"],
+            freq="2M",
+        )
+
+        pi = period_range(start="1/1/11", end="12/31/11", freq="2M")
+        tm.assert_index_equal(pi, expected)
+        assert pi.freq == offsets.MonthEnd(2)
+        assert pi.freqstr == "2M"
+
+        pi = period_range(start="1/1/11", periods=6, freq="2M")
+        tm.assert_index_equal(pi, expected)
+        assert pi.freq == offsets.MonthEnd(2)
+        assert pi.freqstr == "2M"
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    @pytest.mark.filterwarnings("ignore:Period with BDay freq:FutureWarning")
+    def test_iteration(self):
+        index = period_range(start="1/1/10", periods=4, freq="B")
+
+        result = list(index)
+        assert isinstance(result[0], Period)
+        assert result[0].freq == index.freq
+
+    def test_with_multi_index(self):
+        # #1705
+        index = date_range("1/1/2012", periods=4, freq="12h")
+        index_as_arrays = [index.to_period(freq="D"), index.hour]
+
+        s = Series([0, 1, 2, 3], index_as_arrays)
+
+        assert isinstance(s.index.levels[0], PeriodIndex)
+
+        assert isinstance(s.index.values[0][0], Period)
+
+    def test_map(self):
+        # test_map_dictlike generally tests
+
+        index = PeriodIndex([2005, 2007, 2009], freq="Y")
+        result = index.map(lambda x: x.ordinal)
+        exp = Index([x.ordinal for x in index])
+        tm.assert_index_equal(result, exp)
+
+
+def test_maybe_convert_timedelta():
+    pi = PeriodIndex(["2000", "2001"], freq="D")
+    offset = offsets.Day(2)
+    assert pi._maybe_convert_timedelta(offset) == 2
+    assert pi._maybe_convert_timedelta(2) == 2
+
+    offset = offsets.BusinessDay()
+    msg = r"Input has different freq=B from PeriodIndex\(freq=D\)"
+    with pytest.raises(ValueError, match=msg):
+        pi._maybe_convert_timedelta(offset)
+
+
+@pytest.mark.parametrize("array", [True, False])
+def test_dunder_array(array):
+    obj = PeriodIndex(["2000-01-01", "2001-01-01"], freq="D")
+    if array:
+        obj = obj._data
+
+    expected = np.array([obj[0], obj[1]], dtype=object)
+    result = np.array(obj)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = np.asarray(obj)
+    tm.assert_numpy_array_equal(result, expected)
+
+    expected = obj.asi8
+    for dtype in ["i8", "int64", np.int64]:
+        result = np.array(obj, dtype=dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.asarray(obj, dtype=dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+    for dtype in ["float64", "int32", "uint64"]:
+        msg = "argument must be"
+        with pytest.raises(TypeError, match=msg):
+            np.array(obj, dtype=dtype)
+        with pytest.raises(TypeError, match=msg):
+            np.array(obj, dtype=getattr(np, dtype))
diff --git a/pandas/pandas/tests/indexes/period/test_period_range.py b/pandas/pandas/tests/indexes/period/test_period_range.py
new file mode 100644
index 0000000000000..51b03024ce272
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_period_range.py
@@ -0,0 +1,237 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NaT,
+    Period,
+    PeriodIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodRangeKeywords:
+    def test_required_arguments(self):
+        msg = (
+            "Of the three parameters: start, end, and periods, exactly two "
+            "must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            period_range("2011-1-1", "2012-1-1", "B")
+
+    def test_required_arguments2(self):
+        start = Period("02-Apr-2005", "D")
+        msg = (
+            "Of the three parameters: start, end, and periods, exactly two "
+            "must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            period_range(start=start)
+
+    def test_required_arguments3(self):
+        # not enough params
+        msg = (
+            "Of the three parameters: start, end, and periods, "
+            "exactly two must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            period_range(start="2017Q1")
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(end="2017Q1")
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(periods=5)
+
+        with pytest.raises(ValueError, match=msg):
+            period_range()
+
+    def test_required_arguments_too_many(self):
+        msg = (
+            "Of the three parameters: start, end, and periods, "
+            "exactly two must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            period_range(start="2017Q1", end="2018Q1", periods=8, freq="Q")
+
+    def test_start_end_non_nat(self):
+        # start/end NaT
+        msg = "start and end must not be NaT"
+        with pytest.raises(ValueError, match=msg):
+            period_range(start=NaT, end="2018Q1")
+        with pytest.raises(ValueError, match=msg):
+            period_range(start=NaT, end="2018Q1", freq="Q")
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(start="2017Q1", end=NaT)
+        with pytest.raises(ValueError, match=msg):
+            period_range(start="2017Q1", end=NaT, freq="Q")
+
+    def test_periods_requires_integer(self):
+        # invalid periods param
+        msg = "periods must be an integer, got foo"
+        with pytest.raises(TypeError, match=msg):
+            period_range(start="2017Q1", periods="foo")
+
+
+class TestPeriodRange:
+    @pytest.mark.parametrize(
+        "freq_offset, freq_period",
+        [
+            ("D", "D"),
+            ("W", "W"),
+            ("QE", "Q"),
+            ("YE", "Y"),
+        ],
+    )
+    def test_construction_from_string(self, freq_offset, freq_period):
+        # non-empty
+        expected = date_range(
+            start="2017-01-01", periods=5, freq=freq_offset, name="foo"
+        ).to_period()
+        start, end = str(expected[0]), str(expected[-1])
+
+        result = period_range(start=start, end=end, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(start=start, periods=5, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(end=end, periods=5, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # empty
+        expected = PeriodIndex([], freq=freq_period, name="foo")
+
+        result = period_range(start=start, periods=0, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(end=end, periods=0, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(start=end, end=start, freq=freq_period, name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_construction_from_string_monthly(self):
+        # non-empty
+        expected = date_range(
+            start="2017-01-01", periods=5, freq="ME", name="foo"
+        ).to_period()
+        start, end = str(expected[0]), str(expected[-1])
+
+        result = period_range(start=start, end=end, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(start=start, periods=5, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(end=end, periods=5, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # empty
+        expected = PeriodIndex([], freq="M", name="foo")
+
+        result = period_range(start=start, periods=0, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(end=end, periods=0, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(start=end, end=start, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_construction_from_period(self):
+        # upsampling
+        start, end = Period("2017Q1", freq="Q"), Period("2018Q1", freq="Q")
+        expected = date_range(
+            start="2017-03-31", end="2018-03-31", freq="ME", name="foo"
+        ).to_period()
+        result = period_range(start=start, end=end, freq="M", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # downsampling
+        start = Period("2017-1", freq="M")
+        end = Period("2019-12", freq="M")
+        expected = date_range(
+            start="2017-01-31", end="2019-12-31", freq="QE", name="foo"
+        ).to_period()
+        result = period_range(start=start, end=end, freq="Q", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # test for issue # 21793
+        start = Period("2017Q1", freq="Q")
+        end = Period("2018Q1", freq="Q")
+        idx = period_range(start=start, end=end, freq="Q", name="foo")
+        result = idx == idx.values
+        expected = np.array([True, True, True, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        # empty
+        expected = PeriodIndex([], freq="W", name="foo")
+
+        result = period_range(start=start, periods=0, freq="W", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(end=end, periods=0, freq="W", name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = period_range(start=end, end=start, freq="W", name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_mismatched_start_end_freq_raises(self):
+        depr_msg = "Period with BDay freq is deprecated"
+
+        end_w = Period("2006-12-31", "1W")
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            start_b = Period("02-Apr-2005", "B")
+            end_b = Period("2005-05-01", "B")
+
+        msg = "start and end must have same freq"
+        with pytest.raises(ValueError, match=msg):
+            with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+                period_range(start=start_b, end=end_w)
+
+        # without mismatch we are OK
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            period_range(start=start_b, end=end_b)
+
+
+class TestPeriodRangeDisallowedFreqs:
+    def test_constructor_U(self):
+        # U was used as undefined period
+        with pytest.raises(ValueError, match="Invalid frequency: X"):
+            period_range("2007-1-1", periods=500, freq="X")
+
+    @pytest.mark.parametrize("freq_depr", ["2MIN", "2US", "2NS"])
+    def test_uppercase_freq_deprecated_from_time_series(self, freq_depr):
+        # GH#52536, GH#54939
+        msg = f"'{freq_depr[1:]}' is deprecated and will be removed in a "
+        f"future version. Please use '{freq_depr.lower()[1:]}' instead."
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            period_range("2020-01-01 00:00:00 00:00", periods=2, freq=freq_depr)
+
+    @pytest.mark.parametrize("freq", ["2m", "2q-sep", "2y", "2H", "2S"])
+    def test_incorrect_case_freq_from_time_series_raises(self, freq):
+        # GH#52536, GH#54939, GH#59143
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(freq=freq, start="1/1/2001", end="12/1/2009")
+
+    @pytest.mark.parametrize("freq", ["2A", "2a", "2A-AUG", "2A-aug"])
+    def test_A_raises_from_time_series(self, freq):
+        msg = f"Invalid frequency: {freq}"
+
+        with pytest.raises(ValueError, match=msg):
+            period_range(freq=freq, start="1/1/2001", end="12/1/2009")
+
+    @pytest.mark.parametrize("freq", ["2w"])
+    def test_lowercase_freq_from_time_series_deprecated(self, freq):
+        # GH#52536, GH#54939
+        msg = f"'{freq[1:]}' is deprecated and will be removed in a "
+        f"future version. Please use '{freq.upper()[1:]}' instead."
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            period_range(freq=freq, start="1/1/2001", end="12/1/2009")
diff --git a/pandas/pandas/tests/indexes/period/test_pickle.py b/pandas/pandas/tests/indexes/period/test_pickle.py
new file mode 100644
index 0000000000000..7d359fdabb6f1
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_pickle.py
@@ -0,0 +1,26 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NaT,
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+class TestPickle:
+    @pytest.mark.parametrize("freq", ["D", "M", "Y"])
+    def test_pickle_round_trip(self, freq):
+        idx = PeriodIndex(["2016-05-16", "NaT", NaT, np.nan], freq=freq)
+        result = tm.round_trip_pickle(idx)
+        tm.assert_index_equal(result, idx)
+
+    def test_pickle_freq(self):
+        # GH#2891
+        prng = period_range("1/1/2011", "1/1/2012", freq="M")
+        new_prng = tm.round_trip_pickle(prng)
+        assert new_prng.freq == offsets.MonthEnd()
+        assert new_prng.freqstr == "M"
diff --git a/pandas/pandas/tests/indexes/period/test_resolution.py b/pandas/pandas/tests/indexes/period/test_resolution.py
new file mode 100644
index 0000000000000..680bdaa2e2a44
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_resolution.py
@@ -0,0 +1,23 @@
+import pytest
+
+import pandas as pd
+
+
+class TestResolution:
+    @pytest.mark.parametrize(
+        "freq,expected",
+        [
+            ("Y", "year"),
+            ("Q", "quarter"),
+            ("M", "month"),
+            ("D", "day"),
+            ("h", "hour"),
+            ("min", "minute"),
+            ("s", "second"),
+            ("ms", "millisecond"),
+            ("us", "microsecond"),
+        ],
+    )
+    def test_resolution(self, freq, expected):
+        idx = pd.period_range(start="2013-04-01", periods=30, freq=freq)
+        assert idx.resolution == expected
diff --git a/pandas/pandas/tests/indexes/period/test_scalar_compat.py b/pandas/pandas/tests/indexes/period/test_scalar_compat.py
new file mode 100644
index 0000000000000..d8afd29ff31c5
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_scalar_compat.py
@@ -0,0 +1,38 @@
+"""Tests for PeriodIndex behaving like a vectorized Period scalar"""
+
+import pytest
+
+from pandas import (
+    Timedelta,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodIndexOps:
+    def test_start_time(self):
+        # GH#17157
+        index = period_range(freq="M", start="2016-01-01", end="2016-05-31")
+        expected_index = date_range("2016-01-01", end="2016-05-31", freq="MS")
+        tm.assert_index_equal(index.start_time, expected_index)
+
+    def test_end_time(self):
+        # GH#17157
+        index = period_range(freq="M", start="2016-01-01", end="2016-05-31")
+        expected_index = date_range("2016-01-01", end="2016-05-31", freq="ME")
+        expected_index += Timedelta(1, "D") - Timedelta(1, "ns")
+        tm.assert_index_equal(index.end_time, expected_index)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_end_time_business_friday(self):
+        # GH#34449
+        pi = period_range("1990-01-05", freq="B", periods=1)
+        result = pi.end_time
+
+        dti = date_range("1990-01-05", freq="D", periods=1)._with_freq(None)
+        expected = dti + Timedelta(days=1, nanoseconds=-1)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/period/test_searchsorted.py b/pandas/pandas/tests/indexes/period/test_searchsorted.py
new file mode 100644
index 0000000000000..9b02a2f35fd01
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_searchsorted.py
@@ -0,0 +1,80 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import IncompatibleFrequency
+
+from pandas import (
+    NaT,
+    Period,
+    PeriodIndex,
+)
+import pandas._testing as tm
+
+
+class TestSearchsorted:
+    @pytest.mark.parametrize("freq", ["D", "2D"])
+    def test_searchsorted(self, freq):
+        pidx = PeriodIndex(
+            ["2014-01-01", "2014-01-02", "2014-01-03", "2014-01-04", "2014-01-05"],
+            freq=freq,
+        )
+
+        p1 = Period("2014-01-01", freq=freq)
+        assert pidx.searchsorted(p1) == 0
+
+        p2 = Period("2014-01-04", freq=freq)
+        assert pidx.searchsorted(p2) == 3
+
+        assert pidx.searchsorted(NaT) == 5
+
+        msg = "Input has different freq=h from PeriodArray"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            pidx.searchsorted(Period("2014-01-01", freq="h"))
+
+        msg = "Input has different freq=5D from PeriodArray"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            pidx.searchsorted(Period("2014-01-01", freq="5D"))
+
+    def test_searchsorted_different_argument_classes(self, listlike_box):
+        pidx = PeriodIndex(
+            ["2014-01-01", "2014-01-02", "2014-01-03", "2014-01-04", "2014-01-05"],
+            freq="D",
+        )
+        result = pidx.searchsorted(listlike_box(pidx))
+        expected = np.arange(len(pidx), dtype=result.dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = pidx._data.searchsorted(listlike_box(pidx))
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_searchsorted_invalid(self):
+        pidx = PeriodIndex(
+            ["2014-01-01", "2014-01-02", "2014-01-03", "2014-01-04", "2014-01-05"],
+            freq="D",
+        )
+
+        other = np.array([0, 1], dtype=np.int64)
+
+        msg = "|".join(
+            [
+                "searchsorted requires compatible dtype or scalar",
+                "value should be a 'Period', 'NaT', or array of those. Got",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(other)
+
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(other.astype("timedelta64[ns]"))
+
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(np.timedelta64(4))
+
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(np.timedelta64("NaT", "ms"))
+
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(np.datetime64(4, "ns"))
+
+        with pytest.raises(TypeError, match=msg):
+            pidx.searchsorted(np.datetime64("NaT", "ns"))
diff --git a/pandas/pandas/tests/indexes/period/test_setops.py b/pandas/pandas/tests/indexes/period/test_setops.py
new file mode 100644
index 0000000000000..2fa7e8cd0d2df
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_setops.py
@@ -0,0 +1,363 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    PeriodIndex,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+def _permute(obj):
+    return obj.take(np.random.default_rng(2).permutation(len(obj)))
+
+
+class TestPeriodIndex:
+    def test_union(self, sort):
+        # union
+        other1 = period_range("1/1/2000", freq="D", periods=5)
+        rng1 = period_range("1/6/2000", freq="D", periods=5)
+        expected1 = PeriodIndex(
+            [
+                "2000-01-06",
+                "2000-01-07",
+                "2000-01-08",
+                "2000-01-09",
+                "2000-01-10",
+                "2000-01-01",
+                "2000-01-02",
+                "2000-01-03",
+                "2000-01-04",
+                "2000-01-05",
+            ],
+            freq="D",
+        )
+
+        rng2 = period_range("1/1/2000", freq="D", periods=5)
+        other2 = period_range("1/4/2000", freq="D", periods=5)
+        expected2 = period_range("1/1/2000", freq="D", periods=8)
+
+        rng3 = period_range("1/1/2000", freq="D", periods=5)
+        other3 = PeriodIndex([], freq="D")
+        expected3 = period_range("1/1/2000", freq="D", periods=5)
+
+        rng4 = period_range("2000-01-01 09:00", freq="h", periods=5)
+        other4 = period_range("2000-01-02 09:00", freq="h", periods=5)
+        expected4 = PeriodIndex(
+            [
+                "2000-01-01 09:00",
+                "2000-01-01 10:00",
+                "2000-01-01 11:00",
+                "2000-01-01 12:00",
+                "2000-01-01 13:00",
+                "2000-01-02 09:00",
+                "2000-01-02 10:00",
+                "2000-01-02 11:00",
+                "2000-01-02 12:00",
+                "2000-01-02 13:00",
+            ],
+            freq="h",
+        )
+
+        rng5 = PeriodIndex(
+            ["2000-01-01 09:01", "2000-01-01 09:03", "2000-01-01 09:05"], freq="min"
+        )
+        other5 = PeriodIndex(
+            ["2000-01-01 09:01", "2000-01-01 09:05", "2000-01-01 09:08"], freq="min"
+        )
+        expected5 = PeriodIndex(
+            [
+                "2000-01-01 09:01",
+                "2000-01-01 09:03",
+                "2000-01-01 09:05",
+                "2000-01-01 09:08",
+            ],
+            freq="min",
+        )
+
+        rng6 = period_range("2000-01-01", freq="M", periods=7)
+        other6 = period_range("2000-04-01", freq="M", periods=7)
+        expected6 = period_range("2000-01-01", freq="M", periods=10)
+
+        rng7 = period_range("2003-01-01", freq="Y", periods=5)
+        other7 = period_range("1998-01-01", freq="Y", periods=8)
+        expected7 = PeriodIndex(
+            [
+                "2003",
+                "2004",
+                "2005",
+                "2006",
+                "2007",
+                "1998",
+                "1999",
+                "2000",
+                "2001",
+                "2002",
+            ],
+            freq="Y",
+        )
+
+        rng8 = PeriodIndex(
+            ["1/3/2000", "1/2/2000", "1/1/2000", "1/5/2000", "1/4/2000"], freq="D"
+        )
+        other8 = period_range("1/6/2000", freq="D", periods=5)
+        expected8 = PeriodIndex(
+            [
+                "1/3/2000",
+                "1/2/2000",
+                "1/1/2000",
+                "1/5/2000",
+                "1/4/2000",
+                "1/6/2000",
+                "1/7/2000",
+                "1/8/2000",
+                "1/9/2000",
+                "1/10/2000",
+            ],
+            freq="D",
+        )
+
+        for rng, other, expected in [
+            (rng1, other1, expected1),
+            (rng2, other2, expected2),
+            (rng3, other3, expected3),
+            (rng4, other4, expected4),
+            (rng5, other5, expected5),
+            (rng6, other6, expected6),
+            (rng7, other7, expected7),
+            (rng8, other8, expected8),
+        ]:
+            result_union = rng.union(other, sort=sort)
+            if sort is None:
+                expected = expected.sort_values()
+            tm.assert_index_equal(result_union, expected)
+
+    def test_union_misc(self, sort):
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+
+        result = index[:-5].union(index[10:], sort=sort)
+        tm.assert_index_equal(result, index)
+
+        # not in order
+        result = _permute(index[:-5]).union(_permute(index[10:]), sort=sort)
+        if sort is False:
+            tm.assert_index_equal(result.sort_values(), index)
+        else:
+            tm.assert_index_equal(result, index)
+
+        # cast if different frequencies
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+        index2 = period_range("1/1/2000", "1/20/2000", freq="W-WED")
+        result = index.union(index2, sort=sort)
+        expected = index.astype(object).union(index2.astype(object), sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection(self, sort):
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+
+        result = index[:-5].intersection(index[10:], sort=sort)
+        tm.assert_index_equal(result, index[10:-5])
+
+        # not in order
+        left = _permute(index[:-5])
+        right = _permute(index[10:])
+        result = left.intersection(right, sort=sort)
+        if sort is False:
+            tm.assert_index_equal(result.sort_values(), index[10:-5])
+        else:
+            tm.assert_index_equal(result, index[10:-5])
+
+        # cast if different frequencies
+        index = period_range("1/1/2000", "1/20/2000", freq="D")
+        index2 = period_range("1/1/2000", "1/20/2000", freq="W-WED")
+
+        result = index.intersection(index2, sort=sort)
+        expected = pd.Index([], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        index3 = period_range("1/1/2000", "1/20/2000", freq="2D")
+        result = index.intersection(index3, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_cases(self, sort):
+        base = period_range("6/1/2000", "6/30/2000", freq="D", name="idx")
+
+        # if target has the same name, it is preserved
+        rng2 = period_range("5/15/2000", "6/20/2000", freq="D", name="idx")
+        expected2 = period_range("6/1/2000", "6/20/2000", freq="D", name="idx")
+
+        # if target name is different, it will be reset
+        rng3 = period_range("5/15/2000", "6/20/2000", freq="D", name="other")
+        expected3 = period_range("6/1/2000", "6/20/2000", freq="D", name=None)
+
+        rng4 = period_range("7/1/2000", "7/31/2000", freq="D", name="idx")
+        expected4 = PeriodIndex([], name="idx", freq="D")
+
+        for rng, expected in [
+            (rng2, expected2),
+            (rng3, expected3),
+            (rng4, expected4),
+        ]:
+            result = base.intersection(rng, sort=sort)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+        # non-monotonic
+        base = PeriodIndex(
+            ["2011-01-05", "2011-01-04", "2011-01-02", "2011-01-03"],
+            freq="D",
+            name="idx",
+        )
+
+        rng2 = PeriodIndex(
+            ["2011-01-04", "2011-01-02", "2011-02-02", "2011-02-03"],
+            freq="D",
+            name="idx",
+        )
+        expected2 = PeriodIndex(["2011-01-04", "2011-01-02"], freq="D", name="idx")
+
+        rng3 = PeriodIndex(
+            ["2011-01-04", "2011-01-02", "2011-02-02", "2011-02-03"],
+            freq="D",
+            name="other",
+        )
+        expected3 = PeriodIndex(["2011-01-04", "2011-01-02"], freq="D", name=None)
+
+        rng4 = period_range("7/1/2000", "7/31/2000", freq="D", name="idx")
+        expected4 = PeriodIndex([], freq="D", name="idx")
+
+        for rng, expected in [
+            (rng2, expected2),
+            (rng3, expected3),
+            (rng4, expected4),
+        ]:
+            result = base.intersection(rng, sort=sort)
+            if sort is None:
+                expected = expected.sort_values()
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == "D"
+
+        # empty same freq
+        rng = date_range("6/1/2000", "6/15/2000", freq="min")
+        result = rng[0:0].intersection(rng)
+        assert len(result) == 0
+
+        result = rng.intersection(rng[0:0])
+        assert len(result) == 0
+
+    def test_difference(self, sort):
+        # diff
+        period_rng = ["1/3/2000", "1/2/2000", "1/1/2000", "1/5/2000", "1/4/2000"]
+        rng1 = PeriodIndex(period_rng, freq="D")
+        other1 = period_range("1/6/2000", freq="D", periods=5)
+        expected1 = rng1
+
+        rng2 = PeriodIndex(period_rng, freq="D")
+        other2 = period_range("1/4/2000", freq="D", periods=5)
+        expected2 = PeriodIndex(["1/3/2000", "1/2/2000", "1/1/2000"], freq="D")
+
+        rng3 = PeriodIndex(period_rng, freq="D")
+        other3 = PeriodIndex([], freq="D")
+        expected3 = rng3
+
+        period_rng = [
+            "2000-01-01 10:00",
+            "2000-01-01 09:00",
+            "2000-01-01 12:00",
+            "2000-01-01 11:00",
+            "2000-01-01 13:00",
+        ]
+        rng4 = PeriodIndex(period_rng, freq="h")
+        other4 = period_range("2000-01-02 09:00", freq="h", periods=5)
+        expected4 = rng4
+
+        rng5 = PeriodIndex(
+            ["2000-01-01 09:03", "2000-01-01 09:01", "2000-01-01 09:05"], freq="min"
+        )
+        other5 = PeriodIndex(["2000-01-01 09:01", "2000-01-01 09:05"], freq="min")
+        expected5 = PeriodIndex(["2000-01-01 09:03"], freq="min")
+
+        period_rng = [
+            "2000-02-01",
+            "2000-01-01",
+            "2000-06-01",
+            "2000-07-01",
+            "2000-05-01",
+            "2000-03-01",
+            "2000-04-01",
+        ]
+        rng6 = PeriodIndex(period_rng, freq="M")
+        other6 = period_range("2000-04-01", freq="M", periods=7)
+        expected6 = PeriodIndex(["2000-02-01", "2000-01-01", "2000-03-01"], freq="M")
+
+        period_rng = ["2003", "2007", "2006", "2005", "2004"]
+        rng7 = PeriodIndex(period_rng, freq="Y")
+        other7 = period_range("1998-01-01", freq="Y", periods=8)
+        expected7 = PeriodIndex(["2007", "2006"], freq="Y")
+
+        for rng, other, expected in [
+            (rng1, other1, expected1),
+            (rng2, other2, expected2),
+            (rng3, other3, expected3),
+            (rng4, other4, expected4),
+            (rng5, other5, expected5),
+            (rng6, other6, expected6),
+            (rng7, other7, expected7),
+        ]:
+            result_difference = rng.difference(other, sort=sort)
+            if sort is None and len(other):
+                # We dont sort (yet?) when empty GH#24959
+                expected = expected.sort_values()
+            tm.assert_index_equal(result_difference, expected)
+
+    def test_difference_freq(self, sort):
+        # GH14323: difference of Period MUST preserve frequency
+        # but the ability to union results must be preserved
+
+        index = period_range("20160920", "20160925", freq="D")
+
+        other = period_range("20160921", "20160924", freq="D")
+        expected = PeriodIndex(["20160920", "20160925"], freq="D")
+        idx_diff = index.difference(other, sort)
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+        other = period_range("20160922", "20160925", freq="D")
+        idx_diff = index.difference(other, sort)
+        expected = PeriodIndex(["20160920", "20160921"], freq="D")
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+    def test_intersection_equal_duplicates(self):
+        # GH#38302
+        idx = period_range("2011-01-01", periods=2)
+        idx_dup = idx.append(idx)
+        result = idx_dup.intersection(idx_dup)
+        tm.assert_index_equal(result, idx)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_union_duplicates(self):
+        # GH#36289
+        idx = period_range("2011-01-01", periods=2)
+        idx_dup = idx.append(idx)
+
+        idx2 = period_range("2011-01-02", periods=2)
+        idx2_dup = idx2.append(idx2)
+        result = idx_dup.union(idx2_dup)
+
+        expected = PeriodIndex(
+            [
+                "2011-01-01",
+                "2011-01-01",
+                "2011-01-02",
+                "2011-01-02",
+                "2011-01-03",
+                "2011-01-03",
+            ],
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/period/test_tools.py b/pandas/pandas/tests/indexes/period/test_tools.py
new file mode 100644
index 0000000000000..f507e64d88b06
--- /dev/null
+++ b/pandas/pandas/tests/indexes/period/test_tools.py
@@ -0,0 +1,52 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Period,
+    PeriodIndex,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestPeriodRepresentation:
+    """
+    Wish to match NumPy units
+    """
+
+    @pytest.mark.parametrize(
+        "freq, base_date",
+        [
+            ("W-THU", "1970-01-01"),
+            ("D", "1970-01-01"),
+            ("B", "1970-01-01"),
+            ("h", "1970-01-01"),
+            ("min", "1970-01-01"),
+            ("s", "1970-01-01"),
+            ("ms", "1970-01-01"),
+            ("us", "1970-01-01"),
+            ("ns", "1970-01-01"),
+            ("M", "1970-01"),
+            ("Y", 1970),
+        ],
+    )
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_freq(self, freq, base_date):
+        rng = period_range(start=base_date, periods=10, freq=freq)
+        exp = np.arange(10, dtype=np.int64)
+
+        tm.assert_numpy_array_equal(rng.asi8, exp)
+
+
+class TestPeriodIndexConversion:
+    def test_tolist(self):
+        index = period_range(freq="Y", start="1/1/2001", end="12/1/2009")
+        rs = index.tolist()
+        for x in rs:
+            assert isinstance(x, Period)
+
+        recon = PeriodIndex(rs)
+        tm.assert_index_equal(index, recon)
diff --git a/pandas/pandas/tests/indexes/ranges/__init__.py b/pandas/pandas/tests/indexes/ranges/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/ranges/test_constructors.py b/pandas/pandas/tests/indexes/ranges/test_constructors.py
new file mode 100644
index 0000000000000..5e6f16075ae63
--- /dev/null
+++ b/pandas/pandas/tests/indexes/ranges/test_constructors.py
@@ -0,0 +1,164 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    RangeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestRangeIndexConstructors:
+    @pytest.mark.parametrize("name", [None, "foo"])
+    @pytest.mark.parametrize(
+        "args, kwargs, start, stop, step",
+        [
+            ((5,), {}, 0, 5, 1),
+            ((1, 5), {}, 1, 5, 1),
+            ((1, 5, 2), {}, 1, 5, 2),
+            ((0,), {}, 0, 0, 1),
+            ((0, 0), {}, 0, 0, 1),
+            ((), {"start": 0}, 0, 0, 1),
+            ((), {"stop": 0}, 0, 0, 1),
+        ],
+    )
+    def test_constructor(self, args, kwargs, start, stop, step, name):
+        result = RangeIndex(*args, name=name, **kwargs)
+        expected = Index(np.arange(start, stop, step, dtype=np.int64), name=name)
+        assert isinstance(result, RangeIndex)
+        assert result.name is name
+        assert result._range == range(start, stop, step)
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+    def test_constructor_invalid_args(self):
+        msg = "RangeIndex\\(\\.\\.\\.\\) must be called with integers"
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex()
+
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex(name="Foo")
+
+        # we don't allow on a bare Index
+        msg = (
+            r"Index\(\.\.\.\) must be called with a collection of some "
+            r"kind, 0 was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            Index(0)
+
+    @pytest.mark.parametrize(
+        "args",
+        [
+            Index(["a", "b"]),
+            Series(["a", "b"]),
+            np.array(["a", "b"]),
+            [],
+            np.arange(0, 10),
+            np.array([1]),
+            [1],
+        ],
+    )
+    def test_constructor_additional_invalid_args(self, args):
+        msg = f"Value needs to be a scalar value, was type {type(args).__name__}"
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex(args)
+
+    @pytest.mark.parametrize("args", ["foo", datetime(2000, 1, 1, 0, 0)])
+    def test_constructor_invalid_args_wrong_type(self, args):
+        msg = f"Wrong type {type(args)} for value {args}"
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex(args)
+
+    def test_constructor_same(self):
+        # pass thru w and w/o copy
+        index = RangeIndex(1, 5, 2)
+        result = RangeIndex(index, copy=False)
+        assert result.identical(index)
+
+        result = RangeIndex(index, copy=True)
+        tm.assert_index_equal(result, index, exact=True)
+
+        result = RangeIndex(index)
+        tm.assert_index_equal(result, index, exact=True)
+
+        with pytest.raises(
+            ValueError,
+            match="Incorrect `dtype` passed: expected signed integer, received float64",
+        ):
+            RangeIndex(index, dtype="float64")
+
+    def test_constructor_range_object(self):
+        result = RangeIndex(range(1, 5, 2))
+        expected = RangeIndex(1, 5, 2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_constructor_range(self):
+        result = RangeIndex.from_range(range(1, 5, 2))
+        expected = RangeIndex(1, 5, 2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = RangeIndex.from_range(range(5, 6))
+        expected = RangeIndex(5, 6, 1)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # an invalid range
+        result = RangeIndex.from_range(range(5, 1))
+        expected = RangeIndex(0, 0, 1)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = RangeIndex.from_range(range(5))
+        expected = RangeIndex(0, 5, 1)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = Index(range(1, 5, 2))
+        expected = RangeIndex(1, 5, 2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        msg = (
+            r"(RangeIndex.)?from_range\(\) got an unexpected keyword argument( 'copy')?"
+        )
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex.from_range(range(10), copy=True)
+
+    def test_constructor_name(self):
+        # GH#12288
+        orig = RangeIndex(10)
+        orig.name = "original"
+
+        copy = RangeIndex(orig)
+        copy.name = "copy"
+
+        assert orig.name == "original"
+        assert copy.name == "copy"
+
+        new = Index(copy)
+        assert new.name == "copy"
+
+        new.name = "new"
+        assert orig.name == "original"
+        assert copy.name == "copy"
+        assert new.name == "new"
+
+    def test_constructor_corner(self):
+        arr = np.array([1, 2, 3, 4], dtype=object)
+        index = RangeIndex(1, 5)
+        assert index.values.dtype == np.int64
+        expected = Index(arr).astype("int64")
+
+        tm.assert_index_equal(index, expected, exact="equiv")
+
+        # non-int raise Exception
+        with pytest.raises(TypeError, match=r"Wrong type \<class 'str'\>"):
+            RangeIndex("1", "10", "1")
+        with pytest.raises(TypeError, match=r"Wrong type \<class 'float'\>"):
+            RangeIndex(1.1, 10.2, 1.3)
+
+        # invalid passed type
+        with pytest.raises(
+            ValueError,
+            match="Incorrect `dtype` passed: expected signed integer, received float64",
+        ):
+            RangeIndex(1, 5, dtype="float64")
diff --git a/pandas/pandas/tests/indexes/ranges/test_indexing.py b/pandas/pandas/tests/indexes/ranges/test_indexing.py
new file mode 100644
index 0000000000000..6202074a11d78
--- /dev/null
+++ b/pandas/pandas/tests/indexes/ranges/test_indexing.py
@@ -0,0 +1,137 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+class TestGetIndexer:
+    def test_get_indexer(self):
+        index = RangeIndex(start=0, stop=20, step=2)
+        target = RangeIndex(10)
+        indexer = index.get_indexer(target)
+        expected = np.array([0, -1, 1, -1, 2, -1, 3, -1, 4, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+    def test_get_indexer_pad(self):
+        index = RangeIndex(start=0, stop=20, step=2)
+        target = RangeIndex(10)
+        indexer = index.get_indexer(target, method="pad")
+        expected = np.array([0, 0, 1, 1, 2, 2, 3, 3, 4, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+    def test_get_indexer_backfill(self):
+        index = RangeIndex(start=0, stop=20, step=2)
+        target = RangeIndex(10)
+        indexer = index.get_indexer(target, method="backfill")
+        expected = np.array([0, 1, 1, 2, 2, 3, 3, 4, 4, 5], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected)
+
+    def test_get_indexer_limit(self):
+        # GH#28631
+        idx = RangeIndex(4)
+        target = RangeIndex(6)
+        result = idx.get_indexer(target, method="pad", limit=1)
+        expected = np.array([0, 1, 2, 3, 3, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("stop", [0, -1, -2])
+    def test_get_indexer_decreasing(self, stop):
+        # GH#28678
+        index = RangeIndex(7, stop, -3)
+        result = index.get_indexer(range(9))
+        expected = np.array([-1, 2, -1, -1, 1, -1, -1, 0, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestTake:
+    def test_take_preserve_name(self):
+        index = RangeIndex(1, 5, name="foo")
+        taken = index.take([3, 0, 1])
+        assert index.name == taken.name
+
+    def test_take_fill_value(self):
+        # GH#12631
+        idx = RangeIndex(1, 4, name="xxx")
+        result = idx.take(np.array([1, 0, -1]))
+        expected = Index([2, 1, 3], dtype=np.int64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        msg = "Unable to fill values because RangeIndex cannot contain NA"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -1]), fill_value=True)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = Index([2, 1, 3], dtype=np.int64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        msg = "Unable to fill values because RangeIndex cannot contain NA"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+    def test_take_raises_index_error(self):
+        idx = RangeIndex(1, 4, name="xxx")
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+        msg = "index -4 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -4]))
+
+        # no errors
+        result = idx.take(np.array([1, -3]))
+        expected = Index([2, 1], dtype=np.int64, name="xxx")
+        tm.assert_index_equal(result, expected)
+
+    def test_take_accepts_empty_array(self):
+        idx = RangeIndex(1, 4, name="foo")
+        result = idx.take(np.array([]))
+        expected = Index([], dtype=np.int64, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        # empty index
+        idx = RangeIndex(0, name="foo")
+        result = idx.take(np.array([]))
+        expected = Index([], dtype=np.int64, name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_take_accepts_non_int64_array(self):
+        idx = RangeIndex(1, 4, name="foo")
+        result = idx.take(np.array([2, 1], dtype=np.uint32))
+        expected = Index([3, 2], dtype=np.int64, name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_take_when_index_has_step(self):
+        idx = RangeIndex(1, 11, 3, name="foo")  # [1, 4, 7, 10]
+        result = idx.take(np.array([1, 0, -1, -4]))
+        expected = Index([4, 1, 10, 1], dtype=np.int64, name="foo")
+        tm.assert_index_equal(result, expected)
+
+    def test_take_when_index_has_negative_step(self):
+        idx = RangeIndex(11, -4, -2, name="foo")  # [11, 9, 7, 5, 3, 1, -1, -3]
+        result = idx.take(np.array([1, 0, -1, -8]))
+        expected = Index([9, 11, -3, 11], dtype=np.int64, name="foo")
+        tm.assert_index_equal(result, expected)
+
+
+class TestWhere:
+    def test_where_putmask_range_cast(self):
+        # GH#43240
+        idx = RangeIndex(0, 5, name="test")
+
+        mask = np.array([True, True, False, False, False])
+        result = idx.putmask(mask, 10)
+        expected = Index([10, 10, 2, 3, 4], dtype=np.int64, name="test")
+        tm.assert_index_equal(result, expected)
+
+        result = idx.where(~mask, 10)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/ranges/test_join.py b/pandas/pandas/tests/indexes/ranges/test_join.py
new file mode 100644
index 0000000000000..09db30b1d4c51
--- /dev/null
+++ b/pandas/pandas/tests/indexes/ranges/test_join.py
@@ -0,0 +1,235 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+class TestJoin:
+    def test_join_outer(self):
+        # join with Index[int64]
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index(np.arange(25, 14, -1, dtype=np.int64))
+
+        res, lidx, ridx = index.join(other, how="outer", return_indexers=True)
+        noidx_res = index.join(other, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        eres = Index(
+            [0, 2, 4, 6, 8, 10, 12, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25]
+        )
+        elidx = np.array(
+            [0, 1, 2, 3, 4, 5, 6, 7, -1, 8, -1, 9, -1, -1, -1, -1, -1, -1, -1],
+            dtype=np.intp,
+        )
+        eridx = np.array(
+            [-1, -1, -1, -1, -1, -1, -1, -1, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0],
+            dtype=np.intp,
+        )
+
+        assert isinstance(res, Index) and res.dtype == np.dtype(np.int64)
+        assert not isinstance(res, RangeIndex)
+        tm.assert_index_equal(res, eres, exact=True)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # join with RangeIndex
+        other = RangeIndex(25, 14, -1)
+
+        res, lidx, ridx = index.join(other, how="outer", return_indexers=True)
+        noidx_res = index.join(other, how="outer")
+        tm.assert_index_equal(res, noidx_res)
+
+        assert isinstance(res, Index) and res.dtype == np.int64
+        assert not isinstance(res, RangeIndex)
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_inner(self):
+        # Join with non-RangeIndex
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index(np.arange(25, 14, -1, dtype=np.int64))
+
+        res, lidx, ridx = index.join(other, how="inner", return_indexers=True)
+
+        # no guarantee of sortedness, so sort for comparison purposes
+        ind = res.argsort()
+        res = res.take(ind)
+        lidx = lidx.take(ind)
+        ridx = ridx.take(ind)
+
+        eres = Index([16, 18])
+        elidx = np.array([8, 9], dtype=np.intp)
+        eridx = np.array([9, 7], dtype=np.intp)
+
+        assert isinstance(res, Index) and res.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # Join two RangeIndex
+        other = RangeIndex(25, 14, -1)
+
+        res, lidx, ridx = index.join(other, how="inner", return_indexers=True)
+
+        assert isinstance(res, RangeIndex)
+        tm.assert_index_equal(res, eres, exact="equiv")
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_left(self):
+        # Join with Index[int64]
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index(np.arange(25, 14, -1, dtype=np.int64))
+
+        res, lidx, ridx = index.join(other, how="left", return_indexers=True)
+        eres = index
+        eridx = np.array([-1, -1, -1, -1, -1, -1, -1, -1, 9, 7], dtype=np.intp)
+
+        assert isinstance(res, RangeIndex)
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+        # Join withRangeIndex
+        other = Index(np.arange(25, 14, -1, dtype=np.int64))
+
+        res, lidx, ridx = index.join(other, how="left", return_indexers=True)
+
+        assert isinstance(res, RangeIndex)
+        tm.assert_index_equal(res, eres)
+        assert lidx is None
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_right(self):
+        # Join with Index[int64]
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index(np.arange(25, 14, -1, dtype=np.int64))
+
+        res, lidx, ridx = index.join(other, how="right", return_indexers=True)
+        eres = other
+        elidx = np.array([-1, -1, -1, -1, -1, -1, -1, 9, -1, 8, -1], dtype=np.intp)
+
+        assert isinstance(other, Index) and other.dtype == np.int64
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        assert ridx is None
+
+        # Join withRangeIndex
+        other = RangeIndex(25, 14, -1)
+
+        res, lidx, ridx = index.join(other, how="right", return_indexers=True)
+        eres = other
+
+        assert isinstance(other, RangeIndex)
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        assert ridx is None
+
+    def test_join_non_int_index(self):
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index([3, 6, 7, 8, 10], dtype=object)
+
+        outer = index.join(other, how="outer")
+        outer2 = other.join(index, how="outer")
+        expected = Index([0, 2, 3, 4, 6, 7, 8, 10, 12, 14, 16, 18])
+        tm.assert_index_equal(outer, outer2)
+        tm.assert_index_equal(outer, expected)
+
+        inner = index.join(other, how="inner")
+        inner2 = other.join(index, how="inner")
+        expected = Index([6, 8, 10])
+        tm.assert_index_equal(inner, inner2)
+        tm.assert_index_equal(inner, expected)
+
+        left = index.join(other, how="left")
+        tm.assert_index_equal(left, index.astype(object))
+
+        left2 = other.join(index, how="left")
+        tm.assert_index_equal(left2, other)
+
+        right = index.join(other, how="right")
+        tm.assert_index_equal(right, other)
+
+        right2 = other.join(index, how="right")
+        tm.assert_index_equal(right2, index.astype(object))
+
+    def test_join_non_unique(self):
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index([4, 4, 3, 3])
+
+        res, lidx, ridx = index.join(other, return_indexers=True)
+
+        eres = Index([0, 2, 4, 4, 6, 8, 10, 12, 14, 16, 18])
+        elidx = np.array([0, 1, 2, 2, 3, 4, 5, 6, 7, 8, 9], dtype=np.intp)
+        eridx = np.array([-1, -1, 0, 1, -1, -1, -1, -1, -1, -1, -1], dtype=np.intp)
+
+        tm.assert_index_equal(res, eres)
+        tm.assert_numpy_array_equal(lidx, elidx)
+        tm.assert_numpy_array_equal(ridx, eridx)
+
+    def test_join_self(self, join_type):
+        index = RangeIndex(start=0, stop=20, step=2)
+        joined = index.join(index, how=join_type)
+        assert index is joined
+
+
+@pytest.mark.parametrize(
+    "left, right, expected, expected_lidx, expected_ridx, how",
+    [
+        [RangeIndex(2), RangeIndex(3), RangeIndex(2), None, [0, 1], "left"],
+        [RangeIndex(2), RangeIndex(2), RangeIndex(2), None, None, "left"],
+        [RangeIndex(2), RangeIndex(20, 22), RangeIndex(2), None, [-1, -1], "left"],
+        [RangeIndex(2), RangeIndex(3), RangeIndex(3), [0, 1, -1], None, "right"],
+        [RangeIndex(2), RangeIndex(2), RangeIndex(2), None, None, "right"],
+        [
+            RangeIndex(2),
+            RangeIndex(20, 22),
+            RangeIndex(20, 22),
+            [-1, -1],
+            None,
+            "right",
+        ],
+        [RangeIndex(2), RangeIndex(3), RangeIndex(2), [0, 1], [0, 1], "inner"],
+        [RangeIndex(2), RangeIndex(2), RangeIndex(2), None, None, "inner"],
+        [RangeIndex(2), RangeIndex(1, 3), RangeIndex(1, 2), [1], [0], "inner"],
+        [RangeIndex(2), RangeIndex(3), RangeIndex(3), [0, 1, -1], [0, 1, 2], "outer"],
+        [RangeIndex(2), RangeIndex(2), RangeIndex(2), None, None, "outer"],
+        [
+            RangeIndex(2),
+            RangeIndex(2, 4),
+            RangeIndex(4),
+            [0, 1, -1, -1],
+            [-1, -1, 0, 1],
+            "outer",
+        ],
+        [RangeIndex(2), RangeIndex(0), RangeIndex(2), None, [-1, -1], "left"],
+        [RangeIndex(2), RangeIndex(0), RangeIndex(0), [], None, "right"],
+        [RangeIndex(2), RangeIndex(0), RangeIndex(0), [], None, "inner"],
+        [RangeIndex(2), RangeIndex(0), RangeIndex(2), None, [-1, -1], "outer"],
+    ],
+)
+@pytest.mark.parametrize(
+    "right_type", [RangeIndex, lambda x: Index(list(x), dtype=x.dtype)]
+)
+def test_join_preserves_rangeindex(
+    left, right, expected, expected_lidx, expected_ridx, how, right_type
+):
+    result, lidx, ridx = left.join(right_type(right), how=how, return_indexers=True)
+    tm.assert_index_equal(result, expected, exact=True)
+
+    if expected_lidx is None:
+        assert lidx is expected_lidx
+    else:
+        exp_lidx = np.array(expected_lidx, dtype=np.intp)
+        tm.assert_numpy_array_equal(lidx, exp_lidx)
+
+    if expected_ridx is None:
+        assert ridx is expected_ridx
+    else:
+        exp_ridx = np.array(expected_ridx, dtype=np.intp)
+        tm.assert_numpy_array_equal(ridx, exp_ridx)
diff --git a/pandas/pandas/tests/indexes/ranges/test_range.py b/pandas/pandas/tests/indexes/ranges/test_range.py
new file mode 100644
index 0000000000000..1f9df30d60c11
--- /dev/null
+++ b/pandas/pandas/tests/indexes/ranges/test_range.py
@@ -0,0 +1,909 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import ensure_platform_int
+
+import pandas as pd
+from pandas import (
+    Index,
+    RangeIndex,
+)
+import pandas._testing as tm
+from pandas.core.indexes.range import min_fitting_element
+
+
+class TestRangeIndex:
+    @pytest.fixture
+    def simple_index(self):
+        return RangeIndex(start=0, stop=20, step=2)
+
+    def test_constructor_unwraps_index(self):
+        result = RangeIndex(1, 3)
+        expected = np.array([1, 2], dtype=np.int64)
+        tm.assert_numpy_array_equal(result._data, expected)
+
+    def test_can_hold_identifiers(self, simple_index):
+        idx = simple_index
+        key = idx[0]
+        assert idx._can_hold_identifiers_and_holds_name(key) is False
+
+    def test_too_many_names(self, simple_index):
+        index = simple_index
+        with pytest.raises(ValueError, match="^Length"):
+            index.names = ["roger", "harold"]
+
+    @pytest.mark.parametrize(
+        "index, start, stop, step",
+        [
+            (RangeIndex(5), 0, 5, 1),
+            (RangeIndex(0, 5), 0, 5, 1),
+            (RangeIndex(5, step=2), 0, 5, 2),
+            (RangeIndex(1, 5, 2), 1, 5, 2),
+        ],
+    )
+    def test_start_stop_step_attrs(self, index, start, stop, step):
+        # GH 25710
+        assert index.start == start
+        assert index.stop == stop
+        assert index.step == step
+
+    def test_copy(self):
+        i = RangeIndex(5, name="Foo")
+        i_copy = i.copy()
+        assert i_copy is not i
+        assert i_copy.identical(i)
+        assert i_copy._range == range(0, 5, 1)
+        assert i_copy.name == "Foo"
+
+    def test_repr(self):
+        i = RangeIndex(5, name="Foo")
+        result = repr(i)
+        expected = "RangeIndex(start=0, stop=5, step=1, name='Foo')"
+        assert result == expected
+
+        result = eval(result)
+        tm.assert_index_equal(result, i, exact=True)
+
+        i = RangeIndex(5, 0, -1)
+        result = repr(i)
+        expected = "RangeIndex(start=5, stop=0, step=-1)"
+        assert result == expected
+
+        result = eval(result)
+        tm.assert_index_equal(result, i, exact=True)
+
+    def test_insert(self):
+        idx = RangeIndex(5, name="Foo")
+        result = idx[1:4]
+
+        # test 0th element
+        tm.assert_index_equal(idx[0:4], result.insert(0, idx[0]), exact="equiv")
+
+        # GH 18295 (test missing)
+        expected = Index([0, np.nan, 1, 2, 3, 4], dtype=np.float64)
+        for na in [np.nan, None, pd.NA]:
+            result = RangeIndex(5).insert(1, na)
+            tm.assert_index_equal(result, expected)
+
+        result = RangeIndex(5).insert(1, pd.NaT)
+        expected = Index([0, pd.NaT, 1, 2, 3, 4], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_edges_preserves_rangeindex(self):
+        idx = Index(range(4, 9, 2))
+
+        result = idx.insert(0, 2)
+        expected = Index(range(2, 9, 2))
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.insert(3, 10)
+        expected = Index(range(4, 11, 2))
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_insert_middle_preserves_rangeindex(self):
+        # insert in the middle
+        idx = Index(range(0, 3, 2))
+        result = idx.insert(1, 1)
+        expected = Index(range(3))
+        tm.assert_index_equal(result, expected, exact=True)
+
+        idx = idx * 2
+        result = idx.insert(1, 2)
+        expected = expected * 2
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete(self):
+        idx = RangeIndex(5, name="Foo")
+        expected = idx[1:]
+        result = idx.delete(0)
+        tm.assert_index_equal(result, expected, exact=True)
+        assert result.name == expected.name
+
+        expected = idx[:-1]
+        result = idx.delete(-1)
+        tm.assert_index_equal(result, expected, exact=True)
+        assert result.name == expected.name
+
+        msg = "index 5 is out of bounds for axis 0 with size 5"
+        with pytest.raises((IndexError, ValueError), match=msg):
+            # either depending on numpy version
+            result = idx.delete(len(idx))
+
+    def test_delete_preserves_rangeindex(self):
+        idx = Index(range(2), name="foo")
+
+        result = idx.delete([1])
+        expected = Index(range(1), name="foo")
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(1)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete_preserves_rangeindex_middle(self):
+        idx = Index(range(3), name="foo")
+        result = idx.delete(1)
+        expected = idx[::2]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(-2)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete_preserves_rangeindex_list_at_end(self):
+        idx = RangeIndex(0, 6, 1)
+
+        loc = [2, 3, 4, 5]
+        result = idx.delete(loc)
+        expected = idx[:2]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(loc[::-1])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete_preserves_rangeindex_list_middle(self):
+        idx = RangeIndex(0, 6, 1)
+
+        loc = [1, 2, 3, 4]
+        result = idx.delete(loc)
+        expected = RangeIndex(0, 6, 5)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(loc[::-1])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete_all_preserves_rangeindex(self):
+        idx = RangeIndex(0, 6, 1)
+
+        loc = [0, 1, 2, 3, 4, 5]
+        result = idx.delete(loc)
+        expected = idx[:0]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(loc[::-1])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_delete_not_preserving_rangeindex(self):
+        idx = RangeIndex(0, 6, 1)
+
+        loc = [0, 3, 5]
+        result = idx.delete(loc)
+        expected = Index([1, 2, 4])
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.delete(loc[::-1])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_view(self):
+        i = RangeIndex(0, name="Foo")
+        i_view = i.view()
+        assert i_view.name == "Foo"
+
+        i_view = i.view("i8")
+        tm.assert_numpy_array_equal(i.values, i_view)
+
+    def test_dtype(self, simple_index):
+        index = simple_index
+        assert index.dtype == np.int64
+
+    def test_cache(self):
+        # GH 26565, GH26617, GH35432, GH53387
+        # This test checks whether _cache has been set.
+        # Calling RangeIndex._cache["_data"] creates an int64 array of the same length
+        # as the RangeIndex and stores it in _cache.
+        idx = RangeIndex(0, 100, 10)
+
+        assert idx._cache == {}
+
+        repr(idx)
+        assert idx._cache == {}
+
+        str(idx)
+        assert idx._cache == {}
+
+        idx.get_loc(20)
+        assert idx._cache == {}
+
+        90 in idx  # True
+        assert idx._cache == {}
+
+        91 in idx  # False
+        assert idx._cache == {}
+
+        idx.all()
+        assert idx._cache == {}
+
+        idx.any()
+        assert idx._cache == {}
+
+        for _ in idx:
+            pass
+        assert idx._cache == {}
+
+        df = pd.DataFrame({"a": range(10)}, index=idx)
+
+        # df.__repr__ should not populate index cache
+        str(df)
+        assert idx._cache == {}
+
+        df.loc[50]
+        assert idx._cache == {}
+
+        with pytest.raises(KeyError, match="51"):
+            df.loc[51]
+        assert idx._cache == {}
+
+        df.loc[10:50]
+        assert idx._cache == {}
+
+        df.iloc[5:10]
+        assert idx._cache == {}
+
+        # after calling take, _cache may contain other keys, but not "_data"
+        idx.take([3, 0, 1])
+        assert "_data" not in idx._cache
+
+        df.loc[[50]]
+        assert "_data" not in idx._cache
+
+        df.iloc[[5, 6, 7, 8, 9]]
+        assert "_data" not in idx._cache
+
+        # idx._cache should contain a _data entry after call to idx._data
+        idx._data
+        assert isinstance(idx._data, np.ndarray)
+        assert idx._data is idx._data  # check cached value is reused
+        assert "_data" in idx._cache
+        expected = np.arange(0, 100, 10, dtype="int64")
+        tm.assert_numpy_array_equal(idx._cache["_data"], expected)
+
+    def test_is_monotonic(self):
+        index = RangeIndex(0, 20, 2)
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is False
+        assert index._is_strictly_monotonic_increasing is True
+        assert index._is_strictly_monotonic_decreasing is False
+
+        index = RangeIndex(4, 0, -1)
+        assert index.is_monotonic_increasing is False
+        assert index._is_strictly_monotonic_increasing is False
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_decreasing is True
+
+        index = RangeIndex(1, 2)
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_increasing is True
+        assert index._is_strictly_monotonic_decreasing is True
+
+        index = RangeIndex(2, 1)
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_increasing is True
+        assert index._is_strictly_monotonic_decreasing is True
+
+        index = RangeIndex(1, 1)
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_increasing is True
+        assert index.is_monotonic_decreasing is True
+        assert index._is_strictly_monotonic_increasing is True
+        assert index._is_strictly_monotonic_decreasing is True
+
+    @pytest.mark.parametrize(
+        "left,right",
+        [
+            (RangeIndex(0, 9, 2), RangeIndex(0, 10, 2)),
+            (RangeIndex(0), RangeIndex(1, -1, 3)),
+            (RangeIndex(1, 2, 3), RangeIndex(1, 3, 4)),
+            (RangeIndex(0, -9, -2), RangeIndex(0, -10, -2)),
+        ],
+    )
+    def test_equals_range(self, left, right):
+        assert left.equals(right)
+        assert right.equals(left)
+
+    def test_logical_compat(self, simple_index):
+        idx = simple_index
+        assert idx.all() == idx.values.all()
+        assert idx.any() == idx.values.any()
+
+    def test_identical(self, simple_index):
+        index = simple_index
+        i = Index(index.copy())
+        assert i.identical(index)
+
+        # we don't allow object dtype for RangeIndex
+        if isinstance(index, RangeIndex):
+            return
+
+        same_values_different_type = Index(i, dtype=object)
+        assert not i.identical(same_values_different_type)
+
+        i = index.copy(dtype=object)
+        i = i.rename("foo")
+        same_values = Index(i, dtype=object)
+        assert same_values.identical(index.copy(dtype=object))
+
+        assert not i.identical(index)
+        assert Index(same_values, name="foo", dtype=object).identical(i)
+
+        assert not index.copy(dtype=object).identical(index.copy(dtype="int64"))
+
+    def test_nbytes(self):
+        # memory savings vs int index
+        idx = RangeIndex(0, 1000)
+        assert idx.nbytes < Index(idx._values).nbytes / 10
+
+        # constant memory usage
+        i2 = RangeIndex(0, 10)
+        assert idx.nbytes == i2.nbytes
+
+    @pytest.mark.parametrize(
+        "start,stop,step",
+        [
+            # can't
+            ("foo", "bar", "baz"),
+            # shouldn't
+            ("0", "1", "2"),
+        ],
+    )
+    def test_cant_or_shouldnt_cast(self, start, stop, step):
+        msg = f"Wrong type {type(start)} for value {start}"
+        with pytest.raises(TypeError, match=msg):
+            RangeIndex(start, stop, step)
+
+    def test_view_index(self, simple_index):
+        index = simple_index
+        msg = (
+            "Cannot change data-type for array of references.|"
+            "Cannot change data-type for object array.|"
+        )
+        with pytest.raises(TypeError, match=msg):
+            index.view(Index)
+
+    def test_prevent_casting(self, simple_index):
+        index = simple_index
+        result = index.astype("O")
+        assert result.dtype == np.object_
+
+    def test_repr_roundtrip(self, simple_index):
+        index = simple_index
+        tm.assert_index_equal(eval(repr(index)), index)
+
+    def test_slice_keep_name(self):
+        idx = RangeIndex(1, 2, name="asdf")
+        assert idx.name == idx[1:].name
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            RangeIndex(start=0, stop=20, step=2, name="foo"),
+            RangeIndex(start=18, stop=-1, step=-2, name="bar"),
+        ],
+        ids=["index_inc", "index_dec"],
+    )
+    def test_has_duplicates(self, index):
+        assert index.is_unique
+        assert not index.has_duplicates
+
+    def test_extended_gcd(self, simple_index):
+        index = simple_index
+        result = index._extended_gcd(6, 10)
+        assert result[0] == result[1] * 6 + result[2] * 10
+        assert 2 == result[0]
+
+        result = index._extended_gcd(10, 6)
+        assert 2 == result[1] * 10 + result[2] * 6
+        assert 2 == result[0]
+
+    def test_min_fitting_element(self):
+        result = min_fitting_element(0, 2, 1)
+        assert 2 == result
+
+        result = min_fitting_element(1, 1, 1)
+        assert 1 == result
+
+        result = min_fitting_element(18, -2, 1)
+        assert 2 == result
+
+        result = min_fitting_element(5, -1, 1)
+        assert 1 == result
+
+        big_num = 500000000000000000000000
+
+        result = min_fitting_element(5, 1, big_num)
+        assert big_num == result
+
+    def test_slice_specialised(self, simple_index):
+        index = simple_index
+        index.name = "foo"
+
+        # scalar indexing
+        res = index[1]
+        expected = 2
+        assert res == expected
+
+        res = index[-1]
+        expected = 18
+        assert res == expected
+
+        # slicing
+        # slice value completion
+        index_slice = index[:]
+        expected = index
+        tm.assert_index_equal(index_slice, expected)
+
+        # positive slice values
+        index_slice = index[7:10:2]
+        expected = Index([14, 18], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        # negative slice values
+        index_slice = index[-1:-5:-2]
+        expected = Index([18, 14], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        # stop overshoot
+        index_slice = index[2:100:4]
+        expected = Index([4, 12], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        # reverse
+        index_slice = index[::-1]
+        expected = Index(index.values[::-1], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        index_slice = index[-8::-1]
+        expected = Index([4, 2, 0], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        index_slice = index[-40::-1]
+        expected = Index(np.array([], dtype=np.int64), name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        index_slice = index[40::-1]
+        expected = Index(index.values[40::-1], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+        index_slice = index[10::-1]
+        expected = Index(index.values[::-1], name="foo")
+        tm.assert_index_equal(index_slice, expected, exact="equiv")
+
+    @pytest.mark.parametrize("step", set(range(-5, 6)) - {0})
+    def test_len_specialised(self, step):
+        # make sure that our len is the same as np.arange calc
+        start, stop = (0, 5) if step > 0 else (5, 0)
+
+        arr = np.arange(start, stop, step)
+        index = RangeIndex(start, stop, step)
+        assert len(index) == len(arr)
+
+        index = RangeIndex(stop, start, step)
+        assert len(index) == 0
+
+    @pytest.mark.parametrize(
+        "indices, expected",
+        [
+            ([RangeIndex(1, 12, 5)], RangeIndex(1, 12, 5)),
+            ([RangeIndex(0, 6, 4)], RangeIndex(0, 6, 4)),
+            ([RangeIndex(1, 3), RangeIndex(3, 7)], RangeIndex(1, 7)),
+            ([RangeIndex(1, 5, 2), RangeIndex(5, 6)], RangeIndex(1, 6, 2)),
+            ([RangeIndex(1, 3, 2), RangeIndex(4, 7, 3)], RangeIndex(1, 7, 3)),
+            ([RangeIndex(-4, 3, 2), RangeIndex(4, 7, 2)], RangeIndex(-4, 7, 2)),
+            ([RangeIndex(-4, -8), RangeIndex(-8, -12)], RangeIndex(0, 0)),
+            ([RangeIndex(-4, -8), RangeIndex(3, -4)], RangeIndex(0, 0)),
+            ([RangeIndex(-4, -8), RangeIndex(3, 5)], RangeIndex(3, 5)),
+            ([RangeIndex(-4, -2), RangeIndex(3, 5)], Index([-4, -3, 3, 4])),
+            ([RangeIndex(-2), RangeIndex(3, 5)], RangeIndex(3, 5)),
+            ([RangeIndex(2), RangeIndex(2)], Index([0, 1, 0, 1])),
+            ([RangeIndex(2), RangeIndex(2, 5), RangeIndex(5, 8, 4)], RangeIndex(0, 6)),
+            (
+                [RangeIndex(2), RangeIndex(3, 5), RangeIndex(5, 8, 4)],
+                Index([0, 1, 3, 4, 5]),
+            ),
+            (
+                [RangeIndex(-2, 2), RangeIndex(2, 5), RangeIndex(5, 8, 4)],
+                RangeIndex(-2, 6),
+            ),
+            ([RangeIndex(3), Index([-1, 3, 15])], Index([0, 1, 2, -1, 3, 15])),
+            ([RangeIndex(3), Index([-1, 3.1, 15.0])], Index([0, 1, 2, -1, 3.1, 15.0])),
+            ([RangeIndex(3), Index(["a", None, 14])], Index([0, 1, 2, "a", None, 14])),
+            ([RangeIndex(3, 1), Index(["a", None, 14])], Index(["a", None, 14])),
+        ],
+    )
+    def test_append(self, indices, expected):
+        # GH16212
+        result = indices[0].append(indices[1:])
+        tm.assert_index_equal(result, expected, exact=True)
+
+        if len(indices) == 2:
+            # Append single item rather than list
+            result2 = indices[0].append(indices[1])
+            tm.assert_index_equal(result2, expected, exact=True)
+
+    def test_engineless_lookup(self):
+        # GH 16685
+        # Standard lookup on RangeIndex should not require the engine to be
+        # created
+        idx = RangeIndex(2, 10, 3)
+
+        assert idx.get_loc(5) == 1
+        tm.assert_numpy_array_equal(
+            idx.get_indexer([2, 8]), ensure_platform_int(np.array([0, 2]))
+        )
+        with pytest.raises(KeyError, match="3"):
+            idx.get_loc(3)
+
+        assert "_engine" not in idx._cache
+
+        # Different types of scalars can be excluded immediately, no need to
+        #  use the _engine
+        with pytest.raises(KeyError, match="'a'"):
+            idx.get_loc("a")
+
+        assert "_engine" not in idx._cache
+
+    @pytest.mark.parametrize(
+        "ri",
+        [
+            RangeIndex(0, -1, -1),
+            RangeIndex(0, 1, 1),
+            RangeIndex(1, 3, 2),
+            RangeIndex(0, -1, -2),
+            RangeIndex(-3, -5, -2),
+        ],
+    )
+    def test_append_len_one(self, ri):
+        # GH39401
+        result = ri.append([])
+        tm.assert_index_equal(result, ri, exact=True)
+
+    @pytest.mark.parametrize("base", [RangeIndex(0, 2), Index([0, 1])])
+    def test_isin_range(self, base):
+        # GH#41151
+        values = RangeIndex(0, 1)
+        result = base.isin(values)
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_sort_values_key(self):
+        # GH#43666, GH#52764
+        sort_order = {8: 2, 6: 0, 4: 8, 2: 10, 0: 12}
+        values = RangeIndex(0, 10, 2)
+        result = values.sort_values(key=lambda x: x.map(sort_order))
+        expected = Index([6, 8, 4, 2, 0], dtype="int64")
+        tm.assert_index_equal(result, expected, check_exact=True)
+
+        # check this matches the Series.sort_values behavior
+        ser = values.to_series()
+        result2 = ser.sort_values(key=lambda x: x.map(sort_order))
+        tm.assert_series_equal(result2, expected.to_series(), check_exact=True)
+
+    def test_range_index_rsub_by_const(self):
+        # GH#53255
+        result = 3 - RangeIndex(0, 4, 1)
+        expected = RangeIndex(3, -1, -1)
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "rng, decimals",
+    [
+        [range(5), 0],
+        [range(5), 2],
+        [range(10, 30, 10), -1],
+        [range(30, 10, -10), -1],
+    ],
+)
+def test_range_round_returns_rangeindex(rng, decimals):
+    ri = RangeIndex(rng)
+    expected = ri.copy()
+    result = ri.round(decimals=decimals)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+@pytest.mark.parametrize(
+    "rng, decimals",
+    [
+        [range(10, 30, 1), -1],
+        [range(30, 10, -1), -1],
+        [range(11, 14), -10],
+    ],
+)
+def test_range_round_returns_index(rng, decimals):
+    ri = RangeIndex(rng)
+    expected = Index(list(rng)).round(decimals=decimals)
+    result = ri.round(decimals=decimals)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_reindex_1_value_returns_rangeindex():
+    ri = RangeIndex(0, 10, 2, name="foo")
+    result, result_indexer = ri.reindex([2])
+    expected = RangeIndex(2, 4, 2, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    expected_indexer = np.array([1], dtype=np.intp)
+    tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+
+def test_reindex_empty_returns_rangeindex():
+    ri = RangeIndex(0, 10, 2, name="foo")
+    result, result_indexer = ri.reindex([])
+    expected = RangeIndex(0, 0, 2, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    expected_indexer = np.array([], dtype=np.intp)
+    tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+
+def test_insert_empty_0_loc():
+    ri = RangeIndex(0, step=10, name="foo")
+    result = ri.insert(0, 5)
+    expected = RangeIndex(5, 15, 10, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_append_non_rangeindex_return_rangeindex():
+    ri = RangeIndex(1)
+    result = ri.append(Index([1]))
+    expected = RangeIndex(2)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_append_non_rangeindex_return_index():
+    ri = RangeIndex(1)
+    result = ri.append(Index([1, 3, 4]))
+    expected = Index([0, 1, 3, 4])
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_reindex_returns_rangeindex():
+    ri = RangeIndex(2, name="foo")
+    result, result_indexer = ri.reindex([1, 2, 3])
+    expected = RangeIndex(1, 4, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    expected_indexer = np.array([1, -1, -1], dtype=np.intp)
+    tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+
+def test_reindex_returns_index():
+    ri = RangeIndex(4, name="foo")
+    result, result_indexer = ri.reindex([0, 1, 3])
+    expected = Index([0, 1, 3], name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    expected_indexer = np.array([0, 1, 3], dtype=np.intp)
+    tm.assert_numpy_array_equal(result_indexer, expected_indexer)
+
+
+def test_take_return_rangeindex():
+    ri = RangeIndex(5, name="foo")
+    result = ri.take([])
+    expected = RangeIndex(0, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    result = ri.take([3, 4])
+    expected = RangeIndex(3, 5, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+@pytest.mark.parametrize(
+    "rng, exp_rng",
+    [
+        [range(5), range(3, 4)],
+        [range(0, -10, -2), range(-6, -8, -2)],
+        [range(0, 10, 2), range(6, 8, 2)],
+    ],
+)
+def test_take_1_value_returns_rangeindex(rng, exp_rng):
+    ri = RangeIndex(rng, name="foo")
+    result = ri.take([3])
+    expected = RangeIndex(exp_rng, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_append_one_nonempty_preserve_step():
+    expected = RangeIndex(0, -1, -1)
+    result = RangeIndex(0).append([expected])
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_boolmask_all_true():
+    ri = RangeIndex(3, name="foo")
+    expected = ri.copy()
+    result = ri[[True] * 3]
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_boolmask_all_false():
+    ri = RangeIndex(3, name="foo")
+    result = ri[[False] * 3]
+    expected = RangeIndex(0, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_boolmask_returns_rangeindex():
+    ri = RangeIndex(3, name="foo")
+    result = ri[[False, True, True]]
+    expected = RangeIndex(1, 3, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    result = ri[[True, False, True]]
+    expected = RangeIndex(0, 3, 2, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_boolmask_returns_index():
+    ri = RangeIndex(4, name="foo")
+    result = ri[[True, True, False, True]]
+    expected = Index([0, 1, 3], name="foo")
+    tm.assert_index_equal(result, expected)
+
+
+def test_getitem_boolmask_wrong_length():
+    ri = RangeIndex(4, name="foo")
+    with pytest.raises(IndexError, match="Boolean index has wrong length"):
+        ri[[True]]
+
+
+def test_pos_returns_rangeindex():
+    ri = RangeIndex(2, name="foo")
+    expected = ri.copy()
+    result = +ri
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_neg_returns_rangeindex():
+    ri = RangeIndex(2, name="foo")
+    result = -ri
+    expected = RangeIndex(0, -2, -1, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    ri = RangeIndex(-2, 2, name="foo")
+    result = -ri
+    expected = RangeIndex(2, -2, -1, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+@pytest.mark.parametrize(
+    "rng, exp_rng",
+    [
+        [range(0), range(0)],
+        [range(10), range(10)],
+        [range(-2, 1, 1), range(2, -1, -1)],
+        [range(0, -10, -1), range(0, 10, 1)],
+    ],
+)
+def test_abs_returns_rangeindex(rng, exp_rng):
+    ri = RangeIndex(rng, name="foo")
+    expected = RangeIndex(exp_rng, name="foo")
+    result = abs(ri)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_abs_returns_index():
+    ri = RangeIndex(-2, 2, name="foo")
+    result = abs(ri)
+    expected = Index([2, 1, 0, 1], name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+@pytest.mark.parametrize(
+    "rng",
+    [
+        range(0),
+        range(5),
+        range(0, -5, -1),
+        range(-2, 2, 1),
+        range(2, -2, -2),
+        range(0, 5, 2),
+    ],
+)
+def test_invert_returns_rangeindex(rng):
+    ri = RangeIndex(rng, name="foo")
+    result = ~ri
+    assert isinstance(result, RangeIndex)
+    expected = ~Index(list(rng), name="foo")
+    tm.assert_index_equal(result, expected, exact=False)
+
+
+@pytest.mark.parametrize(
+    "rng",
+    [
+        range(0, 5, 1),
+        range(0, 5, 2),
+        range(10, 15, 1),
+        range(10, 5, -1),
+        range(10, 5, -2),
+        range(5, 0, -1),
+    ],
+)
+@pytest.mark.parametrize("meth", ["argmax", "argmin"])
+def test_arg_min_max(rng, meth):
+    ri = RangeIndex(rng)
+    idx = Index(list(rng))
+    assert getattr(ri, meth)() == getattr(idx, meth)()
+
+
+@pytest.mark.parametrize("meth", ["argmin", "argmax"])
+def test_empty_argmin_argmax_raises(meth):
+    with pytest.raises(ValueError, match=f"attempt to get {meth} of an empty sequence"):
+        getattr(RangeIndex(0), meth)()
+
+
+def test_getitem_integers_return_rangeindex():
+    result = RangeIndex(0, 10, 2, name="foo")[[0, -1]]
+    expected = RangeIndex(start=0, stop=16, step=8, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+    result = RangeIndex(0, 10, 2, name="foo")[[3]]
+    expected = RangeIndex(start=6, stop=8, step=2, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_empty_return_rangeindex():
+    result = RangeIndex(0, 10, 2, name="foo")[[]]
+    expected = RangeIndex(start=0, stop=0, step=1, name="foo")
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_getitem_integers_return_index():
+    result = RangeIndex(0, 10, 2, name="foo")[[0, 1, -1]]
+    expected = Index([0, 2, 8], dtype="int64", name="foo")
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("normalize", [True, False])
+@pytest.mark.parametrize(
+    "rng",
+    [
+        range(3),
+        range(0),
+        range(0, 3, 2),
+        range(3, -3, -2),
+    ],
+)
+def test_value_counts(sort, dropna, ascending, normalize, rng):
+    ri = RangeIndex(rng, name="A")
+    result = ri.value_counts(
+        normalize=normalize, sort=sort, ascending=ascending, dropna=dropna
+    )
+    expected = Index(list(rng), name="A").value_counts(
+        normalize=normalize, sort=sort, ascending=ascending, dropna=dropna
+    )
+    tm.assert_series_equal(result, expected, check_index_type=False)
+
+
+@pytest.mark.parametrize("side", ["left", "right"])
+@pytest.mark.parametrize("value", [0, -5, 5, -3, np.array([-5, -3, 0, 5])])
+def test_searchsorted(side, value):
+    ri = RangeIndex(-3, 3, 2)
+    result = ri.searchsorted(value=value, side=side)
+    expected = Index(list(ri)).searchsorted(value=value, side=side)
+    if isinstance(value, int):
+        assert result == expected
+    else:
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/ranges/test_setops.py b/pandas/pandas/tests/indexes/ranges/test_setops.py
new file mode 100644
index 0000000000000..ac24ff828cb8f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/ranges/test_setops.py
@@ -0,0 +1,493 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+from hypothesis import (
+    assume,
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+class TestRangeIndexSetOps:
+    @pytest.mark.parametrize("dtype", [None, "int64", "uint64"])
+    def test_intersection_mismatched_dtype(self, dtype):
+        # check that we cast to float, not object
+        index = RangeIndex(start=0, stop=20, step=2, name="foo")
+        index = Index(index, dtype=dtype)
+
+        flt = index.astype(np.float64)
+
+        # bc index.equals(flt), we go through fastpath and get RangeIndex back
+        result = index.intersection(flt)
+        tm.assert_index_equal(result, index, exact=True)
+
+        result = flt.intersection(index)
+        tm.assert_index_equal(result, flt, exact=True)
+
+        # neither empty, not-equals
+        result = index.intersection(flt[1:])
+        tm.assert_index_equal(result, flt[1:], exact=True)
+
+        result = flt[1:].intersection(index)
+        tm.assert_index_equal(result, flt[1:], exact=True)
+
+        # empty other
+        result = index.intersection(flt[:0])
+        tm.assert_index_equal(result, flt[:0], exact=True)
+
+        result = flt[:0].intersection(index)
+        tm.assert_index_equal(result, flt[:0], exact=True)
+
+    def test_intersection_empty(self, sort, names):
+        # name retention on empty intersections
+        index = RangeIndex(start=0, stop=20, step=2, name=names[0])
+
+        # empty other
+        result = index.intersection(index[:0].rename(names[1]), sort=sort)
+        tm.assert_index_equal(result, index[:0].rename(names[2]), exact=True)
+
+        # empty self
+        result = index[:0].intersection(index.rename(names[1]), sort=sort)
+        tm.assert_index_equal(result, index[:0].rename(names[2]), exact=True)
+
+    def test_intersection(self, sort):
+        # intersect with Index with dtype int64
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index(np.arange(1, 6))
+        result = index.intersection(other, sort=sort)
+        expected = Index(np.sort(np.intersect1d(index.values, other.values)))
+        tm.assert_index_equal(result, expected)
+
+        result = other.intersection(index, sort=sort)
+        expected = Index(
+            np.sort(np.asarray(np.intersect1d(index.values, other.values)))
+        )
+        tm.assert_index_equal(result, expected)
+
+        # intersect with increasing RangeIndex
+        other = RangeIndex(1, 6)
+        result = index.intersection(other, sort=sort)
+        expected = Index(np.sort(np.intersect1d(index.values, other.values)))
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+        # intersect with decreasing RangeIndex
+        other = RangeIndex(5, 0, -1)
+        result = index.intersection(other, sort=sort)
+        expected = Index(np.sort(np.intersect1d(index.values, other.values)))
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+        # reversed (GH 17296)
+        result = other.intersection(index, sort=sort)
+        tm.assert_index_equal(result, expected, exact="equiv")
+
+        # GH 17296: intersect two decreasing RangeIndexes
+        first = RangeIndex(10, -2, -2)
+        other = RangeIndex(5, -4, -1)
+        expected = RangeIndex(start=4, stop=-2, step=-2)
+        result = first.intersection(other, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        # reversed
+        result = other.intersection(first, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+        index = RangeIndex(5, name="foo")
+
+        # intersect of non-overlapping indices
+        other = RangeIndex(5, 10, 1, name="foo")
+        result = index.intersection(other, sort=sort)
+        expected = RangeIndex(0, 0, 1, name="foo")
+        tm.assert_index_equal(result, expected)
+
+        other = RangeIndex(-1, -5, -1)
+        result = index.intersection(other, sort=sort)
+        expected = RangeIndex(0, 0, 1)
+        tm.assert_index_equal(result, expected)
+
+        # intersection of empty indices
+        other = RangeIndex(0, 0, 1)
+        result = index.intersection(other, sort=sort)
+        expected = RangeIndex(0, 0, 1)
+        tm.assert_index_equal(result, expected)
+
+        result = other.intersection(index, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_non_overlapping_gcd(self, sort, names):
+        # intersection of non-overlapping values based on start value and gcd
+        index = RangeIndex(1, 10, 2, name=names[0])
+        other = RangeIndex(0, 10, 4, name=names[1])
+        result = index.intersection(other, sort=sort)
+        expected = RangeIndex(0, 0, 1, name=names[2])
+        tm.assert_index_equal(result, expected)
+
+    def test_union_noncomparable(self, sort):
+        # corner case, Index with non-int64 dtype
+        index = RangeIndex(start=0, stop=20, step=2)
+        other = Index([datetime.now() + timedelta(i) for i in range(4)], dtype=object)
+        result = index.union(other, sort=sort)
+        expected = Index(np.concatenate((index, other)))
+        tm.assert_index_equal(result, expected)
+
+        result = other.union(index, sort=sort)
+        expected = Index(np.concatenate((other, index)))
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "idx1, idx2, expected_sorted, expected_notsorted",
+        [
+            (
+                RangeIndex(0, 10, 1),
+                RangeIndex(0, 10, 1),
+                RangeIndex(0, 10, 1),
+                RangeIndex(0, 10, 1),
+            ),
+            (
+                RangeIndex(0, 10, 1),
+                RangeIndex(5, 20, 1),
+                RangeIndex(0, 20, 1),
+                RangeIndex(0, 20, 1),
+            ),
+            (
+                RangeIndex(0, 10, 1),
+                RangeIndex(10, 20, 1),
+                RangeIndex(0, 20, 1),
+                RangeIndex(0, 20, 1),
+            ),
+            (
+                RangeIndex(0, -10, -1),
+                RangeIndex(0, -10, -1),
+                RangeIndex(0, -10, -1),
+                RangeIndex(0, -10, -1),
+            ),
+            (
+                RangeIndex(0, -10, -1),
+                RangeIndex(-10, -20, -1),
+                RangeIndex(-19, 1, 1),
+                RangeIndex(0, -20, -1),
+            ),
+            (
+                RangeIndex(0, 10, 2),
+                RangeIndex(1, 10, 2),
+                RangeIndex(0, 10, 1),
+                Index(list(range(0, 10, 2)) + list(range(1, 10, 2))),
+            ),
+            (
+                RangeIndex(0, 11, 2),
+                RangeIndex(1, 12, 2),
+                RangeIndex(0, 12, 1),
+                Index(list(range(0, 11, 2)) + list(range(1, 12, 2))),
+            ),
+            (
+                RangeIndex(0, 21, 4),
+                RangeIndex(-2, 24, 4),
+                RangeIndex(-2, 24, 2),
+                Index(list(range(0, 21, 4)) + list(range(-2, 24, 4))),
+            ),
+            (
+                RangeIndex(0, -20, -2),
+                RangeIndex(-1, -21, -2),
+                RangeIndex(-19, 1, 1),
+                Index(list(range(0, -20, -2)) + list(range(-1, -21, -2))),
+            ),
+            (
+                RangeIndex(0, 100, 5),
+                RangeIndex(0, 100, 20),
+                RangeIndex(0, 100, 5),
+                RangeIndex(0, 100, 5),
+            ),
+            (
+                RangeIndex(0, -100, -5),
+                RangeIndex(5, -100, -20),
+                RangeIndex(-95, 10, 5),
+                Index(list(range(0, -100, -5)) + [5]),
+            ),
+            (
+                RangeIndex(0, -11, -1),
+                RangeIndex(1, -12, -4),
+                RangeIndex(-11, 2, 1),
+                Index(list(range(0, -11, -1)) + [1, -11]),
+            ),
+            (RangeIndex(0), RangeIndex(0), RangeIndex(0), RangeIndex(0)),
+            (
+                RangeIndex(0, -10, -2),
+                RangeIndex(0),
+                RangeIndex(0, -10, -2),
+                RangeIndex(0, -10, -2),
+            ),
+            (
+                RangeIndex(0, 100, 2),
+                RangeIndex(100, 150, 200),
+                RangeIndex(0, 102, 2),
+                RangeIndex(0, 102, 2),
+            ),
+            (
+                RangeIndex(0, -100, -2),
+                RangeIndex(-100, 50, 102),
+                RangeIndex(-100, 4, 2),
+                Index(list(range(0, -100, -2)) + [-100, 2]),
+            ),
+            (
+                RangeIndex(0, -100, -1),
+                RangeIndex(0, -50, -3),
+                RangeIndex(-99, 1, 1),
+                RangeIndex(0, -100, -1),
+            ),
+            (
+                RangeIndex(0, 1, 1),
+                RangeIndex(5, 6, 10),
+                RangeIndex(0, 6, 5),
+                RangeIndex(0, 10, 5),
+            ),
+            (
+                RangeIndex(0, 10, 5),
+                RangeIndex(-5, -6, -20),
+                RangeIndex(-5, 10, 5),
+                Index([0, 5, -5]),
+            ),
+            (
+                RangeIndex(0, 3, 1),
+                RangeIndex(4, 5, 1),
+                Index([0, 1, 2, 4]),
+                Index([0, 1, 2, 4]),
+            ),
+            (
+                RangeIndex(0, 10, 1),
+                Index([], dtype=np.int64),
+                RangeIndex(0, 10, 1),
+                RangeIndex(0, 10, 1),
+            ),
+            (
+                RangeIndex(0),
+                Index([1, 5, 6]),
+                Index([1, 5, 6]),
+                Index([1, 5, 6]),
+            ),
+            # GH 43885
+            (
+                RangeIndex(0, 10),
+                RangeIndex(0, 5),
+                RangeIndex(0, 10),
+                RangeIndex(0, 10),
+            ),
+        ],
+        ids=lambda x: repr(x) if isinstance(x, RangeIndex) else x,
+    )
+    def test_union_sorted(self, idx1, idx2, expected_sorted, expected_notsorted):
+        res1 = idx1.union(idx2, sort=None)
+        tm.assert_index_equal(res1, expected_sorted, exact=True)
+
+        res1 = idx1.union(idx2, sort=False)
+        tm.assert_index_equal(res1, expected_notsorted, exact=True)
+
+        res2 = idx2.union(idx1, sort=None)
+        res3 = Index(idx1._values, name=idx1.name).union(idx2, sort=None)
+        tm.assert_index_equal(res2, expected_sorted, exact=True)
+        tm.assert_index_equal(res3, expected_sorted, exact="equiv")
+
+    def test_union_same_step_misaligned(self):
+        # GH#44019
+        left = RangeIndex(range(0, 20, 4))
+        right = RangeIndex(range(1, 21, 4))
+
+        result = left.union(right)
+        expected = Index([0, 1, 4, 5, 8, 9, 12, 13, 16, 17])
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference(self):
+        # GH#12034 Cases where we operate against another RangeIndex and may
+        #  get back another RangeIndex
+        obj = RangeIndex.from_range(range(1, 10), name="foo")
+
+        result = obj.difference(obj)
+        expected = RangeIndex.from_range(range(0), name="foo")
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = obj.difference(expected.rename("bar"))
+        tm.assert_index_equal(result, obj.rename(None), exact=True)
+
+        result = obj.difference(obj[:3])
+        tm.assert_index_equal(result, obj[3:], exact=True)
+
+        result = obj.difference(obj[-3:])
+        tm.assert_index_equal(result, obj[:-3], exact=True)
+
+        # Flipping the step of 'other' doesn't affect the result, but
+        #  flipping the stepof 'self' does when sort=None
+        result = obj[::-1].difference(obj[-3:])
+        tm.assert_index_equal(result, obj[:-3], exact=True)
+
+        result = obj[::-1].difference(obj[-3:], sort=False)
+        tm.assert_index_equal(result, obj[:-3][::-1], exact=True)
+
+        result = obj[::-1].difference(obj[-3:][::-1])
+        tm.assert_index_equal(result, obj[:-3], exact=True)
+
+        result = obj[::-1].difference(obj[-3:][::-1], sort=False)
+        tm.assert_index_equal(result, obj[:-3][::-1], exact=True)
+
+        result = obj.difference(obj[2:6])
+        expected = Index([1, 2, 7, 8, 9], name="foo")
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference_sort(self):
+        # GH#44085 ensure we respect the sort keyword
+
+        idx = Index(range(4))[::-1]
+        other = Index(range(3, 4))
+
+        result = idx.difference(other)
+        expected = Index(range(3))
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = idx.difference(other, sort=False)
+        expected = expected[::-1]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # case where the intersection is empty
+        other = range(10, 12)
+        result = idx.difference(other, sort=None)
+        expected = idx[::-1]
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference_mismatched_step(self):
+        obj = RangeIndex.from_range(range(1, 10), name="foo")
+
+        result = obj.difference(obj[::2])
+        expected = obj[1::2]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = obj[::-1].difference(obj[::2], sort=False)
+        tm.assert_index_equal(result, expected[::-1], exact=True)
+
+        result = obj.difference(obj[1::2])
+        expected = obj[::2]
+        tm.assert_index_equal(result, expected, exact=True)
+
+        result = obj[::-1].difference(obj[1::2], sort=False)
+        tm.assert_index_equal(result, expected[::-1], exact=True)
+
+    def test_difference_interior_overlap_endpoints_preserved(self):
+        left = RangeIndex(range(4))
+        right = RangeIndex(range(1, 3))
+
+        result = left.difference(right)
+        expected = RangeIndex(0, 4, 3)
+        assert expected.tolist() == [0, 3]
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference_endpoints_overlap_interior_preserved(self):
+        left = RangeIndex(-8, 20, 7)
+        right = RangeIndex(13, -9, -3)
+
+        result = left.difference(right)
+        expected = RangeIndex(-1, 13, 7)
+        assert expected.tolist() == [-1, 6]
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference_interior_non_preserving(self):
+        # case with intersection of length 1 but RangeIndex is not preserved
+        idx = Index(range(10))
+
+        other = idx[3:4]
+        result = idx.difference(other)
+        expected = Index([0, 1, 2, 4, 5, 6, 7, 8, 9])
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # case with other.step / self.step > 2
+        other = idx[::3]
+        result = idx.difference(other)
+        expected = Index([1, 2, 4, 5, 7, 8])
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # cases with only reaching one end of left
+        obj = Index(range(20))
+        other = obj[:10:2]
+        result = obj.difference(other)
+        expected = Index([1, 3, 5, 7, 9] + list(range(10, 20)))
+        tm.assert_index_equal(result, expected, exact=True)
+
+        other = obj[1:11:2]
+        result = obj.difference(other)
+        expected = Index([0, 2, 4, 6, 8, 10] + list(range(11, 20)))
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_symmetric_difference(self):
+        # GH#12034 Cases where we operate against another RangeIndex and may
+        #  get back another RangeIndex
+        left = RangeIndex.from_range(range(1, 10), name="foo")
+
+        result = left.symmetric_difference(left)
+        expected = RangeIndex.from_range(range(0), name="foo")
+        tm.assert_index_equal(result, expected)
+
+        result = left.symmetric_difference(expected.rename("bar"))
+        tm.assert_index_equal(result, left.rename(None))
+
+        result = left[:-2].symmetric_difference(left[2:])
+        expected = Index([1, 2, 8, 9], name="foo")
+        tm.assert_index_equal(result, expected, exact=True)
+
+        right = RangeIndex.from_range(range(10, 15))
+
+        result = left.symmetric_difference(right)
+        expected = RangeIndex.from_range(range(1, 15))
+        tm.assert_index_equal(result, expected)
+
+        result = left.symmetric_difference(right[1:])
+        expected = Index([1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14])
+        tm.assert_index_equal(result, expected, exact=True)
+
+
+def assert_range_or_not_is_rangelike(index):
+    """
+    Check that we either have a RangeIndex or that this index *cannot*
+    be represented as a RangeIndex.
+    """
+    if not isinstance(index, RangeIndex) and len(index) > 0:
+        diff = index[:-1] - index[1:]
+        assert not (diff == diff[0]).all()
+
+
+@given(
+    st.integers(-20, 20),
+    st.integers(-20, 20),
+    st.integers(-20, 20),
+    st.integers(-20, 20),
+    st.integers(-20, 20),
+    st.integers(-20, 20),
+)
+def test_range_difference(start1, stop1, step1, start2, stop2, step2):
+    # test that
+    #  a) we match Index[int64].difference and
+    #  b) we return RangeIndex whenever it is possible to do so.
+    assume(step1 != 0)
+    assume(step2 != 0)
+
+    left = RangeIndex(start1, stop1, step1)
+    right = RangeIndex(start2, stop2, step2)
+
+    result = left.difference(right, sort=None)
+    assert_range_or_not_is_rangelike(result)
+
+    left_int64 = Index(left.to_numpy())
+    right_int64 = Index(right.to_numpy())
+
+    alt = left_int64.difference(right_int64, sort=None)
+    tm.assert_index_equal(result, alt, exact="equiv")
+
+    result = left.difference(right, sort=False)
+    assert_range_or_not_is_rangelike(result)
+
+    alt = left_int64.difference(right_int64, sort=False)
+    tm.assert_index_equal(result, alt, exact="equiv")
diff --git a/pandas/pandas/tests/indexes/string/__init__.py b/pandas/pandas/tests/indexes/string/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/string/test_astype.py b/pandas/pandas/tests/indexes/string/test_astype.py
new file mode 100644
index 0000000000000..0349d85f23167
--- /dev/null
+++ b/pandas/pandas/tests/indexes/string/test_astype.py
@@ -0,0 +1,21 @@
+from pandas import (
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_astype_str_from_bytes():
+    # https://github.com/pandas-dev/pandas/issues/38607
+    # GH#49658 pre-2.0 Index called .values.astype(str) here, which effectively
+    #  did a .decode() on the bytes object.  In 2.0 we go through
+    #  ensure_string_array which does f"{val}"
+    idx = Index(["あ", b"a"], dtype="object")
+    result = idx.astype(str)
+    expected = Index(["あ", "a"], dtype="str")
+    tm.assert_index_equal(result, expected)
+
+    # while we're here, check that Series.astype behaves the same
+    result = Series(idx).astype(str)
+    expected = Series(expected, dtype="str")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/string/test_indexing.py b/pandas/pandas/tests/indexes/string/test_indexing.py
new file mode 100644
index 0000000000000..648ee47ddc34c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/string/test_indexing.py
@@ -0,0 +1,199 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Index
+import pandas._testing as tm
+
+
+def _isnan(val):
+    try:
+        return val is not pd.NA and np.isnan(val)
+    except TypeError:
+        return False
+
+
+def _equivalent_na(dtype, null):
+    if dtype.na_value is pd.NA and null is pd.NA:
+        return True
+    elif _isnan(dtype.na_value) and _isnan(null):
+        return True
+    else:
+        return False
+
+
+class TestGetLoc:
+    def test_get_loc(self, any_string_dtype):
+        index = Index(["a", "b", "c"], dtype=any_string_dtype)
+        assert index.get_loc("b") == 1
+
+    def test_get_loc_raises(self, any_string_dtype):
+        index = Index(["a", "b", "c"], dtype=any_string_dtype)
+        with pytest.raises(KeyError, match="d"):
+            index.get_loc("d")
+
+    def test_get_loc_invalid_value(self, any_string_dtype):
+        index = Index(["a", "b", "c"], dtype=any_string_dtype)
+        with pytest.raises(KeyError, match="1"):
+            index.get_loc(1)
+
+    def test_get_loc_non_unique(self, any_string_dtype):
+        index = Index(["a", "b", "a"], dtype=any_string_dtype)
+        result = index.get_loc("a")
+        expected = np.array([True, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_get_loc_non_missing(self, any_string_dtype, nulls_fixture):
+        index = Index(["a", "b", "c"], dtype=any_string_dtype)
+        with pytest.raises(KeyError):
+            index.get_loc(nulls_fixture)
+
+    def test_get_loc_missing(self, any_string_dtype, nulls_fixture):
+        index = Index(["a", "b", nulls_fixture], dtype=any_string_dtype)
+        assert index.get_loc(nulls_fixture) == 2
+
+
+class TestGetIndexer:
+    @pytest.mark.parametrize(
+        "method,expected",
+        [
+            ("pad", [-1, 0, 1, 1]),
+            ("backfill", [0, 0, 1, -1]),
+        ],
+    )
+    def test_get_indexer_strings(self, any_string_dtype, method, expected):
+        expected = np.array(expected, dtype=np.intp)
+        index = Index(["b", "c"], dtype=any_string_dtype)
+        actual = index.get_indexer(["a", "b", "c", "d"], method=method)
+
+        tm.assert_numpy_array_equal(actual, expected)
+
+    def test_get_indexer_strings_raises(self, any_string_dtype):
+        index = Index(["b", "c"], dtype=any_string_dtype)
+
+        msg = "|".join(
+            [
+                "operation 'sub' not supported for dtype 'str",
+                r"unsupported operand type\(s\) for -: 'str' and 'str'",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(["a", "b", "c", "d"], method="nearest")
+
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(["a", "b", "c", "d"], method="pad", tolerance=2)
+
+        with pytest.raises(TypeError, match=msg):
+            index.get_indexer(
+                ["a", "b", "c", "d"], method="pad", tolerance=[2, 2, 2, 2]
+            )
+
+    @pytest.mark.parametrize("null", [None, np.nan, float("nan"), pd.NA])
+    def test_get_indexer_missing(self, any_string_dtype, null, using_infer_string):
+        # NaT and Decimal("NaN") from null_fixture are not supported for string dtype
+        index = Index(["a", "b", null], dtype=any_string_dtype)
+        result = index.get_indexer(["a", null, "c"])
+        if using_infer_string:
+            expected = np.array([0, 2, -1], dtype=np.intp)
+        elif any_string_dtype == "string" and not _equivalent_na(
+            any_string_dtype, null
+        ):
+            expected = np.array([0, -1, -1], dtype=np.intp)
+        else:
+            expected = np.array([0, 2, -1], dtype=np.intp)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestGetIndexerNonUnique:
+    @pytest.mark.parametrize("null", [None, np.nan, float("nan"), pd.NA])
+    def test_get_indexer_non_unique_nas(
+        self, any_string_dtype, null, using_infer_string
+    ):
+        index = Index(["a", "b", null], dtype=any_string_dtype)
+        indexer, missing = index.get_indexer_non_unique(["a", null])
+
+        if using_infer_string:
+            expected_indexer = np.array([0, 2], dtype=np.intp)
+            expected_missing = np.array([], dtype=np.intp)
+        elif any_string_dtype == "string" and not _equivalent_na(
+            any_string_dtype, null
+        ):
+            expected_indexer = np.array([0, -1], dtype=np.intp)
+            expected_missing = np.array([1], dtype=np.intp)
+        else:
+            expected_indexer = np.array([0, 2], dtype=np.intp)
+            expected_missing = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected_indexer)
+        tm.assert_numpy_array_equal(missing, expected_missing)
+
+        # actually non-unique
+        index = Index(["a", null, "b", null], dtype=any_string_dtype)
+        indexer, missing = index.get_indexer_non_unique(["a", null])
+
+        if using_infer_string:
+            expected_indexer = np.array([0, 1, 3], dtype=np.intp)
+        elif any_string_dtype == "string" and not _equivalent_na(
+            any_string_dtype, null
+        ):
+            pass
+        else:
+            expected_indexer = np.array([0, 1, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(indexer, expected_indexer)
+        tm.assert_numpy_array_equal(missing, expected_missing)
+
+
+class TestSliceLocs:
+    @pytest.mark.parametrize(
+        "in_slice,expected",
+        [
+            # error: Slice index must be an integer or None
+            (pd.IndexSlice[::-1], "yxdcb"),
+            (pd.IndexSlice["b":"y":-1], ""),  # type: ignore[misc]
+            (pd.IndexSlice["b"::-1], "b"),  # type: ignore[misc]
+            (pd.IndexSlice[:"b":-1], "yxdcb"),  # type: ignore[misc]
+            (pd.IndexSlice[:"y":-1], "y"),  # type: ignore[misc]
+            (pd.IndexSlice["y"::-1], "yxdcb"),  # type: ignore[misc]
+            (pd.IndexSlice["y"::-4], "yb"),  # type: ignore[misc]
+            # absent labels
+            (pd.IndexSlice[:"a":-1], "yxdcb"),  # type: ignore[misc]
+            (pd.IndexSlice[:"a":-2], "ydb"),  # type: ignore[misc]
+            (pd.IndexSlice["z"::-1], "yxdcb"),  # type: ignore[misc]
+            (pd.IndexSlice["z"::-3], "yc"),  # type: ignore[misc]
+            (pd.IndexSlice["m"::-1], "dcb"),  # type: ignore[misc]
+            (pd.IndexSlice[:"m":-1], "yx"),  # type: ignore[misc]
+            (pd.IndexSlice["a":"a":-1], ""),  # type: ignore[misc]
+            (pd.IndexSlice["z":"z":-1], ""),  # type: ignore[misc]
+            (pd.IndexSlice["m":"m":-1], ""),  # type: ignore[misc]
+        ],
+    )
+    def test_slice_locs_negative_step(self, in_slice, expected, any_string_dtype):
+        index = Index(list("bcdxy"), dtype=any_string_dtype)
+
+        s_start, s_stop = index.slice_locs(in_slice.start, in_slice.stop, in_slice.step)
+        result = index[s_start : s_stop : in_slice.step]
+        expected = Index(list(expected), dtype=any_string_dtype)
+        tm.assert_index_equal(result, expected)
+
+    def test_slice_locs_negative_step_oob(self, any_string_dtype):
+        index = Index(list("bcdxy"), dtype=any_string_dtype)
+
+        result = index[-10:5:1]
+        tm.assert_index_equal(result, index)
+
+        result = index[4:-10:-1]
+        expected = Index(list("yxdcb"), dtype=any_string_dtype)
+        tm.assert_index_equal(result, expected)
+
+    def test_slice_locs_dup(self, any_string_dtype):
+        index = Index(["a", "a", "b", "c", "d", "d"], dtype=any_string_dtype)
+        assert index.slice_locs("a", "d") == (0, 6)
+        assert index.slice_locs(end="d") == (0, 6)
+        assert index.slice_locs("a", "c") == (0, 4)
+        assert index.slice_locs("b", "d") == (2, 6)
+
+        index2 = index[::-1]
+        assert index2.slice_locs("d", "a") == (0, 6)
+        assert index2.slice_locs(end="a") == (0, 6)
+        assert index2.slice_locs("d", "b") == (0, 4)
+        assert index2.slice_locs("c", "a") == (2, 6)
diff --git a/pandas/pandas/tests/indexes/test_any_index.py b/pandas/pandas/tests/indexes/test_any_index.py
new file mode 100644
index 0000000000000..a4c18732ef258
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_any_index.py
@@ -0,0 +1,167 @@
+"""
+Tests that can be parametrized over _any_ Index object.
+"""
+
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+import pandas._testing as tm
+
+
+def test_boolean_context_compat(index):
+    # GH#7897
+    with pytest.raises(ValueError, match="The truth value of a"):
+        if index:
+            pass
+
+    with pytest.raises(ValueError, match="The truth value of a"):
+        bool(index)
+
+
+def test_hash_error(index):
+    with pytest.raises(TypeError, match=f"unhashable type: '{type(index).__name__}'"):
+        hash(index)
+
+
+def test_mutability(index):
+    if not len(index):
+        pytest.skip("Test doesn't make sense for empty index")
+    msg = "Index does not support mutable operations"
+    with pytest.raises(TypeError, match=msg):
+        index[0] = index[0]
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_map_identity_mapping(index, request):
+    # GH#12766
+
+    result = index.map(lambda x: x)
+    if index.dtype == object and (result.dtype == bool or result.dtype == "string"):
+        assert (index == result).all()
+        # TODO: could work that into the 'exact="equiv"'?
+        return  # FIXME: doesn't belong in this file anymore!
+    tm.assert_index_equal(result, index, exact="equiv")
+
+
+def test_wrong_number_names(index):
+    names = index.nlevels * ["apple", "banana", "carrot"]
+    with pytest.raises(ValueError, match="^Length"):
+        index.names = names
+
+
+def test_view_preserves_name(index):
+    assert index.view().name == index.name
+
+
+def test_ravel(index):
+    # GH#19956 ravel returning ndarray is deprecated, in 2.0 returns a view on self
+    res = index.ravel()
+    tm.assert_index_equal(res, index)
+
+
+class TestConversion:
+    def test_to_series(self, index):
+        # assert that we are creating a copy of the index
+
+        ser = index.to_series()
+        assert ser.values is not index.values
+        assert ser.index is not index
+        assert ser.name == index.name
+
+    def test_to_series_with_arguments(self, index):
+        # GH#18699
+
+        # index kwarg
+        ser = index.to_series(index=index)
+
+        assert ser.values is not index.values
+        assert ser.index is index
+        assert ser.name == index.name
+
+        # name kwarg
+        ser = index.to_series(name="__test")
+
+        assert ser.values is not index.values
+        assert ser.index is not index
+        assert ser.name != index.name
+
+    def test_tolist_matches_list(self, index):
+        assert index.tolist() == list(index)
+
+
+class TestRoundTrips:
+    def test_pickle_roundtrip(self, index):
+        result = tm.round_trip_pickle(index)
+        tm.assert_index_equal(result, index, exact=True)
+        if result.nlevels > 1:
+            # GH#8367 round-trip with timezone
+            assert index.equal_levels(result)
+
+    def test_pickle_preserves_name(self, index):
+        original_name, index.name = index.name, "foo"
+        unpickled = tm.round_trip_pickle(index)
+        assert index.equals(unpickled)
+        index.name = original_name
+
+
+class TestIndexing:
+    def test_get_loc_listlike_raises_invalid_index_error(self, index):
+        # and never TypeError
+        key = np.array([0, 1], dtype=np.intp)
+
+        with pytest.raises(InvalidIndexError, match=r"\[0 1\]"):
+            index.get_loc(key)
+
+        with pytest.raises(InvalidIndexError, match=r"\[False  True\]"):
+            index.get_loc(key.astype(bool))
+
+    def test_getitem_ellipsis(self, index):
+        # GH#21282
+        result = index[...]
+        assert result.equals(index)
+        assert result is not index
+
+    def test_slice_keeps_name(self, index):
+        assert index.name == index[1:].name
+
+    @pytest.mark.parametrize("item", [101, "no_int", 2.5])
+    def test_getitem_error(self, index, item):
+        msg = "|".join(
+            [
+                r"index 101 is out of bounds for axis 0 with size [\d]+",
+                re.escape(
+                    "only integers, slices (`:`), ellipsis (`...`), "
+                    "numpy.newaxis (`None`) and integer or boolean arrays "
+                    "are valid indices"
+                ),
+                "index out of bounds",  # string[pyarrow]
+            ]
+        )
+        with pytest.raises(IndexError, match=msg):
+            index[item]
+
+
+class TestRendering:
+    def test_str(self, index):
+        # test the string repr
+        index.name = "foo"
+        assert "'foo'" in str(index)
+        assert type(index).__name__ in str(index)
+
+
+class TestReductions:
+    def test_argmax_axis_invalid(self, index):
+        # GH#23081
+        msg = r"`axis` must be fewer than the number of dimensions \(1\)"
+        with pytest.raises(ValueError, match=msg):
+            index.argmax(axis=1)
+        with pytest.raises(ValueError, match=msg):
+            index.argmin(axis=2)
+        with pytest.raises(ValueError, match=msg):
+            index.min(axis=-2)
+        with pytest.raises(ValueError, match=msg):
+            index.max(axis=-3)
diff --git a/pandas/pandas/tests/indexes/test_base.py b/pandas/pandas/tests/indexes/test_base.py
new file mode 100644
index 0000000000000..5b75bd9afd6df
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_base.py
@@ -0,0 +1,1719 @@
+from collections import defaultdict
+from datetime import datetime
+from functools import partial
+import math
+import operator
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import IS64
+from pandas.errors import InvalidIndexError
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import (
+    is_any_real_numeric_dtype,
+    is_numeric_dtype,
+    is_object_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    IntervalIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    TimedeltaIndex,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.indexes.api import (
+    Index,
+    MultiIndex,
+    _get_combined_index,
+    ensure_index,
+    ensure_index_from_sequences,
+)
+
+
+class TestIndex:
+    @pytest.fixture
+    def simple_index(self) -> Index:
+        return Index(list("abcde"))
+
+    def test_can_hold_identifiers(self, simple_index):
+        index = simple_index
+        key = index[0]
+        assert index._can_hold_identifiers_and_holds_name(key) is True
+
+    @pytest.mark.parametrize("index", ["datetime"], indirect=True)
+    def test_new_axis(self, index):
+        # TODO: a bunch of scattered tests check this deprecation is enforced.
+        #  de-duplicate/centralize them.
+        with pytest.raises(ValueError, match="Multi-dimensional indexing"):
+            # GH#30588 multi-dimensional indexing deprecated
+            index[None, :]
+
+    def test_constructor_regular(self, index):
+        tm.assert_contains_all(index, index)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_constructor_casting(self, index):
+        # casting
+        arr = np.array(index)
+        new_index = Index(arr)
+        tm.assert_contains_all(arr, new_index)
+        tm.assert_index_equal(index, new_index)
+
+    def test_constructor_copy(self, using_infer_string):
+        index = Index(list("abc"), name="name")
+        arr = np.array(index)
+        new_index = Index(arr, copy=True, name="name")
+        assert isinstance(new_index, Index)
+        assert new_index.name == "name"
+        if using_infer_string:
+            tm.assert_extension_array_equal(
+                new_index.values, pd.array(arr, dtype="str")
+            )
+        else:
+            tm.assert_numpy_array_equal(arr, new_index.values)
+        arr[0] = "SOMEBIGLONGSTRING"
+        assert new_index[0] != "SOMEBIGLONGSTRING"
+
+    @pytest.mark.parametrize("cast_as_obj", [True, False])
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range(
+                "2015-01-01 10:00",
+                freq="D",
+                periods=3,
+                tz="US/Eastern",
+                name="Green Eggs & Ham",
+            ),  # DTI with tz
+            date_range("2015-01-01 10:00", freq="D", periods=3),  # DTI no tz
+            timedelta_range("1 days", freq="D", periods=3),  # td
+            period_range("2015-01-01", freq="D", periods=3),  # period
+        ],
+    )
+    def test_constructor_from_index_dtlike(self, cast_as_obj, index):
+        if cast_as_obj:
+            result = Index(index.astype(object))
+            assert result.dtype == np.dtype(object)
+            if isinstance(index, DatetimeIndex):
+                # GH#23524 check that Index(dti, dtype=object) does not
+                #  incorrectly raise ValueError, and that nanoseconds are not
+                #  dropped
+                index += pd.Timedelta(nanoseconds=50)
+                result = Index(index, dtype=object)
+                assert result.dtype == np.object_
+                assert list(result) == list(index)
+        else:
+            result = Index(index)
+
+            tm.assert_index_equal(result, index)
+
+    @pytest.mark.parametrize(
+        "index,has_tz",
+        [
+            (
+                date_range("2015-01-01 10:00", freq="D", periods=3, tz="US/Eastern"),
+                True,
+            ),  # datetimetz
+            (timedelta_range("1 days", freq="D", periods=3), False),  # td
+            (period_range("2015-01-01", freq="D", periods=3), False),  # period
+        ],
+    )
+    def test_constructor_from_series_dtlike(self, index, has_tz):
+        result = Index(Series(index))
+        tm.assert_index_equal(result, index)
+
+        if has_tz:
+            assert result.tz == index.tz
+
+    def test_constructor_from_series_freq(self):
+        # GH 6273
+        # create from a series, passing a freq
+        dts = ["1-1-1990", "2-1-1990", "3-1-1990", "4-1-1990", "5-1-1990"]
+        expected = DatetimeIndex(dts, freq="MS")
+
+        s = Series(pd.to_datetime(dts))
+        result = DatetimeIndex(s, freq="MS")
+
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_from_frame_series_freq(self, using_infer_string):
+        # GH 6273
+        # create from a series, passing a freq
+        dts = ["1-1-1990", "2-1-1990", "3-1-1990", "4-1-1990", "5-1-1990"]
+        expected = DatetimeIndex(dts, freq="MS")
+
+        df = DataFrame(np.random.default_rng(2).random((5, 3)))
+        df["date"] = dts
+        result = DatetimeIndex(df["date"], freq="MS")
+        dtype = object if not using_infer_string else "str"
+        assert df["date"].dtype == dtype
+        expected.name = "date"
+        tm.assert_index_equal(result, expected)
+
+        expected = Series(dts, name="date")
+        tm.assert_series_equal(df["date"], expected)
+
+        # GH 6274
+        # infer freq of same
+        if not using_infer_string:
+            # Doesn't work with arrow strings
+            freq = pd.infer_freq(df["date"])
+            assert freq == "MS"
+
+    def test_constructor_int_dtype_nan(self):
+        # see gh-15187
+        data = [np.nan]
+        expected = Index(data, dtype=np.float64)
+        result = Index(data, dtype="float")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "klass,dtype,na_val",
+        [
+            (Index, np.float64, np.nan),
+            (DatetimeIndex, "datetime64[s]", pd.NaT),
+        ],
+    )
+    def test_index_ctor_infer_nan_nat(self, klass, dtype, na_val):
+        # GH 13467
+        na_list = [na_val, na_val]
+        expected = klass(na_list)
+        assert expected.dtype == dtype
+
+        result = Index(na_list)
+        tm.assert_index_equal(result, expected)
+
+        result = Index(np.array(na_list))
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "vals,dtype",
+        [
+            ([1, 2, 3, 4, 5], "int"),
+            ([1.1, np.nan, 2.2, 3.0], "float"),
+            (["A", "B", "C", np.nan], "obj"),
+        ],
+    )
+    def test_constructor_simple_new(self, vals, dtype):
+        index = Index(vals, name=dtype)
+        result = index._simple_new(index.values, dtype)
+        tm.assert_index_equal(result, index)
+
+    @pytest.mark.parametrize("attr", ["values", "asi8"])
+    @pytest.mark.parametrize("klass", [Index, DatetimeIndex])
+    def test_constructor_dtypes_datetime(self, tz_naive_fixture, attr, klass):
+        # Test constructing with a datetimetz dtype
+        # .values produces numpy datetimes, so these are considered naive
+        # .asi8 produces integers, so these are considered epoch timestamps
+        # ^the above will be true in a later version. Right now we `.view`
+        # the i8 values as NS_DTYPE, effectively treating them as wall times.
+        index = date_range("2011-01-01", periods=5)
+        arg = getattr(index, attr)
+        index = index.tz_localize(tz_naive_fixture)
+        dtype = index.dtype
+
+        # As of 2.0 astype raises on dt64.astype(dt64tz)
+        err = tz_naive_fixture is not None
+        msg = "Cannot use .astype to convert from timezone-naive dtype to"
+
+        if attr == "asi8":
+            result = DatetimeIndex(arg).tz_localize(tz_naive_fixture)
+            tm.assert_index_equal(result, index)
+        elif klass is Index:
+            with pytest.raises(TypeError, match="unexpected keyword"):
+                klass(arg, tz=tz_naive_fixture)
+        else:
+            result = klass(arg, tz=tz_naive_fixture)
+            tm.assert_index_equal(result, index)
+
+        if attr == "asi8":
+            if err:
+                with pytest.raises(TypeError, match=msg):
+                    DatetimeIndex(arg).astype(dtype)
+            else:
+                result = DatetimeIndex(arg).astype(dtype)
+                tm.assert_index_equal(result, index)
+        else:
+            result = klass(arg, dtype=dtype)
+            tm.assert_index_equal(result, index)
+
+        if attr == "asi8":
+            result = DatetimeIndex(list(arg)).tz_localize(tz_naive_fixture)
+            tm.assert_index_equal(result, index)
+        elif klass is Index:
+            with pytest.raises(TypeError, match="unexpected keyword"):
+                klass(arg, tz=tz_naive_fixture)
+        else:
+            result = klass(list(arg), tz=tz_naive_fixture)
+            tm.assert_index_equal(result, index)
+
+        if attr == "asi8":
+            if err:
+                with pytest.raises(TypeError, match=msg):
+                    DatetimeIndex(list(arg)).astype(dtype)
+            else:
+                result = DatetimeIndex(list(arg)).astype(dtype)
+                tm.assert_index_equal(result, index)
+        else:
+            result = klass(list(arg), dtype=dtype)
+            tm.assert_index_equal(result, index)
+
+    @pytest.mark.parametrize("attr", ["values", "asi8"])
+    @pytest.mark.parametrize("klass", [Index, TimedeltaIndex])
+    def test_constructor_dtypes_timedelta(self, attr, klass):
+        index = timedelta_range("1 days", periods=5)
+        index = index._with_freq(None)  # won't be preserved by constructors
+        dtype = index.dtype
+
+        values = getattr(index, attr)
+
+        result = klass(values, dtype=dtype)
+        tm.assert_index_equal(result, index)
+
+        result = klass(list(values), dtype=dtype)
+        tm.assert_index_equal(result, index)
+
+    @pytest.mark.parametrize("value", [[], iter([]), (_ for _ in [])])
+    @pytest.mark.parametrize(
+        "klass",
+        [
+            Index,
+            CategoricalIndex,
+            DatetimeIndex,
+            TimedeltaIndex,
+        ],
+    )
+    def test_constructor_empty(self, value, klass):
+        empty = klass(value)
+        assert isinstance(empty, klass)
+        assert not len(empty)
+
+    @pytest.mark.parametrize(
+        "empty,klass",
+        [
+            (PeriodIndex([], freq="D"), PeriodIndex),
+            (PeriodIndex(iter([]), freq="D"), PeriodIndex),
+            (PeriodIndex((_ for _ in []), freq="D"), PeriodIndex),
+            (RangeIndex(step=1), RangeIndex),
+            (MultiIndex(levels=[[1, 2], ["blue", "red"]], codes=[[], []]), MultiIndex),
+        ],
+    )
+    def test_constructor_empty_special(self, empty, klass):
+        assert isinstance(empty, klass)
+        assert not len(empty)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            "datetime",
+            "float64",
+            "float32",
+            "int64",
+            "int32",
+            "period",
+            "range",
+            "repeats",
+            "timedelta",
+            "tuples",
+            "uint64",
+            "uint32",
+        ],
+        indirect=True,
+    )
+    def test_view_with_args(self, index):
+        index.view("i8")
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            "string",
+            pytest.param("categorical", marks=pytest.mark.xfail(reason="gh-25464")),
+            "bool-object",
+            "bool-dtype",
+            "empty",
+        ],
+        indirect=True,
+    )
+    def test_view_with_args_object_array_raises(self, index):
+        if index.dtype == bool:
+            msg = "When changing to a larger dtype"
+            with pytest.raises(ValueError, match=msg):
+                index.view("i8")
+        else:
+            msg = (
+                r"Cannot change data-type for array of references\.|"
+                r"Cannot change data-type for object array\.|"
+                r"Cannot change data-type for array of strings\.|"
+            )
+            with pytest.raises(TypeError, match=msg):
+                index.view("i8")
+
+    @pytest.mark.parametrize(
+        "index",
+        ["int64", "int32", "range"],
+        indirect=True,
+    )
+    def test_astype(self, index):
+        casted = index.astype("i8")
+
+        # it works!
+        casted.get_loc(5)
+
+        # pass on name
+        index.name = "foobar"
+        casted = index.astype("i8")
+        assert casted.name == "foobar"
+
+    def test_equals_object(self):
+        # same
+        assert Index(["a", "b", "c"]).equals(Index(["a", "b", "c"]))
+
+    @pytest.mark.parametrize(
+        "comp", [Index(["a", "b"]), Index(["a", "b", "d"]), ["a", "b", "c"]]
+    )
+    def test_not_equals_object(self, comp):
+        assert not Index(["a", "b", "c"]).equals(comp)
+
+    def test_identical(self):
+        # index
+        i1 = Index(["a", "b", "c"])
+        i2 = Index(["a", "b", "c"])
+
+        assert i1.identical(i2)
+
+        i1 = i1.rename("foo")
+        assert i1.equals(i2)
+        assert not i1.identical(i2)
+
+        i2 = i2.rename("foo")
+        assert i1.identical(i2)
+
+        i3 = Index([("a", "a"), ("a", "b"), ("b", "a")])
+        i4 = Index([("a", "a"), ("a", "b"), ("b", "a")], tupleize_cols=False)
+        assert not i3.identical(i4)
+
+    def test_is_(self):
+        ind = Index(range(10))
+        assert ind.is_(ind)
+        assert ind.is_(ind.view().view().view().view())
+        assert not ind.is_(Index(range(10)))
+        assert not ind.is_(ind.copy())
+        assert not ind.is_(ind.copy(deep=False))
+        assert not ind.is_(ind[:])
+        assert not ind.is_(np.array(range(10)))
+
+        # quasi-implementation dependent
+        assert ind.is_(ind.view())
+        ind2 = ind.view()
+        ind2.name = "bob"
+        assert ind.is_(ind2)
+        assert ind2.is_(ind)
+        # doesn't matter if Indices are *actually* views of underlying data,
+        assert not ind.is_(Index(ind.values))
+        arr = np.array(range(1, 11))
+        ind1 = Index(arr, copy=False)
+        ind2 = Index(arr, copy=False)
+        assert not ind1.is_(ind2)
+
+    def test_asof_numeric_vs_bool_raises(self):
+        left = Index([1, 2, 3])
+        right = Index([True, False], dtype=object)
+
+        msg = "Cannot compare dtypes int64 and bool"
+        with pytest.raises(TypeError, match=msg):
+            left.asof(right[0])
+        # TODO: should right.asof(left[0]) also raise?
+
+        with pytest.raises(InvalidIndexError, match=re.escape(str(right))):
+            left.asof(right)
+
+        with pytest.raises(InvalidIndexError, match=re.escape(str(left))):
+            right.asof(left)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_booleanindex(self, index):
+        bool_index = np.ones(len(index), dtype=bool)
+        bool_index[5:30:2] = False
+
+        sub_index = index[bool_index]
+
+        for i, val in enumerate(sub_index):
+            assert sub_index.get_loc(val) == i
+
+        sub_index = index[list(bool_index)]
+        for i, val in enumerate(sub_index):
+            assert sub_index.get_loc(val) == i
+
+    def test_fancy(self, simple_index):
+        index = simple_index
+        sl = index[[1, 2, 3]]
+        for i in sl:
+            assert i == sl[sl.get_loc(i)]
+
+    @pytest.mark.parametrize(
+        "index",
+        ["string", "int64", "int32", "uint64", "uint32", "float64", "float32"],
+        indirect=True,
+    )
+    @pytest.mark.parametrize("dtype", [int, np.bool_])
+    def test_empty_fancy(self, index, dtype, request, using_infer_string):
+        if dtype is np.bool_ and using_infer_string and index.dtype == "string":
+            request.applymarker(pytest.mark.xfail(reason="numpy behavior is buggy"))
+        empty_arr = np.array([], dtype=dtype)
+        empty_index = type(index)([], dtype=index.dtype)
+
+        assert index[[]].identical(empty_index)
+        if dtype == np.bool_:
+            with pytest.raises(ValueError, match="length of the boolean indexer"):
+                assert index[empty_arr].identical(empty_index)
+        else:
+            assert index[empty_arr].identical(empty_index)
+
+    @pytest.mark.parametrize(
+        "index",
+        ["string", "int64", "int32", "uint64", "uint32", "float64", "float32"],
+        indirect=True,
+    )
+    def test_empty_fancy_raises(self, index):
+        # DatetimeIndex is excluded, because it overrides getitem and should
+        # be tested separately.
+        empty_farr = np.array([], dtype=np.float64)
+        empty_index = type(index)([], dtype=index.dtype)
+
+        assert index[[]].identical(empty_index)
+        # np.ndarray only accepts ndarray of int & bool dtypes, so should Index
+        msg = r"arrays used as indices must be of integer"
+        with pytest.raises(IndexError, match=msg):
+            index[empty_farr]
+
+    def test_union_dt_as_obj(self, simple_index):
+        # TODO: Replace with fixturesult
+        index = simple_index
+        date_index = date_range("2019-01-01", periods=10)
+        first_cat = index.union(date_index)
+        second_cat = index.union(index)
+
+        appended = Index(np.append(index, date_index.astype("O")))
+
+        tm.assert_index_equal(first_cat, appended)
+        tm.assert_index_equal(second_cat, index)
+        tm.assert_contains_all(index, first_cat)
+        tm.assert_contains_all(index, second_cat)
+        tm.assert_contains_all(date_index, first_cat)
+
+    def test_map_with_tuples(self):
+        # GH 12766
+
+        # Test that returning a single tuple from an Index
+        #   returns an Index.
+        index = Index(np.arange(3), dtype=np.int64)
+        result = index.map(lambda x: (x,))
+        expected = Index([(i,) for i in index])
+        tm.assert_index_equal(result, expected)
+
+        # Test that returning a tuple from a map of a single index
+        #   returns a MultiIndex object.
+        result = index.map(lambda x: (x, x == 1))
+        expected = MultiIndex.from_tuples([(i, i == 1) for i in index])
+        tm.assert_index_equal(result, expected)
+
+    def test_map_with_tuples_mi(self):
+        # Test that returning a single object from a MultiIndex
+        #   returns an Index.
+        first_level = ["foo", "bar", "baz"]
+        multi_index = MultiIndex.from_tuples(zip(first_level, [1, 2, 3]))
+        reduced_index = multi_index.map(lambda x: x[0])
+        tm.assert_index_equal(reduced_index, Index(first_level))
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("2020-01-01", freq="D", periods=10),
+            period_range("2020-01-01", freq="D", periods=10),
+            timedelta_range("1 day", periods=10),
+        ],
+    )
+    def test_map_tseries_indices_return_index(self, index):
+        expected = Index([1] * 10)
+        result = index.map(lambda x: 1)
+        tm.assert_index_equal(expected, result)
+
+    def test_map_tseries_indices_accsr_return_index(self):
+        date_index = DatetimeIndex(
+            date_range("2020-01-01", periods=24, freq="h"), name="hourly"
+        )
+        result = date_index.map(lambda x: x.hour)
+        expected = Index(np.arange(24, dtype="int64"), name="hourly")
+        tm.assert_index_equal(result, expected, exact=True)
+
+    @pytest.mark.parametrize(
+        "mapper",
+        [
+            lambda values, index: {i: e for e, i in zip(values, index)},
+            lambda values, index: Series(values, index),
+        ],
+    )
+    def test_map_dictlike_simple(self, mapper):
+        # GH 12756
+        expected = Index(["foo", "bar", "baz"])
+        index = Index(np.arange(3), dtype=np.int64)
+        result = index.map(mapper(expected.values, index))
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "mapper",
+        [
+            lambda values, index: {i: e for e, i in zip(values, index)},
+            lambda values, index: Series(values, index),
+        ],
+    )
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_map_dictlike(self, index, mapper, request):
+        # GH 12756
+        if isinstance(index, CategoricalIndex):
+            pytest.skip("Tested in test_categorical")
+        elif not index.is_unique:
+            pytest.skip("Cannot map duplicated index")
+
+        rng = np.arange(len(index), 0, -1, dtype=np.int64)
+
+        if index.empty:
+            # to match proper result coercion for uints
+            expected = Index([])
+        elif is_numeric_dtype(index.dtype):
+            expected = index._constructor(rng, dtype=index.dtype)
+        elif type(index) is Index and index.dtype != object:
+            # i.e. EA-backed, for now just Nullable
+            expected = Index(rng, dtype=index.dtype)
+        else:
+            expected = Index(rng)
+
+        result = index.map(mapper(expected, index))
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "mapper",
+        [Series(["foo", 2.0, "baz"], index=[0, 2, -1]), {0: "foo", 2: 2.0, -1: "baz"}],
+    )
+    def test_map_with_non_function_missing_values(self, mapper):
+        # GH 12756
+        expected = Index([2.0, np.nan, "foo"])
+        result = Index([2, 1, 0]).map(mapper)
+
+        tm.assert_index_equal(expected, result)
+
+    def test_map_na_exclusion(self):
+        index = Index([1.5, np.nan, 3, np.nan, 5])
+
+        result = index.map(lambda x: x * 2, na_action="ignore")
+        expected = index * 2
+        tm.assert_index_equal(result, expected)
+
+    def test_map_defaultdict(self):
+        index = Index([1, 2, 3])
+        default_dict = defaultdict(lambda: "blank")
+        default_dict[1] = "stuff"
+        result = index.map(default_dict)
+        expected = Index(["stuff", "blank", "blank"])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("name,expected", [("foo", "foo"), ("bar", None)])
+    def test_append_empty_preserve_name(self, name, expected):
+        left = Index([], name="foo")
+        right = Index([1, 2, 3], name=name)
+
+        result = left.append(right)
+        assert result.name == expected
+
+    @pytest.mark.parametrize(
+        "index, expected",
+        [
+            ("string", False),
+            ("bool-object", False),
+            ("bool-dtype", False),
+            ("categorical", False),
+            ("int64", True),
+            ("int32", True),
+            ("uint64", True),
+            ("uint32", True),
+            ("datetime", False),
+            ("float64", True),
+            ("float32", True),
+        ],
+        indirect=["index"],
+    )
+    def test_is_numeric(self, index, expected):
+        assert is_any_real_numeric_dtype(index) is expected
+
+    @pytest.mark.parametrize(
+        "index, expected",
+        [
+            ("string", True),
+            ("bool-object", True),
+            ("bool-dtype", False),
+            ("categorical", False),
+            ("int64", False),
+            ("int32", False),
+            ("uint64", False),
+            ("uint32", False),
+            ("datetime", False),
+            ("float64", False),
+            ("float32", False),
+        ],
+        indirect=["index"],
+    )
+    def test_is_object(self, index, expected, using_infer_string):
+        if using_infer_string and index.dtype == "string" and expected:
+            expected = False
+        assert is_object_dtype(index) is expected
+
+    def test_summary(self, index):
+        index._summary()
+
+    def test_logical_compat(self, all_boolean_reductions, simple_index):
+        index = simple_index
+        left = getattr(index, all_boolean_reductions)()
+        assert left == getattr(index.values, all_boolean_reductions)()
+        right = getattr(index.to_series(), all_boolean_reductions)()
+        # left might not match right exactly in e.g. string cases where the
+        # because we use np.any/all instead of .any/all
+        assert bool(left) == bool(right)
+
+    @pytest.mark.parametrize(
+        "index", ["string", "int64", "int32", "float64", "float32"], indirect=True
+    )
+    def test_drop_by_str_label(self, index):
+        n = len(index)
+        drop = index[list(range(5, 10))]
+        dropped = index.drop(drop)
+
+        expected = index[list(range(5)) + list(range(10, n))]
+        tm.assert_index_equal(dropped, expected)
+
+        dropped = index.drop(index[0])
+        expected = index[1:]
+        tm.assert_index_equal(dropped, expected)
+
+    @pytest.mark.parametrize(
+        "index", ["string", "int64", "int32", "float64", "float32"], indirect=True
+    )
+    @pytest.mark.parametrize("keys", [["foo", "bar"], ["1", "bar"]])
+    def test_drop_by_str_label_raises_missing_keys(self, index, keys):
+        with pytest.raises(KeyError, match=""):
+            index.drop(keys)
+
+    @pytest.mark.parametrize(
+        "index", ["string", "int64", "int32", "float64", "float32"], indirect=True
+    )
+    def test_drop_by_str_label_errors_ignore(self, index):
+        n = len(index)
+        drop = index[list(range(5, 10))]
+        mixed = drop.tolist() + ["foo"]
+        dropped = index.drop(mixed, errors="ignore")
+
+        expected = index[list(range(5)) + list(range(10, n))]
+        tm.assert_index_equal(dropped, expected)
+
+        dropped = index.drop(["foo", "bar"], errors="ignore")
+        expected = index[list(range(n))]
+        tm.assert_index_equal(dropped, expected)
+
+    def test_drop_by_numeric_label_loc(self):
+        # TODO: Parametrize numeric and str tests after self.strIndex fixture
+        index = Index([1, 2, 3])
+        dropped = index.drop(1)
+        expected = Index([2, 3])
+
+        tm.assert_index_equal(dropped, expected)
+
+    def test_drop_by_numeric_label_raises_missing_keys(self):
+        index = Index([1, 2, 3])
+        with pytest.raises(KeyError, match=""):
+            index.drop([3, 4])
+
+    @pytest.mark.parametrize(
+        "key,expected", [(4, Index([1, 2, 3])), ([3, 4, 5], Index([1, 2]))]
+    )
+    def test_drop_by_numeric_label_errors_ignore(self, key, expected):
+        index = Index([1, 2, 3])
+        dropped = index.drop(key, errors="ignore")
+
+        tm.assert_index_equal(dropped, expected)
+
+    @pytest.mark.parametrize(
+        "values",
+        [["a", "b", ("c", "d")], ["a", ("c", "d"), "b"], [("c", "d"), "a", "b"]],
+    )
+    @pytest.mark.parametrize("to_drop", [[("c", "d"), "a"], ["a", ("c", "d")]])
+    def test_drop_tuple(self, values, to_drop):
+        # GH 18304
+        index = Index(values)
+        expected = Index(["b"], dtype=object)
+
+        result = index.drop(to_drop)
+        tm.assert_index_equal(result, expected)
+
+        removed = index.drop(to_drop[0])
+        for drop_me in to_drop[1], [to_drop[1]]:
+            result = removed.drop(drop_me)
+            tm.assert_index_equal(result, expected)
+
+        removed = index.drop(to_drop[1])
+        msg = rf"\"\[{re.escape(to_drop[1].__repr__())}\] not found in axis\""
+        for drop_me in to_drop[1], [to_drop[1]]:
+            with pytest.raises(KeyError, match=msg):
+                removed.drop(drop_me)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_drop_with_duplicates_in_index(self, index):
+        # GH38051
+        if len(index) == 0 or isinstance(index, MultiIndex):
+            pytest.skip("Test doesn't make sense for empty MultiIndex")
+        if isinstance(index, IntervalIndex) and not IS64:
+            pytest.skip("Cannot test IntervalIndex with int64 dtype on 32 bit platform")
+        index = index.unique().repeat(2)
+        expected = index[2:]
+        result = index.drop(index[0])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "attr",
+        [
+            "is_monotonic_increasing",
+            "is_monotonic_decreasing",
+            "_is_strictly_monotonic_increasing",
+            "_is_strictly_monotonic_decreasing",
+        ],
+    )
+    def test_is_monotonic_incomparable(self, attr):
+        index = Index([5, datetime.now(), 7])
+        assert not getattr(index, attr)
+
+    @pytest.mark.parametrize("values", [["foo", "bar", "quux"], {"foo", "bar", "quux"}])
+    @pytest.mark.parametrize(
+        "index,expected",
+        [
+            (["qux", "baz", "foo", "bar"], [False, False, True, True]),
+            ([], []),  # empty
+        ],
+    )
+    def test_isin(self, values, index, expected):
+        index = Index(index)
+        result = index.isin(values)
+        expected = np.array(expected, dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_isin_nan_common_object(
+        self, nulls_fixture, nulls_fixture2, using_infer_string
+    ):
+        # Test cartesian product of null fixtures and ensure that we don't
+        # mangle the various types (save a corner case with PyPy)
+        idx = Index(["a", nulls_fixture])
+
+        # all nans are the same
+        if (
+            isinstance(nulls_fixture, float)
+            and isinstance(nulls_fixture2, float)
+            and math.isnan(nulls_fixture)
+            and math.isnan(nulls_fixture2)
+        ):
+            tm.assert_numpy_array_equal(
+                idx.isin([nulls_fixture2]),
+                np.array([False, True]),
+            )
+
+        elif nulls_fixture is nulls_fixture2:  # should preserve NA type
+            tm.assert_numpy_array_equal(
+                idx.isin([nulls_fixture2]),
+                np.array([False, True]),
+            )
+
+        elif using_infer_string and idx.dtype == "string":
+            tm.assert_numpy_array_equal(
+                idx.isin([nulls_fixture2]),
+                np.array([False, True]),
+            )
+
+        else:
+            tm.assert_numpy_array_equal(
+                idx.isin([nulls_fixture2]),
+                np.array([False, False]),
+            )
+
+    def test_isin_nan_common_float64(self, nulls_fixture, float_numpy_dtype):
+        dtype = float_numpy_dtype
+
+        if nulls_fixture is pd.NaT or nulls_fixture is pd.NA:
+            # Check 1) that we cannot construct a float64 Index with this value
+            #  and 2) that with an NaN we do not have .isin(nulls_fixture)
+            msg = (
+                r"float\(\) argument must be a string or a (real )?number, "
+                f"not {type(nulls_fixture).__name__!r}"
+            )
+            with pytest.raises(TypeError, match=msg):
+                Index([1.0, nulls_fixture], dtype=dtype)
+
+            idx = Index([1.0, np.nan], dtype=dtype)
+            assert not idx.isin([nulls_fixture]).any()
+            return
+
+        idx = Index([1.0, nulls_fixture], dtype=dtype)
+        res = idx.isin([np.nan])
+        tm.assert_numpy_array_equal(res, np.array([False, True]))
+
+        # we cannot compare NaT with NaN
+        res = idx.isin([pd.NaT])
+        tm.assert_numpy_array_equal(res, np.array([False, False]))
+
+    @pytest.mark.parametrize("level", [0, -1])
+    @pytest.mark.parametrize(
+        "index",
+        [
+            ["qux", "baz", "foo", "bar"],
+            np.array([1.0, 2.0, 3.0, 4.0], dtype=np.float64),
+        ],
+    )
+    def test_isin_level_kwarg(self, level, index):
+        index = Index(index)
+        values = index.tolist()[-2:] + ["nonexisting"]
+
+        expected = np.array([False, False, True, True])
+        tm.assert_numpy_array_equal(expected, index.isin(values, level=level))
+
+        index.name = "foobar"
+        tm.assert_numpy_array_equal(expected, index.isin(values, level="foobar"))
+
+    def test_isin_level_kwarg_bad_level_raises(self, index):
+        for level in [10, index.nlevels, -(index.nlevels + 1)]:
+            with pytest.raises(IndexError, match="Too many levels"):
+                index.isin([], level=level)
+
+    @pytest.mark.parametrize("label", [1.0, "foobar", "xyzzy", np.nan])
+    def test_isin_level_kwarg_bad_label_raises(self, label, index):
+        if isinstance(index, MultiIndex):
+            index = index.rename(["foo", "bar"] + index.names[2:])
+            msg = f"'Level {label} not found'"
+        else:
+            index = index.rename("foo")
+            msg = rf"Requested level \({label}\) does not match index name \(foo\)"
+        with pytest.raises(KeyError, match=msg):
+            index.isin([], level=label)
+
+    @pytest.mark.parametrize("empty", [[], Series(dtype=object), np.array([])])
+    def test_isin_empty(self, empty):
+        # see gh-16991
+        index = Index(["a", "b"])
+        expected = np.array([False, False])
+
+        result = index.isin(empty)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_isin_string_null(self, string_dtype_no_object):
+        # GH#55821
+        index = Index(["a", "b"], dtype=string_dtype_no_object)
+        result = index.isin([None])
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            [1, 2, 3, 4],
+            [1.0, 2.0, 3.0, 4.0],
+            [True, True, True, True],
+            ["foo", "bar", "baz", "qux"],
+            date_range("2018-01-01", freq="D", periods=4),
+        ],
+    )
+    def test_boolean_cmp(self, values):
+        index = Index(values)
+        result = index == values
+        expected = np.array([True, True, True, True], dtype=bool)
+
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    @pytest.mark.parametrize("name,level", [(None, 0), ("a", "a")])
+    def test_get_level_values(self, index, name, level):
+        expected = index.copy()
+        if name:
+            expected.name = name
+
+        result = expected.get_level_values(level)
+        tm.assert_index_equal(result, expected)
+
+    def test_slice_keep_name(self):
+        index = Index(["a", "b"], name="asdf")
+        assert index.name == index[1:].name
+
+    def test_slice_is_unique(self):
+        # GH 57911
+        index = Index([1, 1, 2, 3, 4])
+        assert not index.is_unique
+        filtered_index = index[2:].copy()
+        assert filtered_index.is_unique
+
+    def test_slice_is_montonic(self):
+        """Test that is_monotonic_decreasing is correct on slices."""
+        # GH 57911
+        index = Index([1, 2, 3, 3])
+        assert not index.is_monotonic_decreasing
+
+        filtered_index = index[2:].copy()
+        assert filtered_index.is_monotonic_decreasing
+        assert filtered_index.is_monotonic_increasing
+
+        filtered_index = index[1:].copy()
+        assert not filtered_index.is_monotonic_decreasing
+        assert filtered_index.is_monotonic_increasing
+
+        filtered_index = index[:].copy()
+        assert not filtered_index.is_monotonic_decreasing
+        assert filtered_index.is_monotonic_increasing
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            "string",
+            "datetime",
+            "int64",
+            "int32",
+            "uint64",
+            "uint32",
+            "float64",
+            "float32",
+        ],
+        indirect=True,
+    )
+    def test_join_self(self, index, join_type):
+        result = index.join(index, how=join_type)
+        expected = index
+        if join_type == "outer":
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["strip", "rstrip", "lstrip"])
+    def test_str_attribute(self, method):
+        # GH9068
+        index = Index([" jack", "jill ", " jesse ", "frank"])
+        expected = Index([getattr(str, method)(x) for x in index.values])
+
+        result = getattr(index.str, method)()
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(range(5)),
+            date_range("2020-01-01", periods=10),
+            MultiIndex.from_tuples([("foo", "1"), ("bar", "3")]),
+            period_range(start="2000", end="2010", freq="Y"),
+        ],
+    )
+    def test_str_attribute_raises(self, index):
+        with pytest.raises(AttributeError, match="only use .str accessor"):
+            index.str.repeat(2)
+
+    @pytest.mark.parametrize(
+        "expand,expected",
+        [
+            (None, Index([["a", "b", "c"], ["d", "e"], ["f"]])),
+            (False, Index([["a", "b", "c"], ["d", "e"], ["f"]])),
+            (
+                True,
+                MultiIndex.from_tuples(
+                    [("a", "b", "c"), ("d", "e", np.nan), ("f", np.nan, np.nan)]
+                ),
+            ),
+        ],
+    )
+    def test_str_split(self, expand, expected):
+        index = Index(["a b c", "d e", "f"])
+        if expand is not None:
+            result = index.str.split(expand=expand)
+        else:
+            result = index.str.split()
+
+        tm.assert_index_equal(result, expected)
+
+    def test_str_bool_return(self):
+        # test boolean case, should return np.array instead of boolean Index
+        index = Index(["a1", "a2", "b1", "b2"])
+        result = index.str.startswith("a")
+        expected = np.array([True, True, False, False])
+
+        tm.assert_numpy_array_equal(result, expected)
+        assert isinstance(result, np.ndarray)
+
+    def test_str_bool_series_indexing(self):
+        index = Index(["a1", "a2", "b1", "b2"])
+        s = Series(range(4), index=index)
+
+        result = s[s.index.str.startswith("a")]
+        expected = Series(range(2), index=["a1", "a2"])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index,expected", [(list("abcd"), True), (range(4), False)]
+    )
+    def test_tab_completion(self, index, expected):
+        # GH 9910
+        index = Index(index)
+        result = "str" in dir(index)
+        assert result == expected
+
+    def test_indexing_doesnt_change_class(self):
+        index = Index([1, 2, 3, "a", "b", "c"])
+
+        assert index[1:3].identical(Index([2, 3], dtype=np.object_))
+        assert index[[0, 1]].identical(Index([1, 2], dtype=np.object_))
+
+    def test_outer_join_sort(self):
+        left_index = Index(np.random.default_rng(2).permutation(15))
+        right_index = date_range("2020-01-01", periods=10)
+
+        with tm.assert_produces_warning(RuntimeWarning, match="not supported between"):
+            result = left_index.join(right_index, how="outer")
+
+        with tm.assert_produces_warning(RuntimeWarning, match="not supported between"):
+            expected = left_index.astype(object).union(right_index.astype(object))
+
+        tm.assert_index_equal(result, expected)
+
+    def test_take_fill_value(self):
+        # GH 12631
+        index = Index(list("ABC"), name="xxx")
+        result = index.take(np.array([1, 0, -1]))
+        expected = Index(list("BAC"), name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = index.take(np.array([1, 0, -1]), fill_value=True)
+        expected = Index(["B", "A", np.nan], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = index.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = Index(["B", "A", "C"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+    def test_take_fill_value_none_raises(self):
+        index = Index(list("ABC"), name="xxx")
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            index.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            index.take(np.array([1, 0, -5]), fill_value=True)
+
+    def test_take_bad_bounds_raises(self):
+        index = Index(list("ABC"), name="xxx")
+        with pytest.raises(IndexError, match="out of bounds"):
+            index.take(np.array([1, -5]))
+
+    @pytest.mark.parametrize("name", [None, "foobar"])
+    @pytest.mark.parametrize(
+        "labels",
+        [
+            [],
+            np.array([]),
+            ["A", "B", "C"],
+            ["C", "B", "A"],
+            np.array(["A", "B", "C"]),
+            np.array(["C", "B", "A"]),
+            # Must preserve name even if dtype changes
+            date_range("20130101", periods=3).values,
+            date_range("20130101", periods=3).tolist(),
+        ],
+    )
+    def test_reindex_preserves_name_if_target_is_list_or_ndarray(self, name, labels):
+        # GH6552
+        index = Index([0, 1, 2])
+        index.name = name
+        assert index.reindex(labels)[0].name == name
+
+    @pytest.mark.parametrize("labels", [[], np.array([]), np.array([], dtype=np.int64)])
+    def test_reindex_preserves_type_if_target_is_empty_list_or_array(self, labels):
+        # GH7774
+        index = Index(list("abc"))
+        assert index.reindex(labels)[0].dtype.type == index.dtype.type
+
+    def test_reindex_doesnt_preserve_type_if_target_is_empty_index(self):
+        # GH7774
+        index = Index(list("abc"))
+        labels = DatetimeIndex([])
+        dtype = np.datetime64
+        assert index.reindex(labels)[0].dtype.type == dtype
+
+    def test_reindex_doesnt_preserve_type_if_target_is_empty_index_numeric(
+        self, any_real_numpy_dtype
+    ):
+        # GH7774
+        dtype = any_real_numpy_dtype
+        index = Index(list("abc"))
+        labels = Index([], dtype=dtype)
+        assert index.reindex(labels)[0].dtype == dtype
+
+    def test_reindex_no_type_preserve_target_empty_mi(self):
+        index = Index(list("abc"))
+        result = index.reindex(
+            MultiIndex([Index([], np.int64), Index([], np.float64)], [[], []])
+        )[0]
+        assert result.levels[0].dtype.type == np.int64
+        assert result.levels[1].dtype.type == np.float64
+
+    def test_reindex_ignoring_level(self):
+        # GH#35132
+        idx = Index([1, 2, 3], name="x")
+        idx2 = Index([1, 2, 3, 4], name="x")
+        expected = Index([1, 2, 3, 4], name="x")
+        result, _ = idx.reindex(idx2, level="x")
+        tm.assert_index_equal(result, expected)
+
+    def test_groupby(self):
+        index = Index(range(5))
+        result = index.groupby(np.array([1, 1, 2, 2, 2]))
+        expected = {1: Index([0, 1]), 2: Index([2, 3, 4])}
+
+        tm.assert_dict_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "mi,expected",
+        [
+            (MultiIndex.from_tuples([(1, 2), (4, 5)]), np.array([True, True])),
+            (MultiIndex.from_tuples([(1, 2), (4, 6)]), np.array([True, False])),
+        ],
+    )
+    def test_equals_op_multiindex(self, mi, expected):
+        # GH9785
+        # test comparisons of multiindex
+        df = DataFrame(
+            [3, 6],
+            columns=["c"],
+            index=MultiIndex.from_arrays([[1, 4], [2, 5]], names=["a", "b"]),
+        )
+
+        result = df.index == mi
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_equals_op_multiindex_identify(self):
+        df = DataFrame(
+            [3, 6],
+            columns=["c"],
+            index=MultiIndex.from_arrays([[1, 4], [2, 5]], names=["a", "b"]),
+        )
+
+        result = df.index == df.index
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            MultiIndex.from_tuples([(1, 2), (4, 5), (8, 9)]),
+            Index(["foo", "bar", "baz"]),
+        ],
+    )
+    def test_equals_op_mismatched_multiindex_raises(self, index):
+        df = DataFrame(
+            [3, 6],
+            columns=["c"],
+            index=MultiIndex.from_arrays([[1, 4], [2, 5]], names=["a", "b"]),
+        )
+
+        with pytest.raises(ValueError, match="Lengths must match"):
+            df.index == index
+
+    def test_equals_op_index_vs_mi_same_length(self, using_infer_string):
+        mi = MultiIndex.from_tuples([(1, 2), (4, 5), (8, 9)])
+        index = Index(["foo", "bar", "baz"])
+
+        result = mi == index
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dt_conv, arg",
+        [
+            (pd.to_datetime, ["2000-01-01", "2000-01-02"]),
+            (pd.to_timedelta, ["01:02:03", "01:02:04"]),
+        ],
+    )
+    def test_dt_conversion_preserves_name(self, dt_conv, arg):
+        # GH 10875
+        index = Index(arg, name="label")
+        assert index.name == dt_conv(index).name
+
+    def test_cached_properties_not_settable(self):
+        index = Index([1, 2, 3])
+        with pytest.raises(AttributeError, match="Can't set attribute"):
+            index.is_unique = False
+
+    def test_tab_complete_warning(self, ip):
+        # https://github.com/pandas-dev/pandas/issues/16409
+        pytest.importorskip("IPython", minversion="6.0.0")
+        from IPython.core.completer import provisionalcompleter
+
+        code = "import pandas as pd; idx = pd.Index([1, 2])"
+        ip.run_cell(code)
+
+        # GH 31324 newer jedi version raises Deprecation warning;
+        #  appears resolved 2021-02-02
+        with tm.assert_produces_warning(None, raise_on_extra_warnings=False):
+            with provisionalcompleter("ignore"):
+                list(ip.Completer.completions("idx.", 4))
+
+    def test_contains_method_removed(self, index):
+        # GH#30103 method removed for all types except IntervalIndex
+        if isinstance(index, IntervalIndex):
+            index.contains(1)
+        else:
+            msg = f"'{type(index).__name__}' object has no attribute 'contains'"
+            with pytest.raises(AttributeError, match=msg):
+                index.contains(1)
+
+    def test_sortlevel(self):
+        index = Index([5, 4, 3, 2, 1])
+        with pytest.raises(Exception, match="ascending must be a single bool value or"):
+            index.sortlevel(ascending="True")
+
+        with pytest.raises(
+            Exception, match="ascending must be a list of bool values of length 1"
+        ):
+            index.sortlevel(ascending=[True, True])
+
+        with pytest.raises(Exception, match="ascending must be a bool value"):
+            index.sortlevel(ascending=["True"])
+
+        expected = Index([1, 2, 3, 4, 5])
+        result = index.sortlevel(ascending=[True])
+        tm.assert_index_equal(result[0], expected)
+
+        expected = Index([1, 2, 3, 4, 5])
+        result = index.sortlevel(ascending=True)
+        tm.assert_index_equal(result[0], expected)
+
+        expected = Index([5, 4, 3, 2, 1])
+        result = index.sortlevel(ascending=False)
+        tm.assert_index_equal(result[0], expected)
+
+    def test_sortlevel_na_position(self):
+        # GH#51612
+        idx = Index([1, np.nan])
+        result = idx.sortlevel(na_position="first")[0]
+        expected = Index([np.nan, 1])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "periods, expected_results",
+        [
+            (1, [np.nan, 10, 10, 10, 10]),
+            (2, [np.nan, np.nan, 20, 20, 20]),
+            (3, [np.nan, np.nan, np.nan, 30, 30]),
+        ],
+    )
+    def test_index_diff(self, periods, expected_results):
+        # GH#19708
+        idx = Index([10, 20, 30, 40, 50])
+        result = idx.diff(periods)
+        expected = Index(expected_results)
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "decimals, expected_results",
+        [
+            (0, [1.0, 2.0, 3.0]),
+            (1, [1.2, 2.3, 3.5]),
+            (2, [1.23, 2.35, 3.46]),
+        ],
+    )
+    def test_index_round(self, decimals, expected_results):
+        # GH#19708
+        idx = Index([1.234, 2.345, 3.456])
+        result = idx.round(decimals)
+        expected = Index(expected_results)
+
+        tm.assert_index_equal(result, expected)
+
+
+class TestMixedIntIndex:
+    # Mostly the tests from common.py for which the results differ
+    # in py2 and py3 because ints and strings are uncomparable in py3
+    # (GH 13514)
+    @pytest.fixture
+    def simple_index(self) -> Index:
+        return Index([0, "a", 1, "b", 2, "c"])
+
+    def test_argsort(self, simple_index):
+        index = simple_index
+        with pytest.raises(TypeError, match="'>|<' not supported"):
+            index.argsort()
+
+    def test_numpy_argsort(self, simple_index):
+        index = simple_index
+        with pytest.raises(TypeError, match="'>|<' not supported"):
+            np.argsort(index)
+
+    def test_copy_name(self, simple_index):
+        # Check that "name" argument passed at initialization is honoured
+        # GH12309
+        index = simple_index
+
+        first = type(index)(index, copy=True, name="mario")
+        second = type(first)(first, copy=False)
+
+        # Even though "copy=False", we want a new object.
+        assert first is not second
+        tm.assert_index_equal(first, second)
+
+        assert first.name == "mario"
+        assert second.name == "mario"
+
+        s1 = Series(2, index=first)
+        s2 = Series(3, index=second[:-1])
+
+        s3 = s1 * s2
+
+        assert s3.index.name == "mario"
+
+    def test_copy_name2(self):
+        # Check that adding a "name" parameter to the copy is honored
+        # GH14302
+        index = Index([1, 2], name="MyName")
+        index1 = index.copy()
+
+        tm.assert_index_equal(index, index1)
+
+        index2 = index.copy(name="NewName")
+        tm.assert_index_equal(index, index2, check_names=False)
+        assert index.name == "MyName"
+        assert index2.name == "NewName"
+
+    def test_unique_na(self):
+        idx = Index([2, np.nan, 2, 1], name="my_index")
+        expected = Index([2, np.nan, 1], name="my_index")
+        result = idx.unique()
+        tm.assert_index_equal(result, expected)
+
+    def test_logical_compat(self, simple_index):
+        index = simple_index
+        assert index.all() == index.values.all()
+        assert index.any() == index.values.any()
+
+    @pytest.mark.parametrize("how", ["any", "all"])
+    @pytest.mark.parametrize("dtype", [None, object, "category"])
+    @pytest.mark.parametrize(
+        "vals,expected",
+        [
+            ([1, 2, 3], [1, 2, 3]),
+            ([1.0, 2.0, 3.0], [1.0, 2.0, 3.0]),
+            ([1.0, 2.0, np.nan, 3.0], [1.0, 2.0, 3.0]),
+            (["A", "B", "C"], ["A", "B", "C"]),
+            (["A", np.nan, "B", "C"], ["A", "B", "C"]),
+        ],
+    )
+    def test_dropna(self, how, dtype, vals, expected):
+        # GH 6194
+        index = Index(vals, dtype=dtype)
+        result = index.dropna(how=how)
+        expected = Index(expected, dtype=dtype)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("how", ["any", "all"])
+    @pytest.mark.parametrize(
+        "index,expected",
+        [
+            (
+                DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"]),
+                DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"]),
+            ),
+            (
+                DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03", pd.NaT]),
+                DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"]),
+            ),
+            (
+                TimedeltaIndex(["1 days", "2 days", "3 days"]),
+                TimedeltaIndex(["1 days", "2 days", "3 days"]),
+            ),
+            (
+                TimedeltaIndex([pd.NaT, "1 days", "2 days", "3 days", pd.NaT]),
+                TimedeltaIndex(["1 days", "2 days", "3 days"]),
+            ),
+            (
+                PeriodIndex(["2012-02", "2012-04", "2012-05"], freq="M"),
+                PeriodIndex(["2012-02", "2012-04", "2012-05"], freq="M"),
+            ),
+            (
+                PeriodIndex(["2012-02", "2012-04", "NaT", "2012-05"], freq="M"),
+                PeriodIndex(["2012-02", "2012-04", "2012-05"], freq="M"),
+            ),
+        ],
+    )
+    def test_dropna_dt_like(self, how, index, expected):
+        result = index.dropna(how=how)
+        tm.assert_index_equal(result, expected)
+
+    def test_dropna_invalid_how_raises(self):
+        msg = "invalid how option: xxx"
+        with pytest.raises(ValueError, match=msg):
+            Index([1, 2, 3]).dropna(how="xxx")
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index([np.nan]),
+            Index([np.nan, 1]),
+            Index([1, 2, np.nan]),
+            Index(["a", "b", np.nan]),
+            pd.to_datetime(["NaT"]),
+            pd.to_datetime(["NaT", "2000-01-01"]),
+            pd.to_datetime(["2000-01-01", "NaT", "2000-01-02"]),
+            pd.to_timedelta(["1 day", "NaT"]),
+        ],
+    )
+    def test_is_monotonic_na(self, index):
+        assert index.is_monotonic_increasing is False
+        assert index.is_monotonic_decreasing is False
+        assert index._is_strictly_monotonic_increasing is False
+        assert index._is_strictly_monotonic_decreasing is False
+
+    @pytest.mark.parametrize("dtype", ["f8", "m8[ns]", "M8[us]"])
+    @pytest.mark.parametrize("unique_first", [True, False])
+    def test_is_monotonic_unique_na(self, dtype, unique_first):
+        # GH 55755
+        index = Index([None, 1, 1], dtype=dtype)
+        if unique_first:
+            assert index.is_unique is False
+            assert index.is_monotonic_increasing is False
+            assert index.is_monotonic_decreasing is False
+        else:
+            assert index.is_monotonic_increasing is False
+            assert index.is_monotonic_decreasing is False
+            assert index.is_unique is False
+
+    def test_int_name_format(self, frame_or_series):
+        index = Index(["a", "b", "c"], name=0)
+        result = frame_or_series(list(range(3)), index=index)
+        assert "0" in repr(result)
+
+    def test_str_to_bytes_raises(self):
+        # GH 26447
+        index = Index([str(x) for x in range(10)])
+        msg = "^'str' object cannot be interpreted as an integer$"
+        with pytest.raises(TypeError, match=msg):
+            bytes(index)
+
+    @pytest.mark.filterwarnings("ignore:elementwise comparison failed:FutureWarning")
+    def test_index_with_tuple_bool(self):
+        # GH34123
+        # TODO: also this op right now produces FutureWarning from numpy
+        #  https://github.com/numpy/numpy/issues/11521
+        idx = Index([("a", "b"), ("b", "c"), ("c", "a")])
+        result = idx == ("c", "a")
+        expected = np.array([False, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestIndexUtils:
+    @pytest.mark.parametrize(
+        "data, names, expected",
+        [
+            ([[1, 2, 4]], None, Index([1, 2, 4])),
+            ([[1, 2, 4]], ["name"], Index([1, 2, 4], name="name")),
+            ([[1, 2, 3]], None, RangeIndex(1, 4)),
+            ([[1, 2, 3]], ["name"], RangeIndex(1, 4, name="name")),
+            (
+                [["a", "a"], ["c", "d"]],
+                None,
+                MultiIndex([["a"], ["c", "d"]], [[0, 0], [0, 1]]),
+            ),
+            (
+                [["a", "a"], ["c", "d"]],
+                ["L1", "L2"],
+                MultiIndex([["a"], ["c", "d"]], [[0, 0], [0, 1]], names=["L1", "L2"]),
+            ),
+        ],
+    )
+    def test_ensure_index_from_sequences(self, data, names, expected):
+        result = ensure_index_from_sequences(data, names)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_ensure_index_mixed_closed_intervals(self):
+        # GH27172
+        intervals = [
+            pd.Interval(0, 1, closed="left"),
+            pd.Interval(1, 2, closed="right"),
+            pd.Interval(2, 3, closed="neither"),
+            pd.Interval(3, 4, closed="both"),
+        ]
+        result = ensure_index(intervals)
+        expected = Index(intervals, dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_ensure_index_uint64(self):
+        # with both 0 and a large-uint64, np.array will infer to float64
+        #  https://github.com/numpy/numpy/issues/19146
+        #  but a more accurate choice would be uint64
+        values = [0, np.iinfo(np.uint64).max]
+
+        result = ensure_index(values)
+        assert list(result) == values
+
+        expected = Index(values, dtype="uint64")
+        tm.assert_index_equal(result, expected)
+
+    def test_get_combined_index(self):
+        result = _get_combined_index([])
+        expected = RangeIndex(0)
+        tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "opname",
+    [
+        "eq",
+        "ne",
+        "le",
+        "lt",
+        "ge",
+        "gt",
+        "add",
+        "radd",
+        "sub",
+        "rsub",
+        "mul",
+        "rmul",
+        "truediv",
+        "rtruediv",
+        "floordiv",
+        "rfloordiv",
+        "pow",
+        "rpow",
+        "mod",
+        "divmod",
+    ],
+)
+def test_generated_op_names(opname, index):
+    opname = f"__{opname}__"
+    method = getattr(index, opname)
+    assert method.__name__ == opname
+
+
+@pytest.mark.parametrize(
+    "klass",
+    [
+        partial(CategoricalIndex, data=[1]),
+        partial(DatetimeIndex, data=["2020-01-01"]),
+        partial(PeriodIndex, data=["2020-01-01"]),
+        partial(TimedeltaIndex, data=["1 day"]),
+        partial(RangeIndex, start=range(1)),
+        partial(IntervalIndex, data=[pd.Interval(0, 1)]),
+        partial(Index, data=["a"], dtype=object),
+        partial(MultiIndex, levels=[1], codes=[0]),
+    ],
+)
+def test_index_subclass_constructor_wrong_kwargs(klass):
+    # GH #19348
+    with pytest.raises(TypeError, match="unexpected keyword argument"):
+        klass(foo="bar")
+
+
+def test_deprecated_fastpath():
+    msg = "[Uu]nexpected keyword argument"
+    with pytest.raises(TypeError, match=msg):
+        Index(np.array(["a", "b"], dtype=object), name="test", fastpath=True)
+
+    with pytest.raises(TypeError, match=msg):
+        Index(np.array([1, 2, 3], dtype="int64"), name="test", fastpath=True)
+
+    with pytest.raises(TypeError, match=msg):
+        RangeIndex(0, 5, 2, name="test", fastpath=True)
+
+    with pytest.raises(TypeError, match=msg):
+        CategoricalIndex(["a", "b", "c"], name="test", fastpath=True)
+
+
+def test_shape_of_invalid_index():
+    # Pre-2.0, it was possible to create "invalid" index objects backed by
+    # a multi-dimensional array (see https://github.com/pandas-dev/pandas/issues/27125
+    # about this). However, as long as this is not solved in general,this test ensures
+    # that the returned shape is consistent with this underlying array for
+    # compat with matplotlib (see https://github.com/pandas-dev/pandas/issues/27775)
+    idx = Index([0, 1, 2, 3])
+    with pytest.raises(ValueError, match="Multi-dimensional indexing"):
+        # GH#30588 multi-dimensional indexing deprecated
+        idx[:, None]
+
+
+@pytest.mark.parametrize("dtype", [None, np.int64, np.uint64, np.float64])
+def test_validate_1d_input(dtype):
+    # GH#27125 check that we do not have >1-dimensional input
+    msg = "Index data must be 1-dimensional"
+
+    arr = np.arange(8).reshape(2, 2, 2)
+    with pytest.raises(ValueError, match=msg):
+        Index(arr, dtype=dtype)
+
+    df = DataFrame(arr.reshape(4, 2))
+    with pytest.raises(ValueError, match=msg):
+        Index(df, dtype=dtype)
+
+    # GH#13601 trying to assign a multi-dimensional array to an index is not allowed
+    ser = Series(0, range(4))
+    with pytest.raises(ValueError, match=msg):
+        ser.index = np.array([[2, 3]] * 4, dtype=dtype)
+
+
+@pytest.mark.parametrize(
+    "klass, extra_kwargs",
+    [
+        [Index, {}],
+        *[[lambda x: Index(x, dtype=dtyp), {}] for dtyp in tm.ALL_REAL_NUMPY_DTYPES],
+        [DatetimeIndex, {}],
+        [TimedeltaIndex, {}],
+        [PeriodIndex, {"freq": "Y"}],
+    ],
+)
+def test_construct_from_memoryview(klass, extra_kwargs):
+    # GH 13120
+    result = klass(memoryview(np.arange(2000, 2005)), **extra_kwargs)
+    expected = klass(list(range(2000, 2005)), **extra_kwargs)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+@pytest.mark.parametrize("op", [operator.lt, operator.gt])
+def test_nan_comparison_same_object(op):
+    # GH#47105
+    idx = Index([np.nan])
+    expected = np.array([False])
+
+    result = op(idx, idx)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = op(idx, idx.copy())
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_is_monotonic_pyarrow_list_type():
+    # GH 57333
+    import pyarrow as pa
+
+    idx = Index([[1], [2, 3]], dtype=pd.ArrowDtype(pa.list_(pa.int64())))
+    assert not idx.is_monotonic_increasing
+    assert not idx.is_monotonic_decreasing
diff --git a/pandas/pandas/tests/indexes/test_common.py b/pandas/pandas/tests/indexes/test_common.py
new file mode 100644
index 0000000000000..bf16554871efc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_common.py
@@ -0,0 +1,528 @@
+"""
+Collection of tests asserting things that should be true for
+any index subclass except for MultiIndex. Makes use of the `index_flat`
+fixture defined in pandas/conftest.py.
+"""
+
+from copy import (
+    copy,
+    deepcopy,
+)
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import IS64
+from pandas.compat.numpy import np_version_gte1p25
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_numeric_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+class TestCommon:
+    @pytest.mark.parametrize("name", [None, "new_name"])
+    def test_to_frame(self, name, index_flat):
+        # see GH#15230, GH#22580
+        idx = index_flat
+
+        if name:
+            idx_name = name
+        else:
+            idx_name = idx.name or 0
+
+        df = idx.to_frame(name=idx_name)
+
+        assert df.index is idx
+        assert len(df.columns) == 1
+        assert df.columns[0] == idx_name
+
+        df = idx.to_frame(index=False, name=idx_name)
+        assert df.index is not idx
+
+    def test_droplevel(self, index_flat):
+        # GH 21115
+        # MultiIndex is tested separately in test_multi.py
+        index = index_flat
+
+        assert index.droplevel([]).equals(index)
+
+        for level in [index.name, [index.name]]:
+            if isinstance(index.name, tuple) and level is index.name:
+                # GH 21121 : droplevel with tuple name
+                continue
+            msg = (
+                "Cannot remove 1 levels from an index with 1 levels: at least one "
+                "level must be left."
+            )
+            with pytest.raises(ValueError, match=msg):
+                index.droplevel(level)
+
+        for level in "wrong", ["wrong"]:
+            with pytest.raises(
+                KeyError,
+                match=r"'Requested level \(wrong\) does not match index name \(None\)'",
+            ):
+                index.droplevel(level)
+
+    def test_constructor_non_hashable_name(self, index_flat):
+        # GH 20527
+        index = index_flat
+
+        message = "Index.name must be a hashable type"
+        renamed = [["1"]]
+
+        # With .rename()
+        with pytest.raises(TypeError, match=message):
+            index.rename(name=renamed)
+
+        # With .set_names()
+        with pytest.raises(TypeError, match=message):
+            index.set_names(names=renamed)
+
+    def test_constructor_unwraps_index(self, index_flat):
+        a = index_flat
+        # Passing dtype is necessary for Index([True, False], dtype=object)
+        #  case.
+        b = type(a)(a, dtype=a.dtype)
+        tm.assert_equal(a._data, b._data)
+
+    def test_to_flat_index(self, index_flat):
+        # 22866
+        index = index_flat
+
+        result = index.to_flat_index()
+        tm.assert_index_equal(result, index)
+
+    def test_set_name_methods(self, index_flat):
+        # MultiIndex tested separately
+        index = index_flat
+        new_name = "This is the new name for this index"
+
+        original_name = index.name
+        new_ind = index.set_names([new_name])
+        assert new_ind.name == new_name
+        assert index.name == original_name
+        res = index.rename(new_name, inplace=True)
+
+        # should return None
+        assert res is None
+        assert index.name == new_name
+        assert index.names == [new_name]
+        with pytest.raises(ValueError, match="Level must be None"):
+            index.set_names("a", level=0)
+
+        # rename in place just leaves tuples and other containers alone
+        name = ("A", "B")
+        index.rename(name, inplace=True)
+        assert index.name == name
+        assert index.names == [name]
+
+    @pytest.mark.xfail
+    def test_set_names_single_label_no_level(self, index_flat):
+        with pytest.raises(TypeError, match="list-like"):
+            # should still fail even if it would be the right length
+            index_flat.set_names("a")
+
+    def test_copy_and_deepcopy(self, index_flat):
+        index = index_flat
+
+        for func in (copy, deepcopy):
+            idx_copy = func(index)
+            assert idx_copy is not index
+            assert idx_copy.equals(index)
+
+        new_copy = index.copy(deep=True, name="banana")
+        assert new_copy.name == "banana"
+
+    def test_copy_name(self, index_flat):
+        # GH#12309: Check that the "name" argument
+        # passed at initialization is honored.
+        index = index_flat
+
+        first = type(index)(index, copy=True, name="mario")
+        second = type(first)(first, copy=False)
+
+        # Even though "copy=False", we want a new object.
+        assert first is not second
+        tm.assert_index_equal(first, second)
+
+        # Not using tm.assert_index_equal() since names differ.
+        assert index.equals(first)
+
+        assert first.name == "mario"
+        assert second.name == "mario"
+
+        # TODO: belongs in series arithmetic tests?
+        s1 = pd.Series(2, index=first)
+        s2 = pd.Series(3, index=second[:-1])
+        # See GH#13365
+        s3 = s1 * s2
+        assert s3.index.name == "mario"
+
+    def test_copy_name2(self, index_flat):
+        # GH#35592
+        index = index_flat
+
+        assert index.copy(name="mario").name == "mario"
+
+        with pytest.raises(ValueError, match="Length of new names must be 1, got 2"):
+            index.copy(name=["mario", "luigi"])
+
+        msg = f"{type(index).__name__}.name must be a hashable type"
+        with pytest.raises(TypeError, match=msg):
+            index.copy(name=[["mario"]])
+
+    def test_unique_level(self, index_flat):
+        # don't test a MultiIndex here (as its tested separated)
+        index = index_flat
+
+        # GH 17896
+        expected = index.drop_duplicates()
+        for level in [0, index.name, None]:
+            result = index.unique(level=level)
+            tm.assert_index_equal(result, expected)
+
+        msg = "Too many levels: Index has only 1 level, not 4"
+        with pytest.raises(IndexError, match=msg):
+            index.unique(level=3)
+
+        msg = (
+            rf"Requested level \(wrong\) does not match index name "
+            rf"\({re.escape(index.name.__repr__())}\)"
+        )
+        with pytest.raises(KeyError, match=msg):
+            index.unique(level="wrong")
+
+    def test_unique(self, index_flat):
+        # MultiIndex tested separately
+        index = index_flat
+        if not len(index):
+            pytest.skip("Skip check for empty Index and MultiIndex")
+
+        idx = index[[0] * 5]
+        idx_unique = index[[0]]
+
+        # We test against `idx_unique`, so first we make sure it's unique
+        # and doesn't contain nans.
+        assert idx_unique.is_unique is True
+        try:
+            assert idx_unique.hasnans is False
+        except NotImplementedError:
+            pass
+
+        result = idx.unique()
+        tm.assert_index_equal(
+            result, idx_unique, exact=not isinstance(index, RangeIndex)
+        )
+
+        # nans:
+        if not index._can_hold_na:
+            pytest.skip("Skip na-check if index cannot hold na")
+
+        vals = index._values[[0] * 5]
+        vals[0] = np.nan
+
+        vals_unique = vals[:2]
+        idx_nan = index._shallow_copy(vals)
+        idx_unique_nan = index._shallow_copy(vals_unique)
+        assert idx_unique_nan.is_unique is True
+
+        assert idx_nan.dtype == index.dtype
+        assert idx_unique_nan.dtype == index.dtype
+
+        expected = idx_unique_nan
+        for pos, i in enumerate([idx_nan, idx_unique_nan]):
+            result = i.unique()
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:Period with BDay freq:FutureWarning")
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_searchsorted_monotonic(self, index_flat, request):
+        # GH17271
+        index = index_flat
+        # not implemented for tuple searches in MultiIndex
+        # or Intervals searches in IntervalIndex
+        if isinstance(index, pd.IntervalIndex):
+            mark = pytest.mark.xfail(
+                reason="IntervalIndex.searchsorted does not support Interval arg",
+                raises=NotImplementedError,
+            )
+            request.applymarker(mark)
+
+        # nothing to test if the index is empty
+        if index.empty:
+            pytest.skip("Skip check for empty Index")
+        value = index[0]
+
+        # determine the expected results (handle dupes for 'right')
+        expected_left, expected_right = 0, (index == value).argmin()
+        if expected_right == 0:
+            # all values are the same, expected_right should be length
+            expected_right = len(index)
+
+        if index.is_monotonic_increasing or index.is_monotonic_decreasing:
+            ssm_left = index._searchsorted_monotonic(value, side="left")
+            assert expected_left == ssm_left
+
+            ssm_right = index._searchsorted_monotonic(value, side="right")
+            assert expected_right == ssm_right
+
+            ss_left = index.searchsorted(value, side="left")
+            assert expected_left == ss_left
+
+            ss_right = index.searchsorted(value, side="right")
+            assert expected_right == ss_right
+        else:
+            # non-monotonic should raise.
+            msg = "index must be monotonic increasing or decreasing"
+            with pytest.raises(ValueError, match=msg):
+                index._searchsorted_monotonic(value, side="left")
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_drop_duplicates(self, index_flat, keep):
+        # MultiIndex is tested separately
+        index = index_flat
+        if isinstance(index, RangeIndex):
+            pytest.skip(
+                "RangeIndex is tested in test_drop_duplicates_no_duplicates "
+                "as it cannot hold duplicates"
+            )
+        if len(index) == 0:
+            pytest.skip(
+                "empty index is tested in test_drop_duplicates_no_duplicates "
+                "as it cannot hold duplicates"
+            )
+
+        # make unique index
+        holder = type(index)
+        unique_values = list(set(index))
+        dtype = index.dtype if is_numeric_dtype(index) else None
+        unique_idx = holder(unique_values, dtype=dtype)
+
+        # make duplicated index
+        n = len(unique_idx)
+        duplicated_selection = np.random.default_rng(2).choice(n, int(n * 1.5))
+        idx = holder(unique_idx.values[duplicated_selection])
+
+        # Series.duplicated is tested separately
+        expected_duplicated = (
+            pd.Series(duplicated_selection).duplicated(keep=keep).values
+        )
+        tm.assert_numpy_array_equal(idx.duplicated(keep=keep), expected_duplicated)
+
+        # Series.drop_duplicates is tested separately
+        expected_dropped = holder(pd.Series(idx).drop_duplicates(keep=keep))
+        tm.assert_index_equal(idx.drop_duplicates(keep=keep), expected_dropped)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_drop_duplicates_no_duplicates(self, index_flat):
+        # MultiIndex is tested separately
+        index = index_flat
+
+        # make unique index
+        if isinstance(index, RangeIndex):
+            # RangeIndex cannot have duplicates
+            unique_idx = index
+        else:
+            holder = type(index)
+            unique_values = list(set(index))
+            dtype = index.dtype if is_numeric_dtype(index) else None
+            unique_idx = holder(unique_values, dtype=dtype)
+
+        # check on unique index
+        expected_duplicated = np.array([False] * len(unique_idx), dtype="bool")
+        tm.assert_numpy_array_equal(unique_idx.duplicated(), expected_duplicated)
+        result_dropped = unique_idx.drop_duplicates()
+        tm.assert_index_equal(result_dropped, unique_idx)
+        # validate shallow copy
+        assert result_dropped is not unique_idx
+
+    def test_drop_duplicates_inplace(self, index):
+        msg = r"drop_duplicates\(\) got an unexpected keyword argument"
+        with pytest.raises(TypeError, match=msg):
+            index.drop_duplicates(inplace=True)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_has_duplicates(self, index_flat):
+        # MultiIndex tested separately in:
+        #   tests/indexes/multi/test_unique_and_duplicates.
+        index = index_flat
+        holder = type(index)
+        if not len(index) or isinstance(index, RangeIndex):
+            # MultiIndex tested separately in:
+            #   tests/indexes/multi/test_unique_and_duplicates.
+            # RangeIndex is unique by definition.
+            pytest.skip("Skip check for empty Index, MultiIndex, and RangeIndex")
+
+        idx = holder([index[0]] * 5)
+        assert idx.is_unique is False
+        assert idx.has_duplicates is True
+
+    @pytest.mark.parametrize(
+        "dtype",
+        ["int64", "uint64", "float64", "category", "datetime64[ns]", "timedelta64[ns]"],
+    )
+    def test_astype_preserves_name(self, index, dtype):
+        # https://github.com/pandas-dev/pandas/issues/32013
+        if isinstance(index, MultiIndex):
+            index.names = ["idx" + str(i) for i in range(index.nlevels)]
+        else:
+            index.name = "idx"
+
+        warn = None
+        if index.dtype.kind == "c" and dtype in ["float64", "int64", "uint64"]:
+            # imaginary components discarded
+            if np_version_gte1p25:
+                warn = np.exceptions.ComplexWarning
+            else:
+                warn = np.ComplexWarning
+
+        is_pyarrow_str = str(index.dtype) == "string[pyarrow]" and dtype == "category"
+        try:
+            # Some of these conversions cannot succeed so we use a try / except
+            with tm.assert_produces_warning(
+                warn,
+                raise_on_extra_warnings=is_pyarrow_str,
+                check_stacklevel=False,
+            ):
+                result = index.astype(dtype)
+        except (ValueError, TypeError, NotImplementedError, SystemError):
+            return
+
+        if isinstance(index, MultiIndex):
+            assert result.names == index.names
+        else:
+            assert result.name == index.name
+
+    def test_hasnans_isnans(self, index_flat):
+        # GH#11343, added tests for hasnans / isnans
+        index = index_flat
+
+        # cases in indices doesn't include NaN
+        idx = index.copy(deep=True)
+        expected = np.array([False] * len(idx), dtype=bool)
+        tm.assert_numpy_array_equal(idx._isnan, expected)
+        assert idx.hasnans is False
+
+        idx = index.copy(deep=True)
+        values = idx._values
+
+        if len(index) == 0:
+            return
+        elif is_integer_dtype(index.dtype):
+            return
+        elif index.dtype == bool:
+            # values[1] = np.nan below casts to True!
+            return
+
+        values[1] = np.nan
+
+        idx = type(index)(values)
+
+        expected = np.array([False] * len(idx), dtype=bool)
+        expected[1] = True
+        tm.assert_numpy_array_equal(idx._isnan, expected)
+        assert idx.hasnans is True
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.parametrize("na_position", [None, "middle"])
+def test_sort_values_invalid_na_position(index_with_missing, na_position):
+    with pytest.raises(ValueError, match=f"invalid na_position: {na_position}"):
+        index_with_missing.sort_values(na_position=na_position)
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.parametrize("na_position", ["first", "last"])
+def test_sort_values_with_missing(index_with_missing, na_position, request):
+    # GH 35584. Test that sort_values works with missing values,
+    # sort non-missing and place missing according to na_position
+
+    if isinstance(index_with_missing, CategoricalIndex):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="missing value sorting order not well-defined", strict=False
+            )
+        )
+
+    missing_count = np.sum(index_with_missing.isna())
+    not_na_vals = index_with_missing[index_with_missing.notna()].values
+    sorted_values = np.sort(not_na_vals)
+    if na_position == "first":
+        sorted_values = np.concatenate([[None] * missing_count, sorted_values])
+    else:
+        sorted_values = np.concatenate([sorted_values, [None] * missing_count])
+
+    # Explicitly pass dtype needed for Index backed by EA e.g. IntegerArray
+    expected = type(index_with_missing)(sorted_values, dtype=index_with_missing.dtype)
+
+    result = index_with_missing.sort_values(na_position=na_position)
+    tm.assert_index_equal(result, expected)
+
+
+def test_sort_values_natsort_key():
+    # GH#56081
+    def split_convert(s):
+        return tuple(int(x) for x in s.split("."))
+
+    idx = pd.Index(["1.9", "2.0", "1.11", "1.10"])
+    expected = pd.Index(["1.9", "1.10", "1.11", "2.0"])
+    result = idx.sort_values(key=lambda x: tuple(map(split_convert, x)))
+    tm.assert_index_equal(result, expected)
+
+
+def test_ndarray_compat_properties(index):
+    if isinstance(index, PeriodIndex) and not IS64:
+        pytest.skip("Overflow")
+    idx = index
+    assert idx.T.equals(idx)
+    assert idx.transpose().equals(idx)
+
+    values = idx.values
+
+    assert idx.shape == values.shape
+    assert idx.ndim == values.ndim
+    assert idx.size == values.size
+
+    if not isinstance(index, (RangeIndex, MultiIndex)):
+        # These two are not backed by an ndarray
+        assert idx.nbytes == values.nbytes
+
+    # test for validity
+    idx.nbytes
+    idx.values.nbytes
+
+
+def test_compare_read_only_array():
+    # GH#57130
+    arr = np.array([], dtype=object)
+    arr.flags.writeable = False
+    idx = pd.Index(arr)
+    result = idx > 69
+    assert result.dtype == bool
+
+
+def test_to_frame_column_rangeindex():
+    idx = pd.Index([1])
+    result = idx.to_frame().columns
+    expected = RangeIndex(1)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_to_frame_name_tuple_multiindex():
+    idx = pd.Index([1])
+    result = idx.to_frame(name=(1, 2))
+    expected = pd.DataFrame([1], columns=MultiIndex.from_arrays([[1], [2]]), index=idx)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/test_datetimelike.py b/pandas/pandas/tests/indexes/test_datetimelike.py
new file mode 100644
index 0000000000000..e45d11e6286e2
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_datetimelike.py
@@ -0,0 +1,170 @@
+"""generic datetimelike tests"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestDatetimeLike:
+    @pytest.fixture(
+        params=[
+            pd.period_range("20130101", periods=5, freq="D"),
+            pd.TimedeltaIndex(
+                [
+                    "0 days 01:00:00",
+                    "1 days 01:00:00",
+                    "2 days 01:00:00",
+                    "3 days 01:00:00",
+                    "4 days 01:00:00",
+                ],
+                dtype="timedelta64[ns]",
+                freq="D",
+            ),
+            pd.DatetimeIndex(
+                ["2013-01-01", "2013-01-02", "2013-01-03", "2013-01-04", "2013-01-05"],
+                dtype="datetime64[ns]",
+                freq="D",
+            ),
+        ]
+    )
+    def simple_index(self, request):
+        return request.param
+
+    def test_isin(self, simple_index):
+        index = simple_index[:4]
+        result = index.isin(index)
+        assert result.all()
+
+        result = index.isin(list(index))
+        assert result.all()
+
+        result = index.isin([index[2], 5])
+        expected = np.array([False, False, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_argsort_matches_array(self, simple_index):
+        idx = simple_index
+        idx = idx.insert(1, pd.NaT)
+
+        result = idx.argsort()
+        expected = idx._data.argsort()
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_can_hold_identifiers(self, simple_index):
+        idx = simple_index
+        key = idx[0]
+        assert idx._can_hold_identifiers_and_holds_name(key) is False
+
+    def test_shift_identity(self, simple_index):
+        idx = simple_index
+        tm.assert_index_equal(idx, idx.shift(0))
+
+    def test_shift_empty(self, simple_index):
+        # GH#14811
+        idx = simple_index[:0]
+        tm.assert_index_equal(idx, idx.shift(1))
+
+    def test_str(self, simple_index):
+        # test the string repr
+        idx = simple_index.copy()
+        idx.name = "foo"
+        assert f"length={len(idx)}" not in str(idx)
+        assert "'foo'" in str(idx)
+        assert type(idx).__name__ in str(idx)
+
+        if hasattr(idx, "tz"):
+            if idx.tz is not None:
+                assert idx.tz in str(idx)
+        if isinstance(idx, pd.PeriodIndex):
+            assert f"dtype='period[{idx.freqstr}]'" in str(idx)
+        else:
+            assert f"freq='{idx.freqstr}'" in str(idx)
+
+    def test_view(self, simple_index):
+        idx = simple_index
+
+        result = type(simple_index)(idx)
+        tm.assert_index_equal(result, idx)
+
+        msg = (
+            "Cannot change data-type for array of references.|"
+            "Cannot change data-type for object array.|"
+        )
+        with pytest.raises(TypeError, match=msg):
+            idx.view(type(simple_index))
+
+    def test_map_callable(self, simple_index):
+        index = simple_index
+        expected = index + index.freq
+        result = index.map(lambda x: x + index.freq)
+        tm.assert_index_equal(result, expected)
+
+        # map to NaT
+        result = index.map(lambda x: pd.NaT if x == index[0] else x)
+        expected = pd.Index([pd.NaT] + index[1:].tolist())
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "mapper",
+        [
+            lambda values, index: {i: e for e, i in zip(values, index)},
+            lambda values, index: pd.Series(values, index, dtype=object),
+        ],
+    )
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_map_dictlike(self, mapper, simple_index):
+        index = simple_index
+        expected = index + index.freq
+
+        # don't compare the freqs
+        if isinstance(expected, (pd.DatetimeIndex, pd.TimedeltaIndex)):
+            expected = expected._with_freq(None)
+
+        result = index.map(mapper(expected, index))
+        tm.assert_index_equal(result, expected)
+
+        expected = pd.Index([pd.NaT] + index[1:].tolist())
+        result = index.map(mapper(expected, index))
+        tm.assert_index_equal(result, expected)
+
+        # empty map; these map to np.nan because we cannot know
+        # to re-infer things
+        expected = pd.Index([np.nan] * len(index))
+        result = index.map(mapper([], []))
+        tm.assert_index_equal(result, expected)
+
+    def test_getitem_preserves_freq(self, simple_index):
+        index = simple_index
+        assert index.freq is not None
+
+        result = index[:]
+        assert result.freq == index.freq
+
+    def test_where_cast_str(self, simple_index):
+        index = simple_index
+
+        mask = np.ones(len(index), dtype=bool)
+        mask[-1] = False
+
+        result = index.where(mask, str(index[0]))
+        expected = index.where(mask, index[0])
+        tm.assert_index_equal(result, expected)
+
+        result = index.where(mask, [str(index[0])])
+        tm.assert_index_equal(result, expected)
+
+        expected = index.astype(object).where(mask, "foo")
+        result = index.where(mask, "foo")
+        tm.assert_index_equal(result, expected)
+
+        result = index.where(mask, ["foo"])
+        tm.assert_index_equal(result, expected)
+
+    def test_diff(self, unit):
+        # GH 55080
+        dti = pd.to_datetime([10, 20, 30], unit=unit).as_unit(unit)
+        result = dti.diff(1)
+        expected = pd.to_timedelta([pd.NaT, 10, 10], unit=unit).as_unit(unit)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/test_engines.py b/pandas/pandas/tests/indexes/test_engines.py
new file mode 100644
index 0000000000000..468c2240c8192
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_engines.py
@@ -0,0 +1,192 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+
+import pandas as pd
+
+
+@pytest.fixture(
+    params=[
+        (libindex.Int64Engine, np.int64),
+        (libindex.Int32Engine, np.int32),
+        (libindex.Int16Engine, np.int16),
+        (libindex.Int8Engine, np.int8),
+        (libindex.UInt64Engine, np.uint64),
+        (libindex.UInt32Engine, np.uint32),
+        (libindex.UInt16Engine, np.uint16),
+        (libindex.UInt8Engine, np.uint8),
+        (libindex.Float64Engine, np.float64),
+        (libindex.Float32Engine, np.float32),
+    ],
+    ids=lambda x: x[0].__name__,
+)
+def numeric_indexing_engine_type_and_dtype(request):
+    return request.param
+
+
+class TestDatetimeEngine:
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            pd.Timedelta(pd.Timestamp("2016-01-01").asm8.view("m8[ns]")),
+            pd.Timestamp("2016-01-01")._value,
+            pd.Timestamp("2016-01-01").to_pydatetime(),
+            pd.Timestamp("2016-01-01").to_datetime64(),
+        ],
+    )
+    def test_not_contains_requires_timestamp(self, scalar):
+        dti1 = pd.date_range("2016-01-01", periods=3)
+        dti2 = dti1.insert(1, pd.NaT)  # non-monotonic
+        dti3 = dti1.insert(3, dti1[0])  # non-unique
+        dti4 = pd.date_range("2016-01-01", freq="ns", periods=2_000_000)
+        dti5 = dti4.insert(0, dti4[0])  # over size threshold, not unique
+
+        msg = "|".join([re.escape(str(scalar)), re.escape(repr(scalar))])
+        for dti in [dti1, dti2, dti3, dti4, dti5]:
+            with pytest.raises(TypeError, match=msg):
+                scalar in dti._engine
+
+            with pytest.raises(KeyError, match=msg):
+                dti._engine.get_loc(scalar)
+
+
+class TestTimedeltaEngine:
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            pd.Timestamp(pd.Timedelta(days=42).asm8.view("datetime64[ns]")),
+            pd.Timedelta(days=42)._value,
+            pd.Timedelta(days=42).to_pytimedelta(),
+            pd.Timedelta(days=42).to_timedelta64(),
+        ],
+    )
+    def test_not_contains_requires_timedelta(self, scalar):
+        tdi1 = pd.timedelta_range("42 days", freq="9h", periods=1234)
+        tdi2 = tdi1.insert(1, pd.NaT)  # non-monotonic
+        tdi3 = tdi1.insert(3, tdi1[0])  # non-unique
+        tdi4 = pd.timedelta_range("42 days", freq="ns", periods=2_000_000)
+        tdi5 = tdi4.insert(0, tdi4[0])  # over size threshold, not unique
+
+        msg = "|".join([re.escape(str(scalar)), re.escape(repr(scalar))])
+        for tdi in [tdi1, tdi2, tdi3, tdi4, tdi5]:
+            with pytest.raises(TypeError, match=msg):
+                scalar in tdi._engine
+
+            with pytest.raises(KeyError, match=msg):
+                tdi._engine.get_loc(scalar)
+
+
+class TestNumericEngine:
+    def test_is_monotonic(self, numeric_indexing_engine_type_and_dtype):
+        engine_type, dtype = numeric_indexing_engine_type_and_dtype
+        num = 1000
+        arr = np.array([1] * num + [2] * num + [3] * num, dtype=dtype)
+
+        # monotonic increasing
+        engine = engine_type(arr)
+        assert engine.is_monotonic_increasing is True
+        assert engine.is_monotonic_decreasing is False
+
+        # monotonic decreasing
+        engine = engine_type(arr[::-1])
+        assert engine.is_monotonic_increasing is False
+        assert engine.is_monotonic_decreasing is True
+
+        # neither monotonic increasing or decreasing
+        arr = np.array([1] * num + [2] * num + [1] * num, dtype=dtype)
+        engine = engine_type(arr[::-1])
+        assert engine.is_monotonic_increasing is False
+        assert engine.is_monotonic_decreasing is False
+
+    def test_is_unique(self, numeric_indexing_engine_type_and_dtype):
+        engine_type, dtype = numeric_indexing_engine_type_and_dtype
+
+        # unique
+        arr = np.array([1, 3, 2], dtype=dtype)
+        engine = engine_type(arr)
+        assert engine.is_unique is True
+
+        # not unique
+        arr = np.array([1, 2, 1], dtype=dtype)
+        engine = engine_type(arr)
+        assert engine.is_unique is False
+
+    def test_get_loc(self, numeric_indexing_engine_type_and_dtype):
+        engine_type, dtype = numeric_indexing_engine_type_and_dtype
+
+        # unique
+        arr = np.array([1, 2, 3], dtype=dtype)
+        engine = engine_type(arr)
+        assert engine.get_loc(2) == 1
+
+        # monotonic
+        num = 1000
+        arr = np.array([1] * num + [2] * num + [3] * num, dtype=dtype)
+        engine = engine_type(arr)
+        assert engine.get_loc(2) == slice(1000, 2000)
+
+        # not monotonic
+        arr = np.array([1, 2, 3] * num, dtype=dtype)
+        engine = engine_type(arr)
+        expected = np.array([False, True, False] * num, dtype=bool)
+        result = engine.get_loc(2)
+        assert (result == expected).all()
+
+
+class TestObjectEngine:
+    engine_type = libindex.ObjectEngine
+    dtype = np.object_
+    values = list("abc")
+
+    def test_is_monotonic(self):
+        num = 1000
+        arr = np.array(["a"] * num + ["a"] * num + ["c"] * num, dtype=self.dtype)
+
+        # monotonic increasing
+        engine = self.engine_type(arr)
+        assert engine.is_monotonic_increasing is True
+        assert engine.is_monotonic_decreasing is False
+
+        # monotonic decreasing
+        engine = self.engine_type(arr[::-1])
+        assert engine.is_monotonic_increasing is False
+        assert engine.is_monotonic_decreasing is True
+
+        # neither monotonic increasing or decreasing
+        arr = np.array(["a"] * num + ["b"] * num + ["a"] * num, dtype=self.dtype)
+        engine = self.engine_type(arr[::-1])
+        assert engine.is_monotonic_increasing is False
+        assert engine.is_monotonic_decreasing is False
+
+    def test_is_unique(self):
+        # unique
+        arr = np.array(self.values, dtype=self.dtype)
+        engine = self.engine_type(arr)
+        assert engine.is_unique is True
+
+        # not unique
+        arr = np.array(["a", "b", "a"], dtype=self.dtype)
+        engine = self.engine_type(arr)
+        assert engine.is_unique is False
+
+    def test_get_loc(self):
+        # unique
+        arr = np.array(self.values, dtype=self.dtype)
+        engine = self.engine_type(arr)
+        assert engine.get_loc("b") == 1
+
+        # monotonic
+        num = 1000
+        arr = np.array(["a"] * num + ["b"] * num + ["c"] * num, dtype=self.dtype)
+        engine = self.engine_type(arr)
+        assert engine.get_loc("b") == slice(1000, 2000)
+
+        # not monotonic
+        arr = np.array(self.values * num, dtype=self.dtype)
+        engine = self.engine_type(arr)
+        expected = np.array([False, True, False] * num, dtype=bool)
+        result = engine.get_loc("b")
+        assert (result == expected).all()
diff --git a/pandas/pandas/tests/indexes/test_frozen.py b/pandas/pandas/tests/indexes/test_frozen.py
new file mode 100644
index 0000000000000..ace66b5b06a51
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_frozen.py
@@ -0,0 +1,113 @@
+import re
+
+import pytest
+
+from pandas.core.indexes.frozen import FrozenList
+
+
+@pytest.fixture
+def lst():
+    return [1, 2, 3, 4, 5]
+
+
+@pytest.fixture
+def container(lst):
+    return FrozenList(lst)
+
+
+@pytest.fixture
+def unicode_container():
+    return FrozenList(["\u05d0", "\u05d1", "c"])
+
+
+class TestFrozenList:
+    def check_mutable_error(self, *args, **kwargs):
+        # Pass whatever function you normally would to pytest.raises
+        # (after the Exception kind).
+        mutable_regex = re.compile("does not support mutable operations")
+        msg = "'(_s)?re.(SRE_)?Pattern' object is not callable"
+        with pytest.raises(TypeError, match=msg):
+            mutable_regex(*args, **kwargs)
+
+    def test_no_mutable_funcs(self, container):
+        def setitem():
+            container[0] = 5
+
+        self.check_mutable_error(setitem)
+
+        def setslice():
+            container[1:2] = 3
+
+        self.check_mutable_error(setslice)
+
+        def delitem():
+            del container[0]
+
+        self.check_mutable_error(delitem)
+
+        def delslice():
+            del container[0:3]
+
+        self.check_mutable_error(delslice)
+
+        mutable_methods = ("extend", "pop", "remove", "insert")
+
+        for meth in mutable_methods:
+            self.check_mutable_error(getattr(container, meth))
+
+    def test_slicing_maintains_type(self, container, lst):
+        result = container[1:2]
+        expected = lst[1:2]
+        self.check_result(result, expected)
+
+    def check_result(self, result, expected):
+        assert isinstance(result, FrozenList)
+        assert result == expected
+
+    def test_string_methods_dont_fail(self, container):
+        repr(container)
+        str(container)
+        bytes(container)
+
+    def test_tricky_container(self, unicode_container):
+        repr(unicode_container)
+        str(unicode_container)
+
+    def test_add(self, container, lst):
+        result = container + (1, 2, 3)
+        expected = FrozenList(lst + [1, 2, 3])
+        self.check_result(result, expected)
+
+        result = (1, 2, 3) + container
+        expected = FrozenList([1, 2, 3] + lst)
+        self.check_result(result, expected)
+
+    def test_iadd(self, container, lst):
+        q = r = container
+
+        q += [5]
+        self.check_result(q, lst + [5])
+
+        # Other shouldn't be mutated.
+        self.check_result(r, lst)
+
+    def test_union(self, container, lst):
+        result = container.union((1, 2, 3))
+        expected = FrozenList(lst + [1, 2, 3])
+        self.check_result(result, expected)
+
+    def test_difference(self, container):
+        result = container.difference([2])
+        expected = FrozenList([1, 3, 4, 5])
+        self.check_result(result, expected)
+
+    def test_difference_dupe(self):
+        result = FrozenList([1, 2, 3, 2]).difference([2])
+        expected = FrozenList([1, 3])
+        self.check_result(result, expected)
+
+    def test_tricky_container_to_bytes_raises(self, unicode_container):
+        # GH 26447
+        msg = "^'str' object cannot be interpreted as an integer$"
+        with pytest.raises(TypeError, match=msg):
+            bytes(unicode_container)
diff --git a/pandas/pandas/tests/indexes/test_index_new.py b/pandas/pandas/tests/indexes/test_index_new.py
new file mode 100644
index 0000000000000..dd228e6b713b5
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_index_new.py
@@ -0,0 +1,425 @@
+"""
+Tests for the Index constructor conducting inference.
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.timezones import maybe_get_tz
+
+from pandas import (
+    NA,
+    Categorical,
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    TimedeltaIndex,
+    Timestamp,
+    array,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestIndexConstructorInference:
+    def test_object_all_bools(self):
+        # GH#49594 match Series behavior on ndarray[object] of all bools
+        arr = np.array([True, False], dtype=object)
+        res = Index(arr)
+        assert res.dtype == object
+
+        # since the point is matching Series behavior, let's double check
+        assert Series(arr).dtype == object
+
+    def test_object_all_complex(self):
+        # GH#49594 match Series behavior on ndarray[object] of all complex
+        arr = np.array([complex(1), complex(2)], dtype=object)
+        res = Index(arr)
+        assert res.dtype == object
+
+        # since the point is matching Series behavior, let's double check
+        assert Series(arr).dtype == object
+
+    @pytest.mark.parametrize("val", [NaT, None, np.nan, float("nan")])
+    def test_infer_nat(self, val):
+        # GH#49340 all NaT/None/nan and at least 1 NaT -> datetime64[ns],
+        #  matching Series behavior
+        values = [NaT, val]
+
+        idx = Index(values)
+        assert idx.dtype == "datetime64[s]" and idx.isna().all()
+
+        idx = Index(values[::-1])
+        assert idx.dtype == "datetime64[s]" and idx.isna().all()
+
+        idx = Index(np.array(values, dtype=object))
+        assert idx.dtype == "datetime64[s]" and idx.isna().all()
+
+        idx = Index(np.array(values, dtype=object)[::-1])
+        assert idx.dtype == "datetime64[s]" and idx.isna().all()
+
+    @pytest.mark.parametrize("na_value", [None, np.nan])
+    @pytest.mark.parametrize("vtype", [list, tuple, iter])
+    def test_construction_list_tuples_nan(self, na_value, vtype):
+        # GH#18505 : valid tuples containing NaN
+        values = [(1, "two"), (3.0, na_value)]
+        result = Index(vtype(values))
+        expected = MultiIndex.from_tuples(values)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_int_dtype_float(self, any_int_numpy_dtype):
+        # GH#18400
+        expected = Index([0, 1, 2, 3], dtype=any_int_numpy_dtype)
+        result = Index([0.0, 1.0, 2.0, 3.0], dtype=any_int_numpy_dtype)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("cast_index", [True, False])
+    @pytest.mark.parametrize(
+        "vals", [[True, False, True], np.array([True, False, True], dtype=bool)]
+    )
+    def test_constructor_dtypes_to_object(self, cast_index, vals):
+        if cast_index:
+            index = Index(vals, dtype=bool)
+        else:
+            index = Index(vals)
+
+        assert type(index) is Index
+        assert index.dtype == bool
+
+    def test_constructor_categorical_to_object(self):
+        # GH#32167 Categorical data and dtype=object should return object-dtype
+        ci = CategoricalIndex(range(5))
+        result = Index(ci, dtype=object)
+        assert not isinstance(result, CategoricalIndex)
+
+    def test_constructor_infer_periodindex(self):
+        xp = period_range("2012-1-1", freq="M", periods=3)
+        rs = Index(xp)
+        tm.assert_index_equal(rs, xp)
+        assert isinstance(rs, PeriodIndex)
+
+    def test_from_list_of_periods(self):
+        rng = period_range("1/1/2000", periods=20, freq="D")
+        periods = list(rng)
+
+        result = Index(periods)
+        assert isinstance(result, PeriodIndex)
+
+    @pytest.mark.parametrize("pos", [0, 1])
+    @pytest.mark.parametrize(
+        "klass,dtype,ctor",
+        [
+            (DatetimeIndex, "datetime64[ns]", np.datetime64("nat")),
+            (TimedeltaIndex, "timedelta64[ns]", np.timedelta64("nat")),
+        ],
+    )
+    def test_constructor_infer_nat_dt_like(
+        self, pos, klass, dtype, ctor, nulls_fixture, request
+    ):
+        if isinstance(nulls_fixture, Decimal):
+            # We dont cast these to datetime64/timedelta64
+            pytest.skip(
+                f"We don't cast {type(nulls_fixture).__name__} to "
+                "datetime64/timedelta64"
+            )
+
+        expected = klass([NaT, NaT])
+        if dtype[0] == "d":
+            # we infer all-NaT as second resolution
+            expected = expected.astype("M8[ns]")
+        assert expected.dtype == dtype
+        data = [ctor]
+        data.insert(pos, nulls_fixture)
+
+        if nulls_fixture is NA:
+            expected = Index([NA, NaT])
+            mark = pytest.mark.xfail(reason="Broken with np.NaT ctor; see GH 31884")
+            request.applymarker(mark)
+
+        result = Index(data)
+
+        tm.assert_index_equal(result, expected)
+
+        result = Index(np.array(data, dtype=object))
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("swap_objs", [True, False])
+    def test_constructor_mixed_nat_objs_infers_object(self, swap_objs):
+        # mixed np.datetime64/timedelta64 nat results in object
+        data = [np.datetime64("nat"), np.timedelta64("nat")]
+        if swap_objs:
+            data = data[::-1]
+
+        expected = Index(data, dtype=object)
+        tm.assert_index_equal(Index(data), expected)
+        tm.assert_index_equal(Index(np.array(data, dtype=object)), expected)
+
+    @pytest.mark.parametrize("swap_objs", [True, False])
+    def test_constructor_datetime_and_datetime64(self, swap_objs):
+        data = [Timestamp(2021, 6, 8, 9, 42), np.datetime64("now")]
+        if swap_objs:
+            data = data[::-1]
+        expected = DatetimeIndex(data)
+
+        tm.assert_index_equal(Index(data), expected)
+        tm.assert_index_equal(Index(np.array(data, dtype=object)), expected)
+
+    def test_constructor_datetimes_mixed_tzs(self):
+        # https://github.com/pandas-dev/pandas/pull/55793/files#r1383719998
+        tz = maybe_get_tz("US/Central")
+        dt1 = datetime(2020, 1, 1, tzinfo=tz)
+        dt2 = datetime(2020, 1, 1, tzinfo=timezone.utc)
+        result = Index([dt1, dt2])
+        expected = Index([dt1, dt2], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+
+class TestDtypeEnforced:
+    # check we don't silently ignore the dtype keyword
+
+    def test_constructor_object_dtype_with_ea_data(self, any_numeric_ea_dtype):
+        # GH#45206
+        arr = array([0], dtype=any_numeric_ea_dtype)
+
+        idx = Index(arr, dtype=object)
+        assert idx.dtype == object
+
+    @pytest.mark.parametrize("dtype", [object, "float64", "uint64", "category"])
+    def test_constructor_range_values_mismatched_dtype(self, dtype):
+        rng = Index(range(5))
+
+        result = Index(rng, dtype=dtype)
+        assert result.dtype == dtype
+
+        result = Index(range(5), dtype=dtype)
+        assert result.dtype == dtype
+
+    @pytest.mark.parametrize("dtype", [object, "float64", "uint64", "category"])
+    def test_constructor_categorical_values_mismatched_non_ea_dtype(self, dtype):
+        cat = Categorical([1, 2, 3])
+
+        result = Index(cat, dtype=dtype)
+        assert result.dtype == dtype
+
+    def test_constructor_categorical_values_mismatched_dtype(self):
+        dti = date_range("2016-01-01", periods=3)
+        cat = Categorical(dti)
+        result = Index(cat, dti.dtype)
+        tm.assert_index_equal(result, dti)
+
+        dti2 = dti.tz_localize("Asia/Tokyo")
+        cat2 = Categorical(dti2)
+        result = Index(cat2, dti2.dtype)
+        tm.assert_index_equal(result, dti2)
+
+        ii = IntervalIndex.from_breaks(range(5))
+        cat3 = Categorical(ii)
+        result = Index(cat3, dtype=ii.dtype)
+        tm.assert_index_equal(result, ii)
+
+    def test_constructor_ea_values_mismatched_categorical_dtype(self):
+        dti = date_range("2016-01-01", periods=3)
+        result = Index(dti, dtype="category")
+        expected = CategoricalIndex(dti)
+        tm.assert_index_equal(result, expected)
+
+        dti2 = date_range("2016-01-01", periods=3, tz="US/Pacific")
+        result = Index(dti2, dtype="category")
+        expected = CategoricalIndex(dti2)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_period_values_mismatched_dtype(self):
+        pi = period_range("2016-01-01", periods=3, freq="D")
+        result = Index(pi, dtype="category")
+        expected = CategoricalIndex(pi)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_timedelta64_values_mismatched_dtype(self):
+        # check we don't silently ignore the dtype keyword
+        tdi = timedelta_range("4 Days", periods=5)
+        result = Index(tdi, dtype="category")
+        expected = CategoricalIndex(tdi)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_interval_values_mismatched_dtype(self):
+        dti = date_range("2016-01-01", periods=3)
+        ii = IntervalIndex.from_breaks(dti)
+        result = Index(ii, dtype="category")
+        expected = CategoricalIndex(ii)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_datetime64_values_mismatched_period_dtype(self):
+        dti = date_range("2016-01-01", periods=3)
+        result = Index(dti, dtype="Period[D]")
+        expected = dti.to_period("D")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["int64", "uint64"])
+    def test_constructor_int_dtype_nan_raises(self, dtype):
+        # see GH#15187
+        data = [np.nan]
+        msg = "cannot convert"
+        with pytest.raises(ValueError, match=msg):
+            Index(data, dtype=dtype)
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            [1, 2, 3],
+            np.array([1, 2, 3]),
+            np.array([1, 2, 3], dtype=int),
+            # below should coerce
+            [1.0, 2.0, 3.0],
+            np.array([1.0, 2.0, 3.0], dtype=float),
+        ],
+    )
+    def test_constructor_dtypes_to_int(self, vals, any_int_numpy_dtype):
+        dtype = any_int_numpy_dtype
+        index = Index(vals, dtype=dtype)
+        assert index.dtype == dtype
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            [1, 2, 3],
+            [1.0, 2.0, 3.0],
+            np.array([1.0, 2.0, 3.0]),
+            np.array([1, 2, 3], dtype=int),
+            np.array([1.0, 2.0, 3.0], dtype=float),
+        ],
+    )
+    def test_constructor_dtypes_to_float(self, vals, float_numpy_dtype):
+        dtype = float_numpy_dtype
+        index = Index(vals, dtype=dtype)
+        assert index.dtype == dtype
+
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            [1, 2, 3],
+            np.array([1, 2, 3], dtype=int),
+            np.array(["2011-01-01", "2011-01-02"], dtype="datetime64[ns]"),
+            [datetime(2011, 1, 1), datetime(2011, 1, 2)],
+        ],
+    )
+    def test_constructor_dtypes_to_categorical(self, vals):
+        index = Index(vals, dtype="category")
+        assert isinstance(index, CategoricalIndex)
+
+    @pytest.mark.parametrize("cast_index", [True, False])
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            np.array([np.datetime64("2011-01-01"), np.datetime64("2011-01-02")]),
+            [datetime(2011, 1, 1), datetime(2011, 1, 2)],
+        ],
+    )
+    def test_constructor_dtypes_to_datetime(self, cast_index, vals):
+        vals = Index(vals)
+        if cast_index:
+            index = Index(vals, dtype=object)
+            assert isinstance(index, Index)
+            assert index.dtype == object
+        else:
+            index = Index(vals)
+            assert isinstance(index, DatetimeIndex)
+
+    @pytest.mark.parametrize("cast_index", [True, False])
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            np.array([np.timedelta64(1, "D"), np.timedelta64(1, "D")]),
+            [timedelta(1), timedelta(1)],
+        ],
+    )
+    def test_constructor_dtypes_to_timedelta(self, cast_index, vals):
+        if cast_index:
+            index = Index(vals, dtype=object)
+            assert isinstance(index, Index)
+            assert index.dtype == object
+        else:
+            index = Index(vals)
+            assert isinstance(index, TimedeltaIndex)
+
+    def test_pass_timedeltaindex_to_index(self):
+        rng = timedelta_range("1 days", "10 days")
+        idx = Index(rng, dtype=object)
+
+        expected = Index(rng.to_pytimedelta(), dtype=object)
+
+        tm.assert_numpy_array_equal(idx.values, expected.values)
+
+    def test_pass_datetimeindex_to_index(self):
+        # GH#1396
+        rng = date_range("1/1/2000", "3/1/2000")
+        idx = Index(rng, dtype=object)
+
+        expected = Index(rng.to_pydatetime(), dtype=object)
+
+        tm.assert_numpy_array_equal(idx.values, expected.values)
+
+
+class TestIndexConstructorUnwrapping:
+    # Test passing different arraylike values to pd.Index
+
+    @pytest.mark.parametrize("klass", [Index, DatetimeIndex])
+    def test_constructor_from_series_dt64(self, klass):
+        stamps = [Timestamp("20110101"), Timestamp("20120101"), Timestamp("20130101")]
+        expected = DatetimeIndex(stamps)
+        ser = Series(stamps)
+        result = klass(ser)
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_no_pandas_array(self):
+        ser = Series([1, 2, 3])
+        result = Index(ser.array)
+        expected = Index([1, 2, 3])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "array",
+        [
+            np.arange(5),
+            np.array(["a", "b", "c"]),
+            date_range("2000-01-01", periods=3).values,
+        ],
+    )
+    def test_constructor_ndarray_like(self, array):
+        # GH#5460#issuecomment-44474502
+        # it should be possible to convert any object that satisfies the numpy
+        # ndarray interface directly into an Index
+        class ArrayLike:
+            def __init__(self, array) -> None:
+                self.array = array
+
+            def __array__(self, dtype=None, copy=None) -> np.ndarray:
+                return self.array
+
+        expected = Index(array)
+        result = Index(ArrayLike(array))
+        tm.assert_index_equal(result, expected)
+
+
+class TestIndexConstructionErrors:
+    def test_constructor_overflow_int64(self):
+        # see GH#15832
+        msg = (
+            "The elements provided in the data cannot all be casted to the dtype int64"
+        )
+        with pytest.raises(OverflowError, match=msg):
+            Index([np.iinfo(np.uint64).max - 1], dtype="int64")
diff --git a/pandas/pandas/tests/indexes/test_indexing.py b/pandas/pandas/tests/indexes/test_indexing.py
new file mode 100644
index 0000000000000..1bbffcee3b671
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_indexing.py
@@ -0,0 +1,355 @@
+"""
+test_indexing tests the following Index methods:
+    __getitem__
+    get_loc
+    get_value
+    __contains__
+    take
+    where
+    get_indexer
+    get_indexer_for
+    slice_locs
+    asof_locs
+
+The corresponding tests.indexes.[index_type].test_indexing files
+contain tests for the corresponding methods specific to those Index subclasses.
+"""
+
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_scalar,
+)
+
+from pandas import (
+    NA,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    PeriodIndex,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+
+
+class TestTake:
+    def test_take_invalid_kwargs(self, index):
+        indices = [1, 2]
+
+        msg = r"take\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            index.take(indices, foo=2)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            index.take(indices, out=indices)
+
+        msg = "the 'mode' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            index.take(indices, mode="clip")
+
+    def test_take(self, index):
+        indexer = [4, 3, 0, 2]
+        if len(index) < 5:
+            pytest.skip("Test doesn't make sense since not enough elements")
+
+        result = index.take(indexer)
+        expected = index[indexer]
+        assert result.equals(expected)
+
+        if not isinstance(index, (DatetimeIndex, PeriodIndex, TimedeltaIndex)):
+            # GH 10791
+            msg = r"'(.*Index)' object has no attribute 'freq'"
+            with pytest.raises(AttributeError, match=msg):
+                index.freq
+
+    def test_take_indexer_type(self):
+        # GH#42875
+        integer_index = Index([0, 1, 2, 3])
+        scalar_index = 1
+        msg = "Expected indices to be array-like"
+        with pytest.raises(TypeError, match=msg):
+            integer_index.take(scalar_index)
+
+    def test_take_minus1_without_fill(self, index):
+        # -1 does not get treated as NA unless allow_fill=True is passed
+        if len(index) == 0:
+            # Test is not applicable
+            pytest.skip("Test doesn't make sense for empty index")
+
+        result = index.take([0, 0, -1])
+
+        expected = index.take([0, 0, len(index) - 1])
+        tm.assert_index_equal(result, expected)
+
+
+class TestContains:
+    @pytest.mark.parametrize(
+        "index,val",
+        [
+            ([0, 1, 2], 2),
+            ([0, 1, "2"], "2"),
+            ([0, 1, 2, np.inf, 4], 4),
+            ([0, 1, 2, np.nan, 4], 4),
+            ([0, 1, 2, np.inf], np.inf),
+            ([0, 1, 2, np.nan], np.nan),
+        ],
+    )
+    def test_index_contains(self, index, val):
+        index = Index(index)
+        assert val in index
+
+    @pytest.mark.parametrize(
+        "index,val",
+        [
+            (Index([0, 1, 2]), "2"),
+            (Index([0, 1, "2"]), 2),
+            (Index([0, 1, 2, np.inf]), 4),
+            (Index([0, 1, 2, np.nan]), 4),
+            (Index([0, 1, 2, np.inf]), np.nan),
+            (Index([0, 1, 2, np.nan]), np.inf),
+            # Checking if np.inf in int64 Index should not cause an OverflowError
+            # Related to GH 16957
+            (Index([0, 1, 2], dtype=np.int64), np.inf),
+            (Index([0, 1, 2], dtype=np.int64), np.nan),
+            (Index([0, 1, 2], dtype=np.uint64), np.inf),
+            (Index([0, 1, 2], dtype=np.uint64), np.nan),
+        ],
+    )
+    def test_index_not_contains(self, index, val):
+        assert val not in index
+
+    @pytest.mark.parametrize("val", [0, "2"])
+    def test_mixed_index_contains(self, val):
+        # GH#19860
+        index = Index([0, 1, "2"])
+        assert val in index
+
+    @pytest.mark.parametrize("val", ["1", 2])
+    def test_mixed_index_not_contains(self, index, val):
+        # GH#19860
+        index = Index([0, 1, "2"])
+        assert val not in index
+
+    def test_contains_with_float_index(self, any_real_numpy_dtype):
+        # GH#22085
+        dtype = any_real_numpy_dtype
+        data = [0, 1, 2, 3] if not is_float_dtype(dtype) else [0.1, 1.1, 2.2, 3.3]
+        index = Index(data, dtype=dtype)
+
+        if not is_float_dtype(index.dtype):
+            assert 1.1 not in index
+            assert 1.0 in index
+            assert 1 in index
+        else:
+            assert 1.1 in index
+            assert 1.0 not in index
+            assert 1 not in index
+
+    def test_contains_requires_hashable_raises(self, index):
+        if isinstance(index, MultiIndex):
+            return  # TODO: do we want this to raise?
+
+        msg = "unhashable type: 'list'"
+        with pytest.raises(TypeError, match=msg):
+            [] in index
+
+        msg = "|".join(
+            [
+                r"unhashable type: 'dict'",
+                r"must be real number, not dict",
+                r"an integer is required",
+                r"\{\}",
+                r"pandas\._libs\.interval\.IntervalTree' is not iterable",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            {} in index._engine
+
+
+class TestGetLoc:
+    def test_get_loc_non_hashable(self, index):
+        with pytest.raises(InvalidIndexError, match="[0, 1]"):
+            index.get_loc([0, 1])
+
+    def test_get_loc_non_scalar_hashable(self, index):
+        # GH52877
+        from enum import Enum
+
+        class E(Enum):
+            X1 = "x1"
+
+        assert not is_scalar(E.X1)
+
+        exc = KeyError
+        msg = "<E.X1: 'x1'>"
+        if isinstance(
+            index,
+            (
+                DatetimeIndex,
+                TimedeltaIndex,
+                PeriodIndex,
+                IntervalIndex,
+            ),
+        ):
+            # TODO: make these more consistent?
+            exc = InvalidIndexError
+            msg = "E.X1"
+        with pytest.raises(exc, match=msg):
+            index.get_loc(E.X1)
+
+    def test_get_loc_generator(self, index):
+        exc = KeyError
+        if isinstance(
+            index,
+            (
+                DatetimeIndex,
+                TimedeltaIndex,
+                PeriodIndex,
+                IntervalIndex,
+                MultiIndex,
+            ),
+        ):
+            # TODO: make these more consistent?
+            exc = InvalidIndexError
+        with pytest.raises(exc, match="generator object"):
+            # MultiIndex specifically checks for generator; others for scalar
+            index.get_loc(x for x in range(5))
+
+    def test_get_loc_masked_duplicated_na(self):
+        # GH#48411
+        idx = Index([1, 2, NA, NA], dtype="Int64")
+        result = idx.get_loc(NA)
+        expected = np.array([False, False, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestGetIndexer:
+    def test_get_indexer_base(self, index):
+        if index._index_as_unique:
+            expected = np.arange(index.size, dtype=np.intp)
+            actual = index.get_indexer(index)
+            tm.assert_numpy_array_equal(expected, actual)
+        else:
+            msg = "Reindexing only valid with uniquely valued Index objects"
+            with pytest.raises(InvalidIndexError, match=msg):
+                index.get_indexer(index)
+
+        with pytest.raises(ValueError, match="Invalid fill method"):
+            index.get_indexer(index, method="invalid")
+
+    def test_get_indexer_consistency(self, index):
+        # See GH#16819
+
+        if index._index_as_unique:
+            indexer = index.get_indexer(index[0:2])
+            assert isinstance(indexer, np.ndarray)
+            assert indexer.dtype == np.intp
+        else:
+            msg = "Reindexing only valid with uniquely valued Index objects"
+            with pytest.raises(InvalidIndexError, match=msg):
+                index.get_indexer(index[0:2])
+
+        indexer, _ = index.get_indexer_non_unique(index[0:2])
+        assert isinstance(indexer, np.ndarray)
+        assert indexer.dtype == np.intp
+
+    def test_get_indexer_masked_duplicated_na(self):
+        # GH#48411
+        idx = Index([1, 2, NA, NA], dtype="Int64")
+        result = idx.get_indexer_for(Index([1, NA], dtype="Int64"))
+        expected = np.array([0, 2, 3], dtype=result.dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestConvertSliceIndexer:
+    def test_convert_almost_null_slice(self, index):
+        # slice with None at both ends, but not step
+
+        key = slice(None, None, "foo")
+
+        if isinstance(index, IntervalIndex):
+            msg = "label-based slicing with step!=1 is not supported for IntervalIndex"
+            with pytest.raises(ValueError, match=msg):
+                index._convert_slice_indexer(key, "loc")
+        else:
+            msg = "'>=' not supported between instances of 'str' and 'int'"
+            with pytest.raises(TypeError, match=msg):
+                index._convert_slice_indexer(key, "loc")
+
+
+class TestPutmask:
+    def test_putmask_with_wrong_mask(self, index):
+        # GH#18368
+        if not len(index):
+            pytest.skip("Test doesn't make sense for empty index")
+
+        fill = index[0]
+
+        msg = "putmask: mask and data must be the same size"
+        with pytest.raises(ValueError, match=msg):
+            index.putmask(np.ones(len(index) + 1, np.bool_), fill)
+
+        with pytest.raises(ValueError, match=msg):
+            index.putmask(np.ones(len(index) - 1, np.bool_), fill)
+
+        with pytest.raises(ValueError, match=msg):
+            index.putmask("foo", fill)
+
+
+@pytest.mark.parametrize("idx", [[1, 2, 3], [0.1, 0.2, 0.3], ["a", "b", "c"]])
+def test_getitem_deprecated_float(idx):
+    # https://github.com/pandas-dev/pandas/issues/34191
+    idx = Index(idx)
+    msg = "Indexing with a float is no longer supported"
+    with pytest.raises(IndexError, match=msg):
+        idx[1.0]
+
+
+@pytest.mark.parametrize(
+    "idx,target,expected",
+    [
+        ([np.nan, "var1", np.nan], [np.nan], np.array([0, 2], dtype=np.intp)),
+        (
+            [np.nan, "var1", np.nan],
+            [np.nan, "var1"],
+            np.array([0, 2, 1], dtype=np.intp),
+        ),
+        (
+            np.array([np.nan, "var1", np.nan], dtype=object),
+            [np.nan],
+            np.array([0, 2], dtype=np.intp),
+        ),
+        (
+            DatetimeIndex(["2020-08-05", NaT, NaT]),
+            [NaT],
+            np.array([1, 2], dtype=np.intp),
+        ),
+        (["a", "b", "a", np.nan], [np.nan], np.array([3], dtype=np.intp)),
+        (
+            np.array(["b", np.nan, float("NaN"), "b"], dtype=object),
+            Index([np.nan], dtype=object),
+            np.array([1, 2], dtype=np.intp),
+        ),
+    ],
+)
+def test_get_indexer_non_unique_multiple_nans(idx, target, expected):
+    # GH 35392
+    axis = Index(idx)
+    actual = axis.get_indexer_for(target)
+    tm.assert_numpy_array_equal(actual, expected)
+
+
+def test_get_indexer_non_unique_nans_in_object_dtype_target(nulls_fixture):
+    idx = Index([1.0, 2.0])
+    target = Index([1, nulls_fixture], dtype="object")
+
+    result_idx, result_missing = idx.get_indexer_non_unique(target)
+    tm.assert_numpy_array_equal(result_idx, np.array([0, -1], dtype=np.intp))
+    tm.assert_numpy_array_equal(result_missing, np.array([1], dtype=np.intp))
diff --git a/pandas/pandas/tests/indexes/test_numpy_compat.py b/pandas/pandas/tests/indexes/test_numpy_compat.py
new file mode 100644
index 0000000000000..ace78d77350cb
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_numpy_compat.py
@@ -0,0 +1,189 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    DatetimeIndex,
+    Index,
+    PeriodIndex,
+    TimedeltaIndex,
+    isna,
+)
+import pandas._testing as tm
+from pandas.api.types import (
+    is_complex_dtype,
+    is_numeric_dtype,
+)
+from pandas.core.arrays import BooleanArray
+from pandas.core.indexes.datetimelike import DatetimeIndexOpsMixin
+
+
+def test_numpy_ufuncs_out(index):
+    result = index == index
+
+    out = np.empty(index.shape, dtype=bool)
+    np.equal(index, index, out=out)
+    tm.assert_numpy_array_equal(out, result)
+
+    if not index._is_multi:
+        # same thing on the ExtensionArray
+        out = np.empty(index.shape, dtype=bool)
+        np.equal(index.array, index.array, out=out)
+        tm.assert_numpy_array_equal(out, result)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        np.exp,
+        np.exp2,
+        np.expm1,
+        np.log,
+        np.log2,
+        np.log10,
+        np.log1p,
+        np.sqrt,
+        np.sin,
+        np.cos,
+        np.tan,
+        np.arcsin,
+        np.arccos,
+        np.arctan,
+        np.sinh,
+        np.cosh,
+        np.tanh,
+        np.arcsinh,
+        np.arccosh,
+        np.arctanh,
+        np.deg2rad,
+        np.rad2deg,
+    ],
+    ids=lambda x: x.__name__,
+)
+def test_numpy_ufuncs_basic(index, func):
+    # test ufuncs of numpy, see:
+    # https://numpy.org/doc/stable/reference/ufuncs.html
+
+    if isinstance(index, DatetimeIndexOpsMixin):
+        with tm.external_error_raised((TypeError, AttributeError)):
+            with np.errstate(all="ignore"):
+                func(index)
+    elif is_numeric_dtype(index) and not (
+        is_complex_dtype(index) and func in [np.deg2rad, np.rad2deg]
+    ):
+        # coerces to float (e.g. np.sin)
+        with np.errstate(all="ignore"):
+            result = func(index)
+            arr_result = func(index.values)
+            if arr_result.dtype == np.float16:
+                arr_result = arr_result.astype(np.float32)
+            exp = Index(arr_result, name=index.name)
+
+        tm.assert_index_equal(result, exp)
+        if isinstance(index.dtype, np.dtype) and is_numeric_dtype(index):
+            if is_complex_dtype(index):
+                assert result.dtype == index.dtype
+            elif index.dtype in ["bool", "int8", "uint8"]:
+                assert result.dtype in ["float16", "float32"]
+            elif index.dtype in ["int16", "uint16", "float32"]:
+                assert result.dtype == "float32"
+            else:
+                assert result.dtype == "float64"
+        else:
+            # e.g. np.exp with Int64 -> Float64
+            assert type(result) is Index
+    # raise AttributeError or TypeError
+    elif len(index) == 0:
+        pass
+    else:
+        with tm.external_error_raised((TypeError, AttributeError)):
+            with np.errstate(all="ignore"):
+                func(index)
+
+
+@pytest.mark.parametrize(
+    "func", [np.isfinite, np.isinf, np.isnan, np.signbit], ids=lambda x: x.__name__
+)
+def test_numpy_ufuncs_other(index, func):
+    # test ufuncs of numpy, see:
+    # https://numpy.org/doc/stable/reference/ufuncs.html
+    if isinstance(index, (DatetimeIndex, TimedeltaIndex)):
+        if func in (np.isfinite, np.isinf, np.isnan):
+            # numpy 1.18 changed isinf and isnan to not raise on dt64/td64
+            result = func(index)
+            assert isinstance(result, np.ndarray)
+
+            out = np.empty(index.shape, dtype=bool)
+            func(index, out=out)
+            tm.assert_numpy_array_equal(out, result)
+        else:
+            with tm.external_error_raised(TypeError):
+                func(index)
+
+    elif isinstance(index, PeriodIndex):
+        with tm.external_error_raised(TypeError):
+            func(index)
+
+    elif is_numeric_dtype(index) and not (
+        is_complex_dtype(index) and func is np.signbit
+    ):
+        # Results in bool array
+        result = func(index)
+        if not isinstance(index.dtype, np.dtype):
+            # e.g. Int64 we expect to get BooleanArray back
+            assert isinstance(result, BooleanArray)
+        else:
+            assert isinstance(result, np.ndarray)
+
+        out = np.empty(index.shape, dtype=bool)
+        func(index, out=out)
+
+        if not isinstance(index.dtype, np.dtype):
+            tm.assert_numpy_array_equal(out, result._data)
+        else:
+            tm.assert_numpy_array_equal(out, result)
+
+    elif len(index) == 0:
+        pass
+    else:
+        with tm.external_error_raised(TypeError):
+            func(index)
+
+
+@pytest.mark.parametrize("func", [np.maximum, np.minimum])
+def test_numpy_ufuncs_reductions(index, func, request):
+    # TODO: overlap with tests.series.test_ufunc.test_reductions
+    if len(index) == 0:
+        pytest.skip("Test doesn't make sense for empty index.")
+
+    if isinstance(index, CategoricalIndex) and index.dtype.ordered is False:
+        with pytest.raises(TypeError, match="is not ordered for"):
+            func.reduce(index)
+        return
+    else:
+        result = func.reduce(index)
+
+    if func is np.maximum:
+        expected = index.max(skipna=False)
+    else:
+        expected = index.min(skipna=False)
+        # TODO: do we have cases both with and without NAs?
+
+    assert type(result) is type(expected)
+    if isna(result):
+        assert isna(expected)
+    else:
+        assert result == expected
+
+
+@pytest.mark.parametrize("func", [np.bitwise_and, np.bitwise_or, np.bitwise_xor])
+def test_numpy_ufuncs_bitwise(func):
+    # https://github.com/pandas-dev/pandas/issues/46769
+    idx1 = Index([1, 2, 3, 4], dtype="int64")
+    idx2 = Index([3, 4, 5, 6], dtype="int64")
+
+    with tm.assert_produces_warning(None):
+        result = func(idx1, idx2)
+
+    expected = Index(func(idx1.values, idx2.values))
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/test_old_base.py b/pandas/pandas/tests/indexes/test_old_base.py
new file mode 100644
index 0000000000000..5f36b8c3f5dbf
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_old_base.py
@@ -0,0 +1,982 @@
+from __future__ import annotations
+
+from datetime import datetime
+import weakref
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_numeric_dtype,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    CategoricalIndex,
+    DatetimeIndex,
+    DatetimeTZDtype,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    StringDtype,
+    TimedeltaIndex,
+    isna,
+    period_range,
+)
+import pandas._testing as tm
+import pandas.core.algorithms as algos
+from pandas.core.arrays import BaseMaskedArray
+
+
+class TestBase:
+    @pytest.fixture(
+        params=[
+            RangeIndex(start=0, stop=20, step=2),
+            Index(np.arange(5, dtype=np.float64)),
+            Index(np.arange(5, dtype=np.float32)),
+            Index(np.arange(5, dtype=np.uint64)),
+            Index(range(0, 20, 2), dtype=np.int64),
+            Index(range(0, 20, 2), dtype=np.int32),
+            Index(range(0, 20, 2), dtype=np.int16),
+            Index(range(0, 20, 2), dtype=np.int8),
+            Index(list("abcde")),
+            Index([0, "a", 1, "b", 2, "c"]),
+            period_range("20130101", periods=5, freq="D"),
+            TimedeltaIndex(
+                [
+                    "0 days 01:00:00",
+                    "1 days 01:00:00",
+                    "2 days 01:00:00",
+                    "3 days 01:00:00",
+                    "4 days 01:00:00",
+                ],
+                dtype="timedelta64[ns]",
+                freq="D",
+            ),
+            DatetimeIndex(
+                ["2013-01-01", "2013-01-02", "2013-01-03", "2013-01-04", "2013-01-05"],
+                dtype="datetime64[ns]",
+                freq="D",
+            ),
+            IntervalIndex.from_breaks(range(11), closed="right"),
+        ]
+    )
+    def simple_index(self, request):
+        return request.param
+
+    def test_pickle_compat_construction(self, simple_index):
+        # need an object to create with
+        if isinstance(simple_index, RangeIndex):
+            pytest.skip("RangeIndex() is a valid constructor")
+        msg = "|".join(
+            [
+                r"Index\(\.\.\.\) must be called with a collection of some "
+                r"kind, None was passed",
+                r"DatetimeIndex\(\) must be called with a collection of some "
+                r"kind, None was passed",
+                r"TimedeltaIndex\(\) must be called with a collection of some "
+                r"kind, None was passed",
+                r"__new__\(\) missing 1 required positional argument: 'data'",
+                r"__new__\(\) takes at least 2 arguments \(1 given\)",
+                r"'NoneType' object is not iterable",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            type(simple_index)()
+
+    def test_shift(self, simple_index):
+        # GH8083 test the base class for shift
+        if isinstance(simple_index, (DatetimeIndex, TimedeltaIndex, PeriodIndex)):
+            pytest.skip("Tested in test_ops/test_arithmetic")
+        idx = simple_index
+        msg = (
+            f"This method is only implemented for DatetimeIndex, PeriodIndex and "
+            f"TimedeltaIndex; Got type {type(idx).__name__}"
+        )
+        with pytest.raises(NotImplementedError, match=msg):
+            idx.shift(1)
+        with pytest.raises(NotImplementedError, match=msg):
+            idx.shift(1, 2)
+
+    def test_constructor_name_unhashable(self, simple_index):
+        # GH#29069 check that name is hashable
+        # See also same-named test in tests.series.test_constructors
+        idx = simple_index
+        with pytest.raises(TypeError, match="Index.name must be a hashable type"):
+            type(idx)(idx, name=[])
+
+    def test_create_index_existing_name(self, simple_index):
+        # GH11193, when an existing index is passed, and a new name is not
+        # specified, the new index should inherit the previous object name
+        expected = simple_index.copy()
+        if not isinstance(expected, MultiIndex):
+            expected.name = "foo"
+            result = Index(expected)
+            tm.assert_index_equal(result, expected)
+
+            result = Index(expected, name="bar")
+            expected.name = "bar"
+            tm.assert_index_equal(result, expected)
+        else:
+            expected.names = ["foo", "bar"]
+            result = Index(expected)
+            tm.assert_index_equal(
+                result,
+                Index(
+                    Index(
+                        [
+                            ("foo", "one"),
+                            ("foo", "two"),
+                            ("bar", "one"),
+                            ("baz", "two"),
+                            ("qux", "one"),
+                            ("qux", "two"),
+                        ],
+                        dtype="object",
+                    ),
+                    names=["foo", "bar"],
+                ),
+            )
+
+            result = Index(expected, names=["A", "B"])
+            tm.assert_index_equal(
+                result,
+                Index(
+                    Index(
+                        [
+                            ("foo", "one"),
+                            ("foo", "two"),
+                            ("bar", "one"),
+                            ("baz", "two"),
+                            ("qux", "one"),
+                            ("qux", "two"),
+                        ],
+                        dtype="object",
+                    ),
+                    names=["A", "B"],
+                ),
+            )
+
+    def test_numeric_compat(self, simple_index):
+        idx = simple_index
+        # Check that this doesn't cover MultiIndex case, if/when it does,
+        #  we can remove multi.test_compat.test_numeric_compat
+        assert not isinstance(idx, MultiIndex)
+        if type(idx) is Index:
+            pytest.skip("Not applicable for Index")
+        if is_numeric_dtype(simple_index.dtype) or isinstance(
+            simple_index, TimedeltaIndex
+        ):
+            pytest.skip("Tested elsewhere.")
+
+        typ = type(idx._data).__name__
+        cls = type(idx).__name__
+        lmsg = "|".join(
+            [
+                rf"unsupported operand type\(s\) for \*: '{typ}' and 'int'",
+                "cannot perform (__mul__|__truediv__|__floordiv__) with "
+                f"this index type: ({cls}|{typ})",
+            ]
+        )
+        with pytest.raises(TypeError, match=lmsg):
+            idx * 1
+        rmsg = "|".join(
+            [
+                rf"unsupported operand type\(s\) for \*: 'int' and '{typ}'",
+                "cannot perform (__rmul__|__rtruediv__|__rfloordiv__) with "
+                f"this index type: ({cls}|{typ})",
+            ]
+        )
+        with pytest.raises(TypeError, match=rmsg):
+            1 * idx
+
+        div_err = lmsg.replace("*", "/")
+        with pytest.raises(TypeError, match=div_err):
+            idx / 1
+        div_err = rmsg.replace("*", "/")
+        with pytest.raises(TypeError, match=div_err):
+            1 / idx
+
+        floordiv_err = lmsg.replace("*", "//")
+        with pytest.raises(TypeError, match=floordiv_err):
+            idx // 1
+        floordiv_err = rmsg.replace("*", "//")
+        with pytest.raises(TypeError, match=floordiv_err):
+            1 // idx
+
+    def test_logical_compat(self, simple_index):
+        if simple_index.dtype in (object, "string"):
+            pytest.skip("Tested elsewhere.")
+        idx = simple_index
+        if idx.dtype.kind in "iufcbm":
+            assert idx.all() == idx._values.all()
+            assert idx.all() == idx.to_series().all()
+            assert idx.any() == idx._values.any()
+            assert idx.any() == idx.to_series().any()
+        else:
+            msg = "does not support operation '(any|all)'"
+            with pytest.raises(TypeError, match=msg):
+                idx.all()
+            with pytest.raises(TypeError, match=msg):
+                idx.any()
+
+    def test_repr_roundtrip(self, simple_index):
+        if isinstance(simple_index, IntervalIndex):
+            pytest.skip(f"Not a valid repr for {type(simple_index).__name__}")
+        idx = simple_index
+        tm.assert_index_equal(eval(repr(idx)), idx)
+
+    def test_repr_max_seq_item_setting(self, simple_index):
+        # GH10182
+        if isinstance(simple_index, IntervalIndex):
+            pytest.skip(f"Not a valid repr for {type(simple_index).__name__}")
+        idx = simple_index
+        idx = idx.repeat(50)
+        with pd.option_context("display.max_seq_items", None):
+            repr(idx)
+            assert "..." not in str(idx)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_ensure_copied_data(self, index):
+        # Check the "copy" argument of each Index.__new__ is honoured
+        # GH12309
+        init_kwargs = {}
+        if isinstance(index, PeriodIndex):
+            # Needs "freq" specification:
+            init_kwargs["freq"] = index.freq
+        elif isinstance(index, (RangeIndex, MultiIndex, CategoricalIndex)):
+            pytest.skip(
+                "RangeIndex cannot be initialized from data, "
+                "MultiIndex and CategoricalIndex are tested separately"
+            )
+        elif index.dtype == object and index.inferred_type in ["boolean", "string"]:
+            init_kwargs["dtype"] = index.dtype
+
+        index_type = type(index)
+        result = index_type(index.values, copy=True, **init_kwargs)
+        if isinstance(index.dtype, DatetimeTZDtype):
+            result = result.tz_localize("UTC").tz_convert(index.tz)
+        if isinstance(index, (DatetimeIndex, TimedeltaIndex)):
+            index = index._with_freq(None)
+
+        tm.assert_index_equal(index, result)
+
+        if isinstance(index, PeriodIndex):
+            # .values an object array of Period, thus copied
+            result = index_type.from_ordinals(ordinals=index.asi8, **init_kwargs)
+            tm.assert_numpy_array_equal(index.asi8, result.asi8, check_same="same")
+        elif isinstance(index, IntervalIndex):
+            # checked in test_interval.py
+            pass
+        elif type(index) is Index and not isinstance(index.dtype, np.dtype):
+            result = index_type(index.values, copy=False, **init_kwargs)
+            tm.assert_index_equal(result, index)
+
+            if isinstance(index._values, BaseMaskedArray):
+                assert np.shares_memory(index._values._data, result._values._data)
+                tm.assert_numpy_array_equal(
+                    index._values._data, result._values._data, check_same="same"
+                )
+                assert np.shares_memory(index._values._mask, result._values._mask)
+                tm.assert_numpy_array_equal(
+                    index._values._mask, result._values._mask, check_same="same"
+                )
+            elif (
+                isinstance(index.dtype, StringDtype) and index.dtype.storage == "python"
+            ):
+                assert np.shares_memory(index._values._ndarray, result._values._ndarray)
+                tm.assert_numpy_array_equal(
+                    index._values._ndarray, result._values._ndarray, check_same="same"
+                )
+            elif (
+                isinstance(index.dtype, StringDtype)
+                and index.dtype.storage == "pyarrow"
+            ):
+                assert tm.shares_memory(result._values, index._values)
+            else:
+                raise NotImplementedError(index.dtype)
+        else:
+            result = index_type(index.values, copy=False, **init_kwargs)
+            tm.assert_numpy_array_equal(index.values, result.values, check_same="same")
+
+    def test_memory_usage(self, index):
+        index._engine.clear_mapping()
+        result = index.memory_usage()
+        if index.empty:
+            # we report 0 for no-length
+            assert result == 0
+            return
+
+        # non-zero length
+        index.get_loc(index[0])
+        result2 = index.memory_usage()
+        result3 = index.memory_usage(deep=True)
+
+        # RangeIndex, IntervalIndex
+        # don't have engines
+        # Index[EA] has engine but it does not have a Hashtable .mapping
+        if not isinstance(index, (RangeIndex, IntervalIndex)) and not (
+            type(index) is Index and not isinstance(index.dtype, np.dtype)
+        ):
+            assert result2 > result
+
+        if index.inferred_type == "object":
+            assert result3 > result2
+
+    def test_memory_usage_doesnt_trigger_engine(self, index):
+        index._cache.clear()
+        assert "_engine" not in index._cache
+
+        res_without_engine = index.memory_usage()
+        assert "_engine" not in index._cache
+
+        # explicitly load and cache the engine
+        _ = index._engine
+        assert "_engine" in index._cache
+
+        res_with_engine = index.memory_usage()
+
+        # the empty engine doesn't affect the result even when initialized with values,
+        # because engine.sizeof() doesn't consider the content of engine.values
+        assert res_with_engine == res_without_engine
+
+        if len(index) == 0:
+            assert res_without_engine == 0
+            assert res_with_engine == 0
+        else:
+            assert res_without_engine > 0
+            assert res_with_engine > 0
+
+    def test_argsort(self, index):
+        if isinstance(index, CategoricalIndex):
+            pytest.skip(f"{type(self).__name__} separately tested")
+
+        result = index.argsort()
+        expected = np.array(index).argsort()
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    def test_numpy_argsort(self, index):
+        result = np.argsort(index)
+        expected = index.argsort()
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = np.argsort(index, kind="mergesort")
+        expected = index.argsort(kind="mergesort")
+        tm.assert_numpy_array_equal(result, expected)
+
+        # these are the only two types that perform
+        # pandas compatibility input validation - the
+        # rest already perform separate (or no) such
+        # validation via their 'values' attribute as
+        # defined in pandas.core.indexes/base.py - they
+        # cannot be changed at the moment due to
+        # backwards compatibility concerns
+        if isinstance(index, (CategoricalIndex, RangeIndex)):
+            msg = "the 'axis' parameter is not supported"
+            with pytest.raises(ValueError, match=msg):
+                np.argsort(index, axis=1)
+
+            msg = "the 'order' parameter is not supported"
+            with pytest.raises(ValueError, match=msg):
+                np.argsort(index, order=("a", "b"))
+
+    def test_repeat(self, simple_index):
+        rep = 2
+        idx = simple_index.copy()
+        new_index_cls = idx._constructor
+        expected = new_index_cls(idx.values.repeat(rep), name=idx.name)
+        tm.assert_index_equal(idx.repeat(rep), expected)
+
+        idx = simple_index
+        rep = np.arange(len(idx))
+        expected = new_index_cls(idx.values.repeat(rep), name=idx.name)
+        tm.assert_index_equal(idx.repeat(rep), expected)
+
+    def test_numpy_repeat(self, simple_index):
+        rep = 2
+        idx = simple_index
+        expected = idx.repeat(rep)
+        tm.assert_index_equal(np.repeat(idx, rep), expected)
+
+        msg = "the 'axis' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.repeat(idx, rep, axis=0)
+
+    def test_where(self, listlike_box, simple_index):
+        if isinstance(simple_index, (IntervalIndex, PeriodIndex)) or is_numeric_dtype(
+            simple_index.dtype
+        ):
+            pytest.skip("Tested elsewhere.")
+        klass = listlike_box
+
+        idx = simple_index
+        if isinstance(idx, (DatetimeIndex, TimedeltaIndex)):
+            # where does not preserve freq
+            idx = idx._with_freq(None)
+
+        cond = [True] * len(idx)
+        result = idx.where(klass(cond))
+        expected = idx
+        tm.assert_index_equal(result, expected)
+
+        cond = [False] + [True] * len(idx[1:])
+        expected = Index([idx._na_value] + idx[1:].tolist(), dtype=idx.dtype)
+        result = idx.where(klass(cond))
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_base(self, index):
+        # GH#51363
+        trimmed = index[1:4]
+
+        if not len(index):
+            pytest.skip("Not applicable for empty index")
+
+        result = trimmed.insert(0, index[0])
+        assert index[0:4].equals(result)
+
+    def test_insert_out_of_bounds(self, index, using_infer_string):
+        # TypeError/IndexError matches what np.insert raises in these cases
+
+        if len(index) > 0:
+            err = TypeError
+        else:
+            err = IndexError
+        if len(index) == 0:
+            # 0 vs 0.5 in error message varies with numpy version
+            msg = "index (0|0.5) is out of bounds for axis 0 with size 0"
+        else:
+            msg = "slice indices must be integers or None or have an __index__ method"
+
+        if using_infer_string:
+            if index.dtype == "string" or index.dtype == "category":
+                msg = "loc must be an integer between"
+            elif index.dtype == "object" and len(index) == 0:
+                msg = "loc must be an integer between"
+                err = TypeError
+
+        with pytest.raises(err, match=msg):
+            index.insert(0.5, "foo")
+
+        msg = "|".join(
+            [
+                r"index -?\d+ is out of bounds for axis 0 with size \d+",
+                "loc must be an integer between",
+            ]
+        )
+        with pytest.raises(IndexError, match=msg):
+            index.insert(len(index) + 1, 1)
+
+        with pytest.raises(IndexError, match=msg):
+            index.insert(-len(index) - 1, 1)
+
+    def test_delete_base(self, index):
+        if not len(index):
+            pytest.skip("Not applicable for empty index")
+
+        if isinstance(index, RangeIndex):
+            # tested in class
+            pytest.skip(f"{type(self).__name__} tested elsewhere")
+
+        expected = index[1:]
+        result = index.delete(0)
+        assert result.equals(expected)
+        assert result.name == expected.name
+
+        expected = index[:-1]
+        result = index.delete(-1)
+        assert result.equals(expected)
+        assert result.name == expected.name
+
+        length = len(index)
+        msg = f"index {length} is out of bounds for axis 0 with size {length}"
+        with pytest.raises(IndexError, match=msg):
+            index.delete(length)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_equals(self, index):
+        if isinstance(index, IntervalIndex):
+            pytest.skip(f"{type(index).__name__} tested elsewhere")
+
+        is_ea_idx = type(index) is Index and not isinstance(index.dtype, np.dtype)
+
+        assert index.equals(index)
+        assert index.equals(index.copy())
+        if not is_ea_idx:
+            # doesn't hold for e.g. IntegerDtype
+            assert index.equals(index.astype(object))
+
+        assert not index.equals(list(index))
+        assert not index.equals(np.array(index))
+
+        # Cannot pass in non-int64 dtype to RangeIndex
+        if not isinstance(index, RangeIndex) and not is_ea_idx:
+            same_values = Index(index, dtype=object)
+            assert index.equals(same_values)
+            assert same_values.equals(index)
+
+        if index.nlevels == 1:
+            # do not test MultiIndex
+            assert not index.equals(Series(index))
+
+    def test_equals_op(self, simple_index):
+        # GH9947, GH10637
+        index_a = simple_index
+
+        n = len(index_a)
+        index_b = index_a[0:-1]
+        index_c = index_a[0:-1].append(index_a[-2:-1])
+        index_d = index_a[0:1]
+
+        msg = "Lengths must match|could not be broadcast"
+        with pytest.raises(ValueError, match=msg):
+            index_a == index_b
+        expected1 = np.array([True] * n)
+        expected2 = np.array([True] * (n - 1) + [False])
+        tm.assert_numpy_array_equal(index_a == index_a, expected1)
+        tm.assert_numpy_array_equal(index_a == index_c, expected2)
+
+        # test comparisons with numpy arrays
+        array_a = np.array(index_a)
+        array_b = np.array(index_a[0:-1])
+        array_c = np.array(index_a[0:-1].append(index_a[-2:-1]))
+        array_d = np.array(index_a[0:1])
+        with pytest.raises(ValueError, match=msg):
+            index_a == array_b
+        tm.assert_numpy_array_equal(index_a == array_a, expected1)
+        tm.assert_numpy_array_equal(index_a == array_c, expected2)
+
+        # test comparisons with Series
+        series_a = Series(array_a)
+        series_b = Series(array_b)
+        series_c = Series(array_c)
+        series_d = Series(array_d)
+        with pytest.raises(ValueError, match=msg):
+            index_a == series_b
+
+        tm.assert_numpy_array_equal(index_a == series_a, expected1)
+        tm.assert_numpy_array_equal(index_a == series_c, expected2)
+
+        # cases where length is 1 for one of them
+        with pytest.raises(ValueError, match="Lengths must match"):
+            index_a == index_d
+        with pytest.raises(ValueError, match="Lengths must match"):
+            index_a == series_d
+        with pytest.raises(ValueError, match="Lengths must match"):
+            index_a == array_d
+        msg = "Can only compare identically-labeled Series objects"
+        with pytest.raises(ValueError, match=msg):
+            series_a == series_d
+        with pytest.raises(ValueError, match="Lengths must match"):
+            series_a == array_d
+
+        # comparing with a scalar should broadcast; note that we are excluding
+        # MultiIndex because in this case each item in the index is a tuple of
+        # length 2, and therefore is considered an array of length 2 in the
+        # comparison instead of a scalar
+        if not isinstance(index_a, MultiIndex):
+            expected3 = np.array([False] * (len(index_a) - 2) + [True, False])
+            # assuming the 2nd to last item is unique in the data
+            item = index_a[-2]
+            tm.assert_numpy_array_equal(index_a == item, expected3)
+            tm.assert_series_equal(series_a == item, Series(expected3))
+
+    def test_fillna(self, index):
+        # GH 11343
+        if len(index) == 0:
+            pytest.skip("Not relevant for empty index")
+        elif index.dtype == bool:
+            pytest.skip(f"{index.dtype} cannot hold NAs")
+        elif isinstance(index, Index) and is_integer_dtype(index.dtype):
+            pytest.skip(f"Not relevant for Index with {index.dtype}")
+        elif isinstance(index, MultiIndex):
+            idx = index.copy(deep=True)
+            msg = "fillna is not defined for MultiIndex"
+            with pytest.raises(NotImplementedError, match=msg):
+                idx.fillna(idx[0])
+        else:
+            idx = index.copy(deep=True)
+            result = idx.fillna(idx[0])
+            tm.assert_index_equal(result, idx)
+            assert result is not idx
+
+            msg = "'value' must be a scalar, passed: "
+            with pytest.raises(TypeError, match=msg):
+                idx.fillna([idx[0]])
+
+            idx = index.copy(deep=True)
+            values = idx._values
+
+            values[1] = np.nan
+
+            idx = type(index)(values)
+
+            expected = np.array([False] * len(idx), dtype=bool)
+            expected[1] = True
+            tm.assert_numpy_array_equal(idx._isnan, expected)
+            assert idx.hasnans is True
+
+    def test_nulls(self, index):
+        # this is really a smoke test for the methods
+        # as these are adequately tested for function elsewhere
+        if len(index) == 0:
+            tm.assert_numpy_array_equal(index.isna(), np.array([], dtype=bool))
+        elif isinstance(index, MultiIndex):
+            idx = index.copy()
+            msg = "isna is not defined for MultiIndex"
+            with pytest.raises(NotImplementedError, match=msg):
+                idx.isna()
+        elif not index.hasnans:
+            tm.assert_numpy_array_equal(index.isna(), np.zeros(len(index), dtype=bool))
+            tm.assert_numpy_array_equal(index.notna(), np.ones(len(index), dtype=bool))
+        else:
+            result = isna(index)
+            tm.assert_numpy_array_equal(index.isna(), result)
+            tm.assert_numpy_array_equal(index.notna(), ~result)
+
+    def test_empty(self, simple_index):
+        # GH 15270
+        idx = simple_index
+        assert not idx.empty
+        assert idx[:0].empty
+
+    def test_join_self_unique(self, join_type, simple_index):
+        idx = simple_index
+        if idx.is_unique:
+            joined = idx.join(idx, how=join_type)
+            expected = simple_index
+            if join_type == "outer":
+                expected = algos.safe_sort(expected)
+            tm.assert_index_equal(joined, expected)
+
+    def test_map(self, simple_index):
+        # callable
+        if isinstance(simple_index, (TimedeltaIndex, PeriodIndex)):
+            pytest.skip("Tested elsewhere.")
+        idx = simple_index
+
+        result = idx.map(lambda x: x)
+        # RangeIndex are equivalent to the similar Index with int64 dtype
+        tm.assert_index_equal(result, idx, exact="equiv")
+
+    @pytest.mark.parametrize(
+        "mapper",
+        [
+            lambda values, index: {i: e for e, i in zip(values, index)},
+            lambda values, index: Series(values, index),
+        ],
+    )
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_map_dictlike(self, mapper, simple_index, request):
+        idx = simple_index
+        if isinstance(idx, (DatetimeIndex, TimedeltaIndex, PeriodIndex)):
+            pytest.skip("Tested elsewhere.")
+
+        identity = mapper(idx.values, idx)
+
+        result = idx.map(identity)
+        # RangeIndex are equivalent to the similar Index with int64 dtype
+        tm.assert_index_equal(result, idx, exact="equiv")
+
+        # empty mappable
+        dtype = None
+        if idx.dtype.kind == "f":
+            dtype = idx.dtype
+
+        expected = Index([np.nan] * len(idx), dtype=dtype)
+        result = idx.map(mapper(expected, idx))
+        tm.assert_index_equal(result, expected)
+
+    def test_map_str(self, simple_index):
+        # GH 31202
+        if isinstance(simple_index, CategoricalIndex):
+            pytest.skip("See test_map.py")
+        idx = simple_index
+        result = idx.map(str)
+        expected = Index([str(x) for x in idx])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("copy", [True, False])
+    @pytest.mark.parametrize("name", [None, "foo"])
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_astype_category(self, copy, name, ordered, simple_index):
+        # GH 18630
+        idx = simple_index
+        if name:
+            idx = idx.rename(name)
+
+        # standard categories
+        dtype = CategoricalDtype(ordered=ordered)
+        result = idx.astype(dtype, copy=copy)
+        expected = CategoricalIndex(idx, name=name, ordered=ordered)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        # non-standard categories
+        dtype = CategoricalDtype(idx.unique().tolist()[:-1], ordered)
+        result = idx.astype(dtype, copy=copy)
+        expected = CategoricalIndex(idx, name=name, dtype=dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        if ordered is False:
+            # dtype='category' defaults to ordered=False, so only test once
+            result = idx.astype("category", copy=copy)
+            expected = CategoricalIndex(idx, name=name)
+            tm.assert_index_equal(result, expected, exact=True)
+
+    def test_is_unique(self, simple_index):
+        # initialize a unique index
+        index = simple_index.drop_duplicates()
+        assert index.is_unique is True
+
+        # empty index should be unique
+        index_empty = index[:0]
+        assert index_empty.is_unique is True
+
+        # test basic dupes
+        index_dup = index.insert(0, index[0])
+        assert index_dup.is_unique is False
+
+        # single NA should be unique
+        index_na = index.insert(0, np.nan)
+        assert index_na.is_unique is True
+
+        # multiple NA should not be unique
+        index_na_dup = index_na.insert(0, np.nan)
+        assert index_na_dup.is_unique is False
+
+    @pytest.mark.arm_slow
+    def test_engine_reference_cycle(self, simple_index):
+        # GH27585
+        index = simple_index.copy()
+        ref = weakref.ref(index)
+        index._engine
+        del index
+        assert ref() is None
+
+    def test_getitem_2d_deprecated(self, simple_index):
+        # GH#30588, GH#31479
+        if isinstance(simple_index, IntervalIndex):
+            pytest.skip("Tested elsewhere")
+        idx = simple_index
+        msg = "Multi-dimensional indexing|too many|only"
+        with pytest.raises((ValueError, IndexError), match=msg):
+            idx[:, None]
+
+        if not isinstance(idx, RangeIndex):
+            # GH#44051 RangeIndex already raised pre-2.0 with a different message
+            with pytest.raises((ValueError, IndexError), match=msg):
+                idx[True]
+            with pytest.raises((ValueError, IndexError), match=msg):
+                idx[False]
+        else:
+            msg = "only integers, slices"
+            with pytest.raises(IndexError, match=msg):
+                idx[True]
+            with pytest.raises(IndexError, match=msg):
+                idx[False]
+
+    def test_copy_shares_cache(self, simple_index):
+        # GH32898, GH36840
+        idx = simple_index
+        idx.get_loc(idx[0])  # populates the _cache.
+        copy = idx.copy()
+
+        assert copy._cache is idx._cache
+
+    def test_shallow_copy_shares_cache(self, simple_index):
+        # GH32669, GH36840
+        idx = simple_index
+        idx.get_loc(idx[0])  # populates the _cache.
+        shallow_copy = idx._view()
+
+        assert shallow_copy._cache is idx._cache
+
+        shallow_copy = idx._shallow_copy(idx._data)
+        assert shallow_copy._cache is not idx._cache
+        assert shallow_copy._cache == {}
+
+    def test_index_groupby(self, simple_index):
+        idx = simple_index[:5]
+        to_groupby = np.array([1, 2, np.nan, 2, 1])
+        tm.assert_dict_equal(
+            idx.groupby(to_groupby), {1.0: idx[[0, 4]], 2.0: idx[[1, 3]]}
+        )
+
+        to_groupby = DatetimeIndex(
+            [
+                datetime(2011, 11, 1),
+                datetime(2011, 12, 1),
+                pd.NaT,
+                datetime(2011, 12, 1),
+                datetime(2011, 11, 1),
+            ],
+            tz="UTC",
+        ).values
+
+        ex_keys = [Timestamp("2011-11-01"), Timestamp("2011-12-01")]
+        expected = {ex_keys[0]: idx[[0, 4]], ex_keys[1]: idx[[1, 3]]}
+        tm.assert_dict_equal(idx.groupby(to_groupby), expected)
+
+    def test_append_preserves_dtype(self, simple_index):
+        # In particular Index with dtype float32
+        index = simple_index
+        N = len(index)
+
+        result = index.append(index)
+        assert result.dtype == index.dtype
+
+        tm.assert_index_equal(result[:N], index, exact=False, check_exact=True)
+        tm.assert_index_equal(result[N:], index, exact=False, check_exact=True)
+
+        alt = index.take(list(range(N)) * 2)
+        tm.assert_index_equal(result, alt, check_exact=True)
+
+    def test_inv(self, simple_index, using_infer_string):
+        idx = simple_index
+
+        if idx.dtype.kind in ["i", "u"]:
+            res = ~idx
+            expected = Index(~idx.values, name=idx.name)
+            tm.assert_index_equal(res, expected)
+
+            # check that we are matching Series behavior
+            res2 = ~Series(idx)
+            tm.assert_series_equal(res2, Series(expected))
+        else:
+            if idx.dtype.kind == "f":
+                msg = "ufunc 'invert' not supported for the input types"
+            else:
+                msg = "bad operand|__invert__ is not supported for string dtype"
+            with pytest.raises(TypeError, match=msg):
+                ~idx
+
+            # check that we get the same behavior with Series
+            with pytest.raises(TypeError, match=msg):
+                ~Series(idx)
+
+
+class TestNumericBase:
+    @pytest.fixture(
+        params=[
+            RangeIndex(start=0, stop=20, step=2),
+            Index(np.arange(5, dtype=np.float64)),
+            Index(np.arange(5, dtype=np.float32)),
+            Index(np.arange(5, dtype=np.uint64)),
+            Index(range(0, 20, 2), dtype=np.int64),
+            Index(range(0, 20, 2), dtype=np.int32),
+            Index(range(0, 20, 2), dtype=np.int16),
+            Index(range(0, 20, 2), dtype=np.int8),
+        ]
+    )
+    def simple_index(self, request):
+        return request.param
+
+    def test_constructor_unwraps_index(self, simple_index):
+        if isinstance(simple_index, RangeIndex):
+            pytest.skip("Tested elsewhere.")
+        index_cls = type(simple_index)
+        dtype = simple_index.dtype
+
+        idx = Index([1, 2], dtype=dtype)
+        result = index_cls(idx)
+        expected = np.array([1, 2], dtype=idx.dtype)
+        tm.assert_numpy_array_equal(result._data, expected)
+
+    def test_can_hold_identifiers(self, simple_index):
+        idx = simple_index
+        key = idx[0]
+        assert idx._can_hold_identifiers_and_holds_name(key) is False
+
+    def test_view(self, simple_index):
+        if isinstance(simple_index, RangeIndex):
+            pytest.skip("Tested elsewhere.")
+        index_cls = type(simple_index)
+        dtype = simple_index.dtype
+
+        idx = index_cls([], dtype=dtype, name="Foo")
+        idx_view = idx.view()
+        assert idx_view.name == "Foo"
+
+        idx_view = idx.view(dtype)
+        tm.assert_index_equal(idx, index_cls(idx_view, name="Foo"), exact=True)
+
+        msg = (
+            "Cannot change data-type for array of references.|"
+            "Cannot change data-type for object array.|"
+        )
+        with pytest.raises(TypeError, match=msg):
+            # GH#55709
+            idx.view(index_cls)
+
+    def test_insert_non_na(self, simple_index):
+        # GH#43921 inserting an element that we know we can hold should
+        #  not change dtype or type (except for RangeIndex)
+        index = simple_index
+
+        result = index.insert(0, index[0])
+
+        expected = Index([index[0]] + list(index), dtype=index.dtype)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_insert_na(self, nulls_fixture, simple_index):
+        # GH 18295 (test missing)
+        index = simple_index
+        na_val = nulls_fixture
+
+        if na_val is pd.NaT:
+            expected = Index([index[0], pd.NaT] + list(index[1:]), dtype=object)
+        else:
+            expected = Index([index[0], np.nan] + list(index[1:]))
+            # GH#43921 we preserve float dtype
+            if index.dtype.kind == "f":
+                expected = Index(expected, dtype=index.dtype)
+
+        result = index.insert(1, na_val)
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_arithmetic_explicit_conversions(self, simple_index):
+        # GH 8608
+        # add/sub are overridden explicitly for Float/Int Index
+        index_cls = type(simple_index)
+        if index_cls is RangeIndex:
+            idx = RangeIndex(5)
+        else:
+            idx = index_cls(np.arange(5, dtype="int64"))
+
+        # float conversions
+        arr = np.arange(5, dtype="int64") * 3.2
+        expected = Index(arr, dtype=np.float64)
+        fidx = idx * 3.2
+        tm.assert_index_equal(fidx, expected)
+        fidx = 3.2 * idx
+        tm.assert_index_equal(fidx, expected)
+
+        # interops with numpy arrays
+        expected = Index(arr, dtype=np.float64)
+        a = np.zeros(5, dtype="float64")
+        result = fidx - a
+        tm.assert_index_equal(result, expected)
+
+        expected = Index(-arr, dtype=np.float64)
+        a = np.zeros(5, dtype="float64")
+        result = a - fidx
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("complex_dtype", [np.complex64, np.complex128])
+    def test_astype_to_complex(self, complex_dtype, simple_index):
+        result = simple_index.astype(complex_dtype)
+
+        assert type(result) is Index and result.dtype == complex_dtype
+
+    def test_cast_string(self, simple_index):
+        if isinstance(simple_index, RangeIndex):
+            pytest.skip("casting of strings not relevant for RangeIndex")
+        result = type(simple_index)(["0", "1", "2"], dtype=simple_index.dtype)
+        expected = type(simple_index)([0, 1, 2], dtype=simple_index.dtype)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/test_setops.py b/pandas/pandas/tests/indexes/test_setops.py
new file mode 100644
index 0000000000000..7cc74f4b3405c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_setops.py
@@ -0,0 +1,975 @@
+"""
+The tests in this package are to ensure the proper resultant dtypes of
+set operations.
+"""
+
+from datetime import datetime
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.cast import find_common_type
+
+from pandas import (
+    CategoricalDtype,
+    CategoricalIndex,
+    DatetimeTZDtype,
+    Index,
+    MultiIndex,
+    PeriodDtype,
+    RangeIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.api.types import (
+    is_signed_integer_dtype,
+    pandas_dtype,
+)
+
+
+def equal_contents(arr1, arr2) -> bool:
+    """
+    Checks if the set of unique elements of arr1 and arr2 are equivalent.
+    """
+    return frozenset(arr1) == frozenset(arr2)
+
+
+@pytest.fixture(
+    params=tm.ALL_REAL_NUMPY_DTYPES
+    + [
+        "object",
+        "category",
+        "datetime64[ns]",
+        "timedelta64[ns]",
+    ]
+)
+def any_dtype_for_small_pos_integer_indexes(request):
+    """
+    Dtypes that can be given to an Index with small positive integers.
+
+    This means that for any dtype `x` in the params list, `Index([1, 2, 3], dtype=x)` is
+    valid and gives the correct Index (sub-)class.
+    """
+    return request.param
+
+
+@pytest.fixture
+def index_flat2(index_flat):
+    return index_flat
+
+
+def test_union_same_types(index):
+    # Union with a non-unique, non-monotonic index raises error
+    # Only needed for bool index factory
+    idx1 = index.sort_values()
+    idx2 = index.sort_values()
+    assert idx1.union(idx2).dtype == idx1.dtype
+
+
+def test_union_different_types(index_flat, index_flat2, request):
+    # This test only considers combinations of indices
+    # GH 23525
+    idx1 = index_flat
+    idx2 = index_flat2
+
+    if (
+        not idx1.is_unique
+        and not idx2.is_unique
+        and idx1.dtype.kind == "i"
+        and idx2.dtype.kind == "b"
+    ) or (
+        not idx2.is_unique
+        and not idx1.is_unique
+        and idx2.dtype.kind == "i"
+        and idx1.dtype.kind == "b"
+    ):
+        # Each condition had idx[1|2].is_monotonic_decreasing
+        # but failed when e.g.
+        # idx1 = Index(
+        # [True, True, True, True, True, True, True, True, False, False], dtype='bool'
+        # )
+        # idx2 = Index([0, 0, 1, 1, 2, 2], dtype='int64')
+        mark = pytest.mark.xfail(
+            reason="GH#44000 True==1", raises=ValueError, strict=False
+        )
+        request.applymarker(mark)
+
+    common_dtype = find_common_type([idx1.dtype, idx2.dtype])
+
+    warn = None
+    msg = "'<' not supported between"
+    if not len(idx1) or not len(idx2):
+        pass
+    elif (idx1.dtype.kind == "c" and (not lib.is_np_dtype(idx2.dtype, "iufc"))) or (
+        idx2.dtype.kind == "c" and (not lib.is_np_dtype(idx1.dtype, "iufc"))
+    ):
+        # complex objects non-sortable
+        warn = RuntimeWarning
+    elif (
+        isinstance(idx1.dtype, PeriodDtype) and isinstance(idx2.dtype, CategoricalDtype)
+    ) or (
+        isinstance(idx2.dtype, PeriodDtype) and isinstance(idx1.dtype, CategoricalDtype)
+    ):
+        warn = FutureWarning
+        msg = r"PeriodDtype\[B\] is deprecated"
+        mark = pytest.mark.xfail(
+            reason="Warning not produced on all builds",
+            raises=AssertionError,
+            strict=False,
+        )
+        request.applymarker(mark)
+
+    any_uint64 = np.uint64 in (idx1.dtype, idx2.dtype)
+    idx1_signed = is_signed_integer_dtype(idx1.dtype)
+    idx2_signed = is_signed_integer_dtype(idx2.dtype)
+
+    # Union with a non-unique, non-monotonic index raises error
+    # This applies to the boolean index
+    idx1 = idx1.sort_values()
+    idx2 = idx2.sort_values()
+
+    with tm.assert_produces_warning(warn, match=msg):
+        res1 = idx1.union(idx2)
+        res2 = idx2.union(idx1)
+
+    if any_uint64 and (idx1_signed or idx2_signed):
+        assert res1.dtype == np.dtype("O")
+        assert res2.dtype == np.dtype("O")
+    else:
+        assert res1.dtype == common_dtype
+        assert res2.dtype == common_dtype
+
+
+@pytest.mark.parametrize(
+    "idx1,idx2",
+    [
+        (Index(np.arange(5), dtype=np.int64), RangeIndex(5)),
+        (Index(np.arange(5), dtype=np.float64), Index(np.arange(5), dtype=np.int64)),
+        (Index(np.arange(5), dtype=np.float64), RangeIndex(5)),
+        (Index(np.arange(5), dtype=np.float64), Index(np.arange(5), dtype=np.uint64)),
+    ],
+)
+def test_compatible_inconsistent_pairs(idx1, idx2):
+    # GH 23525
+    res1 = idx1.union(idx2)
+    res2 = idx2.union(idx1)
+
+    assert res1.dtype in (idx1.dtype, idx2.dtype)
+    assert res2.dtype in (idx1.dtype, idx2.dtype)
+
+
+@pytest.mark.parametrize(
+    "left, right, expected",
+    [
+        ("int64", "int64", "int64"),
+        ("int64", "uint64", "object"),
+        ("int64", "float64", "float64"),
+        ("uint64", "float64", "float64"),
+        ("uint64", "uint64", "uint64"),
+        ("float64", "float64", "float64"),
+        ("datetime64[ns]", "int64", "object"),
+        ("datetime64[ns]", "uint64", "object"),
+        ("datetime64[ns]", "float64", "object"),
+        ("datetime64[ns, CET]", "int64", "object"),
+        ("datetime64[ns, CET]", "uint64", "object"),
+        ("datetime64[ns, CET]", "float64", "object"),
+        ("Period[D]", "int64", "object"),
+        ("Period[D]", "uint64", "object"),
+        ("Period[D]", "float64", "object"),
+    ],
+)
+@pytest.mark.parametrize("names", [("foo", "foo", "foo"), ("foo", "bar", None)])
+def test_union_dtypes(left, right, expected, names):
+    left = pandas_dtype(left)
+    right = pandas_dtype(right)
+    a = Index([], dtype=left, name=names[0])
+    b = Index([], dtype=right, name=names[1])
+    result = a.union(b)
+    assert result.dtype == expected
+    assert result.name == names[2]
+
+    # Testing name retention
+    # TODO: pin down desired dtype; do we want it to be commutative?
+    result = a.intersection(b)
+    assert result.name == names[2]
+
+
+@pytest.mark.parametrize("values", [[1, 2, 2, 3], [3, 3]])
+def test_intersection_duplicates(values):
+    # GH#31326
+    a = Index(values)
+    b = Index([3, 3])
+    result = a.intersection(b)
+    expected = Index([3])
+    tm.assert_index_equal(result, expected)
+
+
+class TestSetOps:
+    # Set operation tests shared by all indexes in the `index` fixture
+    @pytest.mark.parametrize("case", [0.5, "xxx"])
+    @pytest.mark.parametrize(
+        "method", ["intersection", "union", "difference", "symmetric_difference"]
+    )
+    def test_set_ops_error_cases(self, case, method, index):
+        # non-iterable input
+        msg = "Input must be Index or array-like"
+        with pytest.raises(TypeError, match=msg):
+            getattr(index, method)(case)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_intersection_base(self, index):
+        if isinstance(index, CategoricalIndex):
+            pytest.skip(f"Not relevant for {type(index).__name__}")
+
+        first = index[:5].unique()
+        second = index[:3].unique()
+        intersect = first.intersection(second)
+        tm.assert_index_equal(intersect, second)
+
+        if isinstance(index.dtype, DatetimeTZDtype):
+            # The second.values below will drop tz, so the rest of this test
+            #  is not applicable.
+            return
+
+        # GH#10149
+        cases = [second.to_numpy(), second.to_series(), second.to_list()]
+        for case in cases:
+            result = first.intersection(case)
+            assert equal_contents(result, second)
+
+        if isinstance(index, MultiIndex):
+            msg = "other must be a MultiIndex or a list of tuples"
+            with pytest.raises(TypeError, match=msg):
+                first.intersection([1, 2, 3])
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_union_base(self, index):
+        index = index.unique()
+        first = index[3:]
+        second = index[:5]
+        everything = index
+
+        union = first.union(second)
+        tm.assert_index_equal(union.sort_values(), everything.sort_values())
+
+        if isinstance(index.dtype, DatetimeTZDtype):
+            # The second.values below will drop tz, so the rest of this test
+            #  is not applicable.
+            return
+
+        # GH#10149
+        cases = [second.to_numpy(), second.to_series(), second.to_list()]
+        for case in cases:
+            result = first.union(case)
+            assert equal_contents(result, everything)
+
+        if isinstance(index, MultiIndex):
+            msg = "other must be a MultiIndex or a list of tuples"
+            with pytest.raises(TypeError, match=msg):
+                first.union([1, 2, 3])
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_difference_base(self, sort, index):
+        first = index[2:]
+        second = index[:4]
+        if index.inferred_type == "boolean":
+            # i think (TODO: be sure) there assumptions baked in about
+            #  the index fixture that don't hold here?
+            answer = set(first).difference(set(second))
+        elif isinstance(index, CategoricalIndex):
+            answer = []
+        else:
+            answer = index[4:]
+        result = first.difference(second, sort)
+        assert equal_contents(result, answer)
+
+        # GH#10149
+        cases = [second.to_numpy(), second.to_series(), second.to_list()]
+        for case in cases:
+            result = first.difference(case, sort)
+            assert equal_contents(result, answer)
+
+        if isinstance(index, MultiIndex):
+            msg = "other must be a MultiIndex or a list of tuples"
+            with pytest.raises(TypeError, match=msg):
+                first.difference([1, 2, 3], sort)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_symmetric_difference(self, index, using_infer_string, request):
+        if (
+            using_infer_string
+            and index.dtype == "object"
+            and index.inferred_type == "string"
+        ):
+            request.applymarker(pytest.mark.xfail(reason="TODO: infer_string"))
+        if isinstance(index, CategoricalIndex):
+            pytest.skip(f"Not relevant for {type(index).__name__}")
+        if len(index) < 2:
+            pytest.skip("Too few values for test")
+        if index[0] in index[1:] or index[-1] in index[:-1]:
+            # index fixture has e.g. an index of bools that does not satisfy this,
+            #  another with [0, 0, 1, 1, 2, 2]
+            pytest.skip("Index values no not satisfy test condition.")
+
+        first = index[1:]
+        second = index[:-1]
+        answer = index[[0, -1]]
+        result = first.symmetric_difference(second)
+        tm.assert_index_equal(result.sort_values(), answer.sort_values())
+
+        # GH#10149
+        cases = [second.to_numpy(), second.to_series(), second.to_list()]
+        for case in cases:
+            result = first.symmetric_difference(case)
+            assert equal_contents(result, answer)
+
+        if isinstance(index, MultiIndex):
+            msg = "other must be a MultiIndex or a list of tuples"
+            with pytest.raises(TypeError, match=msg):
+                first.symmetric_difference([1, 2, 3])
+
+    @pytest.mark.parametrize(
+        "fname, sname, expected_name",
+        [
+            ("A", "A", "A"),
+            ("A", "B", None),
+            ("A", None, None),
+            (None, "B", None),
+            (None, None, None),
+        ],
+    )
+    def test_corner_union(self, index_flat, fname, sname, expected_name):
+        # GH#9943, GH#9862
+        # Test unions with various name combinations
+        # Do not test MultiIndex or repeats
+        if not index_flat.is_unique:
+            index = index_flat.unique()
+        else:
+            index = index_flat
+
+        # Test copy.union(copy)
+        first = index.copy().set_names(fname)
+        second = index.copy().set_names(sname)
+        union = first.union(second)
+        expected = index.copy().set_names(expected_name)
+        tm.assert_index_equal(union, expected)
+
+        # Test copy.union(empty)
+        first = index.copy().set_names(fname)
+        second = index.drop(index).set_names(sname)
+        union = first.union(second)
+        expected = index.copy().set_names(expected_name)
+        tm.assert_index_equal(union, expected)
+
+        # Test empty.union(copy)
+        first = index.drop(index).set_names(fname)
+        second = index.copy().set_names(sname)
+        union = first.union(second)
+        expected = index.copy().set_names(expected_name)
+        tm.assert_index_equal(union, expected)
+
+        # Test empty.union(empty)
+        first = index.drop(index).set_names(fname)
+        second = index.drop(index).set_names(sname)
+        union = first.union(second)
+        expected = index.drop(index).set_names(expected_name)
+        tm.assert_index_equal(union, expected)
+
+    @pytest.mark.parametrize(
+        "fname, sname, expected_name",
+        [
+            ("A", "A", "A"),
+            ("A", "B", None),
+            ("A", None, None),
+            (None, "B", None),
+            (None, None, None),
+        ],
+    )
+    def test_union_unequal(self, index_flat, fname, sname, expected_name):
+        if not index_flat.is_unique:
+            index = index_flat.unique()
+        else:
+            index = index_flat
+
+        # test copy.union(subset) - need sort for unicode and string
+        first = index.copy().set_names(fname)
+        second = index[1:].set_names(sname)
+        union = first.union(second).sort_values()
+        expected = index.set_names(expected_name).sort_values()
+        tm.assert_index_equal(union, expected)
+
+    @pytest.mark.parametrize(
+        "fname, sname, expected_name",
+        [
+            ("A", "A", "A"),
+            ("A", "B", None),
+            ("A", None, None),
+            (None, "B", None),
+            (None, None, None),
+        ],
+    )
+    def test_corner_intersect(self, index_flat, fname, sname, expected_name):
+        # GH#35847
+        # Test intersections with various name combinations
+        if not index_flat.is_unique:
+            index = index_flat.unique()
+        else:
+            index = index_flat
+
+        # Test copy.intersection(copy)
+        first = index.copy().set_names(fname)
+        second = index.copy().set_names(sname)
+        intersect = first.intersection(second)
+        expected = index.copy().set_names(expected_name)
+        tm.assert_index_equal(intersect, expected)
+
+        # Test copy.intersection(empty)
+        first = index.copy().set_names(fname)
+        second = index.drop(index).set_names(sname)
+        intersect = first.intersection(second)
+        expected = index.drop(index).set_names(expected_name)
+        tm.assert_index_equal(intersect, expected)
+
+        # Test empty.intersection(copy)
+        first = index.drop(index).set_names(fname)
+        second = index.copy().set_names(sname)
+        intersect = first.intersection(second)
+        expected = index.drop(index).set_names(expected_name)
+        tm.assert_index_equal(intersect, expected)
+
+        # Test empty.intersection(empty)
+        first = index.drop(index).set_names(fname)
+        second = index.drop(index).set_names(sname)
+        intersect = first.intersection(second)
+        expected = index.drop(index).set_names(expected_name)
+        tm.assert_index_equal(intersect, expected)
+
+    @pytest.mark.parametrize(
+        "fname, sname, expected_name",
+        [
+            ("A", "A", "A"),
+            ("A", "B", None),
+            ("A", None, None),
+            (None, "B", None),
+            (None, None, None),
+        ],
+    )
+    def test_intersect_unequal(self, index_flat, fname, sname, expected_name):
+        if not index_flat.is_unique:
+            index = index_flat.unique()
+        else:
+            index = index_flat
+
+        # test copy.intersection(subset) - need sort for unicode and string
+        first = index.copy().set_names(fname)
+        second = index[1:].set_names(sname)
+        intersect = first.intersection(second).sort_values()
+        expected = index[1:].set_names(expected_name).sort_values()
+        tm.assert_index_equal(intersect, expected)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_intersection_name_retention_with_nameless(self, index):
+        if isinstance(index, MultiIndex):
+            index = index.rename(list(range(index.nlevels)))
+        else:
+            index = index.rename("foo")
+
+        other = np.asarray(index)
+
+        result = index.intersection(other)
+        assert result.name == index.name
+
+        # empty other, same dtype
+        result = index.intersection(other[:0])
+        assert result.name == index.name
+
+        # empty `self`
+        result = index[:0].intersection(other)
+        assert result.name == index.name
+
+    def test_difference_preserves_type_empty(self, index, sort):
+        # GH#20040
+        # If taking difference of a set and itself, it
+        # needs to preserve the type of the index
+        if not index.is_unique:
+            pytest.skip("Not relevant since index is not unique")
+        result = index.difference(index, sort=sort)
+        expected = index[:0]
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_difference_name_retention_equals(self, index, names):
+        if isinstance(index, MultiIndex):
+            names = [[x] * index.nlevels for x in names]
+        index = index.rename(names[0])
+        other = index.rename(names[1])
+
+        assert index.equals(other)
+
+        result = index.difference(other)
+        expected = index[:0].rename(names[2])
+        tm.assert_index_equal(result, expected)
+
+    def test_intersection_difference_match_empty(self, index, sort):
+        # GH#20040
+        # Test that the intersection of an index with an
+        # empty index produces the same index as the difference
+        # of an index with itself.  Test for all types
+        if not index.is_unique:
+            pytest.skip("Not relevant because index is not unique")
+        inter = index.intersection(index[:0])
+        diff = index.difference(index, sort=sort)
+        tm.assert_index_equal(inter, diff, exact=True)
+
+
+@pytest.mark.filterwarnings("ignore:invalid value encountered in cast:RuntimeWarning")
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.parametrize(
+    "method", ["intersection", "union", "difference", "symmetric_difference"]
+)
+def test_setop_with_categorical(index_flat, sort, method, using_infer_string):
+    # MultiIndex tested separately in tests.indexes.multi.test_setops
+    index = index_flat
+
+    other = index.astype("category")
+    exact = "equiv" if isinstance(index, RangeIndex) else True
+
+    result = getattr(index, method)(other, sort=sort)
+    expected = getattr(index, method)(index, sort=sort)
+    if (
+        using_infer_string
+        and index.empty
+        and method in ("union", "symmetric_difference")
+    ):
+        expected = expected.astype("category")
+    tm.assert_index_equal(result, expected, exact=exact)
+
+    result = getattr(index, method)(other[:5], sort=sort)
+    expected = getattr(index, method)(index[:5], sort=sort)
+    if (
+        using_infer_string
+        and index.empty
+        and method in ("union", "symmetric_difference")
+    ):
+        expected = expected.astype("category")
+    tm.assert_index_equal(result, expected, exact=exact)
+
+
+def test_intersection_duplicates_all_indexes(index):
+    # GH#38743
+    if index.empty:
+        # No duplicates in empty indexes
+        pytest.skip("Not relevant for empty Index")
+
+    idx = index
+    idx_non_unique = idx[[0, 0, 1, 2]]
+
+    assert idx.intersection(idx_non_unique).equals(idx_non_unique.intersection(idx))
+    assert idx.intersection(idx_non_unique).is_unique
+
+
+def test_union_duplicate_index_subsets_of_each_other(
+    any_dtype_for_small_pos_integer_indexes,
+):
+    # GH#31326
+    dtype = any_dtype_for_small_pos_integer_indexes
+    a = Index([1, 2, 2, 3], dtype=dtype)
+    b = Index([3, 3, 4], dtype=dtype)
+
+    expected = Index([1, 2, 2, 3, 3, 4], dtype=dtype)
+    if isinstance(a, CategoricalIndex):
+        expected = Index([1, 2, 2, 3, 3, 4])
+    result = a.union(b)
+    tm.assert_index_equal(result, expected)
+    result = a.union(b, sort=False)
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_with_duplicate_index_and_non_monotonic(
+    any_dtype_for_small_pos_integer_indexes,
+):
+    # GH#36289
+    dtype = any_dtype_for_small_pos_integer_indexes
+    a = Index([1, 0, 0], dtype=dtype)
+    b = Index([0, 1], dtype=dtype)
+    expected = Index([0, 0, 1], dtype=dtype)
+
+    result = a.union(b)
+    tm.assert_index_equal(result, expected)
+
+    result = b.union(a)
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_duplicate_index_different_dtypes():
+    # GH#36289
+    a = Index([1, 2, 2, 3])
+    b = Index(["1", "0", "0"])
+    expected = Index([1, 2, 2, 3, "1", "0", "0"])
+    result = a.union(b, sort=False)
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_same_value_duplicated_in_both():
+    # GH#36289
+    a = Index([0, 0, 1])
+    b = Index([0, 0, 1, 2])
+    result = a.union(b)
+    expected = Index([0, 0, 1, 2])
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("dup", [1, np.nan])
+def test_union_nan_in_both(dup):
+    # GH#36289
+    a = Index([np.nan, 1, 2, 2])
+    b = Index([np.nan, dup, 1, 2])
+    result = a.union(b, sort=False)
+    expected = Index([np.nan, dup, 1.0, 2.0, 2.0])
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_rangeindex_sort_true():
+    # GH 53490
+    idx1 = RangeIndex(1, 100, 6)
+    idx2 = RangeIndex(1, 50, 3)
+    result = idx1.union(idx2, sort=True)
+    expected = Index(
+        [
+            1,
+            4,
+            7,
+            10,
+            13,
+            16,
+            19,
+            22,
+            25,
+            28,
+            31,
+            34,
+            37,
+            40,
+            43,
+            46,
+            49,
+            55,
+            61,
+            67,
+            73,
+            79,
+            85,
+            91,
+            97,
+        ]
+    )
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_with_duplicate_index_not_subset_and_non_monotonic(
+    any_dtype_for_small_pos_integer_indexes,
+):
+    # GH#36289
+    dtype = any_dtype_for_small_pos_integer_indexes
+    a = Index([1, 0, 2], dtype=dtype)
+    b = Index([0, 0, 1], dtype=dtype)
+    expected = Index([0, 0, 1, 2], dtype=dtype)
+    if isinstance(a, CategoricalIndex):
+        expected = Index([0, 0, 1, 2])
+
+    result = a.union(b)
+    tm.assert_index_equal(result, expected)
+
+    result = b.union(a)
+    tm.assert_index_equal(result, expected)
+
+
+def test_union_int_categorical_with_nan():
+    ci = CategoricalIndex([1, 2, np.nan])
+    assert ci.categories.dtype.kind == "i"
+
+    idx = Index([1, 2])
+
+    result = idx.union(ci)
+    expected = Index([1, 2, np.nan], dtype=np.float64)
+    tm.assert_index_equal(result, expected)
+
+    result = ci.union(idx)
+    tm.assert_index_equal(result, expected)
+
+
+class TestSetOpsUnsorted:
+    # These may eventually belong in a dtype-specific test_setops, or
+    #  parametrized over a more general fixture
+    def test_intersect_str_dates(self):
+        dt_dates = [datetime(2012, 2, 9), datetime(2012, 2, 22)]
+
+        index1 = Index(dt_dates, dtype=object)
+        index2 = Index(["aa"], dtype=object)
+        result = index2.intersection(index1)
+
+        expected = Index([], dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_intersection(self, index, sort):
+        first = index[:20]
+        second = index[:10]
+        intersect = first.intersection(second, sort=sort)
+        if sort in (None, False):
+            tm.assert_index_equal(intersect.sort_values(), second.sort_values())
+        else:
+            tm.assert_index_equal(intersect, second)
+
+        # Corner cases
+        inter = first.intersection(first, sort=sort)
+        assert inter is first
+
+    @pytest.mark.parametrize(
+        "index2_name,keeps_name",
+        [
+            ("index", True),  # preserve same name
+            ("other", False),  # drop diff names
+            (None, False),
+        ],
+    )
+    def test_intersection_name_preservation(self, index2_name, keeps_name, sort):
+        index2 = Index([3, 4, 5, 6, 7], name=index2_name)
+        index1 = Index([1, 2, 3, 4, 5], name="index")
+        expected = Index([3, 4, 5])
+        result = index1.intersection(index2, sort)
+
+        if keeps_name:
+            expected.name = "index"
+
+        assert result.name == expected.name
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    @pytest.mark.parametrize(
+        "first_name,second_name,expected_name",
+        [("A", "A", "A"), ("A", "B", None), (None, "B", None)],
+    )
+    def test_intersection_name_preservation2(
+        self, index, first_name, second_name, expected_name, sort
+    ):
+        first = index[5:20]
+        second = index[:10]
+        first.name = first_name
+        second.name = second_name
+        intersect = first.intersection(second, sort=sort)
+        assert intersect.name == expected_name
+
+    def test_chained_union(self, sort):
+        # Chained unions handles names correctly
+        i1 = Index([1, 2], name="i1")
+        i2 = Index([5, 6], name="i2")
+        i3 = Index([3, 4], name="i3")
+        union = i1.union(i2.union(i3, sort=sort), sort=sort)
+        expected = i1.union(i2, sort=sort).union(i3, sort=sort)
+        tm.assert_index_equal(union, expected)
+
+        j1 = Index([1, 2], name="j1")
+        j2 = Index([], name="j2")
+        j3 = Index([], name="j3")
+        union = j1.union(j2.union(j3, sort=sort), sort=sort)
+        expected = j1.union(j2, sort=sort).union(j3, sort=sort)
+        tm.assert_index_equal(union, expected)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_union(self, index, sort):
+        first = index[5:20]
+        second = index[:10]
+        everything = index[:20]
+
+        union = first.union(second, sort=sort)
+        if sort in (None, False):
+            tm.assert_index_equal(union.sort_values(), everything.sort_values())
+        else:
+            tm.assert_index_equal(union, everything)
+
+    @pytest.mark.parametrize("klass", [np.array, Series, list])
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_union_from_iterables(self, index, klass, sort):
+        # GH#10149
+        first = index[5:20]
+        second = index[:10]
+        everything = index[:20]
+
+        case = klass(second.values)
+        result = first.union(case, sort=sort)
+        if sort in (None, False):
+            tm.assert_index_equal(result.sort_values(), everything.sort_values())
+        else:
+            tm.assert_index_equal(result, everything)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_union_identity(self, index, sort):
+        first = index[5:20]
+
+        union = first.union(first, sort=sort)
+        # i.e. identity is not preserved when sort is True
+        assert (union is first) is (not sort)
+
+        # This should no longer be the same object, since [] is not consistent,
+        # both objects will be recast to dtype('O')
+        union = first.union(Index([], dtype=first.dtype), sort=sort)
+        assert (union is first) is (not sort)
+
+        union = Index([], dtype=first.dtype).union(first, sort=sort)
+        assert (union is first) is (not sort)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    @pytest.mark.parametrize("second_name,expected", [(None, None), ("name", "name")])
+    def test_difference_name_preservation(self, index, second_name, expected, sort):
+        first = index[5:20]
+        second = index[:10]
+        answer = index[10:20]
+
+        first.name = "name"
+        second.name = second_name
+        result = first.difference(second, sort=sort)
+
+        if sort is True:
+            tm.assert_index_equal(result, answer)
+        else:
+            answer.name = second_name
+            tm.assert_index_equal(result.sort_values(), answer.sort_values())
+
+        if expected is None:
+            assert result.name is None
+        else:
+            assert result.name == expected
+
+    def test_difference_empty_arg(self, index, sort):
+        first = index.copy()
+        first = first[5:20]
+        first.name = "name"
+        result = first.difference([], sort)
+        expected = index[5:20].unique()
+        expected.name = "name"
+        tm.assert_index_equal(result, expected)
+
+    def test_difference_should_not_compare(self):
+        # GH 55113
+        left = Index([1, 1])
+        right = Index([True])
+        result = left.difference(right)
+        expected = Index([1])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_difference_identity(self, index, sort):
+        first = index[5:20]
+        first.name = "name"
+        result = first.difference(first, sort)
+
+        assert len(result) == 0
+        assert result.name == first.name
+
+    @pytest.mark.parametrize("index", ["string"], indirect=True)
+    def test_difference_sort(self, index, sort):
+        first = index[5:20]
+        second = index[:10]
+
+        result = first.difference(second, sort)
+        expected = index[10:20]
+
+        if sort is None:
+            expected = expected.sort_values()
+
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("opname", ["difference", "symmetric_difference"])
+    def test_difference_incomparable(self, opname):
+        a = Index([3, Timestamp("2000"), 1])
+        b = Index([2, Timestamp("1999"), 1])
+        op = operator.methodcaller(opname, b)
+
+        with tm.assert_produces_warning(RuntimeWarning, match="not supported between"):
+            # sort=None, the default
+            result = op(a)
+        expected = Index([3, Timestamp("2000"), 2, Timestamp("1999")])
+        if opname == "difference":
+            expected = expected[:2]
+        tm.assert_index_equal(result, expected)
+
+        # sort=False
+        op = operator.methodcaller(opname, b, sort=False)
+        result = op(a)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("opname", ["difference", "symmetric_difference"])
+    def test_difference_incomparable_true(self, opname):
+        a = Index([3, Timestamp("2000"), 1])
+        b = Index([2, Timestamp("1999"), 1])
+        op = operator.methodcaller(opname, b, sort=True)
+
+        msg = "'<' not supported between instances of 'Timestamp' and 'int'"
+        with pytest.raises(TypeError, match=msg):
+            op(a)
+
+    def test_symmetric_difference_mi(self, sort):
+        index1 = MultiIndex.from_tuples(zip(["foo", "bar", "baz"], [1, 2, 3]))
+        index2 = MultiIndex.from_tuples([("foo", 1), ("bar", 3)])
+        result = index1.symmetric_difference(index2, sort=sort)
+        expected = MultiIndex.from_tuples([("bar", 2), ("baz", 3), ("bar", 3)])
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index2,expected",
+        [
+            ([0, 1, np.nan], [2.0, 3.0, 0.0]),
+            ([0, 1], [np.nan, 2.0, 3.0, 0.0]),
+        ],
+    )
+    def test_symmetric_difference_missing(self, index2, expected, sort):
+        index2 = Index(index2)
+        expected = Index(expected)
+        # GH#13514 change: {nan} - {nan} == {}
+        # (GH#6444, sorting of nans, is no longer an issue)
+        index1 = Index([1, np.nan, 2, 3])
+
+        result = index1.symmetric_difference(index2, sort=sort)
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+
+    def test_symmetric_difference_non_index(self, sort):
+        index1 = Index([1, 2, 3, 4], name="index1")
+        index2 = np.array([2, 3, 4, 5])
+        expected = Index([1, 5], name="index1")
+        result = index1.symmetric_difference(index2, sort=sort)
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+        assert result.name == "index1"
+
+        result = index1.symmetric_difference(index2, result_name="new_name", sort=sort)
+        expected.name = "new_name"
+        if sort in (None, True):
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
+        assert result.name == "new_name"
+
+    def test_union_ea_dtypes(self, any_numeric_ea_and_arrow_dtype):
+        # GH#51365
+        idx = Index([1, 2, 3], dtype=any_numeric_ea_and_arrow_dtype)
+        idx2 = Index([3, 4, 5], dtype=any_numeric_ea_and_arrow_dtype)
+        result = idx.union(idx2)
+        expected = Index([1, 2, 3, 4, 5], dtype=any_numeric_ea_and_arrow_dtype)
+        tm.assert_index_equal(result, expected)
+
+    def test_union_string_array(self, any_string_dtype):
+        idx1 = Index(["a"], dtype=any_string_dtype)
+        idx2 = Index(["b"], dtype=any_string_dtype)
+        result = idx1.union(idx2)
+        expected = Index(["a", "b"], dtype=any_string_dtype)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/test_subclass.py b/pandas/pandas/tests/indexes/test_subclass.py
new file mode 100644
index 0000000000000..a8ba8c3090cf2
--- /dev/null
+++ b/pandas/pandas/tests/indexes/test_subclass.py
@@ -0,0 +1,41 @@
+"""
+Tests involving custom Index subclasses
+"""
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+class CustomIndex(Index):
+    def __new__(cls, data, name=None):
+        # assert that this index class cannot hold strings
+        if any(isinstance(val, str) for val in data):
+            raise TypeError("CustomIndex cannot hold strings")
+
+        if name is None and hasattr(data, "name"):
+            name = data.name
+        data = np.array(data, dtype="O")
+
+        return cls._simple_new(data, name)
+
+
+def test_insert_fallback_to_base_index():
+    # https://github.com/pandas-dev/pandas/issues/47071
+
+    idx = CustomIndex([1, 2, 3])
+    result = idx.insert(0, "string")
+    expected = Index(["string", 1, 2, 3], dtype=object)
+    tm.assert_index_equal(result, expected)
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((2, 3)),
+        columns=idx,
+        index=Index([1, 2], name="string"),
+    )
+    result = df.reset_index()
+    tm.assert_index_equal(result.columns, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/__init__.py b/pandas/pandas/tests/indexes/timedeltas/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/__init__.py b/pandas/pandas/tests/indexes/timedeltas/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_astype.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_astype.py
new file mode 100644
index 0000000000000..5166cadae499e
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_astype.py
@@ -0,0 +1,181 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    NaT,
+    Timedelta,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import TimedeltaArray
+
+
+class TestTimedeltaIndex:
+    def test_astype_object(self):
+        idx = timedelta_range(start="1 days", periods=4, freq="D", name="idx")
+        expected_list = [
+            Timedelta("1 days"),
+            Timedelta("2 days"),
+            Timedelta("3 days"),
+            Timedelta("4 days"),
+        ]
+        result = idx.astype(object)
+        expected = Index(expected_list, dtype=object, name="idx")
+        tm.assert_index_equal(result, expected)
+        assert idx.tolist() == expected_list
+
+    def test_astype_object_with_nat(self):
+        idx = TimedeltaIndex(
+            [timedelta(days=1), timedelta(days=2), NaT, timedelta(days=4)], name="idx"
+        )
+        expected_list = [
+            Timedelta("1 days"),
+            Timedelta("2 days"),
+            NaT,
+            Timedelta("4 days"),
+        ]
+        result = idx.astype(object)
+        expected = Index(expected_list, dtype=object, name="idx")
+        tm.assert_index_equal(result, expected)
+        assert idx.tolist() == expected_list
+
+    def test_astype(self, using_infer_string):
+        # GH 13149, GH 13209
+        idx = TimedeltaIndex([1e14, "NaT", NaT, np.nan], name="idx")
+
+        result = idx.astype(object)
+        expected = Index(
+            [Timedelta("1 days 03:46:40")] + [NaT] * 3, dtype=object, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.astype(np.int64)
+        expected = Index(
+            [100000000000000] + [-9223372036854775808] * 3, dtype=np.int64, name="idx"
+        )
+        tm.assert_index_equal(result, expected)
+
+        result = idx.astype(str)
+        if using_infer_string:
+            expected = Index(
+                [str(x) if x is not NaT else None for x in idx], name="idx", dtype="str"
+            )
+        else:
+            expected = Index([str(x) for x in idx], name="idx", dtype=object)
+        tm.assert_index_equal(result, expected)
+
+        rng = timedelta_range("1 days", periods=10)
+        result = rng.astype("i8")
+        tm.assert_index_equal(result, Index(rng.asi8))
+        tm.assert_numpy_array_equal(rng.asi8, result.values)
+
+    def test_astype_uint(self):
+        arr = timedelta_range("1h", periods=2)
+
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint64")
+        with pytest.raises(TypeError, match=r"Do obj.astype\('int64'\)"):
+            arr.astype("uint32")
+
+    def test_astype_timedelta64(self):
+        # GH 13149, GH 13209
+        idx = TimedeltaIndex([1e14, "NaT", NaT, np.nan])
+
+        msg = (
+            r"Cannot convert from timedelta64\[ns\] to timedelta64. "
+            "Supported resolutions are 's', 'ms', 'us', 'ns'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.astype("timedelta64")
+
+        result = idx.astype("timedelta64[ns]")
+        tm.assert_index_equal(result, idx)
+        assert result is not idx
+
+        result = idx.astype("timedelta64[ns]", copy=False)
+        tm.assert_index_equal(result, idx)
+        assert result is idx
+
+    def test_astype_to_td64d_raises(self, index_or_series):
+        # We don't support "D" reso
+        scalar = Timedelta(days=31)
+        td = index_or_series(
+            [scalar, scalar, scalar + timedelta(minutes=5, seconds=3), NaT],
+            dtype="m8[ns]",
+        )
+        msg = (
+            r"Cannot convert from timedelta64\[ns\] to timedelta64\[D\]. "
+            "Supported resolutions are 's', 'ms', 'us', 'ns'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            td.astype("timedelta64[D]")
+
+    def test_astype_ms_to_s(self, index_or_series):
+        scalar = Timedelta(days=31)
+        td = index_or_series(
+            [scalar, scalar, scalar + timedelta(minutes=5, seconds=3), NaT],
+            dtype="m8[ns]",
+        )
+
+        exp_values = np.asarray(td).astype("m8[s]")
+        exp_tda = TimedeltaArray._simple_new(exp_values, dtype=exp_values.dtype)
+        expected = index_or_series(exp_tda)
+        assert expected.dtype == "m8[s]"
+        result = td.astype("timedelta64[s]")
+        tm.assert_equal(result, expected)
+
+    def test_astype_freq_conversion(self):
+        # pre-2.0 td64 astype converted to float64. now for supported units
+        #  (s, ms, us, ns) this converts to the requested dtype.
+        # This matches TDA and Series
+        tdi = timedelta_range("1 Day", periods=30)
+
+        res = tdi.astype("m8[s]")
+        exp_values = np.asarray(tdi).astype("m8[s]")
+        exp_tda = TimedeltaArray._simple_new(
+            exp_values, dtype=exp_values.dtype, freq=tdi.freq
+        )
+        expected = Index(exp_tda)
+        assert expected.dtype == "m8[s]"
+        tm.assert_index_equal(res, expected)
+
+        # check this matches Series and TimedeltaArray
+        res = tdi._data.astype("m8[s]")
+        tm.assert_equal(res, expected._values)
+
+        res = tdi.to_series().astype("m8[s]")
+        tm.assert_equal(res._values, expected._values._with_freq(None))
+
+    @pytest.mark.parametrize("dtype", [float, "datetime64", "datetime64[ns]"])
+    def test_astype_raises(self, dtype):
+        # GH 13149, GH 13209
+        idx = TimedeltaIndex([1e14, "NaT", NaT, np.nan])
+        msg = "Cannot cast TimedeltaIndex to dtype"
+        with pytest.raises(TypeError, match=msg):
+            idx.astype(dtype)
+
+    def test_astype_category(self):
+        obj = timedelta_range("1h", periods=2, freq="h")
+
+        result = obj.astype("category")
+        expected = pd.CategoricalIndex([Timedelta("1h"), Timedelta("2h")])
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype("category")
+        expected = expected.values
+        tm.assert_categorical_equal(result, expected)
+
+    def test_astype_array_fallback(self):
+        obj = timedelta_range("1h", periods=2)
+        result = obj.astype(bool)
+        expected = Index(np.array([True, True]))
+        tm.assert_index_equal(result, expected)
+
+        result = obj._data.astype(bool)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_factorize.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_factorize.py
new file mode 100644
index 0000000000000..24ab3888412d0
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_factorize.py
@@ -0,0 +1,40 @@
+import numpy as np
+
+from pandas import (
+    TimedeltaIndex,
+    factorize,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexFactorize:
+    def test_factorize(self):
+        idx1 = TimedeltaIndex(["1 day", "1 day", "2 day", "2 day", "3 day", "3 day"])
+
+        exp_arr = np.array([0, 0, 1, 1, 2, 2], dtype=np.intp)
+        exp_idx = TimedeltaIndex(["1 day", "2 day", "3 day"])
+
+        arr, idx = idx1.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+        arr, idx = idx1.factorize(sort=True)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, exp_idx)
+        assert idx.freq == exp_idx.freq
+
+    def test_factorize_preserves_freq(self):
+        # GH#38120 freq should be preserved
+        idx3 = timedelta_range("1 day", periods=4, freq="s")
+        exp_arr = np.array([0, 1, 2, 3], dtype=np.intp)
+        arr, idx = idx3.factorize()
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, idx3)
+        assert idx.freq == idx3.freq
+
+        arr, idx = factorize(idx3)
+        tm.assert_numpy_array_equal(arr, exp_arr)
+        tm.assert_index_equal(idx, idx3)
+        assert idx.freq == idx3.freq
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_fillna.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_fillna.py
new file mode 100644
index 0000000000000..40aa95d0a4605
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_fillna.py
@@ -0,0 +1,22 @@
+from pandas import (
+    Index,
+    NaT,
+    Timedelta,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+
+
+class TestFillNA:
+    def test_fillna_timedelta(self):
+        # GH#11343
+        idx = TimedeltaIndex(["1 day", NaT, "3 day"])
+
+        exp = TimedeltaIndex(["1 day", "2 day", "3 day"])
+        tm.assert_index_equal(idx.fillna(Timedelta("2 day")), exp)
+
+        exp = TimedeltaIndex(["1 day", "3 hour", "3 day"])
+        idx.fillna(Timedelta("3 hour"))
+
+        exp = Index([Timedelta("1 day"), "x", Timedelta("3 day")], dtype=object)
+        tm.assert_index_equal(idx.fillna("x"), exp)
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_insert.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_insert.py
new file mode 100644
index 0000000000000..f8164102815f6
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_insert.py
@@ -0,0 +1,145 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+from pandas import (
+    Index,
+    Timedelta,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexInsert:
+    def test_insert(self):
+        idx = TimedeltaIndex(["4day", "1day", "2day"], name="idx")
+
+        result = idx.insert(2, timedelta(days=5))
+        exp = TimedeltaIndex(["4day", "1day", "5day", "2day"], name="idx")
+        tm.assert_index_equal(result, exp)
+
+        # insertion of non-datetime should coerce to object index
+        result = idx.insert(1, "inserted")
+        expected = Index(
+            [Timedelta("4day"), "inserted", Timedelta("1day"), Timedelta("2day")],
+            name="idx",
+        )
+        assert not isinstance(result, TimedeltaIndex)
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+
+        idx = timedelta_range("1day 00:00:01", periods=3, freq="s", name="idx")
+
+        # preserve freq
+        expected_0 = TimedeltaIndex(
+            ["1day", "1day 00:00:01", "1day 00:00:02", "1day 00:00:03"],
+            name="idx",
+            freq="s",
+        )
+        expected_3 = TimedeltaIndex(
+            ["1day 00:00:01", "1day 00:00:02", "1day 00:00:03", "1day 00:00:04"],
+            name="idx",
+            freq="s",
+        )
+
+        # reset freq to None
+        expected_1_nofreq = TimedeltaIndex(
+            ["1day 00:00:01", "1day 00:00:01", "1day 00:00:02", "1day 00:00:03"],
+            name="idx",
+            freq=None,
+        )
+        expected_3_nofreq = TimedeltaIndex(
+            ["1day 00:00:01", "1day 00:00:02", "1day 00:00:03", "1day 00:00:05"],
+            name="idx",
+            freq=None,
+        )
+
+        cases = [
+            (0, Timedelta("1day"), expected_0),
+            (-3, Timedelta("1day"), expected_0),
+            (3, Timedelta("1day 00:00:04"), expected_3),
+            (1, Timedelta("1day 00:00:01"), expected_1_nofreq),
+            (3, Timedelta("1day 00:00:05"), expected_3_nofreq),
+        ]
+
+        for n, d, expected in cases:
+            result = idx.insert(n, d)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+    @pytest.mark.parametrize(
+        "null", [None, np.nan, np.timedelta64("NaT"), pd.NaT, pd.NA]
+    )
+    def test_insert_nat(self, null):
+        # GH 18295 (test missing)
+        idx = timedelta_range("1day", "3day")
+        result = idx.insert(1, null)
+        expected = TimedeltaIndex(["1day", pd.NaT, "2day", "3day"])
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_invalid_na(self):
+        idx = TimedeltaIndex(["4day", "1day", "2day"], name="idx")
+
+        item = np.datetime64("NaT")
+        result = idx.insert(0, item)
+
+        expected = Index([item] + list(idx), dtype=object, name="idx")
+        tm.assert_index_equal(result, expected)
+
+        # Also works if we pass a different dt64nat object
+        item2 = np.datetime64("NaT")
+        result = idx.insert(0, item2)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "item", [0, np.int64(0), np.float64(0), np.array(0), np.datetime64(456, "us")]
+    )
+    def test_insert_mismatched_types_raises(self, item):
+        # GH#33703 dont cast these to td64
+        tdi = TimedeltaIndex(["4day", "1day", "2day"], name="idx")
+
+        result = tdi.insert(1, item)
+
+        expected = Index(
+            [tdi[0], lib.item_from_zerodim(item)] + list(tdi[1:]),
+            dtype=object,
+            name="idx",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_castable_str(self):
+        idx = timedelta_range("1day", "3day")
+
+        result = idx.insert(0, "1 Day")
+
+        expected = TimedeltaIndex([idx[0]] + list(idx))
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_non_castable_str(self):
+        idx = timedelta_range("1day", "3day")
+
+        result = idx.insert(0, "foo")
+
+        expected = Index(["foo"] + list(idx), dtype=object)
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_empty(self):
+        # Corner case inserting with length zero doesn't raise IndexError
+        # GH#33573 for freq preservation
+        idx = timedelta_range("1 Day", periods=3)
+        td = idx[0]
+
+        result = idx[:0].insert(0, td)
+        assert result.freq == "D"
+
+        with pytest.raises(IndexError, match="loc must be an integer between"):
+            result = idx[:0].insert(1, td)
+
+        with pytest.raises(IndexError, match="loc must be an integer between"):
+            result = idx[:0].insert(-1, td)
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_repeat.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_repeat.py
new file mode 100644
index 0000000000000..2a9b58d1bf322
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_repeat.py
@@ -0,0 +1,34 @@
+import numpy as np
+
+from pandas import (
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestRepeat:
+    def test_repeat(self):
+        index = timedelta_range("1 days", periods=2, freq="D")
+        exp = TimedeltaIndex(["1 days", "1 days", "2 days", "2 days"])
+        for res in [index.repeat(2), np.repeat(index, 2)]:
+            tm.assert_index_equal(res, exp)
+            assert res.freq is None
+
+        index = TimedeltaIndex(["1 days", "NaT", "3 days"])
+        exp = TimedeltaIndex(
+            [
+                "1 days",
+                "1 days",
+                "1 days",
+                "NaT",
+                "NaT",
+                "NaT",
+                "3 days",
+                "3 days",
+                "3 days",
+            ]
+        )
+        for res in [index.repeat(3), np.repeat(index, 3)]:
+            tm.assert_index_equal(res, exp)
+            assert res.freq is None
diff --git a/pandas/pandas/tests/indexes/timedeltas/methods/test_shift.py b/pandas/pandas/tests/indexes/timedeltas/methods/test_shift.py
new file mode 100644
index 0000000000000..9bbf06dc51a0c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/methods/test_shift.py
@@ -0,0 +1,76 @@
+import pytest
+
+from pandas.errors import NullFrequencyError
+
+import pandas as pd
+from pandas import TimedeltaIndex
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexShift:
+    # -------------------------------------------------------------
+    # TimedeltaIndex.shift is used by __add__/__sub__
+
+    def test_tdi_shift_empty(self):
+        # GH#9903
+        idx = TimedeltaIndex([], name="xxx")
+        tm.assert_index_equal(idx.shift(0, freq="h"), idx)
+        tm.assert_index_equal(idx.shift(3, freq="h"), idx)
+
+    def test_tdi_shift_hours(self):
+        # GH#9903
+        idx = TimedeltaIndex(["5 hours", "6 hours", "9 hours"], name="xxx")
+        tm.assert_index_equal(idx.shift(0, freq="h"), idx)
+        exp = TimedeltaIndex(["8 hours", "9 hours", "12 hours"], name="xxx")
+        tm.assert_index_equal(idx.shift(3, freq="h"), exp)
+        exp = TimedeltaIndex(["2 hours", "3 hours", "6 hours"], name="xxx")
+        tm.assert_index_equal(idx.shift(-3, freq="h"), exp)
+
+    def test_tdi_shift_minutes(self):
+        # GH#9903
+        idx = TimedeltaIndex(["5 hours", "6 hours", "9 hours"], name="xxx")
+        tm.assert_index_equal(idx.shift(0, freq="min"), idx)
+        exp = TimedeltaIndex(["05:03:00", "06:03:00", "9:03:00"], name="xxx")
+        tm.assert_index_equal(idx.shift(3, freq="min"), exp)
+        exp = TimedeltaIndex(["04:57:00", "05:57:00", "8:57:00"], name="xxx")
+        tm.assert_index_equal(idx.shift(-3, freq="min"), exp)
+
+    def test_tdi_shift_int(self):
+        # GH#8083
+        tdi = pd.to_timedelta(range(5), unit="D")
+        trange = tdi._with_freq("infer") + pd.offsets.Hour(1)
+        result = trange.shift(1)
+        expected = TimedeltaIndex(
+            [
+                "1 days 01:00:00",
+                "2 days 01:00:00",
+                "3 days 01:00:00",
+                "4 days 01:00:00",
+                "5 days 01:00:00",
+            ],
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_tdi_shift_nonstandard_freq(self):
+        # GH#8083
+        tdi = pd.to_timedelta(range(5), unit="D")
+        trange = tdi._with_freq("infer") + pd.offsets.Hour(1)
+        result = trange.shift(3, freq="2D 1s")
+        expected = TimedeltaIndex(
+            [
+                "6 days 01:00:03",
+                "7 days 01:00:03",
+                "8 days 01:00:03",
+                "9 days 01:00:03",
+                "10 days 01:00:03",
+            ],
+            freq="D",
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_shift_no_freq(self):
+        # GH#19147
+        tdi = TimedeltaIndex(["1 days 01:00:00", "2 days 01:00:00"], freq=None)
+        with pytest.raises(NullFrequencyError, match="Cannot shift with no freq"):
+            tdi.shift(2)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_arithmetic.py b/pandas/pandas/tests/indexes/timedeltas/test_arithmetic.py
new file mode 100644
index 0000000000000..a431e10dc18ab
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_arithmetic.py
@@ -0,0 +1,51 @@
+# Arithmetic tests for TimedeltaIndex are generally about the result's `freq` attribute.
+# Other cases can be shared in tests.arithmetic.test_timedelta64
+import numpy as np
+
+from pandas import (
+    NaT,
+    Timedelta,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexArithmetic:
+    def test_arithmetic_zero_freq(self):
+        # GH#51575 don't get a .freq with freq.n = 0
+        tdi = timedelta_range(0, periods=100, freq="ns")
+        result = tdi / 2
+        assert result.freq is None
+        expected = tdi[:50].repeat(2)
+        tm.assert_index_equal(result, expected)
+
+        result2 = tdi // 2
+        assert result2.freq is None
+        expected2 = expected
+        tm.assert_index_equal(result2, expected2)
+
+        result3 = tdi * 0
+        assert result3.freq is None
+        expected3 = tdi[:1].repeat(100)
+        tm.assert_index_equal(result3, expected3)
+
+    def test_tdi_division(self, index_or_series):
+        # doc example
+
+        scalar = Timedelta(days=31)
+        td = index_or_series(
+            [scalar, scalar, scalar + Timedelta(minutes=5, seconds=3), NaT],
+            dtype="m8[ns]",
+        )
+
+        result = td / np.timedelta64(1, "D")
+        expected = index_or_series(
+            [31, 31, (31 * 86400 + 5 * 60 + 3) / 86400.0, np.nan]
+        )
+        tm.assert_equal(result, expected)
+
+        result = td / np.timedelta64(1, "s")
+        expected = index_or_series(
+            [31 * 86400, 31 * 86400, 31 * 86400 + 5 * 60 + 3, np.nan]
+        )
+        tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_constructors.py b/pandas/pandas/tests/indexes/timedeltas/test_constructors.py
new file mode 100644
index 0000000000000..ace0ab7990138
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_constructors.py
@@ -0,0 +1,266 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Timedelta,
+    TimedeltaIndex,
+    timedelta_range,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.core.arrays.timedeltas import TimedeltaArray
+
+
+class TestTimedeltaIndex:
+    def test_array_of_dt64_nat_raises(self):
+        # GH#39462
+        nat = np.datetime64("NaT", "ns")
+        arr = np.array([nat], dtype=object)
+
+        msg = "Invalid type for timedelta scalar"
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaIndex(arr)
+
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaArray._from_sequence(arr, dtype="m8[ns]")
+
+        with pytest.raises(TypeError, match=msg):
+            to_timedelta(arr)
+
+    def test_int64_nocopy(self):
+        # GH#23539 check that a copy isn't made when we pass int64 data
+        #  and copy=False
+        arr = np.arange(10, dtype=np.int64)
+        tdi = TimedeltaIndex(arr, copy=False)
+        assert tdi._data._ndarray.base is arr
+
+    def test_infer_from_tdi(self):
+        # GH#23539
+        # fast-path for inferring a frequency if the passed data already
+        #  has one
+        tdi = timedelta_range("1 second", periods=10**7, freq="1s")
+
+        result = TimedeltaIndex(tdi, freq="infer")
+        assert result.freq == tdi.freq
+
+        # check that inferred_freq was not called by checking that the
+        #  value has not been cached
+        assert "inferred_freq" not in getattr(result, "_cache", {})
+
+    def test_infer_from_tdi_mismatch(self):
+        # GH#23539
+        # fast-path for invalidating a frequency if the passed data already
+        #  has one and it does not match the `freq` input
+        tdi = timedelta_range("1 second", periods=100, freq="1s")
+
+        msg = (
+            "Inferred frequency .* from passed values does "
+            "not conform to passed frequency"
+        )
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaIndex(tdi, freq="D")
+
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaIndex(tdi._data, freq="D")
+
+    def test_dt64_data_invalid(self):
+        # GH#23539
+        # passing tz-aware DatetimeIndex raises, naive or ndarray[datetime64]
+        #  raise as of GH#29794
+        dti = pd.date_range("2016-01-01", periods=3)
+
+        msg = "cannot be converted to timedelta64"
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaIndex(dti.tz_localize("Europe/Brussels"))
+
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaIndex(dti)
+
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaIndex(np.asarray(dti))
+
+    def test_float64_ns_rounded(self):
+        # GH#23539 without specifying a unit, floats are regarded as nanos,
+        #  and fractional portions are truncated
+        tdi = TimedeltaIndex([2.3, 9.7])
+        expected = TimedeltaIndex([2, 9])
+        tm.assert_index_equal(tdi, expected)
+
+        # integral floats are non-lossy
+        tdi = TimedeltaIndex([2.0, 9.0])
+        expected = TimedeltaIndex([2, 9])
+        tm.assert_index_equal(tdi, expected)
+
+        # NaNs get converted to NaT
+        tdi = TimedeltaIndex([2.0, np.nan])
+        expected = TimedeltaIndex([Timedelta(nanoseconds=2), pd.NaT])
+        tm.assert_index_equal(tdi, expected)
+
+    def test_float64_unit_conversion(self):
+        # GH#23539
+        tdi = to_timedelta([1.5, 2.25], unit="D")
+        expected = TimedeltaIndex([Timedelta(days=1.5), Timedelta(days=2.25)])
+        tm.assert_index_equal(tdi, expected)
+
+    def test_construction_base_constructor(self):
+        arr = [Timedelta("1 days"), pd.NaT, Timedelta("3 days")]
+        tm.assert_index_equal(pd.Index(arr), TimedeltaIndex(arr))
+        tm.assert_index_equal(pd.Index(np.array(arr)), TimedeltaIndex(np.array(arr)))
+
+        arr = [np.nan, pd.NaT, Timedelta("1 days")]
+        tm.assert_index_equal(pd.Index(arr), TimedeltaIndex(arr))
+        tm.assert_index_equal(pd.Index(np.array(arr)), TimedeltaIndex(np.array(arr)))
+
+    def test_constructor(self):
+        expected = TimedeltaIndex(
+            [
+                "1 days",
+                "1 days 00:00:05",
+                "2 days",
+                "2 days 00:00:02",
+                "0 days 00:00:03",
+            ]
+        )
+        result = TimedeltaIndex(
+            [
+                "1 days",
+                "1 days, 00:00:05",
+                np.timedelta64(2, "D"),
+                timedelta(days=2, seconds=2),
+                pd.offsets.Second(3),
+            ]
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_constructor_iso(self):
+        # GH #21877
+        expected = timedelta_range("1s", periods=9, freq="s")
+        durations = [f"P0DT0H0M{i}S" for i in range(1, 10)]
+        result = to_timedelta(durations)
+        tm.assert_index_equal(result, expected)
+
+    def test_timedelta_range_fractional_period(self):
+        msg = "periods must be an integer"
+        with pytest.raises(TypeError, match=msg):
+            timedelta_range("1 days", periods=10.5)
+
+    def test_constructor_coverage(self):
+        msg = "periods must be an integer, got foo"
+        with pytest.raises(TypeError, match=msg):
+            timedelta_range(start="1 days", periods="foo", freq="D")
+
+        msg = (
+            r"TimedeltaIndex\(\.\.\.\) must be called with a collection of some kind, "
+            "'1 days' was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            TimedeltaIndex("1 days")
+
+        # generator expression
+        gen = (timedelta(i) for i in range(10))
+        result = TimedeltaIndex(gen)
+        expected = TimedeltaIndex([timedelta(i) for i in range(10)])
+        tm.assert_index_equal(result, expected)
+
+        # NumPy string array
+        strings = np.array(["1 days", "2 days", "3 days"])
+        result = TimedeltaIndex(strings)
+        expected = to_timedelta([1, 2, 3], unit="D")
+        tm.assert_index_equal(result, expected)
+
+        from_ints = TimedeltaIndex(expected.asi8)
+        tm.assert_index_equal(from_ints, expected)
+
+        # non-conforming freq
+        msg = (
+            "Inferred frequency None from passed values does not conform to "
+            "passed frequency D"
+        )
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaIndex(["1 days", "2 days", "4 days"], freq="D")
+
+        msg = (
+            "Of the four parameters: start, end, periods, and freq, exactly "
+            "three must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(periods=10, freq="D")
+
+    def test_constructor_name(self):
+        idx = timedelta_range(start="1 days", periods=1, freq="D", name="TEST")
+        assert idx.name == "TEST"
+
+        # GH10025
+        idx2 = TimedeltaIndex(idx, name="something else")
+        assert idx2.name == "something else"
+
+    def test_constructor_no_precision_raises(self):
+        # GH-24753, GH-24739
+
+        msg = "with no precision is not allowed"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaIndex(["2000"], dtype="timedelta64")
+
+        msg = "The 'timedelta64' dtype has no unit. Please pass in"
+        with pytest.raises(ValueError, match=msg):
+            pd.Index(["2000"], dtype="timedelta64")
+
+    def test_constructor_wrong_precision_raises(self):
+        msg = "Supported timedelta64 resolutions are 's', 'ms', 'us', 'ns'"
+        with pytest.raises(ValueError, match=msg):
+            TimedeltaIndex(["2000"], dtype="timedelta64[D]")
+
+        # "timedelta64[us]" was unsupported pre-2.0, but now this works.
+        tdi = TimedeltaIndex(["2000"], dtype="timedelta64[us]")
+        assert tdi.dtype == "m8[us]"
+
+    def test_explicit_none_freq(self):
+        # Explicitly passing freq=None is respected
+        tdi = timedelta_range(1, periods=5)
+        assert tdi.freq is not None
+
+        result = TimedeltaIndex(tdi, freq=None)
+        assert result.freq is None
+
+        result = TimedeltaIndex(tdi._data, freq=None)
+        assert result.freq is None
+
+    def test_from_categorical(self):
+        tdi = timedelta_range(1, periods=5)
+
+        cat = pd.Categorical(tdi)
+
+        result = TimedeltaIndex(cat)
+        tm.assert_index_equal(result, tdi)
+
+        ci = pd.CategoricalIndex(tdi)
+        result = TimedeltaIndex(ci)
+        tm.assert_index_equal(result, tdi)
+
+    @pytest.mark.parametrize(
+        "unit,unit_depr",
+        [
+            ("W", "w"),
+            ("D", "d"),
+            ("min", "MIN"),
+            ("s", "S"),
+            ("h", "H"),
+            ("ms", "MS"),
+            ("us", "US"),
+        ],
+    )
+    def test_unit_deprecated(self, unit, unit_depr):
+        # GH#52536, GH#59051
+        msg = f"'{unit_depr}' is deprecated and will be removed in a future version."
+
+        expected = TimedeltaIndex([f"1{unit}", f"2{unit}"])
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = TimedeltaIndex([f"1{unit_depr}", f"2{unit_depr}"])
+        tm.assert_index_equal(result, expected)
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            tdi = to_timedelta([1, 2], unit=unit_depr)
+        tm.assert_index_equal(tdi, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_delete.py b/pandas/pandas/tests/indexes/timedeltas/test_delete.py
new file mode 100644
index 0000000000000..f49af7cd0befd
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_delete.py
@@ -0,0 +1,71 @@
+from pandas import (
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexDelete:
+    def test_delete(self):
+        idx = timedelta_range(start="1 Days", periods=5, freq="D", name="idx")
+
+        # preserve freq
+        expected_0 = timedelta_range(start="2 Days", periods=4, freq="D", name="idx")
+        expected_4 = timedelta_range(start="1 Days", periods=4, freq="D", name="idx")
+
+        # reset freq to None
+        expected_1 = TimedeltaIndex(
+            ["1 day", "3 day", "4 day", "5 day"], freq=None, name="idx"
+        )
+
+        cases = {
+            0: expected_0,
+            -5: expected_0,
+            -1: expected_4,
+            4: expected_4,
+            1: expected_1,
+        }
+        for n, expected in cases.items():
+            result = idx.delete(n)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+        with tm.external_error_raised((IndexError, ValueError)):
+            # either depending on numpy version
+            idx.delete(5)
+
+    def test_delete_slice(self):
+        idx = timedelta_range(start="1 days", periods=10, freq="D", name="idx")
+
+        # preserve freq
+        expected_0_2 = timedelta_range(start="4 days", periods=7, freq="D", name="idx")
+        expected_7_9 = timedelta_range(start="1 days", periods=7, freq="D", name="idx")
+
+        # reset freq to None
+        expected_3_5 = TimedeltaIndex(
+            ["1 D", "2 D", "3 D", "7 D", "8 D", "9 D", "10D"], freq=None, name="idx"
+        )
+
+        cases = {
+            (0, 1, 2): expected_0_2,
+            (7, 8, 9): expected_7_9,
+            (3, 4, 5): expected_3_5,
+        }
+        for n, expected in cases.items():
+            result = idx.delete(n)
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+            result = idx.delete(slice(n[0], n[-1] + 1))
+            tm.assert_index_equal(result, expected)
+            assert result.name == expected.name
+            assert result.freq == expected.freq
+
+    def test_delete_doesnt_infer_freq(self):
+        # GH#30655 behavior matches DatetimeIndex
+
+        tdi = TimedeltaIndex(["1 Day", "2 Days", None, "3 Days", "4 Days"])
+        result = tdi.delete(2)
+        assert result.freq is None
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_formats.py b/pandas/pandas/tests/indexes/timedeltas/test_formats.py
new file mode 100644
index 0000000000000..607336060cbbc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_formats.py
@@ -0,0 +1,106 @@
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    TimedeltaIndex,
+)
+
+
+class TestTimedeltaIndexRendering:
+    def test_repr_round_days_non_nano(self):
+        # GH#55405
+        # we should get "1 days", not "1 days 00:00:00" with non-nano
+        tdi = TimedeltaIndex(["1 days"], freq="D").as_unit("s")
+        result = repr(tdi)
+        expected = "TimedeltaIndex(['1 days'], dtype='timedelta64[s]', freq='D')"
+        assert result == expected
+
+        result2 = repr(Series(tdi))
+        expected2 = "0   1 days\ndtype: timedelta64[s]"
+        assert result2 == expected2
+
+    @pytest.mark.parametrize("method", ["__repr__", "__str__"])
+    def test_representation(self, method):
+        idx1 = TimedeltaIndex([], freq="D")
+        idx2 = TimedeltaIndex(["1 days"], freq="D")
+        idx3 = TimedeltaIndex(["1 days", "2 days"], freq="D")
+        idx4 = TimedeltaIndex(["1 days", "2 days", "3 days"], freq="D")
+        idx5 = TimedeltaIndex(["1 days 00:00:01", "2 days", "3 days"])
+
+        exp1 = "TimedeltaIndex([], dtype='timedelta64[ns]', freq='D')"
+
+        exp2 = "TimedeltaIndex(['1 days'], dtype='timedelta64[ns]', freq='D')"
+
+        exp3 = "TimedeltaIndex(['1 days', '2 days'], dtype='timedelta64[ns]', freq='D')"
+
+        exp4 = (
+            "TimedeltaIndex(['1 days', '2 days', '3 days'], "
+            "dtype='timedelta64[ns]', freq='D')"
+        )
+
+        exp5 = (
+            "TimedeltaIndex(['1 days 00:00:01', '2 days 00:00:00', "
+            "'3 days 00:00:00'], dtype='timedelta64[ns]', freq=None)"
+        )
+
+        with pd.option_context("display.width", 300):
+            for idx, expected in zip(
+                [idx1, idx2, idx3, idx4, idx5], [exp1, exp2, exp3, exp4, exp5]
+            ):
+                result = getattr(idx, method)()
+                assert result == expected
+
+    # TODO: this is a Series.__repr__ test
+    def test_representation_to_series(self):
+        idx1 = TimedeltaIndex([], freq="D")
+        idx2 = TimedeltaIndex(["1 days"], freq="D")
+        idx3 = TimedeltaIndex(["1 days", "2 days"], freq="D")
+        idx4 = TimedeltaIndex(["1 days", "2 days", "3 days"], freq="D")
+        idx5 = TimedeltaIndex(["1 days 00:00:01", "2 days", "3 days"])
+
+        exp1 = """Series([], dtype: timedelta64[ns])"""
+
+        exp2 = "0   1 days\ndtype: timedelta64[ns]"
+
+        exp3 = "0   1 days\n1   2 days\ndtype: timedelta64[ns]"
+
+        exp4 = "0   1 days\n1   2 days\n2   3 days\ndtype: timedelta64[ns]"
+
+        exp5 = (
+            "0   1 days 00:00:01\n"
+            "1   2 days 00:00:00\n"
+            "2   3 days 00:00:00\n"
+            "dtype: timedelta64[ns]"
+        )
+
+        with pd.option_context("display.width", 300):
+            for idx, expected in zip(
+                [idx1, idx2, idx3, idx4, idx5], [exp1, exp2, exp3, exp4, exp5]
+            ):
+                result = repr(Series(idx))
+                assert result == expected
+
+    def test_summary(self):
+        # GH#9116
+        idx1 = TimedeltaIndex([], freq="D")
+        idx2 = TimedeltaIndex(["1 days"], freq="D")
+        idx3 = TimedeltaIndex(["1 days", "2 days"], freq="D")
+        idx4 = TimedeltaIndex(["1 days", "2 days", "3 days"], freq="D")
+        idx5 = TimedeltaIndex(["1 days 00:00:01", "2 days", "3 days"])
+
+        exp1 = "TimedeltaIndex: 0 entries\nFreq: D"
+
+        exp2 = "TimedeltaIndex: 1 entries, 1 days to 1 days\nFreq: D"
+
+        exp3 = "TimedeltaIndex: 2 entries, 1 days to 2 days\nFreq: D"
+
+        exp4 = "TimedeltaIndex: 3 entries, 1 days to 3 days\nFreq: D"
+
+        exp5 = "TimedeltaIndex: 3 entries, 1 days 00:00:01 to 3 days 00:00:00"
+
+        for idx, expected in zip(
+            [idx1, idx2, idx3, idx4, idx5], [exp1, exp2, exp3, exp4, exp5]
+        ):
+            result = idx._summary()
+            assert result == expected
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_freq_attr.py b/pandas/pandas/tests/indexes/timedeltas/test_freq_attr.py
new file mode 100644
index 0000000000000..1912c49d3000f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_freq_attr.py
@@ -0,0 +1,72 @@
+import pytest
+
+from pandas import TimedeltaIndex
+
+from pandas.tseries.offsets import (
+    DateOffset,
+    Day,
+    Hour,
+    MonthEnd,
+)
+
+
+class TestFreq:
+    @pytest.mark.parametrize("values", [["0 days", "2 days", "4 days"], []])
+    @pytest.mark.parametrize("freq", ["2D", Day(2), "48h", Hour(48)])
+    def test_freq_setter(self, values, freq):
+        # GH#20678
+        idx = TimedeltaIndex(values)
+
+        # can set to an offset, converting from string if necessary
+        idx._data.freq = freq
+        assert idx.freq == freq
+        assert isinstance(idx.freq, DateOffset)
+
+        # can reset to None
+        idx._data.freq = None
+        assert idx.freq is None
+
+    def test_with_freq_empty_requires_tick(self):
+        idx = TimedeltaIndex([])
+
+        off = MonthEnd(1)
+        msg = "TimedeltaArray/Index freq must be a Tick"
+        with pytest.raises(TypeError, match=msg):
+            idx._with_freq(off)
+        with pytest.raises(TypeError, match=msg):
+            idx._data._with_freq(off)
+
+    def test_freq_setter_errors(self):
+        # GH#20678
+        idx = TimedeltaIndex(["0 days", "2 days", "4 days"])
+
+        # setting with an incompatible freq
+        msg = (
+            "Inferred frequency 2D from passed values does not conform to "
+            "passed frequency 5D"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx._data.freq = "5D"
+
+        # setting with a non-fixed frequency
+        msg = r"<2 \* BusinessDays> is a non-fixed frequency"
+        with pytest.raises(ValueError, match=msg):
+            idx._data.freq = "2B"
+
+        # setting with non-freq string
+        with pytest.raises(ValueError, match="Invalid frequency"):
+            idx._data.freq = "foo"
+
+    def test_freq_view_safe(self):
+        # Setting the freq for one TimedeltaIndex shouldn't alter the freq
+        #  for another that views the same data
+
+        tdi = TimedeltaIndex(["0 days", "2 days", "4 days"], freq="2D")
+        tda = tdi._data
+
+        tdi2 = TimedeltaIndex(tda)._with_freq(None)
+        assert tdi2.freq is None
+
+        # Original was not altered
+        assert tdi.freq == "2D"
+        assert tda.freq == "2D"
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_indexing.py b/pandas/pandas/tests/indexes/timedeltas/test_indexing.py
new file mode 100644
index 0000000000000..426083cb6b67c
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_indexing.py
@@ -0,0 +1,350 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    NaT,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    notna,
+    offsets,
+    timedelta_range,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+
+class TestGetItem:
+    def test_getitem_slice_keeps_name(self):
+        # GH#4226, GH#59051
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            tdi = timedelta_range("1d", "5d", freq="h", name="timebucket")
+        assert tdi[1:].name == tdi.name
+
+    def test_getitem(self):
+        idx1 = timedelta_range("1 day", "31 day", freq="D", name="idx")
+
+        for idx in [idx1]:
+            result = idx[0]
+            assert result == Timedelta("1 day")
+
+            result = idx[0:5]
+            expected = timedelta_range("1 day", "5 day", freq="D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx[0:10:2]
+            expected = timedelta_range("1 day", "9 day", freq="2D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx[-20:-5:3]
+            expected = timedelta_range("12 day", "24 day", freq="3D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx[4::-1]
+            expected = TimedeltaIndex(
+                ["5 day", "4 day", "3 day", "2 day", "1 day"], freq="-1D", name="idx"
+            )
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+    @pytest.mark.parametrize(
+        "key",
+        [
+            Timestamp("1970-01-01"),
+            Timestamp("1970-01-02"),
+            datetime(1970, 1, 1),
+            Timestamp("1970-01-03").to_datetime64(),
+            # non-matching NA values
+            np.datetime64("NaT"),
+        ],
+    )
+    def test_timestamp_invalid_key(self, key):
+        # GH#20464
+        tdi = timedelta_range(0, periods=10)
+        with pytest.raises(KeyError, match=re.escape(repr(key))):
+            tdi.get_loc(key)
+
+
+class TestGetLoc:
+    def test_get_loc_key_unit_mismatch(self):
+        idx = to_timedelta(["0 days", "1 days", "2 days"])
+        key = idx[1].as_unit("ms")
+        loc = idx.get_loc(key)
+        assert loc == 1
+
+    def test_get_loc_key_unit_mismatch_not_castable(self):
+        tdi = to_timedelta(["0 days", "1 days", "2 days"]).astype("m8[s]")
+        assert tdi.dtype == "m8[s]"
+        key = tdi[0].as_unit("ns") + Timedelta(1)
+
+        with pytest.raises(KeyError, match=r"Timedelta\('0 days 00:00:00.000000001'\)"):
+            tdi.get_loc(key)
+
+        assert key not in tdi
+
+    def test_get_loc(self):
+        idx = to_timedelta(["0 days", "1 days", "2 days"])
+
+        # GH 16909
+        assert idx.get_loc(idx[1].to_timedelta64()) == 1
+
+        # GH 16896
+        assert idx.get_loc("0 days") == 0
+
+    def test_get_loc_nat(self):
+        tidx = TimedeltaIndex(["1 days 01:00:00", "NaT", "2 days 01:00:00"])
+
+        assert tidx.get_loc(NaT) == 1
+        assert tidx.get_loc(None) == 1
+        assert tidx.get_loc(float("nan")) == 1
+        assert tidx.get_loc(np.nan) == 1
+
+
+class TestGetIndexer:
+    def test_get_indexer(self):
+        idx = to_timedelta(["0 days", "1 days", "2 days"])
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(idx), np.array([0, 1, 2], dtype=np.intp)
+        )
+
+        target = to_timedelta(["-1 hour", "12 hours", "1 day 1 hour"])
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "pad"), np.array([-1, 0, 1], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "backfill"), np.array([0, 1, 2], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            idx.get_indexer(target, "nearest"), np.array([0, 1, 1], dtype=np.intp)
+        )
+
+        res = idx.get_indexer(target, "nearest", tolerance=Timedelta("1 hour"))
+        tm.assert_numpy_array_equal(res, np.array([0, -1, 1], dtype=np.intp))
+
+
+class TestWhere:
+    def test_where_doesnt_retain_freq(self):
+        tdi = timedelta_range("1 day", periods=3, freq="D", name="idx")
+        cond = [True, True, False]
+        expected = TimedeltaIndex([tdi[0], tdi[1], tdi[0]], freq=None, name="idx")
+
+        result = tdi.where(cond, tdi[::-1])
+        tm.assert_index_equal(result, expected)
+
+    def test_where_invalid_dtypes(self, fixed_now_ts):
+        tdi = timedelta_range("1 day", periods=3, freq="D", name="idx")
+
+        tail = tdi[2:].tolist()
+        i2 = Index([NaT, NaT] + tail)
+        mask = notna(i2)
+
+        expected = Index([NaT._value, NaT._value] + tail, dtype=object, name="idx")
+        assert isinstance(expected[0], int)
+        result = tdi.where(mask, i2.asi8)
+        tm.assert_index_equal(result, expected)
+
+        ts = i2 + fixed_now_ts
+        expected = Index([ts[0], ts[1]] + tail, dtype=object, name="idx")
+        result = tdi.where(mask, ts)
+        tm.assert_index_equal(result, expected)
+
+        per = (i2 + fixed_now_ts).to_period("D")
+        expected = Index([per[0], per[1]] + tail, dtype=object, name="idx")
+        result = tdi.where(mask, per)
+        tm.assert_index_equal(result, expected)
+
+        ts = fixed_now_ts
+        expected = Index([ts, ts] + tail, dtype=object, name="idx")
+        result = tdi.where(mask, ts)
+        tm.assert_index_equal(result, expected)
+
+    def test_where_mismatched_nat(self):
+        tdi = timedelta_range("1 day", periods=3, freq="D", name="idx")
+        cond = np.array([True, False, False])
+
+        dtnat = np.datetime64("NaT", "ns")
+        expected = Index([tdi[0], dtnat, dtnat], dtype=object, name="idx")
+        assert expected[2] is dtnat
+        result = tdi.where(cond, dtnat)
+        tm.assert_index_equal(result, expected)
+
+
+class TestTake:
+    def test_take(self):
+        # GH 10295
+        idx1 = timedelta_range("1 day", "31 day", freq="D", name="idx")
+
+        for idx in [idx1]:
+            result = idx.take([0])
+            assert result == Timedelta("1 day")
+
+            result = idx.take([-1])
+            assert result == Timedelta("31 day")
+
+            result = idx.take([0, 1, 2])
+            expected = timedelta_range("1 day", "3 day", freq="D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx.take([0, 2, 4])
+            expected = timedelta_range("1 day", "5 day", freq="2D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx.take([7, 4, 1])
+            expected = timedelta_range("8 day", "2 day", freq="-3D", name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq == expected.freq
+
+            result = idx.take([3, 2, 5])
+            expected = TimedeltaIndex(["4 day", "3 day", "6 day"], name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq is None
+
+            result = idx.take([-3, 2, 5])
+            expected = TimedeltaIndex(["29 day", "3 day", "6 day"], name="idx")
+            tm.assert_index_equal(result, expected)
+            assert result.freq is None
+
+    def test_take_invalid_kwargs(self):
+        idx = timedelta_range("1 day", "31 day", freq="D", name="idx")
+        indices = [1, 6, 5, 9, 10, 13, 15, 3]
+
+        msg = r"take\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            idx.take(indices, foo=2)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, out=indices)
+
+        msg = "the 'mode' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            idx.take(indices, mode="clip")
+
+    def test_take_equiv_getitem(self):
+        tds = ["1day 02:00:00", "1 day 04:00:00", "1 day 10:00:00"]
+        idx = timedelta_range(start="1D", end="2D", freq="h", name="idx")
+        expected = TimedeltaIndex(tds, freq=None, name="idx")
+
+        taken1 = idx.take([2, 4, 10])
+        taken2 = idx[[2, 4, 10]]
+
+        for taken in [taken1, taken2]:
+            tm.assert_index_equal(taken, expected)
+            assert isinstance(taken, TimedeltaIndex)
+            assert taken.freq is None
+            assert taken.name == expected.name
+
+    def test_take_fill_value(self):
+        # GH 12631
+        idx = TimedeltaIndex(["1 days", "2 days", "3 days"], name="xxx")
+        result = idx.take(np.array([1, 0, -1]))
+        expected = TimedeltaIndex(["2 days", "1 days", "3 days"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # fill_value
+        result = idx.take(np.array([1, 0, -1]), fill_value=True)
+        expected = TimedeltaIndex(["2 days", "1 days", "NaT"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        # allow_fill=False
+        result = idx.take(np.array([1, 0, -1]), allow_fill=False, fill_value=True)
+        expected = TimedeltaIndex(["2 days", "1 days", "3 days"], name="xxx")
+        tm.assert_index_equal(result, expected)
+
+        msg = (
+            "When allow_fill=True and fill_value is not None, all indices must be >= -1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -2]), fill_value=True)
+        with pytest.raises(ValueError, match=msg):
+            idx.take(np.array([1, 0, -5]), fill_value=True)
+
+        msg = "index -5 is out of bounds for (axis 0 with )?size 3"
+        with pytest.raises(IndexError, match=msg):
+            idx.take(np.array([1, -5]))
+
+
+class TestMaybeCastSliceBound:
+    @pytest.fixture(params=["increasing", "decreasing", None])
+    def monotonic(self, request):
+        return request.param
+
+    @pytest.fixture
+    def tdi(self, monotonic):
+        tdi = timedelta_range("1 Day", periods=10)
+        if monotonic == "decreasing":
+            tdi = tdi[::-1]
+        elif monotonic is None:
+            taker = np.arange(10, dtype=np.intp)
+            np.random.default_rng(2).shuffle(taker)
+            tdi = tdi.take(taker)
+        return tdi
+
+    def test_maybe_cast_slice_bound_invalid_str(self, tdi):
+        # test the low-level _maybe_cast_slice_bound and that we get the
+        #  expected exception+message all the way up the stack
+        msg = (
+            "cannot do slice indexing on TimedeltaIndex with these "
+            r"indexers \[foo\] of type str"
+        )
+        with pytest.raises(TypeError, match=msg):
+            tdi._maybe_cast_slice_bound("foo", side="left")
+        with pytest.raises(TypeError, match=msg):
+            tdi.get_slice_bound("foo", side="left")
+        with pytest.raises(TypeError, match=msg):
+            tdi.slice_locs("foo", None, None)
+
+    def test_slice_invalid_str_with_timedeltaindex(
+        self, tdi, frame_or_series, indexer_sl
+    ):
+        obj = frame_or_series(range(10), index=tdi)
+
+        msg = (
+            "cannot do slice indexing on TimedeltaIndex with these "
+            r"indexers \[foo\] of type str"
+        )
+        with pytest.raises(TypeError, match=msg):
+            indexer_sl(obj)["foo":]
+        with pytest.raises(TypeError, match=msg):
+            indexer_sl(obj)["foo":-1]
+        with pytest.raises(TypeError, match=msg):
+            indexer_sl(obj)[:"foo"]
+        with pytest.raises(TypeError, match=msg):
+            indexer_sl(obj)[tdi[0] : "foo"]
+
+
+class TestContains:
+    def test_contains_nonunique(self):
+        # GH#9512
+        for vals in (
+            [0, 1, 0],
+            [0, 0, -1],
+            [0, -1, -1],
+            ["00:01:00", "00:01:00", "00:02:00"],
+            ["00:01:00", "00:01:00", "00:00:01"],
+        ):
+            idx = TimedeltaIndex(vals)
+            assert idx[0] in idx
+
+    def test_contains(self):
+        # Checking for any NaT-like objects
+        # GH#13603, GH#59051
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            td = to_timedelta(range(5), unit="d") + offsets.Hour(1)
+        for v in [NaT, None, float("nan"), np.nan]:
+            assert v not in td
+
+        td = to_timedelta([NaT])
+        for v in [NaT, None, float("nan"), np.nan]:
+            assert v in td
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_join.py b/pandas/pandas/tests/indexes/timedeltas/test_join.py
new file mode 100644
index 0000000000000..cbd7a5de71b10
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_join.py
@@ -0,0 +1,47 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    Timedelta,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestJoin:
+    def test_append_join_nondatetimeindex(self):
+        rng = timedelta_range("1 days", periods=10)
+        idx = Index(["a", "b", "c", "d"])
+
+        result = rng.append(idx)
+        assert isinstance(result[0], Timedelta)
+
+        # it works
+        rng.join(idx, how="outer")
+
+    def test_join_self(self, join_type):
+        index = timedelta_range("1 day", periods=10)
+        joined = index.join(index, how=join_type)
+        tm.assert_index_equal(index, joined)
+
+    def test_does_not_convert_mixed_integer(self):
+        df = DataFrame(np.ones((5, 5)), columns=timedelta_range("1 day", periods=5))
+
+        cols = df.columns.join(df.index, how="outer")
+        joined = cols.join(df.columns)
+        assert cols.dtype == np.dtype("O")
+        assert cols.dtype == joined.dtype
+        tm.assert_index_equal(cols, joined)
+
+    def test_join_preserves_freq(self):
+        # GH#32157
+        tdi = timedelta_range("1 day", periods=10)
+        result = tdi[:5].join(tdi[5:], how="outer")
+        assert result.freq == tdi.freq
+        tm.assert_index_equal(result, tdi)
+
+        result = tdi[:5].join(tdi[6:], how="outer")
+        assert result.freq is None
+        expected = tdi.delete(5)
+        tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_ops.py b/pandas/pandas/tests/indexes/timedeltas/test_ops.py
new file mode 100644
index 0000000000000..f6013baf86edc
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_ops.py
@@ -0,0 +1,14 @@
+from pandas import (
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndexOps:
+    def test_infer_freq(self, freq_sample):
+        # GH#11018
+        idx = timedelta_range("1", freq=freq_sample, periods=10)
+        result = TimedeltaIndex(idx.asi8, freq="infer")
+        tm.assert_index_equal(idx, result)
+        assert result.freq == freq_sample
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_pickle.py b/pandas/pandas/tests/indexes/timedeltas/test_pickle.py
new file mode 100644
index 0000000000000..befe709728bdd
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_pickle.py
@@ -0,0 +1,11 @@
+from pandas import timedelta_range
+import pandas._testing as tm
+
+
+class TestPickle:
+    def test_pickle_after_set_freq(self):
+        tdi = timedelta_range("1 day", periods=4, freq="s")
+        tdi = tdi._with_freq(None)
+
+        res = tm.round_trip_pickle(tdi)
+        tm.assert_index_equal(res, tdi)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_scalar_compat.py b/pandas/pandas/tests/indexes/timedeltas/test_scalar_compat.py
new file mode 100644
index 0000000000000..9a00c556dc515
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_scalar_compat.py
@@ -0,0 +1,146 @@
+"""
+Tests for TimedeltaIndex methods behaving like their Timedelta counterparts
+"""
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import INVALID_FREQ_ERR_MSG
+
+from pandas import (
+    Index,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestVectorizedTimedelta:
+    def test_tdi_total_seconds(self):
+        # GH#10939
+        # test index
+        rng = timedelta_range("1 days, 10:11:12.100123456", periods=2, freq="s")
+        expt = [
+            1 * 86400 + 10 * 3600 + 11 * 60 + 12 + 100123456.0 / 1e9,
+            1 * 86400 + 10 * 3600 + 11 * 60 + 13 + 100123456.0 / 1e9,
+        ]
+        tm.assert_almost_equal(rng.total_seconds(), Index(expt))
+
+        # test Series
+        ser = Series(rng)
+        s_expt = Series(expt, index=[0, 1])
+        tm.assert_series_equal(ser.dt.total_seconds(), s_expt)
+
+        # with nat
+        ser[1] = np.nan
+        s_expt = Series(
+            [1 * 86400 + 10 * 3600 + 11 * 60 + 12 + 100123456.0 / 1e9, np.nan],
+            index=[0, 1],
+        )
+        tm.assert_series_equal(ser.dt.total_seconds(), s_expt)
+
+    def test_tdi_total_seconds_all_nat(self):
+        # with both nat
+        ser = Series([np.nan, np.nan], dtype="timedelta64[ns]")
+        result = ser.dt.total_seconds()
+        expected = Series([np.nan, np.nan])
+        tm.assert_series_equal(result, expected)
+
+    def test_tdi_round(self):
+        td = timedelta_range(start="16801 days", periods=5, freq="30Min")
+        elt = td[1]
+
+        expected_rng = TimedeltaIndex(
+            [
+                Timedelta("16801 days 00:00:00"),
+                Timedelta("16801 days 00:00:00"),
+                Timedelta("16801 days 01:00:00"),
+                Timedelta("16801 days 02:00:00"),
+                Timedelta("16801 days 02:00:00"),
+            ]
+        )
+        expected_elt = expected_rng[1]
+
+        tm.assert_index_equal(td.round(freq="h"), expected_rng)
+        assert elt.round(freq="h") == expected_elt
+
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            td.round(freq="foo")
+        with pytest.raises(ValueError, match=msg):
+            elt.round(freq="foo")
+
+        msg = "<MonthEnd> is a non-fixed frequency"
+        with pytest.raises(ValueError, match=msg):
+            td.round(freq="ME")
+        with pytest.raises(ValueError, match=msg):
+            elt.round(freq="ME")
+
+    @pytest.mark.parametrize(
+        "freq,msg",
+        [
+            ("YE", "<YearEnd: month=12> is a non-fixed frequency"),
+            ("ME", "<MonthEnd> is a non-fixed frequency"),
+            ("foobar", "Invalid frequency: foobar"),
+        ],
+    )
+    def test_tdi_round_invalid(self, freq, msg):
+        t1 = timedelta_range("1 days", periods=3, freq="1 min 2 s 3 us")
+
+        with pytest.raises(ValueError, match=msg):
+            t1.round(freq)
+        with pytest.raises(ValueError, match=msg):
+            # Same test for TimedeltaArray
+            t1._data.round(freq)
+
+    # TODO: de-duplicate with test_tdi_round
+    def test_round(self):
+        t1 = timedelta_range("1 days", periods=3, freq="1 min 2 s 3 us")
+        t2 = -1 * t1
+        t1a = timedelta_range("1 days", periods=3, freq="1 min 2 s")
+        t1c = TimedeltaIndex(np.array([1, 1, 1], "m8[D]")).as_unit("ns")
+
+        # note that negative times round DOWN! so don't give whole numbers
+        msg = "'d' is deprecated and will be removed in a future version."
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            for freq, s1, s2 in [
+                ("ns", t1, t2),
+                ("us", t1, t2),
+                (
+                    "ms",
+                    t1a,
+                    TimedeltaIndex(
+                        ["-1 days +00:00:00", "-2 days +23:58:58", "-2 days +23:57:56"]
+                    ),
+                ),
+                (
+                    "s",
+                    t1a,
+                    TimedeltaIndex(
+                        ["-1 days +00:00:00", "-2 days +23:58:58", "-2 days +23:57:56"]
+                    ),
+                ),
+                ("12min", t1c, TimedeltaIndex(["-1 days", "-1 days", "-1 days"])),
+                ("h", t1c, TimedeltaIndex(["-1 days", "-1 days", "-1 days"])),
+                ("d", t1c, -1 * t1c),
+            ]:
+                r1 = t1.round(freq)
+                r2 = t2.round(freq)
+
+            tm.assert_index_equal(r1, s1)
+            tm.assert_index_equal(r2, s2)
+
+    def test_components(self):
+        rng = timedelta_range("1 days, 10:11:12", periods=2, freq="s")
+        rng.components
+
+        # with nat
+        s = Series(rng)
+        s[1] = np.nan
+
+        result = s.dt.components
+        assert not result.iloc[0].isna().all()
+        assert result.iloc[1].isna().all()
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_searchsorted.py b/pandas/pandas/tests/indexes/timedeltas/test_searchsorted.py
new file mode 100644
index 0000000000000..710571ef38397
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_searchsorted.py
@@ -0,0 +1,28 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    TimedeltaIndex,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestSearchSorted:
+    def test_searchsorted_different_argument_classes(self, listlike_box):
+        idx = TimedeltaIndex(["1 day", "2 days", "3 days"])
+        result = idx.searchsorted(listlike_box(idx))
+        expected = np.arange(len(idx), dtype=result.dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = idx._data.searchsorted(listlike_box(idx))
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arg", [[1, 2], ["a", "b"], [Timestamp("2020-01-01", tz="Europe/London")] * 2]
+    )
+    def test_searchsorted_invalid_argument_dtype(self, arg):
+        idx = TimedeltaIndex(["1 day", "2 days", "3 days"])
+        msg = "value should be a 'Timedelta', 'NaT', or array of those. Got"
+        with pytest.raises(TypeError, match=msg):
+            idx.searchsorted(arg)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_setops.py b/pandas/pandas/tests/indexes/timedeltas/test_setops.py
new file mode 100644
index 0000000000000..ae88caf18fdae
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_setops.py
@@ -0,0 +1,257 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Index,
+    TimedeltaIndex,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import Hour
+
+
+class TestTimedeltaIndex:
+    def test_union(self):
+        i1 = timedelta_range("1day", periods=5)
+        i2 = timedelta_range("3day", periods=5)
+        result = i1.union(i2)
+        expected = timedelta_range("1day", periods=7)
+        tm.assert_index_equal(result, expected)
+
+        i1 = Index(np.arange(0, 20, 2, dtype=np.int64))
+        i2 = timedelta_range(start="1 day", periods=10, freq="D")
+        i1.union(i2)  # Works
+        i2.union(i1)  # Fails with "AttributeError: can't set attribute"
+
+    def test_union_sort_false(self):
+        tdi = timedelta_range("1day", periods=5)
+
+        left = tdi[3:]
+        right = tdi[:3]
+
+        # Check that we are testing the desired code path
+        assert left._can_fast_union(right)
+
+        result = left.union(right)
+        tm.assert_index_equal(result, tdi)
+
+        result = left.union(right, sort=False)
+        expected = TimedeltaIndex(["4 Days", "5 Days", "1 Days", "2 Day", "3 Days"])
+        tm.assert_index_equal(result, expected)
+
+    def test_union_coverage(self):
+        # GH#59051
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            idx = TimedeltaIndex(["3d", "1d", "2d"])
+        ordered = TimedeltaIndex(idx.sort_values(), freq="infer")
+        result = ordered.union(idx)
+        tm.assert_index_equal(result, ordered)
+
+        result = ordered[:0].union(ordered)
+        tm.assert_index_equal(result, ordered)
+        assert result.freq == ordered.freq
+
+    def test_union_bug_1730(self):
+        rng_a = timedelta_range("1 day", periods=4, freq="3h")
+        rng_b = timedelta_range("1 day", periods=4, freq="4h")
+
+        result = rng_a.union(rng_b)
+        exp = TimedeltaIndex(sorted(set(rng_a) | set(rng_b)))
+        tm.assert_index_equal(result, exp)
+
+    def test_union_bug_1745(self):
+        left = TimedeltaIndex(["1 day 15:19:49.695000"])
+        right = TimedeltaIndex(
+            ["2 day 13:04:21.322000", "1 day 15:27:24.873000", "1 day 15:31:05.350000"]
+        )
+
+        result = left.union(right)
+        exp = TimedeltaIndex(sorted(set(left) | set(right)))
+        tm.assert_index_equal(result, exp)
+
+    def test_union_bug_4564(self):
+        left = timedelta_range("1 day", "30D")
+        right = left + pd.offsets.Minute(15)
+
+        result = left.union(right)
+        exp = TimedeltaIndex(sorted(set(left) | set(right)))
+        tm.assert_index_equal(result, exp)
+
+    def test_union_freq_infer(self):
+        # When taking the union of two TimedeltaIndexes, we infer
+        #  a freq even if the arguments don't have freq.  This matches
+        #  DatetimeIndex behavior.
+        tdi = timedelta_range("1 Day", periods=5)
+        left = tdi[[0, 1, 3, 4]]
+        right = tdi[[2, 3, 1]]
+
+        assert left.freq is None
+        assert right.freq is None
+
+        result = left.union(right)
+        tm.assert_index_equal(result, tdi)
+        assert result.freq == "D"
+
+    def test_intersection_bug_1708(self):
+        index_1 = timedelta_range("1 day", periods=4, freq="h")
+        index_2 = index_1 + pd.offsets.Hour(5)
+
+        result = index_1.intersection(index_2)
+        assert len(result) == 0
+
+        index_1 = timedelta_range("1 day", periods=4, freq="h")
+        index_2 = index_1 + pd.offsets.Hour(1)
+
+        result = index_1.intersection(index_2)
+        expected = timedelta_range("1 day 01:00:00", periods=3, freq="h")
+        tm.assert_index_equal(result, expected)
+        assert result.freq == expected.freq
+
+    def test_intersection_equal(self, sort):
+        # GH 24471 Test intersection outcome given the sort keyword
+        # for equal indices intersection should return the original index
+        first = timedelta_range("1 day", periods=4, freq="h")
+        second = timedelta_range("1 day", periods=4, freq="h")
+        intersect = first.intersection(second, sort=sort)
+        if sort is None:
+            tm.assert_index_equal(intersect, second.sort_values())
+        tm.assert_index_equal(intersect, second)
+
+        # Corner cases
+        inter = first.intersection(first, sort=sort)
+        assert inter is first
+
+    @pytest.mark.parametrize("period_1, period_2", [(0, 4), (4, 0)])
+    def test_intersection_zero_length(self, period_1, period_2, sort):
+        # GH 24471 test for non overlap the intersection should be zero length
+        index_1 = timedelta_range("1 day", periods=period_1, freq="h")
+        index_2 = timedelta_range("1 day", periods=period_2, freq="h")
+        expected = timedelta_range("1 day", periods=0, freq="h")
+        result = index_1.intersection(index_2, sort=sort)
+        tm.assert_index_equal(result, expected)
+
+    def test_zero_length_input_index(self, sort):
+        # GH 24966 test for 0-len intersections are copied
+        index_1 = timedelta_range("1 day", periods=0, freq="h")
+        index_2 = timedelta_range("1 day", periods=3, freq="h")
+        result = index_1.intersection(index_2, sort=sort)
+        assert index_1 is not result
+        assert index_2 is not result
+        tm.assert_copy(result, index_1)
+
+    @pytest.mark.parametrize(
+        "rng, expected",
+        # if target has the same name, it is preserved
+        [
+            (
+                timedelta_range("1 day", periods=5, freq="h", name="idx"),
+                timedelta_range("1 day", periods=4, freq="h", name="idx"),
+            ),
+            # if target name is different, it will be reset
+            (
+                timedelta_range("1 day", periods=5, freq="h", name="other"),
+                timedelta_range("1 day", periods=4, freq="h", name=None),
+            ),
+            # if no overlap exists return empty index
+            (
+                timedelta_range("1 day", periods=10, freq="h", name="idx")[5:],
+                TimedeltaIndex([], freq="h", name="idx"),
+            ),
+        ],
+    )
+    def test_intersection(self, rng, expected, sort):
+        # GH 4690 (with tz)
+        base = timedelta_range("1 day", periods=4, freq="h", name="idx")
+        result = base.intersection(rng, sort=sort)
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+        assert result.freq == expected.freq
+
+    @pytest.mark.parametrize(
+        "rng, expected",
+        # part intersection works
+        [
+            (
+                TimedeltaIndex(["5 hour", "2 hour", "4 hour", "9 hour"], name="idx"),
+                TimedeltaIndex(["2 hour", "4 hour"], name="idx"),
+            ),
+            # reordered part intersection
+            (
+                TimedeltaIndex(["2 hour", "5 hour", "5 hour", "1 hour"], name="other"),
+                TimedeltaIndex(["1 hour", "2 hour"], name=None),
+            ),
+            # reversed index
+            (
+                TimedeltaIndex(["1 hour", "2 hour", "4 hour", "3 hour"], name="idx")[
+                    ::-1
+                ],
+                TimedeltaIndex(["1 hour", "2 hour", "4 hour", "3 hour"], name="idx"),
+            ),
+        ],
+    )
+    def test_intersection_non_monotonic(self, rng, expected, sort):
+        # 24471 non-monotonic
+        base = TimedeltaIndex(["1 hour", "2 hour", "4 hour", "3 hour"], name="idx")
+        result = base.intersection(rng, sort=sort)
+        if sort is None:
+            expected = expected.sort_values()
+        tm.assert_index_equal(result, expected)
+        assert result.name == expected.name
+
+        # if reversed order, frequency is still the same
+        if all(base == rng[::-1]) and sort is None:
+            assert isinstance(result.freq, Hour)
+        else:
+            assert result.freq is None
+
+
+class TestTimedeltaIndexDifference:
+    def test_difference_freq(self, sort):
+        # GH14323: Difference of TimedeltaIndex should not preserve frequency
+
+        index = timedelta_range("0 days", "5 days", freq="D")
+
+        other = timedelta_range("1 days", "4 days", freq="D")
+        expected = TimedeltaIndex(["0 days", "5 days"], freq=None)
+        idx_diff = index.difference(other, sort)
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+        # preserve frequency when the difference is a contiguous
+        # subset of the original range
+        other = timedelta_range("2 days", "5 days", freq="D")
+        idx_diff = index.difference(other, sort)
+        expected = TimedeltaIndex(["0 days", "1 days"], freq="D")
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+    def test_difference_sort(self, sort):
+        index = TimedeltaIndex(
+            ["5 days", "3 days", "2 days", "4 days", "1 days", "0 days"]
+        )
+
+        other = timedelta_range("1 days", "4 days", freq="D")
+        idx_diff = index.difference(other, sort)
+
+        expected = TimedeltaIndex(["5 days", "0 days"], freq=None)
+
+        if sort is None:
+            expected = expected.sort_values()
+
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
+
+        other = timedelta_range("2 days", "5 days", freq="D")
+        idx_diff = index.difference(other, sort)
+        expected = TimedeltaIndex(["1 days", "0 days"], freq=None)
+
+        if sort is None:
+            expected = expected.sort_values()
+
+        tm.assert_index_equal(idx_diff, expected)
+        tm.assert_attr_equal("freq", idx_diff, expected)
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_timedelta.py b/pandas/pandas/tests/indexes/timedeltas/test_timedelta.py
new file mode 100644
index 0000000000000..2066be8976e7f
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_timedelta.py
@@ -0,0 +1,61 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    Series,
+    Timedelta,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaIndex:
+    def test_misc_coverage(self):
+        rng = timedelta_range("1 day", periods=5)
+        result = rng.groupby(rng.days)
+        assert isinstance(next(iter(result.values()))[0], Timedelta)
+
+    def test_map(self):
+        # test_map_dictlike generally tests
+
+        rng = timedelta_range("1 day", periods=10)
+
+        f = lambda x: x.days
+        result = rng.map(f)
+        exp = Index([f(x) for x in rng], dtype=np.int64)
+        tm.assert_index_equal(result, exp)
+
+    def test_fields(self):
+        rng = timedelta_range("1 days, 10:11:12.100123456", periods=2, freq="s")
+        tm.assert_index_equal(rng.days, Index([1, 1], dtype=np.int64))
+        tm.assert_index_equal(
+            rng.seconds,
+            Index([10 * 3600 + 11 * 60 + 12, 10 * 3600 + 11 * 60 + 13], dtype=np.int32),
+        )
+        tm.assert_index_equal(
+            rng.microseconds,
+            Index([100 * 1000 + 123, 100 * 1000 + 123], dtype=np.int32),
+        )
+        tm.assert_index_equal(rng.nanoseconds, Index([456, 456], dtype=np.int32))
+
+        msg = "'TimedeltaIndex' object has no attribute '{}'"
+        with pytest.raises(AttributeError, match=msg.format("hours")):
+            rng.hours
+        with pytest.raises(AttributeError, match=msg.format("minutes")):
+            rng.minutes
+        with pytest.raises(AttributeError, match=msg.format("milliseconds")):
+            rng.milliseconds
+
+        # with nat
+        s = Series(rng)
+        s[1] = np.nan
+
+        tm.assert_series_equal(s.dt.days, Series([1, np.nan], index=range(2)))
+        tm.assert_series_equal(
+            s.dt.seconds, Series([10 * 3600 + 11 * 60 + 12, np.nan], index=range(2))
+        )
+
+        # preserve name (GH15589)
+        rng.name = "name"
+        assert rng.days.name == "name"
diff --git a/pandas/pandas/tests/indexes/timedeltas/test_timedelta_range.py b/pandas/pandas/tests/indexes/timedeltas/test_timedelta_range.py
new file mode 100644
index 0000000000000..6f3d29fb4240a
--- /dev/null
+++ b/pandas/pandas/tests/indexes/timedeltas/test_timedelta_range.py
@@ -0,0 +1,154 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Timedelta,
+    timedelta_range,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import (
+    Day,
+    Second,
+)
+
+
+class TestTimedeltas:
+    def test_timedelta_range_unit(self):
+        # GH#49824
+        tdi = timedelta_range("0 Days", periods=10, freq="100000D", unit="s")
+        exp_arr = (np.arange(10, dtype="i8") * 100_000).view("m8[D]").astype("m8[s]")
+        tm.assert_numpy_array_equal(tdi.to_numpy(), exp_arr)
+
+    def test_timedelta_range(self):
+        expected = to_timedelta(np.arange(5), unit="D")
+        result = timedelta_range("0 days", periods=5, freq="D")
+        tm.assert_index_equal(result, expected)
+
+        expected = to_timedelta(np.arange(11), unit="D")
+        result = timedelta_range("0 days", "10 days", freq="D")
+        tm.assert_index_equal(result, expected)
+
+        expected = to_timedelta(np.arange(5), unit="D") + Second(2) + Day()
+        result = timedelta_range("1 days, 00:00:02", "5 days, 00:00:02", freq="D")
+        tm.assert_index_equal(result, expected)
+
+        expected = to_timedelta([1, 3, 5, 7, 9], unit="D") + Second(2)
+        result = timedelta_range("1 days, 00:00:02", periods=5, freq="2D")
+        tm.assert_index_equal(result, expected)
+
+        expected = to_timedelta(np.arange(50), unit="min") * 30
+        result = timedelta_range("0 days", freq="30min", periods=50)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("depr_unit, unit", [("H", "hour"), ("S", "second")])
+    def test_timedelta_units_H_S_deprecated(self, depr_unit, unit):
+        # GH#52536
+        depr_msg = (
+            f"'{depr_unit}' is deprecated and will be removed in a future version."
+        )
+        expected = to_timedelta(np.arange(5), unit=unit)
+        with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+            result = to_timedelta(np.arange(5), unit=depr_unit)
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("unit", ["T", "t", "L", "l", "U", "u", "N", "n"])
+    def test_timedelta_unit_T_L_U_N_raises(self, unit):
+        msg = f"invalid unit abbreviation: {unit}"
+
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(np.arange(5), unit=unit)
+
+    @pytest.mark.parametrize(
+        "periods, freq", [(3, "2D"), (5, "D"), (6, "19h12min"), (7, "16h"), (9, "12h")]
+    )
+    def test_linspace_behavior(self, periods, freq):
+        # GH 20976
+        result = timedelta_range(start="0 days", end="4 days", periods=periods)
+        expected = timedelta_range(start="0 days", end="4 days", freq=freq)
+        tm.assert_index_equal(result, expected)
+
+    def test_timedelta_range_H_raises(self):
+        # GH#52536
+        msg = "Invalid frequency: H"
+
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(start="0 days", end="4 days", freq="19H12min")
+
+    def test_timedelta_range_T_raises(self):
+        msg = "Invalid frequency: T"
+
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(start="0 days", end="4 days", freq="19h12T")
+
+    def test_errors(self):
+        # not enough params
+        msg = (
+            "Of the four parameters: start, end, periods, and freq, "
+            "exactly three must be specified"
+        )
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(start="0 days")
+
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(end="5 days")
+
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(periods=2)
+
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range()
+
+        # too many params
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(start="0 days", end="5 days", periods=10, freq="h")
+
+    @pytest.mark.parametrize(
+        "start, end, freq, expected_periods",
+        [
+            ("1D", "10D", "2D", (10 - 1) // 2 + 1),
+            ("2D", "30D", "3D", (30 - 2) // 3 + 1),
+            ("2s", "50s", "5s", (50 - 2) // 5 + 1),
+            # tests that worked before GH 33498:
+            ("4D", "16D", "3D", (16 - 4) // 3 + 1),
+            ("8D", "16D", "40s", (16 * 3600 * 24 - 8 * 3600 * 24) // 40 + 1),
+        ],
+    )
+    def test_timedelta_range_freq_divide_end(self, start, end, freq, expected_periods):
+        # GH 33498 only the cases where `(end % freq) == 0` used to fail
+        res = timedelta_range(start=start, end=end, freq=freq)
+        assert Timedelta(start) == res[0]
+        assert Timedelta(end) >= res[-1]
+        assert len(res) == expected_periods
+
+    def test_timedelta_range_infer_freq(self):
+        # https://github.com/pandas-dev/pandas/issues/35897
+        result = timedelta_range("0s", "1s", periods=31)
+        assert result.freq is None
+
+    @pytest.mark.parametrize(
+        "freq_depr, start, end",
+        [
+            (
+                "3.5l",
+                "05:03:01",
+                "05:03:10",
+            ),
+            (
+                "2.5T",
+                "5 hours",
+                "5 hours 8 minutes",
+            ),
+            (
+                "3.5S",
+                "05:03:01",
+                "05:03:10",
+            ),
+        ],
+    )
+    def test_timedelta_range_removed_freq(self, freq_depr, start, end):
+        # GH#59143
+        msg = f"Invalid frequency: {freq_depr}"
+        with pytest.raises(ValueError, match=msg):
+            timedelta_range(start=start, end=end, freq=freq_depr)
diff --git a/pandas/pandas/tests/indexing/__init__.py b/pandas/pandas/tests/indexing/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexing/common.py b/pandas/pandas/tests/indexing/common.py
new file mode 100644
index 0000000000000..a33fb1e6979ec
--- /dev/null
+++ b/pandas/pandas/tests/indexing/common.py
@@ -0,0 +1,41 @@
+"""common utilities"""
+
+from __future__ import annotations
+
+from typing import (
+    Any,
+    Literal,
+)
+
+
+def _mklbl(prefix: str, n: int):
+    return [f"{prefix}{i}" for i in range(n)]
+
+
+def check_indexing_smoketest_or_raises(
+    obj,
+    method: Literal["iloc", "loc"],
+    key: Any,
+    axes: Literal[0, 1] | None = None,
+    fails=None,
+) -> None:
+    if axes is None:
+        axes_list = [0, 1]
+    else:
+        assert axes in [0, 1]
+        axes_list = [axes]
+
+    for ax in axes_list:
+        if ax < obj.ndim:
+            # create a tuple accessor
+            new_axes = [slice(None)] * obj.ndim
+            new_axes[ax] = key
+            axified = tuple(new_axes)
+            try:
+                getattr(obj, method).__getitem__(axified)
+            except (IndexError, TypeError, KeyError) as detail:
+                # if we are in fails, the ok, otherwise raise it
+                if fails is not None:
+                    if isinstance(detail, fails):
+                        return
+                raise
diff --git a/pandas/pandas/tests/indexing/interval/__init__.py b/pandas/pandas/tests/indexing/interval/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexing/interval/test_interval.py b/pandas/pandas/tests/indexing/interval/test_interval.py
new file mode 100644
index 0000000000000..6bcebefa6c696
--- /dev/null
+++ b/pandas/pandas/tests/indexing/interval/test_interval.py
@@ -0,0 +1,224 @@
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+from pandas.compat import WASM
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    IntervalIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestIntervalIndex:
+    @pytest.fixture
+    def series_with_interval_index(self):
+        return Series(np.arange(5), IntervalIndex.from_breaks(np.arange(6)))
+
+    def test_getitem_with_scalar(self, series_with_interval_index, indexer_sl):
+        ser = series_with_interval_index.copy()
+
+        expected = ser.iloc[:3]
+        tm.assert_series_equal(expected, indexer_sl(ser)[:3])
+        tm.assert_series_equal(expected, indexer_sl(ser)[:2.5])
+        tm.assert_series_equal(expected, indexer_sl(ser)[0.1:2.5])
+        if indexer_sl is tm.loc:
+            tm.assert_series_equal(expected, ser.loc[-1:3])
+
+        expected = ser.iloc[1:4]
+        tm.assert_series_equal(expected, indexer_sl(ser)[[1.5, 2.5, 3.5]])
+        tm.assert_series_equal(expected, indexer_sl(ser)[[2, 3, 4]])
+        tm.assert_series_equal(expected, indexer_sl(ser)[[1.5, 3, 4]])
+
+        expected = ser.iloc[2:5]
+        tm.assert_series_equal(expected, indexer_sl(ser)[ser >= 2])
+
+    @pytest.mark.parametrize("direction", ["increasing", "decreasing"])
+    def test_getitem_nonoverlapping_monotonic(self, direction, closed, indexer_sl):
+        tpls = [(0, 1), (2, 3), (4, 5)]
+        if direction == "decreasing":
+            tpls = tpls[::-1]
+
+        idx = IntervalIndex.from_tuples(tpls, closed=closed)
+        ser = Series(list("abc"), idx)
+
+        for key, expected in zip(idx.left, ser):
+            if idx.closed_left:
+                assert indexer_sl(ser)[key] == expected
+            else:
+                with pytest.raises(KeyError, match=str(key)):
+                    indexer_sl(ser)[key]
+
+        for key, expected in zip(idx.right, ser):
+            if idx.closed_right:
+                assert indexer_sl(ser)[key] == expected
+            else:
+                with pytest.raises(KeyError, match=str(key)):
+                    indexer_sl(ser)[key]
+
+        for key, expected in zip(idx.mid, ser):
+            assert indexer_sl(ser)[key] == expected
+
+    def test_getitem_non_matching(self, series_with_interval_index, indexer_sl):
+        ser = series_with_interval_index.copy()
+
+        # this is a departure from our current
+        # indexing scheme, but simpler
+        with pytest.raises(KeyError, match=r"\[-1\] not in index"):
+            indexer_sl(ser)[[-1, 3, 4, 5]]
+
+        with pytest.raises(KeyError, match=r"\[-1\] not in index"):
+            indexer_sl(ser)[[-1, 3]]
+
+    def test_loc_getitem_large_series(self, monkeypatch):
+        size_cutoff = 20
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+            ser = Series(
+                np.arange(size_cutoff),
+                index=IntervalIndex.from_breaks(np.arange(size_cutoff + 1)),
+            )
+
+            result1 = ser.loc[:8]
+            result2 = ser.loc[0:8]
+            result3 = ser.loc[0:8:1]
+        tm.assert_series_equal(result1, result2)
+        tm.assert_series_equal(result1, result3)
+
+    def test_loc_getitem_frame(self):
+        # CategoricalIndex with IntervalIndex categories
+        df = DataFrame({"A": range(10)})
+        ser = pd.cut(df.A, 5)
+        df["B"] = ser
+        df = df.set_index("B")
+
+        result = df.loc[4]
+        expected = df.iloc[4:6]
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(KeyError, match="10"):
+            df.loc[10]
+
+        # single list-like
+        result = df.loc[[4]]
+        expected = df.iloc[4:6]
+        tm.assert_frame_equal(result, expected)
+
+        # non-unique
+        result = df.loc[[4, 5]]
+        expected = df.take([4, 5, 4, 5])
+        tm.assert_frame_equal(result, expected)
+
+        msg = (
+            r"None of \[Index\(\[10\], dtype='object', name='B'\)\] "
+            r"are in the \[index\]"
+        )
+        with pytest.raises(KeyError, match=msg):
+            df.loc[[10]]
+
+        # partial missing
+        with pytest.raises(KeyError, match=r"\[10\] not in index"):
+            df.loc[[10, 4]]
+
+    def test_getitem_interval_with_nans(self, frame_or_series, indexer_sl):
+        # GH#41831
+
+        index = IntervalIndex([np.nan, np.nan])
+        key = index[:-1]
+
+        obj = frame_or_series(range(2), index=index)
+        if frame_or_series is DataFrame and indexer_sl is tm.setitem:
+            obj = obj.T
+
+        result = indexer_sl(obj)[key]
+        expected = obj
+
+        tm.assert_equal(result, expected)
+
+    def test_setitem_interval_with_slice(self):
+        # GH#54722
+        ii = IntervalIndex.from_breaks(range(4, 15))
+        ser = Series(range(10), index=ii)
+
+        orig = ser.copy()
+
+        # This should be a no-op (used to raise)
+        ser.loc[1:3] = 20
+        tm.assert_series_equal(ser, orig)
+
+        ser.loc[6:8] = 19
+        orig.iloc[1:4] = 19
+        tm.assert_series_equal(ser, orig)
+
+        ser2 = Series(range(5), index=ii[::2])
+        orig2 = ser2.copy()
+
+        # this used to raise
+        ser2.loc[6:8] = 22  # <- raises on main, sets on branch
+        orig2.iloc[1] = 22
+        tm.assert_series_equal(ser2, orig2)
+
+        ser2.loc[5:7] = 21
+        orig2.iloc[:2] = 21
+        tm.assert_series_equal(ser2, orig2)
+
+
+class TestIntervalIndexInsideMultiIndex:
+    def test_mi_intervalindex_slicing_with_scalar(self):
+        # GH#27456
+        ii = IntervalIndex.from_arrays(
+            [0, 1, 10, 11, 0, 1, 10, 11], [1, 2, 11, 12, 1, 2, 11, 12], name="MP"
+        )
+        idx = pd.MultiIndex.from_arrays(
+            [
+                pd.Index(["FC", "FC", "FC", "FC", "OWNER", "OWNER", "OWNER", "OWNER"]),
+                pd.Index(
+                    ["RID1", "RID1", "RID2", "RID2", "RID1", "RID1", "RID2", "RID2"]
+                ),
+                ii,
+            ]
+        )
+
+        idx.names = ["Item", "RID", "MP"]
+        df = DataFrame({"value": [1, 2, 3, 4, 5, 6, 7, 8]})
+        df.index = idx
+
+        query_df = DataFrame(
+            {
+                "Item": ["FC", "OWNER", "FC", "OWNER", "OWNER"],
+                "RID": ["RID1", "RID1", "RID1", "RID2", "RID2"],
+                "MP": [0.2, 1.5, 1.6, 11.1, 10.9],
+            }
+        )
+
+        query_df = query_df.sort_index()
+
+        idx = pd.MultiIndex.from_arrays([query_df.Item, query_df.RID, query_df.MP])
+        query_df.index = idx
+        result = df.value.loc[query_df.index]
+
+        # the IntervalIndex level is indexed with floats, which map to
+        #  the intervals containing them.  Matching the behavior we would get
+        #  with _only_ an IntervalIndex, we get an IntervalIndex level back.
+        sliced_level = ii.take([0, 1, 1, 3, 2])
+        expected_index = pd.MultiIndex.from_arrays(
+            [idx.get_level_values(0), idx.get_level_values(1), sliced_level]
+        )
+        expected = Series([1, 6, 2, 8, 7], index=expected_index, name="value")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.xfail(WASM, reason="GH 23440")
+    @pytest.mark.parametrize("base", [101, 1010])
+    def test_reindex_behavior_with_interval_index(self, base):
+        # GH 51826
+
+        ser = Series(
+            range(base),
+            index=IntervalIndex.from_arrays(range(base), range(1, base + 1)),
+        )
+        expected_result = Series([np.nan, 0], index=[np.nan, 1.0], dtype=float)
+        result = ser.reindex(index=[np.nan, 1.0])
+        tm.assert_series_equal(result, expected_result)
diff --git a/pandas/pandas/tests/indexing/interval/test_interval_new.py b/pandas/pandas/tests/indexing/interval/test_interval_new.py
new file mode 100644
index 0000000000000..051dc7b98f2aa
--- /dev/null
+++ b/pandas/pandas/tests/indexing/interval/test_interval_new.py
@@ -0,0 +1,235 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+
+from pandas import (
+    Index,
+    Interval,
+    IntervalIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestIntervalIndex:
+    @pytest.fixture
+    def series_with_interval_index(self):
+        """
+        Fixture providing a Series with an IntervalIndex.
+        """
+        return Series(np.arange(5), IntervalIndex.from_breaks(np.arange(6)))
+
+    def test_loc_with_interval(self, series_with_interval_index, indexer_sl):
+        # loc with single label / list of labels:
+        #   - Intervals: only exact matches
+        #   - scalars: those that contain it
+
+        ser = series_with_interval_index.copy()
+
+        expected = 0
+        result = indexer_sl(ser)[Interval(0, 1)]
+        assert result == expected
+
+        expected = ser.iloc[3:5]
+        result = indexer_sl(ser)[[Interval(3, 4), Interval(4, 5)]]
+        tm.assert_series_equal(expected, result)
+
+        # missing or not exact
+        with pytest.raises(KeyError, match=re.escape("Interval(3, 5, closed='left')")):
+            indexer_sl(ser)[Interval(3, 5, closed="left")]
+
+        with pytest.raises(KeyError, match=re.escape("Interval(3, 5, closed='right')")):
+            indexer_sl(ser)[Interval(3, 5)]
+
+        with pytest.raises(
+            KeyError, match=re.escape("Interval(-2, 0, closed='right')")
+        ):
+            indexer_sl(ser)[Interval(-2, 0)]
+
+        with pytest.raises(KeyError, match=re.escape("Interval(5, 6, closed='right')")):
+            indexer_sl(ser)[Interval(5, 6)]
+
+    def test_loc_with_scalar(self, series_with_interval_index, indexer_sl):
+        # loc with single label / list of labels:
+        #   - Intervals: only exact matches
+        #   - scalars: those that contain it
+
+        ser = series_with_interval_index.copy()
+
+        assert indexer_sl(ser)[1] == 0
+        assert indexer_sl(ser)[1.5] == 1
+        assert indexer_sl(ser)[2] == 1
+
+        expected = ser.iloc[1:4]
+        tm.assert_series_equal(expected, indexer_sl(ser)[[1.5, 2.5, 3.5]])
+        tm.assert_series_equal(expected, indexer_sl(ser)[[2, 3, 4]])
+        tm.assert_series_equal(expected, indexer_sl(ser)[[1.5, 3, 4]])
+
+        expected = ser.iloc[[1, 1, 2, 1]]
+        tm.assert_series_equal(expected, indexer_sl(ser)[[1.5, 2, 2.5, 1.5]])
+
+        expected = ser.iloc[2:5]
+        tm.assert_series_equal(expected, indexer_sl(ser)[ser >= 2])
+
+    def test_loc_with_slices(self, series_with_interval_index, indexer_sl):
+        # loc with slices:
+        #   - Interval objects: only works with exact matches
+        #   - scalars: only works for non-overlapping, monotonic intervals,
+        #     and start/stop select location based on the interval that
+        #     contains them:
+        #    (slice_loc(start, stop) == (idx.get_loc(start), idx.get_loc(stop))
+
+        ser = series_with_interval_index.copy()
+
+        # slice of interval
+
+        expected = ser.iloc[:3]
+        result = indexer_sl(ser)[Interval(0, 1) : Interval(2, 3)]
+        tm.assert_series_equal(expected, result)
+
+        expected = ser.iloc[3:]
+        result = indexer_sl(ser)[Interval(3, 4) :]
+        tm.assert_series_equal(expected, result)
+
+        msg = "Interval objects are not currently supported"
+        with pytest.raises(NotImplementedError, match=msg):
+            indexer_sl(ser)[Interval(3, 6) :]
+
+        with pytest.raises(NotImplementedError, match=msg):
+            indexer_sl(ser)[Interval(3, 4, closed="left") :]
+
+    def test_slice_step_ne1(self, series_with_interval_index):
+        # GH#31658 slice of scalar with step != 1
+        ser = series_with_interval_index.copy()
+        expected = ser.iloc[0:4:2]
+
+        result = ser[0:4:2]
+        tm.assert_series_equal(result, expected)
+
+        result2 = ser[0:4][::2]
+        tm.assert_series_equal(result2, expected)
+
+    def test_slice_float_start_stop(self, series_with_interval_index):
+        # GH#31658 slicing with integers is positional, with floats is not
+        #  supported
+        ser = series_with_interval_index.copy()
+
+        msg = "label-based slicing with step!=1 is not supported for IntervalIndex"
+        with pytest.raises(ValueError, match=msg):
+            ser[1.5:9.5:2]
+
+    def test_slice_interval_step(self, series_with_interval_index):
+        # GH#31658 allows for integer step!=1, not Interval step
+        ser = series_with_interval_index.copy()
+        msg = "label-based slicing with step!=1 is not supported for IntervalIndex"
+        with pytest.raises(ValueError, match=msg):
+            ser[0 : 4 : Interval(0, 1)]
+
+    def test_loc_with_overlap(self, indexer_sl):
+        idx = IntervalIndex.from_tuples([(1, 5), (3, 7)])
+        ser = Series(range(len(idx)), index=idx)
+
+        # scalar
+        expected = ser
+        result = indexer_sl(ser)[4]
+        tm.assert_series_equal(expected, result)
+
+        result = indexer_sl(ser)[[4]]
+        tm.assert_series_equal(expected, result)
+
+        # interval
+        expected = 0
+        result = indexer_sl(ser)[Interval(1, 5)]
+        assert expected == result
+
+        expected = ser
+        result = indexer_sl(ser)[[Interval(1, 5), Interval(3, 7)]]
+        tm.assert_series_equal(expected, result)
+
+        with pytest.raises(KeyError, match=re.escape("Interval(3, 5, closed='right')")):
+            indexer_sl(ser)[Interval(3, 5)]
+
+        msg = (
+            r"None of \[IntervalIndex\(\[\(3, 5\]\], "
+            r"dtype='interval\[int64, right\]'\)\] are in the \[index\]"
+        )
+        with pytest.raises(KeyError, match=msg):
+            indexer_sl(ser)[[Interval(3, 5)]]
+
+        # slices with interval (only exact matches)
+        expected = ser
+        result = indexer_sl(ser)[Interval(1, 5) : Interval(3, 7)]
+        tm.assert_series_equal(expected, result)
+
+        msg = (
+            "'can only get slices from an IntervalIndex if bounds are "
+            "non-overlapping and all monotonic increasing or decreasing'"
+        )
+        with pytest.raises(KeyError, match=msg):
+            indexer_sl(ser)[Interval(1, 6) : Interval(3, 8)]
+
+        if indexer_sl is tm.loc:
+            # slices with scalar raise for overlapping intervals
+            # TODO KeyError is the appropriate error?
+            with pytest.raises(KeyError, match=msg):
+                ser.loc[1:4]
+
+    def test_non_unique(self, indexer_sl):
+        idx = IntervalIndex.from_tuples([(1, 3), (3, 7)])
+        ser = Series(range(len(idx)), index=idx)
+
+        result = indexer_sl(ser)[Interval(1, 3)]
+        assert result == 0
+
+        result = indexer_sl(ser)[[Interval(1, 3)]]
+        expected = ser.iloc[0:1]
+        tm.assert_series_equal(expected, result)
+
+    def test_non_unique_moar(self, indexer_sl):
+        idx = IntervalIndex.from_tuples([(1, 3), (1, 3), (3, 7)])
+        ser = Series(range(len(idx)), index=idx)
+
+        expected = ser.iloc[[0, 1]]
+        result = indexer_sl(ser)[Interval(1, 3)]
+        tm.assert_series_equal(expected, result)
+
+        expected = ser
+        result = indexer_sl(ser)[Interval(1, 3) :]
+        tm.assert_series_equal(expected, result)
+
+        expected = ser.iloc[[0, 1]]
+        result = indexer_sl(ser)[[Interval(1, 3)]]
+        tm.assert_series_equal(expected, result)
+
+    def test_loc_getitem_missing_key_error_message(
+        self, frame_or_series, series_with_interval_index
+    ):
+        # GH#27365
+        ser = series_with_interval_index.copy()
+        obj = frame_or_series(ser)
+        with pytest.raises(KeyError, match=r"\[6\]"):
+            obj.loc[[4, 5, 6]]
+
+
+@pytest.mark.xfail(WASM, reason="GH 23440")
+@pytest.mark.parametrize(
+    "intervals",
+    [
+        ([Interval(-np.inf, 0.0), Interval(0.0, 1.0)]),
+        ([Interval(-np.inf, -2.0), Interval(-2.0, -1.0)]),
+        ([Interval(-1.0, 0.0), Interval(0.0, np.inf)]),
+        ([Interval(1.0, 2.0), Interval(2.0, np.inf)]),
+    ],
+)
+def test_repeating_interval_index_with_infs(intervals):
+    # GH 46658
+
+    interval_index = Index(intervals * 51)
+
+    expected = np.arange(1, 102, 2, dtype=np.intp)
+    result = interval_index.get_indexer_for([intervals[1]])
+
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/__init__.py b/pandas/pandas/tests/indexing/multiindex/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/indexing/multiindex/test_chaining_and_caching.py b/pandas/pandas/tests/indexing/multiindex/test_chaining_and_caching.py
new file mode 100644
index 0000000000000..c7ed21a2cc001
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_chaining_and_caching.py
@@ -0,0 +1,87 @@
+import numpy as np
+
+from pandas._libs import index as libindex
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    RangeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_detect_chained_assignment():
+    # Inplace ops, originally from:
+    # https://stackoverflow.com/questions/20508968/series-fillna-in-a-multiindex-dataframe-does-not-fill-is-this-a-bug
+    a = [12, 23]
+    b = [123, None]
+    c = [1234, 2345]
+    d = [12345, 23456]
+    tuples = [("eyes", "left"), ("eyes", "right"), ("ears", "left"), ("ears", "right")]
+    events = {
+        ("eyes", "left"): a,
+        ("eyes", "right"): b,
+        ("ears", "left"): c,
+        ("ears", "right"): d,
+    }
+    multiind = MultiIndex.from_tuples(tuples, names=["part", "side"])
+    zed = DataFrame(events, index=["a", "b"], columns=multiind)
+
+    with tm.raises_chained_assignment_error():
+        zed["eyes"]["right"].fillna(value=555, inplace=True)
+
+
+def test_cache_updating():
+    # 5216
+    # make sure that we don't try to set a dead cache
+    a = np.random.default_rng(2).random((10, 3))
+    df = DataFrame(a, columns=["x", "y", "z"])
+    df_original = df.copy()
+    tuples = [(i, j) for i in range(5) for j in range(2)]
+    index = MultiIndex.from_tuples(tuples)
+    df.index = index
+
+    # setting via chained assignment
+    # but actually works, since everything is a view
+
+    with tm.raises_chained_assignment_error():
+        df.loc[0]["z"].iloc[0] = 1.0
+
+    assert df.loc[(0, 0), "z"] == df_original.loc[0, "z"]
+
+    # correct setting
+    df.loc[(0, 0), "z"] = 2
+    result = df.loc[(0, 0), "z"]
+    assert result == 2
+
+
+def test_indexer_caching(monkeypatch):
+    # GH5727
+    # make sure that indexers are in the _internal_names_set
+    size_cutoff = 20
+    with monkeypatch.context():
+        monkeypatch.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+        index = MultiIndex.from_arrays([np.arange(size_cutoff), np.arange(size_cutoff)])
+        s = Series(np.zeros(size_cutoff), index=index)
+
+        # setitem
+        s[s == 0] = 1
+    expected = Series(np.ones(size_cutoff), index=index)
+    tm.assert_series_equal(s, expected)
+
+
+def test_set_names_only_clears_level_cache():
+    mi = MultiIndex.from_arrays([range(4), range(4)], names=["a", "b"])
+    mi.dtypes
+    mi.is_monotonic_increasing
+    mi._engine
+    mi.levels
+    old_cache_keys = sorted(mi._cache.keys())
+    assert old_cache_keys == ["_engine", "dtypes", "is_monotonic_increasing", "levels"]
+    mi.names = ["A", "B"]
+    new_cache_keys = sorted(mi._cache.keys())
+    assert new_cache_keys == ["_engine", "dtypes", "is_monotonic_increasing"]
+    new_levels = mi.levels
+    tm.assert_index_equal(new_levels[0], RangeIndex(4, name="A"))
+    tm.assert_index_equal(new_levels[1], RangeIndex(4, name="B"))
diff --git a/pandas/pandas/tests/indexing/multiindex/test_datetime.py b/pandas/pandas/tests/indexing/multiindex/test_datetime.py
new file mode 100644
index 0000000000000..d325971e7baf6
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_datetime.py
@@ -0,0 +1,50 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    Series,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+def test_multiindex_period_datetime():
+    # GH4861, using datetime in period of multiindex raises exception
+
+    idx1 = Index(["a", "a", "a", "b", "b"])
+    idx2 = period_range("2012-01", periods=len(idx1), freq="M")
+    s = Series(np.random.default_rng(2).standard_normal(len(idx1)), [idx1, idx2])
+
+    # try Period as index
+    expected = s.iloc[0]
+    result = s.loc["a", Period("2012-01")]
+    assert result == expected
+
+    # try datetime as index
+    result = s.loc["a", datetime(2012, 1, 1)]
+    assert result == expected
+
+
+def test_multiindex_datetime_columns():
+    # GH35015, using datetime as column indices raises exception
+
+    mi = MultiIndex.from_tuples(
+        [(to_datetime("02/29/2020"), to_datetime("03/01/2020"))], names=["a", "b"]
+    )
+
+    df = DataFrame([], columns=mi)
+
+    expected_df = DataFrame(
+        [],
+        columns=MultiIndex.from_arrays(
+            [[to_datetime("02/29/2020")], [to_datetime("03/01/2020")]], names=["a", "b"]
+        ),
+    )
+
+    tm.assert_frame_equal(df, expected_df)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_getitem.py b/pandas/pandas/tests/indexing/multiindex/test_getitem.py
new file mode 100644
index 0000000000000..b86e233110e88
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_getitem.py
@@ -0,0 +1,410 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.indexing import IndexingError
+
+# ----------------------------------------------------------------------------
+# test indexing of Series with multi-level Index
+# ----------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize(
+    "access_method",
+    [lambda s, x: s[:, x], lambda s, x: s.loc[:, x], lambda s, x: s.xs(x, level=1)],
+)
+@pytest.mark.parametrize(
+    "level1_value, expected",
+    [(0, Series([1], index=[0])), (1, Series([2, 3], index=[1, 2]))],
+)
+def test_series_getitem_multiindex(access_method, level1_value, expected):
+    # GH 6018
+    # series regression getitem with a multi-index
+
+    mi = MultiIndex.from_tuples([(0, 0), (1, 1), (2, 1)], names=["A", "B"])
+    ser = Series([1, 2, 3], index=mi)
+    expected.index.name = "A"
+
+    result = access_method(ser, level1_value)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("level0_value", ["D", "A"])
+def test_series_getitem_duplicates_multiindex(level0_value):
+    # GH 5725 the 'A' happens to be a valid Timestamp so the doesn't raise
+    # the appropriate error, only in PY3 of course!
+
+    index = MultiIndex(
+        levels=[[level0_value, "B", "C"], [0, 26, 27, 37, 57, 67, 75, 82]],
+        codes=[[0, 0, 0, 1, 2, 2, 2, 2, 2, 2], [1, 3, 4, 6, 0, 2, 2, 3, 5, 7]],
+        names=["tag", "day"],
+    )
+    arr = np.random.default_rng(2).standard_normal((len(index), 1))
+    df = DataFrame(arr, index=index, columns=["val"])
+
+    # confirm indexing on missing value raises KeyError
+    if level0_value != "A":
+        with pytest.raises(KeyError, match=r"^'A'$"):
+            df.val["A"]
+
+    with pytest.raises(KeyError, match=r"^'X'$"):
+        df.val["X"]
+
+    result = df.val[level0_value]
+    expected = Series(
+        arr.ravel()[0:3], name="val", index=Index([26, 37, 57], name="day")
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_getitem(multiindex_year_month_day_dataframe_random_data, indexer_sl):
+    s = multiindex_year_month_day_dataframe_random_data["A"]
+    expected = s.reindex(s.index[42:65])
+    expected.index = expected.index.droplevel(0).droplevel(0)
+
+    result = indexer_sl(s)[2000, 3]
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_getitem_returns_scalar(
+    multiindex_year_month_day_dataframe_random_data, indexer_sl
+):
+    s = multiindex_year_month_day_dataframe_random_data["A"]
+    expected = s.iloc[49]
+
+    result = indexer_sl(s)[2000, 3, 10]
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "indexer,expected_error,expected_error_msg",
+    [
+        (lambda s: s.__getitem__((2000, 3, 4)), KeyError, r"^\(2000, 3, 4\)$"),
+        (lambda s: s[(2000, 3, 4)], KeyError, r"^\(2000, 3, 4\)$"),
+        (lambda s: s.loc[(2000, 3, 4)], KeyError, r"^\(2000, 3, 4\)$"),
+        (lambda s: s.loc[(2000, 3, 4, 5)], IndexingError, "Too many indexers"),
+        (lambda s: s.__getitem__(len(s)), KeyError, ""),  # match should include len(s)
+        (lambda s: s[len(s)], KeyError, ""),  # match should include len(s)
+        (
+            lambda s: s.iloc[len(s)],
+            IndexError,
+            "single positional indexer is out-of-bounds",
+        ),
+    ],
+)
+def test_series_getitem_indexing_errors(
+    multiindex_year_month_day_dataframe_random_data,
+    indexer,
+    expected_error,
+    expected_error_msg,
+):
+    s = multiindex_year_month_day_dataframe_random_data["A"]
+    with pytest.raises(expected_error, match=expected_error_msg):
+        indexer(s)
+
+
+def test_series_getitem_corner_generator(
+    multiindex_year_month_day_dataframe_random_data,
+):
+    s = multiindex_year_month_day_dataframe_random_data["A"]
+    result = s[(x > 0 for x in s)]
+    expected = s[s > 0]
+    tm.assert_series_equal(result, expected)
+
+
+# ----------------------------------------------------------------------------
+# test indexing of DataFrame with multi-level Index
+# ----------------------------------------------------------------------------
+
+
+def test_getitem_simple(multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data.T
+    expected = df.values[:, 0]
+    result = df["foo", "one"].values
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer,expected_error_msg",
+    [
+        (lambda df: df[("foo", "four")], r"^\('foo', 'four'\)$"),
+        (lambda df: df["foobar"], r"^'foobar'$"),
+    ],
+)
+def test_frame_getitem_simple_key_error(
+    multiindex_dataframe_random_data, indexer, expected_error_msg
+):
+    df = multiindex_dataframe_random_data.T
+    with pytest.raises(KeyError, match=expected_error_msg):
+        indexer(df)
+
+
+def test_tuple_string_column_names():
+    # GH#50372
+    mi = MultiIndex.from_tuples([("a", "aa"), ("a", "ab"), ("b", "ba"), ("b", "bb")])
+    df = DataFrame([range(4), range(1, 5), range(2, 6)], columns=mi)
+    df["single_index"] = 0
+
+    df_flat = df.copy()
+    df_flat.columns = df_flat.columns.to_flat_index()
+    df_flat["new_single_index"] = 0
+
+    result = df_flat[[("a", "aa"), "new_single_index"]]
+    expected = DataFrame(
+        [[0, 0], [1, 0], [2, 0]], columns=Index([("a", "aa"), "new_single_index"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_getitem_multicolumn_empty_level():
+    df = DataFrame({"a": ["1", "2", "3"], "b": ["2", "3", "4"]})
+    df.columns = [
+        ["level1 item1", "level1 item2"],
+        ["", "level2 item2"],
+        ["level3 item1", "level3 item2"],
+    ]
+
+    result = df["level1 item1"]
+    expected = DataFrame(
+        [["1"], ["2"], ["3"]], index=df.index, columns=["level3 item1"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer,expected_slice",
+    [
+        (lambda df: df["foo"], slice(3)),
+        (lambda df: df["bar"], slice(3, 5)),
+        (lambda df: df.loc[:, "bar"], slice(3, 5)),
+    ],
+)
+def test_frame_getitem_toplevel(
+    multiindex_dataframe_random_data, indexer, expected_slice
+):
+    df = multiindex_dataframe_random_data.T
+    expected = df.reindex(columns=df.columns[expected_slice])
+    expected.columns = expected.columns.droplevel(0)
+    result = indexer(df)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_mixed_depth_get():
+    arrays = [
+        ["a", "top", "top", "routine1", "routine1", "routine2"],
+        ["", "OD", "OD", "result1", "result2", "result1"],
+        ["", "wx", "wy", "", "", ""],
+    ]
+
+    tuples = sorted(zip(*arrays))
+    index = MultiIndex.from_tuples(tuples)
+    df = DataFrame(np.random.default_rng(2).standard_normal((4, 6)), columns=index)
+
+    result = df["a"]
+    expected = df["a", "", ""].rename("a")
+    tm.assert_series_equal(result, expected)
+
+    result = df["routine1", "result1"]
+    expected = df["routine1", "result1", ""]
+    expected = expected.rename(("routine1", "result1"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_frame_getitem_nan_multiindex(nulls_fixture):
+    # GH#29751
+    # loc on a multiindex containing nan values
+    n = nulls_fixture  # for code readability
+    cols = ["a", "b", "c"]
+    df = DataFrame(
+        [[11, n, 13], [21, n, 23], [31, n, 33], [41, n, 43]],
+        columns=cols,
+    ).set_index(["a", "b"])
+    df["c"] = df["c"].astype("int64")
+
+    idx = (21, n)
+    result = df.loc[:idx]
+    expected = DataFrame([[11, n, 13], [21, n, 23]], columns=cols).set_index(["a", "b"])
+    expected["c"] = expected["c"].astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+    result = df.loc[idx:]
+    expected = DataFrame(
+        [[21, n, 23], [31, n, 33], [41, n, 43]], columns=cols
+    ).set_index(["a", "b"])
+    expected["c"] = expected["c"].astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+    idx1, idx2 = (21, n), (31, n)
+    result = df.loc[idx1:idx2]
+    expected = DataFrame([[21, n, 23], [31, n, 33]], columns=cols).set_index(["a", "b"])
+    expected["c"] = expected["c"].astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer,expected",
+    [
+        (
+            (["b"], ["bar", np.nan]),
+            (
+                DataFrame(
+                    [[2, 3], [5, 6]],
+                    columns=MultiIndex.from_tuples([("b", "bar"), ("b", np.nan)]),
+                    dtype="int64",
+                )
+            ),
+        ),
+        (
+            (["a", "b"]),
+            (
+                DataFrame(
+                    [[1, 2, 3], [4, 5, 6]],
+                    columns=MultiIndex.from_tuples(
+                        [("a", "foo"), ("b", "bar"), ("b", np.nan)]
+                    ),
+                    dtype="int64",
+                )
+            ),
+        ),
+        (
+            (["b"]),
+            (
+                DataFrame(
+                    [[2, 3], [5, 6]],
+                    columns=MultiIndex.from_tuples([("b", "bar"), ("b", np.nan)]),
+                    dtype="int64",
+                )
+            ),
+        ),
+        (
+            (["b"], ["bar"]),
+            (
+                DataFrame(
+                    [[2], [5]],
+                    columns=MultiIndex.from_tuples([("b", "bar")]),
+                    dtype="int64",
+                )
+            ),
+        ),
+        (
+            (["b"], [np.nan]),
+            (
+                DataFrame(
+                    [[3], [6]],
+                    columns=MultiIndex(
+                        codes=[[1], [-1]], levels=[["a", "b"], ["bar", "foo"]]
+                    ),
+                    dtype="int64",
+                )
+            ),
+        ),
+        (("b", np.nan), Series([3, 6], dtype="int64", name=("b", np.nan))),
+    ],
+)
+def test_frame_getitem_nan_cols_multiindex(
+    indexer,
+    expected,
+    nulls_fixture,
+):
+    # Slicing MultiIndex including levels with nan values, for more information
+    # see GH#25154
+    df = DataFrame(
+        [[1, 2, 3], [4, 5, 6]],
+        columns=MultiIndex.from_tuples(
+            [("a", "foo"), ("b", "bar"), ("b", nulls_fixture)]
+        ),
+        dtype="int64",
+    )
+
+    result = df.loc[:, indexer]
+    tm.assert_equal(result, expected)
+
+
+# ----------------------------------------------------------------------------
+# test indexing of DataFrame with multi-level Index with duplicates
+# ----------------------------------------------------------------------------
+
+
+@pytest.fixture
+def dataframe_with_duplicate_index():
+    """Fixture for DataFrame used in tests for gh-4145 and gh-4146"""
+    data = [["a", "d", "e", "c", "f", "b"], [1, 4, 5, 3, 6, 2], [1, 4, 5, 3, 6, 2]]
+    index = ["h1", "h3", "h5"]
+    columns = MultiIndex(
+        levels=[["A", "B"], ["A1", "A2", "B1", "B2"]],
+        codes=[[0, 0, 0, 1, 1, 1], [0, 3, 3, 0, 1, 2]],
+        names=["main", "sub"],
+    )
+    return DataFrame(data, index=index, columns=columns)
+
+
+@pytest.mark.parametrize(
+    "indexer", [lambda df: df[("A", "A1")], lambda df: df.loc[:, ("A", "A1")]]
+)
+def test_frame_mi_access(dataframe_with_duplicate_index, indexer):
+    # GH 4145
+    df = dataframe_with_duplicate_index
+    index = Index(["h1", "h3", "h5"])
+    columns = MultiIndex.from_tuples([("A", "A1")], names=["main", "sub"])
+    expected = DataFrame([["a", 1, 1]], index=columns, columns=index).T
+
+    result = indexer(df)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_mi_access_returns_series(dataframe_with_duplicate_index):
+    # GH 4146, not returning a block manager when selecting a unique index
+    # from a duplicate index
+    # as of 4879, this returns a Series (which is similar to what happens
+    # with a non-unique)
+    df = dataframe_with_duplicate_index
+    expected = Series(["a", 1, 1], index=["h1", "h3", "h5"], name="A1")
+    result = df["A"]["A1"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_frame_mi_access_returns_frame(dataframe_with_duplicate_index):
+    # selecting a non_unique from the 2nd level
+    df = dataframe_with_duplicate_index
+    expected = DataFrame(
+        [["d", 4, 4], ["e", 5, 5]],
+        index=Index(["B2", "B2"], name="sub"),
+        columns=["h1", "h3", "h5"],
+    ).T
+    result = df["A"]["B2"]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_mi_empty_slice():
+    # GH 15454
+    df = DataFrame(0, index=range(2), columns=MultiIndex.from_product([[1], [2]]))
+    result = df[[]]
+    expected = DataFrame(
+        index=[0, 1], columns=MultiIndex(levels=[[1], [2]], codes=[[], []])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_empty_multiindex():
+    # GH#36936
+    arrays = [["a", "a", "b", "a"], ["a", "a", "b", "b"]]
+    index = MultiIndex.from_arrays(arrays, names=("idx1", "idx2"))
+    df = DataFrame([1, 2, 3, 4], index=index, columns=["value"])
+
+    # loc on empty multiindex == loc with False mask
+    empty_multiindex = df.loc[df.loc[:, "value"] == 0, :].index
+    result = df.loc[empty_multiindex, :]
+    expected = df.loc[[False] * len(df.index), :]
+    tm.assert_frame_equal(result, expected)
+
+    # replacing value with loc on empty multiindex
+    df.loc[df.loc[df.loc[:, "value"] == 0].index, "value"] = 5
+    result = df
+    expected = DataFrame([1, 2, 3, 4], index=index, columns=["value"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_iloc.py b/pandas/pandas/tests/indexing/multiindex/test_iloc.py
new file mode 100644
index 0000000000000..8939ecc78000b
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_iloc.py
@@ -0,0 +1,171 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def simple_multiindex_dataframe():
+    """
+    Factory function to create simple 3 x 3 dataframe with
+    both columns and row MultiIndex using supplied data or
+    random data by default.
+    """
+
+    data = np.random.default_rng(2).standard_normal((3, 3))
+    return DataFrame(
+        data, columns=[[2, 2, 4], [6, 8, 10]], index=[[4, 4, 8], [8, 10, 12]]
+    )
+
+
+@pytest.mark.parametrize(
+    "indexer, expected",
+    [
+        (
+            lambda df: df.iloc[0],
+            lambda arr: Series(arr[0], index=[[2, 2, 4], [6, 8, 10]], name=(4, 8)),
+        ),
+        (
+            lambda df: df.iloc[2],
+            lambda arr: Series(arr[2], index=[[2, 2, 4], [6, 8, 10]], name=(8, 12)),
+        ),
+        (
+            lambda df: df.iloc[:, 2],
+            lambda arr: Series(arr[:, 2], index=[[4, 4, 8], [8, 10, 12]], name=(4, 10)),
+        ),
+    ],
+)
+def test_iloc_returns_series(indexer, expected, simple_multiindex_dataframe):
+    df = simple_multiindex_dataframe
+    arr = df.values
+    result = indexer(df)
+    expected = expected(arr)
+    tm.assert_series_equal(result, expected)
+
+
+def test_iloc_returns_dataframe(simple_multiindex_dataframe):
+    df = simple_multiindex_dataframe
+    result = df.iloc[[0, 1]]
+    expected = df.xs(4, drop_level=False)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_iloc_returns_scalar(simple_multiindex_dataframe):
+    df = simple_multiindex_dataframe
+    arr = df.values
+    result = df.iloc[2, 2]
+    expected = arr[2, 2]
+    assert result == expected
+
+
+def test_iloc_getitem_multiple_items():
+    # GH 5528
+    tup = zip(*[["a", "a", "b", "b"], ["x", "y", "x", "y"]])
+    index = MultiIndex.from_tuples(tup)
+    df = DataFrame(np.random.default_rng(2).standard_normal((4, 4)), index=index)
+    result = df.iloc[[2, 3]]
+    expected = df.xs("b", drop_level=False)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_iloc_getitem_labels():
+    # this is basically regular indexing
+    arr = np.random.default_rng(2).standard_normal((4, 3))
+    df = DataFrame(
+        arr,
+        columns=[["i", "i", "j"], ["A", "A", "B"]],
+        index=[["i", "i", "j", "k"], ["X", "X", "Y", "Y"]],
+    )
+    result = df.iloc[2, 2]
+    expected = arr[2, 2]
+    assert result == expected
+
+
+def test_frame_getitem_slice(multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data
+    result = df.iloc[:4]
+    expected = df[:4]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_frame_setitem_slice(multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data
+    df.iloc[:4] = 0
+
+    assert (df.values[:4] == 0).all()
+    assert (df.values[4:] != 0).all()
+
+
+def test_indexing_ambiguity_bug_1678():
+    # GH 1678
+    columns = MultiIndex.from_tuples(
+        [("Ohio", "Green"), ("Ohio", "Red"), ("Colorado", "Green")]
+    )
+    index = MultiIndex.from_tuples([("a", 1), ("a", 2), ("b", 1), ("b", 2)])
+
+    df = DataFrame(np.arange(12).reshape((4, 3)), index=index, columns=columns)
+
+    result = df.iloc[:, 1]
+    expected = df.loc[:, ("Ohio", "Red")]
+    tm.assert_series_equal(result, expected)
+
+
+def test_iloc_integer_locations():
+    # GH 13797
+    data = [
+        ["str00", "str01"],
+        ["str10", "str11"],
+        ["str20", "srt21"],
+        ["str30", "str31"],
+        ["str40", "str41"],
+    ]
+
+    index = MultiIndex.from_tuples(
+        [("CC", "A"), ("CC", "B"), ("CC", "B"), ("BB", "a"), ("BB", "b")]
+    )
+
+    expected = DataFrame(data)
+    df = DataFrame(data, index=index)
+
+    result = DataFrame([[df.iloc[r, c] for c in range(2)] for r in range(5)])
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, indexes, values, expected_k",
+    [
+        # test without indexer value in first level of MultiIndex
+        ([[2, 22, 5], [2, 33, 6]], [0, -1, 1], [2, 3, 1], [7, 10]),
+        # test like code sample 1 in the issue
+        ([[1, 22, 555], [1, 33, 666]], [0, -1, 1], [200, 300, 100], [755, 1066]),
+        # test like code sample 2 in the issue
+        ([[1, 3, 7], [2, 4, 8]], [0, -1, 1], [10, 10, 1000], [17, 1018]),
+        # test like code sample 3 in the issue
+        ([[1, 11, 4], [2, 22, 5], [3, 33, 6]], [0, -1, 1], [4, 7, 10], [8, 15, 13]),
+    ],
+)
+def test_iloc_setitem_int_multiindex_series(data, indexes, values, expected_k):
+    # GH17148
+    df = DataFrame(data=data, columns=["i", "j", "k"])
+    df = df.set_index(["i", "j"])
+
+    series = df.k.copy()
+    for i, v in zip(indexes, values):
+        series.iloc[i] += v
+
+    df["k"] = expected_k
+    expected = df.k
+    tm.assert_series_equal(series, expected)
+
+
+def test_getitem_iloc(multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data
+    result = df.iloc[2]
+    expected = df.xs(df.index[2])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_indexing_slow.py b/pandas/pandas/tests/indexing/multiindex/test_indexing_slow.py
new file mode 100644
index 0000000000000..c6fc1659500e6
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_indexing_slow.py
@@ -0,0 +1,118 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def m():
+    return 5
+
+
+@pytest.fixture
+def n():
+    return 100
+
+
+@pytest.fixture
+def cols():
+    return ["jim", "joe", "jolie", "joline", "jolia"]
+
+
+@pytest.fixture
+def vals(n):
+    vals = [
+        np.random.default_rng(2).integers(0, 10, n),
+        np.random.default_rng(2).choice(list("abcdefghij"), n),
+        np.random.default_rng(2).choice(
+            pd.date_range("20141009", periods=10).tolist(), n
+        ),
+        np.random.default_rng(2).choice(list("ZYXWVUTSRQ"), n),
+        np.random.default_rng(2).standard_normal(n),
+    ]
+    vals = list(map(tuple, zip(*vals)))
+    return vals
+
+
+@pytest.fixture
+def keys(n, m, vals):
+    # bunch of keys for testing
+    keys = [
+        np.random.default_rng(2).integers(0, 11, m),
+        np.random.default_rng(2).choice(list("abcdefghijk"), m),
+        np.random.default_rng(2).choice(
+            pd.date_range("20141009", periods=11).tolist(), m
+        ),
+        np.random.default_rng(2).choice(list("ZYXWVUTSRQP"), m),
+    ]
+    keys = list(map(tuple, zip(*keys)))
+    keys += [t[:-1] for t in vals[:: n // m]]
+    return keys
+
+
+# covers both unique index and non-unique index
+@pytest.fixture
+def df(vals, cols):
+    return DataFrame(vals, columns=cols)
+
+
+@pytest.fixture
+def a(df):
+    return pd.concat([df, df])
+
+
+@pytest.fixture
+def b(df, cols):
+    return df.drop_duplicates(subset=cols[:-1])
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+@pytest.mark.parametrize("lexsort_depth", list(range(5)))
+@pytest.mark.parametrize("frame_fixture", ["a", "b"])
+def test_multiindex_get_loc(request, lexsort_depth, keys, frame_fixture, cols):
+    # GH7724, GH2646
+
+    frame = request.getfixturevalue(frame_fixture)
+    if lexsort_depth == 0:
+        df = frame.copy(deep=False)
+    else:
+        df = frame.sort_values(by=cols[:lexsort_depth])
+
+    mi = df.set_index(cols[:-1])
+    assert not mi.index._lexsort_depth < lexsort_depth
+    for key in keys:
+        mask = np.ones(len(df), dtype=bool)
+
+        # test for all partials of this key
+        for i, k in enumerate(key):
+            mask &= df.iloc[:, i] == k
+
+            if not mask.any():
+                assert key[: i + 1] not in mi.index
+                continue
+
+            assert key[: i + 1] in mi.index
+            right = df[mask].copy(deep=False)
+
+            if i + 1 != len(key):  # partial key
+                return_value = right.drop(cols[: i + 1], axis=1, inplace=True)
+                assert return_value is None
+                return_value = right.set_index(cols[i + 1 : -1], inplace=True)
+                assert return_value is None
+                tm.assert_frame_equal(mi.loc[key[: i + 1]], right)
+
+            else:  # full key
+                return_value = right.set_index(cols[:-1], inplace=True)
+                assert return_value is None
+                if len(right) == 1:  # single hit
+                    right = Series(
+                        right["jolia"].values, name=right.index[0], index=["jolia"]
+                    )
+                    tm.assert_series_equal(mi.loc[key[: i + 1]], right)
+                else:  # multi hit
+                    tm.assert_frame_equal(mi.loc[key[: i + 1]], right)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_loc.py b/pandas/pandas/tests/indexing/multiindex/test_loc.py
new file mode 100644
index 0000000000000..1d3258ab18a61
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_loc.py
@@ -0,0 +1,1005 @@
+import numpy as np
+import pytest
+
+from pandas.errors import IndexingError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def frame_random_data_integer_multi_index():
+    levels = [[0, 1], [0, 1, 2]]
+    codes = [[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]]
+    index = MultiIndex(levels=levels, codes=codes)
+    return DataFrame(np.random.default_rng(2).standard_normal((6, 2)), index=index)
+
+
+class TestMultiIndexLoc:
+    def test_loc_setitem_frame_with_multiindex(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        frame.loc[("bar", "two"), "B"] = 5
+        assert frame.loc[("bar", "two"), "B"] == 5
+
+        # with integer labels
+        df = frame.copy()
+        df.columns = list(range(3))
+        df.loc[("bar", "two"), 1] = 7
+        assert df.loc[("bar", "two"), 1] == 7
+
+    def test_loc_getitem_general(self, performance_warning, any_real_numpy_dtype):
+        # GH#2817
+        dtype = any_real_numpy_dtype
+        data = {
+            "amount": {0: 700, 1: 600, 2: 222, 3: 333, 4: 444},
+            "col": {0: 3.5, 1: 3.5, 2: 4.0, 3: 4.0, 4: 4.0},
+            "num": {0: 12, 1: 11, 2: 12, 3: 12, 4: 12},
+        }
+        df = DataFrame(data)
+        df = df.astype({"col": dtype, "num": dtype})
+        df = df.set_index(keys=["col", "num"])
+        key = 4.0, 12
+
+        with tm.assert_produces_warning(performance_warning):
+            tm.assert_frame_equal(df.loc[key], df.iloc[2:])
+
+        # this is ok
+        return_value = df.sort_index(inplace=True)
+        assert return_value is None
+        res = df.loc[key]
+
+        # col has float dtype, result should be float64 Index
+        col_arr = np.array([4.0] * 3, dtype=dtype)
+        year_arr = np.array([12] * 3, dtype=dtype)
+        index = MultiIndex.from_arrays([col_arr, year_arr], names=["col", "num"])
+        expected = DataFrame({"amount": [222, 333, 444]}, index=index)
+        tm.assert_frame_equal(res, expected)
+
+    def test_loc_getitem_multiindex_missing_label_raises(self):
+        # GH#21593
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=[[2, 2, 4], [6, 8, 10]],
+            index=[[4, 4, 8], [8, 10, 12]],
+        )
+
+        with pytest.raises(KeyError, match=r"^2$"):
+            df.loc[2]
+
+    def test_loc_getitem_list_of_tuples_with_multiindex(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        ser = multiindex_year_month_day_dataframe_random_data["A"]
+        expected = ser.reindex(ser.index[49:51])
+        result = ser.loc[[(2000, 3, 10), (2000, 3, 13)]]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_series(self):
+        # GH14730
+        # passing a series as a key with a MultiIndex
+        index = MultiIndex.from_product([[1, 2, 3], ["A", "B", "C"]])
+        x = Series(index=index, data=range(9), dtype=np.float64)
+        y = Series([1, 3])
+        expected = Series(
+            data=[0, 1, 2, 6, 7, 8],
+            index=MultiIndex.from_product([[1, 3], ["A", "B", "C"]]),
+            dtype=np.float64,
+        )
+        result = x.loc[y]
+        tm.assert_series_equal(result, expected)
+
+        result = x.loc[[1, 3]]
+        tm.assert_series_equal(result, expected)
+
+        # GH15424
+        y1 = Series([1, 3], index=[1, 2])
+        result = x.loc[y1]
+        tm.assert_series_equal(result, expected)
+
+        empty = Series(data=[], dtype=np.float64)
+        expected = Series(
+            [],
+            index=MultiIndex(levels=index.levels, codes=[[], []]),
+            dtype=np.float64,
+        )
+        result = x.loc[empty]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_array(self):
+        # GH15434
+        # passing an array as a key with a MultiIndex
+        index = MultiIndex.from_product([[1, 2, 3], ["A", "B", "C"]])
+        x = Series(index=index, data=range(9), dtype=np.float64)
+        y = np.array([1, 3])
+        expected = Series(
+            data=[0, 1, 2, 6, 7, 8],
+            index=MultiIndex.from_product([[1, 3], ["A", "B", "C"]]),
+            dtype=np.float64,
+        )
+        result = x.loc[y]
+        tm.assert_series_equal(result, expected)
+
+        # empty array:
+        empty = np.array([])
+        expected = Series(
+            [],
+            index=MultiIndex(levels=index.levels, codes=[[], []]),
+            dtype="float64",
+        )
+        result = x.loc[empty]
+        tm.assert_series_equal(result, expected)
+
+        # 0-dim array (scalar):
+        scalar = np.int64(1)
+        expected = Series(data=[0, 1, 2], index=["A", "B", "C"], dtype=np.float64)
+        result = x.loc[scalar]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_multiindex_labels(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=[["i", "i", "j"], ["A", "A", "B"]],
+            index=[["i", "i", "j"], ["X", "X", "Y"]],
+        )
+
+        # the first 2 rows
+        expected = df.iloc[[0, 1]].droplevel(0)
+        result = df.loc["i"]
+        tm.assert_frame_equal(result, expected)
+
+        # 2nd (last) column
+        expected = df.iloc[:, [2]].droplevel(0, axis=1)
+        result = df.loc[:, "j"]
+        tm.assert_frame_equal(result, expected)
+
+        # bottom right corner
+        expected = df.iloc[[2], [2]].droplevel(0).droplevel(0, axis=1)
+        result = df.loc["j"].loc[:, "j"]
+        tm.assert_frame_equal(result, expected)
+
+        # with a tuple
+        expected = df.iloc[[0, 1]]
+        result = df.loc[("i", "X")]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_multiindex_ints(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=[[2, 2, 4], [6, 8, 10]],
+            index=[[4, 4, 8], [8, 10, 12]],
+        )
+        expected = df.iloc[[0, 1]].droplevel(0)
+        result = df.loc[4]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_multiindex_missing_label_raises(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=[[2, 2, 4], [6, 8, 10]],
+            index=[[4, 4, 8], [8, 10, 12]],
+        )
+
+        with pytest.raises(KeyError, match=r"^2$"):
+            df.loc[2]
+
+    @pytest.mark.parametrize("key, pos", [([2, 4], [0, 1]), ([2], []), ([2, 3], [])])
+    def test_loc_multiindex_list_missing_label(self, key, pos):
+        # GH 27148 - lists with missing labels _do_ raise
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=[[2, 2, 4], [6, 8, 10]],
+            index=[[4, 4, 8], [8, 10, 12]],
+        )
+
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[key]
+
+    def test_loc_multiindex_too_many_dims_raises(self):
+        # GH 14885
+        s = Series(
+            range(8),
+            index=MultiIndex.from_product([["a", "b"], ["c", "d"], ["e", "f"]]),
+        )
+
+        with pytest.raises(KeyError, match=r"^\('a', 'b'\)$"):
+            s.loc["a", "b"]
+        with pytest.raises(KeyError, match=r"^\('a', 'd', 'g'\)$"):
+            s.loc["a", "d", "g"]
+        with pytest.raises(IndexingError, match="Too many indexers"):
+            s.loc["a", "d", "g", "j"]
+
+    def test_loc_multiindex_indexer_none(self):
+        # GH6788
+        # multi-index indexer is None (meaning take all)
+        attributes = ["Attribute" + str(i) for i in range(1)]
+        attribute_values = ["Value" + str(i) for i in range(5)]
+
+        index = MultiIndex.from_product([attributes, attribute_values])
+        df = 0.1 * np.random.default_rng(2).standard_normal((10, 1 * 5)) + 0.5
+        df = DataFrame(df, columns=index)
+        result = df[attributes]
+        tm.assert_frame_equal(result, df)
+
+        # GH 7349
+        # loc with a multi-index seems to be doing fallback
+        df = DataFrame(
+            np.arange(12).reshape(-1, 1),
+            index=MultiIndex.from_product([[1, 2, 3, 4], [1, 2, 3]]),
+        )
+
+        expected = df.loc[([1, 2],), :]
+        result = df.loc[[1, 2]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_multiindex_incomplete(self):
+        # GH 7399
+        # incomplete indexers
+        s = Series(
+            np.arange(15, dtype="int64"),
+            MultiIndex.from_product([range(5), ["a", "b", "c"]]),
+        )
+        expected = s.loc[:, "a":"c"]
+
+        result = s.loc[0:4, "a":"c"]
+        tm.assert_series_equal(result, expected)
+
+        result = s.loc[:4, "a":"c"]
+        tm.assert_series_equal(result, expected)
+
+        result = s.loc[0:, "a":"c"]
+        tm.assert_series_equal(result, expected)
+
+        # GH 7400
+        # multiindexer getitem with list of indexers skips wrong element
+        s = Series(
+            np.arange(15, dtype="int64"),
+            MultiIndex.from_product([range(5), ["a", "b", "c"]]),
+        )
+        expected = s.iloc[[6, 7, 8, 12, 13, 14]]
+        result = s.loc[2:4:2, "a":"c"]
+        tm.assert_series_equal(result, expected)
+
+    def test_get_loc_single_level(self):
+        single_level = MultiIndex(
+            levels=[["foo", "bar", "baz", "qux"]], codes=[[0, 1, 2, 3]], names=["first"]
+        )
+        s = Series(
+            np.random.default_rng(2).standard_normal(len(single_level)),
+            index=single_level,
+        )
+        for k in single_level.values:
+            s[k]
+
+    def test_loc_getitem_int_slice(self):
+        # GH 3053
+        # loc should treat integer slices like label slices
+
+        index = MultiIndex.from_product([[6, 7, 8], ["a", "b"]])
+        df = DataFrame(np.random.default_rng(2).standard_normal((6, 6)), index, index)
+        result = df.loc[6:8, :]
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+        index = MultiIndex.from_product([[10, 20, 30], ["a", "b"]])
+        df = DataFrame(np.random.default_rng(2).standard_normal((6, 6)), index, index)
+        result = df.loc[20:30, :]
+        expected = df.iloc[2:]
+        tm.assert_frame_equal(result, expected)
+
+        # doc examples
+        result = df.loc[10, :]
+        expected = df.iloc[0:2]
+        expected.index = ["a", "b"]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[:, 10]
+        expected = df[10]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "indexer_type_1", (list, tuple, set, slice, np.ndarray, Series, Index)
+    )
+    @pytest.mark.parametrize(
+        "indexer_type_2", (list, tuple, set, slice, np.ndarray, Series, Index)
+    )
+    def test_loc_getitem_nested_indexer(self, indexer_type_1, indexer_type_2):
+        # GH #19686
+        # .loc should work with nested indexers which can be
+        # any list-like objects (see `is_list_like` (`pandas.api.types`)) or slices
+
+        def convert_nested_indexer(indexer_type, keys):
+            if indexer_type == np.ndarray:
+                return np.array(keys)
+            if indexer_type == slice:
+                return slice(*keys)
+            return indexer_type(keys)
+
+        a = [10, 20, 30]
+        b = [1, 2, 3]
+        index = MultiIndex.from_product([a, b])
+        df = DataFrame(
+            np.arange(len(index), dtype="int64"), index=index, columns=["Data"]
+        )
+
+        keys = ([10, 20], [2, 3])
+        types = (indexer_type_1, indexer_type_2)
+
+        # check indexers with all the combinations of nested objects
+        # of all the valid types
+        indexer = tuple(
+            convert_nested_indexer(indexer_type, k)
+            for indexer_type, k in zip(types, keys)
+        )
+        if indexer_type_1 is set or indexer_type_2 is set:
+            with pytest.raises(TypeError, match="as an indexer is not supported"):
+                df.loc[indexer, "Data"]
+
+            return
+        else:
+            result = df.loc[indexer, "Data"]
+        expected = Series(
+            [1, 2, 4, 5], name="Data", index=MultiIndex.from_product(keys)
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_multiindex_loc_one_dimensional_tuple(self, frame_or_series):
+        # GH#37711
+        mi = MultiIndex.from_tuples([("a", "A"), ("b", "A")])
+        obj = frame_or_series([1, 2], index=mi)
+        obj.loc[("a",)] = 0
+        expected = frame_or_series([0, 2], index=mi)
+        tm.assert_equal(obj, expected)
+
+    @pytest.mark.parametrize("indexer", [("a",), ("a")])
+    def test_multiindex_one_dimensional_tuple_columns(self, indexer):
+        # GH#37711
+        mi = MultiIndex.from_tuples([("a", "A"), ("b", "A")])
+        obj = DataFrame([1, 2], index=mi)
+        obj.loc[indexer, :] = 0
+        expected = DataFrame([0, 2], index=mi)
+        tm.assert_frame_equal(obj, expected)
+
+    @pytest.mark.parametrize(
+        "indexer, exp_value", [(slice(None), 1.0), ((1, 2), np.nan)]
+    )
+    def test_multiindex_setitem_columns_enlarging(self, indexer, exp_value):
+        # GH#39147
+        mi = MultiIndex.from_tuples([(1, 2), (3, 4)])
+        df = DataFrame([[1, 2], [3, 4]], index=mi, columns=["a", "b"])
+        df.loc[indexer, ["c", "d"]] = 1.0
+        expected = DataFrame(
+            [[1, 2, 1.0, 1.0], [3, 4, exp_value, exp_value]],
+            index=mi,
+            columns=["a", "b", "c", "d"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_multiindex_setitem_axis_set(self):
+        # GH#58116
+        dates = pd.date_range("2001-01-01", freq="D", periods=2)
+        ids = ["i1", "i2", "i3"]
+        index = MultiIndex.from_product([dates, ids], names=["date", "identifier"])
+        df = DataFrame(0.0, index=index, columns=["A", "B"])
+        df.loc(axis=0)["2001-01-01", ["i1", "i3"]] = None
+
+        expected = DataFrame(
+            [
+                [None, None],
+                [0.0, 0.0],
+                [None, None],
+                [0.0, 0.0],
+                [0.0, 0.0],
+                [0.0, 0.0],
+            ],
+            index=index,
+            columns=["A", "B"],
+        )
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_sorted_multiindex_after_union(self):
+        # GH#44752
+        midx = MultiIndex.from_product(
+            [pd.date_range("20110101", periods=2), Index(["a", "b"])]
+        )
+        ser1 = Series(1, index=midx)
+        ser2 = Series(1, index=midx[:2])
+        df = pd.concat([ser1, ser2], axis=1)
+        expected = df.copy()
+        result = df.loc["2011-01-01":"2011-01-02"]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({0: ser1, 1: ser2})
+        result = df.loc["2011-01-01":"2011-01-02"]
+        tm.assert_frame_equal(result, expected)
+
+        df = pd.concat([ser1, ser2.reindex(ser1.index)], axis=1)
+        result = df.loc["2011-01-01":"2011-01-02"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_no_second_level_index(self):
+        # GH#43599
+        df = DataFrame(
+            index=MultiIndex.from_product([list("ab"), list("cd"), list("e")]),
+            columns=["Val"],
+        )
+        res = df.loc[np.s_[:, "c", :]]
+        expected = DataFrame(
+            index=MultiIndex.from_product([list("ab"), list("e")]), columns=["Val"]
+        )
+        tm.assert_frame_equal(res, expected)
+
+    def test_loc_multi_index_key_error(self):
+        # GH 51892
+        df = DataFrame(
+            {
+                (1, 2): ["a", "b", "c"],
+                (1, 3): ["d", "e", "f"],
+                (2, 2): ["g", "h", "i"],
+                (2, 4): ["j", "k", "l"],
+            }
+        )
+        with pytest.raises(KeyError, match=r"(1, 4)"):
+            df.loc[0, (1, 4)]
+
+
+@pytest.mark.parametrize(
+    "indexer, pos",
+    [
+        ([], []),  # empty ok
+        (["A"], slice(3)),
+        (["A", "D"], []),  # "D" isn't present -> raise
+        (["D", "E"], []),  # no values found -> raise
+        (["D"], []),  # same, with single item list: GH 27148
+        (pd.IndexSlice[:, ["foo"]], slice(2, None, 3)),
+        (pd.IndexSlice[:, ["foo", "bah"]], slice(2, None, 3)),
+    ],
+)
+def test_loc_getitem_duplicates_multiindex_missing_indexers(indexer, pos):
+    # GH 7866
+    # multi-index slicing with missing indexers
+    idx = MultiIndex.from_product(
+        [["A", "B", "C"], ["foo", "bar", "baz"]], names=["one", "two"]
+    )
+    ser = Series(np.arange(9, dtype="int64"), index=idx).sort_index()
+    expected = ser.iloc[pos]
+
+    if expected.size == 0 and indexer != []:
+        with pytest.raises(KeyError, match=str(indexer)):
+            ser.loc[indexer]
+    elif indexer == (slice(None), ["foo", "bah"]):
+        # "bah" is not in idx.levels[1], raising KeyError enforced in 2.0
+        with pytest.raises(KeyError, match="'bah'"):
+            ser.loc[indexer]
+    else:
+        result = ser.loc[indexer]
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("columns_indexer", [([], slice(None)), (["foo"], [])])
+def test_loc_getitem_duplicates_multiindex_empty_indexer(columns_indexer):
+    # GH 8737
+    # empty indexer
+    multi_index = MultiIndex.from_product((["foo", "bar", "baz"], ["alpha", "beta"]))
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 6)),
+        index=range(5),
+        columns=multi_index,
+    )
+    df = df.sort_index(level=0, axis=1)
+
+    expected = DataFrame(index=range(5), columns=multi_index.reindex([])[0])
+    result = df.loc[:, columns_indexer]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_getitem_duplicates_multiindex_non_scalar_type_object():
+    # regression from < 0.14.0
+    # GH 7914
+    df = DataFrame(
+        [[np.mean, np.median], ["mean", "median"]],
+        columns=MultiIndex.from_tuples([("functs", "mean"), ("functs", "median")]),
+        index=["function", "name"],
+    )
+    result = df.loc["function", ("functs", "mean")]
+    expected = np.mean
+    assert result == expected
+
+
+def test_loc_getitem_tuple_plus_slice():
+    # GH 671
+    df = DataFrame(
+        {
+            "a": np.arange(10),
+            "b": np.arange(10),
+            "c": np.random.default_rng(2).standard_normal(10),
+            "d": np.random.default_rng(2).standard_normal(10),
+        }
+    ).set_index(["a", "b"])
+    expected = df.loc[0, 0]
+    result = df.loc[(0, 0), :]
+    tm.assert_series_equal(result, expected)
+
+
+def test_loc_getitem_int(frame_random_data_integer_multi_index):
+    df = frame_random_data_integer_multi_index
+    result = df.loc[1]
+    expected = df[-3:]
+    expected.index = expected.index.droplevel(0)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_getitem_int_raises_exception(frame_random_data_integer_multi_index):
+    df = frame_random_data_integer_multi_index
+    with pytest.raises(KeyError, match=r"^3$"):
+        df.loc[3]
+
+
+def test_loc_getitem_lowerdim_corner(multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data
+
+    # test setup - check key not in dataframe
+    with pytest.raises(KeyError, match=r"^\('bar', 'three'\)$"):
+        df.loc[("bar", "three"), "B"]
+
+    # in theory should be inserting in a sorted space????
+    df.loc[("bar", "three"), "B"] = 0
+    expected = 0
+    result = df.sort_index().loc[("bar", "three"), "B"]
+    assert result == expected
+
+
+def test_loc_setitem_single_column_slice():
+    # case from https://github.com/pandas-dev/pandas/issues/27841
+    df = DataFrame(
+        "string",
+        index=list("abcd"),
+        columns=MultiIndex.from_product([["Main"], ("another", "one")]),
+    )
+    df["labels"] = "a"
+    df.loc[:, "labels"] = df.index
+    tm.assert_numpy_array_equal(np.asarray(df["labels"]), np.asarray(df.index))
+
+    # test with non-object block
+    df = DataFrame(
+        np.nan,
+        index=range(4),
+        columns=MultiIndex.from_tuples([("A", "1"), ("A", "2"), ("B", "1")]),
+    )
+    expected = df.copy()
+    df.loc[:, "B"] = np.arange(4)
+    expected.iloc[:, 2] = np.arange(4)
+    tm.assert_frame_equal(df, expected)
+
+
+def test_loc_nan_multiindex(using_infer_string):
+    # GH 5286
+    tups = [
+        ("Good Things", "C", np.nan),
+        ("Good Things", "R", np.nan),
+        ("Bad Things", "C", np.nan),
+        ("Bad Things", "T", np.nan),
+        ("Okay Things", "N", "B"),
+        ("Okay Things", "N", "D"),
+        ("Okay Things", "B", np.nan),
+        ("Okay Things", "D", np.nan),
+    ]
+    df = DataFrame(
+        np.ones((8, 4)),
+        columns=Index(["d1", "d2", "d3", "d4"]),
+        index=MultiIndex.from_tuples(tups, names=["u1", "u2", "u3"]),
+    )
+    result = df.loc["Good Things"].loc["C"]
+    expected = DataFrame(
+        np.ones((1, 4)),
+        index=Index(
+            [np.nan],
+            dtype="object" if not using_infer_string else "str",
+            name="u3",
+        ),
+        columns=Index(["d1", "d2", "d3", "d4"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_period_string_indexing():
+    # GH 9892
+    a = pd.period_range("2013Q1", "2013Q4", freq="Q")
+    i = (1111, 2222, 3333)
+    idx = MultiIndex.from_product((a, i), names=("Period", "CVR"))
+    df = DataFrame(
+        index=idx,
+        columns=(
+            "OMS",
+            "OMK",
+            "RES",
+            "DRIFT_IND",
+            "OEVRIG_IND",
+            "FIN_IND",
+            "VARE_UD",
+            "LOEN_UD",
+            "FIN_UD",
+        ),
+    )
+    result = df.loc[("2013Q1", 1111), "OMS"]
+
+    alt = df.loc[(a[0], 1111), "OMS"]
+    assert np.isnan(alt)
+
+    # Because the resolution of the string matches, it is an exact lookup,
+    #  not a slice
+    assert np.isnan(result)
+
+    alt = df.loc[("2013Q1", 1111), "OMS"]
+    assert np.isnan(alt)
+
+
+def test_loc_datetime_mask_slicing():
+    # GH 16699
+    dt_idx = pd.to_datetime(["2017-05-04", "2017-05-05"])
+    m_idx = MultiIndex.from_product([dt_idx, dt_idx], names=["Idx1", "Idx2"])
+    df = DataFrame(
+        data=[[1, 2], [3, 4], [5, 6], [7, 6]], index=m_idx, columns=["C1", "C2"]
+    )
+    result = df.loc[(dt_idx[0], (df.index.get_level_values(1) > "2017-05-04")), "C1"]
+    expected = Series(
+        [3],
+        name="C1",
+        index=MultiIndex.from_tuples(
+            [(pd.Timestamp("2017-05-04"), pd.Timestamp("2017-05-05"))],
+            names=["Idx1", "Idx2"],
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_loc_datetime_series_tuple_slicing():
+    # https://github.com/pandas-dev/pandas/issues/35858
+    date = pd.Timestamp("2000")
+    ser = Series(
+        1,
+        index=MultiIndex.from_tuples([("a", date)], names=["a", "b"]),
+        name="c",
+    )
+    result = ser.loc[:, [date]]
+    tm.assert_series_equal(result, ser)
+
+
+def test_loc_with_mi_indexer():
+    # https://github.com/pandas-dev/pandas/issues/35351
+    df = DataFrame(
+        data=[["a", 1], ["a", 0], ["b", 1], ["c", 2]],
+        index=MultiIndex.from_tuples(
+            [(0, 1), (1, 0), (1, 1), (1, 1)], names=["index", "date"]
+        ),
+        columns=["author", "price"],
+    )
+    idx = MultiIndex.from_tuples([(0, 1), (1, 1)], names=["index", "date"])
+    result = df.loc[idx, :]
+    expected = DataFrame(
+        [["a", 1], ["b", 1], ["c", 2]],
+        index=MultiIndex.from_tuples([(0, 1), (1, 1), (1, 1)], names=["index", "date"]),
+        columns=["author", "price"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_mi_with_level1_named_0():
+    # GH#37194
+    dti = pd.date_range("2016-01-01", periods=3, tz="US/Pacific")
+
+    ser = Series(range(3), index=dti)
+    df = ser.to_frame()
+    df[1] = dti
+
+    df2 = df.set_index(0, append=True)
+    assert df2.index.names == (None, 0)
+    df2.index.get_loc(dti[0])  # smoke test
+
+    result = df2.loc[dti[0]]
+    expected = df2.iloc[[0]].droplevel(None)
+    tm.assert_frame_equal(result, expected)
+
+    ser2 = df2[1]
+    assert ser2.index.names == (None, 0)
+
+    result = ser2.loc[dti[0]]
+    expected = ser2.iloc[[0]].droplevel(None)
+    tm.assert_series_equal(result, expected)
+
+
+def test_getitem_str_slice():
+    # GH#15928
+    df = DataFrame(
+        [
+            ["20160525 13:30:00.023", "MSFT", "51.95", "51.95"],
+            ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+            ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+            ["20160525 13:30:00.131", "AAPL", "98.61", "98.62"],
+            ["20160525 13:30:00.135", "MSFT", "51.92", "51.95"],
+            ["20160525 13:30:00.135", "AAPL", "98.61", "98.62"],
+        ],
+        columns="time,ticker,bid,ask".split(","),
+    )
+    df2 = df.set_index(["ticker", "time"]).sort_index()
+
+    res = df2.loc[("AAPL", slice("2016-05-25 13:30:00")), :].droplevel(0)
+    expected = df2.loc["AAPL"].loc[slice("2016-05-25 13:30:00"), :]
+    tm.assert_frame_equal(res, expected)
+
+
+def test_3levels_leading_period_index():
+    # GH#24091
+    pi = pd.PeriodIndex(
+        ["20181101 1100", "20181101 1200", "20181102 1300", "20181102 1400"],
+        name="datetime",
+        freq="D",
+    )
+    lev2 = ["A", "A", "Z", "W"]
+    lev3 = ["B", "C", "Q", "F"]
+    mi = MultiIndex.from_arrays([pi, lev2, lev3])
+
+    ser = Series(range(4), index=mi, dtype=np.float64)
+    result = ser.loc[(pi[0], "A", "B")]
+    assert result == 0.0
+
+
+class TestKeyErrorsWithMultiIndex:
+    def test_missing_keys_raises_keyerror(self):
+        # GH#27420 KeyError, not TypeError
+        df = DataFrame(np.arange(12).reshape(4, 3), columns=["A", "B", "C"])
+        df2 = df.set_index(["A", "B"])
+
+        with pytest.raises(KeyError, match="1"):
+            df2.loc[(1, 6)]
+
+    def test_missing_key_raises_keyerror2(self):
+        # GH#21168 KeyError, not "IndexingError: Too many indexers"
+        ser = Series(-1, index=MultiIndex.from_product([[0, 1]] * 2))
+
+        with pytest.raises(KeyError, match=r"\(0, 3\)"):
+            ser.loc[0, 3]
+
+    def test_missing_key_combination(self):
+        # GH: 19556
+        mi = MultiIndex.from_arrays(
+            [
+                np.array(["a", "a", "b", "b"]),
+                np.array(["1", "2", "2", "3"]),
+                np.array(["c", "d", "c", "d"]),
+            ],
+            names=["one", "two", "three"],
+        )
+        df = DataFrame(np.random.default_rng(2).random((4, 3)), index=mi)
+        msg = r"\('b', '1', slice\(None, None, None\)\)"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[("b", "1", slice(None)), :]
+        with pytest.raises(KeyError, match=msg):
+            df.index.get_locs(("b", "1", slice(None)))
+        with pytest.raises(KeyError, match=r"\('b', '1'\)"):
+            df.loc[("b", "1"), :]
+
+
+def test_getitem_loc_commutability(multiindex_year_month_day_dataframe_random_data):
+    df = multiindex_year_month_day_dataframe_random_data
+    ser = df["A"]
+    result = ser[2000, 5]
+    expected = df.loc[2000, 5]["A"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_loc_with_nan():
+    # GH: 27104
+    df = DataFrame(
+        {"col": [1, 2, 5], "ind1": ["a", "d", np.nan], "ind2": [1, 4, 5]}
+    ).set_index(["ind1", "ind2"])
+    result = df.loc[["a"]]
+    expected = DataFrame(
+        {"col": [1]}, index=MultiIndex.from_tuples([("a", 1)], names=["ind1", "ind2"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.loc["a"]
+    expected = DataFrame({"col": [1]}, index=Index([1], name="ind2"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_getitem_non_found_tuple():
+    # GH: 25236
+    df = DataFrame([[1, 2, 3, 4]], columns=["a", "b", "c", "d"]).set_index(
+        ["a", "b", "c"]
+    )
+    with pytest.raises(KeyError, match=r"\(2\.0, 2\.0, 3\.0\)"):
+        df.loc[(2.0, 2.0, 3.0)]
+
+
+def test_get_loc_datetime_index():
+    # GH#24263
+    index = pd.date_range("2001-01-01", periods=100)
+    mi = MultiIndex.from_arrays([index])
+    # Check if get_loc matches for Index and MultiIndex
+    assert mi.get_loc("2001-01") == slice(0, 31, None)
+    assert index.get_loc("2001-01") == slice(0, 31, None)
+
+    loc = mi[::2].get_loc("2001-01")
+    expected = index[::2].get_loc("2001-01")
+    assert loc == expected
+
+    loc = mi.repeat(2).get_loc("2001-01")
+    expected = index.repeat(2).get_loc("2001-01")
+    assert loc == expected
+
+    loc = mi.append(mi).get_loc("2001-01")
+    expected = index.append(index).get_loc("2001-01")
+    # TODO: standardize return type for MultiIndex.get_loc
+    tm.assert_numpy_array_equal(loc.nonzero()[0], expected)
+
+
+def test_loc_setitem_indexer_differently_ordered():
+    # GH#34603
+    mi = MultiIndex.from_product([["a", "b"], [0, 1]])
+    df = DataFrame([[1, 2], [3, 4], [5, 6], [7, 8]], index=mi)
+
+    indexer = ("a", [1, 0])
+    df.loc[indexer, :] = np.array([[9, 10], [11, 12]])
+    expected = DataFrame([[11, 12], [9, 10], [5, 6], [7, 8]], index=mi)
+    tm.assert_frame_equal(df, expected)
+
+
+def test_loc_getitem_index_differently_ordered_slice_none():
+    # GH#31330
+    df = DataFrame(
+        [[1, 2], [3, 4], [5, 6], [7, 8]],
+        index=[["a", "a", "b", "b"], [1, 2, 1, 2]],
+        columns=["a", "b"],
+    )
+    result = df.loc[(slice(None), [2, 1]), :]
+    expected = DataFrame(
+        [[3, 4], [7, 8], [1, 2], [5, 6]],
+        index=[["a", "b", "a", "b"], [2, 2, 1, 1]],
+        columns=["a", "b"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("indexer", [[1, 2, 7, 6, 2, 3, 8, 7], [1, 2, 7, 6, 3, 8]])
+def test_loc_getitem_index_differently_ordered_slice_none_duplicates(indexer):
+    # GH#40978
+    df = DataFrame(
+        [1] * 8,
+        index=MultiIndex.from_tuples(
+            [(1, 1), (1, 2), (1, 7), (1, 6), (2, 2), (2, 3), (2, 8), (2, 7)]
+        ),
+        columns=["a"],
+    )
+    result = df.loc[(slice(None), indexer), :]
+    expected = DataFrame(
+        [1] * 8,
+        index=[[1, 1, 2, 1, 2, 1, 2, 2], [1, 2, 2, 7, 7, 6, 3, 8]],
+        columns=["a"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.loc[df.index.isin(indexer, level=1), :]
+    tm.assert_frame_equal(result, df)
+
+
+def test_loc_getitem_drops_levels_for_one_row_dataframe():
+    # GH#10521 "x" and "z" are both scalar indexing, so those levels are dropped
+    mi = MultiIndex.from_arrays([["x"], ["y"], ["z"]], names=["a", "b", "c"])
+    df = DataFrame({"d": [0]}, index=mi)
+    expected = df.droplevel([0, 2])
+    result = df.loc["x", :, "z"]
+    tm.assert_frame_equal(result, expected)
+
+    ser = Series([0], index=mi)
+    result = ser.loc["x", :, "z"]
+    expected = Series([0], index=Index(["y"], name="b"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_mi_columns_loc_list_label_order():
+    # GH 10710
+    cols = MultiIndex.from_product([["A", "B", "C"], [1, 2]])
+    df = DataFrame(np.zeros((5, 6)), columns=cols)
+    result = df.loc[:, ["B", "A"]]
+    expected = DataFrame(
+        np.zeros((5, 4)),
+        columns=MultiIndex.from_tuples([("B", 1), ("B", 2), ("A", 1), ("A", 2)]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_mi_partial_indexing_list_raises():
+    # GH 13501
+    frame = DataFrame(
+        np.arange(12).reshape((4, 3)),
+        index=[["a", "a", "b", "b"], [1, 2, 1, 2]],
+        columns=[["Ohio", "Ohio", "Colorado"], ["Green", "Red", "Green"]],
+    )
+    frame.index.names = ["key1", "key2"]
+    frame.columns.names = ["state", "color"]
+    with pytest.raises(KeyError, match="\\[2\\] not in index"):
+        frame.loc[["b", 2], "Colorado"]
+
+
+def test_mi_indexing_list_nonexistent_raises():
+    # GH 15452
+    s = Series(range(4), index=MultiIndex.from_product([[1, 2], ["a", "b"]]))
+    with pytest.raises(KeyError, match="\\['not' 'found'\\] not in index"):
+        s.loc[["not", "found"]]
+
+
+def test_mi_add_cell_missing_row_non_unique():
+    # GH 16018
+    result = DataFrame(
+        [[1, 2, 5, 6], [3, 4, 7, 8]],
+        index=["a", "a"],
+        columns=MultiIndex.from_product([[1, 2], ["A", "B"]]),
+    )
+    result.loc["c"] = -1
+    result.loc["c", (1, "A")] = 3
+    result.loc["d", (1, "A")] = 3
+    expected = DataFrame(
+        [
+            [1.0, 2.0, 5.0, 6.0],
+            [3.0, 4.0, 7.0, 8.0],
+            [3.0, -1.0, -1, -1],
+            [3.0, np.nan, np.nan, np.nan],
+        ],
+        index=["a", "a", "c", "d"],
+        columns=MultiIndex.from_product([[1, 2], ["A", "B"]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_get_scalar_casting_to_float():
+    # GH#41369
+    df = DataFrame(
+        {"a": 1.0, "b": 2}, index=MultiIndex.from_arrays([[3], [4]], names=["c", "d"])
+    )
+    result = df.loc[(3, 4), "b"]
+    assert result == 2
+    assert isinstance(result, np.int64)
+    result = df.loc[[(3, 4)], "b"].iloc[0]
+    assert result == 2
+    assert isinstance(result, np.int64)
+
+
+def test_loc_empty_single_selector_with_names():
+    # GH 19517
+    idx = MultiIndex.from_product([["a", "b"], ["A", "B"]], names=[1, 0])
+    s2 = Series(index=idx, dtype=np.float64)
+    result = s2.loc["a"]
+    expected = Series([np.nan, np.nan], index=Index(["A", "B"], name=0))
+    tm.assert_series_equal(result, expected)
+
+
+def test_loc_keyerror_rightmost_key_missing():
+    # GH 20951
+
+    df = DataFrame(
+        {
+            "A": [100, 100, 200, 200, 300, 300],
+            "B": [10, 10, 20, 21, 31, 33],
+            "C": range(6),
+        }
+    )
+    df = df.set_index(["A", "B"])
+    with pytest.raises(KeyError, match="^1$"):
+        df.loc[(100, 1)]
+
+
+def test_multindex_series_loc_with_tuple_label():
+    # GH#43908
+    mi = MultiIndex.from_tuples([(1, 2), (3, (4, 5))])
+    ser = Series([1, 2], index=mi)
+    result = ser.loc[(3, (4, 5))]
+    assert result == 2
diff --git a/pandas/pandas/tests/indexing/multiindex/test_multiindex.py b/pandas/pandas/tests/indexing/multiindex/test_multiindex.py
new file mode 100644
index 0000000000000..7140ad7d1e9f5
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_multiindex.py
@@ -0,0 +1,251 @@
+import numpy as np
+import pytest
+
+import pandas._libs.index as libindex
+
+import pandas as pd
+from pandas import (
+    CategoricalDtype,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.arrays.boolean import BooleanDtype
+
+
+class TestMultiIndexBasic:
+    def test_multiindex_perf_warn(self, performance_warning):
+        df = DataFrame(
+            {
+                "jim": [0, 0, 1, 1],
+                "joe": ["x", "x", "z", "y"],
+                "jolie": np.random.default_rng(2).random(4),
+            }
+        ).set_index(["jim", "joe"])
+
+        with tm.assert_produces_warning(performance_warning):
+            df.loc[(1, "z")]
+
+        df = df.iloc[[2, 1, 3, 0]]
+        with tm.assert_produces_warning(performance_warning):
+            df.loc[(0,)]
+
+    @pytest.mark.parametrize("offset", [-5, 5])
+    def test_indexing_over_hashtable_size_cutoff(self, monkeypatch, offset):
+        size_cutoff = 20
+        n = size_cutoff + offset
+
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+            s = Series(np.arange(n), MultiIndex.from_arrays((["a"] * n, np.arange(n))))
+
+            # hai it works!
+            assert s[("a", 5)] == 5
+            assert s[("a", 6)] == 6
+            assert s[("a", 7)] == 7
+
+    def test_multi_nan_indexing(self):
+        # GH 3588
+        df = DataFrame(
+            {
+                "a": ["R1", "R2", np.nan, "R4"],
+                "b": ["C1", "C2", "C3", "C4"],
+                "c": [10, 15, np.nan, 20],
+            }
+        )
+        result = df.set_index(["a", "b"], drop=False)
+        expected = DataFrame(
+            {
+                "a": ["R1", "R2", np.nan, "R4"],
+                "b": ["C1", "C2", "C3", "C4"],
+                "c": [10, 15, np.nan, 20],
+            },
+            index=[
+                Index(["R1", "R2", np.nan, "R4"], name="a"),
+                Index(["C1", "C2", "C3", "C4"], name="b"),
+            ],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_exclusive_nat_column_indexing(self):
+        # GH 38025
+        # test multi indexing when one column exclusively contains NaT values
+        df = DataFrame(
+            {
+                "a": [pd.NaT, pd.NaT, pd.NaT, pd.NaT],
+                "b": ["C1", "C2", "C3", "C4"],
+                "c": [10, 15, np.nan, 20],
+            }
+        )
+        df = df.set_index(["a", "b"])
+        expected = DataFrame(
+            {
+                "c": [10, 15, np.nan, 20],
+            },
+            index=[
+                Index([pd.NaT, pd.NaT, pd.NaT, pd.NaT], name="a"),
+                Index(["C1", "C2", "C3", "C4"], name="b"),
+            ],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_nested_tuples_duplicates(self):
+        # GH#30892
+
+        dti = pd.to_datetime(["20190101", "20190101", "20190102"])
+        idx = Index(["a", "a", "c"])
+        mi = MultiIndex.from_arrays([dti, idx], names=["index1", "index2"])
+
+        df = DataFrame({"c1": [1, 2, 3], "c2": [np.nan, np.nan, np.nan]}, index=mi)
+
+        expected = DataFrame({"c1": df["c1"], "c2": [1.0, 1.0, np.nan]}, index=mi)
+
+        df2 = df.copy(deep=True)
+        df2.loc[(dti[0], "a"), "c2"] = 1.0
+        tm.assert_frame_equal(df2, expected)
+
+        df3 = df.copy(deep=True)
+        df3.loc[[(dti[0], "a")], "c2"] = 1.0
+        tm.assert_frame_equal(df3, expected)
+
+    def test_multiindex_with_datatime_level_preserves_freq(self):
+        # https://github.com/pandas-dev/pandas/issues/35563
+        idx = Index(range(2), name="A")
+        dti = pd.date_range("2020-01-01", periods=7, freq="D", name="B")
+        mi = MultiIndex.from_product([idx, dti])
+        df = DataFrame(np.random.default_rng(2).standard_normal((14, 2)), index=mi)
+        result = df.loc[0].index
+        tm.assert_index_equal(result, dti)
+        assert result.freq == dti.freq
+
+    def test_multiindex_complex(self):
+        # GH#42145
+        complex_data = [1 + 2j, 4 - 3j, 10 - 1j]
+        non_complex_data = [3, 4, 5]
+        result = DataFrame(
+            {
+                "x": complex_data,
+                "y": non_complex_data,
+                "z": non_complex_data,
+            }
+        )
+        result.set_index(["x", "y"], inplace=True)
+        expected = DataFrame(
+            {"z": non_complex_data},
+            index=MultiIndex.from_arrays(
+                [complex_data, non_complex_data],
+                names=("x", "y"),
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_rename_multiindex_with_duplicates(self):
+        # GH 38015
+        mi = MultiIndex.from_tuples([("A", "cat"), ("B", "cat"), ("B", "cat")])
+        df = DataFrame(index=mi)
+        df = df.rename(index={"A": "Apple"}, level=0)
+
+        mi2 = MultiIndex.from_tuples([("Apple", "cat"), ("B", "cat"), ("B", "cat")])
+        expected = DataFrame(index=mi2)
+        tm.assert_frame_equal(df, expected)
+
+    def test_series_align_multiindex_with_nan_overlap_only(self):
+        # GH 38439
+        mi1 = MultiIndex.from_arrays([[81.0, np.nan], [np.nan, np.nan]])
+        mi2 = MultiIndex.from_arrays([[np.nan, 82.0], [np.nan, np.nan]])
+        ser1 = Series([1, 2], index=mi1)
+        ser2 = Series([1, 2], index=mi2)
+        result1, result2 = ser1.align(ser2)
+
+        mi = MultiIndex.from_arrays([[81.0, 82.0, np.nan], [np.nan, np.nan, np.nan]])
+        expected1 = Series([1.0, np.nan, 2.0], index=mi)
+        expected2 = Series([np.nan, 2.0, 1.0], index=mi)
+
+        tm.assert_series_equal(result1, expected1)
+        tm.assert_series_equal(result2, expected2)
+
+    def test_series_align_multiindex_with_nan(self):
+        # GH 38439
+        mi1 = MultiIndex.from_arrays([[81.0, np.nan], [np.nan, np.nan]])
+        mi2 = MultiIndex.from_arrays([[np.nan, 81.0], [np.nan, np.nan]])
+        ser1 = Series([1, 2], index=mi1)
+        ser2 = Series([1, 2], index=mi2)
+        result1, result2 = ser1.align(ser2)
+
+        mi = MultiIndex.from_arrays([[81.0, np.nan], [np.nan, np.nan]])
+        expected1 = Series([1, 2], index=mi)
+        expected2 = Series([2, 1], index=mi)
+
+        tm.assert_series_equal(result1, expected1)
+        tm.assert_series_equal(result2, expected2)
+
+    def test_nunique_smoke(self):
+        # GH 34019
+        n = DataFrame([[1, 2], [1, 2]]).set_index([0, 1]).index.nunique()
+        assert n == 1
+
+    def test_multiindex_repeated_keys(self):
+        # GH19414
+        tm.assert_series_equal(
+            Series([1, 2], MultiIndex.from_arrays([["a", "b"]])).loc[
+                ["a", "a", "b", "b"]
+            ],
+            Series([1, 1, 2, 2], MultiIndex.from_arrays([["a", "a", "b", "b"]])),
+        )
+
+    def test_multiindex_with_na_missing_key(self):
+        # GH46173
+        df = DataFrame.from_dict(
+            {
+                ("foo",): [1, 2, 3],
+                ("bar",): [5, 6, 7],
+                (None,): [8, 9, 0],
+            }
+        )
+        with pytest.raises(KeyError, match="missing_key"):
+            df[[("missing_key",)]]
+
+    def test_multiindex_dtype_preservation(self):
+        # GH51261
+        columns = MultiIndex.from_tuples([("A", "B")], names=["lvl1", "lvl2"])
+        df = DataFrame(["value"], columns=columns).astype("category")
+        df_no_multiindex = df["A"]
+        assert isinstance(df_no_multiindex["B"].dtype, CategoricalDtype)
+
+        # geopandas 1763 analogue
+        df = DataFrame(
+            [[1, 0], [0, 1]],
+            columns=[
+                ["foo", "foo"],
+                ["location", "location"],
+                ["x", "y"],
+            ],
+        ).assign(bools=Series([True, False], dtype="boolean"))
+        assert isinstance(df["bools"].dtype, BooleanDtype)
+
+    def test_multiindex_from_tuples_with_nan(self):
+        # GH#23578
+        result = MultiIndex.from_tuples([("a", "b", "c"), np.nan, ("d", "", "")])
+        expected = MultiIndex.from_tuples(
+            [("a", "b", "c"), (np.nan, np.nan, np.nan), ("d", "", "")]
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("operation", ["div", "mul", "add", "sub"])
+    def test_groupyby_rename_categories_operation_with_multiindex(self, operation):
+        # GH#51500
+        data = DataFrame(
+            [["C", "B", "B"], ["B", "A", "A"], ["B", "A", "B"]], columns=["0", "1", "2"]
+        )
+        data["0"] = data["0"].astype("category")
+        data["0"] = data["0"].cat.rename_categories({"C": "B", "B": "C"})
+
+        a = data.groupby(by=["0", "1"])["2"].value_counts()
+        b = data.groupby(by=["0", "1"]).size()
+
+        result = getattr(a, operation)(b)
+        expected = getattr(a, operation)(b.sort_index(ascending=False))
+
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_partial.py b/pandas/pandas/tests/indexing/multiindex/test_partial.py
new file mode 100644
index 0000000000000..dbfabf7666d25
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_partial.py
@@ -0,0 +1,255 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    MultiIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestMultiIndexPartial:
+    def test_getitem_partial_int(self):
+        # GH 12416
+        # with single item
+        l1 = [10, 20]
+        l2 = ["a", "b"]
+        df = DataFrame(index=range(2), columns=MultiIndex.from_product([l1, l2]))
+        expected = DataFrame(index=range(2), columns=l2)
+        result = df[20]
+        tm.assert_frame_equal(result, expected)
+
+        # with list
+        expected = DataFrame(
+            index=range(2), columns=MultiIndex.from_product([l1[1:], l2])
+        )
+        result = df[[20]]
+        tm.assert_frame_equal(result, expected)
+
+        # missing item:
+        with pytest.raises(KeyError, match="1"):
+            df[1]
+        with pytest.raises(KeyError, match=r"'\[1\] not in index'"):
+            df[[1]]
+
+    def test_series_slice_partial(self):
+        pass
+
+    def test_xs_partial(
+        self,
+        multiindex_dataframe_random_data,
+        multiindex_year_month_day_dataframe_random_data,
+    ):
+        frame = multiindex_dataframe_random_data
+        ymd = multiindex_year_month_day_dataframe_random_data
+        result = frame.xs("foo")
+        result2 = frame.loc["foo"]
+        expected = frame.T["foo"].T
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result, result2)
+
+        result = ymd.xs((2000, 4))
+        expected = ymd.loc[2000, 4]
+        tm.assert_frame_equal(result, expected)
+
+        # ex from #1796
+        index = MultiIndex(
+            levels=[["foo", "bar"], ["one", "two"], [-1, 1]],
+            codes=[
+                [0, 0, 0, 0, 1, 1, 1, 1],
+                [0, 0, 1, 1, 0, 0, 1, 1],
+                [0, 1, 0, 1, 0, 1, 0, 1],
+            ],
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)),
+            index=index,
+            columns=list("abcd"),
+        )
+
+        result = df.xs(("foo", "one"))
+        expected = df.loc["foo", "one"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_partial(self, multiindex_year_month_day_dataframe_random_data):
+        ymd = multiindex_year_month_day_dataframe_random_data
+        ymd = ymd.T
+        result = ymd[2000, 2]
+
+        expected = ymd.reindex(columns=ymd.columns[ymd.columns.codes[1] == 1])
+        expected.columns = expected.columns.droplevel(0).droplevel(0)
+        tm.assert_frame_equal(result, expected)
+
+    def test_fancy_slice_partial(
+        self,
+        multiindex_dataframe_random_data,
+        multiindex_year_month_day_dataframe_random_data,
+    ):
+        frame = multiindex_dataframe_random_data
+        result = frame.loc["bar":"baz"]
+        expected = frame[3:7]
+        tm.assert_frame_equal(result, expected)
+
+        ymd = multiindex_year_month_day_dataframe_random_data
+        result = ymd.loc[(2000, 2) : (2000, 4)]
+        lev = ymd.index.codes[1]
+        expected = ymd[(lev >= 1) & (lev <= 3)]
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_partial_column_select(self):
+        idx = MultiIndex(
+            codes=[[0, 0, 0], [0, 1, 1], [1, 0, 1]],
+            levels=[["a", "b"], ["x", "y"], ["p", "q"]],
+        )
+        df = DataFrame(np.random.default_rng(2).random((3, 2)), index=idx)
+
+        result = df.loc[("a", "y"), :]
+        expected = df.loc[("a", "y")]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[("a", "y"), [1, 0]]
+        expected = df.loc[("a", "y")][[1, 0]]
+        tm.assert_frame_equal(result, expected)
+
+        with pytest.raises(KeyError, match=r"\('a', 'foo'\)"):
+            df.loc[("a", "foo"), :]
+
+    def test_partial_set(
+        self,
+        multiindex_year_month_day_dataframe_random_data,
+    ):
+        # GH #397
+        ymd = multiindex_year_month_day_dataframe_random_data
+        df = ymd.copy()
+        exp = ymd.copy()
+        df.loc[2000, 4] = 0
+        exp.iloc[65:85] = 0
+        tm.assert_frame_equal(df, exp)
+
+        with tm.raises_chained_assignment_error():
+            df["A"].loc[2000, 4] = 1
+        df.loc[(2000, 4), "A"] = 1
+        exp.iloc[65:85, 0] = 1
+        tm.assert_frame_equal(df, exp)
+
+        df.loc[2000] = 5
+        exp.iloc[:100] = 5
+        tm.assert_frame_equal(df, exp)
+
+        # this works...for now
+        with tm.raises_chained_assignment_error():
+            df["A"].iloc[14] = 5
+        assert df["A"].iloc[14] == exp["A"].iloc[14]
+
+    @pytest.mark.parametrize("dtype", [int, float])
+    def test_getitem_intkey_leading_level(
+        self, multiindex_year_month_day_dataframe_random_data, dtype
+    ):
+        # GH#33355 dont fall-back to positional when leading level is int
+        ymd = multiindex_year_month_day_dataframe_random_data
+        levels = ymd.index.levels
+        ymd.index = ymd.index.set_levels([levels[0].astype(dtype)] + levels[1:])
+        ser = ymd["A"]
+        mi = ser.index
+        assert isinstance(mi, MultiIndex)
+        if dtype is int:
+            assert mi.levels[0].dtype == np.dtype(int)
+        else:
+            assert mi.levels[0].dtype == np.float64
+
+        assert 14 not in mi.levels[0]
+        assert not mi.levels[0]._should_fallback_to_positional
+        assert not mi._should_fallback_to_positional
+
+        with pytest.raises(KeyError, match="14"):
+            ser[14]
+
+    # ---------------------------------------------------------------------
+
+    def test_setitem_multiple_partial(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        expected = frame.copy()
+        result = frame.copy()
+        result.loc[["foo", "bar"]] = 0
+        expected.loc["foo"] = 0
+        expected.loc["bar"] = 0
+        tm.assert_frame_equal(result, expected)
+
+        expected = frame.copy()
+        result = frame.copy()
+        result.loc["foo":"bar"] = 0
+        expected.loc["foo"] = 0
+        expected.loc["bar"] = 0
+        tm.assert_frame_equal(result, expected)
+
+        expected = frame["A"].copy()
+        result = frame["A"].copy()
+        result.loc[["foo", "bar"]] = 0
+        expected.loc["foo"] = 0
+        expected.loc["bar"] = 0
+        tm.assert_series_equal(result, expected)
+
+        expected = frame["A"].copy()
+        result = frame["A"].copy()
+        result.loc["foo":"bar"] = 0
+        expected.loc["foo"] = 0
+        expected.loc["bar"] = 0
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "indexer, exp_idx, exp_values",
+        [
+            (
+                slice("2019-2", None),
+                DatetimeIndex(["2019-02-01"], dtype="M8[ns]"),
+                [2, 3],
+            ),
+            (
+                slice(None, "2019-2"),
+                date_range("2019", periods=2, freq="MS"),
+                [0, 1, 2, 3],
+            ),
+        ],
+    )
+    def test_partial_getitem_loc_datetime(self, indexer, exp_idx, exp_values):
+        # GH: 25165
+        date_idx = date_range("2019", periods=2, freq="MS")
+        df = DataFrame(
+            list(range(4)),
+            index=MultiIndex.from_product([date_idx, [0, 1]], names=["x", "y"]),
+        )
+        expected = DataFrame(
+            exp_values,
+            index=MultiIndex.from_product([exp_idx, [0, 1]], names=["x", "y"]),
+        )
+        result = df[indexer]
+        tm.assert_frame_equal(result, expected)
+        result = df.loc[indexer]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc(axis=0)[indexer]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[indexer, :]
+        tm.assert_frame_equal(result, expected)
+
+        df2 = df.swaplevel(0, 1).sort_index()
+        expected = expected.swaplevel(0, 1).sort_index()
+
+        result = df2.loc[:, indexer, :]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_loc_getitem_partial_both_axis():
+    # gh-12660
+    iterables = [["a", "b"], [2, 1]]
+    columns = MultiIndex.from_product(iterables, names=["col1", "col2"])
+    rows = MultiIndex.from_product(iterables, names=["row1", "row2"])
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((4, 4)), index=rows, columns=columns
+    )
+    expected = df.iloc[:2, 2:].droplevel("row1").droplevel("col1", axis=1)
+    result = df.loc["a", "b"]
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_setitem.py b/pandas/pandas/tests/indexing/multiindex/test_setitem.py
new file mode 100644
index 0000000000000..d732cb4d7fbbc
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_setitem.py
@@ -0,0 +1,535 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    date_range,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+
+def assert_equal(a, b):
+    assert a == b
+
+
+class TestMultiIndexSetItem:
+    def check(self, target, indexers, value, compare_fn=assert_equal, expected=None):
+        target.loc[indexers] = value
+        result = target.loc[indexers]
+        if expected is None:
+            expected = value
+        compare_fn(result, expected)
+
+    def test_setitem_multiindex(self):
+        # GH#7190
+        cols = ["A", "w", "l", "a", "x", "X", "d", "profit"]
+        index = MultiIndex.from_product(
+            [np.arange(0, 100), np.arange(0, 80)], names=["time", "firm"]
+        )
+        t, n = 0, 2
+
+        df = DataFrame(
+            np.nan,
+            columns=cols,
+            index=index,
+        )
+        self.check(target=df, indexers=((t, n), "X"), value=0)
+
+        df = DataFrame(-999, columns=cols, index=index)
+        self.check(target=df, indexers=((t, n), "X"), value=1)
+
+        df = DataFrame(columns=cols, index=index)
+        self.check(target=df, indexers=((t, n), "X"), value=2)
+
+        # gh-7218: assigning with 0-dim arrays
+        df = DataFrame(-999, columns=cols, index=index)
+        self.check(
+            target=df,
+            indexers=((t, n), "X"),
+            value=np.array(3),
+            expected=3,
+        )
+
+    def test_setitem_multiindex2(self):
+        # GH#5206
+        df = DataFrame(
+            np.arange(25).reshape(5, 5), columns="A,B,C,D,E".split(","), dtype=float
+        )
+        df["F"] = 99
+        row_selection = df["A"] % 2 == 0
+        col_selection = ["B", "C"]
+        df.loc[row_selection, col_selection] = df["F"]
+        output = DataFrame(99.0, index=[0, 2, 4], columns=["B", "C"])
+        tm.assert_frame_equal(df.loc[row_selection, col_selection], output)
+        self.check(
+            target=df,
+            indexers=(row_selection, col_selection),
+            value=df["F"],
+            compare_fn=tm.assert_frame_equal,
+            expected=output,
+        )
+
+    def test_setitem_multiindex3(self):
+        # GH#11372
+        idx = MultiIndex.from_product(
+            [["A", "B", "C"], date_range("2015-01-01", "2015-04-01", freq="MS")]
+        )
+        cols = MultiIndex.from_product(
+            [["foo", "bar"], date_range("2016-01-01", "2016-02-01", freq="MS")]
+        )
+
+        df = DataFrame(
+            np.random.default_rng(2).random((12, 4)), index=idx, columns=cols
+        )
+
+        subidx = MultiIndex.from_arrays(
+            [["A", "A"], date_range("2015-01-01", "2015-02-01", freq="MS")]
+        )
+        subcols = MultiIndex.from_arrays(
+            [["foo", "foo"], date_range("2016-01-01", "2016-02-01", freq="MS")]
+        )
+
+        vals = DataFrame(
+            np.random.default_rng(2).random((2, 2)), index=subidx, columns=subcols
+        )
+        self.check(
+            target=df,
+            indexers=(subidx, subcols),
+            value=vals,
+            compare_fn=tm.assert_frame_equal,
+        )
+        # set all columns
+        vals = DataFrame(
+            np.random.default_rng(2).random((2, 4)), index=subidx, columns=cols
+        )
+        self.check(
+            target=df,
+            indexers=(subidx, slice(None, None, None)),
+            value=vals,
+            compare_fn=tm.assert_frame_equal,
+        )
+        # identity
+        copy = df.copy()
+        self.check(
+            target=df,
+            indexers=(df.index, df.columns),
+            value=df,
+            compare_fn=tm.assert_frame_equal,
+            expected=copy,
+        )
+
+    def test_multiindex_setitem(self):
+        # GH 3738
+        # setting with a multi-index right hand side
+        arrays = [
+            np.array(["bar", "bar", "baz", "qux", "qux", "bar"]),
+            np.array(["one", "two", "one", "one", "two", "one"]),
+            np.arange(0, 6, 1),
+        ]
+
+        df_orig = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 3)),
+            index=arrays,
+            columns=["A", "B", "C"],
+        ).sort_index()
+
+        expected = df_orig.loc[["bar"]] * 2
+        df = df_orig.copy()
+        df.loc[["bar"]] *= 2
+        tm.assert_frame_equal(df.loc[["bar"]], expected)
+
+        # raise because these have differing levels
+        msg = "cannot align on a multi-index with out specifying the join levels"
+        with pytest.raises(TypeError, match=msg):
+            df.loc["bar"] *= 2
+
+    def test_multiindex_setitem2(self):
+        # from SO
+        # https://stackoverflow.com/questions/24572040/pandas-access-the-level-of-multiindex-for-inplace-operation
+        df_orig = DataFrame.from_dict(
+            {
+                "price": {
+                    ("DE", "Coal", "Stock"): 2,
+                    ("DE", "Gas", "Stock"): 4,
+                    ("DE", "Elec", "Demand"): 1,
+                    ("FR", "Gas", "Stock"): 5,
+                    ("FR", "Solar", "SupIm"): 0,
+                    ("FR", "Wind", "SupIm"): 0,
+                }
+            }
+        )
+        df_orig.index = MultiIndex.from_tuples(
+            df_orig.index, names=["Sit", "Com", "Type"]
+        )
+
+        expected = df_orig.copy()
+        expected.iloc[[0, 1, 3]] *= 2
+
+        idx = pd.IndexSlice
+        df = df_orig.copy()
+        df.loc[idx[:, :, "Stock"], :] *= 2
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[idx[:, :, "Stock"], "price"] *= 2
+        tm.assert_frame_equal(df, expected)
+
+    def test_multiindex_assignment(self):
+        # GH3777 part 2
+
+        # mixed dtype
+        df = DataFrame(
+            np.random.default_rng(2).integers(5, 10, size=9).reshape(3, 3),
+            columns=list("abc"),
+            index=[[4, 4, 8], [8, 10, 12]],
+        )
+        df["d"] = np.nan
+        arr = np.array([0.0, 1.0])
+
+        df.loc[4, "d"] = arr
+        tm.assert_series_equal(df.loc[4, "d"], Series(arr, index=[8, 10], name="d"))
+
+    def test_multiindex_assignment_single_dtype(self):
+        # GH3777 part 2b
+        # single dtype
+        arr = np.array([0.0, 1.0])
+
+        df = DataFrame(
+            np.random.default_rng(2).integers(5, 10, size=9).reshape(3, 3),
+            columns=list("abc"),
+            index=[[4, 4, 8], [8, 10, 12]],
+            dtype=np.int64,
+        )
+
+        # arr can be losslessly cast to int, so this setitem is inplace
+        df.loc[4, "c"] = arr
+        exp = Series(arr, index=[8, 10], name="c", dtype="int64")
+        result = df.loc[4, "c"]
+        tm.assert_series_equal(result, exp)
+
+        # arr + 0.5 cannot be cast losslessly to int, so we upcast
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[4, "c"] = arr + 0.5
+        # Upcast so that we can add .5
+        df = df.astype({"c": "float64"})
+        df.loc[4, "c"] = arr + 0.5
+
+        # scalar ok
+        df.loc[4, "c"] = 10
+        exp = Series(10, index=[8, 10], name="c", dtype="float64")
+        tm.assert_series_equal(df.loc[4, "c"], exp)
+
+        # invalid assignments
+        msg = "Must have equal len keys and value when setting with an iterable"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[4, "c"] = [0, 1, 2, 3]
+
+        with pytest.raises(ValueError, match=msg):
+            df.loc[4, "c"] = [0]
+
+        # But with a length-1 listlike column indexer this behaves like
+        #  `df.loc[4, "c"] = 0
+        df.loc[4, ["c"]] = [0]
+        assert (df.loc[4, "c"] == 0).all()
+
+    def test_groupby_example(self):
+        # groupby example
+        NUM_ROWS = 100
+        NUM_COLS = 10
+        col_names = ["A" + num for num in map(str, np.arange(NUM_COLS).tolist())]
+        index_cols = col_names[:5]
+
+        df = DataFrame(
+            np.random.default_rng(2).integers(5, size=(NUM_ROWS, NUM_COLS)),
+            dtype=np.int64,
+            columns=col_names,
+        )
+        df = df.set_index(index_cols).sort_index()
+        grp = df.groupby(level=index_cols[:4])
+        df["new_col"] = np.nan
+
+        # we are actually operating on a copy here
+        # but in this case, that's ok
+        for name, df2 in grp:
+            new_vals = np.arange(df2.shape[0])
+            df.loc[name, "new_col"] = new_vals
+
+    def test_series_setitem(self, multiindex_year_month_day_dataframe_random_data):
+        ymd = multiindex_year_month_day_dataframe_random_data
+        s = ymd["A"]
+
+        s[2000, 3] = np.nan
+        assert isna(s.values[42:65]).all()
+        assert notna(s.values[:42]).all()
+        assert notna(s.values[65:]).all()
+
+        s[2000, 3, 10] = np.nan
+        assert isna(s.iloc[49])
+
+        with pytest.raises(KeyError, match="49"):
+            # GH#33355 dont fall-back to positional when leading level is int
+            s[49]
+
+    def test_frame_getitem_setitem_boolean(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        df = frame.T.copy()
+        values = df.values.copy()
+
+        result = df[df > 0]
+        expected = df.where(df > 0)
+        tm.assert_frame_equal(result, expected)
+
+        df[df > 0] = 5
+        values[values > 0] = 5
+        tm.assert_almost_equal(df.values, values)
+
+        df[df == 5] = 0
+        values[values == 5] = 0
+        tm.assert_almost_equal(df.values, values)
+
+        # a df that needs alignment first
+        df[df[:-1] < 0] = 2
+        np.putmask(values[:-1], values[:-1] < 0, 2)
+        tm.assert_almost_equal(df.values, values)
+
+        with pytest.raises(TypeError, match="boolean values only"):
+            df[df * 0] = 2
+
+    def test_frame_getitem_setitem_multislice(self):
+        levels = [["t1", "t2"], ["a", "b", "c"]]
+        codes = [[0, 0, 0, 1, 1], [0, 1, 2, 0, 1]]
+        midx = MultiIndex(codes=codes, levels=levels, names=[None, "id"])
+        df = DataFrame({"value": [1, 2, 3, 7, 8]}, index=midx)
+
+        result = df.loc[:, "value"]
+        tm.assert_series_equal(df["value"], result)
+
+        result = df.loc[df.index[1:3], "value"]
+        tm.assert_series_equal(df["value"][1:3], result)
+
+        result = df.loc[:, :]
+        tm.assert_frame_equal(df, result)
+
+        result = df
+        df.loc[:, "value"] = 10
+        result["value"] = 10
+        tm.assert_frame_equal(df, result)
+
+        df.loc[:, :] = 10
+        tm.assert_frame_equal(df, result)
+
+    def test_frame_setitem_multi_column(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=[["a", "a", "b", "b"], [0, 1, 0, 1]],
+        )
+
+        cp = df.copy()
+        cp["a"] = cp["b"]
+        tm.assert_frame_equal(cp["a"], cp["b"])
+
+        # set with ndarray
+        cp = df.copy()
+        cp["a"] = cp["b"].values
+        tm.assert_frame_equal(cp["a"], cp["b"])
+
+    def test_frame_setitem_multi_column2(self):
+        # ---------------------------------------
+        # GH#1803
+        columns = MultiIndex.from_tuples([("A", "1"), ("A", "2"), ("B", "1")])
+        df = DataFrame(index=[1, 3, 5], columns=columns)
+
+        # Works, but adds a column instead of updating the two existing ones
+        df["A"] = 0.0  # Doesn't work
+        assert (df["A"].values == 0).all()
+
+        # it broadcasts
+        df["B", "1"] = [1, 2, 3]
+        df["A"] = df["B", "1"]
+
+        sliced_a1 = df["A", "1"]
+        sliced_a2 = df["A", "2"]
+        sliced_b1 = df["B", "1"]
+        tm.assert_series_equal(sliced_a1, sliced_b1, check_names=False)
+        tm.assert_series_equal(sliced_a2, sliced_b1, check_names=False)
+        assert sliced_a1.name == ("A", "1")
+        assert sliced_a2.name == ("A", "2")
+        assert sliced_b1.name == ("B", "1")
+
+    def test_loc_getitem_tuple_plus_columns(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        # GH #1013
+        ymd = multiindex_year_month_day_dataframe_random_data
+        df = ymd[:5]
+
+        result = df.loc[(2000, 1, 6), ["A", "B", "C"]]
+        expected = df.loc[2000, 1, 6][["A", "B", "C"]]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+    def test_loc_getitem_setitem_slice_integers(self, frame_or_series):
+        index = MultiIndex(
+            levels=[[0, 1, 2], [0, 2]], codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]]
+        )
+
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 4)),
+            index=index,
+            columns=["a", "b", "c", "d"],
+        )
+        obj = tm.get_obj(obj, frame_or_series)
+
+        res = obj.loc[1:2]
+        exp = obj.reindex(obj.index[2:])
+        tm.assert_equal(res, exp)
+
+        obj.loc[1:2] = 7
+        assert (obj.loc[1:2] == 7).values.all()
+
+    def test_setitem_change_dtype(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        dft = frame.T
+        s = dft["foo", "two"]
+        dft["foo", "two"] = s > s.median()
+        tm.assert_series_equal(dft["foo", "two"], s > s.median())
+        # assert isinstance(dft._data.blocks[1].items, MultiIndex)
+
+        reindexed = dft.reindex(columns=[("foo", "two")])
+        tm.assert_series_equal(reindexed["foo", "two"], s > s.median())
+
+    def test_set_column_scalar_with_loc(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        subset = frame.index[[1, 4, 5]]
+
+        frame.loc[subset] = 99
+        assert (frame.loc[subset].values == 99).all()
+
+        frame_original = frame.copy()
+        col = frame["B"]
+        col[subset] = 97
+        # chained setitem doesn't work with CoW
+        tm.assert_frame_equal(frame, frame_original)
+
+    def test_nonunique_assignment_1750(self):
+        df = DataFrame(
+            [[1, 1, "x", "X"], [1, 1, "y", "Y"], [1, 2, "z", "Z"]], columns=list("ABCD")
+        )
+
+        df = df.set_index(["A", "B"])
+        mi = MultiIndex.from_tuples([(1, 1)])
+
+        df.loc[mi, "C"] = "_"
+
+        assert (df.xs((1, 1))["C"] == "_").all()
+
+    def test_astype_assignment_with_dups(self):
+        # GH 4686
+        # assignment with dups that has a dtype change
+        cols = MultiIndex.from_tuples([("A", "1"), ("B", "1"), ("A", "2")])
+        df = DataFrame(np.arange(3).reshape((1, 3)), columns=cols, dtype=object)
+        index = df.index.copy()
+
+        df["A"] = df["A"].astype(np.float64)
+        tm.assert_index_equal(df.index, index)
+
+    def test_setitem_nonmonotonic(self):
+        # https://github.com/pandas-dev/pandas/issues/31449
+        index = MultiIndex.from_tuples(
+            [("a", "c"), ("b", "x"), ("a", "d")], names=["l1", "l2"]
+        )
+        df = DataFrame(data=[0, 1, 2], index=index, columns=["e"])
+        df.loc["a", "e"] = np.arange(99, 101, dtype="int64")
+        expected = DataFrame({"e": [99, 1, 100]}, index=index)
+        tm.assert_frame_equal(df, expected)
+
+
+class TestSetitemWithExpansionMultiIndex:
+    def test_setitem_new_column_mixed_depth(self):
+        arrays = [
+            ["a", "top", "top", "routine1", "routine1", "routine2"],
+            ["", "OD", "OD", "result1", "result2", "result1"],
+            ["", "wx", "wy", "", "", ""],
+        ]
+
+        tuples = sorted(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples)
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 6)), columns=index)
+
+        result = df.copy()
+        expected = df.copy()
+        result["b"] = [1, 2, 3, 4]
+        expected["b", "", ""] = [1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_new_column_all_na(self):
+        # GH#1534
+        mix = MultiIndex.from_tuples([("1a", "2a"), ("1a", "2b"), ("1a", "2c")])
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], index=mix)
+        s = Series({(1, 1): 1, (1, 2): 2})
+        df["new"] = s
+        assert df["new"].isna().all()
+
+    def test_setitem_enlargement_keep_index_names(self):
+        # GH#53053
+        mi = MultiIndex.from_tuples([(1, 2, 3)], names=["i1", "i2", "i3"])
+        df = DataFrame(data=[[10, 20, 30]], index=mi, columns=["A", "B", "C"])
+        df.loc[(0, 0, 0)] = df.loc[(1, 2, 3)]
+        mi_expected = MultiIndex.from_tuples(
+            [(1, 2, 3), (0, 0, 0)], names=["i1", "i2", "i3"]
+        )
+        expected = DataFrame(
+            data=[[10, 20, 30], [10, 20, 30]],
+            index=mi_expected,
+            columns=["A", "B", "C"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+
+def test_frame_setitem_view_direct(multiindex_dataframe_random_data):
+    # this works because we are modifying the underlying array
+    # really a no-no
+    df = multiindex_dataframe_random_data.T
+    with pytest.raises(ValueError, match="read-only"):
+        df["foo"].values[:] = 0
+    assert (df["foo"].values != 0).all()
+
+
+def test_frame_setitem_copy_raises(multiindex_dataframe_random_data):
+    # will raise/warn as its chained assignment
+    df = multiindex_dataframe_random_data.T
+    with tm.raises_chained_assignment_error():
+        df["foo"]["one"] = 2
+
+
+def test_frame_setitem_copy_no_write(multiindex_dataframe_random_data):
+    frame = multiindex_dataframe_random_data.T
+    expected = frame
+    df = frame.copy()
+    with tm.raises_chained_assignment_error():
+        df["foo"]["one"] = 2
+
+    tm.assert_frame_equal(df, expected)
+
+
+def test_frame_setitem_partial_multiindex():
+    # GH 54875
+    df = DataFrame(
+        {
+            "a": [1, 2, 3],
+            "b": [3, 4, 5],
+            "c": 6,
+            "d": 7,
+        }
+    ).set_index(["a", "b", "c"])
+    ser = Series(8, index=df.index.droplevel("c"))
+    result = df.copy()
+    result["d"] = ser
+    expected = df.copy()
+    expected["d"] = 8
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_slice.py b/pandas/pandas/tests/indexing/multiindex/test_slice.py
new file mode 100644
index 0000000000000..7f298e9bdd375
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_slice.py
@@ -0,0 +1,798 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.errors import UnsortedIndexError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.tests.indexing.common import _mklbl
+
+
+class TestMultiIndexSlicers:
+    def test_per_axis_per_level_getitem(self):
+        # GH6134
+        # example test case
+        ix = MultiIndex.from_product(
+            [_mklbl("A", 5), _mklbl("B", 7), _mklbl("C", 4), _mklbl("D", 2)]
+        )
+        df = DataFrame(np.arange(len(ix.to_numpy())), index=ix)
+
+        result = df.loc[(slice("A1", "A3"), slice(None), ["C1", "C3"]), :]
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if a in ("A1", "A2", "A3") and c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if a in ("A1", "A2", "A3") and c in ("C1", "C2", "C3")
+            ]
+        ]
+        result = df.loc[(slice("A1", "A3"), slice(None), slice("C1", "C3")), :]
+        tm.assert_frame_equal(result, expected)
+
+        # test multi-index slicing with per axis and per index controls
+        index = MultiIndex.from_tuples(
+            [("A", 1), ("A", 2), ("A", 3), ("B", 1)], names=["one", "two"]
+        )
+        columns = MultiIndex.from_tuples(
+            [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")],
+            names=["lvl0", "lvl1"],
+        )
+
+        df = DataFrame(
+            np.arange(16, dtype="int64").reshape(4, 4), index=index, columns=columns
+        )
+        df = df.sort_index(axis=0).sort_index(axis=1)
+
+        # identity
+        result = df.loc[(slice(None), slice(None)), :]
+        tm.assert_frame_equal(result, df)
+        result = df.loc[(slice(None), slice(None)), (slice(None), slice(None))]
+        tm.assert_frame_equal(result, df)
+        result = df.loc[:, (slice(None), slice(None))]
+        tm.assert_frame_equal(result, df)
+
+        # index
+        result = df.loc[(slice(None), [1]), :]
+        expected = df.iloc[[0, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[(slice(None), 1), :]
+        expected = df.iloc[[0, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # columns
+        result = df.loc[:, (slice(None), ["foo"])]
+        expected = df.iloc[:, [1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # both
+        result = df.loc[(slice(None), 1), (slice(None), ["foo"])]
+        expected = df.iloc[[0, 3], [1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc["A", "a"]
+        expected = DataFrame(
+            {"bar": [1, 5, 9], "foo": [0, 4, 8]},
+            index=Index([1, 2, 3], name="two"),
+            columns=Index(["bar", "foo"], name="lvl1"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[(slice(None), [1, 2]), :]
+        expected = df.iloc[[0, 1, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        # multi-level series
+        s = Series(np.arange(len(ix.to_numpy())), index=ix)
+        result = s.loc["A1":"A3", :, ["C1", "C3"]]
+        expected = s.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in s.index.values
+                if a in ("A1", "A2", "A3") and c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_series_equal(result, expected)
+
+        # boolean indexers
+        result = df.loc[(slice(None), df.loc[:, ("a", "bar")] > 5), :]
+        expected = df.iloc[[2, 3]]
+        tm.assert_frame_equal(result, expected)
+
+        msg = (
+            "cannot index with a boolean indexer "
+            "that is not the same length as the index"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.loc[(slice(None), np.array([True, False])), :]
+
+        with pytest.raises(KeyError, match=r"\[1\] not in index"):
+            # slice(None) is on the index, [1] is on the columns, but 1 is
+            #  not in the columns, so we raise
+            #  This used to treat [1] as positional GH#16396
+            df.loc[slice(None), [1]]
+
+        # not lexsorted
+        assert df.index._lexsort_depth == 2
+        df = df.sort_index(level=1, axis=0)
+        assert df.index._lexsort_depth == 0
+
+        msg = (
+            "MultiIndex slicing requires the index to be "
+            r"lexsorted: slicing on levels \[1\], lexsort depth 0"
+        )
+        with pytest.raises(UnsortedIndexError, match=msg):
+            df.loc[(slice(None), slice("bar")), :]
+
+        # GH 16734: not sorted, but no real slicing
+        result = df.loc[(slice(None), df.loc[:, ("a", "bar")] > 5), :]
+        tm.assert_frame_equal(result, df.iloc[[1, 3], :])
+
+    def test_multiindex_slicers_non_unique(self):
+        # GH 7106
+        # non-unique mi index support
+        df = (
+            DataFrame(
+                {
+                    "A": ["foo", "foo", "foo", "foo"],
+                    "B": ["a", "a", "a", "a"],
+                    "C": [1, 2, 1, 3],
+                    "D": [1, 2, 3, 4],
+                }
+            )
+            .set_index(["A", "B", "C"])
+            .sort_index()
+        )
+        assert not df.index.is_unique
+        expected = (
+            DataFrame({"A": ["foo", "foo"], "B": ["a", "a"], "C": [1, 1], "D": [1, 3]})
+            .set_index(["A", "B", "C"])
+            .sort_index()
+        )
+        result = df.loc[(slice(None), slice(None), 1), :]
+        tm.assert_frame_equal(result, expected)
+
+        # this is equivalent of an xs expression
+        result = df.xs(1, level=2, drop_level=False)
+        tm.assert_frame_equal(result, expected)
+
+        df = (
+            DataFrame(
+                {
+                    "A": ["foo", "foo", "foo", "foo"],
+                    "B": ["a", "a", "a", "a"],
+                    "C": [1, 2, 1, 2],
+                    "D": [1, 2, 3, 4],
+                }
+            )
+            .set_index(["A", "B", "C"])
+            .sort_index()
+        )
+        assert not df.index.is_unique
+        expected = (
+            DataFrame({"A": ["foo", "foo"], "B": ["a", "a"], "C": [1, 1], "D": [1, 3]})
+            .set_index(["A", "B", "C"])
+            .sort_index()
+        )
+        result = df.loc[(slice(None), slice(None), 1), :]
+        assert not result.index.is_unique
+        tm.assert_frame_equal(result, expected)
+
+        # GH12896
+        # numpy-implementation dependent bug
+        ints = [
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            12,
+            13,
+            14,
+            14,
+            16,
+            17,
+            18,
+            19,
+            200000,
+            200000,
+        ]
+        n = len(ints)
+        idx = MultiIndex.from_arrays([["a"] * n, ints])
+        result = Series([1] * n, index=idx)
+        result = result.sort_index()
+        result = result.loc[(slice(None), slice(100000))]
+        expected = Series([1] * (n - 2), index=idx[:-2]).sort_index()
+        tm.assert_series_equal(result, expected)
+
+    def test_multiindex_slicers_datetimelike(self):
+        # GH 7429
+        # buggy/inconsistent behavior when slicing with datetime-like
+        dates = [datetime(2012, 1, 1, 12, 12, 12) + timedelta(days=i) for i in range(6)]
+        freq = [1, 2]
+        index = MultiIndex.from_product([dates, freq], names=["date", "frequency"])
+
+        df = DataFrame(
+            np.arange(6 * 2 * 4, dtype="int64").reshape(-1, 4),
+            index=index,
+            columns=list("ABCD"),
+        )
+
+        # multi-axis slicing
+        idx = pd.IndexSlice
+        expected = df.iloc[[0, 2, 4], [0, 1]]
+        result = df.loc[
+            (
+                slice(
+                    Timestamp("2012-01-01 12:12:12"), Timestamp("2012-01-03 12:12:12")
+                ),
+                slice(1, 1),
+            ),
+            slice("A", "B"),
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[
+            (
+                idx[
+                    Timestamp("2012-01-01 12:12:12") : Timestamp("2012-01-03 12:12:12")
+                ],
+                idx[1:1],
+            ),
+            slice("A", "B"),
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[
+            (
+                slice(
+                    Timestamp("2012-01-01 12:12:12"), Timestamp("2012-01-03 12:12:12")
+                ),
+                1,
+            ),
+            slice("A", "B"),
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        # with strings
+        result = df.loc[
+            (slice("2012-01-01 12:12:12", "2012-01-03 12:12:12"), slice(1, 1)),
+            slice("A", "B"),
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[
+            (idx["2012-01-01 12:12:12":"2012-01-03 12:12:12"], 1), idx["A", "B"]
+        ]
+        tm.assert_frame_equal(result, expected)
+
+    def test_multiindex_slicers_edges(self):
+        # GH 8132
+        # various edge cases
+        df = DataFrame(
+            {
+                "A": ["A0"] * 5 + ["A1"] * 5 + ["A2"] * 5,
+                "B": ["B0", "B0", "B1", "B1", "B2"] * 3,
+                "DATE": [
+                    "2013-06-11",
+                    "2013-07-02",
+                    "2013-07-09",
+                    "2013-07-30",
+                    "2013-08-06",
+                    "2013-06-11",
+                    "2013-07-02",
+                    "2013-07-09",
+                    "2013-07-30",
+                    "2013-08-06",
+                    "2013-09-03",
+                    "2013-10-01",
+                    "2013-07-09",
+                    "2013-08-06",
+                    "2013-09-03",
+                ],
+                "VALUES": [22, 35, 14, 9, 4, 40, 18, 4, 2, 5, 1, 2, 3, 4, 2],
+            }
+        )
+
+        df["DATE"] = pd.to_datetime(df["DATE"])
+        df1 = df.set_index(["A", "B", "DATE"])
+        df1 = df1.sort_index()
+
+        # A1 - Get all values under "A0" and "A1"
+        result = df1.loc[(slice("A1")), :]
+        expected = df1.iloc[0:10]
+        tm.assert_frame_equal(result, expected)
+
+        # A2 - Get all values from the start to "A2"
+        result = df1.loc[(slice("A2")), :]
+        expected = df1
+        tm.assert_frame_equal(result, expected)
+
+        # A3 - Get all values under "B1" or "B2"
+        result = df1.loc[(slice(None), slice("B1", "B2")), :]
+        expected = df1.iloc[[2, 3, 4, 7, 8, 9, 12, 13, 14]]
+        tm.assert_frame_equal(result, expected)
+
+        # A4 - Get all values between 2013-07-02 and 2013-07-09
+        result = df1.loc[(slice(None), slice(None), slice("20130702", "20130709")), :]
+        expected = df1.iloc[[1, 2, 6, 7, 12]]
+        tm.assert_frame_equal(result, expected)
+
+        # B1 - Get all values in B0 that are also under A0, A1 and A2
+        result = df1.loc[(slice("A2"), slice("B0")), :]
+        expected = df1.iloc[[0, 1, 5, 6, 10, 11]]
+        tm.assert_frame_equal(result, expected)
+
+        # B2 - Get all values in B0, B1 and B2 (similar to what #2 is doing for
+        # the As)
+        result = df1.loc[(slice(None), slice("B2")), :]
+        expected = df1
+        tm.assert_frame_equal(result, expected)
+
+        # B3 - Get all values from B1 to B2 and up to 2013-08-06
+        result = df1.loc[(slice(None), slice("B1", "B2"), slice("2013-08-06")), :]
+        expected = df1.iloc[[2, 3, 4, 7, 8, 9, 12, 13]]
+        tm.assert_frame_equal(result, expected)
+
+        # B4 - Same as A4 but the start of the date slice is not a key.
+        #      shows indexing on a partial selection slice
+        result = df1.loc[(slice(None), slice(None), slice("20130701", "20130709")), :]
+        expected = df1.iloc[[1, 2, 6, 7, 12]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_per_axis_per_level_doc_examples(self):
+        # test index maker
+        idx = pd.IndexSlice
+
+        # from indexing.rst / advanced
+        index = MultiIndex.from_product(
+            [_mklbl("A", 4), _mklbl("B", 2), _mklbl("C", 4), _mklbl("D", 2)]
+        )
+        columns = MultiIndex.from_tuples(
+            [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")],
+            names=["lvl0", "lvl1"],
+        )
+        df = DataFrame(
+            np.arange(len(index) * len(columns), dtype="int64").reshape(
+                (len(index), len(columns))
+            ),
+            index=index,
+            columns=columns,
+        )
+        result = df.loc[(slice("A1", "A3"), slice(None), ["C1", "C3"]), :]
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if a in ("A1", "A2", "A3") and c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_frame_equal(result, expected)
+        result = df.loc[idx["A1":"A3", :, ["C1", "C3"]], :]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[(slice(None), slice(None), ["C1", "C3"]), :]
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_frame_equal(result, expected)
+        result = df.loc[idx[:, :, ["C1", "C3"]], :]
+        tm.assert_frame_equal(result, expected)
+
+        # not sorted
+        msg = (
+            "MultiIndex slicing requires the index to be lexsorted: "
+            r"slicing on levels \[1\], lexsort depth 1"
+        )
+        with pytest.raises(UnsortedIndexError, match=msg):
+            df.loc["A1", ("a", slice("foo"))]
+
+        # GH 16734: not sorted, but no real slicing
+        tm.assert_frame_equal(
+            df.loc["A1", (slice(None), "foo")], df.loc["A1"].iloc[:, [0, 2]]
+        )
+
+        df = df.sort_index(axis=1)
+
+        # slicing
+        df.loc["A1", (slice(None), "foo")]
+        df.loc[(slice(None), slice(None), ["C1", "C3"]), (slice(None), "foo")]
+
+        # setitem
+        df.loc(axis=0)[:, :, ["C1", "C3"]] = -10
+
+    def test_loc_axis_arguments(self):
+        index = MultiIndex.from_product(
+            [_mklbl("A", 4), _mklbl("B", 2), _mklbl("C", 4), _mklbl("D", 2)]
+        )
+        columns = MultiIndex.from_tuples(
+            [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")],
+            names=["lvl0", "lvl1"],
+        )
+        df = (
+            DataFrame(
+                np.arange(len(index) * len(columns), dtype="int64").reshape(
+                    (len(index), len(columns))
+                ),
+                index=index,
+                columns=columns,
+            )
+            .sort_index()
+            .sort_index(axis=1)
+        )
+
+        # axis 0
+        result = df.loc(axis=0)["A1":"A3", :, ["C1", "C3"]]
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if a in ("A1", "A2", "A3") and c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc(axis="index")[:, :, ["C1", "C3"]]
+        expected = df.loc[
+            [
+                (
+                    a,
+                    b,
+                    c,
+                    d,
+                )
+                for a, b, c, d in df.index.values
+                if c in ("C1", "C3")
+            ]
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        # axis 1
+        result = df.loc(axis=1)[:, "foo"]
+        expected = df.loc[:, (slice(None), "foo")]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc(axis="columns")[:, "foo"]
+        expected = df.loc[:, (slice(None), "foo")]
+        tm.assert_frame_equal(result, expected)
+
+        # invalid axis
+        for i in [-1, 2, "foo"]:
+            msg = f"No axis named {i} for object type DataFrame"
+            with pytest.raises(ValueError, match=msg):
+                df.loc(axis=i)[:, :, ["C1", "C3"]]
+
+    def test_loc_axis_single_level_multi_col_indexing_multiindex_col_df(self):
+        # GH29519
+        df = DataFrame(
+            np.arange(27).reshape(3, 9),
+            columns=MultiIndex.from_product([["a1", "a2", "a3"], ["b1", "b2", "b3"]]),
+        )
+        result = df.loc(axis=1)["a1":"a2"]
+        expected = df.iloc[:, :-3]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_axis_single_level_single_col_indexing_multiindex_col_df(self):
+        # GH29519
+        df = DataFrame(
+            np.arange(27).reshape(3, 9),
+            columns=MultiIndex.from_product([["a1", "a2", "a3"], ["b1", "b2", "b3"]]),
+        )
+        result = df.loc(axis=1)["a1"]
+        expected = df.iloc[:, :3]
+        expected.columns = ["b1", "b2", "b3"]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_ax_single_level_indexer_simple_df(self):
+        # GH29519
+        # test single level indexing on single index column data frame
+        df = DataFrame(np.arange(9).reshape(3, 3), columns=["a", "b", "c"])
+        result = df.loc(axis=1)["a"]
+        expected = Series(np.array([0, 3, 6]), name="a")
+        tm.assert_series_equal(result, expected)
+
+    def test_per_axis_per_level_setitem(self):
+        # test index maker
+        idx = pd.IndexSlice
+
+        # test multi-index slicing with per axis and per index controls
+        index = MultiIndex.from_tuples(
+            [("A", 1), ("A", 2), ("A", 3), ("B", 1)], names=["one", "two"]
+        )
+        columns = MultiIndex.from_tuples(
+            [("a", "foo"), ("a", "bar"), ("b", "foo"), ("b", "bah")],
+            names=["lvl0", "lvl1"],
+        )
+
+        df_orig = DataFrame(
+            np.arange(16, dtype="int64").reshape(4, 4), index=index, columns=columns
+        )
+        df_orig = df_orig.sort_index(axis=0).sort_index(axis=1)
+
+        # identity
+        df = df_orig.copy()
+        df.loc[(slice(None), slice(None)), :] = 100
+        expected = df_orig.copy()
+        expected.iloc[:, :] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc(axis=0)[:, :] = 100
+        expected = df_orig.copy()
+        expected.iloc[:, :] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[(slice(None), slice(None)), (slice(None), slice(None))] = 100
+        expected = df_orig.copy()
+        expected.iloc[:, :] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[:, (slice(None), slice(None))] = 100
+        expected = df_orig.copy()
+        expected.iloc[:, :] = 100
+        tm.assert_frame_equal(df, expected)
+
+        # index
+        df = df_orig.copy()
+        df.loc[(slice(None), [1]), :] = 100
+        expected = df_orig.copy()
+        expected.iloc[[0, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), :] = 100
+        expected = df_orig.copy()
+        expected.iloc[[0, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc(axis=0)[:, 1] = 100
+        expected = df_orig.copy()
+        expected.iloc[[0, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        # columns
+        df = df_orig.copy()
+        df.loc[:, (slice(None), ["foo"])] = 100
+        expected = df_orig.copy()
+        expected.iloc[:, [1, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        # both
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), (slice(None), ["foo"])] = 100
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[idx[:, 1], idx[:, ["foo"]]] = 100
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc["A", "a"] = 100
+        expected = df_orig.copy()
+        expected.iloc[0:3, 0:2] = 100
+        tm.assert_frame_equal(df, expected)
+
+        # setting with a list-like
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), (slice(None), ["foo"])] = np.array(
+            [[100, 100], [100, 100]], dtype="int64"
+        )
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] = 100
+        tm.assert_frame_equal(df, expected)
+
+        # not enough values
+        df = df_orig.copy()
+
+        msg = "setting an array element with a sequence."
+        with pytest.raises(ValueError, match=msg):
+            df.loc[(slice(None), 1), (slice(None), ["foo"])] = np.array(
+                [[100], [100, 100]], dtype="int64"
+            )
+
+        msg = "Must have equal len keys and value when setting with an iterable"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[(slice(None), 1), (slice(None), ["foo"])] = np.array(
+                [100, 100, 100, 100], dtype="int64"
+            )
+
+        # with an alignable rhs
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), (slice(None), ["foo"])] = (
+            df.loc[(slice(None), 1), (slice(None), ["foo"])] * 5
+        )
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] = expected.iloc[[0, 3], [1, 3]] * 5
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), (slice(None), ["foo"])] *= df.loc[
+            (slice(None), 1), (slice(None), ["foo"])
+        ]
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] *= expected.iloc[[0, 3], [1, 3]]
+        tm.assert_frame_equal(df, expected)
+
+        rhs = df_orig.loc[(slice(None), 1), (slice(None), ["foo"])].copy()
+        rhs.loc[:, ("c", "bah")] = 10
+        df = df_orig.copy()
+        df.loc[(slice(None), 1), (slice(None), ["foo"])] *= rhs
+        expected = df_orig.copy()
+        expected.iloc[[0, 3], [1, 3]] *= expected.iloc[[0, 3], [1, 3]]
+        tm.assert_frame_equal(df, expected)
+
+    def test_multiindex_label_slicing_with_negative_step(self):
+        ser = Series(
+            np.arange(20), MultiIndex.from_product([list("abcde"), np.arange(4)])
+        )
+        SLC = pd.IndexSlice
+
+        tm.assert_indexing_slices_equivalent(ser, SLC[::-1], SLC[::-1])
+
+        tm.assert_indexing_slices_equivalent(ser, SLC["d"::-1], SLC[15::-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[("d",) :: -1], SLC[15::-1])
+
+        tm.assert_indexing_slices_equivalent(ser, SLC[:"d":-1], SLC[:11:-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[: ("d",) : -1], SLC[:11:-1])
+
+        tm.assert_indexing_slices_equivalent(ser, SLC["d":"b":-1], SLC[15:3:-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[("d",) : "b" : -1], SLC[15:3:-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC["d" : ("b",) : -1], SLC[15:3:-1])
+        tm.assert_indexing_slices_equivalent(
+            ser, SLC[("d",) : ("b",) : -1], SLC[15:3:-1]
+        )
+        tm.assert_indexing_slices_equivalent(ser, SLC["b":"d":-1], SLC[:0])
+
+        tm.assert_indexing_slices_equivalent(ser, SLC[("c", 2) :: -1], SLC[10::-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[: ("c", 2) : -1], SLC[:9:-1])
+        tm.assert_indexing_slices_equivalent(
+            ser, SLC[("e", 0) : ("c", 2) : -1], SLC[16:9:-1]
+        )
+
+    def test_multiindex_slice_first_level(self):
+        # GH 12697
+        freq = ["a", "b", "c", "d"]
+        idx = MultiIndex.from_product([freq, range(500)])
+        df = DataFrame(list(range(2000)), index=idx, columns=["Test"])
+        df_slice = df.loc[pd.IndexSlice[:, 30:70], :]
+        result = df_slice.loc["a"]
+        expected = DataFrame(list(range(30, 71)), columns=["Test"], index=range(30, 71))
+        tm.assert_frame_equal(result, expected)
+        result = df_slice.loc["d"]
+        expected = DataFrame(
+            list(range(1530, 1571)), columns=["Test"], index=range(30, 71)
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_int_series_slicing(self, multiindex_year_month_day_dataframe_random_data):
+        ymd = multiindex_year_month_day_dataframe_random_data
+        s = ymd["A"]
+        result = s[5:]
+        expected = s.reindex(s.index[5:])
+        tm.assert_series_equal(result, expected)
+
+        s = ymd["A"].copy()
+        exp = ymd["A"].copy()
+        s[5:] = 0
+        exp.iloc[5:] = 0
+        tm.assert_numpy_array_equal(s.values, exp.values)
+
+        result = ymd[5:]
+        expected = ymd.reindex(s.index[5:])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, loc, iloc",
+        [
+            # dtype = int, step = -1
+            ("int", slice(None, None, -1), slice(None, None, -1)),
+            ("int", slice(3, None, -1), slice(3, None, -1)),
+            ("int", slice(None, 1, -1), slice(None, 0, -1)),
+            ("int", slice(3, 1, -1), slice(3, 0, -1)),
+            # dtype = int, step = -2
+            ("int", slice(None, None, -2), slice(None, None, -2)),
+            ("int", slice(3, None, -2), slice(3, None, -2)),
+            ("int", slice(None, 1, -2), slice(None, 0, -2)),
+            ("int", slice(3, 1, -2), slice(3, 0, -2)),
+            # dtype = str, step = -1
+            ("str", slice(None, None, -1), slice(None, None, -1)),
+            ("str", slice("d", None, -1), slice(3, None, -1)),
+            ("str", slice(None, "b", -1), slice(None, 0, -1)),
+            ("str", slice("d", "b", -1), slice(3, 0, -1)),
+            # dtype = str, step = -2
+            ("str", slice(None, None, -2), slice(None, None, -2)),
+            ("str", slice("d", None, -2), slice(3, None, -2)),
+            ("str", slice(None, "b", -2), slice(None, 0, -2)),
+            ("str", slice("d", "b", -2), slice(3, 0, -2)),
+        ],
+    )
+    def test_loc_slice_negative_stepsize(self, dtype, loc, iloc):
+        # GH#38071
+        labels = {
+            "str": list("abcde"),
+            "int": range(5),
+        }[dtype]
+
+        mi = MultiIndex.from_arrays([labels] * 2)
+        df = DataFrame(1.0, index=mi, columns=["A"])
+
+        SLC = pd.IndexSlice
+
+        expected = df.iloc[iloc, :]
+        result_get_loc = df.loc[SLC[loc], :]
+        result_get_locs_level_0 = df.loc[SLC[loc, :], :]
+        result_get_locs_level_1 = df.loc[SLC[:, loc], :]
+
+        tm.assert_frame_equal(result_get_loc, expected)
+        tm.assert_frame_equal(result_get_locs_level_0, expected)
+        tm.assert_frame_equal(result_get_locs_level_1, expected)
diff --git a/pandas/pandas/tests/indexing/multiindex/test_sorted.py b/pandas/pandas/tests/indexing/multiindex/test_sorted.py
new file mode 100644
index 0000000000000..cf3fa5296c97c
--- /dev/null
+++ b/pandas/pandas/tests/indexing/multiindex/test_sorted.py
@@ -0,0 +1,153 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    MultiIndex,
+    Series,
+    array,
+)
+import pandas._testing as tm
+
+
+class TestMultiIndexSorted:
+    def test_getitem_multilevel_index_tuple_not_sorted(self):
+        index_columns = list("abc")
+        df = DataFrame(
+            [[0, 1, 0, "x"], [0, 0, 1, "y"]], columns=index_columns + ["data"]
+        )
+        df = df.set_index(index_columns)
+        query_index = df.index[:1]
+        rs = df.loc[query_index, "data"]
+
+        xp_idx = MultiIndex.from_tuples([(0, 1, 0)], names=["a", "b", "c"])
+        xp = Series(["x"], index=xp_idx, name="data")
+        tm.assert_series_equal(rs, xp)
+
+    def test_getitem_slice_not_sorted(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        df = frame.sort_index(level=1).T
+
+        # buglet with int typechecking
+        result = df.iloc[:, : np.int32(3)]
+        expected = df.reindex(columns=df.columns[:3])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("key", [None, lambda x: x])
+    def test_frame_getitem_not_sorted2(self, key):
+        # 13431
+        df = DataFrame(
+            {
+                "col1": ["b", "d", "b", "a"],
+                "col2": [3, 1, 1, 2],
+                "data": ["one", "two", "three", "four"],
+            }
+        )
+
+        df2 = df.set_index(["col1", "col2"])
+        df2_original = df2.copy()
+
+        df2.index = df2.index.set_levels(["b", "d", "a"], level="col1")
+        df2.index = df2.index.set_codes([0, 1, 0, 2], level="col1")
+        assert not df2.index.is_monotonic_increasing
+
+        assert df2_original.index.equals(df2.index)
+        expected = df2.sort_index(key=key)
+        assert expected.index.is_monotonic_increasing
+
+        result = df2.sort_index(level=0, key=key)
+        assert result.index.is_monotonic_increasing
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_values_key(self):
+        arrays = [
+            ["bar", "bar", "baz", "baz", "qux", "qux", "foo", "foo"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = zip(*arrays)
+        index = MultiIndex.from_tuples(tuples)
+        index = index.sort_values(  # sort by third letter
+            key=lambda x: x.map(lambda entry: entry[2])
+        )
+        result = DataFrame(range(8), index=index)
+
+        arrays = [
+            ["foo", "foo", "bar", "bar", "qux", "qux", "baz", "baz"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = zip(*arrays)
+        index = MultiIndex.from_tuples(tuples)
+        expected = DataFrame(range(8), index=index)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_argsort_with_na(self):
+        # GH48495
+        arrays = [
+            array([2, NA, 1], dtype="Int64"),
+            array([1, 2, 3], dtype="Int64"),
+        ]
+        index = MultiIndex.from_arrays(arrays)
+        result = index.argsort()
+        expected = np.array([2, 0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_sort_values_with_na(self):
+        # GH48495
+        arrays = [
+            array([2, NA, 1], dtype="Int64"),
+            array([1, 2, 3], dtype="Int64"),
+        ]
+        index = MultiIndex.from_arrays(arrays)
+        index = index.sort_values()
+        result = DataFrame(range(3), index=index)
+
+        arrays = [
+            array([1, 2, NA], dtype="Int64"),
+            array([3, 1, 2], dtype="Int64"),
+        ]
+        index = MultiIndex.from_arrays(arrays)
+        expected = DataFrame(range(3), index=index)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_getitem_not_sorted(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        df = frame.T
+        df["foo", "four"] = "foo"
+
+        arrays = [np.array(x) for x in zip(*df.columns.values)]
+
+        result = df["foo"]
+        result2 = df.loc[:, "foo"]
+        expected = df.reindex(columns=df.columns[arrays[0] == "foo"])
+        expected.columns = expected.columns.droplevel(0)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        df = df.T
+        result = df.xs("foo")
+        result2 = df.loc["foo"]
+        expected = df.reindex(df.index[arrays[0] == "foo"])
+        expected.index = expected.index.droplevel(0)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_series_getitem_not_sorted(self):
+        arrays = [
+            ["bar", "bar", "baz", "baz", "qux", "qux", "foo", "foo"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = zip(*arrays)
+        index = MultiIndex.from_tuples(tuples)
+        s = Series(np.random.default_rng(2).standard_normal(8), index=index)
+
+        arrays = [np.array(x) for x in zip(*index.values)]
+
+        result = s["qux"]
+        result2 = s.loc["qux"]
+        expected = s[arrays[0] == "qux"]
+        expected.index = expected.index.droplevel(0)
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
diff --git a/pandas/pandas/tests/indexing/test_at.py b/pandas/pandas/tests/indexing/test_at.py
new file mode 100644
index 0000000000000..e80acc230a320
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_at.py
@@ -0,0 +1,252 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+from pandas import (
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    MultiIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+def test_at_timezone():
+    # https://github.com/pandas-dev/pandas/issues/33544
+    result = DataFrame({"foo": [datetime(2000, 1, 1)]})
+    with pytest.raises(TypeError, match="Invalid value"):
+        result.at[0, "foo"] = datetime(2000, 1, 2, tzinfo=timezone.utc)
+
+
+def test_selection_methods_of_assigned_col():
+    # GH 29282
+    df = DataFrame(data={"a": [1, 2, 3], "b": [4, 5, 6]})
+    df2 = DataFrame(data={"c": [7, 8, 9]}, index=[2, 1, 0])
+    df["c"] = df2["c"]
+    df.at[1, "c"] = 11
+    result = df
+    expected = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [9, 11, 7]})
+    tm.assert_frame_equal(result, expected)
+    result = df.at[1, "c"]
+    assert result == 11
+
+    result = df["c"]
+    expected = Series([9, 11, 7], name="c")
+    tm.assert_series_equal(result, expected)
+
+    result = df[["c"]]
+    expected = DataFrame({"c": [9, 11, 7]})
+    tm.assert_frame_equal(result, expected)
+
+
+class TestAtSetItem:
+    def test_at_setitem_item_cache_cleared(self):
+        # GH#22372 Note the multi-step construction is necessary to trigger
+        #  the original bug. pandas/issues/22372#issuecomment-413345309
+        df = DataFrame(index=[0])
+        df["x"] = 1
+        df["cost"] = 2
+
+        # accessing df["cost"] adds "cost" to the _item_cache
+        df["cost"]
+
+        # This loc[[0]] lookup used to call _consolidate_inplace at the
+        #  BlockManager level, which failed to clear the _item_cache
+        df.loc[[0]]
+
+        df.at[0, "x"] = 4
+        df.at[0, "cost"] = 789
+
+        expected = DataFrame({"x": [4], "cost": 789}, index=[0])
+        tm.assert_frame_equal(df, expected)
+
+        # And in particular, check that the _item_cache has updated correctly.
+        tm.assert_series_equal(df["cost"], expected["cost"])
+
+    def test_at_setitem_mixed_index_assignment(self):
+        # GH#19860
+        ser = Series([1, 2, 3, 4, 5], index=["a", "b", "c", 1, 2])
+        ser.at["a"] = 11
+        assert ser.iat[0] == 11
+        ser.at[1] = 22
+        assert ser.iat[3] == 22
+
+    def test_at_setitem_categorical_missing(self):
+        df = DataFrame(
+            index=range(3), columns=range(3), dtype=CategoricalDtype(["foo", "bar"])
+        )
+        df.at[1, 1] = "foo"
+
+        expected = DataFrame(
+            [
+                [np.nan, np.nan, np.nan],
+                [np.nan, "foo", np.nan],
+                [np.nan, np.nan, np.nan],
+            ],
+            dtype=CategoricalDtype(["foo", "bar"]),
+        )
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_at_setitem_multiindex(self):
+        df = DataFrame(
+            np.zeros((3, 2), dtype="int64"),
+            columns=MultiIndex.from_tuples([("a", 0), ("a", 1)]),
+        )
+        df.at[0, "a"] = 10
+        expected = DataFrame(
+            [[10, 10], [0, 0], [0, 0]],
+            columns=MultiIndex.from_tuples([("a", 0), ("a", 1)]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("row", (Timestamp("2019-01-01"), "2019-01-01"))
+    def test_at_datetime_index(self, row):
+        # Set float64 dtype to avoid upcast when setting .5
+        df = DataFrame(
+            data=[[1] * 2], index=DatetimeIndex(data=["2019-01-01", "2019-01-02"])
+        ).astype({0: "float64"})
+        expected = DataFrame(
+            data=[[0.5, 1], [1.0, 1]],
+            index=DatetimeIndex(data=["2019-01-01", "2019-01-02"]),
+        )
+
+        df.at[row, 0] = 0.5
+        tm.assert_frame_equal(df, expected)
+
+
+class TestAtSetItemWithExpansion:
+    def test_at_setitem_expansion_series_dt64tz_value(self, tz_naive_fixture):
+        # GH#25506
+        ts = (
+            Timestamp("2017-08-05 00:00:00+0100", tz=tz_naive_fixture)
+            if tz_naive_fixture is not None
+            else Timestamp("2017-08-05 00:00:00+0100")
+        )
+        result = Series(ts)
+        result.at[1] = ts
+        expected = Series([ts, ts])
+        tm.assert_series_equal(result, expected)
+
+
+class TestAtWithDuplicates:
+    def test_at_with_duplicate_axes_requires_scalar_lookup(self):
+        # GH#33041 check that falling back to loc doesn't allow non-scalar
+        #  args to slip in
+
+        arr = np.random.default_rng(2).standard_normal(6).reshape(3, 2)
+        df = DataFrame(arr, columns=["A", "A"])
+
+        msg = "Invalid call for scalar access"
+        with pytest.raises(ValueError, match=msg):
+            df.at[[1, 2]]
+        with pytest.raises(ValueError, match=msg):
+            df.at[1, ["A"]]
+        with pytest.raises(ValueError, match=msg):
+            df.at[:, "A"]
+
+        with pytest.raises(ValueError, match=msg):
+            df.at[[1, 2]] = 1
+        with pytest.raises(ValueError, match=msg):
+            df.at[1, ["A"]] = 1
+        with pytest.raises(ValueError, match=msg):
+            df.at[:, "A"] = 1
+
+
+class TestAtErrors:
+    # TODO: De-duplicate/parametrize
+    #  test_at_series_raises_key_error2, test_at_frame_raises_key_error2
+
+    def test_at_series_raises_key_error(self, indexer_al):
+        # GH#31724 .at should match .loc
+
+        ser = Series([1, 2, 3], index=[3, 2, 1])
+        result = indexer_al(ser)[1]
+        assert result == 3
+
+        with pytest.raises(KeyError, match="a"):
+            indexer_al(ser)["a"]
+
+    def test_at_frame_raises_key_error(self, indexer_al):
+        # GH#31724 .at should match .loc
+
+        df = DataFrame({0: [1, 2, 3]}, index=[3, 2, 1])
+
+        result = indexer_al(df)[1, 0]
+        assert result == 3
+
+        with pytest.raises(KeyError, match="a"):
+            indexer_al(df)["a", 0]
+
+        with pytest.raises(KeyError, match="a"):
+            indexer_al(df)[1, "a"]
+
+    def test_at_series_raises_key_error2(self, indexer_al):
+        # at should not fallback
+        # GH#7814
+        # GH#31724 .at should match .loc
+        ser = Series([1, 2, 3], index=list("abc"))
+        result = indexer_al(ser)["a"]
+        assert result == 1
+
+        with pytest.raises(KeyError, match="^0$"):
+            indexer_al(ser)[0]
+
+    def test_at_frame_raises_key_error2(self, indexer_al):
+        # GH#31724 .at should match .loc
+        df = DataFrame({"A": [1, 2, 3]}, index=list("abc"))
+        result = indexer_al(df)["a", "A"]
+        assert result == 1
+
+        with pytest.raises(KeyError, match="^0$"):
+            indexer_al(df)["a", 0]
+
+    def test_at_frame_multiple_columns(self):
+        # GH#48296 - at shouldn't modify multiple columns
+        df = DataFrame({"a": [1, 2], "b": [3, 4]})
+        new_row = [6, 7]
+        with pytest.raises(
+            InvalidIndexError,
+            match=f"You can only assign a scalar value not a \\{type(new_row)}",
+        ):
+            df.at[5] = new_row
+
+    def test_at_getitem_mixed_index_no_fallback(self):
+        # GH#19860
+        ser = Series([1, 2, 3, 4, 5], index=["a", "b", "c", 1, 2])
+        with pytest.raises(KeyError, match="^0$"):
+            ser.at[0]
+        with pytest.raises(KeyError, match="^4$"):
+            ser.at[4]
+
+    def test_at_categorical_integers(self):
+        # CategoricalIndex with integer categories that don't happen to match
+        #  the Categorical's codes
+        ci = CategoricalIndex([3, 4])
+
+        arr = np.arange(4).reshape(2, 2)
+        frame = DataFrame(arr, index=ci)
+
+        for df in [frame, frame.T]:
+            for key in [0, 1]:
+                with pytest.raises(KeyError, match=str(key)):
+                    df.at[key, key]
+
+    def test_at_applied_for_rows(self):
+        # GH#48729 .at should raise InvalidIndexError when assigning rows
+        df = DataFrame(index=["a"], columns=["col1", "col2"])
+        new_row = [123, 15]
+        with pytest.raises(
+            InvalidIndexError,
+            match=f"You can only assign a scalar value not a \\{type(new_row)}",
+        ):
+            df.at["a"] = new_row
diff --git a/pandas/pandas/tests/indexing/test_categorical.py b/pandas/pandas/tests/indexing/test_categorical.py
new file mode 100644
index 0000000000000..c9f29b2cb55fe
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_categorical.py
@@ -0,0 +1,573 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    Interval,
+    Series,
+    Timedelta,
+    Timestamp,
+    option_context,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        {
+            "A": np.arange(6, dtype="int64"),
+        },
+        index=CategoricalIndex(
+            list("aabbca"), dtype=CategoricalDtype(list("cab")), name="B"
+        ),
+    )
+
+
+@pytest.fixture
+def df2():
+    return DataFrame(
+        {
+            "A": np.arange(6, dtype="int64"),
+        },
+        index=CategoricalIndex(
+            list("aabbca"), dtype=CategoricalDtype(list("cabe")), name="B"
+        ),
+    )
+
+
+class TestCategoricalIndex:
+    def test_loc_scalar(self, df):
+        dtype = CategoricalDtype(list("cab"))
+        result = df.loc["a"]
+        bidx = Series(list("aaa"), name="B").astype(dtype)
+        assert bidx.dtype == dtype
+
+        expected = DataFrame({"A": [0, 1, 5]}, index=Index(bidx))
+        tm.assert_frame_equal(result, expected)
+
+        df = df.copy()
+        df.loc["a"] = 20
+        bidx2 = Series(list("aabbca"), name="B").astype(dtype)
+        assert bidx2.dtype == dtype
+        expected = DataFrame(
+            {
+                "A": [20, 20, 2, 3, 4, 20],
+            },
+            index=Index(bidx2),
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # value not in the categories
+        with pytest.raises(KeyError, match=r"^'d'$"):
+            df.loc["d"]
+
+        df2 = df.copy()
+        expected = df2.copy()
+        expected.index = expected.index.astype(object)
+        expected.loc["d"] = 10
+        df2.loc["d"] = 10
+        tm.assert_frame_equal(df2, expected)
+
+    def test_loc_setitem_with_expansion_non_category(self, df):
+        # Setting-with-expansion with a new key "d" that is not among caegories
+        df.loc["a"] = 20
+
+        # Setting a new row on an existing column
+        df3 = df.copy()
+        df3.loc["d", "A"] = 10
+        bidx3 = Index(list("aabbcad"), name="B")
+        expected3 = DataFrame(
+            {
+                "A": [20, 20, 2, 3, 4, 20, 10.0],
+            },
+            index=Index(bidx3),
+        )
+        tm.assert_frame_equal(df3, expected3)
+
+        # Setting a new row _and_ new column
+        df4 = df.copy()
+        df4.loc["d", "C"] = 10
+        expected3 = DataFrame(
+            {
+                "A": [20, 20, 2, 3, 4, 20, np.nan],
+                "C": [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, 10],
+            },
+            index=Index(bidx3),
+        )
+        tm.assert_frame_equal(df4, expected3)
+
+    def test_loc_getitem_scalar_non_category(self, df):
+        with pytest.raises(KeyError, match="^1$"):
+            df.loc[1]
+
+    def test_slicing(self):
+        cat = Series(Categorical([1, 2, 3, 4]))
+        reverse = cat[::-1]
+        exp = np.array([4, 3, 2, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(reverse.__array__(), exp)
+
+        df = DataFrame({"value": (np.arange(100) + 1).astype("int64")})
+        df["D"] = pd.cut(df.value, bins=[0, 25, 50, 75, 100])
+
+        expected = Series([11, Interval(0, 25)], index=["value", "D"], name=10)
+        result = df.iloc[10]
+        tm.assert_series_equal(result, expected)
+
+        expected = DataFrame(
+            {"value": np.arange(11, 21).astype("int64")},
+            index=np.arange(10, 20).astype("int64"),
+        )
+        expected["D"] = pd.cut(expected.value, bins=[0, 25, 50, 75, 100])
+        result = df.iloc[10:20]
+        tm.assert_frame_equal(result, expected)
+
+        expected = Series([9, Interval(0, 25)], index=["value", "D"], name=8)
+        result = df.loc[8]
+        tm.assert_series_equal(result, expected)
+
+    def test_slicing_and_getting_ops(self):
+        # systematically test the slicing operations:
+        #  for all slicing ops:
+        #   - returning a dataframe
+        #   - returning a column
+        #   - returning a row
+        #   - returning a single value
+
+        cats = Categorical(
+            ["a", "c", "b", "c", "c", "c", "c"], categories=["a", "b", "c"]
+        )
+        idx = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values = [1, 2, 3, 4, 5, 6, 7]
+        df = DataFrame({"cats": cats, "values": values}, index=idx)
+
+        # the expected values
+        cats2 = Categorical(["b", "c"], categories=["a", "b", "c"])
+        idx2 = Index(["j", "k"])
+        values2 = [3, 4]
+
+        # 2:4,: | "j":"k",:
+        exp_df = DataFrame({"cats": cats2, "values": values2}, index=idx2)
+
+        # :,"cats" | :,0
+        exp_col = Series(cats, index=idx, name="cats")
+
+        # "j",: | 2,:
+        exp_row = Series(["b", 3], index=["cats", "values"], dtype="object", name="j")
+
+        # "j","cats | 2,0
+        exp_val = "b"
+
+        # iloc
+        # frame
+        res_df = df.iloc[2:4, :]
+        tm.assert_frame_equal(res_df, exp_df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+        # row
+        res_row = df.iloc[2, :]
+        tm.assert_series_equal(res_row, exp_row)
+        assert isinstance(res_row["cats"], str)
+
+        # col
+        res_col = df.iloc[:, 0]
+        tm.assert_series_equal(res_col, exp_col)
+        assert isinstance(res_col.dtype, CategoricalDtype)
+
+        # single value
+        res_val = df.iloc[2, 0]
+        assert res_val == exp_val
+
+        # loc
+        # frame
+        res_df = df.loc["j":"k", :]
+        tm.assert_frame_equal(res_df, exp_df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+        # row
+        res_row = df.loc["j", :]
+        tm.assert_series_equal(res_row, exp_row)
+        assert isinstance(res_row["cats"], str)
+
+        # col
+        res_col = df.loc[:, "cats"]
+        tm.assert_series_equal(res_col, exp_col)
+        assert isinstance(res_col.dtype, CategoricalDtype)
+
+        # single value
+        res_val = df.loc["j", "cats"]
+        assert res_val == exp_val
+
+        # single value
+        res_val = df.loc["j", df.columns[0]]
+        assert res_val == exp_val
+
+        # iat
+        res_val = df.iat[2, 0]
+        assert res_val == exp_val
+
+        # at
+        res_val = df.at["j", "cats"]
+        assert res_val == exp_val
+
+        # fancy indexing
+        exp_fancy = df.iloc[[2]]
+
+        res_fancy = df[df["cats"] == "b"]
+        tm.assert_frame_equal(res_fancy, exp_fancy)
+        res_fancy = df[df["values"] == 3]
+        tm.assert_frame_equal(res_fancy, exp_fancy)
+
+        # get_value
+        res_val = df.at["j", "cats"]
+        assert res_val == exp_val
+
+        # i : int, slice, or sequence of integers
+        res_row = df.iloc[2]
+        tm.assert_series_equal(res_row, exp_row)
+        assert isinstance(res_row["cats"], str)
+
+        res_df = df.iloc[slice(2, 4)]
+        tm.assert_frame_equal(res_df, exp_df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+        res_df = df.iloc[[2, 3]]
+        tm.assert_frame_equal(res_df, exp_df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+        res_col = df.iloc[:, 0]
+        tm.assert_series_equal(res_col, exp_col)
+        assert isinstance(res_col.dtype, CategoricalDtype)
+
+        res_df = df.iloc[:, slice(0, 2)]
+        tm.assert_frame_equal(res_df, df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+        res_df = df.iloc[:, [0, 1]]
+        tm.assert_frame_equal(res_df, df)
+        assert isinstance(res_df["cats"].dtype, CategoricalDtype)
+
+    def test_slicing_doc_examples(self):
+        # GH 7918
+        cats = Categorical(
+            ["a", "b", "b", "b", "c", "c", "c"], categories=["a", "b", "c"]
+        )
+        idx = Index(["h", "i", "j", "k", "l", "m", "n"])
+        values = [1, 2, 2, 2, 3, 4, 5]
+        df = DataFrame({"cats": cats, "values": values}, index=idx)
+
+        result = df.iloc[2:4, :]
+        expected = DataFrame(
+            {
+                "cats": Categorical(["b", "b"], categories=["a", "b", "c"]),
+                "values": [2, 2],
+            },
+            index=["j", "k"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[2:4, :].dtypes
+        expected = Series(["category", "int64"], ["cats", "values"], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        result = df.loc["h":"j", "cats"]
+        expected = Series(
+            Categorical(["a", "b", "b"], categories=["a", "b", "c"]),
+            index=["h", "i", "j"],
+            name="cats",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = df.loc["h":"j", df.columns[0:1]]
+        expected = DataFrame(
+            {"cats": Categorical(["a", "b", "b"], categories=["a", "b", "c"])},
+            index=["h", "i", "j"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_listlike_labels(self, df):
+        # list of labels
+        result = df.loc[["c", "a"]]
+        expected = df.iloc[[4, 0, 1, 5]]
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_loc_getitem_listlike_unused_category(self, df2):
+        # GH#37901 a label that is in index.categories but not in index
+        # listlike containing an element in the categories but not in the values
+        with pytest.raises(KeyError, match=re.escape("['e'] not in index")):
+            df2.loc[["a", "b", "e"]]
+
+    def test_loc_getitem_label_unused_category(self, df2):
+        # element in the categories but not in the values
+        with pytest.raises(KeyError, match=r"^'e'$"):
+            df2.loc["e"]
+
+    def test_loc_getitem_non_category(self, df2):
+        # not all labels in the categories
+        with pytest.raises(KeyError, match=re.escape("['d'] not in index")):
+            df2.loc[["a", "d"]]
+
+    def test_loc_setitem_expansion_label_unused_category(self, df2):
+        # assigning with a label that is in the categories but not in the index
+        df = df2.copy()
+        df.loc["e"] = 20
+        result = df.loc[["a", "b", "e"]]
+        exp_index = CategoricalIndex(list("aaabbe"), categories=list("cabe"), name="B")
+        expected = DataFrame({"A": [0, 1, 5, 2, 3, 20]}, index=exp_index)
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_listlike_dtypes(self):
+        # GH 11586
+
+        # unique categories and codes
+        index = CategoricalIndex(["a", "b", "c"])
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=index)
+
+        # unique slice
+        res = df.loc[["a", "b"]]
+        exp_index = CategoricalIndex(["a", "b"], categories=index.categories)
+        exp = DataFrame({"A": [1, 2], "B": [4, 5]}, index=exp_index)
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        # duplicated slice
+        res = df.loc[["a", "a", "b"]]
+
+        exp_index = CategoricalIndex(["a", "a", "b"], categories=index.categories)
+        exp = DataFrame({"A": [1, 1, 2], "B": [4, 4, 5]}, index=exp_index)
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        with pytest.raises(KeyError, match=re.escape("['x'] not in index")):
+            df.loc[["a", "x"]]
+
+    def test_loc_listlike_dtypes_duplicated_categories_and_codes(self):
+        # duplicated categories and codes
+        index = CategoricalIndex(["a", "b", "a"])
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=index)
+
+        # unique slice
+        res = df.loc[["a", "b"]]
+        exp = DataFrame(
+            {"A": [1, 3, 2], "B": [4, 6, 5]}, index=CategoricalIndex(["a", "a", "b"])
+        )
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        # duplicated slice
+        res = df.loc[["a", "a", "b"]]
+        exp = DataFrame(
+            {"A": [1, 3, 1, 3, 2], "B": [4, 6, 4, 6, 5]},
+            index=CategoricalIndex(["a", "a", "a", "a", "b"]),
+        )
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        with pytest.raises(KeyError, match=re.escape("['x'] not in index")):
+            df.loc[["a", "x"]]
+
+    def test_loc_listlike_dtypes_unused_category(self):
+        # contains unused category
+        index = CategoricalIndex(["a", "b", "a", "c"], categories=list("abcde"))
+        df = DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8]}, index=index)
+
+        res = df.loc[["a", "b"]]
+        exp = DataFrame(
+            {"A": [1, 3, 2], "B": [5, 7, 6]},
+            index=CategoricalIndex(["a", "a", "b"], categories=list("abcde")),
+        )
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        # duplicated slice
+        res = df.loc[["a", "a", "b"]]
+        exp = DataFrame(
+            {"A": [1, 3, 1, 3, 2], "B": [5, 7, 5, 7, 6]},
+            index=CategoricalIndex(["a", "a", "a", "a", "b"], categories=list("abcde")),
+        )
+        tm.assert_frame_equal(res, exp, check_index_type=True)
+
+        with pytest.raises(KeyError, match=re.escape("['x'] not in index")):
+            df.loc[["a", "x"]]
+
+    def test_loc_getitem_listlike_unused_category_raises_keyerror(self):
+        # key that is an *unused* category raises
+        index = CategoricalIndex(["a", "b", "a", "c"], categories=list("abcde"))
+        df = DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8]}, index=index)
+
+        with pytest.raises(KeyError, match="e"):
+            # For comparison, check the scalar behavior
+            df.loc["e"]
+
+        with pytest.raises(KeyError, match=re.escape("['e'] not in index")):
+            df.loc[["a", "e"]]
+
+    def test_ix_categorical_index(self):
+        # GH 12531
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=list("ABC"),
+            columns=list("XYZ"),
+        )
+        cdf = df.copy()
+        cdf.index = CategoricalIndex(df.index)
+        cdf.columns = CategoricalIndex(df.columns)
+
+        expect = Series(df.loc["A", :], index=cdf.columns, name="A")
+        tm.assert_series_equal(cdf.loc["A", :], expect)
+
+        expect = Series(df.loc[:, "X"], index=cdf.index, name="X")
+        tm.assert_series_equal(cdf.loc[:, "X"], expect)
+
+        exp_index = CategoricalIndex(list("AB"), categories=["A", "B", "C"])
+        expect = DataFrame(df.loc[["A", "B"], :], columns=cdf.columns, index=exp_index)
+        tm.assert_frame_equal(cdf.loc[["A", "B"], :], expect)
+
+        exp_columns = CategoricalIndex(list("XY"), categories=["X", "Y", "Z"])
+        expect = DataFrame(df.loc[:, ["X", "Y"]], index=cdf.index, columns=exp_columns)
+        tm.assert_frame_equal(cdf.loc[:, ["X", "Y"]], expect)
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    def test_ix_categorical_index_non_unique(self, infer_string):
+        # non-unique
+        with option_context("future.infer_string", infer_string):
+            df = DataFrame(
+                np.random.default_rng(2).standard_normal((3, 3)),
+                index=list("ABA"),
+                columns=list("XYX"),
+            )
+            cdf = df.copy()
+            cdf.index = CategoricalIndex(df.index)
+            cdf.columns = CategoricalIndex(df.columns)
+
+            exp_index = CategoricalIndex(list("AA"), categories=["A", "B"])
+            expect = DataFrame(df.loc["A", :], columns=cdf.columns, index=exp_index)
+            tm.assert_frame_equal(cdf.loc["A", :], expect)
+
+            exp_columns = CategoricalIndex(list("XX"), categories=["X", "Y"])
+            expect = DataFrame(df.loc[:, "X"], index=cdf.index, columns=exp_columns)
+            tm.assert_frame_equal(cdf.loc[:, "X"], expect)
+
+            expect = DataFrame(
+                df.loc[["A", "B"], :],
+                columns=cdf.columns,
+                index=CategoricalIndex(list("AAB")),
+            )
+            tm.assert_frame_equal(cdf.loc[["A", "B"], :], expect)
+
+            expect = DataFrame(
+                df.loc[:, ["X", "Y"]],
+                index=cdf.index,
+                columns=CategoricalIndex(list("XXY")),
+            )
+            tm.assert_frame_equal(cdf.loc[:, ["X", "Y"]], expect)
+
+    def test_loc_slice(self, df):
+        # GH9748
+        msg = (
+            "cannot do slice indexing on CategoricalIndex with these "
+            r"indexers \[1\] of type int"
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.loc[1:5]
+
+        result = df.loc["b":"c"]
+        expected = df.iloc[[2, 3, 4]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_and_at_with_categorical_index(self):
+        # GH 20629
+        df = DataFrame(
+            [[1, 2], [3, 4], [5, 6]], index=CategoricalIndex(["A", "B", "C"])
+        )
+
+        s = df[0]
+        assert s.loc["A"] == 1
+        assert s.at["A"] == 1
+
+        assert df.loc["B", 1] == 4
+        assert df.at["B", 1] == 4
+
+    @pytest.mark.parametrize(
+        "idx_values",
+        [
+            # python types
+            [1, 2, 3],
+            [-1, -2, -3],
+            [1.5, 2.5, 3.5],
+            [-1.5, -2.5, -3.5],
+            # numpy int/uint
+            *(np.array([1, 2, 3], dtype=dtype) for dtype in tm.ALL_INT_NUMPY_DTYPES),
+            # numpy floats
+            *(np.array([1.5, 2.5, 3.5], dtype=dtyp) for dtyp in tm.FLOAT_NUMPY_DTYPES),
+            # numpy object
+            np.array([1, "b", 3.5], dtype=object),
+            # pandas scalars
+            [Interval(1, 4), Interval(4, 6), Interval(6, 9)],
+            [Timestamp(2019, 1, 1), Timestamp(2019, 2, 1), Timestamp(2019, 3, 1)],
+            [Timedelta(1, "D"), Timedelta(2, "D"), Timedelta(3, "D")],
+            # pandas Integer arrays
+            *(pd.array([1, 2, 3], dtype=dtype) for dtype in tm.ALL_INT_EA_DTYPES),
+            # other pandas arrays
+            pd.IntervalIndex.from_breaks([1, 4, 6, 9]).array,
+            pd.date_range("2019-01-01", periods=3).array,
+            pd.timedelta_range(start="1D", periods=3).array,
+        ],
+    )
+    def test_loc_getitem_with_non_string_categories(self, idx_values, ordered):
+        # GH-17569
+        cat_idx = CategoricalIndex(idx_values, ordered=ordered)
+        df = DataFrame({"A": ["foo", "bar", "baz"]}, index=cat_idx)
+        sl = slice(idx_values[0], idx_values[1])
+
+        # scalar selection
+        result = df.loc[idx_values[0]]
+        expected = Series(["foo"], index=["A"], name=idx_values[0])
+        tm.assert_series_equal(result, expected)
+
+        # list selection
+        result = df.loc[idx_values[:2]]
+        expected = DataFrame(["foo", "bar"], index=cat_idx[:2], columns=["A"])
+        tm.assert_frame_equal(result, expected)
+
+        # slice selection
+        result = df.loc[sl]
+        expected = DataFrame(["foo", "bar"], index=cat_idx[:2], columns=["A"])
+        tm.assert_frame_equal(result, expected)
+
+        # scalar assignment
+        result = df.copy()
+        result.loc[idx_values[0]] = "qux"
+        expected = DataFrame({"A": ["qux", "bar", "baz"]}, index=cat_idx)
+        tm.assert_frame_equal(result, expected)
+
+        # list assignment
+        result = df.copy()
+        result.loc[idx_values[:2], "A"] = ["qux", "qux2"]
+        expected = DataFrame({"A": ["qux", "qux2", "baz"]}, index=cat_idx)
+        tm.assert_frame_equal(result, expected)
+
+        # slice assignment
+        result = df.copy()
+        result.loc[sl, "A"] = ["qux", "qux2"]
+        expected = DataFrame({"A": ["qux", "qux2", "baz"]}, index=cat_idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_categorical_with_nan(self):
+        # GH#41933
+        ci = CategoricalIndex(["A", "B", np.nan])
+
+        ser = Series(range(3), index=ci)
+
+        assert ser[np.nan] == 2
+        assert ser.loc[np.nan] == 2
+
+        df = DataFrame(ser)
+        assert df.loc[np.nan, 0] == 2
+        assert df.loc[np.nan][0] == 2
diff --git a/pandas/pandas/tests/indexing/test_chaining_and_caching.py b/pandas/pandas/tests/indexing/test_chaining_and_caching.py
new file mode 100644
index 0000000000000..64d8068fa9291
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_chaining_and_caching.py
@@ -0,0 +1,392 @@
+from string import ascii_letters
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+    option_context,
+)
+import pandas._testing as tm
+
+msg = "A value is trying to be set on a copy of a slice from a DataFrame"
+
+
+class TestCaching:
+    def test_slice_consolidate_invalidate_item_cache(self):
+        # this is chained assignment, but will 'work'
+        with option_context("chained_assignment", None):
+            # #3970
+            df = DataFrame({"aa": np.arange(5), "bb": [2.2] * 5})
+
+            # Creates a second float block
+            df["cc"] = 0.0
+
+            # caches a reference to the 'bb' series
+            df["bb"]
+
+            # Assignment to wrong series
+            with tm.raises_chained_assignment_error():
+                df["bb"].iloc[0] = 0.17
+            tm.assert_almost_equal(df["bb"][0], 2.2)
+
+    @pytest.mark.parametrize("do_ref", [True, False])
+    def test_setitem_cache_updating(self, do_ref):
+        # GH 5424
+        cont = ["one", "two", "three", "four", "five", "six", "seven"]
+
+        df = DataFrame({"a": cont, "b": cont[3:] + cont[:3], "c": np.arange(7)})
+
+        # ref the cache
+        if do_ref:
+            df.loc[0, "c"]
+
+        # set it
+        df.loc[7, "c"] = 1
+
+        assert df.loc[0, "c"] == 0.0
+        assert df.loc[7, "c"] == 1.0
+
+    def test_setitem_cache_updating_slices(self):
+        # GH 7084
+        # not updating cache on series setting with slices
+        expected = DataFrame(
+            {"A": [600, 600, 600]}, index=date_range("5/7/2014", "5/9/2014")
+        )
+        out = DataFrame({"A": [0, 0, 0]}, index=date_range("5/7/2014", "5/9/2014"))
+        df = DataFrame({"C": ["A", "A", "A"], "D": [100, 200, 300]})
+
+        # loop through df to update out
+        six = Timestamp("5/7/2014")
+        eix = Timestamp("5/9/2014")
+        for ix, row in df.iterrows():
+            out.loc[six:eix, row["C"]] = out.loc[six:eix, row["C"]] + row["D"]
+
+        tm.assert_frame_equal(out, expected)
+        tm.assert_series_equal(out["A"], expected["A"])
+
+        # try via a chain indexing
+        # this actually works
+        out = DataFrame({"A": [0, 0, 0]}, index=date_range("5/7/2014", "5/9/2014"))
+        out_original = out.copy()
+        for ix, row in df.iterrows():
+            v = out[row["C"]][six:eix] + row["D"]
+            with tm.raises_chained_assignment_error():
+                out[row["C"]][six:eix] = v
+
+        tm.assert_frame_equal(out, out_original)
+        tm.assert_series_equal(out["A"], out_original["A"])
+
+        out = DataFrame({"A": [0, 0, 0]}, index=date_range("5/7/2014", "5/9/2014"))
+        for ix, row in df.iterrows():
+            out.loc[six:eix, row["C"]] += row["D"]
+
+        tm.assert_frame_equal(out, expected)
+        tm.assert_series_equal(out["A"], expected["A"])
+
+    def test_altering_series_clears_parent_cache(self):
+        # GH #33675
+        df = DataFrame([[1, 2], [3, 4]], index=["a", "b"], columns=["A", "B"])
+        ser = df["A"]
+
+        # Adding a new entry to ser swaps in a new array, so "A" needs to
+        #  be removed from df._item_cache
+        ser["c"] = 5
+        assert len(ser) == 3
+        assert df["A"] is not ser
+        assert len(df["A"]) == 2
+
+
+class TestChaining:
+    def test_setitem_chained_setfault(self):
+        # GH6026
+        data = ["right", "left", "left", "left", "right", "left", "timeout"]
+
+        df = DataFrame({"response": np.array(data)})
+        mask = df.response == "timeout"
+        with tm.raises_chained_assignment_error():
+            df.response[mask] = "none"
+        tm.assert_frame_equal(df, DataFrame({"response": data}))
+
+        recarray = np.rec.fromarrays([data], names=["response"])
+        df = DataFrame(recarray)
+        mask = df.response == "timeout"
+        with tm.raises_chained_assignment_error():
+            df.response[mask] = "none"
+        tm.assert_frame_equal(df, DataFrame({"response": data}))
+
+        df = DataFrame({"response": data, "response1": data})
+        df_original = df.copy()
+        mask = df.response == "timeout"
+        with tm.raises_chained_assignment_error():
+            df.response[mask] = "none"
+        tm.assert_frame_equal(df, df_original)
+
+        # GH 6056
+        expected = DataFrame({"A": [np.nan, "bar", "bah", "foo", "bar"]})
+        df = DataFrame({"A": np.array(["foo", "bar", "bah", "foo", "bar"])})
+        with tm.raises_chained_assignment_error():
+            df["A"].iloc[0] = np.nan
+        expected = DataFrame({"A": ["foo", "bar", "bah", "foo", "bar"]})
+        result = df.head()
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({"A": np.array(["foo", "bar", "bah", "foo", "bar"])})
+        with tm.raises_chained_assignment_error():
+            df.A.iloc[0] = np.nan
+        result = df.head()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment(self):
+        with option_context("chained_assignment", "raise"):
+            # work with the chain
+            df = DataFrame(
+                np.arange(4).reshape(2, 2), columns=list("AB"), dtype="int64"
+            )
+            df_original = df.copy()
+
+            with tm.raises_chained_assignment_error():
+                df["A"][0] = -5
+            with tm.raises_chained_assignment_error():
+                df["A"][1] = -6
+            tm.assert_frame_equal(df, df_original)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_raises(self):
+        # test with the chaining
+        df = DataFrame(
+            {
+                "A": Series(range(2), dtype="int64"),
+                "B": np.array(np.arange(2, 4), dtype=np.float64),
+            }
+        )
+        df_original = df.copy()
+        with tm.raises_chained_assignment_error():
+            df["A"][0] = -5
+        with tm.raises_chained_assignment_error():
+            df["A"][1] = -6
+        tm.assert_frame_equal(df, df_original)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_fails(self):
+        # Using a copy (the chain), fails
+        df = DataFrame(
+            {
+                "A": Series(range(2), dtype="int64"),
+                "B": np.array(np.arange(2, 4), dtype=np.float64),
+            }
+        )
+
+        with tm.raises_chained_assignment_error():
+            df.loc[0]["A"] = -5
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_doc_example(self):
+        # Doc example
+        df = DataFrame(
+            {
+                "a": ["one", "one", "two", "three", "two", "one", "six"],
+                "c": Series(range(7), dtype="int64"),
+            }
+        )
+
+        indexer = df.a.str.startswith("o")
+        with tm.raises_chained_assignment_error():
+            df[indexer]["c"] = 42
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_object_dtype(self):
+        df = DataFrame(
+            {"A": Series(["aaa", "bbb", "ccc"], dtype=object), "B": [1, 2, 3]}
+        )
+        df_original = df.copy()
+
+        with tm.raises_chained_assignment_error():
+            df["A"][0] = 111
+        tm.assert_frame_equal(df, df_original)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_is_copy_pickle(self, temp_file):
+        # gh-5475: Make sure that is_copy is picked up reconstruction
+        df = DataFrame({"A": [1, 2]})
+
+        path = str(temp_file)
+        df.to_pickle(path)
+        df2 = pd.read_pickle(path)
+        df2["B"] = df2["A"]
+        df2["B"] = df2["A"]
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_str(self):
+        idxs = np.random.default_rng(2).integers(len(ascii_letters), size=(100, 2))
+        idxs.sort(axis=1)
+        strings = [ascii_letters[x[0] : x[1]] for x in idxs]
+
+        df = DataFrame(strings, columns=["letters"])
+        indexer = df.letters.apply(lambda x: len(x) > 10)
+        df.loc[indexer, "letters"] = df.loc[indexer, "letters"].apply(str.lower)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_sorting(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        ser = df.iloc[:, 0].sort_values()
+
+        tm.assert_series_equal(ser, df.iloc[:, 0].sort_values())
+        tm.assert_series_equal(ser, df[0].sort_values())
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_false_positives(self):
+        # see gh-6025: false positives
+        df = DataFrame({"column1": ["a", "a", "a"], "column2": [4, 8, 9]})
+        str(df)
+
+        df["column1"] = df["column1"] + "b"
+        str(df)
+
+        df = df[df["column2"] != 8]
+        str(df)
+
+        df["column1"] = df["column1"] + "c"
+        str(df)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_undefined_column(self):
+        # from SO:
+        # https://stackoverflow.com/questions/24054495/potential-bug-setting-value-for-undefined-column-using-iloc
+        df = DataFrame(np.arange(0, 9), columns=["count"])
+        df["group"] = "b"
+        df_original = df.copy()
+        with tm.raises_chained_assignment_error():
+            df.iloc[0:5]["group"] = "a"
+        tm.assert_frame_equal(df, df_original)
+
+    @pytest.mark.arm_slow
+    def test_detect_chained_assignment_changing_dtype(self):
+        # Mixed type setting but same dtype & changing dtype
+        df = DataFrame(
+            {
+                "A": date_range("20130101", periods=5),
+                "B": np.random.default_rng(2).standard_normal(5),
+                "C": np.arange(5, dtype="int64"),
+                "D": ["a", "b", "c", "d", "e"],
+            }
+        )
+        df_original = df.copy()
+
+        with tm.raises_chained_assignment_error():
+            df.loc[2]["D"] = "foo"
+        with tm.raises_chained_assignment_error():
+            df.loc[2]["C"] = "foo"
+        tm.assert_frame_equal(df, df_original)
+        # TODO: Use tm.raises_chained_assignment_error() when PDEP-6 is enforced
+        with pytest.raises(TypeError, match="Invalid value"):
+            with tm.raises_chained_assignment_error():
+                df["C"][2] = "foo"
+
+    def test_setting_with_copy_bug(self):
+        # operating on a copy
+        df = DataFrame(
+            {"a": list(range(4)), "b": list("ab.."), "c": ["a", "b", np.nan, "d"]}
+        )
+        df_original = df.copy()
+        mask = pd.isna(df.c)
+        with tm.raises_chained_assignment_error():
+            df[["c"]][mask] = df[["b"]][mask]
+        tm.assert_frame_equal(df, df_original)
+
+    def test_setting_with_copy_bug_no_warning(self):
+        # invalid warning as we are returning a new object
+        # GH 8730
+        df1 = DataFrame({"x": Series(["a", "b", "c"]), "y": Series(["d", "e", "f"])})
+        df2 = df1[["x"]]
+
+        # this should not raise
+        df2["y"] = ["g", "h", "i"]
+
+    def test_detect_chained_assignment_warnings_errors(self):
+        df = DataFrame({"A": ["aaa", "bbb", "ccc"], "B": [1, 2, 3]})
+        with tm.raises_chained_assignment_error():
+            df.loc[0]["A"] = 111
+
+    @pytest.mark.parametrize("rhs", [3, DataFrame({0: [1, 2, 3, 4]})])
+    def test_detect_chained_assignment_warning_stacklevel(self, rhs):
+        # GH#42570
+        df = DataFrame(np.arange(25).reshape(5, 5))
+        df_original = df.copy()
+        chained = df.loc[:3]
+        chained[2] = rhs
+        tm.assert_frame_equal(df, df_original)
+
+    def test_chained_getitem_with_lists(self):
+        # GH6394
+        # Regression in chained getitem indexing with embedded list-like from
+        # 0.12
+
+        df = DataFrame({"A": 5 * [np.zeros(3)], "B": 5 * [np.ones(3)]})
+        expected = df["A"].iloc[2]
+        result = df.loc[2, "A"]
+        tm.assert_numpy_array_equal(result, expected)
+        result2 = df.iloc[2]["A"]
+        tm.assert_numpy_array_equal(result2, expected)
+        result3 = df["A"].loc[2]
+        tm.assert_numpy_array_equal(result3, expected)
+        result4 = df["A"].iloc[2]
+        tm.assert_numpy_array_equal(result4, expected)
+
+    def test_cache_updating(self):
+        # GH 4939, make sure to update the cache on setitem
+
+        df = DataFrame(
+            np.zeros((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+        )
+        df["A"]  # cache series
+        df.loc["Hello Friend"] = df.iloc[0]
+        assert "Hello Friend" in df["A"].index
+        assert "Hello Friend" in df["B"].index
+
+    def test_cache_updating2(self):
+        # 10264
+        df = DataFrame(
+            np.zeros((5, 5), dtype="int64"),
+            columns=["a", "b", "c", "d", "e"],
+            index=range(5),
+        )
+        df["f"] = 0
+        df_orig = df.copy()
+        with pytest.raises(ValueError, match="read-only"):
+            df.f.values[3] = 1
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_iloc_setitem_chained_assignment(self):
+        # GH#3970
+        with option_context("chained_assignment", None):
+            df = DataFrame({"aa": range(5), "bb": [2.2] * 5})
+            df["cc"] = 0.0
+
+            ck = [True] * len(df)
+
+            with tm.raises_chained_assignment_error():
+                df["bb"].iloc[0] = 0.13
+
+            # GH#3970 this lookup used to break the chained setting to 0.15
+            df.iloc[ck]
+
+            with tm.raises_chained_assignment_error():
+                df["bb"].iloc[0] = 0.15
+
+            assert df["bb"].iloc[0] == 2.2
+
+    def test_getitem_loc_assignment_slice_state(self):
+        # GH 13569
+        df = DataFrame({"a": [10, 20, 30]})
+        with tm.raises_chained_assignment_error():
+            df["a"].loc[4] = 40
+        tm.assert_frame_equal(df, DataFrame({"a": [10, 20, 30]}))
+        tm.assert_series_equal(df["a"], Series([10, 20, 30], name="a"))
diff --git a/pandas/pandas/tests/indexing/test_check_indexer.py b/pandas/pandas/tests/indexing/test_check_indexer.py
new file mode 100644
index 0000000000000..975a31b873792
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_check_indexer.py
@@ -0,0 +1,105 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.indexers import check_array_indexer
+
+
+@pytest.mark.parametrize(
+    "indexer, expected",
+    [
+        # integer
+        ([1, 2], np.array([1, 2], dtype=np.intp)),
+        (np.array([1, 2], dtype="int64"), np.array([1, 2], dtype=np.intp)),
+        (pd.array([1, 2], dtype="Int32"), np.array([1, 2], dtype=np.intp)),
+        (pd.Index([1, 2]), np.array([1, 2], dtype=np.intp)),
+        # boolean
+        ([True, False, True], np.array([True, False, True], dtype=np.bool_)),
+        (np.array([True, False, True]), np.array([True, False, True], dtype=np.bool_)),
+        (
+            pd.array([True, False, True], dtype="boolean"),
+            np.array([True, False, True], dtype=np.bool_),
+        ),
+        # other
+        ([], np.array([], dtype=np.intp)),
+    ],
+)
+def test_valid_input(indexer, expected):
+    arr = np.array([1, 2, 3])
+    result = check_array_indexer(arr, indexer)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer", [[True, False, None], pd.array([True, False, None], dtype="boolean")]
+)
+def test_boolean_na_returns_indexer(indexer):
+    # https://github.com/pandas-dev/pandas/issues/31503
+    arr = np.array([1, 2, 3])
+
+    result = check_array_indexer(arr, indexer)
+    expected = np.array([True, False, False], dtype=bool)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer",
+    [
+        [True, False],
+        pd.array([True, False], dtype="boolean"),
+        np.array([True, False], dtype=np.bool_),
+    ],
+)
+def test_bool_raise_length(indexer):
+    arr = np.array([1, 2, 3])
+
+    msg = "Boolean index has wrong length"
+    with pytest.raises(IndexError, match=msg):
+        check_array_indexer(arr, indexer)
+
+
+@pytest.mark.parametrize(
+    "indexer", [[0, 1, None], pd.array([0, 1, pd.NA], dtype="Int64")]
+)
+def test_int_raise_missing_values(indexer):
+    arr = np.array([1, 2, 3])
+
+    msg = "Cannot index with an integer indexer containing NA values"
+    with pytest.raises(ValueError, match=msg):
+        check_array_indexer(arr, indexer)
+
+
+@pytest.mark.parametrize(
+    "indexer",
+    [
+        [0.0, 1.0],
+        np.array([1.0, 2.0], dtype="float64"),
+        np.array([True, False], dtype=object),
+        pd.Index([True, False], dtype=object),
+    ],
+)
+def test_raise_invalid_array_dtypes(indexer):
+    arr = np.array([1, 2, 3])
+
+    msg = "arrays used as indices must be of integer or boolean type"
+    with pytest.raises(IndexError, match=msg):
+        check_array_indexer(arr, indexer)
+
+
+def test_raise_nullable_string_dtype(nullable_string_dtype):
+    indexer = pd.array(["a", "b"], dtype=nullable_string_dtype)
+    arr = np.array([1, 2, 3])
+
+    msg = "arrays used as indices must be of integer or boolean type"
+    with pytest.raises(IndexError, match=msg):
+        check_array_indexer(arr, indexer)
+
+
+@pytest.mark.parametrize("indexer", [None, Ellipsis, slice(0, 3), (None,)])
+def test_pass_through_non_array_likes(indexer):
+    arr = np.array([1, 2, 3])
+
+    result = check_array_indexer(arr, indexer)
+    assert result == indexer
diff --git a/pandas/pandas/tests/indexing/test_coercion.py b/pandas/pandas/tests/indexing/test_coercion.py
new file mode 100644
index 0000000000000..d5002a47c3447
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_coercion.py
@@ -0,0 +1,907 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+import itertools
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    is_platform_windows,
+)
+from pandas.compat.numpy import np_version_gt2
+
+import pandas as pd
+import pandas._testing as tm
+
+###############################################################
+# Index / Series common tests which may trigger dtype coercions
+###############################################################
+
+
+@pytest.fixture(autouse=True, scope="class")
+def check_comprehensiveness(request):
+    # Iterate over combination of dtype, method and klass
+    # and ensure that each are contained within a collected test
+    cls = request.cls
+    combos = itertools.product(cls.klasses, cls.dtypes, [cls.method])
+
+    def has_test(combo):
+        klass, dtype, method = combo
+        cls_funcs = request.node.session.items
+        return any(
+            klass in x.name and dtype in x.name and method in x.name for x in cls_funcs
+        )
+
+    opts = request.config.option
+    if opts.lf or opts.keyword:
+        # If we are running with "last-failed" or -k foo, we expect to only
+        #  run a subset of tests.
+        yield
+
+    else:
+        for combo in combos:
+            if not has_test(combo):
+                raise AssertionError(
+                    f"test method is not defined: {cls.__name__}, {combo}"
+                )
+
+        yield
+
+
+class CoercionBase:
+    klasses = ["index", "series"]
+    dtypes = [
+        "object",
+        "int64",
+        "float64",
+        "complex128",
+        "bool",
+        "datetime64",
+        "datetime64tz",
+        "timedelta64",
+        "period",
+    ]
+
+    @property
+    def method(self):
+        raise NotImplementedError(self)
+
+
+class TestSetitemCoercion(CoercionBase):
+    method = "setitem"
+
+    # disable comprehensiveness tests, as most of these have been moved to
+    #  tests.series.indexing.test_setitem in SetitemCastingEquivalents subclasses.
+    klasses: list[str] = []
+
+    def test_setitem_series_no_coercion_from_values_list(self):
+        # GH35865 - int casted to str when internally calling np.array(ser.values)
+        ser = pd.Series(["a", 1])
+        ser[:] = list(ser.values)
+
+        expected = pd.Series(["a", 1])
+
+        tm.assert_series_equal(ser, expected)
+
+    def _assert_setitem_index_conversion(
+        self, original_series, loc_key, expected_index, expected_dtype
+    ):
+        """test index's coercion triggered by assign key"""
+        temp = original_series.copy()
+        # GH#33469 pre-2.0 with int loc_key and temp.index.dtype == np.float64
+        #  `temp[loc_key] = 5` treated loc_key as positional
+        temp[loc_key] = 5
+        exp = pd.Series([1, 2, 3, 4, 5], index=expected_index)
+        tm.assert_series_equal(temp, exp)
+        # check dtype explicitly for sure
+        assert temp.index.dtype == expected_dtype
+
+        temp = original_series.copy()
+        temp.loc[loc_key] = 5
+        exp = pd.Series([1, 2, 3, 4, 5], index=expected_index)
+        tm.assert_series_equal(temp, exp)
+        # check dtype explicitly for sure
+        assert temp.index.dtype == expected_dtype
+
+    @pytest.mark.parametrize(
+        "val,exp_dtype", [("x", object), (5, IndexError), (1.1, object)]
+    )
+    def test_setitem_index_object(self, val, exp_dtype):
+        obj = pd.Series([1, 2, 3, 4], index=pd.Index(list("abcd"), dtype=object))
+        assert obj.index.dtype == object
+
+        exp_index = pd.Index(list("abcd") + [val], dtype=object)
+        self._assert_setitem_index_conversion(obj, val, exp_index, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "val,exp_dtype", [(5, np.int64), (1.1, np.float64), ("x", object)]
+    )
+    def test_setitem_index_int64(self, val, exp_dtype):
+        obj = pd.Series([1, 2, 3, 4])
+        assert obj.index.dtype == np.int64
+
+        exp_index = pd.Index([0, 1, 2, 3, val])
+        self._assert_setitem_index_conversion(obj, val, exp_index, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "val,exp_dtype", [(5, np.float64), (5.1, np.float64), ("x", object)]
+    )
+    def test_setitem_index_float64(self, val, exp_dtype, request):
+        obj = pd.Series([1, 2, 3, 4], index=[1.1, 2.1, 3.1, 4.1])
+        assert obj.index.dtype == np.float64
+
+        exp_index = pd.Index([1.1, 2.1, 3.1, 4.1, val])
+        self._assert_setitem_index_conversion(obj, val, exp_index, exp_dtype)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_series_period(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_complex128(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_bool(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_datetime64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_datetime64tz(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_timedelta64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_setitem_index_period(self):
+        raise NotImplementedError
+
+
+class TestInsertIndexCoercion(CoercionBase):
+    klasses = ["index"]
+    method = "insert"
+
+    def _assert_insert_conversion(self, original, value, expected, expected_dtype):
+        """test coercion triggered by insert"""
+        target = original.copy()
+        res = target.insert(1, value)
+        tm.assert_index_equal(res, expected)
+        assert res.dtype == expected_dtype
+
+    @pytest.mark.parametrize(
+        "insert, coerced_val, coerced_dtype",
+        [
+            (1, 1, object),
+            (1.1, 1.1, object),
+            (False, False, object),
+            ("x", "x", object),
+        ],
+    )
+    def test_insert_index_object(self, insert, coerced_val, coerced_dtype):
+        obj = pd.Index(list("abcd"), dtype=object)
+        assert obj.dtype == object
+
+        exp = pd.Index(["a", coerced_val, "b", "c", "d"], dtype=object)
+        self._assert_insert_conversion(obj, insert, exp, coerced_dtype)
+
+    @pytest.mark.parametrize(
+        "insert, coerced_val, coerced_dtype",
+        [
+            (1, 1, None),
+            (1.1, 1.1, np.float64),
+            (False, False, object),  # GH#36319
+            ("x", "x", object),
+        ],
+    )
+    def test_insert_int_index(
+        self, any_int_numpy_dtype, insert, coerced_val, coerced_dtype
+    ):
+        dtype = any_int_numpy_dtype
+        obj = pd.Index([1, 2, 3, 4], dtype=dtype)
+        coerced_dtype = coerced_dtype if coerced_dtype is not None else dtype
+
+        exp = pd.Index([1, coerced_val, 2, 3, 4], dtype=coerced_dtype)
+        self._assert_insert_conversion(obj, insert, exp, coerced_dtype)
+
+    @pytest.mark.parametrize(
+        "insert, coerced_val, coerced_dtype",
+        [
+            (1, 1.0, None),
+            # When float_numpy_dtype=float32, this is not the case
+            # see the correction below
+            (1.1, 1.1, np.float64),
+            (False, False, object),  # GH#36319
+            ("x", "x", object),
+        ],
+    )
+    def test_insert_float_index(
+        self, float_numpy_dtype, insert, coerced_val, coerced_dtype
+    ):
+        dtype = float_numpy_dtype
+        obj = pd.Index([1.0, 2.0, 3.0, 4.0], dtype=dtype)
+        coerced_dtype = coerced_dtype if coerced_dtype is not None else dtype
+
+        if np_version_gt2 and dtype == "float32" and coerced_val == 1.1:
+            # Hack, in the 2nd test case, since 1.1 can be losslessly cast to float32
+            # the expected dtype will be float32 if the original dtype was float32
+            coerced_dtype = np.float32
+        exp = pd.Index([1.0, coerced_val, 2.0, 3.0, 4.0], dtype=coerced_dtype)
+        self._assert_insert_conversion(obj, insert, exp, coerced_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [
+            (pd.Timestamp("2012-01-01"), "datetime64[ns]"),
+            (pd.Timestamp("2012-01-01", tz="US/Eastern"), "datetime64[ns, US/Eastern]"),
+        ],
+        ids=["datetime64", "datetime64tz"],
+    )
+    @pytest.mark.parametrize(
+        "insert_value",
+        [pd.Timestamp("2012-01-01"), pd.Timestamp("2012-01-01", tz="Asia/Tokyo"), 1],
+    )
+    def test_insert_index_datetimes(self, fill_val, exp_dtype, insert_value):
+        obj = pd.DatetimeIndex(
+            ["2011-01-01", "2011-01-02", "2011-01-03", "2011-01-04"], tz=fill_val.tz
+        ).as_unit("ns")
+        assert obj.dtype == exp_dtype
+
+        exp = pd.DatetimeIndex(
+            ["2011-01-01", fill_val.date(), "2011-01-02", "2011-01-03", "2011-01-04"],
+            tz=fill_val.tz,
+        ).as_unit("ns")
+        self._assert_insert_conversion(obj, fill_val, exp, exp_dtype)
+
+        if fill_val.tz:
+            # mismatched tzawareness
+            ts = pd.Timestamp("2012-01-01")
+            result = obj.insert(1, ts)
+            expected = obj.astype(object).insert(1, ts)
+            assert expected.dtype == object
+            tm.assert_index_equal(result, expected)
+
+            ts = pd.Timestamp("2012-01-01", tz="Asia/Tokyo")
+            result = obj.insert(1, ts)
+            # once deprecation is enforced:
+            expected = obj.insert(1, ts.tz_convert(obj.dtype.tz))
+            assert expected.dtype == obj.dtype
+            tm.assert_index_equal(result, expected)
+
+        else:
+            # mismatched tzawareness
+            ts = pd.Timestamp("2012-01-01", tz="Asia/Tokyo")
+            result = obj.insert(1, ts)
+            expected = obj.astype(object).insert(1, ts)
+            assert expected.dtype == object
+            tm.assert_index_equal(result, expected)
+
+        item = 1
+        result = obj.insert(1, item)
+        expected = obj.astype(object).insert(1, item)
+        assert expected[1] == item
+        assert expected.dtype == object
+        tm.assert_index_equal(result, expected)
+
+    def test_insert_index_timedelta64(self):
+        obj = pd.TimedeltaIndex(["1 day", "2 day", "3 day", "4 day"])
+        assert obj.dtype == "timedelta64[ns]"
+
+        # timedelta64 + timedelta64 => timedelta64
+        exp = pd.TimedeltaIndex(["1 day", "10 day", "2 day", "3 day", "4 day"])
+        self._assert_insert_conversion(
+            obj, pd.Timedelta("10 day"), exp, "timedelta64[ns]"
+        )
+
+        for item in [pd.Timestamp("2012-01-01"), 1]:
+            result = obj.insert(1, item)
+            expected = obj.astype(object).insert(1, item)
+            assert expected.dtype == object
+            tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "insert, coerced_val, coerced_dtype",
+        [
+            (pd.Period("2012-01", freq="M"), "2012-01", "period[M]"),
+            (pd.Timestamp("2012-01-01"), pd.Timestamp("2012-01-01"), object),
+            (1, 1, object),
+            ("x", "x", object),
+        ],
+    )
+    def test_insert_index_period(self, insert, coerced_val, coerced_dtype):
+        obj = pd.PeriodIndex(["2011-01", "2011-02", "2011-03", "2011-04"], freq="M")
+        assert obj.dtype == "period[M]"
+
+        data = [
+            pd.Period("2011-01", freq="M"),
+            coerced_val,
+            pd.Period("2011-02", freq="M"),
+            pd.Period("2011-03", freq="M"),
+            pd.Period("2011-04", freq="M"),
+        ]
+        if isinstance(insert, pd.Period):
+            exp = pd.PeriodIndex(data, freq="M")
+            self._assert_insert_conversion(obj, insert, exp, coerced_dtype)
+
+            # string that can be parsed to appropriate PeriodDtype
+            self._assert_insert_conversion(obj, str(insert), exp, coerced_dtype)
+
+        else:
+            result = obj.insert(0, insert)
+            expected = obj.astype(object).insert(0, insert)
+            tm.assert_index_equal(result, expected)
+
+            # TODO: ATM inserting '2012-01-01 00:00:00' when we have obj.freq=="M"
+            #  casts that string to Period[M], not clear that is desirable
+            if not isinstance(insert, pd.Timestamp):
+                # non-castable string
+                result = obj.insert(0, str(insert))
+                expected = obj.astype(object).insert(0, str(insert))
+                tm.assert_index_equal(result, expected)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_insert_index_complex128(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_insert_index_bool(self):
+        raise NotImplementedError
+
+
+class TestWhereCoercion(CoercionBase):
+    method = "where"
+    _cond = np.array([True, False, True, False])
+
+    def _assert_where_conversion(
+        self, original, cond, values, expected, expected_dtype
+    ):
+        """test coercion triggered by where"""
+        target = original.copy()
+        res = target.where(cond, values)
+        tm.assert_equal(res, expected)
+        assert res.dtype == expected_dtype
+
+    def _construct_exp(self, obj, klass, fill_val, exp_dtype):
+        if fill_val is True:
+            values = klass([True, False, True, True])
+        elif isinstance(fill_val, (datetime, np.datetime64)):
+            values = pd.date_range(fill_val, periods=4)
+        else:
+            values = klass(x * fill_val for x in [5, 6, 7, 8])
+
+        exp = klass([obj[0], values[1], obj[2], values[3]], dtype=exp_dtype)
+        return values, exp
+
+    def _run_test(self, obj, fill_val, klass, exp_dtype):
+        cond = klass(self._cond)
+
+        exp = klass([obj[0], fill_val, obj[2], fill_val], dtype=exp_dtype)
+        self._assert_where_conversion(obj, cond, fill_val, exp, exp_dtype)
+
+        values, exp = self._construct_exp(obj, klass, fill_val, exp_dtype)
+        self._assert_where_conversion(obj, cond, values, exp, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [(1, object), (1.1, object), (1 + 1j, object), (True, object)],
+    )
+    def test_where_object(self, index_or_series, fill_val, exp_dtype):
+        klass = index_or_series
+        obj = klass(list("abcd"), dtype=object)
+        assert obj.dtype == object
+        self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [(1, np.int64), (1.1, np.float64), (1 + 1j, np.complex128), (True, object)],
+    )
+    def test_where_int64(self, index_or_series, fill_val, exp_dtype, request):
+        klass = index_or_series
+
+        obj = klass([1, 2, 3, 4])
+        assert obj.dtype == np.int64
+        self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val, exp_dtype",
+        [(1, np.float64), (1.1, np.float64), (1 + 1j, np.complex128), (True, object)],
+    )
+    def test_where_float64(self, index_or_series, fill_val, exp_dtype, request):
+        klass = index_or_series
+
+        obj = klass([1.1, 2.2, 3.3, 4.4])
+        assert obj.dtype == np.float64
+        self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [
+            (1, np.complex128),
+            (1.1, np.complex128),
+            (1 + 1j, np.complex128),
+            (True, object),
+        ],
+    )
+    def test_where_complex128(self, index_or_series, fill_val, exp_dtype):
+        klass = index_or_series
+        obj = klass([1 + 1j, 2 + 2j, 3 + 3j, 4 + 4j], dtype=np.complex128)
+        assert obj.dtype == np.complex128
+        self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [(1, object), (1.1, object), (1 + 1j, object), (True, np.bool_)],
+    )
+    def test_where_series_bool(self, index_or_series, fill_val, exp_dtype):
+        klass = index_or_series
+
+        obj = klass([True, False, True, False])
+        assert obj.dtype == np.bool_
+        self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,exp_dtype",
+        [
+            (pd.Timestamp("2012-01-01"), "datetime64[ns]"),
+            (pd.Timestamp("2012-01-01", tz="US/Eastern"), object),
+        ],
+        ids=["datetime64", "datetime64tz"],
+    )
+    def test_where_datetime64(self, index_or_series, fill_val, exp_dtype):
+        klass = index_or_series
+
+        obj = klass(pd.date_range("2011-01-01", periods=4, freq="D")._with_freq(None))
+        assert obj.dtype == "datetime64[ns]"
+
+        fv = fill_val
+        # do the check with each of the available datetime scalars
+        if exp_dtype == "datetime64[ns]":
+            for scalar in [fv, fv.to_pydatetime(), fv.to_datetime64()]:
+                self._run_test(obj, scalar, klass, exp_dtype)
+        else:
+            for scalar in [fv, fv.to_pydatetime()]:
+                self._run_test(obj, fill_val, klass, exp_dtype)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_where_index_complex128(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_where_index_bool(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_where_series_timedelta64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_where_series_period(self):
+        raise NotImplementedError
+
+    @pytest.mark.parametrize(
+        "value", [pd.Timedelta(days=9), timedelta(days=9), np.timedelta64(9, "D")]
+    )
+    def test_where_index_timedelta64(self, value):
+        tdi = pd.timedelta_range("1 Day", periods=4)
+        cond = np.array([True, False, False, True])
+
+        expected = pd.TimedeltaIndex(["1 Day", value, value, "4 Days"])
+        result = tdi.where(cond, value)
+        tm.assert_index_equal(result, expected)
+
+        # wrong-dtyped NaT
+        dtnat = np.datetime64("NaT", "ns")
+        expected = pd.Index([tdi[0], dtnat, dtnat, tdi[3]], dtype=object)
+        assert expected[1] is dtnat
+
+        result = tdi.where(cond, dtnat)
+        tm.assert_index_equal(result, expected)
+
+    def test_where_index_period(self):
+        dti = pd.date_range("2016-01-01", periods=3, freq="QS")
+        pi = dti.to_period("Q")
+
+        cond = np.array([False, True, False])
+
+        # Passing a valid scalar
+        value = pi[-1] + pi.freq * 10
+        expected = pd.PeriodIndex([value, pi[1], value])
+        result = pi.where(cond, value)
+        tm.assert_index_equal(result, expected)
+
+        # Case passing ndarray[object] of Periods
+        other = np.asarray(pi + pi.freq * 10, dtype=object)
+        result = pi.where(cond, other)
+        expected = pd.PeriodIndex([other[0], pi[1], other[2]])
+        tm.assert_index_equal(result, expected)
+
+        # Passing a mismatched scalar -> casts to object
+        td = pd.Timedelta(days=4)
+        expected = pd.Index([td, pi[1], td], dtype=object)
+        result = pi.where(cond, td)
+        tm.assert_index_equal(result, expected)
+
+        per = pd.Period("2020-04-21", "D")
+        expected = pd.Index([per, pi[1], per], dtype=object)
+        result = pi.where(cond, per)
+        tm.assert_index_equal(result, expected)
+
+
+class TestFillnaSeriesCoercion(CoercionBase):
+    # not indexing, but place here for consistency
+
+    method = "fillna"
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_has_comprehensive_tests(self):
+        raise NotImplementedError
+
+    def _assert_fillna_conversion(self, original, value, expected, expected_dtype):
+        """test coercion triggered by fillna"""
+        target = original.copy()
+        res = target.fillna(value)
+        tm.assert_equal(res, expected)
+        assert res.dtype == expected_dtype
+
+    @pytest.mark.parametrize(
+        "fill_val, fill_dtype",
+        [(1, object), (1.1, object), (1 + 1j, object), (True, object)],
+    )
+    def test_fillna_object(self, index_or_series, fill_val, fill_dtype):
+        klass = index_or_series
+        obj = klass(["a", np.nan, "c", "d"], dtype=object)
+        assert obj.dtype == object
+
+        exp = klass(["a", fill_val, "c", "d"], dtype=object)
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,fill_dtype",
+        [(1, np.float64), (1.1, np.float64), (1 + 1j, np.complex128), (True, object)],
+    )
+    def test_fillna_float64(self, index_or_series, fill_val, fill_dtype):
+        klass = index_or_series
+        obj = klass([1.1, np.nan, 3.3, 4.4])
+        assert obj.dtype == np.float64
+
+        exp = klass([1.1, fill_val, 3.3, 4.4])
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,fill_dtype",
+        [
+            (1, np.complex128),
+            (1.1, np.complex128),
+            (1 + 1j, np.complex128),
+            (True, object),
+        ],
+    )
+    def test_fillna_complex128(self, index_or_series, fill_val, fill_dtype):
+        klass = index_or_series
+        obj = klass([1 + 1j, np.nan, 3 + 3j, 4 + 4j], dtype=np.complex128)
+        assert obj.dtype == np.complex128
+
+        exp = klass([1 + 1j, fill_val, 3 + 3j, 4 + 4j])
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,fill_dtype",
+        [
+            (pd.Timestamp("2012-01-01"), "datetime64[s]"),
+            (pd.Timestamp("2012-01-01", tz="US/Eastern"), object),
+            (1, object),
+            ("x", object),
+        ],
+        ids=["datetime64", "datetime64tz", "object", "object"],
+    )
+    def test_fillna_datetime(self, index_or_series, fill_val, fill_dtype):
+        klass = index_or_series
+        obj = klass(
+            [
+                pd.Timestamp("2011-01-01"),
+                pd.NaT,
+                pd.Timestamp("2011-01-03"),
+                pd.Timestamp("2011-01-04"),
+            ]
+        )
+        assert obj.dtype == "datetime64[s]"
+
+        exp = klass(
+            [
+                pd.Timestamp("2011-01-01"),
+                fill_val,
+                pd.Timestamp("2011-01-03"),
+                pd.Timestamp("2011-01-04"),
+            ]
+        )
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val,fill_dtype",
+        [
+            (pd.Timestamp("2012-01-01", tz="US/Eastern"), "datetime64[s, US/Eastern]"),
+            (pd.Timestamp("2012-01-01"), object),
+            # pre-2.0 with a mismatched tz we would get object result
+            (pd.Timestamp("2012-01-01", tz="Asia/Tokyo"), "datetime64[s, US/Eastern]"),
+            (1, object),
+            ("x", object),
+        ],
+    )
+    def test_fillna_datetime64tz(self, index_or_series, fill_val, fill_dtype):
+        klass = index_or_series
+        tz = "US/Eastern"
+
+        obj = klass(
+            [
+                pd.Timestamp("2011-01-01", tz=tz),
+                pd.NaT,
+                pd.Timestamp("2011-01-03", tz=tz),
+                pd.Timestamp("2011-01-04", tz=tz),
+            ]
+        )
+        assert obj.dtype == "datetime64[s, US/Eastern]"
+
+        if getattr(fill_val, "tz", None) is None:
+            fv = fill_val
+        else:
+            fv = fill_val.tz_convert(tz)
+        exp = klass(
+            [
+                pd.Timestamp("2011-01-01", tz=tz),
+                fv,
+                pd.Timestamp("2011-01-03", tz=tz),
+                pd.Timestamp("2011-01-04", tz=tz),
+            ]
+        )
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.parametrize(
+        "fill_val",
+        [
+            1,
+            1.1,
+            1 + 1j,
+            True,
+            pd.Interval(1, 2, closed="left"),
+            pd.Timestamp("2012-01-01", tz="US/Eastern"),
+            pd.Timestamp("2012-01-01"),
+            pd.Timedelta(days=1),
+            pd.Period("2016-01-01", "D"),
+        ],
+    )
+    def test_fillna_interval(self, index_or_series, fill_val):
+        ii = pd.interval_range(1.0, 5.0, closed="right").insert(1, np.nan)
+        assert isinstance(ii.dtype, pd.IntervalDtype)
+        obj = index_or_series(ii)
+
+        exp = index_or_series([ii[0], fill_val, ii[2], ii[3], ii[4]], dtype=object)
+
+        fill_dtype = object
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_series_int64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_index_int64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_series_bool(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_index_bool(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_series_timedelta64(self):
+        raise NotImplementedError
+
+    @pytest.mark.parametrize(
+        "fill_val",
+        [
+            1,
+            1.1,
+            1 + 1j,
+            True,
+            pd.Interval(1, 2, closed="left"),
+            pd.Timestamp("2012-01-01", tz="US/Eastern"),
+            pd.Timestamp("2012-01-01"),
+            pd.Timedelta(days=1),
+            pd.Period("2016-01-01", "W"),
+        ],
+    )
+    def test_fillna_series_period(self, index_or_series, fill_val):
+        pi = pd.period_range("2016-01-01", periods=4, freq="D").insert(1, pd.NaT)
+        assert isinstance(pi.dtype, pd.PeriodDtype)
+        obj = index_or_series(pi)
+
+        exp = index_or_series([pi[0], fill_val, pi[2], pi[3], pi[4]], dtype=object)
+
+        fill_dtype = object
+        self._assert_fillna_conversion(obj, fill_val, exp, fill_dtype)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_index_timedelta64(self):
+        raise NotImplementedError
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_fillna_index_period(self):
+        raise NotImplementedError
+
+
+class TestReplaceSeriesCoercion(CoercionBase):
+    klasses = ["series"]
+    method = "replace"
+
+    rep: dict[str, list] = {}
+    rep["object"] = ["a", "b"]
+    rep["int64"] = [4, 5]
+    rep["float64"] = [1.1, 2.2]
+    rep["complex128"] = [1 + 1j, 2 + 2j]
+    rep["bool"] = [True, False]
+    rep["datetime64[ns]"] = [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-01-03")]
+
+    for tz in ["UTC", "US/Eastern"]:
+        # to test tz => different tz replacement
+        key = f"datetime64[ns, {tz}]"
+        rep[key] = [
+            pd.Timestamp("2011-01-01", tz=tz),
+            pd.Timestamp("2011-01-03", tz=tz),
+        ]
+
+    rep["timedelta64[ns]"] = [pd.Timedelta("1 day"), pd.Timedelta("2 day")]
+
+    @pytest.fixture(params=["dict", "series"])
+    def how(self, request):
+        return request.param
+
+    @pytest.fixture(
+        params=[
+            "object",
+            "int64",
+            "float64",
+            "complex128",
+            "bool",
+            "datetime64[ns]",
+            "datetime64[ns, UTC]",
+            "datetime64[ns, US/Eastern]",
+            "timedelta64[ns]",
+        ]
+    )
+    def from_key(self, request):
+        return request.param
+
+    @pytest.fixture(
+        params=[
+            "object",
+            "int64",
+            "float64",
+            "complex128",
+            "bool",
+            "datetime64[ns]",
+            "datetime64[ns, UTC]",
+            "datetime64[ns, US/Eastern]",
+            "timedelta64[ns]",
+        ],
+        ids=[
+            "object",
+            "int64",
+            "float64",
+            "complex128",
+            "bool",
+            "datetime64",
+            "datetime64tz",
+            "datetime64tz",
+            "timedelta64",
+        ],
+    )
+    def to_key(self, request):
+        return request.param
+
+    @pytest.fixture
+    def replacer(self, how, from_key, to_key):
+        """
+        Object we will pass to `Series.replace`
+        """
+        if how == "dict":
+            replacer = dict(zip(self.rep[from_key], self.rep[to_key]))
+        elif how == "series":
+            replacer = pd.Series(self.rep[to_key], index=self.rep[from_key])
+        else:
+            raise ValueError
+        return replacer
+
+    def test_replace_series(self, how, to_key, from_key, replacer):
+        index = pd.Index([3, 4], name="xxx")
+        obj = pd.Series(self.rep[from_key], index=index, name="yyy")
+        obj = obj.astype(from_key)
+        assert obj.dtype == from_key
+
+        if from_key.startswith("datetime") and to_key.startswith("datetime"):
+            # tested below
+            return
+        elif from_key in ["datetime64[ns, US/Eastern]", "datetime64[ns, UTC]"]:
+            # tested below
+            return
+
+        if (from_key == "float64" and to_key in ("int64")) or (
+            from_key == "complex128" and to_key in ("int64", "float64")
+        ):
+            if not IS64 or is_platform_windows():
+                pytest.skip(f"32-bit platform buggy: {from_key} -> {to_key}")
+
+            # Expected: do not downcast by replacement
+            exp = pd.Series(self.rep[to_key], index=index, name="yyy", dtype=from_key)
+
+        else:
+            exp = pd.Series(self.rep[to_key], index=index, name="yyy")
+
+        result = obj.replace(replacer)
+        tm.assert_series_equal(result, exp, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "to_key",
+        ["timedelta64[ns]", "bool", "object", "complex128", "float64", "int64"],
+        indirect=True,
+    )
+    @pytest.mark.parametrize(
+        "from_key", ["datetime64[ns, UTC]", "datetime64[ns, US/Eastern]"], indirect=True
+    )
+    def test_replace_series_datetime_tz(
+        self, how, to_key, from_key, replacer, using_infer_string
+    ):
+        index = pd.Index([3, 4], name="xyz")
+        obj = pd.Series(self.rep[from_key], index=index, name="yyy").dt.as_unit("ns")
+        assert obj.dtype == from_key
+
+        exp = pd.Series(self.rep[to_key], index=index, name="yyy")
+        if using_infer_string and to_key == "object":
+            assert exp.dtype == "string"
+        else:
+            assert exp.dtype == to_key
+
+        result = obj.replace(replacer)
+        tm.assert_series_equal(result, exp, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "to_key",
+        ["datetime64[ns]", "datetime64[ns, UTC]", "datetime64[ns, US/Eastern]"],
+        indirect=True,
+    )
+    @pytest.mark.parametrize(
+        "from_key",
+        ["datetime64[ns]", "datetime64[ns, UTC]", "datetime64[ns, US/Eastern]"],
+        indirect=True,
+    )
+    def test_replace_series_datetime_datetime(self, how, to_key, from_key, replacer):
+        index = pd.Index([3, 4], name="xyz")
+        obj = pd.Series(self.rep[from_key], index=index, name="yyy").dt.as_unit("ns")
+        assert obj.dtype == from_key
+
+        exp = pd.Series(self.rep[to_key], index=index, name="yyy")
+        if isinstance(obj.dtype, pd.DatetimeTZDtype) and isinstance(
+            exp.dtype, pd.DatetimeTZDtype
+        ):
+            # with mismatched tzs, we retain the original dtype as of 2.0
+            exp = exp.astype(obj.dtype)
+        elif to_key == from_key:
+            exp = exp.dt.as_unit("ns")
+
+        result = obj.replace(replacer)
+        tm.assert_series_equal(result, exp, check_dtype=False)
+
+    @pytest.mark.xfail(reason="Test not implemented")
+    def test_replace_series_period(self):
+        raise NotImplementedError
diff --git a/pandas/pandas/tests/indexing/test_datetime.py b/pandas/pandas/tests/indexing/test_datetime.py
new file mode 100644
index 0000000000000..af7533399ea74
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_datetime.py
@@ -0,0 +1,191 @@
+import re
+
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeIndex:
+    def test_get_loc_naive_dti_aware_str_deprecated(self):
+        # GH#46903
+        ts = Timestamp("20130101")._value
+        dti = pd.DatetimeIndex([ts + 50 + i for i in range(100)])
+        ser = Series(range(100), index=dti)
+
+        key = "2013-01-01 00:00:00.000000050+0000"
+        msg = re.escape(repr(key))
+        with pytest.raises(KeyError, match=msg):
+            ser[key]
+
+        with pytest.raises(KeyError, match=msg):
+            dti.get_loc(key)
+
+    def test_indexing_with_datetime_tz(self):
+        # GH#8260
+        # support datetime64 with tz
+
+        idx = Index(date_range("20130101", periods=3, tz="US/Eastern"), name="foo")
+        dr = date_range("20130110", periods=3)
+        df = DataFrame({"A": idx, "B": dr})
+        df["C"] = idx
+        df.iloc[1, 1] = pd.NaT
+        df.iloc[1, 2] = pd.NaT
+
+        expected = Series(
+            [Timestamp("2013-01-02 00:00:00-0500", tz="US/Eastern"), pd.NaT, pd.NaT],
+            index=list("ABC"),
+            dtype="object",
+            name=1,
+        )
+
+        # indexing
+        result = df.iloc[1]
+        tm.assert_series_equal(result, expected)
+        result = df.loc[1]
+        tm.assert_series_equal(result, expected)
+
+    def test_indexing_fast_xs(self):
+        # indexing - fast_xs
+        df = DataFrame({"a": date_range("2014-01-01", periods=10, tz="UTC")})
+        result = df.iloc[5]
+        expected = Series(
+            [Timestamp("2014-01-06 00:00:00+0000", tz="UTC")],
+            index=["a"],
+            name=5,
+            dtype="M8[ns, UTC]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = df.loc[5]
+        tm.assert_series_equal(result, expected)
+
+        # indexing - boolean
+        result = df[df.a > df.a[3]]
+        expected = df.iloc[4:]
+        tm.assert_frame_equal(result, expected)
+
+    def test_consistency_with_tz_aware_scalar(self):
+        # xef gh-12938
+        # various ways of indexing the same tz-aware scalar
+        df = Series([Timestamp("2016-03-30 14:35:25", tz="Europe/Brussels")]).to_frame()
+
+        df = pd.concat([df, df]).reset_index(drop=True)
+        expected = Timestamp("2016-03-30 14:35:25+0200", tz="Europe/Brussels")
+
+        result = df[0][0]
+        assert result == expected
+
+        result = df.iloc[0, 0]
+        assert result == expected
+
+        result = df.loc[0, 0]
+        assert result == expected
+
+        result = df.iat[0, 0]
+        assert result == expected
+
+        result = df.at[0, 0]
+        assert result == expected
+
+        result = df[0].loc[0]
+        assert result == expected
+
+        result = df[0].at[0]
+        assert result == expected
+
+    def test_indexing_with_datetimeindex_tz(self, indexer_sl):
+        # GH 12050
+        # indexing on a series with a datetimeindex with tz
+        index = date_range("2015-01-01", periods=2, tz="utc")
+
+        ser = Series(range(2), index=index, dtype="int64")
+
+        # list-like indexing
+
+        for sel in (index, list(index)):
+            # getitem
+            result = indexer_sl(ser)[sel]
+            expected = ser.copy()
+            if sel is not index:
+                expected.index = expected.index._with_freq(None)
+            tm.assert_series_equal(result, expected)
+
+            # setitem
+            result = ser.copy()
+            indexer_sl(result)[sel] = 1
+            expected = Series(1, index=index)
+            tm.assert_series_equal(result, expected)
+
+        # single element indexing
+
+        # getitem
+        assert indexer_sl(ser)[index[1]] == 1
+
+        # setitem
+        result = ser.copy()
+        indexer_sl(result)[index[1]] = 5
+        expected = Series([0, 5], index=index)
+        tm.assert_series_equal(result, expected)
+
+    def test_nanosecond_getitem_setitem_with_tz(self):
+        # GH 11679
+        data = ["2016-06-28 08:30:00.123456789"]
+        index = pd.DatetimeIndex(data, dtype="datetime64[ns, America/Chicago]")
+        df = DataFrame({"a": [10]}, index=index)
+        result = df.loc[df.index[0]]
+        expected = Series(10, index=["a"], name=df.index[0])
+        tm.assert_series_equal(result, expected)
+
+        result = df.copy()
+        result.loc[df.index[0], "a"] = -1
+        expected = DataFrame(-1, index=index, columns=["a"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_getitem_str_slice_millisecond_resolution(self, frame_or_series):
+        # GH#33589
+
+        keys = [
+            "2017-10-25T16:25:04.151",
+            "2017-10-25T16:25:04.252",
+            "2017-10-25T16:50:05.237",
+            "2017-10-25T16:50:05.238",
+        ]
+        obj = frame_or_series(
+            [1, 2, 3, 4],
+            index=[Timestamp(x) for x in keys],
+        )
+        result = obj[keys[1] : keys[2]]
+        expected = frame_or_series(
+            [2, 3],
+            index=[
+                Timestamp(keys[1]),
+                Timestamp(keys[2]),
+            ],
+        )
+        tm.assert_equal(result, expected)
+
+    def test_getitem_pyarrow_index(self, frame_or_series):
+        # GH 53644
+        pytest.importorskip("pyarrow")
+        obj = frame_or_series(
+            range(5),
+            index=date_range("2020", freq="D", periods=5).astype(
+                "timestamp[us][pyarrow]"
+            ),
+        )
+        result = obj.loc[obj.index[:-3]]
+        expected = frame_or_series(
+            range(2),
+            index=date_range("2020", freq="D", periods=2).astype(
+                "timestamp[us][pyarrow]"
+            ),
+        )
+        tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/test_floats.py b/pandas/pandas/tests/indexing/test_floats.py
new file mode 100644
index 0000000000000..f9b9e8a6c7c28
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_floats.py
@@ -0,0 +1,674 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    RangeIndex,
+    Series,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+def gen_obj(klass, index):
+    if klass is Series:
+        obj = Series(np.arange(len(index)), index=index)
+    else:
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), len(index))),
+            index=index,
+            columns=index,
+        )
+    return obj
+
+
+class TestFloatIndexers:
+    def check(self, result, original, indexer, getitem):
+        """
+        comparator for results
+        we need to take care if we are indexing on a
+        Series or a frame
+        """
+        if isinstance(original, Series):
+            expected = original.iloc[indexer]
+        elif getitem:
+            expected = original.iloc[:, indexer]
+        else:
+            expected = original.iloc[indexer]
+
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcde")),
+            Index(list("abcde"), dtype="category"),
+            date_range("2020-01-01", periods=5),
+            timedelta_range("1 day", periods=5),
+            period_range("2020-01-01", periods=5),
+        ],
+    )
+    def test_scalar_non_numeric(self, index, frame_or_series, indexer_sl):
+        # GH 4892
+        # float_indexers should raise exceptions
+        # on appropriate Index types & accessors
+
+        s = gen_obj(frame_or_series, index)
+
+        # getting
+        with pytest.raises(KeyError, match="^3.0$"):
+            indexer_sl(s)[3.0]
+
+        # contains
+        assert 3.0 not in s
+
+        s2 = s.copy()
+        indexer_sl(s2)[3.0] = 10
+
+        if indexer_sl is tm.setitem:
+            assert 3.0 in s2.axes[-1]
+        elif indexer_sl is tm.loc:
+            assert 3.0 in s2.axes[0]
+        else:
+            assert 3.0 not in s2.axes[0]
+            assert 3.0 not in s2.axes[-1]
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcde")),
+            Index(list("abcde"), dtype="category"),
+            date_range("2020-01-01", periods=5),
+            timedelta_range("1 day", periods=5),
+            period_range("2020-01-01", periods=5),
+        ],
+    )
+    def test_scalar_non_numeric_series_fallback(self, index):
+        # starting in 3.0, integer keys are always treated as labels, no longer
+        #  fall back to positional.
+        s = Series(np.arange(len(index)), index=index)
+
+        with pytest.raises(KeyError, match="3"):
+            s[3]
+        with pytest.raises(KeyError, match="^3.0$"):
+            s[3.0]
+
+    def test_scalar_with_mixed(self, indexer_sl):
+        s2 = Series([1, 2, 3], index=["a", "b", "c"])
+        s3 = Series([1, 2, 3], index=["a", "b", 1.5])
+
+        # lookup in a pure string index with an invalid indexer
+
+        with pytest.raises(KeyError, match="^1.0$"):
+            indexer_sl(s2)[1.0]
+
+        with pytest.raises(KeyError, match=r"^1\.0$"):
+            indexer_sl(s2)[1.0]
+
+        result = indexer_sl(s2)["b"]
+        expected = 2
+        assert result == expected
+
+        # mixed index so we have label
+        # indexing
+        with pytest.raises(KeyError, match="^1.0$"):
+            indexer_sl(s3)[1.0]
+
+        if indexer_sl is not tm.loc:
+            # as of 3.0, __getitem__ no longer falls back to positional
+            with pytest.raises(KeyError, match="^1$"):
+                s3[1]
+
+        with pytest.raises(KeyError, match=r"^1\.0$"):
+            indexer_sl(s3)[1.0]
+
+        result = indexer_sl(s3)[1.5]
+        expected = 3
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "index", [Index(np.arange(5), dtype=np.int64), RangeIndex(5)]
+    )
+    def test_scalar_integer(self, index, frame_or_series, indexer_sl):
+        getitem = indexer_sl is not tm.loc
+
+        # test how scalar float indexers work on int indexes
+
+        # integer index
+        i = index
+        obj = gen_obj(frame_or_series, i)
+
+        # coerce to equal int
+
+        result = indexer_sl(obj)[3.0]
+        self.check(result, obj, 3, getitem)
+
+        if isinstance(obj, Series):
+
+            def compare(x, y):
+                assert x == y
+
+            expected = 100
+        else:
+            compare = tm.assert_series_equal
+            if getitem:
+                expected = Series(100, index=range(len(obj)), name=3)
+            else:
+                expected = Series(100.0, index=range(len(obj)), name=3)
+
+        s2 = obj.copy()
+        indexer_sl(s2)[3.0] = 100
+
+        result = indexer_sl(s2)[3.0]
+        compare(result, expected)
+
+        result = indexer_sl(s2)[3]
+        compare(result, expected)
+
+    @pytest.mark.parametrize(
+        "index", [Index(np.arange(5), dtype=np.int64), RangeIndex(5)]
+    )
+    def test_scalar_integer_contains_float(self, index, frame_or_series):
+        # contains
+        # integer index
+        obj = gen_obj(frame_or_series, index)
+
+        # coerce to equal int
+        assert 3.0 in obj
+
+    def test_scalar_float(self, frame_or_series):
+        # scalar float indexers work on a float index
+        index = Index(np.arange(5.0))
+        s = gen_obj(frame_or_series, index)
+
+        # assert all operations except for iloc are ok
+        indexer = index[3]
+        for idxr in [tm.loc, tm.setitem]:
+            getitem = idxr is not tm.loc
+
+            # getting
+            result = idxr(s)[indexer]
+            self.check(result, s, 3, getitem)
+
+            # setting
+            s2 = s.copy()
+
+            result = idxr(s2)[indexer]
+            self.check(result, s, 3, getitem)
+
+            # random float is a KeyError
+            with pytest.raises(KeyError, match=r"^3\.5$"):
+                idxr(s)[3.5]
+
+        # contains
+        assert 3.0 in s
+
+        # iloc succeeds with an integer
+        expected = s.iloc[3]
+        s2 = s.copy()
+
+        s2.iloc[3] = expected
+        result = s2.iloc[3]
+        self.check(result, s, 3, False)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcde"), dtype=object),
+            date_range("2020-01-01", periods=5),
+            timedelta_range("1 day", periods=5),
+            period_range("2020-01-01", periods=5),
+        ],
+    )
+    @pytest.mark.parametrize("idx", [slice(3.0, 4), slice(3, 4.0), slice(3.0, 4.0)])
+    def test_slice_non_numeric(self, index, idx, frame_or_series, indexer_sli):
+        # GH 4892
+        # float_indexers should raise exceptions
+        # on appropriate Index types & accessors
+
+        s = gen_obj(frame_or_series, index)
+
+        # getitem
+        if indexer_sli is tm.iloc:
+            msg = (
+                "cannot do positional indexing "
+                rf"on {type(index).__name__} with these indexers \[(3|4)\.0\] of "
+                "type float"
+            )
+        else:
+            msg = (
+                "cannot do slice indexing "
+                rf"on {type(index).__name__} with these indexers "
+                r"\[(3|4)(\.0)?\] "
+                r"of type (float|int)"
+            )
+        with pytest.raises(TypeError, match=msg):
+            indexer_sli(s)[idx]
+
+        # setitem
+        if indexer_sli is tm.iloc:
+            # otherwise we keep the same message as above
+            msg = "slice indices must be integers or None or have an __index__ method"
+        with pytest.raises(TypeError, match=msg):
+            indexer_sli(s)[idx] = 0
+
+    def test_slice_integer(self):
+        # same as above, but for Integer based indexes
+        # these coerce to a like integer
+        # oob indicates if we are out of bounds
+        # of positional indexing
+        for index, oob in [
+            (Index(np.arange(5, dtype=np.int64)), False),
+            (RangeIndex(5), False),
+            (Index(np.arange(5, dtype=np.int64) + 10), True),
+        ]:
+            # s is an in-range index
+            s = Series(range(5), index=index)
+
+            # getitem
+            for idx in [slice(3.0, 4), slice(3, 4.0), slice(3.0, 4.0)]:
+                result = s.loc[idx]
+
+                # these are all label indexing
+                # except getitem which is positional
+                # empty
+                if oob:
+                    indexer = slice(0, 0)
+                else:
+                    indexer = slice(3, 5)
+                self.check(result, s, indexer, False)
+
+            # getitem out-of-bounds
+            for idx in [slice(-6, 6), slice(-6.0, 6.0)]:
+                result = s.loc[idx]
+
+                # these are all label indexing
+                # except getitem which is positional
+                # empty
+                if oob:
+                    indexer = slice(0, 0)
+                else:
+                    indexer = slice(-6, 6)
+                self.check(result, s, indexer, False)
+
+            # positional indexing
+            msg = (
+                "cannot do slice indexing "
+                rf"on {type(index).__name__} with these indexers \[-6\.0\] of "
+                "type float"
+            )
+            with pytest.raises(TypeError, match=msg):
+                s[slice(-6.0, 6.0)]
+
+            # getitem odd floats
+            for idx, res1 in [
+                (slice(2.5, 4), slice(3, 5)),
+                (slice(2, 3.5), slice(2, 4)),
+                (slice(2.5, 3.5), slice(3, 4)),
+            ]:
+                result = s.loc[idx]
+                if oob:
+                    res = slice(0, 0)
+                else:
+                    res = res1
+
+                self.check(result, s, res, False)
+
+                # positional indexing
+                msg = (
+                    "cannot do slice indexing "
+                    rf"on {type(index).__name__} with these indexers \[(2|3)\.5\] of "
+                    "type float"
+                )
+                with pytest.raises(TypeError, match=msg):
+                    s[idx]
+
+    @pytest.mark.parametrize("idx", [slice(2, 4.0), slice(2.0, 4), slice(2.0, 4.0)])
+    def test_integer_positional_indexing(self, idx):
+        """make sure that we are raising on positional indexing
+        w.r.t. an integer index
+        """
+        s = Series(range(2, 6), index=range(2, 6))
+
+        result = s[2:4]
+        expected = s.iloc[2:4]
+        tm.assert_series_equal(result, expected)
+
+        klass = RangeIndex
+        msg = (
+            "cannot do (slice|positional) indexing "
+            rf"on {klass.__name__} with these indexers \[(2|4)\.0\] of "
+            "type float"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s[idx]
+        with pytest.raises(TypeError, match=msg):
+            s.iloc[idx]
+
+    @pytest.mark.parametrize(
+        "index", [Index(np.arange(5), dtype=np.int64), RangeIndex(5)]
+    )
+    def test_slice_integer_frame_getitem(self, index):
+        # similar to above, but on the getitem dim (of a DataFrame)
+        s = DataFrame(np.random.default_rng(2).standard_normal((5, 2)), index=index)
+
+        # getitem
+        for idx in [slice(0.0, 1), slice(0, 1.0), slice(0.0, 1.0)]:
+            result = s.loc[idx]
+            indexer = slice(0, 2)
+            self.check(result, s, indexer, False)
+
+            # positional indexing
+            msg = (
+                "cannot do slice indexing "
+                rf"on {type(index).__name__} with these indexers \[(0|1)\.0\] of "
+                "type float"
+            )
+            with pytest.raises(TypeError, match=msg):
+                s[idx]
+
+        # getitem out-of-bounds
+        for idx in [slice(-10, 10), slice(-10.0, 10.0)]:
+            result = s.loc[idx]
+            self.check(result, s, slice(-10, 10), True)
+
+        # positional indexing
+        msg = (
+            "cannot do slice indexing "
+            rf"on {type(index).__name__} with these indexers \[-10\.0\] of "
+            "type float"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s[slice(-10.0, 10.0)]
+
+        # getitem odd floats
+        for idx, res in [
+            (slice(0.5, 1), slice(1, 2)),
+            (slice(0, 0.5), slice(0, 1)),
+            (slice(0.5, 1.5), slice(1, 2)),
+        ]:
+            result = s.loc[idx]
+            self.check(result, s, res, False)
+
+            # positional indexing
+            msg = (
+                "cannot do slice indexing "
+                rf"on {type(index).__name__} with these indexers \[0\.5\] of "
+                "type float"
+            )
+            with pytest.raises(TypeError, match=msg):
+                s[idx]
+
+    @pytest.mark.parametrize("idx", [slice(3.0, 4), slice(3, 4.0), slice(3.0, 4.0)])
+    @pytest.mark.parametrize(
+        "index", [Index(np.arange(5), dtype=np.int64), RangeIndex(5)]
+    )
+    def test_float_slice_getitem_with_integer_index_raises(self, idx, index):
+        # similar to above, but on the getitem dim (of a DataFrame)
+        s = DataFrame(np.random.default_rng(2).standard_normal((5, 2)), index=index)
+
+        # setitem
+        sc = s.copy()
+        sc.loc[idx] = 0
+        result = sc.loc[idx].values.ravel()
+        assert (result == 0).all()
+
+        # positional indexing
+        msg = (
+            "cannot do slice indexing "
+            rf"on {type(index).__name__} with these indexers \[(3|4)\.0\] of "
+            "type float"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s[idx] = 0
+
+        with pytest.raises(TypeError, match=msg):
+            s[idx]
+
+    @pytest.mark.parametrize("idx", [slice(3.0, 4), slice(3, 4.0), slice(3.0, 4.0)])
+    def test_slice_float(self, idx, frame_or_series, indexer_sl):
+        # same as above, but for floats
+        index = Index(np.arange(5.0)) + 0.1
+        s = gen_obj(frame_or_series, index)
+
+        expected = s.iloc[3:4]
+
+        # getitem
+        result = indexer_sl(s)[idx]
+        assert isinstance(result, type(s))
+        tm.assert_equal(result, expected)
+
+        # setitem
+        s2 = s.copy()
+        indexer_sl(s2)[idx] = 0
+        result = indexer_sl(s2)[idx].values.ravel()
+        assert (result == 0).all()
+
+    def test_floating_index_doc_example(self):
+        index = Index([1.5, 2, 3, 4.5, 5])
+        s = Series(range(5), index=index)
+        assert s[3] == 2
+        assert s.loc[3] == 2
+        assert s.iloc[3] == 3
+
+    def test_floating_misc(self, indexer_sl):
+        # related 236
+        # scalar/slicing of a float index
+        s = Series(np.arange(5), index=np.arange(5) * 2.5, dtype=np.int64)
+
+        # label based slicing
+        result = indexer_sl(s)[1.0:3.0]
+        expected = Series(1, index=[2.5])
+        tm.assert_series_equal(result, expected)
+
+        # exact indexing when found
+
+        result = indexer_sl(s)[5.0]
+        assert result == 2
+
+        result = indexer_sl(s)[5]
+        assert result == 2
+
+        # value not found (and no fallbacking at all)
+
+        # scalar integers
+        with pytest.raises(KeyError, match=r"^4$"):
+            indexer_sl(s)[4]
+
+        # fancy floats/integers create the correct entry (as nan)
+        # fancy tests
+        expected = Series([2, 0], index=Index([5.0, 0.0], dtype=np.float64))
+        for fancy_idx in [[5.0, 0.0], np.array([5.0, 0.0])]:  # float
+            tm.assert_series_equal(indexer_sl(s)[fancy_idx], expected)
+
+        expected = Series([2, 0], index=Index([5, 0], dtype="float64"))
+        for fancy_idx in [[5, 0], np.array([5, 0])]:
+            tm.assert_series_equal(indexer_sl(s)[fancy_idx], expected)
+
+        # all should return the same as we are slicing 'the same'
+        result2 = indexer_sl(s)[2.0:5.0]
+        result3 = indexer_sl(s)[2.0:5]
+        result4 = indexer_sl(s)[2.1:5]
+        tm.assert_series_equal(result2, result3)
+        tm.assert_series_equal(result2, result4)
+
+        # list selection
+        result1 = indexer_sl(s)[[0.0, 5, 10]]
+        result2 = s.iloc[[0, 2, 4]]
+        tm.assert_series_equal(result1, result2)
+
+        with pytest.raises(KeyError, match="not in index"):
+            indexer_sl(s)[[1.6, 5, 10]]
+
+        with pytest.raises(KeyError, match="not in index"):
+            indexer_sl(s)[[0, 1, 2]]
+
+        result = indexer_sl(s)[[2.5, 5]]
+        tm.assert_series_equal(result, Series([1, 2], index=[2.5, 5.0]))
+
+        result = indexer_sl(s)[[2.5]]
+        tm.assert_series_equal(result, Series([1], index=[2.5]))
+
+    def test_floatindex_slicing_bug(self, float_numpy_dtype):
+        # GH 5557, related to slicing a float index
+        dtype = float_numpy_dtype
+        ser = {
+            256: 2321.0,
+            1: 78.0,
+            2: 2716.0,
+            3: 0.0,
+            4: 369.0,
+            5: 0.0,
+            6: 269.0,
+            7: 0.0,
+            8: 0.0,
+            9: 0.0,
+            10: 3536.0,
+            11: 0.0,
+            12: 24.0,
+            13: 0.0,
+            14: 931.0,
+            15: 0.0,
+            16: 101.0,
+            17: 78.0,
+            18: 9643.0,
+            19: 0.0,
+            20: 0.0,
+            21: 0.0,
+            22: 63761.0,
+            23: 0.0,
+            24: 446.0,
+            25: 0.0,
+            26: 34773.0,
+            27: 0.0,
+            28: 729.0,
+            29: 78.0,
+            30: 0.0,
+            31: 0.0,
+            32: 3374.0,
+            33: 0.0,
+            34: 1391.0,
+            35: 0.0,
+            36: 361.0,
+            37: 0.0,
+            38: 61808.0,
+            39: 0.0,
+            40: 0.0,
+            41: 0.0,
+            42: 6677.0,
+            43: 0.0,
+            44: 802.0,
+            45: 0.0,
+            46: 2691.0,
+            47: 0.0,
+            48: 3582.0,
+            49: 0.0,
+            50: 734.0,
+            51: 0.0,
+            52: 627.0,
+            53: 70.0,
+            54: 2584.0,
+            55: 0.0,
+            56: 324.0,
+            57: 0.0,
+            58: 605.0,
+            59: 0.0,
+            60: 0.0,
+            61: 0.0,
+            62: 3989.0,
+            63: 10.0,
+            64: 42.0,
+            65: 0.0,
+            66: 904.0,
+            67: 0.0,
+            68: 88.0,
+            69: 70.0,
+            70: 8172.0,
+            71: 0.0,
+            72: 0.0,
+            73: 0.0,
+            74: 64902.0,
+            75: 0.0,
+            76: 347.0,
+            77: 0.0,
+            78: 36605.0,
+            79: 0.0,
+            80: 379.0,
+            81: 70.0,
+            82: 0.0,
+            83: 0.0,
+            84: 3001.0,
+            85: 0.0,
+            86: 1630.0,
+            87: 7.0,
+            88: 364.0,
+            89: 0.0,
+            90: 67404.0,
+            91: 9.0,
+            92: 0.0,
+            93: 0.0,
+            94: 7685.0,
+            95: 0.0,
+            96: 1017.0,
+            97: 0.0,
+            98: 2831.0,
+            99: 0.0,
+            100: 2963.0,
+            101: 0.0,
+            102: 854.0,
+            103: 0.0,
+            104: 0.0,
+            105: 0.0,
+            106: 0.0,
+            107: 0.0,
+            108: 0.0,
+            109: 0.0,
+            110: 0.0,
+            111: 0.0,
+            112: 0.0,
+            113: 0.0,
+            114: 0.0,
+            115: 0.0,
+            116: 0.0,
+            117: 0.0,
+            118: 0.0,
+            119: 0.0,
+            120: 0.0,
+            121: 0.0,
+            122: 0.0,
+            123: 0.0,
+            124: 0.0,
+            125: 0.0,
+            126: 67744.0,
+            127: 22.0,
+            128: 264.0,
+            129: 0.0,
+            260: 197.0,
+            268: 0.0,
+            265: 0.0,
+            269: 0.0,
+            261: 0.0,
+            266: 1198.0,
+            267: 0.0,
+            262: 2629.0,
+            258: 775.0,
+            257: 0.0,
+            263: 0.0,
+            259: 0.0,
+            264: 163.0,
+            250: 10326.0,
+            251: 0.0,
+            252: 1228.0,
+            253: 0.0,
+            254: 2769.0,
+            255: 0.0,
+        }
+
+        # smoke test for the repr
+        s = Series(ser, dtype=dtype)
+        result = s.value_counts()
+        assert result.index.dtype == dtype
+        str(result)
diff --git a/pandas/pandas/tests/indexing/test_iat.py b/pandas/pandas/tests/indexing/test_iat.py
new file mode 100644
index 0000000000000..bb9252a50eb2a
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_iat.py
@@ -0,0 +1,30 @@
+import numpy as np
+
+from pandas import (
+    DataFrame,
+    Series,
+    period_range,
+)
+
+
+def test_iat(float_frame):
+    for i, row in enumerate(float_frame.index):
+        for j, col in enumerate(float_frame.columns):
+            result = float_frame.iat[i, j]
+            expected = float_frame.at[row, col]
+            assert result == expected
+
+
+def test_iat_duplicate_columns():
+    # https://github.com/pandas-dev/pandas/issues/11754
+    df = DataFrame([[1, 2]], columns=["x", "x"])
+    assert df.iat[0, 0] == 1
+
+
+def test_iat_getitem_series_with_period_index():
+    # GH#4390, iat incorrectly indexing
+    index = period_range("1/1/2001", periods=10)
+    ser = Series(np.random.default_rng(2).standard_normal(10), index=index)
+    expected = ser[index[0]]
+    result = ser.iat[0]
+    assert expected == result
diff --git a/pandas/pandas/tests/indexing/test_iloc.py b/pandas/pandas/tests/indexing/test_iloc.py
new file mode 100644
index 0000000000000..2f6998a85c80b
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_iloc.py
@@ -0,0 +1,1450 @@
+"""test positional based indexing with iloc"""
+
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import IndexingError
+
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    DataFrame,
+    Index,
+    Interval,
+    NaT,
+    Series,
+    Timestamp,
+    array,
+    concat,
+    date_range,
+    interval_range,
+    isna,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.api.types import is_scalar
+from pandas.tests.indexing.common import check_indexing_smoketest_or_raises
+
+# We pass through the error message from numpy
+_slice_iloc_msg = re.escape(
+    "only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) "
+    "and integer or boolean arrays are valid indices"
+)
+
+
+class TestiLoc:
+    @pytest.mark.parametrize("key", [2, -1, [0, 1, 2]])
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcd"), dtype=object),
+            Index([2, 4, "null", 8], dtype=object),
+            date_range("20130101", periods=4),
+            Index(range(0, 8, 2), dtype=np.float64),
+            Index([]),
+        ],
+    )
+    def test_iloc_getitem_int_and_list_int(self, key, frame_or_series, index, request):
+        obj = frame_or_series(range(len(index)), index=index)
+        check_indexing_smoketest_or_raises(
+            obj,
+            "iloc",
+            key,
+            fails=IndexError,
+        )
+
+        # array of ints (GH5006), make sure that a single indexer is returning
+        # the correct type
+
+
+class TestiLocBaseIndependent:
+    """Tests Independent Of Base Class"""
+
+    @pytest.mark.parametrize(
+        "key",
+        [
+            slice(None),
+            slice(3),
+            range(3),
+            [0, 1, 2],
+            Index(range(3)),
+            np.asarray([0, 1, 2]),
+        ],
+    )
+    def test_iloc_setitem_fullcol_categorical(self, indexer_li, key):
+        frame = DataFrame({0: range(3)}, dtype=object)
+
+        cat = Categorical(["alpha", "beta", "gamma"])
+
+        assert frame._mgr.blocks[0]._can_hold_element(cat)
+
+        df = frame.copy()
+        orig_vals = df.values
+
+        indexer_li(df)[key, 0] = cat
+
+        expected = DataFrame({0: cat}).astype(object)
+        assert np.shares_memory(df[0].values, orig_vals)
+
+        tm.assert_frame_equal(df, expected)
+
+        # check we dont have a view on cat (may be undesired GH#39986)
+        df.iloc[0, 0] = "gamma"
+        assert cat[0] != "gamma"
+
+        # pre-2.0 with mixed dataframe ("split" path) we always overwrote the
+        #  column.  as of 2.0 we correctly write "into" the column, so
+        #  we retain the object dtype.
+        frame = DataFrame({0: np.array([0, 1, 2], dtype=object), 1: range(3)})
+        df = frame.copy()
+        indexer_li(df)[key, 0] = cat
+        expected = DataFrame({0: Series(cat.astype(object), dtype=object), 1: range(3)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_ea_inplace(self, frame_or_series, index_or_series_or_array):
+        # GH#38952 Case with not setting a full column
+        #  IntegerArray without NAs
+        arr = array([1, 2, 3, 4])
+        obj = frame_or_series(arr.to_numpy("i8"))
+
+        if frame_or_series is Series:
+            values = obj.values
+        else:
+            values = obj._mgr.blocks[0].values
+
+        if frame_or_series is Series:
+            obj.iloc[:2] = index_or_series_or_array(arr[2:])
+        else:
+            obj.iloc[:2, 0] = index_or_series_or_array(arr[2:])
+
+        expected = frame_or_series(np.array([3, 4, 3, 4], dtype="i8"))
+        tm.assert_equal(obj, expected)
+
+        # Check that we are actually in-place
+        if frame_or_series is Series:
+            assert obj.values is not values
+            assert np.shares_memory(obj.values, values)
+        else:
+            assert np.shares_memory(obj[0].values, values)
+
+    def test_is_scalar_access(self):
+        # GH#32085 index with duplicates doesn't matter for _is_scalar_access
+        index = Index([1, 2, 1])
+        ser = Series(range(3), index=index)
+
+        assert ser.iloc._is_scalar_access((1,))
+
+        df = ser.to_frame()
+        assert df.iloc._is_scalar_access((1, 0))
+
+    def test_iloc_exceeds_bounds(self):
+        # GH6296
+        # iloc should allow indexers that exceed the bounds
+        df = DataFrame(np.random.default_rng(2).random((20, 5)), columns=list("ABCDE"))
+
+        # lists of positions should raise IndexError!
+        msg = "positional indexers are out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[:, [0, 1, 2, 3, 4, 5]]
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[[1, 30]]
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[[1, -30]]
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[[100]]
+
+        s = df["A"]
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[[100]]
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[[-100]]
+
+        # still raise on a single indexer
+        msg = "single positional indexer is out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[30]
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[-30]
+
+        # GH10779
+        # single positive/negative indexer exceeding Series bounds should raise
+        # an IndexError
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[30]
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[-30]
+
+        # slices are ok
+        result = df.iloc[:, 4:10]  # 0 < start < len < stop
+        expected = df.iloc[:, 4:]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, -4:-10]  # stop < 0 < start < len
+        expected = df.iloc[:, :0]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, 10:4:-1]  # 0 < stop < len < start (down)
+        expected = df.iloc[:, :4:-1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, 4:-10:-1]  # stop < 0 < start < len (down)
+        expected = df.iloc[:, 4::-1]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, -10:4]  # start < 0 < stop < len
+        expected = df.iloc[:, :4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, 10:4]  # 0 < stop < len < start
+        expected = df.iloc[:, :0]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, -10:-11:-1]  # stop < start < 0 < len (down)
+        expected = df.iloc[:, :0]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, 10:11]  # 0 < len < start < stop
+        expected = df.iloc[:, :0]
+        tm.assert_frame_equal(result, expected)
+
+        # slice bounds exceeding is ok
+        result = s.iloc[18:30]
+        expected = s.iloc[18:]
+        tm.assert_series_equal(result, expected)
+
+        result = s.iloc[30:]
+        expected = s.iloc[:0]
+        tm.assert_series_equal(result, expected)
+
+        result = s.iloc[30::-1]
+        expected = s.iloc[::-1]
+        tm.assert_series_equal(result, expected)
+
+        # doc example
+        dfl = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("AB")
+        )
+        tm.assert_frame_equal(
+            dfl.iloc[:, 2:3],
+            DataFrame(index=dfl.index, columns=Index([], dtype=dfl.columns.dtype)),
+        )
+        tm.assert_frame_equal(dfl.iloc[:, 1:3], dfl.iloc[:, [1]])
+        tm.assert_frame_equal(dfl.iloc[4:6], dfl.iloc[[4]])
+
+        msg = "positional indexers are out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            dfl.iloc[[4, 5, 6]]
+        msg = "single positional indexer is out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            dfl.iloc[:, 4]
+
+    @pytest.mark.parametrize("index,columns", [(np.arange(20), list("ABCDE"))])
+    @pytest.mark.parametrize(
+        "index_vals,column_vals",
+        [
+            ([slice(None), ["A", "D"]]),
+            (["1", "2"], slice(None)),
+            ([datetime(2019, 1, 1)], slice(None)),
+        ],
+    )
+    def test_iloc_non_integer_raises(self, index, columns, index_vals, column_vals):
+        # GH 25753
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), len(columns))),
+            index=index,
+            columns=columns,
+        )
+        msg = ".iloc requires numeric indexers, got"
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[index_vals, column_vals]
+
+    def test_iloc_getitem_invalid_scalar(self, frame_or_series):
+        # GH 21982
+
+        obj = DataFrame(np.arange(100).reshape(10, 10))
+        obj = tm.get_obj(obj, frame_or_series)
+
+        with pytest.raises(TypeError, match="Cannot index by location index"):
+            obj.iloc["a"]
+
+    def test_iloc_array_not_mutating_negative_indices(self):
+        # GH 21867
+        array_with_neg_numbers = np.array([1, 2, -1])
+        array_copy = array_with_neg_numbers.copy()
+        df = DataFrame(
+            {"A": [100, 101, 102], "B": [103, 104, 105], "C": [106, 107, 108]},
+            index=[1, 2, 3],
+        )
+        df.iloc[array_with_neg_numbers]
+        tm.assert_numpy_array_equal(array_with_neg_numbers, array_copy)
+        df.iloc[:, array_with_neg_numbers]
+        tm.assert_numpy_array_equal(array_with_neg_numbers, array_copy)
+
+    def test_iloc_getitem_neg_int_can_reach_first_index(self):
+        # GH10547 and GH10779
+        # negative integers should be able to reach index 0
+        df = DataFrame({"A": [2, 3, 5], "B": [7, 11, 13]})
+        s = df["A"]
+
+        expected = df.iloc[0]
+        result = df.iloc[-3]
+        tm.assert_series_equal(result, expected)
+
+        expected = df.iloc[[0]]
+        result = df.iloc[[-3]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = s.iloc[0]
+        result = s.iloc[-3]
+        assert result == expected
+
+        expected = s.iloc[[0]]
+        result = s.iloc[[-3]]
+        tm.assert_series_equal(result, expected)
+
+        # check the length 1 Series case highlighted in GH10547
+        expected = Series(["a"], index=["A"])
+        result = expected.iloc[[-1]]
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_getitem_dups(self):
+        # GH 6766
+        df1 = DataFrame([{"A": None, "B": 1}, {"A": 2, "B": 2}])
+        df2 = DataFrame([{"A": 3, "B": 3}, {"A": 4, "B": 4}])
+        df = concat([df1, df2], axis=1)
+
+        # cross-sectional indexing
+        result = df.iloc[0, 0]
+        assert isna(result)
+
+        result = df.iloc[0, :]
+        expected = Series([np.nan, 1, 3, 3], index=["A", "B", "A", "B"], name=0)
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_getitem_array(self):
+        df = DataFrame(
+            [
+                {"A": 1, "B": 2, "C": 3},
+                {"A": 100, "B": 200, "C": 300},
+                {"A": 1000, "B": 2000, "C": 3000},
+            ]
+        )
+
+        expected = DataFrame([{"A": 1, "B": 2, "C": 3}])
+        tm.assert_frame_equal(df.iloc[[0]], expected)
+
+        expected = DataFrame([{"A": 1, "B": 2, "C": 3}, {"A": 100, "B": 200, "C": 300}])
+        tm.assert_frame_equal(df.iloc[[0, 1]], expected)
+
+        expected = DataFrame([{"B": 2, "C": 3}, {"B": 2000, "C": 3000}], index=[0, 2])
+        result = df.iloc[[0, 2], [1, 2]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_getitem_bool(self):
+        df = DataFrame(
+            [
+                {"A": 1, "B": 2, "C": 3},
+                {"A": 100, "B": 200, "C": 300},
+                {"A": 1000, "B": 2000, "C": 3000},
+            ]
+        )
+
+        expected = DataFrame([{"A": 1, "B": 2, "C": 3}, {"A": 100, "B": 200, "C": 300}])
+        result = df.iloc[[True, True, False]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            [{"A": 1, "B": 2, "C": 3}, {"A": 1000, "B": 2000, "C": 3000}], index=[0, 2]
+        )
+        result = df.iloc[lambda x: x.index % 2 == 0]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("index", [[True, False], [True, False, True, False]])
+    def test_iloc_getitem_bool_diff_len(self, index):
+        # GH26658
+        s = Series([1, 2, 3])
+        msg = f"Boolean index has wrong length: {len(index)} instead of {len(s)}"
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[index]
+
+    def test_iloc_getitem_slice(self):
+        df = DataFrame(
+            [
+                {"A": 1, "B": 2, "C": 3},
+                {"A": 100, "B": 200, "C": 300},
+                {"A": 1000, "B": 2000, "C": 3000},
+            ]
+        )
+
+        expected = DataFrame([{"A": 1, "B": 2, "C": 3}, {"A": 100, "B": 200, "C": 300}])
+        result = df.iloc[:2]
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame([{"A": 100, "B": 200}], index=[1])
+        result = df.iloc[1:2, 0:2]
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            [{"A": 1, "C": 3}, {"A": 100, "C": 300}, {"A": 1000, "C": 3000}]
+        )
+        result = df.iloc[:, lambda df: [0, 2]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_getitem_slice_dups(self):
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=["A", "A", "B", "B"],
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).integers(0, 10, size=20).reshape(10, 2),
+            columns=["A", "C"],
+        )
+
+        # axis=1
+        df = concat([df1, df2], axis=1)
+        tm.assert_frame_equal(df.iloc[:, :4], df1)
+        tm.assert_frame_equal(df.iloc[:, 4:], df2)
+
+        df = concat([df2, df1], axis=1)
+        tm.assert_frame_equal(df.iloc[:, :2], df2)
+        tm.assert_frame_equal(df.iloc[:, 2:], df1)
+
+        exp = concat([df2, df1.iloc[:, [0]]], axis=1)
+        tm.assert_frame_equal(df.iloc[:, 0:3], exp)
+
+        # axis=0
+        df = concat([df, df], axis=0)
+        tm.assert_frame_equal(df.iloc[0:10, :2], df2)
+        tm.assert_frame_equal(df.iloc[0:10, 2:], df1)
+        tm.assert_frame_equal(df.iloc[10:, :2], df2)
+        tm.assert_frame_equal(df.iloc[10:, 2:], df1)
+
+    def test_iloc_setitem(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=np.arange(0, 8, 2),
+            columns=np.arange(0, 12, 3),
+        )
+
+        df.iloc[1, 1] = 1
+        result = df.iloc[1, 1]
+        assert result == 1
+
+        df.iloc[:, 2:3] = 0
+        expected = df.iloc[:, 2:3]
+        result = df.iloc[:, 2:3]
+        tm.assert_frame_equal(result, expected)
+
+        # GH5771
+        s = Series(0, index=[4, 5, 6])
+        s.iloc[1:2] += 1
+        expected = Series([0, 1, 0], index=[4, 5, 6])
+        tm.assert_series_equal(s, expected)
+
+    def test_iloc_setitem_axis_argument(self):
+        # GH45032
+        df = DataFrame([[6, "c", 10], [7, "d", 11], [8, "e", 12]])
+        df[1] = df[1].astype(object)
+        expected = DataFrame([[6, "c", 10], [7, "d", 11], [5, 5, 5]])
+        expected[1] = expected[1].astype(object)
+        df.iloc(axis=0)[2] = 5
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame([[6, "c", 10], [7, "d", 11], [8, "e", 12]])
+        df[1] = df[1].astype(object)
+        expected = DataFrame([[6, "c", 5], [7, "d", 5], [8, "e", 5]])
+        expected[1] = expected[1].astype(object)
+        df.iloc(axis=1)[2] = 5
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_list(self):
+        # setitem with an iloc list
+        df = DataFrame(
+            np.arange(9).reshape((3, 3)), index=["A", "B", "C"], columns=["A", "B", "C"]
+        )
+        df.iloc[[0, 1], [1, 2]]
+        df.iloc[[0, 1], [1, 2]] += 100
+
+        expected = DataFrame(
+            np.array([0, 101, 102, 3, 104, 105, 6, 7, 8]).reshape((3, 3)),
+            index=["A", "B", "C"],
+            columns=["A", "B", "C"],
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_pandas_object(self):
+        # GH 17193
+        s_orig = Series([0, 1, 2, 3])
+        expected = Series([0, -1, -2, 3])
+
+        s = s_orig.copy()
+        s.iloc[Series([1, 2])] = [-1, -2]
+        tm.assert_series_equal(s, expected)
+
+        s = s_orig.copy()
+        s.iloc[Index([1, 2])] = [-1, -2]
+        tm.assert_series_equal(s, expected)
+
+    def test_iloc_setitem_dups(self):
+        # GH 6766
+        # iloc with a mask aligning from another iloc
+        df1 = DataFrame([{"A": None, "B": 1}, {"A": 2, "B": 2}])
+        df2 = DataFrame([{"A": 3, "B": 3}, {"A": 4, "B": 4}])
+        df = concat([df1, df2], axis=1)
+
+        expected = df.fillna(3)
+        inds = np.isnan(df.iloc[:, 0])
+        mask = inds[inds].index
+        df.iloc[mask, 0] = df.iloc[mask, 2]
+        tm.assert_frame_equal(df, expected)
+
+        # del a dup column across blocks
+        expected = DataFrame({0: [1, 2], 1: [3, 4]})
+        expected.columns = ["B", "B"]
+        del df["A"]
+        tm.assert_frame_equal(df, expected)
+
+        # assign back to self
+        df.iloc[[0, 1], [0, 1]] = df.iloc[[0, 1], [0, 1]]
+        tm.assert_frame_equal(df, expected)
+
+        # reversed x 2
+        df.iloc[[1, 0], [0, 1]] = df.iloc[[1, 0], [0, 1]].reset_index(drop=True)
+        df.iloc[[1, 0], [0, 1]] = df.iloc[[1, 0], [0, 1]].reset_index(drop=True)
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_frame_duplicate_columns_multiple_blocks(self):
+        # Same as the "assign back to self" check in test_iloc_setitem_dups
+        #  but on a DataFrame with multiple blocks
+        df = DataFrame([[0, 1], [2, 3]], columns=["B", "B"])
+
+        # setting float values that can be held by existing integer arrays
+        #  is inplace
+        df.iloc[:, 0] = df.iloc[:, 0].astype("f8")
+        assert len(df._mgr.blocks) == 1
+
+        # if the assigned values cannot be held by existing integer arrays,
+        #  we raise
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.iloc[:, 0] = df.iloc[:, 0] + 0.5
+
+        expected = df.copy()
+
+        # assign back to self
+        df.iloc[[0, 1], [0, 1]] = df.iloc[[0, 1], [0, 1]]
+
+        tm.assert_frame_equal(df, expected)
+
+    # TODO: GH#27620 this test used to compare iloc against ix; check if this
+    #  is redundant with another test comparing iloc against loc
+    def test_iloc_getitem_frame(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            index=range(0, 20, 2),
+            columns=range(0, 8, 2),
+        )
+
+        result = df.iloc[2]
+        exp = df.loc[4]
+        tm.assert_series_equal(result, exp)
+
+        result = df.iloc[2, 2]
+        exp = df.loc[4, 4]
+        assert result == exp
+
+        # slice
+        result = df.iloc[4:8]
+        expected = df.loc[8:14]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[:, 2:3]
+        expected = df.loc[:, 4:5]
+        tm.assert_frame_equal(result, expected)
+
+        # list of integers
+        result = df.iloc[[0, 1, 3]]
+        expected = df.loc[[0, 2, 6]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.iloc[[0, 1, 3], [0, 1]]
+        expected = df.loc[[0, 2, 6], [0, 2]]
+        tm.assert_frame_equal(result, expected)
+
+        # neg indices
+        result = df.iloc[[-1, 1, 3], [-1, 1]]
+        expected = df.loc[[18, 2, 6], [6, 2]]
+        tm.assert_frame_equal(result, expected)
+
+        # dups indices
+        result = df.iloc[[-1, -1, 1, 3], [-1, 1]]
+        expected = df.loc[[18, 18, 2, 6], [6, 2]]
+        tm.assert_frame_equal(result, expected)
+
+        # with index-like
+        s = Series(index=range(1, 5), dtype=object)
+        result = df.iloc[s.index]
+        expected = df.loc[[2, 4, 6, 8]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_getitem_labelled_frame(self):
+        # try with labelled frame
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            index=list("abcdefghij"),
+            columns=list("ABCD"),
+        )
+
+        result = df.iloc[1, 1]
+        exp = df.loc["b", "B"]
+        assert result == exp
+
+        result = df.iloc[:, 2:3]
+        expected = df.loc[:, ["C"]]
+        tm.assert_frame_equal(result, expected)
+
+        # negative indexing
+        result = df.iloc[-1, -1]
+        exp = df.loc["j", "D"]
+        assert result == exp
+
+        # out-of-bounds exception
+        msg = "index 5 is out of bounds for axis 0 with size 4|index out of bounds"
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[10, 5]
+
+        # trying to use a label
+        msg = (
+            r"Location based indexing can only have \[integer, integer "
+            r"slice \(START point is INCLUDED, END point is EXCLUDED\), "
+            r"listlike of integers, boolean array\] types"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.iloc["j", "D"]
+
+    def test_iloc_getitem_doc_issue(self):
+        # multi axis slicing issue with single block
+        # surfaced in GH 6059
+
+        arr = np.random.default_rng(2).standard_normal((6, 4))
+        index = date_range("20130101", periods=6)
+        columns = list("ABCD")
+        df = DataFrame(arr, index=index, columns=columns)
+
+        # defines ref_locs
+        df.describe()
+
+        result = df.iloc[3:5, 0:2]
+
+        expected = DataFrame(arr[3:5, 0:2], index=index[3:5], columns=columns[0:2])
+        tm.assert_frame_equal(result, expected)
+
+        # for dups
+        df.columns = list("aaaa")
+        result = df.iloc[3:5, 0:2]
+
+        expected = DataFrame(arr[3:5, 0:2], index=index[3:5], columns=list("aa"))
+        tm.assert_frame_equal(result, expected)
+
+        # related
+        arr = np.random.default_rng(2).standard_normal((6, 4))
+        index = list(range(0, 12, 2))
+        columns = list(range(0, 8, 2))
+        df = DataFrame(arr, index=index, columns=columns)
+
+        df._mgr.blocks[0].mgr_locs
+        result = df.iloc[1:5, 2:4]
+        expected = DataFrame(arr[1:5, 2:4], index=index[1:5], columns=columns[2:4])
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_setitem_series(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            index=list("abcdefghij"),
+            columns=list("ABCD"),
+        )
+
+        df.iloc[1, 1] = 1
+        result = df.iloc[1, 1]
+        assert result == 1
+
+        df.iloc[:, 2:3] = 0
+        expected = df.iloc[:, 2:3]
+        result = df.iloc[:, 2:3]
+        tm.assert_frame_equal(result, expected)
+
+        s = Series(np.random.default_rng(2).standard_normal(10), index=range(0, 20, 2))
+
+        s.iloc[1] = 1
+        result = s.iloc[1]
+        assert result == 1
+
+        s.iloc[:4] = 0
+        expected = s.iloc[:4]
+        result = s.iloc[:4]
+        tm.assert_series_equal(result, expected)
+
+        s = Series([-1] * 6)
+        s.iloc[0::2] = [0, 2, 4]
+        s.iloc[1::2] = [1, 3, 5]
+        result = s
+        expected = Series([0, 1, 2, 3, 4, 5])
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_setitem_list_of_lists(self):
+        # GH 7551
+        # list-of-list is set incorrectly in mixed vs. single dtyped frames
+        df = DataFrame(
+            {"A": np.arange(5, dtype="int64"), "B": np.arange(5, 10, dtype="int64")}
+        )
+        df.iloc[2:4] = [[10, 11], [12, 13]]
+        expected = DataFrame({"A": [0, 1, 10, 12, 4], "B": [5, 6, 11, 13, 9]})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(
+            {"A": ["a", "b", "c", "d", "e"], "B": np.arange(5, 10, dtype="int64")}
+        )
+        df.iloc[2:4] = [["x", 11], ["y", 13]]
+        expected = DataFrame({"A": ["a", "b", "x", "y", "e"], "B": [5, 6, 11, 13, 9]})
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("indexer", [[0], slice(None, 1, None), np.array([0])])
+    @pytest.mark.parametrize("value", [["Z"], np.array(["Z"])])
+    def test_iloc_setitem_with_scalar_index(self, indexer, value):
+        # GH #19474
+        # assigning like "df.iloc[0, [0]] = ['Z']" should be evaluated
+        # elementwisely, not using "setter('A', ['Z'])".
+
+        # Set object type to avoid upcast when setting "Z"
+        df = DataFrame([[1, 2], [3, 4]], columns=["A", "B"]).astype({"A": object})
+        df.iloc[0, indexer] = value
+        result = df.iloc[0, 0]
+
+        assert is_scalar(result) and result == "Z"
+
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    def test_iloc_mask(self):
+        # GH 3631, iloc with a mask (of a series) should raise
+        df = DataFrame(list(range(5)), index=list("ABCDE"), columns=["a"])
+        mask = df.a % 2 == 0
+        msg = "iLocation based boolean indexing cannot use an indexable as a mask"
+        with pytest.raises(ValueError, match=msg):
+            df.iloc[mask]
+        mask.index = range(len(mask))
+        msg = "iLocation based boolean indexing on an integer type is not available"
+        with pytest.raises(NotImplementedError, match=msg):
+            df.iloc[mask]
+
+        # ndarray ok
+        result = df.iloc[np.array([True] * len(mask), dtype=bool)]
+        tm.assert_frame_equal(result, df)
+
+        # the possibilities
+        locs = np.arange(4)
+        nums = 2**locs
+        reps = [bin(num) for num in nums]
+        df = DataFrame({"locs": locs, "nums": nums}, reps)
+
+        expected = {
+            (None, ""): "0b1100",
+            (None, ".loc"): "0b1100",
+            (None, ".iloc"): "0b1100",
+            ("index", ""): "0b11",
+            ("index", ".loc"): "0b11",
+            ("index", ".iloc"): (
+                "iLocation based boolean indexing cannot use an indexable as a mask"
+            ),
+            ("locs", ""): "Unalignable boolean Series provided as indexer "
+            "(index of the boolean Series and of the indexed "
+            "object do not match).",
+            ("locs", ".loc"): "Unalignable boolean Series provided as indexer "
+            "(index of the boolean Series and of the "
+            "indexed object do not match).",
+            ("locs", ".iloc"): (
+                "iLocation based boolean indexing on an integer type is not available"
+            ),
+        }
+
+        # UserWarnings from reindex of a boolean mask
+        for idx in [None, "index", "locs"]:
+            mask = (df.nums > 2).values
+            if idx:
+                mask_index = getattr(df, idx)[::-1]
+                mask = Series(mask, list(mask_index))
+            for method in ["", ".loc", ".iloc"]:
+                try:
+                    if method:
+                        accessor = getattr(df, method[1:])
+                    else:
+                        accessor = df
+                    answer = str(bin(accessor[mask]["nums"].sum()))
+                except (ValueError, IndexingError, NotImplementedError) as err:
+                    answer = str(err)
+
+                key = (
+                    idx,
+                    method,
+                )
+                r = expected.get(key)
+                if r != answer:
+                    raise AssertionError(
+                        f"[{key}] does not match [{answer}], received [{r}]"
+                    )
+
+    def test_iloc_non_unique_indexing(self):
+        # GH 4017, non-unique indexing (on the axis)
+        df = DataFrame({"A": [0.1] * 3000, "B": [1] * 3000})
+        idx = np.arange(30) * 99
+        expected = df.iloc[idx]
+
+        df3 = concat([df, 2 * df, 3 * df])
+        result = df3.iloc[idx]
+
+        tm.assert_frame_equal(result, expected)
+
+        df2 = DataFrame({"A": [0.1] * 1000, "B": [1] * 1000})
+        df2 = concat([df2, 2 * df2, 3 * df2])
+
+        with pytest.raises(KeyError, match="not in index"):
+            df2.loc[idx]
+
+    def test_iloc_empty_list_indexer_is_ok(self):
+        df = DataFrame(
+            np.ones((5, 2)),
+            index=Index([f"i-{i}" for i in range(5)], name="a"),
+            columns=Index([f"i-{i}" for i in range(2)], name="a"),
+        )
+        # vertical empty
+        tm.assert_frame_equal(
+            df.iloc[:, []],
+            df.iloc[:, :0],
+            check_index_type=True,
+            check_column_type=True,
+        )
+        # horizontal empty
+        tm.assert_frame_equal(
+            df.iloc[[], :],
+            df.iloc[:0, :],
+            check_index_type=True,
+            check_column_type=True,
+        )
+        # horizontal empty
+        tm.assert_frame_equal(
+            df.iloc[[]], df.iloc[:0, :], check_index_type=True, check_column_type=True
+        )
+
+    def test_identity_slice_returns_new_object(self):
+        # GH13873
+        original_df = DataFrame({"a": [1, 2, 3]})
+        sliced_df = original_df.iloc[:]
+        assert sliced_df is not original_df
+
+        # should be a shallow copy
+        assert np.shares_memory(original_df["a"], sliced_df["a"])
+
+        # Setting using .loc[:, "a"] sets inplace so alters both sliced and orig
+        # depending on CoW
+        original_df.loc[:, "a"] = [4, 4, 4]
+        assert (sliced_df["a"] == [1, 2, 3]).all()
+
+        original_series = Series([1, 2, 3, 4, 5, 6])
+        sliced_series = original_series.iloc[:]
+        assert sliced_series is not original_series
+
+        # should also be a shallow copy
+        original_series[:3] = [7, 8, 9]
+        # shallow copy not updated (CoW)
+        assert all(sliced_series[:3] == [1, 2, 3])
+
+    def test_indexing_zerodim_np_array(self):
+        # GH24919
+        df = DataFrame([[1, 2], [3, 4]])
+        result = df.iloc[np.array(0)]
+        s = Series([1, 2], name=0)
+        tm.assert_series_equal(result, s)
+
+    def test_series_indexing_zerodim_np_array(self):
+        # GH24919
+        s = Series([1, 2])
+        result = s.iloc[np.array(0)]
+        assert result == 1
+
+    def test_iloc_setitem_categorical_updates_inplace(self):
+        # Mixed dtype ensures we go through take_split_path in setitem_with_indexer
+        cat = Categorical(["A", "B", "C"])
+        df = DataFrame({1: cat, 2: [1, 2, 3]}, copy=False)
+
+        assert tm.shares_memory(df[1], cat)
+
+        # With the enforcement of GH#45333 in 2.0, this modifies original
+        #  values inplace
+        df.iloc[:, 0] = cat[::-1]
+
+        assert tm.shares_memory(df[1], cat)
+        expected = Categorical(["C", "B", "A"], categories=["A", "B", "C"])
+        tm.assert_categorical_equal(cat, expected)
+
+    def test_iloc_with_boolean_operation(self):
+        # GH 20627
+        result = DataFrame([[0, 1], [2, 3], [4, 5], [6, np.nan]])
+        result.iloc[result.index <= 2] *= 2
+        expected = DataFrame([[0, 2], [4, 6], [8, 10], [6, np.nan]])
+        tm.assert_frame_equal(result, expected)
+
+        result.iloc[result.index > 2] *= 2
+        expected = DataFrame([[0, 2], [4, 6], [8, 10], [12, np.nan]])
+        tm.assert_frame_equal(result, expected)
+
+        result.iloc[[True, True, False, False]] *= 2
+        expected = DataFrame([[0, 4], [8, 12], [8, 10], [12, np.nan]])
+        tm.assert_frame_equal(result, expected)
+
+        result.iloc[[False, False, True, True]] /= 2
+        expected = DataFrame([[0, 4.0], [8, 12.0], [4, 5.0], [6, np.nan]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_getitem_singlerow_slice_categoricaldtype_gives_series(self):
+        # GH#29521
+        df = DataFrame({"x": Categorical("a b c d e".split())})
+        result = df.iloc[0]
+        raw_cat = Categorical(["a"], categories=["a", "b", "c", "d", "e"])
+        expected = Series(raw_cat, index=["x"], name=0, dtype="category")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_getitem_categorical_values(self):
+        # GH#14580
+        # test iloc() on Series with Categorical data
+
+        ser = Series([1, 2, 3]).astype("category")
+
+        # get slice
+        result = ser.iloc[0:2]
+        expected = Series([1, 2]).astype(CategoricalDtype([1, 2, 3]))
+        tm.assert_series_equal(result, expected)
+
+        # get list of indexes
+        result = ser.iloc[[0, 1]]
+        expected = Series([1, 2]).astype(CategoricalDtype([1, 2, 3]))
+        tm.assert_series_equal(result, expected)
+
+        # get boolean array
+        result = ser.iloc[[True, False, False]]
+        expected = Series([1]).astype(CategoricalDtype([1, 2, 3]))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("value", [None, NaT, np.nan])
+    def test_iloc_setitem_td64_values_cast_na(self, value):
+        # GH#18586
+        series = Series([0, 1, 2], dtype="timedelta64[ns]")
+        series.iloc[0] = value
+        expected = Series([NaT, 1, 2], dtype="timedelta64[ns]")
+        tm.assert_series_equal(series, expected)
+
+    @pytest.mark.parametrize("not_na", [Interval(0, 1), "a", 1.0])
+    def test_setitem_mix_of_nan_and_interval(self, not_na, nulls_fixture):
+        # GH#27937
+        dtype = CategoricalDtype(categories=[not_na])
+        ser = Series(
+            [nulls_fixture, nulls_fixture, nulls_fixture, nulls_fixture], dtype=dtype
+        )
+        ser.iloc[:3] = [nulls_fixture, not_na, nulls_fixture]
+        exp = Series([nulls_fixture, not_na, nulls_fixture, nulls_fixture], dtype=dtype)
+        tm.assert_series_equal(ser, exp)
+
+    def test_iloc_setitem_empty_frame_raises_with_3d_ndarray(self):
+        idx = Index([])
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), len(idx))),
+            index=idx,
+            columns=idx,
+        )
+        nd3 = np.random.default_rng(2).integers(5, size=(2, 2, 2))
+
+        msg = f"Cannot set values with ndim > {obj.ndim}"
+        with pytest.raises(ValueError, match=msg):
+            obj.iloc[nd3] = 0
+
+    def test_iloc_getitem_read_only_values(self, indexer_li):
+        # GH#10043 this is fundamentally a test for iloc, but test loc while
+        #  we're here
+        rw_array = np.eye(10)
+        rw_df = DataFrame(rw_array)
+
+        ro_array = np.eye(10)
+        ro_array.setflags(write=False)
+        ro_df = DataFrame(ro_array)
+
+        tm.assert_frame_equal(
+            indexer_li(rw_df)[[1, 2, 3]], indexer_li(ro_df)[[1, 2, 3]]
+        )
+        tm.assert_frame_equal(indexer_li(rw_df)[[1]], indexer_li(ro_df)[[1]])
+        tm.assert_series_equal(indexer_li(rw_df)[1], indexer_li(ro_df)[1])
+        tm.assert_frame_equal(indexer_li(rw_df)[1:3], indexer_li(ro_df)[1:3])
+
+    def test_iloc_getitem_readonly_key(self):
+        # GH#17192 iloc with read-only array raising TypeError
+        df = DataFrame({"data": np.ones(100, dtype="float64")})
+        indices = np.array([1, 3, 6])
+        indices.flags.writeable = False
+
+        result = df.iloc[indices]
+        expected = df.loc[[1, 3, 6]]
+        tm.assert_frame_equal(result, expected)
+
+        result = df["data"].iloc[indices]
+        expected = df["data"].loc[[1, 3, 6]]
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_assign_series_to_df_cell(self):
+        # GH 37593
+        df = DataFrame(columns=["a"], index=[0])
+        df.iloc[0, 0] = Series([1, 2, 3])
+        expected = DataFrame({"a": [Series([1, 2, 3])]}, columns=["a"], index=[0])
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("klass", [list, np.array])
+    def test_iloc_setitem_bool_indexer(self, klass):
+        # GH#36741
+        df = DataFrame({"flag": ["x", "y", "z"], "value": [1, 3, 4]})
+        indexer = klass([True, False, False])
+        df.iloc[indexer, 1] = df.iloc[indexer, 1] * 2
+        expected = DataFrame({"flag": ["x", "y", "z"], "value": [2, 3, 4]})
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("indexer", [[1], slice(1, 2)])
+    def test_iloc_setitem_pure_position_based(self, indexer):
+        # GH#22046
+        df1 = DataFrame({"a2": [11, 12, 13], "b2": [14, 15, 16]})
+        df2 = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "c": [7, 8, 9]})
+        df2.iloc[:, indexer] = df1.iloc[:, [0]]
+        expected = DataFrame({"a": [1, 2, 3], "b": [11, 12, 13], "c": [7, 8, 9]})
+        tm.assert_frame_equal(df2, expected)
+
+    def test_iloc_setitem_dictionary_value(self):
+        # GH#37728
+        df = DataFrame({"x": [1, 2], "y": [2, 2]})
+        rhs = {"x": 9, "y": 99}
+        df.iloc[1] = rhs
+        expected = DataFrame({"x": [1, 9], "y": [2, 99]})
+        tm.assert_frame_equal(df, expected)
+
+        # GH#38335 same thing, mixed dtypes
+        df = DataFrame({"x": [1, 2], "y": [2.0, 2.0]})
+        df.iloc[1] = rhs
+        expected = DataFrame({"x": [1, 9], "y": [2.0, 99.0]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_getitem_float_duplicates(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            index=[0.1, 0.2, 0.2],
+            columns=list("abc"),
+        )
+        expect = df.iloc[1:]
+        tm.assert_frame_equal(df.loc[0.2], expect)
+
+        expect = df.iloc[1:, 0]
+        tm.assert_series_equal(df.loc[0.2, "a"], expect)
+
+        df.index = [1, 0.2, 0.2]
+        expect = df.iloc[1:]
+        tm.assert_frame_equal(df.loc[0.2], expect)
+
+        expect = df.iloc[1:, 0]
+        tm.assert_series_equal(df.loc[0.2, "a"], expect)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 3)),
+            index=[1, 0.2, 0.2, 1],
+            columns=list("abc"),
+        )
+        expect = df.iloc[1:-1]
+        tm.assert_frame_equal(df.loc[0.2], expect)
+
+        expect = df.iloc[1:-1, 0]
+        tm.assert_series_equal(df.loc[0.2, "a"], expect)
+
+        df.index = [0.1, 0.2, 2, 0.2]
+        expect = df.iloc[[1, -1]]
+        tm.assert_frame_equal(df.loc[0.2], expect)
+
+        expect = df.iloc[[1, -1], 0]
+        tm.assert_series_equal(df.loc[0.2, "a"], expect)
+
+    def test_iloc_setitem_custom_object(self):
+        # iloc with an object
+        class TO:
+            def __init__(self, value) -> None:
+                self.value = value
+
+            def __str__(self) -> str:
+                return f"[{self.value}]"
+
+            __repr__ = __str__
+
+            def __eq__(self, other) -> bool:
+                return self.value == other.value
+
+            def view(self):
+                return self
+
+        df = DataFrame(index=[0, 1], columns=[0])
+        df.iloc[1, 0] = TO(1)
+        df.iloc[1, 0] = TO(2)
+
+        result = DataFrame(index=[0, 1], columns=[0])
+        result.iloc[1, 0] = TO(2)
+
+        tm.assert_frame_equal(result, df)
+
+        # remains object dtype even after setting it back
+        df = DataFrame(index=[0, 1], columns=[0])
+        df.iloc[1, 0] = TO(1)
+        df.iloc[1, 0] = np.nan
+        result = DataFrame(index=[0, 1], columns=[0])
+
+        tm.assert_frame_equal(result, df)
+
+    def test_iloc_getitem_with_duplicates(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((3, 3)),
+            columns=list("ABC"),
+            index=list("aab"),
+        )
+
+        result = df.iloc[0]
+        assert isinstance(result, Series)
+        tm.assert_almost_equal(result.values, df.values[0])
+
+        result = df.T.iloc[:, 0]
+        assert isinstance(result, Series)
+        tm.assert_almost_equal(result.values, df.values[0])
+
+    def test_iloc_getitem_with_duplicates2(self):
+        # GH#2259
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=[1, 1, 2])
+        result = df.iloc[:, [0]]
+        expected = df.take([0], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_iloc_interval(self):
+        # GH#17130
+        df = DataFrame({Interval(1, 2): [1, 2]})
+
+        result = df.iloc[0]
+        expected = Series({Interval(1, 2): 1}, name=0)
+        tm.assert_series_equal(result, expected)
+
+        result = df.iloc[:, 0]
+        expected = Series([1, 2], name=Interval(1, 2))
+        tm.assert_series_equal(result, expected)
+
+        result = df.copy()
+        result.iloc[:, 0] += 1
+        expected = DataFrame({Interval(1, 2): [2, 3]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("indexing_func", [list, np.array])
+    @pytest.mark.parametrize("rhs_func", [list, np.array])
+    def test_loc_setitem_boolean_list(self, rhs_func, indexing_func):
+        # GH#20438 testing specifically list key, not arraylike
+        ser = Series([0, 1, 2])
+        ser.iloc[indexing_func([True, False, True])] = rhs_func([5, 10])
+        expected = Series([5, 1, 10])
+        tm.assert_series_equal(ser, expected)
+
+        df = DataFrame({"a": [0, 1, 2]})
+        df.iloc[indexing_func([True, False, True])] = rhs_func([[5], [10]])
+        expected = DataFrame({"a": [5, 1, 10]})
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_getitem_slice_negative_step_ea_block(self):
+        # GH#44551
+        df = DataFrame({"A": [1, 2, 3]}, dtype="Int64")
+
+        res = df.iloc[:, ::-1]
+        tm.assert_frame_equal(res, df)
+
+        df["B"] = "foo"
+        res = df.iloc[:, ::-1]
+        expected = DataFrame({"B": df["B"], "A": df["A"]})
+        tm.assert_frame_equal(res, expected)
+
+    def test_iloc_setitem_2d_ndarray_into_ea_block(self):
+        # GH#44703
+        df = DataFrame({"status": ["a", "b", "c"]}, dtype="category")
+        df.iloc[np.array([0, 1]), np.array([0])] = np.array([["a"], ["a"]])
+
+        expected = DataFrame({"status": ["a", "a", "c"]}, dtype=df["status"].dtype)
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_getitem_int_single_ea_block_view(self):
+        # GH#45241
+        # TODO: make an extension interface test for this?
+        arr = interval_range(1, 10.0)._values
+        df = DataFrame(arr)
+
+        # ser should be a *view* on the DataFrame data
+        ser = df.iloc[2]
+
+        # if we have a view, then changing arr[2] should also change ser[0]
+        assert arr[2] != arr[-1]  # otherwise the rest isn't meaningful
+        arr[2] = arr[-1]
+        assert ser[0] == arr[-1]
+
+    def test_iloc_setitem_multicolumn_to_datetime(self, using_infer_string):
+        # GH#20511
+        df = DataFrame({"A": ["2022-01-01", "2022-01-02"], "B": ["2021", "2022"]})
+
+        if using_infer_string:
+            with pytest.raises(TypeError, match="Invalid value"):
+                df.iloc[:, [0]] = DataFrame({"A": to_datetime(["2021", "2022"])})
+        else:
+            df.iloc[:, [0]] = DataFrame({"A": to_datetime(["2021", "2022"])})
+            expected = DataFrame(
+                {
+                    "A": [
+                        Timestamp("2021-01-01 00:00:00"),
+                        Timestamp("2022-01-01 00:00:00"),
+                    ],
+                    "B": ["2021", "2022"],
+                }
+            )
+            tm.assert_frame_equal(df, expected, check_dtype=False)
+
+
+class TestILocErrors:
+    # NB: this test should work for _any_ Series we can pass as
+    #  series_with_simple_index
+    def test_iloc_float_raises(self, series_with_simple_index, frame_or_series):
+        # GH#4892
+        # float_indexers should raise exceptions
+        # on appropriate Index types & accessors
+        # this duplicates the code below
+        # but is specifically testing for the error
+        # message
+
+        obj = series_with_simple_index
+        if frame_or_series is DataFrame:
+            obj = obj.to_frame()
+
+        msg = "Cannot index by location index with a non-integer key"
+        with pytest.raises(TypeError, match=msg):
+            obj.iloc[3.0]
+
+        with pytest.raises(IndexError, match=_slice_iloc_msg):
+            obj.iloc[3.0] = 0
+
+    def test_iloc_getitem_setitem_fancy_exceptions(self, float_frame):
+        with pytest.raises(IndexingError, match="Too many indexers"):
+            float_frame.iloc[:, :, :]
+
+        with pytest.raises(IndexError, match="too many indices for array"):
+            # GH#32257 we let numpy do validation, get their exception
+            float_frame.iloc[:, :, :] = 1
+
+    def test_iloc_frame_indexer(self):
+        # GH#39004
+        df = DataFrame({"a": [1, 2, 3]})
+        indexer = DataFrame({"a": [True, False, True]})
+        msg = "DataFrame indexer for .iloc is not supported. Consider using .loc"
+        with pytest.raises(TypeError, match=msg):
+            df.iloc[indexer] = 1
+
+        msg = (
+            "DataFrame indexer is not allowed for .iloc\n"
+            "Consider using .loc for automatic alignment."
+        )
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[indexer]
+
+
+class TestILocSetItemDuplicateColumns:
+    def test_iloc_setitem_scalar_duplicate_columns(self):
+        # GH#15686, duplicate columns and mixed dtype
+        df1 = DataFrame([{"A": None, "B": 1}, {"A": 2, "B": 2}])
+        df2 = DataFrame([{"A": 3, "B": 3}, {"A": 4, "B": 4}])
+        df = concat([df1, df2], axis=1)
+        df.iloc[0, 0] = -1
+
+        assert df.iloc[0, 0] == -1
+        assert df.iloc[0, 2] == 3
+        assert df.dtypes.iloc[2] == np.int64
+
+    def test_iloc_setitem_list_duplicate_columns(self):
+        # GH#22036 setting with same-sized list
+        df = DataFrame([[0, "str", "str2"]], columns=["a", "b", "b"])
+
+        df.iloc[:, 2] = ["str3"]
+
+        expected = DataFrame([[0, "str", "str3"]], columns=["a", "b", "b"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_setitem_series_duplicate_columns(self):
+        df = DataFrame(
+            np.arange(8, dtype=np.int64).reshape(2, 4), columns=["A", "B", "A", "B"]
+        )
+        df.iloc[:, 0] = df.iloc[:, 0].astype(np.float64)
+        assert df.dtypes.iloc[2] == np.int64
+
+    @pytest.mark.parametrize(
+        ["dtypes", "init_value", "expected_value"],
+        [("int64", "0", 0), ("float", "1.2", 1.2)],
+    )
+    def test_iloc_setitem_dtypes_duplicate_columns(
+        self, dtypes, init_value, expected_value
+    ):
+        # GH#22035
+        df = DataFrame(
+            [[init_value, "str", "str2"]], columns=["a", "b", "b"], dtype=object
+        )
+
+        # with the enforcement of GH#45333 in 2.0, this sets values inplace,
+        #  so we retain object dtype
+        df.iloc[:, 0] = df.iloc[:, 0].astype(dtypes)
+
+        expected_df = DataFrame(
+            [[expected_value, "str", "str2"]],
+            columns=["a", "b", "b"],
+            dtype=object,
+        )
+        tm.assert_frame_equal(df, expected_df)
+
+
+class TestILocCallable:
+    def test_frame_iloc_getitem_callable(self):
+        # GH#11485
+        df = DataFrame({"X": [1, 2, 3, 4], "Y": list("aabb")}, index=list("ABCD"))
+
+        # return location
+        res = df.iloc[lambda x: [1, 3]]
+        tm.assert_frame_equal(res, df.iloc[[1, 3]])
+
+        res = df.iloc[lambda x: [1, 3], :]
+        tm.assert_frame_equal(res, df.iloc[[1, 3], :])
+
+        res = df.iloc[lambda x: [1, 3], lambda x: 0]
+        tm.assert_series_equal(res, df.iloc[[1, 3], 0])
+
+        res = df.iloc[lambda x: [1, 3], lambda x: [0]]
+        tm.assert_frame_equal(res, df.iloc[[1, 3], [0]])
+
+        # mixture
+        res = df.iloc[[1, 3], lambda x: 0]
+        tm.assert_series_equal(res, df.iloc[[1, 3], 0])
+
+        res = df.iloc[[1, 3], lambda x: [0]]
+        tm.assert_frame_equal(res, df.iloc[[1, 3], [0]])
+
+        res = df.iloc[lambda x: [1, 3], 0]
+        tm.assert_series_equal(res, df.iloc[[1, 3], 0])
+
+        res = df.iloc[lambda x: [1, 3], [0]]
+        tm.assert_frame_equal(res, df.iloc[[1, 3], [0]])
+
+    def test_frame_iloc_setitem_callable(self):
+        # GH#11485
+        df = DataFrame(
+            {"X": [1, 2, 3, 4], "Y": Series(list("aabb"), dtype=object)},
+            index=list("ABCD"),
+        )
+
+        # return location
+        res = df.copy()
+        res.iloc[lambda x: [1, 3]] = 0
+        exp = df.copy()
+        exp.iloc[[1, 3]] = 0
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[lambda x: [1, 3], :] = -1
+        exp = df.copy()
+        exp.iloc[[1, 3], :] = -1
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[lambda x: [1, 3], lambda x: 0] = 5
+        exp = df.copy()
+        exp.iloc[[1, 3], 0] = 5
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[lambda x: [1, 3], lambda x: [0]] = 25
+        exp = df.copy()
+        exp.iloc[[1, 3], [0]] = 25
+        tm.assert_frame_equal(res, exp)
+
+        # mixture
+        res = df.copy()
+        res.iloc[[1, 3], lambda x: 0] = -3
+        exp = df.copy()
+        exp.iloc[[1, 3], 0] = -3
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[[1, 3], lambda x: [0]] = -5
+        exp = df.copy()
+        exp.iloc[[1, 3], [0]] = -5
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[lambda x: [1, 3], 0] = 10
+        exp = df.copy()
+        exp.iloc[[1, 3], 0] = 10
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.iloc[lambda x: [1, 3], [0]] = [-5, -5]
+        exp = df.copy()
+        exp.iloc[[1, 3], [0]] = [-5, -5]
+        tm.assert_frame_equal(res, exp)
+
+
+class TestILocSeries:
+    def test_iloc(self):
+        ser = Series(
+            np.random.default_rng(2).standard_normal(10), index=list(range(0, 20, 2))
+        )
+        ser_original = ser.copy()
+
+        for i in range(len(ser)):
+            result = ser.iloc[i]
+            exp = ser[ser.index[i]]
+            tm.assert_almost_equal(result, exp)
+
+        # pass a slice
+        result = ser.iloc[slice(1, 3)]
+        expected = ser.loc[2:4]
+        tm.assert_series_equal(result, expected)
+
+        # test slice is a view
+        with tm.assert_produces_warning(None):
+            # GH#45324 make sure we aren't giving a spurious FutureWarning
+            result[:] = 0
+        tm.assert_series_equal(ser, ser_original)
+
+        # list of integers
+        result = ser.iloc[[0, 2, 3, 4, 5]]
+        expected = ser.reindex(ser.index[[0, 2, 3, 4, 5]])
+        tm.assert_series_equal(result, expected)
+
+    def test_iloc_getitem_nonunique(self):
+        ser = Series([0, 1, 2], index=[0, 1, 0])
+        assert ser.iloc[2] == 2
+
+    def test_iloc_setitem_pure_position_based(self):
+        # GH#22046
+        ser1 = Series([1, 2, 3])
+        ser2 = Series([4, 5, 6], index=[1, 0, 2])
+        ser1.iloc[1:3] = ser2.iloc[1:3]
+        expected = Series([1, 5, 6])
+        tm.assert_series_equal(ser1, expected)
+
+    def test_iloc_nullable_int64_size_1_nan(self):
+        # GH 31861
+        result = DataFrame({"a": ["test"], "b": [np.nan]})
+        with pytest.raises(TypeError, match="Invalid value"):
+            result.loc[:, "b"] = result.loc[:, "b"].astype("Int64")
diff --git a/pandas/pandas/tests/indexing/test_indexers.py b/pandas/pandas/tests/indexing/test_indexers.py
new file mode 100644
index 0000000000000..ddc5c039160d5
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_indexers.py
@@ -0,0 +1,61 @@
+# Tests aimed at pandas.core.indexers
+import numpy as np
+import pytest
+
+from pandas.core.indexers import (
+    is_scalar_indexer,
+    length_of_indexer,
+    validate_indices,
+)
+
+
+def test_length_of_indexer():
+    arr = np.zeros(4, dtype=bool)
+    arr[0] = 1
+    result = length_of_indexer(arr)
+    assert result == 1
+
+
+def test_is_scalar_indexer():
+    indexer = (0, 1)
+    assert is_scalar_indexer(indexer, 2)
+    assert not is_scalar_indexer(indexer[0], 2)
+
+    indexer = (np.array([2]), 1)
+    assert not is_scalar_indexer(indexer, 2)
+
+    indexer = (np.array([2]), np.array([3]))
+    assert not is_scalar_indexer(indexer, 2)
+
+    indexer = (np.array([2]), np.array([3, 4]))
+    assert not is_scalar_indexer(indexer, 2)
+
+    assert not is_scalar_indexer(slice(None), 1)
+
+    indexer = 0
+    assert is_scalar_indexer(indexer, 1)
+
+    indexer = (0,)
+    assert is_scalar_indexer(indexer, 1)
+
+
+class TestValidateIndices:
+    def test_validate_indices_ok(self):
+        indices = np.asarray([0, 1])
+        validate_indices(indices, 2)
+        validate_indices(indices[:0], 0)
+        validate_indices(np.array([-1, -1]), 0)
+
+    def test_validate_indices_low(self):
+        indices = np.asarray([0, -2])
+        with pytest.raises(ValueError, match="'indices' contains"):
+            validate_indices(indices, 2)
+
+    def test_validate_indices_high(self):
+        indices = np.asarray([0, 1, 2])
+        with pytest.raises(IndexError, match="indices are out"):
+            validate_indices(indices, 2)
+
+    def test_validate_indices_empty(self):
+        with pytest.raises(IndexError, match="indices are out"):
+            validate_indices(np.array([0, 1]), 0)
diff --git a/pandas/pandas/tests/indexing/test_indexing.py b/pandas/pandas/tests/indexing/test_indexing.py
new file mode 100644
index 0000000000000..fb7e6649c534f
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_indexing.py
@@ -0,0 +1,1115 @@
+"""test fancy indexing & misc"""
+
+import array
+from datetime import datetime
+import re
+import weakref
+
+import numpy as np
+import pytest
+
+from pandas.errors import IndexingError
+
+from pandas.core.dtypes.common import (
+    is_float_dtype,
+    is_integer_dtype,
+    is_object_dtype,
+)
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    date_range,
+    offsets,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.tests.indexing.common import _mklbl
+from pandas.tests.indexing.test_floats import gen_obj
+
+# ------------------------------------------------------------------------
+# Indexing test cases
+
+
+class TestFancy:
+    """pure get/set item & fancy indexing"""
+
+    def test_setitem_ndarray_1d(self):
+        # GH5508
+
+        # len of indexer vs length of the 1d ndarray
+        df = DataFrame(index=Index(np.arange(1, 11), dtype=np.int64))
+        df["foo"] = np.zeros(10, dtype=np.float64)
+        df["bar"] = np.zeros(10, dtype=complex)
+
+        # invalid
+        msg = "Must have equal len keys and value when setting with an iterable"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[df.index[2:5], "bar"] = np.array([2.33j, 1.23 + 0.1j, 2.2, 1.0])
+
+        # valid
+        df.loc[df.index[2:6], "bar"] = np.array([2.33j, 1.23 + 0.1j, 2.2, 1.0])
+
+        result = df.loc[df.index[2:6], "bar"]
+        expected = Series(
+            [2.33j, 1.23 + 0.1j, 2.2, 1.0], index=[3, 4, 5, 6], name="bar"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_ndarray_1d_2(self):
+        # GH5508
+
+        # dtype getting changed?
+        df = DataFrame(index=Index(np.arange(1, 11)))
+        df["foo"] = np.zeros(10, dtype=np.float64)
+        df["bar"] = np.zeros(10, dtype=complex)
+
+        msg = "Must have equal len keys and value when setting with an iterable"
+        with pytest.raises(ValueError, match=msg):
+            df[2:5] = np.arange(1, 4) * 1j
+
+    @pytest.mark.filterwarnings(
+        "ignore:Series.__getitem__ treating keys as positions is deprecated:"
+        "FutureWarning"
+    )
+    def test_getitem_ndarray_3d(self, index, frame_or_series, indexer_sli):
+        # GH 25567
+        obj = gen_obj(frame_or_series, index)
+        idxr = indexer_sli(obj)
+        nd3 = np.random.default_rng(2).integers(5, size=(2, 2, 2))
+
+        msgs = []
+        if frame_or_series is Series and indexer_sli in [tm.setitem, tm.iloc]:
+            msgs.append(r"Wrong number of dimensions. values.ndim > ndim \[3 > 1\]")
+        if frame_or_series is Series or indexer_sli is tm.iloc:
+            msgs.append(r"Buffer has wrong number of dimensions \(expected 1, got 3\)")
+        if indexer_sli is tm.loc or (
+            frame_or_series is Series and indexer_sli is tm.setitem
+        ):
+            msgs.append("Cannot index with multidimensional key")
+        if frame_or_series is DataFrame and indexer_sli is tm.setitem:
+            msgs.append("Index data must be 1-dimensional")
+        if isinstance(index, pd.IntervalIndex) and indexer_sli is tm.iloc:
+            msgs.append("Index data must be 1-dimensional")
+        if isinstance(index, (pd.TimedeltaIndex, pd.DatetimeIndex, pd.PeriodIndex)):
+            msgs.append("Data must be 1-dimensional")
+        if len(index) == 0 or isinstance(index, pd.MultiIndex):
+            msgs.append("positional indexers are out-of-bounds")
+        if type(index) is Index and not isinstance(index._values, np.ndarray):
+            # e.g. Int64
+            msgs.append("values must be a 1D array")
+
+            # string[pyarrow]
+            msgs.append("only handle 1-dimensional arrays")
+
+        msg = "|".join(msgs)
+
+        potential_errors = (IndexError, ValueError, NotImplementedError)
+        with pytest.raises(potential_errors, match=msg):
+            idxr[nd3]
+
+    @pytest.mark.filterwarnings(
+        "ignore:Series.__setitem__ treating keys as positions is deprecated:"
+        "FutureWarning"
+    )
+    def test_setitem_ndarray_3d(self, index, frame_or_series, indexer_sli):
+        # GH 25567
+        obj = gen_obj(frame_or_series, index)
+        idxr = indexer_sli(obj)
+        nd3 = np.random.default_rng(2).integers(5, size=(2, 2, 2))
+
+        if indexer_sli is tm.iloc:
+            err = ValueError
+            msg = f"Cannot set values with ndim > {obj.ndim}"
+        else:
+            err = ValueError
+            msg = "|".join(
+                [
+                    r"Buffer has wrong number of dimensions \(expected 1, got 3\)",
+                    "Cannot set values with ndim > 1",
+                    "Index data must be 1-dimensional",
+                    "Data must be 1-dimensional",
+                    "Array conditional must be same shape as self",
+                ]
+            )
+
+        with pytest.raises(err, match=msg):
+            idxr[nd3] = 0
+
+    def test_getitem_ndarray_0d(self):
+        # GH#24924
+        key = np.array(0)
+
+        # dataframe __getitem__
+        df = DataFrame([[1, 2], [3, 4]])
+        result = df[key]
+        expected = Series([1, 3], name=0)
+        tm.assert_series_equal(result, expected)
+
+        # series __getitem__
+        ser = Series([1, 2])
+        result = ser[key]
+        assert result == 1
+
+    def test_inf_upcast(self):
+        # GH 16957
+        # We should be able to use np.inf as a key
+        # np.inf should cause an index to convert to float
+
+        # Test with np.inf in rows
+        df = DataFrame(columns=[0])
+        df.loc[1] = 1
+        df.loc[2] = 2
+        df.loc[np.inf] = 3
+
+        # make sure we can look up the value
+        assert df.loc[np.inf, 0] == 3
+
+        result = df.index
+        expected = Index([1, 2, np.inf], dtype=np.float64)
+        tm.assert_index_equal(result, expected)
+
+    def test_setitem_dtype_upcast(self):
+        # GH3216
+        df = DataFrame([{"a": 1}, {"a": 3, "b": 2}])
+        df["c"] = np.nan
+        assert df["c"].dtype == np.float64
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[0, "c"] = "foo"
+
+    @pytest.mark.parametrize("val", [3.14, "wxyz"])
+    def test_setitem_dtype_upcast2(self, val):
+        # GH10280
+        df = DataFrame(
+            np.arange(6, dtype="int64").reshape(2, 3),
+            index=list("ab"),
+            columns=["foo", "bar", "baz"],
+        )
+
+        left = df.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            left.loc["a", "bar"] = val
+
+    def test_setitem_dtype_upcast3(self):
+        left = DataFrame(
+            np.arange(6, dtype="int64").reshape(2, 3) / 10.0,
+            index=list("ab"),
+            columns=["foo", "bar", "baz"],
+        )
+        with pytest.raises(TypeError, match="Invalid value"):
+            left.loc["a", "bar"] = "wxyz"
+
+    def test_dups_fancy_indexing(self):
+        # GH 3455
+
+        df = DataFrame(np.eye(3), columns=["a", "a", "b"])
+        result = df[["b", "a"]].columns
+        expected = Index(["b", "a", "a"])
+        tm.assert_index_equal(result, expected)
+
+    def test_dups_fancy_indexing_across_dtypes(self):
+        # across dtypes
+        df = DataFrame([[1, 2, 1.0, 2.0, 3.0, "foo", "bar"]], columns=list("aaaaaaa"))
+        result = DataFrame([[1, 2, 1.0, 2.0, 3.0, "foo", "bar"]])
+        result.columns = list("aaaaaaa")  # GH#3468
+
+        # GH#3509 smoke tests for indexing with duplicate columns
+        df.iloc[:, 4]
+        result.iloc[:, 4]
+
+        tm.assert_frame_equal(df, result)
+
+    def test_dups_fancy_indexing_not_in_order(self):
+        # GH 3561, dups not in selected order
+        df = DataFrame(
+            {"test": [5, 7, 9, 11], "test1": [4.0, 5, 6, 7], "other": list("abcd")},
+            index=["A", "A", "B", "C"],
+        )
+        rows = ["C", "B"]
+        expected = DataFrame(
+            {"test": [11, 9], "test1": [7.0, 6], "other": ["d", "c"]}, index=rows
+        )
+        result = df.loc[rows]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[Index(rows)]
+        tm.assert_frame_equal(result, expected)
+
+        rows = ["C", "B", "E"]
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[rows]
+
+        # see GH5553, make sure we use the right indexer
+        rows = ["F", "G", "H", "C", "B", "E"]
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[rows]
+
+    def test_dups_fancy_indexing_only_missing_label(self, using_infer_string):
+        # List containing only missing label
+        dfnu = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), index=list("AABCD")
+        )
+        if using_infer_string:
+            with pytest.raises(
+                KeyError,
+                match=re.escape(
+                    "\"None of [Index(['E'], dtype='str')] are in the [index]\""
+                ),
+            ):
+                dfnu.loc[["E"]]
+        else:
+            with pytest.raises(
+                KeyError,
+                match=re.escape(
+                    "\"None of [Index(['E'], dtype='object')] are in the [index]\""
+                ),
+            ):
+                dfnu.loc[["E"]]
+
+    @pytest.mark.parametrize("vals", [[0, 1, 2], list("abc")])
+    def test_dups_fancy_indexing_missing_label(self, vals):
+        # GH 4619; duplicate indexer with missing label
+        df = DataFrame({"A": vals})
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[[0, 8, 0]]
+
+    def test_dups_fancy_indexing_non_unique(self):
+        # non unique with non unique selector
+        df = DataFrame({"test": [5, 7, 9, 11]}, index=["A", "A", "B", "C"])
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[["A", "A", "E"]]
+
+    def test_dups_fancy_indexing2(self):
+        # GH 5835
+        # dups on index and missing values
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 5)),
+            columns=["A", "B", "B", "B", "A"],
+        )
+
+        with pytest.raises(KeyError, match="not in index"):
+            df.loc[:, ["A", "B", "C"]]
+
+    def test_dups_fancy_indexing3(self):
+        # GH 6504, multi-axis indexing
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((9, 2)),
+            index=[1, 1, 1, 2, 2, 2, 3, 3, 3],
+            columns=["a", "b"],
+        )
+
+        expected = df.iloc[0:6]
+        result = df.loc[[1, 2]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = df
+        result = df.loc[:, ["a", "b"]]
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.iloc[0:6, :]
+        result = df.loc[[1, 2], ["a", "b"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_duplicate_int_indexing(self, indexer_sl):
+        # GH 17347
+        ser = Series(range(3), index=[1, 1, 3])
+        expected = Series(range(2), index=[1, 1])
+        result = indexer_sl(ser)[[1]]
+        tm.assert_series_equal(result, expected)
+
+    def test_indexing_mixed_frame_bug(self):
+        # GH3492
+        df = DataFrame(
+            {"a": {1: "aaa", 2: "bbb", 3: "ccc"}, "b": {1: 111, 2: 222, 3: 333}}
+        )
+
+        # this works, new column is created correctly
+        df["test"] = df["a"].apply(lambda x: "_" if x == "aaa" else x)
+
+        # this does not work, ie column test is not changed
+        idx = df["test"] == "_"
+        temp = df.loc[idx, "a"].apply(lambda x: "-----" if x == "aaa" else x)
+        df.loc[idx, "test"] = temp
+        assert df.iloc[0, 2] == "-----"
+
+    def test_multitype_list_index_access(self):
+        # GH 10610
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 5)), columns=["a"] + [20, 21, 22, 23]
+        )
+
+        with pytest.raises(KeyError, match=re.escape("'[26, -8] not in index'")):
+            df[[22, 26, -8]]
+        assert df[21].shape[0] == df.shape[0]
+
+    def test_set_index_nan(self):
+        # GH 3586
+        df = DataFrame(
+            {
+                "PRuid": {
+                    17: "nonQC",
+                    18: "nonQC",
+                    19: "nonQC",
+                    20: "10",
+                    21: "11",
+                    22: "12",
+                    23: "13",
+                    24: "24",
+                    25: "35",
+                    26: "46",
+                    27: "47",
+                    28: "48",
+                    29: "59",
+                    30: "10",
+                },
+                "QC": {
+                    17: 0.0,
+                    18: 0.0,
+                    19: 0.0,
+                    20: np.nan,
+                    21: np.nan,
+                    22: np.nan,
+                    23: np.nan,
+                    24: 1.0,
+                    25: np.nan,
+                    26: np.nan,
+                    27: np.nan,
+                    28: np.nan,
+                    29: np.nan,
+                    30: np.nan,
+                },
+                "data": {
+                    17: 7.9544899999999998,
+                    18: 8.0142609999999994,
+                    19: 7.8591520000000008,
+                    20: 0.86140349999999999,
+                    21: 0.87853110000000001,
+                    22: 0.8427041999999999,
+                    23: 0.78587700000000005,
+                    24: 0.73062459999999996,
+                    25: 0.81668560000000001,
+                    26: 0.81927080000000008,
+                    27: 0.80705009999999999,
+                    28: 0.81440240000000008,
+                    29: 0.80140849999999997,
+                    30: 0.81307740000000006,
+                },
+                "year": {
+                    17: 2006,
+                    18: 2007,
+                    19: 2008,
+                    20: 1985,
+                    21: 1985,
+                    22: 1985,
+                    23: 1985,
+                    24: 1985,
+                    25: 1985,
+                    26: 1985,
+                    27: 1985,
+                    28: 1985,
+                    29: 1985,
+                    30: 1986,
+                },
+            }
+        ).reset_index()
+
+        result = (
+            df.set_index(["year", "PRuid", "QC"])
+            .reset_index()
+            .reindex(columns=df.columns)
+        )
+        tm.assert_frame_equal(result, df)
+
+    def test_multi_assign(self):
+        # GH 3626, an assignment of a sub-df to a df
+        # set float64 to avoid upcast when setting nan
+        df = DataFrame(
+            {
+                "FC": ["a", "b", "a", "b", "a", "b"],
+                "PF": [0, 0, 0, 0, 1, 1],
+                "col1": list(range(6)),
+                "col2": list(range(6, 12)),
+            }
+        ).astype({"col2": "float64"})
+        df.iloc[1, 0] = np.nan
+        df2 = df.copy()
+
+        mask = ~df2.FC.isna()
+        cols = ["col1", "col2"]
+
+        dft = df2 * 2
+        dft.iloc[3, 3] = np.nan
+
+        expected = DataFrame(
+            {
+                "FC": ["a", np.nan, "a", "b", "a", "b"],
+                "PF": [0, 0, 0, 0, 1, 1],
+                "col1": Series([0, 1, 4, 6, 8, 10]),
+                "col2": [12, 7, 16, np.nan, 20, 22],
+            }
+        )
+
+        # frame on rhs
+        df2.loc[mask, cols] = dft.loc[mask, cols]
+        tm.assert_frame_equal(df2, expected)
+
+        # with an ndarray on rhs
+        # coerces to float64 because values has float64 dtype
+        # GH 14001
+        expected = DataFrame(
+            {
+                "FC": ["a", np.nan, "a", "b", "a", "b"],
+                "PF": [0, 0, 0, 0, 1, 1],
+                "col1": [0, 1, 4, 6, 8, 10],
+                "col2": [12, 7, 16, np.nan, 20, 22],
+            }
+        )
+        df2 = df.copy()
+        df2.loc[mask, cols] = dft.loc[mask, cols].values
+        tm.assert_frame_equal(df2, expected)
+
+    def test_multi_assign_broadcasting_rhs(self):
+        # broadcasting on the rhs is required
+        df = DataFrame(
+            {
+                "A": [1, 2, 0, 0, 0],
+                "B": [0, 0, 0, 10, 11],
+                "C": [0, 0, 0, 10, 11],
+                "D": [3, 4, 5, 6, 7],
+            }
+        )
+
+        expected = df.copy()
+        mask = expected["A"] == 0
+        for col in ["A", "B"]:
+            expected.loc[mask, col] = df["D"]
+
+        df.loc[df["A"] == 0, ["A", "B"]] = df["D"].copy()
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_list(self):
+        # GH 6043
+        # iloc with a list
+        df = DataFrame(index=[0, 1], columns=[0])
+        df.iloc[1, 0] = [1, 2, 3]
+        df.iloc[1, 0] = [1, 2]
+
+        result = DataFrame(index=[0, 1], columns=[0])
+        result.iloc[1, 0] = [1, 2]
+
+        tm.assert_frame_equal(result, df)
+
+    def test_string_slice(self):
+        # GH 14424
+        # string indexing against datetimelike with object
+        # dtype should properly raises KeyError
+        df = DataFrame([1], Index([pd.Timestamp("2011-01-01")], dtype=object))
+        assert df.index._is_all_dates
+        with pytest.raises(KeyError, match="'2011'"):
+            df["2011"]
+
+        with pytest.raises(KeyError, match="'2011'"):
+            df.loc["2011", 0]
+
+    def test_string_slice_empty(self):
+        # GH 14424
+
+        df = DataFrame()
+        assert not df.index._is_all_dates
+        with pytest.raises(KeyError, match="'2011'"):
+            df["2011"]
+
+        with pytest.raises(KeyError, match="^0$"):
+            df.loc["2011", 0]
+
+    def test_astype_assignment(self, using_infer_string):
+        # GH4312 (iloc)
+        df_orig = DataFrame(
+            [["1", "2", "3", ".4", 5, 6.0, "foo"]], columns=list("ABCDEFG")
+        )
+        df_orig[list("ABCDG")] = df_orig[list("ABCDG")].astype(object)
+
+        df = df_orig.copy()
+
+        # with the enforcement of GH#45333 in 2.0, this setting is attempted inplace,
+        #  so object dtype is retained
+        df.iloc[:, 0:2] = df.iloc[:, 0:2].astype(np.int64)
+        expected = DataFrame(
+            [[1, 2, "3", ".4", 5, 6.0, "foo"]], columns=list("ABCDEFG")
+        )
+        expected[list("CDG")] = expected[list("CDG")].astype(object)
+        expected["A"] = expected["A"].astype(object)
+        expected["B"] = expected["B"].astype(object)
+        tm.assert_frame_equal(df, expected)
+
+        # GH5702 (loc)
+        df = df_orig.copy()
+        df.loc[:, "A"] = df.loc[:, "A"].astype(np.int64)
+        expected = DataFrame(
+            [[1, "2", "3", ".4", 5, 6.0, "foo"]], columns=list("ABCDEFG")
+        )
+        expected[list("ABCDG")] = expected[list("ABCDG")].astype(object)
+        tm.assert_frame_equal(df, expected)
+
+        df = df_orig.copy()
+
+        df.loc[:, ["B", "C"]] = df.loc[:, ["B", "C"]].astype(np.int64)
+        expected = DataFrame(
+            [["1", 2, 3, ".4", 5, 6.0, "foo"]], columns=list("ABCDEFG")
+        )
+        expected[list("ABCDG")] = expected[list("ABCDG")].astype(object)
+        tm.assert_frame_equal(df, expected)
+
+    def test_astype_assignment_full_replacements(self):
+        # full replacements / no nans
+        df = DataFrame({"A": [1.0, 2.0, 3.0, 4.0]})
+
+        # With the enforcement of GH#45333 in 2.0, this assignment occurs inplace,
+        #  so float64 is retained
+        df.iloc[:, 0] = df["A"].astype(np.int64)
+        expected = DataFrame({"A": [1.0, 2.0, 3.0, 4.0]})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame({"A": [1.0, 2.0, 3.0, 4.0]})
+        df.loc[:, "A"] = df["A"].astype(np.int64)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("indexer", [tm.getitem, tm.loc])
+    def test_index_type_coercion(self, indexer):
+        # GH 11836
+        # if we have an index type and set it with something that looks
+        # to numpy like the same, but is actually, not
+        # (e.g. setting with a float or string '0')
+        # then we need to coerce to object
+
+        # integer indexes
+        for s in [Series(range(5)), Series(range(5), index=range(1, 6))]:
+            assert is_integer_dtype(s.index)
+
+            s2 = s.copy()
+            indexer(s2)[0.1] = 0
+            assert is_float_dtype(s2.index)
+            assert indexer(s2)[0.1] == 0
+
+            s2 = s.copy()
+            indexer(s2)[0.0] = 0
+            exp = s.index
+            if 0 not in s:
+                exp = Index(s.index.tolist() + [0])
+            tm.assert_index_equal(s2.index, exp)
+
+            s2 = s.copy()
+            indexer(s2)["0"] = 0
+            assert is_object_dtype(s2.index)
+
+        for s in [Series(range(5), index=np.arange(5.0))]:
+            assert is_float_dtype(s.index)
+
+            s2 = s.copy()
+            indexer(s2)[0.1] = 0
+            assert is_float_dtype(s2.index)
+            assert indexer(s2)[0.1] == 0
+
+            s2 = s.copy()
+            indexer(s2)[0.0] = 0
+            tm.assert_index_equal(s2.index, s.index)
+
+            s2 = s.copy()
+            indexer(s2)["0"] = 0
+            assert is_object_dtype(s2.index)
+
+
+class TestMisc:
+    def test_float_index_to_mixed(self):
+        df = DataFrame(
+            {
+                0.0: np.random.default_rng(2).random(10),
+                1.0: np.random.default_rng(2).random(10),
+            }
+        )
+        df["a"] = 10
+
+        expected = DataFrame({0.0: df[0.0], 1.0: df[1.0], "a": [10] * 10})
+        tm.assert_frame_equal(expected, df)
+
+    def test_float_index_non_scalar_assignment(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [3, 4, 5]}, index=[1.0, 2.0, 3.0])
+        df.loc[df.index[:2]] = 1
+        expected = DataFrame({"a": [1, 1, 3], "b": [1, 1, 5]}, index=df.index)
+        tm.assert_frame_equal(expected, df)
+
+    def test_loc_setitem_fullindex_views(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [3, 4, 5]}, index=[1.0, 2.0, 3.0])
+        df2 = df.copy()
+        df.loc[df.index] = df.loc[df.index]
+        tm.assert_frame_equal(df, df2)
+
+    def test_rhs_alignment(self):
+        # GH8258, tests that both rows & columns are aligned to what is
+        # assigned to. covers both uniform data-type & multi-type cases
+        def run_tests(df, rhs, right_loc, right_iloc):
+            # label, index, slice
+            lbl_one, idx_one, slice_one = list("bcd"), [1, 2, 3], slice(1, 4)
+            lbl_two, idx_two, slice_two = ["joe", "jolie"], [1, 2], slice(1, 3)
+
+            left = df.copy()
+            left.loc[lbl_one, lbl_two] = rhs
+            tm.assert_frame_equal(left, right_loc)
+
+            left = df.copy()
+            left.iloc[idx_one, idx_two] = rhs
+            tm.assert_frame_equal(left, right_iloc)
+
+            left = df.copy()
+            left.iloc[slice_one, slice_two] = rhs
+            tm.assert_frame_equal(left, right_iloc)
+
+        xs = np.arange(20).reshape(5, 4)
+        cols = ["jim", "joe", "jolie", "joline"]
+        df = DataFrame(xs, columns=cols, index=list("abcde"), dtype="int64")
+
+        # right hand side; permute the indices and multiply by -2
+        rhs = -2 * df.iloc[3:0:-1, 2:0:-1]
+
+        # expected `right` result; just multiply by -2
+        right_iloc = df.copy()
+        right_iloc["joe"] = [1, 14, 10, 6, 17]
+        right_iloc["jolie"] = [2, 13, 9, 5, 18]
+        right_iloc.iloc[1:4, 1:3] *= -2
+        right_loc = df.copy()
+        right_loc.iloc[1:4, 1:3] *= -2
+
+        # run tests with uniform dtypes
+        run_tests(df, rhs, right_loc, right_iloc)
+
+        # make frames multi-type & re-run tests
+        for frame in [df, rhs, right_loc, right_iloc]:
+            frame["joe"] = frame["joe"].astype("float64")
+            frame["jolie"] = frame["jolie"].map(lambda x: f"@{x}")
+        right_iloc["joe"] = [1.0, "@-28", "@-20", "@-12", 17.0]
+        right_iloc["jolie"] = ["@2", -26.0, -18.0, -10.0, "@18"]
+        with pytest.raises(TypeError, match="Invalid value"):
+            run_tests(df, rhs, right_loc, right_iloc)
+
+    @pytest.mark.parametrize(
+        "idx", [_mklbl("A", 20), np.arange(20) + 100, np.linspace(100, 150, 20)]
+    )
+    def test_str_label_slicing_with_negative_step(self, idx):
+        SLC = pd.IndexSlice
+
+        idx = Index(idx)
+        ser = Series(np.arange(20), index=idx)
+        tm.assert_indexing_slices_equivalent(ser, SLC[idx[9] :: -1], SLC[9::-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[: idx[9] : -1], SLC[:8:-1])
+        tm.assert_indexing_slices_equivalent(
+            ser, SLC[idx[13] : idx[9] : -1], SLC[13:8:-1]
+        )
+        tm.assert_indexing_slices_equivalent(ser, SLC[idx[9] : idx[13] : -1], SLC[:0])
+
+    def test_slice_with_zero_step_raises(self, index, indexer_sl, frame_or_series):
+        obj = frame_or_series(np.arange(len(index)), index=index)
+        with pytest.raises(ValueError, match="slice step cannot be zero"):
+            indexer_sl(obj)[::0]
+
+    def test_loc_setitem_indexing_assignment_dict_already_exists(self):
+        index = Index([-5, 0, 5], name="z")
+        df = DataFrame({"x": [1, 2, 6], "y": [2, 2, 8]}, index=index)
+        expected = df.copy()
+        rhs = {"x": 9, "y": 99}
+        df.loc[5] = rhs
+        expected.loc[5] = [9, 99]
+        tm.assert_frame_equal(df, expected)
+
+        # GH#38335 same thing, mixed dtypes
+        df = DataFrame({"x": [1, 2, 6], "y": [2.0, 2.0, 8.0]}, index=index)
+        df.loc[5] = rhs
+        expected = DataFrame({"x": [1, 2, 9], "y": [2.0, 2.0, 99.0]}, index=index)
+        tm.assert_frame_equal(df, expected)
+
+    def test_iloc_getitem_indexing_dtypes_on_empty(self):
+        # Check that .iloc returns correct dtypes GH9983
+        df = DataFrame({"a": [1, 2, 3], "b": ["b", "b2", "b3"]})
+        df2 = df.iloc[[], :]
+
+        assert df2.loc[:, "a"].dtype == np.int64
+        tm.assert_series_equal(df2.loc[:, "a"], df2.iloc[:, 0])
+
+    @pytest.mark.parametrize("size", [5, 999999, 1000000])
+    def test_loc_range_in_series_indexing(self, size):
+        # range can cause an indexing error
+        # GH 11652
+        s = Series(index=range(size), dtype=np.float64)
+        s.loc[range(1)] = 42
+        tm.assert_series_equal(s.loc[range(1)], Series(42.0, index=range(1)))
+
+        s.loc[range(2)] = 43
+        tm.assert_series_equal(s.loc[range(2)], Series(43.0, index=range(2)))
+
+    def test_partial_boolean_frame_indexing(self):
+        # GH 17170
+        df = DataFrame(
+            np.arange(9.0).reshape(3, 3), index=list("abc"), columns=list("ABC")
+        )
+        index_df = DataFrame(1, index=list("ab"), columns=list("AB"))
+        result = df[index_df.notnull()]
+        expected = DataFrame(
+            np.array([[0.0, 1.0, np.nan], [3.0, 4.0, np.nan], [np.nan] * 3]),
+            index=list("abc"),
+            columns=list("ABC"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_no_reference_cycle(self):
+        df = DataFrame({"a": [0, 1], "b": [2, 3]})
+        for name in ("loc", "iloc", "at", "iat"):
+            getattr(df, name)
+        wr = weakref.ref(df)
+        del df
+        assert wr() is None
+
+    def test_label_indexing_on_nan(self, nulls_fixture):
+        # GH 32431
+        df = Series([1, "{1,2}", 1, nulls_fixture])
+        vc = df.value_counts(dropna=False)
+        result1 = vc.loc[nulls_fixture]
+        result2 = vc[nulls_fixture]
+
+        expected = 1
+        assert result1 == expected
+        assert result2 == expected
+
+
+class TestDataframeNoneCoercion:
+    EXPECTED_SINGLE_ROW_RESULTS = [
+        # For numeric series, we should coerce to NaN.
+        ([1, 2, 3], [np.nan, 2, 3], FutureWarning),
+        ([1.0, 2.0, 3.0], [np.nan, 2.0, 3.0], None),
+        # For datetime series, we should coerce to NaT.
+        (
+            [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            [NaT, datetime(2000, 1, 2), datetime(2000, 1, 3)],
+            None,
+        ),
+        # For objects, we should preserve the None value.
+        (["foo", "bar", "baz"], [None, "bar", "baz"], None),
+    ]
+
+    @pytest.mark.parametrize("expected", EXPECTED_SINGLE_ROW_RESULTS)
+    def test_coercion_with_loc(self, expected):
+        start_data, expected_result, warn = expected
+
+        start_dataframe = DataFrame({"foo": start_data})
+        start_dataframe.loc[0, ["foo"]] = None
+
+        expected_dataframe = DataFrame({"foo": expected_result})
+        tm.assert_frame_equal(start_dataframe, expected_dataframe)
+
+    @pytest.mark.parametrize("expected", EXPECTED_SINGLE_ROW_RESULTS)
+    def test_coercion_with_setitem_and_dataframe(self, expected):
+        start_data, expected_result, warn = expected
+
+        start_dataframe = DataFrame({"foo": start_data})
+        start_dataframe[start_dataframe["foo"] == start_dataframe["foo"][0]] = None
+
+        expected_dataframe = DataFrame({"foo": expected_result})
+        tm.assert_frame_equal(start_dataframe, expected_dataframe)
+
+    @pytest.mark.parametrize("expected", EXPECTED_SINGLE_ROW_RESULTS)
+    def test_none_coercion_loc_and_dataframe(self, expected):
+        start_data, expected_result, warn = expected
+
+        start_dataframe = DataFrame({"foo": start_data})
+        start_dataframe.loc[start_dataframe["foo"] == start_dataframe["foo"][0]] = None
+
+        expected_dataframe = DataFrame({"foo": expected_result})
+        tm.assert_frame_equal(start_dataframe, expected_dataframe)
+
+    def test_none_coercion_mixed_dtypes(self):
+        start_dataframe = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": [1.0, 2.0, 3.0],
+                "c": [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                "d": ["a", "b", "c"],
+            }
+        )
+        start_dataframe.iloc[0] = None
+
+        exp = DataFrame(
+            {
+                "a": [np.nan, 2, 3],
+                "b": [np.nan, 2.0, 3.0],
+                "c": [NaT, datetime(2000, 1, 2), datetime(2000, 1, 3)],
+                "d": [None, "b", "c"],
+            }
+        )
+        tm.assert_frame_equal(start_dataframe, exp)
+
+
+class TestDatetimelikeCoercion:
+    def test_setitem_dt64_string_scalar(self, tz_naive_fixture, indexer_sli):
+        # dispatching _can_hold_element to underlying DatetimeArray
+        tz = tz_naive_fixture
+
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        ser = Series(dti.copy(deep=True))
+
+        values = ser._values
+
+        newval = "2018-01-01"
+        values._validate_setitem_value(newval)
+
+        indexer_sli(ser)[0] = newval
+
+        if tz is None:
+            # TODO(EA2D): we can make this no-copy in tz-naive case too
+            assert ser.dtype == dti.dtype
+            assert ser._values._ndarray is values._ndarray
+        else:
+            assert ser._values is values
+
+    @pytest.mark.parametrize("box", [list, np.array, pd.array, pd.Categorical, Index])
+    @pytest.mark.parametrize(
+        "key", [[0, 1], slice(0, 2), np.array([True, True, False])]
+    )
+    def test_setitem_dt64_string_values(self, tz_naive_fixture, indexer_sli, key, box):
+        # dispatching _can_hold_element to underling DatetimeArray
+        tz = tz_naive_fixture
+
+        if isinstance(key, slice) and indexer_sli is tm.loc:
+            key = slice(0, 1)
+
+        dti = date_range("2016-01-01", periods=3, tz=tz)
+        ser = Series(dti.copy(deep=True))
+
+        values = ser._values
+
+        newvals = box(["2019-01-01", "2010-01-02"])
+        values._validate_setitem_value(newvals)
+
+        indexer_sli(ser)[key] = newvals
+
+        if tz is None:
+            # TODO(EA2D): we can make this no-copy in tz-naive case too
+            assert ser.dtype == dti.dtype
+            assert ser._values._ndarray is values._ndarray
+        else:
+            assert ser._values is values
+
+    @pytest.mark.parametrize("scalar", ["3 Days", offsets.Hour(4)])
+    def test_setitem_td64_scalar(self, indexer_sli, scalar):
+        # dispatching _can_hold_element to underling TimedeltaArray
+        tdi = timedelta_range("1 Day", periods=3)
+        ser = Series(tdi.copy(deep=True))
+
+        values = ser._values
+        values._validate_setitem_value(scalar)
+
+        indexer_sli(ser)[0] = scalar
+        assert ser._values._ndarray is values._ndarray
+
+    @pytest.mark.parametrize("box", [list, np.array, pd.array, pd.Categorical, Index])
+    @pytest.mark.parametrize(
+        "key", [[0, 1], slice(0, 2), np.array([True, True, False])]
+    )
+    def test_setitem_td64_string_values(self, indexer_sli, key, box):
+        # dispatching _can_hold_element to underling TimedeltaArray
+        if isinstance(key, slice) and indexer_sli is tm.loc:
+            key = slice(0, 1)
+
+        tdi = timedelta_range("1 Day", periods=3)
+        ser = Series(tdi.copy(deep=True))
+
+        values = ser._values
+
+        newvals = box(["10 Days", "44 hours"])
+        values._validate_setitem_value(newvals)
+
+        indexer_sli(ser)[key] = newvals
+        assert ser._values._ndarray is values._ndarray
+
+
+def test_extension_array_cross_section():
+    # A cross-section of a homogeneous EA should be an EA
+    df = DataFrame(
+        {
+            "A": pd.array([1, 2], dtype="Int64"),
+            "B": pd.array([3, 4], dtype="Int64"),
+        },
+        index=["a", "b"],
+    )
+    expected = Series(pd.array([1, 3], dtype="Int64"), index=["A", "B"], name="a")
+    result = df.loc["a"]
+    tm.assert_series_equal(result, expected)
+
+    result = df.iloc[0]
+    tm.assert_series_equal(result, expected)
+
+
+def test_extension_array_cross_section_converts():
+    # all numeric columns -> numeric series
+    df = DataFrame(
+        {
+            "A": pd.array([1, 2], dtype="Int64"),
+            "B": np.array([1, 2], dtype="int64"),
+        },
+        index=["a", "b"],
+    )
+    result = df.loc["a"]
+    expected = Series([1, 1], dtype="Int64", index=["A", "B"], name="a")
+    tm.assert_series_equal(result, expected)
+
+    result = df.iloc[0]
+    tm.assert_series_equal(result, expected)
+
+    # mixed columns -> object series
+    df = DataFrame(
+        {"A": pd.array([1, 2], dtype="Int64"), "B": np.array(["a", "b"])},
+        index=["a", "b"],
+    )
+    result = df.loc["a"]
+    expected = Series([1, "a"], dtype=object, index=["A", "B"], name="a")
+    tm.assert_series_equal(result, expected)
+
+    result = df.iloc[0]
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser, keys",
+    [(Series([10]), (0, 0)), (Series([1, 2, 3], index=list("abc")), (0, 1))],
+)
+def test_ser_tup_indexer_exceeds_dimensions(ser, keys, indexer_li):
+    # GH#13831
+    exp_err, exp_msg = IndexingError, "Too many indexers"
+    with pytest.raises(exp_err, match=exp_msg):
+        indexer_li(ser)[keys]
+
+    if indexer_li == tm.iloc:
+        # For iloc.__setitem__ we let numpy handle the error reporting.
+        exp_err, exp_msg = IndexError, "too many indices for array"
+
+    with pytest.raises(exp_err, match=exp_msg):
+        indexer_li(ser)[keys] = 0
+
+
+def test_ser_list_indexer_exceeds_dimensions(indexer_li):
+    # GH#13831
+    # Make sure an exception is raised when a tuple exceeds the dimension of the series,
+    # but not list when a list is used.
+    ser = Series([10])
+    res = indexer_li(ser)[[0, 0]]
+    exp = Series([10, 10], index=Index([0, 0]))
+    tm.assert_series_equal(res, exp)
+
+
+@pytest.mark.parametrize(
+    "value", [(0, 1), [0, 1], np.array([0, 1]), array.array("b", [0, 1])]
+)
+def test_scalar_setitem_with_nested_value(value):
+    # For numeric data, we try to unpack and thus raise for mismatching length
+    df = DataFrame({"A": [1, 2, 3]})
+    msg = "|".join(
+        [
+            "Must have equal len keys and value",
+            "setting an array element with a sequence",
+        ]
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.loc[0, "B"] = value
+
+    # TODO For object dtype this happens as well, but should we rather preserve
+    # the nested data and set as such?
+    df = DataFrame({"A": [1, 2, 3], "B": np.array([1, "a", "b"], dtype=object)})
+    with pytest.raises(ValueError, match="Must have equal len keys and value"):
+        df.loc[0, "B"] = value
+    # if isinstance(value, np.ndarray):
+    #     assert (df.loc[0, "B"] == value).all()
+    # else:
+    #     assert df.loc[0, "B"] == value
+
+
+@pytest.mark.parametrize(
+    "value", [(0, 1), [0, 1], np.array([0, 1]), array.array("b", [0, 1])]
+)
+def test_scalar_setitem_series_with_nested_value(value, indexer_sli):
+    # For numeric data, we try to unpack and thus raise for mismatching length
+    ser = Series([1, 2, 3])
+    with pytest.raises(ValueError, match="setting an array element with a sequence"):
+        indexer_sli(ser)[0] = value
+
+    # but for object dtype we preserve the nested data and set as such
+    ser = Series([1, "a", "b"], dtype=object)
+    indexer_sli(ser)[0] = value
+    if isinstance(value, np.ndarray):
+        assert (ser.loc[0] == value).all()
+    else:
+        assert ser.loc[0] == value
+
+
+@pytest.mark.parametrize(
+    "value", [(0.0,), [0.0], np.array([0.0]), array.array("d", [0.0])]
+)
+def test_scalar_setitem_with_nested_value_length1(value):
+    # https://github.com/pandas-dev/pandas/issues/46268
+
+    # For numeric data, assigning length-1 array to scalar position gets unpacked
+    df = DataFrame({"A": [1, 2, 3]})
+    df.loc[0, "B"] = value
+    expected = DataFrame({"A": [1, 2, 3], "B": [0.0, np.nan, np.nan]})
+    tm.assert_frame_equal(df, expected)
+
+    # but for object dtype we preserve the nested data
+    df = DataFrame({"A": [1, 2, 3], "B": np.array([1, "a", "b"], dtype=object)})
+    df.loc[0, "B"] = value
+    if isinstance(value, np.ndarray):
+        assert (df.loc[0, "B"] == value).all()
+    else:
+        assert df.loc[0, "B"] == value
+
+
+@pytest.mark.parametrize(
+    "value", [(0.0,), [0.0], np.array([0.0]), array.array("d", [0.0])]
+)
+def test_scalar_setitem_series_with_nested_value_length1(value, indexer_sli):
+    # For numeric data, assigning length-1 array to scalar position gets unpacked
+    # TODO this only happens in case of ndarray, should we make this consistent
+    # for all list-likes? (as happens for DataFrame.(i)loc, see test above)
+    ser = Series([1.0, 2.0, 3.0])
+    if isinstance(value, np.ndarray):
+        indexer_sli(ser)[0] = value
+        expected = Series([0.0, 2.0, 3.0])
+        tm.assert_series_equal(ser, expected)
+    else:
+        with pytest.raises(
+            ValueError, match="setting an array element with a sequence"
+        ):
+            indexer_sli(ser)[0] = value
+
+    # but for object dtype we preserve the nested data
+    ser = Series([1, "a", "b"], dtype=object)
+    indexer_sli(ser)[0] = value
+    if isinstance(value, np.ndarray):
+        assert (ser.loc[0] == value).all()
+    else:
+        assert ser.loc[0] == value
+
+
+def test_object_dtype_series_set_series_element():
+    # GH 48933
+    s1 = Series(dtype="O", index=["a", "b"])
+
+    s1["a"] = Series()
+    s1.loc["b"] = Series()
+
+    tm.assert_series_equal(s1.loc["a"], Series())
+    tm.assert_series_equal(s1.loc["b"], Series())
+
+    s2 = Series(dtype="O", index=["a", "b"])
+
+    s2.iloc[1] = Series()
+    tm.assert_series_equal(s2.iloc[1], Series())
diff --git a/pandas/pandas/tests/indexing/test_loc.py b/pandas/pandas/tests/indexing/test_loc.py
new file mode 100644
index 0000000000000..17e610bda93e4
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_loc.py
@@ -0,0 +1,3317 @@
+"""test label based indexing with loc"""
+
+from collections import namedtuple
+import contextlib
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+import re
+
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+from pandas.errors import IndexingError
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IndexSlice,
+    MultiIndex,
+    Period,
+    PeriodIndex,
+    Series,
+    SparseDtype,
+    Timedelta,
+    Timestamp,
+    date_range,
+    timedelta_range,
+    to_datetime,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.api.types import is_scalar
+from pandas.core.indexing import _one_ellipsis_message
+from pandas.tests.indexing.common import check_indexing_smoketest_or_raises
+
+
+@pytest.mark.parametrize(
+    "series, new_series, expected_ser",
+    [
+        [[np.nan, np.nan, "b"], ["a", np.nan, np.nan], [False, True, True]],
+        [[np.nan, "b"], ["a", np.nan], [False, True]],
+    ],
+)
+def test_not_change_nan_loc(series, new_series, expected_ser):
+    # GH 28403
+    df = DataFrame({"A": series})
+    df.loc[:, "A"] = new_series
+    expected = DataFrame({"A": expected_ser})
+    tm.assert_frame_equal(df.isna(), expected)
+    tm.assert_frame_equal(df.notna(), ~expected)
+
+
+class TestLoc:
+    def test_none_values_on_string_columns(self, using_infer_string):
+        # Issue #32218
+        df = DataFrame(["1", "2", None], columns=["a"], dtype=object)
+        assert df.loc[2, "a"] is None
+
+        df = DataFrame(["1", "2", None], columns=["a"], dtype="str")
+        if using_infer_string:
+            assert np.isnan(df.loc[2, "a"])
+        else:
+            assert df.loc[2, "a"] is None
+
+    def test_loc_getitem_int(self, frame_or_series):
+        # int label
+        obj = frame_or_series(range(3), index=Index(list("abc"), dtype=object))
+        check_indexing_smoketest_or_raises(obj, "loc", 2, fails=KeyError)
+
+    def test_loc_getitem_label(self, frame_or_series):
+        # label
+        obj = frame_or_series()
+        check_indexing_smoketest_or_raises(obj, "loc", "c", fails=KeyError)
+
+    @pytest.mark.parametrize("key", ["f", 20])
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcd"), dtype=object),
+            Index([2, 4, "null", 8], dtype=object),
+            date_range("20130101", periods=4),
+            Index(range(0, 8, 2), dtype=np.float64),
+            Index([]),
+        ],
+    )
+    def test_loc_getitem_label_out_of_range(self, key, index, frame_or_series):
+        obj = frame_or_series(range(len(index)), index=index)
+        # out of range label
+        check_indexing_smoketest_or_raises(obj, "loc", key, fails=KeyError)
+
+    @pytest.mark.parametrize("key", [[0, 1, 2], [1, 3.0, "A"]])
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_loc_getitem_label_list(self, key, dtype, frame_or_series):
+        obj = frame_or_series(range(3), index=Index([0, 1, 2], dtype=dtype))
+        # list of labels
+        check_indexing_smoketest_or_raises(obj, "loc", key, fails=KeyError)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            None,
+            Index([0, 1, 2], dtype=np.int64),
+            Index([0, 1, 2], dtype=np.uint64),
+            Index([0, 1, 2], dtype=np.float64),
+            MultiIndex.from_arrays([range(3), range(3)]),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "key", [[0, 1, 2], [0, 2, 10], [3, 6, 7], [(1, 3), (1, 4), (2, 5)]]
+    )
+    def test_loc_getitem_label_list_with_missing(self, key, index, frame_or_series):
+        if index is None:
+            obj = frame_or_series()
+        else:
+            obj = frame_or_series(range(len(index)), index=index)
+        check_indexing_smoketest_or_raises(obj, "loc", key, fails=KeyError)
+
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64])
+    def test_loc_getitem_label_list_fails(self, dtype, frame_or_series):
+        # fails
+        obj = frame_or_series(range(3), Index([0, 1, 2], dtype=dtype))
+        check_indexing_smoketest_or_raises(
+            obj, "loc", [20, 30, 40], axes=1, fails=KeyError
+        )
+
+    def test_loc_getitem_bool(self, frame_or_series):
+        obj = frame_or_series()
+        # boolean indexers
+        b = [True, False, True, False]
+
+        check_indexing_smoketest_or_raises(obj, "loc", b, fails=IndexError)
+
+    @pytest.mark.parametrize(
+        "slc, indexes, axes, fails",
+        [
+            [
+                slice(1, 3),
+                [
+                    Index(list("abcd"), dtype=object),
+                    Index([2, 4, "null", 8], dtype=object),
+                    None,
+                    date_range("20130101", periods=4),
+                    Index(range(0, 12, 3), dtype=np.float64),
+                ],
+                None,
+                TypeError,
+            ],
+            [
+                slice("20130102", "20130104"),
+                [date_range("20130101", periods=4)],
+                1,
+                TypeError,
+            ],
+            [slice(2, 8), [Index([2, 4, "null", 8], dtype=object)], 0, TypeError],
+            [slice(2, 8), [Index([2, 4, "null", 8], dtype=object)], 1, KeyError],
+            [slice(2, 4, 2), [Index([2, 4, "null", 8], dtype=object)], 0, TypeError],
+        ],
+    )
+    def test_loc_getitem_label_slice(self, slc, indexes, axes, fails, frame_or_series):
+        # label slices (with ints)
+
+        # real label slices
+
+        # GH 14316
+        for index in indexes:
+            if index is None:
+                obj = frame_or_series()
+            else:
+                obj = frame_or_series(range(len(index)), index=index)
+            check_indexing_smoketest_or_raises(
+                obj,
+                "loc",
+                slc,
+                axes=axes,
+                fails=fails,
+            )
+
+    def test_setitem_from_duplicate_axis(self):
+        # GH#34034
+        df = DataFrame(
+            [[20, "a"], [200, "a"], [200, "a"]],
+            columns=["col1", "col2"],
+            index=[10, 1, 1],
+        )
+        df.loc[1, "col1"] = np.arange(2)
+        expected = DataFrame(
+            [[20, "a"], [0, "a"], [1, "a"]], columns=["col1", "col2"], index=[10, 1, 1]
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_column_types_consistent(self):
+        # GH 26779
+        df = DataFrame(
+            data={
+                "channel": [1, 2, 3],
+                "A": ["String 1", np.nan, "String 2"],
+                "B": [
+                    Timestamp("2019-06-11 11:00:00"),
+                    pd.NaT,
+                    Timestamp("2019-06-11 12:00:00"),
+                ],
+            }
+        )
+        df2 = DataFrame(
+            data={"A": ["String 3"], "B": [Timestamp("2019-06-11 12:00:00")]}
+        )
+        # Change Columns A and B to df2.values wherever Column A is NaN
+        df.loc[df["A"].isna(), ["A", "B"]] = df2.values
+        expected = DataFrame(
+            data={
+                "channel": [1, 2, 3],
+                "A": ["String 1", "String 3", "String 2"],
+                "B": [
+                    Timestamp("2019-06-11 11:00:00"),
+                    Timestamp("2019-06-11 12:00:00"),
+                    Timestamp("2019-06-11 12:00:00"),
+                ],
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "obj, key, exp",
+        [
+            (
+                DataFrame([[1]], columns=Index([False])),
+                IndexSlice[:, False],
+                Series([1], name=False),
+            ),
+            (Series([1], index=Index([False])), False, [1]),
+            (DataFrame([[1]], index=Index([False])), False, Series([1], name=False)),
+        ],
+    )
+    def test_loc_getitem_single_boolean_arg(self, obj, key, exp):
+        # GH 44322
+        res = obj.loc[key]
+        if isinstance(exp, (DataFrame, Series)):
+            tm.assert_equal(res, exp)
+        else:
+            assert res == exp
+
+
+class TestLocBaseIndependent:
+    # Tests for loc that do not depend on subclassing Base
+    def test_loc_npstr(self):
+        # GH#45580
+        df = DataFrame(index=date_range("2021", "2022"))
+        result = df.loc[np.array(["2021/6/1"])[0] :]
+        expected = df.iloc[151:]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "msg, key",
+        [
+            (r"Period\('2019', 'Y-DEC'\), 'foo', 'bar'", (Period(2019), "foo", "bar")),
+            (r"Period\('2019', 'Y-DEC'\), 'y1', 'bar'", (Period(2019), "y1", "bar")),
+            (r"Period\('2019', 'Y-DEC'\), 'foo', 'z1'", (Period(2019), "foo", "z1")),
+            (
+                r"Period\('2018', 'Y-DEC'\), Period\('2016', 'Y-DEC'\), 'bar'",
+                (Period(2018), Period(2016), "bar"),
+            ),
+            (r"Period\('2018', 'Y-DEC'\), 'foo', 'y1'", (Period(2018), "foo", "y1")),
+            (
+                r"Period\('2017', 'Y-DEC'\), 'foo', Period\('2015', 'Y-DEC'\)",
+                (Period(2017), "foo", Period(2015)),
+            ),
+            (r"Period\('2017', 'Y-DEC'\), 'z1', 'bar'", (Period(2017), "z1", "bar")),
+        ],
+    )
+    def test_contains_raise_error_if_period_index_is_in_multi_index(self, msg, key):
+        # GH#20684
+        """
+        parse_datetime_string_with_reso return parameter if type not matched.
+        PeriodIndex.get_loc takes returned value from parse_datetime_string_with_reso
+        as a tuple.
+        If first argument is Period and a tuple has 3 items,
+        process go on not raise exception
+        """
+        df = DataFrame(
+            {
+                "A": [Period(2019), "x1", "x2"],
+                "B": [Period(2018), Period(2016), "y1"],
+                "C": [Period(2017), "z1", Period(2015)],
+                "V1": [1, 2, 3],
+                "V2": [10, 20, 30],
+            }
+        ).set_index(["A", "B", "C"])
+        with pytest.raises(KeyError, match=msg):
+            df.loc[key]
+
+    def test_loc_getitem_missing_unicode_key(self):
+        df = DataFrame({"a": [1]})
+        with pytest.raises(KeyError, match="\u05d0"):
+            df.loc[:, "\u05d0"]  # should not raise UnicodeEncodeError
+
+    def test_loc_getitem_dups(self):
+        # GH 5678
+        # repeated getitems on a dup index returning a ndarray
+        df = DataFrame(
+            np.random.default_rng(2).random((20, 5)),
+            index=["ABCDE"[x % 5] for x in range(20)],
+        )
+        expected = df.loc["A", 0]
+        result = df.loc[:, 0].loc["A"]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_dups2(self):
+        # GH4726
+        # dup indexing with iloc/loc
+        df = DataFrame(
+            [[1, 2, "foo", "bar", Timestamp("20130101")]],
+            columns=["a", "a", "a", "a", "a"],
+            index=[1],
+        )
+        expected = Series(
+            [1, 2, "foo", "bar", Timestamp("20130101")],
+            index=["a", "a", "a", "a", "a"],
+            name=1,
+        )
+
+        result = df.iloc[0]
+        tm.assert_series_equal(result, expected)
+
+        result = df.loc[1]
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_setitem_dups(self):
+        # GH 6541
+        df_orig = DataFrame(
+            {
+                "me": list("rttti"),
+                "foo": list("aaade"),
+                "bar": np.arange(5, dtype="float64") * 1.34 + 2,
+                "bar2": np.arange(5, dtype="float64") * -0.34 + 2,
+            }
+        ).set_index("me")
+
+        indexer = (
+            "r",
+            ["bar", "bar2"],
+        )
+        df = df_orig.copy()
+        df.loc[indexer] *= 2.0
+        tm.assert_series_equal(df.loc[indexer], 2.0 * df_orig.loc[indexer])
+
+        indexer = (
+            "r",
+            "bar",
+        )
+        df = df_orig.copy()
+        df.loc[indexer] *= 2.0
+        assert df.loc[indexer] == 2.0 * df_orig.loc[indexer]
+
+        indexer = (
+            "t",
+            ["bar", "bar2"],
+        )
+        df = df_orig.copy()
+        df.loc[indexer] *= 2.0
+        tm.assert_frame_equal(df.loc[indexer], 2.0 * df_orig.loc[indexer])
+
+    def test_loc_setitem_slice(self):
+        # GH10503
+
+        # assigning the same type should not change the type
+        df1 = DataFrame({"a": [0, 1, 1], "b": Series([100, 200, 300], dtype="uint32")})
+        ix = df1["a"] == 1
+        newb1 = df1.loc[ix, "b"] + 1
+        df1.loc[ix, "b"] = newb1
+        expected = DataFrame(
+            {"a": [0, 1, 1], "b": Series([100, 201, 301], dtype="uint32")}
+        )
+        tm.assert_frame_equal(df1, expected)
+
+        # assigning a new type should get the inferred type
+        df2 = DataFrame({"a": [0, 1, 1], "b": [100, 200, 300]}, dtype="uint64")
+        ix = df1["a"] == 1
+        newb2 = df2.loc[ix, "b"]
+        with pytest.raises(TypeError, match="Invalid value"):
+            df1.loc[ix, "b"] = newb2
+
+    def test_loc_setitem_dtype(self):
+        # GH31340
+        df = DataFrame({"id": ["A"], "a": [1.2], "b": [0.0], "c": [-2.5]})
+        cols = ["a", "b", "c"]
+        df.loc[:, cols] = df.loc[:, cols].astype("float32")
+
+        # pre-2.0 this setting would swap in new arrays, in 2.0 it is correctly
+        #  in-place, consistent with non-split-path
+        expected = DataFrame(
+            {
+                "id": ["A"],
+                "a": np.array([1.2], dtype="float64"),
+                "b": np.array([0.0], dtype="float64"),
+                "c": np.array([-2.5], dtype="float64"),
+            }
+        )  # id is inferred as object
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_getitem_label_list_with_missing(self):
+        s = Series(range(3), index=["a", "b", "c"])
+
+        # consistency
+        with pytest.raises(KeyError, match="not in index"):
+            s[["a", "d"]]
+
+        s = Series(range(3))
+        with pytest.raises(KeyError, match="not in index"):
+            s[[0, 3]]
+
+    @pytest.mark.parametrize("index", [[True, False], [True, False, True, False]])
+    def test_loc_getitem_bool_diff_len(self, index):
+        # GH26658
+        s = Series([1, 2, 3])
+        msg = f"Boolean index has wrong length: {len(index)} instead of {len(s)}"
+        with pytest.raises(IndexError, match=msg):
+            s.loc[index]
+
+    def test_loc_getitem_int_slice(self):
+        # TODO: test something here?
+        pass
+
+    def test_loc_to_fail(self):
+        # GH3449
+        df = DataFrame(
+            np.random.default_rng(2).random((3, 3)),
+            index=["a", "b", "c"],
+            columns=["e", "f", "g"],
+        )
+
+        msg = (
+            rf"\"None of \[Index\(\[1, 2\], dtype='{np.dtype(int)}'\)\] are "
+            r"in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            df.loc[[1, 2], [1, 2]]
+
+    def test_loc_to_fail2(self):
+        # GH  7496
+        # loc should not fallback
+
+        s = Series(dtype=object)
+        s.loc[1] = 1
+        s.loc["a"] = 2
+
+        with pytest.raises(KeyError, match=r"^-1$"):
+            s.loc[-1]
+
+        msg = (
+            rf"\"None of \[Index\(\[-1, -2\], dtype='{np.dtype(int)}'\)\] are "
+            r"in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            s.loc[[-1, -2]]
+
+        msg = r"\"None of \[Index\(\['4'\], dtype='object'\)\] are in the \[index\]\""
+        with pytest.raises(KeyError, match=msg):
+            s.loc[Index(["4"], dtype=object)]
+
+        s.loc[-1] = 3
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[-1, -2]]
+
+        s["a"] = 2
+        msg = (
+            rf"\"None of \[Index\(\[-2\], dtype='{np.dtype(int)}'\)\] are "
+            r"in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            s.loc[[-2]]
+
+        del s["a"]
+
+        with pytest.raises(KeyError, match=msg):
+            s.loc[[-2]] = 0
+
+    def test_loc_to_fail3(self):
+        # inconsistency between .loc[values] and .loc[values,:]
+        # GH 7999
+        df = DataFrame([["a"], ["b"]], index=[1, 2], columns=["value"])
+
+        msg = (
+            rf"\"None of \[Index\(\[3\], dtype='{np.dtype(int)}'\)\] are "
+            r"in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            df.loc[[3], :]
+
+        with pytest.raises(KeyError, match=msg):
+            df.loc[[3]]
+
+    def test_loc_getitem_list_with_fail(self):
+        # 15747
+        # should KeyError if *any* missing labels
+
+        s = Series([1, 2, 3])
+
+        s.loc[[2]]
+
+        msg = "None of [RangeIndex(start=3, stop=4, step=1)] are in the [index]"
+        with pytest.raises(KeyError, match=re.escape(msg)):
+            s.loc[[3]]
+
+        # a non-match and a match
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[2, 3]]
+
+    def test_loc_index(self):
+        # gh-17131
+        # a boolean index should index like a boolean numpy array
+
+        df = DataFrame(
+            np.random.default_rng(2).random(size=(5, 10)),
+            index=["alpha_0", "alpha_1", "alpha_2", "beta_0", "beta_1"],
+        )
+
+        mask = df.index.map(lambda x: "alpha" in x)
+        expected = df.loc[np.array(mask)]
+
+        result = df.loc[mask]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[mask.values]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[pd.array(mask, dtype="boolean")]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_general(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((4, 4)),
+            columns=["A", "B", "C", "D"],
+            index=["A", "B", "C", "D"],
+        )
+
+        # want this to work
+        result = df.loc[:, "A":"B"].iloc[0:2, :]
+        assert (result.columns == ["A", "B"]).all()
+        assert (result.index == ["A", "B"]).all()
+
+        # mixed type
+        result = DataFrame({"a": [Timestamp("20130101")], "b": [1]}).iloc[0]
+        expected = Series([Timestamp("20130101"), 1], index=["a", "b"], name=0)
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == object
+
+    @pytest.fixture
+    def frame_for_consistency(self):
+        return DataFrame(
+            {
+                "date": date_range("2000-01-01", "2000-01-5"),
+                "val": Series(range(5), dtype=np.int64),
+            }
+        )
+
+    @pytest.mark.parametrize(
+        "val",
+        [0, np.array(0, dtype=np.int64), np.array([0, 0, 0, 0, 0], dtype=np.int64)],
+    )
+    def test_loc_setitem_consistency(self, frame_for_consistency, val):
+        # GH 6149
+        # coerce similarly for setitem and loc when rows have a null-slice
+        df = frame_for_consistency.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, "date"] = val
+
+    def test_loc_setitem_consistency_dt64_to_str(self, frame_for_consistency):
+        # GH 6149
+        # coerce similarly for setitem and loc when rows have a null-slice
+
+        df = frame_for_consistency.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, "date"] = "foo"
+
+    def test_loc_setitem_consistency_dt64_to_float(self, frame_for_consistency):
+        # GH 6149
+        # coerce similarly for setitem and loc when rows have a null-slice
+        df = frame_for_consistency.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, "date"] = 1.0
+
+    def test_loc_setitem_consistency_single_row(self):
+        # GH 15494
+        # setting on frame with single row
+        df = DataFrame({"date": Series([Timestamp("20180101")])})
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, "date"] = "string"
+
+    def test_loc_setitem_consistency_empty(self):
+        # empty (essentially noops)
+        # before the enforcement of #45333 in 2.0, the loc.setitem here would
+        #  change the dtype of df.x to int64
+        expected = DataFrame(columns=["x", "y"])
+        df = DataFrame(columns=["x", "y"])
+        with tm.assert_produces_warning(None):
+            df.loc[:, "x"] = 1
+        tm.assert_frame_equal(df, expected)
+
+        # setting with setitem swaps in a new array, so changes the dtype
+        df = DataFrame(columns=["x", "y"])
+        df["x"] = 1
+        expected["x"] = expected["x"].astype(np.int64)
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_consistency_slice_column_len(self, using_infer_string):
+        # .loc[:,column] setting with slice == len of the column
+        # GH10408
+        levels = [
+            ["Region_1"] * 4,
+            ["Site_1", "Site_1", "Site_2", "Site_2"],
+            [3987227376, 3980680971, 3977723249, 3977723089],
+        ]
+        mi = MultiIndex.from_arrays(levels, names=["Region", "Site", "RespondentID"])
+
+        clevels = [
+            ["Respondent", "Respondent", "Respondent", "OtherCat", "OtherCat"],
+            ["Something", "StartDate", "EndDate", "Yes/No", "SomethingElse"],
+        ]
+        cols = MultiIndex.from_arrays(clevels, names=["Level_0", "Level_1"])
+
+        values = [
+            ["A", "5/25/2015 10:59", "5/25/2015 11:22", "Yes", np.nan],
+            ["A", "5/21/2015 9:40", "5/21/2015 9:52", "Yes", "Yes"],
+            ["A", "5/20/2015 8:27", "5/20/2015 8:41", "Yes", np.nan],
+            ["A", "5/20/2015 8:33", "5/20/2015 9:09", "Yes", "No"],
+        ]
+        df = DataFrame(values, index=mi, columns=cols)
+
+        ctx = contextlib.nullcontext()
+        if using_infer_string:
+            ctx = pytest.raises(TypeError, match="Invalid value")
+
+        with ctx:
+            df.loc[:, ("Respondent", "StartDate")] = to_datetime(
+                df.loc[:, ("Respondent", "StartDate")]
+            )
+        with ctx:
+            df.loc[:, ("Respondent", "EndDate")] = to_datetime(
+                df.loc[:, ("Respondent", "EndDate")]
+            )
+
+        if using_infer_string:
+            # infer-objects won't infer stuff anymore
+            return
+
+        df = df.infer_objects()
+
+        # Adding a new key
+        df.loc[:, ("Respondent", "Duration")] = (
+            df.loc[:, ("Respondent", "EndDate")]
+            - df.loc[:, ("Respondent", "StartDate")]
+        )
+
+        # timedelta64[m] -> float, so this cannot be done inplace, so
+        #  no warning
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[:, ("Respondent", "Duration")] = df.loc[
+                :, ("Respondent", "Duration")
+            ] / Timedelta(60_000_000_000)
+
+    @pytest.mark.parametrize("unit", ["Y", "M", "D", "h", "m", "s", "ms", "us"])
+    def test_loc_assign_non_ns_datetime(self, unit):
+        # GH 27395, non-ns dtype assignment via .loc should work
+        # and return the same result when using simple assignment
+        df = DataFrame(
+            {
+                "timestamp": [
+                    np.datetime64("2017-02-11 12:41:29"),
+                    np.datetime64("1991-11-07 04:22:37"),
+                ]
+            }
+        )
+
+        df.loc[:, unit] = df.loc[:, "timestamp"].values.astype(f"datetime64[{unit}]")
+        df["expected"] = df.loc[:, "timestamp"].values.astype(f"datetime64[{unit}]")
+        expected = Series(df.loc[:, "expected"], name=unit)
+        tm.assert_series_equal(df.loc[:, unit], expected)
+
+    def test_loc_modify_datetime(self):
+        # see gh-28837
+        df = DataFrame.from_dict(
+            {"date": [1485264372711, 1485265925110, 1540215845888, 1540282121025]}
+        )
+
+        df["date_dt"] = to_datetime(df["date"], unit="ms", cache=True).dt.as_unit("ms")
+
+        df.loc[:, "date_dt_cp"] = df.loc[:, "date_dt"]
+        df.loc[[2, 3], "date_dt_cp"] = df.loc[[2, 3], "date_dt"]
+
+        expected = DataFrame(
+            [
+                [1485264372711, "2017-01-24 13:26:12.711", "2017-01-24 13:26:12.711"],
+                [1485265925110, "2017-01-24 13:52:05.110", "2017-01-24 13:52:05.110"],
+                [1540215845888, "2018-10-22 13:44:05.888", "2018-10-22 13:44:05.888"],
+                [1540282121025, "2018-10-23 08:08:41.025", "2018-10-23 08:08:41.025"],
+            ],
+            columns=["date", "date_dt", "date_dt_cp"],
+        )
+
+        columns = ["date_dt", "date_dt_cp"]
+        expected[columns] = expected[columns].apply(to_datetime)
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame_with_reindex(self):
+        # GH#6254 setting issue
+        df = DataFrame(index=[3, 5, 4], columns=["A"], dtype=float)
+        df.loc[[4, 3, 5], "A"] = np.array([1, 2, 3], dtype="int64")
+
+        # setting integer values into a float dataframe with loc is inplace,
+        #  so we retain float dtype
+        ser = Series([2, 3, 1], index=[3, 5, 4], dtype=float)
+        expected = DataFrame({"A": ser})
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame_with_reindex_mixed(self):
+        # GH#40480
+        df = DataFrame(index=[3, 5, 4], columns=["A", "B"], dtype=float)
+        df["B"] = "string"
+        df.loc[[4, 3, 5], "A"] = np.array([1, 2, 3], dtype="int64")
+        ser = Series([2, 3, 1], index=[3, 5, 4], dtype="int64")
+        # pre-2.0 this setting swapped in a new array, now it is inplace
+        #  consistent with non-split-path
+        expected = DataFrame({"A": ser.astype(float)})
+        expected["B"] = "string"
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame_with_inverted_slice(self):
+        # GH#40480
+        df = DataFrame(index=[1, 2, 3], columns=["A", "B"], dtype=float)
+        df["B"] = "string"
+        df.loc[slice(3, 0, -1), "A"] = np.array([1, 2, 3], dtype="int64")
+        # pre-2.0 this setting swapped in a new array, now it is inplace
+        #  consistent with non-split-path
+        expected = DataFrame({"A": [3.0, 2.0, 1.0], "B": "string"}, index=[1, 2, 3])
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_empty_frame(self):
+        # GH#6252 setting with an empty frame
+        keys1 = ["@" + str(i) for i in range(5)]
+        val1 = np.arange(5, dtype="int64")
+
+        keys2 = ["@" + str(i) for i in range(4)]
+        val2 = np.arange(4, dtype="int64")
+
+        index = list(set(keys1).union(keys2))
+        df = DataFrame(index=index)
+        df["A"] = np.nan
+        df.loc[keys1, "A"] = val1
+
+        df["B"] = np.nan
+        df.loc[keys2, "B"] = val2
+
+        # Because df["A"] was initialized as float64, setting values into it
+        #  is inplace, so that dtype is retained
+        sera = Series(val1, index=keys1, dtype=np.float64)
+        serb = Series(val2, index=keys2)
+        expected = DataFrame({"A": sera, "B": serb}, columns=Index(["A", "B"])).reindex(
+            index=index
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=list("abcd"),
+            columns=list("ABCD"),
+        )
+
+        result = df.iloc[0, 0]
+
+        df.loc["a", "A"] = 1
+        result = df.loc["a", "A"]
+        assert result == 1
+
+        result = df.iloc[0, 0]
+        assert result == 1
+
+        df.loc[:, "B":"D"] = 0
+        expected = df.loc[:, "B":"D"]
+        result = df.iloc[:, 1:]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_setitem_frame_nan_int_coercion_invalid(self):
+        # GH 8669
+        # invalid coercion of nan -> int
+        df = DataFrame({"A": [1, 2, 3], "B": np.nan})
+        df.loc[df.B > df.A, "B"] = df.A
+        expected = DataFrame({"A": [1, 2, 3], "B": np.nan})
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame_mixed_labels(self):
+        # GH 6546
+        # setting with mixed labels
+        df = DataFrame({1: [1, 2], 2: [3, 4], "a": ["a", "b"]})
+
+        result = df.loc[0, [1, 2]]
+        expected = Series(
+            [1, 3], index=Index([1, 2], dtype=object), dtype=object, name=0
+        )
+        tm.assert_series_equal(result, expected)
+
+        expected = DataFrame({1: [5, 2], 2: [6, 4], "a": ["a", "b"]})
+        df.loc[0, [1, 2]] = [5, 6]
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_frame_multiples(self):
+        # multiple setting
+        df = DataFrame(
+            {"A": ["foo", "bar", "baz"], "B": Series(range(3), dtype=np.int64)}
+        )
+        rhs = df.loc[1:2]
+        rhs.index = df.index[0:2]
+        df.loc[0:1] = rhs
+        expected = DataFrame(
+            {"A": ["bar", "baz", "baz"], "B": Series([1, 2, 2], dtype=np.int64)}
+        )
+        tm.assert_frame_equal(df, expected)
+
+        # multiple setting with frame on rhs (with M8)
+        df = DataFrame(
+            {
+                "date": date_range("2000-01-01", "2000-01-5"),
+                "val": Series(range(5), dtype=np.int64),
+            }
+        )
+        expected = DataFrame(
+            {
+                "date": [
+                    Timestamp("20000101"),
+                    Timestamp("20000102"),
+                    Timestamp("20000101"),
+                    Timestamp("20000102"),
+                    Timestamp("20000103"),
+                ],
+                "val": Series([0, 1, 0, 1, 2], dtype=np.int64),
+            }
+        )
+        expected["date"] = expected["date"].astype("M8[ns]")
+        rhs = df.loc[0:2]
+        rhs.index = df.index[2:5]
+        df.loc[2:4] = rhs
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "indexer", [["A"], slice(None, "A", None), np.array(["A"])]
+    )
+    @pytest.mark.parametrize("value", [["Z"], np.array(["Z"])])
+    def test_loc_setitem_with_scalar_index(self, indexer, value):
+        # GH #19474
+        # assigning like "df.loc[0, ['A']] = ['Z']" should be evaluated
+        # elementwisely, not using "setter('A', ['Z'])".
+
+        # Set object dtype to avoid upcast when setting 'Z'
+        df = DataFrame([[1, 2], [3, 4]], columns=["A", "B"]).astype({"A": object})
+        df.loc[0, indexer] = value
+        result = df.loc[0, "A"]
+
+        assert is_scalar(result) and result == "Z"
+
+    @pytest.mark.parametrize(
+        "index,box,expected",
+        [
+            (
+                ([0, 2], ["A", "B", "C", "D"]),
+                7,
+                DataFrame(
+                    [[7, 7, 7, 7], [3, 4, np.nan, np.nan], [7, 7, 7, 7]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                (1, ["C", "D"]),
+                [7, 8],
+                DataFrame(
+                    [[1, 2, np.nan, np.nan], [3, 4, 7, 8], [5, 6, np.nan, np.nan]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                (1, ["A", "B", "C"]),
+                np.array([7, 8, 9], dtype=np.int64),
+                DataFrame(
+                    [[1, 2, np.nan], [7, 8, 9], [5, 6, np.nan]], columns=["A", "B", "C"]
+                ),
+            ),
+            (
+                (slice(1, 3, None), ["B", "C", "D"]),
+                [[7, 8, 9], [10, 11, 12]],
+                DataFrame(
+                    [[1, 2, np.nan, np.nan], [3, 7, 8, 9], [5, 10, 11, 12]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                (slice(1, 3, None), ["C", "A", "D"]),
+                np.array([[7, 8, 9], [10, 11, 12]], dtype=np.int64),
+                DataFrame(
+                    [[1, 2, np.nan, np.nan], [8, 4, 7, 9], [11, 6, 10, 12]],
+                    columns=["A", "B", "C", "D"],
+                ),
+            ),
+            (
+                (slice(None, None, None), ["A", "C"]),
+                DataFrame([[7, 8], [9, 10], [11, 12]], columns=["A", "C"]),
+                DataFrame(
+                    [[7, 2, 8], [9, 4, 10], [11, 6, 12]], columns=["A", "B", "C"]
+                ),
+            ),
+        ],
+    )
+    def test_loc_setitem_missing_columns(self, index, box, expected):
+        # GH 29334
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], columns=["A", "B"])
+
+        df.loc[index] = box
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_coercion(self):
+        # GH#12411
+        df = DataFrame({"date": [Timestamp("20130101").tz_localize("UTC"), pd.NaT]})
+        expected = df.dtypes
+
+        result = df.iloc[[0]]
+        tm.assert_series_equal(result.dtypes, expected)
+
+        result = df.iloc[[1]]
+        tm.assert_series_equal(result.dtypes, expected)
+
+    def test_loc_coercion2(self):
+        # GH#12045
+        df = DataFrame({"date": [datetime(2012, 1, 1), datetime(1012, 1, 2)]})
+        expected = df.dtypes
+
+        result = df.iloc[[0]]
+        tm.assert_series_equal(result.dtypes, expected)
+
+        result = df.iloc[[1]]
+        tm.assert_series_equal(result.dtypes, expected)
+
+    def test_loc_coercion3(self):
+        # GH#11594
+        df = DataFrame({"text": ["some words"] + [None] * 9})
+        expected = df.dtypes
+
+        result = df.iloc[0:2]
+        tm.assert_series_equal(result.dtypes, expected)
+
+        result = df.iloc[3:]
+        tm.assert_series_equal(result.dtypes, expected)
+
+    def test_setitem_new_key_tz(self, indexer_sl):
+        # GH#12862 should not raise on assigning the second value
+        vals = [
+            to_datetime(42).tz_localize("UTC"),
+            to_datetime(666).tz_localize("UTC"),
+        ]
+        expected = Series(vals, index=Index(["foo", "bar"]))
+
+        ser = Series(dtype=object)
+        indexer_sl(ser)["foo"] = vals[0]
+        indexer_sl(ser)["bar"] = vals[1]
+
+        tm.assert_series_equal(ser, expected)
+
+    def test_loc_non_unique(self):
+        # GH3659
+        # non-unique indexer with loc slice
+        # https://groups.google.com/forum/?fromgroups#!topic/pydata/zTm2No0crYs
+
+        # these are going to raise because the we are non monotonic
+        df = DataFrame(
+            {"A": [1, 2, 3, 4, 5, 6], "B": [3, 4, 5, 6, 7, 8]}, index=[0, 1, 0, 1, 2, 3]
+        )
+        msg = "'Cannot get left slice bound for non-unique label: 1'"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[1:]
+        msg = "'Cannot get left slice bound for non-unique label: 0'"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[0:]
+        msg = "'Cannot get left slice bound for non-unique label: 1'"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[1:2]
+
+        # monotonic are ok
+        df = DataFrame(
+            {"A": [1, 2, 3, 4, 5, 6], "B": [3, 4, 5, 6, 7, 8]}, index=[0, 1, 0, 1, 2, 3]
+        ).sort_index(axis=0)
+        result = df.loc[1:]
+        expected = DataFrame({"A": [2, 4, 5, 6], "B": [4, 6, 7, 8]}, index=[1, 1, 2, 3])
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[0:]
+        tm.assert_frame_equal(result, df)
+
+        result = df.loc[1:2]
+        expected = DataFrame({"A": [2, 4, 5], "B": [4, 6, 7]}, index=[1, 1, 2])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.arm_slow
+    @pytest.mark.parametrize("length, l2", [[900, 100], [900000, 100000]])
+    def test_loc_non_unique_memory_error(self, length, l2):
+        # GH 4280
+        # non_unique index with a large selection triggers a memory error
+
+        columns = list("ABCDEFG")
+
+        df = pd.concat(
+            [
+                DataFrame(
+                    np.random.default_rng(2).standard_normal((length, len(columns))),
+                    index=np.arange(length),
+                    columns=columns,
+                ),
+                DataFrame(np.ones((l2, len(columns))), index=[0] * l2, columns=columns),
+            ]
+        )
+
+        assert df.index.is_unique is False
+
+        mask = np.arange(l2)
+        result = df.loc[mask]
+        expected = pd.concat(
+            [
+                df.take([0]),
+                DataFrame(
+                    np.ones((len(mask), len(columns))),
+                    index=[0] * len(mask),
+                    columns=columns,
+                ),
+                df.take(mask[1:]),
+            ]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_name(self):
+        # GH 3880
+        df = DataFrame([[1, 1], [1, 1]])
+        df.index.name = "index_name"
+        result = df.iloc[[0, 1]].index.name
+        assert result == "index_name"
+
+        result = df.loc[[0, 1]].index.name
+        assert result == "index_name"
+
+    def test_loc_empty_list_indexer_is_ok(self):
+        df = DataFrame(
+            np.ones((5, 2)),
+            index=Index([f"i-{i}" for i in range(5)], name="a"),
+            columns=Index([f"i-{i}" for i in range(2)], name="a"),
+        )
+        # vertical empty
+        tm.assert_frame_equal(
+            df.loc[:, []], df.iloc[:, :0], check_index_type=True, check_column_type=True
+        )
+        # horizontal empty
+        tm.assert_frame_equal(
+            df.loc[[], :], df.iloc[:0, :], check_index_type=True, check_column_type=True
+        )
+        # horizontal empty
+        tm.assert_frame_equal(
+            df.loc[[]], df.iloc[:0, :], check_index_type=True, check_column_type=True
+        )
+
+    def test_identity_slice_returns_new_object(self):
+        # GH13873
+
+        original_df = DataFrame({"a": [1, 2, 3]})
+        sliced_df = original_df.loc[:]
+        assert sliced_df is not original_df
+        assert original_df[:] is not original_df
+        assert original_df.loc[:, :] is not original_df
+
+        # should be a shallow copy
+        assert np.shares_memory(original_df["a"]._values, sliced_df["a"]._values)
+
+        # Setting using .loc[:, "a"] sets inplace so alters both sliced and orig
+        # depending on CoW
+        original_df.loc[:, "a"] = [4, 4, 4]
+        assert (sliced_df["a"] == [1, 2, 3]).all()
+
+        # These should not return copies
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        assert df[0] is not df.loc[:, 0]
+
+        # Same tests for Series
+        original_series = Series([1, 2, 3, 4, 5, 6])
+        sliced_series = original_series.loc[:]
+        assert sliced_series is not original_series
+        assert original_series[:] is not original_series
+
+        original_series[:3] = [7, 8, 9]
+        assert all(sliced_series[:3] == [1, 2, 3])
+
+    def test_loc_copy_vs_view(self, request):
+        # GH 15631
+        x = DataFrame(zip(range(3), range(3)), columns=["a", "b"])
+
+        y = x.copy()
+        q = y.loc[:, "a"]
+        q += 2
+
+        tm.assert_frame_equal(x, y)
+
+        z = x.copy()
+        q = z.loc[x.index, "a"]
+        q += 2
+
+        tm.assert_frame_equal(x, z)
+
+    def test_loc_uint64(self):
+        # GH20722
+        # Test whether loc accept uint64 max value as index.
+        umax = np.iinfo("uint64").max
+        ser = Series([1, 2], index=[umax - 1, umax])
+
+        result = ser.loc[umax - 1]
+        expected = ser.iloc[0]
+        assert result == expected
+
+        result = ser.loc[[umax - 1]]
+        expected = ser.iloc[[0]]
+        tm.assert_series_equal(result, expected)
+
+        result = ser.loc[[umax - 1, umax]]
+        tm.assert_series_equal(result, ser)
+
+    def test_loc_uint64_disallow_negative(self):
+        # GH#41775
+        umax = np.iinfo("uint64").max
+        ser = Series([1, 2], index=[umax - 1, umax])
+
+        with pytest.raises(KeyError, match="-1"):
+            # don't wrap around
+            ser.loc[-1]
+
+        with pytest.raises(KeyError, match="-1"):
+            # don't wrap around
+            ser.loc[[-1]]
+
+    def test_loc_setitem_empty_append_expands_rows(self):
+        # GH6173, various appends to an empty dataframe
+
+        data = [1, 2, 3]
+        expected = DataFrame(
+            {"x": data, "y": np.array([np.nan] * len(data), dtype=object)}
+        )
+
+        # appends to fit length of data
+        df = DataFrame(columns=["x", "y"])
+        df.loc[:, "x"] = data
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_empty_append_expands_rows_mixed_dtype(self):
+        # GH#37932 same as test_loc_setitem_empty_append_expands_rows
+        #  but with mixed dtype so we go through take_split_path
+        data = [1, 2, 3]
+        expected = DataFrame(
+            {"x": data, "y": np.array([np.nan] * len(data), dtype=object)}
+        )
+
+        df = DataFrame(columns=["x", "y"])
+        df["x"] = df["x"].astype(np.int64)
+        df.loc[:, "x"] = data
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_empty_append_single_value(self):
+        # only appends one value
+        expected = DataFrame({"x": [1.0], "y": [np.nan]})
+        df = DataFrame(columns=["x", "y"], dtype=float)
+        df.loc[0, "x"] = expected.loc[0, "x"]
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_empty_append_raises(self):
+        # GH6173, various appends to an empty dataframe
+
+        data = [1, 2]
+        df = DataFrame(columns=["x", "y"])
+        df.index = df.index.astype(np.int64)
+        msg = r"None of .*Index.* are in the \[index\]"
+        with pytest.raises(KeyError, match=msg):
+            df.loc[[0, 1], "x"] = data
+
+        msg = "setting an array element with a sequence."
+        with pytest.raises(ValueError, match=msg):
+            df.loc[0:2, "x"] = data
+
+    def test_indexing_zerodim_np_array(self):
+        # GH24924
+        df = DataFrame([[1, 2], [3, 4]])
+        result = df.loc[np.array(0)]
+        s = Series([1, 2], name=0)
+        tm.assert_series_equal(result, s)
+
+    def test_series_indexing_zerodim_np_array(self):
+        # GH24924
+        s = Series([1, 2])
+        result = s.loc[np.array(0)]
+        assert result == 1
+
+    def test_loc_reverse_assignment(self):
+        # GH26939
+        data = [1, 2, 3, 4, 5, 6] + [None] * 4
+        expected = Series(data, index=range(2010, 2020))
+
+        result = Series(index=range(2010, 2020), dtype=np.float64)
+        result.loc[2015:2010:-1] = [6, 5, 4, 3, 2, 1]
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_setitem_str_to_small_float_conversion_type(self, using_infer_string):
+        # GH#20388
+
+        col_data = [str(np.random.default_rng(2).random() * 1e-12) for _ in range(5)]
+        result = DataFrame(col_data, columns=["A"])
+        expected = DataFrame(col_data, columns=["A"])
+        tm.assert_frame_equal(result, expected)
+
+        # assigning with loc/iloc attempts to set the values inplace, which
+        #  in this case is successful
+        if using_infer_string:
+            with pytest.raises(TypeError, match="Invalid value"):
+                result.loc[result.index, "A"] = [float(x) for x in col_data]
+        else:
+            result.loc[result.index, "A"] = [float(x) for x in col_data]
+            expected = DataFrame(col_data, columns=["A"], dtype=float).astype(object)
+            tm.assert_frame_equal(result, expected)
+
+        # assigning the entire column using __setitem__ swaps in the new array
+        # GH#???
+        result["A"] = [float(x) for x in col_data]
+        expected = DataFrame(col_data, columns=["A"], dtype=float)
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_time_object(self, frame_or_series):
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        mask = (rng.hour == 9) & (rng.minute == 30)
+
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 3)), index=rng
+        )
+        obj = tm.get_obj(obj, frame_or_series)
+
+        result = obj.loc[time(9, 30)]
+        exp = obj.loc[mask]
+        tm.assert_equal(result, exp)
+
+        chunk = obj.loc["1/4/2000":]
+        result = chunk.loc[time(9, 30)]
+        expected = result[-1:]
+
+        # Without resetting the freqs, these are 5 min and 1440 min, respectively
+        result.index = result.index._with_freq(None)
+        expected.index = expected.index._with_freq(None)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("spmatrix_t", ["coo_matrix", "csc_matrix", "csr_matrix"])
+    @pytest.mark.parametrize("dtype", [np.complex128, np.float64, np.int64, bool])
+    def test_loc_getitem_range_from_spmatrix(self, spmatrix_t, dtype):
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        spmatrix_t = getattr(sp_sparse, spmatrix_t)
+
+        # The bug is triggered by a sparse matrix with purely sparse columns.  So the
+        # recipe below generates a rectangular matrix of dimension (5, 7) where all the
+        # diagonal cells are ones, meaning the last two columns are purely sparse.
+        rows, cols = 5, 7
+        spmatrix = spmatrix_t(np.eye(rows, cols, dtype=dtype), dtype=dtype)
+        df = DataFrame.sparse.from_spmatrix(spmatrix)
+
+        # regression test for GH#34526
+        itr_idx = range(2, rows)
+        result = np.nan_to_num(df.loc[itr_idx].values)
+        expected = spmatrix.toarray()[itr_idx]
+        tm.assert_numpy_array_equal(result, expected)
+
+        # regression test for GH#34540
+        result = df.loc[itr_idx].dtypes.values
+        expected = np.full(cols, SparseDtype(dtype))
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_loc_getitem_listlike_all_retains_sparse(self):
+        df = DataFrame({"A": pd.array([0, 0], dtype=SparseDtype("int64"))})
+        result = df.loc[[0, 1]]
+        tm.assert_frame_equal(result, df)
+
+    def test_loc_getitem_sparse_frame(self):
+        # GH34687
+        sp_sparse = pytest.importorskip("scipy.sparse")
+
+        df = DataFrame.sparse.from_spmatrix(sp_sparse.eye(5, dtype=np.int64))
+        result = df.loc[range(2)]
+        expected = DataFrame(
+            [[1, 0, 0, 0, 0], [0, 1, 0, 0, 0]],
+            dtype=SparseDtype(np.int64),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.loc[range(2)].loc[range(1)]
+        expected = DataFrame([[1, 0, 0, 0, 0]], dtype=SparseDtype(np.int64))
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_sparse_series(self):
+        # GH34687
+        s = Series([1.0, 0.0, 0.0, 0.0, 0.0], dtype=SparseDtype("float64", 0.0))
+
+        result = s.loc[range(2)]
+        expected = Series([1.0, 0.0], dtype=SparseDtype("float64", 0.0))
+        tm.assert_series_equal(result, expected)
+
+        result = s.loc[range(3)].loc[range(2)]
+        expected = Series([1.0, 0.0], dtype=SparseDtype("float64", 0.0))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("indexer", ["loc", "iloc"])
+    def test_getitem_single_row_sparse_df(self, indexer):
+        # GH#46406
+        df = DataFrame([[1.0, 0.0, 1.5], [0.0, 2.0, 0.0]], dtype=SparseDtype(float))
+        result = getattr(df, indexer)[0]
+        expected = Series([1.0, 0.0, 1.5], dtype=SparseDtype(float), name=0)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("key_type", [iter, np.array, Series, Index])
+    def test_loc_getitem_iterable(self, float_frame, key_type):
+        idx = key_type(["A", "B", "C"])
+        result = float_frame.loc[:, idx]
+        expected = float_frame.loc[:, ["A", "B", "C"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_timedelta_0seconds(self):
+        # GH#10583
+        df = DataFrame(np.random.default_rng(2).normal(size=(10, 4)))
+        df.index = timedelta_range(start="0s", periods=10, freq="s")
+        expected = df.loc[Timedelta("0s") :, :]
+        result = df.loc["0s":, :]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("val,expected", [(2**63 - 1, 1), (2**63, 2)])
+    def test_loc_getitem_uint64_scalar(self, val, expected):
+        # see GH#19399
+        df = DataFrame([1, 2], index=[2**63 - 1, 2**63])
+        result = df.loc[val]
+        expected = Series([expected])
+        expected.name = val
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_setitem_int_label_with_float_index(self, float_numpy_dtype):
+        # note labels are floats
+        dtype = float_numpy_dtype
+        ser = Series(["a", "b", "c"], index=Index([0, 0.5, 1], dtype=dtype))
+        expected = ser.copy()
+
+        ser.loc[1] = "zoo"
+        expected.iloc[2] = "zoo"
+
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize(
+        "indexer, expected",
+        [
+            # The test name is a misnomer in the 0 case as df.index[indexer]
+            #  is a scalar.
+            (0, [20, 1, 2, 3, 4, 5, 6, 7, 8, 9]),
+            (slice(4, 8), [0, 1, 2, 3, 20, 20, 20, 20, 8, 9]),
+            ([3, 5], [0, 1, 2, 20, 4, 20, 6, 7, 8, 9]),
+        ],
+    )
+    def test_loc_setitem_listlike_with_timedelta64index(self, indexer, expected):
+        # GH#16637
+        tdi = to_timedelta(range(10), unit="s")
+        df = DataFrame({"x": range(10)}, dtype="int64", index=tdi)
+
+        df.loc[df.index[indexer], "x"] = 20
+
+        expected = DataFrame(
+            expected,
+            index=tdi,
+            columns=["x"],
+            dtype="int64",
+        )
+
+        tm.assert_frame_equal(expected, df)
+
+    def test_loc_setitem_categorical_values_partial_column_slice(self):
+        # Assigning a Category to parts of a int/... column uses the values of
+        # the Categorical
+        df = DataFrame({"a": [1, 1, 1, 1, 1], "b": list("aaaaa")})
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[1:2, "a"] = Categorical(["b", "b"], categories=["a", "b"])
+            df.loc[2:3, "b"] = Categorical(["b", "b"], categories=["a", "b"])
+
+    def test_loc_setitem_single_row_categorical(self, using_infer_string):
+        # GH#25495
+        df = DataFrame({"Alpha": ["a"], "Numeric": [0]})
+        categories = Categorical(df["Alpha"], categories=["a", "b", "c"])
+
+        # pre-2.0 this swapped in a new array, in 2.0 it operates inplace,
+        #  consistent with non-split-path
+        df.loc[:, "Alpha"] = categories
+
+        result = df["Alpha"]
+        expected = Series(categories, index=df.index, name="Alpha").astype(
+            object if not using_infer_string else "str"
+        )
+        tm.assert_series_equal(result, expected)
+
+        # double-check that the non-loc setting retains categoricalness
+        df["Alpha"] = categories
+        tm.assert_series_equal(df["Alpha"], Series(categories, name="Alpha"))
+
+    def test_loc_setitem_datetime_coercion(self):
+        # GH#1048
+        df = DataFrame({"c": [Timestamp("2010-10-01")] * 3})
+        df.loc[0:1, "c"] = np.datetime64("2008-08-08")
+        assert Timestamp("2008-08-08") == df.loc[0, "c"]
+        assert Timestamp("2008-08-08") == df.loc[1, "c"]
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[2, "c"] = date(2005, 5, 5)
+
+    @pytest.mark.parametrize("idxer", ["var", ["var"]])
+    def test_loc_setitem_datetimeindex_tz(self, idxer, tz_naive_fixture):
+        # GH#11365
+        tz = tz_naive_fixture
+        idx = date_range(start="2015-07-12", periods=3, freq="h", tz=tz)
+        expected = DataFrame(1.2, index=idx, columns=["var"])
+        # if result started off with object dtype, then the .loc.__setitem__
+        #  below would retain object dtype
+        result = DataFrame(index=idx, columns=["var"], dtype=np.float64)
+        if idxer == "var":
+            with pytest.raises(TypeError, match="Invalid value"):
+                result.loc[:, idxer] = expected
+        else:
+            # See https://github.com/pandas-dev/pandas/issues/56223
+            result.loc[:, idxer] = expected
+            tm.assert_frame_equal(result, expected)
+
+    def test_loc_setitem_time_key(self):
+        index = date_range("2012-01-01", "2012-01-05", freq="30min")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(index), 5)), index=index
+        )
+        akey = time(12, 0, 0)
+        bkey = slice(time(13, 0, 0), time(14, 0, 0))
+        ainds = [24, 72, 120, 168]
+        binds = [26, 27, 28, 74, 75, 76, 122, 123, 124, 170, 171, 172]
+
+        result = df.copy()
+        result.loc[akey] = 0
+        result = result.loc[akey]
+        expected = df.loc[akey].copy()
+        expected.loc[:] = 0
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.loc[akey] = 0
+        result.loc[akey] = df.iloc[ainds]
+        tm.assert_frame_equal(result, df)
+
+        result = df.copy()
+        result.loc[bkey] = 0
+        result = result.loc[bkey]
+        expected = df.loc[bkey].copy()
+        expected.loc[:] = 0
+        tm.assert_frame_equal(result, expected)
+
+        result = df.copy()
+        result.loc[bkey] = 0
+        result.loc[bkey] = df.iloc[binds]
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize("key", ["A", ["A"], ("A", slice(None))])
+    def test_loc_setitem_unsorted_multiindex_columns(self, key):
+        # GH#38601
+        mi = MultiIndex.from_tuples([("A", 4), ("B", "3"), ("A", "2")])
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=mi)
+        obj = df.copy()
+        obj.loc[:, key] = np.zeros((2, 2), dtype="int64")
+        expected = DataFrame([[0, 2, 0], [0, 5, 0]], columns=mi)
+        tm.assert_frame_equal(obj, expected)
+
+        df = df.sort_index(axis=1)
+        df.loc[:, key] = np.zeros((2, 2), dtype="int64")
+        expected = expected.sort_index(axis=1)
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_uint_drop(self, any_int_numpy_dtype):
+        # see GH#18311
+        # assigning series.loc[0] = 4 changed series.dtype to int
+        series = Series([1, 2, 3], dtype=any_int_numpy_dtype)
+        series.loc[0] = 4
+        expected = Series([4, 2, 3], dtype=any_int_numpy_dtype)
+        tm.assert_series_equal(series, expected)
+
+    def test_loc_setitem_td64_non_nano(self):
+        # GH#14155
+        ser = Series(10 * [np.timedelta64(10, "m")])
+        ser.loc[[1, 2, 3]] = np.timedelta64(20, "m")
+        expected = Series(10 * [np.timedelta64(10, "m")])
+        expected.loc[[1, 2, 3]] = Timedelta(np.timedelta64(20, "m"))
+        tm.assert_series_equal(ser, expected)
+
+    def test_loc_setitem_2d_to_1d_raises(self):
+        data = np.random.default_rng(2).standard_normal((2, 2))
+        # float64 dtype to avoid upcast when trying to set float data
+        ser = Series(range(2), dtype="float64")
+
+        msg = "setting an array element with a sequence."
+        with pytest.raises(ValueError, match=msg):
+            ser.loc[range(2)] = data
+
+        with pytest.raises(ValueError, match=msg):
+            ser.loc[:] = data
+
+    def test_loc_getitem_interval_index(self):
+        # GH#19977
+        index = pd.interval_range(start=0, periods=3)
+        df = DataFrame(
+            [[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=index, columns=["A", "B", "C"]
+        )
+
+        expected = 1
+        result = df.loc[0.5, "A"]
+        tm.assert_almost_equal(result, expected)
+
+    def test_loc_getitem_interval_index2(self):
+        # GH#19977
+        index = pd.interval_range(start=0, periods=3, closed="both")
+        df = DataFrame(
+            [[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=index, columns=["A", "B", "C"]
+        )
+
+        index_exp = pd.interval_range(start=0, periods=2, freq=1, closed="both")
+        expected = Series([1, 4], index=index_exp, name="A")
+        result = df.loc[1, "A"]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("tpl", [(1,), (1, 2)])
+    def test_loc_getitem_index_single_double_tuples(self, tpl):
+        # GH#20991
+        idx = Index(
+            [(1,), (1, 2)],
+            name="A",
+            tupleize_cols=False,
+        )
+        df = DataFrame(index=idx)
+
+        result = df.loc[[tpl]]
+        idx = Index([tpl], name="A", tupleize_cols=False)
+        expected = DataFrame(index=idx)
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_index_namedtuple(self):
+        IndexType = namedtuple("IndexType", ["a", "b"])
+        idx1 = IndexType("foo", "bar")
+        idx2 = IndexType("baz", "bof")
+        index = Index([idx1, idx2], name="composite_index", tupleize_cols=False)
+        df = DataFrame([(1, 2), (3, 4)], index=index, columns=["A", "B"])
+
+        result = df.loc[IndexType("foo", "bar")]["A"]
+        assert result == 1
+
+    def test_loc_setitem_single_column_mixed(self, using_infer_string):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)),
+            index=["a", "b", "c", "d", "e"],
+            columns=["foo", "bar", "baz"],
+        )
+        df["str"] = "qux"
+        df.loc[df.index[::2], "str"] = np.nan
+        expected = Series(
+            [np.nan, "qux", np.nan, "qux", np.nan],
+            dtype=object if not using_infer_string else "str",
+        ).values
+        tm.assert_almost_equal(df["str"].values, expected)
+
+    def test_loc_setitem_cast2(self):
+        # GH#7704
+        # dtype conversion on setting
+        df = DataFrame(np.random.default_rng(2).random((30, 3)), columns=tuple("ABC"))
+        df["event"] = np.nan
+        with pytest.raises(TypeError, match="Invalid value"):
+            df.loc[10, "event"] = "foo"
+
+    def test_loc_setitem_cast3(self):
+        # Test that data type is preserved . GH#5782
+        df = DataFrame({"one": np.arange(6, dtype=np.int8)})
+        df.loc[1, "one"] = 6
+        assert df.dtypes.one == np.dtype(np.int8)
+        df.one = np.int8(7)
+        assert df.dtypes.one == np.dtype(np.int8)
+
+    def test_loc_setitem_range_key(self, frame_or_series):
+        # GH#45479 don't treat range key as positional
+        obj = frame_or_series(range(5), index=[3, 4, 1, 0, 2])
+
+        values = [9, 10, 11]
+        if obj.ndim == 2:
+            values = [[9], [10], [11]]
+
+        obj.loc[range(3)] = values
+
+        expected = frame_or_series([0, 1, 10, 9, 11], index=obj.index)
+        tm.assert_equal(obj, expected)
+
+    def test_loc_setitem_numpy_frame_categorical_value(self):
+        # GH#52927
+        df = DataFrame({"a": [1, 1, 1, 1, 1], "b": ["a", "a", "a", "a", "a"]})
+        df.loc[1:2, "a"] = Categorical([2, 2], categories=[1, 2])
+
+        expected = DataFrame({"a": [1, 2, 2, 1, 1], "b": ["a", "a", "a", "a", "a"]})
+        tm.assert_frame_equal(df, expected)
+
+
+class TestLocWithEllipsis:
+    @pytest.fixture
+    def indexer(self, indexer_li):
+        # Test iloc while we're here
+        return indexer_li
+
+    @pytest.fixture
+    def obj(self, series_with_simple_index, frame_or_series):
+        obj = series_with_simple_index
+        if frame_or_series is not Series:
+            obj = obj.to_frame()
+        return obj
+
+    def test_loc_iloc_getitem_ellipsis(self, obj, indexer):
+        result = indexer(obj)[...]
+        tm.assert_equal(result, obj)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_loc_iloc_getitem_leading_ellipses(self, series_with_simple_index, indexer):
+        obj = series_with_simple_index
+        key = 0 if (indexer is tm.iloc or len(obj) == 0) else obj.index[0]
+
+        if indexer is tm.loc and obj.index.inferred_type == "boolean":
+            # passing [False] will get interpreted as a boolean mask
+            # TODO: should it?  unambiguous when lengths dont match?
+            return
+        if indexer is tm.loc and isinstance(obj.index, MultiIndex):
+            msg = "MultiIndex does not support indexing with Ellipsis"
+            with pytest.raises(NotImplementedError, match=msg):
+                result = indexer(obj)[..., [key]]
+
+        elif len(obj) != 0:
+            result = indexer(obj)[..., [key]]
+            expected = indexer(obj)[[key]]
+            tm.assert_series_equal(result, expected)
+
+        key2 = 0 if indexer is tm.iloc else obj.name
+        df = obj.to_frame()
+        result = indexer(df)[..., [key2]]
+        expected = indexer(df)[:, [key2]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_iloc_getitem_ellipses_only_one_ellipsis(self, obj, indexer):
+        # GH37750
+        key = 0 if (indexer is tm.iloc or len(obj) == 0) else obj.index[0]
+
+        with pytest.raises(IndexingError, match=_one_ellipsis_message):
+            indexer(obj)[..., ...]
+
+        with pytest.raises(IndexingError, match=_one_ellipsis_message):
+            indexer(obj)[..., [key], ...]
+
+        with pytest.raises(IndexingError, match=_one_ellipsis_message):
+            indexer(obj)[..., ..., key]
+
+        # one_ellipsis_message takes precedence over "Too many indexers"
+        #  only when the first key is Ellipsis
+        with pytest.raises(IndexingError, match="Too many indexers"):
+            indexer(obj)[key, ..., ...]
+
+
+class TestLocWithMultiIndex:
+    @pytest.mark.parametrize(
+        "keys, expected",
+        [
+            (["b", "a"], [["b", "b", "a", "a"], [1, 2, 1, 2]]),
+            (["a", "b"], [["a", "a", "b", "b"], [1, 2, 1, 2]]),
+            ((["a", "b"], [1, 2]), [["a", "a", "b", "b"], [1, 2, 1, 2]]),
+            ((["a", "b"], [2, 1]), [["a", "a", "b", "b"], [2, 1, 2, 1]]),
+            ((["b", "a"], [2, 1]), [["b", "b", "a", "a"], [2, 1, 2, 1]]),
+            ((["b", "a"], [1, 2]), [["b", "b", "a", "a"], [1, 2, 1, 2]]),
+            ((["c", "a"], [2, 1]), [["c", "a", "a"], [1, 2, 1]]),
+        ],
+    )
+    @pytest.mark.parametrize("dim", ["index", "columns"])
+    def test_loc_getitem_multilevel_index_order(self, dim, keys, expected):
+        # GH#22797
+        # Try to respect order of keys given for MultiIndex.loc
+        kwargs = {dim: [["c", "a", "a", "b", "b"], [1, 1, 2, 1, 2]]}
+        df = DataFrame(np.arange(25).reshape(5, 5), **kwargs)
+        exp_index = MultiIndex.from_arrays(expected)
+        if dim == "index":
+            res = df.loc[keys, :]
+            tm.assert_index_equal(res.index, exp_index)
+        elif dim == "columns":
+            res = df.loc[:, keys]
+            tm.assert_index_equal(res.columns, exp_index)
+
+    def test_loc_preserve_names(self, multiindex_year_month_day_dataframe_random_data):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        result = ymd.loc[2000]
+        result2 = ymd["A"].loc[2000]
+        assert result.index.names == ymd.index.names[1:]
+        assert result2.index.names == ymd.index.names[1:]
+
+        result = ymd.loc[2000, 2]
+        result2 = ymd["A"].loc[2000, 2]
+        assert result.index.name == ymd.index.names[2]
+        assert result2.index.name == ymd.index.names[2]
+
+    def test_loc_getitem_multiindex_nonunique_len_zero(self):
+        # GH#13691
+        mi = MultiIndex.from_product([[0], [1, 1]])
+        ser = Series(0, index=mi)
+
+        res = ser.loc[[]]
+
+        expected = ser[:0]
+        tm.assert_series_equal(res, expected)
+
+        res2 = ser.loc[ser.iloc[0:0]]
+        tm.assert_series_equal(res2, expected)
+
+    def test_loc_getitem_access_none_value_in_multiindex(self):
+        # GH#34318: test that you can access a None value using .loc
+        #  through a Multiindex
+
+        ser = Series([None], MultiIndex.from_arrays([["Level1"], ["Level2"]]))
+        result = ser.loc[("Level1", "Level2")]
+        assert result is None
+
+        midx = MultiIndex.from_product([["Level1"], ["Level2_a", "Level2_b"]])
+        ser = Series([None] * len(midx), dtype=object, index=midx)
+        result = ser.loc[("Level1", "Level2_a")]
+        assert result is None
+
+        ser = Series([1] * len(midx), dtype=object, index=midx)
+        result = ser.loc[("Level1", "Level2_a")]
+        assert result == 1
+
+    def test_loc_setitem_multiindex_slice(self):
+        # GH 34870
+
+        index = MultiIndex.from_tuples(
+            zip(
+                ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+                ["one", "two", "one", "two", "one", "two", "one", "two"],
+            ),
+            names=["first", "second"],
+        )
+
+        result = Series([1, 1, 1, 1, 1, 1, 1, 1], index=index)
+        result.loc[("baz", "one") : ("foo", "two")] = 100
+
+        expected = Series([1, 1, 100, 100, 100, 100, 1, 1], index=index)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_slice_datetime_objs_with_datetimeindex(self):
+        times = date_range("2000-01-01", freq="10min", periods=100000)
+        ser = Series(range(100000), times)
+        result = ser.loc[datetime(1900, 1, 1) : datetime(2100, 1, 1)]
+        tm.assert_series_equal(result, ser)
+
+    def test_loc_getitem_datetime_string_with_datetimeindex(self):
+        # GH 16710
+        df = DataFrame(
+            {"a": range(10), "b": range(10)},
+            index=date_range("2010-01-01", "2010-01-10"),
+        )
+        result = df.loc[["2010-01-01", "2010-01-05"], ["a", "b"]]
+        expected = DataFrame(
+            {"a": [0, 4], "b": [0, 4]},
+            index=DatetimeIndex(["2010-01-01", "2010-01-05"]).as_unit("ns"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_sorted_index_level_with_duplicates(self):
+        # GH#4516 sorting a MultiIndex with duplicates and multiple dtypes
+        mi = MultiIndex.from_tuples(
+            [
+                ("foo", "bar"),
+                ("foo", "bar"),
+                ("bah", "bam"),
+                ("bah", "bam"),
+                ("foo", "bar"),
+                ("bah", "bam"),
+            ],
+            names=["A", "B"],
+        )
+        df = DataFrame(
+            [
+                [1.0, 1],
+                [2.0, 2],
+                [3.0, 3],
+                [4.0, 4],
+                [5.0, 5],
+                [6.0, 6],
+            ],
+            index=mi,
+            columns=["C", "D"],
+        )
+        df = df.sort_index(level=0)
+
+        expected = DataFrame(
+            [[1.0, 1], [2.0, 2], [5.0, 5]], columns=["C", "D"], index=mi.take([0, 1, 4])
+        )
+
+        result = df.loc[("foo", "bar")]
+        tm.assert_frame_equal(result, expected)
+
+    def test_additional_element_to_categorical_series_loc(self):
+        # GH#47677
+        result = Series(["a", "b", "c"], dtype="category")
+        result.loc[3] = 0
+        expected = Series(["a", "b", "c", 0], dtype="object")
+        tm.assert_series_equal(result, expected)
+
+    def test_additional_categorical_element_loc(self):
+        # GH#47677
+        result = Series(["a", "b", "c"], dtype="category")
+        result.loc[3] = "a"
+        expected = Series(["a", "b", "c", "a"], dtype="category")
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_set_nan_in_categorical_series(self, any_numeric_ea_dtype):
+        # GH#47677
+        srs = Series(
+            [1, 2, 3],
+            dtype=CategoricalDtype(Index([1, 2, 3], dtype=any_numeric_ea_dtype)),
+        )
+        # enlarge
+        srs.loc[3] = np.nan
+        expected = Series(
+            [1, 2, 3, np.nan],
+            dtype=CategoricalDtype(Index([1, 2, 3], dtype=any_numeric_ea_dtype)),
+        )
+        tm.assert_series_equal(srs, expected)
+        # set into
+        srs.loc[1] = np.nan
+        expected = Series(
+            [1, np.nan, 3, np.nan],
+            dtype=CategoricalDtype(Index([1, 2, 3], dtype=any_numeric_ea_dtype)),
+        )
+        tm.assert_series_equal(srs, expected)
+
+    @pytest.mark.parametrize("na", (np.nan, pd.NA, None, pd.NaT))
+    def test_loc_consistency_series_enlarge_set_into(self, na):
+        # GH#47677
+        srs_enlarge = Series(["a", "b", "c"], dtype="category")
+        srs_enlarge.loc[3] = na
+
+        srs_setinto = Series(["a", "b", "c", "a"], dtype="category")
+        srs_setinto.loc[3] = na
+
+        tm.assert_series_equal(srs_enlarge, srs_setinto)
+        expected = Series(["a", "b", "c", na], dtype="category")
+        tm.assert_series_equal(srs_enlarge, expected)
+
+    def test_loc_getitem_preserves_index_level_category_dtype(self):
+        # GH#15166
+        df = DataFrame(
+            data=np.arange(2, 22, 2),
+            index=MultiIndex(
+                levels=[CategoricalIndex(["a", "b"]), range(10)],
+                codes=[[0] * 5 + [1] * 5, range(10)],
+                names=["Index1", "Index2"],
+            ),
+        )
+
+        expected = CategoricalIndex(
+            ["a", "b"],
+            categories=["a", "b"],
+            ordered=False,
+            name="Index1",
+            dtype="category",
+        )
+
+        result = df.index.levels[0]
+        tm.assert_index_equal(result, expected)
+
+        result = df.loc[["a"]].index.levels[0]
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("lt_value", [30, 10])
+    def test_loc_multiindex_levels_contain_values_not_in_index_anymore(self, lt_value):
+        # GH#41170
+        df = DataFrame({"a": [12, 23, 34, 45]}, index=[list("aabb"), [0, 1, 2, 3]])
+        with pytest.raises(KeyError, match=r"\['b'\] not in index"):
+            df.loc[df["a"] < lt_value, :].loc[["b"], :]
+
+    def test_loc_multiindex_null_slice_na_level(self):
+        # GH#42055
+        lev1 = np.array([np.nan, np.nan])
+        lev2 = ["bar", "baz"]
+        mi = MultiIndex.from_arrays([lev1, lev2])
+        ser = Series([0, 1], index=mi)
+        result = ser.loc[:, "bar"]
+
+        # TODO: should we have name="bar"?
+        expected = Series([0], index=[np.nan])
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_drops_level(self):
+        # Based on test_series_varied_multiindex_alignment, where
+        #  this used to fail to drop the first level
+        mi = MultiIndex.from_product(
+            [list("ab"), list("xy"), [1, 2]], names=["ab", "xy", "num"]
+        )
+        ser = Series(range(8), index=mi)
+
+        loc_result = ser.loc["a", :, :]
+        expected = ser.index.droplevel(0)[:4]
+        tm.assert_index_equal(loc_result.index, expected)
+
+
+class TestLocSetitemWithExpansion:
+    def test_loc_setitem_with_expansion_large_dataframe(self, monkeypatch):
+        # GH#10692
+        size_cutoff = 50
+        with monkeypatch.context():
+            monkeypatch.setattr(libindex, "_SIZE_CUTOFF", size_cutoff)
+            result = DataFrame({"x": range(size_cutoff)}, dtype="int64")
+            result.loc[size_cutoff] = size_cutoff
+        expected = DataFrame({"x": range(size_cutoff + 1)}, dtype="int64")
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_setitem_empty_series(self):
+        # GH#5226
+
+        # partially set with an empty object series
+        ser = Series(dtype=object)
+        ser.loc[1] = 1
+        tm.assert_series_equal(ser, Series([1], index=range(1, 2)))
+        ser.loc[3] = 3
+        tm.assert_series_equal(ser, Series([1, 3], index=[1, 3]))
+
+    def test_loc_setitem_empty_series_float(self):
+        # GH#5226
+
+        # partially set with an empty object series
+        ser = Series(dtype=object)
+        ser.loc[1] = 1.0
+        tm.assert_series_equal(ser, Series([1.0], index=range(1, 2)))
+        ser.loc[3] = 3.0
+        tm.assert_series_equal(ser, Series([1.0, 3.0], index=[1, 3]))
+
+    def test_loc_setitem_empty_series_str_idx(self):
+        # GH#5226
+
+        # partially set with an empty object series
+        ser = Series(dtype=object)
+        ser.loc["foo"] = 1
+        tm.assert_series_equal(ser, Series([1], index=Index(["foo"])))
+        ser.loc["bar"] = 3
+        tm.assert_series_equal(ser, Series([1, 3], index=Index(["foo", "bar"])))
+        ser.loc[3] = 4
+        tm.assert_series_equal(ser, Series([1, 3, 4], index=Index(["foo", "bar", 3])))
+
+    def test_loc_setitem_incremental_with_dst(self):
+        # GH#20724
+        base = datetime(2015, 11, 1, tzinfo=gettz("US/Pacific"))
+        idxs = [base + timedelta(seconds=i * 900) for i in range(16)]
+        result = Series([0], index=[idxs[0]])
+        for ts in idxs:
+            result.loc[ts] = 1
+        expected = Series(1, index=idxs)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "conv",
+        [
+            lambda x: x,
+            lambda x: x.to_datetime64(),
+            lambda x: x.to_pydatetime(),
+            lambda x: np.datetime64(x),
+        ],
+        ids=["self", "to_datetime64", "to_pydatetime", "np.datetime64"],
+    )
+    def test_loc_setitem_datetime_keys_cast(self, conv, using_infer_string):
+        # GH#9516, GH#51363 changed in 3.0 to not cast on Index.insert
+        dt1 = Timestamp("20130101 09:00:00")
+        dt2 = Timestamp("20130101 10:00:00")
+        df = DataFrame()
+        df.loc[conv(dt1), "one"] = 100
+        df.loc[conv(dt2), "one"] = 200
+
+        expected = DataFrame(
+            {"one": [100.0, 200.0]},
+            index=Index(
+                [conv(dt1), conv(dt2)], dtype=None if using_infer_string else object
+            ),
+            columns=Index(["one"]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_categorical_column_retains_dtype(self, ordered):
+        # GH16360
+        result = DataFrame({"A": [1]})
+        result.loc[:, "B"] = Categorical(["b"], ordered=ordered)
+        expected = DataFrame({"A": [1], "B": Categorical(["b"], ordered=ordered)})
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_setitem_with_expansion_and_existing_dst(self):
+        # GH#18308
+        start = Timestamp("2017-10-29 00:00:00+0200", tz="Europe/Madrid")
+        end = Timestamp("2017-10-29 03:00:00+0100", tz="Europe/Madrid")
+        ts = Timestamp("2016-10-10 03:00:00", tz="Europe/Madrid")
+        idx = date_range(start, end, inclusive="left", freq="h")
+        assert ts not in idx  # i.e. result.loc setitem is with-expansion
+
+        result = DataFrame(index=idx, columns=["value"])
+        result.loc[ts, "value"] = 12
+        expected = DataFrame(
+            [np.nan] * len(idx) + [12],
+            index=idx.append(DatetimeIndex([ts])),
+            columns=["value"],
+            dtype=object,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_setitem_with_expansion(self):
+        # indexing - setting an element
+        df = DataFrame(
+            data=to_datetime(["2015-03-30 20:12:32", "2015-03-12 00:11:11"]),
+            columns=["time"],
+        )
+        df["new_col"] = ["new", "old"]
+        df.time = df.set_index("time").index.tz_localize("UTC")
+        v = df[df.new_col == "new"].set_index("time").index.tz_convert("US/Pacific")
+
+        # pre-2.0  trying to set a single element on a part of a different
+        #  timezone converted to object; in 2.0 it retains dtype
+        df2 = df.copy()
+        df2.loc[df2.new_col == "new", "time"] = v
+
+        expected = Series([v[0].tz_convert("UTC"), df.loc[1, "time"]], name="time")
+        tm.assert_series_equal(df2.time, expected)
+
+        v = df.loc[df.new_col == "new", "time"] + Timedelta("1s").as_unit("s")
+        df.loc[df.new_col == "new", "time"] = v
+        tm.assert_series_equal(df.loc[df.new_col == "new", "time"], v)
+
+    def test_loc_setitem_with_expansion_inf_upcast_empty(self):
+        # Test with np.inf in columns
+        df = DataFrame()
+        df.loc[0, 0] = 1
+        df.loc[1, 1] = 2
+        df.loc[0, np.inf] = 3
+
+        result = df.columns
+        expected = Index([0, 1, np.inf], dtype=np.float64)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:indexing past lexsort depth")
+    def test_loc_setitem_with_expansion_nonunique_index(self, index):
+        # GH#40096
+        if not len(index):
+            pytest.skip("Not relevant for empty Index")
+
+        index = index.repeat(2)  # ensure non-unique
+        N = len(index)
+        arr = np.arange(N).astype(np.int64)
+
+        orig = DataFrame(arr, index=index)
+
+        # key that will requiring object-dtype casting in the index
+        key = "kapow"
+        assert key not in index  # otherwise test is invalid
+        # TODO: using a tuple key breaks here in many cases
+
+        exp_index = index.insert(len(index), key)
+        if isinstance(index, MultiIndex):
+            assert exp_index[-1][0] == key
+        else:
+            assert exp_index[-1] == key
+        exp_data = np.arange(N + 1).astype(np.float64)
+        expected = DataFrame(exp_data, index=exp_index)
+
+        # Add new row, but no new columns
+        df = orig.copy()
+        df.loc[key, 0] = N
+        tm.assert_frame_equal(df, expected)
+
+        # add new row on a Series
+        ser = orig.copy()[0]
+        ser.loc[key] = N
+        # the series machinery lets us preserve int dtype instead of float
+        expected = expected[0].astype(np.int64)
+        tm.assert_series_equal(ser, expected)
+
+        # add new row and new column
+        df = orig.copy()
+        df.loc[key, 1] = N
+        expected = DataFrame(
+            {0: list(arr) + [np.nan], 1: [np.nan] * N + [float(N)]},
+            index=exp_index,
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_with_expansion_preserves_nullable_int(
+        self, any_numeric_ea_dtype
+    ):
+        # GH#42099
+        ser = Series([0, 1, 2, 3], dtype=any_numeric_ea_dtype)
+        df = DataFrame({"data": ser})
+
+        result = DataFrame(index=df.index)
+        result.loc[df.index, "data"] = ser
+
+        tm.assert_frame_equal(result, df, check_column_type=False)
+
+        result = DataFrame(index=df.index)
+        result.loc[df.index, "data"] = ser._values
+        tm.assert_frame_equal(result, df, check_column_type=False)
+
+    def test_loc_setitem_ea_not_full_column(self):
+        # GH#39163
+        df = DataFrame({"A": range(5)})
+
+        val = date_range("2016-01-01", periods=3, tz="US/Pacific")
+
+        df.loc[[0, 1, 2], "B"] = val
+
+        bex = val.append(DatetimeIndex([pd.NaT, pd.NaT], dtype=val.dtype))
+        expected = DataFrame({"A": range(5), "B": bex})
+        assert expected.dtypes["B"] == val.dtype
+        tm.assert_frame_equal(df, expected)
+
+
+class TestLocCallable:
+    def test_frame_loc_getitem_callable(self):
+        # GH#11485
+        df = DataFrame({"A": [1, 2, 3, 4], "B": list("aabb"), "C": [1, 2, 3, 4]})
+        # iloc cannot use boolean Series (see GH3635)
+
+        # return bool indexer
+        res = df.loc[lambda x: x.A > 2]
+        tm.assert_frame_equal(res, df.loc[df.A > 2])
+
+        res = df.loc[lambda x: x.B == "b", :]
+        tm.assert_frame_equal(res, df.loc[df.B == "b", :])
+
+        res = df.loc[lambda x: x.A > 2, lambda x: x.columns == "B"]
+        tm.assert_frame_equal(res, df.loc[df.A > 2, [False, True, False]])
+
+        res = df.loc[lambda x: x.A > 2, lambda x: "B"]
+        tm.assert_series_equal(res, df.loc[df.A > 2, "B"])
+
+        res = df.loc[lambda x: x.A > 2, lambda x: ["A", "B"]]
+        tm.assert_frame_equal(res, df.loc[df.A > 2, ["A", "B"]])
+
+        res = df.loc[lambda x: x.A == 2, lambda x: ["A", "B"]]
+        tm.assert_frame_equal(res, df.loc[df.A == 2, ["A", "B"]])
+
+        # scalar
+        res = df.loc[lambda x: 1, lambda x: "A"]
+        assert res == df.loc[1, "A"]
+
+    def test_frame_loc_getitem_callable_mixture(self):
+        # GH#11485
+        df = DataFrame({"A": [1, 2, 3, 4], "B": list("aabb"), "C": [1, 2, 3, 4]})
+
+        res = df.loc[lambda x: x.A > 2, ["A", "B"]]
+        tm.assert_frame_equal(res, df.loc[df.A > 2, ["A", "B"]])
+
+        res = df.loc[[2, 3], lambda x: ["A", "B"]]
+        tm.assert_frame_equal(res, df.loc[[2, 3], ["A", "B"]])
+
+        res = df.loc[3, lambda x: ["A", "B"]]
+        tm.assert_series_equal(res, df.loc[3, ["A", "B"]])
+
+    def test_frame_loc_getitem_callable_labels(self):
+        # GH#11485
+        df = DataFrame({"X": [1, 2, 3, 4], "Y": list("aabb")}, index=list("ABCD"))
+
+        # return label
+        res = df.loc[lambda x: ["A", "C"]]
+        tm.assert_frame_equal(res, df.loc[["A", "C"]])
+
+        res = df.loc[lambda x: ["A", "C"], :]
+        tm.assert_frame_equal(res, df.loc[["A", "C"], :])
+
+        res = df.loc[lambda x: ["A", "C"], lambda x: "X"]
+        tm.assert_series_equal(res, df.loc[["A", "C"], "X"])
+
+        res = df.loc[lambda x: ["A", "C"], lambda x: ["X"]]
+        tm.assert_frame_equal(res, df.loc[["A", "C"], ["X"]])
+
+        # mixture
+        res = df.loc[["A", "C"], lambda x: "X"]
+        tm.assert_series_equal(res, df.loc[["A", "C"], "X"])
+
+        res = df.loc[["A", "C"], lambda x: ["X"]]
+        tm.assert_frame_equal(res, df.loc[["A", "C"], ["X"]])
+
+        res = df.loc[lambda x: ["A", "C"], "X"]
+        tm.assert_series_equal(res, df.loc[["A", "C"], "X"])
+
+        res = df.loc[lambda x: ["A", "C"], ["X"]]
+        tm.assert_frame_equal(res, df.loc[["A", "C"], ["X"]])
+
+    def test_frame_loc_setitem_callable(self):
+        # GH#11485
+        df = DataFrame(
+            {"X": [1, 2, 3, 4], "Y": Series(list("aabb"), dtype=object)},
+            index=list("ABCD"),
+        )
+
+        # return label
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"]] = -20
+        exp = df.copy()
+        exp.loc[["A", "C"]] = -20
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"], :] = 20
+        exp = df.copy()
+        exp.loc[["A", "C"], :] = 20
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"], lambda x: "X"] = -1
+        exp = df.copy()
+        exp.loc[["A", "C"], "X"] = -1
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"], lambda x: ["X"]] = [5, 10]
+        exp = df.copy()
+        exp.loc[["A", "C"], ["X"]] = [5, 10]
+        tm.assert_frame_equal(res, exp)
+
+        # mixture
+        res = df.copy()
+        res.loc[["A", "C"], lambda x: "X"] = np.array([-1, -2])
+        exp = df.copy()
+        exp.loc[["A", "C"], "X"] = np.array([-1, -2])
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[["A", "C"], lambda x: ["X"]] = 10
+        exp = df.copy()
+        exp.loc[["A", "C"], ["X"]] = 10
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"], "X"] = -2
+        exp = df.copy()
+        exp.loc[["A", "C"], "X"] = -2
+        tm.assert_frame_equal(res, exp)
+
+        res = df.copy()
+        res.loc[lambda x: ["A", "C"], ["X"]] = -4
+        exp = df.copy()
+        exp.loc[["A", "C"], ["X"]] = -4
+        tm.assert_frame_equal(res, exp)
+
+
+class TestPartialStringSlicing:
+    def test_loc_getitem_partial_string_slicing_datetimeindex(self):
+        # GH#35509
+        df = DataFrame(
+            {"col1": ["a", "b", "c"], "col2": [1, 2, 3]},
+            index=to_datetime(["2020-08-01", "2020-07-02", "2020-08-05"]),
+        )
+        expected = DataFrame(
+            {"col1": ["a", "c"], "col2": [1, 3]},
+            index=to_datetime(["2020-08-01", "2020-08-05"]),
+        )
+        result = df.loc["2020-08"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_partial_string_slicing_with_periodindex(self):
+        pi = pd.period_range(start="2017-01-01", end="2018-01-01", freq="M")
+        ser = pi.to_series()
+        result = ser.loc[:"2017-12"]
+        expected = ser.iloc[:-1]
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_partial_string_slicing_with_timedeltaindex(self):
+        ix = timedelta_range(start="1 day", end="2 days", freq="1h")
+        ser = ix.to_series()
+        result = ser.loc[:"1 days"]
+        expected = ser.iloc[:-1]
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_str_timedeltaindex(self):
+        # GH#16896
+        df = DataFrame({"x": range(3)}, index=to_timedelta(range(3), unit="days"))
+        expected = df.iloc[0]
+        sliced = df.loc["0 days"]
+        tm.assert_series_equal(sliced, expected)
+
+    @pytest.mark.parametrize("indexer_end", [None, "2020-01-02 23:59:59.999999999"])
+    def test_loc_getitem_partial_slice_non_monotonicity(
+        self, tz_aware_fixture, indexer_end, frame_or_series
+    ):
+        # GH#33146
+        obj = frame_or_series(
+            [1] * 5,
+            index=DatetimeIndex(
+                [
+                    Timestamp("2019-12-30"),
+                    Timestamp("2020-01-01"),
+                    Timestamp("2019-12-25"),
+                    Timestamp("2020-01-02 23:59:59.999999999"),
+                    Timestamp("2019-12-19"),
+                ],
+                tz=tz_aware_fixture,
+            ),
+        )
+        expected = frame_or_series(
+            [1] * 2,
+            index=DatetimeIndex(
+                [
+                    Timestamp("2020-01-01"),
+                    Timestamp("2020-01-02 23:59:59.999999999"),
+                ],
+                tz=tz_aware_fixture,
+            ),
+        )
+        indexer = slice("2020-01-01", indexer_end)
+
+        result = obj[indexer]
+        tm.assert_equal(result, expected)
+
+        result = obj.loc[indexer]
+        tm.assert_equal(result, expected)
+
+
+class TestLabelSlicing:
+    def test_loc_getitem_slicing_datetimes_frame(self):
+        # GH#7523
+
+        # unique
+        df_unique = DataFrame(
+            np.arange(4.0, dtype="float64"),
+            index=[datetime(2001, 1, i, 10, 00) for i in [1, 2, 3, 4]],
+        )
+
+        # duplicates
+        df_dups = DataFrame(
+            np.arange(5.0, dtype="float64"),
+            index=[datetime(2001, 1, i, 10, 00) for i in [1, 2, 2, 3, 4]],
+        )
+
+        for df in [df_unique, df_dups]:
+            result = df.loc[datetime(2001, 1, 1, 10) :]
+            tm.assert_frame_equal(result, df)
+            result = df.loc[: datetime(2001, 1, 4, 10)]
+            tm.assert_frame_equal(result, df)
+            result = df.loc[datetime(2001, 1, 1, 10) : datetime(2001, 1, 4, 10)]
+            tm.assert_frame_equal(result, df)
+
+            result = df.loc[datetime(2001, 1, 1, 11) :]
+            expected = df.iloc[1:]
+            tm.assert_frame_equal(result, expected)
+            result = df.loc["20010101 11":]
+            tm.assert_frame_equal(result, expected)
+
+    def test_loc_getitem_label_slice_across_dst(self):
+        # GH#21846
+        idx = date_range(
+            "2017-10-29 01:30:00", tz="Europe/Berlin", periods=5, freq="30 min"
+        )
+        series2 = Series([0, 1, 2, 3, 4], index=idx)
+
+        t_1 = Timestamp("2017-10-29 02:30:00+02:00", tz="Europe/Berlin")
+        t_2 = Timestamp("2017-10-29 02:00:00+01:00", tz="Europe/Berlin")
+        result = series2.loc[t_1:t_2]
+        expected = Series([2, 3], index=idx[2:4])
+        tm.assert_series_equal(result, expected)
+
+        result = series2[t_1]
+        expected = 2
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            pd.period_range(start="2017-01-01", end="2018-01-01", freq="M"),
+            timedelta_range(start="1 day", end="2 days", freq="1h"),
+        ],
+    )
+    def test_loc_getitem_label_slice_period_timedelta(self, index):
+        ser = index.to_series()
+        result = ser.loc[: index[-2]]
+        expected = ser.iloc[:-1]
+
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_getitem_slice_floats_inexact(self):
+        index = [52195.504153, 52196.303147, 52198.369883]
+        df = DataFrame(np.random.default_rng(2).random((3, 2)), index=index)
+
+        s1 = df.loc[52195.1:52196.5]
+        assert len(s1) == 2
+
+        s1 = df.loc[52195.1:52196.6]
+        assert len(s1) == 2
+
+        s1 = df.loc[52195.1:52198.9]
+        assert len(s1) == 3
+
+    def test_loc_getitem_float_slice_floatindex(self, float_numpy_dtype):
+        dtype = float_numpy_dtype
+        ser = Series(
+            np.random.default_rng(2).random(10), index=np.arange(10, 20, dtype=dtype)
+        )
+
+        assert len(ser.loc[12.0:]) == 8
+        assert len(ser.loc[12.5:]) == 7
+
+        idx = np.arange(10, 20, dtype=dtype)
+        idx[2] = 12.2
+        ser.index = idx
+        assert len(ser.loc[12.0:]) == 8
+        assert len(ser.loc[12.5:]) == 7
+
+    @pytest.mark.parametrize(
+        "start,stop, expected_slice",
+        [
+            [np.timedelta64(0, "ns"), None, slice(0, 11)],
+            [np.timedelta64(1, "D"), np.timedelta64(6, "D"), slice(1, 7)],
+            [None, np.timedelta64(4, "D"), slice(0, 5)],
+        ],
+    )
+    def test_loc_getitem_slice_label_td64obj(self, start, stop, expected_slice):
+        # GH#20393
+        ser = Series(range(11), timedelta_range("0 days", "10 days"))
+        result = ser.loc[slice(start, stop)]
+        expected = ser.iloc[expected_slice]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("start", ["2018", "2020"])
+    def test_loc_getitem_slice_unordered_dt_index(self, frame_or_series, start):
+        obj = frame_or_series(
+            [1, 2, 3],
+            index=[Timestamp("2016"), Timestamp("2019"), Timestamp("2017")],
+        )
+        with pytest.raises(
+            KeyError, match="Value based partial slicing on non-monotonic"
+        ):
+            obj.loc[start:"2022"]
+
+    @pytest.mark.parametrize("value", [1, 1.5])
+    def test_loc_getitem_slice_labels_int_in_object_index(self, frame_or_series, value):
+        # GH: 26491
+        obj = frame_or_series(range(4), index=[value, "first", 2, "third"])
+        result = obj.loc[value:"third"]
+        expected = frame_or_series(range(4), index=[value, "first", 2, "third"])
+        tm.assert_equal(result, expected)
+
+    def test_loc_getitem_slice_columns_mixed_dtype(self):
+        # GH: 20975
+        df = DataFrame({"test": 1, 1: 2, 2: 3}, index=[0])
+        expected = DataFrame(
+            data=[[2, 3]], index=[0], columns=Index([1, 2], dtype=object)
+        )
+        tm.assert_frame_equal(df.loc[:, 1:], expected)
+
+
+class TestLocBooleanLabelsAndSlices:
+    @pytest.mark.parametrize("bool_value", [True, False])
+    def test_loc_bool_incompatible_index_raises(
+        self, index, frame_or_series, bool_value
+    ):
+        # GH20432
+        message = f"{bool_value}: boolean label can not be used without a boolean index"
+        if index.inferred_type != "boolean":
+            obj = frame_or_series(index=index, dtype="object")
+            with pytest.raises(KeyError, match=message):
+                obj.loc[bool_value]
+
+    @pytest.mark.parametrize("bool_value", [True, False])
+    def test_loc_bool_should_not_raise(self, frame_or_series, bool_value):
+        obj = frame_or_series(
+            index=Index([True, False], dtype="boolean"), dtype="object"
+        )
+        obj.loc[bool_value]
+
+    def test_loc_bool_slice_raises(self, index, frame_or_series):
+        # GH20432
+        message = (
+            r"slice\(True, False, None\): boolean values can not be used in a slice"
+        )
+        obj = frame_or_series(index=index, dtype="object")
+        with pytest.raises(TypeError, match=message):
+            obj.loc[True:False]
+
+
+class TestLocBooleanMask:
+    def test_loc_setitem_bool_mask_timedeltaindex(self):
+        # GH#14946
+        df = DataFrame({"x": range(10)})
+        df.index = to_timedelta(range(10), unit="s")
+        conditions = [df["x"] > 3, df["x"] == 3, df["x"] < 3]
+        expected_data = [
+            [0, 1, 2, 3, 10, 10, 10, 10, 10, 10],
+            [0, 1, 2, 10, 4, 5, 6, 7, 8, 9],
+            [10, 10, 10, 3, 4, 5, 6, 7, 8, 9],
+        ]
+        for cond, data in zip(conditions, expected_data):
+            result = df.copy()
+            result.loc[cond, "x"] = 10
+
+            expected = DataFrame(
+                data,
+                index=to_timedelta(range(10), unit="s"),
+                columns=["x"],
+                dtype="int64",
+            )
+            tm.assert_frame_equal(expected, result)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_loc_setitem_mask_with_datetimeindex_tz(self, tz):
+        # GH#16889
+        # support .loc with alignment and tz-aware DatetimeIndex
+        mask = np.array([True, False, True, False])
+
+        idx = date_range("20010101", periods=4, tz=tz)
+        df = DataFrame({"a": np.arange(4)}, index=idx).astype("float64")
+
+        result = df.copy()
+        result.loc[mask, :] = df.loc[mask, :]
+        tm.assert_frame_equal(result, df)
+
+        result = df.copy()
+        result.loc[mask] = df.loc[mask]
+        tm.assert_frame_equal(result, df)
+
+    def test_loc_setitem_mask_and_label_with_datetimeindex(self):
+        # GH#9478
+        # a datetimeindex alignment issue with partial setting
+        df = DataFrame(
+            np.arange(6.0).reshape(3, 2),
+            columns=list("AB"),
+            index=date_range("1/1/2000", periods=3, freq="1h"),
+        )
+        expected = df.copy()
+        expected["C"] = [expected.index[0]] + [pd.NaT, pd.NaT]
+
+        mask = df.A < 1
+        df.loc[mask, "C"] = df.loc[mask].index
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_mask_td64_series_value(self):
+        # GH#23462 key list of bools, value is a Series
+        td1 = Timedelta(0)
+        td2 = Timedelta(28767471428571405)
+        df = DataFrame({"col": Series([td1, td2])})
+        df_copy = df.copy()
+        ser = Series([td1])
+
+        expected = df["col"].iloc[1]._value
+        df.loc[[True, False]] = ser
+        result = df["col"].iloc[1]._value
+
+        assert expected == result
+        tm.assert_frame_equal(df, df_copy)
+
+    def test_loc_setitem_boolean_and_column(self, float_frame):
+        expected = float_frame.copy()
+        mask = float_frame["A"] > 0
+
+        float_frame.loc[mask, "B"] = 0
+
+        values = expected.values.copy()
+        values[mask.values, 1] = 0
+        expected = DataFrame(values, index=expected.index, columns=expected.columns)
+        tm.assert_frame_equal(float_frame, expected)
+
+    def test_loc_setitem_ndframe_values_alignment(self):
+        # GH#45501
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df.loc[[False, False, True], ["a"]] = DataFrame(
+            {"a": [10, 20, 30]}, index=[2, 1, 0]
+        )
+
+        expected = DataFrame({"a": [1, 2, 10], "b": [4, 5, 6]})
+        tm.assert_frame_equal(df, expected)
+
+        # same thing with Series RHS
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df.loc[[False, False, True], ["a"]] = Series([10, 11, 12], index=[2, 1, 0])
+        tm.assert_frame_equal(df, expected)
+
+        # same thing but setting "a" instead of ["a"]
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df.loc[[False, False, True], "a"] = Series([10, 11, 12], index=[2, 1, 0])
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+        df_orig = df.copy()
+        ser = df["a"]
+        ser.loc[[False, False, True]] = Series([10, 11, 12], index=[2, 1, 0])
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_loc_indexer_empty_broadcast(self):
+        # GH#51450
+        df = DataFrame({"a": [], "b": []}, dtype=object)
+        expected = df.copy()
+        df.loc[np.array([], dtype=np.bool_), ["a"]] = df["a"].copy()
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_indexer_all_false_broadcast(self):
+        # GH#51450
+        df = DataFrame({"a": ["x"], "b": ["y"]}, dtype=object)
+        expected = df.copy()
+        df.loc[np.array([False], dtype=np.bool_), ["a"]] = df["b"].copy()
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_indexer_length_one(self):
+        # GH#51435
+        df = DataFrame({"a": ["x"], "b": ["y"]}, dtype=object)
+        expected = DataFrame({"a": ["y"], "b": ["y"]}, dtype=object)
+        df.loc[np.array([True], dtype=np.bool_), ["a"]] = df["b"].copy()
+        tm.assert_frame_equal(df, expected)
+
+
+class TestLocListlike:
+    @pytest.mark.parametrize("box", [lambda x: x, np.asarray, list])
+    def test_loc_getitem_list_of_labels_categoricalindex_with_na(self, box):
+        # passing a list can include valid categories _or_ NA values
+        ci = CategoricalIndex(["A", "B", np.nan])
+        ser = Series(range(3), index=ci)
+
+        result = ser.loc[box(ci)]
+        tm.assert_series_equal(result, ser)
+
+        result = ser[box(ci)]
+        tm.assert_series_equal(result, ser)
+
+        result = ser.to_frame().loc[box(ci)]
+        tm.assert_frame_equal(result, ser.to_frame())
+
+        ser2 = ser[:-1]
+        ci2 = ci[1:]
+        # but if there are no NAs present, this should raise KeyError
+        msg = "not in index"
+        with pytest.raises(KeyError, match=msg):
+            ser2.loc[box(ci2)]
+
+        with pytest.raises(KeyError, match=msg):
+            ser2[box(ci2)]
+
+        with pytest.raises(KeyError, match=msg):
+            ser2.to_frame().loc[box(ci2)]
+
+    def test_loc_getitem_series_label_list_missing_values(self):
+        # gh-11428
+        key = np.array(
+            ["2001-01-04", "2001-01-02", "2001-01-04", "2001-01-14"], dtype="datetime64"
+        )
+        ser = Series([2, 5, 8, 11], date_range("2001-01-01", freq="D", periods=4))
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[key]
+
+    def test_loc_getitem_series_label_list_missing_integer_values(self):
+        # GH: 25927
+        ser = Series(
+            index=np.array([9730701000001104, 10049011000001109]),
+            data=np.array([999000011000001104, 999000011000001104]),
+        )
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[np.array([9730701000001104, 10047311000001102])]
+
+    @pytest.mark.parametrize("to_period", [True, False])
+    def test_loc_getitem_listlike_of_datetimelike_keys(self, to_period):
+        # GH#11497
+
+        idx = date_range("2011-01-01", "2011-01-02", freq="D", name="idx")
+        if to_period:
+            idx = idx.to_period("D")
+        ser = Series([0.1, 0.2], index=idx, name="s")
+
+        keys = [Timestamp("2011-01-01"), Timestamp("2011-01-02")]
+        if to_period:
+            keys = [x.to_period("D") for x in keys]
+        result = ser.loc[keys]
+        exp = Series([0.1, 0.2], index=idx, name="s")
+        if not to_period:
+            exp.index = exp.index._with_freq(None)
+        tm.assert_series_equal(result, exp, check_index_type=True)
+
+        keys = [
+            Timestamp("2011-01-02"),
+            Timestamp("2011-01-02"),
+            Timestamp("2011-01-01"),
+        ]
+        if to_period:
+            keys = [x.to_period("D") for x in keys]
+        exp = Series(
+            [0.2, 0.2, 0.1], index=Index(keys, name="idx", dtype=idx.dtype), name="s"
+        )
+        result = ser.loc[keys]
+        tm.assert_series_equal(result, exp, check_index_type=True)
+
+        keys = [
+            Timestamp("2011-01-03"),
+            Timestamp("2011-01-02"),
+            Timestamp("2011-01-03"),
+        ]
+        if to_period:
+            keys = [x.to_period("D") for x in keys]
+
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[keys]
+
+    def test_loc_named_index(self):
+        # GH 42790
+        df = DataFrame(
+            [[1, 2], [4, 5], [7, 8]],
+            index=["cobra", "viper", "sidewinder"],
+            columns=["max_speed", "shield"],
+        )
+        expected = df.iloc[:2]
+        expected.index.name = "foo"
+        result = df.loc[Index(["cobra", "viper"], name="foo")]
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "columns, column_key, expected_columns",
+    [
+        ([2011, 2012, 2013], [2011, 2012], [0, 1]),
+        ([2011, 2012, "All"], [2011, 2012], [0, 1]),
+        ([2011, 2012, "All"], [2011, "All"], [0, 2]),
+    ],
+)
+def test_loc_getitem_label_list_integer_labels(columns, column_key, expected_columns):
+    # gh-14836
+    df = DataFrame(
+        np.random.default_rng(2).random((3, 3)), columns=columns, index=list("ABC")
+    )
+    expected = df.iloc[:, expected_columns]
+    result = df.loc[["A", "B", "C"], column_key]
+
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+
+def test_loc_setitem_float_intindex():
+    # GH 8720
+    rand_data = np.random.default_rng(2).standard_normal((8, 4))
+    result = DataFrame(rand_data)
+    result.loc[:, 0.5] = np.nan
+    expected_data = np.hstack((rand_data, np.array([np.nan] * 8).reshape(8, 1)))
+    expected = DataFrame(expected_data, columns=[0.0, 1.0, 2.0, 3.0, 0.5])
+    tm.assert_frame_equal(result, expected)
+
+    result = DataFrame(rand_data)
+    result.loc[:, 0.5] = np.nan
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_axis_1_slice():
+    # GH 10586
+    cols = [(yr, m) for yr in [2014, 2015] for m in [7, 8, 9, 10]]
+    df = DataFrame(
+        np.ones((10, 8)),
+        index=tuple("ABCDEFGHIJ"),
+        columns=MultiIndex.from_tuples(cols),
+    )
+    result = df.loc(axis=1)[(2014, 9) : (2015, 8)]
+    expected = DataFrame(
+        np.ones((10, 4)),
+        index=tuple("ABCDEFGHIJ"),
+        columns=MultiIndex.from_tuples([(2014, 9), (2014, 10), (2015, 7), (2015, 8)]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_set_dataframe_multiindex():
+    # GH 14592
+    expected = DataFrame(
+        "a", index=range(2), columns=MultiIndex.from_product([range(2), range(2)])
+    )
+    result = expected.copy()
+    result.loc[0, [(0, 1)]] = result.loc[0, [(0, 1)]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_loc_mixed_int_float():
+    # GH#19456
+    ser = Series(range(2), Index([1, 2.0], dtype=object))
+
+    result = ser.loc[1]
+    assert result == 0
+
+
+def test_loc_with_positional_slice_raises():
+    # GH#31840
+    ser = Series(range(4), index=["A", "B", "C", "D"])
+
+    with pytest.raises(TypeError, match="Slicing a positional slice with .loc"):
+        ser.loc[:3] = 2
+
+
+def test_loc_slice_disallows_positional():
+    # GH#16121, GH#24612, GH#31810
+    dti = date_range("2016-01-01", periods=3)
+    df = DataFrame(np.random.default_rng(2).random((3, 2)), index=dti)
+
+    ser = df[0]
+
+    msg = (
+        "cannot do slice indexing on DatetimeIndex with these "
+        r"indexers \[1\] of type int"
+    )
+
+    for obj in [df, ser]:
+        with pytest.raises(TypeError, match=msg):
+            obj.loc[1:3]
+
+        with pytest.raises(TypeError, match="Slicing a positional slice with .loc"):
+            # GH#31840 enforce incorrect behavior
+            obj.loc[1:3] = 1
+
+    with pytest.raises(TypeError, match=msg):
+        df.loc[1:3, 1]
+
+    with pytest.raises(TypeError, match="Slicing a positional slice with .loc"):
+        # GH#31840 enforce incorrect behavior
+        df.loc[1:3, 1] = 2
+
+
+def test_loc_datetimelike_mismatched_dtypes():
+    # GH#32650 dont mix and match datetime/timedelta/period dtypes
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 3)),
+        columns=["a", "b", "c"],
+        index=date_range("2012", freq="h", periods=5),
+    )
+    # create dataframe with non-unique DatetimeIndex
+    df = df.iloc[[0, 2, 2, 3]].copy()
+
+    dti = df.index
+    tdi = pd.TimedeltaIndex(dti.asi8)  # matching i8 values
+
+    msg = r"None of \[TimedeltaIndex.* are in the \[index\]"
+    with pytest.raises(KeyError, match=msg):
+        df.loc[tdi]
+
+    with pytest.raises(KeyError, match=msg):
+        df["a"].loc[tdi]
+
+
+def test_loc_with_period_index_indexer():
+    # GH#4125
+    idx = pd.period_range("2002-01", "2003-12", freq="M")
+    df = DataFrame(np.random.default_rng(2).standard_normal((24, 10)), index=idx)
+    tm.assert_frame_equal(df, df.loc[idx])
+    tm.assert_frame_equal(df, df.loc[list(idx)])
+    tm.assert_frame_equal(df, df.loc[list(idx)])
+    tm.assert_frame_equal(df.iloc[0:5], df.loc[idx[0:5]])
+    tm.assert_frame_equal(df, df.loc[list(idx)])
+
+
+def test_loc_setitem_multiindex_timestamp():
+    # GH#13831
+    vals = np.random.default_rng(2).standard_normal((8, 6))
+    idx = date_range("1/1/2000", periods=8)
+    cols = ["A", "B", "C", "D", "E", "F"]
+    exp = DataFrame(vals, index=idx, columns=cols)
+    exp.loc[exp.index[1], ("A", "B")] = np.nan
+    vals[1][0:2] = np.nan
+    res = DataFrame(vals, index=idx, columns=cols)
+    tm.assert_frame_equal(res, exp)
+
+
+def test_loc_getitem_multiindex_tuple_level():
+    # GH#27591
+    lev1 = ["a", "b", "c"]
+    lev2 = [(0, 1), (1, 0)]
+    lev3 = [0, 1]
+    cols = MultiIndex.from_product([lev1, lev2, lev3], names=["x", "y", "z"])
+    df = DataFrame(6, index=range(5), columns=cols)
+
+    # the lev2[0] here should be treated as a single label, not as a sequence
+    #  of labels
+    result = df.loc[:, (lev1[0], lev2[0], lev3[0])]
+
+    # TODO: i think this actually should drop levels
+    expected = df.iloc[:, :1]
+    tm.assert_frame_equal(result, expected)
+
+    alt = df.xs((lev1[0], lev2[0], lev3[0]), level=[0, 1, 2], axis=1)
+    tm.assert_frame_equal(alt, expected)
+
+    # same thing on a Series
+    ser = df.iloc[0]
+    expected2 = ser.iloc[:1]
+
+    alt2 = ser.xs((lev1[0], lev2[0], lev3[0]), level=[0, 1, 2], axis=0)
+    tm.assert_series_equal(alt2, expected2)
+
+    result2 = ser.loc[lev1[0], lev2[0], lev3[0]]
+    assert result2 == 6
+
+
+def test_loc_getitem_nullable_index_with_duplicates():
+    # GH#34497
+    df = DataFrame(
+        data=np.array([[1, 2, 3, 4], [5, 6, 7, 8], [1, 2, np.nan, np.nan]]).T,
+        columns=["a", "b", "c"],
+        dtype="Int64",
+    )
+    df2 = df.set_index("c")
+    assert df2.index.dtype == "Int64"
+
+    res = df2.loc[1]
+    expected = Series([1, 5], index=df2.columns, dtype="Int64", name=1)
+    tm.assert_series_equal(res, expected)
+
+    # pd.NA and duplicates in an object-dtype Index
+    df2.index = df2.index.astype(object)
+    res = df2.loc[1]
+    tm.assert_series_equal(res, expected)
+
+
+@pytest.mark.parametrize("value", [300, np.uint16(300), np.int16(300)])
+def test_loc_setitem_uint8_upcast(value):
+    # GH#26049
+
+    df = DataFrame([1, 2, 3, 4], columns=["col1"], dtype="uint8")
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.loc[2, "col1"] = value  # value that can't be held in uint8
+
+
+@pytest.mark.parametrize(
+    "fill_val,exp_dtype",
+    [
+        (Timestamp("2022-01-06"), "datetime64[ns]"),
+        (Timestamp("2022-01-07", tz="US/Eastern"), "datetime64[ns, US/Eastern]"),
+    ],
+)
+def test_loc_setitem_using_datetimelike_str_as_index(fill_val, exp_dtype):
+    data = ["2022-01-02", "2022-01-03", "2022-01-04", fill_val.date()]
+    index = DatetimeIndex(data, tz=fill_val.tz, dtype=exp_dtype)
+    df = DataFrame([10, 11, 12, 14], columns=["a"], index=index)
+    # adding new row using an unexisting datetime-like str index
+    df.loc["2022-01-08", "a"] = 13
+
+    data.append("2022-01-08")
+    expected_index = DatetimeIndex(data, dtype=exp_dtype)
+    tm.assert_index_equal(df.index, expected_index, exact=True)
+
+
+def test_loc_set_int_dtype():
+    # GH#23326
+    df = DataFrame([list("abc")])
+    df.loc[:, "col1"] = 5
+
+    expected = DataFrame({0: ["a"], 1: ["b"], 2: ["c"], "col1": [5]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.filterwarnings(r"ignore:Period with BDay freq is deprecated:FutureWarning")
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_loc_periodindex_3_levels():
+    # GH#24091
+    p_index = PeriodIndex(
+        ["20181101 1100", "20181101 1200", "20181102 1300", "20181102 1400"],
+        name="datetime",
+        freq="B",
+    )
+    mi_series = DataFrame(
+        [["A", "B", 1.0], ["A", "C", 2.0], ["Z", "Q", 3.0], ["W", "F", 4.0]],
+        index=p_index,
+        columns=["ONE", "TWO", "VALUES"],
+    )
+    mi_series = mi_series.set_index(["ONE", "TWO"], append=True)["VALUES"]
+    assert mi_series.loc[(p_index[0], "A", "B")] == 1.0
+
+
+def test_loc_setitem_pyarrow_strings():
+    # GH#52319
+    pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {
+            "strings": Series(["A", "B", "C"], dtype="string[pyarrow]"),
+            "ids": Series([True, True, False]),
+        }
+    )
+    new_value = Series(["X", "Y"])
+    df.loc[df.ids, "strings"] = new_value
+
+    expected_df = DataFrame(
+        {
+            "strings": Series(["X", "Y", "C"], dtype="string[pyarrow]"),
+            "ids": Series([True, True, False]),
+        }
+    )
+
+    tm.assert_frame_equal(df, expected_df)
+
+
+class TestLocSeries:
+    @pytest.mark.parametrize("val,expected", [(2**63 - 1, 3), (2**63, 4)])
+    def test_loc_uint64(self, val, expected):
+        # see GH#19399
+        ser = Series({2**63 - 1: 3, 2**63: 4})
+        assert ser.loc[val] == expected
+
+    def test_loc_getitem(self, string_series, datetime_series):
+        inds = string_series.index[[3, 4, 7]]
+        tm.assert_series_equal(string_series.loc[inds], string_series.reindex(inds))
+        tm.assert_series_equal(string_series.iloc[5::2], string_series[5::2])
+
+        # slice with indices
+        d1, d2 = datetime_series.index[[5, 15]]
+        result = datetime_series.loc[d1:d2]
+        expected = datetime_series.truncate(d1, d2)
+        tm.assert_series_equal(result, expected)
+
+        # boolean
+        mask = string_series > string_series.median()
+        tm.assert_series_equal(string_series.loc[mask], string_series[mask])
+
+        # ask for index value
+        assert datetime_series.loc[d1] == datetime_series[d1]
+        assert datetime_series.loc[d2] == datetime_series[d2]
+
+    def test_loc_getitem_not_monotonic(self, datetime_series):
+        d1, d2 = datetime_series.index[[5, 15]]
+
+        ts2 = datetime_series[::2].iloc[[1, 2, 0]]
+
+        msg = r"Timestamp\('2000-01-10 00:00:00'\)"
+        with pytest.raises(KeyError, match=msg):
+            ts2.loc[d1:d2]
+        with pytest.raises(KeyError, match=msg):
+            ts2.loc[d1:d2] = 0
+
+    def test_loc_getitem_setitem_integer_slice_keyerrors(self):
+        ser = Series(
+            np.random.default_rng(2).standard_normal(10), index=list(range(0, 20, 2))
+        )
+
+        # this is OK
+        cp = ser.copy()
+        cp.iloc[4:10] = 0
+        assert (cp.iloc[4:10] == 0).all()
+
+        # so is this
+        cp = ser.copy()
+        cp.iloc[3:11] = 0
+        assert (cp.iloc[3:11] == 0).values.all()
+
+        result = ser.iloc[2:6]
+        result2 = ser.loc[3:11]
+        expected = ser.reindex([4, 6, 8, 10])
+
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
+
+        # non-monotonic, raise KeyError
+        s2 = ser.iloc[list(range(5)) + list(range(9, 4, -1))]
+        with pytest.raises(KeyError, match=r"^3$"):
+            s2.loc[3:11]
+        with pytest.raises(KeyError, match=r"^3$"):
+            s2.loc[3:11] = 0
+
+    def test_loc_getitem_iterator(self, string_series):
+        idx = iter(string_series.index[:10])
+        result = string_series.loc[idx]
+        tm.assert_series_equal(result, string_series[:10])
+
+    def test_loc_setitem_boolean(self, string_series):
+        mask = string_series > string_series.median()
+
+        result = string_series.copy()
+        result.loc[mask] = 0
+        expected = string_series
+        expected[mask] = 0
+        tm.assert_series_equal(result, expected)
+
+    def test_loc_setitem_corner(self, string_series):
+        inds = list(string_series.index[[5, 8, 12]])
+        string_series.loc[inds] = 5
+        msg = r"\['foo'\] not in index"
+        with pytest.raises(KeyError, match=msg):
+            string_series.loc[inds + ["foo"]] = 5
+
+    def test_basic_setitem_with_labels(self, datetime_series):
+        indices = datetime_series.index[[5, 10, 15]]
+
+        cp = datetime_series.copy()
+        exp = datetime_series.copy()
+        cp[indices] = 0
+        exp.loc[indices] = 0
+        tm.assert_series_equal(cp, exp)
+
+        cp = datetime_series.copy()
+        exp = datetime_series.copy()
+        cp[indices[0] : indices[2]] = 0
+        exp.loc[indices[0] : indices[2]] = 0
+        tm.assert_series_equal(cp, exp)
+
+    def test_loc_setitem_listlike_of_ints(self):
+        # integer indexes, be careful
+        ser = Series(
+            np.random.default_rng(2).standard_normal(10), index=list(range(0, 20, 2))
+        )
+        inds = [0, 4, 6]
+        arr_inds = np.array([0, 4, 6])
+
+        cp = ser.copy()
+        exp = ser.copy()
+        ser[inds] = 0
+        ser.loc[inds] = 0
+        tm.assert_series_equal(cp, exp)
+
+        cp = ser.copy()
+        exp = ser.copy()
+        ser[arr_inds] = 0
+        ser.loc[arr_inds] = 0
+        tm.assert_series_equal(cp, exp)
+
+        inds_notfound = [0, 4, 5, 6]
+        arr_inds_notfound = np.array([0, 4, 5, 6])
+        msg = r"\[5\] not in index"
+        with pytest.raises(KeyError, match=msg):
+            ser[inds_notfound] = 0
+        with pytest.raises(Exception, match=msg):
+            ser[arr_inds_notfound] = 0
+
+    def test_loc_setitem_dt64tz_values(self):
+        # GH#12089
+        ser = Series(
+            date_range("2011-01-01", periods=3, tz="US/Eastern"),
+            index=["a", "b", "c"],
+        )
+        s2 = ser.copy()
+        expected = Timestamp("2011-01-03", tz="US/Eastern")
+        s2.loc["a"] = expected
+        result = s2.loc["a"]
+        assert result == expected
+
+        s2 = ser.copy()
+        s2.iloc[0] = expected
+        result = s2.iloc[0]
+        assert result == expected
+
+        s2 = ser.copy()
+        s2["a"] = expected
+        result = s2["a"]
+        assert result == expected
+
+    @pytest.mark.parametrize("array_fn", [np.array, pd.array, list, tuple])
+    @pytest.mark.parametrize("size", [0, 4, 5, 6])
+    def test_loc_iloc_setitem_with_listlike(self, size, array_fn):
+        # GH37748
+        # testing insertion, in a Series of size N (here 5), of a listlike object
+        # of size  0, N-1, N, N+1
+
+        arr = array_fn([0] * size)
+        expected = Series([arr, 0, 0, 0, 0], index=list("abcde"), dtype=object)
+
+        ser = Series(0, index=list("abcde"), dtype=object)
+        ser.loc["a"] = arr
+        tm.assert_series_equal(ser, expected)
+
+        ser = Series(0, index=list("abcde"), dtype=object)
+        ser.iloc[0] = arr
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize("indexer", [IndexSlice["A", :], ("A", slice(None))])
+    def test_loc_series_getitem_too_many_dimensions(self, indexer):
+        # GH#35349
+        ser = Series(
+            index=MultiIndex.from_tuples([("A", "0"), ("A", "1"), ("B", "0")]),
+            data=[21, 22, 23],
+        )
+        msg = "Too many indexers"
+        with pytest.raises(IndexingError, match=msg):
+            ser.loc[indexer, :]
+
+        with pytest.raises(IndexingError, match=msg):
+            ser.loc[indexer, :] = 1
+
+    def test_loc_setitem(self, string_series):
+        inds = string_series.index[[3, 4, 7]]
+
+        result = string_series.copy()
+        result.loc[inds] = 5
+
+        expected = string_series.copy()
+        expected.iloc[[3, 4, 7]] = 5
+        tm.assert_series_equal(result, expected)
+
+        result.iloc[5:10] = 10
+        expected[5:10] = 10
+        tm.assert_series_equal(result, expected)
+
+        # set slice with indices
+        d1, d2 = string_series.index[[5, 15]]
+        result.loc[d1:d2] = 6
+        expected[5:16] = 6  # because it's inclusive
+        tm.assert_series_equal(result, expected)
+
+        # set index value
+        string_series.loc[d1] = 4
+        string_series.loc[d2] = 6
+        assert string_series[d1] == 4
+        assert string_series[d2] == 6
+
+    @pytest.mark.parametrize("dtype", ["object", "string"])
+    def test_loc_assign_dict_to_row(self, dtype):
+        # GH41044
+        df = DataFrame({"A": ["abc", "def"], "B": ["ghi", "jkl"]}, dtype=dtype)
+        df.loc[0, :] = {"A": "newA", "B": "newB"}
+
+        expected = DataFrame({"A": ["newA", "def"], "B": ["newB", "jkl"]}, dtype=dtype)
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_dict_timedelta_multiple_set(self):
+        # GH 16309
+        result = DataFrame(columns=["time", "value"])
+        result.loc[1] = {"time": Timedelta(6, unit="s"), "value": "foo"}
+        result.loc[1] = {"time": Timedelta(6, unit="s"), "value": "foo"}
+        expected = DataFrame(
+            [[Timedelta(6, unit="s"), "foo"]], columns=["time", "value"], index=[1]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_set_multiple_items_in_multiple_new_columns(self):
+        # GH 25594
+        df = DataFrame(index=[1, 2], columns=["a"])
+        df.loc[1, ["b", "c"]] = [6, 7]
+
+        expected = DataFrame(
+            {
+                "a": Series([np.nan, np.nan], dtype="object"),
+                "b": [6, np.nan],
+                "c": [7, np.nan],
+            },
+            index=[1, 2],
+        )
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_getitem_loc_str_periodindex(self):
+        # GH#33964
+        msg = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            index = pd.period_range(start="2000", periods=20, freq="B")
+            series = Series(range(20), index=index)
+            assert series.loc["2000-01-14"] == 9
+
+    def test_loc_nonunique_masked_index(self):
+        # GH 57027
+        ids = list(range(11))
+        index = Index(ids * 1000, dtype="Int64")
+        df = DataFrame({"val": np.arange(len(index), dtype=np.intp)}, index=index)
+        result = df.loc[ids]
+        expected = DataFrame(
+            {"val": index.argsort(kind="stable").astype(np.intp)},
+            index=Index(np.array(ids).repeat(1000), dtype="Int64"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_loc_index_alignment_for_series(self):
+        # GH #56024
+        df = DataFrame({"a": [1, 2], "b": [3, 4]})
+        other = Series([200, 999], index=[1, 0])
+        df.loc[:, "a"] = other
+        expected = DataFrame({"a": [999, 200], "b": [3, 4]})
+        tm.assert_frame_equal(expected, df)
+
+    def test_loc_reindexing_of_empty_index(self):
+        # GH 57735
+        df = DataFrame(index=[1, 1, 2, 2], data=["1", "1", "2", "2"])
+        df.loc[Series([False] * 4, index=df.index, name=0), 0] = df[0]
+        expected = DataFrame(index=[1, 1, 2, 2], data=["1", "1", "2", "2"])
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_matching_index(self):
+        # GH 25548
+        s = Series(0.0, index=list("abcd"))
+        s1 = Series(1.0, index=list("ab"))
+        s2 = Series(2.0, index=list("xy"))
+
+        # Test matching indices
+        s.loc[["a", "b"]] = s1
+
+        result = s[["a", "b"]]
+        expected = s1
+        tm.assert_series_equal(result, expected)
+
+        # Test unmatched indices
+        s.loc[["a", "b"]] = s2
+
+        result = s[["a", "b"]]
+        expected = Series([np.nan, np.nan], index=["a", "b"])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/test_na_indexing.py b/pandas/pandas/tests/indexing/test_na_indexing.py
new file mode 100644
index 0000000000000..d4ad350a64e4d
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_na_indexing.py
@@ -0,0 +1,74 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "values, dtype",
+    [
+        ([], "object"),
+        ([1, 2, 3], "int64"),
+        ([1.0, 2.0, 3.0], "float64"),
+        (["a", "b", "c"], "object"),
+        (["a", "b", "c"], "string"),
+        ([1, 2, 3], "datetime64[ns]"),
+        ([1, 2, 3], "datetime64[ns, CET]"),
+        ([1, 2, 3], "timedelta64[ns]"),
+        (["2000", "2001", "2002"], "Period[D]"),
+        ([1, 0, 3], "Sparse"),
+        ([pd.Interval(0, 1), pd.Interval(1, 2), pd.Interval(3, 4)], "interval"),
+    ],
+)
+@pytest.mark.parametrize(
+    "mask", [[True, False, False], [True, True, True], [False, False, False]]
+)
+@pytest.mark.parametrize("indexer_class", [list, pd.array, pd.Index, pd.Series])
+@pytest.mark.parametrize("frame", [True, False])
+def test_series_mask_boolean(values, dtype, mask, indexer_class, frame):
+    # In case len(values) < 3
+    index = ["a", "b", "c"][: len(values)]
+    mask = mask[: len(values)]
+
+    obj = pd.Series(values, dtype=dtype, index=index)
+    if frame:
+        if len(values) == 0:
+            # Otherwise obj is an empty DataFrame with shape (0, 1)
+            obj = pd.DataFrame(dtype=dtype, index=index)
+        else:
+            obj = obj.to_frame()
+
+    if indexer_class is pd.array:
+        mask = pd.array(mask, dtype="boolean")
+    elif indexer_class is pd.Series:
+        mask = pd.Series(mask, index=obj.index, dtype="boolean")
+    else:
+        mask = indexer_class(mask)
+
+    expected = obj[mask]
+
+    result = obj[mask]
+    tm.assert_equal(result, expected)
+
+    if indexer_class is pd.Series:
+        msg = "iLocation based boolean indexing cannot use an indexable as a mask"
+        with pytest.raises(ValueError, match=msg):
+            result = obj.iloc[mask]
+    else:
+        result = obj.iloc[mask]
+        tm.assert_equal(result, expected)
+
+    result = obj.loc[mask]
+    tm.assert_equal(result, expected)
+
+
+def test_na_treated_as_false(frame_or_series, indexer_sli):
+    # https://github.com/pandas-dev/pandas/issues/31503
+    obj = frame_or_series([1, 2, 3])
+
+    mask = pd.array([True, False, None], dtype="boolean")
+
+    result = indexer_sli(obj)[mask]
+    expected = indexer_sli(obj)[mask.fillna(False)]
+
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/test_partial.py b/pandas/pandas/tests/indexing/test_partial.py
new file mode 100644
index 0000000000000..6f20d0e4e7cbf
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_partial.py
@@ -0,0 +1,694 @@
+"""
+test setting *parts* of objects both positionally and label based
+
+TODO: these should be split among the indexer tests
+"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Period,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestEmptyFrameSetitemExpansion:
+    def test_empty_frame_setitem_index_name_retained(self):
+        # GH#31368 empty frame has non-None index.name -> retained
+        df = DataFrame({}, index=pd.RangeIndex(0, name="df_index"))
+        series = Series(1.23, index=pd.RangeIndex(4, name="series_index"))
+
+        df["series"] = series
+        expected = DataFrame(
+            {"series": [1.23] * 4},
+            index=pd.RangeIndex(4, name="df_index"),
+            columns=Index(["series"]),
+        )
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_empty_frame_setitem_index_name_inherited(self):
+        # GH#36527 empty frame has None index.name -> not retained
+        df = DataFrame()
+        series = Series(1.23, index=pd.RangeIndex(4, name="series_index"))
+        df["series"] = series
+        expected = DataFrame(
+            {"series": [1.23] * 4},
+            index=pd.RangeIndex(4, name="series_index"),
+            columns=Index(["series"]),
+        )
+        tm.assert_frame_equal(df, expected)
+
+    def test_loc_setitem_zerolen_series_columns_align(self):
+        # columns will align
+        df = DataFrame(columns=["A", "B"])
+        df.loc[0] = Series(1, index=range(4))
+        expected = DataFrame(columns=["A", "B"], index=[0], dtype=np.float64)
+        tm.assert_frame_equal(df, expected)
+
+        # columns will align
+        df = DataFrame(columns=["A", "B"])
+        df.loc[0] = Series(1, index=["B"])
+
+        exp = DataFrame([[np.nan, 1]], columns=["A", "B"], index=[0], dtype="float64")
+        tm.assert_frame_equal(df, exp)
+
+    def test_loc_setitem_zerolen_list_length_must_match_columns(self):
+        # list-like must conform
+        df = DataFrame(columns=["A", "B"])
+
+        msg = "cannot set a row with mismatched columns"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[0] = [1, 2, 3]
+
+        df = DataFrame(columns=["A", "B"])
+        df.loc[3] = [6, 7]  # length matches len(df.columns) --> OK!
+
+        exp = DataFrame([[6, 7]], index=[3], columns=["A", "B"], dtype=np.int64)
+        tm.assert_frame_equal(df, exp)
+
+    def test_partial_set_empty_frame(self):
+        # partially set with an empty object
+        # frame
+        df = DataFrame()
+
+        msg = "cannot set a frame with no defined columns"
+
+        with pytest.raises(ValueError, match=msg):
+            df.loc[1] = 1
+
+        with pytest.raises(ValueError, match=msg):
+            df.loc[1] = Series([1], index=["foo"])
+
+        msg = "cannot set a frame with no defined index and a scalar"
+        with pytest.raises(ValueError, match=msg):
+            df.loc[:, 1] = 1
+
+    def test_partial_set_empty_frame2(self):
+        # these work as they don't really change
+        # anything but the index
+        # GH#5632
+        expected = DataFrame(columns=Index(["foo"]), index=Index([], dtype="object"))
+
+        df = DataFrame(index=Index([], dtype="object"))
+        df["foo"] = Series([], dtype="object")
+
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(index=Index([]))
+        df["foo"] = Series(df.index)
+
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(index=Index([]))
+        df["foo"] = df.index
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame3(self):
+        expected = DataFrame(columns=Index(["foo"]), index=Index([], dtype="int64"))
+        expected["foo"] = expected["foo"].astype("float64")
+
+        df = DataFrame(index=Index([], dtype="int64"))
+        df["foo"] = []
+
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(index=Index([], dtype="int64"))
+        df["foo"] = Series(np.arange(len(df)), dtype="float64")
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame4(self):
+        df = DataFrame(index=Index([], dtype="int64"))
+        df["foo"] = range(len(df))
+
+        expected = DataFrame(columns=Index(["foo"]), index=Index([], dtype="int64"))
+        # range is int-dtype-like, so we get int64 dtype
+        expected["foo"] = expected["foo"].astype("int64")
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame5(self):
+        df = DataFrame()
+        tm.assert_index_equal(df.columns, pd.RangeIndex(0))
+        df2 = DataFrame()
+        df2[1] = Series([1], index=["foo"])
+        df.loc[:, 1] = Series([1], index=["foo"])
+        tm.assert_frame_equal(df, DataFrame([[1]], index=["foo"], columns=[1]))
+        tm.assert_frame_equal(df, df2)
+
+    def test_partial_set_empty_frame_no_index(self):
+        # no index to start
+        expected = DataFrame({0: Series(1, index=range(4))}, columns=["A", "B", 0])
+
+        df = DataFrame(columns=["A", "B"])
+        df[0] = Series(1, index=range(4))
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(columns=["A", "B"])
+        df.loc[:, 0] = Series(1, index=range(4))
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame_row(self):
+        # GH#5720, GH#5744
+        # don't create rows when empty
+        expected = DataFrame(columns=["A", "B", "New"], index=Index([], dtype="int64"))
+        expected["A"] = expected["A"].astype("int64")
+        expected["B"] = expected["B"].astype("float64")
+        expected["New"] = expected["New"].astype("float64")
+
+        df = DataFrame({"A": [1, 2, 3], "B": [1.2, 4.2, 5.2]})
+        y = df[df.A > 5]
+        y["New"] = np.nan
+        tm.assert_frame_equal(y, expected)
+
+        expected = DataFrame(columns=["a", "b", "c c", "d"])
+        expected["d"] = expected["d"].astype("int64")
+        df = DataFrame(columns=["a", "b", "c c"])
+        df["d"] = 3
+        tm.assert_frame_equal(df, expected)
+        tm.assert_series_equal(df["c c"], Series(name="c c", dtype=object))
+
+        # reindex columns is ok
+        df = DataFrame({"A": [1, 2, 3], "B": [1.2, 4.2, 5.2]})
+        y = df[df.A > 5]
+        result = y.reindex(columns=["A", "B", "C"])
+        expected = DataFrame(columns=["A", "B", "C"])
+        expected["A"] = expected["A"].astype("int64")
+        expected["B"] = expected["B"].astype("float64")
+        expected["C"] = expected["C"].astype("float64")
+        tm.assert_frame_equal(result, expected)
+
+    def test_partial_set_empty_frame_set_series(self):
+        # GH#5756
+        # setting with empty Series
+        df = DataFrame(Series(dtype=object))
+        expected = DataFrame({0: Series(dtype=object)})
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(Series(name="foo", dtype=object))
+        expected = DataFrame({"foo": Series(dtype=object)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame_empty_copy_assignment(self):
+        # GH#5932
+        # copy on empty with assignment fails
+        df = DataFrame(index=[0])
+        df = df.copy()
+        df["a"] = 0
+        expected = DataFrame(0, index=[0], columns=Index(["a"]))
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_empty_frame_empty_consistencies(self, using_infer_string):
+        # GH#6171
+        # consistency on empty frames
+        df = DataFrame(columns=["x", "y"])
+        df["x"] = [1, 2]
+        expected = DataFrame({"x": [1, 2], "y": [np.nan, np.nan]})
+        tm.assert_frame_equal(df, expected, check_dtype=False)
+
+        df = DataFrame(columns=["x", "y"])
+        df["x"] = ["1", "2"]
+        expected = DataFrame(
+            {
+                "x": Series(
+                    ["1", "2"],
+                    dtype=object if not using_infer_string else "str",
+                ),
+                "y": Series([np.nan, np.nan], dtype=object),
+            }
+        )
+        tm.assert_frame_equal(df, expected)
+
+        df = DataFrame(columns=["x", "y"])
+        df.loc[0, "x"] = 1
+        expected = DataFrame({"x": [1], "y": [np.nan]})
+        tm.assert_frame_equal(df, expected, check_dtype=False)
+
+
+class TestPartialSetting:
+    def test_partial_setting(self):
+        # GH2578, allow ix and friends to partially set
+
+        # series
+        s_orig = Series([1, 2, 3])
+
+        s = s_orig.copy()
+        s[5] = 5
+        expected = Series([1, 2, 3, 5], index=[0, 1, 2, 5])
+        tm.assert_series_equal(s, expected)
+
+        s = s_orig.copy()
+        s.loc[5] = 5
+        expected = Series([1, 2, 3, 5], index=[0, 1, 2, 5])
+        tm.assert_series_equal(s, expected)
+
+        s = s_orig.copy()
+        s[5] = 5.0
+        expected = Series([1, 2, 3, 5.0], index=[0, 1, 2, 5])
+        tm.assert_series_equal(s, expected)
+
+        s = s_orig.copy()
+        s.loc[5] = 5.0
+        expected = Series([1, 2, 3, 5.0], index=[0, 1, 2, 5])
+        tm.assert_series_equal(s, expected)
+
+        # iloc/iat raise
+        s = s_orig.copy()
+
+        msg = "iloc cannot enlarge its target object"
+        with pytest.raises(IndexError, match=msg):
+            s.iloc[3] = 5.0
+
+        msg = "index 3 is out of bounds for axis 0 with size 3"
+        with pytest.raises(IndexError, match=msg):
+            s.iat[3] = 5.0
+
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+    def test_partial_setting_frame(self):
+        df_orig = DataFrame(
+            np.arange(6).reshape(3, 2), columns=["A", "B"], dtype="int64"
+        )
+
+        # iloc/iat raise
+        df = df_orig.copy()
+
+        msg = "iloc cannot enlarge its target object"
+        with pytest.raises(IndexError, match=msg):
+            df.iloc[4, 2] = 5.0
+
+        msg = "index 2 is out of bounds for axis 0 with size 2"
+        with pytest.raises(IndexError, match=msg):
+            df.iat[4, 2] = 5.0
+
+        # row setting where it exists
+        expected = DataFrame({"A": [0, 4, 4], "B": [1, 5, 5]})
+        df = df_orig.copy()
+        df.iloc[1] = df.iloc[2]
+        tm.assert_frame_equal(df, expected)
+
+        expected = DataFrame({"A": [0, 4, 4], "B": [1, 5, 5]})
+        df = df_orig.copy()
+        df.loc[1] = df.loc[2]
+        tm.assert_frame_equal(df, expected)
+
+        # like 2578, partial setting with dtype preservation
+        expected = DataFrame({"A": [0, 2, 4, 4], "B": [1, 3, 5, 5]})
+        df = df_orig.copy()
+        df.loc[3] = df.loc[2]
+        tm.assert_frame_equal(df, expected)
+
+        # single dtype frame, overwrite
+        expected = DataFrame({"A": [0, 2, 4], "B": [0, 2, 4]})
+        df = df_orig.copy()
+        df.loc[:, "B"] = df.loc[:, "A"]
+        tm.assert_frame_equal(df, expected)
+
+        # mixed dtype frame, overwrite
+        expected = DataFrame({"A": [0, 2, 4], "B": Series([0.0, 2.0, 4.0])})
+        df = df_orig.copy()
+        df["B"] = df["B"].astype(np.float64)
+        # as of 2.0, df.loc[:, "B"] = ... attempts (and here succeeds) at
+        #  setting inplace
+        df.loc[:, "B"] = df.loc[:, "A"]
+        tm.assert_frame_equal(df, expected)
+
+        # single dtype frame, partial setting
+        expected = df_orig.copy()
+        expected["C"] = df["A"]
+        df = df_orig.copy()
+        df.loc[:, "C"] = df.loc[:, "A"]
+        tm.assert_frame_equal(df, expected)
+
+        # mixed frame, partial setting
+        expected = df_orig.copy()
+        expected["C"] = df["A"]
+        df = df_orig.copy()
+        df.loc[:, "C"] = df.loc[:, "A"]
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_setting2(self):
+        # GH 8473
+        dates = date_range("1/1/2000", periods=8)
+        df_orig = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)),
+            index=dates,
+            columns=["A", "B", "C", "D"],
+        )
+
+        expected = pd.concat(
+            [df_orig, DataFrame({"A": 7}, index=dates[-1:] + dates.freq)], sort=True
+        )
+        df = df_orig.copy()
+        df.loc[dates[-1] + dates.freq, "A"] = 7
+        tm.assert_frame_equal(df, expected)
+        df = df_orig.copy()
+        df.at[dates[-1] + dates.freq, "A"] = 7
+        tm.assert_frame_equal(df, expected)
+
+        exp_other = DataFrame({0: 7}, index=dates[-1:] + dates.freq)
+        expected = pd.concat([df_orig, exp_other], axis=1)
+
+        df = df_orig.copy()
+        df.loc[dates[-1] + dates.freq, 0] = 7
+        tm.assert_frame_equal(df, expected)
+        df = df_orig.copy()
+        df.at[dates[-1] + dates.freq, 0] = 7
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_setting_mixed_dtype(self):
+        # in a mixed dtype environment, try to preserve dtypes
+        # by appending
+        df = DataFrame([[True, 1], [False, 2]], columns=["female", "fitness"])
+
+        s = df.loc[1].copy()
+        s.name = 2
+        expected = pd.concat([df, DataFrame(s).T.infer_objects()])
+
+        df.loc[2] = df.loc[1]
+        tm.assert_frame_equal(df, expected)
+
+    def test_series_partial_set(self):
+        # partial set with new index
+        # Regression from GH4825
+        ser = Series([0.1, 0.2], index=[1, 2])
+
+        # loc equiv to .reindex
+        expected = Series([np.nan, 0.2, np.nan], index=[3, 2, 3])
+        with pytest.raises(KeyError, match=r"not in index"):
+            ser.loc[[3, 2, 3]]
+
+        result = ser.reindex([3, 2, 3])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        expected = Series([np.nan, 0.2, np.nan, np.nan], index=[3, 2, 3, "x"])
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[[3, 2, 3, "x"]]
+
+        result = ser.reindex([3, 2, 3, "x"])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        expected = Series([0.2, 0.2, 0.1], index=[2, 2, 1])
+        result = ser.loc[[2, 2, 1]]
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        expected = Series([0.2, 0.2, np.nan, 0.1], index=[2, 2, "x", 1])
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[[2, 2, "x", 1]]
+
+        result = ser.reindex([2, 2, "x", 1])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        # raises as nothing is in the index
+        msg = (
+            rf"\"None of \[Index\(\[3, 3, 3\], dtype='{np.dtype(int)}'\)\] "
+            r"are in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            ser.loc[[3, 3, 3]]
+
+        expected = Series([0.2, 0.2, np.nan], index=[2, 2, 3])
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[[2, 2, 3]]
+
+        result = ser.reindex([2, 2, 3])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        s = Series([0.1, 0.2, 0.3], index=[1, 2, 3])
+        expected = Series([0.3, np.nan, np.nan], index=[3, 4, 4])
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[3, 4, 4]]
+
+        result = s.reindex([3, 4, 4])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        s = Series([0.1, 0.2, 0.3, 0.4], index=[1, 2, 3, 4])
+        expected = Series([np.nan, 0.3, 0.3], index=[5, 3, 3])
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[5, 3, 3]]
+
+        result = s.reindex([5, 3, 3])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        s = Series([0.1, 0.2, 0.3, 0.4], index=[1, 2, 3, 4])
+        expected = Series([np.nan, 0.4, 0.4], index=[5, 4, 4])
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[5, 4, 4]]
+
+        result = s.reindex([5, 4, 4])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        s = Series([0.1, 0.2, 0.3, 0.4], index=[4, 5, 6, 7])
+        expected = Series([0.4, np.nan, np.nan], index=[7, 2, 2])
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[7, 2, 2]]
+
+        result = s.reindex([7, 2, 2])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        s = Series([0.1, 0.2, 0.3, 0.4], index=[1, 2, 3, 4])
+        expected = Series([0.4, np.nan, np.nan], index=[4, 5, 5])
+        with pytest.raises(KeyError, match="not in index"):
+            s.loc[[4, 5, 5]]
+
+        result = s.reindex([4, 5, 5])
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        # iloc
+        expected = Series([0.2, 0.2, 0.1, 0.1], index=[2, 2, 1, 1])
+        result = ser.iloc[[1, 1, 0, 0]]
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_series_partial_set_with_name(self):
+        # GH 11497
+
+        idx = Index([1, 2], dtype="int64", name="idx")
+        ser = Series([0.1, 0.2], index=idx, name="s")
+
+        # loc
+        with pytest.raises(KeyError, match=r"\[3\] not in index"):
+            ser.loc[[3, 2, 3]]
+
+        with pytest.raises(KeyError, match=r"not in index"):
+            ser.loc[[3, 2, 3, "x"]]
+
+        exp_idx = Index([2, 2, 1], dtype="int64", name="idx")
+        expected = Series([0.2, 0.2, 0.1], index=exp_idx, name="s")
+        result = ser.loc[[2, 2, 1]]
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        with pytest.raises(KeyError, match=r"\['x'\] not in index"):
+            ser.loc[[2, 2, "x", 1]]
+
+        # raises as nothing is in the index
+        msg = (
+            rf"\"None of \[Index\(\[3, 3, 3\], dtype='{np.dtype(int)}', "
+            r"name='idx'\)\] are in the \[index\]\""
+        )
+        with pytest.raises(KeyError, match=msg):
+            ser.loc[[3, 3, 3]]
+
+        with pytest.raises(KeyError, match="not in index"):
+            ser.loc[[2, 2, 3]]
+
+        idx = Index([1, 2, 3], dtype="int64", name="idx")
+        with pytest.raises(KeyError, match="not in index"):
+            Series([0.1, 0.2, 0.3], index=idx, name="s").loc[[3, 4, 4]]
+
+        idx = Index([1, 2, 3, 4], dtype="int64", name="idx")
+        with pytest.raises(KeyError, match="not in index"):
+            Series([0.1, 0.2, 0.3, 0.4], index=idx, name="s").loc[[5, 3, 3]]
+
+        idx = Index([1, 2, 3, 4], dtype="int64", name="idx")
+        with pytest.raises(KeyError, match="not in index"):
+            Series([0.1, 0.2, 0.3, 0.4], index=idx, name="s").loc[[5, 4, 4]]
+
+        idx = Index([4, 5, 6, 7], dtype="int64", name="idx")
+        with pytest.raises(KeyError, match="not in index"):
+            Series([0.1, 0.2, 0.3, 0.4], index=idx, name="s").loc[[7, 2, 2]]
+
+        idx = Index([1, 2, 3, 4], dtype="int64", name="idx")
+        with pytest.raises(KeyError, match="not in index"):
+            Series([0.1, 0.2, 0.3, 0.4], index=idx, name="s").loc[[4, 5, 5]]
+
+        # iloc
+        exp_idx = Index([2, 2, 1, 1], dtype="int64", name="idx")
+        expected = Series([0.2, 0.2, 0.1, 0.1], index=exp_idx, name="s")
+        result = ser.iloc[[1, 1, 0, 0]]
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    @pytest.mark.parametrize("key", [100, 100.0])
+    def test_setitem_with_expansion_numeric_into_datetimeindex(self, key):
+        # GH#4940 inserting non-strings
+        orig = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df = orig.copy()
+
+        df.loc[key, :] = df.iloc[0]
+        ex_index = Index(list(orig.index) + [key], dtype=object, name=orig.index.name)
+        ex_data = np.concatenate([orig.values, df.iloc[[0]].values], axis=0)
+        expected = DataFrame(ex_data, index=ex_index, columns=orig.columns)
+
+        tm.assert_frame_equal(df, expected)
+
+    def test_partial_set_invalid(self):
+        # GH 4940
+        # allow only setting of 'valid' values
+
+        orig = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+
+        # allow object conversion here
+        df = orig.copy()
+        df.loc["a", :] = df.iloc[0]
+        ser = Series(df.iloc[0], name="a")
+        exp = pd.concat([orig, DataFrame(ser).T.infer_objects()])
+        tm.assert_frame_equal(df, exp)
+        tm.assert_index_equal(df.index, Index(orig.index.tolist() + ["a"]))
+        assert df.index.dtype == "object"
+
+    @pytest.mark.parametrize(
+        "idx,labels,expected_idx",
+        [
+            (
+                period_range(start="2000", periods=20, freq="D"),
+                ["2000-01-04", "2000-01-08", "2000-01-12"],
+                [
+                    Period("2000-01-04", freq="D"),
+                    Period("2000-01-08", freq="D"),
+                    Period("2000-01-12", freq="D"),
+                ],
+            ),
+            (
+                date_range(start="2000", periods=20, freq="D", unit="s"),
+                ["2000-01-04", "2000-01-08", "2000-01-12"],
+                [
+                    Timestamp("2000-01-04"),
+                    Timestamp("2000-01-08"),
+                    Timestamp("2000-01-12"),
+                ],
+            ),
+            (
+                pd.timedelta_range(start="1 day", periods=20),
+                ["4D", "8D", "12D"],
+                [pd.Timedelta("4 day"), pd.Timedelta("8 day"), pd.Timedelta("12 day")],
+            ),
+        ],
+    )
+    def test_loc_with_list_of_strings_representing_datetimes(
+        self, idx, labels, expected_idx, frame_or_series
+    ):
+        # GH 11278
+        obj = frame_or_series(range(20), index=idx)
+
+        expected_value = [3, 7, 11]
+        expected = frame_or_series(expected_value, expected_idx)
+
+        tm.assert_equal(expected, obj.loc[labels])
+        if frame_or_series is Series:
+            tm.assert_series_equal(expected, obj[labels])
+
+    @pytest.mark.parametrize(
+        "idx,labels",
+        [
+            (
+                period_range(start="2000", periods=20, freq="D"),
+                ["2000-01-04", "2000-01-30"],
+            ),
+            (
+                date_range(start="2000", periods=20, freq="D"),
+                ["2000-01-04", "2000-01-30"],
+            ),
+            (pd.timedelta_range(start="1 day", periods=20), ["3 day", "30 day"]),
+        ],
+    )
+    def test_loc_with_list_of_strings_representing_datetimes_missing_value(
+        self, idx, labels
+    ):
+        # GH 11278
+        ser = Series(range(20), index=idx)
+        df = DataFrame(range(20), index=idx)
+        msg = r"not in index"
+
+        with pytest.raises(KeyError, match=msg):
+            ser.loc[labels]
+        with pytest.raises(KeyError, match=msg):
+            ser[labels]
+        with pytest.raises(KeyError, match=msg):
+            df.loc[labels]
+
+    @pytest.mark.parametrize(
+        "idx,labels,msg",
+        [
+            (
+                period_range(start="2000", periods=20, freq="D"),
+                Index(["4D", "8D"], dtype=object),
+                (
+                    r"None of \[Index\(\['4D', '8D'\], dtype='object'\)\] "
+                    r"are in the \[index\]"
+                ),
+            ),
+            (
+                date_range(start="2000", periods=20, freq="D"),
+                Index(["4D", "8D"], dtype=object),
+                (
+                    r"None of \[Index\(\['4D', '8D'\], dtype='object'\)\] "
+                    r"are in the \[index\]"
+                ),
+            ),
+            (
+                pd.timedelta_range(start="1 day", periods=20),
+                Index(["2000-01-04", "2000-01-08"], dtype=object),
+                (
+                    r"None of \[Index\(\['2000-01-04', '2000-01-08'\], "
+                    r"dtype='object'\)\] are in the \[index\]"
+                ),
+            ),
+        ],
+    )
+    def test_loc_with_list_of_strings_representing_datetimes_not_matched_type(
+        self, idx, labels, msg
+    ):
+        # GH 11278
+        ser = Series(range(20), index=idx)
+        df = DataFrame(range(20), index=idx)
+
+        with pytest.raises(KeyError, match=msg):
+            ser.loc[labels]
+        with pytest.raises(KeyError, match=msg):
+            ser[labels]
+        with pytest.raises(KeyError, match=msg):
+            df.loc[labels]
+
+
+class TestStringSlicing:
+    def test_slice_irregular_datetime_index_with_nan(self):
+        # GH36953
+        index = pd.to_datetime(["2012-01-01", "2012-01-02", "2012-01-03", None])
+        df = DataFrame(range(len(index)), index=index)
+        expected = DataFrame(range(len(index[:3])), index=index[:3])
+        with pytest.raises(KeyError, match="non-existing keys is not allowed"):
+            # Upper bound is not in index (which is unordered)
+            # GH53983
+            # GH37819
+            df["2012-01-01":"2012-01-04"]
+        # Need this precision for right bound since the right slice
+        # bound is "rounded" up to the largest timepoint smaller than
+        # the next "resolution"-step of the provided point.
+        # e.g. 2012-01-03 is rounded up to 2012-01-04 - 1ns
+        result = df["2012-01-01":"2012-01-03 00:00:00.000000000"]
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/indexing/test_scalar.py b/pandas/pandas/tests/indexing/test_scalar.py
new file mode 100644
index 0000000000000..730fe584d7f07
--- /dev/null
+++ b/pandas/pandas/tests/indexing/test_scalar.py
@@ -0,0 +1,318 @@
+"""test scalar indexing, including at and iat"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+import itertools
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def generate_indices(f, values=False):
+    """
+    generate the indices
+    if values is True , use the axis values
+    is False, use the range
+    """
+    axes = f.axes
+    if values:
+        axes = (list(range(len(ax))) for ax in axes)
+
+    return itertools.product(*axes)
+
+
+class TestScalar:
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64])
+    def test_iat_set_ints(self, dtype, frame_or_series):
+        f = frame_or_series(range(3), index=Index([0, 1, 2], dtype=dtype))
+        indices = generate_indices(f, True)
+        for i in indices:
+            f.iat[i] = 1
+            expected = f.values[i]
+            tm.assert_almost_equal(expected, 1)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcd"), dtype=object),
+            date_range("20130101", periods=4),
+            Index(range(0, 8, 2), dtype=np.float64),
+        ],
+    )
+    def test_iat_set_other(self, index, frame_or_series):
+        f = frame_or_series(range(len(index)), index=index)
+        msg = "iAt based indexing can only have integer indexers"
+        idx = next(generate_indices(f, False))
+        with pytest.raises(ValueError, match=msg):
+            f.iat[idx] = 1
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("abcd"), dtype=object),
+            date_range("20130101", periods=4),
+            Index(range(0, 8, 2), dtype=np.float64),
+            Index(range(0, 8, 2), dtype=np.uint64),
+            Index(range(0, 8, 2), dtype=np.int64),
+        ],
+    )
+    def test_at_set_ints_other(self, index, frame_or_series):
+        f = frame_or_series(range(len(index)), index=index)
+        indices = generate_indices(f, False)
+        for i in indices:
+            f.at[i] = 1
+            expected = f.loc[i]
+            tm.assert_almost_equal(expected, 1)
+
+
+class TestAtAndiAT:
+    # at and iat tests that don't need Base class
+
+    def test_float_index_at_iat(self):
+        ser = Series([1, 2, 3], index=[0.1, 0.2, 0.3])
+        for el, item in ser.items():
+            assert ser.at[el] == item
+        for i in range(len(ser)):
+            assert ser.iat[i] == i + 1
+
+    def test_at_iat_coercion(self):
+        # as timestamp is not a tuple!
+        dates = date_range("1/1/2000", periods=8)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)),
+            index=dates,
+            columns=["A", "B", "C", "D"],
+        )
+        s = df["A"]
+
+        result = s.at[dates[5]]
+        xp = s.values[5]
+        assert result == xp
+
+    @pytest.mark.parametrize(
+        "ser, expected",
+        [
+            [
+                Series(["2014-01-01", "2014-02-02"], dtype="datetime64[ns]"),
+                Timestamp("2014-02-02"),
+            ],
+            [
+                Series(["1 days", "2 days"], dtype="timedelta64[ns]"),
+                Timedelta("2 days"),
+            ],
+        ],
+    )
+    def test_iloc_iat_coercion_datelike(self, indexer_ial, ser, expected):
+        # GH 7729
+        # make sure we are boxing the returns
+        result = indexer_ial(ser)[1]
+        assert result == expected
+
+    def test_imethods_with_dups(self):
+        # GH6493
+        # iat/iloc with dups
+
+        s = Series(range(5), index=[1, 1, 2, 2, 3], dtype="int64")
+        result = s.iloc[2]
+        assert result == 2
+        result = s.iat[2]
+        assert result == 2
+
+        msg = "index 10 is out of bounds for axis 0 with size 5"
+        with pytest.raises(IndexError, match=msg):
+            s.iat[10]
+        msg = "index -10 is out of bounds for axis 0 with size 5"
+        with pytest.raises(IndexError, match=msg):
+            s.iat[-10]
+
+        result = s.iloc[[2, 3]]
+        expected = Series([2, 3], [2, 2], dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+        df = s.to_frame()
+        result = df.iloc[2]
+        expected = Series(2, index=[0], name=2)
+        tm.assert_series_equal(result, expected)
+
+        result = df.iat[2, 0]
+        assert result == 2
+
+    def test_frame_at_with_duplicate_axes(self):
+        # GH#33041
+        arr = np.random.default_rng(2).standard_normal(6).reshape(3, 2)
+        df = DataFrame(arr, columns=["A", "A"])
+
+        result = df.at[0, "A"]
+        expected = df.iloc[0].copy()
+
+        tm.assert_series_equal(result, expected)
+
+        result = df.T.at["A", 0]
+        tm.assert_series_equal(result, expected)
+
+        # setter
+        df.at[1, "A"] = 2
+        expected = Series([2.0, 2.0], index=["A", "A"], name=1)
+        tm.assert_series_equal(df.iloc[1], expected)
+
+    def test_at_getitem_dt64tz_values(self):
+        # gh-15822
+        df = DataFrame(
+            {
+                "name": ["John", "Anderson"],
+                "date": [
+                    Timestamp(2017, 3, 13, 13, 32, 56),
+                    Timestamp(2017, 2, 16, 12, 10, 3),
+                ],
+            }
+        )
+        df["date"] = df["date"].dt.tz_localize("Asia/Shanghai")
+
+        expected = Timestamp("2017-03-13 13:32:56+0800", tz="Asia/Shanghai")
+
+        result = df.loc[0, "date"]
+        assert result == expected
+
+        result = df.at[0, "date"]
+        assert result == expected
+
+    def test_mixed_index_at_iat_loc_iloc_series(self):
+        # GH 19860
+        s = Series([1, 2, 3, 4, 5], index=["a", "b", "c", 1, 2])
+        for el, item in s.items():
+            assert s.at[el] == s.loc[el] == item
+        for i in range(len(s)):
+            assert s.iat[i] == s.iloc[i] == i + 1
+
+        with pytest.raises(KeyError, match="^4$"):
+            s.at[4]
+        with pytest.raises(KeyError, match="^4$"):
+            s.loc[4]
+
+    def test_mixed_index_at_iat_loc_iloc_dataframe(self):
+        # GH 19860
+        df = DataFrame(
+            [[0, 1, 2, 3, 4], [5, 6, 7, 8, 9]], columns=["a", "b", "c", 1, 2]
+        )
+        for rowIdx, row in df.iterrows():
+            for el, item in row.items():
+                assert df.at[rowIdx, el] == df.loc[rowIdx, el] == item
+
+        for row in range(2):
+            for i in range(5):
+                assert df.iat[row, i] == df.iloc[row, i] == row * 5 + i
+
+        with pytest.raises(KeyError, match="^3$"):
+            df.at[0, 3]
+        with pytest.raises(KeyError, match="^3$"):
+            df.loc[0, 3]
+
+    def test_iat_setter_incompatible_assignment(self):
+        # GH 23236
+        result = DataFrame({"a": [0.0, 1.0], "b": [4, 5]})
+        result.iat[0, 0] = None
+        expected = DataFrame({"a": [None, 1], "b": [4, 5]})
+        tm.assert_frame_equal(result, expected)
+
+
+def test_iat_dont_wrap_object_datetimelike():
+    # GH#32809 .iat calls go through DataFrame._get_value, should not
+    #  call maybe_box_datetimelike
+    dti = date_range("2016-01-01", periods=3)
+    tdi = dti - dti
+    ser = Series(dti.to_pydatetime(), dtype=object)
+    ser2 = Series(tdi.to_pytimedelta(), dtype=object)
+    df = DataFrame({"A": ser, "B": ser2})
+    assert (df.dtypes == object).all()
+
+    for result in [df.at[0, "A"], df.iat[0, 0], df.loc[0, "A"], df.iloc[0, 0]]:
+        assert result is ser[0]
+        assert isinstance(result, datetime)
+        assert not isinstance(result, Timestamp)
+
+    for result in [df.at[1, "B"], df.iat[1, 1], df.loc[1, "B"], df.iloc[1, 1]]:
+        assert result is ser2[1]
+        assert isinstance(result, timedelta)
+        assert not isinstance(result, Timedelta)
+
+
+def test_at_with_tuple_index_get():
+    # GH 26989
+    # DataFrame.at getter works with Index of tuples
+    df = DataFrame({"a": [1, 2]}, index=[(1, 2), (3, 4)])
+    assert df.index.nlevels == 1
+    assert df.at[(1, 2), "a"] == 1
+
+    # Series.at getter works with Index of tuples
+    series = df["a"]
+    assert series.index.nlevels == 1
+    assert series.at[(1, 2)] == 1
+
+
+@pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+def test_at_with_tuple_index_set():
+    # GH 26989
+    # DataFrame.at setter works with Index of tuples
+    df = DataFrame({"a": [1, 2]}, index=[(1, 2), (3, 4)])
+    assert df.index.nlevels == 1
+    df.at[(1, 2), "a"] = 2
+    assert df.at[(1, 2), "a"] == 2
+
+    # Series.at setter works with Index of tuples
+    series = df["a"]
+    assert series.index.nlevels == 1
+    series.at[1, 2] = 3
+    assert series.at[1, 2] == 3
+
+
+class TestMultiIndexScalar:
+    def test_multiindex_at_get(self):
+        # GH 26989
+        # DataFrame.at and DataFrame.loc getter works with MultiIndex
+        df = DataFrame({"a": [1, 2]}, index=[[1, 2], [3, 4]])
+        assert df.index.nlevels == 2
+        assert df.at[(1, 3), "a"] == 1
+        assert df.loc[(1, 3), "a"] == 1
+
+        # Series.at and Series.loc getter works with MultiIndex
+        series = df["a"]
+        assert series.index.nlevels == 2
+        assert series.at[1, 3] == 1
+        assert series.loc[1, 3] == 1
+
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+    def test_multiindex_at_set(self):
+        # GH 26989
+        # DataFrame.at and DataFrame.loc setter works with MultiIndex
+        df = DataFrame({"a": [1, 2]}, index=[[1, 2], [3, 4]])
+        assert df.index.nlevels == 2
+        df.at[(1, 3), "a"] = 3
+        assert df.at[(1, 3), "a"] == 3
+        df.loc[(1, 3), "a"] = 4
+        assert df.loc[(1, 3), "a"] == 4
+
+        # Series.at and Series.loc setter works with MultiIndex
+        series = df["a"]
+        assert series.index.nlevels == 2
+        series.at[1, 3] = 5
+        assert series.at[1, 3] == 5
+        series.loc[1, 3] = 6
+        assert series.loc[1, 3] == 6
+
+    def test_multiindex_at_get_one_level(self):
+        # GH#38053
+        s2 = Series((0, 1), index=[[False, True]])
+        result = s2.at[False]
+        assert result == 0
diff --git a/pandas/pandas/tests/interchange/__init__.py b/pandas/pandas/tests/interchange/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/interchange/test_impl.py b/pandas/pandas/tests/interchange/test_impl.py
new file mode 100644
index 0000000000000..a41d7dec8b496
--- /dev/null
+++ b/pandas/pandas/tests/interchange/test_impl.py
@@ -0,0 +1,651 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+from pandas.compat import (
+    is_ci_environment,
+    is_platform_windows,
+)
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.interchange.column import PandasColumn
+from pandas.core.interchange.dataframe_protocol import (
+    ColumnNullType,
+    DtypeKind,
+)
+from pandas.core.interchange.from_dataframe import from_dataframe
+from pandas.core.interchange.utils import ArrowCTypes
+
+
+@pytest.mark.parametrize("data", [("ordered", True), ("unordered", False)])
+def test_categorical_dtype(data):
+    data_categorical = {
+        "ordered": pd.Categorical(list("testdata") * 30, ordered=True),
+        "unordered": pd.Categorical(list("testdata") * 30, ordered=False),
+    }
+    df = pd.DataFrame({"A": (data_categorical[data[0]])})
+
+    col = df.__dataframe__().get_column_by_name("A")
+    assert col.dtype[0] == DtypeKind.CATEGORICAL
+    assert col.null_count == 0
+    assert col.describe_null == (ColumnNullType.USE_SENTINEL, -1)
+    assert col.num_chunks() == 1
+    desc_cat = col.describe_categorical
+    assert desc_cat["is_ordered"] == data[1]
+    assert desc_cat["is_dictionary"] is True
+    assert isinstance(desc_cat["categories"], PandasColumn)
+    tm.assert_series_equal(
+        desc_cat["categories"]._col, pd.Series(["a", "d", "e", "s", "t"])
+    )
+
+    tm.assert_frame_equal(df, from_dataframe(df.__dataframe__()))
+
+
+def test_categorical_pyarrow():
+    # GH 49889
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+
+    arr = ["Mon", "Tue", "Mon", "Wed", "Mon", "Thu", "Fri", "Sat", "Sun"]
+    table = pa.table({"weekday": pa.array(arr).dictionary_encode()})
+    exchange_df = table.__dataframe__()
+    result = from_dataframe(exchange_df)
+    weekday = pd.Categorical(
+        arr, categories=["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"]
+    )
+    expected = pd.DataFrame({"weekday": weekday})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_empty_categorical_pyarrow():
+    # https://github.com/pandas-dev/pandas/issues/53077
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+
+    arr = [None]
+    table = pa.table({"arr": pa.array(arr, "float64").dictionary_encode()})
+    exchange_df = table.__dataframe__()
+    result = pd.api.interchange.from_dataframe(exchange_df)
+    expected = pd.DataFrame({"arr": pd.Categorical([np.nan])})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_large_string_pyarrow():
+    # GH 52795
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+
+    arr = ["Mon", "Tue"]
+    table = pa.table({"weekday": pa.array(arr, "large_string")})
+    exchange_df = table.__dataframe__()
+    result = from_dataframe(exchange_df)
+    expected = pd.DataFrame({"weekday": ["Mon", "Tue"]})
+    tm.assert_frame_equal(result, expected)
+
+    # check round-trip
+    assert pa.Table.equals(pa.interchange.from_dataframe(result), table)
+
+
+@pytest.mark.parametrize(
+    ("offset", "length", "expected_values"),
+    [
+        (0, None, [3.3, float("nan"), 2.1]),
+        (1, None, [float("nan"), 2.1]),
+        (2, None, [2.1]),
+        (0, 2, [3.3, float("nan")]),
+        (0, 1, [3.3]),
+        (1, 1, [float("nan")]),
+    ],
+)
+def test_bitmasks_pyarrow(offset, length, expected_values):
+    # GH 52795
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+
+    arr = [3.3, None, 2.1]
+    table = pa.table({"arr": arr}).slice(offset, length)
+    exchange_df = table.__dataframe__()
+    result = from_dataframe(exchange_df)
+    expected = pd.DataFrame({"arr": expected_values})
+    tm.assert_frame_equal(result, expected)
+
+    # check round-trip
+    assert pa.Table.equals(pa.interchange.from_dataframe(result), table)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        lambda: np.random.default_rng(2).integers(-100, 100),
+        lambda: np.random.default_rng(2).integers(1, 100),
+        lambda: np.random.default_rng(2).random(),
+        lambda: np.random.default_rng(2).choice([True, False]),
+        lambda: datetime(
+            year=np.random.default_rng(2).integers(1900, 2100),
+            month=np.random.default_rng(2).integers(1, 12),
+            day=np.random.default_rng(2).integers(1, 20),
+        ),
+    ],
+)
+def test_dataframe(data):
+    NCOLS, NROWS = 10, 20
+    data = {
+        f"col{int((i - NCOLS / 2) % NCOLS + 1)}": [data() for _ in range(NROWS)]
+        for i in range(NCOLS)
+    }
+    df = pd.DataFrame(data)
+
+    df2 = df.__dataframe__()
+
+    assert df2.num_columns() == NCOLS
+    assert df2.num_rows() == NROWS
+
+    assert list(df2.column_names()) == list(data.keys())
+
+    indices = (0, 2)
+    names = tuple(list(data.keys())[idx] for idx in indices)
+
+    result = from_dataframe(df2.select_columns(indices))
+    expected = from_dataframe(df2.select_columns_by_name(names))
+    tm.assert_frame_equal(result, expected)
+
+    assert isinstance(result.attrs["_INTERCHANGE_PROTOCOL_BUFFERS"], list)
+    assert isinstance(expected.attrs["_INTERCHANGE_PROTOCOL_BUFFERS"], list)
+
+
+def test_missing_from_masked():
+    df = pd.DataFrame(
+        {
+            "x": np.array([1.0, 2.0, 3.0, 4.0, 0.0]),
+            "y": np.array([1.5, 2.5, 3.5, 4.5, 0]),
+            "z": np.array([1.0, 0.0, 1.0, 1.0, 1.0]),
+        }
+    )
+
+    rng = np.random.default_rng(2)
+    dict_null = {col: rng.integers(low=0, high=len(df)) for col in df.columns}
+    for col, num_nulls in dict_null.items():
+        null_idx = df.index[
+            rng.choice(np.arange(len(df)), size=num_nulls, replace=False)
+        ]
+        df.loc[null_idx, col] = None
+
+    df2 = df.__dataframe__()
+
+    assert df2.get_column_by_name("x").null_count == dict_null["x"]
+    assert df2.get_column_by_name("y").null_count == dict_null["y"]
+    assert df2.get_column_by_name("z").null_count == dict_null["z"]
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        {"x": [1.5, 2.5, 3.5], "y": [9.2, 10.5, 11.8]},
+        {"x": [1, 2, 0], "y": [9.2, 10.5, 11.8]},
+        {
+            "x": np.array([True, True, False]),
+            "y": np.array([1, 2, 0]),
+            "z": np.array([9.2, 10.5, 11.8]),
+        },
+    ],
+)
+def test_mixed_data(data):
+    df = pd.DataFrame(data)
+    df2 = df.__dataframe__()
+
+    for col_name in df.columns:
+        assert df2.get_column_by_name(col_name).null_count == 0
+
+
+def test_mixed_missing():
+    df = pd.DataFrame(
+        {
+            "x": np.array([True, None, False, None, True]),
+            "y": np.array([None, 2, None, 1, 2]),
+            "z": np.array([9.2, 10.5, None, 11.8, None]),
+        }
+    )
+
+    df2 = df.__dataframe__()
+
+    for col_name in df.columns:
+        assert df2.get_column_by_name(col_name).null_count == 2
+
+
+def test_string():
+    string_data = {
+        "separator data": [
+            "abC|DeF,Hik",
+            "234,3245.67",
+            "gSaf,qWer|Gre",
+            "asd3,4sad|",
+            np.nan,
+        ]
+    }
+    test_str_data = string_data["separator data"] + [""]
+    df = pd.DataFrame({"A": test_str_data})
+    col = df.__dataframe__().get_column_by_name("A")
+
+    assert col.size() == 6
+    assert col.null_count == 1
+    assert col.dtype[0] == DtypeKind.STRING
+    assert col.describe_null == (ColumnNullType.USE_BYTEMASK, 0)
+
+    df_sliced = df[1:]
+    col = df_sliced.__dataframe__().get_column_by_name("A")
+    assert col.size() == 5
+    assert col.null_count == 1
+    assert col.dtype[0] == DtypeKind.STRING
+    assert col.describe_null == (ColumnNullType.USE_BYTEMASK, 0)
+
+
+def test_nonstring_object():
+    df = pd.DataFrame({"A": ["a", 10, 1.0, ()]})
+    col = df.__dataframe__().get_column_by_name("A")
+    with pytest.raises(NotImplementedError, match="not supported yet"):
+        col.dtype
+
+
+def test_datetime():
+    df = pd.DataFrame({"A": [pd.Timestamp("2022-01-01"), pd.NaT]})
+    col = df.__dataframe__().get_column_by_name("A")
+
+    assert col.size() == 2
+    assert col.null_count == 1
+    assert col.dtype[0] == DtypeKind.DATETIME
+    assert col.describe_null == (ColumnNullType.USE_SENTINEL, iNaT)
+
+    tm.assert_frame_equal(df, from_dataframe(df.__dataframe__()))
+
+
+def test_categorical_to_numpy_dlpack():
+    # https://github.com/pandas-dev/pandas/issues/48393
+    df = pd.DataFrame({"A": pd.Categorical(["a", "b", "a"])})
+    col = df.__dataframe__().get_column_by_name("A")
+    result = np.from_dlpack(col.get_buffers()["data"][0])
+    expected = np.array([0, 1, 0], dtype="int8")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("data", [{}, {"a": []}])
+def test_empty_pyarrow(data):
+    # GH 53155
+    pytest.importorskip("pyarrow", "11.0.0")
+    from pyarrow.interchange import from_dataframe as pa_from_dataframe
+
+    expected = pd.DataFrame(data)
+    arrow_df = pa_from_dataframe(expected)
+    result = from_dataframe(arrow_df)
+    tm.assert_frame_equal(result, expected, check_column_type=False)
+
+
+def test_multi_chunk_pyarrow() -> None:
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+    n_legs = pa.chunked_array([[2, 2, 4], [4, 5, 100]])
+    names = ["n_legs"]
+    table = pa.table([n_legs], names=names)
+    with pytest.raises(
+        RuntimeError,
+        match="Cannot do zero copy conversion into multi-column DataFrame block",
+    ):
+        pd.api.interchange.from_dataframe(table, allow_copy=False)
+
+
+def test_multi_chunk_column() -> None:
+    pytest.importorskip("pyarrow", "11.0.0")
+    ser = pd.Series([1, 2, None], dtype="Int64[pyarrow]")
+    df = pd.concat([ser, ser], ignore_index=True).to_frame("a")
+    df_orig = df.copy()
+    with pytest.raises(
+        RuntimeError, match="Found multi-chunk pyarrow array, but `allow_copy` is False"
+    ):
+        pd.api.interchange.from_dataframe(df.__dataframe__(allow_copy=False))
+    result = pd.api.interchange.from_dataframe(df.__dataframe__(allow_copy=True))
+    # Interchange protocol defaults to creating numpy-backed columns, so currently this
+    # is 'float64'.
+    expected = pd.DataFrame({"a": [1.0, 2.0, None, 1.0, 2.0, None]}, dtype="float64")
+    tm.assert_frame_equal(result, expected)
+
+    # Check that the rechunking we did didn't modify the original DataFrame.
+    tm.assert_frame_equal(df, df_orig)
+    assert len(df["a"].array._pa_array.chunks) == 2
+    assert len(df_orig["a"].array._pa_array.chunks) == 2
+
+
+def test_timestamp_ns_pyarrow():
+    # GH 56712
+    pytest.importorskip("pyarrow", "11.0.0")
+    timestamp_args = {
+        "year": 2000,
+        "month": 1,
+        "day": 1,
+        "hour": 1,
+        "minute": 1,
+        "second": 1,
+    }
+    df = pd.Series(
+        [datetime(**timestamp_args)],
+        dtype="timestamp[ns][pyarrow]",
+        name="col0",
+    ).to_frame()
+
+    dfi = df.__dataframe__()
+    result = pd.api.interchange.from_dataframe(dfi)["col0"].item()
+
+    expected = pd.Timestamp(**timestamp_args)
+    assert result == expected
+
+
+@pytest.mark.parametrize("tz", ["UTC", "US/Pacific"])
+def test_datetimetzdtype(tz, unit):
+    # GH 54239
+    tz_data = (
+        pd.date_range("2018-01-01", periods=5, freq="D").tz_localize(tz).as_unit(unit)
+    )
+    df = pd.DataFrame({"ts_tz": tz_data})
+    tm.assert_frame_equal(df, from_dataframe(df.__dataframe__()))
+
+
+def test_interchange_from_non_pandas_tz_aware(request):
+    # GH 54239, 54287
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+    import pyarrow.compute as pc
+
+    if is_platform_windows() and is_ci_environment():
+        mark = pytest.mark.xfail(
+            raises=pa.ArrowInvalid,
+            reason=(
+                "TODO: Set ARROW_TIMEZONE_DATABASE environment variable "
+                "on CI to path to the tzdata for pyarrow."
+            ),
+        )
+        request.applymarker(mark)
+
+    arr = pa.array([datetime(2020, 1, 1), None, datetime(2020, 1, 2)])
+    arr = pc.assume_timezone(arr, "Asia/Kathmandu")
+    table = pa.table({"arr": arr})
+    exchange_df = table.__dataframe__()
+    result = from_dataframe(exchange_df)
+
+    expected = pd.DataFrame(
+        ["2020-01-01 00:00:00+05:45", "NaT", "2020-01-02 00:00:00+05:45"],
+        columns=["arr"],
+        dtype="datetime64[us, Asia/Kathmandu]",
+    )
+    tm.assert_frame_equal(expected, result)
+
+
+def test_interchange_from_corrected_buffer_dtypes(monkeypatch) -> None:
+    # https://github.com/pandas-dev/pandas/issues/54781
+    df = pd.DataFrame({"a": ["foo", "bar"]}).__dataframe__()
+    interchange = df.__dataframe__()
+    column = interchange.get_column_by_name("a")
+    buffers = column.get_buffers()
+    buffers_data = buffers["data"]
+    buffer_dtype = buffers_data[1]
+    buffer_dtype = (
+        DtypeKind.UINT,
+        8,
+        ArrowCTypes.UINT8,
+        buffer_dtype[3],
+    )
+    buffers["data"] = (buffers_data[0], buffer_dtype)
+    column.get_buffers = lambda: buffers
+    interchange.get_column_by_name = lambda _: column
+    monkeypatch.setattr(df, "__dataframe__", lambda allow_copy: interchange)
+    pd.api.interchange.from_dataframe(df)
+
+
+def test_empty_string_column():
+    # https://github.com/pandas-dev/pandas/issues/56703
+    df = pd.DataFrame({"a": []}, dtype=str)
+    df2 = df.__dataframe__()
+    result = pd.api.interchange.from_dataframe(df2)
+    tm.assert_frame_equal(df, result)
+
+
+def test_large_string():
+    # GH#56702
+    pytest.importorskip("pyarrow")
+    df = pd.DataFrame({"a": ["x"]}, dtype="large_string[pyarrow]")
+    result = pd.api.interchange.from_dataframe(df.__dataframe__())
+    expected = pd.DataFrame({"a": ["x"]}, dtype="str")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_non_str_names():
+    # https://github.com/pandas-dev/pandas/issues/56701
+    df = pd.Series([1, 2, 3], name=0).to_frame()
+    names = df.__dataframe__().column_names()
+    assert names == ["0"]
+
+
+def test_non_str_names_w_duplicates():
+    # https://github.com/pandas-dev/pandas/issues/56701
+    df = pd.DataFrame({"0": [1, 2, 3], 0: [4, 5, 6]})
+    dfi = df.__dataframe__()
+    with pytest.raises(
+        TypeError,
+        match=(
+            "Expected a Series, got a DataFrame. This likely happened because you "
+            "called __dataframe__ on a DataFrame which, after converting column "
+            r"names to string, resulted in duplicated names: Index\(\['0', '0'\], "
+            r"dtype='(str|object)'\). Please rename these columns before using the "
+            "interchange protocol."
+        ),
+    ):
+        pd.api.interchange.from_dataframe(dfi, allow_copy=False)
+
+
+@pytest.mark.parametrize(
+    ("data", "dtype", "expected_dtype"),
+    [
+        ([1, 2, None], "Int64", "int64"),
+        ([1, 2, None], "Int64[pyarrow]", "int64"),
+        ([1, 2, None], "Int8", "int8"),
+        ([1, 2, None], "Int8[pyarrow]", "int8"),
+        (
+            [1, 2, None],
+            "UInt64",
+            "uint64",
+        ),
+        (
+            [1, 2, None],
+            "UInt64[pyarrow]",
+            "uint64",
+        ),
+        ([1.0, 2.25, None], "Float32", "float32"),
+        ([1.0, 2.25, None], "Float32[pyarrow]", "float32"),
+        ([True, False, None], "boolean", "bool"),
+        ([True, False, None], "boolean[pyarrow]", "bool"),
+        (["much ado", "about", None], pd.StringDtype(na_value=np.nan), "large_string"),
+        (["much ado", "about", None], "string[pyarrow]", "large_string"),
+        (
+            [datetime(2020, 1, 1), datetime(2020, 1, 2), None],
+            "timestamp[ns][pyarrow]",
+            "timestamp[ns]",
+        ),
+        (
+            [datetime(2020, 1, 1), datetime(2020, 1, 2), None],
+            "timestamp[us][pyarrow]",
+            "timestamp[us]",
+        ),
+        (
+            [
+                datetime(2020, 1, 1, tzinfo=timezone.utc),
+                datetime(2020, 1, 2, tzinfo=timezone.utc),
+                None,
+            ],
+            "timestamp[us, Asia/Kathmandu][pyarrow]",
+            "timestamp[us, tz=Asia/Kathmandu]",
+        ),
+    ],
+)
+def test_pandas_nullable_with_missing_values(
+    data: list, dtype: str, expected_dtype: str
+) -> None:
+    # https://github.com/pandas-dev/pandas/issues/57643
+    # https://github.com/pandas-dev/pandas/issues/57664
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+    import pyarrow.interchange as pai
+
+    if expected_dtype == "timestamp[us, tz=Asia/Kathmandu]":
+        expected_dtype = pa.timestamp("us", "Asia/Kathmandu")
+
+    df = pd.DataFrame({"a": data}, dtype=dtype)
+    result = pai.from_dataframe(df.__dataframe__())["a"]
+    assert result.type == expected_dtype
+    assert result[0].as_py() == data[0]
+    assert result[1].as_py() == data[1]
+    assert result[2].as_py() is None
+
+
+@pytest.mark.parametrize(
+    ("data", "dtype", "expected_dtype"),
+    [
+        ([1, 2, 3], "Int64", "int64"),
+        ([1, 2, 3], "Int64[pyarrow]", "int64"),
+        ([1, 2, 3], "Int8", "int8"),
+        ([1, 2, 3], "Int8[pyarrow]", "int8"),
+        (
+            [1, 2, 3],
+            "UInt64",
+            "uint64",
+        ),
+        (
+            [1, 2, 3],
+            "UInt64[pyarrow]",
+            "uint64",
+        ),
+        ([1.0, 2.25, 5.0], "Float32", "float32"),
+        ([1.0, 2.25, 5.0], "Float32[pyarrow]", "float32"),
+        ([True, False, False], "boolean", "bool"),
+        ([True, False, False], "boolean[pyarrow]", "bool"),
+        (
+            ["much ado", "about", "nothing"],
+            pd.StringDtype(na_value=np.nan),
+            "large_string",
+        ),
+        (["much ado", "about", "nothing"], "string[pyarrow]", "large_string"),
+        (
+            [datetime(2020, 1, 1), datetime(2020, 1, 2), datetime(2020, 1, 3)],
+            "timestamp[ns][pyarrow]",
+            "timestamp[ns]",
+        ),
+        (
+            [datetime(2020, 1, 1), datetime(2020, 1, 2), datetime(2020, 1, 3)],
+            "timestamp[us][pyarrow]",
+            "timestamp[us]",
+        ),
+        (
+            [
+                datetime(2020, 1, 1, tzinfo=timezone.utc),
+                datetime(2020, 1, 2, tzinfo=timezone.utc),
+                datetime(2020, 1, 3, tzinfo=timezone.utc),
+            ],
+            "timestamp[us, Asia/Kathmandu][pyarrow]",
+            "timestamp[us, tz=Asia/Kathmandu]",
+        ),
+    ],
+)
+def test_pandas_nullable_without_missing_values(
+    data: list, dtype: str, expected_dtype: str
+) -> None:
+    # https://github.com/pandas-dev/pandas/issues/57643
+    pa = pytest.importorskip("pyarrow", "11.0.0")
+    import pyarrow.interchange as pai
+
+    if expected_dtype == "timestamp[us, tz=Asia/Kathmandu]":
+        expected_dtype = pa.timestamp("us", "Asia/Kathmandu")
+
+    df = pd.DataFrame({"a": data}, dtype=dtype)
+    result = pai.from_dataframe(df.__dataframe__())["a"]
+    assert result.type == expected_dtype
+    assert result[0].as_py() == data[0]
+    assert result[1].as_py() == data[1]
+    assert result[2].as_py() == data[2]
+
+
+def test_string_validity_buffer() -> None:
+    # https://github.com/pandas-dev/pandas/issues/57761
+    pytest.importorskip("pyarrow", "11.0.0")
+    df = pd.DataFrame({"a": ["x"]}, dtype="large_string[pyarrow]")
+    result = df.__dataframe__().get_column_by_name("a").get_buffers()["validity"]
+    assert result is None
+
+
+def test_string_validity_buffer_no_missing() -> None:
+    # https://github.com/pandas-dev/pandas/issues/57762
+    pytest.importorskip("pyarrow", "11.0.0")
+    df = pd.DataFrame({"a": ["x", None]}, dtype="large_string[pyarrow]")
+    validity = df.__dataframe__().get_column_by_name("a").get_buffers()["validity"]
+    assert validity is not None
+    result = validity[1]
+    expected = (DtypeKind.BOOL, 1, ArrowCTypes.BOOL, "=")
+    assert result == expected
+
+
+def test_empty_dataframe():
+    # https://github.com/pandas-dev/pandas/issues/56700
+    df = pd.DataFrame({"a": []}, dtype="int8")
+    dfi = df.__dataframe__()
+    result = pd.api.interchange.from_dataframe(dfi, allow_copy=False)
+    expected = pd.DataFrame({"a": []}, dtype="int8")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("data", "expected_dtype", "expected_buffer_dtype"),
+    [
+        (
+            pd.Series(["a", "b", "a"], dtype="category"),
+            (DtypeKind.CATEGORICAL, 8, "c", "="),
+            (DtypeKind.INT, 8, "c", "|"),
+        ),
+        (
+            pd.Series(
+                [datetime(2022, 1, 1), datetime(2022, 1, 2), datetime(2022, 1, 3)],
+                dtype="M8[ns]",
+            ),
+            (DtypeKind.DATETIME, 64, "tsn:", "="),
+            (DtypeKind.INT, 64, ArrowCTypes.INT64, "="),
+        ),
+        (
+            pd.Series(["a", "bc", None]),
+            (DtypeKind.STRING, 8, ArrowCTypes.STRING, "="),
+            (DtypeKind.UINT, 8, ArrowCTypes.UINT8, "="),
+        ),
+        (
+            pd.Series([1, 2, 3]),
+            (DtypeKind.INT, 64, ArrowCTypes.INT64, "="),
+            (DtypeKind.INT, 64, ArrowCTypes.INT64, "="),
+        ),
+        (
+            pd.Series([1.5, 2, 3]),
+            (DtypeKind.FLOAT, 64, ArrowCTypes.FLOAT64, "="),
+            (DtypeKind.FLOAT, 64, ArrowCTypes.FLOAT64, "="),
+        ),
+    ],
+)
+def test_buffer_dtype_categorical(
+    data: pd.Series,
+    expected_dtype: tuple[DtypeKind, int, str, str],
+    expected_buffer_dtype: tuple[DtypeKind, int, str, str],
+) -> None:
+    # https://github.com/pandas-dev/pandas/issues/54781
+    df = pd.DataFrame({"data": data})
+    dfi = df.__dataframe__()
+    col = dfi.get_column_by_name("data")
+    assert col.dtype == expected_dtype
+    assert col.get_buffers()["data"][1] == expected_buffer_dtype
+
+
+def test_from_dataframe_list_dtype():
+    pa = pytest.importorskip("pyarrow", "14.0.0")
+    data = {"a": [[1, 2], [4, 5, 6]]}
+    tbl = pa.table(data)
+    result = from_dataframe(tbl)
+    expected = pd.DataFrame(data)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/interchange/test_spec_conformance.py b/pandas/pandas/tests/interchange/test_spec_conformance.py
new file mode 100644
index 0000000000000..55e42ed2023cd
--- /dev/null
+++ b/pandas/pandas/tests/interchange/test_spec_conformance.py
@@ -0,0 +1,176 @@
+"""
+A verbatim copy (vendored) of the spec tests.
+Taken from https://github.com/data-apis/dataframe-api
+"""
+
+import ctypes
+import math
+
+import pytest
+
+import pandas as pd
+
+
+@pytest.fixture
+def df_from_dict():
+    def maker(dct, is_categorical=False):
+        df = pd.DataFrame(dct)
+        return df.astype("category") if is_categorical else df
+
+    return maker
+
+
+@pytest.mark.parametrize(
+    "test_data",
+    [
+        {"a": ["foo", "bar"], "b": ["baz", "qux"]},
+        {"a": [1.5, 2.5, 3.5], "b": [9.2, 10.5, 11.8]},
+        {"A": [1, 2, 3, 4], "B": [1, 2, 3, 4]},
+    ],
+    ids=["str_data", "float_data", "int_data"],
+)
+def test_only_one_dtype(test_data, df_from_dict):
+    columns = list(test_data.keys())
+    df = df_from_dict(test_data)
+    dfX = df.__dataframe__()
+
+    column_size = len(test_data[columns[0]])
+    for column in columns:
+        null_count = dfX.get_column_by_name(column).null_count
+        assert null_count == 0
+        assert isinstance(null_count, int)
+        assert dfX.get_column_by_name(column).size() == column_size
+        assert dfX.get_column_by_name(column).offset == 0
+
+
+def test_mixed_dtypes(df_from_dict):
+    df = df_from_dict(
+        {
+            "a": [1, 2, 3],  # dtype kind INT = 0
+            "b": [3, 4, 5],  # dtype kind INT = 0
+            "c": [1.5, 2.5, 3.5],  # dtype kind FLOAT = 2
+            "d": [9, 10, 11],  # dtype kind INT = 0
+            "e": [True, False, True],  # dtype kind BOOLEAN = 20
+            "f": ["a", "", "c"],  # dtype kind STRING = 21
+        }
+    )
+    dfX = df.__dataframe__()
+    # for meanings of dtype[0] see the spec; we cannot import the spec here as this
+    # file is expected to be vendored *anywhere*;
+    # values for dtype[0] are explained above
+    columns = {"a": 0, "b": 0, "c": 2, "d": 0, "e": 20, "f": 21}
+
+    for column, kind in columns.items():
+        colX = dfX.get_column_by_name(column)
+        assert colX.null_count == 0
+        assert isinstance(colX.null_count, int)
+        assert colX.size() == 3
+        assert colX.offset == 0
+
+        assert colX.dtype[0] == kind
+
+    assert dfX.get_column_by_name("c").dtype[1] == 64
+
+
+def test_na_float(df_from_dict):
+    df = df_from_dict({"a": [1.0, math.nan, 2.0]})
+    dfX = df.__dataframe__()
+    colX = dfX.get_column_by_name("a")
+    assert colX.null_count == 1
+    assert isinstance(colX.null_count, int)
+
+
+def test_noncategorical(df_from_dict):
+    df = df_from_dict({"a": [1, 2, 3]})
+    dfX = df.__dataframe__()
+    colX = dfX.get_column_by_name("a")
+    with pytest.raises(TypeError, match=".*categorical.*"):
+        colX.describe_categorical
+
+
+def test_categorical(df_from_dict):
+    df = df_from_dict(
+        {"weekday": ["Mon", "Tue", "Mon", "Wed", "Mon", "Thu", "Fri", "Sat", "Sun"]},
+        is_categorical=True,
+    )
+
+    colX = df.__dataframe__().get_column_by_name("weekday")
+    categorical = colX.describe_categorical
+    assert isinstance(categorical["is_ordered"], bool)
+    assert isinstance(categorical["is_dictionary"], bool)
+
+
+def test_dataframe(df_from_dict):
+    df = df_from_dict(
+        {"x": [True, True, False], "y": [1, 2, 0], "z": [9.2, 10.5, 11.8]}
+    )
+    dfX = df.__dataframe__()
+
+    assert dfX.num_columns() == 3
+    assert dfX.num_rows() == 3
+    assert dfX.num_chunks() == 1
+    assert list(dfX.column_names()) == ["x", "y", "z"]
+    assert list(dfX.select_columns((0, 2)).column_names()) == list(
+        dfX.select_columns_by_name(("x", "z")).column_names()
+    )
+
+
+@pytest.mark.parametrize(["size", "n_chunks"], [(10, 3), (12, 3), (12, 5)])
+def test_df_get_chunks(size, n_chunks, df_from_dict):
+    df = df_from_dict({"x": list(range(size))})
+    dfX = df.__dataframe__()
+    chunks = list(dfX.get_chunks(n_chunks))
+    assert len(chunks) == n_chunks
+    assert sum(chunk.num_rows() for chunk in chunks) == size
+
+
+@pytest.mark.parametrize(["size", "n_chunks"], [(10, 3), (12, 3), (12, 5)])
+def test_column_get_chunks(size, n_chunks, df_from_dict):
+    df = df_from_dict({"x": list(range(size))})
+    dfX = df.__dataframe__()
+    chunks = list(dfX.get_column(0).get_chunks(n_chunks))
+    assert len(chunks) == n_chunks
+    assert sum(chunk.size() for chunk in chunks) == size
+
+
+def test_get_columns(df_from_dict):
+    df = df_from_dict({"a": [0, 1], "b": [2.5, 3.5]})
+    dfX = df.__dataframe__()
+    for colX in dfX.get_columns():
+        assert colX.size() == 2
+        assert colX.num_chunks() == 1
+    # for meanings of dtype[0] see the spec; we cannot import the spec here as this
+    # file is expected to be vendored *anywhere*
+    assert dfX.get_column(0).dtype[0] == 0  # INT
+    assert dfX.get_column(1).dtype[0] == 2  # FLOAT
+
+
+def test_buffer(df_from_dict):
+    arr = [0, 1, -1]
+    df = df_from_dict({"a": arr})
+    dfX = df.__dataframe__()
+    colX = dfX.get_column(0)
+    bufX = colX.get_buffers()
+
+    dataBuf, dataDtype = bufX["data"]
+
+    assert dataBuf.bufsize > 0
+    assert dataBuf.ptr != 0
+    device, _ = dataBuf.__dlpack_device__()
+
+    # for meanings of dtype[0] see the spec; we cannot import the spec here as this
+    # file is expected to be vendored *anywhere*
+    assert dataDtype[0] == 0  # INT
+
+    if device == 1:  # CPU-only as we're going to directly read memory here
+        bitwidth = dataDtype[1]
+        ctype = {
+            8: ctypes.c_int8,
+            16: ctypes.c_int16,
+            32: ctypes.c_int32,
+            64: ctypes.c_int64,
+        }[bitwidth]
+
+        for idx, truth in enumerate(arr):
+            val = ctype.from_address(dataBuf.ptr + idx * (bitwidth // 8)).value
+            assert val == truth, f"Buffer at index {idx} mismatch"
diff --git a/pandas/pandas/tests/interchange/test_utils.py b/pandas/pandas/tests/interchange/test_utils.py
new file mode 100644
index 0000000000000..a47bc2752ff32
--- /dev/null
+++ b/pandas/pandas/tests/interchange/test_utils.py
@@ -0,0 +1,89 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas.core.interchange.utils import dtype_to_arrow_c_fmt
+
+# TODO: use ArrowSchema to get reference C-string.
+# At the time, there is no way to access ArrowSchema holding a type format string
+# from python. The only way to access it is to export the structure to a C-pointer,
+# see DataType._export_to_c() method defined in
+# https://github.com/apache/arrow/blob/master/python/pyarrow/types.pxi
+
+
+@pytest.mark.parametrize(
+    "pandas_dtype, c_string",
+    [
+        (np.dtype("bool"), "b"),
+        (np.dtype("int8"), "c"),
+        (np.dtype("uint8"), "C"),
+        (np.dtype("int16"), "s"),
+        (np.dtype("uint16"), "S"),
+        (np.dtype("int32"), "i"),
+        (np.dtype("uint32"), "I"),
+        (np.dtype("int64"), "l"),
+        (np.dtype("uint64"), "L"),
+        (np.dtype("float16"), "e"),
+        (np.dtype("float32"), "f"),
+        (np.dtype("float64"), "g"),
+        (pd.Series(["a"]).dtype, "u"),
+        (
+            pd.Series([0]).astype("datetime64[ns]").dtype,
+            "tsn:",
+        ),
+        (pd.CategoricalDtype(["a"]), "l"),
+        (np.dtype("O"), "u"),
+    ],
+)
+def test_dtype_to_arrow_c_fmt(pandas_dtype, c_string):  # PR01
+    """Test ``dtype_to_arrow_c_fmt`` utility function."""
+    assert dtype_to_arrow_c_fmt(pandas_dtype) == c_string
+
+
+@pytest.mark.parametrize(
+    "pa_dtype, args_kwargs, c_string",
+    [
+        ["null", {}, "n"],
+        ["bool_", {}, "b"],
+        ["uint8", {}, "C"],
+        ["uint16", {}, "S"],
+        ["uint32", {}, "I"],
+        ["uint64", {}, "L"],
+        ["int8", {}, "c"],
+        ["int16", {}, "S"],
+        ["int32", {}, "i"],
+        ["int64", {}, "l"],
+        ["float16", {}, "e"],
+        ["float32", {}, "f"],
+        ["float64", {}, "g"],
+        ["string", {}, "u"],
+        ["binary", {}, "z"],
+        ["time32", ("s",), "tts"],
+        ["time32", ("ms",), "ttm"],
+        ["time64", ("us",), "ttu"],
+        ["time64", ("ns",), "ttn"],
+        ["date32", {}, "tdD"],
+        ["date64", {}, "tdm"],
+        ["timestamp", {"unit": "s"}, "tss:"],
+        ["timestamp", {"unit": "ms"}, "tsm:"],
+        ["timestamp", {"unit": "us"}, "tsu:"],
+        ["timestamp", {"unit": "ns"}, "tsn:"],
+        ["timestamp", {"unit": "ns", "tz": "UTC"}, "tsn:UTC"],
+        ["duration", ("s",), "tDs"],
+        ["duration", ("ms",), "tDm"],
+        ["duration", ("us",), "tDu"],
+        ["duration", ("ns",), "tDn"],
+        ["decimal128", {"precision": 4, "scale": 2}, "d:4,2"],
+    ],
+)
+def test_dtype_to_arrow_c_fmt_arrowdtype(pa_dtype, args_kwargs, c_string):
+    # GH 52323
+    pa = pytest.importorskip("pyarrow")
+    if not args_kwargs:
+        pa_type = getattr(pa, pa_dtype)()
+    elif isinstance(args_kwargs, tuple):
+        pa_type = getattr(pa, pa_dtype)(*args_kwargs)
+    else:
+        pa_type = getattr(pa, pa_dtype)(**args_kwargs)
+    arrow_type = pd.ArrowDtype(pa_type)
+    assert dtype_to_arrow_c_fmt(arrow_type) == c_string
diff --git a/pandas/pandas/tests/internals/__init__.py b/pandas/pandas/tests/internals/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/internals/test_api.py b/pandas/pandas/tests/internals/test_api.py
new file mode 100644
index 0000000000000..fc222f6987466
--- /dev/null
+++ b/pandas/pandas/tests/internals/test_api.py
@@ -0,0 +1,167 @@
+"""
+Tests for the pseudo-public API implemented in internals/api.py and exposed
+in core.internals
+"""
+
+import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.api.internals import create_dataframe_from_blocks
+from pandas.core import internals
+from pandas.core.internals import api
+
+
+def test_internals_api():
+    assert internals.make_block is api.make_block
+
+
+def test_namespace():
+    # SUBJECT TO CHANGE
+
+    modules = [
+        "blocks",
+        "concat",
+        "managers",
+        "construction",
+        "api",
+        "ops",
+    ]
+    expected = [
+        "make_block",
+        "BlockManager",
+        "SingleBlockManager",
+        "concatenate_managers",
+    ]
+
+    result = [x for x in dir(internals) if not x.startswith("__")]
+    assert set(result) == set(expected + modules)
+
+
+@pytest.mark.parametrize(
+    "name",
+    [
+        "Block",
+        "ExtensionBlock",
+    ],
+)
+def test_deprecations(name):
+    # GH#55139
+    msg = f"{name} is deprecated.* Use public APIs instead"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        getattr(internals, name)
+
+
+def test_make_block_2d_with_dti():
+    # GH#41168
+    dti = pd.date_range("2012", periods=3, tz="UTC")
+
+    msg = "make_block is deprecated"
+    with tm.assert_produces_warning(DeprecationWarning, match=msg):
+        blk = api.make_block(dti, placement=[0])
+
+    assert blk.shape == (1, 3)
+    assert blk.values.shape == (1, 3)
+
+
+def test_create_block_manager_from_blocks_deprecated():
+    # GH#33892
+    # If they must, downstream packages should get this from internals.api,
+    #  not internals.
+    msg = (
+        "create_block_manager_from_blocks is deprecated and will be "
+        "removed in a future version. Use public APIs instead"
+    )
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        internals.create_block_manager_from_blocks
+
+
+def test_create_dataframe_from_blocks(float_frame):
+    block = float_frame._mgr.blocks[0]
+    index = float_frame.index.copy()
+    columns = float_frame.columns.copy()
+
+    result = create_dataframe_from_blocks(
+        [(block.values, block.mgr_locs.as_array)], index=index, columns=columns
+    )
+    tm.assert_frame_equal(result, float_frame)
+
+
+def test_create_dataframe_from_blocks_types():
+    df = pd.DataFrame(
+        {
+            "int": list(range(1, 4)),
+            "uint": np.arange(3, 6).astype("uint8"),
+            "float": [2.0, np.nan, 3.0],
+            "bool": np.array([True, False, True]),
+            "boolean": pd.array([True, False, None], dtype="boolean"),
+            "string": list("abc"),
+            "datetime": pd.date_range("20130101", periods=3),
+            "datetimetz": pd.date_range("20130101", periods=3).tz_localize(
+                "Europe/Brussels"
+            ),
+            "timedelta": pd.timedelta_range("1 day", periods=3),
+            "period": pd.period_range("2012-01-01", periods=3, freq="D"),
+            "categorical": pd.Categorical(["a", "b", "a"]),
+            "interval": pd.IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4)]),
+        }
+    )
+
+    result = create_dataframe_from_blocks(
+        [(block.values, block.mgr_locs.as_array) for block in df._mgr.blocks],
+        index=df.index,
+        columns=df.columns,
+    )
+    tm.assert_frame_equal(result, df)
+
+
+def test_create_dataframe_from_blocks_datetimelike():
+    # extension dtypes that have an exact matching numpy dtype can also be
+    # be passed as a numpy array
+    index, columns = pd.RangeIndex(3), pd.Index(["a", "b", "c", "d"])
+
+    block_array1 = np.arange(
+        datetime.datetime(2020, 1, 1),
+        datetime.datetime(2020, 1, 7),
+        step=datetime.timedelta(1),
+    ).reshape((2, 3))
+    block_array2 = np.arange(
+        datetime.timedelta(1), datetime.timedelta(7), step=datetime.timedelta(1)
+    ).reshape((2, 3))
+    result = create_dataframe_from_blocks(
+        [(block_array1, np.array([0, 2])), (block_array2, np.array([1, 3]))],
+        index=index,
+        columns=columns,
+    )
+    expected = pd.DataFrame(
+        {
+            "a": pd.date_range("2020-01-01", periods=3, unit="us"),
+            "b": pd.timedelta_range("1 days", periods=3, unit="us"),
+            "c": pd.date_range("2020-01-04", periods=3, unit="us"),
+            "d": pd.timedelta_range("4 days", periods=3, unit="us"),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "array",
+    [
+        pd.date_range("2020-01-01", periods=3),
+        pd.date_range("2020-01-01", periods=3, tz="UTC"),
+        pd.period_range("2012-01-01", periods=3, freq="D"),
+        pd.timedelta_range("1 day", periods=3),
+    ],
+)
+def test_create_dataframe_from_blocks_1dEA(array):
+    # ExtensionArrays can be passed as 1D even if stored under the hood as 2D
+    df = pd.DataFrame({"a": array})
+
+    block = df._mgr.blocks[0]
+    result = create_dataframe_from_blocks(
+        [(block.values[0], block.mgr_locs.as_array)], index=df.index, columns=df.columns
+    )
+    tm.assert_frame_equal(result, df)
diff --git a/pandas/pandas/tests/internals/test_internals.py b/pandas/pandas/tests/internals/test_internals.py
new file mode 100644
index 0000000000000..c1f71c6de92dd
--- /dev/null
+++ b/pandas/pandas/tests/internals/test_internals.py
@@ -0,0 +1,1412 @@
+from datetime import (
+    date,
+    datetime,
+)
+import itertools
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.internals import BlockPlacement
+from pandas.compat import IS64
+
+from pandas.core.dtypes.common import is_scalar
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    period_range,
+)
+import pandas._testing as tm
+import pandas.core.algorithms as algos
+from pandas.core.arrays import (
+    DatetimeArray,
+    SparseArray,
+    TimedeltaArray,
+)
+from pandas.core.internals import (
+    BlockManager,
+    SingleBlockManager,
+    make_block,
+)
+from pandas.core.internals.blocks import (
+    ensure_block_shape,
+    maybe_coerce_values,
+    new_block,
+)
+
+
+@pytest.fixture(params=[new_block, make_block])
+def block_maker(request):
+    """
+    Fixture to test both the internal new_block and pseudo-public make_block.
+    """
+    return request.param
+
+
+@pytest.fixture
+def mgr():
+    return create_mgr(
+        "a: f8; b: object; c: f8; d: object; e: f8;"
+        "f: bool; g: i8; h: complex; i: datetime-1; j: datetime-2;"
+        "k: M8[ns, US/Eastern]; l: M8[ns, CET];"
+    )
+
+
+def assert_block_equal(left, right):
+    tm.assert_numpy_array_equal(left.values, right.values)
+    assert left.dtype == right.dtype
+    assert isinstance(left.mgr_locs, BlockPlacement)
+    assert isinstance(right.mgr_locs, BlockPlacement)
+    tm.assert_numpy_array_equal(left.mgr_locs.as_array, right.mgr_locs.as_array)
+
+
+def get_numeric_mat(shape):
+    arr = np.arange(shape[0])
+    return np.lib.stride_tricks.as_strided(
+        x=arr, shape=shape, strides=(arr.itemsize,) + (0,) * (len(shape) - 1)
+    ).copy()
+
+
+N = 10
+
+
+def create_block(typestr, placement, item_shape=None, num_offset=0, maker=new_block):
+    """
+    Supported typestr:
+
+        * float, f8, f4, f2
+        * int, i8, i4, i2, i1
+        * uint, u8, u4, u2, u1
+        * complex, c16, c8
+        * bool
+        * object, string, O
+        * datetime, dt, M8[ns], M8[ns, tz]
+        * timedelta, td, m8[ns]
+        * sparse (SparseArray with fill_value=0.0)
+        * sparse_na (SparseArray with fill_value=np.nan)
+        * category, category2
+
+    """
+    placement = BlockPlacement(placement)
+    num_items = len(placement)
+
+    if item_shape is None:
+        item_shape = (N,)
+
+    shape = (num_items,) + item_shape
+
+    mat = get_numeric_mat(shape)
+
+    if typestr in (
+        "float",
+        "f8",
+        "f4",
+        "f2",
+        "int",
+        "i8",
+        "i4",
+        "i2",
+        "i1",
+        "uint",
+        "u8",
+        "u4",
+        "u2",
+        "u1",
+    ):
+        values = mat.astype(typestr) + num_offset
+    elif typestr in ("complex", "c16", "c8"):
+        values = 1.0j * (mat.astype(typestr) + num_offset)
+    elif typestr in ("object", "string", "O"):
+        values = np.reshape([f"A{i:d}" for i in mat.ravel() + num_offset], shape)
+    elif typestr in ("b", "bool"):
+        values = np.ones(shape, dtype=np.bool_)
+    elif typestr in ("datetime", "dt", "M8[ns]"):
+        values = (mat * 1e9).astype("M8[ns]")
+    elif typestr.startswith("M8[ns"):
+        # datetime with tz
+        m = re.search(r"M8\[ns,\s*(\w+\/?\w*)\]", typestr)
+        assert m is not None, f"incompatible typestr -> {typestr}"
+        tz = m.groups()[0]
+        assert num_items == 1, "must have only 1 num items for a tz-aware"
+        values = DatetimeIndex(np.arange(N) * 10**9, tz=tz)._data
+        values = ensure_block_shape(values, ndim=len(shape))
+    elif typestr in ("timedelta", "td", "m8[ns]"):
+        values = (mat * 1).astype("m8[ns]")
+    elif typestr in ("category",):
+        values = Categorical([1, 1, 2, 2, 3, 3, 3, 3, 4, 4])
+    elif typestr in ("category2",):
+        values = Categorical(["a", "a", "a", "a", "b", "b", "c", "c", "c", "d"])
+    elif typestr in ("sparse", "sparse_na"):
+        if shape[-1] != 10:
+            # We also are implicitly assuming this in the category cases above
+            raise NotImplementedError
+
+        assert all(s == 1 for s in shape[:-1])
+        if typestr.endswith("_na"):
+            fill_value = np.nan
+        else:
+            fill_value = 0.0
+        values = SparseArray(
+            [fill_value, fill_value, 1, 2, 3, fill_value, 4, 5, fill_value, 6],
+            fill_value=fill_value,
+        )
+        arr = values.sp_values.view()
+        arr += num_offset - 1
+    else:
+        raise ValueError(f'Unsupported typestr: "{typestr}"')
+
+    values = maybe_coerce_values(values)
+    return maker(values, placement=placement, ndim=len(shape))
+
+
+def create_single_mgr(typestr, num_rows=None):
+    if num_rows is None:
+        num_rows = N
+
+    return SingleBlockManager(
+        create_block(typestr, placement=slice(0, num_rows), item_shape=()),
+        Index(np.arange(num_rows)),
+    )
+
+
+def create_mgr(descr, item_shape=None):
+    """
+    Construct BlockManager from string description.
+
+    String description syntax looks similar to np.matrix initializer.  It looks
+    like this::
+
+        a,b,c: f8; d,e,f: i8
+
+    Rules are rather simple:
+
+    * see list of supported datatypes in `create_block` method
+    * components are semicolon-separated
+    * each component is `NAME,NAME,NAME: DTYPE_ID`
+    * whitespace around colons & semicolons are removed
+    * components with same DTYPE_ID are combined into single block
+    * to force multiple blocks with same dtype, use '-SUFFIX'::
+
+        "a:f8-1; b:f8-2; c:f8-foobar"
+
+    """
+    if item_shape is None:
+        item_shape = (N,)
+
+    offset = 0
+    mgr_items = []
+    block_placements = {}
+    for d in descr.split(";"):
+        d = d.strip()
+        if not len(d):
+            continue
+        names, blockstr = d.partition(":")[::2]
+        blockstr = blockstr.strip()
+        names = names.strip().split(",")
+
+        mgr_items.extend(names)
+        placement = list(np.arange(len(names)) + offset)
+        try:
+            block_placements[blockstr].extend(placement)
+        except KeyError:
+            block_placements[blockstr] = placement
+        offset += len(names)
+
+    mgr_items = Index(mgr_items)
+
+    blocks = []
+    num_offset = 0
+    for blockstr, placement in block_placements.items():
+        typestr = blockstr.split("-")[0]
+        blocks.append(
+            create_block(
+                typestr, placement, item_shape=item_shape, num_offset=num_offset
+            )
+        )
+        num_offset += len(placement)
+
+    sblocks = sorted(blocks, key=lambda b: b.mgr_locs[0])
+    return BlockManager(
+        tuple(sblocks),
+        [mgr_items] + [Index(np.arange(n)) for n in item_shape],
+    )
+
+
+@pytest.fixture
+def fblock():
+    return create_block("float", [0, 2, 4])
+
+
+class TestBlock:
+    def test_constructor(self):
+        int32block = create_block("i4", [0])
+        assert int32block.dtype == np.int32
+
+    @pytest.mark.parametrize(
+        "typ, data",
+        [
+            ["float", [0, 2, 4]],
+            ["complex", [7]],
+            ["object", [1, 3]],
+            ["bool", [5]],
+        ],
+    )
+    def test_pickle(self, typ, data):
+        blk = create_block(typ, data)
+        assert_block_equal(tm.round_trip_pickle(blk), blk)
+
+    def test_mgr_locs(self, fblock):
+        assert isinstance(fblock.mgr_locs, BlockPlacement)
+        tm.assert_numpy_array_equal(
+            fblock.mgr_locs.as_array, np.array([0, 2, 4], dtype=np.intp)
+        )
+
+    def test_attrs(self, fblock):
+        assert fblock.shape == fblock.values.shape
+        assert fblock.dtype == fblock.values.dtype
+        assert len(fblock) == len(fblock.values)
+
+    def test_copy(self, fblock):
+        cop = fblock.copy()
+        assert cop is not fblock
+        assert_block_equal(fblock, cop)
+
+    def test_delete(self, fblock):
+        newb = fblock.copy()
+        locs = newb.mgr_locs
+        nb = newb.delete(0)[0]
+        assert newb.mgr_locs is locs
+
+        assert nb is not newb
+
+        tm.assert_numpy_array_equal(
+            nb.mgr_locs.as_array, np.array([2, 4], dtype=np.intp)
+        )
+        assert not (newb.values[0] == 1).all()
+        assert (nb.values[0] == 1).all()
+
+        newb = fblock.copy()
+        locs = newb.mgr_locs
+        nb = newb.delete(1)
+        assert len(nb) == 2
+        assert newb.mgr_locs is locs
+
+        tm.assert_numpy_array_equal(
+            nb[0].mgr_locs.as_array, np.array([0], dtype=np.intp)
+        )
+        tm.assert_numpy_array_equal(
+            nb[1].mgr_locs.as_array, np.array([4], dtype=np.intp)
+        )
+        assert not (newb.values[1] == 2).all()
+        assert (nb[1].values[0] == 2).all()
+
+        newb = fblock.copy()
+        nb = newb.delete(2)
+        assert len(nb) == 1
+        tm.assert_numpy_array_equal(
+            nb[0].mgr_locs.as_array, np.array([0, 2], dtype=np.intp)
+        )
+        assert (nb[0].values[1] == 1).all()
+
+        newb = fblock.copy()
+
+        with pytest.raises(IndexError, match=None):
+            newb.delete(3)
+
+    def test_delete_datetimelike(self):
+        # dont use np.delete on values, as that will coerce from DTA/TDA to ndarray
+        arr = np.arange(20, dtype="i8").reshape(5, 4).view("m8[ns]")
+        df = DataFrame(arr)
+        blk = df._mgr.blocks[0]
+        assert isinstance(blk.values, TimedeltaArray)
+
+        nb = blk.delete(1)
+        assert len(nb) == 2
+        assert isinstance(nb[0].values, TimedeltaArray)
+        assert isinstance(nb[1].values, TimedeltaArray)
+
+        df = DataFrame(arr.view("M8[ns]"))
+        blk = df._mgr.blocks[0]
+        assert isinstance(blk.values, DatetimeArray)
+
+        nb = blk.delete([1, 3])
+        assert len(nb) == 2
+        assert isinstance(nb[0].values, DatetimeArray)
+        assert isinstance(nb[1].values, DatetimeArray)
+
+    def test_split(self):
+        # GH#37799
+        values = np.random.default_rng(2).standard_normal((3, 4))
+        blk = new_block(values, placement=BlockPlacement([3, 1, 6]), ndim=2)
+        result = list(blk._split())
+
+        # check that we get views, not copies
+        values[:] = -9999
+        assert (blk.values == -9999).all()
+
+        assert len(result) == 3
+        expected = [
+            new_block(values[[0]], placement=BlockPlacement([3]), ndim=2),
+            new_block(values[[1]], placement=BlockPlacement([1]), ndim=2),
+            new_block(values[[2]], placement=BlockPlacement([6]), ndim=2),
+        ]
+        for res, exp in zip(result, expected):
+            assert_block_equal(res, exp)
+
+
+class TestBlockManager:
+    def test_attrs(self):
+        mgr = create_mgr("a,b,c: f8-1; d,e,f: f8-2")
+        assert mgr.nblocks == 2
+        assert len(mgr) == 6
+
+    def test_duplicate_ref_loc_failure(self):
+        tmp_mgr = create_mgr("a:bool; a: f8")
+
+        axes, blocks = tmp_mgr.axes, tmp_mgr.blocks
+
+        blocks[0].mgr_locs = BlockPlacement(np.array([0]))
+        blocks[1].mgr_locs = BlockPlacement(np.array([0]))
+
+        # test trying to create block manager with overlapping ref locs
+
+        msg = "Gaps in blk ref_locs"
+
+        mgr = BlockManager(blocks, axes)
+        with pytest.raises(AssertionError, match=msg):
+            mgr._rebuild_blknos_and_blklocs()
+
+        blocks[0].mgr_locs = BlockPlacement(np.array([0]))
+        blocks[1].mgr_locs = BlockPlacement(np.array([1]))
+        mgr = BlockManager(blocks, axes)
+        mgr.iget(1)
+
+    def test_pickle(self, mgr):
+        mgr2 = tm.round_trip_pickle(mgr)
+        tm.assert_frame_equal(
+            DataFrame._from_mgr(mgr, axes=mgr.axes),
+            DataFrame._from_mgr(mgr2, axes=mgr2.axes),
+        )
+
+        # GH2431
+        assert hasattr(mgr2, "_is_consolidated")
+        assert hasattr(mgr2, "_known_consolidated")
+
+        # reset to False on load
+        assert not mgr2._is_consolidated
+        assert not mgr2._known_consolidated
+
+    @pytest.mark.parametrize("mgr_string", ["a,a,a:f8", "a: f8; a: i8"])
+    def test_non_unique_pickle(self, mgr_string):
+        mgr = create_mgr(mgr_string)
+        mgr2 = tm.round_trip_pickle(mgr)
+        tm.assert_frame_equal(
+            DataFrame._from_mgr(mgr, axes=mgr.axes),
+            DataFrame._from_mgr(mgr2, axes=mgr2.axes),
+        )
+
+    def test_categorical_block_pickle(self):
+        mgr = create_mgr("a: category")
+        mgr2 = tm.round_trip_pickle(mgr)
+        tm.assert_frame_equal(
+            DataFrame._from_mgr(mgr, axes=mgr.axes),
+            DataFrame._from_mgr(mgr2, axes=mgr2.axes),
+        )
+
+        smgr = create_single_mgr("category")
+        smgr2 = tm.round_trip_pickle(smgr)
+        tm.assert_series_equal(
+            Series()._constructor_from_mgr(smgr, axes=smgr.axes),
+            Series()._constructor_from_mgr(smgr2, axes=smgr2.axes),
+        )
+
+    def test_iget(self):
+        cols = Index(list("abc"))
+        values = np.random.default_rng(2).random((3, 3))
+        block = new_block(
+            values=values.copy(),
+            placement=BlockPlacement(np.arange(3, dtype=np.intp)),
+            ndim=values.ndim,
+        )
+        mgr = BlockManager(blocks=(block,), axes=[cols, Index(np.arange(3))])
+
+        tm.assert_almost_equal(mgr.iget(0).internal_values(), values[0])
+        tm.assert_almost_equal(mgr.iget(1).internal_values(), values[1])
+        tm.assert_almost_equal(mgr.iget(2).internal_values(), values[2])
+
+    def test_set(self):
+        mgr = create_mgr("a,b,c: int", item_shape=(3,))
+
+        mgr.insert(len(mgr.items), "d", np.array(["foo"] * 3))
+        mgr.iset(1, np.array(["bar"] * 3))
+        tm.assert_numpy_array_equal(mgr.iget(0).internal_values(), np.array([0] * 3))
+        tm.assert_numpy_array_equal(
+            mgr.iget(1).internal_values(), np.array(["bar"] * 3, dtype=np.object_)
+        )
+        tm.assert_numpy_array_equal(mgr.iget(2).internal_values(), np.array([2] * 3))
+        tm.assert_numpy_array_equal(
+            mgr.iget(3).internal_values(), np.array(["foo"] * 3, dtype=np.object_)
+        )
+
+    def test_set_change_dtype(self, mgr):
+        mgr.insert(len(mgr.items), "baz", np.zeros(N, dtype=bool))
+
+        mgr.iset(mgr.items.get_loc("baz"), np.repeat("foo", N))
+        idx = mgr.items.get_loc("baz")
+        assert mgr.iget(idx).dtype == np.object_
+
+        mgr2 = mgr.consolidate()
+        mgr2.iset(mgr2.items.get_loc("baz"), np.repeat("foo", N))
+        idx = mgr2.items.get_loc("baz")
+        assert mgr2.iget(idx).dtype == np.object_
+
+        mgr2.insert(
+            len(mgr2.items),
+            "quux",
+            np.random.default_rng(2).standard_normal(N).astype(int),
+        )
+        idx = mgr2.items.get_loc("quux")
+        assert mgr2.iget(idx).dtype == np.dtype(int)
+
+        mgr2.iset(
+            mgr2.items.get_loc("quux"), np.random.default_rng(2).standard_normal(N)
+        )
+        assert mgr2.iget(idx).dtype == np.float64
+
+    def test_copy(self, mgr):
+        cp = mgr.copy(deep=False)
+        for blk, cp_blk in zip(mgr.blocks, cp.blocks):
+            # view assertion
+            tm.assert_equal(cp_blk.values, blk.values)
+            if isinstance(blk.values, np.ndarray):
+                assert cp_blk.values.base is blk.values.base
+            else:
+                # DatetimeTZBlock has DatetimeIndex values
+                assert cp_blk.values._ndarray.base is blk.values._ndarray.base
+
+        # copy(deep=True) consolidates, so the block-wise assertions will
+        #  fail is mgr is not consolidated
+        mgr._consolidate_inplace()
+        cp = mgr.copy(deep=True)
+        for blk, cp_blk in zip(mgr.blocks, cp.blocks):
+            bvals = blk.values
+            cpvals = cp_blk.values
+
+            tm.assert_equal(cpvals, bvals)
+
+            if isinstance(cpvals, np.ndarray):
+                lbase = cpvals.base
+                rbase = bvals.base
+            else:
+                lbase = cpvals._ndarray.base
+                rbase = bvals._ndarray.base
+
+            # copy assertion we either have a None for a base or in case of
+            # some blocks it is an array (e.g. datetimetz), but was copied
+            if isinstance(cpvals, DatetimeArray):
+                assert (lbase is None and rbase is None) or (lbase is not rbase)
+            elif not isinstance(cpvals, np.ndarray):
+                assert lbase is not rbase
+            else:
+                assert lbase is None and rbase is None
+
+    def test_sparse(self):
+        mgr = create_mgr("a: sparse-1; b: sparse-2")
+        assert mgr.as_array().dtype == np.float64
+
+    def test_sparse_mixed(self):
+        mgr = create_mgr("a: sparse-1; b: sparse-2; c: f8")
+        assert len(mgr.blocks) == 3
+        assert isinstance(mgr, BlockManager)
+
+    @pytest.mark.parametrize(
+        "mgr_string, dtype",
+        [("c: f4; d: f2", np.float32), ("c: f4; d: f2; e: f8", np.float64)],
+    )
+    def test_as_array_float(self, mgr_string, dtype):
+        mgr = create_mgr(mgr_string)
+        assert mgr.as_array().dtype == dtype
+
+    @pytest.mark.parametrize(
+        "mgr_string, dtype",
+        [
+            ("a: bool-1; b: bool-2", np.bool_),
+            ("a: i8-1; b: i8-2; c: i4; d: i2; e: u1", np.int64),
+            ("c: i4; d: i2; e: u1", np.int32),
+        ],
+    )
+    def test_as_array_int_bool(self, mgr_string, dtype):
+        mgr = create_mgr(mgr_string)
+        assert mgr.as_array().dtype == dtype
+
+    def test_as_array_datetime(self):
+        mgr = create_mgr("h: datetime-1; g: datetime-2")
+        assert mgr.as_array().dtype == "M8[ns]"
+
+    def test_as_array_datetime_tz(self):
+        mgr = create_mgr("h: M8[ns, US/Eastern]; g: M8[ns, CET]")
+        assert mgr.iget(0).dtype == "datetime64[ns, US/Eastern]"
+        assert mgr.iget(1).dtype == "datetime64[ns, CET]"
+        assert mgr.as_array().dtype == "object"
+
+    @pytest.mark.parametrize("t", ["float16", "float32", "float64", "int32", "int64"])
+    def test_astype(self, t):
+        # coerce all
+        mgr = create_mgr("c: f4; d: f2; e: f8")
+
+        t = np.dtype(t)
+        tmgr = mgr.astype(t)
+        assert tmgr.iget(0).dtype.type == t
+        assert tmgr.iget(1).dtype.type == t
+        assert tmgr.iget(2).dtype.type == t
+
+        # mixed
+        mgr = create_mgr("a,b: object; c: bool; d: datetime; e: f4; f: f2; g: f8")
+
+        t = np.dtype(t)
+        tmgr = mgr.astype(t, errors="ignore")
+        assert tmgr.iget(2).dtype.type == t
+        assert tmgr.iget(4).dtype.type == t
+        assert tmgr.iget(5).dtype.type == t
+        assert tmgr.iget(6).dtype.type == t
+
+        assert tmgr.iget(0).dtype.type == np.object_
+        assert tmgr.iget(1).dtype.type == np.object_
+        if t != np.int64:
+            assert tmgr.iget(3).dtype.type == np.datetime64
+        else:
+            assert tmgr.iget(3).dtype.type == t
+
+    def test_convert(self, using_infer_string):
+        def _compare(old_mgr, new_mgr):
+            """compare the blocks, numeric compare ==, object don't"""
+            old_blocks = set(old_mgr.blocks)
+            new_blocks = set(new_mgr.blocks)
+            assert len(old_blocks) == len(new_blocks)
+
+            # compare non-numeric
+            for b in old_blocks:
+                found = False
+                for nb in new_blocks:
+                    if (b.values == nb.values).all():
+                        found = True
+                        break
+                assert found
+
+            for b in new_blocks:
+                found = False
+                for ob in old_blocks:
+                    if (b.values == ob.values).all():
+                        found = True
+                        break
+                assert found
+
+        # noops
+        mgr = create_mgr("f: i8; g: f8")
+        new_mgr = mgr.convert()
+        _compare(mgr, new_mgr)
+
+        # convert
+        mgr = create_mgr("a,b,foo: object; f: i8; g: f8")
+        mgr.iset(0, np.array(["1"] * N, dtype=np.object_))
+        mgr.iset(1, np.array(["2."] * N, dtype=np.object_))
+        mgr.iset(2, np.array(["foo."] * N, dtype=np.object_))
+        new_mgr = mgr.convert()
+        dtype = "str" if using_infer_string else np.object_
+        assert new_mgr.iget(0).dtype == dtype
+        assert new_mgr.iget(1).dtype == dtype
+        assert new_mgr.iget(2).dtype == dtype
+        assert new_mgr.iget(3).dtype == np.int64
+        assert new_mgr.iget(4).dtype == np.float64
+
+        mgr = create_mgr(
+            "a,b,foo: object; f: i4; bool: bool; dt: datetime; i: i8; g: f8; h: f2"
+        )
+        mgr.iset(0, np.array(["1"] * N, dtype=np.object_))
+        mgr.iset(1, np.array(["2."] * N, dtype=np.object_))
+        mgr.iset(2, np.array(["foo."] * N, dtype=np.object_))
+        new_mgr = mgr.convert()
+        assert new_mgr.iget(0).dtype == dtype
+        assert new_mgr.iget(1).dtype == dtype
+        assert new_mgr.iget(2).dtype == dtype
+        assert new_mgr.iget(3).dtype == np.int32
+        assert new_mgr.iget(4).dtype == np.bool_
+        assert new_mgr.iget(5).dtype.type, np.datetime64
+        assert new_mgr.iget(6).dtype == np.int64
+        assert new_mgr.iget(7).dtype == np.float64
+        assert new_mgr.iget(8).dtype == np.float16
+
+    def test_interleave(self):
+        # self
+        for dtype in ["f8", "i8", "object", "bool", "complex", "M8[ns]", "m8[ns]"]:
+            mgr = create_mgr(f"a: {dtype}")
+            assert mgr.as_array().dtype == dtype
+            mgr = create_mgr(f"a: {dtype}; b: {dtype}")
+            assert mgr.as_array().dtype == dtype
+
+    @pytest.mark.parametrize(
+        "mgr_string, dtype",
+        [
+            ("a: category", "i8"),
+            ("a: category; b: category", "i8"),
+            ("a: category; b: category2", "object"),
+            ("a: category2", "object"),
+            ("a: category2; b: category2", "object"),
+            ("a: f8", "f8"),
+            ("a: f8; b: i8", "f8"),
+            ("a: f4; b: i8", "f8"),
+            ("a: f4; b: i8; d: object", "object"),
+            ("a: bool; b: i8", "object"),
+            ("a: complex", "complex"),
+            ("a: f8; b: category", "object"),
+            ("a: M8[ns]; b: category", "object"),
+            ("a: M8[ns]; b: bool", "object"),
+            ("a: M8[ns]; b: i8", "object"),
+            ("a: m8[ns]; b: bool", "object"),
+            ("a: m8[ns]; b: i8", "object"),
+            ("a: M8[ns]; b: m8[ns]", "object"),
+        ],
+    )
+    def test_interleave_dtype(self, mgr_string, dtype):
+        # will be converted according the actual dtype of the underlying
+        mgr = create_mgr("a: category")
+        assert mgr.as_array().dtype == "i8"
+        mgr = create_mgr("a: category; b: category2")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: category2")
+        assert mgr.as_array().dtype == "object"
+
+        # combinations
+        mgr = create_mgr("a: f8")
+        assert mgr.as_array().dtype == "f8"
+        mgr = create_mgr("a: f8; b: i8")
+        assert mgr.as_array().dtype == "f8"
+        mgr = create_mgr("a: f4; b: i8")
+        assert mgr.as_array().dtype == "f8"
+        mgr = create_mgr("a: f4; b: i8; d: object")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: bool; b: i8")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: complex")
+        assert mgr.as_array().dtype == "complex"
+        mgr = create_mgr("a: f8; b: category")
+        assert mgr.as_array().dtype == "f8"
+        mgr = create_mgr("a: M8[ns]; b: category")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: M8[ns]; b: bool")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: M8[ns]; b: i8")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: m8[ns]; b: bool")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: m8[ns]; b: i8")
+        assert mgr.as_array().dtype == "object"
+        mgr = create_mgr("a: M8[ns]; b: m8[ns]")
+        assert mgr.as_array().dtype == "object"
+
+    def test_consolidate_ordering_issues(self, mgr):
+        mgr.iset(mgr.items.get_loc("f"), np.random.default_rng(2).standard_normal(N))
+        mgr.iset(mgr.items.get_loc("d"), np.random.default_rng(2).standard_normal(N))
+        mgr.iset(mgr.items.get_loc("b"), np.random.default_rng(2).standard_normal(N))
+        mgr.iset(mgr.items.get_loc("g"), np.random.default_rng(2).standard_normal(N))
+        mgr.iset(mgr.items.get_loc("h"), np.random.default_rng(2).standard_normal(N))
+
+        # we have datetime/tz blocks in mgr
+        cons = mgr.consolidate()
+        assert cons.nblocks == 4
+        cons = mgr.consolidate().get_numeric_data()
+        assert cons.nblocks == 1
+        assert isinstance(cons.blocks[0].mgr_locs, BlockPlacement)
+        tm.assert_numpy_array_equal(
+            cons.blocks[0].mgr_locs.as_array, np.arange(len(cons.items), dtype=np.intp)
+        )
+
+    def test_reindex_items(self):
+        # mgr is not consolidated, f8 & f8-2 blocks
+        mgr = create_mgr("a: f8; b: i8; c: f8; d: i8; e: f8; f: bool; g: f8-2")
+
+        reindexed = mgr.reindex_axis(["g", "c", "a", "d"], axis=0)
+        # reindex_axis does not consolidate_inplace, as that risks failing to
+        #  invalidate _item_cache
+        assert not reindexed.is_consolidated()
+
+        tm.assert_index_equal(reindexed.items, Index(["g", "c", "a", "d"]))
+        tm.assert_almost_equal(
+            mgr.iget(6).internal_values(), reindexed.iget(0).internal_values()
+        )
+        tm.assert_almost_equal(
+            mgr.iget(2).internal_values(), reindexed.iget(1).internal_values()
+        )
+        tm.assert_almost_equal(
+            mgr.iget(0).internal_values(), reindexed.iget(2).internal_values()
+        )
+        tm.assert_almost_equal(
+            mgr.iget(3).internal_values(), reindexed.iget(3).internal_values()
+        )
+
+    def test_get_numeric_data(self):
+        mgr = create_mgr(
+            "int: int; float: float; complex: complex;"
+            "str: object; bool: bool; obj: object; dt: datetime",
+            item_shape=(3,),
+        )
+        mgr.iset(5, np.array([1, 2, 3], dtype=np.object_))
+
+        numeric = mgr.get_numeric_data()
+        tm.assert_index_equal(numeric.items, Index(["int", "float", "complex", "bool"]))
+        tm.assert_almost_equal(
+            mgr.iget(mgr.items.get_loc("float")).internal_values(),
+            numeric.iget(numeric.items.get_loc("float")).internal_values(),
+        )
+
+        # Check sharing
+        numeric.iset(
+            numeric.items.get_loc("float"),
+            np.array([100.0, 200.0, 300.0]),
+            inplace=True,
+        )
+        tm.assert_almost_equal(
+            mgr.iget(mgr.items.get_loc("float")).internal_values(),
+            np.array([1.0, 1.0, 1.0]),
+        )
+
+    def test_get_bool_data(self):
+        mgr = create_mgr(
+            "int: int; float: float; complex: complex;"
+            "str: object; bool: bool; obj: object; dt: datetime",
+            item_shape=(3,),
+        )
+        mgr.iset(6, np.array([True, False, True], dtype=np.object_))
+
+        bools = mgr.get_bool_data()
+        tm.assert_index_equal(bools.items, Index(["bool"]))
+        tm.assert_almost_equal(
+            mgr.iget(mgr.items.get_loc("bool")).internal_values(),
+            bools.iget(bools.items.get_loc("bool")).internal_values(),
+        )
+
+        bools.iset(0, np.array([True, False, True]), inplace=True)
+        tm.assert_numpy_array_equal(
+            mgr.iget(mgr.items.get_loc("bool")).internal_values(),
+            np.array([True, True, True]),
+        )
+
+    def test_unicode_repr_doesnt_raise(self):
+        repr(create_mgr("b,\u05d0: object"))
+
+    @pytest.mark.parametrize(
+        "mgr_string", ["a,b,c: i8-1; d,e,f: i8-2", "a,a,a: i8-1; b,b,b: i8-2"]
+    )
+    def test_equals(self, mgr_string):
+        # unique items
+        bm1 = create_mgr(mgr_string)
+        bm2 = BlockManager(bm1.blocks[::-1], bm1.axes)
+        assert bm1.equals(bm2)
+
+    @pytest.mark.parametrize(
+        "mgr_string",
+        [
+            "a:i8;b:f8",  # basic case
+            "a:i8;b:f8;c:c8;d:b",  # many types
+            "a:i8;e:dt;f:td;g:string",  # more types
+            "a:i8;b:category;c:category2",  # categories
+            "c:sparse;d:sparse_na;b:f8",  # sparse
+        ],
+    )
+    def test_equals_block_order_different_dtypes(self, mgr_string):
+        # GH 9330
+        bm = create_mgr(mgr_string)
+        block_perms = itertools.permutations(bm.blocks)
+        for bm_perm in block_perms:
+            bm_this = BlockManager(tuple(bm_perm), bm.axes)
+            assert bm.equals(bm_this)
+            assert bm_this.equals(bm)
+
+    def test_single_mgr_ctor(self):
+        mgr = create_single_mgr("f8", num_rows=5)
+        assert mgr.external_values().tolist() == [0.0, 1.0, 2.0, 3.0, 4.0]
+
+    @pytest.mark.parametrize("value", [1, "True", [1, 2, 3], 5.0])
+    def test_validate_bool_args(self, value):
+        bm1 = create_mgr("a,b,c: i8-1; d,e,f: i8-2")
+
+        msg = (
+            'For argument "inplace" expected type bool, '
+            f"received type {type(value).__name__}."
+        )
+        with pytest.raises(ValueError, match=msg):
+            bm1.replace_list([1], [2], inplace=value)
+
+    def test_iset_split_block(self):
+        bm = create_mgr("a,b,c: i8; d: f8")
+        bm._iset_split_block(0, np.array([0]))
+        tm.assert_numpy_array_equal(
+            bm.blklocs, np.array([0, 0, 1, 0], dtype="int64" if IS64 else "int32")
+        )
+        # First indexer currently does not have a block associated with it in case
+        tm.assert_numpy_array_equal(
+            bm.blknos, np.array([0, 0, 0, 1], dtype="int64" if IS64 else "int32")
+        )
+        assert len(bm.blocks) == 2
+
+    def test_iset_split_block_values(self):
+        bm = create_mgr("a,b,c: i8; d: f8")
+        bm._iset_split_block(0, np.array([0]), np.array([list(range(10))]))
+        tm.assert_numpy_array_equal(
+            bm.blklocs, np.array([0, 0, 1, 0], dtype="int64" if IS64 else "int32")
+        )
+        # First indexer currently does not have a block associated with it in case
+        tm.assert_numpy_array_equal(
+            bm.blknos, np.array([0, 2, 2, 1], dtype="int64" if IS64 else "int32")
+        )
+        assert len(bm.blocks) == 3
+
+
+def _as_array(mgr):
+    if mgr.ndim == 1:
+        return mgr.external_values()
+    return mgr.as_array().T
+
+
+class TestIndexing:
+    # Nosetests-style data-driven tests.
+    #
+    # This test applies different indexing routines to block managers and
+    # compares the outcome to the result of same operations on np.ndarray.
+    #
+    # NOTE: sparse (SparseBlock with fill_value != np.nan) fail a lot of tests
+    #       and are disabled.
+
+    MANAGERS = [
+        create_single_mgr("f8", N),
+        create_single_mgr("i8", N),
+        # 2-dim
+        create_mgr("a,b,c,d,e,f: f8", item_shape=(N,)),
+        create_mgr("a,b,c,d,e,f: i8", item_shape=(N,)),
+        create_mgr("a,b: f8; c,d: i8; e,f: string", item_shape=(N,)),
+        create_mgr("a,b: f8; c,d: i8; e,f: f8", item_shape=(N,)),
+    ]
+
+    @pytest.mark.parametrize("mgr", MANAGERS)
+    def test_get_slice(self, mgr):
+        def assert_slice_ok(mgr, axis, slobj):
+            mat = _as_array(mgr)
+
+            # we maybe using an ndarray to test slicing and
+            # might not be the full length of the axis
+            if isinstance(slobj, np.ndarray):
+                ax = mgr.axes[axis]
+                if len(ax) and len(slobj) and len(slobj) != len(ax):
+                    slobj = np.concatenate(
+                        [slobj, np.zeros(len(ax) - len(slobj), dtype=bool)]
+                    )
+
+            if isinstance(slobj, slice):
+                sliced = mgr.get_slice(slobj, axis=axis)
+            elif (
+                mgr.ndim == 1
+                and axis == 0
+                and isinstance(slobj, np.ndarray)
+                and slobj.dtype == bool
+            ):
+                sliced = mgr.get_rows_with_mask(slobj)
+            else:
+                # BlockManager doesn't support non-slice, SingleBlockManager
+                #  doesn't support axis > 0
+                raise TypeError(slobj)
+
+            mat_slobj = (slice(None),) * axis + (slobj,)
+            tm.assert_numpy_array_equal(
+                mat[mat_slobj], _as_array(sliced), check_dtype=False
+            )
+            tm.assert_index_equal(mgr.axes[axis][slobj], sliced.axes[axis])
+
+        assert mgr.ndim <= 2, mgr.ndim
+        for ax in range(mgr.ndim):
+            # slice
+            assert_slice_ok(mgr, ax, slice(None))
+            assert_slice_ok(mgr, ax, slice(3))
+            assert_slice_ok(mgr, ax, slice(100))
+            assert_slice_ok(mgr, ax, slice(1, 4))
+            assert_slice_ok(mgr, ax, slice(3, 0, -2))
+
+            if mgr.ndim < 2:
+                # 2D only support slice objects
+
+                # boolean mask
+                assert_slice_ok(mgr, ax, np.ones(mgr.shape[ax], dtype=np.bool_))
+                assert_slice_ok(mgr, ax, np.zeros(mgr.shape[ax], dtype=np.bool_))
+
+                if mgr.shape[ax] >= 3:
+                    assert_slice_ok(mgr, ax, np.arange(mgr.shape[ax]) % 3 == 0)
+                    assert_slice_ok(
+                        mgr, ax, np.array([True, True, False], dtype=np.bool_)
+                    )
+
+    @pytest.mark.parametrize("mgr", MANAGERS)
+    def test_take(self, mgr):
+        def assert_take_ok(mgr, axis, indexer):
+            mat = _as_array(mgr)
+            taken = mgr.take(indexer, axis)
+            tm.assert_numpy_array_equal(
+                np.take(mat, indexer, axis), _as_array(taken), check_dtype=False
+            )
+            tm.assert_index_equal(mgr.axes[axis].take(indexer), taken.axes[axis])
+
+        for ax in range(mgr.ndim):
+            # take/fancy indexer
+            assert_take_ok(mgr, ax, indexer=np.array([], dtype=np.intp))
+            assert_take_ok(mgr, ax, indexer=np.array([0, 0, 0], dtype=np.intp))
+            assert_take_ok(
+                mgr, ax, indexer=np.array(list(range(mgr.shape[ax])), dtype=np.intp)
+            )
+
+            if mgr.shape[ax] >= 3:
+                assert_take_ok(mgr, ax, indexer=np.array([0, 1, 2], dtype=np.intp))
+                assert_take_ok(mgr, ax, indexer=np.array([-1, -2, -3], dtype=np.intp))
+
+    @pytest.mark.parametrize("mgr", MANAGERS)
+    @pytest.mark.parametrize("fill_value", [None, np.nan, 100.0])
+    def test_reindex_axis(self, fill_value, mgr):
+        def assert_reindex_axis_is_ok(mgr, axis, new_labels, fill_value):
+            mat = _as_array(mgr)
+            indexer = mgr.axes[axis].get_indexer_for(new_labels)
+
+            reindexed = mgr.reindex_axis(new_labels, axis, fill_value=fill_value)
+            tm.assert_numpy_array_equal(
+                algos.take_nd(mat, indexer, axis, fill_value=fill_value),
+                _as_array(reindexed),
+                check_dtype=False,
+            )
+            tm.assert_index_equal(reindexed.axes[axis], new_labels)
+
+        for ax in range(mgr.ndim):
+            assert_reindex_axis_is_ok(mgr, ax, Index([]), fill_value)
+            assert_reindex_axis_is_ok(mgr, ax, mgr.axes[ax], fill_value)
+            assert_reindex_axis_is_ok(mgr, ax, mgr.axes[ax][[0, 0, 0]], fill_value)
+            assert_reindex_axis_is_ok(mgr, ax, Index(["foo", "bar", "baz"]), fill_value)
+            assert_reindex_axis_is_ok(
+                mgr, ax, Index(["foo", mgr.axes[ax][0], "baz"]), fill_value
+            )
+
+            if mgr.shape[ax] >= 3:
+                assert_reindex_axis_is_ok(mgr, ax, mgr.axes[ax][:-3], fill_value)
+                assert_reindex_axis_is_ok(mgr, ax, mgr.axes[ax][-3::-1], fill_value)
+                assert_reindex_axis_is_ok(
+                    mgr, ax, mgr.axes[ax][[0, 1, 2, 0, 1, 2]], fill_value
+                )
+
+    @pytest.mark.parametrize("mgr", MANAGERS)
+    @pytest.mark.parametrize("fill_value", [None, np.nan, 100.0])
+    def test_reindex_indexer(self, fill_value, mgr):
+        def assert_reindex_indexer_is_ok(mgr, axis, new_labels, indexer, fill_value):
+            mat = _as_array(mgr)
+            reindexed_mat = algos.take_nd(mat, indexer, axis, fill_value=fill_value)
+            reindexed = mgr.reindex_indexer(
+                new_labels, indexer, axis, fill_value=fill_value
+            )
+            tm.assert_numpy_array_equal(
+                reindexed_mat, _as_array(reindexed), check_dtype=False
+            )
+            tm.assert_index_equal(reindexed.axes[axis], new_labels)
+
+        for ax in range(mgr.ndim):
+            assert_reindex_indexer_is_ok(
+                mgr, ax, Index([]), np.array([], dtype=np.intp), fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr, ax, mgr.axes[ax], np.arange(mgr.shape[ax]), fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr,
+                ax,
+                Index(["foo"] * mgr.shape[ax]),
+                np.arange(mgr.shape[ax]),
+                fill_value,
+            )
+            assert_reindex_indexer_is_ok(
+                mgr, ax, mgr.axes[ax][::-1], np.arange(mgr.shape[ax]), fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr, ax, mgr.axes[ax], np.arange(mgr.shape[ax])[::-1], fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr, ax, Index(["foo", "bar", "baz"]), np.array([0, 0, 0]), fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr, ax, Index(["foo", "bar", "baz"]), np.array([-1, 0, -1]), fill_value
+            )
+            assert_reindex_indexer_is_ok(
+                mgr,
+                ax,
+                Index(["foo", mgr.axes[ax][0], "baz"]),
+                np.array([-1, -1, -1]),
+                fill_value,
+            )
+
+            if mgr.shape[ax] >= 3:
+                assert_reindex_indexer_is_ok(
+                    mgr,
+                    ax,
+                    Index(["foo", "bar", "baz"]),
+                    np.array([0, 1, 2]),
+                    fill_value,
+                )
+
+
+class TestBlockPlacement:
+    @pytest.mark.parametrize(
+        "slc, expected",
+        [
+            (slice(0, 4), 4),
+            (slice(0, 4, 2), 2),
+            (slice(0, 3, 2), 2),
+            (slice(0, 1, 2), 1),
+            (slice(1, 0, -1), 1),
+        ],
+    )
+    def test_slice_len(self, slc, expected):
+        assert len(BlockPlacement(slc)) == expected
+
+    @pytest.mark.parametrize("slc", [slice(1, 1, 0), slice(1, 2, 0)])
+    def test_zero_step_raises(self, slc):
+        msg = "slice step cannot be zero"
+        with pytest.raises(ValueError, match=msg):
+            BlockPlacement(slc)
+
+    def test_slice_canonize_negative_stop(self):
+        # GH#37524 negative stop is OK with negative step and positive start
+        slc = slice(3, -1, -2)
+
+        bp = BlockPlacement(slc)
+        assert bp.indexer == slice(3, None, -2)
+
+    @pytest.mark.parametrize(
+        "slc",
+        [
+            slice(None, None),
+            slice(10, None),
+            slice(None, None, -1),
+            slice(None, 10, -1),
+            # These are "unbounded" because negative index will
+            #  change depending on container shape.
+            slice(-1, None),
+            slice(None, -1),
+            slice(-1, -1),
+            slice(-1, None, -1),
+            slice(None, -1, -1),
+            slice(-1, -1, -1),
+        ],
+    )
+    def test_unbounded_slice_raises(self, slc):
+        msg = "unbounded slice"
+        with pytest.raises(ValueError, match=msg):
+            BlockPlacement(slc)
+
+    @pytest.mark.parametrize(
+        "slc",
+        [
+            slice(0, 0),
+            slice(100, 0),
+            slice(100, 100),
+            slice(100, 100, -1),
+            slice(0, 100, -1),
+        ],
+    )
+    def test_not_slice_like_slices(self, slc):
+        assert not BlockPlacement(slc).is_slice_like
+
+    @pytest.mark.parametrize(
+        "arr, slc",
+        [
+            ([0], slice(0, 1, 1)),
+            ([100], slice(100, 101, 1)),
+            ([0, 1, 2], slice(0, 3, 1)),
+            ([0, 5, 10], slice(0, 15, 5)),
+            ([0, 100], slice(0, 200, 100)),
+            ([2, 1], slice(2, 0, -1)),
+        ],
+    )
+    def test_array_to_slice_conversion(self, arr, slc):
+        assert BlockPlacement(arr).as_slice == slc
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            [],
+            [-1],
+            [-1, -2, -3],
+            [-10],
+            [-1, 0, 1, 2],
+            [-2, 0, 2, 4],
+            [1, 0, -1],
+            [1, 1, 1],
+        ],
+    )
+    def test_not_slice_like_arrays(self, arr):
+        assert not BlockPlacement(arr).is_slice_like
+
+    @pytest.mark.parametrize(
+        "slc, expected",
+        [(slice(0, 3), [0, 1, 2]), (slice(0, 0), []), (slice(3, 0), [])],
+    )
+    def test_slice_iter(self, slc, expected):
+        assert list(BlockPlacement(slc)) == expected
+
+    @pytest.mark.parametrize(
+        "slc, arr",
+        [
+            (slice(0, 3), [0, 1, 2]),
+            (slice(0, 0), []),
+            (slice(3, 0), []),
+            (slice(3, 0, -1), [3, 2, 1]),
+        ],
+    )
+    def test_slice_to_array_conversion(self, slc, arr):
+        tm.assert_numpy_array_equal(
+            BlockPlacement(slc).as_array, np.asarray(arr, dtype=np.intp)
+        )
+
+    def test_blockplacement_add(self):
+        bpl = BlockPlacement(slice(0, 5))
+        assert bpl.add(1).as_slice == slice(1, 6, 1)
+        assert bpl.add(np.arange(5)).as_slice == slice(0, 10, 2)
+        assert list(bpl.add(np.arange(5, 0, -1))) == [5, 5, 5, 5, 5]
+
+    @pytest.mark.parametrize(
+        "val, inc, expected",
+        [
+            (slice(0, 0), 0, []),
+            (slice(1, 4), 0, [1, 2, 3]),
+            (slice(3, 0, -1), 0, [3, 2, 1]),
+            ([1, 2, 4], 0, [1, 2, 4]),
+            (slice(0, 0), 10, []),
+            (slice(1, 4), 10, [11, 12, 13]),
+            (slice(3, 0, -1), 10, [13, 12, 11]),
+            ([1, 2, 4], 10, [11, 12, 14]),
+            (slice(0, 0), -1, []),
+            (slice(1, 4), -1, [0, 1, 2]),
+            ([1, 2, 4], -1, [0, 1, 3]),
+        ],
+    )
+    def test_blockplacement_add_int(self, val, inc, expected):
+        assert list(BlockPlacement(val).add(inc)) == expected
+
+    @pytest.mark.parametrize("val", [slice(1, 4), [1, 2, 4]])
+    def test_blockplacement_add_int_raises(self, val):
+        msg = "iadd causes length change"
+        with pytest.raises(ValueError, match=msg):
+            BlockPlacement(val).add(-10)
+
+
+class TestCanHoldElement:
+    @pytest.fixture(
+        params=[
+            lambda x: x,
+            lambda x: x.to_series(),
+            lambda x: x._data,
+            lambda x: list(x),
+            lambda x: x.astype(object),
+            lambda x: np.asarray(x),
+            lambda x: x[0],
+            lambda x: x[:0],
+        ]
+    )
+    def element(self, request):
+        """
+        Functions that take an Index and return an element that should have
+        blk._can_hold_element(element) for a Block with this index's dtype.
+        """
+        return request.param
+
+    def test_datetime_block_can_hold_element(self):
+        block = create_block("datetime", [0])
+
+        assert block._can_hold_element([])
+
+        # We will check that block._can_hold_element iff arr.__setitem__ works
+        arr = pd.array(block.values.ravel())
+
+        # coerce None
+        assert block._can_hold_element(None)
+        arr[0] = None
+        assert arr[0] is pd.NaT
+
+        # coerce different types of datetime objects
+        vals = [np.datetime64("2010-10-10"), datetime(2010, 10, 10)]
+        for val in vals:
+            assert block._can_hold_element(val)
+            arr[0] = val
+
+        val = date(2010, 10, 10)
+        assert not block._can_hold_element(val)
+
+        msg = (
+            "value should be a 'Timestamp', 'NaT', "
+            "or array of those. Got 'date' instead."
+        )
+        with pytest.raises(TypeError, match=msg):
+            arr[0] = val
+
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_interval_can_hold_element_emptylist(self, dtype, element):
+        arr = np.array([1, 3, 4], dtype=dtype)
+        ii = IntervalIndex.from_breaks(arr)
+        blk = new_block(ii._data, BlockPlacement([1]), ndim=2)
+
+        assert blk._can_hold_element([])
+        # TODO: check this holds for all blocks
+
+    @pytest.mark.parametrize("dtype", [np.int64, np.uint64, np.float64])
+    def test_interval_can_hold_element(self, dtype, element):
+        arr = np.array([1, 3, 4, 9], dtype=dtype)
+        ii = IntervalIndex.from_breaks(arr)
+        blk = new_block(ii._data, BlockPlacement([1]), ndim=2)
+
+        elem = element(ii)
+        self.check_series_setitem(elem, ii, True)
+        assert blk._can_hold_element(elem)
+
+        # Careful: to get the expected Series-inplace behavior we need
+        # `elem` to not have the same length as `arr`
+        ii2 = IntervalIndex.from_breaks(arr[:-1], closed="neither")
+        elem = element(ii2)
+        with pytest.raises(TypeError, match="Invalid value"):
+            self.check_series_setitem(elem, ii, False)
+        assert not blk._can_hold_element(elem)
+
+        ii3 = IntervalIndex.from_breaks([Timestamp(1), Timestamp(3), Timestamp(4)])
+        elem = element(ii3)
+        with pytest.raises(TypeError, match="Invalid value"):
+            self.check_series_setitem(elem, ii, False)
+        assert not blk._can_hold_element(elem)
+
+        ii4 = IntervalIndex.from_breaks([Timedelta(1), Timedelta(3), Timedelta(4)])
+        elem = element(ii4)
+        with pytest.raises(TypeError, match="Invalid value"):
+            self.check_series_setitem(elem, ii, False)
+        assert not blk._can_hold_element(elem)
+
+    def test_period_can_hold_element_emptylist(self):
+        pi = period_range("2016", periods=3, freq="Y")
+        blk = new_block(pi._data.reshape(1, 3), BlockPlacement([1]), ndim=2)
+
+        assert blk._can_hold_element([])
+
+    def test_period_can_hold_element(self, element):
+        pi = period_range("2016", periods=3, freq="Y")
+
+        elem = element(pi)
+        self.check_series_setitem(elem, pi, True)
+
+        # Careful: to get the expected Series-inplace behavior we need
+        # `elem` to not have the same length as `arr`
+        pi2 = pi.asfreq("D")[:-1]
+        elem = element(pi2)
+        with pytest.raises(TypeError, match="Invalid value"):
+            self.check_series_setitem(elem, pi, False)
+
+        dti = pi.to_timestamp("s")[:-1]
+        elem = element(dti)
+        with pytest.raises(TypeError, match="Invalid value"):
+            self.check_series_setitem(elem, pi, False)
+
+    def check_can_hold_element(self, obj, elem, inplace: bool):
+        blk = obj._mgr.blocks[0]
+        if inplace:
+            assert blk._can_hold_element(elem)
+        else:
+            assert not blk._can_hold_element(elem)
+
+    def check_series_setitem(self, elem, index: Index, inplace: bool):
+        arr = index._data.copy()
+        ser = Series(arr, copy=False)
+
+        self.check_can_hold_element(ser, elem, inplace)
+
+        if is_scalar(elem):
+            ser[0] = elem
+        else:
+            ser[: len(elem)] = elem
+
+        if inplace:
+            assert ser.array is arr  # i.e. setting was done inplace
+        else:
+            assert ser.dtype == object
+
+
+class TestShouldStore:
+    def test_should_store_categorical(self):
+        cat = Categorical(["A", "B", "C"])
+        df = DataFrame(cat)
+        blk = df._mgr.blocks[0]
+
+        # matching dtype
+        assert blk.should_store(cat)
+        assert blk.should_store(cat[:-1])
+
+        # different dtype
+        assert not blk.should_store(cat.as_ordered())
+
+        # ndarray instead of Categorical
+        assert not blk.should_store(np.asarray(cat))
+
+
+def test_validate_ndim():
+    values = np.array([1.0, 2.0])
+    placement = BlockPlacement(slice(2))
+    msg = r"Wrong number of dimensions. values.ndim != ndim \[1 != 2\]"
+
+    depr_msg = "make_block is deprecated"
+    with pytest.raises(ValueError, match=msg):
+        with tm.assert_produces_warning(DeprecationWarning, match=depr_msg):
+            make_block(values, placement, ndim=2)
+
+
+def test_block_shape():
+    idx = Index([0, 1, 2, 3, 4])
+    a = Series([1, 2, 3]).reindex(idx)
+    b = Series(Categorical([1, 2, 3])).reindex(idx)
+
+    assert a._mgr.blocks[0].mgr_locs.indexer == b._mgr.blocks[0].mgr_locs.indexer
+
+
+def test_make_block_no_pandas_array(block_maker):
+    # https://github.com/pandas-dev/pandas/pull/24866
+    arr = pd.arrays.NumpyExtensionArray(np.array([1, 2]))
+
+    depr_msg = "make_block is deprecated"
+    warn = DeprecationWarning if block_maker is make_block else None
+
+    # NumpyExtensionArray, no dtype
+    with tm.assert_produces_warning(warn, match=depr_msg):
+        result = block_maker(arr, BlockPlacement(slice(len(arr))), ndim=arr.ndim)
+    assert result.dtype.kind in ["i", "u"]
+
+    if block_maker is make_block:
+        # new_block requires caller to unwrap NumpyExtensionArray
+        assert result.is_extension is False
+
+        # NumpyExtensionArray, NumpyEADtype
+        with tm.assert_produces_warning(warn, match=depr_msg):
+            result = block_maker(arr, slice(len(arr)), dtype=arr.dtype, ndim=arr.ndim)
+        assert result.dtype.kind in ["i", "u"]
+        assert result.is_extension is False
+
+        # new_block no longer accepts dtype keyword
+        # ndarray, NumpyEADtype
+        with tm.assert_produces_warning(warn, match=depr_msg):
+            result = block_maker(
+                arr.to_numpy(), slice(len(arr)), dtype=arr.dtype, ndim=arr.ndim
+            )
+        assert result.dtype.kind in ["i", "u"]
+        assert result.is_extension is False
diff --git a/pandas/pandas/tests/io/__init__.py b/pandas/pandas/tests/io/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/conftest.py b/pandas/pandas/tests/io/conftest.py
new file mode 100644
index 0000000000000..a5ddda9d66e7a
--- /dev/null
+++ b/pandas/pandas/tests/io/conftest.py
@@ -0,0 +1,225 @@
+import shlex
+import subprocess
+import time
+import uuid
+
+import pytest
+
+from pandas.compat import (
+    is_ci_environment,
+    is_platform_arm,
+    is_platform_mac,
+    is_platform_windows,
+)
+import pandas.util._test_decorators as td
+
+import pandas.io.common as icom
+from pandas.io.parsers import read_csv
+
+
+@pytest.fixture
+def compression_to_extension():
+    return {value: key for key, value in icom.extension_to_compression.items()}
+
+
+@pytest.fixture
+def tips_file(datapath):
+    """Path to the tips dataset"""
+    return datapath("io", "data", "csv", "tips.csv")
+
+
+@pytest.fixture
+def jsonl_file(datapath):
+    """Path to a JSONL dataset"""
+    return datapath("io", "parser", "data", "items.jsonl")
+
+
+@pytest.fixture
+def salaries_table(datapath):
+    """DataFrame with the salaries dataset"""
+    return read_csv(datapath("io", "parser", "data", "salaries.csv"), sep="\t")
+
+
+@pytest.fixture
+def feather_file(datapath):
+    return datapath("io", "data", "feather", "feather-0_3_1.feather")
+
+
+@pytest.fixture
+def xml_file(datapath):
+    return datapath("io", "data", "xml", "books.xml")
+
+
+@pytest.fixture
+def s3_base(worker_id, monkeypatch):
+    """
+    Fixture for mocking S3 interaction.
+
+    Sets up moto server in separate process locally
+    Return url for motoserver/moto CI service
+    """
+    pytest.importorskip("s3fs")
+    pytest.importorskip("boto3")
+
+    # temporary workaround as moto fails for botocore >= 1.11 otherwise,
+    # see https://github.com/spulec/moto/issues/1924 & 1952
+    monkeypatch.setenv("AWS_ACCESS_KEY_ID", "foobar_key")
+    monkeypatch.setenv("AWS_SECRET_ACCESS_KEY", "foobar_secret")
+    if is_ci_environment():
+        if is_platform_arm() or is_platform_mac() or is_platform_windows():
+            # NOT RUN on Windows/macOS, only Ubuntu
+            # - subprocess in CI can cause timeouts
+            # - GitHub Actions do not support
+            #   container services for the above OSs
+            pytest.skip(
+                "S3 tests do not have a corresponding service on "
+                "Windows or macOS platforms"
+            )
+        else:
+            # set in .github/workflows/unit-tests.yml
+            yield "http://localhost:5000"
+    else:
+        requests = pytest.importorskip("requests")
+        pytest.importorskip("moto")
+        pytest.importorskip("flask")  # server mode needs flask too
+
+        # Launching moto in server mode, i.e., as a separate process
+        # with an S3 endpoint on localhost
+
+        worker_id = "5" if worker_id == "master" else worker_id.lstrip("gw")
+        endpoint_port = f"555{worker_id}"
+        endpoint_uri = f"http://127.0.0.1:{endpoint_port}/"
+
+        # pipe to null to avoid logging in terminal
+        with subprocess.Popen(
+            shlex.split(f"moto_server s3 -p {endpoint_port}"),
+            stdout=subprocess.DEVNULL,
+            stderr=subprocess.DEVNULL,
+        ) as proc:
+            timeout = 5
+            while timeout > 0:
+                try:
+                    # OK to go once server is accepting connections
+                    r = requests.get(endpoint_uri)
+                    if r.ok:
+                        break
+                except Exception:
+                    pass
+                timeout -= 0.1
+                time.sleep(0.1)
+            yield endpoint_uri
+
+            proc.terminate()
+
+
+@pytest.fixture
+def s3so(s3_base):
+    return {"client_kwargs": {"endpoint_url": s3_base}}
+
+
+@pytest.fixture
+def s3_resource(s3_base):
+    import boto3
+
+    s3 = boto3.resource("s3", endpoint_url=s3_base)
+    return s3
+
+
+@pytest.fixture
+def s3_public_bucket(s3_resource):
+    bucket = s3_resource.Bucket(f"pandas-test-{uuid.uuid4()}")
+    bucket.create()
+    yield bucket
+    bucket.objects.delete()
+    bucket.delete()
+
+
+@pytest.fixture
+def s3_public_bucket_with_data(
+    s3_public_bucket, tips_file, jsonl_file, feather_file, xml_file
+):
+    """
+    The following datasets
+    are loaded.
+
+    - tips.csv
+    - tips.csv.gz
+    - tips.csv.bz2
+    - items.jsonl
+    """
+    test_s3_files = [
+        ("tips#1.csv", tips_file),
+        ("tips.csv", tips_file),
+        ("tips.csv.gz", tips_file + ".gz"),
+        ("tips.csv.bz2", tips_file + ".bz2"),
+        ("items.jsonl", jsonl_file),
+        ("simple_dataset.feather", feather_file),
+        ("books.xml", xml_file),
+    ]
+    for s3_key, file_name in test_s3_files:
+        with open(file_name, "rb") as f:
+            s3_public_bucket.put_object(Key=s3_key, Body=f)
+    return s3_public_bucket
+
+
+@pytest.fixture
+def s3_private_bucket(s3_resource):
+    bucket = s3_resource.Bucket(f"cant_get_it-{uuid.uuid4()}")
+    bucket.create(ACL="private")
+    yield bucket
+    bucket.objects.delete()
+    bucket.delete()
+
+
+@pytest.fixture
+def s3_private_bucket_with_data(
+    s3_private_bucket, tips_file, jsonl_file, feather_file, xml_file
+):
+    """
+    The following datasets
+    are loaded.
+
+    - tips.csv
+    - tips.csv.gz
+    - tips.csv.bz2
+    - items.jsonl
+    """
+    test_s3_files = [
+        ("tips#1.csv", tips_file),
+        ("tips.csv", tips_file),
+        ("tips.csv.gz", tips_file + ".gz"),
+        ("tips.csv.bz2", tips_file + ".bz2"),
+        ("items.jsonl", jsonl_file),
+        ("simple_dataset.feather", feather_file),
+        ("books.xml", xml_file),
+    ]
+    for s3_key, file_name in test_s3_files:
+        with open(file_name, "rb") as f:
+            s3_private_bucket.put_object(Key=s3_key, Body=f)
+    return s3_private_bucket
+
+
+_compression_formats_params = [
+    (".no_compress", None),
+    ("", None),
+    (".gz", "gzip"),
+    (".GZ", "gzip"),
+    (".bz2", "bz2"),
+    (".BZ2", "bz2"),
+    (".zip", "zip"),
+    (".ZIP", "zip"),
+    (".xz", "xz"),
+    (".XZ", "xz"),
+    pytest.param((".zst", "zstd"), marks=td.skip_if_no("zstandard")),
+    pytest.param((".ZST", "zstd"), marks=td.skip_if_no("zstandard")),
+]
+
+
+@pytest.fixture(params=_compression_formats_params[1:])
+def compression_format(request):
+    return request.param
+
+
+@pytest.fixture(params=_compression_formats_params)
+def compression_ext(request):
+    return request.param[0]
diff --git a/pandas/pandas/tests/io/data/csv/banklist.csv b/pandas/pandas/tests/io/data/csv/banklist.csv
new file mode 100644
index 0000000000000..e7900830140d2
--- /dev/null
+++ b/pandas/pandas/tests/io/data/csv/banklist.csv
@@ -0,0 +1,507 @@
+Bank Name,City,ST,CERT,Acquiring Institution,Closing Date,Updated Date
+Banks of Wisconsin d/b/a Bank of Kenosha,Kenosha,WI,35386,"North Shore Bank, FSB",31-May-13,31-May-13
+Central Arizona Bank,Scottsdale,AZ,34527,Western State Bank,14-May-13,20-May-13
+Sunrise Bank,Valdosta,GA,58185,Synovus Bank,10-May-13,21-May-13
+Pisgah Community Bank,Asheville,NC,58701,"Capital Bank, N.A.",10-May-13,14-May-13
+Douglas County Bank,Douglasville,GA,21649,Hamilton State Bank,26-Apr-13,16-May-13
+Parkway Bank,Lenoir,NC,57158,"CertusBank, National Association",26-Apr-13,17-May-13
+Chipola Community Bank,Marianna,FL,58034,First Federal Bank of Florida,19-Apr-13,16-May-13
+Heritage Bank of North Florida,Orange Park,FL,26680,FirstAtlantic Bank,19-Apr-13,16-May-13
+First Federal Bank,Lexington,KY,29594,Your Community Bank,19-Apr-13,23-Apr-13
+Gold Canyon Bank,Gold Canyon,AZ,58066,"First Scottsdale Bank, National Association",5-Apr-13,9-Apr-13
+Frontier Bank,LaGrange,GA,16431,HeritageBank of the South,8-Mar-13,26-Mar-13
+Covenant Bank,Chicago,IL,22476,Liberty Bank and Trust Company,15-Feb-13,4-Mar-13
+1st Regents Bank,Andover,MN,57157,First Minnesota Bank,18-Jan-13,28-Feb-13
+Westside Community Bank,University Place,WA,33997,Sunwest Bank,11-Jan-13,24-Jan-13
+Community Bank of the Ozarks,Sunrise Beach,MO,27331,Bank of Sullivan,14-Dec-12,24-Jan-13
+Hometown Community Bank,Braselton,GA,57928,"CertusBank, National Association",16-Nov-12,24-Jan-13
+Citizens First National Bank,Princeton,IL,3731,Heartland Bank and Trust Company,2-Nov-12,24-Jan-13
+Heritage Bank of Florida,Lutz,FL,35009,Centennial Bank,2-Nov-12,24-Jan-13
+NOVA Bank,Berwyn,PA,27148,No Acquirer,26-Oct-12,24-Jan-13
+Excel Bank,Sedalia,MO,19189,Simmons First National Bank,19-Oct-12,24-Jan-13
+First East Side Savings Bank,Tamarac,FL,28144,Stearns Bank N.A.,19-Oct-12,24-Jan-13
+GulfSouth Private Bank,Destin,FL,58073,SmartBank,19-Oct-12,24-Jan-13
+First United Bank,Crete,IL,20685,"Old Plank Trail Community Bank, National Association",28-Sep-12,15-Nov-12
+Truman Bank,St. Louis,MO,27316,Simmons First National Bank,14-Sep-12,17-Dec-12
+First Commercial Bank,Bloomington,MN,35246,Republic Bank & Trust Company,7-Sep-12,17-Dec-12
+Waukegan Savings Bank,Waukegan,IL,28243,First Midwest Bank,3-Aug-12,11-Oct-12
+Jasper Banking Company,Jasper,GA,16240,Stearns Bank N.A.,27-Jul-12,17-Dec-12
+Second Federal Savings and Loan Association of Chicago,Chicago,IL,27986,Hinsdale Bank & Trust Company,20-Jul-12,14-Jan-13
+Heartland Bank,Leawood,KS,1361,Metcalf Bank,20-Jul-12,17-Dec-12
+First Cherokee State Bank,Woodstock,GA,32711,Community & Southern Bank,20-Jul-12,31-Oct-12
+Georgia Trust Bank,Buford,GA,57847,Community & Southern Bank,20-Jul-12,17-Dec-12
+The Royal Palm Bank of Florida,Naples,FL,57096,First National Bank of the Gulf Coast,20-Jul-12,7-Jan-13
+Glasgow Savings Bank,Glasgow,MO,1056,Regional Missouri Bank,13-Jul-12,11-Oct-12
+Montgomery Bank & Trust,Ailey,GA,19498,Ameris Bank,6-Jul-12,31-Oct-12
+The Farmers Bank of Lynchburg,Lynchburg,TN,1690,Clayton Bank and Trust,15-Jun-12,31-Oct-12
+Security Exchange Bank,Marietta,GA,35299,Fidelity Bank,15-Jun-12,10-Oct-12
+Putnam State Bank,Palatka,FL,27405,Harbor Community Bank,15-Jun-12,10-Oct-12
+Waccamaw Bank,Whiteville,NC,34515,First Community Bank,8-Jun-12,8-Nov-12
+Farmers' and Traders' State Bank,Shabbona,IL,9257,First State Bank,8-Jun-12,10-Oct-12
+Carolina Federal Savings Bank,Charleston,SC,35372,Bank of North Carolina,8-Jun-12,31-Oct-12
+First Capital Bank,Kingfisher,OK,416,F & M Bank,8-Jun-12,10-Oct-12
+"Alabama Trust Bank, National Association",Sylacauga,AL,35224,Southern States Bank,18-May-12,20-May-13
+"Security Bank, National Association",North Lauderdale,FL,23156,Banesco USA,4-May-12,31-Oct-12
+Palm Desert National Bank,Palm Desert,CA,23632,Pacific Premier Bank,27-Apr-12,17-May-13
+Plantation Federal Bank,Pawleys Island,SC,32503,First Federal Bank,27-Apr-12,17-May-13
+"Inter Savings Bank, fsb D/B/A InterBank, fsb",Maple Grove,MN,31495,Great Southern Bank,27-Apr-12,17-May-13
+HarVest Bank of Maryland,Gaithersburg,MD,57766,Sonabank,27-Apr-12,17-May-13
+Bank of the Eastern Shore,Cambridge,MD,26759,No Acquirer,27-Apr-12,17-Oct-12
+"Fort Lee Federal Savings Bank, FSB",Fort Lee,NJ,35527,Alma Bank,20-Apr-12,17-May-13
+Fidelity Bank,Dearborn,MI,33883,The Huntington National Bank,30-Mar-12,16-May-13
+Premier Bank,Wilmette,IL,35419,International Bank of Chicago,23-Mar-12,17-Oct-12
+Covenant Bank & Trust,Rock Spring,GA,58068,"Stearns Bank, N.A.",23-Mar-12,31-Oct-12
+New City Bank,Chicago,IL,57597,No Acquirer,9-Mar-12,29-Oct-12
+Global Commerce Bank,Doraville,GA,34046,Metro City Bank,2-Mar-12,31-Oct-12
+Home Savings of America,Little Falls,MN,29178,No Acquirer,24-Feb-12,17-Dec-12
+Central Bank of Georgia,Ellaville,GA,5687,Ameris Bank,24-Feb-12,9-Aug-12
+SCB Bank,Shelbyville,IN,29761,"First Merchants Bank, National Association",10-Feb-12,25-Mar-13
+Charter National Bank and Trust,Hoffman Estates,IL,23187,"Barrington Bank & Trust Company, National Association",10-Feb-12,25-Mar-13
+BankEast,Knoxville,TN,19869,U.S.Bank National Association,27-Jan-12,8-Mar-13
+Patriot Bank Minnesota,Forest Lake,MN,34823,First Resource Bank,27-Jan-12,12-Sep-12
+Tennessee Commerce Bank,Franklin,TN,35296,Republic Bank & Trust Company,27-Jan-12,20-Nov-12
+First Guaranty Bank and Trust Company of Jacksonville,Jacksonville,FL,16579,"CenterState Bank of Florida, N.A.",27-Jan-12,12-Sep-12
+American Eagle Savings Bank,Boothwyn,PA,31581,"Capital Bank, N.A.",20-Jan-12,25-Jan-13
+The First State Bank,Stockbridge,GA,19252,Hamilton State Bank,20-Jan-12,25-Jan-13
+Central Florida State Bank,Belleview,FL,57186,"CenterState Bank of Florida, N.A.",20-Jan-12,25-Jan-13
+Western National Bank,Phoenix,AZ,57917,Washington Federal,16-Dec-11,13-Aug-12
+Premier Community Bank of the Emerald Coast,Crestview,FL,58343,Summit Bank,16-Dec-11,12-Sep-12
+Central Progressive Bank,Lacombe,LA,19657,First NBC Bank,18-Nov-11,13-Aug-12
+Polk County Bank,Johnston,IA,14194,Grinnell State Bank,18-Nov-11,15-Aug-12
+Community Bank of Rockmart,Rockmart,GA,57860,Century Bank of Georgia,10-Nov-11,13-Aug-12
+SunFirst Bank,Saint George,UT,57087,Cache Valley Bank,4-Nov-11,16-Nov-12
+"Mid City Bank, Inc.",Omaha,NE,19397,Premier Bank,4-Nov-11,15-Aug-12
+All American Bank,Des Plaines,IL,57759,International Bank of Chicago,28-Oct-11,15-Aug-12
+Community Banks of Colorado,Greenwood Village,CO,21132,"Bank Midwest, N.A.",21-Oct-11,2-Jan-13
+Community Capital Bank,Jonesboro,GA,57036,State Bank and Trust Company,21-Oct-11,8-Nov-12
+Decatur First Bank,Decatur,GA,34392,Fidelity Bank,21-Oct-11,8-Nov-12
+Old Harbor Bank,Clearwater,FL,57537,1st United Bank,21-Oct-11,8-Nov-12
+Country Bank,Aledo,IL,35395,Blackhawk Bank & Trust,14-Oct-11,15-Aug-12
+First State Bank,Cranford,NJ,58046,Northfield Bank,14-Oct-11,8-Nov-12
+"Blue Ridge Savings Bank, Inc.",Asheville,NC,32347,Bank of North Carolina,14-Oct-11,8-Nov-12
+Piedmont Community Bank,Gray,GA,57256,State Bank and Trust Company,14-Oct-11,22-Jan-13
+Sun Security Bank,Ellington,MO,20115,Great Southern Bank,7-Oct-11,7-Nov-12
+The RiverBank,Wyoming,MN,10216,Central Bank,7-Oct-11,7-Nov-12
+First International Bank,Plano,TX,33513,American First National Bank,30-Sep-11,9-Oct-12
+Citizens Bank of Northern California,Nevada City,CA,33983,Tri Counties Bank,23-Sep-11,9-Oct-12
+Bank of the Commonwealth,Norfolk,VA,20408,Southern Bank and Trust Company,23-Sep-11,9-Oct-12
+The First National Bank of Florida,Milton,FL,25155,CharterBank,9-Sep-11,6-Sep-12
+CreekSide Bank,Woodstock,GA,58226,Georgia Commerce Bank,2-Sep-11,6-Sep-12
+Patriot Bank of Georgia,Cumming,GA,58273,Georgia Commerce Bank,2-Sep-11,2-Nov-12
+First Choice Bank,Geneva,IL,57212,Inland Bank & Trust,19-Aug-11,15-Aug-12
+First Southern National Bank,Statesboro,GA,57239,Heritage Bank of the South,19-Aug-11,2-Nov-12
+Lydian Private Bank,Palm Beach,FL,35356,"Sabadell United Bank, N.A.",19-Aug-11,2-Nov-12
+Public Savings Bank,Huntingdon Valley,PA,34130,"Capital Bank, N.A.",18-Aug-11,15-Aug-12
+The First National Bank of Olathe,Olathe,KS,4744,Enterprise Bank & Trust,12-Aug-11,23-Aug-12
+Bank of Whitman,Colfax,WA,22528,Columbia State Bank,5-Aug-11,16-Aug-12
+Bank of Shorewood,Shorewood,IL,22637,Heartland Bank and Trust Company,5-Aug-11,16-Aug-12
+Integra Bank National Association,Evansville,IN,4392,Old National Bank,29-Jul-11,16-Aug-12
+"BankMeridian, N.A.",Columbia,SC,58222,SCBT National Association,29-Jul-11,2-Nov-12
+Virginia Business Bank,Richmond,VA,58283,Xenith Bank,29-Jul-11,9-Oct-12
+Bank of Choice,Greeley,CO,2994,"Bank Midwest, N.A.",22-Jul-11,12-Sep-12
+LandMark Bank of Florida,Sarasota,FL,35244,American Momentum Bank,22-Jul-11,2-Nov-12
+Southshore Community Bank,Apollo Beach,FL,58056,American Momentum Bank,22-Jul-11,2-Nov-12
+Summit Bank,Prescott,AZ,57442,The Foothills Bank,15-Jul-11,16-Aug-12
+First Peoples Bank,Port St. Lucie,FL,34870,"Premier American Bank, N.A.",15-Jul-11,2-Nov-12
+High Trust Bank,Stockbridge,GA,19554,Ameris Bank,15-Jul-11,2-Nov-12
+One Georgia Bank,Atlanta,GA,58238,Ameris Bank,15-Jul-11,2-Nov-12
+Signature Bank,Windsor,CO,57835,Points West Community Bank,8-Jul-11,26-Oct-12
+Colorado Capital Bank,Castle Rock,CO,34522,First-Citizens Bank & Trust Company,8-Jul-11,15-Jan-13
+First Chicago Bank & Trust,Chicago,IL,27935,Northbrook Bank & Trust Company,8-Jul-11,9-Sep-12
+Mountain Heritage Bank,Clayton,GA,57593,First American Bank and Trust Company,24-Jun-11,2-Nov-12
+First Commercial Bank of Tampa Bay,Tampa,FL,27583,Stonegate Bank,17-Jun-11,2-Nov-12
+McIntosh State Bank,Jackson,GA,19237,Hamilton State Bank,17-Jun-11,2-Nov-12
+Atlantic Bank and Trust,Charleston,SC,58420,"First Citizens Bank and Trust Company, Inc.",3-Jun-11,31-Oct-12
+First Heritage Bank,Snohomish,WA,23626,Columbia State Bank,27-May-11,28-Jan-13
+Summit Bank,Burlington,WA,513,Columbia State Bank,20-May-11,22-Jan-13
+First Georgia Banking Company,Franklin,GA,57647,"CertusBank, National Association",20-May-11,13-Nov-12
+Atlantic Southern Bank,Macon,GA,57213,"CertusBank, National Association",20-May-11,31-Oct-12
+Coastal Bank,Cocoa Beach,FL,34898,"Florida Community Bank, a division of Premier American Bank, N.A.",6-May-11,30-Nov-12
+Community Central Bank,Mount Clemens,MI,34234,Talmer Bank & Trust,29-Apr-11,16-Aug-12
+The Park Avenue Bank,Valdosta,GA,19797,Bank of the Ozarks,29-Apr-11,30-Nov-12
+First Choice Community Bank,Dallas,GA,58539,Bank of the Ozarks,29-Apr-11,22-Jan-13
+Cortez Community Bank,Brooksville,FL,57625,"Florida Community Bank, a division of Premier American Bank, N.A.",29-Apr-11,30-Nov-12
+First National Bank of Central Florida,Winter Park,FL,26297,"Florida Community Bank, a division of Premier American Bank, N.A.",29-Apr-11,30-Nov-12
+Heritage Banking Group,Carthage,MS,14273,Trustmark National Bank,15-Apr-11,30-Nov-12
+Rosemount National Bank,Rosemount,MN,24099,Central Bank,15-Apr-11,16-Aug-12
+Superior Bank,Birmingham,AL,17750,"Superior Bank, National Association",15-Apr-11,30-Nov-12
+Nexity Bank,Birmingham,AL,19794,AloStar Bank of Commerce,15-Apr-11,4-Sep-12
+New Horizons Bank,East Ellijay,GA,57705,Citizens South Bank,15-Apr-11,16-Aug-12
+Bartow County Bank,Cartersville,GA,21495,Hamilton State Bank,15-Apr-11,22-Jan-13
+Nevada Commerce Bank,Las Vegas,NV,35418,City National Bank,8-Apr-11,9-Sep-12
+Western Springs National Bank and Trust,Western Springs,IL,10086,Heartland Bank and Trust Company,8-Apr-11,22-Jan-13
+The Bank of Commerce,Wood Dale,IL,34292,Advantage National Bank Group,25-Mar-11,22-Jan-13
+Legacy Bank,Milwaukee,WI,34818,Seaway Bank and Trust Company,11-Mar-11,12-Sep-12
+First National Bank of Davis,Davis,OK,4077,The Pauls Valley National Bank,11-Mar-11,20-Aug-12
+Valley Community Bank,St. Charles,IL,34187,First State Bank,25-Feb-11,12-Sep-12
+"San Luis Trust Bank, FSB",San Luis Obispo,CA,34783,First California Bank,18-Feb-11,20-Aug-12
+Charter Oak Bank,Napa,CA,57855,Bank of Marin,18-Feb-11,12-Sep-12
+Citizens Bank of Effingham,Springfield,GA,34601,Heritage Bank of the South,18-Feb-11,2-Nov-12
+Habersham Bank,Clarkesville,GA,151,SCBT National Association,18-Feb-11,2-Nov-12
+Canyon National Bank,Palm Springs,CA,34692,Pacific Premier Bank,11-Feb-11,12-Sep-12
+Badger State Bank,Cassville,WI,13272,Royal Bank,11-Feb-11,12-Sep-12
+Peoples State Bank,Hamtramck,MI,14939,First Michigan Bank,11-Feb-11,22-Jan-13
+Sunshine State Community Bank,Port Orange,FL,35478,"Premier American Bank, N.A.",11-Feb-11,2-Nov-12
+Community First Bank Chicago,Chicago,IL,57948,Northbrook Bank & Trust Company,4-Feb-11,20-Aug-12
+North Georgia Bank,Watkinsville,GA,35242,BankSouth,4-Feb-11,2-Nov-12
+American Trust Bank,Roswell,GA,57432,Renasant Bank,4-Feb-11,31-Oct-12
+First Community Bank,Taos,NM,12261,"U.S. Bank, N.A.",28-Jan-11,12-Sep-12
+FirsTier Bank,Louisville,CO,57646,No Acquirer,28-Jan-11,12-Sep-12
+Evergreen State Bank,Stoughton,WI,5328,McFarland State Bank,28-Jan-11,12-Sep-12
+The First State Bank,Camargo,OK,2303,Bank 7,28-Jan-11,12-Sep-12
+United Western Bank,Denver,CO,31293,First-Citizens Bank & Trust Company,21-Jan-11,12-Sep-12
+The Bank of Asheville,Asheville,NC,34516,First Bank,21-Jan-11,2-Nov-12
+CommunitySouth Bank & Trust,Easley,SC,57868,"CertusBank, National Association",21-Jan-11,2-Nov-12
+Enterprise Banking Company,McDonough,GA,19758,No Acquirer,21-Jan-11,2-Nov-12
+Oglethorpe Bank,Brunswick,GA,57440,Bank of the Ozarks,14-Jan-11,2-Nov-12
+Legacy Bank,Scottsdale,AZ,57820,Enterprise Bank & Trust,7-Jan-11,12-Sep-12
+First Commercial Bank of Florida,Orlando,FL,34965,First Southern Bank,7-Jan-11,2-Nov-12
+Community National Bank,Lino Lakes,MN,23306,Farmers & Merchants Savings Bank,17-Dec-10,20-Aug-12
+First Southern Bank,Batesville,AR,58052,Southern Bank,17-Dec-10,20-Aug-12
+"United Americas Bank, N.A.",Atlanta,GA,35065,State Bank and Trust Company,17-Dec-10,2-Nov-12
+"Appalachian Community Bank, FSB",McCaysville,GA,58495,Peoples Bank of East Tennessee,17-Dec-10,31-Oct-12
+Chestatee State Bank,Dawsonville,GA,34578,Bank of the Ozarks,17-Dec-10,2-Nov-12
+"The Bank of Miami,N.A.",Coral Gables,FL,19040,1st United Bank,17-Dec-10,2-Nov-12
+Earthstar Bank,Southampton,PA,35561,Polonia Bank,10-Dec-10,20-Aug-12
+Paramount Bank,Farmington Hills,MI,34673,Level One Bank,10-Dec-10,20-Aug-12
+First Banking Center,Burlington,WI,5287,First Michigan Bank,19-Nov-10,20-Aug-12
+Allegiance Bank of North America,Bala Cynwyd,PA,35078,VIST Bank,19-Nov-10,20-Aug-12
+Gulf State Community Bank,Carrabelle,FL,20340,Centennial Bank,19-Nov-10,2-Nov-12
+Copper Star Bank,Scottsdale,AZ,35463,"Stearns Bank, N.A.",12-Nov-10,20-Aug-12
+Darby Bank & Trust Co.,Vidalia,GA,14580,Ameris Bank,12-Nov-10,15-Jan-13
+Tifton Banking Company,Tifton,GA,57831,Ameris Bank,12-Nov-10,2-Nov-12
+First Vietnamese American Bank,Westminster,CA,57885,Grandpoint Bank,5-Nov-10,12-Sep-12
+Pierce Commercial Bank,Tacoma,WA,34411,Heritage Bank,5-Nov-10,20-Aug-12
+Western Commercial Bank,Woodland Hills,CA,58087,First California Bank,5-Nov-10,12-Sep-12
+K Bank,Randallstown,MD,31263,Manufacturers and Traders Trust Company (M&T Bank),5-Nov-10,20-Aug-12
+"First Arizona Savings, A FSB",Scottsdale,AZ,32582,No Acquirer,22-Oct-10,20-Aug-12
+Hillcrest Bank,Overland Park,KS,22173,"Hillcrest Bank, N.A.",22-Oct-10,20-Aug-12
+First Suburban National Bank,Maywood,IL,16089,Seaway Bank and Trust Company,22-Oct-10,20-Aug-12
+The First National Bank of Barnesville,Barnesville,GA,2119,United Bank,22-Oct-10,2-Nov-12
+The Gordon Bank,Gordon,GA,33904,Morris Bank,22-Oct-10,2-Nov-12
+Progress Bank of Florida,Tampa,FL,32251,Bay Cities Bank,22-Oct-10,2-Nov-12
+First Bank of Jacksonville,Jacksonville,FL,27573,Ameris Bank,22-Oct-10,2-Nov-12
+Premier Bank,Jefferson City,MO,34016,Providence Bank,15-Oct-10,20-Aug-12
+WestBridge Bank and Trust Company,Chesterfield,MO,58205,Midland States Bank,15-Oct-10,20-Aug-12
+"Security Savings Bank, F.S.B.",Olathe,KS,30898,Simmons First National Bank,15-Oct-10,20-Aug-12
+Shoreline Bank,Shoreline,WA,35250,GBC International Bank,1-Oct-10,20-Aug-12
+Wakulla Bank,Crawfordville,FL,21777,Centennial Bank,1-Oct-10,2-Nov-12
+North County Bank,Arlington,WA,35053,Whidbey Island Bank,24-Sep-10,20-Aug-12
+Haven Trust Bank Florida,Ponte Vedra Beach,FL,58308,First Southern Bank,24-Sep-10,5-Nov-12
+Maritime Savings Bank,West Allis,WI,28612,"North Shore Bank, FSB",17-Sep-10,20-Aug-12
+Bramble Savings Bank,Milford,OH,27808,Foundation Bank,17-Sep-10,20-Aug-12
+The Peoples Bank,Winder,GA,182,Community & Southern Bank,17-Sep-10,5-Nov-12
+First Commerce Community Bank,Douglasville,GA,57448,Community & Southern Bank,17-Sep-10,15-Jan-13
+Bank of Ellijay,Ellijay,GA,58197,Community & Southern Bank,17-Sep-10,15-Jan-13
+ISN Bank,Cherry Hill,NJ,57107,Customers Bank,17-Sep-10,22-Aug-12
+Horizon Bank,Bradenton,FL,35061,Bank of the Ozarks,10-Sep-10,5-Nov-12
+Sonoma Valley Bank,Sonoma,CA,27259,Westamerica Bank,20-Aug-10,12-Sep-12
+Los Padres Bank,Solvang,CA,32165,Pacific Western Bank,20-Aug-10,12-Sep-12
+Butte Community Bank,Chico,CA,33219,"Rabobank, N.A.",20-Aug-10,12-Sep-12
+Pacific State Bank,Stockton,CA,27090,"Rabobank, N.A.",20-Aug-10,12-Sep-12
+ShoreBank,Chicago,IL,15640,Urban Partnership Bank,20-Aug-10,16-May-13
+Imperial Savings and Loan Association,Martinsville,VA,31623,"River Community Bank, N.A.",20-Aug-10,24-Aug-12
+Independent National Bank,Ocala,FL,27344,"CenterState Bank of Florida, N.A.",20-Aug-10,5-Nov-12
+Community National Bank at Bartow,Bartow,FL,25266,"CenterState Bank of Florida, N.A.",20-Aug-10,5-Nov-12
+Palos Bank and Trust Company,Palos Heights,IL,17599,First Midwest Bank,13-Aug-10,22-Aug-12
+Ravenswood Bank,Chicago,IL,34231,Northbrook Bank & Trust Company,6-Aug-10,22-Aug-12
+LibertyBank,Eugene,OR,31964,Home Federal Bank,30-Jul-10,22-Aug-12
+The Cowlitz Bank,Longview,WA,22643,Heritage Bank,30-Jul-10,22-Aug-12
+Coastal Community Bank,Panama City Beach,FL,9619,Centennial Bank,30-Jul-10,5-Nov-12
+Bayside Savings Bank,Port Saint Joe,FL,57669,Centennial Bank,30-Jul-10,5-Nov-12
+Northwest Bank & Trust,Acworth,GA,57658,State Bank and Trust Company,30-Jul-10,5-Nov-12
+Home Valley Bank,Cave Junction,OR,23181,South Valley Bank & Trust,23-Jul-10,12-Sep-12
+SouthwestUSA Bank,Las Vegas,NV,35434,Plaza Bank,23-Jul-10,22-Aug-12
+Community Security Bank,New Prague,MN,34486,Roundbank,23-Jul-10,12-Sep-12
+Thunder Bank,Sylvan Grove,KS,10506,The Bennington State Bank,23-Jul-10,13-Sep-12
+Williamsburg First National Bank,Kingstree,SC,17837,"First Citizens Bank and Trust Company, Inc.",23-Jul-10,5-Nov-12
+Crescent Bank and Trust Company,Jasper,GA,27559,Renasant Bank,23-Jul-10,5-Nov-12
+Sterling Bank,Lantana,FL,32536,IBERIABANK,23-Jul-10,5-Nov-12
+"Mainstreet Savings Bank, FSB",Hastings,MI,28136,Commercial Bank,16-Jul-10,13-Sep-12
+Olde Cypress Community Bank,Clewiston,FL,28864,"CenterState Bank of Florida, N.A.",16-Jul-10,5-Nov-12
+Turnberry Bank,Aventura,FL,32280,NAFH National Bank,16-Jul-10,5-Nov-12
+Metro Bank of Dade County,Miami,FL,25172,NAFH National Bank,16-Jul-10,5-Nov-12
+First National Bank of the South,Spartanburg,SC,35383,NAFH National Bank,16-Jul-10,5-Nov-12
+Woodlands Bank,Bluffton,SC,32571,Bank of the Ozarks,16-Jul-10,5-Nov-12
+Home National Bank,Blackwell,OK,11636,RCB Bank,9-Jul-10,10-Dec-12
+USA Bank,Port Chester,NY,58072,New Century Bank,9-Jul-10,14-Sep-12
+Ideal Federal Savings Bank,Baltimore,MD,32456,No Acquirer,9-Jul-10,14-Sep-12
+Bay National Bank,Baltimore,MD,35462,"Bay Bank, FSB",9-Jul-10,15-Jan-13
+High Desert State Bank,Albuquerque,NM,35279,First American Bank,25-Jun-10,14-Sep-12
+First National Bank,Savannah,GA,34152,"The Savannah Bank, N.A.",25-Jun-10,5-Nov-12
+Peninsula Bank,Englewood,FL,26563,"Premier American Bank, N.A.",25-Jun-10,5-Nov-12
+Nevada Security Bank,Reno,NV,57110,Umpqua Bank,18-Jun-10,23-Aug-12
+Washington First International Bank,Seattle,WA,32955,East West Bank,11-Jun-10,14-Sep-12
+TierOne Bank,Lincoln,NE,29341,Great Western Bank,4-Jun-10,14-Sep-12
+Arcola Homestead Savings Bank,Arcola,IL,31813,No Acquirer,4-Jun-10,14-Sep-12
+First National Bank,Rosedale,MS,15814,The Jefferson Bank,4-Jun-10,5-Nov-12
+Sun West Bank,Las Vegas,NV,34785,City National Bank,28-May-10,14-Sep-12
+"Granite Community Bank, NA",Granite Bay,CA,57315,Tri Counties Bank,28-May-10,14-Sep-12
+Bank of Florida - Tampa,Tampa,FL,57814,EverBank,28-May-10,5-Nov-12
+Bank of Florida - Southwest,Naples,FL,35106,EverBank,28-May-10,5-Nov-12
+Bank of Florida - Southeast,Fort Lauderdale,FL,57360,EverBank,28-May-10,5-Nov-12
+Pinehurst Bank,Saint Paul,MN,57735,Coulee Bank,21-May-10,26-Oct-12
+Midwest Bank and Trust Company,Elmwood Park,IL,18117,"FirstMerit Bank, N.A.",14-May-10,23-Aug-12
+Southwest Community Bank,Springfield,MO,34255,Simmons First National Bank,14-May-10,23-Aug-12
+New Liberty Bank,Plymouth,MI,35586,Bank of Ann Arbor,14-May-10,23-Aug-12
+Satilla Community Bank,Saint Marys,GA,35114,Ameris Bank,14-May-10,5-Nov-12
+1st Pacific Bank of California,San Diego,CA,35517,City National Bank,7-May-10,13-Dec-12
+Towne Bank of Arizona,Mesa,AZ,57697,Commerce Bank of Arizona,7-May-10,23-Aug-12
+Access Bank,Champlin,MN,16476,PrinsBank,7-May-10,23-Aug-12
+The Bank of Bonifay,Bonifay,FL,14246,First Federal Bank of Florida,7-May-10,5-Nov-12
+Frontier Bank,Everett,WA,22710,"Union Bank, N.A.",30-Apr-10,15-Jan-13
+BC National Banks,Butler,MO,17792,Community First Bank,30-Apr-10,23-Aug-12
+Champion Bank,Creve Coeur,MO,58362,BankLiberty,30-Apr-10,23-Aug-12
+CF Bancorp,Port Huron,MI,30005,First Michigan Bank,30-Apr-10,15-Jan-13
+Westernbank Puerto Rico,Mayaguez,PR,31027,Banco Popular de Puerto Rico,30-Apr-10,5-Nov-12
+R-G Premier Bank of Puerto Rico,Hato Rey,PR,32185,Scotiabank de Puerto Rico,30-Apr-10,5-Nov-12
+Eurobank,San Juan,PR,27150,Oriental Bank and Trust,30-Apr-10,5-Nov-12
+Wheatland Bank,Naperville,IL,58429,Wheaton Bank & Trust,23-Apr-10,23-Aug-12
+Peotone Bank and Trust Company,Peotone,IL,10888,First Midwest Bank,23-Apr-10,23-Aug-12
+Lincoln Park Savings Bank,Chicago,IL,30600,Northbrook Bank & Trust Company,23-Apr-10,23-Aug-12
+New Century Bank,Chicago,IL,34821,"MB Financial Bank, N.A.",23-Apr-10,23-Aug-12
+Citizens Bank and Trust Company of Chicago,Chicago,IL,34658,Republic Bank of Chicago,23-Apr-10,23-Aug-12
+Broadway Bank,Chicago,IL,22853,"MB Financial Bank, N.A.",23-Apr-10,23-Aug-12
+"Amcore Bank, National Association",Rockford,IL,3735,Harris N.A.,23-Apr-10,23-Aug-12
+City Bank,Lynnwood,WA,21521,Whidbey Island Bank,16-Apr-10,14-Sep-12
+Tamalpais Bank,San Rafael,CA,33493,"Union Bank, N.A.",16-Apr-10,23-Aug-12
+Innovative Bank,Oakland,CA,23876,Center Bank,16-Apr-10,23-Aug-12
+Butler Bank,Lowell,MA,26619,People's United Bank,16-Apr-10,23-Aug-12
+Riverside National Bank of Florida,Fort Pierce,FL,24067,"TD Bank, N.A.",16-Apr-10,5-Nov-12
+AmericanFirst Bank,Clermont,FL,57724,"TD Bank, N.A.",16-Apr-10,31-Oct-12
+First Federal Bank of North Florida,Palatka,FL,28886,"TD Bank, N.A.",16-Apr-10,15-Jan-13
+Lakeside Community Bank,Sterling Heights,MI,34878,No Acquirer,16-Apr-10,23-Aug-12
+Beach First National Bank,Myrtle Beach,SC,34242,Bank of North Carolina,9-Apr-10,5-Nov-12
+Desert Hills Bank,Phoenix,AZ,57060,New York Community Bank,26-Mar-10,23-Aug-12
+Unity National Bank,Cartersville,GA,34678,Bank of the Ozarks,26-Mar-10,14-Sep-12
+Key West Bank,Key West,FL,34684,Centennial Bank,26-Mar-10,23-Aug-12
+McIntosh Commercial Bank,Carrollton,GA,57399,CharterBank,26-Mar-10,23-Aug-12
+State Bank of Aurora,Aurora,MN,8221,Northern State Bank,19-Mar-10,23-Aug-12
+First Lowndes Bank,Fort Deposit,AL,24957,First Citizens Bank,19-Mar-10,23-Aug-12
+Bank of Hiawassee,Hiawassee,GA,10054,Citizens South Bank,19-Mar-10,23-Aug-12
+Appalachian Community Bank,Ellijay,GA,33989,Community & Southern Bank,19-Mar-10,31-Oct-12
+Advanta Bank Corp.,Draper,UT,33535,No Acquirer,19-Mar-10,14-Sep-12
+Century Security Bank,Duluth,GA,58104,Bank of Upson,19-Mar-10,23-Aug-12
+American National Bank,Parma,OH,18806,The National Bank and Trust Company,19-Mar-10,23-Aug-12
+Statewide Bank,Covington,LA,29561,Home Bank,12-Mar-10,23-Aug-12
+Old Southern Bank,Orlando,FL,58182,Centennial Bank,12-Mar-10,23-Aug-12
+The Park Avenue Bank,New York,NY,27096,Valley National Bank,12-Mar-10,23-Aug-12
+LibertyPointe Bank,New York,NY,58071,Valley National Bank,11-Mar-10,23-Aug-12
+Centennial Bank,Ogden,UT,34430,No Acquirer,5-Mar-10,14-Sep-12
+Waterfield Bank,Germantown,MD,34976,No Acquirer,5-Mar-10,23-Aug-12
+Bank of Illinois,Normal,IL,9268,Heartland Bank and Trust Company,5-Mar-10,23-Aug-12
+Sun American Bank,Boca Raton,FL,27126,First-Citizens Bank & Trust Company,5-Mar-10,23-Aug-12
+Rainier Pacific Bank,Tacoma,WA,38129,Umpqua Bank,26-Feb-10,23-Aug-12
+Carson River Community Bank,Carson City,NV,58352,Heritage Bank of Nevada,26-Feb-10,15-Jan-13
+"La Jolla Bank, FSB",La Jolla,CA,32423,"OneWest Bank, FSB",19-Feb-10,24-Aug-12
+George Washington Savings Bank,Orland Park,IL,29952,"FirstMerit Bank, N.A.",19-Feb-10,24-Aug-12
+The La Coste National Bank,La Coste,TX,3287,Community National Bank,19-Feb-10,14-Sep-12
+Marco Community Bank,Marco Island,FL,57586,Mutual of Omaha Bank,19-Feb-10,24-Aug-12
+1st American State Bank of Minnesota,Hancock,MN,15448,"Community Development Bank, FSB",5-Feb-10,24-Aug-12
+American Marine Bank,Bainbridge Island,WA,16730,Columbia State Bank,29-Jan-10,24-Aug-12
+First Regional Bank,Los Angeles,CA,23011,First-Citizens Bank & Trust Company,29-Jan-10,24-Aug-12
+Community Bank and Trust,Cornelia,GA,5702,SCBT National Association,29-Jan-10,15-Jan-13
+"Marshall Bank, N.A.",Hallock,MN,16133,United Valley Bank,29-Jan-10,23-Aug-12
+Florida Community Bank,Immokalee,FL,5672,"Premier American Bank, N.A.",29-Jan-10,15-Jan-13
+First National Bank of Georgia,Carrollton,GA,16480,Community & Southern Bank,29-Jan-10,13-Dec-12
+Columbia River Bank,The Dalles,OR,22469,Columbia State Bank,22-Jan-10,14-Sep-12
+Evergreen Bank,Seattle,WA,20501,Umpqua Bank,22-Jan-10,15-Jan-13
+Charter Bank,Santa Fe,NM,32498,Charter Bank,22-Jan-10,23-Aug-12
+Bank of Leeton,Leeton,MO,8265,"Sunflower Bank, N.A.",22-Jan-10,15-Jan-13
+Premier American Bank,Miami,FL,57147,"Premier American Bank, N.A.",22-Jan-10,13-Dec-12
+Barnes Banking Company,Kaysville,UT,1252,No Acquirer,15-Jan-10,23-Aug-12
+St. Stephen State Bank,St. Stephen,MN,17522,First State Bank of St. Joseph,15-Jan-10,23-Aug-12
+Town Community Bank & Trust,Antioch,IL,34705,First American Bank,15-Jan-10,23-Aug-12
+Horizon Bank,Bellingham,WA,22977,Washington Federal Savings and Loan Association,8-Jan-10,23-Aug-12
+"First Federal Bank of California, F.S.B.",Santa Monica,CA,28536,"OneWest Bank, FSB",18-Dec-09,23-Aug-12
+Imperial Capital Bank,La Jolla,CA,26348,City National Bank,18-Dec-09,5-Sep-12
+Independent Bankers' Bank,Springfield,IL,26820,The Independent BankersBank (TIB),18-Dec-09,23-Aug-12
+New South Federal Savings Bank,Irondale,AL,32276,Beal Bank,18-Dec-09,23-Aug-12
+Citizens State Bank,New Baltimore,MI,1006,No Acquirer,18-Dec-09,5-Nov-12
+Peoples First Community Bank,Panama City,FL,32167,Hancock Bank,18-Dec-09,5-Nov-12
+RockBridge Commercial Bank,Atlanta,GA,58315,No Acquirer,18-Dec-09,5-Nov-12
+SolutionsBank,Overland Park,KS,4731,Arvest Bank,11-Dec-09,23-Aug-12
+"Valley Capital Bank, N.A.",Mesa,AZ,58399,Enterprise Bank & Trust,11-Dec-09,23-Aug-12
+"Republic Federal Bank, N.A.",Miami,FL,22846,1st United Bank,11-Dec-09,5-Nov-12
+Greater Atlantic Bank,Reston,VA,32583,Sonabank,4-Dec-09,5-Nov-12
+Benchmark Bank,Aurora,IL,10440,"MB Financial Bank, N.A.",4-Dec-09,23-Aug-12
+AmTrust Bank,Cleveland,OH,29776,New York Community Bank,4-Dec-09,5-Nov-12
+The Tattnall Bank,Reidsville,GA,12080,Heritage Bank of the South,4-Dec-09,5-Nov-12
+First Security National Bank,Norcross,GA,26290,State Bank and Trust Company,4-Dec-09,5-Nov-12
+The Buckhead Community Bank,Atlanta,GA,34663,State Bank and Trust Company,4-Dec-09,5-Nov-12
+Commerce Bank of Southwest Florida,Fort Myers,FL,58016,Central Bank,20-Nov-09,5-Nov-12
+Pacific Coast National Bank,San Clemente,CA,57914,Sunwest Bank,13-Nov-09,22-Aug-12
+Orion Bank,Naples,FL,22427,IBERIABANK,13-Nov-09,5-Nov-12
+"Century Bank, F.S.B.",Sarasota,FL,32267,IBERIABANK,13-Nov-09,22-Aug-12
+United Commercial Bank,San Francisco,CA,32469,East West Bank,6-Nov-09,5-Nov-12
+Gateway Bank of St. Louis,St. Louis,MO,19450,Central Bank of Kansas City,6-Nov-09,22-Aug-12
+Prosperan Bank,Oakdale,MN,35074,"Alerus Financial, N.A.",6-Nov-09,22-Aug-12
+Home Federal Savings Bank,Detroit,MI,30329,Liberty Bank and Trust Company,6-Nov-09,22-Aug-12
+United Security Bank,Sparta,GA,22286,Ameris Bank,6-Nov-09,15-Jan-13
+North Houston Bank,Houston,TX,18776,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+Madisonville State Bank,Madisonville,TX,33782,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+Citizens National Bank,Teague,TX,25222,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+Park National Bank,Chicago,IL,11677,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+Pacific National Bank,San Francisco,CA,30006,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+California National Bank,Los Angeles,CA,34659,U.S. Bank N.A.,30-Oct-09,5-Sep-12
+San Diego National Bank,San Diego,CA,23594,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+Community Bank of Lemont,Lemont,IL,35291,U.S. Bank N.A.,30-Oct-09,15-Jan-13
+"Bank USA, N.A.",Phoenix,AZ,32218,U.S. Bank N.A.,30-Oct-09,22-Aug-12
+First DuPage Bank,Westmont,IL,35038,First Midwest Bank,23-Oct-09,22-Aug-12
+Riverview Community Bank,Otsego,MN,57525,Central Bank,23-Oct-09,22-Aug-12
+Bank of Elmwood,Racine,WI,18321,Tri City National Bank,23-Oct-09,22-Aug-12
+Flagship National Bank,Bradenton,FL,35044,First Federal Bank of Florida,23-Oct-09,22-Aug-12
+Hillcrest Bank Florida,Naples,FL,58336,Stonegate Bank,23-Oct-09,22-Aug-12
+American United Bank,Lawrenceville,GA,57794,Ameris Bank,23-Oct-09,5-Sep-12
+Partners Bank,Naples,FL,57959,Stonegate Bank,23-Oct-09,15-Jan-13
+San Joaquin Bank,Bakersfield,CA,23266,Citizens Business Bank,16-Oct-09,22-Aug-12
+Southern Colorado National Bank,Pueblo,CO,57263,Legacy Bank,2-Oct-09,5-Sep-12
+Jennings State Bank,Spring Grove,MN,11416,Central Bank,2-Oct-09,21-Aug-12
+Warren Bank,Warren,MI,34824,The Huntington National Bank,2-Oct-09,21-Aug-12
+Georgian Bank,Atlanta,GA,57151,"First Citizens Bank and Trust Company, Inc.",25-Sep-09,21-Aug-12
+"Irwin Union Bank, F.S.B.",Louisville,KY,57068,"First Financial Bank, N.A.",18-Sep-09,5-Sep-12
+Irwin Union Bank and Trust Company,Columbus,IN,10100,"First Financial Bank, N.A.",18-Sep-09,21-Aug-12
+Venture Bank,Lacey,WA,22868,First-Citizens Bank & Trust Company,11-Sep-09,21-Aug-12
+Brickwell Community Bank,Woodbury,MN,57736,CorTrust Bank N.A.,11-Sep-09,15-Jan-13
+"Corus Bank, N.A.",Chicago,IL,13693,"MB Financial Bank, N.A.",11-Sep-09,21-Aug-12
+First State Bank,Flagstaff,AZ,34875,Sunwest Bank,4-Sep-09,15-Jan-13
+Platinum Community Bank,Rolling Meadows,IL,35030,No Acquirer,4-Sep-09,21-Aug-12
+Vantus Bank,Sioux City,IN,27732,Great Southern Bank,4-Sep-09,21-Aug-12
+InBank,Oak Forest,IL,20203,"MB Financial Bank, N.A.",4-Sep-09,21-Aug-12
+First Bank of Kansas City,Kansas City,MO,25231,Great American Bank,4-Sep-09,21-Aug-12
+Affinity Bank,Ventura,CA,27197,Pacific Western Bank,28-Aug-09,21-Aug-12
+Mainstreet Bank,Forest Lake,MN,1909,Central Bank,28-Aug-09,21-Aug-12
+Bradford Bank,Baltimore,MD,28312,Manufacturers and Traders Trust Company (M&T Bank),28-Aug-09,15-Jan-13
+Guaranty Bank,Austin,TX,32618,BBVA Compass,21-Aug-09,21-Aug-12
+CapitalSouth Bank,Birmingham,AL,22130,IBERIABANK,21-Aug-09,15-Jan-13
+First Coweta Bank,Newnan,GA,57702,United Bank,21-Aug-09,15-Jan-13
+ebank,Atlanta,GA,34682,"Stearns Bank, N.A.",21-Aug-09,21-Aug-12
+Community Bank of Nevada,Las Vegas,NV,34043,No Acquirer,14-Aug-09,21-Aug-12
+Community Bank of Arizona,Phoenix,AZ,57645,MidFirst Bank,14-Aug-09,21-Aug-12
+"Union Bank, National Association",Gilbert,AZ,34485,MidFirst Bank,14-Aug-09,21-Aug-12
+Colonial Bank,Montgomery,AL,9609,"Branch Banking & Trust Company, (BB&T)",14-Aug-09,5-Sep-12
+Dwelling House Savings and Loan Association,Pittsburgh,PA,31559,"PNC Bank, N.A.",14-Aug-09,15-Jan-13
+Community First Bank,Prineville,OR,23268,Home Federal Bank,7-Aug-09,15-Jan-13
+Community National Bank of Sarasota County,Venice,FL,27183,"Stearns Bank, N.A.",7-Aug-09,20-Aug-12
+First State Bank,Sarasota,FL,27364,"Stearns Bank, N.A.",7-Aug-09,20-Aug-12
+Mutual Bank,Harvey,IL,18659,United Central Bank,31-Jul-09,20-Aug-12
+First BankAmericano,Elizabeth,NJ,34270,Crown Bank,31-Jul-09,20-Aug-12
+Peoples Community Bank,West Chester,OH,32288,"First Financial Bank, N.A.",31-Jul-09,20-Aug-12
+Integrity Bank,Jupiter,FL,57604,Stonegate Bank,31-Jul-09,20-Aug-12
+First State Bank of Altus,Altus,OK,9873,Herring Bank,31-Jul-09,20-Aug-12
+Security Bank of Jones County,Gray,GA,8486,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Security Bank of Houston County,Perry,GA,27048,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Security Bank of Bibb County,Macon,GA,27367,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Security Bank of North Metro,Woodstock,GA,57105,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Security Bank of North Fulton,Alpharetta,GA,57430,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Security Bank of Gwinnett County,Suwanee,GA,57346,State Bank and Trust Company,24-Jul-09,20-Aug-12
+Waterford Village Bank,Williamsville,NY,58065,"Evans Bank, N.A.",24-Jul-09,20-Aug-12
+Temecula Valley Bank,Temecula,CA,34341,First-Citizens Bank & Trust Company,17-Jul-09,20-Aug-12
+Vineyard Bank,Rancho Cucamonga,CA,23556,California Bank & Trust,17-Jul-09,20-Aug-12
+BankFirst,Sioux Falls,SD,34103,"Alerus Financial, N.A.",17-Jul-09,20-Aug-12
+First Piedmont Bank,Winder,GA,34594,First American Bank and Trust Company,17-Jul-09,15-Jan-13
+Bank of Wyoming,Thermopolis,WY,22754,Central Bank & Trust,10-Jul-09,20-Aug-12
+Founders Bank,Worth,IL,18390,The PrivateBank and Trust Company,2-Jul-09,20-Aug-12
+Millennium State Bank of Texas,Dallas,TX,57667,State Bank of Texas,2-Jul-09,26-Oct-12
+First National Bank of Danville,Danville,IL,3644,"First Financial Bank, N.A.",2-Jul-09,20-Aug-12
+Elizabeth State Bank,Elizabeth,IL,9262,Galena State Bank and Trust Company,2-Jul-09,20-Aug-12
+Rock River Bank,Oregon,IL,15302,The Harvard State Bank,2-Jul-09,20-Aug-12
+First State Bank of Winchester,Winchester,IL,11710,The First National Bank of Beardstown,2-Jul-09,20-Aug-12
+John Warner Bank,Clinton,IL,12093,State Bank of Lincoln,2-Jul-09,20-Aug-12
+Mirae Bank,Los Angeles,CA,57332,Wilshire State Bank,26-Jun-09,20-Aug-12
+MetroPacific Bank,Irvine,CA,57893,Sunwest Bank,26-Jun-09,20-Aug-12
+Horizon Bank,Pine City,MN,9744,"Stearns Bank, N.A.",26-Jun-09,20-Aug-12
+Neighborhood Community Bank,Newnan,GA,35285,CharterBank,26-Jun-09,20-Aug-12
+Community Bank of West Georgia,Villa Rica,GA,57436,No Acquirer,26-Jun-09,17-Aug-12
+First National Bank of Anthony,Anthony,KS,4614,Bank of Kansas,19-Jun-09,17-Aug-12
+Cooperative Bank,Wilmington,NC,27837,First Bank,19-Jun-09,17-Aug-12
+Southern Community Bank,Fayetteville,GA,35251,United Community Bank,19-Jun-09,17-Aug-12
+Bank of Lincolnwood,Lincolnwood,IL,17309,Republic Bank of Chicago,5-Jun-09,17-Aug-12
+Citizens National Bank,Macomb,IL,5757,Morton Community Bank,22-May-09,4-Sep-12
+Strategic Capital Bank,Champaign,IL,35175,Midland States Bank,22-May-09,4-Sep-12
+"BankUnited, FSB",Coral Gables,FL,32247,BankUnited,21-May-09,17-Aug-12
+Westsound Bank,Bremerton,WA,34843,Kitsap Bank,8-May-09,4-Sep-12
+America West Bank,Layton,UT,35461,Cache Valley Bank,1-May-09,17-Aug-12
+Citizens Community Bank,Ridgewood,NJ,57563,North Jersey Community Bank,1-May-09,4-Sep-12
+"Silverton Bank, NA",Atlanta,GA,26535,No Acquirer,1-May-09,17-Aug-12
+First Bank of Idaho,Ketchum,ID,34396,"U.S. Bank, N.A.",24-Apr-09,17-Aug-12
+First Bank of Beverly Hills,Calabasas,CA,32069,No Acquirer,24-Apr-09,4-Sep-12
+Michigan Heritage Bank,Farmington Hills,MI,34369,Level One Bank,24-Apr-09,17-Aug-12
+American Southern Bank,Kennesaw,GA,57943,Bank of North Georgia,24-Apr-09,17-Aug-12
+Great Basin Bank of Nevada,Elko,NV,33824,Nevada State Bank,17-Apr-09,4-Sep-12
+American Sterling Bank,Sugar Creek,MO,8266,Metcalf Bank,17-Apr-09,31-Aug-12
+New Frontier Bank,Greeley,CO,34881,No Acquirer,10-Apr-09,4-Sep-12
+Cape Fear Bank,Wilmington,NC,34639,First Federal Savings and Loan Association,10-Apr-09,17-Aug-12
+Omni National Bank,Atlanta,GA,22238,No Acquirer,27-Mar-09,17-Aug-12
+"TeamBank, NA",Paola,KS,4754,Great Southern Bank,20-Mar-09,17-Aug-12
+Colorado National Bank,Colorado Springs,CO,18896,Herring Bank,20-Mar-09,17-Aug-12
+FirstCity Bank,Stockbridge,GA,18243,No Acquirer,20-Mar-09,17-Aug-12
+Freedom Bank of Georgia,Commerce,GA,57558,Northeast Georgia Bank,6-Mar-09,17-Aug-12
+Security Savings Bank,Henderson,NV,34820,Bank of Nevada,27-Feb-09,7-Sep-12
+Heritage Community Bank,Glenwood,IL,20078,"MB Financial Bank, N.A.",27-Feb-09,17-Aug-12
+Silver Falls Bank,Silverton,OR,35399,Citizens Bank,20-Feb-09,17-Aug-12
+Pinnacle Bank of Oregon,Beaverton,OR,57342,Washington Trust Bank of Spokane,13-Feb-09,17-Aug-12
+Corn Belt Bank & Trust Co.,Pittsfield,IL,16500,The Carlinville National Bank,13-Feb-09,17-Aug-12
+Riverside Bank of the Gulf Coast,Cape Coral,FL,34563,TIB Bank,13-Feb-09,17-Aug-12
+Sherman County Bank,Loup City,NE,5431,Heritage Bank,13-Feb-09,17-Aug-12
+County Bank,Merced,CA,22574,Westamerica Bank,6-Feb-09,4-Sep-12
+Alliance Bank,Culver City,CA,23124,California Bank & Trust,6-Feb-09,16-Aug-12
+FirstBank Financial Services,McDonough,GA,57017,Regions Bank,6-Feb-09,16-Aug-12
+Ocala National Bank,Ocala,FL,26538,"CenterState Bank of Florida, N.A.",30-Jan-09,4-Sep-12
+Suburban FSB,Crofton,MD,30763,Bank of Essex,30-Jan-09,16-Aug-12
+MagnetBank,Salt Lake City,UT,58001,No Acquirer,30-Jan-09,16-Aug-12
+1st Centennial Bank,Redlands,CA,33025,First California Bank,23-Jan-09,16-Aug-12
+Bank of Clark County,Vancouver,WA,34959,Umpqua Bank,16-Jan-09,16-Aug-12
+National Bank of Commerce,Berkeley,IL,19733,Republic Bank of Chicago,16-Jan-09,16-Aug-12
+Sanderson State Bank,Sanderson,TX,11568,The Pecos County State Bank,12-Dec-08,4-Sep-12
+Haven Trust Bank,Duluth,GA,35379,"Branch Banking & Trust Company, (BB&T)",12-Dec-08,16-Aug-12
+First Georgia Community Bank,Jackson,GA,34301,United Bank,5-Dec-08,16-Aug-12
+PFF Bank & Trust,Pomona,CA,28344,"U.S. Bank, N.A.",21-Nov-08,4-Jan-13
+Downey Savings & Loan,Newport Beach,CA,30968,"U.S. Bank, N.A.",21-Nov-08,4-Jan-13
+Community Bank,Loganville,GA,16490,Bank of Essex,21-Nov-08,4-Sep-12
+Security Pacific Bank,Los Angeles,CA,23595,Pacific Western Bank,7-Nov-08,28-Aug-12
+"Franklin Bank, SSB",Houston,TX,26870,Prosperity Bank,7-Nov-08,16-Aug-12
+Freedom Bank,Bradenton,FL,57930,Fifth Third Bank,31-Oct-08,16-Aug-12
+Alpha Bank & Trust,Alpharetta,GA,58241,"Stearns Bank, N.A.",24-Oct-08,16-Aug-12
+Meridian Bank,Eldred,IL,13789,National Bank,10-Oct-08,31-May-12
+Main Street Bank,Northville,MI,57654,Monroe Bank & Trust,10-Oct-08,16-Aug-12
+Washington Mutual Bank,Henderson,NV,32633,JP Morgan Chase Bank,25-Sep-08,16-Aug-12
+Ameribank,Northfork,WV,6782,The Citizens Savings Bank,19-Sep-08,16-Aug-12
+Silver State Bank,Henderson,NV,34194,Nevada State Bank,5-Sep-08,16-Aug-12
+Integrity Bank,Alpharetta,GA,35469,Regions Bank,29-Aug-08,16-Aug-12
+Columbian Bank & Trust,Topeka,KS,22728,Citizens Bank & Trust,22-Aug-08,16-Aug-12
+First Priority Bank,Bradenton,FL,57523,SunTrust Bank,1-Aug-08,16-Aug-12
+"First Heritage Bank, NA",Newport Beach,CA,57961,Mutual of Omaha Bank,25-Jul-08,28-Aug-12
+First National Bank of Nevada,Reno,NV,27011,Mutual of Omaha Bank,25-Jul-08,28-Aug-12
+IndyMac Bank,Pasadena,CA,29730,"OneWest Bank, FSB",11-Jul-08,28-Aug-12
+"First Integrity Bank, NA",Staples,MN,12736,First International Bank and Trust,30-May-08,28-Aug-12
+"ANB Financial, NA",Bentonville,AR,33901,Pulaski Bank and Trust Company,9-May-08,28-Aug-12
+Hume Bank,Hume,MO,1971,Security Bank,7-Mar-08,28-Aug-12
+Douglass National Bank,Kansas City,MO,24660,Liberty Bank and Trust Company,25-Jan-08,26-Oct-12
+Miami Valley Bank,Lakeview,OH,16848,The Citizens Banking Company,4-Oct-07,28-Aug-12
+NetBank,Alpharetta,GA,32575,ING DIRECT,28-Sep-07,28-Aug-12
+Metropolitan Savings Bank,Pittsburgh,PA,35353,Allegheny Valley Bank of Pittsburgh,2-Feb-07,27-Oct-10
+Bank of Ephraim,Ephraim,UT,1249,Far West Bank,25-Jun-04,9-Apr-08
+Reliance Bank,White Plains,NY,26778,Union State Bank,19-Mar-04,9-Apr-08
+Guaranty National Bank of Tallahassee,Tallahassee,FL,26838,Hancock Bank of Florida,12-Mar-04,5-Jun-12
+Dollar Savings Bank,Newark,NJ,31330,No Acquirer,14-Feb-04,9-Apr-08
+Pulaski Savings Bank,Philadelphia,PA,27203,Earthstar Bank,14-Nov-03,22-Jul-05
+First National Bank of Blanchardville,Blanchardville,WI,11639,The Park Bank,9-May-03,5-Jun-12
+Southern Pacific Bank,Torrance,CA,27094,Beal Bank,7-Feb-03,20-Oct-08
+Farmers Bank of Cheneyville,Cheneyville,LA,16445,Sabine State Bank & Trust,17-Dec-02,20-Oct-04
+Bank of Alamo,Alamo,TN,9961,No Acquirer,8-Nov-02,18-Mar-05
+AmTrade International Bank,Atlanta,GA,33784,No Acquirer,30-Sep-02,11-Sep-06
+Universal Federal Savings Bank,Chicago,IL,29355,Chicago Community Bank,27-Jun-02,9-Apr-08
+Connecticut Bank of Commerce,Stamford,CT,19183,Hudson United Bank,26-Jun-02,14-Feb-12
+New Century Bank,Shelby Township,MI,34979,No Acquirer,28-Mar-02,18-Mar-05
+Net 1st National Bank,Boca Raton,FL,26652,Bank Leumi USA,1-Mar-02,9-Apr-08
+"NextBank, NA",Phoenix,AZ,22314,No Acquirer,7-Feb-02,27-Aug-10
+Oakwood Deposit Bank Co.,Oakwood,OH,8966,The State Bank & Trust Company,1-Feb-02,25-Oct-12
+Bank of Sierra Blanca,Sierra Blanca,TX,22002,The Security State Bank of Pecos,18-Jan-02,6-Nov-03
+"Hamilton Bank, NA",Miami,FL,24382,Israel Discount Bank of New York,11-Jan-02,5-Jun-12
+Sinclair National Bank,Gravette,AR,34248,Delta Trust & Bank,7-Sep-01,10-Feb-04
+"Superior Bank, FSB",Hinsdale,IL,32646,"Superior Federal, FSB",27-Jul-01,5-Jun-12
+Malta National Bank,Malta,OH,6629,North Valley Bank,3-May-01,18-Nov-02
+First Alliance Bank & Trust Co.,Manchester,NH,34264,Southern New Hampshire Bank & Trust,2-Feb-01,18-Feb-03
+National State Bank of Metropolis,Metropolis,IL,3815,Banterra Bank of Marion,14-Dec-00,17-Mar-05
+Bank of Honolulu,Honolulu,HI,21029,Bank of the Orient,13-Oct-00,17-Mar-05
diff --git a/pandas/pandas/tests/io/data/csv/iris.csv b/pandas/pandas/tests/io/data/csv/iris.csv
new file mode 100644
index 0000000000000..c19b9c3688515
--- /dev/null
+++ b/pandas/pandas/tests/io/data/csv/iris.csv
@@ -0,0 +1,151 @@
+SepalLength,SepalWidth,PetalLength,PetalWidth,Name
+5.1,3.5,1.4,0.2,Iris-setosa
+4.9,3.0,1.4,0.2,Iris-setosa
+4.7,3.2,1.3,0.2,Iris-setosa
+4.6,3.1,1.5,0.2,Iris-setosa
+5.0,3.6,1.4,0.2,Iris-setosa
+5.4,3.9,1.7,0.4,Iris-setosa
+4.6,3.4,1.4,0.3,Iris-setosa
+5.0,3.4,1.5,0.2,Iris-setosa
+4.4,2.9,1.4,0.2,Iris-setosa
+4.9,3.1,1.5,0.1,Iris-setosa
+5.4,3.7,1.5,0.2,Iris-setosa
+4.8,3.4,1.6,0.2,Iris-setosa
+4.8,3.0,1.4,0.1,Iris-setosa
+4.3,3.0,1.1,0.1,Iris-setosa
+5.8,4.0,1.2,0.2,Iris-setosa
+5.7,4.4,1.5,0.4,Iris-setosa
+5.4,3.9,1.3,0.4,Iris-setosa
+5.1,3.5,1.4,0.3,Iris-setosa
+5.7,3.8,1.7,0.3,Iris-setosa
+5.1,3.8,1.5,0.3,Iris-setosa
+5.4,3.4,1.7,0.2,Iris-setosa
+5.1,3.7,1.5,0.4,Iris-setosa
+4.6,3.6,1.0,0.2,Iris-setosa
+5.1,3.3,1.7,0.5,Iris-setosa
+4.8,3.4,1.9,0.2,Iris-setosa
+5.0,3.0,1.6,0.2,Iris-setosa
+5.0,3.4,1.6,0.4,Iris-setosa
+5.2,3.5,1.5,0.2,Iris-setosa
+5.2,3.4,1.4,0.2,Iris-setosa
+4.7,3.2,1.6,0.2,Iris-setosa
+4.8,3.1,1.6,0.2,Iris-setosa
+5.4,3.4,1.5,0.4,Iris-setosa
+5.2,4.1,1.5,0.1,Iris-setosa
+5.5,4.2,1.4,0.2,Iris-setosa
+4.9,3.1,1.5,0.1,Iris-setosa
+5.0,3.2,1.2,0.2,Iris-setosa
+5.5,3.5,1.3,0.2,Iris-setosa
+4.9,3.1,1.5,0.1,Iris-setosa
+4.4,3.0,1.3,0.2,Iris-setosa
+5.1,3.4,1.5,0.2,Iris-setosa
+5.0,3.5,1.3,0.3,Iris-setosa
+4.5,2.3,1.3,0.3,Iris-setosa
+4.4,3.2,1.3,0.2,Iris-setosa
+5.0,3.5,1.6,0.6,Iris-setosa
+5.1,3.8,1.9,0.4,Iris-setosa
+4.8,3.0,1.4,0.3,Iris-setosa
+5.1,3.8,1.6,0.2,Iris-setosa
+4.6,3.2,1.4,0.2,Iris-setosa
+5.3,3.7,1.5,0.2,Iris-setosa
+5.0,3.3,1.4,0.2,Iris-setosa
+7.0,3.2,4.7,1.4,Iris-versicolor
+6.4,3.2,4.5,1.5,Iris-versicolor
+6.9,3.1,4.9,1.5,Iris-versicolor
+5.5,2.3,4.0,1.3,Iris-versicolor
+6.5,2.8,4.6,1.5,Iris-versicolor
+5.7,2.8,4.5,1.3,Iris-versicolor
+6.3,3.3,4.7,1.6,Iris-versicolor
+4.9,2.4,3.3,1.0,Iris-versicolor
+6.6,2.9,4.6,1.3,Iris-versicolor
+5.2,2.7,3.9,1.4,Iris-versicolor
+5.0,2.0,3.5,1.0,Iris-versicolor
+5.9,3.0,4.2,1.5,Iris-versicolor
+6.0,2.2,4.0,1.0,Iris-versicolor
+6.1,2.9,4.7,1.4,Iris-versicolor
+5.6,2.9,3.6,1.3,Iris-versicolor
+6.7,3.1,4.4,1.4,Iris-versicolor
+5.6,3.0,4.5,1.5,Iris-versicolor
+5.8,2.7,4.1,1.0,Iris-versicolor
+6.2,2.2,4.5,1.5,Iris-versicolor
+5.6,2.5,3.9,1.1,Iris-versicolor
+5.9,3.2,4.8,1.8,Iris-versicolor
+6.1,2.8,4.0,1.3,Iris-versicolor
+6.3,2.5,4.9,1.5,Iris-versicolor
+6.1,2.8,4.7,1.2,Iris-versicolor
+6.4,2.9,4.3,1.3,Iris-versicolor
+6.6,3.0,4.4,1.4,Iris-versicolor
+6.8,2.8,4.8,1.4,Iris-versicolor
+6.7,3.0,5.0,1.7,Iris-versicolor
+6.0,2.9,4.5,1.5,Iris-versicolor
+5.7,2.6,3.5,1.0,Iris-versicolor
+5.5,2.4,3.8,1.1,Iris-versicolor
+5.5,2.4,3.7,1.0,Iris-versicolor
+5.8,2.7,3.9,1.2,Iris-versicolor
+6.0,2.7,5.1,1.6,Iris-versicolor
+5.4,3.0,4.5,1.5,Iris-versicolor
+6.0,3.4,4.5,1.6,Iris-versicolor
+6.7,3.1,4.7,1.5,Iris-versicolor
+6.3,2.3,4.4,1.3,Iris-versicolor
+5.6,3.0,4.1,1.3,Iris-versicolor
+5.5,2.5,4.0,1.3,Iris-versicolor
+5.5,2.6,4.4,1.2,Iris-versicolor
+6.1,3.0,4.6,1.4,Iris-versicolor
+5.8,2.6,4.0,1.2,Iris-versicolor
+5.0,2.3,3.3,1.0,Iris-versicolor
+5.6,2.7,4.2,1.3,Iris-versicolor
+5.7,3.0,4.2,1.2,Iris-versicolor
+5.7,2.9,4.2,1.3,Iris-versicolor
+6.2,2.9,4.3,1.3,Iris-versicolor
+5.1,2.5,3.0,1.1,Iris-versicolor
+5.7,2.8,4.1,1.3,Iris-versicolor
+6.3,3.3,6.0,2.5,Iris-virginica
+5.8,2.7,5.1,1.9,Iris-virginica
+7.1,3.0,5.9,2.1,Iris-virginica
+6.3,2.9,5.6,1.8,Iris-virginica
+6.5,3.0,5.8,2.2,Iris-virginica
+7.6,3.0,6.6,2.1,Iris-virginica
+4.9,2.5,4.5,1.7,Iris-virginica
+7.3,2.9,6.3,1.8,Iris-virginica
+6.7,2.5,5.8,1.8,Iris-virginica
+7.2,3.6,6.1,2.5,Iris-virginica
+6.5,3.2,5.1,2.0,Iris-virginica
+6.4,2.7,5.3,1.9,Iris-virginica
+6.8,3.0,5.5,2.1,Iris-virginica
+5.7,2.5,5.0,2.0,Iris-virginica
+5.8,2.8,5.1,2.4,Iris-virginica
+6.4,3.2,5.3,2.3,Iris-virginica
+6.5,3.0,5.5,1.8,Iris-virginica
+7.7,3.8,6.7,2.2,Iris-virginica
+7.7,2.6,6.9,2.3,Iris-virginica
+6.0,2.2,5.0,1.5,Iris-virginica
+6.9,3.2,5.7,2.3,Iris-virginica
+5.6,2.8,4.9,2.0,Iris-virginica
+7.7,2.8,6.7,2.0,Iris-virginica
+6.3,2.7,4.9,1.8,Iris-virginica
+6.7,3.3,5.7,2.1,Iris-virginica
+7.2,3.2,6.0,1.8,Iris-virginica
+6.2,2.8,4.8,1.8,Iris-virginica
+6.1,3.0,4.9,1.8,Iris-virginica
+6.4,2.8,5.6,2.1,Iris-virginica
+7.2,3.0,5.8,1.6,Iris-virginica
+7.4,2.8,6.1,1.9,Iris-virginica
+7.9,3.8,6.4,2.0,Iris-virginica
+6.4,2.8,5.6,2.2,Iris-virginica
+6.3,2.8,5.1,1.5,Iris-virginica
+6.1,2.6,5.6,1.4,Iris-virginica
+7.7,3.0,6.1,2.3,Iris-virginica
+6.3,3.4,5.6,2.4,Iris-virginica
+6.4,3.1,5.5,1.8,Iris-virginica
+6.0,3.0,4.8,1.8,Iris-virginica
+6.9,3.1,5.4,2.1,Iris-virginica
+6.7,3.1,5.6,2.4,Iris-virginica
+6.9,3.1,5.1,2.3,Iris-virginica
+5.8,2.7,5.1,1.9,Iris-virginica
+6.8,3.2,5.9,2.3,Iris-virginica
+6.7,3.3,5.7,2.5,Iris-virginica
+6.7,3.0,5.2,2.3,Iris-virginica
+6.3,2.5,5.0,1.9,Iris-virginica
+6.5,3.0,5.2,2.0,Iris-virginica
+6.2,3.4,5.4,2.3,Iris-virginica
+5.9,3.0,5.1,1.8,Iris-virginica
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/data/csv/test1.csv b/pandas/pandas/tests/io/data/csv/test1.csv
new file mode 100644
index 0000000000000..4bdb62943c4c8
--- /dev/null
+++ b/pandas/pandas/tests/io/data/csv/test1.csv
@@ -0,0 +1,8 @@
+index,A,B,C,D
+2000-01-03 00:00:00,0.980268513777,3.68573087906,-0.364216805298,-1.15973806169
+2000-01-04 00:00:00,1.04791624281,-0.0412318367011,-0.16181208307,0.212549316967
+2000-01-05 00:00:00,0.498580885705,0.731167677815,-0.537677223318,1.34627041952
+2000-01-06 00:00:00,1.12020151869,1.56762092543,0.00364077397681,0.67525259227
+2000-01-07 00:00:00,-0.487094399463,0.571454623474,-1.6116394093,0.103468562917
+2000-01-10 00:00:00,0.836648671666,0.246461918642,0.588542635376,1.0627820613
+2000-01-11 00:00:00,-0.157160753327,1.34030689438,1.19577795622,-1.09700699751
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/data/csv/test1.csv.bz2 b/pandas/pandas/tests/io/data/csv/test1.csv.bz2
new file mode 100644
index 0000000000000..f96f26a8e7419
Binary files /dev/null and b/pandas/pandas/tests/io/data/csv/test1.csv.bz2 differ
diff --git a/pandas/pandas/tests/io/data/csv/test1.csv.gz b/pandas/pandas/tests/io/data/csv/test1.csv.gz
new file mode 100644
index 0000000000000..1336db6e2af7e
Binary files /dev/null and b/pandas/pandas/tests/io/data/csv/test1.csv.gz differ
diff --git a/pandas/pandas/tests/io/data/csv/test_mmap.csv b/pandas/pandas/tests/io/data/csv/test_mmap.csv
new file mode 100644
index 0000000000000..cc2cd7c30349b
--- /dev/null
+++ b/pandas/pandas/tests/io/data/csv/test_mmap.csv
@@ -0,0 +1,5 @@
+a,b,c
+1,one,I
+2,two,II
+
+3,three,III
diff --git a/pandas/pandas/tests/io/data/csv/tips.csv b/pandas/pandas/tests/io/data/csv/tips.csv
new file mode 100644
index 0000000000000..856a65a69e647
--- /dev/null
+++ b/pandas/pandas/tests/io/data/csv/tips.csv
@@ -0,0 +1,245 @@
+total_bill,tip,sex,smoker,day,time,size
+16.99,1.01,Female,No,Sun,Dinner,2
+10.34,1.66,Male,No,Sun,Dinner,3
+21.01,3.5,Male,No,Sun,Dinner,3
+23.68,3.31,Male,No,Sun,Dinner,2
+24.59,3.61,Female,No,Sun,Dinner,4
+25.29,4.71,Male,No,Sun,Dinner,4
+8.77,2.0,Male,No,Sun,Dinner,2
+26.88,3.12,Male,No,Sun,Dinner,4
+15.04,1.96,Male,No,Sun,Dinner,2
+14.78,3.23,Male,No,Sun,Dinner,2
+10.27,1.71,Male,No,Sun,Dinner,2
+35.26,5.0,Female,No,Sun,Dinner,4
+15.42,1.57,Male,No,Sun,Dinner,2
+18.43,3.0,Male,No,Sun,Dinner,4
+14.83,3.02,Female,No,Sun,Dinner,2
+21.58,3.92,Male,No,Sun,Dinner,2
+10.33,1.67,Female,No,Sun,Dinner,3
+16.29,3.71,Male,No,Sun,Dinner,3
+16.97,3.5,Female,No,Sun,Dinner,3
+20.65,3.35,Male,No,Sat,Dinner,3
+17.92,4.08,Male,No,Sat,Dinner,2
+20.29,2.75,Female,No,Sat,Dinner,2
+15.77,2.23,Female,No,Sat,Dinner,2
+39.42,7.58,Male,No,Sat,Dinner,4
+19.82,3.18,Male,No,Sat,Dinner,2
+17.81,2.34,Male,No,Sat,Dinner,4
+13.37,2.0,Male,No,Sat,Dinner,2
+12.69,2.0,Male,No,Sat,Dinner,2
+21.7,4.3,Male,No,Sat,Dinner,2
+19.65,3.0,Female,No,Sat,Dinner,2
+9.55,1.45,Male,No,Sat,Dinner,2
+18.35,2.5,Male,No,Sat,Dinner,4
+15.06,3.0,Female,No,Sat,Dinner,2
+20.69,2.45,Female,No,Sat,Dinner,4
+17.78,3.27,Male,No,Sat,Dinner,2
+24.06,3.6,Male,No,Sat,Dinner,3
+16.31,2.0,Male,No,Sat,Dinner,3
+16.93,3.07,Female,No,Sat,Dinner,3
+18.69,2.31,Male,No,Sat,Dinner,3
+31.27,5.0,Male,No,Sat,Dinner,3
+16.04,2.24,Male,No,Sat,Dinner,3
+17.46,2.54,Male,No,Sun,Dinner,2
+13.94,3.06,Male,No,Sun,Dinner,2
+9.68,1.32,Male,No,Sun,Dinner,2
+30.4,5.6,Male,No,Sun,Dinner,4
+18.29,3.0,Male,No,Sun,Dinner,2
+22.23,5.0,Male,No,Sun,Dinner,2
+32.4,6.0,Male,No,Sun,Dinner,4
+28.55,2.05,Male,No,Sun,Dinner,3
+18.04,3.0,Male,No,Sun,Dinner,2
+12.54,2.5,Male,No,Sun,Dinner,2
+10.29,2.6,Female,No,Sun,Dinner,2
+34.81,5.2,Female,No,Sun,Dinner,4
+9.94,1.56,Male,No,Sun,Dinner,2
+25.56,4.34,Male,No,Sun,Dinner,4
+19.49,3.51,Male,No,Sun,Dinner,2
+38.01,3.0,Male,Yes,Sat,Dinner,4
+26.41,1.5,Female,No,Sat,Dinner,2
+11.24,1.76,Male,Yes,Sat,Dinner,2
+48.27,6.73,Male,No,Sat,Dinner,4
+20.29,3.21,Male,Yes,Sat,Dinner,2
+13.81,2.0,Male,Yes,Sat,Dinner,2
+11.02,1.98,Male,Yes,Sat,Dinner,2
+18.29,3.76,Male,Yes,Sat,Dinner,4
+17.59,2.64,Male,No,Sat,Dinner,3
+20.08,3.15,Male,No,Sat,Dinner,3
+16.45,2.47,Female,No,Sat,Dinner,2
+3.07,1.0,Female,Yes,Sat,Dinner,1
+20.23,2.01,Male,No,Sat,Dinner,2
+15.01,2.09,Male,Yes,Sat,Dinner,2
+12.02,1.97,Male,No,Sat,Dinner,2
+17.07,3.0,Female,No,Sat,Dinner,3
+26.86,3.14,Female,Yes,Sat,Dinner,2
+25.28,5.0,Female,Yes,Sat,Dinner,2
+14.73,2.2,Female,No,Sat,Dinner,2
+10.51,1.25,Male,No,Sat,Dinner,2
+17.92,3.08,Male,Yes,Sat,Dinner,2
+27.2,4.0,Male,No,Thur,Lunch,4
+22.76,3.0,Male,No,Thur,Lunch,2
+17.29,2.71,Male,No,Thur,Lunch,2
+19.44,3.0,Male,Yes,Thur,Lunch,2
+16.66,3.4,Male,No,Thur,Lunch,2
+10.07,1.83,Female,No,Thur,Lunch,1
+32.68,5.0,Male,Yes,Thur,Lunch,2
+15.98,2.03,Male,No,Thur,Lunch,2
+34.83,5.17,Female,No,Thur,Lunch,4
+13.03,2.0,Male,No,Thur,Lunch,2
+18.28,4.0,Male,No,Thur,Lunch,2
+24.71,5.85,Male,No,Thur,Lunch,2
+21.16,3.0,Male,No,Thur,Lunch,2
+28.97,3.0,Male,Yes,Fri,Dinner,2
+22.49,3.5,Male,No,Fri,Dinner,2
+5.75,1.0,Female,Yes,Fri,Dinner,2
+16.32,4.3,Female,Yes,Fri,Dinner,2
+22.75,3.25,Female,No,Fri,Dinner,2
+40.17,4.73,Male,Yes,Fri,Dinner,4
+27.28,4.0,Male,Yes,Fri,Dinner,2
+12.03,1.5,Male,Yes,Fri,Dinner,2
+21.01,3.0,Male,Yes,Fri,Dinner,2
+12.46,1.5,Male,No,Fri,Dinner,2
+11.35,2.5,Female,Yes,Fri,Dinner,2
+15.38,3.0,Female,Yes,Fri,Dinner,2
+44.3,2.5,Female,Yes,Sat,Dinner,3
+22.42,3.48,Female,Yes,Sat,Dinner,2
+20.92,4.08,Female,No,Sat,Dinner,2
+15.36,1.64,Male,Yes,Sat,Dinner,2
+20.49,4.06,Male,Yes,Sat,Dinner,2
+25.21,4.29,Male,Yes,Sat,Dinner,2
+18.24,3.76,Male,No,Sat,Dinner,2
+14.31,4.0,Female,Yes,Sat,Dinner,2
+14.0,3.0,Male,No,Sat,Dinner,2
+7.25,1.0,Female,No,Sat,Dinner,1
+38.07,4.0,Male,No,Sun,Dinner,3
+23.95,2.55,Male,No,Sun,Dinner,2
+25.71,4.0,Female,No,Sun,Dinner,3
+17.31,3.5,Female,No,Sun,Dinner,2
+29.93,5.07,Male,No,Sun,Dinner,4
+10.65,1.5,Female,No,Thur,Lunch,2
+12.43,1.8,Female,No,Thur,Lunch,2
+24.08,2.92,Female,No,Thur,Lunch,4
+11.69,2.31,Male,No,Thur,Lunch,2
+13.42,1.68,Female,No,Thur,Lunch,2
+14.26,2.5,Male,No,Thur,Lunch,2
+15.95,2.0,Male,No,Thur,Lunch,2
+12.48,2.52,Female,No,Thur,Lunch,2
+29.8,4.2,Female,No,Thur,Lunch,6
+8.52,1.48,Male,No,Thur,Lunch,2
+14.52,2.0,Female,No,Thur,Lunch,2
+11.38,2.0,Female,No,Thur,Lunch,2
+22.82,2.18,Male,No,Thur,Lunch,3
+19.08,1.5,Male,No,Thur,Lunch,2
+20.27,2.83,Female,No,Thur,Lunch,2
+11.17,1.5,Female,No,Thur,Lunch,2
+12.26,2.0,Female,No,Thur,Lunch,2
+18.26,3.25,Female,No,Thur,Lunch,2
+8.51,1.25,Female,No,Thur,Lunch,2
+10.33,2.0,Female,No,Thur,Lunch,2
+14.15,2.0,Female,No,Thur,Lunch,2
+16.0,2.0,Male,Yes,Thur,Lunch,2
+13.16,2.75,Female,No,Thur,Lunch,2
+17.47,3.5,Female,No,Thur,Lunch,2
+34.3,6.7,Male,No,Thur,Lunch,6
+41.19,5.0,Male,No,Thur,Lunch,5
+27.05,5.0,Female,No,Thur,Lunch,6
+16.43,2.3,Female,No,Thur,Lunch,2
+8.35,1.5,Female,No,Thur,Lunch,2
+18.64,1.36,Female,No,Thur,Lunch,3
+11.87,1.63,Female,No,Thur,Lunch,2
+9.78,1.73,Male,No,Thur,Lunch,2
+7.51,2.0,Male,No,Thur,Lunch,2
+14.07,2.5,Male,No,Sun,Dinner,2
+13.13,2.0,Male,No,Sun,Dinner,2
+17.26,2.74,Male,No,Sun,Dinner,3
+24.55,2.0,Male,No,Sun,Dinner,4
+19.77,2.0,Male,No,Sun,Dinner,4
+29.85,5.14,Female,No,Sun,Dinner,5
+48.17,5.0,Male,No,Sun,Dinner,6
+25.0,3.75,Female,No,Sun,Dinner,4
+13.39,2.61,Female,No,Sun,Dinner,2
+16.49,2.0,Male,No,Sun,Dinner,4
+21.5,3.5,Male,No,Sun,Dinner,4
+12.66,2.5,Male,No,Sun,Dinner,2
+16.21,2.0,Female,No,Sun,Dinner,3
+13.81,2.0,Male,No,Sun,Dinner,2
+17.51,3.0,Female,Yes,Sun,Dinner,2
+24.52,3.48,Male,No,Sun,Dinner,3
+20.76,2.24,Male,No,Sun,Dinner,2
+31.71,4.5,Male,No,Sun,Dinner,4
+10.59,1.61,Female,Yes,Sat,Dinner,2
+10.63,2.0,Female,Yes,Sat,Dinner,2
+50.81,10.0,Male,Yes,Sat,Dinner,3
+15.81,3.16,Male,Yes,Sat,Dinner,2
+7.25,5.15,Male,Yes,Sun,Dinner,2
+31.85,3.18,Male,Yes,Sun,Dinner,2
+16.82,4.0,Male,Yes,Sun,Dinner,2
+32.9,3.11,Male,Yes,Sun,Dinner,2
+17.89,2.0,Male,Yes,Sun,Dinner,2
+14.48,2.0,Male,Yes,Sun,Dinner,2
+9.6,4.0,Female,Yes,Sun,Dinner,2
+34.63,3.55,Male,Yes,Sun,Dinner,2
+34.65,3.68,Male,Yes,Sun,Dinner,4
+23.33,5.65,Male,Yes,Sun,Dinner,2
+45.35,3.5,Male,Yes,Sun,Dinner,3
+23.17,6.5,Male,Yes,Sun,Dinner,4
+40.55,3.0,Male,Yes,Sun,Dinner,2
+20.69,5.0,Male,No,Sun,Dinner,5
+20.9,3.5,Female,Yes,Sun,Dinner,3
+30.46,2.0,Male,Yes,Sun,Dinner,5
+18.15,3.5,Female,Yes,Sun,Dinner,3
+23.1,4.0,Male,Yes,Sun,Dinner,3
+15.69,1.5,Male,Yes,Sun,Dinner,2
+19.81,4.19,Female,Yes,Thur,Lunch,2
+28.44,2.56,Male,Yes,Thur,Lunch,2
+15.48,2.02,Male,Yes,Thur,Lunch,2
+16.58,4.0,Male,Yes,Thur,Lunch,2
+7.56,1.44,Male,No,Thur,Lunch,2
+10.34,2.0,Male,Yes,Thur,Lunch,2
+43.11,5.0,Female,Yes,Thur,Lunch,4
+13.0,2.0,Female,Yes,Thur,Lunch,2
+13.51,2.0,Male,Yes,Thur,Lunch,2
+18.71,4.0,Male,Yes,Thur,Lunch,3
+12.74,2.01,Female,Yes,Thur,Lunch,2
+13.0,2.0,Female,Yes,Thur,Lunch,2
+16.4,2.5,Female,Yes,Thur,Lunch,2
+20.53,4.0,Male,Yes,Thur,Lunch,4
+16.47,3.23,Female,Yes,Thur,Lunch,3
+26.59,3.41,Male,Yes,Sat,Dinner,3
+38.73,3.0,Male,Yes,Sat,Dinner,4
+24.27,2.03,Male,Yes,Sat,Dinner,2
+12.76,2.23,Female,Yes,Sat,Dinner,2
+30.06,2.0,Male,Yes,Sat,Dinner,3
+25.89,5.16,Male,Yes,Sat,Dinner,4
+48.33,9.0,Male,No,Sat,Dinner,4
+13.27,2.5,Female,Yes,Sat,Dinner,2
+28.17,6.5,Female,Yes,Sat,Dinner,3
+12.9,1.1,Female,Yes,Sat,Dinner,2
+28.15,3.0,Male,Yes,Sat,Dinner,5
+11.59,1.5,Male,Yes,Sat,Dinner,2
+7.74,1.44,Male,Yes,Sat,Dinner,2
+30.14,3.09,Female,Yes,Sat,Dinner,4
+12.16,2.2,Male,Yes,Fri,Lunch,2
+13.42,3.48,Female,Yes,Fri,Lunch,2
+8.58,1.92,Male,Yes,Fri,Lunch,1
+15.98,3.0,Female,No,Fri,Lunch,3
+13.42,1.58,Male,Yes,Fri,Lunch,2
+16.27,2.5,Female,Yes,Fri,Lunch,2
+10.09,2.0,Female,Yes,Fri,Lunch,2
+20.45,3.0,Male,No,Sat,Dinner,4
+13.28,2.72,Male,No,Sat,Dinner,2
+22.12,2.88,Female,Yes,Sat,Dinner,2
+24.01,2.0,Male,Yes,Sat,Dinner,4
+15.69,3.0,Male,Yes,Sat,Dinner,3
+11.61,3.39,Male,No,Sat,Dinner,2
+10.77,1.47,Male,No,Sat,Dinner,2
+15.53,3.0,Male,Yes,Sat,Dinner,2
+10.07,1.25,Male,No,Sat,Dinner,2
+12.6,1.0,Male,Yes,Sat,Dinner,2
+32.83,1.17,Male,Yes,Sat,Dinner,2
+35.83,4.67,Female,No,Sat,Dinner,3
+29.03,5.92,Male,No,Sat,Dinner,3
+27.18,2.0,Female,Yes,Sat,Dinner,2
+22.67,2.0,Male,Yes,Sat,Dinner,2
+17.82,1.75,Male,No,Sat,Dinner,2
+18.78,3.0,Female,No,Thur,Dinner,2
diff --git a/pandas/pandas/tests/io/data/csv/tips.csv.bz2 b/pandas/pandas/tests/io/data/csv/tips.csv.bz2
new file mode 100644
index 0000000000000..1452896b05e9d
Binary files /dev/null and b/pandas/pandas/tests/io/data/csv/tips.csv.bz2 differ
diff --git a/pandas/pandas/tests/io/data/csv/tips.csv.gz b/pandas/pandas/tests/io/data/csv/tips.csv.gz
new file mode 100644
index 0000000000000..3a131068b2a38
Binary files /dev/null and b/pandas/pandas/tests/io/data/csv/tips.csv.gz differ
diff --git a/pandas/pandas/tests/io/data/excel/blank.ods b/pandas/pandas/tests/io/data/excel/blank.ods
new file mode 100644
index 0000000000000..7ded3c3c1d688
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/blank.xls b/pandas/pandas/tests/io/data/excel/blank.xls
new file mode 100644
index 0000000000000..952c76f045e8a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/blank.xlsb b/pandas/pandas/tests/io/data/excel/blank.xlsb
new file mode 100644
index 0000000000000..d72fd68ab3dbf
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/blank.xlsm b/pandas/pandas/tests/io/data/excel/blank.xlsm
new file mode 100644
index 0000000000000..c3c2074276d12
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/blank.xlsx b/pandas/pandas/tests/io/data/excel/blank.xlsx
new file mode 100644
index 0000000000000..73f6ba6d29af3
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/blank_with_header.ods b/pandas/pandas/tests/io/data/excel/blank_with_header.ods
new file mode 100644
index 0000000000000..0a2e696267fda
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank_with_header.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/blank_with_header.xls b/pandas/pandas/tests/io/data/excel/blank_with_header.xls
new file mode 100644
index 0000000000000..a2e6350808147
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank_with_header.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/blank_with_header.xlsb b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsb
new file mode 100644
index 0000000000000..3c241513d221a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/blank_with_header.xlsm b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsm
new file mode 100644
index 0000000000000..5b2ed0055b243
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/blank_with_header.xlsx b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsx
new file mode 100644
index 0000000000000..7b30edb13be2c
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/blank_with_header.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/chartsheet.xls b/pandas/pandas/tests/io/data/excel/chartsheet.xls
new file mode 100644
index 0000000000000..7d027400fbd52
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/chartsheet.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/chartsheet.xlsb b/pandas/pandas/tests/io/data/excel/chartsheet.xlsb
new file mode 100644
index 0000000000000..805087280f851
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/chartsheet.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/chartsheet.xlsm b/pandas/pandas/tests/io/data/excel/chartsheet.xlsm
new file mode 100644
index 0000000000000..aadb48d6f4824
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/chartsheet.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/chartsheet.xlsx b/pandas/pandas/tests/io/data/excel/chartsheet.xlsx
new file mode 100644
index 0000000000000..c8d5e7afb3d07
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/chartsheet.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/df_empty.xlsx b/pandas/pandas/tests/io/data/excel/df_empty.xlsx
new file mode 100644
index 0000000000000..d65a92b10e293
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_empty.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/df_equals.xlsx b/pandas/pandas/tests/io/data/excel/df_equals.xlsx
new file mode 100644
index 0000000000000..d65a92b10e293
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_equals.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/df_header_oob.xlsx b/pandas/pandas/tests/io/data/excel/df_header_oob.xlsx
new file mode 100644
index 0000000000000..1e26091cd2ace
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_header_oob.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.ods b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.ods
new file mode 100644
index 0000000000000..66558c16319fc
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xls b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xls
new file mode 100644
index 0000000000000..472ad75901286
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsb b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsb
new file mode 100755
index 0000000000000..5052102c6655d
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsm b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsm
new file mode 100644
index 0000000000000..51edc7f94f9d8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsx b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsx
new file mode 100644
index 0000000000000..ec4e49add4233
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/df_mangle_dup_col_dtypes.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/dimension_large.xlsx b/pandas/pandas/tests/io/data/excel/dimension_large.xlsx
new file mode 100644
index 0000000000000..d57abdf2fbbae
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/dimension_large.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/dimension_missing.xlsx b/pandas/pandas/tests/io/data/excel/dimension_missing.xlsx
new file mode 100644
index 0000000000000..9274896689a72
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/dimension_missing.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/dimension_small.xlsx b/pandas/pandas/tests/io/data/excel/dimension_small.xlsx
new file mode 100644
index 0000000000000..78ce4723ebef4
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/dimension_small.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/empty_trailing_rows.xlsx b/pandas/pandas/tests/io/data/excel/empty_trailing_rows.xlsx
new file mode 100644
index 0000000000000..920b03915a3c8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/empty_trailing_rows.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/empty_with_blank_row.xlsx b/pandas/pandas/tests/io/data/excel/empty_with_blank_row.xlsx
new file mode 100644
index 0000000000000..fe3bcfcc269d7
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/empty_with_blank_row.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/gh-35802.ods b/pandas/pandas/tests/io/data/excel/gh-35802.ods
new file mode 100755
index 0000000000000..f3ad061f1d995
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/gh-35802.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/gh-36122.ods b/pandas/pandas/tests/io/data/excel/gh-36122.ods
new file mode 100755
index 0000000000000..3dfdaf976da45
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/gh-36122.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/ints_spelled_with_decimals.xlsx b/pandas/pandas/tests/io/data/excel/ints_spelled_with_decimals.xlsx
new file mode 100644
index 0000000000000..a667be86283b8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/ints_spelled_with_decimals.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/invalid_value_type.ods b/pandas/pandas/tests/io/data/excel/invalid_value_type.ods
new file mode 100644
index 0000000000000..75a7a40b25d79
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/invalid_value_type.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/multiindex_no_index_names.xlsx b/pandas/pandas/tests/io/data/excel/multiindex_no_index_names.xlsx
new file mode 100755
index 0000000000000..3913ffce5befb
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/multiindex_no_index_names.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/one_col_blank_line.ods b/pandas/pandas/tests/io/data/excel/one_col_blank_line.ods
new file mode 100644
index 0000000000000..df5fbcfaa0357
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/one_col_blank_line.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/one_col_blank_line.xls b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xls
new file mode 100644
index 0000000000000..dcf2ebecded61
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsb b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsb
new file mode 100644
index 0000000000000..9257d016c762a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsm b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsm
new file mode 100644
index 0000000000000..c249901ecc10e
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsx b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsx
new file mode 100644
index 0000000000000..2538e406d2e77
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/one_col_blank_line.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test1.ods b/pandas/pandas/tests/io/data/excel/test1.ods
new file mode 100644
index 0000000000000..5dc0e83456264
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test1.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test1.xls b/pandas/pandas/tests/io/data/excel/test1.xls
new file mode 100644
index 0000000000000..faf5dc84700c9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test1.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test1.xlsb b/pandas/pandas/tests/io/data/excel/test1.xlsb
new file mode 100644
index 0000000000000..d0b8a1f2735bd
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test1.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test1.xlsm b/pandas/pandas/tests/io/data/excel/test1.xlsm
new file mode 100644
index 0000000000000..28f4f27e4e1b1
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test1.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test1.xlsx b/pandas/pandas/tests/io/data/excel/test1.xlsx
new file mode 100644
index 0000000000000..862574e05a114
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test1.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test2.ods b/pandas/pandas/tests/io/data/excel/test2.ods
new file mode 100644
index 0000000000000..2a90db839026b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test2.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test2.xls b/pandas/pandas/tests/io/data/excel/test2.xls
new file mode 100644
index 0000000000000..dadeb7c2453af
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test2.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test2.xlsb b/pandas/pandas/tests/io/data/excel/test2.xlsb
new file mode 100644
index 0000000000000..e19a0f1e067c8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test2.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test2.xlsm b/pandas/pandas/tests/io/data/excel/test2.xlsm
new file mode 100644
index 0000000000000..31cfba7ede082
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test2.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test2.xlsx b/pandas/pandas/tests/io/data/excel/test2.xlsx
new file mode 100644
index 0000000000000..94dd951e0bb84
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test2.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test3.ods b/pandas/pandas/tests/io/data/excel/test3.ods
new file mode 100644
index 0000000000000..dc78781caa6e9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test3.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test3.xls b/pandas/pandas/tests/io/data/excel/test3.xls
new file mode 100644
index 0000000000000..f73943d677951
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test3.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test3.xlsb b/pandas/pandas/tests/io/data/excel/test3.xlsb
new file mode 100644
index 0000000000000..617d27630e8a0
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test3.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test3.xlsm b/pandas/pandas/tests/io/data/excel/test3.xlsm
new file mode 100644
index 0000000000000..54b7ef456a9ea
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test3.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test3.xlsx b/pandas/pandas/tests/io/data/excel/test3.xlsx
new file mode 100644
index 0000000000000..c16755c25fabd
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test3.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test4.ods b/pandas/pandas/tests/io/data/excel/test4.ods
new file mode 100644
index 0000000000000..c73a20d8b0562
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test4.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test4.xls b/pandas/pandas/tests/io/data/excel/test4.xls
new file mode 100644
index 0000000000000..10a6ab1cca6a4
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test4.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test4.xlsb b/pandas/pandas/tests/io/data/excel/test4.xlsb
new file mode 100644
index 0000000000000..2e5bb229939be
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test4.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test4.xlsm b/pandas/pandas/tests/io/data/excel/test4.xlsm
new file mode 100644
index 0000000000000..a5a2ff8caadd6
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test4.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test4.xlsx b/pandas/pandas/tests/io/data/excel/test4.xlsx
new file mode 100644
index 0000000000000..6fb62272caf09
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test4.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test5.ods b/pandas/pandas/tests/io/data/excel/test5.ods
new file mode 100644
index 0000000000000..5872e2624d033
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test5.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test5.xls b/pandas/pandas/tests/io/data/excel/test5.xls
new file mode 100644
index 0000000000000..f484f54a60c61
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test5.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test5.xlsb b/pandas/pandas/tests/io/data/excel/test5.xlsb
new file mode 100644
index 0000000000000..022ebef25aee2
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test5.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test5.xlsm b/pandas/pandas/tests/io/data/excel/test5.xlsm
new file mode 100644
index 0000000000000..7b8d489f24321
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test5.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test5.xlsx b/pandas/pandas/tests/io/data/excel/test5.xlsx
new file mode 100644
index 0000000000000..8660913c74812
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test5.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test6.xls b/pandas/pandas/tests/io/data/excel/test6.xls
new file mode 100644
index 0000000000000..e43a1a67510d8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test6.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_cell_annotation.ods b/pandas/pandas/tests/io/data/excel/test_cell_annotation.ods
new file mode 100644
index 0000000000000..f00a88fe681ef
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_cell_annotation.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_converters.ods b/pandas/pandas/tests/io/data/excel/test_converters.ods
new file mode 100644
index 0000000000000..0216fb16311d8
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_converters.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_converters.xls b/pandas/pandas/tests/io/data/excel/test_converters.xls
new file mode 100644
index 0000000000000..c0aa9d903adad
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_converters.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_converters.xlsb b/pandas/pandas/tests/io/data/excel/test_converters.xlsb
new file mode 100644
index 0000000000000..c39c33d8dd94f
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_converters.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_converters.xlsm b/pandas/pandas/tests/io/data/excel/test_converters.xlsm
new file mode 100644
index 0000000000000..eaf0b1d0219c5
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_converters.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_converters.xlsx b/pandas/pandas/tests/io/data/excel/test_converters.xlsx
new file mode 100644
index 0000000000000..e21bc5fbf9ee2
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_converters.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_datetime_mi.ods b/pandas/pandas/tests/io/data/excel/test_datetime_mi.ods
new file mode 100644
index 0000000000000..c37c35060c650
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_datetime_mi.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_datetime_mi.xls b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xls
new file mode 100644
index 0000000000000..aeade05855919
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsb b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsb
new file mode 100644
index 0000000000000..0984c020a4c54
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsm b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsm
new file mode 100644
index 0000000000000..55fb88912afb9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsx b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsx
new file mode 100644
index 0000000000000..0ffee0a8b79a3
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_datetime_mi.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_decimal.ods b/pandas/pandas/tests/io/data/excel/test_decimal.ods
new file mode 100644
index 0000000000000..308a851809dde
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_decimal.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_decimal.xls b/pandas/pandas/tests/io/data/excel/test_decimal.xls
new file mode 100644
index 0000000000000..ce34667873cb5
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_decimal.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_decimal.xlsb b/pandas/pandas/tests/io/data/excel/test_decimal.xlsb
new file mode 100644
index 0000000000000..addfd1480a190
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_decimal.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_decimal.xlsm b/pandas/pandas/tests/io/data/excel/test_decimal.xlsm
new file mode 100644
index 0000000000000..7dd6b1e7da036
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_decimal.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_decimal.xlsx b/pandas/pandas/tests/io/data/excel/test_decimal.xlsx
new file mode 100644
index 0000000000000..0cedf3899a566
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_decimal.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_index_name_pre17.ods b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.ods
new file mode 100644
index 0000000000000..56638c983d944
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xls b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xls
new file mode 100644
index 0000000000000..2ab13105e7925
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsb b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsb
new file mode 100644
index 0000000000000..5251b8f3b3194
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsm b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsm
new file mode 100644
index 0000000000000..33c0d7949531c
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsx b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsx
new file mode 100644
index 0000000000000..ce66c40cda141
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_index_name_pre17.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_multisheet.ods b/pandas/pandas/tests/io/data/excel/test_multisheet.ods
new file mode 100644
index 0000000000000..39058e67b4d5b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_multisheet.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_multisheet.xls b/pandas/pandas/tests/io/data/excel/test_multisheet.xls
new file mode 100644
index 0000000000000..bb6cb35e34a57
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_multisheet.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_multisheet.xlsb b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsb
new file mode 100644
index 0000000000000..39b15568a7121
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_multisheet.xlsm b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsm
new file mode 100644
index 0000000000000..eaee98a801cf0
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_multisheet.xlsx b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsx
new file mode 100644
index 0000000000000..c22771232961f
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_multisheet.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_newlines.ods b/pandas/pandas/tests/io/data/excel/test_newlines.ods
new file mode 100644
index 0000000000000..262529800351c
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_newlines.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_none_type.xlsx b/pandas/pandas/tests/io/data/excel/test_none_type.xlsx
new file mode 100644
index 0000000000000..38aaf72ddfc8f
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_none_type.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_spaces.ods b/pandas/pandas/tests/io/data/excel/test_spaces.ods
new file mode 100644
index 0000000000000..375e839c8c221
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_spaces.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_spaces.xls b/pandas/pandas/tests/io/data/excel/test_spaces.xls
new file mode 100644
index 0000000000000..316db172360d0
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_spaces.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_spaces.xlsb b/pandas/pandas/tests/io/data/excel/test_spaces.xlsb
new file mode 100644
index 0000000000000..e38b6c2d8f170
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_spaces.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_spaces.xlsm b/pandas/pandas/tests/io/data/excel/test_spaces.xlsm
new file mode 100644
index 0000000000000..a41ebe5bb0e65
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_spaces.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_spaces.xlsx b/pandas/pandas/tests/io/data/excel/test_spaces.xlsx
new file mode 100644
index 0000000000000..9071543c4739b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_spaces.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_squeeze.ods b/pandas/pandas/tests/io/data/excel/test_squeeze.ods
new file mode 100644
index 0000000000000..10ccf0da2693e
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_squeeze.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_squeeze.xls b/pandas/pandas/tests/io/data/excel/test_squeeze.xls
new file mode 100644
index 0000000000000..2524b975bdce6
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_squeeze.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_squeeze.xlsb b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsb
new file mode 100644
index 0000000000000..6aadd727e957b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_squeeze.xlsm b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsm
new file mode 100644
index 0000000000000..fb19b5a40efb7
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_squeeze.xlsx b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsx
new file mode 100644
index 0000000000000..84141fc80ef11
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_squeeze.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_types.ods b/pandas/pandas/tests/io/data/excel/test_types.ods
new file mode 100644
index 0000000000000..c9a82bfff810b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_types.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/test_types.xls b/pandas/pandas/tests/io/data/excel/test_types.xls
new file mode 100644
index 0000000000000..2d387603a8307
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_types.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/test_types.xlsb b/pandas/pandas/tests/io/data/excel/test_types.xlsb
new file mode 100644
index 0000000000000..e7403aa288263
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_types.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/test_types.xlsm b/pandas/pandas/tests/io/data/excel/test_types.xlsm
new file mode 100644
index 0000000000000..4c8c10e40effe
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_types.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/test_types.xlsx b/pandas/pandas/tests/io/data/excel/test_types.xlsx
new file mode 100644
index 0000000000000..561bb2c5d6714
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_types.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/test_unempty_cells.ods b/pandas/pandas/tests/io/data/excel/test_unempty_cells.ods
new file mode 100644
index 0000000000000..52458753bae06
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/test_unempty_cells.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/testdateoverflow.ods b/pandas/pandas/tests/io/data/excel/testdateoverflow.ods
new file mode 100644
index 0000000000000..bb05267865303
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdateoverflow.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/testdateoverflow.xls b/pandas/pandas/tests/io/data/excel/testdateoverflow.xls
new file mode 100644
index 0000000000000..d63942f1b5f64
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdateoverflow.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsb b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsb
new file mode 100644
index 0000000000000..3d279396924b9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsm b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsm
new file mode 100644
index 0000000000000..0207549692e6a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsx b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsx
new file mode 100644
index 0000000000000..15f0e8825339e
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdateoverflow.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/testdtype.ods b/pandas/pandas/tests/io/data/excel/testdtype.ods
new file mode 100644
index 0000000000000..91145f807c9d9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdtype.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/testdtype.xls b/pandas/pandas/tests/io/data/excel/testdtype.xls
new file mode 100644
index 0000000000000..f63357524324f
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdtype.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/testdtype.xlsb b/pandas/pandas/tests/io/data/excel/testdtype.xlsb
new file mode 100644
index 0000000000000..1c1d45f0d783b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdtype.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/testdtype.xlsm b/pandas/pandas/tests/io/data/excel/testdtype.xlsm
new file mode 100644
index 0000000000000..20e658288d5ac
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdtype.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/testdtype.xlsx b/pandas/pandas/tests/io/data/excel/testdtype.xlsx
new file mode 100644
index 0000000000000..7c65263c373a3
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testdtype.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/testmultiindex.ods b/pandas/pandas/tests/io/data/excel/testmultiindex.ods
new file mode 100644
index 0000000000000..dca8d70abdc24
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testmultiindex.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/testmultiindex.xls b/pandas/pandas/tests/io/data/excel/testmultiindex.xls
new file mode 100644
index 0000000000000..c91698be29b13
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testmultiindex.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/testmultiindex.xlsb b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsb
new file mode 100644
index 0000000000000..a693e0c66afc2
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/testmultiindex.xlsm b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsm
new file mode 100644
index 0000000000000..5a2a4ea35f0d9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/testmultiindex.xlsx b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsx
new file mode 100644
index 0000000000000..a6174445bb83a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testmultiindex.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/testskiprows.ods b/pandas/pandas/tests/io/data/excel/testskiprows.ods
new file mode 100644
index 0000000000000..443602a2c3f98
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testskiprows.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/testskiprows.xls b/pandas/pandas/tests/io/data/excel/testskiprows.xls
new file mode 100644
index 0000000000000..21ccd30ec62da
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testskiprows.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/testskiprows.xlsb b/pandas/pandas/tests/io/data/excel/testskiprows.xlsb
new file mode 100644
index 0000000000000..a5ff4ed22e70c
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testskiprows.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/testskiprows.xlsm b/pandas/pandas/tests/io/data/excel/testskiprows.xlsm
new file mode 100644
index 0000000000000..f5889ded4637a
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testskiprows.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/testskiprows.xlsx b/pandas/pandas/tests/io/data/excel/testskiprows.xlsx
new file mode 100644
index 0000000000000..2d7ce943a7214
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/testskiprows.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1900.ods b/pandas/pandas/tests/io/data/excel/times_1900.ods
new file mode 100644
index 0000000000000..79e031c721ea3
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1900.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1900.xls b/pandas/pandas/tests/io/data/excel/times_1900.xls
new file mode 100644
index 0000000000000..e9a62b2c25da9
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1900.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1900.xlsb b/pandas/pandas/tests/io/data/excel/times_1900.xlsb
new file mode 100644
index 0000000000000..ceb7bccb0c66e
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1900.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1900.xlsm b/pandas/pandas/tests/io/data/excel/times_1900.xlsm
new file mode 100644
index 0000000000000..1ffdbe223453b
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1900.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1900.xlsx b/pandas/pandas/tests/io/data/excel/times_1900.xlsx
new file mode 100644
index 0000000000000..3702289b256fd
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1900.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1904.ods b/pandas/pandas/tests/io/data/excel/times_1904.ods
new file mode 100644
index 0000000000000..b47a949d3b715
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1904.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1904.xls b/pandas/pandas/tests/io/data/excel/times_1904.xls
new file mode 100644
index 0000000000000..ac70787c358a5
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1904.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1904.xlsb b/pandas/pandas/tests/io/data/excel/times_1904.xlsb
new file mode 100644
index 0000000000000..e426dc959da49
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1904.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1904.xlsm b/pandas/pandas/tests/io/data/excel/times_1904.xlsm
new file mode 100644
index 0000000000000..e884eca1e7c74
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1904.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/times_1904.xlsx b/pandas/pandas/tests/io/data/excel/times_1904.xlsx
new file mode 100644
index 0000000000000..1a13468e59d1c
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/times_1904.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/trailing_blanks.ods b/pandas/pandas/tests/io/data/excel/trailing_blanks.ods
new file mode 100644
index 0000000000000..a56fbfe452387
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/trailing_blanks.ods differ
diff --git a/pandas/pandas/tests/io/data/excel/trailing_blanks.xls b/pandas/pandas/tests/io/data/excel/trailing_blanks.xls
new file mode 100644
index 0000000000000..32aeb3fe36b05
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/trailing_blanks.xls differ
diff --git a/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsb b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsb
new file mode 100644
index 0000000000000..b40b390a48f38
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsb differ
diff --git a/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsm b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsm
new file mode 100644
index 0000000000000..9f8ca5370ef81
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsm differ
diff --git a/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsx b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsx
new file mode 100644
index 0000000000000..8f1a739be9078
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/trailing_blanks.xlsx differ
diff --git a/pandas/pandas/tests/io/data/excel/writertable.odt b/pandas/pandas/tests/io/data/excel/writertable.odt
new file mode 100644
index 0000000000000..113bd651e8cd0
Binary files /dev/null and b/pandas/pandas/tests/io/data/excel/writertable.odt differ
diff --git a/pandas/pandas/tests/io/data/feather/feather-0_3_1.feather b/pandas/pandas/tests/io/data/feather/feather-0_3_1.feather
new file mode 100644
index 0000000000000..5a2c7b3dcc684
Binary files /dev/null and b/pandas/pandas/tests/io/data/feather/feather-0_3_1.feather differ
diff --git a/pandas/pandas/tests/io/data/fixed_width/fixed_width_format.txt b/pandas/pandas/tests/io/data/fixed_width/fixed_width_format.txt
new file mode 100644
index 0000000000000..bb487d8de7ef9
--- /dev/null
+++ b/pandas/pandas/tests/io/data/fixed_width/fixed_width_format.txt
@@ -0,0 +1,3 @@
+A   B   C
+1   2   3
+4   5   6
diff --git a/pandas/pandas/tests/io/data/html/banklist.html b/pandas/pandas/tests/io/data/html/banklist.html
new file mode 100644
index 0000000000000..a0562989ad3a4
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html/banklist.html
@@ -0,0 +1,4886 @@
+<!DOCTYPE html><!-- HTML5 -->
+<html lang="en-US">
+<!-- Content language is American English. -->
+<head>
+<title>FDIC: Failed Bank List</title>
+<!-- Meta Tags -->
+<meta charset="UTF-8">
+<!-- Unicode character encoding -->
+<meta http-equiv="X-UA-Compatible" content="IE=edge">
+<!-- Turns off IE Compatibility Mode -->
+<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1">
+<!-- Makes it so phones don't auto zoom out. -->
+<meta name="author" content="DRR">
+<meta http-equiv="keywords" name="keywords" content="banks, financial institutions, failed, failure, closing, deposits, depositors, banking services, assuming institution, acquiring institution, claims">
+<!-- CSS -->
+<link rel="stylesheet" type="text/css" href="/responsive/css/responsive.css">
+<link rel="stylesheet" type="text/css" href="banklist.css">
+</head>
+<body>
+
+<!-- START of Header -->
+<script type="text/javascript" src="/responsive/header/js/header.js"></script>
+<link rel="stylesheet" type="text/css" href="/responsive/header/css/header.css" />
+<!-- googleac.html includes Autocomplete functionality -->
+<!-- Autocomplete files -->
+<link rel="stylesheet" type="text/css" href="/responsive/header/css/jquery.autocomplete.css" />
+<script type="text/javascript" src="/responsive/js/jquery-1.4.1.min.js"></script>
+<script type="text/javascript" src="/responsive/header/js/jquery.autocomplete-1.4.2.js"></script>
+<script type="text/javascript">
+function findValue(li) {
+	if( li == null ) return alert("No match!");
+
+	// if coming from an AJAX call, let's use the Id as the value
+	if( !!li.extra ) var sValue = li.extra[0];
+
+	// otherwise, let's just display the value in the text box
+	else var sValue = li.selectValue;
+
+	$('#googlesearch').submit();
+
+}
+function findValue2(li) {
+	if( li == null ) return alert("No match!");
+
+	// if coming from an AJAX call, let's use the Id as the value
+	if( !!li.extra ) var sValue = li.extra[0];
+
+	// otherwise, let's just display the value in the text box
+	else var sValue = li.selectValue;
+
+	$('#googlesearch2').submit();
+}
+function selectItem(li) {
+	findValue(li);
+}
+function selectItem2(li) {
+	findValue2(li);
+}
+
+$().ready(function() {
+
+	function log(event, data, formatted) {
+		$("<li>").html( !data ? "No match!" : "Selected: " + formatted).appendTo("#result");
+	}
+
+	function formatItem(row) {
+		return row[0] + " (<strong>id: " + row[1] + "</strong>)";
+	}
+	function formatResult(row) {
+		return row[0].replace(/(<.+?>)/gi, '');
+	}
+
+	$("#newSearch").autocomplete("/searchjs.asp", {
+		width: 179,
+		autoFill: false,
+		//delay:10,
+		minChars:2,
+		cacheLength: 10,
+		onFindValue:findValue,
+		onItemSelect: selectItem,
+		selectFirst: false
+
+	});
+
+	$("#search2").autocomplete("/searchjs.asp", {
+		width: 160,
+		autoFill: false,
+		//delay:10,
+		minChars:2,
+		cacheLength: 10,
+		onFindValue:findValue2,
+		onItemSelect: selectItem2,
+		selectFirst: false
+
+	});
+
+});
+
+</script>
+<!-- END CODE NEEDED TO MAKE THE SEARCH BOX WORK -->
+
+<!-- FORESEE Code -->
+<script type="text/javascript" src="/foresee/foresee-trigger.js"></script>
+
+<a href="#after_header" class="responsive_header-skip_header">Skip Header</a>
+<header>
+<div id="responsive_header">
+	<div id="responsive_header-right_side">
+		<ul id="responsive_header-links">
+			<li id="responsive_header-twitter" title="Visit FDIC on Twitter"><a tabindex="1" href="/social.html?site=http://twitter.com/FDICgov">Visit FDIC on Twitter</a></li>
+			<li id="responsive_header-facebook" title="Visit FDIC on Facebook"><a tabindex="1" href="/social.html?site=http://www.facebook.com/FDICgov">Visit FDIC on Facebook</a></li>
+			<li id="responsive_header-fdicchannel" title="Visit FDIC on YouTube"><a tabindex="1" href="/social.html?site=http://www.youtube.com/user/FDICchannel">Visit FDIC on YouTube</a></li>
+			<li id="responsive_header-rss" title="FDIC RSS Feed"><a tabindex="1" href="/rss.html">FDIC RSS Feed</a></li>
+			<li id="responsive_header-subscribe" title="Subscribe to FDIC alerts"><a tabindex="1" href="http://service.govdelivery.com/service/multi_subscribe.html?code=USFDIC">Subscribe to FDIC alerts</a></li>
+		</ul>
+		<div id="responsive_header-search">
+			<a href="/search/advanced.html" class="search" title="Advanced Search">Advanced Search</a>
+			<form id="googlesearch" action="http://search.fdic.gov/search" method="get" name="Search box for FDIC.gov">
+				<fieldset>
+					<div class="form">
+						<label for="q">Search FDIC.gov</label>
+						<input tabindex="1" id="newSearch" name="q" class="field" type="text" style="outline: 0 none;" value="Search FDIC..." onblur="if(this.value == '') {this.value = 'Search FDIC...';}" onfocus="if(this.value == 'Search FDIC...') {this.value = '';}" />
+						<input tabindex="1" id="searchsubmit" class="submit" alt="Search Icon" title="Search Icon" type="submit" value="" />
+						<input value="date:D:L:d1" name="sort" type="hidden" />
+
+						<input value="xml_no_dtd" name="output" type="hidden" />
+						<input value="UTF-8" name="ie" type="hidden" />
+						<input value="UTF-8" name="oe" type="hidden" />
+						<input value="wwwGOV" name="client" type="hidden" />
+						<input value="wwwGOV" name="proxystylesheet" type="hidden" />
+						<input value="default" name="site" type="hidden" />
+					</div>
+				</fieldset>
+			</form>
+		</div>
+	</div>
+	<!-- close right side -->
+	<a id="responsive_header-fdic_logo" href="/" title="FDIC Homepage">FDIC Homepage</a>
+	<h1>Federal Deposit<br>Insurance Corporation</h1>
+	<h2>Each depositor insured to at least $250,000 per insured bank</h2>
+	<div class="clear"></div>
+	<nav>
+		<div id="responsive_header_nav">
+			<div id="responsive_header-topnav">
+				<div id="responsive_header-topnav-downarrow" onclick="show_rwdnav(this)"></div>
+				<ul id="responsive_header-topnav-list">
+					<li id="responsive_header-topnav-home" title="Home" onmouseover="show_responsive_header_subnav(this)"><a href="/">Home</a></li>
+					<li id="responsive_header-topnav-deposit" title="Deposit Insurance" onmouseover="show_responsive_header_subnav(this)"><a href="/deposit/">Deposit Insurance</a></li>
+					<li id="responsive_header-topnav-consumers" title="Consumer Protection" onmouseover="show_responsive_header_subnav(this)"><a href="/consumers/">Consumer Protection</a></li>
+					<li id="responsive_header-topnav-bank" title="Industry Analysis" onmouseover="show_responsive_header_subnav(this)"><a href="/bank/">Industry Analysis</a></li>
+					<li id="responsive_header-topnav-regulations" title="Regulations &amp; Examinations" onmouseover="show_responsive_header_subnav(this)"><a href="/regulations/">Regulations &amp; Examinations</a></li>
+					<li id="responsive_header-topnav-buying" title="Asset Sales" onmouseover="show_responsive_header_subnav(this)"><a href="/buying/">Asset Sales</a></li>
+					<li id="responsive_header-topnav-news" title="News &amp; Events" onmouseover="show_responsive_header_subnav(this)"><a href="/news/">News &amp; Events</a></li>
+					<li id="responsive_header-topnav-about" title="About FDIC" onmouseover="show_responsive_header_subnav(this)"><a href="/about/">About FDIC</a></li>
+				</ul>
+				<div class="clear"></div>
+			</div>
+			<div id="responsive_header-topnav_subnav">
+				<div id="responsive_header-topnav_subnav-downarrow" onclick="show_rwdnav(this)"></div>
+				<ul id="responsive_header-topnav-home_subnav"><li><a>&nbsp;</a></li></ul>
+				<ul id="responsive_header-topnav-deposit_subnav">
+					<li title="BankFind"><a href="http://research.fdic.gov/bankfind/">BankFind</a></li>
+					<li title="Are My Deposits Insured?"><a href="/deposit/deposits/">Are My Deposits Insured?</a></li>
+					<li title="Uninsured Investments"><a href="/deposit/investments/">Uninsured Investments</a></li>
+					<li title="The Deposit Insurance Fund"><a href="/deposit/insurance/index.html">The Deposit Insurance Fund</a></li>
+					<li title="International Deposit Insurance"><a href="/deposit/deposits/international/">International Deposit Insurance</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-consumers_subnav">
+					<li title="Consumer News &amp; Information"><a href="/consumers/consumer/">Consumer News &amp; Information</a></li>
+					<li title="Loans &amp; Mortgages"><a href="/consumers/loans/">Loans &amp; Mortgages</a></li>
+					<li title="Banking &amp; Your Money"><a href="/consumers/banking/">Banking &amp; Your Money</a></li>
+					<li title="Financial Education &amp; Literacy"><a href="/consumers/education/">Financial Education &amp; Literacy</a></li>
+					<li title="Community Affairs"><a href="/consumers/community/">Community Affairs</a></li>
+					<li title="Identity Theft &amp; Fraud"><a href="/consumers/theft/">Identity Theft &amp; Fraud</a></li>
+					<li title="Consumer Financial Privacy"><a href="/consumers/privacy/">Consumer Financial Privacy</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-bank_subnav">
+					<li title="Bank Data &amp; Statistics"><a href="/bank/statistical/">Bank Data &amp; Statistics</a></li>
+					<li title="Research &amp; Analysis"><a href="/bank/analytical/">Research &amp; Analysis</a></li>
+					<li title="Failed Banks"><a href="/bank/individual/failed/">Failed Banks</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-regulations_subnav">
+					<li title="Bank Examinations"><a href="/regulations/examinations/">Bank Examinations</a></li>
+					<li title="Laws &amp; Regulations"><a href="/regulations/laws/">Laws &amp; Regulations</a></li>
+					<li title="Resources for Bank Officers &amp; Directors"><a href="/regulations/resources/">Resources for Bank Officers &amp; Directors</a></li>
+					<li title="FDICconnect"><a href="http://www.fdicconnect.gov/">FDIC<em>connect</em></a></li>
+					<li title="Required Financial Reports"><a href="/regulations/required/">Required Financial Reports</a></li>
+					<li title="Examiner Training Programs"><a href="/regulations/examiner/">Examiner Training Programs</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-buying_subnav">
+					<li title="Loan Sales"><a href="/buying/loan/">Loan Sales</a></li>
+					<li title="Real Estate Sales"><a href="/buying/owned/">Real Estate and Property Marketplace</a></li>
+					<li title="Financial Asset Sales"><a href="/buying/financial/">Financial Asset Sales</a></li>
+					<li title="Servicing Sales Announcements"><a href="/buying/servicing/">Servicing Sales Announcements</a></li>
+					<li title="Other Asset Sales"><a href="/buying/otherasset/">Other Asset Sales</a></li>
+					<li title="Historical Sales"><a href="/buying/historical/">Historical Sales</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-news_subnav">
+					<li title="Press Releases"><a href="/news/news/press/2013/">Press Releases</a></li>
+					<li title="Online Press Room"><a href="https://fdicsurvey.inquisiteasp.com/fdic/cgi-bin/qwebcorporate.dll?M58TRS">Online Press Room</a></li>
+					<li title="Conferences &amp; Events"><a href="/news/conferences/">Conferences &amp; Events</a></li>
+					<li title="Financial Institution Letters"><a href="/news/news/financial/2013/">Financial Institution Letters</a></li>
+					<li title="Special Alerts"><a href="/news/news/SpecialAlert/2012/">Special Alerts</a></li>
+					<li title="Letters to the Editor/Opinion Editorials"><a href="/news/letters/">Letters to the Editor/Opinion Editorials</a></li>
+					<li title="Speeches &amp; Testimony"><a href="/news/news/speeches/chairman/">Speeches &amp; Testimony</a></li>
+				</ul>
+				<ul id="responsive_header-topnav-about_subnav">
+					<li title="Mission &amp; Purpose"><a href="/about/index.html#1">Mission &amp; Purpose</a></span></li>
+					<li title="Advisory Committees"><a href="/about/index.html#2">Advisory Committees</a></span></li>
+					<li title="Careers with the FDIC"><a href="/about/index.html#3">Careers with the FDIC</a></span></li>
+					<li title="Management Team"><a href="/about/index.html#4">Management Team</a></span></li>
+					<li title="Plans &amp; Reports"><a href="/about/index.html#5">Plans &amp; Reports</a></span></li>
+					<li title="What We Can Do for You"><a href="/about/index.html#6">What We Can Do for You</a></span></li>
+					<li title="Diversity at the FDIC"><a href="/about/index.html#7">Diversity at the FDIC</a></span></li>
+				</ul>
+			</div><!-- Close subnav -->
+			<div class="clear"></div>
+		</div>
+	</nav>
+</div>
+</header>
+<a id="after_header" name="after_header"></a>
+<script type="text/javascript">
+prepare_responsive_header_nav();
+</script>
+<!-- END of Header -->
+
+<div id="breadcrumbs"><a href="/">Home</a> &gt; <a href="/bank/">Industry Analysis</a> &gt; <a href="/bank/individual/failed/">Failed Banks</a> &gt; Failed Bank List</div>
+
+<div id="content" class="failed_bank_list">
+
+	<h1 class="page_title">Failed Bank List</h1>
+
+	<p>The FDIC is often appointed as receiver for failed banks. This page contains useful information for the customers and vendors of these banks. This includes information on the acquiring bank (if applicable), how your accounts and loans are affected, and how vendors can file claims against the receivership. <a href="http://www2.fdic.gov/drrip/cs/index.asp">Failed Financial Institution Contact Search</a> displays point of contact information related to failed banks.</p>
+
+	<p>This list includes banks which have failed since October 1, 2000. To search for banks that failed prior to those on this page, visit this link: <a href="http://www2.fdic.gov/hsob/SelectRpt.asp?EntryTyp=30">Failures and Assistance Transactions</a></p>
+
+	<p><a href="banklist.csv">Failed Bank List</a> - CSV file (Updated on Mondays. Also opens in Excel - <a href="/excel.html">Excel Help</a>)</p>
+
+	<p class="small_screen_warning">Due to the small screen size some information is no longer visible.<br>Full information available when viewed on a larger screen.</p>
+
+	<script type="text/javascript">
+	<!--
+	document.writeln("<p><em>Click arrows next to headers to sort in Ascending or Descending order.</em></p>");
+	//-->
+	</script>
+
+	<div id="table_wrapper">
+	<table id="table" class="sortable">
+		<thead>
+			<tr>
+				<th id="institution" scope="col">Bank Name</th>
+				<th id="city" class="nosort" scope="col">City</th>
+				<th id="state" scope="col">ST</th>
+				<th id="cert" class="nosort" scope="col">CERT</th>
+				<th id="ai" scope="col">Acquiring Institution</th>
+				<th id="closing" scope="col">Closing Date</th>
+				<th id="updated" scope="col">Updated Date</th>
+			</tr>
+		</thead>
+		<tbody>
+			<tr>
+			  <td class="institution"><a href="kenosha.html">Banks of Wisconsin d/b/a Bank of Kenosha</a></td>
+			  <td class="city">Kenosha</td>
+			  <td class="state">WI</td>
+			  <td class="cert">35386</td>
+			  <td class="ai">North Shore Bank, FSB</td>
+			  <td class="closing">May 31, 2013</td>
+			  <td class="updated">May 31, 2013</td>
+			</tr>
+            <tr>
+			  <td class="institution"><a href="centralaz.html">Central Arizona Bank</a></td>
+			  <td class="city">Scottsdale</td>
+			  <td class="state">AZ</td>
+			  <td class="cert">34527</td>
+			  <td class="ai">Western State Bank</td>
+			  <td class="closing">May 14, 2013</td>
+			  <td class="updated">May 20, 2013</td>
+			</tr>
+			<tr>
+			  <td class="institution"><a href="sunrisebank.html">Sunrise Bank</a></td>
+			  <td class="city">Valdosta</td>
+			  <td class="state">GA</td>
+			  <td class="cert">58185</td>
+			  <td class="ai">Synovus Bank</td>
+			  <td class="closing">May 10, 2013</td>
+			  <td class="updated">May 21, 2013</td>
+			</tr>
+			<tr>
+			  <td class="institution"><a href="pisgahcommbk.html">Pisgah Community Bank</a></td>
+			  <td class="city">Asheville</td>
+			  <td class="state">NC</td>
+			  <td class="cert">58701</td>
+			  <td class="ai">Capital Bank, N.A.</td>
+			  <td class="closing">May 10, 2013</td>
+			  <td class="updated">May 14, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="douglascb.html">Douglas County Bank</a></td>
+				<td class="city">Douglasville</td>
+				<td class="state">GA</td>
+				<td class="cert">21649</td>
+				<td class="ai">Hamilton State Bank</td>
+				<td class="closing">April 26, 2013</td>
+				<td class="updated">May 16, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="parkway.html">Parkway Bank</a></td>
+				<td class="city">Lenoir</td>
+				<td class="state">NC</td>
+				<td class="cert">57158</td>
+				<td class="ai">CertusBank, National Association</td>
+				<td class="closing">April 26, 2013</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="chipola.html">Chipola Community Bank</a></td>
+				<td class="city">Marianna</td>
+				<td class="state">FL</td>
+				<td class="cert">58034</td>
+				<td class="ai">First Federal Bank of Florida</td>
+				<td class="closing">April 19, 2013</td>
+				<td class="updated">May 16, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heritagebank-fl.html">Heritage Bank of North Florida</a></td>
+				<td class="city">Orange Park</td>
+				<td class="state">FL</td>
+				<td class="cert">26680</td>
+				<td class="ai">FirstAtlantic Bank</td>
+				<td class="closing">April 19, 2013</td>
+				<td class="updated">May 16, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstfederal-ky.html">First Federal Bank</a></td>
+				<td class="city">Lexington</td>
+				<td class="state">KY</td>
+				<td class="cert">29594</td>
+				<td class="ai">Your Community Bank</td>
+				<td class="closing">April 19, 2013</td>
+				<td class="updated">April 23, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="goldcanyon.html">Gold Canyon Bank</a></td>
+				<td class="city">Gold Canyon</td>
+				<td class="state">AZ</td>
+				<td class="cert">58066</td>
+				<td class="ai">First Scottsdale Bank, National Association</td>
+				<td class="closing">April 5, 2013</td>
+				<td class="updated">April 9, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="frontier-ga.html">Frontier Bank</a></td>
+				<td class="city">LaGrange</td>
+				<td class="state">GA</td>
+				<td class="cert">16431</td>
+				<td class="ai">HeritageBank of the South</td>
+				<td class="closing">March 8, 2013</td>
+				<td class="updated">March 26, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="covenant-il.html">Covenant Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">22476</td>
+				<td class="ai">Liberty Bank and Trust Company</td>
+				<td class="closing">February 15, 2013</td>
+				<td class="updated">March 4, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="1stregents.html">1st Regents Bank</a></td>
+				<td class="city">Andover</td>
+				<td class="state">MN</td>
+				<td class="cert">57157</td>
+				<td class="ai">First Minnesota Bank</td>
+				<td class="closing">January 18, 2013</td>
+				<td class="updated">February 28, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westside.html">Westside Community Bank</a></td>
+				<td class="city">University Place</td>
+				<td class="state">WA</td>
+				<td class="cert">33997</td>
+				<td class="ai">Sunwest Bank</td>
+				<td class="closing">January 11, 2013</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cmbkozarks.html">Community Bank of the Ozarks</a></td>
+				<td class="city">Sunrise Beach</td>
+				<td class="state">MO</td>
+				<td class="cert">27331</td>
+				<td class="ai">Bank of Sullivan</td>
+				<td class="closing">December 14, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="hometown.html">Hometown Community Bank</a></td>
+				<td class="city">Braselton</td>
+				<td class="state">GA</td>
+				<td class="cert">57928</td>
+				<td class="ai">CertusBank, National Association</td>
+				<td class="closing">November 16, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cfnb.html">Citizens First National Bank</a></td>
+				<td class="city">Princeton</td>
+				<td class="state">IL</td>
+				<td class="cert">3731</td>
+				<td class="ai">Heartland Bank and Trust Company</td>
+				<td class="closing">November 2, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heritage_fl.html">Heritage Bank of Florida</a></td>
+				<td class="city">Lutz</td>
+				<td class="state">FL</td>
+				<td class="cert">35009</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">November 2, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="novabank.html">NOVA Bank</a></td>
+				<td class="city">Berwyn</td>
+				<td class="state">PA</td>
+				<td class="cert">27148</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">October 26, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="excelbank.html">Excel Bank</a></td>
+				<td class="city">Sedalia</td>
+				<td class="state">MO</td>
+				<td class="cert">19189</td>
+				<td class="ai">Simmons First National Bank</td>
+				<td class="closing">October 19, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firsteastside.html">First East Side Savings Bank</a></td>
+				<td class="city">Tamarac</td>
+				<td class="state">FL</td>
+				<td class="cert">28144</td>
+				<td class="ai">Stearns Bank N.A.</td>
+				<td class="closing">October 19, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="gulfsouth.html">GulfSouth Private Bank</a></td>
+				<td class="city">Destin</td>
+				<td class="state">FL</td>
+				<td class="cert">58073</td>
+				<td class="ai">SmartBank</td>
+				<td class="closing">October 19, 2012</td>
+				<td class="updated">January 24, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstunited.html">First United Bank</a></td>
+				<td class="city">Crete</td>
+				<td class="state">IL</td>
+				<td class="cert">20685</td>
+				<td class="ai">Old Plank Trail Community Bank, National Association</td>
+				<td class="closing">September 28, 2012</td>
+				<td class="updated">November 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="truman.html">Truman Bank</a></td>
+				<td class="city">St. Louis</td>
+				<td class="state">MO</td>
+				<td class="cert">27316</td>
+				<td class="ai">Simmons First National Bank</td>
+				<td class="closing">September 14, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcommbk_mn.html">First Commercial Bank</a></td>
+				<td class="city">Bloomington</td>
+				<td class="state">MN</td>
+				<td class="cert">35246</td>
+				<td class="ai">Republic Bank &amp; Trust Company</td>
+				<td class="closing">September 7, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="waukegan.html">Waukegan Savings Bank</a></td>
+				<td class="city">Waukegan</td>
+				<td class="state">IL</td>
+				<td class="cert">28243</td>
+				<td class="ai">First Midwest Bank</td>
+				<td class="closing">August 3, 2012</td>
+				<td class="updated">October 11, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="jasper.html">Jasper Banking Company</a></td>
+				<td class="city">Jasper</td>
+				<td class="state">GA</td>
+				<td class="cert">16240</td>
+				<td class="ai">Stearns Bank N.A.</td>
+				<td class="closing">July 27, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="secondfederal.html">Second Federal Savings and Loan Association of Chicago</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">27986</td>
+				<td class="ai">Hinsdale Bank &amp; Trust Company</td>
+				<td class="closing">July 20, 2012</td>
+				<td class="updated">January 14, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heartland.html">Heartland Bank</a></td>
+				<td class="city">Leawood</td>
+				<td class="state">KS</td>
+				<td class="cert">1361</td>
+				<td class="ai">Metcalf Bank</td>
+				<td class="closing">July 20, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cherokee.html">First Cherokee State Bank</a></td>
+				<td class="city">Woodstock</td>
+				<td class="state">GA</td>
+				<td class="cert">32711</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">July 20, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="georgiatrust.html">Georgia Trust Bank</a></td>
+				<td class="city">Buford</td>
+				<td class="state">GA</td>
+				<td class="cert">57847</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">July 20, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="royalpalm.html">The Royal Palm Bank of Florida</a></td>
+				<td class="city">Naples</td>
+				<td class="state">FL</td>
+				<td class="cert">57096</td>
+				<td class="ai">First National Bank of the Gulf Coast</td>
+				<td class="closing">July 20, 2012</td>
+				<td class="updated">January 7, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="glasgow.html">Glasgow Savings Bank</a></td>
+				<td class="city">Glasgow</td>
+				<td class="state">MO</td>
+				<td class="cert">1056</td>
+				<td class="ai">Regional Missouri Bank</td>
+				<td class="closing">July 13, 2012</td>
+				<td class="updated">October 11, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="montgomery.html">Montgomery Bank &amp; Trust</a></td>
+				<td class="city">Ailey</td>
+				<td class="state">GA</td>
+				<td class="cert">19498</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">July 6, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="farmersbank.html">The Farmers Bank of Lynchburg</a></td>
+				<td class="city">Lynchburg</td>
+				<td class="state">TN</td>
+				<td class="cert">1690</td>
+				<td class="ai">Clayton Bank and Trust</td>
+				<td class="closing">June 15, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="securityexchange.html">Security Exchange Bank</a></td>
+				<td class="city">Marietta</td>
+				<td class="state">GA</td>
+				<td class="cert">35299</td>
+				<td class="ai">Fidelity Bank</td>
+				<td class="closing">June 15, 2012</td>
+				<td class="updated">October 10, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="putnam.html">Putnam State Bank</a></td>
+				<td class="city">Palatka</td>
+				<td class="state">FL</td>
+				<td class="cert">27405</td>
+				<td class="ai">Harbor Community Bank</td>
+				<td class="closing">June 15, 2012</td>
+				<td class="updated">October 10, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="waccamaw.html">Waccamaw Bank</a></td>
+				<td class="city">Whiteville</td>
+				<td class="state">NC</td>
+				<td class="cert">34515</td>
+				<td class="ai">First Community Bank</td>
+				<td class="closing">June 8, 2012</td>
+				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ftsb.html">Farmers' and Traders' State Bank</a></td>
+				<td class="city">Shabbona</td>
+				<td class="state">IL</td>
+				<td class="cert">9257</td>
+				<td class="ai">First State Bank</td>
+				<td class="closing">June 8, 2012</td>
+				<td class="updated">October 10, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="carolina.html">Carolina Federal Savings Bank</a></td>
+				<td class="city">Charleston</td>
+				<td class="state">SC</td>
+				<td class="cert">35372</td>
+				<td class="ai">Bank of North Carolina</td>
+				<td class="closing">June 8, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcapital.html">First Capital Bank</a></td>
+				<td class="city">Kingfisher</td>
+				<td class="state">OK</td>
+				<td class="cert">416</td>
+				<td class="ai">F &amp; M Bank</td>
+				<td class="closing">June 8, 2012</td>
+				<td class="updated">October 10, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="alabamatrust.html">Alabama Trust Bank, National Association</a></td>
+				<td class="city">Sylacauga</td>
+				<td class="state">AL</td>
+				<td class="cert">35224</td>
+				<td class="ai">Southern States Bank</td>
+				<td class="closing">May 18, 2012</td>
+				<td class="updated">May 20, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="securitybank.html">Security Bank, National Association</a></td>
+				<td class="city">North Lauderdale</td>
+				<td class="state">FL</td>
+				<td class="cert">23156</td>
+				<td class="ai">Banesco USA</td>
+				<td class="closing">May 4, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="palmdesert.html">Palm Desert National Bank</a></td>
+				<td class="city">Palm Desert</td>
+				<td class="state">CA</td>
+				<td class="cert">23632</td>
+				<td class="ai">Pacific Premier Bank</td>
+				<td class="closing">April 27, 2012</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="plantation.html">Plantation Federal Bank</a></td>
+				<td class="city">Pawleys Island</td>
+				<td class="state">SC</td>
+				<td class="cert">32503</td>
+				<td class="ai">First Federal Bank</td>
+				<td class="closing">April 27, 2012</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="interbank.html">Inter Savings Bank, fsb D/B/A InterBank, fsb</a></td>
+				<td class="city">Maple Grove</td>
+				<td class="state">MN</td>
+				<td class="cert">31495</td>
+				<td class="ai">Great Southern Bank</td>
+				<td class="closing">April 27, 2012</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="harvest.html">HarVest Bank of Maryland</a></td>
+				<td class="city">Gaithersburg</td>
+				<td class="state">MD</td>
+				<td class="cert">57766</td>
+				<td class="ai">Sonabank</td>
+				<td class="closing">April 27, 2012</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="easternshore.html">Bank of the Eastern Shore</a></td>
+				<td class="city">Cambridge</td>
+				<td class="state">MD</td>
+				<td class="cert">26759</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">April 27, 2012</td>
+				<td class="updated">October 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fortlee.html">Fort Lee Federal Savings Bank, FSB</a></td>
+				<td class="city">Fort Lee</td>
+				<td class="state">NJ</td>
+				<td class="cert">35527</td>
+				<td class="ai">Alma Bank</td>
+				<td class="closing">April 20, 2012</td>
+				<td class="updated">May 17, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fidelity.html">Fidelity Bank</a></td>
+				<td class="city">Dearborn</td>
+				<td class="state">MI</td>
+				<td class="cert">33883</td>
+				<td class="ai">The Huntington National Bank</td>
+				<td class="closing">March 30, 2012</td>
+				<td class="updated">May 16, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="premier-il.html">Premier Bank</a></td>
+				<td class="city">Wilmette</td>
+				<td class="state">IL</td>
+				<td class="cert">35419</td>
+				<td class="ai">International Bank of Chicago</td>
+				<td class="closing">March 23, 2012</td>
+				<td class="updated">October 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="covenant.html">Covenant Bank &amp; Trust</a></td>
+				<td class="city">Rock Spring</td>
+				<td class="state">GA</td>
+				<td class="cert">58068</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">March 23, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newcity.html">New City Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">57597</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 9, 2012</td>
+				<td class="updated">October 29, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="global.html">Global Commerce Bank</a></td>
+				<td class="city">Doraville</td>
+				<td class="state">GA</td>
+				<td class="cert">34046</td>
+				<td class="ai">Metro City Bank</td>
+				<td class="closing">March 2, 2012</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="homesvgs.html">Home Savings of America</a></td>
+				<td class="city">Little Falls</td>
+				<td class="state">MN</td>
+				<td class="cert">29178</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">February 24, 2012</td>
+				<td class="updated">December 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cbg.html">Central Bank of Georgia</a></td>
+				<td class="city">Ellaville</td>
+				<td class="state">GA</td>
+				<td class="cert">5687</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">February 24, 2012</td>
+				<td class="updated">August 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="scbbank.html">SCB Bank</a></td>
+				<td class="city">Shelbyville</td>
+				<td class="state">IN</td>
+				<td class="cert">29761</td>
+				<td class="ai">First Merchants Bank, National Association</td>
+				<td class="closing">February 10, 2012</td>
+				<td class="updated">March 25, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cnbt.html">Charter National Bank and Trust</a></td>
+				<td class="city">Hoffman Estates</td>
+				<td class="state">IL</td>
+				<td class="cert">23187</td>
+				<td class="ai">Barrington Bank &amp; Trust Company, National Association</td>
+				<td class="closing">February 10, 2012</td>
+				<td class="updated">March 25, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankeast.html">BankEast</a></td>
+				<td class="city">Knoxville</td>
+				<td class="state">TN</td>
+				<td class="cert">19869</td>
+				<td class="ai">U.S.Bank National Association</td>
+				<td class="closing">January 27, 2012</td>
+				<td class="updated">March 8, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="patriot-mn.html">Patriot Bank Minnesota</a></td>
+				<td class="city">Forest Lake</td>
+				<td class="state">MN</td>
+				<td class="cert">34823</td>
+				<td class="ai">First Resource Bank</td>
+				<td class="closing">January 27, 2012</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="tcb.html">Tennessee Commerce Bank</a></td>
+				<td class="city">Franklin</td>
+				<td class="state">TN</td>
+				<td class="cert">35296</td>
+				<td class="ai">Republic Bank &amp; Trust Company</td>
+				<td class="closing">January 27, 2012</td>
+				<td class="updated">November 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fgbtcj.html">First Guaranty Bank and Trust Company of Jacksonville</a></td>
+				<td class="city">Jacksonville</td>
+				<td class="state">FL</td>
+				<td class="cert">16579</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">January 27, 2012</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americaneagle.html">American Eagle Savings Bank</a></td>
+				<td class="city">Boothwyn</td>
+				<td class="state">PA</td>
+				<td class="cert">31581</td>
+				<td class="ai">Capital Bank, N.A.</td>
+				<td class="closing">January 20, 2012</td>
+				<td class="updated">January 25, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firststatebank-ga.html">The First State Bank</a></td>
+				<td class="city">Stockbridge</td>
+				<td class="state">GA</td>
+				<td class="cert">19252</td>
+				<td class="ai">Hamilton State Bank</td>
+				<td class="closing">January 20, 2012</td>
+				<td class="updated">January 25, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cfsb.html">Central Florida State Bank</a></td>
+				<td class="city">Belleview</td>
+				<td class="state">FL</td>
+				<td class="cert">57186</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">January 20, 2012</td>
+				<td class="updated">January 25, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westernnatl.html">Western National Bank</a></td>
+				<td class="city">Phoenix</td>
+				<td class="state">AZ</td>
+				<td class="cert">57917</td>
+				<td class="ai">Washington Federal</td>
+				<td class="closing">December 16, 2011</td>
+				<td class="updated">August 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="premier-fl.html">Premier Community Bank of the Emerald Coast</a></td>
+				<td class="city">Crestview</td>
+				<td class="state">FL</td>
+				<td class="cert">58343</td>
+				<td class="ai">Summit Bank</td>
+				<td class="closing">December 16, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="centralprog.html">Central Progressive Bank</a></td>
+				<td class="city">Lacombe</td>
+				<td class="state">LA</td>
+				<td class="cert">19657</td>
+				<td class="ai">First NBC Bank</td>
+				<td class="closing">November 18, 2011</td>
+				<td class="updated">August 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="polkcounty.html">Polk County Bank</a></td>
+				<td class="city">Johnston</td>
+				<td class="state">IA</td>
+				<td class="cert">14194</td>
+				<td class="ai">Grinnell State Bank</td>
+				<td class="closing">November 18, 2011</td>
+				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="rockmart.html">Community Bank of Rockmart</a></td>
+				<td class="city">Rockmart</td>
+				<td class="state">GA</td>
+				<td class="cert">57860</td>
+				<td class="ai">Century Bank of Georgia</td>
+				<td class="closing">November 10, 2011</td>
+				<td class="updated">August 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sunfirst.html">SunFirst Bank</a></td>
+				<td class="city">Saint George</td>
+				<td class="state">UT</td>
+				<td class="cert">57087</td>
+				<td class="ai">Cache Valley Bank</td>
+				<td class="closing">November 4, 2011</td>
+				<td class="updated">November 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="midcity.html">Mid City Bank, Inc.</a></td>
+				<td class="city">Omaha</td>
+				<td class="state">NE</td>
+				<td class="cert">19397</td>
+				<td class="ai">Premier Bank</td>
+				<td class="closing">November 4, 2011</td>
+				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="allamerican.html ">All American Bank</a></td>
+				<td class="city">Des Plaines</td>
+				<td class="state">IL</td>
+				<td class="cert">57759</td>
+				<td class="ai">International Bank of Chicago</td>
+				<td class="closing">October 28, 2011</td>
+				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commbanksco.html">Community Banks of Colorado</a></td>
+				<td class="city">Greenwood Village</td>
+				<td class="state">CO</td>
+				<td class="cert">21132</td>
+				<td class="ai">Bank Midwest, N.A.</td>
+				<td class="closing">October 21, 2011</td>
+				<td class="updated">January 2, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commcapbk.html">Community Capital Bank</a></td>
+				<td class="city">Jonesboro</td>
+				<td class="state">GA</td>
+				<td class="cert">57036</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">October 21, 2011</td>
+				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="decatur.html">Decatur First Bank</a></td>
+				<td class="city">Decatur</td>
+				<td class="state">GA</td>
+				<td class="cert">34392</td>
+				<td class="ai">Fidelity Bank</td>
+				<td class="closing">October 21, 2011</td>
+				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="oldharbor.html">Old Harbor Bank</a></td>
+				<td class="city">Clearwater</td>
+				<td class="state">FL</td>
+				<td class="cert">57537</td>
+				<td class="ai">1st United Bank</td>
+				<td class="closing">October 21, 2011</td>
+				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="countrybank.html">Country Bank</a></td>
+ 				<td class="city">Aledo</td>
+ 				<td class="state">IL</td>
+ 				<td class="cert">35395</td>
+ 				<td class="ai">Blackhawk Bank &amp; Trust</td>
+ 				<td class="closing">October 14, 2011</td>
+ 				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firststatebank-nj.html">First State Bank</a></td>
+ 				<td class="city">Cranford</td>
+ 				<td class="state">NJ</td>
+ 				<td class="cert">58046</td>
+ 				<td class="ai">Northfield Bank</td>
+ 				<td class="closing">October 14, 2011</td>
+ 				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="blueridge.html">Blue Ridge Savings Bank, Inc.</a></td>
+ 				<td class="city">Asheville</td>
+ 				<td class="state">NC</td>
+ 				<td class="cert">32347</td>
+ 				<td class="ai">Bank of North Carolina</td>
+ 				<td class="closing">October 14, 2011</td>
+ 				<td class="updated">November 8, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="piedmont-ga.html">Piedmont Community Bank</a></td>
+ 				<td class="city">Gray</td>
+ 				<td class="state">GA</td>
+ 				<td class="cert">57256</td>
+ 				<td class="ai">State Bank and Trust Company</td>
+ 				<td class="closing">October 14, 2011</td>
+ 				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sunsecurity.html">Sun Security Bank</a></td>
+ 				<td class="city">Ellington</td>
+ 				<td class="state">MO</td>
+ 				<td class="cert">20115</td>
+ 				<td class="ai">Great Southern Bank</td>
+ 				<td class="closing">October 7, 2011</td>
+ 				<td class="updated">November 7, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="riverbank.html">The RiverBank</a></td>
+ 				<td class="city">Wyoming</td>
+ 				<td class="state">MN</td>
+ 				<td class="cert">10216</td>
+ 				<td class="ai">Central Bank</td>
+ 				<td class="closing">October 7, 2011</td>
+ 				<td class="updated">November 7, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstintlbank.html">First International Bank</a></td>
+ 				<td class="city">Plano</td>
+ 				<td class="state">TX</td>
+ 				<td class="cert">33513</td>
+ 				<td class="ai">American First National Bank</td>
+ 				<td class="closing">September 30, 2011</td>
+ 				<td class="updated">October 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cbnc.html">Citizens Bank of Northern California</a></td>
+				<td class="city">Nevada City</td>
+				<td class="state">CA</td>
+				<td class="cert">33983</td>
+				<td class="ai">Tri Counties Bank</td>
+				<td class="closing">September 23, 2011</td>
+				<td class="updated">October 9, 2012</td>
+			</tr>
+			<tr>
+  				<td class="institution"><a href="boc-va.html">Bank of the Commonwealth</a></td>
+				<td class="city">Norfolk</td>
+				<td class="state">VA</td>
+				<td class="cert">20408</td>
+				<td class="ai">Southern Bank and Trust Company</td>
+				<td class="closing">September 23, 2011</td>
+				<td class="updated">October 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fnbf.html">The First National Bank of Florida</a></td>
+				<td class="city">Milton</td>
+				<td class="state">FL</td>
+				<td class="cert">25155</td>
+				<td class="ai">CharterBank</td>
+				<td class="closing">September 9, 2011</td>
+				<td class="updated">September 6, 2012</td>
+			</tr>
+			<tr>
+  				<td class="institution"><a href="creekside.html">CreekSide Bank</a></td>
+				<td class="city">Woodstock</td>
+				<td class="state">GA</td>
+				<td class="cert">58226</td>
+				<td class="ai">Georgia Commerce Bank</td>
+				<td class="closing">September 2, 2011</td>
+				<td class="updated">September 6, 2012</td>
+			</tr>
+			<tr>
+  				<td class="institution"><a href="patriot.html">Patriot Bank of Georgia</a></td>
+				<td class="city">Cumming</td>
+				<td class="state">GA</td>
+				<td class="cert">58273</td>
+				<td class="ai">Georgia Commerce Bank</td>
+				<td class="closing">September 2, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstchoice-il.html">First Choice Bank</a></td>
+				<td class="city">Geneva</td>
+				<td class="state">IL</td>
+				<td class="cert">57212</td>
+				<td class="ai">Inland Bank &amp; Trust</td>
+				<td class="closing">August 19, 2011</td>
+				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstsouthern-ga.html">First Southern National Bank</a></td>
+				<td class="city">Statesboro</td>
+				<td class="state">GA</td>
+				<td class="cert">57239</td>
+				<td class="ai">Heritage Bank of the South</td>
+				<td class="closing">August 19, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lydian.html">Lydian Private Bank</a></td>
+				<td class="city">Palm Beach</td>
+				<td class="state">FL</td>
+				<td class="cert">35356</td>
+				<td class="ai">Sabadell United Bank, N.A.</td>
+				<td class="closing">August 19, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="publicsvgs.html">Public Savings Bank</a></td>
+				<td class="city">Huntingdon Valley</td>
+				<td class="state">PA</td>
+				<td class="cert">34130</td>
+				<td class="ai">Capital Bank, N.A.</td>
+				<td class="closing">August 18, 2011</td>
+				<td class="updated">August 15, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fnbo.html">The First National Bank of Olathe</a></td>
+				<td class="city">Olathe</td>
+				<td class="state">KS</td>
+				<td class="cert">4744</td>
+				<td class="ai">Enterprise Bank &amp; Trust</td>
+				<td class="closing">August 12, 2011</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="whitman.html">Bank of Whitman</a></td>
+				<td class="city">Colfax</td>
+				<td class="state">WA</td>
+				<td class="cert">22528</td>
+				<td class="ai">Columbia State Bank</td>
+				<td class="closing">August 5, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="shorewood.html">Bank of Shorewood</a></td>
+				<td class="city">Shorewood</td>
+				<td class="state">IL</td>
+				<td class="cert">22637</td>
+				<td class="ai">Heartland Bank and Trust Company</td>
+				<td class="closing">August 5, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="integra.html">Integra Bank National Association</a></td>
+				<td class="city">Evansville</td>
+				<td class="state">IN</td>
+				<td class="cert">4392</td>
+				<td class="ai">Old National Bank</td>
+				<td class="closing">July 29, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankmeridian.html">BankMeridian, N.A.</a></td>
+				<td class="city">Columbia</td>
+				<td class="state">SC</td>
+				<td class="cert">58222</td>
+				<td class="ai">SCBT National Association</td>
+				<td class="closing">July 29, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="vbb.html">Virginia Business Bank</a></td>
+				<td class="city">Richmond</td>
+				<td class="state">VA</td>
+				<td class="cert">58283</td>
+				<td class="ai">Xenith Bank</td>
+				<td class="closing">July 29, 2011</td>
+				<td class="updated">October 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofchoice.html">Bank of Choice</a></td>
+				<td class="city">Greeley</td>
+				<td class="state">CO</td>
+				<td class="cert">2994</td>
+				<td class="ai">Bank Midwest, N.A.</td>
+				<td class="closing">July 22, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="landmark.html">LandMark Bank of Florida</a></td>
+				<td class="city">Sarasota</td>
+				<td class="state">FL</td>
+				<td class="cert">35244</td>
+				<td class="ai">American Momentum Bank</td>
+				<td class="closing">July 22, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="southshore.html">Southshore Community Bank</a></td>
+				<td class="city">Apollo Beach</td>
+				<td class="state">FL</td>
+				<td class="cert">58056</td>
+				<td class="ai">American Momentum Bank</td>
+				<td class="closing">July 22, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="summitbank.html">Summit Bank</a></td>
+				<td class="city">Prescott</td>
+				<td class="state">AZ</td>
+				<td class="cert">57442</td>
+				<td class="ai">The Foothills Bank</td>
+				<td class="closing">July 15, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstpeoples.html">First Peoples Bank</a></td>
+				<td class="city">Port St. Lucie</td>
+				<td class="state">FL</td>
+				<td class="cert">34870</td>
+				<td class="ai">Premier American Bank, N.A.</td>
+				<td class="closing">July 15, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="hightrust.html">High Trust Bank</a></td>
+				<td class="city">Stockbridge</td>
+				<td class="state">GA</td>
+				<td class="cert">19554</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">July 15, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="onegeorgia.html">One Georgia Bank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">58238</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">July 15, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="signaturebank.html">Signature Bank</a></td>
+				<td class="city">Windsor</td>
+				<td class="state">CO</td>
+				<td class="cert">57835</td>
+				<td class="ai">Points West Community Bank</td>
+				<td class="closing">July 8, 2011</td>
+				<td class="updated">October 26, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="coloradocapital.html">Colorado Capital Bank</a></td>
+				<td class="city">Castle Rock</td>
+				<td class="state">CO</td>
+				<td class="cert">34522</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">July 8, 2011</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstchicago.html">First Chicago Bank &amp; Trust</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">27935</td>
+				<td class="ai">Northbrook Bank &amp; Trust Company</td>
+				<td class="closing">July 8, 2011</td>
+				<td class="updated">September 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mountain.html">Mountain Heritage Bank</a></td>
+				<td class="city">Clayton</td>
+				<td class="state">GA</td>
+				<td class="cert">57593</td>
+				<td class="ai">First American Bank and Trust Company</td>
+				<td class="closing">June 24, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fcbtb.html">First Commercial Bank of Tampa Bay</a></td>
+				<td class="city">Tampa</td>
+				<td class="state">FL</td>
+				<td class="cert">27583</td>
+				<td class="ai">Stonegate Bank</td>
+				<td class="closing">June 17, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mcintoshstate.html">McIntosh State Bank</a></td>
+				<td class="city">Jackson</td>
+				<td class="state">GA</td>
+				<td class="cert">19237</td>
+				<td class="ai">Hamilton State Bank</td>
+				<td class="closing">June 17, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="atlanticbanktrust.html">Atlantic Bank and Trust</a></td>
+				<td class="city">Charleston</td>
+				<td class="state">SC</td>
+				<td class="cert">58420</td>
+				<td class="ai">First Citizens Bank and Trust Company, Inc.</td>
+				<td class="closing">June 3, 2011</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstheritage.html">First Heritage Bank</a></td>
+				<td class="city">Snohomish</td>
+				<td class="state">WA</td>
+				<td class="cert">23626</td>
+				<td class="ai">Columbia State Bank</td>
+				<td class="closing">May 27, 2011</td>
+				<td class="updated">January 28, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="summit.html">Summit Bank</a></td>
+				<td class="city">Burlington</td>
+				<td class="state">WA</td>
+				<td class="cert">513</td>
+				<td class="ai">Columbia State Bank</td>
+				<td class="closing">May 20, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fgbc.html">First Georgia Banking Company</a></td>
+				<td class="city">Franklin</td>
+				<td class="state">GA</td>
+				<td class="cert">57647</td>
+				<td class="ai">CertusBank, National Association</td>
+				<td class="closing">May 20, 2011</td>
+				<td class="updated">November 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="atlanticsthrn.html">Atlantic Southern Bank</a></td>
+				<td class="city">Macon</td>
+				<td class="state">GA</td>
+				<td class="cert">57213</td>
+				<td class="ai">CertusBank, National Association</td>
+				<td class="closing">May 20, 2011</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="coastal_fl.html">Coastal Bank</a></td>
+				<td class="city">Cocoa Beach</td>
+				<td class="state">FL</td>
+				<td class="cert">34898</td>
+				<td class="ai">Florida Community Bank, a division of Premier American Bank, N.A.</td>
+				<td class="closing">May 6, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="communitycentral.html">Community Central Bank</a></td>
+				<td class="city">Mount Clemens</td>
+				<td class="state">MI</td>
+				<td class="cert">34234</td>
+				<td class="ai">Talmer Bank &amp; Trust</td>
+				<td class="closing">April 29, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="parkavenue_ga.html">The Park Avenue Bank</a></td>
+				<td class="city">Valdosta</td>
+				<td class="state">GA</td>
+				<td class="cert">19797</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">April 29, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstchoice.html">First Choice Community Bank</a></td>
+				<td class="city">Dallas</td>
+				<td class="state">GA</td>
+				<td class="cert">58539</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">April 29, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cortez.html">Cortez Community Bank</a></td>
+				<td class="city">Brooksville</td>
+				<td class="state">FL</td>
+				<td class="cert">57625</td>
+				<td class="ai">Florida Community Bank, a division of Premier American Bank, N.A.</td>
+				<td class="closing">April 29, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fnbcf.html">First National Bank of Central Florida</a></td>
+				<td class="city">Winter Park</td>
+				<td class="state">FL</td>
+				<td class="cert">26297</td>
+				<td class="ai">Florida Community Bank, a division of Premier American Bank, N.A.</td>
+				<td class="closing">April 29, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heritage_ms.html">Heritage Banking Group</a></td>
+				<td class="city">Carthage</td>
+				<td class="state">MS</td>
+				<td class="cert">14273</td>
+				<td class="ai">Trustmark National Bank</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="rosemount.html">Rosemount National Bank</a></td>
+				<td class="city">Rosemount</td>
+				<td class="state">MN</td>
+				<td class="cert">24099</td>
+				<td class="ai">Central Bank</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="superior_al.html">Superior Bank</a></td>
+				<td class="city">Birmingham</td>
+				<td class="state">AL</td>
+				<td class="cert">17750</td>
+				<td class="ai">Superior Bank, National Association</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">November 30, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="nexity.html">Nexity Bank</a></td>
+				<td class="city">Birmingham</td>
+				<td class="state">AL</td>
+				<td class="cert">19794</td>
+				<td class="ai">AloStar Bank of Commerce</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newhorizons.html">New Horizons Bank</a></td>
+				<td class="city">East Ellijay</td>
+				<td class="state">GA</td>
+				<td class="cert">57705</td>
+				<td class="ai">Citizens South Bank</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bartow.html">Bartow County Bank</a></td>
+				<td class="city">Cartersville</td>
+				<td class="state">GA</td>
+				<td class="cert">21495</td>
+				<td class="ai">Hamilton State Bank</td>
+				<td class="closing">April 15, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="nevadacommerce.html">Nevada Commerce Bank</a></td>
+				<td class="city">Las Vegas</td>
+				<td class="state">NV</td>
+				<td class="cert">35418</td>
+				<td class="ai">City National Bank</td>
+				<td class="closing">April 8, 2011</td>
+				<td class="updated">September 9, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westernsprings.html">Western Springs National Bank and Trust</a></td>
+				<td class="city">Western Springs</td>
+				<td class="state">IL</td>
+				<td class="cert">10086</td>
+				<td class="ai">Heartland Bank and Trust Company</td>
+				<td class="closing">April 8, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofcommerce.html">The Bank of Commerce</a></td>
+				<td class="city">Wood Dale</td>
+				<td class="state">IL</td>
+				<td class="cert">34292</td>
+				<td class="ai">Advantage National Bank Group</td>
+				<td class="closing">March 25, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="legacy-wi.html">Legacy Bank</a></td>
+				<td class="city">Milwaukee</td>
+				<td class="state">WI</td>
+				<td class="cert">34818</td>
+				<td class="ai">Seaway Bank and Trust Company</td>
+				<td class="closing">March 11, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstnatldavis.html">First National Bank of Davis</a></td>
+				<td class="city">Davis</td>
+				<td class="state">OK</td>
+				<td class="cert">4077</td>
+				<td class="ai">The Pauls Valley National Bank</td>
+				<td class="closing">March 11, 2011</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="valleycomm.html">Valley Community Bank</a></td>
+				<td class="city">St. Charles</td>
+				<td class="state">IL</td>
+				<td class="cert">34187</td>
+				<td class="ai">First State Bank</td>
+				<td class="closing">February 25, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sanluistrust.html">San Luis Trust Bank, FSB</a></td>
+				<td class="city">San Luis Obispo</td>
+				<td class="state">CA</td>
+				<td class="cert">34783</td>
+				<td class="ai">First California Bank</td>
+				<td class="closing">February 18, 2011</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="charteroak.html">Charter Oak Bank</a></td>
+				<td class="city">Napa</td>
+				<td class="state">CA</td>
+				<td class="cert">57855</td>
+				<td class="ai">Bank of Marin</td>
+				<td class="closing">February 18, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizensbk_ga.html">Citizens Bank of Effingham</a></td>
+				<td class="city">Springfield</td>
+				<td class="state">GA</td>
+				<td class="cert">34601</td>
+				<td class="ai">Heritage Bank of the South</td>
+				<td class="closing">February 18, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="habersham.html">Habersham Bank</a></td>
+				<td class="city">Clarkesville</td>
+				<td class="state">GA</td>
+				<td class="cert">151</td>
+				<td class="ai">SCBT National Association</td>
+				<td class="closing">February 18, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="canyonstate.html">Canyon National Bank</a></td>
+				<td class="city">Palm Springs</td>
+				<td class="state">CA</td>
+				<td class="cert">34692</td>
+				<td class="ai">Pacific Premier Bank</td>
+				<td class="closing">February 11, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="badgerstate.html">Badger State Bank</a></td>
+				<td class="city">Cassville</td>
+				<td class="state">WI</td>
+				<td class="cert">13272</td>
+				<td class="ai">Royal Bank</td>
+				<td class="closing">February 11, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peoplesstatebank.html">Peoples State Bank</a></td>
+				<td class="city">Hamtramck</td>
+				<td class="state">MI</td>
+				<td class="cert">14939</td>
+				<td class="ai">First Michigan Bank</td>
+				<td class="closing">February 11, 2011</td>
+				<td class="updated">January 22, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sunshinestate.html">Sunshine State Community Bank</a></td>
+				<td class="city">Port Orange</td>
+				<td class="state">FL</td>
+				<td class="cert">35478</td>
+				<td class="ai">Premier American Bank, N.A.</td>
+				<td class="closing">February 11, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commfirst_il.html">Community First Bank Chicago</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">57948</td>
+				<td class="ai">Northbrook Bank &amp; Trust Company</td>
+				<td class="closing">February 4, 2011</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="northgabank.html">North Georgia Bank</a></td>
+				<td class="city">Watkinsville</td>
+				<td class="state">GA</td>
+				<td class="cert">35242</td>
+				<td class="ai">BankSouth</td>
+				<td class="closing">February 4, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americantrust.html">American Trust Bank</a></td>
+				<td class="city">Roswell</td>
+				<td class="state">GA</td>
+				<td class="cert">57432</td>
+				<td class="ai">Renasant Bank</td>
+				<td class="closing">February 4, 2011</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcomm_nm.html">First Community Bank</a></td>
+				<td class="city">Taos</td>
+				<td class="state">NM</td>
+				<td class="cert">12261</td>
+				<td class="ai">U.S. Bank, N.A.</td>
+				<td class="closing">January 28, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstier.html">FirsTier Bank</a></td>
+				<td class="city">Louisville</td>
+				<td class="state">CO</td>
+				<td class="cert">57646</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">January 28, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="evergreenstatewi.html">Evergreen State Bank</a></td>
+				<td class="city">Stoughton</td>
+				<td class="state">WI</td>
+				<td class="cert">5328</td>
+				<td class="ai">McFarland State Bank</td>
+				<td class="closing">January 28, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firststatebank_ok.html">The First State Bank</a></td>
+				<td class="city">Camargo</td>
+				<td class="state">OK</td>
+				<td class="cert">2303</td>
+				<td class="ai">Bank 7</td>
+				<td class="closing">January 28, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="unitedwestern.html">United Western Bank</a></td>
+				<td class="city">Denver</td>
+				<td class="state">CO</td>
+				<td class="cert">31293</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">January 21, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofasheville.html">The Bank of Asheville</a></td>
+				<td class="city">Asheville</td>
+				<td class="state">NC</td>
+				<td class="cert">34516</td>
+				<td class="ai">First Bank</td>
+				<td class="closing">January 21, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commsouth.html">CommunitySouth Bank &amp; Trust</a></td>
+				<td class="city">Easley</td>
+				<td class="state">SC</td>
+				<td class="cert">57868</td>
+				<td class="ai">CertusBank, National Association</td>
+				<td class="closing">January 21, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="enterprise.html">Enterprise Banking Company</a></td>
+				<td class="city">McDonough</td>
+				<td class="state">GA</td>
+				<td class="cert">19758</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">January 21, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="oglethorpe.html">Oglethorpe Bank</a></td>
+				<td class="city">Brunswick</td>
+				<td class="state">GA</td>
+				<td class="cert">57440</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">January 14, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="legacybank.html">Legacy Bank</a></td>
+				<td class="city">Scottsdale</td>
+				<td class="state">AZ</td>
+				<td class="cert">57820</td>
+				<td class="ai">Enterprise Bank &amp; Trust</td>
+				<td class="closing">January 7, 2011</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcommercial.html">First Commercial Bank of Florida</a></td>
+				<td class="city">Orlando</td>
+				<td class="state">FL</td>
+				<td class="cert">34965</td>
+				<td class="ai">First Southern Bank</td>
+				<td class="closing">January 7, 2011</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="communitynatl.html">Community National Bank</a></td>
+				<td class="city">Lino Lakes</td>
+				<td class="state">MN</td>
+				<td class="cert">23306</td>
+				<td class="ai">Farmers &amp; Merchants Savings Bank</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstsouthern.html">First Southern Bank</a></td>
+				<td class="city">Batesville</td>
+				<td class="state">AR</td>
+				<td class="cert">58052</td>
+				<td class="ai">Southern Bank</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="unitedamericas.html">United Americas Bank, N.A.</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">35065</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="appalachianga.html">Appalachian Community Bank, FSB</a></td>
+				<td class="city">McCaysville</td>
+				<td class="state">GA</td>
+				<td class="cert">58495</td>
+				<td class="ai">Peoples Bank of East Tennessee</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="chestatee.html">Chestatee State Bank</a></td>
+				<td class="city">Dawsonville</td>
+				<td class="state">GA</td>
+				<td class="cert">34578</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofmiami.html">The Bank of Miami,N.A.</a></td>
+				<td class="city">Coral Gables</td>
+				<td class="state">FL</td>
+				<td class="cert">19040</td>
+				<td class="ai">1st United Bank</td>
+				<td class="closing">December 17, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="earthstar.html">Earthstar Bank</a></td>
+				<td class="city">Southampton</td>
+				<td class="state">PA</td>
+				<td class="cert">35561</td>
+				<td class="ai">Polonia Bank</td>
+				<td class="closing">December 10, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="paramount.html">Paramount Bank</a></td>
+				<td class="city">Farmington Hills</td>
+				<td class="state">MI</td>
+				<td class="cert">34673</td>
+				<td class="ai">Level One Bank</td>
+				<td class="closing">December 10, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstbanking.html">First Banking Center</a></td>
+				<td class="city">Burlington</td>
+				<td class="state">WI</td>
+				<td class="cert">5287</td>
+				<td class="ai">First Michigan Bank</td>
+				<td class="closing">November 19, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="allegbank.html">Allegiance Bank of North America</a></td>
+				<td class="city">Bala Cynwyd</td>
+				<td class="state">PA</td>
+				<td class="cert">35078</td>
+				<td class="ai">VIST Bank</td>
+				<td class="closing">November 19, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="gulfstate.html">Gulf State Community Bank</a></td>
+				<td class="city">Carrabelle</td>
+				<td class="state">FL</td>
+				<td class="cert">20340</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">November 19, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="copperstar.html">Copper Star Bank</a></td>
+				<td class="city">Scottsdale</td>
+				<td class="state">AZ</td>
+				<td class="cert">35463</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">November 12, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="darbybank.html">Darby Bank &amp; Trust Co.</a></td>
+				<td class="city">Vidalia</td>
+				<td class="state">GA</td>
+				<td class="cert">14580</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">November 12, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="tifton.html">Tifton Banking Company</a></td>
+				<td class="city">Tifton</td>
+				<td class="state">GA</td>
+				<td class="cert">57831</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">November 12, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstvietnamese.html">First Vietnamese American Bank</a><br><a href="firstvietnamese_viet.pdf">In Vietnamese</a></td>
+				<td class="city">Westminster</td>
+				<td class="state">CA</td>
+				<td class="cert">57885</td>
+				<td class="ai">Grandpoint Bank</td>
+				<td class="closing">November 5, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="piercecommercial.html">Pierce Commercial Bank</a></td>
+				<td class="city">Tacoma</td>
+				<td class="state">WA</td>
+				<td class="cert">34411</td>
+				<td class="ai">Heritage Bank</td>
+				<td class="closing">November 5, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westerncommercial_ca.html">Western Commercial Bank</a></td>
+				<td class="city">Woodland Hills</td>
+				<td class="state">CA</td>
+				<td class="cert">58087</td>
+				<td class="ai">First California Bank</td>
+				<td class="closing">November 5, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="kbank.html">K Bank</a></td>
+				<td class="city">Randallstown</td>
+				<td class="state">MD</td>
+				<td class="cert">31263</td>
+				<td class="ai">Manufacturers and Traders Trust Company (M&amp;T Bank)</td>
+				<td class="closing">November 5, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstazfsb.html">First Arizona Savings, A FSB</a></td>
+				<td class="city">Scottsdale</td>
+				<td class="state">AZ</td>
+				<td class="cert">32582</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="hillcrest_ks.html">Hillcrest Bank</a></td>
+				<td class="city">Overland Park</td>
+				<td class="state">KS</td>
+				<td class="cert">22173</td>
+				<td class="ai">Hillcrest Bank, N.A.</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstsuburban.html">First Suburban National Bank</a></td>
+				<td class="city">Maywood</td>
+				<td class="state">IL</td>
+				<td class="cert">16089</td>
+				<td class="ai">Seaway Bank and Trust Company</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fnbbarnesville.html">The First National Bank of Barnesville</a></td>
+				<td class="city">Barnesville</td>
+				<td class="state">GA</td>
+				<td class="cert">2119</td>
+				<td class="ai">United Bank</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="gordon.html">The Gordon Bank</a></td>
+				<td class="city">Gordon</td>
+				<td class="state">GA</td>
+				<td class="cert">33904</td>
+				<td class="ai">Morris Bank</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="progress_fl.html">Progress Bank of Florida</a></td>
+				<td class="city">Tampa</td>
+				<td class="state">FL</td>
+				<td class="cert">32251</td>
+				<td class="ai">Bay Cities Bank</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstbankjacksonville.html">First Bank of Jacksonville</a></td>
+				<td class="city">Jacksonville</td>
+				<td class="state">FL</td>
+				<td class="cert">27573</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">October 22, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="premier_mo.html">Premier Bank</a></td>
+				<td class="city">Jefferson City</td>
+				<td class="state">MO</td>
+				<td class="cert">34016</td>
+				<td class="ai">Providence Bank</td>
+				<td class="closing">October 15, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westbridge.html">WestBridge Bank and Trust Company</a></td>
+				<td class="city">Chesterfield</td>
+				<td class="state">MO</td>
+				<td class="cert">58205</td>
+				<td class="ai">Midland States Bank</td>
+				<td class="closing">October 15, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="securitysavingsfsb.html">Security Savings Bank, F.S.B.</a></td>
+				<td class="city">Olathe</td>
+				<td class="state">KS</td>
+				<td class="cert">30898</td>
+				<td class="ai">Simmons First National Bank</td>
+				<td class="closing">October 15, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="shoreline.html">Shoreline Bank</a></td>
+				<td class="city">Shoreline</td>
+				<td class="state">WA</td>
+				<td class="cert">35250</td>
+				<td class="ai">GBC International Bank</td>
+				<td class="closing">October 1, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="wakulla.html">Wakulla Bank</a></td>
+				<td class="city">Crawfordville</td>
+				<td class="state">FL</td>
+				<td class="cert">21777</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">October 1, 2010</td>
+				<td class="updated">November 2, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="northcounty.html">North County Bank</a></td>
+				<td class="city">Arlington</td>
+				<td class="state">WA</td>
+				<td class="cert">35053</td>
+				<td class="ai">Whidbey Island Bank</td>
+				<td class="closing">September 24, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="haventrust_fl.html">Haven Trust Bank Florida</a></td>
+				<td class="city">Ponte Vedra Beach</td>
+				<td class="state">FL</td>
+				<td class="cert">58308</td>
+				<td class="ai">First Southern Bank</td>
+				<td class="closing">September 24, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="maritimesavings.html">Maritime Savings Bank</a></td>
+				<td class="city">West Allis</td>
+				<td class="state">WI</td>
+				<td class="cert">28612</td>
+				<td class="ai">North Shore Bank, FSB</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bramblesavings.html">Bramble Savings Bank</a></td>
+				<td class="city">Milford</td>
+				<td class="state">OH</td>
+				<td class="cert">27808</td>
+				<td class="ai">Foundation Bank</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peoplesbank_ga.html">The Peoples Bank</a></td>
+				<td class="city">Winder</td>
+				<td class="state">GA</td>
+				<td class="cert">182</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcommerce_ga.html">First Commerce Community Bank</a></td>
+				<td class="city">Douglasville</td>
+				<td class="state">GA</td>
+				<td class="cert">57448</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ellijay.html">Bank of Ellijay</a></td>
+				<td class="city">Ellijay</td>
+				<td class="state">GA</td>
+				<td class="cert">58197</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="isnbank.html">ISN Bank</a></td>
+				<td class="city">Cherry Hill</td>
+				<td class="state">NJ</td>
+				<td class="cert">57107</td>
+				<td class="ai">Customers Bank</td>
+				<td class="closing">September 17, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="horizonfl.html">Horizon Bank</a></td>
+				<td class="city">Bradenton</td>
+				<td class="state">FL</td>
+				<td class="cert">35061</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">September 10, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sonoma.html">Sonoma Valley Bank</a></td>
+				<td class="city">Sonoma</td>
+				<td class="state">CA</td>
+				<td class="cert">27259</td>
+				<td class="ai">Westamerica Bank</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lospadres.html">Los Padres Bank</a></td>
+				<td class="city">Solvang</td>
+				<td class="state">CA</td>
+				<td class="cert">32165</td>
+				<td class="ai">Pacific Western Bank</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="butte.html">Butte Community Bank</a></td>
+				<td class="city">Chico</td>
+				<td class="state">CA</td>
+				<td class="cert">33219</td>
+				<td class="ai">Rabobank, N.A.</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pacificbk.html">Pacific State Bank</a></td>
+				<td class="city">Stockton</td>
+				<td class="state">CA</td>
+				<td class="cert">27090</td>
+				<td class="ai">Rabobank, N.A.</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="shorebank.html">ShoreBank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">15640</td>
+				<td class="ai">Urban Partnership Bank</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">May 16, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="imperialsvgs.html">Imperial Savings and Loan Association</a></td>
+				<td class="city">Martinsville</td>
+				<td class="state">VA</td>
+				<td class="cert">31623</td>
+				<td class="ai">River Community Bank, N.A.</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="inatbank.html">Independent National Bank</a></td>
+				<td class="city">Ocala</td>
+				<td class="state">FL</td>
+				<td class="cert">27344</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cnbbartow.html">Community National Bank at Bartow</a></td>
+				<td class="city">Bartow</td>
+				<td class="state">FL</td>
+				<td class="cert">25266</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">August 20, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="palosbank.html">Palos Bank and Trust Company</a></td>
+				<td class="city">Palos Heights</td>
+				<td class="state">IL</td>
+				<td class="cert">17599</td>
+				<td class="ai">First Midwest Bank</td>
+				<td class="closing">August 13, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ravenswood.html">Ravenswood Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">34231</td>
+				<td class="ai">Northbrook Bank &amp; Trust Company</td>
+				<td class="closing">August 6, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="libertyor.html">LibertyBank</a></td>
+				<td class="city">Eugene</td>
+				<td class="state">OR</td>
+				<td class="cert">31964</td>
+				<td class="ai">Home Federal Bank</td>
+				<td class="closing">July 30, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cowlitz.html">The Cowlitz Bank</a></td>
+				<td class="city">Longview</td>
+				<td class="state">WA</td>
+				<td class="cert">22643</td>
+				<td class="ai">Heritage Bank</td>
+				<td class="closing">July 30, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="coastal.html">Coastal Community Bank</a></td>
+				<td class="city">Panama City Beach</td>
+				<td class="state">FL</td>
+				<td class="cert">9619</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">July 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bayside.html">Bayside Savings Bank</a></td>
+				<td class="city">Port Saint Joe</td>
+				<td class="state">FL</td>
+				<td class="cert">57669</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">July 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="northwestga.html">Northwest Bank &amp; Trust</a></td>
+				<td class="city">Acworth</td>
+				<td class="state">GA</td>
+				<td class="cert">57658</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="homevalleyor.html">Home Valley Bank</a></td>
+				<td class="city">Cave Junction</td>
+				<td class="state">OR</td>
+				<td class="cert">23181</td>
+				<td class="ai">South Valley Bank &amp; Trust</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="southwestusanv.html">SouthwestUSA Bank</a></td>
+				<td class="city">Las Vegas</td>
+				<td class="state">NV</td>
+				<td class="cert">35434</td>
+				<td class="ai">Plaza Bank</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="communitysecmn.html">Community Security Bank</a></td>
+				<td class="city">New Prague</td>
+				<td class="state">MN</td>
+				<td class="cert">34486</td>
+				<td class="ai">Roundbank</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">September 12, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="thunderbankks.html">Thunder Bank</a></td>
+				<td class="city">Sylvan Grove</td>
+				<td class="state">KS</td>
+				<td class="cert">10506</td>
+				<td class="ai">The Bennington State Bank</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">September 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="williamsburgsc.html">Williamsburg First National Bank</a></td>
+				<td class="city">Kingstree</td>
+				<td class="state">SC</td>
+				<td class="cert">17837</td>
+				<td class="ai">First Citizens Bank and Trust Company, Inc.</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="crescentga.html">Crescent Bank and Trust Company</a></td>
+				<td class="city">Jasper</td>
+				<td class="state">GA</td>
+				<td class="cert">27559</td>
+				<td class="ai">Renasant Bank</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sterlingfl.html">Sterling Bank</a></td>
+				<td class="city">Lantana</td>
+				<td class="state">FL</td>
+				<td class="cert">32536</td>
+				<td class="ai">IBERIABANK</td>
+				<td class="closing">July 23, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mainstsvgs.html">Mainstreet Savings Bank, FSB</a></td>
+				<td class="city">Hastings</td>
+				<td class="state">MI</td>
+				<td class="cert">28136</td>
+				<td class="ai">Commercial Bank</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">September 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="oldecypress.html">Olde Cypress Community Bank</a></td>
+				<td class="city">Clewiston</td>
+				<td class="state">FL</td>
+				<td class="cert">28864</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="turnberry.html">Turnberry Bank</a></td>
+				<td class="city">Aventura</td>
+				<td class="state">FL</td>
+				<td class="cert">32280</td>
+				<td class="ai">NAFH National Bank</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="metrobankfl.html">Metro Bank of Dade County</a></td>
+				<td class="city">Miami</td>
+				<td class="state">FL</td>
+				<td class="cert">25172</td>
+				<td class="ai">NAFH National Bank</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstnatlsc.html">First National Bank of the South</a></td>
+				<td class="city">Spartanburg</td>
+				<td class="state">SC</td>
+				<td class="cert">35383</td>
+				<td class="ai">NAFH National Bank</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="woodlands.html">Woodlands Bank</a></td>
+				<td class="city">Bluffton</td>
+				<td class="state">SC</td>
+				<td class="cert">32571</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">July 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="homenatlok.html">Home National Bank</a></td>
+				<td class="city">Blackwell</td>
+				<td class="state">OK</td>
+				<td class="cert">11636</td>
+				<td class="ai">RCB Bank</td>
+				<td class="closing">July 9, 2010</td>
+				<td class="updated">December 10, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="usabankny.html">USA Bank</a></td>
+				<td class="city">Port Chester</td>
+				<td class="state">NY</td>
+				<td class="cert">58072</td>
+				<td class="ai">New Century Bank</td>
+				<td class="closing">July 9, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="idealfedsvngsmd.html">Ideal Federal Savings Bank</a></td>
+				<td class="city">Baltimore</td>
+				<td class="state">MD</td>
+				<td class="cert">32456</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">July 9, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="baynatlmd.html">Bay National Bank</a></td>
+				<td class="city">Baltimore</td>
+				<td class="state">MD</td>
+				<td class="cert">35462</td>
+				<td class="ai">Bay Bank, FSB</td>
+				<td class="closing">July 9, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="highdesertnm.html">High Desert State Bank</a></td>
+				<td class="city">Albuquerque</td>
+				<td class="state">NM</td>
+				<td class="cert">35279</td>
+				<td class="ai">First American Bank</td>
+				<td class="closing">June 25, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstnatga.html">First National Bank</a></td>
+				<td class="city">Savannah</td>
+				<td class="state">GA</td>
+				<td class="cert">34152</td>
+				<td class="ai">The Savannah Bank, N.A.</td>
+				<td class="closing">June 25, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peninsulafl.html">Peninsula Bank</a></td>
+				<td class="city">Englewood</td>
+				<td class="state">FL</td>
+				<td class="cert">26563</td>
+				<td class="ai">Premier American Bank, N.A.</td>
+				<td class="closing">June 25, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="nevsecbank.html">Nevada Security Bank</a></td>
+				<td class="city">Reno</td>
+				<td class="state">NV</td>
+				<td class="cert">57110</td>
+				<td class="ai">Umpqua Bank</td>
+				<td class="closing">June 18, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="washfirstintl.html">Washington First International Bank</a></td>
+				<td class="city">Seattle</td>
+				<td class="state">WA</td>
+				<td class="cert">32955</td>
+				<td class="ai">East West Bank</td>
+				<td class="closing">June 11, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="tieronebankne.html">TierOne Bank</a></td>
+				<td class="city">Lincoln</td>
+				<td class="state">NE</td>
+				<td class="cert">29341</td>
+				<td class="ai">Great Western Bank</td>
+				<td class="closing">June 4, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="arcolail.html">Arcola Homestead Savings Bank</a></td>
+				<td class="city">Arcola</td>
+				<td class="state">IL</td>
+				<td class="cert">31813</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">June 4, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstnatms.html">First National Bank</a></td>
+				<td class="city">Rosedale</td>
+				<td class="state">MS</td>
+				<td class="cert">15814</td>
+				<td class="ai">The Jefferson Bank</td>
+				<td class="closing">June 4, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="swbnevada.html">Sun West Bank</a></td>
+				<td class="city">Las Vegas</td>
+				<td class="state">NV</td>
+				<td class="cert">34785</td>
+				<td class="ai">City National Bank</td>
+				<td class="closing">May 28, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="graniteca.html">Granite Community Bank, NA</a></td>
+				<td class="city">Granite Bay</td>
+				<td class="state">CA</td>
+				<td class="cert">57315</td>
+				<td class="ai">Tri Counties Bank</td>
+				<td class="closing">May 28, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankoffloridatb.html">Bank of Florida - Tampa</a></td>
+				<td class="city">Tampa</td>
+				<td class="state">FL</td>
+				<td class="cert">57814</td>
+				<td class="ai">EverBank</td>
+				<td class="closing">May 28, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankoffloridasw.html">Bank of Florida - Southwest</a></td>
+				<td class="city">Naples</td>
+				<td class="state">FL</td>
+				<td class="cert">35106</td>
+				<td class="ai">EverBank</td>
+				<td class="closing">May 28, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankoffloridase.html">Bank of Florida - Southeast</a></td>
+				<td class="city">Fort Lauderdale</td>
+				<td class="state">FL</td>
+				<td class="cert">57360</td>
+				<td class="ai">EverBank</td>
+				<td class="closing">May 28, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pinehurstmn.html">Pinehurst Bank</a></td>
+				<td class="city">Saint Paul</td>
+				<td class="state">MN</td>
+				<td class="cert">57735</td>
+				<td class="ai">Coulee Bank</td>
+				<td class="closing">May 21, 2010</td>
+				<td class="updated">October 26, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="midwestil.html">Midwest Bank and Trust Company</a></td>
+				<td class="city">Elmwood Park</td>
+				<td class="state">IL</td>
+				<td class="cert">18117</td>
+				<td class="ai">FirstMerit Bank, N.A.</td>
+				<td class="closing">May 14, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="swcmntymo.html">Southwest Community Bank</a></td>
+				<td class="city">Springfield</td>
+				<td class="state">MO</td>
+				<td class="cert">34255</td>
+				<td class="ai">Simmons First National Bank</td>
+				<td class="closing">May 14, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newlibertymi.html">New Liberty Bank</a></td>
+				<td class="city">Plymouth</td>
+				<td class="state">MI</td>
+				<td class="cert">35586</td>
+				<td class="ai">Bank of Ann Arbor</td>
+				<td class="closing">May 14, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="satillacmntyga.html">Satilla Community Bank</a></td>
+				<td class="city">Saint Marys</td>
+				<td class="state">GA</td>
+				<td class="cert">35114</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">May 14, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="1stpacific.html">1st Pacific Bank of California</a></td>
+				<td class="city">San Diego</td>
+				<td class="state">CA</td>
+				<td class="cert">35517</td>
+				<td class="ai">City National Bank</td>
+				<td class="closing">May 7, 2010</td>
+				<td class="updated">December 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="townebank.html">Towne Bank of Arizona</a></td>
+				<td class="city">Mesa</td>
+				<td class="state">AZ</td>
+				<td class="cert">57697</td>
+				<td class="ai">Commerce Bank of Arizona</td>
+				<td class="closing">May 7, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="accessbank.html">Access Bank</a></td>
+				<td class="city">Champlin</td>
+				<td class="state">MN</td>
+				<td class="cert">16476</td>
+				<td class="ai">PrinsBank</td>
+				<td class="closing">May 7, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bonifay.html">The Bank of Bonifay</a></td>
+				<td class="city">Bonifay</td>
+				<td class="state">FL</td>
+				<td class="cert">14246</td>
+				<td class="ai">First Federal Bank of Florida</td>
+				<td class="closing">May 7, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="frontier.html">Frontier Bank</a></td>
+				<td class="city">Everett</td>
+				<td class="state">WA</td>
+				<td class="cert">22710</td>
+				<td class="ai">Union Bank, N.A.</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bc-natl.html">BC National Banks</a></td>
+				<td class="city">Butler</td>
+				<td class="state">MO</td>
+				<td class="cert">17792</td>
+				<td class="ai">Community First Bank</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="champion.html">Champion Bank</a></td>
+				<td class="city">Creve Coeur</td>
+				<td class="state">MO</td>
+				<td class="cert">58362</td>
+				<td class="ai">BankLiberty</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cfbancorp.html">CF Bancorp</a></td>
+				<td class="city">Port Huron</td>
+				<td class="state">MI</td>
+				<td class="cert">30005</td>
+				<td class="ai">First Michigan Bank</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westernbank-puertorico.html">Westernbank Puerto Rico</a><br><a href="westernbank-puertorico_spanish.html">En Espanol</a></td>
+				<td class="city">Mayaguez</td>
+				<td class="state">PR</td>
+				<td class="cert">31027</td>
+				<td class="ai">Banco Popular de Puerto Rico</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="r-gpremier-puertorico.html">R-G Premier Bank of Puerto Rico</a><br><a href="r-gpremier-puertorico_spanish.html">En Espanol</a></td>
+				<td class="city">Hato Rey</td>
+				<td class="state">PR</td>
+				<td class="cert">32185</td>
+				<td class="ai">Scotiabank de Puerto Rico</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="eurobank-puertorico.html">Eurobank</a><br><a href="eurobank-puertorico_spanish.html">En Espanol</a></td>
+				<td class="city">San Juan</td>
+				<td class="state">PR</td>
+				<td class="cert">27150</td>
+				<td class="ai">Oriental Bank and Trust</td>
+				<td class="closing">April 30, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="wheatland.html">Wheatland Bank</a></td>
+				<td class="city">Naperville</td>
+				<td class="state">IL</td>
+				<td class="cert">58429</td>
+				<td class="ai">Wheaton Bank &amp; Trust</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peotone.html">Peotone Bank and Trust Company</a></td>
+				<td class="city">Peotone</td>
+				<td class="state">IL</td>
+				<td class="cert">10888</td>
+				<td class="ai">First Midwest Bank</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lincoln-park.html">Lincoln Park Savings Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">30600</td>
+				<td class="ai">Northbrook Bank &amp; Trust Company</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="new-century-il.html">New Century Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">34821</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizens-bank.html">Citizens Bank and Trust Company of Chicago</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">34658</td>
+				<td class="ai">Republic Bank of Chicago</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="broadway.html">Broadway Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">22853</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amcore.html">Amcore Bank, National Association</a></td>
+				<td class="city">Rockford</td>
+				<td class="state">IL</td>
+				<td class="cert">3735</td>
+				<td class="ai">Harris N.A.</td>
+				<td class="closing">April 23, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citybank.html">City Bank</a></td>
+				<td class="city">Lynnwood</td>
+				<td class="state">WA</td>
+				<td class="cert">21521</td>
+				<td class="ai">Whidbey Island Bank</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="tamalpais.html">Tamalpais Bank</a></td>
+				<td class="city">San Rafael</td>
+				<td class="state">CA</td>
+				<td class="cert">33493</td>
+				<td class="ai">Union Bank, N.A.</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="innovative.html">Innovative Bank</a></td>
+				<td class="city">Oakland</td>
+				<td class="state">CA</td>
+				<td class="cert">23876</td>
+				<td class="ai">Center Bank</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="butlerbank.html">Butler Bank</a></td>
+				<td class="city">Lowell</td>
+				<td class="state">MA</td>
+				<td class="cert">26619</td>
+				<td class="ai">People's United Bank</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="riverside-natl.html">Riverside National Bank of Florida</a></td>
+				<td class="city">Fort Pierce</td>
+				<td class="state">FL</td>
+				<td class="cert">24067</td>
+				<td class="ai">TD Bank, N.A.</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americanfirst.html">AmericanFirst Bank</a></td>
+				<td class="city">Clermont</td>
+				<td class="state">FL</td>
+				<td class="cert">57724</td>
+				<td class="ai">TD Bank, N.A.</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ffbnf.html">First Federal Bank of North Florida</a></td>
+				<td class="city">Palatka</td>
+				<td class="state">FL</td>
+				<td class="cert">28886</td>
+				<td class="ai">TD Bank, N.A.</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lakeside-comm.html">Lakeside Community Bank</a></td>
+				<td class="city">Sterling Heights</td>
+				<td class="state">MI</td>
+				<td class="cert">34878</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">April 16, 2010</td>
+				<td class="updated">August 23, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="beachfirst.html">Beach First National Bank</a></td>
+				<td class="city">Myrtle Beach</td>
+				<td class="state">SC</td>
+				<td class="cert">34242</td>
+				<td class="ai">Bank of North Carolina</td>
+				<td class="closing">April 9, 2010</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="deserthills.html">Desert Hills Bank</a></td>
+				<td class="city">Phoenix</td>
+				<td class="state">AZ</td>
+				<td class="cert">57060</td>
+				<td class="ai">New York Community Bank</td>
+				<td class="closing">March 26, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="unity-natl.html">Unity National Bank</a></td>
+				<td class="city">Cartersville</td>
+				<td class="state">GA</td>
+				<td class="cert">34678</td>
+				<td class="ai">Bank of the Ozarks</td>
+				<td class="closing">March 26, 2010</td>
+				<td class="updated">September 14, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="key-west.html">Key West Bank</a></td>
+				<td class="city">Key West</td>
+				<td class="state">FL</td>
+				<td class="cert">34684</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">March 26, 2010</td>
+				<td class="updated">August 23, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="mcintosh.html">McIntosh Commercial Bank</a></td>
+				<td class="city">Carrollton</td>
+				<td class="state">GA</td>
+				<td class="cert">57399</td>
+				<td class="ai">CharterBank</td>
+				<td class="closing">March 26, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="state-aurora.html">State Bank of Aurora</a></td>
+				<td class="city">Aurora</td>
+				<td class="state">MN</td>
+				<td class="cert">8221</td>
+				<td class="ai">Northern State Bank</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstlowndes.html">First Lowndes Bank</a></td>
+				<td class="city">Fort Deposit</td>
+				<td class="state">AL</td>
+				<td class="cert">24957</td>
+				<td class="ai">First Citizens Bank</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofhiawassee.html">Bank of Hiawassee</a></td>
+				<td class="city">Hiawassee</td>
+				<td class="state">GA</td>
+				<td class="cert">10054</td>
+				<td class="ai">Citizens South Bank</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="appalachian.html">Appalachian Community Bank</a></td>
+				<td class="city">Ellijay</td>
+				<td class="state">GA</td>
+				<td class="cert">33989</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">October 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="advanta-ut.html">Advanta Bank Corp.</a></td>
+				<td class="city">Draper</td>
+				<td class="state">UT</td>
+				<td class="cert">33535</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cent-security.html">Century Security Bank</a></td>
+				<td class="city">Duluth</td>
+				<td class="state">GA</td>
+				<td class="cert">58104</td>
+				<td class="ai">Bank of Upson</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amer-natl-oh.html">American National Bank</a></td>
+				<td class="city">Parma</td>
+				<td class="state">OH</td>
+				<td class="cert">18806</td>
+				<td class="ai">The National Bank and Trust Company</td>
+				<td class="closing">March 19, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="statewide.html">Statewide Bank</a></td>
+				<td class="city">Covington</td>
+				<td class="state">LA</td>
+				<td class="cert">29561</td>
+				<td class="ai">Home Bank</td>
+				<td class="closing">March 12, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="oldsouthern.html">Old Southern Bank</a></td>
+				<td class="city">Orlando</td>
+				<td class="state">FL</td>
+				<td class="cert">58182</td>
+				<td class="ai">Centennial Bank</td>
+				<td class="closing">March 12, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="parkavenue-ny.html">The Park Avenue Bank</a></td>
+				<td class="city">New York</td>
+				<td class="state">NY</td>
+				<td class="cert">27096</td>
+				<td class="ai">Valley National Bank</td>
+				<td class="closing">March 12, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="libertypointe.html">LibertyPointe Bank</a></td>
+				<td class="city">New York</td>
+				<td class="state">NY</td>
+				<td class="cert">58071</td>
+				<td class="ai">Valley National Bank</td>
+				<td class="closing">March 11, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="centennial-ut.html">Centennial Bank</a></td>
+				<td class="city">Ogden</td>
+				<td class="state">UT</td>
+				<td class="cert">34430</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 5, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="waterfield.html">Waterfield Bank</a></td>
+				<td class="city">Germantown</td>
+				<td class="state">MD</td>
+				<td class="cert">34976</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 5, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofillinois.html">Bank of Illinois</a></td>
+				<td class="city">Normal</td>
+				<td class="state">IL</td>
+				<td class="cert">9268</td>
+				<td class="ai">Heartland Bank and Trust Company</td>
+				<td class="closing">March 5, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sunamerican.html">Sun American Bank</a></td>
+				<td class="city">Boca Raton</td>
+				<td class="state">FL</td>
+				<td class="cert">27126</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">March 5, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="rainier.html">Rainier Pacific Bank</a></td>
+				<td class="city">Tacoma</td>
+				<td class="state">WA</td>
+				<td class="cert">38129</td>
+				<td class="ai">Umpqua Bank</td>
+				<td class="closing">February 26, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="carsonriver.html">Carson River Community Bank</a></td>
+				<td class="city">Carson City</td>
+				<td class="state">NV</td>
+				<td class="cert">58352</td>
+				<td class="ai">Heritage Bank of Nevada</td>
+				<td class="closing">February 26, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lajolla.html">La Jolla Bank, FSB</a></td>
+				<td class="city">La Jolla</td>
+				<td class="state">CA</td>
+				<td class="cert">32423</td>
+				<td class="ai">OneWest Bank, FSB</td>
+				<td class="closing">February 19, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="georgewashington.html">George Washington Savings Bank</a></td>
+				<td class="city">Orland Park</td>
+				<td class="state">IL</td>
+				<td class="cert">29952</td>
+				<td class="ai">FirstMerit Bank, N.A.</td>
+				<td class="closing">February 19, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lacoste.html">The La Coste National Bank</a></td>
+				<td class="city">La Coste</td>
+				<td class="state">TX</td>
+				<td class="cert">3287</td>
+				<td class="ai">Community National Bank</td>
+				<td class="closing">February 19, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="marco.html">Marco Community Bank</a></td>
+				<td class="city">Marco Island</td>
+				<td class="state">FL</td>
+				<td class="cert">57586</td>
+				<td class="ai">Mutual of Omaha Bank</td>
+				<td class="closing">February 19, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="1stamerican.html">1st American State Bank of Minnesota</a></td>
+				<td class="city">Hancock</td>
+				<td class="state">MN</td>
+				<td class="cert">15448</td>
+				<td class="ai">Community Development Bank, FSB</td>
+				<td class="closing">February 5, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americanmarine.html">American Marine Bank</a></td>
+				<td class="city">Bainbridge Island</td>
+				<td class="state">WA</td>
+				<td class="cert">16730</td>
+				<td class="ai">Columbia State Bank</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstregional.html">First Regional Bank</a></td>
+				<td class="city">Los Angeles</td>
+				<td class="state">CA</td>
+				<td class="cert">23011</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">August 24, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cbt-cornelia.html">Community Bank and Trust</a></td>
+				<td class="city">Cornelia</td>
+				<td class="state">GA</td>
+				<td class="cert">5702</td>
+				<td class="ai">SCBT National Association</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="marshall-mn.html">Marshall Bank, N.A.</a></td>
+				<td class="city">Hallock</td>
+				<td class="state">MN</td>
+				<td class="cert">16133</td>
+				<td class="ai">United Valley Bank</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="floridacommunity.html">Florida Community Bank</a></td>
+				<td class="city">Immokalee</td>
+				<td class="state">FL</td>
+				<td class="cert">5672</td>
+				<td class="ai">Premier American Bank, N.A.</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstnational-carrollton.html">First National Bank of Georgia</a></td>
+				<td class="city">Carrollton</td>
+				<td class="state">GA</td>
+				<td class="cert">16480</td>
+				<td class="ai">Community &amp; Southern Bank</td>
+				<td class="closing">January 29, 2010</td>
+				<td class="updated">December 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="columbiariver.html">Columbia River Bank</a></td>
+				<td class="city">The Dalles</td>
+				<td class="state">OR</td>
+				<td class="cert">22469</td>
+				<td class="ai">Columbia State Bank</td>
+				<td class="closing">January 22, 2010</td>
+				<td class="updated">September 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="evergreen-wa.html">Evergreen Bank</a></td>
+				<td class="city">Seattle</td>
+				<td class="state">WA</td>
+				<td class="cert">20501</td>
+				<td class="ai">Umpqua Bank</td>
+				<td class="closing">January 22, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="charter-nm.html">Charter Bank</a></td>
+				<td class="city">Santa Fe</td>
+				<td class="state">NM</td>
+				<td class="cert">32498</td>
+				<td class="ai">Charter Bank</td>
+				<td class="closing">January 22, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="leeton.html">Bank of Leeton</a></td>
+				<td class="city">Leeton</td>
+				<td class="state">MO</td>
+				<td class="cert">8265</td>
+				<td class="ai">Sunflower Bank, N.A.</td>
+				<td class="closing">January 22, 2010</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="premieramerican.html">Premier American Bank</a></td>
+				<td class="city">Miami</td>
+				<td class="state">FL</td>
+				<td class="cert">57147</td>
+				<td class="ai">Premier American Bank, N.A.</td>
+				<td class="closing">January 22, 2010</td>
+				<td class="updated">December 13, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="barnes.html">Barnes Banking Company</a></td>
+				<td class="city">Kaysville</td>
+				<td class="state">UT</td>
+				<td class="cert">1252</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">January 15, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ststephen.html">St. Stephen State Bank</a></td>
+				<td class="city">St. Stephen</td>
+				<td class="state">MN</td>
+				<td class="cert">17522</td>
+				<td class="ai">First State Bank of St. Joseph</td>
+				<td class="closing">January 15, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="towncommunity.html">Town Community Bank &amp; Trust</a></td>
+				<td class="city">Antioch</td>
+				<td class="state">IL</td>
+				<td class="cert">34705</td>
+				<td class="ai">First American Bank</td>
+				<td class="closing">January 15, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="horizon-wa.html">Horizon Bank</a></td>
+				<td class="city">Bellingham</td>
+				<td class="state">WA</td>
+				<td class="cert">22977</td>
+				<td class="ai">Washington Federal Savings and Loan Association</td>
+				<td class="closing">January 8, 2010</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstfederal-ca.html">First Federal Bank of California, F.S.B.</a></td>
+				<td class="city">Santa Monica</td>
+				<td class="state">CA</td>
+				<td class="cert">28536</td>
+				<td class="ai">OneWest Bank, FSB</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="imperialcapital.html">Imperial Capital Bank</a></td>
+				<td class="city">La Jolla</td>
+				<td class="state">CA</td>
+				<td class="cert">26348</td>
+				<td class="ai">City National Bank</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ibb.html">Independent Bankers' Bank</a></td>
+				<td class="city">Springfield</td>
+				<td class="state">IL</td>
+				<td class="cert">26820</td>
+				<td class="ai">The Independent BankersBank (TIB)</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newsouth.html">New South Federal Savings Bank</a></td>
+				<td class="city">Irondale</td>
+				<td class="state">AL</td>
+				<td class="cert">32276</td>
+				<td class="ai">Beal Bank</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizensstate-mi.html">Citizens State Bank</a></td>
+				<td class="city">New Baltimore</td>
+				<td class="state">MI</td>
+				<td class="cert">1006</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peoplesfirst-fl.html">Peoples First Community Bank</a></td>
+				<td class="city">Panama City</td>
+				<td class="state">FL</td>
+				<td class="cert">32167</td>
+				<td class="ai">Hancock Bank</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="rockbridge.html">RockBridge Commercial Bank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">58315</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">December 18, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="solutions.html">SolutionsBank</a></td>
+				<td class="city">Overland Park</td>
+				<td class="state">KS</td>
+				<td class="cert">4731</td>
+				<td class="ai">Arvest Bank</td>
+				<td class="closing">December 11, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="valleycapital.html">Valley Capital Bank, N.A.</a></td>
+				<td class="city">Mesa</td>
+				<td class="state">AZ</td>
+				<td class="cert">58399</td>
+				<td class="ai">Enterprise Bank &amp; Trust</td>
+				<td class="closing">December 11, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="republicfederal.html">Republic Federal Bank, N.A.</a></td>
+				<td class="city">Miami</td>
+				<td class="state">FL</td>
+				<td class="cert">22846</td>
+				<td class="ai">1st United Bank</td>
+				<td class="closing">December 11, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="atlantic-va.html">Greater Atlantic Bank</a></td>
+				<td class="city">Reston</td>
+				<td class="state">VA</td>
+				<td class="cert">32583</td>
+				<td class="ai">Sonabank</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="benchmark-il.html">Benchmark Bank</a></td>
+				<td class="city">Aurora</td>
+				<td class="state">IL</td>
+				<td class="cert">10440</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">August 23, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amtrust.html">AmTrust Bank</a></td>
+				<td class="city">Cleveland</td>
+				<td class="state">OH</td>
+				<td class="cert">29776</td>
+				<td class="ai">New York Community Bank</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="tattnall.html">The Tattnall Bank</a></td>
+				<td class="city">Reidsville</td>
+				<td class="state">GA</td>
+				<td class="cert">12080</td>
+				<td class="ai">Heritage Bank of the South</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstsecurity.html">First Security National Bank</a></td>
+				<td class="city">Norcross</td>
+				<td class="state">GA</td>
+				<td class="cert">26290</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="buckheadcommunity.html">The Buckhead Community Bank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">34663</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">December 4, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commercesw-fl.html">Commerce Bank of Southwest Florida</a></td>
+				<td class="city">Fort Myers</td>
+				<td class="state">FL</td>
+				<td class="cert">58016</td>
+				<td class="ai">Central Bank</td>
+				<td class="closing">November 20, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pacificcoastnatl.html">Pacific Coast National Bank</a></td>
+				<td class="city">San Clemente</td>
+				<td class="state">CA</td>
+				<td class="cert">57914</td>
+				<td class="ai">Sunwest Bank</td>
+				<td class="closing">November 13, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="orion-fl.html">Orion Bank</a></td>
+				<td class="city">Naples</td>
+				<td class="state">FL</td>
+				<td class="cert">22427</td>
+				<td class="ai">IBERIABANK</td>
+				<td class="closing">November 13, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="centuryfsb.html">Century Bank, F.S.B.</a></td>
+				<td class="city">Sarasota</td>
+				<td class="state">FL</td>
+				<td class="cert">32267</td>
+				<td class="ai">IBERIABANK</td>
+				<td class="closing">November 13, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ucb.html">United Commercial Bank</a></td>
+				<td class="city">San Francisco</td>
+				<td class="state">CA</td>
+				<td class="cert">32469</td>
+				<td class="ai">East West Bank</td>
+				<td class="closing">November 6, 2009</td>
+				<td class="updated">November 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="gateway-mo.html">Gateway Bank of St. Louis</a></td>
+				<td class="city">St. Louis</td>
+				<td class="state">MO</td>
+				<td class="cert">19450</td>
+				<td class="ai">Central Bank of Kansas City</td>
+				<td class="closing">November 6, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="prosperan.html">Prosperan Bank</a></td>
+				<td class="city">Oakdale</td>
+				<td class="state">MN</td>
+				<td class="cert">35074</td>
+				<td class="ai">Alerus Financial, N.A.</td>
+				<td class="closing">November 6, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="homefsb-mi.html">Home Federal Savings Bank</a></td>
+				<td class="city">Detroit</td>
+				<td class="state">MI</td>
+				<td class="cert">30329</td>
+				<td class="ai">Liberty Bank and Trust Company</td>
+				<td class="closing">November 6, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="unitedsecurity-ga.html">United Security Bank</a></td>
+				<td class="city">Sparta</td>
+				<td class="state">GA</td>
+				<td class="cert">22286</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">November 6, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="northhouston-tx.html">North Houston Bank</a></td>
+				<td class="city">Houston</td>
+				<td class="state">TX</td>
+				<td class="cert">18776</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="madisonville-tx.html">Madisonville State Bank</a></td>
+				<td class="city">Madisonville</td>
+				<td class="state">TX</td>
+				<td class="cert">33782</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizens-teague.html">Citizens National Bank</a></td>
+				<td class="city">Teague</td>
+				<td class="state">TX</td>
+				<td class="cert">25222</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="park-il.html">Park National Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">11677</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pacificnational-ca.html">Pacific National Bank</a></td>
+				<td class="city">San Francisco</td>
+				<td class="state">CA</td>
+				<td class="cert">30006</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="calnational.html">California National Bank</a></td>
+				<td class="city">Los Angeles</td>
+				<td class="state">CA</td>
+				<td class="cert">34659</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sandiegonational.html">San Diego National Bank</a></td>
+				<td class="city">San Diego</td>
+				<td class="state">CA</td>
+				<td class="cert">23594</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community-lemont.html">Community Bank of Lemont</a></td>
+				<td class="city">Lemont</td>
+				<td class="state">IL</td>
+				<td class="cert">35291</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankusa-az.html">Bank USA, N.A.</a></td>
+				<td class="city">Phoenix</td>
+				<td class="state">AZ</td>
+				<td class="cert">32218</td>
+				<td class="ai">U.S. Bank N.A.</td>
+				<td class="closing">October 30, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstdupage.html">First DuPage Bank</a></td>
+				<td class="city">Westmont</td>
+				<td class="state">IL</td>
+				<td class="cert">35038</td>
+				<td class="ai">First Midwest Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="riverview-mn.html">Riverview Community Bank</a></td>
+				<td class="city">Otsego</td>
+				<td class="state">MN</td>
+				<td class="cert">57525</td>
+				<td class="ai">Central Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="elmwood.html">Bank of Elmwood</a></td>
+				<td class="city">Racine</td>
+				<td class="state">WI</td>
+				<td class="cert">18321</td>
+				<td class="ai">Tri City National Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="flagship.html">Flagship National Bank</a></td>
+				<td class="city">Bradenton</td>
+				<td class="state">FL</td>
+				<td class="cert">35044</td>
+				<td class="ai">First Federal Bank of Florida</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="hillcrest-fl.html">Hillcrest Bank Florida</a></td>
+				<td class="city">Naples</td>
+				<td class="state">FL</td>
+				<td class="cert">58336</td>
+				<td class="ai">Stonegate Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americanunited.html">American United Bank</a></td>
+				<td class="city">Lawrenceville</td>
+				<td class="state">GA</td>
+				<td class="cert">57794</td>
+				<td class="ai">Ameris Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="partners-fl.html">Partners Bank</a></td>
+				<td class="city">Naples</td>
+				<td class="state">FL</td>
+				<td class="cert">57959</td>
+				<td class="ai">Stonegate Bank</td>
+				<td class="closing">October 23, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sanjoaquin.html">San Joaquin Bank</a></td>
+				<td class="city">Bakersfield</td>
+				<td class="state">CA</td>
+				<td class="cert">23266</td>
+				<td class="ai">Citizens Business Bank</td>
+				<td class="closing">October 16, 2009</td>
+				<td class="updated">August 22, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="scnb-co.html">Southern Colorado National Bank</a></td>
+				<td class="city">Pueblo</td>
+				<td class="state">CO</td>
+				<td class="cert">57263</td>
+				<td class="ai">Legacy Bank</td>
+				<td class="closing">October 2, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="jennings-mn.html">Jennings State Bank</a></td>
+				<td class="city">Spring Grove</td>
+				<td class="state">MN</td>
+				<td class="cert">11416</td>
+				<td class="ai">Central Bank</td>
+				<td class="closing">October 2, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="warren-mi.html">Warren Bank</a></td>
+				<td class="city">Warren</td>
+				<td class="state">MI</td>
+				<td class="cert">34824</td>
+				<td class="ai">The Huntington National Bank</td>
+				<td class="closing">October 2, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="georgian.html">Georgian Bank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">57151</td>
+				<td class="ai">First Citizens Bank and Trust Company, Inc.</td>
+				<td class="closing">September 25, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="irwin-ky.html">Irwin Union Bank, F.S.B.</a></td>
+				<td class="city">Louisville</td>
+				<td class="state">KY</td>
+				<td class="cert">57068</td>
+				<td class="ai">First Financial Bank, N.A.</td>
+				<td class="closing">September 18, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="irwin-in.html">Irwin Union Bank and Trust Company</a></td>
+				<td class="city">Columbus</td>
+				<td class="state">IN</td>
+				<td class="cert">10100</td>
+				<td class="ai">First Financial Bank, N.A.</td>
+				<td class="closing">September 18, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="venture-wa.html">Venture Bank</a></td>
+				<td class="city">Lacey</td>
+				<td class="state">WA</td>
+				<td class="cert">22868</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">September 11, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="brickwell-mn.html">Brickwell Community Bank</a></td>
+				<td class="city">Woodbury</td>
+				<td class="state">MN</td>
+				<td class="cert">57736</td>
+				<td class="ai">CorTrust Bank N.A.</td>
+				<td class="closing">September 11, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="corus.html">Corus Bank, N.A.</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">13693</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">September 11, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firststate-az.html">First State Bank</a></td>
+				<td class="city">Flagstaff</td>
+				<td class="state">AZ</td>
+				<td class="cert">34875</td>
+				<td class="ai">Sunwest Bank</td>
+				<td class="closing">September 4, 2009</td>
+				<td class="updated">January 15, 2013</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="platinum-il.html">Platinum Community Bank</a></td>
+				<td class="city">Rolling Meadows</td>
+				<td class="state">IL</td>
+				<td class="cert">35030</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">September 4, 2009</td>
+				<td class="updated">August 21, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="vantus.html">Vantus Bank</a></td>
+				<td class="city">Sioux City</td>
+				<td class="state">IN</td>
+				<td class="cert">27732</td>
+				<td class="ai">Great Southern Bank</td>
+				<td class="closing">September 4, 2009</td>
+				<td class="updated">August 21, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="inbank.html">InBank</a></td>
+				<td class="city">Oak Forest</td>
+				<td class="state">IL</td>
+				<td class="cert">20203</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">September 4, 2009</td>
+				<td class="updated">August 21, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="firstbankkc-mo.html">First Bank of Kansas City</a></td>
+				<td class="city">Kansas City</td>
+				<td class="state">MO</td>
+				<td class="cert">25231</td>
+				<td class="ai">Great American Bank</td>
+				<td class="closing">September 4, 2009</td>
+				<td class="updated">August 21, 2012</td>
+	  			</tr>
+			<tr>
+				<td class="institution"><a href="affinity-ca.html">Affinity Bank</a></td>
+				<td class="city">Ventura</td>
+				<td class="state">CA</td>
+				<td class="cert">27197</td>
+				<td class="ai">Pacific Western Bank</td>
+				<td class="closing">August 28, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mainstreet-mn.html">Mainstreet Bank</a></td>
+				<td class="city">Forest Lake</td>
+				<td class="state">MN</td>
+				<td class="cert">1909</td>
+				<td class="ai">Central Bank</td>
+				<td class="closing">August 28, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bradford-md.html">Bradford Bank</a></td>
+				<td class="city">Baltimore</td>
+				<td class="state">MD</td>
+				<td class="cert">28312</td>
+				<td class="ai">Manufacturers and Traders Trust Company (M&amp;T Bank)</td>
+				<td class="closing">August 28, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="guaranty-tx.html">Guaranty Bank</a></td>
+				<td class="city">Austin</td>
+				<td class="state">TX</td>
+				<td class="cert">32618</td>
+				<td class="ai">BBVA Compass</td>
+				<td class="closing">August 21, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="capitalsouth.html">CapitalSouth Bank</a></td>
+				<td class="city">Birmingham</td>
+				<td class="state">AL</td>
+				<td class="cert">22130</td>
+				<td class="ai">IBERIABANK</td>
+				<td class="closing">August 21, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="coweta.html">First Coweta Bank</a></td>
+				<td class="city">Newnan</td>
+				<td class="state">GA</td>
+				<td class="cert">57702</td>
+				<td class="ai">United Bank</td>
+				<td class="closing">August 21, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ebank.html">ebank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">34682</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">August 21, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community-nv.html">Community Bank of Nevada</a></td>
+				<td class="city">Las Vegas</td>
+				<td class="state">NV</td>
+				<td class="cert">34043</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">August 14, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community-az.html">Community Bank of Arizona</a></td>
+				<td class="city">Phoenix</td>
+				<td class="state">AZ</td>
+				<td class="cert">57645</td>
+				<td class="ai">MidFirst Bank</td>
+				<td class="closing">August 14, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="union-az.html">Union Bank, National Association</a></td>
+				<td class="city">Gilbert</td>
+				<td class="state">AZ</td>
+				<td class="cert">34485</td>
+				<td class="ai">MidFirst Bank</td>
+				<td class="closing">August 14, 2009</td>
+				<td class="updated">August 21, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="colonial-al.html">Colonial Bank</a></td>
+				<td class="city">Montgomery</td>
+				<td class="state">AL</td>
+				<td class="cert">9609</td>
+				<td class="ai">Branch Banking &amp; Trust Company, (BB&amp;T)</td>
+				<td class="closing">August 14, 2009</td>
+				<td class="updated">September 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="dwelling.html">Dwelling House Savings and Loan Association</a></td>
+				<td class="city">Pittsburgh</td>
+				<td class="state">PA</td>
+				<td class="cert">31559</td>
+				<td class="ai">PNC Bank, N.A.</td>
+				<td class="closing">August 14, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community-prineville.html">Community First Bank</a></td>
+				<td class="city">Prineville</td>
+				<td class="state">OR</td>
+				<td class="cert">23268</td>
+				<td class="ai">Home Federal Bank</td>
+				<td class="closing">August 7, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community-venice.html">Community National Bank of Sarasota County</a></td>
+				<td class="city">Venice</td>
+				<td class="state">FL</td>
+				<td class="cert">27183</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">August 7, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fsb-sarasota.html">First State Bank</a></td>
+				<td class="city">Sarasota</td>
+				<td class="state">FL</td>
+				<td class="cert">27364</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">August 7, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mutual-harvey.html">Mutual Bank</a></td>
+				<td class="city">Harvey</td>
+				<td class="state">IL</td>
+				<td class="cert">18659</td>
+				<td class="ai">United Central Bank</td>
+				<td class="closing">July 31, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americano.html">First BankAmericano</a></td>
+				<td class="city">Elizabeth</td>
+				<td class="state">NJ</td>
+				<td class="cert">34270</td>
+				<td class="ai">Crown Bank</td>
+				<td class="closing">July 31, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="peoplescommunity-oh.html">Peoples Community Bank</a></td>
+				<td class="city">West Chester</td>
+				<td class="state">OH</td>
+				<td class="cert">32288</td>
+				<td class="ai">First Financial Bank, N.A.</td>
+				<td class="closing">July 31, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="integrity-fl.html">Integrity Bank</a></td>
+				<td class="city">Jupiter</td>
+				<td class="state">FL</td>
+				<td class="cert">57604</td>
+				<td class="ai">Stonegate Bank</td>
+				<td class="closing">July 31, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fsb-altus.html">First State Bank of Altus</a></td>
+				<td class="city">Altus</td>
+				<td class="state">OK</td>
+				<td class="cert">9873</td>
+				<td class="ai">Herring Bank</td>
+				<td class="closing">July 31, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-jones.html">Security Bank of Jones County</a></td>
+				<td class="city">Gray</td>
+				<td class="state">GA</td>
+				<td class="cert">8486</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-houston.html">Security Bank of Houston County</a></td>
+				<td class="city">Perry</td>
+				<td class="state">GA</td>
+				<td class="cert">27048</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-bibb.html">Security Bank of Bibb County</a></td>
+				<td class="city">Macon</td>
+				<td class="state">GA</td>
+				<td class="cert">27367</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-metro.html">Security Bank of North Metro</a></td>
+				<td class="city">Woodstock</td>
+				<td class="state">GA</td>
+				<td class="cert">57105</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-fulton.html">Security Bank of North Fulton</a></td>
+				<td class="city">Alpharetta</td>
+				<td class="state">GA</td>
+				<td class="cert">57430</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sb-gwinnett.html">Security Bank of Gwinnett County</a></td>
+				<td class="city">Suwanee</td>
+				<td class="state">GA</td>
+				<td class="cert">57346</td>
+				<td class="ai">State Bank and Trust Company</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="waterford.html">Waterford Village Bank</a></td>
+				<td class="city">Williamsville</td>
+				<td class="state">NY</td>
+				<td class="cert">58065</td>
+				<td class="ai">Evans Bank, N.A.</td>
+				<td class="closing">July 24, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="temecula.html">Temecula Valley Bank</a></td>
+				<td class="city">Temecula</td>
+				<td class="state">CA</td>
+				<td class="cert">34341</td>
+				<td class="ai">First-Citizens Bank &amp; Trust Company</td>
+				<td class="closing">July 17, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="vineyard.html">Vineyard Bank</a></td>
+				<td class="city">Rancho Cucamonga</td>
+				<td class="state">CA</td>
+				<td class="cert">23556</td>
+				<td class="ai">California Bank &amp; Trust</td>
+				<td class="closing">July 17, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankfirst.html">BankFirst</a></td>
+				<td class="city">Sioux Falls</td>
+				<td class="state">SD</td>
+				<td class="cert">34103</td>
+				<td class="ai">Alerus Financial, N.A.</td>
+				<td class="closing">July 17, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="piedmont.html">First Piedmont Bank</a></td>
+				<td class="city">Winder</td>
+				<td class="state">GA</td>
+				<td class="cert">34594</td>
+				<td class="ai">First American Bank and Trust Company</td>
+				<td class="closing">July 17, 2009</td>
+				<td class="updated">January 15, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="wyoming.html">Bank of Wyoming</a></td>
+				<td class="city">Thermopolis</td>
+				<td class="state">WY</td>
+				<td class="cert">22754</td>
+				<td class="ai">Central Bank &amp; Trust</td>
+				<td class="closing">July 10, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="founders.html">Founders Bank</a></td>
+				<td class="city">Worth</td>
+				<td class="state">IL</td>
+				<td class="cert">18390</td>
+				<td class="ai">The PrivateBank and Trust Company</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="millennium.html">Millennium State Bank of Texas</a></td>
+				<td class="city">Dallas</td>
+				<td class="state">TX</td>
+				<td class="cert">57667</td>
+				<td class="ai">State Bank of Texas</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">October 26, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="danville.html">First National Bank of Danville</a></td>
+				<td class="city">Danville</td>
+				<td class="state">IL</td>
+				<td class="cert">3644</td>
+				<td class="ai">First Financial Bank, N.A.</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="elizabeth.html">Elizabeth State Bank</a></td>
+				<td class="city">Elizabeth</td>
+				<td class="state">IL</td>
+				<td class="cert">9262</td>
+				<td class="ai">Galena State Bank and Trust Company</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="rockriver.html">Rock River Bank</a></td>
+				<td class="city">Oregon</td>
+				<td class="state">IL</td>
+				<td class="cert">15302</td>
+				<td class="ai">The Harvard State Bank</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="winchester.html">First State Bank of Winchester</a></td>
+				<td class="city">Winchester</td>
+				<td class="state">IL</td>
+				<td class="cert">11710</td>
+				<td class="ai">The First National Bank of Beardstown</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="warner.html">John Warner Bank</a></td>
+				<td class="city">Clinton</td>
+				<td class="state">IL</td>
+				<td class="cert">12093</td>
+				<td class="ai">State Bank of Lincoln</td>
+				<td class="closing">July 2, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mirae.html">Mirae Bank</a></td>
+				<td class="city">Los Angeles</td>
+				<td class="state">CA</td>
+				<td class="cert">57332</td>
+				<td class="ai">Wilshire State Bank</td>
+				<td class="closing">June 26, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="metropacific.html">MetroPacific Bank</a></td>
+				<td class="city">Irvine</td>
+				<td class="state">CA</td>
+				<td class="cert">57893</td>
+				<td class="ai">Sunwest Bank</td>
+				<td class="closing">June 26, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="horizon.html">Horizon Bank</a></td>
+				<td class="city">Pine City</td>
+				<td class="state">MN</td>
+				<td class="cert">9744</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">June 26, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="neighbor.html">Neighborhood Community Bank</a></td>
+				<td class="city">Newnan</td>
+				<td class="state">GA</td>
+				<td class="cert">35285</td>
+				<td class="ai">CharterBank</td>
+				<td class="closing">June 26, 2009</td>
+				<td class="updated">August 20, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="communityga.html">Community Bank of West Georgia</a></td>
+				<td class="city">Villa Rica</td>
+				<td class="state">GA</td>
+				<td class="cert">57436</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">June 26, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="anthony.html">First National Bank of Anthony</a></td>
+				<td class="city">Anthony</td>
+				<td class="state">KS</td>
+				<td class="cert">4614</td>
+				<td class="ai">Bank of Kansas</td>
+				<td class="closing">June 19, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cooperative.html">Cooperative Bank</a></td>
+				<td class="city">Wilmington</td>
+				<td class="state">NC</td>
+				<td class="cert">27837</td>
+				<td class="ai">First Bank</td>
+				<td class="closing">June 19, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="scb.html">Southern Community Bank</a></td>
+				<td class="city">Fayetteville</td>
+				<td class="state">GA</td>
+				<td class="cert">35251</td>
+				<td class="ai">United Community Bank</td>
+				<td class="closing">June 19, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="lincolnwood.html">Bank of Lincolnwood</a></td>
+				<td class="city">Lincolnwood</td>
+				<td class="state">IL</td>
+				<td class="cert">17309</td>
+				<td class="ai">Republic Bank of Chicago</td>
+				<td class="closing">June 5, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizensnational.html">Citizens National Bank</a></td>
+				<td class="city">Macomb</td>
+				<td class="state">IL</td>
+				<td class="cert">5757</td>
+				<td class="ai">Morton Community Bank</td>
+				<td class="closing">May 22, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="strategiccapital.html">Strategic Capital Bank</a></td>
+				<td class="city">Champaign</td>
+				<td class="state">IL</td>
+				<td class="cert">35175</td>
+				<td class="ai">Midland States Bank</td>
+				<td class="closing">May 22, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankunited.html">BankUnited, FSB</a></td>
+				<td class="city">Coral Gables</td>
+				<td class="state">FL</td>
+				<td class="cert">32247</td>
+				<td class="ai">BankUnited</td>
+				<td class="closing">May 21, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="westsound.html">Westsound Bank</a></td>
+				<td class="city">Bremerton</td>
+				<td class="state">WA</td>
+				<td class="cert">34843</td>
+				<td class="ai">Kitsap Bank</td>
+				<td class="closing">May 8, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="americawest.html">America West Bank</a></td>
+				<td class="city">Layton</td>
+				<td class="state">UT</td>
+				<td class="cert">35461</td>
+				<td class="ai">Cache Valley Bank</td>
+				<td class="closing">May 1, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="citizens.html">Citizens Community Bank</a></td>
+				<td class="city">Ridgewood</td>
+				<td class="state">NJ</td>
+				<td class="cert">57563</td>
+				<td class="ai">North Jersey Community Bank</td>
+				<td class="closing">May 1, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="silverton.html">Silverton Bank, NA</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">26535</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">May 1, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstbankidaho.html">First Bank of Idaho</a></td>
+				<td class="city">Ketchum</td>
+				<td class="state">ID</td>
+				<td class="cert">34396</td>
+				<td class="ai">U.S. Bank, N.A.</td>
+				<td class="closing">April 24, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="beverlyhills.html">First Bank of Beverly Hills</a></td>
+				<td class="city">Calabasas</td>
+				<td class="state">CA</td>
+				<td class="cert">32069</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">April 24, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="michiganheritage.html">Michigan Heritage Bank</a></td>
+				<td class="city">Farmington Hills</td>
+				<td class="state">MI</td>
+				<td class="cert">34369</td>
+				<td class="ai">Level One Bank</td>
+				<td class="closing">April 24, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amsouthern.html">American Southern Bank</a></td>
+				<td class="city">Kennesaw</td>
+				<td class="state">GA</td>
+				<td class="cert">57943</td>
+				<td class="ai">Bank of North Georgia</td>
+				<td class="closing">April 24, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="greatbasin.html">Great Basin Bank of Nevada</a></td>
+				<td class="city">Elko</td>
+				<td class="state">NV</td>
+				<td class="cert">33824</td>
+				<td class="ai">Nevada State Bank</td>
+				<td class="closing">April 17, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amsterling.html">American Sterling Bank</a></td>
+				<td class="city">Sugar Creek</td>
+				<td class="state">MO</td>
+				<td class="cert">8266</td>
+				<td class="ai">Metcalf Bank</td>
+				<td class="closing">April 17, 2009</td>
+				<td class="updated">August 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newfrontier.html">New Frontier Bank</a></td>
+				<td class="city">Greeley</td>
+				<td class="state">CO</td>
+				<td class="cert">34881</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">April 10, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="capefear.html">Cape Fear Bank</a></td>
+				<td class="city">Wilmington</td>
+				<td class="state">NC</td>
+				<td class="cert">34639</td>
+				<td class="ai">First Federal Savings and Loan Association</td>
+				<td class="closing">April 10, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="omni.html">Omni National Bank</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">22238</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 27, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="teambank.html">TeamBank, NA</a></td>
+				<td class="city">Paola</td>
+				<td class="state">KS</td>
+				<td class="cert">4754</td>
+				<td class="ai">Great Southern Bank</td>
+				<td class="closing">March 20, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="coloradonational.html">Colorado National Bank</a></td>
+				<td class="city">Colorado Springs</td>
+				<td class="state">CO</td>
+				<td class="cert">18896</td>
+				<td class="ai">Herring Bank</td>
+				<td class="closing">March 20, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstcity.html">FirstCity Bank</a></td>
+				<td class="city">Stockbridge</td>
+				<td class="state">GA</td>
+				<td class="cert">18243</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 20, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="freedomga.html">Freedom Bank of Georgia</a></td>
+				<td class="city">Commerce</td>
+				<td class="state">GA</td>
+				<td class="cert">57558</td>
+				<td class="ai">Northeast Georgia Bank</td>
+				<td class="closing">March 6, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="securitysavings.html">Security Savings Bank</a></td>
+				<td class="city">Henderson</td>
+				<td class="state">NV</td>
+				<td class="cert">34820</td>
+				<td class="ai">Bank of Nevada</td>
+				<td class="closing">February 27, 2009</td>
+				<td class="updated">September 7, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heritagebank.html">Heritage Community Bank</a></td>
+				<td class="city">Glenwood</td>
+				<td class="state">IL</td>
+				<td class="cert">20078</td>
+				<td class="ai">MB Financial Bank, N.A.</td>
+				<td class="closing">February 27, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="silverfalls.html">Silver Falls Bank</a></td>
+				<td class="city">Silverton</td>
+				<td class="state">OR</td>
+				<td class="cert">35399</td>
+				<td class="ai">Citizens Bank</td>
+				<td class="closing">February 20, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pinnacle.html">Pinnacle Bank of Oregon</a></td>
+				<td class="city">Beaverton</td>
+				<td class="state">OR</td>
+				<td class="cert">57342</td>
+				<td class="ai">Washington Trust Bank of Spokane</td>
+				<td class="closing">February 13, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cornbelt.html">Corn Belt Bank &amp; Trust Co.</a></td>
+				<td class="city">Pittsfield</td>
+				<td class="state">IL</td>
+				<td class="cert">16500</td>
+				<td class="ai">The Carlinville National Bank</td>
+				<td class="closing">February 13, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="riverside.html">Riverside Bank of the Gulf Coast</a></td>
+				<td class="city">Cape Coral</td>
+				<td class="state">FL</td>
+				<td class="cert">34563</td>
+				<td class="ai">TIB Bank</td>
+				<td class="closing">February 13, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sherman.html">Sherman County Bank</a></td>
+				<td class="city">Loup City</td>
+				<td class="state">NE</td>
+				<td class="cert">5431</td>
+				<td class="ai">Heritage Bank</td>
+				<td class="closing">February 13, 2009</td>
+				<td class="updated">August 17, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="county.html">County Bank</a></td>
+				<td class="city">Merced</td>
+				<td class="state">CA</td>
+				<td class="cert">22574</td>
+				<td class="ai">Westamerica Bank</td>
+				<td class="closing">February 6, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="alliance.html">Alliance Bank</a></td>
+				<td class="city">Culver City</td>
+				<td class="state">CA</td>
+				<td class="cert">23124</td>
+				<td class="ai">California Bank &amp; Trust</td>
+				<td class="closing">February 6, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstbank.html">FirstBank Financial Services</a></td>
+				<td class="city">McDonough</td>
+				<td class="state">GA</td>
+				<td class="cert">57017</td>
+				<td class="ai">Regions Bank</td>
+				<td class="closing">February 6, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ocala.html">Ocala National Bank</a></td>
+				<td class="city">Ocala</td>
+				<td class="state">FL</td>
+				<td class="cert">26538</td>
+				<td class="ai">CenterState Bank of Florida, N.A.</td>
+				<td class="closing">January 30, 2009</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="suburban.html">Suburban FSB</a></td>
+				<td class="city">Crofton</td>
+				<td class="state">MD</td>
+				<td class="cert">30763</td>
+				<td class="ai">Bank of Essex</td>
+				<td class="closing">January 30, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="magnet.html">MagnetBank</a></td>
+				<td class="city">Salt Lake City</td>
+				<td class="state">UT</td>
+				<td class="cert">58001</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">January 30, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="centennial.html">1st Centennial Bank</a></td>
+				<td class="city">Redlands</td>
+				<td class="state">CA</td>
+				<td class="cert">33025</td>
+				<td class="ai">First California Bank</td>
+				<td class="closing">January 23, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="clark.html">Bank of Clark County</a></td>
+				<td class="city">Vancouver</td>
+				<td class="state">WA</td>
+				<td class="cert">34959</td>
+				<td class="ai">Umpqua Bank</td>
+				<td class="closing">January 16, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="commerce.html">National Bank of Commerce</a></td>
+				<td class="city">Berkeley</td>
+				<td class="state">IL</td>
+				<td class="cert">19733</td>
+				<td class="ai">Republic Bank of Chicago</td>
+				<td class="closing">January 16, 2009</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sanderson.html">Sanderson State Bank</a><br><a href="sanderson_spanish.html">En Espanol</a></td>
+				<td class="city">Sanderson</td>
+				<td class="state">TX</td>
+				<td class="cert">11568</td>
+				<td class="ai">The Pecos County State Bank</td>
+				<td class="closing">December 12, 2008</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="haventrust.html">Haven Trust Bank</a></td>
+				<td class="city">Duluth</td>
+				<td class="state">GA</td>
+				<td class="cert">35379</td>
+				<td class="ai">Branch Banking &amp; Trust Company, (BB&amp;T)</td>
+				<td class="closing">December 12, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstga.html">First Georgia Community Bank</a></td>
+				<td class="city">Jackson</td>
+				<td class="state">GA</td>
+				<td class="cert">34301</td>
+				<td class="ai">United Bank</td>
+				<td class="closing">December 5, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pff.html">PFF Bank &amp; Trust</a></td>
+				<td class="city">Pomona</td>
+				<td class="state">CA</td>
+				<td class="cert">28344</td>
+				<td class="ai">U.S. Bank, N.A.</td>
+				<td class="closing">November 21, 2008</td>
+				<td class="updated">January 4, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="downey.html">Downey Savings &amp; Loan</a></td>
+				<td class="city">Newport Beach</td>
+				<td class="state">CA</td>
+				<td class="cert">30968</td>
+				<td class="ai">U.S. Bank, N.A.</td>
+				<td class="closing">November 21, 2008</td>
+				<td class="updated">January 4, 2013</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="community.html">Community Bank</a></td>
+				<td class="city">Loganville</td>
+				<td class="state">GA</td>
+				<td class="cert">16490</td>
+				<td class="ai">Bank of Essex</td>
+				<td class="closing">November 21, 2008</td>
+				<td class="updated">September 4, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="securitypacific.html">Security Pacific Bank</a></td>
+				<td class="city">Los Angeles</td>
+				<td class="state">CA</td>
+				<td class="cert">23595</td>
+				<td class="ai">Pacific Western Bank</td>
+				<td class="closing">November 7, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="franklinbank.html">Franklin Bank, SSB</a></td>
+				<td class="city">Houston</td>
+				<td class="state">TX</td>
+				<td class="cert">26870</td>
+				<td class="ai">Prosperity Bank</td>
+				<td class="closing">November 7, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="freedom.html">Freedom Bank</a></td>
+				<td class="city">Bradenton</td>
+				<td class="state">FL</td>
+				<td class="cert">57930</td>
+				<td class="ai">Fifth Third Bank</td>
+				<td class="closing">October 31, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="alpha.html">Alpha Bank &amp; Trust</a></td>
+				<td class="city">Alpharetta</td>
+				<td class="state">GA</td>
+				<td class="cert">58241</td>
+				<td class="ai">Stearns Bank, N.A.</td>
+				<td class="closing">October 24, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="meridian.html">Meridian Bank</a></td>
+				<td class="city">Eldred</td>
+				<td class="state">IL</td>
+				<td class="cert">13789</td>
+				<td class="ai">National Bank</td>
+				<td class="closing">October 10, 2008</td>
+				<td class="updated">May 31, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="mainstreet.html">Main Street Bank</a></td>
+				<td class="city">Northville</td>
+				<td class="state">MI</td>
+				<td class="cert">57654</td>
+				<td class="ai">Monroe Bank &amp; Trust</td>
+				<td class="closing">October 10, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="wamu.html">Washington Mutual Bank<br>(Including its subsidiary Washington Mutual Bank FSB)</a></td>
+				<td class="city">Henderson</td>
+				<td class="state">NV</td>
+				<td class="cert">32633</td>
+				<td class="ai">JP Morgan Chase Bank</td>
+				<td class="closing">September 25, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ameribank.html">Ameribank</a></td>
+				<td class="city">Northfork</td>
+				<td class="state">WV</td>
+				<td class="cert">6782</td>
+				<td class="ai">The Citizens Savings Bank<br><br>Pioneer Community Bank, Inc.</td>
+				<td class="closing">September 19, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="silverstate.html">Silver State Bank</a><br><a href="silverstatesp.html">En Espanol</a></td>
+				<td class="city">Henderson</td>
+				<td class="state">NV</td>
+				<td class="cert">34194</td>
+				<td class="ai">Nevada State Bank</td>
+				<td class="closing">September 5, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="integrity.html">Integrity Bank</a></td>
+				<td class="city">Alpharetta</td>
+				<td class="state">GA</td>
+				<td class="cert">35469</td>
+				<td class="ai">Regions Bank</td>
+				<td class="closing">August 29, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="columbian.html">Columbian Bank &amp; Trust</a></td>
+				<td class="city">Topeka</td>
+				<td class="state">KS</td>
+				<td class="cert">22728</td>
+				<td class="ai">Citizens Bank &amp; Trust</td>
+				<td class="closing">August 22, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstprioritybank.html">First Priority Bank</a></td>
+				<td class="city">Bradenton</td>
+				<td class="state">FL</td>
+				<td class="cert">57523</td>
+				<td class="ai">SunTrust Bank</td>
+				<td class="closing">August 1, 2008</td>
+				<td class="updated">August 16, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="heritage.html">First Heritage Bank, NA</a></td>
+				<td class="city">Newport Beach</td>
+				<td class="state">CA</td>
+				<td class="cert">57961</td>
+				<td class="ai">Mutual of Omaha Bank</td>
+				<td class="closing">July 25, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="fnbnv.html">First National Bank of Nevada</a></td>
+				<td class="city">Reno</td>
+				<td class="state">NV</td>
+				<td class="cert">27011</td>
+				<td class="ai">Mutual of Omaha Bank</td>
+				<td class="closing">July 25, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="IndyMac.html">IndyMac Bank</a></td>
+				<td class="city">Pasadena</td>
+				<td class="state">CA</td>
+				<td class="cert">29730</td>
+				<td class="ai">OneWest Bank, FSB</td>
+				<td class="closing">July 11, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="first_integrity_bank.html">First Integrity Bank, NA</a></td>
+				<td class="city">Staples</td>
+				<td class="state">MN</td>
+				<td class="cert">12736</td>
+				<td class="ai">First International Bank and Trust</td>
+				<td class="closing">May 30, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="anb.html">ANB Financial, NA</a></td>
+				<td class="city">Bentonville</td>
+				<td class="state">AR</td>
+				<td class="cert">33901</td>
+				<td class="ai">Pulaski Bank and Trust Company</td>
+				<td class="closing">May 9, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="Hume.html">Hume Bank</a></td>
+				<td class="city">Hume</td>
+				<td class="state">MO</td>
+				<td class="cert">1971</td>
+				<td class="ai">Security Bank</td>
+				<td class="closing">March 7, 2008</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="Douglass.html">Douglass National Bank</a></td>
+				<td class="city">Kansas City</td>
+				<td class="state">MO</td>
+				<td class="cert">24660</td>
+				<td class="ai">Liberty Bank and Trust Company</td>
+				<td class="closing">January 25, 2008</td>
+				<td class="updated">October 26, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="MiamiValley.html">Miami Valley Bank</a></td>
+				<td class="city">Lakeview</td>
+				<td class="state">OH</td>
+				<td class="cert">16848</td>
+				<td class="ai">The Citizens Banking Company</td>
+				<td class="closing">October 4, 2007</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="NetBank.html">NetBank</a></td>
+				<td class="city">Alpharetta</td>
+				<td class="state">GA</td>
+				<td class="cert">32575</td>
+				<td class="ai">ING DIRECT</td>
+				<td class="closing">September 28, 2007</td>
+				<td class="updated">August 28, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="MetropolitanSB.html">Metropolitan Savings Bank</a></td>
+				<td class="city">Pittsburgh</td>
+				<td class="state">PA</td>
+				<td class="cert">35353</td>
+				<td class="ai">Allegheny Valley Bank of Pittsburgh</td>
+				<td class="closing">February 2, 2007</td>
+				<td class="updated">October 27, 2010</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="ephraim.html">Bank of Ephraim</a></td>
+				<td class="city">Ephraim</td>
+				<td class="state">UT</td>
+				<td class="cert">1249</td>
+				<td class="ai">Far West Bank</td>
+				<td class="closing">June 25, 2004</td>
+				<td class="updated">April 9, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="reliance.html">Reliance Bank</a></td>
+				<td class="city">White Plains</td>
+				<td class="state">NY</td>
+				<td class="cert">26778</td>
+				<td class="ai">Union State Bank</td>
+				<td class="closing">March 19, 2004</td>
+				<td class="updated">April 9, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="gnb.html">Guaranty National Bank of Tallahassee</a></td>
+				<td class="city">Tallahassee</td>
+				<td class="state">FL</td>
+				<td class="cert">26838</td>
+				<td class="ai">Hancock Bank of Florida</td>
+				<td class="closing">March 12, 2004</td>
+				<td class="updated">June 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="dollar.html">Dollar Savings Bank</a></td>
+				<td class="city">Newark</td>
+				<td class="state">NJ</td>
+				<td class="cert">31330</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">February 14, 2004</td>
+				<td class="updated">April 9, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="pulaski.html">Pulaski Savings Bank</a></td>
+				<td class="city">Philadelphia</td>
+				<td class="state">PA</td>
+				<td class="cert">27203</td>
+				<td class="ai">Earthstar Bank</td>
+				<td class="closing">November 14, 2003</td>
+				<td class="updated">July 22, 2005</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="blanchardville.html">First National Bank of Blanchardville</a></td>
+				<td class="city">Blanchardville</td>
+				<td class="state">WI</td>
+				<td class="cert">11639</td>
+				<td class="ai">The Park Bank</td>
+				<td class="closing">May 9, 2003</td>
+				<td class="updated">June 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="spbank.html">Southern Pacific Bank</a></td>
+				<td class="city">Torrance</td>
+				<td class="state">CA</td>
+				<td class="cert">27094</td>
+				<td class="ai">Beal Bank</td>
+				<td class="closing">February 7, 2003</td>
+				<td class="updated">October 20, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="farmers.html">Farmers Bank of Cheneyville</a></td>
+				<td class="city">Cheneyville</td>
+				<td class="state">LA</td>
+				<td class="cert">16445</td>
+				<td class="ai">Sabine State Bank &amp; Trust</td>
+				<td class="closing">December 17, 2002</td>
+				<td class="updated">October 20, 2004</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="bankofalamo.html">Bank of Alamo</a></td>
+				<td class="city">Alamo</td>
+				<td class="state">TN</td>
+				<td class="cert">9961</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">November 8, 2002</td>
+				<td class="updated">March 18, 2005</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="amtrade.html">AmTrade International Bank</a><br><a href="amtrade-spanish.html">En Espanol</a></td>
+				<td class="city">Atlanta</td>
+				<td class="state">GA</td>
+				<td class="cert">33784</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">September 30, 2002</td>
+				<td class="updated">September 11, 2006</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="universal.html">Universal Federal Savings Bank</a></td>
+				<td class="city">Chicago</td>
+				<td class="state">IL</td>
+				<td class="cert">29355</td>
+				<td class="ai">Chicago Community Bank</td>
+				<td class="closing">June 27, 2002</td>
+				<td class="updated">April 9, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="cbc.html">Connecticut Bank of Commerce</a></td>
+				<td class="city">Stamford</td>
+				<td class="state">CT</td>
+				<td class="cert">19183</td>
+				<td class="ai">Hudson United Bank</td>
+				<td class="closing">June 26, 2002</td>
+				<td class="updated">February 14, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="newcentury.html">New Century Bank</a></td>
+				<td class="city">Shelby Township</td>
+				<td class="state">MI</td>
+				<td class="cert">34979</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">March 28, 2002</td>
+				<td class="updated">March 18, 2005</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="netfirst.html">Net 1st National Bank</a></td>
+				<td class="city">Boca Raton</td>
+				<td class="state">FL</td>
+				<td class="cert">26652</td>
+				<td class="ai">Bank Leumi USA</td>
+				<td class="closing">March 1, 2002</td>
+				<td class="updated">April 9, 2008</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="nextbank.html">NextBank, NA</a></td>
+				<td class="city">Phoenix</td>
+				<td class="state">AZ</td>
+				<td class="cert">22314</td>
+				<td class="ai">No Acquirer</td>
+				<td class="closing">February 7, 2002</td>
+				<td class="updated">August 27, 2010</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="Oakwood.html">Oakwood Deposit Bank Co.</a></td>
+				<td class="city">Oakwood</td>
+				<td class="state">OH</td>
+				<td class="cert">8966</td>
+				<td class="ai">The State Bank &amp; Trust Company</td>
+				<td class="closing">February 1, 2002</td>
+				<td class="updated">October 25, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sierrablanca.html">Bank of Sierra Blanca</a></td>
+				<td class="city">Sierra Blanca</td>
+				<td class="state">TX</td>
+				<td class="cert">22002</td>
+				<td class="ai">The Security State Bank of Pecos</td>
+				<td class="closing">January 18, 2002</td>
+				<td class="updated">November 6, 2003</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="hamilton.html">Hamilton Bank, NA</a><br><a href="hamilton-spanish.html">En Espanol</a></td>
+				<td class="city">Miami</td>
+				<td class="state">FL</td>
+				<td class="cert">24382</td>
+				<td class="ai">Israel Discount Bank of New York</td>
+				<td class="closing">January 11, 2002</td>
+				<td class="updated">June 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="sinclair.html">Sinclair National Bank</a></td>
+				<td class="city">Gravette</td>
+				<td class="state">AR</td>
+				<td class="cert">34248</td>
+				<td class="ai">Delta Trust &amp; Bank</td>
+				<td class="closing">September 7, 2001</td>
+				<td class="updated">February 10, 2004</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="superior.html">Superior Bank, FSB</a></td>
+				<td class="city">Hinsdale</td>
+				<td class="state">IL</td>
+				<td class="cert">32646</td>
+				<td class="ai">Superior Federal, FSB</td>
+				<td class="closing">July 27, 2001</td>
+				<td class="updated">June 5, 2012</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="Malta.html">Malta National Bank</a></td>
+				<td class="city">Malta</td>
+				<td class="state">OH</td>
+				<td class="cert">6629</td>
+				<td class="ai">North Valley Bank</td>
+				<td class="closing">May 3, 2001</td>
+				<td class="updated">November 18, 2002</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="firstalliance.html">First Alliance Bank &amp; Trust Co.</a></td>
+				<td class="city">Manchester</td>
+				<td class="state">NH</td>
+				<td class="cert">34264</td>
+				<td class="ai">Southern New Hampshire Bank &amp; Trust</td>
+				<td class="closing">February 2, 2001</td>
+				<td class="updated">February 18, 2003</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="nsb.html">National State Bank of Metropolis</a></td>
+				<td class="city">Metropolis</td>
+				<td class="state">IL</td>
+				<td class="cert">3815</td>
+				<td class="ai">Banterra Bank of Marion</td>
+				<td class="closing">December 14, 2000</td>
+				<td class="updated">March 17, 2005</td>
+			</tr>
+			<tr>
+				<td class="institution"><a href="boh.html">Bank of Honolulu</a></td>
+				<td class="city">Honolulu</td>
+				<td class="state">HI</td>
+				<td class="cert">21029</td>
+				<td class="ai">Bank of the Orient</td>
+				<td class="closing">October 13, 2000</td>
+				<td class="updated">March 17, 2005</td>
+			</tr>
+		</tbody>
+	</table>
+	</div>
+
+</div>
+<div id="page_foot">
+	<div class="date">Last Updated 05/31/2013</div>
+	<div class="email"><a href="mailto:cservicefdicdal@fdic.gov">cservicefdicdal@fdic.gov</a></div>
+	<div class="clear"></div>
+</div>
+
+<!-- START of Footer -->
+<footer>
+<link rel="stylesheet" type="text/css" href="/responsive/footer/css/footer.css" />
+<div id="responsive_footer">
+	<div id="responsive_footer-full">
+		<ul>
+			<li><a href="/" title="Home">Home</a></li>
+			<li><a href="/about/contact/ask/" title="Contact Us">Contact Us</a></li>
+			<li><a href="/search/" title="Search">Search</a></li>
+			<li><a href="/help/" title="Help">Help</a></li>
+			<li><a href="/sitemap/" title="SiteMap">SiteMap</a></li>
+			<li><a href="/regulations/laws/forms/" title="Forms">Forms</a></li>
+			<li><a href="/quicklinks/spanish.html" title="En Espa&ntilde;ol">En Espa&ntilde;ol</a></li>
+		</ul>
+		<hr>
+		<ul>
+			<li><a href="/about/policies/" title="Website Policies">Website Policies</a></li>
+			<li><a href="/about/privacy/policy/" title="Privacy Policy">Privacy Policy</a></li>
+			<li><a href="/plainlanguage/" title="Privacy Policy">Plain Writing Act of 2010 </a></li>
+			<li><a href="http://www.usa.gov/" title="USA.gov">USA.gov</a></li>
+			<li><a href="http://www.fdicoig.gov/" title="FDIC Office of Inspector General">FDIC Office of Inspector General</a></li>
+		</ul>
+		<hr>
+		<ul>
+			<li><a href="/about/freedom/" title="Freedom of Information Act (FOIA) Service Center">Freedom of Information Act (FOIA) Service Center</a></li>
+			<li><a href="/open/" title="FDIC Open Government Webpage">FDIC Open Government Webpage</a></li>
+			<li><a href="/about/diversity/nofear/" title="No FEAR Act Data">No FEAR Act Data</a></li>
+		</ul>
+	</div>
+	<div id="responsive_footer-small">
+		<ul>
+			<li><a href="/" title="Home">Home</a></li>
+			<li><a href="/about/contact/ask/" title="Contact Us">Contact Us</a></li>
+			<li><a href="/about/policies/" title="Website Policies">Website Policies</a></li>
+			<li><a href="/search/" title="Search">Search</a></li>
+		</ul>
+    </div>
+</div>
+</footer>
+<!-- START Omniture SiteCatalyst Code -->
+<script language="JavaScript" type="text/javascript" src="/js/s_code_v1.js"></script>
+<script type="text/javascript">
+/************* DO NOT ALTER ANYTHING BELOW THIS LINE ! **************/
+var s_code=s.t();if(s_code)document.write(s_code)</script>
+<script type="text/javascript">
+if(navigator.appVersion.indexOf('MSIE')>=0)document.write(unescape('%3C')+'\!-'+'-')
+</script>
+<noscript>
+<a href="http://www.omniture.com" title="Web Analytics">
+<img src="http://fdic.122.2o7.net/b/ss/fdicgovprod/1/H.21--NS/0?[AQB]%26cl=Session%26AQE" height="1" width="1" border="0" alt="" /></a></li>
+</noscript>
+<!--/DO NOT REMOVE/-->
+<!-- END Omniture SiteCatalyst Code -->
+<!-- END of Footer -->
+
+<script type="text/javascript" src="/responsive/js/jquery.tablesorter.js"></script>
+<script type="text/javascript" src="banklist.js"></script>
+
+</body>
+</html>
diff --git a/pandas/pandas/tests/io/data/html/spam.html b/pandas/pandas/tests/io/data/html/spam.html
new file mode 100644
index 0000000000000..a8e445ff1e176
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html/spam.html
@@ -0,0 +1,797 @@
+
+<!DOCTYPE html>
+<!--[if lt IE 7 ]> <html lang="en" class="no-js ie6"> <![endif]-->
+<!--[if IE 7 ]>    <html lang="en" class="no-js ie7"> <![endif]-->
+<!--[if IE 8 ]>    <html lang="en" class="no-js ie8"> <![endif]-->
+<!--[if IE 9 ]>    <html lang="en" class="no-js ie9"> <![endif]-->
+<!--[if (gt IE 9)|!(IE)]><!--> <html lang="en" class="no-js"><!--<![endif]-->
+
+
+<html>
+    <head>
+
+        <title>Show Foods</title>
+        <link rel="shortcut icon" href="/ndb/static/images/favicon.ico" type="image/x-icon" />
+
+
+
+
+
+
+         <link rel='stylesheet' type='text/css' href='/ndb/plugins/richui-0.8/css/autocomplete.css' />
+<script type='text/javascript' src='/ndb/plugins/richui-0.8/js/yui/yahoo-dom-event/yahoo-dom-event.js'></script>
+<script type='text/javascript' src='/ndb/plugins/richui-0.8/js/yui/connection/connection-min.js'></script>
+<script type='text/javascript' src='/ndb/plugins/richui-0.8/js/yui/datasource/datasource-min.js'></script>
+<script type='text/javascript' src='/ndb/plugins/richui-0.8/js/yui/animation/animation-min.js'></script>
+<script type='text/javascript' src='/ndb/plugins/richui-0.8/js/yui/autocomplete/autocomplete-min.js'></script>
+
+<link rel="stylesheet" href="/ndb/static/css/main.css" />
+
+      	<script type="text/JavaScript">
+		 var _gaq = _gaq || [];
+		 // NAL
+		  _gaq.push(['_setAccount', 'UA-28627214-1']);
+		  _gaq.push(['_setDomainName', 'nal.usda.gov']);
+		  _gaq.push(['_setAllowLinker', true]);
+		  _gaq.push(['_trackPageview']);
+		 //
+		// _gaq.push(['_setAccount', 'UA-3876418-1']);
+		//  _gaq.push(['_trackPageview']);
+		  // for NDB
+		  _gaq.push(['_setAccount', 'UA-36442725-1']);
+		  _gaq.push(['_trackPageview']);
+			// USDA servers
+		  _gaq.push(['_setAccount', 'UA-466807-3']);
+		  _gaq.push(['_setDomainName', 'usda.gov']);
+		  _gaq.push(['_setAllowLinker', true]);
+		  _gaq.push(['_trackPageview']);
+		  //
+		  _gaq.push(['a._setAccount', 'UA-27627304-18']);
+		  _gaq.push(['a._setDomainName', 'usda.gov']);
+		  _gaq.push(['a._setAllowLinker', true]);
+		  _gaq.push(['a._trackPageview']);
+			//
+		  _gaq.push(['b._setAccount', 'UA-27627304-1']);
+		  _gaq.push(['b._setDomainName', 'usda.gov']);
+		  _gaq.push(['b._setAllowLinker', true]);
+		  _gaq.push(['b._trackPageview']);
+
+		  (function() {
+		    var ga = document.createElement('script'); ga.type =
+		'text/javascript'; ga.async = true;
+		    ga.src = ('https:' == document.location.protocol ? 'https://ssl' :
+		'http://www') + '.google-analytics.com/ga.js';
+		    var s = document.getElementsByTagName('script')[0];
+		s.parentNode.insertBefore(ga, s);
+		  })();
+	</script>
+
+
+
+        <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
+        <meta name="layout" content="main"/>
+
+
+
+
+
+
+
+          <script src="/ndb/static/plugins/yui-2.8.2.1/js/yui/yahoo-dom-event/yahoo-dom-event.js" type="text/javascript" ></script>
+<script src="/ndb/static/plugins/yui-2.8.2.1/js/yui/element/element-min.js" type="text/javascript" ></script>
+<script src="/ndb/static/plugins/yui-2.8.2.1/js/yui/animation/animation-min.js" type="text/javascript" ></script>
+<script src="/ndb/static/plugins/yui-2.8.2.1/js/yui/connection/connection-min.js" type="text/javascript" ></script>
+<script src="/ndb/static/plugins/yui-2.8.2.1/js/yui/dragdrop/dragdrop-min.js" type="text/javascript" ></script>
+<script src="/ndb/static/bundle-bundle_yui-container_head.js" type="text/javascript" ></script>
+<link href="/ndb/static/bundle-bundle_yui-container_head.css" type="text/css" rel="stylesheet" media="screen, projection" />
+<link href="/ndb/static/plugins/yui-2.8.2.1/js/yui/fonts/fonts-min.css" type="text/css" rel="stylesheet" media="screen, projection" />
+<script src="/ndb/static/plugins/grails-ui-1.2.3/js/grailsui/grailsui.js" type="text/javascript" ></script>
+<link href="/ndb/static/plugins/jquery-ui-1.8.24/jquery-ui/themes/ui-lightness/jquery-ui-1.8.24.custom.css" type="text/css" rel="stylesheet" media="screen, projection" />
+<script src="/ndb/static/plugins/jquery-1.8.0/js/jquery/jquery-1.8.0.min.js" type="text/javascript" ></script>
+<script src="/ndb/static/plugins/jquery-ui-1.8.24/jquery-ui/js/jquery-ui-1.8.24.custom.min.js" type="text/javascript" ></script>
+<script src="/ndb/static/bundle-bundle_yui-menu_head.js" type="text/javascript" ></script>
+<link href="/ndb/static/bundle-bundle_yui-menu_head.css" type="text/css" rel="stylesheet" media="screen, projection" />
+<script src="/ndb/static/bundle-bundle_yui-button_head.js" type="text/javascript" ></script>
+<link href="/ndb/static/bundle-bundle_yui-button_head.css" type="text/css" rel="stylesheet" media="screen, projection" />
+
+    </head>
+   <body class="yui-skin-sam">
+   <div class="section clearfix" >
+    	<div id="name-and-slogan" style="padding-left:15px;" >
+         <a href="http://www.ars.usda.gov"><img id="masthead-map" usemap="#masthead-map" src="/ndb/static/images/masthead.jpg" alt="National Nutrient Database" border="0" /></a>
+         <map id="masthead-map" name="masthead-map">
+<area shape="rect" coords="4,2,54,52" href="http://www.usda.gov" alt="" title="USDA Website"    />
+<area shape="rect" coords="66,1,128,49" href="http://www.ars.usda.gov" alt="" title="Agricultural Research Service Website"    />
+<area shape="rect" coords="127,0,336,50" href="http://www.nal.usda.gov" alt="" title="National Agricultural Library Website"    />
+<area shape="rect" coords="470,2,679,52" href="http://www.ars.usda.gov/main/site_main.htm?modecode=12-35-45-00" alt="" title="Nutrient Data Laboratory Website"    />
+<area shape="rect" coords="702,6,742,47" href="http://fnic.nal.usda.gov" alt="" title="Food and Nutrition Information Center Website"    />
+</map>
+
+
+    </div>
+
+
+
+        </div>
+        	<div id='site-slogan'  align='left'>
+				National Nutrient Database for Standard Reference<br>Release 25
+			</div>
+			<div class="bodywrapper">
+
+		 <div class="nav">
+          <span class="menuButton"> <a href="http://www.ars.usda.gov/main/site_main.htm?modecode=12-35-45-00" class="home" title="Go the NDL home page">NDL Home</a></span>
+        <span class="menuButton"><a href="/ndb/search/list" class="list" name="menu-advanced" title="Browse the foods list">Foods List</a></span>
+        <span class="menuButton"><a href="/ndb/beef/show" class="calc" title="Use the ground beef calculator">Ground Beef Calculator</a></span>
+        <span class="menuButton"><a href="http://www.ars.usda.gov/SP2UserFiles/Place/12354500/Data/SR25/sr25_doc.pdf" class="docs" title="View and download release documentation" target="_help">SR25 Documentation</a></span>
+        <span class="menuButton"><a href="/ndb/help/index" class="help" target="_help" title="Read help on how to use the website">Help</a></span>
+       </div>
+
+
+
+
+
+     <div id="view-name">Basic Report</div>
+
+        <div class="body">
+            <h1>Nutrient data for 07908, Luncheon meat, pork with ham, minced, canned, includes SPAM (Hormel)
+
+
+              </h1>
+             <div class="menuButton" >
+            <a href="/ndb/search/list?fg=&amp;man=&amp;lfacet=&amp;count=&amp;max=25&amp;sort=&amp;qlookup=spam&amp;offset=&amp;format=Abridged&amp;new=" name="search" class="previous" title="Return to results list">Return to Search Results</a><script type='text/javascript'> var myTooltip = new YAHOO.widget.Tooltip("myTooltip", { context:"null" } );</script>
+
+
+
+            	<a href="/ndb/foods/show/1732?fg=&amp;man=&amp;lfacet=&amp;count=&amp;max=25&amp;sort=&amp;qlookup=spam&amp;offset=&amp;format=Full&amp;new=" name="full" title="View Full Report">Full Report (All Nutrients)</a><script type='text/javascript'> var myTooltip = new YAHOO.widget.Tooltip("myTooltip", { context:"null" } );</script>
+
+
+            	 <a href="/ndb/foods/show/1732?fg=&amp;man=&amp;lfacet=&amp;count=&amp;max=25&amp;sort=&amp;qlookup=spam&amp;offset=&amp;format=Stats&amp;new=" name="stats" title="View Statistics Report">Statistics Report</a><script type='text/javascript'> var myTooltip = new YAHOO.widget.Tooltip("myTooltip", { context:"null" } );</script>
+
+            </div>
+
+
+            <div class="dialog">
+
+
+        <div class="null">
+            <div id="measuresHelpDialog">
+                <div class="hd">Modifying household measures</div>
+                <div class="bd">
+
+                	<div id="helpDiv"></div>
+
+                </div>
+            </div>
+        </div>
+        <script>
+            function init_dlg_measuresHelpDialog() {
+                // Instantiate the Dialog
+                GRAILSUI.measuresHelpDialog = new YAHOO.widget.Dialog("measuresHelpDialog",
+                { 'width': '600px',
+'class': 'helpDialog',
+'draggable': true,
+'modal': false,
+'fixedcenter': true,
+'visible': false,
+'params': [],
+'constraintoviewport': true,
+'buttons': [{'text': 'OK',
+handler: function() {this.cancel();},
+'isDefault': true}] });
+                GRAILSUI.measuresHelpDialog.render(document.body);
+
+
+            }
+            YAHOO.util.Event.onDOMReady(init_dlg_measuresHelpDialog);
+        </script>
+
+
+               <!--  NUTRIENT DATA TABLE -->
+               <form action="/ndb/foods/show/1732" method="get" >
+               <input type="hidden" name="fg" value="" id="fg" />
+               <input type="hidden" name="man" value="" id="man" />
+               <input type="hidden" name="lfacet" value="" id="lfacet" />
+               <input type="hidden" name="count" value="" id="count" />
+               <input type="hidden" name="max" value="25" id="max" />
+               <input type="hidden" name="qlookup" value="spam" id="qlookup" />
+               <input type="hidden" name="offset" value="" id="offset" />
+               <input type="hidden" name="sort" value="" id="sort" />
+               <input type="hidden" name="format" value="Abridged" id="format" />
+
+
+              		<div class="nutlist">
+
+
+              	<p style="font-style:italic;font-size:.8em">Nutrient values and weights are for edible portion</p>
+
+
+	<table>
+                <thead>
+
+                <tr><td colspan="6" style="vertical-align:middle;text-align:center;height:2em;" class="buttons"><input type="submit" name="_action_show" value="Apply Changes" class="calc" title="Click to recalculate measures" id="1732" /><a href="/ndb/help/contextHelp/measures" onclick="jQuery.ajax({type:'POST', url:'/ndb/help/contextHelp/measures',success:function(data,textStatus){jQuery('#helpDiv').html(data);},error:function(XMLHttpRequest,textStatus,errorThrown){},complete:function(XMLHttpRequest,textStatus){GRAILSUI.measuresHelpDialog.show();}});return false;" controller="help" action="contextHelp" id="measures"><img title="Click for more information on calculating household measures" src="/ndb/static/images/skin/help.png" alt="Help" border="0" style="vertical-align:middle"/></a></span></td></tr>
+                <th style="vertical-align:middle">Nutrient</th>
+				<th style="vertical-align:middle" >Unit</th>
+                <th style="vertical-align:middle"><input type="text" name="Qv" style="width:30px;text-align:right;border-style:inset;height:15px" maxlength="5" value="1" id="Qv" /><br/>Value per 100.0g</th>
+
+
+
+
+  <th style="width:130px;line-height:1.2em;text-align:center">
+  	<input type="text" name="Q3483" style="width:30px;text-align:right;border-style:inset;height:15px" maxlength="5" value="2.0" id="Q3483" />
+  	<br>
+
+  	oz 1 NLEA serving
+  	<br>56g
+  	<!--
+  	-->
+  	</th>
+
+                </thead>
+                <tbody>
+
+                <tr class="even" >
+                <td style="font-weight:bold" colspan="6" bgcolor="#dddddd" >Proximates</td>
+                </tr>
+
+
+	                	<tr class="odd">
+	                	<td >Water
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">51.70</td>
+
+
+	                		<td style="text-align:right;">28.95</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Energy
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">kcal</td>
+	                	<td style="text-align:right;">315</td>
+
+
+	                		<td style="text-align:right;">176</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Protein
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">13.40</td>
+
+
+	                		<td style="text-align:right;">7.50</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Total lipid (fat)
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">26.60</td>
+
+
+	                		<td style="text-align:right;">14.90</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Carbohydrate, by difference
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">4.60</td>
+
+
+	                		<td style="text-align:right;">2.58</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Fiber, total dietary
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">0.0</td>
+
+
+	                		<td style="text-align:right;">0.0</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Sugars, total
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">0.00</td>
+
+
+	                		<td style="text-align:right;">0.00</td>
+
+
+	                	</tr>
+
+
+
+                <tr class="even" >
+                <td style="font-weight:bold" colspan="6" bgcolor="#dddddd" >Minerals</td>
+                </tr>
+
+
+	                	<tr class="odd">
+	                	<td >Calcium, Ca
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0</td>
+
+
+	                		<td style="text-align:right;">0</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Iron, Fe
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.64</td>
+
+
+	                		<td style="text-align:right;">0.36</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Magnesium, Mg
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">14</td>
+
+
+	                		<td style="text-align:right;">8</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Phosphorus, P
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">151</td>
+
+
+	                		<td style="text-align:right;">85</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Potassium, K
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">409</td>
+
+
+	                		<td style="text-align:right;">229</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Sodium, Na
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">1411</td>
+
+
+	                		<td style="text-align:right;">790</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Zinc, Zn
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">1.59</td>
+
+
+	                		<td style="text-align:right;">0.89</td>
+
+
+	                	</tr>
+
+
+
+                <tr class="even" >
+                <td style="font-weight:bold" colspan="6" bgcolor="#dddddd" >Vitamins</td>
+                </tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin C, total ascorbic acid
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.0</td>
+
+
+	                		<td style="text-align:right;">0.0</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Thiamin
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.317</td>
+
+
+	                		<td style="text-align:right;">0.178</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Riboflavin
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.176</td>
+
+
+	                		<td style="text-align:right;">0.099</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Niacin
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">3.530</td>
+
+
+	                		<td style="text-align:right;">1.977</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin B-6
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.218</td>
+
+
+	                		<td style="text-align:right;">0.122</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Folate, DFE
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">µg</td>
+	                	<td style="text-align:right;">3</td>
+
+
+	                		<td style="text-align:right;">2</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin B-12
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">µg</td>
+	                	<td style="text-align:right;">0.45</td>
+
+
+	                		<td style="text-align:right;">0.25</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Vitamin A, RAE
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">µg</td>
+	                	<td style="text-align:right;">0</td>
+
+
+	                		<td style="text-align:right;">0</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin A, IU
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">IU</td>
+	                	<td style="text-align:right;">0</td>
+
+
+	                		<td style="text-align:right;">0</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Vitamin E (alpha-tocopherol)
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0.42</td>
+
+
+	                		<td style="text-align:right;">0.24</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin D (D2 + D3)
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">µg</td>
+	                	<td style="text-align:right;">0.6</td>
+
+
+	                		<td style="text-align:right;">0.3</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Vitamin D
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">IU</td>
+	                	<td style="text-align:right;">26</td>
+
+
+	                		<td style="text-align:right;">15</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Vitamin K (phylloquinone)
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">µg</td>
+	                	<td style="text-align:right;">0.0</td>
+
+
+	                		<td style="text-align:right;">0.0</td>
+
+
+	                	</tr>
+
+
+
+                <tr class="even" >
+                <td style="font-weight:bold" colspan="6" bgcolor="#dddddd" >Lipids</td>
+                </tr>
+
+
+	                	<tr class="odd">
+	                	<td >Fatty acids, total saturated
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">9.987</td>
+
+
+	                		<td style="text-align:right;">5.593</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Fatty acids, total monounsaturated
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">13.505</td>
+
+
+	                		<td style="text-align:right;">7.563</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="odd">
+	                	<td >Fatty acids, total polyunsaturated
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">g</td>
+	                	<td style="text-align:right;">2.019</td>
+
+
+	                		<td style="text-align:right;">1.131</td>
+
+
+	                	</tr>
+
+
+	                	<tr class="even">
+	                	<td >Cholesterol
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">71</td>
+
+
+	                		<td style="text-align:right;">40</td>
+
+
+	                	</tr>
+
+
+
+                <tr class="even" >
+                <td style="font-weight:bold" colspan="6" bgcolor="#dddddd" >Other</td>
+                </tr>
+
+
+	                	<tr class="odd">
+	                	<td >Caffeine
+
+
+	                	</td>
+
+	                	<td style="text-align:center;">mg</td>
+	                	<td style="text-align:right;">0</td>
+
+
+	                		<td style="text-align:right;">0</td>
+
+
+	                	</tr>
+
+
+
+                </tbody>
+                </table>
+
+                </div>
+               </form>
+
+
+
+
+
+            </div>
+
+        </div>
+
+        <script src="/ndb/static/js/application.js" type="text/javascript" ></script>
+
+
+        </div>
+        <div class="footer">
+        National Nutrient Database for Standard Reference<br>Release 25
+			&nbsp;&nbsp;Software v.1.2.2
+		</div>
+    </body>
+</html>
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/data/html/valid_markup.html b/pandas/pandas/tests/io/data/html/valid_markup.html
new file mode 100644
index 0000000000000..0130e9ed9d5f3
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html/valid_markup.html
@@ -0,0 +1,62 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN">
+<html>
+    <head>
+        <meta name="generator" content=
+        "HTML Tidy for Linux (vers 25 March 2009), see www.w3.org">
+        <title></title>
+    </head>
+    <body>
+        <table border="1" class="dataframe">
+            <thead>
+                <tr style="text-align: right;">
+                    <th></th>
+                    <th>a</th>
+                    <th>b</th>
+                </tr>
+            </thead>
+            <tbody>
+                <tr>
+                    <th>0</th>
+                    <td>6</td>
+                    <td>7</td>
+                </tr>
+                <tr>
+                    <th>1</th>
+                    <td>4</td>
+                    <td>0</td>
+                </tr>
+                <tr>
+                    <th>2</th>
+                    <td>9</td>
+                    <td>4</td>
+                </tr>
+                <tr>
+                    <th>3</th>
+                    <td>7</td>
+                    <td>0</td>
+                </tr>
+            </tbody>
+        </table>
+        <table border="1" class="dataframe">
+            <thead>
+                <tr style="text-align: right;">
+                    <th></th>
+                    <th>a</th>
+                    <th>b</th>
+                </tr>
+            </thead>
+            <tbody>
+                <tr>
+                    <th>0</th>
+                    <td>6</td>
+                    <td>7</td>
+                </tr>
+                <tr>
+                    <th>1</th>
+                    <td>4</td>
+                    <td>0</td>
+                </tr>
+            </tbody>
+        </table>
+    </body>
+</html>
diff --git a/pandas/pandas/tests/io/data/html/wikipedia_states.html b/pandas/pandas/tests/io/data/html/wikipedia_states.html
new file mode 100644
index 0000000000000..f1a4c4d29c72e
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html/wikipedia_states.html
@@ -0,0 +1,1756 @@
+<!DOCTYPE html>
+<html lang="en" dir="ltr" class="client-nojs">
+<head>
+<meta charset="UTF-8" />
+<title>List of U.S. states and territories by area - Wikipedia, the free encyclopedia</title>
+<meta name="generator" content="MediaWiki 1.24wmf14" />
+<link rel="alternate" href="android-app://org.wikipedia/http/en.m.wikipedia.org/wiki/List_of_U.S._states_and_territories_by_area" />
+<link rel="alternate" type="application/x-wiki" title="Edit this page" href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit" />
+<link rel="edit" title="Edit this page" href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit" />
+<link rel="apple-touch-icon" href="//bits.wikimedia.org/apple-touch/wikipedia.png" />
+<link rel="shortcut icon" href="//bits.wikimedia.org/favicon/wikipedia.ico" />
+<link rel="search" type="application/opensearchdescription+xml" href="/w/opensearch_desc.php" title="Wikipedia (en)" />
+<link rel="EditURI" type="application/rsd+xml" href="//en.wikipedia.org/w/api.php?action=rsd" />
+<link rel="copyright" href="//creativecommons.org/licenses/by-sa/3.0/" />
+<link rel="alternate" type="application/atom+xml" title="Wikipedia Atom feed" href="/w/index.php?title=Special:RecentChanges&amp;feed=atom" />
+<link rel="canonical" href="http://en.wikipedia.org/wiki/List_of_U.S._states_and_territories_by_area" />
+<link rel="stylesheet" href="//bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&amp;lang=en&amp;modules=ext.gadget.DRN-wizard%2CReferenceTooltips%2Ccharinsert%2CrefToolbar%2Cteahouse%7Cext.rtlcite%2Cwikihiero%7Cext.uls.nojs%7Cext.visualEditor.viewPageTarget.noscript%7Cmediawiki.legacy.commonPrint%2Cshared%7Cmediawiki.skinning.interface%7Cmediawiki.ui.button%7Cskins.vector.styles%7Cwikibase.client.init&amp;only=styles&amp;skin=vector&amp;*" />
+<meta name="ResourceLoaderDynamicStyles" content="" />
+<link rel="stylesheet" href="//bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&amp;lang=en&amp;modules=site&amp;only=styles&amp;skin=vector&amp;*" />
+<style>a:lang(ar),a:lang(kk-arab),a:lang(mzn),a:lang(ps),a:lang(ur){text-decoration:none}
+/* cache key: enwiki:resourceloader:filter:minify-css:7:3904d24a08aa08f6a68dc338f9be277e */</style>
+<script src="//bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&amp;lang=en&amp;modules=startup&amp;only=scripts&amp;skin=vector&amp;*"></script>
+<script>if(window.mw){
+mw.config.set({"wgCanonicalNamespace":"","wgCanonicalSpecialPageName":false,"wgNamespaceNumber":0,"wgPageName":"List_of_U.S._states_and_territories_by_area","wgTitle":"List of U.S. states and territories by area","wgCurRevisionId":614847271,"wgRevisionId":614847271,"wgArticleId":87513,"wgIsArticle":true,"wgIsRedirect":false,"wgAction":"view","wgUserName":null,"wgUserGroups":["*"],"wgCategories":["Geography of the United States","Lists of states of the United States"],"wgBreakFrames":false,"wgPageContentLanguage":"en","wgPageContentModel":"wikitext","wgSeparatorTransformTable":["",""],"wgDigitTransformTable":["",""],"wgDefaultDateFormat":"dmy","wgMonthNames":["","January","February","March","April","May","June","July","August","September","October","November","December"],"wgMonthNamesShort":["","Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec"],"wgRelevantPageName":"List_of_U.S._states_and_territories_by_area","wgIsProbablyEditable":true,"wgRestrictionEdit":[],"wgRestrictionMove":[],"wgWikiEditorEnabledModules":{"toolbar":true,"dialogs":true,"hidesig":true,"preview":false,"previewDialog":false,"publish":false},"wgBetaFeaturesFeatures":[],"wgMediaViewerOnClick":true,"wgVisualEditor":{"isPageWatched":false,"magnifyClipIconURL":"//bits.wikimedia.org/static-1.24wmf14/skins/common/images/magnify-clip.png","pageLanguageCode":"en","pageLanguageDir":"ltr","svgMaxSize":2048,"namespacesWithSubpages":{"6":0,"8":0,"1":true,"2":true,"3":true,"4":true,"5":true,"7":true,"9":true,"10":true,"11":true,"12":true,"13":true,"14":true,"15":true,"100":true,"101":true,"102":true,"103":true,"104":true,"105":true,"106":true,"107":true,"108":true,"109":true,"110":true,"111":true,"447":true,"2600":false,"828":true,"829":true}},"wikilove-recipient":"","wikilove-anon":0,"wgGuidedTourHelpGuiderUrl":"Help:Guided tours/guider","wgULSAcceptLanguageList":["en-us"],"wgULSCurrentAutonym":"English","wgFlaggedRevsParams":{"tags":{"status":{"levels":1,"quality":2,"pristine":3}}},"wgStableRevisionId":null,"wgCategoryTreePageCategoryOptions":"{\"mode\":0,\"hideprefix\":20,\"showcount\":true,\"namespaces\":false}","wgNoticeProject":"wikipedia","wgWikibaseItemId":"Q150340"});
+}</script><script>if(window.mw){
+mw.loader.implement("user.options",function($,jQuery){mw.user.options.set({"ccmeonemails":0,"cols":80,"date":"default","diffonly":0,"disablemail":0,"editfont":"default","editondblclick":0,"editsectiononrightclick":0,"enotifminoredits":0,"enotifrevealaddr":0,"enotifusertalkpages":1,"enotifwatchlistpages":0,"extendwatchlist":0,"fancysig":0,"forceeditsummary":0,"gender":"unknown","hideminor":0,"hidepatrolled":0,"imagesize":2,"math":0,"minordefault":0,"newpageshidepatrolled":0,"nickname":"","norollbackdiff":0,"numberheadings":0,"previewonfirst":0,"previewontop":1,"rcdays":7,"rclimit":50,"rows":25,"showhiddencats":false,"shownumberswatching":1,"showtoolbar":1,"skin":"vector","stubthreshold":0,"thumbsize":4,"underline":2,"uselivepreview":0,"usenewrc":0,"watchcreations":1,"watchdefault":0,"watchdeletion":0,"watchlistdays":3,"watchlisthideanons":0,"watchlisthidebots":0,"watchlisthideliu":0,"watchlisthideminor":0,"watchlisthideown":0,"watchlisthidepatrolled":0,"watchmoves":0,"wllimit":250,
+"useeditwarning":1,"prefershttps":1,"flaggedrevssimpleui":1,"flaggedrevsstable":0,"flaggedrevseditdiffs":true,"flaggedrevsviewdiffs":false,"usebetatoolbar":1,"usebetatoolbar-cgd":1,"multimediaviewer-enable":true,"visualeditor-enable":0,"visualeditor-betatempdisable":0,"visualeditor-enable-experimental":0,"visualeditor-enable-language":0,"visualeditor-hidebetawelcome":0,"wikilove-enabled":1,"mathJax":false,"echo-subscriptions-web-page-review":true,"echo-subscriptions-email-page-review":false,"ep_showtoplink":false,"ep_bulkdelorgs":false,"ep_bulkdelcourses":true,"ep_showdyk":true,"echo-subscriptions-web-education-program":true,"echo-subscriptions-email-education-program":false,"echo-notify-show-link":true,"echo-show-alert":true,"echo-email-frequency":0,"echo-email-format":"html","echo-subscriptions-email-system":true,"echo-subscriptions-web-system":true,"echo-subscriptions-email-user-rights":true,"echo-subscriptions-web-user-rights":true,"echo-subscriptions-email-other":false,
+"echo-subscriptions-web-other":true,"echo-subscriptions-email-edit-user-talk":false,"echo-subscriptions-web-edit-user-talk":true,"echo-subscriptions-email-reverted":false,"echo-subscriptions-web-reverted":true,"echo-subscriptions-email-article-linked":false,"echo-subscriptions-web-article-linked":false,"echo-subscriptions-email-mention":false,"echo-subscriptions-web-mention":true,"echo-subscriptions-web-edit-thank":true,"echo-subscriptions-email-edit-thank":false,"echo-subscriptions-web-flow-discussion":true,"echo-subscriptions-email-flow-discussion":false,"gettingstarted-task-toolbar-show-intro":true,"uls-preferences":"","language":"en","variant-gan":"gan","variant-iu":"iu","variant-kk":"kk","variant-ku":"ku","variant-shi":"shi","variant-sr":"sr","variant-tg":"tg","variant-uz":"uz","variant-zh":"zh","searchNs0":true,"searchNs1":false,"searchNs2":false,"searchNs3":false,"searchNs4":false,"searchNs5":false,"searchNs6":false,"searchNs7":false,"searchNs8":false,"searchNs9":false,
+"searchNs10":false,"searchNs11":false,"searchNs12":false,"searchNs13":false,"searchNs14":false,"searchNs15":false,"searchNs100":false,"searchNs101":false,"searchNs108":false,"searchNs109":false,"searchNs118":false,"searchNs119":false,"searchNs446":false,"searchNs447":false,"searchNs710":false,"searchNs711":false,"searchNs828":false,"searchNs829":false,"searchNs2600":false,"gadget-teahouse":1,"gadget-ReferenceTooltips":1,"gadget-DRN-wizard":1,"gadget-charinsert":1,"gadget-refToolbar":1,"gadget-mySandbox":1,"variant":"en"});},{},{});mw.loader.implement("user.tokens",function($,jQuery){mw.user.tokens.set({"editToken":"+\\","patrolToken":false,"watchToken":false});},{},{});
+/* cache key: enwiki:resourceloader:filter:minify-js:7:ffff827f827051d73171f6b2dc70d368 */
+}</script>
+<script>if(window.mw){
+mw.loader.load(["mediawiki.page.startup","mediawiki.legacy.wikibits","mediawiki.legacy.ajax","ext.centralauth.centralautologin","mmv.head","ext.visualEditor.viewPageTarget.init","ext.uls.init","ext.uls.interface","ext.centralNotice.bannerController","skins.vector.js"]);
+}</script>
+<link rel="dns-prefetch" href="//meta.wikimedia.org" />
+<!--[if lt IE 7]><style type="text/css">body{behavior:url("/w/static-1.24wmf14/skins/Vector/csshover.min.htc")}</style><![endif]-->
+</head>
+<body class="mediawiki ltr sitedir-ltr ns-0 ns-subject page-List_of_U_S_states_and_territories_by_area skin-vector action-view vector-animateLayout">
+		<div id="mw-page-base" class="noprint"></div>
+		<div id="mw-head-base" class="noprint"></div>
+		<div id="content" class="mw-body" role="main">
+			<a id="top"></a>
+
+			<div id="mw-js-message" style="display:none;"></div>
+							<div id="siteNotice"><!-- CentralNotice --></div>
+						<h1 id="firstHeading" class="firstHeading" lang="en"><span dir="auto">List of U.S. states and territories by area</span></h1>
+						<div id="bodyContent" class="mw-body-content">
+									<div id="siteSub">From Wikipedia, the free encyclopedia</div>
+								<div id="contentSub"></div>
+												<div id="jump-to-nav" class="mw-jump">
+					Jump to:					<a href="#mw-navigation">navigation</a>, 					<a href="#p-search">search</a>
+				</div>
+				<div id="mw-content-text" lang="en" dir="ltr" class="mw-content-ltr"><div class="thumb tright">
+<div class="thumbinner" style="width:222px;"><a href="/wiki/File:Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg/220px-Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg" width="220" height="126" class="thumbimage" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg/330px-Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/2/28/Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg/440px-Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg 2x" data-file-width="1015" data-file-height="580" /></a>
+<div class="thumbcaption">
+<div class="magnify"><a href="/wiki/File:Image_shows_the_50_states_by_area-_Check_the_legend_for_more_details-_2014-06-29_05-36.jpg" class="internal" title="Enlarge"><img src="//bits.wikimedia.org/static-1.24wmf14/skins/common/images/magnify-clip.png" width="15" height="11" alt="" /></a></div>
+Image shows the 50 states by area. Check the legend for more details.</div>
+</div>
+</div>
+<p>This is a complete <b>list of the <a href="/wiki/U.S._state" title="U.S. state">states of the United States</a> and its major <a href="/wiki/Territories_of_the_United_States" title="Territories of the United States">territories</a></b> ordered by <i>total area</i>, <i>land area</i>, and <i>water area</i>. The water area figures include inland, coastal, <a href="/wiki/Great_Lakes" title="Great Lakes">Great Lakes</a>, and <a href="/wiki/Territorial_waters" title="Territorial waters">territorial waters</a>. Glaciers and intermittent water features are counted as land area.<sup id="cite_ref-1" class="reference"><a href="#cite_note-1"><span>[</span>1<span>]</span></a></sup></p>
+<p></p>
+<div id="toc" class="toc">
+<div id="toctitle">
+<h2>Contents</h2>
+</div>
+<ul>
+<li class="toclevel-1 tocsection-1"><a href="#Area_by_state.2Fterritory"><span class="tocnumber">1</span> <span class="toctext">Area by state/territory</span></a></li>
+<li class="toclevel-1 tocsection-2"><a href="#Area_by_division"><span class="tocnumber">2</span> <span class="toctext">Area by division</span></a></li>
+<li class="toclevel-1 tocsection-3"><a href="#Area_by_region"><span class="tocnumber">3</span> <span class="toctext">Area by region</span></a></li>
+<li class="toclevel-1 tocsection-4"><a href="#See_also"><span class="tocnumber">4</span> <span class="toctext">See also</span></a></li>
+<li class="toclevel-1 tocsection-5"><a href="#Notes"><span class="tocnumber">5</span> <span class="toctext">Notes</span></a></li>
+<li class="toclevel-1 tocsection-6"><a href="#References"><span class="tocnumber">6</span> <span class="toctext">References</span></a></li>
+<li class="toclevel-1 tocsection-7"><a href="#External_links"><span class="tocnumber">7</span> <span class="toctext">External links</span></a></li>
+</ul>
+</div>
+<p></p>
+<div style="clear:both;"></div>
+<h2><span class="mw-headline" id="Area_by_state.2Fterritory">Area by state/territory</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=1" title="Edit section: Area by state/territory">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<table class="wikitable sortable">
+<tr>
+<th></th>
+<th colspan="3">Total area<sup id="cite_ref-2010census_2-0" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Land area<sup id="cite_ref-2010census_2-1" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Water<sup id="cite_ref-2010census_2-2" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+</tr>
+<tr>
+<th>State/territory</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% land</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% water</th>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/e/e6/Flag_of_Alaska.svg/21px-Flag_of_Alaska.svg.png" width="21" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/e6/Flag_of_Alaska.svg/33px-Flag_of_Alaska.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/e6/Flag_of_Alaska.svg/43px-Flag_of_Alaska.svg.png 2x" data-file-width="1416" data-file-height="1000" />&#160;</span><a href="/wiki/Alaska" title="Alaska">Alaska</a></td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">665,384.04</td>
+<td align="right">1,723,337</td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">570,640.95</td>
+<td align="right">1,477,953</td>
+<td align="right"><span style="display:none" class="sortkey">7001857600000000000</span>85.76%</td>
+<td align="right">94,743.10</td>
+<td align="right">245,384</td>
+<td align="right"><span style="display:none" class="sortkey">7001142400000000000</span>14.24%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Texas.svg/23px-Flag_of_Texas.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Texas.svg/35px-Flag_of_Texas.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Texas.svg/45px-Flag_of_Texas.svg.png 2x" data-file-width="1080" data-file-height="720" />&#160;</span><a href="/wiki/Texas" title="Texas">Texas</a></td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">268,596.46</td>
+<td align="right">695,662</td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">261,231.71</td>
+<td align="right">676,587</td>
+<td align="right"><span style="display:none" class="sortkey">7001972600000000000</span>97.26%</td>
+<td align="right">7,364.75</td>
+<td align="right">19,075</td>
+<td align="right"><span style="display:none" class="sortkey">7000274000000000000</span>2.74%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_California.svg/23px-Flag_of_California.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_California.svg/35px-Flag_of_California.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_California.svg/45px-Flag_of_California.svg.png 2x" data-file-width="900" data-file-height="600" />&#160;</span><a href="/wiki/California" title="California">California</a></td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">163,694.74</td>
+<td align="right">423,967</td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">155,779.22</td>
+<td align="right">403,466</td>
+<td align="right"><span style="display:none" class="sortkey">7001951600000000000</span>95.16%</td>
+<td align="right">7,915.52</td>
+<td align="right">20,501</td>
+<td align="right"><span style="display:none" class="sortkey">7000484000000000000</span>4.84%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/c/cb/Flag_of_Montana.svg/23px-Flag_of_Montana.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/c/cb/Flag_of_Montana.svg/35px-Flag_of_Montana.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/c/cb/Flag_of_Montana.svg/45px-Flag_of_Montana.svg.png 2x" data-file-width="615" data-file-height="410" />&#160;</span><a href="/wiki/Montana" title="Montana">Montana</a></td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">147,039.71</td>
+<td align="right">380,831</td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">145,545.80</td>
+<td align="right">376,962</td>
+<td align="right"><span style="display:none" class="sortkey">7001989800000000000</span>98.98%</td>
+<td align="right">1,493.91</td>
+<td align="right">3,869</td>
+<td align="right"><span style="display:none" class="sortkey">7000102000000000000</span>1.02%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Flag_of_New_Mexico.svg/23px-Flag_of_New_Mexico.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Flag_of_New_Mexico.svg/35px-Flag_of_New_Mexico.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Flag_of_New_Mexico.svg/45px-Flag_of_New_Mexico.svg.png 2x" data-file-width="1200" data-file-height="800" />&#160;</span><a href="/wiki/New_Mexico" title="New Mexico">New Mexico</a></td>
+<td align="center"><span style="" class="sortkey">!B9983905620875&#160;</span>5</td>
+<td align="right">121,590.30</td>
+<td align="right">314,917</td>
+<td align="center"><span style="" class="sortkey">!B9983905620875&#160;</span>5</td>
+<td align="right">121,298.15</td>
+<td align="right">314,161</td>
+<td align="right"><span style="display:none" class="sortkey">7001997600000000000</span>99.76%</td>
+<td align="right">292.15</td>
+<td align="right">757</td>
+<td align="right"><span style="display:none" class="sortkey">6999240000000000000</span>0.24%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arizona.svg/23px-Flag_of_Arizona.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arizona.svg/35px-Flag_of_Arizona.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arizona.svg/45px-Flag_of_Arizona.svg.png 2x" data-file-width="900" data-file-height="600" />&#160;</span><a href="/wiki/Arizona" title="Arizona">Arizona</a></td>
+<td align="center"><span style="" class="sortkey">!B9982082405307&#160;</span>6</td>
+<td align="right">113,990.30</td>
+<td align="right">295,234</td>
+<td align="center"><span style="" class="sortkey">!B9982082405307&#160;</span>6</td>
+<td align="right">113,594.08</td>
+<td align="right">294,207</td>
+<td align="right"><span style="display:none" class="sortkey">7001996500000000000</span>99.65%</td>
+<td align="right">396.22</td>
+<td align="right">1,026</td>
+<td align="right"><span style="display:none" class="sortkey">6999350000000000000</span>0.35%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f1/Flag_of_Nevada.svg/23px-Flag_of_Nevada.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f1/Flag_of_Nevada.svg/35px-Flag_of_Nevada.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f1/Flag_of_Nevada.svg/45px-Flag_of_Nevada.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/Nevada" title="Nevada">Nevada</a></td>
+<td align="center"><span style="" class="sortkey">!B9980540898509&#160;</span>7</td>
+<td align="right">110,571.82</td>
+<td align="right">286,380</td>
+<td align="center"><span style="" class="sortkey">!B9980540898509&#160;</span>7</td>
+<td align="right">109,781.18</td>
+<td align="right">284,332</td>
+<td align="right"><span style="display:none" class="sortkey">7001992800000000000</span>99.28%</td>
+<td align="right">790.65</td>
+<td align="right">2,048</td>
+<td align="right"><span style="display:none" class="sortkey">6999720000000000000</span>0.72%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/46/Flag_of_Colorado.svg/23px-Flag_of_Colorado.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/46/Flag_of_Colorado.svg/35px-Flag_of_Colorado.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/46/Flag_of_Colorado.svg/45px-Flag_of_Colorado.svg.png 2x" data-file-width="1800" data-file-height="1200" />&#160;</span><a href="/wiki/Colorado" title="Colorado">Colorado</a></td>
+<td align="center"><span style="" class="sortkey">!B9979205584583&#160;</span>8</td>
+<td align="right">104,093.67</td>
+<td align="right">269,601</td>
+<td align="center"><span style="" class="sortkey">!B9979205584583&#160;</span>8</td>
+<td align="right">103,641.89</td>
+<td align="right">268,431</td>
+<td align="right"><span style="display:none" class="sortkey">7001995700000099999</span>99.57%</td>
+<td align="right">451.78</td>
+<td align="right">1,170</td>
+<td align="right"><span style="display:none" class="sortkey">6999430000000000000</span>0.43%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Oregon.svg/23px-Flag_of_Oregon.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Oregon.svg/35px-Flag_of_Oregon.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Oregon.svg/46px-Flag_of_Oregon.svg.png 2x" data-file-width="750" data-file-height="450" />&#160;</span><a href="/wiki/Oregon" title="Oregon">Oregon</a></td>
+<td align="center"><span style="" class="sortkey">!B9978027754226&#160;</span>9</td>
+<td align="right">98,378.54</td>
+<td align="right">254,799</td>
+<td align="center"><span style="" class="sortkey">!B9976974149070&#160;</span>10</td>
+<td align="right">95,988.01</td>
+<td align="right">248,608</td>
+<td align="right"><span style="display:none" class="sortkey">7001975700000099999</span>97.57%</td>
+<td align="right">2,390.53</td>
+<td align="right">6,191</td>
+<td align="right"><span style="display:none" class="sortkey">7000243000000000000</span>2.43%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/bc/Flag_of_Wyoming.svg/22px-Flag_of_Wyoming.svg.png" width="22" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/bc/Flag_of_Wyoming.svg/33px-Flag_of_Wyoming.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/bc/Flag_of_Wyoming.svg/43px-Flag_of_Wyoming.svg.png 2x" data-file-width="1000" data-file-height="700" />&#160;</span><a href="/wiki/Wyoming" title="Wyoming">Wyoming</a></td>
+<td align="center"><span style="" class="sortkey">!B9976974149070&#160;</span>10</td>
+<td align="right">97,813.01</td>
+<td align="right">253,335</td>
+<td align="center"><span style="" class="sortkey">!B9978027754226&#160;</span>9</td>
+<td align="right">97,093.14</td>
+<td align="right">251,470</td>
+<td align="right"><span style="display:none" class="sortkey">7001992600000000000</span>99.26%</td>
+<td align="right">719.87</td>
+<td align="right">1,864</td>
+<td align="right"><span style="display:none" class="sortkey">6999740000000000000</span>0.74%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/b5/Flag_of_Michigan.svg/23px-Flag_of_Michigan.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/b5/Flag_of_Michigan.svg/35px-Flag_of_Michigan.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/b5/Flag_of_Michigan.svg/45px-Flag_of_Michigan.svg.png 2x" data-file-width="685" data-file-height="457" />&#160;</span><a href="/wiki/Michigan" title="Michigan">Michigan</a></td>
+<td align="center"><span style="" class="sortkey">!B9976021047272&#160;</span>11</td>
+<td align="right">96,713.51</td>
+<td align="right">250,487</td>
+<td align="center"><span style="" class="sortkey">!B9969089575466&#160;</span>22</td>
+<td align="right">56,538.90</td>
+<td align="right">146,435</td>
+<td align="right"><span style="display:none" class="sortkey">7001584600000000000</span>58.46%</td>
+<td align="right">40,174.61</td>
+<td align="right">104,052</td>
+<td align="right"><span style="display:none" class="sortkey">7001415400000000000</span>41.54%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Minnesota.svg/23px-Flag_of_Minnesota.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Minnesota.svg/35px-Flag_of_Minnesota.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/b9/Flag_of_Minnesota.svg/46px-Flag_of_Minnesota.svg.png 2x" data-file-width="500" data-file-height="318" />&#160;</span><a href="/wiki/Minnesota" title="Minnesota">Minnesota</a></td>
+<td align="center"><span style="" class="sortkey">!B9975150933502&#160;</span>12</td>
+<td align="right">86,935.83</td>
+<td align="right">225,163</td>
+<td align="center"><span style="" class="sortkey">!B9973609426703&#160;</span>14</td>
+<td align="right">79,626.74</td>
+<td align="right">206,232</td>
+<td align="right"><span style="display:none" class="sortkey">7001915900000000000</span>91.59%</td>
+<td align="right">7,309.09</td>
+<td align="right">18,930</td>
+<td align="right"><span style="display:none" class="sortkey">7000841000000000000</span>8.41%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Flag_of_Utah.png/23px-Flag_of_Utah.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Flag_of_Utah.png/35px-Flag_of_Utah.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Flag_of_Utah.png/46px-Flag_of_Utah.png 2x" data-file-width="2875" data-file-height="1725" />&#160;</span><a href="/wiki/Utah" title="Utah">Utah</a></td>
+<td align="center"><span style="" class="sortkey">!B9974350506425&#160;</span>13</td>
+<td align="right">84,896.88</td>
+<td align="right">219,882</td>
+<td align="center"><span style="" class="sortkey">!B9975150933502&#160;</span>12</td>
+<td align="right">82,169.62</td>
+<td align="right">212,818</td>
+<td align="right"><span style="display:none" class="sortkey">7001967900000000000</span>96.79%</td>
+<td align="right">2,727.26</td>
+<td align="right">7,064</td>
+<td align="right"><span style="display:none" class="sortkey">7000321000000000000</span>3.21%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/a/a4/Flag_of_Idaho.svg/19px-Flag_of_Idaho.svg.png" width="19" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/a/a4/Flag_of_Idaho.svg/29px-Flag_of_Idaho.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/a/a4/Flag_of_Idaho.svg/38px-Flag_of_Idaho.svg.png 2x" data-file-width="660" data-file-height="520" />&#160;</span><a href="/wiki/Idaho" title="Idaho">Idaho</a></td>
+<td align="center"><span style="" class="sortkey">!B9973609426703&#160;</span>14</td>
+<td align="right">83,568.95</td>
+<td align="right">216,443</td>
+<td align="center"><span style="" class="sortkey">!B9976021047272&#160;</span>11</td>
+<td align="right">82,643.12</td>
+<td align="right">214,045</td>
+<td align="right"><span style="display:none" class="sortkey">7001988900000000000</span>98.89%</td>
+<td align="right">925.83</td>
+<td align="right">2,398</td>
+<td align="right"><span style="display:none" class="sortkey">7000111000000000000</span>1.11%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/d/da/Flag_of_Kansas.svg/23px-Flag_of_Kansas.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/d/da/Flag_of_Kansas.svg/35px-Flag_of_Kansas.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/d/da/Flag_of_Kansas.svg/46px-Flag_of_Kansas.svg.png 2x" data-file-width="5400" data-file-height="3240" />&#160;</span><a href="/wiki/Kansas" title="Kansas">Kansas</a></td>
+<td align="center"><span style="" class="sortkey">!B9972919497988&#160;</span>15</td>
+<td align="right">82,278.36</td>
+<td align="right">213,100</td>
+<td align="center"><span style="" class="sortkey">!B9974350506425&#160;</span>13</td>
+<td align="right">81,758.72</td>
+<td align="right">211,754</td>
+<td align="right"><span style="display:none" class="sortkey">7001993700000000000</span>99.37%</td>
+<td align="right">519.64</td>
+<td align="right">1,346</td>
+<td align="right"><span style="display:none" class="sortkey">6999630000000000000</span>0.63%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/4d/Flag_of_Nebraska.svg/23px-Flag_of_Nebraska.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/4d/Flag_of_Nebraska.svg/35px-Flag_of_Nebraska.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/4d/Flag_of_Nebraska.svg/46px-Flag_of_Nebraska.svg.png 2x" data-file-width="750" data-file-height="450" />&#160;</span><a href="/wiki/Nebraska" title="Nebraska">Nebraska</a></td>
+<td align="center"><span style="" class="sortkey">!B9972274112777&#160;</span>16</td>
+<td align="right">77,347.81</td>
+<td align="right">200,330</td>
+<td align="center"><span style="" class="sortkey">!B9972919497988&#160;</span>15</td>
+<td align="right">76,824.17</td>
+<td align="right">198,974</td>
+<td align="right"><span style="display:none" class="sortkey">7001993200000099999</span>99.32%</td>
+<td align="right">523.64</td>
+<td align="right">1,356</td>
+<td align="right"><span style="display:none" class="sortkey">6999680000000000000</span>0.68%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_South_Dakota.svg/23px-Flag_of_South_Dakota.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_South_Dakota.svg/35px-Flag_of_South_Dakota.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_South_Dakota.svg/46px-Flag_of_South_Dakota.svg.png 2x" data-file-width="720" data-file-height="450" />&#160;</span><a href="/wiki/South_Dakota" title="South Dakota">South Dakota</a></td>
+<td align="center"><span style="" class="sortkey">!B9971667866559&#160;</span>17</td>
+<td align="right">77,115.68</td>
+<td align="right">199,729</td>
+<td align="center"><span style="" class="sortkey">!B9972274112777&#160;</span>16</td>
+<td align="right">75,811.00</td>
+<td align="right">196,350</td>
+<td align="right"><span style="display:none" class="sortkey">7001983100000000000</span>98.31%</td>
+<td align="right">1,304.68</td>
+<td align="right">3,379</td>
+<td align="right"><span style="display:none" class="sortkey">7000169000000000000</span>1.69%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Washington.svg/23px-Flag_of_Washington.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Washington.svg/35px-Flag_of_Washington.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Washington.svg/46px-Flag_of_Washington.svg.png 2x" data-file-width="1106" data-file-height="658" />&#160;</span><a href="/wiki/Washington_(state)" title="Washington (state)">Washington</a></td>
+<td align="center"><span style="" class="sortkey">!B9971096282421&#160;</span>18</td>
+<td align="right">71,297.95</td>
+<td align="right">184,661</td>
+<td align="center"><span style="" class="sortkey">!B9970042677264&#160;</span>20</td>
+<td align="right">66,455.52</td>
+<td align="right">172,119</td>
+<td align="right"><span style="display:none" class="sortkey">7001932100000099999</span>93.21%</td>
+<td align="right">4,842.43</td>
+<td align="right">12,542</td>
+<td align="right"><span style="display:none" class="sortkey">7000679000000000000</span>6.79%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/e/ee/Flag_of_North_Dakota.svg/20px-Flag_of_North_Dakota.svg.png" width="20" height="16" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/ee/Flag_of_North_Dakota.svg/30px-Flag_of_North_Dakota.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/ee/Flag_of_North_Dakota.svg/40px-Flag_of_North_Dakota.svg.png 2x" data-file-width="575" data-file-height="450" />&#160;</span><a href="/wiki/North_Dakota" title="North Dakota">North Dakota</a></td>
+<td align="center"><span style="" class="sortkey">!B9970555610208&#160;</span>19</td>
+<td align="right">70,698.32</td>
+<td align="right">183,108</td>
+<td align="center"><span style="" class="sortkey">!B9971667866559&#160;</span>17</td>
+<td align="right">69,000.80</td>
+<td align="right">178,711</td>
+<td align="right"><span style="display:none" class="sortkey">7001976000000000000</span>97.60%</td>
+<td align="right">1,697.52</td>
+<td align="right">4,397</td>
+<td align="right"><span style="display:none" class="sortkey">7000240000000000000</span>2.40%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/6/6e/Flag_of_Oklahoma.svg/23px-Flag_of_Oklahoma.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/6/6e/Flag_of_Oklahoma.svg/35px-Flag_of_Oklahoma.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/6/6e/Flag_of_Oklahoma.svg/45px-Flag_of_Oklahoma.svg.png 2x" data-file-width="675" data-file-height="450" />&#160;</span><a href="/wiki/Oklahoma" title="Oklahoma">Oklahoma</a></td>
+<td align="center"><span style="" class="sortkey">!B9970042677264&#160;</span>20</td>
+<td align="right">69,898.87</td>
+<td align="right">181,037</td>
+<td align="center"><span style="" class="sortkey">!B9970555610208&#160;</span>19</td>
+<td align="right">68,594.92</td>
+<td align="right">177,660</td>
+<td align="right"><span style="display:none" class="sortkey">7001981300000000000</span>98.13%</td>
+<td align="right">1,303.95</td>
+<td align="right">3,377</td>
+<td align="right"><span style="display:none" class="sortkey">7000187000000000000</span>1.87%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/5/5a/Flag_of_Missouri.svg/23px-Flag_of_Missouri.svg.png" width="23" height="13" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/5/5a/Flag_of_Missouri.svg/35px-Flag_of_Missouri.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/5/5a/Flag_of_Missouri.svg/46px-Flag_of_Missouri.svg.png 2x" data-file-width="2400" data-file-height="1400" />&#160;</span><a href="/wiki/Missouri" title="Missouri">Missouri</a></td>
+<td align="center"><span style="" class="sortkey">!B9969554775622&#160;</span>21</td>
+<td align="right">69,706.99</td>
+<td align="right">180,540</td>
+<td align="center"><span style="" class="sortkey">!B9971096282421&#160;</span>18</td>
+<td align="right">68,741.52</td>
+<td align="right">178,040</td>
+<td align="right"><span style="display:none" class="sortkey">7001986100000000000</span>98.61%</td>
+<td align="right">965.47</td>
+<td align="right">2,501</td>
+<td align="right"><span style="display:none" class="sortkey">7000138990000099999</span>1.39%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Florida.svg/23px-Flag_of_Florida.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Florida.svg/35px-Flag_of_Florida.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Florida.svg/45px-Flag_of_Florida.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/Florida" title="Florida">Florida</a></td>
+<td align="center"><span style="" class="sortkey">!B9969089575466&#160;</span>22</td>
+<td align="right">65,757.70</td>
+<td align="right">170,312</td>
+<td align="center"><span style="" class="sortkey">!B9967419034619&#160;</span>26</td>
+<td align="right">53,624.76</td>
+<td align="right">138,887</td>
+<td align="right"><span style="display:none" class="sortkey">7001815500000000000</span>81.55%</td>
+<td align="right">12,132.94</td>
+<td align="right">31,424</td>
+<td align="right"><span style="display:none" class="sortkey">7001184500000000000</span>18.45%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_Wisconsin.svg/23px-Flag_of_Wisconsin.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_Wisconsin.svg/35px-Flag_of_Wisconsin.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_Wisconsin.svg/45px-Flag_of_Wisconsin.svg.png 2x" data-file-width="675" data-file-height="450" />&#160;</span><a href="/wiki/Wisconsin" title="Wisconsin">Wisconsin</a></td>
+<td align="center"><span style="" class="sortkey">!B9968645057840&#160;</span>23</td>
+<td align="right">65,496.38</td>
+<td align="right">169,635</td>
+<td align="center"><span style="" class="sortkey">!B9967811241751&#160;</span>25</td>
+<td align="right">54,157.80</td>
+<td align="right">140,268</td>
+<td align="right"><span style="display:none" class="sortkey">7001826900000000000</span>82.69%</td>
+<td align="right">11,338.57</td>
+<td align="right">29,367</td>
+<td align="right"><span style="display:none" class="sortkey">7001173109999900000</span>17.31%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Georgia_%28U.S._state%29.svg/23px-Flag_of_Georgia_%28U.S._state%29.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Georgia_%28U.S._state%29.svg/35px-Flag_of_Georgia_%28U.S._state%29.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/5/54/Flag_of_Georgia_%28U.S._state%29.svg/46px-Flag_of_Georgia_%28U.S._state%29.svg.png 2x" data-file-width="6912" data-file-height="4320" />&#160;</span><a href="/wiki/Georgia_(U.S._state)" title="Georgia (U.S. state)">Georgia</a></td>
+<td align="center"><span style="" class="sortkey">!B9968219461696&#160;</span>24</td>
+<td align="right">59,425.15</td>
+<td align="right">153,910</td>
+<td align="center"><span style="" class="sortkey">!B9969554775622&#160;</span>21</td>
+<td align="right">57,513.49</td>
+<td align="right">148,959</td>
+<td align="right"><span style="display:none" class="sortkey">7001967800000000000</span>96.78%</td>
+<td align="right">1,911.66</td>
+<td align="right">4,951</td>
+<td align="right"><span style="display:none" class="sortkey">7000322000000000000</span>3.22%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_Illinois.svg/23px-Flag_of_Illinois.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_Illinois.svg/35px-Flag_of_Illinois.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/0/01/Flag_of_Illinois.svg/46px-Flag_of_Illinois.svg.png 2x" data-file-width="778" data-file-height="460" />&#160;</span><a href="/wiki/Illinois" title="Illinois">Illinois</a></td>
+<td align="center"><span style="" class="sortkey">!B9967811241751&#160;</span>25</td>
+<td align="right">57,913.55</td>
+<td align="right">149,995</td>
+<td align="center"><span style="" class="sortkey">!B9968219461696&#160;</span>24</td>
+<td align="right">55,518.93</td>
+<td align="right">143,793</td>
+<td align="right"><span style="display:none" class="sortkey">7001958700000000000</span>95.87%</td>
+<td align="right">2,394.62</td>
+<td align="right">6,202</td>
+<td align="right"><span style="display:none" class="sortkey">7000413000000000000</span>4.13%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/a/aa/Flag_of_Iowa.svg/22px-Flag_of_Iowa.svg.png" width="22" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/a/aa/Flag_of_Iowa.svg/34px-Flag_of_Iowa.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/a/aa/Flag_of_Iowa.svg/44px-Flag_of_Iowa.svg.png 2x" data-file-width="670" data-file-height="459" />&#160;</span><a href="/wiki/Iowa" title="Iowa">Iowa</a></td>
+<td align="center"><span style="" class="sortkey">!B9967419034619&#160;</span>26</td>
+<td align="right">56,272.81</td>
+<td align="right">145,746</td>
+<td align="center"><span style="" class="sortkey">!B9968645057840&#160;</span>23</td>
+<td align="right">55,857.13</td>
+<td align="right">144,669</td>
+<td align="right"><span style="display:none" class="sortkey">7001992600000000000</span>99.26%</td>
+<td align="right">415.68</td>
+<td align="right">1,077</td>
+<td align="right"><span style="display:none" class="sortkey">6999740000000000000</span>0.74%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_New_York.svg/23px-Flag_of_New_York.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_New_York.svg/35px-Flag_of_New_York.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/1/1a/Flag_of_New_York.svg/46px-Flag_of_New_York.svg.png 2x" data-file-width="900" data-file-height="450" />&#160;</span><a href="/wiki/New_York" title="New York">New York</a></td>
+<td align="center"><span style="" class="sortkey">!B9967041631339&#160;</span>27</td>
+<td align="right">54,554.98</td>
+<td align="right">141,297</td>
+<td align="center"><span style="" class="sortkey">!B9965988026183&#160;</span>30</td>
+<td align="right">47,126.40</td>
+<td align="right">122,057</td>
+<td align="right"><span style="display:none" class="sortkey">7001863800000000000</span>86.38%</td>
+<td align="right">7,428.58</td>
+<td align="right">19,240</td>
+<td align="right"><span style="display:none" class="sortkey">7001136200000099999</span>13.62%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Flag_of_North_Carolina.svg/23px-Flag_of_North_Carolina.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Flag_of_North_Carolina.svg/35px-Flag_of_North_Carolina.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Flag_of_North_Carolina.svg/45px-Flag_of_North_Carolina.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/North_Carolina" title="North Carolina">North Carolina</a></td>
+<td align="center"><span style="" class="sortkey">!B9966677954898&#160;</span>28</td>
+<td align="right">53,819.16</td>
+<td align="right">139,391</td>
+<td align="center"><span style="" class="sortkey">!B9966327041700&#160;</span>29</td>
+<td align="right">48,617.91</td>
+<td align="right">125,920</td>
+<td align="right"><span style="display:none" class="sortkey">7001903400000000000</span>90.34%</td>
+<td align="right">5,201.25</td>
+<td align="right">13,471</td>
+<td align="right"><span style="display:none" class="sortkey">7000966000000000000</span>9.66%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arkansas.svg/23px-Flag_of_Arkansas.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arkansas.svg/35px-Flag_of_Arkansas.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/9/9d/Flag_of_Arkansas.svg/45px-Flag_of_Arkansas.svg.png 2x" data-file-width="450" data-file-height="300" />&#160;</span><a href="/wiki/Arkansas" title="Arkansas">Arkansas</a></td>
+<td align="center"><span style="" class="sortkey">!B9966327041700&#160;</span>29</td>
+<td align="right">53,178.55</td>
+<td align="right">137,732</td>
+<td align="center"><span style="" class="sortkey">!B9967041631339&#160;</span>27</td>
+<td align="right">52,035.48</td>
+<td align="right">134,771</td>
+<td align="right"><span style="display:none" class="sortkey">7001978500000000000</span>97.85%</td>
+<td align="right">1,143.07</td>
+<td align="right">2,961</td>
+<td align="right"><span style="display:none" class="sortkey">7000215000000000000</span>2.15%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/5/5c/Flag_of_Alabama.svg/23px-Flag_of_Alabama.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/5/5c/Flag_of_Alabama.svg/35px-Flag_of_Alabama.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/5/5c/Flag_of_Alabama.svg/45px-Flag_of_Alabama.svg.png 2x" data-file-width="600" data-file-height="400" />&#160;</span><a href="/wiki/Alabama" title="Alabama">Alabama</a></td>
+<td align="center"><span style="" class="sortkey">!B9965988026183&#160;</span>30</td>
+<td align="right">52,420.07</td>
+<td align="right">135,767</td>
+<td align="center"><span style="" class="sortkey">!B9966677954898&#160;</span>28</td>
+<td align="right">50,645.33</td>
+<td align="right">131,171</td>
+<td align="right"><span style="display:none" class="sortkey">7001966100000000000</span>96.61%</td>
+<td align="right">1,774.74</td>
+<td align="right">4,597</td>
+<td align="right"><span style="display:none" class="sortkey">7000339000000000000</span>3.39%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_Louisiana.svg/23px-Flag_of_Louisiana.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_Louisiana.svg/35px-Flag_of_Louisiana.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_Louisiana.svg/46px-Flag_of_Louisiana.svg.png 2x" data-file-width="990" data-file-height="630" />&#160;</span><a href="/wiki/Louisiana" title="Louisiana">Louisiana</a></td>
+<td align="center"><span style="" class="sortkey">!B9965660127955&#160;</span>31</td>
+<td align="right">52,378.13</td>
+<td align="right">135,659</td>
+<td align="center"><span style="" class="sortkey">!B9965034924385&#160;</span>33</td>
+<td align="right">43,203.90</td>
+<td align="right">111,898</td>
+<td align="right"><span style="display:none" class="sortkey">7001824800000000000</span>82.48%</td>
+<td align="right">9,174.23</td>
+<td align="right">23,761</td>
+<td align="right"><span style="display:none" class="sortkey">7001175200000000000</span>17.52%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/42/Flag_of_Mississippi.svg/23px-Flag_of_Mississippi.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/42/Flag_of_Mississippi.svg/35px-Flag_of_Mississippi.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/42/Flag_of_Mississippi.svg/45px-Flag_of_Mississippi.svg.png 2x" data-file-width="900" data-file-height="600" />&#160;</span><a href="/wiki/Mississippi" title="Mississippi">Mississippi</a></td>
+<td align="center"><span style="" class="sortkey">!B9965342640972&#160;</span>32</td>
+<td align="right">48,431.78</td>
+<td align="right">125,438</td>
+<td align="center"><span style="" class="sortkey">!B9965660127955&#160;</span>31</td>
+<td align="right">46,923.27</td>
+<td align="right">121,531</td>
+<td align="right"><span style="display:none" class="sortkey">7001968900000000000</span>96.89%</td>
+<td align="right">1,508.51</td>
+<td align="right">3,907</td>
+<td align="right"><span style="display:none" class="sortkey">7000311000000000000</span>3.11%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Pennsylvania.svg/23px-Flag_of_Pennsylvania.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Pennsylvania.svg/35px-Flag_of_Pennsylvania.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Flag_of_Pennsylvania.svg/45px-Flag_of_Pennsylvania.svg.png 2x" data-file-width="675" data-file-height="450" />&#160;</span><a href="/wiki/Pennsylvania" title="Pennsylvania">Pennsylvania</a></td>
+<td align="center"><span style="" class="sortkey">!B9965034924385&#160;</span>33</td>
+<td align="right">46,054.35</td>
+<td align="right">119,280</td>
+<td align="center"><span style="" class="sortkey">!B9965342640972&#160;</span>32</td>
+<td align="right">44,742.70</td>
+<td align="right">115,883</td>
+<td align="right"><span style="display:none" class="sortkey">7001971500000000000</span>97.15%</td>
+<td align="right">1,311.64</td>
+<td align="right">3,397</td>
+<td align="right"><span style="display:none" class="sortkey">7000285000000000000</span>2.85%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Flag_of_Ohio.svg/23px-Flag_of_Ohio.svg.png" width="23" height="14" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Flag_of_Ohio.svg/35px-Flag_of_Ohio.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Flag_of_Ohio.svg/46px-Flag_of_Ohio.svg.png 2x" data-file-width="520" data-file-height="320" />&#160;</span><a href="/wiki/Ohio" title="Ohio">Ohio</a></td>
+<td align="center"><span style="" class="sortkey">!B9964736394753&#160;</span>34</td>
+<td align="right">44,825.58</td>
+<td align="right">116,098</td>
+<td align="center"><span style="" class="sortkey">!B9964446519385&#160;</span>35</td>
+<td align="right">40,860.69</td>
+<td align="right">105,829</td>
+<td align="right"><span style="display:none" class="sortkey">7001911500000000000</span>91.15%</td>
+<td align="right">3,964.89</td>
+<td align="right">10,269</td>
+<td align="right"><span style="display:none" class="sortkey">7000885000000000000</span>8.85%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/47/Flag_of_Virginia.svg/22px-Flag_of_Virginia.svg.png" width="22" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/47/Flag_of_Virginia.svg/34px-Flag_of_Virginia.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/47/Flag_of_Virginia.svg/44px-Flag_of_Virginia.svg.png 2x" data-file-width="670" data-file-height="460" />&#160;</span><a href="/wiki/Virginia" title="Virginia">Virginia</a></td>
+<td align="center"><span style="" class="sortkey">!B9964446519385&#160;</span>35</td>
+<td align="right">42,774.93</td>
+<td align="right">110,787</td>
+<td align="center"><span style="" class="sortkey">!B9964164810615&#160;</span>36</td>
+<td align="right">39,490.09</td>
+<td align="right">102,279</td>
+<td align="right"><span style="display:none" class="sortkey">7001923200000099999</span>92.32%</td>
+<td align="right">3,284.84</td>
+<td align="right">8,508</td>
+<td align="right"><span style="display:none" class="sortkey">7000768000000000000</span>7.68%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/9e/Flag_of_Tennessee.svg/23px-Flag_of_Tennessee.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/9/9e/Flag_of_Tennessee.svg/35px-Flag_of_Tennessee.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/9/9e/Flag_of_Tennessee.svg/46px-Flag_of_Tennessee.svg.png 2x" data-file-width="500" data-file-height="300" />&#160;</span><a href="/wiki/Tennessee" title="Tennessee">Tennessee</a></td>
+<td align="center"><span style="" class="sortkey">!B9964164810615&#160;</span>36</td>
+<td align="right">42,144.25</td>
+<td align="right">109,153</td>
+<td align="center"><span style="" class="sortkey">!B9964736394753&#160;</span>34</td>
+<td align="right">41,234.90</td>
+<td align="right">106,798</td>
+<td align="right"><span style="display:none" class="sortkey">7001978400000000000</span>97.84%</td>
+<td align="right">909.36</td>
+<td align="right">2,355</td>
+<td align="right"><span style="display:none" class="sortkey">7000216000000000000</span>2.16%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/8/8d/Flag_of_Kentucky.svg/23px-Flag_of_Kentucky.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/8/8d/Flag_of_Kentucky.svg/35px-Flag_of_Kentucky.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/8/8d/Flag_of_Kentucky.svg/46px-Flag_of_Kentucky.svg.png 2x" data-file-width="950" data-file-height="500" />&#160;</span><a href="/wiki/Kentucky" title="Kentucky">Kentucky</a></td>
+<td align="center"><span style="" class="sortkey">!B9963890820873&#160;</span>37</td>
+<td align="right">40,407.80</td>
+<td align="right">104,656</td>
+<td align="center"><span style="" class="sortkey">!B9963890820873&#160;</span>37</td>
+<td align="right">39,486.34</td>
+<td align="right">102,269</td>
+<td align="right"><span style="display:none" class="sortkey">7001977200000000000</span>97.72%</td>
+<td align="right">921.46</td>
+<td align="right">2,387</td>
+<td align="right"><span style="display:none" class="sortkey">7000227990000099999</span>2.28%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/a/ac/Flag_of_Indiana.svg/23px-Flag_of_Indiana.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/a/ac/Flag_of_Indiana.svg/35px-Flag_of_Indiana.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/a/ac/Flag_of_Indiana.svg/45px-Flag_of_Indiana.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/Indiana" title="Indiana">Indiana</a></td>
+<td align="center"><span style="" class="sortkey">!B9963624138402&#160;</span>38</td>
+<td align="right">36,419.55</td>
+<td align="right">94,326</td>
+<td align="center"><span style="" class="sortkey">!B9963624138402&#160;</span>38</td>
+<td align="right">35,826.11</td>
+<td align="right">92,789</td>
+<td align="right"><span style="display:none" class="sortkey">7001983700000000000</span>98.37%</td>
+<td align="right">593.44</td>
+<td align="right">1,537</td>
+<td align="right"><span style="display:none" class="sortkey">7000162990000000000</span>1.63%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/3/35/Flag_of_Maine.svg/23px-Flag_of_Maine.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/3/35/Flag_of_Maine.svg/35px-Flag_of_Maine.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/3/35/Flag_of_Maine.svg/45px-Flag_of_Maine.svg.png 2x" data-file-width="687" data-file-height="458" />&#160;</span><a href="/wiki/Maine" title="Maine">Maine</a></td>
+<td align="center"><span style="" class="sortkey">!B9963364383538&#160;</span>39</td>
+<td align="right">35,379.74</td>
+<td align="right">91,633</td>
+<td align="center"><span style="" class="sortkey">!B9963364383538&#160;</span>39</td>
+<td align="right">30,842.92</td>
+<td align="right">79,883</td>
+<td align="right"><span style="display:none" class="sortkey">7001871800000000000</span>87.18%</td>
+<td align="right">4,536.82</td>
+<td align="right">11,750</td>
+<td align="right"><span style="display:none" class="sortkey">7001128200000000000</span>12.82%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/6/69/Flag_of_South_Carolina.svg/23px-Flag_of_South_Carolina.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/6/69/Flag_of_South_Carolina.svg/35px-Flag_of_South_Carolina.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/6/69/Flag_of_South_Carolina.svg/45px-Flag_of_South_Carolina.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/South_Carolina" title="South Carolina">South Carolina</a></td>
+<td align="center"><span style="" class="sortkey">!B9963111205458&#160;</span>40</td>
+<td align="right">32,020.49</td>
+<td align="right">82,933</td>
+<td align="center"><span style="" class="sortkey">!B9963111205458&#160;</span>40</td>
+<td align="right">30,060.70</td>
+<td align="right">77,857</td>
+<td align="right"><span style="display:none" class="sortkey">7001938800000000000</span>93.88%</td>
+<td align="right">1,959.79</td>
+<td align="right">5,076</td>
+<td align="right"><span style="display:none" class="sortkey">7000612000000000000</span>6.12%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_West_Virginia.svg/23px-Flag_of_West_Virginia.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_West_Virginia.svg/35px-Flag_of_West_Virginia.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/2/22/Flag_of_West_Virginia.svg/46px-Flag_of_West_Virginia.svg.png 2x" data-file-width="760" data-file-height="400" />&#160;</span><a href="/wiki/West_Virginia" title="West Virginia">West Virginia</a></td>
+<td align="center"><span style="" class="sortkey">!B9962864279332&#160;</span>41</td>
+<td align="right">24,230.04</td>
+<td align="right">62,756</td>
+<td align="center"><span style="" class="sortkey">!B9962864279332&#160;</span>41</td>
+<td align="right">24,038.21</td>
+<td align="right">62,259</td>
+<td align="right"><span style="display:none" class="sortkey">7001992100000099999</span>99.21%</td>
+<td align="right">191.83</td>
+<td align="right">497</td>
+<td align="right"><span style="display:none" class="sortkey">6999790000000000000</span>0.79%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/a/a0/Flag_of_Maryland.svg/23px-Flag_of_Maryland.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/a/a0/Flag_of_Maryland.svg/35px-Flag_of_Maryland.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/a/a0/Flag_of_Maryland.svg/45px-Flag_of_Maryland.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/Maryland" title="Maryland">Maryland</a></td>
+<td align="center"><span style="" class="sortkey">!B9962623303817&#160;</span>42</td>
+<td align="right">12,405.93</td>
+<td align="right">32,131</td>
+<td align="center"><span style="" class="sortkey">!B9962623303817&#160;</span>42</td>
+<td align="right">9,707.24</td>
+<td align="right">25,142</td>
+<td align="right"><span style="display:none" class="sortkey">7001782500000000000</span>78.25%</td>
+<td align="right">2,698.69</td>
+<td align="right">6,990</td>
+<td align="right"><span style="display:none" class="sortkey">7001217500000000000</span>21.75%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/e/ef/Flag_of_Hawaii.svg/23px-Flag_of_Hawaii.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/ef/Flag_of_Hawaii.svg/35px-Flag_of_Hawaii.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/ef/Flag_of_Hawaii.svg/46px-Flag_of_Hawaii.svg.png 2x" data-file-width="1200" data-file-height="600" />&#160;</span><a href="/wiki/Hawaii" title="Hawaii">Hawaii</a></td>
+<td align="center"><span style="" class="sortkey">!B9962387998843&#160;</span>43</td>
+<td align="right">10,931.72</td>
+<td align="right">28,313</td>
+<td align="center"><span style="" class="sortkey">!B9961498523982&#160;</span>47</td>
+<td align="right">6,422.63</td>
+<td align="right">16,635</td>
+<td align="right"><span style="display:none" class="sortkey">7001587500000000000</span>58.75%</td>
+<td align="right">4,509.09</td>
+<td align="right">11,678</td>
+<td align="right"><span style="display:none" class="sortkey">7001412500000000000</span>41.25%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f2/Flag_of_Massachusetts.svg/23px-Flag_of_Massachusetts.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f2/Flag_of_Massachusetts.svg/35px-Flag_of_Massachusetts.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f2/Flag_of_Massachusetts.svg/46px-Flag_of_Massachusetts.svg.png 2x" data-file-width="1500" data-file-height="900" />&#160;</span><a href="/wiki/Massachusetts" title="Massachusetts">Massachusetts</a></td>
+<td align="center"><span style="" class="sortkey">!B9962158103660&#160;</span>44</td>
+<td align="right">10,554.39</td>
+<td align="right">27,336</td>
+<td align="center"><span style="" class="sortkey">!B9961933375102&#160;</span>45</td>
+<td align="right">7,800.06</td>
+<td align="right">20,202</td>
+<td align="right"><span style="display:none" class="sortkey">7001739000000000000</span>73.90%</td>
+<td align="right">2,754.33</td>
+<td align="right">7,134</td>
+<td align="right"><span style="display:none" class="sortkey">7001261000000000000</span>26.10%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/49/Flag_of_Vermont.svg/23px-Flag_of_Vermont.svg.png" width="23" height="14" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/49/Flag_of_Vermont.svg/35px-Flag_of_Vermont.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/49/Flag_of_Vermont.svg/46px-Flag_of_Vermont.svg.png 2x" data-file-width="750" data-file-height="450" />&#160;</span><a href="/wiki/Vermont" title="Vermont">Vermont</a></td>
+<td align="center"><span style="" class="sortkey">!B9961933375102&#160;</span>45</td>
+<td align="right">9,616.36</td>
+<td align="right">24,906</td>
+<td align="center"><span style="" class="sortkey">!B9962387998843&#160;</span>43</td>
+<td align="right">9,216.66</td>
+<td align="right">23,871</td>
+<td align="right"><span style="display:none" class="sortkey">7001958400000000000</span>95.84%</td>
+<td align="right">399.71</td>
+<td align="right">1,035</td>
+<td align="right"><span style="display:none" class="sortkey">7000416000000000000</span>4.16%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_New_Hampshire.svg/23px-Flag_of_New_Hampshire.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_New_Hampshire.svg/35px-Flag_of_New_Hampshire.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_New_Hampshire.svg/45px-Flag_of_New_Hampshire.svg.png 2x" data-file-width="660" data-file-height="440" />&#160;</span><a href="/wiki/New_Hampshire" title="New Hampshire">New Hampshire</a></td>
+<td align="center"><span style="" class="sortkey">!B9961713586035&#160;</span>46</td>
+<td align="right">9,349.16</td>
+<td align="right">24,214</td>
+<td align="center"><span style="" class="sortkey">!B9962158103660&#160;</span>44</td>
+<td align="right">8,952.65</td>
+<td align="right">23,187</td>
+<td align="right"><span style="display:none" class="sortkey">7001957600000000000</span>95.76%</td>
+<td align="right">396.51</td>
+<td align="right">1,027</td>
+<td align="right"><span style="display:none" class="sortkey">7000424000000000000</span>4.24%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/92/Flag_of_New_Jersey.svg/23px-Flag_of_New_Jersey.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/9/92/Flag_of_New_Jersey.svg/35px-Flag_of_New_Jersey.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/9/92/Flag_of_New_Jersey.svg/45px-Flag_of_New_Jersey.svg.png 2x" data-file-width="750" data-file-height="500" />&#160;</span><a href="/wiki/New_Jersey" title="New Jersey">New Jersey</a></td>
+<td align="center"><span style="" class="sortkey">!B9961498523982&#160;</span>47</td>
+<td align="right">8,722.58</td>
+<td align="right">22,591</td>
+<td align="center"><span style="" class="sortkey">!B9961713586035&#160;</span>46</td>
+<td align="right">7,354.22</td>
+<td align="right">19,047</td>
+<td align="right"><span style="display:none" class="sortkey">7001843100000000000</span>84.31%</td>
+<td align="right">1,368.36</td>
+<td align="right">3,544</td>
+<td align="right"><span style="display:none" class="sortkey">7001156900000000000</span>15.69%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/96/Flag_of_Connecticut.svg/20px-Flag_of_Connecticut.svg.png" width="20" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/9/96/Flag_of_Connecticut.svg/30px-Flag_of_Connecticut.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/9/96/Flag_of_Connecticut.svg/40px-Flag_of_Connecticut.svg.png 2x" data-file-width="594" data-file-height="459" />&#160;</span><a href="/wiki/Connecticut" title="Connecticut">Connecticut</a></td>
+<td align="center"><span style="" class="sortkey">!B9961287989890&#160;</span>48</td>
+<td align="right">5,543.41</td>
+<td align="right">14,357</td>
+<td align="center"><span style="" class="sortkey">!B9961287989890&#160;</span>48</td>
+<td align="right">4,842.36</td>
+<td align="right">12,542</td>
+<td align="right"><span style="display:none" class="sortkey">7001873500000000000</span>87.35%</td>
+<td align="right">701.06</td>
+<td align="right">1,816</td>
+<td align="right"><span style="display:none" class="sortkey">7001126500000000000</span>12.65%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/c/c6/Flag_of_Delaware.svg/23px-Flag_of_Delaware.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/c/c6/Flag_of_Delaware.svg/35px-Flag_of_Delaware.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/c/c6/Flag_of_Delaware.svg/45px-Flag_of_Delaware.svg.png 2x" data-file-width="600" data-file-height="400" />&#160;</span><a href="/wiki/Delaware" title="Delaware">Delaware</a></td>
+<td align="center"><span style="" class="sortkey">!B9961081797018&#160;</span>49</td>
+<td align="right">2,488.72</td>
+<td align="right">6,446</td>
+<td align="center"><span style="" class="sortkey">!B9961081797018&#160;</span>49</td>
+<td align="right">1,948.54</td>
+<td align="right">5,047</td>
+<td align="right"><span style="display:none" class="sortkey">7001782900000000000</span>78.29%</td>
+<td align="right">540.18</td>
+<td align="right">1,399</td>
+<td align="right"><span style="display:none" class="sortkey">7001217100000000000</span>21.71%</td>
+</tr>
+<tr valign="top">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f3/Flag_of_Rhode_Island.svg/18px-Flag_of_Rhode_Island.svg.png" width="18" height="17" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f3/Flag_of_Rhode_Island.svg/28px-Flag_of_Rhode_Island.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f3/Flag_of_Rhode_Island.svg/36px-Flag_of_Rhode_Island.svg.png 2x" data-file-width="531" data-file-height="496" />&#160;</span><a href="/wiki/Rhode_Island" title="Rhode Island">Rhode Island</a></td>
+<td align="center"><span style="" class="sortkey">!B9960879769945&#160;</span>50</td>
+<td align="right">1,544.89</td>
+<td align="right">4,001</td>
+<td align="center"><span style="" class="sortkey">!B9960879769945&#160;</span>50</td>
+<td align="right">1,033.81</td>
+<td align="right">2,678</td>
+<td align="right"><span style="display:none" class="sortkey">7001669200000000000</span>66.92%</td>
+<td align="right">511.07</td>
+<td align="right">1,324</td>
+<td align="right"><span style="display:none" class="sortkey">7001330800000000000</span>33.08%</td>
+</tr>
+<tr valign="top" style="background:lightgreen">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/3/3e/Flag_of_Washington%2C_D.C..svg/23px-Flag_of_Washington%2C_D.C..svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/3/3e/Flag_of_Washington%2C_D.C..svg/35px-Flag_of_Washington%2C_D.C..svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/3/3e/Flag_of_Washington%2C_D.C..svg/46px-Flag_of_Washington%2C_D.C..svg.png 2x" data-file-width="800" data-file-height="400" />&#160;</span><a href="/wiki/Washington,_D.C." title="Washington, D.C.">District of Columbia</a></td>
+<td align="center"></td>
+<td align="right">68.34</td>
+<td align="right">177</td>
+<td align="center"></td>
+<td align="right">61.05</td>
+<td align="right">158</td>
+<td align="right"><span style="display:none" class="sortkey">7001893300000000000</span>89.33%</td>
+<td align="right">7.29</td>
+<td align="right">19</td>
+<td align="right"><span style="display:none" class="sortkey">7001106700000000000</span>10.67%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_Puerto_Rico.svg/23px-Flag_of_Puerto_Rico.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_Puerto_Rico.svg/35px-Flag_of_Puerto_Rico.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/2/28/Flag_of_Puerto_Rico.svg/45px-Flag_of_Puerto_Rico.svg.png 2x" data-file-width="900" data-file-height="600" />&#160;</span><a href="/wiki/Puerto_Rico" title="Puerto Rico">Puerto Rico</a></td>
+<td align="center"></td>
+<td align="right">5,324.84</td>
+<td align="right">13,791</td>
+<td align="center"></td>
+<td align="right">3,423.78</td>
+<td align="right">8,868</td>
+<td align="right"><span style="display:none" class="sortkey">7001643000000000000</span>64.30%</td>
+<td align="right">1,901.07</td>
+<td align="right">4,924</td>
+<td align="right"><span style="display:none" class="sortkey">7001357000000000000</span>35.70%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_the_Northern_Mariana_Islands.svg/23px-Flag_of_the_Northern_Mariana_Islands.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_the_Northern_Mariana_Islands.svg/35px-Flag_of_the_Northern_Mariana_Islands.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/e0/Flag_of_the_Northern_Mariana_Islands.svg/46px-Flag_of_the_Northern_Mariana_Islands.svg.png 2x" data-file-width="1000" data-file-height="500" />&#160;</span><a href="/wiki/Northern_Mariana_Islands" title="Northern Mariana Islands">Northern Mariana Islands</a></td>
+<td align="center"></td>
+<td align="right">1,975.57</td>
+<td align="right">5,117</td>
+<td align="center"></td>
+<td align="right">182.33</td>
+<td align="right">472</td>
+<td align="right"><span style="display:none" class="sortkey">7000923000000000000</span>9.23%</td>
+<td align="right">1,793.24</td>
+<td align="right">4,644</td>
+<td align="right"><span style="display:none" class="sortkey">7001907700000000000</span>90.77%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Flag_of_the_United_States_Virgin_Islands.svg/23px-Flag_of_the_United_States_Virgin_Islands.svg.png" width="23" height="15" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Flag_of_the_United_States_Virgin_Islands.svg/35px-Flag_of_the_United_States_Virgin_Islands.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Flag_of_the_United_States_Virgin_Islands.svg/45px-Flag_of_the_United_States_Virgin_Islands.svg.png 2x" data-file-width="744" data-file-height="496" />&#160;</span><a href="/wiki/United_States_Virgin_Islands" title="United States Virgin Islands">United States Virgin Islands</a></td>
+<td align="center"></td>
+<td align="right">732.93</td>
+<td align="right">1,898</td>
+<td align="center"></td>
+<td align="right">134.32</td>
+<td align="right">348</td>
+<td align="right"><span style="display:none" class="sortkey">7001183309999999999</span>18.33%</td>
+<td align="right">598.61</td>
+<td align="right">1,550</td>
+<td align="right"><span style="display:none" class="sortkey">7001816700000000000</span>81.67%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/8/87/Flag_of_American_Samoa.svg/23px-Flag_of_American_Samoa.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/8/87/Flag_of_American_Samoa.svg/35px-Flag_of_American_Samoa.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/8/87/Flag_of_American_Samoa.svg/46px-Flag_of_American_Samoa.svg.png 2x" data-file-width="1000" data-file-height="500" />&#160;</span><a href="/wiki/American_Samoa" title="American Samoa">American Samoa</a></td>
+<td align="center"></td>
+<td align="right">581.05</td>
+<td align="right">1,505</td>
+<td align="center"></td>
+<td align="right">76.46</td>
+<td align="right">198</td>
+<td align="right"><span style="display:none" class="sortkey">7001131600000000000</span>13.16%</td>
+<td align="right">504.60</td>
+<td align="right">1,307</td>
+<td align="right"><span style="display:none" class="sortkey">7001868400000000000</span>86.84%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/0/07/Flag_of_Guam.svg/23px-Flag_of_Guam.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/0/07/Flag_of_Guam.svg/35px-Flag_of_Guam.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/0/07/Flag_of_Guam.svg/46px-Flag_of_Guam.svg.png 2x" data-file-width="820" data-file-height="440" />&#160;</span><a href="/wiki/Guam" title="Guam">Guam</a></td>
+<td align="center"></td>
+<td align="right">570.62</td>
+<td align="right">1,478</td>
+<td align="center"></td>
+<td align="right">209.80</td>
+<td align="right">543</td>
+<td align="right"><span style="display:none" class="sortkey">7001367700000000000</span>36.77%</td>
+<td align="right">360.82</td>
+<td align="right">935</td>
+<td align="right"><span style="display:none" class="sortkey">7001632300000000000</span>63.23%</td>
+</tr>
+<tr valign="top" style="background:beige">
+<td><span class="flagicon"><a href="/wiki/United_States" title="United States"><img alt="United States" src="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/23px-Flag_of_the_United_States.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/35px-Flag_of_the_United_States.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/46px-Flag_of_the_United_States.svg.png 2x" data-file-width="1235" data-file-height="650" /></a></span> <i><a href="/wiki/United_States_Minor_Outlying_Islands" title="United States Minor Outlying Islands">Minor Outlying Islands</a></i><sup id="cite_ref-2000census_3-0" class="reference"><a href="#cite_note-2000census-3"><span>[</span>3<span>]</span></a></sup><sup id="cite_ref-4" class="reference"><a href="#cite_note-4"><span>[</span>a<span>]</span></a></sup></td>
+<td align="center"></td>
+<td align="right">16.0</td>
+<td align="right">41</td>
+<td align="center"></td>
+<td align="right">16.0</td>
+<td align="right">41</td>
+<td align="center">—</td>
+<td align="center">—</td>
+<td align="center">—</td>
+<td align="center">—</td>
+</tr>
+<tr class="sortbottom" valign="top" style="background: #D0E6FF;">
+<td><span class="flagicon"><a href="/wiki/United_States" title="United States"><img alt="United States" src="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/23px-Flag_of_the_United_States.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/35px-Flag_of_the_United_States.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/46px-Flag_of_the_United_States.svg.png 2x" data-file-width="1235" data-file-height="650" /></a></span> <b><a href="/wiki/Contiguous_United_States" title="Contiguous United States">Contiguous United States</a></b></td>
+<td align="center"><b>Total</b></td>
+<td align="right">3,120,426.47</td>
+<td align="right">8,081,867</td>
+<td align="center"></td>
+<td align="right">2,954,841.42</td>
+<td align="right">7,653,004</td>
+<td align="right"><span style="display:none" class="sortkey">7001946900000000000</span>94.69%</td>
+<td align="right">165,584.6</td>
+<td align="right">428,862</td>
+<td align="right"><span style="display:none" class="sortkey">7000530990000099999</span>5.31%</td>
+</tr>
+<tr class="sortbottom" valign="top" style="background: #D0E6FF;">
+<td><span class="flagicon"><a href="/wiki/United_States" title="United States"><img alt="United States" src="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/23px-Flag_of_the_United_States.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/35px-Flag_of_the_United_States.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/46px-Flag_of_the_United_States.svg.png 2x" data-file-width="1235" data-file-height="650" /></a></span> <b>50 states and D.C.</b></td>
+<td align="center"><b>Total</b></td>
+<td align="right">3,796,742.23</td>
+<td align="right">9,833,517</td>
+<td align="center"></td>
+<td align="right">3,531,905.43</td>
+<td align="right">9,147,593</td>
+<td align="right"><span style="display:none" class="sortkey">7001930200000000000</span>93.02%</td>
+<td align="right">264,836.79</td>
+<td align="right">685,924</td>
+<td align="right"><span style="display:none" class="sortkey">7000698000000000000</span>6.98%</td>
+</tr>
+<tr class="sortbottom" valign="top" style="background: #D0E6FF;">
+<td><span class="flagicon"><a href="/wiki/United_States" title="United States"><img alt="United States" src="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/23px-Flag_of_the_United_States.svg.png" width="23" height="12" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/35px-Flag_of_the_United_States.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/46px-Flag_of_the_United_States.svg.png 2x" data-file-width="1235" data-file-height="650" /></a></span> <b>All U.S. territory</b></td>
+<td align="center"><b>Total</b></td>
+<td align="right">3,805,943.26</td>
+<td align="right">9,857,348</td>
+<td align="center"></td>
+<td align="right">3,535,948.12</td>
+<td align="right">9,158,064</td>
+<td align="right"><span style="display:none" class="sortkey">7001929100000000000</span>92.91%</td>
+<td align="right">269,995.13</td>
+<td align="right">699,284</td>
+<td align="right"><span style="display:none" class="sortkey">7000709000000000000</span>7.09%</td>
+</tr>
+</table>
+<h2><span class="mw-headline" id="Area_by_division">Area by division</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=2" title="Edit section: Area by division">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<table class="wikitable sortable">
+<tr>
+<th></th>
+<th colspan="3">Total area<sup id="cite_ref-2010census_2-4" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Land area<sup id="cite_ref-2010census_2-5" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Water<sup id="cite_ref-2010census_2-6" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+</tr>
+<tr>
+<th>Division</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% land</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% water</th>
+</tr>
+<tr>
+<td><a href="/wiki/East_North_Central_States" title="East North Central States">East North Central</a></td>
+<td align="center"><span style="" class="sortkey">!B9983905620875&#160;</span>5</td>
+<td align="right">301,368.57</td>
+<td align="right">780,541</td>
+<td align="center"><span style="" class="sortkey">!B9982082405307&#160;</span>6</td>
+<td align="right">242,902.44</td>
+<td align="right">629,114</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">58,466.13</td>
+<td align="right">151,427</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/East_South_Central_States" title="East South Central States">East South Central</a></td>
+<td align="center"><span style="" class="sortkey">!B9980540898509&#160;</span>7</td>
+<td align="right">183,403.89</td>
+<td align="right">475,014</td>
+<td align="center"><span style="" class="sortkey">!B9980540898509&#160;</span>7</td>
+<td align="right">178,289.83</td>
+<td align="right">461,769</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9978027754226&#160;</span>9</td>
+<td align="right">5,114.60</td>
+<td align="right">13,247</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Mid-Atlantic_States" title="Mid-Atlantic States" class="mw-redirect">Middle Atlantic</a></td>
+<td align="center"><span style="" class="sortkey">!B9979205584583&#160;</span>8</td>
+<td align="right">109,331.89</td>
+<td align="right">283,168</td>
+<td align="center"><span style="" class="sortkey">!B9979205584583&#160;</span>8</td>
+<td align="right">99,223.32</td>
+<td align="right">256,987</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9982082405307&#160;</span>6</td>
+<td align="right">10,108.57</td>
+<td align="right">26,181</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Mountain_States" title="Mountain States">Mountain</a></td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">863,564.63</td>
+<td align="right">2,236,622</td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">855,766.98</td>
+<td align="right">2,216,426</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9979205584583&#160;</span>8</td>
+<td align="right">7,797.65</td>
+<td align="right">20,196</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/New_England" title="New England">New England</a></td>
+<td align="center"><span style="" class="sortkey">!B9978027754226&#160;</span>9</td>
+<td align="right">71,987.96</td>
+<td align="right">186,448</td>
+<td align="center"><span style="" class="sortkey">!B9978027754226&#160;</span>9</td>
+<td align="right">62,668.46</td>
+<td align="right">162,311</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9980540898509&#160;</span>7</td>
+<td align="right">9,299.50</td>
+<td align="right">24,086</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Pacific_States" title="Pacific States">Pacific</a></td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">1,009,687.00</td>
+<td align="right">2,615,077</td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">895,286.33</td>
+<td align="right">2,318,781</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">114,400.67</td>
+<td align="right">296,296</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/South_Atlantic_States" title="South Atlantic States">South Atlantic</a></td>
+<td align="center"><span style="" class="sortkey">!B9982082405307&#160;</span>6</td>
+<td align="right">292,990.46</td>
+<td align="right">758,842</td>
+<td align="center"><span style="" class="sortkey">!B9983905620875&#160;</span>5</td>
+<td align="right">265,061.97</td>
+<td align="right">686,507</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">27,928.49</td>
+<td align="right">72,334</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/West_North_Central_States" title="West North Central States">West North Central</a></td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">520,355.80</td>
+<td align="right">1,347,715</td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">507,620.08</td>
+<td align="right">1,314,730</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9983905620875&#160;</span>5</td>
+<td align="right">12,735.72</td>
+<td align="right">32,985</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/West_South_Central_States" title="West South Central States">West South Central</a></td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">444,052.01</td>
+<td align="right">1,150,089</td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">425,066.01</td>
+<td align="right">1,100,916</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">18,986.00</td>
+<td align="right">49,174</td>
+<td></td>
+</tr>
+</table>
+<h2><span class="mw-headline" id="Area_by_region">Area by region</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=3" title="Edit section: Area by region">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<table class="wikitable sortable">
+<tr>
+<th></th>
+<th colspan="3">Total area<sup id="cite_ref-2010census_2-7" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Land area<sup id="cite_ref-2010census_2-8" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+<th colspan="4">Water<sup id="cite_ref-2010census_2-9" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></th>
+</tr>
+<tr>
+<th>Region</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% land</th>
+<th>Rank</th>
+<th>sq mi</th>
+<th>km²</th>
+<th>&#160;% water</th>
+</tr>
+<tr>
+<td><a href="/wiki/Midwestern_United_States" title="Midwestern United States">Midwest</a></td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">821,724.38</td>
+<td align="right">2,128,256</td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">750,522.52</td>
+<td align="right">1,943,844</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">71,201.86</td>
+<td align="right">184,412</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Northeastern_United_States" title="Northeastern United States">Northeast</a></td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">181,319.85</td>
+<td align="right">469,616</td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">161,911.78</td>
+<td align="right">419,350</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9986137056388&#160;</span>4</td>
+<td align="right">19,408.07</td>
+<td align="right">50,267</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Southern_United_States" title="Southern United States">South</a></td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">920,446.37</td>
+<td align="right">2,383,945</td>
+<td align="center"><span style="" class="sortkey">!B9993068528194&#160;</span>2</td>
+<td align="right">868,417.82</td>
+<td align="right">2,249,192</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!B9989013877113&#160;</span>3</td>
+<td align="right">52,028.55</td>
+<td align="right">134,753</td>
+<td></td>
+</tr>
+<tr>
+<td><a href="/wiki/Western_United_States" title="Western United States">West</a></td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">1,873,251.63</td>
+<td align="right">4,851,699</td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">1,751,053.31</td>
+<td align="right">4,535,207</td>
+<td></td>
+<td align="center"><span style="" class="sortkey">!C&#160;</span>1</td>
+<td align="right">122,198.32</td>
+<td align="right">316,492</td>
+<td></td>
+</tr>
+</table>
+<ul class="gallery mw-gallery-traditional">
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:28.5px auto;"><a href="/wiki/File:US_States_by_Total_Area.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/b/bb/US_States_by_Total_Area.svg/120px-US_States_by_Total_Area.svg.png" width="120" height="93" data-file-width="932" data-file-height="723" /></a></div>
+</div>
+<div class="gallerytext">
+<p>U.S. states by total area</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:28.5px auto;"><a href="/wiki/File:US_States_by_Total_Land_Area.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/5/5b/US_States_by_Total_Land_Area.svg/120px-US_States_by_Total_Land_Area.svg.png" width="120" height="93" data-file-width="932" data-file-height="723" /></a></div>
+</div>
+<div class="gallerytext">
+<p>U.S. states by land area</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:27.5px auto;"><a href="/wiki/File:US_States_by_Water_Area.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/7/70/US_States_by_Water_Area.svg/120px-US_States_by_Water_Area.svg.png" width="120" height="95" data-file-width="966" data-file-height="764" /></a></div>
+</div>
+<div class="gallerytext">
+<p>U.S. states by water area</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:27.5px auto;"><a href="/wiki/File:US_States_by_Water_Percentage.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/2d/US_States_by_Water_Percentage.svg/120px-US_States_by_Water_Percentage.svg.png" width="120" height="95" data-file-width="966" data-file-height="764" /></a></div>
+</div>
+<div class="gallerytext">
+<p>U.S. states by water percentage</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:36px auto;"><a href="/wiki/File:Map_of_USA_AK_full.png" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/8/8c/Map_of_USA_AK_full.png/120px-Map_of_USA_AK_full.png" width="120" height="78" data-file-width="284" data-file-height="184" /></a></div>
+</div>
+<div class="gallerytext">
+<p>Alaska is the largest state by total area, land area, and water area</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:37px auto;"><a href="/wiki/File:Alaska-Size.png" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/4e/Alaska-Size.png/120px-Alaska-Size.png" width="120" height="76" data-file-width="764" data-file-height="485" /></a></div>
+</div>
+<div class="gallerytext">
+<p>The area of <a href="/wiki/Alaska" title="Alaska">Alaska</a> is <span style="display:none" class="sortkey">7001180000000000000</span>18% of the area of the <a href="/wiki/United_States" title="United States">United States</a> and <span style="display:none" class="sortkey">7001210000000000000</span>21% of the area of the <a href="/wiki/Contiguous_United_States" title="Contiguous United States">contiguous United States</a></p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:36px auto;"><a href="/wiki/File:Map_of_USA_TX.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/c/cc/Map_of_USA_TX.svg/120px-Map_of_USA_TX.svg.png" width="120" height="78" data-file-width="286" data-file-height="186" /></a></div>
+</div>
+<div class="gallerytext">
+<p>The second largest state, Texas, is only <span style="display:none" class="sortkey">7001400000000000000</span>40% of the total area of the largest state, Alaska</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:36px auto;"><a href="/wiki/File:Map_of_USA_highlighting_Rhode_Island.png" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/91/Map_of_USA_highlighting_Rhode_Island.png/120px-Map_of_USA_highlighting_Rhode_Island.png" width="120" height="78" data-file-width="280" data-file-height="183" /></a></div>
+</div>
+<div class="gallerytext">
+<p>Rhode Island is the smallest state by total area and land area</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:15px auto;"><a href="/wiki/File:Map_of_California_highlighting_San_Bernardino_County.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/7/77/Map_of_California_highlighting_San_Bernardino_County.svg/104px-Map_of_California_highlighting_San_Bernardino_County.svg.png" width="104" height="120" data-file-width="9164" data-file-height="10536" /></a></div>
+</div>
+<div class="gallerytext">
+<p><a href="/wiki/San_Bernardino_County" title="San Bernardino County" class="mw-redirect">San Bernardino County</a> is the largest <a href="/wiki/County" title="County">county</a> in the U.S. and is larger than each of the nine smallest states, including larger than the four smallest states combined. (Although some of Alaska's boroughs and census areas are larger, they are not true counties.)</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:25.5px auto;"><a href="/wiki/File:Michigan.svg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/9/93/Michigan.svg/120px-Michigan.svg.png" width="120" height="99" data-file-width="741" data-file-height="612" /></a></div>
+</div>
+<div class="gallerytext">
+<p>Michigan is second (after Alaska) in water area, and first in water percentage</p>
+</div>
+</div>
+</li>
+<li class="gallerybox" style="width: 155px">
+<div style="width: 155px">
+<div class="thumb" style="width: 150px;">
+<div style="margin:22.5px auto;"><a href="/wiki/File:STS-95_Florida_From_Space.jpg" class="image"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/2/2e/STS-95_Florida_From_Space.jpg/120px-STS-95_Florida_From_Space.jpg" width="120" height="105" data-file-width="3000" data-file-height="2624" /></a></div>
+</div>
+<div class="gallerytext">
+<p>Florida is mostly a peninsula, and has the third largest water area and seventh largest water area percentage</p>
+</div>
+</div>
+</li>
+</ul>
+<h2><span class="mw-headline" id="See_also">See also</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=4" title="Edit section: See also">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<div class="noprint tright portal" style="border:solid #aaa 1px;margin:0.5em 0 0.5em 1em;">
+<table style="background:#f9f9f9;font-size:85%;line-height:110%;max-width:175px;">
+<tr valign="middle">
+<td style="text-align:center;"><a href="/wiki/File:Flag_of_the_United_States.svg" class="image"><img alt="Portal icon" src="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/32px-Flag_of_the_United_States.svg.png" width="32" height="17" class="thumbborder" srcset="//upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/48px-Flag_of_the_United_States.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/a/a4/Flag_of_the_United_States.svg/64px-Flag_of_the_United_States.svg.png 2x" data-file-width="1235" data-file-height="650" /></a></td>
+<td style="padding:0 0.2em;vertical-align:middle;font-style:italic;font-weight:bold;"><a href="/wiki/Portal:United_States" title="Portal:United States">United States portal</a></td>
+</tr>
+</table>
+</div>
+<ul>
+<li><a href="/wiki/List_of_the_largest_country_subdivisions_by_area" title="List of the largest country subdivisions by area" class="mw-redirect">List of the largest country subdivisions by area</a></li>
+<li><a href="/wiki/List_of_political_and_geographic_subdivisions_by_total_area" title="List of political and geographic subdivisions by total area">List of political and geographic subdivisions by total area</a></li>
+</ul>
+<h2><span class="mw-headline" id="Notes">Notes</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=5" title="Edit section: Notes">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<div class="reflist columns references-column-width" style="-moz-column-width: 30em; -webkit-column-width: 30em; column-width: 30em; list-style-type: lower-alpha;">
+<ol class="references">
+<li id="cite_note-4"><span class="mw-cite-backlink"><b><a href="#cite_ref-4">^</a></b></span> <span class="reference-text">Areas were not published in the 2010 census, unlike previous years, as the U.S. Census Bureau no longer collects data on the Minor Outlying Islands.<sup id="cite_ref-2010census_2-3" class="reference"><a href="#cite_note-2010census-2"><span>[</span>2<span>]</span></a></sup></span></li>
+</ol>
+</div>
+<h2><span class="mw-headline" id="References">References</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=6" title="Edit section: References">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<div class="reflist columns references-column-width" style="-moz-column-width: 30em; -webkit-column-width: 30em; column-width: 30em; list-style-type: decimal;">
+<ol class="references">
+<li id="cite_note-1"><span class="mw-cite-backlink"><b><a href="#cite_ref-1">^</a></b></span> <span class="reference-text"><a rel="nofollow" class="external text" href="http://www.census.gov/geo/www/tiger/glossry2.pdf">Census 2000 Geographic Terms and Concepts</a>, Census 2000 Geography Glossary, U.S. Census Bureau. Accessed 2007-07-10</span></li>
+<li id="cite_note-2010census-2"><span class="mw-cite-backlink">^ <a href="#cite_ref-2010census_2-0"><sup><i><b>a</b></i></sup></a> <a href="#cite_ref-2010census_2-1"><sup><i><b>b</b></i></sup></a> <a href="#cite_ref-2010census_2-2"><sup><i><b>c</b></i></sup></a> <a href="#cite_ref-2010census_2-3"><sup><i><b>d</b></i></sup></a> <a href="#cite_ref-2010census_2-4"><sup><i><b>e</b></i></sup></a> <a href="#cite_ref-2010census_2-5"><sup><i><b>f</b></i></sup></a> <a href="#cite_ref-2010census_2-6"><sup><i><b>g</b></i></sup></a> <a href="#cite_ref-2010census_2-7"><sup><i><b>h</b></i></sup></a> <a href="#cite_ref-2010census_2-8"><sup><i><b>i</b></i></sup></a> <a href="#cite_ref-2010census_2-9"><sup><i><b>j</b></i></sup></a></span> <span class="reference-text"><span class="citation web"><a rel="nofollow" class="external text" href="http://www.census.gov/prod/cen2010/cph-2-1.pdf">"United States Summary: 2010, Population and Housing Unit Counts, 2010 Census of Population and Housing"</a> (PDF). <a href="/wiki/United_States_Census_Bureau" title="United States Census Bureau">United States Census Bureau</a>. September 2012. pp.&#160;V–2, 1 &amp; 41 (Tables 1 &amp; 18)<span class="reference-accessdate">. Retrieved February 7, 2014</span>.</span><span title="ctx_ver=Z39.88-2004&amp;rfr_id=info%3Asid%2Fen.wikipedia.org%3AList+of+U.S.+states+and+territories+by+area&amp;rft.btitle=United+States+Summary%3A+2010%2C+Population+and+Housing+Unit+Counts%2C+2010+Census+of+Population+and+Housing&amp;rft.date=September+2012&amp;rft.genre=book&amp;rft_id=http%3A%2F%2Fwww.census.gov%2Fprod%2Fcen2010%2Fcph-2-1.pdf&amp;rft.pages=V-2%2C+1+%26+41+%28Tables+1+%26+18%29&amp;rft.pub=United+States+Census+Bureau&amp;rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook" class="Z3988"><span style="display:none;">&#160;</span></span></span></li>
+<li id="cite_note-2000census-3"><span class="mw-cite-backlink"><b><a href="#cite_ref-2000census_3-0">^</a></b></span> <span class="reference-text"><span class="citation web"><a rel="nofollow" class="external text" href="http://www.census.gov/prod/cen2000/phc3-us-pt1.pdf">"United States Summary: 2010, Population and Housing Unit Counts, 2000 Census of Population and Housing"</a> (PDF). United States Census Bureau. April 2004. p.&#160;1 (Table 1)<span class="reference-accessdate">. Retrieved February 10, 2014</span>.</span><span title="ctx_ver=Z39.88-2004&amp;rfr_id=info%3Asid%2Fen.wikipedia.org%3AList+of+U.S.+states+and+territories+by+area&amp;rft.btitle=United+States+Summary%3A+2010%2C+Population+and+Housing+Unit+Counts%2C+2000+Census+of+Population+and+Housing&amp;rft.date=April+2004&amp;rft.genre=book&amp;rft_id=http%3A%2F%2Fwww.census.gov%2Fprod%2Fcen2000%2Fphc3-us-pt1.pdf&amp;rft.pages=1+%28Table+1%29&amp;rft.pub=United+States+Census+Bureau&amp;rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook" class="Z3988"><span style="display:none;">&#160;</span></span></span></li>
+</ol>
+</div>
+<h2><span class="mw-headline" id="External_links">External links</span><span class="mw-editsection"><span class="mw-editsection-bracket">[</span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit&amp;section=7" title="Edit section: External links">edit</a><span class="mw-editsection-bracket">]</span></span></h2>
+<table class="metadata plainlinks mbox-small" style="padding:0.25em 0.5em 0.5em 0.75em;border:1px solid #aaa;background:#f9f9f9;">
+<tr style="height:25px;">
+<td colspan="2" style="padding-bottom:0.5em;border-bottom:1px solid #aaa;margin:auto;text-align:center;">Find more about <b>area</b> at Wikipedia's <a href="/wiki/Wikipedia:Wikimedia_sister_projects" title="Wikipedia:Wikimedia sister projects">sister projects</a></td>
+</tr>
+<tr style="height:25px;">
+<td style="padding-top:0.75em;"><a href="//en.wiktionary.org/wiki/Special:Search/area" title="Search Wiktionary"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Wiktionary-logo-en.svg/23px-Wiktionary-logo-en.svg.png" width="23" height="25" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Wiktionary-logo-en.svg/35px-Wiktionary-logo-en.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/f8/Wiktionary-logo-en.svg/46px-Wiktionary-logo-en.svg.png 2x" data-file-width="1000" data-file-height="1089" /></a></td>
+<td style="padding-top:0.75em;"><a href="//en.wiktionary.org/wiki/Special:Search/area" class="extiw" title="wikt:Special:Search/area">Definitions and translations</a> from Wiktionary</td>
+</tr>
+<tr style="height:25px;">
+<td><a href="//commons.wikimedia.org/wiki/Special:Search/area" title="Search Commons"><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/4/4a/Commons-logo.svg/18px-Commons-logo.svg.png" width="18" height="25" srcset="//upload.wikimedia.org/wikipedia/en/thumb/4/4a/Commons-logo.svg/28px-Commons-logo.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/4/4a/Commons-logo.svg/37px-Commons-logo.svg.png 2x" data-file-width="1024" data-file-height="1376" /></a></td>
+<td><a href="//commons.wikimedia.org/wiki/Special:Search/area" class="extiw" title="commons:Special:Search/area">Media</a> from Commons</td>
+</tr>
+<tr style="height:25px;">
+<td><a href="//en.wikiquote.org/wiki/Special:Search/area" title="Search Wikiquote"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikiquote-logo.svg/21px-Wikiquote-logo.svg.png" width="21" height="25" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikiquote-logo.svg/32px-Wikiquote-logo.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikiquote-logo.svg/42px-Wikiquote-logo.svg.png 2x" data-file-width="300" data-file-height="355" /></a></td>
+<td><a href="//en.wikiquote.org/wiki/Special:Search/area" class="extiw" title="q:Special:Search/area">Quotations</a> from Wikiquote</td>
+</tr>
+<tr style="height:25px;">
+<td><a href="//en.wikisource.org/wiki/Special:Search/area" title="Search Wikisource"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Wikisource-logo.svg/24px-Wikisource-logo.svg.png" width="24" height="25" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Wikisource-logo.svg/36px-Wikisource-logo.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/4/4c/Wikisource-logo.svg/48px-Wikisource-logo.svg.png 2x" data-file-width="410" data-file-height="430" /></a></td>
+<td><a href="//en.wikisource.org/wiki/Special:Search/area" class="extiw" title="s:Special:Search/area">Source texts</a> from Wikisource</td>
+</tr>
+<tr style="height:25px;">
+<td><a href="//en.wikibooks.org/wiki/Special:Search/area" title="Search Wikibooks"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikibooks-logo.svg/25px-Wikibooks-logo.svg.png" width="25" height="25" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikibooks-logo.svg/38px-Wikibooks-logo.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/f/fa/Wikibooks-logo.svg/50px-Wikibooks-logo.svg.png 2x" data-file-width="300" data-file-height="300" /></a></td>
+<td><a href="//en.wikibooks.org/wiki/Special:Search/area" class="extiw" title="b:Special:Search/area">Textbooks</a> from Wikibooks</td>
+</tr>
+<tr style="height:25px;">
+<td><a href="//en.wikiversity.org/wiki/Special:Search/area" title="Search Wikiversity"><img alt="" src="//upload.wikimedia.org/wikipedia/commons/thumb/1/1b/Wikiversity-logo-en.svg/25px-Wikiversity-logo-en.svg.png" width="25" height="23" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/1/1b/Wikiversity-logo-en.svg/38px-Wikiversity-logo-en.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/1/1b/Wikiversity-logo-en.svg/50px-Wikiversity-logo-en.svg.png 2x" data-file-width="1000" data-file-height="900" /></a></td>
+<td><a href="//en.wikiversity.org/wiki/Special:Search/area" class="extiw" title="v:Special:Search/area">Learning resources</a> from Wikiversity</td>
+</tr>
+</table>
+<table cellspacing="0" class="navbox" style="border-spacing:0;">
+<tr>
+<td style="padding:2px;">
+<table cellspacing="0" class="nowraplinks collapsible autocollapse navbox-inner" style="border-spacing:0;background:transparent;color:inherit;">
+<tr>
+<th scope="col" class="navbox-title" colspan="2">
+<div class="plainlinks hlist navbar mini">
+<ul>
+<li class="nv-view"><a href="/wiki/Template:USStateLists" title="Template:USStateLists"><span title="View this template" style=";;background:none transparent;border:none;;">v</span></a></li>
+<li class="nv-talk"><a href="/wiki/Template_talk:USStateLists" title="Template talk:USStateLists"><span title="Discuss this template" style=";;background:none transparent;border:none;;">t</span></a></li>
+<li class="nv-edit"><a class="external text" href="//en.wikipedia.org/w/index.php?title=Template:USStateLists&amp;action=edit"><span title="Edit this template" style=";;background:none transparent;border:none;;">e</span></a></li>
+</ul>
+</div>
+<div style="font-size:110%;"><a href="/wiki/List_of_U.S._state_lists" title="List of U.S. state lists" class="mw-redirect">United States state-related lists</a></div>
+</th>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<td class="navbox-abovebelow" colspan="2">
+<div><a href="/wiki/List_of_states_and_territories_of_the_United_States" title="List of states and territories of the United States">List of states and territories of the United States</a></div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Demographics</th>
+<td class="navbox-list navbox-odd hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_educational_attainment" title="List of U.S. states by educational attainment">Educational attainment</a></li>
+<li><a href="/wiki/Irreligion_in_the_United_States#Demographics" title="Irreligion in the United States">Irreligion</a></li>
+<li><a href="/wiki/List_of_U.S._states%27_largest_cities_by_population" title="List of U.S. states' largest cities by population">Five most populous cities</a></li>
+<li><a href="/wiki/List_of_the_most_populous_counties_by_U.S._state" title="List of the most populous counties by U.S. state">Most populous county</a></li>
+<li><a href="/wiki/List_of_U.S._states_and_territories_by_population#States_and_territories" title="List of U.S. states and territories by population">Population</a>
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_population_density" title="List of U.S. states by population density">Density</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_population_growth_rate" title="List of U.S. states by population growth rate">Growth rate</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_historical_population" title="List of U.S. states by historical population">Historical</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_African-American_population" title="List of U.S. states by African-American population">African American</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_Amish_population" title="List of U.S. states by Amish population">Amish</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_Hispanic_and_Latino_population" title="List of U.S. states by Hispanic and Latino population">Hispanic and Latino</a></li>
+<li><a href="/wiki/Spanish_language_in_the_United_States#Geographic_distribution" title="Spanish language in the United States">Spanish-speaking</a></li>
+</ul>
+</li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Economy</th>
+<td class="navbox-list navbox-even hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_the_number_of_billionaires" title="List of U.S. states by the number of billionaires">Billionaires</a></li>
+<li><a href="/wiki/List_of_U.S._state_budgets" title="List of U.S. state budgets">Budgets</a></li>
+<li><a href="/wiki/Federal_tax_revenue_by_state" title="Federal tax revenue by state">Federal tax revenue</a></li>
+<li><a href="/wiki/Federal_taxation_and_spending_by_state" title="Federal taxation and spending by state">Federal net taxation less spending</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_GDP" title="List of U.S. states by GDP">Gross domestic product</a>
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_economic_growth_rate" title="List of U.S. states by economic growth rate">Growth rate</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_GDP_per_capita" title="List of U.S. states by GDP per capita">Per capita</a></li>
+</ul>
+</li>
+<li><a href="/wiki/List_of_U.S._states_by_income" title="List of U.S. states by income">Income</a>
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_income#States_ranked_by_median_household_income" title="List of U.S. states by income">Household</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_income#States_ranked_by_per_capita_income" title="List of U.S. states by income">Per capita</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_Gini_coefficient" title="List of U.S. states by Gini coefficient">Inequality</a></li>
+</ul>
+</li>
+<li><a href="/wiki/Union_affiliation_by_U.S._state" title="Union affiliation by U.S. state">Labor affiliation</a></li>
+<li><a href="/wiki/List_of_U.S._state_minimum_wages" title="List of U.S. state minimum wages" class="mw-redirect">Minimum wages</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_poverty_rate" title="List of U.S. states by poverty rate">Poverty rates</a></li>
+<li><a href="/wiki/Sales_taxes_in_the_United_States#By_jurisdiction" title="Sales taxes in the United States">Sales taxes</a></li>
+<li><a href="/wiki/State_tax_levels_in_the_United_States" title="State tax levels in the United States">State income taxes</a>
+<ul>
+<li><a href="/wiki/State_income_tax#U.S._States_with_a_flat_rate_individual_income_tax" title="State income tax">Flat rate</a></li>
+<li><a href="/wiki/State_income_tax#U.S._States_with_no_individual_income_tax" title="State income tax">None</a></li>
+</ul>
+</li>
+<li><a href="/wiki/List_of_U.S._states_by_unemployment_rate" title="List of U.S. states by unemployment rate">Unemployment rates</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Environment</th>
+<td class="navbox-list navbox-odd hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_carbon_dioxide_emissions" title="List of U.S. states by carbon dioxide emissions">Carbon dioxide emissions</a></li>
+<li><a href="/wiki/List_of_U.S._state_parks" title="List of U.S. state parks">Parks</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_electricity_production_from_renewable_sources" title="List of U.S. states by electricity production from renewable sources">Renewable energy</a></li>
+<li><a href="/wiki/List_of_Superfund_sites_in_the_United_States" title="List of Superfund sites in the United States">Superfund sites</a></li>
+<li><a href="/wiki/List_of_U.S._state_and_tribal_wilderness_areas" title="List of U.S. state and tribal wilderness areas">Wildernesses</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Geography</th>
+<td class="navbox-list navbox-even hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><strong class="selflink">Area</strong></li>
+<li><a href="/wiki/List_of_U.S._states_by_coastline" title="List of U.S. states by coastline">Coastline</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_elevation" title="List of U.S. states by elevation">Elevations</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Government</th>
+<td class="navbox-list navbox-odd hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/Wikipedia:List_of_U.S._state_portals" title="Wikipedia:List of U.S. state portals">Agencies</a></li>
+<li><a href="/wiki/State_attorney_general" title="State attorney general">Attorneys general</a></li>
+<li><a href="/wiki/List_of_capitals_in_the_United_States" title="List of capitals in the United States">Capitals</a>
+<ul>
+<li><a href="/wiki/List_of_capitals_in_the_United_States#Historical_state_capitals" title="List of capitals in the United States">Historical</a></li>
+</ul>
+</li>
+<li><a href="/wiki/List_of_state_capitols_in_the_United_States" title="List of state capitols in the United States">Capitol buildings</a></li>
+<li><a href="/wiki/Comparison_of_U.S._state_governments" title="Comparison of U.S. state governments">Comparison</a></li>
+<li><a href="/wiki/List_of_counties_by_U.S._state" title="List of counties by U.S. state">Counties</a>
+<ul>
+<li><a href="/wiki/Index_of_U.S._counties" title="Index of U.S. counties">Alphabetical</a></li>
+<li><a href="/wiki/County_(United_States)#Number_of_county_equivalents_per_state" title="County (United States)">Number</a></li>
+</ul>
+</li>
+<li><a href="/wiki/List_of_current_United_States_governors" title="List of current United States governors">Governors</a></li>
+<li><a href="/wiki/List_of_United_States_state_legislatures" title="List of United States state legislatures">Legislatures</a></li>
+<li><a href="/wiki/List_of_U.S._state_libraries_and_archives" title="List of U.S. state libraries and archives">Libraries and archives</a></li>
+<li><a href="/wiki/Languages_of_the_United_States#Official_language_status" title="Languages of the United States">Official languages</a></li>
+<li><a href="/wiki/List_of_U.S._states%27_Poets_Laureate" title="List of U.S. states' Poets Laureate">Poets laureate</a></li>
+<li><a href="/wiki/State_supreme_court" title="State supreme court">State supreme courts</a></li>
+<li><a href="/wiki/State_treasurer" title="State treasurer">State treasurers</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Health</th>
+<td class="navbox-list navbox-even hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/List_of_hospitals_in_the_United_States" title="List of hospitals in the United States">Hospitals</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_American_Human_Development_Index" title="List of U.S. states by American Human Development Index">American HDI</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_life_expectancy" title="List of U.S. states by life expectancy">Life expectancy</a></li>
+<li><a href="/wiki/Obesity_in_the_United_States#Prevalence_by_state" title="Obesity in the United States">Obesity rates</a></li>
+<li><a href="/wiki/List_of_U.S._states_and_territories_by_fertility_rate" title="List of U.S. states and territories by fertility rate">Total fertility rates</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">History</th>
+<td class="navbox-list navbox-odd hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/Historic_regions_of_the_United_States" title="Historic regions of the United States">Historic regions of the United States</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_date_of_statehood" title="List of U.S. states by date of statehood">Date of statehood</a></li>
+<li><a href="/wiki/List_of_U.S._state_name_etymologies" title="List of U.S. state name etymologies">Etymologies</a></li>
+<li><a href="/wiki/List_of_capitals_in_the_United_States#Historical_state_capitals" title="List of capitals in the United States">Historical capitals</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_historical_population" title="List of U.S. states by historical population">Historical population</a></li>
+<li><a href="/wiki/List_of_U.S._state_historical_societies_and_museums" title="List of U.S. state historical societies and museums">Historical societies and museums</a></li>
+<li><a href="/wiki/List_of_U.S._National_Historic_Landmarks_by_state" title="List of U.S. National Historic Landmarks by state">National Historic Landmarks</a></li>
+<li><a href="/wiki/List_of_U.S._states_that_were_never_U.S._territories" title="List of U.S. states that were never U.S. territories">Never territories</a></li>
+<li><a href="/wiki/List_of_U.S._state_partition_proposals" title="List of U.S. state partition proposals">Partition proposals</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_date_of_statehood#Table" title="List of U.S. states by date of statehood">Preceding entities</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Law</th>
+<td class="navbox-list navbox-even hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/Abortion_in_the_United_States_by_state#State_table" title="Abortion in the United States by state">Abortion</a></li>
+<li><a href="/wiki/Ages_of_consent_in_North_America#United_States" title="Ages of consent in North America">Age of consent</a></li>
+<li><a href="/wiki/Alcohol_laws_of_the_United_States" title="Alcohol laws of the United States">Alcohol</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_Alford_plea_usage#U.S._states" title="List of U.S. states by Alford plea usage">Alford plea</a></li>
+<li><a href="/wiki/Restrictions_on_cell_phone_use_while_driving_in_the_United_States" title="Restrictions on cell phone use while driving in the United States">Cell phones and driving laws</a></li>
+<li><a href="/wiki/List_of_U.S._state_constitutions" title="List of U.S. state constitutions" class="mw-redirect">Constitutions</a></li>
+<li><a href="/wiki/Gun_laws_in_the_United_States_by_state" title="Gun laws in the United States by state">Firearms</a>
+<ul>
+<li><a href="/wiki/Gun_violence_in_the_United_States_by_state" title="Gun violence in the United States by state">Homicide</a></li>
+</ul>
+</li>
+<li><a href="/wiki/List_of_United_States_state_and_local_law_enforcement_agencies" title="List of United States state and local law enforcement agencies">Law enforcement agencies</a></li>
+<li><a href="/wiki/List_of_U.S._state_legal_codes" title="List of U.S. state legal codes">Legal codes</a></li>
+<li><a href="/wiki/Legality_of_cannabis_by_US_state" title="Legality of cannabis by US state" class="mw-redirect">Legality of cannabis</a></li>
+<li><a href="/wiki/United_States_Peace_Index" title="United States Peace Index">Peace Index</a></li>
+<li><a href="/wiki/List_of_United_States_state_prisons" title="List of United States state prisons">Prisons</a>
+<ul>
+<li><a href="/wiki/List_of_U.S._states_by_incarceration_rate" title="List of U.S. states by incarceration rate">Incarceration rate</a></li>
+</ul>
+</li>
+<li><a href="/wiki/Same-sex_marriage_status_in_the_United_States_by_state" title="Same-sex marriage status in the United States by state" class="mw-redirect">Same-sex marriage</a>
+<ul>
+<li><a href="/wiki/List_of_U.S._state_constitutional_amendments_banning_same-sex_unions_by_type" title="List of U.S. state constitutional amendments banning same-sex unions by type">Constitutional bans</a></li>
+<li><a href="/wiki/Same-sex_marriage_law_in_the_United_States_by_state" title="Same-sex marriage law in the United States by state">Law</a></li>
+</ul>
+</li>
+<li><a href="/wiki/Seat_belt_legislation_in_the_United_States#The_laws_by_state" title="Seat belt legislation in the United States">Seat belt laws</a></li>
+<li><a href="/wiki/List_of_U.S._state_constitutional_provisions_allowing_self-representation_in_state_courts" title="List of U.S. state constitutional provisions allowing self-representation in state courts">Self-representation</a></li>
+<li><a href="/wiki/List_of_smoking_bans_in_the_United_States" title="List of smoking bans in the United States">Smoking</a></li>
+</ul>
+</div>
+</td>
+</tr>
+<tr style="height:2px;">
+<td colspan="2"></td>
+</tr>
+<tr>
+<th scope="row" class="navbox-group">Miscellaneous</th>
+<td class="navbox-list navbox-odd hlist" style="text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px;">
+<div style="padding:0em 0.25em;">
+<ul>
+<li><a href="/wiki/List_of_U.S._state_abbreviations" title="List of U.S. state abbreviations">Abbreviations</a></li>
+<li><a href="/wiki/List_of_U.S._state_abbreviations" title="List of U.S. state abbreviations">Codes</a></li>
+<li><a href="/wiki/List_of_demonyms_for_U.S._states" title="List of demonyms for U.S. states">Demonyms</a></li>
+<li><a href="/wiki/List_of_U.S._state,_district,_and_territorial_insignia" title="List of U.S. state, district, and territorial insignia">Insignia</a></li>
+<li><a href="/wiki/List_of_U.S._states_by_vehicles_per_capita" title="List of U.S. states by vehicles per capita">Motor vehicles</a></li>
+<li><a href="/wiki/Wikipedia:List_of_U.S._state_portals" title="Wikipedia:List of U.S. state portals">Portals</a></li>
+<li><a href="/wiki/Lists_of_United_States_state_insignia" title="Lists of United States state insignia" class="mw-redirect">Symbols</a></li>
+<li><a href="/wiki/List_of_tallest_buildings_by_U.S._state" title="List of tallest buildings by U.S. state">Tallest buildings</a></li>
+<li><a href="/wiki/List_of_time_zones_by_U.S._state" title="List of time zones by U.S. state" class="mw-redirect">Time zones</a></li>
+<li><a href="/wiki/List_of_fictional_U.S._states" title="List of fictional U.S. states">Fictional states</a></li>
+</ul>
+</div>
+</td>
+</tr>
+</table>
+</td>
+</tr>
+</table>
+
+
+<!--
+NewPP limit report
+Parsed by terbium
+CPU time usage: 6.220 seconds
+Real time usage: 7.420 seconds
+Preprocessor visited node count: 33194/1000000
+Preprocessor generated node count: 33781/1500000
+Post‐expand include size: 222845/2048000 bytes
+Template argument size: 25771/2048000 bytes
+Highest expansion depth: 18/40
+Expensive parser function count: 2/500
+Lua time usage: 0.493/10.000 seconds
+Lua memory usage: 4.17 MB/50 MB
+-->
+
+<!-- Saved in parser cache with key enwiki:pcache:idhash:87513-0!*!0!!en!4!* and timestamp 20140725194944 and revision id 614847271
+ -->
+<noscript><img src="//en.wikipedia.org/wiki/Special:CentralAutoLogin/start?type=1x1" alt="" title="" width="1" height="1" style="border: none; position: absolute;" /></noscript></div>									<div class="printfooter">
+						Retrieved from "<a dir="ltr" href="http://en.wikipedia.org/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;oldid=614847271">http://en.wikipedia.org/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;oldid=614847271</a>"					</div>
+													<div id='catlinks' class='catlinks'><div id="mw-normal-catlinks" class="mw-normal-catlinks"><a href="/wiki/Help:Category" title="Help:Category">Categories</a>: <ul><li><a href="/wiki/Category:Geography_of_the_United_States" title="Category:Geography of the United States">Geography of the United States</a></li><li><a href="/wiki/Category:Lists_of_states_of_the_United_States" title="Category:Lists of states of the United States">Lists of states of the United States</a></li></ul></div></div>												<div class="visualClear"></div>
+							</div>
+		</div>
+		<div id="mw-navigation">
+			<h2>Navigation menu</h2>
+
+			<div id="mw-head">
+									<div id="p-personal" role="navigation" class="" aria-labelledby="p-personal-label">
+						<h3 id="p-personal-label">Personal tools</h3>
+						<ul>
+							<li id="pt-createaccount"><a href="/w/index.php?title=Special:UserLogin&amp;returnto=List+of+U.S.+states+and+territories+by+area&amp;type=signup">Create account</a></li><li id="pt-login"><a href="/w/index.php?title=Special:UserLogin&amp;returnto=List+of+U.S.+states+and+territories+by+area" title="You're encouraged to log in; however, it's not mandatory. [o]" accesskey="o">Log in</a></li>						</ul>
+					</div>
+									<div id="left-navigation">
+										<div id="p-namespaces" role="navigation" class="vectorTabs" aria-labelledby="p-namespaces-label">
+						<h3 id="p-namespaces-label">Namespaces</h3>
+						<ul>
+															<li  id="ca-nstab-main" class="selected"><span><a href="/wiki/List_of_U.S._states_and_territories_by_area"  title="View the content page [c]" accesskey="c">Article</a></span></li>
+															<li  id="ca-talk"><span><a href="/wiki/Talk:List_of_U.S._states_and_territories_by_area"  title="Discussion about the content page [t]" accesskey="t">Talk</a></span></li>
+													</ul>
+					</div>
+										<div id="p-variants" role="navigation" class="vectorMenu emptyPortlet" aria-labelledby="p-variants-label">
+						<h3 id="mw-vector-current-variant">
+													</h3>
+
+						<h3 id="p-variants-label"><span>Variants</span><a href="#"></a></h3>
+
+						<div class="menu">
+							<ul>
+															</ul>
+						</div>
+					</div>
+									</div>
+				<div id="right-navigation">
+										<div id="p-views" role="navigation" class="vectorTabs" aria-labelledby="p-views-label">
+						<h3 id="p-views-label">Views</h3>
+						<ul>
+															<li id="ca-view" class="selected"><span><a href="/wiki/List_of_U.S._states_and_territories_by_area" >Read</a></span></li>
+															<li id="ca-edit"><span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=edit"  title="You can edit this page. Please use the preview button before saving [e]" accesskey="e">Edit</a></span></li>
+															<li id="ca-history" class="collapsible"><span><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=history"  title="Past versions of this page [h]" accesskey="h">View history</a></span></li>
+													</ul>
+					</div>
+										<div id="p-cactions" role="navigation" class="vectorMenu emptyPortlet" aria-labelledby="p-cactions-label">
+						<h3 id="p-cactions-label"><span>More</span><a href="#"></a></h3>
+
+						<div class="menu">
+							<ul>
+															</ul>
+						</div>
+					</div>
+										<div id="p-search" role="search">
+						<h3>
+							<label for="searchInput">Search</label>
+						</h3>
+
+						<form action="/w/index.php" id="searchform">
+														<div id="simpleSearch">
+															<input type="search" name="search" placeholder="Search" title="Search Wikipedia [f]" accesskey="f" id="searchInput" /><input type="hidden" value="Special:Search" name="title" /><input type="submit" name="fulltext" value="Search" title="Search Wikipedia for this text" id="mw-searchButton" class="searchButton mw-fallbackSearchButton" /><input type="submit" name="go" value="Go" title="Go to a page with this exact name if one exists" id="searchButton" class="searchButton" />								</div>
+						</form>
+					</div>
+									</div>
+			</div>
+			<div id="mw-panel">
+				<div id="p-logo" role="banner"><a style="background-image: url(//upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png);" href="/wiki/Main_Page"  title="Visit the main page"></a></div>
+						<div class="portal" role="navigation" id='p-navigation' aria-labelledby='p-navigation-label'>
+			<h3 id='p-navigation-label'>Navigation</h3>
+
+			<div class="body">
+									<ul>
+													<li id="n-mainpage-description"><a href="/wiki/Main_Page" title="Visit the main page [z]" accesskey="z">Main page</a></li>
+													<li id="n-contents"><a href="/wiki/Portal:Contents" title="Guides to browsing Wikipedia">Contents</a></li>
+													<li id="n-featuredcontent"><a href="/wiki/Portal:Featured_content" title="Featured content – the best of Wikipedia">Featured content</a></li>
+													<li id="n-currentevents"><a href="/wiki/Portal:Current_events" title="Find background information on current events">Current events</a></li>
+													<li id="n-randompage"><a href="/wiki/Special:Random" title="Load a random article [x]" accesskey="x">Random article</a></li>
+													<li id="n-sitesupport"><a href="https://donate.wikimedia.org/wiki/Special:FundraiserRedirector?utm_source=donate&amp;utm_medium=sidebar&amp;utm_campaign=C13_en.wikipedia.org&amp;uselang=en" title="Support us">Donate to Wikipedia</a></li>
+													<li id="n-shoplink"><a href="//shop.wikimedia.org" title="Visit the Wikimedia Shop">Wikimedia Shop</a></li>
+											</ul>
+							</div>
+		</div>
+			<div class="portal" role="navigation" id='p-interaction' aria-labelledby='p-interaction-label'>
+			<h3 id='p-interaction-label'>Interaction</h3>
+
+			<div class="body">
+									<ul>
+													<li id="n-help"><a href="/wiki/Help:Contents" title="Guidance on how to use and edit Wikipedia">Help</a></li>
+													<li id="n-aboutsite"><a href="/wiki/Wikipedia:About" title="Find out about Wikipedia">About Wikipedia</a></li>
+													<li id="n-portal"><a href="/wiki/Wikipedia:Community_portal" title="About the project, what you can do, where to find things">Community portal</a></li>
+													<li id="n-recentchanges"><a href="/wiki/Special:RecentChanges" title="A list of recent changes in the wiki [r]" accesskey="r">Recent changes</a></li>
+													<li id="n-contactpage"><a href="//en.wikipedia.org/wiki/Wikipedia:Contact_us">Contact page</a></li>
+											</ul>
+							</div>
+		</div>
+			<div class="portal" role="navigation" id='p-tb' aria-labelledby='p-tb-label'>
+			<h3 id='p-tb-label'>Tools</h3>
+
+			<div class="body">
+									<ul>
+													<li id="t-whatlinkshere"><a href="/wiki/Special:WhatLinksHere/List_of_U.S._states_and_territories_by_area" title="List of all English Wikipedia pages containing links to this page [j]" accesskey="j">What links here</a></li>
+													<li id="t-recentchangeslinked"><a href="/wiki/Special:RecentChangesLinked/List_of_U.S._states_and_territories_by_area" title="Recent changes in pages linked from this page [k]" accesskey="k">Related changes</a></li>
+													<li id="t-upload"><a href="/wiki/Wikipedia:File_Upload_Wizard" title="Upload files [u]" accesskey="u">Upload file</a></li>
+													<li id="t-specialpages"><a href="/wiki/Special:SpecialPages" title="A list of all special pages [q]" accesskey="q">Special pages</a></li>
+													<li id="t-permalink"><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;oldid=614847271" title="Permanent link to this revision of the page">Permanent link</a></li>
+													<li id="t-info"><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;action=info">Page information</a></li>
+													<li id="t-wikibase"><a href="//www.wikidata.org/wiki/Q150340" title="Link to connected data repository item [g]" accesskey="g">Wikidata item</a></li>
+						<li id="t-cite"><a href="/w/index.php?title=Special:Cite&amp;page=List_of_U.S._states_and_territories_by_area&amp;id=614847271" title="Information on how to cite this page">Cite this page</a></li>					</ul>
+							</div>
+		</div>
+			<div class="portal" role="navigation" id='p-coll-print_export' aria-labelledby='p-coll-print_export-label'>
+			<h3 id='p-coll-print_export-label'>Print/export</h3>
+
+			<div class="body">
+									<ul>
+													<li id="coll-create_a_book"><a href="/w/index.php?title=Special:Book&amp;bookcmd=book_creator&amp;referer=List+of+U.S.+states+and+territories+by+area">Create a book</a></li>
+													<li id="coll-download-as-rl"><a href="/w/index.php?title=Special:Book&amp;bookcmd=render_article&amp;arttitle=List+of+U.S.+states+and+territories+by+area&amp;oldid=614847271&amp;writer=rl">Download as PDF</a></li>
+													<li id="t-print"><a href="/w/index.php?title=List_of_U.S._states_and_territories_by_area&amp;printable=yes" title="Printable version of this page [p]" accesskey="p">Printable version</a></li>
+											</ul>
+							</div>
+		</div>
+			<div class="portal" role="navigation" id='p-lang' aria-labelledby='p-lang-label'>
+			<h3 id='p-lang-label'>Languages</h3>
+
+			<div class="body">
+									<ul>
+													<li class="interlanguage-link interwiki-ar"><a href="//ar.wikipedia.org/wiki/%D9%85%D9%84%D8%AD%D9%82:%D9%82%D8%A7%D8%A6%D9%85%D8%A9_%D8%A7%D9%84%D9%88%D9%84%D8%A7%D9%8A%D8%A7%D8%AA_%D8%A7%D9%84%D8%A3%D9%85%D8%B1%D9%8A%D9%83%D9%8A%D8%A9_%D8%AD%D8%B3%D8%A8_%D8%A7%D9%84%D9%85%D8%B3%D8%A7%D8%AD%D8%A9" title="ملحق:قائمة الولايات الأمريكية حسب المساحة – Arabic" lang="ar" hreflang="ar">العربية</a></li>
+													<li class="interlanguage-link interwiki-bg"><a href="//bg.wikipedia.org/wiki/%D0%A1%D0%BF%D0%B8%D1%81%D1%8A%D0%BA_%D0%BD%D0%B0_%D1%89%D0%B0%D1%82%D0%B8%D1%82%D0%B5_%D0%B2_%D0%A1%D0%90%D0%A9_%D0%BF%D0%BE_%D0%BF%D0%BB%D0%BE%D1%89" title="Списък на щатите в САЩ по площ – Bulgarian" lang="bg" hreflang="bg">Български</a></li>
+													<li class="interlanguage-link interwiki-bar"><a href="//bar.wikipedia.org/wiki/Bundesstootn_vo_da_USA_noch_Fl%C3%A4chn" title="Bundesstootn vo da USA noch Flächn – Bavarian" lang="bar" hreflang="bar">Boarisch</a></li>
+													<li class="interlanguage-link interwiki-ca"><a href="//ca.wikipedia.org/wiki/Llista_d%27estats_dels_Estats_Units_per_superf%C3%ADcie" title="Llista d'estats dels Estats Units per superfície – Catalan" lang="ca" hreflang="ca">Català</a></li>
+													<li class="interlanguage-link interwiki-cs"><a href="//cs.wikipedia.org/wiki/Seznam_st%C3%A1t%C5%AF_a_teritori%C3%AD_USA_podle_rozlohy" title="Seznam států a teritorií USA podle rozlohy – Czech" lang="cs" hreflang="cs">Čeština</a></li>
+													<li class="interlanguage-link interwiki-cy"><a href="//cy.wikipedia.org/wiki/Rhestr_taleithau%27r_Unol_Daleithiau_yn_%C3%B4l_arwynebedd" title="Rhestr taleithau'r Unol Daleithiau yn ôl arwynebedd – Welsh" lang="cy" hreflang="cy">Cymraeg</a></li>
+													<li class="interlanguage-link interwiki-da"><a href="//da.wikipedia.org/wiki/USA%27s_delstater_efter_areal" title="USA's delstater efter areal – Danish" lang="da" hreflang="da">Dansk</a></li>
+													<li class="interlanguage-link interwiki-de"><a href="//de.wikipedia.org/wiki/Liste_der_Bundesstaaten_der_Vereinigten_Staaten_nach_Fl%C3%A4che" title="Liste der Bundesstaaten der Vereinigten Staaten nach Fläche – German" lang="de" hreflang="de">Deutsch</a></li>
+													<li class="interlanguage-link interwiki-el"><a href="//el.wikipedia.org/wiki/%CE%9A%CE%B1%CF%84%CE%AC%CE%BB%CE%BF%CE%B3%CE%BF%CF%82_%CF%80%CE%BF%CE%BB%CE%B9%CF%84%CE%B5%CE%B9%CF%8E%CE%BD_%CE%BA%CE%B1%CE%B9_%CE%B5%CE%B4%CE%B1%CF%86%CF%8E%CE%BD_%CF%84%CF%89%CE%BD_%CE%97%CE%A0%CE%91_%CE%B1%CE%BD%CE%AC_%CE%AD%CE%BA%CF%84%CE%B1%CF%83%CE%B7" title="Κατάλογος πολιτειών και εδαφών των ΗΠΑ ανά έκταση – Greek" lang="el" hreflang="el">Ελληνικά</a></li>
+													<li class="interlanguage-link interwiki-es"><a href="//es.wikipedia.org/wiki/Anexo:Estados_de_los_Estados_Unidos_por_superficie" title="Anexo:Estados de los Estados Unidos por superficie – Spanish" lang="es" hreflang="es">Español</a></li>
+													<li class="interlanguage-link interwiki-fr"><a href="//fr.wikipedia.org/wiki/%C3%89tats_des_%C3%89tats-Unis_par_superficie" title="États des États-Unis par superficie – French" lang="fr" hreflang="fr">Français</a></li>
+													<li class="interlanguage-link interwiki-ga"><a href="//ga.wikipedia.org/wiki/Liosta_st%C3%A1it_SAM_de_r%C3%A9ir_achair" title="Liosta stáit SAM de réir achair – Irish" lang="ga" hreflang="ga">Gaeilge</a></li>
+													<li class="interlanguage-link interwiki-gl"><a href="//gl.wikipedia.org/wiki/Lista_dos_estados_dos_EUA_por_%C3%A1rea" title="Lista dos estados dos EUA por área – Galician" lang="gl" hreflang="gl">Galego</a></li>
+													<li class="interlanguage-link interwiki-ilo"><a href="//ilo.wikipedia.org/wiki/Listaan_dagiti_estado_ken_territorio_iti_Estados_Unidos_babaen_ti_kadakkel" title="Listaan dagiti estado ken territorio iti Estados Unidos babaen ti kadakkel – Iloko" lang="ilo" hreflang="ilo">Ilokano</a></li>
+													<li class="interlanguage-link interwiki-it"><a href="//it.wikipedia.org/wiki/Stati_degli_Stati_Uniti_d%27America_per_superficie" title="Stati degli Stati Uniti d'America per superficie – Italian" lang="it" hreflang="it">Italiano</a></li>
+													<li class="interlanguage-link interwiki-jv"><a href="//jv.wikipedia.org/wiki/Dhaptar_negara_bag%C3%A9an_Am%C3%A9rika_Sar%C3%A9kat_miturut_jembar_wewengkon" title="Dhaptar negara bagéan Amérika Sarékat miturut jembar wewengkon – Javanese" lang="jv" hreflang="jv">Basa Jawa</a></li>
+													<li class="interlanguage-link interwiki-la"><a href="//la.wikipedia.org/wiki/Index_civitatum_Americae_per_superficiem" title="Index civitatum Americae per superficiem – Latin" lang="la" hreflang="la">Latina</a></li>
+													<li class="interlanguage-link interwiki-hu"><a href="//hu.wikipedia.org/wiki/Az_Amerikai_Egyes%C3%BClt_%C3%81llamok_tag%C3%A1llamainak_list%C3%A1ja_ter%C3%BClet%C3%BCk_szerint" title="Az Amerikai Egyesült Államok tagállamainak listája területük szerint – Hungarian" lang="hu" hreflang="hu">Magyar</a></li>
+													<li class="interlanguage-link interwiki-mk"><a href="//mk.wikipedia.org/wiki/%D0%A1%D0%BF%D0%B8%D1%81%D0%BE%D0%BA_%D0%BD%D0%B0_%D1%81%D0%BE%D1%98%D1%83%D0%B7%D0%BD%D0%B8_%D0%B4%D1%80%D0%B6%D0%B0%D0%B2%D0%B8_%D0%B8_%D1%82%D0%B5%D1%80%D0%B8%D1%82%D0%BE%D1%80%D0%B8%D0%B8_%D0%BD%D0%B0_%D0%A1%D0%90%D0%94_%D0%BF%D0%BE_%D0%BF%D0%BE%D0%B2%D1%80%D1%88%D0%B8%D0%BD%D0%B0" title="Список на сојузни држави и територии на САД по површина – Macedonian" lang="mk" hreflang="mk">Македонски</a></li>
+													<li class="interlanguage-link interwiki-no"><a href="//no.wikipedia.org/wiki/Liste_over_USAs_delstater_etter_areal" title="Liste over USAs delstater etter areal – Norwegian (bokmål)" lang="no" hreflang="no">Norsk bokmål</a></li>
+													<li class="interlanguage-link interwiki-pl"><a href="//pl.wikipedia.org/wiki/Stany_USA_wed%C5%82ug_powierzchni" title="Stany USA według powierzchni – Polish" lang="pl" hreflang="pl">Polski</a></li>
+													<li class="interlanguage-link interwiki-pt"><a href="//pt.wikipedia.org/wiki/Anexo:Lista_de_estados_e_territ%C3%B3rios_dos_Estados_Unidos_por_%C3%A1rea" title="Anexo:Lista de estados e territórios dos Estados Unidos por área – Portuguese" lang="pt" hreflang="pt">Português</a></li>
+													<li class="interlanguage-link interwiki-ro"><a href="//ro.wikipedia.org/wiki/List%C4%83_a_statelor_SUA_ordonate_dup%C4%83_m%C4%83rimea_suprafe%C8%9Bei" title="Listă a statelor SUA ordonate după mărimea suprafeței – Romanian" lang="ro" hreflang="ro">Română</a></li>
+													<li class="interlanguage-link interwiki-sq"><a href="//sq.wikipedia.org/wiki/Renditja_e_Shteteve_t%C3%AB_SHBA-s%C3%AB_sipas_Sip%C3%ABrfaqes" title="Renditja e Shteteve të SHBA-së sipas Sipërfaqes – Albanian" lang="sq" hreflang="sq">Shqip</a></li>
+													<li class="interlanguage-link interwiki-simple"><a href="//simple.wikipedia.org/wiki/List_of_U.S._states_by_area" title="List of U.S. states by area – Simple English" lang="simple" hreflang="simple">Simple English</a></li>
+													<li class="interlanguage-link interwiki-sk"><a href="//sk.wikipedia.org/wiki/Zoznam_%C5%A1t%C3%A1tov_a_terit%C3%B3ri%C3%AD_USA_pod%C4%BEa_rozlohy" title="Zoznam štátov a teritórií USA podľa rozlohy – Slovak" lang="sk" hreflang="sk">Slovenčina</a></li>
+													<li class="interlanguage-link interwiki-sr"><a href="//sr.wikipedia.org/wiki/%D0%A1%D0%BF%D0%B8%D1%81%D0%B0%D0%BA_%D1%81%D0%B0%D0%B2%D0%B5%D0%B7%D0%BD%D0%B8%D1%85_%D0%B4%D1%80%D0%B6%D0%B0%D0%B2%D0%B0_%D0%B8_%D1%82%D0%B5%D1%80%D0%B8%D1%82%D0%BE%D1%80%D0%B8%D1%98%D0%B0_%D0%A1%D0%90%D0%94_%D0%BF%D0%BE_%D0%BF%D0%BE%D0%B2%D1%80%D1%88%D0%B8%D0%BD%D0%B8" title="Списак савезних држава и територија САД по површини – Serbian" lang="sr" hreflang="sr">Српски / srpski</a></li>
+													<li class="interlanguage-link interwiki-fi"><a href="//fi.wikipedia.org/wiki/Luettelo_Yhdysvaltain_osavaltioista_pinta-alan_mukaan" title="Luettelo Yhdysvaltain osavaltioista pinta-alan mukaan – Finnish" lang="fi" hreflang="fi">Suomi</a></li>
+													<li class="interlanguage-link interwiki-sv"><a href="//sv.wikipedia.org/wiki/Lista_%C3%B6ver_USA:s_delstater_och_territorier_efter_yta" title="Lista över USA:s delstater och territorier efter yta – Swedish" lang="sv" hreflang="sv">Svenska</a></li>
+													<li class="interlanguage-link interwiki-ur"><a href="//ur.wikipedia.org/wiki/%D9%81%DB%81%D8%B1%D8%B3%D8%AA_%D8%A7%D9%85%D8%B1%DB%8C%DA%A9%DB%8C_%D8%B1%DB%8C%D8%A7%D8%B3%D8%AA%DB%8C%DA%BA_%D8%A7%D9%88%D8%B1_%D8%B9%D9%84%D8%A7%D9%82%DB%81_%D8%AC%D8%A7%D8%AA_%D8%A8%D9%84%D8%AD%D8%A7%D8%B8_%D8%B1%D9%82%D8%A8%DB%81" title="فہرست امریکی ریاستیں اور علاقہ جات بلحاظ رقبہ – Urdu" lang="ur" hreflang="ur">اردو</a></li>
+													<li class="interlanguage-link interwiki-vi"><a href="//vi.wikipedia.org/wiki/Danh_s%C3%A1ch_ti%E1%BB%83u_bang_Hoa_K%E1%BB%B3_theo_di%E1%BB%87n_t%C3%ADch" title="Danh sách tiểu bang Hoa Kỳ theo diện tích – Vietnamese" lang="vi" hreflang="vi">Tiếng Việt</a></li>
+													<li class="interlanguage-link interwiki-zh"><a href="//zh.wikipedia.org/wiki/%E7%BE%8E%E5%9B%BD%E5%90%84%E5%B7%9E%E9%9D%A2%E7%A7%AF%E5%88%97%E8%A1%A8" title="美国各州面积列表 – Chinese" lang="zh" hreflang="zh">中文</a></li>
+													<li class="uls-p-lang-dummy"><a href="#"></a></li>
+											</ul>
+				<div class='after-portlet after-portlet-lang'><span class="wb-langlinks-edit wb-langlinks-link"><a action="edit" href="//www.wikidata.org/wiki/Q150340#sitelinks-wikipedia" text="Edit links" title="Edit interlanguage links" class="wbc-editpage">Edit links</a></span></div>			</div>
+		</div>
+				</div>
+		</div>
+		<div id="footer" role="contentinfo">
+							<ul id="footer-info">
+											<li id="footer-info-lastmod"> This page was last modified on 29 June 2014 at 05:36.<br /></li>
+											<li id="footer-info-copyright">Text is available under the <a rel="license" href="//en.wikipedia.org/wiki/Wikipedia:Text_of_Creative_Commons_Attribution-ShareAlike_3.0_Unported_License">Creative Commons Attribution-ShareAlike License</a><a rel="license" href="//creativecommons.org/licenses/by-sa/3.0/" style="display:none;"></a>;
+additional terms may apply.  By using this site, you agree to the <a href="//wikimediafoundation.org/wiki/Terms_of_Use">Terms of Use</a> and <a href="//wikimediafoundation.org/wiki/Privacy_policy">Privacy Policy</a>. Wikipedia® is a registered trademark of the <a href="//www.wikimediafoundation.org/">Wikimedia Foundation, Inc.</a>, a non-profit organization.</li>
+									</ul>
+							<ul id="footer-places">
+											<li id="footer-places-privacy"><a href="//wikimediafoundation.org/wiki/Privacy_policy" title="wikimedia:Privacy policy">Privacy policy</a></li>
+											<li id="footer-places-about"><a href="/wiki/Wikipedia:About" title="Wikipedia:About">About Wikipedia</a></li>
+											<li id="footer-places-disclaimer"><a href="/wiki/Wikipedia:General_disclaimer" title="Wikipedia:General disclaimer">Disclaimers</a></li>
+											<li id="footer-places-contact"><a href="//en.wikipedia.org/wiki/Wikipedia:Contact_us">Contact Wikipedia</a></li>
+											<li id="footer-places-developers"><a href="https://www.mediawiki.org/wiki/Special:MyLanguage/How_to_contribute">Developers</a></li>
+											<li id="footer-places-mobileview"><a href="//en.m.wikipedia.org/wiki/List_of_U.S._states_and_territories_by_area" class="noprint stopMobileRedirectToggle">Mobile view</a></li>
+									</ul>
+										<ul id="footer-icons" class="noprint">
+											<li id="footer-copyrightico">
+															<a href="//wikimediafoundation.org/"><img src="//bits.wikimedia.org/images/wikimedia-button.png" width="88" height="31" alt="Wikimedia Foundation"/></a>
+													</li>
+											<li id="footer-poweredbyico">
+															<a href="//www.mediawiki.org/"><img src="//bits.wikimedia.org/static-1.24wmf14/skins/common/images/poweredby_mediawiki_88x31.png" alt="Powered by MediaWiki" width="88" height="31" /></a>
+													</li>
+									</ul>
+						<div style="clear:both"></div>
+		</div>
+		<script>/*<![CDATA[*/window.jQuery && jQuery.ready();/*]]>*/</script><script>if(window.mw){
+mw.loader.state({"site":"loading","user":"ready","user.groups":"ready"});
+}</script>
+<script>if(window.mw){
+mw.loader.load(["ext.cite","mediawiki.toc","mobile.desktop","mediawiki.action.view.postEdit","mediawiki.user","mediawiki.hidpi","mediawiki.page.ready","mediawiki.searchSuggest","ext.gadget.teahouse","ext.gadget.ReferenceTooltips","ext.gadget.DRN-wizard","ext.gadget.charinsert","ext.gadget.refToolbar","mmv.bootstrap.autostart","ext.eventLogging.subscriber","ext.navigationTiming","schema.UniversalLanguageSelector","ext.uls.eventlogger","ext.uls.interlanguage"],null,true);
+}</script>
+<script src="//bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&amp;lang=en&amp;modules=site&amp;only=scripts&amp;skin=vector&amp;*"></script>
+<script>if(window.mw){
+mw.config.set({"wgBackendResponseTime":215,"wgHostname":"mw1025"});
+}</script>
+	</body>
+</html>
diff --git a/pandas/pandas/tests/io/data/html_encoding/chinese_utf-16.html b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-16.html
new file mode 100644
index 0000000000000..59fffc0d19c57
Binary files /dev/null and b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-16.html differ
diff --git a/pandas/pandas/tests/io/data/html_encoding/chinese_utf-32.html b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-32.html
new file mode 100644
index 0000000000000..365c44bf08ea1
Binary files /dev/null and b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-32.html differ
diff --git a/pandas/pandas/tests/io/data/html_encoding/chinese_utf-8.html b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-8.html
new file mode 100644
index 0000000000000..ad1ca33a78a65
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html_encoding/chinese_utf-8.html
@@ -0,0 +1,26 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td> 漊煻獌</td>
+      <td> 漊煻獌</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td> 袟袘觕</td>
+      <td> 袟袘觕</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td> 埱娵徖</td>
+      <td> 埱娵徖</td>
+    </tr>
+  </tbody>
+</table>
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/data/html_encoding/letz_latin1.html b/pandas/pandas/tests/io/data/html_encoding/letz_latin1.html
new file mode 100644
index 0000000000000..7b4b99cb33388
--- /dev/null
+++ b/pandas/pandas/tests/io/data/html_encoding/letz_latin1.html
@@ -0,0 +1,26 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>  G�t</td>
+      <td>  G�t</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>   m�</td>
+      <td>   m�</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td> iech</td>
+      <td> iech</td>
+    </tr>
+  </tbody>
+</table>
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/data/legacy_pickle/1.1.0/1.1.0_x86_64_darwin_3.8.5.pickle b/pandas/pandas/tests/io/data/legacy_pickle/1.1.0/1.1.0_x86_64_darwin_3.8.5.pickle
new file mode 100644
index 0000000000000..f8df9afff6565
Binary files /dev/null and b/pandas/pandas/tests/io/data/legacy_pickle/1.1.0/1.1.0_x86_64_darwin_3.8.5.pickle differ
diff --git a/pandas/pandas/tests/io/data/legacy_pickle/1.2.4/empty_frame_v1_2_4-GH#42345.pkl b/pandas/pandas/tests/io/data/legacy_pickle/1.2.4/empty_frame_v1_2_4-GH#42345.pkl
new file mode 100644
index 0000000000000..255a745dd9021
Binary files /dev/null and b/pandas/pandas/tests/io/data/legacy_pickle/1.2.4/empty_frame_v1_2_4-GH#42345.pkl differ
diff --git a/pandas/pandas/tests/io/data/legacy_pickle/1.3.5/1.3.5_x86_64_darwin_3.10.13.pickle b/pandas/pandas/tests/io/data/legacy_pickle/1.3.5/1.3.5_x86_64_darwin_3.10.13.pickle
new file mode 100644
index 0000000000000..47152e776b8c7
Binary files /dev/null and b/pandas/pandas/tests/io/data/legacy_pickle/1.3.5/1.3.5_x86_64_darwin_3.10.13.pickle differ
diff --git a/pandas/pandas/tests/io/data/legacy_pickle/1.4.2/1.4.2_x86_64_linux_3.9.7.pickle b/pandas/pandas/tests/io/data/legacy_pickle/1.4.2/1.4.2_x86_64_linux_3.9.7.pickle
new file mode 100644
index 0000000000000..403ccf99c06d1
Binary files /dev/null and b/pandas/pandas/tests/io/data/legacy_pickle/1.4.2/1.4.2_x86_64_linux_3.9.7.pickle differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.decimal.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.decimal.orc
new file mode 100644
index 0000000000000..cb0f7b9d767a3
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.decimal.orc differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.emptyFile.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.emptyFile.orc
new file mode 100644
index 0000000000000..ecdadcbff1346
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.emptyFile.orc differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.test1.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.test1.orc
new file mode 100644
index 0000000000000..4fb0beff86897
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.test1.orc differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate1900.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate1900.orc
new file mode 100644
index 0000000000000..f51ffdbd03a43
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate1900.orc differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate2038.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate2038.orc
new file mode 100644
index 0000000000000..cd11fa8a4e91d
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.testDate2038.orc differ
diff --git a/pandas/pandas/tests/io/data/orc/TestOrcFile.testSnappy.orc b/pandas/pandas/tests/io/data/orc/TestOrcFile.testSnappy.orc
new file mode 100644
index 0000000000000..aa6cc9c9ba1a7
Binary files /dev/null and b/pandas/pandas/tests/io/data/orc/TestOrcFile.testSnappy.orc differ
diff --git a/pandas/pandas/tests/io/data/parquet/simple.parquet b/pandas/pandas/tests/io/data/parquet/simple.parquet
new file mode 100644
index 0000000000000..2862a91f508ea
Binary files /dev/null and b/pandas/pandas/tests/io/data/parquet/simple.parquet differ
diff --git a/pandas/pandas/tests/io/data/pickle/categorical.0.25.0.pickle b/pandas/pandas/tests/io/data/pickle/categorical.0.25.0.pickle
new file mode 100644
index 0000000000000..b756060c83d94
Binary files /dev/null and b/pandas/pandas/tests/io/data/pickle/categorical.0.25.0.pickle differ
diff --git a/pandas/pandas/tests/io/data/spss/labelled-num-na.sav b/pandas/pandas/tests/io/data/spss/labelled-num-na.sav
new file mode 100755
index 0000000000000..fbe6ee7767240
Binary files /dev/null and b/pandas/pandas/tests/io/data/spss/labelled-num-na.sav differ
diff --git a/pandas/pandas/tests/io/data/spss/labelled-num.sav b/pandas/pandas/tests/io/data/spss/labelled-num.sav
new file mode 100755
index 0000000000000..bfab052089d7e
Binary files /dev/null and b/pandas/pandas/tests/io/data/spss/labelled-num.sav differ
diff --git a/pandas/pandas/tests/io/data/spss/labelled-str.sav b/pandas/pandas/tests/io/data/spss/labelled-str.sav
new file mode 100755
index 0000000000000..b96a9c00fcec1
Binary files /dev/null and b/pandas/pandas/tests/io/data/spss/labelled-str.sav differ
diff --git a/pandas/pandas/tests/io/data/spss/umlauts.sav b/pandas/pandas/tests/io/data/spss/umlauts.sav
new file mode 100755
index 0000000000000..e99cf1267bebe
Binary files /dev/null and b/pandas/pandas/tests/io/data/spss/umlauts.sav differ
diff --git a/pandas/pandas/tests/io/data/stata/stata1_119.dta.gz b/pandas/pandas/tests/io/data/stata/stata1_119.dta.gz
new file mode 100644
index 0000000000000..0f75d8b92db14
Binary files /dev/null and b/pandas/pandas/tests/io/data/stata/stata1_119.dta.gz differ
diff --git a/pandas/pandas/tests/io/data/stata/stata3.csv b/pandas/pandas/tests/io/data/stata/stata3.csv
new file mode 100644
index 0000000000000..25175f7f706ed
--- /dev/null
+++ b/pandas/pandas/tests/io/data/stata/stata3.csv
@@ -0,0 +1,204 @@
+"year","quarter","realgdp","realcons","realinv","realgovt","realdpi","cpi","m1","tbilrate","unemp","pop","infl","realint"
+1959,1,2710.349,1707.4,286.898,470.045,1886.9,28.980,139.7,2.82,5.8,177.146,0,0
+1959,2,2778.801,1733.7,310.859,481.301,1919.7,29.150,141.7,3.08,5.1,177.830,2.34,0.74
+1959,3,2775.488,1751.8,289.226,491.260,1916.4,29.350,140.5,3.82,5.3,178.657,2.74,1.09
+1959,4,2785.204,1753.7,299.356,484.052,1931.3,29.370,140,4.33,5.6,179.386,0.27,4.06
+1960,1,2847.699,1770.5,331.722,462.199,1955.5,29.540,139.6,3.50,5.2,180.007,2.31,1.19
+1960,2,2834.390,1792.9,298.152,460.400,1966.1,29.550,140.2,2.68,5.2,180.671,0.14,2.55
+1960,3,2839.022,1785.8,296.375,474.676,1967.8,29.750,140.9,2.36,5.6,181.528,2.7,-0.34
+1960,4,2802.616,1788.2,259.764,476.434,1966.6,29.840,141.1,2.29,6.3,182.287,1.21,1.08
+1961,1,2819.264,1787.7,266.405,475.854,1984.5,29.810,142.1,2.37,6.8,182.992,-0.4,2.77
+1961,2,2872.005,1814.3,286.246,480.328,2014.4,29.920,142.9,2.29,7,183.691,1.47,0.81
+1961,3,2918.419,1823.1,310.227,493.828,2041.9,29.980,144.1,2.32,6.8,184.524,0.8,1.52
+1961,4,2977.830,1859.6,315.463,502.521,2082.0,30.040,145.2,2.60,6.2,185.242,0.8,1.8
+1962,1,3031.241,1879.4,334.271,520.960,2101.7,30.210,146.4,2.73,5.6,185.874,2.26,0.47
+1962,2,3064.709,1902.5,331.039,523.066,2125.2,30.220,146.5,2.78,5.5,186.538,0.13,2.65
+1962,3,3093.047,1917.9,336.962,538.838,2137.0,30.380,146.7,2.78,5.6,187.323,2.11,0.67
+1962,4,3100.563,1945.1,325.650,535.912,2154.6,30.440,148.3,2.87,5.5,188.013,0.79,2.08
+1963,1,3141.087,1958.2,343.721,522.917,2172.5,30.480,149.7,2.90,5.8,188.580,0.53,2.38
+1963,2,3180.447,1976.9,348.730,518.108,2193.1,30.690,151.3,3.03,5.7,189.242,2.75,0.29
+1963,3,3240.332,2003.8,360.102,546.893,2217.9,30.750,152.6,3.38,5.5,190.028,0.78,2.6
+1963,4,3264.967,2020.6,364.534,532.383,2254.6,30.940,153.7,3.52,5.6,190.668,2.46,1.06
+1964,1,3338.246,2060.5,379.523,529.686,2299.6,30.950,154.8,3.51,5.5,191.245,0.13,3.38
+1964,2,3376.587,2096.7,377.778,526.175,2362.1,31.020,156.8,3.47,5.2,191.889,0.9,2.57
+1964,3,3422.469,2135.2,386.754,522.008,2392.7,31.120,159.2,3.53,5,192.631,1.29,2.25
+1964,4,3431.957,2141.2,389.910,514.603,2420.4,31.280,160.7,3.76,5,193.223,2.05,1.71
+1965,1,3516.251,2188.8,429.145,508.006,2447.4,31.380,162,3.93,4.9,193.709,1.28,2.65
+1965,2,3563.960,2213.0,429.119,508.931,2474.5,31.580,163.1,3.84,4.7,194.303,2.54,1.3
+1965,3,3636.285,2251.0,444.444,529.446,2542.6,31.650,166,3.93,4.4,194.997,0.89,3.04
+1965,4,3724.014,2314.3,446.493,544.121,2594.1,31.880,169.1,4.35,4.1,195.539,2.9,1.46
+1966,1,3815.423,2348.5,484.244,556.593,2618.4,32.280,171.8,4.62,3.9,195.999,4.99,-0.37
+1966,2,3828.124,2354.5,475.408,571.371,2624.7,32.450,170.3,4.65,3.8,196.560,2.1,2.55
+1966,3,3853.301,2381.5,470.697,594.514,2657.8,32.850,171.2,5.23,3.8,197.207,4.9,0.33
+1966,4,3884.520,2391.4,472.957,599.528,2688.2,32.900,171.9,5.00,3.7,197.736,0.61,4.39
+1967,1,3918.740,2405.3,460.007,640.682,2728.4,33.100,174.2,4.22,3.8,198.206,2.42,1.8
+1967,2,3919.556,2438.1,440.393,631.430,2750.8,33.400,178.1,3.78,3.8,198.712,3.61,0.17
+1967,3,3950.826,2450.6,453.033,641.504,2777.1,33.700,181.6,4.42,3.8,199.311,3.58,0.84
+1967,4,3980.970,2465.7,462.834,640.234,2797.4,34.100,184.3,4.90,3.9,199.808,4.72,0.18
+1968,1,4063.013,2524.6,472.907,651.378,2846.2,34.400,186.6,5.18,3.7,200.208,3.5,1.67
+1968,2,4131.998,2563.3,492.026,646.145,2893.5,34.900,190.5,5.50,3.5,200.706,5.77,-0.28
+1968,3,4160.267,2611.5,476.053,640.615,2899.3,35.300,194,5.21,3.5,201.290,4.56,0.65
+1968,4,4178.293,2623.5,480.998,636.729,2918.4,35.700,198.7,5.85,3.4,201.760,4.51,1.34
+1969,1,4244.100,2652.9,512.686,633.224,2923.4,36.300,200.7,6.08,3.4,202.161,6.67,-0.58
+1969,2,4256.460,2669.8,508.601,623.160,2952.9,36.800,201.7,6.49,3.4,202.677,5.47,1.02
+1969,3,4283.378,2682.7,520.360,623.613,3012.9,37.300,202.9,7.02,3.6,203.302,5.4,1.63
+1969,4,4263.261,2704.1,492.334,606.900,3034.9,37.900,206.2,7.64,3.6,203.849,6.38,1.26
+1970,1,4256.573,2720.7,476.925,594.888,3050.1,38.500,206.7,6.76,4.2,204.401,6.28,0.47
+1970,2,4264.289,2733.2,478.419,576.257,3103.5,38.900,208,6.66,4.8,205.052,4.13,2.52
+1970,3,4302.259,2757.1,486.594,567.743,3145.4,39.400,212.9,6.15,5.2,205.788,5.11,1.04
+1970,4,4256.637,2749.6,458.406,564.666,3135.1,39.900,215.5,4.86,5.8,206.466,5.04,-0.18
+1971,1,4374.016,2802.2,517.935,542.709,3197.3,40.100,220,3.65,5.9,207.065,2,1.65
+1971,2,4398.829,2827.9,533.986,534.905,3245.3,40.600,224.9,4.76,5.9,207.661,4.96,-0.19
+1971,3,4433.943,2850.4,541.010,532.646,3259.7,40.900,227.2,4.70,6,208.345,2.94,1.75
+1971,4,4446.264,2897.8,524.085,516.140,3294.2,41.200,230.1,3.87,6,208.917,2.92,0.95
+1972,1,4525.769,2936.5,561.147,518.192,3314.9,41.500,235.6,3.55,5.8,209.386,2.9,0.64
+1972,2,4633.101,2992.6,595.495,526.473,3346.1,41.800,238.8,3.86,5.7,209.896,2.88,0.98
+1972,3,4677.503,3038.8,603.970,498.116,3414.6,42.200,245,4.47,5.6,210.479,3.81,0.66
+1972,4,4754.546,3110.1,607.104,496.540,3550.5,42.700,251.5,5.09,5.3,210.985,4.71,0.38
+1973,1,4876.166,3167.0,645.654,504.838,3590.7,43.700,252.7,5.98,5,211.420,9.26,-3.28
+1973,2,4932.571,3165.4,675.837,497.033,3626.2,44.200,257.5,7.19,4.9,211.909,4.55,2.64
+1973,3,4906.252,3176.7,649.412,475.897,3644.4,45.600,259,8.06,4.8,212.475,12.47,-4.41
+1973,4,4953.050,3167.4,674.253,476.174,3688.9,46.800,263.8,7.68,4.8,212.932,10.39,-2.71
+1974,1,4909.617,3139.7,631.230,491.043,3632.3,48.100,267.2,7.80,5.1,213.361,10.96,-3.16
+1974,2,4922.188,3150.6,628.102,490.177,3601.1,49.300,269.3,7.89,5.2,213.854,9.86,-1.96
+1974,3,4873.520,3163.6,592.672,492.586,3612.4,51.000,272.3,8.16,5.6,214.451,13.56,-5.4
+1974,4,4854.340,3117.3,598.306,496.176,3596.0,52.300,273.9,6.96,6.6,214.931,10.07,-3.11
+1975,1,4795.295,3143.4,493.212,490.603,3581.9,53.000,276.2,5.53,8.2,215.353,5.32,0.22
+1975,2,4831.942,3195.8,476.085,486.679,3749.3,54.000,283.7,5.57,8.9,215.973,7.48,-1.91
+1975,3,4913.328,3241.4,516.402,498.836,3698.6,54.900,285.4,6.27,8.5,216.587,6.61,-0.34
+1975,4,4977.511,3275.7,530.596,500.141,3736.0,55.800,288.4,5.26,8.3,217.095,6.5,-1.24
+1976,1,5090.663,3341.2,585.541,495.568,3791.0,56.100,294.7,4.91,7.7,217.528,2.14,2.77
+1976,2,5128.947,3371.8,610.513,494.532,3822.2,57.000,297.2,5.28,7.6,218.035,6.37,-1.09
+1976,3,5154.072,3407.5,611.646,493.141,3856.7,57.900,302,5.05,7.7,218.644,6.27,-1.22
+1976,4,5191.499,3451.8,615.898,494.415,3884.4,58.700,308.3,4.57,7.8,219.179,5.49,-0.92
+1977,1,5251.762,3491.3,646.198,498.509,3887.5,60.000,316,4.60,7.5,219.684,8.76,-4.16
+1977,2,5356.131,3510.6,696.141,506.695,3931.8,60.800,320.2,5.06,7.1,220.239,5.3,-0.24
+1977,3,5451.921,3544.1,734.078,509.605,3990.8,61.600,326.4,5.82,6.9,220.904,5.23,0.59
+1977,4,5450.793,3597.5,713.356,504.584,4071.2,62.700,334.4,6.20,6.6,221.477,7.08,-0.88
+1978,1,5469.405,3618.5,727.504,506.314,4096.4,63.900,339.9,6.34,6.3,221.991,7.58,-1.24
+1978,2,5684.569,3695.9,777.454,518.366,4143.4,65.500,347.6,6.72,6,222.585,9.89,-3.18
+1978,3,5740.300,3711.4,801.452,520.199,4177.1,67.100,353.3,7.64,6,223.271,9.65,-2.01
+1978,4,5816.222,3741.3,819.689,524.782,4209.8,68.500,358.6,9.02,5.9,223.865,8.26,0.76
+1979,1,5825.949,3760.2,819.556,525.524,4255.9,70.600,368,9.42,5.9,224.438,12.08,-2.66
+1979,2,5831.418,3758.0,817.660,532.040,4226.1,73.000,377.2,9.30,5.7,225.055,13.37,-4.07
+1979,3,5873.335,3794.9,801.742,531.232,4250.3,75.200,380.8,10.49,5.9,225.801,11.88,-1.38
+1979,4,5889.495,3805.0,786.817,531.126,4284.3,78.000,385.8,11.94,5.9,226.451,14.62,-2.68
+1980,1,5908.467,3798.4,781.114,548.115,4296.2,80.900,383.8,13.75,6.3,227.061,14.6,-0.85
+1980,2,5787.373,3712.2,710.640,561.895,4236.1,82.600,394,7.90,7.3,227.726,8.32,-0.42
+1980,3,5776.617,3752.0,656.477,554.292,4279.7,84.700,409,10.34,7.7,228.417,10.04,0.3
+1980,4,5883.460,3802.0,723.220,556.130,4368.1,87.200,411.3,14.75,7.4,228.937,11.64,3.11
+1981,1,6005.717,3822.8,795.091,567.618,4358.1,89.100,427.4,13.95,7.4,229.403,8.62,5.32
+1981,2,5957.795,3822.8,757.240,584.540,4358.6,91.500,426.9,15.33,7.4,229.966,10.63,4.69
+1981,3,6030.184,3838.3,804.242,583.890,4455.4,93.400,428.4,14.58,7.4,230.641,8.22,6.36
+1981,4,5955.062,3809.3,773.053,590.125,4464.4,94.400,442.7,11.33,8.2,231.157,4.26,7.07
+1982,1,5857.333,3833.9,692.514,591.043,4469.6,95.000,447.1,12.95,8.8,231.645,2.53,10.42
+1982,2,5889.074,3847.7,691.900,596.403,4500.8,97.500,448,11.97,9.4,232.188,10.39,1.58
+1982,3,5866.370,3877.2,683.825,605.370,4520.6,98.100,464.5,8.10,9.9,232.816,2.45,5.65
+1982,4,5871.001,3947.9,622.930,623.307,4536.4,97.900,477.2,7.96,10.7,233.322,-0.82,8.77
+1983,1,5944.020,3986.6,645.110,630.873,4572.2,98.800,493.2,8.22,10.4,233.781,3.66,4.56
+1983,2,6077.619,4065.7,707.372,644.322,4605.5,99.800,507.8,8.69,10.1,234.307,4.03,4.66
+1983,3,6197.468,4137.6,754.937,662.412,4674.7,100.800,517.2,8.99,9.4,234.907,3.99,5.01
+1983,4,6325.574,4203.2,834.427,639.197,4771.1,102.100,525.1,8.89,8.5,235.385,5.13,3.76
+1984,1,6448.264,4239.2,921.763,644.635,4875.4,103.300,535,9.43,7.9,235.839,4.67,4.76
+1984,2,6559.594,4299.9,952.841,664.839,4959.4,104.100,540.9,9.94,7.5,236.348,3.09,6.85
+1984,3,6623.343,4333.0,974.989,662.294,5036.6,105.100,543.7,10.19,7.4,236.976,3.82,6.37
+1984,4,6677.264,4390.1,958.993,684.282,5084.5,105.700,557,8.14,7.3,237.468,2.28,5.87
+1985,1,6740.275,4464.6,927.375,691.613,5072.0,107.000,570.4,8.25,7.3,237.900,4.89,3.36
+1985,2,6797.344,4505.2,943.383,708.524,5172.7,107.700,589.1,7.17,7.3,238.466,2.61,4.56
+1985,3,6903.523,4590.8,932.959,732.305,5140.7,108.500,607.8,7.13,7.2,239.113,2.96,4.17
+1985,4,6955.918,4600.9,969.434,732.026,5193.9,109.900,621.4,7.14,7,239.638,5.13,2.01
+1986,1,7022.757,4639.3,967.442,728.125,5255.8,108.700,641,6.56,7,240.094,-4.39,10.95
+1986,2,7050.969,4688.7,945.972,751.334,5315.5,109.500,670.3,6.06,7.2,240.651,2.93,3.13
+1986,3,7118.950,4770.7,916.315,779.770,5343.3,110.200,694.9,5.31,7,241.274,2.55,2.76
+1986,4,7153.359,4799.4,917.736,767.671,5346.5,111.400,730.2,5.44,6.8,241.784,4.33,1.1
+1987,1,7193.019,4792.1,945.776,772.247,5379.4,112.700,743.9,5.61,6.6,242.252,4.64,0.97
+1987,2,7269.510,4856.3,947.100,782.962,5321.0,113.800,743,5.67,6.3,242.804,3.89,1.79
+1987,3,7332.558,4910.4,948.055,783.804,5416.2,115.000,756.2,6.19,6,243.446,4.2,1.99
+1987,4,7458.022,4922.2,1021.980,795.467,5493.1,116.000,756.2,5.76,5.9,243.981,3.46,2.29
+1988,1,7496.600,5004.4,964.398,773.851,5562.1,117.200,768.1,5.76,5.7,244.445,4.12,1.64
+1988,2,7592.881,5040.8,987.858,765.980,5614.3,118.500,781.4,6.48,5.5,245.021,4.41,2.07
+1988,3,7632.082,5080.6,994.204,760.245,5657.5,119.900,783.3,7.22,5.5,245.693,4.7,2.52
+1988,4,7733.991,5140.4,1007.371,783.065,5708.5,121.200,785.7,8.03,5.3,246.224,4.31,3.72
+1989,1,7806.603,5159.3,1045.975,767.024,5773.4,123.100,779.2,8.67,5.2,246.721,6.22,2.44
+1989,2,7865.016,5182.4,1033.753,784.275,5749.8,124.500,777.8,8.15,5.2,247.342,4.52,3.63
+1989,3,7927.393,5236.1,1021.604,791.819,5787.0,125.400,786.6,7.76,5.3,248.067,2.88,4.88
+1989,4,7944.697,5261.7,1011.119,787.844,5831.3,127.500,795.4,7.65,5.4,248.659,6.64,1.01
+1990,1,8027.693,5303.3,1021.070,799.681,5875.1,128.900,806.2,7.80,5.3,249.306,4.37,3.44
+1990,2,8059.598,5320.8,1021.360,800.639,5913.9,130.500,810.1,7.70,5.3,250.132,4.93,2.76
+1990,3,8059.476,5341.0,997.319,793.513,5918.1,133.400,819.8,7.33,5.7,251.057,8.79,-1.46
+1990,4,7988.864,5299.5,934.248,800.525,5878.2,134.700,827.2,6.67,6.1,251.889,3.88,2.79
+1991,1,7950.164,5284.4,896.210,806.775,5896.3,135.100,843.2,5.83,6.6,252.643,1.19,4.65
+1991,2,8003.822,5324.7,891.704,809.081,5941.1,136.200,861.5,5.54,6.8,253.493,3.24,2.29
+1991,3,8037.538,5345.0,913.904,793.987,5953.6,137.200,878,5.18,6.9,254.435,2.93,2.25
+1991,4,8069.046,5342.6,948.891,778.378,5992.4,138.300,910.4,4.14,7.1,255.214,3.19,0.95
+1992,1,8157.616,5434.5,927.796,778.568,6082.9,139.400,943.8,3.88,7.4,255.992,3.17,0.71
+1992,2,8244.294,5466.7,988.912,777.762,6129.5,140.500,963.2,3.50,7.6,256.894,3.14,0.36
+1992,3,8329.361,5527.1,999.135,786.639,6160.6,141.700,1003.8,2.97,7.6,257.861,3.4,-0.44
+1992,4,8417.016,5594.6,1030.758,787.064,6248.2,142.800,1030.4,3.12,7.4,258.679,3.09,0.02
+1993,1,8432.485,5617.2,1054.979,762.901,6156.5,143.800,1047.6,2.92,7.2,259.414,2.79,0.13
+1993,2,8486.435,5671.1,1063.263,752.158,6252.3,144.500,1084.5,3.02,7.1,260.255,1.94,1.08
+1993,3,8531.108,5732.7,1062.514,744.227,6265.7,145.600,1113,3.00,6.8,261.163,3.03,-0.04
+1993,4,8643.769,5783.7,1118.583,748.102,6358.1,146.300,1131.6,3.05,6.6,261.919,1.92,1.13
+1994,1,8727.919,5848.1,1166.845,721.288,6332.6,147.200,1141.1,3.48,6.6,262.631,2.45,1.02
+1994,2,8847.303,5891.5,1234.855,717.197,6440.6,148.400,1150.5,4.20,6.2,263.436,3.25,0.96
+1994,3,8904.289,5938.7,1212.655,736.890,6487.9,149.400,1150.1,4.68,6,264.301,2.69,2
+1994,4,9003.180,5997.3,1269.190,716.702,6574.0,150.500,1151.4,5.53,5.6,265.044,2.93,2.6
+1995,1,9025.267,6004.3,1282.090,715.326,6616.6,151.800,1149.3,5.72,5.5,265.755,3.44,2.28
+1995,2,9044.668,6053.5,1247.610,712.492,6617.2,152.600,1145.4,5.52,5.7,266.557,2.1,3.42
+1995,3,9120.684,6107.6,1235.601,707.649,6666.8,153.500,1137.3,5.32,5.7,267.456,2.35,2.97
+1995,4,9184.275,6150.6,1270.392,681.081,6706.2,154.700,1123.5,5.17,5.6,268.151,3.11,2.05
+1996,1,9247.188,6206.9,1287.128,695.265,6777.7,156.100,1124.8,4.91,5.5,268.853,3.6,1.31
+1996,2,9407.052,6277.1,1353.795,705.172,6850.6,157.000,1112.4,5.09,5.5,269.667,2.3,2.79
+1996,3,9488.879,6314.6,1422.059,692.741,6908.9,158.200,1086.1,5.04,5.3,270.581,3.05,2
+1996,4,9592.458,6366.1,1418.193,690.744,6946.8,159.400,1081.5,4.99,5.3,271.360,3.02,1.97
+1997,1,9666.235,6430.2,1451.304,681.445,7008.9,159.900,1063.8,5.10,5.2,272.083,1.25,3.85
+1997,2,9809.551,6456.2,1543.976,693.525,7061.5,160.400,1066.2,5.01,5,272.912,1.25,3.76
+1997,3,9932.672,6566.0,1571.426,691.261,7142.4,161.500,1065.5,5.02,4.9,273.852,2.73,2.29
+1997,4,10008.874,6641.1,1596.523,690.311,7241.5,162.000,1074.4,5.11,4.7,274.626,1.24,3.88
+1998,1,10103.425,6707.2,1672.732,668.783,7406.2,162.200,1076.1,5.02,4.6,275.304,0.49,4.53
+1998,2,10194.277,6822.6,1652.716,687.184,7512.0,163.200,1075,4.98,4.4,276.115,2.46,2.52
+1998,3,10328.787,6913.1,1700.071,681.472,7591.0,163.900,1086,4.49,4.5,277.003,1.71,2.78
+1998,4,10507.575,7019.1,1754.743,688.147,7646.5,164.700,1097.8,4.38,4.4,277.790,1.95,2.43
+1999,1,10601.179,7088.3,1809.993,683.601,7698.4,165.900,1101.9,4.39,4.3,278.451,2.9,1.49
+1999,2,10684.049,7199.9,1803.674,683.594,7716.0,166.700,1098.7,4.54,4.3,279.295,1.92,2.62
+1999,3,10819.914,7286.4,1848.949,697.936,7765.9,168.100,1102.3,4.75,4.2,280.203,3.35,1.41
+1999,4,11014.254,7389.2,1914.567,713.445,7887.7,169.300,1121.9,5.20,4.1,280.976,2.85,2.35
+2000,1,11043.044,7501.3,1887.836,685.216,8053.4,170.900,1113.5,5.63,4,281.653,3.76,1.87
+2000,2,11258.454,7571.8,2018.529,712.641,8135.9,172.700,1103,5.81,3.9,282.385,4.19,1.62
+2000,3,11267.867,7645.9,1986.956,698.827,8222.3,173.900,1098.7,6.07,4,283.190,2.77,3.3
+2000,4,11334.544,7713.5,1987.845,695.597,8234.6,175.600,1097.7,5.70,3.9,283.900,3.89,1.81
+2001,1,11297.171,7744.3,1882.691,710.403,8296.5,176.400,1114.9,4.39,4.2,284.550,1.82,2.57
+2001,2,11371.251,7773.5,1876.650,725.623,8273.7,177.400,1139.7,3.54,4.4,285.267,2.26,1.28
+2001,3,11340.075,7807.7,1837.074,730.493,8484.5,177.600,1166,2.72,4.8,286.047,0.45,2.27
+2001,4,11380.128,7930.0,1731.189,739.318,8385.5,177.700,1190.9,1.74,5.5,286.728,0.23,1.51
+2002,1,11477.868,7957.3,1789.327,756.915,8611.6,179.300,1185.9,1.75,5.7,287.328,3.59,-1.84
+2002,2,11538.770,7997.8,1810.779,774.408,8658.9,180.000,1199.5,1.70,5.8,288.028,1.56,0.14
+2002,3,11596.430,8052.0,1814.531,786.673,8629.2,181.200,1204,1.61,5.7,288.783,2.66,-1.05
+2002,4,11598.824,8080.6,1813.219,799.967,8649.6,182.600,1226.8,1.20,5.8,289.421,3.08,-1.88
+2003,1,11645.819,8122.3,1813.141,800.196,8681.3,183.200,1248.4,1.14,5.9,290.019,1.31,-0.17
+2003,2,11738.706,8197.8,1823.698,838.775,8812.5,183.700,1287.9,0.96,6.2,290.704,1.09,-0.13
+2003,3,11935.461,8312.1,1889.883,839.598,8935.4,184.900,1297.3,0.94,6.1,291.449,2.6,-1.67
+2003,4,12042.817,8358.0,1959.783,845.722,8986.4,186.300,1306.1,0.90,5.8,292.057,3.02,-2.11
+2004,1,12127.623,8437.6,1970.015,856.570,9025.9,187.400,1332.1,0.94,5.7,292.635,2.35,-1.42
+2004,2,12213.818,8483.2,2055.580,861.440,9115.0,189.100,1340.5,1.21,5.6,293.310,3.61,-2.41
+2004,3,12303.533,8555.8,2082.231,876.385,9175.9,190.800,1361,1.63,5.4,294.066,3.58,-1.95
+2004,4,12410.282,8654.2,2125.152,865.596,9303.4,191.800,1366.6,2.20,5.4,294.741,2.09,0.11
+2005,1,12534.113,8719.0,2170.299,869.204,9189.6,193.800,1357.8,2.69,5.3,295.308,4.15,-1.46
+2005,2,12587.535,8802.9,2131.468,870.044,9253.0,194.700,1366.6,3.01,5.1,295.994,1.85,1.16
+2005,3,12683.153,8865.6,2154.949,890.394,9308.0,199.200,1375,3.52,5,296.770,9.14,-5.62
+2005,4,12748.699,8888.5,2232.193,875.557,9358.7,199.400,1380.6,4.00,4.9,297.435,0.4,3.6
+2006,1,12915.938,8986.6,2264.721,900.511,9533.8,200.700,1380.5,4.51,4.7,298.061,2.6,1.91
+2006,2,12962.462,9035.0,2261.247,892.839,9617.3,202.700,1369.2,4.82,4.7,298.766,3.97,0.85
+2006,3,12965.916,9090.7,2229.636,892.002,9662.5,201.900,1369.4,4.90,4.7,299.593,-1.58,6.48
+2006,4,13060.679,9181.6,2165.966,894.404,9788.8,203.574,1373.6,4.92,4.4,300.320,3.3,1.62
+2007,1,13099.901,9265.1,2132.609,882.766,9830.2,205.920,1379.7,4.95,4.5,300.977,4.58,0.36
+2007,2,13203.977,9291.5,2162.214,898.713,9842.7,207.338,1370,4.72,4.5,301.714,2.75,1.97
+2007,3,13321.109,9335.6,2166.491,918.983,9883.9,209.133,1379.2,4.00,4.7,302.509,3.45,0.55
+2007,4,13391.249,9363.6,2123.426,925.110,9886.2,212.495,1377.4,3.01,4.8,303.204,6.38,-3.37
+2008,1,13366.865,9349.6,2082.886,943.372,9826.8,213.997,1384,1.56,4.9,303.803,2.82,-1.26
+2008,2,13415.266,9351.0,2026.518,961.280,10059.0,218.610,1409.3,1.74,5.4,304.483,8.53,-6.79
+2008,3,13324.600,9267.7,1990.693,991.551,9838.3,216.889,1474.7,1.17,6,305.270,-3.16,4.33
+2008,4,13141.920,9195.3,1857.661,1007.273,9920.4,212.174,1576.5,0.12,6.9,305.952,-8.79,8.91
+2009,1,12925.410,9209.2,1558.494,996.287,9926.4,212.671,1592.8,0.22,8.1,306.547,0.94,-0.71
+2009,2,12901.504,9189.0,1456.678,1023.528,10077.5,214.469,1653.6,0.18,9.2,307.226,3.37,-3.19
+2009,3,12990.341,9256.0,1486.398,1044.088,10040.6,216.385,1673.9,0.12,9.6,308.013,3.56,-3.44
diff --git a/pandas/pandas/tests/io/data/stata/stata5.csv b/pandas/pandas/tests/io/data/stata/stata5.csv
new file mode 100644
index 0000000000000..8eb0c2854a740
--- /dev/null
+++ b/pandas/pandas/tests/io/data/stata/stata5.csv
@@ -0,0 +1,19 @@
+byte_,int_,long_,float_,double_,date_td,string_,string_1
+0,0,0,0,0,,"a","a"
+1,1,1,1,1,,"ab","b"
+-1,-1,-1,-1,-1,,"abc","c"
+100,32740,-2147483647,-1.70100000027769e+38,-2.0000000000000e+307,1970-01-01,"abcdefghijklmnop","d"
+-127,-32767,2147483620,1.70100000027769e+38,8.0000000000000e+307,1970-01-02,"abcdefghijklmnopqrstuvwxyz","e"
+,0,,,,2014-01-01,"ABCDEFGHIJKLMNOPQRSTUVWXYZ","f"
+0,,,,,2114-01-01,"1234567890","1"
+,,0,,,2014-12-31,"This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted","2"
+.a,.a,.a,.a,.a,2012-02-29,"!","A"
+100,32740,-2.15e+09,-1.70e+38,-2.0e+307,01jan1970,"abcdefghijklmnop","d"
+-127,-32767,2.15e+09,1.70e+38,8.0e+307,02jan1970,"abcdefghijklmnopqrstuvwxyz","e"
+,0,,,,01jan2014,"ABCDEFGHIJKLMNOPQRSTUVWXYZ","f"
+0,,,,,01jan2114,"1234567890","1"
+,,0,,,31dec2014,"This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted","2"
+.a,.a,.a,.a,.a,29feb2012,"!","A"
+.z,.z,.z,.z,.z,,"&","Z"
+,,,0,,,"1.23","!"
+,,,,0,,"10jan1970","."
diff --git a/pandas/pandas/tests/io/data/stata/stata6.csv b/pandas/pandas/tests/io/data/stata/stata6.csv
new file mode 100644
index 0000000000000..27a1dc64f530b
--- /dev/null
+++ b/pandas/pandas/tests/io/data/stata/stata6.csv
@@ -0,0 +1,6 @@
+byte_,int_,long_,float_,double_,date_td,string_,string_1
+0,0,0,0,0,1960-01-01,"a","a"
+1,1,1,1,1,3014-12-31,"ab","b"
+-1,-1,-1,-1,-1,2014-12-31,"abc","c"
+100,32740,-2147483647,-1.7010000002777e+38,-2.000000000000e+307,1970-01-01,"This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted by Stata. This string has 244 characters, so that ir is the maximum length permitted","d"
+-127,-32767,2147483620,1.7010000002777e+38,8.000000000000e+307,1970-01-02,"abcdefghijklmnopqrstuvwxyz","e"
diff --git a/pandas/pandas/tests/io/data/tar/test-csv.tar b/pandas/pandas/tests/io/data/tar/test-csv.tar
new file mode 100644
index 0000000000000..c3b3091348426
Binary files /dev/null and b/pandas/pandas/tests/io/data/tar/test-csv.tar differ
diff --git a/pandas/pandas/tests/io/data/xml/baby_names.xml b/pandas/pandas/tests/io/data/xml/baby_names.xml
new file mode 100644
index 0000000000000..b4797b79d7112
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/baby_names.xml
@@ -0,0 +1,53 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<data>
+  <row>
+    <rank>1</rank>
+    <malename>Jos�</malename>
+    <femalename>Sof�a</femalename>
+  </row>
+  <row>
+    <rank>2</rank>
+    <malename>Luis</malename>
+    <femalename>Valentina</femalename>
+  </row>
+  <row>
+    <rank>3</rank>
+    <malename>Carlos</malename>
+    <femalename>Isabella</femalename>
+  </row>
+  <row>
+    <rank>4</rank>
+    <malename>Juan</malename>
+    <femalename>Camila</femalename>
+  </row>
+  <row>
+    <rank>5</rank>
+    <malename>Jorge</malename>
+    <femalename>Valeria</femalename>
+  </row>
+  <row>
+    <rank>6</rank>
+    <malename>Pedro</malename>
+    <femalename>Mariana</femalename>
+  </row>
+  <row>
+    <rank>7</rank>
+    <malename>Jes�s</malename>
+    <femalename>Gabriela</femalename>
+  </row>
+  <row>
+    <rank>8</rank>
+    <malename>Manuel</malename>
+    <femalename>Sara</femalename>
+  </row>
+  <row>
+    <rank>9</rank>
+    <malename>Santiago</malename>
+    <femalename>Daniella</femalename>
+  </row>
+  <row>
+    <rank>10</rank>
+    <malename>Sebasti�n</malename>
+    <femalename>Mar�a Jos�</femalename>
+  </row>
+</data>
diff --git a/pandas/pandas/tests/io/data/xml/books.xml b/pandas/pandas/tests/io/data/xml/books.xml
new file mode 100644
index 0000000000000..666ce60e9a2be
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/books.xml
@@ -0,0 +1,21 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<bookstore>
+  <book category="cooking">
+    <title lang="en">Everyday Italian</title>
+    <author>Giada De Laurentiis</author>
+    <year>2005</year>
+    <price>30.00</price>
+  </book>
+  <book category="children">
+    <title lang="en">Harry Potter</title>
+    <author>J K. Rowling</author>
+    <year>2005</year>
+    <price>29.99</price>
+  </book>
+  <book category="web">
+    <title lang="en">Learning XML</title>
+    <author>Erik T. Ray</author>
+    <year>2003</year>
+    <price>39.95</price>
+  </book>
+</bookstore>
diff --git a/pandas/pandas/tests/io/data/xml/cta_rail_lines.kml b/pandas/pandas/tests/io/data/xml/cta_rail_lines.kml
new file mode 100644
index 0000000000000..c031137ee7b20
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/cta_rail_lines.kml
@@ -0,0 +1,92 @@
+<kml xmlns="http://www.opengis.net/kml/2.2"
+     xmlns:gx="http://www.google.com/kml/ext/2.2"
+     xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+     xsi:schemaLocation="http://www.opengis.net/kml/2.2 http://schemas.opengis.net/kml/2.2.0/ogckml22.xsd http://www.google.com/kml/ext/2.2 http://code.google.com/apis/kml/schema/kml22gx.xsd">
+   <Document id="CTA_RailLines">
+      <name>CTA_RailLines</name>
+      <Snippet/>
+      <Folder id="FeatureLayer0">
+         <name>CTA_RailLines</name>
+         <Snippet/>
+         <Placemark id="ID_00001">
+            <name>Blue Line (Forest Park)</name>
+            <Snippet/>
+            <description><![CDATA[<html xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:msxsl="urn:schemas-microsoft-com:xslt"> <head> <META http-equiv="Content-Type" content="text/html"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> </head> <body style="margin:0px 0px 0px 0px;overflow:auto;background:#FFFFFF;"> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-collapse:collapse;padding:3px 3px 3px 3px"> <tr style="text-align:center;font-weight:bold;background:#9CBCE2"> <td>Blue Line (Forest Park)</td> </tr> <tr> <td> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-spacing:0px; padding:3px 3px 3px 3px"> <tr> <td>OBJECTID_1</td> <td>1</td> </tr> <tr bgcolor="#D4E4F3"> <td>ASSET_ID</td> <td>21100001</td> </tr> <tr> <td>LINES</td> <td>Blue Line (Forest Park)</td> </tr> <tr bgcolor="#D4E4F3"> <td>DESCRIPTIO</td> <td>Oak Park to Austin</td> </tr> <tr> <td>TYPE</td> <td>Elevated or at Grade</td> </tr> <tr bgcolor="#D4E4F3"> <td>LEGEND</td> <td>BL</td> </tr> <tr> <td>ALT_LEGEND</td> <td>BL</td> </tr> <tr bgcolor="#D4E4F3"> <td>BRANCH</td> <td>Blue Line Forest Park</td> </tr> <tr> <td>SHAPE.LEN</td> <td>4060.368778</td> </tr> </table> </td> </tr> </table> </body> </html>]]></description>
+            <styleUrl>#LineStyle01</styleUrl>
+            <MultiGeometry>
+               <LineString>
+                  <extrude>0</extrude>
+                  <altitudeMode>clampedToGround</altitudeMode>
+                  <coordinates>-87.77678526964958,41.8708863930319,0 -87.77826234150609,41.87097820122218,0 -87.78251583439344,41.87130129991005,0 -87.78418294588424,41.87145055520308,0 -87.7872369165933,41.8717239119163,0 -87.79160214925886,41.87210797280065,0</coordinates>
+               </LineString>
+            </MultiGeometry>
+         </Placemark>
+         <Placemark id="ID_00002">
+            <name>Red, Purple Line</name>
+            <Snippet/>
+            <description><![CDATA[<html xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:msxsl="urn:schemas-microsoft-com:xslt"> <head> <META http-equiv="Content-Type" content="text/html"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> </head> <body style="margin:0px 0px 0px 0px;overflow:auto;background:#FFFFFF;"> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-collapse:collapse;padding:3px 3px 3px 3px"> <tr style="text-align:center;font-weight:bold;background:#9CBCE2"> <td>Red, Purple Line</td> </tr> <tr> <td> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-spacing:0px; padding:3px 3px 3px 3px"> <tr> <td>OBJECTID_1</td> <td>2</td> </tr> <tr bgcolor="#D4E4F3"> <td>ASSET_ID</td> <td>21100002</td> </tr> <tr> <td>LINES</td> <td>Red, Purple Line</td> </tr> <tr bgcolor="#D4E4F3"> <td>DESCRIPTIO</td> <td>Lawrence to Wilson</td> </tr> <tr> <td>TYPE</td> <td>Elevated or at Grade</td> </tr> <tr bgcolor="#D4E4F3"> <td>LEGEND</td> <td>RD</td> </tr> <tr> <td>ALT_LEGEND</td> <td>RDPR</td> </tr> <tr bgcolor="#D4E4F3"> <td>BRANCH</td> <td>Red Line North Side</td> </tr> <tr> <td>SHAPE.LEN</td> <td>1800.132896</td> </tr> </table> </td> </tr> </table> </body> </html>]]></description>
+            <styleUrl>#LineStyle01</styleUrl>
+            <MultiGeometry>
+               <LineString>
+                  <extrude>0</extrude>
+                  <altitudeMode>clampedToGround</altitudeMode>
+                  <coordinates>-87.65758750947528,41.96427269188822,0 -87.65802133507393,41.96581929055245,0 -87.65819033925305,41.96621846093642,0 -87.6583189819129,41.96650362897086,0 -87.65835858701473,41.96669002089185,0 -87.65838428411853,41.96688150295095,0 -87.65842208882658,41.96745896091846,0 -87.65846556843937,41.9683761425439,0 -87.65849296214573,41.96913893870342,0</coordinates>
+               </LineString>
+            </MultiGeometry>
+         </Placemark>
+         <Placemark id="ID_00003">
+            <name>Red, Purple Line</name>
+            <Snippet/>
+            <description><![CDATA[<html xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:msxsl="urn:schemas-microsoft-com:xslt"> <head> <META http-equiv="Content-Type" content="text/html"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> </head> <body style="margin:0px 0px 0px 0px;overflow:auto;background:#FFFFFF;"> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-collapse:collapse;padding:3px 3px 3px 3px"> <tr style="text-align:center;font-weight:bold;background:#9CBCE2"> <td>Red, Purple Line</td> </tr> <tr> <td> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-spacing:0px; padding:3px 3px 3px 3px"> <tr> <td>OBJECTID_1</td> <td>3</td> </tr> <tr bgcolor="#D4E4F3"> <td>ASSET_ID</td> <td>21100003</td> </tr> <tr> <td>LINES</td> <td>Red, Purple Line</td> </tr> <tr bgcolor="#D4E4F3"> <td>DESCRIPTIO</td> <td>Wilson to Sheridan</td> </tr> <tr> <td>TYPE</td> <td>Elevated or at Grade</td> </tr> <tr bgcolor="#D4E4F3"> <td>LEGEND</td> <td>RD</td> </tr> <tr> <td>ALT_LEGEND</td> <td>RDPR</td> </tr> <tr bgcolor="#D4E4F3"> <td>BRANCH</td> <td>Red Line North Side</td> </tr> <tr> <td>SHAPE.LEN</td> <td>4256.243677</td> </tr> </table> </td> </tr> </table> </body> </html>]]></description>
+            <styleUrl>#LineStyle01</styleUrl>
+            <MultiGeometry>
+               <LineString>
+                  <extrude>0</extrude>
+                  <altitudeMode>clampedToGround</altitudeMode>
+                  <coordinates>-87.65492939166126,41.95377494531437,0 -87.65557043199591,41.95376544118533,0 -87.65606302030132,41.95376391658746,0 -87.65623502146268,41.95377379126367,0 -87.65634748981634,41.95380103566435,0 -87.65646537904269,41.95387703994676,0 -87.65656532461145,41.95396622645799,0 -87.65664760856414,41.95404201996044,0 -87.65671750555913,41.95416647054043,0 -87.65673983607117,41.95429949810849,0 -87.65673866475777,41.95441024240925,0 -87.6567690255541,41.95490657227902,0 -87.65683672482363,41.95692259283837,0 -87.6568900886376,41.95861070983142,0 -87.65699865558875,41.96181418669004,0 -87.65756347177603,41.96397045777844,0 -87.65758750947528,41.96427269188822,0</coordinates>
+               </LineString>
+            </MultiGeometry>
+         </Placemark>
+         <Placemark id="ID_00004">
+            <name>Red, Purple Line</name>
+            <Snippet/>
+            <description><![CDATA[<html xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:msxsl="urn:schemas-microsoft-com:xslt"> <head> <META http-equiv="Content-Type" content="text/html"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> </head> <body style="margin:0px 0px 0px 0px;overflow:auto;background:#FFFFFF;"> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-collapse:collapse;padding:3px 3px 3px 3px"> <tr style="text-align:center;font-weight:bold;background:#9CBCE2"> <td>Red, Purple Line</td> </tr> <tr> <td> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-spacing:0px; padding:3px 3px 3px 3px"> <tr> <td>OBJECTID_1</td> <td>4</td> </tr> <tr bgcolor="#D4E4F3"> <td>ASSET_ID</td> <td>21100004</td> </tr> <tr> <td>LINES</td> <td>Red, Purple Line</td> </tr> <tr bgcolor="#D4E4F3"> <td>DESCRIPTIO</td> <td>Sheridan to Addison</td> </tr> <tr> <td>TYPE</td> <td>Elevated or at Grade</td> </tr> <tr bgcolor="#D4E4F3"> <td>LEGEND</td> <td>RD</td> </tr> <tr> <td>ALT_LEGEND</td> <td>RDPR</td> </tr> <tr bgcolor="#D4E4F3"> <td>BRANCH</td> <td>Red Line North Side</td> </tr> <tr> <td>SHAPE.LEN</td> <td>2581.713736</td> </tr> </table> </td> </tr> </table> </body> </html>]]></description>
+            <styleUrl>#LineStyle01</styleUrl>
+            <MultiGeometry>
+               <LineString>
+                  <extrude>0</extrude>
+                  <altitudeMode>clampedToGround</altitudeMode>
+                  <coordinates>-87.65362593118043,41.94742799535678,0 -87.65363554415794,41.94819886386848,0 -87.6536456393239,41.95059994675451,0 -87.65365831235026,41.95108288489359,0 -87.6536604873874,41.9519954657554,0 -87.65362592053201,41.95245597302328,0 -87.65367158496069,41.95311153649393,0 -87.65368468595476,41.9533202828916,0 -87.65369271253692,41.95343095587119,0 -87.65373335834569,41.95351536301472,0 -87.65378605844126,41.95358212680591,0 -87.65385067928185,41.95364452823767,0 -87.6539390793817,41.95370263886964,0 -87.6540786298351,41.95373403675265,0 -87.65430648647626,41.9537535411832,0 -87.65492939166126,41.95377494531437,0</coordinates>
+               </LineString>
+            </MultiGeometry>
+         </Placemark>
+         <Placemark id="ID_00005">
+            <name>Red, Purple Line</name>
+            <Snippet/>
+            <description><![CDATA[<html xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:msxsl="urn:schemas-microsoft-com:xslt"> <head> <META http-equiv="Content-Type" content="text/html"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> </head> <body style="margin:0px 0px 0px 0px;overflow:auto;background:#FFFFFF;"> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-collapse:collapse;padding:3px 3px 3px 3px"> <tr style="text-align:center;font-weight:bold;background:#9CBCE2"> <td>Red, Purple Line</td> </tr> <tr> <td> <table style="font-family:Arial,Verdana,Times;font-size:12px;text-align:left;width:100%;border-spacing:0px; padding:3px 3px 3px 3px"> <tr> <td>OBJECTID_1</td> <td>5</td> </tr> <tr bgcolor="#D4E4F3"> <td>ASSET_ID</td> <td>21100005</td> </tr> <tr> <td>LINES</td> <td>Red, Purple Line</td> </tr> <tr bgcolor="#D4E4F3"> <td>DESCRIPTIO</td> <td>Addison to Clark Junction</td> </tr> <tr> <td>TYPE</td> <td>Elevated or at Grade</td> </tr> <tr bgcolor="#D4E4F3"> <td>LEGEND</td> <td>RD</td> </tr> <tr> <td>ALT_LEGEND</td> <td>RDPR</td> </tr> <tr bgcolor="#D4E4F3"> <td>BRANCH</td> <td>Red Line North Side</td> </tr> <tr> <td>SHAPE.LEN</td> <td>1918.716686</td> </tr> </table> </td> </tr> </table> </body> </html>]]></description>
+            <styleUrl>#LineStyle01</styleUrl>
+            <MultiGeometry>
+               <LineString>
+                  <extrude>0</extrude>
+                  <altitudeMode>clampedToGround</altitudeMode>
+                  <coordinates>-87.65345391792157,41.94217681262115,0 -87.65342448305786,41.94237224420864,0 -87.65339745703922,41.94268217746244,0 -87.65337753982941,41.94288140770284,0 -87.65336256753105,41.94317369618263,0 -87.65338799707138,41.94357253961736,0 -87.65340240886648,41.94389158188269,0 -87.65341837392448,41.94406444407721,0 -87.65342275247338,41.94421065714904,0 -87.65347469646018,41.94434829382345,0 -87.65351486483024,41.94447699917548,0 -87.65353483605053,41.9453896864472,0 -87.65361975532807,41.94689193720703,0 -87.65362593118043,41.94742799535678,0</coordinates>
+               </LineString>
+            </MultiGeometry>
+         </Placemark>
+      </Folder>
+      <Style id="LineStyle01">
+         <LabelStyle>
+            <color>00000000</color>
+            <scale>0.000000</scale>
+         </LabelStyle>
+         <LineStyle>
+            <color>ff899e00</color>
+            <width>1.000000</width>
+         </LineStyle>
+         <PolyStyle>
+            <color>00000000</color>
+            <outline>0</outline>
+         </PolyStyle>
+      </Style>
+   </Document>
+</kml>
diff --git a/pandas/pandas/tests/io/data/xml/doc_ch_utf.xml b/pandas/pandas/tests/io/data/xml/doc_ch_utf.xml
new file mode 100644
index 0000000000000..fde215b89646b
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/doc_ch_utf.xml
@@ -0,0 +1,29 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE fragmentdoc [
+  <!ELEMENT qafragment (qa+)>
+  <!ELEMENT qa ( q, a )>
+  <!ELEMENT q ( #PCDATA | title | name)*>
+  <!ATTLIST q speaker CDATA #REQUIRED>
+  <!ELEMENT a ( #PCDATA | title | name)*>
+  <!ATTLIST a speaker CDATA #REQUIRED>
+  <!ELEMENT name (#PCDATA)>
+  <!ELEMENT title (#PCDATA)>
+  <!ENTITY C4-4F71 "Sorry, this is Big5 only">
+]>
+
+<qafragment>
+	<qa>
+		<問 speaker="Opponent">問  若箇是邪而言破邪 何者是正而道(Sorry, this is Big5 only)申正</問>
+		<答 speaker="吉藏">答  邪既無量 正亦多途  大略為言不出二種 謂有得與無得 有得是邪須破 無得是正須申
+		故<title>大品經</title> <name>善吉</name> 致問 何等是菩薩道 何等非菩薩道
+		<name>佛</name>答云  有所得非菩薩道 無所得是菩薩道</答>
+	</qa>
+	<qa>
+		<問 speaker="Opponent">問 既破有得申無得 亦應但破性執申假名以不</問>
+		<a speaker="吉藏">答 性執是有得 假名是無得  今破有得申無得 即是破性執申假名也</a>
+	</qa>
+	<qa>
+		<問 speaker="Opponent">問 既破性申假 亦應但破有申無 若有無兩洗 亦應性假雙破耶</問>
+		<答 speaker="吉藏">答  不例  有無皆是性 所以須雙破 既分性假異 故有破不破</答>
+	</qa>
+</qafragment>
diff --git a/pandas/pandas/tests/io/data/xml/flatten_doc.xsl b/pandas/pandas/tests/io/data/xml/flatten_doc.xsl
new file mode 100644
index 0000000000000..a9d62d180beaf
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/flatten_doc.xsl
@@ -0,0 +1,18 @@
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                              xmlns:k="http://www.opengis.net/kml/2.2">
+    <xsl:output method="xml" omit-xml-declaration="yes"
+                cdata-section-elements="k:description" indent="yes"/>
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="node()|@*">
+     <xsl:copy>
+       <xsl:apply-templates select="node()|@*"/>
+     </xsl:copy>
+    </xsl:template>
+
+    <xsl:template match="k:MultiGeometry|k:LineString">
+        <xsl:apply-templates select='*'/>
+    </xsl:template>
+
+    <xsl:template match="k:description|k:Snippet|k:Style"/>
+</xsl:stylesheet>
diff --git a/pandas/pandas/tests/io/data/xml/row_field_output.xsl b/pandas/pandas/tests/io/data/xml/row_field_output.xsl
new file mode 100644
index 0000000000000..5a0f0e655a78e
--- /dev/null
+++ b/pandas/pandas/tests/io/data/xml/row_field_output.xsl
@@ -0,0 +1,19 @@
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+	<xsl:output method="xml" encoding="utf-8" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+	<xsl:template match="@*|node()">
+		<xsl:copy>
+		    <xsl:apply-templates select="@*|node()"/>
+		</xsl:copy>
+	</xsl:template>
+
+	<xsl:template match="row/*">
+	    <field>
+    	    <xsl:attribute name="field">
+    	        <xsl:value-of select="name()"/>
+    	    </xsl:attribute>
+    	    <xsl:value-of select="text()"/>
+	    </field>
+	</xsl:template>
+</xsl:stylesheet>
diff --git a/pandas/pandas/tests/io/excel/__init__.py b/pandas/pandas/tests/io/excel/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/excel/test_odf.py b/pandas/pandas/tests/io/excel/test_odf.py
new file mode 100644
index 0000000000000..f01827fa4ca2f
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_odf.py
@@ -0,0 +1,72 @@
+import functools
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+pytest.importorskip("odf")
+
+
+@pytest.fixture(autouse=True)
+def cd_and_set_engine(monkeypatch, datapath):
+    func = functools.partial(pd.read_excel, engine="odf")
+    monkeypatch.setattr(pd, "read_excel", func)
+    monkeypatch.chdir(datapath("io", "data", "excel"))
+
+
+def test_read_invalid_types_raises():
+    # the invalid_value_type.ods required manually editing
+    # of the included content.xml file
+    with pytest.raises(ValueError, match="Unrecognized type awesome_new_type"):
+        pd.read_excel("invalid_value_type.ods")
+
+
+def test_read_writer_table():
+    # Also test reading tables from an text OpenDocument file
+    # (.odt)
+    index = pd.Index(["Row 1", "Row 2", "Row 3"], name="Header")
+    expected = pd.DataFrame(
+        [[1, np.nan, 7], [2, np.nan, 8], [3, np.nan, 9]],
+        index=index,
+        columns=["Column 1", "Unnamed: 2", "Column 3"],
+    )
+
+    result = pd.read_excel("writertable.odt", sheet_name="Table1", index_col=0)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_newlines_between_xml_elements_table():
+    # GH#45598
+    expected = pd.DataFrame(
+        [[1.0, 4.0, 7], [np.nan, np.nan, 8], [3.0, 6.0, 9]],
+        columns=["Column 1", "Column 2", "Column 3"],
+    )
+
+    result = pd.read_excel("test_newlines.ods")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_unempty_cells():
+    expected = pd.DataFrame(
+        [1, np.nan, 3, np.nan, 5],
+        columns=["Column 1"],
+    )
+
+    result = pd.read_excel("test_unempty_cells.ods")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_cell_annotation():
+    expected = pd.DataFrame(
+        ["test", np.nan, "test 3"],
+        columns=["Column 1"],
+    )
+
+    result = pd.read_excel("test_cell_annotation.ods")
+
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/excel/test_odswriter.py b/pandas/pandas/tests/io/excel/test_odswriter.py
new file mode 100644
index 0000000000000..7843bb59f97cf
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_odswriter.py
@@ -0,0 +1,106 @@
+from datetime import (
+    date,
+    datetime,
+)
+import re
+import uuid
+
+import pytest
+
+import pandas as pd
+
+from pandas.io.excel import ExcelWriter
+
+odf = pytest.importorskip("odf")
+
+
+@pytest.fixture
+def ext():
+    return ".ods"
+
+
+@pytest.fixture
+def tmp_excel(ext, tmp_path):
+    tmp = tmp_path / f"{uuid.uuid4()}{ext}"
+    tmp.touch()
+    return str(tmp)
+
+
+def test_write_append_mode_raises(tmp_excel):
+    msg = "Append mode is not supported with odf!"
+
+    with pytest.raises(ValueError, match=msg):
+        ExcelWriter(tmp_excel, engine="odf", mode="a")
+
+
+@pytest.mark.parametrize("engine_kwargs", [None, {"kwarg": 1}])
+def test_engine_kwargs(tmp_excel, engine_kwargs):
+    # GH 42286
+    # GH 43445
+    # test for error: OpenDocumentSpreadsheet does not accept any arguments
+    if engine_kwargs is not None:
+        error = re.escape(
+            "OpenDocumentSpreadsheet() got an unexpected keyword argument 'kwarg'"
+        )
+        with pytest.raises(
+            TypeError,
+            match=error,
+        ):
+            ExcelWriter(tmp_excel, engine="odf", engine_kwargs=engine_kwargs)
+    else:
+        with ExcelWriter(tmp_excel, engine="odf", engine_kwargs=engine_kwargs) as _:
+            pass
+
+
+def test_book_and_sheets_consistent(tmp_excel):
+    # GH#45687 - Ensure sheets is updated if user modifies book
+    with ExcelWriter(tmp_excel) as writer:
+        assert writer.sheets == {}
+        table = odf.table.Table(name="test_name")
+        writer.book.spreadsheet.addElement(table)
+        assert writer.sheets == {"test_name": table}
+
+
+@pytest.mark.parametrize(
+    ["value", "cell_value_type", "cell_value_attribute", "cell_value"],
+    argvalues=[
+        (True, "boolean", "boolean-value", "true"),
+        ("test string", "string", "string-value", "test string"),
+        (1, "float", "value", "1"),
+        (1.5, "float", "value", "1.5"),
+        (
+            datetime(2010, 10, 10, 10, 10, 10),
+            "date",
+            "date-value",
+            "2010-10-10T10:10:10",
+        ),
+        (date(2010, 10, 10), "date", "date-value", "2010-10-10"),
+    ],
+)
+def test_cell_value_type(
+    tmp_excel, value, cell_value_type, cell_value_attribute, cell_value
+):
+    # GH#54994 ODS: cell attributes should follow specification
+    # http://docs.oasis-open.org/office/v1.2/os/OpenDocument-v1.2-os-part1.html#refTable13
+    from odf.namespaces import OFFICENS
+    from odf.table import (
+        TableCell,
+        TableRow,
+    )
+
+    table_cell_name = TableCell().qname
+
+    pd.DataFrame([[value]]).to_excel(tmp_excel, header=False, index=False)
+
+    with pd.ExcelFile(tmp_excel) as wb:
+        sheet = wb._reader.get_sheet_by_index(0)
+        sheet_rows = sheet.getElementsByType(TableRow)
+        sheet_cells = [
+            x
+            for x in sheet_rows[0].childNodes
+            if hasattr(x, "qname") and x.qname == table_cell_name
+        ]
+
+        cell = sheet_cells[0]
+        assert cell.attributes.get((OFFICENS, "value-type")) == cell_value_type
+        assert cell.attributes.get((OFFICENS, cell_value_attribute)) == cell_value
diff --git a/pandas/pandas/tests/io/excel/test_openpyxl.py b/pandas/pandas/tests/io/excel/test_openpyxl.py
new file mode 100644
index 0000000000000..5b4bbb9e686d3
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_openpyxl.py
@@ -0,0 +1,431 @@
+import contextlib
+from pathlib import Path
+import re
+import uuid
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+
+from pandas.io.excel import (
+    ExcelWriter,
+    _OpenpyxlWriter,
+)
+from pandas.io.excel._openpyxl import OpenpyxlReader
+
+openpyxl = pytest.importorskip("openpyxl")
+
+
+@pytest.fixture
+def ext():
+    return ".xlsx"
+
+
+@pytest.fixture
+def tmp_excel(ext, tmp_path):
+    tmp = tmp_path / f"{uuid.uuid4()}{ext}"
+    tmp.touch()
+    return str(tmp)
+
+
+def test_to_excel_styleconverter():
+    from openpyxl import styles
+
+    hstyle = {
+        "font": {"color": "00FF0000", "bold": True},
+        "borders": {"top": "thin", "right": "thin", "bottom": "thin", "left": "thin"},
+        "alignment": {"horizontal": "center", "vertical": "top"},
+        "fill": {"patternType": "solid", "fgColor": {"rgb": "006666FF", "tint": 0.3}},
+        "number_format": {"format_code": "0.00"},
+        "protection": {"locked": True, "hidden": False},
+    }
+
+    font_color = styles.Color("00FF0000")
+    font = styles.Font(bold=True, color=font_color)
+    side = styles.Side(style=styles.borders.BORDER_THIN)
+    border = styles.Border(top=side, right=side, bottom=side, left=side)
+    alignment = styles.Alignment(horizontal="center", vertical="top")
+    fill_color = styles.Color(rgb="006666FF", tint=0.3)
+    fill = styles.PatternFill(patternType="solid", fgColor=fill_color)
+
+    number_format = "0.00"
+
+    protection = styles.Protection(locked=True, hidden=False)
+
+    kw = _OpenpyxlWriter._convert_to_style_kwargs(hstyle)
+    assert kw["font"] == font
+    assert kw["border"] == border
+    assert kw["alignment"] == alignment
+    assert kw["fill"] == fill
+    assert kw["number_format"] == number_format
+    assert kw["protection"] == protection
+
+
+def test_write_cells_merge_styled(tmp_excel):
+    from pandas.io.formats.excel import ExcelCell
+
+    sheet_name = "merge_styled"
+
+    sty_b1 = {"font": {"color": "00FF0000"}}
+    sty_a2 = {"font": {"color": "0000FF00"}}
+
+    initial_cells = [
+        ExcelCell(col=1, row=0, val=42, style=sty_b1),
+        ExcelCell(col=0, row=1, val=99, style=sty_a2),
+    ]
+
+    sty_merged = {"font": {"color": "000000FF", "bold": True}}
+    sty_kwargs = _OpenpyxlWriter._convert_to_style_kwargs(sty_merged)
+    openpyxl_sty_merged = sty_kwargs["font"]
+    merge_cells = [
+        ExcelCell(
+            col=0, row=0, val="pandas", mergestart=1, mergeend=1, style=sty_merged
+        )
+    ]
+
+    with _OpenpyxlWriter(tmp_excel) as writer:
+        writer._write_cells(initial_cells, sheet_name=sheet_name)
+        writer._write_cells(merge_cells, sheet_name=sheet_name)
+
+        wks = writer.sheets[sheet_name]
+    xcell_b1 = wks["B1"]
+    xcell_a2 = wks["A2"]
+    assert xcell_b1.font == openpyxl_sty_merged
+    assert xcell_a2.font == openpyxl_sty_merged
+
+
+@pytest.mark.parametrize("iso_dates", [True, False])
+def test_engine_kwargs_write(tmp_excel, iso_dates):
+    # GH 42286 GH 43445
+    engine_kwargs = {"iso_dates": iso_dates}
+    with ExcelWriter(
+        tmp_excel, engine="openpyxl", engine_kwargs=engine_kwargs
+    ) as writer:
+        assert writer.book.iso_dates == iso_dates
+        # ExcelWriter won't allow us to close without writing something
+        DataFrame().to_excel(writer)
+
+
+def test_engine_kwargs_append_invalid(tmp_excel):
+    # GH 43445
+    # test whether an invalid engine kwargs actually raises
+    DataFrame(["hello", "world"]).to_excel(tmp_excel)
+    with pytest.raises(
+        TypeError,
+        match=re.escape(
+            "load_workbook() got an unexpected keyword argument 'apple_banana'"
+        ),
+    ):
+        with ExcelWriter(
+            tmp_excel,
+            engine="openpyxl",
+            mode="a",
+            engine_kwargs={"apple_banana": "fruit"},
+        ) as writer:
+            # ExcelWriter needs us to write something to close properly
+            DataFrame(["good"]).to_excel(writer, sheet_name="Sheet2")
+
+
+@pytest.mark.parametrize("data_only, expected", [(True, 0), (False, "=1+1")])
+def test_engine_kwargs_append_data_only(tmp_excel, data_only, expected):
+    # GH 43445
+    # tests whether the data_only engine_kwarg actually works well for
+    # openpyxl's load_workbook
+    DataFrame(["=1+1"]).to_excel(tmp_excel)
+    with ExcelWriter(
+        tmp_excel, engine="openpyxl", mode="a", engine_kwargs={"data_only": data_only}
+    ) as writer:
+        assert writer.sheets["Sheet1"]["B2"].value == expected
+        # ExcelWriter needs us to writer something to close properly?
+        DataFrame().to_excel(writer, sheet_name="Sheet2")
+
+    # ensure that data_only also works for reading
+    #  and that formulas/values roundtrip
+    assert (
+        pd.read_excel(
+            tmp_excel,
+            sheet_name="Sheet1",
+            engine="openpyxl",
+            engine_kwargs={"data_only": data_only},
+        ).iloc[0, 1]
+        == expected
+    )
+
+
+@pytest.mark.parametrize("kwarg_name", ["read_only", "data_only"])
+@pytest.mark.parametrize("kwarg_value", [True, False])
+def test_engine_kwargs_append_reader(datapath, ext, kwarg_name, kwarg_value):
+    # GH 55027
+    # test that `read_only` and `data_only` can be passed to
+    #  `openpyxl.reader.excel.load_workbook` via `engine_kwargs`
+    filename = datapath("io", "data", "excel", "test1" + ext)
+    with contextlib.closing(
+        OpenpyxlReader(filename, engine_kwargs={kwarg_name: kwarg_value})
+    ) as reader:
+        assert getattr(reader.book, kwarg_name) == kwarg_value
+
+
+@pytest.mark.parametrize(
+    "mode,expected", [("w", ["baz"]), ("a", ["foo", "bar", "baz"])]
+)
+def test_write_append_mode(tmp_excel, mode, expected):
+    df = DataFrame([1], columns=["baz"])
+
+    wb = openpyxl.Workbook()
+    wb.worksheets[0].title = "foo"
+    wb.worksheets[0]["A1"].value = "foo"
+    wb.create_sheet("bar")
+    wb.worksheets[1]["A1"].value = "bar"
+    wb.save(tmp_excel)
+
+    with ExcelWriter(tmp_excel, engine="openpyxl", mode=mode) as writer:
+        df.to_excel(writer, sheet_name="baz", index=False)
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb2:
+        result = [sheet.title for sheet in wb2.worksheets]
+        assert result == expected
+
+        for index, cell_value in enumerate(expected):
+            assert wb2.worksheets[index]["A1"].value == cell_value
+
+
+@pytest.mark.parametrize(
+    "if_sheet_exists,num_sheets,expected",
+    [
+        ("new", 2, ["apple", "banana"]),
+        ("replace", 1, ["pear"]),
+        ("overlay", 1, ["pear", "banana"]),
+    ],
+)
+def test_if_sheet_exists_append_modes(tmp_excel, if_sheet_exists, num_sheets, expected):
+    # GH 40230
+    df1 = DataFrame({"fruit": ["apple", "banana"]})
+    df2 = DataFrame({"fruit": ["pear"]})
+
+    df1.to_excel(tmp_excel, engine="openpyxl", sheet_name="foo", index=False)
+    with ExcelWriter(
+        tmp_excel, engine="openpyxl", mode="a", if_sheet_exists=if_sheet_exists
+    ) as writer:
+        df2.to_excel(writer, sheet_name="foo", index=False)
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        assert len(wb.sheetnames) == num_sheets
+        assert wb.sheetnames[0] == "foo"
+        result = pd.read_excel(wb, "foo", engine="openpyxl")
+        assert list(result["fruit"]) == expected
+        if len(wb.sheetnames) == 2:
+            result = pd.read_excel(wb, wb.sheetnames[1], engine="openpyxl")
+            tm.assert_frame_equal(result, df2)
+
+
+@pytest.mark.parametrize(
+    "startrow, startcol, greeting, goodbye",
+    [
+        (0, 0, ["poop", "world"], ["goodbye", "people"]),
+        (0, 1, ["hello", "world"], ["poop", "people"]),
+        (1, 0, ["hello", "poop"], ["goodbye", "people"]),
+        (1, 1, ["hello", "world"], ["goodbye", "poop"]),
+    ],
+)
+def test_append_overlay_startrow_startcol(
+    tmp_excel, startrow, startcol, greeting, goodbye
+):
+    df1 = DataFrame({"greeting": ["hello", "world"], "goodbye": ["goodbye", "people"]})
+    df2 = DataFrame(["poop"])
+
+    df1.to_excel(tmp_excel, engine="openpyxl", sheet_name="poo", index=False)
+    with ExcelWriter(
+        tmp_excel, engine="openpyxl", mode="a", if_sheet_exists="overlay"
+    ) as writer:
+        # use startrow+1 because we don't have a header
+        df2.to_excel(
+            writer,
+            index=False,
+            header=False,
+            startrow=startrow + 1,
+            startcol=startcol,
+            sheet_name="poo",
+        )
+
+    result = pd.read_excel(tmp_excel, sheet_name="poo", engine="openpyxl")
+    expected = DataFrame({"greeting": greeting, "goodbye": goodbye})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "if_sheet_exists,msg",
+    [
+        (
+            "invalid",
+            "'invalid' is not valid for if_sheet_exists. Valid options "
+            "are 'error', 'new', 'replace' and 'overlay'.",
+        ),
+        (
+            "error",
+            "Sheet 'foo' already exists and if_sheet_exists is set to 'error'.",
+        ),
+        (
+            None,
+            "Sheet 'foo' already exists and if_sheet_exists is set to 'error'.",
+        ),
+    ],
+)
+def test_if_sheet_exists_raises(tmp_excel, if_sheet_exists, msg):
+    # GH 40230
+    df = DataFrame({"fruit": ["pear"]})
+    df.to_excel(tmp_excel, sheet_name="foo", engine="openpyxl")
+    with pytest.raises(ValueError, match=re.escape(msg)):
+        with ExcelWriter(
+            tmp_excel, engine="openpyxl", mode="a", if_sheet_exists=if_sheet_exists
+        ) as writer:
+            df.to_excel(writer, sheet_name="foo")
+
+
+def test_to_excel_with_openpyxl_engine(tmp_excel):
+    # GH 29854
+    df1 = DataFrame({"A": np.linspace(1, 10, 10)})
+    df2 = DataFrame({"B": np.linspace(1, 20, 10)})
+    df = pd.concat([df1, df2], axis=1)
+    styled = df.style.map(
+        lambda val: f"color: {'red' if val < 0 else 'black'}"
+    ).highlight_max()
+
+    styled.to_excel(tmp_excel, engine="openpyxl")
+
+
+@pytest.mark.parametrize("read_only", [True, False])
+def test_read_workbook(datapath, ext, read_only):
+    # GH 39528
+    filename = datapath("io", "data", "excel", "test1" + ext)
+    with contextlib.closing(
+        openpyxl.load_workbook(filename, read_only=read_only)
+    ) as wb:
+        result = pd.read_excel(wb, engine="openpyxl")
+    expected = pd.read_excel(filename)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "header, expected_data",
+    [
+        (
+            0,
+            {
+                "Title": [np.nan, "A", 1, 2, 3],
+                "Unnamed: 1": [np.nan, "B", 4, 5, 6],
+                "Unnamed: 2": [np.nan, "C", 7, 8, 9],
+            },
+        ),
+        (2, {"A": [1, 2, 3], "B": [4, 5, 6], "C": [7, 8, 9]}),
+    ],
+)
+@pytest.mark.parametrize(
+    "filename", ["dimension_missing", "dimension_small", "dimension_large"]
+)
+# When read_only is None, use read_excel instead of a workbook
+@pytest.mark.parametrize("read_only", [True, False, None])
+def test_read_with_bad_dimension(
+    datapath, ext, header, expected_data, filename, read_only
+):
+    # GH 38956, 39001 - no/incorrect dimension information
+    path = datapath("io", "data", "excel", f"{filename}{ext}")
+    if read_only is None:
+        result = pd.read_excel(path, header=header)
+    else:
+        with contextlib.closing(
+            openpyxl.load_workbook(path, read_only=read_only)
+        ) as wb:
+            result = pd.read_excel(wb, engine="openpyxl", header=header)
+    expected = DataFrame(expected_data)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_append_mode_file(tmp_excel):
+    # GH 39576
+    df = DataFrame()
+
+    df.to_excel(tmp_excel, engine="openpyxl")
+
+    with ExcelWriter(
+        tmp_excel, mode="a", engine="openpyxl", if_sheet_exists="new"
+    ) as writer:
+        df.to_excel(writer)
+
+    # make sure that zip files are not concatenated by making sure that
+    # "docProps/app.xml" only occurs twice in the file
+    data = Path(tmp_excel).read_bytes()
+    first = data.find(b"docProps/app.xml")
+    second = data.find(b"docProps/app.xml", first + 1)
+    third = data.find(b"docProps/app.xml", second + 1)
+    assert second != -1 and third == -1
+
+
+# When read_only is None, use read_excel instead of a workbook
+@pytest.mark.parametrize("read_only", [True, False, None])
+def test_read_with_empty_trailing_rows(datapath, ext, read_only):
+    # GH 39181
+    path = datapath("io", "data", "excel", f"empty_trailing_rows{ext}")
+    if read_only is None:
+        result = pd.read_excel(path)
+    else:
+        with contextlib.closing(
+            openpyxl.load_workbook(path, read_only=read_only)
+        ) as wb:
+            result = pd.read_excel(wb, engine="openpyxl")
+    expected = DataFrame(
+        {
+            "Title": [np.nan, "A", 1, 2, 3],
+            "Unnamed: 1": [np.nan, "B", 4, 5, 6],
+            "Unnamed: 2": [np.nan, "C", 7, 8, 9],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# When read_only is None, use read_excel instead of a workbook
+@pytest.mark.parametrize("read_only", [True, False, None])
+def test_read_empty_with_blank_row(datapath, ext, read_only):
+    # GH 39547 - empty excel file with a row that has no data
+    path = datapath("io", "data", "excel", f"empty_with_blank_row{ext}")
+    if read_only is None:
+        result = pd.read_excel(path)
+    else:
+        with contextlib.closing(
+            openpyxl.load_workbook(path, read_only=read_only)
+        ) as wb:
+            result = pd.read_excel(wb, engine="openpyxl")
+    expected = DataFrame()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_book_and_sheets_consistent(tmp_excel):
+    # GH#45687 - Ensure sheets is updated if user modifies book
+    with ExcelWriter(tmp_excel, engine="openpyxl") as writer:
+        assert writer.sheets == {}
+        sheet = writer.book.create_sheet("test_name", 0)
+        assert writer.sheets == {"test_name": sheet}
+
+
+def test_ints_spelled_with_decimals(datapath, ext):
+    # GH 46988 - openpyxl returns this sheet with floats
+    path = datapath("io", "data", "excel", f"ints_spelled_with_decimals{ext}")
+    result = pd.read_excel(path)
+    expected = DataFrame(range(2, 12), columns=[1])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_multiindex_header_no_index_names(datapath, ext):
+    # GH#47487
+    path = datapath("io", "data", "excel", f"multiindex_no_index_names{ext}")
+    result = pd.read_excel(path, index_col=[0, 1, 2], header=[0, 1, 2])
+    expected = DataFrame(
+        [[np.nan, "x", "x", "x"], ["x", np.nan, np.nan, np.nan]],
+        columns=pd.MultiIndex.from_tuples(
+            [("X", "Y", "A1"), ("X", "Y", "A2"), ("XX", "YY", "B1"), ("XX", "YY", "B2")]
+        ),
+        index=pd.MultiIndex.from_tuples([("A", "AA", "AAA"), ("A", "BB", "BBB")]),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/excel/test_readers.py b/pandas/pandas/tests/io/excel/test_readers.py
new file mode 100644
index 0000000000000..140cf39b26556
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_readers.py
@@ -0,0 +1,1744 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    time,
+)
+from functools import partial
+from io import BytesIO
+import os
+from pathlib import Path
+import platform
+import re
+from urllib.error import URLError
+import uuid
+from zipfile import BadZipFile
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    read_csv,
+)
+import pandas._testing as tm
+
+read_ext_params = [".xls", ".xlsx", ".xlsm", ".xlsb", ".ods"]
+engine_params = [
+    # Add any engines to test here
+    # When defusedxml is installed it triggers deprecation warnings for
+    # xlrd and openpyxl, so catch those here
+    pytest.param(
+        "xlrd",
+        marks=[
+            td.skip_if_no("xlrd"),
+        ],
+    ),
+    pytest.param(
+        "openpyxl",
+        marks=[
+            td.skip_if_no("openpyxl"),
+        ],
+    ),
+    pytest.param(
+        None,
+        marks=[
+            td.skip_if_no("xlrd"),
+        ],
+    ),
+    pytest.param("pyxlsb", marks=td.skip_if_no("pyxlsb")),
+    pytest.param("odf", marks=td.skip_if_no("odf")),
+    pytest.param("calamine", marks=td.skip_if_no("python_calamine")),
+]
+
+
+def _is_valid_engine_ext_pair(engine, read_ext: str) -> bool:
+    """
+    Filter out invalid (engine, ext) pairs instead of skipping, as that
+    produces 500+ pytest.skips.
+    """
+    engine = engine.values[0]
+    if engine == "openpyxl" and read_ext == ".xls":
+        return False
+    if engine == "odf" and read_ext != ".ods":
+        return False
+    if read_ext == ".ods" and engine not in {"odf", "calamine"}:
+        return False
+    if engine == "pyxlsb" and read_ext != ".xlsb":
+        return False
+    if read_ext == ".xlsb" and engine not in {"pyxlsb", "calamine"}:
+        return False
+    if engine == "xlrd" and read_ext != ".xls":
+        return False
+    return True
+
+
+def _transfer_marks(engine, read_ext):
+    """
+    engine gives us a pytest.param object with some marks, read_ext is just
+    a string.  We need to generate a new pytest.param inheriting the marks.
+    """
+    values = engine.values + (read_ext,)
+    new_param = pytest.param(values, marks=engine.marks)
+    return new_param
+
+
+@pytest.fixture(
+    params=[
+        _transfer_marks(eng, ext)
+        for eng in engine_params
+        for ext in read_ext_params
+        if _is_valid_engine_ext_pair(eng, ext)
+    ],
+    ids=str,
+)
+def engine_and_read_ext(request):
+    """
+    Fixture for Excel reader engine and read_ext, only including valid pairs.
+    """
+    return request.param
+
+
+@pytest.fixture
+def engine(engine_and_read_ext):
+    engine, read_ext = engine_and_read_ext
+    return engine
+
+
+@pytest.fixture
+def read_ext(engine_and_read_ext):
+    engine, read_ext = engine_and_read_ext
+    return read_ext
+
+
+@pytest.fixture
+def tmp_excel(read_ext, tmp_path):
+    tmp = tmp_path / f"{uuid.uuid4()}{read_ext}"
+    tmp.touch()
+    return str(tmp)
+
+
+@pytest.fixture
+def df_ref(datapath):
+    """
+    Obtain the reference data from read_csv with the Python engine.
+    """
+    filepath = datapath("io", "data", "csv", "test1.csv")
+    df_ref = read_csv(filepath, index_col=0, parse_dates=True, engine="python")
+    return df_ref
+
+
+def get_exp_unit(read_ext: str, engine: str | None) -> str:
+    unit = "us"
+    if (read_ext == ".ods") ^ (engine == "calamine"):
+        unit = "s"
+    return unit
+
+
+def adjust_expected(expected: DataFrame, read_ext: str, engine: str | None) -> None:
+    expected.index.name = None
+    unit = get_exp_unit(read_ext, engine)
+    # error: "Index" has no attribute "as_unit"
+    expected.index = expected.index.as_unit(unit)  # type: ignore[attr-defined]
+
+
+def xfail_datetimes_with_pyxlsb(engine, request):
+    if engine == "pyxlsb":
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Sheets containing datetimes not supported by pyxlsb"
+            )
+        )
+
+
+class TestReaders:
+    @pytest.mark.parametrize("col", [[True, None, False], [True], [True, False]])
+    def test_read_excel_type_check(self, col, tmp_excel, read_ext):
+        # GH 58159
+        if read_ext in (".xlsb", ".xls"):
+            pytest.skip(f"No engine for filetype: '{read_ext}'")
+        df = DataFrame({"bool_column": col}, dtype="boolean")
+        df.to_excel(tmp_excel, index=False)
+        df2 = pd.read_excel(tmp_excel, dtype={"bool_column": "boolean"})
+        tm.assert_frame_equal(df, df2)
+
+    def test_pass_none_type(self, datapath):
+        # GH 58159
+        f_path = datapath("io", "data", "excel", "test_none_type.xlsx")
+
+        with pd.ExcelFile(f_path) as excel:
+            parsed = pd.read_excel(
+                excel,
+                sheet_name="Sheet1",
+                keep_default_na=True,
+                na_values=["nan", "None", "abcd"],
+                dtype="boolean",
+                engine="openpyxl",
+            )
+        expected = DataFrame(
+            {"Test": [True, None, False, None, False, None, True]},
+            dtype="boolean",
+        )
+
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.fixture(autouse=True)
+    def cd_and_set_engine(self, engine, datapath, monkeypatch):
+        """
+        Change directory and set engine for read_excel calls.
+        """
+        func = partial(pd.read_excel, engine=engine)
+        monkeypatch.chdir(datapath("io", "data", "excel"))
+        monkeypatch.setattr(pd, "read_excel", func)
+
+    def test_engine_used(self, read_ext, engine, monkeypatch):
+        # GH 38884
+        def parser(self, *args, **kwargs):
+            return self.engine
+
+        monkeypatch.setattr(pd.ExcelFile, "parse", parser)
+
+        expected_defaults = {
+            "xlsx": "openpyxl",
+            "xlsm": "openpyxl",
+            "xlsb": "pyxlsb",
+            "xls": "xlrd",
+            "ods": "odf",
+        }
+
+        with open("test1" + read_ext, "rb") as f:
+            result = pd.read_excel(f)
+
+        if engine is not None:
+            expected = engine
+        else:
+            expected = expected_defaults[read_ext[1:]]
+        assert result == expected
+
+    def test_engine_kwargs(self, read_ext, engine):
+        # GH#52214
+        expected_defaults = {
+            "xlsx": {"foo": "abcd"},
+            "xlsm": {"foo": 123},
+            "xlsb": {"foo": "True"},
+            "xls": {"foo": True},
+            "ods": {"foo": "abcd"},
+        }
+
+        if engine in {"xlrd", "pyxlsb"}:
+            msg = re.escape(r"open_workbook() got an unexpected keyword argument 'foo'")
+        elif engine == "odf":
+            msg = re.escape(r"load() got an unexpected keyword argument 'foo'")
+        else:
+            msg = re.escape(r"load_workbook() got an unexpected keyword argument 'foo'")
+
+        if engine is not None:
+            with pytest.raises(TypeError, match=msg):
+                pd.read_excel(
+                    "test1" + read_ext,
+                    sheet_name="Sheet1",
+                    index_col=0,
+                    engine_kwargs=expected_defaults[read_ext[1:]],
+                )
+
+    def test_usecols_int(self, read_ext):
+        # usecols as int
+        msg = "Passing an integer for `usecols`"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel(
+                "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols=3
+            )
+
+        # usecols as int
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel(
+                "test1" + read_ext,
+                sheet_name="Sheet2",
+                skiprows=[1],
+                index_col=0,
+                usecols=3,
+            )
+
+    def test_usecols_list(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref[["B", "C"]]
+        adjust_expected(expected, read_ext, engine)
+
+        df1 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols=[0, 2, 3]
+        )
+        df2 = pd.read_excel(
+            "test1" + read_ext,
+            sheet_name="Sheet2",
+            skiprows=[1],
+            index_col=0,
+            usecols=[0, 2, 3],
+        )
+
+        # TODO add index to xls file)
+        tm.assert_frame_equal(df1, expected)
+        tm.assert_frame_equal(df2, expected)
+
+    def test_usecols_str(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref[["A", "B", "C"]]
+        adjust_expected(expected, read_ext, engine)
+
+        df2 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols="A:D"
+        )
+        df3 = pd.read_excel(
+            "test1" + read_ext,
+            sheet_name="Sheet2",
+            skiprows=[1],
+            index_col=0,
+            usecols="A:D",
+        )
+
+        # TODO add index to xls, read xls ignores index name ?
+        tm.assert_frame_equal(df2, expected)
+        tm.assert_frame_equal(df3, expected)
+
+        expected = df_ref[["B", "C"]]
+        adjust_expected(expected, read_ext, engine)
+
+        df2 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols="A,C,D"
+        )
+        df3 = pd.read_excel(
+            "test1" + read_ext,
+            sheet_name="Sheet2",
+            skiprows=[1],
+            index_col=0,
+            usecols="A,C,D",
+        )
+        # TODO add index to xls file
+        tm.assert_frame_equal(df2, expected)
+        tm.assert_frame_equal(df3, expected)
+
+        df2 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols="A,C:D"
+        )
+        df3 = pd.read_excel(
+            "test1" + read_ext,
+            sheet_name="Sheet2",
+            skiprows=[1],
+            index_col=0,
+            usecols="A,C:D",
+        )
+        tm.assert_frame_equal(df2, expected)
+        tm.assert_frame_equal(df3, expected)
+
+    @pytest.mark.parametrize(
+        "usecols", [[0, 1, 3], [0, 3, 1], [1, 0, 3], [1, 3, 0], [3, 0, 1], [3, 1, 0]]
+    )
+    def test_usecols_diff_positional_int_columns_order(
+        self, request, engine, read_ext, usecols, df_ref
+    ):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref[["A", "C"]]
+        adjust_expected(expected, read_ext, engine)
+
+        result = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols=usecols
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("usecols", [["B", "D"], ["D", "B"]])
+    def test_usecols_diff_positional_str_columns_order(self, read_ext, usecols, df_ref):
+        expected = df_ref[["B", "D"]]
+        expected.index = range(len(expected))
+
+        result = pd.read_excel("test1" + read_ext, sheet_name="Sheet1", usecols=usecols)
+        tm.assert_frame_equal(result, expected)
+
+    def test_read_excel_without_slicing(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref
+        adjust_expected(expected, read_ext, engine)
+
+        result = pd.read_excel("test1" + read_ext, sheet_name="Sheet1", index_col=0)
+        tm.assert_frame_equal(result, expected)
+
+    def test_usecols_excel_range_str(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref[["C", "D"]]
+        adjust_expected(expected, read_ext, engine)
+
+        result = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, usecols="A,D:E"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_usecols_excel_range_str_invalid(self, read_ext):
+        msg = "Invalid column name: E1"
+
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel("test1" + read_ext, sheet_name="Sheet1", usecols="D:E1")
+
+    def test_index_col_label_error(self, read_ext):
+        msg = "list indices must be integers.*, not str"
+
+        with pytest.raises(TypeError, match=msg):
+            pd.read_excel(
+                "test1" + read_ext,
+                sheet_name="Sheet1",
+                index_col=["A"],
+                usecols=["A", "C"],
+            )
+
+    def test_index_col_str(self, read_ext):
+        # see gh-52716
+        result = pd.read_excel("test1" + read_ext, sheet_name="Sheet3", index_col="A")
+        expected = DataFrame(
+            columns=["B", "C", "D", "E", "F"], index=Index([], name="A")
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_index_col_empty(self, read_ext):
+        # see gh-9208
+        result = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet3", index_col=["A", "B", "C"]
+        )
+        expected = DataFrame(
+            columns=["D", "E", "F"],
+            index=MultiIndex(levels=[[]] * 3, codes=[[]] * 3, names=["A", "B", "C"]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("index_col", [None, 2])
+    def test_index_col_with_unnamed(self, read_ext, index_col):
+        # see gh-18792
+        result = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet4", index_col=index_col
+        )
+        expected = DataFrame(
+            [["i1", "a", "x"], ["i2", "b", "y"]], columns=["Unnamed: 0", "col1", "col2"]
+        )
+        if index_col:
+            expected = expected.set_index(expected.columns[index_col])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_usecols_pass_non_existent_column(self, read_ext):
+        msg = (
+            "Usecols do not match columns, "
+            "columns expected but not found: "
+            r"\['E'\]"
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel("test1" + read_ext, usecols=["E"])
+
+    def test_usecols_wrong_type(self, read_ext):
+        msg = (
+            "'usecols' must either be list-like of "
+            "all strings, all unicode, all integers or a callable."
+        )
+
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel("test1" + read_ext, usecols=["E1", 0])
+
+    def test_excel_stop_iterator(self, read_ext):
+        parsed = pd.read_excel("test2" + read_ext, sheet_name="Sheet1")
+        expected = DataFrame([["aaaa", "bbbbb"]], columns=["Test", "Test1"])
+        tm.assert_frame_equal(parsed, expected)
+
+    def test_excel_cell_error_na(self, request, engine, read_ext):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        # https://github.com/tafia/calamine/issues/355
+        if engine == "calamine" and read_ext == ".ods":
+            request.applymarker(
+                pytest.mark.xfail(reason="Calamine can't extract error from ods files")
+            )
+
+        parsed = pd.read_excel("test3" + read_ext, sheet_name="Sheet1")
+        expected = DataFrame([[np.nan]], columns=["Test"])
+        tm.assert_frame_equal(parsed, expected)
+
+    def test_excel_table(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref
+        adjust_expected(expected, read_ext, engine)
+
+        df1 = pd.read_excel("test1" + read_ext, sheet_name="Sheet1", index_col=0)
+        df2 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet2", skiprows=[1], index_col=0
+        )
+        # TODO add index to file
+        tm.assert_frame_equal(df1, expected)
+        tm.assert_frame_equal(df2, expected)
+
+        df3 = pd.read_excel(
+            "test1" + read_ext, sheet_name="Sheet1", index_col=0, skipfooter=1
+        )
+        tm.assert_frame_equal(df3, df1.iloc[:-1])
+
+    def test_reader_special_dtypes(self, request, engine, read_ext):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        unit = get_exp_unit(read_ext, engine)
+        expected = DataFrame.from_dict(
+            {
+                "IntCol": [1, 2, -3, 4, 0],
+                "FloatCol": [1.25, 2.25, 1.83, 1.92, 0.0000000005],
+                "BoolCol": [True, False, True, True, False],
+                "StrCol": [1, 2, 3, 4, 5],
+                "Str2Col": ["a", 3, "c", "d", "e"],
+                "DateCol": Index(
+                    [
+                        datetime(2013, 10, 30),
+                        datetime(2013, 10, 31),
+                        datetime(1905, 1, 1),
+                        datetime(2013, 12, 14),
+                        datetime(2015, 3, 14),
+                    ],
+                    dtype=f"M8[{unit}]",
+                ),
+            },
+        )
+        basename = "test_types"
+
+        # should read in correctly and infer types
+        actual = pd.read_excel(basename + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, expected)
+
+        # if not coercing number, then int comes in as float
+        float_expected = expected.copy()
+        float_expected.loc[float_expected.index[1], "Str2Col"] = 3.0
+        actual = pd.read_excel(basename + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, float_expected)
+
+        # check setting Index (assuming xls and xlsx are the same here)
+        for icol, name in enumerate(expected.columns):
+            actual = pd.read_excel(
+                basename + read_ext, sheet_name="Sheet1", index_col=icol
+            )
+            exp = expected.set_index(name)
+            tm.assert_frame_equal(actual, exp)
+
+        expected["StrCol"] = expected["StrCol"].apply(str)
+        actual = pd.read_excel(
+            basename + read_ext, sheet_name="Sheet1", converters={"StrCol": str}
+        )
+        tm.assert_frame_equal(actual, expected)
+
+    # GH8212 - support for converters and missing values
+    def test_reader_converters(self, read_ext):
+        basename = "test_converters"
+
+        expected = DataFrame.from_dict(
+            {
+                "IntCol": [1, 2, -3, -1000, 0],
+                "FloatCol": [12.5, np.nan, 18.3, 19.2, 0.000000005],
+                "BoolCol": ["Found", "Found", "Found", "Not found", "Found"],
+                "StrCol": ["1", np.nan, "3", "4", "5"],
+            }
+        )
+
+        converters = {
+            "IntCol": lambda x: int(x) if x != "" else -1000,
+            "FloatCol": lambda x: 10 * x if x else np.nan,
+            2: lambda x: "Found" if x != "" else "Not found",
+            3: lambda x: str(x) if x else "",
+        }
+
+        # should read in correctly and set types of single cells (not array
+        # dtypes)
+        actual = pd.read_excel(
+            basename + read_ext, sheet_name="Sheet1", converters=converters
+        )
+        tm.assert_frame_equal(actual, expected)
+
+    def test_reader_dtype(self, read_ext):
+        # GH 8212
+        basename = "testdtype"
+        actual = pd.read_excel(basename + read_ext)
+
+        expected = DataFrame(
+            {
+                "a": [1, 2, 3, 4],
+                "b": [2.5, 3.5, 4.5, 5.5],
+                "c": [1, 2, 3, 4],
+                "d": [1.0, 2.0, np.nan, 4.0],
+            }
+        )
+
+        tm.assert_frame_equal(actual, expected)
+
+        actual = pd.read_excel(
+            basename + read_ext, dtype={"a": "float64", "b": "float32", "c": str}
+        )
+
+        expected["a"] = expected["a"].astype("float64")
+        expected["b"] = expected["b"].astype("float32")
+        expected["c"] = Series(["001", "002", "003", "004"], dtype="str")
+        tm.assert_frame_equal(actual, expected)
+
+        msg = "Unable to convert column d to type int64"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel(basename + read_ext, dtype={"d": "int64"})
+
+    @pytest.mark.parametrize(
+        "dtype,expected",
+        [
+            (
+                None,
+                {
+                    "a": [1, 2, 3, 4],
+                    "b": [2.5, 3.5, 4.5, 5.5],
+                    "c": [1, 2, 3, 4],
+                    "d": [1.0, 2.0, np.nan, 4.0],
+                },
+            ),
+            (
+                {"a": "float64", "b": "float32", "c": str, "d": str},
+                {
+                    "a": Series([1, 2, 3, 4], dtype="float64"),
+                    "b": Series([2.5, 3.5, 4.5, 5.5], dtype="float32"),
+                    "c": Series(["001", "002", "003", "004"], dtype="str"),
+                    "d": Series(["1", "2", np.nan, "4"], dtype="str"),
+                },
+            ),
+        ],
+    )
+    def test_reader_dtype_str(self, read_ext, dtype, expected):
+        # see gh-20377
+        basename = "testdtype"
+
+        actual = pd.read_excel(basename + read_ext, dtype=dtype)
+        expected = DataFrame(expected)
+        tm.assert_frame_equal(actual, expected)
+
+    def test_dtype_backend(self, read_ext, dtype_backend, engine, tmp_excel):
+        # GH#36712
+        if read_ext in (".xlsb", ".xls"):
+            pytest.skip(f"No engine for filetype: '{read_ext}'")
+
+        df = DataFrame(
+            {
+                "a": Series([1, 3], dtype="Int64"),
+                "b": Series([2.5, 4.5], dtype="Float64"),
+                "c": Series([True, False], dtype="boolean"),
+                "d": Series(["a", "b"], dtype="string"),
+                "e": Series([pd.NA, 6], dtype="Int64"),
+                "f": Series([pd.NA, 7.5], dtype="Float64"),
+                "g": Series([pd.NA, True], dtype="boolean"),
+                "h": Series([pd.NA, "a"], dtype="string"),
+                "i": Series([pd.Timestamp("2019-12-31")] * 2),
+                "j": Series([pd.NA, pd.NA], dtype="Int64"),
+            }
+        )
+        df.to_excel(tmp_excel, sheet_name="test", index=False)
+        result = pd.read_excel(
+            tmp_excel, sheet_name="test", dtype_backend=dtype_backend
+        )
+        if dtype_backend == "pyarrow":
+            import pyarrow as pa
+
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(df[col], from_pandas=True))
+                    for col in df.columns
+                }
+            )
+            # pyarrow by default infers timestamp resolution as us, not ns
+            expected["i"] = ArrowExtensionArray(
+                expected["i"].array._pa_array.cast(pa.timestamp(unit="us"))
+            )
+            # pyarrow supports a null type, so don't have to default to Int64
+            expected["j"] = ArrowExtensionArray(pa.array([None, None]))
+        else:
+            expected = df
+            unit = get_exp_unit(read_ext, engine)
+            expected["i"] = expected["i"].astype(f"M8[{unit}]")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_dtype_backend_and_dtype(self, read_ext, tmp_excel):
+        # GH#36712
+        if read_ext in (".xlsb", ".xls"):
+            pytest.skip(f"No engine for filetype: '{read_ext}'")
+
+        df = DataFrame({"a": [np.nan, 1.0], "b": [2.5, np.nan]})
+        df.to_excel(tmp_excel, sheet_name="test", index=False)
+        result = pd.read_excel(
+            tmp_excel,
+            sheet_name="test",
+            dtype_backend="numpy_nullable",
+            dtype="float64",
+        )
+        tm.assert_frame_equal(result, df)
+
+    def test_dtype_backend_string(self, read_ext, string_storage, tmp_excel):
+        # GH#36712
+        if read_ext in (".xlsb", ".xls"):
+            pytest.skip(f"No engine for filetype: '{read_ext}'")
+
+        df = DataFrame(
+            {
+                "a": np.array(["a", "b"], dtype=np.object_),
+                "b": np.array(["x", pd.NA], dtype=np.object_),
+            }
+        )
+        df.to_excel(tmp_excel, sheet_name="test", index=False)
+
+        with pd.option_context("mode.string_storage", string_storage):
+            result = pd.read_excel(
+                tmp_excel, sheet_name="test", dtype_backend="numpy_nullable"
+            )
+
+        expected = DataFrame(
+            {
+                "a": Series(["a", "b"], dtype=pd.StringDtype(string_storage)),
+                "b": Series(["x", None], dtype=pd.StringDtype(string_storage)),
+            }
+        )
+        # the storage of the str columns' Index is also affected by the
+        # string_storage setting -> ignore that for checking the result
+        tm.assert_frame_equal(result, expected, check_column_type=False)
+
+    @pytest.mark.parametrize("dtypes, exp_value", [({}, 1), ({"a.1": "int64"}, 1)])
+    def test_dtype_mangle_dup_cols(self, read_ext, dtypes, exp_value):
+        # GH#35211
+        basename = "df_mangle_dup_col_dtypes"
+        dtype_dict = {"a": object, **dtypes}
+        dtype_dict_copy = dtype_dict.copy()
+        # GH#42462
+        result = pd.read_excel(basename + read_ext, dtype=dtype_dict)
+        expected = DataFrame(
+            {
+                "a": Series([1], dtype=object),
+                "a.1": Series([exp_value], dtype=object if not dtypes else None),
+            }
+        )
+        assert dtype_dict == dtype_dict_copy, "dtype dict changed"
+        tm.assert_frame_equal(result, expected)
+
+    def test_reader_spaces(self, read_ext):
+        # see gh-32207
+        basename = "test_spaces"
+
+        actual = pd.read_excel(basename + read_ext)
+        expected = DataFrame(
+            {
+                "testcol": [
+                    "this is great",
+                    "4    spaces",
+                    "1 trailing ",
+                    " 1 leading",
+                    "2  spaces  multiple  times",
+                ]
+            }
+        )
+        tm.assert_frame_equal(actual, expected)
+
+    # gh-36122, gh-35802
+    @pytest.mark.parametrize(
+        "basename,expected",
+        [
+            ("gh-35802", DataFrame({"COLUMN": ["Test (1)"]})),
+            ("gh-36122", DataFrame(columns=["got 2nd sa"])),
+        ],
+    )
+    def test_read_excel_ods_nested_xml(self, engine, read_ext, basename, expected):
+        # see gh-35802
+        if engine != "odf":
+            pytest.skip(f"Skipped for engine: {engine}")
+
+        actual = pd.read_excel(basename + read_ext)
+        tm.assert_frame_equal(actual, expected)
+
+    def test_reading_all_sheets(self, read_ext):
+        # Test reading all sheet names by setting sheet_name to None,
+        # Ensure a dict is returned.
+        # See PR #9450
+        basename = "test_multisheet"
+        dfs = pd.read_excel(basename + read_ext, sheet_name=None)
+        # ensure this is not alphabetical to test order preservation
+        expected_keys = ["Charlie", "Alpha", "Beta"]
+        tm.assert_contains_all(expected_keys, dfs.keys())
+        # Issue 9930
+        # Ensure sheet order is preserved
+        assert expected_keys == list(dfs.keys())
+
+    def test_reading_multiple_specific_sheets(self, read_ext):
+        # Test reading specific sheet names by specifying a mixed list
+        # of integers and strings, and confirm that duplicated sheet
+        # references (positions/names) are removed properly.
+        # Ensure a dict is returned
+        # See PR #9450
+        basename = "test_multisheet"
+        # Explicitly request duplicates. Only the set should be returned.
+        expected_keys = [2, "Charlie", "Charlie"]
+        dfs = pd.read_excel(basename + read_ext, sheet_name=expected_keys)
+        expected_keys = list(set(expected_keys))
+        tm.assert_contains_all(expected_keys, dfs.keys())
+        assert len(expected_keys) == len(dfs.keys())
+
+    def test_reading_all_sheets_with_blank(self, read_ext):
+        # Test reading all sheet names by setting sheet_name to None,
+        # In the case where some sheets are blank.
+        # Issue #11711
+        basename = "blank_with_header"
+        dfs = pd.read_excel(basename + read_ext, sheet_name=None)
+        expected_keys = ["Sheet1", "Sheet2", "Sheet3"]
+        tm.assert_contains_all(expected_keys, dfs.keys())
+
+    # GH6403
+    def test_read_excel_blank(self, read_ext):
+        actual = pd.read_excel("blank" + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, DataFrame())
+
+    def test_read_excel_blank_with_header(self, read_ext):
+        expected = DataFrame(columns=["col_1", "col_2"])
+        actual = pd.read_excel("blank_with_header" + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_exception_message_includes_sheet_name(self, read_ext):
+        # GH 48706
+        with pytest.raises(ValueError, match=r" \(sheet: Sheet1\)$"):
+            pd.read_excel("blank_with_header" + read_ext, header=[1], sheet_name=None)
+        with pytest.raises(ZeroDivisionError, match=r" \(sheet: Sheet1\)$"):
+            pd.read_excel("test1" + read_ext, usecols=lambda x: 1 / 0, sheet_name=None)
+
+    @pytest.mark.filterwarnings("ignore:Cell A4 is marked:UserWarning:openpyxl")
+    def test_date_conversion_overflow(self, request, engine, read_ext):
+        # GH 10001 : pandas.ExcelFile ignore parse_dates=False
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = DataFrame(
+            [
+                [pd.Timestamp("2016-03-12"), "Marc Johnson"],
+                [pd.Timestamp("2016-03-16"), "Jack Black"],
+                [1e20, "Timothy Brown"],
+            ],
+            columns=["DateColWithBigInt", "StringCol"],
+        )
+
+        if engine == "openpyxl":
+            request.applymarker(
+                pytest.mark.xfail(reason="Maybe not supported by openpyxl")
+            )
+
+        if engine is None and read_ext in (".xlsx", ".xlsm"):
+            # GH 35029
+            request.applymarker(
+                pytest.mark.xfail(reason="Defaults to openpyxl, maybe not supported")
+            )
+
+        result = pd.read_excel("testdateoverflow" + read_ext)
+        tm.assert_frame_equal(result, expected)
+
+    def test_sheet_name(self, request, read_ext, engine, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        filename = "test1"
+        sheet_name = "Sheet1"
+
+        expected = df_ref
+        adjust_expected(expected, read_ext, engine)
+
+        df1 = pd.read_excel(
+            filename + read_ext, sheet_name=sheet_name, index_col=0
+        )  # doc
+        df2 = pd.read_excel(filename + read_ext, index_col=0, sheet_name=sheet_name)
+
+        tm.assert_frame_equal(df1, expected)
+        tm.assert_frame_equal(df2, expected)
+
+    def test_excel_read_buffer(self, read_ext):
+        pth = "test1" + read_ext
+        expected = pd.read_excel(pth, sheet_name="Sheet1", index_col=0)
+        with open(pth, "rb") as f:
+            actual = pd.read_excel(f, sheet_name="Sheet1", index_col=0)
+            tm.assert_frame_equal(expected, actual)
+
+    def test_bad_engine_raises(self):
+        bad_engine = "foo"
+        with pytest.raises(ValueError, match="Unknown engine: foo"):
+            pd.read_excel("", engine=bad_engine)
+
+    @pytest.mark.parametrize(
+        "sheet_name",
+        [3, [0, 3], [3, 0], "Sheet4", ["Sheet1", "Sheet4"], ["Sheet4", "Sheet1"]],
+    )
+    def test_bad_sheetname_raises(self, read_ext, sheet_name):
+        # GH 39250
+        msg = "Worksheet index 3 is invalid|Worksheet named 'Sheet4' not found"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel("blank" + read_ext, sheet_name=sheet_name)
+
+    def test_missing_file_raises(self, read_ext):
+        bad_file = f"foo{read_ext}"
+        # CI tests with other languages, translates to "No such file or directory"
+        match = "|".join(
+            [
+                "(No such file or directory",
+                "没有那个文件或目录",
+                "File o directory non esistente)",
+            ]
+        )
+        with pytest.raises(FileNotFoundError, match=match):
+            pd.read_excel(bad_file)
+
+    def test_corrupt_bytes_raises(self, engine):
+        bad_stream = b"foo"
+        if engine is None:
+            error = ValueError
+            msg = (
+                "Excel file format cannot be determined, you must "
+                "specify an engine manually."
+            )
+        elif engine == "xlrd":
+            from xlrd import XLRDError
+
+            error = XLRDError
+            msg = (
+                "Unsupported format, or corrupt file: Expected BOF record; found b'foo'"
+            )
+        elif engine == "calamine":
+            from python_calamine import CalamineError
+
+            error = CalamineError
+            msg = "Cannot detect file format"
+        else:
+            error = BadZipFile
+            msg = "File is not a zip file"
+        with pytest.raises(error, match=msg):
+            pd.read_excel(BytesIO(bad_stream))
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_read_from_http_url(self, httpserver, read_ext):
+        with open("test1" + read_ext, "rb") as f:
+            httpserver.serve_content(content=f.read())
+        url_table = pd.read_excel(httpserver.url)
+        local_table = pd.read_excel("test1" + read_ext)
+        tm.assert_frame_equal(url_table, local_table)
+
+    @td.skip_if_not_us_locale
+    @pytest.mark.single_cpu
+    def test_read_from_s3_url(self, read_ext, s3_public_bucket, s3so):
+        # Bucket created in tests/io/conftest.py
+        with open("test1" + read_ext, "rb") as f:
+            s3_public_bucket.put_object(Key="test1" + read_ext, Body=f)
+
+        url = f"s3://{s3_public_bucket.name}/test1" + read_ext
+
+        url_table = pd.read_excel(url, storage_options=s3so)
+        local_table = pd.read_excel("test1" + read_ext)
+        tm.assert_frame_equal(url_table, local_table)
+
+    @pytest.mark.single_cpu
+    def test_read_from_s3_object(self, read_ext, s3_public_bucket, s3so):
+        # GH 38788
+        # Bucket created in tests/io/conftest.py
+        with open("test1" + read_ext, "rb") as f:
+            s3_public_bucket.put_object(Key="test1" + read_ext, Body=f)
+
+        import s3fs
+
+        s3 = s3fs.S3FileSystem(**s3so)
+
+        with s3.open(f"s3://{s3_public_bucket.name}/test1" + read_ext) as f:
+            url_table = pd.read_excel(f)
+
+        local_table = pd.read_excel("test1" + read_ext)
+        tm.assert_frame_equal(url_table, local_table)
+
+    @pytest.mark.slow
+    def test_read_from_file_url(self, read_ext, datapath):
+        # FILE
+        localtable = os.path.join(datapath("io", "data", "excel"), "test1" + read_ext)
+        local_table = pd.read_excel(localtable)
+
+        try:
+            url_table = pd.read_excel("file://localhost/" + localtable)
+        except URLError:
+            # fails on some systems
+            platform_info = " ".join(platform.uname()).strip()
+            pytest.skip(f"failing on {platform_info}")
+
+        tm.assert_frame_equal(url_table, local_table)
+
+    def test_read_from_pathlib_path(self, read_ext):
+        # GH12655
+        str_path = "test1" + read_ext
+        expected = pd.read_excel(str_path, sheet_name="Sheet1", index_col=0)
+
+        path_obj = Path("test1" + read_ext)
+        actual = pd.read_excel(path_obj, sheet_name="Sheet1", index_col=0)
+
+        tm.assert_frame_equal(expected, actual)
+
+    def test_close_from_py_localpath(self, read_ext):
+        # GH31467
+        str_path = os.path.join("test1" + read_ext)
+        with open(str_path, "rb") as f:
+            x = pd.read_excel(f, sheet_name="Sheet1", index_col=0)
+            del x
+            # should not throw an exception because the passed file was closed
+            f.read()
+
+    def test_reader_seconds(self, request, engine, read_ext):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        # GH 55045
+        if engine == "calamine" and read_ext == ".ods":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="ODS file contains bad datetime (seconds as text)"
+                )
+            )
+
+        # Test reading times with and without milliseconds. GH5945.
+        expected = DataFrame.from_dict(
+            {
+                "Time": [
+                    time(1, 2, 3),
+                    time(2, 45, 56, 100000),
+                    time(4, 29, 49, 200000),
+                    time(6, 13, 42, 300000),
+                    time(7, 57, 35, 400000),
+                    time(9, 41, 28, 500000),
+                    time(11, 25, 21, 600000),
+                    time(13, 9, 14, 700000),
+                    time(14, 53, 7, 800000),
+                    time(16, 37, 0, 900000),
+                    time(18, 20, 54),
+                ]
+            }
+        )
+
+        actual = pd.read_excel("times_1900" + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, expected)
+
+        actual = pd.read_excel("times_1904" + read_ext, sheet_name="Sheet1")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_multiindex(self, request, engine, read_ext):
+        # see gh-4679
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        unit = get_exp_unit(read_ext, engine)
+
+        mi = MultiIndex.from_product([["foo", "bar"], ["a", "b"]])
+        mi_file = "testmultiindex" + read_ext
+
+        # "mi_column" sheet
+        expected = DataFrame(
+            [
+                [1, 2.5, pd.Timestamp("2015-01-01"), True],
+                [2, 3.5, pd.Timestamp("2015-01-02"), False],
+                [3, 4.5, pd.Timestamp("2015-01-03"), False],
+                [4, 5.5, pd.Timestamp("2015-01-04"), True],
+            ],
+            columns=mi,
+        )
+        expected[mi[2]] = expected[mi[2]].astype(f"M8[{unit}]")
+
+        actual = pd.read_excel(
+            mi_file, sheet_name="mi_column", header=[0, 1], index_col=0
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # "mi_index" sheet
+        expected.index = mi
+        expected.columns = ["a", "b", "c", "d"]
+
+        actual = pd.read_excel(mi_file, sheet_name="mi_index", index_col=[0, 1])
+        tm.assert_frame_equal(actual, expected)
+
+        # "both" sheet
+        expected.columns = mi
+
+        actual = pd.read_excel(
+            mi_file, sheet_name="both", index_col=[0, 1], header=[0, 1]
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # "mi_index_name" sheet
+        expected.columns = ["a", "b", "c", "d"]
+        expected.index = mi.set_names(["ilvl1", "ilvl2"])
+
+        actual = pd.read_excel(mi_file, sheet_name="mi_index_name", index_col=[0, 1])
+        tm.assert_frame_equal(actual, expected)
+
+        # "mi_column_name" sheet
+        expected.index = range(4)
+        expected.columns = mi.set_names(["c1", "c2"])
+        actual = pd.read_excel(
+            mi_file, sheet_name="mi_column_name", header=[0, 1], index_col=0
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # see gh-11317
+        # "name_with_int" sheet
+        expected.columns = mi.set_levels([1, 2], level=1).set_names(["c1", "c2"])
+
+        actual = pd.read_excel(
+            mi_file, sheet_name="name_with_int", index_col=0, header=[0, 1]
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # "both_name" sheet
+        expected.columns = mi.set_names(["c1", "c2"])
+        expected.index = mi.set_names(["ilvl1", "ilvl2"])
+
+        actual = pd.read_excel(
+            mi_file, sheet_name="both_name", index_col=[0, 1], header=[0, 1]
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # "both_skiprows" sheet
+        actual = pd.read_excel(
+            mi_file,
+            sheet_name="both_name_skiprows",
+            index_col=[0, 1],
+            header=[0, 1],
+            skiprows=2,
+        )
+        tm.assert_frame_equal(actual, expected)
+
+    @pytest.mark.parametrize(
+        "sheet_name,idx_lvl2",
+        [
+            ("both_name_blank_after_mi_name", [np.nan, "b", "a", "b"]),
+            ("both_name_multiple_blanks", [np.nan] * 4),
+        ],
+    )
+    def test_read_excel_multiindex_blank_after_name(
+        self, request, engine, read_ext, sheet_name, idx_lvl2
+    ):
+        # GH34673
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        mi_file = "testmultiindex" + read_ext
+        mi = MultiIndex.from_product([["foo", "bar"], ["a", "b"]], names=["c1", "c2"])
+
+        unit = get_exp_unit(read_ext, engine)
+        expected = DataFrame(
+            [
+                [1, 2.5, pd.Timestamp("2015-01-01"), True],
+                [2, 3.5, pd.Timestamp("2015-01-02"), False],
+                [3, 4.5, pd.Timestamp("2015-01-03"), False],
+                [4, 5.5, pd.Timestamp("2015-01-04"), True],
+            ],
+            columns=mi,
+            index=MultiIndex.from_arrays(
+                (["foo", "foo", "bar", "bar"], idx_lvl2),
+                names=["ilvl1", "ilvl2"],
+            ),
+        )
+        expected[mi[2]] = expected[mi[2]].astype(f"M8[{unit}]")
+        result = pd.read_excel(
+            mi_file,
+            sheet_name=sheet_name,
+            index_col=[0, 1],
+            header=[0, 1],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_read_excel_multiindex_header_only(self, read_ext):
+        # see gh-11733.
+        #
+        # Don't try to parse a header name if there isn't one.
+        mi_file = "testmultiindex" + read_ext
+        result = pd.read_excel(mi_file, sheet_name="index_col_none", header=[0, 1])
+
+        exp_columns = MultiIndex.from_product([("A", "B"), ("key", "val")])
+        expected = DataFrame([[1, 2, 3, 4]] * 2, columns=exp_columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_excel_old_index_format(self, read_ext):
+        # see gh-4679
+        filename = "test_index_name_pre17" + read_ext
+
+        # We detect headers to determine if index names exist, so
+        # that "index" name in the "names" version of the data will
+        # now be interpreted as rows that include null data.
+        data = np.array(
+            [
+                [np.nan, np.nan, np.nan, np.nan, np.nan],
+                ["R0C0", "R0C1", "R0C2", "R0C3", "R0C4"],
+                ["R1C0", "R1C1", "R1C2", "R1C3", "R1C4"],
+                ["R2C0", "R2C1", "R2C2", "R2C3", "R2C4"],
+                ["R3C0", "R3C1", "R3C2", "R3C3", "R3C4"],
+                ["R4C0", "R4C1", "R4C2", "R4C3", "R4C4"],
+            ],
+            dtype=object,
+        )
+        columns = ["C_l0_g0", "C_l0_g1", "C_l0_g2", "C_l0_g3", "C_l0_g4"]
+        mi = MultiIndex(
+            levels=[
+                ["R0", "R_l0_g0", "R_l0_g1", "R_l0_g2", "R_l0_g3", "R_l0_g4"],
+                ["R1", "R_l1_g0", "R_l1_g1", "R_l1_g2", "R_l1_g3", "R_l1_g4"],
+            ],
+            codes=[[0, 1, 2, 3, 4, 5], [0, 1, 2, 3, 4, 5]],
+            names=[None, None],
+        )
+        si = Index(
+            ["R0", "R_l0_g0", "R_l0_g1", "R_l0_g2", "R_l0_g3", "R_l0_g4"], name=None
+        )
+
+        expected = DataFrame(data, index=si, columns=columns)
+
+        actual = pd.read_excel(filename, sheet_name="single_names", index_col=0)
+        tm.assert_frame_equal(actual, expected)
+
+        expected.index = mi
+
+        actual = pd.read_excel(filename, sheet_name="multi_names", index_col=[0, 1])
+        tm.assert_frame_equal(actual, expected)
+
+        # The analogous versions of the "names" version data
+        # where there are explicitly no names for the indices.
+        data = np.array(
+            [
+                ["R0C0", "R0C1", "R0C2", "R0C3", "R0C4"],
+                ["R1C0", "R1C1", "R1C2", "R1C3", "R1C4"],
+                ["R2C0", "R2C1", "R2C2", "R2C3", "R2C4"],
+                ["R3C0", "R3C1", "R3C2", "R3C3", "R3C4"],
+                ["R4C0", "R4C1", "R4C2", "R4C3", "R4C4"],
+            ]
+        )
+        columns = ["C_l0_g0", "C_l0_g1", "C_l0_g2", "C_l0_g3", "C_l0_g4"]
+        mi = MultiIndex(
+            levels=[
+                ["R_l0_g0", "R_l0_g1", "R_l0_g2", "R_l0_g3", "R_l0_g4"],
+                ["R_l1_g0", "R_l1_g1", "R_l1_g2", "R_l1_g3", "R_l1_g4"],
+            ],
+            codes=[[0, 1, 2, 3, 4], [0, 1, 2, 3, 4]],
+            names=[None, None],
+        )
+        si = Index(["R_l0_g0", "R_l0_g1", "R_l0_g2", "R_l0_g3", "R_l0_g4"], name=None)
+
+        expected = DataFrame(data, index=si, columns=columns)
+
+        actual = pd.read_excel(filename, sheet_name="single_no_names", index_col=0)
+        tm.assert_frame_equal(actual, expected)
+
+        expected.index = mi
+
+        actual = pd.read_excel(filename, sheet_name="multi_no_names", index_col=[0, 1])
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_bool_header_arg(self, read_ext):
+        # GH 6114
+        msg = "Passing a bool to header is invalid"
+        for arg in [True, False]:
+            with pytest.raises(TypeError, match=msg):
+                pd.read_excel("test1" + read_ext, header=arg)
+
+    def test_read_excel_skiprows(self, request, engine, read_ext):
+        # GH 4903
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        unit = get_exp_unit(read_ext, engine)
+
+        actual = pd.read_excel(
+            "testskiprows" + read_ext, sheet_name="skiprows_list", skiprows=[0, 2]
+        )
+        expected = DataFrame(
+            [
+                [1, 2.5, pd.Timestamp("2015-01-01"), True],
+                [2, 3.5, pd.Timestamp("2015-01-02"), False],
+                [3, 4.5, pd.Timestamp("2015-01-03"), False],
+                [4, 5.5, pd.Timestamp("2015-01-04"), True],
+            ],
+            columns=["a", "b", "c", "d"],
+        )
+        expected["c"] = expected["c"].astype(f"M8[{unit}]")
+        tm.assert_frame_equal(actual, expected)
+
+        actual = pd.read_excel(
+            "testskiprows" + read_ext,
+            sheet_name="skiprows_list",
+            skiprows=np.array([0, 2]),
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        # GH36435
+        actual = pd.read_excel(
+            "testskiprows" + read_ext,
+            sheet_name="skiprows_list",
+            skiprows=lambda x: x in [0, 2],
+        )
+        tm.assert_frame_equal(actual, expected)
+
+        actual = pd.read_excel(
+            "testskiprows" + read_ext,
+            sheet_name="skiprows_list",
+            skiprows=3,
+            names=["a", "b", "c", "d"],
+        )
+        expected = DataFrame(
+            [
+                # [1, 2.5, pd.Timestamp("2015-01-01"), True],
+                [2, 3.5, pd.Timestamp("2015-01-02"), False],
+                [3, 4.5, pd.Timestamp("2015-01-03"), False],
+                [4, 5.5, pd.Timestamp("2015-01-04"), True],
+            ],
+            columns=["a", "b", "c", "d"],
+        )
+        expected["c"] = expected["c"].astype(f"M8[{unit}]")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_skiprows_callable_not_in(self, request, engine, read_ext):
+        # GH 4903
+        xfail_datetimes_with_pyxlsb(engine, request)
+        unit = get_exp_unit(read_ext, engine)
+
+        actual = pd.read_excel(
+            "testskiprows" + read_ext,
+            sheet_name="skiprows_list",
+            skiprows=lambda x: x not in [1, 3, 5],
+        )
+        expected = DataFrame(
+            [
+                [1, 2.5, pd.Timestamp("2015-01-01"), True],
+                # [2, 3.5, pd.Timestamp("2015-01-02"), False],
+                [3, 4.5, pd.Timestamp("2015-01-03"), False],
+                # [4, 5.5, pd.Timestamp("2015-01-04"), True],
+            ],
+            columns=["a", "b", "c", "d"],
+        )
+        expected["c"] = expected["c"].astype(f"M8[{unit}]")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_nrows(self, read_ext):
+        # GH 16645
+        num_rows_to_pull = 5
+        actual = pd.read_excel("test1" + read_ext, nrows=num_rows_to_pull)
+        expected = pd.read_excel("test1" + read_ext)
+        expected = expected[:num_rows_to_pull]
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_nrows_greater_than_nrows_in_file(self, read_ext):
+        # GH 16645
+        expected = pd.read_excel("test1" + read_ext)
+        num_records_in_file = len(expected)
+        num_rows_to_pull = num_records_in_file + 10
+        actual = pd.read_excel("test1" + read_ext, nrows=num_rows_to_pull)
+        tm.assert_frame_equal(actual, expected)
+
+    def test_read_excel_nrows_non_integer_parameter(self, read_ext):
+        # GH 16645
+        msg = "'nrows' must be an integer >=0"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel("test1" + read_ext, nrows="5")
+
+    @pytest.mark.parametrize(
+        "filename,sheet_name,header,index_col,skiprows",
+        [
+            ("testmultiindex", "mi_column", [0, 1], 0, None),
+            ("testmultiindex", "mi_index", None, [0, 1], None),
+            ("testmultiindex", "both", [0, 1], [0, 1], None),
+            ("testmultiindex", "mi_column_name", [0, 1], 0, None),
+            ("testskiprows", "skiprows_list", None, None, [0, 2]),
+            ("testskiprows", "skiprows_list", None, None, lambda x: x in (0, 2)),
+        ],
+    )
+    def test_read_excel_nrows_params(
+        self, read_ext, filename, sheet_name, header, index_col, skiprows
+    ):
+        """
+        For various parameters, we should get the same result whether we
+        limit the rows during load (nrows=3) or after (df.iloc[:3]).
+        """
+        # GH 46894
+        expected = pd.read_excel(
+            filename + read_ext,
+            sheet_name=sheet_name,
+            header=header,
+            index_col=index_col,
+            skiprows=skiprows,
+        ).iloc[:3]
+        actual = pd.read_excel(
+            filename + read_ext,
+            sheet_name=sheet_name,
+            header=header,
+            index_col=index_col,
+            skiprows=skiprows,
+            nrows=3,
+        )
+        tm.assert_frame_equal(actual, expected)
+
+    def test_deprecated_kwargs(self, read_ext):
+        with pytest.raises(TypeError, match="but 3 positional arguments"):
+            pd.read_excel("test1" + read_ext, "Sheet1", 0)
+
+    def test_no_header_with_list_index_col(self, read_ext):
+        # GH 31783
+        file_name = "testmultiindex" + read_ext
+        data = [("B", "B"), ("key", "val"), (3, 4), (3, 4)]
+        idx = MultiIndex.from_tuples(
+            [("A", "A"), ("key", "val"), (1, 2), (1, 2)], names=(0, 1)
+        )
+        expected = DataFrame(data, index=idx, columns=(2, 3))
+        result = pd.read_excel(
+            file_name, sheet_name="index_col_none", index_col=[0, 1], header=None
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_one_col_noskip_blank_line(self, read_ext):
+        # GH 39808
+        file_name = "one_col_blank_line" + read_ext
+        data = [0.5, np.nan, 1, 2]
+        expected = DataFrame(data, columns=["numbers"])
+        result = pd.read_excel(file_name)
+        tm.assert_frame_equal(result, expected)
+
+    def test_multiheader_two_blank_lines(self, read_ext):
+        # GH 40442
+        file_name = "testmultiindex" + read_ext
+        columns = MultiIndex.from_tuples([("a", "A"), ("b", "B")])
+        data = [[np.nan, np.nan], [np.nan, np.nan], [1, 3], [2, 4]]
+        expected = DataFrame(data, columns=columns)
+        result = pd.read_excel(
+            file_name, sheet_name="mi_column_empty_rows", header=[0, 1]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_trailing_blanks(self, read_ext):
+        """
+        Sheets can contain blank cells with no data. Some of our readers
+        were including those cells, creating many empty rows and columns
+        """
+        file_name = "trailing_blanks" + read_ext
+        result = pd.read_excel(file_name)
+        assert result.shape == (3, 3)
+
+    def test_ignore_chartsheets_by_str(self, request, engine, read_ext):
+        # GH 41448
+        if read_ext == ".ods":
+            pytest.skip("chartsheets do not exist in the ODF format")
+        if engine == "pyxlsb":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="pyxlsb can't distinguish chartsheets from worksheets"
+                )
+            )
+        with pytest.raises(ValueError, match="Worksheet named 'Chart1' not found"):
+            pd.read_excel("chartsheet" + read_ext, sheet_name="Chart1")
+
+    def test_ignore_chartsheets_by_int(self, request, engine, read_ext):
+        # GH 41448
+        if read_ext == ".ods":
+            pytest.skip("chartsheets do not exist in the ODF format")
+        if engine == "pyxlsb":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="pyxlsb can't distinguish chartsheets from worksheets"
+                )
+            )
+        with pytest.raises(
+            ValueError, match="Worksheet index 1 is invalid, 1 worksheets found"
+        ):
+            pd.read_excel("chartsheet" + read_ext, sheet_name=1)
+
+    def test_euro_decimal_format(self, read_ext):
+        # copied from read_csv
+        result = pd.read_excel("test_decimal" + read_ext, decimal=",", skiprows=1)
+        expected = DataFrame(
+            [
+                [1, 1521.1541, 187101.9543, "ABC", "poi", 4.738797819],
+                [2, 121.12, 14897.76, "DEF", "uyt", 0.377320872],
+                [3, 878.158, 108013.434, "GHI", "rez", 2.735694704],
+            ],
+            columns=["Id", "Number1", "Number2", "Text1", "Text2", "Number3"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+class TestExcelFileRead:
+    def test_raises_bytes_input(self, engine, read_ext):
+        # GH 53830
+        msg = "Expected file path name or file-like object"
+        with pytest.raises(TypeError, match=msg):
+            with open("test1" + read_ext, "rb") as f:
+                pd.read_excel(f.read(), engine=engine)
+
+    @pytest.fixture(autouse=True)
+    def cd_and_set_engine(self, engine, datapath, monkeypatch):
+        """
+        Change directory and set engine for ExcelFile objects.
+        """
+        func = partial(pd.ExcelFile, engine=engine)
+        monkeypatch.chdir(datapath("io", "data", "excel"))
+        monkeypatch.setattr(pd, "ExcelFile", func)
+
+    def test_engine_used(self, read_ext, engine):
+        expected_defaults = {
+            "xlsx": "openpyxl",
+            "xlsm": "openpyxl",
+            "xlsb": "pyxlsb",
+            "xls": "xlrd",
+            "ods": "odf",
+        }
+
+        with pd.ExcelFile("test1" + read_ext) as excel:
+            result = excel.engine
+
+        if engine is not None:
+            expected = engine
+        else:
+            expected = expected_defaults[read_ext[1:]]
+        assert result == expected
+
+    def test_excel_passes_na(self, read_ext):
+        with pd.ExcelFile("test4" + read_ext) as excel:
+            parsed = pd.read_excel(
+                excel, sheet_name="Sheet1", keep_default_na=False, na_values=["apple"]
+            )
+        expected = DataFrame(
+            [["NA"], [1], ["NA"], [np.nan], ["rabbit"]], columns=["Test"]
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+        with pd.ExcelFile("test4" + read_ext) as excel:
+            parsed = pd.read_excel(
+                excel, sheet_name="Sheet1", keep_default_na=True, na_values=["apple"]
+            )
+        expected = DataFrame(
+            [[np.nan], [1], [np.nan], [np.nan], ["rabbit"]], columns=["Test"]
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+        # 13967
+        with pd.ExcelFile("test5" + read_ext) as excel:
+            parsed = pd.read_excel(
+                excel, sheet_name="Sheet1", keep_default_na=False, na_values=["apple"]
+            )
+        expected = DataFrame(
+            [["1.#QNAN"], [1], ["nan"], [np.nan], ["rabbit"]], columns=["Test"]
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+        with pd.ExcelFile("test5" + read_ext) as excel:
+            parsed = pd.read_excel(
+                excel, sheet_name="Sheet1", keep_default_na=True, na_values=["apple"]
+            )
+        expected = DataFrame(
+            [[np.nan], [1], [np.nan], [np.nan], ["rabbit"]], columns=["Test"]
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.mark.parametrize("na_filter", [None, True, False])
+    def test_excel_passes_na_filter(self, read_ext, na_filter):
+        # gh-25453
+        kwargs = {}
+
+        if na_filter is not None:
+            kwargs["na_filter"] = na_filter
+
+        with pd.ExcelFile("test5" + read_ext) as excel:
+            parsed = pd.read_excel(
+                excel,
+                sheet_name="Sheet1",
+                keep_default_na=True,
+                na_values=["apple"],
+                **kwargs,
+            )
+
+        if na_filter is False:
+            expected = [["1.#QNAN"], [1], ["nan"], ["apple"], ["rabbit"]]
+        else:
+            expected = [[np.nan], [1], [np.nan], [np.nan], ["rabbit"]]
+
+        expected = DataFrame(expected, columns=["Test"])
+        tm.assert_frame_equal(parsed, expected)
+
+    def test_excel_table_sheet_by_index(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref
+        adjust_expected(expected, read_ext, engine)
+
+        with pd.ExcelFile("test1" + read_ext) as excel:
+            df1 = pd.read_excel(excel, sheet_name=0, index_col=0)
+            df2 = pd.read_excel(excel, sheet_name=1, skiprows=[1], index_col=0)
+        tm.assert_frame_equal(df1, expected)
+        tm.assert_frame_equal(df2, expected)
+
+        with pd.ExcelFile("test1" + read_ext) as excel:
+            df1 = excel.parse(0, index_col=0)
+            df2 = excel.parse(1, skiprows=[1], index_col=0)
+        tm.assert_frame_equal(df1, expected)
+        tm.assert_frame_equal(df2, expected)
+
+        with pd.ExcelFile("test1" + read_ext) as excel:
+            df3 = pd.read_excel(excel, sheet_name=0, index_col=0, skipfooter=1)
+        tm.assert_frame_equal(df3, df1.iloc[:-1])
+
+        with pd.ExcelFile("test1" + read_ext) as excel:
+            df3 = excel.parse(0, index_col=0, skipfooter=1)
+
+        tm.assert_frame_equal(df3, df1.iloc[:-1])
+
+    def test_sheet_name(self, request, engine, read_ext, df_ref):
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        expected = df_ref
+        adjust_expected(expected, read_ext, engine)
+
+        filename = "test1"
+        sheet_name = "Sheet1"
+
+        with pd.ExcelFile(filename + read_ext) as excel:
+            df1_parse = excel.parse(sheet_name=sheet_name, index_col=0)  # doc
+
+        with pd.ExcelFile(filename + read_ext) as excel:
+            df2_parse = excel.parse(index_col=0, sheet_name=sheet_name)
+
+        tm.assert_frame_equal(df1_parse, expected)
+        tm.assert_frame_equal(df2_parse, expected)
+
+    @pytest.mark.parametrize(
+        "sheet_name",
+        [3, [0, 3], [3, 0], "Sheet4", ["Sheet1", "Sheet4"], ["Sheet4", "Sheet1"]],
+    )
+    def test_bad_sheetname_raises(self, read_ext, sheet_name):
+        # GH 39250
+        msg = "Worksheet index 3 is invalid|Worksheet named 'Sheet4' not found"
+        with pytest.raises(ValueError, match=msg):
+            with pd.ExcelFile("blank" + read_ext) as excel:
+                excel.parse(sheet_name=sheet_name)
+
+    def test_excel_read_buffer(self, engine, read_ext):
+        pth = "test1" + read_ext
+        expected = pd.read_excel(pth, sheet_name="Sheet1", index_col=0, engine=engine)
+
+        with open(pth, "rb") as f:
+            with pd.ExcelFile(f) as xls:
+                actual = pd.read_excel(xls, sheet_name="Sheet1", index_col=0)
+
+        tm.assert_frame_equal(expected, actual)
+
+    def test_reader_closes_file(self, engine, read_ext):
+        with open("test1" + read_ext, "rb") as f:
+            with pd.ExcelFile(f) as xlsx:
+                # parses okay
+                pd.read_excel(xlsx, sheet_name="Sheet1", index_col=0, engine=engine)
+
+        assert f.closed
+
+    def test_conflicting_excel_engines(self, read_ext):
+        # GH 26566
+        msg = "Engine should not be specified when passing an ExcelFile"
+
+        with pd.ExcelFile("test1" + read_ext) as xl:
+            with pytest.raises(ValueError, match=msg):
+                pd.read_excel(xl, engine="foo")
+
+    def test_excel_read_binary(self, engine, read_ext):
+        # GH 15914
+        expected = pd.read_excel("test1" + read_ext, engine=engine)
+
+        with open("test1" + read_ext, "rb") as f:
+            data = f.read()
+
+        actual = pd.read_excel(BytesIO(data), engine=engine)
+        tm.assert_frame_equal(expected, actual)
+
+    def test_excel_read_binary_via_read_excel(self, read_ext, engine):
+        # GH 38424
+        with open("test1" + read_ext, "rb") as f:
+            result = pd.read_excel(f, engine=engine)
+        expected = pd.read_excel("test1" + read_ext, engine=engine)
+        tm.assert_frame_equal(result, expected)
+
+    def test_read_excel_header_index_out_of_range(self, engine):
+        # GH#43143
+        with open("df_header_oob.xlsx", "rb") as f:
+            with pytest.raises(ValueError, match="exceeds maximum"):
+                pd.read_excel(f, header=[0, 1])
+
+    @pytest.mark.parametrize("filename", ["df_empty.xlsx", "df_equals.xlsx"])
+    def test_header_with_index_col(self, filename):
+        # GH 33476
+        idx = Index(["Z"], name="I2")
+        cols = MultiIndex.from_tuples([("A", "B"), ("A", "B.1")], names=["I11", "I12"])
+        expected = DataFrame([[1, 3]], index=idx, columns=cols, dtype="int64")
+        result = pd.read_excel(
+            filename, sheet_name="Sheet1", index_col=0, header=[0, 1]
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_read_datetime_multiindex(self, request, engine, read_ext):
+        # GH 34748
+        xfail_datetimes_with_pyxlsb(engine, request)
+
+        f = "test_datetime_mi" + read_ext
+        with pd.ExcelFile(f) as excel:
+            actual = pd.read_excel(excel, header=[0, 1], index_col=0, engine=engine)
+
+        unit = get_exp_unit(read_ext, engine)
+
+        dti = pd.DatetimeIndex(["2020-02-29", "2020-03-01"], dtype=f"M8[{unit}]")
+        expected_column_index = MultiIndex.from_arrays(
+            [dti[:1], dti[1:]],
+            names=[
+                dti[0].to_pydatetime(),
+                dti[1].to_pydatetime(),
+            ],
+        )
+        expected = DataFrame([], index=[], columns=expected_column_index)
+
+        tm.assert_frame_equal(expected, actual)
+
+    def test_engine_invalid_option(self, read_ext):
+        # read_ext includes the '.' hence the weird formatting
+        with pytest.raises(ValueError, match="Value must be one of *"):
+            with pd.option_context(f"io.excel{read_ext}.reader", "abc"):
+                pass
+
+    def test_ignore_chartsheets(self, request, engine, read_ext):
+        # GH 41448
+        if read_ext == ".ods":
+            pytest.skip("chartsheets do not exist in the ODF format")
+        if engine == "pyxlsb":
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="pyxlsb can't distinguish chartsheets from worksheets"
+                )
+            )
+        with pd.ExcelFile("chartsheet" + read_ext) as excel:
+            assert excel.sheet_names == ["Sheet1"]
+
+    def test_corrupt_files_closed(self, engine, tmp_excel):
+        # GH41778
+        errors = (BadZipFile,)
+        if engine is None:
+            pytest.skip(f"Invalid test for engine={engine}")
+        elif engine == "xlrd":
+            import xlrd
+
+            errors = (BadZipFile, xlrd.biffh.XLRDError)
+        elif engine == "calamine":
+            from python_calamine import CalamineError
+
+            errors = (CalamineError,)
+
+        Path(tmp_excel).write_text("corrupt", encoding="utf-8")
+        with tm.assert_produces_warning(False):
+            try:
+                pd.ExcelFile(tmp_excel, engine=engine)
+            except errors:
+                pass
diff --git a/pandas/pandas/tests/io/excel/test_style.py b/pandas/pandas/tests/io/excel/test_style.py
new file mode 100644
index 0000000000000..0e13b2f94ed58
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_style.py
@@ -0,0 +1,361 @@
+import contextlib
+import time
+import uuid
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Timestamp,
+    period_range,
+    read_excel,
+)
+import pandas._testing as tm
+
+from pandas.io.excel import ExcelWriter
+from pandas.io.formats.excel import ExcelFormatter
+
+pytest.importorskip("jinja2")
+# jinja2 is currently required for Styler.__init__(). Technically Styler.to_excel
+# could compute styles and render to excel without jinja2, since there is no
+# 'template' file, but this needs the import error to delayed until render time.
+
+
+@pytest.fixture
+def tmp_excel(tmp_path):
+    tmp = tmp_path / f"{uuid.uuid4()}.xlsx"
+    tmp.touch()
+    return str(tmp)
+
+
+def assert_equal_cell_styles(cell1, cell2):
+    # TODO: should find a better way to check equality
+    assert cell1.alignment.__dict__ == cell2.alignment.__dict__
+    assert cell1.border.__dict__ == cell2.border.__dict__
+    assert cell1.fill.__dict__ == cell2.fill.__dict__
+    assert cell1.font.__dict__ == cell2.font.__dict__
+    assert cell1.number_format == cell2.number_format
+    assert cell1.protection.__dict__ == cell2.protection.__dict__
+
+
+def test_styler_default_values(tmp_excel):
+    # GH 54154
+    openpyxl = pytest.importorskip("openpyxl")
+    df = DataFrame([{"A": 1, "B": 2, "C": 3}, {"A": 1, "B": 2, "C": 3}])
+
+    with ExcelWriter(tmp_excel, engine="openpyxl") as writer:
+        df.to_excel(writer, sheet_name="custom")
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        # Check font, spacing, indentation
+        assert wb["custom"].cell(1, 1).font.bold is False
+        assert wb["custom"].cell(1, 1).alignment.horizontal is None
+        assert wb["custom"].cell(1, 1).alignment.vertical is None
+
+        # Check border
+        assert wb["custom"].cell(1, 1).border.bottom.color is None
+        assert wb["custom"].cell(1, 1).border.top.color is None
+        assert wb["custom"].cell(1, 1).border.left.color is None
+        assert wb["custom"].cell(1, 1).border.right.color is None
+
+
+@pytest.mark.parametrize("engine", ["xlsxwriter", "openpyxl"])
+def test_styler_to_excel_unstyled(engine, tmp_excel):
+    # compare DataFrame.to_excel and Styler.to_excel when no styles applied
+    pytest.importorskip(engine)
+    df = DataFrame(np.random.default_rng(2).standard_normal((2, 2)))
+    with ExcelWriter(tmp_excel, engine=engine) as writer:
+        df.to_excel(writer, sheet_name="dataframe")
+        df.style.to_excel(writer, sheet_name="unstyled")
+
+    openpyxl = pytest.importorskip("openpyxl")  # test loading only with openpyxl
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        for col1, col2 in zip(wb["dataframe"].columns, wb["unstyled"].columns):
+            assert len(col1) == len(col2)
+            for cell1, cell2 in zip(col1, col2):
+                assert cell1.value == cell2.value
+                assert_equal_cell_styles(cell1, cell2)
+
+
+shared_style_params = [
+    (
+        "background-color: #111222",
+        ["fill", "fgColor", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    (
+        "color: #111222",
+        ["font", "color", "value"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    ("font-family: Arial;", ["font", "name"], "arial"),
+    ("font-weight: bold;", ["font", "b"], True),
+    ("font-style: italic;", ["font", "i"], True),
+    ("text-decoration: underline;", ["font", "u"], "single"),
+    ("number-format: $??,???.00;", ["number_format"], "$??,???.00"),
+    ("text-align: left;", ["alignment", "horizontal"], "left"),
+    (
+        "vertical-align: bottom;",
+        ["alignment", "vertical"],
+        {"xlsxwriter": None, "openpyxl": "bottom"},  # xlsxwriter Fails
+    ),
+    ("vertical-align: middle;", ["alignment", "vertical"], "center"),
+    # Border widths
+    ("border-left: 2pt solid red", ["border", "left", "style"], "medium"),
+    ("border-left: 1pt dotted red", ["border", "left", "style"], "dotted"),
+    ("border-left: 2pt dotted red", ["border", "left", "style"], "mediumDashDotDot"),
+    ("border-left: 1pt dashed red", ["border", "left", "style"], "dashed"),
+    ("border-left: 2pt dashed red", ["border", "left", "style"], "mediumDashed"),
+    ("border-left: 1pt solid red", ["border", "left", "style"], "thin"),
+    ("border-left: 3pt solid red", ["border", "left", "style"], "thick"),
+    # Border expansion
+    (
+        "border-left: 2pt solid #111222",
+        ["border", "left", "color", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    ("border: 1pt solid red", ["border", "top", "style"], "thin"),
+    (
+        "border: 1pt solid #111222",
+        ["border", "top", "color", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    ("border: 1pt solid red", ["border", "right", "style"], "thin"),
+    (
+        "border: 1pt solid #111222",
+        ["border", "right", "color", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    ("border: 1pt solid red", ["border", "bottom", "style"], "thin"),
+    (
+        "border: 1pt solid #111222",
+        ["border", "bottom", "color", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    ("border: 1pt solid red", ["border", "left", "style"], "thin"),
+    (
+        "border: 1pt solid #111222",
+        ["border", "left", "color", "rgb"],
+        {"xlsxwriter": "FF111222", "openpyxl": "00111222"},
+    ),
+    # Border styles
+    (
+        "border-left-style: hair; border-left-color: black",
+        ["border", "left", "style"],
+        "hair",
+    ),
+]
+
+
+def test_styler_custom_style(tmp_excel):
+    # GH 54154
+    css_style = "background-color: #111222"
+    openpyxl = pytest.importorskip("openpyxl")
+    df = DataFrame([{"A": 1, "B": 2}, {"A": 1, "B": 2}])
+
+    with ExcelWriter(tmp_excel, engine="openpyxl") as writer:
+        styler = df.style.map(lambda x: css_style)
+        styler.to_excel(writer, sheet_name="custom", index=False)
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        # Check font, spacing, indentation
+        assert wb["custom"].cell(1, 1).font.bold is False
+        assert wb["custom"].cell(1, 1).alignment.horizontal is None
+        assert wb["custom"].cell(1, 1).alignment.vertical is None
+
+        # Check border
+        assert wb["custom"].cell(1, 1).border.bottom.color is None
+        assert wb["custom"].cell(1, 1).border.top.color is None
+        assert wb["custom"].cell(1, 1).border.left.color is None
+        assert wb["custom"].cell(1, 1).border.right.color is None
+
+        # Check background color
+        assert wb["custom"].cell(2, 1).fill.fgColor.index == "00111222"
+        assert wb["custom"].cell(3, 1).fill.fgColor.index == "00111222"
+        assert wb["custom"].cell(2, 2).fill.fgColor.index == "00111222"
+        assert wb["custom"].cell(3, 2).fill.fgColor.index == "00111222"
+
+
+@pytest.mark.parametrize("engine", ["xlsxwriter", "openpyxl"])
+@pytest.mark.parametrize("css, attrs, expected", shared_style_params)
+def test_styler_to_excel_basic(engine, css, attrs, expected, tmp_excel):
+    pytest.importorskip(engine)
+    df = DataFrame(np.random.default_rng(2).standard_normal((1, 1)))
+    styler = df.style.map(lambda x: css)
+
+    with ExcelWriter(tmp_excel, engine=engine) as writer:
+        df.to_excel(writer, sheet_name="dataframe")
+        styler.to_excel(writer, sheet_name="styled")
+
+    openpyxl = pytest.importorskip("openpyxl")  # test loading only with openpyxl
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        # test unstyled data cell does not have expected styles
+        # test styled cell has expected styles
+        u_cell, s_cell = wb["dataframe"].cell(2, 2), wb["styled"].cell(2, 2)
+    for attr in attrs:
+        u_cell, s_cell = getattr(u_cell, attr, None), getattr(s_cell, attr)
+
+    if isinstance(expected, dict):
+        assert u_cell is None or u_cell != expected[engine]
+        assert s_cell == expected[engine]
+    else:
+        assert u_cell is None or u_cell != expected
+        assert s_cell == expected
+
+
+@pytest.mark.parametrize("engine", ["xlsxwriter", "openpyxl"])
+@pytest.mark.parametrize("css, attrs, expected", shared_style_params)
+def test_styler_to_excel_basic_indexes(engine, css, attrs, expected, tmp_excel):
+    pytest.importorskip(engine)
+    df = DataFrame(np.random.default_rng(2).standard_normal((1, 1)))
+
+    styler = df.style
+    styler.map_index(lambda x: css, axis=0)
+    styler.map_index(lambda x: css, axis=1)
+
+    null_styler = df.style
+    null_styler.map(lambda x: "null: css;")
+    null_styler.map_index(lambda x: "null: css;", axis=0)
+    null_styler.map_index(lambda x: "null: css;", axis=1)
+
+    with ExcelWriter(tmp_excel, engine=engine) as writer:
+        null_styler.to_excel(writer, sheet_name="null_styled")
+        styler.to_excel(writer, sheet_name="styled")
+
+    openpyxl = pytest.importorskip("openpyxl")  # test loading only with openpyxl
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        # test null styled index cells does not have expected styles
+        # test styled cell has expected styles
+        ui_cell, si_cell = wb["null_styled"].cell(2, 1), wb["styled"].cell(2, 1)
+        uc_cell, sc_cell = wb["null_styled"].cell(1, 2), wb["styled"].cell(1, 2)
+    for attr in attrs:
+        ui_cell, si_cell = getattr(ui_cell, attr, None), getattr(si_cell, attr)
+        uc_cell, sc_cell = getattr(uc_cell, attr, None), getattr(sc_cell, attr)
+
+    if isinstance(expected, dict):
+        assert ui_cell is None or ui_cell != expected[engine]
+        assert si_cell == expected[engine]
+        assert uc_cell is None or uc_cell != expected[engine]
+        assert sc_cell == expected[engine]
+    else:
+        assert ui_cell is None or ui_cell != expected
+        assert si_cell == expected
+        assert uc_cell is None or uc_cell != expected
+        assert sc_cell == expected
+
+
+# From https://openpyxl.readthedocs.io/en/stable/api/openpyxl.styles.borders.html
+# Note: Leaving behavior of "width"-type styles undefined; user should use border-width
+# instead
+excel_border_styles = [
+    # "thin",
+    "dashed",
+    "mediumDashDot",
+    "dashDotDot",
+    "hair",
+    "dotted",
+    "mediumDashDotDot",
+    # "medium",
+    "double",
+    "dashDot",
+    "slantDashDot",
+    # "thick",
+    "mediumDashed",
+]
+
+
+@pytest.mark.parametrize("engine", ["xlsxwriter", "openpyxl"])
+@pytest.mark.parametrize("border_style", excel_border_styles)
+def test_styler_to_excel_border_style(engine, border_style, tmp_excel):
+    css = f"border-left: {border_style} black thin"
+    attrs = ["border", "left", "style"]
+    expected = border_style
+
+    pytest.importorskip(engine)
+    df = DataFrame(np.random.default_rng(2).standard_normal((1, 1)))
+    styler = df.style.map(lambda x: css)
+
+    with ExcelWriter(tmp_excel, engine=engine) as writer:
+        df.to_excel(writer, sheet_name="dataframe")
+        styler.to_excel(writer, sheet_name="styled")
+
+    openpyxl = pytest.importorskip("openpyxl")  # test loading only with openpyxl
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        # test unstyled data cell does not have expected styles
+        # test styled cell has expected styles
+        u_cell, s_cell = wb["dataframe"].cell(2, 2), wb["styled"].cell(2, 2)
+    for attr in attrs:
+        u_cell, s_cell = getattr(u_cell, attr, None), getattr(s_cell, attr)
+
+    if isinstance(expected, dict):
+        assert u_cell is None or u_cell != expected[engine]
+        assert s_cell == expected[engine]
+    else:
+        assert u_cell is None or u_cell != expected
+        assert s_cell == expected
+
+
+def test_styler_custom_converter(tmp_excel):
+    openpyxl = pytest.importorskip("openpyxl")
+
+    def custom_converter(css):
+        return {"font": {"color": {"rgb": "111222"}}}
+
+    df = DataFrame(np.random.default_rng(2).standard_normal((1, 1)))
+    styler = df.style.map(lambda x: "color: #888999")
+    with ExcelWriter(tmp_excel, engine="openpyxl") as writer:
+        ExcelFormatter(styler, style_converter=custom_converter).write(
+            writer, sheet_name="custom"
+        )
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as wb:
+        assert wb["custom"].cell(2, 2).font.color.value == "00111222"
+
+
+@pytest.mark.single_cpu
+@td.skip_if_not_us_locale
+def test_styler_to_s3(s3_public_bucket, s3so):
+    # GH#46381
+
+    mock_bucket_name, target_file = s3_public_bucket.name, "test.xlsx"
+    df = DataFrame({"x": [1, 2, 3], "y": [2, 4, 6]})
+    styler = df.style.set_sticky(axis="index")
+    styler.to_excel(f"s3://{mock_bucket_name}/{target_file}", storage_options=s3so)
+    timeout = 5
+    while True:
+        if target_file in (obj.key for obj in s3_public_bucket.objects.all()):
+            break
+        time.sleep(0.1)
+        timeout -= 0.1
+        assert timeout > 0, "Timed out waiting for file to appear on moto"
+        result = read_excel(
+            f"s3://{mock_bucket_name}/{target_file}", index_col=0, storage_options=s3so
+        )
+        tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("merge_cells", [True, False, "columns"])
+def test_format_hierarchical_rows_periodindex(merge_cells):
+    # GH#60099
+    df = DataFrame(
+        {"A": [1, 2]},
+        index=MultiIndex.from_arrays(
+            [
+                period_range(start="2006-10-06", end="2006-10-07", freq="D"),
+                ["X", "Y"],
+            ],
+            names=["date", "category"],
+        ),
+    )
+    formatter = ExcelFormatter(df, merge_cells=merge_cells)
+    formatted_cells = formatter._format_hierarchical_rows()
+
+    for cell in formatted_cells:
+        if cell.row != 0 and cell.col == 0:
+            assert isinstance(cell.val, Timestamp), (
+                "Period should be converted to Timestamp"
+            )
diff --git a/pandas/pandas/tests/io/excel/test_writers.py b/pandas/pandas/tests/io/excel/test_writers.py
new file mode 100644
index 0000000000000..ced4feb9e7eb9
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_writers.py
@@ -0,0 +1,1611 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+from decimal import Decimal
+from functools import partial
+from io import BytesIO
+import os
+import re
+import uuid
+
+import numpy as np
+import pytest
+
+from pandas.compat._optional import import_optional_dependency
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    date_range,
+    option_context,
+    period_range,
+)
+import pandas._testing as tm
+
+from pandas.io.excel import (
+    ExcelFile,
+    ExcelWriter,
+    _OpenpyxlWriter,
+    _XlsxWriter,
+    register_writer,
+)
+from pandas.io.excel._util import _writers
+
+
+def get_exp_unit(path: str) -> str:
+    if path.endswith(".ods"):
+        return "s"
+    return "us"
+
+
+@pytest.fixture
+def frame(float_frame):
+    """
+    Returns the first ten items in fixture "float_frame".
+    """
+    return float_frame[:10]
+
+
+@pytest.fixture(params=[True, False, "columns"])
+def merge_cells(request):
+    return request.param
+
+
+@pytest.fixture
+def tmp_excel(ext, tmp_path):
+    """
+    Fixture to open file for use in each test case.
+    """
+    tmp = tmp_path / f"{uuid.uuid4()}{ext}"
+    tmp.touch()
+    return str(tmp)
+
+
+@pytest.fixture
+def set_engine(engine, ext):
+    """
+    Fixture to set engine for use in each test case.
+
+    Rather than requiring `engine=...` to be provided explicitly as an
+    argument in each test, this fixture sets a global option to dictate
+    which engine should be used to write Excel files. After executing
+    the test it rolls back said change to the global option.
+    """
+    option_name = f"io.excel.{ext.strip('.')}.writer"
+    with option_context(option_name, engine):
+        yield
+
+
+@pytest.mark.parametrize(
+    "ext",
+    [
+        pytest.param(".xlsx", marks=[td.skip_if_no("openpyxl"), td.skip_if_no("xlrd")]),
+        pytest.param(".xlsm", marks=[td.skip_if_no("openpyxl"), td.skip_if_no("xlrd")]),
+        pytest.param(
+            ".xlsx", marks=[td.skip_if_no("xlsxwriter"), td.skip_if_no("xlrd")]
+        ),
+        pytest.param(".ods", marks=td.skip_if_no("odf")),
+    ],
+)
+class TestRoundTrip:
+    @pytest.mark.parametrize(
+        "header,expected",
+        [(None, [np.nan] * 4), (0, {"Unnamed: 0": [np.nan] * 3})],
+    )
+    def test_read_one_empty_col_no_header(self, tmp_excel, header, expected):
+        # xref gh-12292
+        filename = "no_header"
+        df = DataFrame([["", 1, 100], ["", 2, 200], ["", 3, 300], ["", 4, 400]])
+
+        df.to_excel(tmp_excel, sheet_name=filename, index=False, header=False)
+        result = pd.read_excel(
+            tmp_excel, sheet_name=filename, usecols=[0], header=header
+        )
+        expected = DataFrame(expected)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "header,expected_extra",
+        [(None, [0]), (0, [])],
+    )
+    def test_read_one_empty_col_with_header(self, tmp_excel, header, expected_extra):
+        filename = "with_header"
+        df = DataFrame([["", 1, 100], ["", 2, 200], ["", 3, 300], ["", 4, 400]])
+
+        df.to_excel(tmp_excel, sheet_name="with_header", index=False, header=True)
+        result = pd.read_excel(
+            tmp_excel, sheet_name=filename, usecols=[0], header=header
+        )
+        expected = DataFrame(expected_extra + [np.nan] * 4)
+        tm.assert_frame_equal(result, expected)
+
+    def test_set_column_names_in_parameter(self, tmp_excel):
+        # GH 12870 : pass down column names associated with
+        # keyword argument names
+        refdf = DataFrame([[1, "foo"], [2, "bar"], [3, "baz"]], columns=["a", "b"])
+
+        with ExcelWriter(tmp_excel) as writer:
+            refdf.to_excel(writer, sheet_name="Data_no_head", header=False, index=False)
+            refdf.to_excel(writer, sheet_name="Data_with_head", index=False)
+
+        refdf.columns = ["A", "B"]
+
+        with ExcelFile(tmp_excel) as reader:
+            xlsdf_no_head = pd.read_excel(
+                reader, sheet_name="Data_no_head", header=None, names=["A", "B"]
+            )
+            xlsdf_with_head = pd.read_excel(
+                reader,
+                sheet_name="Data_with_head",
+                index_col=None,
+                names=["A", "B"],
+            )
+
+        tm.assert_frame_equal(xlsdf_no_head, refdf)
+        tm.assert_frame_equal(xlsdf_with_head, refdf)
+
+    def test_creating_and_reading_multiple_sheets(self, tmp_excel):
+        # see gh-9450
+        #
+        # Test reading multiple sheets, from a runtime
+        # created Excel file with multiple sheets.
+        def tdf(col_sheet_name):
+            d, i = [11, 22, 33], [1, 2, 3]
+            return DataFrame(d, i, columns=[col_sheet_name])
+
+        sheets = ["AAA", "BBB", "CCC"]
+
+        dfs = [tdf(s) for s in sheets]
+        dfs = dict(zip(sheets, dfs))
+
+        with ExcelWriter(tmp_excel) as ew:
+            for sheetname, df in dfs.items():
+                df.to_excel(ew, sheet_name=sheetname)
+
+        dfs_returned = pd.read_excel(tmp_excel, sheet_name=sheets, index_col=0)
+
+        for s in sheets:
+            tm.assert_frame_equal(dfs[s], dfs_returned[s])
+
+    def test_read_excel_multiindex_empty_level(self, tmp_excel):
+        # see gh-12453
+        df = DataFrame(
+            {
+                ("One", "x"): {0: 1},
+                ("Two", "X"): {0: 3},
+                ("Two", "Y"): {0: 7},
+                ("Zero", ""): {0: 0},
+            }
+        )
+
+        expected = DataFrame(
+            {
+                ("One", "x"): {0: 1},
+                ("Two", "X"): {0: 3},
+                ("Two", "Y"): {0: 7},
+                ("Zero", "Unnamed: 4_level_1"): {0: 0},
+            }
+        )
+
+        df.to_excel(tmp_excel)
+        actual = pd.read_excel(tmp_excel, header=[0, 1], index_col=0)
+        tm.assert_frame_equal(actual, expected)
+
+        df = DataFrame(
+            {
+                ("Beg", ""): {0: 0},
+                ("Middle", "x"): {0: 1},
+                ("Tail", "X"): {0: 3},
+                ("Tail", "Y"): {0: 7},
+            }
+        )
+
+        expected = DataFrame(
+            {
+                ("Beg", "Unnamed: 1_level_1"): {0: 0},
+                ("Middle", "x"): {0: 1},
+                ("Tail", "X"): {0: 3},
+                ("Tail", "Y"): {0: 7},
+            }
+        )
+
+        df.to_excel(tmp_excel)
+        actual = pd.read_excel(tmp_excel, header=[0, 1], index_col=0)
+        tm.assert_frame_equal(actual, expected)
+
+    @pytest.mark.parametrize("c_idx_names", ["a", None])
+    @pytest.mark.parametrize("r_idx_names", ["b", None])
+    @pytest.mark.parametrize("c_idx_levels", [1, 3])
+    @pytest.mark.parametrize("r_idx_levels", [1, 3])
+    def test_excel_multindex_roundtrip(
+        self,
+        tmp_excel,
+        c_idx_names,
+        r_idx_names,
+        c_idx_levels,
+        r_idx_levels,
+    ):
+        # see gh-4679
+        # Empty name case current read in as
+        # unnamed levels, not Nones.
+        check_names = bool(r_idx_names) or r_idx_levels <= 1
+
+        if c_idx_levels == 1:
+            columns = Index(list("abcde"))
+        else:
+            columns = MultiIndex.from_arrays(
+                [range(5) for _ in range(c_idx_levels)],
+                names=[f"{c_idx_names}-{i}" for i in range(c_idx_levels)],
+            )
+        if r_idx_levels == 1:
+            index = Index(list("ghijk"))
+        else:
+            index = MultiIndex.from_arrays(
+                [range(5) for _ in range(r_idx_levels)],
+                names=[f"{r_idx_names}-{i}" for i in range(r_idx_levels)],
+            )
+        df = DataFrame(
+            1.1 * np.ones((5, 5)),
+            columns=columns,
+            index=index,
+        )
+        df.to_excel(tmp_excel)
+
+        act = pd.read_excel(
+            tmp_excel,
+            index_col=list(range(r_idx_levels)),
+            header=list(range(c_idx_levels)),
+        )
+        tm.assert_frame_equal(df, act, check_names=check_names)
+
+        df.iloc[0, :] = np.nan
+        df.to_excel(tmp_excel)
+
+        act = pd.read_excel(
+            tmp_excel,
+            index_col=list(range(r_idx_levels)),
+            header=list(range(c_idx_levels)),
+        )
+        tm.assert_frame_equal(df, act, check_names=check_names)
+
+        df.iloc[-1, :] = np.nan
+        df.to_excel(tmp_excel)
+        act = pd.read_excel(
+            tmp_excel,
+            index_col=list(range(r_idx_levels)),
+            header=list(range(c_idx_levels)),
+        )
+        tm.assert_frame_equal(df, act, check_names=check_names)
+
+    def test_read_excel_parse_dates(self, tmp_excel):
+        # see gh-11544, gh-12051
+        df = DataFrame(
+            {"col": [1, 2, 3], "date_strings": date_range("2012-01-01", periods=3)}
+        )
+        df2 = df.copy()
+        df2["date_strings"] = df2["date_strings"].dt.strftime("%m/%d/%Y")
+
+        df2.to_excel(tmp_excel)
+
+        res = pd.read_excel(tmp_excel, index_col=0)
+        tm.assert_frame_equal(df2, res)
+
+        res = pd.read_excel(tmp_excel, parse_dates=["date_strings"], index_col=0)
+        expected = df[:]
+        expected["date_strings"] = expected["date_strings"].astype("M8[s]")
+        tm.assert_frame_equal(res, expected)
+
+        res = pd.read_excel(
+            tmp_excel, parse_dates=["date_strings"], date_format="%m/%d/%Y", index_col=0
+        )
+        expected["date_strings"] = expected["date_strings"].astype("M8[s]")
+        tm.assert_frame_equal(expected, res)
+
+    def test_multiindex_interval_datetimes(self, tmp_excel):
+        # GH 30986
+        midx = MultiIndex.from_arrays(
+            [
+                range(4),
+                pd.interval_range(
+                    start=pd.Timestamp("2020-01-01"), periods=4, freq="6ME"
+                ),
+            ]
+        )
+        df = DataFrame(range(4), index=midx)
+        df.to_excel(tmp_excel)
+        result = pd.read_excel(tmp_excel, index_col=[0, 1])
+        expected = DataFrame(
+            range(4),
+            MultiIndex.from_arrays(
+                [
+                    range(4),
+                    [
+                        "(2020-01-31 00:00:00, 2020-07-31 00:00:00]",
+                        "(2020-07-31 00:00:00, 2021-01-31 00:00:00]",
+                        "(2021-01-31 00:00:00, 2021-07-31 00:00:00]",
+                        "(2021-07-31 00:00:00, 2022-01-31 00:00:00]",
+                    ],
+                ]
+            ),
+            columns=Index([0]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("merge_cells", [True, False, "columns"])
+    def test_excel_round_trip_with_periodindex(self, tmp_excel, merge_cells):
+        # GH#60099
+        df = DataFrame(
+            {"A": [1, 2]},
+            index=MultiIndex.from_arrays(
+                [
+                    period_range(start="2006-10-06", end="2006-10-07", freq="D"),
+                    ["X", "Y"],
+                ],
+                names=["date", "category"],
+            ),
+        )
+        df.to_excel(tmp_excel, merge_cells=merge_cells)
+        result = pd.read_excel(tmp_excel, index_col=[0, 1])
+        expected = DataFrame(
+            {"A": [1, 2]},
+            MultiIndex.from_arrays(
+                [
+                    [
+                        pd.to_datetime("2006-10-06 00:00:00"),
+                        pd.to_datetime("2006-10-07 00:00:00"),
+                    ],
+                    ["X", "Y"],
+                ],
+                names=["date", "category"],
+            ),
+        )
+        time_format = (
+            "datetime64[s]" if tmp_excel.endswith(".ods") else "datetime64[us]"
+        )
+        expected.index = expected.index.set_levels(
+            expected.index.levels[0].astype(time_format), level=0
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "engine,ext",
+    [
+        pytest.param(
+            "openpyxl",
+            ".xlsx",
+            marks=[td.skip_if_no("openpyxl"), td.skip_if_no("xlrd")],
+        ),
+        pytest.param(
+            "openpyxl",
+            ".xlsm",
+            marks=[td.skip_if_no("openpyxl"), td.skip_if_no("xlrd")],
+        ),
+        pytest.param(
+            "xlsxwriter",
+            ".xlsx",
+            marks=[td.skip_if_no("xlsxwriter"), td.skip_if_no("xlrd")],
+        ),
+        pytest.param("odf", ".ods", marks=td.skip_if_no("odf")),
+    ],
+)
+@pytest.mark.usefixtures("set_engine")
+class TestExcelWriter:
+    def test_excel_sheet_size(self, tmp_excel):
+        # GH 26080
+        breaking_row_count = 2**20 + 1
+        breaking_col_count = 2**14 + 1
+        # purposely using two arrays to prevent memory issues while testing
+        row_arr = np.zeros(shape=(breaking_row_count, 1))
+        col_arr = np.zeros(shape=(1, breaking_col_count))
+        row_df = DataFrame(row_arr)
+        col_df = DataFrame(col_arr)
+
+        msg = "sheet is too large"
+        with pytest.raises(ValueError, match=msg):
+            row_df.to_excel(tmp_excel)
+
+        with pytest.raises(ValueError, match=msg):
+            col_df.to_excel(tmp_excel)
+
+    def test_excel_sheet_by_name_raise(self, tmp_excel):
+        gt = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=Index(list(range(10))),
+        )
+        gt.to_excel(tmp_excel)
+
+        with ExcelFile(tmp_excel) as xl:
+            df = pd.read_excel(xl, sheet_name=0, index_col=0)
+
+        tm.assert_frame_equal(gt, df)
+
+        msg = "Worksheet named '0' not found"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_excel(xl, "0")
+
+    def test_excel_writer_context_manager(self, frame, tmp_excel):
+        with ExcelWriter(tmp_excel) as writer:
+            frame.to_excel(writer, sheet_name="Data1")
+            frame2 = frame.copy()
+            frame2.columns = frame.columns[::-1]
+            frame2.to_excel(writer, sheet_name="Data2")
+
+        with ExcelFile(tmp_excel) as reader:
+            found_df = pd.read_excel(reader, sheet_name="Data1", index_col=0)
+            found_df2 = pd.read_excel(reader, sheet_name="Data2", index_col=0)
+
+            tm.assert_frame_equal(found_df, frame)
+            tm.assert_frame_equal(found_df2, frame2)
+
+    def test_roundtrip(self, frame, tmp_excel):
+        frame = frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+        # test roundtrip
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        recons = pd.read_excel(tmp_excel, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(frame, recons)
+
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+        recons = pd.read_excel(tmp_excel, sheet_name="test1", index_col=None)
+        recons.index = frame.index
+        tm.assert_frame_equal(frame, recons)
+
+        frame.to_excel(tmp_excel, sheet_name="test1", na_rep="NA")
+        recons = pd.read_excel(
+            tmp_excel, sheet_name="test1", index_col=0, na_values=["NA"]
+        )
+        tm.assert_frame_equal(frame, recons)
+
+        # GH 3611
+        frame.to_excel(tmp_excel, sheet_name="test1", na_rep="88")
+        recons = pd.read_excel(
+            tmp_excel, sheet_name="test1", index_col=0, na_values=["88"]
+        )
+        tm.assert_frame_equal(frame, recons)
+
+        frame.to_excel(tmp_excel, sheet_name="test1", na_rep="88")
+        recons = pd.read_excel(
+            tmp_excel, sheet_name="test1", index_col=0, na_values=[88, 88.0]
+        )
+        tm.assert_frame_equal(frame, recons)
+
+        # GH 6573
+        frame.to_excel(tmp_excel, sheet_name="Sheet1")
+        recons = pd.read_excel(tmp_excel, index_col=0)
+        tm.assert_frame_equal(frame, recons)
+
+        frame.to_excel(tmp_excel, sheet_name="0")
+        recons = pd.read_excel(tmp_excel, index_col=0)
+        tm.assert_frame_equal(frame, recons)
+
+        # GH 8825 Pandas Series should provide to_excel method
+        s = frame["A"]
+        s.to_excel(tmp_excel)
+        recons = pd.read_excel(tmp_excel, index_col=0)
+        tm.assert_frame_equal(s.to_frame(), recons)
+
+    def test_mixed(self, frame, tmp_excel):
+        mixed_frame = frame.copy()
+        mixed_frame["foo"] = "bar"
+
+        mixed_frame.to_excel(tmp_excel, sheet_name="test1")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(mixed_frame, recons)
+
+    def test_ts_frame(self, tmp_excel):
+        unit = get_exp_unit(tmp_excel)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+
+        # freq doesn't round-trip
+        index = pd.DatetimeIndex(np.asarray(df.index), freq=None)
+        df.index = index
+
+        expected = df[:]
+        expected.index = expected.index.as_unit(unit)
+
+        df.to_excel(tmp_excel, sheet_name="test1")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(expected, recons)
+
+    def test_basics_with_nan(self, frame, tmp_excel):
+        frame = frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+    @pytest.mark.parametrize("np_type", [np.int8, np.int16, np.int32, np.int64])
+    def test_int_types(self, np_type, tmp_excel):
+        # Test np.int values read come back as int
+        # (rather than float which is Excel's format).
+        df = DataFrame(
+            np.random.default_rng(2).integers(-10, 10, size=(10, 2)),
+            dtype=np_type,
+            index=Index(list(range(10))),
+        )
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+
+        int_frame = df.astype(np.int64)
+        tm.assert_frame_equal(int_frame, recons)
+
+        recons2 = pd.read_excel(tmp_excel, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(int_frame, recons2)
+
+    @pytest.mark.parametrize("np_type", [np.float16, np.float32, np.float64])
+    def test_float_types(self, np_type, tmp_excel):
+        # Test np.float values read come back as float.
+        df = DataFrame(
+            np.random.default_rng(2).random(10),
+            dtype=np_type,
+            index=Index(list(range(10))),
+        )
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0).astype(
+                np_type
+            )
+
+        tm.assert_frame_equal(df, recons)
+
+    def test_bool_types(self, tmp_excel):
+        # Test np.bool_ values read come back as float.
+        df = DataFrame([1, 0, True, False], dtype=np.bool_, index=Index(list(range(4))))
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0).astype(
+                np.bool_
+            )
+
+        tm.assert_frame_equal(df, recons)
+
+    def test_inf_roundtrip(self, tmp_excel):
+        df = DataFrame([(1, np.inf), (2, 3), (5, -np.inf)], index=Index(list(range(3))))
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+
+        tm.assert_frame_equal(df, recons)
+
+    def test_sheets(self, frame, tmp_excel):
+        # freq doesn't round-trip
+        unit = get_exp_unit(tmp_excel)
+        tsframe = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+
+        index = pd.DatetimeIndex(np.asarray(tsframe.index), freq=None)
+        tsframe.index = index
+
+        expected = tsframe[:]
+        expected.index = expected.index.as_unit(unit)
+
+        frame = frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+        # Test writing to separate sheets
+        with ExcelWriter(tmp_excel) as writer:
+            frame.to_excel(writer, sheet_name="test1")
+            tsframe.to_excel(writer, sheet_name="test2")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+            tm.assert_frame_equal(frame, recons)
+            recons = pd.read_excel(reader, sheet_name="test2", index_col=0)
+        tm.assert_frame_equal(expected, recons)
+        assert 2 == len(reader.sheet_names)
+        assert "test1" == reader.sheet_names[0]
+        assert "test2" == reader.sheet_names[1]
+
+    def test_colaliases(self, frame, tmp_excel):
+        frame = frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+        # column aliases
+        col_aliases = Index(["AA", "X", "Y", "Z"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=col_aliases)
+        with ExcelFile(tmp_excel) as reader:
+            rs = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        xp = frame.copy()
+        xp.columns = col_aliases
+        tm.assert_frame_equal(xp, rs)
+
+    def test_roundtrip_indexlabels(self, merge_cells, frame, tmp_excel):
+        frame = frame.copy()
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
+
+        frame.to_excel(tmp_excel, sheet_name="test1")
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+        # test index_label
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2))) >= 0
+        df.to_excel(
+            tmp_excel, sheet_name="test1", index_label=["test"], merge_cells=merge_cells
+        )
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0).astype(
+                np.int64
+            )
+        df.index.names = ["test"]
+        assert df.index.names == recons.index.names
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2))) >= 0
+        df.to_excel(
+            tmp_excel,
+            sheet_name="test1",
+            index_label=["test", "dummy", "dummy2"],
+            merge_cells=merge_cells,
+        )
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0).astype(
+                np.int64
+            )
+        df.index.names = ["test"]
+        assert df.index.names == recons.index.names
+
+        df = (
+            DataFrame(
+                np.random.default_rng(2).standard_normal((10, 2)),
+                index=Index(list(range(10))),
+            )
+            >= 0
+        )
+        df.to_excel(
+            tmp_excel, sheet_name="test1", index_label="test", merge_cells=merge_cells
+        )
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0).astype(
+                np.int64
+            )
+        df.index.names = ["test"]
+        tm.assert_frame_equal(df, recons.astype(bool))
+
+        frame.to_excel(
+            tmp_excel,
+            sheet_name="test1",
+            columns=["A", "B", "C", "D"],
+            index=False,
+            merge_cells=merge_cells,
+        )
+        # take 'A' and 'B' as indexes (same row as cols 'C', 'D')
+        df = frame.copy()
+        df = df.set_index(["A", "B"])
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=[0, 1])
+        tm.assert_frame_equal(df, recons)
+
+    def test_excel_roundtrip_indexname(self, merge_cells, tmp_excel):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        df.index.name = "foo"
+
+        df.to_excel(tmp_excel, merge_cells=merge_cells)
+
+        with ExcelFile(tmp_excel) as xf:
+            result = pd.read_excel(xf, sheet_name=xf.sheet_names[0], index_col=0)
+
+        tm.assert_frame_equal(result, df)
+        assert result.index.name == "foo"
+
+    def test_excel_roundtrip_datetime(self, merge_cells, tmp_excel):
+        # datetime.date, not sure what to test here exactly
+        unit = get_exp_unit(tmp_excel)
+
+        # freq does not round-trip
+        tsframe = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        index = pd.DatetimeIndex(np.asarray(tsframe.index), freq=None)
+        tsframe.index = index
+
+        tsf = tsframe.copy()
+
+        tsf.index = [x.date() for x in tsframe.index]
+        tsf.to_excel(tmp_excel, sheet_name="test1", merge_cells=merge_cells)
+
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+
+        expected = tsframe[:]
+        expected.index = expected.index.as_unit(unit)
+        tm.assert_frame_equal(expected, recons)
+
+    def test_excel_date_datetime_format(self, ext, tmp_excel, tmp_path):
+        # see gh-4133
+        #
+        # Excel output format strings
+        unit = get_exp_unit(tmp_excel)
+        df = DataFrame(
+            [
+                [date(2014, 1, 31), date(1999, 9, 24)],
+                [datetime(1998, 5, 26, 23, 33, 4), datetime(2014, 2, 28, 13, 5, 13)],
+            ],
+            index=["DATE", "DATETIME"],
+            columns=["X", "Y"],
+        )
+        df_expected = DataFrame(
+            [
+                [datetime(2014, 1, 31), datetime(1999, 9, 24)],
+                [datetime(1998, 5, 26, 23, 33, 4), datetime(2014, 2, 28, 13, 5, 13)],
+            ],
+            index=["DATE", "DATETIME"],
+            columns=["X", "Y"],
+        )
+        df_expected = df_expected.astype(f"M8[{unit}]")
+
+        filename2 = tmp_path / f"tmp2{ext}"
+        filename2.touch()
+        with ExcelWriter(tmp_excel) as writer1:
+            df.to_excel(writer1, sheet_name="test1")
+
+        with ExcelWriter(
+            filename2,
+            date_format="DD.MM.YYYY",
+            datetime_format="DD.MM.YYYY HH-MM-SS",
+        ) as writer2:
+            df.to_excel(writer2, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader1:
+            rs1 = pd.read_excel(reader1, sheet_name="test1", index_col=0)
+
+        with ExcelFile(filename2) as reader2:
+            rs2 = pd.read_excel(reader2, sheet_name="test1", index_col=0)
+
+        # TODO: why do we get different units?
+        rs2 = rs2.astype(f"M8[{unit}]")
+
+        tm.assert_frame_equal(rs1, rs2)
+
+        # Since the reader returns a datetime object for dates,
+        # we need to use df_expected to check the result.
+        tm.assert_frame_equal(rs2, df_expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:invalid value encountered in cast:RuntimeWarning"
+    )
+    def test_to_excel_interval_no_labels(self, tmp_excel, using_infer_string):
+        # see gh-19242
+        #
+        # Test writing Interval without labels.
+        df = DataFrame(
+            np.random.default_rng(2).integers(-10, 10, size=(20, 1)), dtype=np.int64
+        )
+        expected = df.copy()
+
+        df["new"] = pd.cut(df[0], 10)
+        expected["new"] = pd.cut(expected[0], 10).astype(
+            str if not using_infer_string else "str"
+        )
+
+        df.to_excel(tmp_excel, sheet_name="test1")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(expected, recons)
+
+    def test_to_excel_interval_labels(self, tmp_excel):
+        # see gh-19242
+        #
+        # Test writing Interval with labels.
+        df = DataFrame(
+            np.random.default_rng(2).integers(-10, 10, size=(20, 1)), dtype=np.int64
+        )
+        expected = df.copy()
+        intervals = pd.cut(
+            df[0], 10, labels=["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"]
+        )
+        df["new"] = intervals
+        expected["new"] = pd.Series(list(intervals))
+
+        df.to_excel(tmp_excel, sheet_name="test1")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(expected, recons)
+
+    def test_to_excel_timedelta(self, tmp_excel):
+        # see gh-19242, gh-9155
+        #
+        # Test writing timedelta to xls.
+        df = DataFrame(
+            np.random.default_rng(2).integers(-10, 10, size=(20, 1)),
+            columns=["A"],
+            dtype=np.int64,
+        )
+        expected = df.copy()
+
+        df["new"] = df["A"].apply(lambda x: timedelta(seconds=x))
+        expected["new"] = expected["A"].apply(
+            lambda x: timedelta(seconds=x).total_seconds() / 86400
+        )
+
+        df.to_excel(tmp_excel, sheet_name="test1")
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(expected, recons)
+
+    def test_to_excel_periodindex(self, tmp_excel):
+        # xp has a PeriodIndex
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        xp = df.resample("ME").mean().to_period("M")
+
+        xp.to_excel(tmp_excel, sheet_name="sht1")
+
+        with ExcelFile(tmp_excel) as reader:
+            rs = pd.read_excel(reader, sheet_name="sht1", index_col=0)
+        tm.assert_frame_equal(xp, rs.to_period("M"))
+
+    def test_to_excel_multiindex(self, merge_cells, frame, tmp_excel):
+        arrays = np.arange(len(frame.index) * 2, dtype=np.int64).reshape(2, -1)
+        new_index = MultiIndex.from_arrays(arrays, names=["first", "second"])
+        frame.index = new_index
+
+        frame.to_excel(tmp_excel, sheet_name="test1", header=False)
+        frame.to_excel(tmp_excel, sheet_name="test1", columns=["A", "B"])
+
+        # round trip
+        frame.to_excel(tmp_excel, sheet_name="test1", merge_cells=merge_cells)
+        with ExcelFile(tmp_excel) as reader:
+            df = pd.read_excel(reader, sheet_name="test1", index_col=[0, 1])
+        tm.assert_frame_equal(frame, df)
+
+    # GH13511
+    def test_to_excel_multiindex_nan_label(self, merge_cells, tmp_excel):
+        df = DataFrame(
+            {
+                "A": [None, 2, 3],
+                "B": [10, 20, 30],
+                "C": np.random.default_rng(2).random(3),
+            }
+        )
+        df = df.set_index(["A", "B"])
+
+        df.to_excel(tmp_excel, merge_cells=merge_cells)
+        df1 = pd.read_excel(tmp_excel, index_col=[0, 1])
+        tm.assert_frame_equal(df, df1)
+
+    # Test for Issue 11328. If column indices are integers, make
+    # sure they are handled correctly for either setting of
+    # merge_cells
+    def test_to_excel_multiindex_cols(self, merge_cells, tmp_excel):
+        # GH#11328
+        frame = DataFrame(
+            {
+                "A": [1, 2, 3],
+                "B": [4, 5, 6],
+                "C": [7, 8, 9],
+            }
+        )
+        arrays = np.arange(len(frame.index) * 2, dtype=np.int64).reshape(2, -1)
+        new_index = MultiIndex.from_arrays(arrays, names=["first", "second"])
+        frame.index = new_index
+
+        new_cols_index = MultiIndex.from_tuples([(40, 1), (40, 2), (50, 1)])
+        frame.columns = new_cols_index
+        frame.to_excel(tmp_excel, sheet_name="test1", merge_cells=merge_cells)
+
+        # Check round trip
+        with ExcelFile(tmp_excel) as reader:
+            result = pd.read_excel(
+                reader, sheet_name="test1", header=[0, 1], index_col=[0, 1]
+            )
+        tm.assert_frame_equal(result, frame)
+
+        # GH#60274
+        # Check with header/index_col None to determine which cells were merged
+        with ExcelFile(tmp_excel) as reader:
+            result = pd.read_excel(
+                reader, sheet_name="test1", header=None, index_col=None
+            )
+        expected = DataFrame(
+            {
+                0: [np.nan, np.nan, "first", 0, 1, 2],
+                1: [np.nan, np.nan, "second", 3, 4, 5],
+                2: [40.0, 1.0, np.nan, 1.0, 2.0, 3.0],
+                3: [np.nan, 2.0, np.nan, 4.0, 5.0, 6.0],
+                4: [50.0, 1.0, np.nan, 7.0, 8.0, 9.0],
+            }
+        )
+        if not merge_cells:
+            # MultiIndex column value is repeated
+            expected.loc[0, 3] = 40.0
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_excel_multiindex_dates(self, merge_cells, tmp_excel):
+        # try multiindex with dates
+        unit = get_exp_unit(tmp_excel)
+        tsframe = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        tsframe.index = MultiIndex.from_arrays(
+            [
+                tsframe.index.as_unit(unit),
+                np.arange(len(tsframe.index), dtype=np.int64),
+            ],
+            names=["time", "foo"],
+        )
+
+        tsframe.to_excel(tmp_excel, sheet_name="test1", merge_cells=merge_cells)
+        with ExcelFile(tmp_excel) as reader:
+            recons = pd.read_excel(reader, sheet_name="test1", index_col=[0, 1])
+
+        tm.assert_frame_equal(tsframe, recons)
+        assert recons.index.names == ("time", "foo")
+
+    def test_to_excel_multiindex_no_write_index(self, tmp_excel):
+        # Test writing and re-reading a MI without the index. GH 5616.
+
+        # Initial non-MI frame.
+        frame1 = DataFrame({"a": [10, 20], "b": [30, 40], "c": [50, 60]})
+
+        # Add a MI.
+        frame2 = frame1.copy()
+        multi_index = MultiIndex.from_tuples([(70, 80), (90, 100)])
+        frame2.index = multi_index
+
+        # Write out to Excel without the index.
+        frame2.to_excel(tmp_excel, sheet_name="test1", index=False)
+
+        # Read it back in.
+        with ExcelFile(tmp_excel) as reader:
+            frame3 = pd.read_excel(reader, sheet_name="test1")
+
+        # Test that it is the same as the initial frame.
+        tm.assert_frame_equal(frame1, frame3)
+
+    def test_to_excel_empty_multiindex(self, tmp_excel):
+        # GH 19543.
+        expected = DataFrame([], columns=[0, 1, 2])
+
+        df = DataFrame([], index=MultiIndex.from_tuples([], names=[0, 1]), columns=[2])
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        with ExcelFile(tmp_excel) as reader:
+            result = pd.read_excel(reader, sheet_name="test1")
+        tm.assert_frame_equal(
+            result, expected, check_index_type=False, check_dtype=False
+        )
+
+    def test_to_excel_empty_multiindex_both_axes(self, tmp_excel):
+        # GH 57696
+        df = DataFrame(
+            [],
+            index=MultiIndex.from_tuples([], names=[0, 1]),
+            columns=MultiIndex.from_tuples([("A", "B")]),
+        )
+        df.to_excel(tmp_excel)
+        result = pd.read_excel(tmp_excel, header=[0, 1], index_col=[0, 1])
+        tm.assert_frame_equal(result, df)
+
+    def test_to_excel_float_format(self, tmp_excel):
+        df = DataFrame(
+            [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        df.to_excel(tmp_excel, sheet_name="test1", float_format="%.2f")
+
+        with ExcelFile(tmp_excel) as reader:
+            result = pd.read_excel(reader, sheet_name="test1", index_col=0)
+
+        expected = DataFrame(
+            [[0.12, 0.23, 0.57], [12.32, 123123.20, 321321.20]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_excel_datatypes_preserved(self, tmp_excel):
+        # Test that when writing and reading Excel with dtype=object,
+        # datatypes are preserved, except Decimals which should be
+        # stored as floats
+
+        # see gh-49598
+        df = DataFrame(
+            [
+                [1.23, "1.23", Decimal("1.23")],
+                [4.56, "4.56", Decimal("4.56")],
+            ],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        df.to_excel(tmp_excel)
+
+        with ExcelFile(tmp_excel) as reader:
+            result = pd.read_excel(reader, index_col=0, dtype=object)
+
+        expected = DataFrame(
+            [
+                [1.23, "1.23", 1.23],
+                [4.56, "4.56", 4.56],
+            ],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+            dtype=object,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_excel_output_encoding(self, tmp_excel):
+        # Avoid mixed inferred_type.
+        df = DataFrame(
+            [["\u0192", "\u0193", "\u0194"], ["\u0195", "\u0196", "\u0197"]],
+            index=["A\u0192", "B"],
+            columns=["X\u0193", "Y", "Z"],
+        )
+
+        df.to_excel(tmp_excel, sheet_name="TestSheet")
+        result = pd.read_excel(tmp_excel, sheet_name="TestSheet", index_col=0)
+        tm.assert_frame_equal(result, df)
+
+    def test_to_excel_unicode_filename(self, ext, tmp_path):
+        filename = tmp_path / f"\u0192u.{ext}"
+        filename.touch()
+        df = DataFrame(
+            [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        df.to_excel(filename, sheet_name="test1", float_format="%.2f")
+
+        with ExcelFile(filename) as reader:
+            result = pd.read_excel(reader, sheet_name="test1", index_col=0)
+
+        expected = DataFrame(
+            [[0.12, 0.23, 0.57], [12.32, 123123.20, 321321.20]],
+            index=["A", "B"],
+            columns=["X", "Y", "Z"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("use_headers", [True, False])
+    @pytest.mark.parametrize("r_idx_nlevels", [1, 2, 3])
+    @pytest.mark.parametrize("c_idx_nlevels", [1, 2, 3])
+    def test_excel_010_hemstring(
+        self, merge_cells, c_idx_nlevels, r_idx_nlevels, use_headers, tmp_excel
+    ):
+        def roundtrip(data, header=True, parser_hdr=0, index=True):
+            data.to_excel(
+                tmp_excel, header=header, merge_cells=merge_cells, index=index
+            )
+
+            with ExcelFile(tmp_excel) as xf:
+                return pd.read_excel(
+                    xf, sheet_name=xf.sheet_names[0], header=parser_hdr
+                )
+
+        # Basic test.
+        parser_header = 0 if use_headers else None
+        res = roundtrip(DataFrame([0]), use_headers, parser_header)
+
+        assert res.shape == (1, 2)
+        assert res.iloc[0, 0] is not np.nan
+
+        # More complex tests with multi-index.
+        nrows = 5
+        ncols = 3
+
+        # ensure limited functionality in 0.10
+        # override of gh-2370 until sorted out in 0.11
+
+        if c_idx_nlevels == 1:
+            columns = Index([f"a-{i}" for i in range(ncols)], dtype=object)
+        else:
+            columns = MultiIndex.from_arrays(
+                [range(ncols) for _ in range(c_idx_nlevels)],
+                names=[f"i-{i}" for i in range(c_idx_nlevels)],
+            )
+        if r_idx_nlevels == 1:
+            index = Index([f"b-{i}" for i in range(nrows)], dtype=object)
+        else:
+            index = MultiIndex.from_arrays(
+                [range(nrows) for _ in range(r_idx_nlevels)],
+                names=[f"j-{i}" for i in range(r_idx_nlevels)],
+            )
+
+        df = DataFrame(
+            np.ones((nrows, ncols)),
+            columns=columns,
+            index=index,
+        )
+
+        # This if will be removed once multi-column Excel writing
+        # is implemented. For now fixing gh-9794.
+        if c_idx_nlevels > 1:
+            msg = (
+                "Writing to Excel with MultiIndex columns and no index "
+                "\\('index'=False\\) is not yet implemented."
+            )
+            with pytest.raises(NotImplementedError, match=msg):
+                roundtrip(df, use_headers, index=False)
+        else:
+            res = roundtrip(df, use_headers)
+
+            if use_headers:
+                assert res.shape == (nrows, ncols + r_idx_nlevels)
+            else:
+                # First row taken as columns.
+                assert res.shape == (nrows - 1, ncols + r_idx_nlevels)
+
+            # No NaNs.
+            for r in range(len(res.index)):
+                for c in range(len(res.columns)):
+                    assert res.iloc[r, c] is not np.nan
+
+    def test_duplicated_columns(self, tmp_excel):
+        # see gh-5235
+        df = DataFrame([[1, 2, 3], [1, 2, 3], [1, 2, 3]], columns=["A", "B", "B"])
+        df.to_excel(tmp_excel, sheet_name="test1")
+        expected = DataFrame(
+            [[1, 2, 3], [1, 2, 3], [1, 2, 3]], columns=["A", "B", "B.1"]
+        )
+
+        # By default, we mangle.
+        result = pd.read_excel(tmp_excel, sheet_name="test1", index_col=0)
+        tm.assert_frame_equal(result, expected)
+
+        # see gh-11007, gh-10970
+        df = DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]], columns=["A", "B", "A", "B"])
+        df.to_excel(tmp_excel, sheet_name="test1")
+
+        result = pd.read_excel(tmp_excel, sheet_name="test1", index_col=0)
+        expected = DataFrame(
+            [[1, 2, 3, 4], [5, 6, 7, 8]], columns=["A", "B", "A.1", "B.1"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # see gh-10982
+        df.to_excel(tmp_excel, sheet_name="test1", index=False, header=False)
+        result = pd.read_excel(tmp_excel, sheet_name="test1", header=None)
+
+        expected = DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_swapped_columns(self, tmp_excel):
+        # Test for issue #5427.
+        write_frame = DataFrame({"A": [1, 1, 1], "B": [2, 2, 2]})
+        write_frame.to_excel(tmp_excel, sheet_name="test1", columns=["B", "A"])
+
+        read_frame = pd.read_excel(tmp_excel, sheet_name="test1", header=0)
+
+        tm.assert_series_equal(write_frame["A"], read_frame["A"])
+        tm.assert_series_equal(write_frame["B"], read_frame["B"])
+
+    def test_invalid_columns(self, tmp_excel):
+        # see gh-10982
+        write_frame = DataFrame({"A": [1, 1, 1], "B": [2, 2, 2]})
+
+        with pytest.raises(KeyError, match="Not all names specified"):
+            write_frame.to_excel(tmp_excel, sheet_name="test1", columns=["B", "C"])
+
+        with pytest.raises(
+            KeyError, match="'passes columns are not ALL present dataframe'"
+        ):
+            write_frame.to_excel(tmp_excel, sheet_name="test1", columns=["C", "D"])
+
+    @pytest.mark.parametrize(
+        "to_excel_index,read_excel_index_col",
+        [
+            (True, 0),  # Include index in write to file
+            (False, None),  # Dont include index in write to file
+        ],
+    )
+    def test_write_subset_columns(
+        self, tmp_excel, to_excel_index, read_excel_index_col
+    ):
+        # GH 31677
+        write_frame = DataFrame({"A": [1, 1, 1], "B": [2, 2, 2], "C": [3, 3, 3]})
+        write_frame.to_excel(
+            tmp_excel,
+            sheet_name="col_subset_bug",
+            columns=["A", "B"],
+            index=to_excel_index,
+        )
+
+        expected = write_frame[["A", "B"]]
+        read_frame = pd.read_excel(
+            tmp_excel, sheet_name="col_subset_bug", index_col=read_excel_index_col
+        )
+
+        tm.assert_frame_equal(expected, read_frame)
+
+    def test_comment_arg(self, tmp_excel):
+        # see gh-18735
+        #
+        # Test the comment argument functionality to pd.read_excel.
+
+        # Create file to read in.
+        df = DataFrame({"A": ["one", "#one", "one"], "B": ["two", "two", "#two"]})
+        df.to_excel(tmp_excel, sheet_name="test_c")
+
+        # Read file without comment arg.
+        result1 = pd.read_excel(tmp_excel, sheet_name="test_c", index_col=0)
+
+        result1.iloc[1, 0] = None
+        result1.iloc[1, 1] = None
+        result1.iloc[2, 1] = None
+
+        result2 = pd.read_excel(
+            tmp_excel, sheet_name="test_c", comment="#", index_col=0
+        )
+        tm.assert_frame_equal(result1, result2)
+
+    def test_comment_default(self, tmp_excel):
+        # Re issue #18735
+        # Test the comment argument default to pd.read_excel
+
+        # Create file to read in
+        df = DataFrame({"A": ["one", "#one", "one"], "B": ["two", "two", "#two"]})
+        df.to_excel(tmp_excel, sheet_name="test_c")
+
+        # Read file with default and explicit comment=None
+        result1 = pd.read_excel(tmp_excel, sheet_name="test_c")
+        result2 = pd.read_excel(tmp_excel, sheet_name="test_c", comment=None)
+        tm.assert_frame_equal(result1, result2)
+
+    def test_comment_used(self, tmp_excel):
+        # see gh-18735
+        #
+        # Test the comment argument is working as expected when used.
+
+        # Create file to read in.
+        df = DataFrame({"A": ["one", "#one", "one"], "B": ["two", "two", "#two"]})
+        df.to_excel(tmp_excel, sheet_name="test_c")
+
+        # Test read_frame_comment against manually produced expected output.
+        expected = DataFrame({"A": ["one", None, "one"], "B": ["two", None, None]})
+        result = pd.read_excel(tmp_excel, sheet_name="test_c", comment="#", index_col=0)
+        tm.assert_frame_equal(result, expected)
+
+    def test_comment_empty_line(self, tmp_excel):
+        # Re issue #18735
+        # Test that pd.read_excel ignores commented lines at the end of file
+
+        df = DataFrame({"a": ["1", "#2"], "b": ["2", "3"]})
+        df.to_excel(tmp_excel, index=False)
+
+        # Test that all-comment lines at EoF are ignored
+        expected = DataFrame({"a": [1], "b": [2]})
+        result = pd.read_excel(tmp_excel, comment="#")
+        tm.assert_frame_equal(result, expected)
+
+    def test_datetimes(self, tmp_excel):
+        # Test writing and reading datetimes. For issue #9139. (xref #9185)
+        unit = get_exp_unit(tmp_excel)
+        datetimes = [
+            datetime(2013, 1, 13, 1, 2, 3),
+            datetime(2013, 1, 13, 2, 45, 56),
+            datetime(2013, 1, 13, 4, 29, 49),
+            datetime(2013, 1, 13, 6, 13, 42),
+            datetime(2013, 1, 13, 7, 57, 35),
+            datetime(2013, 1, 13, 9, 41, 28),
+            datetime(2013, 1, 13, 11, 25, 21),
+            datetime(2013, 1, 13, 13, 9, 14),
+            datetime(2013, 1, 13, 14, 53, 7),
+            datetime(2013, 1, 13, 16, 37, 0),
+            datetime(2013, 1, 13, 18, 20, 52),
+        ]
+
+        write_frame = DataFrame({"A": datetimes})
+        write_frame.to_excel(tmp_excel, sheet_name="Sheet1")
+        read_frame = pd.read_excel(tmp_excel, sheet_name="Sheet1", header=0)
+
+        expected = write_frame.astype(f"M8[{unit}]")
+        tm.assert_series_equal(expected["A"], read_frame["A"])
+
+    def test_bytes_io(self, engine):
+        # see gh-7074
+        with BytesIO() as bio:
+            df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+
+            # Pass engine explicitly, as there is no file path to infer from.
+            with ExcelWriter(bio, engine=engine) as writer:
+                df.to_excel(writer)
+
+            bio.seek(0)
+            reread_df = pd.read_excel(bio, index_col=0)
+            tm.assert_frame_equal(df, reread_df)
+
+    def test_engine_kwargs(self, engine, tmp_excel):
+        # GH#52368
+        df = DataFrame([{"A": 1, "B": 2}, {"A": 3, "B": 4}])
+
+        msgs = {
+            "odf": r"OpenDocumentSpreadsheet() got an unexpected keyword "
+            r"argument 'foo'",
+            "openpyxl": r"__init__() got an unexpected keyword argument 'foo'",
+            "xlsxwriter": r"__init__() got an unexpected keyword argument 'foo'",
+        }
+
+        msgs["openpyxl"] = (
+            "Workbook.__init__() got an unexpected keyword argument 'foo'"
+        )
+        msgs["xlsxwriter"] = (
+            "Workbook.__init__() got an unexpected keyword argument 'foo'"
+        )
+
+        # Handle change in error message for openpyxl (write and append mode)
+        if engine == "openpyxl" and not os.path.exists(tmp_excel):
+            msgs["openpyxl"] = (
+                r"load_workbook() got an unexpected keyword argument 'foo'"
+            )
+
+        with pytest.raises(TypeError, match=re.escape(msgs[engine])):
+            df.to_excel(
+                tmp_excel,
+                engine=engine,
+                engine_kwargs={"foo": "bar"},
+            )
+
+    def test_write_lists_dict(self, tmp_excel):
+        # see gh-8188.
+        df = DataFrame(
+            {
+                "mixed": ["a", ["b", "c"], {"d": "e", "f": 2}],
+                "numeric": [1, 2, 3.0],
+                "str": ["apple", "banana", "cherry"],
+            }
+        )
+        df.to_excel(tmp_excel, sheet_name="Sheet1")
+        read = pd.read_excel(tmp_excel, sheet_name="Sheet1", header=0, index_col=0)
+
+        expected = df.copy()
+        expected.mixed = expected.mixed.apply(str)
+        expected.numeric = expected.numeric.astype("int64")
+
+        tm.assert_frame_equal(read, expected)
+
+    def test_render_as_column_name(self, tmp_excel):
+        # see gh-34331
+        df = DataFrame({"render": [1, 2], "data": [3, 4]})
+        df.to_excel(tmp_excel, sheet_name="Sheet1")
+        read = pd.read_excel(tmp_excel, "Sheet1", index_col=0)
+        expected = df
+        tm.assert_frame_equal(read, expected)
+
+    def test_true_and_false_value_options(self, tmp_excel):
+        # see gh-13347
+        df = DataFrame([["foo", "bar"]], columns=["col1", "col2"], dtype=object)
+        with option_context("future.no_silent_downcasting", True):
+            expected = df.replace({"foo": True, "bar": False}).astype("bool")
+
+        df.to_excel(tmp_excel)
+        read_frame = pd.read_excel(
+            tmp_excel, true_values=["foo"], false_values=["bar"], index_col=0
+        )
+        tm.assert_frame_equal(read_frame, expected)
+
+    def test_freeze_panes(self, tmp_excel):
+        # see gh-15160
+        expected = DataFrame([[1, 2], [3, 4]], columns=["col1", "col2"])
+        expected.to_excel(tmp_excel, sheet_name="Sheet1", freeze_panes=(1, 1))
+
+        result = pd.read_excel(tmp_excel, index_col=0)
+        tm.assert_frame_equal(result, expected)
+
+    def test_path_path_lib(self, engine, ext):
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        writer = partial(df.to_excel, engine=engine)
+
+        reader = partial(pd.read_excel, index_col=0)
+        result = tm.round_trip_pathlib(writer, reader, path=f"foo{ext}")
+        tm.assert_frame_equal(result, df)
+
+    def test_merged_cell_custom_objects(self, tmp_excel):
+        # see GH-27006
+        mi = MultiIndex.from_tuples(
+            [
+                (pd.Period("2018"), pd.Period("2018Q1")),
+                (pd.Period("2018"), pd.Period("2018Q2")),
+            ]
+        )
+        expected = DataFrame(np.ones((2, 2), dtype="int64"), columns=mi)
+        expected.to_excel(tmp_excel)
+        result = pd.read_excel(tmp_excel, header=[0, 1], index_col=0)
+        # need to convert PeriodIndexes to standard Indexes for assert equal
+        expected.columns = expected.columns.set_levels(
+            [[str(i) for i in mi.levels[0]], [str(i) for i in mi.levels[1]]],
+            level=[0, 1],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [None, object])
+    def test_raise_when_saving_timezones(self, dtype, tz_aware_fixture, tmp_excel):
+        # GH 27008, GH 7056
+        tz = tz_aware_fixture
+        data = pd.Timestamp("2019", tz=tz)
+        df = DataFrame([data], dtype=dtype)
+        with pytest.raises(ValueError, match="Excel does not support"):
+            df.to_excel(tmp_excel)
+
+        data = data.to_pydatetime()
+        df = DataFrame([data], dtype=dtype)
+        with pytest.raises(ValueError, match="Excel does not support"):
+            df.to_excel(tmp_excel)
+
+    def test_excel_duplicate_columns_with_names(self, tmp_excel):
+        # GH#39695
+        df = DataFrame({"A": [0, 1], "B": [10, 11]})
+        df.to_excel(tmp_excel, columns=["A", "B", "A"], index=False)
+
+        result = pd.read_excel(tmp_excel)
+        expected = DataFrame([[0, 10, 0], [1, 11, 1]], columns=["A", "B", "A.1"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_if_sheet_exists_raises(self, tmp_excel):
+        # GH 40230
+        msg = "if_sheet_exists is only valid in append mode (mode='a')"
+
+        with pytest.raises(ValueError, match=re.escape(msg)):
+            ExcelWriter(tmp_excel, if_sheet_exists="replace")
+
+    def test_excel_writer_empty_frame(self, engine, tmp_excel):
+        # GH#45793
+        with ExcelWriter(tmp_excel, engine=engine) as writer:
+            DataFrame().to_excel(writer)
+        result = pd.read_excel(tmp_excel)
+        expected = DataFrame()
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_excel_empty_frame(self, engine, tmp_excel):
+        # GH#45793
+        DataFrame().to_excel(tmp_excel, engine=engine)
+        result = pd.read_excel(tmp_excel)
+        expected = DataFrame()
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_excel_raising_warning_when_cell_character_exceed_limit(self):
+        # GH#56954
+        df = DataFrame({"A": ["a" * 32768]})
+        msg = r"Cell contents too long \(32768\), truncated to 32767 characters"
+        with tm.assert_produces_warning(
+            UserWarning, match=msg, raise_on_extra_warnings=False
+        ):
+            buf = BytesIO()
+            df.to_excel(buf)
+
+
+class TestExcelWriterEngineTests:
+    @pytest.mark.parametrize(
+        "klass,ext",
+        [
+            pytest.param(_XlsxWriter, ".xlsx", marks=td.skip_if_no("xlsxwriter")),
+            pytest.param(_OpenpyxlWriter, ".xlsx", marks=td.skip_if_no("openpyxl")),
+        ],
+    )
+    def test_ExcelWriter_dispatch(self, klass, ext, tmp_excel):
+        with ExcelWriter(tmp_excel) as writer:
+            if ext == ".xlsx" and bool(
+                import_optional_dependency("xlsxwriter", errors="ignore")
+            ):
+                # xlsxwriter has preference over openpyxl if both installed
+                assert isinstance(writer, _XlsxWriter)
+            else:
+                assert isinstance(writer, klass)
+
+    def test_ExcelWriter_dispatch_raises(self):
+        with pytest.raises(ValueError, match="No engine"):
+            ExcelWriter("nothing")
+
+    def test_register_writer(self, tmp_path):
+        class DummyClass(ExcelWriter):
+            called_save = False
+            called_write_cells = False
+            called_sheets = False
+            _supported_extensions = ("xlsx", "xls")
+            _engine = "dummy"
+
+            def book(self):
+                pass
+
+            def _save(self):
+                type(self).called_save = True
+
+            def _write_cells(self, *args, **kwargs):
+                type(self).called_write_cells = True
+
+            @property
+            def sheets(self):
+                type(self).called_sheets = True
+
+            @classmethod
+            def assert_called_and_reset(cls):
+                assert cls.called_save
+                assert cls.called_write_cells
+                assert not cls.called_sheets
+                cls.called_save = False
+                cls.called_write_cells = False
+
+        register_writer(DummyClass)
+
+        with option_context("io.excel.xlsx.writer", "dummy"):
+            filepath = tmp_path / "something.xlsx"
+            filepath.touch()
+            with ExcelWriter(filepath) as writer:
+                assert isinstance(writer, DummyClass)
+            df = DataFrame(
+                ["a"],
+                columns=Index(["b"], name="foo"),
+                index=Index(["c"], name="bar"),
+            )
+            df.to_excel(filepath)
+            DummyClass.assert_called_and_reset()
+
+        filepath2 = tmp_path / "something2.xlsx"
+        filepath2.touch()
+        df.to_excel(filepath2, engine="dummy")
+        DummyClass.assert_called_and_reset()
+
+
+@td.skip_if_no("xlrd")
+@td.skip_if_no("openpyxl")
+class TestFSPath:
+    def test_excelfile_fspath(self, tmp_path):
+        path = tmp_path / "foo.xlsx"
+        path.touch()
+        df = DataFrame({"A": [1, 2]})
+        df.to_excel(path)
+        with ExcelFile(path) as xl:
+            result = os.fspath(xl)
+        assert result == str(path)
+
+    def test_excelwriter_fspath(self, tmp_path):
+        path = tmp_path / "foo.xlsx"
+        path.touch()
+        with ExcelWriter(path) as writer:
+            assert os.fspath(writer) == str(path)
+
+
+@pytest.mark.parametrize("klass", _writers.values())
+def test_subclass_attr(klass):
+    # testing that subclasses of ExcelWriter don't have public attributes (issue 49602)
+    attrs_base = {name for name in dir(ExcelWriter) if not name.startswith("_")}
+    attrs_klass = {name for name in dir(klass) if not name.startswith("_")}
+    assert not attrs_base.symmetric_difference(attrs_klass)
diff --git a/pandas/pandas/tests/io/excel/test_xlrd.py b/pandas/pandas/tests/io/excel/test_xlrd.py
new file mode 100644
index 0000000000000..c49cbaf7fb26c
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_xlrd.py
@@ -0,0 +1,71 @@
+import io
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+from pandas.io.excel import ExcelFile
+from pandas.io.excel._base import inspect_excel_format
+
+xlrd = pytest.importorskip("xlrd")
+
+
+@pytest.fixture
+def read_ext_xlrd():
+    """
+    Valid extensions for reading Excel files with xlrd.
+
+    Similar to read_ext, but excludes .ods, .xlsb, and for xlrd>2 .xlsx, .xlsm
+    """
+    return ".xls"
+
+
+def test_read_xlrd_book(read_ext_xlrd, datapath):
+    engine = "xlrd"
+    sheet_name = "Sheet1"
+    pth = datapath("io", "data", "excel", "test1.xls")
+    with xlrd.open_workbook(pth) as book:
+        with ExcelFile(book, engine=engine) as xl:
+            result = pd.read_excel(xl, sheet_name=sheet_name, index_col=0)
+
+        expected = pd.read_excel(
+            book, sheet_name=sheet_name, engine=engine, index_col=0
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_xlsx_fails(datapath):
+    # GH 29375
+    from xlrd.biffh import XLRDError
+
+    path = datapath("io", "data", "excel", "test1.xlsx")
+    with pytest.raises(XLRDError, match="Excel xlsx file; not supported"):
+        pd.read_excel(path, engine="xlrd")
+
+
+def test_nan_in_xls(datapath):
+    # GH 54564
+    path = datapath("io", "data", "excel", "test6.xls")
+
+    expected = pd.DataFrame({0: np.r_[0, 2].astype("int64"), 1: np.r_[1, np.nan]})
+
+    result = pd.read_excel(path, header=None)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "file_header",
+    [
+        b"\x09\x00\x04\x00\x07\x00\x10\x00",
+        b"\x09\x02\x06\x00\x00\x00\x10\x00",
+        b"\x09\x04\x06\x00\x00\x00\x10\x00",
+        b"\xd0\xcf\x11\xe0\xa1\xb1\x1a\xe1",
+    ],
+)
+def test_read_old_xls_files(file_header):
+    # GH 41226
+    f = io.BytesIO(file_header)
+    assert inspect_excel_format(f) == "xls"
diff --git a/pandas/pandas/tests/io/excel/test_xlsxwriter.py b/pandas/pandas/tests/io/excel/test_xlsxwriter.py
new file mode 100644
index 0000000000000..b2e6c845e5019
--- /dev/null
+++ b/pandas/pandas/tests/io/excel/test_xlsxwriter.py
@@ -0,0 +1,86 @@
+import contextlib
+import uuid
+
+import pytest
+
+from pandas import DataFrame
+
+from pandas.io.excel import ExcelWriter
+
+xlsxwriter = pytest.importorskip("xlsxwriter")
+
+
+@pytest.fixture
+def ext():
+    return ".xlsx"
+
+
+@pytest.fixture
+def tmp_excel(ext, tmp_path):
+    tmp = tmp_path / f"{uuid.uuid4()}{ext}"
+    tmp.touch()
+    return str(tmp)
+
+
+def test_column_format(tmp_excel):
+    # Test that column formats are applied to cells. Test for issue #9167.
+    # Applicable to xlsxwriter only.
+    openpyxl = pytest.importorskip("openpyxl")
+
+    frame = DataFrame({"A": [123456, 123456], "B": [123456, 123456]})
+
+    with ExcelWriter(tmp_excel) as writer:
+        frame.to_excel(writer)
+
+        # Add a number format to col B and ensure it is applied to cells.
+        num_format = "#,##0"
+        write_workbook = writer.book
+        write_worksheet = write_workbook.worksheets()[0]
+        col_format = write_workbook.add_format({"num_format": num_format})
+        write_worksheet.set_column("B:B", None, col_format)
+
+    with contextlib.closing(openpyxl.load_workbook(tmp_excel)) as read_workbook:
+        try:
+            read_worksheet = read_workbook["Sheet1"]
+        except TypeError:
+            # compat
+            read_worksheet = read_workbook.get_sheet_by_name(name="Sheet1")
+
+    # Get the number format from the cell.
+    try:
+        cell = read_worksheet["B2"]
+    except TypeError:
+        # compat
+        cell = read_worksheet.cell("B2")
+
+    try:
+        read_num_format = cell.number_format
+    except AttributeError:
+        read_num_format = cell.style.number_format._format_code
+
+    assert read_num_format == num_format
+
+
+def test_write_append_mode_raises(tmp_excel):
+    msg = "Append mode is not supported with xlsxwriter!"
+
+    with pytest.raises(ValueError, match=msg):
+        ExcelWriter(tmp_excel, engine="xlsxwriter", mode="a")
+
+
+@pytest.mark.parametrize("nan_inf_to_errors", [True, False])
+def test_engine_kwargs(tmp_excel, nan_inf_to_errors):
+    # GH 42286
+    engine_kwargs = {"options": {"nan_inf_to_errors": nan_inf_to_errors}}
+    with ExcelWriter(
+        tmp_excel, engine="xlsxwriter", engine_kwargs=engine_kwargs
+    ) as writer:
+        assert writer.book.nan_inf_to_errors == nan_inf_to_errors
+
+
+def test_book_and_sheets_consistent(tmp_excel):
+    # GH#45687 - Ensure sheets is updated if user modifies book
+    with ExcelWriter(tmp_excel, engine="xlsxwriter") as writer:
+        assert writer.sheets == {}
+        sheet = writer.book.add_worksheet("test_name")
+        assert writer.sheets == {"test_name": sheet}
diff --git a/pandas/pandas/tests/io/formats/__init__.py b/pandas/pandas/tests/io/formats/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/formats/data/html/datetime64_hourformatter.html b/pandas/pandas/tests/io/formats/data/html/datetime64_hourformatter.html
new file mode 100644
index 0000000000000..c92b7218eba76
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/datetime64_hourformatter.html
@@ -0,0 +1,18 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>hod</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>10:10</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>12:12</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/datetime64_monthformatter.html b/pandas/pandas/tests/io/formats/data/html/datetime64_monthformatter.html
new file mode 100644
index 0000000000000..589c8fba858a5
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/datetime64_monthformatter.html
@@ -0,0 +1,18 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>months</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>2016-01</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>2016-02</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/escape_disabled.html b/pandas/pandas/tests/io/formats/data/html/escape_disabled.html
new file mode 100644
index 0000000000000..260a04d26108b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/escape_disabled.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>co<l1</th>
+      <th>co>l2</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>str<ing1 &amp;</th>
+      <td><b>bold</b></td>
+      <td><b>bold</b></td>
+    </tr>
+    <tr>
+      <th>stri>ng2 &amp;</th>
+      <td><b>bold</b></td>
+      <td><b>bold</b></td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/escaped.html b/pandas/pandas/tests/io/formats/data/html/escaped.html
new file mode 100644
index 0000000000000..d68bdd3df79c4
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/escaped.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>co&lt;l1</th>
+      <th>co&gt;l2</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>str&lt;ing1 &amp;amp;</th>
+      <td>&lt;type 'str'&gt;</td>
+      <td>&lt;type 'str'&gt;</td>
+    </tr>
+    <tr>
+      <th>stri&gt;ng2 &amp;amp;</th>
+      <td>&lt;type 'str'&gt;</td>
+      <td>&lt;type 'str'&gt;</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh12031_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh12031_expected_output.html
new file mode 100644
index 0000000000000..896e154a2b324
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh12031_expected_output.html
@@ -0,0 +1,22 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>A</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>6,0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>3,1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2,2</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh13828_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh13828_expected_output.html
new file mode 100644
index 0000000000000..690d638c31d5b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh13828_expected_output.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>Group</th>
+      <th>Data</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>A</td>
+      <td>1.22</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>A</td>
+      <td>{na_rep}</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_1.html b/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_1.html
new file mode 100644
index 0000000000000..4cfd8785de825
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_1.html
@@ -0,0 +1,274 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>n</th>
+    </tr>
+    <tr>
+      <th>a</th>
+      <th>b</th>
+      <th>c</th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="21" valign="top">100</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>6</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">20</th>
+      <th>1</th>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>8</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>9</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>10</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>11</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>12</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>13</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>14</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>16</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>17</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>18</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>19</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>20</td>
+    </tr>
+    <tr>
+      <th rowspan="19" valign="top">200</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>21</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>22</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>23</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>24</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>25</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>26</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>27</td>
+    </tr>
+    <tr>
+      <th rowspan="5" valign="top">20</th>
+      <th>1</th>
+      <td>28</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>29</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>33</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>34</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>35</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>36</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>37</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>38</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>39</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>40</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>41</td>
+    </tr>
+    <tr>
+      <th rowspan="21" valign="top">300</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>42</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>43</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>44</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>45</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>46</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>47</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>48</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">20</th>
+      <th>1</th>
+      <td>49</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>50</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>51</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>52</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>53</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>54</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>60</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>61</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>62</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_2.html b/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_2.html
new file mode 100644
index 0000000000000..d4e7fd9bd8135
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh14882_expected_output_2.html
@@ -0,0 +1,258 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>n</th>
+    </tr>
+    <tr>
+      <th>a</th>
+      <th>b</th>
+      <th>c</th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="21" valign="top">100</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>6</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">20</th>
+      <th>1</th>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>8</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>9</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>10</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>11</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>12</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>13</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>14</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>16</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>17</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>18</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>19</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>20</td>
+    </tr>
+    <tr>
+      <th rowspan="15" valign="top">200</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>21</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>22</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>23</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>24</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>25</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>26</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>27</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>35</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>36</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>37</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>38</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>39</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>40</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>41</td>
+    </tr>
+    <tr>
+      <th rowspan="21" valign="top">300</th>
+      <th rowspan="7" valign="top">10</th>
+      <th>1</th>
+      <td>42</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>43</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>44</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>45</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>46</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>47</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>48</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">20</th>
+      <th>1</th>
+      <td>49</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>50</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>51</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>52</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>53</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>54</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th rowspan="7" valign="top">30</th>
+      <th>1</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>60</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>61</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>62</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh14998_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh14998_expected_output.html
new file mode 100644
index 0000000000000..62b96493a8ecd
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh14998_expected_output.html
@@ -0,0 +1,12 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>A</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>1</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh15019_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh15019_expected_output.html
new file mode 100644
index 0000000000000..5fb9d960f4465
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh15019_expected_output.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>1.764052</td>
+      <td>0.400157</td>
+    </tr>
+    <tr>
+      <td>0.978738</td>
+      <td>2.240893</td>
+    </tr>
+    <tr>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <td>0.950088</td>
+      <td>-0.151357</td>
+    </tr>
+    <tr>
+      <td>-0.103219</td>
+      <td>0.410599</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh21625_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh21625_expected_output.html
new file mode 100644
index 0000000000000..a87e4ca301d9d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh21625_expected_output.html
@@ -0,0 +1,14 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>x</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0.200</td>
+    </tr>
+  </tbody>
+</table>
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/formats/data/html/gh22270_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh22270_expected_output.html
new file mode 100644
index 0000000000000..6694c43dc9e68
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh22270_expected_output.html
@@ -0,0 +1,14 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>x</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>100</td>
+    </tr>
+  </tbody>
+</table>
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/formats/data/html/gh22579_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh22579_expected_output.html
new file mode 100644
index 0000000000000..425b0f915ed16
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh22579_expected_output.html
@@ -0,0 +1,76 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th colspan="2" halign="left">a</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th>c</th>
+      <th>d</th>
+      <th>c</th>
+      <th>d</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>10</td>
+      <td>10</td>
+      <td>10</td>
+    </tr>
+    <tr>
+      <td>1</td>
+      <td>11</td>
+      <td>11</td>
+      <td>11</td>
+    </tr>
+    <tr>
+      <td>2</td>
+      <td>12</td>
+      <td>12</td>
+      <td>12</td>
+    </tr>
+    <tr>
+      <td>3</td>
+      <td>13</td>
+      <td>13</td>
+      <td>13</td>
+    </tr>
+    <tr>
+      <td>4</td>
+      <td>14</td>
+      <td>14</td>
+      <td>14</td>
+    </tr>
+    <tr>
+      <td>5</td>
+      <td>15</td>
+      <td>15</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <td>6</td>
+      <td>16</td>
+      <td>16</td>
+      <td>16</td>
+    </tr>
+    <tr>
+      <td>7</td>
+      <td>17</td>
+      <td>17</td>
+      <td>17</td>
+    </tr>
+    <tr>
+      <td>8</td>
+      <td>18</td>
+      <td>18</td>
+      <td>18</td>
+    </tr>
+    <tr>
+      <td>9</td>
+      <td>19</td>
+      <td>19</td>
+      <td>19</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh22783_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh22783_expected_output.html
new file mode 100644
index 0000000000000..107db43c48639
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh22783_expected_output.html
@@ -0,0 +1,27 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>3</th>
+      <th>4</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>1.764052</td>
+      <td>0.400157</td>
+      <td>...</td>
+      <td>2.240893</td>
+      <td>1.867558</td>
+    </tr>
+    <tr>
+      <td>-0.977278</td>
+      <td>0.950088</td>
+      <td>...</td>
+      <td>-0.103219</td>
+      <td>0.410599</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh22783_named_columns_index.html b/pandas/pandas/tests/io/formats/data/html/gh22783_named_columns_index.html
new file mode 100644
index 0000000000000..55ab290920cc5
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh22783_named_columns_index.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>3</th>
+      <th>4</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th></th>
+      <td>1.764052</td>
+      <td>0.400157</td>
+      <td>...</td>
+      <td>2.240893</td>
+      <td>1.867558</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>-0.977278</td>
+      <td>0.950088</td>
+      <td>...</td>
+      <td>-0.103219</td>
+      <td>0.410599</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh40024_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh40024_expected_output.html
new file mode 100644
index 0000000000000..0877c29525d2c
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh40024_expected_output.html
@@ -0,0 +1,18 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>x</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>1,000</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>test</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh47103_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh47103_expected_output.html
new file mode 100644
index 0000000000000..f8c54cb85c815
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh47103_expected_output.html
@@ -0,0 +1,16 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+      <th>b</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>1</td>
+      <td>[1, 2, 3]</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh6131_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh6131_expected_output.html
new file mode 100644
index 0000000000000..cb3a3363ff016
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh6131_expected_output.html
@@ -0,0 +1,46 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th colspan="3" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>a</th>
+      <th>aa</th>
+      <th>...</th>
+      <th>ac</th>
+    </tr>
+    <tr>
+      <th>b</th>
+      <th>c</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>ba</th>
+      <th>ca</th>
+      <td>1.0</td>
+      <td>...</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>bb</th>
+      <th>cb</th>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>bc</th>
+      <th>cc</th>
+      <td>NaN</td>
+      <td>...</td>
+      <td>3.0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/gh8452_expected_output.html b/pandas/pandas/tests/io/formats/data/html/gh8452_expected_output.html
new file mode 100644
index 0000000000000..81ce397a201e0
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/gh8452_expected_output.html
@@ -0,0 +1,28 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th colspan="2" halign="left">a</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th>c</th>
+      <th>d</th>
+      <th>c</th>
+      <th>d</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>3</td>
+      <td>5</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <td>1</td>
+      <td>4</td>
+      <td>6</td>
+      <td>4</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_12.html b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_12.html
new file mode 100644
index 0000000000000..4eb3f5319749d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_12.html
@@ -0,0 +1,70 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>56</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>57</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>58</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>59</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>60</th>
+      <td>60</td>
+    </tr>
+  </tbody>
+</table>
+<p>61 rows × 1 columns</p>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_4.html b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_4.html
new file mode 100644
index 0000000000000..2b1d97aec517c
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_10_min_rows_4.html
@@ -0,0 +1,46 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>59</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>60</th>
+      <td>60</td>
+    </tr>
+  </tbody>
+</table>
+<p>61 rows × 1 columns</p>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_12_min_rows_None.html b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_12_min_rows_None.html
new file mode 100644
index 0000000000000..a539e5a4884a1
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_12_min_rows_None.html
@@ -0,0 +1,78 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>55</th>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>56</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>57</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>58</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>59</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>60</th>
+      <td>60</td>
+    </tr>
+  </tbody>
+</table>
+<p>61 rows × 1 columns</p>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_None_min_rows_12.html b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_None_min_rows_12.html
new file mode 100644
index 0000000000000..3e680a505c6d6
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_max_rows_None_min_rows_12.html
@@ -0,0 +1,269 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>6</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>8</th>
+      <td>8</td>
+    </tr>
+    <tr>
+      <th>9</th>
+      <td>9</td>
+    </tr>
+    <tr>
+      <th>10</th>
+      <td>10</td>
+    </tr>
+    <tr>
+      <th>11</th>
+      <td>11</td>
+    </tr>
+    <tr>
+      <th>12</th>
+      <td>12</td>
+    </tr>
+    <tr>
+      <th>13</th>
+      <td>13</td>
+    </tr>
+    <tr>
+      <th>14</th>
+      <td>14</td>
+    </tr>
+    <tr>
+      <th>15</th>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>16</th>
+      <td>16</td>
+    </tr>
+    <tr>
+      <th>17</th>
+      <td>17</td>
+    </tr>
+    <tr>
+      <th>18</th>
+      <td>18</td>
+    </tr>
+    <tr>
+      <th>19</th>
+      <td>19</td>
+    </tr>
+    <tr>
+      <th>20</th>
+      <td>20</td>
+    </tr>
+    <tr>
+      <th>21</th>
+      <td>21</td>
+    </tr>
+    <tr>
+      <th>22</th>
+      <td>22</td>
+    </tr>
+    <tr>
+      <th>23</th>
+      <td>23</td>
+    </tr>
+    <tr>
+      <th>24</th>
+      <td>24</td>
+    </tr>
+    <tr>
+      <th>25</th>
+      <td>25</td>
+    </tr>
+    <tr>
+      <th>26</th>
+      <td>26</td>
+    </tr>
+    <tr>
+      <th>27</th>
+      <td>27</td>
+    </tr>
+    <tr>
+      <th>28</th>
+      <td>28</td>
+    </tr>
+    <tr>
+      <th>29</th>
+      <td>29</td>
+    </tr>
+    <tr>
+      <th>30</th>
+      <td>30</td>
+    </tr>
+    <tr>
+      <th>31</th>
+      <td>31</td>
+    </tr>
+    <tr>
+      <th>32</th>
+      <td>32</td>
+    </tr>
+    <tr>
+      <th>33</th>
+      <td>33</td>
+    </tr>
+    <tr>
+      <th>34</th>
+      <td>34</td>
+    </tr>
+    <tr>
+      <th>35</th>
+      <td>35</td>
+    </tr>
+    <tr>
+      <th>36</th>
+      <td>36</td>
+    </tr>
+    <tr>
+      <th>37</th>
+      <td>37</td>
+    </tr>
+    <tr>
+      <th>38</th>
+      <td>38</td>
+    </tr>
+    <tr>
+      <th>39</th>
+      <td>39</td>
+    </tr>
+    <tr>
+      <th>40</th>
+      <td>40</td>
+    </tr>
+    <tr>
+      <th>41</th>
+      <td>41</td>
+    </tr>
+    <tr>
+      <th>42</th>
+      <td>42</td>
+    </tr>
+    <tr>
+      <th>43</th>
+      <td>43</td>
+    </tr>
+    <tr>
+      <th>44</th>
+      <td>44</td>
+    </tr>
+    <tr>
+      <th>45</th>
+      <td>45</td>
+    </tr>
+    <tr>
+      <th>46</th>
+      <td>46</td>
+    </tr>
+    <tr>
+      <th>47</th>
+      <td>47</td>
+    </tr>
+    <tr>
+      <th>48</th>
+      <td>48</td>
+    </tr>
+    <tr>
+      <th>49</th>
+      <td>49</td>
+    </tr>
+    <tr>
+      <th>50</th>
+      <td>50</td>
+    </tr>
+    <tr>
+      <th>51</th>
+      <td>51</td>
+    </tr>
+    <tr>
+      <th>52</th>
+      <td>52</td>
+    </tr>
+    <tr>
+      <th>53</th>
+      <td>53</td>
+    </tr>
+    <tr>
+      <th>54</th>
+      <td>54</td>
+    </tr>
+    <tr>
+      <th>55</th>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>56</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>57</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>58</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>59</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>60</th>
+      <td>60</td>
+    </tr>
+  </tbody>
+</table>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_no_truncation.html b/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_no_truncation.html
new file mode 100644
index 0000000000000..10f6247e37def
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_no_truncation.html
@@ -0,0 +1,105 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>6</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>8</th>
+      <td>8</td>
+    </tr>
+    <tr>
+      <th>9</th>
+      <td>9</td>
+    </tr>
+    <tr>
+      <th>10</th>
+      <td>10</td>
+    </tr>
+    <tr>
+      <th>11</th>
+      <td>11</td>
+    </tr>
+    <tr>
+      <th>12</th>
+      <td>12</td>
+    </tr>
+    <tr>
+      <th>13</th>
+      <td>13</td>
+    </tr>
+    <tr>
+      <th>14</th>
+      <td>14</td>
+    </tr>
+    <tr>
+      <th>15</th>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>16</th>
+      <td>16</td>
+    </tr>
+    <tr>
+      <th>17</th>
+      <td>17</td>
+    </tr>
+    <tr>
+      <th>18</th>
+      <td>18</td>
+    </tr>
+    <tr>
+      <th>19</th>
+      <td>19</td>
+    </tr>
+  </tbody>
+</table>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_truncated.html b/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_truncated.html
new file mode 100644
index 0000000000000..4eb3f5319749d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/html_repr_min_rows_default_truncated.html
@@ -0,0 +1,70 @@
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>a</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>56</th>
+      <td>56</td>
+    </tr>
+    <tr>
+      <th>57</th>
+      <td>57</td>
+    </tr>
+    <tr>
+      <th>58</th>
+      <td>58</td>
+    </tr>
+    <tr>
+      <th>59</th>
+      <td>59</td>
+    </tr>
+    <tr>
+      <th>60</th>
+      <td>60</td>
+    </tr>
+  </tbody>
+</table>
+<p>61 rows × 1 columns</p>
+</div>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_1.html b/pandas/pandas/tests/io/formats/data/html/index_1.html
new file mode 100644
index 0000000000000..41221865a7cb7
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_1.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>foo</th>
+      <td>1</td>
+      <td>1.2</td>
+      <td>one</td>
+    </tr>
+    <tr>
+      <th>bar</th>
+      <td>2</td>
+      <td>3.4</td>
+      <td>two</td>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <td>3</td>
+      <td>5.6</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_2.html b/pandas/pandas/tests/io/formats/data/html/index_2.html
new file mode 100644
index 0000000000000..a86ba80a69bb1
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_2.html
@@ -0,0 +1,26 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>1</td>
+      <td>1.2</td>
+      <td>one</td>
+    </tr>
+    <tr>
+      <td>2</td>
+      <td>3.4</td>
+      <td>two</td>
+    </tr>
+    <tr>
+      <td>3</td>
+      <td>5.6</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_3.html b/pandas/pandas/tests/io/formats/data/html/index_3.html
new file mode 100644
index 0000000000000..02edba4961bc7
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_3.html
@@ -0,0 +1,36 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+    <tr>
+      <th>idx</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>foo</th>
+      <td>1</td>
+      <td>1.2</td>
+      <td>one</td>
+    </tr>
+    <tr>
+      <th>bar</th>
+      <td>2</td>
+      <td>3.4</td>
+      <td>two</td>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <td>3</td>
+      <td>5.6</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_4.html b/pandas/pandas/tests/io/formats/data/html/index_4.html
new file mode 100644
index 0000000000000..0d1bf9ffcd717
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_4.html
@@ -0,0 +1,33 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">foo</th>
+      <th>car</th>
+      <td>1</td>
+      <td>1.2</td>
+      <td>one</td>
+    </tr>
+    <tr>
+      <th>bike</th>
+      <td>2</td>
+      <td>3.4</td>
+      <td>two</td>
+    </tr>
+    <tr>
+      <th>bar</th>
+      <th>car</th>
+      <td>3</td>
+      <td>5.6</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_5.html b/pandas/pandas/tests/io/formats/data/html/index_5.html
new file mode 100644
index 0000000000000..c5ac12ecd630e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_5.html
@@ -0,0 +1,40 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+    <tr>
+      <th>idx1</th>
+      <th>idx2</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">foo</th>
+      <th>car</th>
+      <td>1</td>
+      <td>1.2</td>
+      <td>one</td>
+    </tr>
+    <tr>
+      <th>bike</th>
+      <td>2</td>
+      <td>3.4</td>
+      <td>two</td>
+    </tr>
+    <tr>
+      <th>bar</th>
+      <th>car</th>
+      <td>3</td>
+      <td>5.6</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_formatter.html b/pandas/pandas/tests/io/formats/data/html/index_formatter.html
new file mode 100644
index 0000000000000..7a2f8a9f52a04
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_formatter.html
@@ -0,0 +1,31 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>foo</th>
+      <th>None</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>a</th>
+      <td>0</td>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>b</th>
+      <td>2</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>4</td>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>d</th>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_multi.html
new file mode 100644
index 0000000000000..817b54d77f8b1
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_multi.html
@@ -0,0 +1,34 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th>columns.name.0</th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>columns.name.1</th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+    <tr>
+      <th>index.name.0</th>
+      <th>index.name.1</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_standard.html
new file mode 100644
index 0000000000000..e85965f14075d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_named_standard.html
@@ -0,0 +1,29 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>index.name.0</th>
+      <th>index.name.1</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_none.html b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_none.html
new file mode 100644
index 0000000000000..8c41d2e29f2c0
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_none.html
@@ -0,0 +1,23 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>index.name.0</th>
+      <th>index.name.1</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..7af63e893b12e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_multi.html
@@ -0,0 +1,34 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+    <tr>
+      <th>index.name.0</th>
+      <th>index.name.1</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..2f7837864bf88
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_multi_columns_unnamed_standard.html
@@ -0,0 +1,29 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>index.name.0</th>
+      <th>index.name.1</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_multi.html
new file mode 100644
index 0000000000000..ca9b8bd834a9c
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_multi.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>columns.name.0</th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th>columns.name.1</th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_standard.html
new file mode 100644
index 0000000000000..6478c99ad85e9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_named_standard.html
@@ -0,0 +1,26 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_none.html b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_none.html
new file mode 100644
index 0000000000000..432d8e06d5784
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_none.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..d7660872177dc
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_multi.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..4810f66018d3b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_named_standard_columns_unnamed_standard.html
@@ -0,0 +1,26 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_multi.html
new file mode 100644
index 0000000000000..e111f55be7d25
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_multi.html
@@ -0,0 +1,25 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>columns.name.0</th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th>columns.name.1</th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_standard.html
new file mode 100644
index 0000000000000..d3a9ba017b43e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_none_columns_named_standard.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_none_columns_none.html b/pandas/pandas/tests/io/formats/data/html/index_none_columns_none.html
new file mode 100644
index 0000000000000..44899858d9519
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_none_columns_none.html
@@ -0,0 +1,12 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..b21a618328b1b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_multi.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..1249fa5605099
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_none_columns_unnamed_standard.html
@@ -0,0 +1,18 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_multi.html
new file mode 100644
index 0000000000000..95c38c9c8fd28
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_multi.html
@@ -0,0 +1,28 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th>columns.name.0</th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>columns.name.1</th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_standard.html
new file mode 100644
index 0000000000000..9583a21f55f01
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_named_standard.html
@@ -0,0 +1,23 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_none.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_none.html
new file mode 100644
index 0000000000000..81da7c3619abc
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_none.html
@@ -0,0 +1,15 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..f620259037b60
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_multi.html
@@ -0,0 +1,28 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..2ca18c288437b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_multi_columns_unnamed_standard.html
@@ -0,0 +1,23 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th>b</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>c</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_multi.html
new file mode 100644
index 0000000000000..ed3360f898afd
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_multi.html
@@ -0,0 +1,25 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>columns.name.0</th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th>columns.name.1</th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_standard.html
new file mode 100644
index 0000000000000..54da03858a9a4
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_named_standard.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_none.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_none.html
new file mode 100644
index 0000000000000..3d958afe4a4ac
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_none.html
@@ -0,0 +1,14 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..b57fafbe0ca40
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_multi.html
@@ -0,0 +1,25 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>b</th>
+      <th>c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..235ca61a9e63d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/index_unnamed_standard_columns_unnamed_standard.html
@@ -0,0 +1,21 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/justify.html b/pandas/pandas/tests/io/formats/data/html/justify.html
new file mode 100644
index 0000000000000..33e4b5715260e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/justify.html
@@ -0,0 +1,30 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: {justify};">
+      <th></th>
+      <th>A</th>
+      <th>B</th>
+      <th>C</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>6</td>
+      <td>1</td>
+      <td>223442</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>30000</td>
+      <td>2</td>
+      <td>0</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2</td>
+      <td>70000</td>
+      <td>1</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_1.html b/pandas/pandas/tests/io/formats/data/html/multiindex_1.html
new file mode 100644
index 0000000000000..88db177545972
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_1.html
@@ -0,0 +1,32 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>CL0</th>
+      <th colspan="2" halign="left">0</th>
+      <th colspan="2" halign="left">1</th>
+    </tr>
+    <tr>
+      <th>CL1</th>
+      <th>0</th>
+      <th>1</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>a</td>
+      <td>b</td>
+      <td>c</td>
+      <td>d</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>e</td>
+      <td>f</td>
+      <td>g</td>
+      <td>h</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_2.html b/pandas/pandas/tests/io/formats/data/html/multiindex_2.html
new file mode 100644
index 0000000000000..289ea2202d6b9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_2.html
@@ -0,0 +1,34 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>2</th>
+      <th>3</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>a</td>
+      <td>b</td>
+      <td>c</td>
+      <td>d</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>e</td>
+      <td>f</td>
+      <td>g</td>
+      <td>h</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_1.html b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_1.html
new file mode 100644
index 0000000000000..5b5bcf9ce0a96
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_1.html
@@ -0,0 +1,40 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">0</th>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>2</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">1</th>
+      <th>0</th>
+      <td>4</td>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_2.html b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_2.html
new file mode 100644
index 0000000000000..fd4c6bd23dae2
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_2.html
@@ -0,0 +1,46 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th>foo</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>0</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">0</th>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>2</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">1</th>
+      <th>0</th>
+      <td>4</td>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_1.html b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_1.html
new file mode 100644
index 0000000000000..42a5ea5eb5899
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_1.html
@@ -0,0 +1,42 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>0</th>
+      <th>1</th>
+      <td>2</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <th>0</th>
+      <td>4</td>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <th>1</th>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_2.html b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_2.html
new file mode 100644
index 0000000000000..2be61392e8573
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/multiindex_sparsify_false_multi_sparse_2.html
@@ -0,0 +1,48 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th>foo</th>
+      <th>0</th>
+      <th>1</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>0</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+    </tr>
+    <tr>
+      <th>0</th>
+      <th>1</th>
+      <td>2</td>
+      <td>3</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <th>0</th>
+      <td>4</td>
+      <td>5</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <th>1</th>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/render_links_false.html b/pandas/pandas/tests/io/formats/data/html/render_links_false.html
new file mode 100644
index 0000000000000..6feb403d63051
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/render_links_false.html
@@ -0,0 +1,24 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>foo</th>
+      <th>bar</th>
+      <th>None</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>https://pandas.pydata.org/?q1=a&amp;q2=b</td>
+      <td>pydata.org</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>www.pydata.org</td>
+      <td>pydata.org</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/render_links_true.html b/pandas/pandas/tests/io/formats/data/html/render_links_true.html
new file mode 100644
index 0000000000000..3eb53f3160a77
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/render_links_true.html
@@ -0,0 +1,24 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>foo</th>
+      <th>bar</th>
+      <th>None</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td><a href="https://pandas.pydata.org/?q1=a&q2=b" target="_blank">https://pandas.pydata.org/?q1=a&amp;q2=b</a></td>
+      <td>pydata.org</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>0</td>
+      <td>www.pydata.org</td>
+      <td>pydata.org</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_multi.html
new file mode 100644
index 0000000000000..e66d3c816e67d
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_multi.html
@@ -0,0 +1,88 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>foo</th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>baz</th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th>baz</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_standard.html
new file mode 100644
index 0000000000000..536b371145081
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_named_standard.html
@@ -0,0 +1,72 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th>baz</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_none.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_none.html
new file mode 100644
index 0000000000000..0f262495b6c6b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_none.html
@@ -0,0 +1,62 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th>baz</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..d472cdecb12c9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_multi.html
@@ -0,0 +1,88 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th>baz</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..31c71ca3e59f6
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_multi_columns_unnamed_standard.html
@@ -0,0 +1,72 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th></th>
+      <th>baz</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_multi.html
new file mode 100644
index 0000000000000..779e84f6ee6d1
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_multi.html
@@ -0,0 +1,74 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>foo</th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_standard.html
new file mode 100644
index 0000000000000..b86454f5fb11f
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_named_standard.html
@@ -0,0 +1,62 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_none.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_none.html
new file mode 100644
index 0000000000000..d294a507dbce4
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_none.html
@@ -0,0 +1,54 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..24b776e18bef9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_multi.html
@@ -0,0 +1,74 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..a0ca960207ac0
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_named_standard_columns_unnamed_standard.html
@@ -0,0 +1,62 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+    <tr>
+      <th>index.name</th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_multi.html
new file mode 100644
index 0000000000000..6640db4cf8704
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_multi.html
@@ -0,0 +1,66 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>foo</th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_standard.html
new file mode 100644
index 0000000000000..364a0b98d6548
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_named_standard.html
@@ -0,0 +1,54 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th></th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th></th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_none.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_none.html
new file mode 100644
index 0000000000000..e2af1ba42e940
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_none.html
@@ -0,0 +1,39 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..8c9a9e244277b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_multi.html
@@ -0,0 +1,58 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..b9dcf52619490
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_none_columns_unnamed_standard.html
@@ -0,0 +1,48 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_multi.html
new file mode 100644
index 0000000000000..0590d0dea6669
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_multi.html
@@ -0,0 +1,78 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>foo</th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>baz</th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_standard.html
new file mode 100644
index 0000000000000..28a2d964675a3
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_named_standard.html
@@ -0,0 +1,62 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_none.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_none.html
new file mode 100644
index 0000000000000..387ac51b17634
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_none.html
@@ -0,0 +1,50 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..30cd85904be4e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_multi.html
@@ -0,0 +1,78 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..81edece220408
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_multi_columns_unnamed_standard.html
@@ -0,0 +1,62 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th></th>
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">a</th>
+      <th rowspan="2" valign="top">c</th>
+      <th>e</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">b</th>
+      <th rowspan="2" valign="top">d</th>
+      <th>e</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>f</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_multi.html
new file mode 100644
index 0000000000000..2acacfed3a6d0
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_multi.html
@@ -0,0 +1,66 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th>foo</th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_standard.html
new file mode 100644
index 0000000000000..c9bacdbd241a6
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_named_standard.html
@@ -0,0 +1,54 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th>columns.name</th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_none.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_none.html
new file mode 100644
index 0000000000000..f2696f7d6b46a
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_none.html
@@ -0,0 +1,44 @@
+<table border="1" class="dataframe">
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_multi.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_multi.html
new file mode 100644
index 0000000000000..37e731520c7d9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_multi.html
@@ -0,0 +1,66 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">a</th>
+      <th>...</th>
+      <th colspan="2" halign="left">b</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th colspan="2" halign="left">c</th>
+      <th>...</th>
+      <th colspan="2" halign="left">d</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th>e</th>
+      <th>f</th>
+      <th>...</th>
+      <th>e</th>
+      <th>f</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_standard.html b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_standard.html
new file mode 100644
index 0000000000000..3241ff41c5c58
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/trunc_df_index_unnamed_standard_columns_unnamed_standard.html
@@ -0,0 +1,54 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>6</th>
+      <th>7</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0</td>
+      <td>1</td>
+      <td>...</td>
+      <td>6</td>
+      <td>7</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>8</td>
+      <td>9</td>
+      <td>...</td>
+      <td>14</td>
+      <td>15</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>48</td>
+      <td>49</td>
+      <td>...</td>
+      <td>54</td>
+      <td>55</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>56</td>
+      <td>57</td>
+      <td>...</td>
+      <td>62</td>
+      <td>63</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/truncate.html b/pandas/pandas/tests/io/formats/data/html/truncate.html
new file mode 100644
index 0000000000000..a5eb8c5cdbb9b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/truncate.html
@@ -0,0 +1,86 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>0</th>
+      <th>1</th>
+      <th>...</th>
+      <th>18</th>
+      <th>19</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>2001-01-01</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-02</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-03</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-04</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>2001-01-17</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-18</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-19</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>2001-01-20</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/truncate_formatter.html b/pandas/pandas/tests/io/formats/data/html/truncate_formatter.html
new file mode 100644
index 0000000000000..7615ef89d85d1
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/truncate_formatter.html
@@ -0,0 +1,36 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>A</th>
+      <th>...</th>
+      <th>D</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>1_mod</td>
+      <td>...</td>
+      <td>4</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>5_mod</td>
+      <td>...</td>
+      <td>8</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>9_mod</td>
+      <td>...</td>
+      <td>12</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>13_mod</td>
+      <td>...</td>
+      <td>16</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/truncate_multi_index.html b/pandas/pandas/tests/io/formats/data/html/truncate_multi_index.html
new file mode 100644
index 0000000000000..8a295d66db130
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/truncate_multi_index.html
@@ -0,0 +1,101 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th colspan="2" halign="left">bar</th>
+      <th>baz</th>
+      <th>...</th>
+      <th>foo</th>
+      <th colspan="2" halign="left">qux</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>one</th>
+      <th>two</th>
+      <th>one</th>
+      <th>...</th>
+      <th>two</th>
+      <th>one</th>
+      <th>two</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th rowspan="2" valign="top">bar</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th rowspan="2" valign="top">qux</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/truncate_multi_index_sparse_off.html b/pandas/pandas/tests/io/formats/data/html/truncate_multi_index_sparse_off.html
new file mode 100644
index 0000000000000..6a7e1b5a59e3b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/truncate_multi_index_sparse_off.html
@@ -0,0 +1,105 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>bar</th>
+      <th>bar</th>
+      <th>baz</th>
+      <th>...</th>
+      <th>foo</th>
+      <th>qux</th>
+      <th>qux</th>
+    </tr>
+    <tr>
+      <th></th>
+      <th></th>
+      <th>one</th>
+      <th>two</th>
+      <th>one</th>
+      <th>...</th>
+      <th>two</th>
+      <th>one</th>
+      <th>two</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>bar</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>bar</th>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>baz</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>...</th>
+      <th>...</th>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+      <td>...</td>
+    </tr>
+    <tr>
+      <th>foo</th>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>qux</th>
+      <th>one</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+    <tr>
+      <th>qux</th>
+      <th>two</th>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>...</td>
+      <td>NaN</td>
+      <td>NaN</td>
+      <td>NaN</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/unicode_1.html b/pandas/pandas/tests/io/formats/data/html/unicode_1.html
new file mode 100644
index 0000000000000..72b810181bade
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/unicode_1.html
@@ -0,0 +1,50 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>σ</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>0.0</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>1.0</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>2.0</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>3.0</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>4.0</td>
+    </tr>
+    <tr>
+      <th>5</th>
+      <td>5.0</td>
+    </tr>
+    <tr>
+      <th>6</th>
+      <td>6.0</td>
+    </tr>
+    <tr>
+      <th>7</th>
+      <td>7.0</td>
+    </tr>
+    <tr>
+      <th>8</th>
+      <td>8.0</td>
+    </tr>
+    <tr>
+      <th>9</th>
+      <td>9.0</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/unicode_2.html b/pandas/pandas/tests/io/formats/data/html/unicode_2.html
new file mode 100644
index 0000000000000..79c088093e539
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/unicode_2.html
@@ -0,0 +1,14 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>A</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>σ</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/various_dtypes_formatted.html b/pandas/pandas/tests/io/formats/data/html/various_dtypes_formatted.html
new file mode 100644
index 0000000000000..7d2ede3379213
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/various_dtypes_formatted.html
@@ -0,0 +1,36 @@
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>i</th>
+      <th>f</th>
+      <th>I</th>
+      <th>s</th>
+      <th>b</th>
+      <th>c</th>
+      <th>o</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+      <td>formatted</td>
+    </tr>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/data/html/with_classes.html b/pandas/pandas/tests/io/formats/data/html/with_classes.html
new file mode 100644
index 0000000000000..8cee3f0c7052b
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/data/html/with_classes.html
@@ -0,0 +1,9 @@
+<table border="1" class="dataframe sortable draggable">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+    </tr>
+  </thead>
+  <tbody>
+  </tbody>
+</table>
diff --git a/pandas/pandas/tests/io/formats/style/__init__.py b/pandas/pandas/tests/io/formats/style/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/formats/style/test_bar.py b/pandas/pandas/tests/io/formats/style/test_bar.py
new file mode 100644
index 0000000000000..d28c7c566d851
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_bar.py
@@ -0,0 +1,359 @@
+import io
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    read_csv,
+)
+
+pytest.importorskip("jinja2")
+
+
+def bar_grad(a=None, b=None, c=None, d=None):
+    """Used in multiple tests to simplify formatting of expected result"""
+    ret = [("width", "10em")]
+    if all(x is None for x in [a, b, c, d]):
+        return ret
+    return ret + [
+        (
+            "background",
+            f"linear-gradient(90deg,{','.join([x for x in [a, b, c, d] if x])})",
+        )
+    ]
+
+
+def no_bar():
+    return bar_grad()
+
+
+def bar_to(x, color="#d65f5f"):
+    return bar_grad(f" {color} {x:.1f}%", f" transparent {x:.1f}%")
+
+
+def bar_from_to(x, y, color="#d65f5f"):
+    return bar_grad(
+        f" transparent {x:.1f}%",
+        f" {color} {x:.1f}%",
+        f" {color} {y:.1f}%",
+        f" transparent {y:.1f}%",
+    )
+
+
+@pytest.fixture
+def df_pos():
+    return DataFrame([[1], [2], [3]])
+
+
+@pytest.fixture
+def df_neg():
+    return DataFrame([[-1], [-2], [-3]])
+
+
+@pytest.fixture
+def df_mix():
+    return DataFrame([[-3], [1], [2]])
+
+
+@pytest.mark.parametrize(
+    "align, exp",
+    [
+        ("left", [no_bar(), bar_to(50), bar_to(100)]),
+        ("right", [bar_to(100), bar_from_to(50, 100), no_bar()]),
+        ("mid", [bar_to(33.33), bar_to(66.66), bar_to(100)]),
+        ("zero", [bar_from_to(50, 66.7), bar_from_to(50, 83.3), bar_from_to(50, 100)]),
+        ("mean", [bar_to(50), no_bar(), bar_from_to(50, 100)]),
+        (2.0, [bar_to(50), no_bar(), bar_from_to(50, 100)]),
+        (np.median, [bar_to(50), no_bar(), bar_from_to(50, 100)]),
+    ],
+)
+def test_align_positive_cases(df_pos, align, exp):
+    # test different align cases for all positive values
+    result = df_pos.style.bar(align=align)._compute().ctx
+    expected = {(0, 0): exp[0], (1, 0): exp[1], (2, 0): exp[2]}
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "align, exp",
+    [
+        ("left", [bar_to(100), bar_to(50), no_bar()]),
+        ("right", [no_bar(), bar_from_to(50, 100), bar_to(100)]),
+        ("mid", [bar_from_to(66.66, 100), bar_from_to(33.33, 100), bar_to(100)]),
+        ("zero", [bar_from_to(33.33, 50), bar_from_to(16.66, 50), bar_to(50)]),
+        ("mean", [bar_from_to(50, 100), no_bar(), bar_to(50)]),
+        (-2.0, [bar_from_to(50, 100), no_bar(), bar_to(50)]),
+        (np.median, [bar_from_to(50, 100), no_bar(), bar_to(50)]),
+    ],
+)
+def test_align_negative_cases(df_neg, align, exp):
+    # test different align cases for all negative values
+    result = df_neg.style.bar(align=align)._compute().ctx
+    expected = {(0, 0): exp[0], (1, 0): exp[1], (2, 0): exp[2]}
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "align, exp",
+    [
+        ("left", [no_bar(), bar_to(80), bar_to(100)]),
+        ("right", [bar_to(100), bar_from_to(80, 100), no_bar()]),
+        ("mid", [bar_to(60), bar_from_to(60, 80), bar_from_to(60, 100)]),
+        ("zero", [bar_to(50), bar_from_to(50, 66.66), bar_from_to(50, 83.33)]),
+        ("mean", [bar_to(50), bar_from_to(50, 66.66), bar_from_to(50, 83.33)]),
+        (-0.0, [bar_to(50), bar_from_to(50, 66.66), bar_from_to(50, 83.33)]),
+        (np.nanmedian, [bar_to(50), no_bar(), bar_from_to(50, 62.5)]),
+    ],
+)
+@pytest.mark.parametrize("nans", [True, False])
+def test_align_mixed_cases(df_mix, align, exp, nans):
+    # test different align cases for mixed positive and negative values
+    # also test no impact of NaNs and no_bar
+    expected = {(0, 0): exp[0], (1, 0): exp[1], (2, 0): exp[2]}
+    if nans:
+        df_mix.loc[3, :] = np.nan
+        expected.update({(3, 0): no_bar()})
+    result = df_mix.style.bar(align=align)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "align, exp",
+    [
+        (
+            "left",
+            {
+                "index": [[no_bar(), no_bar()], [bar_to(100), bar_to(100)]],
+                "columns": [[no_bar(), bar_to(100)], [no_bar(), bar_to(100)]],
+                "none": [[no_bar(), bar_to(33.33)], [bar_to(66.66), bar_to(100)]],
+            },
+        ),
+        (
+            "mid",
+            {
+                "index": [[bar_to(33.33), bar_to(50)], [bar_to(100), bar_to(100)]],
+                "columns": [[bar_to(50), bar_to(100)], [bar_to(75), bar_to(100)]],
+                "none": [[bar_to(25), bar_to(50)], [bar_to(75), bar_to(100)]],
+            },
+        ),
+        (
+            "zero",
+            {
+                "index": [
+                    [bar_from_to(50, 66.66), bar_from_to(50, 75)],
+                    [bar_from_to(50, 100), bar_from_to(50, 100)],
+                ],
+                "columns": [
+                    [bar_from_to(50, 75), bar_from_to(50, 100)],
+                    [bar_from_to(50, 87.5), bar_from_to(50, 100)],
+                ],
+                "none": [
+                    [bar_from_to(50, 62.5), bar_from_to(50, 75)],
+                    [bar_from_to(50, 87.5), bar_from_to(50, 100)],
+                ],
+            },
+        ),
+        (
+            2,
+            {
+                "index": [
+                    [bar_to(50), no_bar()],
+                    [bar_from_to(50, 100), bar_from_to(50, 100)],
+                ],
+                "columns": [
+                    [bar_to(50), no_bar()],
+                    [bar_from_to(50, 75), bar_from_to(50, 100)],
+                ],
+                "none": [
+                    [bar_from_to(25, 50), no_bar()],
+                    [bar_from_to(50, 75), bar_from_to(50, 100)],
+                ],
+            },
+        ),
+    ],
+)
+@pytest.mark.parametrize("axis", ["index", "columns", "none"])
+def test_align_axis(align, exp, axis):
+    # test all axis combinations with positive values and different aligns
+    data = DataFrame([[1, 2], [3, 4]])
+    result = (
+        data.style.bar(align=align, axis=None if axis == "none" else axis)
+        ._compute()
+        .ctx
+    )
+    expected = {
+        (0, 0): exp[axis][0][0],
+        (0, 1): exp[axis][0][1],
+        (1, 0): exp[axis][1][0],
+        (1, 1): exp[axis][1][1],
+    }
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "values, vmin, vmax",
+    [
+        ("positive", 1.5, 2.5),
+        ("negative", -2.5, -1.5),
+        ("mixed", -2.5, 1.5),
+    ],
+)
+@pytest.mark.parametrize("nullify", [None, "vmin", "vmax"])  # test min/max separately
+@pytest.mark.parametrize("align", ["left", "right", "zero", "mid"])
+def test_vmin_vmax_clipping(df_pos, df_neg, df_mix, values, vmin, vmax, nullify, align):
+    # test that clipping occurs if any vmin > data_values or vmax < data_values
+    if align == "mid":  # mid acts as left or right in each case
+        if values == "positive":
+            align = "left"
+        elif values == "negative":
+            align = "right"
+    df = {"positive": df_pos, "negative": df_neg, "mixed": df_mix}[values]
+    vmin = None if nullify == "vmin" else vmin
+    vmax = None if nullify == "vmax" else vmax
+
+    clip_df = df.where(df <= (vmax if vmax else 999), other=vmax)
+    clip_df = clip_df.where(clip_df >= (vmin if vmin else -999), other=vmin)
+
+    result = (
+        df.style.bar(align=align, vmin=vmin, vmax=vmax, color=["red", "green"])
+        ._compute()
+        .ctx
+    )
+    expected = clip_df.style.bar(align=align, color=["red", "green"])._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "values, vmin, vmax",
+    [
+        ("positive", 0.5, 4.5),
+        ("negative", -4.5, -0.5),
+        ("mixed", -4.5, 4.5),
+    ],
+)
+@pytest.mark.parametrize("nullify", [None, "vmin", "vmax"])  # test min/max separately
+@pytest.mark.parametrize("align", ["left", "right", "zero", "mid"])
+def test_vmin_vmax_widening(df_pos, df_neg, df_mix, values, vmin, vmax, nullify, align):
+    # test that widening occurs if any vmax > data_values or vmin < data_values
+    if align == "mid":  # mid acts as left or right in each case
+        if values == "positive":
+            align = "left"
+        elif values == "negative":
+            align = "right"
+    df = {"positive": df_pos, "negative": df_neg, "mixed": df_mix}[values]
+    vmin = None if nullify == "vmin" else vmin
+    vmax = None if nullify == "vmax" else vmax
+
+    expand_df = df.copy()
+    expand_df.loc[3, :], expand_df.loc[4, :] = vmin, vmax
+
+    result = (
+        df.style.bar(align=align, vmin=vmin, vmax=vmax, color=["red", "green"])
+        ._compute()
+        .ctx
+    )
+    expected = expand_df.style.bar(align=align, color=["red", "green"])._compute().ctx
+    assert result.items() <= expected.items()
+
+
+def test_numerics():
+    # test data is pre-selected for numeric values
+    data = DataFrame([[1, "a"], [2, "b"]])
+    result = data.style.bar()._compute().ctx
+    assert (0, 1) not in result
+    assert (1, 1) not in result
+
+
+@pytest.mark.parametrize(
+    "align, exp",
+    [
+        ("left", [no_bar(), bar_to(100, "green")]),
+        ("right", [bar_to(100, "red"), no_bar()]),
+        ("mid", [bar_to(25, "red"), bar_from_to(25, 100, "green")]),
+        ("zero", [bar_from_to(33.33, 50, "red"), bar_from_to(50, 100, "green")]),
+    ],
+)
+def test_colors_mixed(align, exp):
+    data = DataFrame([[-1], [3]])
+    result = data.style.bar(align=align, color=["red", "green"])._compute().ctx
+    assert result == {(0, 0): exp[0], (1, 0): exp[1]}
+
+
+def test_bar_align_height():
+    # test when keyword height is used 'no-repeat center' and 'background-size' present
+    data = DataFrame([[1], [2]])
+    result = data.style.bar(align="left", height=50)._compute().ctx
+    bg_s = "linear-gradient(90deg, #d65f5f 100.0%, transparent 100.0%) no-repeat center"
+    expected = {
+        (0, 0): [("width", "10em")],
+        (1, 0): [
+            ("width", "10em"),
+            ("background", bg_s),
+            ("background-size", "100% 50.0%"),
+        ],
+    }
+    assert result == expected
+
+
+def test_bar_value_error_raises():
+    df = DataFrame({"A": [-100, -60, -30, -20]})
+
+    msg = "`align` should be in {'left', 'right', 'mid', 'mean', 'zero'} or"
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(align="poorly", color=["#d65f5f", "#5fba7d"]).to_html()
+
+    msg = r"`width` must be a value in \[0, 100\]"
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(width=200).to_html()
+
+    msg = r"`height` must be a value in \[0, 100\]"
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(height=200).to_html()
+
+
+def test_bar_color_and_cmap_error_raises():
+    df = DataFrame({"A": [1, 2, 3, 4]})
+    msg = "`color` and `cmap` cannot both be given"
+    # Test that providing both color and cmap raises a ValueError
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color="#d65f5f", cmap="viridis").to_html()
+
+
+def test_bar_invalid_color_type_error_raises():
+    df = DataFrame({"A": [1, 2, 3, 4]})
+    msg = (
+        r"`color` must be string or list or tuple of 2 strings,"
+        r"\(eg: color=\['#d65f5f', '#5fba7d'\]\)"
+    )
+    # Test that providing an invalid color type raises a ValueError
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color=123).to_html()
+
+    # Test that providing a color list with more than two elements raises a ValueError
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color=["#d65f5f", "#5fba7d", "#abcdef"]).to_html()
+
+
+def test_styler_bar_with_NA_values():
+    df1 = DataFrame({"A": [1, 2, NA, 4]})
+    df2 = DataFrame([[NA, NA], [NA, NA]])
+    expected_substring = "style type="
+    html_output1 = df1.style.bar(subset="A").to_html()
+    html_output2 = df2.style.bar(align="left", axis=None).to_html()
+    assert expected_substring in html_output1
+    assert expected_substring in html_output2
+
+
+def test_style_bar_with_pyarrow_NA_values():
+    pytest.importorskip("pyarrow")
+    data = """name,age,test1,test2,teacher
+        Adam,15,95.0,80,Ashby
+        Bob,16,81.0,82,Ashby
+        Dave,16,89.0,84,Jones
+        Fred,15,,88,Jones"""
+    df = read_csv(io.StringIO(data), dtype_backend="pyarrow")
+    expected_substring = "style type="
+    html_output = df.style.bar(subset="test1").to_html()
+    assert expected_substring in html_output
diff --git a/pandas/pandas/tests/io/formats/style/test_exceptions.py b/pandas/pandas/tests/io/formats/style/test_exceptions.py
new file mode 100644
index 0000000000000..d52e3a37e7693
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_exceptions.py
@@ -0,0 +1,44 @@
+import pytest
+
+jinja2 = pytest.importorskip("jinja2")
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        data=[[0, -0.609], [1, -1.228]],
+        columns=["A", "B"],
+        index=["x", "y"],
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+def test_concat_bad_columns(styler):
+    msg = "`other.data` must have same columns as `Styler.data"
+    with pytest.raises(ValueError, match=msg):
+        styler.concat(DataFrame([[1, 2]]).style)
+
+
+def test_concat_bad_type(styler):
+    msg = "`other` must be of type `Styler`"
+    with pytest.raises(TypeError, match=msg):
+        styler.concat(DataFrame([[1, 2]]))
+
+
+def test_concat_bad_index_levels(styler, df):
+    df = df.copy()
+    df.index = MultiIndex.from_tuples([(0, 0), (1, 1)])
+    msg = "number of index levels must be same in `other`"
+    with pytest.raises(ValueError, match=msg):
+        styler.concat(df.style)
diff --git a/pandas/pandas/tests/io/formats/style/test_format.py b/pandas/pandas/tests/io/formats/style/test_format.py
new file mode 100644
index 0000000000000..ae68fcf9ef1fc
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_format.py
@@ -0,0 +1,661 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    IndexSlice,
+    MultiIndex,
+    NaT,
+    Timestamp,
+    option_context,
+)
+
+pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+from pandas.io.formats.style_render import _str_escape
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        data=[[0, -0.609], [1, -1.228]],
+        columns=["A", "B"],
+        index=["x", "y"],
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+@pytest.fixture
+def df_multi():
+    return (
+        DataFrame(
+            data=np.arange(16).reshape(4, 4),
+            columns=MultiIndex.from_product([["A", "B"], ["a", "b"]]),
+            index=MultiIndex.from_product([["X", "Y"], ["x", "y"]]),
+        )
+        .rename_axis(["0_0", "0_1"], axis=0)
+        .rename_axis(["1_0", "1_1"], axis=1)
+    )
+
+
+@pytest.fixture
+def styler_multi(df_multi):
+    return Styler(df_multi, uuid_len=0)
+
+
+def test_display_format(styler):
+    ctx = styler.format("{:0.1f}")._translate(True, True)
+    assert all(["display_value" in c for c in row] for row in ctx["body"])
+    assert all([len(c["display_value"]) <= 3 for c in row[1:]] for row in ctx["body"])
+    assert len(ctx["body"][0][1]["display_value"].lstrip("-")) <= 3
+
+
+@pytest.mark.parametrize("index", [True, False])
+@pytest.mark.parametrize("columns", [True, False])
+def test_display_format_index(styler, index, columns):
+    exp_index = ["x", "y"]
+    if index:
+        styler.format_index(lambda v: v.upper(), axis=0)  # test callable
+        exp_index = ["X", "Y"]
+
+    exp_columns = ["A", "B"]
+    if columns:
+        styler.format_index("*{}*", axis=1)  # test string
+        exp_columns = ["*A*", "*B*"]
+
+    ctx = styler._translate(True, True)
+
+    for r, row in enumerate(ctx["body"]):
+        assert row[0]["display_value"] == exp_index[r]
+
+    for c, col in enumerate(ctx["head"][1:]):
+        assert col["display_value"] == exp_columns[c]
+
+
+def test_format_dict(styler):
+    ctx = styler.format({"A": "{:0.1f}", "B": "{0:.2%}"})._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "0.0"
+    assert ctx["body"][0][2]["display_value"] == "-60.90%"
+
+
+def test_format_index_dict(styler):
+    ctx = styler.format_index({0: lambda v: v.upper()})._translate(True, True)
+    for i, val in enumerate(["X", "Y"]):
+        assert ctx["body"][i][0]["display_value"] == val
+
+
+def test_format_string(styler):
+    ctx = styler.format("{:.2f}")._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "0.00"
+    assert ctx["body"][0][2]["display_value"] == "-0.61"
+    assert ctx["body"][1][1]["display_value"] == "1.00"
+    assert ctx["body"][1][2]["display_value"] == "-1.23"
+
+
+def test_format_callable(styler):
+    ctx = styler.format(lambda v: "neg" if v < 0 else "pos")._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "pos"
+    assert ctx["body"][0][2]["display_value"] == "neg"
+    assert ctx["body"][1][1]["display_value"] == "pos"
+    assert ctx["body"][1][2]["display_value"] == "neg"
+
+
+def test_format_with_na_rep():
+    # GH 21527 28358
+    df = DataFrame([[None, None], [1.1, 1.2]], columns=["A", "B"])
+
+    ctx = df.style.format(None, na_rep="-")._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "-"
+    assert ctx["body"][0][2]["display_value"] == "-"
+
+    ctx = df.style.format("{:.2%}", na_rep="-")._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "-"
+    assert ctx["body"][0][2]["display_value"] == "-"
+    assert ctx["body"][1][1]["display_value"] == "110.00%"
+    assert ctx["body"][1][2]["display_value"] == "120.00%"
+
+    ctx = df.style.format("{:.2%}", na_rep="-", subset=["B"])._translate(True, True)
+    assert ctx["body"][0][2]["display_value"] == "-"
+    assert ctx["body"][1][2]["display_value"] == "120.00%"
+
+
+def test_format_index_with_na_rep():
+    df = DataFrame([[1, 2, 3, 4, 5]], columns=["A", None, np.nan, NaT, NA])
+    ctx = df.style.format_index(None, na_rep="--", axis=1)._translate(True, True)
+    assert ctx["head"][0][1]["display_value"] == "A"
+    for i in [2, 3, 4, 5]:
+        assert ctx["head"][0][i]["display_value"] == "--"
+
+
+def test_format_non_numeric_na():
+    # GH 21527 28358
+    df = DataFrame(
+        {
+            "object": [None, np.nan, "foo"],
+            "datetime": [None, NaT, Timestamp("20120101")],
+        }
+    )
+    ctx = df.style.format(None, na_rep="-")._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "-"
+    assert ctx["body"][0][2]["display_value"] == "-"
+    assert ctx["body"][1][1]["display_value"] == "-"
+    assert ctx["body"][1][2]["display_value"] == "-"
+
+
+@pytest.mark.parametrize(
+    "func, attr, kwargs",
+    [
+        ("format", "_display_funcs", {}),
+        ("format_index", "_display_funcs_index", {"axis": 0}),
+        ("format_index", "_display_funcs_columns", {"axis": 1}),
+    ],
+)
+def test_format_clear(styler, func, attr, kwargs):
+    assert (0, 0) not in getattr(styler, attr)  # using default
+    getattr(styler, func)("{:.2f}", **kwargs)
+    assert (0, 0) in getattr(styler, attr)  # formatter is specified
+    getattr(styler, func)(**kwargs)
+    assert (0, 0) not in getattr(styler, attr)  # formatter cleared to default
+
+
+@pytest.mark.parametrize(
+    "escape, exp",
+    [
+        ("html", "&lt;&gt;&amp;&#34;%$#_{}~^\\~ ^ \\ "),
+        (
+            "latex",
+            '<>\\&"\\%\\$\\#\\_\\{\\}\\textasciitilde \\textasciicircum '
+            "\\textbackslash \\textasciitilde \\space \\textasciicircum \\space "
+            "\\textbackslash \\space ",
+        ),
+    ],
+)
+def test_format_escape_html(escape, exp):
+    chars = '<>&"%$#_{}~^\\~ ^ \\ '
+    df = DataFrame([[chars]])
+
+    s = Styler(df, uuid_len=0).format("&{0}&", escape=None)
+    expected = f'<td id="T__row0_col0" class="data row0 col0" >&{chars}&</td>'
+    assert expected in s.to_html()
+
+    # only the value should be escaped before passing to the formatter
+    s = Styler(df, uuid_len=0).format("&{0}&", escape=escape)
+    expected = f'<td id="T__row0_col0" class="data row0 col0" >&{exp}&</td>'
+    assert expected in s.to_html()
+
+    # also test format_index()
+    styler = Styler(DataFrame(columns=[chars]), uuid_len=0)
+    styler.format_index("&{0}&", escape=None, axis=1)
+    assert styler._translate(True, True)["head"][0][1]["display_value"] == f"&{chars}&"
+    styler.format_index("&{0}&", escape=escape, axis=1)
+    assert styler._translate(True, True)["head"][0][1]["display_value"] == f"&{exp}&"
+
+
+@pytest.mark.parametrize(
+    "chars, expected",
+    [
+        (
+            r"$ \$&%#_{}~^\ $ &%#_{}~^\ $",
+            "".join(
+                [
+                    r"$ \$&%#_{}~^\ $ ",
+                    r"\&\%\#\_\{\}\textasciitilde \textasciicircum ",
+                    r"\textbackslash \space \$",
+                ]
+            ),
+        ),
+        (
+            r"\( &%#_{}~^\ \) &%#_{}~^\ \(",
+            "".join(
+                [
+                    r"\( &%#_{}~^\ \) ",
+                    r"\&\%\#\_\{\}\textasciitilde \textasciicircum ",
+                    r"\textbackslash \space \textbackslash (",
+                ]
+            ),
+        ),
+        (
+            r"$\&%#_{}^\$",
+            r"\$\textbackslash \&\%\#\_\{\}\textasciicircum \textbackslash \$",
+        ),
+        (
+            r"$ \frac{1}{2} $ \( \frac{1}{2} \)",
+            "".join(
+                [
+                    r"$ \frac{1}{2} $",
+                    r" \textbackslash ( \textbackslash frac\{1\}\{2\} \textbackslash )",
+                ]
+            ),
+        ),
+    ],
+)
+def test_format_escape_latex_math(chars, expected):
+    # GH 51903
+    # latex-math escape works for each DataFrame cell separately. If we have
+    # a combination of dollar signs and brackets, the dollar sign would apply.
+    df = DataFrame([[chars]])
+    s = df.style.format("{0}", escape="latex-math")
+    assert s._translate(True, True)["body"][0][1]["display_value"] == expected
+
+
+def test_format_escape_na_rep():
+    # tests the na_rep is not escaped
+    df = DataFrame([['<>&"', None]])
+    s = Styler(df, uuid_len=0).format("X&{0}>X", escape="html", na_rep="&")
+    ex = '<td id="T__row0_col0" class="data row0 col0" >X&&lt;&gt;&amp;&#34;>X</td>'
+    expected2 = '<td id="T__row0_col1" class="data row0 col1" >&</td>'
+    assert ex in s.to_html()
+    assert expected2 in s.to_html()
+
+    # also test for format_index()
+    df = DataFrame(columns=['<>&"', None])
+    styler = Styler(df, uuid_len=0)
+    styler.format_index("X&{0}>X", escape="html", na_rep="&", axis=1)
+    ctx = styler._translate(True, True)
+    assert ctx["head"][0][1]["display_value"] == "X&&lt;&gt;&amp;&#34;>X"
+    assert ctx["head"][0][2]["display_value"] == "&"
+
+
+def test_format_escape_floats(styler):
+    # test given formatter for number format is not impacted by escape
+    s = styler.format("{:.1f}", escape="html")
+    for expected in [">0.0<", ">1.0<", ">-1.2<", ">-0.6<"]:
+        assert expected in s.to_html()
+    # tests precision of floats is not impacted by escape
+    s = styler.format(precision=1, escape="html")
+    for expected in [">0<", ">1<", ">-1.2<", ">-0.6<"]:
+        assert expected in s.to_html()
+
+
+@pytest.mark.parametrize("formatter", [5, True, [2.0]])
+@pytest.mark.parametrize("func", ["format", "format_index"])
+def test_format_raises(styler, formatter, func):
+    with pytest.raises(TypeError, match="expected str or callable"):
+        getattr(styler, func)(formatter)
+
+
+@pytest.mark.parametrize(
+    "precision, expected",
+    [
+        (1, ["1.0", "2.0", "3.2", "4.6"]),
+        (2, ["1.00", "2.01", "3.21", "4.57"]),
+        (3, ["1.000", "2.009", "3.212", "4.566"]),
+    ],
+)
+def test_format_with_precision(precision, expected):
+    # Issue #13257
+    df = DataFrame([[1.0, 2.0090, 3.2121, 4.566]], columns=[1.0, 2.0090, 3.2121, 4.566])
+    styler = Styler(df)
+    styler.format(precision=precision)
+    styler.format_index(precision=precision, axis=1)
+
+    ctx = styler._translate(True, True)
+    for col, exp in enumerate(expected):
+        assert ctx["body"][0][col + 1]["display_value"] == exp  # format test
+        assert ctx["head"][0][col + 1]["display_value"] == exp  # format_index test
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+@pytest.mark.parametrize(
+    "level, expected",
+    [
+        (0, ["X", "X", "_", "_"]),  # level int
+        ("zero", ["X", "X", "_", "_"]),  # level name
+        (1, ["_", "_", "X", "X"]),  # other level int
+        ("one", ["_", "_", "X", "X"]),  # other level name
+        ([0, 1], ["X", "X", "X", "X"]),  # both levels
+        ([0, "zero"], ["X", "X", "_", "_"]),  # level int and name simultaneous
+        ([0, "one"], ["X", "X", "X", "X"]),  # both levels as int and name
+        (["one", "zero"], ["X", "X", "X", "X"]),  # both level names, reversed
+    ],
+)
+def test_format_index_level(axis, level, expected):
+    midx = MultiIndex.from_arrays([["_", "_"], ["_", "_"]], names=["zero", "one"])
+    df = DataFrame([[1, 2], [3, 4]])
+    if axis == 0:
+        df.index = midx
+    else:
+        df.columns = midx
+
+    styler = df.style.format_index(lambda v: "X", level=level, axis=axis)
+    ctx = styler._translate(True, True)
+
+    if axis == 0:  # compare index
+        result = [ctx["body"][s][0]["display_value"] for s in range(2)]
+        result += [ctx["body"][s][1]["display_value"] for s in range(2)]
+    else:  # compare columns
+        result = [ctx["head"][0][s + 1]["display_value"] for s in range(2)]
+        result += [ctx["head"][1][s + 1]["display_value"] for s in range(2)]
+
+    assert expected == result
+
+
+def test_format_subset():
+    df = DataFrame([[0.1234, 0.1234], [1.1234, 1.1234]], columns=["a", "b"])
+    ctx = df.style.format(
+        {"a": "{:0.1f}", "b": "{0:.2%}"}, subset=IndexSlice[0, :]
+    )._translate(True, True)
+    expected = "0.1"
+    raw_11 = "1.123400"
+    assert ctx["body"][0][1]["display_value"] == expected
+    assert ctx["body"][1][1]["display_value"] == raw_11
+    assert ctx["body"][0][2]["display_value"] == "12.34%"
+
+    ctx = df.style.format("{:0.1f}", subset=IndexSlice[0, :])._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == expected
+    assert ctx["body"][1][1]["display_value"] == raw_11
+
+    ctx = df.style.format("{:0.1f}", subset=IndexSlice["a"])._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == expected
+    assert ctx["body"][0][2]["display_value"] == "0.123400"
+
+    ctx = df.style.format("{:0.1f}", subset=IndexSlice[0, "a"])._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == expected
+    assert ctx["body"][1][1]["display_value"] == raw_11
+
+    ctx = df.style.format("{:0.1f}", subset=IndexSlice[[0, 1], ["a"]])._translate(
+        True, True
+    )
+    assert ctx["body"][0][1]["display_value"] == expected
+    assert ctx["body"][1][1]["display_value"] == "1.1"
+    assert ctx["body"][0][2]["display_value"] == "0.123400"
+    assert ctx["body"][1][2]["display_value"] == raw_11
+
+
+@pytest.mark.parametrize("formatter", [None, "{:,.1f}"])
+@pytest.mark.parametrize("decimal", [".", "*"])
+@pytest.mark.parametrize("precision", [None, 2])
+@pytest.mark.parametrize("func, col", [("format", 1), ("format_index", 0)])
+def test_format_thousands(formatter, decimal, precision, func, col):
+    styler = DataFrame([[1000000.123456789]], index=[1000000.123456789]).style
+    result = getattr(styler, func)(  # testing float
+        thousands="_", formatter=formatter, decimal=decimal, precision=precision
+    )._translate(True, True)
+    assert "1_000_000" in result["body"][0][col]["display_value"]
+
+    styler = DataFrame([[1000000]], index=[1000000]).style
+    result = getattr(styler, func)(  # testing int
+        thousands="_", formatter=formatter, decimal=decimal, precision=precision
+    )._translate(True, True)
+    assert "1_000_000" in result["body"][0][col]["display_value"]
+
+    styler = DataFrame([[1 + 1000000.123456789j]], index=[1 + 1000000.123456789j]).style
+    result = getattr(styler, func)(  # testing complex
+        thousands="_", formatter=formatter, decimal=decimal, precision=precision
+    )._translate(True, True)
+    assert "1_000_000" in result["body"][0][col]["display_value"]
+
+
+@pytest.mark.parametrize("formatter", [None, "{:,.4f}"])
+@pytest.mark.parametrize("thousands", [None, ",", "*"])
+@pytest.mark.parametrize("precision", [None, 4])
+@pytest.mark.parametrize("func, col", [("format", 1), ("format_index", 0)])
+def test_format_decimal(formatter, thousands, precision, func, col):
+    styler = DataFrame([[1000000.123456789]], index=[1000000.123456789]).style
+    result = getattr(styler, func)(  # testing float
+        decimal="_", formatter=formatter, thousands=thousands, precision=precision
+    )._translate(True, True)
+    assert "000_123" in result["body"][0][col]["display_value"]
+
+    styler = DataFrame([[1 + 1000000.123456789j]], index=[1 + 1000000.123456789j]).style
+    result = getattr(styler, func)(  # testing complex
+        decimal="_", formatter=formatter, thousands=thousands, precision=precision
+    )._translate(True, True)
+    assert "000_123" in result["body"][0][col]["display_value"]
+
+
+def test_str_escape_error():
+    msg = "`escape` only permitted in {'html', 'latex', 'latex-math'}, got "
+    with pytest.raises(ValueError, match=msg):
+        _str_escape("text", "bad_escape")
+
+    with pytest.raises(ValueError, match=msg):
+        _str_escape("text", [])
+
+    _str_escape(2.00, "bad_escape")  # OK since dtype is float
+
+
+def test_long_int_formatting():
+    df = DataFrame(data=[[1234567890123456789]], columns=["test"])
+    styler = df.style
+    ctx = styler._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "1234567890123456789"
+
+    styler = df.style.format(thousands="_")
+    ctx = styler._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] == "1_234_567_890_123_456_789"
+
+
+def test_format_options():
+    df = DataFrame({"int": [2000, 1], "float": [1.009, None], "str": ["&<", "&~"]})
+    ctx = df.style._translate(True, True)
+
+    # test option: na_rep
+    assert ctx["body"][1][2]["display_value"] == "nan"
+    with option_context("styler.format.na_rep", "MISSING"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][1][2]["display_value"] == "MISSING"
+
+    # test option: decimal and precision
+    assert ctx["body"][0][2]["display_value"] == "1.009000"
+    with option_context("styler.format.decimal", "_"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][0][2]["display_value"] == "1_009000"
+    with option_context("styler.format.precision", 2):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][0][2]["display_value"] == "1.01"
+
+    # test option: thousands
+    assert ctx["body"][0][1]["display_value"] == "2000"
+    with option_context("styler.format.thousands", "_"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][0][1]["display_value"] == "2_000"
+
+    # test option: escape
+    assert ctx["body"][0][3]["display_value"] == "&<"
+    assert ctx["body"][1][3]["display_value"] == "&~"
+    with option_context("styler.format.escape", "html"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][0][3]["display_value"] == "&amp;&lt;"
+    with option_context("styler.format.escape", "latex"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][1][3]["display_value"] == "\\&\\textasciitilde "
+    with option_context("styler.format.escape", "latex-math"):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][1][3]["display_value"] == "\\&\\textasciitilde "
+
+    # test option: formatter
+    with option_context("styler.format.formatter", {"int": "{:,.2f}"}):
+        ctx_with_op = df.style._translate(True, True)
+        assert ctx_with_op["body"][0][1]["display_value"] == "2,000.00"
+
+
+def test_precision_zero(df):
+    styler = Styler(df, precision=0)
+    ctx = styler._translate(True, True)
+    assert ctx["body"][0][2]["display_value"] == "-1"
+    assert ctx["body"][1][2]["display_value"] == "-1"
+
+
+@pytest.mark.parametrize(
+    "formatter, exp",
+    [
+        (lambda x: f"{x:.3f}", "9.000"),
+        ("{:.2f}", "9.00"),
+        ({0: "{:.1f}"}, "9.0"),
+        (None, "9"),
+    ],
+)
+def test_formatter_options_validator(formatter, exp):
+    df = DataFrame([[9]])
+    with option_context("styler.format.formatter", formatter):
+        assert f" {exp} " in df.style.to_latex()
+
+
+def test_formatter_options_raises():
+    msg = "Value must be an instance of"
+    with pytest.raises(ValueError, match=msg):
+        with option_context("styler.format.formatter", ["bad", "type"]):
+            DataFrame().style.to_latex()
+
+
+def test_1level_multiindex():
+    # GH 43383
+    midx = MultiIndex.from_product([[1, 2]], names=[""])
+    df = DataFrame(-1, index=midx, columns=[0, 1])
+    ctx = df.style._translate(True, True)
+    assert ctx["body"][0][0]["display_value"] == "1"
+    assert ctx["body"][0][0]["is_visible"] is True
+    assert ctx["body"][1][0]["display_value"] == "2"
+    assert ctx["body"][1][0]["is_visible"] is True
+
+
+def test_boolean_format():
+    # gh 46384: booleans do not collapse to integer representation on display
+    df = DataFrame([[True, False]])
+    ctx = df.style._translate(True, True)
+    assert ctx["body"][0][1]["display_value"] is True
+    assert ctx["body"][0][2]["display_value"] is False
+
+
+@pytest.mark.parametrize(
+    "hide, labels",
+    [
+        (False, [1, 2]),
+        (True, [1, 2, 3, 4]),
+    ],
+)
+def test_relabel_raise_length(styler_multi, hide, labels):
+    if hide:
+        styler_multi.hide(axis=0, subset=[("X", "x"), ("Y", "y")])
+    with pytest.raises(ValueError, match="``labels`` must be of length equal"):
+        styler_multi.relabel_index(labels=labels)
+
+
+def test_relabel_index(styler_multi):
+    labels = [(1, 2), (3, 4)]
+    styler_multi.hide(axis=0, subset=[("X", "x"), ("Y", "y")])
+    styler_multi.relabel_index(labels=labels)
+    ctx = styler_multi._translate(True, True)
+    assert {"value": "X", "display_value": 1}.items() <= ctx["body"][0][0].items()
+    assert {"value": "y", "display_value": 2}.items() <= ctx["body"][0][1].items()
+    assert {"value": "Y", "display_value": 3}.items() <= ctx["body"][1][0].items()
+    assert {"value": "x", "display_value": 4}.items() <= ctx["body"][1][1].items()
+
+
+def test_relabel_columns(styler_multi):
+    labels = [(1, 2), (3, 4)]
+    styler_multi.hide(axis=1, subset=[("A", "a"), ("B", "b")])
+    styler_multi.relabel_index(axis=1, labels=labels)
+    ctx = styler_multi._translate(True, True)
+    assert {"value": "A", "display_value": 1}.items() <= ctx["head"][0][3].items()
+    assert {"value": "B", "display_value": 3}.items() <= ctx["head"][0][4].items()
+    assert {"value": "b", "display_value": 2}.items() <= ctx["head"][1][3].items()
+    assert {"value": "a", "display_value": 4}.items() <= ctx["head"][1][4].items()
+
+
+def test_relabel_roundtrip(styler):
+    styler.relabel_index(["{}", "{}"])
+    ctx = styler._translate(True, True)
+    assert {"value": "x", "display_value": "x"}.items() <= ctx["body"][0][0].items()
+    assert {"value": "y", "display_value": "y"}.items() <= ctx["body"][1][0].items()
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+@pytest.mark.parametrize(
+    "level, expected",
+    [
+        (0, ["X", "one"]),  # level int
+        ("zero", ["X", "one"]),  # level name
+        (1, ["zero", "X"]),  # other level int
+        ("one", ["zero", "X"]),  # other level name
+        ([0, 1], ["X", "X"]),  # both levels
+        ([0, "zero"], ["X", "one"]),  # level int and name simultaneous
+        ([0, "one"], ["X", "X"]),  # both levels as int and name
+        (["one", "zero"], ["X", "X"]),  # both level names, reversed
+    ],
+)
+def test_format_index_names_level(axis, level, expected):
+    midx = MultiIndex.from_arrays([["_", "_"], ["_", "_"]], names=["zero", "one"])
+    df = DataFrame([[1, 2], [3, 4]])
+    if axis == 0:
+        df.index = midx
+    else:
+        df.columns = midx
+
+    styler = df.style.format_index_names(lambda v: "X", level=level, axis=axis)
+    ctx = styler._translate(True, True)
+
+    if axis == 0:  # compare index
+        result = [ctx["head"][1][s]["display_value"] for s in range(2)]
+    else:  # compare columns
+        result = [ctx["head"][s][0]["display_value"] for s in range(2)]
+    assert expected == result
+
+
+@pytest.mark.parametrize(
+    "attr, kwargs",
+    [
+        ("_display_funcs_index_names", {"axis": 0}),
+        ("_display_funcs_column_names", {"axis": 1}),
+    ],
+)
+def test_format_index_names_clear(styler, attr, kwargs):
+    assert 0 not in getattr(styler, attr)  # using default
+    styler.format_index_names("{:.2f}", **kwargs)
+    assert 0 in getattr(styler, attr)  # formatter is specified
+    styler.format_index_names(**kwargs)
+    assert 0 not in getattr(styler, attr)  # formatter cleared to default
+
+
+@pytest.mark.parametrize("axis", [0, 1])
+def test_format_index_names_callable(styler_multi, axis):
+    ctx = styler_multi.format_index_names(
+        lambda v: v.replace("_", "A"), axis=axis
+    )._translate(True, True)
+    result = [
+        ctx["head"][2][0]["display_value"],
+        ctx["head"][2][1]["display_value"],
+        ctx["head"][0][1]["display_value"],
+        ctx["head"][1][1]["display_value"],
+    ]
+    if axis == 0:
+        expected = ["0A0", "0A1", "1_0", "1_1"]
+    else:
+        expected = ["0_0", "0_1", "1A0", "1A1"]
+    assert result == expected
+
+
+def test_format_index_names_dict(styler_multi):
+    ctx = (
+        styler_multi.format_index_names({"0_0": "{:<<5}"})
+        .format_index_names({"1_1": "{:>>4}"}, axis=1)
+        ._translate(True, True)
+    )
+    assert ctx["head"][2][0]["display_value"] == "0_0<<"
+    assert ctx["head"][1][1]["display_value"] == ">1_1"
+
+
+def test_format_index_names_with_hidden_levels(styler_multi):
+    ctx = styler_multi._translate(True, True)
+    full_head_height = len(ctx["head"])
+    full_head_width = len(ctx["head"][0])
+    assert full_head_height == 3
+    assert full_head_width == 6
+
+    ctx = (
+        styler_multi.hide(axis=0, level=1)
+        .hide(axis=1, level=1)
+        .format_index_names("{:>>4}", axis=1)
+        .format_index_names("{:!<5}")
+        ._translate(True, True)
+    )
+    assert len(ctx["head"]) == full_head_height - 1
+    assert len(ctx["head"][0]) == full_head_width - 1
+    assert ctx["head"][0][0]["display_value"] == ">1_0"
+    assert ctx["head"][1][0]["display_value"] == "0_0!!"
diff --git a/pandas/pandas/tests/io/formats/style/test_highlight.py b/pandas/pandas/tests/io/formats/style/test_highlight.py
new file mode 100644
index 0000000000000..5d19e9c14d534
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_highlight.py
@@ -0,0 +1,219 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    IndexSlice,
+)
+
+pytest.importorskip("jinja2")
+
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture(params=[(None, "float64"), (NA, "Int64")])
+def df(request):
+    # GH 45804
+    return DataFrame(
+        {"A": [0, np.nan, 10], "B": [1, request.param[0], 2]}, dtype=request.param[1]
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+def test_highlight_null(styler):
+    result = styler.highlight_null()._compute().ctx
+    expected = {
+        (1, 0): [("background-color", "red")],
+        (1, 1): [("background-color", "red")],
+    }
+    assert result == expected
+
+
+def test_highlight_null_subset(styler):
+    # GH 31345
+    result = (
+        styler.highlight_null(color="red", subset=["A"])
+        .highlight_null(color="green", subset=["B"])
+        ._compute()
+        .ctx
+    )
+    expected = {
+        (1, 0): [("background-color", "red")],
+        (1, 1): [("background-color", "green")],
+    }
+    assert result == expected
+
+
+@pytest.mark.parametrize("f", ["highlight_min", "highlight_max"])
+def test_highlight_minmax_basic(df, f):
+    expected = {
+        (0, 1): [("background-color", "red")],
+        # ignores NaN row,
+        (2, 0): [("background-color", "red")],
+    }
+    if f == "highlight_min":
+        df = -df
+    result = getattr(df.style, f)(axis=1, color="red")._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize("f", ["highlight_min", "highlight_max"])
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"axis": None, "color": "red"},  # test axis
+        {"axis": 0, "subset": ["A"], "color": "red"},  # test subset and ignores NaN
+        {"axis": None, "props": "background-color: red"},  # test props
+    ],
+)
+def test_highlight_minmax_ext(df, f, kwargs):
+    expected = {(2, 0): [("background-color", "red")]}
+    if f == "highlight_min":
+        df = -df
+    result = getattr(df.style, f)(**kwargs)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize("f", ["highlight_min", "highlight_max"])
+@pytest.mark.parametrize("axis", [None, 0, 1])
+def test_highlight_minmax_nulls(f, axis):
+    # GH 42750
+    expected = {
+        (1, 0): [("background-color", "yellow")],
+        (1, 1): [("background-color", "yellow")],
+    }
+    if axis == 1:
+        expected.update({(2, 1): [("background-color", "yellow")]})
+
+    if f == "highlight_max":
+        df = DataFrame({"a": [NA, 1, None], "b": [np.nan, 1, -1]})
+    else:
+        df = DataFrame({"a": [NA, -1, None], "b": [np.nan, -1, 1]})
+
+    result = getattr(df.style, f)(axis=axis)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"left": 0, "right": 1},  # test basic range
+        {"left": 0, "right": 1, "props": "background-color: yellow"},  # test props
+        {"left": -100, "right": 100, "subset": IndexSlice[[0, 1], :]},  # test subset
+        {"left": 0, "subset": IndexSlice[[0, 1], :]},  # test no right
+        {"right": 1},  # test no left
+        {"left": [0, 0, 11], "axis": 0},  # test left as sequence
+        {"left": DataFrame({"A": [0, 0, 11], "B": [1, 1, 11]}), "axis": None},  # axis
+        {"left": 0, "right": [0, 1], "axis": 1},  # test sequence right
+    ],
+)
+def test_highlight_between(styler, kwargs):
+    expected = {
+        (0, 0): [("background-color", "yellow")],
+        (0, 1): [("background-color", "yellow")],
+    }
+    result = styler.highlight_between(**kwargs)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "arg, map, axis",
+    [
+        ("left", [1, 2], 0),  # 0 axis has 3 elements not 2
+        ("left", [1, 2, 3], 1),  # 1 axis has 2 elements not 3
+        ("left", np.array([[1, 2], [1, 2]]), None),  # df is (2,3) not (2,2)
+        ("right", [1, 2], 0),  # same tests as above for 'right' not 'left'
+        ("right", [1, 2, 3], 1),  # ..
+        ("right", np.array([[1, 2], [1, 2]]), None),  # ..
+    ],
+)
+def test_highlight_between_raises(arg, styler, map, axis):
+    msg = f"supplied '{arg}' is not correct shape"
+    with pytest.raises(ValueError, match=msg):
+        styler.highlight_between(**{arg: map, "axis": axis})._compute()
+
+
+def test_highlight_between_raises2(styler):
+    msg = "values can be 'both', 'left', 'right', or 'neither'"
+    with pytest.raises(ValueError, match=msg):
+        styler.highlight_between(inclusive="badstring")._compute()
+
+    with pytest.raises(ValueError, match=msg):
+        styler.highlight_between(inclusive=1)._compute()
+
+
+@pytest.mark.parametrize(
+    "inclusive, expected",
+    [
+        (
+            "both",
+            {
+                (0, 0): [("background-color", "yellow")],
+                (0, 1): [("background-color", "yellow")],
+            },
+        ),
+        ("neither", {}),
+        ("left", {(0, 0): [("background-color", "yellow")]}),
+        ("right", {(0, 1): [("background-color", "yellow")]}),
+    ],
+)
+def test_highlight_between_inclusive(styler, inclusive, expected):
+    kwargs = {"left": 0, "right": 1, "subset": IndexSlice[[0, 1], :]}
+    result = styler.highlight_between(**kwargs, inclusive=inclusive)._compute()
+    assert result.ctx == expected
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"q_left": 0.5, "q_right": 1, "axis": 0},  # base case
+        {"q_left": 0.5, "q_right": 1, "axis": None},  # test axis
+        {"q_left": 0, "q_right": 1, "subset": IndexSlice[2, :]},  # test subset
+        {"q_left": 0.5, "axis": 0},  # test no high
+        {"q_right": 1, "subset": IndexSlice[2, :], "axis": 1},  # test no low
+        {"q_left": 0.5, "axis": 0, "props": "background-color: yellow"},  # tst prop
+    ],
+)
+def test_highlight_quantile(styler, kwargs):
+    expected = {
+        (2, 0): [("background-color", "yellow")],
+        (2, 1): [("background-color", "yellow")],
+    }
+    result = styler.highlight_quantile(**kwargs)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "f,kwargs",
+    [
+        ("highlight_min", {"axis": 1, "subset": IndexSlice[1, :]}),
+        ("highlight_max", {"axis": 0, "subset": [0]}),
+        ("highlight_quantile", {"axis": None, "q_left": 0.6, "q_right": 0.8}),
+        ("highlight_between", {"subset": [0]}),
+    ],
+)
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        int,
+        float,
+        "datetime64[ns]",
+        str,
+        "timedelta64[ns]",
+    ],
+)
+def test_all_highlight_dtypes(f, kwargs, dtype):
+    df = DataFrame([[0, 10], [20, 30]], dtype=dtype)
+    if f == "highlight_quantile" and isinstance(df.iloc[0, 0], (str)):
+        return None  # quantile incompatible with str
+    if f == "highlight_between":
+        kwargs["left"] = df.iloc[1, 0]  # set the range low for testing
+
+    expected = {(1, 0): [("background-color", "yellow")]}
+    result = getattr(df.style, f)(**kwargs)._compute().ctx
+    assert result == expected
diff --git a/pandas/pandas/tests/io/formats/style/test_html.py b/pandas/pandas/tests/io/formats/style/test_html.py
new file mode 100644
index 0000000000000..752b9b391f9cb
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_html.py
@@ -0,0 +1,1034 @@
+from textwrap import (
+    dedent,
+    indent,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    option_context,
+)
+
+jinja2 = pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def env():
+    loader = jinja2.PackageLoader("pandas", "io/formats/templates")
+    env = jinja2.Environment(loader=loader, trim_blocks=True)
+    return env
+
+
+@pytest.fixture
+def styler():
+    return Styler(DataFrame([[2.61], [2.69]], index=["a", "b"], columns=["A"]))
+
+
+@pytest.fixture
+def styler_mi():
+    midx = MultiIndex.from_product([["a", "b"], ["c", "d"]])
+    return Styler(DataFrame(np.arange(16).reshape(4, 4), index=midx, columns=midx))
+
+
+@pytest.fixture
+def styler_multi():
+    df = DataFrame(
+        data=np.arange(16).reshape(4, 4),
+        columns=MultiIndex.from_product([["A", "B"], ["a", "b"]], names=["A&", "b&"]),
+        index=MultiIndex.from_product([["X", "Y"], ["x", "y"]], names=["X>", "y_"]),
+    )
+    return Styler(df)
+
+
+@pytest.fixture
+def tpl_style(env):
+    return env.get_template("html_style.tpl")
+
+
+@pytest.fixture
+def tpl_table(env):
+    return env.get_template("html_table.tpl")
+
+
+def test_html_template_extends_options():
+    # make sure if templates are edited tests are updated as are setup fixtures
+    # to understand the dependency
+    with open("pandas/io/formats/templates/html.tpl", encoding="utf-8") as file:
+        result = file.read()
+    assert "{% include html_style_tpl %}" in result
+    assert "{% include html_table_tpl %}" in result
+
+
+def test_exclude_styles(styler):
+    result = styler.to_html(exclude_styles=True, doctype_html=True)
+    expected = dedent(
+        """\
+        <!DOCTYPE html>
+        <html>
+        <head>
+        <meta charset="utf-8">
+        </head>
+        <body>
+        <table>
+          <thead>
+            <tr>
+              <th >&nbsp;</th>
+              <th >A</th>
+            </tr>
+          </thead>
+          <tbody>
+            <tr>
+              <th >a</th>
+              <td >2.610000</td>
+            </tr>
+            <tr>
+              <th >b</th>
+              <td >2.690000</td>
+            </tr>
+          </tbody>
+        </table>
+        </body>
+        </html>
+        """
+    )
+    assert result == expected
+
+
+def test_w3_html_format(styler):
+    styler.set_uuid("").set_table_styles([{"selector": "th", "props": "att2:v2;"}]).map(
+        lambda x: "att1:v1;"
+    ).set_table_attributes('class="my-cls1" style="attr3:v3;"').set_td_classes(
+        DataFrame(["my-cls2"], index=["a"], columns=["A"])
+    ).format("{:.1f}").set_caption("A comprehensive test")
+    expected = dedent(
+        """\
+        <style type="text/css">
+        #T_ th {
+          att2: v2;
+        }
+        #T__row0_col0, #T__row1_col0 {
+          att1: v1;
+        }
+        </style>
+        <table id="T_" class="my-cls1" style="attr3:v3;">
+          <caption>A comprehensive test</caption>
+          <thead>
+            <tr>
+              <th class="blank level0" >&nbsp;</th>
+              <th id="T__level0_col0" class="col_heading level0 col0" >A</th>
+            </tr>
+          </thead>
+          <tbody>
+            <tr>
+              <th id="T__level0_row0" class="row_heading level0 row0" >a</th>
+              <td id="T__row0_col0" class="data row0 col0 my-cls2" >2.6</td>
+            </tr>
+            <tr>
+              <th id="T__level0_row1" class="row_heading level0 row1" >b</th>
+              <td id="T__row1_col0" class="data row1 col0" >2.7</td>
+            </tr>
+          </tbody>
+        </table>
+        """
+    )
+    assert expected == styler.to_html()
+
+
+def test_colspan_w3():
+    # GH 36223
+    df = DataFrame(data=[[1, 2]], columns=[["l0", "l0"], ["l1a", "l1b"]])
+    styler = Styler(df, uuid="_", cell_ids=False)
+    assert '<th class="col_heading level0 col0" colspan="2">l0</th>' in styler.to_html()
+
+
+def test_rowspan_w3():
+    # GH 38533
+    df = DataFrame(data=[[1, 2]], index=[["l0", "l0"], ["l1a", "l1b"]])
+    styler = Styler(df, uuid="_", cell_ids=False)
+    assert '<th class="row_heading level0 row0" rowspan="2">l0</th>' in styler.to_html()
+
+
+def test_styles(styler):
+    styler.set_uuid("abc")
+    styler.set_table_styles([{"selector": "td", "props": "color: red;"}])
+    result = styler.to_html(doctype_html=True)
+    expected = dedent(
+        """\
+        <!DOCTYPE html>
+        <html>
+        <head>
+        <meta charset="utf-8">
+        <style type="text/css">
+        #T_abc td {
+          color: red;
+        }
+        </style>
+        </head>
+        <body>
+        <table id="T_abc">
+          <thead>
+            <tr>
+              <th class="blank level0" >&nbsp;</th>
+              <th id="T_abc_level0_col0" class="col_heading level0 col0" >A</th>
+            </tr>
+          </thead>
+          <tbody>
+            <tr>
+              <th id="T_abc_level0_row0" class="row_heading level0 row0" >a</th>
+              <td id="T_abc_row0_col0" class="data row0 col0" >2.610000</td>
+            </tr>
+            <tr>
+              <th id="T_abc_level0_row1" class="row_heading level0 row1" >b</th>
+              <td id="T_abc_row1_col0" class="data row1 col0" >2.690000</td>
+            </tr>
+          </tbody>
+        </table>
+        </body>
+        </html>
+        """
+    )
+    assert result == expected
+
+
+def test_doctype(styler):
+    result = styler.to_html(doctype_html=False)
+    assert "<html>" not in result
+    assert "<body>" not in result
+    assert "<!DOCTYPE html>" not in result
+    assert "<head>" not in result
+
+
+def test_doctype_encoding(styler):
+    with option_context("styler.render.encoding", "ASCII"):
+        result = styler.to_html(doctype_html=True)
+        assert '<meta charset="ASCII">' in result
+        result = styler.to_html(doctype_html=True, encoding="ANSI")
+        assert '<meta charset="ANSI">' in result
+
+
+def test_bold_headers_arg(styler):
+    result = styler.to_html(bold_headers=True)
+    assert "th {\n  font-weight: bold;\n}" in result
+    result = styler.to_html()
+    assert "th {\n  font-weight: bold;\n}" not in result
+
+
+def test_caption_arg(styler):
+    result = styler.to_html(caption="foo bar")
+    assert "<caption>foo bar</caption>" in result
+    result = styler.to_html()
+    assert "<caption>foo bar</caption>" not in result
+
+
+def test_block_names(tpl_style, tpl_table):
+    # catch accidental removal of a block
+    expected_style = {
+        "before_style",
+        "style",
+        "table_styles",
+        "before_cellstyle",
+        "cellstyle",
+    }
+    expected_table = {
+        "before_table",
+        "table",
+        "caption",
+        "thead",
+        "tbody",
+        "after_table",
+        "before_head_rows",
+        "head_tr",
+        "after_head_rows",
+        "before_rows",
+        "tr",
+        "after_rows",
+    }
+    result1 = set(tpl_style.blocks)
+    assert result1 == expected_style
+
+    result2 = set(tpl_table.blocks)
+    assert result2 == expected_table
+
+
+def test_from_custom_template_table(tmpdir):
+    p = tmpdir.mkdir("tpl").join("myhtml_table.tpl")
+    p.write(
+        dedent(
+            """\
+            {% extends "html_table.tpl" %}
+            {% block table %}
+            <h1>{{custom_title}}</h1>
+            {{ super() }}
+            {% endblock table %}"""
+        )
+    )
+    result = Styler.from_custom_template(str(tmpdir.join("tpl")), "myhtml_table.tpl")
+    assert issubclass(result, Styler)
+    assert result.env is not Styler.env
+    assert result.template_html_table is not Styler.template_html_table
+    styler = result(DataFrame({"A": [1, 2]}))
+    assert "<h1>My Title</h1>\n\n\n<table" in styler.to_html(custom_title="My Title")
+
+
+def test_from_custom_template_style(tmpdir):
+    p = tmpdir.mkdir("tpl").join("myhtml_style.tpl")
+    p.write(
+        dedent(
+            """\
+            {% extends "html_style.tpl" %}
+            {% block style %}
+            <link rel="stylesheet" href="mystyle.css">
+            {{ super() }}
+            {% endblock style %}"""
+        )
+    )
+    result = Styler.from_custom_template(
+        str(tmpdir.join("tpl")), html_style="myhtml_style.tpl"
+    )
+    assert issubclass(result, Styler)
+    assert result.env is not Styler.env
+    assert result.template_html_style is not Styler.template_html_style
+    styler = result(DataFrame({"A": [1, 2]}))
+    assert '<link rel="stylesheet" href="mystyle.css">\n\n<style' in styler.to_html()
+
+
+def test_caption_as_sequence(styler):
+    styler.set_caption(("full cap", "short cap"))
+    assert "<caption>full cap</caption>" in styler.to_html()
+
+
+@pytest.mark.parametrize("index", [False, True])
+@pytest.mark.parametrize("columns", [False, True])
+@pytest.mark.parametrize("index_name", [True, False])
+def test_sticky_basic(styler, index, columns, index_name):
+    if index_name:
+        styler.index.name = "some text"
+    if index:
+        styler.set_sticky(axis=0)
+    if columns:
+        styler.set_sticky(axis=1)
+
+    left_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  left: 0px;\n  z-index: {1};\n}}"
+    )
+    top_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  top: {1}px;\n  z-index: {2};\n{3}}}"
+    )
+
+    res = styler.set_uuid("").to_html()
+
+    # test index stickys over thead and tbody
+    assert (left_css.format("thead tr th:nth-child(1)", "3 !important") in res) is index
+    assert (left_css.format("tbody tr th:nth-child(1)", "1") in res) is index
+
+    # test column stickys including if name row
+    assert (
+        top_css.format("thead tr:nth-child(1) th", "0", "2", "  height: 25px;\n") in res
+    ) is (columns and index_name)
+    assert (
+        top_css.format("thead tr:nth-child(2) th", "25", "2", "  height: 25px;\n")
+        in res
+    ) is (columns and index_name)
+    assert (top_css.format("thead tr:nth-child(1) th", "0", "2", "") in res) is (
+        columns and not index_name
+    )
+
+
+@pytest.mark.parametrize("index", [False, True])
+@pytest.mark.parametrize("columns", [False, True])
+def test_sticky_mi(styler_mi, index, columns):
+    if index:
+        styler_mi.set_sticky(axis=0)
+    if columns:
+        styler_mi.set_sticky(axis=1)
+
+    left_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  left: {1}px;\n  min-width: 75px;\n  max-width: 75px;\n  z-index: {2};\n}}"
+    )
+    top_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  top: {1}px;\n  height: 25px;\n  z-index: {2};\n}}"
+    )
+
+    res = styler_mi.set_uuid("").to_html()
+
+    # test the index stickys for thead and tbody over both levels
+    assert (
+        left_css.format("thead tr th:nth-child(1)", "0", "3 !important") in res
+    ) is index
+    assert (left_css.format("tbody tr th.level0", "0", "1") in res) is index
+    assert (
+        left_css.format("thead tr th:nth-child(2)", "75", "3 !important") in res
+    ) is index
+    assert (left_css.format("tbody tr th.level1", "75", "1") in res) is index
+
+    # test the column stickys for each level row
+    assert (top_css.format("thead tr:nth-child(1) th", "0", "2") in res) is columns
+    assert (top_css.format("thead tr:nth-child(2) th", "25", "2") in res) is columns
+
+
+@pytest.mark.parametrize("index", [False, True])
+@pytest.mark.parametrize("columns", [False, True])
+@pytest.mark.parametrize("levels", [[1], ["one"], "one"])
+def test_sticky_levels(styler_mi, index, columns, levels):
+    styler_mi.index.names, styler_mi.columns.names = ["zero", "one"], ["zero", "one"]
+    if index:
+        styler_mi.set_sticky(axis=0, levels=levels)
+    if columns:
+        styler_mi.set_sticky(axis=1, levels=levels)
+
+    left_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  left: {1}px;\n  min-width: 75px;\n  max-width: 75px;\n  z-index: {2};\n}}"
+    )
+    top_css = (
+        "#T_ {0} {{\n  position: sticky;\n  background-color: inherit;\n"
+        "  top: {1}px;\n  height: 25px;\n  z-index: {2};\n}}"
+    )
+
+    res = styler_mi.set_uuid("").to_html()
+
+    # test no sticking of level0
+    assert "#T_ thead tr th:nth-child(1)" not in res
+    assert "#T_ tbody tr th.level0" not in res
+    assert "#T_ thead tr:nth-child(1) th" not in res
+
+    # test sticking level1
+    assert (
+        left_css.format("thead tr th:nth-child(2)", "0", "3 !important") in res
+    ) is index
+    assert (left_css.format("tbody tr th.level1", "0", "1") in res) is index
+    assert (top_css.format("thead tr:nth-child(2) th", "0", "2") in res) is columns
+
+
+def test_sticky_raises(styler):
+    with pytest.raises(ValueError, match="No axis named bad for object type DataFrame"):
+        styler.set_sticky(axis="bad")
+
+
+@pytest.mark.parametrize(
+    "sparse_index, sparse_columns",
+    [(True, True), (True, False), (False, True), (False, False)],
+)
+def test_sparse_options(sparse_index, sparse_columns):
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=ridx, columns=cidx)
+    styler = df.style
+
+    default_html = styler.to_html()  # defaults under pd.options to (True , True)
+
+    with option_context(
+        "styler.sparse.index", sparse_index, "styler.sparse.columns", sparse_columns
+    ):
+        html1 = styler.to_html()
+        assert (html1 == default_html) is (sparse_index and sparse_columns)
+    html2 = styler.to_html(sparse_index=sparse_index, sparse_columns=sparse_columns)
+    assert html1 == html2
+
+
+@pytest.mark.parametrize("index", [True, False])
+@pytest.mark.parametrize("columns", [True, False])
+def test_map_header_cell_ids(styler, index, columns):
+    # GH 41893
+    func = lambda v: "attr: val;"
+    styler.uuid, styler.cell_ids = "", False
+    if index:
+        styler.map_index(func, axis="index")
+    if columns:
+        styler.map_index(func, axis="columns")
+
+    result = styler.to_html()
+
+    # test no data cell ids
+    assert '<td class="data row0 col0" >2.610000</td>' in result
+    assert '<td class="data row1 col0" >2.690000</td>' in result
+
+    # test index header ids where needed and css styles
+    assert (
+        '<th id="T__level0_row0" class="row_heading level0 row0" >a</th>' in result
+    ) is index
+    assert (
+        '<th id="T__level0_row1" class="row_heading level0 row1" >b</th>' in result
+    ) is index
+    assert ("#T__level0_row0, #T__level0_row1 {\n  attr: val;\n}" in result) is index
+
+    # test column header ids where needed and css styles
+    assert (
+        '<th id="T__level0_col0" class="col_heading level0 col0" >A</th>' in result
+    ) is columns
+    assert ("#T__level0_col0 {\n  attr: val;\n}" in result) is columns
+
+
+@pytest.mark.parametrize("rows", [True, False])
+@pytest.mark.parametrize("cols", [True, False])
+def test_maximums(styler_mi, rows, cols):
+    result = styler_mi.to_html(
+        max_rows=2 if rows else None,
+        max_columns=2 if cols else None,
+    )
+
+    assert ">5</td>" in result  # [[0,1], [4,5]] always visible
+    assert (">8</td>" in result) is not rows  # first trimmed vertical element
+    assert (">2</td>" in result) is not cols  # first trimmed horizontal element
+
+
+def test_replaced_css_class_names():
+    css = {
+        "row_heading": "ROWHEAD",
+        # "col_heading": "COLHEAD",
+        "index_name": "IDXNAME",
+        # "col": "COL",
+        "row": "ROW",
+        # "col_trim": "COLTRIM",
+        "row_trim": "ROWTRIM",
+        "level": "LEVEL",
+        "data": "DATA",
+        "blank": "BLANK",
+    }
+    midx = MultiIndex.from_product([["a", "b"], ["c", "d"]])
+    styler_mi = Styler(
+        DataFrame(np.arange(16).reshape(4, 4), index=midx, columns=midx),
+        uuid_len=0,
+    ).set_table_styles(css_class_names=css)
+    styler_mi.index.names = ["n1", "n2"]
+    styler_mi.hide(styler_mi.index[1:], axis=0)
+    styler_mi.hide(styler_mi.columns[1:], axis=1)
+    styler_mi.map_index(lambda v: "color: red;", axis=0)
+    styler_mi.map_index(lambda v: "color: green;", axis=1)
+    styler_mi.map(lambda v: "color: blue;")
+    expected = dedent(
+        """\
+    <style type="text/css">
+    #T__ROW0_col0 {
+      color: blue;
+    }
+    #T__LEVEL0_ROW0, #T__LEVEL1_ROW0 {
+      color: red;
+    }
+    #T__LEVEL0_col0, #T__LEVEL1_col0 {
+      color: green;
+    }
+    </style>
+    <table id="T_">
+      <thead>
+        <tr>
+          <th class="BLANK" >&nbsp;</th>
+          <th class="IDXNAME LEVEL0" >n1</th>
+          <th id="T__LEVEL0_col0" class="col_heading LEVEL0 col0" >a</th>
+        </tr>
+        <tr>
+          <th class="BLANK" >&nbsp;</th>
+          <th class="IDXNAME LEVEL1" >n2</th>
+          <th id="T__LEVEL1_col0" class="col_heading LEVEL1 col0" >c</th>
+        </tr>
+        <tr>
+          <th class="IDXNAME LEVEL0" >n1</th>
+          <th class="IDXNAME LEVEL1" >n2</th>
+          <th class="BLANK col0" >&nbsp;</th>
+        </tr>
+      </thead>
+      <tbody>
+        <tr>
+          <th id="T__LEVEL0_ROW0" class="ROWHEAD LEVEL0 ROW0" >a</th>
+          <th id="T__LEVEL1_ROW0" class="ROWHEAD LEVEL1 ROW0" >c</th>
+          <td id="T__ROW0_col0" class="DATA ROW0 col0" >0</td>
+        </tr>
+      </tbody>
+    </table>
+    """
+    )
+    result = styler_mi.to_html()
+    assert result == expected
+
+
+def test_include_css_style_rules_only_for_visible_cells(styler_mi):
+    # GH 43619
+    result = (
+        styler_mi.set_uuid("")
+        .map(lambda v: "color: blue;")
+        .hide(styler_mi.data.columns[1:], axis="columns")
+        .hide(styler_mi.data.index[1:], axis="index")
+        .to_html()
+    )
+    expected_styles = dedent(
+        """\
+        <style type="text/css">
+        #T__row0_col0 {
+          color: blue;
+        }
+        </style>
+        """
+    )
+    assert expected_styles in result
+
+
+def test_include_css_style_rules_only_for_visible_index_labels(styler_mi):
+    # GH 43619
+    result = (
+        styler_mi.set_uuid("")
+        .map_index(lambda v: "color: blue;", axis="index")
+        .hide(styler_mi.data.columns, axis="columns")
+        .hide(styler_mi.data.index[1:], axis="index")
+        .to_html()
+    )
+    expected_styles = dedent(
+        """\
+        <style type="text/css">
+        #T__level0_row0, #T__level1_row0 {
+          color: blue;
+        }
+        </style>
+        """
+    )
+    assert expected_styles in result
+
+
+def test_include_css_style_rules_only_for_visible_column_labels(styler_mi):
+    # GH 43619
+    result = (
+        styler_mi.set_uuid("")
+        .map_index(lambda v: "color: blue;", axis="columns")
+        .hide(styler_mi.data.columns[1:], axis="columns")
+        .hide(styler_mi.data.index, axis="index")
+        .to_html()
+    )
+    expected_styles = dedent(
+        """\
+        <style type="text/css">
+        #T__level0_col0, #T__level1_col0 {
+          color: blue;
+        }
+        </style>
+        """
+    )
+    assert expected_styles in result
+
+
+def test_hiding_index_columns_multiindex_alignment():
+    # gh 43644
+    midx = MultiIndex.from_product(
+        [["i0", "j0"], ["i1"], ["i2", "j2"]], names=["i-0", "i-1", "i-2"]
+    )
+    cidx = MultiIndex.from_product(
+        [["c0"], ["c1", "d1"], ["c2", "d2"]], names=["c-0", "c-1", "c-2"]
+    )
+    df = DataFrame(np.arange(16).reshape(4, 4), index=midx, columns=cidx)
+    styler = Styler(df, uuid_len=0)
+    styler.hide(level=1, axis=0).hide(level=0, axis=1)
+    styler.hide([("j0", "i1", "j2")], axis=0)
+    styler.hide([("c0", "d1", "d2")], axis=1)
+    result = styler.to_html()
+    expected = dedent(
+        """\
+    <style type="text/css">
+    </style>
+    <table id="T_">
+      <thead>
+        <tr>
+          <th class="blank" >&nbsp;</th>
+          <th class="index_name level1" >c-1</th>
+          <th id="T__level1_col0" class="col_heading level1 col0" colspan="2">c1</th>
+          <th id="T__level1_col2" class="col_heading level1 col2" >d1</th>
+        </tr>
+        <tr>
+          <th class="blank" >&nbsp;</th>
+          <th class="index_name level2" >c-2</th>
+          <th id="T__level2_col0" class="col_heading level2 col0" >c2</th>
+          <th id="T__level2_col1" class="col_heading level2 col1" >d2</th>
+          <th id="T__level2_col2" class="col_heading level2 col2" >c2</th>
+        </tr>
+        <tr>
+          <th class="index_name level0" >i-0</th>
+          <th class="index_name level2" >i-2</th>
+          <th class="blank col0" >&nbsp;</th>
+          <th class="blank col1" >&nbsp;</th>
+          <th class="blank col2" >&nbsp;</th>
+        </tr>
+      </thead>
+      <tbody>
+        <tr>
+          <th id="T__level0_row0" class="row_heading level0 row0" rowspan="2">i0</th>
+          <th id="T__level2_row0" class="row_heading level2 row0" >i2</th>
+          <td id="T__row0_col0" class="data row0 col0" >0</td>
+          <td id="T__row0_col1" class="data row0 col1" >1</td>
+          <td id="T__row0_col2" class="data row0 col2" >2</td>
+        </tr>
+        <tr>
+          <th id="T__level2_row1" class="row_heading level2 row1" >j2</th>
+          <td id="T__row1_col0" class="data row1 col0" >4</td>
+          <td id="T__row1_col1" class="data row1 col1" >5</td>
+          <td id="T__row1_col2" class="data row1 col2" >6</td>
+        </tr>
+        <tr>
+          <th id="T__level0_row2" class="row_heading level0 row2" >j0</th>
+          <th id="T__level2_row2" class="row_heading level2 row2" >i2</th>
+          <td id="T__row2_col0" class="data row2 col0" >8</td>
+          <td id="T__row2_col1" class="data row2 col1" >9</td>
+          <td id="T__row2_col2" class="data row2 col2" >10</td>
+        </tr>
+      </tbody>
+    </table>
+    """
+    )
+    assert result == expected
+
+
+def test_hiding_index_columns_multiindex_trimming():
+    # gh 44272
+    df = DataFrame(np.arange(64).reshape(8, 8))
+    df.columns = MultiIndex.from_product([[0, 1, 2, 3], [0, 1]])
+    df.index = MultiIndex.from_product([[0, 1, 2, 3], [0, 1]])
+    df.index.names, df.columns.names = ["a", "b"], ["c", "d"]
+    styler = Styler(df, cell_ids=False, uuid_len=0)
+    styler.hide([(0, 0), (0, 1), (1, 0)], axis=1).hide([(0, 0), (0, 1), (1, 0)], axis=0)
+    with option_context("styler.render.max_rows", 4, "styler.render.max_columns", 4):
+        result = styler.to_html()
+
+    expected = dedent(
+        """\
+    <style type="text/css">
+    </style>
+    <table id="T_">
+      <thead>
+        <tr>
+          <th class="blank" >&nbsp;</th>
+          <th class="index_name level0" >c</th>
+          <th class="col_heading level0 col3" >1</th>
+          <th class="col_heading level0 col4" colspan="2">2</th>
+          <th class="col_heading level0 col6" >3</th>
+        </tr>
+        <tr>
+          <th class="blank" >&nbsp;</th>
+          <th class="index_name level1" >d</th>
+          <th class="col_heading level1 col3" >1</th>
+          <th class="col_heading level1 col4" >0</th>
+          <th class="col_heading level1 col5" >1</th>
+          <th class="col_heading level1 col6" >0</th>
+          <th class="col_heading level1 col_trim" >...</th>
+        </tr>
+        <tr>
+          <th class="index_name level0" >a</th>
+          <th class="index_name level1" >b</th>
+          <th class="blank col3" >&nbsp;</th>
+          <th class="blank col4" >&nbsp;</th>
+          <th class="blank col5" >&nbsp;</th>
+          <th class="blank col6" >&nbsp;</th>
+          <th class="blank col7 col_trim" >&nbsp;</th>
+        </tr>
+      </thead>
+      <tbody>
+        <tr>
+          <th class="row_heading level0 row3" >1</th>
+          <th class="row_heading level1 row3" >1</th>
+          <td class="data row3 col3" >27</td>
+          <td class="data row3 col4" >28</td>
+          <td class="data row3 col5" >29</td>
+          <td class="data row3 col6" >30</td>
+          <td class="data row3 col_trim" >...</td>
+        </tr>
+        <tr>
+          <th class="row_heading level0 row4" rowspan="2">2</th>
+          <th class="row_heading level1 row4" >0</th>
+          <td class="data row4 col3" >35</td>
+          <td class="data row4 col4" >36</td>
+          <td class="data row4 col5" >37</td>
+          <td class="data row4 col6" >38</td>
+          <td class="data row4 col_trim" >...</td>
+        </tr>
+        <tr>
+          <th class="row_heading level1 row5" >1</th>
+          <td class="data row5 col3" >43</td>
+          <td class="data row5 col4" >44</td>
+          <td class="data row5 col5" >45</td>
+          <td class="data row5 col6" >46</td>
+          <td class="data row5 col_trim" >...</td>
+        </tr>
+        <tr>
+          <th class="row_heading level0 row6" >3</th>
+          <th class="row_heading level1 row6" >0</th>
+          <td class="data row6 col3" >51</td>
+          <td class="data row6 col4" >52</td>
+          <td class="data row6 col5" >53</td>
+          <td class="data row6 col6" >54</td>
+          <td class="data row6 col_trim" >...</td>
+        </tr>
+        <tr>
+          <th class="row_heading level0 row_trim" >...</th>
+          <th class="row_heading level1 row_trim" >...</th>
+          <td class="data col3 row_trim" >...</td>
+          <td class="data col4 row_trim" >...</td>
+          <td class="data col5 row_trim" >...</td>
+          <td class="data col6 row_trim" >...</td>
+          <td class="data row_trim col_trim" >...</td>
+        </tr>
+      </tbody>
+    </table>
+    """
+    )
+
+    assert result == expected
+
+
+@pytest.mark.parametrize("type", ["data", "index"])
+@pytest.mark.parametrize(
+    "text, exp, found",
+    [
+        ("no link, just text", False, ""),
+        ("subdomain not www: sub.web.com", False, ""),
+        ("www subdomain: www.web.com other", True, "www.web.com"),
+        ("scheme full structure: http://www.web.com", True, "http://www.web.com"),
+        ("scheme no top-level: http://www.web", True, "http://www.web"),
+        ("no scheme, no top-level: www.web", False, "www.web"),
+        ("https scheme: https://www.web.com", True, "https://www.web.com"),
+        ("ftp scheme: ftp://www.web", True, "ftp://www.web"),
+        ("ftps scheme: ftps://www.web", True, "ftps://www.web"),
+        ("subdirectories: www.web.com/directory", True, "www.web.com/directory"),
+        ("Multiple domains: www.1.2.3.4", True, "www.1.2.3.4"),
+        ("with port: http://web.com:80", True, "http://web.com:80"),
+        (
+            "full net_loc scheme: http://user:pass@web.com",
+            True,
+            "http://user:pass@web.com",
+        ),
+        (
+            "with valid special chars: http://web.com/,.':;~!@#$*()[]",
+            True,
+            "http://web.com/,.':;~!@#$*()[]",
+        ),
+    ],
+)
+def test_rendered_links(type, text, exp, found):
+    if type == "data":
+        df = DataFrame([text])
+        styler = df.style.format(hyperlinks="html")
+    else:
+        df = DataFrame([0], index=[text])
+        styler = df.style.format_index(hyperlinks="html")
+
+    rendered = f'<a href="{found}" target="_blank">{found}</a>'
+    result = styler.to_html()
+    assert (rendered in result) is exp
+    assert (text in result) is not exp  # test conversion done when expected and not
+
+
+def test_multiple_rendered_links():
+    links = ("www.a.b", "http://a.c", "https://a.d", "ftp://a.e")
+    df = DataFrame(["text {} {} text {} {}".format(*links)])
+    result = df.style.format(hyperlinks="html").to_html()
+    href = '<a href="{0}" target="_blank">{0}</a>'
+    for link in links:
+        assert href.format(link) in result
+    assert href.format("text") not in result
+
+
+def test_concat(styler):
+    other = styler.data.agg(["mean"]).style
+    styler.concat(other).set_uuid("X")
+    result = styler.to_html()
+    fp = "foot0_"
+    expected = dedent(
+        f"""\
+    <tr>
+      <th id="T_X_level0_row1" class="row_heading level0 row1" >b</th>
+      <td id="T_X_row1_col0" class="data row1 col0" >2.690000</td>
+    </tr>
+    <tr>
+      <th id="T_X_level0_{fp}row0" class="{fp}row_heading level0 {fp}row0" >mean</th>
+      <td id="T_X_{fp}row0_col0" class="{fp}data {fp}row0 col0" >2.650000</td>
+    </tr>
+  </tbody>
+</table>
+    """
+    )
+    assert expected in result
+
+
+def test_concat_recursion(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["mean"]), precision=3)
+    styler3 = Styler(df.agg(["mean"]), precision=4)
+    styler1.concat(styler2.concat(styler3)).set_uuid("X")
+    result = styler.to_html()
+    # notice that the second concat (last <tr> of the output html),
+    # there are two `foot_` in the id and class
+    fp1 = "foot0_"
+    fp2 = "foot0_foot0_"
+    expected = dedent(
+        f"""\
+    <tr>
+      <th id="T_X_level0_row1" class="row_heading level0 row1" >b</th>
+      <td id="T_X_row1_col0" class="data row1 col0" >2.690000</td>
+    </tr>
+    <tr>
+      <th id="T_X_level0_{fp1}row0" class="{fp1}row_heading level0 {fp1}row0" >mean</th>
+      <td id="T_X_{fp1}row0_col0" class="{fp1}data {fp1}row0 col0" >2.650</td>
+    </tr>
+    <tr>
+      <th id="T_X_level0_{fp2}row0" class="{fp2}row_heading level0 {fp2}row0" >mean</th>
+      <td id="T_X_{fp2}row0_col0" class="{fp2}data {fp2}row0 col0" >2.6500</td>
+    </tr>
+  </tbody>
+</table>
+    """
+    )
+    assert expected in result
+
+
+def test_concat_chain(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["mean"]), precision=3)
+    styler3 = Styler(df.agg(["mean"]), precision=4)
+    styler1.concat(styler2).concat(styler3).set_uuid("X")
+    result = styler.to_html()
+    fp1 = "foot0_"
+    fp2 = "foot1_"
+    expected = dedent(
+        f"""\
+    <tr>
+      <th id="T_X_level0_row1" class="row_heading level0 row1" >b</th>
+      <td id="T_X_row1_col0" class="data row1 col0" >2.690000</td>
+    </tr>
+    <tr>
+      <th id="T_X_level0_{fp1}row0" class="{fp1}row_heading level0 {fp1}row0" >mean</th>
+      <td id="T_X_{fp1}row0_col0" class="{fp1}data {fp1}row0 col0" >2.650</td>
+    </tr>
+    <tr>
+      <th id="T_X_level0_{fp2}row0" class="{fp2}row_heading level0 {fp2}row0" >mean</th>
+      <td id="T_X_{fp2}row0_col0" class="{fp2}data {fp2}row0 col0" >2.6500</td>
+    </tr>
+  </tbody>
+</table>
+    """
+    )
+    assert expected in result
+
+
+def test_concat_combined():
+    def html_lines(foot_prefix: str):
+        assert foot_prefix.endswith("_") or foot_prefix == ""
+        fp = foot_prefix
+        return indent(
+            dedent(
+                f"""\
+        <tr>
+          <th id="T_X_level0_{fp}row0" class="{fp}row_heading level0 {fp}row0" >a</th>
+          <td id="T_X_{fp}row0_col0" class="{fp}data {fp}row0 col0" >2.610000</td>
+        </tr>
+        <tr>
+          <th id="T_X_level0_{fp}row1" class="{fp}row_heading level0 {fp}row1" >b</th>
+          <td id="T_X_{fp}row1_col0" class="{fp}data {fp}row1 col0" >2.690000</td>
+        </tr>
+        """
+            ),
+            prefix=" " * 4,
+        )
+
+    df = DataFrame([[2.61], [2.69]], index=["a", "b"], columns=["A"])
+    s1 = df.style.highlight_max(color="red")
+    s2 = df.style.highlight_max(color="green")
+    s3 = df.style.highlight_max(color="blue")
+    s4 = df.style.highlight_max(color="yellow")
+
+    result = s1.concat(s2).concat(s3.concat(s4)).set_uuid("X").to_html()
+    expected_css = dedent(
+        """\
+        <style type="text/css">
+        #T_X_row1_col0 {
+          background-color: red;
+        }
+        #T_X_foot0_row1_col0 {
+          background-color: green;
+        }
+        #T_X_foot1_row1_col0 {
+          background-color: blue;
+        }
+        #T_X_foot1_foot0_row1_col0 {
+          background-color: yellow;
+        }
+        </style>
+        """
+    )
+    expected_table = (
+        dedent(
+            """\
+            <table id="T_X">
+              <thead>
+                <tr>
+                  <th class="blank level0" >&nbsp;</th>
+                  <th id="T_X_level0_col0" class="col_heading level0 col0" >A</th>
+                </tr>
+              </thead>
+              <tbody>
+            """
+        )
+        + html_lines("")
+        + html_lines("foot0_")
+        + html_lines("foot1_")
+        + html_lines("foot1_foot0_")
+        + dedent(
+            """\
+              </tbody>
+            </table>
+            """
+        )
+    )
+    assert expected_css + expected_table == result
+
+
+def test_to_html_na_rep_non_scalar_data(datapath):
+    # GH47103
+    df = DataFrame([{"a": 1, "b": [1, 2, 3], "c": np.nan}])
+    result = df.style.format(na_rep="-").to_html(table_uuid="test")
+    expected = """\
+<style type="text/css">
+</style>
+<table id="T_test">
+  <thead>
+    <tr>
+      <th class="blank level0" >&nbsp;</th>
+      <th id="T_test_level0_col0" class="col_heading level0 col0" >a</th>
+      <th id="T_test_level0_col1" class="col_heading level0 col1" >b</th>
+      <th id="T_test_level0_col2" class="col_heading level0 col2" >c</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th id="T_test_level0_row0" class="row_heading level0 row0" >0</th>
+      <td id="T_test_row0_col0" class="data row0 col0" >1</td>
+      <td id="T_test_row0_col1" class="data row0 col1" >[1, 2, 3]</td>
+      <td id="T_test_row0_col2" class="data row0 col2" >-</td>
+    </tr>
+  </tbody>
+</table>
+"""
+    assert result == expected
+
+
+@pytest.mark.parametrize("escape_axis_0", [True, False])
+@pytest.mark.parametrize("escape_axis_1", [True, False])
+def test_format_index_names(styler_multi, escape_axis_0, escape_axis_1):
+    if escape_axis_0:
+        styler_multi.format_index_names(axis=0, escape="html")
+        expected_index = ["X&gt;", "y_"]
+    else:
+        expected_index = ["X>", "y_"]
+
+    if escape_axis_1:
+        styler_multi.format_index_names(axis=1, escape="html")
+        expected_columns = ["A&amp;", "b&amp;"]
+    else:
+        expected_columns = ["A&", "b&"]
+
+    result = styler_multi.to_html(table_uuid="test")
+    for expected_str in expected_index + expected_columns:
+        assert f"{expected_str}</th>" in result
diff --git a/pandas/pandas/tests/io/formats/style/test_matplotlib.py b/pandas/pandas/tests/io/formats/style/test_matplotlib.py
new file mode 100644
index 0000000000000..490bd45bfb2ee
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_matplotlib.py
@@ -0,0 +1,305 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    IndexSlice,
+    Series,
+)
+
+mpl = pytest.importorskip("matplotlib")
+pytest.importorskip("jinja2")
+
+from pandas.io.formats.style import Styler
+
+pytestmark = pytest.mark.usefixtures("mpl_cleanup")
+
+
+@pytest.fixture
+def df():
+    return DataFrame([[1, 2], [2, 4]], columns=["A", "B"])
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+@pytest.fixture
+def df_blank():
+    return DataFrame([[0, 0], [0, 0]], columns=["A", "B"], index=["X", "Y"])
+
+
+@pytest.fixture
+def styler_blank(df_blank):
+    return Styler(df_blank, uuid_len=0)
+
+
+@pytest.mark.parametrize("f", ["background_gradient", "text_gradient"])
+def test_function_gradient(styler, f):
+    for c_map in [None, "YlOrRd"]:
+        result = getattr(styler, f)(cmap=c_map)._compute().ctx
+        assert all("#" in x[0][1] for x in result.values())
+        assert result[(0, 0)] == result[(0, 1)]
+        assert result[(1, 0)] == result[(1, 1)]
+
+
+@pytest.mark.parametrize("f", ["background_gradient", "text_gradient"])
+def test_background_gradient_color(styler, f):
+    result = getattr(styler, f)(subset=IndexSlice[1, "A"])._compute().ctx
+    if f == "background_gradient":
+        assert result[(1, 0)] == [("background-color", "#fff7fb"), ("color", "#000000")]
+    elif f == "text_gradient":
+        assert result[(1, 0)] == [("color", "#fff7fb")]
+
+
+@pytest.mark.parametrize(
+    "axis, expected",
+    [
+        (0, ["low", "low", "high", "high"]),
+        (1, ["low", "high", "low", "high"]),
+        (None, ["low", "mid", "mid", "high"]),
+    ],
+)
+@pytest.mark.parametrize("f", ["background_gradient", "text_gradient"])
+def test_background_gradient_axis(styler, axis, expected, f):
+    if f == "background_gradient":
+        colors = {
+            "low": [("background-color", "#f7fbff"), ("color", "#000000")],
+            "mid": [("background-color", "#abd0e6"), ("color", "#000000")],
+            "high": [("background-color", "#08306b"), ("color", "#f1f1f1")],
+        }
+    elif f == "text_gradient":
+        colors = {
+            "low": [("color", "#f7fbff")],
+            "mid": [("color", "#abd0e6")],
+            "high": [("color", "#08306b")],
+        }
+    result = getattr(styler, f)(cmap="Blues", axis=axis)._compute().ctx
+    for i, cell in enumerate([(0, 0), (0, 1), (1, 0), (1, 1)]):
+        assert result[cell] == colors[expected[i]]
+
+
+@pytest.mark.parametrize(
+    "cmap, expected",
+    [
+        (
+            "PuBu",
+            {
+                (4, 5): [("background-color", "#86b0d3"), ("color", "#000000")],
+                (4, 6): [("background-color", "#83afd3"), ("color", "#f1f1f1")],
+            },
+        ),
+        (
+            "YlOrRd",
+            {
+                (4, 8): [("background-color", "#fd913e"), ("color", "#000000")],
+                (4, 9): [("background-color", "#fd8f3d"), ("color", "#f1f1f1")],
+            },
+        ),
+        (
+            None,
+            {
+                (7, 0): [("background-color", "#48c16e"), ("color", "#f1f1f1")],
+                (7, 1): [("background-color", "#4cc26c"), ("color", "#000000")],
+            },
+        ),
+    ],
+)
+def test_text_color_threshold(cmap, expected):
+    # GH 39888
+    df = DataFrame(np.arange(100).reshape(10, 10))
+    result = df.style.background_gradient(cmap=cmap, axis=None)._compute().ctx
+    for k in expected.keys():
+        assert result[k] == expected[k]
+
+
+def test_background_gradient_vmin_vmax():
+    # GH 12145
+    df = DataFrame(range(5))
+    ctx = df.style.background_gradient(vmin=1, vmax=3)._compute().ctx
+    assert ctx[(0, 0)] == ctx[(1, 0)]
+    assert ctx[(4, 0)] == ctx[(3, 0)]
+
+
+def test_background_gradient_int64():
+    # GH 28869
+    df1 = Series(range(3)).to_frame()
+    df2 = Series(range(3), dtype="Int64").to_frame()
+    ctx1 = df1.style.background_gradient()._compute().ctx
+    ctx2 = df2.style.background_gradient()._compute().ctx
+    assert ctx2[(0, 0)] == ctx1[(0, 0)]
+    assert ctx2[(1, 0)] == ctx1[(1, 0)]
+    assert ctx2[(2, 0)] == ctx1[(2, 0)]
+
+
+@pytest.mark.parametrize(
+    "axis, gmap, expected",
+    [
+        (
+            0,
+            [1, 2],
+            {
+                (0, 0): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (1, 0): [("background-color", "#023858"), ("color", "#f1f1f1")],
+                (0, 1): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (1, 1): [("background-color", "#023858"), ("color", "#f1f1f1")],
+            },
+        ),
+        (
+            1,
+            [1, 2],
+            {
+                (0, 0): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (1, 0): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (0, 1): [("background-color", "#023858"), ("color", "#f1f1f1")],
+                (1, 1): [("background-color", "#023858"), ("color", "#f1f1f1")],
+            },
+        ),
+        (
+            None,
+            np.array([[2, 1], [1, 2]]),
+            {
+                (0, 0): [("background-color", "#023858"), ("color", "#f1f1f1")],
+                (1, 0): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (0, 1): [("background-color", "#fff7fb"), ("color", "#000000")],
+                (1, 1): [("background-color", "#023858"), ("color", "#f1f1f1")],
+            },
+        ),
+    ],
+)
+def test_background_gradient_gmap_array(styler_blank, axis, gmap, expected):
+    # tests when gmap is given as a sequence and converted to ndarray
+    result = styler_blank.background_gradient(axis=axis, gmap=gmap)._compute().ctx
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "gmap, axis", [([1, 2, 3], 0), ([1, 2], 1), (np.array([[1, 2], [1, 2]]), None)]
+)
+def test_background_gradient_gmap_array_raises(gmap, axis):
+    # test when gmap as converted ndarray is bad shape
+    df = DataFrame([[0, 0, 0], [0, 0, 0]])
+    msg = "supplied 'gmap' is not correct shape"
+    with pytest.raises(ValueError, match=msg):
+        df.style.background_gradient(gmap=gmap, axis=axis)._compute()
+
+
+@pytest.mark.parametrize(
+    "gmap",
+    [
+        DataFrame(  # reverse the columns
+            [[2, 1], [1, 2]], columns=["B", "A"], index=["X", "Y"]
+        ),
+        DataFrame(  # reverse the index
+            [[2, 1], [1, 2]], columns=["A", "B"], index=["Y", "X"]
+        ),
+        DataFrame(  # reverse the index and columns
+            [[1, 2], [2, 1]], columns=["B", "A"], index=["Y", "X"]
+        ),
+        DataFrame(  # add unnecessary columns
+            [[1, 2, 3], [2, 1, 3]], columns=["A", "B", "C"], index=["X", "Y"]
+        ),
+        DataFrame(  # add unnecessary index
+            [[1, 2], [2, 1], [3, 3]], columns=["A", "B"], index=["X", "Y", "Z"]
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    "subset, exp_gmap",  # exp_gmap is underlying map DataFrame should conform to
+    [
+        (None, [[1, 2], [2, 1]]),
+        (["A"], [[1], [2]]),  # slice only column "A" in data and gmap
+        (["B", "A"], [[2, 1], [1, 2]]),  # reverse the columns in data
+        (IndexSlice["X", :], [[1, 2]]),  # slice only index "X" in data and gmap
+        (IndexSlice[["Y", "X"], :], [[2, 1], [1, 2]]),  # reverse the index in data
+    ],
+)
+def test_background_gradient_gmap_dataframe_align(styler_blank, gmap, subset, exp_gmap):
+    # test gmap given as DataFrame that it aligns to the data including subset
+    expected = styler_blank.background_gradient(axis=None, gmap=exp_gmap, subset=subset)
+    result = styler_blank.background_gradient(axis=None, gmap=gmap, subset=subset)
+    assert expected._compute().ctx == result._compute().ctx
+
+
+@pytest.mark.parametrize(
+    "gmap, axis, exp_gmap",
+    [
+        (Series([2, 1], index=["Y", "X"]), 0, [[1, 1], [2, 2]]),  # reverse the index
+        (Series([2, 1], index=["B", "A"]), 1, [[1, 2], [1, 2]]),  # reverse the cols
+        (Series([1, 2, 3], index=["X", "Y", "Z"]), 0, [[1, 1], [2, 2]]),  # add idx
+        (Series([1, 2, 3], index=["A", "B", "C"]), 1, [[1, 2], [1, 2]]),  # add col
+    ],
+)
+def test_background_gradient_gmap_series_align(styler_blank, gmap, axis, exp_gmap):
+    # test gmap given as Series that it aligns to the data including subset
+    expected = styler_blank.background_gradient(axis=None, gmap=exp_gmap)._compute()
+    result = styler_blank.background_gradient(axis=axis, gmap=gmap)._compute()
+    assert expected.ctx == result.ctx
+
+
+@pytest.mark.parametrize("axis", [1, 0])
+def test_background_gradient_gmap_wrong_dataframe(styler_blank, axis):
+    # test giving a gmap in DataFrame but with wrong axis
+    gmap = DataFrame([[1, 2], [2, 1]], columns=["A", "B"], index=["X", "Y"])
+    msg = "'gmap' is a DataFrame but underlying data for operations is a Series"
+    with pytest.raises(ValueError, match=msg):
+        styler_blank.background_gradient(gmap=gmap, axis=axis)._compute()
+
+
+def test_background_gradient_gmap_wrong_series(styler_blank):
+    # test giving a gmap in Series form but with wrong axis
+    msg = "'gmap' is a Series but underlying data for operations is a DataFrame"
+    gmap = Series([1, 2], index=["X", "Y"])
+    with pytest.raises(ValueError, match=msg):
+        styler_blank.background_gradient(gmap=gmap, axis=None)._compute()
+
+
+def test_background_gradient_nullable_dtypes():
+    # GH 50712
+    df1 = DataFrame([[1], [0], [np.nan]], dtype=float)
+    df2 = DataFrame([[1], [0], [None]], dtype="Int64")
+
+    ctx1 = df1.style.background_gradient()._compute().ctx
+    ctx2 = df2.style.background_gradient()._compute().ctx
+    assert ctx1 == ctx2
+
+
+@pytest.mark.parametrize(
+    "cmap",
+    ["PuBu", mpl.colormaps["PuBu"]],
+)
+def test_bar_colormap(cmap):
+    data = DataFrame([[1, 2], [3, 4]])
+    ctx = data.style.bar(cmap=cmap, axis=None)._compute().ctx
+    pubu_colors = {
+        (0, 0): "#d0d1e6",
+        (1, 0): "#056faf",
+        (0, 1): "#73a9cf",
+        (1, 1): "#023858",
+    }
+    for k, v in pubu_colors.items():
+        assert v in ctx[k][1][1]
+
+
+def test_bar_color_raises(df):
+    msg = "`color` must be string or list or tuple of 2 strings"
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color={"a", "b"}).to_html()
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color=["a", "b", "c"]).to_html()
+
+    msg = "`color` and `cmap` cannot both be given"
+    with pytest.raises(ValueError, match=msg):
+        df.style.bar(color="something", cmap="something else").to_html()
+
+
+@pytest.mark.parametrize("plot_method", ["scatter", "hexbin"])
+def test_pass_colormap_instance(df, plot_method):
+    # https://github.com/pandas-dev/pandas/issues/49374
+    cmap = mpl.colors.ListedColormap([[1, 1, 1], [0, 0, 0]])
+    df["c"] = df.A + df.B
+    kwargs = {"x": "A", "y": "B", "c": "c", "colormap": cmap}
+    if plot_method == "hexbin":
+        kwargs["C"] = kwargs.pop("c")
+    getattr(df.plot, plot_method)(**kwargs)
diff --git a/pandas/pandas/tests/io/formats/style/test_non_unique.py b/pandas/pandas/tests/io/formats/style/test_non_unique.py
new file mode 100644
index 0000000000000..e4d31fe21f2c9
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_non_unique.py
@@ -0,0 +1,140 @@
+from textwrap import dedent
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    IndexSlice,
+)
+
+pytest.importorskip("jinja2")
+
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        index=["i", "j", "j"],
+        columns=["c", "d", "d"],
+        dtype=float,
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+def test_format_non_unique(df):
+    # GH 41269
+
+    # test dict
+    html = df.style.format({"d": "{:.1f}"}).to_html()
+    for val in ["1.000000<", "4.000000<", "7.000000<"]:
+        assert val in html
+    for val in ["2.0<", "3.0<", "5.0<", "6.0<", "8.0<", "9.0<"]:
+        assert val in html
+
+    # test subset
+    html = df.style.format(precision=1, subset=IndexSlice["j", "d"]).to_html()
+    for val in ["1.000000<", "4.000000<", "7.000000<", "2.000000<", "3.000000<"]:
+        assert val in html
+    for val in ["5.0<", "6.0<", "8.0<", "9.0<"]:
+        assert val in html
+
+
+@pytest.mark.parametrize("func", ["apply", "map"])
+def test_apply_map_non_unique_raises(df, func):
+    # GH 41269
+    if func == "apply":
+        op = lambda s: ["color: red;"] * len(s)
+    else:
+        op = lambda v: "color: red;"
+
+    with pytest.raises(KeyError, match="`Styler.apply` and `.map` are not"):
+        getattr(df.style, func)(op)._compute()
+
+
+def test_table_styles_dict_non_unique_index(styler):
+    styles = styler.set_table_styles(
+        {"j": [{"selector": "td", "props": "a: v;"}]}, axis=1
+    ).table_styles
+    assert styles == [
+        {"selector": "td.row1", "props": [("a", "v")]},
+        {"selector": "td.row2", "props": [("a", "v")]},
+    ]
+
+
+def test_table_styles_dict_non_unique_columns(styler):
+    styles = styler.set_table_styles(
+        {"d": [{"selector": "td", "props": "a: v;"}]}, axis=0
+    ).table_styles
+    assert styles == [
+        {"selector": "td.col1", "props": [("a", "v")]},
+        {"selector": "td.col2", "props": [("a", "v")]},
+    ]
+
+
+def test_tooltips_non_unique_raises(styler):
+    # ttips has unique keys
+    ttips = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "d"], index=["a", "b"])
+    styler.set_tooltips(ttips=ttips)  # OK
+
+    # ttips has non-unique columns
+    ttips = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "c"], index=["a", "b"])
+    with pytest.raises(KeyError, match="Tooltips render only if `ttips` has unique"):
+        styler.set_tooltips(ttips=ttips)
+
+    # ttips has non-unique index
+    ttips = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "d"], index=["a", "a"])
+    with pytest.raises(KeyError, match="Tooltips render only if `ttips` has unique"):
+        styler.set_tooltips(ttips=ttips)
+
+
+def test_set_td_classes_non_unique_raises(styler):
+    # classes has unique keys
+    classes = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "d"], index=["a", "b"])
+    styler.set_td_classes(classes=classes)  # OK
+
+    # classes has non-unique columns
+    classes = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "c"], index=["a", "b"])
+    with pytest.raises(KeyError, match="Classes render only if `classes` has unique"):
+        styler.set_td_classes(classes=classes)
+
+    # classes has non-unique index
+    classes = DataFrame([["1", "2"], ["3", "4"]], columns=["c", "d"], index=["a", "a"])
+    with pytest.raises(KeyError, match="Classes render only if `classes` has unique"):
+        styler.set_td_classes(classes=classes)
+
+
+def test_hide_columns_non_unique(styler):
+    ctx = styler.hide(["d"], axis="columns")._translate(True, True)
+
+    assert ctx["head"][0][1]["display_value"] == "c"
+    assert ctx["head"][0][1]["is_visible"] is True
+
+    assert ctx["head"][0][2]["display_value"] == "d"
+    assert ctx["head"][0][2]["is_visible"] is False
+
+    assert ctx["head"][0][3]["display_value"] == "d"
+    assert ctx["head"][0][3]["is_visible"] is False
+
+    assert ctx["body"][0][1]["is_visible"] is True
+    assert ctx["body"][0][2]["is_visible"] is False
+    assert ctx["body"][0][3]["is_visible"] is False
+
+
+def test_latex_non_unique(styler):
+    result = styler.to_latex()
+    assert result == dedent(
+        """\
+        \\begin{tabular}{lrrr}
+         & c & d & d \\\\
+        i & 1.000000 & 2.000000 & 3.000000 \\\\
+        j & 4.000000 & 5.000000 & 6.000000 \\\\
+        j & 7.000000 & 8.000000 & 9.000000 \\\\
+        \\end{tabular}
+    """
+    )
diff --git a/pandas/pandas/tests/io/formats/style/test_style.py b/pandas/pandas/tests/io/formats/style/test_style.py
new file mode 100644
index 0000000000000..b7dcfde327b83
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_style.py
@@ -0,0 +1,1603 @@
+import contextlib
+import copy
+import re
+from textwrap import dedent
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    IndexSlice,
+    MultiIndex,
+    Series,
+    option_context,
+)
+import pandas._testing as tm
+
+jinja2 = pytest.importorskip("jinja2")
+from pandas.io.formats.style import (  # isort:skip
+    Styler,
+)
+from pandas.io.formats.style_render import (
+    _get_level_lengths,
+    _get_trimming_maximums,
+    maybe_convert_css_to_tuples,
+    non_reducing_slice,
+)
+
+
+@pytest.fixture
+def mi_df():
+    return DataFrame(
+        [[1, 2], [3, 4]],
+        index=MultiIndex.from_product([["i0"], ["i1_a", "i1_b"]]),
+        columns=MultiIndex.from_product([["c0"], ["c1_a", "c1_b"]]),
+        dtype=int,
+    )
+
+
+@pytest.fixture
+def mi_styler(mi_df):
+    return Styler(mi_df, uuid_len=0)
+
+
+@pytest.fixture
+def mi_styler_comp(mi_styler):
+    # comprehensively add features to mi_styler
+    mi_styler = mi_styler._copy(deepcopy=True)
+    mi_styler.css = {**mi_styler.css, "row": "ROW", "col": "COL"}
+    mi_styler.uuid_len = 5
+    mi_styler.uuid = "abcde"
+    mi_styler.set_caption("capt")
+    mi_styler.set_table_styles([{"selector": "a", "props": "a:v;"}])
+    mi_styler.hide(axis="columns")
+    mi_styler.hide([("c0", "c1_a")], axis="columns", names=True)
+    mi_styler.hide(axis="index")
+    mi_styler.hide([("i0", "i1_a")], axis="index", names=True)
+    mi_styler.set_table_attributes('class="box"')
+    other = mi_styler.data.agg(["mean"])
+    other.index = MultiIndex.from_product([[""], other.index])
+    mi_styler.concat(other.style)
+    mi_styler.format(na_rep="MISSING", precision=3)
+    mi_styler.format_index(precision=2, axis=0)
+    mi_styler.format_index(precision=4, axis=1)
+    mi_styler.highlight_max(axis=None)
+    mi_styler.map_index(lambda x: "color: white;", axis=0)
+    mi_styler.map_index(lambda x: "color: black;", axis=1)
+    mi_styler.set_td_classes(
+        DataFrame(
+            [["a", "b"], ["a", "c"]], index=mi_styler.index, columns=mi_styler.columns
+        )
+    )
+    mi_styler.set_tooltips(
+        DataFrame(
+            [["a2", "b2"], ["a2", "c2"]],
+            index=mi_styler.index,
+            columns=mi_styler.columns,
+        )
+    )
+    mi_styler.format_index_names(escape="html", axis=0)
+    mi_styler.format_index_names(escape="html", axis=1)
+    return mi_styler
+
+
+@pytest.fixture
+def blank_value():
+    return "&nbsp;"
+
+
+@pytest.fixture
+def df():
+    df = DataFrame({"A": [0, 1], "B": np.random.default_rng(2).standard_normal(2)})
+    return df
+
+
+@pytest.fixture
+def styler(df):
+    df = DataFrame({"A": [0, 1], "B": np.random.default_rng(2).standard_normal(2)})
+    return Styler(df)
+
+
+@pytest.mark.parametrize(
+    "sparse_columns, exp_cols",
+    [
+        (
+            True,
+            [
+                {"is_visible": True, "attributes": 'colspan="2"', "value": "c0"},
+                {"is_visible": False, "attributes": "", "value": "c0"},
+            ],
+        ),
+        (
+            False,
+            [
+                {"is_visible": True, "attributes": "", "value": "c0"},
+                {"is_visible": True, "attributes": "", "value": "c0"},
+            ],
+        ),
+    ],
+)
+def test_mi_styler_sparsify_columns(mi_styler, sparse_columns, exp_cols):
+    exp_l1_c0 = {"is_visible": True, "attributes": "", "display_value": "c1_a"}
+    exp_l1_c1 = {"is_visible": True, "attributes": "", "display_value": "c1_b"}
+
+    ctx = mi_styler._translate(True, sparse_columns)
+
+    assert exp_cols[0].items() <= ctx["head"][0][2].items()
+    assert exp_cols[1].items() <= ctx["head"][0][3].items()
+    assert exp_l1_c0.items() <= ctx["head"][1][2].items()
+    assert exp_l1_c1.items() <= ctx["head"][1][3].items()
+
+
+@pytest.mark.parametrize(
+    "sparse_index, exp_rows",
+    [
+        (
+            True,
+            [
+                {"is_visible": True, "attributes": 'rowspan="2"', "value": "i0"},
+                {"is_visible": False, "attributes": "", "value": "i0"},
+            ],
+        ),
+        (
+            False,
+            [
+                {"is_visible": True, "attributes": "", "value": "i0"},
+                {"is_visible": True, "attributes": "", "value": "i0"},
+            ],
+        ),
+    ],
+)
+def test_mi_styler_sparsify_index(mi_styler, sparse_index, exp_rows):
+    exp_l1_r0 = {"is_visible": True, "attributes": "", "display_value": "i1_a"}
+    exp_l1_r1 = {"is_visible": True, "attributes": "", "display_value": "i1_b"}
+
+    ctx = mi_styler._translate(sparse_index, True)
+
+    assert exp_rows[0].items() <= ctx["body"][0][0].items()
+    assert exp_rows[1].items() <= ctx["body"][1][0].items()
+    assert exp_l1_r0.items() <= ctx["body"][0][1].items()
+    assert exp_l1_r1.items() <= ctx["body"][1][1].items()
+
+
+def test_mi_styler_sparsify_options(mi_styler):
+    with option_context("styler.sparse.index", False):
+        html1 = mi_styler.to_html()
+    with option_context("styler.sparse.index", True):
+        html2 = mi_styler.to_html()
+
+    assert html1 != html2
+
+    with option_context("styler.sparse.columns", False):
+        html1 = mi_styler.to_html()
+    with option_context("styler.sparse.columns", True):
+        html2 = mi_styler.to_html()
+
+    assert html1 != html2
+
+
+@pytest.mark.parametrize(
+    "rn, cn, max_els, max_rows, max_cols, exp_rn, exp_cn",
+    [
+        (100, 100, 100, None, None, 12, 6),  # reduce to (12, 6) < 100 elements
+        (1000, 3, 750, None, None, 250, 3),  # dynamically reduce rows to 250, keep cols
+        (4, 1000, 500, None, None, 4, 125),  # dynamically reduce cols to 125, keep rows
+        (1000, 3, 750, 10, None, 10, 3),  # overwrite above dynamics with max_row
+        (4, 1000, 500, None, 5, 4, 5),  # overwrite above dynamics with max_col
+        (100, 100, 700, 50, 50, 25, 25),  # rows cols below given maxes so < 700 elmts
+    ],
+)
+def test_trimming_maximum(rn, cn, max_els, max_rows, max_cols, exp_rn, exp_cn):
+    rn, cn = _get_trimming_maximums(
+        rn, cn, max_els, max_rows, max_cols, scaling_factor=0.5
+    )
+    assert (rn, cn) == (exp_rn, exp_cn)
+
+
+@pytest.mark.parametrize(
+    "option, val",
+    [
+        ("styler.render.max_elements", 6),
+        ("styler.render.max_rows", 3),
+    ],
+)
+def test_render_trimming_rows(option, val):
+    # test auto and specific trimming of rows
+    df = DataFrame(np.arange(120).reshape(60, 2))
+    with option_context(option, val):
+        ctx = df.style._translate(True, True)
+    assert len(ctx["head"][0]) == 3  # index + 2 data cols
+    assert len(ctx["body"]) == 4  # 3 data rows + trimming row
+    assert len(ctx["body"][0]) == 3  # index + 2 data cols
+
+
+@pytest.mark.parametrize(
+    "option, val",
+    [
+        ("styler.render.max_elements", 6),
+        ("styler.render.max_columns", 2),
+    ],
+)
+def test_render_trimming_cols(option, val):
+    # test auto and specific trimming of cols
+    df = DataFrame(np.arange(30).reshape(3, 10))
+    with option_context(option, val):
+        ctx = df.style._translate(True, True)
+    assert len(ctx["head"][0]) == 4  # index + 2 data cols + trimming col
+    assert len(ctx["body"]) == 3  # 3 data rows
+    assert len(ctx["body"][0]) == 4  # index + 2 data cols + trimming col
+
+
+def test_render_trimming_mi():
+    midx = MultiIndex.from_product([[1, 2], [1, 2, 3]])
+    df = DataFrame(np.arange(36).reshape(6, 6), columns=midx, index=midx)
+    with option_context("styler.render.max_elements", 4):
+        ctx = df.style._translate(True, True)
+
+    assert len(ctx["body"][0]) == 5  # 2 indexes + 2 data cols + trimming row
+    assert {"attributes": 'rowspan="2"'}.items() <= ctx["body"][0][0].items()
+    assert {"class": "data row0 col_trim"}.items() <= ctx["body"][0][4].items()
+    assert {"class": "data row_trim col_trim"}.items() <= ctx["body"][2][4].items()
+    assert len(ctx["body"]) == 3  # 2 data rows + trimming row
+
+
+def test_render_empty_mi():
+    # GH 43305
+    df = DataFrame(index=MultiIndex.from_product([["A"], [0, 1]], names=[None, "one"]))
+    expected = dedent(
+        """\
+    >
+      <thead>
+        <tr>
+          <th class="index_name level0" >&nbsp;</th>
+          <th class="index_name level1" >one</th>
+        </tr>
+      </thead>
+    """
+    )
+    assert expected in df.style.to_html()
+
+
+@pytest.mark.parametrize("comprehensive", [True, False])
+@pytest.mark.parametrize("render", [True, False])
+@pytest.mark.parametrize("deepcopy", [True, False])
+def test_copy(comprehensive, render, deepcopy, mi_styler, mi_styler_comp):
+    styler = mi_styler_comp if comprehensive else mi_styler
+    styler.uuid_len = 5
+
+    s2 = copy.deepcopy(styler) if deepcopy else copy.copy(styler)  # make copy and check
+    assert s2 is not styler
+
+    if render:
+        styler.to_html()
+
+    excl = [
+        "cellstyle_map",  # render time vars..
+        "cellstyle_map_columns",
+        "cellstyle_map_index",
+        "template_latex",  # render templates are class level
+        "template_html",
+        "template_html_style",
+        "template_html_table",
+    ]
+    if not deepcopy:  # check memory locations are equal for all included attributes
+        for attr in [a for a in styler.__dict__ if (not callable(a) and a not in excl)]:
+            assert id(getattr(s2, attr)) == id(getattr(styler, attr))
+    else:  # check memory locations are different for nested or mutable vars
+        shallow = [
+            "data",
+            "columns",
+            "index",
+            "uuid_len",
+            "uuid",
+            "caption",
+            "cell_ids",
+            "hide_index_",
+            "hide_columns_",
+            "hide_index_names",
+            "hide_column_names",
+            "table_attributes",
+        ]
+        for attr in shallow:
+            assert id(getattr(s2, attr)) == id(getattr(styler, attr))
+
+        for attr in [
+            a
+            for a in styler.__dict__
+            if (not callable(a) and a not in excl and a not in shallow)
+        ]:
+            if getattr(s2, attr) is None:
+                assert id(getattr(s2, attr)) == id(getattr(styler, attr))
+            else:
+                assert id(getattr(s2, attr)) != id(getattr(styler, attr))
+
+
+@pytest.mark.parametrize("deepcopy", [True, False])
+def test_inherited_copy(mi_styler, deepcopy):
+    # Ensure that the inherited class is preserved when a Styler object is copied.
+    # GH 52728
+    class CustomStyler(Styler):
+        pass
+
+    custom_styler = CustomStyler(mi_styler.data)
+    custom_styler_copy = (
+        copy.deepcopy(custom_styler) if deepcopy else copy.copy(custom_styler)
+    )
+    assert isinstance(custom_styler_copy, CustomStyler)
+
+
+def test_clear(mi_styler_comp):
+    # NOTE: if this test fails for new features then 'mi_styler_comp' should be updated
+    # to ensure proper testing of the 'copy', 'clear', 'export' methods with new feature
+    # GH 40675
+    styler = mi_styler_comp
+    styler._compute()  # execute applied methods
+
+    clean_copy = Styler(styler.data, uuid=styler.uuid)
+
+    excl = [
+        "data",
+        "index",
+        "columns",
+        "uuid",
+        "uuid_len",  # uuid is set to be the same on styler and clean_copy
+        "cell_ids",
+        "cellstyle_map",  # execution time only
+        "cellstyle_map_columns",  # execution time only
+        "cellstyle_map_index",  # execution time only
+        "template_latex",  # render templates are class level
+        "template_html",
+        "template_html_style",
+        "template_html_table",
+    ]
+    # tests vars are not same vals on obj and clean copy before clear (except for excl)
+    for attr in [a for a in styler.__dict__ if not (callable(a) or a in excl)]:
+        res = getattr(styler, attr) == getattr(clean_copy, attr)
+        if hasattr(res, "__iter__") and len(res) > 0:
+            assert not all(res)  # some element in iterable differs
+        elif hasattr(res, "__iter__") and len(res) == 0:
+            pass  # empty array
+        else:
+            assert not res  # explicit var differs
+
+    # test vars have same vales on obj and clean copy after clearing
+    styler.clear()
+    for attr in [a for a in styler.__dict__ if not callable(a)]:
+        res = getattr(styler, attr) == getattr(clean_copy, attr)
+        assert all(res) if hasattr(res, "__iter__") else res
+
+
+def test_export(mi_styler_comp, mi_styler):
+    exp_attrs = [
+        "_todo",
+        "hide_index_",
+        "hide_index_names",
+        "hide_columns_",
+        "hide_column_names",
+        "table_attributes",
+        "table_styles",
+        "css",
+    ]
+    for attr in exp_attrs:
+        check = getattr(mi_styler, attr) == getattr(mi_styler_comp, attr)
+        assert not (
+            all(check) if (hasattr(check, "__iter__") and len(check) > 0) else check
+        )
+
+    export = mi_styler_comp.export()
+    used = mi_styler.use(export)
+    for attr in exp_attrs:
+        check = getattr(used, attr) == getattr(mi_styler_comp, attr)
+        assert all(check) if (hasattr(check, "__iter__") and len(check) > 0) else check
+
+    used.to_html()
+
+
+def test_hide_raises(mi_styler):
+    msg = "`subset` and `level` cannot be passed simultaneously"
+    with pytest.raises(ValueError, match=msg):
+        mi_styler.hide(axis="index", subset="something", level="something else")
+
+    msg = "`level` must be of type `int`, `str` or list of such"
+    with pytest.raises(ValueError, match=msg):
+        mi_styler.hide(axis="index", level={"bad": 1, "type": 2})
+
+
+@pytest.mark.parametrize("level", [1, "one", [1], ["one"]])
+def test_hide_index_level(mi_styler, level):
+    mi_styler.index.names, mi_styler.columns.names = ["zero", "one"], ["zero", "one"]
+    ctx = mi_styler.hide(axis="index", level=level)._translate(False, True)
+    assert len(ctx["head"][0]) == 3
+    assert len(ctx["head"][1]) == 3
+    assert len(ctx["head"][2]) == 4
+    assert ctx["head"][2][0]["is_visible"]
+    assert not ctx["head"][2][1]["is_visible"]
+
+    assert ctx["body"][0][0]["is_visible"]
+    assert not ctx["body"][0][1]["is_visible"]
+    assert ctx["body"][1][0]["is_visible"]
+    assert not ctx["body"][1][1]["is_visible"]
+
+
+@pytest.mark.parametrize("level", [1, "one", [1], ["one"]])
+@pytest.mark.parametrize("names", [True, False])
+def test_hide_columns_level(mi_styler, level, names):
+    mi_styler.columns.names = ["zero", "one"]
+    if names:
+        mi_styler.index.names = ["zero", "one"]
+    ctx = mi_styler.hide(axis="columns", level=level)._translate(True, False)
+    assert len(ctx["head"]) == (2 if names else 1)
+
+
+@pytest.mark.parametrize("method", ["map", "apply"])
+@pytest.mark.parametrize("axis", ["index", "columns"])
+def test_apply_map_header(method, axis):
+    # GH 41893
+    df = DataFrame({"A": [0, 0], "B": [1, 1]}, index=["C", "D"])
+    func = {
+        "apply": lambda s: ["attr: val" if ("A" in v or "C" in v) else "" for v in s],
+        "map": lambda v: "attr: val" if ("A" in v or "C" in v) else "",
+    }
+
+    # test execution added to todo
+    result = getattr(df.style, f"{method}_index")(func[method], axis=axis)
+    assert len(result._todo) == 1
+    assert len(getattr(result, f"ctx_{axis}")) == 0
+
+    # test ctx object on compute
+    result._compute()
+    expected = {
+        (0, 0): [("attr", "val")],
+    }
+    assert getattr(result, f"ctx_{axis}") == expected
+
+
+@pytest.mark.parametrize("method", ["apply", "map"])
+@pytest.mark.parametrize("axis", ["index", "columns"])
+def test_apply_map_header_mi(mi_styler, method, axis):
+    # GH 41893
+    func = {
+        "apply": lambda s: ["attr: val;" if "b" in v else "" for v in s],
+        "map": lambda v: "attr: val" if "b" in v else "",
+    }
+    result = getattr(mi_styler, f"{method}_index")(func[method], axis=axis)._compute()
+    expected = {(1, 1): [("attr", "val")]}
+    assert getattr(result, f"ctx_{axis}") == expected
+
+
+def test_apply_map_header_raises(mi_styler):
+    # GH 41893
+    with pytest.raises(ValueError, match="No axis named bad for object type DataFrame"):
+        mi_styler.map_index(lambda v: "attr: val;", axis="bad")._compute()
+
+
+class TestStyler:
+    def test_init_non_pandas(self):
+        msg = "``data`` must be a Series or DataFrame"
+        with pytest.raises(TypeError, match=msg):
+            Styler([1, 2, 3])
+
+    def test_init_series(self):
+        result = Styler(Series([1, 2]))
+        assert result.data.ndim == 2
+
+    def test_repr_html_ok(self, styler):
+        styler._repr_html_()
+
+    def test_repr_html_mathjax(self, styler):
+        # gh-19824 / 41395
+        assert "tex2jax_ignore" not in styler._repr_html_()
+        assert "mathjax_ignore" not in styler._repr_html_()
+
+        with option_context("styler.html.mathjax", False):
+            assert "tex2jax_ignore" in styler._repr_html_()
+            assert "mathjax_ignore" in styler._repr_html_()
+
+    def test_update_ctx(self, styler):
+        styler._update_ctx(DataFrame({"A": ["color: red", "color: blue"]}))
+        expected = {(0, 0): [("color", "red")], (1, 0): [("color", "blue")]}
+        assert styler.ctx == expected
+
+    def test_update_ctx_flatten_multi_and_trailing_semi(self, styler):
+        attrs = DataFrame({"A": ["color: red; foo: bar", "color:blue ; foo: baz;"]})
+        styler._update_ctx(attrs)
+        expected = {
+            (0, 0): [("color", "red"), ("foo", "bar")],
+            (1, 0): [("color", "blue"), ("foo", "baz")],
+        }
+        assert styler.ctx == expected
+
+    def test_render(self):
+        df = DataFrame({"A": [0, 1]})
+        style = lambda x: Series(["color: red", "color: blue"], name=x.name)
+        s = Styler(df, uuid="AB").apply(style)
+        s.to_html()
+        # it worked?
+
+    def test_multiple_render(self, df):
+        # GH 39396
+        s = Styler(df, uuid_len=0).map(lambda x: "color: red;", subset=["A"])
+        s.to_html()  # do 2 renders to ensure css styles not duplicated
+        assert (
+            '<style type="text/css">\n#T__row0_col0, #T__row1_col0 {\n'
+            "  color: red;\n}\n</style>" in s.to_html()
+        )
+
+    def test_render_empty_dfs(self):
+        empty_df = DataFrame()
+        es = Styler(empty_df)
+        es.to_html()
+        # An index but no columns
+        DataFrame(columns=["a"]).style.to_html()
+        # A column but no index
+        DataFrame(index=["a"]).style.to_html()
+        # No IndexError raised?
+
+    def test_render_double(self):
+        df = DataFrame({"A": [0, 1]})
+        style = lambda x: Series(
+            ["color: red; border: 1px", "color: blue; border: 2px"], name=x.name
+        )
+        s = Styler(df, uuid="AB").apply(style)
+        s.to_html()
+        # it worked?
+
+    def test_set_properties(self):
+        df = DataFrame({"A": [0, 1]})
+        result = df.style.set_properties(color="white", size="10px")._compute().ctx
+        # order is deterministic
+        v = [("color", "white"), ("size", "10px")]
+        expected = {(0, 0): v, (1, 0): v}
+        assert result.keys() == expected.keys()
+        for v1, v2 in zip(result.values(), expected.values()):
+            assert sorted(v1) == sorted(v2)
+
+    def test_set_properties_subset(self):
+        df = DataFrame({"A": [0, 1]})
+        result = (
+            df.style.set_properties(subset=IndexSlice[0, "A"], color="white")
+            ._compute()
+            .ctx
+        )
+        expected = {(0, 0): [("color", "white")]}
+        assert result == expected
+
+    def test_empty_index_name_doesnt_display(self, blank_value):
+        # https://github.com/pandas-dev/pandas/pull/12090#issuecomment-180695902
+        df = DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+        result = df.style._translate(True, True)
+        assert len(result["head"]) == 1
+        expected = {
+            "class": "blank level0",
+            "type": "th",
+            "value": blank_value,
+            "is_visible": True,
+            "display_value": blank_value,
+        }
+        assert expected.items() <= result["head"][0][0].items()
+
+    def test_index_name(self):
+        # https://github.com/pandas-dev/pandas/issues/11655
+        df = DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+        result = df.set_index("A").style._translate(True, True)
+        expected = {
+            "class": "index_name level0",
+            "type": "th",
+            "value": "A",
+            "is_visible": True,
+            "display_value": "A",
+        }
+        assert expected.items() <= result["head"][1][0].items()
+
+    def test_numeric_columns(self):
+        # https://github.com/pandas-dev/pandas/issues/12125
+        # smoke test for _translate
+        df = DataFrame({0: [1, 2, 3]})
+        df.style._translate(True, True)
+
+    def test_apply_axis(self):
+        df = DataFrame({"A": [0, 0], "B": [1, 1]})
+        f = lambda x: [f"val: {x.max()}" for v in x]
+        result = df.style.apply(f, axis=1)
+        assert len(result._todo) == 1
+        assert len(result.ctx) == 0
+        result._compute()
+        expected = {
+            (0, 0): [("val", "1")],
+            (0, 1): [("val", "1")],
+            (1, 0): [("val", "1")],
+            (1, 1): [("val", "1")],
+        }
+        assert result.ctx == expected
+
+        result = df.style.apply(f, axis=0)
+        expected = {
+            (0, 0): [("val", "0")],
+            (0, 1): [("val", "1")],
+            (1, 0): [("val", "0")],
+            (1, 1): [("val", "1")],
+        }
+        result._compute()
+        assert result.ctx == expected
+        result = df.style.apply(f)  # default
+        result._compute()
+        assert result.ctx == expected
+
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_apply_series_return(self, axis):
+        # GH 42014
+        df = DataFrame([[1, 2], [3, 4]], index=["X", "Y"], columns=["X", "Y"])
+
+        # test Series return where len(Series) < df.index or df.columns but labels OK
+        func = lambda s: Series(["color: red;"], index=["Y"])
+        result = df.style.apply(func, axis=axis)._compute().ctx
+        assert result[(1, 1)] == [("color", "red")]
+        assert result[(1 - axis, axis)] == [("color", "red")]
+
+        # test Series return where labels align but different order
+        func = lambda s: Series(["color: red;", "color: blue;"], index=["Y", "X"])
+        result = df.style.apply(func, axis=axis)._compute().ctx
+        assert result[(0, 0)] == [("color", "blue")]
+        assert result[(1, 1)] == [("color", "red")]
+        assert result[(1 - axis, axis)] == [("color", "red")]
+        assert result[(axis, 1 - axis)] == [("color", "blue")]
+
+    @pytest.mark.parametrize("index", [False, True])
+    @pytest.mark.parametrize("columns", [False, True])
+    def test_apply_dataframe_return(self, index, columns):
+        # GH 42014
+        df = DataFrame([[1, 2], [3, 4]], index=["X", "Y"], columns=["X", "Y"])
+        idxs = ["X", "Y"] if index else ["Y"]
+        cols = ["X", "Y"] if columns else ["Y"]
+        df_styles = DataFrame("color: red;", index=idxs, columns=cols)
+        result = df.style.apply(lambda x: df_styles, axis=None)._compute().ctx
+
+        assert result[(1, 1)] == [("color", "red")]  # (Y,Y) styles always present
+        assert (result[(0, 1)] == [("color", "red")]) is index  # (X,Y) only if index
+        assert (result[(1, 0)] == [("color", "red")]) is columns  # (Y,X) only if cols
+        assert (result[(0, 0)] == [("color", "red")]) is (index and columns)  # (X,X)
+
+    @pytest.mark.parametrize(
+        "slice_",
+        [
+            IndexSlice[:],
+            IndexSlice[:, ["A"]],
+            IndexSlice[[1], :],
+            IndexSlice[[1], ["A"]],
+            IndexSlice[:2, ["A", "B"]],
+        ],
+    )
+    @pytest.mark.parametrize("axis", [0, 1])
+    def test_apply_subset(self, slice_, axis, df):
+        def h(x, color="bar"):
+            return Series(f"color: {color}", index=x.index, name=x.name)
+
+        result = df.style.apply(h, axis=axis, subset=slice_, color="baz")._compute().ctx
+        expected = {
+            (r, c): [("color", "baz")]
+            for r, row in enumerate(df.index)
+            for c, col in enumerate(df.columns)
+            if row in df.loc[slice_].index and col in df.loc[slice_].columns
+        }
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "slice_",
+        [
+            IndexSlice[:],
+            IndexSlice[:, ["A"]],
+            IndexSlice[[1], :],
+            IndexSlice[[1], ["A"]],
+            IndexSlice[:2, ["A", "B"]],
+        ],
+    )
+    def test_map_subset(self, slice_, df):
+        result = df.style.map(lambda x: "color:baz;", subset=slice_)._compute().ctx
+        expected = {
+            (r, c): [("color", "baz")]
+            for r, row in enumerate(df.index)
+            for c, col in enumerate(df.columns)
+            if row in df.loc[slice_].index and col in df.loc[slice_].columns
+        }
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "slice_",
+        [
+            IndexSlice[:, IndexSlice["x", "A"]],
+            IndexSlice[:, IndexSlice[:, "A"]],
+            IndexSlice[:, IndexSlice[:, ["A", "C"]]],  # missing col element
+            IndexSlice[IndexSlice["a", 1], :],
+            IndexSlice[IndexSlice[:, 1], :],
+            IndexSlice[IndexSlice[:, [1, 3]], :],  # missing row element
+            IndexSlice[:, ("x", "A")],
+            IndexSlice[("a", 1), :],
+        ],
+    )
+    def test_map_subset_multiindex(self, slice_):
+        # GH 19861
+        # edited for GH 33562
+        if (
+            isinstance(slice_[-1], tuple)
+            and isinstance(slice_[-1][-1], list)
+            and "C" in slice_[-1][-1]
+        ):
+            ctx = pytest.raises(KeyError, match="C")
+        elif (
+            isinstance(slice_[0], tuple)
+            and isinstance(slice_[0][1], list)
+            and 3 in slice_[0][1]
+        ):
+            ctx = pytest.raises(KeyError, match="3")
+        else:
+            ctx = contextlib.nullcontext()
+
+        idx = MultiIndex.from_product([["a", "b"], [1, 2]])
+        col = MultiIndex.from_product([["x", "y"], ["A", "B"]])
+        df = DataFrame(np.random.default_rng(2).random((4, 4)), columns=col, index=idx)
+
+        with ctx:
+            df.style.map(lambda x: "color: red;", subset=slice_).to_html()
+
+    def test_map_subset_multiindex_code(self):
+        # https://github.com/pandas-dev/pandas/issues/25858
+        # Checks styler.map works with multindex when codes are provided
+        codes = np.array([[0, 0, 1, 1], [0, 1, 0, 1]])
+        columns = MultiIndex(
+            levels=[["a", "b"], ["%", "#"]], codes=codes, names=["", ""]
+        )
+        df = DataFrame(
+            [[1, -1, 1, 1], [-1, 1, 1, 1]], index=["hello", "world"], columns=columns
+        )
+        pct_subset = IndexSlice[:, IndexSlice[:, "%":"%"]]
+
+        def color_negative_red(val):
+            color = "red" if val < 0 else "black"
+            return f"color: {color}"
+
+        df.loc[pct_subset]
+        df.style.map(color_negative_red, subset=pct_subset)
+
+    @pytest.mark.parametrize(
+        "stylefunc", ["background_gradient", "bar", "text_gradient"]
+    )
+    def test_subset_for_boolean_cols(self, stylefunc):
+        # GH47838
+        df = DataFrame(
+            [
+                [1, 2],
+                [3, 4],
+            ],
+            columns=[False, True],
+        )
+        styled = getattr(df.style, stylefunc)()
+        styled._compute()
+        assert set(styled.ctx) == {(0, 0), (0, 1), (1, 0), (1, 1)}
+
+    def test_empty(self):
+        df = DataFrame({"A": [1, 0]})
+        s = df.style
+        s.ctx = {(0, 0): [("color", "red")], (1, 0): [("", "")]}
+
+        result = s._translate(True, True)["cellstyle"]
+        expected = [
+            {"props": [("color", "red")], "selectors": ["row0_col0"]},
+            {"props": [("", "")], "selectors": ["row1_col0"]},
+        ]
+        assert result == expected
+
+    def test_duplicate(self):
+        df = DataFrame({"A": [1, 0]})
+        s = df.style
+        s.ctx = {(0, 0): [("color", "red")], (1, 0): [("color", "red")]}
+
+        result = s._translate(True, True)["cellstyle"]
+        expected = [
+            {"props": [("color", "red")], "selectors": ["row0_col0", "row1_col0"]}
+        ]
+        assert result == expected
+
+    def test_init_with_na_rep(self):
+        # GH 21527 28358
+        df = DataFrame([[None, None], [1.1, 1.2]], columns=["A", "B"])
+
+        ctx = Styler(df, na_rep="NA")._translate(True, True)
+        assert ctx["body"][0][1]["display_value"] == "NA"
+        assert ctx["body"][0][2]["display_value"] == "NA"
+
+    def test_caption(self, df):
+        styler = Styler(df, caption="foo")
+        result = styler.to_html()
+        assert all(["caption" in result, "foo" in result])
+
+        styler = df.style
+        result = styler.set_caption("baz")
+        assert styler is result
+        assert styler.caption == "baz"
+
+    def test_uuid(self, df):
+        styler = Styler(df, uuid="abc123")
+        result = styler.to_html()
+        assert "abc123" in result
+
+        styler = df.style
+        result = styler.set_uuid("aaa")
+        assert result is styler
+        assert result.uuid == "aaa"
+
+    def test_unique_id(self):
+        # See https://github.com/pandas-dev/pandas/issues/16780
+        df = DataFrame({"a": [1, 3, 5, 6], "b": [2, 4, 12, 21]})
+        result = df.style.to_html(uuid="test")
+        assert "test" in result
+        ids = re.findall('id="(.*?)"', result)
+        assert np.unique(ids).size == len(ids)
+
+    def test_table_styles(self, df):
+        style = [{"selector": "th", "props": [("foo", "bar")]}]  # default format
+        styler = Styler(df, table_styles=style)
+        result = " ".join(styler.to_html().split())
+        assert "th { foo: bar; }" in result
+
+        styler = df.style
+        result = styler.set_table_styles(style)
+        assert styler is result
+        assert styler.table_styles == style
+
+        # GH 39563
+        style = [{"selector": "th", "props": "foo:bar;"}]  # css string format
+        styler = df.style.set_table_styles(style)
+        result = " ".join(styler.to_html().split())
+        assert "th { foo: bar; }" in result
+
+    def test_table_styles_multiple(self, df):
+        ctx = df.style.set_table_styles(
+            [
+                {"selector": "th,td", "props": "color:red;"},
+                {"selector": "tr", "props": "color:green;"},
+            ]
+        )._translate(True, True)["table_styles"]
+        assert ctx == [
+            {"selector": "th", "props": [("color", "red")]},
+            {"selector": "td", "props": [("color", "red")]},
+            {"selector": "tr", "props": [("color", "green")]},
+        ]
+
+    def test_table_styles_dict_multiple_selectors(self, df):
+        # GH 44011
+        result = df.style.set_table_styles(
+            {
+                "B": [
+                    {"selector": "th,td", "props": [("border-left", "2px solid black")]}
+                ]
+            }
+        )._translate(True, True)["table_styles"]
+
+        expected = [
+            {"selector": "th.col1", "props": [("border-left", "2px solid black")]},
+            {"selector": "td.col1", "props": [("border-left", "2px solid black")]},
+        ]
+
+        assert result == expected
+
+    def test_maybe_convert_css_to_tuples(self):
+        expected = [("a", "b"), ("c", "d e")]
+        assert maybe_convert_css_to_tuples("a:b;c:d e;") == expected
+        assert maybe_convert_css_to_tuples("a: b ;c:  d e  ") == expected
+        expected = []
+        assert maybe_convert_css_to_tuples("") == expected
+
+        # issue #59623
+        expected = [("a", "b"), ("c", "url('data:123')")]
+        assert maybe_convert_css_to_tuples("a:b;c: url('data:123');") == expected
+
+        # if no value, return attr and empty string
+        expected = [("a", ""), ("c", "")]
+        assert maybe_convert_css_to_tuples("a:;c: ") == expected
+
+    def test_maybe_convert_css_to_tuples_err(self):
+        msg = (
+            "Styles supplied as string must follow CSS rule formats, "
+            "for example 'attr: val;'. 'err' was given."
+        )
+        with pytest.raises(ValueError, match=msg):
+            maybe_convert_css_to_tuples("err")
+
+    def test_table_attributes(self, df):
+        attributes = 'class="foo" data-bar'
+        styler = Styler(df, table_attributes=attributes)
+        result = styler.to_html()
+        assert 'class="foo" data-bar' in result
+
+        result = df.style.set_table_attributes(attributes).to_html()
+        assert 'class="foo" data-bar' in result
+
+    def test_apply_none(self):
+        def f(x):
+            return DataFrame(
+                np.where(x == x.max(), "color: red", ""),
+                index=x.index,
+                columns=x.columns,
+            )
+
+        result = DataFrame([[1, 2], [3, 4]]).style.apply(f, axis=None)._compute().ctx
+        assert result[(1, 1)] == [("color", "red")]
+
+    def test_trim(self, df):
+        result = df.style.to_html()  # trim=True
+        assert result.count("#") == 0
+
+        result = df.style.highlight_max().to_html()
+        assert result.count("#") == len(df.columns)
+
+    def test_export(self, df, styler):
+        f = lambda x: "color: red" if x > 0 else "color: blue"
+        g = lambda x, z: f"color: {z}"
+        style1 = styler
+        style1.map(f).map(g, z="b").highlight_max()._compute()  # = render
+        result = style1.export()
+        style2 = df.style
+        style2.use(result)
+        assert style1._todo == style2._todo
+        style2.to_html()
+
+    def test_bad_apply_shape(self):
+        df = DataFrame([[1, 2], [3, 4]], index=["A", "B"], columns=["X", "Y"])
+
+        msg = "resulted in the apply method collapsing to a Series."
+        with pytest.raises(ValueError, match=msg):
+            df.style._apply(lambda x: "x")
+
+        msg = "created invalid {} labels"
+        with pytest.raises(ValueError, match=msg.format("index")):
+            df.style._apply(lambda x: [""])
+
+        with pytest.raises(ValueError, match=msg.format("index")):
+            df.style._apply(lambda x: ["", "", "", ""])
+
+        with pytest.raises(ValueError, match=msg.format("index")):
+            df.style._apply(lambda x: Series(["a:v;", ""], index=["A", "C"]), axis=0)
+
+        with pytest.raises(ValueError, match=msg.format("columns")):
+            df.style._apply(lambda x: ["", "", ""], axis=1)
+
+        with pytest.raises(ValueError, match=msg.format("columns")):
+            df.style._apply(lambda x: Series(["a:v;", ""], index=["X", "Z"]), axis=1)
+
+        msg = "returned ndarray with wrong shape"
+        with pytest.raises(ValueError, match=msg):
+            df.style._apply(lambda x: np.array([[""], [""]]), axis=None)
+
+    def test_apply_bad_return(self):
+        def f(x):
+            return ""
+
+        df = DataFrame([[1, 2], [3, 4]])
+        msg = (
+            "must return a DataFrame or ndarray when passed to `Styler.apply` "
+            "with axis=None"
+        )
+        with pytest.raises(TypeError, match=msg):
+            df.style._apply(f, axis=None)
+
+    @pytest.mark.parametrize("axis", ["index", "columns"])
+    def test_apply_bad_labels(self, axis):
+        def f(x):
+            return DataFrame(**{axis: ["bad", "labels"]})
+
+        df = DataFrame([[1, 2], [3, 4]])
+        msg = f"created invalid {axis} labels."
+        with pytest.raises(ValueError, match=msg):
+            df.style._apply(f, axis=None)
+
+    def test_get_level_lengths(self):
+        index = MultiIndex.from_product([["a", "b"], [0, 1, 2]])
+        expected = {
+            (0, 0): 3,
+            (0, 3): 3,
+            (1, 0): 1,
+            (1, 1): 1,
+            (1, 2): 1,
+            (1, 3): 1,
+            (1, 4): 1,
+            (1, 5): 1,
+        }
+        result = _get_level_lengths(index, sparsify=True, max_index=100)
+        tm.assert_dict_equal(result, expected)
+
+        expected = {
+            (0, 0): 1,
+            (0, 1): 1,
+            (0, 2): 1,
+            (0, 3): 1,
+            (0, 4): 1,
+            (0, 5): 1,
+            (1, 0): 1,
+            (1, 1): 1,
+            (1, 2): 1,
+            (1, 3): 1,
+            (1, 4): 1,
+            (1, 5): 1,
+        }
+        result = _get_level_lengths(index, sparsify=False, max_index=100)
+        tm.assert_dict_equal(result, expected)
+
+    def test_get_level_lengths_un_sorted(self):
+        index = MultiIndex.from_arrays([[1, 1, 2, 1], ["a", "b", "b", "d"]])
+        expected = {
+            (0, 0): 2,
+            (0, 2): 1,
+            (0, 3): 1,
+            (1, 0): 1,
+            (1, 1): 1,
+            (1, 2): 1,
+            (1, 3): 1,
+        }
+        result = _get_level_lengths(index, sparsify=True, max_index=100)
+        tm.assert_dict_equal(result, expected)
+
+        expected = {
+            (0, 0): 1,
+            (0, 1): 1,
+            (0, 2): 1,
+            (0, 3): 1,
+            (1, 0): 1,
+            (1, 1): 1,
+            (1, 2): 1,
+            (1, 3): 1,
+        }
+        result = _get_level_lengths(index, sparsify=False, max_index=100)
+        tm.assert_dict_equal(result, expected)
+
+    def test_mi_sparse_index_names(self, blank_value):
+        # Test the class names and displayed value are correct on rendering MI names
+        df = DataFrame(
+            {"A": [1, 2]},
+            index=MultiIndex.from_arrays(
+                [["a", "a"], [0, 1]], names=["idx_level_0", "idx_level_1"]
+            ),
+        )
+        result = df.style._translate(True, True)
+        head = result["head"][1]
+        expected = [
+            {
+                "class": "index_name level0",
+                "display_value": "idx_level_0",
+                "is_visible": True,
+            },
+            {
+                "class": "index_name level1",
+                "display_value": "idx_level_1",
+                "is_visible": True,
+            },
+            {
+                "class": "blank col0",
+                "display_value": blank_value,
+                "is_visible": True,
+            },
+        ]
+        for i, expected_dict in enumerate(expected):
+            assert expected_dict.items() <= head[i].items()
+
+    def test_mi_sparse_column_names(self, blank_value):
+        df = DataFrame(
+            np.arange(16).reshape(4, 4),
+            index=MultiIndex.from_arrays(
+                [["a", "a", "b", "a"], [0, 1, 1, 2]],
+                names=["idx_level_0", "idx_level_1"],
+            ),
+            columns=MultiIndex.from_arrays(
+                [["C1", "C1", "C2", "C2"], [1, 0, 1, 0]], names=["colnam_0", "colnam_1"]
+            ),
+        )
+        result = Styler(df, cell_ids=False)._translate(True, True)
+
+        for level in [0, 1]:
+            head = result["head"][level]
+            expected = [
+                {
+                    "class": "blank",
+                    "display_value": blank_value,
+                    "is_visible": True,
+                },
+                {
+                    "class": f"index_name level{level}",
+                    "display_value": f"colnam_{level}",
+                    "is_visible": True,
+                },
+            ]
+            for i, expected_dict in enumerate(expected):
+                assert expected_dict.items() <= head[i].items()
+
+    def test_hide_column_headers(self, df, styler):
+        ctx = styler.hide(axis="columns")._translate(True, True)
+        assert len(ctx["head"]) == 0  # no header entries with an unnamed index
+
+        df.index.name = "some_name"
+        ctx = df.style.hide(axis="columns")._translate(True, True)
+        assert len(ctx["head"]) == 1
+        # index names still visible, changed in #42101, reverted in 43404
+
+    def test_hide_single_index(self, df):
+        # GH 14194
+        # single unnamed index
+        ctx = df.style._translate(True, True)
+        assert ctx["body"][0][0]["is_visible"]
+        assert ctx["head"][0][0]["is_visible"]
+        ctx2 = df.style.hide(axis="index")._translate(True, True)
+        assert not ctx2["body"][0][0]["is_visible"]
+        assert not ctx2["head"][0][0]["is_visible"]
+
+        # single named index
+        ctx3 = df.set_index("A").style._translate(True, True)
+        assert ctx3["body"][0][0]["is_visible"]
+        assert len(ctx3["head"]) == 2  # 2 header levels
+        assert ctx3["head"][0][0]["is_visible"]
+
+        ctx4 = df.set_index("A").style.hide(axis="index")._translate(True, True)
+        assert not ctx4["body"][0][0]["is_visible"]
+        assert len(ctx4["head"]) == 1  # only 1 header levels
+        assert not ctx4["head"][0][0]["is_visible"]
+
+    def test_hide_multiindex(self):
+        # GH 14194
+        df = DataFrame(
+            {"A": [1, 2], "B": [1, 2]},
+            index=MultiIndex.from_arrays(
+                [["a", "a"], [0, 1]], names=["idx_level_0", "idx_level_1"]
+            ),
+        )
+        ctx1 = df.style._translate(True, True)
+        # tests for 'a' and '0'
+        assert ctx1["body"][0][0]["is_visible"]
+        assert ctx1["body"][0][1]["is_visible"]
+        # check for blank header rows
+        assert len(ctx1["head"][0]) == 4  # two visible indexes and two data columns
+
+        ctx2 = df.style.hide(axis="index")._translate(True, True)
+        # tests for 'a' and '0'
+        assert not ctx2["body"][0][0]["is_visible"]
+        assert not ctx2["body"][0][1]["is_visible"]
+        # check for blank header rows
+        assert len(ctx2["head"][0]) == 3  # one hidden (col name) and two data columns
+        assert not ctx2["head"][0][0]["is_visible"]
+
+    def test_hide_columns_single_level(self, df):
+        # GH 14194
+        # test hiding single column
+        ctx = df.style._translate(True, True)
+        assert ctx["head"][0][1]["is_visible"]
+        assert ctx["head"][0][1]["display_value"] == "A"
+        assert ctx["head"][0][2]["is_visible"]
+        assert ctx["head"][0][2]["display_value"] == "B"
+        assert ctx["body"][0][1]["is_visible"]  # col A, row 1
+        assert ctx["body"][1][2]["is_visible"]  # col B, row 1
+
+        ctx = df.style.hide("A", axis="columns")._translate(True, True)
+        assert not ctx["head"][0][1]["is_visible"]
+        assert not ctx["body"][0][1]["is_visible"]  # col A, row 1
+        assert ctx["body"][1][2]["is_visible"]  # col B, row 1
+
+        # test hiding multiple columns
+        ctx = df.style.hide(["A", "B"], axis="columns")._translate(True, True)
+        assert not ctx["head"][0][1]["is_visible"]
+        assert not ctx["head"][0][2]["is_visible"]
+        assert not ctx["body"][0][1]["is_visible"]  # col A, row 1
+        assert not ctx["body"][1][2]["is_visible"]  # col B, row 1
+
+    def test_hide_columns_index_mult_levels(self):
+        # GH 14194
+        # setup dataframe with multiple column levels and indices
+        i1 = MultiIndex.from_arrays(
+            [["a", "a"], [0, 1]], names=["idx_level_0", "idx_level_1"]
+        )
+        i2 = MultiIndex.from_arrays(
+            [["b", "b"], [0, 1]], names=["col_level_0", "col_level_1"]
+        )
+        df = DataFrame([[1, 2], [3, 4]], index=i1, columns=i2)
+        ctx = df.style._translate(True, True)
+        # column headers
+        assert ctx["head"][0][2]["is_visible"]
+        assert ctx["head"][1][2]["is_visible"]
+        assert ctx["head"][1][3]["display_value"] == "1"
+        # indices
+        assert ctx["body"][0][0]["is_visible"]
+        # data
+        assert ctx["body"][1][2]["is_visible"]
+        assert ctx["body"][1][2]["display_value"] == "3"
+        assert ctx["body"][1][3]["is_visible"]
+        assert ctx["body"][1][3]["display_value"] == "4"
+
+        # hide top column level, which hides both columns
+        ctx = df.style.hide("b", axis="columns")._translate(True, True)
+        assert not ctx["head"][0][2]["is_visible"]  # b
+        assert not ctx["head"][1][2]["is_visible"]  # 0
+        assert not ctx["body"][1][2]["is_visible"]  # 3
+        assert ctx["body"][0][0]["is_visible"]  # index
+
+        # hide first column only
+        ctx = df.style.hide([("b", 0)], axis="columns")._translate(True, True)
+        assert not ctx["head"][0][2]["is_visible"]  # b
+        assert ctx["head"][0][3]["is_visible"]  # b
+        assert not ctx["head"][1][2]["is_visible"]  # 0
+        assert not ctx["body"][1][2]["is_visible"]  # 3
+        assert ctx["body"][1][3]["is_visible"]
+        assert ctx["body"][1][3]["display_value"] == "4"
+
+        # hide second column and index
+        ctx = df.style.hide([("b", 1)], axis=1).hide(axis=0)._translate(True, True)
+        assert not ctx["body"][0][0]["is_visible"]  # index
+        assert len(ctx["head"][0]) == 3
+        assert ctx["head"][0][1]["is_visible"]  # b
+        assert ctx["head"][1][1]["is_visible"]  # 0
+        assert not ctx["head"][1][2]["is_visible"]  # 1
+        assert not ctx["body"][1][3]["is_visible"]  # 4
+        assert ctx["body"][1][2]["is_visible"]
+        assert ctx["body"][1][2]["display_value"] == "3"
+
+        # hide top row level, which hides both rows so body empty
+        ctx = df.style.hide("a", axis="index")._translate(True, True)
+        assert ctx["body"] == []
+
+        # hide first row only
+        ctx = df.style.hide(("a", 0), axis="index")._translate(True, True)
+        for i in [0, 1, 2, 3]:
+            assert "row1" in ctx["body"][0][i]["class"]  # row0 not included in body
+            assert ctx["body"][0][i]["is_visible"]
+
+    def test_pipe(self, df):
+        def set_caption_from_template(styler, a, b):
+            return styler.set_caption(f"Dataframe with a = {a} and b = {b}")
+
+        styler = df.style.pipe(set_caption_from_template, "A", b="B")
+        assert "Dataframe with a = A and b = B" in styler.to_html()
+
+        # Test with an argument that is a (callable, keyword_name) pair.
+        def f(a, b, styler):
+            return (a, b, styler)
+
+        styler = df.style
+        result = styler.pipe((f, "styler"), a=1, b=2)
+        assert result == (1, 2, styler)
+
+    def test_no_cell_ids(self):
+        # GH 35588
+        # GH 35663
+        df = DataFrame(data=[[0]])
+        styler = Styler(df, uuid="_", cell_ids=False)
+        styler.to_html()
+        s = styler.to_html()  # render twice to ensure ctx is not updated
+        assert s.find('<td class="data row0 col0" >') != -1
+
+    @pytest.mark.parametrize(
+        "classes",
+        [
+            DataFrame(
+                data=[["", "test-class"], [np.nan, None]],
+                columns=["A", "B"],
+                index=["a", "b"],
+            ),
+            DataFrame(data=[["test-class"]], columns=["B"], index=["a"]),
+            DataFrame(data=[["test-class", "unused"]], columns=["B", "C"], index=["a"]),
+        ],
+    )
+    def test_set_data_classes(self, classes):
+        # GH 36159
+        df = DataFrame(data=[[0, 1], [2, 3]], columns=["A", "B"], index=["a", "b"])
+        s = Styler(df, uuid_len=0, cell_ids=False).set_td_classes(classes).to_html()
+        assert '<td class="data row0 col0" >0</td>' in s
+        assert '<td class="data row0 col1 test-class" >1</td>' in s
+        assert '<td class="data row1 col0" >2</td>' in s
+        assert '<td class="data row1 col1" >3</td>' in s
+        # GH 39317
+        s = Styler(df, uuid_len=0, cell_ids=True).set_td_classes(classes).to_html()
+        assert '<td id="T__row0_col0" class="data row0 col0" >0</td>' in s
+        assert '<td id="T__row0_col1" class="data row0 col1 test-class" >1</td>' in s
+        assert '<td id="T__row1_col0" class="data row1 col0" >2</td>' in s
+        assert '<td id="T__row1_col1" class="data row1 col1" >3</td>' in s
+
+    def test_set_data_classes_reindex(self):
+        # GH 39317
+        df = DataFrame(
+            data=[[0, 1, 2], [3, 4, 5], [6, 7, 8]], columns=[0, 1, 2], index=[0, 1, 2]
+        )
+        classes = DataFrame(
+            data=[["mi", "ma"], ["mu", "mo"]],
+            columns=[0, 2],
+            index=[0, 2],
+        )
+        s = Styler(df, uuid_len=0).set_td_classes(classes).to_html()
+        assert '<td id="T__row0_col0" class="data row0 col0 mi" >0</td>' in s
+        assert '<td id="T__row0_col2" class="data row0 col2 ma" >2</td>' in s
+        assert '<td id="T__row1_col1" class="data row1 col1" >4</td>' in s
+        assert '<td id="T__row2_col0" class="data row2 col0 mu" >6</td>' in s
+        assert '<td id="T__row2_col2" class="data row2 col2 mo" >8</td>' in s
+
+    def test_chaining_table_styles(self):
+        # GH 35607
+        df = DataFrame(data=[[0, 1], [1, 2]], columns=["A", "B"])
+        styler = df.style.set_table_styles(
+            [{"selector": "", "props": [("background-color", "yellow")]}]
+        ).set_table_styles(
+            [{"selector": ".col0", "props": [("background-color", "blue")]}],
+            overwrite=False,
+        )
+        assert len(styler.table_styles) == 2
+
+    def test_column_and_row_styling(self):
+        # GH 35607
+        df = DataFrame(data=[[0, 1], [1, 2]], columns=["A", "B"])
+        s = Styler(df, uuid_len=0)
+        s = s.set_table_styles({"A": [{"selector": "", "props": [("color", "blue")]}]})
+        assert "#T_ .col0 {\n  color: blue;\n}" in s.to_html()
+        s = s.set_table_styles(
+            {0: [{"selector": "", "props": [("color", "blue")]}]}, axis=1
+        )
+        assert "#T_ .row0 {\n  color: blue;\n}" in s.to_html()
+
+    @pytest.mark.parametrize("len_", [1, 5, 32, 33, 100])
+    def test_uuid_len(self, len_):
+        # GH 36345
+        df = DataFrame(data=[["A"]])
+        s = Styler(df, uuid_len=len_, cell_ids=False).to_html()
+        strt = s.find('id="T_')
+        end = s[strt + 6 :].find('"')
+        if len_ > 32:
+            assert end == 32
+        else:
+            assert end == len_
+
+    @pytest.mark.parametrize("len_", [-2, "bad", None])
+    def test_uuid_len_raises(self, len_):
+        # GH 36345
+        df = DataFrame(data=[["A"]])
+        msg = "``uuid_len`` must be an integer in range \\[0, 32\\]."
+        with pytest.raises(TypeError, match=msg):
+            Styler(df, uuid_len=len_, cell_ids=False).to_html()
+
+    @pytest.mark.parametrize(
+        "slc",
+        [
+            IndexSlice[:, :],
+            IndexSlice[:, 1],
+            IndexSlice[1, :],
+            IndexSlice[[1], [1]],
+            IndexSlice[1, [1]],
+            IndexSlice[[1], 1],
+            IndexSlice[1],
+            IndexSlice[1, 1],
+            slice(None, None, None),
+            [0, 1],
+            np.array([0, 1]),
+            Series([0, 1]),
+        ],
+    )
+    def test_non_reducing_slice(self, slc):
+        df = DataFrame([[0, 1], [2, 3]])
+
+        tslice_ = non_reducing_slice(slc)
+        assert isinstance(df.loc[tslice_], DataFrame)
+
+    @pytest.mark.parametrize("box", [list, Series, np.array])
+    def test_list_slice(self, box):
+        # like dataframe getitem
+        subset = box(["A"])
+
+        df = DataFrame({"A": [1, 2], "B": [3, 4]}, index=["A", "B"])
+        expected = IndexSlice[:, ["A"]]
+
+        result = non_reducing_slice(subset)
+        tm.assert_frame_equal(df.loc[result], df.loc[expected])
+
+    def test_non_reducing_slice_on_multiindex(self):
+        # GH 19861
+        dic = {
+            ("a", "d"): [1, 4],
+            ("a", "c"): [2, 3],
+            ("b", "c"): [3, 2],
+            ("b", "d"): [4, 1],
+        }
+        df = DataFrame(dic, index=[0, 1])
+        idx = IndexSlice
+        slice_ = idx[:, idx["b", "d"]]
+        tslice_ = non_reducing_slice(slice_)
+
+        result = df.loc[tslice_]
+        expected = DataFrame({("b", "d"): [4, 1]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "slice_",
+        [
+            IndexSlice[:, :],
+            # check cols
+            IndexSlice[:, IndexSlice[["a"]]],  # inferred deeper need list
+            IndexSlice[:, IndexSlice[["a"], ["c"]]],  # inferred deeper need list
+            IndexSlice[:, IndexSlice["a", "c", :]],
+            IndexSlice[:, IndexSlice["a", :, "e"]],
+            IndexSlice[:, IndexSlice[:, "c", "e"]],
+            IndexSlice[:, IndexSlice["a", ["c", "d"], :]],  # check list
+            IndexSlice[:, IndexSlice["a", ["c", "d", "-"], :]],  # don't allow missing
+            IndexSlice[:, IndexSlice["a", ["c", "d", "-"], "e"]],  # no slice
+            # check rows
+            IndexSlice[IndexSlice[["U"]], :],  # inferred deeper need list
+            IndexSlice[IndexSlice[["U"], ["W"]], :],  # inferred deeper need list
+            IndexSlice[IndexSlice["U", "W", :], :],
+            IndexSlice[IndexSlice["U", :, "Y"], :],
+            IndexSlice[IndexSlice[:, "W", "Y"], :],
+            IndexSlice[IndexSlice[:, "W", ["Y", "Z"]], :],  # check list
+            IndexSlice[IndexSlice[:, "W", ["Y", "Z", "-"]], :],  # don't allow missing
+            IndexSlice[IndexSlice["U", "W", ["Y", "Z", "-"]], :],  # no slice
+            # check simultaneous
+            IndexSlice[IndexSlice[:, "W", "Y"], IndexSlice["a", "c", :]],
+        ],
+    )
+    def test_non_reducing_multi_slice_on_multiindex(self, slice_):
+        # GH 33562
+        cols = MultiIndex.from_product([["a", "b"], ["c", "d"], ["e", "f"]])
+        idxs = MultiIndex.from_product([["U", "V"], ["W", "X"], ["Y", "Z"]])
+        df = DataFrame(np.arange(64).reshape(8, 8), columns=cols, index=idxs)
+
+        for lvl in [0, 1]:
+            key = slice_[lvl]
+            if isinstance(key, tuple):
+                for subkey in key:
+                    if isinstance(subkey, list) and "-" in subkey:
+                        # not present in the index level, raises KeyError since 2.0
+                        with pytest.raises(KeyError, match="-"):
+                            df.loc[slice_]
+                        return
+
+        expected = df.loc[slice_]
+        result = df.loc[non_reducing_slice(slice_)]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_hidden_index_names(mi_df):
+    mi_df.index.names = ["Lev0", "Lev1"]
+    mi_styler = mi_df.style
+    ctx = mi_styler._translate(True, True)
+    assert len(ctx["head"]) == 3  # 2 column index levels + 1 index names row
+
+    mi_styler.hide(axis="index", names=True)
+    ctx = mi_styler._translate(True, True)
+    assert len(ctx["head"]) == 2  # index names row is unparsed
+    for i in range(4):
+        assert ctx["body"][0][i]["is_visible"]  # 2 index levels + 2 data values visible
+
+    mi_styler.hide(axis="index", level=1)
+    ctx = mi_styler._translate(True, True)
+    assert len(ctx["head"]) == 2  # index names row is still hidden
+    assert ctx["body"][0][0]["is_visible"] is True
+    assert ctx["body"][0][1]["is_visible"] is False
+
+
+def test_hidden_column_names(mi_df):
+    mi_df.columns.names = ["Lev0", "Lev1"]
+    mi_styler = mi_df.style
+    ctx = mi_styler._translate(True, True)
+    assert ctx["head"][0][1]["display_value"] == "Lev0"
+    assert ctx["head"][1][1]["display_value"] == "Lev1"
+
+    mi_styler.hide(names=True, axis="columns")
+    ctx = mi_styler._translate(True, True)
+    assert ctx["head"][0][1]["display_value"] == "&nbsp;"
+    assert ctx["head"][1][1]["display_value"] == "&nbsp;"
+
+    mi_styler.hide(level=0, axis="columns")
+    ctx = mi_styler._translate(True, True)
+    assert len(ctx["head"]) == 1  # no index names and only one visible column headers
+    assert ctx["head"][0][1]["display_value"] == "&nbsp;"
+
+
+@pytest.mark.parametrize("caption", [1, ("a", "b", "c"), (1, "s")])
+def test_caption_raises(mi_styler, caption):
+    msg = "`caption` must be either a string or 2-tuple of strings."
+    with pytest.raises(ValueError, match=msg):
+        mi_styler.set_caption(caption)
+
+
+def test_hiding_headers_over_index_no_sparsify():
+    # GH 43464
+    midx = MultiIndex.from_product([[1, 2], ["a", "a", "b"]])
+    df = DataFrame(9, index=midx, columns=[0])
+    ctx = df.style._translate(False, False)
+    assert len(ctx["body"]) == 6
+    ctx = df.style.hide((1, "a"), axis=0)._translate(False, False)
+    assert len(ctx["body"]) == 4
+    assert "row2" in ctx["body"][0][0]["class"]
+
+
+def test_hiding_headers_over_columns_no_sparsify():
+    # GH 43464
+    midx = MultiIndex.from_product([[1, 2], ["a", "a", "b"]])
+    df = DataFrame(9, columns=midx, index=[0])
+    ctx = df.style._translate(False, False)
+    for ix in [(0, 1), (0, 2), (1, 1), (1, 2)]:
+        assert ctx["head"][ix[0]][ix[1]]["is_visible"] is True
+    ctx = df.style.hide((1, "a"), axis="columns")._translate(False, False)
+    for ix in [(0, 1), (0, 2), (1, 1), (1, 2)]:
+        assert ctx["head"][ix[0]][ix[1]]["is_visible"] is False
+
+
+def test_get_level_lengths_mi_hidden():
+    # GH 43464
+    index = MultiIndex.from_arrays([[1, 1, 1, 2, 2, 2], ["a", "a", "b", "a", "a", "b"]])
+    expected = {
+        (0, 2): 1,
+        (0, 3): 1,
+        (0, 4): 1,
+        (0, 5): 1,
+        (1, 2): 1,
+        (1, 3): 1,
+        (1, 4): 1,
+        (1, 5): 1,
+    }
+    result = _get_level_lengths(
+        index,
+        sparsify=False,
+        max_index=100,
+        hidden_elements=[0, 1, 0, 1],  # hidden element can repeat if duplicated index
+    )
+    tm.assert_dict_equal(result, expected)
+
+
+def test_row_trimming_hide_index():
+    # gh 43703
+    df = DataFrame([[1], [2], [3], [4], [5]])
+    with option_context("styler.render.max_rows", 2):
+        ctx = df.style.hide([0, 1], axis="index")._translate(True, True)
+    assert len(ctx["body"]) == 3
+    for r, val in enumerate(["3", "4", "..."]):
+        assert ctx["body"][r][1]["display_value"] == val
+
+
+def test_row_trimming_hide_index_mi():
+    # gh 44247
+    df = DataFrame([[1], [2], [3], [4], [5]])
+    df.index = MultiIndex.from_product([[0], [0, 1, 2, 3, 4]])
+    with option_context("styler.render.max_rows", 2):
+        ctx = df.style.hide([(0, 0), (0, 1)], axis="index")._translate(True, True)
+    assert len(ctx["body"]) == 3
+
+    # level 0 index headers (sparsified)
+    assert {"value": 0, "attributes": 'rowspan="2"', "is_visible": True}.items() <= ctx[
+        "body"
+    ][0][0].items()
+    assert {"value": 0, "attributes": "", "is_visible": False}.items() <= ctx["body"][
+        1
+    ][0].items()
+    assert {"value": "...", "is_visible": True}.items() <= ctx["body"][2][0].items()
+
+    for r, val in enumerate(["2", "3", "..."]):
+        assert ctx["body"][r][1]["display_value"] == val  # level 1 index headers
+    for r, val in enumerate(["3", "4", "..."]):
+        assert ctx["body"][r][2]["display_value"] == val  # data values
+
+
+def test_col_trimming_hide_columns():
+    # gh 44272
+    df = DataFrame([[1, 2, 3, 4, 5]])
+    with option_context("styler.render.max_columns", 2):
+        ctx = df.style.hide([0, 1], axis="columns")._translate(True, True)
+
+    assert len(ctx["head"][0]) == 6  # blank, [0, 1 (hidden)], [2 ,3 (visible)], + trim
+    for c, vals in enumerate([(1, False), (2, True), (3, True), ("...", True)]):
+        assert ctx["head"][0][c + 2]["value"] == vals[0]
+        assert ctx["head"][0][c + 2]["is_visible"] == vals[1]
+
+    assert len(ctx["body"][0]) == 6  # index + 2 hidden + 2 visible + trimming col
+
+
+def test_no_empty_apply(mi_styler):
+    # 45313
+    mi_styler.apply(lambda s: ["a:v;"] * 2, subset=[False, False])
+    mi_styler._compute()
+
+
+@pytest.mark.parametrize("format", ["html", "latex", "string"])
+def test_output_buffer(mi_styler, format):
+    # gh 47053
+    with tm.ensure_clean(f"delete_me.{format}") as f:
+        getattr(mi_styler, f"to_{format}")(f)
diff --git a/pandas/pandas/tests/io/formats/style/test_to_latex.py b/pandas/pandas/tests/io/formats/style/test_to_latex.py
new file mode 100644
index 0000000000000..eb221686dd165
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_to_latex.py
@@ -0,0 +1,1091 @@
+from textwrap import dedent
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    option_context,
+)
+
+pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+from pandas.io.formats.style_render import (
+    _parse_latex_cell_styles,
+    _parse_latex_css_conversion,
+    _parse_latex_header_span,
+    _parse_latex_table_styles,
+    _parse_latex_table_wrapping,
+)
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        {"A": [0, 1], "B": [-0.61, -1.22], "C": Series(["ab", "cd"], dtype=object)}
+    )
+
+
+@pytest.fixture
+def df_ext():
+    return DataFrame(
+        {"A": [0, 1, 2], "B": [-0.61, -1.22, -2.22], "C": ["ab", "cd", "de"]}
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0, precision=2)
+
+
+def test_minimal_latex_tabular(styler):
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+         & A & B & C \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{tabular}
+        """
+    )
+    assert styler.to_latex() == expected
+
+
+def test_tabular_hrules(styler):
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+        \\toprule
+         & A & B & C \\\\
+        \\midrule
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\bottomrule
+        \\end{tabular}
+        """
+    )
+    assert styler.to_latex(hrules=True) == expected
+
+
+def test_tabular_custom_hrules(styler):
+    styler.set_table_styles(
+        [
+            {"selector": "toprule", "props": ":hline"},
+            {"selector": "bottomrule", "props": ":otherline"},
+        ]
+    )  # no midrule
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+        \\hline
+         & A & B & C \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\otherline
+        \\end{tabular}
+        """
+    )
+    assert styler.to_latex() == expected
+
+
+def test_column_format(styler):
+    # default setting is already tested in `test_latex_minimal_tabular`
+    styler.set_table_styles([{"selector": "column_format", "props": ":cccc"}])
+
+    assert "\\begin{tabular}{rrrr}" in styler.to_latex(column_format="rrrr")
+    styler.set_table_styles([{"selector": "column_format", "props": ":r|r|cc"}])
+    assert "\\begin{tabular}{r|r|cc}" in styler.to_latex()
+
+
+def test_siunitx_cols(styler):
+    expected = dedent(
+        """\
+        \\begin{tabular}{lSSl}
+        {} & {A} & {B} & {C} \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{tabular}
+        """
+    )
+    assert styler.to_latex(siunitx=True) == expected
+
+
+def test_position(styler):
+    assert "\\begin{table}[h!]" in styler.to_latex(position="h!")
+    assert "\\end{table}" in styler.to_latex(position="h!")
+    styler.set_table_styles([{"selector": "position", "props": ":b!"}])
+    assert "\\begin{table}[b!]" in styler.to_latex()
+    assert "\\end{table}" in styler.to_latex()
+
+
+@pytest.mark.parametrize("env", [None, "longtable"])
+def test_label(styler, env):
+    assert "\n\\label{text}" in styler.to_latex(label="text", environment=env)
+    styler.set_table_styles([{"selector": "label", "props": ":{more §text}"}])
+    assert "\n\\label{more :text}" in styler.to_latex(environment=env)
+
+
+def test_position_float_raises(styler):
+    msg = "`position_float` should be one of 'raggedright', 'raggedleft', 'centering',"
+    with pytest.raises(ValueError, match=msg):
+        styler.to_latex(position_float="bad_string")
+
+    msg = "`position_float` cannot be used in 'longtable' `environment`"
+    with pytest.raises(ValueError, match=msg):
+        styler.to_latex(position_float="centering", environment="longtable")
+
+
+@pytest.mark.parametrize("label", [(None, ""), ("text", "\\label{text}")])
+@pytest.mark.parametrize("position", [(None, ""), ("h!", "{table}[h!]")])
+@pytest.mark.parametrize("caption", [(None, ""), ("text", "\\caption{text}")])
+@pytest.mark.parametrize("column_format", [(None, ""), ("rcrl", "{tabular}{rcrl}")])
+@pytest.mark.parametrize("position_float", [(None, ""), ("centering", "\\centering")])
+def test_kwargs_combinations(
+    styler, label, position, caption, column_format, position_float
+):
+    result = styler.to_latex(
+        label=label[0],
+        position=position[0],
+        caption=caption[0],
+        column_format=column_format[0],
+        position_float=position_float[0],
+    )
+    assert label[1] in result
+    assert position[1] in result
+    assert caption[1] in result
+    assert column_format[1] in result
+    assert position_float[1] in result
+
+
+def test_custom_table_styles(styler):
+    styler.set_table_styles(
+        [
+            {"selector": "mycommand", "props": ":{myoptions}"},
+            {"selector": "mycommand2", "props": ":{myoptions2}"},
+        ]
+    )
+    expected = dedent(
+        """\
+        \\begin{table}
+        \\mycommand{myoptions}
+        \\mycommand2{myoptions2}
+        """
+    )
+    assert expected in styler.to_latex()
+
+
+def test_cell_styling(styler):
+    styler.highlight_max(props="itshape:;Huge:--wrap;")
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+         & A & B & C \\\\
+        0 & 0 & \\itshape {\\Huge -0.61} & ab \\\\
+        1 & \\itshape {\\Huge 1} & -1.22 & \\itshape {\\Huge cd} \\\\
+        \\end{tabular}
+        """
+    )
+    assert expected == styler.to_latex()
+
+
+def test_multiindex_columns(df):
+    cidx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df.columns = cidx
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+         & \\multicolumn{2}{r}{A} & B \\\\
+         & a & b & c \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{tabular}
+        """
+    )
+    s = df.style.format(precision=2)
+    assert expected == s.to_latex()
+
+    # non-sparse
+    expected = dedent(
+        """\
+        \\begin{tabular}{lrrl}
+         & A & A & B \\\\
+         & a & b & c \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{tabular}
+        """
+    )
+    s = df.style.format(precision=2)
+    assert expected == s.to_latex(sparse_columns=False)
+
+
+def test_multiindex_row(df_ext):
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index = ridx
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrl}
+         &  & A & B & C \\\\
+        \\multirow[c]{2}{*}{A} & a & 0 & -0.61 & ab \\\\
+         & b & 1 & -1.22 & cd \\\\
+        B & c & 2 & -2.22 & de \\\\
+        \\end{tabular}
+        """
+    )
+    styler = df_ext.style.format(precision=2)
+    result = styler.to_latex()
+    assert expected == result
+
+    # non-sparse
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrl}
+         &  & A & B & C \\\\
+        A & a & 0 & -0.61 & ab \\\\
+        A & b & 1 & -1.22 & cd \\\\
+        B & c & 2 & -2.22 & de \\\\
+        \\end{tabular}
+        """
+    )
+    result = styler.to_latex(sparse_index=False)
+    assert expected == result
+
+
+def test_multirow_naive(df_ext):
+    ridx = MultiIndex.from_tuples([("X", "x"), ("X", "y"), ("Y", "z")])
+    df_ext.index = ridx
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrl}
+         &  & A & B & C \\\\
+        X & x & 0 & -0.61 & ab \\\\
+         & y & 1 & -1.22 & cd \\\\
+        Y & z & 2 & -2.22 & de \\\\
+        \\end{tabular}
+        """
+    )
+    styler = df_ext.style.format(precision=2)
+    result = styler.to_latex(multirow_align="naive")
+    assert expected == result
+
+
+def test_multiindex_row_and_col(df_ext):
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index, df_ext.columns = ridx, cidx
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrl}
+         &  & \\multicolumn{2}{l}{Z} & Y \\\\
+         &  & a & b & c \\\\
+        \\multirow[b]{2}{*}{A} & a & 0 & -0.61 & ab \\\\
+         & b & 1 & -1.22 & cd \\\\
+        B & c & 2 & -2.22 & de \\\\
+        \\end{tabular}
+        """
+    )
+    styler = df_ext.style.format(precision=2)
+    result = styler.to_latex(multirow_align="b", multicol_align="l")
+    assert result == expected
+
+    # non-sparse
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrl}
+         &  & Z & Z & Y \\\\
+         &  & a & b & c \\\\
+        A & a & 0 & -0.61 & ab \\\\
+        A & b & 1 & -1.22 & cd \\\\
+        B & c & 2 & -2.22 & de \\\\
+        \\end{tabular}
+        """
+    )
+    result = styler.to_latex(sparse_index=False, sparse_columns=False)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "multicol_align, siunitx, header",
+    [
+        ("naive-l", False, " & A & &"),
+        ("naive-r", False, " & & & A"),
+        ("naive-l", True, "{} & {A} & {} & {}"),
+        ("naive-r", True, "{} & {} & {} & {A}"),
+    ],
+)
+def test_multicol_naive(df, multicol_align, siunitx, header):
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("A", "c")])
+    df.columns = ridx
+    level1 = " & a & b & c" if not siunitx else "{} & {a} & {b} & {c}"
+    col_format = "lrrl" if not siunitx else "lSSl"
+    expected = dedent(
+        f"""\
+        \\begin{{tabular}}{{{col_format}}}
+        {header} \\\\
+        {level1} \\\\
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{{tabular}}
+        """
+    )
+    styler = df.style.format(precision=2)
+    result = styler.to_latex(multicol_align=multicol_align, siunitx=siunitx)
+    assert expected == result
+
+
+def test_multi_options(df_ext):
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index, df_ext.columns = ridx, cidx
+    styler = df_ext.style.format(precision=2)
+
+    expected = dedent(
+        """\
+     &  & \\multicolumn{2}{r}{Z} & Y \\\\
+     &  & a & b & c \\\\
+    \\multirow[c]{2}{*}{A} & a & 0 & -0.61 & ab \\\\
+    """
+    )
+    result = styler.to_latex()
+    assert expected in result
+
+    with option_context("styler.latex.multicol_align", "l"):
+        assert " &  & \\multicolumn{2}{l}{Z} & Y \\\\" in styler.to_latex()
+
+    with option_context("styler.latex.multirow_align", "b"):
+        assert "\\multirow[b]{2}{*}{A} & a & 0 & -0.61 & ab \\\\" in styler.to_latex()
+
+
+def test_multiindex_columns_hidden():
+    df = DataFrame([[1, 2, 3, 4]])
+    df.columns = MultiIndex.from_tuples([("A", 1), ("A", 2), ("A", 3), ("B", 1)])
+    s = df.style
+    assert "{tabular}{lrrrr}" in s.to_latex()
+    s.set_table_styles([])  # reset the position command
+    s.hide([("A", 2)], axis="columns")
+    assert "{tabular}{lrrr}" in s.to_latex()
+
+
+@pytest.mark.parametrize(
+    "option, value",
+    [
+        ("styler.sparse.index", True),
+        ("styler.sparse.index", False),
+        ("styler.sparse.columns", True),
+        ("styler.sparse.columns", False),
+    ],
+)
+def test_sparse_options(df_ext, option, value):
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index, df_ext.columns = ridx, cidx
+    styler = df_ext.style
+
+    latex1 = styler.to_latex()
+    with option_context(option, value):
+        latex2 = styler.to_latex()
+    assert (latex1 == latex2) is value
+
+
+def test_hidden_index(styler):
+    styler.hide(axis="index")
+    expected = dedent(
+        """\
+        \\begin{tabular}{rrl}
+        A & B & C \\\\
+        0 & -0.61 & ab \\\\
+        1 & -1.22 & cd \\\\
+        \\end{tabular}
+        """
+    )
+    assert styler.to_latex() == expected
+
+
+@pytest.mark.parametrize("environment", ["table", "figure*", None])
+def test_comprehensive(df_ext, environment):
+    # test as many low level features simultaneously as possible
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index, df_ext.columns = ridx, cidx
+    stlr = df_ext.style
+    stlr.set_caption("mycap")
+    stlr.set_table_styles(
+        [
+            {"selector": "label", "props": ":{fig§item}"},
+            {"selector": "position", "props": ":h!"},
+            {"selector": "position_float", "props": ":centering"},
+            {"selector": "column_format", "props": ":rlrlr"},
+            {"selector": "toprule", "props": ":toprule"},
+            {"selector": "midrule", "props": ":midrule"},
+            {"selector": "bottomrule", "props": ":bottomrule"},
+            {"selector": "rowcolors", "props": ":{3}{pink}{}"},  # custom command
+        ]
+    )
+    stlr.highlight_max(axis=0, props="textbf:--rwrap;cellcolor:[rgb]{1,1,0.6}--rwrap")
+    stlr.highlight_max(axis=None, props="Huge:--wrap;", subset=[("Z", "a"), ("Z", "b")])
+
+    expected = (
+        """\
+\\begin{table}[h!]
+\\centering
+\\caption{mycap}
+\\label{fig:item}
+\\rowcolors{3}{pink}{}
+\\begin{tabular}{rlrlr}
+\\toprule
+ &  & \\multicolumn{2}{r}{Z} & Y \\\\
+ &  & a & b & c \\\\
+\\midrule
+\\multirow[c]{2}{*}{A} & a & 0 & \\textbf{\\cellcolor[rgb]{1,1,0.6}{-0.61}} & ab \\\\
+ & b & 1 & -1.22 & cd \\\\
+B & c & \\textbf{\\cellcolor[rgb]{1,1,0.6}{{\\Huge 2}}} & -2.22 & """
+        """\
+\\textbf{\\cellcolor[rgb]{1,1,0.6}{de}} \\\\
+\\bottomrule
+\\end{tabular}
+\\end{table}
+"""
+    ).replace("table", environment if environment else "table")
+    result = stlr.format(precision=2).to_latex(environment=environment)
+    assert result == expected
+
+
+def test_environment_option(styler):
+    with option_context("styler.latex.environment", "bar-env"):
+        assert "\\begin{bar-env}" in styler.to_latex()
+        assert "\\begin{foo-env}" in styler.to_latex(environment="foo-env")
+
+
+def test_parse_latex_table_styles(styler):
+    styler.set_table_styles(
+        [
+            {"selector": "foo", "props": [("attr", "value")]},
+            {"selector": "bar", "props": [("attr", "overwritten")]},
+            {"selector": "bar", "props": [("attr", "baz"), ("attr2", "ignored")]},
+            {"selector": "label", "props": [("", "{fig§item}")]},
+        ]
+    )
+    assert _parse_latex_table_styles(styler.table_styles, "bar") == "baz"
+
+    # test '§' replaced by ':' [for CSS compatibility]
+    assert _parse_latex_table_styles(styler.table_styles, "label") == "{fig:item}"
+
+
+def test_parse_latex_cell_styles_basic():  # test nesting
+    cell_style = [("itshape", "--rwrap"), ("cellcolor", "[rgb]{0,1,1}--rwrap")]
+    expected = "\\itshape{\\cellcolor[rgb]{0,1,1}{text}}"
+    assert _parse_latex_cell_styles(cell_style, "text") == expected
+
+
+@pytest.mark.parametrize(
+    "wrap_arg, expected",
+    [  # test wrapping
+        ("", "\\<command><options> <display_value>"),
+        ("--wrap", "{\\<command><options> <display_value>}"),
+        ("--nowrap", "\\<command><options> <display_value>"),
+        ("--lwrap", "{\\<command><options>} <display_value>"),
+        ("--dwrap", "{\\<command><options>}{<display_value>}"),
+        ("--rwrap", "\\<command><options>{<display_value>}"),
+    ],
+)
+def test_parse_latex_cell_styles_braces(wrap_arg, expected):
+    cell_style = [("<command>", f"<options>{wrap_arg}")]
+    assert _parse_latex_cell_styles(cell_style, "<display_value>") == expected
+
+
+def test_parse_latex_header_span():
+    cell = {"attributes": 'colspan="3"', "display_value": "text", "cellstyle": []}
+    expected = "\\multicolumn{3}{Y}{text}"
+    assert _parse_latex_header_span(cell, "X", "Y") == expected
+
+    cell = {"attributes": 'rowspan="5"', "display_value": "text", "cellstyle": []}
+    expected = "\\multirow[X]{5}{*}{text}"
+    assert _parse_latex_header_span(cell, "X", "Y") == expected
+
+    cell = {"display_value": "text", "cellstyle": []}
+    assert _parse_latex_header_span(cell, "X", "Y") == "text"
+
+    cell = {"display_value": "text", "cellstyle": [("bfseries", "--rwrap")]}
+    assert _parse_latex_header_span(cell, "X", "Y") == "\\bfseries{text}"
+
+
+def test_parse_latex_table_wrapping(styler):
+    styler.set_table_styles(
+        [
+            {"selector": "toprule", "props": ":value"},
+            {"selector": "bottomrule", "props": ":value"},
+            {"selector": "midrule", "props": ":value"},
+            {"selector": "column_format", "props": ":value"},
+        ]
+    )
+    assert _parse_latex_table_wrapping(styler.table_styles, styler.caption) is False
+    assert _parse_latex_table_wrapping(styler.table_styles, "some caption") is True
+    styler.set_table_styles(
+        [
+            {"selector": "not-ignored", "props": ":value"},
+        ],
+        overwrite=False,
+    )
+    assert _parse_latex_table_wrapping(styler.table_styles, None) is True
+
+
+def test_short_caption(styler):
+    result = styler.to_latex(caption=("full cap", "short cap"))
+    assert "\\caption[short cap]{full cap}" in result
+
+
+@pytest.mark.parametrize(
+    "css, expected",
+    [
+        ([("color", "red")], [("color", "{red}")]),  # test color and input format types
+        (
+            [("color", "rgb(128, 128, 128 )")],
+            [("color", "[rgb]{0.502, 0.502, 0.502}")],
+        ),
+        (
+            [("color", "rgb(128, 50%, 25% )")],
+            [("color", "[rgb]{0.502, 0.500, 0.250}")],
+        ),
+        (
+            [("color", "rgba(128,128,128,1)")],
+            [("color", "[rgb]{0.502, 0.502, 0.502}")],
+        ),
+        ([("color", "#FF00FF")], [("color", "[HTML]{FF00FF}")]),
+        ([("color", "#F0F")], [("color", "[HTML]{FF00FF}")]),
+        ([("font-weight", "bold")], [("bfseries", "")]),  # test font-weight and types
+        ([("font-weight", "bolder")], [("bfseries", "")]),
+        ([("font-weight", "normal")], []),
+        ([("background-color", "red")], [("cellcolor", "{red}--lwrap")]),
+        (
+            [("background-color", "#FF00FF")],  # test background-color command and wrap
+            [("cellcolor", "[HTML]{FF00FF}--lwrap")],
+        ),
+        ([("font-style", "italic")], [("itshape", "")]),  # test font-style and types
+        ([("font-style", "oblique")], [("slshape", "")]),
+        ([("font-style", "normal")], []),
+        ([("color", "red /*--dwrap*/")], [("color", "{red}--dwrap")]),  # css comments
+        ([("background-color", "red /* --dwrap */")], [("cellcolor", "{red}--dwrap")]),
+    ],
+)
+def test_parse_latex_css_conversion(css, expected):
+    result = _parse_latex_css_conversion(css)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "env, inner_env",
+    [
+        (None, "tabular"),
+        ("table", "tabular"),
+        ("longtable", "longtable"),
+    ],
+)
+@pytest.mark.parametrize(
+    "convert, exp", [(True, "bfseries"), (False, "font-weightbold")]
+)
+def test_parse_latex_css_convert_minimal(styler, env, inner_env, convert, exp):
+    # parameters ensure longtable template is also tested
+    styler.highlight_max(props="font-weight:bold;")
+    result = styler.to_latex(convert_css=convert, environment=env)
+    expected = dedent(
+        f"""\
+        0 & 0 & \\{exp} -0.61 & ab \\\\
+        1 & \\{exp} 1 & -1.22 & \\{exp} cd \\\\
+        \\end{{{inner_env}}}
+    """
+    )
+    assert expected in result
+
+
+def test_parse_latex_css_conversion_option():
+    css = [("command", "option--latex--wrap")]
+    expected = [("command", "option--wrap")]
+    result = _parse_latex_css_conversion(css)
+    assert result == expected
+
+
+def test_styler_object_after_render(styler):
+    # GH 42320
+    pre_render = styler._copy(deepcopy=True)
+    styler.to_latex(
+        column_format="rllr",
+        position="h",
+        position_float="centering",
+        hrules=True,
+        label="my lab",
+        caption="my cap",
+    )
+
+    assert pre_render.table_styles == styler.table_styles
+    assert pre_render.caption == styler.caption
+
+
+def test_longtable_comprehensive(styler):
+    result = styler.to_latex(
+        environment="longtable", hrules=True, label="fig:A", caption=("full", "short")
+    )
+    expected = dedent(
+        """\
+        \\begin{longtable}{lrrl}
+        \\caption[short]{full} \\label{fig:A} \\\\
+        \\toprule
+         & A & B & C \\\\
+        \\midrule
+        \\endfirsthead
+        \\caption[]{full} \\\\
+        \\toprule
+         & A & B & C \\\\
+        \\midrule
+        \\endhead
+        \\midrule
+        \\multicolumn{4}{r}{Continued on next page} \\\\
+        \\midrule
+        \\endfoot
+        \\bottomrule
+        \\endlastfoot
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{longtable}
+    """
+    )
+    assert result == expected
+
+
+def test_longtable_minimal(styler):
+    result = styler.to_latex(environment="longtable")
+    expected = dedent(
+        """\
+        \\begin{longtable}{lrrl}
+         & A & B & C \\\\
+        \\endfirsthead
+         & A & B & C \\\\
+        \\endhead
+        \\multicolumn{4}{r}{Continued on next page} \\\\
+        \\endfoot
+        \\endlastfoot
+        0 & 0 & -0.61 & ab \\\\
+        1 & 1 & -1.22 & cd \\\\
+        \\end{longtable}
+    """
+    )
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "sparse, exp, siunitx",
+    [
+        (True, "{} & \\multicolumn{2}{r}{A} & {B}", True),
+        (False, "{} & {A} & {A} & {B}", True),
+        (True, " & \\multicolumn{2}{r}{A} & B", False),
+        (False, " & A & A & B", False),
+    ],
+)
+def test_longtable_multiindex_columns(df, sparse, exp, siunitx):
+    cidx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df.columns = cidx
+    with_si = "{} & {a} & {b} & {c} \\\\"
+    without_si = " & a & b & c \\\\"
+    expected = dedent(
+        f"""\
+        \\begin{{longtable}}{{l{"SS" if siunitx else "rr"}l}}
+        {exp} \\\\
+        {with_si if siunitx else without_si}
+        \\endfirsthead
+        {exp} \\\\
+        {with_si if siunitx else without_si}
+        \\endhead
+        """
+    )
+    result = df.style.to_latex(
+        environment="longtable", sparse_columns=sparse, siunitx=siunitx
+    )
+    assert expected in result
+
+
+@pytest.mark.parametrize(
+    "caption, cap_exp",
+    [
+        ("full", ("{full}", "")),
+        (("full", "short"), ("{full}", "[short]")),
+    ],
+)
+@pytest.mark.parametrize("label, lab_exp", [(None, ""), ("tab:A", " \\label{tab:A}")])
+def test_longtable_caption_label(styler, caption, cap_exp, label, lab_exp):
+    cap_exp1 = f"\\caption{cap_exp[1]}{cap_exp[0]}"
+    cap_exp2 = f"\\caption[]{cap_exp[0]}"
+
+    expected = dedent(
+        f"""\
+        {cap_exp1}{lab_exp} \\\\
+         & A & B & C \\\\
+        \\endfirsthead
+        {cap_exp2} \\\\
+        """
+    )
+    assert expected in styler.to_latex(
+        environment="longtable", caption=caption, label=label
+    )
+
+
+@pytest.mark.parametrize("index", [True, False])
+@pytest.mark.parametrize(
+    "columns, siunitx",
+    [
+        (True, True),
+        (True, False),
+        (False, False),
+    ],
+)
+def test_apply_map_header_render_mi(df_ext, index, columns, siunitx):
+    cidx = MultiIndex.from_tuples([("Z", "a"), ("Z", "b"), ("Y", "c")])
+    ridx = MultiIndex.from_tuples([("A", "a"), ("A", "b"), ("B", "c")])
+    df_ext.index, df_ext.columns = ridx, cidx
+    styler = df_ext.style
+
+    func = lambda v: "bfseries: --rwrap" if "A" in v or "Z" in v or "c" in v else None
+
+    if index:
+        styler.map_index(func, axis="index")
+    if columns:
+        styler.map_index(func, axis="columns")
+
+    result = styler.to_latex(siunitx=siunitx)
+
+    expected_index = dedent(
+        """\
+    \\multirow[c]{2}{*}{\\bfseries{A}} & a & 0 & -0.610000 & ab \\\\
+    \\bfseries{} & b & 1 & -1.220000 & cd \\\\
+    B & \\bfseries{c} & 2 & -2.220000 & de \\\\
+    """
+    )
+    assert (expected_index in result) is index
+
+    exp_cols_si = dedent(
+        """\
+    {} & {} & \\multicolumn{2}{r}{\\bfseries{Z}} & {Y} \\\\
+    {} & {} & {a} & {b} & {\\bfseries{c}} \\\\
+    """
+    )
+    exp_cols_no_si = """\
+ &  & \\multicolumn{2}{r}{\\bfseries{Z}} & Y \\\\
+ &  & a & b & \\bfseries{c} \\\\
+"""
+    assert ((exp_cols_si if siunitx else exp_cols_no_si) in result) is columns
+
+
+def test_repr_option(styler):
+    assert "<style" in styler._repr_html_()[:6]
+    assert styler._repr_latex_() is None
+    with option_context("styler.render.repr", "latex"):
+        assert "\\begin{tabular}" in styler._repr_latex_()[:15]
+        assert styler._repr_html_() is None
+
+
+@pytest.mark.parametrize("option", ["hrules"])
+def test_bool_options(styler, option):
+    with option_context(f"styler.latex.{option}", False):
+        latex_false = styler.to_latex()
+    with option_context(f"styler.latex.{option}", True):
+        latex_true = styler.to_latex()
+    assert latex_false != latex_true  # options are reactive under to_latex(*no_args)
+
+
+def test_siunitx_basic_headers(styler):
+    assert "{} & {A} & {B} & {C} \\\\" in styler.to_latex(siunitx=True)
+    assert " & A & B & C \\\\" in styler.to_latex()  # default siunitx=False
+
+
+@pytest.mark.parametrize("axis", ["index", "columns"])
+def test_css_convert_apply_index(styler, axis):
+    styler.map_index(lambda x: "font-weight: bold;", axis=axis)
+    for label in getattr(styler, axis):
+        assert f"\\bfseries {label}" in styler.to_latex(convert_css=True)
+
+
+def test_hide_index_latex(styler):
+    # GH 43637
+    styler.hide([0], axis=0)
+    result = styler.to_latex()
+    expected = dedent(
+        """\
+    \\begin{tabular}{lrrl}
+     & A & B & C \\\\
+    1 & 1 & -1.22 & cd \\\\
+    \\end{tabular}
+    """
+    )
+    assert expected == result
+
+
+def test_latex_hiding_index_columns_multiindex_alignment():
+    # gh 43644
+    midx = MultiIndex.from_product(
+        [["i0", "j0"], ["i1"], ["i2", "j2"]], names=["i-0", "i-1", "i-2"]
+    )
+    cidx = MultiIndex.from_product(
+        [["c0"], ["c1", "d1"], ["c2", "d2"]], names=["c-0", "c-1", "c-2"]
+    )
+    df = DataFrame(np.arange(16).reshape(4, 4), index=midx, columns=cidx)
+    styler = Styler(df, uuid_len=0)
+    styler.hide(level=1, axis=0).hide(level=0, axis=1)
+    styler.hide([("i0", "i1", "i2")], axis=0)
+    styler.hide([("c0", "c1", "c2")], axis=1)
+    styler.map(lambda x: "color:{red};" if x == 5 else "")
+    styler.map_index(lambda x: "color:{blue};" if "j" in x else "")
+    result = styler.to_latex()
+    expected = dedent(
+        """\
+        \\begin{tabular}{llrrr}
+         & c-1 & c1 & \\multicolumn{2}{r}{d1} \\\\
+         & c-2 & d2 & c2 & d2 \\\\
+        i-0 & i-2 &  &  &  \\\\
+        i0 & \\color{blue} j2 & \\color{red} 5 & 6 & 7 \\\\
+        \\multirow[c]{2}{*}{\\color{blue} j0} & i2 & 9 & 10 & 11 \\\\
+        \\color{blue}  & \\color{blue} j2 & 13 & 14 & 15 \\\\
+        \\end{tabular}
+        """
+    )
+    assert result == expected
+
+
+def test_rendered_links():
+    # note the majority of testing is done in test_html.py: test_rendered_links
+    # these test only the alternative latex format is functional
+    df = DataFrame(["text www.domain.com text"])
+    result = df.style.format(hyperlinks="latex").to_latex()
+    assert r"text \href{www.domain.com}{www.domain.com} text" in result
+
+
+def test_apply_index_hidden_levels():
+    # gh 45156
+    styler = DataFrame(
+        [[1]],
+        index=MultiIndex.from_tuples([(0, 1)], names=["l0", "l1"]),
+        columns=MultiIndex.from_tuples([(0, 1)], names=["c0", "c1"]),
+    ).style
+    styler.hide(level=1)
+    styler.map_index(lambda v: "color: red;", level=0, axis=1)
+    result = styler.to_latex(convert_css=True)
+    expected = dedent(
+        """\
+        \\begin{tabular}{lr}
+        c0 & \\color{red} 0 \\\\
+        c1 & 1 \\\\
+        l0 &  \\\\
+        0 & 1 \\\\
+        \\end{tabular}
+        """
+    )
+    assert result == expected
+
+
+@pytest.mark.parametrize("clines", ["bad", "index", "skip-last", "all", "data"])
+def test_clines_validation(clines, styler):
+    msg = f"`clines` value of {clines} is invalid."
+    with pytest.raises(ValueError, match=msg):
+        styler.to_latex(clines=clines)
+
+
+@pytest.mark.parametrize(
+    "clines, exp",
+    [
+        ("all;index", "\n\\cline{1-1}"),
+        ("all;data", "\n\\cline{1-2}"),
+        ("skip-last;index", ""),
+        ("skip-last;data", ""),
+        (None, ""),
+    ],
+)
+@pytest.mark.parametrize("env", ["table", "longtable"])
+def test_clines_index(clines, exp, env):
+    df = DataFrame([[1], [2], [3], [4]])
+    result = df.style.to_latex(clines=clines, environment=env)
+    expected = f"""\
+0 & 1 \\\\{exp}
+1 & 2 \\\\{exp}
+2 & 3 \\\\{exp}
+3 & 4 \\\\{exp}
+"""
+    assert expected in result
+
+
+@pytest.mark.parametrize(
+    "clines, expected",
+    [
+        (
+            None,
+            dedent(
+                """\
+            \\multirow[c]{2}{*}{A} & X & 1 \\\\
+             & Y & 2 \\\\
+            \\multirow[c]{2}{*}{B} & X & 3 \\\\
+             & Y & 4 \\\\
+            """
+            ),
+        ),
+        (
+            "skip-last;index",
+            dedent(
+                """\
+            \\multirow[c]{2}{*}{A} & X & 1 \\\\
+             & Y & 2 \\\\
+            \\cline{1-2}
+            \\multirow[c]{2}{*}{B} & X & 3 \\\\
+             & Y & 4 \\\\
+            \\cline{1-2}
+            """
+            ),
+        ),
+        (
+            "skip-last;data",
+            dedent(
+                """\
+            \\multirow[c]{2}{*}{A} & X & 1 \\\\
+             & Y & 2 \\\\
+            \\cline{1-3}
+            \\multirow[c]{2}{*}{B} & X & 3 \\\\
+             & Y & 4 \\\\
+            \\cline{1-3}
+            """
+            ),
+        ),
+        (
+            "all;index",
+            dedent(
+                """\
+            \\multirow[c]{2}{*}{A} & X & 1 \\\\
+            \\cline{2-2}
+             & Y & 2 \\\\
+            \\cline{1-2} \\cline{2-2}
+            \\multirow[c]{2}{*}{B} & X & 3 \\\\
+            \\cline{2-2}
+             & Y & 4 \\\\
+            \\cline{1-2} \\cline{2-2}
+            """
+            ),
+        ),
+        (
+            "all;data",
+            dedent(
+                """\
+            \\multirow[c]{2}{*}{A} & X & 1 \\\\
+            \\cline{2-3}
+             & Y & 2 \\\\
+            \\cline{1-3} \\cline{2-3}
+            \\multirow[c]{2}{*}{B} & X & 3 \\\\
+            \\cline{2-3}
+             & Y & 4 \\\\
+            \\cline{1-3} \\cline{2-3}
+            """
+            ),
+        ),
+    ],
+)
+@pytest.mark.parametrize("env", ["table"])
+def test_clines_multiindex(clines, expected, env):
+    # also tests simultaneously with hidden rows and a hidden multiindex level
+    midx = MultiIndex.from_product([["A", "-", "B"], [0], ["X", "Y"]])
+    df = DataFrame([[1], [2], [99], [99], [3], [4]], index=midx)
+    styler = df.style
+    styler.hide([("-", 0, "X"), ("-", 0, "Y")])
+    styler.hide(level=1)
+    result = styler.to_latex(clines=clines, environment=env)
+    assert expected in result
+
+
+def test_col_format_len(styler):
+    # gh 46037
+    result = styler.to_latex(environment="longtable", column_format="lrr{10cm}")
+    expected = r"\multicolumn{4}{r}{Continued on next page} \\"
+    assert expected in result
+
+
+def test_concat(styler):
+    result = styler.concat(styler.data.agg(["sum"]).style).to_latex()
+    expected = dedent(
+        """\
+    \\begin{tabular}{lrrl}
+     & A & B & C \\\\
+    0 & 0 & -0.61 & ab \\\\
+    1 & 1 & -1.22 & cd \\\\
+    sum & 1 & -1.830000 & abcd \\\\
+    \\end{tabular}
+    """
+    )
+    assert result == expected
+
+
+def test_concat_recursion():
+    # tests hidden row recursion and applied styles
+    styler1 = DataFrame([[1], [9]]).style.hide([1]).highlight_min(color="red")
+    styler2 = DataFrame([[9], [2]]).style.hide([0]).highlight_min(color="green")
+    styler3 = DataFrame([[3], [9]]).style.hide([1]).highlight_min(color="blue")
+
+    result = styler1.concat(styler2.concat(styler3)).to_latex(convert_css=True)
+    expected = dedent(
+        """\
+    \\begin{tabular}{lr}
+     & 0 \\\\
+    0 & {\\cellcolor{red}} 1 \\\\
+    1 & {\\cellcolor{green}} 2 \\\\
+    0 & {\\cellcolor{blue}} 3 \\\\
+    \\end{tabular}
+    """
+    )
+    assert result == expected
+
+
+def test_concat_chain():
+    # tests hidden row recursion and applied styles
+    styler1 = DataFrame([[1], [9]]).style.hide([1]).highlight_min(color="red")
+    styler2 = DataFrame([[9], [2]]).style.hide([0]).highlight_min(color="green")
+    styler3 = DataFrame([[3], [9]]).style.hide([1]).highlight_min(color="blue")
+
+    result = styler1.concat(styler2).concat(styler3).to_latex(convert_css=True)
+    expected = dedent(
+        """\
+    \\begin{tabular}{lr}
+     & 0 \\\\
+    0 & {\\cellcolor{red}} 1 \\\\
+    1 & {\\cellcolor{green}} 2 \\\\
+    0 & {\\cellcolor{blue}} 3 \\\\
+    \\end{tabular}
+    """
+    )
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "columns, expected",
+    [
+        (
+            None,
+            dedent(
+                """\
+            \\begin{tabular}{l}
+            \\end{tabular}
+            """
+            ),
+        ),
+        (
+            ["a", "b", "c"],
+            dedent(
+                """\
+            \\begin{tabular}{llll}
+             & a & b & c \\\\
+            \\end{tabular}
+            """
+            ),
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    "clines", [None, "all;data", "all;index", "skip-last;data", "skip-last;index"]
+)
+def test_empty_clines(columns, expected: str, clines: str):
+    # GH 47203
+    df = DataFrame(columns=columns)
+    result = df.style.to_latex(clines=clines)
+    assert result == expected
diff --git a/pandas/pandas/tests/io/formats/style/test_to_string.py b/pandas/pandas/tests/io/formats/style/test_to_string.py
new file mode 100644
index 0000000000000..e6f84cc9e9cd8
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_to_string.py
@@ -0,0 +1,96 @@
+from textwrap import dedent
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+
+pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        {"A": [0, 1], "B": [-0.61, -1.22], "C": Series(["ab", "cd"], dtype=object)}
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0, precision=2)
+
+
+def test_basic_string(styler):
+    result = styler.to_string()
+    expected = dedent(
+        """\
+     A B C
+    0 0 -0.61 ab
+    1 1 -1.22 cd
+    """
+    )
+    assert result == expected
+
+
+def test_string_delimiter(styler):
+    result = styler.to_string(delimiter=";")
+    expected = dedent(
+        """\
+    ;A;B;C
+    0;0;-0.61;ab
+    1;1;-1.22;cd
+    """
+    )
+    assert result == expected
+
+
+def test_concat(styler):
+    result = styler.concat(styler.data.agg(["sum"]).style).to_string()
+    expected = dedent(
+        """\
+     A B C
+    0 0 -0.61 ab
+    1 1 -1.22 cd
+    sum 1 -1.830000 abcd
+    """
+    )
+    assert result == expected
+
+
+def test_concat_recursion(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["sum"]), uuid_len=0, precision=3)
+    styler3 = Styler(df.agg(["sum"]), uuid_len=0, precision=4)
+    result = styler1.concat(styler2.concat(styler3)).to_string()
+    expected = dedent(
+        """\
+     A B C
+    0 0 -0.61 ab
+    1 1 -1.22 cd
+    sum 1 -1.830 abcd
+    sum 1 -1.8300 abcd
+    """
+    )
+    assert result == expected
+
+
+def test_concat_chain(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["sum"]), uuid_len=0, precision=3)
+    styler3 = Styler(df.agg(["sum"]), uuid_len=0, precision=4)
+    result = styler1.concat(styler2).concat(styler3).to_string()
+    expected = dedent(
+        """\
+     A B C
+    0 0 -0.61 ab
+    1 1 -1.22 cd
+    sum 1 -1.830 abcd
+    sum 1 -1.8300 abcd
+    """
+    )
+    assert result == expected
diff --git a/pandas/pandas/tests/io/formats/style/test_to_typst.py b/pandas/pandas/tests/io/formats/style/test_to_typst.py
new file mode 100644
index 0000000000000..2365119c9c4dc
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_to_typst.py
@@ -0,0 +1,96 @@
+from textwrap import dedent
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+
+pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        {"A": [0, 1], "B": [-0.61, -1.22], "C": Series(["ab", "cd"], dtype=object)}
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0, precision=2)
+
+
+def test_basic_table(styler):
+    result = styler.to_typst()
+    expected = dedent(
+        """\
+    #table(
+      columns: 4,
+      [], [A], [B], [C],
+
+      [0], [0], [-0.61], [ab],
+      [1], [1], [-1.22], [cd],
+    )"""
+    )
+    assert result == expected
+
+
+def test_concat(styler):
+    result = styler.concat(styler.data.agg(["sum"]).style).to_typst()
+    expected = dedent(
+        """\
+    #table(
+      columns: 4,
+      [], [A], [B], [C],
+
+      [0], [0], [-0.61], [ab],
+      [1], [1], [-1.22], [cd],
+      [sum], [1], [-1.830000], [abcd],
+    )"""
+    )
+    assert result == expected
+
+
+def test_concat_recursion(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["sum"]), uuid_len=0, precision=3)
+    styler3 = Styler(df.agg(["sum"]), uuid_len=0, precision=4)
+    result = styler1.concat(styler2.concat(styler3)).to_typst()
+    expected = dedent(
+        """\
+    #table(
+      columns: 4,
+      [], [A], [B], [C],
+
+      [0], [0], [-0.61], [ab],
+      [1], [1], [-1.22], [cd],
+      [sum], [1], [-1.830], [abcd],
+      [sum], [1], [-1.8300], [abcd],
+    )"""
+    )
+    assert result == expected
+
+
+def test_concat_chain(styler):
+    df = styler.data
+    styler1 = styler
+    styler2 = Styler(df.agg(["sum"]), uuid_len=0, precision=3)
+    styler3 = Styler(df.agg(["sum"]), uuid_len=0, precision=4)
+    result = styler1.concat(styler2).concat(styler3).to_typst()
+    expected = dedent(
+        """\
+    #table(
+      columns: 4,
+      [], [A], [B], [C],
+
+      [0], [0], [-0.61], [ab],
+      [1], [1], [-1.22], [cd],
+      [sum], [1], [-1.830], [abcd],
+      [sum], [1], [-1.8300], [abcd],
+    )"""
+    )
+    assert result == expected
diff --git a/pandas/pandas/tests/io/formats/style/test_tooltip.py b/pandas/pandas/tests/io/formats/style/test_tooltip.py
new file mode 100644
index 0000000000000..f1071c949299e
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/style/test_tooltip.py
@@ -0,0 +1,179 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+
+pytest.importorskip("jinja2")
+from pandas.io.formats.style import Styler
+
+
+@pytest.fixture
+def df():
+    return DataFrame(
+        data=[[0, 1, 2], [3, 4, 5], [6, 7, 8]],
+        columns=["A", "B", "C"],
+        index=["x", "y", "z"],
+    )
+
+
+@pytest.fixture
+def styler(df):
+    return Styler(df, uuid_len=0)
+
+
+@pytest.mark.parametrize(
+    "data, columns, index",
+    [
+        # Test basic reindex and ignoring blank
+        ([["Min", "Max"], [np.nan, ""]], ["A", "C"], ["x", "y"]),
+        # Test non-referenced columns, reversed col names, short index
+        ([["Max", "Min", "Bad-Col"]], ["C", "A", "D"], ["x"]),
+    ],
+)
+def test_tooltip_render(data, columns, index, styler):
+    ttips = DataFrame(data=data, columns=columns, index=index)
+
+    # GH 21266
+    result = styler.set_tooltips(ttips).to_html()
+
+    # test tooltip table level class
+    assert "#T_ .pd-t {\n  visibility: hidden;\n" in result
+
+    # test 'Min' tooltip added
+    assert "#T_ #T__row0_col0:hover .pd-t {\n  visibility: visible;\n}" in result
+    assert '#T_ #T__row0_col0 .pd-t::after {\n  content: "Min";\n}' in result
+    assert 'class="data row0 col0" >0<span class="pd-t"></span></td>' in result
+
+    # test 'Max' tooltip added
+    assert "#T_ #T__row0_col2:hover .pd-t {\n  visibility: visible;\n}" in result
+    assert '#T_ #T__row0_col2 .pd-t::after {\n  content: "Max";\n}' in result
+    assert 'class="data row0 col2" >2<span class="pd-t"></span></td>' in result
+
+    # test Nan, empty string and bad column ignored
+    assert "#T_ #T__row1_col0:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row0_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col2:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "Bad-Col" not in result
+
+
+def test_tooltip_ignored(styler):
+    # GH 21266
+    result = styler.to_html()  # no set_tooltips() creates no <span>
+    assert '<style type="text/css">\n</style>' in result
+    assert '<span class="pd-t"></span>' not in result
+    assert 'title="' not in result
+
+
+def test_tooltip_css_class(styler):
+    # GH 21266
+    result = styler.set_tooltips(
+        DataFrame([["tooltip"]], index=["x"], columns=["A"]),
+        css_class="other-class",
+        props=[("color", "green")],
+    ).to_html()
+    assert "#T_ .other-class {\n  color: green;\n" in result
+    assert '#T_ #T__row0_col0 .other-class::after {\n  content: "tooltip";\n' in result
+
+    # GH 39563
+    result = styler.set_tooltips(  # set_tooltips overwrites previous
+        DataFrame([["tooltip"]], index=["x"], columns=["A"]),
+        css_class="another-class",
+        props="color:green;color:red;",
+    ).to_html()
+    assert "#T_ .another-class {\n  color: green;\n  color: red;\n}" in result
+
+
+@pytest.mark.parametrize(
+    "data, columns, index",
+    [
+        # Test basic reindex and ignoring blank
+        ([["Min", "Max"], [np.nan, ""]], ["A", "C"], ["x", "y"]),
+        # Test non-referenced columns, reversed col names, short index
+        ([["Max", "Min", "Bad-Col"]], ["C", "A", "D"], ["x"]),
+    ],
+)
+def test_tooltip_render_as_title(data, columns, index, styler):
+    ttips = DataFrame(data=data, columns=columns, index=index)
+    # GH 56605
+    result = styler.set_tooltips(ttips, as_title_attribute=True).to_html()
+
+    # test css not added
+    assert "#T_ .pd-t {\n  visibility: hidden;\n" not in result
+
+    # test 'Min' tooltip added as title attribute and css does not exist
+    assert "#T_ #T__row0_col0:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert '#T_ #T__row0_col0 .pd-t::after {\n  content: "Min";\n}' not in result
+    assert 'class="data row0 col0"  title="Min">0</td>' in result
+
+    # test 'Max' tooltip added as title attribute and css does not exist
+    assert "#T_ #T__row0_col2:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert '#T_ #T__row0_col2 .pd-t::after {\n  content: "Max";\n}' not in result
+    assert 'class="data row0 col2"  title="Max">2</td>' in result
+
+    # test Nan, empty string and bad column ignored
+    assert "#T_ #T__row1_col0:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row0_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col2:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "Bad-Col" not in result
+    assert 'class="data row0 col1" >1</td>' in result
+    assert 'class="data row1 col0" >3</td>' in result
+    assert 'class="data row1 col1" >4</td>' in result
+    assert 'class="data row1 col2" >5</td>' in result
+    assert 'class="data row2 col0" >6</td>' in result
+    assert 'class="data row2 col1" >7</td>' in result
+    assert 'class="data row2 col2" >8</td>' in result
+
+
+def test_tooltip_render_as_title_with_hidden_index_level():
+    df = DataFrame(
+        data=[[0, 1, 2], [3, 4, 5], [6, 7, 8]],
+        columns=["A", "B", "C"],
+        index=MultiIndex.from_arrays(
+            [["x", "y", "z"], [1, 2, 3], ["aa", "bb", "cc"]],
+            names=["alpha", "num", "char"],
+        ),
+    )
+    ttips = DataFrame(
+        # Test basic reindex and ignoring blank, and hide level 2 (num) from index
+        data=[["Min", "Max"], [np.nan, ""]],
+        columns=["A", "C"],
+        index=MultiIndex.from_arrays(
+            [["x", "y"], [1, 2], ["aa", "bb"]], names=["alpha", "num", "char"]
+        ),
+    )
+    styler = Styler(df, uuid_len=0)
+    styler = styler.hide(axis=0, level=-1, names=True)
+    # GH 56605
+    result = styler.set_tooltips(ttips, as_title_attribute=True).to_html()
+
+    # test css not added
+    assert "#T_ .pd-t {\n  visibility: hidden;\n" not in result
+
+    # test 'Min' tooltip added as title attribute and css does not exist
+    assert "#T_ #T__row0_col0:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert '#T_ #T__row0_col0 .pd-t::after {\n  content: "Min";\n}' not in result
+    assert 'class="data row0 col0"  title="Min">0</td>' in result
+
+    # test 'Max' tooltip added as title attribute and css does not exist
+    assert "#T_ #T__row0_col2:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert '#T_ #T__row0_col2 .pd-t::after {\n  content: "Max";\n}' not in result
+    assert 'class="data row0 col2"  title="Max">2</td>' in result
+
+    # test Nan, empty string and bad column ignored
+    assert "#T_ #T__row1_col0:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row0_col1:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "#T_ #T__row1_col2:hover .pd-t {\n  visibility: visible;\n}" not in result
+    assert "Bad-Col" not in result
+    assert 'class="data row0 col1" >1</td>' in result
+    assert 'class="data row1 col0" >3</td>' in result
+    assert 'class="data row1 col1" >4</td>' in result
+    assert 'class="data row1 col2" >5</td>' in result
+    assert 'class="data row2 col0" >6</td>' in result
+    assert 'class="data row2 col1" >7</td>' in result
+    assert 'class="data row2 col2" >8</td>' in result
diff --git a/pandas/pandas/tests/io/formats/test_console.py b/pandas/pandas/tests/io/formats/test_console.py
new file mode 100644
index 0000000000000..dd7b57df9baed
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_console.py
@@ -0,0 +1,72 @@
+import locale
+
+import pytest
+
+from pandas._config import detect_console_encoding
+
+
+class MockEncoding:
+    """
+    Used to add a side effect when accessing the 'encoding' property. If the
+    side effect is a str in nature, the value will be returned. Otherwise, the
+    side effect should be an exception that will be raised.
+    """
+
+    def __init__(self, encoding) -> None:
+        super().__init__()
+        self.val = encoding
+
+    @property
+    def encoding(self):
+        return self.raise_or_return(self.val)
+
+    @staticmethod
+    def raise_or_return(val):
+        if isinstance(val, str):
+            return val
+        else:
+            raise val
+
+
+@pytest.mark.parametrize("empty,filled", [["stdin", "stdout"], ["stdout", "stdin"]])
+def test_detect_console_encoding_from_stdout_stdin(monkeypatch, empty, filled):
+    # Ensures that when sys.stdout.encoding or sys.stdin.encoding is used when
+    # they have values filled.
+    # GH 21552
+    with monkeypatch.context() as context:
+        context.setattr(f"sys.{empty}", MockEncoding(""))
+        context.setattr(f"sys.{filled}", MockEncoding(filled))
+        assert detect_console_encoding() == filled
+
+
+@pytest.mark.parametrize("encoding", [AttributeError, OSError, "ascii"])
+def test_detect_console_encoding_fallback_to_locale(monkeypatch, encoding):
+    # GH 21552
+    with monkeypatch.context() as context:
+        context.setattr("locale.getpreferredencoding", lambda: "foo")
+        context.setattr("sys.stdout", MockEncoding(encoding))
+        assert detect_console_encoding() == "foo"
+
+
+@pytest.mark.parametrize(
+    "std,locale",
+    [
+        ["ascii", "ascii"],
+        ["ascii", locale.Error],
+        [AttributeError, "ascii"],
+        [AttributeError, locale.Error],
+        [OSError, "ascii"],
+        [OSError, locale.Error],
+    ],
+)
+def test_detect_console_encoding_fallback_to_default(monkeypatch, std, locale):
+    # When both the stdout/stdin encoding and locale preferred encoding checks
+    # fail (or return 'ascii', we should default to the sys default encoding.
+    # GH 21552
+    with monkeypatch.context() as context:
+        context.setattr(
+            "locale.getpreferredencoding", lambda: MockEncoding.raise_or_return(locale)
+        )
+        context.setattr("sys.stdout", MockEncoding(std))
+        context.setattr("sys.getdefaultencoding", lambda: "sysDefaultEncoding")
+        assert detect_console_encoding() == "sysDefaultEncoding"
diff --git a/pandas/pandas/tests/io/formats/test_css.py b/pandas/pandas/tests/io/formats/test_css.py
new file mode 100644
index 0000000000000..642a562704344
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_css.py
@@ -0,0 +1,288 @@
+import pytest
+
+from pandas.errors import CSSWarning
+
+import pandas._testing as tm
+
+from pandas.io.formats.css import CSSResolver
+
+
+def assert_resolves(css, props, inherited=None):
+    resolve = CSSResolver()
+    actual = resolve(css, inherited=inherited)
+    assert props == actual
+
+
+def assert_same_resolution(css1, css2, inherited=None):
+    resolve = CSSResolver()
+    resolved1 = resolve(css1, inherited=inherited)
+    resolved2 = resolve(css2, inherited=inherited)
+    assert resolved1 == resolved2
+
+
+@pytest.mark.parametrize(
+    "name,norm,abnorm",
+    [
+        (
+            "whitespace",
+            "hello: world; foo: bar",
+            " \t hello \t :\n  world \n  ;  \n foo: \tbar\n\n",
+        ),
+        ("case", "hello: world; foo: bar", "Hello: WORLD; foO: bar"),
+        ("empty-decl", "hello: world; foo: bar", "; hello: world;; foo: bar;\n; ;"),
+        ("empty-list", "", ";"),
+    ],
+)
+def test_css_parse_normalisation(name, norm, abnorm):
+    assert_same_resolution(norm, abnorm)
+
+
+@pytest.mark.parametrize(
+    "invalid_css,remainder,msg",
+    [
+        # No colon
+        ("hello-world", "", "expected a colon"),
+        ("border-style: solid; hello-world", "border-style: solid", "expected a colon"),
+        (
+            "border-style: solid; hello-world; font-weight: bold",
+            "border-style: solid; font-weight: bold",
+            "expected a colon",
+        ),
+        # Unclosed string fail
+        # Invalid size
+        ("font-size: blah", "font-size: 1em", "Unhandled size"),
+        ("font-size: 1a2b", "font-size: 1em", "Unhandled size"),
+        ("font-size: 1e5pt", "font-size: 1em", "Unhandled size"),
+        ("font-size: 1+6pt", "font-size: 1em", "Unhandled size"),
+        ("font-size: 1unknownunit", "font-size: 1em", "Unhandled size"),
+        ("font-size: 10", "font-size: 1em", "Unhandled size"),
+        ("font-size: 10 pt", "font-size: 1em", "Unhandled size"),
+        # Too many args
+        ("border-top: 1pt solid red green", "border-top: 1pt solid green", "Too many"),
+    ],
+)
+def test_css_parse_invalid(invalid_css, remainder, msg):
+    with tm.assert_produces_warning(CSSWarning, match=msg):
+        assert_same_resolution(invalid_css, remainder)
+
+
+@pytest.mark.parametrize(
+    "shorthand,expansions",
+    [
+        ("margin", ["margin-top", "margin-right", "margin-bottom", "margin-left"]),
+        ("padding", ["padding-top", "padding-right", "padding-bottom", "padding-left"]),
+        (
+            "border-width",
+            [
+                "border-top-width",
+                "border-right-width",
+                "border-bottom-width",
+                "border-left-width",
+            ],
+        ),
+        (
+            "border-color",
+            [
+                "border-top-color",
+                "border-right-color",
+                "border-bottom-color",
+                "border-left-color",
+            ],
+        ),
+        (
+            "border-style",
+            [
+                "border-top-style",
+                "border-right-style",
+                "border-bottom-style",
+                "border-left-style",
+            ],
+        ),
+    ],
+)
+def test_css_side_shorthands(shorthand, expansions):
+    top, right, bottom, left = expansions
+
+    assert_resolves(
+        f"{shorthand}: 1pt", {top: "1pt", right: "1pt", bottom: "1pt", left: "1pt"}
+    )
+
+    assert_resolves(
+        f"{shorthand}: 1pt 4pt", {top: "1pt", right: "4pt", bottom: "1pt", left: "4pt"}
+    )
+
+    assert_resolves(
+        f"{shorthand}: 1pt 4pt 2pt",
+        {top: "1pt", right: "4pt", bottom: "2pt", left: "4pt"},
+    )
+
+    assert_resolves(
+        f"{shorthand}: 1pt 4pt 2pt 0pt",
+        {top: "1pt", right: "4pt", bottom: "2pt", left: "0pt"},
+    )
+
+    with tm.assert_produces_warning(CSSWarning, match="Could not expand"):
+        assert_resolves(f"{shorthand}: 1pt 1pt 1pt 1pt 1pt", {})
+
+
+@pytest.mark.parametrize(
+    "shorthand,sides",
+    [
+        ("border-top", ["top"]),
+        ("border-right", ["right"]),
+        ("border-bottom", ["bottom"]),
+        ("border-left", ["left"]),
+        ("border", ["top", "right", "bottom", "left"]),
+    ],
+)
+def test_css_border_shorthand_sides(shorthand, sides):
+    def create_border_dict(sides, color=None, style=None, width=None):
+        resolved = {}
+        for side in sides:
+            if color:
+                resolved[f"border-{side}-color"] = color
+            if style:
+                resolved[f"border-{side}-style"] = style
+            if width:
+                resolved[f"border-{side}-width"] = width
+        return resolved
+
+    assert_resolves(
+        f"{shorthand}: 1pt red solid", create_border_dict(sides, "red", "solid", "1pt")
+    )
+
+
+@pytest.mark.parametrize(
+    "prop, expected",
+    [
+        ("1pt red solid", ("red", "solid", "1pt")),
+        ("red 1pt solid", ("red", "solid", "1pt")),
+        ("red solid 1pt", ("red", "solid", "1pt")),
+        ("solid 1pt red", ("red", "solid", "1pt")),
+        ("red solid", ("red", "solid", "1.500000pt")),
+        # Note: color=black is not CSS conforming
+        # (See https://drafts.csswg.org/css-backgrounds/#border-shorthands)
+        ("1pt solid", ("black", "solid", "1pt")),
+        ("1pt red", ("red", "none", "1pt")),
+        ("red", ("red", "none", "1.500000pt")),
+        ("1pt", ("black", "none", "1pt")),
+        ("solid", ("black", "solid", "1.500000pt")),
+        # Sizes
+        ("1em", ("black", "none", "12pt")),
+    ],
+)
+def test_css_border_shorthands(prop, expected):
+    color, style, width = expected
+
+    assert_resolves(
+        f"border-left: {prop}",
+        {
+            "border-left-color": color,
+            "border-left-style": style,
+            "border-left-width": width,
+        },
+    )
+
+
+@pytest.mark.parametrize(
+    "style,inherited,equiv",
+    [
+        ("margin: 1px; margin: 2px", "", "margin: 2px"),
+        ("margin: 1px", "margin: 2px", "margin: 1px"),
+        ("margin: 1px; margin: inherit", "margin: 2px", "margin: 2px"),
+        (
+            "margin: 1px; margin-top: 2px",
+            "",
+            "margin-left: 1px; margin-right: 1px; margin-bottom: 1px; margin-top: 2px",
+        ),
+        ("margin-top: 2px", "margin: 1px", "margin: 1px; margin-top: 2px"),
+        ("margin: 1px", "margin-top: 2px", "margin: 1px"),
+        (
+            "margin: 1px; margin-top: inherit",
+            "margin: 2px",
+            "margin: 1px; margin-top: 2px",
+        ),
+    ],
+)
+def test_css_precedence(style, inherited, equiv):
+    resolve = CSSResolver()
+    inherited_props = resolve(inherited)
+    style_props = resolve(style, inherited=inherited_props)
+    equiv_props = resolve(equiv)
+    assert style_props == equiv_props
+
+
+@pytest.mark.parametrize(
+    "style,equiv",
+    [
+        (
+            "margin: 1px; margin-top: inherit",
+            "margin-bottom: 1px; margin-right: 1px; margin-left: 1px",
+        ),
+        ("margin-top: inherit", ""),
+        ("margin-top: initial", ""),
+    ],
+)
+def test_css_none_absent(style, equiv):
+    assert_same_resolution(style, equiv)
+
+
+@pytest.mark.parametrize(
+    "size,resolved",
+    [
+        ("xx-small", "6pt"),
+        ("x-small", f"{7.5:f}pt"),
+        ("small", f"{9.6:f}pt"),
+        ("medium", "12pt"),
+        ("large", f"{13.5:f}pt"),
+        ("x-large", "18pt"),
+        ("xx-large", "24pt"),
+        ("8px", "6pt"),
+        ("1.25pc", "15pt"),
+        (".25in", "18pt"),
+        ("02.54cm", "72pt"),
+        ("25.4mm", "72pt"),
+        ("101.6q", "72pt"),
+    ],
+)
+@pytest.mark.parametrize("relative_to", [None, "16pt"])  # invariant to inherited size
+def test_css_absolute_font_size(size, relative_to, resolved):
+    if relative_to is None:
+        inherited = None
+    else:
+        inherited = {"font-size": relative_to}
+    assert_resolves(f"font-size: {size}", {"font-size": resolved}, inherited=inherited)
+
+
+@pytest.mark.parametrize(
+    "size,relative_to,resolved",
+    [
+        ("1em", None, "12pt"),
+        ("1.0em", None, "12pt"),
+        ("1.25em", None, "15pt"),
+        ("1em", "16pt", "16pt"),
+        ("1.0em", "16pt", "16pt"),
+        ("1.25em", "16pt", "20pt"),
+        ("1rem", "16pt", "12pt"),
+        ("1.0rem", "16pt", "12pt"),
+        ("1.25rem", "16pt", "15pt"),
+        ("100%", None, "12pt"),
+        ("125%", None, "15pt"),
+        ("100%", "16pt", "16pt"),
+        ("125%", "16pt", "20pt"),
+        ("2ex", None, "12pt"),
+        ("2.0ex", None, "12pt"),
+        ("2.50ex", None, "15pt"),
+        ("inherit", "16pt", "16pt"),
+        ("smaller", None, "10pt"),
+        ("smaller", "18pt", "15pt"),
+        ("larger", None, f"{14.4:f}pt"),
+        ("larger", "15pt", "18pt"),
+    ],
+)
+def test_css_relative_font_size(size, relative_to, resolved):
+    if relative_to is None:
+        inherited = None
+    else:
+        inherited = {"font-size": relative_to}
+    assert_resolves(f"font-size: {size}", {"font-size": resolved}, inherited=inherited)
diff --git a/pandas/pandas/tests/io/formats/test_eng_formatting.py b/pandas/pandas/tests/io/formats/test_eng_formatting.py
new file mode 100644
index 0000000000000..6d581b5b92e0c
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_eng_formatting.py
@@ -0,0 +1,254 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    reset_option,
+    set_eng_float_format,
+)
+
+from pandas.io.formats.format import EngFormatter
+
+
+@pytest.fixture(autouse=True)
+def reset_float_format():
+    yield
+    reset_option("display.float_format")
+
+
+class TestEngFormatter:
+    def test_eng_float_formatter2(self, float_frame):
+        df = float_frame
+        df.loc[5] = 0
+
+        set_eng_float_format()
+        repr(df)
+
+        set_eng_float_format(use_eng_prefix=True)
+        repr(df)
+
+        set_eng_float_format(accuracy=0)
+        repr(df)
+
+    def test_eng_float_formatter(self):
+        df = DataFrame({"A": [1.41, 141.0, 14100, 1410000.0]})
+
+        set_eng_float_format()
+        result = df.to_string()
+        expected = (
+            "             A\n"
+            "0    1.410E+00\n"
+            "1  141.000E+00\n"
+            "2   14.100E+03\n"
+            "3    1.410E+06"
+        )
+        assert result == expected
+
+        set_eng_float_format(use_eng_prefix=True)
+        result = df.to_string()
+        expected = "         A\n0    1.410\n1  141.000\n2  14.100k\n3   1.410M"
+        assert result == expected
+
+        set_eng_float_format(accuracy=0)
+        result = df.to_string()
+        expected = "         A\n0    1E+00\n1  141E+00\n2   14E+03\n3    1E+06"
+        assert result == expected
+
+    def compare(self, formatter, input, output):
+        formatted_input = formatter(input)
+        assert formatted_input == output
+
+    def compare_all(self, formatter, in_out):
+        """
+        Parameters:
+        -----------
+        formatter: EngFormatter under test
+        in_out: list of tuples. Each tuple = (number, expected_formatting)
+
+        It is tested if 'formatter(number) == expected_formatting'.
+        *number* should be >= 0 because formatter(-number) == fmt is also
+        tested. *fmt* is derived from *expected_formatting*
+        """
+        for input, output in in_out:
+            self.compare(formatter, input, output)
+            self.compare(formatter, -input, "-" + output[1:])
+
+    def test_exponents_with_eng_prefix(self):
+        formatter = EngFormatter(accuracy=3, use_eng_prefix=True)
+        f = np.sqrt(2)
+        in_out = [
+            (f * 10**-24, " 1.414y"),
+            (f * 10**-23, " 14.142y"),
+            (f * 10**-22, " 141.421y"),
+            (f * 10**-21, " 1.414z"),
+            (f * 10**-20, " 14.142z"),
+            (f * 10**-19, " 141.421z"),
+            (f * 10**-18, " 1.414a"),
+            (f * 10**-17, " 14.142a"),
+            (f * 10**-16, " 141.421a"),
+            (f * 10**-15, " 1.414f"),
+            (f * 10**-14, " 14.142f"),
+            (f * 10**-13, " 141.421f"),
+            (f * 10**-12, " 1.414p"),
+            (f * 10**-11, " 14.142p"),
+            (f * 10**-10, " 141.421p"),
+            (f * 10**-9, " 1.414n"),
+            (f * 10**-8, " 14.142n"),
+            (f * 10**-7, " 141.421n"),
+            (f * 10**-6, " 1.414u"),
+            (f * 10**-5, " 14.142u"),
+            (f * 10**-4, " 141.421u"),
+            (f * 10**-3, " 1.414m"),
+            (f * 10**-2, " 14.142m"),
+            (f * 10**-1, " 141.421m"),
+            (f * 10**0, " 1.414"),
+            (f * 10**1, " 14.142"),
+            (f * 10**2, " 141.421"),
+            (f * 10**3, " 1.414k"),
+            (f * 10**4, " 14.142k"),
+            (f * 10**5, " 141.421k"),
+            (f * 10**6, " 1.414M"),
+            (f * 10**7, " 14.142M"),
+            (f * 10**8, " 141.421M"),
+            (f * 10**9, " 1.414G"),
+            (f * 10**10, " 14.142G"),
+            (f * 10**11, " 141.421G"),
+            (f * 10**12, " 1.414T"),
+            (f * 10**13, " 14.142T"),
+            (f * 10**14, " 141.421T"),
+            (f * 10**15, " 1.414P"),
+            (f * 10**16, " 14.142P"),
+            (f * 10**17, " 141.421P"),
+            (f * 10**18, " 1.414E"),
+            (f * 10**19, " 14.142E"),
+            (f * 10**20, " 141.421E"),
+            (f * 10**21, " 1.414Z"),
+            (f * 10**22, " 14.142Z"),
+            (f * 10**23, " 141.421Z"),
+            (f * 10**24, " 1.414Y"),
+            (f * 10**25, " 14.142Y"),
+            (f * 10**26, " 141.421Y"),
+        ]
+        self.compare_all(formatter, in_out)
+
+    def test_exponents_without_eng_prefix(self):
+        formatter = EngFormatter(accuracy=4, use_eng_prefix=False)
+        f = np.pi
+        in_out = [
+            (f * 10**-24, " 3.1416E-24"),
+            (f * 10**-23, " 31.4159E-24"),
+            (f * 10**-22, " 314.1593E-24"),
+            (f * 10**-21, " 3.1416E-21"),
+            (f * 10**-20, " 31.4159E-21"),
+            (f * 10**-19, " 314.1593E-21"),
+            (f * 10**-18, " 3.1416E-18"),
+            (f * 10**-17, " 31.4159E-18"),
+            (f * 10**-16, " 314.1593E-18"),
+            (f * 10**-15, " 3.1416E-15"),
+            (f * 10**-14, " 31.4159E-15"),
+            (f * 10**-13, " 314.1593E-15"),
+            (f * 10**-12, " 3.1416E-12"),
+            (f * 10**-11, " 31.4159E-12"),
+            (f * 10**-10, " 314.1593E-12"),
+            (f * 10**-9, " 3.1416E-09"),
+            (f * 10**-8, " 31.4159E-09"),
+            (f * 10**-7, " 314.1593E-09"),
+            (f * 10**-6, " 3.1416E-06"),
+            (f * 10**-5, " 31.4159E-06"),
+            (f * 10**-4, " 314.1593E-06"),
+            (f * 10**-3, " 3.1416E-03"),
+            (f * 10**-2, " 31.4159E-03"),
+            (f * 10**-1, " 314.1593E-03"),
+            (f * 10**0, " 3.1416E+00"),
+            (f * 10**1, " 31.4159E+00"),
+            (f * 10**2, " 314.1593E+00"),
+            (f * 10**3, " 3.1416E+03"),
+            (f * 10**4, " 31.4159E+03"),
+            (f * 10**5, " 314.1593E+03"),
+            (f * 10**6, " 3.1416E+06"),
+            (f * 10**7, " 31.4159E+06"),
+            (f * 10**8, " 314.1593E+06"),
+            (f * 10**9, " 3.1416E+09"),
+            (f * 10**10, " 31.4159E+09"),
+            (f * 10**11, " 314.1593E+09"),
+            (f * 10**12, " 3.1416E+12"),
+            (f * 10**13, " 31.4159E+12"),
+            (f * 10**14, " 314.1593E+12"),
+            (f * 10**15, " 3.1416E+15"),
+            (f * 10**16, " 31.4159E+15"),
+            (f * 10**17, " 314.1593E+15"),
+            (f * 10**18, " 3.1416E+18"),
+            (f * 10**19, " 31.4159E+18"),
+            (f * 10**20, " 314.1593E+18"),
+            (f * 10**21, " 3.1416E+21"),
+            (f * 10**22, " 31.4159E+21"),
+            (f * 10**23, " 314.1593E+21"),
+            (f * 10**24, " 3.1416E+24"),
+            (f * 10**25, " 31.4159E+24"),
+            (f * 10**26, " 314.1593E+24"),
+        ]
+        self.compare_all(formatter, in_out)
+
+    def test_rounding(self):
+        formatter = EngFormatter(accuracy=3, use_eng_prefix=True)
+        in_out = [
+            (5.55555, " 5.556"),
+            (55.5555, " 55.556"),
+            (555.555, " 555.555"),
+            (5555.55, " 5.556k"),
+            (55555.5, " 55.556k"),
+            (555555, " 555.555k"),
+        ]
+        self.compare_all(formatter, in_out)
+
+        formatter = EngFormatter(accuracy=1, use_eng_prefix=True)
+        in_out = [
+            (5.55555, " 5.6"),
+            (55.5555, " 55.6"),
+            (555.555, " 555.6"),
+            (5555.55, " 5.6k"),
+            (55555.5, " 55.6k"),
+            (555555, " 555.6k"),
+        ]
+        self.compare_all(formatter, in_out)
+
+        formatter = EngFormatter(accuracy=0, use_eng_prefix=True)
+        in_out = [
+            (5.55555, " 6"),
+            (55.5555, " 56"),
+            (555.555, " 556"),
+            (5555.55, " 6k"),
+            (55555.5, " 56k"),
+            (555555, " 556k"),
+        ]
+        self.compare_all(formatter, in_out)
+
+        formatter = EngFormatter(accuracy=3, use_eng_prefix=True)
+        result = formatter(0)
+        assert result == " 0.000"
+
+    def test_nan(self):
+        # Issue #11981
+
+        formatter = EngFormatter(accuracy=1, use_eng_prefix=True)
+        result = formatter(np.nan)
+        assert result == "NaN"
+
+        df = DataFrame(
+            {
+                "a": [1.5, 10.3, 20.5],
+                "b": [50.3, 60.67, 70.12],
+                "c": [100.2, 101.33, 120.33],
+            }
+        )
+        pt = df.pivot_table(values="a", index="b", columns="c")
+        set_eng_float_format(accuracy=1)
+        result = pt.to_string()
+        assert "NaN" in result
+
+    def test_inf(self):
+        # Issue #11981
+
+        formatter = EngFormatter(accuracy=1, use_eng_prefix=True)
+        result = formatter(np.inf)
+        assert result == "inf"
diff --git a/pandas/pandas/tests/io/formats/test_format.py b/pandas/pandas/tests/io/formats/test_format.py
new file mode 100644
index 0000000000000..86682e8160762
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_format.py
@@ -0,0 +1,2305 @@
+"""
+Tests for the file pandas.io.formats.format, *not* tests for general formatting
+of pandas objects.
+"""
+
+from datetime import datetime
+from io import StringIO
+import re
+from shutil import get_terminal_size
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+    get_option,
+    option_context,
+    read_csv,
+    reset_option,
+)
+
+from pandas.io.formats import printing
+import pandas.io.formats.format as fmt
+
+
+def has_info_repr(df):
+    r = repr(df)
+    c1 = r.split("\n")[0].startswith("<class")
+    c2 = r.split("\n")[0].startswith(r"&lt;class")  # _repr_html_
+    return c1 or c2
+
+
+def has_non_verbose_info_repr(df):
+    has_info = has_info_repr(df)
+    r = repr(df)
+
+    # 1. <class>
+    # 2. Index
+    # 3. Columns
+    # 4. dtype
+    # 5. memory usage
+    # 6. trailing newline
+    nv = len(r.split("\n")) == 6
+    return has_info and nv
+
+
+def has_horizontally_truncated_repr(df):
+    try:  # Check header row
+        fst_line = np.array(repr(df).splitlines()[0].split())
+        cand_col = np.where(fst_line == "...")[0][0]
+    except IndexError:
+        return False
+    # Make sure each row has this ... in the same place
+    r = repr(df)
+    for ix, _ in enumerate(r.splitlines()):
+        if not r.split()[cand_col] == "...":
+            return False
+    return True
+
+
+def has_vertically_truncated_repr(df):
+    r = repr(df)
+    only_dot_row = False
+    for row in r.splitlines():
+        if re.match(r"^[\.\ ]+$", row):
+            only_dot_row = True
+    return only_dot_row
+
+
+def has_truncated_repr(df):
+    return has_horizontally_truncated_repr(df) or has_vertically_truncated_repr(df)
+
+
+def has_doubly_truncated_repr(df):
+    return has_horizontally_truncated_repr(df) and has_vertically_truncated_repr(df)
+
+
+def has_expanded_repr(df):
+    r = repr(df)
+    for line in r.split("\n"):
+        if line.endswith("\\"):
+            return True
+    return False
+
+
+class TestDataFrameFormatting:
+    def test_repr_truncation(self):
+        max_len = 20
+        with option_context("display.max_colwidth", max_len):
+            df = DataFrame(
+                {
+                    "A": np.random.default_rng(2).standard_normal(10),
+                    "B": [
+                        "a"
+                        * np.random.default_rng(2).integers(max_len - 1, max_len + 1)
+                        for _ in range(10)
+                    ],
+                }
+            )
+            r = repr(df)
+            r = r[r.find("\n") + 1 :]
+
+            adj = printing.get_adjustment()
+
+            for line, value in zip(r.split("\n"), df["B"]):
+                if adj.len(value) + 1 > max_len:
+                    assert "..." in line
+                else:
+                    assert "..." not in line
+
+        with option_context("display.max_colwidth", 999999):
+            assert "..." not in repr(df)
+
+        with option_context("display.max_colwidth", max_len + 2):
+            assert "..." not in repr(df)
+
+    def test_repr_truncation_preserves_na(self):
+        # https://github.com/pandas-dev/pandas/issues/55630
+        df = DataFrame({"a": [pd.NA for _ in range(10)]})
+        with option_context("display.max_rows", 2, "display.show_dimensions", False):
+            assert repr(df) == "       a\n0   <NA>\n..   ...\n9   <NA>"
+
+    def test_repr_truncation_dataframe_attrs(self):
+        # GH#60455
+        df = DataFrame([[0] * 10])
+        df.attrs["b"] = DataFrame([])
+        with option_context("display.max_columns", 2, "display.show_dimensions", False):
+            assert repr(df) == "   0  ...  9\n0  0  ...  0"
+
+    def test_repr_truncation_series_with_dataframe_attrs(self):
+        # GH#60568
+        ser = Series([0] * 10)
+        ser.attrs["b"] = DataFrame([])
+        with option_context("display.max_rows", 2, "display.show_dimensions", False):
+            assert repr(ser) == "0    0\n    ..\n9    0\ndtype: int64"
+
+    def test_max_colwidth_negative_int_raises(self):
+        # Deprecation enforced from:
+        # https://github.com/pandas-dev/pandas/issues/31532
+        with pytest.raises(
+            ValueError, match="Value must be a nonnegative integer or None"
+        ):
+            with option_context("display.max_colwidth", -1):
+                pass
+
+    def test_repr_chop_threshold(self):
+        df = DataFrame([[0.1, 0.5], [0.5, -0.1]])
+        reset_option("display.chop_threshold")  # default None
+        assert repr(df) == "     0    1\n0  0.1  0.5\n1  0.5 -0.1"
+
+        with option_context("display.chop_threshold", 0.2):
+            assert repr(df) == "     0    1\n0  0.0  0.5\n1  0.5  0.0"
+
+        with option_context("display.chop_threshold", 0.6):
+            assert repr(df) == "     0    1\n0  0.0  0.0\n1  0.0  0.0"
+
+        with option_context("display.chop_threshold", None):
+            assert repr(df) == "     0    1\n0  0.1  0.5\n1  0.5 -0.1"
+
+    def test_repr_chop_threshold_column_below(self):
+        # GH 6839: validation case
+
+        df = DataFrame([[10, 20, 30, 40], [8e-10, -1e-11, 2e-9, -2e-11]]).T
+
+        with option_context("display.chop_threshold", 0):
+            assert repr(df) == (
+                "      0             1\n"
+                "0  10.0  8.000000e-10\n"
+                "1  20.0 -1.000000e-11\n"
+                "2  30.0  2.000000e-09\n"
+                "3  40.0 -2.000000e-11"
+            )
+
+        with option_context("display.chop_threshold", 1e-8):
+            assert repr(df) == (
+                "      0             1\n"
+                "0  10.0  0.000000e+00\n"
+                "1  20.0  0.000000e+00\n"
+                "2  30.0  0.000000e+00\n"
+                "3  40.0  0.000000e+00"
+            )
+
+        with option_context("display.chop_threshold", 5e-11):
+            assert repr(df) == (
+                "      0             1\n"
+                "0  10.0  8.000000e-10\n"
+                "1  20.0  0.000000e+00\n"
+                "2  30.0  2.000000e-09\n"
+                "3  40.0  0.000000e+00"
+            )
+
+    def test_repr_no_backslash(self):
+        with option_context("mode.sim_interactive", True):
+            df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+            assert "\\" not in repr(df)
+
+    def test_expand_frame_repr(self):
+        df_small = DataFrame("hello", index=[0], columns=[0])
+        df_wide = DataFrame("hello", index=[0], columns=range(10))
+        df_tall = DataFrame("hello", index=range(30), columns=range(5))
+
+        with option_context("mode.sim_interactive", True):
+            with option_context(
+                "display.max_columns",
+                10,
+                "display.width",
+                20,
+                "display.max_rows",
+                20,
+                "display.show_dimensions",
+                True,
+            ):
+                with option_context("display.expand_frame_repr", True):
+                    assert not has_truncated_repr(df_small)
+                    assert not has_expanded_repr(df_small)
+                    assert not has_truncated_repr(df_wide)
+                    assert has_expanded_repr(df_wide)
+                    assert has_vertically_truncated_repr(df_tall)
+                    assert has_expanded_repr(df_tall)
+
+                with option_context("display.expand_frame_repr", False):
+                    assert not has_truncated_repr(df_small)
+                    assert not has_expanded_repr(df_small)
+                    assert not has_horizontally_truncated_repr(df_wide)
+                    assert not has_expanded_repr(df_wide)
+                    assert has_vertically_truncated_repr(df_tall)
+                    assert not has_expanded_repr(df_tall)
+
+    def test_repr_non_interactive(self):
+        # in non interactive mode, there can be no dependency on the
+        # result of terminal auto size detection
+        df = DataFrame("hello", index=range(1000), columns=range(5))
+
+        with option_context(
+            "mode.sim_interactive", False, "display.width", 0, "display.max_rows", 5000
+        ):
+            assert not has_truncated_repr(df)
+            assert not has_expanded_repr(df)
+
+    def test_repr_truncates_terminal_size(self, monkeypatch):
+        # see gh-21180
+
+        terminal_size = (118, 96)
+        monkeypatch.setattr(
+            "pandas.io.formats.format.get_terminal_size", lambda: terminal_size
+        )
+
+        index = range(5)
+        columns = MultiIndex.from_tuples(
+            [
+                ("This is a long title with > 37 chars.", "cat"),
+                ("This is a loooooonger title with > 43 chars.", "dog"),
+            ]
+        )
+        df = DataFrame(1, index=index, columns=columns)
+
+        result = repr(df)
+
+        h1, h2 = result.split("\n")[:2]
+        assert "long" in h1
+        assert "loooooonger" in h1
+        assert "cat" in h2
+        assert "dog" in h2
+
+        # regular columns
+        df2 = DataFrame({"A" * 41: [1, 2], "B" * 41: [1, 2]})
+        result = repr(df2)
+
+        assert df2.columns[0] in result.split("\n")[0]
+
+    def test_repr_truncates_terminal_size_full(self, monkeypatch):
+        # GH 22984 ensure entire window is filled
+        terminal_size = (80, 24)
+        df = DataFrame(np.random.default_rng(2).random((1, 7)))
+
+        monkeypatch.setattr(
+            "pandas.io.formats.format.get_terminal_size", lambda: terminal_size
+        )
+        assert "..." not in str(df)
+
+    def test_repr_truncation_column_size(self):
+        # dataframe with last column very wide -> check it is not used to
+        # determine size of truncation (...) column
+        df = DataFrame(
+            {
+                "a": [108480, 30830],
+                "b": [12345, 12345],
+                "c": [12345, 12345],
+                "d": [12345, 12345],
+                "e": ["a" * 50] * 2,
+            }
+        )
+        assert "..." in str(df)
+        assert "    ...    " not in str(df)
+
+    def test_repr_max_columns_max_rows(self):
+        term_width, term_height = get_terminal_size()
+        if term_width < 10 or term_height < 10:
+            pytest.skip(f"terminal size too small, {term_width} x {term_height}")
+
+        def mkframe(n):
+            index = [f"{i:05d}" for i in range(n)]
+            return DataFrame(0, index, index)
+
+        df6 = mkframe(6)
+        df10 = mkframe(10)
+        with option_context("mode.sim_interactive", True):
+            with option_context("display.width", term_width * 2):
+                with option_context("display.max_rows", 5, "display.max_columns", 5):
+                    assert not has_expanded_repr(mkframe(4))
+                    assert not has_expanded_repr(mkframe(5))
+                    assert not has_expanded_repr(df6)
+                    assert has_doubly_truncated_repr(df6)
+
+                with option_context("display.max_rows", 20, "display.max_columns", 10):
+                    # Out off max_columns boundary, but no extending
+                    # since not exceeding width
+                    assert not has_expanded_repr(df6)
+                    assert not has_truncated_repr(df6)
+
+                with option_context("display.max_rows", 9, "display.max_columns", 10):
+                    # out vertical bounds can not result in expanded repr
+                    assert not has_expanded_repr(df10)
+                    assert has_vertically_truncated_repr(df10)
+
+            # width=None in terminal, auto detection
+            with option_context(
+                "display.max_columns",
+                100,
+                "display.max_rows",
+                term_width * 20,
+                "display.width",
+                None,
+            ):
+                df = mkframe((term_width // 7) - 2)
+                assert not has_expanded_repr(df)
+                df = mkframe((term_width // 7) + 2)
+                printing.pprint_thing(df._repr_fits_horizontal_())
+                assert has_expanded_repr(df)
+
+    def test_repr_min_rows(self):
+        df = DataFrame({"a": range(20)})
+
+        # default setting no truncation even if above min_rows
+        assert ".." not in repr(df)
+        assert ".." not in df._repr_html_()
+
+        df = DataFrame({"a": range(61)})
+
+        # default of max_rows 60 triggers truncation if above
+        assert ".." in repr(df)
+        assert ".." in df._repr_html_()
+
+        with option_context("display.max_rows", 10, "display.min_rows", 4):
+            # truncated after first two rows
+            assert ".." in repr(df)
+            assert "2  " not in repr(df)
+            assert "..." in df._repr_html_()
+            assert "<td>2</td>" not in df._repr_html_()
+
+        with option_context("display.max_rows", 12, "display.min_rows", None):
+            # when set to None, follow value of max_rows
+            assert "5    5" in repr(df)
+            assert "<td>5</td>" in df._repr_html_()
+
+        with option_context("display.max_rows", 10, "display.min_rows", 12):
+            # when set value higher as max_rows, use the minimum
+            assert "5    5" not in repr(df)
+            assert "<td>5</td>" not in df._repr_html_()
+
+        with option_context("display.max_rows", None, "display.min_rows", 12):
+            # max_rows of None -> never truncate
+            assert ".." not in repr(df)
+            assert ".." not in df._repr_html_()
+
+    @pytest.mark.parametrize(
+        "data, format_option, expected_values",
+        [
+            (12345.6789, "{:12.3f}", "12345.679"),
+            (None, "{:.3f}", "None"),
+            ("", "{:.2f}", ""),
+            (112345.6789, "{:6.3f}", "112345.679"),
+            ("foo      foo", None, "foo&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;foo"),
+            (" foo", None, "foo"),
+            (
+                "foo foo       foo",
+                None,
+                "foo foo&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; foo",
+            ),  # odd no.of spaces
+            (
+                "foo foo    foo",
+                None,
+                "foo foo&nbsp;&nbsp;&nbsp;&nbsp;foo",
+            ),  # even no.of spaces
+        ],
+    )
+    def test_repr_float_formatting_html_output(
+        self, data, format_option, expected_values
+    ):
+        if format_option is not None:
+            with option_context("display.float_format", format_option.format):
+                df = DataFrame({"A": [data]})
+                html_output = df._repr_html_()
+                assert expected_values in html_output
+        else:
+            df = DataFrame({"A": [data]})
+            html_output = df._repr_html_()
+            assert expected_values in html_output
+
+    def test_str_max_colwidth(self):
+        # GH 7856
+        df = DataFrame(
+            [
+                {
+                    "a": "foo",
+                    "b": "bar",
+                    "c": "uncomfortably long line with lots of stuff",
+                    "d": 1,
+                },
+                {"a": "foo", "b": "bar", "c": "stuff", "d": 1},
+            ]
+        )
+        df.set_index(["a", "b", "c"])
+        assert str(df) == (
+            "     a    b                                           c  d\n"
+            "0  foo  bar  uncomfortably long line with lots of stuff  1\n"
+            "1  foo  bar                                       stuff  1"
+        )
+        with option_context("max_colwidth", 20):
+            assert str(df) == (
+                "     a    b                    c  d\n"
+                "0  foo  bar  uncomfortably lo...  1\n"
+                "1  foo  bar                stuff  1"
+            )
+
+    def test_auto_detect(self):
+        term_width, term_height = get_terminal_size()
+        fac = 1.05  # Arbitrary large factor to exceed term width
+        cols = range(int(term_width * fac))
+        index = range(10)
+        df = DataFrame(index=index, columns=cols)
+        with option_context("mode.sim_interactive", True):
+            with option_context("display.max_rows", None):
+                with option_context("display.max_columns", None):
+                    # Wrap around with None
+                    assert has_expanded_repr(df)
+            with option_context("display.max_rows", 0):
+                with option_context("display.max_columns", 0):
+                    # Truncate with auto detection.
+                    assert has_horizontally_truncated_repr(df)
+
+            index = range(int(term_height * fac))
+            df = DataFrame(index=index, columns=cols)
+            with option_context("display.max_rows", 0):
+                with option_context("display.max_columns", None):
+                    # Wrap around with None
+                    assert has_expanded_repr(df)
+                    # Truncate vertically
+                    assert has_vertically_truncated_repr(df)
+
+            with option_context("display.max_rows", None):
+                with option_context("display.max_columns", 0):
+                    assert has_horizontally_truncated_repr(df)
+
+    def test_to_string_repr_unicode2(self):
+        idx = Index(["abc", "\u03c3a", "aegdvg"])
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        rs = repr(ser).split("\n")
+        line_len = len(rs[0])
+        for line in rs[1:]:
+            try:
+                line = line.decode(get_option("display.encoding"))
+            except AttributeError:
+                pass
+            if not line.startswith("dtype:"):
+                assert len(line) == line_len
+
+    def test_east_asian_unicode_false(self):
+        # not aligned properly because of east asian width
+
+        # mid col
+        df = DataFrame(
+            {"a": ["あ", "いいい", "う", "ええええええ"], "b": [1, 222, 33333, 4]},
+            index=["a", "bb", "c", "ddd"],
+        )
+        expected = (
+            "          a      b\na         あ      1\n"
+            "bb      いいい    222\nc         う  33333\n"
+            "ddd  ええええええ      4"
+        )
+        assert repr(df) == expected
+
+        # last col
+        df = DataFrame(
+            {"a": [1, 222, 33333, 4], "b": ["あ", "いいい", "う", "ええええええ"]},
+            index=["a", "bb", "c", "ddd"],
+        )
+        expected = (
+            "         a       b\na        1       あ\n"
+            "bb     222     いいい\nc    33333       う\n"
+            "ddd      4  ええええええ"
+        )
+        assert repr(df) == expected
+
+        # all col
+        df = DataFrame(
+            {
+                "a": ["あああああ", "い", "う", "えええ"],
+                "b": ["あ", "いいい", "う", "ええええええ"],
+            },
+            index=["a", "bb", "c", "ddd"],
+        )
+        expected = (
+            "         a       b\na    あああああ       あ\n"
+            "bb       い     いいい\nc        う       う\n"
+            "ddd    えええ  ええええええ"
+        )
+        assert repr(df) == expected
+
+        # column name
+        df = DataFrame(
+            {
+                "b": ["あ", "いいい", "う", "ええええええ"],
+                "あああああ": [1, 222, 33333, 4],
+            },
+            index=["a", "bb", "c", "ddd"],
+        )
+        expected = (
+            "          b  あああああ\na         あ      1\n"
+            "bb      いいい    222\nc         う  33333\n"
+            "ddd  ええええええ      4"
+        )
+        assert repr(df) == expected
+
+        # index
+        df = DataFrame(
+            {
+                "a": ["あああああ", "い", "う", "えええ"],
+                "b": ["あ", "いいい", "う", "ええええええ"],
+            },
+            index=["あああ", "いいいいいい", "うう", "え"],
+        )
+        expected = (
+            "            a       b\nあああ     あああああ       あ\n"
+            "いいいいいい      い     いいい\nうう          う       う\n"
+            "え         えええ  ええええええ"
+        )
+        assert repr(df) == expected
+
+        # index name
+        df = DataFrame(
+            {
+                "a": ["あああああ", "い", "う", "えええ"],
+                "b": ["あ", "いいい", "う", "ええええええ"],
+            },
+            index=Index(["あ", "い", "うう", "え"], name="おおおお"),
+        )
+        expected = (
+            "          a       b\n"
+            "おおおお               \n"
+            "あ     あああああ       あ\n"
+            "い         い     いいい\n"
+            "うう        う       う\n"
+            "え       えええ  ええええええ"
+        )
+        assert repr(df) == expected
+
+        # all
+        df = DataFrame(
+            {
+                "あああ": ["あああ", "い", "う", "えええええ"],
+                "いいいいい": ["あ", "いいい", "う", "ええ"],
+            },
+            index=Index(["あ", "いいい", "うう", "え"], name="お"),
+        )
+        expected = (
+            "       あああ いいいいい\n"
+            "お               \n"
+            "あ      あああ     あ\n"
+            "いいい      い   いいい\n"
+            "うう       う     う\n"
+            "え    えええええ    ええ"
+        )
+        assert repr(df) == expected
+
+        # MultiIndex
+        idx = MultiIndex.from_tuples(
+            [("あ", "いい"), ("う", "え"), ("おおお", "かかかか"), ("き", "くく")]
+        )
+        df = DataFrame(
+            {
+                "a": ["あああああ", "い", "う", "えええ"],
+                "b": ["あ", "いいい", "う", "ええええええ"],
+            },
+            index=idx,
+        )
+        expected = (
+            "              a       b\n"
+            "あ   いい    あああああ       あ\n"
+            "う   え         い     いいい\n"
+            "おおお かかかか      う       う\n"
+            "き   くく      えええ  ええええええ"
+        )
+        assert repr(df) == expected
+
+        # truncate
+        with option_context("display.max_rows", 3, "display.max_columns", 3):
+            df = DataFrame(
+                {
+                    "a": ["あああああ", "い", "う", "えええ"],
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                    "c": ["お", "か", "ききき", "くくくくくく"],
+                    "ああああ": ["さ", "し", "す", "せ"],
+                },
+                columns=["a", "b", "c", "ああああ"],
+            )
+
+            expected = (
+                "        a  ... ああああ\n0   あああああ  ...    さ\n"
+                "..    ...  ...  ...\n3     えええ  ...    せ\n"
+                "\n[4 rows x 4 columns]"
+            )
+            assert repr(df) == expected
+
+            df.index = ["あああ", "いいいい", "う", "aaa"]
+            expected = (
+                "         a  ... ああああ\nあああ  あああああ  ...    さ\n"
+                "..     ...  ...  ...\naaa    えええ  ...    せ\n"
+                "\n[4 rows x 4 columns]"
+            )
+            assert repr(df) == expected
+
+    def test_east_asian_unicode_true(self):
+        # Enable Unicode option -----------------------------------------
+        with option_context("display.unicode.east_asian_width", True):
+            # mid col
+            df = DataFrame(
+                {"a": ["あ", "いいい", "う", "ええええええ"], "b": [1, 222, 33333, 4]},
+                index=["a", "bb", "c", "ddd"],
+            )
+            expected = (
+                "                a      b\na              あ      1\n"
+                "bb         いいい    222\nc              う  33333\n"
+                "ddd  ええええええ      4"
+            )
+            assert repr(df) == expected
+
+            # last col
+            df = DataFrame(
+                {"a": [1, 222, 33333, 4], "b": ["あ", "いいい", "う", "ええええええ"]},
+                index=["a", "bb", "c", "ddd"],
+            )
+            expected = (
+                "         a             b\na        1            あ\n"
+                "bb     222        いいい\nc    33333            う\n"
+                "ddd      4  ええええええ"
+            )
+            assert repr(df) == expected
+
+            # all col
+            df = DataFrame(
+                {
+                    "a": ["あああああ", "い", "う", "えええ"],
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                },
+                index=["a", "bb", "c", "ddd"],
+            )
+            expected = (
+                "              a             b\n"
+                "a    あああああ            あ\n"
+                "bb           い        いいい\n"
+                "c            う            う\n"
+                "ddd      えええ  ええええええ"
+            )
+            assert repr(df) == expected
+
+            # column name
+            df = DataFrame(
+                {
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                    "あああああ": [1, 222, 33333, 4],
+                },
+                index=["a", "bb", "c", "ddd"],
+            )
+            expected = (
+                "                b  あああああ\n"
+                "a              あ           1\n"
+                "bb         いいい         222\n"
+                "c              う       33333\n"
+                "ddd  ええええええ           4"
+            )
+            assert repr(df) == expected
+
+            # index
+            df = DataFrame(
+                {
+                    "a": ["あああああ", "い", "う", "えええ"],
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                },
+                index=["あああ", "いいいいいい", "うう", "え"],
+            )
+            expected = (
+                "                       a             b\n"
+                "あああ        あああああ            あ\n"
+                "いいいいいい          い        いいい\n"
+                "うう                  う            う\n"
+                "え                えええ  ええええええ"
+            )
+            assert repr(df) == expected
+
+            # index name
+            df = DataFrame(
+                {
+                    "a": ["あああああ", "い", "う", "えええ"],
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                },
+                index=Index(["あ", "い", "うう", "え"], name="おおおお"),
+            )
+            expected = (
+                "                   a             b\n"
+                "おおおお                          \n"
+                "あ        あああああ            あ\n"
+                "い                い        いいい\n"
+                "うう              う            う\n"
+                "え            えええ  ええええええ"
+            )
+            assert repr(df) == expected
+
+            # all
+            df = DataFrame(
+                {
+                    "あああ": ["あああ", "い", "う", "えええええ"],
+                    "いいいいい": ["あ", "いいい", "う", "ええ"],
+                },
+                index=Index(["あ", "いいい", "うう", "え"], name="お"),
+            )
+            expected = (
+                "            あああ いいいいい\n"
+                "お                           \n"
+                "あ          あああ         あ\n"
+                "いいい          い     いいい\n"
+                "うう            う         う\n"
+                "え      えええええ       ええ"
+            )
+            assert repr(df) == expected
+
+            # MultiIndex
+            idx = MultiIndex.from_tuples(
+                [("あ", "いい"), ("う", "え"), ("おおお", "かかかか"), ("き", "くく")]
+            )
+            df = DataFrame(
+                {
+                    "a": ["あああああ", "い", "う", "えええ"],
+                    "b": ["あ", "いいい", "う", "ええええええ"],
+                },
+                index=idx,
+            )
+            expected = (
+                "                          a             b\n"
+                "あ     いい      あああああ            あ\n"
+                "う     え                い        いいい\n"
+                "おおお かかかか          う            う\n"
+                "き     くく          えええ  ええええええ"
+            )
+            assert repr(df) == expected
+
+            # truncate
+            with option_context("display.max_rows", 3, "display.max_columns", 3):
+                df = DataFrame(
+                    {
+                        "a": ["あああああ", "い", "う", "えええ"],
+                        "b": ["あ", "いいい", "う", "ええええええ"],
+                        "c": ["お", "か", "ききき", "くくくくくく"],
+                        "ああああ": ["さ", "し", "す", "せ"],
+                    },
+                    columns=["a", "b", "c", "ああああ"],
+                )
+
+                expected = (
+                    "             a  ... ああああ\n"
+                    "0   あああああ  ...       さ\n"
+                    "..         ...  ...      ...\n"
+                    "3       えええ  ...       せ\n"
+                    "\n[4 rows x 4 columns]"
+                )
+                assert repr(df) == expected
+
+                df.index = ["あああ", "いいいい", "う", "aaa"]
+                expected = (
+                    "                 a  ... ああああ\n"
+                    "あああ  あああああ  ...       さ\n"
+                    "...            ...  ...      ...\n"
+                    "aaa         えええ  ...       せ\n"
+                    "\n[4 rows x 4 columns]"
+                )
+                assert repr(df) == expected
+
+            # ambiguous unicode
+            df = DataFrame(
+                {
+                    "b": ["あ", "いいい", "¡¡", "ええええええ"],
+                    "あああああ": [1, 222, 33333, 4],
+                },
+                index=["a", "bb", "c", "¡¡¡"],
+            )
+            expected = (
+                "                b  あああああ\n"
+                "a              あ           1\n"
+                "bb         いいい         222\n"
+                "c              ¡¡       33333\n"
+                "¡¡¡  ええええええ           4"
+            )
+            assert repr(df) == expected
+
+    def test_to_string_buffer_all_unicode(self):
+        buf = StringIO()
+
+        empty = DataFrame({"c/\u03c3": Series(dtype=object)})
+        nonempty = DataFrame({"c/\u03c3": Series([1, 2, 3])})
+
+        print(empty, file=buf)
+        print(nonempty, file=buf)
+
+        # this should work
+        buf.getvalue()
+
+    @pytest.mark.parametrize(
+        "index_scalar",
+        [
+            "a" * 10,
+            1,
+            Timestamp(2020, 1, 1),
+            pd.Period("2020-01-01"),
+        ],
+    )
+    @pytest.mark.parametrize("h", [10, 20])
+    @pytest.mark.parametrize("w", [10, 20])
+    def test_to_string_truncate_indices(self, index_scalar, h, w):
+        with option_context("display.expand_frame_repr", False):
+            df = DataFrame(
+                index=[index_scalar] * h, columns=[str(i) * 10 for i in range(w)]
+            )
+            with option_context("display.max_rows", 15):
+                if h == 20:
+                    assert has_vertically_truncated_repr(df)
+                else:
+                    assert not has_vertically_truncated_repr(df)
+            with option_context("display.max_columns", 15):
+                if w == 20:
+                    assert has_horizontally_truncated_repr(df)
+                else:
+                    assert not has_horizontally_truncated_repr(df)
+            with option_context("display.max_rows", 15, "display.max_columns", 15):
+                if h == 20 and w == 20:
+                    assert has_doubly_truncated_repr(df)
+                else:
+                    assert not has_doubly_truncated_repr(df)
+
+    def test_to_string_truncate_multilevel(self):
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        df = DataFrame(index=arrays, columns=arrays)
+        with option_context("display.max_rows", 7, "display.max_columns", 7):
+            assert has_doubly_truncated_repr(df)
+
+    @pytest.mark.parametrize("dtype", ["object", "datetime64[us]"])
+    def test_truncate_with_different_dtypes(self, dtype):
+        # 11594, 12045
+        # when truncated the dtypes of the splits can differ
+
+        # 11594
+        ser = Series(
+            [datetime(2012, 1, 1)] * 10
+            + [datetime(1012, 1, 2)]
+            + [datetime(2012, 1, 3)] * 10,
+            dtype=dtype,
+        )
+
+        with option_context("display.max_rows", 8):
+            result = str(ser)
+        assert dtype in result
+
+    def test_truncate_with_different_dtypes2(self):
+        # 12045
+        df = DataFrame({"text": ["some words"] + [None] * 9}, dtype=object)
+
+        with option_context("display.max_rows", 8, "display.max_columns", 3):
+            result = str(df)
+            assert "None" in result
+            assert "NaN" not in result
+
+    def test_truncate_with_different_dtypes_multiindex(self):
+        # GH#13000
+        df = DataFrame({"Vals": range(100)})
+        frame = pd.concat([df], keys=["Sweep"], names=["Sweep", "Index"])
+        result = repr(frame)
+
+        result2 = repr(frame.iloc[:5])
+        assert result.startswith(result2)
+
+    def test_datetimelike_frame(self):
+        # GH 12211
+        df = DataFrame({"date": [Timestamp("20130101").tz_localize("UTC")] + [NaT] * 5})
+
+        with option_context("display.max_rows", 5):
+            result = str(df)
+            assert "2013-01-01 00:00:00+00:00" in result
+            assert "NaT" in result
+            assert "..." in result
+            assert "[6 rows x 1 columns]" in result
+
+        dts = [Timestamp("2011-01-01", tz="US/Eastern")] * 5 + [NaT] * 5
+        df = DataFrame({"dt": dts, "x": [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]})
+        with option_context("display.max_rows", 5):
+            expected = (
+                "                          dt   x\n"
+                "0  2011-01-01 00:00:00-05:00   1\n"
+                "1  2011-01-01 00:00:00-05:00   2\n"
+                "..                       ...  ..\n"
+                "8                        NaT   9\n"
+                "9                        NaT  10\n\n"
+                "[10 rows x 2 columns]"
+            )
+            assert repr(df) == expected
+
+        dts = [NaT] * 5 + [Timestamp("2011-01-01", tz="US/Eastern")] * 5
+        df = DataFrame({"dt": dts, "x": [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]})
+        with option_context("display.max_rows", 5):
+            expected = (
+                "                          dt   x\n"
+                "0                        NaT   1\n"
+                "1                        NaT   2\n"
+                "..                       ...  ..\n"
+                "8  2011-01-01 00:00:00-05:00   9\n"
+                "9  2011-01-01 00:00:00-05:00  10\n\n"
+                "[10 rows x 2 columns]"
+            )
+            assert repr(df) == expected
+
+        dts = [Timestamp("2011-01-01", tz="Asia/Tokyo")] * 5 + [
+            Timestamp("2011-01-01", tz="US/Eastern")
+        ] * 5
+        df = DataFrame({"dt": dts, "x": [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]})
+        with option_context("display.max_rows", 5):
+            expected = (
+                "                           dt   x\n"
+                "0   2011-01-01 00:00:00+09:00   1\n"
+                "1   2011-01-01 00:00:00+09:00   2\n"
+                "..                        ...  ..\n"
+                "8   2011-01-01 00:00:00-05:00   9\n"
+                "9   2011-01-01 00:00:00-05:00  10\n\n"
+                "[10 rows x 2 columns]"
+            )
+            assert repr(df) == expected
+
+    @pytest.mark.parametrize(
+        "start_date",
+        [
+            "2017-01-01 23:59:59.999999999",
+            "2017-01-01 23:59:59.99999999",
+            "2017-01-01 23:59:59.9999999",
+            "2017-01-01 23:59:59.999999",
+            "2017-01-01 23:59:59.99999",
+            "2017-01-01 23:59:59.9999",
+        ],
+    )
+    def test_datetimeindex_highprecision(self, start_date):
+        # GH19030
+        # Check that high-precision time values for the end of day are
+        # included in repr for DatetimeIndex
+        df = DataFrame({"A": date_range(start=start_date, freq="D", periods=5)})
+        result = str(df)
+        assert start_date in result
+
+        dti = date_range(start=start_date, freq="D", periods=5)
+        df = DataFrame({"A": range(5)}, index=dti)
+        result = str(df.index)
+        assert start_date in result
+
+    def test_string_repr_encoding(self, datapath):
+        filepath = datapath("io", "parser", "data", "unicode_series.csv")
+        df = read_csv(filepath, header=None, encoding="latin1")
+        repr(df)
+        repr(df[1])
+
+    def test_repr_corner(self):
+        # representing infs poses no problems
+        df = DataFrame({"foo": [-np.inf, np.inf]})
+        repr(df)
+
+    def test_frame_info_encoding(self):
+        index = ["'Til There Was You (1997)", "ldum klaka (Cold Fever) (1994)"]
+        with option_context("display.max_rows", 1):
+            df = DataFrame(columns=["a", "b", "c"], index=index)
+            repr(df)
+            repr(df.T)
+
+    def test_wide_repr(self):
+        with option_context(
+            "mode.sim_interactive",
+            True,
+            "display.show_dimensions",
+            True,
+            "display.max_columns",
+            20,
+        ):
+            max_cols = get_option("display.max_columns")
+            df = DataFrame([["a" * 25] * (max_cols - 1)] * 10)
+            with option_context("display.expand_frame_repr", False):
+                rep_str = repr(df)
+
+            assert f"10 rows x {max_cols - 1} columns" in rep_str
+            with option_context("display.expand_frame_repr", True):
+                wide_repr = repr(df)
+            assert rep_str != wide_repr
+
+            with option_context("display.width", 120):
+                wider_repr = repr(df)
+                assert len(wider_repr) < len(wide_repr)
+
+    def test_wide_repr_wide_columns(self):
+        with option_context("mode.sim_interactive", True, "display.max_columns", 20):
+            df = DataFrame(
+                np.random.default_rng(2).standard_normal((5, 3)),
+                columns=["a" * 90, "b" * 90, "c" * 90],
+            )
+            rep_str = repr(df)
+
+            assert len(rep_str.splitlines()) == 20
+
+    def test_wide_repr_named(self):
+        with option_context("mode.sim_interactive", True, "display.max_columns", 20):
+            max_cols = get_option("display.max_columns")
+            df = DataFrame([["a" * 25] * (max_cols - 1)] * 10)
+            df.index.name = "DataFrame Index"
+            with option_context("display.expand_frame_repr", False):
+                rep_str = repr(df)
+            with option_context("display.expand_frame_repr", True):
+                wide_repr = repr(df)
+            assert rep_str != wide_repr
+
+            with option_context("display.width", 150):
+                wider_repr = repr(df)
+                assert len(wider_repr) < len(wide_repr)
+
+            for line in wide_repr.splitlines()[1::13]:
+                assert "DataFrame Index" in line
+
+    def test_wide_repr_multiindex(self):
+        with option_context("mode.sim_interactive", True, "display.max_columns", 20):
+            midx = MultiIndex.from_arrays([["a" * 5] * 10] * 2)
+            max_cols = get_option("display.max_columns")
+            df = DataFrame([["a" * 25] * (max_cols - 1)] * 10, index=midx)
+            df.index.names = ["Level 0", "Level 1"]
+            with option_context("display.expand_frame_repr", False):
+                rep_str = repr(df)
+            with option_context("display.expand_frame_repr", True):
+                wide_repr = repr(df)
+            assert rep_str != wide_repr
+
+            with option_context("display.width", 150):
+                wider_repr = repr(df)
+                assert len(wider_repr) < len(wide_repr)
+
+            for line in wide_repr.splitlines()[1::13]:
+                assert "Level 0 Level 1" in line
+
+    def test_wide_repr_multiindex_cols(self):
+        with option_context("mode.sim_interactive", True, "display.max_columns", 20):
+            max_cols = get_option("display.max_columns")
+            midx = MultiIndex.from_arrays([["a" * 5] * 10] * 2)
+            mcols = MultiIndex.from_arrays([["b" * 3] * (max_cols - 1)] * 2)
+            df = DataFrame(
+                [["c" * 25] * (max_cols - 1)] * 10, index=midx, columns=mcols
+            )
+            df.index.names = ["Level 0", "Level 1"]
+            with option_context("display.expand_frame_repr", False):
+                rep_str = repr(df)
+            with option_context("display.expand_frame_repr", True):
+                wide_repr = repr(df)
+            assert rep_str != wide_repr
+
+        with option_context("display.width", 150, "display.max_columns", 20):
+            wider_repr = repr(df)
+            assert len(wider_repr) < len(wide_repr)
+
+    def test_wide_repr_unicode(self):
+        with option_context("mode.sim_interactive", True, "display.max_columns", 20):
+            max_cols = 20
+            df = DataFrame([["a" * 25] * 10] * (max_cols - 1))
+            with option_context("display.expand_frame_repr", False):
+                rep_str = repr(df)
+            with option_context("display.expand_frame_repr", True):
+                wide_repr = repr(df)
+            assert rep_str != wide_repr
+
+            with option_context("display.width", 150):
+                wider_repr = repr(df)
+                assert len(wider_repr) < len(wide_repr)
+
+    def test_wide_repr_wide_long_columns(self):
+        with option_context("mode.sim_interactive", True):
+            df = DataFrame({"a": ["a" * 30, "b" * 30], "b": ["c" * 70, "d" * 80]})
+
+            result = repr(df)
+            assert "ccccc" in result
+            assert "ddddd" in result
+
+    def test_long_series(self):
+        n = 1000
+        s = Series(
+            np.random.default_rng(2).integers(-50, 50, n),
+            index=[f"s{x:04d}" for x in range(n)],
+            dtype="int64",
+        )
+
+        str_rep = str(s)
+        nmatches = len(re.findall("dtype", str_rep))
+        assert nmatches == 1
+
+    def test_to_string_ascii_error(self):
+        data = [
+            (
+                "0  ",
+                "                        .gitignore ",
+                "     5 ",
+                " \xe2\x80\xa2\xe2\x80\xa2\xe2\x80\xa2\xe2\x80\xa2\xe2\x80\xa2",
+            )
+        ]
+        df = DataFrame(data)
+
+        # it works!
+        repr(df)
+
+    def test_show_dimensions(self):
+        df = DataFrame(123, index=range(10, 15), columns=range(30))
+
+        with option_context(
+            "display.max_rows",
+            10,
+            "display.max_columns",
+            40,
+            "display.width",
+            500,
+            "display.expand_frame_repr",
+            "info",
+            "display.show_dimensions",
+            True,
+        ):
+            assert "5 rows" in str(df)
+            assert "5 rows" in df._repr_html_()
+        with option_context(
+            "display.max_rows",
+            10,
+            "display.max_columns",
+            40,
+            "display.width",
+            500,
+            "display.expand_frame_repr",
+            "info",
+            "display.show_dimensions",
+            False,
+        ):
+            assert "5 rows" not in str(df)
+            assert "5 rows" not in df._repr_html_()
+        with option_context(
+            "display.max_rows",
+            2,
+            "display.max_columns",
+            2,
+            "display.width",
+            500,
+            "display.expand_frame_repr",
+            "info",
+            "display.show_dimensions",
+            "truncate",
+        ):
+            assert "5 rows" in str(df)
+            assert "5 rows" in df._repr_html_()
+        with option_context(
+            "display.max_rows",
+            10,
+            "display.max_columns",
+            40,
+            "display.width",
+            500,
+            "display.expand_frame_repr",
+            "info",
+            "display.show_dimensions",
+            "truncate",
+        ):
+            assert "5 rows" not in str(df)
+            assert "5 rows" not in df._repr_html_()
+
+    def test_info_repr(self):
+        # GH#21746 For tests inside a terminal (i.e. not CI) we need to detect
+        # the terminal size to ensure that we try to print something "too big"
+        term_width, term_height = get_terminal_size()
+
+        max_rows = 60
+        max_cols = 20 + (max(term_width, 80) - 80) // 4
+        # Long
+        h, w = max_rows + 1, max_cols - 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        assert has_vertically_truncated_repr(df)
+        with option_context("display.large_repr", "info"):
+            assert has_info_repr(df)
+
+        # Wide
+        h, w = max_rows - 1, max_cols + 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        assert has_horizontally_truncated_repr(df)
+        with option_context(
+            "display.large_repr", "info", "display.max_columns", max_cols
+        ):
+            assert has_info_repr(df)
+
+    def test_info_repr_max_cols(self):
+        # GH #6939
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+        with option_context(
+            "display.large_repr",
+            "info",
+            "display.max_columns",
+            1,
+            "display.max_info_columns",
+            4,
+        ):
+            assert has_non_verbose_info_repr(df)
+
+        with option_context(
+            "display.large_repr",
+            "info",
+            "display.max_columns",
+            1,
+            "display.max_info_columns",
+            5,
+        ):
+            assert not has_non_verbose_info_repr(df)
+
+        # FIXME: don't leave commented-out
+        # test verbose overrides
+        # set_option('display.max_info_columns', 4)  # exceeded
+
+    def test_pprint_pathological_object(self):
+        """
+        If the test fails, it at least won't hang.
+        """
+
+        class A:
+            def __getitem__(self, key):
+                return 3  # obviously simplified
+
+        df = DataFrame([A()])
+        repr(df)  # just don't die
+
+    def test_float_trim_zeros(self):
+        vals = [
+            2.08430917305e10,
+            3.52205017305e10,
+            2.30674817305e10,
+            2.03954217305e10,
+            5.59897817305e10,
+        ]
+        skip = True
+        for line in repr(DataFrame({"A": vals})).split("\n")[:-2]:
+            if line.startswith("dtype:"):
+                continue
+            if _three_digit_exp():
+                assert ("+010" in line) or skip
+            else:
+                assert ("+10" in line) or skip
+            skip = False
+
+    @pytest.mark.parametrize(
+        "data, expected",
+        [
+            (["3.50"], "0    3.50\ndtype: object"),
+            ([1.20, "1.00"], "0     1.2\n1    1.00\ndtype: object"),
+            ([np.nan], "0   NaN\ndtype: float64"),
+            ([None], "0    None\ndtype: object"),
+            (["3.50", np.nan], "0    3.50\n1     NaN\ndtype: object"),
+            ([3.50, np.nan], "0    3.5\n1    NaN\ndtype: float64"),
+            ([3.50, np.nan, "3.50"], "0     3.5\n1     NaN\n2    3.50\ndtype: object"),
+            ([3.50, None, "3.50"], "0     3.5\n1    None\n2    3.50\ndtype: object"),
+        ],
+    )
+    def test_repr_str_float_truncation(self, data, expected, using_infer_string):
+        # GH#38708
+        series = Series(data, dtype=object if "3.50" in data else None)
+        result = repr(series)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "float_format,expected",
+        [
+            ("{:,.0f}".format, "0   1,000\n1    test\ndtype: object"),
+            ("{:.4f}".format, "0   1000.0000\n1        test\ndtype: object"),
+        ],
+    )
+    def test_repr_float_format_in_object_col(self, float_format, expected):
+        # GH#40024
+        df = Series([1000.0, "test"])
+        with option_context("display.float_format", float_format):
+            result = repr(df)
+
+        assert result == expected
+
+    def test_period(self):
+        # GH 12615
+        df = DataFrame(
+            {
+                "A": pd.period_range("2013-01", periods=4, freq="M"),
+                "B": [
+                    pd.Period("2011-01", freq="M"),
+                    pd.Period("2011-02-01", freq="D"),
+                    pd.Period("2011-03-01 09:00", freq="h"),
+                    pd.Period("2011-04", freq="M"),
+                ],
+                "C": list("abcd"),
+            }
+        )
+        exp = (
+            "         A                 B  C\n"
+            "0  2013-01           2011-01  a\n"
+            "1  2013-02        2011-02-01  b\n"
+            "2  2013-03  2011-03-01 09:00  c\n"
+            "3  2013-04           2011-04  d"
+        )
+        assert str(df) == exp
+
+    @pytest.mark.parametrize(
+        "length, max_rows, min_rows, expected",
+        [
+            (10, 10, 10, 10),
+            (10, 10, None, 10),
+            (10, 8, None, 8),
+            (20, 30, 10, 30),  # max_rows > len(frame), hence max_rows
+            (50, 30, 10, 10),  # max_rows < len(frame), hence min_rows
+            (100, 60, 10, 10),  # same
+            (60, 60, 10, 60),  # edge case
+            (61, 60, 10, 10),  # edge case
+        ],
+    )
+    def test_max_rows_fitted(self, length, min_rows, max_rows, expected):
+        """Check that display logic is correct.
+
+        GH #37359
+
+        See description here:
+        https://pandas.pydata.org/docs/dev/user_guide/options.html#frequently-used-options
+        """
+        formatter = fmt.DataFrameFormatter(
+            DataFrame(np.random.default_rng(2).random((length, 3))),
+            max_rows=max_rows,
+            min_rows=min_rows,
+        )
+        result = formatter.max_rows_fitted
+        assert result == expected
+
+
+def gen_series_formatting():
+    s1 = Series(["a"] * 100)
+    s2 = Series(["ab"] * 100)
+    s3 = Series(["a", "ab", "abc", "abcd", "abcde", "abcdef"])
+    s4 = s3[::-1]
+    test_sers = {"onel": s1, "twol": s2, "asc": s3, "desc": s4}
+    return test_sers
+
+
+class TestSeriesFormatting:
+    def test_freq_name_separation(self):
+        s = Series(
+            np.random.default_rng(2).standard_normal(10),
+            index=date_range("1/1/2000", periods=10),
+            name=0,
+        )
+
+        result = repr(s)
+        assert "Freq: D, Name: 0" in result
+
+    def test_unicode_name_in_footer(self):
+        s = Series([1, 2], name="\u05e2\u05d1\u05e8\u05d9\u05ea")
+        sf = fmt.SeriesFormatter(s, name="\u05e2\u05d1\u05e8\u05d9\u05ea")
+        sf._get_footer()  # should not raise exception
+
+    @pytest.mark.xfail(using_string_dtype(), reason="Fixup when arrow is default")
+    def test_east_asian_unicode_series(self):
+        # not aligned properly because of east asian width
+
+        # unicode index
+        s = Series(["a", "bb", "CCC", "D"], index=["あ", "いい", "ううう", "ええええ"])
+        expected = "".join(
+            [
+                "あ         a\n",
+                "いい       bb\n",
+                "ううう     CCC\n",
+                "ええええ      D\ndtype: object",
+            ]
+        )
+        assert repr(s) == expected
+
+        # unicode values
+        s = Series(["あ", "いい", "ううう", "ええええ"], index=["a", "bb", "c", "ddd"])
+        expected = "".join(
+            [
+                "a         あ\n",
+                "bb       いい\n",
+                "c       ううう\n",
+                "ddd    ええええ\n",
+                "dtype: object",
+            ]
+        )
+
+        assert repr(s) == expected
+
+        # both
+        s = Series(
+            ["あ", "いい", "ううう", "ええええ"],
+            index=["ああ", "いいいい", "う", "えええ"],
+        )
+        expected = "".join(
+            [
+                "ああ         あ\n",
+                "いいいい      いい\n",
+                "う        ううう\n",
+                "えええ     ええええ\n",
+                "dtype: object",
+            ]
+        )
+
+        assert repr(s) == expected
+
+        # unicode footer
+        s = Series(
+            ["あ", "いい", "ううう", "ええええ"],
+            index=["ああ", "いいいい", "う", "えええ"],
+            name="おおおおおおお",
+        )
+        expected = (
+            "ああ         あ\nいいいい      いい\nう        ううう\n"
+            "えええ     ええええ\nName: おおおおおおお, dtype: object"
+        )
+        assert repr(s) == expected
+
+        # MultiIndex
+        idx = MultiIndex.from_tuples(
+            [("あ", "いい"), ("う", "え"), ("おおお", "かかかか"), ("き", "くく")]
+        )
+        s = Series([1, 22, 3333, 44444], index=idx)
+        expected = (
+            "あ    いい          1\n"
+            "う    え          22\n"
+            "おおお  かかかか     3333\n"
+            "き    くく      44444\ndtype: int64"
+        )
+        assert repr(s) == expected
+
+        # object dtype, shorter than unicode repr
+        s = Series([1, 22, 3333, 44444], index=[1, "AB", np.nan, "あああ"])
+        expected = (
+            "1          1\nAB        22\nNaN     3333\nあああ    44444\ndtype: int64"
+        )
+        assert repr(s) == expected
+
+        # object dtype, longer than unicode repr
+        s = Series(
+            [1, 22, 3333, 44444], index=[1, "AB", Timestamp("2011-01-01"), "あああ"]
+        )
+        expected = (
+            "1                          1\n"
+            "AB                        22\n"
+            "2011-01-01 00:00:00     3333\n"
+            "あああ                    44444\ndtype: int64"
+        )
+        assert repr(s) == expected
+
+        # truncate
+        with option_context("display.max_rows", 3):
+            s = Series(["あ", "いい", "ううう", "ええええ"], name="おおおおおおお")
+
+            expected = (
+                "0       あ\n     ... \n"
+                "3    ええええ\n"
+                "Name: おおおおおおお, Length: 4, dtype: object"
+            )
+            assert repr(s) == expected
+
+            s.index = ["ああ", "いいいい", "う", "えええ"]
+            expected = (
+                "ああ        あ\n       ... \n"
+                "えええ    ええええ\n"
+                "Name: おおおおおおお, Length: 4, dtype: object"
+            )
+            assert repr(s) == expected
+
+        # Enable Unicode option -----------------------------------------
+        with option_context("display.unicode.east_asian_width", True):
+            # unicode index
+            s = Series(
+                ["a", "bb", "CCC", "D"],
+                index=["あ", "いい", "ううう", "ええええ"],
+            )
+            expected = (
+                "あ            a\nいい         bb\nううう      CCC\n"
+                "ええええ      D\ndtype: object"
+            )
+            assert repr(s) == expected
+
+            # unicode values
+            s = Series(
+                ["あ", "いい", "ううう", "ええええ"],
+                index=["a", "bb", "c", "ddd"],
+            )
+            expected = (
+                "a            あ\nbb         いい\nc        ううう\n"
+                "ddd    ええええ\ndtype: object"
+            )
+            assert repr(s) == expected
+            # both
+            s = Series(
+                ["あ", "いい", "ううう", "ええええ"],
+                index=["ああ", "いいいい", "う", "えええ"],
+            )
+            expected = (
+                "ああ              あ\n"
+                "いいいい        いい\n"
+                "う            ううう\n"
+                "えええ      ええええ\ndtype: object"
+            )
+            assert repr(s) == expected
+
+            # unicode footer
+            s = Series(
+                ["あ", "いい", "ううう", "ええええ"],
+                index=["ああ", "いいいい", "う", "えええ"],
+                name="おおおおおおお",
+            )
+            expected = (
+                "ああ              あ\n"
+                "いいいい        いい\n"
+                "う            ううう\n"
+                "えええ      ええええ\n"
+                "Name: おおおおおおお, dtype: object"
+            )
+            assert repr(s) == expected
+
+            # MultiIndex
+            idx = MultiIndex.from_tuples(
+                [("あ", "いい"), ("う", "え"), ("おおお", "かかかか"), ("き", "くく")]
+            )
+            s = Series([1, 22, 3333, 44444], index=idx)
+            expected = (
+                "あ      いい            1\n"
+                "う      え             22\n"
+                "おおお  かかかか     3333\n"
+                "き      くく        44444\n"
+                "dtype: int64"
+            )
+            assert repr(s) == expected
+
+            # object dtype, shorter than unicode repr
+            s = Series([1, 22, 3333, 44444], index=[1, "AB", np.nan, "あああ"])
+            expected = (
+                "1             1\nAB           22\nNaN        3333\n"
+                "あああ    44444\ndtype: int64"
+            )
+            assert repr(s) == expected
+
+            # object dtype, longer than unicode repr
+            s = Series(
+                [1, 22, 3333, 44444],
+                index=[1, "AB", Timestamp("2011-01-01"), "あああ"],
+            )
+            expected = (
+                "1                          1\n"
+                "AB                        22\n"
+                "2011-01-01 00:00:00     3333\n"
+                "あああ                 44444\ndtype: int64"
+            )
+            assert repr(s) == expected
+
+            # truncate
+            with option_context("display.max_rows", 3):
+                s = Series(["あ", "いい", "ううう", "ええええ"], name="おおおおおおお")
+                expected = (
+                    "0          あ\n       ...   \n"
+                    "3    ええええ\n"
+                    "Name: おおおおおおお, Length: 4, dtype: object"
+                )
+                assert repr(s) == expected
+
+                s.index = ["ああ", "いいいい", "う", "えええ"]
+                expected = (
+                    "ああ            あ\n"
+                    "            ...   \n"
+                    "えええ    ええええ\n"
+                    "Name: おおおおおおお, Length: 4, dtype: object"
+                )
+                assert repr(s) == expected
+
+            # ambiguous unicode
+            s = Series(
+                ["¡¡", "い¡¡", "ううう", "ええええ"],
+                index=["ああ", "¡¡¡¡いい", "¡¡", "えええ"],
+            )
+            expected = (
+                "ああ              ¡¡\n"
+                "¡¡¡¡いい        い¡¡\n"
+                "¡¡            ううう\n"
+                "えええ      ええええ\ndtype: object"
+            )
+            assert repr(s) == expected
+
+    def test_float_trim_zeros(self):
+        vals = [
+            2.08430917305e10,
+            3.52205017305e10,
+            2.30674817305e10,
+            2.03954217305e10,
+            5.59897817305e10,
+        ]
+        for line in repr(Series(vals)).split("\n"):
+            if line.startswith("dtype:"):
+                continue
+            if _three_digit_exp():
+                assert "+010" in line
+            else:
+                assert "+10" in line
+
+    @pytest.mark.parametrize(
+        "start_date",
+        [
+            "2017-01-01 23:59:59.999999999",
+            "2017-01-01 23:59:59.99999999",
+            "2017-01-01 23:59:59.9999999",
+            "2017-01-01 23:59:59.999999",
+            "2017-01-01 23:59:59.99999",
+            "2017-01-01 23:59:59.9999",
+        ],
+    )
+    def test_datetimeindex_highprecision(self, start_date):
+        # GH19030
+        # Check that high-precision time values for the end of day are
+        # included in repr for DatetimeIndex
+        s1 = Series(date_range(start=start_date, freq="D", periods=5))
+        result = str(s1)
+        assert start_date in result
+
+        dti = date_range(start=start_date, freq="D", periods=5)
+        s2 = Series(3, index=dti)
+        result = str(s2.index)
+        assert start_date in result
+
+    def test_mixed_datetime64(self):
+        df = DataFrame({"A": [1, 2], "B": ["2012-01-01", "2012-01-02"]})
+        df["B"] = pd.to_datetime(df.B)
+
+        result = repr(df.loc[0])
+        assert "2012-01-01" in result
+
+    def test_period(self):
+        # GH 12615
+        index = pd.period_range("2013-01", periods=6, freq="M")
+        s = Series(np.arange(6, dtype="int64"), index=index)
+        exp = (
+            "2013-01    0\n"
+            "2013-02    1\n"
+            "2013-03    2\n"
+            "2013-04    3\n"
+            "2013-05    4\n"
+            "2013-06    5\n"
+            "Freq: M, dtype: int64"
+        )
+        assert str(s) == exp
+
+        s = Series(index)
+        exp = (
+            "0    2013-01\n"
+            "1    2013-02\n"
+            "2    2013-03\n"
+            "3    2013-04\n"
+            "4    2013-05\n"
+            "5    2013-06\n"
+            "dtype: period[M]"
+        )
+        assert str(s) == exp
+
+        # periods with mixed freq
+        s = Series(
+            [
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02-01", freq="D"),
+                pd.Period("2011-03-01 09:00", freq="h"),
+            ]
+        )
+        exp = (
+            "0             2011-01\n1          2011-02-01\n"
+            "2    2011-03-01 09:00\ndtype: object"
+        )
+        assert str(s) == exp
+
+    def test_max_multi_index_display(self):
+        # GH 7101
+
+        # doc example (indexing.rst)
+
+        # multi-index
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = list(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples, names=["first", "second"])
+        s = Series(np.random.default_rng(2).standard_normal(8), index=index)
+
+        with option_context("display.max_rows", 10):
+            assert len(str(s).split("\n")) == 10
+        with option_context("display.max_rows", 3):
+            assert len(str(s).split("\n")) == 5
+        with option_context("display.max_rows", 2):
+            assert len(str(s).split("\n")) == 5
+        with option_context("display.max_rows", 1):
+            assert len(str(s).split("\n")) == 4
+        with option_context("display.max_rows", 0):
+            assert len(str(s).split("\n")) == 10
+
+        # index
+        s = Series(np.random.default_rng(2).standard_normal(8), None)
+
+        with option_context("display.max_rows", 10):
+            assert len(str(s).split("\n")) == 9
+        with option_context("display.max_rows", 3):
+            assert len(str(s).split("\n")) == 4
+        with option_context("display.max_rows", 2):
+            assert len(str(s).split("\n")) == 4
+        with option_context("display.max_rows", 1):
+            assert len(str(s).split("\n")) == 3
+        with option_context("display.max_rows", 0):
+            assert len(str(s).split("\n")) == 9
+
+    # Make sure #8532 is fixed
+    def test_consistent_format(self):
+        s = Series([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0.9999, 1, 1] * 10)
+        with option_context("display.max_rows", 10, "display.show_dimensions", False):
+            res = repr(s)
+        exp = (
+            "0      1.0000\n1      1.0000\n2      1.0000\n3      "
+            "1.0000\n4      1.0000\n        ...  \n125    "
+            "1.0000\n126    1.0000\n127    0.9999\n128    "
+            "1.0000\n129    1.0000\ndtype: float64"
+        )
+        assert res == exp
+
+    def chck_ncols(self, s):
+        lines = [
+            line for line in repr(s).split("\n") if not re.match(r"[^\.]*\.+", line)
+        ][:-1]
+        ncolsizes = len({len(line.strip()) for line in lines})
+        assert ncolsizes == 1
+
+    @pytest.mark.xfail(using_string_dtype(), reason="change when arrow is default")
+    def test_format_explicit(self):
+        test_sers = gen_series_formatting()
+        with option_context("display.max_rows", 4, "display.show_dimensions", False):
+            res = repr(test_sers["onel"])
+            exp = "0     a\n1     a\n     ..\n98    a\n99    a\ndtype: object"
+            assert exp == res
+            res = repr(test_sers["twol"])
+            exp = "0     ab\n1     ab\n      ..\n98    ab\n99    ab\ndtype: object"
+            assert exp == res
+            res = repr(test_sers["asc"])
+            exp = (
+                "0         a\n1        ab\n      ...  \n4     abcde\n5    "
+                "abcdef\ndtype: object"
+            )
+            assert exp == res
+            res = repr(test_sers["desc"])
+            exp = (
+                "5    abcdef\n4     abcde\n      ...  \n1        ab\n0         "
+                "a\ndtype: object"
+            )
+            assert exp == res
+
+    def test_ncols(self):
+        test_sers = gen_series_formatting()
+        for s in test_sers.values():
+            self.chck_ncols(s)
+
+    def test_max_rows_eq_one(self):
+        s = Series(range(10), dtype="int64")
+        with option_context("display.max_rows", 1):
+            strrepr = repr(s).split("\n")
+        exp1 = ["0", "0"]
+        res1 = strrepr[0].split()
+        assert exp1 == res1
+        exp2 = [".."]
+        res2 = strrepr[1].split()
+        assert exp2 == res2
+
+    def test_truncate_ndots(self):
+        def getndots(s):
+            return len(re.match(r"[^\.]*(\.*)", s).groups()[0])
+
+        s = Series([0, 2, 3, 6])
+        with option_context("display.max_rows", 2):
+            strrepr = repr(s).replace("\n", "")
+        assert getndots(strrepr) == 2
+
+        s = Series([0, 100, 200, 400])
+        with option_context("display.max_rows", 2):
+            strrepr = repr(s).replace("\n", "")
+        assert getndots(strrepr) == 3
+
+    def test_show_dimensions(self):
+        # gh-7117
+        s = Series(range(5))
+
+        assert "Length" not in repr(s)
+
+        with option_context("display.max_rows", 4):
+            assert "Length" in repr(s)
+
+        with option_context("display.show_dimensions", True):
+            assert "Length" in repr(s)
+
+        with option_context("display.max_rows", 4, "display.show_dimensions", False):
+            assert "Length" not in repr(s)
+
+    def test_repr_min_rows(self):
+        s = Series(range(20))
+
+        # default setting no truncation even if above min_rows
+        assert ".." not in repr(s)
+
+        s = Series(range(61))
+
+        # default of max_rows 60 triggers truncation if above
+        assert ".." in repr(s)
+
+        with option_context("display.max_rows", 10, "display.min_rows", 4):
+            # truncated after first two rows
+            assert ".." in repr(s)
+            assert "2  " not in repr(s)
+
+        with option_context("display.max_rows", 12, "display.min_rows", None):
+            # when set to None, follow value of max_rows
+            assert "5      5" in repr(s)
+
+        with option_context("display.max_rows", 10, "display.min_rows", 12):
+            # when set value higher as max_rows, use the minimum
+            assert "5      5" not in repr(s)
+
+        with option_context("display.max_rows", None, "display.min_rows", 12):
+            # max_rows of None -> never truncate
+            assert ".." not in repr(s)
+
+
+class TestGenericArrayFormatter:
+    def test_1d_array(self):
+        # _GenericArrayFormatter is used on types for which there isn't a dedicated
+        # formatter. np.bool_ is one of those types.
+        obj = fmt._GenericArrayFormatter(np.array([True, False]))
+        res = obj.get_result()
+        assert len(res) == 2
+        # Results should be right-justified.
+        assert res[0] == "  True"
+        assert res[1] == " False"
+
+    def test_2d_array(self):
+        obj = fmt._GenericArrayFormatter(np.array([[True, False], [False, True]]))
+        res = obj.get_result()
+        assert len(res) == 2
+        assert res[0] == " [True, False]"
+        assert res[1] == " [False, True]"
+
+    def test_3d_array(self):
+        obj = fmt._GenericArrayFormatter(
+            np.array([[[True, True], [False, False]], [[False, True], [True, False]]])
+        )
+        res = obj.get_result()
+        assert len(res) == 2
+        assert res[0] == " [[True, True], [False, False]]"
+        assert res[1] == " [[False, True], [True, False]]"
+
+    def test_2d_extension_type(self):
+        # GH 33770
+
+        # Define a stub extension type with just enough code to run Series.__repr__()
+        class DtypeStub(pd.api.extensions.ExtensionDtype):
+            @property
+            def type(self):
+                return np.ndarray
+
+            @property
+            def name(self):
+                return "DtypeStub"
+
+        class ExtTypeStub(pd.api.extensions.ExtensionArray):
+            def __len__(self) -> int:
+                return 2
+
+            def __getitem__(self, ix):
+                return [ix == 1, ix == 0]
+
+            @property
+            def dtype(self):
+                return DtypeStub()
+
+        series = Series(ExtTypeStub(), copy=False)
+        res = repr(series)  # This line crashed before #33770 was fixed.
+        expected = "\n".join(
+            ["0    [False True]", "1    [True False]", "dtype: DtypeStub"]
+        )
+        assert res == expected
+
+
+def _three_digit_exp():
+    return f"{1.7e8:.4g}" == "1.7e+008"
+
+
+class TestFloatArrayFormatter:
+    def test_misc(self):
+        obj = fmt.FloatArrayFormatter(np.array([], dtype=np.float64))
+        result = obj.get_result()
+        assert len(result) == 0
+
+    def test_format(self):
+        obj = fmt.FloatArrayFormatter(np.array([12, 0], dtype=np.float64))
+        result = obj.get_result()
+        assert result[0] == " 12.0"
+        assert result[1] == "  0.0"
+
+    def test_output_display_precision_trailing_zeroes(self):
+        # Issue #20359: trimming zeros while there is no decimal point
+
+        # Happens when display precision is set to zero
+        with option_context("display.precision", 0):
+            s = Series([840.0, 4200.0])
+            expected_output = "0     840\n1    4200\ndtype: float64"
+            assert str(s) == expected_output
+
+    @pytest.mark.parametrize(
+        "value,expected",
+        [
+            ([9.4444], "   0\n0  9"),
+            ([0.49], "       0\n0  5e-01"),
+            ([10.9999], "    0\n0  11"),
+            ([9.5444, 9.6], "    0\n0  10\n1  10"),
+            ([0.46, 0.78, -9.9999], "       0\n0  5e-01\n1  8e-01\n2 -1e+01"),
+        ],
+    )
+    def test_set_option_precision(self, value, expected):
+        # Issue #30122
+        # Precision was incorrectly shown
+
+        with option_context("display.precision", 0):
+            df_value = DataFrame(value)
+            assert str(df_value) == expected
+
+    def test_output_significant_digits(self):
+        # Issue #9764
+
+        # In case default display precision changes:
+        with option_context("display.precision", 6):
+            # DataFrame example from issue #9764
+            d = DataFrame(
+                {
+                    "col1": [
+                        9.999e-8,
+                        1e-7,
+                        1.0001e-7,
+                        2e-7,
+                        4.999e-7,
+                        5e-7,
+                        5.0001e-7,
+                        6e-7,
+                        9.999e-7,
+                        1e-6,
+                        1.0001e-6,
+                        2e-6,
+                        4.999e-6,
+                        5e-6,
+                        5.0001e-6,
+                        6e-6,
+                    ]
+                }
+            )
+
+            expected_output = {
+                (0, 6): "           col1\n"
+                "0  9.999000e-08\n"
+                "1  1.000000e-07\n"
+                "2  1.000100e-07\n"
+                "3  2.000000e-07\n"
+                "4  4.999000e-07\n"
+                "5  5.000000e-07",
+                (1, 6): "           col1\n"
+                "1  1.000000e-07\n"
+                "2  1.000100e-07\n"
+                "3  2.000000e-07\n"
+                "4  4.999000e-07\n"
+                "5  5.000000e-07",
+                (1, 8): "           col1\n"
+                "1  1.000000e-07\n"
+                "2  1.000100e-07\n"
+                "3  2.000000e-07\n"
+                "4  4.999000e-07\n"
+                "5  5.000000e-07\n"
+                "6  5.000100e-07\n"
+                "7  6.000000e-07",
+                (8, 16): "            col1\n"
+                "8   9.999000e-07\n"
+                "9   1.000000e-06\n"
+                "10  1.000100e-06\n"
+                "11  2.000000e-06\n"
+                "12  4.999000e-06\n"
+                "13  5.000000e-06\n"
+                "14  5.000100e-06\n"
+                "15  6.000000e-06",
+                (9, 16): "        col1\n"
+                "9   0.000001\n"
+                "10  0.000001\n"
+                "11  0.000002\n"
+                "12  0.000005\n"
+                "13  0.000005\n"
+                "14  0.000005\n"
+                "15  0.000006",
+            }
+
+            for (start, stop), v in expected_output.items():
+                assert str(d[start:stop]) == v
+
+    def test_too_long(self):
+        # GH 10451
+        with option_context("display.precision", 4):
+            # need both a number > 1e6 and something that normally formats to
+            # having length > display.precision + 6
+            df = DataFrame({"x": [12345.6789]})
+            assert str(df) == "            x\n0  12345.6789"
+            df = DataFrame({"x": [2e6]})
+            assert str(df) == "           x\n0  2000000.0"
+            df = DataFrame({"x": [12345.6789, 2e6]})
+            assert str(df) == "            x\n0  1.2346e+04\n1  2.0000e+06"
+
+
+class TestTimedelta64Formatter:
+    def test_days(self):
+        x = pd.to_timedelta(list(range(5)) + [NaT], unit="D")._values
+        result = fmt._Timedelta64Formatter(x).get_result()
+        assert result[0].strip() == "0 days"
+        assert result[1].strip() == "1 days"
+
+        result = fmt._Timedelta64Formatter(x[1:2]).get_result()
+        assert result[0].strip() == "1 days"
+
+        result = fmt._Timedelta64Formatter(x).get_result()
+        assert result[0].strip() == "0 days"
+        assert result[1].strip() == "1 days"
+
+        result = fmt._Timedelta64Formatter(x[1:2]).get_result()
+        assert result[0].strip() == "1 days"
+
+    def test_days_neg(self):
+        x = pd.to_timedelta(list(range(5)) + [NaT], unit="D")._values
+        result = fmt._Timedelta64Formatter(-x).get_result()
+        assert result[0].strip() == "0 days"
+        assert result[1].strip() == "-1 days"
+
+    def test_subdays(self):
+        y = pd.to_timedelta(list(range(5)) + [NaT], unit="s")._values
+        result = fmt._Timedelta64Formatter(y).get_result()
+        assert result[0].strip() == "0 days 00:00:00"
+        assert result[1].strip() == "0 days 00:00:01"
+
+    def test_subdays_neg(self):
+        y = pd.to_timedelta(list(range(5)) + [NaT], unit="s")._values
+        result = fmt._Timedelta64Formatter(-y).get_result()
+        assert result[0].strip() == "0 days 00:00:00"
+        assert result[1].strip() == "-1 days +23:59:59"
+
+    def test_zero(self):
+        x = pd.to_timedelta(list(range(1)) + [NaT], unit="D")._values
+        result = fmt._Timedelta64Formatter(x).get_result()
+        assert result[0].strip() == "0 days"
+
+        x = pd.to_timedelta(list(range(1)), unit="D")._values
+        result = fmt._Timedelta64Formatter(x).get_result()
+        assert result[0].strip() == "0 days"
+
+
+class TestDatetime64Formatter:
+    def test_mixed(self):
+        x = Series([datetime(2013, 1, 1), datetime(2013, 1, 1, 12), NaT])._values
+        result = fmt._Datetime64Formatter(x).get_result()
+        assert result[0].strip() == "2013-01-01 00:00:00"
+        assert result[1].strip() == "2013-01-01 12:00:00"
+
+    def test_dates(self):
+        x = Series([datetime(2013, 1, 1), datetime(2013, 1, 2), NaT])._values
+        result = fmt._Datetime64Formatter(x).get_result()
+        assert result[0].strip() == "2013-01-01"
+        assert result[1].strip() == "2013-01-02"
+
+    def test_date_nanos(self):
+        x = Series([Timestamp(200)])._values
+        result = fmt._Datetime64Formatter(x).get_result()
+        assert result[0].strip() == "1970-01-01 00:00:00.000000200"
+
+    def test_dates_display(self):
+        # 10170
+        # make sure that we are consistently display date formatting
+        x = Series(date_range("20130101 09:00:00", periods=5, freq="D"))
+        x.iloc[1] = np.nan
+        result = fmt._Datetime64Formatter(x._values).get_result()
+        assert result[0].strip() == "2013-01-01 09:00:00"
+        assert result[1].strip() == "NaT"
+        assert result[4].strip() == "2013-01-05 09:00:00"
+
+        x = Series(date_range("20130101 09:00:00", periods=5, freq="s"))
+        x.iloc[1] = np.nan
+        result = fmt._Datetime64Formatter(x._values).get_result()
+        assert result[0].strip() == "2013-01-01 09:00:00"
+        assert result[1].strip() == "NaT"
+        assert result[4].strip() == "2013-01-01 09:00:04"
+
+        x = Series(date_range("20130101 09:00:00", periods=5, freq="ms"))
+        x.iloc[1] = np.nan
+        result = fmt._Datetime64Formatter(x._values).get_result()
+        assert result[0].strip() == "2013-01-01 09:00:00.000"
+        assert result[1].strip() == "NaT"
+        assert result[4].strip() == "2013-01-01 09:00:00.004"
+
+        x = Series(date_range("20130101 09:00:00", periods=5, freq="us"))
+        x.iloc[1] = np.nan
+        result = fmt._Datetime64Formatter(x._values).get_result()
+        assert result[0].strip() == "2013-01-01 09:00:00.000000"
+        assert result[1].strip() == "NaT"
+        assert result[4].strip() == "2013-01-01 09:00:00.000004"
+
+        x = Series(date_range("20130101 09:00:00", periods=5, freq="ns"))
+        x.iloc[1] = np.nan
+        result = fmt._Datetime64Formatter(x._values).get_result()
+        assert result[0].strip() == "2013-01-01 09:00:00.000000000"
+        assert result[1].strip() == "NaT"
+        assert result[4].strip() == "2013-01-01 09:00:00.000000004"
+
+    def test_datetime64formatter_yearmonth(self):
+        x = Series([datetime(2016, 1, 1), datetime(2016, 2, 2)])._values
+
+        def format_func(x):
+            return x.strftime("%Y-%m")
+
+        formatter = fmt._Datetime64Formatter(x, formatter=format_func)
+        result = formatter.get_result()
+        assert result == ["2016-01", "2016-02"]
+
+    def test_datetime64formatter_hoursecond(self):
+        x = Series(
+            pd.to_datetime(["10:10:10.100", "12:12:12.120"], format="%H:%M:%S.%f")
+        )._values
+
+        def format_func(x):
+            return x.strftime("%H:%M")
+
+        formatter = fmt._Datetime64Formatter(x, formatter=format_func)
+        result = formatter.get_result()
+        assert result == ["10:10", "12:12"]
+
+    def test_datetime64formatter_tz_ms(self):
+        x = (
+            Series(
+                np.array(["2999-01-01", "2999-01-02", "NaT"], dtype="datetime64[ms]")
+            )
+            .dt.tz_localize("US/Pacific")
+            ._values
+        )
+        result = fmt._Datetime64TZFormatter(x).get_result()
+        assert result[0].strip() == "2999-01-01 00:00:00-08:00"
+        assert result[1].strip() == "2999-01-02 00:00:00-08:00"
+
+
+class TestFormatPercentiles:
+    @pytest.mark.parametrize(
+        "percentiles, expected",
+        [
+            (
+                [0.01999, 0.02001, 0.5, 0.666666, 0.9999],
+                ["1.999%", "2.001%", "50%", "66.667%", "99.99%"],
+            ),
+            (
+                [0, 0.5, 0.02001, 0.5, 0.666666, 0.9999],
+                ["0%", "50%", "2.0%", "50%", "66.67%", "99.99%"],
+            ),
+            ([0.281, 0.29, 0.57, 0.58], ["28.1%", "29%", "57%", "58%"]),
+            ([0.28, 0.29, 0.57, 0.58], ["28%", "29%", "57%", "58%"]),
+            (
+                [0.9, 0.99, 0.999, 0.9999, 0.99999],
+                ["90%", "99%", "99.9%", "99.99%", "99.999%"],
+            ),
+        ],
+    )
+    def test_format_percentiles(self, percentiles, expected):
+        result = fmt.format_percentiles(percentiles)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "percentiles",
+        [
+            ([0.1, np.nan, 0.5]),
+            ([-0.001, 0.1, 0.5]),
+            ([2, 0.1, 0.5]),
+            ([0.1, 0.5, "a"]),
+        ],
+    )
+    def test_error_format_percentiles(self, percentiles):
+        msg = r"percentiles should all be in the interval \[0,1\]"
+        with pytest.raises(ValueError, match=msg):
+            fmt.format_percentiles(percentiles)
+
+    def test_format_percentiles_integer_idx(self):
+        # Issue #26660
+        result = fmt.format_percentiles(np.linspace(0, 1, 10 + 1))
+        expected = [
+            "0%",
+            "10%",
+            "20%",
+            "30%",
+            "40%",
+            "50%",
+            "60%",
+            "70%",
+            "80%",
+            "90%",
+            "100%",
+        ]
+        assert result == expected
+
+
+@pytest.mark.parametrize("method", ["to_string", "to_html", "to_latex"])
+@pytest.mark.parametrize(
+    "encoding, data",
+    [(None, "abc"), ("utf-8", "abc"), ("gbk", "造成输出中文显示乱码"), ("foo", "abc")],
+)
+@pytest.mark.parametrize("filepath_or_buffer_id", ["string", "pathlike", "buffer"])
+def test_filepath_or_buffer_arg(
+    method,
+    tmp_path,
+    encoding,
+    data,
+    filepath_or_buffer_id,
+):
+    if filepath_or_buffer_id == "buffer":
+        filepath_or_buffer = StringIO()
+    elif filepath_or_buffer_id == "pathlike":
+        filepath_or_buffer = tmp_path / "foo"
+    else:
+        filepath_or_buffer = str(tmp_path / "foo")
+
+    df = DataFrame([data])
+    if method in ["to_latex"]:  # uses styler implementation
+        pytest.importorskip("jinja2")
+
+    if filepath_or_buffer_id not in ["string", "pathlike"] and encoding is not None:
+        with pytest.raises(
+            ValueError, match="buf is not a file name and encoding is specified."
+        ):
+            getattr(df, method)(buf=filepath_or_buffer, encoding=encoding)
+    elif encoding == "foo":
+        with pytest.raises(LookupError, match="unknown encoding"):
+            getattr(df, method)(buf=filepath_or_buffer, encoding=encoding)
+    else:
+        expected = getattr(df, method)()
+        getattr(df, method)(buf=filepath_or_buffer, encoding=encoding)
+        encoding = encoding or "utf-8"
+        if filepath_or_buffer_id == "string":
+            with open(filepath_or_buffer, encoding=encoding) as f:
+                result = f.read()
+        elif filepath_or_buffer_id == "pathlike":
+            result = filepath_or_buffer.read_text(encoding=encoding)
+        elif filepath_or_buffer_id == "buffer":
+            result = filepath_or_buffer.getvalue()
+        assert result == expected
+    if filepath_or_buffer_id == "buffer":
+        assert not filepath_or_buffer.closed
+
+
+@pytest.mark.parametrize("method", ["to_string", "to_html", "to_latex"])
+def test_filepath_or_buffer_bad_arg_raises(float_frame, method):
+    if method in ["to_latex"]:  # uses styler implementation
+        pytest.importorskip("jinja2")
+    msg = "buf is not a file name and it has no write method"
+    with pytest.raises(TypeError, match=msg):
+        getattr(float_frame, method)(buf=object())
diff --git a/pandas/pandas/tests/io/formats/test_ipython_compat.py b/pandas/pandas/tests/io/formats/test_ipython_compat.py
new file mode 100644
index 0000000000000..8512f41396906
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_ipython_compat.py
@@ -0,0 +1,90 @@
+import numpy as np
+
+import pandas._config.config as cf
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+
+
+class TestTableSchemaRepr:
+    def test_publishes(self, ip):
+        ipython = ip.instance(config=ip.config)
+        df = DataFrame({"A": [1, 2]})
+        objects = [df["A"], df]  # dataframe / series
+        expected_keys = [
+            {"text/plain", "application/vnd.dataresource+json"},
+            {"text/plain", "text/html", "application/vnd.dataresource+json"},
+        ]
+
+        opt = cf.option_context("display.html.table_schema", True)
+        last_obj = None
+        for obj, expected in zip(objects, expected_keys):
+            last_obj = obj
+            with opt:
+                formatted = ipython.display_formatter.format(obj)
+            assert set(formatted[0].keys()) == expected
+
+        with_latex = cf.option_context("styler.render.repr", "latex")
+
+        with opt, with_latex:
+            formatted = ipython.display_formatter.format(last_obj)
+
+        expected = {
+            "text/plain",
+            "text/html",
+            "text/latex",
+            "application/vnd.dataresource+json",
+        }
+        assert set(formatted[0].keys()) == expected
+
+    def test_publishes_not_implemented(self, ip):
+        # column MultiIndex
+        # GH#15996
+        midx = MultiIndex.from_product([["A", "B"], ["a", "b", "c"]])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, len(midx))), columns=midx
+        )
+
+        opt = cf.option_context("display.html.table_schema", True)
+
+        with opt:
+            formatted = ip.instance(config=ip.config).display_formatter.format(df)
+
+        expected = {"text/plain", "text/html"}
+        assert set(formatted[0].keys()) == expected
+
+    def test_config_on(self):
+        df = DataFrame({"A": [1, 2]})
+        with cf.option_context("display.html.table_schema", True):
+            result = df._repr_data_resource_()
+
+        assert result is not None
+
+    def test_config_default_off(self):
+        df = DataFrame({"A": [1, 2]})
+        with cf.option_context("display.html.table_schema", False):
+            result = df._repr_data_resource_()
+
+        assert result is None
+
+    def test_enable_data_resource_formatter(self, ip):
+        # GH#10491
+        formatters = ip.instance(config=ip.config).display_formatter.formatters
+        mimetype = "application/vnd.dataresource+json"
+
+        with cf.option_context("display.html.table_schema", True):
+            assert "application/vnd.dataresource+json" in formatters
+            assert formatters[mimetype].enabled
+
+        # still there, just disabled
+        assert "application/vnd.dataresource+json" in formatters
+        assert not formatters[mimetype].enabled
+
+        # able to re-set
+        with cf.option_context("display.html.table_schema", True):
+            assert "application/vnd.dataresource+json" in formatters
+            assert formatters[mimetype].enabled
+            # smoke test that it works
+            ip.instance(config=ip.config).display_formatter.format(cf)
diff --git a/pandas/pandas/tests/io/formats/test_printing.py b/pandas/pandas/tests/io/formats/test_printing.py
new file mode 100644
index 0000000000000..7d154235d2c4a
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_printing.py
@@ -0,0 +1,172 @@
+# Note! This file is aimed specifically at pandas.io.formats.printing utility
+# functions, not the general printing of pandas objects.
+from collections.abc import Mapping
+import string
+
+import pytest
+
+import pandas._config.config as cf
+
+import pandas as pd
+
+from pandas.io.formats import printing
+
+
+@pytest.mark.parametrize(
+    "input_names, expected_names",
+    [
+        (["'a b"], "['\\'a b']"),  # Escape leading quote
+        (["test's b"], "['test\\'s b']"),  # Escape apostrophe
+        (["'test' b"], "['\\'test\\' b']"),  # Escape surrounding quotes
+        (["test b'"], "['test b\\'']"),  # Escape single quote
+        (["test\n' b"], "['test\\n\\' b']"),  # Escape quotes, preserve newline
+    ],
+)
+def test_formatted_index_names(input_names, expected_names):
+    # GH#60190
+    df = pd.DataFrame({name: [1, 2, 3] for name in input_names}).set_index(input_names)
+    formatted_names = str(df.index.names)
+
+    assert formatted_names == expected_names
+
+
+def test_adjoin():
+    data = [["a", "b", "c"], ["dd", "ee", "ff"], ["ggg", "hhh", "iii"]]
+    expected = "a  dd  ggg\nb  ee  hhh\nc  ff  iii"
+
+    adjoined = printing.adjoin(2, *data)
+
+    assert adjoined == expected
+
+
+class MyMapping(Mapping):
+    def __getitem__(self, key):
+        return 4
+
+    def __iter__(self):
+        return iter(["a", "b"])
+
+    def __len__(self):
+        return 2
+
+
+class TestPPrintThing:
+    def test_repr_binary_type(self):
+        letters = string.ascii_letters
+        try:
+            raw = bytes(letters, encoding=cf.get_option("display.encoding"))
+        except TypeError:
+            raw = bytes(letters)
+        b = str(raw.decode("utf-8"))
+        res = printing.pprint_thing(b, quote_strings=True)
+        assert res == repr(b)
+        res = printing.pprint_thing(b, quote_strings=False)
+        assert res == b
+
+    def test_repr_obeys_max_seq_limit(self):
+        with cf.option_context("display.max_seq_items", 2000):
+            assert len(printing.pprint_thing(list(range(1000)))) > 1000
+
+        with cf.option_context("display.max_seq_items", 5):
+            assert len(printing.pprint_thing(list(range(1000)))) < 100
+
+        with cf.option_context("display.max_seq_items", 1):
+            assert len(printing.pprint_thing(list(range(1000)))) < 9
+
+    def test_repr_set(self):
+        assert printing.pprint_thing({1}) == "{1}"
+
+    def test_repr_dict(self):
+        assert printing.pprint_thing({"a": 4, "b": 4}) == "{'a': 4, 'b': 4}"
+
+    def test_repr_mapping(self):
+        assert printing.pprint_thing(MyMapping()) == "{'a': 4, 'b': 4}"
+
+    def test_repr_frozenset(self):
+        assert printing.pprint_thing(frozenset([1, 2])) == "frozenset({1, 2})"
+
+
+class TestFormatBase:
+    def test_adjoin(self):
+        data = [["a", "b", "c"], ["dd", "ee", "ff"], ["ggg", "hhh", "iii"]]
+        expected = "a  dd  ggg\nb  ee  hhh\nc  ff  iii"
+
+        adjoined = printing.adjoin(2, *data)
+
+        assert adjoined == expected
+
+    def test_adjoin_unicode(self):
+        data = [["あ", "b", "c"], ["dd", "ええ", "ff"], ["ggg", "hhh", "いいい"]]
+        expected = "あ  dd  ggg\nb  ええ  hhh\nc  ff  いいい"
+        adjoined = printing.adjoin(2, *data)
+        assert adjoined == expected
+
+        adj = printing._EastAsianTextAdjustment()
+
+        expected = """あ  dd    ggg
+b   ええ  hhh
+c   ff    いいい"""
+
+        adjoined = adj.adjoin(2, *data)
+        assert adjoined == expected
+        cols = adjoined.split("\n")
+        assert adj.len(cols[0]) == 13
+        assert adj.len(cols[1]) == 13
+        assert adj.len(cols[2]) == 16
+
+        expected = """あ       dd         ggg
+b        ええ       hhh
+c        ff         いいい"""
+
+        adjoined = adj.adjoin(7, *data)
+        assert adjoined == expected
+        cols = adjoined.split("\n")
+        assert adj.len(cols[0]) == 23
+        assert adj.len(cols[1]) == 23
+        assert adj.len(cols[2]) == 26
+
+    def test_justify(self):
+        adj = printing._EastAsianTextAdjustment()
+
+        def just(x, *args, **kwargs):
+            # wrapper to test single str
+            return adj.justify([x], *args, **kwargs)[0]
+
+        assert just("abc", 5, mode="left") == "abc  "
+        assert just("abc", 5, mode="center") == " abc "
+        assert just("abc", 5, mode="right") == "  abc"
+        assert just("abc", 5, mode="left") == "abc  "
+        assert just("abc", 5, mode="center") == " abc "
+        assert just("abc", 5, mode="right") == "  abc"
+
+        assert just("パンダ", 5, mode="left") == "パンダ"
+        assert just("パンダ", 5, mode="center") == "パンダ"
+        assert just("パンダ", 5, mode="right") == "パンダ"
+
+        assert just("パンダ", 10, mode="left") == "パンダ    "
+        assert just("パンダ", 10, mode="center") == "  パンダ  "
+        assert just("パンダ", 10, mode="right") == "    パンダ"
+
+    def test_east_asian_len(self):
+        adj = printing._EastAsianTextAdjustment()
+
+        assert adj.len("abc") == 3
+        assert adj.len("abc") == 3
+
+        assert adj.len("パンダ") == 6
+        assert adj.len("ﾊﾟﾝﾀﾞ") == 5
+        assert adj.len("パンダpanda") == 11
+        assert adj.len("ﾊﾟﾝﾀﾞpanda") == 10
+
+    def test_ambiguous_width(self):
+        adj = printing._EastAsianTextAdjustment()
+        assert adj.len("¡¡ab") == 4
+
+        with cf.option_context("display.unicode.ambiguous_as_wide", True):
+            adj = printing._EastAsianTextAdjustment()
+            assert adj.len("¡¡ab") == 6
+
+        data = [["あ", "b", "c"], ["dd", "ええ", "ff"], ["ggg", "¡¡ab", "いいい"]]
+        expected = "あ  dd    ggg \nb   ええ  ¡¡ab\nc   ff    いいい"
+        adjoined = adj.adjoin(2, *data)
+        assert adjoined == expected
diff --git a/pandas/pandas/tests/io/formats/test_to_csv.py b/pandas/pandas/tests/io/formats/test_to_csv.py
new file mode 100644
index 0000000000000..6d762fdeb8d79
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_csv.py
@@ -0,0 +1,743 @@
+import io
+import os
+import sys
+from zipfile import ZipFile
+
+from _csv import Error
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    compat,
+)
+import pandas._testing as tm
+
+
+class TestToCSV:
+    def test_to_csv_with_single_column(self):
+        # see gh-18676, https://bugs.python.org/issue32255
+        #
+        # Python's CSV library adds an extraneous '""'
+        # before the newline when the NaN-value is in
+        # the first row. Otherwise, only the newline
+        # character is added. This behavior is inconsistent
+        # and was patched in https://bugs.python.org/pull_request4672.
+        df1 = DataFrame([None, 1])
+        expected1 = """\
+""
+1.0
+"""
+        with tm.ensure_clean("test.csv") as path:
+            df1.to_csv(path, header=None, index=None)
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected1
+
+        df2 = DataFrame([1, None])
+        expected2 = """\
+1.0
+""
+"""
+        with tm.ensure_clean("test.csv") as path:
+            df2.to_csv(path, header=None, index=None)
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected2
+
+    def test_to_csv_default_encoding(self):
+        # GH17097
+        df = DataFrame({"col": ["AAAAA", "ÄÄÄÄÄ", "ßßßßß", "聞聞聞聞聞"]})
+
+        with tm.ensure_clean("test.csv") as path:
+            # the default to_csv encoding is uft-8.
+            df.to_csv(path)
+            tm.assert_frame_equal(pd.read_csv(path, index_col=0), df)
+
+    def test_to_csv_quotechar(self):
+        df = DataFrame({"col": [1, 2]})
+        expected = """\
+"","col"
+"0","1"
+"1","2"
+"""
+
+        with tm.ensure_clean("test.csv") as path:
+            df.to_csv(path, quoting=1)  # 1=QUOTE_ALL
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+        expected = """\
+$$,$col$
+$0$,$1$
+$1$,$2$
+"""
+
+        with tm.ensure_clean("test.csv") as path:
+            df.to_csv(path, quoting=1, quotechar="$")
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+        with tm.ensure_clean("test.csv") as path:
+            with pytest.raises(TypeError, match="quotechar"):
+                df.to_csv(path, quoting=1, quotechar=None)
+
+    def test_to_csv_doublequote(self):
+        df = DataFrame({"col": ['a"a', '"bb"']})
+        expected = '''\
+"","col"
+"0","a""a"
+"1","""bb"""
+'''
+
+        with tm.ensure_clean("test.csv") as path:
+            df.to_csv(path, quoting=1, doublequote=True)  # QUOTE_ALL
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+        with tm.ensure_clean("test.csv") as path:
+            with pytest.raises(Error, match="escapechar"):
+                df.to_csv(path, doublequote=False)  # no escapechar set
+
+    def test_to_csv_escapechar(self):
+        df = DataFrame({"col": ['a"a', '"bb"']})
+        expected = """\
+"","col"
+"0","a\\"a"
+"1","\\"bb\\""
+"""
+
+        with tm.ensure_clean("test.csv") as path:  # QUOTE_ALL
+            df.to_csv(path, quoting=1, doublequote=False, escapechar="\\")
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+        df = DataFrame({"col": ["a,a", ",bb,"]})
+        expected = """\
+,col
+0,a\\,a
+1,\\,bb\\,
+"""
+
+        with tm.ensure_clean("test.csv") as path:
+            df.to_csv(path, quoting=3, escapechar="\\")  # QUOTE_NONE
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+    def test_csv_to_string(self):
+        df = DataFrame({"col": [1, 2]})
+        expected_rows = [",col", "0,1", "1,2"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv() == expected
+
+    def test_to_csv_decimal(self):
+        # see gh-781
+        df = DataFrame({"col1": [1], "col2": ["a"], "col3": [10.1]})
+
+        expected_rows = [",col1,col2,col3", "0,1,a,10.1"]
+        expected_default = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv() == expected_default
+
+        expected_rows = [";col1;col2;col3", "0;1;a;10,1"]
+        expected_european_excel = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv(decimal=",", sep=";") == expected_european_excel
+
+        expected_rows = [",col1,col2,col3", "0,1,a,10.10"]
+        expected_float_format_default = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv(float_format="%.2f") == expected_float_format_default
+
+        expected_rows = [";col1;col2;col3", "0;1;a;10,10"]
+        expected_float_format = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert (
+            df.to_csv(decimal=",", sep=";", float_format="%.2f")
+            == expected_float_format
+        )
+
+        # see gh-11553: testing if decimal is taken into account for '0.0'
+        df = DataFrame({"a": [0, 1.1], "b": [2.2, 3.3], "c": 1})
+
+        expected_rows = ["a,b,c", "0^0,2^2,1", "1^1,3^3,1"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv(index=False, decimal="^") == expected
+
+        # same but for an index
+        assert df.set_index("a").to_csv(decimal="^") == expected
+
+        # same for a multi-index
+        assert df.set_index(["a", "b"]).to_csv(decimal="^") == expected
+
+    def test_to_csv_float_format(self):
+        # testing if float_format is taken into account for the index
+        # GH 11553
+        df = DataFrame({"a": [0, 1], "b": [2.2, 3.3], "c": 1})
+
+        expected_rows = ["a,b,c", "0,2.20,1", "1,3.30,1"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.set_index("a").to_csv(float_format="%.2f") == expected
+
+        # same for a multi-index
+        assert df.set_index(["a", "b"]).to_csv(float_format="%.2f") == expected
+
+    def test_to_csv_na_rep(self):
+        # see gh-11553
+        #
+        # Testing if NaN values are correctly represented in the index.
+        df = DataFrame({"a": [0, np.nan], "b": [0, 1], "c": [2, 3]})
+        expected_rows = ["a,b,c", "0.0,0,2", "_,1,3"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        assert df.set_index("a").to_csv(na_rep="_") == expected
+        assert df.set_index(["a", "b"]).to_csv(na_rep="_") == expected
+
+        # now with an index containing only NaNs
+        df = DataFrame({"a": np.nan, "b": [0, 1], "c": [2, 3]})
+        expected_rows = ["a,b,c", "_,0,2", "_,1,3"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        assert df.set_index("a").to_csv(na_rep="_") == expected
+        assert df.set_index(["a", "b"]).to_csv(na_rep="_") == expected
+
+        # check if na_rep parameter does not break anything when no NaN
+        df = DataFrame({"a": 0, "b": [0, 1], "c": [2, 3]})
+        expected_rows = ["a,b,c", "0,0,2", "0,1,3"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        assert df.set_index("a").to_csv(na_rep="_") == expected
+        assert df.set_index(["a", "b"]).to_csv(na_rep="_") == expected
+
+        csv = pd.Series(["a", pd.NA, "c"]).to_csv(na_rep="ZZZZZ")
+        expected = tm.convert_rows_list_to_csv_str([",0", "0,a", "1,ZZZZZ", "2,c"])
+        assert expected == csv
+
+    def test_to_csv_na_rep_nullable_string(self, nullable_string_dtype):
+        # GH 29975
+        # Make sure full na_rep shows up when a dtype is provided
+        expected = tm.convert_rows_list_to_csv_str([",0", "0,a", "1,ZZZZZ", "2,c"])
+        csv = pd.Series(["a", pd.NA, "c"], dtype=nullable_string_dtype).to_csv(
+            na_rep="ZZZZZ"
+        )
+        assert expected == csv
+
+    def test_to_csv_date_format(self):
+        # GH 10209
+        df_sec = DataFrame({"A": pd.date_range("20130101", periods=5, freq="s")})
+        df_day = DataFrame({"A": pd.date_range("20130101", periods=5, freq="D")})
+
+        expected_rows = [
+            ",A",
+            "0,2013-01-01 00:00:00",
+            "1,2013-01-01 00:00:01",
+            "2,2013-01-01 00:00:02",
+            "3,2013-01-01 00:00:03",
+            "4,2013-01-01 00:00:04",
+        ]
+        expected_default_sec = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df_sec.to_csv() == expected_default_sec
+
+        expected_rows = [
+            ",A",
+            "0,2013-01-01 00:00:00",
+            "1,2013-01-02 00:00:00",
+            "2,2013-01-03 00:00:00",
+            "3,2013-01-04 00:00:00",
+            "4,2013-01-05 00:00:00",
+        ]
+        expected_ymdhms_day = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df_day.to_csv(date_format="%Y-%m-%d %H:%M:%S") == expected_ymdhms_day
+
+        expected_rows = [
+            ",A",
+            "0,2013-01-01",
+            "1,2013-01-01",
+            "2,2013-01-01",
+            "3,2013-01-01",
+            "4,2013-01-01",
+        ]
+        expected_ymd_sec = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df_sec.to_csv(date_format="%Y-%m-%d") == expected_ymd_sec
+
+        expected_rows = [
+            ",A",
+            "0,2013-01-01",
+            "1,2013-01-02",
+            "2,2013-01-03",
+            "3,2013-01-04",
+            "4,2013-01-05",
+        ]
+        expected_default_day = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df_day.to_csv() == expected_default_day
+        assert df_day.to_csv(date_format="%Y-%m-%d") == expected_default_day
+
+        # see gh-7791
+        #
+        # Testing if date_format parameter is taken into account
+        # for multi-indexed DataFrames.
+        df_sec["B"] = 0
+        df_sec["C"] = 1
+
+        expected_rows = ["A,B,C", "2013-01-01,0,1.0"]
+        expected_ymd_sec = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        df_sec_grouped = df_sec.groupby([pd.Grouper(key="A", freq="1h"), "B"])
+        assert df_sec_grouped.mean().to_csv(date_format="%Y-%m-%d") == expected_ymd_sec
+
+    def test_to_csv_different_datetime_formats(self):
+        # GH#21734
+        df = DataFrame(
+            {
+                "date": pd.to_datetime("1970-01-01"),
+                "datetime": pd.date_range("1970-01-01", periods=2, freq="h"),
+            }
+        )
+        expected_rows = [
+            "date,datetime",
+            "1970-01-01,1970-01-01 00:00:00",
+            "1970-01-01,1970-01-01 01:00:00",
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert df.to_csv(index=False) == expected
+
+    def test_to_csv_date_format_in_categorical(self):
+        # GH#40754
+        ser = pd.Series(pd.to_datetime(["2021-03-27", pd.NaT], format="%Y-%m-%d"))
+        ser = ser.astype("category")
+        expected = tm.convert_rows_list_to_csv_str(["0", "2021-03-27", '""'])
+        assert ser.to_csv(index=False) == expected
+
+        ser = pd.Series(
+            pd.date_range(
+                start="2021-03-27", freq="D", periods=1, tz="Europe/Berlin"
+            ).append(pd.DatetimeIndex([pd.NaT]))
+        )
+        ser = ser.astype("category")
+        assert ser.to_csv(index=False, date_format="%Y-%m-%d") == expected
+
+    def test_to_csv_float_ea_float_format(self):
+        # GH#45991
+        df = DataFrame({"a": [1.1, 2.02, pd.NA, 6.000006], "b": "c"})
+        df["a"] = df["a"].astype("Float64")
+        result = df.to_csv(index=False, float_format="%.5f")
+        expected = tm.convert_rows_list_to_csv_str(
+            ["a,b", "1.10000,c", "2.02000,c", ",c", "6.00001,c"]
+        )
+        assert result == expected
+
+    def test_to_csv_float_ea_no_float_format(self):
+        # GH#45991
+        df = DataFrame({"a": [1.1, 2.02, pd.NA, 6.000006], "b": "c"})
+        df["a"] = df["a"].astype("Float64")
+        result = df.to_csv(index=False)
+        expected = tm.convert_rows_list_to_csv_str(
+            ["a,b", "1.1,c", "2.02,c", ",c", "6.000006,c"]
+        )
+        assert result == expected
+
+    def test_to_csv_multi_index(self):
+        # see gh-6618
+        df = DataFrame([1], columns=pd.MultiIndex.from_arrays([[1], [2]]))
+
+        exp_rows = [",1", ",2", "0,1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv() == exp
+
+        exp_rows = ["1", "2", "1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv(index=False) == exp
+
+        df = DataFrame(
+            [1],
+            columns=pd.MultiIndex.from_arrays([[1], [2]]),
+            index=pd.MultiIndex.from_arrays([[1], [2]]),
+        )
+
+        exp_rows = [",,1", ",,2", "1,2,1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv() == exp
+
+        exp_rows = ["1", "2", "1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv(index=False) == exp
+
+        df = DataFrame([1], columns=pd.MultiIndex.from_arrays([["foo"], ["bar"]]))
+
+        exp_rows = [",foo", ",bar", "0,1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv() == exp
+
+        exp_rows = ["foo", "bar", "1"]
+        exp = tm.convert_rows_list_to_csv_str(exp_rows)
+        assert df.to_csv(index=False) == exp
+
+    @pytest.mark.parametrize(
+        "ind,expected",
+        [
+            (
+                pd.MultiIndex(levels=[[1.0]], codes=[[0]], names=["x"]),
+                "x,data\n1.0,1\n",
+            ),
+            (
+                pd.MultiIndex(
+                    levels=[[1.0], [2.0]], codes=[[0], [0]], names=["x", "y"]
+                ),
+                "x,y,data\n1.0,2.0,1\n",
+            ),
+        ],
+    )
+    def test_to_csv_single_level_multi_index(self, ind, expected, frame_or_series):
+        # see gh-19589
+        obj = frame_or_series(pd.Series([1], ind, name="data"))
+
+        result = obj.to_csv(lineterminator="\n", header=True)
+        assert result == expected
+
+    def test_to_csv_string_array_ascii(self):
+        # GH 10813
+        str_array = [{"names": ["foo", "bar"]}, {"names": ["baz", "qux"]}]
+        df = DataFrame(str_array)
+        expected_ascii = """\
+,names
+0,"['foo', 'bar']"
+1,"['baz', 'qux']"
+"""
+        with tm.ensure_clean("str_test.csv") as path:
+            df.to_csv(path, encoding="ascii")
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected_ascii
+
+    def test_to_csv_string_array_utf8(self):
+        # GH 10813
+        str_array = [{"names": ["foo", "bar"]}, {"names": ["baz", "qux"]}]
+        df = DataFrame(str_array)
+        expected_utf8 = """\
+,names
+0,"['foo', 'bar']"
+1,"['baz', 'qux']"
+"""
+        with tm.ensure_clean("unicode_test.csv") as path:
+            df.to_csv(path, encoding="utf-8")
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected_utf8
+
+    def test_to_csv_string_with_lf(self):
+        # GH 20353
+        data = {"int": [1, 2, 3], "str_lf": ["abc", "d\nef", "g\nh\n\ni"]}
+        df = DataFrame(data)
+        with tm.ensure_clean("lf_test.csv") as path:
+            # case 1: The default line terminator(=os.linesep)(PR 21406)
+            os_linesep = os.linesep.encode("utf-8")
+            expected_noarg = (
+                b"int,str_lf"
+                + os_linesep
+                + b"1,abc"
+                + os_linesep
+                + b'2,"d\nef"'
+                + os_linesep
+                + b'3,"g\nh\n\ni"'
+                + os_linesep
+            )
+            df.to_csv(path, index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_noarg
+        with tm.ensure_clean("lf_test.csv") as path:
+            # case 2: LF as line terminator
+            expected_lf = b'int,str_lf\n1,abc\n2,"d\nef"\n3,"g\nh\n\ni"\n'
+            df.to_csv(path, lineterminator="\n", index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_lf
+        with tm.ensure_clean("lf_test.csv") as path:
+            # case 3: CRLF as line terminator
+            # 'lineterminator' should not change inner element
+            expected_crlf = b'int,str_lf\r\n1,abc\r\n2,"d\nef"\r\n3,"g\nh\n\ni"\r\n'
+            df.to_csv(path, lineterminator="\r\n", index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_crlf
+
+    def test_to_csv_string_with_crlf(self):
+        # GH 20353
+        data = {"int": [1, 2, 3], "str_crlf": ["abc", "d\r\nef", "g\r\nh\r\n\r\ni"]}
+        df = DataFrame(data)
+        with tm.ensure_clean("crlf_test.csv") as path:
+            # case 1: The default line terminator(=os.linesep)(PR 21406)
+            os_linesep = os.linesep.encode("utf-8")
+            expected_noarg = (
+                b"int,str_crlf"
+                + os_linesep
+                + b"1,abc"
+                + os_linesep
+                + b'2,"d\r\nef"'
+                + os_linesep
+                + b'3,"g\r\nh\r\n\r\ni"'
+                + os_linesep
+            )
+            df.to_csv(path, index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_noarg
+        with tm.ensure_clean("crlf_test.csv") as path:
+            # case 2: LF as line terminator
+            expected_lf = b'int,str_crlf\n1,abc\n2,"d\r\nef"\n3,"g\r\nh\r\n\r\ni"\n'
+            df.to_csv(path, lineterminator="\n", index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_lf
+        with tm.ensure_clean("crlf_test.csv") as path:
+            # case 3: CRLF as line terminator
+            # 'lineterminator' should not change inner element
+            expected_crlf = (
+                b'int,str_crlf\r\n1,abc\r\n2,"d\r\nef"\r\n3,"g\r\nh\r\n\r\ni"\r\n'
+            )
+            df.to_csv(path, lineterminator="\r\n", index=False)
+            with open(path, "rb") as f:
+                assert f.read() == expected_crlf
+
+    def test_to_csv_stdout_file(self, capsys):
+        # GH 21561
+        df = DataFrame([["foo", "bar"], ["baz", "qux"]], columns=["name_1", "name_2"])
+        expected_rows = [",name_1,name_2", "0,foo,bar", "1,baz,qux"]
+        expected_ascii = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        df.to_csv(sys.stdout, encoding="ascii")
+        captured = capsys.readouterr()
+
+        assert captured.out == expected_ascii
+        assert not sys.stdout.closed
+
+    @pytest.mark.xfail(
+        compat.is_platform_windows(),
+        reason=(
+            "Especially in Windows, file stream should not be passed"
+            "to csv writer without newline='' option."
+            "(https://docs.python.org/3/library/csv.html#csv.writer)"
+        ),
+    )
+    def test_to_csv_write_to_open_file(self):
+        # GH 21696
+        df = DataFrame({"a": ["x", "y", "z"]})
+        expected = """\
+manual header
+x
+y
+z
+"""
+        with tm.ensure_clean("test.txt") as path:
+            with open(path, "w", encoding="utf-8") as f:
+                f.write("manual header\n")
+                df.to_csv(f, header=None, index=None)
+            with open(path, encoding="utf-8") as f:
+                assert f.read() == expected
+
+    def test_to_csv_write_to_open_file_with_newline_py3(self):
+        # see gh-21696
+        # see gh-20353
+        df = DataFrame({"a": ["x", "y", "z"]})
+        expected_rows = ["x", "y", "z"]
+        expected = "manual header\n" + tm.convert_rows_list_to_csv_str(expected_rows)
+        with tm.ensure_clean("test.txt") as path:
+            with open(path, "w", newline="", encoding="utf-8") as f:
+                f.write("manual header\n")
+                df.to_csv(f, header=None, index=None)
+
+            with open(path, "rb") as f:
+                assert f.read() == bytes(expected, "utf-8")
+
+    @pytest.mark.parametrize("to_infer", [True, False])
+    @pytest.mark.parametrize("read_infer", [True, False])
+    def test_to_csv_compression(
+        self, compression_only, read_infer, to_infer, compression_to_extension
+    ):
+        # see gh-15008
+        compression = compression_only
+
+        # We'll complete file extension subsequently.
+        filename = "test."
+        filename += compression_to_extension[compression]
+
+        df = DataFrame({"A": [1]})
+
+        to_compression = "infer" if to_infer else compression
+        read_compression = "infer" if read_infer else compression
+
+        with tm.ensure_clean(filename) as path:
+            df.to_csv(path, compression=to_compression)
+            result = pd.read_csv(path, index_col=0, compression=read_compression)
+            tm.assert_frame_equal(result, df)
+
+    def test_to_csv_compression_dict(self, compression_only):
+        # GH 26023
+        method = compression_only
+        df = DataFrame({"ABC": [1]})
+        filename = "to_csv_compress_as_dict."
+        extension = {
+            "gzip": "gz",
+            "zstd": "zst",
+        }.get(method, method)
+        filename += extension
+        with tm.ensure_clean(filename) as path:
+            df.to_csv(path, compression={"method": method})
+            read_df = pd.read_csv(path, index_col=0)
+            tm.assert_frame_equal(read_df, df)
+
+    def test_to_csv_compression_dict_no_method_raises(self):
+        # GH 26023
+        df = DataFrame({"ABC": [1]})
+        compression = {"some_option": True}
+        msg = "must have key 'method'"
+
+        with tm.ensure_clean("out.zip") as path:
+            with pytest.raises(ValueError, match=msg):
+                df.to_csv(path, compression=compression)
+
+    @pytest.mark.parametrize("compression", ["zip", "infer"])
+    @pytest.mark.parametrize("archive_name", ["test_to_csv.csv", "test_to_csv.zip"])
+    def test_to_csv_zip_arguments(self, compression, archive_name):
+        # GH 26023
+        df = DataFrame({"ABC": [1]})
+        with tm.ensure_clean("to_csv_archive_name.zip") as path:
+            df.to_csv(
+                path, compression={"method": compression, "archive_name": archive_name}
+            )
+            with ZipFile(path) as zp:
+                assert len(zp.filelist) == 1
+                archived_file = zp.filelist[0].filename
+                assert archived_file == archive_name
+
+    @pytest.mark.parametrize(
+        "filename,expected_arcname",
+        [
+            ("archive.csv", "archive.csv"),
+            ("archive.tsv", "archive.tsv"),
+            ("archive.csv.zip", "archive.csv"),
+            ("archive.tsv.zip", "archive.tsv"),
+            ("archive.zip", "archive"),
+        ],
+    )
+    def test_to_csv_zip_infer_name(self, tmp_path, filename, expected_arcname):
+        # GH 39465
+        df = DataFrame({"ABC": [1]})
+        path = tmp_path / filename
+        df.to_csv(path, compression="zip")
+        with ZipFile(path) as zp:
+            assert len(zp.filelist) == 1
+            archived_file = zp.filelist[0].filename
+            assert archived_file == expected_arcname
+
+    @pytest.mark.parametrize("df_new_type", ["Int64"])
+    def test_to_csv_na_rep_long_string(self, df_new_type):
+        # see gh-25099
+        df = DataFrame({"c": [float("nan")] * 3})
+        df = df.astype(df_new_type)
+        expected_rows = ["c", "mynull", "mynull", "mynull"]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+
+        result = df.to_csv(index=False, na_rep="mynull", encoding="ascii")
+
+        assert expected == result
+
+    def test_to_csv_timedelta_precision(self):
+        # GH 6783
+        s = pd.Series([1, 1]).astype("timedelta64[ns]")
+        buf = io.StringIO()
+        s.to_csv(buf)
+        result = buf.getvalue()
+        expected_rows = [
+            ",0",
+            "0,0 days 00:00:00.000000001",
+            "1,0 days 00:00:00.000000001",
+        ]
+        expected = tm.convert_rows_list_to_csv_str(expected_rows)
+        assert result == expected
+
+    def test_na_rep_truncated(self):
+        # https://github.com/pandas-dev/pandas/issues/31447
+        result = pd.Series(range(8, 12)).to_csv(na_rep="-")
+        expected = tm.convert_rows_list_to_csv_str([",0", "0,8", "1,9", "2,10", "3,11"])
+        assert result == expected
+
+        result = pd.Series([True, False]).to_csv(na_rep="nan")
+        expected = tm.convert_rows_list_to_csv_str([",0", "0,True", "1,False"])
+        assert result == expected
+
+        result = pd.Series([1.1, 2.2]).to_csv(na_rep=".")
+        expected = tm.convert_rows_list_to_csv_str([",0", "0,1.1", "1,2.2"])
+        assert result == expected
+
+    @pytest.mark.parametrize("errors", ["surrogatepass", "ignore", "replace"])
+    def test_to_csv_errors(self, errors):
+        # GH 22610
+        data = ["\ud800foo"]
+        ser = pd.Series(data, index=Index(data, dtype=object), dtype=object)
+        with tm.ensure_clean("test.csv") as path:
+            ser.to_csv(path, errors=errors)
+        # No use in reading back the data as it is not the same anymore
+        # due to the error handling
+
+    @pytest.mark.parametrize("mode", ["wb", "w"])
+    def test_to_csv_binary_handle(self, mode):
+        """
+        Binary file objects should work (if 'mode' contains a 'b') or even without
+        it in most cases.
+
+        GH 35058 and GH 19827
+        """
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        with tm.ensure_clean() as path:
+            with open(path, mode="w+b") as handle:
+                df.to_csv(handle, mode=mode)
+            tm.assert_frame_equal(df, pd.read_csv(path, index_col=0))
+
+    @pytest.mark.parametrize("mode", ["wb", "w"])
+    def test_to_csv_encoding_binary_handle(self, mode):
+        """
+        Binary file objects should honor a specified encoding.
+
+        GH 23854 and GH 13068 with binary handles
+        """
+        # example from GH 23854
+        content = "a, b, 🐟".encode("utf-8-sig")
+        buffer = io.BytesIO(content)
+        df = pd.read_csv(buffer, encoding="utf-8-sig")
+
+        buffer = io.BytesIO()
+        df.to_csv(buffer, mode=mode, encoding="utf-8-sig", index=False)
+        buffer.seek(0)  # tests whether file handle wasn't closed
+        assert buffer.getvalue().startswith(content)
+
+        # example from GH 13068
+        with tm.ensure_clean() as path:
+            with open(path, "w+b") as handle:
+                DataFrame().to_csv(handle, mode=mode, encoding="utf-8-sig")
+
+                handle.seek(0)
+                assert handle.read().startswith(b'\xef\xbb\xbf""')
+
+
+def test_to_csv_iterative_compression_name(compression):
+    # GH 38714
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    with tm.ensure_clean() as path:
+        df.to_csv(path, compression=compression, chunksize=1)
+        tm.assert_frame_equal(
+            pd.read_csv(path, compression=compression, index_col=0), df
+        )
+
+
+def test_to_csv_iterative_compression_buffer(compression):
+    # GH 38714
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    with io.BytesIO() as buffer:
+        df.to_csv(buffer, compression=compression, chunksize=1)
+        buffer.seek(0)
+        tm.assert_frame_equal(
+            pd.read_csv(buffer, compression=compression, index_col=0), df
+        )
+        assert not buffer.closed
diff --git a/pandas/pandas/tests/io/formats/test_to_excel.py b/pandas/pandas/tests/io/formats/test_to_excel.py
new file mode 100644
index 0000000000000..b40201b9ba1e6
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_excel.py
@@ -0,0 +1,430 @@
+"""Tests formatting as writer-agnostic ExcelCells
+
+ExcelFormatter is tested implicitly in pandas/tests/io/excel
+"""
+
+import string
+
+import pytest
+
+from pandas.errors import CSSWarning
+
+import pandas._testing as tm
+
+from pandas.io.formats.excel import (
+    CssExcelCell,
+    CSSToExcelConverter,
+)
+
+
+@pytest.mark.parametrize(
+    "css,expected",
+    [
+        # FONT
+        # - name
+        ("font-family: foo,bar", {"font": {"name": "foo"}}),
+        ('font-family: "foo bar",baz', {"font": {"name": "foo bar"}}),
+        ("font-family: foo,\nbar", {"font": {"name": "foo"}}),
+        ("font-family: foo, bar,    baz", {"font": {"name": "foo"}}),
+        ("font-family: bar, foo", {"font": {"name": "bar"}}),
+        ("font-family: 'foo bar', baz", {"font": {"name": "foo bar"}}),
+        ("font-family: 'foo \\'bar', baz", {"font": {"name": "foo 'bar"}}),
+        ('font-family: "foo \\"bar", baz', {"font": {"name": 'foo "bar'}}),
+        ('font-family: "foo ,bar", baz', {"font": {"name": "foo ,bar"}}),
+        # - family
+        ("font-family: serif", {"font": {"name": "serif", "family": 1}}),
+        ("font-family: Serif", {"font": {"name": "serif", "family": 1}}),
+        ("font-family: roman, serif", {"font": {"name": "roman", "family": 1}}),
+        ("font-family: roman, sans-serif", {"font": {"name": "roman", "family": 2}}),
+        ("font-family: roman, sans serif", {"font": {"name": "roman"}}),
+        ("font-family: roman, sansserif", {"font": {"name": "roman"}}),
+        ("font-family: roman, cursive", {"font": {"name": "roman", "family": 4}}),
+        ("font-family: roman, fantasy", {"font": {"name": "roman", "family": 5}}),
+        # - size
+        ("font-size: 1em", {"font": {"size": 12}}),
+        ("font-size: xx-small", {"font": {"size": 6}}),
+        ("font-size: x-small", {"font": {"size": 7.5}}),
+        ("font-size: small", {"font": {"size": 9.6}}),
+        ("font-size: medium", {"font": {"size": 12}}),
+        ("font-size: large", {"font": {"size": 13.5}}),
+        ("font-size: x-large", {"font": {"size": 18}}),
+        ("font-size: xx-large", {"font": {"size": 24}}),
+        ("font-size: 50%", {"font": {"size": 6}}),
+        # - bold
+        ("font-weight: 100", {"font": {"bold": False}}),
+        ("font-weight: 200", {"font": {"bold": False}}),
+        ("font-weight: 300", {"font": {"bold": False}}),
+        ("font-weight: 400", {"font": {"bold": False}}),
+        ("font-weight: normal", {"font": {"bold": False}}),
+        ("font-weight: lighter", {"font": {"bold": False}}),
+        ("font-weight: bold", {"font": {"bold": True}}),
+        ("font-weight: bolder", {"font": {"bold": True}}),
+        ("font-weight: 700", {"font": {"bold": True}}),
+        ("font-weight: 800", {"font": {"bold": True}}),
+        ("font-weight: 900", {"font": {"bold": True}}),
+        # - italic
+        ("font-style: italic", {"font": {"italic": True}}),
+        ("font-style: oblique", {"font": {"italic": True}}),
+        # - underline
+        ("text-decoration: underline", {"font": {"underline": "single"}}),
+        ("text-decoration: overline", {}),
+        ("text-decoration: none", {}),
+        # - strike
+        ("text-decoration: line-through", {"font": {"strike": True}}),
+        (
+            "text-decoration: underline line-through",
+            {"font": {"strike": True, "underline": "single"}},
+        ),
+        (
+            "text-decoration: underline; text-decoration: line-through",
+            {"font": {"strike": True}},
+        ),
+        # - color
+        ("color: red", {"font": {"color": "FF0000"}}),
+        ("color: #ff0000", {"font": {"color": "FF0000"}}),
+        ("color: #f0a", {"font": {"color": "FF00AA"}}),
+        # - shadow
+        ("text-shadow: none", {"font": {"shadow": False}}),
+        ("text-shadow: 0px -0em 0px #CCC", {"font": {"shadow": False}}),
+        ("text-shadow: 0px -0em 0px #999", {"font": {"shadow": False}}),
+        ("text-shadow: 0px -0em 0px", {"font": {"shadow": False}}),
+        ("text-shadow: 2px -0em 0px #CCC", {"font": {"shadow": True}}),
+        ("text-shadow: 0px -2em 0px #CCC", {"font": {"shadow": True}}),
+        ("text-shadow: 0px -0em 2px #CCC", {"font": {"shadow": True}}),
+        ("text-shadow: 0px -0em 2px", {"font": {"shadow": True}}),
+        ("text-shadow: 0px -2em", {"font": {"shadow": True}}),
+        # FILL
+        # - color, fillType
+        (
+            "background-color: red",
+            {"fill": {"fgColor": "FF0000", "patternType": "solid"}},
+        ),
+        (
+            "background-color: #ff0000",
+            {"fill": {"fgColor": "FF0000", "patternType": "solid"}},
+        ),
+        (
+            "background-color: #f0a",
+            {"fill": {"fgColor": "FF00AA", "patternType": "solid"}},
+        ),
+        # BORDER
+        # - style
+        (
+            "border-style: solid",
+            {
+                "border": {
+                    "top": {"style": "medium"},
+                    "bottom": {"style": "medium"},
+                    "left": {"style": "medium"},
+                    "right": {"style": "medium"},
+                }
+            },
+        ),
+        (
+            "border-style: solid; border-width: thin",
+            {
+                "border": {
+                    "top": {"style": "thin"},
+                    "bottom": {"style": "thin"},
+                    "left": {"style": "thin"},
+                    "right": {"style": "thin"},
+                }
+            },
+        ),
+        (
+            "border-top-style: solid; border-top-width: thin",
+            {"border": {"top": {"style": "thin"}}},
+        ),
+        (
+            "border-top-style: solid; border-top-width: 1pt",
+            {"border": {"top": {"style": "thin"}}},
+        ),
+        ("border-top-style: solid", {"border": {"top": {"style": "medium"}}}),
+        (
+            "border-top-style: solid; border-top-width: medium",
+            {"border": {"top": {"style": "medium"}}},
+        ),
+        (
+            "border-top-style: solid; border-top-width: 2pt",
+            {"border": {"top": {"style": "medium"}}},
+        ),
+        (
+            "border-top-style: solid; border-top-width: thick",
+            {"border": {"top": {"style": "thick"}}},
+        ),
+        (
+            "border-top-style: solid; border-top-width: 4pt",
+            {"border": {"top": {"style": "thick"}}},
+        ),
+        (
+            "border-top-style: dotted",
+            {"border": {"top": {"style": "mediumDashDotDot"}}},
+        ),
+        (
+            "border-top-style: dotted; border-top-width: thin",
+            {"border": {"top": {"style": "dotted"}}},
+        ),
+        ("border-top-style: dashed", {"border": {"top": {"style": "mediumDashed"}}}),
+        (
+            "border-top-style: dashed; border-top-width: thin",
+            {"border": {"top": {"style": "dashed"}}},
+        ),
+        ("border-top-style: double", {"border": {"top": {"style": "double"}}}),
+        # - color
+        (
+            "border-style: solid; border-color: #0000ff",
+            {
+                "border": {
+                    "top": {"style": "medium", "color": "0000FF"},
+                    "right": {"style": "medium", "color": "0000FF"},
+                    "bottom": {"style": "medium", "color": "0000FF"},
+                    "left": {"style": "medium", "color": "0000FF"},
+                }
+            },
+        ),
+        (
+            "border-top-style: double; border-top-color: blue",
+            {"border": {"top": {"style": "double", "color": "0000FF"}}},
+        ),
+        (
+            "border-top-style: solid; border-top-color: #06c",
+            {"border": {"top": {"style": "medium", "color": "0066CC"}}},
+        ),
+        (
+            "border-top-color: blue",
+            {"border": {"top": {"color": "0000FF", "style": "none"}}},
+        ),
+        # ALIGNMENT
+        # - horizontal
+        ("text-align: center", {"alignment": {"horizontal": "center"}}),
+        ("text-align: left", {"alignment": {"horizontal": "left"}}),
+        ("text-align: right", {"alignment": {"horizontal": "right"}}),
+        ("text-align: justify", {"alignment": {"horizontal": "justify"}}),
+        # - vertical
+        ("vertical-align: top", {"alignment": {"vertical": "top"}}),
+        ("vertical-align: text-top", {"alignment": {"vertical": "top"}}),
+        ("vertical-align: middle", {"alignment": {"vertical": "center"}}),
+        ("vertical-align: bottom", {"alignment": {"vertical": "bottom"}}),
+        ("vertical-align: text-bottom", {"alignment": {"vertical": "bottom"}}),
+        # - wrap_text
+        ("white-space: nowrap", {"alignment": {"wrap_text": False}}),
+        ("white-space: pre", {"alignment": {"wrap_text": False}}),
+        ("white-space: pre-line", {"alignment": {"wrap_text": False}}),
+        ("white-space: normal", {"alignment": {"wrap_text": True}}),
+        # NUMBER FORMAT
+        ("number-format: 0%", {"number_format": {"format_code": "0%"}}),
+        (
+            "number-format: 0§[Red](0)§-§@;",
+            {"number_format": {"format_code": "0;[red](0);-;@"}},  # GH 46152
+        ),
+    ],
+)
+def test_css_to_excel(css, expected):
+    convert = CSSToExcelConverter()
+    assert expected == convert(css)
+
+
+def test_css_to_excel_multiple():
+    convert = CSSToExcelConverter()
+    actual = convert(
+        """
+        font-weight: bold;
+        text-decoration: underline;
+        color: red;
+        border-width: thin;
+        text-align: center;
+        vertical-align: top;
+        unused: something;
+    """
+    )
+    assert {
+        "font": {"bold": True, "underline": "single", "color": "FF0000"},
+        "border": {
+            "top": {"style": "thin"},
+            "right": {"style": "thin"},
+            "bottom": {"style": "thin"},
+            "left": {"style": "thin"},
+        },
+        "alignment": {"horizontal": "center", "vertical": "top"},
+    } == actual
+
+
+@pytest.mark.parametrize(
+    "css,inherited,expected",
+    [
+        ("font-weight: bold", "", {"font": {"bold": True}}),
+        ("", "font-weight: bold", {"font": {"bold": True}}),
+        (
+            "font-weight: bold",
+            "font-style: italic",
+            {"font": {"bold": True, "italic": True}},
+        ),
+        ("font-style: normal", "font-style: italic", {"font": {"italic": False}}),
+        ("font-style: inherit", "", {}),
+        (
+            "font-style: normal; font-style: inherit",
+            "font-style: italic",
+            {"font": {"italic": True}},
+        ),
+    ],
+)
+def test_css_to_excel_inherited(css, inherited, expected):
+    convert = CSSToExcelConverter(inherited)
+    assert expected == convert(css)
+
+
+@pytest.mark.parametrize(
+    "input_color,output_color",
+    (
+        list(CSSToExcelConverter.NAMED_COLORS.items())
+        + [("#" + rgb, rgb) for rgb in CSSToExcelConverter.NAMED_COLORS.values()]
+        + [("#F0F", "FF00FF"), ("#ABC", "AABBCC")]
+    ),
+)
+def test_css_to_excel_good_colors(input_color, output_color):
+    # see gh-18392
+    css = (
+        f"border-top-color: {input_color}; "
+        f"border-right-color: {input_color}; "
+        f"border-bottom-color: {input_color}; "
+        f"border-left-color: {input_color}; "
+        f"background-color: {input_color}; "
+        f"color: {input_color}"
+    )
+
+    expected = {}
+
+    expected["fill"] = {"patternType": "solid", "fgColor": output_color}
+
+    expected["font"] = {"color": output_color}
+
+    expected["border"] = {
+        k: {"color": output_color, "style": "none"}
+        for k in ("top", "right", "bottom", "left")
+    }
+
+    with tm.assert_produces_warning(None):
+        convert = CSSToExcelConverter()
+        assert expected == convert(css)
+
+
+@pytest.mark.parametrize("input_color", [None, "not-a-color"])
+def test_css_to_excel_bad_colors(input_color):
+    # see gh-18392
+    css = (
+        f"border-top-color: {input_color}; "
+        f"border-right-color: {input_color}; "
+        f"border-bottom-color: {input_color}; "
+        f"border-left-color: {input_color}; "
+        f"background-color: {input_color}; "
+        f"color: {input_color}"
+    )
+
+    expected = {}
+
+    if input_color is not None:
+        expected["fill"] = {"patternType": "solid"}
+
+    with tm.assert_produces_warning(CSSWarning, match="Unhandled color format"):
+        convert = CSSToExcelConverter()
+        assert expected == convert(css)
+
+
+def tests_css_named_colors_valid():
+    upper_hexs = set(map(str.upper, string.hexdigits))
+    for color in CSSToExcelConverter.NAMED_COLORS.values():
+        assert len(color) == 6 and all(c in upper_hexs for c in color)
+
+
+def test_css_named_colors_from_mpl_present():
+    mpl_colors = pytest.importorskip("matplotlib.colors")
+
+    pd_colors = CSSToExcelConverter.NAMED_COLORS
+    for name, color in mpl_colors.CSS4_COLORS.items():
+        assert name in pd_colors and pd_colors[name] == color[1:]
+
+
+@pytest.mark.parametrize(
+    "styles,expected",
+    [
+        ([("color", "green"), ("color", "red")], "color: red;"),
+        ([("font-weight", "bold"), ("font-weight", "normal")], "font-weight: normal;"),
+        ([("text-align", "center"), ("TEXT-ALIGN", "right")], "text-align: right;"),
+    ],
+)
+def test_css_excel_cell_precedence(styles, expected):
+    """It applies favors latter declarations over former declarations"""
+    # See GH 47371
+    converter = CSSToExcelConverter()
+    converter._call_cached.cache_clear()
+    css_styles = {(0, 0): styles}
+    cell = CssExcelCell(
+        row=0,
+        col=0,
+        val="",
+        style=None,
+        css_styles=css_styles,
+        css_row=0,
+        css_col=0,
+        css_converter=converter,
+    )
+    converter._call_cached.cache_clear()
+
+    assert cell.style == converter(expected)
+
+
+@pytest.mark.parametrize(
+    "styles,cache_hits,cache_misses",
+    [
+        ([[("color", "green"), ("color", "red"), ("color", "green")]], 0, 1),
+        (
+            [
+                [("font-weight", "bold")],
+                [("font-weight", "normal"), ("font-weight", "bold")],
+            ],
+            1,
+            1,
+        ),
+        ([[("text-align", "center")], [("TEXT-ALIGN", "center")]], 1, 1),
+        (
+            [
+                [("font-weight", "bold"), ("text-align", "center")],
+                [("font-weight", "bold"), ("text-align", "left")],
+            ],
+            0,
+            2,
+        ),
+        (
+            [
+                [("font-weight", "bold"), ("text-align", "center")],
+                [("font-weight", "bold"), ("text-align", "left")],
+                [("font-weight", "bold"), ("text-align", "center")],
+            ],
+            1,
+            2,
+        ),
+    ],
+)
+def test_css_excel_cell_cache(styles, cache_hits, cache_misses):
+    """It caches unique cell styles"""
+    # See GH 47371
+    converter = CSSToExcelConverter()
+    converter._call_cached.cache_clear()
+
+    css_styles = {(0, i): _style for i, _style in enumerate(styles)}
+    for css_row, css_col in css_styles:
+        CssExcelCell(
+            row=0,
+            col=0,
+            val="",
+            style=None,
+            css_styles=css_styles,
+            css_row=css_row,
+            css_col=css_col,
+            css_converter=converter,
+        )
+    cache_info = converter._call_cached.cache_info()
+    converter._call_cached.cache_clear()
+
+    assert cache_info.hits == cache_hits
+    assert cache_info.misses == cache_misses
diff --git a/pandas/pandas/tests/io/formats/test_to_html.py b/pandas/pandas/tests/io/formats/test_to_html.py
new file mode 100644
index 0000000000000..9c75314b66fa2
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_html.py
@@ -0,0 +1,1167 @@
+from datetime import datetime
+from io import StringIO
+import itertools
+import re
+import textwrap
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    get_option,
+    option_context,
+)
+
+import pandas.io.formats.format as fmt
+
+lorem_ipsum = (
+    "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod "
+    "tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim "
+    "veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex "
+    "ea commodo consequat. Duis aute irure dolor in reprehenderit in "
+    "voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur "
+    "sint occaecat cupidatat non proident, sunt in culpa qui officia "
+    "deserunt mollit anim id est laborum."
+)
+
+
+def expected_html(datapath, name):
+    """
+    Read HTML file from formats data directory.
+
+    Parameters
+    ----------
+    datapath : pytest fixture
+        The datapath fixture injected into a test by pytest.
+    name : str
+        The name of the HTML file without the suffix.
+
+    Returns
+    -------
+    str : contents of HTML file.
+    """
+    filename = ".".join([name, "html"])
+    filepath = datapath("io", "formats", "data", "html", filename)
+    with open(filepath, encoding="utf-8") as f:
+        html = f.read()
+    return html.rstrip()
+
+
+@pytest.fixture(params=["mixed", "empty"])
+def biggie_df_fixture(request):
+    """Fixture for a big mixed Dataframe and an empty Dataframe"""
+    if request.param == "mixed":
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(200),
+                "B": Index([f"{i}?!" for i in range(200)]),
+            },
+            index=np.arange(200),
+        )
+        df.loc[:20, "A"] = np.nan
+        df.loc[:20, "B"] = np.nan
+        return df
+    elif request.param == "empty":
+        df = DataFrame(index=np.arange(200))
+        return df
+
+
+@pytest.fixture(params=fmt.VALID_JUSTIFY_PARAMETERS)
+def justify(request):
+    return request.param
+
+
+@pytest.mark.parametrize("col_space", [30, 50])
+def test_to_html_with_col_space(col_space):
+    df = DataFrame(np.random.default_rng(2).random(size=(1, 3)))
+    # check that col_space affects HTML generation
+    # and be very brittle about it.
+    result = df.to_html(col_space=col_space)
+    hdrs = [x for x in result.split(r"\n") if re.search(r"<th[>\s]", x)]
+    assert len(hdrs) > 0
+    for h in hdrs:
+        assert "min-width" in h
+        assert str(col_space) in h
+
+
+def test_to_html_with_column_specific_col_space_raises():
+    df = DataFrame(
+        np.random.default_rng(2).random(size=(3, 3)), columns=["a", "b", "c"]
+    )
+
+    msg = (
+        "Col_space length\\(\\d+\\) should match DataFrame number of columns\\(\\d+\\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.to_html(col_space=[30, 40])
+
+    with pytest.raises(ValueError, match=msg):
+        df.to_html(col_space=[30, 40, 50, 60])
+
+    msg = "unknown column"
+    with pytest.raises(ValueError, match=msg):
+        df.to_html(col_space={"a": "foo", "b": 23, "d": 34})
+
+
+def test_to_html_with_column_specific_col_space():
+    df = DataFrame(
+        np.random.default_rng(2).random(size=(3, 3)), columns=["a", "b", "c"]
+    )
+
+    result = df.to_html(col_space={"a": "2em", "b": 23})
+    hdrs = [x for x in result.split("\n") if re.search(r"<th[>\s]", x)]
+    assert 'min-width: 2em;">a</th>' in hdrs[1]
+    assert 'min-width: 23px;">b</th>' in hdrs[2]
+    assert "<th>c</th>" in hdrs[3]
+
+    result = df.to_html(col_space=["1em", 2, 3])
+    hdrs = [x for x in result.split("\n") if re.search(r"<th[>\s]", x)]
+    assert 'min-width: 1em;">a</th>' in hdrs[1]
+    assert 'min-width: 2px;">b</th>' in hdrs[2]
+    assert 'min-width: 3px;">c</th>' in hdrs[3]
+
+
+def test_to_html_with_empty_string_label():
+    # GH 3547, to_html regards empty string labels as repeated labels
+    data = {"c1": ["a", "b"], "c2": ["a", ""], "data": [1, 2]}
+    df = DataFrame(data).set_index(["c1", "c2"])
+    result = df.to_html()
+    assert "rowspan" not in result
+
+
+@pytest.mark.parametrize(
+    "df_data,expected",
+    [
+        ({"\u03c3": np.arange(10.0)}, "unicode_1"),
+        ({"A": ["\u03c3"]}, "unicode_2"),
+    ],
+)
+def test_to_html_unicode(df_data, expected, datapath):
+    df = DataFrame(df_data)
+    expected = expected_html(datapath, expected)
+    result = df.to_html()
+    assert result == expected
+
+
+def test_to_html_encoding(float_frame, tmp_path):
+    # GH 28663
+    path = tmp_path / "test.html"
+    float_frame.to_html(path, encoding="gbk")
+    with open(str(path), encoding="gbk") as f:
+        assert float_frame.to_html() == f.read()
+
+
+def test_to_html_decimal(datapath):
+    # GH 12031
+    df = DataFrame({"A": [6.0, 3.1, 2.2]})
+    result = df.to_html(decimal=",")
+    expected = expected_html(datapath, "gh12031_expected_output")
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "kwargs,string,expected",
+    [
+        ({}, "<type 'str'>", "escaped"),
+        ({"escape": False}, "<b>bold</b>", "escape_disabled"),
+    ],
+)
+def test_to_html_escaped(kwargs, string, expected, datapath):
+    a = "str<ing1 &amp;"
+    b = "stri>ng2 &amp;"
+
+    test_dict = {"co<l1": {a: string, b: string}, "co>l2": {a: string, b: string}}
+    result = DataFrame(test_dict).to_html(**kwargs)
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+@pytest.mark.parametrize("index_is_named", [True, False])
+def test_to_html_multiindex_index_false(index_is_named, datapath):
+    # GH 8452
+    df = DataFrame(
+        {"a": range(2), "b": range(3, 5), "c": range(5, 7), "d": range(3, 5)}
+    )
+    df.columns = MultiIndex.from_product([["a", "b"], ["c", "d"]])
+    if index_is_named:
+        df.index = Index(df.index.values, name="idx")
+    result = df.to_html(index=False)
+    expected = expected_html(datapath, "gh8452_expected_output")
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "multi_sparse,expected",
+    [
+        (False, "multiindex_sparsify_false_multi_sparse_1"),
+        (False, "multiindex_sparsify_false_multi_sparse_2"),
+        (True, "multiindex_sparsify_1"),
+        (True, "multiindex_sparsify_2"),
+    ],
+)
+def test_to_html_multiindex_sparsify(multi_sparse, expected, datapath):
+    index = MultiIndex.from_arrays([[0, 0, 1, 1], [0, 1, 0, 1]], names=["foo", None])
+    df = DataFrame([[0, 1], [2, 3], [4, 5], [6, 7]], index=index)
+    if expected.endswith("2"):
+        df.columns = index[::2]
+    with option_context("display.multi_sparse", multi_sparse):
+        result = df.to_html()
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "max_rows,expected",
+    [
+        (60, "gh14882_expected_output_1"),
+        # Test that ... appears in a middle level
+        (56, "gh14882_expected_output_2"),
+    ],
+)
+def test_to_html_multiindex_odd_even_truncate(max_rows, expected, datapath):
+    # GH 14882 - Issue on truncation with odd length DataFrame
+    index = MultiIndex.from_product(
+        [[100, 200, 300], [10, 20, 30], [1, 2, 3, 4, 5, 6, 7]], names=["a", "b", "c"]
+    )
+    df = DataFrame({"n": range(len(index))}, index=index)
+    result = df.to_html(max_rows=max_rows)
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "df,formatters,expected",
+    [
+        (
+            DataFrame(
+                [[0, 1], [2, 3], [4, 5], [6, 7]],
+                columns=Index(["foo", None], dtype=object),
+                index=np.arange(4),
+            ),
+            {"__index__": lambda x: "abcd"[x]},
+            "index_formatter",
+        ),
+        (
+            DataFrame({"months": [datetime(2016, 1, 1), datetime(2016, 2, 2)]}),
+            {"months": lambda x: x.strftime("%Y-%m")},
+            "datetime64_monthformatter",
+        ),
+        (
+            DataFrame(
+                {
+                    "hod": pd.to_datetime(
+                        ["10:10:10.100", "12:12:12.120"], format="%H:%M:%S.%f"
+                    )
+                }
+            ),
+            {"hod": lambda x: x.strftime("%H:%M")},
+            "datetime64_hourformatter",
+        ),
+        (
+            DataFrame(
+                {
+                    "i": pd.Series([1, 2], dtype="int64"),
+                    "f": pd.Series([1, 2], dtype="float64"),
+                    "I": pd.Series([1, 2], dtype="Int64"),
+                    "s": pd.Series([1, 2], dtype="string"),
+                    "b": pd.Series([True, False], dtype="boolean"),
+                    "c": pd.Series(["a", "b"], dtype=pd.CategoricalDtype(["a", "b"])),
+                    "o": pd.Series([1, "2"], dtype=object),
+                }
+            ),
+            [lambda x: "formatted"] * 7,
+            "various_dtypes_formatted",
+        ),
+    ],
+)
+def test_to_html_formatters(df, formatters, expected, datapath):
+    expected = expected_html(datapath, expected)
+    result = df.to_html(formatters=formatters)
+    assert result == expected
+
+
+def test_to_html_regression_GH6098():
+    df = DataFrame(
+        {
+            "clé1": ["a", "a", "b", "b", "a"],
+            "clé2": ["1er", "2ème", "1er", "2ème", "1er"],
+            "données1": np.random.default_rng(2).standard_normal(5),
+            "données2": np.random.default_rng(2).standard_normal(5),
+        }
+    )
+
+    # it works
+    df.pivot_table(index=["clé1"], columns=["clé2"])._repr_html_()
+
+
+def test_to_html_truncate(datapath):
+    index = pd.date_range(start="20010101", freq="D", periods=20)
+    df = DataFrame(index=index, columns=range(20))
+    result = df.to_html(max_rows=8, max_cols=4)
+    expected = expected_html(datapath, "truncate")
+    assert result == expected
+
+
+@pytest.mark.parametrize("size", [1, 5])
+def test_html_invalid_formatters_arg_raises(size):
+    # issue-28469
+    df = DataFrame(columns=["a", "b", "c"])
+    msg = "Formatters length({}) should match DataFrame number of columns(3)"
+    with pytest.raises(ValueError, match=re.escape(msg.format(size))):
+        df.to_html(formatters=["{}".format] * size)
+
+
+def test_to_html_truncate_formatter(datapath):
+    # issue-25955
+    data = [
+        {"A": 1, "B": 2, "C": 3, "D": 4},
+        {"A": 5, "B": 6, "C": 7, "D": 8},
+        {"A": 9, "B": 10, "C": 11, "D": 12},
+        {"A": 13, "B": 14, "C": 15, "D": 16},
+    ]
+
+    df = DataFrame(data)
+    fmt = lambda x: str(x) + "_mod"
+    formatters = [fmt, fmt, None, None]
+    result = df.to_html(formatters=formatters, max_cols=3)
+    expected = expected_html(datapath, "truncate_formatter")
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "sparsify,expected",
+    [(True, "truncate_multi_index"), (False, "truncate_multi_index_sparse_off")],
+)
+def test_to_html_truncate_multi_index(sparsify, expected, datapath):
+    arrays = [
+        ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+        ["one", "two", "one", "two", "one", "two", "one", "two"],
+    ]
+    df = DataFrame(index=arrays, columns=arrays)
+    result = df.to_html(max_rows=7, max_cols=7, sparsify=sparsify)
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "option,result,expected",
+    [
+        (None, lambda df: df.to_html(), "1"),
+        (None, lambda df: df.to_html(border=2), "2"),
+        (2, lambda df: df.to_html(), "2"),
+        (2, lambda df: df._repr_html_(), "2"),
+    ],
+)
+def test_to_html_border(option, result, expected):
+    df = DataFrame({"A": [1, 2]})
+    if option is None:
+        result = result(df)
+    else:
+        with option_context("display.html.border", option):
+            result = result(df)
+    expected = f'border="{expected}"'
+    assert expected in result
+
+
+@pytest.mark.parametrize("biggie_df_fixture", ["mixed"], indirect=True)
+def test_to_html(biggie_df_fixture):
+    # TODO: split this test
+    df = biggie_df_fixture
+    s = df.to_html()
+
+    buf = StringIO()
+    retval = df.to_html(buf=buf)
+    assert retval is None
+    assert buf.getvalue() == s
+
+    assert isinstance(s, str)
+
+    df.to_html(columns=["B", "A"], col_space=17)
+    df.to_html(columns=["B", "A"], formatters={"A": lambda x: f"{x:.1f}"})
+
+    df.to_html(columns=["B", "A"], float_format=str)
+    df.to_html(columns=["B", "A"], col_space=12, float_format=str)
+
+
+@pytest.mark.parametrize("biggie_df_fixture", ["empty"], indirect=True)
+def test_to_html_empty_dataframe(biggie_df_fixture):
+    df = biggie_df_fixture
+    df.to_html()
+
+
+def test_to_html_filename(biggie_df_fixture, tmpdir):
+    df = biggie_df_fixture
+    expected = df.to_html()
+    path = tmpdir.join("test.html")
+    df.to_html(path)
+    result = path.read()
+    assert result == expected
+
+
+def test_to_html_with_no_bold():
+    df = DataFrame({"x": np.random.default_rng(2).standard_normal(5)})
+    html = df.to_html(bold_rows=False)
+    result = html[html.find("</thead>")]
+    assert "<strong" not in result
+
+
+def test_to_html_columns_arg(float_frame):
+    result = float_frame.to_html(columns=["A"])
+    assert "<th>B</th>" not in result
+
+
+@pytest.mark.parametrize(
+    "columns,justify,expected",
+    [
+        (
+            MultiIndex.from_arrays(
+                [np.arange(2).repeat(2), np.mod(range(4), 2)],
+                names=["CL0", "CL1"],
+            ),
+            "left",
+            "multiindex_1",
+        ),
+        (
+            MultiIndex.from_arrays([np.arange(4), np.mod(range(4), 2)]),
+            "right",
+            "multiindex_2",
+        ),
+    ],
+)
+def test_to_html_multiindex(columns, justify, expected, datapath):
+    df = DataFrame([list("abcd"), list("efgh")], columns=columns)
+    result = df.to_html(justify=justify)
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+def test_to_html_justify(justify, datapath):
+    df = DataFrame(
+        {"A": [6, 30000, 2], "B": [1, 2, 70000], "C": [223442, 0, 1]},
+        columns=["A", "B", "C"],
+    )
+    result = df.to_html(justify=justify)
+    expected = expected_html(datapath, "justify").format(justify=justify)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "justify", ["super-right", "small-left", "noinherit", "tiny", "pandas"]
+)
+def test_to_html_invalid_justify(justify):
+    # GH 17527
+    df = DataFrame()
+    msg = "Invalid value for justify parameter"
+
+    with pytest.raises(ValueError, match=msg):
+        df.to_html(justify=justify)
+
+
+class TestHTMLIndex:
+    @pytest.fixture
+    def df(self):
+        index = ["foo", "bar", "baz"]
+        df = DataFrame(
+            {"A": [1, 2, 3], "B": [1.2, 3.4, 5.6], "C": ["one", "two", np.nan]},
+            columns=["A", "B", "C"],
+            index=index,
+        )
+        return df
+
+    @pytest.fixture
+    def expected_without_index(self, datapath):
+        return expected_html(datapath, "index_2")
+
+    def test_to_html_flat_index_without_name(
+        self, datapath, df, expected_without_index
+    ):
+        expected_with_index = expected_html(datapath, "index_1")
+        assert df.to_html() == expected_with_index
+
+        result = df.to_html(index=False)
+        for i in df.index:
+            assert i not in result
+        assert result == expected_without_index
+
+    def test_to_html_flat_index_with_name(self, datapath, df, expected_without_index):
+        df.index = Index(["foo", "bar", "baz"], name="idx")
+        expected_with_index = expected_html(datapath, "index_3")
+        assert df.to_html() == expected_with_index
+        assert df.to_html(index=False) == expected_without_index
+
+    def test_to_html_multiindex_without_names(
+        self, datapath, df, expected_without_index
+    ):
+        tuples = [("foo", "car"), ("foo", "bike"), ("bar", "car")]
+        df.index = MultiIndex.from_tuples(tuples)
+
+        expected_with_index = expected_html(datapath, "index_4")
+        assert df.to_html() == expected_with_index
+
+        result = df.to_html(index=False)
+        for i in ["foo", "bar", "car", "bike"]:
+            assert i not in result
+        # must be the same result as normal index
+        assert result == expected_without_index
+
+    def test_to_html_multiindex_with_names(self, datapath, df, expected_without_index):
+        tuples = [("foo", "car"), ("foo", "bike"), ("bar", "car")]
+        df.index = MultiIndex.from_tuples(tuples, names=["idx1", "idx2"])
+        expected_with_index = expected_html(datapath, "index_5")
+        assert df.to_html() == expected_with_index
+        assert df.to_html(index=False) == expected_without_index
+
+
+@pytest.mark.parametrize("classes", ["sortable draggable", ["sortable", "draggable"]])
+def test_to_html_with_classes(classes, datapath):
+    df = DataFrame()
+    expected = expected_html(datapath, "with_classes")
+    result = df.to_html(classes=classes)
+    assert result == expected
+
+
+def test_to_html_no_index_max_rows(datapath):
+    # GH 14998
+    df = DataFrame({"A": [1, 2, 3, 4]})
+    result = df.to_html(index=False, max_rows=1)
+    expected = expected_html(datapath, "gh14998_expected_output")
+    assert result == expected
+
+
+def test_to_html_multiindex_max_cols(datapath):
+    # GH 6131
+    index = MultiIndex(
+        levels=[["ba", "bb", "bc"], ["ca", "cb", "cc"]],
+        codes=[[0, 1, 2], [0, 1, 2]],
+        names=["b", "c"],
+    )
+    columns = MultiIndex(
+        levels=[["d"], ["aa", "ab", "ac"]],
+        codes=[[0, 0, 0], [0, 1, 2]],
+        names=[None, "a"],
+    )
+    data = np.array(
+        [[1.0, np.nan, np.nan], [np.nan, 2.0, np.nan], [np.nan, np.nan, 3.0]]
+    )
+    df = DataFrame(data, index, columns)
+    result = df.to_html(max_cols=2)
+    expected = expected_html(datapath, "gh6131_expected_output")
+    assert result == expected
+
+
+def test_to_html_multi_indexes_index_false(datapath):
+    # GH 22579
+    df = DataFrame(
+        {"a": range(10), "b": range(10, 20), "c": range(10, 20), "d": range(10, 20)}
+    )
+    df.columns = MultiIndex.from_product([["a", "b"], ["c", "d"]])
+    df.index = MultiIndex.from_product([["a", "b"], ["c", "d", "e", "f", "g"]])
+    result = df.to_html(index=False)
+    expected = expected_html(datapath, "gh22579_expected_output")
+    assert result == expected
+
+
+@pytest.mark.parametrize("index_names", [True, False])
+@pytest.mark.parametrize("header", [True, False])
+@pytest.mark.parametrize("index", [True, False])
+@pytest.mark.parametrize(
+    "column_index, column_type",
+    [
+        (Index([0, 1]), "unnamed_standard"),
+        (Index([0, 1], name="columns.name"), "named_standard"),
+        (MultiIndex.from_product([["a"], ["b", "c"]]), "unnamed_multi"),
+        (
+            MultiIndex.from_product(
+                [["a"], ["b", "c"]], names=["columns.name.0", "columns.name.1"]
+            ),
+            "named_multi",
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    "row_index, row_type",
+    [
+        (Index([0, 1]), "unnamed_standard"),
+        (Index([0, 1], name="index.name"), "named_standard"),
+        (MultiIndex.from_product([["a"], ["b", "c"]]), "unnamed_multi"),
+        (
+            MultiIndex.from_product(
+                [["a"], ["b", "c"]], names=["index.name.0", "index.name.1"]
+            ),
+            "named_multi",
+        ),
+    ],
+)
+def test_to_html_basic_alignment(
+    datapath, row_index, row_type, column_index, column_type, index, header, index_names
+):
+    # GH 22747, GH 22579
+    df = DataFrame(np.zeros((2, 2), dtype=int), index=row_index, columns=column_index)
+    result = df.to_html(index=index, header=header, index_names=index_names)
+
+    if not index:
+        row_type = "none"
+    elif not index_names and row_type.startswith("named"):
+        row_type = "un" + row_type
+
+    if not header:
+        column_type = "none"
+    elif not index_names and column_type.startswith("named"):
+        column_type = "un" + column_type
+
+    filename = "index_" + row_type + "_columns_" + column_type
+    expected = expected_html(datapath, filename)
+    assert result == expected
+
+
+@pytest.mark.parametrize("index_names", [True, False])
+@pytest.mark.parametrize("header", [True, False])
+@pytest.mark.parametrize("index", [True, False])
+@pytest.mark.parametrize(
+    "column_index, column_type",
+    [
+        (Index(np.arange(8)), "unnamed_standard"),
+        (Index(np.arange(8), name="columns.name"), "named_standard"),
+        (
+            MultiIndex.from_product([["a", "b"], ["c", "d"], ["e", "f"]]),
+            "unnamed_multi",
+        ),
+        (
+            MultiIndex.from_product(
+                [["a", "b"], ["c", "d"], ["e", "f"]], names=["foo", None, "baz"]
+            ),
+            "named_multi",
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    "row_index, row_type",
+    [
+        (Index(np.arange(8)), "unnamed_standard"),
+        (Index(np.arange(8), name="index.name"), "named_standard"),
+        (
+            MultiIndex.from_product([["a", "b"], ["c", "d"], ["e", "f"]]),
+            "unnamed_multi",
+        ),
+        (
+            MultiIndex.from_product(
+                [["a", "b"], ["c", "d"], ["e", "f"]], names=["foo", None, "baz"]
+            ),
+            "named_multi",
+        ),
+    ],
+)
+def test_to_html_alignment_with_truncation(
+    datapath, row_index, row_type, column_index, column_type, index, header, index_names
+):
+    # GH 22747, GH 22579
+    df = DataFrame(np.arange(64).reshape(8, 8), index=row_index, columns=column_index)
+    result = df.to_html(
+        max_rows=4, max_cols=4, index=index, header=header, index_names=index_names
+    )
+
+    if not index:
+        row_type = "none"
+    elif not index_names and row_type.startswith("named"):
+        row_type = "un" + row_type
+
+    if not header:
+        column_type = "none"
+    elif not index_names and column_type.startswith("named"):
+        column_type = "un" + column_type
+
+    filename = "trunc_df_index_" + row_type + "_columns_" + column_type
+    expected = expected_html(datapath, filename)
+    assert result == expected
+
+
+@pytest.mark.parametrize("index", [False, 0])
+def test_to_html_truncation_index_false_max_rows(datapath, index):
+    # GH 15019
+    data = [
+        [1.764052, 0.400157],
+        [0.978738, 2.240893],
+        [1.867558, -0.977278],
+        [0.950088, -0.151357],
+        [-0.103219, 0.410599],
+    ]
+    df = DataFrame(data)
+    result = df.to_html(max_rows=4, index=index)
+    expected = expected_html(datapath, "gh15019_expected_output")
+    assert result == expected
+
+
+@pytest.mark.parametrize("index", [False, 0])
+@pytest.mark.parametrize(
+    "col_index_named, expected_output",
+    [(False, "gh22783_expected_output"), (True, "gh22783_named_columns_index")],
+)
+def test_to_html_truncation_index_false_max_cols(
+    datapath, index, col_index_named, expected_output
+):
+    # GH 22783
+    data = [
+        [1.764052, 0.400157, 0.978738, 2.240893, 1.867558],
+        [-0.977278, 0.950088, -0.151357, -0.103219, 0.410599],
+    ]
+    df = DataFrame(data)
+    if col_index_named:
+        df.columns.rename("columns.name", inplace=True)
+    result = df.to_html(max_cols=4, index=index)
+    expected = expected_html(datapath, expected_output)
+    assert result == expected
+
+
+@pytest.mark.parametrize("notebook", [True, False])
+def test_to_html_notebook_has_style(notebook):
+    df = DataFrame({"A": [1, 2, 3]})
+    result = df.to_html(notebook=notebook)
+
+    if notebook:
+        assert "tbody tr th:only-of-type" in result
+        assert "vertical-align: middle;" in result
+        assert "thead th" in result
+    else:
+        assert "tbody tr th:only-of-type" not in result
+        assert "vertical-align: middle;" not in result
+        assert "thead th" not in result
+
+
+def test_to_html_with_index_names_false():
+    # GH 16493
+    df = DataFrame({"A": [1, 2]}, index=Index(["a", "b"], name="myindexname"))
+    result = df.to_html(index_names=False)
+    assert "myindexname" not in result
+
+
+def test_to_html_with_id():
+    # GH 8496
+    df = DataFrame({"A": [1, 2]}, index=Index(["a", "b"], name="myindexname"))
+    result = df.to_html(index_names=False, table_id="TEST_ID")
+    assert ' id="TEST_ID"' in result
+
+
+@pytest.mark.parametrize(
+    "value,float_format,expected",
+    [
+        (0.19999, "%.3f", "gh21625_expected_output"),
+        (100.0, "%.0f", "gh22270_expected_output"),
+    ],
+)
+def test_to_html_float_format_no_fixed_width(value, float_format, expected, datapath):
+    # GH 21625, GH 22270
+    df = DataFrame({"x": [value]})
+    expected = expected_html(datapath, expected)
+    result = df.to_html(float_format=float_format)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "render_links,expected",
+    [(True, "render_links_true"), (False, "render_links_false")],
+)
+def test_to_html_render_links(render_links, expected, datapath):
+    # GH 2679
+    data = [
+        [0, "https://pandas.pydata.org/?q1=a&q2=b", "pydata.org"],
+        [0, "www.pydata.org", "pydata.org"],
+    ]
+    df = DataFrame(data, columns=Index(["foo", "bar", None], dtype=object))
+
+    result = df.to_html(render_links=render_links)
+    expected = expected_html(datapath, expected)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "method,expected",
+    [
+        ("to_html", lambda x: lorem_ipsum),
+        ("_repr_html_", lambda x: lorem_ipsum[: x - 4] + "..."),  # regression case
+    ],
+)
+@pytest.mark.parametrize("max_colwidth", [10, 20, 50, 100])
+def test_ignore_display_max_colwidth(method, expected, max_colwidth):
+    # see gh-17004
+    df = DataFrame([lorem_ipsum])
+    with option_context("display.max_colwidth", max_colwidth):
+        result = getattr(df, method)()
+    expected = expected(max_colwidth)
+    assert expected in result
+
+
+@pytest.mark.parametrize("classes", [True, 0])
+def test_to_html_invalid_classes_type(classes):
+    # GH 25608
+    df = DataFrame()
+    msg = "classes must be a string, list, or tuple"
+
+    with pytest.raises(TypeError, match=msg):
+        df.to_html(classes=classes)
+
+
+def test_to_html_round_column_headers():
+    # GH 17280
+    df = DataFrame([1], columns=[0.55555])
+    with option_context("display.precision", 3):
+        html = df.to_html(notebook=False)
+        notebook = df.to_html(notebook=True)
+    assert "0.55555" in html
+    assert "0.556" in notebook
+
+
+@pytest.mark.parametrize("unit", ["100px", "10%", "5em", 150])
+def test_to_html_with_col_space_units(unit):
+    # GH 25941
+    df = DataFrame(np.random.default_rng(2).random(size=(1, 3)))
+    result = df.to_html(col_space=unit)
+    result = result.split("tbody")[0]
+    hdrs = [x for x in result.split("\n") if re.search(r"<th[>\s]", x)]
+    if isinstance(unit, int):
+        unit = str(unit) + "px"
+    for h in hdrs:
+        expected = f'<th style="min-width: {unit};">'
+        assert expected in h
+
+
+class TestReprHTML:
+    def test_html_repr_min_rows_default(self, datapath):
+        # gh-27991
+
+        # default setting no truncation even if above min_rows
+        df = DataFrame({"a": range(20)})
+        result = df._repr_html_()
+        expected = expected_html(datapath, "html_repr_min_rows_default_no_truncation")
+        assert result == expected
+
+        # default of max_rows 60 triggers truncation if above
+        df = DataFrame({"a": range(61)})
+        result = df._repr_html_()
+        expected = expected_html(datapath, "html_repr_min_rows_default_truncated")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "max_rows,min_rows,expected",
+        [
+            # truncated after first two rows
+            (10, 4, "html_repr_max_rows_10_min_rows_4"),
+            # when set to None, follow value of max_rows
+            (12, None, "html_repr_max_rows_12_min_rows_None"),
+            # when set value higher as max_rows, use the minimum
+            (10, 12, "html_repr_max_rows_10_min_rows_12"),
+            # max_rows of None -> never truncate
+            (None, 12, "html_repr_max_rows_None_min_rows_12"),
+        ],
+    )
+    def test_html_repr_min_rows(self, datapath, max_rows, min_rows, expected):
+        # gh-27991
+
+        df = DataFrame({"a": range(61)})
+        expected = expected_html(datapath, expected)
+        with option_context("display.max_rows", max_rows, "display.min_rows", min_rows):
+            result = df._repr_html_()
+        assert result == expected
+
+    def test_repr_html_ipython_config(self, ip):
+        code = textwrap.dedent(
+            """\
+        from pandas import DataFrame
+        df = DataFrame({"A": [1, 2]})
+        df._repr_html_()
+
+        cfg = get_ipython().config
+        cfg['IPKernelApp']['parent_appname']
+        df._repr_html_()
+        """
+        )
+        result = ip.run_cell(code, silent=True)
+        assert not result.error_in_exec
+
+    def test_info_repr_html(self):
+        max_rows = 60
+        max_cols = 20
+        # Long
+        h, w = max_rows + 1, max_cols - 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        assert r"&lt;class" not in df._repr_html_()
+        with option_context("display.large_repr", "info"):
+            assert r"&lt;class" in df._repr_html_()
+
+        # Wide
+        h, w = max_rows - 1, max_cols + 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        assert "<class" not in df._repr_html_()
+        with option_context(
+            "display.large_repr", "info", "display.max_columns", max_cols
+        ):
+            assert "&lt;class" in df._repr_html_()
+
+    def test_fake_qtconsole_repr_html(self, float_frame):
+        df = float_frame
+
+        def get_ipython():
+            return {"config": {"KernelApp": {"parent_appname": "ipython-qtconsole"}}}
+
+        repstr = df._repr_html_()
+        assert repstr is not None
+
+        with option_context("display.max_rows", 5, "display.max_columns", 2):
+            repstr = df._repr_html_()
+
+        assert "class" in repstr  # info fallback
+
+    def test_repr_html(self, float_frame):
+        df = float_frame
+        df._repr_html_()
+
+        with option_context("display.max_rows", 1, "display.max_columns", 1):
+            df._repr_html_()
+
+        with option_context("display.notebook_repr_html", False):
+            df._repr_html_()
+
+        df = DataFrame([[1, 2], [3, 4]])
+        with option_context("display.show_dimensions", True):
+            assert "2 rows" in df._repr_html_()
+        with option_context("display.show_dimensions", False):
+            assert "2 rows" not in df._repr_html_()
+
+    def test_repr_html_mathjax(self):
+        df = DataFrame([[1, 2], [3, 4]])
+        assert "tex2jax_ignore" not in df._repr_html_()
+        assert "mathjax_ignore" not in df._repr_html_()
+
+        with option_context("display.html.use_mathjax", False):
+            assert "tex2jax_ignore" in df._repr_html_()
+            assert "mathjax_ignore" in df._repr_html_()
+
+    def test_repr_html_wide(self):
+        max_cols = 20
+        df = DataFrame([["a" * 25] * (max_cols - 1)] * 10)
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            assert "..." not in df._repr_html_()
+
+        wide_df = DataFrame([["a" * 25] * (max_cols + 1)] * 10)
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            assert "..." in wide_df._repr_html_()
+
+    def test_repr_html_wide_multiindex_cols(self):
+        max_cols = 20
+
+        mcols = MultiIndex.from_product(
+            [np.arange(max_cols // 2), ["foo", "bar"]], names=["first", "second"]
+        )
+        df = DataFrame([["a" * 25] * len(mcols)] * 10, columns=mcols)
+        reg_repr = df._repr_html_()
+        assert "..." not in reg_repr
+
+        mcols = MultiIndex.from_product(
+            (np.arange(1 + (max_cols // 2)), ["foo", "bar"]), names=["first", "second"]
+        )
+        df = DataFrame([["a" * 25] * len(mcols)] * 10, columns=mcols)
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            assert "..." in df._repr_html_()
+
+    def test_repr_html_long(self):
+        with option_context("display.max_rows", 60):
+            max_rows = get_option("display.max_rows")
+            h = max_rows - 1
+            df = DataFrame({"A": np.arange(1, 1 + h), "B": np.arange(41, 41 + h)})
+            reg_repr = df._repr_html_()
+            assert ".." not in reg_repr
+            assert str(41 + max_rows // 2) in reg_repr
+
+            h = max_rows + 1
+            df = DataFrame({"A": np.arange(1, 1 + h), "B": np.arange(41, 41 + h)})
+            long_repr = df._repr_html_()
+            assert ".." in long_repr
+            assert str(41 + max_rows // 2) not in long_repr
+            assert f"{h} rows " in long_repr
+            assert "2 columns" in long_repr
+
+    def test_repr_html_float(self):
+        with option_context("display.max_rows", 60):
+            max_rows = get_option("display.max_rows")
+            h = max_rows - 1
+            df = DataFrame(
+                {
+                    "idx": np.linspace(-10, 10, h),
+                    "A": np.arange(1, 1 + h),
+                    "B": np.arange(41, 41 + h),
+                }
+            ).set_index("idx")
+            reg_repr = df._repr_html_()
+            assert ".." not in reg_repr
+            assert f"<td>{40 + h}</td>" in reg_repr
+
+            h = max_rows + 1
+            df = DataFrame(
+                {
+                    "idx": np.linspace(-10, 10, h),
+                    "A": np.arange(1, 1 + h),
+                    "B": np.arange(41, 41 + h),
+                }
+            ).set_index("idx")
+            long_repr = df._repr_html_()
+            assert ".." in long_repr
+            assert "<td>31</td>" not in long_repr
+            assert f"{h} rows " in long_repr
+            assert "2 columns" in long_repr
+
+    def test_repr_html_long_multiindex(self):
+        max_rows = 60
+        max_L1 = max_rows // 2
+
+        tuples = list(itertools.product(np.arange(max_L1), ["foo", "bar"]))
+        idx = MultiIndex.from_tuples(tuples, names=["first", "second"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((max_L1 * 2, 2)),
+            index=idx,
+            columns=["A", "B"],
+        )
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            reg_repr = df._repr_html_()
+        assert "..." not in reg_repr
+
+        tuples = list(itertools.product(np.arange(max_L1 + 1), ["foo", "bar"]))
+        idx = MultiIndex.from_tuples(tuples, names=["first", "second"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal(((max_L1 + 1) * 2, 2)),
+            index=idx,
+            columns=["A", "B"],
+        )
+        long_repr = df._repr_html_()
+        assert "..." in long_repr
+
+    def test_repr_html_long_and_wide(self):
+        max_cols = 20
+        max_rows = 60
+
+        h, w = max_rows - 1, max_cols - 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            assert "..." not in df._repr_html_()
+
+        h, w = max_rows + 1, max_cols + 1
+        df = DataFrame({k: np.arange(1, 1 + h) for k in np.arange(w)})
+        with option_context("display.max_rows", 60, "display.max_columns", 20):
+            assert "..." in df._repr_html_()
+
+
+def test_to_html_multilevel(multiindex_year_month_day_dataframe_random_data):
+    ymd = multiindex_year_month_day_dataframe_random_data
+
+    ymd.columns.name = "foo"
+    ymd.to_html()
+    ymd.T.to_html()
+
+
+@pytest.mark.parametrize("na_rep", ["NaN", "Ted"])
+def test_to_html_na_rep_and_float_format(na_rep, datapath):
+    # https://github.com/pandas-dev/pandas/issues/13828
+    df = DataFrame(
+        [
+            ["A", 1.2225],
+            ["A", None],
+        ],
+        columns=["Group", "Data"],
+    )
+    result = df.to_html(na_rep=na_rep, float_format="{:.2f}".format)
+    expected = expected_html(datapath, "gh13828_expected_output")
+    expected = expected.format(na_rep=na_rep)
+    assert result == expected
+
+
+def test_to_html_na_rep_non_scalar_data(datapath):
+    # GH47103
+    df = DataFrame([{"a": 1, "b": [1, 2, 3]}])
+    result = df.to_html(na_rep="-")
+    expected = expected_html(datapath, "gh47103_expected_output")
+    assert result == expected
+
+
+def test_to_html_float_format_object_col(datapath):
+    # GH#40024
+    df = DataFrame(data={"x": [1000.0, "test"]})
+    result = df.to_html(float_format=lambda x: f"{x:,.0f}")
+    expected = expected_html(datapath, "gh40024_expected_output")
+    assert result == expected
+
+
+def test_to_html_multiindex_col_with_colspace():
+    # GH#53885
+    df = DataFrame([[1, 2]])
+    df.columns = MultiIndex.from_tuples([(1, 1), (2, 1)])
+    result = df.to_html(col_space=100)
+    expected = (
+        '<table border="1" class="dataframe">\n'
+        "  <thead>\n"
+        "    <tr>\n"
+        '      <th style="min-width: 100px;"></th>\n'
+        '      <th style="min-width: 100px;">1</th>\n'
+        '      <th style="min-width: 100px;">2</th>\n'
+        "    </tr>\n"
+        "    <tr>\n"
+        '      <th style="min-width: 100px;"></th>\n'
+        '      <th style="min-width: 100px;">1</th>\n'
+        '      <th style="min-width: 100px;">1</th>\n'
+        "    </tr>\n"
+        "  </thead>\n"
+        "  <tbody>\n"
+        "    <tr>\n"
+        "      <th>0</th>\n"
+        "      <td>1</td>\n"
+        "      <td>2</td>\n"
+        "    </tr>\n"
+        "  </tbody>\n"
+        "</table>"
+    )
+    assert result == expected
+
+
+def test_to_html_tuple_col_with_colspace():
+    # GH#53885
+    df = DataFrame({("a", "b"): [1], "b": [2]})
+    result = df.to_html(col_space=100)
+    expected = (
+        '<table border="1" class="dataframe">\n'
+        "  <thead>\n"
+        '    <tr style="text-align: right;">\n'
+        '      <th style="min-width: 100px;"></th>\n'
+        '      <th style="min-width: 100px;">(a, b)</th>\n'
+        '      <th style="min-width: 100px;">b</th>\n'
+        "    </tr>\n"
+        "  </thead>\n"
+        "  <tbody>\n"
+        "    <tr>\n"
+        "      <th>0</th>\n"
+        "      <td>1</td>\n"
+        "      <td>2</td>\n"
+        "    </tr>\n"
+        "  </tbody>\n"
+        "</table>"
+    )
+    assert result == expected
+
+
+def test_to_html_empty_complex_array():
+    # GH#54167
+    df = DataFrame({"x": np.array([], dtype="complex")})
+    result = df.to_html(col_space=100)
+    expected = (
+        '<table border="1" class="dataframe">\n'
+        "  <thead>\n"
+        '    <tr style="text-align: right;">\n'
+        '      <th style="min-width: 100px;"></th>\n'
+        '      <th style="min-width: 100px;">x</th>\n'
+        "    </tr>\n"
+        "  </thead>\n"
+        "  <tbody>\n"
+        "  </tbody>\n"
+        "</table>"
+    )
+    assert result == expected
diff --git a/pandas/pandas/tests/io/formats/test_to_latex.py b/pandas/pandas/tests/io/formats/test_to_latex.py
new file mode 100644
index 0000000000000..8d46442611719
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_latex.py
@@ -0,0 +1,1492 @@
+import codecs
+from datetime import datetime
+from textwrap import dedent
+
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+pytest.importorskip("jinja2")
+
+
+def _dedent(string):
+    """Dedent without new line in the beginning.
+
+    Built-in textwrap.dedent would keep new line character in the beginning
+    of multi-line string starting from the new line.
+    This version drops the leading new line character.
+    """
+    return dedent(string).lstrip()
+
+
+@pytest.fixture
+def df_short():
+    """Short dataframe for testing table/tabular/longtable LaTeX env."""
+    return DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+
+
+class TestToLatex:
+    def test_to_latex_to_file(self, float_frame):
+        with tm.ensure_clean("test.tex") as path:
+            float_frame.to_latex(path)
+            with open(path, encoding="utf-8") as f:
+                assert float_frame.to_latex() == f.read()
+
+    def test_to_latex_to_file_utf8_with_encoding(self):
+        # test with utf-8 and encoding option (GH 7061)
+        df = DataFrame([["au\xdfgangen"]])
+        with tm.ensure_clean("test.tex") as path:
+            df.to_latex(path, encoding="utf-8")
+            with codecs.open(path, "r", encoding="utf-8") as f:
+                assert df.to_latex() == f.read()
+
+    def test_to_latex_to_file_utf8_without_encoding(self):
+        # test with utf-8 without encoding option
+        df = DataFrame([["au\xdfgangen"]])
+        with tm.ensure_clean("test.tex") as path:
+            df.to_latex(path)
+            with codecs.open(path, "r", encoding="utf-8") as f:
+                assert df.to_latex() == f.read()
+
+    def test_to_latex_tabular_with_index(self):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_tabular_without_index(self):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(index=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{rl}
+            \toprule
+            a & b \\
+            \midrule
+            1 & b1 \\
+            2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "bad_column_format",
+        [5, 1.2, ["l", "r"], ("r", "c"), {"r", "c", "l"}, {"a": "r", "b": "l"}],
+    )
+    def test_to_latex_bad_column_format(self, bad_column_format):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        msg = r"`column_format` must be str or unicode"
+        with pytest.raises(ValueError, match=msg):
+            df.to_latex(column_format=bad_column_format)
+
+    def test_to_latex_column_format_just_works(self, float_frame):
+        # GH Bug #9402
+        float_frame.to_latex(column_format="lcr")
+
+    def test_to_latex_column_format(self):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(column_format="lcr")
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lcr}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_float_format_object_col(self):
+        # GH#40024
+        ser = Series([1000.0, "test"])
+        result = ser.to_latex(float_format="{:,.0f}".format)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{ll}
+            \toprule
+             & 0 \\
+            \midrule
+            0 & 1,000 \\
+            1 & test \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_empty_tabular(self):
+        df = DataFrame()
+        result = df.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{l}
+            \toprule
+            \midrule
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_series(self):
+        s = Series(["a", "b", "c"])
+        result = s.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{ll}
+            \toprule
+             & 0 \\
+            \midrule
+            0 & a \\
+            1 & b \\
+            2 & c \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_midrule_location(self):
+        # GH 18326
+        df = DataFrame({"a": [1, 2]})
+        df.index.name = "foo"
+        result = df.to_latex(index_names=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lr}
+            \toprule
+             & a \\
+            \midrule
+            0 & 1 \\
+            1 & 2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexLongtable:
+    def test_to_latex_empty_longtable(self):
+        df = DataFrame()
+        result = df.to_latex(longtable=True)
+        expected = _dedent(
+            r"""
+            \begin{longtable}{l}
+            \toprule
+            \midrule
+            \endfirsthead
+            \toprule
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{0}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_with_index(self):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(longtable=True)
+        expected = _dedent(
+            r"""
+            \begin{longtable}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            \endfirsthead
+            \toprule
+             & a & b \\
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{3}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_without_index(self):
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(index=False, longtable=True)
+        expected = _dedent(
+            r"""
+            \begin{longtable}{rl}
+            \toprule
+            a & b \\
+            \midrule
+            \endfirsthead
+            \toprule
+            a & b \\
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{2}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            1 & b1 \\
+            2 & b2 \\
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "df_data, expected_number",
+        [
+            ({"a": [1, 2]}, 1),
+            ({"a": [1, 2], "b": [3, 4]}, 2),
+            ({"a": [1, 2], "b": [3, 4], "c": [5, 6]}, 3),
+        ],
+    )
+    def test_to_latex_longtable_continued_on_next_page(self, df_data, expected_number):
+        df = DataFrame(df_data)
+        result = df.to_latex(index=False, longtable=True)
+        assert rf"\multicolumn{{{expected_number}}}" in result
+
+
+class TestToLatexHeader:
+    def test_to_latex_no_header_with_index(self):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(header=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_no_header_without_index(self):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(index=False, header=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{rl}
+            \toprule
+            \midrule
+            1 & b1 \\
+            2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_specified_header_with_index(self):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(header=["AA", "BB"])
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & AA & BB \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_specified_header_without_index(self):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(header=["AA", "BB"], index=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{rl}
+            \toprule
+            AA & BB \\
+            \midrule
+            1 & b1 \\
+            2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "header, num_aliases",
+        [
+            (["A"], 1),
+            (("B",), 1),
+            (("Col1", "Col2", "Col3"), 3),
+            (("Col1", "Col2", "Col3", "Col4"), 4),
+        ],
+    )
+    def test_to_latex_number_of_items_in_header_missmatch_raises(
+        self,
+        header,
+        num_aliases,
+    ):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        msg = f"Writing 2 cols but got {num_aliases} aliases"
+        with pytest.raises(ValueError, match=msg):
+            df.to_latex(header=header)
+
+    def test_to_latex_decimal(self):
+        # GH 12031
+        df = DataFrame({"a": [1.0, 2.1], "b": ["b1", "b2"]})
+        result = df.to_latex(decimal=",")
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1,000000 & b1 \\
+            1 & 2,100000 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexBold:
+    def test_to_latex_bold_rows(self):
+        # GH 16707
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(bold_rows=True)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            \textbf{0} & 1 & b1 \\
+            \textbf{1} & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_no_bold_rows(self):
+        # GH 16707
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(bold_rows=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexCaptionLabel:
+    @pytest.fixture
+    def caption_table(self):
+        """Caption for table/tabular LaTeX environment."""
+        return "a table in a \\texttt{table/tabular} environment"
+
+    @pytest.fixture
+    def short_caption(self):
+        """Short caption for testing \\caption[short_caption]{full_caption}."""
+        return "a table"
+
+    @pytest.fixture
+    def label_table(self):
+        """Label for table/tabular LaTeX environment."""
+        return "tab:table_tabular"
+
+    @pytest.fixture
+    def caption_longtable(self):
+        """Caption for longtable LaTeX environment."""
+        return "a table in a \\texttt{longtable} environment"
+
+    @pytest.fixture
+    def label_longtable(self):
+        """Label for longtable LaTeX environment."""
+        return "tab:longtable"
+
+    def test_to_latex_caption_only(self, df_short, caption_table):
+        # GH 25436
+        result = df_short.to_latex(caption=caption_table)
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \caption{a table in a \texttt{table/tabular} environment}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_label_only(self, df_short, label_table):
+        # GH 25436
+        result = df_short.to_latex(label=label_table)
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \label{tab:table_tabular}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_caption_and_label(self, df_short, caption_table, label_table):
+        # GH 25436
+        result = df_short.to_latex(caption=caption_table, label=label_table)
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \caption{a table in a \texttt{table/tabular} environment}
+            \label{tab:table_tabular}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_caption_and_shortcaption(
+        self,
+        df_short,
+        caption_table,
+        short_caption,
+    ):
+        result = df_short.to_latex(caption=(caption_table, short_caption))
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \caption[a table]{a table in a \texttt{table/tabular} environment}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_caption_and_shortcaption_list_is_ok(self, df_short):
+        caption = ("Long-long-caption", "Short")
+        result_tuple = df_short.to_latex(caption=caption)
+        result_list = df_short.to_latex(caption=list(caption))
+        assert result_tuple == result_list
+
+    def test_to_latex_caption_shortcaption_and_label(
+        self,
+        df_short,
+        caption_table,
+        short_caption,
+        label_table,
+    ):
+        # test when the short_caption is provided alongside caption and label
+        result = df_short.to_latex(
+            caption=(caption_table, short_caption),
+            label=label_table,
+        )
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \caption[a table]{a table in a \texttt{table/tabular} environment}
+            \label{tab:table_tabular}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "bad_caption",
+        [
+            ("full_caption", "short_caption", "extra_string"),
+            ("full_caption", "short_caption", 1),
+            ("full_caption", "short_caption", None),
+            ("full_caption",),
+            (None,),
+        ],
+    )
+    def test_to_latex_bad_caption_raises(self, bad_caption):
+        # test that wrong number of params is raised
+        df = DataFrame({"a": [1]})
+        msg = "`caption` must be either a string or 2-tuple of strings"
+        with pytest.raises(ValueError, match=msg):
+            df.to_latex(caption=bad_caption)
+
+    def test_to_latex_two_chars_caption(self, df_short):
+        # test that two chars caption is handled correctly
+        # it must not be unpacked into long_caption, short_caption.
+        result = df_short.to_latex(caption="xy")
+        expected = _dedent(
+            r"""
+            \begin{table}
+            \caption{xy}
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_caption_only(self, df_short, caption_longtable):
+        # GH 25436
+        # test when no caption and no label is provided
+        # is performed by test_to_latex_longtable()
+        result = df_short.to_latex(longtable=True, caption=caption_longtable)
+        expected = _dedent(
+            r"""
+            \begin{longtable}{lrl}
+            \caption{a table in a \texttt{longtable} environment} \\
+            \toprule
+             & a & b \\
+            \midrule
+            \endfirsthead
+            \caption[]{a table in a \texttt{longtable} environment} \\
+            \toprule
+             & a & b \\
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{3}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_label_only(self, df_short, label_longtable):
+        # GH 25436
+        result = df_short.to_latex(longtable=True, label=label_longtable)
+        expected = _dedent(
+            r"""
+            \begin{longtable}{lrl}
+            \label{tab:longtable} \\
+            \toprule
+             & a & b \\
+            \midrule
+            \endfirsthead
+            \toprule
+             & a & b \\
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{3}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_caption_and_label(
+        self,
+        df_short,
+        caption_longtable,
+        label_longtable,
+    ):
+        # GH 25436
+        result = df_short.to_latex(
+            longtable=True,
+            caption=caption_longtable,
+            label=label_longtable,
+        )
+        expected = _dedent(
+            r"""
+        \begin{longtable}{lrl}
+        \caption{a table in a \texttt{longtable} environment} \label{tab:longtable} \\
+        \toprule
+         & a & b \\
+        \midrule
+        \endfirsthead
+        \caption[]{a table in a \texttt{longtable} environment} \\
+        \toprule
+         & a & b \\
+        \midrule
+        \endhead
+        \midrule
+        \multicolumn{3}{r}{Continued on next page} \\
+        \midrule
+        \endfoot
+        \bottomrule
+        \endlastfoot
+        0 & 1 & b1 \\
+        1 & 2 & b2 \\
+        \end{longtable}
+        """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_caption_shortcaption_and_label(
+        self,
+        df_short,
+        caption_longtable,
+        short_caption,
+        label_longtable,
+    ):
+        # test when the caption, the short_caption and the label are provided
+        result = df_short.to_latex(
+            longtable=True,
+            caption=(caption_longtable, short_caption),
+            label=label_longtable,
+        )
+        expected = _dedent(
+            r"""
+\begin{longtable}{lrl}
+\caption[a table]{a table in a \texttt{longtable} environment} \label{tab:longtable} \\
+\toprule
+ & a & b \\
+\midrule
+\endfirsthead
+\caption[]{a table in a \texttt{longtable} environment} \\
+\toprule
+ & a & b \\
+\midrule
+\endhead
+\midrule
+\multicolumn{3}{r}{Continued on next page} \\
+\midrule
+\endfoot
+\bottomrule
+\endlastfoot
+0 & 1 & b1 \\
+1 & 2 & b2 \\
+\end{longtable}
+"""
+        )
+        assert result == expected
+
+
+class TestToLatexEscape:
+    @pytest.fixture
+    def df_with_symbols(self):
+        """Dataframe with special characters for testing chars escaping."""
+        a = "a"
+        b = "b"
+        return DataFrame({"co$e^x$": {a: "a", b: "b"}, "co^l1": {a: "a", b: "b"}})
+
+    def test_to_latex_escape_false(self, df_with_symbols):
+        result = df_with_symbols.to_latex(escape=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lll}
+            \toprule
+             & co$e^x$ & co^l1 \\
+            \midrule
+            a & a & a \\
+            b & b & b \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_escape_default(self, df_with_symbols):
+        # gh50871: in v2.0 escape is False by default (styler.format.escape=None)
+        default = df_with_symbols.to_latex()
+        specified_true = df_with_symbols.to_latex(escape=True)
+        assert default != specified_true
+
+    def test_to_latex_special_escape(self):
+        df = DataFrame([r"a\b\c", r"^a^b^c", r"~a~b~c"])
+        result = df.to_latex(escape=True)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{ll}
+            \toprule
+             & 0 \\
+            \midrule
+            0 & a\textbackslash b\textbackslash c \\
+            1 & \textasciicircum a\textasciicircum b\textasciicircum c \\
+            2 & \textasciitilde a\textasciitilde b\textasciitilde c \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_escape_special_chars(self):
+        special_characters = ["&", "%", "$", "#", "_", "{", "}", "~", "^", "\\"]
+        df = DataFrame(data=special_characters)
+        result = df.to_latex(escape=True)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{ll}
+            \toprule
+             & 0 \\
+            \midrule
+            0 & \& \\
+            1 & \% \\
+            2 & \$ \\
+            3 & \# \\
+            4 & \_ \\
+            5 & \{ \\
+            6 & \} \\
+            7 & \textasciitilde  \\
+            8 & \textasciicircum  \\
+            9 & \textbackslash  \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_specified_header_special_chars_without_escape(self):
+        # GH 7124
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(header=["$A$", "$B$"], escape=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrl}
+            \toprule
+             & $A$ & $B$ \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexPosition:
+    def test_to_latex_position(self):
+        the_position = "h"
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(position=the_position)
+        expected = _dedent(
+            r"""
+            \begin{table}[h]
+            \begin{tabular}{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \bottomrule
+            \end{tabular}
+            \end{table}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_longtable_position(self):
+        the_position = "t"
+        df = DataFrame({"a": [1, 2], "b": ["b1", "b2"]})
+        result = df.to_latex(longtable=True, position=the_position)
+        expected = _dedent(
+            r"""
+            \begin{longtable}[t]{lrl}
+            \toprule
+             & a & b \\
+            \midrule
+            \endfirsthead
+            \toprule
+             & a & b \\
+            \midrule
+            \endhead
+            \midrule
+            \multicolumn{3}{r}{Continued on next page} \\
+            \midrule
+            \endfoot
+            \bottomrule
+            \endlastfoot
+            0 & 1 & b1 \\
+            1 & 2 & b2 \\
+            \end{longtable}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexFormatters:
+    def test_to_latex_with_formatters(self):
+        df = DataFrame(
+            {
+                "datetime64": [
+                    datetime(2016, 1, 1),
+                    datetime(2016, 2, 5),
+                    datetime(2016, 3, 3),
+                ],
+                "float": [1.0, 2.0, 3.0],
+                "int": [1, 2, 3],
+                "object": [(1, 2), True, False],
+            }
+        )
+
+        formatters = {
+            "datetime64": lambda x: x.strftime("%Y-%m"),
+            "float": lambda x: f"[{x: 4.1f}]",
+            "int": lambda x: f"0x{x:x}",
+            "object": lambda x: f"-{x!s}-",
+            "__index__": lambda x: f"index: {x}",
+        }
+        result = df.to_latex(formatters=dict(formatters))
+
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llrrl}
+            \toprule
+             & datetime64 & float & int & object \\
+            \midrule
+            index: 0 & 2016-01 & [ 1.0] & 0x1 & -(1, 2)- \\
+            index: 1 & 2016-02 & [ 2.0] & 0x2 & -True- \\
+            index: 2 & 2016-03 & [ 3.0] & 0x3 & -False- \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_float_format_no_fixed_width_3decimals(self):
+        # GH 21625
+        df = DataFrame({"x": [0.19999]})
+        result = df.to_latex(float_format="%.3f")
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lr}
+            \toprule
+             & x \\
+            \midrule
+            0 & 0.200 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_float_format_no_fixed_width_integer(self):
+        # GH 22270
+        df = DataFrame({"x": [100.0]})
+        result = df.to_latex(float_format="%.0f")
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lr}
+            \toprule
+             & x \\
+            \midrule
+            0 & 100 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize("na_rep", ["NaN", "Ted"])
+    def test_to_latex_na_rep_and_float_format(self, na_rep):
+        df = DataFrame(
+            [
+                ["A", 1.2225],
+                ["A", None],
+            ],
+            columns=["Group", "Data"],
+        )
+        result = df.to_latex(na_rep=na_rep, float_format="{:.2f}".format)
+        expected = _dedent(
+            rf"""
+            \begin{{tabular}}{{llr}}
+            \toprule
+             & Group & Data \\
+            \midrule
+            0 & A & 1.22 \\
+            1 & A & {na_rep} \\
+            \bottomrule
+            \end{{tabular}}
+            """
+        )
+        assert result == expected
+
+
+class TestToLatexMultiindex:
+    @pytest.fixture
+    def multiindex_frame(self):
+        """Multiindex dataframe for testing multirow LaTeX macros."""
+        return DataFrame.from_dict(
+            {
+                ("c1", 0): Series({x: x for x in range(4)}),
+                ("c1", 1): Series({x: x + 4 for x in range(4)}),
+                ("c2", 0): Series({x: x for x in range(4)}),
+                ("c2", 1): Series({x: x + 4 for x in range(4)}),
+                ("c3", 0): Series({x: x for x in range(4)}),
+            }
+        ).T
+
+    @pytest.fixture
+    def multicolumn_frame(self):
+        """Multicolumn dataframe for testing multicolumn LaTeX macros."""
+        return DataFrame(
+            {
+                ("c1", 0): {x: x for x in range(5)},
+                ("c1", 1): {x: x + 5 for x in range(5)},
+                ("c2", 0): {x: x for x in range(5)},
+                ("c2", 1): {x: x + 5 for x in range(5)},
+                ("c3", 0): {x: x for x in range(5)},
+            }
+        )
+
+    def test_to_latex_multindex_header(self):
+        # GH 16718
+        df = DataFrame({"a": [0], "b": [1], "c": [2], "d": [3]})
+        df = df.set_index(["a", "b"])
+        observed = df.to_latex(header=["r1", "r2"], multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llrr}
+            \toprule
+             &  & r1 & r2 \\
+            a & b &  &  \\
+            \midrule
+            0 & 1 & 2 & 3 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert observed == expected
+
+    def test_to_latex_multiindex_empty_name(self):
+        # GH 18669
+        mi = pd.MultiIndex.from_product([[1, 2]], names=[""])
+        df = DataFrame(-1, index=mi, columns=range(4))
+        observed = df.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrrrr}
+            \toprule
+             & 0 & 1 & 2 & 3 \\
+             &  &  &  &  \\
+            \midrule
+            1 & -1 & -1 & -1 & -1 \\
+            2 & -1 & -1 & -1 & -1 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert observed == expected
+
+    def test_to_latex_multiindex_column_tabular(self):
+        df = DataFrame({("x", "y"): ["a"]})
+        result = df.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{ll}
+            \toprule
+             & x \\
+             & y \\
+            \midrule
+            0 & a \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multiindex_small_tabular(self):
+        df = DataFrame({("x", "y"): ["a"]}).T
+        result = df.to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lll}
+            \toprule
+             &  & 0 \\
+            \midrule
+            x & y & a \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multiindex_tabular(self, multiindex_frame):
+        result = multiindex_frame.to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llrrrr}
+            \toprule
+             &  & 0 & 1 & 2 & 3 \\
+            \midrule
+            c1 & 0 & 0 & 1 & 2 & 3 \\
+             & 1 & 4 & 5 & 6 & 7 \\
+            c2 & 0 & 0 & 1 & 2 & 3 \\
+             & 1 & 4 & 5 & 6 & 7 \\
+            c3 & 0 & 0 & 1 & 2 & 3 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multicolumn_tabular(self, multiindex_frame):
+        # GH 14184
+        df = multiindex_frame.T
+        df.columns.names = ["a", "b"]
+        result = df.to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrrrrr}
+            \toprule
+            a & \multicolumn{2}{r}{c1} & \multicolumn{2}{r}{c2} & c3 \\
+            b & 0 & 1 & 0 & 1 & 0 \\
+            \midrule
+            0 & 0 & 4 & 0 & 4 & 0 \\
+            1 & 1 & 5 & 1 & 5 & 1 \\
+            2 & 2 & 6 & 2 & 6 & 2 \\
+            3 & 3 & 7 & 3 & 7 & 3 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_index_has_name_tabular(self):
+        # GH 10660
+        df = DataFrame({"a": [0, 0, 1, 1], "b": list("abab"), "c": [1, 2, 3, 4]})
+        result = df.set_index(["a", "b"]).to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llr}
+            \toprule
+             &  & c \\
+            a & b &  \\
+            \midrule
+            0 & a & 1 \\
+             & b & 2 \\
+            1 & a & 3 \\
+             & b & 4 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_groupby_tabular(self):
+        # GH 10660
+        df = DataFrame({"a": [0, 0, 1, 1], "b": list("abab"), "c": [1, 2, 3, 4]})
+        result = (
+            df.groupby("a")
+            .describe()
+            .to_latex(float_format="{:.1f}".format, escape=True)
+        )
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrrrrrrrr}
+            \toprule
+             & \multicolumn{8}{r}{c} \\
+             & count & mean & std & min & 25\% & 50\% & 75\% & max \\
+            a &  &  &  &  &  &  &  &  \\
+            \midrule
+            0 & 2.0 & 1.5 & 0.7 & 1.0 & 1.2 & 1.5 & 1.8 & 2.0 \\
+            1 & 2.0 & 3.5 & 0.7 & 3.0 & 3.2 & 3.5 & 3.8 & 4.0 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multiindex_dupe_level(self):
+        # see gh-14484
+        #
+        # If an index is repeated in subsequent rows, it should be
+        # replaced with a blank in the created table. This should
+        # ONLY happen if all higher order indices (to the left) are
+        # equal too. In this test, 'c' has to be printed both times
+        # because the higher order index 'A' != 'B'.
+        df = DataFrame(
+            index=pd.MultiIndex.from_tuples([("A", "c"), ("B", "c")]), columns=["col"]
+        )
+        result = df.to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lll}
+            \toprule
+             &  & col \\
+            \midrule
+            A & c & NaN \\
+            B & c & NaN \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multicolumn_default(self, multicolumn_frame):
+        result = multicolumn_frame.to_latex()
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrrrrr}
+            \toprule
+             & \multicolumn{2}{r}{c1} & \multicolumn{2}{r}{c2} & c3 \\
+             & 0 & 1 & 0 & 1 & 0 \\
+            \midrule
+            0 & 0 & 5 & 0 & 5 & 0 \\
+            1 & 1 & 6 & 1 & 6 & 1 \\
+            2 & 2 & 7 & 2 & 7 & 2 \\
+            3 & 3 & 8 & 3 & 8 & 3 \\
+            4 & 4 & 9 & 4 & 9 & 4 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multicolumn_false(self, multicolumn_frame):
+        result = multicolumn_frame.to_latex(multicolumn=False, multicolumn_format="l")
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lrrrrr}
+            \toprule
+             & c1 & & c2 & & c3 \\
+             & 0 & 1 & 0 & 1 & 0 \\
+            \midrule
+            0 & 0 & 5 & 0 & 5 & 0 \\
+            1 & 1 & 6 & 1 & 6 & 1 \\
+            2 & 2 & 7 & 2 & 7 & 2 \\
+            3 & 3 & 8 & 3 & 8 & 3 \\
+            4 & 4 & 9 & 4 & 9 & 4 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multirow_true(self, multicolumn_frame):
+        result = multicolumn_frame.T.to_latex(multirow=True)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llrrrrr}
+            \toprule
+             &  & 0 & 1 & 2 & 3 & 4 \\
+            \midrule
+            \multirow[t]{2}{*}{c1} & 0 & 0 & 1 & 2 & 3 & 4 \\
+             & 1 & 5 & 6 & 7 & 8 & 9 \\
+            \cline{1-7}
+            \multirow[t]{2}{*}{c2} & 0 & 0 & 1 & 2 & 3 & 4 \\
+             & 1 & 5 & 6 & 7 & 8 & 9 \\
+            \cline{1-7}
+            c3 & 0 & 0 & 1 & 2 & 3 & 4 \\
+            \cline{1-7}
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multicolumnrow_with_multicol_format(self, multicolumn_frame):
+        multicolumn_frame.index = multicolumn_frame.T.index
+        result = multicolumn_frame.T.to_latex(
+            multirow=True,
+            multicolumn=True,
+            multicolumn_format="c",
+        )
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llrrrrr}
+            \toprule
+             &  & \multicolumn{2}{c}{c1} & \multicolumn{2}{c}{c2} & c3 \\
+             &  & 0 & 1 & 0 & 1 & 0 \\
+            \midrule
+            \multirow[t]{2}{*}{c1} & 0 & 0 & 1 & 2 & 3 & 4 \\
+             & 1 & 5 & 6 & 7 & 8 & 9 \\
+            \cline{1-7}
+            \multirow[t]{2}{*}{c2} & 0 & 0 & 1 & 2 & 3 & 4 \\
+             & 1 & 5 & 6 & 7 & 8 & 9 \\
+            \cline{1-7}
+            c3 & 0 & 0 & 1 & 2 & 3 & 4 \\
+            \cline{1-7}
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize("name0", [None, "named0"])
+    @pytest.mark.parametrize("name1", [None, "named1"])
+    @pytest.mark.parametrize("axes", [[0], [1], [0, 1]])
+    def test_to_latex_multiindex_names(self, name0, name1, axes):
+        # GH 18667
+        names = [name0, name1]
+        mi = pd.MultiIndex.from_product([[1, 2], [3, 4]])
+        df = DataFrame(-1, index=mi.copy(), columns=mi.copy())
+        for idx in axes:
+            df.axes[idx].names = names
+
+        idx_names = tuple(n or "" for n in names)
+        idx_names_row = (
+            f"{idx_names[0]} & {idx_names[1]} &  &  &  &  \\\\\n"
+            if (0 in axes and any(names))
+            else ""
+        )
+        col_names = [n if (bool(n) and 1 in axes) else "" for n in names]
+        observed = df.to_latex(multirow=False)
+        expected = r"""\begin{tabular}{llrrrr}
+\toprule
+ & %s & \multicolumn{2}{r}{1} & \multicolumn{2}{r}{2} \\
+ & %s & 3 & 4 & 3 & 4 \\
+%s\midrule
+1 & 3 & -1 & -1 & -1 & -1 \\
+ & 4 & -1 & -1 & -1 & -1 \\
+2 & 3 & -1 & -1 & -1 & -1 \\
+ & 4 & -1 & -1 & -1 & -1 \\
+\bottomrule
+\end{tabular}
+""" % tuple(list(col_names) + [idx_names_row])
+        assert observed == expected
+
+    @pytest.mark.parametrize("one_row", [True, False])
+    def test_to_latex_multiindex_nans(self, one_row):
+        # GH 14249
+        df = DataFrame({"a": [None, 1], "b": [2, 3], "c": [4, 5]})
+        if one_row:
+            df = df.iloc[[0]]
+        observed = df.set_index(["a", "b"]).to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llr}
+            \toprule
+             &  & c \\
+            a & b &  \\
+            \midrule
+            NaN & 2 & 4 \\
+            """
+        )
+        if not one_row:
+            expected += r"""1.000000 & 3 & 5 \\
+"""
+        expected += r"""\bottomrule
+\end{tabular}
+"""
+        assert observed == expected
+
+    def test_to_latex_non_string_index(self):
+        # GH 19981
+        df = DataFrame([[1, 2, 3]] * 2).set_index([0, 1])
+        result = df.to_latex(multirow=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{llr}
+            \toprule
+             &  & 2 \\
+            0 & 1 &  \\
+            \midrule
+            1 & 2 & 3 \\
+             & 2 & 3 \\
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multiindex_multirow(self):
+        # GH 16719
+        mi = pd.MultiIndex.from_product(
+            [[0.0, 1.0], [3.0, 2.0, 1.0], ["0", "1"]], names=["i", "val0", "val1"]
+        )
+        df = DataFrame(index=mi)
+        result = df.to_latex(multirow=True, escape=False)
+        expected = _dedent(
+            r"""
+            \begin{tabular}{lll}
+            \toprule
+            i & val0 & val1 \\
+            \midrule
+            \multirow[t]{6}{*}{0.000000} & \multirow[t]{2}{*}{3.000000} & 0 \\
+             &  & 1 \\
+            \cline{2-3}
+             & \multirow[t]{2}{*}{2.000000} & 0 \\
+             &  & 1 \\
+            \cline{2-3}
+             & \multirow[t]{2}{*}{1.000000} & 0 \\
+             &  & 1 \\
+            \cline{1-3} \cline{2-3}
+            \multirow[t]{6}{*}{1.000000} & \multirow[t]{2}{*}{3.000000} & 0 \\
+             &  & 1 \\
+            \cline{2-3}
+             & \multirow[t]{2}{*}{2.000000} & 0 \\
+             &  & 1 \\
+            \cline{2-3}
+             & \multirow[t]{2}{*}{1.000000} & 0 \\
+             &  & 1 \\
+            \cline{1-3} \cline{2-3}
+            \bottomrule
+            \end{tabular}
+            """
+        )
+        assert result == expected
+
+    def test_to_latex_multiindex_format_single_index_hidden(self):
+        # GH 52218
+        df = DataFrame(
+            {
+                "A": [1, 2],
+                "B": [4, 5],
+            }
+        )
+        result = (
+            df.style.hide(axis="index")
+            .map_index(lambda v: "textbf:--rwrap;", axis="columns")
+            .to_latex()
+        )
+        expected = _dedent(r"""
+            \begin{tabular}{rr}
+            \textbf{A} & \textbf{B} \\
+            1 & 4 \\
+            2 & 5 \\
+            \end{tabular}
+            """)
+        assert result == expected
+
+    def test_to_latex_multiindex_format_triple_index_two_hidden(self):
+        # GH 52218
+        arrays = [
+            ["A", "A", "B", "B"],
+            ["one", "two", "one", "two"],
+            ["x", "x", "y", "y"],
+        ]
+        index = pd.MultiIndex.from_arrays(
+            arrays, names=["Level 0", "Level 1", "Level 2"]
+        )
+        df = DataFrame(
+            [[0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0]],
+            index=index,
+            columns=["C1", "C2", "C3"],
+        )
+        result = (
+            df.style.hide(axis="index", level=[0, 1])
+            .map_index(lambda v: "textbf:--rwrap;", axis="columns")
+            .to_latex()
+        )
+        expected = _dedent(r"""
+            \begin{tabular}{lrrr}
+             & \textbf{C1} & \textbf{C2} & \textbf{C3} \\
+            Level 2 &  &  &  \\
+            x & 0 & 0 & 0 \\
+            x & 0 & 0 & 0 \\
+            y & 0 & 0 & 0 \\
+            y & 0 & 0 & 0 \\
+            \end{tabular}
+            """)
+        assert result == expected
+
+    def test_to_latex_multiindex_format_triple_index_all_hidden(self):
+        # GH 52218
+        arrays = [
+            ["A", "A", "B", "B"],
+            ["one", "two", "one", "two"],
+            ["x", "x", "y", "y"],
+        ]
+        index = pd.MultiIndex.from_arrays(
+            arrays, names=["Level 0", "Level 1", "Level 2"]
+        )
+        df = DataFrame(
+            [[0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0]],
+            index=index,
+            columns=["C1", "C2", "C3"],
+        )
+        result = (
+            df.style.hide(axis="index", level=[0, 1, 2])
+            .map_index(lambda v: "textbf:--rwrap;", axis="columns")
+            .to_latex()
+        )
+        expected = _dedent(r"""
+            \begin{tabular}{rrr}
+            \textbf{C1} & \textbf{C2} & \textbf{C3} \\
+            0 & 0 & 0 \\
+            0 & 0 & 0 \\
+            0 & 0 & 0 \\
+            0 & 0 & 0 \\
+            \end{tabular}
+            """)
+        assert result == expected
diff --git a/pandas/pandas/tests/io/formats/test_to_markdown.py b/pandas/pandas/tests/io/formats/test_to_markdown.py
new file mode 100644
index 0000000000000..f3d9b88cc91e2
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_markdown.py
@@ -0,0 +1,100 @@
+from io import StringIO
+
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+pytest.importorskip("tabulate")
+
+
+def test_keyword_deprecation():
+    # GH 57280
+    msg = (
+        "Starting with pandas version 4.0 all arguments of to_markdown "
+        "except for the argument 'buf' will be keyword-only."
+    )
+    s = pd.Series()
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        s.to_markdown(None, "wt")
+
+
+def test_simple():
+    buf = StringIO()
+    df = pd.DataFrame([1, 2, 3])
+    df.to_markdown(buf=buf)
+    result = buf.getvalue()
+    assert (
+        result == "|    |   0 |\n|---:|----:|\n|  0 |   1 |\n|  1 |   2 |\n|  2 |   3 |"
+    )
+
+
+def test_empty_frame():
+    buf = StringIO()
+    df = pd.DataFrame({"id": [], "first_name": [], "last_name": []}).set_index("id")
+    df.to_markdown(buf=buf)
+    result = buf.getvalue()
+    assert result == (
+        "| id   | first_name   | last_name   |\n|------|--------------|-------------|"
+    )
+
+
+def test_other_tablefmt():
+    buf = StringIO()
+    df = pd.DataFrame([1, 2, 3])
+    df.to_markdown(buf=buf, tablefmt="jira")
+    result = buf.getvalue()
+    assert result == "||    ||   0 ||\n|  0 |   1 |\n|  1 |   2 |\n|  2 |   3 |"
+
+
+def test_other_headers():
+    buf = StringIO()
+    df = pd.DataFrame([1, 2, 3])
+    df.to_markdown(buf=buf, headers=["foo", "bar"])
+    result = buf.getvalue()
+    assert result == (
+        "|   foo |   bar |\n|------:|------:|\n|     0 "
+        "|     1 |\n|     1 |     2 |\n|     2 |     3 |"
+    )
+
+
+def test_series():
+    buf = StringIO()
+    s = pd.Series([1, 2, 3], name="foo")
+    s.to_markdown(buf=buf)
+    result = buf.getvalue()
+    assert result == (
+        "|    |   foo |\n|---:|------:|\n|  0 |     1 |\n|  1 |     2 |\n|  2 |     3 |"
+    )
+
+
+def test_no_buf():
+    df = pd.DataFrame([1, 2, 3])
+    result = df.to_markdown()
+    assert (
+        result == "|    |   0 |\n|---:|----:|\n|  0 |   1 |\n|  1 |   2 |\n|  2 |   3 |"
+    )
+
+
+@pytest.mark.parametrize("index", [True, False])
+def test_index(index):
+    # GH 32667
+
+    df = pd.DataFrame([1, 2, 3])
+
+    result = df.to_markdown(index=index)
+
+    if index:
+        expected = (
+            "|    |   0 |\n|---:|----:|\n|  0 |   1 |\n|  1 |   2 |\n|  2 |   3 |"
+        )
+    else:
+        expected = "|   0 |\n|----:|\n|   1 |\n|   2 |\n|   3 |"
+    assert result == expected
+
+
+def test_showindex_disallowed_in_kwargs():
+    # GH 32667; disallowing showindex in kwargs enforced in 2.0
+    df = pd.DataFrame([1, 2, 3])
+    with pytest.raises(ValueError, match="Pass 'index' instead of 'showindex"):
+        df.to_markdown(index=True, showindex=True)
diff --git a/pandas/pandas/tests/io/formats/test_to_string.py b/pandas/pandas/tests/io/formats/test_to_string.py
new file mode 100644
index 0000000000000..63c975fd831e7
--- /dev/null
+++ b/pandas/pandas/tests/io/formats/test_to_string.py
@@ -0,0 +1,1224 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+from io import StringIO
+import re
+import sys
+from textwrap import dedent
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    get_option,
+    option_context,
+    read_csv,
+    timedelta_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+def _three_digit_exp():
+    return f"{1.7e8:.4g}" == "1.7e+008"
+
+
+class TestDataFrameToStringFormatters:
+    def test_keyword_deprecation(self):
+        # GH 57280
+        msg = (
+            "Starting with pandas version 4.0 all arguments of to_string "
+            "except for the argument 'buf' will be keyword-only."
+        )
+        s = Series(["a", "b"])
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            s.to_string(None, "NaN")
+
+    def test_to_string_masked_ea_with_formatter(self):
+        # GH#39336
+        df = DataFrame(
+            {
+                "a": Series([0.123456789, 1.123456789], dtype="Float64"),
+                "b": Series([1, 2], dtype="Int64"),
+            }
+        )
+        result = df.to_string(formatters=["{:.2f}".format, "{:.2f}".format])
+        expected = dedent(
+            """\
+                  a     b
+            0  0.12  1.00
+            1  1.12  2.00"""
+        )
+        assert result == expected
+
+    def test_to_string_with_formatters(self):
+        df = DataFrame(
+            {
+                "int": [1, 2, 3],
+                "float": [1.0, 2.0, 3.0],
+                "object": [(1, 2), True, False],
+            },
+            columns=["int", "float", "object"],
+        )
+
+        formatters = [
+            ("int", lambda x: f"0x{x:x}"),
+            ("float", lambda x: f"[{x: 4.1f}]"),
+            ("object", lambda x: f"-{x!s}-"),
+        ]
+        result = df.to_string(formatters=dict(formatters))
+        result2 = df.to_string(formatters=list(zip(*formatters))[1])
+        assert result == (
+            "  int  float    object\n"
+            "0 0x1 [ 1.0]  -(1, 2)-\n"
+            "1 0x2 [ 2.0]    -True-\n"
+            "2 0x3 [ 3.0]   -False-"
+        )
+        assert result == result2
+
+    def test_to_string_with_datetime64_monthformatter(self):
+        months = [datetime(2016, 1, 1), datetime(2016, 2, 2)]
+        x = DataFrame({"months": months})
+
+        def format_func(x):
+            return x.strftime("%Y-%m")
+
+        result = x.to_string(formatters={"months": format_func})
+        expected = dedent(
+            """\
+            months
+            0 2016-01
+            1 2016-02"""
+        )
+        assert result.strip() == expected
+
+    def test_to_string_with_datetime64_hourformatter(self):
+        x = DataFrame(
+            {"hod": to_datetime(["10:10:10.100", "12:12:12.120"], format="%H:%M:%S.%f")}
+        )
+
+        def format_func(x):
+            return x.strftime("%H:%M")
+
+        result = x.to_string(formatters={"hod": format_func})
+        expected = dedent(
+            """\
+            hod
+            0 10:10
+            1 12:12"""
+        )
+        assert result.strip() == expected
+
+    def test_to_string_with_formatters_unicode(self):
+        df = DataFrame({"c/\u03c3": [1, 2, 3]})
+        result = df.to_string(formatters={"c/\u03c3": str})
+        expected = dedent(
+            """\
+              c/\u03c3
+            0   1
+            1   2
+            2   3"""
+        )
+        assert result == expected
+
+    def test_to_string_index_formatter(self):
+        df = DataFrame([range(5), range(5, 10), range(10, 15)])
+        rs = df.to_string(formatters={"__index__": lambda x: "abc"[x]})
+        xp = dedent(
+            """\
+            0   1   2   3   4
+        a   0   1   2   3   4
+        b   5   6   7   8   9
+        c  10  11  12  13  14"""
+        )
+        assert rs == xp
+
+    def test_no_extra_space(self):
+        # GH#52690: Check that no extra space is given
+        col1 = "TEST"
+        col2 = "PANDAS"
+        col3 = "to_string"
+        expected = f"{col1:<6s} {col2:<7s} {col3:<10s}"
+        df = DataFrame([{"col1": "TEST", "col2": "PANDAS", "col3": "to_string"}])
+        d = {"col1": "{:<6s}".format, "col2": "{:<7s}".format, "col3": "{:<10s}".format}
+        result = df.to_string(index=False, header=False, formatters=d)
+        assert result == expected
+
+
+class TestDataFrameToStringColSpace:
+    def test_to_string_with_column_specific_col_space_raises(self):
+        df = DataFrame(
+            np.random.default_rng(2).random(size=(3, 3)), columns=["a", "b", "c"]
+        )
+
+        msg = (
+            "Col_space length\\(\\d+\\) should match "
+            "DataFrame number of columns\\(\\d+\\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.to_string(col_space=[30, 40])
+
+        with pytest.raises(ValueError, match=msg):
+            df.to_string(col_space=[30, 40, 50, 60])
+
+        msg = "unknown column"
+        with pytest.raises(ValueError, match=msg):
+            df.to_string(col_space={"a": "foo", "b": 23, "d": 34})
+
+    def test_to_string_with_column_specific_col_space(self):
+        df = DataFrame(
+            np.random.default_rng(2).random(size=(3, 3)), columns=["a", "b", "c"]
+        )
+
+        result = df.to_string(col_space={"a": 10, "b": 11, "c": 12})
+        # 3 separating space + each col_space for (id, a, b, c)
+        assert len(result.split("\n")[1]) == (3 + 1 + 10 + 11 + 12)
+
+        result = df.to_string(col_space=[10, 11, 12])
+        assert len(result.split("\n")[1]) == (3 + 1 + 10 + 11 + 12)
+
+    def test_to_string_with_col_space(self):
+        df = DataFrame(np.random.default_rng(2).random(size=(1, 3)))
+        c10 = len(df.to_string(col_space=10).split("\n")[1])
+        c20 = len(df.to_string(col_space=20).split("\n")[1])
+        c30 = len(df.to_string(col_space=30).split("\n")[1])
+        assert c10 < c20 < c30
+
+        # GH#8230
+        # col_space wasn't being applied with header=False
+        with_header = df.to_string(col_space=20)
+        with_header_row1 = with_header.splitlines()[1]
+        no_header = df.to_string(col_space=20, header=False)
+        assert len(with_header_row1) == len(no_header)
+
+    def test_to_string_repr_tuples(self):
+        buf = StringIO()
+
+        df = DataFrame({"tups": list(zip(range(10), range(10)))})
+        repr(df)
+        df.to_string(col_space=10, buf=buf)
+
+
+class TestDataFrameToStringHeader:
+    def test_to_string_header_false(self):
+        # GH#49230
+        df = DataFrame([1, 2])
+        df.index.name = "a"
+        s = df.to_string(header=False)
+        expected = "a   \n0  1\n1  2"
+        assert s == expected
+
+        df = DataFrame([[1, 2], [3, 4]])
+        df.index.name = "a"
+        s = df.to_string(header=False)
+        expected = "a      \n0  1  2\n1  3  4"
+        assert s == expected
+
+    def test_to_string_multindex_header(self):
+        # GH#16718
+        df = DataFrame({"a": [0], "b": [1], "c": [2], "d": [3]}).set_index(["a", "b"])
+        res = df.to_string(header=["r1", "r2"])
+        exp = "    r1 r2\na b      \n0 1  2  3"
+        assert res == exp
+
+    def test_to_string_no_header(self):
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(header=False)
+        expected = "0  1  4\n1  2  5\n2  3  6"
+
+        assert df_s == expected
+
+    def test_to_string_specified_header(self):
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(header=["X", "Y"])
+        expected = "   X  Y\n0  1  4\n1  2  5\n2  3  6"
+
+        assert df_s == expected
+
+        msg = "Writing 2 cols but got 1 aliases"
+        with pytest.raises(ValueError, match=msg):
+            df.to_string(header=["X"])
+
+
+class TestDataFrameToStringLineWidth:
+    def test_to_string_line_width(self):
+        df = DataFrame(123, index=range(10, 15), columns=range(30))
+        lines = df.to_string(line_width=80)
+        assert max(len(line) for line in lines.split("\n")) == 80
+
+    def test_to_string_line_width_no_index(self):
+        # GH#13998, GH#22505
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, index=False)
+        expected = " x  \\\n 1   \n 2   \n 3   \n\n y  \n 4  \n 5  \n 6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, 33], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, index=False)
+        expected = " x  \\\n11   \n22   \n33   \n\n y  \n 4  \n 5  \n 6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, -33], "y": [4, 5, -6]})
+
+        df_s = df.to_string(line_width=1, index=False)
+        expected = "  x  \\\n 11   \n 22   \n-33   \n\n y  \n 4  \n 5  \n-6  "
+
+        assert df_s == expected
+
+    def test_to_string_line_width_no_header(self):
+        # GH#53054
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, header=False)
+        expected = "0  1  \\\n1  2   \n2  3   \n\n0  4  \n1  5  \n2  6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, 33], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, header=False)
+        expected = "0  11  \\\n1  22   \n2  33   \n\n0  4  \n1  5  \n2  6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, -33], "y": [4, 5, -6]})
+
+        df_s = df.to_string(line_width=1, header=False)
+        expected = "0  11  \\\n1  22   \n2 -33   \n\n0  4  \n1  5  \n2 -6  "
+
+        assert df_s == expected
+
+    def test_to_string_line_width_with_both_index_and_header(self):
+        # GH#53054
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1)
+        expected = (
+            "   x  \\\n0  1   \n1  2   \n2  3   \n\n   y  \n0  4  \n1  5  \n2  6  "
+        )
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, 33], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1)
+        expected = (
+            "    x  \\\n0  11   \n1  22   \n2  33   \n\n   y  \n0  4  \n1  5  \n2  6  "
+        )
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, -33], "y": [4, 5, -6]})
+
+        df_s = df.to_string(line_width=1)
+        expected = (
+            "    x  \\\n0  11   \n1  22   \n2 -33   \n\n   y  \n0  4  \n1  5  \n2 -6  "
+        )
+
+        assert df_s == expected
+
+    def test_to_string_line_width_no_index_no_header(self):
+        # GH#53054
+        df = DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, index=False, header=False)
+        expected = "1  \\\n2   \n3   \n\n4  \n5  \n6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, 33], "y": [4, 5, 6]})
+
+        df_s = df.to_string(line_width=1, index=False, header=False)
+        expected = "11  \\\n22   \n33   \n\n4  \n5  \n6  "
+
+        assert df_s == expected
+
+        df = DataFrame({"x": [11, 22, -33], "y": [4, 5, -6]})
+
+        df_s = df.to_string(line_width=1, index=False, header=False)
+        expected = " 11  \\\n 22   \n-33   \n\n 4  \n 5  \n-6  "
+
+        assert df_s == expected
+
+
+class TestToStringNumericFormatting:
+    def test_to_string_float_format_no_fixed_width(self):
+        # GH#21625
+        df = DataFrame({"x": [0.19999]})
+        expected = "      x\n0 0.200"
+        assert df.to_string(float_format="%.3f") == expected
+
+        # GH#22270
+        df = DataFrame({"x": [100.0]})
+        expected = "    x\n0 100"
+        assert df.to_string(float_format="%.0f") == expected
+
+    def test_to_string_small_float_values(self):
+        df = DataFrame({"a": [1.5, 1e-17, -5.5e-7]})
+
+        result = df.to_string()
+        # sadness per above
+        if _three_digit_exp():
+            expected = (
+                "               a\n0  1.500000e+000\n1  1.000000e-017\n2 -5.500000e-007"
+            )
+        else:
+            expected = (
+                "              a\n0  1.500000e+00\n1  1.000000e-17\n2 -5.500000e-07"
+            )
+        assert result == expected
+
+        # but not all exactly zero
+        df = df * 0
+        result = df.to_string()
+        expected = "   0\n0  0\n1  0\n2 -0"
+        # TODO: assert that these match??
+
+    def test_to_string_complex_float_formatting(self):
+        # GH #25514, 25745
+        with option_context("display.precision", 5):
+            df = DataFrame(
+                {
+                    "x": [
+                        (0.4467846931321966 + 0.0715185102060818j),
+                        (0.2739442392974528 + 0.23515228785438969j),
+                        (0.26974928742135185 + 0.3250604054898979j),
+                        (-1j),
+                    ]
+                }
+            )
+            result = df.to_string()
+            expected = (
+                "                  x\n0  0.44678+0.07152j\n"
+                "1  0.27394+0.23515j\n"
+                "2  0.26975+0.32506j\n"
+                "3 -0.00000-1.00000j"
+            )
+            assert result == expected
+
+    def test_to_string_complex_float_formatting_with_exponents(self):
+        # GH #60393
+        with option_context("display.precision", 6):
+            df = DataFrame(
+                {
+                    "x": [
+                        (1.8816e-09 + 0j),
+                        (1.8816e-09 + 3.39676e-09j),
+                    ]
+                }
+            )
+            result = df.to_string()
+            expected = (
+                "                            x\n0  1.881600e-09+0.000000e+00j\n"
+                "1  1.881600e-09+3.396760e-09j"
+            )
+            assert result == expected
+
+    def test_to_string_format_inf(self):
+        # GH#24861
+        df = DataFrame(
+            {
+                "A": [-np.inf, np.inf, -1, -2.1234, 3, 4],
+                "B": [-np.inf, np.inf, "foo", "foooo", "fooooo", "bar"],
+            }
+        )
+        result = df.to_string()
+
+        expected = (
+            "        A       B\n"
+            "0    -inf    -inf\n"
+            "1     inf     inf\n"
+            "2 -1.0000     foo\n"
+            "3 -2.1234   foooo\n"
+            "4  3.0000  fooooo\n"
+            "5  4.0000     bar"
+        )
+        assert result == expected
+
+        df = DataFrame(
+            {
+                "A": [-np.inf, np.inf, -1.0, -2.0, 3.0, 4.0],
+                "B": [-np.inf, np.inf, "foo", "foooo", "fooooo", "bar"],
+            }
+        )
+        result = df.to_string()
+
+        expected = (
+            "     A       B\n"
+            "0 -inf    -inf\n"
+            "1  inf     inf\n"
+            "2 -1.0     foo\n"
+            "3 -2.0   foooo\n"
+            "4  3.0  fooooo\n"
+            "5  4.0     bar"
+        )
+        assert result == expected
+
+    def test_to_string_int_formatting(self):
+        df = DataFrame({"x": [-15, 20, 25, -35]})
+        assert issubclass(df["x"].dtype.type, np.integer)
+
+        output = df.to_string()
+        expected = "    x\n0 -15\n1  20\n2  25\n3 -35"
+        assert output == expected
+
+    def test_to_string_float_formatting(self):
+        with option_context(
+            "display.precision",
+            5,
+            "display.notebook_repr_html",
+            False,
+        ):
+            df = DataFrame(
+                {"x": [0, 0.25, 3456.000, 12e45, 1.64e6, 1.7e8, 1.253456, np.pi, -1e6]}
+            )
+
+            df_s = df.to_string()
+
+            if _three_digit_exp():
+                expected = (
+                    "              x\n0  0.00000e+000\n1  2.50000e-001\n"
+                    "2  3.45600e+003\n3  1.20000e+046\n4  1.64000e+006\n"
+                    "5  1.70000e+008\n6  1.25346e+000\n7  3.14159e+000\n"
+                    "8 -1.00000e+006"
+                )
+            else:
+                expected = (
+                    "             x\n0  0.00000e+00\n1  2.50000e-01\n"
+                    "2  3.45600e+03\n3  1.20000e+46\n4  1.64000e+06\n"
+                    "5  1.70000e+08\n6  1.25346e+00\n7  3.14159e+00\n"
+                    "8 -1.00000e+06"
+                )
+            assert df_s == expected
+
+            df = DataFrame({"x": [3234, 0.253]})
+            df_s = df.to_string()
+
+            expected = "          x\n0  3234.000\n1     0.253"
+            assert df_s == expected
+
+        assert get_option("display.precision") == 6
+
+        df = DataFrame({"x": [1e9, 0.2512]})
+        df_s = df.to_string()
+
+        if _three_digit_exp():
+            expected = "               x\n0  1.000000e+009\n1  2.512000e-001"
+        else:
+            expected = "              x\n0  1.000000e+09\n1  2.512000e-01"
+        assert df_s == expected
+
+
+class TestDataFrameToString:
+    def test_to_string_decimal(self):
+        # GH#23614
+        df = DataFrame({"A": [6.0, 3.1, 2.2]})
+        expected = "     A\n0  6,0\n1  3,1\n2  2,2"
+        assert df.to_string(decimal=",") == expected
+
+    def test_to_string_left_justify_cols(self):
+        df = DataFrame({"x": [3234, 0.253]})
+        df_s = df.to_string(justify="left")
+        expected = "   x       \n0  3234.000\n1     0.253"
+        assert df_s == expected
+
+    def test_to_string_format_na(self):
+        df = DataFrame(
+            {
+                "A": [np.nan, -1, -2.1234, 3, 4],
+                "B": [np.nan, "foo", "foooo", "fooooo", "bar"],
+            }
+        )
+        result = df.to_string()
+
+        expected = (
+            "        A       B\n"
+            "0     NaN     NaN\n"
+            "1 -1.0000     foo\n"
+            "2 -2.1234   foooo\n"
+            "3  3.0000  fooooo\n"
+            "4  4.0000     bar"
+        )
+        assert result == expected
+
+        df = DataFrame(
+            {
+                "A": [np.nan, -1.0, -2.0, 3.0, 4.0],
+                "B": [np.nan, "foo", "foooo", "fooooo", "bar"],
+            }
+        )
+        result = df.to_string()
+
+        expected = (
+            "     A       B\n"
+            "0  NaN     NaN\n"
+            "1 -1.0     foo\n"
+            "2 -2.0   foooo\n"
+            "3  3.0  fooooo\n"
+            "4  4.0     bar"
+        )
+        assert result == expected
+
+    def test_to_string_with_dict_entries(self):
+        df = DataFrame({"A": [{"a": 1, "b": 2}]})
+
+        val = df.to_string()
+        assert "'a': 1" in val
+        assert "'b': 2" in val
+
+    def test_to_string_with_categorical_columns(self):
+        # GH#35439
+        data = [[4, 2], [3, 2], [4, 3]]
+        cols = ["aaaaaaaaa", "b"]
+        df = DataFrame(data, columns=cols)
+        df_cat_cols = DataFrame(data, columns=CategoricalIndex(cols))
+
+        assert df.to_string() == df_cat_cols.to_string()
+
+    def test_repr_embedded_ndarray(self):
+        arr = np.empty(10, dtype=[("err", object)])
+        for i in range(len(arr)):
+            arr["err"][i] = np.random.default_rng(2).standard_normal(i)
+
+        df = DataFrame(arr)
+        repr(df["err"])
+        repr(df)
+        df.to_string()
+
+    def test_to_string_truncate(self):
+        # GH 9784 - dont truncate when calling DataFrame.to_string
+        df = DataFrame(
+            [
+                {
+                    "a": "foo",
+                    "b": "bar",
+                    "c": "let's make this a very VERY long line that is longer "
+                    "than the default 50 character limit",
+                    "d": 1,
+                },
+                {"a": "foo", "b": "bar", "c": "stuff", "d": 1},
+            ]
+        )
+        df.set_index(["a", "b", "c"])
+        assert df.to_string() == (
+            "     a    b                                         "
+            "                                                c  d\n"
+            "0  foo  bar  let's make this a very VERY long line t"
+            "hat is longer than the default 50 character limit  1\n"
+            "1  foo  bar                                         "
+            "                                            stuff  1"
+        )
+        with option_context("max_colwidth", 20):
+            # the display option has no effect on the to_string method
+            assert df.to_string() == (
+                "     a    b                                         "
+                "                                                c  d\n"
+                "0  foo  bar  let's make this a very VERY long line t"
+                "hat is longer than the default 50 character limit  1\n"
+                "1  foo  bar                                         "
+                "                                            stuff  1"
+            )
+        assert df.to_string(max_colwidth=20) == (
+            "     a    b                    c  d\n"
+            "0  foo  bar  let's make this ...  1\n"
+            "1  foo  bar                stuff  1"
+        )
+
+    @pytest.mark.parametrize(
+        "input_array, expected",
+        [
+            ({"A": ["a"]}, "A\na"),
+            ({"A": ["a", "b"], "B": ["c", "dd"]}, "A  B\na  c\nb dd"),
+            ({"A": ["a", 1], "B": ["aa", 1]}, "A  B\na aa\n1  1"),
+        ],
+    )
+    def test_format_remove_leading_space_dataframe(self, input_array, expected):
+        # GH#24980
+        df = DataFrame(input_array).to_string(index=False)
+        assert df == expected
+
+    @pytest.mark.parametrize(
+        "data,expected",
+        [
+            (
+                {"col1": [1, 2], "col2": [3, 4]},
+                "   col1  col2\n0     1     3\n1     2     4",
+            ),
+            (
+                {"col1": ["Abc", 0.756], "col2": [np.nan, 4.5435]},
+                "    col1    col2\n0    Abc     NaN\n1  0.756  4.5435",
+            ),
+            (
+                {"col1": [np.nan, "a"], "col2": [0.009, 3.543], "col3": ["Abc", 23]},
+                "  col1   col2 col3\n0  NaN  0.009  Abc\n1    a  3.543   23",
+            ),
+        ],
+    )
+    def test_to_string_max_rows_zero(self, data, expected):
+        # GH#35394
+        result = DataFrame(data=data).to_string(max_rows=0)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "max_cols, max_rows, expected",
+        [
+            (
+                10,
+                None,
+                " 0   1   2   3   4   ...  6   7   8   9   10\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0",
+            ),
+            (
+                None,
+                2,
+                " 0   1   2   3   4   5   6   7   8   9   10\n"
+                "  0   0   0   0   0   0   0   0   0   0   0\n"
+                " ..  ..  ..  ..  ..  ..  ..  ..  ..  ..  ..\n"
+                "  0   0   0   0   0   0   0   0   0   0   0",
+            ),
+            (
+                10,
+                2,
+                " 0   1   2   3   4   ...  6   7   8   9   10\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0\n"
+                " ..  ..  ..  ..  ..  ...  ..  ..  ..  ..  ..\n"
+                "  0   0   0   0   0  ...   0   0   0   0   0",
+            ),
+            (
+                9,
+                2,
+                " 0   1   2   3   ...  7   8   9   10\n"
+                "  0   0   0   0  ...   0   0   0   0\n"
+                " ..  ..  ..  ..  ...  ..  ..  ..  ..\n"
+                "  0   0   0   0  ...   0   0   0   0",
+            ),
+            (
+                1,
+                1,
+                " 0  ...\n 0  ...\n..  ...",
+            ),
+        ],
+    )
+    def test_truncation_no_index(self, max_cols, max_rows, expected):
+        df = DataFrame([[0] * 11] * 4)
+        assert (
+            df.to_string(index=False, max_cols=max_cols, max_rows=max_rows) == expected
+        )
+
+    def test_to_string_no_index(self):
+        # GH#16839, GH#13032
+        df = DataFrame({"x": [11, 22], "y": [33, -44], "z": ["AAA", "   "]})
+
+        df_s = df.to_string(index=False)
+        # Leading space is expected for positive numbers.
+        expected = " x   y   z\n11  33 AAA\n22 -44    "
+        assert df_s == expected
+
+        df_s = df[["y", "x", "z"]].to_string(index=False)
+        expected = "  y  x   z\n 33 11 AAA\n-44 22    "
+        assert df_s == expected
+
+    def test_to_string_unicode_columns(self, float_frame):
+        df = DataFrame({"\u03c3": np.arange(10.0)})
+
+        buf = StringIO()
+        df.to_string(buf=buf)
+        buf.getvalue()
+
+        buf = StringIO()
+        df.info(buf=buf)
+        buf.getvalue()
+
+        result = float_frame.to_string()
+        assert isinstance(result, str)
+
+    @pytest.mark.parametrize("na_rep", ["NaN", "Ted"])
+    def test_to_string_na_rep_and_float_format(self, na_rep):
+        # GH#13828
+        df = DataFrame([["A", 1.2225], ["A", None]], columns=["Group", "Data"])
+        result = df.to_string(na_rep=na_rep, float_format="{:.2f}".format)
+        expected = dedent(
+            f"""\
+               Group  Data
+             0     A  1.22
+             1     A   {na_rep}"""
+        )
+        assert result == expected
+
+    def test_to_string_string_dtype(self):
+        # GH#50099
+        pytest.importorskip("pyarrow")
+        df = DataFrame(
+            {"x": ["foo", "bar", "baz"], "y": ["a", "b", "c"], "z": [1, 2, 3]}
+        )
+        df = df.astype(
+            {"x": "string[pyarrow]", "y": "string[python]", "z": "int64[pyarrow]"}
+        )
+        result = df.dtypes.to_string()
+        expected = dedent(
+            """\
+            x    string[pyarrow]
+            y     string[python]
+            z     int64[pyarrow]"""
+        )
+        assert result == expected
+
+    def test_to_string_utf8_columns(self):
+        n = "\u05d0".encode()
+        df = DataFrame([1, 2], columns=[n])
+
+        with option_context("display.max_rows", 1):
+            repr(df)
+
+    def test_to_string_unicode_two(self):
+        dm = DataFrame({"c/\u03c3": []})
+        buf = StringIO()
+        dm.to_string(buf)
+
+    def test_to_string_unicode_three(self):
+        dm = DataFrame(["\xc2"])
+        buf = StringIO()
+        dm.to_string(buf)
+
+    def test_to_string_with_float_index(self):
+        index = Index([1.5, 2, 3, 4, 5])
+        df = DataFrame(np.arange(5), index=index)
+
+        result = df.to_string()
+        expected = "     0\n1.5  0\n2.0  1\n3.0  2\n4.0  3\n5.0  4"
+        assert result == expected
+
+    def test_to_string(self):
+        # big mixed
+        biggie = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(200),
+                "B": Index([f"{i}?!" for i in range(200)]),
+            },
+        )
+
+        biggie.loc[:20, "A"] = np.nan
+        biggie.loc[:20, "B"] = np.nan
+        s = biggie.to_string()
+
+        buf = StringIO()
+        retval = biggie.to_string(buf=buf)
+        assert retval is None
+        assert buf.getvalue() == s
+
+        assert isinstance(s, str)
+
+        # print in right order
+        result = biggie.to_string(
+            columns=["B", "A"], col_space=17, float_format="%.5f".__mod__
+        )
+        lines = result.split("\n")
+        header = lines[0].strip().split()
+        joined = "\n".join([re.sub(r"\s+", " ", x).strip() for x in lines[1:]])
+        recons = read_csv(StringIO(joined), names=header, header=None, sep=" ")
+        tm.assert_series_equal(recons["B"], biggie["B"])
+        assert recons["A"].count() == biggie["A"].count()
+        assert (np.abs(recons["A"].dropna() - biggie["A"].dropna()) < 0.1).all()
+
+        # FIXME: don't leave commented-out
+        # expected = ['B', 'A']
+        # assert header == expected
+
+        result = biggie.to_string(columns=["A"], col_space=17)
+        header = result.split("\n")[0].strip().split()
+        expected = ["A"]
+        assert header == expected
+
+        biggie.to_string(columns=["B", "A"], formatters={"A": lambda x: f"{x:.1f}"})
+
+        biggie.to_string(columns=["B", "A"], float_format=str)
+        biggie.to_string(columns=["B", "A"], col_space=12, float_format=str)
+
+        frame = DataFrame(index=np.arange(200))
+        frame.to_string()
+
+    # TODO: split or simplify this test?
+    @pytest.mark.xfail(using_string_dtype(), reason="fix when arrow is default")
+    def test_to_string_index_with_nan(self):
+        # GH#2850
+        df = DataFrame(
+            {
+                "id1": {0: "1a3", 1: "9h4"},
+                "id2": {0: np.nan, 1: "d67"},
+                "id3": {0: "78d", 1: "79d"},
+                "value": {0: 123, 1: 64},
+            }
+        )
+
+        # multi-index
+        y = df.set_index(["id1", "id2", "id3"])
+        result = y.to_string()
+        expected = (
+            "             value\nid1 id2 id3       \n"
+            "1a3 NaN 78d    123\n9h4 d67 79d     64"
+        )
+        assert result == expected
+
+        # index
+        y = df.set_index("id2")
+        result = y.to_string()
+        expected = (
+            "     id1  id3  value\nid2                 \n"
+            "NaN  1a3  78d    123\nd67  9h4  79d     64"
+        )
+        assert result == expected
+
+        # with append (this failed in 0.12)
+        y = df.set_index(["id1", "id2"]).set_index("id3", append=True)
+        result = y.to_string()
+        expected = (
+            "             value\nid1 id2 id3       \n"
+            "1a3 NaN 78d    123\n9h4 d67 79d     64"
+        )
+        assert result == expected
+
+        # all-nan in mi
+        df2 = df.copy()
+        df2.loc[:, "id2"] = np.nan
+        y = df2.set_index("id2")
+        result = y.to_string()
+        expected = (
+            "     id1  id3  value\nid2                 \n"
+            "NaN  1a3  78d    123\nNaN  9h4  79d     64"
+        )
+        assert result == expected
+
+        # partial nan in mi
+        df2 = df.copy()
+        df2.loc[:, "id2"] = np.nan
+        y = df2.set_index(["id2", "id3"])
+        result = y.to_string()
+        expected = (
+            "         id1  value\nid2 id3            \n"
+            "NaN 78d  1a3    123\n    79d  9h4     64"
+        )
+        assert result == expected
+
+        df = DataFrame(
+            {
+                "id1": {0: np.nan, 1: "9h4"},
+                "id2": {0: np.nan, 1: "d67"},
+                "id3": {0: np.nan, 1: "79d"},
+                "value": {0: 123, 1: 64},
+            }
+        )
+
+        y = df.set_index(["id1", "id2", "id3"])
+        result = y.to_string()
+        expected = (
+            "             value\nid1 id2 id3       \n"
+            "NaN NaN NaN    123\n9h4 d67 79d     64"
+        )
+        assert result == expected
+
+    def test_to_string_nonunicode_nonascii_alignment(self):
+        df = DataFrame([["aa\xc3\xa4\xc3\xa4", 1], ["bbbb", 2]])
+        rep_str = df.to_string()
+        lines = rep_str.split("\n")
+        assert len(lines[1]) == len(lines[2])
+
+    def test_unicode_problem_decoding_as_ascii(self):
+        df = DataFrame({"c/\u03c3": Series({"test": np.nan})})
+        str(df.to_string())
+
+    def test_to_string_repr_unicode(self):
+        buf = StringIO()
+
+        unicode_values = ["\u03c3"] * 10
+        unicode_values = np.array(unicode_values, dtype=object)
+        df = DataFrame({"unicode": unicode_values})
+        df.to_string(col_space=10, buf=buf)
+
+        # it works!
+        repr(df)
+        # it works even if sys.stdin in None
+        _stdin = sys.stdin
+        try:
+            sys.stdin = None
+            repr(df)
+        finally:
+            sys.stdin = _stdin
+
+    def test_nested_dataframe(self):
+        df1 = DataFrame({"level1": [["row1"], ["row2"]]})
+        df2 = DataFrame({"level3": [{"level2": df1}]})
+        result = df2.to_string()
+        expected = "                   level3\n0  {'level2': ['level1']}"
+        assert result == expected
+
+
+class TestSeriesToString:
+    def test_to_string_without_index(self):
+        # GH#11729 Test index=False option
+        ser = Series([1, 2, 3, 4])
+        result = ser.to_string(index=False)
+        expected = "\n".join(["1", "2", "3", "4"])
+        assert result == expected
+
+    def test_to_string_name(self):
+        ser = Series(range(100), dtype="int64")
+        ser.name = "myser"
+        res = ser.to_string(max_rows=2, name=True)
+        exp = "0      0\n      ..\n99    99\nName: myser"
+        assert res == exp
+        res = ser.to_string(max_rows=2, name=False)
+        exp = "0      0\n      ..\n99    99"
+        assert res == exp
+
+    def test_to_string_dtype(self):
+        ser = Series(range(100), dtype="int64")
+        res = ser.to_string(max_rows=2, dtype=True)
+        exp = "0      0\n      ..\n99    99\ndtype: int64"
+        assert res == exp
+        res = ser.to_string(max_rows=2, dtype=False)
+        exp = "0      0\n      ..\n99    99"
+        assert res == exp
+
+    def test_to_string_length(self):
+        ser = Series(range(100), dtype="int64")
+        res = ser.to_string(max_rows=2, length=True)
+        exp = "0      0\n      ..\n99    99\nLength: 100"
+        assert res == exp
+
+    def test_to_string_na_rep(self):
+        ser = Series(index=range(100), dtype=np.float64)
+        res = ser.to_string(na_rep="foo", max_rows=2)
+        exp = "0    foo\n      ..\n99   foo"
+        assert res == exp
+
+    def test_to_string_float_format(self):
+        ser = Series(range(10), dtype="float64")
+        res = ser.to_string(float_format=lambda x: f"{x:2.1f}", max_rows=2)
+        exp = "0   0.0\n     ..\n9   9.0"
+        assert res == exp
+
+    def test_to_string_header(self):
+        ser = Series(range(10), dtype="int64")
+        ser.index.name = "foo"
+        res = ser.to_string(header=True, max_rows=2)
+        exp = "foo\n0    0\n    ..\n9    9"
+        assert res == exp
+        res = ser.to_string(header=False, max_rows=2)
+        exp = "0    0\n    ..\n9    9"
+        assert res == exp
+
+    def test_to_string_empty_col(self):
+        # GH#13653
+        ser = Series(["", "Hello", "World", "", "", "Mooooo", "", ""])
+        res = ser.to_string(index=False)
+        exp = "      \n Hello\n World\n      \n      \nMooooo\n      \n      "
+        assert re.match(exp, res)
+
+    def test_to_string_timedelta64(self):
+        Series(np.array([1100, 20], dtype="timedelta64[ns]")).to_string()
+
+        ser = Series(date_range("2012-1-1", periods=3, freq="D"))
+
+        # GH#2146
+
+        # adding NaTs
+        y = ser - ser.shift(1)
+        result = y.to_string()
+        assert "1 days" in result
+        assert "00:00:00" not in result
+        assert "NaT" in result
+
+        # with frac seconds
+        o = Series([datetime(2012, 1, 1, microsecond=150)] * 3)
+        y = ser - o
+        result = y.to_string()
+        assert "-1 days +23:59:59.999850" in result
+
+        # rounding?
+        o = Series([datetime(2012, 1, 1, 1)] * 3)
+        y = ser - o
+        result = y.to_string()
+        assert "-1 days +23:00:00" in result
+        assert "1 days 23:00:00" in result
+
+        o = Series([datetime(2012, 1, 1, 1, 1)] * 3)
+        y = ser - o
+        result = y.to_string()
+        assert "-1 days +22:59:00" in result
+        assert "1 days 22:59:00" in result
+
+        o = Series([datetime(2012, 1, 1, 1, 1, microsecond=150)] * 3)
+        y = ser - o
+        result = y.to_string()
+        assert "-1 days +22:58:59.999850" in result
+        assert "0 days 22:58:59.999850" in result
+
+        # neg time
+        td = timedelta(minutes=5, seconds=3)
+        s2 = Series(date_range("2012-1-1", periods=3, freq="D")) + td
+        y = ser - s2
+        result = y.to_string()
+        assert "-1 days +23:54:57" in result
+
+        td = timedelta(microseconds=550)
+        s2 = Series(date_range("2012-1-1", periods=3, freq="D")) + td
+        y = ser - td
+        result = y.to_string()
+        assert "2012-01-01 23:59:59.999450" in result
+
+        # no boxing of the actual elements
+        td = Series(timedelta_range("1 days", periods=3))
+        result = td.to_string()
+        assert result == "0   1 days\n1   2 days\n2   3 days"
+
+    def test_to_string(self):
+        ts = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10, freq="B"),
+        )
+        buf = StringIO()
+
+        s = ts.to_string()
+
+        retval = ts.to_string(buf=buf)
+        assert retval is None
+        assert buf.getvalue().strip() == s
+
+        # pass float_format
+        format = "%.4f".__mod__
+        result = ts.to_string(float_format=format)
+        result = [x.split()[1] for x in result.split("\n")[:-1]]
+        expected = [format(x) for x in ts]
+        assert result == expected
+
+        # empty string
+        result = ts[:0].to_string()
+        assert result == "Series([], Freq: B)"
+
+        result = ts[:0].to_string(length=0)
+        assert result == "Series([], Freq: B)"
+
+        # name and length
+        cp = ts.copy()
+        cp.name = "foo"
+        result = cp.to_string(length=True, name=True, dtype=True)
+        last_line = result.split("\n")[-1].strip()
+        assert last_line == (f"Freq: B, Name: foo, Length: {len(cp)}, dtype: float64")
+
+    @pytest.mark.parametrize(
+        "input_array, expected",
+        [
+            ("a", "a"),
+            (["a", "b"], "a\nb"),
+            ([1, "a"], "1\na"),
+            (1, "1"),
+            ([0, -1], " 0\n-1"),
+            (1.0, "1.0"),
+            ([" a", " b"], " a\n b"),
+            ([".1", "1"], ".1\n 1"),
+            (["10", "-10"], " 10\n-10"),
+        ],
+    )
+    def test_format_remove_leading_space_series(self, input_array, expected):
+        # GH: 24980
+        ser = Series(input_array)
+        result = ser.to_string(index=False)
+        assert result == expected
+
+    def test_to_string_complex_number_trims_zeros(self):
+        ser = Series([1.000000 + 1.000000j, 1.0 + 1.0j, 1.05 + 1.0j])
+        result = ser.to_string()
+        expected = dedent(
+            """\
+            0    1.00+1.00j
+            1    1.00+1.00j
+            2    1.05+1.00j"""
+        )
+        assert result == expected
+
+    def test_nullable_float_to_string(self, float_ea_dtype):
+        # https://github.com/pandas-dev/pandas/issues/36775
+        dtype = float_ea_dtype
+        ser = Series([0.0, 1.0, None], dtype=dtype)
+        result = ser.to_string()
+        expected = dedent(
+            """\
+            0     0.0
+            1     1.0
+            2    <NA>"""
+        )
+        assert result == expected
+
+    def test_nullable_int_to_string(self, any_int_ea_dtype):
+        # https://github.com/pandas-dev/pandas/issues/36775
+        dtype = any_int_ea_dtype
+        ser = Series([0, 1, None], dtype=dtype)
+        result = ser.to_string()
+        expected = dedent(
+            """\
+            0       0
+            1       1
+            2    <NA>"""
+        )
+        assert result == expected
+
+    def test_to_string_mixed(self):
+        ser = Series(["foo", np.nan, -1.23, 4.56])
+        result = ser.to_string()
+        expected = "".join(["0     foo\n", "1     NaN\n", "2   -1.23\n", "3    4.56"])
+        assert result == expected
+
+        # but don't count NAs as floats
+        ser = Series(["foo", np.nan, "bar", "baz"])
+        result = ser.to_string()
+        expected = "".join(["0    foo\n", "1    NaN\n", "2    bar\n", "3    baz"])
+        assert result == expected
+
+        ser = Series(["foo", 5, "bar", "baz"])
+        result = ser.to_string()
+        expected = "".join(["0    foo\n", "1      5\n", "2    bar\n", "3    baz"])
+        assert result == expected
+
+    def test_to_string_float_na_spacing(self):
+        ser = Series([0.0, 1.5678, 2.0, -3.0, 4.0])
+        ser[::2] = np.nan
+
+        result = ser.to_string()
+        expected = "0       NaN\n1    1.5678\n2       NaN\n3   -3.0000\n4       NaN"
+        assert result == expected
+
+    def test_to_string_with_datetimeindex(self):
+        index = date_range("20130102", periods=6)
+        ser = Series(1, index=index)
+        result = ser.to_string()
+        assert "2013-01-02" in result
+
+        # nat in index
+        s2 = Series(2, index=[Timestamp("20130111"), NaT])
+        ser = concat([s2, ser])
+        result = ser.to_string()
+        assert "NaT" in result
+
+        # nat in summary
+        result = str(s2.index)
+        assert "NaT" in result
diff --git a/pandas/pandas/tests/io/generate_legacy_storage_files.py b/pandas/pandas/tests/io/generate_legacy_storage_files.py
new file mode 100644
index 0000000000000..9bfd8eb9d51d5
--- /dev/null
+++ b/pandas/pandas/tests/io/generate_legacy_storage_files.py
@@ -0,0 +1,342 @@
+"""
+self-contained to write legacy storage pickle files
+
+To use this script. Create an environment where you want
+generate pickles, say its for 0.20.3, with your pandas clone
+in ~/pandas
+
+. activate pandas_0.20.3
+cd ~/pandas/pandas
+
+$ python -m tests.io.generate_legacy_storage_files \
+    tests/io/data/legacy_pickle/0.20.3/ pickle
+
+This script generates a storage file for the current arch, system,
+and python version
+  pandas version: 0.20.3
+  output dir    : pandas/pandas/tests/io/data/legacy_pickle/0.20.3/
+  storage format: pickle
+created pickle file: 0.20.3_x86_64_darwin_3.5.2.pickle
+
+The idea here is you are using the *current* version of the
+generate_legacy_storage_files with an *older* version of pandas to
+generate a pickle file. We will then check this file into a current
+branch, and test using test_pickle.py. This will load the *older*
+pickles and test versus the current data that is generated
+(with main). These are then compared.
+
+If we have cases where we changed the signature (e.g. we renamed
+offset -> freq in Timestamp). Then we have to conditionally execute
+in the generate_legacy_storage_files.py to make it
+run under the older AND the newer version.
+
+"""
+
+from datetime import timedelta
+import os
+import pickle
+import platform as pl
+import sys
+
+# Remove script directory from path, otherwise Python will try to
+# import the JSON test directory as the json module
+sys.path.pop(0)
+
+import numpy as np
+
+import pandas
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    NaT,
+    Period,
+    RangeIndex,
+    Series,
+    Timestamp,
+    bdate_range,
+    date_range,
+    interval_range,
+    period_range,
+    timedelta_range,
+)
+from pandas.arrays import SparseArray
+
+from pandas.tseries.offsets import (
+    FY5253,
+    BusinessDay,
+    BusinessHour,
+    CustomBusinessDay,
+    DateOffset,
+    Day,
+    Easter,
+    Hour,
+    LastWeekOfMonth,
+    Minute,
+    MonthBegin,
+    MonthEnd,
+    QuarterBegin,
+    QuarterEnd,
+    SemiMonthBegin,
+    SemiMonthEnd,
+    Week,
+    WeekOfMonth,
+    YearBegin,
+    YearEnd,
+)
+
+
+def _create_sp_series():
+    nan = np.nan
+
+    # nan-based
+    arr = np.arange(15, dtype=np.float64)
+    arr[7:12] = nan
+    arr[-1:] = nan
+
+    bseries = Series(SparseArray(arr, kind="block"))
+    bseries.name = "bseries"
+    return bseries
+
+
+def _create_sp_tsseries():
+    nan = np.nan
+
+    # nan-based
+    arr = np.arange(15, dtype=np.float64)
+    arr[7:12] = nan
+    arr[-1:] = nan
+
+    date_index = bdate_range("1/1/2011", periods=len(arr))
+    bseries = Series(SparseArray(arr, kind="block"), index=date_index)
+    bseries.name = "btsseries"
+    return bseries
+
+
+def _create_sp_frame():
+    nan = np.nan
+
+    data = {
+        "A": [nan, nan, nan, 0, 1, 2, 3, 4, 5, 6],
+        "B": [0, 1, 2, nan, nan, nan, 3, 4, 5, 6],
+        "C": np.arange(10).astype(np.int64),
+        "D": [0, 1, 2, 3, 4, 5, nan, nan, nan, nan],
+    }
+
+    dates = bdate_range("1/1/2011", periods=10)
+    return DataFrame(data, index=dates).apply(SparseArray)
+
+
+def create_pickle_data():
+    """create the pickle data"""
+    data = {
+        "A": [0.0, 1.0, 2.0, 3.0, np.nan],
+        "B": [0, 1, 0, 1, 0],
+        "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+        "D": date_range("1/1/2009", periods=5),
+        "E": [0.0, 1, Timestamp("20100101"), "foo", 2.0],
+    }
+
+    scalars = {"timestamp": Timestamp("20130101"), "period": Period("2012", "M")}
+
+    index = {
+        "int": Index(np.arange(10)),
+        "date": date_range("20130101", periods=10),
+        "period": period_range("2013-01-01", freq="M", periods=10),
+        "float": Index(np.arange(10, dtype=np.float64)),
+        "uint": Index(np.arange(10, dtype=np.uint64)),
+        "timedelta": timedelta_range("00:00:00", freq="30min", periods=10),
+    }
+
+    index["range"] = RangeIndex(10)
+
+    index["interval"] = interval_range(0, periods=10)
+
+    mi = {
+        "reg2": MultiIndex.from_tuples(
+            tuple(
+                zip(
+                    *[
+                        ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+                        ["one", "two", "one", "two", "one", "two", "one", "two"],
+                    ]
+                )
+            ),
+            names=["first", "second"],
+        )
+    }
+
+    series = {
+        "float": Series(data["A"]),
+        "int": Series(data["B"]),
+        "mixed": Series(data["E"]),
+        "ts": Series(
+            np.arange(10).astype(np.int64), index=date_range("20130101", periods=10)
+        ),
+        "mi": Series(
+            np.arange(5).astype(np.float64),
+            index=MultiIndex.from_tuples(
+                tuple(zip(*[[1, 1, 2, 2, 2], [3, 4, 3, 4, 5]])), names=["one", "two"]
+            ),
+        ),
+        "dup": Series(np.arange(5).astype(np.float64), index=["A", "B", "C", "D", "A"]),
+        "cat": Series(Categorical(["foo", "bar", "baz"])),
+        "dt": Series(date_range("20130101", periods=5)),
+        "dt_tz": Series(date_range("20130101", periods=5, tz="US/Eastern")),
+        "period": Series([Period("2000Q1")] * 5),
+    }
+
+    mixed_dup_df = DataFrame(data)
+    mixed_dup_df.columns = list("ABCDA")
+    frame = {
+        "float": DataFrame({"A": series["float"], "B": series["float"] + 1}),
+        "int": DataFrame({"A": series["int"], "B": series["int"] + 1}),
+        "mixed": DataFrame({k: data[k] for k in ["A", "B", "C", "D"]}),
+        "mi": DataFrame(
+            {"A": np.arange(5).astype(np.float64), "B": np.arange(5).astype(np.int64)},
+            index=MultiIndex.from_tuples(
+                tuple(
+                    zip(
+                        *[
+                            ["bar", "bar", "baz", "baz", "baz"],
+                            ["one", "two", "one", "two", "three"],
+                        ]
+                    )
+                ),
+                names=["first", "second"],
+            ),
+        ),
+        "dup": DataFrame(
+            np.arange(15).reshape(5, 3).astype(np.float64), columns=["A", "B", "A"]
+        ),
+        "cat_onecol": DataFrame({"A": Categorical(["foo", "bar"])}),
+        "cat_and_float": DataFrame(
+            {
+                "A": Categorical(["foo", "bar", "baz"]),
+                "B": np.arange(3).astype(np.int64),
+            }
+        ),
+        "mixed_dup": mixed_dup_df,
+        "dt_mixed_tzs": DataFrame(
+            {
+                "A": Timestamp("20130102", tz="US/Eastern"),
+                "B": Timestamp("20130603", tz="CET"),
+            },
+            index=range(5),
+        ),
+        "dt_mixed2_tzs": DataFrame(
+            {
+                "A": Timestamp("20130102", tz="US/Eastern"),
+                "B": Timestamp("20130603", tz="CET"),
+                "C": Timestamp("20130603", tz="UTC"),
+            },
+            index=range(5),
+        ),
+    }
+
+    cat = {
+        "int8": Categorical(list("abcdefg")),
+        "int16": Categorical(np.arange(1000)),
+        "int32": Categorical(np.arange(10000)),
+    }
+
+    timestamp = {
+        "normal": Timestamp("2011-01-01"),
+        "nat": NaT,
+        "tz": Timestamp("2011-01-01", tz="US/Eastern"),
+    }
+
+    off = {
+        "DateOffset": DateOffset(years=1),
+        "DateOffset_h_ns": DateOffset(hour=6, nanoseconds=5824),
+        "BusinessDay": BusinessDay(offset=timedelta(seconds=9)),
+        "BusinessHour": BusinessHour(normalize=True, n=6, end="15:14"),
+        "CustomBusinessDay": CustomBusinessDay(weekmask="Mon Fri"),
+        "SemiMonthBegin": SemiMonthBegin(day_of_month=9),
+        "SemiMonthEnd": SemiMonthEnd(day_of_month=24),
+        "MonthBegin": MonthBegin(1),
+        "MonthEnd": MonthEnd(1),
+        "QuarterBegin": QuarterBegin(1),
+        "QuarterEnd": QuarterEnd(1),
+        "Day": Day(1),
+        "YearBegin": YearBegin(1),
+        "YearEnd": YearEnd(1),
+        "Week": Week(1),
+        "Week_Tues": Week(2, normalize=False, weekday=1),
+        "WeekOfMonth": WeekOfMonth(week=3, weekday=4),
+        "LastWeekOfMonth": LastWeekOfMonth(n=1, weekday=3),
+        "FY5253": FY5253(n=2, weekday=6, startingMonth=7, variation="last"),
+        "Easter": Easter(),
+        "Hour": Hour(1),
+        "Minute": Minute(1),
+    }
+
+    return {
+        "series": series,
+        "frame": frame,
+        "index": index,
+        "scalars": scalars,
+        "mi": mi,
+        "sp_series": {"float": _create_sp_series(), "ts": _create_sp_tsseries()},
+        "sp_frame": {"float": _create_sp_frame()},
+        "cat": cat,
+        "timestamp": timestamp,
+        "offsets": off,
+    }
+
+
+def platform_name():
+    return "_".join(
+        [
+            str(pandas.__version__),
+            str(pl.machine()),
+            str(pl.system().lower()),
+            str(pl.python_version()),
+        ]
+    )
+
+
+def write_legacy_pickles(output_dir):
+    version = pandas.__version__
+
+    print(
+        "This script generates a storage file for the current arch, system, "
+        "and python version"
+    )
+    print(f"  pandas version: {version}")
+    print(f"  output dir    : {output_dir}")
+    print("  storage format: pickle")
+
+    pth = f"{platform_name()}.pickle"
+
+    with open(os.path.join(output_dir, pth), "wb") as fh:
+        pickle.dump(create_pickle_data(), fh, pickle.DEFAULT_PROTOCOL)
+
+    print(f"created pickle file: {pth}")
+
+
+def write_legacy_file():
+    # force our cwd to be the first searched
+    sys.path.insert(0, "")
+
+    if not 3 <= len(sys.argv) <= 4:
+        sys.exit(
+            "Specify output directory and storage type: generate_legacy_"
+            "storage_files.py <output_dir> <storage_type> "
+        )
+
+    output_dir = str(sys.argv[1])
+    storage_type = str(sys.argv[2])
+
+    if not os.path.exists(output_dir):
+        os.mkdir(output_dir)
+
+    if storage_type == "pickle":
+        write_legacy_pickles(output_dir=output_dir)
+    else:
+        sys.exit("storage_type must be one of {'pickle'}")
+
+
+if __name__ == "__main__":
+    write_legacy_file()
diff --git a/pandas/pandas/tests/io/json/__init__.py b/pandas/pandas/tests/io/json/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/json/conftest.py b/pandas/pandas/tests/io/json/conftest.py
new file mode 100644
index 0000000000000..4e848cd48b42d
--- /dev/null
+++ b/pandas/pandas/tests/io/json/conftest.py
@@ -0,0 +1,9 @@
+import pytest
+
+
+@pytest.fixture(params=["split", "records", "index", "columns", "values"])
+def orient(request):
+    """
+    Fixture for orients excluding the table format.
+    """
+    return request.param
diff --git a/pandas/pandas/tests/io/json/data/line_delimited.json b/pandas/pandas/tests/io/json/data/line_delimited.json
new file mode 100644
index 0000000000000..be84245329583
--- /dev/null
+++ b/pandas/pandas/tests/io/json/data/line_delimited.json
@@ -0,0 +1,3 @@
+    {"a": 1, "b": 2}
+    {"a": 3, "b": 4}
+    {"a": 5, "b": 6}
diff --git a/pandas/pandas/tests/io/json/data/teams.csv b/pandas/pandas/tests/io/json/data/teams.csv
new file mode 100644
index 0000000000000..fe509fb7d7c31
--- /dev/null
+++ b/pandas/pandas/tests/io/json/data/teams.csv
@@ -0,0 +1,2716 @@
+yearID,lgID,teamID,franchID,divID,Rank,G,Ghome,W,L,DivWin,WCWin,LgWin,WSWin,R,AB,H,2B,3B,HR,BB,SO,SB,CS,HBP,SF,RA,ER,ERA,CG,SHO,SV,IPouts,HA,HRA,BBA,SOA,E,DP,FP,name,park,attendance,BPF,PPF,teamIDBR,teamIDlahman45,teamIDretro
+1871,NA,BS1,BNA,,3,31,,20,10,,,N,,401,1372,426,70,37,3,60,19,73,,,,303,109,3.5500000000,22,1,3,828,367,2,42,23,225,,0.83,Boston Red Stockings,South End Grounds I,,103,98,BOS,BS1,BS1
+1871,NA,CH1,CNA,,2,28,,19,9,,,N,,302,1196,323,52,21,10,60,22,69,,,,241,77,2.7600000000,25,0,1,753,308,6,28,22,218,,0.82,Chicago White Stockings,Union Base-Ball Grounds,,104,102,CHI,CH1,CH1
+1871,NA,CL1,CFC,,8,29,,10,19,,,N,,249,1186,328,35,40,7,26,25,18,,,,341,116,4.1100000000,23,0,0,762,346,13,53,34,223,,0.81,Cleveland Forest Citys,National Association Grounds,,96,100,CLE,CL1,CL1
+1871,NA,FW1,KEK,,7,19,,7,12,,,N,,137,746,178,19,8,2,33,9,16,,,,243,97,5.1700000000,19,1,0,507,261,5,21,17,163,,0.80,Fort Wayne Kekiongas,Hamilton Field,,101,107,KEK,FW1,FW1
+1871,NA,NY2,NNA,,5,33,,16,17,,,N,,302,1404,403,43,21,1,33,15,46,,,,313,121,3.7200000000,32,1,0,879,373,7,42,22,227,,0.83,New York Mutuals,Union Grounds (Brooklyn),,90,88,NYU,NY2,NY2
+1871,NA,PH1,PNA,,1,28,,21,7,,,Y,,376,1281,410,66,27,9,46,23,56,,,,266,137,4.9500000000,27,0,0,747,329,3,53,16,194,,0.84,Philadelphia Athletics,Jefferson Street Grounds,,102,98,ATH,PH1,PH1
+1871,NA,RC1,ROK,,9,25,,4,21,,,N,,231,1036,274,44,25,3,38,30,53,,,,287,108,4.3000000000,23,1,0,678,315,3,34,16,220,,0.82,Rockford Forest Citys,Agricultural Society Fair Grounds,,97,99,ROK,RC1,RC1
+1871,NA,TRO,TRO,,6,29,,13,15,,,N,,351,1248,384,51,34,6,49,19,62,,,,362,153,5.5100000000,28,0,0,750,431,4,75,12,198,,0.84,Troy Haymakers,Haymakers' Grounds,,101,100,TRO,TRO,TRO
+1871,NA,WS3,OLY,,4,32,,15,15,,,N,,310,1353,375,54,26,6,48,13,48,,,,303,137,4.3700000000,32,0,0,846,371,4,45,13,217,,0.85,Washington Olympics,Olympics Grounds,,94,98,OLY,WS3,WS3
+1872,NA,BL1,BLC,,2,58,,35,19,,,N,,617,2576,747,94,35,14,27,28,35,15,,,434,173,3.0200000000,48,1,1,1545,566,3,63,0,432,,0.82,Baltimore Canaries,Newington Park,,106,102,BAL,BL1,BL1
+1872,NA,BR1,ECK,,9,29,,3,26,,,N,,152,1075,235,26,6,0,14,29,8,4,,,413,165,5.7300000000,28,0,0,777,503,6,24,0,278,,0.79,Brooklyn Eckfords,Union Grounds,,87,96,ECK,BR1,BR1
+1872,NA,BR2,BRA,,6,37,,9,28,,,N,,237,1466,370,46,10,0,19,24,17,14,,,473,189,5.0600000000,37,0,0,1008,561,6,19,0,358,,0.81,Brooklyn Atlantics,Capitoline Grounds,,115,122,BRA,BR2,BR2
+1872,NA,BS1,BNA,,1,48,,39,8,,,Y,,521,2137,677,114,31,7,28,26,47,14,,,236,95,1.9900000000,41,3,1,1290,438,0,27,0,263,,0.87,Boston Red Stockings,South End Grounds I,,105,100,BOS,BS1,BS1
+1872,NA,CL1,CFC,,7,22,,6,16,,,N,,174,935,272,37,6,0,17,13,12,3,,,254,101,4.5700000000,15,0,0,597,289,6,24,0,168,,0.82,Cleveland Forest Citys,National Association Grounds,,96,100,CLE,CL1,CL1
+1872,NA,MID,MAN,,8,24,,5,19,,,N,,220,1014,305,31,8,1,5,12,5,3,,,348,140,5.9700000000,22,0,0,633,368,5,13,0,224,,0.80,Middletown Mansfields,Mansfield Club Grounds,,98,103,MAN,MID,MID
+1872,NA,NY2,NNA,,3,56,,34,20,,,N,,523,2423,668,86,13,4,55,52,56,21,,,362,145,2.5500000000,54,3,0,1536,618,2,32,0,326,,0.86,New York Mutuals,Union Grounds (Brooklyn),,93,92,NYU,NY2,NY2
+1872,NA,PH1,PNA,,4,47,,30,14,,,N,,539,2140,678,77,23,4,69,47,58,32,,,349,140,3.0100000000,47,1,0,1257,513,3,26,0,298,,0.85,Philadelphia Athletics,Jefferson Street Grounds,,102,100,ATH,PH1,PH1
+1872,NA,TRO,TRO,,5,25,,15,10,,,N,,273,1124,329,56,11,5,7,14,9,7,,,191,77,3.0800000000,17,2,0,675,282,2,10,0,144,,0.86,Troy Haymakers,Haymakers' Grounds,,101,100,TRO,TRO,TRO
+1872,NA,WS3,OLY,,10,9,,2,7,,,N,,54,363,93,12,3,0,4,4,0,3,,,140,56,6.3800000000,9,0,0,237,147,0,5,0,96,,0.78,Washington Olympics,Olympics Grounds,,94,98,OLY,WS3,WS3
+1872,NA,WS4,NAT,,11,11,,0,11,,,N,,80,460,105,3,2,0,1,3,0,0,,,190,76,6.9100000000,11,0,0,297,194,2,3,0,100,,0.76,Washington Nationals,Nationals Grounds,,129,141,NAT,WS4,WS4
+1873,NA,BL1,BLC,,3,57,,34,22,,,N,,644,2563,810,119,39,12,40,25,21,10,,,451,170,3.0100000000,55,1,0,1524,680,4,42,34,366,,0.85,Baltimore Canaries,Newington Park,,101,102,BAL,BL1,BL1
+1873,NA,BL4,MAR,,9,6,,0,6,,,N,,26,211,33,5,0,0,0,0,0,0,,,152,48,8.0000000000,6,0,0,162,144,4,0,0,70,,0.76,Baltimore Marylands,Newington Park,,77,99,MAR,BL4,BL4
+1873,NA,BR2,BRA,,6,55,,17,37,,,N,,366,2210,588,60,27,6,53,43,18,9,,,549,221,3.9800000000,52,1,0,1500,737,8,42,15,505,,0.82,Brooklyn Atlantics,Union Grounds,,90,94,BRA,BR2,BR2
+1873,NA,BS1,BNA,,1,60,,43,16,,,Y,,739,2755,930,137,46,13,62,24,39,27,,,460,154,2.5900000000,48,1,3,1608,708,5,35,31,434,,0.83,Boston Red Stockings,South End Grounds I,,106,100,BOS,BS1,BS1
+1873,NA,ELI,RES,,8,23,,2,21,,,N,,98,868,204,21,8,0,8,22,2,1,,,299,74,3.2200000000,22,0,0,621,342,7,9,8,247,,0.78,Elizabeth Resolutes,Waverly Fairgrounds,,92,103,RES,ELI,ELI
+1873,NA,NY2,NNA,,4,53,,29,24,,,N,,424,2214,622,67,41,5,42,22,15,5,,,385,139,2.6200000000,48,2,0,1431,539,5,69,76,419,,0.82,New York Mutuals,Union Grounds (Brooklyn),,98,96,NYU,NY2,NY2
+1873,NA,PH1,PNA,,5,52,,28,23,,,N,,474,2266,683,76,22,4,35,32,29,24,,,403,160,3.0300000000,44,3,1,1425,553,4,58,41,383,,0.84,Philadelphia Athletics,Jefferson Street Grounds,,108,105,ATH,PH1,PH1
+1873,NA,PH2,PWS,,2,53,,36,17,,,N,,526,2325,645,83,20,8,62,39,44,14,,,396,148,2.7700000000,50,0,0,1443,627,3,44,28,361,,0.84,Philadelphia Whites,Jefferson Street Grounds,,103,101,PHI,PH2,PH2
+1873,NA,WS5,WBL,,7,39,,8,31,,,N,,283,1563,408,45,25,3,19,33,5,5,,,485,181,4.7100000000,39,0,0,1038,593,11,22,7,240,,0.84,Washington Blue Legs,Olympics Grounds,,96,102,WAS,WS5,WS5
+1874,NA,BL1,BLC,,8,47,,9,38,,,N,,227,1781,435,53,8,1,24,0,,,,,505,201,4.3100000000,43,0,0,1260,643,4,37,0,355,,0.81,Baltimore Canaries,Newington Park,,102,104,BAL,BL1,BL1
+1874,NA,BR2,BRA,,6,56,,22,33,,,N,,301,2165,497,50,11,1,32,0,,,,,449,180,3.2000000000,56,1,0,1518,621,16,13,0,500,,0.82,Brooklyn Atlantics,Union Grounds,,89,94,BRA,BR2,BR2
+1874,NA,BS1,BNA,,1,71,,52,18,,,Y,,735,3122,981,130,57,18,32,0,,,,,415,166,2.3600000000,65,4,3,1899,777,1,27,0,457,,0.85,Boston Red Stockings,South End Grounds I,,105,98,BOS,BS1,BS1
+1874,NA,CH2,CNA,,5,59,,28,31,,,N,,418,2457,690,89,3,3,34,0,,,,,480,192,3.2400000000,58,3,0,1599,696,7,48,0,446,,0.83,Chicago White Stockings,23rd Street Grounds,,101,102,CHI,CH2,CH2
+1874,NA,HR1,HNA,,7,53,,16,37,,,N,,371,2140,587,92,19,2,26,0,,,,,471,189,3.5400000000,45,0,1,1443,660,2,23,0,501,,0.80,Hartford Dark Blues,Hartford Ball Club Grounds,,106,106,HAR,HR1,HR1
+1874,NA,NY2,NNA,,2,65,,42,23,,,N,,500,2719,709,100,31,8,38,0,,,,,377,150,2.3000000000,62,4,0,1758,658,3,40,0,438,,0.84,New York Mutuals,Union Grounds (Brooklyn),,103,102,NYU,NY2,NY2
+1874,NA,PH1,PNA,,3,55,,33,22,,,N,,441,2260,646,87,17,6,24,0,,,,,344,138,2.5500000000,55,0,0,1461,514,5,29,0,396,,0.83,Philadelphia Athletics,Jefferson Street Grounds,,108,105,ATH,PH1,PH1
+1874,NA,PH2,PWS,,4,58,,29,29,,,N,,476,2432,690,74,51,3,28,0,,,,,428,172,2.9500000000,56,3,0,1572,666,4,21,0,518,,0.80,Philadelphia Whites,Jefferson Street Grounds,,103,101,PHI,PH2,PH2
+1875,NA,BR2,BRA,,11,44,,2,42,,,N,,132,1547,304,32,9,2,10,0,,,,,438,174,3.9500000000,31,0,0,1188,531,9,22,0,426,,0.80,Brooklyn Atlantics,Union Grounds,,88,94,BRA,BR2,BR2
+1875,NA,BS1,BNA,,1,82,,71,8,,,Y,,831,3516,1124,157,54,14,31,0,,,,,343,138,1.7000000000,60,10,17,2196,751,2,29,0,464,,0.86,Boston Red Stockings,South End Grounds I,,103,96,BOS,BS1,BS1
+1875,NA,CH2,CNA,,6,69,,30,37,,,N,,379,2658,697,85,13,0,24,0,,,,,416,167,2.4000000000,65,7,0,1875,652,0,28,0,385,,0.86,Chicago White Stockings,23rd Street Grounds,,101,102,CHI,CH2,CH2
+1875,NA,HR1,HNA,,3,86,,54,28,,,N,,557,3342,873,98,34,2,40,0,,,,,343,138,1.6100000000,83,13,0,2313,706,4,16,0,432,,0.88,Hartford Dark Blues,Hartford Ball Club Grounds,,107,105,HAR,HR1,HR1
+1875,NA,KEO,WES,,13,13,,1,12,,,N,,45,449,81,9,6,0,1,0,,,,,88,35,2.8100000000,13,0,0,336,110,0,14,0,68,,0.85,Keokuk Westerns,Perry Park,,105,109,WES,KEO,KEO
+1875,NA,NH1,NHV,,8,47,,7,40,,,N,,170,1708,374,42,12,2,13,0,,,,,397,158,3.3400000000,40,0,0,1278,500,5,22,0,437,,0.81,New Haven Elm Citys,Hamilton Park,,87,93,NHV,NH1,NH1
+1875,NA,NY2,NNA,,6,71,,30,38,,,N,,328,2683,626,84,24,7,21,0,,,,,425,170,2.4000000000,70,2,0,1911,716,3,24,0,526,,0.83,New York Mutuals,Union Grounds (Brooklyn),,99,100,NYU,NY2,NY2
+1875,NA,PH1,PNA,,2,77,,53,20,,,N,,699,3245,940,126,61,10,40,0,,,,,402,161,2.1100000000,75,6,0,2061,775,4,40,0,418,,0.87,Philadelphia Athletics,Jefferson Street Grounds,,107,106,ATH,PH1,PH1
+1875,NA,PH2,PWS,,5,70,,37,31,,,N,,470,2717,681,65,28,5,22,0,,,,,376,150,2.1500000000,64,5,0,1881,650,6,30,0,475,,0.84,Philadelphia Whites,Jefferson Street Grounds,,104,102,PHI,PH2,PH2
+1875,NA,PH3,CEN,,11,14,,2,12,,,N,,70,529,126,19,3,0,8,0,,,,,138,55,3.9300000000,14,0,0,378,170,0,3,0,164,,0.76,Philadelphia Centennials,Centennial Grounds,,92,98,CEN,PH3,PH3
+1875,NA,SL1,SLR,,10,19,,4,15,,,N,,60,688,138,20,1,0,14,0,,,,,161,64,3.3700000000,16,2,0,513,207,0,3,0,150,,0.83,St. Louis Red Stockings,Red Stocking Baseball Park,,90,98,SLR,SL1,SL1
+1875,NA,SL2,SNA,,4,70,,39,29,,,N,,386,2679,641,84,27,0,33,0,,,,,369,148,2.1100000000,67,5,1,1890,634,3,20,0,424,,0.86,St. Louis Brown Stockings,Grand Avenue Park,,89,86,STL,SL2,SL2
+1875,NA,WS6,WNT,,9,28,,5,23,,,N,,107,997,194,12,11,0,5,0,,,,,338,135,4.8600000000,23,0,0,750,397,6,11,0,263,,0.79,Washington Nationals,Olympics Grounds,,96,106,WAS,WS6,WS6
+1876,NL,BSN,ATL,,4,70,,39,31,,,N,,471,2722,723,96,24,9,58,98,,,,,450,176,2.5100000000,49,3,7,1896,732,7,104,77,422,42,0.86,Boston Red Caps,South End Grounds I,,102,98,BSN,BSN,BSN
+1876,NL,CHN,CHC,,1,66,,52,14,,,Y,,624,2748,926,131,32,8,70,45,,,,,257,116,1.7600000000,59,9,4,1776,608,6,29,51,282,33,0.89,Chicago White Stockings,23rd Street Grounds,,112,106,CHC,CHN,CHN
+1876,NL,CN1,CNR,,8,65,,9,56,,,N,,238,2372,555,51,12,4,41,136,,,,,579,238,3.6200000000,57,0,0,1773,850,9,34,60,469,45,0.84,Cincinnati Reds,Avenue Grounds,,86,95,CIN,CN1,CN1
+1876,NL,HAR,HAR,,2,69,,47,21,,,N,,429,2664,711,96,22,2,39,78,,,,,261,116,1.6700000000,69,11,0,1872,570,2,27,114,337,27,0.88,Hartford Dark Blues,Hartford Ball Club Grounds,,110,106,HAR,HAR,HAR
+1876,NL,LS1,LGR,,5,69,,30,36,,,N,,280,2570,641,68,14,6,24,98,,,,,344,121,1.6900000000,67,5,0,1929,605,3,38,125,396,44,0.87,Louisville Grays,Louisville Baseball Park,,117,117,LOU,LS1,LS1
+1876,NL,NY3,NYU,,6,57,,21,35,,,N,,260,2180,494,39,15,2,18,35,,,,,412,173,2.9400000000,56,2,0,1590,718,8,24,37,475,18,0.82,New York Mutuals,Union Grounds (Brooklyn),,96,100,NYU,NY3,NY3
+1876,NL,PHN,ATH,,7,60,,14,45,,,N,,378,2387,646,79,35,7,27,36,,,,,534,197,3.2200000000,53,1,2,1650,783,2,41,22,456,32,0.83,Philadelphia Athletics,Jefferson Street Grounds,,104,104,ATH,PHN,PHN
+1876,NL,SL3,SBS,,3,64,,45,19,,,N,,386,2478,642,73,27,2,59,63,,,,,229,78,1.2200000000,63,16,0,1731,472,3,39,103,268,33,0.90,St. Louis Brown Stockings,Sportsman's Park I,,93,92,STL,SL3,SL3
+1877,NL,BSN,ATL,,1,61,,42,18,,,Y,,419,2368,700,91,37,4,65,121,,,,,263,131,2.1500000000,61,7,0,1644,557,5,38,177,289,36,0.88,Boston Red Caps,South End Grounds I,,104,100,BSN,BSN,BSN
+1877,NL,CHN,CHC,,5,60,,26,33,,,N,,366,2273,633,79,30,0,57,111,,,,,375,200,3.3700000000,45,3,3,1602,630,7,58,92,313,43,0.88,Chicago White Stockings,23rd Street Grounds,,112,106,CHC,CHN,CHN
+1877,NL,CN1,CNR,,6,58,,15,42,,,N,,291,2135,545,72,34,6,78,110,,,,,485,240,4.1900000000,48,1,1,1545,747,4,61,85,393,33,0.85,Cincinnati Reds,Avenue Grounds,,89,94,CIN,CN1,CN1
+1877,NL,HAR,HAR,,3,60,,31,27,,,N,,341,2358,637,63,31,4,30,97,,,,,311,140,2.3200000000,59,4,0,1632,572,2,56,99,313,32,0.88,Hartford Dark Blues,Union Grounds (Brooklyn),,90,87,HAR,HAR,HAR
+1877,NL,LS1,LGR,,2,61,,35,25,,,N,,339,2355,659,75,36,9,58,140,,,,,288,140,2.2500000000,61,4,0,1677,617,4,41,141,267,37,0.90,Louisville Grays,Louisville Baseball Park,,117,117,LOU,LS1,LS1
+1877,NL,SL3,SBS,,4,60,,28,32,,,N,,284,2178,531,51,36,1,57,147,,,,,318,160,2.6600000000,52,1,0,1623,582,2,92,132,281,29,0.89,St. Louis Brown Stockings,Sportsman's Park I,,95,93,STL,SL3,SL3
+1878,NL,BSN,ATL,,1,60,,41,19,,,Y,,298,2220,535,75,25,2,35,154,,,,,241,140,2.3200000000,58,9,0,1632,595,6,38,184,228,48,0.91,Boston Red Caps,South End Grounds I,,108,102,BSN,BSN,BSN
+1878,NL,CHN,CHC,,4,61,,30,30,,,N,,371,2333,677,91,20,3,88,157,,,,,331,145,2.3700000000,61,1,0,1653,577,4,35,175,304,37,0.89,Chicago White Stockings,Lake Front Park I,,106,105,CHC,CHN,CHN
+1878,NL,CN1,CNR,,2,61,,37,23,,,N,,333,2281,629,67,22,5,58,141,,,,,281,112,1.8400000000,61,6,0,1644,546,2,63,220,254,37,0.90,Cincinnati Reds,Avenue Grounds,,91,92,CIN,CN1,CN1
+1878,NL,IN1,IBL,,5,63,,24,36,,,N,,293,2300,542,76,15,3,64,197,,,,,328,149,2.3200000000,59,2,1,1734,621,3,87,182,290,37,0.89,Indianapolis Blues,South Street Park,,87,89,IND,IN1,IN1
+1878,NL,ML2,MLG,,6,61,,15,45,,,N,,256,2212,552,65,20,2,69,214,,,,,386,158,2.6000000000,54,1,0,1641,589,3,55,147,376,32,0.86,Milwaukee Grays,Eclipse Park II,,106,113,MLG,ML2,ML2
+1878,NL,PRO,PRO,,3,62,,33,27,,,N,,353,2298,604,107,30,8,50,218,,,,,337,147,2.3800000000,59,6,0,1668,609,5,86,173,311,42,0.89,Providence Grays,Messer Street Grounds,,100,96,PRO,PRO,PRO
+1879,NL,BFN,BUF,,3,79,,46,32,,,N,,394,2906,733,105,54,2,78,314,,,,,365,185,2.3400000000,78,8,0,2139,698,3,47,198,331,62,0.90,Buffalo Bisons,Riverside Park,,102,105,BUF,BFN,BFN
+1879,NL,BSN,ATL,,2,84,,54,30,,,N,,562,3217,883,138,51,20,90,222,,,,,348,183,2.1900000000,80,13,0,2259,757,9,46,230,319,58,0.91,Boston Red Caps,South End Grounds I,,102,100,BSN,BSN,BSN
+1879,NL,CHN,CHC,,4,83,,46,33,,,N,,437,3116,808,167,32,3,73,294,,,,,411,203,2.4600000000,82,6,0,2232,762,5,57,211,381,52,0.90,Chicago White Stockings,Lake Front Park I,,106,103,CHC,CHN,CHN
+1879,NL,CL2,CBL,,6,82,,27,55,,,N,,322,2987,666,116,29,4,37,214,,,,,461,218,2.6500000000,79,3,0,2223,818,4,116,287,405,42,0.88,Cleveland Blues,Kennard Street Park,,99,100,CLV,CL2,CL2
+1879,NL,CN1,CNR,,5,81,,43,37,,,N,,485,3085,813,127,53,8,66,207,,,,,464,185,2.2900000000,79,4,0,2178,756,11,81,246,426,48,0.87,Cincinnati Reds,Avenue Grounds,,95,94,CIN,CN1,CN1
+1879,NL,PRO,PRO,,1,85,,59,25,,,Y,,612,3392,1003,142,55,12,91,172,,,,,355,188,2.1800000000,73,3,2,2328,765,9,62,329,382,41,0.90,Providence Grays,Messer Street Grounds,,99,95,PRO,PRO,PRO
+1879,NL,SR1,SYR,,7,71,,22,48,,,N,,276,2611,592,61,19,5,28,238,,,,,462,230,3.1900000000,64,5,0,1947,775,4,52,132,398,37,0.87,Syracuse Stars,Newell Park,,89,95,SYR,SR1,SR1
+1879,NL,TRN,TRT,,8,77,,19,56,,,N,,321,2841,673,102,24,4,45,182,,,,,543,216,2.8000000000,75,3,0,2085,840,13,47,210,460,44,0.87,Troy Trojans,Putnam Grounds,,93,100,TRO,TRN,TRN
+1880,NL,BFN,BUF,,7,85,,24,58,,,N,,331,2962,669,104,37,3,90,327,,,,,502,254,3.0900000000,72,6,1,2217,879,10,78,186,403,55,0.89,Buffalo Bisons,Riverside Park,,102,103,BUF,BFN,BFN
+1880,NL,BSN,ATL,,6,86,,40,44,,,N,,416,3080,779,134,41,20,105,221,,,,,456,255,3.0800000000,70,4,0,2232,840,2,86,187,368,54,0.90,Boston Red Caps,South End Grounds I,,97,96,BSN,BSN,BSN
+1880,NL,CHN,CHC,,1,86,,67,17,,,Y,,538,3135,876,164,39,4,104,217,,,,,317,166,1.9300000000,80,8,3,2325,622,8,129,367,329,41,0.91,Chicago White Stockings,Lake Front Park I,,106,102,CHC,CHN,CHN
+1880,NL,CL2,CBL,,3,85,,47,37,,,N,,387,3002,726,130,52,7,76,237,,,,,337,160,1.9000000000,83,7,1,2277,685,4,98,289,330,52,0.91,Cleveland Blues,Kennard Street Park,,98,99,CLV,CL2,CL2
+1880,NL,CN1,CNR,,8,83,,21,59,,,N,,296,2895,649,91,36,7,75,267,,,,,472,193,2.4400000000,79,3,0,2139,785,10,88,208,427,49,0.87,Cincinnati Reds,Bank Street Grounds,,98,104,CIN,CN1,CN1
+1880,NL,PRO,PRO,,2,87,,52,32,,,N,,419,3196,793,114,34,8,89,186,,,,,299,146,1.6400000000,75,13,2,2397,663,7,51,286,357,53,0.91,Providence Grays,Messer Street Grounds,,97,93,PRO,PRO,PRO
+1880,NL,TRN,TRT,,4,83,,41,42,,,N,,392,3007,755,114,37,5,120,260,,,,,438,225,2.7400000000,81,4,0,2214,763,8,113,173,366,58,0.90,Troy Trojans,Haymakers' Grounds,,105,106,TRO,TRN,TRN
+1880,NL,WOR,WOR,,5,85,,40,43,,,N,,412,3024,699,129,52,8,81,278,,,,,370,192,2.2700000000,68,7,5,2286,709,13,97,297,355,49,0.90,Worcester Ruby Legs,Worcester Driving Park Grounds,,108,109,WOR,WOR,WOR
+1881,NL,BFN,BUF,,3,83,,45,38,,,N,,440,3019,797,157,50,12,108,270,,,,,447,234,2.8400000000,72,5,0,2226,881,9,89,185,408,48,0.89,Buffalo Bisons,Riverside Park,,99,100,BUF,BFN,BFN
+1881,NL,BSN,ATL,,6,83,,38,45,,,N,,349,2916,733,121,27,5,110,193,,,,,410,220,2.7100000000,72,6,3,2190,763,9,143,199,293,54,0.91,Boston Red Caps,South End Grounds I,,96,96,BSN,BSN,BSN
+1881,NL,CHN,CHC,,1,84,,56,28,,,Y,,550,3114,918,157,36,12,140,224,,,,,379,201,2.4300000000,81,9,0,2232,722,14,122,228,309,54,0.91,Chicago White Stockings,Lake Front Park I,,105,99,CHC,CHN,CHN
+1881,NL,CL2,CBL,,7,85,,36,48,,,N,,392,3117,796,120,39,7,132,224,,,,,414,226,2.6800000000,82,2,0,2280,737,9,126,240,348,68,0.90,Cleveland Blues,Kennard Street Park,,95,95,CLV,CL2,CL2
+1881,NL,DTN,DTN,,4,84,,41,43,,,N,,439,2995,780,131,53,17,136,250,,,,,429,219,2.6500000000,83,10,0,2232,785,8,137,265,338,80,0.90,Detroit Wolverines,Recreation Park,,104,105,DTN,DTN,DTN
+1881,NL,PRO,PRO,,2,85,,47,37,,,N,,447,3077,780,144,37,11,146,214,,,,,426,202,2.4000000000,76,7,0,2271,756,5,138,264,389,66,0.89,Providence Grays,Messer Street Grounds,,98,96,PRO,PRO,PRO
+1881,NL,TRN,TRT,,5,85,,39,45,,,N,,399,3046,754,124,31,5,140,240,,,,,429,254,2.9700000000,85,8,0,2310,813,11,159,207,311,70,0.91,Troy Trojans,Haymakers' Grounds,,105,106,TRO,TRN,TRN
+1881,NL,WOR,WOR,,8,83,,32,50,,,N,,410,3093,781,114,31,7,121,169,,,,,492,290,3.5400000000,80,5,0,2211,882,11,120,196,353,50,0.90,Worcester Ruby Legs,Worcester Driving Park Grounds,,105,108,WOR,WOR,WOR
+1882,AA,BL2,BLO,,6,74,,19,54,,,N,,273,2583,535,60,24,4,72,215,,,,,515,278,3.8700000000,64,1,0,1938,760,15,108,113,488,41,0.85,Baltimore Orioles,Newington Park,,90,100,BAL,BL2,BL2
+1882,AA,CN2,CIN,,1,80,,55,25,,,Y,,489,3007,795,95,47,5,102,204,,,,,268,132,1.6500000000,77,11,0,2163,609,7,125,165,322,41,0.90,Cincinnati Red Stockings,Bank Street Grounds,,106,99,CIN,CN2,CN2
+1882,AA,LS2,LOU,,2,80,,42,38,,,N,,443,2806,728,110,28,9,128,193,,,,,352,156,2.0300000000,73,6,0,2079,637,6,112,240,385,37,0.89,Louisville Eclipse,Eclipse Park I,,95,93,LOU,LS2,LS2
+1882,AA,PH4,PHA,,3,75,,41,34,,,N,,406,2707,660,89,21,5,125,164,,,,,389,219,2.9700000000,72,2,0,1989,682,13,99,190,361,36,0.89,Philadelphia Athletics,Oakdale Park,,110,110,PHA,PH4,PH4
+1882,AA,PT1,PIT,,4,79,,39,39,,,N,,428,2904,730,110,59,18,90,183,,,,,418,216,2.7900000000,77,2,0,2088,694,4,82,252,397,40,0.88,Pittsburg Alleghenys,Exposition Park I,,100,99,PIT,PT1,PT1
+1882,AA,SL4,STL,,5,80,,37,43,,,N,,399,2865,663,87,41,11,112,226,,,,,496,223,2.9200000000,75,3,1,2064,729,7,103,225,446,41,0.87,St. Louis Brown Stockings,Sportsman's Park I,,104,104,STL,SL4,SL4
+1882,NL,BFN,BUF,,3,84,,45,39,,,N,,500,3128,858,146,47,18,116,228,,,,,461,266,3.2500000000,79,3,0,2211,778,16,114,287,315,42,0.91,Buffalo Bisons,Riverside Park,,102,101,BUF,BFN,BFN
+1882,NL,BSN,ATL,,3,85,,45,39,,,N,,472,3118,823,114,50,15,134,244,,,,,414,233,2.8000000000,81,4,0,2247,738,10,77,352,295,37,0.91,Boston Red Caps,South End Grounds I,,100,98,BSN,BSN,BSN
+1882,NL,CHN,CHC,,1,84,,55,29,,,Y,,604,3225,892,209,54,15,142,262,,,,,353,188,2.2200000000,83,7,0,2289,667,13,102,279,376,54,0.89,Chicago White Stockings,Lake Front Park I/Lake Front Park II,,105,99,CHC,CHN,CHN
+1882,NL,CL2,CBL,,5,84,,42,40,,,N,,402,3009,716,139,40,20,122,261,,,,,411,229,2.7400000000,81,4,0,2253,743,22,132,232,355,71,0.90,Cleveland Blues,Kennard Street Park,,97,97,CLV,CL2,CL2
+1882,NL,DTN,DTN,,5,86,,42,41,,,N,,407,3144,724,117,44,19,122,308,,,,,488,263,2.9800000000,82,7,0,2379,808,19,129,354,377,44,0.89,Detroit Wolverines,Recreation Park,,100,101,DTN,DTN,DTN
+1882,NL,PRO,PRO,,2,84,,52,32,,,N,,463,3104,776,121,53,11,102,255,,,,,356,190,2.2700000000,80,10,1,2256,690,12,87,273,371,67,0.90,Providence Grays,Messer Street Grounds,,100,98,PRO,PRO,PRO
+1882,NL,TRN,TRT,,7,85,,35,48,,,N,,430,3057,747,116,59,12,109,298,,,,,522,259,3.0800000000,81,6,0,2271,837,13,168,189,432,70,0.88,Troy Trojans,Troy Ball Club Grounds,,95,98,TRO,TRN,TRN
+1882,NL,WOR,WOR,,8,84,,18,66,,,N,,379,2984,689,109,57,16,113,303,,,,,652,308,3.7600000000,75,0,0,2214,964,21,151,195,469,66,0.87,Worcester Ruby Legs,Worcester Driving Park Grounds,,102,107,WOR,WOR,WOR
+1883,AA,BL2,BLO,,8,96,,28,68,,,N,,471,3534,870,125,49,5,162,331,,,,,742,383,4.0800000000,86,1,0,2532,943,12,190,290,624,,0.85,Baltimore Orioles,Oriole Park,,103,105,BAL,BL2,BL2
+1883,AA,CL5,CBK,,6,97,,32,65,,,N,,476,3553,854,101,79,15,134,410,,,,,659,370,3.9600000000,90,4,0,2520,980,16,211,222,535,,0.87,Columbus Buckeyes,Recreation Park I,,92,93,COL,CL5,CL5
+1883,AA,CN2,CIN,,3,98,,61,37,,,N,,662,3669,961,122,74,34,139,261,,,,,413,217,2.2600000000,96,8,0,2598,766,17,168,215,360,,0.90,Cincinnati Red Stockings,Bank Street Grounds,,106,99,CIN,CN2,CN2
+1883,AA,LS2,LOU,,5,98,,52,45,,,N,,564,3553,891,114,64,14,140,305,,,,,562,340,3.5100000000,96,7,0,2619,987,7,110,269,471,,0.88,Louisville Eclipse,Eclipse Park I,,93,91,LOU,LS2,LS2
+1883,AA,NY4,NYP,,4,97,,54,42,,,N,,498,3534,883,111,58,6,142,259,,,,,405,282,2.9000000000,97,6,0,2622,749,12,123,480,391,,0.90,New York Metropolitans,Polo Grounds I West Diamond,,109,106,NYP,NY4,NY4
+1883,AA,PH4,PHA,,1,98,,66,32,,,Y,,720,3714,972,149,50,20,194,268,,,,,547,279,2.8800000000,92,1,0,2619,921,22,95,347,584,,0.86,Philadelphia Athletics,Jefferson Street Grounds,,108,105,PHA,PH4,PH4
+1883,AA,PT1,PIT,,7,98,,31,67,,,N,,525,3609,892,120,58,13,162,345,,,,,728,445,4.6200000000,82,1,1,2601,1140,21,151,271,506,,0.88,Pittsburg Alleghenys,Exposition Park I,,94,98,PIT,PT1,PT1
+1883,AA,SL4,STL,,2,98,,65,33,,,N,,549,3495,892,118,46,7,125,240,,,,,409,218,2.2300000000,93,9,1,2637,729,7,150,325,388,,0.90,St. Louis Browns,Sportsman's Park I,,106,106,STL,SL4,SL4
+1883,NL,BFN,BUF,,5,98,,52,45,,,N,,614,3729,1058,184,59,8,147,342,,,,,576,317,3.3200000000,90,5,2,2577,971,12,101,362,445,,0.89,Buffalo Bisons,Riverside Park,,102,101,BUF,BFN,BFN
+1883,NL,BSN,ATL,,1,98,,63,35,,,Y,,669,3657,1010,209,86,34,123,423,,,,,456,244,2.5500000000,89,6,3,2580,853,11,90,538,382,,0.90,Boston Beaneaters,South End Grounds I,,102,98,BSN,BSN,BSN
+1883,NL,CHN,CHC,,2,98,,59,39,,,N,,679,3658,1000,277,61,13,129,399,,,,,540,266,2.7800000000,91,5,1,2586,942,21,123,299,543,,0.87,Chicago White Stockings,Lake Front Park II,,108,105,CHC,CHN,CHN
+1883,NL,CL2,CBL,,4,100,,55,42,,,N,,476,3457,852,184,38,8,139,374,,,,,443,217,2.2200000000,92,5,2,2637,818,7,217,402,389,,0.90,Cleveland Blues,Kennard Street Park,,99,100,CLV,CL2,CL2
+1883,NL,DTN,DTN,,7,101,,40,58,,,N,,524,3726,931,164,48,13,166,378,,,,,650,356,3.5800000000,89,5,2,2682,1026,22,184,324,444,,0.89,Detroit Wolverines,Recreation Park,,95,99,DTN,DTN,DTN
+1883,NL,NY1,SFG,,6,98,,46,50,,,N,,530,3524,900,139,69,24,127,297,,,,,577,283,2.9400000000,87,5,0,2598,907,19,170,323,468,,0.88,New York Gothams,Polo Grounds I,,99,99,NYG,NY1,NY1
+1883,NL,PHI,PHI,,8,99,,17,81,,,N,,437,3576,859,181,48,3,141,355,,,,,887,513,5.3400000000,91,3,0,2592,1267,20,125,253,639,,0.85,Philadelphia Quakers,Recreation Park,,91,98,PHI,PHI,PHI
+1883,NL,PRO,PRO,,3,98,,58,40,,,N,,636,3685,1001,189,59,21,149,309,,,,,436,229,2.3700000000,88,4,1,2613,827,12,111,376,419,,0.90,Providence Grays,Messer Street Grounds,,103,99,PRO,PRO,PRO
+1884,AA,BL2,BLO,,6,108,,63,43,,,N,N,636,3845,896,133,84,32,211,545,,,,,515,288,2.7100000000,105,8,1,2865,869,16,219,635,461,,0.89,Baltimore Orioles,Oriole Park,,104,107,BAL,BL2,BL2
+1884,AA,BR3,LAD,,9,109,,40,64,,,N,N,476,3763,845,112,47,16,179,417,,,,,644,399,3.7900000000,105,6,0,2844,996,20,163,378,508,,0.88,Brooklyn Atlantics,Washington Park I,,100,102,BRO,BR3,BR3
+1884,AA,CL5,CBK,,2,110,,69,39,,,N,N,585,3759,901,107,96,40,196,404,,,,,459,286,2.6800000000,102,8,1,2886,815,22,172,526,433,,0.90,Columbus Buckeyes,Recreation Park I,,92,93,COL,CL5,CL5
+1884,AA,CN2,CIN,,5,112,,68,41,,,N,N,754,4090,1037,109,96,36,154,404,,,,,512,364,3.3300000000,111,11,0,2949,956,27,181,308,412,,0.91,Cincinnati Red Stockings,League Park I in Cincinnati,,104,102,CIN,CN2,CN2
+1884,AA,IN2,IHO,,12,110,,29,78,,,N,N,462,3813,890,129,62,20,125,560,,,,,755,437,4.2000000000,107,2,0,2811,1001,30,199,479,515,,0.88,Indianapolis Hoosiers,Seventh Street Park,,97,100,IND,IN2,IN2
+1884,AA,LS2,LOU,,3,110,,68,40,,,N,N,573,3957,1004,152,69,17,146,408,,,,,425,238,2.1700000000,101,6,0,2967,836,9,97,470,426,,0.91,Louisville Eclipse,Eclipse Park I,,95,95,LOU,LS2,LS2
+1884,AA,NY4,NYP,,1,112,,75,32,,,Y,N,734,4012,1052,155,64,22,203,315,,,,,423,269,2.4600000000,111,9,0,2952,800,15,119,611,441,,0.90,New York Metropolitans,Polo Grounds I West Diamond,,100,96,NYP,NY4,NY4
+1884,AA,PH4,PHA,,7,108,,61,46,,,N,N,700,3959,1057,167,100,26,153,425,,,,,546,360,3.4200000000,105,5,0,2844,920,16,127,530,458,,0.90,Philadelphia Athletics,Jefferson Street Grounds,,107,104,PHA,PH4,PH4
+1884,AA,PT1,PIT,,11,110,,30,78,,,N,N,406,3689,777,105,50,2,143,411,,,,,725,456,4.3500000000,108,4,0,2829,1059,25,216,338,523,,0.88,Pittsburg Alleghenys,Recreation Park,,97,101,PIT,PT1,PT1
+1884,AA,RIC,RIC,,10,46,,12,30,,,N,N,194,1469,325,40,33,7,53,284,,,,,294,186,4.5200000000,45,1,0,1110,402,14,52,167,238,,0.87,Richmond Virginians,Allen Pasture,,99,102,RIC,RIC,RIC
+1884,AA,SL4,STL,,4,110,,67,40,,,N,N,658,3952,986,151,60,11,174,339,,,,,539,293,2.6700000000,99,8,0,2961,881,16,172,477,489,,0.90,St. Louis Browns,Sportsman's Park I,,104,101,STL,SL4,SL4
+1884,AA,TL1,TOL,,8,110,,46,58,,,N,N,463,3712,859,153,48,8,157,541,,,,,571,322,3.0600000000,103,9,1,2838,885,12,169,501,469,,0.90,Toledo Blue Stockings,League Park,,103,105,TOL,TL1,TL1
+1884,AA,WS7,WST,,13,63,,12,51,,,N,N,248,2166,434,61,24,6,102,377,,,,,481,242,4.0100000000,62,3,0,1629,643,21,110,235,399,,0.85,Washington Nationals,Athletic Park,,88,93,WAS,WS7,WS7
+1884,NL,BFN,BUF,,3,115,,64,47,,,N,N,700,4197,1099,163,69,39,215,458,,,,,626,328,2.9500000000,108,14,1,3003,1041,46,189,534,463,,0.90,Buffalo Bisons,Olympic Park I,,103,105,BUF,BFN,BFN
+1884,NL,BSN,ATL,,2,116,,73,38,,,N,N,684,4189,1063,179,60,36,207,660,,,,,468,285,2.4700000000,109,14,2,3111,932,30,135,742,374,,0.92,Boston Beaneaters,South End Grounds I,,99,97,BSN,BSN,BSN
+1884,NL,CHN,CHC,,5,113,,62,50,,,N,N,834,4182,1176,162,50,142,264,469,,,,,647,336,3.0300000000,106,9,0,2991,1028,83,231,472,595,,0.88,Chicago White Stockings,Lake Front Park II,,108,105,CHC,CHN,CHN
+1884,NL,CL2,CBL,,7,113,,35,77,,,N,N,458,3934,934,147,49,16,170,576,,,,,716,379,3.4300000000,107,7,0,2982,1046,35,269,482,512,,0.89,Cleveland Blues,Kennard Street Park,,103,106,CLV,CL2,CL2
+1884,NL,DTN,DTN,,8,114,,28,84,,,N,N,445,3970,825,114,47,31,207,699,,,,,736,370,3.3800000000,109,3,0,2952,1097,36,245,488,514,,0.89,Detroit Wolverines,Recreation Park,,94,97,DTN,DTN,DTN
+1884,NL,NY1,SFG,,4,116,,62,50,,,N,N,693,4124,1053,149,67,23,249,492,,,,,623,352,3.1200000000,111,4,0,3042,1011,28,326,567,514,,0.89,New York Gothams,Polo Grounds I,,105,102,NYG,NY1,NY1
+1884,NL,PHI,PHI,,6,113,,39,73,,,N,N,549,3998,934,149,39,14,209,512,,,,,824,428,3.9300000000,106,3,1,2943,1090,38,254,411,536,,0.88,Philadelphia Quakers,Recreation Park,,94,99,PHI,PHI,PHI
+1884,NL,PRO,PRO,,1,114,,84,28,,,Y,Y,665,4093,987,153,43,21,300,469,,,,,388,185,1.6100000000,107,16,2,3108,825,26,172,639,398,,0.91,Providence Grays,Messer Street Grounds,,99,96,PRO,PRO,PRO
+1884,UA,ALT,ALT,,10,25,,6,19,,,N,,90,899,223,30,6,2,22,130,,,,,216,114,4.6800000000,20,0,0,657,292,3,52,93,156,,0.86,Altoona Mountain City,,,101,109,ALT,ALT,ALT
+1884,UA,BLU,BLU,,4,106,,58,47,,,N,,662,3883,952,150,26,17,144,652,,,,,627,316,3.0100000000,92,4,0,2838,1002,24,177,628,610,,0.87,Baltimore Monumentals,,,109,109,BLU,BLU,BLU
+1884,UA,BSU,BRD,,5,111,,58,51,,,N,,636,3940,928,168,32,19,128,787,,,,,558,286,2.7000000000,100,5,1,2859,885,17,110,753,633,,0.86,Boston Reds,,,99,98,BOS,BSU,BSU
+1884,UA,CHU,CPI,,6,93,,41,50,,,N,,438,3212,742,127,26,10,119,505,,,,,482,243,2.7200000000,86,6,0,2409,743,12,137,679,459,,0.88,Chicago/Pittsburgh (Union League),,,98,99,CPI,CHU,CHU
+1884,UA,CNU,COR,,3,105,,69,36,,,N,,703,3786,1027,118,63,26,147,482,,,,,466,242,2.3800000000,95,11,1,2742,831,17,90,503,532,,0.88,Cincinnati Outlaw Reds,,,111,107,COR,CNU,CNU
+1884,UA,KCU,KCU,,11,82,,16,63,,,N,,311,2802,557,104,15,6,123,529,,,,,618,317,4.0600000000,70,0,0,2106,862,14,127,334,520,,0.86,Kansas City Cowboys,,,87,92,KCC,KCU,KCU
+1884,UA,MLU,MLU,,2,12,,8,4,,,N,,53,395,88,25,0,0,20,70,,,,,34,26,2.2500000000,12,3,0,312,49,1,13,139,53,,0.89,Milwaukee Brewers,,,60,60,MIL,MLU,MLU
+1884,UA,PHU,PHK,,8,67,,21,46,,,N,,414,2518,618,108,35,7,103,405,,,,,545,305,4.6300000000,64,1,0,1779,726,7,105,310,501,,0.84,Philadelphia Keystones,,,91,94,PHK,PHU,PHU
+1884,UA,SLU,SLM,,1,114,,94,19,,,Y,,887,4285,1251,259,41,32,181,542,,,,,429,216,1.9600000000,104,8,6,2979,838,9,110,550,554,,0.88,St. Louis Maroons,,,99,98,SLM,SLU,SLU
+1884,UA,SPU,STP,,9,9,,2,6,,,N,,24,272,49,13,1,0,7,47,,,,,57,25,3.1700000000,7,1,0,213,72,1,27,44,47,,0.87,St. Paul White Caps,,,60,60,STP,SPU,SPU
+1884,UA,WIL,WIL,,12,18,,2,16,,,N,,35,521,91,8,8,2,22,123,,,,,114,48,3.0400000000,15,0,0,426,165,4,18,113,104,,0.85,Wilmington Quicksteps,,,103,109,WIL,WIL,WIL
+1884,UA,WSU,WNA,,7,114,,47,65,,,N,,572,3926,931,120,26,4,118,558,,,,,679,364,3.4400000000,94,5,0,2859,992,16,168,684,550,,0.87,Washington Nationals,,,96,98,WHS,WSU,WSU
+1885,AA,BL2,BLO,,8,110,,41,68,,,N,N,541,3820,837,124,59,17,279,529,,,,,683,421,3.9000000000,103,2,4,2913,1059,12,222,395,419,,0.90,Baltimore Orioles,Oriole Park,,99,100,BAL,BL2,BL2
+1885,AA,BR3,LAD,,5,112,,53,59,,,N,N,624,3943,966,121,65,14,238,324,,,,,650,381,3.4600000000,110,3,1,2973,955,27,211,436,436,,0.91,Brooklyn Grays,Washington Park I,,100,102,BRO,BR3,BR3
+1885,AA,CN2,CIN,,2,112,,63,49,,,N,N,642,4050,1046,108,77,26,153,420,,,,,575,362,3.2600000000,102,7,1,2997,998,24,250,330,417,,0.91,Cincinnati Red Stockings,League Park I in Cincinnati,,102,100,CIN,CN2,CN2
+1885,AA,LS2,LOU,,6,112,,53,59,,,N,N,564,3969,986,126,83,19,152,448,,,,,598,298,2.6800000000,109,3,1,3006,927,13,217,462,460,,0.90,Louisville Colonels,Eclipse Park I,,100,100,LOU,LS2,LS2
+1885,AA,NY4,NYP,,7,108,,44,64,,,N,N,526,3731,921,123,57,21,217,428,,,,,688,432,4.1500000000,103,2,0,2811,1015,36,204,408,453,,0.90,New York Metropolitans,Polo Grounds I West Diamond,,84,86,NYP,NY4,NY4
+1885,AA,PH4,PHA,,4,113,,55,57,,,N,N,764,4142,1099,169,76,30,223,410,,,,,691,360,3.2300000000,105,5,0,3009,1038,11,212,506,482,,0.90,Philadelphia Athletics,Jefferson Street Grounds,,106,106,PHA,PH4,PH4
+1885,AA,PT1,PIT,,3,111,,56,55,,,N,N,547,3975,955,123,79,5,189,537,,,,,539,328,2.9200000000,104,8,0,3033,918,14,201,454,422,,0.91,Pittsburg Alleghenys,Recreation Park,,100,99,PIT,PT1,PT1
+1885,AA,SL4,STL,,1,112,,79,33,,,Y,N,677,3972,979,132,57,17,234,282,,,,,461,272,2.4400000000,111,11,0,3006,879,12,168,378,381,,0.92,St. Louis Browns,Sportsman's Park I,,105,101,STL,SL4,SL4
+1885,NL,BFN,BUF,,7,112,,38,74,,,N,N,495,3900,980,149,50,23,179,380,,,,,761,456,4.2900000000,107,4,1,2868,1175,31,234,320,464,,0.90,Buffalo Bisons,Olympic Park I,,103,105,BUF,BFN,BFN
+1885,NL,BSN,ATL,,5,113,,46,66,,,N,N,528,3950,915,144,53,22,190,522,,,,,589,330,3.0300000000,111,10,0,2943,1045,26,188,480,445,,0.90,Boston Beaneaters,South End Grounds I,,96,95,BSN,BSN,BSN
+1885,NL,CHN,CHC,,1,113,,87,25,,,Y,N,834,4093,1079,184,75,54,340,429,,,,,470,251,2.2300000000,108,14,4,3045,868,37,202,458,497,,0.90,Chicago White Stockings,West Side Park I,,114,107,CHC,CHN,CHN
+1885,NL,DTN,DTN,,6,108,,41,67,,,N,N,514,3773,917,149,66,25,216,451,,,,,582,305,2.8800000000,105,6,1,2862,966,18,224,475,445,,0.90,Detroit Wolverines,Recreation Park,,100,101,DTN,DTN,DTN
+1885,NL,NY1,SFG,,2,112,,85,27,,,N,N,691,4029,1085,150,82,16,221,312,,,,,370,190,1.7200000000,109,16,1,2982,758,11,265,516,331,,0.92,New York Giants,Polo Grounds I,,101,97,NYG,NY1,NY1
+1885,NL,PHI,PHI,,3,111,,56,54,,,N,N,513,3893,891,156,35,20,220,401,,,,,511,259,2.3900000000,108,10,0,2928,860,18,218,378,447,,0.90,Philadelphia Quakers,Recreation Park,,97,98,PHI,PHI,PHI
+1885,NL,PRO,PRO,,4,110,,53,57,,,N,N,442,3727,820,114,30,6,265,430,,,,,531,289,2.7100000000,108,8,0,2880,912,18,235,371,459,,0.90,Providence Grays,Messer Street Grounds,,97,95,PRO,PRO,PRO
+1885,NL,SL5,SLM,,8,111,,36,72,,,N,N,390,3758,829,121,21,8,214,412,,,,,593,361,3.3700000000,107,4,0,2895,935,15,278,337,398,,0.91,St. Louis Maroons,Sportsman's Park I,,97,98,SLM,SL5,SL5
+1886,AA,BL2,BLO,,8,139,,48,83,,,N,N,625,4639,945,124,51,8,379,603,269,,,,878,547,4.0800000000,134,5,0,3618,1197,25,403,805,524,,0.90,Baltimore Orioles,Oriole Park,,98,99,BAL,BL2,BL2
+1886,AA,BR3,LAD,,3,141,,76,61,,,N,N,832,5053,1261,196,80,16,433,523,248,,,,832,469,3.4200000000,138,6,0,3702,1202,17,464,540,610,,0.90,Brooklyn Grays,Washington Park I,,100,100,BRO,BR3,BR3
+1886,AA,CN2,CIN,,5,141,,65,73,,,N,N,883,4915,1225,145,95,45,374,633,185,,,,865,579,4.1800000000,129,3,0,3741,1267,25,481,495,580,,0.90,Cincinnati Red Stockings,League Park I in Cincinnati,,103,102,CIN,CN2,CN2
+1886,AA,LS2,LOU,,4,138,,66,70,,,N,N,833,4921,1294,182,88,20,410,558,202,,,,805,412,3.0700000000,131,5,2,3627,1109,16,432,720,593,,0.90,Louisville Colonels,Eclipse Park I,,106,106,LOU,LS2,LS2
+1886,AA,NY4,NYP,,7,137,,53,82,,,N,N,628,4683,1047,108,72,18,330,578,120,,,,766,461,3.5000000000,134,5,0,3558,1148,23,386,559,545,,0.90,New York Metropolitans,St. George Cricket Grounds,,96,100,NYP,NY4,NY4
+1886,AA,PH4,PHA,,6,139,,63,72,,,N,N,772,4856,1142,192,82,21,378,697,284,,,,942,537,3.9700000000,134,4,0,3654,1308,35,388,513,637,,0.89,Philadelphia Athletics,Jefferson Street Grounds,,101,102,PHA,PH4,PH4
+1886,AA,PT1,PIT,,2,140,,80,57,,,N,N,810,4854,1171,186,96,16,478,713,260,,,,647,386,2.8300000000,137,15,1,3678,1130,10,299,515,487,,0.91,Pittsburg Alleghenys,Recreation Park,,98,97,PIT,PT1,PT1
+1886,AA,SL4,STL,,1,139,,93,46,,,Y,Y,944,5009,1365,206,85,20,400,425,336,,,,592,340,2.4900000000,134,14,2,3687,1087,13,329,583,494,,0.91,St. Louis Browns,Sportsman's Park I,,105,100,STL,SL4,SL4
+1886,NL,BSN,ATL,,5,118,,56,61,,,N,N,657,4180,1085,151,59,24,250,537,156,,,,661,370,3.2400000000,116,3,0,3087,1049,33,298,511,465,,0.90,Boston Beaneaters,South End Grounds I,,97,97,BSN,BSN,BSN
+1886,NL,CHN,CHC,,1,126,,90,34,,,Y,N,900,4378,1223,198,87,53,460,513,213,,,,555,310,2.5400000000,116,8,3,3291,988,49,262,647,475,,0.91,Chicago White Stockings,West Side Park I,,114,109,CHC,CHN,CHN
+1886,NL,DTN,DTN,,2,126,,87,36,,,N,N,829,4501,1260,176,81,53,374,426,194,,,,538,349,2.8500000000,122,8,0,3309,995,20,270,592,373,,0.92,Detroit Wolverines,Recreation Park,,103,100,DTN,DTN,DTN
+1886,NL,KCN,KCN,,7,126,,30,91,,,N,N,494,4236,967,177,48,19,269,608,96,,,,872,573,4.8400000000,117,4,0,3198,1345,27,246,442,455,,0.90,Kansas City Cowboys,Association Park,,106,114,KCN,KCN,KCN
+1886,NL,NY1,SFG,,3,124,,75,44,,,N,N,692,4298,1156,175,68,21,237,410,155,,,,558,337,2.8600000000,119,3,1,3186,1029,23,280,588,359,,0.92,New York Giants,Polo Grounds I,,103,99,NYG,NY1,NY1
+1886,NL,PHI,PHI,,4,119,,71,43,,,N,N,621,4072,976,145,66,26,282,516,226,,,,498,284,2.4500000000,110,10,2,3135,923,29,264,540,393,,0.92,Philadelphia Quakers,Recreation Park,,100,99,PHI,PHI,PHI
+1886,NL,SL5,SLM,,6,126,,43,79,,,N,N,547,4250,1001,183,46,30,235,656,156,,,,712,388,3.2400000000,118,6,0,3231,1050,34,392,501,452,,0.91,St. Louis Maroons,Sportsman's Park I,,94,97,SLM,SL5,SL5
+1886,NL,WS8,WNL,,8,125,,28,92,,,N,N,445,4082,856,135,51,23,265,582,143,,,,791,497,4.3000000000,116,4,0,3123,1147,34,379,500,458,,0.91,Washington Nationals,Swampdoodle Grounds,,94,99,WHS,WS8,WS8
+1887,AA,BL2,BLO,,3,141,,77,58,,,N,N,975,4825,1337,202,100,31,469,334,545,,,,861,525,3.8700000000,132,8,0,3660,1288,16,418,470,549,,0.90,Baltimore Orioles,Oriole Park,,94,95,BAL,BL2,BL2
+1887,AA,BR3,LAD,,6,138,,60,74,,,N,N,904,4913,1281,200,82,25,456,365,409,,,,918,589,4.4700000000,132,3,3,3555,1348,27,454,332,562,,0.90,Brooklyn Grays,Washington Park I,,101,100,BRO,BR3,BR3
+1887,AA,CL3,CLV,,8,133,,39,92,,,N,N,729,4649,1170,178,77,14,375,463,355,,,,1112,630,4.9900000000,127,2,1,3408,1472,34,533,332,576,,0.89,Cleveland Blues,National League Park II,,97,101,CLE,CL3,CL3
+1887,AA,CN2,CIN,,2,136,,81,54,,,N,N,892,4797,1285,179,102,37,382,366,527,,,,745,470,3.5800000000,129,11,1,3546,1202,28,396,330,484,,0.91,Cincinnati Red Stockings,League Park I in Cincinnati,,102,101,CIN,CN2,CN2
+1887,AA,LS2,LOU,,4,139,,76,60,,,N,N,956,4916,1420,194,98,27,436,356,466,,,,854,511,3.8200000000,133,3,1,3615,1274,31,357,544,574,,0.90,Louisville Colonels,Eclipse Park I,,102,102,LOU,LS2,LS2
+1887,AA,NY4,NYP,,7,138,,44,89,,,N,N,754,4820,1197,193,66,21,439,463,305,,,,1093,692,5.2800000000,132,1,0,3540,1545,39,406,316,628,,0.89,New York Metropolitans,St. George Cricket Grounds,,96,100,NYP,NY4,NY4
+1887,AA,PH4,PHA,,5,137,,64,69,,,N,N,893,4954,1370,231,84,29,321,388,476,,,,890,605,4.5900000000,131,5,1,3558,1227,29,433,417,528,,0.90,Philadelphia Athletics,Jefferson Street Grounds,,100,100,PHA,PH4,PH4
+1887,AA,SL4,STL,,1,138,,95,40,,,Y,N,1131,5048,1550,261,78,39,442,340,581,,,,761,502,3.7700000000,132,7,2,3597,1254,19,323,334,481,,0.91,St. Louis Browns,Sportsman's Park I,,111,106,STL,SL4,SL4
+1887,NL,BSN,ATL,,5,127,,61,60,,,N,N,831,4531,1255,185,94,53,340,392,373,,,,792,539,4.4100000000,123,4,1,3300,1226,55,396,254,522,,0.90,Boston Beaneaters,South End Grounds I,,98,97,BSN,BSN,BSN
+1887,NL,CHN,CHC,,3,127,,71,50,,,N,N,813,4350,1177,178,98,80,407,400,382,,,,716,433,3.4600000000,117,4,3,3378,1156,55,338,510,472,,0.91,Chicago White Stockings,West Side Park I,,113,110,CHC,CHN,CHN
+1887,NL,DTN,DTN,,1,127,,79,45,,,Y,Y,969,4689,1404,213,126,55,352,258,267,,,,714,490,3.9500000000,122,3,1,3348,1172,52,344,337,384,,0.92,Detroit Wolverines,Recreation Park,,104,100,DTN,DTN,DTN
+1887,NL,IN3,IND,,8,127,,37,89,,,N,N,628,4368,1080,162,70,33,300,379,334,,,,965,633,5.2400000000,118,4,1,3264,1289,60,431,245,479,,0.91,Indianapolis Hoosiers,Athletic Park I,,96,102,IND,IN3,IN3
+1887,NL,NY1,SFG,,4,129,,68,55,,,N,N,816,4516,1259,167,93,48,361,326,415,,,,723,441,3.5700000000,123,5,1,3339,1096,27,373,415,430,,0.92,New York Giants,Polo Grounds I,,96,93,NYG,NY1,NY1
+1887,NL,PHI,PHI,,2,128,,75,48,,,N,N,901,4630,1269,213,89,47,385,346,355,,,,702,436,3.4700000000,119,7,1,3396,1173,48,305,435,471,,0.91,Philadelphia Quakers,Philadelphia Baseball Grounds,,106,104,PHI,PHI,PHI
+1887,NL,PIT,PIT,,6,125,,55,69,,,N,N,621,4414,1141,183,78,20,319,381,221,,,,750,507,4.1200000000,123,3,0,3324,1287,39,246,248,425,,0.92,Pittsburg Alleghenys,Recreation Park,,93,93,PIT,PIT,PIT
+1887,NL,WS8,WNL,,7,126,,46,76,,,N,N,601,4314,1039,149,63,47,269,339,334,,,,818,507,4.1900000000,124,4,0,3270,1216,47,299,396,469,,0.91,Washington Nationals,Swampdoodle Grounds,,95,100,WHS,WS8,WS8
+1888,AA,BL2,BLO,,5,137,,57,80,,,N,N,653,4656,1068,162,70,19,298,479,326,,,,779,504,3.7800000000,130,3,0,3600,1162,23,419,525,461,,0.92,Baltimore Orioles,Oriole Park,,98,98,BAL,BL2,BL2
+1888,AA,BR3,LAD,,2,143,,88,52,,,N,N,758,4871,1177,172,70,25,353,439,334,,,,584,333,2.3300000000,138,9,0,3858,1059,15,285,577,502,,0.91,Brooklyn Bridegrooms,Washington Park I,,101,100,BRO,BR3,BR3
+1888,AA,CL3,CLV,,6,135,,50,82,,,N,N,651,4603,1076,128,59,12,315,559,353,,,,839,484,3.7200000000,131,6,1,3513,1235,38,389,500,480,,0.91,Cleveland Blues,National League Park II,,97,101,CLE,CL3,CL3
+1888,AA,CN2,CIN,,4,137,,80,54,,,N,N,745,4801,1161,132,82,32,345,555,469,,,,628,375,2.7300000000,132,10,2,3711,1103,19,310,539,456,,0.92,Cincinnati Red Stockings,League Park I in Cincinnati,,105,104,CIN,CN2,CN2
+1888,AA,KC2,KCC,,8,132,,43,89,,,N,N,579,4588,1000,142,61,19,288,604,257,,,,896,552,4.2900000000,128,4,0,3471,1306,32,401,381,507,,0.91,Kansas City Cowboys,Association Park I,,106,112,KCC,KC2,KC2
+1888,AA,LS2,LOU,,7,139,,48,87,,,N,N,689,4881,1177,183,67,14,322,604,318,,,,870,445,3.2500000000,133,6,0,3693,1264,28,281,599,609,,0.90,Louisville Colonels,Eclipse Park I,,98,100,LOU,LS2,LS2
+1888,AA,PH4,PHA,,3,136,,81,52,,,N,N,827,4828,1209,183,89,31,303,473,434,,,,594,323,2.4100000000,133,13,0,3624,988,14,324,596,475,,0.91,Philadelphia Athletics,Jefferson Street Grounds,,100,98,PHA,PH4,PH4
+1888,AA,SL4,STL,,1,137,,92,43,,,Y,N,789,4755,1189,149,47,36,410,521,468,,,,501,281,2.0900000000,132,12,0,3636,939,19,225,517,430,,0.92,St. Louis Browns,Sportsman's Park I,,111,106,STL,SL4,SL4
+1888,NL,BSN,ATL,,4,137,,70,64,,,N,N,669,4834,1183,167,89,56,282,524,293,,,,619,355,2.6100000000,134,7,0,3675,1104,36,269,484,494,,0.91,Boston Beaneaters,South End Grounds II,,104,102,BSN,BSN,BSN
+1888,NL,CHN,CHC,,2,136,,77,58,,,N,N,734,4616,1201,147,95,77,290,563,287,,,,659,390,2.9600000000,123,13,1,3558,1139,63,308,588,417,,0.92,Chicago White Stockings,West Side Park I,,107,106,CHC,CHN,CHN
+1888,NL,DTN,DTN,,5,134,,68,63,,,N,N,721,4849,1275,177,72,51,307,396,193,,,,629,365,2.7400000000,130,10,1,3597,1115,44,183,522,463,,0.91,Detroit Wolverines,Recreation Park,,100,97,DTN,DTN,DTN
+1888,NL,IN3,IND,,7,136,,50,85,,,N,N,603,4623,1100,180,33,34,236,492,350,,,,731,502,3.8100000000,132,6,0,3561,1260,64,308,388,449,,0.92,Indianapolis Hoosiers,Athletic Park II,,102,104,IND,IN3,IN3
+1888,NL,NY1,SFG,,1,138,,84,47,,,Y,Y,659,4747,1149,130,76,55,270,456,314,,,,479,263,1.9600000000,133,20,1,3624,907,27,307,726,430,,0.92,New York Giants,Polo Grounds I,,99,96,NYG,NY1,NY1
+1888,NL,PHI,PHI,,3,132,,69,61,,,N,N,535,4528,1021,151,46,16,268,485,246,,,,509,309,2.3800000000,125,16,3,3501,1072,26,196,519,424,,0.92,Philadelphia Quakers,Philadelphia Baseball Grounds,,105,104,PHI,PHI,PHI
+1888,NL,PIT,PIT,,6,139,,66,68,,,N,N,534,4713,1070,150,49,14,194,583,287,,,,580,357,2.6700000000,135,13,0,3609,1190,23,223,367,416,,0.92,Pittsburg Alleghenys,Recreation Park,,92,93,PIT,PIT,PIT
+1888,NL,WS8,WNL,,8,136,,48,86,,,N,N,482,4546,944,98,49,30,246,499,331,,,,731,464,3.5400000000,133,6,0,3537,1157,50,298,406,474,,0.91,Washington Nationals,Swampdoodle Grounds,,94,98,WHS,WS8,WS8
+1889,AA,BL2,BLO,,5,139,,70,65,,,N,N,791,4756,1209,155,68,20,418,536,311,,,,795,472,3.5600000000,128,10,1,3576,1168,27,424,540,536,,0.90,Baltimore Orioles,Oriole Park,,98,99,BAL,BL2,BL2
+1889,AA,BR3,LAD,,1,140,,93,44,,,Y,N,995,4815,1265,188,79,47,550,401,389,,,,706,486,3.6100000000,120,10,1,3636,1205,33,400,471,421,,0.92,Brooklyn Bridegrooms,Washington Park I,,95,91,BRO,BR3,BR3
+1889,AA,CL6,CLS,,6,140,,60,78,,,N,N,779,4816,1247,171,95,36,507,609,304,,,,924,585,4.3900000000,114,9,4,3597,1274,33,551,610,488,,0.91,Columbus Solons,Recreation Park II,,95,94,COL,CL6,CL6
+1889,AA,CN2,CIN,,4,141,,76,63,,,N,N,897,4844,1307,197,96,52,452,511,462,,,,769,483,3.5000000000,114,3,8,3729,1270,35,475,562,440,,0.92,Cincinnati Red Stockings,League Park I in Cincinnati,,105,103,CIN,CN2,CN2
+1889,AA,KC2,KCC,,7,139,,55,82,,,N,N,852,4947,1256,162,76,18,430,626,472,,,,1031,583,4.3600000000,128,0,2,3612,1373,51,457,447,611,,0.89,Kansas City Cowboys,Exposition Park,,105,108,KCC,KC2,KC2
+1889,AA,LS2,LOU,,8,140,,27,111,,,N,N,632,4955,1249,170,75,22,320,521,203,,,,1091,655,4.8100000000,127,2,1,3678,1529,43,475,451,584,,0.90,Louisville Colonels,Eclipse Park I,,98,100,LOU,LS2,LS2
+1889,AA,PH4,PHA,,3,138,,75,58,,,N,N,880,4868,1339,239,65,43,534,496,252,,,,787,470,3.5300000000,130,9,1,3597,1200,35,509,479,465,,0.92,Philadelphia Athletics,Jefferson Street Grounds,,98,99,PHA,PH4,PH4
+1889,AA,SL4,STL,,2,141,,90,45,,,N,N,957,4939,1312,211,64,58,493,477,336,,,,680,412,3.0000000000,121,7,3,3711,1166,39,413,617,438,,0.92,St. Louis Browns,Sportsman's Park I,,113,110,STL,SL4,SL4
+1889,NL,BSN,ATL,,2,133,,83,45,,,N,N,826,4628,1251,196,54,42,471,450,331,,,,626,435,3.3600000000,121,10,4,3498,1152,41,413,497,413,,0.92,Boston Beaneaters,South End Grounds II,,106,104,BSN,BSN,BSN
+1889,NL,CHN,CHC,,3,136,,67,65,,,N,N,867,4849,1274,184,66,79,518,516,243,,,,814,513,3.7300000000,123,6,2,3711,1313,71,408,434,463,,0.92,Chicago White Stockings,West Side Park I,,105,103,CHC,CHN,CHN
+1889,NL,CL4,CLV,,6,136,,61,72,,,N,N,656,4673,1167,131,59,25,429,417,237,,,,720,484,3.6600000000,132,6,1,3573,1182,36,519,435,365,,0.93,Cleveland Spiders,National League Park,,104,105,CLV,CL4,CL4
+1889,NL,IN3,IND,,7,135,,59,75,,,N,N,819,4879,1356,228,35,62,377,447,252,,,,894,633,4.8500000000,109,3,2,3522,1365,73,420,408,419,,0.92,Indianapolis Hoosiers,Athletic Park II,,102,104,IND,IN3,IN3
+1889,NL,NY1,SFG,,1,131,,83,43,,,Y,Y,935,4671,1319,208,77,52,538,386,292,,,,708,444,3.4700000000,118,6,3,3453,1073,38,523,558,436,,0.92,New York Giants,Polo Grounds II,,104,101,NYG,NY1,NY1
+1889,NL,PHI,PHI,,4,130,,63,64,,,N,N,742,4695,1248,215,52,44,393,353,269,,,,748,512,4.0000000000,106,4,2,3459,1288,33,428,443,466,,0.91,Philadelphia Quakers,Philadelphia Baseball Grounds,,109,108,PHI,PHI,PHI
+1889,NL,PIT,PIT,,5,134,,61,71,,,N,N,726,4748,1202,209,65,42,420,467,231,,,,801,566,4.5100000000,125,5,1,3390,1296,42,374,345,384,,0.93,Pittsburg Alleghenys,Recreation Park,,91,93,PIT,PIT,PIT
+1889,NL,WS8,WNL,,8,127,,41,83,,,N,N,632,4395,1105,151,57,25,466,456,232,,,,892,574,4.6800000000,113,1,0,3309,1261,37,527,388,519,,0.90,Washington Nationals,Swampdoodle Grounds,,94,98,WHS,WS8,WS8
+1890,AA,BL3,BLO,,6,38,,15,19,,,N,N,182,1213,278,34,16,2,125,152,101,,,,192,140,4.0000000000,36,1,0,945,307,3,123,134,109,,0.92,Baltimore Orioles,Oriole Park,,109,110,BAL,BL3,BL3
+1890,AA,BR4,BRG,,9,100,,26,73,,,N,N,492,3475,769,116,47,13,328,456,182,,,,733,460,4.7100000000,96,0,0,2637,1011,21,421,230,394,,0.91,Brooklyn Gladiators,Ridgewood Park,,97,101,BRG,BR4,BR4
+1890,AA,CL6,CLS,,2,140,,79,55,,,N,N,831,4741,1225,159,77,16,545,557,353,,,,617,403,2.9900000000,120,14,3,3642,976,20,471,624,393,,0.93,Columbus Solons,Recreation Park II,,93,93,COL,CL6,CL6
+1890,AA,LS2,LOU,,1,136,,88,44,,,Y,N,819,4687,1310,156,65,15,410,460,341,,,,588,344,2.5700000000,114,13,7,3618,1120,18,293,587,380,,0.93,Louisville Colonels,Eclipse Park I,,98,99,LOU,LS2,LS2
+1890,AA,PH4,PHA,,8,132,,54,78,,,N,N,702,4490,1057,181,51,24,475,540,305,,,,945,657,5.2200000000,119,3,2,3396,1405,17,514,461,453,,0.91,Philadelphia Athletics,Jefferson Street Grounds,,98,99,PHA,PH4,PH4
+1890,AA,RC2,ROC,,5,133,,63,63,,,N,N,709,4553,1088,131,64,31,446,538,310,,,,711,459,3.5600000000,122,5,2,3483,1115,19,530,477,416,,0.92,Rochester Broncos,Culver Field I,,93,92,ROC,RC2,RC2
+1890,AA,SL4,STL,,3,139,,78,58,,,N,N,870,4800,1308,178,73,48,474,490,307,,,,736,487,3.6700000000,118,4,1,3585,1127,38,447,733,472,,0.91,St. Louis Browns,Sportsman's Park I,,114,111,STL,SL4,SL4
+1890,AA,SR2,SYS,,7,128,,55,72,,,N,N,698,4469,1158,151,59,14,457,482,292,,,,831,603,4.9800000000,115,5,0,3267,1158,28,518,454,391,,0.92,Syracuse Stars,Star Park II,,90,91,SYR,SR2,SR2
+1890,AA,TL2,TLM,,4,134,,68,64,,,N,N,739,4575,1152,152,108,24,486,558,421,,,,689,458,3.5600000000,122,4,2,3477,1122,23,429,533,419,,0.92,Toledo Maumees,Speranza Park,,103,102,TOL,TL2,TL2
+1890,NL,BRO,LAD,,1,129,,86,43,,,Y,N,884,4419,1166,184,75,43,517,361,349,,,,620,389,3.0600000000,115,6,2,3435,1102,27,401,403,319,,0.94,Brooklyn Bridegrooms,Washington Park II,121412,101,97,BRO,BRO,BRO
+1890,NL,BSN,ATL,,5,134,,76,57,,,N,N,763,4722,1220,175,62,31,530,515,285,,,,593,386,2.9300000000,132,13,1,3561,1132,27,354,506,358,,0.93,Boston Beaneaters,South End Grounds II,147539,108,105,BSN,BSN,BSN
+1890,NL,CHN,CHC,,2,139,,83,53,,,N,N,847,4891,1271,147,59,67,516,514,329,,,,692,445,3.2400000000,126,6,3,3711,1103,41,481,504,347,,0.93,Chicago Colts,West Side Park I,102536,104,102,CHC,CHN,CHN
+1890,NL,CIN,CIN,,4,134,,77,55,,,N,N,753,4644,1204,150,120,27,433,377,312,,,,633,369,2.7900000000,124,9,1,3570,1097,41,407,488,381,,0.93,Cincinnati Reds,League Park I in Cincinnati,131980,101,101,CIN,CIN,CIN
+1890,NL,CL4,CLV,,7,136,,44,88,,,N,N,630,4633,1073,132,59,21,497,474,152,,,,832,543,4.1300000000,129,2,0,3552,1322,33,462,306,405,,0.92,Cleveland Spiders,National League Park,47478,94,97,CLV,CL4,CL4
+1890,NL,NY1,SFG,,6,135,,63,68,,,N,N,713,4832,1250,208,89,25,350,479,289,,,,698,400,3.0600000000,115,6,1,3531,1029,14,607,612,440,,0.92,New York Giants,Polo Grounds II,60667,95,95,NYG,NY1,NY1
+1890,NL,PHI,PHI,,3,133,,78,53,,,N,N,823,4707,1267,220,78,23,522,403,335,,,,707,440,3.3200000000,122,9,2,3582,1210,22,486,507,396,,0.92,Philadelphia Phillies,Philadelphia Baseball Grounds,148366,103,102,PHI,PHI,PHI
+1890,NL,PIT,PIT,,8,138,,23,113,,,N,N,597,4739,1088,160,43,20,408,458,208,,,,1235,780,5.9700000000,119,3,0,3528,1520,52,573,381,607,,0.89,Pittsburg Alleghenys,Recreation Park,16064,88,92,PIT,PIT,PIT
+1890,PL,BFP,BFB,,8,134,,36,96,,,N,,793,4795,1249,180,64,20,541,367,160,,,,1199,775,6.1100000000,125,2,0,3423,1499,67,673,351,491,,0.91,Buffalo Bisons,,,92,97,BUF,BFP,BFP
+1890,PL,BRP,BWW,,2,133,,76,56,,,N,,964,4887,1354,186,93,34,502,369,272,,,,893,520,3.9500000000,111,4,7,3552,1334,26,570,377,531,,0.90,Brooklyn Ward's Wonders,,,106,105,BWW,BRP,BRP
+1890,PL,BSP,BRS,,1,130,,81,48,,,Y,,992,4626,1306,223,76,54,652,435,412,,,,767,479,3.7900000000,105,6,2,3411,1291,49,467,345,459,,0.91,Boston Reds,,,103,99,BOS,BSP,BSP
+1890,PL,CHP,CHP,,4,138,,75,62,,,N,,886,4968,1311,200,95,31,492,410,276,,,,770,459,3.3900000000,124,5,2,3657,1238,27,503,460,492,,0.91,Chicago Pirates,,,104,103,CHI,CHP,CHP
+1890,PL,CLP,CLI,,7,131,,55,75,,,N,,849,4804,1373,213,94,27,509,345,180,,,,1027,537,4.2300000000,115,1,0,3429,1386,45,571,325,532,,0.90,Cleveland Infants,,,92,94,CLE,CLP,CLP
+1890,PL,NYP,NYI,,3,132,,74,57,,,N,,1018,4913,1393,204,97,66,486,364,231,,,,875,543,4.1700000000,111,3,6,3516,1219,37,569,449,449,,0.92,New York Giants,,,108,107,NYI,NYP,NYP
+1890,PL,PHP,PHQ,,5,132,,68,63,,,N,,941,4855,1348,187,113,49,431,321,203,,,,855,519,4.0500000000,118,4,2,3462,1292,33,495,361,510,,0.91,Philadelphia Athletics,,,103,101,PHQ,PHP,PHP
+1890,PL,PTP,PBB,,6,128,,60,68,,,N,,835,4577,1192,168,113,35,569,375,249,,,,892,523,4.2200000000,121,7,0,3348,1267,32,334,318,512,,0.90,Pittsburgh Burghers,,,92,92,PBB,PTP,PTP
+1891,AA,BL3,BLO,,4,139,,71,64,,,N,,850,4771,1217,142,99,30,551,553,342,,,,798,464,3.4300000000,118,6,2,3651,1238,33,472,408,503,,0.91,Baltimore Orioles,Union Park,,101,100,BAL,BL3,BL3
+1891,AA,BS2,BRS,,1,139,,93,42,,,Y,,1028,4889,1341,163,100,52,651,499,447,,,,675,410,3.0300000000,108,9,7,3657,1158,42,497,524,392,,0.93,Boston Reds,Congress Street Grounds,,103,99,BOS,BS2,BS2
+1891,AA,CL6,CLS,,6,138,,61,76,,,N,,702,4697,1113,154,61,20,529,530,280,,,,777,505,3.7500000000,118,6,0,3639,1141,29,588,502,379,,0.93,Columbus Solons,Recreation Park II,,95,93,COL,CL6,CL6
+1891,AA,CN3,CKK,,7,102,,43,57,,,N,,549,3574,838,105,58,28,428,385,164,,,,643,344,3.4300000000,86,2,1,2706,921,20,446,331,389,,0.91,Cincinnati Kelly's Killers,East End Park,,108,110,CKK,CN3,CN3
+1891,AA,LS2,LOU,,8,141,,54,83,,,N,,713,4833,1247,130,69,17,443,473,230,,,,890,582,4.2700000000,128,9,1,3678,1353,33,464,485,458,,0.92,Louisville Colonels,Eclipse Park I,,97,97,LOU,LS2,LS2
+1891,AA,ML3,MLA,,3,36,,21,15,,,N,,227,1271,332,58,15,13,107,114,47,,,,156,86,2.5000000000,35,3,0,927,291,6,120,137,116,,0.92,Milwaukee Brewers,Athletic Field,,120,118,MIL,ML3,ML3
+1891,AA,PH4,PHQ,,5,143,,73,66,,,N,,817,5039,1301,182,123,55,447,548,149,,,,794,549,4.0100000000,135,3,0,3699,1274,35,520,533,389,,0.93,Philadelphia Athletics,Athletic Park,,103,103,PHA,PH4,PH4
+1891,AA,SL4,STL,,2,141,,85,51,,,N,,976,5005,1330,169,51,58,625,440,283,,,,753,444,3.2700000000,103,8,5,3666,1106,50,576,621,466,,0.92,St. Louis Browns,Sportsman's Park I,,108,108,STL,SL4,SL4
+1891,AA,WS9,WAS,,9,139,,44,91,,,N,,691,4715,1183,147,84,19,468,485,219,,,,1067,634,4.8300000000,123,2,2,3543,1420,44,566,486,589,,0.90,Washington Statesmen,Boundary Field,,95,100,WAS,WS9,WS9
+1891,NL,BRO,LAD,,6,137,,61,76,,,N,,765,4748,1233,200,69,23,464,435,337,,,,820,516,3.8600000000,121,8,3,3612,1272,40,459,407,432,,0.92,Brooklyn Grooms,Eastern Park,181477,99,99,BRO,BRO,BRO
+1891,NL,BSN,ATL,,1,140,,87,51,,,Y,,847,4956,1264,181,80,54,532,538,289,,,,658,381,2.7600000000,126,9,6,3723,1223,51,364,525,358,,0.93,Boston Beaneaters,South End Grounds II,184472,112,109,BSN,BSN,BSN
+1891,NL,CHN,CHC,,2,137,,82,53,,,N,,832,4873,1233,159,88,60,526,457,238,,,,730,470,3.4700000000,114,6,3,3660,1207,53,475,477,397,,0.93,Chicago Colts,South Side Park I,181431,106,105,CHC,CHN,CHN
+1891,NL,CIN,CIN,,7,138,,56,81,,,N,,646,4791,1158,148,90,40,414,439,244,,,,790,480,3.5500000000,125,6,1,3654,1234,40,465,393,409,,0.93,Cincinnati Reds,League Park I in Cincinnati,97500,101,101,CIN,CIN,CIN
+1891,NL,CL4,CLV,,5,141,,65,74,,,N,,835,5074,1294,183,88,22,519,464,242,,,,888,484,3.5000000000,118,1,3,3732,1371,24,466,400,485,,0.92,Cleveland Spiders,League Park I,132000,104,103,CLV,CL4,CL4
+1891,NL,NY1,SFG,,3,136,,71,61,,,N,,754,4833,1271,189,72,46,438,394,224,,,,711,400,2.9900000000,117,11,3,3612,1098,27,593,651,384,,0.93,New York Giants,Polo Grounds III,210568,96,96,NYG,NY1,NY1
+1891,NL,PHI,PHI,,4,138,,68,69,,,N,,756,4929,1244,180,51,21,482,412,232,,,,773,509,3.7300000000,105,3,5,3687,1280,29,505,343,443,,0.92,Philadelphia Phillies,Philadelphia Baseball Grounds,217282,103,102,PHI,PHI,PHI
+1891,NL,PIT,PIT,,8,137,,55,80,,,N,,679,4794,1148,148,71,29,427,503,205,,,,744,384,2.8900000000,122,7,2,3591,1160,31,465,446,475,,0.91,Pittsburgh Pirates,Exposition Park,128000,97,98,PIT,PIT,PIT
+1892,NL,BLN,BLO,,12,152,,46,101,,,N,,779,5296,1342,160,111,30,499,480,227,,,,1020,617,4.2800000000,131,2,2,3894,1537,51,536,437,584,,0.91,Baltimore Orioles,Union Park,93589,103,104,BLN,BLN,BLN
+1892,NL,BRO,LAD,,3,158,,95,59,,,N,,935,5485,1439,183,105,30,629,506,409,,,,733,507,3.2500000000,132,12,5,4215,1285,26,600,597,398,,0.94,Brooklyn Grooms,Eastern Park,183727,96,96,BRO,BRO,BRO
+1892,NL,BSN,ATL,,1,152,,102,48,,,Y,,862,5301,1324,203,51,34,526,488,338,,,,649,425,2.8600000000,142,15,1,4008,1156,41,460,509,454,,0.92,Boston Beaneaters,South End Grounds II,146421,109,107,BSN,BSN,BSN
+1892,NL,CHN,CHC,,7,147,,70,76,,,N,,635,5063,1188,149,92,26,427,482,233,,,,735,456,3.1600000000,133,6,1,3894,1269,35,424,518,424,,0.93,Chicago Colts,South Side Park I,109067,94,94,CHC,CHN,CHN
+1892,NL,CIN,CIN,,5,155,,82,68,,,N,,766,5349,1288,155,75,44,503,474,270,,,,731,485,3.1700000000,130,8,2,4131,1327,39,535,437,402,,0.93,Cincinnati Reds,League Park I in Cincinnati,196473,99,99,CIN,CIN,CIN
+1892,NL,CL4,CLV,,2,153,,93,56,,,N,,855,5412,1375,196,96,26,552,536,225,,,,613,358,2.4100000000,140,11,2,4008,1178,28,413,472,406,,0.93,Cleveland Spiders,League Park I,139928,104,103,CLV,CL4,CL4
+1892,NL,LS3,LOU,,9,154,,63,89,,,N,,649,5334,1208,133,61,18,433,508,275,,,,804,500,3.3400000000,147,9,0,4038,1358,26,447,430,469,,0.92,Louisville Colonels,Eclipse Park I,131159,91,93,LOU,LS3,LS3
+1892,NL,NY1,SFG,,8,153,,71,80,,,N,,811,5291,1326,173,85,39,510,469,301,,,,826,483,3.2900000000,139,5,1,3966,1165,32,635,641,565,,0.91,New York Giants,Polo Grounds III,130566,99,98,NYG,NY1,NY1
+1892,NL,PHI,PHI,,4,155,,87,66,,,N,,860,5413,1420,225,95,50,528,515,216,,,,690,449,2.9300000000,131,10,5,4137,1297,24,492,502,393,,0.93,Philadelphia Phillies,Philadelphia Baseball Grounds,193731,100,99,PHI,PHI,PHI
+1892,NL,PIT,PIT,,6,155,,80,73,,,N,,802,5469,1288,143,108,38,435,453,222,,,,796,464,3.1000000000,130,3,1,4041,1300,28,537,455,480,,0.92,Pittsburgh Pirates,Exposition Park,177205,100,100,PIT,PIT,PIT
+1892,NL,SLN,STL,,11,155,,56,94,,,N,,703,5259,1187,138,53,45,607,491,209,,,,922,627,4.2000000000,139,4,1,4032,1466,47,543,478,453,,0.92,St. Louis Browns,Sportsman's Park I,192442,105,104,STL,SLN,SLN
+1892,NL,WAS,WAS,,10,153,,58,93,,,N,,731,5204,1245,149,78,37,529,553,276,,,,869,506,3.4600000000,129,5,3,3945,1293,40,556,479,547,,0.91,Washington Senators,Boundary Field,128279,97,99,WHS,WSN,WSN
+1893,NL,BLN,BLO,,8,130,,60,70,,,N,,820,4651,1281,164,86,27,537,323,233,,,,893,620,4.9700000000,104,1,2,3369,1325,29,534,275,384,,0.92,Baltimore Orioles,Union Park,143000,102,102,BLN,BLN,BLN
+1893,NL,BRO,LAD,,7,130,,65,63,,,N,,775,4511,1200,173,83,45,473,296,213,,,,845,583,4.5500000000,109,3,3,3462,1262,41,547,297,385,,0.93,Brooklyn Grooms,Eastern Park,235000,95,95,BRO,BRO,BRO
+1893,NL,BSN,ATL,,1,131,,86,43,,,Y,,1008,4678,1358,178,50,65,561,292,243,,,,795,572,4.4300000000,114,2,2,3489,1314,66,402,253,353,,0.93,Boston Beaneaters,South End Grounds II,193300,108,106,BSN,BSN,BSN
+1893,NL,CHN,CHC,,9,128,,56,71,,,N,,829,4664,1299,186,93,32,465,262,255,,,,874,597,4.8100000000,101,4,5,3351,1278,26,553,273,421,,0.92,Chicago Colts,South Side Park I,223500,103,104,CHC,CHN,CHN
+1893,NL,CIN,CIN,,6,131,,65,63,,,N,,759,4617,1195,161,65,29,532,256,238,,,,814,593,4.5500000000,97,4,5,3516,1305,38,549,258,321,,0.94,Cincinnati Reds,League Park I in Cincinnati,194250,102,103,CIN,CIN,CIN
+1893,NL,CL4,CLV,,3,129,,73,55,,,N,,976,4747,1425,222,98,32,532,229,252,,,,839,532,4.2000000000,110,2,2,3420,1361,35,356,242,389,,0.93,Cleveland Spiders,League Park I,130000,106,105,CLV,CL4,CL4
+1893,NL,LS3,LOU,,11,126,,50,75,,,N,,759,4566,1185,177,73,19,485,306,203,,,,942,708,5.9000000000,113,4,1,3240,1431,38,479,190,328,,0.93,Louisville Colonels,Eclipse Park II,53683,92,94,LOU,LS3,LS3
+1893,NL,NY1,SFG,,5,136,,68,64,,,N,,941,4858,1424,182,101,61,504,279,299,,,,845,577,4.2900000000,111,6,4,3633,1271,36,581,395,432,,0.92,New York Giants,Polo Grounds III,290000,101,100,NYG,NY1,NY1
+1893,NL,PHI,PHI,,4,133,,72,57,,,N,,1011,5151,1553,246,90,80,468,335,202,,,,841,618,4.6800000000,107,4,2,3567,1359,30,521,283,318,,0.94,Philadelphia Phillies,Philadelphia Baseball Grounds,293019,102,101,PHI,PHI,PHI
+1893,NL,PIT,PIT,,2,131,,81,48,,,N,,970,4834,1447,176,127,37,537,273,210,,,,766,529,4.0800000000,104,8,2,3501,1232,29,504,280,346,,0.93,Pittsburgh Pirates,Exposition Park,184000,98,98,PIT,PIT,PIT
+1893,NL,SLN,STL,,10,135,,57,75,,,N,,745,4879,1288,152,98,10,524,251,250,,,,829,544,4.0600000000,114,3,4,3621,1292,38,542,301,398,,0.93,St. Louis Browns,Robison Field,195000,100,101,STL,SLN,SLN
+1893,NL,WAS,WAS,,12,130,,40,89,,,N,,722,4742,1260,180,83,23,524,237,154,,,,1032,704,5.5600000000,110,2,0,3417,1485,54,574,292,497,,0.91,Washington Senators,Boundary Field,90000,97,99,WHS,WSN,WSN
+1894,NL,BLN,BLO,,1,129,,89,39,,,Y,,1171,4799,1647,271,150,33,516,200,324,,,,819,620,5.0000000000,97,1,11,3348,1371,31,472,275,293,,0.94,Baltimore Orioles,Union Park,328000,104,103,BLN,BLN,BLN
+1894,NL,BRO,LAD,,5,134,,70,61,,,N,,1021,4816,1507,228,130,42,466,294,282,,,,1007,711,5.5100000000,105,3,5,3486,1447,41,555,285,390,,0.92,Brooklyn Grooms,Eastern Park,214000,93,93,BRO,BRO,BRO
+1894,NL,BSN,ATL,,3,133,,83,49,,,N,,1220,5011,1658,272,94,103,535,261,241,,,,1002,701,5.4100000000,108,3,1,3498,1529,89,411,262,415,,0.92,Boston Beaneaters,South End Grounds II / Congress Street Grounds / South End Grounds III,152800,113,111,BSN,BSN,BSN
+1894,NL,CHN,CHC,,8,135,,57,75,,,N,,1041,4960,1555,265,86,65,496,298,327,,,,1066,725,5.6800000000,117,0,0,3444,1561,43,557,281,452,,0.91,Chicago Colts,West Side Park II,239000,106,106,CHC,CHN,CHN
+1894,NL,CIN,CIN,,10,132,,55,75,,,N,,910,4671,1374,224,67,61,508,252,215,,,,1085,763,5.9900000000,110,4,3,3441,1585,85,491,219,420,,0.92,Cincinnati Reds,League Park II in Cincinnati,158000,103,104,CIN,CIN,CIN
+1894,NL,CL4,CLV,,6,130,,68,61,,,N,,932,4764,1442,241,90,37,471,301,220,,,,896,621,4.9700000000,106,6,1,3372,1390,54,435,254,344,,0.93,Cleveland Spiders,League Park I,82000,104,103,CLV,CL4,CL4
+1894,NL,LS3,LOU,,12,130,,36,94,,,N,,692,4482,1206,173,88,42,350,364,217,,,,1001,664,5.4500000000,113,2,1,3288,1462,39,475,258,428,,0.92,Louisville Colonels,Eclipse Park II,75000,93,96,LOU,LS3,LS3
+1894,NL,NY1,SFG,,2,137,,88,44,,,N,,940,4806,1446,197,96,43,476,217,319,,,,789,516,3.8300000000,111,5,5,3636,1292,37,539,395,443,,0.92,New York Giants,Polo Grounds III,387000,99,99,NYG,NY1,NY1
+1894,NL,PHI,PHI,,4,129,,71,57,,,N,,1143,4967,1732,252,131,40,496,245,273,,,,966,704,5.6300000000,102,3,4,3375,1482,62,469,262,338,,0.93,Philadelphia Phillies,Philadelphia Baseball Grounds,352773,96,94,PHI,PHI,PHI
+1894,NL,PIT,PIT,,7,132,,65,65,,,N,,955,4676,1458,222,124,48,434,208,256,,,,972,724,5.6000000000,106,2,0,3492,1552,39,457,304,354,,0.93,Pittsburgh Pirates,Exposition Park,159000,99,98,PIT,PIT,PIT
+1894,NL,SLN,STL,,9,133,,56,76,,,N,,771,4610,1320,171,113,54,442,289,190,,,,953,682,5.2900000000,114,2,0,3483,1418,48,500,319,426,,0.92,St. Louis Browns,Robison Field,155000,100,102,STL,SLN,SLN
+1894,NL,WAS,WAS,,11,132,,45,87,,,N,,882,4581,1317,218,118,59,617,375,249,,,,1122,678,5.5100000000,101,0,4,3321,1573,59,446,190,499,,0.90,Washington Senators,Boundary Field,125000,97,99,WHS,WSN,WSN
+1895,NL,BLN,BLO,,1,132,,87,43,,,Y,,1009,4725,1530,235,89,25,355,243,310,,,,646,479,3.8000000000,104,10,4,3402,1216,31,430,244,288,,0.94,Baltimore Orioles,Union Park,293000,103,100,BLN,BLN,BLN
+1895,NL,BRO,LAD,,5,133,,71,60,,,N,,867,4717,1330,189,77,39,397,318,183,,,,834,631,4.9400000000,103,5,6,3450,1360,41,395,216,325,,0.94,Brooklyn Grooms,Eastern Park,230000,92,92,BRO,BRO,BRO
+1895,NL,BSN,ATL,,6,132,,71,60,,,N,,907,4715,1369,197,57,54,500,236,199,,,,826,557,4.2700000000,116,4,4,3525,1364,56,363,370,364,,0.93,Boston Beaneaters,South End Grounds III,242000,105,104,BSN,BSN,BSN
+1895,NL,CHN,CHC,,4,133,,72,58,,,N,,866,4708,1401,171,85,55,422,344,260,,,,854,597,4.6700000000,119,3,1,3450,1422,38,432,297,401,,0.92,Chicago Colts,West Side Park II,382300,107,107,CHC,CHN,CHN
+1895,NL,CIN,CIN,,8,132,,66,64,,,N,,903,4684,1395,235,105,36,414,249,326,,,,854,613,4.8100000000,97,2,6,3441,1451,39,362,245,377,,0.93,Cincinnati Reds,League Park II in Cincinnati,281000,104,104,CIN,CIN,CIN
+1895,NL,CL4,CLV,,2,131,,84,46,,,N,,917,4658,1423,194,67,29,472,361,187,,,,720,497,3.9100000000,108,6,3,3429,1272,33,346,326,348,,0.93,Cleveland Spiders,League Park I,143000,106,104,CLV,CL4,CL4
+1895,NL,LS3,LOU,,12,133,,35,96,,,N,,698,4724,1320,171,73,34,346,323,156,,,,1090,732,5.9000000000,104,3,1,3351,1520,40,470,245,477,,0.91,Louisville Colonels,Eclipse Park II,92000,94,97,LOU,LS3,LS3
+1895,NL,NY1,SFG,,9,132,,66,65,,,N,,852,4605,1324,191,90,32,454,292,292,,,,834,575,4.5100000000,115,6,1,3441,1359,34,415,409,438,,0.92,New York Giants,Polo Grounds III,240000,98,97,NYG,NY1,NY1
+1895,NL,PHI,PHI,,3,133,,78,53,,,N,,1068,5037,1664,272,73,61,463,262,276,,,,957,706,5.4700000000,106,2,7,3483,1467,36,485,330,369,,0.93,Philadelphia Phillies,Baker Bowl,474971,100,100,PHI,PHI,PHI
+1895,NL,PIT,PIT,,7,134,,71,61,,,N,,811,4645,1349,190,89,26,376,299,257,,,,787,527,4.0500000000,106,4,6,3513,1263,17,500,382,392,,0.93,Pittsburgh Pirates,Exposition Park,188000,95,95,PIT,PIT,PIT
+1895,NL,SLN,STL,,11,135,,39,92,,,N,,747,4781,1344,155,88,38,384,279,205,,,,1032,737,5.7600000000,105,1,1,3456,1562,64,439,280,380,,0.93,St. Louis Browns,Robison Field,170000,99,101,STL,SLN,SLN
+1895,NL,WAS,WAS,,10,132,,43,85,,,N,,837,4577,1314,207,101,55,518,396,237,,,,1048,646,5.2800000000,99,0,5,3303,1507,55,465,258,444,,0.91,Washington Senators,Boundary Field,153000,99,101,WHS,WSN,WSN
+1896,NL,BLN,BLO,,1,132,,90,39,,,Y,,995,4719,1548,207,100,23,386,201,441,,,,662,476,3.6700000000,115,9,1,3504,1281,22,339,302,296,,0.94,Baltimore Orioles,Union Park,249448,102,98,BLN,BLN,BLN
+1896,NL,BRO,LAD,,10,133,,58,73,,,N,,692,4548,1292,174,87,28,344,269,198,,,,764,540,4.2500000000,97,3,1,3432,1353,39,400,259,297,,0.94,Brooklyn Bridegrooms,Eastern Park,201000,94,95,BRO,BRO,BRO
+1896,NL,BSN,ATL,,4,132,,74,57,,,N,,860,4717,1416,175,74,36,414,274,241,,,,761,485,3.7800000000,110,6,3,3465,1254,57,397,277,368,,0.93,Boston Beaneaters,South End Grounds III,240000,106,104,BSN,BSN,BSN
+1896,NL,CHN,CHC,,5,132,,71,57,,,N,,815,4582,1311,182,97,34,409,290,332,,,,799,569,4.4100000000,118,2,1,3483,1302,30,467,353,367,,0.93,Chicago Colts,West Side Park II,317500,104,104,CHC,CHN,CHN
+1896,NL,CIN,CIN,,3,128,,77,50,,,N,,783,4360,1283,204,73,20,382,226,350,,,,620,452,3.6700000000,105,12,4,3324,1240,27,310,219,252,,0.95,Cincinnati Reds,League Park II in Cincinnati,373000,107,106,CIN,CIN,CIN
+1896,NL,CL4,CLV,,2,135,,80,48,,,N,,840,4856,1463,207,72,28,436,316,175,,,,650,459,3.4600000000,113,9,5,3585,1363,27,280,336,280,,0.95,Cleveland Spiders,League Park I,152000,106,104,CLV,CL4,CL4
+1896,NL,LS3,LOU,,12,134,,38,93,,,N,,653,4588,1197,142,80,37,371,427,195,,,,997,653,5.1200000000,108,1,4,3444,1398,48,541,288,475,,0.91,Louisville Colonels,Eclipse Park II,133000,96,99,LOU,LS3,LS3
+1896,NL,NY1,SFG,,7,133,,64,67,,,N,,829,4661,1383,159,87,40,439,271,274,,,,821,573,4.5400000000,104,1,2,3408,1303,33,403,312,365,,0.93,New York Giants,Polo Grounds III,274000,97,96,NYG,NY1,NY1
+1896,NL,PHI,PHI,,8,130,,62,68,,,N,,890,4680,1382,234,84,49,438,297,191,,,,891,645,5.2000000000,107,3,2,3351,1473,39,387,243,313,,0.94,Philadelphia Phillies,Baker Bowl,357025,99,99,PHI,PHI,PHI
+1896,NL,PIT,PIT,,6,131,,66,63,,,N,,787,4701,1371,169,94,27,387,286,217,,,,741,554,4.3000000000,108,8,1,3477,1286,18,439,362,317,,0.94,Pittsburgh Pirates,Exposition Park,197000,96,96,PIT,PIT,PIT
+1896,NL,SLN,STL,,11,131,,40,90,,,N,,593,4520,1162,134,78,37,332,300,185,,,,929,669,5.3300000000,115,1,1,3390,1448,40,456,279,345,,0.93,St. Louis Browns,Robison Field,184000,96,100,STL,SLN,SLN
+1896,NL,WAS,WAS,,9,133,,58,73,,,N,,818,4639,1328,179,79,45,516,365,258,,,,920,582,4.6100000000,106,2,3,3408,1435,24,435,292,398,,0.92,Washington Senators,Boundary Field,223000,100,101,WHS,WSN,WSN
+1897,NL,BLN,BLO,,2,136,,90,40,,,N,,964,4872,1584,243,66,19,437,256,401,,,,674,472,3.5500000000,118,3,0,3591,1296,18,382,361,277,,0.95,Baltimore Orioles,Union Park,273046,100,97,BLN,BLN,BLN
+1897,NL,BRO,LAD,,7,136,,61,71,,,N,,802,4810,1343,202,72,24,351,255,187,,,,845,610,4.6000000000,114,4,2,3582,1417,34,410,256,364,,0.93,Brooklyn Bridegrooms,Eastern Park,220831,94,95,BRO,BRO,BRO
+1897,NL,BSN,ATL,,1,135,,93,39,,,Y,,1025,4937,1574,230,83,45,423,262,233,,,,665,484,3.6500000000,115,8,7,3582,1273,39,393,329,272,,0.95,Boston Beaneaters,South End Grounds III,334800,106,104,BSN,BSN,BSN
+1897,NL,CHN,CHC,,9,138,,59,73,,,N,,832,4803,1356,189,97,38,430,317,264,,,,894,602,4.5300000000,131,2,1,3591,1485,30,433,361,394,,0.93,Chicago Colts,West Side Park II,327160,104,104,CHC,CHN,CHN
+1897,NL,CIN,CIN,,4,134,,76,56,,,N,,763,4524,1311,219,69,22,380,218,194,,,,705,525,4.0900000000,100,4,2,3468,1375,18,329,270,273,,0.94,Cincinnati Reds,League Park II in Cincinnati,336800,107,106,CIN,CIN,CIN
+1897,NL,CL4,CLV,,5,132,,69,62,,,N,,773,4604,1374,192,88,16,435,344,181,,,,680,491,3.9500000000,111,6,0,3357,1297,32,289,277,257,,0.95,Cleveland Spiders,League Park I,115250,110,109,CLV,CL4,CL4
+1897,NL,LS3,LOU,,11,134,,52,78,,,N,,669,4520,1197,160,70,40,370,453,195,,,,859,559,4.4200000000,114,2,0,3414,1363,39,459,267,395,,0.92,Louisville Colonels,Eclipse Park II,145210,97,99,LOU,LS3,LS3
+1897,NL,NY1,SFG,,3,137,,83,48,,,N,,895,4844,1449,188,84,31,404,327,328,,,,695,458,3.4700000000,118,8,3,3561,1214,26,486,456,397,,0.93,New York Giants,Polo Grounds III,390340,97,96,NYG,NY1,NY1
+1897,NL,PHI,PHI,,10,134,,55,77,,,N,,752,4756,1392,213,83,40,399,299,163,,,,792,590,4.6000000000,115,4,2,3465,1415,28,364,253,296,,0.94,Philadelphia Phillies,Baker Bowl,290027,97,97,PHI,PHI,PHI
+1897,NL,PIT,PIT,,8,135,,60,71,,,N,,676,4590,1266,140,108,25,359,334,170,,,,835,598,4.6700000000,112,2,2,3459,1397,22,318,342,346,,0.93,Pittsburgh Pirates,Exposition Park,165950,96,97,PIT,PIT,PIT
+1897,NL,SLN,STL,,12,132,,29,102,,,N,,588,4642,1277,149,67,31,354,314,172,,,,1083,778,6.2100000000,109,1,1,3381,1584,54,453,207,375,,0.93,St. Louis Browns,Robison Field,136400,98,102,STL,SLN,SLN
+1897,NL,WAS,WAS,,6,135,,61,71,,,N,,781,4636,1376,194,77,36,374,348,208,,,,793,511,4.0100000000,102,7,6,3444,1383,27,400,348,369,,0.93,Washington Senators,Boundary Field,151028,99,101,WHS,WSN,WSN
+1898,NL,BLN,BLO,,2,154,,96,53,,,N,,933,5242,1584,154,77,12,519,316,250,,,,623,426,2.9000000000,138,12,0,3969,1236,17,400,422,326,,0.94,Baltimore Orioles,Union Park,123416,102,99,BLN,BLN,BLN
+1898,NL,BRO,LAD,,10,149,,54,91,,,N,,638,5126,1314,156,66,17,328,314,130,,,,811,578,4.0100000000,134,1,0,3894,1446,34,476,294,334,,0.94,Brooklyn Bridegrooms,Washington Park III,122514,100,99,BRO,BRO,BRO
+1898,NL,BSN,ATL,,1,152,,102,47,,,Y,,872,5276,1531,190,55,53,405,303,172,,,,614,444,2.9800000000,127,9,8,4020,1186,37,470,432,299,,0.95,Boston Beaneaters,South End Grounds III,229275,105,102,BSN,BSN,BSN
+1898,NL,CHN,CHC,,4,152,,85,65,,,N,,828,5219,1431,175,84,18,476,394,220,,,,679,422,2.8300000000,137,13,0,4026,1357,17,364,323,412,,0.93,Chicago Orphans,West Side Park II,424352,100,99,CHC,CHN,CHN
+1898,NL,CIN,CIN,,3,157,,92,60,,,N,,831,5334,1448,207,101,19,455,300,165,,,,740,539,3.5000000000,131,10,2,4155,1484,16,449,294,325,,0.95,Cincinnati Reds,League Park II in Cincinnati,336378,107,106,CIN,CIN,CIN
+1898,NL,CL4,CLV,,5,156,,81,68,,,N,,730,5246,1379,162,56,18,545,306,93,,,,683,474,3.2000000000,142,9,0,4002,1429,26,309,339,296,,0.95,Cleveland Spiders,League Park I,70496,96,95,CLV,CL4,CL4
+1898,NL,LS3,LOU,,9,154,,70,81,,,N,,728,5193,1389,150,71,32,375,429,235,,,,833,628,4.2400000000,137,4,0,4002,1457,33,470,271,382,,0.93,Louisville Colonels,Eclipse Park II,128980,99,99,LOU,LS3,LS3
+1898,NL,NY1,SFG,,7,157,,77,73,,,N,,837,5349,1422,190,86,34,428,372,214,,,,800,517,3.4400000000,141,9,1,4059,1359,21,587,558,447,,0.93,New York Giants,Polo Grounds III,265414,97,96,NYG,NY1,NY1
+1898,NL,PHI,PHI,,6,150,,78,71,,,N,,823,5118,1431,238,81,33,472,382,182,,,,784,532,3.7200000000,129,10,0,3864,1440,23,399,325,379,,0.93,Philadelphia Phillies,Baker Bowl,265414,96,95,PHI,PHI,PHI
+1898,NL,PIT,PIT,,8,152,,72,76,,,N,,634,5087,1313,140,88,14,336,343,107,,,,694,501,3.4100000000,131,10,3,3969,1400,14,346,330,340,,0.94,Pittsburgh Pirates,Exposition Park,150900,98,99,PIT,PIT,PIT
+1898,NL,SLN,STL,,12,154,,39,111,,,N,,571,5214,1290,149,55,13,383,402,104,,,,929,666,4.5300000000,133,0,2,3972,1584,32,372,288,388,,0.93,St. Louis Browns,Robison Field,151700,102,105,STL,SLN,SLN
+1898,NL,WAS,WAS,,11,155,,51,101,,,N,,704,5257,1423,177,80,36,370,386,197,,,,939,656,4.5200000000,129,0,1,3921,1577,29,450,371,442,,0.92,Washington Senators,Boundary Field,103250,100,101,WHS,WSN,WSN
+1899,NL,BLN,BLO,,4,152,,86,62,,,N,,827,5073,1509,204,71,17,418,383,364,,,,691,480,3.3100000000,132,10,4,3912,1403,13,349,294,308,,0.94,Baltimore Orioles,Union Park,121935,105,103,BLN,BLN,BLN
+1899,NL,BRO,LAD,,1,150,,101,47,,,Y,,892,4937,1436,178,97,27,477,263,271,,,,658,458,3.2500000000,121,9,9,3807,1320,32,463,331,314,,0.94,Brooklyn Superbas,Washington Park III,269641,102,102,BRO,BRO,BRO
+1899,NL,BSN,ATL,,2,153,,95,57,,,N,,858,5290,1517,178,90,39,431,269,185,,,,645,488,3.2600000000,138,13,4,4044,1273,44,432,385,295,,0.95,Boston Beaneaters,South End Grounds III,200384,109,108,BSN,BSN,BSN
+1899,NL,CHN,CHC,,8,152,,75,73,,,N,,812,5148,1428,173,82,27,406,342,247,,,,763,498,3.3700000000,147,8,1,3993,1433,20,330,313,428,,0.93,Chicago Orphans,West Side Park II,352130,97,97,CHC,CHN,CHN
+1899,NL,CIN,CIN,,6,156,,83,67,,,N,,856,5225,1439,194,105,13,485,295,228,,,,770,560,3.7000000000,130,8,5,4083,1484,26,370,360,339,,0.94,Cincinnati Reds,League Park II in Cincinnati,259536,102,102,CIN,CIN,CIN
+1899,NL,CL4,CLV,,12,154,,20,134,,,N,,529,5279,1333,142,50,12,289,280,127,,,,1252,895,6.3700000000,138,0,0,3792,1844,43,527,215,388,,0.93,Cleveland Spiders,League Park I,6088,89,96,CLV,CL4,CL4
+1899,NL,LS3,LOU,,9,155,,75,77,,,N,,827,5307,1484,192,68,40,436,375,233,,,,775,518,3.4500000000,134,5,2,4053,1509,33,323,287,382,,0.93,Louisville Colonels,Eclipse Park II,109319,100,100,LOU,LS3,LS3
+1899,NL,NY1,SFG,,10,152,,60,90,,,N,,734,5092,1431,161,65,23,387,360,234,,,,863,609,4.2900000000,138,4,0,3834,1454,19,628,397,433,,0.93,New York Giants,Polo Grounds III,121384,97,97,NYG,NY1,NY1
+1899,NL,PHI,PHI,,3,154,,94,58,,,N,,916,5353,1613,241,83,31,441,341,212,,,,743,514,3.4700000000,129,15,2,3999,1398,17,370,281,379,,0.94,Philadelphia Phillies,Baker Bowl,388933,97,96,PHI,PHI,PHI
+1899,NL,PIT,PIT,,7,154,,76,73,,,N,,834,5450,1574,196,122,27,384,345,179,,,,765,546,3.6000000000,117,9,4,4092,1464,27,437,334,361,,0.94,Pittsburgh Pirates,Exposition Park,251834,100,99,PIT,PIT,PIT
+1899,NL,SLN,STL,,5,155,,84,67,,,N,,819,5304,1514,172,88,47,468,262,210,,,,739,500,3.3600000000,134,7,1,4020,1476,41,321,331,399,,0.93,St. Louis Perfectos,Robison Field,373909,102,103,STL,SLN,SLN
+1899,NL,WAS,WAS,,11,155,,54,98,,,N,,743,5256,1429,162,87,47,350,341,176,,,,983,712,4.9300000000,131,3,0,3900,1649,35,422,328,403,,0.93,Washington Senators,Boundary Field,86392,99,101,WHS,WSN,WSN
+1900,NL,BRO,LAD,,1,142,,82,54,,,Y,,816,4860,1423,199,81,26,421,272,274,,,,722,529,3.8900000000,104,8,4,3675,1370,30,405,300,303,102,0.94,Brooklyn Superbas,Washington Park III,183000,106,104,BRO,BRO,BRO
+1900,NL,BSN,ATL,,4,142,,66,72,,,N,,778,4952,1403,163,68,48,395,278,182,,,,739,513,3.7200000000,116,8,2,3720,1263,59,463,340,271,86,0.95,Boston Beaneaters,South End Grounds III,202000,112,111,BSN,BSN,BSN
+1900,NL,CHN,CHC,,6,146,,65,75,,,N,,635,4907,1276,202,51,33,343,383,189,,,,751,456,3.2300000000,137,9,1,3813,1375,21,324,357,418,98,0.93,Chicago Orphans,West Side Park II,248577,96,98,CHC,CHN,CHN
+1900,NL,CIN,CIN,,7,144,,62,77,,,N,,703,5026,1335,178,83,33,333,408,183,,,,745,542,3.8300000000,118,9,1,3822,1383,28,404,399,341,120,0.94,Cincinnati Reds,League Park II in Cincinnati,170000,98,99,CIN,CIN,CIN
+1900,NL,NY1,SFG,,8,141,,60,78,,,N,,713,4724,1317,177,61,23,369,343,236,,,,823,531,3.9600000000,113,4,0,3621,1423,26,442,277,439,124,0.92,New York Giants,Polo Grounds III,190000,95,98,NYG,NY1,NY1
+1900,NL,PHI,PHI,,3,141,,75,63,,,N,,810,4969,1439,187,82,29,440,374,205,,,,792,571,4.1200000000,116,7,3,3744,1506,29,402,284,330,125,0.94,Philadelphia Phillies,Baker Bowl,301913,99,98,PHI,PHI,PHI
+1900,NL,PIT,PIT,,2,140,,79,60,,,N,,733,4817,1312,185,100,26,327,321,174,,,,612,418,3.0600000000,114,11,1,3687,1232,24,295,415,322,106,0.94,Pittsburgh Pirates,Exposition Park,264000,101,98,PIT,PIT,PIT
+1900,NL,SLN,STL,,5,142,,65,75,,,N,,744,4877,1420,141,81,36,406,318,243,,,,748,507,3.7500000000,117,12,0,3651,1373,37,299,325,331,73,0.94,St. Louis Cardinals,Robison Field,270000,99,98,STL,SLN,SLN
+1901,AL,BLA,NYY,,5,135,66,68,65,,,N,,760,4589,1348,179,111,24,369,377,207,,,,750,480,3.7300000000,115,4,3,3474,1313,21,344,271,401,76,0.92,Baltimore Orioles,Oriole Park,141952,104,105,BLA,BLA,BLA
+1901,AL,BOS,BOS,,2,138,69,79,57,,,N,,759,4866,1353,183,104,37,331,282,157,,,,608,411,3.0400000000,123,7,1,3651,1178,33,294,396,329,104,0.94,Boston Americans,Huntington Avenue Grounds,289448,98,96,BOS,BOS,BOS
+1901,AL,CHA,CHW,,1,137,71,83,53,,,Y,,819,4725,1303,173,89,32,475,337,280,,,,631,403,2.9800000000,110,11,2,3654,1250,27,312,394,345,100,0.94,Chicago White Sox,South Side Park II,354350,97,95,CHW,CHA,CHA
+1901,AL,CLE,CLE,,7,138,69,54,82,,,N,,667,4833,1311,197,68,12,243,326,125,,,,831,541,4.1200000000,122,7,4,3546,1365,22,464,334,329,99,0.94,Cleveland Blues,League Park I,131380,96,97,CLE,CLE,CLE
+1901,AL,DET,DET,,3,136,70,74,61,,,N,,741,4676,1303,180,80,29,380,346,204,,,,694,436,3.3000000000,118,8,2,3564,1328,22,313,307,410,127,0.93,Detroit Tigers,Bennett Park,259430,105,105,DET,DET,DET
+1901,AL,MLA,BAL,,8,139,70,48,89,,,N,,641,4795,1250,192,66,26,325,384,176,,,,828,549,4.0600000000,107,3,4,3654,1383,32,395,376,393,106,0.93,Milwaukee Brewers,Lloyd Street Grounds,139034,95,98,MLA,MLA,MLA
+1901,AL,PHA,OAK,,4,137,66,74,62,,,N,,805,4882,1409,239,87,35,301,344,173,,,,761,533,4.0000000000,124,6,1,3600,1346,20,374,350,331,93,0.94,Philadelphia Athletics,Columbia Park,206329,104,103,PHA,PHA,PHA
+1901,AL,WS1,MIN,,6,138,68,61,72,,,N,,682,4772,1282,191,83,33,356,340,127,,,,771,538,4.0900000000,118,8,2,3549,1396,51,284,308,323,97,0.94,Washington Senators,American League Park I,161661,98,100,WSH,WS1,WS1
+1901,NL,BRO,LAD,,3,137,68,79,57,,,N,,744,4879,1399,206,93,32,312,449,178,,,,600,423,3.1400000000,111,7,5,3639,1244,18,435,583,281,99,0.95,Brooklyn Superbas,Washington Park III,198200,103,101,BRO,BRO,BRO
+1901,NL,BSN,ATL,,5,140,70,69,69,,,N,,531,4746,1180,135,36,28,303,519,158,,,,556,407,2.9000000000,128,11,0,3789,1196,29,349,558,279,89,0.95,Boston Beaneaters,South End Grounds III,146502,109,109,BSN,BSN,BSN
+1901,NL,CHN,CHC,,6,140,70,53,86,,,N,,578,4844,1250,153,61,18,314,532,204,,,,699,459,3.3300000000,131,2,0,3723,1348,27,324,586,336,87,0.94,Chicago Orphans,West Side Park II,205071,96,98,CHC,CHN,CHN
+1901,NL,CIN,CIN,,8,142,72,52,87,,,N,,561,4914,1232,173,70,38,323,584,137,,,,818,586,4.1700000000,126,4,0,3795,1469,51,365,542,355,102,0.94,Cincinnati Reds,League Park II in Cincinnati,205728,94,96,CIN,CIN,CIN
+1901,NL,NY1,SFG,,7,141,71,52,85,,,N,,544,4839,1225,167,46,19,303,575,133,,,,755,530,3.8700000000,118,11,1,3696,1389,24,377,542,343,81,0.94,New York Giants,Polo Grounds III,297650,96,100,NYG,NY1,NY1
+1901,NL,PHI,PHI,,2,140,69,83,57,,,N,,668,4793,1275,194,58,24,430,549,199,,,,543,397,2.8700000000,125,15,2,3738,1221,19,259,480,262,65,0.95,Philadelphia Phillies,Baker Bowl,234937,102,102,PHI,PHI,PHI
+1901,NL,PIT,PIT,,1,140,69,90,49,,,Y,,776,4913,1407,182,92,29,386,493,203,,,,534,357,2.5800000000,119,15,4,3732,1198,20,244,505,285,97,0.95,Pittsburgh Pirates,Exposition Park,251955,103,98,PIT,PIT,PIT
+1901,NL,SLN,STL,,4,142,72,76,64,,,N,,792,5039,1430,187,94,39,314,540,190,,,,689,519,3.6800000000,118,5,5,3807,1333,39,332,445,305,108,0.94,St. Louis Cardinals,Robison Field,379988,95,96,STL,SLN,SLN
+1902,AL,BLA,NYY,,8,141,64,50,88,,,N,,715,4760,1318,202,107,33,417,429,189,,,,848,582,4.3300000000,119,3,1,3630,1531,30,354,258,347,109,0.93,Baltimore Orioles,Oriole Park,174606,104,105,BLA,BLA,BLA
+1902,AL,BOS,BOS,,3,138,71,77,60,,,N,,664,4875,1356,195,95,42,275,375,132,,,,600,415,3.0200000000,123,6,1,3714,1217,27,326,431,255,101,0.95,Boston Americans,Huntington Avenue Grounds,348567,103,100,BOS,BOS,BOS
+1902,AL,CHA,CHW,,4,138,72,74,60,,,N,,675,4654,1248,170,50,14,411,381,265,,,,602,463,3.4100000000,116,11,0,3663,1269,30,331,346,259,125,0.95,Chicago White Sox,South Side Park II,337898,96,95,CHW,CHA,CHA
+1902,AL,CLE,CLE,,5,137,65,69,67,,,N,,686,4840,1401,248,68,33,308,356,140,,,,667,439,3.2800000000,116,16,3,3612,1199,26,411,361,287,96,0.95,Cleveland Bronchos,League Park I,275395,96,96,CLE,CLE,CLE
+1902,AL,DET,DET,,7,137,67,52,83,,,N,,566,4644,1167,141,55,22,359,287,130,,,,657,471,3.5600000000,116,9,3,3570,1267,20,370,245,332,111,0.94,Detroit Tigers,Bennett Park,189469,105,105,DET,DET,DET
+1902,AL,PHA,OAK,,1,137,73,83,53,,,Y,,775,4762,1369,235,67,38,343,293,201,,,,636,445,3.2900000000,114,5,2,3648,1292,33,368,455,266,75,0.95,Philadelphia Athletics,Columbia Park,420078,104,103,PHA,PHA,PHA
+1902,AL,SLA,BAL,,2,140,73,78,58,,,N,,619,4736,1254,208,61,29,373,327,137,,,,607,462,3.3400000000,120,7,2,3732,1273,36,343,348,274,122,0.95,St. Louis Browns,Sportsman's Park III,272283,99,99,SLB,SLA,SLA
+1902,AL,WS1,MIN,,6,138,68,61,75,,,N,,707,4734,1338,262,66,47,329,296,121,,,,790,585,4.3600000000,130,2,1,3621,1403,56,312,300,316,70,0.94,Washington Senators,American League Park I,188158,101,104,WSH,WS1,WS1
+1902,NL,BRO,LAD,,2,141,69,75,63,,,N,,564,4845,1242,147,49,19,319,489,145,,,,519,375,2.6900000000,132,14,3,3768,1113,10,363,536,275,79,0.95,Brooklyn Superbas,Washington Park III,199868,101,99,BRO,BRO,BRO
+1902,NL,BSN,ATL,,3,142,72,73,64,,,N,,572,4728,1178,142,39,14,398,481,189,,,,516,365,2.6100000000,124,14,4,3777,1233,16,372,523,236,90,0.95,Boston Beaneaters,South End Grounds III,116960,101,102,BSN,BSN,BSN
+1902,NL,CHN,CHC,,5,141,70,68,69,,,N,,530,4802,1200,131,40,6,353,565,222,,,,501,313,2.2100000000,132,17,2,3825,1235,9,279,437,311,111,0.94,Chicago Orphans,West Side Park II,263700,97,97,CHC,CHN,CHN
+1902,NL,CIN,CIN,,4,141,70,70,70,,,N,,633,4908,1383,188,77,18,297,465,131,,,,566,368,2.6700000000,130,9,1,3717,1228,15,352,430,322,118,0.94,Cincinnati Reds,Palace of the Fans,217300,109,108,CIN,CIN,CIN
+1902,NL,NY1,SFG,,8,139,71,48,88,,,N,,401,4571,1088,147,34,8,252,530,187,,,,590,384,2.8200000000,118,11,1,3678,1193,16,332,501,330,104,0.94,New York Giants,Polo Grounds III,302875,99,101,NYG,NY1,NY1
+1902,NL,PHI,PHI,,7,138,69,56,81,,,N,,484,4615,1139,110,43,5,356,481,108,,,,649,471,3.5000000000,118,8,3,3633,1323,12,334,504,305,81,0.94,Philadelphia Phillies,Baker Bowl,112066,104,105,PHI,PHI,PHI
+1902,NL,PIT,PIT,,1,142,71,103,36,,,Y,,775,4926,1410,189,95,18,372,446,222,,,,440,323,2.3000000000,131,21,3,3792,1142,4,250,564,247,87,0.95,Pittsburgh Pirates,Exposition Park,243826,104,99,PIT,PIT,PIT
+1902,NL,SLN,STL,,6,140,70,56,78,,,N,,517,4751,1226,116,37,10,273,438,158,,,,695,473,3.4700000000,112,7,4,3681,1399,16,338,400,323,107,0.94,St. Louis Cardinals,Robison Field,226417,96,99,STL,SLN,SLN
+1903,AL,BOS,BOS,,1,141,70,91,47,,,Y,Y,708,4919,1336,222,113,48,262,561,141,,,,504,358,2.5700000000,123,20,4,3765,1142,23,269,579,235,86,0.95,Boston Americans,Huntington Avenue Grounds,379338,105,102,BOS,BOS,BOS
+1903,AL,CHA,CHW,,7,138,70,60,77,,,N,N,516,4670,1152,176,49,14,325,537,180,,,,613,414,3.0200000000,114,9,4,3705,1233,23,287,391,297,85,0.94,Chicago White Sox,South Side Park II,286183,96,95,CHW,CHA,CHA
+1903,AL,CLE,CLE,,3,140,74,77,63,,,N,N,639,4773,1265,231,95,31,259,595,175,,,,579,377,2.7300000000,125,20,1,3729,1161,16,271,521,322,99,0.94,Cleveland Naps,League Park I,311280,98,96,CLE,CLE,CLE
+1903,AL,DET,DET,,5,137,65,65,71,,,N,N,567,4582,1229,162,91,12,292,526,128,,,,539,365,2.7500000000,123,15,2,3588,1169,19,336,554,281,82,0.95,Detroit Tigers,Bennett Park,224523,96,97,DET,DET,DET
+1903,AL,NYA,NYY,,4,136,67,72,62,,,N,N,579,4565,1136,193,62,18,332,465,160,,,,573,411,3.0800000000,111,7,2,3603,1171,19,245,463,264,87,0.95,New York Highlanders,Hilltop Park,211808,106,105,NYY,NYA,NYA
+1903,AL,PHA,OAK,,2,137,67,75,60,,,N,N,597,4673,1236,227,68,32,268,513,157,,,,519,400,2.9800000000,112,10,1,3621,1124,20,315,728,217,66,0.96,Philadelphia Athletics,Columbia Park,422473,105,103,PHA,PHA,PHA
+1903,AL,SLA,BAL,,6,139,70,65,74,,,N,N,500,4639,1133,166,68,12,271,539,101,,,,525,376,2.7700000000,124,12,3,3666,1220,26,237,511,268,94,0.95,St. Louis Browns,Sportsman's Park III,380405,97,98,SLB,SLA,SLA
+1903,AL,WS1,MIN,,8,140,71,43,94,,,N,N,437,4613,1066,172,72,17,257,463,131,,,,691,519,3.8200000000,122,6,3,3669,1333,38,306,452,262,86,0.95,Washington Senators,American League Park I,128878,102,105,WSH,WS1,WS1
+1903,NL,BRO,LAD,,5,139,73,70,66,,,N,N,667,4534,1201,177,56,15,522,,273,,,,682,467,3.4400000000,118,11,4,3663,1276,18,377,438,284,98,0.95,Brooklyn Superbas,Washington Park III,224670,97,98,BRO,BRO,BRO
+1903,NL,BSN,ATL,,6,140,68,58,80,,,N,N,578,4682,1145,176,47,25,398,,159,,,,699,456,3.3400000000,125,8,1,3684,1310,30,460,516,351,89,0.93,Boston Beaneaters,South End Grounds III,143155,96,98,BSN,BSN,BSN
+1903,NL,CHN,CHC,,3,139,73,82,56,,,N,N,695,4733,1300,191,62,9,422,,259,,,,599,382,2.7700000000,117,6,6,3720,1182,14,354,451,314,78,0.94,Chicago Cubs,West Side Park II,386205,97,96,CHC,CHN,CHN
+1903,NL,CIN,CIN,,4,141,76,74,65,,,N,N,765,4857,1399,228,92,28,403,,144,,,,656,420,3.0700000000,126,11,1,3690,1277,14,378,480,312,84,0.94,Cincinnati Reds,Palace of the Fans,351680,111,109,CIN,CIN,CIN
+1903,NL,NY1,SFG,,2,142,70,84,55,,,N,N,729,4741,1290,181,49,20,379,,264,,,,567,414,2.9500000000,115,8,8,3786,1257,20,371,628,287,87,0.95,New York Giants,Polo Grounds III,579530,104,102,NYG,NY1,NY1
+1903,NL,PHI,PHI,,7,139,61,49,86,,,N,N,617,4781,1283,186,62,12,338,,120,,,,738,533,3.9600000000,126,5,3,3636,1347,21,425,381,299,76,0.94,Philadelphia Phillies,Baker Bowl,151729,92,95,PHI,PHI,PHI
+1903,NL,PIT,PIT,,1,141,70,91,49,,,Y,N,793,4988,1429,208,110,34,364,,172,,,,613,404,2.9100000000,117,16,5,3753,1215,9,384,454,289,100,0.95,Pittsburgh Pirates,Exposition Park,326855,103,99,PIT,PIT,PIT
+1903,NL,SLN,STL,,8,139,69,43,94,,,N,N,505,4689,1176,138,65,8,277,,171,,,,795,494,3.6700000000,111,4,2,3636,1353,25,430,419,321,111,0.94,St. Louis Cardinals,Robison Field,226538,97,100,STL,SLN,SLN
+1904,AL,BOS,BOS,,1,157,81,95,59,,,Y,,608,5231,1294,194,105,26,347,570,101,,,,466,331,2.1200000000,148,21,1,4218,1208,31,233,612,236,83,0.96,Boston Americans,Huntington Avenue Grounds,623295,106,103,BOS,BOS,BOS
+1904,AL,CHA,CHW,,3,156,78,89,65,,,N,,600,5027,1217,193,68,14,373,586,216,,,,482,353,2.3000000000,134,26,3,4140,1161,13,303,550,238,95,0.96,Chicago White Sox,South Side Park II,557123,96,95,CHW,CHA,CHA
+1904,AL,CLE,CLE,,4,154,78,86,65,,,N,,647,5152,1340,225,90,27,307,714,178,,,,482,334,2.2200000000,141,20,0,4068,1273,10,285,627,255,86,0.95,Cleveland Naps,League Park I,264749,99,98,CLE,CLE,CLE
+1904,AL,DET,DET,,7,162,79,62,90,,,N,,505,5321,1231,154,69,11,344,635,112,,,,627,440,2.7700000000,143,15,2,4290,1345,16,433,556,273,92,0.95,Detroit Tigers,Bennett Park,177796,97,98,DET,DET,DET
+1904,AL,NYA,NYY,,2,155,75,92,59,,,N,,598,5220,1354,195,91,27,312,548,163,,,,526,394,2.5700000000,123,15,1,4140,1180,29,311,684,275,90,0.95,New York Highlanders,Hilltop Park,438919,105,104,NYY,NYA,NYA
+1904,AL,PHA,OAK,,5,155,79,81,70,,,N,,557,5088,1266,197,77,31,313,605,137,,,,503,355,2.3500000000,136,26,0,4083,1149,13,366,887,250,67,0.95,Philadelphia Athletics,Columbia Park,512294,105,103,PHA,PHA,PHA
+1904,AL,SLA,BAL,,6,156,78,65,87,,,N,,481,5291,1266,153,53,10,332,609,150,,,,604,443,2.8300000000,135,13,1,4230,1335,25,333,577,267,78,0.96,St. Louis Browns,Sportsman's Park III,318108,94,96,SLB,SLA,SLA
+1904,AL,WS1,MIN,,8,157,78,38,113,,,N,,437,5149,1170,171,57,10,283,759,150,,,,743,547,3.6200000000,137,7,4,4077,1487,19,347,533,313,97,0.95,Washington Senators,American League Park II,131744,98,102,WSH,WS1,WS1
+1904,NL,BRO,LAD,,6,154,76,56,97,,,N,,497,4917,1142,159,53,15,411,,205,,,,614,401,2.7000000000,135,12,2,4011,1281,27,414,453,343,87,0.94,Brooklyn Superbas,Washington Park III,214600,97,100,BRO,BRO,BRO
+1904,NL,BSN,ATL,,7,155,79,55,98,,,N,,491,5135,1217,153,50,24,316,,143,,,,749,514,3.4300000000,136,13,0,4044,1405,25,500,544,338,91,0.94,Boston Beaneaters,South End Grounds III,140694,96,101,BSN,BSN,BSN
+1904,NL,CHN,CHC,,2,156,78,93,60,,,N,,599,5210,1294,157,62,22,298,,227,,,,517,353,2.3000000000,139,18,6,4149,1150,16,402,618,282,89,0.95,Chicago Cubs,West Side Park II,439100,100,98,CHC,CHN,CHN
+1904,NL,CIN,CIN,,3,157,79,88,65,,,N,,695,5231,1332,189,92,21,399,,179,,,,547,362,2.3400000000,142,12,2,4176,1256,13,343,502,301,81,0.95,Cincinnati Reds,Palace of the Fans,391915,109,107,CIN,CIN,CIN
+1904,NL,NY1,SFG,,1,158,84,106,47,,,Y,,744,5150,1347,202,65,31,434,,283,,,,476,337,2.1700000000,127,21,15,4188,1151,36,349,707,289,93,0.95,New York Giants,Polo Grounds III,609826,104,100,NYG,NY1,NY1
+1904,NL,PHI,PHI,,8,155,73,52,100,,,N,,571,5103,1268,170,54,23,377,,159,,,,784,504,3.3900000000,131,10,2,4017,1418,22,425,469,401,93,0.93,Philadelphia Phillies,Baker Bowl,140771,99,100,PHI,PHI,PHI
+1904,NL,PIT,PIT,,4,156,78,87,66,,,N,,675,5160,1333,164,102,15,391,,178,,,,592,433,2.8900000000,133,15,1,4044,1273,13,379,455,291,93,0.95,Pittsburgh Pirates,Exposition Park,340615,103,100,PIT,PIT,PIT
+1904,NL,SLN,STL,,5,155,76,75,79,,,N,,602,5104,1292,175,66,24,343,,199,,,,595,401,2.6400000000,146,7,2,4104,1286,23,319,529,307,83,0.95,St. Louis Cardinals,Robison Field,386750,96,99,STL,SLN,SLN
+1905,AL,BOS,BOS,,4,153,77,78,74,,,N,N,579,5049,1179,165,69,29,486,,131,,,,564,428,2.8400000000,124,14,1,4068,1198,33,292,652,290,75,0.95,Boston Americans,Huntington Avenue Grounds,468828,101,102,BOS,BOS,BOS
+1905,AL,CHA,CHW,,2,158,82,92,60,,,N,N,612,5114,1213,200,55,11,439,,194,,,,451,316,1.9900000000,131,15,0,4281,1163,11,329,613,217,64,0.96,Chicago White Sox,South Side Park II,687419,96,93,CHW,CHA,CHA
+1905,AL,CLE,CLE,,5,155,77,76,78,,,N,N,567,5166,1318,211,72,18,286,,188,,,,587,432,2.8500000000,140,16,0,4089,1251,23,334,555,233,84,0.96,Cleveland Naps,League Park I,316306,101,99,CLE,CLE,CLE
+1905,AL,DET,DET,,3,154,76,79,74,,,N,N,512,4971,1209,190,54,13,375,,129,,,,602,424,2.8300000000,124,17,1,4044,1226,11,474,578,264,80,0.95,Detroit Tigers,Bennett Park,193384,101,103,DET,DET,DET
+1905,AL,NYA,NYY,,6,152,75,71,78,,,N,N,586,4957,1228,163,61,23,360,,200,,,,622,440,2.9300000000,88,16,4,4059,1235,26,396,642,293,88,0.95,New York Highlanders,Hilltop Park,309100,111,110,NYY,NYA,NYA
+1905,AL,PHA,OAK,,1,152,74,92,56,,,Y,N,623,5146,1310,256,51,24,376,,190,,,,492,337,2.1900000000,117,19,0,4149,1137,21,409,895,265,95,0.95,Philadelphia Athletics,Columbia Park,554576,102,100,PHA,PHA,PHA
+1905,AL,SLA,BAL,,8,156,79,54,99,,,N,N,511,5204,1205,153,49,16,362,,130,,,,608,421,2.7400000000,134,10,2,4152,1245,19,389,633,284,78,0.95,St. Louis Browns,Sportsman's Park III,339112,95,96,SLB,SLA,SLA
+1905,AL,WS1,MIN,,7,154,77,64,87,,,N,N,559,5015,1121,193,68,22,298,,169,,,,623,434,2.8700000000,118,12,1,4086,1250,12,385,539,319,76,0.95,Washington Senators,American League Park II,252027,96,100,WSH,WS1,WS1
+1905,NL,BRO,LAD,,8,155,77,48,104,,,N,N,506,5100,1255,154,60,29,327,,186,,,,807,563,3.7600000000,125,7,3,4041,1416,24,476,556,408,101,0.93,Brooklyn Superbas,Washington Park III,227924,93,96,BRO,BRO,BRO
+1905,NL,BSN,ATL,,7,156,76,51,103,,,N,N,468,5190,1217,148,52,17,302,,132,,,,733,541,3.5200000000,139,14,0,4149,1390,36,433,533,311,89,0.95,Boston Beaneaters,South End Grounds III,150003,98,103,BSN,BSN,BSN
+1905,NL,CHN,CHC,,3,155,81,92,61,,,N,N,667,5108,1249,157,82,12,448,,267,,,,442,319,2.0400000000,133,23,2,4221,1135,14,385,627,246,99,0.96,Chicago Cubs,West Side Park II,509900,104,99,CHC,CHN,CHN
+1905,NL,CIN,CIN,,5,155,79,79,74,,,N,N,735,5205,1401,160,101,27,434,,181,,,,698,457,3.0100000000,119,10,2,4095,1409,22,439,547,310,122,0.95,Cincinnati Reds,Palace of the Fans,313927,111,110,CIN,CIN,CIN
+1905,NL,NY1,SFG,,1,155,76,105,48,,,Y,Y,780,5094,1392,191,88,39,517,,291,,,,505,364,2.3900000000,117,18,15,4110,1160,25,364,760,253,93,0.96,New York Giants,Polo Grounds III,552700,102,98,NYG,NY1,NY1
+1905,NL,PHI,PHI,,4,155,76,83,69,,,N,N,708,5243,1362,187,82,16,406,,180,,,,602,436,2.8100000000,119,12,5,4194,1303,21,411,516,275,99,0.95,Philadelphia Phillies,Baker Bowl,317932,96,97,PHI,PHI,PHI
+1905,NL,PIT,PIT,,2,155,78,96,57,,,N,N,692,5213,1385,190,91,22,382,,202,,,,570,439,2.8600000000,113,12,6,4146,1270,12,389,512,255,112,0.96,Pittsburgh Pirates,Exposition Park,369124,103,100,PIT,PIT,PIT
+1905,NL,SLN,STL,,6,154,77,58,96,,,N,N,535,5066,1254,140,85,20,391,,162,,,,734,537,3.5900000000,135,10,2,4041,1431,28,367,411,274,83,0.95,St. Louis Cardinals,Robison Field,292800,97,99,STL,SLN,SLN
+1906,AL,BOS,BOS,,8,155,77,49,105,,,N,N,463,5168,1223,160,75,13,298,,99,,,,706,524,3.4100000000,124,6,6,4146,1360,37,285,549,333,84,0.94,Boston Americans,Huntington Avenue Grounds,410209,100,102,BOS,BOS,BOS
+1906,AL,CHA,CHW,,1,154,79,93,58,,,Y,Y,570,4925,1133,152,52,7,453,,214,,,,460,325,2.1300000000,117,32,3,4125,1212,11,255,543,243,80,0.96,Chicago White Sox,South Side Park II,585202,98,95,CHW,CHA,CHA
+1906,AL,CLE,CLE,,3,157,79,89,64,,,N,N,663,5426,1516,240,73,12,330,,203,,,,482,328,2.0900000000,133,27,4,4236,1197,16,365,530,217,111,0.96,Cleveland Naps,League Park I,325733,99,97,CLE,CLE,CLE
+1906,AL,DET,DET,,6,151,78,71,78,,,N,N,518,4930,1195,154,64,10,333,,206,,,,599,454,3.0600000000,128,7,4,4002,1398,14,389,469,259,86,0.95,Detroit Tigers,Bennett Park,174043,103,103,DET,DET,DET
+1906,AL,NYA,NYY,,2,155,76,90,61,,,N,N,644,5095,1354,166,77,17,331,,192,,,,543,419,2.7800000000,99,18,5,4071,1236,21,351,605,272,69,0.95,New York Highlanders,Hilltop Park,434700,110,110,NYY,NYA,NYA
+1906,AL,PHA,OAK,,4,149,73,78,67,,,N,N,561,4883,1206,213,49,32,385,,165,,,,543,382,2.6000000000,107,19,4,3966,1135,9,425,749,267,86,0.95,Philadelphia Athletics,Columbia Park,489129,103,101,PHA,PHA,PHA
+1906,AL,SLA,BAL,,5,154,76,76,73,,,N,N,558,5030,1244,145,60,20,366,,221,,,,498,336,2.2300000000,133,17,5,4071,1132,14,314,558,289,80,0.95,St. Louis Browns,Sportsman's Park III,389157,96,96,SLB,SLA,SLA
+1906,AL,WS1,MIN,,7,151,75,55,95,,,N,N,518,4956,1180,144,65,26,306,,233,,,,664,477,3.2500000000,115,13,1,3966,1331,15,451,558,273,78,0.95,Washington Senators,American League Park II,129903,95,98,WSH,WS1,WS1
+1906,NL,BRO,LAD,,5,153,76,66,86,,,N,N,496,4897,1156,141,68,25,388,,175,,,,625,469,3.1300000000,119,22,11,4044,1255,15,453,476,283,73,0.95,Brooklyn Superbas,Washington Park III,277400,92,96,BRO,BRO,BRO
+1906,NL,BSN,ATL,,8,152,76,49,102,,,N,N,408,4925,1115,136,43,16,356,,93,,,,649,465,3.1400000000,137,10,0,4002,1291,24,436,562,337,102,0.94,Boston Beaneaters,South End Grounds III,143280,97,102,BSN,BSN,BSN
+1906,NL,CHN,CHC,,1,155,79,116,36,,,Y,N,705,5018,1316,181,71,20,448,,283,,,,381,270,1.7500000000,125,30,10,4164,1018,12,446,702,190,100,0.97,Chicago Cubs,West Side Park II,654300,106,100,CHC,CHN,CHN
+1906,NL,CIN,CIN,,6,155,78,64,87,,,N,N,533,5025,1198,140,71,16,395,,170,,,,582,409,2.6900000000,126,12,5,4107,1248,14,470,567,260,97,0.95,Cincinnati Reds,Palace of the Fans,330056,104,104,CIN,CIN,CIN
+1906,NL,NY1,SFG,,2,153,75,96,56,,,N,N,625,4768,1217,162,53,15,563,,288,,,,510,369,2.4900000000,105,19,18,4002,1207,13,394,639,228,84,0.96,New York Giants,Polo Grounds III,402850,102,99,NYG,NY1,NY1
+1906,NL,PHI,PHI,,4,154,77,71,82,,,N,N,528,4911,1183,197,47,12,432,,180,,,,564,388,2.5800000000,108,21,5,4062,1201,18,436,500,271,83,0.95,Philadelphia Phillies,Baker Bowl,294680,100,99,PHI,PHI,PHI
+1906,NL,PIT,PIT,,3,154,77,93,60,,,N,N,623,5030,1313,164,67,12,424,,162,,,,470,333,2.2100000000,116,27,2,4074,1234,13,309,532,220,109,0.96,Pittsburgh Pirates,Exposition Park,394877,104,102,PIT,PIT,PIT
+1906,NL,SLN,STL,,7,154,77,52,98,,,N,N,470,5075,1195,137,69,10,361,,110,,,,607,457,3.0400000000,118,4,2,4062,1246,17,479,559,272,92,0.95,St. Louis Cardinals,Robison Field,283770,96,100,STL,SLN,SLN
+1907,AL,BOS,BOS,,7,155,78,59,90,,,N,N,464,5235,1224,154,48,18,305,,125,,,,558,385,2.4500000000,100,17,7,4242,1222,22,337,517,269,103,0.95,Boston Americans,Huntington Avenue Grounds,436777,99,101,BOS,BOS,BOS
+1907,AL,CHA,CHW,,3,157,79,87,64,,,N,N,588,5070,1205,149,33,5,421,,175,,,,474,347,2.2200000000,112,17,9,4218,1279,13,305,604,232,101,0.96,Chicago White Sox,South Side Park II,666307,97,95,CHW,CHA,CHA
+1907,AL,CLE,CLE,,4,158,82,85,67,,,N,N,530,5068,1221,182,68,11,335,,193,,,,525,350,2.2600000000,127,20,5,4176,1253,8,362,513,264,137,0.96,Cleveland Naps,League Park I,382046,101,99,CLE,CLE,CLE
+1907,AL,DET,DET,,1,153,79,92,58,,,Y,N,694,5204,1383,179,75,11,315,,192,,,,532,355,2.3300000000,120,15,7,4110,1281,8,380,512,264,79,0.95,Detroit Tigers,Bennett Park,297079,103,102,DET,DET,DET
+1907,AL,NYA,NYY,,5,152,75,70,78,,,N,N,605,5044,1258,150,67,15,304,,206,,,,665,449,3.0300000000,93,10,5,3999,1327,13,428,511,336,79,0.94,New York Highlanders,Hilltop Park,350020,108,110,NYY,NYA,NYA
+1907,AL,PHA,OAK,,2,150,73,88,57,,,N,N,582,5008,1274,223,43,22,384,,137,,,,511,354,2.3500000000,106,27,6,4062,1106,13,378,789,263,67,0.95,Philadelphia Athletics,Columbia Park,625581,102,102,PHA,PHA,PHA
+1907,AL,SLA,BAL,,6,155,76,69,83,,,N,N,542,5224,1324,154,63,10,370,,144,,,,555,400,2.6100000000,129,15,9,4143,1254,17,352,463,266,97,0.95,St. Louis Browns,Sportsman's Park III,419025,100,99,SLB,SLA,SLA
+1907,AL,WS1,MIN,,8,154,75,49,102,,,N,N,506,5112,1243,134,58,12,390,,223,,,,691,467,3.1100000000,106,12,5,4053,1383,10,344,570,309,69,0.95,Washington Senators,American League Park II,221929,92,95,WSH,WS1,WS1
+1907,NL,BRO,LAD,,5,153,77,65,83,,,N,N,446,4895,1135,142,63,18,336,,121,,,,522,359,2.3800000000,125,20,1,4068,1218,16,463,479,262,94,0.95,Brooklyn Superbas,Washington Park III,312500,92,95,BRO,BRO,BRO
+1907,NL,BSN,ATL,,7,152,74,58,90,,,N,N,502,5020,1222,142,61,22,413,,118,,,,652,495,3.3300000000,121,9,2,4014,1324,28,458,426,249,128,0.96,Boston Doves,South End Grounds III,203221,100,103,BSN,BSN,BSN
+1907,NL,CHN,CHC,,1,155,76,107,45,,,Y,Y,574,4892,1224,162,48,13,435,,235,,,,390,264,1.7300000000,114,32,8,4119,1054,11,402,586,211,110,0.96,Chicago Cubs,West Side Park II,422550,106,101,CHC,CHN,CHN
+1907,NL,CIN,CIN,,6,156,81,66,87,,,N,N,526,4966,1226,126,90,15,372,,158,,,,519,362,2.4100000000,118,10,2,4053,1223,16,444,481,227,118,0.96,Cincinnati Reds,Palace of the Fans,317500,104,105,CIN,CIN,CIN
+1907,NL,NY1,SFG,,4,155,77,82,71,,,N,N,574,4874,1222,160,48,23,516,,205,,,,510,373,2.4500000000,109,22,13,4113,1219,25,369,655,230,75,0.96,New York Giants,Polo Grounds III,538350,103,100,NYG,NY1,NY1
+1907,NL,PHI,PHI,,3,149,75,83,64,,,N,N,512,4725,1113,162,65,12,424,,154,,,,476,351,2.4300000000,110,21,4,3897,1095,13,422,499,256,104,0.95,Philadelphia Phillies,Baker Bowl,341216,99,98,PHI,PHI,PHI
+1907,NL,PIT,PIT,,2,157,77,91,63,,,N,N,634,4957,1261,133,78,19,469,,264,,,,510,348,2.3000000000,111,24,5,4089,1207,12,368,497,253,75,0.95,Pittsburgh Pirates,Exposition Park,319506,102,99,PIT,PIT,PIT
+1907,NL,SLN,STL,,8,155,79,52,101,,,N,N,419,5008,1163,121,51,19,312,,125,,,,608,410,2.7000000000,127,19,2,4095,1212,20,500,594,350,105,0.94,St. Louis Cardinals,Robison Field,185377,97,101,STL,SLN,SLN
+1908,AL,BOS,BOS,,5,155,77,75,79,,,N,N,564,5048,1239,117,88,14,289,,167,,,,513,350,2.2800000000,102,12,7,4140,1200,18,364,624,296,71,0.95,Boston Red Sox,Huntington Avenue Grounds,473048,103,103,BOS,BOS,BOS
+1908,AL,CHA,CHW,,3,156,78,88,64,,,N,N,537,5027,1127,145,41,3,463,,209,,,,470,349,2.2200000000,107,23,10,4242,1170,11,284,623,233,82,0.96,Chicago White Sox,South Side Park II,636096,98,97,CHW,CHA,CHA
+1908,AL,CLE,CLE,,2,157,78,90,64,,,N,N,568,5108,1221,188,58,18,364,,177,,,,457,320,2.0200000000,108,18,5,4272,1172,16,328,548,257,95,0.96,Cleveland Naps,League Park I,422262,101,100,CLE,CLE,CLE
+1908,AL,DET,DET,,1,154,78,90,63,,,Y,N,647,5115,1347,199,86,19,320,,165,,,,547,366,2.4000000000,119,15,5,4122,1313,12,318,553,298,95,0.95,Detroit Tigers,Bennett Park,436199,104,101,DET,DET,DET
+1908,AL,NYA,NYY,,8,155,77,51,103,,,N,N,459,5046,1190,142,50,13,288,,231,,,,713,480,3.1600000000,90,11,3,4098,1288,26,458,585,338,78,0.94,New York Highlanders,Hilltop Park,305500,101,103,NYY,NYA,NYA
+1908,AL,PHA,OAK,,6,157,78,68,85,,,N,N,486,5065,1131,183,50,21,368,,116,,,,562,398,2.5600000000,102,23,4,4200,1194,10,410,741,272,68,0.95,Philadelphia Athletics,Columbia Park,455062,107,107,PHA,PHA,PHA
+1908,AL,SLA,BAL,,4,155,78,83,69,,,N,N,544,5151,1261,173,52,20,343,,126,,,,483,334,2.1500000000,107,15,5,4191,1151,7,387,607,237,97,0.96,St. Louis Browns,Sportsman's Park III,618947,101,100,SLB,SLA,SLA
+1908,AL,WS1,MIN,,7,155,78,67,85,,,N,N,479,5041,1186,132,74,8,368,,170,,,,539,362,2.3400000000,106,15,7,4173,1236,16,348,649,276,89,0.95,Washington Senators,American League Park II,264252,92,96,WSH,WS1,WS1
+1908,NL,BRO,LAD,,7,154,77,53,101,,,N,N,377,4897,1044,110,60,28,323,,113,,,,516,376,2.4700000000,118,20,4,4107,1165,17,444,535,241,66,0.96,Brooklyn Superbas,Washington Park III,275600,96,99,BRO,BRO,BRO
+1908,NL,BSN,ATL,,6,156,78,63,91,,,N,N,537,5131,1228,137,43,17,414,,134,,,,622,435,2.7900000000,92,14,1,4212,1262,29,423,416,252,90,0.96,Boston Doves,South End Grounds III,253750,98,102,BSN,BSN,BSN
+1908,NL,CHN,CHC,,1,158,78,99,55,,,Y,Y,624,5085,1267,196,56,19,418,,212,,,,461,341,2.1400000000,108,29,12,4299,1137,20,437,668,206,76,0.96,Chicago Cubs,West Side Park II,665325,104,100,CHC,CHN,CHN
+1908,NL,CIN,CIN,,5,155,77,73,81,,,N,N,489,4879,1108,129,77,14,372,,196,,,,544,364,2.3700000000,110,17,8,4152,1218,19,415,433,255,72,0.95,Cincinnati Reds,Palace of the Fans,399200,97,98,CIN,CIN,CIN
+1908,NL,NY1,SFG,,2,157,80,98,56,,,N,N,652,5006,1339,182,43,20,494,,181,,,,456,336,2.1400000000,95,25,18,4233,1214,26,288,656,248,79,0.96,New York Giants,Polo Grounds III,910000,105,102,NYG,NY1,NY1
+1908,NL,PHI,PHI,,4,155,78,83,71,,,N,N,504,5012,1223,194,68,11,334,,200,,,,445,325,2.1000000000,116,22,6,4179,1167,8,379,476,238,75,0.96,Philadelphia Phillies,Baker Bowl,420660,103,102,PHI,PHI,PHI
+1908,NL,PIT,PIT,,2,155,77,98,56,,,N,N,585,5109,1263,162,98,25,420,,186,,,,469,330,2.1200000000,100,24,9,4206,1142,16,406,468,225,74,0.96,Pittsburgh Pirates,Exposition Park,382444,100,98,PIT,PIT,PIT
+1908,NL,SLN,STL,,8,154,77,49,105,,,N,N,371,4959,1105,134,57,17,282,,150,,,,626,401,2.6400000000,97,13,4,4104,1217,16,430,528,349,68,0.94,St. Louis Cardinals,Robison Field,205129,95,100,STL,SLN,SLN
+1909,AL,BOS,BOS,,3,152,75,88,63,,,N,N,597,4979,1307,151,69,20,348,,215,,,,550,391,2.5900000000,75,11,15,4080,1214,18,384,555,291,95,0.95,Boston Red Sox,Huntington Avenue Grounds,668965,102,101,BOS,BOS,BOS
+1909,AL,CHA,CHW,,4,159,81,78,74,,,N,N,492,5017,1110,145,56,4,441,,211,,,,463,326,2.0500000000,115,26,4,4290,1182,8,340,669,247,101,0.96,Chicago White Sox,South Side Park II,478400,96,95,CHW,CHA,CHA
+1909,AL,CLE,CLE,,6,155,77,71,82,,,N,N,493,5048,1216,173,81,10,283,,174,,,,532,363,2.4000000000,110,15,3,4083,1212,9,348,568,277,110,0.95,Cleveland Naps,League Park I,354627,104,103,CLE,CLE,CLE
+1909,AL,DET,DET,,1,158,78,98,54,,,Y,N,666,5095,1360,209,58,19,397,,280,,,,493,357,2.2600000000,117,17,12,4260,1254,16,359,528,268,87,0.96,Detroit Tigers,Bennett Park,490490,104,102,DET,DET,DET
+1909,AL,NYA,NYY,,5,153,77,74,77,,,N,N,590,4981,1234,143,61,16,407,,187,,,,587,398,2.6500000000,94,18,7,4050,1223,21,422,597,328,94,0.94,New York Highlanders,Hilltop Park,501000,101,102,NYY,NYA,NYA
+1909,AL,PHA,OAK,,2,153,76,95,58,,,N,N,605,4905,1257,186,88,21,403,,205,,,,408,296,1.9300000000,110,27,3,4134,1069,9,386,728,245,92,0.96,Philadelphia Athletics,Shibe Park,674915,102,97,PHA,PHA,PHA
+1909,AL,SLA,BAL,,7,154,79,61,89,,,N,N,441,4964,1151,116,45,10,331,,136,,,,575,433,2.8800000000,105,21,4,4062,1287,16,383,620,267,107,0.95,St. Louis Browns,Sportsman's Park IV,366274,93,98,SLB,SLA,SLA
+1909,AL,WS1,MIN,,8,156,77,42,110,,,N,N,380,4983,1112,149,41,9,321,,136,,,,656,464,3.0400000000,99,11,2,4122,1288,12,424,653,280,100,0.95,Washington Senators,American League Park II,205199,95,98,WSH,WS1,WS1
+1909,NL,BRO,LAD,,6,155,79,55,98,,,N,N,444,5056,1157,176,59,16,330,,141,,,,627,477,3.1000000000,126,18,3,4152,1277,32,528,594,276,86,0.95,Brooklyn Superbas,Washington Park III,321300,96,100,BRO,BRO,BRO
+1909,NL,BSN,ATL,,8,155,76,45,108,,,N,N,435,5017,1121,124,43,15,400,,135,,,,683,487,3.2000000000,98,13,6,4110,1329,23,543,414,340,101,0.94,Boston Doves,South End Grounds III,195188,104,108,BSN,BSN,BSN
+1909,NL,CHN,CHC,,2,155,77,104,49,,,N,N,635,4999,1227,203,60,20,420,,187,,,,390,272,1.7500000000,111,32,11,4197,1094,6,364,680,244,95,0.96,Chicago Cubs,West Side Park II,633480,103,98,CHC,CHN,CHN
+1909,NL,CIN,CIN,,4,157,80,77,76,,,N,N,606,5088,1273,159,72,22,478,,280,,,,599,394,2.5200000000,91,10,8,4221,1233,5,510,477,309,120,0.95,Cincinnati Reds,Palace of the Fans,424643,99,100,CIN,CIN,CIN
+1909,NL,NY1,SFG,,3,158,77,92,61,,,N,N,623,5218,1327,173,68,26,530,,234,,,,546,363,2.2700000000,105,17,15,4320,1248,28,397,735,305,99,0.95,New York Giants,Polo Grounds III,783700,102,98,NYG,NY1,NY1
+1909,NL,PHI,PHI,,5,154,77,74,79,,,N,N,516,5034,1228,185,53,12,369,,185,,,,518,377,2.4400000000,89,17,6,4173,1190,23,472,612,241,97,0.96,Philadelphia Phillies,Baker Bowl,303177,101,100,PHI,PHI,PHI
+1909,NL,PIT,PIT,,1,154,78,110,42,,,Y,Y,699,5129,1332,218,92,25,479,,185,,,,447,322,2.0700000000,93,21,11,4203,1174,12,320,490,222,100,0.96,Pittsburgh Pirates,Exposition Park/Forbes Field,534950,105,99,PIT,PIT,PIT
+1909,NL,SLN,STL,,7,154,76,54,98,,,N,N,583,5108,1242,148,56,15,568,,161,,,,731,522,3.4100000000,84,5,4,4137,1368,22,483,435,321,90,0.95,St. Louis Cardinals,Robison Field,299982,94,97,STL,SLN,SLN
+1910,AL,BOS,BOS,,4,158,80,81,72,,,N,N,638,5204,1350,175,87,43,430,,194,,,,564,389,2.4500000000,100,12,6,4290,1236,30,414,670,308,80,0.95,Boston Red Sox,Huntington Avenue Grounds,584619,102,101,BOS,BOS,BOS
+1910,AL,CHA,CHW,,6,156,79,68,85,,,N,N,457,5024,1058,115,58,7,403,,183,,,,479,321,2.0300000000,103,23,7,4263,1130,16,381,785,314,100,0.95,Chicago White Sox,South Side Park II/Comiskey Park,552084,95,95,CHW,CHA,CHA
+1910,AL,CLE,CLE,,5,161,80,71,81,,,N,N,548,5385,1316,188,64,9,366,,189,,,,657,469,2.8800000000,92,13,5,4401,1392,10,488,617,246,112,0.96,Cleveland Naps,League Park II,293456,101,103,CLE,CLE,CLE
+1910,AL,DET,DET,,3,155,78,86,68,,,N,N,679,5039,1317,190,72,28,459,,249,,,,582,432,2.8200000000,108,17,5,4140,1257,34,460,532,281,79,0.95,Detroit Tigers,Bennett Park,391288,107,104,DET,DET,DET
+1910,AL,NYA,NYY,,2,156,77,88,63,,,N,N,626,5051,1254,164,75,20,464,,288,,,,557,406,2.6100000000,110,14,8,4197,1238,16,364,654,286,95,0.95,New York Highlanders,Hilltop Park,355857,106,105,NYY,NYA,NYA
+1910,AL,PHA,OAK,,1,155,78,102,48,,,Y,Y,673,5156,1373,191,105,19,409,,207,,,,441,283,1.7900000000,123,24,5,4263,1103,8,450,789,230,117,0.96,Philadelphia Athletics,Shibe Park,588905,99,94,PHA,PHA,PHA
+1910,AL,SLA,BAL,,8,158,79,47,107,,,N,N,451,5077,1105,131,60,12,415,,169,,,,743,478,3.0900000000,101,9,3,4173,1356,14,532,557,385,113,0.94,St. Louis Browns,Sportsman's Park IV,249889,94,98,SLB,SLA,SLA
+1910,AL,WS1,MIN,,7,157,77,66,85,,,N,N,501,4989,1175,145,46,9,449,,192,,,,550,375,2.4600000000,119,19,3,4119,1215,19,375,674,264,99,0.95,Washington Senators,American League Park II,254591,95,99,WSH,WS1,WS1
+1910,NL,BRO,LAD,,6,156,80,64,90,,,N,N,497,5125,1174,166,73,25,434,706,151,,,,623,484,3.0700000000,103,15,5,4260,1331,17,545,555,234,125,0.96,Brooklyn Superbas,Washington Park III,279321,97,100,BRO,BRO,BRO
+1910,NL,BSN,ATL,,8,157,78,53,100,,,N,N,495,5123,1260,173,49,31,359,540,152,,,,701,497,3.2200000000,72,12,9,4170,1328,36,599,531,305,137,0.95,Boston Doves,South End Grounds III,149027,103,109,BSN,BSN,BSN
+1910,NL,CHN,CHC,,1,154,77,104,50,,,Y,N,712,4977,1333,219,84,34,542,501,173,,,,499,384,2.5100000000,100,25,13,4134,1171,18,474,609,229,110,0.96,Chicago Cubs,West Side Park II,526152,99,95,CHC,CHN,CHN
+1910,NL,CIN,CIN,,5,156,77,75,79,,,N,N,620,5121,1326,150,79,23,529,515,310,,,,684,474,3.0800000000,86,16,11,4158,1334,27,528,497,291,103,0.95,Cincinnati Reds,Palace of the Fans,380622,96,97,CIN,CIN,CIN
+1910,NL,NY1,SFG,,2,155,79,91,63,,,N,N,715,5061,1391,204,83,31,562,489,282,,,,567,414,2.6800000000,96,9,10,4173,1290,30,397,717,282,117,0.95,New York Giants,Polo Grounds III,511785,100,98,NYG,NY1,NY1
+1910,NL,PHI,PHI,,4,157,78,78,75,,,N,N,674,5171,1319,223,71,22,506,559,199,,,,639,478,3.0500000000,84,17,9,4233,1297,36,547,657,258,132,0.96,Philadelphia Phillies,Baker Bowl,296597,103,103,PHI,PHI,PHI
+1910,NL,PIT,PIT,,3,154,76,86,67,,,N,N,655,5125,1364,214,83,33,437,524,148,,,,576,433,2.8300000000,73,13,12,4128,1254,20,392,479,245,102,0.96,Pittsburgh Pirates,Forbes Field,436586,107,104,PIT,PIT,PIT
+1910,NL,SLN,STL,,7,153,76,63,90,,,N,N,639,4912,1217,167,70,15,655,581,179,,,,718,562,3.7800000000,81,4,14,4011,1396,30,541,466,257,109,0.95,St. Louis Cardinals,Robison Field,355668,96,98,STL,SLN,SLN
+1911,AL,BOS,BOS,,4,153,76,78,75,,,N,N,680,5014,1379,203,66,35,506,,190,,,,643,411,2.7400000000,87,10,8,4053,1309,21,473,711,322,93,0.94,Boston Red Sox,Huntington Avenue Grounds,503961,99,98,BOS,BOS,BOS
+1911,AL,CHA,CHW,,5,154,78,77,74,,,N,N,719,5213,1401,179,92,20,385,,201,,,,624,457,2.9700000000,85,17,11,4158,1349,22,384,752,255,98,0.96,Chicago White Sox,Comiskey Park,583208,98,97,CHW,CHA,CHA
+1911,AL,CLE,CLE,,3,156,77,80,73,,,N,N,691,5321,1501,238,81,20,354,,209,,,,712,519,3.3600000000,93,6,6,4170,1382,17,552,675,302,108,0.95,Cleveland Naps,League Park II,406296,101,102,CLE,CLE,CLE
+1911,AL,DET,DET,,2,154,76,89,65,,,N,N,831,5294,1544,230,96,30,471,,276,,,,776,575,3.7300000000,108,8,3,4161,1514,28,460,538,312,78,0.95,Detroit Tigers,Bennett Park,484988,105,104,DET,DET,DET
+1911,AL,NYA,NYY,,6,153,77,76,76,,,N,N,684,5052,1374,190,96,25,493,,269,,,,724,535,3.5400000000,91,5,3,4080,1404,26,406,667,320,99,0.94,New York Highlanders,Hilltop Park,302444,106,107,NYY,NYA,NYA
+1911,AL,PHA,OAK,,1,152,75,101,50,,,Y,Y,861,5199,1540,237,93,35,424,,226,,,,601,460,3.0100000000,97,13,13,4125,1343,17,487,739,228,100,0.96,Philadelphia Athletics,Shibe Park,605749,98,95,PHA,PHA,PHA
+1911,AL,SLA,BAL,,8,152,78,45,107,,,N,N,567,4996,1192,187,63,17,460,,125,,,,812,571,3.8600000000,92,8,1,3996,1465,28,463,383,358,104,0.94,St. Louis Browns,Sportsman's Park IV,207984,96,101,SLB,SLA,SLA
+1911,AL,WS1,MIN,,7,154,77,64,90,,,N,N,625,5065,1308,159,54,16,466,,215,,,,766,529,3.5200000000,106,13,3,4059,1471,39,410,628,306,90,0.95,Washington Senators,Griffith Stadium I,244884,98,98,WSH,WS1,WS1
+1911,NL,BRO,LAD,,7,154,74,64,86,,,N,N,539,5059,1198,151,71,28,425,683,184,,,,659,516,3.3900000000,81,13,10,4113,1310,27,566,533,243,112,0.96,Brooklyn Dodgers,Washington Park III,269000,96,98,BRO,BRO,BRO
+1911,NL,BSN,ATL,,8,156,75,44,107,,,N,N,699,5308,1417,249,54,37,554,577,169,,,,1021,776,5.0800000000,73,5,7,4122,1570,76,672,486,347,110,0.94,Boston Rustlers,South End Grounds III,116000,107,111,BSN,BSN,BSN
+1911,NL,CHN,CHC,,2,157,84,92,62,,,N,N,757,5130,1335,218,101,54,585,617,214,,,,607,455,2.9000000000,85,12,16,4233,1270,26,525,582,260,114,0.96,Chicago Cubs,West Side Park II,576000,100,97,CHC,CHN,CHN
+1911,NL,CIN,CIN,,6,159,82,70,83,,,N,N,682,5291,1379,180,105,21,578,594,289,,,,706,516,3.2600000000,77,4,12,4275,1410,36,476,557,297,108,0.95,Cincinnati Reds,Palace of the Fans,300000,97,98,CIN,CIN,CIN
+1911,NL,NY1,SFG,,1,154,75,99,54,,,Y,N,756,5006,1399,225,103,41,530,506,347,,,,542,409,2.6900000000,95,19,13,4104,1267,33,369,771,251,86,0.96,New York Giants,Polo Grounds III/Polo Grounds IV,675000,103,99,NYG,NY1,NY1
+1911,NL,PHI,PHI,,4,153,76,79,73,,,N,N,658,5044,1307,214,56,60,490,588,153,,,,669,503,3.3000000000,90,20,10,4119,1285,43,598,697,231,113,0.96,Philadelphia Phillies,Baker Bowl,416000,101,101,PHI,PHI,PHI
+1911,NL,PIT,PIT,,3,155,78,85,69,,,N,N,744,5137,1345,206,106,49,525,583,160,,,,557,435,2.8400000000,91,13,11,4140,1249,36,375,605,231,131,0.96,Pittsburgh Pirates,Forbes Field,432000,104,101,PIT,PIT,PIT
+1911,NL,SLN,STL,,5,158,79,75,74,,,N,N,671,5132,1295,199,86,26,592,650,175,,,,745,573,3.6800000000,88,6,10,4206,1296,39,701,561,262,106,0.95,St. Louis Cardinals,Robison Field,447768,97,99,STL,SLN,SLN
+1912,AL,BOS,BOS,,1,154,78,105,47,,,Y,Y,799,5071,1404,269,84,29,565,,185,,,,544,418,2.7600000000,108,18,6,4086,1243,18,385,712,266,88,0.95,Boston Red Sox,Fenway Park I,597096,105,102,BOS,BOS,BOS
+1912,AL,CHA,CHW,,4,158,78,78,76,,,N,N,639,5182,1321,174,80,17,423,,205,,,,648,480,3.0600000000,85,14,16,4239,1398,26,426,698,289,102,0.95,Chicago White Sox,Comiskey Park,602241,97,97,CHW,CHA,CHA
+1912,AL,CLE,CLE,,5,155,77,75,78,,,N,N,677,5133,1403,219,77,12,407,,194,,,,681,496,3.3000000000,94,7,7,4056,1367,15,523,622,286,124,0.95,Cleveland Naps,League Park II,336844,103,103,CLE,CLE,CLE
+1912,AL,DET,DET,,6,154,76,69,84,,,N,N,720,5143,1376,189,86,19,530,,270,,,,777,573,3.7700000000,107,7,5,4101,1438,16,521,512,330,91,0.95,Detroit Tigers,Navin Field,402870,97,98,DET,DET,DET
+1912,AL,NYA,NYY,,8,153,76,50,102,,,N,N,630,5092,1320,168,79,18,463,,247,,,,842,613,4.1300000000,105,5,3,4005,1448,28,436,637,376,77,0.94,New York Highlanders,Hilltop Park,242194,106,108,NYY,NYA,NYA
+1912,AL,PHA,OAK,,3,153,77,90,62,,,N,N,779,5111,1442,204,108,22,485,,258,,,,658,501,3.3200000000,95,11,9,4071,1273,12,518,601,263,115,0.95,Philadelphia Athletics,Shibe Park,517653,97,93,PHA,PHA,PHA
+1912,AL,SLA,BAL,,7,157,79,53,101,,,N,N,552,5080,1262,166,71,19,449,,176,,,,764,564,3.7100000000,85,8,5,4107,1433,17,442,547,340,127,0.94,St. Louis Browns,Sportsman's Park IV,214070,96,100,SLB,SLA,SLA
+1912,AL,WS1,MIN,,2,154,78,91,61,,,N,N,699,5075,1298,202,86,20,472,,274,,,,581,411,2.6900000000,98,11,7,4128,1219,24,525,828,297,92,0.95,Washington Senators,Griffith Stadium I,350663,101,101,WSH,WS1,WS1
+1912,NL,BRO,LAD,,7,153,76,58,95,,,N,N,651,5141,1377,220,73,32,490,584,179,,,,754,549,3.6400000000,71,10,8,4071,1399,45,510,553,255,96,0.95,Brooklyn Dodgers,Washington Park III,243000,96,98,BRO,BRO,BRO
+1912,NL,BSN,ATL,,8,155,79,52,101,,,N,N,693,5361,1465,227,68,35,454,690,137,,,,861,644,4.1700000000,88,5,5,4170,1544,43,521,542,295,129,0.95,Boston Braves,South End Grounds III,121000,100,104,BSN,BSN,BSN
+1912,NL,CHN,CHC,,3,152,78,91,59,,,N,N,756,5048,1398,245,90,43,560,615,164,,,,668,516,3.4200000000,80,15,9,4074,1307,33,493,554,249,125,0.96,Chicago Cubs,West Side Park II,514000,100,97,CHC,CHN,CHN
+1912,NL,CIN,CIN,,4,155,77,75,78,,,N,N,656,5115,1310,183,89,21,479,492,248,,,,722,523,3.4200000000,86,13,10,4131,1455,28,452,561,247,102,0.96,Cincinnati Reds,Crosley Field,344000,97,99,CIN,CIN,CIN
+1912,NL,NY1,SFG,,1,154,76,103,48,,,Y,N,823,5067,1451,231,89,47,514,497,319,,,,571,392,2.5800000000,93,8,15,4107,1352,36,338,652,280,123,0.95,New York Giants,Polo Grounds IV,638000,104,100,NYG,NY1,NY1
+1912,NL,PHI,PHI,,5,152,75,73,79,,,N,N,670,5077,1354,244,68,43,464,615,159,,,,688,489,3.2500000000,81,10,9,4065,1381,43,515,616,231,98,0.96,Philadelphia Phillies,Baker Bowl,250000,106,106,PHI,PHI,PHI
+1912,NL,PIT,PIT,,2,152,75,93,58,,,N,N,751,5252,1493,222,129,39,420,514,177,,,,565,439,2.8500000000,94,18,7,4155,1268,28,497,664,169,125,0.97,Pittsburgh Pirates,Forbes Field,384000,99,96,PIT,PIT,PIT
+1912,NL,SLN,STL,,6,153,77,63,90,,,N,N,659,5092,1366,190,77,27,508,620,193,,,,830,579,3.8500000000,61,6,12,4059,1466,31,560,487,272,113,0.95,St. Louis Cardinals,Robison Field,241759,98,101,STL,SLN,SLN
+1913,AL,BOS,BOS,,4,151,75,79,71,,,N,N,631,4965,1334,220,101,17,466,534,189,,,,610,444,2.9400000000,83,12,10,4074,1323,6,442,710,238,84,0.96,Boston Red Sox,Fenway Park I,437194,103,100,BOS,BOS,BOS
+1913,AL,CHA,CHW,,5,153,77,78,74,,,N,N,488,4822,1139,157,66,24,398,550,156,,,,498,352,2.3300000000,84,17,8,4080,1190,10,438,602,255,104,0.96,Chicago White Sox,Comiskey Park,644501,99,100,CHW,CHA,CHA
+1913,AL,CLE,CLE,,3,155,80,86,66,,,N,N,633,5031,1349,206,74,16,420,557,191,,,,536,391,2.5400000000,93,18,5,4158,1278,19,502,689,242,124,0.96,Cleveland Naps,League Park II,541000,103,103,CLE,CLE,CLE
+1913,AL,DET,DET,,6,153,76,66,87,,,N,N,624,5064,1344,180,101,24,496,501,218,,,,716,511,3.3800000000,90,4,7,4080,1359,13,504,468,297,105,0.95,Detroit Tigers,Navin Field,398502,99,99,DET,DET,DET
+1913,AL,NYA,NYY,,7,153,75,57,94,,,N,N,529,4880,1157,155,45,8,534,617,203,,,,668,488,3.2700000000,75,8,7,4032,1318,31,455,530,286,94,0.95,New York Yankees,Polo Grounds IV,357551,100,102,NYY,NYA,NYA
+1913,AL,PHA,OAK,,1,153,76,96,57,,,Y,Y,794,5044,1412,223,80,33,534,547,221,,,,592,479,3.1900000000,69,17,22,4053,1200,24,532,630,212,108,0.96,Philadelphia Athletics,Shibe Park,571896,98,94,PHA,PHA,PHA
+1913,AL,SLA,BAL,,8,155,77,57,96,,,N,N,528,5031,1193,179,73,18,455,769,209,,,,642,470,3.0600000000,104,14,5,4146,1369,21,454,476,301,125,0.95,St. Louis Browns,Sportsman's Park IV,250330,97,100,SLB,SLA,SLA
+1913,AL,WS1,MIN,,2,155,78,90,64,,,N,N,596,5074,1281,156,81,19,440,595,287,,,,561,423,2.7300000000,78,23,20,4188,1177,35,465,758,262,122,0.96,Washington Senators,Griffith Stadium I,325831,102,101,WSH,WS1,WS1
+1913,NL,BRO,LAD,,6,152,77,65,84,,,N,N,595,5165,1394,193,86,39,361,555,188,,,,613,477,3.1300000000,71,9,7,4119,1287,33,439,548,243,125,0.96,Brooklyn Superbas,Ebbets Field,347000,103,103,BRO,BRO,BRO
+1913,NL,BSN,ATL,,5,154,77,69,82,,,N,N,641,5145,1318,191,60,32,488,640,177,,,,690,487,3.1900000000,105,13,3,4119,1343,37,419,597,273,82,0.95,Boston Braves,South End Grounds III,208000,100,102,BSN,BSN,BSN
+1913,NL,CHN,CHC,,3,155,76,88,65,,,N,N,720,5022,1289,195,96,59,554,634,181,,,,630,477,3.1300000000,89,12,15,4116,1330,39,478,556,260,106,0.95,Chicago Cubs,West Side Park II,419000,100,99,CHC,CHN,CHN
+1913,NL,CIN,CIN,,7,156,78,64,89,,,N,N,607,5132,1339,170,96,27,458,579,226,,,,717,531,3.4600000000,71,10,10,4140,1398,40,456,522,251,104,0.96,Cincinnati Reds,Crosley Field,258000,100,101,CIN,CIN,CIN
+1913,NL,NY1,SFG,,1,156,81,101,51,,,Y,N,684,5218,1427,226,71,30,444,501,296,,,,515,382,2.4200000000,82,12,17,4266,1276,38,315,651,252,107,0.96,New York Giants,Polo Grounds IV,630000,101,98,NYG,NY1,NY1
+1913,NL,PHI,PHI,,2,159,78,88,63,,,N,N,693,5400,1433,257,78,73,383,578,156,,,,636,509,3.1500000000,77,20,11,4365,1407,40,512,667,213,112,0.96,Philadelphia Phillies,Baker Bowl,470000,104,104,PHI,PHI,PHI
+1913,NL,PIT,PIT,,4,155,79,78,71,,,N,N,673,5252,1383,210,86,35,391,545,181,,,,585,451,2.9000000000,74,9,7,4200,1344,26,434,590,226,94,0.96,Pittsburgh Pirates,Forbes Field,296000,96,94,PIT,PIT,PIT
+1913,NL,SLN,STL,,8,153,74,51,99,,,N,N,528,4967,1229,152,72,15,451,573,171,,,,755,635,4.2300000000,74,6,11,4053,1426,57,477,465,219,113,0.96,St. Louis Cardinals,Robison Field,203531,98,101,STL,SLN,SLN
+1914,AL,BOS,BOS,,2,159,79,91,62,,,N,N,589,5117,1278,226,85,18,490,549,177,176,,,510,374,2.3600000000,88,24,8,4281,1207,18,393,602,242,99,0.96,Boston Red Sox,Fenway Park I,481359,100,98,BOS,BOS,BOS
+1914,AL,CHA,CHW,,6,157,81,70,84,,,N,N,487,5040,1205,161,71,19,408,609,167,152,,,560,385,2.4800000000,74,17,11,4194,1207,15,401,660,288,90,0.95,Chicago White Sox,Comiskey Park,469290,99,98,CHW,CHA,CHA
+1914,AL,CLE,CLE,,8,157,79,51,102,,,N,N,538,5157,1262,178,70,10,450,685,167,157,,,709,496,3.2100000000,69,9,3,4173,1365,10,666,688,298,119,0.95,Cleveland Naps,League Park II,185997,104,105,CLE,CLE,CLE
+1914,AL,DET,DET,,4,157,78,80,73,,,N,N,615,5102,1318,195,84,25,557,537,211,154,,,618,449,2.8600000000,81,14,12,4236,1285,17,498,567,285,101,0.95,Detroit Tigers,Navin Field,416225,103,102,DET,DET,DET
+1914,AL,NYA,NYY,,6,157,78,70,84,,,N,N,537,4992,1144,149,52,12,577,711,251,191,,,550,436,2.8100000000,98,9,5,4191,1277,30,390,563,223,93,0.96,New York Yankees,Polo Grounds IV,359477,100,101,NYY,NYA,NYA
+1914,AL,PHA,OAK,,1,158,78,99,53,,,Y,N,749,5126,1392,165,80,29,545,517,231,188,,,529,434,2.7800000000,89,24,17,4212,1264,18,521,720,213,116,0.96,Philadelphia Athletics,Shibe Park,346641,96,95,PHA,PHA,PHA
+1914,AL,SLA,BAL,,5,159,81,71,82,,,N,N,523,5101,1241,185,75,17,423,863,233,189,,,615,446,2.8500000000,81,15,11,4230,1309,20,540,553,317,114,0.95,St. Louis Browns,Sportsman's Park IV,244714,96,99,SLB,SLA,SLA
+1914,AL,WS1,MIN,,3,158,77,81,73,,,N,N,572,5108,1245,176,81,18,470,640,220,163,,,519,401,2.5400000000,75,25,20,4260,1170,20,520,784,252,116,0.96,Washington Senators,Griffith Stadium I,243888,104,103,WSH,WS1,WS1
+1914,FL,BLF,BLT,,3,160,,84,70,,,N,,645,5120,1374,222,67,32,487,589,152,,,,628,484,3.1300000000,88,15,13,4176,1389,34,392,732,259,105,0.96,Baltimore Terrapins,,,103,105,BAL,BLF,BLF
+1914,FL,BRF,BTT,,5,157,,77,77,,,N,,662,5221,1402,225,85,42,404,665,220,,,,677,512,3.3300000000,91,11,9,4155,1375,31,559,636,282,120,0.95,Brooklyn Tip-Tops,,,99,100,BTT,BRF,BRF
+1914,FL,BUF,BFL,,4,155,,80,71,,,N,,620,5064,1264,177,74,38,430,761,228,,,,602,487,3.1600000000,89,15,16,4161,1249,45,505,662,242,109,0.96,Buffalo Buffeds,,,102,103,BUF,BUF,BUF
+1914,FL,CHF,CHH,,2,157,,87,67,,,N,,621,5098,1314,227,50,52,520,645,171,,,,517,385,2.4400000000,93,17,8,4260,1204,43,393,650,249,113,0.96,Chicago Chi-Feds,Wrigley Field,,94,92,CHI,CHF,CHF
+1914,FL,IND,NEW,,1,157,,88,65,,,Y,,762,5176,1474,230,90,33,470,641,273,,,,622,475,3.0600000000,104,15,9,4191,1352,29,476,664,275,113,0.95,Indianapolis Hoosiers,,,111,108,IND,IND,IND
+1914,FL,KCF,KCP,,6,154,,67,84,,,N,,644,5127,1369,226,77,39,399,621,171,,,,683,516,3.4100000000,82,10,12,4083,1387,37,445,600,279,135,0.95,Kansas City Packers,,,97,97,KCP,KCF,KCF
+1914,FL,PTF,PBS,,7,154,,64,86,,,N,,605,5114,1339,180,90,34,410,575,153,,,,698,542,3.5600000000,97,9,6,4110,1416,39,444,510,253,92,0.96,Pittsburgh Rebels,,,99,99,PBS,PTF,PTF
+1914,FL,SLF,SLI,,8,154,,62,89,,,N,,565,5078,1254,193,65,26,503,662,113,,,,697,545,3.5900000000,97,9,6,4101,1418,38,409,661,269,94,0.95,St. Louis Terriers,,,105,105,SLM,SLF,SLF
+1914,NL,BRO,LAD,,5,154,79,75,79,,,N,N,622,5152,1386,172,90,31,376,559,173,,,,618,429,2.8200000000,80,11,11,4104,1282,36,466,605,248,112,0.96,Brooklyn Robins,Ebbets Field,122671,102,102,BRO,BRO,BRO
+1914,NL,BSN,ATL,,1,158,79,94,59,,,Y,Y,657,5206,1307,213,60,35,502,617,139,,,,548,433,2.7400000000,104,19,6,4263,1272,38,477,606,237,143,0.96,Boston Braves,South End Grounds III,382913,104,102,BSN,BSN,BSN
+1914,NL,CHN,CHC,,4,156,76,78,76,,,N,N,605,5050,1229,199,74,42,501,577,164,,,,638,418,2.7100000000,70,14,11,4167,1169,37,528,651,310,87,0.95,Chicago Cubs,West Side Park II,202516,100,100,CHC,CHN,CHN
+1914,NL,CIN,CIN,,8,157,77,60,94,,,N,N,530,4991,1178,142,64,16,441,627,224,,,,651,453,2.9400000000,74,15,15,4161,1259,30,489,607,314,113,0.95,Cincinnati Reds,Crosley Field,100791,103,105,CIN,CIN,CIN
+1914,NL,NY1,SFG,,2,156,80,84,70,,,N,N,672,5146,1363,222,59,30,447,479,239,,,,576,454,2.9400000000,88,20,9,4170,1298,47,367,563,253,119,0.96,New York Giants,Polo Grounds IV,364313,97,95,NYG,NY1,NY1
+1914,NL,PHI,PHI,,6,154,78,74,80,,,N,N,651,5110,1345,211,52,62,472,570,145,,,,687,469,3.0600000000,85,14,7,4137,1403,26,452,650,324,81,0.95,Philadelphia Phillies,Baker Bowl,138474,106,105,PHI,PHI,PHI
+1914,NL,PIT,PIT,,7,158,77,69,85,,,N,N,503,5145,1197,148,79,18,416,608,147,,,,540,422,2.7000000000,86,10,11,4215,1272,27,392,488,223,96,0.96,Pittsburgh Pirates,Forbes Field,139620,96,95,PIT,PIT,PIT
+1914,NL,SLN,STL,,3,157,79,81,72,,,N,N,558,5046,1249,203,65,33,445,618,204,,,,540,377,2.3800000000,84,16,12,4272,1279,26,422,531,239,109,0.96,St. Louis Cardinals,Robison Field,256099,99,100,STL,SLN,SLN
+1915,AL,BOS,BOS,,1,155,76,101,50,,,Y,Y,669,5024,1308,202,76,14,527,476,118,117,,,499,371,2.3900000000,81,19,15,4191,1164,18,446,634,226,95,0.96,Boston Red Sox,Fenway Park I,539885,97,95,BOS,BOS,BOS
+1915,AL,CHA,CHW,,3,155,79,93,61,,,N,N,717,4914,1269,163,102,25,583,575,233,183,,,509,378,2.4300000000,91,16,9,4203,1242,14,350,635,222,95,0.96,Chicago White Sox,Comiskey Park,539461,103,102,CHW,CHA,CHA
+1915,AL,CLE,CLE,,7,154,77,57,95,,,N,N,539,5034,1210,169,79,20,490,681,138,117,,,670,477,3.1300000000,62,11,10,4116,1287,18,518,610,272,82,0.95,Cleveland Indians,League Park II,159285,102,104,CLE,CLE,CLE
+1915,AL,DET,DET,,2,156,77,100,54,,,N,N,778,5128,1372,207,94,23,681,527,241,146,,,597,449,2.8600000000,86,10,19,4239,1259,14,492,550,258,107,0.96,Detroit Tigers,Navin Field,476105,105,103,DET,DET,DET
+1915,AL,NYA,NYY,,5,154,82,69,83,,,N,N,584,4982,1162,167,50,31,570,669,198,133,,,588,470,3.0600000000,101,12,2,4146,1272,41,517,559,215,118,0.96,New York Yankees,Polo Grounds IV,256035,100,100,NYY,NYA,NYA
+1915,AL,PHA,OAK,,8,154,74,43,109,,,N,N,545,5081,1204,183,72,16,436,634,127,89,,,888,643,4.2900000000,78,6,2,4044,1358,22,827,588,338,118,0.94,Philadelphia Athletics,Shibe Park,146223,96,99,PHA,PHA,PHA
+1915,AL,SLA,BAL,,6,159,76,63,91,,,N,N,521,5112,1255,166,65,19,472,765,202,160,,,680,474,3.0400000000,76,6,7,4209,1256,21,612,566,336,144,0.94,St. Louis Browns,Sportsman's Park IV,150358,96,98,SLB,SLA,SLA
+1915,AL,WS1,MIN,,4,155,80,85,68,,,N,N,569,5029,1225,152,79,12,458,541,186,106,,,491,358,2.3100000000,87,21,13,4179,1161,12,455,715,230,101,0.96,Washington Senators,Griffith Stadium I,167332,102,101,WSH,WS1,WS1
+1915,FL,BLF,BLT,,8,154,,47,107,,,N,,550,5060,1235,196,53,36,470,641,128,,,,760,598,3.9600000000,85,5,7,4080,1455,52,466,570,273,140,0.95,Baltimore Terrapins,,,103,105,BAL,BLF,BLF
+1915,FL,BRF,BTT,,7,153,,70,82,,,N,,647,5035,1348,205,75,36,473,654,249,,,,673,507,3.3700000000,78,10,16,4065,1299,27,536,467,285,103,0.95,Brooklyn Tip-Tops,,,99,100,BTT,BRF,BRF
+1915,FL,BUF,BFL,,6,153,,74,78,,,N,,574,5065,1261,193,68,40,420,587,184,,,,634,511,3.3800000000,79,14,11,4080,1271,35,553,594,232,112,0.96,Buffalo Blues,,,102,103,BUF,BUF,BUF
+1915,FL,CHF,CHH,,2,155,,86,66,,,Y,,640,5133,1320,185,77,50,444,590,161,,,,538,410,2.6400000000,97,21,10,4191,1232,33,402,576,227,102,0.96,Chicago Whales,Wrigley Field,,94,92,CHI,CHF,CHF
+1915,FL,KCF,KCP,,4,153,,81,72,,,N,,547,4937,1206,200,66,28,368,503,144,,,,551,426,2.8200000000,95,16,11,4077,1210,29,390,526,246,96,0.96,Kansas City Packers,,,97,97,KCP,KCF,KCF
+1915,FL,NEW,NEW,,5,155,,80,72,,,N,,585,5097,1283,210,80,17,438,550,184,,,,562,406,2.6000000000,100,16,7,4218,1308,15,453,581,239,124,0.96,Newark Pepper,,,94,94,NEW,NEW,NEW
+1915,FL,PTF,PBS,,3,156,,86,67,,,N,,592,5040,1318,180,80,20,448,561,224,,,,524,428,2.7900000000,88,16,12,4146,1273,37,441,517,174,98,0.97,Pittsburgh Rebels,,,99,99,PBS,PTF,PTF
+1915,FL,SLF,SLI,,1,159,,87,67,,,N,,634,5145,1344,199,81,23,576,502,195,,,,527,433,2.7300000000,94,24,9,4278,1267,22,396,698,207,111,0.96,St. Louis Terriers,,,105,105,SLM,SLF,SLF
+1915,NL,BRO,LAD,,3,154,78,80,72,,,N,N,536,5120,1268,165,75,14,313,496,131,126,,,560,411,2.6600000000,87,16,8,4167,1252,29,473,499,238,96,0.96,Brooklyn Robins,Ebbets Field,297766,102,101,BRO,BRO,BRO
+1915,NL,BSN,ATL,,2,157,78,83,69,,,N,N,582,5070,1219,231,57,17,549,620,121,98,,,545,401,2.5700000000,95,17,13,4215,1257,23,366,630,213,115,0.96,Boston Braves,Fenway Park I / Braves Field,376283,98,97,BSN,BSN,BSN
+1915,NL,CHN,CHC,,4,156,77,73,80,,,N,N,570,5114,1246,212,66,53,393,639,166,124,,,620,483,3.1100000000,71,18,8,4197,1272,28,480,657,268,94,0.95,Chicago Cubs,West Side Park II,217058,100,101,CHC,CHN,CHN
+1915,NL,CIN,CIN,,7,160,79,71,83,,,N,N,516,5231,1323,194,84,15,360,512,156,142,,,585,452,2.8400000000,80,19,12,4296,1304,28,497,572,224,148,0.96,Cincinnati Reds,Crosley Field,218878,102,104,CIN,CIN,CIN
+1915,NL,NY1,SFG,,8,155,76,69,83,,,N,N,582,5218,1312,195,68,24,315,547,155,137,,,628,479,3.1100000000,78,15,9,4155,1350,40,325,637,256,119,0.96,New York Giants,Polo Grounds IV,391850,94,93,NYG,NY1,NY1
+1915,NL,PHI,PHI,,1,153,76,90,62,,,Y,N,589,4916,1216,202,39,58,460,600,121,113,,,463,331,2.1700000000,98,20,8,4122,1161,26,342,652,216,99,0.96,Philadelphia Phillies,Baker Bowl,449898,101,100,PHI,PHI,PHI
+1915,NL,PIT,PIT,,5,156,79,73,81,,,N,N,557,5113,1259,197,91,24,419,656,182,111,,,520,399,2.6000000000,91,18,11,4140,1229,21,384,544,214,100,0.96,Pittsburgh Pirates,Forbes Field,225743,99,99,PIT,PIT,PIT
+1915,NL,SLN,STL,,6,157,81,72,81,,,N,N,590,5106,1297,159,92,20,457,658,162,144,,,601,450,2.8900000000,79,13,9,4200,1320,30,402,538,235,109,0.96,St. Louis Cardinals,Robison Field,252666,100,101,STL,SLN,SLN
+1916,AL,BOS,BOS,,1,156,78,91,63,,,Y,Y,550,5018,1246,197,56,14,464,482,129,,,,480,389,2.4800000000,76,24,16,4230,1221,10,463,584,183,108,0.97,Boston Red Sox,Fenway Park I,496397,101,98,BOS,BOS,BOS
+1916,AL,CHA,CHW,,2,155,77,89,65,,,N,N,601,5081,1277,194,100,17,447,591,197,,,,497,370,2.3600000000,73,20,15,4236,1189,14,405,644,205,134,0.96,Chicago White Sox,Comiskey Park,679923,101,98,CHW,CHA,CHA
+1916,AL,CLE,CLE,,6,157,78,77,77,,,N,N,630,5064,1264,233,66,16,522,605,160,,,,602,454,2.9000000000,65,9,16,4230,1383,16,467,537,232,130,0.96,Cleveland Indians,League Park II,492106,106,106,CLE,CLE,CLE
+1916,AL,DET,DET,,3,155,77,87,67,,,N,N,670,5193,1371,202,96,17,545,529,190,,,,595,465,2.9700000000,81,8,13,4230,1254,12,578,531,211,110,0.96,Detroit Tigers,Navin Field,616772,104,101,DET,DET,DET
+1916,AL,NYA,NYY,,4,156,79,80,74,,,N,N,577,5198,1277,194,59,35,516,632,179,,,,561,440,2.7700000000,84,12,17,4284,1249,37,476,616,219,119,0.96,New York Yankees,Polo Grounds IV,469211,102,102,NYY,NYA,NYA
+1916,AL,PHA,OAK,,8,154,76,36,117,,,N,N,447,5010,1212,169,65,19,406,631,151,,,,776,585,3.9200000000,94,11,3,4029,1311,26,715,575,312,126,0.95,Philadelphia Athletics,Shibe Park,184471,95,101,PHA,PHA,PHA
+1916,AL,SLA,BAL,,5,158,79,79,75,,,N,N,588,5159,1262,181,50,14,627,640,234,,,,545,414,2.5800000000,74,9,13,4329,1292,15,478,505,248,120,0.96,St. Louis Browns,Sportsman's Park IV,335740,95,97,SLB,SLA,SLA
+1916,AL,WS1,MIN,,7,159,81,76,77,,,N,N,536,5114,1238,170,60,12,535,597,185,,,,543,424,2.6700000000,85,11,7,4290,1271,14,490,706,232,119,0.96,Washington Senators,Griffith Stadium I,177265,99,99,WSH,WS1,WS1
+1916,NL,BRO,LAD,,1,156,78,94,60,,,Y,N,585,5234,1366,195,80,28,355,550,187,,,,471,336,2.1200000000,96,22,9,4281,1201,24,372,634,224,90,0.96,Brooklyn Robins,Ebbets Field,447747,103,102,BRO,BRO,BRO
+1916,NL,BSN,ATL,,3,158,78,89,63,,,N,N,542,5075,1181,166,73,22,437,646,141,,,,453,344,2.1900000000,97,23,11,4245,1206,24,325,644,212,124,0.96,Boston Braves,Braves Field,313495,95,94,BSN,BSN,BSN
+1916,NL,CHN,CHC,,5,156,79,67,86,,,N,N,520,5179,1237,194,56,46,399,662,133,,,,541,417,2.6500000000,72,17,13,4248,1265,32,365,616,277,104,0.95,Chicago Cubs,Wrigley Field,453685,110,111,CHC,CHN,CHN
+1916,NL,CIN,CIN,,7,155,76,60,93,,,N,N,505,5254,1336,187,88,14,362,573,157,,,,617,485,3.1000000000,86,7,6,4224,1356,35,461,569,228,126,0.96,Cincinnati Reds,Crosley Field,255846,98,99,CIN,CIN,CIN
+1916,NL,NY1,SFG,,4,155,78,86,66,,,N,N,597,5152,1305,188,74,42,356,558,206,,,,504,404,2.6000000000,88,22,12,4191,1267,41,310,638,217,108,0.96,New York Giants,Polo Grounds IV,552056,95,93,NYG,NY1,NY1
+1916,NL,PHI,PHI,,2,154,79,91,62,,,N,N,581,4985,1244,223,53,42,399,571,149,,,,489,362,2.3600000000,97,25,9,4146,1238,28,295,601,234,119,0.96,Philadelphia Phillies,Baker Bowl,515365,104,101,PHI,PHI,PHI
+1916,NL,PIT,PIT,,6,157,78,65,89,,,N,N,484,5181,1246,147,91,20,372,618,173,,,,586,435,2.7600000000,88,11,7,4257,1277,24,443,596,260,97,0.95,Pittsburgh Pirates,Forbes Field,289132,101,102,PIT,PIT,PIT
+1916,NL,SLN,STL,,7,153,76,60,93,,,N,N,476,5030,1223,155,74,25,335,651,182,,,,629,473,3.1400000000,58,13,15,4065,1331,31,445,529,273,124,0.95,St. Louis Cardinals,Robison Field,224308,99,101,STL,SLN,SLN
+1917,AL,BOS,BOS,,2,157,80,90,62,,,N,N,555,5048,1243,198,64,14,466,473,105,,,,454,347,2.2000000000,115,15,7,4263,1197,12,413,509,183,116,0.97,Boston Red Sox,Fenway Park I,387856,99,97,BOS,BOS,BOS
+1917,AL,CHA,CHW,,1,156,79,100,54,,,Y,Y,656,5057,1281,152,81,18,522,479,219,,,,464,342,2.1600000000,78,22,21,4272,1236,10,413,517,200,117,0.96,Chicago White Sox,Comiskey Park,684521,102,100,CHW,CHA,CHA
+1917,AL,CLE,CLE,,3,156,78,88,66,,,N,N,584,4994,1224,218,64,13,549,596,210,,,,543,395,2.5200000000,73,20,22,4236,1270,17,438,451,242,136,0.96,Cleveland Indians,League Park II,477298,107,106,CLE,CLE,CLE
+1917,AL,DET,DET,,4,154,76,78,75,,,N,N,639,5093,1317,204,77,25,483,476,163,,,,577,397,2.5600000000,78,20,15,4188,1209,12,504,516,234,95,0.96,Detroit Tigers,Navin Field,457289,100,99,DET,DET,DET
+1917,AL,NYA,NYY,,6,155,75,71,82,,,N,N,524,5136,1226,172,52,27,496,535,136,,,,558,417,2.6600000000,87,10,6,4233,1280,28,427,571,225,129,0.96,New York Yankees,Polo Grounds IV,330294,101,101,NYY,NYA,NYA
+1917,AL,PHA,OAK,,8,154,76,55,98,,,N,N,529,5109,1296,177,62,17,435,519,112,,,,691,496,3.2700000000,80,8,8,4095,1310,23,562,516,251,106,0.96,Philadelphia Athletics,Shibe Park,221432,98,103,PHA,PHA,PHA
+1917,AL,SLA,BAL,,7,155,78,57,97,,,N,N,510,5091,1250,183,63,15,405,540,157,,,,687,492,3.2000000000,66,12,12,4155,1320,19,537,429,281,139,0.95,St. Louis Browns,Sportsman's Park IV,210486,96,98,SLB,SLA,SLA
+1917,AL,WS1,MIN,,5,157,80,74,79,,,N,N,543,5142,1238,173,70,4,500,574,166,,,,566,432,2.7500000000,84,21,10,4239,1217,12,537,637,251,127,0.96,Washington Senators,Griffith Stadium I,89682,99,99,WSH,WS1,WS1
+1917,NL,BRO,LAD,,7,156,78,70,81,,,N,N,511,5251,1299,159,78,25,334,527,130,,,,559,439,2.7800000000,99,8,9,4263,1288,32,405,582,245,102,0.96,Brooklyn Robins,Ebbets Field,221619,103,103,BRO,BRO,BRO
+1917,NL,BSN,ATL,,6,157,77,72,81,,,N,N,536,5201,1280,169,75,22,427,587,155,,,,552,438,2.7700000000,105,22,3,4272,1309,19,371,593,224,122,0.96,Boston Braves,Braves Field,174253,94,94,BSN,BSN,BSN
+1917,NL,CHN,CHC,,5,157,77,74,80,,,N,N,552,5135,1229,194,67,17,415,599,127,,,,567,409,2.6200000000,79,16,9,4212,1303,34,374,654,267,121,0.95,Chicago Cubs,Wrigley Field,360218,108,107,CHC,CHN,CHN
+1917,NL,CIN,CIN,,4,157,80,78,76,,,N,N,601,5251,1385,196,100,26,312,477,153,,,,611,419,2.7000000000,94,12,6,4191,1358,20,402,488,247,120,0.96,Cincinnati Reds,Crosley Field,269056,96,97,CIN,CIN,CIN
+1917,NL,NY1,SFG,,1,158,80,98,56,,,Y,N,635,5211,1360,170,71,39,373,533,162,,,,457,360,2.2700000000,92,18,14,4278,1221,29,327,551,208,122,0.96,New York Giants,Polo Grounds IV,500264,97,94,NYG,NY1,NY1
+1917,NL,PHI,PHI,,2,154,76,87,65,,,N,N,578,5084,1262,225,60,38,435,533,109,,,,500,380,2.4600000000,102,22,5,4167,1258,25,325,616,212,112,0.96,Philadelphia Phillies,Baker Bowl,354428,104,104,PHI,PHI,PHI
+1917,NL,PIT,PIT,,8,157,79,51,103,,,N,N,464,5169,1230,160,61,9,399,580,150,,,,595,474,3.0100000000,84,17,6,4251,1318,14,432,509,251,119,0.96,Pittsburgh Pirates,Forbes Field,192807,103,105,PIT,PIT,PIT
+1917,NL,SLN,STL,,3,154,78,82,70,,,N,N,531,5083,1271,159,93,26,359,652,159,,,,567,469,3.0300000000,66,16,10,4176,1257,29,421,502,221,153,0.96,St. Louis Cardinals,Robison Field,288491,97,99,STL,SLN,SLN
+1918,AL,BOS,BOS,,1,126,70,75,51,,,Y,Y,474,3982,990,159,54,15,406,324,110,,,,380,287,2.3100000000,105,26,2,3360,931,9,380,392,152,89,0.97,Boston Red Sox,Fenway Park I,249513,98,97,BOS,BOS,BOS
+1918,AL,CHA,CHW,,6,124,56,57,67,,,N,N,457,4132,1057,136,55,8,375,358,116,,,,446,342,2.7300000000,76,9,8,3378,1092,9,300,349,167,98,0.96,Chicago White Sox,Comiskey Park,195081,101,98,CHW,CHA,CHA
+1918,AL,CLE,CLE,,2,129,62,73,54,,,N,N,504,4166,1084,176,67,9,491,386,165,,,,447,341,2.6400000000,78,5,13,3483,1126,9,343,364,207,82,0.96,Cleveland Indians,League Park II,295515,109,108,CLE,CLE,CLE
+1918,AL,DET,DET,,7,128,58,55,71,,,N,N,476,4262,1063,141,56,13,452,380,123,,,,557,438,3.4000000000,74,8,7,3480,1130,10,437,374,212,77,0.96,Detroit Tigers,Navin Field,203719,96,96,DET,DET,DET
+1918,AL,NYA,NYY,,4,126,67,60,63,,,N,N,493,4224,1085,160,45,20,367,370,88,,,,475,386,3.0000000000,59,8,13,3471,1103,25,463,370,161,137,0.97,New York Yankees,Polo Grounds IV,282047,102,102,NYY,NYA,NYA
+1918,AL,PHA,OAK,,8,130,68,52,76,,,N,N,412,4278,1039,124,44,22,343,485,83,,,,538,414,3.2200000000,80,13,9,3468,1106,13,486,277,228,136,0.95,Philadelphia Athletics,Shibe Park,177926,101,105,PHA,PHA,PHA
+1918,AL,SLA,BAL,,5,123,53,58,64,,,N,N,426,4019,1040,152,40,5,397,340,138,,,,448,339,2.7500000000,67,8,8,3333,993,11,402,346,190,86,0.96,St. Louis Browns,Sportsman's Park IV,122076,97,99,SLB,SLA,SLA
+1918,AL,WS1,MIN,,3,130,74,72,56,,,N,N,461,4472,1144,156,49,4,376,361,137,,,,412,292,2.1400000000,75,19,8,3684,1021,10,395,505,226,95,0.96,Washington Senators,Griffith Stadium I,182122,99,98,WSH,WS1,WS1
+1918,NL,BRO,LAD,,5,126,54,57,69,,,N,N,360,4212,1052,121,62,10,212,326,113,,,,463,353,2.8100000000,85,17,2,3393,1024,22,320,395,193,74,0.96,Brooklyn Robins,Ebbets Field,83831,100,101,BRO,BRO,BRO
+1918,NL,BSN,ATL,,7,124,52,53,71,,,N,N,424,4162,1014,107,59,13,350,438,83,,,,469,360,2.9000000000,96,13,0,3351,1111,14,277,340,184,89,0.96,Boston Braves,Braves Field,84938,96,97,BSN,BSN,BSN
+1918,NL,CHN,CHC,,1,131,74,84,45,,,Y,N,538,4325,1147,164,53,21,358,343,159,,,,393,290,2.1800000000,92,23,8,3591,1050,13,296,472,188,91,0.96,Chicago Cubs,Wrigley Field,337256,102,101,CHC,CHN,CHN
+1918,NL,CIN,CIN,,3,129,71,68,60,,,N,N,530,4265,1185,165,84,15,304,303,128,,,,496,381,3.0000000000,84,14,6,3426,1136,19,381,321,192,127,0.96,Cincinnati Reds,Crosley Field,163009,98,97,CIN,CIN,CIN
+1918,NL,NY1,SFG,,2,124,56,71,53,,,N,N,480,4164,1081,150,53,13,271,365,130,,,,415,326,2.6400000000,74,18,11,3333,1002,20,228,330,152,78,0.97,New York Giants,Polo Grounds IV,256618,98,95,NYG,NY1,NY1
+1918,NL,PHI,PHI,,6,125,57,55,68,,,N,N,430,4192,1022,158,28,25,346,400,97,,,,507,399,3.1500000000,78,10,6,3417,1086,22,369,312,211,91,0.96,Philadelphia Phillies,Baker Bowl,122266,107,108,PHI,PHI,PHI
+1918,NL,PIT,PIT,,4,126,71,65,60,,,N,N,466,4091,1016,107,72,15,371,285,200,,,,412,314,2.4800000000,85,10,7,3420,1005,13,299,367,179,108,0.96,Pittsburgh Pirates,Forbes Field,213610,103,104,PIT,PIT,PIT
+1918,NL,SLN,STL,,8,131,73,51,78,,,N,N,454,4369,1066,147,64,27,329,461,119,,,,527,392,2.9600000000,72,3,5,3579,1148,16,352,361,220,116,0.96,St. Louis Cardinals,Robison Field,110599,97,98,STL,SLN,SLN
+1919,AL,BOS,BOS,,6,138,66,66,71,,,N,N,564,4548,1188,181,49,33,471,411,108,,,,552,450,3.3100000000,89,15,8,3672,1251,16,421,381,143,118,0.97,Boston Red Sox,Fenway Park I,417291,94,94,BOS,BOS,BOS
+1919,AL,CHA,CHW,,1,140,70,88,52,,,Y,N,667,4675,1343,218,70,25,427,358,150,,,,534,427,3.0400000000,88,14,3,3795,1245,24,342,468,173,116,0.96,Chicago White Sox,Comiskey Park,627186,100,99,CHW,CHA,CHA
+1919,AL,CLE,CLE,,2,139,69,84,55,,,N,N,636,4565,1268,254,72,24,498,367,113,,,,537,407,2.9400000000,80,10,10,3735,1242,19,362,432,201,102,0.96,Cleveland Indians,League Park II,538135,106,104,CLE,CLE,CLE
+1919,AL,DET,DET,,4,140,70,80,60,,,N,N,618,4665,1319,222,84,23,429,427,121,,,,578,461,3.3000000000,85,10,4,3768,1254,35,436,428,205,81,0.96,Detroit Tigers,Navin Field,643805,98,100,DET,DET,DET
+1919,AL,NYA,NYY,,3,141,73,80,59,,,N,N,578,4775,1275,193,49,45,386,479,101,,,,506,403,2.8200000000,85,14,7,3861,1143,47,433,500,193,108,0.96,New York Yankees,Polo Grounds IV,619164,101,99,NYY,NYA,NYA
+1919,AL,PHA,OAK,,8,140,70,36,104,,,N,N,457,4730,1156,175,71,35,349,565,103,,,,742,586,4.2600000000,72,1,3,3717,1371,44,503,417,257,96,0.95,Philadelphia Athletics,Shibe Park,225209,101,106,PHA,PHA,PHA
+1919,AL,SLA,BAL,,5,140,70,67,72,,,N,N,533,4672,1234,187,73,31,391,443,74,,,,567,437,3.1300000000,78,14,4,3768,1255,35,421,415,214,98,0.96,St. Louis Browns,Sportsman's Park IV,349350,102,103,SLB,SLA,SLA
+1919,AL,WS1,MIN,,7,142,72,56,84,,,N,N,533,4757,1238,177,63,24,416,511,142,,,,570,426,3.0100000000,68,13,10,3822,1237,20,451,536,227,86,0.96,Washington Senators,Griffith Stadium I,234096,99,99,WSH,WS1,WS1
+1919,NL,BRO,LAD,,5,141,70,69,71,,,N,N,525,4844,1272,167,66,25,258,405,112,,,,513,389,2.7300000000,98,12,1,3843,1256,21,292,476,218,84,0.96,Brooklyn Robins,Ebbets Field,360721,103,103,BRO,BRO,BRO
+1919,NL,BSN,ATL,,6,140,68,57,82,,,N,N,465,4746,1201,142,62,24,355,481,145,,,,563,447,3.1700000000,79,5,9,3810,1313,29,337,374,204,111,0.96,Boston Braves,Braves Field,167401,95,98,BSN,BSN,BSN
+1919,NL,CHN,CHC,,3,140,71,75,65,,,N,N,454,4581,1174,166,58,21,298,359,150,,,,407,311,2.2100000000,80,21,5,3795,1127,14,294,495,186,87,0.96,Chicago Cubs,Wrigley Field,424430,100,99,CHC,CHN,CHN
+1919,NL,CIN,CIN,,1,140,70,96,44,,,Y,Y,577,4577,1204,135,83,20,405,368,143,,,,401,316,2.2300000000,89,23,9,3822,1104,21,298,407,152,98,0.97,Cincinnati Reds,Crosley Field,532501,97,95,CIN,CIN,CIN
+1919,NL,NY1,SFG,,2,140,69,87,53,,,N,N,605,4664,1254,204,64,40,328,407,157,,,,470,377,2.7000000000,72,11,13,3768,1153,34,305,340,215,96,0.96,New York Giants,Polo Grounds IV,708857,100,97,NYG,NY1,NY1
+1919,NL,PHI,PHI,,8,138,71,47,90,,,N,N,510,4746,1191,208,50,42,323,469,114,,,,699,576,4.1400000000,93,6,2,3756,1391,40,408,397,219,112,0.96,Philadelphia Phillies,Baker Bowl,240424,107,111,PHI,PHI,PHI
+1919,NL,PIT,PIT,,4,139,70,71,68,,,N,N,472,4538,1132,130,82,17,344,381,196,,,,466,400,2.8800000000,91,17,4,3747,1113,23,263,391,160,89,0.96,Pittsburgh Pirates,Forbes Field,276810,104,103,PIT,PIT,PIT
+1919,NL,SLN,STL,,7,138,69,54,83,,,N,N,463,4588,1175,163,52,18,304,418,148,,,,552,437,3.2300000000,55,6,8,3651,1146,25,415,414,217,112,0.96,St. Louis Cardinals,Robison Field,167059,94,96,STL,SLN,SLN
+1920,AL,BOS,BOS,,5,154,76,72,81,,,N,N,650,5199,1397,216,71,22,533,429,98,111,,,698,592,3.8200000000,92,11,6,4185,1481,39,461,481,183,131,0.97,Boston Red Sox,Fenway Park I,402445,96,96,BOS,BOS,BOS
+1920,AL,CHA,CHW,,2,154,77,96,58,,,N,N,794,5328,1574,263,98,37,471,353,112,96,,,665,553,3.5900000000,109,9,10,4158,1467,45,405,438,194,142,0.96,Chicago White Sox,Comiskey Park,833492,99,99,CHW,CHA,CHA
+1920,AL,CLE,CLE,,1,154,78,98,56,,,Y,Y,857,5203,1574,300,95,35,574,379,73,92,,,642,522,3.4100000000,94,11,7,4131,1448,31,401,466,185,124,0.97,Cleveland Indians,League Park II,912832,104,101,CLE,CLE,CLE
+1920,AL,DET,DET,,7,155,78,61,93,,,N,N,652,5215,1408,228,72,30,479,391,76,66,,,833,622,4.0400000000,74,9,7,4155,1487,46,561,483,229,95,0.96,Detroit Tigers,Navin Field,579650,97,98,DET,DET,DET
+1920,AL,NYA,NYY,,3,154,77,95,59,,,N,N,838,5176,1448,268,71,115,539,626,64,82,,,629,505,3.3200000000,88,15,11,4104,1414,48,420,480,193,129,0.97,New York Yankees,Polo Grounds IV,1289422,104,101,NYY,NYA,NYA
+1920,AL,PHA,OAK,,8,156,77,48,106,,,N,N,558,5256,1324,220,49,44,353,593,50,67,,,834,603,3.9300000000,79,6,2,4140,1612,56,461,423,267,126,0.95,Philadelphia Athletics,Shibe Park,287888,101,106,PHA,PHA,PHA
+1920,AL,SLA,BAL,,4,154,78,76,77,,,N,N,797,5358,1651,279,83,50,427,339,118,79,,,766,617,4.0300000000,84,9,14,4134,1481,53,578,444,231,119,0.96,St. Louis Browns,Sportsman's Park IV,419311,103,103,SLB,SLA,SLA
+1920,AL,WS1,MIN,,6,153,76,68,84,,,N,N,723,5251,1526,233,81,36,433,543,161,114,,,802,633,4.1700000000,81,10,10,4101,1521,51,520,418,232,95,0.96,Washington Senators,Griffith Stadium I,359260,97,98,WSH,WS1,WS1
+1920,NL,BRO,LAD,,1,155,78,93,61,,,Y,N,660,5399,1493,205,99,28,359,391,70,80,,,528,415,2.6200000000,89,17,10,4281,1381,25,327,553,226,118,0.96,Brooklyn Robins,Ebbets Field,808722,104,103,BRO,BRO,BRO
+1920,NL,BSN,ATL,,7,153,74,62,90,,,N,N,523,5218,1358,168,86,23,385,488,88,98,,,670,545,3.5400000000,93,14,6,4158,1464,39,415,368,239,125,0.96,Boston Braves,Braves Field,162483,95,97,BSN,BSN,BSN
+1920,NL,CHN,CHC,,5,154,77,75,79,,,N,N,619,5117,1350,223,67,34,428,421,115,129,,,635,504,3.2700000000,95,13,9,4164,1459,37,382,508,225,112,0.96,Chicago Cubs,Wrigley Field,480783,101,102,CHC,CHN,CHN
+1920,NL,CIN,CIN,,3,154,77,82,71,,,N,N,639,5176,1432,169,76,18,382,367,158,128,,,569,448,2.9000000000,90,12,9,4173,1327,26,393,435,200,125,0.96,Cincinnati Reds,Crosley Field,568107,98,97,CIN,CIN,CIN
+1920,NL,NY1,SFG,,2,155,80,86,68,,,N,N,682,5309,1427,210,76,46,432,545,131,113,,,543,438,2.8000000000,86,18,9,4224,1379,44,297,380,209,137,0.96,New York Giants,Polo Grounds IV,929609,100,97,NYG,NY1,NY1
+1920,NL,PHI,PHI,,8,153,77,62,91,,,N,N,565,5264,1385,229,54,64,283,531,100,83,,,714,557,3.6300000000,77,8,11,4140,1480,35,444,419,232,135,0.96,Philadelphia Phillies,Baker Bowl,330998,104,109,PHI,PHI,PHI
+1920,NL,PIT,PIT,,4,155,78,79,75,,,N,N,530,5219,1342,162,90,16,374,405,181,117,,,552,454,2.8900000000,92,17,10,4245,1389,25,280,444,186,119,0.97,Pittsburgh Pirates,Forbes Field,429037,103,103,PIT,PIT,PIT
+1920,NL,SLN,STL,,5,155,76,75,79,,,N,N,675,5495,1589,238,96,32,373,484,126,114,,,682,543,3.4300000000,72,9,12,4278,1488,30,479,529,256,136,0.96,St. Louis Cardinals,Robison Field/Sportsman's Park IV,326836,98,98,STL,SLN,SLN
+1921,AL,BOS,BOS,,5,154,77,75,79,,,N,N,668,5206,1440,248,69,17,428,344,83,65,,,696,603,3.9800000000,88,9,5,4092,1521,53,452,446,157,151,0.97,Boston Red Sox,Fenway Park I,279273,97,99,BOS,BOS,BOS
+1921,AL,CHA,CHW,,7,154,77,62,92,,,N,N,683,5329,1509,242,82,35,445,474,97,93,,,858,749,4.9400000000,84,7,9,4095,1603,52,549,392,199,155,0.96,Chicago White Sox,Comiskey Park,543650,98,98,CHW,CHA,CHA
+1921,AL,CLE,CLE,,2,154,77,94,60,,,N,N,925,5383,1656,355,90,42,623,376,50,42,,,712,597,3.9000000000,81,11,14,4131,1534,43,431,475,204,124,0.96,Cleveland Indians,League Park II,748705,102,100,CLE,CLE,CLE
+1921,AL,DET,DET,,6,154,77,71,82,,,N,N,883,5461,1724,268,100,58,582,376,95,89,,,852,678,4.4000000000,73,4,16,4158,1634,71,495,452,231,107,0.96,Detroit Tigers,Navin Field,661527,99,100,DET,DET,DET
+1921,AL,NYA,NYY,,1,153,78,98,55,,,Y,N,948,5249,1576,285,87,134,588,569,89,64,,,708,579,3.8200000000,92,8,15,4092,1461,51,470,481,220,138,0.96,New York Yankees,Polo Grounds IV,1230696,102,98,NYY,NYA,NYA
+1921,AL,PHA,OAK,,8,155,77,53,100,,,N,N,657,5465,1497,256,64,82,424,565,68,55,,,894,717,4.6100000000,75,2,7,4200,1645,85,548,431,277,144,0.95,Philadelphia Athletics,Shibe Park,344430,100,104,PHA,PHA,PHA
+1921,AL,SLA,BAL,,3,154,77,81,73,,,N,N,835,5442,1655,246,106,67,413,407,92,71,,,845,706,4.6100000000,77,9,9,4137,1541,71,556,477,224,127,0.96,St. Louis Browns,Sportsman's Park IV,355978,106,105,SLB,SLA,SLA
+1921,AL,WS1,MIN,,4,154,76,80,73,,,N,N,704,5294,1468,240,96,42,462,472,111,66,,,738,610,3.9700000000,80,10,10,4149,1568,51,442,452,234,153,0.96,Washington Senators,Griffith Stadium I,456069,95,96,WSH,WS1,WS1
+1921,NL,BRO,LAD,,5,152,78,77,75,,,N,N,667,5263,1476,209,85,59,325,400,91,73,,,681,560,3.7000000000,82,8,12,4089,1556,46,361,471,232,142,0.96,Brooklyn Robins,Ebbets Field,613245,105,104,BRO,BRO,BRO
+1921,NL,BSN,ATL,,4,153,74,79,74,,,N,N,721,5385,1561,209,100,61,377,470,94,100,,,697,600,3.9000000000,74,11,12,4155,1488,54,420,382,199,122,0.96,Boston Braves,Braves Field,318627,94,96,BSN,BSN,BSN
+1921,NL,CHN,CHC,,7,153,76,64,89,,,N,N,668,5321,1553,234,56,37,343,374,70,97,,,773,665,4.3900000000,73,7,7,4089,1605,67,409,441,166,129,0.97,Chicago Cubs,Wrigley Field,410107,100,101,CHC,CHN,CHN
+1921,NL,CIN,CIN,,6,153,76,70,83,,,N,N,618,5112,1421,221,94,20,375,308,117,120,,,649,524,3.4600000000,83,7,9,4089,1500,37,305,408,193,139,0.96,Cincinnati Reds,Crosley Field,311227,95,94,CIN,CIN,CIN
+1921,NL,NY1,SFG,,1,153,79,94,59,,,Y,Y,840,5278,1575,237,93,75,469,390,137,114,,,637,541,3.5500000000,71,9,18,4116,1497,79,295,357,187,155,0.97,New York Giants,Polo Grounds IV,973477,102,98,NYG,NY1,NY1
+1921,NL,PHI,PHI,,8,154,76,51,103,,,N,N,617,5329,1512,238,50,88,294,615,66,80,,,919,671,4.4800000000,82,5,8,4044,1665,79,371,333,294,127,0.95,Philadelphia Phillies,Baker Bowl,273961,107,111,PHI,PHI,PHI
+1921,NL,PIT,PIT,,2,154,76,90,63,,,N,N,692,5379,1533,231,104,37,341,371,134,93,,,595,498,3.1700000000,88,10,10,4245,1448,37,322,500,171,129,0.97,Pittsburgh Pirates,Forbes Field,701567,102,101,PIT,PIT,PIT
+1921,NL,SLN,STL,,3,154,78,87,66,,,N,N,809,5309,1635,260,88,83,382,452,94,94,,,681,551,3.6200000000,70,10,16,4113,1486,61,399,464,219,130,0.96,St. Louis Cardinals,Sportsman's Park IV,384773,97,96,STL,SLN,SLN
+1922,AL,BOS,BOS,,8,154,73,61,93,,,N,N,598,5288,1392,250,55,45,366,455,60,63,,,769,656,4.3000000000,71,10,6,4119,1508,48,503,359,221,139,0.96,Boston Red Sox,Fenway Park I,259184,99,101,BOS,BOS,BOS
+1922,AL,CHA,CHW,,5,155,77,77,77,,,N,N,691,5267,1463,243,62,45,482,463,106,84,,,691,614,3.9400000000,86,13,8,4209,1472,57,529,484,153,132,0.97,Chicago White Sox,Comiskey Park,602860,99,100,CHW,CHA,CHA
+1922,AL,CLE,CLE,,4,155,80,78,76,,,N,N,768,5293,1544,320,73,32,554,331,89,58,,,817,705,4.5900000000,76,14,7,4149,1605,58,464,489,201,140,0.96,Cleveland Indians,League Park II,528145,102,100,CLE,CLE,CLE
+1922,AL,DET,DET,,3,155,77,79,75,,,N,N,828,5360,1641,250,87,54,530,378,78,61,,,791,660,4.2700000000,67,7,15,4173,1554,62,473,461,192,135,0.97,Detroit Tigers,Navin Field,861206,97,96,DET,DET,DET
+1922,AL,NYA,NYY,,1,154,77,94,60,,,Y,N,758,5245,1504,220,75,95,497,532,62,59,,,618,525,3.3900000000,100,7,14,4179,1402,73,423,458,155,122,0.97,New York Yankees,Polo Grounds IV,1026134,102,98,NYY,NYA,NYA
+1922,AL,PHA,OAK,,7,155,78,65,89,,,N,N,705,5211,1409,229,63,111,437,591,60,63,,,830,695,4.5900000000,73,4,6,4086,1573,107,469,373,218,119,0.96,Philadelphia Athletics,Shibe Park,425356,102,105,PHA,PHA,PHA
+1922,AL,SLA,BAL,,2,154,77,93,61,,,N,N,867,5416,1693,291,94,98,473,381,132,73,,,643,523,3.3800000000,79,8,22,4176,1412,71,419,534,196,158,0.96,St. Louis Browns,Sportsman's Park IV,712918,104,103,SLB,SLA,SLA
+1922,AL,WS1,MIN,,6,154,79,69,85,,,N,N,650,5201,1395,229,76,45,458,442,94,54,,,706,577,3.8100000000,84,13,10,4086,1485,49,500,422,201,161,0.96,Washington Senators,Griffith Stadium I,458552,95,96,WSH,WS1,WS1
+1922,NL,BRO,LAD,,6,155,78,76,78,,,N,N,743,5413,1569,235,76,56,339,318,79,60,,,754,623,4.0500000000,82,12,8,4155,1574,74,490,499,208,139,0.96,Brooklyn Robins,Ebbets Field,498865,100,100,BRO,BRO,BRO
+1922,NL,BSN,ATL,,8,154,76,53,100,,,N,N,596,5161,1355,162,73,32,387,451,67,65,,,822,655,4.3700000000,63,7,6,4044,1565,57,489,360,215,121,0.96,Boston Braves,Braves Field,167965,95,97,BSN,BSN,BSN
+1922,NL,CHN,CHC,,5,156,76,80,74,,,N,N,771,5335,1564,248,71,42,525,447,97,108,,,808,674,4.3400000000,74,8,12,4191,1579,77,475,402,202,154,0.96,Chicago Cubs,Wrigley Field,542283,102,102,CHC,CHN,CHN
+1922,NL,CIN,CIN,,2,156,79,86,68,,,N,N,766,5282,1561,226,99,45,436,381,130,136,,,677,543,3.5300000000,88,8,3,4155,1481,49,326,357,205,147,0.96,Cincinnati Reds,Crosley Field,493754,98,97,CIN,CIN,CIN
+1922,NL,NY1,SFG,,1,156,79,93,61,,,Y,Y,852,5454,1661,253,90,80,448,421,116,83,,,658,535,3.4500000000,76,7,15,4188,1454,71,393,388,194,145,0.97,New York Giants,Polo Grounds IV,945809,102,98,NYG,NY1,NY1
+1922,NL,PHI,PHI,,7,154,77,57,96,,,N,N,738,5459,1537,268,55,116,450,611,48,60,,,920,707,4.6400000000,73,6,5,4116,1692,89,460,394,222,152,0.96,Philadelphia Phillies,Baker Bowl,232471,108,113,PHI,PHI,PHI
+1922,NL,PIT,PIT,,3,155,78,85,69,,,N,N,865,5521,1698,239,110,52,423,326,145,59,,,736,613,3.9800000000,88,15,7,4161,1613,52,358,490,187,126,0.97,Pittsburgh Pirates,Forbes Field,523675,102,100,PIT,PIT,PIT
+1922,NL,SLN,STL,,3,154,77,85,69,,,N,N,863,5425,1634,280,88,107,447,425,73,63,,,819,672,4.4400000000,60,8,12,4086,1609,61,447,465,239,122,0.96,St. Louis Cardinals,Sportsman's Park IV,536998,95,94,STL,SLN,SLN
+1923,AL,BOS,BOS,,8,154,78,61,91,,,N,N,584,5181,1354,253,54,34,391,480,77,91,,,809,640,4.2000000000,77,3,11,4116,1534,48,520,412,232,126,0.96,Boston Red Sox,Fenway Park I,229688,100,103,BOS,BOS,BOS
+1923,AL,CHA,CHW,,7,156,75,69,85,,,N,N,692,5246,1463,254,57,42,532,458,191,119,,,741,629,4.0500000000,74,5,11,4191,1512,49,534,467,181,138,0.97,Chicago White Sox,Comiskey Park,573778,99,99,CHW,CHA,CHA
+1923,AL,CLE,CLE,,3,153,78,82,71,,,N,N,888,5290,1594,301,75,59,633,384,79,77,,,746,598,3.9100000000,77,10,11,4128,1517,36,465,407,227,143,0.96,Cleveland Indians,League Park II,558856,100,100,CLE,CLE,CLE
+1923,AL,DET,DET,,2,155,77,83,71,,,N,N,831,5266,1579,270,69,41,596,385,87,62,,,741,624,4.0900000000,61,9,12,4119,1502,58,449,447,200,103,0.96,Detroit Tigers,Navin Field,911377,98,97,DET,DET,DET
+1923,AL,NYA,NYY,,1,152,76,98,54,,,Y,Y,823,5347,1554,231,79,105,521,516,69,74,,,622,555,3.6200000000,101,9,10,4140,1365,68,491,506,144,131,0.97,New York Yankees,Yankee Stadium I,1007066,103,100,NYY,NYA,NYA
+1923,AL,PHA,OAK,,6,153,75,69,83,,,N,N,661,5196,1407,229,64,53,445,517,72,62,,,761,618,4.0800000000,65,7,12,4092,1465,68,550,400,225,127,0.96,Philadelphia Athletics,Shibe Park,534122,101,103,PHA,PHA,PHA
+1923,AL,SLA,BAL,,5,154,78,74,78,,,N,N,688,5298,1489,248,62,82,442,423,64,54,,,720,600,3.9300000000,83,10,10,4119,1430,59,528,488,176,145,0.97,St. Louis Browns,Sportsman's Park IV,430296,106,105,SLB,SLA,SLA
+1923,AL,WS1,MIN,,4,155,79,75,78,,,N,N,720,5244,1436,224,93,26,532,448,102,67,,,747,608,3.9800000000,71,8,16,4122,1527,56,563,474,213,182,0.96,Washington Senators,Griffith Stadium I,357406,95,95,WSH,WS1,WS1
+1923,NL,BRO,LAD,,6,155,78,76,78,,,N,N,753,5476,1559,214,81,62,425,382,71,50,,,741,580,3.7400000000,94,8,5,4188,1503,55,476,548,291,137,0.95,Brooklyn Robins,Ebbets Field,564666,98,97,BRO,BRO,BRO
+1923,NL,BSN,ATL,,7,155,77,54,100,,,N,N,636,5329,1455,213,58,32,429,404,57,80,,,798,651,4.2100000000,54,13,7,4176,1662,64,394,351,230,157,0.96,Boston Braves,Braves Field,227802,95,99,BSN,BSN,BSN
+1923,NL,CHN,CHC,,4,154,77,83,71,,,N,N,756,5259,1516,243,52,90,455,485,181,143,,,704,580,3.8200000000,80,8,11,4098,1419,86,435,408,208,144,0.96,Chicago Cubs,Wrigley Field,703705,100,100,CHC,CHN,CHN
+1923,NL,CIN,CIN,,2,154,78,91,63,,,N,N,708,5278,1506,237,95,45,439,367,96,105,,,629,496,3.2100000000,88,11,9,4173,1465,28,359,450,202,144,0.96,Cincinnati Reds,Crosley Field,575063,98,97,CIN,CIN,CIN
+1923,NL,NY1,SFG,,1,153,77,95,58,,,Y,N,854,5452,1610,248,76,85,487,406,106,70,,,679,597,3.9000000000,62,10,18,4134,1440,82,424,453,176,141,0.97,New York Giants,Polo Grounds IV,820780,100,96,NYG,NY1,NY1
+1923,NL,PHI,PHI,,8,155,75,50,104,,,N,N,748,5491,1528,259,39,112,414,556,70,73,,,1008,816,5.3400000000,68,3,8,4128,1801,100,549,384,213,172,0.96,Philadelphia Phillies,Baker Bowl,228168,111,115,PHI,PHI,PHI
+1923,NL,PIT,PIT,,3,154,77,87,67,,,N,N,786,5405,1592,224,111,49,407,362,154,75,,,696,592,3.8700000000,92,5,9,4128,1513,53,402,414,179,157,0.97,Pittsburgh Pirates,Forbes Field,611082,102,100,PIT,PIT,PIT
+1923,NL,SLN,STL,,5,154,78,79,74,,,N,N,746,5526,1582,274,76,63,438,446,89,61,,,732,601,3.8700000000,77,9,7,4194,1539,70,456,398,230,141,0.96,St. Louis Cardinals,Sportsman's Park IV,338551,97,97,STL,SLN,SLN
+1924,AL,BOS,BOS,,7,157,77,67,87,,,N,N,737,5340,1481,302,63,30,603,417,79,61,,,806,672,4.3500000000,73,8,16,4173,1563,43,523,414,207,124,0.96,Boston Red Sox,Fenway Park I,448556,99,103,BOS,BOS,BOS
+1924,AL,CHA,CHW,,8,154,77,66,87,,,N,N,793,5255,1512,254,58,41,604,418,138,86,,,858,722,4.7400000000,76,1,11,4110,1635,52,512,360,230,136,0.96,Chicago White Sox,Comiskey Park,606658,97,97,CHW,CHA,CHA
+1924,AL,CLE,CLE,,6,153,75,67,86,,,N,N,755,5332,1580,306,59,41,492,371,84,56,,,814,660,4.4000000000,87,7,7,4047,1603,43,503,315,201,130,0.96,Cleveland Indians,League Park II,481905,102,101,CLE,CLE,CLE
+1924,AL,DET,DET,,3,156,78,86,68,,,N,N,849,5389,1604,315,76,35,607,400,100,76,,,796,649,4.1900000000,60,5,20,4182,1586,55,467,441,186,142,0.97,Detroit Tigers,Navin Field,1015136,98,97,DET,DET,DET
+1924,AL,NYA,NYY,,2,153,78,89,63,,,N,N,798,5240,1516,248,86,98,478,420,69,64,,,667,583,3.8600000000,76,13,13,4077,1483,59,522,487,154,131,0.97,New York Yankees,Yankee Stadium I,1053533,100,99,NYY,NYA,NYA
+1924,AL,PHA,OAK,,5,152,75,71,81,,,N,N,685,5184,1459,251,59,63,374,482,79,66,,,778,656,4.3900000000,68,8,10,4035,1527,43,597,371,181,157,0.97,Philadelphia Athletics,Shibe Park,531992,101,101,PHA,PHA,PHA
+1924,AL,SLA,BAL,,4,153,78,74,78,,,N,N,769,5236,1543,266,62,67,465,349,85,85,,,809,687,4.5700000000,65,11,7,4059,1511,68,517,386,176,141,0.97,St. Louis Browns,Sportsman's Park IV,533349,106,107,SLB,SLA,SLA
+1924,AL,WS1,MIN,,1,156,79,92,62,,,Y,Y,755,5304,1558,255,88,22,513,392,115,87,,,613,513,3.3400000000,74,13,25,4149,1329,34,505,469,171,149,0.97,Washington Senators,Griffith Stadium I,584310,98,96,WSH,WS1,WS1
+1924,NL,BRO,LAD,,2,154,77,92,62,,,N,N,717,5339,1534,227,54,72,447,357,34,46,,,675,557,3.6400000000,97,10,5,4128,1432,58,403,638,196,121,0.96,Brooklyn Robins,Ebbets Field,818883,97,97,BRO,BRO,BRO
+1924,NL,BSN,ATL,,8,154,76,53,100,,,N,N,520,5283,1355,194,52,25,354,451,74,68,,,800,683,4.4600000000,66,10,4,4137,1607,49,402,364,168,154,0.97,Boston Braves,Braves Field,177478,95,98,BSN,BSN,BSN
+1924,NL,CHN,CHC,,5,154,78,81,72,,,N,N,698,5134,1419,207,59,66,469,521,137,149,,,699,587,3.8300000000,85,4,6,4140,1459,89,438,416,218,153,0.96,Chicago Cubs,Wrigley Field,716922,101,101,CHC,CHN,CHN
+1924,NL,CIN,CIN,,4,153,76,83,70,,,N,N,649,5301,1539,236,111,36,349,334,103,98,,,579,478,3.1200000000,77,14,9,4134,1408,30,293,451,217,142,0.96,Cincinnati Reds,Crosley Field,473707,99,97,CIN,CIN,CIN
+1924,NL,NY1,SFG,,1,154,77,93,60,,,Y,N,857,5445,1634,269,81,95,467,479,82,53,,,641,554,3.6200000000,71,4,21,4134,1464,77,392,406,186,160,0.97,New York Giants,Polo Grounds IV,844068,98,95,NYG,NY1,NY1
+1924,NL,PHI,PHI,,7,152,76,55,96,,,N,N,676,5306,1459,256,56,94,382,452,57,67,,,849,733,4.8700000000,59,7,10,4062,1689,84,469,349,175,168,0.97,Philadelphia Phillies,Baker Bowl,299818,112,115,PHI,PHI,PHI
+1924,NL,PIT,PIT,,3,153,77,90,63,,,N,N,724,5288,1517,222,122,44,366,396,181,92,,,588,502,3.2700000000,85,15,5,4146,1387,42,323,364,183,161,0.97,Pittsburgh Pirates,Forbes Field,736883,102,99,PIT,PIT,PIT
+1924,NL,SLN,STL,,6,154,77,65,89,,,N,N,740,5349,1552,270,87,67,382,418,86,86,,,750,629,4.1500000000,79,7,6,4092,1528,70,486,393,191,162,0.96,St. Louis Cardinals,Sportsman's Park IV,272885,98,98,STL,SLN,SLN
+1925,AL,BOS,BOS,,8,152,75,47,105,,,N,N,639,5166,1375,257,64,41,513,422,42,56,,,922,732,4.9700000000,68,6,6,3978,1615,67,510,310,265,150,0.95,Boston Red Sox,Fenway Park I,267782,99,103,BOS,BOS,BOS
+1925,AL,CHA,CHW,,5,154,77,79,75,,,N,N,811,5224,1482,299,59,38,662,405,129,88,,,770,660,4.2900000000,71,12,13,4155,1579,69,489,374,198,162,0.96,Chicago White Sox,Comiskey Park,832231,95,95,CHW,CHA,CHA
+1925,AL,CLE,CLE,,6,155,77,70,84,,,N,N,782,5436,1613,285,58,52,520,379,90,77,,,817,684,4.4900000000,93,6,9,4116,1604,41,493,345,212,146,0.96,Cleveland Indians,League Park II,419005,101,101,CLE,CLE,CLE
+1925,AL,DET,DET,,4,156,77,81,73,,,N,N,903,5371,1621,277,84,50,640,386,97,63,,,829,708,4.6100000000,66,2,18,4149,1582,70,556,419,171,143,0.97,Detroit Tigers,Navin Field,820766,98,98,DET,DET,DET
+1925,AL,NYA,NYY,,7,156,79,69,85,,,N,N,706,5353,1471,247,74,110,470,482,67,73,,,774,667,4.3300000000,80,8,13,4161,1560,78,505,492,162,150,0.97,New York Yankees,Yankee Stadium I,697267,98,97,NYY,NYA,NYA
+1925,AL,PHA,OAK,,2,153,77,88,64,,,N,N,831,5399,1659,298,79,76,453,432,67,59,,,713,594,3.8700000000,61,8,18,4143,1468,60,544,495,214,148,0.96,Philadelphia Athletics,Shibe Park,869703,107,106,PHA,PHA,PHA
+1925,AL,SLA,BAL,,3,154,78,82,71,,,N,N,900,5440,1620,304,68,110,498,375,85,78,,,906,754,4.9200000000,67,7,10,4137,1588,99,675,419,219,164,0.96,St. Louis Browns,Sportsman's Park IV,462898,105,106,SLB,SLA,SLA
+1925,AL,WS1,MIN,,1,152,76,96,55,,,Y,N,829,5206,1577,251,71,56,533,427,134,88,,,670,558,3.7000000000,69,10,21,4074,1434,49,543,463,172,166,0.97,Washington Senators,Griffith Stadium I,817199,99,97,WSH,WS1,WS1
+1925,NL,BRO,LAD,,6,153,77,68,85,,,N,N,786,5468,1617,250,80,64,437,383,37,30,,,866,715,4.7700000000,82,4,4,4050,1608,75,477,518,209,130,0.96,Brooklyn Robins,Ebbets Field,659435,97,97,BRO,BRO,BRO
+1925,NL,BSN,ATL,,5,153,76,70,83,,,N,N,708,5365,1567,260,70,41,405,380,77,72,,,802,666,4.3900000000,77,5,4,4098,1567,67,458,351,221,145,0.96,Boston Braves,Braves Field,313528,91,94,BSN,BSN,BSN
+1925,NL,CHN,CHC,,8,154,77,68,86,,,N,N,723,5353,1473,254,70,86,397,470,94,70,,,773,671,4.4100000000,75,5,10,4110,1575,102,485,435,198,161,0.96,Chicago Cubs,Wrigley Field,622610,102,102,CHC,CHN,CHN
+1925,NL,CIN,CIN,,3,153,76,80,73,,,N,N,690,5233,1490,221,90,44,409,327,108,107,,,643,516,3.3800000000,92,11,12,4125,1447,35,324,437,203,161,0.96,Cincinnati Reds,Crosley Field,464920,98,96,CIN,CIN,CIN
+1925,NL,NY1,SFG,,2,152,76,86,66,,,N,N,736,5327,1507,239,61,114,411,494,79,65,,,702,593,3.9400000000,80,6,8,4062,1532,73,408,446,198,129,0.96,New York Giants,Polo Grounds IV,778993,96,95,NYG,NY1,NY1
+1925,NL,PHI,PHI,,6,153,77,68,85,,,N,N,812,5412,1598,288,58,100,456,542,48,59,,,930,753,5.0200000000,69,8,9,4050,1753,117,444,371,211,147,0.96,Philadelphia Phillies,Baker Bowl,304905,108,112,PHI,PHI,PHI
+1925,NL,PIT,PIT,,1,153,77,95,58,,,Y,Y,912,5372,1651,316,105,78,499,363,159,63,,,715,582,3.8700000000,77,2,13,4062,1526,81,387,386,224,171,0.96,Pittsburgh Pirates,Forbes Field,804354,106,104,PIT,PIT,PIT
+1925,NL,SLN,STL,,4,153,76,77,76,,,N,N,828,5329,1592,292,80,109,446,414,70,51,,,764,647,4.3600000000,82,8,7,4005,1480,86,470,428,204,156,0.96,St. Louis Cardinals,Sportsman's Park IV,404959,103,102,STL,SLN,SLN
+1926,AL,BOS,BOS,,8,154,77,46,107,,,N,N,562,5185,1325,249,54,32,465,450,48,51,,,835,714,4.7200000000,53,6,5,4086,1520,45,546,336,182,143,0.97,Boston Red Sox,Fenway Park I,285155,96,101,BOS,BOS,BOS
+1926,AL,CHA,CHW,,5,155,79,81,72,,,N,N,730,5220,1508,314,60,32,556,381,121,77,,,665,573,3.7400000000,85,11,12,4140,1426,47,506,458,161,122,0.97,Chicago White Sox,Comiskey Park,710339,97,97,CHW,CHA,CHA
+1926,AL,CLE,CLE,,2,154,80,88,66,,,N,N,738,5293,1529,333,49,27,455,332,88,43,,,612,519,3.4000000000,96,11,4,4122,1412,49,450,381,172,153,0.97,Cleveland Indians,League Park II,627426,101,101,CLE,CLE,CLE
+1926,AL,DET,DET,,6,157,81,79,75,,,N,N,793,5315,1547,281,90,36,599,423,88,72,,,830,683,4.4100000000,57,10,18,4182,1570,58,555,469,197,151,0.96,Detroit Tigers,Navin Field,711914,101,100,DET,DET,DET
+1926,AL,NYA,NYY,,1,155,75,91,63,,,Y,N,847,5221,1508,262,75,121,642,580,79,60,,,713,588,3.8600000000,63,4,20,4116,1442,56,478,486,209,117,0.96,New York Yankees,Yankee Stadium I,1027675,99,96,NYY,NYA,NYA
+1926,AL,PHA,OAK,,3,150,71,83,67,,,N,N,677,5046,1359,259,65,61,523,452,56,46,,,570,449,3.0000000000,62,10,16,4038,1362,38,451,571,173,131,0.97,Philadelphia Athletics,Shibe Park,714508,106,104,PHA,PHA,PHA
+1926,AL,SLA,BAL,,7,155,79,62,92,,,N,N,682,5259,1449,253,78,72,437,465,62,71,,,845,708,4.6600000000,64,5,9,4104,1549,86,654,337,228,167,0.96,St. Louis Browns,Sportsman's Park IV,283986,105,107,SLB,SLA,SLA
+1926,AL,WS1,MIN,,4,152,74,81,69,,,N,N,802,5223,1525,244,97,43,555,369,122,89,,,761,650,4.3400000000,65,5,26,4044,1489,45,566,418,184,129,0.96,Washington Senators,Griffith Stadium I,551580,98,96,WSH,WS1,WS1
+1926,NL,BRO,LAD,,6,155,76,71,82,,,N,N,623,5130,1348,246,62,40,475,464,76,,,,705,578,3.8200000000,83,5,9,4083,1440,50,472,517,228,95,0.96,Brooklyn Robins,Ebbets Field,650819,98,100,BRO,BRO,BRO
+1926,NL,BSN,ATL,,7,153,77,66,86,,,N,N,624,5216,1444,209,62,16,426,348,81,,,,719,608,4.0100000000,60,9,9,4095,1536,46,455,408,208,150,0.96,Boston Braves,Braves Field,303598,91,93,BSN,BSN,BSN
+1926,NL,CHN,CHC,,4,155,78,82,72,,,N,N,682,5229,1453,291,49,66,445,447,85,,,,602,499,3.2600000000,77,13,14,4134,1407,39,486,508,162,174,0.97,Chicago Cubs,Wrigley Field,885063,102,101,CHC,CHN,CHN
+1926,NL,CIN,CIN,,2,157,77,87,67,,,N,N,747,5320,1541,242,120,35,454,333,51,,,,651,535,3.4200000000,88,14,8,4224,1449,40,324,424,183,160,0.97,Cincinnati Reds,Crosley Field,672987,98,97,CIN,CIN,CIN
+1926,NL,NY1,SFG,,5,151,76,74,77,,,N,N,663,5167,1435,214,58,73,339,420,94,,,,668,562,3.7700000000,61,4,15,4023,1370,70,427,419,186,150,0.97,New York Giants,Polo Grounds IV,700362,99,98,NYG,NY1,NY1
+1926,NL,PHI,PHI,,8,152,76,58,93,,,N,N,687,5254,1479,244,50,75,422,479,47,,,,900,746,5.0300000000,68,5,5,4002,1699,68,454,331,223,153,0.96,Philadelphia Phillies,Baker Bowl,240600,108,112,PHI,PHI,PHI
+1926,NL,PIT,PIT,,3,157,79,84,69,,,N,N,769,5312,1514,243,106,44,434,350,91,,,,689,562,3.6700000000,83,12,18,4137,1422,50,455,387,220,161,0.96,Pittsburgh Pirates,Forbes Field,798542,105,102,PIT,PIT,PIT
+1926,NL,SLN,STL,,1,156,79,89,65,,,Y,Y,817,5381,1541,259,82,90,478,518,83,,,,678,570,3.6700000000,90,10,6,4194,1423,76,397,365,198,141,0.96,St. Louis Cardinals,Sportsman's Park IV,668428,104,103,STL,SLN,SLN
+1927,AL,BOS,BOS,,8,154,78,51,103,,,N,N,597,5207,1348,271,78,28,430,456,82,46,,,856,716,4.7200000000,63,6,7,4098,1603,56,558,381,222,162,0.96,Boston Red Sox,Fenway Park I,305275,98,102,BOS,BOS,BOS
+1927,AL,CHA,CHW,,5,153,75,70,83,,,N,N,662,5157,1433,285,61,36,493,389,90,75,,,708,594,3.9100000000,85,10,8,4101,1467,55,440,365,180,131,0.97,Chicago White Sox,Comiskey Park,614423,98,98,CHW,CHA,CHA
+1927,AL,CLE,CLE,,6,153,77,66,87,,,N,N,668,5202,1471,321,52,26,381,366,63,72,,,766,642,4.2700000000,72,5,8,4059,1542,37,508,366,202,146,0.96,Cleveland Indians,League Park II,373138,101,102,CLE,CLE,CLE
+1927,AL,DET,DET,,4,156,78,82,71,,,N,N,845,5299,1533,282,100,51,587,420,141,73,,,805,638,4.1400000000,75,5,17,4161,1542,52,577,421,209,173,0.96,Detroit Tigers,Navin Field,773716,101,102,DET,DET,DET
+1927,AL,NYA,NYY,,1,155,77,110,44,,,Y,Y,975,5347,1644,291,103,158,635,605,90,64,,,599,494,3.2000000000,82,11,20,4167,1403,42,409,431,196,123,0.96,New York Yankees,Yankee Stadium I,1164015,98,94,NYY,NYA,NYA
+1927,AL,PHA,OAK,,2,155,77,91,63,,,N,N,841,5296,1606,281,70,56,551,326,98,63,,,726,610,3.9700000000,66,8,25,4152,1467,65,442,553,194,124,0.96,Philadelphia Athletics,Shibe Park,605529,105,102,PHA,PHA,PHA
+1927,AL,SLA,BAL,,7,155,78,59,94,,,N,N,724,5220,1440,262,59,55,443,420,91,66,,,904,744,4.9500000000,80,4,8,4059,1592,79,604,385,248,166,0.96,St. Louis Browns,Sportsman's Park IV,247879,103,105,SLB,SLA,SLA
+1927,AL,WS1,MIN,,3,157,79,85,69,,,N,N,782,5389,1549,268,87,29,498,359,133,52,,,730,618,3.9700000000,62,10,23,4206,1434,53,491,497,194,125,0.96,Washington Senators,Griffith Stadium I,528976,99,98,WSH,WS1,WS1
+1927,NL,BRO,LAD,,6,154,74,65,88,,,N,N,541,5193,1314,195,74,39,368,494,106,,,,619,513,3.3600000000,74,7,10,4125,1382,63,418,574,226,117,0.96,Brooklyn Robins,Ebbets Field,637230,100,100,BRO,BRO,BRO
+1927,NL,BSN,ATL,,7,155,74,60,94,,,N,N,651,5370,1498,216,61,37,346,363,100,,,,771,652,4.2200000000,52,3,11,4170,1602,43,468,402,231,130,0.96,Boston Braves,Braves Field,288685,92,94,BSN,BSN,BSN
+1927,NL,CHN,CHC,,4,153,78,85,68,,,N,N,750,5303,1505,266,63,74,481,492,65,,,,661,562,3.6500000000,75,11,5,4155,1439,50,514,465,181,152,0.97,Chicago Cubs,Wrigley Field,1159168,101,99,CHC,CHN,CHN
+1927,NL,CIN,CIN,,5,153,80,75,78,,,N,N,643,5185,1439,222,77,29,402,332,62,,,,653,538,3.5400000000,87,12,12,4104,1472,36,316,407,165,160,0.97,Cincinnati Reds,Crosley Field,442164,97,97,CIN,CIN,CIN
+1927,NL,NY1,SFG,,3,155,74,92,62,,,N,N,817,5372,1594,251,62,109,461,462,73,,,,720,609,3.9700000000,65,7,16,4143,1520,77,453,442,195,160,0.96,New York Giants,Polo Grounds IV,858190,99,98,NYG,NY1,NY1
+1927,NL,PHI,PHI,,8,155,78,51,103,,,N,N,678,5317,1487,216,46,57,434,482,68,,,,903,807,5.3600000000,81,5,6,4065,1710,84,462,377,168,152,0.97,Philadelphia Phillies,Baker Bowl,305420,95,99,PHI,PHI,PHI
+1927,NL,PIT,PIT,,1,156,79,94,60,,,Y,N,817,5397,1648,258,78,54,437,355,65,,,,659,563,3.6600000000,90,10,10,4155,1400,58,418,435,187,130,0.96,Pittsburgh Pirates,Forbes Field,869720,107,105,PIT,PIT,PIT
+1927,NL,SLN,STL,,2,153,80,92,61,,,N,N,754,5207,1450,264,79,84,484,511,110,,,,665,542,3.5700000000,89,14,11,4101,1416,72,363,394,213,170,0.96,St. Louis Cardinals,Sportsman's Park IV,749340,105,102,STL,SLN,SLN
+1928,AL,BOS,BOS,,8,154,74,57,96,,,N,N,589,5132,1356,260,62,38,389,512,99,64,,,770,659,4.3900000000,70,5,9,4056,1492,49,452,407,172,139,0.97,Boston Red Sox,Fenway Park I,396920,96,100,BOS,BOS,BOS
+1928,AL,CHA,CHW,,5,155,78,72,82,,,N,N,656,5207,1405,231,77,24,469,488,139,82,,,725,609,3.9800000000,88,6,11,4134,1518,66,501,418,184,149,0.97,Chicago White Sox,Comiskey Park,494152,98,100,CHW,CHA,CHA
+1928,AL,CLE,CLE,,7,155,77,62,92,,,N,N,674,5386,1535,299,61,34,377,426,50,52,,,830,684,4.4700000000,71,4,15,4134,1615,52,511,416,221,187,0.96,Cleveland Indians,League Park II,375907,101,102,CLE,CLE,CLE
+1928,AL,DET,DET,,6,154,77,68,86,,,N,N,744,5292,1476,265,97,62,469,438,113,77,,,804,659,4.3200000000,65,5,16,4116,1481,58,567,451,212,140,0.96,Detroit Tigers,Navin Field,474323,102,102,DET,DET,DET
+1928,AL,NYA,NYY,,1,154,77,101,53,,,Y,Y,894,5337,1578,269,79,133,562,544,51,51,,,685,571,3.7400000000,82,13,21,4125,1466,59,452,487,199,136,0.96,New York Yankees,Yankee Stadium I,1072132,98,94,NYY,NYA,NYA
+1928,AL,PHA,OAK,,2,153,77,98,55,,,N,N,829,5226,1540,323,75,89,533,442,59,48,,,615,510,3.3600000000,81,15,16,4101,1349,66,424,607,183,124,0.97,Philadelphia Athletics,Shibe Park,689756,103,99,PHA,PHA,PHA
+1928,AL,SLA,BAL,,3,154,77,82,72,,,N,N,772,5217,1431,276,76,63,548,479,76,43,,,742,637,4.1700000000,80,6,15,4122,1487,93,454,456,184,146,0.97,St. Louis Browns,Sportsman's Park IV,339497,103,104,SLB,SLA,SLA
+1928,AL,WS1,MIN,,4,155,80,75,79,,,N,N,718,5320,1510,277,93,40,481,390,110,59,,,705,597,3.8800000000,77,15,10,4152,1420,40,466,462,180,146,0.97,Washington Senators,Griffith Stadium I,378501,99,99,WSH,WS1,WS1
+1928,NL,BRO,LAD,,6,155,77,77,76,,,N,N,665,5243,1393,229,70,66,557,510,81,,,,640,504,3.2500000000,75,16,15,4188,1378,59,468,551,217,113,0.96,Brooklyn Robins,Ebbets Field,664863,99,100,BRO,BRO,BRO
+1928,NL,BSN,ATL,,7,153,76,50,103,,,N,N,631,5228,1439,241,41,52,447,377,60,,,,878,730,4.8300000000,54,1,6,4080,1596,100,524,343,193,141,0.96,Boston Braves,Braves Field,227001,94,97,BSN,BSN,BSN
+1928,NL,CHN,CHC,,3,154,77,91,63,,,N,N,714,5260,1460,251,64,92,508,517,83,,,,615,521,3.4000000000,75,12,14,4140,1383,56,508,531,156,176,0.97,Chicago Cubs,Wrigley Field,1143740,99,97,CHC,CHN,CHN
+1928,NL,CIN,CIN,,5,153,78,78,74,,,N,N,648,5184,1449,229,67,32,386,330,83,,,,686,600,3.9400000000,68,11,11,4113,1516,58,410,355,162,194,0.97,Cincinnati Reds,Crosley Field,490490,98,99,CIN,CIN,CIN
+1928,NL,NY1,SFG,,2,155,77,93,61,,,N,N,807,5459,1600,276,59,118,444,376,62,,,,653,568,3.6700000000,79,7,16,4182,1454,77,405,399,178,175,0.97,New York Giants,Polo Grounds IV,916191,100,98,NYG,NY1,NY1
+1928,NL,PHI,PHI,,8,152,75,43,109,,,N,N,660,5234,1396,257,47,85,503,510,53,,,,957,835,5.5600000000,42,4,11,4056,1660,108,675,404,179,171,0.97,Philadelphia Phillies,Baker Bowl,182168,106,110,PHI,PHI,PHI
+1928,NL,PIT,PIT,,4,152,77,85,67,,,N,N,837,5371,1659,246,100,52,435,352,64,,,,704,594,3.9500000000,82,8,11,4062,1422,66,446,385,201,123,0.96,Pittsburgh Pirates,Forbes Field,495070,104,102,PIT,PIT,PIT
+1928,NL,SLN,STL,,1,154,77,95,59,,,Y,N,807,5357,1505,292,70,113,568,438,82,,,,636,531,3.3800000000,83,4,21,4245,1470,86,399,422,152,134,0.97,St. Louis Cardinals,Sportsman's Park IV,761574,103,102,STL,SLN,SLN
+1929,AL,BOS,BOS,,8,155,78,58,96,,,N,N,605,5160,1377,285,69,28,413,494,85,80,,,803,672,4.4300000000,84,9,5,4098,1537,78,496,416,218,159,0.96,Boston Red Sox,Fenway Park I,394620,97,100,BOS,BOS,BOS
+1929,AL,CHA,CHW,,7,152,76,59,93,,,N,N,627,5248,1406,240,74,37,425,436,109,65,,,792,665,4.4100000000,78,5,7,4071,1481,84,505,328,190,153,0.96,Chicago White Sox,Comiskey Park,426795,98,100,CHW,CHA,CHA
+1929,AL,CLE,CLE,,3,152,76,81,71,,,N,N,717,5187,1525,294,79,62,453,363,75,85,,,736,608,4.0500000000,80,8,10,4056,1570,56,488,389,198,162,0.96,Cleveland Indians,League Park II,536210,103,105,CLE,CLE,CLE
+1929,AL,DET,DET,,6,155,77,70,84,,,N,N,926,5592,1671,339,97,110,521,496,95,72,,,928,766,4.9600000000,82,5,9,4170,1641,73,646,467,242,149,0.96,Detroit Tigers,Navin Field,869318,101,101,DET,DET,DET
+1929,AL,NYA,NYY,,2,154,77,88,66,,,N,N,899,5379,1587,262,74,142,554,518,51,44,,,775,636,4.1900000000,64,12,18,4098,1475,83,485,484,179,152,0.97,New York Yankees,Yankee Stadium I,960148,94,92,NYY,NYA,NYA
+1929,AL,PHA,OAK,,1,151,74,104,46,,,Y,Y,901,5204,1539,288,76,122,543,440,61,38,,,615,519,3.4400000000,70,9,24,4071,1371,73,487,573,148,117,0.97,Philadelphia Athletics,Shibe Park,839176,104,100,PHA,PHA,PHA
+1929,AL,SLA,BAL,,4,154,77,79,73,,,N,N,733,5174,1426,277,64,46,589,431,72,46,,,713,622,4.0800000000,83,15,10,4113,1469,100,462,415,156,148,0.97,St. Louis Browns,Sportsman's Park IV,280697,104,104,SLB,SLA,SLA
+1929,AL,WS1,MIN,,5,153,78,71,81,,,N,N,730,5237,1445,244,66,48,556,400,86,61,,,776,653,4.3400000000,62,3,17,4062,1429,48,496,494,191,156,0.96,Washington Senators,Griffith Stadium I,355506,101,100,WSH,WS1,WS1
+1929,NL,BRO,LAD,,6,153,77,70,83,,,N,N,755,5273,1535,282,69,99,504,454,80,,,,888,742,4.9200000000,59,8,16,4074,1553,92,549,549,192,113,0.96,Brooklyn Robins,Ebbets Field,731886,98,98,BRO,BRO,BRO
+1929,NL,BSN,ATL,,8,154,77,56,98,,,N,N,657,5291,1481,252,77,33,408,432,65,,,,876,769,5.1200000000,78,4,12,4056,1604,103,530,366,204,146,0.96,Boston Braves,Braves Field,372351,95,98,BSN,BSN,BSN
+1929,NL,CHN,CHC,,1,156,78,98,54,,,Y,N,982,5471,1655,310,46,139,589,567,103,,,,758,646,4.1600000000,79,14,21,4194,1542,77,537,548,154,169,0.97,Chicago Cubs,Wrigley Field,1485166,100,97,CHC,CHN,CHN
+1929,NL,CIN,CIN,,7,155,78,66,88,,,N,N,686,5269,1478,258,79,34,412,347,134,,,,760,671,4.4100000000,75,5,8,4107,1558,61,413,347,162,148,0.97,Cincinnati Reds,Crosley Field,295040,95,97,CIN,CIN,CIN
+1929,NL,NY1,SFG,,3,152,77,84,67,,,N,N,897,5388,1594,251,47,136,482,405,85,,,,709,605,3.9700000000,68,9,13,4116,1536,102,387,431,158,163,0.97,New York Giants,Polo Grounds IV,868806,100,97,NYG,NY1,NY1
+1929,NL,PHI,PHI,,5,154,76,71,82,,,N,N,897,5484,1693,305,51,153,573,470,59,,,,1032,918,6.1300000000,45,5,24,4044,1743,122,616,369,189,153,0.96,Philadelphia Phillies,Baker Bowl,281200,106,110,PHI,PHI,PHI
+1929,NL,PIT,PIT,,2,154,76,88,65,,,N,N,904,5490,1663,285,116,60,503,335,94,,,,780,668,4.3600000000,79,5,13,4137,1530,96,439,409,181,136,0.97,Pittsburgh Pirates,Forbes Field,491377,103,102,PIT,PIT,PIT
+1929,NL,SLN,STL,,4,154,77,78,74,,,N,N,831,5364,1569,310,84,100,490,455,72,,,,806,704,4.6600000000,83,6,8,4077,1604,101,474,453,174,149,0.97,St. Louis Cardinals,Sportsman's Park IV,399887,103,101,STL,SLN,SLN
+1930,AL,BOS,BOS,,8,154,76,52,102,,,N,N,612,5286,1393,257,68,47,358,552,42,35,,,814,707,4.6800000000,78,4,5,4080,1515,75,488,356,194,161,0.96,Boston Red Sox,Fenway Park I,444045,96,99,BOS,BOS,BOS
+1930,AL,CHA,CHW,,7,154,78,62,92,,,N,N,729,5419,1496,256,90,63,389,479,74,40,,,884,712,4.7100000000,63,2,10,4083,1629,74,407,471,233,136,0.96,Chicago White Sox,Comiskey Park,406123,96,99,CHW,CHA,CHA
+1930,AL,CLE,CLE,,4,154,77,81,73,,,N,N,890,5439,1654,358,59,72,490,461,51,47,,,915,737,4.8800000000,68,5,14,4080,1663,85,528,441,235,156,0.96,Cleveland Indians,League Park II,528657,103,103,CLE,CLE,CLE
+1930,AL,DET,DET,,5,154,78,75,79,,,N,N,783,5297,1504,298,90,82,461,508,98,70,,,833,706,4.7000000000,68,4,17,4053,1507,86,570,574,188,156,0.96,Detroit Tigers,Navin Field,649450,102,104,DET,DET,DET
+1930,AL,NYA,NYY,,3,154,76,86,68,,,N,N,1062,5448,1683,298,110,152,644,569,91,60,,,898,741,4.8800000000,65,7,15,4101,1566,93,524,572,208,132,0.96,New York Yankees,Yankee Stadium I,1169230,96,93,NYY,NYA,NYA
+1930,AL,PHA,OAK,,1,154,76,102,52,,,Y,Y,951,5345,1573,319,74,125,599,531,48,33,,,751,652,4.2800000000,72,8,21,4113,1457,84,488,672,145,121,0.97,Philadelphia Athletics,Shibe Park,721663,105,101,PHA,PHA,PHA
+1930,AL,SLA,BAL,,6,154,78,64,90,,,N,N,751,5278,1415,289,67,75,497,550,93,71,,,886,772,5.0700000000,68,5,10,4113,1639,124,449,470,189,152,0.96,St. Louis Browns,Sportsman's Park IV,152088,103,104,SLB,SLA,SLA
+1930,AL,WS1,MIN,,2,154,77,94,60,,,N,N,892,5370,1620,300,98,57,537,438,101,67,,,689,602,3.9600000000,78,6,14,4107,1367,52,504,524,155,150,0.97,Washington Senators,Griffith Stadium I,614474,101,99,WSH,WS1,WS1
+1930,NL,BRO,LAD,,4,154,77,86,68,,,N,N,871,5433,1654,303,73,122,481,541,53,,,,738,614,4.0300000000,74,13,15,4116,1480,115,394,526,169,167,0.97,Brooklyn Robins,Ebbets Field,1097329,99,99,BRO,BRO,BRO
+1930,NL,BSN,ATL,,6,154,77,70,84,,,N,N,693,5356,1503,246,78,66,332,397,69,,,,835,743,4.9100000000,71,6,11,4083,1624,117,475,424,178,167,0.97,Boston Braves,Braves Field,464835,97,99,BSN,BSN,BSN
+1930,NL,CHN,CHC,,2,156,79,90,64,,,N,N,998,5581,1722,305,72,171,588,635,70,,,,870,748,4.8000000000,67,6,12,4209,1642,111,528,601,170,167,0.97,Chicago Cubs,Wrigley Field,1463624,101,98,CHC,CHN,CHN
+1930,NL,CIN,CIN,,7,154,77,59,95,,,N,N,665,5245,1475,265,67,74,445,489,48,,,,857,754,5.0800000000,61,6,11,4005,1650,75,394,361,161,164,0.97,Cincinnati Reds,Crosley Field,386727,94,97,CIN,CIN,CIN
+1930,NL,NY1,SFG,,3,154,77,87,67,,,N,N,959,5553,1769,264,83,143,422,382,59,,,,814,698,4.6100000000,64,6,19,4089,1546,117,439,522,164,164,0.97,New York Giants,Polo Grounds IV,868714,98,95,NYG,NY1,NY1
+1930,NL,PHI,PHI,,8,156,77,52,102,,,N,N,944,5667,1783,345,44,126,450,459,34,,,,1199,1023,6.7100000000,54,3,7,4116,1993,142,543,384,239,169,0.96,Philadelphia Phillies,Baker Bowl,299007,107,110,PHI,PHI,PHI
+1930,NL,PIT,PIT,,5,154,77,80,74,,,N,N,891,5346,1622,285,119,86,494,449,76,,,,928,792,5.2400000000,80,7,13,4083,1730,128,438,393,216,164,0.96,Pittsburgh Pirates,Forbes Field,357795,100,100,PIT,PIT,PIT
+1930,NL,SLN,STL,,1,154,77,92,62,,,Y,N,1004,5512,1732,373,89,104,479,496,72,,,,784,673,4.3900000000,63,5,21,4140,1596,87,476,639,183,176,0.97,St. Louis Cardinals,Sportsman's Park IV,508501,104,102,STL,SLN,SLN
+1931,AL,BOS,BOS,,6,153,80,62,90,,,N,N,625,5379,1409,289,34,37,405,565,42,43,,,800,698,4.6000000000,61,5,10,4098,1559,54,473,365,187,127,0.97,Boston Red Sox,Fenway Park I,350975,94,98,BOS,BOS,BOS
+1931,AL,CHA,CHW,,8,156,77,56,97,,,N,N,704,5481,1423,238,69,27,483,445,94,39,,,939,778,5.0400000000,54,6,10,4170,1613,82,588,421,246,131,0.96,Chicago White Sox,Comiskey Park,403550,94,97,CHW,CHA,CHA
+1931,AL,CLE,CLE,,4,155,76,78,76,,,N,N,885,5445,1612,321,69,71,555,433,63,60,,,833,697,4.6300000000,76,6,9,4062,1577,64,561,470,231,143,0.96,Cleveland Indians,League Park II,483027,105,105,CLE,CLE,CLE
+1931,AL,DET,DET,,7,154,77,61,93,,,N,N,651,5430,1456,292,69,43,480,468,117,75,,,836,706,4.5900000000,86,5,6,4152,1549,79,597,511,221,139,0.96,Detroit Tigers,Navin Field,434056,103,104,DET,DET,DET
+1931,AL,NYA,NYY,,2,155,77,94,59,,,N,N,1067,5608,1667,277,78,155,748,554,138,68,,,760,658,4.2000000000,78,4,17,4230,1461,67,543,686,167,131,0.97,New York Yankees,Yankee Stadium I,912437,95,91,NYY,NYA,NYA
+1931,AL,PHA,OAK,,1,153,75,107,45,,,Y,N,858,5377,1544,311,64,118,528,543,25,23,,,626,526,3.4700000000,97,12,16,4095,1342,73,457,574,140,151,0.97,Philadelphia Athletics,Shibe Park,627464,106,103,PHA,PHA,PHA
+1931,AL,SLA,BAL,,5,154,77,63,91,,,N,N,722,5374,1455,287,62,76,488,580,73,80,,,870,720,4.7600000000,65,4,10,4086,1623,84,444,436,231,160,0.96,St. Louis Browns,Sportsman's Park IV,179126,103,105,SLB,SLA,SLA
+1931,AL,WS1,MIN,,3,156,79,92,62,,,N,N,843,5576,1588,308,93,49,481,459,72,64,,,691,582,3.7600000000,60,7,24,4182,1434,73,498,582,142,148,0.97,Washington Senators,Griffith Stadium I,492657,101,98,WSH,WS1,WS1
+1931,NL,BRO,LAD,,4,153,76,79,73,,,N,N,681,5309,1464,240,77,71,409,512,45,,,,673,579,3.8400000000,64,9,18,4068,1520,56,351,546,187,154,0.96,Brooklyn Robins,Ebbets Field,753133,100,99,BRO,BRO,BRO
+1931,NL,BSN,ATL,,7,156,78,64,90,,,N,N,533,5296,1367,221,59,34,368,430,46,,,,680,598,3.9000000000,78,12,9,4140,1465,66,406,419,167,141,0.97,Boston Braves,Braves Field,515005,97,99,BSN,BSN,BSN
+1931,NL,CHN,CHC,,3,156,77,84,70,,,N,N,828,5451,1578,340,66,84,577,641,49,,,,710,611,3.9700000000,80,8,8,4155,1448,54,524,541,169,141,0.97,Chicago Cubs,Wrigley Field,1086422,102,100,CHC,CHN,CHN
+1931,NL,CIN,CIN,,8,154,77,58,96,,,N,N,592,5343,1439,241,70,21,403,463,24,,,,742,631,4.2200000000,70,7,6,4035,1545,51,399,317,165,194,0.97,Cincinnati Reds,Crosley Field,263316,93,96,CIN,CIN,CIN
+1931,NL,NY1,SFG,,2,153,78,87,65,,,N,N,768,5372,1554,251,64,101,383,395,83,,,,599,499,3.3000000000,90,17,12,4080,1341,71,422,570,159,126,0.97,New York Giants,Polo Grounds IV,812163,97,95,NYG,NY1,NY1
+1931,NL,PHI,PHI,,6,155,76,66,88,,,N,N,684,5375,1502,299,52,81,437,492,42,,,,828,692,4.5800000000,60,6,16,4080,1603,75,511,499,204,149,0.96,Philadelphia Phillies,Baker Bowl,284849,108,110,PHI,PHI,PHI
+1931,NL,PIT,PIT,,5,155,78,75,79,,,N,N,636,5360,1425,243,70,41,493,454,59,,,,691,565,3.6600000000,89,9,5,4170,1489,55,442,345,193,167,0.96,Pittsburgh Pirates,Forbes Field,260392,98,99,PIT,PIT,PIT
+1931,NL,SLN,STL,,1,154,78,101,53,,,Y,Y,815,5435,1554,353,74,60,432,475,114,,,,614,531,3.4500000000,80,17,20,4152,1470,65,449,626,160,169,0.97,St. Louis Cardinals,Sportsman's Park IV,608535,105,103,STL,SLN,SLN
+1932,AL,BOS,BOS,,8,154,77,43,111,,,N,N,566,5295,1331,253,57,53,469,539,46,46,,,915,760,5.0200000000,42,3,7,4086,1574,79,612,365,231,165,0.96,Boston Red Sox,Fenway Park I,182150,95,99,BOS,BOS,BOS
+1932,AL,CHA,CHW,,7,152,77,49,102,,,N,N,667,5336,1426,274,56,36,459,386,89,58,,,897,722,4.8200000000,50,2,12,4044,1551,88,580,379,267,170,0.95,Chicago White Sox,Comiskey Park,233198,93,96,CHW,CHA,CHA
+1932,AL,CLE,CLE,,4,153,77,87,65,,,N,N,845,5412,1544,310,74,78,566,454,52,54,,,747,630,4.1200000000,94,6,8,4131,1506,70,446,439,189,129,0.96,Cleveland Indians,League Park II/Cleveland Stadium,468953,109,107,CLE,CLE,CLE
+1932,AL,DET,DET,,5,153,78,76,75,,,N,N,799,5409,1479,291,80,80,486,523,103,49,,,787,651,4.3000000000,67,9,17,4086,1421,89,592,521,186,154,0.96,Detroit Tigers,Navin Field,397157,104,105,DET,DET,DET
+1932,AL,NYA,NYY,,1,156,77,107,47,,,Y,Y,1002,5477,1564,279,82,160,766,527,77,66,,,724,623,3.9800000000,96,11,15,4224,1425,93,561,780,190,124,0.96,New York Yankees,Yankee Stadium I,962320,96,92,NYY,NYA,NYA
+1932,AL,PHA,OAK,,2,154,77,94,60,,,N,N,981,5537,1606,303,52,172,647,630,38,23,,,752,685,4.4500000000,95,10,10,4158,1477,112,511,595,123,142,0.98,Philadelphia Athletics,Shibe Park,405500,104,101,PHA,PHA,PHA
+1932,AL,SLA,BAL,,6,154,75,63,91,,,N,N,736,5449,1502,274,69,67,507,528,69,62,,,898,766,5.0100000000,63,7,11,4128,1592,103,574,496,187,156,0.96,St. Louis Browns,Sportsman's Park IV,112558,105,108,SLB,SLA,SLA
+1932,AL,WS1,MIN,,3,154,77,93,61,,,N,N,840,5515,1565,303,100,61,505,442,70,47,,,716,639,4.1600000000,66,11,22,4149,1463,73,526,437,125,157,0.97,Washington Senators,Griffith Stadium I,371396,99,96,WSH,WS1,WS1
+1932,NL,BRO,LAD,,3,154,78,81,73,,,N,N,752,5433,1538,296,59,110,388,574,61,,,,747,654,4.2700000000,61,7,16,4137,1538,72,403,497,182,169,0.97,Brooklyn Dodgers,Ebbets Field,681827,98,98,BRO,BRO,BRO
+1932,NL,BSN,ATL,,5,155,77,77,77,,,N,N,649,5506,1460,262,53,63,347,496,36,,,,655,555,3.5300000000,72,8,8,4242,1483,61,420,440,152,145,0.97,Boston Braves,Braves Field,507606,96,97,BSN,BSN,BSN
+1932,NL,CHN,CHC,,1,154,77,90,64,,,Y,N,720,5462,1519,296,60,69,398,514,48,,,,633,535,3.4400000000,79,9,7,4203,1444,68,409,527,173,146,0.97,Chicago Cubs,Wrigley Field,974688,100,98,CHC,CHN,CHN
+1932,NL,CIN,CIN,,8,155,77,60,94,,,N,N,575,5443,1429,265,68,47,436,436,35,,,,715,587,3.7900000000,83,6,6,4182,1505,69,276,359,178,129,0.97,Cincinnati Reds,Crosley Field,356950,96,99,CIN,CIN,CIN
+1932,NL,NY1,SFG,,6,154,77,72,82,,,N,N,755,5530,1527,263,54,116,348,391,31,,,,706,585,3.8300000000,57,3,16,4125,1533,112,387,506,191,143,0.96,New York Giants,Polo Grounds IV,484868,98,96,NYG,NY1,NY1
+1932,NL,PHI,PHI,,4,154,77,78,76,,,N,N,844,5510,1608,330,67,122,446,547,71,,,,796,687,4.4700000000,59,4,17,4152,1589,107,450,459,194,133,0.96,Philadelphia Phillies,Baker Bowl,268914,112,114,PHI,PHI,PHI
+1932,NL,PIT,PIT,,2,154,76,86,68,,,N,N,701,5421,1543,274,90,48,358,385,71,,,,711,574,3.7500000000,71,12,12,4131,1472,86,338,377,185,124,0.96,Pittsburgh Pirates,Forbes Field,287262,97,97,PIT,PIT,PIT
+1932,NL,SLN,STL,,6,156,79,72,82,,,N,N,684,5458,1467,307,51,76,420,514,92,,,,717,616,3.9700000000,70,13,9,4188,1533,76,455,681,168,155,0.97,St. Louis Cardinals,Sportsman's Park IV,279219,104,102,STL,SLN,SLN
+1933,AL,BOS,BOS,,7,149,72,63,86,,,N,N,700,5201,1407,294,56,50,525,464,58,37,,,758,641,4.3500000000,60,4,14,3981,1396,75,591,467,206,133,0.96,Boston Red Sox,Fenway Park I,268715,103,103,BOS,BOS,BOS
+1933,AL,CHA,CHW,,6,151,77,67,83,,,N,N,683,5318,1448,231,53,43,538,416,43,46,,,814,678,4.4500000000,53,8,13,4113,1505,85,519,423,186,143,0.97,Chicago White Sox,Comiskey Park,397789,95,98,CHW,CHA,CHA
+1933,AL,CLE,CLE,,4,151,77,75,76,,,N,N,654,5240,1366,218,77,50,448,426,36,40,,,669,556,3.7100000000,74,12,7,4050,1382,60,465,437,157,127,0.97,Cleveland Indians,League Park II/Cleveland Stadium,387936,106,106,CLE,CLE,CLE
+1933,AL,DET,DET,,5,155,78,75,79,,,N,N,722,5502,1479,283,78,57,475,523,68,50,,,733,614,3.9500000000,69,6,17,4194,1415,84,561,575,178,167,0.97,Detroit Tigers,Navin Field,320972,102,101,DET,DET,DET
+1933,AL,NYA,NYY,,2,152,75,91,59,,,N,N,927,5274,1495,241,75,144,700,506,76,59,,,768,656,4.3600000000,70,8,22,4062,1426,66,612,711,164,122,0.97,New York Yankees,Yankee Stadium I,728014,95,91,NYY,NYA,NYA
+1933,AL,PHA,OAK,,3,152,76,79,72,,,N,N,875,5330,1519,298,56,139,625,618,34,34,,,853,718,4.8100000000,69,6,14,4029,1523,77,644,423,203,121,0.96,Philadelphia Athletics,Shibe Park,297138,101,100,PHA,PHA,PHA
+1933,AL,SLA,BAL,,8,153,77,55,96,,,N,N,669,5285,1337,244,64,64,520,556,72,60,,,820,728,4.8200000000,55,7,10,4080,1574,96,531,426,150,162,0.97,St. Louis Browns,Sportsman's Park IV,88113,106,108,SLB,SLA,SLA
+1933,AL,WS1,MIN,,1,153,76,99,53,,,Y,N,850,5524,1586,281,86,60,539,395,65,50,,,665,590,3.8200000000,68,5,26,4167,1415,64,452,447,133,149,0.97,Washington Senators,Griffith Stadium I,437533,99,98,WSH,WS1,WS1
+1933,NL,BRO,LAD,,6,157,80,65,88,,,N,N,617,5367,1413,224,51,62,397,453,82,,,,695,574,3.7300000000,71,9,10,4158,1502,51,374,415,177,120,0.97,Brooklyn Dodgers,Ebbets Field,526815,96,96,BRO,BRO,BRO
+1933,NL,BSN,ATL,,4,156,77,83,71,,,N,N,552,5243,1320,217,56,54,326,428,25,,,,531,461,2.9600000000,85,15,16,4209,1391,54,355,383,130,148,0.97,Boston Braves,Braves Field,517803,92,92,BSN,BSN,BSN
+1933,NL,CHN,CHC,,3,154,79,86,68,,,N,N,646,5255,1422,256,51,72,392,475,52,,,,536,443,2.9300000000,95,16,9,4086,1316,51,413,488,168,163,0.97,Chicago Cubs,Wrigley Field,594112,100,99,CHC,CHN,CHN
+1933,NL,CIN,CIN,,8,153,79,58,94,,,N,N,496,5156,1267,208,37,34,349,354,30,,,,643,514,3.4200000000,74,13,8,4056,1470,47,257,310,177,139,0.97,Cincinnati Reds,Crosley Field,218281,97,101,CIN,CIN,CIN
+1933,NL,NY1,SFG,,1,156,77,91,61,,,Y,Y,636,5461,1437,204,41,82,377,477,31,,,,515,424,2.7100000000,75,23,15,4224,1280,61,400,555,178,156,0.97,New York Giants,Polo Grounds IV,604471,99,97,NYG,NY1,NY1
+1933,NL,PHI,PHI,,7,152,72,60,92,,,N,N,607,5261,1439,240,41,60,381,479,55,,,,760,644,4.3400000000,52,10,13,4008,1563,87,410,341,183,156,0.97,Philadelphia Phillies,Baker Bowl,156421,112,114,PHI,PHI,PHI
+1933,NL,PIT,PIT,,2,154,77,87,67,,,N,N,667,5429,1548,249,84,39,366,334,34,,,,619,499,3.2700000000,70,16,12,4119,1417,54,313,401,166,133,0.97,Pittsburgh Pirates,Forbes Field,288747,100,99,PIT,PIT,PIT
+1933,NL,SLN,STL,,5,154,77,82,71,,,N,N,687,5387,1486,256,61,57,391,528,99,,,,609,517,3.3700000000,73,11,16,4146,1391,55,452,635,160,119,0.97,St. Louis Cardinals,Sportsman's Park IV,256171,106,105,STL,SLN,SLN
+1934,AL,BOS,BOS,,4,153,77,76,76,,,N,N,820,5339,1465,287,70,51,610,535,116,47,,,775,653,4.3200000000,68,9,9,4083,1527,70,543,538,189,141,0.96,Boston Red Sox,Fenway Park II,610640,105,105,BOS,BOS,BOS
+1934,AL,CHA,CHW,,8,153,75,53,99,,,N,N,704,5301,1395,237,40,71,565,524,36,27,,,946,815,5.4100000000,72,5,8,4065,1599,139,628,506,207,126,0.96,Chicago White Sox,Comiskey Park,236559,103,105,CHW,CHA,CHA
+1934,AL,CLE,CLE,,3,154,78,85,69,,,N,N,814,5396,1550,340,46,100,526,433,52,32,,,763,650,4.2800000000,72,8,19,4101,1476,70,582,554,170,164,0.97,Cleveland Indians,League Park II/Cleveland Stadium,391338,101,100,CLE,CLE,CLE
+1934,AL,DET,DET,,1,154,80,101,53,,,Y,N,958,5475,1644,349,53,74,639,528,125,55,,,708,618,4.0600000000,74,13,14,4110,1467,86,488,640,156,150,0.97,Detroit Tigers,Navin Field,919161,100,98,DET,DET,DET
+1934,AL,NYA,NYY,,2,154,77,94,60,,,N,N,842,5368,1494,226,61,135,700,597,71,46,,,669,577,3.7600000000,83,13,10,4146,1349,71,542,656,160,151,0.97,New York Yankees,Yankee Stadium I,854682,94,91,NYY,NYA,NYA
+1934,AL,PHA,OAK,,5,153,76,68,82,,,N,N,764,5317,1491,236,50,144,491,584,57,35,,,838,744,5.0100000000,68,8,8,4011,1429,84,693,480,196,166,0.96,Philadelphia Athletics,Shibe Park,305847,96,98,PHA,PHA,PHA
+1934,AL,SLA,BAL,,6,154,76,67,85,,,N,N,674,5288,1417,252,59,62,514,631,43,31,,,800,674,4.4900000000,50,6,20,4050,1499,94,632,499,187,160,0.96,St. Louis Browns,Sportsman's Park IV,115305,107,110,SLB,SLA,SLA
+1934,AL,WS1,MIN,,7,155,76,66,86,,,N,N,729,5448,1512,278,70,51,570,447,47,42,,,806,718,4.6800000000,61,4,12,4143,1622,74,503,412,161,167,0.97,Washington Senators,Griffith Stadium I,330074,96,96,WSH,WS1,WS1
+1934,NL,BRO,LAD,,6,153,77,71,81,,,N,N,748,5427,1526,284,52,79,548,555,55,,,,795,674,4.4800000000,66,6,12,4062,1540,81,475,520,180,141,0.97,Brooklyn Dodgers,Ebbets Field,434188,95,96,BRO,BRO,BRO
+1934,NL,BSN,ATL,,4,152,75,78,73,,,N,N,683,5370,1460,233,44,83,375,440,30,,,,714,621,4.1100000000,62,12,20,4077,1512,78,405,462,169,120,0.97,Boston Braves,Braves Field,303205,92,94,BSN,BSN,BSN
+1934,NL,CHN,CHC,,3,152,77,86,65,,,N,N,705,5347,1494,263,44,101,375,630,59,,,,639,569,3.7600000000,73,11,9,4083,1432,80,417,633,137,135,0.97,Chicago Cubs,Wrigley Field,707525,99,96,CHC,CHN,CHN
+1934,NL,CIN,CIN,,8,152,78,52,99,,,N,N,590,5361,1428,227,65,55,313,532,34,,,,801,654,4.3700000000,51,3,19,4041,1645,61,389,438,181,136,0.97,Cincinnati Reds,Crosley Field,206773,96,100,CIN,CIN,CIN
+1934,NL,NY1,SFG,,2,153,75,93,60,,,N,N,760,5396,1485,240,41,126,406,526,19,,,,583,486,3.1900000000,68,13,30,4110,1384,75,351,499,179,141,0.97,New York Giants,Polo Grounds IV,730851,98,96,NYG,NY1,NY1
+1934,NL,PHI,PHI,,7,149,71,56,93,,,N,N,675,5218,1480,286,35,56,398,534,52,,,,794,686,4.7600000000,52,8,15,3891,1501,126,437,416,197,140,0.96,Philadelphia Phillies,Baker Bowl,169885,113,116,PHI,PHI,PHI
+1934,NL,PIT,PIT,,5,151,78,74,76,,,N,N,735,5361,1541,281,77,52,440,398,44,,,,713,620,4.2000000000,63,8,8,3987,1523,78,354,487,145,118,0.97,Pittsburgh Pirates,Forbes Field,322622,103,102,PIT,PIT,PIT
+1934,NL,SLN,STL,,1,154,77,95,58,,,Y,Y,799,5502,1582,294,75,104,392,535,69,,,,656,568,3.6900000000,78,15,16,4158,1463,77,411,689,166,141,0.97,St. Louis Cardinals,Sportsman's Park IV,325056,107,104,STL,SLN,SLN
+1935,AL,BOS,BOS,,4,154,79,78,75,,,N,N,718,5288,1458,281,63,69,609,470,91,59,,,732,619,4.0500000000,82,6,11,4128,1520,67,520,470,190,136,0.96,Boston Red Sox,Fenway Park II,558568,107,106,BOS,BOS,BOS
+1935,AL,CHA,CHW,,5,153,77,74,78,,,N,N,738,5314,1460,262,42,74,580,405,46,28,,,750,662,4.3800000000,80,8,8,4080,1443,105,574,436,145,133,0.97,Chicago White Sox,Comiskey Park,470281,102,103,CHW,CHA,CHA
+1935,AL,CLE,CLE,,3,156,77,82,71,,,N,N,776,5534,1573,324,77,93,460,567,63,54,,,739,644,4.1500000000,67,12,21,4188,1527,68,457,498,174,147,0.97,Cleveland Indians,League Park II/Cleveland Stadium,397615,102,102,CLE,CLE,CLE
+1935,AL,DET,DET,,1,152,79,93,58,,,Y,Y,919,5423,1573,301,83,106,627,456,70,45,,,665,579,3.8200000000,87,16,11,4092,1440,78,522,584,126,154,0.97,Detroit Tigers,Navin Field,1034929,96,93,DET,DET,DET
+1935,AL,NYA,NYY,,2,149,74,89,60,,,N,N,818,5214,1462,255,70,104,604,469,68,46,,,632,532,3.6000000000,76,12,13,3993,1276,91,516,594,151,114,0.97,New York Yankees,Yankee Stadium I,657508,95,91,NYY,NYA,NYA
+1935,AL,PHA,OAK,,8,149,72,58,91,,,N,N,710,5269,1470,243,44,112,475,602,43,35,,,869,754,5.1200000000,58,7,10,3978,1486,73,704,469,189,150,0.96,Philadelphia Athletics,Shibe Park,233173,99,102,PHA,PHA,PHA
+1935,AL,SLA,BAL,,7,155,76,65,87,,,N,N,718,5365,1446,291,51,73,593,561,45,25,,,930,807,5.2600000000,42,4,15,4140,1667,92,641,435,186,138,0.97,St. Louis Browns,Sportsman's Park IV,80922,104,107,SLB,SLA,SLA
+1935,AL,WS1,MIN,,6,154,77,67,86,,,N,N,823,5592,1591,255,95,32,596,406,54,37,,,903,804,5.2500000000,67,5,12,4134,1672,89,613,456,171,186,0.97,Washington Senators,Griffith Stadium I,255011,97,98,WSH,WS1,WS1
+1935,NL,BRO,LAD,,5,154,77,70,83,,,N,N,711,5410,1496,235,62,59,430,520,60,,,,767,637,4.2200000000,62,11,20,4074,1519,88,436,480,188,146,0.96,Brooklyn Dodgers,Ebbets Field,470517,98,99,BRO,BRO,BRO
+1935,NL,BSN,ATL,,8,153,75,38,115,,,N,N,575,5309,1396,233,33,75,353,436,20,,,,852,729,4.9300000000,54,6,5,3990,1645,81,404,355,197,101,0.96,Boston Braves,Braves Field,232754,92,94,BSN,BSN,BSN
+1935,NL,CHN,CHC,,1,154,77,100,54,,,Y,N,847,5486,1581,303,62,88,464,471,66,,,,597,505,3.2600000000,81,12,14,4182,1417,85,400,589,186,163,0.97,Chicago Cubs,Wrigley Field,692604,101,98,CHC,CHN,CHN
+1935,NL,CIN,CIN,,6,154,76,68,85,,,N,N,646,5296,1403,244,68,73,392,547,72,,,,772,648,4.3000000000,59,9,13,4068,1490,65,438,500,204,139,0.96,Cincinnati Reds,Crosley Field,448247,96,98,CIN,CIN,CIN
+1935,NL,NY1,SFG,,3,156,79,91,62,,,N,N,770,5623,1608,248,56,123,392,479,32,,,,675,589,3.7800000000,76,10,11,4209,1433,106,411,524,174,129,0.97,New York Giants,Polo Grounds IV,748748,99,96,NYG,NY1,NY1
+1935,NL,PHI,PHI,,7,156,79,64,89,,,N,N,685,5442,1466,249,32,92,392,661,52,,,,871,727,4.7600000000,53,8,15,4122,1652,106,505,475,228,145,0.96,Philadelphia Phillies,Baker Bowl,205470,110,113,PHI,PHI,PHI
+1935,NL,PIT,PIT,,4,153,77,86,67,,,N,N,743,5415,1543,255,90,66,457,437,30,,,,647,519,3.4200000000,76,15,11,4095,1428,63,312,549,190,94,0.96,Pittsburgh Pirates,Forbes Field,352885,103,102,PIT,PIT,PIT
+1935,NL,SLN,STL,,2,154,77,96,58,,,N,N,829,5457,1548,286,59,86,404,521,71,,,,625,541,3.5200000000,73,10,18,4152,1445,68,377,602,163,133,0.97,St. Louis Cardinals,Sportsman's Park IV,506084,104,102,STL,SLN,SLN
+1936,AL,BOS,BOS,,6,155,77,74,80,,,N,N,775,5383,1485,288,62,86,584,465,55,44,,,764,669,4.3900000000,78,11,9,4116,1501,78,552,584,162,139,0.97,Boston Red Sox,Fenway Park II,626895,106,106,BOS,BOS,BOS
+1936,AL,CHA,CHW,,4,153,75,81,70,,,N,N,920,5466,1597,282,56,60,684,417,66,29,,,873,767,5.0600000000,80,5,8,4095,1603,104,578,414,168,174,0.97,Chicago White Sox,Comiskey Park,440810,104,103,CHW,CHA,CHA
+1936,AL,CLE,CLE,,5,157,81,80,74,,,N,N,921,5646,1715,357,82,123,514,470,66,53,,,862,745,4.8300000000,74,6,12,4167,1604,73,607,619,182,154,0.97,Cleveland Indians,League Park II/Cleveland Stadium,500391,103,102,CLE,CLE,CLE
+1936,AL,DET,DET,,2,154,77,83,71,,,N,N,921,5464,1638,326,55,94,640,462,73,49,,,871,756,5.0000000000,76,13,13,4080,1568,100,562,526,154,159,0.97,Detroit Tigers,Navin Field,875948,100,98,DET,DET,DET
+1936,AL,NYA,NYY,,1,155,77,102,51,,,Y,Y,1065,5591,1676,315,83,182,700,594,77,40,,,731,649,4.1700000000,77,6,21,4200,1474,84,663,624,163,148,0.97,New York Yankees,Yankee Stadium I,976913,97,93,NYY,NYA,NYA
+1936,AL,PHA,OAK,,8,154,77,53,100,,,N,N,714,5373,1443,240,60,72,524,590,59,43,,,1045,913,6.0800000000,68,3,12,4056,1645,131,696,405,208,152,0.96,Philadelphia Athletics,Shibe Park,285173,97,101,PHA,PHA,PHA
+1936,AL,SLA,BAL,,7,155,77,57,95,,,N,N,804,5391,1502,299,66,79,625,627,62,20,,,1064,935,6.2400000000,54,3,13,4044,1776,115,609,399,188,143,0.96,St. Louis Browns,Sportsman's Park IV,93267,102,106,SLB,SLA,SLA
+1936,AL,WS1,MIN,,3,153,77,82,71,,,N,N,889,5433,1601,293,84,62,576,398,104,42,,,799,684,4.5800000000,78,8,14,4035,1484,73,588,462,182,163,0.97,Washington Senators,Griffith Stadium I,379525,94,95,WSH,WS1,WS1
+1936,NL,BRO,LAD,,7,156,79,67,87,,,N,N,662,5574,1518,263,43,33,390,458,55,,,,752,620,3.9800000000,59,7,18,4209,1466,84,528,651,202,107,0.96,Brooklyn Dodgers,Ebbets Field,489618,101,103,BRO,BRO,BRO
+1936,NL,BSN,ATL,,6,157,79,71,83,,,N,N,631,5478,1450,207,45,67,433,582,23,,,,715,619,3.9400000000,61,7,13,4239,1566,69,451,421,189,175,0.97,Boston Bees,Braves Field,340585,93,95,BSN,BSN,BSN
+1936,NL,CHN,CHC,,2,154,77,87,67,,,N,N,755,5409,1545,275,36,76,491,462,68,,,,603,544,3.5400000000,77,18,10,4146,1413,77,434,597,146,156,0.97,Chicago Cubs,Wrigley Field,699370,103,99,CHC,CHN,CHN
+1936,NL,CIN,CIN,,5,154,76,74,80,,,N,N,722,5393,1476,224,73,82,410,584,68,,,,760,641,4.2200000000,50,6,23,4101,1576,51,418,459,191,150,0.96,Cincinnati Reds,Crosley Field,466345,93,95,CIN,CIN,CIN
+1936,NL,NY1,SFG,,1,154,78,92,62,,,Y,N,742,5449,1529,237,48,97,431,452,31,,,,621,532,3.4600000000,60,12,22,4155,1458,75,401,500,168,164,0.97,New York Giants,Polo Grounds IV,837952,100,97,NYG,NY1,NY1
+1936,NL,PHI,PHI,,8,154,78,54,100,,,N,N,726,5465,1538,250,46,103,451,586,50,,,,874,704,4.6400000000,51,7,14,4095,1630,87,515,454,237,144,0.95,Philadelphia Phillies,Baker Bowl,249219,111,114,PHI,PHI,PHI
+1936,NL,PIT,PIT,,4,156,76,84,70,,,N,N,804,5586,1596,283,80,60,517,502,37,,,,718,603,3.8900000000,67,5,12,4185,1475,74,379,559,199,113,0.96,Pittsburgh Pirates,Forbes Field,372524,102,101,PIT,PIT,PIT
+1936,NL,SLN,STL,,2,155,77,87,67,,,N,N,795,5537,1554,332,60,88,442,577,69,,,,794,694,4.4700000000,65,5,24,4194,1610,89,434,559,154,134,0.97,St. Louis Cardinals,Sportsman's Park IV,448078,99,98,STL,SLN,SLN
+1937,AL,BOS,BOS,,5,154,74,80,72,,,N,N,821,5354,1506,269,64,100,601,557,79,61,,,775,680,4.4800000000,74,6,14,4098,1518,92,597,682,175,139,0.97,Boston Red Sox,Fenway Park II,559659,104,103,BOS,BOS,BOS
+1937,AL,CHA,CHW,,3,154,77,86,68,,,N,N,780,5277,1478,280,76,67,549,447,70,34,,,730,626,4.1700000000,70,15,21,4053,1435,115,532,533,174,173,0.97,Chicago White Sox,Comiskey Park,589245,100,100,CHW,CHA,CHA
+1937,AL,CLE,CLE,,4,156,78,83,71,,,N,N,817,5353,1499,304,76,103,570,551,78,51,,,768,665,4.3900000000,64,4,15,4092,1529,61,566,630,157,153,0.97,Cleveland Indians,League Park II/Cleveland Stadium,564849,98,97,CLE,CLE,CLE
+1937,AL,DET,DET,,2,155,77,89,65,,,N,N,935,5516,1611,309,62,150,656,711,89,45,,,841,746,4.8700000000,70,6,11,4134,1521,102,635,485,147,149,0.97,Detroit Tigers,Navin Field,1072276,102,101,DET,DET,DET
+1937,AL,NYA,NYY,,1,157,79,102,52,,,Y,Y,979,5487,1554,282,73,174,709,607,60,36,,,671,566,3.6500000000,82,15,21,4188,1417,92,506,652,169,134,0.97,New York Yankees,Yankee Stadium I,998148,102,97,NYY,NYA,NYA
+1937,AL,PHA,OAK,,7,154,79,54,97,,,N,N,699,5228,1398,278,60,94,583,557,95,48,,,854,719,4.8500000000,65,6,9,4005,1490,105,613,469,198,150,0.96,Philadelphia Athletics,Shibe Park,430738,98,101,PHA,PHA,PHA
+1937,AL,SLA,BAL,,8,156,78,46,108,,,N,N,715,5510,1573,327,44,71,514,510,30,27,,,1023,909,6.0000000000,55,2,8,4089,1768,143,653,468,174,166,0.97,St. Louis Browns,Sportsman's Park IV,123121,100,104,SLB,SLA,SLA
+1937,AL,WS1,MIN,,6,158,80,73,80,,,N,N,757,5578,1559,245,84,47,591,503,61,35,,,841,711,4.5800000000,75,5,14,4194,1498,96,671,524,170,181,0.97,Washington Senators,Griffith Stadium I,397799,95,96,WSH,WS1,WS1
+1937,NL,BRO,LAD,,6,155,76,62,91,,,N,N,616,5295,1401,258,53,37,469,583,69,,,,772,625,4.1300000000,63,5,8,4086,1470,68,476,592,216,127,0.96,Brooklyn Dodgers,Ebbets Field,482481,102,104,BRO,BRO,BRO
+1937,NL,BSN,ATL,,5,152,76,79,73,,,N,N,579,5124,1265,200,41,63,485,707,45,,,,556,486,3.2200000000,85,16,10,4077,1344,60,372,387,157,128,0.97,Boston Bees,Braves Field,385339,91,91,BSN,BSN,BSN
+1937,NL,CHN,CHC,,2,154,78,93,61,,,N,N,811,5349,1537,253,74,96,538,496,71,,,,682,609,3.9700000000,73,11,13,4143,1434,91,502,596,151,141,0.97,Chicago Cubs,Wrigley Field,895020,104,102,CHC,CHN,CHN
+1937,NL,CIN,CIN,,8,155,80,56,98,,,N,N,612,5230,1329,215,59,73,437,586,53,,,,707,595,3.9400000000,64,10,18,4074,1428,38,533,581,208,139,0.96,Cincinnati Reds,Crosley Field,411221,95,95,CIN,CIN,CIN
+1937,NL,NY1,SFG,,1,152,75,95,57,,,Y,N,732,5329,1484,251,41,111,412,492,45,,,,602,519,3.4300000000,67,11,17,4083,1341,85,404,653,159,143,0.97,New York Giants,Polo Grounds IV,926887,102,100,NYG,NY1,NY1
+1937,NL,PHI,PHI,,7,155,74,61,92,,,N,N,724,5424,1482,258,37,103,478,640,66,,,,869,770,5.0500000000,59,6,15,4119,1629,116,501,529,184,157,0.97,Philadelphia Phillies,Baker Bowl,212790,109,112,PHI,PHI,PHI
+1937,NL,PIT,PIT,,3,154,78,86,68,,,N,N,704,5433,1550,223,86,47,463,480,32,,,,646,540,3.5600000000,67,12,17,4098,1398,71,428,643,181,135,0.97,Pittsburgh Pirates,Forbes Field,459679,100,98,PIT,PIT,PIT
+1937,NL,SLN,STL,,4,157,80,81,73,,,N,N,789,5476,1543,264,67,94,385,569,78,,,,733,616,3.9800000000,81,10,4,4176,1546,95,448,571,164,127,0.97,St. Louis Cardinals,Sportsman's Park IV,430811,102,101,STL,SLN,SLN
+1938,AL,BOS,BOS,,2,150,75,88,61,,,N,N,902,5229,1566,298,56,98,650,463,55,51,,,751,652,4.4600000000,67,10,15,3948,1472,102,528,484,187,172,0.96,Boston Red Sox,Fenway Park II,646459,105,103,BOS,BOS,BOS
+1938,AL,CHA,CHW,,6,149,73,65,83,,,N,N,709,5199,1439,239,55,67,514,489,56,39,,,752,638,4.3600000000,83,5,9,3948,1449,101,550,432,195,155,0.96,Chicago White Sox,Comiskey Park,338278,103,103,CHW,CHA,CHA
+1938,AL,CLE,CLE,,3,153,76,86,66,,,N,N,847,5356,1506,300,89,113,550,605,83,36,,,782,692,4.6000000000,68,5,17,4059,1416,100,681,717,152,145,0.97,Cleveland Indians,League Park II/Cleveland Stadium,652006,98,97,CLE,CLE,CLE
+1938,AL,DET,DET,,4,155,79,84,70,,,N,N,862,5270,1434,219,52,137,693,581,76,41,,,795,717,4.7900000000,75,3,11,4044,1532,110,608,435,147,172,0.97,Detroit Tigers,Briggs Stadium,799557,107,106,DET,DET,DET
+1938,AL,NYA,NYY,,1,157,79,99,53,,,Y,Y,966,5410,1480,283,63,174,749,616,91,28,,,710,600,3.9100000000,91,11,13,4146,1436,85,566,567,169,169,0.97,New York Yankees,Yankee Stadium I,970916,100,95,NYY,NYA,NYA
+1938,AL,PHA,OAK,,8,154,76,53,99,,,N,N,726,5229,1410,243,62,98,605,590,65,53,,,956,806,5.4800000000,56,4,12,3972,1573,142,599,473,204,119,0.96,Philadelphia Athletics,Shibe Park,385357,96,100,PHA,PHA,PHA
+1938,AL,SLA,BAL,,7,156,77,55,97,,,N,N,755,5333,1498,273,36,92,590,528,51,40,,,962,866,5.8000000000,71,3,7,4032,1584,132,737,632,145,163,0.97,St. Louis Browns,Sportsman's Park IV,130417,99,104,SLB,SLA,SLA
+1938,AL,WS1,MIN,,5,152,78,75,76,,,N,N,814,5474,1602,278,72,85,573,379,65,37,,,873,746,4.9400000000,59,6,11,4080,1472,92,655,515,179,179,0.97,Washington Senators,Griffith Stadium I,522694,93,94,WSH,WS1,WS1
+1938,NL,BRO,LAD,,7,151,74,69,80,,,N,N,704,5142,1322,225,79,61,611,615,66,,,,710,602,4.0700000000,56,12,14,3996,1464,88,446,469,157,148,0.97,Brooklyn Dodgers,Ebbets Field,663087,103,104,BRO,BRO,BRO
+1938,NL,BSN,ATL,,5,153,75,77,75,,,N,N,561,5250,1311,199,39,54,424,548,49,,,,618,521,3.4000000000,83,15,12,4140,1375,66,465,413,173,136,0.97,Boston Bees,Braves Field,341149,89,90,BSN,BSN,BSN
+1938,NL,CHN,CHC,,1,154,77,89,63,,,Y,N,713,5333,1435,242,70,65,522,476,49,,,,598,523,3.3700000000,67,16,18,4188,1414,71,454,583,135,151,0.97,Chicago Cubs,Wrigley Field,951640,103,101,CHC,CHN,CHN
+1938,NL,CIN,CIN,,4,151,77,82,68,,,N,N,723,5391,1495,251,57,110,366,518,19,,,,634,548,3.6200000000,72,11,16,4086,1329,75,463,542,169,133,0.97,Cincinnati Reds,Crosley Field,706756,98,97,CIN,CIN,CIN
+1938,NL,NY1,SFG,,3,152,73,83,67,,,N,N,705,5255,1424,210,36,125,465,528,31,,,,637,543,3.6200000000,59,8,18,4047,1370,87,389,497,168,147,0.97,New York Giants,Polo Grounds IV,799633,101,100,NYG,NY1,NY1
+1938,NL,PHI,PHI,,8,151,75,45,105,,,N,N,550,5192,1318,233,29,40,423,507,38,,,,840,728,4.9300000000,68,3,6,3987,1516,76,582,492,201,135,0.96,Philadelphia Phillies,Baker Bowl/Shibe Park,166111,99,105,PHI,PHI,PHI
+1938,NL,PIT,PIT,,2,152,78,86,64,,,N,N,707,5422,1511,265,66,65,485,409,47,,,,630,530,3.4600000000,57,8,15,4137,1406,71,432,557,163,168,0.97,Pittsburgh Pirates,Forbes Field,641033,100,100,PIT,PIT,PIT
+1938,NL,SLN,STL,,6,156,81,71,80,,,N,N,725,5528,1542,288,74,91,412,492,55,,,,721,591,3.8400000000,58,10,16,4152,1482,77,474,534,199,145,0.96,St. Louis Cardinals,Sportsman's Park IV,291418,105,104,STL,SLN,SLN
+1939,AL,BOS,BOS,,2,152,75,89,62,,,N,N,890,5308,1543,287,57,124,591,505,42,44,,,795,684,4.5600000000,52,4,20,4050,1533,77,543,539,179,147,0.97,Boston Red Sox,Fenway Park II,573070,104,102,BOS,BOS,BOS
+1939,AL,CHA,CHW,,4,155,77,85,69,,,N,N,755,5279,1451,220,56,64,579,502,113,61,,,737,659,4.3100000000,62,5,21,4131,1470,99,454,535,167,140,0.97,Chicago White Sox,Comiskey Park,594104,103,103,CHW,CHA,CHA
+1939,AL,CLE,CLE,,3,154,77,87,67,,,N,N,797,5316,1490,291,79,85,557,574,72,46,,,700,618,4.0800000000,69,10,13,4092,1394,75,602,614,180,148,0.97,Cleveland Indians,League Park II/Cleveland Stadium,563926,97,96,CLE,CLE,CLE
+1939,AL,DET,DET,,5,155,78,81,73,,,N,N,849,5326,1487,277,67,124,620,592,88,38,,,762,652,4.2900000000,64,8,16,4101,1430,104,574,633,198,147,0.96,Detroit Tigers,Briggs Stadium,836279,107,106,DET,DET,DET
+1939,AL,NYA,NYY,,1,152,77,106,45,,,Y,Y,967,5300,1521,259,55,166,701,543,72,37,,,556,496,3.3100000000,87,15,26,4044,1208,85,567,565,126,159,0.97,New York Yankees,Yankee Stadium I,859785,99,95,NYY,NYA,NYA
+1939,AL,PHA,OAK,,7,153,76,55,97,,,N,N,711,5309,1438,282,55,98,503,532,60,34,,,1022,863,5.7900000000,50,6,12,4026,1687,148,579,397,210,131,0.96,Philadelphia Athletics,Shibe Park,395022,97,101,PHA,PHA,PHA
+1939,AL,SLA,BAL,,8,156,78,43,111,,,N,N,733,5422,1453,242,50,91,559,606,48,38,,,1035,916,6.0100000000,56,3,3,4113,1724,133,739,516,199,144,0.96,St. Louis Browns,Sportsman's Park IV,109159,102,105,SLB,SLA,SLA
+1939,AL,WS1,MIN,,6,153,77,65,87,,,N,N,702,5334,1483,249,79,44,547,460,94,47,,,797,692,4.6000000000,72,4,10,4062,1420,75,602,521,205,167,0.96,Washington Senators,Griffith Stadium I,339257,92,94,WSH,WS1,WS1
+1939,NL,BRO,LAD,,3,157,78,84,69,,,N,N,708,5350,1420,265,57,78,564,639,59,,,,645,570,3.6400000000,69,9,13,4230,1431,93,399,528,176,157,0.97,Brooklyn Dodgers,Ebbets Field,955668,105,104,BRO,BRO,BRO
+1939,NL,BSN,ATL,,7,152,73,63,88,,,N,N,572,5286,1395,199,39,56,366,494,41,,,,659,560,3.7100000000,68,11,15,4074,1400,63,513,430,181,178,0.97,Boston Bees,Braves Field,285994,92,94,BSN,BSN,BSN
+1939,NL,CHN,CHC,,4,156,80,84,70,,,N,N,724,5293,1407,263,62,91,523,553,61,,,,678,588,3.8000000000,72,8,13,4176,1504,74,430,584,186,126,0.97,Chicago Cubs,Wrigley Field,726663,102,100,CHC,CHN,CHN
+1939,NL,CIN,CIN,,1,156,81,97,57,,,Y,N,767,5378,1493,269,60,98,500,538,46,,,,595,510,3.2700000000,86,13,9,4209,1340,81,499,637,159,170,0.97,Cincinnati Reds,Crosley Field,981443,102,99,CIN,CIN,CIN
+1939,NL,NY1,SFG,,5,151,74,77,74,,,N,N,703,5129,1395,211,38,116,498,499,26,,,,685,596,4.0700000000,55,6,20,3957,1412,86,477,505,153,152,0.97,New York Giants,Polo Grounds IV,702457,101,101,NYG,NY1,NY1
+1939,NL,PHI,PHI,,8,152,74,45,106,,,N,N,553,5133,1341,232,40,49,421,486,47,,,,856,762,5.1700000000,67,3,12,3978,1502,106,579,447,171,133,0.97,Philadelphia Phillies,Shibe Park,277973,94,100,PHI,PHI,PHI
+1939,NL,PIT,PIT,,6,153,77,68,85,,,N,N,666,5269,1453,261,60,63,477,420,44,,,,721,624,4.1500000000,53,10,15,4062,1537,70,423,464,168,153,0.97,Pittsburgh Pirates,Forbes Field,376734,98,98,PIT,PIT,PIT
+1939,NL,SLN,STL,,2,155,79,92,61,,,N,N,779,5447,1601,332,62,98,475,566,44,,,,633,552,3.5900000000,45,18,32,4152,1377,76,498,603,177,140,0.97,St. Louis Cardinals,Sportsman's Park IV,400245,106,105,STL,SLN,SLN
+1940,AL,BOS,BOS,,4,154,79,82,72,,,N,N,872,5481,1566,301,80,145,590,597,55,49,,,825,749,4.8900000000,51,4,16,4137,1568,124,625,613,173,156,0.97,Boston Red Sox,Fenway Park II,716234,104,102,BOS,BOS,BOS
+1940,AL,CHA,CHW,,4,155,78,82,72,,,N,N,735,5386,1499,238,63,73,496,569,52,60,,,672,576,3.7400000000,83,10,18,4158,1335,111,480,574,185,125,0.96,Chicago White Sox,Comiskey Park,660336,102,101,CHW,CHA,CHA
+1940,AL,CLE,CLE,,2,155,82,89,65,,,N,N,710,5361,1422,287,61,101,519,597,53,36,,,637,555,3.6300000000,72,13,22,4125,1328,86,512,686,148,164,0.97,Cleveland Indians,League Park II/Cleveland Stadium,902576,98,97,CLE,CLE,CLE
+1940,AL,DET,DET,,1,155,79,90,64,,,Y,N,888,5418,1549,312,65,134,664,556,66,39,,,717,613,4.0100000000,59,10,23,4125,1425,102,570,752,194,116,0.96,Detroit Tigers,Briggs Stadium,1112693,109,108,DET,DET,DET
+1940,AL,NYA,NYY,,3,155,76,88,66,,,N,N,817,5286,1371,243,66,155,648,606,59,36,,,671,593,3.8900000000,76,10,14,4119,1389,119,511,559,152,158,0.97,New York Yankees,Yankee Stadium I,988975,96,92,NYY,NYA,NYA
+1940,AL,PHA,OAK,,8,154,71,54,100,,,N,N,703,5304,1391,242,53,105,556,656,48,33,,,932,780,5.2200000000,72,4,12,4035,1543,135,534,488,238,131,0.96,Philadelphia Athletics,Shibe Park,432145,97,101,PHA,PHA,PHA
+1940,AL,SLA,BAL,,6,156,77,67,87,,,N,N,757,5416,1423,278,58,118,556,642,51,40,,,882,781,5.1200000000,64,4,9,4119,1592,113,646,439,158,179,0.97,St. Louis Browns,Sportsman's Park IV,239591,102,105,SLB,SLA,SLA
+1940,AL,WS1,MIN,,7,154,77,64,90,,,N,N,665,5365,1453,266,67,52,468,504,94,40,,,811,688,4.5900000000,74,6,7,4050,1494,93,618,618,192,166,0.96,Washington Senators,Griffith Stadium I,381241,93,95,WSH,WS1,WS1
+1940,NL,BRO,LAD,,2,156,81,88,65,,,N,N,697,5470,1421,256,70,93,522,570,56,,,,621,557,3.5000000000,65,17,14,4299,1366,101,393,639,183,110,0.97,Brooklyn Dodgers,Ebbets Field,975978,108,105,BRO,BRO,BRO
+1940,NL,BSN,ATL,,7,152,75,65,87,,,N,N,623,5329,1366,219,50,59,402,581,48,,,,745,658,4.3600000000,76,9,12,4077,1444,83,573,435,183,169,0.97,Boston Bees,Braves Field,241616,94,96,BSN,BSN,BSN
+1940,NL,CHN,CHC,,5,154,77,75,79,,,N,N,681,5389,1441,272,48,86,482,566,63,,,,636,548,3.5400000000,69,12,14,4176,1418,74,430,564,199,143,0.96,Chicago Cubs,Wrigley Field,534878,98,97,CHC,CHN,CHN
+1940,NL,CIN,CIN,,1,155,77,100,53,,,Y,Y,707,5372,1427,264,38,89,453,503,72,,,,528,477,3.0500000000,91,10,11,4221,1263,73,445,557,117,158,0.98,Cincinnati Reds,Crosley Field,850180,102,99,CIN,CIN,CIN
+1940,NL,NY1,SFG,,6,152,76,72,80,,,N,N,663,5324,1423,201,46,91,453,478,45,,,,659,573,3.7900000000,57,11,18,4080,1383,110,473,606,139,132,0.97,New York Giants,Polo Grounds IV,747852,101,101,NYG,NY1,NY1
+1940,NL,PHI,PHI,,8,153,79,50,103,,,N,N,494,5137,1225,180,35,75,435,527,25,,,,750,663,4.4000000000,66,5,8,4071,1429,92,475,485,181,136,0.97,Philadelphia Phillies,Shibe Park,207177,95,101,PHI,PHI,PHI
+1940,NL,PIT,PIT,,4,156,75,78,76,,,N,N,809,5466,1511,276,68,76,553,494,69,,,,783,672,4.3600000000,49,8,24,4164,1569,72,492,491,217,160,0.96,Pittsburgh Pirates,Forbes Field,507934,99,99,PIT,PIT,PIT
+1940,NL,SLN,STL,,3,156,77,84,69,,,N,N,747,5499,1514,266,61,119,479,610,97,,,,699,594,3.8300000000,71,10,14,4188,1457,83,488,550,174,134,0.97,St. Louis Cardinals,Sportsman's Park IV,324078,106,103,STL,SLN,SLN
+1941,AL,BOS,BOS,,2,155,77,84,70,,,N,N,865,5359,1517,304,55,124,683,567,67,51,,,750,639,4.1900000000,70,8,11,4116,1453,88,611,574,169,139,0.97,Boston Red Sox,Fenway Park II,718497,103,101,BOS,BOS,BOS
+1941,AL,CHA,CHW,,3,156,79,77,77,,,N,N,638,5404,1376,245,47,47,510,476,91,53,,,649,554,3.5200000000,106,14,4,4248,1362,89,521,564,180,145,0.97,Chicago White Sox,Comiskey Park,677077,99,99,CHW,CHA,CHA
+1941,AL,CLE,CLE,,4,155,77,75,79,,,N,N,677,5283,1350,249,84,103,512,605,63,47,,,668,597,3.9000000000,68,10,19,4131,1366,71,660,617,142,158,0.97,Cleveland Indians,League Park II/Cleveland Stadium,745948,95,95,CLE,CLE,CLE
+1941,AL,DET,DET,,4,155,77,75,79,,,N,N,686,5370,1412,247,55,81,602,584,43,28,,,743,641,4.1800000000,52,8,16,4143,1399,80,645,697,186,129,0.96,Detroit Tigers,Briggs Stadium,684915,110,110,DET,DET,DET
+1941,AL,NYA,NYY,,1,156,78,101,53,,,Y,Y,830,5444,1464,243,60,151,616,565,51,33,,,631,548,3.5300000000,75,13,26,4188,1309,81,598,589,165,196,0.97,New York Yankees,Yankee Stadium I,964722,99,95,NYY,NYA,NYA
+1941,AL,PHA,OAK,,8,154,77,64,90,,,N,N,713,5336,1431,240,69,85,574,588,27,36,,,840,733,4.8300000000,64,3,18,4095,1516,136,557,386,200,150,0.96,Philadelphia Athletics,Shibe Park,528894,98,101,PHA,PHA,PHA
+1941,AL,SLA,BAL,,6,157,79,70,84,,,N,N,765,5408,1440,281,58,91,775,552,50,39,,,823,728,4.7200000000,65,7,10,4167,1563,120,549,454,151,156,0.97,St. Louis Browns,Sportsman's Park IV,176240,104,104,SLB,SLA,SLA
+1941,AL,WS1,MIN,,6,156,78,70,84,,,N,N,728,5521,1502,257,80,52,470,488,79,36,,,798,671,4.3500000000,69,8,7,4167,1524,69,603,544,188,169,0.96,Washington Senators,Griffith Stadium I,415663,95,97,WSH,WS1,WS1
+1941,NL,BRO,LAD,,1,157,79,100,54,,,Y,N,800,5485,1494,286,69,101,600,535,36,,,,581,496,3.1400000000,66,17,22,4263,1236,81,495,603,162,125,0.97,Brooklyn Dodgers,Ebbets Field,1214910,106,102,BRO,BRO,BRO
+1941,NL,BSN,ATL,,7,156,76,62,92,,,N,N,592,5414,1357,231,38,48,471,608,61,,,,720,608,3.9500000000,62,10,9,4155,1440,75,554,446,190,174,0.97,Boston Braves,Braves Field,263680,94,97,BSN,BSN,BSN
+1941,NL,CHN,CHC,,6,155,77,70,84,,,N,N,666,5230,1323,239,25,99,559,670,39,,,,670,564,3.7200000000,74,8,9,4092,1431,60,449,548,180,139,0.97,Chicago Cubs,Wrigley Field,545159,96,96,CHC,CHN,CHN
+1941,NL,CIN,CIN,,3,154,79,88,66,,,N,N,616,5218,1288,213,33,64,477,428,68,,,,564,488,3.1700000000,89,19,10,4158,1300,61,510,627,150,147,0.97,Cincinnati Reds,Crosley Field,643513,101,99,CIN,CIN,CIN
+1941,NL,NY1,SFG,,5,156,78,74,79,,,N,N,667,5395,1401,248,35,95,504,518,36,,,,706,609,3.9400000000,55,12,18,4173,1455,90,539,566,160,144,0.97,New York Giants,Polo Grounds IV,763098,103,102,NYG,NY1,NY1
+1941,NL,PHI,PHI,,8,155,76,43,111,,,N,N,501,5233,1277,188,38,64,451,596,65,,,,793,686,4.5000000000,35,4,9,4116,1499,79,606,552,186,147,0.96,Philadelphia Phillies,Shibe Park,231401,95,101,PHI,PHI,PHI
+1941,NL,PIT,PIT,,4,156,78,81,73,,,N,N,690,5297,1417,233,65,56,547,516,59,,,,643,531,3.4800000000,71,8,12,4122,1392,66,492,410,194,130,0.96,Pittsburgh Pirates,Forbes Field,482241,100,100,PIT,PIT,PIT
+1941,NL,SLN,STL,,2,155,79,97,56,,,N,N,734,5457,1482,254,56,70,540,543,47,,,,589,502,3.1900000000,64,15,20,4248,1289,85,502,659,171,146,0.97,St. Louis Cardinals,Sportsman's Park IV,633645,107,104,STL,SLN,SLN
+1942,AL,BOS,BOS,,2,152,77,93,59,,,N,N,761,5248,1451,244,55,103,591,508,68,61,,,594,519,3.4400000000,84,11,17,4074,1260,65,553,500,157,156,0.97,Boston Red Sox,Fenway Park II,730340,104,102,BOS,BOS,BOS
+1942,AL,CHA,CHW,,6,148,70,66,82,,,N,N,538,4949,1215,214,36,25,497,427,114,70,,,609,523,3.5800000000,86,8,8,3942,1304,74,473,432,173,144,0.97,Chicago White Sox,Comiskey Park,425734,97,98,CHW,CHA,CHA
+1942,AL,CLE,CLE,,4,156,80,75,79,,,N,N,590,5317,1344,223,58,50,500,544,69,74,,,659,559,3.5900000000,61,12,11,4206,1353,61,560,448,163,175,0.97,Cleveland Indians,League Park II/Cleveland Stadium,459447,94,94,CLE,CLE,CLE
+1942,AL,DET,DET,,5,156,77,73,81,,,N,N,589,5327,1313,217,37,76,509,476,39,40,,,587,487,3.1300000000,65,12,14,4197,1321,60,598,671,193,142,0.96,Detroit Tigers,Briggs Stadium,580087,108,108,DET,DET,DET
+1942,AL,NYA,NYY,,1,154,77,103,51,,,Y,N,801,5305,1429,223,57,108,591,556,69,33,,,507,445,2.9100000000,88,18,17,4125,1259,71,431,558,142,190,0.97,New York Yankees,Yankee Stadium I,922011,98,94,NYY,NYA,NYA
+1942,AL,PHA,OAK,,8,154,76,55,99,,,N,N,549,5285,1315,213,46,33,440,490,44,45,,,801,679,4.4500000000,67,5,9,4122,1404,89,639,546,188,124,0.96,Philadelphia Athletics,Shibe Park,423487,100,104,PHA,PHA,PHA
+1942,AL,SLA,BAL,,3,151,77,82,69,,,N,N,730,5229,1354,239,62,98,609,607,37,38,,,637,544,3.5900000000,68,12,13,4089,1387,63,505,488,167,143,0.97,St. Louis Browns,Sportsman's Park IV,255617,102,102,SLB,SLA,SLA
+1942,AL,WS1,MIN,,7,151,77,62,89,,,N,N,653,5295,1364,224,49,40,581,536,98,29,,,817,685,4.5800000000,68,12,11,4038,1496,50,558,496,222,133,0.96,Washington Senators,Griffith Stadium I,403493,99,100,WSH,WS1,WS1
+1942,NL,BRO,LAD,,2,155,79,104,50,,,N,N,742,5285,1398,263,34,62,572,484,81,,,,510,441,2.8400000000,67,16,24,4194,1205,73,493,612,138,150,0.97,Brooklyn Dodgers,Ebbets Field,1037765,103,99,BRO,BRO,BRO
+1942,NL,BSN,ATL,,7,150,71,59,89,,,N,N,515,5077,1216,210,19,68,474,507,49,,,,645,557,3.7600000000,68,9,8,4002,1326,82,518,414,139,138,0.97,Boston Braves,Braves Field,285332,97,100,BSN,BSN,BSN
+1942,NL,CHN,CHC,,6,155,78,68,86,,,N,N,591,5352,1360,224,41,75,509,607,63,,,,665,560,3.6000000000,71,10,14,4200,1447,70,525,507,169,169,0.97,Chicago Cubs,Wrigley Field,590972,96,96,CHC,CHN,CHN
+1942,NL,CIN,CIN,,4,154,77,76,76,,,N,N,527,5260,1216,198,39,66,483,549,42,,,,545,442,2.8200000000,80,12,8,4233,1213,47,526,616,175,158,0.97,Cincinnati Reds,Crosley Field,427031,100,99,CIN,CIN,CIN
+1942,NL,NY1,SFG,,3,154,79,85,67,,,N,N,675,5210,1323,162,35,109,558,511,39,,,,600,504,3.3100000000,70,12,13,4110,1299,94,493,497,137,128,0.97,New York Giants,Polo Grounds IV,779621,101,102,NYG,NY1,NY1
+1942,NL,PHI,PHI,,8,151,74,42,109,,,N,N,394,5060,1174,168,37,44,392,488,37,,,,706,614,4.1200000000,51,2,6,4023,1328,61,605,472,193,147,0.96,Philadelphia Phillies,Shibe Park,230183,94,99,PHI,PHI,PHI
+1942,NL,PIT,PIT,,5,151,77,66,81,,,N,N,585,5104,1250,173,49,54,537,536,41,,,,631,537,3.5800000000,64,13,11,4053,1376,62,435,426,182,129,0.96,Pittsburgh Pirates,Forbes Field,448897,103,102,PIT,PIT,PIT
+1942,NL,SLN,STL,,1,156,78,106,48,,,Y,Y,755,5421,1454,282,69,60,551,507,71,,,,482,399,2.5500000000,70,18,15,4230,1192,49,473,651,169,137,0.97,St. Louis Cardinals,Sportsman's Park IV,553552,108,104,STL,SLN,SLN
+1943,AL,BOS,BOS,,7,155,77,68,84,,,N,N,563,5392,1314,223,42,57,486,591,86,61,,,607,547,3.4500000000,62,13,16,4278,1369,61,615,513,153,179,0.97,Boston Red Sox,Fenway Park II,358275,102,101,BOS,BOS,BOS
+1943,AL,CHA,CHW,,4,155,76,82,72,,,N,N,573,5254,1297,193,46,33,561,581,173,87,,,594,498,3.2000000000,70,12,19,4200,1352,54,501,476,166,167,0.97,Chicago White Sox,Comiskey Park,508962,99,101,CHW,CHA,CHA
+1943,AL,CLE,CLE,,3,153,77,82,71,,,N,N,600,5269,1344,246,45,55,567,521,47,58,,,577,492,3.1500000000,64,14,20,4218,1234,52,606,585,157,183,0.97,Cleveland Indians,League Park II/Cleveland Stadium,438894,93,94,CLE,CLE,CLE
+1943,AL,DET,DET,,5,155,78,78,76,,,N,N,632,5364,1401,200,47,77,483,553,40,43,,,560,470,3.0000000000,67,18,20,4233,1226,51,549,706,177,130,0.97,Detroit Tigers,Briggs Stadium,606287,107,106,DET,DET,DET
+1943,AL,NYA,NYY,,1,155,77,98,56,,,Y,Y,669,5282,1350,218,59,100,624,562,46,60,,,542,461,2.9300000000,83,14,13,4245,1229,60,489,653,160,166,0.97,New York Yankees,Yankee Stadium I,618330,101,98,NYY,NYA,NYA
+1943,AL,PHA,OAK,,8,155,79,49,105,,,N,N,497,5244,1219,174,44,26,430,465,55,42,,,717,627,4.0500000000,73,5,13,4182,1421,73,536,503,162,148,0.97,Philadelphia Athletics,Shibe Park,376735,99,103,PHA,PHA,PHA
+1943,AL,SLA,BAL,,6,153,77,72,80,,,N,N,596,5175,1269,229,36,78,569,646,37,43,,,604,525,3.4100000000,64,10,14,4155,1397,74,488,572,152,127,0.97,St. Louis Browns,Sportsman's Park IV,214392,104,102,SLB,SLA,SLA
+1943,AL,WS1,MIN,,2,153,76,84,69,,,N,N,666,5233,1328,245,50,47,605,579,142,55,,,595,490,3.1800000000,61,16,21,4164,1293,48,540,495,179,145,0.97,Washington Senators,Griffith Stadium I,574694,96,97,WSH,WS1,WS1
+1943,NL,BRO,LAD,,3,153,77,81,72,,,N,N,716,5309,1444,263,35,39,580,422,58,,,,674,590,3.8800000000,50,13,22,4107,1326,59,637,588,167,137,0.97,Brooklyn Dodgers,Ebbets Field,661739,101,100,BRO,BRO,BRO
+1943,NL,BSN,ATL,,6,153,77,68,85,,,N,N,465,5196,1213,202,36,39,469,609,56,,,,612,504,3.2500000000,87,13,4,4191,1361,66,441,409,178,139,0.97,Boston Braves,Braves Field,271289,98,100,BSN,BSN,BSN
+1943,NL,CHN,CHC,,5,154,75,74,79,,,N,N,632,5279,1380,207,56,52,574,522,53,,,,600,510,3.3100000000,67,13,14,4158,1379,53,394,513,167,138,0.97,Chicago Cubs,Wrigley Field,508247,98,98,CHC,CHN,CHN
+1943,NL,CIN,CIN,,2,155,78,87,67,,,N,N,608,5329,1362,229,47,43,445,476,49,,,,543,488,3.1300000000,78,18,17,4212,1299,38,579,498,121,193,0.98,Cincinnati Reds,Crosley Field,379122,99,99,CIN,CIN,CIN
+1943,NL,NY1,SFG,,8,156,77,55,98,,,N,N,558,5290,1309,153,33,81,480,470,35,,,,713,632,4.0800000000,35,6,19,4182,1474,80,626,588,167,140,0.97,New York Giants,Polo Grounds IV,466095,100,101,NYG,NY1,NY1
+1943,NL,PHI,PHI,,7,157,78,64,90,,,N,N,571,5297,1321,186,36,66,499,556,29,,,,676,586,3.7900000000,66,10,14,4176,1436,59,451,431,187,143,0.97,Philadelphia Blue Jays,Shibe Park,466975,95,99,PHI,PHI,PHI
+1943,NL,PIT,PIT,,4,157,78,80,74,,,N,N,669,5353,1401,240,73,42,573,566,64,,,,605,480,3.0800000000,74,11,12,4212,1424,44,422,396,170,159,0.97,Pittsburgh Pirates,Forbes Field,498740,104,103,PIT,PIT,PIT
+1943,NL,SLN,STL,,1,157,81,105,49,,,Y,N,679,5438,1515,259,72,70,428,438,40,,,,475,407,2.5700000000,94,21,15,4281,1246,33,477,639,151,183,0.97,St. Louis Cardinals,Sportsman's Park IV,517135,106,101,STL,SLN,SLN
+1944,AL,BOS,BOS,,4,156,78,77,77,,,N,N,739,5400,1456,277,56,69,522,505,60,40,,,676,592,3.8200000000,58,7,17,4182,1404,66,592,524,172,154,0.97,Boston Red Sox,Fenway Park II,506975,99,100,BOS,BOS,BOS
+1944,AL,CHA,CHW,,7,154,77,71,83,,,N,N,543,5292,1307,210,55,23,439,448,66,47,,,662,553,3.5800000000,64,5,17,4170,1411,68,420,481,183,154,0.97,Chicago White Sox,Comiskey Park,563539,98,99,CHW,CHA,CHA
+1944,AL,CLE,CLE,,5,155,78,72,82,,,N,N,643,5481,1458,270,50,70,512,593,48,42,,,677,575,3.6500000000,48,7,18,4257,1428,40,621,524,164,192,0.97,Cleveland Indians,League Park II/Cleveland Stadium,475272,96,96,CLE,CLE,CLE
+1944,AL,DET,DET,,2,156,78,88,66,,,N,N,658,5344,1405,220,44,60,532,500,61,55,,,581,481,3.0900000000,87,20,8,4200,1373,39,452,568,190,184,0.97,Detroit Tigers,Briggs Stadium,923176,105,103,DET,DET,DET
+1944,AL,NYA,NYY,,3,154,78,83,71,,,N,N,674,5331,1410,216,74,96,523,627,91,31,,,617,524,3.3900000000,78,9,13,4170,1351,82,532,529,156,170,0.97,New York Yankees,Yankee Stadium I,789995,104,102,NYY,NYA,NYA
+1944,AL,PHA,OAK,,5,155,76,72,82,,,N,N,525,5312,1364,169,47,36,422,490,42,32,,,594,506,3.2600000000,72,10,14,4191,1345,58,390,534,174,127,0.97,Philadelphia Athletics,Shibe Park,505322,97,101,PHA,PHA,PHA
+1944,AL,SLA,BAL,,1,154,77,89,65,,,Y,N,684,5269,1328,223,45,72,531,604,44,33,,,587,492,3.1700000000,71,16,17,4191,1392,58,469,581,171,142,0.97,St. Louis Browns,Sportsman's Park IV,508644,107,107,SLB,SLA,SLA
+1944,AL,WS1,MIN,,8,154,77,64,90,,,N,N,592,5319,1386,186,42,33,470,477,127,59,,,664,536,3.4900000000,83,13,11,4143,1410,48,475,503,218,156,0.96,Washington Senators,Griffith Stadium I,525235,95,95,WSH,WS1,WS1
+1944,NL,BRO,LAD,,7,155,77,63,91,,,N,N,690,5393,1450,255,51,56,486,451,45,,,,832,711,4.6800000000,50,4,13,4101,1471,75,660,487,197,112,0.96,Brooklyn Dodgers,Ebbets Field,605905,98,98,BRO,BRO,BRO
+1944,NL,BSN,ATL,,6,155,78,65,89,,,N,N,593,5282,1299,250,39,79,456,509,37,,,,674,566,3.6700000000,70,13,12,4164,1430,80,527,454,182,160,0.97,Boston Braves,Braves Field,208691,104,106,BSN,BSN,BSN
+1944,NL,CHN,CHC,,4,157,78,75,79,,,N,N,702,5462,1425,236,46,71,520,521,53,,,,669,558,3.5900000000,70,11,13,4200,1484,75,458,545,185,151,0.97,Chicago Cubs,Wrigley Field,640110,99,97,CHC,CHN,CHN
+1944,NL,CIN,CIN,,3,155,78,89,65,,,N,N,573,5271,1340,229,31,51,423,391,51,,,,537,461,2.9700000000,93,17,12,4194,1292,60,390,369,138,153,0.97,Cincinnati Reds,Crosley Field,409567,96,97,CIN,CIN,CIN
+1944,NL,NY1,SFG,,5,155,75,67,87,,,N,N,682,5306,1398,191,47,93,512,480,39,,,,773,650,4.2900000000,47,4,21,4089,1413,116,587,499,179,128,0.97,New York Giants,Polo Grounds IV,674483,99,101,NYG,NY1,NY1
+1944,NL,PHI,PHI,,8,154,79,61,92,,,N,N,539,5301,1331,199,42,55,470,500,32,,,,658,564,3.6400000000,66,11,6,4185,1407,49,459,496,177,138,0.97,Philadelphia Blue Jays,Shibe Park,369586,96,99,PHI,PHI,PHI
+1944,NL,PIT,PIT,,2,158,81,90,63,,,N,N,744,5428,1441,248,80,70,573,616,87,,,,662,540,3.4400000000,77,10,19,4242,1466,65,435,452,191,122,0.97,Pittsburgh Pirates,Forbes Field,604278,105,103,PIT,PIT,PIT
+1944,NL,SLN,STL,,1,157,77,105,49,,,Y,Y,772,5475,1507,274,59,100,544,473,37,,,,490,423,2.6700000000,89,26,12,4281,1228,55,468,637,112,162,0.98,St. Louis Cardinals,Sportsman's Park IV,461968,103,99,STL,SLN,SLN
+1945,AL,BOS,BOS,,7,157,78,71,83,,,N,N,599,5367,1393,225,44,50,541,534,72,50,,,674,587,3.8000000000,71,15,13,4170,1389,58,656,490,168,198,0.97,Boston Red Sox,Fenway Park II,603794,103,101,BOS,BOS,BOS
+1945,AL,CHA,CHW,,6,150,74,71,78,,,N,N,596,5077,1330,204,55,22,470,467,78,54,,,633,545,3.6900000000,84,13,13,3990,1400,63,448,486,180,139,0.97,Chicago White Sox,Comiskey Park,657981,97,98,CHW,CHA,CHA
+1945,AL,CLE,CLE,,5,147,77,73,72,,,N,N,557,4898,1249,216,48,65,505,578,19,31,,,548,479,3.3100000000,76,14,12,3906,1269,39,501,497,126,149,0.97,Cleveland Indians,League Park II/Cleveland Stadium,558182,95,96,CLE,CLE,CLE
+1945,AL,DET,DET,,1,155,76,88,65,,,Y,Y,633,5257,1345,227,47,77,517,533,60,54,,,565,463,2.9900000000,78,19,16,4179,1305,48,538,588,158,173,0.97,Detroit Tigers,Briggs Stadium,1280341,107,105,DET,DET,DET
+1945,AL,NYA,NYY,,4,152,76,81,71,,,N,N,676,5176,1343,189,61,93,618,567,64,43,,,606,519,3.4500000000,78,9,14,4065,1277,66,485,474,175,170,0.97,New York Yankees,Yankee Stadium I,881845,105,103,NYY,NYA,NYA
+1945,AL,PHA,OAK,,8,153,77,52,98,,,N,N,494,5296,1297,201,37,33,449,463,25,45,,,638,555,3.6200000000,65,11,8,4143,1380,55,571,531,152,160,0.97,Philadelphia Athletics,Shibe Park,462631,98,101,PHA,PHA,PHA
+1945,AL,SLA,BAL,,3,154,76,81,70,,,N,N,597,5227,1302,215,37,63,500,555,25,31,,,548,482,3.1400000000,91,10,8,4146,1307,59,506,570,143,123,0.97,St. Louis Browns,Sportsman's Park IV,482986,106,106,SLB,SLA,SLA
+1945,AL,WS1,MIN,,2,156,78,87,67,,,N,N,622,5326,1375,197,63,27,545,489,110,65,,,562,458,2.9200000000,82,19,11,4236,1307,42,440,550,183,124,0.97,Washington Senators,Griffith Stadium I,652660,91,92,WSH,WS1,WS1
+1945,NL,BRO,LAD,,3,155,78,87,67,,,N,N,795,5418,1468,257,71,57,629,434,75,,,,724,572,3.7000000000,61,7,18,4176,1357,74,586,557,230,144,0.96,Brooklyn Dodgers,Ebbets Field,1059220,99,99,BRO,BRO,BRO
+1945,NL,BSN,ATL,,6,154,75,67,85,,,N,N,721,5441,1453,229,25,101,520,510,82,,,,728,624,4.0400000000,57,7,13,4173,1474,99,557,404,193,160,0.96,Boston Braves,Braves Field,374178,100,101,BSN,BSN,BSN
+1945,NL,CHN,CHC,,1,155,76,98,56,,,Y,N,735,5298,1465,229,52,57,554,462,69,,,,532,452,2.9800000000,86,15,14,4098,1301,57,385,541,119,124,0.98,Chicago Cubs,Wrigley Field,1036386,98,96,CHC,CHN,CHN
+1945,NL,CIN,CIN,,7,154,77,61,93,,,N,N,536,5283,1317,221,26,56,392,532,71,,,,694,607,4.0000000000,77,11,6,4095,1438,70,534,372,146,138,0.97,Cincinnati Reds,Crosley Field,290070,98,99,CIN,CIN,CIN
+1945,NL,NY1,SFG,,5,154,78,78,74,,,N,N,668,5350,1439,175,35,114,501,457,38,,,,700,620,4.0600000000,53,13,21,4122,1401,85,528,530,166,112,0.97,New York Giants,Polo Grounds IV,1016468,101,102,NYG,NY1,NY1
+1945,NL,PHI,PHI,,8,154,77,46,108,,,N,N,548,5203,1278,197,27,56,449,501,54,,,,865,697,4.6400000000,31,4,26,4056,1544,61,608,432,234,150,0.96,Philadelphia Phillies,Shibe Park,285057,96,100,PHI,PHI,PHI
+1945,NL,PIT,PIT,,4,155,79,82,72,,,N,N,753,5343,1425,259,56,72,590,480,81,,,,686,579,3.7600000000,73,8,16,4161,1477,61,455,518,178,141,0.97,Pittsburgh Pirates,Forbes Field,604694,104,104,PIT,PIT,PIT
+1945,NL,SLN,STL,,2,155,78,95,59,,,N,N,756,5487,1498,256,44,64,515,488,55,,,,583,507,3.2400000000,77,18,9,4224,1351,70,497,510,136,150,0.97,St. Louis Cardinals,Sportsman's Park IV,594630,103,99,STL,SLN,SLN
+1946,AL,BOS,BOS,,1,156,78,104,50,,,Y,N,792,5318,1441,268,50,109,687,661,45,36,,,594,524,3.3800000000,79,15,20,4188,1359,89,501,667,139,165,0.97,Boston Red Sox,Fenway Park II,1416944,106,105,BOS,BOS,BOS
+1946,AL,CHA,CHW,,5,155,79,74,80,,,N,N,562,5312,1364,206,44,37,501,600,78,64,,,595,479,3.1000000000,62,9,16,4176,1348,80,508,550,173,170,0.97,Chicago White Sox,Comiskey Park,983403,96,97,CHW,CHA,CHA
+1946,AL,CLE,CLE,,6,156,77,68,86,,,N,N,537,5242,1285,233,56,79,506,697,57,49,,,638,558,3.6200000000,63,16,13,4164,1282,84,649,789,147,147,0.97,Cleveland Indians,League Park II/Cleveland Stadium,1057289,93,94,CLE,CLE,CLE
+1946,AL,DET,DET,,2,155,79,92,62,,,N,N,704,5318,1373,212,41,108,622,616,65,41,,,567,502,3.2200000000,94,18,15,4206,1277,97,497,896,155,138,0.97,Detroit Tigers,Briggs Stadium,1722590,107,105,DET,DET,DET
+1946,AL,NYA,NYY,,3,154,77,87,67,,,N,N,684,5139,1275,208,50,136,627,706,48,35,,,547,473,3.1300000000,68,17,17,4083,1232,66,552,653,147,174,0.97,New York Yankees,Yankee Stadium I,2265512,101,98,NYY,NYA,NYA
+1946,AL,PHA,OAK,,8,155,78,49,105,,,N,N,529,5200,1317,220,51,40,482,594,39,30,,,680,582,3.9000000000,61,10,5,4026,1371,83,577,562,167,141,0.97,Philadelphia Athletics,Shibe Park,621793,98,100,PHA,PHA,PHA
+1946,AL,SLA,BAL,,7,156,77,66,88,,,N,N,621,5373,1350,220,46,84,465,713,23,35,,,710,607,3.9500000000,63,13,12,4146,1465,73,573,574,159,157,0.97,St. Louis Browns,Sportsman's Park IV,526435,106,107,SLB,SLA,SLA
+1946,AL,WS1,MIN,,4,155,76,76,78,,,N,N,608,5337,1388,260,63,60,511,641,51,50,,,706,580,3.7400000000,71,8,10,4188,1459,81,547,537,211,162,0.96,Washington Senators,Griffith Stadium I,1027216,94,95,WSH,WS1,WS1
+1946,NL,BRO,LAD,,2,157,79,96,60,,,N,N,701,5285,1376,233,66,55,691,575,100,,,,570,481,3.0500000000,52,14,28,4254,1280,58,671,647,174,154,0.97,Brooklyn Dodgers,Ebbets Field,1796824,101,99,BRO,BRO,BRO
+1946,NL,BSN,ATL,,4,154,77,81,72,,,N,N,630,5225,1377,238,48,44,558,468,60,,,,592,510,3.3500000000,73,10,12,4113,1291,76,478,566,169,129,0.97,Boston Braves,Braves Field,969673,101,101,BSN,BSN,BSN
+1946,NL,CHN,CHC,,3,155,77,82,71,,,N,N,626,5298,1344,223,50,56,586,599,43,,,,581,501,3.2400000000,59,15,11,4179,1370,58,527,619,146,119,0.97,Chicago Cubs,Wrigley Field,1342970,98,97,CHC,CHN,CHN
+1946,NL,CIN,CIN,,6,156,77,67,87,,,N,N,523,5291,1262,206,33,65,493,604,82,,,,570,484,3.0800000000,69,17,11,4239,1334,70,467,506,152,192,0.97,Cincinnati Reds,Crosley Field,715751,97,99,CIN,CIN,CIN
+1946,NL,NY1,SFG,,8,154,77,61,93,,,N,N,612,5191,1326,176,37,121,532,546,46,,,,685,589,3.9200000000,47,8,13,4059,1313,114,660,581,159,121,0.97,New York Giants,Polo Grounds IV,1219873,100,101,NYG,NY1,NY1
+1946,NL,PHI,PHI,,5,155,78,69,85,,,N,N,560,5233,1351,209,40,80,417,590,41,,,,705,607,3.9900000000,55,11,23,4107,1442,73,542,490,147,144,0.97,Philadelphia Phillies,Shibe Park,1045247,97,100,PHI,PHI,PHI
+1946,NL,PIT,PIT,,7,155,78,63,91,,,N,N,552,5199,1300,202,52,60,592,555,48,,,,668,566,3.7200000000,61,10,6,4110,1406,50,561,458,184,127,0.97,Pittsburgh Pirates,Forbes Field,749962,103,104,PIT,PIT,PIT
+1946,NL,SLN,STL,,1,156,78,98,58,,,Y,Y,712,5372,1426,265,56,81,530,537,58,,,,545,467,3.0100000000,75,18,15,4191,1326,63,493,607,123,167,0.98,St. Louis Cardinals,Sportsman's Park IV,1061807,104,101,STL,SLN,SLN
+1947,AL,BOS,BOS,,3,157,81,83,71,,,N,N,720,5322,1412,206,54,103,666,590,41,35,,,669,589,3.8100000000,64,13,19,4173,1383,84,575,586,137,172,0.97,Boston Red Sox,Fenway Park II,1427315,107,105,BOS,BOS,BOS
+1947,AL,CHA,CHW,,6,155,75,70,84,,,N,N,553,5274,1350,211,41,53,492,527,91,57,,,661,563,3.6400000000,47,11,27,4173,1384,76,603,522,155,180,0.97,Chicago White Sox,Comiskey Park,876948,96,98,CHW,CHA,CHA
+1947,AL,CLE,CLE,,4,157,78,80,74,,,N,N,687,5367,1392,234,51,112,502,609,29,25,,,588,536,3.4400000000,55,13,29,4206,1244,94,628,590,104,178,0.98,Cleveland Indians,Cleveland Stadium,1521978,98,94,CLE,CLE,CLE
+1947,AL,DET,DET,,2,158,80,85,69,,,N,N,714,5276,1363,234,42,103,762,565,52,60,,,642,555,3.5700000000,77,15,18,4194,1382,79,531,648,155,142,0.97,Detroit Tigers,Briggs Stadium,1398093,104,103,DET,DET,DET
+1947,AL,NYA,NYY,,1,155,77,97,57,,,Y,Y,794,5308,1439,230,72,115,610,581,27,23,,,568,518,3.3900000000,73,14,21,4122,1221,95,628,691,109,151,0.98,New York Yankees,Yankee Stadium I,2178937,99,95,NYY,NYA,NYA
+1947,AL,PHA,OAK,,5,156,78,78,76,,,N,N,633,5198,1311,218,52,61,605,563,37,33,,,614,542,3.5100000000,70,12,15,4173,1291,85,597,493,143,161,0.97,Philadelphia Athletics,Shibe Park,911566,102,103,PHA,PHA,PHA
+1947,AL,SLA,BAL,,8,154,77,59,95,,,N,N,564,5145,1238,189,52,90,583,664,69,49,,,744,657,4.3300000000,50,7,13,4095,1426,103,604,552,134,169,0.97,St. Louis Browns,Sportsman's Park IV,320474,102,105,SLB,SLA,SLA
+1947,AL,WS1,MIN,,7,154,77,64,90,,,N,N,496,5112,1234,186,48,42,525,534,53,51,,,675,601,3.9700000000,67,15,12,4086,1408,63,579,551,142,151,0.97,Washington Senators,Griffith Stadium I,850758,97,100,WSH,WS1,WS1
+1947,NL,BRO,LAD,,1,155,78,94,60,,,Y,N,774,5249,1428,241,50,83,732,561,88,,,,668,584,3.8200000000,47,14,34,4125,1299,104,626,592,129,164,0.97,Brooklyn Dodgers,Ebbets Field,1807526,104,101,BRO,BRO,BRO
+1947,NL,BSN,ATL,,3,154,77,86,68,,,N,N,701,5253,1444,265,42,85,558,500,58,,,,622,548,3.6200000000,74,14,13,4086,1342,93,453,494,153,124,0.97,Boston Braves,Braves Field,1277361,98,97,BSN,BSN,BSN
+1947,NL,CHN,CHC,,6,155,79,69,85,,,N,N,567,5305,1373,231,48,71,471,578,22,,,,722,614,4.0400000000,46,8,15,4101,1449,106,618,571,150,159,0.97,Chicago Cubs,Wrigley Field,1364039,96,97,CHC,CHN,CHN
+1947,NL,CIN,CIN,,5,154,77,73,81,,,N,N,681,5299,1372,242,43,95,539,530,46,,,,755,669,4.4100000000,54,13,13,4095,1442,102,589,633,136,134,0.97,Cincinnati Reds,Crosley Field,899975,100,102,CIN,CIN,CIN
+1947,NL,NY1,SFG,,4,155,76,81,73,,,N,N,830,5343,1446,220,48,221,494,568,29,,,,761,672,4.4400000000,58,6,14,4089,1428,122,590,553,155,136,0.97,New York Giants,Polo Grounds IV,1600793,101,100,NYG,NY1,NY1
+1947,NL,PHI,PHI,,7,155,77,62,92,,,N,N,589,5256,1354,210,52,60,464,594,60,,,,687,599,3.9600000000,70,8,14,4086,1399,98,513,514,152,140,0.97,Philadelphia Phillies,Shibe Park,907332,96,98,PHI,PHI,PHI
+1947,NL,PIT,PIT,,7,156,79,62,92,,,N,N,744,5307,1385,216,44,156,607,687,30,,,,817,714,4.6800000000,44,9,13,4122,1488,155,592,530,149,131,0.97,Pittsburgh Pirates,Forbes Field,1283531,103,104,PIT,PIT,PIT
+1947,NL,SLN,STL,,2,156,77,89,65,,,N,N,780,5422,1462,235,65,115,612,511,28,,,,634,548,3.5300000000,65,12,20,4191,1417,106,495,642,127,169,0.97,St. Louis Cardinals,Sportsman's Park IV,1247913,104,101,STL,SLN,SLN
+1948,AL,BOS,BOS,,2,155,78,96,59,,,N,N,907,5363,1471,277,40,121,823,552,38,17,,,720,653,4.2600000000,70,11,13,4137,1445,83,592,513,116,174,0.98,Boston Red Sox,Fenway Park II,1558798,105,102,BOS,BOS,BOS
+1948,AL,CHA,CHW,,8,154,76,51,101,,,N,N,559,5192,1303,172,39,55,595,528,46,47,,,814,731,4.8900000000,35,2,23,4035,1454,89,673,403,154,176,0.97,Chicago White Sox,Comiskey Park,777844,96,99,CHW,CHA,CHA
+1948,AL,CLE,CLE,,1,156,79,97,58,,,Y,Y,840,5446,1534,242,54,155,646,575,54,44,,,568,504,3.2200000000,66,26,30,4227,1246,82,628,595,114,183,0.98,Cleveland Indians,Cleveland Stadium,2620627,97,95,CLE,CLE,CLE
+1948,AL,DET,DET,,5,154,77,78,76,,,N,N,700,5235,1396,219,58,78,671,504,22,32,,,726,635,4.1500000000,60,5,22,4131,1367,92,589,678,155,143,0.97,Detroit Tigers,Briggs Stadium,1743035,103,102,DET,DET,DET
+1948,AL,NYA,NYY,,3,154,77,94,60,,,N,N,857,5324,1480,251,75,139,623,478,24,24,,,633,569,3.7500000000,62,16,24,4095,1289,94,641,654,120,161,0.97,New York Yankees,Yankee Stadium I,2373901,99,95,NYY,NYA,NYA
+1948,AL,PHA,OAK,,4,154,77,84,70,,,N,N,729,5181,1345,231,47,68,726,523,40,32,,,735,673,4.4300000000,74,7,18,4104,1456,86,638,486,113,180,0.98,Philadelphia Athletics,Shibe Park,945076,101,101,PHA,PHA,PHA
+1948,AL,SLA,BAL,,6,155,76,59,94,,,N,N,671,5303,1438,251,62,63,578,572,63,44,,,849,764,5.0100000000,35,4,20,4119,1513,103,737,531,168,190,0.97,St. Louis Browns,Sportsman's Park IV,335564,103,106,SLB,SLA,SLA
+1948,AL,WS1,MIN,,7,154,78,56,97,,,N,N,578,5111,1245,203,75,31,568,572,76,48,,,796,701,4.6500000000,42,4,22,4071,1439,81,734,446,154,144,0.97,Washington Senators,Griffith Stadium I,795254,97,101,WSH,WS1,WS1
+1948,NL,BRO,LAD,,3,155,78,84,70,,,N,N,744,5328,1393,256,54,91,601,684,114,,,,667,580,3.7500000000,52,9,22,4176,1328,119,633,670,160,151,0.97,Brooklyn Dodgers,Ebbets Field,1398967,104,101,BRO,BRO,BRO
+1948,NL,BSN,ATL,,1,154,76,91,62,,,Y,N,739,5297,1458,272,49,95,671,536,43,,,,584,520,3.3700000000,70,10,17,4167,1354,93,430,579,143,132,0.97,Boston Braves,Braves Field,1455439,99,98,BSN,BSN,BSN
+1948,NL,CHN,CHC,,8,155,78,64,90,,,N,N,597,5352,1402,225,44,87,443,578,39,,,,706,602,4.0000000000,51,7,10,4065,1355,89,619,636,172,152,0.97,Chicago Cubs,Wrigley Field,1237792,95,98,CHC,CHN,CHN
+1948,NL,CIN,CIN,,7,153,77,64,89,,,N,N,588,5127,1266,221,37,104,478,586,42,,,,752,667,4.4700000000,40,8,20,4029,1410,104,572,599,153,135,0.97,Cincinnati Reds,Crosley Field,823386,97,99,CIN,CIN,CIN
+1948,NL,NY1,SFG,,5,155,77,78,76,,,N,N,780,5277,1352,210,49,164,599,648,51,,,,704,600,3.9300000000,54,15,21,4119,1425,122,556,527,156,134,0.97,New York Giants,Polo Grounds IV,1459269,101,100,NYG,NY1,NY1
+1948,NL,PHI,PHI,,6,155,76,66,88,,,N,N,591,5287,1367,227,39,91,440,598,68,,,,729,617,4.0800000000,61,6,15,4086,1385,95,556,550,210,126,0.96,Philadelphia Phillies,Shibe Park,767429,98,99,PHI,PHI,PHI
+1948,NL,PIT,PIT,,4,156,80,83,71,,,N,N,706,5286,1388,191,54,108,580,578,68,,,,699,632,4.1500000000,65,5,19,4113,1373,120,564,543,135,150,0.97,Pittsburgh Pirates,Forbes Field,1517021,102,103,PIT,PIT,PIT
+1948,NL,SLN,STL,,2,155,77,85,69,,,N,N,742,5302,1396,238,58,105,594,521,24,,,,646,594,3.9100000000,60,13,18,4104,1392,103,476,625,118,138,0.98,St. Louis Cardinals,Sportsman's Park IV,1111440,106,103,STL,SLN,SLN
+1949,AL,BOS,BOS,,2,155,77,96,58,,,N,N,896,5320,1500,272,36,131,835,510,43,25,,,667,607,3.9700000000,84,16,16,4131,1375,82,661,598,120,207,0.98,Boston Red Sox,Fenway Park II,1596650,108,104,BOS,BOS,BOS
+1949,AL,CHA,CHW,,6,154,77,63,91,,,N,N,648,5204,1340,207,66,43,702,596,62,55,,,737,651,4.3000000000,57,10,17,4089,1362,108,693,502,135,180,0.97,Chicago White Sox,Comiskey Park,937151,96,99,CHW,CHA,CHA
+1949,AL,CLE,CLE,,3,154,77,89,65,,,N,N,675,5221,1358,194,58,112,601,534,44,40,,,574,516,3.3600000000,65,10,19,4149,1275,82,611,594,103,192,0.98,Cleveland Indians,Cleveland Stadium,2233771,98,95,CLE,CLE,CLE
+1949,AL,DET,DET,,4,155,78,87,67,,,N,N,751,5259,1405,215,51,88,751,502,39,52,,,655,584,3.7700000000,70,19,12,4179,1338,102,628,631,132,174,0.97,Detroit Tigers,Briggs Stadium,1821204,101,99,DET,DET,DET
+1949,AL,NYA,NYY,,1,155,78,97,57,,,Y,Y,829,5196,1396,215,60,115,731,539,58,30,,,637,562,3.6900000000,59,12,36,4113,1231,98,812,671,137,195,0.97,New York Yankees,Yankee Stadium I,2283676,100,96,NYY,NYA,NYA
+1949,AL,PHA,OAK,,5,154,77,81,73,,,N,N,726,5123,1331,214,49,82,783,493,36,25,,,725,642,4.2300000000,85,9,11,4095,1359,105,758,490,140,217,0.97,Philadelphia Athletics,Shibe Park,816514,97,98,PHA,PHA,PHA
+1949,AL,SLA,BAL,,7,155,77,53,101,,,N,N,667,5112,1301,213,30,117,631,700,38,39,,,913,776,5.2100000000,43,3,16,4023,1583,113,685,432,165,154,0.97,St. Louis Browns,Sportsman's Park IV,270936,105,109,SLB,SLA,SLA
+1949,AL,WS1,MIN,,8,154,77,50,104,,,N,N,584,5234,1330,207,41,81,593,495,46,33,,,868,762,5.1000000000,44,9,9,4035,1438,79,779,451,161,168,0.97,Washington Senators,Griffith Stadium I,770745,98,101,WSH,WS1,WS1
+1949,NL,BRO,LAD,,1,156,78,97,57,,,Y,N,879,5400,1477,236,47,152,638,570,117,,,,651,594,3.8000000000,62,15,17,4224,1306,132,582,743,124,162,0.97,Brooklyn Dodgers,Ebbets Field,1633747,105,102,BRO,BRO,BRO
+1949,NL,BSN,ATL,,4,157,77,75,79,,,N,N,706,5336,1376,246,33,103,684,656,28,,,,719,621,3.9900000000,68,12,11,4200,1466,110,520,589,149,144,0.97,Boston Braves,Braves Field,1081795,95,94,BSN,BSN,BSN
+1949,NL,CHN,CHC,,8,154,77,61,93,,,N,N,593,5214,1336,212,53,97,396,573,53,,,,773,678,4.5000000000,44,8,17,4071,1487,104,575,544,187,160,0.97,Chicago Cubs,Wrigley Field,1143139,97,99,CHC,CHN,CHN
+1949,NL,CIN,CIN,,7,156,78,62,92,,,N,N,627,5469,1423,264,35,86,429,559,31,,,,770,676,4.3400000000,55,10,6,4203,1423,124,640,538,136,150,0.97,Cincinnati Reds,Crosley Field,707782,101,104,CIN,CIN,CIN
+1949,NL,NY1,SFG,,5,156,79,73,81,,,N,N,736,5308,1383,203,52,147,613,523,43,,,,693,583,3.8200000000,68,10,9,4122,1328,132,544,516,163,134,0.97,New York Giants,Polo Grounds IV,1218446,101,99,NYG,NY1,NY1
+1949,NL,PHI,PHI,,3,154,77,81,73,,,N,N,662,5307,1349,232,55,122,528,670,27,,,,668,601,3.8900000000,58,12,15,4173,1389,104,502,495,158,141,0.97,Philadelphia Phillies,Shibe Park,819698,97,97,PHI,PHI,PHI
+1949,NL,PIT,PIT,,6,154,77,71,83,,,N,N,681,5214,1350,191,41,126,548,554,48,,,,760,689,4.5700000000,53,9,15,4068,1452,142,535,556,132,173,0.97,Pittsburgh Pirates,Forbes Field,1449435,102,104,PIT,PIT,PIT
+1949,NL,SLN,STL,,2,157,79,96,58,,,N,N,766,5463,1513,281,54,102,569,482,17,,,,616,538,3.4400000000,64,13,19,4221,1356,87,507,606,142,149,0.97,St. Louis Cardinals,Sportsman's Park IV,1430676,104,103,STL,SLN,SLN
+1950,AL,BOS,BOS,,3,154,77,94,60,,,N,N,1027,5516,1665,287,61,161,719,582,32,17,,,804,739,4.8800000000,66,6,28,4086,1413,121,748,630,111,181,0.98,Boston Red Sox,Fenway Park II,1344080,111,108,BOS,BOS,BOS
+1950,AL,CHA,CHW,,6,156,79,60,94,,,N,N,625,5260,1368,172,47,93,551,566,19,22,,,749,669,4.4100000000,62,7,9,4095,1370,107,734,566,135,181,0.97,Chicago White Sox,Comiskey Park,781330,97,98,CHW,CHA,CHA
+1950,AL,CLE,CLE,,4,155,77,92,62,,,N,N,806,5263,1417,222,46,164,693,624,40,34,,,654,574,3.7500000000,69,11,16,4134,1289,120,647,674,129,150,0.97,Cleveland Indians,Cleveland Stadium,1727464,97,94,CLE,CLE,CLE
+1950,AL,DET,DET,,2,157,81,95,59,,,N,N,837,5381,1518,285,50,114,722,480,23,40,,,713,644,4.1200000000,72,9,20,4221,1444,141,553,576,120,194,0.98,Detroit Tigers,Briggs Stadium,1951474,103,102,DET,DET,DET
+1950,AL,NYA,NYY,,1,155,77,98,56,,,Y,Y,914,5361,1511,234,70,159,687,463,41,28,,,691,633,4.1500000000,66,12,31,4116,1322,118,708,712,119,188,0.97,New York Yankees,Yankee Stadium I,2081380,97,94,NYY,NYA,NYA
+1950,AL,PHA,OAK,,8,154,77,52,102,,,N,N,670,5212,1361,204,53,100,685,493,42,25,,,913,821,5.4900000000,50,3,18,4038,1528,138,729,466,155,208,0.97,Philadelphia Athletics,Shibe Park,309805,98,99,PHA,PHA,PHA
+1950,AL,SLA,BAL,,7,154,74,58,96,,,N,N,684,5163,1269,235,43,106,690,744,39,40,,,916,789,5.2000000000,56,7,14,4095,1629,129,651,448,196,155,0.96,St. Louis Browns,Sportsman's Park IV,247131,104,108,SLB,SLA,SLA
+1950,AL,WS1,MIN,,5,155,78,67,87,,,N,N,690,5251,1365,190,53,76,671,606,42,25,,,813,706,4.6600000000,59,7,18,4092,1479,99,648,486,167,181,0.97,Washington Senators,Griffith Stadium I,699697,95,98,WSH,WS1,WS1
+1950,NL,BRO,LAD,,2,155,78,89,65,,,N,N,847,5364,1461,247,46,194,607,632,77,,,,724,661,4.2800000000,62,10,21,4167,1397,163,591,772,127,183,0.97,Brooklyn Dodgers,Ebbets Field,1185896,103,99,BRO,BRO,BRO
+1950,NL,BSN,ATL,,4,156,79,83,71,,,N,N,785,5363,1411,246,36,148,615,616,71,,,,736,637,4.1400000000,88,7,10,4155,1411,129,554,615,182,146,0.97,Boston Braves,Braves Field,944391,94,93,BSN,BSN,BSN
+1950,NL,CHN,CHC,,7,154,78,64,89,,,N,N,643,5230,1298,224,47,161,479,767,46,,,,772,652,4.2800000000,55,9,19,4113,1452,130,593,559,198,169,0.96,Chicago Cubs,Wrigley Field,1165944,99,102,CHC,CHN,CHN
+1950,NL,CIN,CIN,,6,153,76,66,87,,,N,N,654,5253,1366,257,27,99,504,497,37,,,,734,651,4.3200000000,67,7,13,4071,1363,145,582,686,140,132,0.97,Cincinnati Reds,Crosley Field,538794,100,102,CIN,CIN,CIN
+1950,NL,NY1,SFG,,3,154,76,86,68,,,N,N,735,5238,1352,204,50,133,627,629,42,,,,643,567,3.7100000000,70,19,15,4125,1268,140,536,596,137,181,0.97,New York Giants,Polo Grounds IV,1008878,101,99,NYG,NY1,NY1
+1950,NL,PHI,PHI,,1,157,78,91,63,,,Y,N,722,5426,1440,225,55,125,535,569,33,,,,624,547,3.5000000000,57,13,27,4218,1324,122,530,620,151,155,0.97,Philadelphia Phillies,Shibe Park,1217035,98,97,PHI,PHI,PHI
+1950,NL,PIT,PIT,,8,154,77,57,96,,,N,N,681,5327,1404,227,59,138,564,693,43,,,,857,754,4.9600000000,42,6,16,4104,1472,152,616,556,136,165,0.97,Pittsburgh Pirates,Forbes Field,1166267,103,106,PIT,PIT,PIT
+1950,NL,SLN,STL,,5,153,76,78,75,,,N,N,693,5215,1353,255,50,102,606,604,23,,,,670,598,3.9700000000,57,10,14,4068,1398,119,535,603,130,172,0.97,St. Louis Cardinals,Sportsman's Park IV,1093411,104,103,STL,SLN,SLN
+1951,AL,BOS,BOS,,3,154,75,87,67,,,N,N,804,5378,1428,233,32,127,756,594,20,21,,,725,644,4.1400000000,46,7,24,4197,1413,100,599,658,138,184,0.97,Boston Red Sox,Fenway Park II,1312282,110,109,BOS,BOS,BOS
+1951,AL,CHA,CHW,,4,155,78,81,73,,,N,N,714,5378,1453,229,64,86,596,524,99,70,,,644,551,3.5000000000,74,11,14,4254,1353,109,549,572,151,176,0.97,Chicago White Sox,Comiskey Park,1328234,98,98,CHW,CHA,CHA
+1951,AL,CLE,CLE,,2,155,77,93,61,,,N,N,696,5250,1346,208,35,140,606,632,52,35,,,594,522,3.3800000000,76,10,19,4173,1287,86,577,642,134,151,0.97,Cleveland Indians,Cleveland Stadium,1704984,95,92,CLE,CLE,CLE
+1951,AL,DET,DET,,5,154,77,73,81,,,N,N,685,5336,1413,231,35,104,568,525,37,34,,,741,660,4.2900000000,51,8,17,4152,1385,102,602,597,163,166,0.97,Detroit Tigers,Briggs Stadium,1132641,100,101,DET,DET,DET
+1951,AL,NYA,NYY,,1,154,78,98,56,,,Y,Y,798,5194,1395,208,48,140,605,547,78,39,,,621,541,3.5600000000,66,24,22,4101,1290,92,562,664,144,190,0.97,New York Yankees,Yankee Stadium I,1950107,97,93,NYY,NYA,NYA
+1951,AL,PHA,OAK,,6,154,79,70,84,,,N,N,736,5277,1381,262,43,102,677,565,47,36,,,745,674,4.4700000000,52,7,22,4074,1421,109,569,437,135,204,0.97,Philadelphia Athletics,Shibe Park,465469,101,103,PHA,PHA,PHA
+1951,AL,SLA,BAL,,8,154,77,52,102,,,N,N,611,5219,1288,223,47,86,521,693,35,38,,,882,789,5.1800000000,56,5,9,4110,1525,131,801,550,171,179,0.97,St. Louis Browns,Sportsman's Park IV,293790,103,107,SLB,SLA,SLA
+1951,AL,WS1,MIN,,7,154,76,62,92,,,N,N,672,5329,1399,242,45,54,560,515,45,38,,,764,681,4.4900000000,58,6,13,4098,1429,110,630,475,160,148,0.97,Washington Senators,Griffith Stadium I,695167,98,100,WSH,WS1,WS1
+1951,NL,BRO,LAD,,2,158,78,97,60,,,N,N,855,5492,1511,249,37,184,603,649,89,70,,,672,613,3.8800000000,64,10,13,4269,1360,150,549,693,128,192,0.97,Brooklyn Dodgers,Ebbets Field,1282628,102,99,BRO,BRO,BRO
+1951,NL,BSN,ATL,,4,155,78,76,78,,,N,N,723,5293,1385,234,37,130,565,617,80,34,,,662,579,3.7500000000,73,16,12,4167,1378,96,595,604,145,157,0.97,Boston Braves,Braves Field,487475,93,93,BSN,BSN,BSN
+1951,NL,CHN,CHC,,8,155,77,62,92,,,N,N,614,5307,1327,200,47,103,477,647,63,30,,,750,668,4.3400000000,48,10,10,4155,1416,125,572,544,181,161,0.97,Chicago Cubs,Wrigley Field,894415,102,104,CHC,CHN,CHN
+1951,NL,CIN,CIN,,6,155,77,68,86,,,N,N,559,5285,1309,215,33,88,415,577,44,40,,,667,571,3.7000000000,55,14,23,4170,1357,119,490,584,140,141,0.97,Cincinnati Reds,Crosley Field,588268,101,103,CIN,CIN,CIN
+1951,NL,NY1,SFG,,1,157,78,98,59,,,Y,N,781,5360,1396,201,53,179,671,624,55,34,,,641,546,3.4800000000,64,9,18,4236,1334,148,482,625,170,175,0.97,New York Giants,Polo Grounds IV,1059539,101,99,NYG,NY1,NY1
+1951,NL,PHI,PHI,,5,154,77,73,81,,,N,N,648,5332,1384,199,47,108,505,525,63,28,,,644,586,3.8100000000,57,19,15,4152,1373,110,497,570,138,146,0.97,Philadelphia Phillies,Shibe Park,937658,99,97,PHI,PHI,PHI
+1951,NL,PIT,PIT,,7,155,78,64,90,,,N,N,689,5318,1372,218,56,137,557,615,29,27,,,845,734,4.7900000000,40,9,22,4140,1479,157,609,580,170,178,0.97,Pittsburgh Pirates,Forbes Field,980590,102,106,PIT,PIT,PIT
+1951,NL,SLN,STL,,3,155,79,81,73,,,N,N,683,5317,1404,230,57,95,569,492,30,30,,,671,609,3.9500000000,58,9,23,4161,1391,119,568,546,125,187,0.98,St. Louis Cardinals,Sportsman's Park IV,1013429,100,100,STL,SLN,SLN
+1952,AL,BOS,BOS,,6,154,77,76,78,,,N,N,668,5246,1338,233,34,113,542,739,59,47,,,658,579,3.8000000000,53,7,24,4116,1332,107,623,624,143,181,0.97,Boston Red Sox,Fenway Park II,1115750,108,107,BOS,BOS,BOS
+1952,AL,CHA,CHW,,3,156,79,81,73,,,N,N,610,5316,1337,199,38,80,541,521,61,38,,,568,511,3.2500000000,53,15,28,4248,1251,86,578,774,123,158,0.98,Chicago White Sox,Comiskey Park,1231675,101,99,CHW,CHA,CHA
+1952,AL,CLE,CLE,,2,155,77,93,61,,,N,N,763,5330,1399,211,49,148,626,749,46,39,,,606,519,3.3200000000,80,19,18,4221,1278,94,556,671,155,141,0.97,Cleveland Indians,Cleveland Stadium,1444607,94,91,CLE,CLE,CLE
+1952,AL,DET,DET,,8,156,77,50,104,,,N,N,557,5258,1278,190,37,103,553,605,27,38,,,738,655,4.2500000000,51,10,14,4164,1394,111,591,702,149,145,0.97,Detroit Tigers,Briggs Stadium,1026846,100,103,DET,DET,DET
+1952,AL,NYA,NYY,,1,154,77,95,59,,,Y,Y,727,5294,1411,221,56,129,566,652,52,42,,,557,482,3.1400000000,72,21,27,4143,1240,94,581,666,127,199,0.97,New York Yankees,Yankee Stadium I,1629665,95,91,NYY,NYA,NYA
+1952,AL,PHA,OAK,,4,155,78,79,75,,,N,N,664,5163,1305,212,35,89,683,561,52,43,,,723,638,4.1500000000,73,11,16,4152,1402,113,526,562,140,148,0.97,Philadelphia Athletics,Shibe Park,627100,106,108,PHA,PHA,PHA
+1952,AL,SLA,BAL,,7,155,78,64,90,,,N,N,604,5353,1340,225,46,82,540,720,30,34,,,733,640,4.1200000000,48,6,18,4197,1388,111,598,581,155,176,0.97,St. Louis Browns,Sportsman's Park IV,518796,103,106,SLB,SLA,SLA
+1952,AL,WS1,MIN,,5,157,78,78,76,,,N,N,598,5357,1282,225,44,50,580,607,48,37,,,608,535,3.3700000000,75,10,15,4287,1405,78,577,574,130,152,0.97,Washington Senators,Griffith Stadium I,699457,97,97,WSH,WS1,WS1
+1952,NL,BRO,LAD,,1,155,80,96,57,,,Y,N,775,5266,1380,199,32,153,663,699,90,49,,,603,549,3.5300000000,45,11,24,4197,1295,121,544,773,105,169,0.98,Brooklyn Dodgers,Ebbets Field,1088704,102,98,BRO,BRO,BRO
+1952,NL,BSN,ATL,,7,155,77,64,89,,,N,N,569,5221,1214,187,31,110,483,711,58,34,,,651,586,3.7800000000,63,11,13,4188,1388,106,525,687,153,143,0.97,Boston Braves,Braves Field,281278,97,98,BSN,BSN,BSN
+1952,NL,CHN,CHC,,5,155,77,77,77,,,N,N,628,5330,1408,223,45,107,422,712,50,40,,,631,551,3.5800000000,59,15,15,4158,1265,101,534,661,146,123,0.97,Chicago Cubs,Wrigley Field,1024826,101,103,CHC,CHN,CHN
+1952,NL,CIN,CIN,,6,154,77,69,85,,,N,N,615,5234,1303,212,45,104,480,709,32,42,,,659,607,4.0100000000,56,11,12,4089,1377,111,517,579,107,145,0.98,Cincinnati Reds,Crosley Field,604197,99,100,CIN,CIN,CIN
+1952,NL,NY1,SFG,,2,154,77,92,62,,,N,N,722,5229,1337,186,56,151,536,672,30,31,,,639,547,3.5900000000,49,12,31,4113,1282,121,538,655,153,175,0.97,New York Giants,Polo Grounds IV,984940,101,99,NYG,NY1,NY1
+1952,NL,PHI,PHI,,4,154,76,87,67,,,N,N,657,5205,1353,237,45,93,540,534,60,41,,,552,473,3.0700000000,80,17,16,4158,1306,95,373,609,150,145,0.97,Philadelphia Phillies,Shibe Park,755417,99,98,PHI,PHI,PHI
+1952,NL,PIT,PIT,,8,155,77,42,112,,,N,N,515,5193,1201,181,30,92,486,724,43,41,,,793,704,4.6500000000,43,5,8,4089,1395,133,615,564,181,167,0.97,Pittsburgh Pirates,Forbes Field,686673,102,107,PIT,PIT,PIT
+1952,NL,SLN,STL,,3,154,77,88,66,,,N,N,677,5200,1386,247,54,97,537,479,33,32,,,630,553,3.6600000000,49,12,27,4083,1274,119,501,712,141,159,0.97,St. Louis Cardinals,Sportsman's Park IV,913113,100,99,STL,SLN,SLN
+1953,AL,BOS,BOS,,4,153,76,84,69,,,N,N,656,5246,1385,255,37,101,496,601,33,45,,,632,546,3.5800000000,41,15,37,4119,1333,92,584,642,147,173,0.97,Boston Red Sox,Fenway Park II,1026133,105,105,BOS,BOS,BOS
+1953,AL,CHA,CHW,,3,156,78,89,65,,,N,N,716,5212,1345,226,53,74,601,530,73,55,,,592,532,3.4100000000,57,17,33,4209,1299,113,583,714,125,144,0.98,Chicago White Sox,Comiskey Park,1191353,103,101,CHW,CHA,CHA
+1953,AL,CLE,CLE,,2,155,78,92,62,,,N,N,770,5285,1426,201,29,160,609,683,33,29,,,627,555,3.6400000000,81,11,15,4119,1311,92,519,586,126,197,0.97,Cleveland Indians,Cleveland Stadium,1069176,98,94,CLE,CLE,CLE
+1953,AL,DET,DET,,6,158,79,60,94,,,N,N,695,5553,1479,259,44,108,506,603,30,35,,,923,825,5.2500000000,50,2,16,4245,1633,154,585,645,131,149,0.97,Detroit Tigers,Briggs Stadium,884658,98,102,DET,DET,DET
+1953,AL,NYA,NYY,,1,151,77,99,52,,,Y,Y,801,5194,1420,226,52,139,656,644,34,44,,,547,483,3.2000000000,50,18,39,4074,1286,94,500,604,126,182,0.97,New York Yankees,Yankee Stadium I,1537811,97,93,NYY,NYA,NYA
+1953,AL,PHA,OAK,,7,157,78,59,95,,,N,N,632,5455,1398,205,38,116,498,602,41,24,,,799,731,4.6700000000,51,7,11,4227,1475,121,594,566,136,161,0.97,Philadelphia Athletics,Connie Mack Stadium,362113,104,107,PHA,PHA,PHA
+1953,AL,SLA,BAL,,8,154,77,54,100,,,N,N,555,5264,1310,214,25,112,507,644,17,34,,,778,688,4.4800000000,28,10,24,4149,1467,101,626,639,152,165,0.97,St. Louis Browns,Sportsman's Park IV,297238,102,106,SLB,SLA,SLA
+1953,AL,WS1,MIN,,5,152,75,76,76,,,N,N,687,5149,1354,230,53,69,596,604,65,36,,,614,547,3.6600000000,76,16,10,4032,1313,112,478,515,120,173,0.97,Washington Senators,Griffith Stadium I,595594,98,98,WSH,WS1,WS1
+1953,NL,BRO,LAD,,1,155,78,105,49,,,Y,N,955,5373,1529,274,59,208,655,686,90,47,,,689,629,4.1000000000,51,11,29,4140,1337,169,509,817,118,161,0.98,Brooklyn Dodgers,Ebbets Field,1163419,103,100,BRO,BRO,BRO
+1953,NL,CHN,CHC,,7,155,77,65,89,,,N,N,633,5272,1372,204,57,137,514,746,49,21,,,835,723,4.7900000000,38,3,22,4077,1491,151,554,623,193,141,0.96,Chicago Cubs,Wrigley Field,763658,102,103,CHC,CHN,CHN
+1953,NL,CIN,CIN,,6,155,78,68,86,,,N,N,714,5343,1396,190,34,166,485,701,25,20,,,788,704,4.6400000000,47,7,15,4095,1484,179,488,506,129,176,0.97,Cincinnati Reds,Crosley Field,548086,100,101,CIN,CIN,CIN
+1953,NL,ML1,ATL,,2,157,79,92,62,,,N,N,738,5349,1422,227,52,156,439,637,46,27,,,589,509,3.3000000000,72,14,15,4161,1282,107,539,738,143,169,0.97,Milwaukee Braves,County Stadium,1826397,94,92,MLN,MLN,MLN
+1953,NL,NY1,SFG,,5,155,77,70,84,,,N,N,768,5362,1452,195,45,176,499,608,31,21,,,747,645,4.2500000000,46,10,20,4095,1403,146,610,647,150,151,0.97,New York Giants,Polo Grounds IV,811518,103,101,NYG,NY1,NY1
+1953,NL,PHI,PHI,,3,156,78,83,71,,,N,N,716,5290,1400,228,62,115,530,597,42,21,,,666,578,3.8000000000,76,13,15,4107,1410,138,410,637,147,161,0.97,Philadelphia Phillies,Connie Mack Stadium,853644,99,98,PHI,PHI,PHI
+1953,NL,PIT,PIT,,8,154,77,50,104,,,N,N,622,5253,1297,178,49,99,524,715,41,39,,,887,788,5.2200000000,49,4,10,4074,1529,168,577,607,163,139,0.97,Pittsburgh Pirates,Forbes Field,572757,99,104,PIT,PIT,PIT
+1953,NL,SLN,STL,,3,157,78,83,71,,,N,N,768,5397,1474,281,56,140,574,617,18,22,,,713,651,4.2300000000,51,11,36,4158,1406,139,533,732,138,161,0.97,St. Louis Cardinals,Sportsman's Park IV,880242,100,99,STL,SLN,SLN
+1954,AL,BAL,BAL,,7,154,77,54,100,,,N,N,483,5206,1309,195,49,52,468,634,30,31,,,668,592,3.8800000000,58,6,8,4119,1279,78,688,668,147,152,0.97,Baltimore Orioles,Memorial Stadium,1060910,92,96,BAL,BAL,BAL
+1954,AL,BOS,BOS,,4,156,79,69,85,,,N,N,700,5399,1436,244,41,123,654,660,51,30,,,728,629,4.0100000000,41,10,22,4236,1434,118,612,707,176,163,0.97,Boston Red Sox,Fenway Park II,931127,111,110,BOS,BOS,BOS
+1954,AL,CHA,CHW,,3,155,78,94,60,,,N,N,711,5168,1382,203,47,94,604,536,98,58,,,521,469,3.0500000000,60,23,33,4149,1255,94,517,701,108,149,0.98,Chicago White Sox,Comiskey Park,1231629,103,100,CHW,CHA,CHA
+1954,AL,CLE,CLE,,1,156,77,111,43,,,Y,N,746,5222,1368,188,39,156,637,668,30,33,,,504,438,2.7800000000,77,12,36,4257,1220,89,486,678,128,148,0.97,Cleveland Indians,Cleveland Stadium,1335472,102,98,CLE,CLE,CLE
+1954,AL,DET,DET,,5,155,77,68,86,,,N,N,584,5233,1351,215,41,90,492,603,48,44,,,664,585,3.8100000000,58,13,13,4149,1375,138,506,603,127,131,0.97,Detroit Tigers,Briggs Stadium,1079847,99,100,DET,DET,DET
+1954,AL,NYA,NYY,,2,155,78,103,51,,,N,N,805,5226,1400,215,59,133,650,632,34,41,,,563,500,3.2600000000,51,16,37,4137,1284,86,552,655,127,198,0.97,New York Yankees,Yankee Stadium I,1475171,97,93,NYY,NYA,NYA
+1954,AL,PHA,OAK,,8,156,77,51,103,,,N,N,542,5206,1228,191,41,94,504,677,30,29,,,875,789,5.1800000000,49,3,13,4113,1523,141,685,555,166,163,0.97,Philadelphia Athletics,Connie Mack Stadium,304666,100,105,PHA,PHA,PHA
+1954,AL,WS1,MIN,,6,155,78,66,88,,,N,N,632,5249,1292,188,69,81,610,719,37,21,,,680,590,3.8400000000,69,10,7,4149,1396,79,573,562,137,172,0.97,Washington Senators,Griffith Stadium I,503542,94,96,WSH,WS1,WS1
+1954,NL,BRO,LAD,,2,154,77,92,62,,,N,N,778,5251,1418,246,56,186,634,625,46,39,,,740,667,4.3100000000,39,8,36,4179,1399,164,533,762,129,138,0.97,Brooklyn Dodgers,Ebbets Field,1020531,104,101,BRO,BRO,BRO
+1954,NL,CHN,CHC,,7,154,77,64,90,,,N,N,700,5359,1412,229,45,159,478,693,46,31,,,766,689,4.5100000000,41,6,19,4122,1375,131,619,622,154,164,0.97,Chicago Cubs,Wrigley Field,748183,101,103,CHC,CHN,CHN
+1954,NL,CIN,CIN,,5,154,77,74,80,,,N,N,729,5234,1369,221,46,147,557,645,47,30,,,763,684,4.5000000000,34,8,27,4101,1491,169,547,537,137,194,0.97,Cincinnati Redlegs,Crosley Field,704167,102,103,CIN,CIN,CIN
+1954,NL,ML1,ATL,,3,154,77,89,65,,,N,N,670,5261,1395,217,41,139,471,619,54,31,,,556,494,3.1900000000,63,13,21,4182,1296,106,553,698,116,171,0.98,Milwaukee Braves,County Stadium,2131388,94,92,MLN,MLN,MLN
+1954,NL,NY1,SFG,,1,154,76,97,57,,,Y,Y,732,5245,1386,194,42,186,522,561,30,23,,,550,477,3.0900000000,45,19,33,4170,1258,113,613,692,154,172,0.97,New York Giants,Polo Grounds IV,1155067,101,100,NYG,NY1,NY1
+1954,NL,PHI,PHI,,4,154,78,75,79,,,N,N,659,5184,1384,243,58,102,604,620,30,27,,,614,544,3.5900000000,78,14,12,4095,1329,133,450,570,145,133,0.97,Philadelphia Phillies,Connie Mack Stadium,738991,100,99,PHI,PHI,PHI
+1954,NL,PIT,PIT,,8,154,77,53,101,,,N,N,557,5088,1260,181,57,76,566,737,21,13,,,845,736,4.9200000000,37,4,15,4038,1510,128,564,525,173,136,0.97,Pittsburgh Pirates,Forbes Field,475494,98,102,PIT,PIT,PIT
+1954,NL,SLN,STL,,6,154,77,72,82,,,N,N,799,5405,1518,285,58,119,582,586,63,46,,,790,695,4.5000000000,40,11,18,4170,1484,170,535,680,146,178,0.97,St. Louis Cardinals,Sportsman's Park IV,1039698,100,101,STL,SLN,SLN
+1955,AL,BAL,BAL,,7,156,79,57,97,,,N,N,540,5257,1263,177,39,54,560,742,34,46,,,754,649,4.2100000000,35,10,20,4164,1403,103,625,595,162,159,0.97,Baltimore Orioles,Memorial Stadium,852039,92,96,BAL,BAL,BAL
+1955,AL,BOS,BOS,,4,154,78,84,70,,,N,N,755,5273,1392,241,39,137,707,733,43,17,,,652,572,3.7200000000,44,9,34,4152,1333,128,582,674,136,140,0.97,Boston Red Sox,Fenway Park II,1203200,110,109,BOS,BOS,BOS
+1955,AL,CHA,CHW,,3,155,77,91,63,,,N,N,725,5220,1401,204,36,116,567,595,69,45,,,557,516,3.3700000000,55,20,23,4134,1301,111,497,720,111,147,0.98,Chicago White Sox,Comiskey Park,1175684,103,99,CHW,CHA,CHA
+1955,AL,CLE,CLE,,2,154,77,93,61,,,N,N,698,5146,1325,195,31,148,723,715,28,24,,,601,522,3.3900000000,45,15,36,4158,1285,111,558,877,108,152,0.98,Cleveland Indians,Cleveland Stadium,1221780,104,101,CLE,CLE,CLE
+1955,AL,DET,DET,,5,154,77,79,75,,,N,N,775,5283,1407,211,38,130,641,583,41,22,,,658,581,3.7900000000,66,16,12,4140,1381,126,517,629,139,159,0.97,Detroit Tigers,Briggs Stadium,1181838,98,98,DET,DET,DET
+1955,AL,KC1,OAK,,6,155,76,63,91,,,N,N,638,5335,1395,189,46,121,463,725,22,36,,,911,822,5.3500000000,29,9,22,4146,1486,175,707,572,146,174,0.97,Kansas City Athletics,Municipal Stadium I,1393054,100,105,KCA,KC1,KC1
+1955,AL,NYA,NYY,,1,154,77,96,58,,,Y,N,762,5161,1342,179,55,175,609,658,55,25,,,569,492,3.2300000000,52,19,33,4116,1163,108,688,731,128,180,0.97,New York Yankees,Yankee Stadium I,1490138,99,95,NYY,NYA,NYA
+1955,AL,WS1,MIN,,8,154,77,53,101,,,N,N,598,5142,1277,178,54,80,538,654,25,32,,,789,695,4.6200000000,37,10,16,4062,1450,99,634,607,154,170,0.97,Washington Senators,Griffith Stadium I,425238,97,100,WSH,WS1,WS1
+1955,NL,BRO,LAD,,1,154,77,98,55,,,Y,Y,857,5193,1406,230,44,201,674,718,79,56,,,650,563,3.6800000000,46,11,37,4134,1296,168,483,773,133,156,0.97,Brooklyn Dodgers,Ebbets Field,1033589,104,101,BRO,BRO,BRO
+1955,NL,CHN,CHC,,6,154,77,72,81,,,N,N,626,5214,1287,187,55,164,428,806,37,35,,,713,638,4.1700000000,47,10,23,4134,1306,153,601,686,147,147,0.97,Chicago Cubs,Wrigley Field,875800,100,102,CHC,CHN,CHN
+1955,NL,CIN,CIN,,5,154,77,75,79,,,N,N,761,5270,1424,216,28,181,556,657,51,36,,,684,598,3.9500000000,38,12,22,4089,1373,161,443,576,139,169,0.97,Cincinnati Redlegs,Crosley Field,693662,105,104,CIN,CIN,CIN
+1955,NL,ML1,ATL,,2,154,77,85,69,,,N,N,743,5277,1377,219,55,182,504,735,42,27,,,668,592,3.8500000000,61,5,12,4149,1339,138,591,654,152,155,0.97,Milwaukee Braves,County Stadium,2005836,96,93,MLN,MLN,MLN
+1955,NL,NY1,SFG,,3,154,79,80,74,,,N,N,702,5288,1377,173,34,169,497,581,38,22,,,673,581,3.7700000000,52,6,14,4158,1347,155,560,721,141,165,0.97,New York Giants,Polo Grounds IV,824112,101,101,NYG,NY1,NY1
+1955,NL,PHI,PHI,,4,154,77,77,77,,,N,N,675,5092,1300,214,50,132,652,673,44,32,,,666,592,3.9300000000,58,11,21,4068,1291,161,477,657,110,117,0.98,Philadelphia Phillies,Connie Mack Stadium,922886,98,98,PHI,PHI,PHI
+1955,NL,PIT,PIT,,8,154,75,60,94,,,N,N,560,5173,1262,210,60,91,471,652,22,22,,,767,664,4.3900000000,41,5,16,4086,1480,142,536,622,166,175,0.97,Pittsburgh Pirates,Forbes Field,469397,98,101,PIT,PIT,PIT
+1955,NL,SLN,STL,,7,154,77,68,86,,,N,N,654,5266,1375,228,36,143,458,597,64,59,,,757,697,4.5600000000,42,10,15,4128,1376,185,549,730,146,152,0.97,St. Louis Cardinals,Sportsman's Park IV,849130,99,100,STL,SLN,SLN
+1956,AL,BAL,BAL,,6,154,77,69,85,,,N,N,571,5090,1242,198,34,91,563,725,39,42,,,705,635,4.2000000000,38,10,24,4080,1362,99,547,715,133,142,0.97,Baltimore Orioles,Memorial Stadium,901201,92,94,BAL,BAL,BAL
+1956,AL,BOS,BOS,,4,155,78,84,70,,,N,N,780,5349,1473,261,45,139,727,687,28,19,,,751,648,4.1700000000,50,8,20,4194,1354,130,668,712,169,168,0.97,Boston Red Sox,Fenway Park II,1137158,113,111,BOS,BOS,BOS
+1956,AL,CHA,CHW,,3,154,77,85,69,,,N,N,776,5286,1412,218,43,128,619,660,70,33,,,634,576,3.7300000000,65,11,13,4167,1351,118,524,722,122,160,0.97,Chicago White Sox,Comiskey Park,1000090,101,99,CHW,CHA,CHA
+1956,AL,CLE,CLE,,2,155,77,88,66,,,N,N,712,5148,1256,199,23,153,681,764,40,32,,,581,511,3.3200000000,67,17,24,4152,1233,116,564,845,129,130,0.97,Cleveland Indians,Cleveland Stadium,865467,103,101,CLE,CLE,CLE
+1956,AL,DET,DET,,5,155,78,82,72,,,N,N,789,5364,1494,209,50,150,644,618,43,26,,,699,622,4.0600000000,62,10,15,4137,1389,140,655,788,140,151,0.97,Detroit Tigers,Briggs Stadium,1051182,101,100,DET,DET,DET
+1956,AL,KC1,OAK,,8,154,77,52,102,,,N,N,619,5256,1325,204,41,112,480,727,40,30,,,831,740,4.8600000000,30,3,18,4110,1424,187,679,636,166,187,0.97,Kansas City Athletics,Municipal Stadium I,1015154,100,104,KCA,KC1,KC1
+1956,AL,NYA,NYY,,1,154,77,97,57,,,Y,Y,857,5312,1433,193,55,190,615,755,51,37,,,631,557,3.6300000000,50,10,35,4146,1285,114,652,732,135,214,0.97,New York Yankees,Yankee Stadium I,1491784,97,93,NYY,NYA,NYA
+1956,AL,WS1,MIN,,7,155,77,59,95,,,N,N,652,5202,1302,198,62,112,690,877,37,34,,,924,810,5.3300000000,36,1,18,4104,1539,171,730,663,168,173,0.97,Washington Senators,Griffith Stadium II,431647,96,100,WSH,WS1,WS1
+1956,NL,BRO,LAD,,1,154,77,93,61,,,Y,N,720,5098,1315,212,36,179,649,738,65,37,,,601,543,3.5700000000,46,12,30,4104,1251,171,441,772,111,149,0.98,Brooklyn Dodgers,Ebbets Field,1213562,109,107,BRO,BRO,BRO
+1956,NL,CHN,CHC,,8,157,80,60,94,,,N,N,597,5260,1281,202,50,142,446,776,55,38,,,708,612,3.9600000000,37,6,17,4176,1325,161,613,744,143,141,0.97,Chicago Cubs,Wrigley Field,720118,99,101,CHC,CHN,CHN
+1956,NL,CIN,CIN,,3,155,77,91,63,,,N,N,775,5291,1406,201,32,221,528,760,45,22,,,658,594,3.8500000000,47,4,29,4167,1406,141,458,653,113,147,0.98,Cincinnati Redlegs,Crosley Field,1125928,106,105,CIN,CIN,CIN
+1956,NL,ML1,ATL,,2,155,77,92,62,,,N,N,709,5207,1350,212,54,177,486,714,29,20,,,569,481,3.1100000000,64,12,27,4179,1295,133,467,639,130,159,0.97,Milwaukee Braves,County Stadium,2046331,94,92,MLN,MLN,MLN
+1956,NL,NY1,SFG,,6,154,77,67,87,,,N,N,540,5190,1268,192,45,145,402,659,67,34,,,650,579,3.7800000000,31,9,28,4134,1287,144,551,765,144,143,0.97,New York Giants,Polo Grounds IV,629179,99,100,NYG,NY1,NY1
+1956,NL,PHI,PHI,,5,154,77,71,83,,,N,N,668,5204,1313,207,49,121,585,673,45,23,,,738,643,4.2000000000,57,4,15,4131,1407,172,437,750,144,140,0.97,Philadelphia Phillies,Connie Mack Stadium,934798,98,99,PHI,PHI,PHI
+1956,NL,PIT,PIT,,7,157,78,66,88,,,N,N,588,5221,1340,199,57,110,383,752,24,33,,,653,572,3.7400000000,37,8,24,4128,1406,142,469,662,162,140,0.97,Pittsburgh Pirates,Forbes Field,949878,97,100,PIT,PIT,PIT
+1956,NL,SLN,STL,,4,156,78,76,78,,,N,N,678,5378,1443,234,49,124,503,622,41,35,,,698,612,3.9700000000,41,12,30,4164,1339,155,546,709,134,172,0.97,St. Louis Cardinals,Sportsman's Park IV,1029773,100,100,STL,SLN,SLN
+1957,AL,BAL,BAL,,5,154,77,76,76,,,N,N,597,5264,1326,191,39,87,504,699,57,35,,,588,541,3.4600000000,44,13,25,4224,1272,95,493,767,109,159,0.98,Baltimore Orioles,Memorial Stadium,1029581,93,94,BAL,BAL,BAL
+1957,AL,BOS,BOS,,3,154,77,82,72,,,N,N,721,5267,1380,231,32,153,624,739,29,21,,,668,593,3.8800000000,55,9,23,4128,1391,116,498,692,149,179,0.97,Boston Red Sox,Fenway Park II,1181087,106,105,BOS,BOS,BOS
+1957,AL,CHA,CHW,,2,155,77,90,64,,,N,N,707,5265,1369,208,41,106,633,745,109,51,,,566,521,3.3500000000,59,16,27,4203,1305,124,470,665,107,169,0.98,Chicago White Sox,Comiskey Park,1135668,101,99,CHW,CHA,CHA
+1957,AL,CLE,CLE,,6,153,77,76,77,,,N,N,682,5171,1304,199,26,140,591,786,40,47,,,722,623,4.0600000000,46,7,23,4140,1381,130,618,807,147,154,0.97,Cleveland Indians,Cleveland Stadium,722256,100,99,CLE,CLE,CLE
+1957,AL,DET,DET,,4,154,77,78,76,,,N,N,614,5348,1376,224,37,116,504,643,36,47,,,614,561,3.5600000000,52,9,21,4251,1330,147,505,756,121,151,0.98,Detroit Tigers,Briggs Stadium,1272346,103,102,DET,DET,DET
+1957,AL,KC1,OAK,,7,154,77,59,94,,,N,N,563,5170,1262,195,40,166,364,760,35,27,,,710,637,4.1900000000,26,6,19,4107,1344,153,565,626,125,162,0.97,Kansas City Athletics,Municipal Stadium I,901067,101,104,KCA,KC1,KC1
+1957,AL,NYA,NYY,,1,154,77,98,56,,,Y,N,723,5271,1412,200,54,145,562,709,49,38,,,534,465,3.0000000000,41,13,42,4185,1198,110,580,810,123,183,0.98,New York Yankees,Yankee Stadium I,1497134,99,95,NYY,NYA,NYA
+1957,AL,WS1,MIN,,8,154,77,55,99,,,N,N,603,5231,1274,215,38,111,527,733,13,38,,,808,742,4.8500000000,31,5,16,4131,1482,149,580,691,128,159,0.97,Washington Senators,Griffith Stadium II,457079,98,102,WSH,WS1,WS1
+1957,NL,BRO,LAD,,3,154,77,84,70,,,N,N,690,5242,1325,188,38,147,550,848,60,34,,,591,521,3.3500000000,44,18,29,4197,1285,144,456,891,127,136,0.97,Brooklyn Dodgers,Ebbets Field,1028258,109,107,BRO,BRO,BRO
+1957,NL,CHN,CHC,,7,156,78,62,92,,,N,N,628,5369,1312,223,31,147,461,989,28,25,,,722,644,4.1300000000,30,5,26,4209,1397,144,601,859,137,140,0.97,Chicago Cubs,Wrigley Field,670629,99,100,CHC,CHN,CHN
+1957,NL,CIN,CIN,,4,154,77,80,74,,,N,N,747,5389,1452,251,33,187,546,752,51,36,,,781,716,4.6200000000,40,5,29,4185,1486,179,429,707,107,139,0.98,Cincinnati Redlegs,Crosley Field,1070850,106,105,CIN,CIN,CIN
+1957,NL,ML1,ATL,,1,155,78,95,59,,,Y,Y,772,5458,1469,221,62,199,461,729,35,16,,,613,544,3.4700000000,60,9,24,4233,1347,124,570,693,120,173,0.98,Milwaukee Braves,County Stadium,2215404,93,90,MLN,MLN,MLN
+1957,NL,NY1,SFG,,6,154,77,69,85,,,N,N,643,5346,1349,171,54,157,447,669,64,38,,,701,623,4.0100000000,35,9,20,4194,1436,150,471,701,161,180,0.97,New York Giants,Polo Grounds IV,653923,100,102,NYG,NY1,NY1
+1957,NL,PHI,PHI,,5,156,78,77,77,,,N,N,623,5241,1311,213,44,117,534,758,57,26,,,656,590,3.7900000000,54,9,23,4203,1363,139,412,858,136,117,0.97,Philadelphia Phillies,Connie Mack Stadium,1146230,97,98,PHI,PHI,PHI
+1957,NL,PIT,PIT,,7,155,77,62,92,,,N,N,586,5402,1447,231,60,92,374,733,46,35,,,696,601,3.8800000000,47,9,15,4185,1463,158,421,663,170,143,0.97,Pittsburgh Pirates,Forbes Field,850732,97,98,PIT,PIT,PIT
+1957,NL,SLN,STL,,2,154,77,87,67,,,N,N,737,5472,1497,235,43,132,493,672,58,44,,,666,593,3.7800000000,46,11,29,4239,1385,140,506,778,131,168,0.97,St. Louis Cardinals,Sportsman's Park IV,1183575,102,102,STL,SLN,SLN
+1958,AL,BAL,BAL,,6,154,78,74,79,,,N,N,521,5111,1233,195,19,108,483,731,33,35,,,575,517,3.4000000000,55,15,28,4107,1277,106,403,749,114,159,0.98,Baltimore Orioles,Memorial Stadium,829991,94,95,BAL,BAL,BAL
+1958,AL,BOS,BOS,,3,155,77,79,75,,,N,N,697,5218,1335,229,30,155,638,820,29,22,,,691,601,3.9200000000,44,5,28,4140,1396,121,521,695,145,172,0.97,Boston Red Sox,Fenway Park II,1077047,106,106,BOS,BOS,BOS
+1958,AL,CHA,CHW,,2,155,77,82,72,,,N,N,634,5249,1348,191,42,101,518,669,101,33,,,615,557,3.6100000000,55,15,25,4167,1296,152,515,751,113,160,0.98,Chicago White Sox,Comiskey Park,797451,99,97,CHW,CHA,CHA
+1958,AL,CLE,CLE,,4,153,76,77,76,,,N,N,694,5201,1340,210,31,161,494,819,50,49,,,635,569,3.7300000000,51,2,20,4119,1283,123,604,766,152,171,0.97,Cleveland Indians,Cleveland Stadium,663805,99,98,CLE,CLE,CLE
+1958,AL,DET,DET,,5,154,77,77,77,,,N,N,659,5194,1384,229,41,109,463,678,48,32,,,606,541,3.5900000000,59,8,19,4071,1294,133,437,797,106,140,0.98,Detroit Tigers,Briggs Stadium,1098924,107,107,DET,DET,DET
+1958,AL,KC1,OAK,,7,156,78,73,81,,,N,N,642,5261,1297,196,50,138,452,747,22,36,,,713,645,4.1500000000,42,9,25,4194,1405,150,467,721,125,166,0.97,Kansas City Athletics,Municipal Stadium I,925090,102,104,KCA,KC1,KC1
+1958,AL,NYA,NYY,,1,155,78,92,62,,,Y,Y,759,5294,1418,212,39,164,537,822,48,32,,,577,493,3.2200000000,53,21,33,4137,1201,116,557,796,128,182,0.97,New York Yankees,Yankee Stadium I,1428438,96,94,NYY,NYA,NYA
+1958,AL,WS1,MIN,,8,156,78,61,93,,,N,N,553,5156,1240,161,38,121,477,751,22,41,,,747,693,4.5300000000,28,6,28,4128,1443,156,558,762,118,163,0.98,Washington Senators,Griffith Stadium II,475288,98,101,WSH,WS1,WS1
+1958,NL,CHN,CHC,,5,154,77,72,82,,,N,N,709,5289,1402,207,49,182,487,853,39,23,,,725,638,4.2200000000,27,5,24,4083,1322,142,619,805,150,161,0.97,Chicago Cubs,Wrigley Field,979904,98,99,CHC,CHN,CHN
+1958,NL,CIN,CIN,,4,154,77,76,78,,,N,N,695,5273,1359,242,40,123,572,765,61,38,,,621,574,3.7300000000,50,7,20,4155,1422,148,419,705,99,148,0.98,Cincinnati Redlegs,Crosley Field,788582,106,105,CIN,CIN,CIN
+1958,NL,LAN,LAD,,7,154,77,71,83,,,N,N,668,5173,1297,166,50,172,495,850,73,47,,,761,679,4.4700000000,30,7,31,4104,1399,173,606,855,145,198,0.97,Los Angeles Dodgers,Los Angeles Memorial Coliseum,1845556,104,104,LAD,LAN,LAN
+1958,NL,ML1,ATL,,1,154,77,92,62,,,Y,N,675,5225,1388,221,21,167,478,646,26,8,,,541,491,3.2100000000,72,16,17,4128,1261,125,426,773,119,152,0.98,Milwaukee Braves,County Stadium,1971101,92,89,MLN,MLN,MLN
+1958,NL,PHI,PHI,,8,154,77,69,85,,,N,N,664,5363,1424,238,56,124,573,871,51,33,,,762,671,4.3200000000,51,6,15,4191,1480,148,446,778,129,136,0.97,Philadelphia Phillies,Connie Mack Stadium,931110,99,100,PHI,PHI,PHI
+1958,NL,PIT,PIT,,2,154,77,84,70,,,N,N,662,5247,1386,229,68,134,396,753,30,15,,,607,541,3.5600000000,43,10,41,4101,1344,123,470,679,133,173,0.97,Pittsburgh Pirates,Forbes Field,1311988,97,98,PIT,PIT,PIT
+1958,NL,SFN,SFG,,3,154,77,80,74,,,N,N,727,5318,1399,250,42,170,531,817,64,29,,,698,614,3.9800000000,38,7,25,4167,1400,166,512,775,152,156,0.97,San Francisco Giants,Seals Stadium,1272625,98,96,SFG,SFN,SFN
+1958,NL,SLN,STL,,5,154,77,72,82,,,N,N,619,5255,1371,216,39,111,533,637,44,43,,,704,632,4.1200000000,45,6,25,4143,1398,158,567,822,152,163,0.97,St. Louis Cardinals,Sportsman's Park IV,1063730,104,105,STL,SLN,SLN
+1959,AL,BAL,BAL,,6,155,78,74,80,,,N,N,551,5208,1240,182,23,109,536,690,36,24,,,621,554,3.5600000000,45,15,30,4200,1290,111,476,735,147,163,0.97,Baltimore Orioles,Memorial Stadium,891926,97,98,BAL,BAL,BAL
+1959,AL,BOS,BOS,,5,154,77,75,79,,,N,N,726,5225,1335,248,28,125,626,810,68,25,,,696,632,4.1700000000,38,9,25,4092,1386,135,589,724,130,167,0.97,Boston Red Sox,Fenway Park II,984102,104,104,BOS,BOS,BOS
+1959,AL,CHA,CHW,,1,156,78,94,60,,,Y,N,669,5297,1325,220,46,97,580,634,113,53,,,588,521,3.2900000000,44,13,36,4275,1297,129,525,761,130,141,0.97,Chicago White Sox,Comiskey Park,1423144,99,98,CHW,CHA,CHA
+1959,AL,CLE,CLE,,2,154,77,89,65,,,N,N,745,5288,1390,216,25,167,433,721,33,36,,,646,576,3.7500000000,58,7,23,4149,1230,148,635,799,127,138,0.97,Cleveland Indians,Cleveland Stadium,1497976,97,96,CLE,CLE,CLE
+1959,AL,DET,DET,,4,154,77,76,78,,,N,N,713,5211,1346,196,30,160,580,737,34,17,,,732,635,4.2000000000,53,9,24,4080,1327,177,432,829,124,131,0.97,Detroit Tigers,Briggs Stadium,1221221,106,105,DET,DET,DET
+1959,AL,KC1,OAK,,7,154,77,66,88,,,N,N,681,5264,1383,231,43,117,481,780,34,24,,,760,657,4.3500000000,44,8,21,4080,1452,148,492,703,159,156,0.97,Kansas City Athletics,Municipal Stadium I,963683,102,104,KCA,KC1,KC1
+1959,AL,NYA,NYY,,3,155,77,79,75,,,N,N,687,5379,1397,224,40,153,457,828,45,22,,,647,560,3.6000000000,38,15,28,4197,1281,120,594,836,131,160,0.97,New York Yankees,Yankee Stadium I,1552030,96,94,NYY,NYA,NYA
+1959,AL,WS1,MIN,,8,154,77,63,91,,,N,N,619,5092,1205,173,32,163,517,881,51,34,,,701,606,4.0100000000,46,10,21,4080,1358,123,467,694,162,140,0.97,Washington Senators,Griffith Stadium II,615372,100,102,WSH,WS1,WS1
+1959,NL,CHN,CHC,,5,155,77,74,80,,,N,N,673,5296,1321,209,44,163,498,911,32,19,,,688,620,4.0100000000,30,11,25,4173,1337,152,519,765,140,142,0.97,Chicago Cubs,Wrigley Field,858255,99,100,CHC,CHN,CHN
+1959,NL,CIN,CIN,,5,154,77,74,80,,,N,N,764,5288,1448,258,34,161,499,763,65,28,,,738,650,4.3100000000,44,7,26,4071,1460,162,456,690,126,157,0.97,Cincinnati Redlegs,Crosley Field,801298,104,103,CIN,CIN,CIN
+1959,NL,LAN,LAD,,1,156,78,88,68,,,Y,Y,705,5282,1360,196,46,148,591,891,84,51,,,670,594,3.7900000000,43,14,26,4233,1317,157,614,1077,114,154,0.98,Los Angeles Dodgers,Los Angeles Memorial Coliseum,2071045,107,107,LAD,LAN,LAN
+1959,NL,ML1,ATL,,2,157,79,86,70,,,N,N,724,5388,1426,216,36,177,488,765,41,14,,,623,546,3.5100000000,69,18,18,4200,1406,128,429,775,123,138,0.97,Milwaukee Braves,County Stadium,1749112,92,90,MLN,MLN,MLN
+1959,NL,PHI,PHI,,8,155,78,64,90,,,N,N,599,5109,1237,196,38,113,498,858,39,46,,,725,642,4.2700000000,54,8,15,4062,1357,150,474,769,154,132,0.97,Philadelphia Phillies,Connie Mack Stadium,802815,101,104,PHI,PHI,PHI
+1959,NL,PIT,PIT,,4,155,77,78,76,,,N,N,651,5369,1414,230,42,112,442,715,32,26,,,680,604,3.9000000000,48,7,17,4179,1432,134,418,730,154,165,0.97,Pittsburgh Pirates,Forbes Field,1359917,99,98,PIT,PIT,PIT
+1959,NL,SFN,SFG,,3,154,77,83,71,,,N,N,705,5281,1377,239,35,167,473,875,81,34,,,613,531,3.4700000000,52,12,23,4128,1279,139,500,873,152,118,0.97,San Francisco Giants,Seals Stadium,1422130,98,96,SFG,SFN,SFN
+1959,NL,SLN,STL,,7,154,77,71,83,,,N,N,641,5317,1432,244,49,118,485,747,65,53,,,725,657,4.3400000000,36,8,21,4089,1427,137,564,846,146,158,0.97,St. Louis Cardinals,Sportsman's Park IV,929953,106,107,STL,SLN,SLN
+1960,AL,BAL,BAL,,2,154,77,89,65,,,N,N,682,5170,1307,206,33,123,596,801,37,24,,,606,538,3.5200000000,48,11,22,4125,1222,117,552,785,107,172,0.98,Baltimore Orioles,Memorial Stadium,1187849,99,98,BAL,BAL,BAL
+1960,AL,BOS,BOS,,7,154,77,65,89,,,N,N,658,5215,1359,234,32,124,570,798,34,28,,,775,699,4.6200000000,34,6,23,4083,1440,127,580,767,140,156,0.97,Boston Red Sox,Fenway Park II,1129866,103,103,BOS,BOS,BOS
+1960,AL,CHA,CHW,,3,154,77,87,67,,,N,N,741,5191,1402,242,38,112,567,648,122,48,,,617,552,3.6000000000,42,11,26,4143,1338,127,533,695,109,175,0.98,Chicago White Sox,Comiskey Park,1644460,100,98,CHW,CHA,CHA
+1960,AL,CLE,CLE,,4,154,77,76,78,,,N,N,667,5296,1415,218,20,127,444,573,58,25,,,693,607,3.9500000000,32,10,30,4146,1308,161,636,771,126,165,0.97,Cleveland Indians,Cleveland Stadium,950985,97,97,CLE,CLE,CLE
+1960,AL,DET,DET,,6,154,77,71,83,,,N,N,633,5202,1243,188,34,150,636,728,66,32,,,644,568,3.6400000000,40,7,25,4215,1336,141,474,824,137,138,0.97,Detroit Tigers,Briggs Stadium,1167669,103,103,DET,DET,DET
+1960,AL,KC1,OAK,,8,155,78,58,96,,,N,N,615,5226,1303,212,34,110,513,744,16,11,,,756,669,4.3800000000,44,4,14,4122,1428,160,525,664,127,149,0.97,Kansas City Athletics,Municipal Stadium I,774944,100,103,KCA,KC1,KC1
+1960,AL,NYA,NYY,,1,155,77,97,57,,,Y,N,746,5290,1377,215,40,193,537,818,37,23,,,627,547,3.5200000000,38,16,42,4194,1225,123,609,712,129,162,0.97,New York Yankees,Yankee Stadium I,1627349,94,92,NYY,NYA,NYA
+1960,AL,WS1,MIN,,5,154,77,73,81,,,N,N,672,5248,1283,205,43,147,584,883,52,43,,,696,589,3.7700000000,34,10,35,4215,1392,130,538,775,165,159,0.97,Washington Senators,Griffith Stadium II,743404,101,102,WSH,WS1,WS1
+1960,NL,CHN,CHC,,7,156,79,60,94,,,N,N,634,5311,1293,213,48,119,531,897,51,34,,,776,678,4.3500000000,36,6,25,4206,1393,152,565,805,143,133,0.97,Chicago Cubs,Wrigley Field,809770,99,100,CHC,CHN,CHN
+1960,NL,CIN,CIN,,6,154,77,67,87,,,N,N,640,5289,1324,230,40,140,512,858,73,37,,,692,618,4.0000000000,33,8,35,4170,1417,134,442,740,125,155,0.97,Cincinnati Reds,Crosley Field,663486,102,102,CIN,CIN,CIN
+1960,NL,LAN,LAD,,4,154,77,82,72,,,N,N,662,5227,1333,216,38,126,529,837,95,53,,,593,528,3.4000000000,46,13,20,4194,1218,154,564,1122,124,142,0.97,Los Angeles Dodgers,Los Angeles Memorial Coliseum,2253887,106,105,LAD,LAN,LAN
+1960,NL,ML1,ATL,,2,154,77,88,66,,,N,N,724,5263,1393,198,48,170,463,793,69,37,,,658,579,3.7600000000,55,13,28,4161,1327,130,518,807,139,137,0.97,Milwaukee Braves,County Stadium,1497799,93,91,MLN,MLN,MLN
+1960,NL,PHI,PHI,,8,154,77,59,95,,,N,N,546,5169,1235,196,44,99,448,1054,45,48,,,691,613,4.0100000000,45,6,16,4125,1423,133,439,736,155,129,0.97,Philadelphia Phillies,Connie Mack Stadium,862205,100,103,PHI,PHI,PHI
+1960,NL,PIT,PIT,,1,155,78,95,59,,,Y,Y,734,5406,1493,236,56,120,486,747,34,24,,,593,543,3.4900000000,47,11,33,4197,1363,105,386,811,128,163,0.97,Pittsburgh Pirates,Forbes Field,1705828,101,100,PIT,PIT,PIT
+1960,NL,SFN,SFG,,5,156,77,79,75,,,N,N,671,5324,1357,220,62,130,467,846,86,45,,,631,534,3.4400000000,55,16,26,4188,1288,107,512,897,165,117,0.97,San Francisco Giants,Candlestick Park,1795356,95,93,SFG,SFN,SFN
+1960,NL,SLN,STL,,3,155,77,86,68,,,N,N,639,5187,1317,213,48,138,501,792,48,35,,,616,554,3.6400000000,37,11,30,4113,1316,127,511,906,141,152,0.97,St. Louis Cardinals,Sportsman's Park IV,1096632,108,109,STL,SLN,SLN
+1961,AL,BAL,BAL,,3,163,82,95,67,,,N,N,691,5481,1393,227,36,149,581,902,39,30,,,588,526,3.2200000000,54,21,33,4413,1226,109,617,926,126,173,0.98,Baltimore Orioles,Memorial Stadium,951089,96,96,BAL,BAL,BAL
+1961,AL,BOS,BOS,,6,163,82,76,86,,,N,N,729,5508,1401,251,37,112,647,847,56,36,,,792,687,4.2900000000,35,6,30,4326,1472,167,679,831,143,140,0.97,Boston Red Sox,Fenway Park II,850589,102,103,BOS,BOS,BOS
+1961,AL,CHA,CHW,,4,163,81,86,76,,,N,N,765,5556,1475,216,46,138,550,612,100,40,,,726,653,4.0600000000,39,3,33,4344,1491,158,498,814,128,138,0.98,Chicago White Sox,Comiskey Park,1146019,99,97,CHW,CHA,CHA
+1961,AL,CLE,CLE,,5,161,81,78,83,,,N,N,737,5609,1493,257,39,150,492,720,34,11,,,752,665,4.1500000000,35,12,23,4329,1426,178,599,801,139,142,0.97,Cleveland Indians,Cleveland Stadium,725547,97,98,CLE,CLE,CLE
+1961,AL,DET,DET,,2,163,82,101,61,,,N,N,841,5561,1481,215,53,180,673,867,98,36,,,671,575,3.5500000000,62,12,30,4377,1404,170,469,836,146,147,0.97,Detroit Tigers,Tiger Stadium,1600710,103,102,DET,DET,DET
+1961,AL,KC1,OAK,,9,162,80,61,100,,,N,N,683,5423,1342,216,47,90,580,772,58,22,,,863,745,4.7400000000,32,5,23,4245,1519,141,629,703,174,160,0.97,Kansas City Athletics,Municipal Stadium I,683817,101,103,KCA,KC1,KC1
+1961,AL,LAA,ANA,,8,162,82,70,91,,,N,N,744,5424,1331,218,22,189,681,1068,37,28,,,784,689,4.3100000000,25,5,34,4314,1391,180,713,973,192,154,0.96,Los Angeles Angels,Wrigley Field (LA),603510,111,112,LAA,LAA,LAA
+1961,AL,MIN,MIN,,7,161,81,70,90,,,N,N,707,5417,1353,215,40,167,597,840,47,43,,,778,681,4.2800000000,49,14,23,4296,1415,163,570,914,174,150,0.97,Minnesota Twins,Metropolitan Stadium,1256723,106,106,MIN,MIN,MIN
+1961,AL,NYA,NYY,,1,163,81,109,53,,,Y,Y,827,5559,1461,194,40,240,543,785,28,18,,,612,558,3.4600000000,47,14,39,4353,1288,137,542,866,124,180,0.98,New York Yankees,Yankee Stadium I,1747725,95,93,NYY,NYA,NYA
+1961,AL,WS2,TEX,,9,161,79,61,100,,,N,N,618,5366,1307,217,44,119,558,917,81,47,,,776,670,4.2300000000,39,8,21,4275,1405,131,586,666,156,171,0.97,Washington Senators,Griffith Stadium II,597287,95,97,WSA,WS2,WS2
+1961,NL,CHN,CHC,,7,156,78,64,90,,,N,N,689,5344,1364,238,51,176,539,1027,35,25,,,800,689,4.4800000000,34,6,25,4155,1492,165,465,755,183,175,0.97,Chicago Cubs,Wrigley Field,673057,101,104,CHC,CHN,CHN
+1961,NL,CIN,CIN,,1,154,77,93,61,,,Y,N,710,5243,1414,247,35,158,423,761,70,33,,,653,575,3.7800000000,46,12,40,4110,1300,147,500,829,134,124,0.97,Cincinnati Reds,Crosley Field,1117603,102,101,CIN,CIN,CIN
+1961,NL,LAN,LAD,,2,154,77,89,65,,,N,N,735,5189,1358,193,40,157,596,796,86,45,,,697,619,4.0400000000,40,10,35,4134,1346,167,544,1105,136,162,0.97,Los Angeles Dodgers,Los Angeles Memorial Coliseum,1804250,108,107,LAD,LAN,LAN
+1961,NL,ML1,ATL,,4,155,77,83,71,,,N,N,712,5288,1365,199,34,188,534,880,70,43,,,656,601,3.8900000000,57,8,16,4173,1357,153,493,652,111,152,0.98,Milwaukee Braves,County Stadium,1101441,94,92,MLN,MLN,MLN
+1961,NL,PHI,PHI,,8,155,78,47,107,,,N,N,584,5213,1265,185,50,103,475,928,56,30,,,796,708,4.6100000000,29,9,13,4149,1452,155,521,775,144,179,0.97,Philadelphia Phillies,Connie Mack Stadium,590039,98,101,PHI,PHI,PHI
+1961,NL,PIT,PIT,,6,154,77,75,79,,,N,N,694,5311,1448,232,57,128,428,721,26,30,,,675,593,3.9200000000,34,9,29,4086,1442,121,400,759,150,189,0.97,Pittsburgh Pirates,Forbes Field,1199128,101,99,PIT,PIT,PIT
+1961,NL,SFN,SFG,,3,155,77,85,69,,,N,N,773,5233,1379,219,32,183,506,764,79,54,,,655,581,3.7700000000,39,9,30,4164,1306,152,502,924,133,126,0.97,San Francisco Giants,Candlestick Park,1390679,97,95,SFG,SFN,SFN
+1961,NL,SLN,STL,,5,155,78,80,74,,,N,N,703,5307,1436,236,51,103,494,745,46,28,,,668,568,3.7400000000,49,10,24,4104,1334,136,570,823,166,165,0.97,St. Louis Cardinals,Sportsman's Park IV,855305,110,110,STL,SLN,SLN
+1962,AL,BAL,BAL,,7,162,82,77,85,,,N,N,652,5491,1363,225,34,156,516,931,45,32,,,680,599,3.6900000000,32,8,33,4386,1373,147,549,898,122,152,0.98,Baltimore Orioles,Memorial Stadium,790254,94,93,BAL,BAL,BAL
+1962,AL,BOS,BOS,,8,160,79,76,84,,,N,N,707,5530,1429,257,53,146,525,923,39,33,,,756,674,4.2200000000,34,12,40,4311,1416,159,632,923,128,152,0.97,Boston Red Sox,Fenway Park II,733080,103,104,BOS,BOS,BOS
+1962,AL,CHA,CHW,,5,162,81,85,77,,,N,N,707,5514,1415,250,56,92,620,674,76,40,,,658,601,3.7300000000,50,13,28,4353,1380,123,537,821,110,153,0.98,Chicago White Sox,Comiskey Park,1131562,100,99,CHW,CHA,CHA
+1962,AL,CLE,CLE,,6,162,81,80,82,,,N,N,682,5484,1341,202,22,180,502,939,35,16,,,745,663,4.1400000000,45,12,31,4323,1410,174,594,780,139,168,0.97,Cleveland Indians,Cleveland Stadium,716076,97,97,CLE,CLE,CLE
+1962,AL,DET,DET,,4,161,82,85,76,,,N,N,758,5456,1352,191,36,209,651,894,69,21,,,692,611,3.8100000000,46,8,35,4329,1452,169,503,873,156,114,0.97,Detroit Tigers,Tiger Stadium,1207881,103,102,DET,DET,DET
+1962,AL,KC1,OAK,,9,162,81,72,90,,,N,N,745,5576,1467,220,58,116,556,803,76,21,,,837,763,4.7900000000,32,4,33,4302,1450,199,655,825,132,131,0.97,Kansas City Athletics,Municipal Stadium I,635675,104,106,KCA,KC1,KC1
+1962,AL,LAA,ANA,,3,162,81,86,76,,,N,N,718,5499,1377,232,35,137,602,917,46,27,,,706,603,3.7000000000,23,15,47,4398,1412,118,616,858,175,153,0.97,Los Angeles Angels,Dodger Stadium,1144063,97,97,LAA,LAA,LAA
+1962,AL,MIN,MIN,,2,163,82,91,71,,,N,N,798,5561,1445,215,39,185,649,823,33,20,,,713,632,3.8900000000,53,11,27,4389,1400,166,493,948,129,173,0.97,Minnesota Twins,Metropolitan Stadium,1433116,104,103,MIN,MIN,MIN
+1962,AL,NYA,NYY,,1,162,80,96,66,,,Y,Y,817,5644,1509,240,29,199,584,842,42,29,,,680,604,3.7000000000,33,10,42,4410,1375,146,499,838,131,151,0.97,New York Yankees,Yankee Stadium I,1493574,97,95,NYY,NYA,NYA
+1962,AL,WS2,TEX,,10,162,80,60,101,,,N,N,599,5484,1370,206,38,132,466,789,99,53,,,716,649,4.0400000000,38,11,13,4335,1400,151,593,771,139,160,0.97,Washington Senators,R.F.K. Stadium,729775,100,102,WSA,WS2,WS2
+1962,NL,CHN,CHC,,9,162,81,59,103,,,N,N,632,5534,1398,196,56,126,504,1044,78,50,,,827,725,4.5400000000,29,4,26,4314,1509,159,601,783,146,171,0.97,Chicago Cubs,Wrigley Field,609802,104,105,CHC,CHN,CHN
+1962,NL,CIN,CIN,,3,162,81,98,64,,,N,N,802,5645,1523,252,40,167,498,903,66,39,,,685,608,3.7500000000,51,13,35,4380,1397,149,567,964,143,144,0.97,Cincinnati Reds,Crosley Field,982095,104,102,CIN,CIN,CIN
+1962,NL,HOU,HOU,,8,162,82,64,96,,,N,N,592,5558,1370,170,47,105,493,806,42,30,,,717,618,3.8300000000,34,9,19,4359,1446,113,471,1047,173,149,0.97,Houston Colt .45's,Colt Stadium,924456,93,95,HOU,HOU,HOU
+1962,NL,LAN,LAD,,2,165,83,102,63,,,N,N,842,5628,1510,192,65,140,572,886,198,43,,,697,599,3.6200000000,44,8,46,4464,1386,115,588,1104,191,144,0.97,Los Angeles Dodgers,Dodger Stadium,2755184,93,91,LAD,LAN,LAN
+1962,NL,ML1,ATL,,5,162,81,86,76,,,N,N,730,5458,1376,204,38,181,581,975,57,27,,,665,586,3.6800000000,59,10,24,4302,1443,151,407,802,123,154,0.98,Milwaukee Braves,County Stadium,766921,97,96,MLN,MLN,MLN
+1962,NL,NYN,NYM,,10,161,80,40,120,,,N,N,617,5492,1318,166,40,139,616,991,59,48,,,948,801,5.0400000000,43,4,10,4290,1577,192,571,772,210,167,0.96,New York Mets,Polo Grounds IV,922530,100,105,NYM,NYN,NYN
+1962,NL,PHI,PHI,,7,161,80,81,80,,,N,N,705,5420,1410,199,39,142,531,923,79,42,,,759,678,4.2800000000,43,7,24,4278,1469,155,574,863,138,167,0.97,Philadelphia Phillies,Connie Mack Stadium,762034,97,98,PHI,PHI,PHI
+1962,NL,PIT,PIT,,4,161,81,93,68,,,N,N,706,5483,1468,240,65,108,432,836,50,39,,,626,536,3.3700000000,40,13,41,4296,1433,118,466,897,152,177,0.97,Pittsburgh Pirates,Forbes Field,1090648,101,100,PIT,PIT,PIT
+1962,NL,SFN,SFG,,1,165,82,103,62,,,Y,N,878,5588,1552,235,32,204,523,822,73,50,,,690,615,3.7900000000,62,10,39,4383,1399,148,503,886,142,153,0.97,San Francisco Giants,Candlestick Park,1592594,99,96,SFG,SFN,SFN
+1962,NL,SLN,STL,,6,163,81,84,78,,,N,N,774,5643,1528,221,31,137,515,846,86,41,,,664,577,3.5500000000,53,17,25,4389,1394,149,517,914,132,170,0.97,St. Louis Cardinals,Sportsman's Park IV,953895,111,109,STL,SLN,SLN
+1963,AL,BAL,BAL,,4,162,81,86,76,,,N,N,644,5448,1359,207,32,146,469,940,97,34,,,621,557,3.4500000000,35,8,43,4356,1353,137,507,913,99,157,0.98,Baltimore Orioles,Memorial Stadium,774343,96,95,BAL,BAL,BAL
+1963,AL,BOS,BOS,,7,161,80,76,85,,,N,N,666,5575,1403,247,34,171,475,954,27,16,,,704,639,3.9700000000,29,7,32,4347,1367,152,539,1009,132,119,0.97,Boston Red Sox,Fenway Park II,942642,103,104,BOS,BOS,BOS
+1963,AL,CHA,CHW,,2,162,82,94,68,,,N,N,683,5508,1379,208,40,114,571,896,64,28,,,544,485,2.9700000000,49,21,39,4407,1311,100,440,932,130,163,0.97,Chicago White Sox,Comiskey Park,1158848,98,96,CHW,CHA,CHA
+1963,AL,CLE,CLE,,5,162,81,79,83,,,N,N,635,5496,1314,214,29,169,469,1102,59,36,,,702,619,3.7900000000,40,14,25,4407,1390,176,478,1018,143,129,0.97,Cleveland Indians,Cleveland Stadium,562507,100,100,CLE,CLE,CLE
+1963,AL,DET,DET,,5,162,81,79,83,,,N,N,700,5500,1388,195,36,148,592,908,73,32,,,703,631,3.9000000000,42,7,28,4368,1407,195,477,930,113,124,0.98,Detroit Tigers,Tiger Stadium,821952,103,103,DET,DET,DET
+1963,AL,KC1,OAK,,8,162,81,73,89,,,N,N,615,5495,1356,225,38,95,529,829,47,26,,,704,635,3.9200000000,35,11,29,4374,1417,156,540,887,124,131,0.98,Kansas City Athletics,Municipal Stadium I,762364,106,108,KCA,KC1,KC1
+1963,AL,LAA,ANA,,9,161,81,70,91,,,N,N,597,5506,1378,208,38,95,448,916,43,30,,,660,569,3.5200000000,30,13,31,4365,1317,120,578,889,163,155,0.97,Los Angeles Angels,Dodger Stadium,821015,94,94,LAA,LAA,LAA
+1963,AL,MIN,MIN,,3,161,81,91,70,,,N,N,767,5531,1408,223,35,225,547,912,32,14,,,602,527,3.2800000000,58,13,30,4338,1322,162,459,941,144,140,0.97,Minnesota Twins,Metropolitan Stadium,1406652,102,101,MIN,MIN,MIN
+1963,AL,NYA,NYY,,1,161,80,104,57,,,Y,N,714,5506,1387,197,35,188,434,808,42,26,,,547,494,3.0700000000,59,19,31,4347,1239,115,476,965,108,162,0.98,New York Yankees,Yankee Stadium I,1308920,100,97,NYY,NYA,NYA
+1963,AL,WS2,TEX,,10,162,80,56,106,,,N,N,578,5446,1237,190,35,138,497,963,68,28,,,812,711,4.4200000000,29,8,25,4341,1486,176,537,744,182,165,0.97,Washington Senators,R.F.K. Stadium,535604,100,103,WSA,WS2,WS2
+1963,NL,CHN,CHC,,7,162,81,82,80,,,N,N,570,5404,1286,205,44,127,439,1049,68,60,,,578,499,3.0800000000,45,15,28,4371,1357,119,400,851,155,172,0.97,Chicago Cubs,Wrigley Field,979551,106,107,CHC,CHN,CHN
+1963,NL,CIN,CIN,,5,162,81,86,76,,,N,N,648,5416,1333,225,44,122,474,960,92,58,,,594,526,3.2900000000,55,22,36,4317,1307,117,425,1048,135,127,0.97,Cincinnati Reds,Crosley Field,858805,103,102,CIN,CIN,CIN
+1963,NL,HOU,HOU,,9,162,81,66,96,,,N,N,464,5384,1184,170,39,62,456,938,39,30,,,640,554,3.4400000000,36,16,20,4350,1341,95,378,937,162,100,0.97,Houston Colt .45's,Colt Stadium,719502,94,96,HOU,HOU,HOU
+1963,NL,LAN,LAD,,1,163,81,99,63,,,Y,Y,640,5428,1361,178,34,110,453,867,124,70,,,550,465,2.8500000000,51,24,29,4407,1329,111,402,1095,158,129,0.97,Los Angeles Dodgers,Dodger Stadium,2538602,93,91,LAD,LAN,LAN
+1963,NL,ML1,ATL,,6,163,82,84,78,,,N,N,677,5518,1345,204,39,139,525,954,75,52,,,603,534,3.2700000000,56,18,25,4413,1327,149,489,924,129,161,0.98,Milwaukee Braves,County Stadium,773018,99,98,MLN,MLN,MLN
+1963,NL,NYN,NYM,,10,162,81,51,111,,,N,N,501,5336,1168,156,35,96,457,1078,41,52,,,774,653,4.1200000000,42,5,12,4281,1452,162,529,806,208,151,0.96,New York Mets,Polo Grounds IV,1080108,100,105,NYM,NYN,NYN
+1963,NL,PHI,PHI,,4,162,81,87,75,,,N,N,642,5524,1390,228,54,126,403,955,56,39,,,578,500,3.0900000000,45,12,31,4371,1262,113,553,1052,142,147,0.97,Philadelphia Phillies,Connie Mack Stadium,907141,99,99,PHI,PHI,PHI
+1963,NL,PIT,PIT,,8,162,81,74,88,,,N,N,567,5536,1385,181,49,108,454,940,57,41,,,595,499,3.1000000000,34,16,33,4344,1350,99,457,900,182,195,0.97,Pittsburgh Pirates,Forbes Field,783648,101,100,PIT,PIT,PIT
+1963,NL,SFN,SFG,,3,162,81,88,74,,,N,N,725,5579,1442,206,35,197,441,889,55,49,,,641,547,3.3500000000,46,9,30,4407,1380,126,464,954,156,113,0.97,San Francisco Giants,Candlestick Park,1571306,99,97,SFG,SFN,SFN
+1963,NL,SLN,STL,,2,162,81,93,69,,,N,N,747,5678,1540,231,66,128,458,915,77,42,,,628,540,3.3200000000,49,17,32,4389,1329,124,463,978,147,136,0.97,St. Louis Cardinals,Sportsman's Park IV,1170546,110,109,STL,SLN,SLN
+1964,AL,BAL,BAL,,3,163,82,97,65,,,N,N,679,5463,1357,229,20,162,537,1019,78,38,,,567,512,3.1600000000,44,17,41,4374,1292,129,456,939,95,159,0.98,Baltimore Orioles,Memorial Stadium,1116215,100,99,BAL,BAL,BAL
+1964,AL,BOS,BOS,,8,162,81,72,90,,,N,N,688,5513,1425,253,29,186,504,917,18,16,,,793,711,4.5000000000,21,9,38,4266,1464,178,571,1094,136,123,0.97,Boston Red Sox,Fenway Park II,883276,105,106,BOS,BOS,BOS
+1964,AL,CHA,CHW,,2,162,81,98,64,,,N,N,642,5491,1356,184,40,106,562,902,75,39,,,501,443,2.7200000000,44,20,45,4401,1216,124,401,955,122,164,0.98,Chicago White Sox,Comiskey Park,1250053,97,95,CHW,CHA,CHA
+1964,AL,CLE,CLE,,6,164,82,79,83,,,N,N,689,5603,1386,208,22,164,500,1063,79,51,,,693,620,3.7500000000,37,16,37,4461,1443,154,565,1162,118,149,0.98,Cleveland Indians,Cleveland Stadium,653293,99,99,CLE,CLE,CLE
+1964,AL,DET,DET,,4,163,82,85,77,,,N,N,699,5513,1394,199,57,157,517,912,60,27,,,678,620,3.8400000000,35,11,35,4359,1343,164,536,993,109,137,0.98,Detroit Tigers,Tiger Stadium,816139,102,101,DET,DET,DET
+1964,AL,KC1,OAK,,10,163,81,57,105,,,N,N,621,5524,1321,216,29,166,548,1104,34,20,,,836,761,4.7100000000,18,6,27,4365,1516,220,614,966,154,152,0.97,Kansas City Athletics,Municipal Stadium I,642478,104,106,KCA,KC1,KC1
+1964,AL,LAA,ANA,,5,162,81,82,80,,,N,N,544,5362,1297,186,27,102,472,920,49,39,,,551,469,2.9100000000,30,28,41,4350,1273,100,530,965,138,168,0.97,Los Angeles Angels,Dodger Stadium,760439,90,90,LAA,LAA,LAA
+1964,AL,MIN,MIN,,6,163,82,79,83,,,N,N,737,5610,1413,227,46,221,553,1019,46,22,,,678,588,3.5800000000,47,4,29,4431,1361,181,545,1099,145,131,0.97,Minnesota Twins,Metropolitan Stadium,1207514,101,99,MIN,MIN,MIN
+1964,AL,NYA,NYY,,1,164,81,99,63,,,Y,N,730,5705,1442,208,35,162,520,976,54,18,,,577,527,3.1500000000,46,18,45,4518,1312,129,504,989,107,158,0.98,New York Yankees,Yankee Stadium I,1305638,101,100,NYY,NYA,NYA
+1964,AL,WS2,TEX,,9,162,81,62,100,,,N,N,578,5396,1246,199,28,125,514,1124,47,30,,,733,635,3.9800000000,27,5,26,4305,1417,172,505,794,127,145,0.97,Washington Senators,R.F.K. Stadium,600106,100,102,WSA,WS2,WS2
+1964,NL,CHN,CHC,,8,162,81,76,86,,,N,N,649,5545,1391,239,50,145,499,1041,70,49,,,724,655,4.0800000000,58,11,19,4335,1510,144,423,737,162,147,0.97,Chicago Cubs,Wrigley Field,751647,104,105,CHC,CHN,CHN
+1964,NL,CIN,CIN,,2,163,82,92,70,,,N,N,660,5561,1383,220,38,130,457,974,90,36,,,566,500,3.0700000000,54,14,35,4401,1306,112,436,1122,130,137,0.97,Cincinnati Reds,Crosley Field,862466,103,102,CIN,CIN,CIN
+1964,NL,HOU,HOU,,9,162,81,66,96,,,N,N,495,5303,1214,162,41,70,381,872,40,48,,,628,541,3.4100000000,30,9,31,4284,1421,105,353,852,149,124,0.97,Houston Colt .45's,Colt Stadium,725773,95,97,HOU,HOU,HOU
+1964,NL,LAN,LAD,,6,164,81,80,82,,,N,N,614,5499,1375,180,39,79,438,893,141,60,,,572,486,2.9500000000,47,19,27,4449,1289,88,458,1062,170,126,0.97,Los Angeles Dodgers,Dodger Stadium,2228751,93,92,LAD,LAN,LAN
+1964,NL,ML1,ATL,,5,162,81,88,74,,,N,N,803,5591,1522,274,32,159,486,825,53,41,,,744,656,4.1200000000,45,14,39,4302,1411,160,452,906,143,139,0.97,Milwaukee Braves,County Stadium,910911,101,100,MLN,MLN,MLN
+1964,NL,NYN,NYM,,10,163,82,53,109,,,N,N,569,5566,1372,195,31,103,353,932,36,31,,,776,679,4.2500000000,40,10,15,4314,1511,130,466,717,166,154,0.97,New York Mets,Shea Stadium,1732597,97,100,NYM,NYN,NYN
+1964,NL,PHI,PHI,,2,162,81,92,70,,,N,N,693,5493,1415,241,51,130,440,924,30,35,,,632,545,3.3600000000,37,17,41,4383,1402,129,440,1009,156,150,0.97,Philadelphia Phillies,Connie Mack Stadium,1425891,99,98,PHI,PHI,PHI
+1964,NL,PIT,PIT,,6,162,81,80,82,,,N,N,663,5566,1469,225,54,121,408,970,39,33,,,636,564,3.5200000000,42,14,29,4329,1429,92,476,951,177,179,0.97,Pittsburgh Pirates,Forbes Field,759496,100,99,PIT,PIT,PIT
+1964,NL,SFN,SFG,,4,162,81,90,72,,,N,N,656,5535,1360,185,38,165,505,900,64,35,,,587,523,3.1900000000,48,17,30,4428,1348,118,480,1023,159,136,0.97,San Francisco Giants,Candlestick Park,1504364,102,101,SFG,SFN,SFN
+1964,NL,SLN,STL,,1,162,81,93,69,,,Y,Y,715,5625,1531,240,53,109,427,925,73,51,,,652,551,3.4300000000,47,10,38,4335,1405,133,410,877,172,147,0.97,St. Louis Cardinals,Sportsman's Park IV,1143294,109,108,STL,SLN,SLN
+1965,AL,BAL,BAL,,3,162,79,94,68,,,N,N,641,5450,1299,227,38,125,529,907,67,31,,,578,489,2.9800000000,32,15,41,4431,1268,120,510,939,126,152,0.98,Baltimore Orioles,Memorial Stadium,781649,102,101,BAL,BAL,BAL
+1965,AL,BOS,BOS,,9,162,81,62,100,,,N,N,669,5487,1378,244,40,165,607,964,47,24,,,791,678,4.2400000000,33,9,25,4317,1443,158,543,993,159,129,0.97,Boston Red Sox,Fenway Park II,652201,106,108,BOS,BOS,BOS
+1965,AL,CAL,ANA,,7,162,80,75,87,,,N,N,527,5354,1279,200,36,92,443,973,107,59,,,569,508,3.1700000000,39,14,33,4323,1259,91,563,847,123,149,0.98,California Angels,Dodger Stadium,566727,97,98,CAL,CAL,CAL
+1965,AL,CHA,CHW,,2,162,81,95,67,,,N,N,647,5509,1354,200,38,125,533,916,50,33,,,555,492,2.9900000000,21,14,53,4443,1261,122,460,946,126,156,0.98,Chicago White Sox,Comiskey Park,1130519,93,92,CHW,CHA,CHA
+1965,AL,CLE,CLE,,5,162,82,87,75,,,N,N,663,5469,1367,198,21,156,506,857,109,46,,,613,535,3.3000000000,41,13,41,4374,1254,129,500,1156,114,127,0.98,Cleveland Indians,Cleveland Stadium,934786,101,101,CLE,CLE,CLE
+1965,AL,DET,DET,,4,162,81,89,73,,,N,N,680,5368,1278,190,27,162,554,952,57,41,,,602,542,3.3500000000,45,14,31,4365,1283,137,509,1069,114,126,0.98,Detroit Tigers,Tiger Stadium,1029645,102,101,DET,DET,DET
+1965,AL,KC1,OAK,,10,162,81,59,103,,,N,N,585,5393,1294,186,59,110,521,1020,110,51,,,755,675,4.2400000000,18,7,32,4299,1399,161,574,882,139,142,0.97,Kansas City Athletics,Municipal Stadium I,528344,99,101,KCA,KC1,KC1
+1965,AL,MIN,MIN,,1,162,81,102,60,,,Y,N,774,5488,1396,257,42,150,554,969,92,33,,,600,508,3.1400000000,32,12,45,4371,1278,166,503,934,172,158,0.97,Minnesota Twins,Metropolitan Stadium,1463258,105,103,MIN,MIN,MIN
+1965,AL,NYA,NYY,,6,162,83,77,85,,,N,N,611,5470,1286,196,31,149,489,951,35,20,,,604,532,3.2800000000,41,11,31,4377,1337,126,511,1001,137,166,0.97,New York Yankees,Yankee Stadium I,1213552,99,98,NYY,NYA,NYA
+1965,AL,WS2,TEX,,8,162,81,70,92,,,N,N,591,5374,1227,179,33,136,570,1125,30,19,,,721,627,3.9300000000,21,8,40,4305,1376,160,633,867,143,148,0.97,Washington Senators,R.F.K. Stadium,560083,98,100,WSA,WS2,WS2
+1965,NL,CHN,CHC,,8,164,83,72,90,,,N,N,635,5540,1316,202,33,134,532,948,65,47,,,723,618,3.7800000000,33,9,35,4416,1470,154,481,855,171,166,0.97,Chicago Cubs,Wrigley Field,641361,102,104,CHC,CHN,CHN
+1965,NL,CIN,CIN,,4,162,81,89,73,,,N,N,825,5658,1544,268,61,183,538,1003,82,40,,,704,628,3.8800000000,43,9,34,4371,1355,136,587,1113,117,142,0.98,Cincinnati Reds,Crosley Field,1047824,107,106,CIN,CIN,CIN
+1965,NL,HOU,HOU,,9,162,81,65,97,,,N,N,569,5483,1299,188,42,97,502,877,90,37,,,711,623,3.8400000000,29,7,26,4383,1459,123,388,931,166,130,0.97,Houston Astros,Astrodome,2151470,92,94,HOU,HOU,HOU
+1965,NL,LAN,LAD,,1,162,81,97,65,,,Y,Y,608,5425,1329,193,32,78,492,891,172,77,,,521,461,2.8100000000,58,23,34,4428,1223,127,425,1079,134,135,0.97,Los Angeles Dodgers,Dodger Stadium,2553577,93,92,LAD,LAN,LAN
+1965,NL,ML1,ATL,,5,162,81,86,76,,,N,N,708,5542,1419,243,28,196,408,976,64,37,,,633,566,3.5200000000,43,4,38,4341,1336,123,541,966,140,145,0.97,Milwaukee Braves,County Stadium,555584,101,99,MLN,MLN,MLN
+1965,NL,NYN,NYM,,10,164,82,50,112,,,N,N,495,5441,1202,203,27,107,392,1129,28,42,,,752,656,4.0600000000,29,11,14,4362,1462,147,498,776,169,153,0.97,New York Mets,Shea Stadium,1768389,96,99,NYM,NYN,NYN
+1965,NL,PHI,PHI,,6,162,80,85,76,,,N,N,654,5528,1380,205,53,144,494,1091,46,32,,,667,576,3.5300000000,50,18,21,4404,1426,116,466,1071,157,153,0.97,Philadelphia Phillies,Connie Mack Stadium,1166376,98,98,PHI,PHI,PHI
+1965,NL,PIT,PIT,,3,163,82,90,72,,,N,N,675,5686,1506,217,57,111,419,1008,51,38,,,580,495,3.0100000000,49,17,27,4437,1324,89,469,882,152,189,0.97,Pittsburgh Pirates,Forbes Field,909279,101,99,PIT,PIT,PIT
+1965,NL,SFN,SFG,,2,163,81,95,67,,,N,N,682,5495,1384,169,43,159,476,844,47,27,,,593,521,3.2000000000,42,17,42,4395,1325,137,408,1060,148,124,0.97,San Francisco Giants,Candlestick Park,1546075,104,102,SFG,SFN,SFN
+1965,NL,SLN,STL,,7,162,81,80,81,,,N,N,707,5579,1415,234,46,109,477,882,100,52,,,674,612,3.7700000000,40,11,35,4383,1414,166,467,916,130,152,0.97,St. Louis Cardinals,Sportsman's Park IV,1241201,110,109,STL,SLN,SLN
+1966,AL,BAL,BAL,,1,160,79,97,63,,,Y,Y,755,5529,1426,243,35,175,514,926,55,43,,,601,541,3.3200000000,23,13,51,4398,1267,127,514,1070,115,142,0.98,Baltimore Orioles,Memorial Stadium,1203366,99,98,BAL,BAL,BAL
+1966,AL,BOS,BOS,,9,162,81,72,90,,,N,N,655,5498,1318,228,44,145,542,1020,35,24,,,731,637,3.9200000000,32,10,31,4389,1402,164,577,977,153,153,0.97,Boston Red Sox,Fenway Park II,811172,111,112,BOS,BOS,BOS
+1966,AL,CAL,ANA,,6,162,81,80,82,,,N,N,604,5360,1244,179,54,122,525,1062,80,54,,,643,576,3.5600000000,31,12,40,4371,1364,136,511,836,136,186,0.97,California Angels,Anaheim Stadium,1400321,97,97,CAL,CAL,CAL
+1966,AL,CHA,CHW,,4,163,81,83,79,,,N,N,574,5348,1235,193,40,87,476,872,153,78,,,517,439,2.6800000000,38,22,34,4425,1229,101,403,896,160,149,0.97,Chicago White Sox,Comiskey Park,990016,93,92,CHW,CHA,CHA
+1966,AL,CLE,CLE,,5,162,81,81,81,,,N,N,574,5474,1300,156,25,155,450,914,53,41,,,586,526,3.2300000000,49,15,28,4401,1260,129,489,1111,138,132,0.97,Cleveland Indians,Cleveland Stadium,903359,100,100,CLE,CLE,CLE
+1966,AL,DET,DET,,3,162,81,88,74,,,N,N,719,5507,1383,224,45,179,551,987,41,34,,,698,622,3.8500000000,36,11,38,4362,1356,185,520,1026,117,142,0.98,Detroit Tigers,Tiger Stadium,1124293,103,102,DET,DET,DET
+1966,AL,KC1,OAK,,7,160,81,74,86,,,N,N,564,5328,1259,212,56,70,421,982,132,50,,,648,568,3.5600000000,19,11,47,4305,1281,106,630,854,139,154,0.97,Kansas City Athletics,Municipal Stadium I,773929,97,99,KCA,KC1,KC1
+1966,AL,MIN,MIN,,2,162,81,89,73,,,N,N,663,5390,1341,219,33,144,513,844,67,42,,,581,500,3.1300000000,52,11,28,4314,1246,139,392,1015,139,118,0.97,Minnesota Twins,Metropolitan Stadium,1259374,107,105,MIN,MIN,MIN
+1966,AL,NYA,NYY,,10,160,82,70,89,,,N,N,611,5330,1254,182,36,162,485,817,49,29,,,612,536,3.4100000000,29,7,32,4245,1318,124,443,842,142,142,0.97,New York Yankees,Yankee Stadium I,1124648,95,96,NYY,NYA,NYA
+1966,AL,WS2,TEX,,8,159,78,71,88,,,N,N,557,5318,1245,185,40,126,450,1069,53,37,,,659,583,3.7000000000,25,6,35,4257,1282,154,448,866,142,139,0.97,Washington Senators,R.F.K. Stadium,576260,99,101,WSA,WS2,WS2
+1966,NL,ATL,ATL,,5,163,82,85,77,,,N,N,782,5617,1476,220,32,207,512,913,59,47,,,683,601,3.6800000000,37,10,36,4407,1430,129,485,884,154,139,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1539801,103,102,ATL,ATL,ATL
+1966,NL,CHN,CHC,,10,162,81,59,103,,,N,N,644,5592,1418,203,43,140,457,998,76,47,,,809,701,4.3300000000,28,6,24,4374,1513,184,479,908,166,132,0.97,Chicago Cubs,Wrigley Field,635891,101,102,CHC,CHN,CHN
+1966,NL,CIN,CIN,,7,160,79,76,84,,,N,N,692,5521,1434,232,33,149,394,877,70,50,,,702,651,4.0800000000,28,10,35,4308,1408,153,490,1043,122,133,0.98,Cincinnati Reds,Crosley Field,742958,110,109,CIN,CIN,CIN
+1966,NL,HOU,HOU,,8,163,81,72,90,,,N,N,612,5511,1405,203,35,112,491,885,90,47,,,695,603,3.7600000000,34,13,26,4329,1468,130,391,929,174,126,0.97,Houston Astros,Astrodome,1872108,93,94,HOU,HOU,HOU
+1966,NL,LAN,LAD,,1,162,81,95,67,,,Y,N,606,5471,1399,201,27,108,430,830,94,64,,,490,424,2.6200000000,52,20,35,4374,1287,84,356,1084,133,128,0.97,Los Angeles Dodgers,Dodger Stadium,2617029,92,91,LAD,LAN,LAN
+1966,NL,NYN,NYM,,9,161,81,66,95,,,N,N,587,5371,1286,187,35,98,446,992,55,46,,,761,661,4.1700000000,37,9,22,4281,1497,166,521,773,159,171,0.97,New York Mets,Shea Stadium,1932693,97,100,NYM,NYN,NYN
+1966,NL,PHI,PHI,,4,162,81,87,75,,,N,N,696,5607,1448,224,49,117,510,969,56,42,,,640,579,3.5700000000,52,15,23,4377,1439,137,412,928,113,147,0.98,Philadelphia Phillies,Connie Mack Stadium,1108201,100,99,PHI,PHI,PHI
+1966,NL,PIT,PIT,,3,162,81,92,70,,,N,N,759,5676,1586,238,66,158,405,1011,64,60,,,641,572,3.5200000000,35,12,43,4389,1445,125,463,898,141,215,0.97,Pittsburgh Pirates,Forbes Field,1196618,100,99,PIT,PIT,PIT
+1966,NL,SFN,SFG,,2,161,81,93,68,,,N,N,675,5539,1373,195,31,181,414,860,29,30,,,626,531,3.2400000000,52,14,27,4428,1370,140,359,973,168,131,0.97,San Francisco Giants,Candlestick Park,1657192,104,103,SFG,SFN,SFN
+1966,NL,SLN,STL,,6,162,81,83,79,,,N,N,571,5480,1377,196,61,108,345,977,144,61,,,577,504,3.1100000000,47,19,32,4377,1345,130,448,892,145,166,0.97,St. Louis Cardinals,Sportsman's Park IV/Busch Stadium II,1712980,100,100,STL,SLN,SLN
+1967,AL,BAL,BAL,,6,161,77,76,85,,,N,N,654,5456,1312,215,44,138,531,1002,54,37,,,592,537,3.3200000000,29,17,36,4371,1218,116,566,1034,124,144,0.98,Baltimore Orioles,Memorial Stadium,955053,100,98,BAL,BAL,BAL
+1967,AL,BOS,BOS,,1,162,81,92,70,,,Y,N,722,5471,1394,216,39,158,522,1020,68,59,,,614,545,3.3600000000,41,9,44,4377,1307,142,477,1010,142,142,0.97,Boston Red Sox,Fenway Park II,1727832,109,109,BOS,BOS,BOS
+1967,AL,CAL,ANA,,5,161,83,84,77,,,N,N,567,5307,1265,170,37,114,453,1021,40,36,,,587,507,3.1900000000,19,14,46,4290,1246,118,525,892,111,135,0.98,California Angels,Anaheim Stadium,1317713,95,96,CAL,CAL,CAL
+1967,AL,CHA,CHW,,4,162,82,89,73,,,N,N,531,5383,1209,181,34,89,480,849,124,82,,,491,406,2.4500000000,36,24,39,4470,1197,87,465,927,137,149,0.97,Chicago White Sox,Comiskey Park,985634,93,93,CHW,CHA,CHA
+1967,AL,CLE,CLE,,8,162,81,75,87,,,N,N,559,5461,1282,213,35,131,413,984,53,65,,,613,533,3.2500000000,49,14,27,4431,1258,120,559,1189,115,138,0.98,Cleveland Indians,Cleveland Stadium,662980,102,102,CLE,CLE,CLE
+1967,AL,DET,DET,,2,163,82,91,71,,,N,N,683,5410,1315,192,36,152,626,994,37,21,,,587,532,3.3200000000,46,17,40,4329,1230,151,472,1038,130,126,0.97,Detroit Tigers,Tiger Stadium,1447143,103,101,DET,DET,DET
+1967,AL,KC1,OAK,,10,161,81,62,99,,,N,N,533,5349,1244,212,50,69,452,1019,132,59,,,660,584,3.6800000000,26,10,34,4284,1265,125,558,990,132,120,0.97,Kansas City Athletics,Municipal Stadium I,726639,97,99,KCA,KC1,KC1
+1967,AL,MIN,MIN,,2,164,81,91,71,,,N,N,671,5458,1309,216,48,131,512,976,55,37,,,590,510,3.1400000000,58,18,24,4383,1336,115,396,1089,132,123,0.97,Minnesota Twins,Metropolitan Stadium,1483547,108,107,MIN,MIN,MIN
+1967,AL,NYA,NYY,,9,163,82,72,90,,,N,N,522,5443,1225,166,17,100,532,1043,63,37,,,621,533,3.2400000000,37,16,27,4440,1375,110,480,898,154,144,0.97,New York Yankees,Yankee Stadium I,1259514,95,96,NYY,NYA,NYA
+1967,AL,WS2,TEX,,6,161,80,76,85,,,N,N,550,5441,1211,168,25,115,472,1037,53,37,,,637,553,3.3800000000,24,14,39,4419,1334,113,495,878,144,167,0.97,Washington Senators,R.F.K. Stadium,770868,96,97,WSA,WS2,WS2
+1967,NL,ATL,ATL,,7,162,81,77,85,,,N,N,631,5450,1307,191,29,158,512,947,55,45,,,640,561,3.4700000000,35,5,32,4362,1377,118,449,862,138,148,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1389222,99,99,ATL,ATL,ATL
+1967,NL,CHN,CHC,,3,162,84,87,74,,,N,N,702,5463,1373,211,49,128,509,912,63,50,,,624,563,3.4800000000,47,7,28,4371,1352,142,463,888,121,143,0.98,Chicago Cubs,Wrigley Field,977226,105,105,CHC,CHN,CHN
+1967,NL,CIN,CIN,,4,162,81,87,75,,,N,N,604,5519,1366,251,54,109,372,969,92,63,,,563,497,3.0500000000,34,18,39,4404,1328,101,498,1065,121,124,0.98,Cincinnati Reds,Crosley Field,958300,112,112,CIN,CIN,CIN
+1967,NL,HOU,HOU,,9,162,81,69,93,,,N,N,626,5506,1372,259,46,93,537,934,88,38,,,742,647,4.0300000000,35,8,21,4335,1444,120,485,1060,157,120,0.97,Houston Astros,Astrodome,1348303,96,97,HOU,HOU,HOU
+1967,NL,LAN,LAD,,8,162,81,73,89,,,N,N,519,5456,1285,203,38,82,485,881,56,47,,,595,525,3.2100000000,41,17,24,4419,1421,93,393,967,160,144,0.97,Los Angeles Dodgers,Dodger Stadium,1664362,91,91,LAD,LAN,LAN
+1967,NL,NYN,NYM,,10,162,78,61,101,,,N,N,498,5417,1288,178,23,83,362,981,58,44,,,672,594,3.7300000000,36,10,19,4299,1369,124,536,893,157,147,0.97,New York Mets,Shea Stadium,1565492,98,100,NYM,NYN,NYN
+1967,NL,PHI,PHI,,5,162,80,82,80,,,N,N,612,5401,1306,221,47,103,545,1033,79,62,,,581,500,3.1000000000,46,17,23,4359,1372,86,403,967,137,174,0.97,Philadelphia Phillies,Connie Mack Stadium,828888,101,101,PHI,PHI,PHI
+1967,NL,PIT,PIT,,6,163,81,81,81,,,N,N,679,5724,1585,193,62,91,387,914,79,37,,,693,606,3.7400000000,35,5,35,4374,1439,108,561,820,141,186,0.97,Pittsburgh Pirates,Forbes Field,907012,100,99,PIT,PIT,PIT
+1967,NL,SFN,SFG,,2,162,82,91,71,,,N,N,652,5524,1354,201,39,140,520,978,22,30,,,551,478,2.9200000000,64,17,25,4422,1283,113,453,990,134,149,0.97,San Francisco Giants,Candlestick Park,1242480,100,99,SFG,SFN,SFN
+1967,NL,SLN,STL,,1,161,81,101,60,,,Y,Y,695,5566,1462,225,40,115,443,919,102,54,,,557,496,3.0500000000,44,17,45,4395,1313,97,431,956,140,127,0.97,St. Louis Cardinals,Busch Stadium II,2090145,99,97,STL,SLN,SLN
+1968,AL,BAL,BAL,,2,162,80,91,71,,,N,N,579,5275,1187,215,28,133,570,1019,78,32,,,497,429,2.6600000000,53,16,31,4353,1111,101,502,1044,120,131,0.98,Baltimore Orioles,Memorial Stadium,943977,101,99,BAL,BAL,BAL
+1968,AL,BOS,BOS,,4,162,81,86,76,,,N,N,614,5303,1253,207,17,125,582,974,76,62,,,611,535,3.3300000000,55,17,31,4341,1303,115,523,972,128,147,0.97,Boston Red Sox,Fenway Park II,1940788,108,107,BOS,BOS,BOS
+1968,AL,CAL,ANA,,8,162,81,67,95,,,N,N,498,5331,1209,170,33,83,447,1080,62,50,,,615,548,3.4300000000,29,11,31,4311,1234,131,519,869,140,156,0.97,California Angels,Anaheim Stadium,1025956,95,97,CAL,CAL,CAL
+1968,AL,CHA,CHW,,8,162,81,67,95,,,N,N,463,5405,1233,169,33,71,397,840,90,50,,,527,449,2.7500000000,20,11,40,4404,1290,97,451,834,151,152,0.97,Chicago White Sox,Comiskey Park,803775,106,107,CHW,CHA,CHA
+1968,AL,CLE,CLE,,3,162,81,86,75,,,N,N,516,5416,1266,210,36,75,427,858,115,61,,,504,433,2.6600000000,48,23,32,4392,1087,98,540,1157,126,130,0.97,Cleveland Indians,Cleveland Stadium,857994,99,100,CLE,CLE,CLE
+1968,AL,DET,DET,,1,164,81,103,59,,,Y,Y,671,5490,1292,190,39,185,521,964,26,32,,,492,448,2.7100000000,59,19,29,4467,1180,129,486,1115,101,133,0.98,Detroit Tigers,Tiger Stadium,2031847,103,101,DET,DET,DET
+1968,AL,MIN,MIN,,7,162,81,79,83,,,N,N,562,5373,1274,207,41,105,445,966,98,54,,,546,460,2.8900000000,46,14,29,4299,1224,92,414,996,170,117,0.97,Minnesota Twins,Metropolitan Stadium,1143257,107,105,MIN,MIN,MIN
+1968,AL,NYA,NYY,,5,164,82,83,79,,,N,N,536,5310,1137,154,34,109,566,958,90,50,,,531,455,2.7900000000,45,14,27,4401,1308,99,424,831,139,142,0.97,New York Yankees,Yankee Stadium I,1185666,96,96,NYY,NYA,NYA
+1968,AL,OAK,OAK,,6,163,83,82,80,,,N,N,569,5406,1300,192,40,94,472,1022,147,61,,,544,475,2.9400000000,45,18,29,4365,1220,124,505,997,145,136,0.97,Oakland Athletics,Oakland Coliseum,837466,94,94,OAK,OAK,OAK
+1968,AL,WS2,TEX,,10,161,81,65,96,,,N,N,524,5400,1208,160,37,124,454,960,29,19,,,665,582,3.6400000000,26,11,28,4317,1402,118,517,826,148,144,0.97,Washington Senators,R.F.K. Stadium,546661,97,98,WSA,WS2,WS2
+1968,NL,ATL,ATL,,5,163,81,81,81,,,N,N,514,5552,1399,179,31,80,414,782,83,44,,,549,478,2.9200000000,44,16,29,4422,1326,87,362,871,125,139,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,1126540,101,101,ATL,ATL,ATL
+1968,NL,CHN,CHC,,3,163,82,84,78,,,N,N,612,5458,1319,203,43,130,415,854,41,30,,,611,551,3.4100000000,46,12,32,4359,1399,138,392,894,119,149,0.98,Chicago Cubs,Wrigley Field,1043409,107,106,CHC,CHN,CHN
+1968,NL,CIN,CIN,,4,163,82,83,79,,,N,N,690,5767,1573,281,36,106,379,938,59,55,,,673,589,3.5600000000,24,16,38,4470,1399,114,573,963,144,144,0.97,Cincinnati Reds,Crosley Field,733354,106,106,CIN,CIN,CIN
+1968,NL,HOU,HOU,,10,162,81,72,90,,,N,N,510,5336,1233,205,28,66,479,988,44,51,,,588,524,3.2600000000,50,12,23,4338,1362,68,479,1021,152,129,0.97,Houston Astros,Astrodome,1312887,97,99,HOU,HOU,HOU
+1968,NL,LAN,LAD,,7,162,81,76,86,,,N,N,470,5354,1234,202,36,67,439,980,57,43,,,509,433,2.6900000000,38,23,31,4344,1293,65,414,994,144,144,0.97,Los Angeles Dodgers,Dodger Stadium,1581093,92,92,LAD,LAN,LAN
+1968,NL,NYN,NYM,,9,163,82,73,89,,,N,N,473,5503,1252,178,30,81,379,1203,72,45,,,499,448,2.7200000000,45,25,32,4449,1250,87,430,1014,133,142,0.97,New York Mets,Shea Stadium,1781657,100,101,NYM,NYN,NYN
+1968,NL,PHI,PHI,,7,162,81,76,86,,,N,N,543,5372,1253,178,30,100,462,1003,58,51,,,615,541,3.3600000000,42,12,27,4344,1416,91,421,935,127,163,0.98,Philadelphia Phillies,Connie Mack Stadium,664546,100,101,PHI,PHI,PHI
+1968,NL,PIT,PIT,,6,163,81,80,82,,,N,N,583,5569,1404,180,44,80,422,953,130,59,,,532,453,2.7400000000,42,19,30,4461,1322,73,485,897,139,162,0.97,Pittsburgh Pirates,Forbes Field,693485,98,97,PIT,PIT,PIT
+1968,NL,SFN,SFG,,2,163,81,88,74,,,N,N,599,5441,1301,162,33,108,508,904,50,37,,,529,442,2.7100000000,77,20,16,4407,1302,86,344,942,162,125,0.97,San Francisco Giants,Candlestick Park,837220,101,100,SFG,SFN,SFN
+1968,NL,SLN,STL,,1,162,81,97,65,,,Y,N,583,5561,1383,227,48,73,378,897,110,45,,,472,409,2.4900000000,63,30,32,4437,1282,82,375,971,140,135,0.97,St. Louis Cardinals,Busch Stadium II,2011167,99,97,STL,SLN,SLN
+1969,AL,BAL,BAL,E,1,162,81,109,53,Y,,Y,N,779,5518,1465,234,29,175,634,806,82,45,,,517,463,2.8300000000,50,20,36,4419,1194,117,498,897,101,145,0.98,Baltimore Orioles,Memorial Stadium,1062069,101,99,BAL,BAL,BAL
+1969,AL,BOS,BOS,E,3,162,81,87,75,N,,N,N,743,5494,1381,234,37,197,658,923,41,47,,,736,639,3.9200000000,30,7,41,4398,1423,155,685,935,157,178,0.97,Boston Red Sox,Fenway Park II,1833246,107,106,BOS,BOS,BOS
+1969,AL,CAL,ANA,W,3,163,81,71,91,N,,N,N,528,5316,1221,151,29,88,516,929,54,39,,,652,566,3.5400000000,25,9,39,4314,1294,126,517,885,135,164,0.97,California Angels,Anaheim Stadium,758388,94,95,CAL,CAL,CAL
+1969,AL,CHA,CHW,W,5,162,81,68,94,N,,N,N,625,5450,1346,210,27,112,552,844,54,22,,,723,672,4.2100000000,29,10,25,4311,1470,146,564,810,122,163,0.98,Chicago White Sox,Comiskey Park,589546,106,107,CHW,CHA,CHA
+1969,AL,CLE,CLE,E,6,161,81,62,99,N,,N,N,573,5365,1272,173,24,119,535,906,85,37,,,717,629,3.9400000000,35,7,22,4311,1330,134,681,1000,142,153,0.97,Cleveland Indians,Cleveland Stadium,619970,103,103,CLE,CLE,CLE
+1969,AL,DET,DET,E,2,162,81,90,72,N,,N,N,701,5441,1316,188,29,182,578,922,35,28,,,601,535,3.3100000000,55,20,28,4365,1250,128,586,1032,126,130,0.97,Detroit Tigers,Tiger Stadium,1577481,105,104,DET,DET,DET
+1969,AL,KCA,KCR,W,4,163,82,69,93,N,,N,N,586,5462,1311,179,32,98,522,901,129,70,,,688,605,3.7200000000,42,10,25,4392,1357,136,560,894,156,114,0.97,Kansas City Royals,Municipal Stadium II,902414,100,101,KCR,KCA,KCA
+1969,AL,MIN,MIN,W,1,162,81,97,65,Y,,N,N,790,5677,1520,246,32,163,599,906,115,70,,,618,539,3.2400000000,41,8,43,4491,1388,119,524,906,150,177,0.97,Minnesota Twins,Metropolitan Stadium,1349328,104,102,MIN,MIN,MIN
+1969,AL,NYA,NYY,E,5,162,80,80,81,N,,N,N,562,5308,1247,210,44,94,565,840,119,74,,,587,517,3.2300000000,53,13,20,4320,1258,118,522,801,131,158,0.97,New York Yankees,Yankee Stadium I,1067996,96,96,NYY,NYA,NYA
+1969,AL,OAK,OAK,W,2,162,81,88,74,N,,N,N,740,5614,1400,210,28,148,617,953,100,39,,,678,610,3.7100000000,42,14,36,4440,1356,163,586,887,137,162,0.97,Oakland Athletics,Oakland Coliseum,778232,95,95,OAK,OAK,OAK
+1969,AL,SE1,MIL,W,6,163,82,64,98,N,,N,N,639,5444,1276,179,27,125,626,1015,167,59,,,799,707,4.3500000000,21,6,33,4389,1490,172,653,963,167,149,0.97,Seattle Pilots,Sicks Stadium,677944,97,100,SEP,SE1,SE1
+1969,AL,WS2,TEX,E,4,162,81,86,76,N,,N,N,694,5447,1365,171,40,148,630,900,52,40,,,644,561,3.4900000000,28,10,41,4341,1310,135,656,835,140,159,0.97,Washington Senators,R.F.K. Stadium,918106,94,95,WSA,WS2,WS2
+1969,NL,ATL,ATL,W,1,162,81,93,69,Y,,N,N,691,5460,1411,195,22,141,485,665,59,48,,,631,567,3.5300000000,38,7,42,4335,1334,144,438,893,115,114,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,1458320,101,101,ATL,ATL,ATL
+1969,NL,CHN,CHC,E,2,163,82,92,70,N,,N,N,720,5530,1400,215,40,142,559,928,30,32,,,611,540,3.3400000000,58,22,27,4362,1366,118,475,1017,136,149,0.97,Chicago Cubs,Wrigley Field,1674993,113,112,CHC,CHN,CHN
+1969,NL,CIN,CIN,W,3,163,81,89,73,N,,N,N,798,5634,1558,224,42,171,474,1042,79,56,,,768,669,4.1100000000,23,11,44,4395,1478,149,611,818,168,158,0.97,Cincinnati Reds,Crosley Field,987991,105,105,CIN,CIN,CIN
+1969,NL,HOU,HOU,W,5,162,81,81,81,N,,N,N,676,5348,1284,208,40,104,699,972,101,58,,,668,574,3.6000000000,52,11,34,4305,1347,111,547,1221,150,136,0.97,Houston Astros,Astrodome,1442995,98,98,HOU,HOU,HOU
+1969,NL,LAN,LAD,W,4,162,81,85,77,N,,N,N,645,5532,1405,185,52,97,484,823,80,51,,,561,499,3.0800000000,47,20,31,4371,1324,122,420,975,109,130,0.98,Los Angeles Dodgers,Dodger Stadium,1784527,93,92,LAD,LAN,LAN
+1969,NL,MON,WSN,E,6,162,81,52,110,N,,N,N,582,5419,1300,202,33,125,529,962,52,52,,,791,686,4.3300000000,26,8,21,4278,1429,145,702,973,184,179,0.97,Montreal Expos,Jarry Park,1212608,100,102,MON,MON,MON
+1969,NL,NYN,NYM,E,1,162,82,100,62,Y,,Y,Y,632,5427,1311,184,41,109,527,1089,66,43,,,541,488,2.9900000000,51,28,35,4404,1217,119,517,1012,122,146,0.98,New York Mets,Shea Stadium,2175373,102,101,NYM,NYN,NYN
+1969,NL,PHI,PHI,E,5,162,81,63,99,N,,N,N,645,5408,1304,227,35,137,549,1130,73,49,,,745,660,4.1400000000,47,14,21,4302,1494,134,570,921,136,157,0.97,Philadelphia Phillies,Connie Mack Stadium,519414,98,99,PHI,PHI,PHI
+1969,NL,PIT,PIT,E,3,162,81,88,74,N,,N,N,725,5626,1557,220,52,119,454,944,74,34,,,652,580,3.6100000000,39,9,33,4335,1348,96,553,1124,155,169,0.97,Pittsburgh Pirates,Forbes Field,769369,98,97,PIT,PIT,PIT
+1969,NL,SDN,SDP,W,6,162,81,52,110,N,,N,N,468,5357,1203,180,42,99,423,1143,45,44,,,746,670,4.2400000000,16,9,25,4266,1454,113,592,764,155,140,0.97,San Diego Padres,Jack Murphy Stadium,512970,96,98,SDP,SDN,SDN
+1969,NL,SFN,SFG,W,2,162,81,90,72,N,,N,N,713,5474,1325,187,28,136,711,1054,71,32,,,636,534,3.2600000000,71,15,17,4419,1381,120,461,906,169,155,0.97,San Francisco Giants,Candlestick Park,873603,99,98,SFG,SFN,SFN
+1969,NL,SLN,STL,E,4,162,80,87,75,N,,N,N,595,5536,1403,228,44,90,503,876,87,49,,,540,477,2.9400000000,63,12,26,4380,1289,99,511,1004,138,144,0.97,St. Louis Cardinals,Busch Stadium II,1682783,100,99,STL,SLN,SLN
+1970,AL,BAL,BAL,E,1,162,81,108,54,Y,,Y,Y,792,5545,1424,213,25,179,717,952,84,39,,,574,517,3.1500000000,60,12,31,4434,1317,139,469,941,117,148,0.98,Baltimore Orioles,Memorial Stadium,1057069,101,98,BAL,BAL,BAL
+1970,AL,BOS,BOS,E,3,162,81,87,75,N,,N,N,786,5535,1450,252,28,203,594,855,50,48,,,722,622,3.8700000000,38,8,44,4338,1391,156,594,1003,156,131,0.97,Boston Red Sox,Fenway Park II,1595278,108,107,BOS,BOS,BOS
+1970,AL,CAL,ANA,W,3,162,81,86,76,N,,N,N,631,5532,1391,197,40,114,447,922,69,27,,,630,565,3.4800000000,21,10,49,4386,1280,154,559,922,127,169,0.98,California Angels,Anaheim Stadium,1077741,96,97,CAL,CAL,CAL
+1970,AL,CHA,CHW,W,6,162,84,56,106,N,,N,N,633,5514,1394,192,20,123,477,872,53,33,,,822,721,4.5400000000,20,6,30,4290,1554,164,556,762,165,187,0.97,Chicago White Sox,Comiskey Park,495355,101,102,CHW,CHA,CHA
+1970,AL,CLE,CLE,E,5,162,81,76,86,N,,N,N,649,5463,1358,197,23,183,503,909,25,36,,,675,630,3.9100000000,34,8,35,4353,1333,163,689,1076,133,168,0.97,Cleveland Indians,Cleveland Stadium,729752,104,105,CLE,CLE,CLE
+1970,AL,DET,DET,E,4,162,81,79,83,N,,N,N,666,5377,1282,207,38,148,656,825,29,30,,,731,658,4.0900000000,33,9,39,4341,1443,153,623,1045,132,142,0.97,Detroit Tigers,Tiger Stadium,1501293,101,101,DET,DET,DET
+1970,AL,KCA,KCR,W,4,162,79,65,97,N,,N,N,611,5503,1341,202,41,97,514,958,97,53,,,705,614,3.7800000000,30,11,25,4389,1346,138,641,915,152,162,0.97,Kansas City Royals,Municipal Stadium II,693047,99,100,KCR,KCA,KCA
+1970,AL,MIN,MIN,W,1,162,81,98,64,Y,,N,N,744,5483,1438,230,41,153,501,905,57,52,,,605,520,3.2300000000,26,12,58,4344,1329,130,486,940,123,130,0.98,Minnesota Twins,Metropolitan Stadium,1261887,103,102,MIN,MIN,MIN
+1970,AL,ML4,MIL,W,4,163,81,65,97,N,,N,N,613,5395,1305,202,24,126,592,985,91,73,,,751,676,4.2100000000,31,2,27,4338,1397,146,587,895,136,142,0.97,Milwaukee Brewers,County Stadium,933690,100,101,MIL,MIL,MIL
+1970,AL,NYA,NYY,E,2,163,81,93,69,N,,N,N,680,5492,1381,208,41,111,588,808,105,61,,,612,530,3.2400000000,36,6,49,4413,1386,130,451,777,130,146,0.98,New York Yankees,Yankee Stadium I,1136879,95,95,NYY,NYA,NYA
+1970,AL,OAK,OAK,W,2,162,81,89,73,N,,N,N,678,5376,1338,208,24,171,584,977,131,68,,,593,529,3.3000000000,33,15,40,4326,1253,134,542,858,141,152,0.97,Oakland Athletics,Oakland Coliseum,778355,97,95,OAK,OAK,OAK
+1970,AL,WS2,TEX,E,6,162,81,70,92,N,,N,N,626,5460,1302,184,28,138,635,989,72,42,,,689,615,3.8000000000,20,11,40,4371,1375,139,611,823,116,173,0.98,Washington Senators,R.F.K. Stadium,824789,95,95,WSA,WS2,WS2
+1970,NL,ATL,ATL,W,5,162,81,76,86,N,,N,N,736,5546,1495,215,24,160,522,736,58,34,,,772,688,4.3300000000,45,9,24,4290,1451,185,478,960,140,118,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1078848,106,106,ATL,ATL,ATL
+1970,NL,CHN,CHC,E,2,162,80,84,78,N,,N,N,806,5491,1424,228,44,179,607,844,39,16,,,679,600,3.7600000000,59,9,25,4305,1402,143,475,1000,137,146,0.97,Chicago Cubs,Wrigley Field,1642705,111,110,CHC,CHN,CHN
+1970,NL,CIN,CIN,W,1,162,81,102,60,Y,,Y,N,775,5540,1498,253,45,191,547,984,115,52,,,681,592,3.6900000000,32,15,60,4332,1370,118,592,843,151,173,0.97,Cincinnati Reds,Crosley Field/Riverfront Stadium,1803568,104,103,CIN,CIN,CIN
+1970,NL,HOU,HOU,W,4,162,81,79,83,N,,N,N,744,5574,1446,250,47,129,598,911,114,41,,,763,684,4.2300000000,36,6,35,4368,1491,131,577,942,138,144,0.97,Houston Astros,Astrodome,1253444,96,96,HOU,HOU,HOU
+1970,NL,LAN,LAD,W,2,161,81,87,74,N,,N,N,749,5606,1515,233,67,87,541,841,138,57,,,684,619,3.8200000000,37,17,42,4374,1394,164,496,880,133,135,0.97,Los Angeles Dodgers,Dodger Stadium,1697142,95,94,LAD,LAN,LAN
+1970,NL,MON,WSN,E,6,162,80,73,89,N,,N,N,687,5411,1284,211,35,136,659,972,65,45,,,807,719,4.5000000000,29,10,32,4314,1434,162,716,914,141,193,0.97,Montreal Expos,Jarry Park,1424683,99,101,MON,MON,MON
+1970,NL,NYN,NYM,E,3,162,82,83,79,N,,N,N,695,5443,1358,211,42,120,684,1062,118,54,,,630,559,3.4500000000,47,10,32,4377,1260,135,575,1064,124,136,0.97,New York Mets,Shea Stadium,2697479,100,99,NYM,NYN,NYN
+1970,NL,PHI,PHI,E,5,161,80,73,88,N,,N,N,594,5456,1299,224,58,101,519,1066,72,64,,,730,677,4.1700000000,24,8,36,4383,1483,132,538,1047,112,134,0.98,Philadelphia Phillies,Connie Mack Stadium,708247,98,99,PHI,PHI,PHI
+1970,NL,PIT,PIT,E,1,162,82,89,73,Y,,N,N,729,5637,1522,235,70,130,444,871,66,34,,,664,597,3.7000000000,36,13,43,4359,1386,106,625,990,137,195,0.97,Pittsburgh Pirates,Forbes Field/Three Rivers Stadium,1341947,97,96,PIT,PIT,PIT
+1970,NL,SDN,SDP,W,6,162,81,63,99,N,,N,N,681,5494,1353,208,36,172,500,1164,60,45,,,788,698,4.3600000000,24,9,32,4320,1483,149,611,886,157,159,0.97,San Diego Padres,Jack Murphy Stadium,643679,96,98,SDP,SDN,SDN
+1970,NL,SFN,SFG,W,3,162,81,86,76,N,,N,N,831,5578,1460,257,35,165,729,1005,83,27,,,826,728,4.5000000000,50,7,30,4371,1514,156,604,931,170,153,0.97,San Francisco Giants,Candlestick Park,740720,100,99,SFG,SFN,SFN
+1970,NL,SLN,STL,E,4,162,81,76,86,N,,N,N,744,5689,1497,218,51,113,569,961,117,47,,,747,665,4.0600000000,51,11,20,4425,1483,102,632,960,150,159,0.97,St. Louis Cardinals,Busch Stadium II,1629736,103,102,STL,SLN,SLN
+1971,AL,BAL,BAL,E,1,158,77,101,57,Y,,Y,N,742,5303,1382,207,25,158,672,844,66,38,,,530,470,2.9900000000,71,15,22,4245,1257,125,416,793,112,148,0.98,Baltimore Orioles,Memorial Stadium,1023037,99,97,BAL,BAL,BAL
+1971,AL,BOS,BOS,E,3,162,80,85,77,N,,N,N,691,5401,1360,246,28,161,552,871,51,34,,,667,609,3.8000000000,44,11,35,4329,1424,136,535,871,116,149,0.98,Boston Red Sox,Fenway Park II,1678732,108,108,BOS,BOS,BOS
+1971,AL,CAL,ANA,W,4,162,81,76,86,N,,N,N,511,5495,1271,213,18,96,441,827,72,34,,,576,510,3.1000000000,39,11,32,4443,1246,101,607,904,131,159,0.98,California Angels,Anaheim Stadium,926373,93,93,CAL,CAL,CAL
+1971,AL,CHA,CHW,W,3,162,81,79,83,N,,N,N,617,5382,1346,185,30,138,562,870,83,65,,,597,503,3.1200000000,46,19,32,4350,1348,100,468,976,160,128,0.97,Chicago White Sox,Comiskey Park,833891,104,104,CHW,CHA,CHA
+1971,AL,CLE,CLE,E,6,162,81,60,102,N,,N,N,543,5467,1303,200,20,109,467,868,57,37,,,747,685,4.2800000000,21,7,32,4320,1352,154,770,937,116,159,0.98,Cleveland Indians,Cleveland Stadium,591361,109,110,CLE,CLE,CLE
+1971,AL,DET,DET,E,2,162,81,91,71,N,,N,N,701,5502,1399,214,38,179,540,854,35,43,,,645,592,3.6300000000,53,11,32,4404,1355,126,609,1000,106,156,0.98,Detroit Tigers,Tiger Stadium,1591073,105,105,DET,DET,DET
+1971,AL,KCA,KCR,W,2,161,81,85,76,N,,N,N,603,5295,1323,225,40,80,490,819,130,46,,,566,513,3.2500000000,34,15,44,4260,1301,84,496,775,132,178,0.97,Kansas City Royals,Municipal Stadium II,910784,99,99,KCR,KCA,KCA
+1971,AL,MIN,MIN,W,5,160,79,74,86,N,,N,N,654,5414,1406,197,31,116,512,846,66,44,,,670,599,3.8100000000,43,9,25,4248,1384,139,529,895,117,134,0.98,Minnesota Twins,Metropolitan Stadium,940858,103,102,MIN,MIN,MIN
+1971,AL,ML4,MIL,W,6,161,82,69,92,N,,N,N,534,5185,1188,160,23,104,543,924,82,53,,,609,532,3.3800000000,32,23,32,4248,1303,130,569,795,138,152,0.97,Milwaukee Brewers,County Stadium,731531,98,100,MIL,MIL,MIL
+1971,AL,NYA,NYY,E,4,162,81,82,80,N,,N,N,648,5413,1377,195,43,97,581,717,75,55,,,641,553,3.4300000000,67,15,12,4356,1382,126,423,707,125,159,0.98,New York Yankees,Yankee Stadium I,1070771,94,94,NYY,NYA,NYA
+1971,AL,OAK,OAK,W,1,161,81,101,60,Y,,N,N,691,5494,1383,195,25,160,542,1018,80,53,,,564,498,3.0500000000,57,18,36,4407,1229,131,501,999,117,157,0.98,Oakland Athletics,Oakland Coliseum,914993,98,97,OAK,OAK,OAK
+1971,AL,WS2,TEX,E,5,159,81,63,96,N,,N,N,537,5290,1219,189,30,86,575,956,68,45,,,660,583,3.7000000000,30,10,26,4254,1376,132,554,762,141,170,0.97,Washington Senators,R.F.K. Stadium,655156,94,95,WSA,WS2,WS2
+1971,NL,ATL,ATL,W,3,162,82,82,80,N,,N,N,643,5575,1434,192,30,153,434,747,57,46,,,699,614,3.7500000000,40,11,31,4422,1529,152,485,823,146,180,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1006320,106,107,ATL,ATL,ATL
+1971,NL,CHN,CHC,E,3,162,81,83,79,N,,N,N,637,5438,1401,202,34,128,527,772,44,32,,,648,579,3.6100000000,75,17,13,4332,1458,132,411,900,126,150,0.98,Chicago Cubs,Wrigley Field,1653007,113,113,CHC,CHN,CHN
+1971,NL,CIN,CIN,W,4,162,81,79,83,N,,N,N,586,5414,1306,203,28,138,438,907,59,33,,,581,537,3.3500000000,27,11,38,4332,1298,112,501,750,103,174,0.98,Cincinnati Reds,Riverfront Stadium,1501122,95,94,CIN,CIN,CIN
+1971,NL,HOU,HOU,W,4,162,81,79,83,N,,N,N,585,5492,1319,230,52,71,478,888,101,51,,,567,512,3.1300000000,43,10,25,4413,1318,75,475,914,104,152,0.98,Houston Astros,Astrodome,1261589,98,97,HOU,HOU,HOU
+1971,NL,LAN,LAD,W,2,162,81,89,73,N,,N,N,663,5523,1469,213,38,95,489,755,76,40,,,587,520,3.2300000000,48,18,33,4347,1363,110,399,853,131,159,0.97,Los Angeles Dodgers,Dodger Stadium,2064594,94,93,LAD,LAN,LAN
+1971,NL,MON,WSN,E,5,162,80,71,90,N,,N,N,622,5335,1312,197,29,88,543,800,51,43,,,729,656,4.1200000000,49,8,25,4302,1418,133,658,829,150,164,0.97,Montreal Expos,Jarry Park,1290963,100,101,MON,MON,MON
+1971,NL,NYN,NYM,E,3,162,81,83,79,N,,N,N,588,5477,1365,203,29,98,547,958,89,43,,,550,487,2.9900000000,42,13,22,4398,1227,100,529,1157,112,135,0.98,New York Mets,Shea Stadium,2266680,98,98,NYM,NYN,NYN
+1971,NL,PHI,PHI,E,6,162,81,67,95,N,,N,N,558,5538,1289,209,35,123,499,1031,63,39,,,688,606,3.7100000000,31,10,25,4410,1396,132,525,838,121,158,0.98,Philadelphia Phillies,Veterans Stadium,1511223,100,102,PHI,PHI,PHI
+1971,NL,PIT,PIT,E,1,162,80,97,65,Y,,Y,Y,788,5674,1555,223,61,154,469,919,65,31,,,599,537,3.3100000000,43,15,48,4383,1426,108,470,813,133,164,0.97,Pittsburgh Pirates,Three Rivers Stadium,1501132,102,100,PIT,PIT,PIT
+1971,NL,SDN,SDP,W,6,161,81,61,100,N,,N,N,486,5366,1250,184,31,96,438,966,70,45,,,610,514,3.2200000000,47,10,17,4314,1351,93,559,923,161,144,0.97,San Diego Padres,Jack Murphy Stadium,557513,93,95,SDP,SDN,SDN
+1971,NL,SFN,SFG,W,1,162,81,90,72,Y,,N,N,706,5461,1348,224,36,140,654,1042,101,36,,,644,536,3.3200000000,45,14,30,4362,1324,128,471,831,177,153,0.97,San Francisco Giants,Candlestick Park,1106043,99,99,SFG,SFN,SFN
+1971,NL,SLN,STL,E,2,163,82,90,72,N,,N,N,739,5610,1542,225,54,95,543,757,124,53,,,699,628,3.8500000000,56,14,22,4401,1482,104,576,911,142,155,0.97,St. Louis Cardinals,Busch Stadium II,1604671,104,104,STL,SLN,SLN
+1972,AL,BAL,BAL,E,3,154,77,80,74,N,,N,N,519,5028,1153,193,29,100,507,935,78,41,,,430,385,2.5300000000,62,20,21,4113,1116,85,395,788,100,150,0.98,Baltimore Orioles,Memorial Stadium,899950,103,101,BAL,BAL,BAL
+1972,AL,BOS,BOS,E,2,155,78,85,70,N,,N,N,640,5208,1289,229,34,124,522,858,66,30,,,620,533,3.4700000000,48,20,25,4146,1309,101,512,918,130,141,0.97,Boston Red Sox,Fenway Park II,1441718,106,106,BOS,BOS,BOS
+1972,AL,CAL,ANA,W,5,155,80,75,80,N,,N,N,454,5165,1249,171,26,78,358,850,57,37,,,533,468,3.0600000000,57,18,16,4131,1109,90,620,1000,114,135,0.98,California Angels,Anaheim Stadium,744190,94,95,CAL,CAL,CAL
+1972,AL,CHA,CHW,W,2,154,78,87,67,N,,N,N,566,5083,1208,170,28,108,511,991,100,52,,,538,480,3.1200000000,36,14,42,4155,1269,94,431,936,135,136,0.97,Chicago White Sox,Comiskey Park,1177318,103,103,CHW,CHA,CHA
+1972,AL,CLE,CLE,E,5,156,77,72,84,N,,N,N,472,5207,1220,187,18,91,420,762,49,53,,,519,457,2.9200000000,47,13,24,4230,1232,123,534,846,111,157,0.98,Cleveland Indians,Cleveland Stadium,626354,104,106,CLE,CLE,CLE
+1972,AL,DET,DET,E,1,156,78,86,70,Y,,N,N,558,5099,1206,179,32,122,483,793,17,21,,,514,456,2.9600000000,46,11,33,4164,1212,101,465,952,96,137,0.98,Detroit Tigers,Tiger Stadium,1892386,103,103,DET,DET,DET
+1972,AL,KCA,KCR,W,4,154,77,76,78,N,,N,N,580,5167,1317,220,26,78,534,711,85,44,,,545,497,3.2400000000,44,16,29,4143,1293,85,405,801,116,164,0.98,Kansas City Royals,Municipal Stadium II,707656,99,98,KCR,KCA,KCA
+1972,AL,MIN,MIN,W,3,154,74,77,77,N,,N,N,537,5234,1277,182,31,93,478,905,53,41,,,535,441,2.8400000000,37,17,34,4197,1188,105,444,838,159,133,0.97,Minnesota Twins,Metropolitan Stadium,797901,105,105,MIN,MIN,MIN
+1972,AL,ML4,MIL,E,6,156,79,65,91,N,,N,N,493,5124,1204,167,22,88,472,868,64,57,,,595,533,3.4500000000,37,14,32,4173,1289,116,486,740,139,145,0.97,Milwaukee Brewers,County Stadium,600440,97,98,MIL,MIL,MIL
+1972,AL,NYA,NYY,E,4,155,77,79,76,N,,N,N,557,5168,1288,201,24,103,491,689,71,42,,,527,465,3.0500000000,35,19,39,4119,1306,87,419,625,134,179,0.97,New York Yankees,Yankee Stadium I,966328,97,96,NYY,NYA,NYA
+1972,AL,OAK,OAK,W,1,155,77,93,62,Y,,Y,Y,604,5200,1248,195,29,134,463,886,87,48,,,457,406,2.5800000000,42,23,43,4251,1170,96,418,862,130,146,0.97,Oakland Athletics,Oakland Coliseum,921323,95,93,OAK,OAK,OAK
+1972,AL,TEX,TEX,W,6,154,77,54,100,N,,N,N,461,5029,1092,166,17,56,503,926,126,73,,,628,539,3.5300000000,11,8,34,4122,1258,92,613,868,166,147,0.97,Texas Rangers,Arlington Stadium,662974,96,98,TEX,TEX,TEX
+1972,NL,ATL,ATL,W,4,155,78,70,84,N,,N,N,628,5278,1363,186,17,144,532,770,47,35,,,730,653,4.2700000000,40,4,27,4131,1412,155,512,732,156,130,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,752973,109,110,ATL,ATL,ATL
+1972,NL,CHN,CHC,E,2,156,77,85,70,N,,N,N,685,5247,1346,206,40,133,565,815,69,47,,,567,500,3.2200000000,54,19,32,4194,1329,112,421,824,132,148,0.97,Chicago Cubs,Wrigley Field,1299163,110,110,CHC,CHN,CHN
+1972,NL,CIN,CIN,W,1,154,76,95,59,Y,,Y,N,707,5241,1317,214,44,124,606,914,140,63,,,557,504,3.2100000000,25,15,60,4236,1313,129,435,806,110,143,0.98,Cincinnati Reds,Riverfront Stadium,1611459,94,93,CIN,CIN,CIN
+1972,NL,HOU,HOU,W,3,153,77,84,69,N,,N,N,708,5267,1359,233,38,134,524,907,111,56,,,636,580,3.7700000000,38,14,31,4155,1340,114,498,971,114,151,0.98,Houston Astros,Astrodome,1469247,98,97,HOU,HOU,HOU
+1972,NL,LAN,LAD,W,2,155,75,85,70,N,,N,N,584,5270,1349,178,39,98,480,786,82,39,,,527,433,2.7800000000,50,23,29,4209,1196,83,429,856,162,145,0.97,Los Angeles Dodgers,Dodger Stadium,1860858,98,97,LAD,LAN,LAN
+1972,NL,MON,WSN,E,5,156,78,70,86,N,,N,N,513,5156,1205,156,22,91,474,828,68,66,,,609,559,3.5900000000,39,11,23,4203,1281,103,579,888,134,141,0.97,Montreal Expos,Jarry Park,1142145,101,102,MON,MON,MON
+1972,NL,NYN,NYM,E,3,156,78,83,73,N,,N,N,528,5135,1154,175,31,105,589,990,41,41,,,578,512,3.2600000000,32,12,41,4242,1263,118,486,1059,116,122,0.98,New York Mets,Shea Stadium,2134185,97,97,NYM,NYN,NYN
+1972,NL,PHI,PHI,E,6,156,79,59,97,N,,N,N,503,5248,1240,200,36,98,487,930,42,50,,,635,569,3.6600000000,43,13,15,4200,1318,117,536,927,116,142,0.98,Philadelphia Phillies,Veterans Stadium,1343329,102,104,PHI,PHI,PHI
+1972,NL,PIT,PIT,E,1,155,78,96,59,Y,,N,N,691,5490,1505,251,47,110,404,871,49,30,,,512,441,2.8100000000,39,15,48,4242,1282,90,433,838,136,171,0.97,Pittsburgh Pirates,Three Rivers Stadium,1427460,99,97,PIT,PIT,PIT
+1972,NL,SDN,SDP,W,6,153,80,58,95,N,,N,N,488,5213,1181,168,38,102,407,976,78,46,,,665,589,3.7800000000,39,17,19,4209,1350,121,618,960,144,146,0.97,San Diego Padres,Jack Murphy Stadium,644273,92,95,SDP,SDN,SDN
+1972,NL,SFN,SFG,W,5,155,77,69,86,N,,N,N,662,5245,1281,211,36,150,480,964,123,45,,,649,568,3.6900000000,44,8,23,4158,1309,130,507,771,154,121,0.97,San Francisco Giants,Candlestick Park,647744,102,102,SFG,SFN,SFN
+1972,NL,SLN,STL,E,4,156,77,75,81,N,,N,N,568,5326,1383,214,42,70,437,793,104,48,,,600,532,3.4200000000,64,13,13,4197,1290,87,531,912,140,146,0.97,St. Louis Cardinals,Busch Stadium II,1196894,99,99,STL,SLN,SLN
+1973,AL,BAL,BAL,E,1,162,81,97,65,Y,,N,N,754,5537,1474,229,48,119,648,752,146,64,,,561,498,3.0700000000,67,14,26,4383,1297,124,475,715,119,184,0.98,Baltimore Orioles,Memorial Stadium,958667,99,98,BAL,BAL,BAL
+1973,AL,BOS,BOS,E,2,162,81,89,73,N,,N,N,738,5513,1472,235,30,147,581,799,114,45,,,647,584,3.6500000000,67,10,33,4320,1417,158,499,808,127,162,0.97,Boston Red Sox,Fenway Park II,1481002,106,106,BOS,BOS,BOS
+1973,AL,CAL,ANA,W,4,162,81,79,83,N,,N,N,629,5505,1395,183,29,93,509,816,59,47,,,657,571,3.5300000000,72,13,19,4368,1351,104,614,1010,156,153,0.97,California Angels,Anaheim Stadium,1058206,92,92,CAL,CAL,CAL
+1973,AL,CHA,CHW,W,5,162,81,77,85,N,,N,N,652,5475,1400,228,38,111,537,952,83,73,,,705,624,3.8600000000,48,15,35,4368,1484,110,574,848,144,165,0.97,Chicago White Sox,Comiskey Park,1302527,105,105,CHW,CHA,CHA
+1973,AL,CLE,CLE,E,6,162,81,71,91,N,,N,N,680,5592,1429,205,29,158,471,793,60,68,,,826,745,4.5800000000,55,9,21,4392,1532,172,602,883,138,174,0.97,Cleveland Indians,Cleveland Stadium,615107,103,104,CLE,CLE,CLE
+1973,AL,DET,DET,E,3,162,81,85,77,N,,N,N,642,5508,1400,213,32,157,509,722,28,30,,,674,627,3.9000000000,39,11,46,4341,1468,154,493,911,112,144,0.98,Detroit Tigers,Tiger Stadium,1724146,106,107,DET,DET,DET
+1973,AL,KCA,KCR,W,2,162,81,88,74,N,,N,N,755,5508,1440,239,40,114,644,696,105,69,,,752,675,4.1900000000,40,7,41,4347,1521,114,617,790,167,192,0.97,Kansas City Royals,Royals Stadium,1345341,106,106,KCR,KCA,KCA
+1973,AL,MIN,MIN,W,3,162,81,81,81,N,,N,N,738,5625,1521,240,44,120,598,954,87,46,,,692,608,3.7700000000,48,18,34,4353,1443,115,519,879,139,147,0.97,Minnesota Twins,Metropolitan Stadium,907499,104,104,MIN,MIN,MIN
+1973,AL,ML4,MIL,E,5,162,81,74,88,N,,N,N,708,5526,1399,229,40,145,563,977,110,66,,,731,643,3.9800000000,50,11,28,4362,1476,119,623,671,145,167,0.97,Milwaukee Brewers,County Stadium,1092158,97,98,MIL,MIL,MIL
+1973,AL,NYA,NYY,E,4,162,81,80,82,N,,N,N,641,5492,1435,212,17,131,489,680,47,43,,,610,530,3.3400000000,47,16,39,4281,1379,109,457,708,156,172,0.97,New York Yankees,Yankee Stadium I,1262103,97,96,NYY,NYA,NYA
+1973,AL,OAK,OAK,W,1,162,81,94,68,Y,,Y,Y,758,5507,1431,216,28,147,595,919,128,57,,,615,533,3.2900000000,46,16,41,4371,1311,143,494,797,137,170,0.97,Oakland Athletics,Oakland Coliseum,1000763,95,93,OAK,OAK,OAK
+1973,AL,TEX,TEX,W,6,162,81,57,105,N,,N,N,619,5488,1397,195,29,110,503,791,91,53,,,844,737,4.6400000000,35,10,27,4290,1514,130,680,831,161,164,0.97,Texas Rangers,Arlington Stadium,686085,95,97,TEX,TEX,TEX
+1973,NL,ATL,ATL,W,5,162,81,76,85,N,,N,N,799,5631,1497,219,34,206,608,870,84,40,,,774,690,4.2500000000,34,9,35,4386,1467,144,575,803,166,142,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,800655,108,108,ATL,ATL,ATL
+1973,NL,CHN,CHC,E,5,161,80,77,84,N,,N,N,614,5363,1322,201,21,117,575,855,65,58,,,655,584,3.6600000000,27,13,40,4311,1471,128,438,885,157,155,0.97,Chicago Cubs,Wrigley Field,1351705,107,107,CHC,CHN,CHN
+1973,NL,CIN,CIN,W,1,162,81,99,63,Y,,N,N,741,5505,1398,232,34,137,639,947,148,55,,,621,556,3.4000000000,39,17,43,4419,1389,135,518,801,115,162,0.98,Cincinnati Reds,Riverfront Stadium,2017601,95,93,CIN,CIN,CIN
+1973,NL,HOU,HOU,W,4,162,81,82,80,N,,N,N,681,5532,1391,216,35,134,469,962,92,48,,,672,608,3.7500000000,45,14,26,4380,1389,111,575,907,112,140,0.98,Houston Astros,Astrodome,1394004,99,99,HOU,HOU,HOU
+1973,NL,LAN,LAD,W,2,162,81,95,66,N,,N,N,675,5604,1473,219,29,110,497,795,109,50,,,565,497,3.0000000000,45,15,38,4473,1270,129,461,961,125,166,0.98,Los Angeles Dodgers,Dodger Stadium,2136192,97,95,LAD,LAN,LAN
+1973,NL,MON,WSN,E,4,162,81,79,83,N,,N,N,668,5369,1345,190,23,125,695,777,77,68,,,702,598,3.7100000000,26,6,38,4353,1356,128,681,866,163,156,0.97,Montreal Expos,Jarry Park,1246863,103,104,MON,MON,MON
+1973,NL,NYN,NYM,E,1,161,81,82,79,Y,,Y,N,608,5457,1345,198,24,85,540,805,27,22,,,588,531,3.2600000000,47,15,40,4395,1345,127,490,1027,126,140,0.98,New York Mets,Shea Stadium,1912390,98,99,NYM,NYN,NYN
+1973,NL,PHI,PHI,E,6,162,81,71,91,N,,N,N,642,5546,1381,218,29,134,476,979,51,47,,,717,642,3.9900000000,49,11,22,4341,1435,131,632,919,134,179,0.97,Philadelphia Phillies,Veterans Stadium,1475934,102,103,PHI,PHI,PHI
+1973,NL,PIT,PIT,E,3,162,81,80,82,N,,N,N,704,5608,1465,257,44,154,432,842,23,30,,,693,601,3.7300000000,26,11,44,4350,1426,110,564,839,151,156,0.97,Pittsburgh Pirates,Three Rivers Stadium,1319913,98,96,PIT,PIT,PIT
+1973,NL,SDN,SDP,W,6,162,81,60,102,N,,N,N,548,5457,1330,198,26,112,401,966,88,36,,,770,661,4.1600000000,34,10,23,4290,1461,157,548,845,170,152,0.97,San Diego Padres,Jack Murphy Stadium,611826,91,94,SDP,SDN,SDN
+1973,NL,SFN,SFG,W,3,162,81,88,74,N,,N,N,739,5537,1452,212,52,161,590,913,112,52,,,702,611,3.7900000000,33,8,44,4356,1442,145,485,787,163,138,0.97,San Francisco Giants,Candlestick Park,834193,105,105,SFG,SFN,SFN
+1973,NL,SLN,STL,E,2,162,81,81,81,N,,N,N,643,5478,1418,240,35,75,531,796,100,46,,,603,527,3.2500000000,42,14,36,4380,1366,105,486,867,158,149,0.97,St. Louis Cardinals,Busch Stadium II,1574046,100,100,STL,SLN,SLN
+1974,AL,BAL,BAL,E,1,162,81,91,71,Y,,N,N,659,5535,1418,226,27,116,509,770,145,58,,,612,536,3.2700000000,57,16,25,4422,1393,101,480,701,128,174,0.98,Baltimore Orioles,Memorial Stadium,962572,97,95,BAL,BAL,BAL
+1974,AL,BOS,BOS,E,3,162,81,84,78,N,,N,N,696,5499,1449,236,31,109,569,811,104,58,,,661,601,3.7200000000,71,12,18,4365,1462,126,463,751,145,156,0.97,Boston Red Sox,Fenway Park II,1556411,108,107,BOS,BOS,BOS
+1974,AL,CAL,ANA,W,6,163,81,68,94,N,,N,N,618,5401,1372,203,31,95,509,801,119,79,,,657,563,3.5200000000,64,13,12,4317,1339,101,649,986,147,150,0.97,California Angels,Anaheim Stadium,917269,94,94,CAL,CAL,CAL
+1974,AL,CHA,CHW,W,4,163,82,80,80,N,,N,N,684,5577,1492,225,23,135,519,858,64,53,,,721,641,3.9400000000,55,11,29,4395,1470,103,548,826,147,188,0.97,Chicago White Sox,Comiskey Park,1149596,103,104,CHW,CHA,CHA
+1974,AL,CLE,CLE,E,4,162,81,77,85,N,,N,N,662,5474,1395,201,19,131,432,756,79,68,,,694,610,3.8000000000,45,9,27,4335,1419,138,479,650,143,157,0.97,Cleveland Indians,Cleveland Stadium,1114262,99,100,CLE,CLE,CLE
+1974,AL,DET,DET,E,6,162,81,72,90,N,,N,N,620,5568,1375,200,35,131,436,784,67,38,,,768,673,4.1600000000,54,7,15,4365,1443,148,621,869,158,155,0.97,Detroit Tigers,Tiger Stadium,1243080,102,104,DET,DET,DET
+1974,AL,KCA,KCR,W,5,162,81,77,85,N,,N,N,667,5582,1448,232,42,89,550,768,146,76,,,662,574,3.5100000000,54,13,17,4413,1477,91,482,731,152,166,0.97,Kansas City Royals,Royals Stadium,1173292,105,105,KCR,KCA,KCA
+1974,AL,MIN,MIN,W,3,163,82,82,80,N,,N,N,673,5632,1530,190,37,111,520,791,74,45,,,669,588,3.6400000000,43,11,29,4365,1436,115,513,934,151,164,0.97,Minnesota Twins,Metropolitan Stadium,662401,104,104,MIN,MIN,MIN
+1974,AL,ML4,MIL,E,5,162,81,76,86,N,,N,N,647,5472,1335,228,49,120,500,909,106,75,,,660,609,3.7600000000,43,11,24,4371,1476,126,493,621,127,168,0.98,Milwaukee Brewers,County Stadium,955741,99,99,MIL,MIL,MIL
+1974,AL,NYA,NYY,E,2,162,81,89,73,N,,N,N,671,5524,1451,220,30,101,515,690,53,35,,,623,535,3.3100000000,53,13,24,4365,1402,104,528,829,142,158,0.97,New York Yankees,Shea Stadium,1273075,99,98,NYY,NYA,NYA
+1974,AL,OAK,OAK,W,1,162,81,90,72,Y,,Y,Y,689,5331,1315,205,37,132,568,876,164,93,,,551,472,2.9500000000,49,12,28,4317,1322,90,430,755,141,154,0.97,Oakland Athletics,Oakland Coliseum,845693,94,92,OAK,OAK,OAK
+1974,AL,TEX,TEX,W,2,161,80,84,76,N,,N,N,690,5449,1482,198,39,99,508,710,113,80,,,698,608,3.8200000000,62,16,12,4299,1423,126,449,871,163,164,0.97,Texas Rangers,Arlington Stadium,1193902,97,98,TEX,TEX,TEX
+1974,NL,ATL,ATL,W,3,163,81,88,74,N,,N,N,661,5533,1375,202,37,120,571,772,72,44,,,563,500,3.0500000000,46,21,22,4422,1343,97,488,772,132,161,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,981085,104,104,ATL,ATL,ATL
+1974,NL,CHN,CHC,E,6,162,81,66,96,N,,N,N,669,5574,1397,221,42,110,621,857,78,73,,,826,697,4.2800000000,23,6,26,4398,1593,122,576,895,199,141,0.96,Chicago Cubs,Wrigley Field,1015378,103,105,CHC,CHN,CHN
+1974,NL,CIN,CIN,W,2,163,82,98,64,N,,N,N,776,5535,1437,271,35,135,693,940,146,49,,,631,555,3.4100000000,34,11,27,4398,1364,126,536,875,134,151,0.97,Cincinnati Reds,Riverfront Stadium,2164307,99,97,CIN,CIN,CIN
+1974,NL,HOU,HOU,W,4,162,81,81,81,N,,N,N,653,5489,1441,222,41,110,471,864,108,65,,,632,557,3.4600000000,36,18,18,4350,1396,84,601,738,108,161,0.98,Houston Astros,Astrodome,1090728,96,96,HOU,HOU,HOU
+1974,NL,LAN,LAD,W,1,162,81,102,60,Y,,Y,N,798,5557,1511,231,34,139,597,820,149,75,,,561,483,2.9700000000,33,19,23,4395,1272,112,464,943,157,122,0.97,Los Angeles Dodgers,Dodger Stadium,2632474,96,94,LAD,LAN,LAN
+1974,NL,MON,WSN,E,4,161,80,79,82,N,,N,N,662,5343,1355,201,29,86,652,812,124,49,,,657,572,3.6000000000,35,8,27,4287,1340,99,544,822,153,157,0.97,Montreal Expos,Jarry Park,1019134,105,106,MON,MON,MON
+1974,NL,NYN,NYM,E,5,162,81,71,91,N,,N,N,572,5468,1286,183,22,96,597,735,43,23,,,646,559,3.4200000000,46,15,14,4410,1433,99,504,908,158,150,0.97,New York Mets,Shea Stadium,1722209,99,99,NYM,NYN,NYN
+1974,NL,PHI,PHI,E,3,162,81,80,82,N,,N,N,676,5494,1434,233,50,95,469,822,115,58,,,701,629,3.9100000000,46,4,19,4341,1394,111,682,892,148,168,0.97,Philadelphia Phillies,Veterans Stadium,1808648,104,104,PHI,PHI,PHI
+1974,NL,PIT,PIT,E,1,162,81,88,74,Y,,N,N,751,5702,1560,238,46,114,514,828,55,31,,,657,568,3.4900000000,51,9,17,4398,1428,93,543,721,162,154,0.97,Pittsburgh Pirates,Three Rivers Stadium,1110552,97,95,PIT,PIT,PIT
+1974,NL,SDN,SDP,W,6,162,81,60,102,N,,N,N,541,5415,1239,196,27,99,564,900,85,45,,,830,735,4.5800000000,25,7,19,4335,1536,124,715,855,170,126,0.97,San Diego Padres,Jack Murphy Stadium,1075399,95,98,SDP,SDN,SDN
+1974,NL,SFN,SFG,W,5,162,81,72,90,N,,N,N,634,5482,1380,228,38,93,548,869,107,51,,,723,604,3.7800000000,27,11,25,4317,1409,116,559,756,175,153,0.97,San Francisco Giants,Candlestick Park,519987,105,105,SFG,SFN,SFN
+1974,NL,SLN,STL,E,2,161,81,86,75,N,,N,N,677,5620,1492,216,46,83,531,752,172,62,,,643,570,3.4800000000,37,13,20,4419,1399,97,616,794,147,192,0.97,St. Louis Cardinals,Busch Stadium II,1838413,100,100,STL,SLN,SLN
+1975,AL,BAL,BAL,E,2,159,77,90,69,N,,N,N,682,5474,1382,224,33,124,580,834,104,55,,,553,511,3.1700000000,70,19,21,4353,1285,110,500,717,107,175,0.98,Baltimore Orioles,Memorial Stadium,1002157,94,93,BAL,BAL,BAL
+1975,AL,BOS,BOS,E,1,160,81,95,65,Y,,Y,N,796,5448,1500,284,44,134,565,741,66,58,,,709,635,3.9800000000,62,11,31,4308,1463,145,490,720,135,142,0.97,Boston Red Sox,Fenway Park II,1748587,110,109,BOS,BOS,BOS
+1975,AL,CAL,ANA,W,6,161,81,72,89,N,,N,N,628,5377,1324,195,41,55,593,811,220,108,,,723,628,3.8900000000,59,19,16,4359,1386,123,613,975,184,164,0.97,California Angels,Anaheim Stadium,1058163,92,93,CAL,CAL,CAL
+1975,AL,CHA,CHW,W,5,161,81,75,86,N,,N,N,655,5490,1400,209,38,94,611,800,101,54,,,703,634,3.9300000000,34,7,39,4356,1489,107,655,799,140,155,0.97,Chicago White Sox,Comiskey Park,750802,102,103,CHW,CHA,CHA
+1975,AL,CLE,CLE,E,4,159,80,79,80,N,,N,N,688,5404,1409,201,25,153,525,667,106,89,,,703,612,3.8400000000,37,6,33,4305,1395,136,599,800,134,156,0.97,Cleveland Indians,Cleveland Stadium,977039,99,99,CLE,CLE,CLE
+1975,AL,DET,DET,E,6,159,80,57,102,N,,N,N,570,5366,1338,171,39,125,383,872,63,57,,,786,662,4.2700000000,52,10,17,4188,1496,137,533,787,173,141,0.97,Detroit Tigers,Tiger Stadium,1058836,104,106,DET,DET,DET
+1975,AL,KCA,KCR,W,2,162,81,91,71,N,,N,N,710,5491,1431,263,58,118,591,675,155,75,,,649,561,3.4700000000,52,11,25,4368,1422,108,498,815,155,151,0.97,Kansas City Royals,Royals Stadium,1151836,103,102,KCR,KCA,KCA
+1975,AL,MIN,MIN,W,4,159,82,76,83,N,,N,N,724,5514,1497,215,28,121,563,746,81,48,,,736,640,4.0500000000,57,7,22,4269,1381,137,617,846,169,147,0.97,Minnesota Twins,Metropolitan Stadium,737156,102,102,MIN,MIN,MIN
+1975,AL,ML4,MIL,E,5,162,81,68,94,N,,N,N,675,5378,1343,242,34,146,553,922,65,64,,,792,690,4.3400000000,36,10,34,4293,1496,133,624,643,180,162,0.97,Milwaukee Brewers,County Stadium,1213357,100,101,MIL,MIL,MIL
+1975,AL,NYA,NYY,E,3,160,78,83,77,N,,N,N,681,5415,1430,230,39,110,486,710,102,59,,,588,521,3.2900000000,70,11,20,4272,1325,104,502,809,135,148,0.97,New York Yankees,Shea Stadium,1288048,99,98,NYY,NYA,NYA
+1975,AL,OAK,OAK,W,1,162,81,98,64,Y,,N,N,758,5415,1376,220,33,151,609,846,183,82,,,606,526,3.2700000000,36,10,44,4344,1267,102,523,784,143,140,0.97,Oakland Athletics,Oakland Coliseum,1075518,98,96,OAK,OAK,OAK
+1975,AL,TEX,TEX,W,3,162,80,79,83,N,,N,N,714,5599,1431,208,17,134,613,863,102,62,,,733,628,3.8600000000,60,16,17,4395,1456,123,518,792,191,173,0.97,Texas Rangers,Arlington Stadium,1127924,99,99,TEX,TEX,TEX
+1975,NL,ATL,ATL,W,5,161,80,67,94,N,,N,N,583,5424,1323,179,28,107,543,759,55,38,,,739,621,3.9100000000,32,4,24,4290,1543,101,519,669,175,147,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,534672,103,104,ATL,ATL,ATL
+1975,NL,CHN,CHC,E,5,162,81,75,87,N,,N,N,712,5470,1419,229,41,95,650,802,67,55,,,827,720,4.4900000000,27,8,33,4332,1587,130,551,850,178,152,0.97,Chicago Cubs,Wrigley Field,1034819,104,106,CHC,CHN,CHN
+1975,NL,CIN,CIN,W,1,162,81,108,54,Y,,Y,Y,840,5581,1515,278,37,124,691,916,168,36,,,586,546,3.3700000000,22,8,50,4377,1422,112,487,663,102,173,0.98,Cincinnati Reds,Riverfront Stadium,2315603,102,99,CIN,CIN,CIN
+1975,NL,HOU,HOU,W,6,162,81,64,97,N,,N,N,664,5515,1401,218,54,84,523,762,133,62,,,711,654,4.0400000000,39,6,25,4374,1436,106,679,839,137,166,0.97,Houston Astros,Astrodome,858002,93,93,HOU,HOU,HOU
+1975,NL,LAN,LAD,W,2,162,81,88,74,N,,N,N,648,5453,1355,217,31,118,611,825,138,52,,,534,477,2.9200000000,51,18,21,4407,1215,104,448,894,127,106,0.97,Los Angeles Dodgers,Dodger Stadium,2539349,96,94,LAD,LAN,LAN
+1975,NL,MON,WSN,E,5,162,81,75,87,N,,N,N,601,5518,1346,216,31,98,579,954,108,58,,,690,612,3.7200000000,30,12,25,4440,1448,102,665,831,180,179,0.97,Montreal Expos,Jarry Park,908292,104,106,MON,MON,MON
+1975,NL,NYN,NYM,E,3,162,81,82,80,N,,N,N,646,5587,1430,217,34,101,501,805,32,26,,,625,552,3.3900000000,40,14,31,4398,1344,99,580,989,151,144,0.97,New York Mets,Shea Stadium,1730566,96,95,NYM,NYN,NYN
+1975,NL,PHI,PHI,E,2,162,81,86,76,N,,N,N,735,5592,1506,283,42,125,610,960,126,57,,,694,618,3.8200000000,33,11,30,4365,1353,111,546,897,152,156,0.97,Philadelphia Phillies,Veterans Stadium,1909233,104,103,PHI,PHI,PHI
+1975,NL,PIT,PIT,E,1,161,80,92,69,Y,,N,N,712,5489,1444,255,47,138,468,832,49,28,,,565,481,3.0100000000,43,14,31,4311,1302,79,551,768,151,147,0.97,Pittsburgh Pirates,Three Rivers Stadium,1270018,99,97,PIT,PIT,PIT
+1975,NL,SDN,SDP,W,4,162,81,71,91,N,,N,N,552,5429,1324,215,22,78,506,754,85,50,,,683,566,3.4800000000,40,12,20,4389,1494,99,521,713,188,163,0.97,San Diego Padres,Jack Murphy Stadium,1281747,94,96,SDP,SDN,SDN
+1975,NL,SFN,SFG,W,3,161,81,80,81,N,,N,N,659,5447,1412,235,45,84,604,775,99,47,,,671,595,3.7400000000,37,9,24,4296,1406,92,612,856,145,164,0.97,San Francisco Giants,Candlestick Park,522919,104,105,SFG,SFN,SFN
+1975,NL,SLN,STL,E,3,163,82,82,80,N,,N,N,662,5597,1527,239,46,81,444,649,116,49,,,689,577,3.5700000000,33,13,36,4362,1452,98,571,824,171,140,0.97,St. Louis Cardinals,Busch Stadium II,1695270,104,105,STL,SLN,SLN
+1976,AL,BAL,BAL,E,2,162,81,88,74,N,,N,N,619,5457,1326,213,28,119,519,883,150,61,,,598,542,3.3200000000,59,16,23,4404,1396,80,489,678,118,157,0.98,Baltimore Orioles,Memorial Stadium,1058609,94,93,BAL,BAL,BAL
+1976,AL,BOS,BOS,E,3,162,81,83,79,N,,N,N,716,5511,1448,257,53,134,500,832,95,70,,,660,570,3.5200000000,49,13,27,4374,1495,109,409,673,139,148,0.97,Boston Red Sox,Fenway Park II,1895846,113,112,BOS,BOS,BOS
+1976,AL,CAL,ANA,W,4,162,81,76,86,N,,N,N,550,5385,1265,210,23,63,534,812,126,80,,,631,551,3.3600000000,64,15,17,4431,1323,95,553,992,150,139,0.97,California Angels,Anaheim Stadium,1006774,93,94,CAL,CAL,CAL
+1976,AL,CHA,CHW,W,6,161,80,64,97,N,,N,N,586,5532,1410,209,46,73,471,739,120,53,,,745,684,4.2500000000,54,10,22,4344,1460,87,600,802,130,155,0.97,Chicago White Sox,Comiskey Park,914945,101,102,CHW,CHA,CHA
+1976,AL,CLE,CLE,E,4,159,79,81,78,N,,N,N,615,5412,1423,189,38,85,479,631,75,69,,,615,552,3.4700000000,30,17,46,4296,1361,80,533,928,121,159,0.98,Cleveland Indians,Cleveland Stadium,948776,98,98,CLE,CLE,CLE
+1976,AL,DET,DET,E,5,161,80,74,87,N,,N,N,609,5441,1401,207,38,101,450,730,107,59,,,709,615,3.8700000000,55,12,20,4293,1426,101,550,738,168,161,0.97,Detroit Tigers,Tiger Stadium,1467020,103,105,DET,DET,DET
+1976,AL,KCA,KCR,W,1,162,81,90,72,Y,,N,N,713,5540,1490,259,57,65,484,650,218,106,,,611,525,3.2100000000,41,12,35,4416,1356,83,493,735,139,147,0.97,Kansas City Royals,Royals Stadium,1680265,102,100,KCR,KCA,KCA
+1976,AL,MIN,MIN,W,3,162,81,85,77,N,,N,N,743,5574,1526,222,51,81,550,714,146,75,,,704,598,3.6900000000,29,11,23,4377,1421,89,610,762,172,182,0.97,Minnesota Twins,Metropolitan Stadium,715394,103,102,MIN,MIN,MIN
+1976,AL,ML4,MIL,E,6,161,81,66,95,N,,N,N,570,5396,1326,170,38,88,511,909,62,61,,,655,580,3.6400000000,45,10,27,4305,1406,99,567,677,152,160,0.97,Milwaukee Brewers,County Stadium,1012164,98,99,MIL,MIL,MIL
+1976,AL,NYA,NYY,E,1,159,80,97,62,Y,,Y,N,730,5555,1496,231,36,120,470,616,163,65,,,575,516,3.1900000000,62,15,37,4365,1300,97,448,674,126,141,0.98,New York Yankees,Yankee Stadium II,2012434,100,98,NYY,NYA,NYA
+1976,AL,OAK,OAK,W,2,161,81,87,74,N,,N,N,686,5353,1319,208,33,113,592,818,341,123,,,598,528,3.2600000000,39,15,29,4377,1412,96,415,711,144,130,0.97,Oakland Athletics,Oakland Coliseum,780593,95,95,OAK,OAK,OAK
+1976,AL,TEX,TEX,W,4,162,81,76,86,N,,N,N,616,5555,1390,213,26,80,568,809,87,45,,,652,564,3.4500000000,63,15,15,4416,1464,106,461,773,156,142,0.97,Texas Rangers,Arlington Stadium,1164982,102,102,TEX,TEX,TEX
+1976,NL,ATL,ATL,W,6,162,81,70,92,N,,N,N,620,5345,1309,170,30,82,589,811,74,61,,,700,617,3.8600000000,33,13,27,4314,1435,86,564,818,167,151,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,818179,106,108,ATL,ATL,ATL
+1976,NL,CHN,CHC,E,4,162,81,75,87,N,,N,N,611,5519,1386,216,24,105,490,834,74,74,,,728,642,3.9300000000,27,12,33,4413,1511,123,490,850,140,145,0.97,Chicago Cubs,Wrigley Field,1026217,108,109,CHC,CHN,CHN
+1976,NL,CIN,CIN,W,1,162,81,102,60,Y,,Y,Y,857,5702,1599,271,63,141,681,902,210,57,,,633,574,3.5100000000,33,12,45,4413,1436,100,491,790,102,157,0.98,Cincinnati Reds,Riverfront Stadium,2629708,102,100,CIN,CIN,CIN
+1976,NL,HOU,HOU,W,3,162,82,80,82,N,,N,N,625,5464,1401,195,50,66,530,719,150,57,,,657,571,3.5600000000,42,17,29,4332,1349,82,662,780,140,155,0.97,Houston Astros,Astrodome,886146,91,91,HOU,HOU,HOU
+1976,NL,LAN,LAD,W,2,162,81,92,70,N,,N,N,608,5472,1371,200,34,91,486,744,144,55,,,543,493,3.0200000000,47,17,28,4410,1330,97,479,747,128,154,0.98,Los Angeles Dodgers,Dodger Stadium,2386301,98,97,LAD,LAN,LAN
+1976,NL,MON,WSN,E,6,162,80,55,107,N,,N,N,531,5428,1275,224,32,94,433,841,86,44,,,734,638,3.9900000000,26,10,21,4320,1442,89,659,783,147,179,0.97,Montreal Expos,Jarry Park,646704,105,107,MON,MON,MON
+1976,NL,NYN,NYM,E,3,162,82,86,76,N,,N,N,615,5415,1334,198,34,102,561,797,66,58,,,538,473,2.9400000000,53,18,25,4347,1248,97,419,1025,131,116,0.97,New York Mets,Shea Stadium,1468754,94,94,NYM,NYN,NYN
+1976,NL,PHI,PHI,E,1,162,81,101,61,Y,,N,N,770,5528,1505,259,45,110,542,793,127,70,,,557,499,3.0800000000,34,9,44,4377,1377,98,397,918,115,148,0.98,Philadelphia Phillies,Veterans Stadium,2480150,104,102,PHI,PHI,PHI
+1976,NL,PIT,PIT,E,2,162,81,92,70,N,,N,N,708,5604,1499,249,56,110,433,807,130,45,,,630,547,3.3600000000,45,12,35,4398,1402,95,460,762,163,142,0.97,Pittsburgh Pirates,Three Rivers Stadium,1025945,100,99,PIT,PIT,PIT
+1976,NL,SDN,SDP,W,5,162,80,73,89,N,,N,N,570,5369,1327,216,37,64,488,716,92,46,,,662,581,3.6500000000,47,11,18,4296,1368,87,543,652,141,148,0.97,San Diego Padres,Jack Murphy Stadium,1458478,92,94,SDP,SDN,SDN
+1976,NL,SFN,SFG,W,4,162,81,74,88,N,,N,N,595,5452,1340,211,37,85,518,778,88,55,,,686,573,3.5300000000,27,18,31,4383,1464,68,518,746,183,153,0.97,San Francisco Giants,Candlestick Park,626868,103,103,SFG,SFN,SFN
+1976,NL,SLN,STL,E,5,162,81,72,90,N,,N,N,629,5516,1432,243,57,63,512,860,123,55,,,671,581,3.6000000000,35,15,26,4359,1416,91,581,731,174,163,0.97,St. Louis Cardinals,Busch Stadium II,1207079,101,101,STL,SLN,SLN
+1977,AL,BAL,BAL,E,2,161,81,97,64,N,,N,N,719,5494,1433,231,25,148,560,945,90,51,,,653,603,3.7400000000,65,11,23,4353,1414,124,494,737,106,189,0.98,Baltimore Orioles,Memorial Stadium,1195769,94,94,BAL,BAL,BAL
+1977,AL,BOS,BOS,E,2,161,80,97,64,N,,N,N,859,5510,1551,258,56,213,528,905,66,47,,,712,652,4.1100000000,40,13,40,4284,1555,158,378,758,132,162,0.97,Boston Red Sox,Fenway Park II,2074549,113,112,BOS,BOS,BOS
+1977,AL,CAL,ANA,W,5,162,81,74,88,N,,N,N,675,5410,1380,233,40,131,542,880,159,89,,,695,594,3.7200000000,53,13,26,4311,1383,136,572,965,147,137,0.97,California Angels,Anaheim Stadium,1432633,96,96,CAL,CAL,CAL
+1977,AL,CHA,CHW,W,3,162,81,90,72,N,,N,N,844,5633,1568,254,52,192,559,666,42,44,,,771,682,4.2500000000,34,3,40,4332,1557,136,516,842,159,125,0.97,Chicago White Sox,Comiskey Park,1657135,100,101,CHW,CHA,CHA
+1977,AL,CLE,CLE,E,5,161,81,71,90,N,,N,N,676,5491,1476,221,46,100,531,688,87,87,,,739,661,4.1000000000,45,8,30,4356,1441,136,550,876,130,145,0.97,Cleveland Indians,Cleveland Stadium,900365,96,97,CLE,CLE,CLE
+1977,AL,DET,DET,E,4,162,81,74,88,N,,N,N,714,5604,1480,228,45,166,452,764,60,46,,,751,669,4.1300000000,44,3,23,4371,1526,162,470,784,140,153,0.97,Detroit Tigers,Tiger Stadium,1359856,105,105,DET,DET,DET
+1977,AL,KCA,KCR,W,1,162,81,102,60,Y,,N,N,822,5594,1549,299,77,146,522,687,170,87,,,651,571,3.5200000000,41,15,42,4380,1377,110,499,850,137,145,0.97,Kansas City Royals,Royals Stadium,1852603,102,100,KCR,KCA,KCA
+1977,AL,MIN,MIN,W,4,161,80,84,77,N,,N,N,867,5639,1588,273,60,123,563,754,105,65,,,776,699,4.3600000000,35,4,25,4326,1546,151,507,737,144,184,0.97,Minnesota Twins,Metropolitan Stadium,1162727,99,99,MIN,MIN,MIN
+1977,AL,ML4,MIL,E,6,162,81,67,95,N,,N,N,639,5517,1425,255,46,125,443,862,85,67,,,765,687,4.3200000000,38,6,25,4293,1461,136,566,719,139,165,0.97,Milwaukee Brewers,County Stadium,1114938,100,100,MIL,MIL,MIL
+1977,AL,NYA,NYY,E,1,162,81,100,62,Y,,Y,Y,831,5605,1576,267,47,184,533,681,93,57,,,651,581,3.6100000000,52,16,34,4347,1395,139,486,758,132,151,0.97,New York Yankees,Yankee Stadium II,2103092,99,97,NYY,NYA,NYA
+1977,AL,OAK,OAK,W,7,161,81,63,98,N,,N,N,605,5358,1284,176,37,117,516,910,176,89,,,749,645,4.0400000000,32,4,26,4308,1459,145,560,788,190,136,0.97,Oakland Athletics,Oakland Coliseum,495599,97,98,OAK,OAK,OAK
+1977,AL,SEA,SEA,W,6,162,81,64,98,N,,N,N,624,5460,1398,218,33,133,426,769,110,67,,,855,769,4.8300000000,18,1,31,4299,1508,194,578,785,147,162,0.97,Seattle Mariners,Kingdome,1338511,99,101,SEA,SEA,SEA
+1977,AL,TEX,TEX,W,2,162,81,94,68,N,,N,N,767,5541,1497,265,39,135,596,904,154,85,,,657,582,3.5600000000,49,17,31,4416,1412,134,471,864,117,156,0.98,Texas Rangers,Arlington Stadium,1250722,101,101,TEX,TEX,TEX
+1977,AL,TOR,TOR,E,7,161,80,54,107,N,,N,N,605,5418,1367,230,41,100,499,819,65,55,,,822,725,4.5700000000,40,3,20,4284,1538,152,623,771,163,133,0.97,Toronto Blue Jays,Exhibition Stadium,1701052,101,103,TOR,TOR,TOR
+1977,NL,ATL,ATL,W,6,162,81,61,101,N,,N,N,678,5534,1404,218,20,139,537,876,82,53,,,895,779,4.8500000000,28,5,31,4335,1581,169,701,915,175,127,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,872464,112,114,ATL,ATL,ATL
+1977,NL,CHN,CHC,E,4,162,81,81,81,N,,N,N,692,5604,1489,271,37,111,534,796,64,45,,,739,654,4.0100000000,16,10,44,4404,1500,128,489,942,153,147,0.97,Chicago Cubs,Wrigley Field,1439834,111,112,CHC,CHN,CHN
+1977,NL,CIN,CIN,W,2,162,81,88,74,N,,N,N,802,5524,1513,269,42,181,600,911,170,64,,,725,672,4.2100000000,33,12,32,4311,1469,156,544,868,95,154,0.98,Cincinnati Reds,Riverfront Stadium,2519670,102,101,CIN,CIN,CIN
+1977,NL,HOU,HOU,W,3,162,81,81,81,N,,N,N,680,5530,1405,263,60,114,515,839,187,72,,,650,576,3.5400000000,37,11,28,4395,1384,110,545,871,142,136,0.97,Houston Astros,Astrodome,1109560,91,91,HOU,HOU,HOU
+1977,NL,LAN,LAD,W,1,162,81,98,64,Y,,Y,N,769,5589,1484,223,28,191,588,896,114,62,,,582,528,3.2200000000,34,13,39,4425,1393,119,438,930,124,160,0.98,Los Angeles Dodgers,Dodger Stadium,2955087,100,98,LAD,LAN,LAN
+1977,NL,MON,WSN,E,5,162,81,75,87,N,,N,N,665,5675,1474,294,50,138,478,877,88,50,,,736,660,4.0100000000,31,11,33,4443,1426,135,579,856,129,128,0.98,Montreal Expos,Stade Olympique,1433757,97,97,MON,MON,MON
+1977,NL,NYN,NYM,E,6,162,79,64,98,N,,N,N,587,5410,1319,227,30,88,529,887,98,81,,,663,600,3.7700000000,27,12,28,4299,1378,118,490,911,134,132,0.97,New York Mets,Shea Stadium,1066825,95,95,NYM,NYN,NYN
+1977,NL,PHI,PHI,E,1,162,81,101,61,Y,,N,N,847,5546,1548,266,56,186,573,806,135,68,,,668,600,3.7100000000,31,7,47,4365,1451,134,482,856,120,168,0.98,Philadelphia Phillies,Veterans Stadium,2700070,105,103,PHI,PHI,PHI
+1977,NL,PIT,PIT,E,2,162,81,96,66,N,,N,N,734,5662,1550,278,57,133,474,878,260,120,,,665,594,3.6100000000,25,15,39,4443,1406,149,485,890,145,137,0.97,Pittsburgh Pirates,Three Rivers Stadium,1237349,102,101,PIT,PIT,PIT
+1977,NL,SDN,SDP,W,5,162,81,69,93,N,,N,N,692,5602,1397,245,49,120,602,1057,133,57,,,834,722,4.4300000000,6,5,44,4398,1556,160,673,827,189,142,0.97,San Diego Padres,Jack Murphy Stadium,1376269,90,91,SDP,SDN,SDN
+1977,NL,SFN,SFG,W,4,162,81,75,87,N,,N,N,673,5497,1392,227,41,134,568,842,90,59,,,711,608,3.7500000000,27,10,33,4377,1501,114,529,854,178,136,0.97,San Francisco Giants,Candlestick Park,700056,99,100,SFG,SFN,SFN
+1977,NL,SLN,STL,E,3,162,83,83,79,N,,N,N,737,5527,1490,252,56,96,489,823,134,112,,,688,612,3.8100000000,26,10,31,4338,1420,139,532,768,139,174,0.97,St. Louis Cardinals,Busch Stadium II,1659287,99,99,STL,SLN,SLN
+1978,AL,BAL,BAL,E,4,161,81,90,71,N,,N,N,659,5422,1397,248,19,154,552,864,75,61,,,633,565,3.5600000000,65,16,33,4287,1340,107,509,754,110,166,0.98,Baltimore Orioles,Memorial Stadium,1051724,94,93,BAL,BAL,BAL
+1978,AL,BOS,BOS,E,2,163,82,99,64,N,,N,N,796,5587,1493,270,46,172,582,835,74,51,,,657,579,3.5400000000,57,15,26,4416,1530,137,464,706,146,171,0.97,Boston Red Sox,Fenway Park II,2320643,112,110,BOS,BOS,BOS
+1978,AL,CAL,ANA,W,2,162,81,87,75,N,,N,N,691,5472,1417,226,28,108,539,682,86,69,,,666,590,3.6500000000,44,13,33,4365,1382,125,599,892,136,136,0.97,California Angels,Anaheim Stadium,1755386,96,96,CAL,CAL,CAL
+1978,AL,CHA,CHW,W,5,161,80,71,90,N,,N,N,634,5393,1423,221,41,106,409,625,83,68,,,731,659,4.2100000000,38,9,33,4227,1380,128,586,710,138,130,0.97,Chicago White Sox,Comiskey Park,1491100,101,102,CHW,CHA,CHA
+1978,AL,CLE,CLE,E,6,159,78,69,90,N,,N,N,639,5365,1400,223,45,106,488,698,64,63,,,694,621,3.9700000000,36,6,28,4221,1397,100,568,739,121,142,0.98,Cleveland Indians,Cleveland Stadium,800584,99,100,CLE,CLE,CLE
+1978,AL,DET,DET,E,5,162,81,86,76,N,,N,N,714,5601,1520,218,34,129,563,695,90,38,,,653,588,3.6400000000,60,12,21,4365,1441,135,503,684,117,177,0.98,Detroit Tigers,Tiger Stadium,1714893,103,103,DET,DET,DET
+1978,AL,KCA,KCR,W,1,162,81,92,70,Y,,N,N,743,5474,1469,305,59,98,498,644,216,84,,,634,550,3.4400000000,53,14,33,4317,1350,108,478,657,150,153,0.97,Kansas City Royals,Royals Stadium,2255493,103,102,KCR,KCA,KCA
+1978,AL,MIN,MIN,W,4,162,81,73,89,N,,N,N,666,5522,1472,259,47,82,604,684,99,56,,,678,598,3.6900000000,48,9,26,4377,1468,102,520,703,146,171,0.97,Minnesota Twins,Metropolitan Stadium,787878,102,102,MIN,MIN,MIN
+1978,AL,ML4,MIL,E,3,162,81,93,69,N,,N,N,804,5536,1530,265,38,173,520,805,95,53,,,650,582,3.6500000000,62,19,24,4308,1442,109,398,577,150,144,0.97,Milwaukee Brewers,County Stadium,1601406,100,100,MIL,MIL,MIL
+1978,AL,NYA,NYY,E,1,163,81,100,63,Y,,Y,Y,735,5583,1489,228,38,125,505,695,98,42,,,582,516,3.1800000000,39,16,36,4380,1321,111,478,817,113,134,0.98,New York Yankees,Yankee Stadium II,2335871,97,96,NYY,NYA,NYA
+1978,AL,OAK,OAK,W,6,162,80,69,93,N,,N,N,532,5321,1304,200,31,100,433,800,144,117,,,690,576,3.6200000000,26,11,29,4299,1401,106,582,750,179,145,0.97,Oakland Athletics,Oakland Coliseum,526999,94,96,OAK,OAK,OAK
+1978,AL,SEA,SEA,W,7,160,81,56,104,N,,N,N,614,5358,1327,229,37,97,522,702,123,47,,,834,736,4.6700000000,28,4,20,4257,1540,155,567,630,141,174,0.97,Seattle Mariners,Kingdome,877440,100,101,SEA,SEA,SEA
+1978,AL,TEX,TEX,W,2,162,82,87,75,N,,N,N,692,5347,1353,216,36,132,624,779,196,91,,,632,544,3.3600000000,54,12,25,4368,1431,108,421,776,153,140,0.97,Texas Rangers,Arlington Stadium,1447963,100,100,TEX,TEX,TEX
+1978,AL,TOR,TOR,E,7,161,81,59,102,N,,N,N,590,5430,1358,217,39,98,448,645,28,52,,,775,721,4.5400000000,35,5,23,4287,1529,149,614,758,130,163,0.97,Toronto Blue Jays,Exhibition Stadium,1562585,101,104,TOR,TOR,TOR
+1978,NL,ATL,ATL,W,6,162,81,69,93,N,,N,N,600,5381,1313,191,39,123,550,874,90,65,,,750,653,4.0800000000,29,12,32,4320,1404,132,624,848,153,126,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,904494,112,114,ATL,ATL,ATL
+1978,NL,CHN,CHC,E,3,162,82,79,83,N,,N,N,664,5532,1461,224,48,72,562,746,110,58,,,724,655,4.0500000000,24,7,38,4365,1475,125,539,768,144,154,0.97,Chicago Cubs,Wrigley Field,1525311,112,112,CHC,CHN,CHN
+1978,NL,CIN,CIN,W,2,161,80,92,69,N,,N,N,710,5392,1378,270,32,136,636,899,137,58,,,688,613,3.8100000000,16,10,46,4344,1437,122,567,908,134,120,0.97,Cincinnati Reds,Riverfront Stadium,2532497,100,100,CIN,CIN,CIN
+1978,NL,HOU,HOU,W,5,162,81,74,88,N,,N,N,605,5458,1408,231,45,70,434,743,178,59,,,634,581,3.6300000000,48,17,23,4320,1328,86,578,930,132,109,0.97,Houston Astros,Astrodome,1126145,92,92,HOU,HOU,HOU
+1978,NL,LAN,LAD,W,1,162,81,95,67,Y,,Y,N,727,5437,1435,251,27,149,610,818,137,52,,,573,499,3.1200000000,46,16,38,4320,1362,107,440,800,140,138,0.97,Los Angeles Dodgers,Dodger Stadium,3347845,100,98,LAD,LAN,LAN
+1978,NL,MON,WSN,E,4,162,80,76,86,N,,N,N,633,5530,1404,269,31,121,396,881,80,42,,,611,549,3.4200000000,42,13,32,4338,1332,117,572,740,134,150,0.97,Montreal Expos,Stade Olympique,1427007,99,98,MON,MON,MON
+1978,NL,NYN,NYM,E,6,162,80,66,96,N,,N,N,607,5433,1332,227,47,86,549,829,100,77,,,690,626,3.8700000000,21,7,26,4365,1447,114,531,775,132,160,0.97,New York Mets,Shea Stadium,1007328,96,97,NYM,NYN,NYN
+1978,NL,PHI,PHI,E,1,162,82,90,72,Y,,N,N,708,5448,1404,248,32,133,552,866,152,58,,,586,531,3.3300000000,38,9,29,4308,1343,118,393,813,104,156,0.98,Philadelphia Phillies,Veterans Stadium,2583389,101,100,PHI,PHI,PHI
+1978,NL,PIT,PIT,E,2,161,81,88,73,N,,N,N,684,5406,1390,239,54,115,480,874,213,90,,,637,547,3.4100000000,30,13,44,4332,1366,103,499,880,166,133,0.97,Pittsburgh Pirates,Three Rivers Stadium,964106,105,104,PIT,PIT,PIT
+1978,NL,SDN,SDP,W,4,162,81,84,78,N,,N,N,591,5360,1349,208,42,75,536,848,152,70,,,598,522,3.2800000000,21,10,55,4299,1385,74,483,744,160,171,0.97,San Diego Padres,Jack Murphy Stadium,1670107,91,92,SDP,SDN,SDN
+1978,NL,SFN,SFG,W,3,162,81,89,73,N,,N,N,613,5364,1331,240,41,117,554,814,87,54,,,594,534,3.3000000000,42,17,29,4365,1377,84,453,840,146,118,0.97,San Francisco Giants,Candlestick Park,1740477,96,96,SFG,SFN,SFN
+1978,NL,SLN,STL,E,5,162,81,69,93,N,,N,N,600,5415,1351,263,44,79,420,713,97,42,,,657,572,3.5800000000,32,13,22,4311,1300,94,600,859,136,155,0.97,St. Louis Cardinals,Busch Stadium II,1278215,99,99,STL,SLN,SLN
+1979,AL,BAL,BAL,E,1,159,79,102,57,Y,,Y,N,757,5371,1401,258,24,181,608,847,99,49,,,582,519,3.2600000000,52,12,30,4302,1279,133,467,786,125,161,0.98,Baltimore Orioles,Memorial Stadium,1681009,97,95,BAL,BAL,BAL
+1979,AL,BOS,BOS,E,3,160,80,91,69,N,,N,N,841,5538,1567,310,34,194,512,708,60,43,,,711,641,4.0300000000,47,11,29,4293,1487,133,463,731,142,166,0.97,Boston Red Sox,Fenway Park II,2353114,106,106,BOS,BOS,BOS
+1979,AL,CAL,ANA,W,1,162,81,88,74,Y,,N,N,866,5550,1563,242,43,164,589,843,100,53,,,768,692,4.3400000000,46,9,33,4308,1463,131,573,820,135,172,0.97,California Angels,Anaheim Stadium,2523575,96,96,CAL,CAL,CAL
+1979,AL,CHA,CHW,W,5,160,79,73,87,N,,N,N,730,5463,1505,290,33,127,454,668,97,62,,,748,642,4.1000000000,28,9,37,4227,1365,114,618,675,173,142,0.97,Chicago White Sox,Comiskey Park,1280702,100,101,CHW,CHA,CHA
+1979,AL,CLE,CLE,E,6,161,81,81,80,N,,N,N,760,5376,1388,206,29,138,657,786,143,90,,,805,727,4.5700000000,28,7,32,4293,1502,138,570,781,126,149,0.97,Cleveland Indians,Cleveland Stadium,1011644,101,101,CLE,CLE,CLE
+1979,AL,DET,DET,E,5,161,80,85,76,N,,N,N,770,5375,1446,221,35,164,575,814,176,86,,,738,675,4.2700000000,25,5,37,4269,1429,167,547,802,120,184,0.98,Detroit Tigers,Tiger Stadium,1630929,103,103,DET,DET,DET
+1979,AL,KCA,KCR,W,2,162,81,85,77,N,,N,N,851,5653,1596,286,79,116,528,675,207,76,,,816,716,4.4500000000,42,7,27,4344,1477,165,536,640,146,160,0.97,Kansas City Royals,Royals Stadium,2261845,102,101,KCR,KCA,KCA
+1979,AL,MIN,MIN,W,4,162,81,82,80,N,,N,N,764,5544,1544,256,46,112,526,693,66,45,,,725,663,4.1300000000,31,6,33,4332,1590,128,452,721,134,203,0.97,Minnesota Twins,Metropolitan Stadium,1070521,104,104,MIN,MIN,MIN
+1979,AL,ML4,MIL,E,2,161,81,95,66,N,,N,N,807,5536,1552,291,41,185,549,745,100,53,,,722,644,4.0300000000,61,12,23,4317,1563,162,381,580,127,153,0.98,Milwaukee Brewers,County Stadium,1918343,100,99,MIL,MIL,MIL
+1979,AL,NYA,NYY,E,4,160,81,89,71,N,,N,N,734,5421,1443,226,40,150,509,590,65,46,,,672,609,3.8300000000,43,10,37,4296,1446,123,455,731,122,183,0.98,New York Yankees,Yankee Stadium II,2537765,98,96,NYY,NYA,NYA
+1979,AL,OAK,OAK,W,7,162,81,54,108,N,,N,N,573,5348,1276,188,32,108,482,751,104,69,,,860,754,4.7500000000,41,4,20,4287,1606,147,654,726,174,137,0.97,Oakland Athletics,Oakland Coliseum,306763,94,96,OAK,OAK,OAK
+1979,AL,SEA,SEA,W,6,162,81,67,95,N,,N,N,711,5544,1490,250,52,132,515,725,126,52,,,820,732,4.5800000000,37,7,26,4314,1567,165,571,736,141,170,0.97,Seattle Mariners,Kingdome,844447,102,104,SEA,SEA,SEA
+1979,AL,TEX,TEX,W,3,162,81,83,79,N,,N,N,750,5562,1549,252,26,140,461,607,79,51,,,698,616,3.8600000000,26,10,42,4311,1371,135,532,773,130,151,0.97,Texas Rangers,Arlington Stadium,1519671,99,98,TEX,TEX,TEX
+1979,AL,TOR,TOR,E,7,162,81,53,109,N,,N,N,613,5423,1362,253,34,95,448,663,75,56,,,862,759,4.8200000000,44,7,11,4251,1537,165,594,613,158,187,0.97,Toronto Blue Jays,Exhibition Stadium,1431651,100,102,TOR,TOR,TOR
+1979,NL,ATL,ATL,W,6,160,79,66,94,N,,N,N,669,5422,1389,220,28,126,490,818,98,50,,,763,653,4.1800000000,32,3,34,4221,1496,132,494,779,183,139,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,769465,107,109,ATL,ATL,ATL
+1979,NL,CHN,CHC,E,5,162,81,80,82,N,,N,N,706,5550,1494,250,43,135,478,762,73,52,,,707,623,3.8800000000,20,11,44,4338,1500,127,521,933,159,163,0.97,Chicago Cubs,Wrigley Field,1648587,110,111,CHC,CHN,CHN
+1979,NL,CIN,CIN,W,1,161,80,90,71,Y,,N,N,731,5477,1445,266,31,132,614,902,99,47,,,644,573,3.5800000000,27,10,40,4320,1415,103,485,773,124,152,0.98,Cincinnati Reds,Riverfront Stadium,2356933,101,101,CIN,CIN,CIN
+1979,NL,HOU,HOU,W,2,162,81,89,73,N,,N,N,583,5394,1382,224,52,49,461,745,190,95,,,582,514,3.2000000000,55,19,31,4341,1278,90,504,854,137,146,0.97,Houston Astros,Astrodome,1900312,94,94,HOU,HOU,HOU
+1979,NL,LAN,LAD,W,3,162,81,79,83,N,,N,N,739,5490,1443,220,24,183,556,834,106,46,,,717,615,3.8300000000,30,6,34,4332,1425,101,555,811,118,123,0.98,Los Angeles Dodgers,Dodger Stadium,2860954,99,98,LAD,LAN,LAN
+1979,NL,MON,WSN,E,2,160,81,95,65,N,,N,N,701,5465,1445,273,42,143,432,890,121,56,,,581,505,3.1400000000,33,18,39,4341,1379,116,450,813,131,123,0.97,Montreal Expos,Stade Olympique,2102173,99,98,MON,MON,MON
+1979,NL,NYN,NYM,E,6,163,82,63,99,N,,N,N,593,5591,1399,255,41,74,498,817,135,79,,,706,632,3.8400000000,16,10,36,4446,1486,120,607,819,140,168,0.97,New York Mets,Shea Stadium,788905,96,97,NYM,NYN,NYN
+1979,NL,PHI,PHI,E,4,163,81,84,78,N,,N,N,683,5463,1453,250,53,119,602,764,128,76,,,718,666,4.1600000000,33,14,29,4323,1455,135,477,787,106,148,0.98,Philadelphia Phillies,Veterans Stadium,2775011,103,102,PHI,PHI,PHI
+1979,NL,PIT,PIT,E,1,163,81,98,64,Y,,Y,Y,775,5661,1541,264,52,148,483,855,180,66,,,643,566,3.4100000000,24,7,52,4479,1424,125,504,904,134,163,0.97,Pittsburgh Pirates,Three Rivers Stadium,1435454,105,105,PIT,PIT,PIT
+1979,NL,SDN,SDP,W,5,161,81,68,93,N,,N,N,603,5446,1316,193,53,93,534,770,100,58,,,681,596,3.6900000000,29,7,25,4359,1438,108,513,779,140,154,0.97,San Diego Padres,Jack Murphy Stadium,1456967,93,94,SDP,SDN,SDN
+1979,NL,SFN,SFG,W,4,162,81,71,91,N,,N,N,672,5395,1328,192,36,125,580,925,140,73,,,751,664,4.1600000000,25,6,34,4308,1484,143,577,880,161,138,0.97,San Francisco Giants,Candlestick Park,1456402,93,94,SFG,SFN,SFN
+1979,NL,SLN,STL,E,3,163,82,86,76,N,,N,N,731,5734,1594,279,63,100,460,838,116,69,,,693,614,3.7200000000,38,10,25,4458,1449,127,501,788,132,166,0.98,St. Louis Cardinals,Busch Stadium II,1627256,102,102,STL,SLN,SLN
+1980,AL,BAL,BAL,E,2,162,81,100,62,N,,N,N,805,5585,1523,258,29,156,587,766,111,38,,,640,590,3.6400000000,42,10,41,4380,1438,134,507,789,95,178,0.98,Baltimore Orioles,Memorial Stadium,1797438,99,98,BAL,BAL,BAL
+1980,AL,BOS,BOS,E,5,160,81,83,77,N,,N,N,757,5603,1588,297,36,162,475,720,79,48,,,767,701,4.3800000000,30,8,43,4323,1557,129,481,696,149,206,0.97,Boston Red Sox,Fenway Park II,1956092,106,105,BOS,BOS,BOS
+1980,AL,CAL,ANA,W,6,160,81,65,95,N,,N,N,698,5443,1442,236,32,106,539,889,91,63,,,797,717,4.5200000000,22,6,30,4284,1548,141,529,725,134,144,0.97,California Angels,Anaheim Stadium,2297327,97,97,CAL,CAL,CAL
+1980,AL,CHA,CHW,W,5,162,81,70,90,N,,N,N,587,5444,1408,255,38,91,399,670,68,54,,,722,625,3.9200000000,32,12,42,4305,1434,108,563,724,171,162,0.97,Chicago White Sox,Comiskey Park,1200365,100,100,CHW,CHA,CHA
+1980,AL,CLE,CLE,E,6,160,79,79,81,N,,N,N,738,5470,1517,221,40,89,617,625,118,58,,,807,743,4.6800000000,35,8,32,4284,1519,137,552,843,105,143,0.98,Cleveland Indians,Cleveland Stadium,1033827,101,102,CLE,CLE,CLE
+1980,AL,DET,DET,E,4,163,82,84,78,N,,N,N,830,5648,1543,232,53,143,645,844,75,68,,,757,693,4.2500000000,40,9,30,4401,1505,152,558,741,129,165,0.97,Detroit Tigers,Tiger Stadium,1785293,102,102,DET,DET,DET
+1980,AL,KCA,KCR,W,1,162,81,97,65,Y,,Y,N,809,5714,1633,266,59,115,508,709,185,43,,,694,621,3.8300000000,37,10,42,4377,1496,129,465,614,141,150,0.97,Kansas City Royals,Royals Stadium,2288714,100,100,KCR,KCA,KCA
+1980,AL,MIN,MIN,W,3,161,80,77,84,N,,N,N,670,5530,1468,252,46,99,436,703,62,46,,,724,634,3.9300000000,35,9,30,4353,1502,120,468,744,146,192,0.97,Minnesota Twins,Metropolitan Stadium,769206,108,108,MIN,MIN,MIN
+1980,AL,ML4,MIL,E,3,162,82,86,76,N,,N,N,811,5653,1555,298,36,203,455,745,131,56,,,682,598,3.7100000000,48,14,30,4350,1530,137,420,575,147,189,0.97,Milwaukee Brewers,County Stadium,1857408,97,96,MIL,MIL,MIL
+1980,AL,NYA,NYY,E,1,162,81,103,59,Y,,N,N,820,5553,1484,239,34,189,643,739,86,36,,,662,582,3.5800000000,29,15,50,4392,1433,102,463,845,138,160,0.97,New York Yankees,Yankee Stadium II,2627417,98,97,NYY,NYA,NYA
+1980,AL,OAK,OAK,W,2,162,81,83,79,N,,N,N,686,5495,1424,212,35,137,506,824,175,82,,,642,566,3.4600000000,94,9,13,4413,1347,142,521,769,130,115,0.97,Oakland Athletics,Oakland Coliseum,842259,93,93,OAK,OAK,OAK
+1980,AL,SEA,SEA,W,7,163,81,59,103,N,,N,N,610,5489,1359,211,35,104,483,727,116,62,,,793,709,4.3800000000,31,7,26,4371,1565,159,540,703,149,189,0.97,Seattle Mariners,Kingdome,836204,101,103,SEA,SEA,SEA
+1980,AL,TEX,TEX,W,4,163,80,76,85,N,,N,N,756,5690,1616,263,27,124,480,589,91,49,,,752,648,4.0200000000,35,6,25,4353,1561,119,519,890,147,169,0.97,Texas Rangers,Arlington Stadium,1198175,97,96,TEX,TEX,TEX
+1980,AL,TOR,TOR,E,7,162,81,67,95,N,,N,N,624,5571,1398,249,53,126,448,813,67,72,,,762,683,4.1900000000,39,9,23,4398,1523,135,635,705,133,206,0.97,Toronto Blue Jays,Exhibition Stadium,1400327,105,107,TOR,TOR,TOR
+1980,NL,ATL,ATL,W,4,161,80,81,80,N,,N,N,630,5402,1352,226,22,144,434,899,73,52,,,660,598,3.7700000000,29,9,37,4284,1397,131,454,696,162,156,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1048411,103,103,ATL,ATL,ATL
+1980,NL,CHN,CHC,E,6,162,81,64,98,N,,N,N,614,5619,1411,251,35,107,471,912,93,64,,,728,639,3.8900000000,13,6,35,4437,1525,109,589,923,174,149,0.97,Chicago Cubs,Wrigley Field,1206776,107,109,CHC,CHN,CHN
+1980,NL,CIN,CIN,W,3,163,82,89,73,N,,N,N,707,5516,1445,256,45,113,537,852,156,43,,,670,624,3.8500000000,30,12,37,4377,1404,113,506,833,106,144,0.98,Cincinnati Reds,Riverfront Stadium,2022450,101,100,CIN,CIN,CIN
+1980,NL,HOU,HOU,W,1,163,81,93,70,Y,,N,N,637,5566,1455,231,67,75,540,755,194,74,,,589,510,3.1000000000,31,18,41,4446,1367,69,466,929,140,145,0.97,Houston Astros,Astrodome,2278217,92,91,HOU,HOU,HOU
+1980,NL,LAN,LAD,W,2,163,82,92,71,N,,N,N,663,5568,1462,209,24,148,492,846,123,72,,,591,532,3.2500000000,24,19,42,4416,1358,105,480,835,123,149,0.98,Los Angeles Dodgers,Dodger Stadium,3249287,99,98,LAD,LAN,LAN
+1980,NL,MON,WSN,E,2,162,80,90,72,N,,N,N,694,5465,1407,250,61,114,547,865,237,82,,,629,563,3.4800000000,33,15,36,4368,1447,100,460,823,144,126,0.97,Montreal Expos,Stade Olympique,2208175,100,99,MON,MON,MON
+1980,NL,NYN,NYM,E,5,162,82,67,95,N,,N,N,611,5478,1407,218,41,61,501,840,158,99,,,702,621,3.8500000000,17,9,33,4353,1473,140,510,886,154,132,0.97,New York Mets,Shea Stadium,1192073,97,98,NYM,NYN,NYN
+1980,NL,PHI,PHI,E,1,162,81,91,71,Y,,Y,Y,728,5625,1517,272,54,117,472,708,140,62,,,639,564,3.4300000000,25,8,40,4440,1419,87,530,889,136,136,0.97,Philadelphia Phillies,Veterans Stadium,2651650,105,105,PHI,PHI,PHI
+1980,NL,PIT,PIT,E,3,162,81,83,79,N,,N,N,666,5517,1469,249,38,116,452,760,209,102,,,646,580,3.5800000000,25,8,43,4374,1422,110,451,832,137,154,0.97,Pittsburgh Pirates,Three Rivers Stadium,1646757,103,102,PIT,PIT,PIT
+1980,NL,SDN,SDP,W,6,163,81,73,89,N,,N,N,591,5540,1410,195,43,67,563,791,239,73,,,654,595,3.6500000000,19,9,39,4398,1474,97,536,728,131,157,0.98,San Diego Padres,Jack Murphy Stadium,1139026,94,95,SDP,SDN,SDN
+1980,NL,SFN,SFG,W,5,161,81,75,86,N,,N,N,573,5368,1310,199,44,80,509,840,100,58,,,634,557,3.4600000000,27,10,35,4344,1446,92,492,811,156,124,0.97,San Francisco Giants,Candlestick Park,1096115,97,98,SFG,SFN,SFN
+1980,NL,SLN,STL,E,4,162,81,74,88,N,,N,N,738,5608,1541,300,49,101,451,781,117,54,,,710,632,3.9300000000,34,9,27,4341,1454,90,495,664,122,174,0.98,St. Louis Cardinals,Busch Stadium II,1385147,104,103,STL,SLN,SLN
+1981,AL,BAL,BAL,E,2,105,55,59,46,N,,N,N,429,3516,883,165,11,88,404,454,41,34,,,437,386,3.7000000000,25,10,23,2820,923,83,347,489,68,114,0.98,Baltimore Orioles,Memorial Stadium,1024247,100,99,BAL,BAL,BAL
+1981,AL,BOS,BOS,E,5,108,53,59,49,N,,N,N,519,3820,1052,168,17,90,378,520,32,31,,,481,418,3.8100000000,19,4,24,2961,983,90,354,536,91,108,0.97,Boston Red Sox,Fenway Park II,1060379,107,106,BOS,BOS,BOS
+1981,AL,CAL,ANA,W,5,110,54,51,59,N,,N,N,476,3688,944,134,16,97,393,571,44,33,,,453,399,3.7000000000,27,5,19,2914,958,81,323,426,101,120,0.97,California Angels,Anaheim Stadium,1441545,100,99,CAL,CAL,CAL
+1981,AL,CHA,CHW,W,3,106,49,54,52,N,,N,N,476,3615,982,135,27,76,322,518,86,44,,,423,363,3.4700000000,20,8,23,2822,891,73,336,529,87,113,0.97,Chicago White Sox,Comiskey Park,946651,98,98,CHW,CHA,CHA
+1981,AL,CLE,CLE,E,6,103,54,52,51,N,,N,N,431,3507,922,150,21,39,343,379,119,37,,,442,401,3.8800000000,33,10,13,2793,989,67,311,569,87,91,0.97,Cleveland Indians,Cleveland Stadium,661395,100,100,CLE,CLE,CLE
+1981,AL,DET,DET,E,3,109,55,60,49,N,,N,N,427,3600,922,148,29,65,404,500,61,37,,,404,380,3.5300000000,33,13,22,2907,840,83,373,476,65,109,0.98,Detroit Tigers,Tiger Stadium,1149144,104,103,DET,DET,DET
+1981,AL,KCA,KCR,W,4,103,47,50,53,Y,,N,N,397,3560,952,169,29,61,301,419,100,53,,,405,365,3.5600000000,24,8,24,2767,909,75,273,404,72,94,0.98,Kansas City Royals,Royals Stadium,1279403,99,98,KCR,KCA,KCA
+1981,AL,MIN,MIN,W,7,110,61,41,68,N,,N,N,378,3676,884,147,36,47,275,497,34,27,,,486,433,3.9800000000,13,6,22,2937,1021,79,376,500,93,103,0.97,Minnesota Twins,Metropolitan Stadium,469090,107,108,MIN,MIN,MIN
+1981,AL,ML4,MIL,E,1,109,49,62,47,Y,,N,N,493,3743,961,173,20,96,300,461,39,36,,,459,428,3.9100000000,11,4,35,2958,994,72,352,448,79,135,0.98,Milwaukee Brewers,County Stadium,874292,95,94,MIL,MIL,MIL
+1981,AL,NYA,NYY,E,4,107,51,59,48,Y,,Y,N,421,3529,889,148,22,100,391,434,47,30,,,343,305,2.9000000000,16,13,30,2844,827,64,287,606,72,100,0.98,New York Yankees,Yankee Stadium II,1614353,98,97,NYY,NYA,NYA
+1981,AL,OAK,OAK,W,1,109,56,64,45,Y,,N,N,458,3677,910,119,26,104,342,647,98,47,,,403,364,3.3000000000,60,13,10,2979,883,80,370,505,81,74,0.98,Oakland Athletics,Oakland Coliseum,1304052,95,95,OAK,OAK,OAK
+1981,AL,SEA,SEA,W,6,110,57,44,65,N,,N,N,426,3780,950,148,13,89,329,553,100,50,,,521,469,4.2300000000,10,6,23,2992,1039,76,360,478,91,122,0.97,Seattle Mariners,Kingdome,636276,105,106,SEA,SEA,SEA
+1981,AL,TEX,TEX,W,2,105,56,57,48,N,,N,N,452,3581,968,178,15,49,295,396,46,41,,,389,355,3.4000000000,23,8,18,2820,891,67,322,488,69,102,0.98,Texas Rangers,Arlington Stadium,850076,94,94,TEX,TEX,TEX
+1981,AL,TOR,TOR,E,7,106,53,37,69,N,,N,N,329,3521,797,137,23,61,284,556,66,57,,,466,404,3.8200000000,20,4,18,2859,908,72,377,451,105,102,0.97,Toronto Blue Jays,Exhibition Stadium,755083,106,108,TOR,TOR,TOR
+1981,NL,ATL,ATL,W,5,107,50,50,56,N,,N,N,395,3642,886,148,22,64,321,540,98,39,,,416,371,3.4500000000,11,6,24,2904,936,62,330,471,102,93,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,535418,102,103,ATL,ATL,ATL
+1981,NL,CHN,CHC,E,6,106,58,38,65,N,,N,N,370,3546,838,138,29,57,342,611,72,41,,,483,426,4.0100000000,6,2,20,2868,983,59,388,532,113,103,0.97,Chicago Cubs,Wrigley Field,565637,104,106,CHC,CHN,CHN
+1981,NL,CIN,CIN,W,1,108,54,66,42,N,,N,N,464,3637,972,190,24,64,375,553,58,37,,,440,400,3.7300000000,25,19,20,2897,863,67,393,593,80,99,0.98,Cincinnati Reds,Riverfront Stadium,1093730,102,102,CIN,CIN,CIN
+1981,NL,HOU,HOU,W,3,110,51,61,49,Y,,N,N,394,3693,948,160,35,45,340,488,81,43,,,331,293,2.6600000000,23,9,25,2970,842,40,300,610,87,81,0.98,Houston Astros,Astrodome,1321282,95,94,HOU,HOU,HOU
+1981,NL,LAN,LAD,W,2,110,56,63,47,Y,,Y,Y,450,3751,984,133,20,82,331,550,73,46,,,356,333,3.0100000000,26,19,24,2991,904,54,302,603,87,101,0.98,Los Angeles Dodgers,Dodger Stadium,2381292,97,96,LAD,LAN,LAN
+1981,NL,MON,WSN,E,2,108,56,60,48,Y,,N,N,443,3591,883,146,28,81,368,498,138,40,,,394,357,3.3000000000,20,12,23,2925,902,58,268,520,81,88,0.98,Montreal Expos,Stade Olympique,1534564,101,100,MON,MON,MON
+1981,NL,NYN,NYM,E,5,105,52,41,62,N,,N,N,348,3493,868,136,35,57,304,603,103,42,,,432,365,3.5500000000,7,3,24,2778,906,74,336,490,130,89,0.96,New York Mets,Shea Stadium,704244,98,99,NYM,NYN,NYN
+1981,NL,PHI,PHI,E,3,107,55,59,48,Y,,N,N,491,3665,1002,165,25,69,372,432,103,46,,,472,432,4.0500000000,19,5,23,2880,967,72,347,580,86,90,0.98,Philadelphia Phillies,Veterans Stadium,1638752,104,104,PHI,PHI,PHI
+1981,NL,PIT,PIT,E,4,103,51,46,56,N,,N,N,407,3576,920,176,30,55,278,494,122,52,,,425,373,3.5600000000,11,5,29,2826,953,60,346,492,86,106,0.97,Pittsburgh Pirates,Three Rivers Stadium,541789,104,104,PIT,PIT,PIT
+1981,NL,SDN,SDP,W,6,110,55,41,69,N,,N,N,382,3757,963,170,35,32,311,525,83,62,,,455,414,3.7200000000,9,6,23,3006,1013,64,414,492,102,117,0.97,San Diego Padres,Jack Murphy Stadium,519161,93,94,SDP,SDN,SDN
+1981,NL,SFN,SFG,W,4,111,53,56,55,N,,N,N,427,3766,941,161,26,63,386,543,89,50,,,414,368,3.2800000000,8,4,33,3028,970,57,393,561,99,102,0.97,San Francisco Giants,Candlestick Park,632274,98,98,SFG,SFN,SFN
+1981,NL,SLN,STL,E,1,103,53,59,43,N,,N,N,464,3537,936,158,45,50,379,495,88,45,,,417,380,3.6300000000,11,5,33,2829,902,52,290,388,82,108,0.98,St. Louis Cardinals,Busch Stadium II,1010247,103,102,STL,SLN,SLN
+1982,AL,BAL,BAL,E,2,163,82,94,68,N,,N,N,774,5557,1478,259,27,179,634,796,49,38,,,687,648,3.9900000000,38,8,34,4386,1436,147,488,719,101,140,0.98,Baltimore Orioles,Memorial Stadium,1613031,100,99,BAL,BAL,BAL
+1982,AL,BOS,BOS,E,3,162,81,89,73,N,,N,N,753,5596,1536,271,31,136,547,736,42,39,,,713,651,4.0300000000,23,11,33,4359,1557,155,478,816,121,172,0.98,Boston Red Sox,Fenway Park II,1950124,107,106,BOS,BOS,BOS
+1982,AL,CAL,ANA,W,1,162,81,93,69,Y,,N,N,814,5532,1518,268,26,186,613,760,55,53,,,670,621,3.8200000000,40,10,27,4392,1436,124,482,728,106,171,0.98,California Angels,Anaheim Stadium,2807360,100,99,CAL,CAL,CAL
+1982,AL,CHA,CHW,W,3,162,80,87,75,N,,N,N,786,5575,1523,266,52,136,533,866,136,58,,,710,619,3.8700000000,30,10,41,4317,1502,99,460,753,154,173,0.97,Chicago White Sox,Comiskey Park,1567787,100,99,CHW,CHA,CHA
+1982,AL,CLE,CLE,E,6,162,81,78,84,N,,N,N,683,5559,1458,225,32,109,651,625,151,68,,,748,670,4.1100000000,31,9,30,4404,1433,122,589,882,123,129,0.98,Cleveland Indians,Cleveland Stadium,1044021,101,101,CLE,CLE,CLE
+1982,AL,DET,DET,E,4,162,81,83,79,N,,N,N,729,5590,1489,237,40,177,470,807,93,66,,,685,613,3.8000000000,45,5,27,4353,1371,172,554,740,115,165,0.98,Detroit Tigers,Tiger Stadium,1636058,100,99,DET,DET,DET
+1982,AL,KCA,KCR,W,2,162,81,90,72,N,,N,N,784,5629,1603,295,58,132,442,758,133,48,,,717,649,4.0800000000,16,12,45,4293,1443,163,471,650,127,140,0.97,Kansas City Royals,Royals Stadium,2284464,100,100,KCR,KCA,KCA
+1982,AL,MIN,MIN,W,7,162,81,60,102,N,,N,N,657,5544,1427,234,44,148,474,887,38,33,,,819,752,4.7200000000,26,7,30,4299,1484,208,643,812,105,162,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,921186,103,104,MIN,MIN,MIN
+1982,AL,ML4,MIL,E,1,163,82,95,67,Y,,Y,N,891,5733,1599,277,41,216,484,714,84,52,,,717,649,3.9800000000,34,6,47,4401,1514,152,511,717,125,185,0.98,Milwaukee Brewers,County Stadium,1978896,94,93,MIL,MIL,MIL
+1982,AL,NYA,NYY,E,5,162,81,79,83,N,,N,N,709,5526,1417,225,37,161,590,719,69,45,,,716,647,3.9900000000,24,8,39,4377,1471,113,491,939,128,158,0.97,New York Yankees,Yankee Stadium II,2041219,98,97,NYY,NYA,NYA
+1982,AL,OAK,OAK,W,5,162,81,68,94,N,,N,N,691,5448,1286,211,27,149,582,948,232,87,,,819,734,4.5400000000,42,6,22,4368,1506,177,648,697,160,140,0.97,Oakland Athletics,Oakland Coliseum,1735489,94,95,OAK,OAK,OAK
+1982,AL,SEA,SEA,W,4,162,81,76,86,N,,N,N,651,5626,1431,259,33,130,456,806,131,82,,,712,636,3.8800000000,23,11,39,4428,1431,173,547,1002,138,158,0.97,Seattle Mariners,Kingdome,1070404,103,105,SEA,SEA,SEA
+1982,AL,TEX,TEX,W,6,162,81,64,98,N,,N,N,590,5445,1354,204,26,115,447,750,63,45,,,749,681,4.2800000000,32,5,24,4293,1554,128,483,690,121,169,0.98,Texas Rangers,Arlington Stadium,1154432,95,95,TEX,TEX,TEX
+1982,AL,TOR,TOR,E,6,162,81,78,84,N,,N,N,651,5526,1447,262,45,106,415,749,118,81,,,701,633,3.9500000000,41,13,25,4329,1428,147,493,776,136,146,0.97,Toronto Blue Jays,Exhibition Stadium,1275978,110,110,TOR,TOR,TOR
+1982,NL,ATL,ATL,W,1,162,81,89,73,Y,,N,N,739,5507,1411,215,22,146,554,869,151,77,,,702,621,3.8200000000,15,11,51,4389,1484,126,502,813,137,186,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1801985,104,104,ATL,ATL,ATL
+1982,NL,CHN,CHC,E,5,162,81,73,89,N,,N,N,676,5531,1436,239,46,102,460,869,132,70,,,709,630,3.9200000000,9,7,43,4341,1510,125,452,764,132,110,0.97,Chicago Cubs,Wrigley Field,1249278,102,103,CHC,CHN,CHN
+1982,NL,CIN,CIN,W,6,162,81,61,101,N,,N,N,545,5479,1375,228,34,82,470,817,131,69,,,661,594,3.6600000000,22,7,31,4380,1414,105,570,998,128,158,0.98,Cincinnati Reds,Riverfront Stadium,1326528,102,102,CIN,CIN,CIN
+1982,NL,HOU,HOU,W,5,162,81,77,85,N,,N,N,569,5440,1342,236,48,74,435,830,140,61,,,620,549,3.4200000000,37,16,31,4338,1338,87,479,899,136,154,0.97,Houston Astros,Astrodome,1558555,92,92,HOU,HOU,HOU
+1982,NL,LAN,LAD,W,2,162,81,88,74,N,,N,N,691,5642,1487,222,32,138,528,804,151,56,,,612,539,3.2600000000,37,16,28,4464,1356,81,468,932,139,131,0.97,Los Angeles Dodgers,Dodger Stadium,3608881,98,97,LAD,LAN,LAN
+1982,NL,MON,WSN,E,3,162,81,86,76,N,,N,N,697,5557,1454,270,38,133,503,816,156,56,,,616,537,3.3100000000,34,10,43,4380,1371,110,448,936,122,117,0.98,Montreal Expos,Stade Olympique,2318292,102,101,MON,MON,MON
+1982,NL,NYN,NYM,E,6,162,81,65,97,N,,N,N,609,5510,1361,227,26,97,456,1005,137,58,,,723,624,3.8800000000,15,5,37,4341,1508,119,582,759,175,134,0.97,New York Mets,Shea Stadium,1323036,99,101,NYM,NYN,NYN
+1982,NL,PHI,PHI,E,2,162,81,89,73,N,,N,N,664,5454,1417,245,25,112,506,831,128,76,,,654,584,3.6100000000,38,13,33,4368,1395,86,472,1002,121,138,0.98,Philadelphia Phillies,Veterans Stadium,2376394,103,102,PHI,PHI,PHI
+1982,NL,PIT,PIT,E,4,162,81,84,78,N,,N,N,724,5614,1535,272,40,134,447,862,161,75,,,696,621,3.8100000000,19,7,39,4398,1434,118,521,933,145,133,0.97,Pittsburgh Pirates,Three Rivers Stadium,1024106,104,104,PIT,PIT,PIT
+1982,NL,SDN,SDP,W,4,162,81,81,81,N,,N,N,675,5575,1435,217,52,81,429,877,165,77,,,658,577,3.5200000000,20,11,41,4428,1348,139,502,765,152,142,0.97,San Diego Padres,Jack Murphy Stadium,1607516,96,96,SDP,SDN,SDN
+1982,NL,SFN,SFG,W,3,162,81,87,75,N,,N,N,673,5499,1393,213,30,133,607,915,130,56,,,687,593,3.6400000000,18,4,45,4395,1507,109,466,810,173,125,0.97,San Francisco Giants,Candlestick Park,1200948,99,99,SFG,SFN,SFN
+1982,NL,SLN,STL,E,1,162,81,92,70,Y,,Y,Y,685,5455,1439,239,52,67,569,805,200,91,,,609,549,3.3700000000,25,10,47,4395,1420,94,502,689,124,169,0.98,St. Louis Cardinals,Busch Stadium II,2111906,101,101,STL,SLN,SLN
+1983,AL,BAL,BAL,E,1,162,81,98,64,Y,,Y,Y,799,5546,1492,283,27,168,601,800,61,33,,,652,586,3.6300000000,36,15,38,4356,1451,130,452,774,121,159,0.98,Baltimore Orioles,Memorial Stadium,2042071,98,98,BAL,BAL,BAL
+1983,AL,BOS,BOS,E,6,162,81,78,84,N,,N,N,724,5590,1512,287,32,142,536,758,30,26,,,775,697,4.3400000000,29,7,42,4338,1572,158,493,767,130,168,0.97,Boston Red Sox,Fenway Park II,1782285,107,107,BOS,BOS,BOS
+1983,AL,CAL,ANA,W,5,162,81,70,92,N,,N,N,722,5640,1467,241,22,154,509,835,41,39,,,779,706,4.3100000000,39,7,23,4422,1636,130,496,668,154,190,0.97,California Angels,Anaheim Stadium,2555016,99,99,CAL,CAL,CAL
+1983,AL,CHA,CHW,W,1,162,81,99,63,Y,,N,N,800,5484,1439,270,42,157,527,888,165,50,,,650,589,3.6700000000,35,12,48,4335,1355,128,447,877,120,158,0.98,Chicago White Sox,Comiskey Park,2132821,104,103,CHW,CHA,CHA
+1983,AL,CLE,CLE,E,7,162,81,70,92,N,,N,N,704,5476,1451,249,31,86,605,691,109,71,,,785,709,4.4300000000,34,8,25,4323,1531,120,529,794,122,174,0.98,Cleveland Indians,Cleveland Stadium,768941,104,105,CLE,CLE,CLE
+1983,AL,DET,DET,E,2,162,81,92,70,N,,N,N,789,5592,1530,283,53,156,508,831,93,53,,,679,613,3.8000000000,42,9,28,4353,1318,170,522,875,124,142,0.98,Detroit Tigers,Tiger Stadium,1829636,97,96,DET,DET,DET
+1983,AL,KCA,KCR,W,2,163,82,79,83,N,,N,N,696,5598,1515,273,54,109,397,722,182,47,,,767,679,4.2500000000,19,8,49,4311,1535,133,471,593,164,178,0.97,Kansas City Royals,Royals Stadium,1963875,100,100,KCR,KCA,KCA
+1983,AL,MIN,MIN,W,5,162,81,70,92,N,,N,N,709,5601,1463,280,41,141,467,802,44,29,,,822,744,4.6600000000,20,5,39,4311,1559,163,580,748,118,170,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,858939,104,105,MIN,MIN,MIN
+1983,AL,ML4,MIL,E,5,162,81,87,75,N,,N,N,764,5620,1556,281,57,132,475,665,101,49,,,708,649,4.0200000000,35,10,43,4362,1513,133,491,689,113,162,0.98,Milwaukee Brewers,County Stadium,2397131,93,92,MIL,MIL,MIL
+1983,AL,NYA,NYY,E,3,162,81,91,71,N,,N,N,770,5631,1535,269,40,153,533,686,84,42,,,703,624,3.8600000000,47,12,32,4368,1449,116,455,892,139,157,0.97,New York Yankees,Yankee Stadium II,2257976,96,95,NYY,NYA,NYA
+1983,AL,OAK,OAK,W,4,162,81,74,88,N,,N,N,708,5516,1447,237,28,121,524,872,235,98,,,782,701,4.3400000000,22,12,33,4362,1462,135,626,719,157,157,0.97,Oakland Athletics,Oakland Coliseum,1294941,93,94,OAK,OAK,OAK
+1983,AL,SEA,SEA,W,7,162,81,60,102,N,,N,N,558,5336,1280,247,31,111,460,840,144,80,,,740,649,4.1200000000,25,9,39,4254,1455,145,544,910,134,159,0.97,Seattle Mariners,Kingdome,813537,104,105,SEA,SEA,SEA
+1983,AL,TEX,TEX,W,3,163,81,77,85,N,,N,N,639,5610,1429,242,33,106,442,767,119,60,,,609,539,3.3100000000,43,11,32,4398,1392,97,471,826,113,151,0.98,Texas Rangers,Arlington Stadium,1363469,97,98,TEX,TEX,TEX
+1983,AL,TOR,TOR,E,4,162,81,89,73,N,,N,N,795,5581,1546,268,58,167,510,810,131,72,,,726,661,4.1200000000,43,8,32,4335,1434,145,517,835,115,148,0.98,Toronto Blue Jays,Exhibition Stadium,1930415,106,106,TOR,TOR,TOR
+1983,NL,ATL,ATL,W,2,162,80,88,74,N,,N,N,746,5472,1489,218,45,130,582,847,146,88,,,640,587,3.6700000000,18,4,48,4320,1412,132,540,895,137,176,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,2119935,108,107,ATL,ATL,ATL
+1983,NL,CHN,CHC,E,5,162,81,71,91,N,,N,N,701,5512,1436,272,42,140,470,868,84,40,,,719,647,4.0800000000,9,10,42,4284,1496,117,498,807,115,164,0.98,Chicago Cubs,Wrigley Field,1479717,104,104,CHC,CHN,CHN
+1983,NL,CIN,CIN,W,6,162,81,74,88,N,,N,N,623,5333,1274,236,35,107,588,1006,154,77,,,710,637,3.9800000000,34,5,29,4323,1365,135,627,934,114,121,0.98,Cincinnati Reds,Riverfront Stadium,1190419,103,104,CIN,CIN,CIN
+1983,NL,HOU,HOU,W,3,162,82,85,77,N,,N,N,643,5502,1412,239,60,97,517,869,164,95,,,646,562,3.4500000000,22,14,48,4398,1276,94,570,904,147,165,0.97,Houston Astros,Astrodome,1351962,93,93,HOU,HOU,HOU
+1983,NL,LAN,LAD,W,1,163,80,91,71,Y,,N,N,654,5440,1358,197,34,146,541,925,166,76,,,609,504,3.1000000000,27,12,40,4392,1336,97,495,1000,168,132,0.97,Los Angeles Dodgers,Dodger Stadium,3510313,99,99,LAD,LAN,LAN
+1983,NL,MON,WSN,E,3,163,81,82,80,N,,N,N,677,5611,1482,297,41,102,509,733,138,44,,,646,585,3.5800000000,38,15,34,4413,1406,120,479,899,116,130,0.98,Montreal Expos,Stade Olympique,2320651,99,99,MON,MON,MON
+1983,NL,NYN,NYM,E,6,162,82,68,94,N,,N,N,575,5444,1314,172,26,112,436,1031,141,64,,,680,593,3.6800000000,18,7,33,4353,1384,97,615,717,151,171,0.97,New York Mets,Shea Stadium,1112774,99,100,NYM,NYN,NYN
+1983,NL,PHI,PHI,E,1,163,82,90,72,Y,,Y,N,696,5426,1352,209,45,125,640,906,143,75,,,635,542,3.3400000000,20,10,41,4383,1429,111,464,1092,152,117,0.97,Philadelphia Phillies,Veterans Stadium,2128339,99,99,PHI,PHI,PHI
+1983,NL,PIT,PIT,E,2,162,81,84,78,N,,N,N,659,5531,1460,238,29,121,497,873,124,77,,,648,577,3.5500000000,25,14,41,4386,1378,109,563,1061,115,165,0.98,Pittsburgh Pirates,Three Rivers Stadium,1225916,103,102,PIT,PIT,PIT
+1983,NL,SDN,SDP,W,4,163,82,81,81,N,,N,N,653,5527,1384,207,34,93,482,822,179,67,,,653,590,3.6200000000,23,5,44,4401,1389,144,528,850,129,135,0.97,San Diego Padres,Jack Murphy Stadium,1539815,97,97,SDP,SDN,SDN
+1983,NL,SFN,SFG,W,5,162,81,79,83,N,,N,N,687,5369,1324,206,30,142,619,990,140,78,,,697,594,3.7000000000,20,9,47,4335,1431,127,520,881,171,109,0.97,San Francisco Giants,Candlestick Park,1251530,97,98,SFG,SFN,SFN
+1983,NL,SLN,STL,E,4,162,81,79,83,N,,N,N,679,5550,1496,262,63,83,543,879,207,89,,,710,615,3.7900000000,22,10,27,4380,1479,115,525,709,152,173,0.97,St. Louis Cardinals,Busch Stadium II,2317914,101,100,STL,SLN,SLN
+1984,AL,BAL,BAL,E,5,162,81,85,77,N,,N,N,681,5456,1374,234,23,160,620,884,51,36,,,667,593,3.7100000000,48,13,32,4317,1393,137,512,714,123,166,0.98,Baltimore Orioles,Memorial Stadium,2045784,98,97,BAL,BAL,BAL
+1984,AL,BOS,BOS,E,4,162,81,86,76,N,,N,N,810,5648,1598,259,45,181,500,842,38,25,,,764,670,4.1800000000,40,12,32,4326,1524,141,517,927,143,128,0.97,Boston Red Sox,Fenway Park II,1661618,105,105,BOS,BOS,BOS
+1984,AL,CAL,ANA,W,2,162,81,81,81,N,,N,N,696,5470,1363,211,30,150,556,928,80,51,,,697,642,3.9600000000,36,12,26,4374,1526,143,474,754,128,170,0.98,California Angels,Anaheim Stadium,2402997,100,100,CAL,CAL,CAL
+1984,AL,CHA,CHW,W,5,162,81,74,88,N,,N,N,679,5513,1360,225,38,172,523,883,109,49,,,736,667,4.1300000000,43,9,32,4362,1416,155,483,840,122,160,0.98,Chicago White Sox,Comiskey Park,2136988,105,104,CHW,CHA,CHA
+1984,AL,CLE,CLE,E,6,163,81,75,87,N,,N,N,761,5643,1498,222,39,123,600,815,126,77,,,766,694,4.2600000000,21,7,35,4401,1523,141,545,803,146,163,0.97,Cleveland Indians,Cleveland Stadium,734079,102,103,CLE,CLE,CLE
+1984,AL,DET,DET,E,1,162,82,104,58,Y,,Y,Y,829,5644,1529,254,46,187,602,941,106,68,,,643,568,3.4900000000,19,8,51,4392,1358,130,489,914,124,162,0.98,Detroit Tigers,Tiger Stadium,2704794,100,98,DET,DET,DET
+1984,AL,KCA,KCR,W,1,162,81,84,78,Y,,N,N,673,5543,1487,269,52,117,400,832,106,64,,,686,629,3.9200000000,18,9,50,4332,1426,136,433,724,129,157,0.98,Kansas City Royals,Royals Stadium,1810018,100,101,KCR,KCA,KCA
+1984,AL,MIN,MIN,W,2,162,81,81,81,N,,N,N,673,5562,1473,259,33,114,437,735,39,30,,,675,615,3.8500000000,32,9,38,4311,1429,159,463,713,120,134,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1598692,104,105,MIN,MIN,MIN
+1984,AL,ML4,MIL,E,7,161,81,67,94,N,,N,N,641,5511,1446,232,36,96,432,673,52,57,,,734,646,4.0600000000,13,7,41,4299,1532,137,480,785,136,156,0.97,Milwaukee Brewers,County Stadium,1608509,96,97,MIL,MIL,MIL
+1984,AL,NYA,NYY,E,3,162,81,87,75,N,,N,N,758,5661,1560,275,32,130,534,673,62,38,,,679,615,3.7800000000,15,12,43,4395,1485,120,518,992,142,177,0.97,New York Yankees,Yankee Stadium II,1821815,96,95,NYY,NYA,NYA
+1984,AL,OAK,OAK,W,4,162,81,77,85,N,,N,N,738,5457,1415,257,29,158,568,871,145,64,,,796,712,4.4800000000,15,6,44,4290,1554,155,592,695,146,159,0.97,Oakland Athletics,Oakland Coliseum,1353281,93,93,OAK,OAK,OAK
+1984,AL,SEA,SEA,W,5,162,81,74,88,N,,N,N,682,5546,1429,244,34,129,519,871,116,62,,,774,691,4.3100000000,26,4,35,4326,1497,138,619,972,126,143,0.97,Seattle Mariners,Kingdome,870372,98,100,SEA,SEA,SEA
+1984,AL,TEX,TEX,W,7,161,80,69,92,N,,N,N,656,5569,1452,227,29,120,420,807,81,50,,,714,625,3.9100000000,38,6,21,4314,1443,148,518,863,138,138,0.97,Texas Rangers,Arlington Stadium,1102471,103,104,TEX,TEX,TEX
+1984,AL,TOR,TOR,E,2,163,81,89,73,N,,N,N,750,5687,1555,275,68,143,460,816,193,67,,,696,628,3.8600000000,34,10,33,4392,1433,140,528,875,123,166,0.98,Toronto Blue Jays,Exhibition Stadium,2110009,104,103,TOR,TOR,TOR
+1984,NL,ATL,ATL,W,2,162,81,80,82,N,,N,N,632,5422,1338,234,27,111,555,896,140,85,,,655,574,3.5700000000,17,7,49,4341,1401,122,525,859,139,153,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1724892,107,107,ATL,ATL,ATL
+1984,NL,CHN,CHC,E,1,161,80,96,65,Y,,N,N,762,5437,1415,240,47,136,567,967,154,66,,,658,598,3.7500000000,19,8,50,4302,1458,99,442,879,121,137,0.98,Chicago Cubs,Wrigley Field,2107655,108,108,CHC,CHN,CHN
+1984,NL,CIN,CIN,W,5,162,81,70,92,N,,N,N,627,5498,1342,238,30,106,566,978,160,63,,,747,675,4.1600000000,25,6,25,4383,1445,128,578,946,139,116,0.97,Cincinnati Reds,Riverfront Stadium,1275887,104,105,CIN,CIN,CIN
+1984,NL,HOU,HOU,W,2,162,81,80,82,N,,N,N,693,5548,1465,222,67,79,494,837,105,61,,,630,535,3.3200000000,24,13,29,4347,1350,91,502,950,133,160,0.97,Houston Astros,Astrodome,1229862,92,92,HOU,HOU,HOU
+1984,NL,LAN,LAD,W,4,162,81,79,83,N,,N,N,580,5399,1316,213,23,102,488,829,109,69,,,600,514,3.1700000000,39,16,27,4380,1381,76,499,1033,163,146,0.97,Los Angeles Dodgers,Dodger Stadium,3134824,98,98,LAD,LAN,LAN
+1984,NL,MON,WSN,E,5,161,81,78,83,N,,N,N,593,5439,1367,242,36,96,470,782,131,38,,,585,526,3.3100000000,19,10,48,4293,1333,114,474,861,132,147,0.97,Montreal Expos,Stade Olympique,1606531,96,96,MON,MON,MON
+1984,NL,NYN,NYM,E,2,162,81,90,72,N,,N,N,652,5438,1400,235,25,107,500,1001,149,54,,,676,577,3.6000000000,12,15,50,4326,1371,104,573,1028,129,154,0.97,New York Mets,Shea Stadium,1842695,99,99,NYM,NYN,NYN
+1984,NL,PHI,PHI,E,4,162,81,81,81,N,,N,N,720,5614,1494,248,51,147,555,1084,186,60,,,690,586,3.6200000000,11,6,35,4374,1416,101,448,904,161,112,0.97,Philadelphia Phillies,Veterans Stadium,2062693,102,101,PHI,PHI,PHI
+1984,NL,PIT,PIT,E,6,162,81,75,87,N,,N,N,615,5537,1412,237,33,98,438,841,96,62,,,567,508,3.1100000000,27,13,34,4410,1344,102,502,995,128,142,0.98,Pittsburgh Pirates,Three Rivers Stadium,773500,101,101,PIT,PIT,PIT
+1984,NL,SDN,SDP,W,1,162,81,92,70,Y,,Y,N,686,5504,1425,207,42,109,472,810,152,68,,,634,565,3.4800000000,13,17,44,4380,1327,122,563,812,138,144,0.97,San Diego Padres,Jack Murphy Stadium,1983904,100,100,SDP,SDN,SDN
+1984,NL,SFN,SFG,W,6,162,81,66,96,N,,N,N,682,5650,1499,229,26,112,528,980,126,76,,,807,713,4.3900000000,9,7,38,4383,1589,125,549,854,169,134,0.97,San Francisco Giants,Candlestick Park,1001545,97,98,SFG,SFN,SFN
+1984,NL,SLN,STL,E,3,162,81,84,78,N,,N,N,652,5433,1369,225,44,75,516,924,220,71,,,645,576,3.5800000000,19,12,51,4347,1427,94,494,808,118,184,0.98,St. Louis Cardinals,Busch Stadium II,2037448,98,98,STL,SLN,SLN
+1985,AL,BAL,BAL,E,4,161,81,83,78,N,,N,N,818,5517,1451,234,22,214,604,908,69,43,,,764,694,4.3800000000,32,6,33,4281,1480,160,568,793,115,168,0.98,Baltimore Orioles,Memorial Stadium,2132387,97,97,BAL,BAL,BAL
+1985,AL,BOS,BOS,E,5,163,81,81,81,N,,N,N,800,5720,1615,292,31,162,562,816,66,27,,,720,659,4.0600000000,35,8,29,4383,1487,130,540,913,145,161,0.97,Boston Red Sox,Fenway Park II,1786633,104,104,BOS,BOS,BOS
+1985,AL,CAL,ANA,W,2,162,79,90,72,N,,N,N,732,5442,1364,215,31,153,648,902,106,51,,,703,633,3.9100000000,22,8,41,4371,1453,171,514,767,112,202,0.98,California Angels,Anaheim Stadium,2567427,100,100,CAL,CAL,CAL
+1985,AL,CHA,CHW,W,3,163,81,85,77,N,,N,N,736,5470,1386,247,37,146,471,843,108,56,,,720,656,4.0700000000,20,8,39,4353,1411,161,569,1023,111,152,0.98,Chicago White Sox,Comiskey Park,1669888,104,104,CHW,CHA,CHA
+1985,AL,CLE,CLE,E,7,162,81,60,102,N,,N,N,729,5527,1465,254,31,116,492,817,132,72,,,861,775,4.9100000000,24,7,28,4263,1556,170,547,702,141,161,0.97,Cleveland Indians,Cleveland Stadium,655181,99,100,CLE,CLE,CLE
+1985,AL,DET,DET,E,3,161,81,84,77,N,,N,N,729,5575,1413,254,45,202,526,926,75,41,,,688,612,3.7800000000,31,11,40,4368,1313,141,556,943,141,152,0.97,Detroit Tigers,Tiger Stadium,2286609,100,98,DET,DET,DET
+1985,AL,KCA,KCR,W,1,162,82,91,71,Y,,Y,Y,687,5500,1384,261,49,154,473,840,128,48,,,639,567,3.4900000000,27,11,41,4383,1433,103,463,846,127,160,0.98,Kansas City Royals,Royals Stadium,2162717,100,100,KCR,KCA,KCA
+1985,AL,MIN,MIN,W,4,162,84,77,85,N,,N,N,705,5509,1453,282,41,141,502,779,68,44,,,782,710,4.4800000000,41,7,34,4278,1468,164,462,767,120,139,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1651814,105,105,MIN,MIN,MIN
+1985,AL,ML4,MIL,E,6,161,80,71,90,N,,N,N,690,5568,1467,250,44,101,462,746,69,34,,,802,701,4.3900000000,34,5,37,4311,1510,175,499,777,142,153,0.97,Milwaukee Brewers,County Stadium,1360265,100,101,MIL,MIL,MIL
+1985,AL,NYA,NYY,E,2,161,80,97,64,N,,N,N,839,5458,1458,272,31,176,620,771,155,53,,,660,590,3.6900000000,25,9,49,4320,1373,157,518,907,123,172,0.97,New York Yankees,Yankee Stadium II,2214587,98,97,NYY,NYA,NYA
+1985,AL,OAK,OAK,W,4,162,79,77,85,N,,N,N,757,5581,1475,230,34,155,508,861,117,58,,,787,712,4.4100000000,10,6,41,4359,1451,172,607,785,140,137,0.97,Oakland Athletics,Oakland Coliseum,1334599,92,92,OAK,OAK,OAK
+1985,AL,SEA,SEA,W,6,162,83,74,88,N,,N,N,719,5521,1410,277,38,171,564,942,94,35,,,818,745,4.6800000000,23,8,30,4296,1456,154,637,868,122,156,0.98,Seattle Mariners,Kingdome,1128696,100,101,SEA,SEA,SEA
+1985,AL,TEX,TEX,W,7,161,80,62,99,N,,N,N,617,5361,1359,213,41,129,530,819,130,76,,,785,715,4.5600000000,18,5,33,4233,1479,173,501,863,119,145,0.98,Texas Rangers,Arlington Stadium,1112497,101,102,TEX,TEX,TEX
+1985,AL,TOR,TOR,E,1,161,80,99,62,Y,,N,N,759,5508,1482,281,53,158,503,807,144,77,,,588,533,3.3100000000,18,9,47,4344,1312,147,484,823,125,164,0.98,Toronto Blue Jays,Exhibition Stadium,2468925,104,103,TOR,TOR,TOR
+1985,NL,ATL,ATL,W,5,162,81,66,96,N,,N,N,632,5526,1359,213,28,126,553,849,72,52,,,781,678,4.1900000000,9,9,29,4371,1512,134,642,776,159,197,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1350137,105,106,ATL,ATL,ATL
+1985,NL,CHN,CHC,E,4,162,81,77,84,N,,N,N,686,5492,1397,239,28,150,562,937,182,49,,,729,667,4.1600000000,20,8,42,4326,1492,156,519,820,134,150,0.97,Chicago Cubs,Wrigley Field,2161534,110,110,CHC,CHN,CHN
+1985,NL,CIN,CIN,W,2,162,81,89,72,N,,N,N,677,5431,1385,249,34,114,576,856,159,70,,,666,598,3.7100000000,24,11,45,4353,1347,131,535,910,121,142,0.98,Cincinnati Reds,Riverfront Stadium,1834619,104,105,CIN,CIN,CIN
+1985,NL,HOU,HOU,W,3,162,81,83,79,N,,N,N,706,5582,1457,261,42,121,477,873,96,56,,,691,593,3.6600000000,17,9,42,4374,1393,119,543,909,152,159,0.97,Houston Astros,Astrodome,1184314,97,96,HOU,HOU,HOU
+1985,NL,LAN,LAD,W,1,162,81,95,67,Y,,N,N,682,5502,1434,226,28,129,539,846,136,58,,,579,482,2.9600000000,37,21,36,4395,1280,102,462,979,166,131,0.97,Los Angeles Dodgers,Dodger Stadium,3264593,97,96,LAD,LAN,LAN
+1985,NL,MON,WSN,E,3,161,81,84,77,N,,N,N,633,5429,1342,242,49,118,492,880,169,77,,,636,575,3.5500000000,13,13,53,4371,1346,99,509,870,121,152,0.98,Montreal Expos,Stade Olympique,1502494,95,95,MON,MON,MON
+1985,NL,NYN,NYM,E,2,162,81,98,64,N,,N,N,695,5549,1425,239,35,134,546,872,117,53,,,568,514,3.1100000000,32,19,37,4464,1306,111,515,1039,115,138,0.98,New York Mets,Shea Stadium,2761601,99,97,NYM,NYN,NYN
+1985,NL,PHI,PHI,E,5,162,81,75,87,N,,N,N,667,5477,1343,238,47,141,527,1095,122,51,,,673,592,3.6800000000,24,9,30,4341,1424,115,596,899,139,142,0.97,Philadelphia Phillies,Veterans Stadium,1830350,103,103,PHI,PHI,PHI
+1985,NL,PIT,PIT,E,6,161,80,57,104,N,,N,N,568,5436,1340,251,28,80,514,842,110,60,,,708,637,3.9700000000,15,6,29,4335,1406,107,584,962,133,127,0.97,Pittsburgh Pirates,Three Rivers Stadium,735900,99,100,PIT,PIT,PIT
+1985,NL,SDN,SDP,W,3,162,81,83,79,N,,N,N,650,5507,1405,241,28,109,513,809,60,39,,,622,548,3.4000000000,26,19,44,4353,1399,127,443,727,124,158,0.98,San Diego Padres,Jack Murphy Stadium,2210352,99,99,SDP,SDN,SDN
+1985,NL,SFN,SFG,W,6,162,81,62,100,N,,N,N,556,5420,1263,217,31,115,488,962,99,55,,,674,581,3.6100000000,13,5,24,4344,1348,125,572,985,144,134,0.97,San Francisco Giants,Candlestick Park,818697,95,96,SFG,SFN,SFN
+1985,NL,SLN,STL,E,1,162,81,101,61,Y,,Y,N,747,5467,1446,245,59,87,586,853,314,96,,,572,504,3.1000000000,37,20,44,4392,1343,98,453,798,108,166,0.98,St. Louis Cardinals,Busch Stadium II,2637563,100,99,STL,SLN,SLN
+1986,AL,BAL,BAL,E,7,162,79,73,89,N,,N,N,708,5524,1425,223,13,169,563,862,64,34,,,760,686,4.3000000000,17,6,39,4308,1451,177,535,954,135,163,0.97,Baltimore Orioles,Memorial Stadium,1973176,98,99,BAL,BAL,BAL
+1986,AL,BOS,BOS,E,1,161,81,95,66,Y,,Y,N,794,5498,1488,320,21,144,595,707,41,34,,,696,624,3.9300000000,36,6,41,4287,1469,167,474,1033,129,146,0.97,Boston Red Sox,Fenway Park II,2147641,101,100,BOS,BOS,BOS
+1986,AL,CAL,ANA,W,1,162,82,92,70,Y,,N,N,786,5433,1387,236,36,167,671,860,109,42,,,684,621,3.8400000000,29,12,40,4368,1356,153,478,955,107,156,0.98,California Angels,Anaheim Stadium,2655872,99,99,CAL,CAL,CAL
+1986,AL,CHA,CHW,W,5,162,81,72,90,N,,N,N,644,5406,1335,197,34,121,487,940,115,54,,,699,630,3.9300000000,18,8,38,4326,1361,143,561,895,117,142,0.98,Chicago White Sox,Comiskey Park,1424313,104,104,CHW,CHA,CHA
+1986,AL,CLE,CLE,E,5,163,81,84,78,N,,N,N,831,5702,1620,270,45,157,456,944,141,54,,,841,736,4.5800000000,31,7,34,4341,1548,167,605,744,156,148,0.97,Cleveland Indians,Cleveland Stadium,1471805,98,99,CLE,CLE,CLE
+1986,AL,DET,DET,E,3,162,81,87,75,N,,N,N,798,5512,1447,234,30,198,613,885,138,58,,,714,645,4.0200000000,33,12,38,4329,1374,183,571,880,108,163,0.98,Detroit Tigers,Tiger Stadium,1899437,100,99,DET,DET,DET
+1986,AL,KCA,KCR,W,3,162,81,76,86,N,,N,N,654,5561,1403,264,45,137,474,919,97,46,,,673,611,3.8200000000,24,13,31,4320,1413,121,479,888,123,153,0.98,Kansas City Royals,Royals Stadium,2320794,102,101,KCR,KCA,KCA
+1986,AL,MIN,MIN,W,6,162,81,71,91,N,,N,N,741,5531,1446,257,39,196,501,977,81,61,,,839,759,4.7700000000,39,6,24,4296,1579,200,503,937,118,168,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1255453,102,102,MIN,MIN,MIN
+1986,AL,ML4,MIL,E,6,161,80,77,84,N,,N,N,667,5461,1393,255,38,127,530,986,100,50,,,734,638,4.0100000000,29,12,32,4293,1478,158,494,952,146,146,0.97,Milwaukee Brewers,County Stadium,1265041,103,104,MIL,MIL,MIL
+1986,AL,NYA,NYY,E,2,162,80,90,72,N,,N,N,797,5570,1512,275,23,188,645,911,139,48,,,738,659,4.1100000000,13,8,58,4329,1461,175,492,878,127,153,0.97,New York Yankees,Yankee Stadium II,2268030,99,98,NYY,NYA,NYA
+1986,AL,OAK,OAK,W,3,162,83,76,86,N,,N,N,731,5435,1370,213,25,163,553,983,139,61,,,760,686,4.3100000000,22,8,37,4299,1334,166,667,937,135,120,0.97,Oakland Athletics,Oakland Coliseum,1314646,92,92,OAK,OAK,OAK
+1986,AL,SEA,SEA,W,7,162,82,67,95,N,,N,N,718,5498,1392,243,41,158,572,1148,93,76,,,835,743,4.6500000000,33,5,27,4317,1590,171,585,944,156,191,0.97,Seattle Mariners,Kingdome,1029045,100,101,SEA,SEA,SEA
+1986,AL,TEX,TEX,W,2,162,81,87,75,N,,N,N,771,5529,1479,248,43,184,511,1088,103,85,,,743,662,4.1100000000,15,8,41,4350,1356,145,736,1059,120,160,0.98,Texas Rangers,Arlington Stadium,1692002,102,103,TEX,TEX,TEX
+1986,AL,TOR,TOR,E,4,163,81,86,76,N,,N,N,809,5716,1540,285,35,181,496,848,110,59,,,733,669,4.0800000000,16,12,44,4428,1467,164,487,1002,100,150,0.98,Toronto Blue Jays,Exhibition Stadium,2455477,103,102,TOR,TOR,TOR
+1986,NL,ATL,ATL,W,6,161,81,72,89,N,,N,N,615,5384,1348,241,24,138,538,904,93,76,,,719,628,3.9700000000,17,5,39,4272,1443,117,576,932,141,181,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,1387181,105,106,ATL,ATL,ATL
+1986,NL,CHN,CHC,E,5,160,80,70,90,N,,N,N,680,5499,1409,258,27,155,508,966,132,62,,,781,721,4.4900000000,11,6,42,4335,1546,143,557,962,124,147,0.98,Chicago Cubs,Wrigley Field,1859102,107,108,CHC,CHN,CHN
+1986,NL,CIN,CIN,W,2,162,81,86,76,N,,N,N,732,5536,1404,237,35,144,586,920,177,53,,,717,638,3.9100000000,14,8,45,4404,1465,136,524,924,140,160,0.97,Cincinnati Reds,Riverfront Stadium,1692432,104,104,CIN,CIN,CIN
+1986,NL,HOU,HOU,W,1,162,81,96,66,Y,,N,N,654,5441,1388,244,32,125,536,916,163,75,,,569,510,3.1500000000,18,19,51,4368,1203,116,523,1160,130,108,0.97,Houston Astros,Astrodome,1734276,97,96,HOU,HOU,HOU
+1986,NL,LAN,LAD,W,5,162,81,73,89,N,,N,N,638,5471,1373,232,14,130,478,966,155,67,,,679,607,3.7600000000,35,14,25,4362,1428,115,499,1051,179,118,0.97,Los Angeles Dodgers,Dodger Stadium,3023208,93,93,LAD,LAN,LAN
+1986,NL,MON,WSN,E,4,161,80,78,83,N,,N,N,637,5508,1401,255,50,110,537,1016,193,95,,,688,616,3.7800000000,15,9,50,4398,1350,119,566,1051,133,132,0.97,Montreal Expos,Stade Olympique,1128981,100,100,MON,MON,MON
+1986,NL,NYN,NYM,E,1,162,81,108,54,Y,,Y,Y,783,5558,1462,261,31,148,631,968,118,48,,,578,513,3.1100000000,27,11,46,4452,1304,103,509,1083,138,145,0.97,New York Mets,Shea Stadium,2767601,98,96,NYM,NYN,NYN
+1986,NL,PHI,PHI,E,2,161,80,86,75,N,,N,N,739,5483,1386,266,39,154,589,1154,153,59,,,713,621,3.8500000000,22,11,39,4353,1473,130,553,874,137,157,0.97,Philadelphia Phillies,Veterans Stadium,1933335,104,104,PHI,PHI,PHI
+1986,NL,PIT,PIT,E,6,162,81,64,98,N,,N,N,663,5456,1366,273,33,111,569,929,152,84,,,700,628,3.9000000000,17,9,30,4350,1397,138,570,924,142,134,0.97,Pittsburgh Pirates,Three Rivers Stadium,1000917,102,103,PIT,PIT,PIT
+1986,NL,SDN,SDP,W,4,162,81,74,88,N,,N,N,656,5515,1442,239,25,136,484,917,96,68,,,723,640,3.9900000000,13,7,32,4329,1406,150,607,934,135,135,0.97,San Diego Padres,Jack Murphy Stadium,1805716,98,98,SDP,SDN,SDN
+1986,NL,SFN,SFG,W,3,162,81,83,79,N,,N,N,698,5501,1394,269,29,114,536,1087,148,93,,,618,540,3.3300000000,18,10,35,4380,1264,121,591,992,143,149,0.97,San Francisco Giants,Candlestick Park,1528748,95,95,SFG,SFN,SFN
+1986,NL,SLN,STL,E,3,161,81,79,82,N,,N,N,601,5378,1270,216,48,58,568,905,262,78,,,611,549,3.3700000000,17,4,46,4398,1364,135,485,761,123,178,0.98,St. Louis Cardinals,Busch Stadium II,2471974,100,99,STL,SLN,SLN
+1987,AL,BAL,BAL,E,6,162,82,67,95,N,,N,N,729,5576,1437,219,20,211,524,939,69,45,,,880,801,5.0100000000,17,6,30,4317,1555,226,547,870,109,174,0.98,Baltimore Orioles,Memorial Stadium,1835692,97,98,BAL,BAL,BAL
+1987,AL,BOS,BOS,E,5,162,80,78,84,N,,N,N,842,5586,1554,273,26,174,606,825,77,45,,,825,761,4.7700000000,47,13,16,4308,1584,190,517,1034,110,158,0.98,Boston Red Sox,Fenway Park II,2231551,103,102,BOS,BOS,BOS
+1987,AL,CAL,ANA,W,6,162,81,75,87,N,,N,N,770,5570,1406,257,26,172,590,926,125,44,,,803,709,4.3800000000,20,7,36,4371,1481,212,504,941,117,162,0.98,California Angels,Anaheim Stadium,2696299,97,97,CAL,CAL,CAL
+1987,AL,CHA,CHW,W,5,162,81,77,85,N,,N,N,748,5538,1427,283,36,173,487,971,138,52,,,746,691,4.3000000000,29,12,37,4341,1436,189,537,792,115,174,0.98,Chicago White Sox,Comiskey Park,1208060,103,103,CHW,CHA,CHA
+1987,AL,CLE,CLE,E,7,162,81,61,101,N,,N,N,742,5606,1476,267,30,187,489,977,140,54,,,957,834,5.2800000000,24,8,25,4266,1566,219,606,849,152,128,0.97,Cleveland Indians,Cleveland Stadium,1077898,100,101,CLE,CLE,CLE
+1987,AL,DET,DET,E,1,162,81,98,64,Y,,N,N,896,5649,1535,274,32,225,653,913,106,50,,,735,650,4.0200000000,33,10,31,4368,1430,180,563,976,122,147,0.98,Detroit Tigers,Tiger Stadium,2061830,96,95,DET,DET,DET
+1987,AL,KCA,KCR,W,2,162,81,83,79,N,,N,N,715,5499,1443,239,40,168,523,1034,125,43,,,691,611,3.8600000000,44,11,26,4272,1424,128,548,923,131,151,0.97,Kansas City Royals,Royals Stadium,2392471,102,102,KCR,KCA,KCA
+1987,AL,MIN,MIN,W,1,162,81,85,77,Y,,Y,Y,786,5441,1422,258,35,196,523,898,113,65,,,806,734,4.6300000000,16,4,39,4281,1465,210,564,990,98,147,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2081976,103,103,MIN,MIN,MIN
+1987,AL,ML4,MIL,E,3,162,81,91,71,N,,N,N,862,5625,1552,272,46,163,598,1040,176,74,,,817,752,4.6200000000,28,6,45,4392,1548,169,529,1039,145,155,0.97,Milwaukee Brewers,County Stadium,1909244,103,103,MIL,MIL,MIL
+1987,AL,NYA,NYY,E,4,162,81,89,73,N,,N,N,788,5511,1445,239,16,196,604,949,105,43,,,758,701,4.3600000000,19,10,47,4338,1475,179,542,900,102,155,0.98,New York Yankees,Yankee Stadium II,2427672,99,99,NYY,NYA,NYA
+1987,AL,OAK,OAK,W,3,162,81,81,81,N,,N,N,806,5511,1432,263,33,199,593,1056,140,63,,,789,694,4.3200000000,18,6,40,4335,1442,176,531,1042,142,122,0.97,Oakland Athletics,Oakland Coliseum,1678921,93,93,OAK,OAK,OAK
+1987,AL,SEA,SEA,W,4,162,81,78,84,N,,N,N,760,5508,1499,282,48,161,500,863,174,73,,,801,713,4.4900000000,39,10,33,4290,1503,199,497,919,121,150,0.98,Seattle Mariners,Kingdome,1134255,105,106,SEA,SEA,SEA
+1987,AL,TEX,TEX,W,6,162,81,75,87,N,,N,N,823,5564,1478,264,35,194,567,1081,120,71,,,849,743,4.6300000000,20,3,27,4332,1388,199,760,1103,148,148,0.97,Texas Rangers,Arlington Stadium,1763053,100,101,TEX,TEX,TEX
+1987,AL,TOR,TOR,E,2,162,81,96,66,N,,N,N,845,5635,1514,277,38,215,555,970,126,50,,,655,604,3.7400000000,18,8,43,4362,1323,158,567,1064,111,148,0.98,Toronto Blue Jays,Exhibition Stadium,2778429,102,101,TOR,TOR,TOR
+1987,NL,ATL,ATL,W,5,161,81,69,92,N,,N,N,747,5428,1401,284,24,152,641,834,135,68,,,829,734,4.6300000000,16,4,32,4281,1529,163,587,837,116,170,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,1217402,104,106,ATL,ATL,ATL
+1987,NL,CHN,CHC,E,6,161,80,76,85,N,,N,N,720,5583,1475,244,33,209,504,1064,109,48,,,801,725,4.5500000000,11,5,48,4302,1524,159,628,1024,130,154,0.97,Chicago Cubs,Wrigley Field,2035130,103,104,CHC,CHN,CHN
+1987,NL,CIN,CIN,W,2,162,81,84,78,N,,N,N,783,5560,1478,262,29,192,514,928,169,46,,,752,684,4.2400000000,7,6,44,4356,1486,170,485,919,128,137,0.97,Cincinnati Reds,Riverfront Stadium,2185205,104,103,CIN,CIN,CIN
+1987,NL,HOU,HOU,W,3,162,81,76,86,N,,N,N,648,5485,1386,238,28,122,526,936,162,46,,,678,615,3.8400000000,13,13,33,4323,1363,141,525,1137,115,113,0.98,Houston Astros,Astrodome,1909902,96,96,HOU,HOU,HOU
+1987,NL,LAN,LAD,W,4,162,81,73,89,N,,N,N,635,5517,1389,236,23,125,445,923,128,59,,,675,601,3.7200000000,29,8,32,4365,1415,130,565,1097,155,144,0.97,Los Angeles Dodgers,Dodger Stadium,2797409,98,98,LAD,LAN,LAN
+1987,NL,MON,WSN,E,3,162,81,91,71,N,,N,N,741,5527,1467,310,39,120,501,918,166,74,,,720,632,3.9200000000,16,8,50,4350,1428,145,446,1012,147,122,0.97,Montreal Expos,Stade Olympique,1850324,103,103,MON,MON,MON
+1987,NL,NYN,NYM,E,2,162,81,92,70,N,,N,N,823,5601,1499,287,34,192,592,1012,159,49,,,698,620,3.8400000000,16,7,51,4362,1407,135,510,1032,137,137,0.97,New York Mets,Shea Stadium,3034129,95,93,NYM,NYN,NYN
+1987,NL,PHI,PHI,E,4,162,81,80,82,N,,N,N,702,5475,1390,248,51,169,587,1109,111,49,,,749,673,4.1800000000,13,7,48,4344,1453,167,587,877,121,137,0.98,Philadelphia Phillies,Veterans Stadium,2100110,103,104,PHI,PHI,PHI
+1987,NL,PIT,PIT,E,4,162,81,80,82,N,,N,N,723,5536,1464,282,45,131,535,914,140,58,,,744,674,4.2000000000,25,13,39,4335,1377,164,562,914,123,147,0.98,Pittsburgh Pirates,Three Rivers Stadium,1161193,100,100,PIT,PIT,PIT
+1987,NL,SDN,SDP,W,6,162,81,65,97,N,,N,N,668,5456,1419,209,48,113,577,992,198,91,,,763,680,4.2700000000,14,10,33,4299,1402,175,602,897,146,135,0.97,San Diego Padres,Jack Murphy Stadium,1454061,96,97,SDP,SDN,SDN
+1987,NL,SFN,SFG,W,1,162,81,90,72,Y,,N,N,783,5608,1458,274,32,205,511,1094,126,97,,,669,601,3.6800000000,19,10,38,4413,1407,146,547,1038,129,183,0.98,San Francisco Giants,Candlestick Park,1917168,96,95,SFG,SFN,SFN
+1987,NL,SLN,STL,E,1,162,81,95,67,Y,,Y,N,798,5500,1449,252,49,94,644,933,248,72,,,693,637,3.9100000000,10,7,48,4398,1484,129,533,873,116,172,0.98,St. Louis Cardinals,Busch Stadium II,3072122,102,102,STL,SLN,SLN
+1988,AL,BAL,BAL,E,7,161,80,54,107,N,,N,N,550,5358,1275,199,20,137,504,869,69,44,,,789,714,4.5400000000,20,7,26,4248,1506,153,523,709,118,172,0.98,Baltimore Orioles,Memorial Stadium,1660738,97,98,BAL,BAL,BAL
+1988,AL,BOS,BOS,E,1,162,81,89,73,Y,,N,N,813,5545,1569,310,39,124,623,728,65,36,,,689,629,3.9700000000,26,14,37,4278,1415,143,493,1085,93,123,0.98,Boston Red Sox,Fenway Park II,2464851,105,104,BOS,BOS,BOS
+1988,AL,CAL,ANA,W,4,162,81,75,87,N,,N,N,714,5582,1458,258,31,124,469,819,86,52,,,771,698,4.3200000000,26,9,33,4365,1503,135,568,817,135,175,0.97,California Angels,Anaheim Stadium,2340925,97,97,CAL,CAL,CAL
+1988,AL,CHA,CHW,W,5,161,81,71,90,N,,N,N,631,5449,1327,224,35,132,446,908,98,46,,,757,659,4.1200000000,11,9,43,4317,1467,138,533,754,154,177,0.97,Chicago White Sox,Comiskey Park,1115749,100,100,CHW,CHA,CHA
+1988,AL,CLE,CLE,E,6,162,81,78,84,N,,N,N,666,5505,1435,235,28,134,416,866,97,50,,,731,663,4.1600000000,35,10,46,4302,1501,120,442,812,124,131,0.98,Cleveland Indians,Cleveland Stadium,1411610,102,103,CLE,CLE,CLE
+1988,AL,DET,DET,E,2,162,81,88,74,N,,N,N,703,5433,1358,213,28,143,588,841,87,42,,,658,596,3.7100000000,34,8,36,4335,1361,150,497,890,109,129,0.98,Detroit Tigers,Tiger Stadium,2081162,97,97,DET,DET,DET
+1988,AL,KCA,KCR,W,3,161,80,84,77,N,,N,N,704,5469,1419,275,40,121,486,944,137,54,,,648,579,3.6500000000,29,12,32,4284,1415,102,465,886,124,147,0.98,Kansas City Royals,Royals Stadium,2350181,101,101,KCR,KCA,KCA
+1988,AL,MIN,MIN,W,2,162,81,91,71,N,,N,N,759,5510,1508,294,31,151,528,832,107,63,,,672,625,3.9300000000,18,9,52,4293,1457,146,453,897,83,155,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,3030672,103,102,MIN,MIN,MIN
+1988,AL,ML4,MIL,E,3,162,81,87,75,N,,N,N,682,5488,1409,258,26,113,439,911,159,55,,,616,555,3.4500000000,30,8,51,4347,1355,125,437,832,120,146,0.98,Milwaukee Brewers,County Stadium,1923238,101,100,MIL,MIL,MIL
+1988,AL,NYA,NYY,E,5,161,80,85,76,N,,N,N,772,5592,1469,272,12,148,588,935,146,39,,,748,689,4.2600000000,16,5,43,4368,1512,157,487,861,134,161,0.97,New York Yankees,Yankee Stadium II,2633701,99,100,NYY,NYA,NYA
+1988,AL,OAK,OAK,W,1,162,81,104,58,Y,,Y,N,800,5602,1474,251,22,156,580,926,129,54,,,620,569,3.4400000000,22,9,64,4467,1376,116,553,983,105,151,0.98,Oakland Athletics,Oakland Coliseum,2287335,97,96,OAK,OAK,OAK
+1988,AL,SEA,SEA,W,7,161,81,68,93,N,,N,N,664,5436,1397,271,27,148,461,787,95,61,,,744,658,4.1500000000,28,11,28,4284,1385,144,558,981,123,168,0.98,Seattle Mariners,Kingdome,1022398,104,105,SEA,SEA,SEA
+1988,AL,TEX,TEX,W,6,161,81,70,91,N,,N,N,637,5479,1378,227,39,112,542,1022,130,57,,,735,647,4.0500000000,41,11,31,4314,1310,129,654,912,129,145,0.97,Texas Rangers,Arlington Stadium,1581901,102,103,TEX,TEX,TEX
+1988,AL,TOR,TOR,E,3,162,81,87,75,N,,N,N,763,5557,1491,271,47,158,521,935,107,36,,,680,612,3.8000000000,16,17,47,4347,1404,143,528,904,110,170,0.98,Toronto Blue Jays,Exhibition Stadium,2595175,100,99,TOR,TOR,TOR
+1988,NL,ATL,ATL,W,6,160,79,54,106,N,,N,N,555,5440,1319,228,28,96,432,848,95,69,,,741,657,4.0900000000,14,4,25,4338,1481,108,524,810,151,138,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,848089,104,106,ATL,ATL,ATL
+1988,NL,CHN,CHC,E,4,163,82,77,85,N,,N,N,660,5675,1481,262,46,113,403,910,120,46,,,694,625,3.8400000000,30,10,29,4392,1494,115,490,897,125,128,0.98,Chicago Cubs,Wrigley Field,2089034,105,105,CHC,CHN,CHN
+1988,NL,CIN,CIN,W,2,161,80,87,74,N,,N,N,641,5426,1334,246,25,122,479,922,207,56,,,596,542,3.3500000000,24,13,43,4365,1271,121,504,934,121,131,0.98,Cincinnati Reds,Riverfront Stadium,2072528,104,104,CIN,CIN,CIN
+1988,NL,HOU,HOU,W,5,162,81,82,80,N,,N,N,617,5494,1338,239,31,96,474,840,198,71,,,631,558,3.4100000000,21,15,40,4422,1339,123,478,1049,138,124,0.97,Houston Astros,Astrodome,1933505,96,96,HOU,HOU,HOU
+1988,NL,LAN,LAD,W,1,162,81,94,67,Y,,Y,Y,628,5431,1346,217,25,99,437,947,131,46,,,544,481,2.9600000000,32,24,49,4389,1291,84,473,1029,142,126,0.97,Los Angeles Dodgers,Dodger Stadium,2980262,98,97,LAD,LAN,LAN
+1988,NL,MON,WSN,E,3,163,81,81,81,N,,N,N,628,5573,1400,260,48,107,454,1053,189,89,,,592,507,3.0800000000,18,12,43,4446,1310,122,476,923,142,145,0.97,Montreal Expos,Stade Olympique,1478659,104,104,MON,MON,MON
+1988,NL,NYN,NYM,E,1,160,80,100,60,Y,,N,N,703,5408,1387,251,24,152,544,842,140,51,,,532,465,2.9100000000,31,22,46,4317,1253,78,404,1100,115,127,0.98,New York Mets,Shea Stadium,3055445,95,94,NYM,NYN,NYN
+1988,NL,PHI,PHI,E,6,162,81,65,96,N,,N,N,597,5403,1294,246,31,106,489,981,112,49,,,734,659,4.1400000000,16,6,36,4299,1447,118,628,859,145,139,0.97,Philadelphia Phillies,Veterans Stadium,1990041,101,103,PHI,PHI,PHI
+1988,NL,PIT,PIT,E,2,160,81,85,75,N,,N,N,651,5379,1327,240,45,110,553,947,119,60,,,616,555,3.4700000000,12,11,46,4320,1349,108,469,790,124,128,0.98,Pittsburgh Pirates,Three Rivers Stadium,1866713,98,98,PIT,PIT,PIT
+1988,NL,SDN,SDP,W,3,161,81,83,78,N,,N,N,594,5366,1325,205,35,94,494,892,123,50,,,583,528,3.2800000000,30,9,39,4347,1332,112,439,885,116,147,0.98,San Diego Padres,Jack Murphy Stadium,1506896,99,99,SDP,SDN,SDN
+1988,NL,SFN,SFG,W,4,162,81,83,79,N,,N,N,670,5450,1353,227,44,113,550,1023,121,78,,,626,551,3.3900000000,25,13,42,4386,1323,99,422,875,129,145,0.98,San Francisco Giants,Candlestick Park,1785297,96,95,SFG,SFN,SFN
+1988,NL,SLN,STL,E,5,162,81,76,86,N,,N,N,578,5518,1373,207,33,71,484,827,234,64,,,633,567,3.4700000000,17,14,42,4410,1387,91,486,881,121,131,0.98,St. Louis Cardinals,Busch Stadium II,2892799,101,101,STL,SLN,SLN
+1989,AL,BAL,BAL,E,2,162,81,87,75,N,,N,N,708,5440,1369,238,33,129,593,957,118,55,,,686,644,4.0000000000,16,7,44,4344,1518,134,486,676,87,163,0.98,Baltimore Orioles,Memorial Stadium,2535208,96,97,BAL,BAL,BAL
+1989,AL,BOS,BOS,E,3,162,81,83,79,N,,N,N,774,5666,1571,326,30,108,643,755,56,35,,,735,651,4.0100000000,14,9,42,4380,1448,131,548,1054,127,162,0.98,Boston Red Sox,Fenway Park II,2510012,107,106,BOS,BOS,BOS
+1989,AL,CAL,ANA,W,3,162,81,91,71,N,,N,N,669,5545,1422,208,37,145,429,1011,89,40,,,578,530,3.2800000000,32,20,38,4362,1384,113,465,897,96,173,0.98,California Angels,Anaheim Stadium,2647291,98,98,CAL,CAL,CAL
+1989,AL,CHA,CHW,W,7,161,80,69,92,N,,N,N,693,5504,1493,262,36,94,464,873,97,52,,,750,668,4.2300000000,9,5,46,4266,1472,144,539,778,151,176,0.97,Chicago White Sox,Comiskey Park,1045651,97,98,CHW,CHA,CHA
+1989,AL,CLE,CLE,E,6,162,81,73,89,N,,N,N,604,5463,1340,221,26,127,499,934,74,51,,,654,589,3.6500000000,23,13,38,4359,1423,107,452,844,118,126,0.98,Cleveland Indians,Cleveland Stadium,1285542,101,102,CLE,CLE,CLE
+1989,AL,DET,DET,E,7,162,81,59,103,N,,N,N,617,5432,1315,198,24,116,585,899,103,50,,,816,718,4.5300000000,24,4,26,4281,1514,150,652,831,130,153,0.97,Detroit Tigers,Tiger Stadium,1543656,98,99,DET,DET,DET
+1989,AL,KCA,KCR,W,2,162,81,92,70,N,,N,N,690,5475,1428,227,41,101,554,897,154,51,,,635,572,3.5500000000,27,13,38,4353,1415,86,455,978,111,139,0.98,Kansas City Royals,Royals Stadium,2477700,100,100,KCR,KCA,KCA
+1989,AL,MIN,MIN,W,5,162,81,80,82,N,,N,N,740,5581,1542,278,35,117,478,743,111,53,,,738,680,4.2800000000,19,8,38,4287,1495,139,500,851,107,141,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2277438,107,107,MIN,MIN,MIN
+1989,AL,ML4,MIL,E,4,162,81,81,81,N,,N,N,707,5473,1415,235,32,126,455,791,165,62,,,679,605,3.8000000000,16,8,45,4296,1463,129,457,812,155,164,0.97,Milwaukee Brewers,County Stadium,1970735,99,99,MIL,MIL,MIL
+1989,AL,NYA,NYY,E,5,161,81,74,87,N,,N,N,698,5458,1470,229,23,130,502,831,137,60,,,792,707,4.5000000000,15,9,44,4242,1550,150,521,787,122,183,0.98,New York Yankees,Yankee Stadium II,2170485,99,99,NYY,NYA,NYA
+1989,AL,OAK,OAK,W,1,162,81,99,63,Y,,Y,Y,712,5416,1414,220,25,127,562,855,157,55,,,576,497,3.0900000000,17,20,57,4344,1287,103,510,930,129,159,0.97,Oakland Athletics,Oakland Coliseum,2667225,97,95,OAK,OAK,OAK
+1989,AL,SEA,SEA,W,6,162,81,73,89,N,,N,N,694,5512,1417,237,29,134,489,838,81,55,,,728,639,4.0000000000,15,10,44,4314,1422,114,560,897,143,168,0.97,Seattle Mariners,Kingdome,1298443,103,104,SEA,SEA,SEA
+1989,AL,TEX,TEX,W,4,162,81,83,79,N,,N,N,695,5458,1433,260,46,122,503,989,101,49,,,714,623,3.9100000000,26,7,44,4302,1279,119,654,1112,136,137,0.97,Texas Rangers,Arlington Stadium,2043993,102,102,TEX,TEX,TEX
+1989,AL,TOR,TOR,E,1,162,81,89,73,Y,,N,N,731,5581,1449,265,40,142,521,923,144,58,,,651,584,3.5800000000,12,12,38,4401,1408,99,478,849,126,164,0.98,Toronto Blue Jays,Exhibition Stadium /Skydome,3375883,94,94,TOR,TOR,TOR
+1989,NL,ATL,ATL,W,6,161,79,63,97,N,,N,N,584,5463,1281,201,22,128,485,996,83,54,,,680,595,3.7000000000,15,8,33,4341,1370,114,468,966,152,124,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,984930,102,104,ATL,ATL,ATL
+1989,NL,CHN,CHC,E,1,162,81,93,69,Y,,N,N,702,5513,1438,235,45,124,472,921,136,57,,,623,556,3.4300000000,18,10,55,4380,1369,106,532,918,119,130,0.98,Chicago Cubs,Wrigley Field,2491942,108,108,CHC,CHN,CHN
+1989,NL,CIN,CIN,W,5,162,81,75,87,N,,N,N,632,5520,1362,243,28,128,493,1028,128,71,,,691,607,3.7300000000,16,9,37,4392,1404,125,559,981,121,108,0.98,Cincinnati Reds,Riverfront Stadium,1979320,104,103,CIN,CIN,CIN
+1989,NL,HOU,HOU,W,3,162,82,86,76,N,,N,N,647,5516,1316,239,28,97,530,860,144,62,,,669,598,3.6400000000,19,12,38,4437,1379,105,551,965,140,121,0.97,Houston Astros,Astrodome,1834908,97,97,HOU,HOU,HOU
+1989,NL,LAN,LAD,W,4,160,81,77,83,N,,N,N,554,5465,1313,241,17,89,507,885,81,54,,,536,480,2.9500000000,25,19,36,4389,1278,95,504,1052,118,153,0.98,Los Angeles Dodgers,Dodger Stadium,2944653,99,98,LAD,LAN,LAN
+1989,NL,MON,WSN,E,4,162,81,81,81,N,,N,N,632,5482,1353,267,30,100,572,958,160,70,,,630,568,3.4800000000,20,13,35,4404,1344,120,519,1059,136,126,0.97,Montreal Expos,Stade Olympique,1783533,102,101,MON,MON,MON
+1989,NL,NYN,NYM,E,2,162,81,87,75,N,,N,N,683,5489,1351,280,21,147,504,934,158,53,,,595,532,3.2900000000,24,12,38,4362,1260,115,532,1108,144,110,0.97,New York Mets,Shea Stadium,2918710,95,93,NYM,NYN,NYN
+1989,NL,PHI,PHI,E,6,163,81,67,95,N,,N,N,629,5447,1324,215,36,123,558,926,106,50,,,735,643,4.0400000000,10,10,33,4299,1408,127,613,899,133,136,0.97,Philadelphia Phillies,Veterans Stadium,1861985,99,101,PHI,PHI,PHI
+1989,NL,PIT,PIT,E,5,164,81,74,88,N,,N,N,637,5539,1334,263,53,95,563,914,155,69,,,680,601,3.6400000000,20,9,40,4461,1394,121,539,827,160,130,0.97,Pittsburgh Pirates,Three Rivers Stadium,1374141,96,96,PIT,PIT,PIT
+1989,NL,SDN,SDP,W,2,162,81,89,73,N,,N,N,642,5422,1360,215,32,120,552,1013,136,67,,,626,547,3.3800000000,21,11,52,4371,1359,133,481,933,152,147,0.97,San Diego Padres,Jack Murphy Stadium,2009031,101,101,SDP,SDN,SDN
+1989,NL,SFN,SFG,W,1,162,81,92,70,Y,,Y,N,699,5469,1365,241,52,141,508,1071,87,54,,,600,534,3.3000000000,12,16,47,4371,1320,120,471,802,113,135,0.98,San Francisco Giants,Candlestick Park,2059701,97,96,SFG,SFN,SFN
+1989,NL,SLN,STL,E,3,164,83,86,76,N,,N,N,632,5492,1418,263,47,73,507,848,155,54,,,608,545,3.3600000000,18,18,43,4383,1330,84,482,844,111,134,0.98,St. Louis Cardinals,Busch Stadium II,3080980,103,103,STL,SLN,SLN
+1990,AL,BAL,BAL,E,5,161,80,76,85,N,,N,N,669,5410,1328,234,22,132,660,962,94,52,,,698,644,4.0400000000,10,5,43,4305,1445,161,537,776,91,151,0.98,Baltimore Orioles,Memorial Stadium,2415189,97,98,BAL,BAL,BAL
+1990,AL,BOS,BOS,E,1,162,81,88,74,Y,,N,N,699,5516,1502,298,31,106,598,795,53,52,,,664,596,3.7200000000,15,13,44,4326,1439,92,519,997,123,154,0.98,Boston Red Sox,Fenway Park II,2528986,105,105,BOS,BOS,BOS
+1990,AL,CAL,ANA,W,4,162,81,80,82,N,,N,N,690,5570,1448,237,27,147,566,1000,69,43,,,706,612,3.7900000000,21,13,42,4362,1482,106,544,944,140,186,0.97,California Angels,Anaheim Stadium,2555688,97,97,CAL,CAL,CAL
+1990,AL,CHA,CHW,W,2,162,80,94,68,N,,N,N,682,5402,1393,251,44,106,478,903,140,90,,,633,581,3.6100000000,17,10,68,4347,1313,106,548,914,124,169,0.98,Chicago White Sox,Comiskey Park,2002357,98,98,CHW,CHA,CHA
+1990,AL,CLE,CLE,E,4,162,81,77,85,N,,N,N,732,5485,1465,266,41,110,458,836,107,52,,,737,675,4.2600000000,12,10,47,4281,1491,163,518,860,117,146,0.98,Cleveland Indians,Cleveland Stadium,1225240,100,100,CLE,CLE,CLE
+1990,AL,DET,DET,E,3,162,81,79,83,N,,N,N,750,5479,1418,241,32,172,634,952,82,57,,,754,698,4.3900000000,15,12,45,4290,1401,154,661,856,131,178,0.97,Detroit Tigers,Tiger Stadium,1495785,101,102,DET,DET,DET
+1990,AL,KCA,KCR,W,6,161,81,75,86,N,,N,N,707,5488,1465,316,44,100,498,879,107,62,,,709,620,3.9300000000,18,8,33,4260,1449,116,560,1006,122,161,0.98,Kansas City Royals,Royals Stadium,2244956,98,98,KCR,KCA,KCA
+1990,AL,MIN,MIN,W,7,162,81,74,88,N,,N,N,666,5499,1458,281,39,100,445,749,96,53,,,729,657,4.1200000000,13,13,43,4305,1509,134,489,872,101,161,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1751584,107,107,MIN,MIN,MIN
+1990,AL,ML4,MIL,E,6,162,81,74,88,N,,N,N,732,5503,1408,247,36,128,519,821,164,72,,,760,655,4.0800000000,23,13,42,4335,1558,121,469,771,149,152,0.97,Milwaukee Brewers,County Stadium,1752900,99,99,MIL,MIL,MIL
+1990,AL,NYA,NYY,E,7,162,81,67,95,N,,N,N,603,5483,1322,208,19,147,427,1027,119,45,,,749,675,4.2100000000,15,6,41,4332,1430,144,618,909,126,164,0.98,New York Yankees,Yankee Stadium II,2006436,100,102,NYY,NYA,NYA
+1990,AL,OAK,OAK,W,1,162,81,103,59,Y,,Y,N,733,5433,1379,209,22,164,651,992,141,54,,,570,514,3.1800000000,18,16,64,4368,1287,123,494,831,87,152,0.98,Oakland Athletics,Oakland Coliseum,2900217,96,95,OAK,OAK,OAK
+1990,AL,SEA,SEA,W,5,162,81,77,85,N,,N,N,640,5474,1419,251,26,107,596,749,105,51,,,680,592,3.6900000000,21,7,41,4329,1319,120,606,1064,130,152,0.97,Seattle Mariners,Kingdome,1509727,101,101,SEA,SEA,SEA
+1990,AL,TEX,TEX,W,3,162,82,83,79,N,,N,N,676,5469,1416,257,27,110,575,1054,115,48,,,696,615,3.8300000000,25,9,36,4332,1343,113,623,997,133,161,0.97,Texas Rangers,Arlington Stadium,2057911,100,101,TEX,TEX,TEX
+1990,AL,TOR,TOR,E,2,162,81,86,76,N,,N,N,767,5589,1479,263,50,167,526,970,111,52,,,661,620,3.8400000000,6,9,48,4362,1434,143,445,892,86,144,0.98,Toronto Blue Jays,Skydome,3885284,106,105,TOR,TOR,TOR
+1990,NL,ATL,ATL,W,6,162,81,65,97,N,,N,N,682,5504,1376,263,26,162,473,1010,92,55,,,821,727,4.5800000000,17,8,30,4287,1527,128,579,938,158,133,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,980129,105,106,ATL,ATL,ATL
+1990,NL,CHN,CHC,E,4,162,81,77,85,N,,N,N,690,5600,1474,240,36,136,406,869,151,50,,,774,695,4.3400000000,13,7,42,4326,1510,121,572,877,122,136,0.98,Chicago Cubs,Wrigley Field,2243791,108,108,CHC,CHN,CHN
+1990,NL,CIN,CIN,W,1,162,81,91,71,Y,,Y,Y,693,5525,1466,284,40,125,466,913,166,66,,,597,548,3.3900000000,14,12,50,4368,1338,124,543,1029,102,126,0.98,Cincinnati Reds,Riverfront Stadium,2400892,105,105,CIN,CIN,CIN
+1990,NL,HOU,HOU,W,4,162,81,75,87,N,,N,N,573,5379,1301,209,32,94,548,997,179,83,,,656,582,3.6100000000,12,6,37,4350,1396,130,496,854,131,124,0.97,Houston Astros,Astrodome,1310927,97,98,HOU,HOU,HOU
+1990,NL,LAN,LAD,W,2,162,81,86,76,N,,N,N,728,5491,1436,222,27,129,538,952,141,65,,,685,596,3.7200000000,29,12,29,4326,1364,137,478,1021,130,123,0.97,Los Angeles Dodgers,Dodger Stadium,3002396,98,97,LAD,LAN,LAN
+1990,NL,MON,WSN,E,3,162,81,85,77,N,,N,N,662,5453,1363,227,43,114,576,1024,235,99,,,598,552,3.3700000000,18,11,50,4419,1349,127,510,991,110,134,0.98,Montreal Expos,Stade Olympique,1373087,96,96,MON,MON,MON
+1990,NL,NYN,NYM,E,2,162,81,91,71,N,,N,N,775,5504,1410,278,21,172,536,851,110,33,,,613,547,3.4200000000,18,14,41,4320,1339,119,444,1217,132,107,0.97,New York Mets,Shea Stadium,2732745,100,99,NYM,NYN,NYN
+1990,NL,PHI,PHI,E,4,162,81,77,85,N,,N,N,646,5535,1410,237,27,103,582,915,108,35,,,729,655,4.0700000000,18,7,35,4347,1381,124,651,840,117,150,0.98,Philadelphia Phillies,Veterans Stadium,1992484,99,100,PHI,PHI,PHI
+1990,NL,PIT,PIT,E,1,162,81,95,67,Y,,N,N,733,5388,1395,288,42,138,582,914,137,52,,,619,547,3.4000000000,18,8,43,4341,1367,135,413,848,134,125,0.97,Pittsburgh Pirates,Three Rivers Stadium,2049908,96,95,PIT,PIT,PIT
+1990,NL,SDN,SDP,W,4,162,81,75,87,N,,N,N,673,5554,1429,243,35,123,509,902,138,59,,,673,597,3.6800000000,21,12,35,4383,1437,147,507,928,141,141,0.97,San Diego Padres,Jack Murphy Stadium,1856396,101,101,SDP,SDN,SDN
+1990,NL,SFN,SFG,W,3,162,81,85,77,N,,N,N,719,5573,1459,221,35,152,488,973,109,56,,,710,656,4.0800000000,14,6,45,4338,1477,131,553,788,107,148,0.98,San Francisco Giants,Candlestick Park,1975528,96,96,SFG,SFN,SFN
+1990,NL,SLN,STL,E,6,162,81,70,92,N,,N,N,599,5462,1398,255,41,73,517,844,221,74,,,698,620,3.8700000000,8,13,39,4329,1432,98,475,833,129,114,0.97,St. Louis Cardinals,Busch Stadium II,2573225,100,100,STL,SLN,SLN
+1991,AL,BAL,BAL,E,6,162,81,67,95,N,,N,N,686,5604,1421,256,29,170,528,974,50,33,,,796,743,4.5900000000,8,8,42,4371,1534,147,504,868,89,172,0.98,Baltimore Orioles,Memorial Stadium,2552753,96,97,BAL,BAL,BAL
+1991,AL,BOS,BOS,E,2,162,81,84,78,N,,N,N,731,5530,1486,305,25,126,593,820,59,39,,,712,641,4.0100000000,15,13,45,4317,1405,147,530,999,116,165,0.98,Boston Red Sox,Fenway Park II,2562435,105,105,BOS,BOS,BOS
+1991,AL,CAL,ANA,W,7,162,81,81,81,N,,N,N,653,5470,1396,245,29,115,448,928,94,56,,,649,591,3.6900000000,18,10,50,4323,1351,141,543,990,102,156,0.98,California Angels,Anaheim Stadium,2416236,99,100,CAL,CAL,CAL
+1991,AL,CHA,CHW,W,2,162,81,87,75,N,,N,N,758,5594,1464,226,39,139,610,896,134,74,,,681,622,3.7900000000,28,8,40,4434,1302,154,601,923,116,151,0.98,Chicago White Sox,Comiskey Park II,2934154,98,97,CHW,CHA,CHA
+1991,AL,CLE,CLE,E,7,162,82,57,105,N,,N,N,576,5470,1390,236,26,79,449,888,84,58,,,759,677,4.2300000000,22,8,33,4323,1551,110,441,862,142,150,0.97,Cleveland Indians,Cleveland Stadium,1051863,101,102,CLE,CLE,CLE
+1991,AL,DET,DET,E,2,162,81,84,78,N,,N,N,817,5547,1372,259,26,209,699,1185,109,47,,,794,727,4.5100000000,18,8,38,4350,1570,148,593,739,104,171,0.98,Detroit Tigers,Tiger Stadium,1641661,102,102,DET,DET,DET
+1991,AL,KCA,KCR,W,6,162,81,82,80,N,,N,N,727,5584,1475,290,41,117,523,969,119,68,,,722,639,3.9200000000,17,12,41,4398,1473,105,529,1004,125,141,0.98,Kansas City Royals,Royals Stadium,2161537,100,101,KCR,KCA,KCA
+1991,AL,MIN,MIN,W,1,162,81,95,67,Y,,Y,Y,776,5556,1557,270,42,140,526,747,107,68,,,652,594,3.6900000000,21,12,53,4347,1402,139,488,876,94,161,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2293842,105,104,MIN,MIN,MIN
+1991,AL,ML4,MIL,E,4,162,80,83,79,N,,N,N,799,5611,1523,247,53,116,556,802,106,68,,,744,673,4.1400000000,23,11,41,4389,1498,147,527,859,118,176,0.98,Milwaukee Brewers,County Stadium,1478729,98,98,MIL,MIL,MIL
+1991,AL,NYA,NYY,E,5,162,81,71,91,N,,N,N,674,5541,1418,249,19,147,473,861,109,36,,,777,709,4.4200000000,3,11,37,4332,1510,152,506,936,133,181,0.97,New York Yankees,Yankee Stadium II,1863733,100,101,NYY,NYA,NYA
+1991,AL,OAK,OAK,W,4,162,81,84,78,N,,N,N,760,5410,1342,246,19,159,642,981,151,64,,,776,733,4.5700000000,14,10,49,4332,1425,155,655,892,107,150,0.98,Oakland Athletics,Oakland Coliseum,2713493,94,93,OAK,OAK,OAK
+1991,AL,SEA,SEA,W,5,162,81,83,79,N,,N,N,702,5494,1400,268,29,126,588,811,97,44,,,674,617,3.7900000000,10,13,48,4392,1387,136,628,1003,110,187,0.98,Seattle Mariners,Kingdome,2147905,100,100,SEA,SEA,SEA
+1991,AL,TEX,TEX,W,3,162,81,85,77,N,,N,N,829,5703,1539,288,31,177,596,1039,102,50,,,814,735,4.4700000000,9,10,41,4437,1486,151,662,1022,134,138,0.97,Texas Rangers,Arlington Stadium,2297720,98,99,TEX,TEX,TEX
+1991,AL,TOR,TOR,E,1,162,81,91,71,Y,,N,N,684,5489,1412,295,45,133,499,1043,148,53,,,622,569,3.5000000000,10,16,60,4386,1301,121,523,971,127,115,0.98,Toronto Blue Jays,Skydome,4001527,104,103,TOR,TOR,TOR
+1991,NL,ATL,ATL,W,1,162,81,94,68,Y,,Y,N,749,5456,1407,255,30,141,563,906,165,76,,,644,563,3.4900000000,18,7,48,4356,1304,118,481,969,138,122,0.97,Atlanta Braves,Atlanta-Fulton County Stadium,2140217,106,106,ATL,ATL,ATL
+1991,NL,CHN,CHC,E,4,160,83,77,83,N,,N,N,695,5522,1395,232,26,159,442,879,123,64,,,734,652,4.0300000000,12,4,40,4368,1415,117,542,927,113,120,0.98,Chicago Cubs,Wrigley Field,2314250,104,105,CHC,CHN,CHN
+1991,NL,CIN,CIN,W,5,162,81,74,88,N,,N,N,689,5501,1419,250,27,164,488,1006,124,56,,,691,613,3.8300000000,7,11,43,4320,1372,127,560,997,125,131,0.97,Cincinnati Reds,Riverfront Stadium,2372377,104,104,CIN,CIN,CIN
+1991,NL,HOU,HOU,W,6,162,81,65,97,N,,N,N,605,5504,1345,240,43,79,502,1027,125,68,,,717,646,4.0000000000,7,13,36,4359,1347,129,651,1033,161,129,0.97,Houston Astros,Astrodome,1196152,93,94,HOU,HOU,HOU
+1991,NL,LAN,LAD,W,2,162,81,93,69,N,,N,N,665,5408,1366,191,29,108,583,957,126,68,,,565,496,3.0600000000,15,14,40,4374,1312,96,500,1028,123,126,0.98,Los Angeles Dodgers,Dodger Stadium,3348170,98,97,LAD,LAN,LAN
+1991,NL,MON,WSN,E,6,161,68,71,90,N,,N,N,579,5412,1329,236,42,95,484,1056,221,100,,,655,582,3.6400000000,12,14,39,4320,1304,111,584,909,131,128,0.97,Montreal Expos,Stade Olympique,934742,99,99,MON,MON,MON
+1991,NL,NYN,NYM,E,5,161,82,77,84,N,,N,N,640,5359,1305,250,24,117,578,789,153,70,,,646,568,3.5600000000,12,11,39,4311,1403,108,410,1028,143,112,0.97,New York Mets,Shea Stadium,2284484,99,99,NYM,NYN,NYN
+1991,NL,PHI,PHI,E,3,162,83,78,84,N,,N,N,629,5521,1332,248,33,111,490,1026,92,30,,,680,627,3.8600000000,16,11,35,4389,1346,111,670,988,116,111,0.98,Philadelphia Phillies,Veterans Stadium,2050012,99,99,PHI,PHI,PHI
+1991,NL,PIT,PIT,E,1,162,84,98,64,Y,,N,N,768,5449,1433,259,50,126,620,901,124,46,,,632,557,3.4400000000,18,11,51,4368,1411,117,401,919,120,134,0.98,Pittsburgh Pirates,Three Rivers Stadium,2065302,99,98,PIT,PIT,PIT
+1991,NL,SDN,SDP,W,3,162,81,84,78,N,,N,N,636,5408,1321,204,36,121,501,1069,101,64,,,646,576,3.5700000000,14,11,47,4356,1385,139,457,921,113,130,0.98,San Diego Padres,Jack Murphy Stadium,1804289,103,103,SDP,SDN,SDN
+1991,NL,SFN,SFG,W,4,162,81,75,87,N,,N,N,649,5463,1345,215,48,141,471,973,95,57,,,697,646,4.0300000000,10,10,45,4326,1397,143,544,905,108,151,0.98,San Francisco Giants,Candlestick Park,1737478,96,97,SFG,SFN,SFN
+1991,NL,SLN,STL,E,2,162,84,84,78,N,,N,N,651,5362,1366,239,53,68,532,857,202,110,,,648,588,3.6900000000,9,5,51,4305,1367,114,454,822,105,133,0.98,St. Louis Cardinals,Busch Stadium II,2448699,100,100,STL,SLN,SLN
+1992,AL,BAL,BAL,E,3,162,81,89,73,N,,N,N,705,5485,1423,243,36,148,647,827,89,48,,,656,617,3.7900000000,20,16,48,4392,1419,124,518,846,93,168,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3567819,102,101,BAL,BAL,BAL
+1992,AL,BOS,BOS,E,7,162,81,73,89,N,,N,N,599,5461,1343,259,21,84,591,865,44,48,,,669,576,3.5800000000,22,13,39,4344,1403,107,535,943,139,170,0.97,Boston Red Sox,Fenway Park II,2468574,106,107,BOS,BOS,BOS
+1992,AL,CAL,ANA,W,5,162,81,72,90,N,,N,N,579,5364,1306,202,20,88,416,882,160,101,,,671,617,3.8400000000,26,13,42,4338,1449,130,532,888,134,172,0.97,California Angels,Anaheim Stadium,2065444,100,101,CAL,CAL,CAL
+1992,AL,CHA,CHW,W,3,162,82,86,76,N,,N,N,738,5498,1434,269,36,110,622,784,160,57,,,690,620,3.8200000000,21,5,52,4383,1400,123,550,810,129,134,0.97,Chicago White Sox,Comiskey Park II,2681156,99,99,CHW,CHA,CHA
+1992,AL,CLE,CLE,E,4,162,81,76,86,N,,N,N,674,5620,1495,227,24,127,448,885,144,67,,,746,671,4.1100000000,13,7,46,4410,1507,159,566,890,141,176,0.97,Cleveland Indians,Cleveland Stadium,1224094,99,100,CLE,CLE,CLE
+1992,AL,DET,DET,E,6,162,80,75,87,N,,N,N,791,5515,1411,256,16,182,675,1055,66,45,,,794,733,4.6000000000,10,4,36,4305,1534,155,564,693,116,164,0.98,Detroit Tigers,Tiger Stadium,1423963,101,100,DET,DET,DET
+1992,AL,KCA,KCR,W,5,162,81,72,90,N,,N,N,610,5501,1411,284,42,75,439,741,131,71,,,667,613,3.8100000000,9,12,44,4341,1426,106,512,834,122,164,0.98,Kansas City Royals,Royals Stadium,1867689,102,103,KCR,KCA,KCA
+1992,AL,MIN,MIN,W,2,162,81,90,72,N,,N,N,747,5582,1544,275,27,104,527,834,123,74,,,653,597,3.7000000000,16,13,50,4359,1391,121,479,923,94,155,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2482428,103,102,MIN,MIN,MIN
+1992,AL,ML4,MIL,E,2,162,81,92,70,N,,N,N,740,5504,1477,272,35,82,511,779,256,115,,,604,555,3.4300000000,19,14,39,4371,1344,127,435,793,89,146,0.98,Milwaukee Brewers,County Stadium,1857351,99,98,MIL,MIL,MIL
+1992,AL,NYA,NYY,E,4,162,81,76,86,N,,N,N,733,5593,1462,281,18,163,536,903,78,37,,,746,679,4.2100000000,20,9,44,4356,1453,129,612,851,112,165,0.98,New York Yankees,Yankee Stadium II,1748737,100,100,NYY,NYA,NYA
+1992,AL,OAK,OAK,W,1,162,81,96,66,Y,,N,N,745,5387,1389,219,24,142,707,831,143,59,,,672,600,3.7300000000,8,9,58,4341,1396,129,601,843,125,158,0.97,Oakland Athletics,Oakland Coliseum,2494160,94,95,OAK,OAK,OAK
+1992,AL,SEA,SEA,W,7,162,81,64,98,N,,N,N,679,5564,1466,278,24,149,474,841,100,55,,,799,731,4.5500000000,21,9,30,4335,1467,129,661,894,112,170,0.98,Seattle Mariners,Kingdome,1651367,100,100,SEA,SEA,SEA
+1992,AL,TEX,TEX,W,4,162,81,77,85,N,,N,N,682,5537,1387,266,23,159,550,1036,81,44,,,753,663,4.0900000000,19,3,42,4380,1471,113,598,1034,153,153,0.97,Texas Rangers,Arlington Stadium,2198231,97,97,TEX,TEX,TEX
+1992,AL,TOR,TOR,E,1,162,81,96,66,Y,,Y,Y,780,5536,1458,265,40,163,561,933,129,39,,,682,626,3.9100000000,18,14,49,4320,1346,124,541,954,93,109,0.98,Toronto Blue Jays,Skydome,4028318,105,104,TOR,TOR,TOR
+1992,NL,ATL,ATL,W,1,162,81,98,64,Y,,Y,N,682,5480,1391,223,48,138,493,924,126,60,,,569,509,3.1400000000,26,24,41,4380,1321,89,489,948,109,121,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,3077400,107,105,ATL,ATL,ATL
+1992,NL,CHN,CHC,E,4,162,81,78,84,N,,N,N,593,5590,1420,221,41,104,417,816,77,51,,,624,553,3.3900000000,16,11,37,4407,1337,107,575,901,114,142,0.98,Chicago Cubs,Wrigley Field,2126720,102,103,CHC,CHN,CHN
+1992,NL,CIN,CIN,W,2,162,81,90,72,N,,N,N,660,5460,1418,281,44,99,563,888,125,65,,,609,557,3.4600000000,9,11,55,4347,1362,109,470,1060,96,128,0.98,Cincinnati Reds,Riverfront Stadium,2315946,103,103,CIN,CIN,CIN
+1992,NL,HOU,HOU,W,4,162,81,81,81,N,,N,N,608,5480,1350,255,38,96,506,1025,139,54,,,668,603,3.7200000000,5,12,45,4377,1386,114,539,978,112,125,0.98,Houston Astros,Astrodome,1211412,95,95,HOU,HOU,HOU
+1992,NL,LAN,LAD,W,6,162,81,63,99,N,,N,N,548,5368,1333,201,34,72,503,899,142,78,,,636,545,3.4100000000,18,13,29,4314,1401,82,553,981,173,136,0.97,Los Angeles Dodgers,Dodger Stadium,2473266,99,99,LAD,LAN,LAN
+1992,NL,MON,WSN,E,2,162,81,87,75,N,,N,N,648,5477,1381,263,37,102,463,976,196,63,,,581,530,3.2500000000,11,14,49,4404,1296,92,525,1014,124,113,0.98,Montreal Expos,Stade Olympique,1669127,99,99,MON,MON,MON
+1992,NL,NYN,NYM,E,5,162,81,72,90,N,,N,N,599,5340,1254,259,17,93,572,956,129,52,,,653,588,3.6600000000,17,13,34,4338,1404,98,482,1025,116,134,0.98,New York Mets,Shea Stadium,1779534,99,99,NYM,NYN,NYN
+1992,NL,PHI,PHI,E,6,162,81,70,92,N,,N,N,686,5500,1392,255,36,118,509,1059,127,31,,,717,652,4.1100000000,27,7,34,4284,1387,113,549,851,131,128,0.97,Philadelphia Phillies,Veterans Stadium,1927448,100,100,PHI,PHI,PHI
+1992,NL,PIT,PIT,E,1,162,81,96,66,Y,,N,N,693,5527,1409,272,54,106,569,872,110,53,,,595,551,3.3500000000,20,20,43,4437,1410,101,455,844,101,144,0.98,Pittsburgh Pirates,Three Rivers Stadium,1829395,99,98,PIT,PIT,PIT
+1992,NL,SDN,SDP,W,3,162,81,82,80,N,,N,N,617,5476,1396,255,30,135,453,864,69,52,,,636,578,3.5600000000,9,11,46,4383,1444,111,439,971,115,127,0.98,San Diego Padres,Jack Murphy Stadium,1721406,101,102,SDP,SDN,SDN
+1992,NL,SFN,SFG,W,5,162,81,72,90,N,,N,N,574,5456,1330,220,36,105,435,1067,112,64,,,647,586,3.6100000000,9,12,30,4383,1385,128,502,927,111,174,0.98,San Francisco Giants,Candlestick Park,1560998,94,94,SFG,SFN,SFN
+1992,NL,SLN,STL,E,3,162,81,83,79,N,,N,N,631,5594,1464,262,44,94,495,996,208,118,,,604,556,3.3800000000,10,9,47,4440,1405,118,400,842,91,146,0.98,St. Louis Cardinals,Busch Stadium II,2418483,97,97,STL,SLN,SLN
+1993,AL,BAL,BAL,E,3,162,81,85,77,N,,N,N,786,5508,1470,287,24,157,655,930,73,54,,,745,691,4.3100000000,21,10,42,4326,1427,153,579,900,100,171,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3644965,104,103,BAL,BAL,BAL
+1993,AL,BOS,BOS,E,5,162,81,80,82,N,,N,N,686,5496,1451,319,29,114,508,871,73,38,,,698,608,3.7700000000,9,11,44,4356,1379,127,552,997,122,155,0.98,Boston Red Sox,Fenway Park II,2422021,106,107,BOS,BOS,BOS
+1993,AL,CAL,ANA,W,5,162,81,71,91,N,,N,N,684,5391,1399,259,24,114,564,930,169,100,,,770,690,4.3400000000,26,6,41,4290,1482,153,550,843,120,161,0.98,California Angels,Anaheim Stadium,2057460,103,104,CAL,CAL,CAL
+1993,AL,CHA,CHW,W,1,162,81,94,68,Y,,N,N,776,5483,1454,228,44,162,604,834,106,57,,,664,598,3.7000000000,16,11,48,4362,1398,125,566,974,112,153,0.98,Chicago White Sox,Comiskey Park II,2581091,98,97,CHW,CHA,CHA
+1993,AL,CLE,CLE,E,6,162,81,76,86,N,,N,N,790,5619,1547,264,31,141,488,843,159,55,,,813,735,4.5800000000,7,8,45,4335,1591,182,591,888,148,174,0.97,Cleveland Indians,Cleveland Stadium,2177908,100,101,CLE,CLE,CLE
+1993,AL,DET,DET,E,3,162,81,85,77,N,,N,N,899,5620,1546,282,38,178,765,1122,104,63,,,837,742,4.6500000000,11,7,36,4308,1547,188,542,828,132,148,0.97,Detroit Tigers,Tiger Stadium,1971421,99,99,DET,DET,DET
+1993,AL,KCA,KCR,W,3,162,81,84,78,N,,N,N,675,5522,1455,294,35,125,428,936,100,75,,,694,649,4.0400000000,16,6,48,4335,1379,105,571,985,96,150,0.98,Kansas City Royals,Kauffman Stadium,1934578,106,106,KCR,KCA,KCA
+1993,AL,MIN,MIN,W,5,162,81,71,91,N,,N,N,693,5601,1480,261,27,121,493,850,83,59,,,830,756,4.7100000000,5,3,44,4332,1591,148,514,901,100,160,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2048673,100,100,MIN,MIN,MIN
+1993,AL,ML4,MIL,E,7,162,81,69,93,N,,N,N,733,5525,1426,240,25,125,555,932,138,93,,,792,715,4.4500000000,26,6,29,4341,1511,153,522,810,131,148,0.97,Milwaukee Brewers,County Stadium,1688080,99,99,MIL,MIL,MIL
+1993,AL,NYA,NYY,E,2,162,81,88,74,N,,N,N,821,5615,1568,294,24,178,629,910,39,35,,,761,695,4.3500000000,11,13,38,4314,1467,170,552,899,101,166,0.98,New York Yankees,Yankee Stadium II,2416942,97,96,NYY,NYA,NYA
+1993,AL,OAK,OAK,W,7,162,81,68,94,N,,N,N,715,5543,1408,260,21,158,622,1048,131,59,,,846,791,4.9000000000,8,2,42,4356,1551,157,680,864,111,161,0.98,Oakland Athletics,Oakland Coliseum,2035025,94,94,OAK,OAK,OAK
+1993,AL,SEA,SEA,W,4,162,81,82,80,N,,N,N,734,5494,1429,272,24,161,624,901,91,68,,,731,678,4.2000000000,22,10,41,4359,1421,135,605,1083,90,173,0.98,Seattle Mariners,Kingdome,2052638,101,102,SEA,SEA,SEA
+1993,AL,TEX,TEX,W,2,162,81,86,76,N,,N,N,835,5510,1472,284,39,181,483,984,113,67,,,751,684,4.2800000000,20,6,45,4314,1476,144,562,957,130,145,0.97,Texas Rangers,Arlington Stadium,2244616,96,96,TEX,TEX,TEX
+1993,AL,TOR,TOR,E,1,162,81,95,67,Y,,Y,Y,847,5579,1556,317,42,159,588,861,170,49,,,742,674,4.2100000000,11,11,50,4323,1441,134,620,1023,107,144,0.98,Toronto Blue Jays,Skydome,4057947,101,100,TOR,TOR,TOR
+1993,NL,ATL,ATL,W,1,162,81,104,58,Y,,N,N,767,5515,1444,239,29,169,560,946,125,48,,,559,508,3.1400000000,18,16,46,4365,1297,101,480,1036,108,146,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,3884720,101,100,ATL,ATL,ATL
+1993,NL,CHN,CHC,E,4,163,82,84,78,N,,N,N,738,5627,1521,259,32,161,446,923,100,43,,,739,673,4.1800000000,8,5,56,4347,1514,153,470,905,115,162,0.98,Chicago Cubs,Wrigley Field,2653763,98,98,CHC,CHN,CHN
+1993,NL,CIN,CIN,W,5,162,81,73,89,N,,N,N,722,5517,1457,261,28,137,485,1025,142,59,,,785,719,4.5100000000,11,8,37,4302,1510,158,508,996,120,133,0.98,Cincinnati Reds,Riverfront Stadium,2453232,101,100,CIN,CIN,CIN
+1993,NL,COL,COL,W,6,162,81,67,95,N,,N,N,758,5517,1507,278,59,142,388,944,146,90,,,967,860,5.4100000000,9,0,35,4293,1664,181,609,913,167,149,0.97,Colorado Rockies,Mile High Stadium,4483350,117,118,COL,COL,COL
+1993,NL,FLO,FLA,E,6,162,81,64,98,N,,N,N,581,5475,1356,197,31,94,498,1054,117,56,,,724,661,4.1300000000,4,5,48,4320,1437,135,598,945,122,130,0.98,Florida Marlins,Joe Robbie Stadium,3064847,104,106,FLA,FLO,FLO
+1993,NL,HOU,HOU,W,3,162,81,85,77,N,,N,N,716,5464,1459,288,37,138,497,911,103,60,,,630,559,3.4900000000,18,14,42,4323,1363,117,476,1056,125,141,0.97,Houston Astros,Astrodome,2084618,96,96,HOU,HOU,HOU
+1993,NL,LAN,LAD,W,4,162,81,81,81,N,,N,N,675,5588,1458,234,28,130,492,937,126,61,,,662,572,3.5000000000,17,9,36,4416,1406,103,567,1043,132,141,0.97,Los Angeles Dodgers,Dodger Stadium,3170393,95,95,LAD,LAN,LAN
+1993,NL,MON,WSN,E,2,163,81,94,68,N,,N,N,732,5493,1410,270,36,122,542,860,228,56,,,682,574,3.5500000000,8,7,61,4368,1369,119,521,934,159,144,0.97,Montreal Expos,Stade Olympique,1641437,105,104,MON,MON,MON
+1993,NL,NYN,NYM,E,7,162,81,59,103,N,,N,N,672,5448,1350,228,37,158,448,879,79,50,,,744,647,4.0500000000,16,8,22,4314,1483,139,434,867,156,143,0.97,New York Mets,Shea Stadium,1873183,99,100,NYM,NYN,NYN
+1993,NL,PHI,PHI,E,1,162,81,97,65,Y,,Y,N,877,5685,1555,297,51,156,665,1049,91,32,,,740,646,3.9500000000,24,11,46,4416,1419,129,573,1117,141,123,0.97,Philadelphia Phillies,Veterans Stadium,3137674,99,98,PHI,PHI,PHI
+1993,NL,PIT,PIT,E,5,162,81,75,87,N,,N,N,707,5549,1482,267,50,110,536,972,92,55,,,806,766,4.7700000000,12,5,34,4335,1557,153,485,832,105,161,0.98,Pittsburgh Pirates,Three Rivers Stadium,1650593,99,100,PIT,PIT,PIT
+1993,NL,SDN,SDP,W,7,162,81,61,101,N,,N,N,679,5503,1386,239,28,153,443,1046,92,41,,,772,675,4.2300000000,8,6,32,4311,1470,148,558,957,160,129,0.97,San Diego Padres,Jack Murphy Stadium,1375432,102,102,SDP,SDN,SDN
+1993,NL,SFN,SFG,W,2,162,81,103,59,N,,N,N,808,5557,1534,269,33,168,516,930,120,65,,,636,584,3.6100000000,4,9,50,4368,1385,168,442,982,101,169,0.98,San Francisco Giants,Candlestick Park,2606354,97,96,SFG,SFN,SFN
+1993,NL,SLN,STL,E,3,162,81,87,75,N,,N,N,758,5551,1508,262,34,118,588,882,153,72,,,744,660,4.0900000000,5,7,54,4359,1553,152,383,775,159,157,0.97,St. Louis Cardinals,Busch Stadium II,2844977,98,98,STL,SLN,SLN
+1994,AL,BAL,BAL,E,2,112,55,63,49,,,,,589,3856,1047,185,20,139,438,655,69,13,,,497,477,4.3100000000,13,3,37,2991,1005,131,351,666,57,103,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2535359,105,104,BAL,BAL,BAL
+1994,AL,BOS,BOS,E,4,115,64,54,61,,,,,552,3940,1038,222,19,120,404,723,81,38,,,621,564,4.9300000000,6,1,30,3087,1104,120,450,729,81,124,0.98,Boston Red Sox,Fenway Park II,1775818,105,105,BOS,BOS,BOS
+1994,AL,CAL,ANA,W,4,115,63,47,68,,,,,543,3943,1042,178,16,120,402,715,65,54,,,660,618,5.4200000000,11,3,21,3081,1149,150,436,682,76,110,0.98,California Angels,Anaheim Stadium,1512622,101,101,CAL,CAL,CAL
+1994,AL,CHA,CHW,C,1,113,53,67,46,,,,,633,3942,1133,175,39,121,497,568,77,27,,,498,444,3.9500000000,13,6,20,3034,964,115,377,754,79,91,0.98,Chicago White Sox,Comiskey Park II,1697398,99,98,CHW,CHA,CHA
+1994,AL,CLE,CLE,C,2,113,51,66,47,,,,,679,4022,1165,240,20,167,382,629,131,48,,,562,493,4.3600000000,17,5,21,3054,1097,94,404,666,90,119,0.98,Cleveland Indians,Jacobs Field,1995174,99,97,CLE,CLE,CLE
+1994,AL,DET,DET,E,5,115,58,53,62,,,,,652,3955,1048,216,25,161,520,897,46,33,,,671,609,5.3800000000,15,1,20,3054,1139,148,449,560,82,90,0.98,Detroit Tigers,Tiger Stadium,1184783,101,101,DET,DET,DET
+1994,AL,KCA,KCR,C,3,115,59,64,51,,,,,574,3911,1051,211,38,100,376,698,140,62,,,532,485,4.2300000000,5,3,38,3093,1018,95,392,717,80,102,0.98,Kansas City Royals,Kauffman Stadium,1400494,104,104,KCR,KCA,KCA
+1994,AL,MIN,MIN,C,4,113,59,53,60,,,,,594,3952,1092,239,23,103,359,635,94,30,,,688,634,5.6800000000,6,2,29,3015,1197,153,388,602,75,99,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1398565,100,102,MIN,MIN,MIN
+1994,AL,ML4,MIL,C,5,115,56,53,62,,,,,547,3978,1045,238,21,99,417,680,59,37,,,586,532,4.6200000000,11,1,23,3108,1071,127,421,577,85,130,0.98,Milwaukee Brewers,County Stadium,1268399,104,105,MIL,MIL,MIL
+1994,AL,NYA,NYY,E,1,113,57,70,43,,,,,670,3986,1155,238,16,139,530,660,55,40,,,534,491,4.3400000000,8,0,31,3057,1045,120,398,656,80,122,0.98,New York Yankees,Yankee Stadium II,1675556,97,96,NYY,NYA,NYA
+1994,AL,OAK,OAK,W,2,114,56,51,63,,,,,549,3885,1009,178,13,113,417,686,91,39,,,589,537,4.8200000000,12,3,23,3009,979,128,510,732,88,105,0.97,Oakland Athletics,Oakland Coliseum,1242692,91,92,OAK,OAK,OAK
+1994,AL,SEA,SEA,W,3,112,44,49,63,,,,,569,3883,1045,211,18,153,372,652,48,21,,,616,546,4.9900000000,13,4,21,2952,1051,109,486,763,95,102,0.97,Seattle Mariners,Kingdome,1104206,102,102,SEA,SEA,SEA
+1994,AL,TEX,TEX,W,1,114,63,52,62,,,,,613,3983,1114,198,27,124,437,730,82,35,,,697,619,5.4500000000,10,3,26,3069,1176,157,394,683,105,106,0.97,Texas Rangers,The Ballpark at Arlington,2503198,100,101,TEX,TEX,TEX
+1994,AL,TOR,TOR,E,3,115,59,55,60,,,,,566,3962,1064,210,30,115,387,691,79,26,,,579,535,4.7000000000,13,4,26,3075,1053,127,482,832,80,105,0.98,Toronto Blue Jays,Skydome,2907933,100,100,TOR,TOR,TOR
+1994,NL,ATL,ATL,E,2,114,55,68,46,,,,,542,3861,1031,198,18,137,377,668,48,31,,,448,407,3.5700000000,16,4,26,3078,929,76,378,865,81,85,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,2539240,102,100,ATL,ATL,ATL
+1994,NL,CHN,CHC,C,5,113,59,49,64,,,,,500,3918,1015,189,26,109,364,750,69,53,,,549,508,4.4700000000,5,1,27,3069,1054,120,392,717,81,110,0.98,Chicago Cubs,Wrigley Field,1845208,99,99,CHC,CHN,CHN
+1994,NL,CIN,CIN,C,1,115,60,66,48,,,,,609,3999,1142,211,36,124,388,738,119,51,,,490,436,3.7800000000,6,2,27,3114,1037,117,339,799,73,91,0.98,Cincinnati Reds,Riverfront Stadium,1897681,99,99,CIN,CIN,CIN
+1994,NL,COL,COL,W,3,117,57,53,64,,,,,573,4006,1098,206,39,125,378,761,91,53,,,638,590,5.1500000000,4,1,28,3093,1185,120,448,703,84,117,0.98,Colorado Rockies,Mile High Stadium,3281511,117,118,COL,COL,COL
+1994,NL,FLO,FLA,E,5,115,59,51,64,,,,,468,3926,1043,180,24,94,349,746,65,26,,,576,508,4.5000000000,5,3,30,3045,1069,120,428,649,95,111,0.97,Florida Marlins,Joe Robbie Stadium,1937467,102,103,FLA,FLO,FLO
+1994,NL,HOU,HOU,C,2,115,59,66,49,,,,,602,3955,1099,252,25,120,394,718,124,44,,,503,454,3.9700000000,9,3,29,3087,1043,102,367,739,72,110,0.98,Houston Astros,Astrodome,1561136,95,94,HOU,HOU,HOU
+1994,NL,LAN,LAD,W,1,114,55,58,56,,,,,532,3904,1055,160,29,115,366,687,74,37,,,509,477,4.2300000000,14,4,20,3042,1041,90,354,732,88,104,0.98,Los Angeles Dodgers,Dodger Stadium,2279355,94,94,LAD,LAN,LAN
+1994,NL,MON,WSN,E,1,114,52,74,40,,,,,585,4000,1111,246,30,108,379,669,137,36,,,454,410,3.5600000000,4,2,46,3108,970,100,288,805,94,90,0.97,Montreal Expos,Stade Olympique,1276250,101,101,MON,MON,MON
+1994,NL,NYN,NYM,E,3,113,53,55,58,,,,,506,3869,966,164,21,117,336,807,25,26,,,526,469,4.1300000000,7,2,35,3069,1069,117,332,640,89,112,0.98,New York Mets,Shea Stadium,1151471,99,99,NYM,NYN,NYN
+1994,NL,PHI,PHI,E,4,115,60,54,61,,,,,521,3927,1028,208,28,80,396,711,67,24,,,497,438,3.8500000000,7,1,30,3072,1028,98,377,699,94,96,0.97,Philadelphia Phillies,Veterans Stadium,2290971,102,102,PHI,PHI,PHI
+1994,NL,PIT,PIT,C,3,114,61,53,61,,,,,466,3864,1001,198,23,80,349,725,53,25,,,580,518,4.6400000000,8,1,24,3015,1094,117,370,650,89,131,0.98,Pittsburgh Pirates,Three Rivers Stadium,1222520,101,102,PIT,PIT,PIT
+1994,NL,SDN,SDP,W,4,117,57,47,70,,,,,479,4068,1117,200,19,92,319,762,79,37,,,531,474,4.0800000000,8,3,27,3135,1008,99,393,862,111,82,0.97,San Diego Padres,Jack Murphy Stadium,953857,97,98,SDP,SDN,SDN
+1994,NL,SFN,SFG,W,2,115,60,55,60,,,,,504,3869,963,159,32,123,364,719,114,40,,,500,454,3.9900000000,2,0,33,3075,1014,122,372,655,68,113,0.98,San Francisco Giants,Candlestick Park,1704608,94,94,SFG,SFN,SFN
+1994,NL,SLN,STL,C,3,115,56,53,61,,,,,535,3902,1026,213,27,108,434,686,76,46,,,621,581,5.1400000000,7,3,29,3054,1154,134,355,632,80,119,0.98,St. Louis Cardinals,Busch Stadium II,1866544,98,99,STL,SLN,SLN
+1995,AL,BAL,BAL,E,3,144,72,71,73,N,N,N,N,704,4837,1267,229,27,173,574,803,92,45,,,640,607,4.3100000000,19,10,29,3801,1165,149,523,930,72,141,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3098475,102,101,BAL,BAL,BAL
+1995,AL,BOS,BOS,E,1,144,72,86,58,Y,N,N,N,791,4997,1399,286,31,175,560,923,99,44,,,698,630,4.3900000000,7,9,39,3876,1338,127,476,888,120,151,0.97,Boston Red Sox,Fenway Park II,2164410,103,103,BOS,BOS,BOS
+1995,AL,CAL,ANA,W,2,145,72,78,67,N,N,N,N,801,5019,1390,252,25,186,564,889,58,39,,,697,645,4.5200000000,8,9,42,3852,1310,163,486,901,95,120,0.98,California Angels,Anaheim Stadium,1748680,99,99,CAL,CAL,CAL
+1995,AL,CHA,CHW,C,3,145,72,68,76,N,N,N,N,755,5060,1417,252,37,146,576,767,110,39,,,758,692,4.8500000000,12,4,36,3852,1374,164,617,892,108,131,0.98,Chicago White Sox,Comiskey Park II,1609773,96,95,CHW,CHA,CHA
+1995,AL,CLE,CLE,C,1,144,72,100,44,Y,N,Y,N,840,5028,1461,279,23,207,542,766,132,53,,,607,554,3.8300000000,10,10,50,3903,1261,135,445,926,101,142,0.98,Cleveland Indians,Jacobs Field,2842745,101,99,CLE,CLE,CLE
+1995,AL,DET,DET,E,4,144,72,60,84,N,N,N,N,654,4865,1204,228,29,159,551,987,73,36,,,844,778,5.4900000000,5,3,38,3825,1509,170,536,729,106,143,0.98,Detroit Tigers,Tiger Stadium,1180979,100,101,DET,DET,DET
+1995,AL,KCA,KCR,C,2,144,72,70,74,N,N,N,N,629,4903,1275,240,35,119,475,849,120,53,,,691,643,4.4900000000,11,10,37,3864,1323,142,503,763,90,168,0.98,Kansas City Royals,Kauffman Stadium,1233530,101,101,KCR,KCA,KCA
+1995,AL,MIN,MIN,C,5,144,72,56,88,N,N,N,N,703,5005,1398,270,34,120,471,916,105,57,,,889,814,5.7600000000,7,2,27,3816,1450,210,533,790,100,141,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1057667,101,102,MIN,MIN,MIN
+1995,AL,ML4,MIL,C,4,144,72,65,79,N,N,N,N,740,5000,1329,249,42,128,502,800,105,40,,,747,689,4.8200000000,7,4,31,3858,1391,146,603,699,105,186,0.98,Milwaukee Brewers,County Stadium,1087560,105,105,MIL,MIL,MIL
+1995,AL,NYA,NYY,E,2,145,73,79,65,N,Y,N,N,749,4947,1365,280,34,122,625,851,50,30,,,688,651,4.5600000000,18,5,35,3852,1286,159,535,908,73,121,0.98,New York Yankees,Yankee Stadium II,1705263,99,98,NYY,NYA,NYA
+1995,AL,OAK,OAK,W,4,144,72,67,77,N,N,N,N,730,4916,1296,228,18,169,565,911,112,46,,,761,697,4.9300000000,8,4,34,3819,1320,153,556,890,101,151,0.98,Oakland Athletics,Oakland Coliseum,1174310,91,91,OAK,OAK,OAK
+1995,AL,SEA,SEA,W,1,145,73,79,66,Y,N,N,N,796,4996,1377,276,20,182,549,871,110,41,,,708,644,4.5000000000,9,8,39,3868,1343,149,591,1068,104,108,0.98,Seattle Mariners,Kingdome,1643203,101,101,SEA,SEA,SEA
+1995,AL,TEX,TEX,W,3,144,72,74,70,N,N,N,N,691,4913,1304,247,24,138,526,877,90,47,,,720,665,4.6600000000,14,4,34,3855,1385,152,514,838,98,156,0.98,Texas Rangers,The Ballpark at Arlington,1985910,103,103,TEX,TEX,TEX
+1995,AL,TOR,TOR,E,5,144,72,56,88,N,N,N,N,642,5036,1309,275,27,140,492,906,75,16,,,777,701,4.8800000000,16,8,22,3876,1336,145,654,894,97,131,0.98,Toronto Blue Jays,Skydome,2826483,99,100,TOR,TOR,TOR
+1995,NL,ATL,ATL,E,1,144,72,90,54,Y,N,Y,Y,645,4814,1202,210,27,168,520,933,73,43,,,540,493,3.4400000000,18,11,34,3873,1184,107,436,1087,100,113,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,2561831,103,102,ATL,ATL,ATL
+1995,NL,CHN,CHC,C,3,144,72,73,71,N,N,N,N,693,4963,1315,267,39,158,440,953,105,37,,,671,597,4.1300000000,6,12,45,3903,1313,162,518,926,115,115,0.97,Chicago Cubs,Wrigley Field,1918265,98,98,CHC,CHN,CHN
+1995,NL,CIN,CIN,C,1,144,72,85,59,Y,N,N,N,747,4903,1326,277,35,161,519,946,190,68,,,623,577,4.0300000000,8,10,38,3867,1270,131,424,903,79,140,0.98,Cincinnati Reds,Riverfront Stadium,1837649,100,99,CIN,CIN,CIN
+1995,NL,COL,COL,W,2,144,72,77,67,N,Y,N,N,785,4994,1406,259,43,200,484,943,125,59,,,783,711,4.9700000000,1,1,43,3864,1443,160,512,891,107,146,0.98,Colorado Rockies,Coors Field,3390037,129,129,COL,COL,COL
+1995,NL,FLO,FLA,E,4,143,71,67,76,N,N,N,N,673,4886,1278,214,29,144,517,916,131,53,,,673,610,4.2700000000,12,7,29,3858,1299,139,562,994,115,143,0.97,Florida Marlins,Joe Robbie Stadium,1700466,101,101,FLA,FLO,FLO
+1995,NL,HOU,HOU,C,2,144,72,76,68,N,N,N,N,747,5097,1403,260,22,109,566,992,176,60,,,674,595,4.0600000000,6,8,32,3960,1357,118,460,1056,121,120,0.97,Houston Astros,Astrodome,1363801,93,93,HOU,HOU,HOU
+1995,NL,LAN,LAD,W,1,144,72,78,66,Y,N,N,N,634,4942,1303,191,31,140,468,1023,127,45,,,609,527,3.6600000000,16,11,37,3885,1188,125,462,1060,130,120,0.97,Los Angeles Dodgers,Dodger Stadium,2766251,91,91,LAD,LAN,LAN
+1995,NL,MON,WSN,E,5,144,72,66,78,N,N,N,N,621,4905,1268,265,24,118,400,901,120,49,,,638,586,4.1100000000,7,9,42,3849,1286,128,416,950,109,119,0.98,Montreal Expos,Stade Olympique,1309618,104,103,MON,MON,MON
+1995,NL,NYN,NYM,E,2,144,72,69,75,N,N,N,N,657,4958,1323,218,34,125,446,994,58,39,,,618,557,3.8800000000,9,9,36,3873,1296,133,401,901,115,125,0.97,New York Mets,Shea Stadium,1273183,96,97,NYM,NYN,NYN
+1995,NL,PHI,PHI,E,2,144,72,69,75,N,N,N,N,615,4950,1296,263,30,94,497,884,72,25,,,658,603,4.2100000000,8,8,41,3870,1241,134,538,980,96,139,0.98,Philadelphia Phillies,Veterans Stadium,2043598,100,101,PHI,PHI,PHI
+1995,NL,PIT,PIT,C,5,144,72,58,86,N,N,N,N,629,4937,1281,245,27,125,456,972,84,55,,,736,666,4.7000000000,11,7,29,3825,1407,130,477,871,121,138,0.97,Pittsburgh Pirates,Three Rivers Stadium,905517,102,103,PIT,PIT,PIT
+1995,NL,SDN,SDP,W,3,144,72,70,74,N,N,N,N,668,4950,1345,231,20,116,447,872,124,46,,,672,589,4.1300000000,6,10,35,3852,1242,142,512,1047,108,130,0.98,San Diego Padres,Jack Murphy Stadium,1041805,97,97,SDP,SDN,SDN
+1995,NL,SFN,SFG,W,4,144,72,67,77,N,N,N,N,652,4971,1256,229,33,152,472,1060,138,46,,,776,698,4.8600000000,12,5,34,3879,1368,173,505,801,106,142,0.98,San Francisco Giants,Candlestick Park,1241500,96,97,SFG,SFN,SFN
+1995,NL,SLN,STL,C,4,143,72,62,81,N,N,N,N,563,4779,1182,238,24,107,436,920,79,46,,,658,575,4.0900000000,4,6,38,3795,1290,135,445,842,113,156,0.98,St. Louis Cardinals,Busch Stadium II,1756727,99,100,STL,SLN,SLN
+1996,AL,BAL,BAL,E,2,163,82,88,74,N,Y,N,N,949,5689,1557,299,29,257,645,915,76,40,,,903,840,5.1500000000,13,2,44,4404,1604,209,597,1047,97,173,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3646950,100,99,BAL,BAL,BAL
+1996,AL,BOS,BOS,E,3,162,81,85,77,N,N,N,N,928,5756,1631,308,31,209,642,1020,91,44,,,921,810,5.0000000000,17,5,37,4374,1606,185,722,1165,135,152,0.97,Boston Red Sox,Fenway Park II,2315231,101,101,BOS,BOS,BOS
+1996,AL,CAL,ANA,W,4,161,81,70,91,N,N,N,N,762,5686,1571,256,24,192,527,974,53,39,,,943,849,5.3100000000,12,8,38,4317,1546,219,662,1052,128,156,0.97,California Angels,Anaheim Stadium,1820521,98,98,CAL,CAL,CAL
+1996,AL,CHA,CHW,C,2,162,81,85,77,N,N,N,N,898,5644,1586,284,33,195,701,927,105,41,,,794,735,4.5300000000,7,5,43,4383,1529,174,616,1039,109,145,0.98,Chicago White Sox,Comiskey Park II,1676403,95,95,CHW,CHA,CHA
+1996,AL,CLE,CLE,C,1,161,80,99,62,Y,N,N,N,952,5681,1665,335,23,218,671,844,160,50,,,769,702,4.3500000000,13,9,46,4356,1530,173,484,1033,124,156,0.98,Cleveland Indians,Jacobs Field,3318174,99,98,CLE,CLE,CLE
+1996,AL,DET,DET,E,5,162,81,53,109,N,N,N,N,783,5530,1413,257,21,204,546,1268,87,50,,,1103,1015,6.3800000000,10,4,22,4296,1699,241,784,957,135,157,0.97,Detroit Tigers,Tiger Stadium,1168610,100,102,DET,DET,DET
+1996,AL,KCA,KCR,C,5,161,80,75,86,N,N,N,N,746,5542,1477,286,38,123,529,943,195,85,,,786,733,4.5500000000,17,8,35,4350,1563,176,460,926,111,184,0.98,Kansas City Royals,Kauffman Stadium,1435997,99,100,KCR,KCA,KCA
+1996,AL,MIN,MIN,C,4,162,82,78,84,N,N,N,N,877,5673,1633,332,47,118,576,958,143,53,,,900,847,5.3000000000,13,5,31,4317,1561,233,581,959,94,142,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1437352,101,102,MIN,MIN,MIN
+1996,AL,ML4,MIL,C,3,162,81,80,82,N,N,N,N,894,5662,1578,304,40,178,624,986,101,48,,,899,831,5.1700000000,6,5,42,4341,1570,213,635,846,134,180,0.97,Milwaukee Brewers,County Stadium,1327155,104,104,MIL,MIL,MIL
+1996,AL,NYA,NYY,E,1,162,80,92,70,Y,N,Y,Y,871,5628,1621,293,28,162,632,909,96,46,,,787,744,4.6500000000,6,9,52,4320,1469,143,610,1139,91,146,0.98,New York Yankees,Yankee Stadium II,2250877,101,100,NYY,NYA,NYA
+1996,AL,OAK,OAK,W,3,162,81,78,84,N,N,N,N,861,5630,1492,283,21,243,640,1114,58,35,,,900,841,5.2000000000,7,5,34,4368,1638,205,644,884,102,195,0.98,Oakland Athletics,Oakland Coliseum,1148380,101,102,OAK,OAK,OAK
+1996,AL,SEA,SEA,W,2,161,81,85,76,N,N,N,N,993,5668,1625,343,19,245,670,1052,90,39,,,895,828,5.2100000000,4,4,34,4293,1562,216,605,1000,110,155,0.98,Seattle Mariners,Kingdome,2723850,100,99,SEA,SEA,SEA
+1996,AL,TEX,TEX,W,1,163,81,90,72,Y,N,N,N,928,5702,1622,323,32,221,660,1041,83,26,,,799,750,4.6600000000,19,6,43,4347,1569,168,582,976,87,150,0.98,Texas Rangers,The Ballpark at Arlington,2889020,105,105,TEX,TEX,TEX
+1996,AL,TOR,TOR,E,4,162,81,74,88,N,N,N,N,766,5599,1451,302,35,177,529,1105,116,38,,,809,735,4.5800000000,19,7,35,4335,1476,187,610,1033,110,187,0.98,Toronto Blue Jays,Skydome,2559573,99,100,TOR,TOR,TOR
+1996,NL,ATL,ATL,E,1,162,81,96,66,Y,N,Y,N,773,5614,1514,264,28,197,530,1032,83,43,,,648,578,3.5400000000,14,9,46,4407,1372,120,451,1245,130,143,0.98,Atlanta Braves,Atlanta-Fulton County Stadium,2901242,106,104,ATL,ATL,ATL
+1996,NL,CHN,CHC,C,4,162,81,76,86,N,N,N,N,772,5531,1388,267,19,175,523,1090,108,50,,,771,705,4.3600000000,10,11,34,4368,1447,184,546,1027,104,147,0.98,Chicago Cubs,Wrigley Field,2219110,103,103,CHC,CHN,CHN
+1996,NL,CIN,CIN,C,3,162,81,81,81,N,N,N,N,778,5455,1398,259,36,191,604,1134,171,63,,,773,694,4.3300000000,6,8,52,4329,1447,167,591,1089,121,145,0.98,Cincinnati Reds,Riverfront Stadium,1861428,101,101,CIN,CIN,CIN
+1996,NL,COL,COL,W,3,162,81,83,79,N,N,N,N,961,5590,1607,297,37,221,527,1108,201,66,,,964,885,5.6000000000,5,5,34,4266,1597,198,624,932,149,167,0.97,Colorado Rockies,Coors Field,3891014,123,124,COL,COL,COL
+1996,NL,FLO,FLA,E,3,162,81,80,82,N,N,N,N,688,5498,1413,240,30,150,553,1122,99,46,,,703,633,3.9500000000,8,13,41,4329,1386,113,598,1050,111,187,0.98,Florida Marlins,Joe Robbie Stadium,1746767,97,97,FLA,FLO,FLO
+1996,NL,HOU,HOU,C,2,162,81,82,80,N,N,N,N,753,5508,1445,297,29,129,554,1057,180,63,,,792,704,4.3800000000,13,4,35,4341,1541,154,539,1163,138,130,0.97,Houston Astros,Astrodome,1975888,92,92,HOU,HOU,HOU
+1996,NL,LAN,LAD,W,2,162,81,90,72,N,Y,N,N,703,5538,1396,215,33,150,516,1190,124,40,,,652,567,3.4800000000,6,9,50,4398,1378,125,534,1212,125,143,0.98,Los Angeles Dodgers,Dodger Stadium,3188454,93,92,LAD,LAN,LAN
+1996,NL,MON,WSN,E,2,162,81,88,74,N,N,N,N,741,5505,1441,297,27,148,492,1077,108,34,,,668,605,3.7800000000,11,7,43,4324,1353,152,482,1206,126,121,0.98,Montreal Expos,Stade Olympique,1616709,102,102,MON,MON,MON
+1996,NL,NYN,NYM,E,4,162,81,71,91,N,N,N,N,746,5618,1515,267,47,147,445,1069,97,48,,,779,675,4.2200000000,10,10,41,4320,1517,159,532,999,157,163,0.97,New York Mets,Shea Stadium,1588323,95,95,NYM,NYN,NYN
+1996,NL,PHI,PHI,E,5,162,81,67,95,N,N,N,N,650,5499,1405,249,39,132,536,1092,117,41,,,790,710,4.4900000000,12,6,42,4269,1463,160,510,1044,115,145,0.98,Philadelphia Phillies,Veterans Stadium,1801677,100,101,PHI,PHI,PHI
+1996,NL,PIT,PIT,C,5,162,80,73,89,N,N,N,N,776,5665,1509,319,33,138,510,989,126,49,,,833,750,4.6400000000,5,7,37,4360,1602,183,479,1044,128,144,0.98,Pittsburgh Pirates,Three Rivers Stadium,1332150,104,104,PIT,PIT,PIT
+1996,NL,SDN,SDP,W,1,162,81,91,71,Y,N,N,N,771,5655,1499,285,24,147,601,1014,109,55,,,682,617,3.7300000000,5,11,47,4467,1395,138,506,1194,118,136,0.98,San Diego Padres,Jack Murphy Stadium,2187886,95,95,SDP,SDN,SDN
+1996,NL,SFN,SFG,W,4,162,82,68,94,N,N,N,N,752,5533,1400,245,21,153,615,1189,113,53,,,862,756,4.7200000000,9,8,35,4326,1520,194,570,997,134,165,0.97,San Francisco Giants,Candlestick Park,1413922,95,96,SFG,SFN,SFN
+1996,NL,SLN,STL,C,1,162,81,88,74,Y,N,N,N,759,5502,1468,281,31,142,495,1089,149,58,,,706,642,3.9800000000,13,11,43,4356,1380,173,539,1050,125,139,0.98,St. Louis Cardinals,Busch Stadium II,2654718,100,100,STL,SLN,SLN
+1997,AL,ANA,ANA,W,2,162,82,84,78,N,N,N,N,829,5628,1531,279,25,161,617,953,126,72,,,794,730,4.5200000000,9,5,39,4362,1506,202,605,1050,126,140,0.98,Anaheim Angels,Edison International Field,1767330,102,102,ANA,ANA,ANA
+1997,AL,BAL,BAL,E,1,162,81,98,64,Y,N,N,N,812,5584,1498,264,22,196,586,952,63,26,,,681,635,3.9100000000,8,10,59,4383,1404,164,563,1139,97,148,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3711132,97,96,BAL,BAL,BAL
+1997,AL,BOS,BOS,E,4,162,81,78,84,N,N,N,N,851,5781,1684,373,32,185,514,1044,68,48,,,857,785,4.8700000000,7,4,40,4355,1567,149,610,987,136,179,0.97,Boston Red Sox,Fenway Park II,2226136,102,102,BOS,BOS,BOS
+1997,AL,CHA,CHW,C,2,161,81,80,81,N,N,N,N,779,5491,1498,260,28,158,569,901,106,52,,,833,749,4.7400000000,6,7,52,4266,1505,175,575,961,125,131,0.97,Chicago White Sox,Comiskey Park II,1864782,96,96,CHW,CHA,CHA
+1997,AL,CLE,CLE,C,1,161,81,86,75,Y,N,Y,N,868,5556,1589,301,22,220,617,955,118,59,,,815,749,4.7300000000,4,3,39,4275,1528,181,575,1036,108,159,0.98,Cleveland Indians,Jacobs Field,3404750,103,102,CLE,CLE,CLE
+1997,AL,DET,DET,E,3,162,81,79,83,N,N,N,N,784,5481,1415,268,32,176,578,1164,161,72,,,790,732,4.5600000000,13,8,42,4335,1476,178,552,982,91,146,0.98,Detroit Tigers,Tiger Stadium,1365157,99,100,DET,DET,DET
+1997,AL,KCA,KCR,C,5,161,80,67,94,N,N,N,N,747,5599,1478,256,35,158,561,1061,130,66,,,820,755,4.7100000000,11,5,29,4329,1530,186,531,961,88,168,0.98,Kansas City Royals,Kauffman Stadium,1517638,101,102,KCR,KCA,KCA
+1997,AL,MIN,MIN,C,4,162,81,68,94,N,N,N,N,772,5634,1522,305,40,132,495,1121,151,52,,,861,800,5.0200000000,10,4,30,4302,1596,187,495,908,100,170,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1411064,101,101,MIN,MIN,MIN
+1997,AL,ML4,MIL,C,3,161,80,78,83,N,N,N,N,681,5444,1415,294,27,135,494,967,103,55,,,742,670,4.2200000000,6,8,44,4282,1419,177,542,1012,125,171,0.98,Milwaukee Brewers,County Stadium,1444027,101,102,MIL,MIL,MIL
+1997,AL,NYA,NYY,E,2,162,80,96,66,N,Y,N,N,891,5710,1636,325,23,161,676,954,99,58,,,688,626,3.8400000000,11,10,51,4401,1463,144,532,1165,105,156,0.98,New York Yankees,Yankee Stadium II,2580325,100,98,NYY,NYA,NYA
+1997,AL,OAK,OAK,W,4,162,81,65,97,N,N,N,N,764,5589,1451,274,23,197,642,1181,71,36,,,946,881,5.4900000000,2,1,38,4335,1734,197,642,953,121,170,0.98,Oakland Athletics,Oakland Coliseum,1264218,96,97,OAK,OAK,OAK
+1997,AL,SEA,SEA,W,1,162,81,90,72,Y,N,N,N,925,5614,1574,312,21,264,626,1110,89,40,,,833,770,4.7900000000,9,8,38,4341,1500,192,598,1207,123,143,0.98,Seattle Mariners,Kingdome,3192237,98,98,SEA,SEA,SEA
+1997,AL,TEX,TEX,W,3,162,81,77,85,N,N,N,N,807,5651,1547,311,27,187,500,1116,72,37,,,823,744,4.6800000000,8,9,33,4289,1598,169,541,925,122,155,0.98,Texas Rangers,The Ballpark at Arlington,2945228,105,105,TEX,TEX,TEX
+1997,AL,TOR,TOR,E,5,162,81,76,86,N,N,N,N,654,5473,1333,275,41,147,487,1138,134,50,,,694,628,3.9200000000,19,16,34,4326,1453,167,497,1150,97,150,0.98,Toronto Blue Jays,Skydome,2589297,99,99,TOR,TOR,TOR
+1997,NL,ATL,ATL,E,1,162,81,101,61,Y,N,N,N,791,5528,1490,268,37,174,597,1160,108,58,,,581,518,3.1800000000,21,17,37,4395,1319,111,450,1196,114,136,0.98,Atlanta Braves,Turner Field,3464488,102,99,ATL,ATL,ATL
+1997,NL,CHN,CHC,C,5,162,81,68,94,N,N,N,N,687,5489,1444,269,39,127,451,1003,116,60,,,759,705,4.4400000000,6,4,37,4287,1451,185,590,1072,115,117,0.98,Chicago Cubs,Wrigley Field,2190308,102,102,CHC,CHN,CHN
+1997,NL,CIN,CIN,C,3,162,81,76,86,N,N,N,N,651,5484,1386,269,27,142,518,1113,190,67,,,764,712,4.4200000000,5,8,49,4347,1408,173,558,1159,106,129,0.98,Cincinnati Reds,Cinergy Field,1785788,102,102,CIN,CIN,CIN
+1997,NL,COL,COL,W,3,162,81,83,79,N,N,N,N,923,5603,1611,269,40,239,562,1060,137,65,,,908,835,5.2500000000,9,5,38,4296,1697,196,566,870,111,202,0.98,Colorado Rockies,Coors Field,3888453,122,123,COL,COL,COL
+1997,NL,FLO,FLA,E,2,162,81,92,70,N,Y,Y,Y,740,5439,1410,272,28,136,686,1074,115,58,,,669,615,3.8300000000,12,10,39,4338,1353,131,639,1188,119,167,0.98,Florida Marlins,Joe Robbie Stadium,2364387,95,96,FLA,FLO,FLO
+1997,NL,HOU,HOU,C,1,162,81,84,78,Y,N,N,N,777,5502,1427,314,40,133,633,1085,171,74,,,660,595,3.6700000000,16,12,37,4377,1379,134,511,1138,131,169,0.97,Houston Astros,Astrodome,2046781,96,95,HOU,HOU,HOU
+1997,NL,LAN,LAD,W,2,162,81,88,74,N,N,N,N,742,5544,1488,242,33,174,498,1079,131,64,,,645,588,3.6300000000,6,6,45,4377,1325,163,546,1232,116,104,0.98,Los Angeles Dodgers,Dodger Stadium,3319504,93,92,LAD,LAN,LAN
+1997,NL,MON,WSN,E,4,162,81,78,84,N,N,N,N,691,5526,1423,339,34,172,420,1084,75,46,,,740,666,4.1400000000,27,14,37,4341,1365,149,557,1138,131,150,0.97,Montreal Expos,Stade Olympique,1497609,99,99,MON,MON,MON
+1997,NL,NYN,NYM,E,3,162,81,88,74,N,N,N,N,777,5524,1448,274,28,153,550,1029,97,74,,,709,640,3.9500000000,7,8,49,4377,1452,160,504,982,117,165,0.98,New York Mets,Shea Stadium,1766174,97,97,NYM,NYN,NYN
+1997,NL,PHI,PHI,E,5,162,81,68,94,N,N,N,N,668,5443,1390,290,35,116,519,1032,92,56,,,840,768,4.8700000000,13,7,35,4260,1441,171,616,1209,105,134,0.98,Philadelphia Phillies,Veterans Stadium,1490638,100,101,PHI,PHI,PHI
+1997,NL,PIT,PIT,C,2,162,81,79,83,N,N,N,N,725,5503,1440,291,52,129,481,1161,160,50,,,760,683,4.2800000000,6,8,41,4308,1503,143,560,1080,131,149,0.97,Pittsburgh Pirates,Three Rivers Stadium,1657022,103,103,PIT,PIT,PIT
+1997,NL,SDN,SDP,W,4,162,81,76,86,N,N,N,N,795,5609,1519,275,16,152,604,1129,140,60,,,891,804,4.9900000000,5,2,43,4350,1581,172,596,1059,125,132,0.98,San Diego Padres,Qualcomm Stadium,2089333,93,93,SDP,SDN,SDN
+1997,NL,SFN,SFG,W,1,162,81,90,72,Y,N,N,N,784,5485,1415,266,37,172,642,1120,121,49,,,793,709,4.4100000000,5,9,45,4338,1494,160,578,1044,120,157,0.98,San Francisco Giants,3Com Park,1690869,98,98,SFG,SFN,SFN
+1997,NL,SLN,STL,C,4,162,81,73,89,N,N,N,N,689,5524,1409,269,39,144,543,1191,164,60,,,708,631,3.9000000000,5,3,39,4367,1422,124,536,1130,123,156,0.98,St. Louis Cardinals,Busch Stadium II,2634014,99,99,STL,SLN,SLN
+1998,AL,ANA,ANA,W,2,162,81,85,77,N,N,N,N,787,5630,1530,314,27,147,510,1028,93,45,,,783,720,4.4900000000,3,5,52,4332,1481,164,630,1091,106,146,0.98,Anaheim Angels,Edison International Field,2519280,102,102,ANA,ANA,ANA
+1998,AL,BAL,BAL,E,4,162,81,79,83,N,N,N,N,817,5565,1520,303,11,214,593,903,86,48,,,785,752,4.7300000000,16,10,37,4293,1505,169,535,1065,81,144,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3684650,98,97,BAL,BAL,BAL
+1998,AL,BOS,BOS,E,2,162,81,92,70,N,Y,N,N,876,5601,1568,338,35,205,541,1049,72,39,,,729,669,4.1900000000,5,8,53,4308,1406,168,504,1025,105,128,0.98,Boston Red Sox,Fenway Park II,2314704,102,101,BOS,BOS,BOS
+1998,AL,CHA,CHW,C,2,163,81,80,82,N,N,N,N,861,5585,1516,291,38,198,551,916,127,46,,,931,837,5.2400000000,8,4,42,4315,1569,211,580,911,140,161,0.97,Chicago White Sox,Comiskey Park II,1391146,98,98,CHW,CHA,CHA
+1998,AL,CLE,CLE,C,1,162,81,89,73,Y,N,N,N,850,5616,1530,334,30,198,630,1061,143,60,,,779,722,4.4500000000,9,4,47,4380,1552,171,563,1037,110,146,0.98,Cleveland Indians,Jacobs Field,3467299,103,102,CLE,CLE,CLE
+1998,AL,DET,DET,C,5,162,81,65,97,N,N,N,N,722,5664,1494,306,29,165,455,1070,122,62,,,863,792,4.9300000000,9,4,32,4338,1551,185,595,947,115,164,0.98,Detroit Tigers,Tiger Stadium,1409391,100,101,DET,DET,DET
+1998,AL,KCA,KCR,C,3,161,80,72,89,N,N,N,N,714,5546,1459,274,40,134,475,984,135,50,,,899,823,5.1600000000,6,7,46,4308,1590,196,568,999,125,172,0.98,Kansas City Royals,Kauffman Stadium,1494875,101,102,KCR,KCA,KCA
+1998,AL,MIN,MIN,C,4,162,81,70,92,N,N,N,N,734,5641,1499,285,32,115,506,915,112,54,,,818,765,4.7600000000,7,9,42,4342,1622,180,457,952,108,135,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1165976,101,102,MIN,MIN,MIN
+1998,AL,NYA,NYY,E,1,162,81,114,48,Y,N,Y,Y,965,5643,1625,290,31,207,653,1025,153,63,,,656,618,3.8200000000,22,16,48,4369,1357,156,466,1080,98,146,0.98,New York Yankees,Yankee Stadium II,2955193,97,95,NYY,NYA,NYA
+1998,AL,OAK,OAK,W,4,162,81,74,88,N,N,N,N,804,5490,1413,295,13,149,633,1122,131,47,,,866,770,4.8300000000,12,4,39,4302,1555,179,529,922,141,155,0.97,Oakland Athletics,Oakland Coliseum,1232343,96,97,OAK,OAK,OAK
+1998,AL,SEA,SEA,W,3,161,81,76,85,N,N,N,N,859,5628,1553,321,28,234,558,1081,115,39,,,855,783,4.9500000000,17,7,31,4272,1530,196,528,1156,125,139,0.97,Seattle Mariners,Kingdome,2651511,99,99,SEA,SEA,SEA
+1998,AL,TBA,TBD,E,5,162,81,63,99,N,N,N,N,620,5555,1450,267,43,111,473,1107,120,73,,,751,697,4.3500000000,7,8,28,4329,1425,171,643,1008,94,178,0.98,Tampa Bay Devil Rays,Tropicana Field,2506293,101,102,TBD,TBA,TBA
+1998,AL,TEX,TEX,W,1,162,81,88,74,Y,N,N,N,940,5672,1637,314,32,201,595,1045,82,47,,,871,795,5.0000000000,10,8,46,4293,1624,164,519,994,122,140,0.98,Texas Rangers,The Ballpark at Arlington,2927399,104,103,TEX,TEX,TEX
+1998,AL,TOR,TOR,E,3,163,81,88,74,N,N,N,N,816,5580,1482,316,19,221,564,1132,184,81,,,768,698,4.2900000000,10,11,47,4395,1443,169,587,1154,125,131,0.97,Toronto Blue Jays,Skydome,2454303,100,99,TOR,TOR,TOR
+1998,NL,ARI,ARI,W,5,162,81,65,97,N,N,N,N,665,5491,1353,235,46,159,489,1239,73,38,,,812,738,4.6400000000,7,6,37,4296,1463,188,489,908,100,125,0.98,Arizona Diamondbacks,Bank One Ballpark,3610290,100,99,ARI,ARI,ARI
+1998,NL,ATL,ATL,E,1,162,81,106,56,Y,N,N,N,826,5484,1489,297,26,215,548,1062,98,43,,,581,519,3.2500000000,24,23,45,4315,1291,117,467,1232,91,139,0.98,Atlanta Braves,Turner Field,3360860,100,98,ATL,ATL,ATL
+1998,NL,CHN,CHC,C,2,163,82,90,73,N,Y,N,N,831,5649,1494,250,34,212,601,1223,65,44,,,792,739,4.5000000000,7,7,56,4431,1528,180,575,1207,101,107,0.98,Chicago Cubs,Wrigley Field,2623194,103,103,CHC,CHN,CHN
+1998,NL,CIN,CIN,C,4,162,81,77,85,N,N,N,N,750,5496,1441,298,28,138,608,1107,95,42,,,760,711,4.4400000000,6,8,42,4323,1400,170,573,1098,122,142,0.98,Cincinnati Reds,Cinergy Field,1793649,102,102,CIN,CIN,CIN
+1998,NL,COL,COL,W,4,162,81,77,85,N,N,N,N,826,5632,1640,333,36,183,469,949,67,47,,,855,796,5.0000000000,9,5,36,4297,1583,174,562,951,102,193,0.98,Colorado Rockies,Coors Field,3792683,121,122,COL,COL,COL
+1998,NL,FLO,FLA,E,5,162,81,54,108,N,N,N,N,667,5558,1381,277,36,114,525,1120,115,57,,,923,837,5.2000000000,11,3,24,4348,1617,182,715,1016,129,177,0.97,Florida Marlins,Joe Robbie Stadium,1730384,94,95,FLA,FLO,FLO
+1998,NL,HOU,HOU,C,1,162,81,102,60,Y,N,N,N,874,5641,1578,326,28,166,621,1122,155,51,,,620,572,3.5000000000,12,11,44,4413,1435,147,465,1187,108,144,0.98,Houston Astros,Astrodome,2458451,98,97,HOU,HOU,HOU
+1998,NL,LAN,LAD,W,3,162,81,83,79,N,N,N,N,669,5459,1374,209,27,159,447,1056,137,53,,,678,613,3.8100000000,16,10,47,4341,1332,135,587,1178,134,154,0.97,Los Angeles Dodgers,Dodger Stadium,3089222,95,95,LAD,LAN,LAN
+1998,NL,MIL,MIL,C,5,162,81,74,88,N,N,N,N,707,5541,1439,266,17,152,532,1039,81,59,,,812,746,4.6300000000,2,2,39,4353,1538,188,550,1063,110,192,0.98,Milwaukee Brewers,County Stadium,1811593,100,101,MIL,ML4,MIL
+1998,NL,MON,WSN,E,4,162,81,65,97,N,N,N,N,644,5418,1348,280,32,147,439,1058,91,46,,,783,696,4.3900000000,4,5,39,4281,1448,156,533,1017,155,127,0.97,Montreal Expos,Stade Olympique,914909,98,99,MON,MON,MON
+1998,NL,NYN,NYM,E,2,162,81,88,74,N,N,N,N,706,5510,1425,289,24,136,572,1049,62,46,,,645,611,3.7700000000,9,16,46,4374,1381,152,532,1129,101,151,0.98,New York Mets,Shea Stadium,2287948,99,99,NYM,NYN,NYN
+1998,NL,PHI,PHI,E,3,162,81,75,87,N,N,N,N,713,5617,1482,286,36,126,508,1080,97,45,,,808,754,4.6400000000,21,11,32,4389,1476,188,544,1176,110,131,0.98,Philadelphia Phillies,Veterans Stadium,1715722,102,103,PHI,PHI,PHI
+1998,NL,PIT,PIT,C,6,163,80,69,93,N,N,N,N,650,5493,1395,271,35,107,393,1060,159,51,,,718,630,3.9100000000,7,10,41,4347,1433,147,530,1112,140,161,0.97,Pittsburgh Pirates,Three Rivers Stadium,1560950,102,103,PIT,PIT,PIT
+1998,NL,SDN,SDP,W,1,162,81,98,64,Y,N,Y,N,749,5490,1390,292,30,167,604,1072,79,37,,,635,587,3.6300000000,14,11,59,4363,1384,139,501,1217,104,155,0.98,San Diego Padres,Qualcomm Stadium,2555874,92,92,SDP,SDN,SDN
+1998,NL,SFN,SFG,W,2,163,81,89,74,N,N,N,N,845,5628,1540,292,26,161,678,1040,102,51,,,739,688,4.1900000000,6,6,44,4431,1457,171,562,1089,101,157,0.98,San Francisco Giants,3Com Park,1925364,95,95,SFG,SFN,SFN
+1998,NL,SLN,STL,C,3,163,82,83,79,N,N,N,N,810,5593,1444,292,30,223,676,1179,133,41,,,782,705,4.3200000000,6,10,44,4409,1513,151,558,972,142,160,0.97,St. Louis Cardinals,Busch Stadium II,3195691,100,100,STL,SLN,SLN
+1999,AL,ANA,ANA,W,4,162,81,70,92,N,N,N,N,711,5494,1404,248,22,158,511,1022,71,45,,,826,762,4.7900000000,4,7,37,4293,1472,177,624,877,106,156,0.98,Anaheim Angels,Edison International Field,2253123,99,100,ANA,ANA,ANA
+1999,AL,BAL,BAL,E,4,162,81,78,84,N,N,N,N,851,5637,1572,299,21,203,615,890,107,46,,,815,761,4.7700000000,17,11,33,4305,1468,198,647,982,89,192,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3433150,96,96,BAL,BAL,BAL
+1999,AL,BOS,BOS,E,2,162,81,94,68,N,Y,N,N,836,5579,1551,334,42,176,597,928,67,39,,,718,638,4.0000000000,6,12,50,4309,1396,160,469,1131,127,131,0.97,Boston Red Sox,Fenway Park II,2446162,104,103,BOS,BOS,BOS
+1999,AL,CHA,CHW,C,2,162,81,75,86,N,N,N,N,777,5644,1563,298,37,162,499,810,110,50,,,870,786,4.9200000000,6,3,39,4314,1608,210,596,968,136,148,0.97,Chicago White Sox,Comiskey Park II,1338851,101,101,CHW,CHA,CHA
+1999,AL,CLE,CLE,C,1,162,81,97,65,Y,N,N,N,1009,5634,1629,309,32,209,743,1099,147,50,,,860,789,4.9000000000,3,6,46,4350,1503,197,634,1120,106,153,0.98,Cleveland Indians,Jacobs Field,3468456,104,103,CLE,CLE,CLE
+1999,AL,DET,DET,C,3,161,81,69,92,N,N,N,N,747,5481,1433,289,34,212,458,1049,108,70,,,882,817,5.1800000000,4,6,33,4261,1528,209,583,976,106,155,0.98,Detroit Tigers,Tiger Stadium,2026441,99,101,DET,DET,DET
+1999,AL,KCA,KCR,C,4,161,80,64,97,N,N,N,N,856,5624,1584,294,52,151,535,932,127,39,,,921,844,5.3500000000,11,3,29,4261,1607,202,643,831,125,189,0.98,Kansas City Royals,Kauffman Stadium,1506068,102,103,KCR,KCA,KCA
+1999,AL,MIN,MIN,C,5,161,81,63,97,N,N,N,N,686,5495,1450,285,30,105,500,978,118,60,,,845,794,5.0200000000,12,8,34,4269,1591,208,487,927,92,144,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1202829,103,104,MIN,MIN,MIN
+1999,AL,NYA,NYY,E,1,162,81,98,64,Y,N,Y,Y,900,5568,1568,302,36,193,718,978,104,57,,,731,660,4.1300000000,6,10,50,4318,1402,158,581,1111,111,130,0.98,New York Yankees,Yankee Stadium II,3292736,98,97,NYY,NYA,NYA
+1999,AL,OAK,OAK,W,2,162,81,87,75,N,N,N,N,893,5519,1430,287,20,235,770,1129,70,37,,,846,749,4.6900000000,6,5,48,4314,1537,160,569,967,122,162,0.98,Oakland Athletics,Oakland Coliseum,1434610,94,94,OAK,OAK,OAK
+1999,AL,SEA,SEA,W,3,162,81,79,83,N,N,N,N,859,5572,1499,263,21,244,610,1095,130,45,,,905,834,5.2400000000,7,6,40,4300,1613,191,684,980,113,180,0.98,Seattle Mariners,Kingdome / Safeco Field,2916346,102,103,SEA,SEA,SEA
+1999,AL,TBA,TBD,E,5,162,81,69,93,N,N,N,N,772,5586,1531,272,29,145,544,1042,73,49,,,913,805,5.0600000000,6,5,45,4297,1606,172,695,1055,135,198,0.97,Tampa Bay Devil Rays,Tropicana Field,1562827,100,102,TBD,TBA,TBA
+1999,AL,TEX,TEX,W,1,162,81,95,67,Y,N,N,N,945,5651,1653,304,29,230,611,937,111,54,,,859,809,5.0700000000,6,9,47,4306,1626,186,509,979,119,167,0.98,Texas Rangers,The Ballpark at Arlington,2771469,105,105,TEX,TEX,TEX
+1999,AL,TOR,TOR,E,3,162,81,84,78,N,N,N,N,883,5642,1580,337,14,212,578,1077,119,48,,,862,786,4.9200000000,14,9,39,4315,1582,191,575,1009,106,164,0.98,Toronto Blue Jays,Skydome,2163464,101,101,TOR,TOR,TOR
+1999,NL,ARI,ARI,W,1,162,81,100,62,Y,N,N,N,908,5658,1566,289,46,216,588,1045,137,39,,,676,615,3.7700000000,16,9,42,4401,1387,176,543,1198,104,129,0.98,Arizona Diamondbacks,Bank One Ballpark,3019654,101,101,ARI,ARI,ARI
+1999,NL,ATL,ATL,E,1,162,81,103,59,Y,N,Y,N,840,5569,1481,309,23,197,608,962,148,66,,,661,593,3.6300000000,9,9,45,4413,1398,142,507,1197,111,122,0.98,Atlanta Braves,Turner Field,3284897,100,98,ATL,ATL,ATL
+1999,NL,CHN,CHC,C,6,162,81,67,95,N,N,N,N,747,5482,1411,255,35,189,571,1170,60,44,,,920,837,5.2700000000,11,6,32,4291,1619,221,529,980,139,135,0.97,Chicago Cubs,Wrigley Field,2813854,97,98,CHC,CHN,CHN
+1999,NL,CIN,CIN,C,2,163,82,96,67,N,N,N,N,865,5649,1536,312,37,209,569,1125,164,54,,,711,648,3.9900000000,6,11,55,4384,1309,190,636,1081,105,137,0.98,Cincinnati Reds,Cinergy Field,2061222,103,103,CIN,CIN,CIN
+1999,NL,COL,COL,W,5,162,81,72,90,N,N,N,N,906,5716,1644,305,39,223,508,863,70,43,,,1028,955,6.0200000000,12,2,33,4285,1700,237,737,1032,118,186,0.98,Colorado Rockies,Coors Field,3481065,127,127,COL,COL,COL
+1999,NL,FLO,FLA,E,5,162,80,64,98,N,N,N,N,691,5578,1465,266,44,128,479,1145,92,46,,,852,781,4.9000000000,6,5,33,4306,1560,171,655,943,127,150,0.97,Florida Marlins,Pro Player Stadium,1369421,93,95,FLA,FLO,FLO
+1999,NL,HOU,HOU,C,1,162,82,97,65,Y,N,N,N,823,5485,1463,293,23,168,728,1138,166,75,,,675,622,3.8400000000,12,8,48,4375,1485,128,478,1204,106,168,0.98,Houston Astros,Astrodome,2706017,100,98,HOU,HOU,HOU
+1999,NL,LAN,LAD,W,3,162,81,77,85,N,N,N,N,793,5567,1480,253,23,187,594,1030,167,68,,,787,718,4.4500000000,8,6,37,4357,1438,192,594,1077,137,134,0.97,Los Angeles Dodgers,Dodger Stadium,3095346,95,94,LAD,LAN,LAN
+1999,NL,MIL,MIL,C,5,161,80,74,87,N,N,N,N,815,5582,1524,299,30,165,658,1065,81,33,,,886,812,5.0700000000,2,5,40,4327,1618,213,616,987,127,142,0.97,Milwaukee Brewers,County Stadium,1701796,99,99,MIL,ML4,MIL
+1999,NL,MON,WSN,E,4,162,81,68,94,N,N,N,N,718,5559,1473,320,47,163,438,939,70,51,,,853,747,4.6900000000,6,4,44,4302,1505,152,572,1043,160,126,0.97,Montreal Expos,Stade Olympique,773277,97,98,MON,MON,MON
+1999,NL,NYN,NYM,E,2,163,81,97,66,N,Y,N,N,853,5572,1553,297,14,181,717,994,150,61,,,711,691,4.2700000000,5,7,49,4369,1372,167,617,1172,68,147,0.98,New York Mets,Shea Stadium,2725668,98,97,NYM,NYN,NYN
+1999,NL,PHI,PHI,E,3,162,81,77,85,N,N,N,N,841,5598,1539,302,44,161,631,1081,125,35,,,846,788,4.9300000000,11,6,32,4314,1494,212,627,1030,100,143,0.98,Philadelphia Phillies,Veterans Stadium,1825337,105,105,PHI,PHI,PHI
+1999,NL,PIT,PIT,C,3,161,81,78,83,N,N,N,N,775,5468,1417,282,40,171,573,1197,112,44,,,782,689,4.3300000000,8,3,34,4297,1444,160,633,1083,147,174,0.97,Pittsburgh Pirates,Three Rivers Stadium,1638023,101,101,PIT,PIT,PIT
+1999,NL,SDN,SDP,W,4,162,81,74,88,N,N,N,N,710,5394,1360,256,22,153,631,1169,174,67,,,781,705,4.4700000000,5,6,43,4260,1454,193,529,1078,129,151,0.97,San Diego Padres,Qualcomm Stadium,2523538,92,92,SDP,SDN,SDN
+1999,NL,SFN,SFG,W,2,162,81,86,76,N,N,N,N,872,5563,1507,307,18,188,696,1028,109,56,,,831,762,4.7100000000,6,3,42,4368,1486,194,655,1076,105,150,0.98,San Francisco Giants,3Com Park,2078399,95,94,SFG,SFN,SFN
+1999,NL,SLN,STL,C,4,161,80,75,86,N,N,N,N,809,5570,1461,274,27,194,613,1202,134,48,,,838,761,4.7400000000,5,3,38,4335,1519,161,667,1025,132,161,0.97,St. Louis Cardinals,Busch Stadium II,3225334,101,101,STL,SLN,SLN
+2000,AL,ANA,ANA,W,3,162,81,82,80,N,N,N,N,864,5628,1574,309,34,236,608,1024,93,52,47,43,869,805,5.0000000000,5,3,46,4344,1534,228,662,846,134,184,0.97,Anaheim Angels,Edison International Field,2066982,102,103,ANA,ANA,ANA
+2000,AL,BAL,BAL,E,4,162,81,74,88,N,N,N,N,794,5549,1508,310,22,184,558,900,126,65,49,54,913,855,5.3700000000,14,6,33,4299,1547,202,665,1017,116,151,0.98,Baltimore Orioles,Oriole Park at Camden Yards,3297031,95,96,BAL,BAL,BAL
+2000,AL,BOS,BOS,E,2,162,81,85,77,N,N,N,N,792,5630,1503,316,32,167,611,1019,43,30,42,48,745,683,4.2300000000,7,12,46,4357,1433,173,499,1121,109,120,0.98,Boston Red Sox,Fenway Park II,2585895,104,103,BOS,BOS,BOS
+2000,AL,CHA,CHW,C,1,162,81,95,67,Y,N,N,N,978,5646,1615,325,33,216,591,960,119,42,53,61,839,751,4.6600000000,5,7,43,4350,1509,195,614,1037,133,190,0.97,Chicago White Sox,Comiskey Park II,1947799,102,102,CHW,CHA,CHA
+2000,AL,CLE,CLE,C,2,162,81,90,72,N,N,N,N,950,5683,1639,310,30,221,685,1057,113,34,51,52,816,775,4.8400000000,6,5,34,4326,1511,173,666,1213,72,149,0.98,Cleveland Indians,Jacobs Field,3456278,101,100,CLE,CLE,CLE
+2000,AL,DET,DET,C,3,162,81,79,83,N,N,N,N,823,5644,1553,307,41,177,562,982,83,38,43,49,827,755,4.7100000000,6,6,44,4329,1583,177,496,978,105,171,0.98,Detroit Tigers,Comerica Park,2438617,95,95,DET,DET,DET
+2000,AL,KCA,KCR,C,4,162,81,77,85,N,N,N,N,879,5709,1644,281,27,150,511,840,121,35,48,70,930,876,5.4800000000,10,6,29,4317,1585,239,693,927,102,185,0.98,Kansas City Royals,Kauffman Stadium,1564847,103,103,KCR,KCA,KCA
+2000,AL,MIN,MIN,C,5,162,81,69,93,N,N,N,N,748,5615,1516,325,49,116,556,1021,90,45,35,51,880,819,5.1500000000,6,4,35,4297,1634,212,516,1042,102,155,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1000760,104,105,MIN,MIN,MIN
+2000,AL,NYA,NYY,E,1,161,80,87,74,Y,N,Y,Y,871,5556,1541,294,25,205,631,1007,99,48,57,50,814,753,4.7600000000,9,6,40,4272,1458,177,577,1040,109,135,0.98,New York Yankees,Yankee Stadium II,3055435,99,98,NYY,NYA,NYA
+2000,AL,OAK,OAK,W,1,161,81,91,70,Y,N,N,N,947,5560,1501,281,23,239,750,1159,40,15,52,44,813,730,4.5800000000,7,11,43,4305,1535,158,615,963,134,164,0.97,Oakland Athletics,Oakland Coliseum,1603744,96,95,OAK,OAK,OAK
+2000,AL,SEA,SEA,W,2,162,81,91,71,N,Y,N,N,907,5497,1481,300,26,198,775,1073,122,56,48,61,780,720,4.5000000000,4,10,44,4324,1442,167,634,998,99,178,0.98,Seattle Mariners,Safeco Field,2914624,95,93,SEA,SEA,SEA
+2000,AL,TBA,TBD,E,5,161,80,69,92,N,N,N,N,733,5505,1414,253,22,162,559,1022,90,46,48,40,842,773,4.8600000000,10,8,38,4293,1553,198,533,955,118,171,0.98,Tampa Bay Devil Rays,Tropicana Field,1449673,99,100,TBD,TBA,TBA
+2000,AL,TEX,TEX,W,4,162,81,71,91,N,N,N,N,848,5648,1601,330,35,173,580,922,69,47,39,48,974,876,5.5200000000,3,4,39,4287,1683,202,661,918,135,162,0.97,Texas Rangers,The Ballpark at Arlington,2588401,102,102,TEX,TEX,TEX
+2000,AL,TOR,TOR,E,3,162,81,83,79,N,N,N,N,861,5677,1562,328,21,244,526,1026,89,34,60,34,908,821,5.1400000000,15,4,37,4311,1615,195,560,978,100,176,0.98,Toronto Blue Jays,Skydome,1705712,103,103,TOR,TOR,TOR
+2000,NL,ARI,ARI,W,3,162,81,85,77,N,N,N,N,792,5527,1466,282,44,179,535,975,97,44,59,58,754,698,4.3500000000,16,8,38,4330,1441,190,500,1220,107,138,0.98,Arizona Diamondbacks,Bank One Ballpark,2942251,105,103,ARI,ARI,ARI
+2000,NL,ATL,ATL,E,1,162,81,95,67,Y,N,N,N,810,5489,1490,274,26,179,595,1010,148,56,59,45,714,648,4.0500000000,13,9,53,4320,1428,165,484,1093,129,138,0.97,Atlanta Braves,Turner Field,3234304,101,99,ATL,ATL,ATL
+2000,NL,CHN,CHC,C,6,162,81,65,97,N,N,N,N,764,5577,1426,272,23,183,632,1120,93,37,54,45,904,849,5.2500000000,10,5,39,4363,1505,231,658,1143,100,139,0.98,Chicago Cubs,Wrigley Field,2789511,97,98,CHC,CHN,CHN
+2000,NL,CIN,CIN,C,2,163,82,85,77,N,N,N,N,825,5635,1545,302,36,200,559,995,100,38,64,58,765,700,4.3300000000,8,7,42,4368,1446,190,659,1015,111,156,0.98,Cincinnati Reds,Cinergy Field,2577371,102,102,CIN,CIN,CIN
+2000,NL,COL,COL,W,4,162,81,82,80,N,N,N,N,968,5660,1664,320,53,161,601,907,131,61,42,75,897,835,5.2600000000,7,2,33,4290,1568,221,588,1001,94,176,0.98,Colorado Rockies,Coors Field,3295129,125,125,COL,COL,COL
+2000,NL,FLO,FLA,E,3,161,81,79,82,N,N,N,N,731,5509,1441,274,29,160,540,1184,168,55,60,51,797,729,4.5900000000,5,4,48,4288,1477,169,650,1051,125,144,0.98,Florida Marlins,Pro Player Stadium,1218326,94,95,FLA,FLO,FLO
+2000,NL,HOU,HOU,C,4,162,81,72,90,N,N,N,N,938,5570,1547,289,36,249,673,1129,114,52,83,61,944,864,5.4100000000,8,2,30,4312,1596,234,598,1064,133,149,0.97,Houston Astros,Enron Field,3056139,107,107,HOU,HOU,HOU
+2000,NL,LAN,LAD,W,2,162,81,86,76,N,N,N,N,798,5481,1408,265,28,211,668,1083,95,42,51,46,729,659,4.1000000000,9,11,36,4335,1379,176,600,1154,135,151,0.97,Los Angeles Dodgers,Dodger Stadium,2880242,94,94,LAD,LAN,LAN
+2000,NL,MIL,MIL,C,3,163,81,73,89,N,N,N,N,740,5563,1366,297,25,177,620,1245,72,44,61,49,826,755,4.6400000000,2,7,29,4398,1501,174,728,967,118,187,0.98,Milwaukee Brewers,County Stadium,1573621,97,98,MIL,ML4,MIL
+2000,NL,MON,WSN,E,4,162,81,67,95,N,N,N,N,738,5535,1475,310,35,178,476,1048,58,48,29,34,902,812,5.1300000000,4,7,39,4273,1575,181,579,1011,132,151,0.97,Montreal Expos,Stade Olympique,926272,103,104,MON,MON,MON
+2000,NL,NYN,NYM,E,2,162,81,94,68,N,Y,Y,N,807,5486,1445,282,20,198,675,1037,66,46,45,51,738,670,4.1600000000,8,10,49,4350,1398,164,574,1164,118,121,0.98,New York Mets,Shea Stadium,2820530,96,96,NYM,NYN,NYN
+2000,NL,PHI,PHI,E,5,162,81,65,97,N,N,N,N,708,5511,1386,304,40,144,611,1117,102,30,44,37,830,763,4.7700000000,8,6,34,4315,1458,201,640,1123,100,136,0.98,Philadelphia Phillies,Veterans Stadium,1612769,100,101,PHI,PHI,PHI
+2000,NL,PIT,PIT,C,5,162,81,69,93,N,N,N,N,793,5643,1506,320,31,168,564,1032,86,40,66,37,888,794,4.9300000000,5,7,27,4347,1554,163,711,1070,132,169,0.97,Pittsburgh Pirates,Three Rivers Stadium,1748908,99,100,PIT,PIT,PIT
+2000,NL,SDN,SDP,W,5,162,81,76,86,N,N,N,N,752,5560,1413,279,37,157,602,1177,131,53,46,43,815,733,4.5200000000,5,5,46,4377,1443,191,649,1071,141,155,0.97,San Diego Padres,Qualcomm Stadium,2352443,92,92,SDP,SDN,SDN
+2000,NL,SFN,SFG,W,1,162,81,97,65,Y,N,N,N,925,5519,1535,304,44,226,709,1032,79,39,51,66,747,675,4.2100000000,9,15,47,4332,1452,151,623,1076,93,173,0.98,San Francisco Giants,PacBell Park,3318800,93,92,SFG,SFN,SFN
+2000,NL,SLN,STL,C,1,162,81,95,67,Y,N,N,N,887,5478,1481,259,25,235,675,1253,87,51,84,53,771,698,4.3800000000,10,7,37,4300,1403,196,606,1100,111,148,0.98,St. Louis Cardinals,Busch Stadium II,3336493,102,101,STL,SLN,SLN
+2001,AL,ANA,ANA,W,3,162,81,75,87,N,N,N,N,691,5551,1447,275,26,158,494,1001,116,52,77,53,730,671,4.2000000000,6,1,43,4313,1452,168,525,947,103,142,0.98,Anaheim Angels,Edison International Field,2000919,101,101,ANA,ANA,ANA
+2001,AL,BAL,BAL,E,4,162,80,63,98,N,N,N,N,687,5472,1359,262,24,136,514,989,133,53,77,49,829,744,4.6700000000,10,6,31,4297,1504,194,528,938,125,137,0.97,Baltimore Orioles,Oriole Park at Camden Yards,3094841,95,96,BAL,BAL,BAL
+2001,AL,BOS,BOS,E,2,161,81,82,79,N,N,N,N,772,5605,1493,316,29,198,520,1131,46,35,70,41,745,667,4.1500000000,3,9,48,4344,1412,146,544,1259,113,129,0.98,Boston Red Sox,Fenway Park II,2625333,102,101,BOS,BOS,BOS
+2001,AL,CHA,CHW,C,3,162,81,83,79,N,N,N,N,798,5464,1463,300,29,214,520,998,123,59,52,51,795,725,4.5500000000,8,7,51,4300,1465,181,500,921,118,149,0.98,Chicago White Sox,Comiskey Park II,1766172,104,103,CHW,CHA,CHA
+2001,AL,CLE,CLE,C,1,162,80,91,71,Y,N,N,N,897,5600,1559,294,37,212,577,1076,79,41,69,62,821,746,4.6400000000,3,4,42,4340,1512,148,573,1218,107,137,0.98,Cleveland Indians,Jacobs Field,3175523,100,100,CLE,CLE,CLE
+2001,AL,DET,DET,C,4,162,81,66,96,N,N,N,N,724,5537,1439,291,60,139,466,972,133,61,51,49,876,795,5.0100000000,16,2,34,4288,1624,180,553,859,131,165,0.97,Detroit Tigers,Comerica Park,1921305,93,95,DET,DET,DET
+2001,AL,KCA,KCR,C,5,162,81,65,97,N,N,N,N,729,5643,1503,277,37,152,406,898,100,42,44,47,858,779,4.8700000000,5,1,30,4320,1537,209,576,911,117,204,0.98,Kansas City Royals,Kauffman Stadium,1536371,107,108,KCR,KCA,KCA
+2001,AL,MIN,MIN,C,2,162,81,85,77,N,N,N,N,771,5560,1514,328,38,164,495,1083,146,67,64,38,766,722,4.5100000000,12,8,45,4324,1494,192,445,965,108,121,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1782929,102,102,MIN,MIN,MIN
+2001,AL,NYA,NYY,E,1,161,80,95,65,Y,N,Y,N,804,5577,1488,289,20,203,519,1035,161,53,64,43,713,649,4.0200000000,7,9,57,4354,1429,158,465,1266,109,132,0.98,New York Yankees,Yankee Stadium II,3264907,102,100,NYY,NYA,NYA
+2001,AL,OAK,OAK,W,2,162,81,102,60,N,Y,N,N,884,5573,1469,334,22,199,640,1021,68,29,88,59,645,583,3.5900000000,13,9,44,4390,1384,153,440,1117,125,151,0.98,Oakland Athletics,Oakland Coliseum,2133277,99,97,OAK,OAK,OAK
+2001,AL,SEA,SEA,W,1,162,81,116,46,Y,N,N,N,927,5680,1637,310,38,169,614,989,174,42,62,70,627,576,3.5400000000,8,14,56,4395,1293,160,465,1051,83,137,0.98,Seattle Mariners,Safeco Field,3507326,94,93,SEA,SEA,SEA
+2001,AL,TBA,TBD,E,5,162,81,62,100,N,N,N,N,672,5524,1426,311,21,121,456,1116,115,52,54,25,887,781,4.9400000000,1,6,30,4271,1513,207,569,1030,139,144,0.97,Tampa Bay Devil Rays,Tropicana Field,1298365,98,100,TBD,TBA,TBA
+2001,AL,TEX,TEX,W,4,162,82,73,89,N,N,N,N,890,5685,1566,326,23,246,548,1093,97,32,75,55,968,913,5.7100000000,4,3,37,4315,1670,222,596,951,114,167,0.98,Texas Rangers,The Ballpark at Arlington,2831021,104,105,TEX,TEX,TEX
+2001,AL,TOR,TOR,E,3,162,82,80,82,N,N,N,N,767,5663,1489,287,36,195,470,1094,156,55,74,43,753,696,4.2800000000,7,10,41,4388,1553,165,490,1041,97,184,0.98,Toronto Blue Jays,Skydome,1915438,102,103,TOR,TOR,TOR
+2001,NL,ARI,ARI,W,1,162,81,92,70,Y,N,Y,Y,818,5595,1494,284,35,208,587,1052,71,38,57,36,677,627,3.8700000000,12,13,34,4379,1352,195,461,1297,84,148,0.98,Arizona Diamondbacks,Bank One Ballpark,2736451,108,107,ARI,ARI,ARI
+2001,NL,ATL,ATL,E,1,162,81,88,74,Y,N,N,N,729,5498,1432,263,24,174,493,1039,85,46,45,52,643,578,3.5900000000,5,13,41,4342,1363,153,499,1133,103,133,0.98,Atlanta Braves,Turner Field,2823530,103,102,ATL,ATL,ATL
+2001,NL,CHN,CHC,C,3,162,81,88,74,N,N,N,N,777,5406,1409,268,32,194,577,1077,67,36,66,53,701,643,4.0300000000,8,6,41,4311,1357,164,550,1344,109,113,0.98,Chicago Cubs,Wrigley Field,2779465,95,95,CHC,CHN,CHN
+2001,NL,CIN,CIN,C,5,162,81,66,96,N,N,N,N,735,5583,1464,304,22,176,468,1172,103,54,65,40,850,765,4.7700000000,2,2,35,4328,1572,198,515,943,138,136,0.97,Cincinnati Reds,Cinergy Field,1879757,105,105,CIN,CIN,CIN
+2001,NL,COL,COL,W,5,162,81,73,89,N,N,N,N,923,5690,1663,324,61,213,511,1027,132,54,61,50,906,841,5.2900000000,8,8,26,4290,1522,239,598,1058,96,167,0.98,Colorado Rockies,Coors Field,3166821,122,122,COL,COL,COL
+2001,NL,FLO,FLA,E,4,162,80,76,86,N,N,N,N,742,5542,1461,325,30,166,470,1145,89,40,67,45,744,691,4.3200000000,5,11,32,4314,1397,151,617,1119,103,174,0.98,Florida Marlins,Pro Player Stadium,1261226,97,97,FLA,FLO,FLO
+2001,NL,HOU,HOU,C,1,162,81,93,69,Y,N,N,N,847,5528,1500,313,29,208,581,1119,64,49,89,56,769,707,4.3700000000,7,6,48,4364,1453,221,486,1228,110,138,0.98,Houston Astros,Enron Field,2904277,107,106,HOU,HOU,HOU
+2001,NL,LAN,LAD,W,3,162,81,86,76,N,N,N,N,758,5493,1399,264,27,206,519,1062,89,42,56,44,744,685,4.2500000000,3,5,46,4352,1387,184,524,1212,116,138,0.98,Los Angeles Dodgers,Dodger Stadium,3017143,92,91,LAD,LAN,LAN
+2001,NL,MIL,MIL,C,4,162,81,68,94,N,N,N,N,740,5488,1378,273,30,209,488,1399,66,36,72,35,806,740,4.6400000000,3,8,28,4309,1452,197,667,1057,103,156,0.98,Milwaukee Brewers,Miller Park,2811041,98,99,MIL,ML4,MIL
+2001,NL,MON,WSN,E,5,162,81,68,94,N,N,N,N,670,5379,1361,320,28,131,478,1071,101,51,60,45,812,745,4.6800000000,5,11,28,4294,1509,190,525,1103,108,139,0.98,Montreal Expos,Stade Olympique,642745,101,102,MON,MON,MON
+2001,NL,NYN,NYM,E,3,162,81,82,80,N,N,N,N,642,5459,1361,273,18,147,545,1062,66,48,65,35,713,654,4.0700000000,6,14,48,4337,1418,186,438,1191,101,132,0.98,New York Mets,Shea Stadium,2658330,96,96,NYM,NYN,NYN
+2001,NL,PHI,PHI,E,2,162,81,86,76,N,N,N,N,746,5497,1431,295,29,164,551,1125,153,47,43,61,719,667,4.1500000000,8,7,47,4336,1417,170,527,1086,91,145,0.98,Philadelphia Phillies,Veterans Stadium,1782054,98,98,PHI,PHI,PHI
+2001,NL,PIT,PIT,C,6,162,81,62,100,N,N,N,N,657,5398,1333,256,25,161,467,1106,93,73,67,35,858,794,5.0500000000,8,9,36,4249,1493,167,549,908,133,168,0.97,Pittsburgh Pirates,PNC Park,2464870,103,105,PIT,PIT,PIT
+2001,NL,SDN,SDP,W,4,162,81,79,83,N,N,N,N,789,5482,1379,273,26,161,678,1273,129,44,41,48,812,724,4.5200000000,5,6,46,4322,1519,219,476,1088,145,127,0.97,San Diego Padres,Qualcomm Stadium,2378128,91,92,SDP,SDN,SDN
+2001,NL,SFN,SFG,W,2,162,81,90,72,N,N,N,N,799,5612,1493,304,40,235,625,1090,57,42,50,54,748,680,4.1800000000,3,8,47,4390,1437,145,579,1080,118,170,0.98,San Francisco Giants,PacBell Park,3311958,93,92,SFG,SFN,SFN
+2001,NL,SLN,STL,C,2,162,82,93,69,N,Y,N,N,814,5450,1469,274,32,199,529,1089,91,35,65,50,684,627,3.9300000000,8,11,38,4306,1389,196,526,1083,110,156,0.98,St. Louis Cardinals,Busch Stadium II,3109578,100,99,STL,SLN,SLN
+2002,AL,ANA,ANA,W,2,162,81,99,63,N,Y,Y,Y,851,5678,1603,333,32,152,462,805,117,51,74,64,644,595,3.6900000000,7,14,54,4357,1345,169,509,999,87,151,0.98,Anaheim Angels,Edison International Field,2305547,100,99,ANA,ANA,ANA
+2002,AL,BAL,BAL,E,4,162,81,67,95,N,N,N,N,667,5491,1353,311,27,165,452,993,110,48,64,49,773,719,4.4600000000,8,3,31,4352,1491,208,549,967,91,173,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2682439,95,96,BAL,BAL,BAL
+2002,AL,BOS,BOS,E,2,162,81,93,69,N,N,N,N,859,5640,1560,348,33,177,545,944,80,28,72,53,665,603,3.7500000000,5,17,51,4338,1339,146,430,1157,104,140,0.98,Boston Red Sox,Fenway Park II,2650862,103,102,BOS,BOS,BOS
+2002,AL,CHA,CHW,C,2,162,81,81,81,N,N,N,N,856,5502,1475,289,29,217,555,952,75,31,49,53,798,716,4.5300000000,7,7,35,4269,1422,190,528,945,97,157,0.98,Chicago White Sox,Comiskey Park II,1676911,102,101,CHW,CHA,CHA
+2002,AL,CLE,CLE,C,3,162,81,74,88,N,N,N,N,739,5423,1349,255,26,192,542,1000,52,37,56,39,837,777,4.9100000000,9,4,34,4274,1508,142,603,1058,113,161,0.98,Cleveland Indians,Jacobs Field,2616940,97,98,CLE,CLE,CLE
+2002,AL,DET,DET,C,5,161,80,55,106,N,N,N,N,575,5406,1340,265,37,124,363,1035,65,44,64,57,864,774,4.9300000000,11,7,33,4242,1593,163,463,794,142,148,0.97,Detroit Tigers,Comerica Park,1503623,92,95,DET,DET,DET
+2002,AL,KCA,KCR,C,4,162,77,62,100,N,N,N,N,737,5535,1415,285,42,140,524,921,140,65,52,51,891,834,5.2100000000,12,6,30,4323,1587,212,572,909,130,153,0.97,Kansas City Royals,Kauffman Stadium,1323036,110,111,KCR,KCA,KCA
+2002,AL,MIN,MIN,C,1,161,81,94,67,Y,N,N,N,768,5582,1518,348,36,167,472,1089,79,62,56,52,712,662,4.1200000000,8,9,47,4334,1454,184,439,1026,74,124,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1924473,100,100,MIN,MIN,MIN
+2002,AL,NYA,NYY,E,1,161,80,103,58,Y,N,N,N,897,5601,1540,314,12,223,640,1171,100,38,72,41,697,625,3.8700000000,9,11,53,4356,1441,144,403,1135,127,117,0.97,New York Yankees,Yankee Stadium II,3465807,100,99,NYY,NYA,NYA
+2002,AL,OAK,OAK,W,1,162,81,103,59,Y,N,N,N,800,5558,1450,279,28,205,609,1008,46,20,68,36,654,593,3.6800000000,9,19,48,4356,1391,135,474,1021,102,144,0.98,Oakland Athletics,Oakland Coliseum,2169811,98,97,OAK,OAK,OAK
+2002,AL,SEA,SEA,W,3,162,81,93,69,N,N,N,N,814,5569,1531,285,31,152,629,1003,137,58,51,72,699,654,4.0700000000,8,12,43,4336,1422,178,441,1063,88,134,0.98,Seattle Mariners,Safeco Field,3542938,97,95,SEA,SEA,SEA
+2002,AL,TBA,TBD,E,5,161,81,55,106,N,N,N,N,673,5604,1418,297,35,133,456,1115,102,45,58,36,918,846,5.2900000000,12,3,25,4321,1567,215,620,925,126,168,0.97,Tampa Bay Devil Rays,Tropicana Field,1065742,97,99,TBD,TBA,TBA
+2002,AL,TEX,TEX,W,4,162,80,72,90,N,N,N,N,843,5618,1510,304,27,230,554,1055,62,34,62,50,882,824,5.1500000000,4,4,33,4319,1528,194,669,1030,99,152,0.98,Texas Rangers,The Ballpark at Arlington,2352397,106,107,TEX,TEX,TEX
+2002,AL,TOR,TOR,E,3,162,81,78,84,N,N,N,N,813,5581,1457,305,38,187,522,1142,71,18,53,57,828,767,4.8000000000,6,6,41,4315,1504,177,590,991,107,159,0.98,Toronto Blue Jays,Skydome,1637900,104,104,TOR,TOR,TOR
+2002,NL,ARI,ARI,W,1,162,81,98,64,Y,N,N,N,819,5508,1471,283,41,165,643,1016,92,46,50,53,674,630,3.9200000000,14,10,40,4340,1361,170,421,1303,89,116,0.98,Arizona Diamondbacks,Bank One Ballpark,3198977,111,111,ARI,ARI,ARI
+2002,NL,ATL,ATL,E,1,161,81,101,59,Y,N,N,N,708,5495,1428,280,25,164,558,1028,76,39,54,49,565,511,3.1300000000,3,15,57,4402,1302,123,554,1058,114,170,0.98,Atlanta Braves,Turner Field,2603484,102,101,ATL,ATL,ATL
+2002,NL,CHN,CHC,C,5,162,78,67,95,N,N,N,N,706,5496,1351,259,29,200,585,1269,63,21,44,39,759,687,4.2900000000,11,9,23,4324,1373,167,606,1333,114,144,0.98,Chicago Cubs,Wrigley Field,2693096,98,98,CHC,CHN,CHN
+2002,NL,CIN,CIN,C,3,162,80,78,84,N,N,N,N,709,5470,1386,297,21,169,583,1188,116,52,66,40,774,690,4.2700000000,2,8,42,4361,1502,173,550,980,120,169,0.98,Cincinnati Reds,Cinergy Field,1855787,104,105,CIN,CIN,CIN
+2002,NL,COL,COL,W,4,162,81,73,89,N,N,N,N,778,5512,1508,283,41,152,497,1043,103,53,56,50,898,825,5.2000000000,1,8,43,4280,1554,225,582,920,112,158,0.98,Colorado Rockies,Coors Field,2737838,116,116,COL,COL,COL
+2002,NL,FLO,FLA,E,4,162,81,79,83,N,N,N,N,699,5496,1433,280,32,146,595,1130,177,73,61,49,763,706,4.3600000000,11,12,36,4369,1449,151,631,1104,106,163,0.98,Florida Marlins,Pro Player Stadium,813118,97,98,FLA,FLO,FLO
+2002,NL,HOU,HOU,C,2,162,81,84,78,N,N,N,N,749,5503,1441,291,32,167,589,1120,71,27,59,37,695,643,4.0000000000,2,11,43,4335,1423,151,546,1219,83,149,0.98,Houston Astros,Minute Maid Park,2517357,106,105,HOU,HOU,HOU
+2002,NL,LAN,LAD,W,3,162,81,92,70,N,N,N,N,713,5554,1464,286,29,155,428,940,96,37,53,44,643,598,3.6900000000,4,15,56,4373,1311,165,555,1132,90,134,0.98,Los Angeles Dodgers,Dodger Stadium,3131255,92,92,LAD,LAN,LAN
+2002,NL,MIL,MIL,C,6,162,81,56,106,N,N,N,N,627,5415,1369,269,29,139,500,1125,94,50,55,34,821,752,4.7300000000,7,4,32,4297,1468,199,666,1026,103,154,0.98,Milwaukee Brewers,Miller Park,1969153,98,99,MIL,ML4,MIL
+2002,NL,MON,WSN,E,2,162,81,83,79,N,N,N,N,735,5479,1432,300,36,162,575,1104,118,64,46,42,718,641,3.9700000000,9,3,39,4359,1475,165,508,1088,139,160,0.97,Montreal Expos,Stade Olympique,812045,103,103,MON,MON,MON
+2002,NL,NYN,NYM,E,5,161,78,75,86,N,N,N,N,690,5496,1409,238,22,160,486,1044,87,42,63,30,703,624,3.8900000000,9,10,36,4328,1408,163,543,1107,144,138,0.97,New York Mets,Shea Stadium,2804838,97,97,NYM,NYN,NYN
+2002,NL,PHI,PHI,E,3,161,79,80,81,N,N,N,N,710,5523,1428,325,41,165,640,1095,104,43,53,39,724,671,4.1700000000,5,9,47,4349,1381,153,570,1075,88,156,0.98,Philadelphia Phillies,Veterans Stadium,1618467,94,94,PHI,PHI,PHI
+2002,NL,PIT,PIT,C,4,161,79,72,89,N,N,N,N,641,5330,1300,263,20,142,537,1109,86,49,73,41,730,664,4.2300000000,2,7,47,4238,1447,163,572,920,115,177,0.98,Pittsburgh Pirates,PNC Park,1784988,102,103,PIT,PIT,PIT
+2002,NL,SDN,SDP,W,5,162,81,66,96,N,N,N,N,662,5515,1393,243,29,136,547,1062,71,44,30,41,815,737,4.6200000000,5,10,40,4309,1522,177,582,1108,128,162,0.97,San Diego Padres,Qualcomm Stadium,2220601,90,91,SDP,SDN,SDN
+2002,NL,SFN,SFG,W,2,162,81,95,66,N,Y,Y,N,783,5497,1465,300,35,198,616,961,74,21,65,52,616,566,3.5400000000,10,13,43,4312,1349,116,523,992,90,166,0.98,San Francisco Giants,PacBell Park,3253203,95,94,SFG,SFN,SFN
+2002,NL,SLN,STL,C,1,162,81,97,65,Y,N,N,N,787,5505,1475,285,26,175,542,927,86,42,67,49,648,595,3.7000000000,4,9,42,4339,1355,141,547,1009,103,168,0.98,St. Louis Cardinals,Busch Stadium II,3011756,98,97,STL,SLN,SLN
+2003,AL,ANA,ANA,W,3,162,82,77,85,N,N,N,N,736,5487,1473,276,33,150,476,838,129,61,56,50,743,680,4.2800000000,5,9,39,4294,1444,190,486,980,105,138,0.98,Anaheim Angels,Edison International Field,3061094,98,97,ANA,ANA,ANA
+2003,AL,BAL,BAL,E,4,163,81,71,91,N,N,N,N,743,5665,1516,277,24,152,431,902,89,36,54,40,820,767,4.7600000000,9,3,41,4349,1579,198,526,981,105,164,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2454523,99,99,BAL,BAL,BAL
+2003,AL,BOS,BOS,E,2,162,81,95,67,N,Y,N,N,961,5769,1667,371,40,238,620,943,88,35,53,64,809,729,4.4800000000,5,6,36,4394,1503,153,488,1141,113,130,0.98,Boston Red Sox,Fenway Park II,2724165,105,103,BOS,BOS,BOS
+2003,AL,CHA,CHW,C,2,162,81,86,76,N,N,N,N,791,5487,1445,303,19,220,519,916,77,29,58,41,715,663,4.1700000000,12,4,36,4293,1364,162,518,1056,93,154,0.98,Chicago White Sox,U.S. Cellular Field,1939524,102,102,CHW,CHA,CHA
+2003,AL,CLE,CLE,C,4,162,81,68,94,N,N,N,N,699,5572,1413,296,26,158,466,1062,86,61,62,41,778,682,4.2100000000,5,7,34,4378,1477,179,501,943,126,178,0.98,Cleveland Indians,Jacobs Field,1730002,96,97,CLE,CLE,CLE
+2003,AL,DET,DET,C,5,162,81,43,119,N,N,N,N,591,5466,1312,201,39,153,443,1099,98,63,47,49,928,847,5.3000000000,3,5,27,4316,1616,195,557,764,138,194,0.97,Detroit Tigers,Comerica Park,1368245,93,95,DET,DET,DET
+2003,AL,KCA,KCR,C,3,162,80,83,79,N,N,N,N,836,5568,1526,288,39,162,476,926,120,42,75,57,867,809,5.0600000000,7,10,36,4316,1569,190,566,865,108,143,0.98,Kansas City Royals,Kauffman Stadium,1779895,107,108,KCR,KCA,KCA
+2003,AL,MIN,MIN,C,1,162,81,90,72,Y,N,N,N,801,5655,1567,318,45,155,512,1027,94,44,63,52,758,716,4.4100000000,7,8,45,4385,1526,187,402,997,87,114,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1946011,101,100,MIN,MIN,MIN
+2003,AL,NYA,NYY,E,1,163,82,101,61,Y,N,Y,N,877,5605,1518,304,14,230,684,1042,98,33,81,35,716,653,4.0200000000,8,12,49,4386,1512,145,375,1119,114,126,0.98,New York Yankees,Yankee Stadium II,3465600,98,97,NYY,NYA,NYA
+2003,AL,OAK,OAK,W,1,162,81,96,66,Y,N,N,N,768,5497,1398,317,24,176,556,898,48,14,59,53,643,582,3.6300000000,16,14,48,4325,1336,140,499,1018,107,145,0.98,Oakland Athletics,Oakland Coliseum,2216596,99,98,OAK,OAK,OAK
+2003,AL,SEA,SEA,W,2,162,81,93,69,N,N,N,N,795,5561,1509,290,33,139,586,989,108,37,53,46,637,602,3.7600000000,8,15,38,4323,1340,173,466,1001,65,159,0.98,Seattle Mariners,Safeco Field,3268509,95,95,SEA,SEA,SEA
+2003,AL,TBA,TBD,E,5,162,81,63,99,N,N,N,N,715,5654,1501,298,38,137,420,1030,142,42,56,50,852,787,4.9300000000,7,7,30,4310,1454,196,639,877,103,158,0.98,Tampa Bay Devil Rays,Tropicana Field,1058695,95,97,TBD,TBA,TBA
+2003,AL,TEX,TEX,W,4,162,81,71,91,N,N,N,N,826,5664,1506,274,36,239,488,1052,65,25,75,42,969,903,5.6700000000,4,3,43,4300,1625,208,603,1009,94,168,0.98,Texas Rangers,The Ballpark at Arlington,2094394,111,111,TEX,TEX,TEX
+2003,AL,TOR,TOR,E,3,162,81,86,76,N,N,N,N,894,5661,1580,357,33,190,546,1081,37,25,90,56,826,748,4.6900000000,14,6,36,4305,1560,184,485,984,117,161,0.98,Toronto Blue Jays,Skydome,1799458,104,104,TOR,TOR,TOR
+2003,NL,ARI,ARI,W,3,162,81,84,78,N,N,N,N,717,5570,1467,303,47,152,531,1006,76,38,45,52,685,621,3.8400000000,7,11,42,4365,1379,150,526,1291,107,132,0.98,Arizona Diamondbacks,Bank One Ballpark,2805542,108,109,ARI,ARI,ARI
+2003,NL,ATL,ATL,E,1,162,81,101,61,Y,N,N,N,907,5670,1608,321,31,235,545,933,68,22,49,49,740,663,4.1000000000,4,7,51,4369,1425,147,555,992,121,166,0.98,Atlanta Braves,Turner Field,2401084,101,100,ATL,ATL,ATL
+2003,NL,CHN,CHC,C,1,162,81,88,74,Y,N,N,N,724,5519,1431,302,24,172,492,1158,73,31,50,46,683,619,3.8300000000,13,14,36,4369,1304,143,617,1404,106,157,0.98,Chicago Cubs,Wrigley Field,2962630,101,101,CHC,CHN,CHN
+2003,NL,CIN,CIN,C,5,162,81,69,93,N,N,N,N,694,5509,1349,239,21,182,524,1326,80,34,79,32,886,818,5.0900000000,4,5,38,4339,1578,209,590,932,141,152,0.97,Cincinnati Reds,Great American Ball Park,2355259,95,96,CIN,CIN,CIN
+2003,NL,COL,COL,W,4,162,81,74,88,N,N,N,N,853,5518,1472,330,31,198,619,1134,63,37,52,38,892,821,5.2000000000,3,4,34,4260,1629,200,552,866,116,165,0.98,Colorado Rockies,Coors Field,2334085,115,116,COL,COL,COL
+2003,NL,FLO,FLA,E,2,162,81,91,71,N,Y,Y,Y,751,5490,1459,292,44,157,515,978,150,74,57,41,692,648,4.0400000000,7,11,36,4336,1415,128,530,1132,78,162,0.98,Florida Marlins,Pro Player Stadium,1303215,98,98,FLA,FLO,FLO
+2003,NL,HOU,HOU,C,2,162,81,87,75,N,N,N,N,805,5583,1466,308,30,191,557,1021,66,30,81,38,677,622,3.8600000000,1,5,50,4350,1350,161,565,1139,95,149,0.98,Houston Astros,Minute Maid Park,2454241,103,102,HOU,HOU,HOU
+2003,NL,LAN,LAD,W,2,162,81,85,77,N,N,N,N,574,5458,1328,260,25,124,407,985,80,36,72,28,556,511,3.1600000000,3,17,58,4373,1254,127,526,1289,119,164,0.98,Los Angeles Dodgers,Dodger Stadium,3138626,94,94,LAD,LAN,LAN
+2003,NL,MIL,MIL,C,6,162,81,68,94,N,N,N,N,714,5548,1423,266,24,196,547,1221,99,39,71,40,873,810,5.0200000000,5,3,44,4356,1590,219,575,1034,114,142,0.98,Milwaukee Brewers,Miller Park,1700354,98,100,MIL,ML4,MIL
+2003,NL,MON,WSN,E,4,162,81,83,79,N,N,N,N,711,5437,1404,294,25,144,522,990,100,39,45,40,716,640,4.0100000000,15,10,42,4313,1467,181,463,1028,102,152,0.98,Montreal Expos,Stade Olympique/Hiram Bithorn Stadium,1025639,105,105,MON,MON,MON
+2003,NL,NYN,NYM,E,5,161,80,66,95,N,N,N,N,642,5341,1317,262,24,124,489,1035,70,31,54,45,754,704,4.4800000000,3,10,38,4240,1497,168,576,907,118,158,0.98,New York Mets,Shea Stadium,2140599,97,98,NYM,NYN,NYN
+2003,NL,PHI,PHI,E,3,162,81,86,76,N,N,N,N,791,5543,1448,325,27,166,651,1155,72,29,55,38,697,648,4.0400000000,9,13,33,4331,1386,142,536,1060,97,146,0.98,Philadelphia Phillies,Veterans Stadium,2259948,93,93,PHI,PHI,PHI
+2003,NL,PIT,PIT,C,4,162,81,75,87,N,N,N,N,753,5581,1492,275,45,163,529,1049,86,37,87,38,801,744,4.6400000000,7,10,44,4333,1527,178,502,926,123,159,0.98,Pittsburgh Pirates,PNC Park,1636751,102,102,PIT,PIT,PIT
+2003,NL,SDN,SDP,W,5,162,81,64,98,N,N,N,N,678,5531,1442,257,32,128,565,1073,76,39,57,42,831,774,4.8700000000,2,10,31,4294,1458,208,611,1091,102,141,0.98,San Diego Padres,Qualcomm Stadium,2030084,91,92,SDP,SDN,SDN
+2003,NL,SFN,SFG,W,1,161,81,100,61,Y,N,N,N,755,5456,1440,281,29,180,593,980,53,37,40,39,638,595,3.7300000000,7,10,43,4312,1349,136,546,1006,80,163,0.98,San Francisco Giants,PacBell Park,3264898,99,98,SFG,SFN,SFN
+2003,NL,SLN,STL,C,3,162,81,85,77,N,N,N,N,876,5672,1580,342,32,196,580,952,82,32,73,54,796,748,4.6000000000,9,10,41,4391,1544,210,508,969,77,138,0.98,St. Louis Cardinals,Busch Stadium II,2910386,97,96,STL,SLN,SLN
+2004,AL,ANA,ANA,W,1,162,81,92,70,Y,N,N,N,836,5675,1603,272,37,162,450,942,143,46,73,41,734,692,4.2800000000,2,11,50,4363,1476,170,502,1164,90,126,0.98,Anaheim Angels,Angels Stadium of Anaheim,3375677,97,97,ANA,ANA,ANA
+2004,AL,BAL,BAL,E,3,162,81,78,84,N,N,N,N,842,5736,1614,319,18,169,528,949,101,41,57,62,830,760,4.7000000000,8,10,27,4366,1488,159,687,1090,110,161,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2744018,98,99,BAL,BAL,BAL
+2004,AL,BOS,BOS,E,2,162,81,98,64,N,Y,Y,Y,949,5720,1613,373,25,222,659,1189,68,30,69,55,768,674,4.1800000000,4,12,36,4354,1430,159,447,1132,118,129,0.98,Boston Red Sox,Fenway Park II,2837294,106,105,BOS,BOS,BOS
+2004,AL,CHA,CHW,C,2,162,81,83,79,N,N,N,N,865,5534,1481,284,19,242,499,1030,78,51,63,42,831,782,4.9100000000,8,8,34,4297,1505,224,527,1013,100,167,0.98,Chicago White Sox,U.S. Cellular Field,1930537,102,102,CHW,CHA,CHA
+2004,AL,CLE,CLE,C,3,162,81,80,82,N,N,N,N,858,5676,1565,345,29,184,606,1009,94,55,78,42,857,784,4.8100000000,8,8,32,4400,1553,201,579,1115,106,152,0.98,Cleveland Indians,Jacobs Field,1814401,94,94,CLE,CLE,CLE
+2004,AL,DET,DET,C,4,162,81,72,90,N,N,N,N,827,5623,1531,284,54,201,518,1144,86,50,50,43,844,788,4.9300000000,7,9,35,4319,1542,190,530,995,144,160,0.97,Detroit Tigers,Comerica Park,1917004,95,96,DET,DET,DET
+2004,AL,KCA,KCR,C,5,162,80,58,104,N,N,N,N,720,5538,1432,261,29,150,461,1057,67,48,76,38,905,813,5.1500000000,6,3,25,4261,1638,208,518,887,131,169,0.97,Kansas City Royals,Kauffman Stadium,1661478,102,103,KCR,KCA,KCA
+2004,AL,MIN,MIN,C,1,162,81,92,70,Y,N,N,N,780,5623,1494,310,24,191,513,982,116,46,64,40,715,661,4.0300000000,4,9,48,4428,1523,167,431,1123,101,158,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,1911490,103,102,MIN,MIN,MIN
+2004,AL,NYA,NYY,E,1,162,81,101,61,Y,N,N,N,897,5527,1483,281,20,242,670,982,84,33,80,50,808,752,4.6900000000,1,5,59,4331,1532,182,445,1058,99,148,0.98,New York Yankees,Yankee Stadium II,3775292,98,97,NYY,NYA,NYA
+2004,AL,OAK,OAK,W,2,162,81,91,71,N,N,N,N,793,5728,1545,336,15,189,608,1061,47,22,55,43,742,682,4.1700000000,10,8,35,4414,1466,164,544,1034,91,172,0.98,Oakland Athletics,Network Associates Coliseum,2201516,99,98,OAK,OAK,OAK
+2004,AL,SEA,SEA,W,4,162,82,63,99,N,N,N,N,698,5722,1544,276,20,136,492,1058,110,42,54,48,823,772,4.7600000000,7,7,28,4378,1498,212,575,1036,103,140,0.98,Seattle Mariners,Safeco Field,2940731,97,98,SEA,SEA,SEA
+2004,AL,TBA,TBD,E,4,161,80,70,91,N,N,N,N,714,5483,1416,278,46,145,469,944,132,42,55,56,842,757,4.8100000000,3,5,35,4251,1459,192,580,923,119,139,0.98,Tampa Bay Devil Rays,Tropicana Field,1274911,96,98,TBD,TBA,TBA
+2004,AL,TEX,TEX,W,3,162,81,89,73,N,N,N,N,860,5615,1492,323,34,227,500,1099,69,36,61,57,794,724,4.5300000000,5,9,52,4319,1536,182,547,979,117,152,0.98,Texas Rangers,The Ballpark at Arlington,2513685,108,109,TEX,TEX,TEX
+2004,AL,TOR,TOR,E,5,161,81,67,94,N,N,N,N,719,5531,1438,290,34,145,513,1083,58,31,71,42,823,775,4.9100000000,6,11,37,4263,1505,181,608,956,91,150,0.98,Toronto Blue Jays,Skydome,1900041,104,104,TOR,TOR,TOR
+2004,NL,ARI,ARI,W,5,162,81,51,111,N,N,N,N,615,5544,1401,295,38,135,441,1022,53,32,35,37,899,794,4.9800000000,5,6,33,4308,1480,197,668,1153,139,144,0.97,Arizona Diamondbacks,Bank One Ballpark,2519560,105,107,ARI,ARI,ARI
+2004,NL,ATL,ATL,E,1,162,81,96,66,Y,N,N,N,803,5570,1503,304,37,178,587,1158,86,32,59,48,668,603,3.7400000000,4,13,48,4350,1475,154,523,1025,116,171,0.98,Atlanta Braves,Turner Field,2327565,101,100,ATL,ATL,ATL
+2004,NL,CHN,CHC,C,3,162,82,89,73,N,N,N,N,789,5628,1508,308,29,235,489,1080,66,28,38,48,665,621,3.8100000000,3,6,42,4396,1363,169,545,1346,86,126,0.98,Chicago Cubs,Wrigley Field,3170154,103,102,CHC,CHN,CHN
+2004,NL,CIN,CIN,C,4,162,81,76,86,N,N,N,N,750,5518,1380,287,28,194,599,1335,77,25,81,25,907,832,5.1900000000,5,8,47,4331,1595,236,572,992,113,123,0.98,Cincinnati Reds,Great American Ball Park,2287250,98,99,CIN,CIN,CIN
+2004,NL,COL,COL,W,4,162,81,68,94,N,N,N,N,833,5577,1531,331,34,202,568,1181,44,33,54,37,923,883,5.5400000000,3,2,36,4306,1634,198,697,947,89,161,0.98,Colorado Rockies,Coors Field,2338069,113,114,COL,COL,COL
+2004,NL,FLO,FLA,E,3,162,80,83,79,N,N,N,N,718,5486,1447,275,32,148,499,968,96,43,58,40,700,655,4.1000000000,6,14,53,4317,1395,166,513,1116,86,153,0.98,Florida Marlins,Pro Player Stadium,1723105,95,95,FLA,FLO,FLO
+2004,NL,HOU,HOU,C,2,162,81,92,70,N,Y,N,N,803,5468,1458,294,36,187,590,999,89,30,61,52,698,650,4.0500000000,2,13,47,4329,1416,174,525,1282,101,136,0.98,Houston Astros,Minute Maid Park,3087872,102,101,HOU,HOU,HOU
+2004,NL,LAN,LAD,W,1,162,81,93,69,Y,N,N,N,761,5542,1450,226,30,203,536,1092,102,41,62,35,684,647,4.0100000000,2,6,51,4360,1386,178,521,1066,73,145,0.98,Los Angeles Dodgers,Dodger Stadium,3488283,95,95,LAD,LAN,LAN
+2004,NL,MIL,MIL,C,6,161,81,67,94,N,N,N,N,634,5483,1358,295,32,135,540,1312,138,40,68,40,757,679,4.2400000000,6,10,42,4326,1440,164,476,1098,117,132,0.98,Milwaukee Brewers,Miller Park,2062382,100,101,MIL,ML4,MIL
+2004,NL,MON,WSN,E,5,162,80,67,95,N,N,N,N,635,5474,1361,276,27,151,496,925,109,38,35,33,769,696,4.3300000000,11,11,31,4341,1477,191,582,1032,99,171,0.98,Montreal Expos,Stade Olympique/Hiram Bithorn Stadium,749550,105,105,MON,MON,MON
+2004,NL,NYN,NYM,E,4,162,81,71,91,N,N,N,N,684,5532,1376,289,20,185,512,1159,107,23,61,34,731,658,4.0900000000,2,6,31,4347,1452,156,592,977,137,144,0.97,New York Mets,Shea Stadium,2318951,99,99,NYM,NYN,NYN
+2004,NL,PHI,PHI,E,2,162,81,86,76,N,N,N,N,840,5643,1505,303,23,215,645,1133,100,27,58,46,781,724,4.4500000000,4,5,43,4388,1488,214,502,1070,81,142,0.98,Philadelphia Phillies,Citizens Bank Park,3250092,105,104,PHI,PHI,PHI
+2004,NL,PIT,PIT,C,5,161,80,72,89,N,N,N,N,680,5483,1428,267,39,142,415,1066,63,40,95,42,744,680,4.2900000000,3,8,46,4284,1451,149,576,1079,103,189,0.98,Pittsburgh Pirates,PNC Park,1580031,99,100,PIT,PIT,PIT
+2004,NL,SDN,SDP,W,3,162,81,87,75,N,N,N,N,768,5573,1521,304,32,139,566,910,52,25,56,66,705,645,4.0300000000,3,8,44,4323,1460,184,422,1079,108,146,0.98,San Diego Padres,Petco Park,3016752,90,90,SDP,SDN,SDN
+2004,NL,SFN,SFG,W,2,162,82,91,71,N,N,N,N,850,5546,1500,314,33,183,705,874,43,23,72,51,770,695,4.2900000000,8,8,46,4371,1481,161,548,1020,101,153,0.98,San Francisco Giants,SBC Park,3256854,101,101,SFG,SFN,SFN
+2004,NL,SLN,STL,C,1,162,81,105,57,Y,N,Y,N,855,5555,1544,319,24,214,548,1085,111,47,51,70,659,605,3.7500000000,4,12,57,4361,1378,169,440,1041,97,154,0.98,St. Louis Cardinals,Busch Stadium II,3048427,100,99,STL,SLN,SLN
+2005,AL,BAL,BAL,E,4,162,81,74,88,N,N,N,N,729,5551,1492,296,27,189,447,902,83,37,54,42,800,724,4.5600000000,2,9,38,4283,1458,180,580,1052,107,154,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2624740,99,99,BAL,BAL,BAL
+2005,AL,BOS,BOS,E,2,162,81,95,67,N,Y,N,N,910,5626,1579,339,21,199,653,1044,45,12,47,63,805,752,4.7400000000,6,8,38,4287,1550,164,440,959,109,135,0.98,Boston Red Sox,Fenway Park II,2847888,104,104,BOS,BOS,BOS
+2005,AL,CHA,CHW,C,1,162,81,99,63,Y,N,Y,Y,741,5529,1450,253,23,200,435,1002,137,67,79,49,645,592,3.6100000000,9,10,54,4427,1392,167,459,1040,94,166,0.98,Chicago White Sox,U.S. Cellular Field,2342833,103,103,CHW,CHA,CHA
+2005,AL,CLE,CLE,C,2,162,81,93,69,N,N,N,N,790,5609,1522,337,30,207,503,1093,62,36,54,50,642,582,3.6100000000,6,10,51,4358,1363,157,413,1050,106,156,0.98,Cleveland Indians,Jacobs Field,2013763,96,96,CLE,CLE,CLE
+2005,AL,DET,DET,C,4,162,81,71,91,N,N,N,N,723,5602,1521,283,45,168,384,1038,66,28,53,52,787,719,4.5100000000,7,2,37,4307,1504,193,461,907,110,171,0.98,Detroit Tigers,Comerica Park,2024431,98,98,DET,DET,DET
+2005,AL,KCA,KCR,C,5,162,81,56,106,N,N,N,N,701,5503,1445,289,34,126,424,1008,53,33,63,50,935,862,5.4900000000,4,4,25,4240,1640,178,580,924,125,163,0.97,Kansas City Royals,Kauffman Stadium,1371181,99,101,KCR,KCA,KCA
+2005,AL,LAA,ANA,W,1,162,81,95,67,Y,N,N,N,761,5624,1520,278,30,147,447,848,161,57,29,39,643,598,3.6800000000,7,11,54,4393,1419,158,443,1126,87,139,0.98,Los Angeles Angels of Anaheim,Angel Stadium,3404686,98,97,LAA,ANA,ANA
+2005,AL,MIN,MIN,C,3,162,81,83,79,N,N,N,N,688,5564,1441,269,32,134,485,978,102,44,59,42,662,604,3.7100000000,9,8,44,4393,1458,169,348,965,102,171,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2034243,102,102,MIN,MIN,MIN
+2005,AL,NYA,NYY,E,1,162,81,95,67,Y,N,N,N,886,5624,1552,259,16,229,637,989,84,27,73,43,789,718,4.5200000000,8,14,46,4292,1495,164,463,985,95,151,0.98,New York Yankees,Yankee Stadium II,4090696,98,97,NYY,NYA,NYA
+2005,AL,OAK,OAK,W,2,162,81,88,74,N,N,N,N,772,5627,1476,310,20,155,537,819,31,22,52,40,658,594,3.6900000000,9,12,38,4351,1315,154,504,1075,88,166,0.98,Oakland Athletics,McAfee Coliseum,2109118,101,100,OAK,OAK,OAK
+2005,AL,SEA,SEA,W,4,162,81,69,93,N,N,N,N,699,5507,1408,289,34,130,466,986,102,47,48,37,751,712,4.4900000000,6,7,39,4283,1483,179,496,892,86,144,0.98,Seattle Mariners,Safeco Field,2725459,95,96,SEA,SEA,SEA
+2005,AL,TBA,TBD,E,5,162,81,67,95,N,N,N,N,750,5552,1519,289,40,157,412,990,151,49,69,51,936,851,5.3900000000,1,4,43,4265,1570,194,615,949,124,139,0.97,Tampa Bay Devil Rays,Tropicana Field,1141669,99,100,TBD,TBA,TBA
+2005,AL,TEX,TEX,W,3,162,81,79,83,N,N,N,N,865,5716,1528,311,29,260,495,1112,67,15,48,32,858,794,4.9600000000,2,6,46,4320,1589,159,522,932,108,149,0.98,Texas Rangers,Ameriquest Field,2525221,106,106,TEX,TEX,TEX
+2005,AL,TOR,TOR,E,3,162,81,80,82,N,N,N,N,775,5581,1480,307,39,136,486,955,72,35,89,56,705,653,4.0600000000,9,8,35,4341,1475,185,444,958,95,154,0.98,Toronto Blue Jays,Rogers Centre,2014995,102,102,TOR,TOR,TOR
+2005,NL,ARI,ARI,W,2,162,81,77,85,N,N,N,N,696,5550,1419,291,27,191,606,1094,67,26,55,45,856,783,4.8400000000,6,10,45,4369,1580,193,537,1038,94,159,0.98,Arizona Diamondbacks,Bank One Ballpark,2059424,103,105,ARI,ARI,ARI
+2005,NL,ATL,ATL,E,1,162,81,90,72,Y,N,N,N,769,5486,1453,308,37,184,534,1084,92,32,45,46,674,639,3.9800000000,8,12,38,4331,1487,145,520,929,86,170,0.98,Atlanta Braves,Turner Field,2521167,101,100,ATL,ATL,ATL
+2005,NL,CHN,CHC,C,4,162,81,79,83,N,N,N,N,703,5584,1506,323,23,194,419,920,65,39,50,37,714,671,4.1900000000,8,10,39,4320,1357,186,576,1256,101,136,0.98,Chicago Cubs,Wrigley Field,3099992,104,104,CHC,CHN,CHN
+2005,NL,CIN,CIN,C,5,163,82,73,89,N,N,N,N,820,5565,1453,335,15,222,611,1303,72,23,62,39,889,820,5.1500000000,2,1,31,4299,1657,219,492,955,104,133,0.98,Cincinnati Reds,Great American Ball Park,1943067,100,101,CIN,CIN,CIN
+2005,NL,COL,COL,W,5,162,81,67,95,N,N,N,N,740,5542,1477,280,34,150,509,1103,65,32,64,34,862,808,5.1300000000,4,4,37,4256,1600,175,604,981,118,158,0.98,Colorado Rockies,Coors Field,1914389,112,113,COL,COL,COL
+2005,NL,FLO,FLA,E,3,162,81,83,79,N,N,N,N,717,5502,1499,306,32,128,512,918,96,38,67,50,732,666,4.1600000000,14,15,42,4327,1459,116,563,1125,103,177,0.98,Florida Marlins,Dolphin Stadium,1852608,94,94,FLA,FLO,FLO
+2005,NL,HOU,HOU,C,2,163,81,89,73,N,Y,Y,N,693,5462,1400,281,32,161,481,1037,115,44,72,42,609,563,3.5100000000,6,11,45,4329,1336,155,440,1164,89,146,0.98,Houston Astros,Minute Maid Park,2804760,101,100,HOU,HOU,HOU
+2005,NL,LAN,LAD,W,4,162,81,71,91,N,N,N,N,685,5433,1374,284,21,149,541,1094,58,35,67,33,755,695,4.3800000000,6,9,40,4282,1434,182,471,1004,106,141,0.98,Los Angeles Dodgers,Dodger Stadium,3603646,98,97,LAD,LAN,LAN
+2005,NL,MIL,MIL,C,3,162,81,81,81,N,N,N,N,726,5448,1413,327,19,175,531,1162,79,34,73,38,697,635,3.9700000000,7,6,46,4314,1382,169,569,1173,119,139,0.98,Milwaukee Brewers,Miller Park,2211023,101,101,MIL,ML4,MIL
+2005,NL,NYN,NYM,E,3,162,81,83,79,N,N,N,N,722,5505,1421,279,32,175,486,1075,153,40,48,38,648,599,3.7600000000,8,11,38,4307,1390,135,491,1012,106,146,0.98,New York Mets,Shea Stadium,2829929,98,97,NYM,NYN,NYN
+2005,NL,PHI,PHI,E,2,162,81,88,74,N,N,N,N,807,5542,1494,282,35,167,639,1083,116,27,56,46,726,672,4.2100000000,4,6,40,4305,1379,189,487,1159,90,132,0.98,Philadelphia Phillies,Citizens Bank Park,2665304,105,104,PHI,PHI,PHI
+2005,NL,PIT,PIT,C,6,162,81,67,95,N,N,N,N,680,5573,1445,292,38,139,471,1092,73,30,72,49,769,706,4.4200000000,4,14,35,4308,1456,162,612,958,117,193,0.98,Pittsburgh Pirates,PNC Park,1817245,99,100,PIT,PIT,PIT
+2005,NL,SDN,SDP,W,1,162,81,82,80,Y,N,N,N,684,5502,1416,269,39,130,600,977,99,44,49,48,726,668,4.1300000000,4,8,45,4366,1452,146,503,1133,109,136,0.98,San Diego Padres,Petco Park,2869787,92,91,SDP,SDN,SDN
+2005,NL,SFN,SFG,W,3,162,81,75,87,N,N,N,N,649,5462,1427,299,26,128,431,901,71,35,49,44,745,695,4.3300000000,4,8,46,4333,1456,151,592,972,90,146,0.98,San Francisco Giants,SBC Park,3181023,101,101,SFG,SFN,SFN
+2005,NL,SLN,STL,C,1,162,81,100,62,Y,N,N,N,805,5538,1494,287,26,170,534,947,83,36,62,35,634,560,3.4900000000,15,14,48,4337,1399,153,443,974,100,196,0.98,St. Louis Cardinals,Busch Stadium II,3538988,101,100,STL,SLN,SLN
+2005,NL,WAS,WSN,E,5,162,81,81,81,N,N,N,N,639,5426,1367,311,32,117,491,1090,45,45,89,45,673,627,3.8700000000,4,9,51,4374,1456,140,539,997,92,156,0.98,Washington Nationals,R.F.K. Stadium,2731993,95,96,WSN,MON,WAS
+2006,AL,BAL,BAL,E,4,162,81,70,92,N,N,N,N,768,5610,1556,288,20,164,474,878,121,32,73,41,899,843,5.3500000000,5,9,35,4257,1579,216,613,1016,102,156,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2153139,99,100,BAL,BAL,BAL
+2006,AL,BOS,BOS,E,3,162,81,86,76,N,N,N,N,820,5619,1510,327,16,192,672,1056,51,23,66,56,825,773,4.8300000000,3,6,46,4324,1570,181,509,1070,66,174,0.98,Boston Red Sox,Fenway Park II,2930588,105,104,BOS,BOS,BOS
+2006,AL,CHA,CHW,C,3,162,81,90,72,N,N,N,N,868,5657,1586,291,20,236,502,1056,93,48,58,57,794,743,4.6100000000,5,11,46,4347,1534,200,433,1012,90,145,0.98,Chicago White Sox,U.S. Cellular Field,2957414,104,104,CHW,CHA,CHA
+2006,AL,CLE,CLE,C,4,162,81,78,84,N,N,N,N,870,5619,1576,351,27,196,556,1204,55,23,54,43,782,698,4.4100000000,13,13,24,4270,1583,166,429,948,118,165,0.98,Cleveland Indians,Jacobs Field,1997995,99,98,CLE,CLE,CLE
+2006,AL,DET,DET,C,2,162,81,95,67,N,Y,Y,N,822,5642,1548,294,40,203,430,1133,60,40,45,36,675,618,3.8400000000,3,16,46,4344,1420,160,489,1003,106,162,0.98,Detroit Tigers,Comerica Park,2595937,100,100,DET,DET,DET
+2006,AL,KCA,KCR,C,5,162,81,62,100,N,N,N,N,757,5589,1515,335,37,124,474,1040,65,34,64,48,971,896,5.6500000000,3,5,35,4279,1648,213,637,904,98,189,0.98,Kansas City Royals,Kauffman Stadium,1372638,101,103,KCR,KCA,KCA
+2006,AL,LAA,ANA,W,2,162,81,89,73,N,N,N,N,766,5609,1539,309,29,159,486,914,148,57,42,53,732,652,4.0400000000,5,12,50,4358,1410,158,471,1164,124,154,0.97,Los Angeles Angels of Anaheim,Angel Stadium,3406790,100,100,LAA,ANA,ANA
+2006,AL,MIN,MIN,C,1,162,81,96,66,Y,N,N,N,801,5602,1608,275,34,143,490,872,101,42,50,55,683,632,3.9500000000,1,6,40,4318,1490,182,356,1164,84,135,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2285018,98,98,MIN,MIN,MIN
+2006,AL,NYA,NYY,E,1,162,81,97,65,Y,N,N,N,930,5651,1608,327,21,210,649,1053,139,35,72,49,767,708,4.4100000000,5,8,43,4331,1463,170,496,1019,104,145,0.98,New York Yankees,Yankee Stadium II,4248067,101,99,NYY,NYA,NYA
+2006,AL,OAK,OAK,W,1,162,81,93,69,Y,N,N,N,771,5500,1429,266,22,175,650,976,61,20,50,56,727,679,4.2100000000,5,11,54,4355,1525,162,529,1003,84,173,0.98,Oakland Athletics,McAfee Coliseum,1976625,97,97,OAK,OAK,OAK
+2006,AL,SEA,SEA,W,4,162,81,78,84,N,N,N,N,756,5670,1540,266,42,172,404,974,106,37,63,38,792,739,4.6000000000,6,6,47,4340,1500,183,560,1067,88,150,0.98,Seattle Mariners,Safeco Field,2481165,97,97,SEA,SEA,SEA
+2006,AL,TBA,TBD,E,5,162,81,61,101,N,N,N,N,689,5474,1395,267,33,190,441,1106,134,52,47,43,856,782,4.9600000000,3,7,33,4261,1600,180,606,979,116,156,0.98,Tampa Bay Devil Rays,Tropicana Field,1368950,99,101,TBD,TBA,TBA
+2006,AL,TEX,TEX,W,3,162,81,80,82,N,N,N,N,835,5659,1571,357,23,183,505,1061,53,24,40,50,784,731,4.6000000000,3,8,42,4294,1558,162,496,972,98,174,0.98,Texas Rangers,Ameriquest Field,2388757,101,101,TEX,TEX,TEX
+2006,AL,TOR,TOR,E,2,162,81,87,75,N,N,N,N,809,5596,1591,348,27,199,514,906,65,33,63,52,754,694,4.3700000000,6,6,42,4285,1447,185,504,1076,99,157,0.98,Toronto Blue Jays,Rogers Centre,2302212,100,100,TOR,TOR,TOR
+2006,NL,ARI,ARI,W,4,162,81,76,86,N,N,N,N,773,5645,1506,331,38,160,504,965,76,30,67,53,788,727,4.4800000000,8,9,34,4379,1503,168,536,1115,104,172,0.98,Arizona Diamondbacks,Chase Field,2091685,105,105,ARI,ARI,ARI
+2006,NL,ATL,ATL,E,3,162,81,79,83,N,N,N,N,849,5583,1510,312,26,222,526,1169,52,35,52,44,805,736,4.6000000000,6,6,38,4324,1529,183,572,1049,99,146,0.98,Atlanta Braves,Turner Field,2550524,100,99,ATL,ATL,ATL
+2006,NL,CHN,CHC,C,6,162,81,66,96,N,N,N,N,716,5587,1496,271,46,166,395,928,121,49,43,37,834,758,4.7400000000,2,7,29,4317,1396,210,687,1250,106,122,0.98,Chicago Cubs,Wrigley Field,3123215,103,103,CHC,CHN,CHN
+2006,NL,CIN,CIN,C,3,162,82,80,82,N,N,N,N,749,5515,1419,291,12,217,614,1192,124,33,59,38,801,725,4.5100000000,9,10,36,4337,1576,213,464,1053,128,139,0.97,Cincinnati Reds,Great American Ball Park,2134607,104,104,CIN,CIN,CIN
+2006,NL,COL,COL,W,4,162,81,76,86,N,N,N,N,813,5562,1504,325,54,157,561,1108,85,50,60,45,812,749,4.6600000000,5,8,34,4342,1549,155,553,952,91,190,0.98,Colorado Rockies,Coors Field,2104362,109,109,COL,COL,COL
+2006,NL,FLO,FLA,E,4,162,81,78,84,N,N,N,N,758,5502,1454,309,42,182,497,1249,110,58,74,42,772,696,4.3700000000,6,6,41,4300,1465,166,622,1088,126,166,0.97,Florida Marlins,Dolphin Stadium,1164134,95,96,FLA,FLO,FLO
+2006,NL,HOU,HOU,C,2,162,81,82,80,N,N,N,N,735,5521,1407,275,27,174,585,1076,79,36,73,46,719,666,4.0800000000,5,12,42,4406,1425,182,480,1160,80,164,0.98,Houston Astros,Minute Maid Park,3022763,100,99,HOU,HOU,HOU
+2006,NL,LAN,LAD,W,2,162,81,88,74,N,Y,N,N,820,5628,1552,307,58,153,601,959,128,49,51,48,751,686,4.2300000000,1,10,40,4381,1524,152,492,1068,115,174,0.98,Los Angeles Dodgers,Dodger Stadium,3758545,101,100,LAD,LAN,LAN
+2006,NL,MIL,MIL,C,4,162,81,75,87,N,N,N,N,730,5433,1400,301,20,180,502,1233,71,37,82,53,833,763,4.8200000000,7,8,43,4277,1454,177,514,1145,117,126,0.98,Milwaukee Brewers,Miller Park,2335643,101,101,MIL,ML4,MIL
+2006,NL,NYN,NYM,E,1,162,81,97,65,Y,N,N,N,834,5558,1469,323,41,200,547,1071,146,35,62,47,731,673,4.1400000000,5,12,43,4384,1402,180,527,1161,104,131,0.98,New York Mets,Shea Stadium,3379535,98,97,NYM,NYN,NYN
+2006,NL,PHI,PHI,E,2,162,81,85,77,N,N,N,N,865,5687,1518,294,41,216,626,1203,92,25,95,44,812,747,4.6000000000,4,6,42,4381,1561,211,512,1138,104,153,0.98,Philadelphia Phillies,Citizens Bank Park,2701815,105,104,PHI,PHI,PHI
+2006,NL,PIT,PIT,C,5,162,81,67,95,N,N,N,N,691,5558,1462,286,17,141,459,1200,68,23,89,49,797,720,4.5200000000,2,10,39,4305,1545,156,620,1060,104,168,0.98,Pittsburgh Pirates,PNC Park,1861549,98,99,PIT,PIT,PIT
+2006,NL,SDN,SDP,W,1,162,81,88,74,Y,N,N,N,731,5576,1465,298,38,161,564,1104,123,31,40,47,679,629,3.8700000000,4,11,50,4391,1385,176,468,1097,92,138,0.98,San Diego Padres,Petco Park,2659757,91,90,SDP,SDN,SDN
+2006,NL,SFN,SFG,W,3,161,81,76,85,N,N,N,N,746,5472,1418,297,52,163,494,891,58,25,53,37,790,735,4.6300000000,7,9,37,4289,1422,153,584,992,91,132,0.98,San Francisco Giants,AT&T Park,3130313,100,100,SFG,SFN,SFN
+2006,NL,SLN,STL,C,1,161,80,83,78,Y,N,Y,Y,781,5522,1484,292,27,184,531,922,59,32,61,40,762,721,4.5400000000,6,9,38,4289,1475,193,504,970,98,170,0.98,St. Louis Cardinals,Busch Stadium III,3407104,99,99,STL,SLN,SLN
+2006,NL,WAS,WSN,E,5,162,81,71,91,N,N,N,N,746,5495,1437,322,22,164,594,1156,123,62,69,49,872,803,5.0300000000,1,3,32,4309,1535,193,584,960,131,123,0.97,Washington Nationals,R.F.K. Stadium,2153056,94,95,WSN,MON,WAS
+2007,AL,BAL,BAL,E,4,162,,69,93,N,N,N,N,756,5631,1529,306,30,142,500,939,144,42,47,47,868,827,5.1700000000,4,9,30,4316,1491,161,696,1087,79,155,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2164822,101,102,BAL,BAL,BAL
+2007,AL,BOS,BOS,E,1,162,,96,66,Y,N,Y,Y,867,5589,1561,352,35,166,689,1042,96,24,64,54,657,618,3.8700000000,5,13,45,4316,1350,151,482,1149,81,145,0.98,Boston Red Sox,Fenway Park II,2970755,106,105,BOS,BOS,BOS
+2007,AL,CHA,CHW,C,4,162,,72,90,N,N,N,N,693,5441,1341,249,20,190,532,1149,78,45,52,35,839,763,4.7700000000,9,9,42,4322,1556,174,499,1015,108,166,0.98,Chicago White Sox,U.S. Cellular Field,2684395,104,104,CHW,CHA,CHA
+2007,AL,CLE,CLE,C,1,162,,96,66,Y,N,N,N,811,5604,1504,305,27,178,590,1202,72,41,80,59,704,658,4.0500000000,9,9,49,4388,1519,146,410,1047,92,166,0.98,Cleveland Indians,Jacobs Field,2275912,101,100,CLE,CLE,CLE
+2007,AL,DET,DET,C,2,162,,88,74,N,N,N,N,887,5757,1652,352,50,177,474,1054,103,30,56,45,797,735,4.5700000000,1,9,44,4342,1498,174,566,1047,99,147,0.98,Detroit Tigers,Comerica Park,3047133,102,101,DET,DET,DET
+2007,AL,KCA,KCR,C,5,162,,69,93,N,N,N,N,706,5534,1447,300,46,102,428,1069,78,44,89,47,778,716,4.4800000000,2,6,36,4312,1547,168,520,993,106,159,0.98,Kansas City Royals,Kauffman Stadium,1616867,100,101,KCR,KCA,KCA
+2007,AL,LAA,ANA,W,1,162,,94,68,Y,N,N,N,822,5554,1578,324,23,123,507,883,139,55,40,65,731,674,4.2300000000,5,9,43,4305,1480,151,477,1156,101,154,0.98,Los Angeles Angels of Anaheim,Angel Stadium,3365632,101,100,LAA,ANA,ANA
+2007,AL,MIN,MIN,C,3,162,,79,83,N,N,N,N,718,5522,1460,273,36,118,512,839,112,30,48,45,725,663,4.1500000000,5,8,38,4310,1505,185,420,1094,95,149,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2296383,95,95,MIN,MIN,MIN
+2007,AL,NYA,NYY,E,2,162,,94,68,N,Y,N,N,968,5717,1656,326,32,201,637,991,123,40,78,54,777,724,4.4900000000,1,5,34,4352,1498,150,578,1009,88,173,0.98,New York Yankees,Yankee Stadium II,4271083,101,100,NYY,NYA,NYA
+2007,AL,OAK,OAK,W,3,162,,76,86,N,N,N,N,741,5577,1430,295,16,171,664,1119,52,20,49,56,758,689,4.2800000000,4,9,36,4344,1468,138,530,1036,90,151,0.98,Oakland Athletics,McAfee Coliseum,1921844,94,94,OAK,OAK,OAK
+2007,AL,SEA,SEA,W,2,162,,88,74,N,N,N,N,794,5684,1629,284,22,153,389,861,81,30,62,40,813,754,4.7300000000,6,12,43,4303,1578,147,546,1020,90,167,0.98,Seattle Mariners,Safeco Field,2672223,96,97,SEA,SEA,SEA
+2007,AL,TBA,TBD,E,5,162,,66,96,N,N,N,N,782,5593,1500,291,36,187,545,1324,131,48,53,52,944,879,5.5300000000,2,2,28,4289,1649,199,568,1194,117,155,0.98,Tampa Bay Devil Rays,Tropicana Field,1387603,98,100,TBD,TBA,TBA
+2007,AL,TEX,TEX,W,4,162,,75,87,N,N,N,N,816,5555,1460,298,36,179,503,1224,88,25,56,42,844,755,4.7500000000,0,6,42,4290,1525,155,668,976,124,178,0.98,Texas Rangers,Rangers Ballpark in Arlington,2353862,101,101,TEX,TEX,TEX
+2007,AL,TOR,TOR,E,3,162,,83,79,N,N,N,N,753,5536,1434,344,24,165,533,1044,57,22,47,48,699,644,4.0000000000,11,9,44,4346,1383,157,479,1067,102,160,0.98,Toronto Blue Jays,Rogers Centre,2360644,100,99,TOR,TOR,TOR
+2007,NL,ARI,ARI,W,1,162,,90,72,Y,N,N,N,712,5398,1350,286,40,171,532,1111,109,24,57,58,732,662,4.1300000000,7,12,51,4323,1446,169,546,1088,106,157,0.98,Arizona Diamondbacks,Chase Field,2325249,107,107,ARI,ARI,ARI
+2007,NL,ATL,ATL,E,3,162,,84,78,N,N,N,N,810,5689,1562,328,27,176,534,1149,64,30,49,47,733,665,4.1100000000,1,6,36,4369,1442,172,537,1106,107,140,0.98,Atlanta Braves,Turner Field,2745207,98,98,ATL,ATL,ATL
+2007,NL,CHN,CHC,C,1,162,,85,77,Y,N,N,N,752,5643,1530,340,28,151,500,1054,86,33,40,37,690,650,4.0400000000,2,10,39,4340,1340,165,573,1211,94,134,0.98,Chicago Cubs,Wrigley Field,3252462,105,104,CHC,CHN,CHN
+2007,NL,CIN,CIN,C,5,162,,72,90,N,N,N,N,783,5607,1496,293,23,204,536,1113,97,31,66,46,853,796,4.9400000000,6,7,34,4349,1605,198,482,1068,95,153,0.98,Cincinnati Reds,Great American Ball Park,2058593,104,104,CIN,CIN,CIN
+2007,NL,COL,COL,W,2,163,,90,73,N,Y,Y,N,860,5691,1591,313,36,171,622,1152,100,31,58,44,758,706,4.3200000000,4,7,39,4416,1497,164,504,967,68,179,0.98,Colorado Rockies,Coors Field,2376250,107,108,COL,COL,COL
+2007,NL,FLO,FLA,E,5,162,,71,91,N,N,N,N,790,5627,1504,340,38,201,521,1332,105,34,82,42,891,793,4.9400000000,0,4,40,4331,1617,176,661,1142,137,159,0.97,Florida Marlins,Dolphin Stadium,1370511,97,98,FLA,FLO,FLO
+2007,NL,HOU,HOU,C,4,162,,73,89,N,N,N,N,723,5605,1457,293,30,167,547,1043,65,33,55,40,813,761,4.6800000000,2,6,38,4394,1566,206,510,1109,103,128,0.98,Houston Astros,Minute Maid Park,3020405,100,100,HOU,HOU,HOU
+2007,NL,LAN,LAD,W,4,162,,82,80,N,N,N,N,735,5613,1544,276,35,129,511,864,137,50,41,55,727,677,4.2000000000,4,6,43,4350,1443,146,518,1184,114,160,0.98,Los Angeles Dodgers,Dodger Stadium,3857036,101,100,LAD,LAN,LAN
+2007,NL,MIL,MIL,C,2,162,,83,79,N,N,N,N,801,5554,1455,310,37,231,501,1137,96,32,76,47,776,708,4.4100000000,3,6,49,4333,1513,161,507,1174,109,143,0.98,Milwaukee Brewers,Miller Park,2869144,100,100,MIL,ML4,MIL
+2007,NL,NYN,NYM,E,2,162,,88,74,N,N,N,N,804,5605,1543,294,27,177,549,981,200,46,54,58,750,687,4.2600000000,2,10,39,4357,1415,165,570,1134,101,124,0.98,New York Mets,Shea Stadium,3853955,98,97,NYM,NYN,NYN
+2007,NL,PHI,PHI,E,1,162,,89,73,Y,N,N,N,892,5688,1558,326,41,213,641,1205,138,19,90,52,821,767,4.7300000000,5,5,42,4375,1555,198,558,1050,89,164,0.98,Philadelphia Phillies,Citizens Bank Park,3108325,104,103,PHI,PHI,PHI
+2007,NL,PIT,PIT,C,6,162,,68,94,N,N,N,N,724,5569,1463,322,31,148,463,1135,68,30,71,51,846,793,4.9300000000,4,5,32,4343,1627,174,518,997,83,189,0.98,Pittsburgh Pirates,PNC Park,1749142,97,98,PIT,PIT,PIT
+2007,NL,SDN,SDP,W,3,163,,89,74,N,N,N,N,741,5612,1408,322,31,171,557,1229,55,24,49,44,666,611,3.7000000000,1,20,45,4454,1406,119,474,1136,92,147,0.98,San Diego Padres,Petco Park,2790074,90,90,SDP,SDN,SDN
+2007,NL,SFN,SFG,W,5,162,,71,91,N,N,N,N,683,5538,1407,267,37,131,532,907,119,33,39,36,720,677,4.1900000000,5,10,37,4361,1442,133,593,1057,88,146,0.98,San Francisco Giants,AT&T Park,3223215,101,101,SFG,SFN,SFN
+2007,NL,SLN,STL,C,3,162,,78,84,N,N,N,N,725,5529,1513,279,13,141,506,909,56,33,54,54,829,741,4.6500000000,2,8,34,4307,1514,168,509,945,121,155,0.98,St. Louis Cardinals,Busch Stadium III,3552180,99,99,STL,SLN,SLN
+2007,NL,WAS,WSN,E,4,162,,73,89,N,N,N,N,673,5520,1415,309,31,123,524,1128,69,23,53,41,783,736,4.5800000000,0,6,46,4340,1502,187,580,931,109,151,0.98,Washington Nationals,R.F.K. Stadium,1943812,94,95,WSN,MON,WAS
+2008,AL,BAL,BAL,E,5,161,,68,93,N,N,N,N,782,5559,1486,322,30,172,533,990,81,37,42,48,869,810,5.1500000000,4,4,35,4266,1538,184,687,922,100,163,0.98,Baltimore Orioles,Oriole Park at Camden Yards,1950075,101,102,BAL,BAL,BAL
+2008,AL,BOS,BOS,E,2,162,,95,67,N,Y,N,N,845,5596,1565,353,33,173,646,1068,120,35,70,62,694,645,4.0100000000,5,16,47,4339,1369,147,548,1185,85,147,0.98,Boston Red Sox,Fenway Park II,3048250,108,106,BOS,BOS,BOS
+2008,AL,CHA,CHW,C,1,163,,89,74,Y,N,N,N,811,5553,1458,296,13,235,540,1016,67,34,63,47,729,658,4.1100000000,4,10,34,4373,1471,156,460,1147,108,154,0.98,Chicago White Sox,U.S. Cellular Field,2500648,105,105,CHW,CHA,CHA
+2008,AL,CLE,CLE,C,3,162,,81,81,N,N,N,N,805,5543,1455,339,22,171,560,1213,77,29,103,49,761,711,4.4600000000,10,13,31,4311,1530,170,444,986,94,183,0.98,Cleveland Indians,Jacobs Field,2169760,98,98,CLE,CLE,CLE
+2008,AL,DET,DET,C,5,162,,74,88,N,N,N,N,821,5641,1529,293,41,200,572,1076,63,31,44,44,857,786,4.9100000000,1,2,34,4335,1541,172,644,991,113,172,0.98,Detroit Tigers,Comerica Park,3202645,103,103,DET,DET,DET
+2008,AL,KCA,KCR,C,4,162,,75,87,N,N,N,N,691,5608,1507,303,28,120,392,1005,79,38,50,36,781,720,4.5000000000,2,8,44,4337,1473,159,515,1085,96,159,0.98,Kansas City Royals,Kauffman Stadium,1578922,98,100,KCR,KCA,KCA
+2008,AL,LAA,ANA,W,1,162,,100,62,Y,N,N,N,765,5540,1486,274,25,159,481,987,129,48,52,50,697,644,4.0000000000,7,10,66,4354,1455,160,457,1106,91,159,0.98,Los Angeles Angels of Anaheim,Angel Stadium,3336747,103,102,LAA,ANA,ANA
+2008,AL,MIN,MIN,C,2,163,,88,75,N,N,N,N,829,5641,1572,298,49,111,529,979,102,42,36,72,745,675,4.1800000000,5,10,42,4377,1568,183,406,995,108,168,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2302431,96,96,MIN,MIN,MIN
+2008,AL,NYA,NYY,E,3,162,,89,73,N,N,N,N,789,5572,1512,289,20,180,535,1015,118,39,80,39,727,685,4.2800000000,1,11,42,4325,1478,143,489,1141,83,141,0.98,New York Yankees,Yankee Stadium II,4298655,103,102,NYY,NYA,NYA
+2008,AL,OAK,OAK,W,3,161,,75,86,N,N,N,N,646,5451,1318,270,23,125,574,1226,88,21,48,35,690,640,4.0100000000,4,7,33,4305,1364,135,576,1061,98,169,0.98,Oakland Athletics,McAfee Coliseum,1665256,95,95,OAK,OAK,OAK
+2008,AL,SEA,SEA,W,4,162,,61,101,N,N,N,N,671,5643,1498,285,20,124,417,890,90,32,38,42,811,754,4.7300000000,4,4,36,4306,1544,161,626,1016,99,159,0.98,Seattle Mariners,Safeco Field,2329702,97,97,SEA,SEA,SEA
+2008,AL,TBA,TBD,E,1,162,,97,65,Y,N,Y,N,774,5541,1443,284,37,180,626,1224,142,50,68,52,671,618,3.8200000000,7,12,52,4373,1349,166,526,1143,90,153,0.98,Tampa Bay Rays,Tropicana Field,1811986,101,101,TBR,TBA,TBA
+2008,AL,TEX,TEX,W,2,162,,79,83,N,N,N,N,901,5728,1619,376,35,194,595,1207,81,25,63,54,967,860,5.3700000000,6,8,36,4326,1647,176,625,963,132,191,0.97,Texas Rangers,Rangers Ballpark in Arlington,1945677,102,102,TEX,TEX,TEX
+2008,AL,TOR,TOR,E,4,162,,86,76,N,N,N,N,714,5503,1453,303,32,126,521,938,80,27,59,56,610,561,3.4900000000,15,13,44,4340,1330,134,467,1184,84,137,0.98,Toronto Blue Jays,Rogers Centre,2399786,98,97,TOR,TOR,TOR
+2008,NL,ARI,ARI,W,2,162,,82,80,N,N,N,N,720,5409,1355,318,47,159,587,1287,58,23,49,43,706,635,3.9900000000,6,9,39,4304,1403,147,451,1229,113,136,0.98,Arizona Diamondbacks,Chase Field,2509924,108,108,ARI,ARI,ARI
+2008,NL,ATL,ATL,E,4,162,,72,90,N,N,N,N,753,5604,1514,316,33,130,618,1023,58,27,42,34,778,714,4.4700000000,2,7,26,4322,1439,156,586,1076,107,149,0.98,Atlanta Braves,Turner Field,2532834,98,98,ATL,ATL,ATL
+2008,NL,CHN,CHC,C,1,161,,97,64,Y,N,N,N,855,5588,1552,329,21,184,636,1186,87,34,50,45,671,624,3.8700000000,2,8,44,4352,1329,160,548,1264,99,118,0.98,Chicago Cubs,Wrigley Field,3300200,108,107,CHC,CHN,CHN
+2008,NL,CIN,CIN,C,5,162,,74,88,N,N,N,N,704,5465,1351,269,24,187,560,1125,85,47,50,41,800,729,4.5500000000,2,6,34,4327,1542,201,557,1227,114,155,0.98,Cincinnati Reds,Great American Ball Park,2058632,102,103,CIN,CIN,CIN
+2008,NL,COL,COL,W,3,162,,74,88,N,N,N,N,747,5557,1462,310,28,160,570,1209,141,37,57,38,822,766,4.7700000000,3,8,36,4338,1547,148,562,1041,96,176,0.98,Colorado Rockies,Coors Field,2650218,109,109,COL,COL,COL
+2008,NL,FLO,FLA,E,3,161,,84,77,N,N,N,N,770,5499,1397,302,28,208,543,1371,76,28,69,46,767,707,4.4400000000,2,8,36,4306,1421,161,586,1127,117,122,0.98,Florida Marlins,Dolphin Stadium,1335076,101,101,FLA,FLO,FLO
+2008,NL,HOU,HOU,C,3,161,,86,75,N,N,N,N,712,5451,1432,284,22,167,449,1051,114,52,52,41,743,691,4.3900000000,4,13,48,4276,1453,197,492,1095,67,141,0.98,Houston Astros,Minute Maid Park,2779487,98,98,HOU,HOU,HOU
+2008,NL,LAN,LAD,W,1,162,,84,78,Y,N,N,N,700,5506,1455,271,29,137,543,1032,126,43,43,38,648,591,3.6800000000,5,11,35,4342,1381,123,480,1205,101,137,0.98,Los Angeles Dodgers,Dodger Stadium,3730553,98,97,LAD,LAN,LAN
+2008,NL,MIL,MIL,C,2,162,,90,72,N,Y,N,N,750,5535,1398,324,35,198,550,1203,108,38,69,43,689,623,3.8700000000,12,10,45,4367,1415,175,528,1110,101,160,0.98,Milwaukee Brewers,Miller Park,3068458,98,98,MIL,ML4,MIL
+2008,NL,NYN,NYM,E,2,162,,89,73,N,N,N,N,799,5606,1491,274,38,172,619,1024,138,36,39,49,715,662,4.0700000000,5,12,43,4393,1415,163,590,1181,83,125,0.98,New York Mets,Shea Stadium,4042045,99,98,NYM,NYN,NYN
+2008,NL,PHI,PHI,E,1,162,,92,70,Y,N,Y,Y,799,5509,1407,291,36,214,586,1117,136,25,67,40,680,625,3.8900000000,4,11,47,4349,1444,160,533,1081,90,142,0.98,Philadelphia Phillies,Citizens Bank Park,3422583,103,102,PHI,PHI,PHI
+2008,NL,PIT,PIT,C,6,162,,67,95,N,N,N,N,735,5628,1454,314,21,153,474,1039,57,19,59,51,884,822,5.1000000000,3,7,34,4365,1631,176,657,963,107,178,0.98,Pittsburgh Pirates,PNC Park,1609076,96,98,PIT,PIT,PIT
+2008,NL,SDN,SDP,W,5,162,,63,99,N,N,N,N,637,5568,1390,264,27,154,518,1259,36,17,53,46,764,714,4.4100000000,3,6,30,4375,1466,165,561,1100,85,148,0.98,San Diego Padres,Petco Park,2427535,88,88,SDP,SDN,SDN
+2008,NL,SFN,SFG,W,4,162,,72,90,N,N,N,N,640,5543,1452,311,37,94,452,1044,108,46,48,44,759,701,4.3800000000,4,12,41,4326,1416,147,652,1240,96,130,0.98,San Francisco Giants,AT&T Park,2863837,102,103,SFG,SFN,SFN
+2008,NL,SLN,STL,C,4,162,,86,76,N,N,N,N,779,5636,1585,283,26,174,577,985,73,32,42,44,725,677,4.2000000000,2,7,42,4362,1517,163,496,957,85,156,0.98,St. Louis Cardinals,Busch Stadium III,3432917,99,99,STL,SLN,SLN
+2008,NL,WAS,WSN,E,5,161,,59,102,N,N,N,N,641,5491,1376,269,26,117,534,1095,81,43,67,36,825,742,4.6600000000,2,8,28,4302,1496,190,588,1063,123,141,0.98,Washington Nationals,Nationals Park,2320400,99,100,WSN,MON,WAS
+2009,AL,BAL,BAL,E,5,162,,64,98,N,N,N,N,741,5618,1508,307,19,160,517,1013,76,37,39,46,876,817,5.1600000000,2,3,31,4287,1633,218,546,933,108,151,0.98,Baltimore Orioles,Oriole Park at Camden Yards,1907163,101,102,BAL,BAL,BAL
+2009,AL,BOS,BOS,E,2,162,,95,67,N,Y,N,N,872,5543,1495,335,25,212,659,1120,126,39,70,51,736,695,4.3500000000,8,11,41,4310,1494,167,530,1230,92,119,0.98,Boston Red Sox,Fenway Park II,3062699,106,105,BOS,BOS,BOS
+2009,AL,CHA,CHW,C,3,162,,79,83,N,N,N,N,724,5463,1410,246,20,184,534,1022,113,49,62,39,732,663,4.1600000000,4,11,36,4319,1438,169,507,1119,128,156,0.98,Chicago White Sox,U.S. Cellular Field,2284163,105,105,CHW,CHA,CHA
+2009,AL,CLE,CLE,C,4,162,,65,97,N,N,N,N,773,5568,1468,314,28,161,582,1211,84,31,81,50,865,806,5.0700000000,5,6,25,4302,1570,183,598,986,114,169,0.98,Cleveland Indians,Jacobs Field,1766242,95,95,CLE,CLE,CLE
+2009,AL,DET,DET,C,2,163,,86,77,N,N,N,N,743,5540,1443,245,35,183,540,1114,72,33,61,39,745,690,4.3400000000,4,9,42,4341,1449,182,594,1102,106,164,0.98,Detroit Tigers,Comerica Park,2567165,101,102,DET,DET,DET
+2009,AL,KCA,KCR,C,4,162,,65,97,N,N,N,N,686,5532,1432,276,51,144,457,1091,88,29,42,32,842,765,4.8300000000,10,9,34,4278,1486,166,600,1153,127,159,0.98,Kansas City Royals,Kauffman Stadium,1797891,97,99,KCR,KCA,KCA
+2009,AL,LAA,ANA,W,1,162,,97,65,Y,N,N,N,883,5622,1604,293,33,173,547,1054,148,63,41,52,761,715,4.4500000000,9,13,51,4335,1513,180,523,1062,99,174,0.98,Los Angeles Angels of Anaheim,Angel Stadium,3240386,99,98,LAA,ANA,ANA
+2009,AL,MIN,MIN,C,1,163,,87,76,Y,N,N,N,817,5608,1539,271,40,172,585,1021,85,32,45,57,765,726,4.5000000000,4,7,48,4359,1542,185,466,1052,89,133,0.98,Minnesota Twins,Hubert H Humphrey Metrodome,2416237,99,98,MIN,MIN,MIN
+2009,AL,NYA,NYY,E,1,162,,103,59,Y,N,Y,Y,915,5660,1604,325,21,244,663,1014,111,28,54,39,753,687,4.2800000000,3,8,51,4350,1386,181,574,1260,101,131,0.98,New York Yankees,Yankee Stadium III,3719358,105,103,NYY,NYA,NYA
+2009,AL,OAK,OAK,W,4,162,,75,87,N,N,N,N,759,5584,1464,307,21,135,527,1046,133,48,50,54,761,685,4.2900000000,2,10,38,4342,1486,156,523,1124,121,153,0.98,Oakland Athletics,Oakland-Alameda Country Coliseum,1408783,99,99,OAK,OAK,OAK
+2009,AL,SEA,SEA,W,3,162,,85,77,N,N,N,N,640,5543,1430,280,19,160,421,1093,89,33,49,44,692,625,3.8700000000,4,10,49,4358,1359,172,534,1043,120,150,0.98,Seattle Mariners,Safeco Field,2195533,94,96,SEA,SEA,SEA
+2009,AL,TBA,TBD,E,3,162,,84,78,N,N,N,N,803,5462,1434,297,36,199,642,1229,194,61,49,45,754,686,4.3600000000,3,5,41,4282,1421,183,515,1125,111,135,0.98,Tampa Bay Rays,Tropicana Field,1874962,98,97,TBR,TBA,TBA
+2009,AL,TEX,TEX,W,2,162,,87,75,N,N,N,N,784,5526,1436,296,27,224,472,1253,149,36,37,51,740,698,4.3800000000,8,11,45,4304,1432,171,531,1016,120,166,0.98,Texas Rangers,Rangers Ballpark in Arlington,2156016,104,104,TEX,TEX,TEX
+2009,AL,TOR,TOR,E,4,162,,75,87,N,N,N,N,798,5696,1516,339,13,209,548,1028,73,23,45,49,771,720,4.4700000000,10,10,25,4353,1509,181,551,1181,88,167,0.98,Toronto Blue Jays,Rogers Centre,1876129,100,99,TOR,TOR,TOR
+2009,NL,ARI,ARI,W,5,162,,70,92,N,N,N,N,720,5565,1408,307,45,173,571,1298,102,40,37,41,782,711,4.4400000000,4,12,36,4343,1470,168,525,1158,151,132,0.97,Arizona Diamondbacks,Chase Field,2128765,105,106,ARI,ARI,ARI
+2009,NL,ATL,ATL,E,3,162,,86,76,N,N,N,N,735,5539,1459,300,20,149,602,1064,58,26,52,47,641,581,3.5700000000,3,10,38,4388,1399,119,530,1232,107,158,0.98,Atlanta Braves,Turner Field,2373631,99,98,ATL,ATL,ATL
+2009,NL,CHN,CHC,C,2,161,,83,78,N,N,N,N,707,5486,1398,293,29,161,592,1185,56,34,59,42,672,616,3.8400000000,3,8,40,4336,1329,160,586,1272,127,144,0.98,Chicago Cubs,Wrigley Field,3168859,107,106,CHC,CHN,CHN
+2009,NL,CIN,CIN,C,4,162,,78,84,N,N,N,N,673,5462,1349,280,25,158,531,1129,96,40,53,41,723,677,4.1800000000,6,12,41,4375,1420,188,577,1069,104,161,0.98,Cincinnati Reds,Great American Ball Park,1747919,100,100,CIN,CIN,CIN
+2009,NL,COL,COL,W,2,162,,92,70,N,Y,N,N,804,5398,1408,300,50,190,660,1277,106,55,47,60,715,675,4.2400000000,5,7,45,4315,1427,141,528,1154,103,144,0.98,Colorado Rockies,Coors Field,2665080,112,112,COL,COL,COL
+2009,NL,FLO,FLA,E,2,162,,87,75,N,N,N,N,772,5572,1493,296,25,159,568,1226,75,35,63,39,766,690,4.3200000000,5,5,45,4339,1425,160,601,1248,118,129,0.98,Florida Marlins,Dolphin Stadium,1464109,102,102,FLA,FLO,FLO
+2009,NL,HOU,HOU,C,5,162,,74,88,N,N,N,N,643,5436,1415,270,32,142,448,990,113,44,43,45,770,722,4.5400000000,5,10,39,4290,1521,176,546,1144,88,161,0.98,Houston Astros,Minute Maid Park,2521076,98,98,HOU,HOU,HOU
+2009,NL,LAN,LAD,W,1,162,,95,67,Y,N,N,N,780,5592,1511,278,39,145,607,1068,116,48,63,44,611,558,3.4100000000,1,9,44,4420,1265,127,584,1272,97,134,0.98,Los Angeles Dodgers,Dodger Stadium,3761655,96,95,LAD,LAN,LAN
+2009,NL,MIL,MIL,C,3,162,,80,82,N,N,N,N,785,5510,1447,281,37,182,610,1231,68,37,71,47,818,770,4.8400000000,1,8,44,4305,1498,207,607,1104,111,150,0.98,Milwaukee Brewers,Miller Park,3037451,98,98,MIL,ML4,MIL
+2009,NL,NYN,NYM,E,4,162,,70,92,N,N,N,N,671,5453,1472,295,49,95,526,928,122,44,36,55,757,705,4.4600000000,3,12,39,4278,1452,158,616,1031,108,134,0.98,New York Mets,Citi Field,3168571,97,97,NYM,NYN,NYN
+2009,NL,PHI,PHI,E,1,162,,93,69,Y,N,Y,N,820,5578,1439,312,35,224,589,1155,119,28,71,45,709,673,4.1600000000,8,9,44,4367,1479,189,489,1153,86,132,0.98,Philadelphia Phillies,Citizens Bank Park,3600693,101,100,PHI,PHI,PHI
+2009,NL,PIT,PIT,C,6,161,,62,99,N,N,N,N,636,5417,1364,289,34,125,499,1142,90,32,46,36,768,723,4.5900000000,5,7,28,4255,1491,152,563,919,80,169,0.98,Pittsburgh Pirates,PNC Park,1577853,98,99,PIT,PIT,PIT
+2009,NL,SDN,SDP,W,4,162,,75,87,N,N,N,N,638,5425,1315,265,31,141,586,1182,82,29,57,36,769,704,4.3700000000,2,9,45,4352,1422,167,603,1187,109,144,0.98,San Diego Padres,Petco Park,1919603,90,90,SDP,SDN,SDN
+2009,NL,SFN,SFG,W,3,162,,88,74,N,N,N,N,657,5493,1411,275,43,122,392,1158,78,28,50,55,611,571,3.5500000000,11,18,41,4338,1268,140,584,1302,99,138,0.98,San Francisco Giants,AT&T Park,2862110,102,102,SFG,SFN,SFN
+2009,NL,SLN,STL,C,1,162,,91,71,Y,N,N,N,730,5465,1436,294,29,160,528,1041,75,31,61,43,640,586,3.6600000000,8,11,43,4322,1407,123,460,1049,113,167,0.98,St. Louis Cardinals,Busch Stadium III,3343252,98,97,STL,SLN,SLN
+2009,NL,WAS,WSN,E,5,162,,59,103,N,N,N,N,710,5493,1416,271,38,156,617,1208,73,40,56,42,874,791,5.0200000000,6,3,33,4273,1533,173,629,911,170,154,0.97,Washington Nationals,Nationals Park,1817226,99,101,WSN,MON,WAS
+2010,AL,BAL,BAL,E,5,162,,66,96,N,N,N,N,613,5554,1440,264,21,133,424,1056,76,34,54,45,785,733,4.5900000000,3,7,35,4309,1508,186,520,1007,123,141,0.98,Baltimore Orioles,Oriole Park at Camden Yards,,101,102,BAL,BAL,BAL
+2010,AL,BOS,BOS,E,3,162,,89,73,N,N,N,N,818,5646,1511,358,22,211,587,1140,68,17,47,46,744,679,4.2000000000,3,9,44,4370,1402,152,580,1207,127,132,0.98,Boston Red Sox,Fenway Park II,,106,105,BOS,BOS,BOS
+2010,AL,CHA,CHW,C,2,162,,88,74,N,N,N,N,752,5484,1467,263,21,177,467,922,160,74,79,38,704,658,4.0900000000,6,11,43,4339,1471,136,490,1149,120,157,0.98,Chicago White Sox,U.S. Cellular Field,,105,105,CHW,CHA,CHA
+2010,AL,CLE,CLE,C,4,162,,69,93,N,N,N,N,646,5487,1362,290,20,128,545,1184,91,33,64,33,752,684,4.3000000000,10,4,34,4299,1477,147,572,967,126,179,0.98,Cleveland Indians,Jacobs Field,,93,93,CLE,CLE,CLE
+2010,AL,DET,DET,C,3,162,,81,81,N,N,N,N,751,5643,1515,308,32,152,546,1147,69,30,41,41,743,690,4.3000000000,6,5,32,4333,1445,142,537,1056,136,171,0.98,Detroit Tigers,Comerica Park,,101,101,DET,DET,DET
+2010,AL,KCA,KCR,C,5,162,,67,95,N,N,N,N,676,5604,1534,279,31,121,471,905,115,50,35,53,845,794,4.9700000000,7,3,44,4310,1553,176,551,1035,138,138,0.98,Kansas City Royals,Kauffman Stadium,,99,101,KCR,KCA,KCA
+2010,AL,LAA,ANA,W,3,162,,80,82,N,N,N,N,681,5488,1363,276,19,155,466,1070,104,52,52,37,702,651,4.0400000000,10,9,39,4348,1422,148,565,1130,131,116,0.98,Los Angeles Angels of Anaheim,Angel Stadium,,98,98,LAA,ANA,ANA
+2010,AL,MIN,MIN,C,1,162,,94,68,Y,N,N,N,781,5568,1521,318,41,142,559,967,68,28,39,53,671,638,3.9500000000,9,13,40,4358,1493,155,383,1048,94,149,0.98,Minnesota Twins,Target Field,,101,100,MIN,MIN,MIN
+2010,AL,NYA,NYY,E,2,162,,95,67,N,Y,N,N,859,5567,1485,275,32,201,662,1136,103,30,73,44,693,651,4.0600000000,3,8,39,4327,1349,179,540,1154,79,161,0.98,New York Yankees,Yankee Stadium III,,105,103,NYY,NYA,NYA
+2010,AL,OAK,OAK,W,2,162,,81,81,N,N,N,N,663,5448,1396,276,30,109,527,1061,156,38,47,51,626,566,3.5800000000,7,17,38,4295,1315,153,512,1070,110,146,0.98,Oakland Athletics,Oakland-Alameda Country Coliseum,,100,100,OAK,OAK,OAK
+2010,AL,SEA,SEA,W,4,162,,61,101,N,N,N,N,513,5409,1274,227,16,101,459,1184,142,39,39,40,698,628,3.9500000000,11,10,38,4314,1402,157,452,973,122,146,0.98,Seattle Mariners,Safeco Field,,93,95,SEA,SEA,SEA
+2010,AL,TBA,TBD,E,1,162,,96,66,Y,N,N,N,802,5439,1343,295,37,160,672,1292,172,47,57,57,649,611,3.7800000000,6,12,51,4361,1347,175,478,1189,95,134,0.98,Tampa Bay Rays,Tropicana Field,,96,95,TBR,TBA,TBA
+2010,AL,TEX,TEX,W,1,162,,90,72,Y,N,Y,N,787,5635,1556,268,25,162,511,986,123,48,45,54,687,636,3.9300000000,7,8,46,4366,1355,162,551,1181,123,132,0.98,Texas Rangers,Rangers Ballpark in Arlington,,105,104,TEX,TEX,TEX
+2010,AL,TOR,TOR,E,4,162,,85,77,N,N,N,N,755,5495,1364,319,21,257,471,1164,58,20,55,34,728,676,4.2300000000,5,11,45,4322,1407,150,539,1184,102,172,0.98,Toronto Blue Jays,Rogers Centre,,100,100,TOR,TOR,TOR
+2010,NL,ARI,ARI,W,5,162,,65,97,N,N,N,N,713,5473,1366,301,34,180,589,1529,86,41,39,41,836,765,4.8100000000,3,3,35,4296,1503,210,548,1070,121,152,0.98,Arizona Diamondbacks,Chase Field,,104,105,ARI,ARI,ARI
+2010,NL,ATL,ATL,E,2,162,,91,71,N,Y,N,N,738,5463,1411,312,25,139,634,1140,63,29,51,35,629,569,3.5700000000,2,9,41,4318,1326,126,505,1241,145,165,0.98,Atlanta Braves,Turner Field,,98,97,ATL,ATL,ATL
+2010,NL,CHN,CHC,C,5,162,,75,87,N,N,N,N,685,5512,1414,298,27,149,479,1236,55,31,50,38,767,668,4.2400000000,1,14,40,4310,1409,154,605,1268,144,135,0.98,Chicago Cubs,Wrigley Field,,108,108,CHC,CHN,CHN
+2010,NL,CIN,CIN,C,1,162,,91,71,Y,N,N,N,790,5579,1515,293,30,188,522,1218,93,43,68,50,685,648,4.0200000000,4,9,43,4359,1404,158,524,1130,86,140,0.98,Cincinnati Reds,Great American Ball Park,,99,99,CIN,CIN,CIN
+2010,NL,COL,COL,W,3,162,,83,79,N,N,N,N,770,5530,1452,270,54,173,585,1274,99,42,47,47,717,663,4.1400000000,6,12,35,4326,1405,139,525,1234,113,182,0.98,Colorado Rockies,Coors Field,,115,115,COL,COL,COL
+2010,NL,FLO,FLA,E,3,162,,80,82,N,N,N,N,719,5531,1403,294,37,152,514,1375,92,26,55,43,717,652,4.0900000000,5,17,39,4315,1433,134,549,1168,141,130,0.98,Florida Marlins,Dolphin Stadium,,105,104,FLA,FLO,FLO
+2010,NL,HOU,HOU,C,4,162,,76,86,N,N,N,N,611,5452,1348,252,25,108,415,1025,100,36,33,29,729,654,4.0900000000,4,11,45,4318,1446,140,548,1210,118,135,0.98,Houston Astros,Minute Maid Park,,95,96,HOU,HOU,HOU
+2010,NL,LAN,LAD,W,4,162,,80,82,N,N,N,N,667,5426,1368,270,29,120,533,1184,92,50,46,50,692,643,4.0100000000,4,16,41,4325,1323,134,539,1274,114,122,0.98,Los Angeles Dodgers,Dodger Stadium,,95,95,LAD,LAN,LAN
+2010,NL,MIL,MIL,C,3,162,,77,85,N,N,N,N,750,5606,1471,293,33,182,546,1216,81,26,81,35,804,733,4.5900000000,3,7,35,4317,1487,173,582,1258,115,141,0.98,Milwaukee Brewers,Miller Park,,98,98,MIL,ML4,MIL
+2010,NL,NYN,NYM,E,4,162,,79,83,N,N,N,N,656,5465,1361,266,40,128,502,1095,130,44,46,57,652,597,3.7300000000,8,19,36,4359,1438,135,545,1106,96,160,0.98,New York Mets,Citi Field,,97,97,NYM,NYN,NYN
+2010,NL,PHI,PHI,E,1,162,,97,65,Y,N,N,N,772,5581,1451,290,34,166,560,1064,108,21,63,43,640,594,3.6800000000,14,21,40,4369,1402,168,416,1183,93,156,0.98,Philadelphia Phillies,Citizens Bank Park,,101,100,PHI,PHI,PHI
+2010,NL,PIT,PIT,C,6,162,,57,105,N,N,N,N,587,5386,1303,276,27,126,463,1207,87,36,33,33,866,784,5.0000000000,1,6,31,4235,1567,167,538,1026,142,119,0.98,Pittsburgh Pirates,PNC Park,,99,101,PIT,PIT,PIT
+2010,NL,SDN,SDP,W,2,162,,90,72,N,N,N,N,665,5434,1338,236,24,132,538,1183,124,50,50,46,581,549,3.4100000000,2,20,49,4369,1305,139,517,1295,85,142,0.98,San Diego Padres,Petco Park,,91,91,SDP,SDN,SDN
+2010,NL,SFN,SFG,W,1,162,,92,70,Y,N,Y,Y,697,5488,1411,284,30,162,487,1099,55,32,50,41,583,546,3.3600000000,6,17,57,4383,1279,134,578,1331,80,110,0.98,San Francisco Giants,AT&T Park,,101,101,SFG,SFN,SFN
+2010,NL,SLN,STL,C,2,162,,86,76,N,N,N,N,736,5542,1456,285,18,150,541,1027,79,41,50,40,641,577,3.5700000000,7,16,32,4361,1412,133,477,1094,109,170,0.98,St. Louis Cardinals,Busch Stadium III,,97,97,STL,SLN,SLN
+2010,NL,WAS,WSN,E,5,162,,69,93,N,N,N,N,655,5418,1355,250,31,149,503,1220,110,41,60,47,742,658,4.1300000000,2,5,37,4305,1469,151,512,1068,142,146,0.98,Washington Nationals,Nationals Park,,99,100,WSN,MON,WAS
+2011,AL,BAL,BAL,E,5,162,,69,93,N,N,N,N,708,5585,1434,273,13,191,452,1120,81,25,52,43,860,786,4.9200000000,3,7,32,4340,1568,210,535,1044,110,159,0.98,Baltimore Orioles,Oriole Park at Camden Yards,1755461,96,97,BAL,BAL,BAL
+2011,AL,BOS,BOS,E,3,162,,90,72,N,N,N,N,875,5710,1600,352,35,203,578,1108,102,42,50,50,737,680,4.2000000000,2,13,36,4372,1366,156,540,1213,92,120,0.99,Boston Red Sox,Fenway Park II,3054001,106,105,BOS,BOS,BOS
+2011,AL,CHA,CHW,C,3,162,,79,83,N,N,N,N,654,5502,1387,252,16,154,475,989,81,53,84,46,706,665,4.1000000000,6,14,42,4380,1463,147,439,1220,79,128,0.99,Chicago White Sox,U.S. Cellular Field,2001117,99,100,CHW,CHA,CHA
+2011,AL,CLE,CLE,C,2,162,,80,82,N,N,N,N,704,5509,1380,290,26,154,494,1269,89,42,65,43,760,683,4.2400000000,2,4,38,4360,1482,153,463,1024,110,131,0.98,Cleveland Indians,Jacobs Field,1840835,96,97,CLE,CLE,CLE
+2011,AL,DET,DET,C,1,162,,95,67,Y,N,N,N,787,5563,1540,297,34,169,521,1143,49,20,39,58,711,647,4.0400000000,4,14,52,4320,1406,149,492,1115,103,146,0.98,Detroit Tigers,Comerica Park,2642045,103,102,DET,DET,DET
+2011,AL,KCA,KCR,C,4,162,,71,91,N,N,N,N,730,5672,1560,325,41,129,442,1006,153,58,39,57,762,716,4.4500000000,2,6,37,4354,1487,163,557,1080,95,149,0.99,Kansas City Royals,Kauffman Stadium,1724450,101,101,KCR,KCA,KCA
+2011,AL,LAA,ANA,W,2,162,,86,76,N,N,N,N,667,5513,1394,289,34,155,442,1086,135,52,51,32,633,581,3.5700000000,12,11,39,4395,1388,142,476,1058,93,157,0.99,Los Angeles Angels of Anaheim,Angel Stadium,3166321,93,93,LAA,ANA,ANA
+2011,AL,MIN,MIN,C,5,162,,63,99,N,N,N,N,619,5487,1357,259,25,103,440,1048,92,39,37,25,804,724,4.6000000000,7,8,32,4265,1564,161,480,940,119,153,0.98,Minnesota Twins,Target Field,3168116,95,97,MIN,MIN,MIN
+2011,AL,NYA,NYY,E,1,162,,97,65,Y,N,N,N,867,5518,1452,267,33,222,627,1138,147,46,74,51,657,605,3.7300000000,5,8,47,4375,1423,152,507,1222,102,140,0.98,New York Yankees,Yankee Stadium III,3653680,108,106,NYY,NYA,NYA
+2011,AL,OAK,OAK,W,3,162,,74,88,N,N,N,N,645,5452,1330,280,29,114,509,1094,117,43,50,57,679,597,3.7100000000,6,12,39,4343,1380,136,519,1160,124,145,0.98,Oakland Athletics,Oakland-Alameda Country Coliseum,1476791,97,98,OAK,OAK,OAK
+2011,AL,SEA,SEA,W,4,162,,67,95,N,N,N,N,556,5421,1263,253,22,109,435,1280,125,40,37,41,675,621,3.9100000000,12,10,39,4299,1369,145,436,1088,108,152,0.98,Seattle Mariners,Safeco Field,1939421,94,95,SEA,SEA,SEA
+2011,AL,TBA,TBD,E,2,162,,91,71,N,Y,N,N,707,5436,1324,273,37,172,571,1193,155,62,73,35,614,577,3.5800000000,15,13,32,4347,1263,161,504,1143,73,139,0.99,Tampa Bay Rays,Tropicana Field,1529188,92,92,TBR,TBA,TBA
+2011,AL,TEX,TEX,W,1,162,,96,66,Y,N,Y,N,855,5659,1599,310,32,210,475,930,143,45,39,49,677,607,3.7900000000,10,19,38,4324,1327,170,461,1179,114,162,0.98,Texas Rangers,Rangers Ballpark in Arlington,2946949,117,115,TEX,TEX,TEX
+2011,AL,TOR,TOR,E,4,162,,81,81,N,N,N,N,743,5559,1384,285,34,186,525,1184,131,52,48,47,761,700,4.3300000000,7,10,33,4376,1433,179,540,1169,110,146,0.98,Toronto Blue Jays,Rogers Centre,1818103,104,104,TOR,TOR,TOR
+2011,NL,ARI,ARI,W,1,162,,94,68,Y,N,N,N,731,5421,1357,293,37,172,531,1249,133,55,61,33,662,609,3.8000000000,5,12,58,4330,1414,159,442,1058,90,130,0.99,Arizona Diamondbacks,Chase Field,2105432,107,106,ARI,ARI,ARI
+2011,NL,ATL,ATL,E,2,162,,89,73,N,N,N,N,641,5528,1345,244,16,173,504,1260,77,44,28,30,605,572,3.4900000000,3,16,52,4439,1332,125,521,1332,83,131,0.99,Atlanta Braves,Turner Field,2372940,98,98,ATL,ATL,ATL
+2011,NL,CHN,CHC,C,5,162,,71,91,N,N,N,N,654,5549,1423,285,36,148,425,1202,69,23,59,35,756,690,4.3400000000,4,5,40,4303,1439,162,580,1224,134,128,0.98,Chicago Cubs,Wrigley Field,3017966,96,97,CHC,CHN,CHN
+2011,NL,CIN,CIN,C,3,162,,79,83,N,N,N,N,735,5612,1438,264,19,183,535,1250,97,50,63,40,720,678,4.1600000000,4,5,39,4403,1414,185,539,1112,91,145,0.99,Cincinnati Reds,Great American Ball Park,2213588,107,106,CIN,CIN,CIN
+2011,NL,COL,COL,W,4,162,,73,89,N,N,N,N,735,5544,1429,274,40,163,555,1201,118,42,57,44,774,713,4.4400000000,5,7,41,4343,1471,176,522,1118,98,156,0.98,Colorado Rockies,Coors Field,2909777,116,116,COL,COL,COL
+2011,NL,FLO,FLA,E,5,162,,72,90,N,N,N,N,625,5508,1358,274,30,149,542,1244,95,41,51,42,702,640,3.9500000000,7,11,40,4379,1403,149,500,1218,93,126,0.99,Florida Marlins,Sun Life Stadium,1477462,99,100,FLA,FLO,FLO
+2011,NL,HOU,HOU,C,6,162,,56,106,N,N,N,N,615,5598,1442,309,28,95,401,1164,118,33,46,37,796,719,4.5100000000,2,6,25,4305,1477,188,560,1191,116,139,0.98,Houston Astros,Minute Maid Park,2067016,102,104,HOU,HOU,HOU
+2011,NL,LAN,LAD,W,3,161,,82,79,N,N,N,N,644,5436,1395,237,28,117,498,1087,126,40,45,43,612,563,3.5600000000,7,17,40,4296,1287,132,507,1265,85,121,0.99,Los Angeles Dodgers,Dodger Stadium,2935139,98,98,LAD,LAN,LAN
+2011,NL,MIL,MIL,C,1,162,,96,66,Y,N,N,N,721,5447,1422,276,31,185,481,1083,94,31,56,44,638,582,3.6400000000,1,13,47,4325,1348,147,440,1257,111,131,0.98,Milwaukee Brewers,Miller Park,3071373,103,102,MIL,ML4,MIL
+2011,NL,NYN,NYM,E,4,162,,77,85,N,N,N,N,718,5600,1477,309,39,108,571,1085,130,35,51,48,742,674,4.1900000000,6,9,43,4344,1482,147,514,1126,116,125,0.98,New York Mets,Citi Field,2352596,98,98,NYM,NYN,NYN
+2011,NL,PHI,PHI,E,1,162,,102,60,Y,N,N,N,713,5579,1409,258,38,153,539,1024,96,24,56,38,529,495,3.0200000000,18,21,47,4431,1320,120,404,1299,74,134,0.99,Philadelphia Phillies,Citizens Bank Park,3680718,105,103,PHI,PHI,PHI
+2011,NL,PIT,PIT,C,4,162,,72,90,N,N,N,N,610,5421,1325,277,35,107,489,1308,108,52,34,44,712,650,4.0500000000,5,11,43,4348,1513,152,535,1031,112,154,0.98,Pittsburgh Pirates,PNC Park,1940429,96,97,PIT,PIT,PIT
+2011,NL,SDN,SDP,W,5,162,,71,91,N,N,N,N,593,5417,1284,247,42,91,501,1320,170,44,48,47,611,551,3.4300000000,0,10,44,4348,1324,125,521,1139,94,138,0.99,San Diego Padres,Petco Park,2143018,92,92,SDP,SDN,SDN
+2011,NL,SFN,SFG,W,2,162,,86,76,N,N,N,N,570,5486,1327,282,24,121,448,1122,85,51,52,43,578,522,3.2100000000,3,12,52,4404,1260,96,559,1316,104,127,0.98,San Francisco Giants,AT&T Park,3387303,89,89,SFG,SFN,SFN
+2011,NL,SLN,STL,C,2,162,,90,72,N,Y,Y,Y,762,5532,1513,308,22,162,542,978,57,39,44,40,692,608,3.7900000000,7,9,47,4386,1461,136,448,1098,116,167,0.98,St. Louis Cardinals,Busch Stadium III,3093954,95,94,STL,SLN,SLN
+2011,NL,WAS,WSN,E,3,161,,80,81,N,N,N,N,624,5441,1319,257,22,154,470,1323,106,38,65,32,643,577,3.5800000000,3,10,49,4348,1403,129,477,1049,104,145,0.98,Washington Nationals,Nationals Park,1940478,100,100,WSN,MON,WAS
+2012,AL,BAL,BAL,E,2,162,81,93,69,N,Y,N,N,712,5560,1375,270,16,214,480,1315,58,29,50,30,705,642,3.9000000000,1,1,55,4449,1433,184,481,1177,106,150,0.98,Baltimore Orioles,Oriole Park at Camden Yards,2102240,102,103,BAL,BAL,BAL
+2012,AL,BOS,BOS,E,5,162,81,69,93,N,N,N,N,734,5604,1459,339,16,165,428,1197,97,31,45,55,806,754,4.7200000000,6,2,35,4329,1449,190,529,1176,101,159,0.98,Boston Red Sox,Fenway Park II,3043003,106,106,BOS,BOS,BOS
+2012,AL,CHA,CHW,C,2,162,81,85,77,N,N,N,N,748,5518,1409,228,29,211,461,1203,109,43,65,36,676,646,4.0200000000,6,2,37,4337,1365,186,503,1246,70,154,0.99,Chicago White Sox,U.S. Cellular Field,1965955,106,106,CHW,CHA,CHA
+2012,AL,CLE,CLE,C,4,162,81,68,94,N,N,N,N,667,5525,1385,266,24,136,555,1087,110,44,59,39,845,766,4.7900000000,2,1,43,4326,1503,174,543,1086,96,156,0.98,Cleveland Indians,Progressive Field,1603596,93,95,CLE,CLE,CLE
+2012,AL,DET,DET,C,1,162,81,88,74,Y,N,Y,N,726,5476,1467,279,39,163,511,1103,59,23,57,39,670,596,3.7700000000,9,3,40,4292,1409,151,438,1318,99,127,0.98,Detroit Tigers,Comerica Park,3028033,104,103,DET,DET,DET
+2012,AL,KCA,KCR,C,3,162,81,72,90,N,N,N,N,676,5636,1492,295,37,131,404,1032,132,38,42,41,746,693,4.3000000000,2,1,44,4354,1504,163,542,1177,113,169,0.98,Kansas City Royals,Kauffman Stadium,1739859,100,100,KCR,KCA,KCA
+2012,AL,LAA,ANA,W,3,162,81,89,73,N,N,N,N,767,5536,1518,273,22,187,449,1113,134,33,47,41,699,640,4.0200000000,6,5,38,4300,1339,186,483,1157,98,138,0.98,Los Angeles Angels of Anaheim,Angel Stadium of Anaheim,3061770,92,92,LAA,ANA,ANA
+2012,AL,MIN,MIN,C,5,162,81,66,96,N,N,N,N,701,5562,1448,270,30,131,505,1069,135,37,53,56,832,762,4.7700000000,3,1,35,4316,1536,198,465,943,107,187,0.98,Minnesota Twins,Target Field,2776354,97,99,MIN,MIN,MIN
+2012,AL,NYA,NYY,E,1,162,81,95,67,Y,N,N,N,804,5524,1462,280,13,245,565,1176,93,27,62,49,668,618,3.8600000000,6,2,51,4336,1401,190,431,1318,74,135,0.99,New York Yankees,Yankee Stadium III,3542406,103,102,NYY,NYA,NYA
+2012,AL,OAK,OAK,W,1,162,81,94,68,Y,N,N,N,713,5527,1315,267,32,195,550,1387,122,32,45,34,614,569,3.5000000000,1,0,47,4410,1360,147,462,1136,111,136,0.98,Oakland Athletics,O.co Coliseum,1679013,97,97,OAK,OAK,OAK
+2012,AL,SEA,SEA,W,4,162,81,75,87,N,N,N,N,619,5494,1285,241,27,149,466,1259,104,35,30,35,651,608,3.7600000000,8,6,43,4370,1359,166,449,1166,72,153,0.99,Seattle Mariners,Safeco Field,1721920,90,91,SEA,SEA,SEA
+2012,AL,TBA,TBD,E,3,162,81,90,72,N,N,N,N,697,5398,1293,250,30,175,571,1323,134,44,58,42,577,518,3.1900000000,7,4,50,4379,1233,139,469,1383,114,154,0.98,Tampa Bay Rays,Tropicana Field,1559681,94,93,TBR,TBA,TBA
+2012,AL,TEX,TEX,W,2,162,81,93,69,N,Y,N,N,808,5590,1526,303,32,200,478,1103,91,44,57,53,707,639,4.0200000000,7,2,43,4326,1378,175,446,1286,85,136,0.99,Texas Rangers,Rangers Ballpark in Arlington,3460280,112,111,TEX,TEX,TEX
+2012,AL,TOR,TOR,E,4,162,81,73,89,N,N,N,N,716,5487,1346,247,22,198,473,1251,123,41,55,45,784,745,4.6400000000,5,4,29,4331,1439,204,574,1142,101,168,0.98,Toronto Blue Jays,Rogers Centre,2099663,103,104,TOR,TOR,TOR
+2012,NL,ARI,ARI,W,3,162,81,81,81,N,N,N,N,734,5462,1416,307,33,165,539,1266,93,51,41,45,688,626,3.9400000000,4,2,39,4301,1432,155,417,1200,90,144,0.99,Arizona Diamondbacks,Chase Field,2177617,105,106,ARI,ARI,ARI
+2012,NL,ATL,ATL,E,2,162,81,94,68,N,Y,N,N,700,5425,1341,263,30,149,567,1289,101,32,34,46,600,549,3.4200000000,5,4,47,4336,1310,145,464,1232,86,146,0.99,Atlanta Braves,Turner Field,2420171,102,101,ATL,ATL,ATL
+2012,NL,CHN,CHC,C,5,162,81,61,101,N,N,N,N,613,5411,1297,265,36,137,447,1235,94,45,43,24,759,708,4.5100000000,1,0,28,4241,1399,175,573,1128,105,148,0.98,Chicago Cubs,Wrigley Field,2882756,98,99,CHC,CHN,CHN
+2012,NL,CIN,CIN,C,1,162,81,97,65,Y,N,N,N,669,5477,1377,296,30,172,481,1266,87,27,47,37,588,540,3.3400000000,9,2,56,4359,1356,152,427,1248,89,113,0.99,Cincinnati Reds,Great American Ball Park,2347251,107,107,CIN,CIN,CIN
+2012,NL,COL,COL,W,5,162,81,64,98,N,N,N,N,758,5577,1526,306,52,166,450,1213,100,40,36,39,890,824,5.2200000000,0,0,36,4266,1637,198,566,1144,122,138,0.98,Colorado Rockies,Coors Field,2630458,120,121,COL,COL,COL
+2012,NL,HOU,HOU,C,6,162,81,55,107,N,N,N,N,583,5407,1276,238,28,146,463,1365,105,46,58,30,794,721,4.5700000000,3,2,31,4270,1493,173,540,1170,118,131,0.98,Houston Astros,Minute Maid Park,1607733,99,101,HOU,HOU,HOU
+2012,NL,LAN,LAD,W,2,162,81,86,76,N,N,N,N,637,5438,1369,269,23,116,481,1156,104,44,52,38,597,538,3.3500000000,2,2,40,4349,1277,122,539,1276,98,137,0.98,Los Angeles Dodgers,Dodger Stadium,3324246,96,96,LAD,LAN,LAN
+2012,NL,MIA,FLA,E,5,162,81,69,93,N,N,N,N,609,5437,1327,261,39,137,484,1228,149,41,35,40,724,655,4.1000000000,5,3,38,4322,1448,133,495,1113,103,154,0.98,Miami Marlins,Marlins Park,2219444,100,100,MIA,FLO,MIA
+2012,NL,MIL,MIL,C,3,162,81,83,79,N,N,N,N,776,5557,1442,300,39,202,466,1240,158,39,90,35,733,682,4.2200000000,0,0,44,4361,1458,169,525,1402,99,132,0.98,Milwaukee Brewers,Miller Park,2831385,104,104,MIL,ML4,MIL
+2012,NL,NYN,NYM,E,4,162,81,74,88,N,N,N,N,650,5450,1357,286,21,139,503,1250,79,38,42,30,709,651,4.0900000000,7,5,36,4302,1368,161,488,1240,101,135,0.98,New York Mets,Citi Field,2242803,96,97,NYM,NYN,NYN
+2012,NL,PHI,PHI,E,3,162,81,81,81,N,N,N,N,684,5544,1414,271,28,158,454,1094,116,23,63,39,680,618,3.8600000000,5,4,42,4354,1387,178,409,1385,101,118,0.98,Philadelphia Phillies,Citizens Bank Park,3565718,102,101,PHI,PHI,PHI
+2012,NL,PIT,PIT,C,4,162,81,79,83,N,N,N,N,651,5412,1313,241,37,170,444,1354,73,52,51,45,674,615,3.9100000000,2,1,45,4300,1357,153,490,1192,112,126,0.98,Pittsburgh Pirates,PNC Park,2091918,93,94,PIT,PIT,PIT
+2012,NL,SDN,SDP,W,4,162,81,76,86,N,N,N,N,651,5422,1339,272,43,121,539,1238,155,46,54,34,710,640,4.0100000000,4,3,43,4304,1356,162,539,1205,121,97,0.98,San Diego Padres,Petco Park,2123721,92,92,SDP,SDN,SDN
+2012,NL,SFN,SFG,W,1,162,81,94,68,Y,N,Y,Y,718,5558,1495,287,57,103,483,1097,118,39,29,61,649,593,3.6800000000,5,4,53,4353,1361,142,489,1237,115,134,0.98,San Francisco Giants,AT&T Park,3377371,88,88,SFG,SFN,SFN
+2012,NL,SLN,STL,C,2,162,81,88,74,N,Y,N,N,765,5622,1526,290,37,159,533,1192,91,37,53,49,648,603,3.7100000000,4,2,42,4388,1420,134,436,1218,107,149,0.98,St. Louis Cardinals,Busch Stadium III,3262109,98,97,STL,SLN,SLN
+2012,NL,WAS,WSN,E,1,162,81,98,64,Y,N,N,N,731,5615,1468,301,25,194,479,1325,105,35,41,36,594,543,3.3400000000,3,1,51,4405,1296,129,497,1325,94,134,0.99,Washington Nationals,Nationals Park,2370794,99,101,WSN,MON,WAS
diff --git a/pandas/pandas/tests/io/json/data/tsframe_iso_v012.json b/pandas/pandas/tests/io/json/data/tsframe_iso_v012.json
new file mode 100644
index 0000000000000..5fa01d5cd902d
--- /dev/null
+++ b/pandas/pandas/tests/io/json/data/tsframe_iso_v012.json
@@ -0,0 +1 @@
+{"A":{"2000-01-03T00:00:00":1.56808523,"2000-01-04T00:00:00":-0.2550111,"2000-01-05T00:00:00":1.51493992,"2000-01-06T00:00:00":-0.02765498,"2000-01-07T00:00:00":0.05951614},"B":{"2000-01-03T00:00:00":0.65727391,"2000-01-04T00:00:00":-0.08072427,"2000-01-05T00:00:00":0.11805825,"2000-01-06T00:00:00":0.44679743,"2000-01-07T00:00:00":-2.69652057},"C":{"2000-01-03T00:00:00":1.81021139,"2000-01-04T00:00:00":-0.03202878,"2000-01-05T00:00:00":1.629455,"2000-01-06T00:00:00":0.33192641,"2000-01-07T00:00:00":1.28163262},"D":{"2000-01-03T00:00:00":-0.17251653,"2000-01-04T00:00:00":-0.17581665,"2000-01-05T00:00:00":-1.31506612,"2000-01-06T00:00:00":-0.27885413,"2000-01-07T00:00:00":0.34703478},"date":{"2000-01-03T00:00:00":"1992-01-06T18:21:32.120000","2000-01-04T00:00:00":"1992-01-06T18:21:32.120000","2000-01-05T00:00:00":"1992-01-06T18:21:32.120000","2000-01-06T00:00:00":"2013-01-01T00:00:00","2000-01-07T00:00:00":"1992-01-06T18:21:32.120000"}}
diff --git a/pandas/pandas/tests/io/json/data/tsframe_v012.json b/pandas/pandas/tests/io/json/data/tsframe_v012.json
new file mode 100644
index 0000000000000..1d6a0a45c028e
--- /dev/null
+++ b/pandas/pandas/tests/io/json/data/tsframe_v012.json
@@ -0,0 +1 @@
+{"A":{"946857600000000000":1.56808523,"946944000000000000":-0.2550111,"947030400000000000":1.51493992,"947116800000000000":-0.02765498,"947203200000000000":0.05951614},"B":{"946857600000000000":0.65727391,"946944000000000000":-0.08072427,"947030400000000000":0.11805825,"947116800000000000":0.44679743,"947203200000000000":-2.69652057},"C":{"946857600000000000":1.81021139,"946944000000000000":-0.03202878,"947030400000000000":1.629455,"947116800000000000":0.33192641,"947203200000000000":1.28163262},"D":{"946857600000000000":-0.17251653,"946944000000000000":-0.17581665,"947030400000000000":-1.31506612,"947116800000000000":-0.27885413,"947203200000000000":0.34703478},"date":{"946857600000000000":694722092120000000,"946944000000000000":694722092120000000,"947030400000000000":694722092120000000,"947116800000000000":1356998400000000000,"947203200000000000":694722092120000000},"modified":{"946857600000000000":694722092120000000,"946944000000000000":null,"947030400000000000":694722092120000000,"947116800000000000":1356998400000000000,"947203200000000000":694722092120000000}}
diff --git a/pandas/pandas/tests/io/json/test_compression.py b/pandas/pandas/tests/io/json/test_compression.py
new file mode 100644
index 0000000000000..953a9246da1cd
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_compression.py
@@ -0,0 +1,131 @@
+from io import (
+    BytesIO,
+    StringIO,
+)
+
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_compression_roundtrip(compression):
+    df = pd.DataFrame(
+        [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+        index=["A", "B"],
+        columns=["X", "Y", "Z"],
+    )
+
+    with tm.ensure_clean() as path:
+        df.to_json(path, compression=compression)
+        tm.assert_frame_equal(df, pd.read_json(path, compression=compression))
+
+        # explicitly ensure file was compressed.
+        with tm.decompress_file(path, compression) as fh:
+            result = fh.read().decode("utf8")
+            data = StringIO(result)
+        tm.assert_frame_equal(df, pd.read_json(data))
+
+
+def test_read_zipped_json(datapath):
+    uncompressed_path = datapath("io", "json", "data", "tsframe_v012.json")
+    uncompressed_df = pd.read_json(uncompressed_path)
+
+    compressed_path = datapath("io", "json", "data", "tsframe_v012.json.zip")
+    compressed_df = pd.read_json(compressed_path, compression="zip")
+
+    tm.assert_frame_equal(uncompressed_df, compressed_df)
+
+
+@td.skip_if_not_us_locale
+@pytest.mark.single_cpu
+@pytest.mark.network
+def test_with_s3_url(compression, s3_public_bucket, s3so):
+    # Bucket created in tests/io/conftest.py
+    df = pd.read_json(StringIO('{"a": [1, 2, 3], "b": [4, 5, 6]}'))
+
+    with tm.ensure_clean() as path:
+        df.to_json(path, compression=compression)
+        with open(path, "rb") as f:
+            s3_public_bucket.put_object(Key="test-1", Body=f)
+
+    roundtripped_df = pd.read_json(
+        f"s3://{s3_public_bucket.name}/test-1",
+        compression=compression,
+        storage_options=s3so,
+    )
+    tm.assert_frame_equal(df, roundtripped_df)
+
+
+def test_lines_with_compression(compression):
+    with tm.ensure_clean() as path:
+        df = pd.read_json(StringIO('{"a": [1, 2, 3], "b": [4, 5, 6]}'))
+        df.to_json(path, orient="records", lines=True, compression=compression)
+        roundtripped_df = pd.read_json(path, lines=True, compression=compression)
+        tm.assert_frame_equal(df, roundtripped_df)
+
+
+def test_chunksize_with_compression(compression):
+    with tm.ensure_clean() as path:
+        df = pd.read_json(StringIO('{"a": ["foo", "bar", "baz"], "b": [4, 5, 6]}'))
+        df.to_json(path, orient="records", lines=True, compression=compression)
+
+        with pd.read_json(
+            path, lines=True, chunksize=1, compression=compression
+        ) as res:
+            roundtripped_df = pd.concat(res)
+        tm.assert_frame_equal(df, roundtripped_df)
+
+
+def test_write_unsupported_compression_type():
+    df = pd.read_json(StringIO('{"a": [1, 2, 3], "b": [4, 5, 6]}'))
+    with tm.ensure_clean() as path:
+        msg = "Unrecognized compression type: unsupported"
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(path, compression="unsupported")
+
+
+def test_read_unsupported_compression_type():
+    with tm.ensure_clean() as path:
+        msg = "Unrecognized compression type: unsupported"
+        with pytest.raises(ValueError, match=msg):
+            pd.read_json(path, compression="unsupported")
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+@pytest.mark.parametrize("to_infer", [True, False])
+@pytest.mark.parametrize("read_infer", [True, False])
+def test_to_json_compression(
+    compression_only, read_infer, to_infer, compression_to_extension, infer_string
+):
+    with pd.option_context("future.infer_string", infer_string):
+        # see gh-15008
+        compression = compression_only
+
+        # We'll complete file extension subsequently.
+        filename = "test."
+        filename += compression_to_extension[compression]
+
+        df = pd.DataFrame({"A": [1]})
+
+        to_compression = "infer" if to_infer else compression
+        read_compression = "infer" if read_infer else compression
+
+        with tm.ensure_clean(filename) as path:
+            df.to_json(path, compression=to_compression)
+            result = pd.read_json(path, compression=read_compression)
+            tm.assert_frame_equal(result, df)
+
+
+def test_to_json_compression_mode(compression):
+    # GH 39985 (read_json does not support user-provided binary files)
+    expected = pd.DataFrame({"A": [1]})
+
+    with BytesIO() as buffer:
+        expected.to_json(buffer, compression=compression)
+        # df = pd.read_json(buffer, compression=compression)
+        # tm.assert_frame_equal(expected, df)
diff --git a/pandas/pandas/tests/io/json/test_deprecated_kwargs.py b/pandas/pandas/tests/io/json/test_deprecated_kwargs.py
new file mode 100644
index 0000000000000..9da682c90a285
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_deprecated_kwargs.py
@@ -0,0 +1,22 @@
+"""
+Tests for the deprecated keyword arguments for `read_json`.
+"""
+
+from io import StringIO
+
+import pandas as pd
+import pandas._testing as tm
+
+from pandas.io.json import read_json
+
+
+def test_good_kwargs():
+    df = pd.DataFrame({"A": [2, 4, 6], "B": [3, 6, 9]}, index=[0, 1, 2])
+
+    with tm.assert_produces_warning(None):
+        data1 = StringIO(df.to_json(orient="split"))
+        tm.assert_frame_equal(df, read_json(data1, orient="split"))
+        data2 = StringIO(df.to_json(orient="columns"))
+        tm.assert_frame_equal(df, read_json(data2, orient="columns"))
+        data3 = StringIO(df.to_json(orient="index"))
+        tm.assert_frame_equal(df, read_json(data3, orient="index"))
diff --git a/pandas/pandas/tests/io/json/test_json_table_schema.py b/pandas/pandas/tests/io/json/test_json_table_schema.py
new file mode 100644
index 0000000000000..7936982e4a055
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_json_table_schema.py
@@ -0,0 +1,883 @@
+"""Tests for Table Schema integration."""
+
+from collections import OrderedDict
+from io import StringIO
+import json
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+    PeriodDtype,
+)
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+
+from pandas.io.json._table_schema import (
+    as_json_table_type,
+    build_table_schema,
+    convert_json_field_to_pandas_type,
+    convert_pandas_type_to_json_field,
+    set_default_names,
+)
+
+
+@pytest.fixture
+def df_schema():
+    return DataFrame(
+        {
+            "A": [1, 2, 3, 4],
+            "B": ["a", "b", "c", "c"],
+            "C": pd.date_range("2016-01-01", freq="D", periods=4),
+            "D": pd.timedelta_range("1h", periods=4, freq="min"),
+        },
+        index=pd.Index(range(4), name="idx"),
+    )
+
+
+@pytest.fixture
+def df_table():
+    return DataFrame(
+        {
+            "A": [1, 2, 3, 4],
+            "B": ["a", "b", "c", "c"],
+            "C": pd.date_range("2016-01-01", freq="D", periods=4),
+            "D": pd.timedelta_range("1h", periods=4, freq="min"),
+            "E": pd.Series(pd.Categorical(["a", "b", "c", "c"])),
+            "F": pd.Series(pd.Categorical(["a", "b", "c", "c"], ordered=True)),
+            "G": [1.0, 2.0, 3, 4.0],
+            "H": pd.date_range("2016-01-01", freq="D", periods=4, tz="US/Central"),
+        },
+        index=pd.Index(range(4), name="idx"),
+    )
+
+
+class TestBuildSchema:
+    def test_build_table_schema(self, df_schema, using_infer_string):
+        result = build_table_schema(df_schema, version=False)
+        expected = {
+            "fields": [
+                {"name": "idx", "type": "integer"},
+                {"name": "A", "type": "integer"},
+                {"name": "B", "type": "string"},
+                {"name": "C", "type": "datetime"},
+                {"name": "D", "type": "duration"},
+            ],
+            "primaryKey": ["idx"],
+        }
+        if using_infer_string:
+            expected["fields"][2] = {"name": "B", "type": "any", "extDtype": "str"}
+        assert result == expected
+        result = build_table_schema(df_schema)
+        assert "pandas_version" in result
+
+    def test_series(self):
+        s = pd.Series([1, 2, 3], name="foo")
+        result = build_table_schema(s, version=False)
+        expected = {
+            "fields": [
+                {"name": "index", "type": "integer"},
+                {"name": "foo", "type": "integer"},
+            ],
+            "primaryKey": ["index"],
+        }
+        assert result == expected
+        result = build_table_schema(s)
+        assert "pandas_version" in result
+
+    def test_series_unnamed(self):
+        result = build_table_schema(pd.Series([1, 2, 3]), version=False)
+        expected = {
+            "fields": [
+                {"name": "index", "type": "integer"},
+                {"name": "values", "type": "integer"},
+            ],
+            "primaryKey": ["index"],
+        }
+        assert result == expected
+
+    def test_multiindex(self, df_schema, using_infer_string):
+        df = df_schema
+        idx = pd.MultiIndex.from_product([("a", "b"), (1, 2)])
+        df.index = idx
+
+        result = build_table_schema(df, version=False)
+        expected = {
+            "fields": [
+                {"name": "level_0", "type": "string"},
+                {"name": "level_1", "type": "integer"},
+                {"name": "A", "type": "integer"},
+                {"name": "B", "type": "string"},
+                {"name": "C", "type": "datetime"},
+                {"name": "D", "type": "duration"},
+            ],
+            "primaryKey": ["level_0", "level_1"],
+        }
+        if using_infer_string:
+            expected["fields"][0] = {
+                "name": "level_0",
+                "type": "any",
+                "extDtype": "str",
+            }
+            expected["fields"][3] = {"name": "B", "type": "any", "extDtype": "str"}
+        assert result == expected
+
+        df.index.names = ["idx0", None]
+        expected["fields"][0]["name"] = "idx0"
+        expected["primaryKey"] = ["idx0", "level_1"]
+        result = build_table_schema(df, version=False)
+        assert result == expected
+
+
+class TestTableSchemaType:
+    @pytest.mark.parametrize("int_type", [int, np.int16, np.int32, np.int64])
+    def test_as_json_table_type_int_data(self, int_type):
+        int_data = [1, 2, 3]
+        assert as_json_table_type(np.array(int_data, dtype=int_type).dtype) == "integer"
+
+    @pytest.mark.parametrize("float_type", [float, np.float16, np.float32, np.float64])
+    def test_as_json_table_type_float_data(self, float_type):
+        float_data = [1.0, 2.0, 3.0]
+        assert (
+            as_json_table_type(np.array(float_data, dtype=float_type).dtype) == "number"
+        )
+
+    @pytest.mark.parametrize("bool_type", [bool, np.bool_])
+    def test_as_json_table_type_bool_data(self, bool_type):
+        bool_data = [True, False]
+        assert (
+            as_json_table_type(np.array(bool_data, dtype=bool_type).dtype) == "boolean"
+        )
+
+    @pytest.mark.parametrize(
+        "date_data",
+        [
+            pd.to_datetime(["2016"]),
+            pd.to_datetime(["2016"], utc=True),
+            pd.Series(pd.to_datetime(["2016"])),
+            pd.Series(pd.to_datetime(["2016"], utc=True)),
+            pd.period_range("2016", freq="Y", periods=3),
+        ],
+    )
+    def test_as_json_table_type_date_data(self, date_data):
+        assert as_json_table_type(date_data.dtype) == "datetime"
+
+    @pytest.mark.parametrize("klass", [pd.Series, pd.Index])
+    def test_as_json_table_type_string_data(self, klass):
+        str_data = klass(["a", "b"], dtype=object)
+        assert as_json_table_type(str_data.dtype) == "string"
+
+    @pytest.mark.parametrize(
+        "cat_data",
+        [
+            pd.Categorical(["a"]),
+            pd.Categorical([1]),
+            pd.Series(pd.Categorical([1])),
+            pd.CategoricalIndex([1]),
+        ],
+    )
+    def test_as_json_table_type_categorical_data(self, cat_data):
+        assert as_json_table_type(cat_data.dtype) == "any"
+
+    # ------
+    # dtypes
+    # ------
+    @pytest.mark.parametrize("int_dtype", [int, np.int16, np.int32, np.int64])
+    def test_as_json_table_type_int_dtypes(self, int_dtype):
+        assert as_json_table_type(int_dtype) == "integer"
+
+    @pytest.mark.parametrize("float_dtype", [float, np.float16, np.float32, np.float64])
+    def test_as_json_table_type_float_dtypes(self, float_dtype):
+        assert as_json_table_type(float_dtype) == "number"
+
+    @pytest.mark.parametrize("bool_dtype", [bool, np.bool_])
+    def test_as_json_table_type_bool_dtypes(self, bool_dtype):
+        assert as_json_table_type(bool_dtype) == "boolean"
+
+    @pytest.mark.parametrize(
+        "date_dtype",
+        [
+            np.dtype("<M8[ns]"),
+            PeriodDtype("D"),
+            DatetimeTZDtype("ns", "US/Central"),
+        ],
+    )
+    def test_as_json_table_type_date_dtypes(self, date_dtype):
+        # TODO: datedate.date? datetime.time?
+        assert as_json_table_type(date_dtype) == "datetime"
+
+    @pytest.mark.parametrize("td_dtype", [np.dtype("<m8[ns]")])
+    def test_as_json_table_type_timedelta_dtypes(self, td_dtype):
+        assert as_json_table_type(td_dtype) == "duration"
+
+    @pytest.mark.parametrize("str_dtype", [object])  # TODO(GH#14904) flesh out dtypes?
+    def test_as_json_table_type_string_dtypes(self, str_dtype):
+        assert as_json_table_type(str_dtype) == "string"
+
+    def test_as_json_table_type_categorical_dtypes(self):
+        assert as_json_table_type(pd.Categorical(["a"]).dtype) == "any"
+        assert as_json_table_type(CategoricalDtype()) == "any"
+
+
+class TestTableOrient:
+    def test_build_series(self):
+        s = pd.Series([1, 2], name="a")
+        s.index.name = "id"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [{"name": "id", "type": "integer"}, {"name": "a", "type": "integer"}]
+
+        schema = {"fields": fields, "primaryKey": ["id"]}
+
+        expected = OrderedDict(
+            [
+                ("schema", schema),
+                (
+                    "data",
+                    [
+                        OrderedDict([("id", 0), ("a", 1)]),
+                        OrderedDict([("id", 1), ("a", 2)]),
+                    ],
+                ),
+            ]
+        )
+
+        assert result == expected
+
+    def test_read_json_from_to_json_results(self):
+        # GH32383
+        df = DataFrame(
+            {
+                "_id": {"row_0": 0},
+                "category": {"row_0": "Goods"},
+                "recommender_id": {"row_0": 3},
+                "recommender_name_jp": {"row_0": "浦田"},
+                "recommender_name_en": {"row_0": "Urata"},
+                "name_jp": {"row_0": "博多人形(松尾吉将まつお よしまさ)"},
+                "name_en": {"row_0": "Hakata Dolls Matsuo"},
+            }
+        )
+
+        result1 = pd.read_json(StringIO(df.to_json()))
+        result2 = DataFrame.from_dict(json.loads(df.to_json()))
+        tm.assert_frame_equal(result1, df)
+        tm.assert_frame_equal(result2, df)
+
+    def test_to_json(self, df_table, using_infer_string):
+        df = df_table
+        df.index.name = "idx"
+        result = df.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "idx", "type": "integer"},
+            {"name": "A", "type": "integer"},
+            {"name": "B", "type": "string"},
+            {"name": "C", "type": "datetime"},
+            {"name": "D", "type": "duration"},
+            {
+                "constraints": {"enum": ["a", "b", "c"]},
+                "name": "E",
+                "ordered": False,
+                "type": "any",
+            },
+            {
+                "constraints": {"enum": ["a", "b", "c"]},
+                "name": "F",
+                "ordered": True,
+                "type": "any",
+            },
+            {"name": "G", "type": "number"},
+            {"name": "H", "type": "datetime", "tz": "US/Central"},
+        ]
+
+        if using_infer_string:
+            fields[2] = {"name": "B", "type": "any", "extDtype": "str"}
+
+        schema = {"fields": fields, "primaryKey": ["idx"]}
+        data = [
+            OrderedDict(
+                [
+                    ("idx", 0),
+                    ("A", 1),
+                    ("B", "a"),
+                    ("C", "2016-01-01T00:00:00.000"),
+                    ("D", "P0DT1H0M0S"),
+                    ("E", "a"),
+                    ("F", "a"),
+                    ("G", 1.0),
+                    ("H", "2016-01-01T06:00:00.000Z"),
+                ]
+            ),
+            OrderedDict(
+                [
+                    ("idx", 1),
+                    ("A", 2),
+                    ("B", "b"),
+                    ("C", "2016-01-02T00:00:00.000"),
+                    ("D", "P0DT1H1M0S"),
+                    ("E", "b"),
+                    ("F", "b"),
+                    ("G", 2.0),
+                    ("H", "2016-01-02T06:00:00.000Z"),
+                ]
+            ),
+            OrderedDict(
+                [
+                    ("idx", 2),
+                    ("A", 3),
+                    ("B", "c"),
+                    ("C", "2016-01-03T00:00:00.000"),
+                    ("D", "P0DT1H2M0S"),
+                    ("E", "c"),
+                    ("F", "c"),
+                    ("G", 3.0),
+                    ("H", "2016-01-03T06:00:00.000Z"),
+                ]
+            ),
+            OrderedDict(
+                [
+                    ("idx", 3),
+                    ("A", 4),
+                    ("B", "c"),
+                    ("C", "2016-01-04T00:00:00.000"),
+                    ("D", "P0DT1H3M0S"),
+                    ("E", "c"),
+                    ("F", "c"),
+                    ("G", 4.0),
+                    ("H", "2016-01-04T06:00:00.000Z"),
+                ]
+            ),
+        ]
+        expected = OrderedDict([("schema", schema), ("data", data)])
+
+        assert result == expected
+
+    def test_to_json_float_index(self):
+        data = pd.Series(1, index=[1.0, 2.0])
+        result = data.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+        result["schema"].pop("pandas_version")
+
+        expected = OrderedDict(
+            [
+                (
+                    "schema",
+                    {
+                        "fields": [
+                            {"name": "index", "type": "number"},
+                            {"name": "values", "type": "integer"},
+                        ],
+                        "primaryKey": ["index"],
+                    },
+                ),
+                (
+                    "data",
+                    [
+                        OrderedDict([("index", 1.0), ("values", 1)]),
+                        OrderedDict([("index", 2.0), ("values", 1)]),
+                    ],
+                ),
+            ]
+        )
+
+        assert result == expected
+
+    def test_to_json_period_index(self):
+        idx = pd.period_range("2016", freq="Q-JAN", periods=2)
+        data = pd.Series(1, idx)
+        result = data.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"freq": "QE-JAN", "name": "index", "type": "datetime"},
+            {"name": "values", "type": "integer"},
+        ]
+
+        schema = {"fields": fields, "primaryKey": ["index"]}
+        data = [
+            OrderedDict([("index", "2015-11-01T00:00:00.000"), ("values", 1)]),
+            OrderedDict([("index", "2016-02-01T00:00:00.000"), ("values", 1)]),
+        ]
+        expected = OrderedDict([("schema", schema), ("data", data)])
+
+        assert result == expected
+
+    def test_to_json_categorical_index(self):
+        data = pd.Series(1, pd.CategoricalIndex(["a", "b"]))
+        result = data.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+        result["schema"].pop("pandas_version")
+
+        expected = OrderedDict(
+            [
+                (
+                    "schema",
+                    {
+                        "fields": [
+                            {
+                                "name": "index",
+                                "type": "any",
+                                "constraints": {"enum": ["a", "b"]},
+                                "ordered": False,
+                            },
+                            {"name": "values", "type": "integer"},
+                        ],
+                        "primaryKey": ["index"],
+                    },
+                ),
+                (
+                    "data",
+                    [
+                        OrderedDict([("index", "a"), ("values", 1)]),
+                        OrderedDict([("index", "b"), ("values", 1)]),
+                    ],
+                ),
+            ]
+        )
+
+        assert result == expected
+
+    def test_date_format_raises(self, df_table):
+        error_msg = (
+            "Trying to write with `orient='table'` and `date_format='epoch'`. Table "
+            "Schema requires dates to be formatted with `date_format='iso'`"
+        )
+        warning_msg = (
+            "'epoch' date format is deprecated and will be removed in a future "
+            "version, please use 'iso' date format instead."
+        )
+        with pytest.raises(ValueError, match=error_msg):
+            with tm.assert_produces_warning(FutureWarning, match=warning_msg):
+                df_table.to_json(orient="table", date_format="epoch")
+
+        # others work
+        df_table.to_json(orient="table", date_format="iso")
+        df_table.to_json(orient="table")
+
+    def test_convert_pandas_type_to_json_field_int(self, index_or_series):
+        kind = index_or_series
+        data = [1, 2, 3]
+        result = convert_pandas_type_to_json_field(kind(data, name="name"))
+        expected = {"name": "name", "type": "integer"}
+        assert result == expected
+
+    def test_convert_pandas_type_to_json_field_float(self, index_or_series):
+        kind = index_or_series
+        data = [1.0, 2.0, 3.0]
+        result = convert_pandas_type_to_json_field(kind(data, name="name"))
+        expected = {"name": "name", "type": "number"}
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "dt_args,extra_exp", [({}, {}), ({"utc": True}, {"tz": "UTC"})]
+    )
+    @pytest.mark.parametrize("wrapper", [None, pd.Series])
+    def test_convert_pandas_type_to_json_field_datetime(
+        self, dt_args, extra_exp, wrapper
+    ):
+        data = [1.0, 2.0, 3.0]
+        data = pd.to_datetime(data, **dt_args)
+        if wrapper is pd.Series:
+            data = pd.Series(data, name="values")
+        result = convert_pandas_type_to_json_field(data)
+        expected = {"name": "values", "type": "datetime"}
+        expected.update(extra_exp)
+        assert result == expected
+
+    def test_convert_pandas_type_to_json_period_range(self):
+        arr = pd.period_range("2016", freq="Y-DEC", periods=4)
+        result = convert_pandas_type_to_json_field(arr)
+        expected = {"name": "values", "type": "datetime", "freq": "YE-DEC"}
+        assert result == expected
+
+    @pytest.mark.parametrize("kind", [pd.Categorical, pd.CategoricalIndex])
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_convert_pandas_type_to_json_field_categorical(self, kind, ordered):
+        data = ["a", "b", "c"]
+        if kind is pd.Categorical:
+            arr = pd.Series(kind(data, ordered=ordered), name="cats")
+        elif kind is pd.CategoricalIndex:
+            arr = kind(data, ordered=ordered, name="cats")
+
+        result = convert_pandas_type_to_json_field(arr)
+        expected = {
+            "name": "cats",
+            "type": "any",
+            "constraints": {"enum": data},
+            "ordered": ordered,
+        }
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "inp,exp",
+        [
+            ({"type": "integer"}, "int64"),
+            ({"type": "number"}, "float64"),
+            ({"type": "boolean"}, "bool"),
+            ({"type": "duration"}, "timedelta64"),
+            ({"type": "datetime"}, "datetime64[ns]"),
+            ({"type": "datetime", "tz": "US/Hawaii"}, "datetime64[ns, US/Hawaii]"),
+            ({"type": "any"}, "object"),
+            (
+                {
+                    "type": "any",
+                    "constraints": {"enum": ["a", "b", "c"]},
+                    "ordered": False,
+                },
+                CategoricalDtype(categories=["a", "b", "c"], ordered=False),
+            ),
+            (
+                {
+                    "type": "any",
+                    "constraints": {"enum": ["a", "b", "c"]},
+                    "ordered": True,
+                },
+                CategoricalDtype(categories=["a", "b", "c"], ordered=True),
+            ),
+            ({"type": "string"}, "object"),
+        ],
+    )
+    def test_convert_json_field_to_pandas_type(self, inp, exp):
+        field = {"name": "foo"}
+        field.update(inp)
+        assert convert_json_field_to_pandas_type(field) == exp
+
+    @pytest.mark.parametrize("inp", ["geopoint", "geojson", "fake_type"])
+    def test_convert_json_field_to_pandas_type_raises(self, inp):
+        field = {"type": inp}
+        with pytest.raises(
+            ValueError, match=f"Unsupported or invalid field type: {inp}"
+        ):
+            convert_json_field_to_pandas_type(field)
+
+    def test_categorical(self):
+        s = pd.Series(pd.Categorical(["a", "b", "a"]))
+        s.index.name = "idx"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "idx", "type": "integer"},
+            {
+                "constraints": {"enum": ["a", "b"]},
+                "name": "values",
+                "ordered": False,
+                "type": "any",
+            },
+        ]
+
+        expected = OrderedDict(
+            [
+                ("schema", {"fields": fields, "primaryKey": ["idx"]}),
+                (
+                    "data",
+                    [
+                        OrderedDict([("idx", 0), ("values", "a")]),
+                        OrderedDict([("idx", 1), ("values", "b")]),
+                        OrderedDict([("idx", 2), ("values", "a")]),
+                    ],
+                ),
+            ]
+        )
+
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "idx,nm,prop",
+        [
+            (pd.Index([1]), "index", "name"),
+            (pd.Index([1], name="myname"), "myname", "name"),
+            (
+                pd.MultiIndex.from_product([("a", "b"), ("c", "d")]),
+                ["level_0", "level_1"],
+                "names",
+            ),
+            (
+                pd.MultiIndex.from_product(
+                    [("a", "b"), ("c", "d")], names=["n1", "n2"]
+                ),
+                ["n1", "n2"],
+                "names",
+            ),
+            (
+                pd.MultiIndex.from_product(
+                    [("a", "b"), ("c", "d")], names=["n1", None]
+                ),
+                ["n1", "level_1"],
+                "names",
+            ),
+        ],
+    )
+    def test_set_names_unset(self, idx, nm, prop):
+        data = pd.Series(1, idx)
+        result = set_default_names(data)
+        assert getattr(result.index, prop) == nm
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            pd.Index([], name="index"),
+            pd.MultiIndex.from_arrays([["foo"], ["bar"]], names=("level_0", "level_1")),
+            pd.MultiIndex.from_arrays([["foo"], ["bar"]], names=("foo", "level_1")),
+        ],
+    )
+    def test_warns_non_roundtrippable_names(self, idx):
+        # GH 19130
+        df = DataFrame(index=idx)
+        df.index.name = "index"
+        with tm.assert_produces_warning(UserWarning, match="not round-trippable"):
+            set_default_names(df)
+
+    def test_timestamp_in_columns(self):
+        df = DataFrame(
+            [[1, 2]], columns=[pd.Timestamp("2016"), pd.Timedelta(10, unit="s")]
+        )
+        result = df.to_json(orient="table")
+        js = json.loads(result)
+        assert js["schema"]["fields"][1]["name"] == "2016-01-01T00:00:00.000"
+        assert js["schema"]["fields"][2]["name"] == "P0DT0H0M10S"
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            pd.Series([1], index=pd.Index([1], name="a"), name="a"),
+            DataFrame({"A": [1]}, index=pd.Index([1], name="A")),
+            DataFrame(
+                {"A": [1]},
+                index=pd.MultiIndex.from_arrays([["a"], [1]], names=["A", "a"]),
+            ),
+        ],
+    )
+    def test_overlapping_names(self, case):
+        with pytest.raises(ValueError, match="Overlapping"):
+            case.to_json(orient="table")
+
+    def test_mi_falsey_name(self):
+        # GH 16203
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            index=pd.MultiIndex.from_product([("A", "B"), ("a", "b")]),
+        )
+        result = [x["name"] for x in build_table_schema(df)["fields"]]
+        assert result == ["level_0", "level_1", 0, 1, 2, 3]
+
+
+class TestTableOrientReader:
+    @pytest.mark.parametrize(
+        "index_nm",
+        [None, "idx", pytest.param("index", marks=pytest.mark.xfail), "level_0"],
+    )
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            {"ints": [1, 2, 3, 4]},
+            {"objects": ["a", "b", "c", "d"]},
+            {"objects": ["1", "2", "3", "4"]},
+            {"date_ranges": pd.date_range("2016-01-01", freq="D", periods=4)},
+            {"categoricals": pd.Series(pd.Categorical(["a", "b", "c", "c"]))},
+            {
+                "ordered_cats": pd.Series(
+                    pd.Categorical(["a", "b", "c", "c"], ordered=True)
+                )
+            },
+            {"floats": [1.0, 2.0, 3.0, 4.0]},
+            {"floats": [1.1, 2.2, 3.3, 4.4]},
+            {"bools": [True, False, False, True]},
+            {
+                "timezones": pd.date_range(
+                    "2016-01-01", freq="D", periods=4, tz="US/Central"
+                )  # added in # GH 35973
+            },
+        ],
+    )
+    def test_read_json_table_orient(self, index_nm, vals):
+        df = DataFrame(vals, index=pd.Index(range(4), name=index_nm))
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
+
+    @pytest.mark.parametrize(
+        "index_nm",
+        [
+            None,
+            "idx",
+            pytest.param(
+                "index",
+                marks=pytest.mark.filterwarnings("ignore:Index name:UserWarning"),
+            ),
+        ],
+    )
+    def test_read_json_table_orient_raises(self, index_nm):
+        vals = {"timedeltas": pd.timedelta_range("1h", periods=4, freq="min")}
+        df = DataFrame(vals, index=pd.Index(range(4), name=index_nm))
+        out = StringIO(df.to_json(orient="table"))
+        with pytest.raises(NotImplementedError, match="can not yet read "):
+            pd.read_json(out, orient="table")
+
+    @pytest.mark.parametrize(
+        "index_nm",
+        [None, "idx", pytest.param("index", marks=pytest.mark.xfail), "level_0"],
+    )
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            {"ints": [1, 2, 3, 4]},
+            {"objects": ["a", "b", "c", "d"]},
+            {"objects": ["1", "2", "3", "4"]},
+            {"date_ranges": pd.date_range("2016-01-01", freq="D", periods=4)},
+            {"categoricals": pd.Series(pd.Categorical(["a", "b", "c", "c"]))},
+            {
+                "ordered_cats": pd.Series(
+                    pd.Categorical(["a", "b", "c", "c"], ordered=True)
+                )
+            },
+            {"floats": [1.0, 2.0, 3.0, 4.0]},
+            {"floats": [1.1, 2.2, 3.3, 4.4]},
+            {"bools": [True, False, False, True]},
+            {
+                "timezones": pd.date_range(
+                    "2016-01-01", freq="D", periods=4, tz="US/Central"
+                )  # added in # GH 35973
+            },
+        ],
+    )
+    def test_read_json_table_period_orient(self, index_nm, vals):
+        df = DataFrame(
+            vals,
+            index=pd.Index(
+                (pd.Period(f"2022Q{q}") for q in range(1, 5)), name=index_nm
+            ),
+        )
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            pd.Index(range(4)),
+            pd.date_range(
+                "2020-08-30",
+                freq="D",
+                periods=4,
+            )._with_freq(None),
+            pd.date_range(
+                "2020-08-30", freq="D", periods=4, tz="US/Central"
+            )._with_freq(None),
+            pd.MultiIndex.from_product(
+                [
+                    pd.date_range("2020-08-30", freq="D", periods=2, tz="US/Central"),
+                    ["x", "y"],
+                ],
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "vals",
+        [
+            {"floats": [1.1, 2.2, 3.3, 4.4]},
+            {"dates": pd.date_range("2020-08-30", freq="D", periods=4)},
+            {
+                "timezones": pd.date_range(
+                    "2020-08-30", freq="D", periods=4, tz="Europe/London"
+                )
+            },
+        ],
+    )
+    def test_read_json_table_timezones_orient(self, idx, vals):
+        # GH 35973
+        df = DataFrame(vals, index=idx)
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
+
+    def test_comprehensive(self):
+        df = DataFrame(
+            {
+                "A": [1, 2, 3, 4],
+                "B": ["a", "b", "c", "c"],
+                "C": pd.date_range("2016-01-01", freq="D", periods=4),
+                # 'D': pd.timedelta_range('1h', periods=4, freq='min'),
+                "E": pd.Series(pd.Categorical(["a", "b", "c", "c"])),
+                "F": pd.Series(pd.Categorical(["a", "b", "c", "c"], ordered=True)),
+                "G": [1.1, 2.2, 3.3, 4.4],
+                "H": pd.date_range("2016-01-01", freq="D", periods=4, tz="US/Central"),
+                "I": [True, False, False, True],
+            },
+            index=pd.Index(range(4), name="idx"),
+        )
+
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
+
+    @pytest.mark.parametrize(
+        "index_names",
+        [[None, None], ["foo", "bar"], ["foo", None], [None, "foo"], ["index", "foo"]],
+    )
+    def test_multiindex(self, index_names):
+        # GH 18912
+        df = DataFrame(
+            [["Arr", "alpha", [1, 2, 3, 4]], ["Bee", "Beta", [10, 20, 30, 40]]],
+            index=[["A", "B"], ["Null", "Eins"]],
+            columns=["Aussprache", "Griechisch", "Args"],
+        )
+        df.index.names = index_names
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
+
+    def test_empty_frame_roundtrip(self):
+        # GH 21287
+        df = DataFrame(columns=["a", "b", "c"])
+        expected = df.copy()
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(expected, result)
+
+    def test_read_json_orient_table_old_schema_version(self):
+        df_json = """
+        {
+            "schema":{
+                "fields":[
+                    {"name":"index","type":"integer"},
+                    {"name":"a","type":"string"}
+                ],
+                "primaryKey":["index"],
+                "pandas_version":"0.20.0"
+            },
+            "data":[
+                {"index":0,"a":1},
+                {"index":1,"a":2.0},
+                {"index":2,"a":"s"}
+            ]
+        }
+        """
+        expected = DataFrame({"a": [1, 2.0, "s"]})
+        result = pd.read_json(StringIO(df_json), orient="table")
+        tm.assert_frame_equal(expected, result)
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "Q", "2Q", "Y", "2Y"])
+    def test_read_json_table_orient_period_depr_freq(self, freq):
+        # GH#9586
+        df = DataFrame(
+            {"ints": [1, 2]},
+            index=pd.PeriodIndex(["2020-01", "2021-06"], freq=freq),
+        )
+        out = StringIO(df.to_json(orient="table"))
+        result = pd.read_json(out, orient="table")
+        tm.assert_frame_equal(df, result)
diff --git a/pandas/pandas/tests/io/json/test_json_table_schema_ext_dtype.py b/pandas/pandas/tests/io/json/test_json_table_schema_ext_dtype.py
new file mode 100644
index 0000000000000..12ae24b064c9d
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_json_table_schema_ext_dtype.py
@@ -0,0 +1,296 @@
+"""Tests for ExtensionDtype Table Schema integration."""
+
+from collections import OrderedDict
+import datetime as dt
+import decimal
+from io import StringIO
+import json
+
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+    array,
+    read_json,
+)
+import pandas._testing as tm
+from pandas.core.arrays.integer import Int64Dtype
+from pandas.core.arrays.string_ import StringDtype
+from pandas.core.series import Series
+from pandas.tests.extension.date import (
+    DateArray,
+    DateDtype,
+)
+from pandas.tests.extension.decimal.array import (
+    DecimalArray,
+    DecimalDtype,
+)
+
+from pandas.io.json._table_schema import (
+    as_json_table_type,
+    build_table_schema,
+)
+
+
+class TestBuildSchema:
+    def test_build_table_schema(self):
+        df = DataFrame(
+            {
+                "A": DateArray([dt.date(2021, 10, 10)]),
+                "B": DecimalArray([decimal.Decimal(10)]),
+                "C": array(["pandas"], dtype="string"),
+                "D": array([10], dtype="Int64"),
+            }
+        )
+        result = build_table_schema(df, version=False)
+        expected = {
+            "fields": [
+                {"name": "index", "type": "integer"},
+                {"name": "A", "type": "any", "extDtype": "DateDtype"},
+                {"name": "B", "type": "number", "extDtype": "decimal"},
+                {"name": "C", "type": "any", "extDtype": "string"},
+                {"name": "D", "type": "integer", "extDtype": "Int64"},
+            ],
+            "primaryKey": ["index"],
+        }
+        assert result == expected
+        result = build_table_schema(df)
+        assert "pandas_version" in result
+
+
+class TestTableSchemaType:
+    @pytest.mark.parametrize("box", [lambda x: x, Series])
+    def test_as_json_table_type_ext_date_array_dtype(self, box):
+        date_data = box(DateArray([dt.date(2021, 10, 10)]))
+        assert as_json_table_type(date_data.dtype) == "any"
+
+    def test_as_json_table_type_ext_date_dtype(self):
+        assert as_json_table_type(DateDtype()) == "any"
+
+    @pytest.mark.parametrize("box", [lambda x: x, Series])
+    def test_as_json_table_type_ext_decimal_array_dtype(self, box):
+        decimal_data = box(DecimalArray([decimal.Decimal(10)]))
+        assert as_json_table_type(decimal_data.dtype) == "number"
+
+    def test_as_json_table_type_ext_decimal_dtype(self):
+        assert as_json_table_type(DecimalDtype()) == "number"
+
+    @pytest.mark.parametrize("box", [lambda x: x, Series])
+    def test_as_json_table_type_ext_string_array_dtype(self, box):
+        string_data = box(array(["pandas"], dtype="string"))
+        assert as_json_table_type(string_data.dtype) == "any"
+
+    def test_as_json_table_type_ext_string_dtype(self):
+        assert as_json_table_type(StringDtype()) == "any"
+
+    @pytest.mark.parametrize("box", [lambda x: x, Series])
+    def test_as_json_table_type_ext_integer_array_dtype(self, box):
+        integer_data = box(array([10], dtype="Int64"))
+        assert as_json_table_type(integer_data.dtype) == "integer"
+
+    def test_as_json_table_type_ext_integer_dtype(self):
+        assert as_json_table_type(Int64Dtype()) == "integer"
+
+
+class TestTableOrient:
+    @pytest.fixture
+    def da(self):
+        """Fixture for creating a DateArray."""
+        return DateArray([dt.date(2021, 10, 10)])
+
+    @pytest.fixture
+    def dc(self):
+        """Fixture for creating a DecimalArray."""
+        return DecimalArray([decimal.Decimal(10)])
+
+    @pytest.fixture
+    def sa(self):
+        """Fixture for creating a StringDtype array."""
+        return array(["pandas"], dtype="string")
+
+    @pytest.fixture
+    def ia(self):
+        """Fixture for creating an Int64Dtype array."""
+        return array([10], dtype="Int64")
+
+    def test_build_date_series(self, da):
+        s = Series(da, name="a")
+        s.index.name = "id"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "id", "type": "integer"},
+            {"name": "a", "type": "any", "extDtype": "DateDtype"},
+        ]
+
+        schema = {"fields": fields, "primaryKey": ["id"]}
+
+        expected = OrderedDict(
+            [
+                ("schema", schema),
+                ("data", [OrderedDict([("id", 0), ("a", "2021-10-10T00:00:00.000")])]),
+            ]
+        )
+
+        assert result == expected
+
+    def test_build_decimal_series(self, dc):
+        s = Series(dc, name="a")
+        s.index.name = "id"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "id", "type": "integer"},
+            {"name": "a", "type": "number", "extDtype": "decimal"},
+        ]
+
+        schema = {"fields": fields, "primaryKey": ["id"]}
+
+        expected = OrderedDict(
+            [
+                ("schema", schema),
+                ("data", [OrderedDict([("id", 0), ("a", "10")])]),
+            ]
+        )
+
+        assert result == expected
+
+    def test_build_string_series(self, sa):
+        s = Series(sa, name="a")
+        s.index.name = "id"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "id", "type": "integer"},
+            {"name": "a", "type": "any", "extDtype": "string"},
+        ]
+
+        schema = {"fields": fields, "primaryKey": ["id"]}
+
+        expected = OrderedDict(
+            [
+                ("schema", schema),
+                ("data", [OrderedDict([("id", 0), ("a", "pandas")])]),
+            ]
+        )
+
+        assert result == expected
+
+    def test_build_int64_series(self, ia):
+        s = Series(ia, name="a")
+        s.index.name = "id"
+        result = s.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            {"name": "id", "type": "integer"},
+            {"name": "a", "type": "integer", "extDtype": "Int64"},
+        ]
+
+        schema = {"fields": fields, "primaryKey": ["id"]}
+
+        expected = OrderedDict(
+            [
+                ("schema", schema),
+                ("data", [OrderedDict([("id", 0), ("a", 10)])]),
+            ]
+        )
+
+        assert result == expected
+
+    def test_to_json(self, da, dc, sa, ia):
+        df = DataFrame(
+            {
+                "A": da,
+                "B": dc,
+                "C": sa,
+                "D": ia,
+            }
+        )
+        df.index.name = "idx"
+        result = df.to_json(orient="table", date_format="iso")
+        result = json.loads(result, object_pairs_hook=OrderedDict)
+
+        assert "pandas_version" in result["schema"]
+        result["schema"].pop("pandas_version")
+
+        fields = [
+            OrderedDict({"name": "idx", "type": "integer"}),
+            OrderedDict({"name": "A", "type": "any", "extDtype": "DateDtype"}),
+            OrderedDict({"name": "B", "type": "number", "extDtype": "decimal"}),
+            OrderedDict({"name": "C", "type": "any", "extDtype": "string"}),
+            OrderedDict({"name": "D", "type": "integer", "extDtype": "Int64"}),
+        ]
+
+        schema = OrderedDict({"fields": fields, "primaryKey": ["idx"]})
+        data = [
+            OrderedDict(
+                [
+                    ("idx", 0),
+                    ("A", "2021-10-10T00:00:00.000"),
+                    ("B", "10"),
+                    ("C", "pandas"),
+                    ("D", 10),
+                ]
+            )
+        ]
+        expected = OrderedDict([("schema", schema), ("data", data)])
+
+        assert result == expected
+
+    def test_json_ext_dtype_reading_roundtrip(self):
+        # GH#40255
+        df = DataFrame(
+            {
+                "a": Series([2, NA], dtype="Int64"),
+                "b": Series([1.5, NA], dtype="Float64"),
+                "c": Series([True, NA], dtype="boolean"),
+            },
+            index=Index([1, NA], dtype="Int64"),
+        )
+        expected = df.copy()
+        data_json = df.to_json(orient="table", indent=4)
+        result = read_json(StringIO(data_json), orient="table")
+        tm.assert_frame_equal(result, expected)
+
+    def test_json_ext_dtype_reading(self):
+        # GH#40255
+        data_json = """{
+            "schema":{
+                "fields":[
+                    {
+                        "name":"a",
+                        "type":"integer",
+                        "extDtype":"Int64"
+                    }
+                ],
+            },
+            "data":[
+                {
+                    "a":2
+                },
+                {
+                    "a":null
+                }
+            ]
+        }"""
+        result = read_json(StringIO(data_json), orient="table")
+        expected = DataFrame({"a": Series([2, NA], dtype="Int64")})
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/json/test_normalize.py b/pandas/pandas/tests/io/json/test_normalize.py
new file mode 100644
index 0000000000000..fdbfbd004617e
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_normalize.py
@@ -0,0 +1,905 @@
+import json
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    json_normalize,
+)
+import pandas._testing as tm
+
+from pandas.io.json._normalize import nested_to_record
+
+
+@pytest.fixture
+def deep_nested():
+    # deeply nested data
+    return [
+        {
+            "country": "USA",
+            "states": [
+                {
+                    "name": "California",
+                    "cities": [
+                        {"name": "San Francisco", "pop": 12345},
+                        {"name": "Los Angeles", "pop": 12346},
+                    ],
+                },
+                {
+                    "name": "Ohio",
+                    "cities": [
+                        {"name": "Columbus", "pop": 1234},
+                        {"name": "Cleveland", "pop": 1236},
+                    ],
+                },
+            ],
+        },
+        {
+            "country": "Germany",
+            "states": [
+                {"name": "Bayern", "cities": [{"name": "Munich", "pop": 12347}]},
+                {
+                    "name": "Nordrhein-Westfalen",
+                    "cities": [
+                        {"name": "Duesseldorf", "pop": 1238},
+                        {"name": "Koeln", "pop": 1239},
+                    ],
+                },
+            ],
+        },
+    ]
+
+
+@pytest.fixture
+def state_data():
+    return [
+        {
+            "counties": [
+                {"name": "Dade", "population": 12345},
+                {"name": "Broward", "population": 40000},
+                {"name": "Palm Beach", "population": 60000},
+            ],
+            "info": {"governor": "Rick Scott"},
+            "shortname": "FL",
+            "state": "Florida",
+        },
+        {
+            "counties": [
+                {"name": "Summit", "population": 1234},
+                {"name": "Cuyahoga", "population": 1337},
+            ],
+            "info": {"governor": "John Kasich"},
+            "shortname": "OH",
+            "state": "Ohio",
+        },
+    ]
+
+
+@pytest.fixture
+def missing_metadata():
+    return [
+        {
+            "name": "Alice",
+            "addresses": [
+                {
+                    "number": 9562,
+                    "street": "Morris St.",
+                    "city": "Massillon",
+                    "state": "OH",
+                    "zip": 44646,
+                }
+            ],
+            "previous_residences": {"cities": [{"city_name": "Foo York City"}]},
+        },
+        {
+            "addresses": [
+                {
+                    "number": 8449,
+                    "street": "Spring St.",
+                    "city": "Elizabethton",
+                    "state": "TN",
+                    "zip": 37643,
+                }
+            ],
+            "previous_residences": {"cities": [{"city_name": "Barmingham"}]},
+        },
+    ]
+
+
+class TestJSONNormalize:
+    def test_simple_records(self):
+        recs = [
+            {"a": 1, "b": 2, "c": 3},
+            {"a": 4, "b": 5, "c": 6},
+            {"a": 7, "b": 8, "c": 9},
+            {"a": 10, "b": 11, "c": 12},
+        ]
+
+        result = json_normalize(recs)
+        expected = DataFrame(recs)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_simple_normalize(self, state_data):
+        result = json_normalize(state_data[0], "counties")
+        expected = DataFrame(state_data[0]["counties"])
+        tm.assert_frame_equal(result, expected)
+
+        result = json_normalize(state_data, "counties")
+
+        expected = []
+        for rec in state_data:
+            expected.extend(rec["counties"])
+        expected = DataFrame(expected)
+
+        tm.assert_frame_equal(result, expected)
+
+        result = json_normalize(state_data, "counties", meta="state")
+        expected["state"] = np.array(["Florida", "Ohio"]).repeat([3, 2])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_fields_list_type_normalize(self):
+        parse_metadata_fields_list_type = [
+            {"values": [1, 2, 3], "metadata": {"listdata": [1, 2]}}
+        ]
+        result = json_normalize(
+            parse_metadata_fields_list_type,
+            record_path=["values"],
+            meta=[["metadata", "listdata"]],
+        )
+        expected = DataFrame(
+            {0: [1, 2, 3], "metadata.listdata": [[1, 2], [1, 2], [1, 2]]}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_empty_array(self):
+        result = json_normalize([])
+        expected = DataFrame()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data, record_path, exception_type",
+        [
+            ([{"a": 0}, {"a": 1}], None, None),
+            ({"a": [{"a": 0}, {"a": 1}]}, "a", None),
+            ('{"a": [{"a": 0}, {"a": 1}]}', None, NotImplementedError),
+            (None, None, NotImplementedError),
+        ],
+    )
+    def test_accepted_input(self, data, record_path, exception_type):
+        if exception_type is not None:
+            with pytest.raises(exception_type, match=""):
+                json_normalize(data, record_path=record_path)
+        else:
+            result = json_normalize(data, record_path=record_path)
+            expected = DataFrame([0, 1], columns=["a"])
+            tm.assert_frame_equal(result, expected)
+
+    def test_simple_normalize_with_separator(self, deep_nested):
+        # GH 14883
+        result = json_normalize({"A": {"A": 1, "B": 2}})
+        expected = DataFrame([[1, 2]], columns=["A.A", "A.B"])
+        tm.assert_frame_equal(result.reindex_like(expected), expected)
+
+        result = json_normalize({"A": {"A": 1, "B": 2}}, sep="_")
+        expected = DataFrame([[1, 2]], columns=["A_A", "A_B"])
+        tm.assert_frame_equal(result.reindex_like(expected), expected)
+
+        result = json_normalize({"A": {"A": 1, "B": 2}}, sep="\u03c3")
+        expected = DataFrame([[1, 2]], columns=["A\u03c3A", "A\u03c3B"])
+        tm.assert_frame_equal(result.reindex_like(expected), expected)
+
+        result = json_normalize(
+            deep_nested,
+            ["states", "cities"],
+            meta=["country", ["states", "name"]],
+            sep="_",
+        )
+        expected = Index(["name", "pop", "country", "states_name"]).sort_values()
+        assert result.columns.sort_values().equals(expected)
+
+    def test_normalize_with_multichar_separator(self):
+        # GH #43831
+        data = {"a": [1, 2], "b": {"b_1": 2, "b_2": (3, 4)}}
+        result = json_normalize(data, sep="__")
+        expected = DataFrame([[[1, 2], 2, (3, 4)]], columns=["a", "b__b_1", "b__b_2"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_value_array_record_prefix(self):
+        # GH 21536
+        result = json_normalize({"A": [1, 2]}, "A", record_prefix="Prefix.")
+        expected = DataFrame([[1], [2]], columns=["Prefix.0"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_object_record_path(self):
+        # GH 22706
+        data = {
+            "state": "Florida",
+            "info": {
+                "governor": "Rick Scott",
+                "counties": [
+                    {"name": "Dade", "population": 12345},
+                    {"name": "Broward", "population": 40000},
+                    {"name": "Palm Beach", "population": 60000},
+                ],
+            },
+        }
+        result = json_normalize(data, record_path=["info", "counties"])
+        expected = DataFrame(
+            [["Dade", 12345], ["Broward", 40000], ["Palm Beach", 60000]],
+            columns=["name", "population"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_more_deeply_nested(self, deep_nested):
+        result = json_normalize(
+            deep_nested, ["states", "cities"], meta=["country", ["states", "name"]]
+        )
+        ex_data = {
+            "country": ["USA"] * 4 + ["Germany"] * 3,
+            "states.name": [
+                "California",
+                "California",
+                "Ohio",
+                "Ohio",
+                "Bayern",
+                "Nordrhein-Westfalen",
+                "Nordrhein-Westfalen",
+            ],
+            "name": [
+                "San Francisco",
+                "Los Angeles",
+                "Columbus",
+                "Cleveland",
+                "Munich",
+                "Duesseldorf",
+                "Koeln",
+            ],
+            "pop": [12345, 12346, 1234, 1236, 12347, 1238, 1239],
+        }
+
+        expected = DataFrame(ex_data, columns=result.columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_shallow_nested(self):
+        data = [
+            {
+                "state": "Florida",
+                "shortname": "FL",
+                "info": {"governor": "Rick Scott"},
+                "counties": [
+                    {"name": "Dade", "population": 12345},
+                    {"name": "Broward", "population": 40000},
+                    {"name": "Palm Beach", "population": 60000},
+                ],
+            },
+            {
+                "state": "Ohio",
+                "shortname": "OH",
+                "info": {"governor": "John Kasich"},
+                "counties": [
+                    {"name": "Summit", "population": 1234},
+                    {"name": "Cuyahoga", "population": 1337},
+                ],
+            },
+        ]
+
+        result = json_normalize(
+            data, "counties", ["state", "shortname", ["info", "governor"]]
+        )
+        ex_data = {
+            "name": ["Dade", "Broward", "Palm Beach", "Summit", "Cuyahoga"],
+            "state": ["Florida"] * 3 + ["Ohio"] * 2,
+            "shortname": ["FL", "FL", "FL", "OH", "OH"],
+            "info.governor": ["Rick Scott"] * 3 + ["John Kasich"] * 2,
+            "population": [12345, 40000, 60000, 1234, 1337],
+        }
+        expected = DataFrame(ex_data, columns=result.columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_nested_meta_path_with_nested_record_path(self, state_data):
+        # GH 27220
+        result = json_normalize(
+            data=state_data,
+            record_path=["counties"],
+            meta=["state", "shortname", ["info", "governor"]],
+            errors="ignore",
+        )
+
+        ex_data = {
+            "name": ["Dade", "Broward", "Palm Beach", "Summit", "Cuyahoga"],
+            "population": [12345, 40000, 60000, 1234, 1337],
+            "state": ["Florida"] * 3 + ["Ohio"] * 2,
+            "shortname": ["FL"] * 3 + ["OH"] * 2,
+            "info.governor": ["Rick Scott"] * 3 + ["John Kasich"] * 2,
+        }
+
+        expected = DataFrame(ex_data)
+        tm.assert_frame_equal(result, expected)
+
+    def test_meta_name_conflict(self):
+        data = [
+            {
+                "foo": "hello",
+                "bar": "there",
+                "data": [
+                    {"foo": "something", "bar": "else"},
+                    {"foo": "something2", "bar": "else2"},
+                ],
+            }
+        ]
+
+        msg = r"Conflicting metadata name (foo|bar), need distinguishing prefix"
+        with pytest.raises(ValueError, match=msg):
+            json_normalize(data, "data", meta=["foo", "bar"])
+
+        result = json_normalize(data, "data", meta=["foo", "bar"], meta_prefix="meta")
+
+        for val in ["metafoo", "metabar", "foo", "bar"]:
+            assert val in result
+
+    def test_meta_parameter_not_modified(self):
+        # GH 18610
+        data = [
+            {
+                "foo": "hello",
+                "bar": "there",
+                "data": [
+                    {"foo": "something", "bar": "else"},
+                    {"foo": "something2", "bar": "else2"},
+                ],
+            }
+        ]
+
+        COLUMNS = ["foo", "bar"]
+        result = json_normalize(data, "data", meta=COLUMNS, meta_prefix="meta")
+
+        assert COLUMNS == ["foo", "bar"]
+        for val in ["metafoo", "metabar", "foo", "bar"]:
+            assert val in result
+
+    def test_record_prefix(self, state_data):
+        result = json_normalize(state_data[0], "counties")
+        expected = DataFrame(state_data[0]["counties"])
+        tm.assert_frame_equal(result, expected)
+
+        result = json_normalize(
+            state_data, "counties", meta="state", record_prefix="county_"
+        )
+
+        expected = []
+        for rec in state_data:
+            expected.extend(rec["counties"])
+        expected = DataFrame(expected)
+        expected = expected.rename(columns=lambda x: "county_" + x)
+        expected["state"] = np.array(["Florida", "Ohio"]).repeat([3, 2])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_non_ascii_key(self):
+        testjson = (
+            b'[{"\xc3\x9cnic\xc3\xb8de":0,"sub":{"A":1, "B":2}},'
+            b'{"\xc3\x9cnic\xc3\xb8de":1,"sub":{"A":3, "B":4}}]'
+        ).decode("utf8")
+
+        testdata = {
+            b"\xc3\x9cnic\xc3\xb8de".decode("utf8"): [0, 1],
+            "sub.A": [1, 3],
+            "sub.B": [2, 4],
+        }
+        expected = DataFrame(testdata)
+
+        result = json_normalize(json.loads(testjson))
+        tm.assert_frame_equal(result, expected)
+
+    def test_missing_field(self):
+        # GH20030:
+        author_missing_data = [
+            {"info": None},
+            {
+                "info": {"created_at": "11/08/1993", "last_updated": "26/05/2012"},
+                "author_name": {"first": "Jane", "last_name": "Doe"},
+            },
+        ]
+        result = json_normalize(author_missing_data)
+        ex_data = [
+            {
+                "info": np.nan,
+                "info.created_at": np.nan,
+                "info.last_updated": np.nan,
+                "author_name.first": np.nan,
+                "author_name.last_name": np.nan,
+            },
+            {
+                "info": None,
+                "info.created_at": "11/08/1993",
+                "info.last_updated": "26/05/2012",
+                "author_name.first": "Jane",
+                "author_name.last_name": "Doe",
+            },
+        ]
+        expected = DataFrame(ex_data)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "max_level,expected",
+        [
+            (
+                0,
+                [
+                    {
+                        "TextField": "Some text",
+                        "UserField": {"Id": "ID001", "Name": "Name001"},
+                        "CreatedBy": {"Name": "User001"},
+                        "Image": {"a": "b"},
+                    },
+                    {
+                        "TextField": "Some text",
+                        "UserField": {"Id": "ID001", "Name": "Name001"},
+                        "CreatedBy": {"Name": "User001"},
+                        "Image": {"a": "b"},
+                    },
+                ],
+            ),
+            (
+                1,
+                [
+                    {
+                        "TextField": "Some text",
+                        "UserField.Id": "ID001",
+                        "UserField.Name": "Name001",
+                        "CreatedBy": {"Name": "User001"},
+                        "Image": {"a": "b"},
+                    },
+                    {
+                        "TextField": "Some text",
+                        "UserField.Id": "ID001",
+                        "UserField.Name": "Name001",
+                        "CreatedBy": {"Name": "User001"},
+                        "Image": {"a": "b"},
+                    },
+                ],
+            ),
+        ],
+    )
+    def test_max_level_with_records_path(self, max_level, expected):
+        # GH23843: Enhanced JSON normalize
+        test_input = [
+            {
+                "CreatedBy": {"Name": "User001"},
+                "Lookup": [
+                    {
+                        "TextField": "Some text",
+                        "UserField": {"Id": "ID001", "Name": "Name001"},
+                    },
+                    {
+                        "TextField": "Some text",
+                        "UserField": {"Id": "ID001", "Name": "Name001"},
+                    },
+                ],
+                "Image": {"a": "b"},
+                "tags": [
+                    {"foo": "something", "bar": "else"},
+                    {"foo": "something2", "bar": "else2"},
+                ],
+            }
+        ]
+
+        result = json_normalize(
+            test_input,
+            record_path=["Lookup"],
+            meta=[["CreatedBy"], ["Image"]],
+            max_level=max_level,
+        )
+        expected_df = DataFrame(data=expected, columns=result.columns.values)
+        tm.assert_equal(expected_df, result)
+
+    def test_nested_flattening_consistent(self):
+        # see gh-21537
+        df1 = json_normalize([{"A": {"B": 1}}])
+        df2 = json_normalize({"dummy": [{"A": {"B": 1}}]}, "dummy")
+
+        # They should be the same.
+        tm.assert_frame_equal(df1, df2)
+
+    def test_nonetype_record_path(self, nulls_fixture):
+        # see gh-30148
+        # should not raise TypeError
+        result = json_normalize(
+            [
+                {"state": "Texas", "info": nulls_fixture},
+                {"state": "Florida", "info": [{"i": 2}]},
+            ],
+            record_path=["info"],
+        )
+        expected = DataFrame({"i": 2}, index=range(1))
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("value", ["false", "true", "{}", "1", '"text"'])
+    def test_non_list_record_path_errors(self, value):
+        # see gh-30148, GH 26284
+        parsed_value = json.loads(value)
+        test_input = {"state": "Texas", "info": parsed_value}
+        test_path = "info"
+        msg = (
+            f"Path must contain list or null, "
+            f"but got {type(parsed_value).__name__} at 'info'"
+        )
+        with pytest.raises(TypeError, match=msg):
+            json_normalize([test_input], record_path=[test_path])
+
+    def test_meta_non_iterable(self):
+        # GH 31507
+        data = """[{"id": 99, "data": [{"one": 1, "two": 2}]}]"""
+
+        result = json_normalize(json.loads(data), record_path=["data"], meta=["id"])
+        expected = DataFrame(
+            {"one": [1], "two": [2], "id": np.array([99], dtype=object)}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_generator(self, state_data):
+        # GH35923 Fix pd.json_normalize to not skip the first element of a
+        # generator input
+        def generator_data():
+            yield from state_data[0]["counties"]
+
+        result = json_normalize(generator_data())
+        expected = DataFrame(state_data[0]["counties"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_top_column_with_leading_underscore(self):
+        # 49861
+        data = {"_id": {"a1": 10, "l2": {"l3": 0}}, "gg": 4}
+        result = json_normalize(data, sep="_")
+        expected = DataFrame([[4, 10, 0]], columns=["gg", "_id_a1", "_id_l2_l3"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_series_index(self, state_data):
+        idx = Index([7, 8])
+        series = Series(state_data, index=idx)
+        result = json_normalize(series)
+        tm.assert_index_equal(result.index, idx)
+        result = json_normalize(series, "counties")
+        tm.assert_index_equal(result.index, idx.repeat([3, 2]))
+
+
+class TestNestedToRecord:
+    def test_flat_stays_flat(self):
+        recs = [{"flat1": 1, "flat2": 2}, {"flat3": 3, "flat2": 4}]
+        result = nested_to_record(recs)
+        expected = recs
+        assert result == expected
+
+    def test_one_level_deep_flattens(self):
+        data = {"flat1": 1, "dict1": {"c": 1, "d": 2}}
+
+        result = nested_to_record(data)
+        expected = {"dict1.c": 1, "dict1.d": 2, "flat1": 1}
+
+        assert result == expected
+
+    def test_nested_flattens(self):
+        data = {
+            "flat1": 1,
+            "dict1": {"c": 1, "d": 2},
+            "nested": {"e": {"c": 1, "d": 2}, "d": 2},
+        }
+
+        result = nested_to_record(data)
+        expected = {
+            "dict1.c": 1,
+            "dict1.d": 2,
+            "flat1": 1,
+            "nested.d": 2,
+            "nested.e.c": 1,
+            "nested.e.d": 2,
+        }
+
+        assert result == expected
+
+    def test_json_normalize_errors(self, missing_metadata):
+        # GH14583:
+        # If meta keys are not always present a new option to set
+        # errors='ignore' has been implemented
+
+        msg = (
+            "Key 'name' not found. To replace missing values of "
+            "'name' with np.nan, pass in errors='ignore'"
+        )
+        with pytest.raises(KeyError, match=msg):
+            json_normalize(
+                data=missing_metadata,
+                record_path="addresses",
+                meta="name",
+                errors="raise",
+            )
+
+    def test_missing_meta(self, missing_metadata):
+        # GH25468
+        # If metadata is nullable with errors set to ignore, the null values
+        # should be numpy.nan values
+        result = json_normalize(
+            data=missing_metadata, record_path="addresses", meta="name", errors="ignore"
+        )
+        ex_data = [
+            [9562, "Morris St.", "Massillon", "OH", 44646, "Alice"],
+            [8449, "Spring St.", "Elizabethton", "TN", 37643, np.nan],
+        ]
+        columns = ["number", "street", "city", "state", "zip", "name"]
+        expected = DataFrame(ex_data, columns=columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_missing_nested_meta(self):
+        # GH44312
+        # If errors="ignore" and nested metadata is null, we should return nan
+        data = {"meta": "foo", "nested_meta": None, "value": [{"rec": 1}, {"rec": 2}]}
+        result = json_normalize(
+            data,
+            record_path="value",
+            meta=["meta", ["nested_meta", "leaf"]],
+            errors="ignore",
+        )
+        ex_data = [[1, "foo", np.nan], [2, "foo", np.nan]]
+        columns = ["rec", "meta", "nested_meta.leaf"]
+        expected = DataFrame(ex_data, columns=columns).astype(
+            {"nested_meta.leaf": object}
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # If errors="raise" and nested metadata is null, we should raise with the
+        # key of the first missing level
+        with pytest.raises(KeyError, match="'leaf' not found"):
+            json_normalize(
+                data,
+                record_path="value",
+                meta=["meta", ["nested_meta", "leaf"]],
+                errors="raise",
+            )
+
+    def test_missing_meta_multilevel_record_path_errors_raise(self, missing_metadata):
+        # GH41876
+        # Ensure errors='raise' works as intended even when a record_path of length
+        # greater than one is passed in
+        msg = (
+            "Key 'name' not found. To replace missing values of "
+            "'name' with np.nan, pass in errors='ignore'"
+        )
+        with pytest.raises(KeyError, match=msg):
+            json_normalize(
+                data=missing_metadata,
+                record_path=["previous_residences", "cities"],
+                meta="name",
+                errors="raise",
+            )
+
+    def test_missing_meta_multilevel_record_path_errors_ignore(self, missing_metadata):
+        # GH41876
+        # Ensure errors='ignore' works as intended even when a record_path of length
+        # greater than one is passed in
+        result = json_normalize(
+            data=missing_metadata,
+            record_path=["previous_residences", "cities"],
+            meta="name",
+            errors="ignore",
+        )
+        ex_data = [
+            ["Foo York City", "Alice"],
+            ["Barmingham", np.nan],
+        ]
+        columns = ["city_name", "name"]
+        expected = DataFrame(ex_data, columns=columns)
+        tm.assert_frame_equal(result, expected)
+
+    def test_donot_drop_nonevalues(self):
+        # GH21356
+        data = [
+            {"info": None, "author_name": {"first": "Smith", "last_name": "Appleseed"}},
+            {
+                "info": {"created_at": "11/08/1993", "last_updated": "26/05/2012"},
+                "author_name": {"first": "Jane", "last_name": "Doe"},
+            },
+        ]
+        result = nested_to_record(data)
+        expected = [
+            {
+                "info": None,
+                "author_name.first": "Smith",
+                "author_name.last_name": "Appleseed",
+            },
+            {
+                "author_name.first": "Jane",
+                "author_name.last_name": "Doe",
+                "info.created_at": "11/08/1993",
+                "info.last_updated": "26/05/2012",
+            },
+        ]
+
+        assert result == expected
+
+    def test_nonetype_top_level_bottom_level(self):
+        # GH21158: If inner level json has a key with a null value
+        # make sure it does not do a new_d.pop twice and except
+        data = {
+            "id": None,
+            "location": {
+                "country": {
+                    "state": {
+                        "id": None,
+                        "town.info": {
+                            "id": None,
+                            "region": None,
+                            "x": 49.151580810546875,
+                            "y": -33.148521423339844,
+                            "z": 27.572303771972656,
+                        },
+                    }
+                }
+            },
+        }
+        result = nested_to_record(data)
+        expected = {
+            "id": None,
+            "location.country.state.id": None,
+            "location.country.state.town.info.id": None,
+            "location.country.state.town.info.region": None,
+            "location.country.state.town.info.x": 49.151580810546875,
+            "location.country.state.town.info.y": -33.148521423339844,
+            "location.country.state.town.info.z": 27.572303771972656,
+        }
+        assert result == expected
+
+    def test_nonetype_multiple_levels(self):
+        # GH21158: If inner level json has a key with a null value
+        # make sure it does not do a new_d.pop twice and except
+        data = {
+            "id": None,
+            "location": {
+                "id": None,
+                "country": {
+                    "id": None,
+                    "state": {
+                        "id": None,
+                        "town.info": {
+                            "region": None,
+                            "x": 49.151580810546875,
+                            "y": -33.148521423339844,
+                            "z": 27.572303771972656,
+                        },
+                    },
+                },
+            },
+        }
+        result = nested_to_record(data)
+        expected = {
+            "id": None,
+            "location.id": None,
+            "location.country.id": None,
+            "location.country.state.id": None,
+            "location.country.state.town.info.region": None,
+            "location.country.state.town.info.x": 49.151580810546875,
+            "location.country.state.town.info.y": -33.148521423339844,
+            "location.country.state.town.info.z": 27.572303771972656,
+        }
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "max_level, expected",
+        [
+            (
+                None,
+                [
+                    {
+                        "CreatedBy.Name": "User001",
+                        "Lookup.TextField": "Some text",
+                        "Lookup.UserField.Id": "ID001",
+                        "Lookup.UserField.Name": "Name001",
+                        "Image.a": "b",
+                    }
+                ],
+            ),
+            (
+                0,
+                [
+                    {
+                        "CreatedBy": {"Name": "User001"},
+                        "Lookup": {
+                            "TextField": "Some text",
+                            "UserField": {"Id": "ID001", "Name": "Name001"},
+                        },
+                        "Image": {"a": "b"},
+                    }
+                ],
+            ),
+            (
+                1,
+                [
+                    {
+                        "CreatedBy.Name": "User001",
+                        "Lookup.TextField": "Some text",
+                        "Lookup.UserField": {"Id": "ID001", "Name": "Name001"},
+                        "Image.a": "b",
+                    }
+                ],
+            ),
+        ],
+    )
+    def test_with_max_level(self, max_level, expected):
+        # GH23843: Enhanced JSON normalize
+        max_level_test_input_data = [
+            {
+                "CreatedBy": {"Name": "User001"},
+                "Lookup": {
+                    "TextField": "Some text",
+                    "UserField": {"Id": "ID001", "Name": "Name001"},
+                },
+                "Image": {"a": "b"},
+            }
+        ]
+        output = nested_to_record(max_level_test_input_data, max_level=max_level)
+        assert output == expected
+
+    def test_with_large_max_level(self):
+        # GH23843: Enhanced JSON normalize
+        max_level = 100
+        input_data = [
+            {
+                "CreatedBy": {
+                    "user": {
+                        "name": {"firstname": "Leo", "LastName": "Thomson"},
+                        "family_tree": {
+                            "father": {
+                                "name": "Father001",
+                                "father": {
+                                    "Name": "Father002",
+                                    "father": {
+                                        "name": "Father003",
+                                        "father": {"Name": "Father004"},
+                                    },
+                                },
+                            }
+                        },
+                    }
+                }
+            }
+        ]
+        expected = [
+            {
+                "CreatedBy.user.name.firstname": "Leo",
+                "CreatedBy.user.name.LastName": "Thomson",
+                "CreatedBy.user.family_tree.father.name": "Father001",
+                "CreatedBy.user.family_tree.father.father.Name": "Father002",
+                "CreatedBy.user.family_tree.father.father.father.name": "Father003",
+                "CreatedBy.user.family_tree.father.father.father.father.Name": "Father004",  # noqa: E501
+            }
+        ]
+        output = nested_to_record(input_data, max_level=max_level)
+        assert output == expected
+
+    def test_series_non_zero_index(self):
+        # GH 19020
+        data = {
+            0: {"id": 1, "name": "Foo", "elements": {"a": 1}},
+            1: {"id": 2, "name": "Bar", "elements": {"b": 2}},
+            2: {"id": 3, "name": "Baz", "elements": {"c": 3}},
+        }
+        s = Series(data)
+        s.index = [1, 2, 3]
+        result = json_normalize(s)
+        expected = DataFrame(
+            {
+                "id": [1, 2, 3],
+                "name": ["Foo", "Bar", "Baz"],
+                "elements.a": [1.0, np.nan, np.nan],
+                "elements.b": [np.nan, 2.0, np.nan],
+                "elements.c": [np.nan, np.nan, 3.0],
+            },
+            index=[1, 2, 3],
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/json/test_pandas.py b/pandas/pandas/tests/io/json/test_pandas.py
new file mode 100644
index 0000000000000..144b36166261b
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_pandas.py
@@ -0,0 +1,2285 @@
+import datetime
+from datetime import timedelta
+from io import StringIO
+import json
+import os
+import sys
+import time
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import IS64
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    NA,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    RangeIndex,
+    Series,
+    Timestamp,
+    date_range,
+    read_json,
+)
+import pandas._testing as tm
+
+from pandas.io.json import ujson_dumps
+
+
+def test_literal_json_raises():
+    # PR 53409
+    jsonl = """{"a": 1, "b": 2}
+        {"a": 3, "b": 4}
+        {"a": 5, "b": 6}
+        {"a": 7, "b": 8}"""
+
+    msg = r".* does not exist"
+
+    with pytest.raises(FileNotFoundError, match=msg):
+        read_json(jsonl, lines=False)
+
+    with pytest.raises(FileNotFoundError, match=msg):
+        read_json('{"a": 1, "b": 2}\n{"b":2, "a" :1}\n', lines=True)
+
+    with pytest.raises(FileNotFoundError, match=msg):
+        read_json(
+            '{"a\\\\":"foo\\\\","b":"bar"}\n{"a\\\\":"foo\\"","b":"bar"}\n',
+            lines=False,
+        )
+
+    with pytest.raises(FileNotFoundError, match=msg):
+        read_json('{"a": 1, "b": 2}\n{"b":2, "a" :1}\n', lines=False)
+
+
+def assert_json_roundtrip_equal(result, expected, orient):
+    if orient in ("records", "values"):
+        expected = expected.reset_index(drop=True)
+    if orient == "values":
+        expected.columns = range(len(expected.columns))
+    tm.assert_frame_equal(result, expected)
+
+
+class TestPandasContainer:
+    @pytest.fixture
+    def datetime_series(self):
+        # Same as usual datetime_series, but with index freq set to None,
+        #  since that doesn't round-trip, see GH#33711
+        ser = Series(
+            1.1 * np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        ser.index = ser.index._with_freq(None)
+        return ser
+
+    @pytest.fixture
+    def datetime_frame(self):
+        # Same as usual datetime_frame, but with index freq set to None,
+        #  since that doesn't round-trip, see GH#33711
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=30, freq="B"),
+        )
+        df.index = df.index._with_freq(None)
+        return df
+
+    def test_frame_double_encoded_labels(self, orient):
+        df = DataFrame(
+            [["a", "b"], ["c", "d"]],
+            index=['index " 1', "index / 2"],
+            columns=["a \\ b", "y / z"],
+        )
+
+        data = StringIO(df.to_json(orient=orient))
+        result = read_json(data, orient=orient)
+        expected = df.copy()
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("orient", ["split", "records", "values"])
+    def test_frame_non_unique_index(self, orient):
+        df = DataFrame([["a", "b"], ["c", "d"]], index=[1, 1], columns=["x", "y"])
+        data = StringIO(df.to_json(orient=orient))
+        result = read_json(data, orient=orient)
+        expected = df.copy()
+
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("orient", ["index", "columns"])
+    def test_frame_non_unique_index_raises(self, orient):
+        df = DataFrame([["a", "b"], ["c", "d"]], index=[1, 1], columns=["x", "y"])
+        msg = f"DataFrame index must be unique for orient='{orient}'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(orient=orient)
+
+    @pytest.mark.parametrize("orient", ["split", "values"])
+    @pytest.mark.parametrize(
+        "data",
+        [
+            [["a", "b"], ["c", "d"]],
+            [[1.5, 2.5], [3.5, 4.5]],
+            [[1, 2.5], [3, 4.5]],
+            [[Timestamp("20130101"), 3.5], [Timestamp("20130102"), 4.5]],
+        ],
+    )
+    def test_frame_non_unique_columns(self, orient, data, request):
+        if isinstance(data[0][0], Timestamp) and orient == "split":
+            mark = pytest.mark.xfail(
+                reason="GH#55827 non-nanosecond dt64 fails to round-trip"
+            )
+            request.applymarker(mark)
+
+        df = DataFrame(data, index=[1, 2], columns=["x", "x"])
+
+        expected_warning = None
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        if df.iloc[:, 0].dtype == "datetime64[s]":
+            expected_warning = FutureWarning
+
+        with tm.assert_produces_warning(expected_warning, match=msg):
+            result = read_json(
+                StringIO(df.to_json(orient=orient)), orient=orient, convert_dates=["x"]
+            )
+        if orient == "values":
+            expected = DataFrame(data)
+            if expected.iloc[:, 0].dtype == "datetime64[s]":
+                # orient == "values" by default will write Timestamp objects out
+                # in milliseconds; these are internally stored in nanosecond,
+                # so divide to get where we need
+                # TODO: a to_epoch method would also solve; see GH 14772
+                expected.isetitem(0, expected.iloc[:, 0].astype(np.int64) // 1000000)
+        elif orient == "split":
+            expected = df
+            expected.columns = ["x", "x.1"]
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("orient", ["index", "columns", "records"])
+    def test_frame_non_unique_columns_raises(self, orient):
+        df = DataFrame([["a", "b"], ["c", "d"]], index=[1, 2], columns=["x", "x"])
+
+        msg = f"DataFrame columns must be unique for orient='{orient}'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(orient=orient)
+
+    def test_frame_default_orient(self, float_frame):
+        assert float_frame.to_json() == float_frame.to_json(orient="columns")
+
+    @pytest.mark.parametrize("dtype", [False, float])
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_simple(self, orient, convert_axes, dtype, float_frame):
+        data = StringIO(float_frame.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes, dtype=dtype)
+
+        expected = float_frame
+
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("dtype", [False, np.int64])
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_intframe(self, orient, convert_axes, dtype, int_frame):
+        data = StringIO(int_frame.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes, dtype=dtype)
+        expected = int_frame
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("dtype", [None, np.float64, int, "U3"])
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_str_axes(self, orient, convert_axes, dtype):
+        df = DataFrame(
+            np.zeros((200, 4)),
+            columns=[str(i) for i in range(4)],
+            index=[str(i) for i in range(200)],
+            dtype=dtype,
+        )
+
+        data = StringIO(df.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes, dtype=dtype)
+
+        expected = df.copy()
+        if not dtype:
+            expected = expected.astype(np.int64)
+
+        # index columns, and records orients cannot fully preserve the string
+        # dtype for axes as the index and column labels are used as keys in
+        # JSON objects. JSON keys are by definition strings, so there's no way
+        # to disambiguate whether those keys actually were strings or numeric
+        # beforehand and numeric wins out.
+        if convert_axes and (orient in ("index", "columns")):
+            expected.columns = expected.columns.astype(np.int64)
+            expected.index = expected.index.astype(np.int64)
+        elif orient == "records" and convert_axes:
+            expected.columns = expected.columns.astype(np.int64)
+        elif convert_axes and orient == "split":
+            expected.columns = expected.columns.astype(np.int64)
+
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_categorical(
+        self, request, orient, convert_axes, using_infer_string
+    ):
+        # TODO: create a better frame to test with and improve coverage
+        if orient in ("index", "columns"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=f"Can't have duplicate index values for orient '{orient}')"
+                )
+            )
+
+        data = {
+            c: np.random.default_rng(i).standard_normal(30)
+            for i, c in enumerate(list("ABCD"))
+        }
+        cat = ["bah"] * 5 + ["bar"] * 5 + ["baz"] * 5 + ["foo"] * 15
+        data["E"] = list(reversed(cat))
+        data["sort"] = np.arange(30, dtype="int64")
+        categorical_frame = DataFrame(data, index=pd.CategoricalIndex(cat, name="E"))
+        data = StringIO(categorical_frame.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes)
+
+        expected = categorical_frame.copy()
+        expected.index = expected.index.astype(
+            str if not using_infer_string else "str"
+        )  # Categorical not preserved
+        expected.index.name = None  # index names aren't preserved in JSON
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_empty(self, orient, convert_axes):
+        empty_frame = DataFrame()
+        data = StringIO(empty_frame.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes)
+        if orient == "split":
+            idx = Index([], dtype=(float if convert_axes else object))
+            expected = DataFrame(index=idx, columns=idx)
+        elif orient in ["index", "columns"]:
+            expected = DataFrame()
+        else:
+            expected = empty_frame.copy()
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_timestamp(self, orient, convert_axes, datetime_frame):
+        # TODO: improve coverage with date_format parameter
+        data = StringIO(datetime_frame.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes)
+        expected = datetime_frame.copy()
+
+        if not convert_axes:  # one off for ts handling
+            # DTI gets converted to epoch values
+            idx = expected.index.view(np.int64) // 1000000
+            if orient != "split":  # TODO: handle consistently across orients
+                idx = idx.astype(str)
+
+            expected.index = idx
+
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_roundtrip_mixed(self, orient, convert_axes):
+        index = Index(["a", "b", "c", "d", "e"])
+        values = {
+            "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+            "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+            "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+            "D": [True, False, True, False, True],
+        }
+
+        df = DataFrame(data=values, index=index)
+
+        data = StringIO(df.to_json(orient=orient))
+        result = read_json(data, orient=orient, convert_axes=convert_axes)
+
+        expected = df.copy()
+        expected = expected.assign(**expected.select_dtypes("number").astype(np.int64))
+
+        assert_json_roundtrip_equal(result, expected, orient)
+
+    @pytest.mark.xfail(
+        reason="#50456 Column multiindex is stored and loaded differently",
+        raises=AssertionError,
+    )
+    @pytest.mark.parametrize(
+        "columns",
+        [
+            [["2022", "2022"], ["JAN", "FEB"]],
+            [["2022", "2023"], ["JAN", "JAN"]],
+            [["2022", "2022"], ["JAN", "JAN"]],
+        ],
+    )
+    def test_roundtrip_multiindex(self, columns):
+        df = DataFrame(
+            [[1, 2], [3, 4]],
+            columns=pd.MultiIndex.from_arrays(columns),
+        )
+        data = StringIO(df.to_json(orient="split"))
+        result = read_json(data, orient="split")
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize(
+        "data,msg,orient",
+        [
+            ('{"key":b:a:d}', "Expected object or value", "columns"),
+            # too few indices
+            (
+                '{"columns":["A","B"],'
+                '"index":["2","3"],'
+                '"data":[[1.0,"1"],[2.0,"2"],[null,"3"]]}',
+                "|".join(
+                    [
+                        r"Length of values \(3\) does not match length of index \(2\)",
+                    ]
+                ),
+                "split",
+            ),
+            # too many columns
+            (
+                '{"columns":["A","B","C"],'
+                '"index":["1","2","3"],'
+                '"data":[[1.0,"1"],[2.0,"2"],[null,"3"]]}',
+                "3 columns passed, passed data had 2 columns",
+                "split",
+            ),
+            # bad key
+            (
+                '{"badkey":["A","B"],'
+                '"index":["2","3"],'
+                '"data":[[1.0,"1"],[2.0,"2"],[null,"3"]]}',
+                r"unexpected key\(s\): badkey",
+                "split",
+            ),
+        ],
+    )
+    def test_frame_from_json_bad_data_raises(self, data, msg, orient):
+        with pytest.raises(ValueError, match=msg):
+            read_json(StringIO(data), orient=orient)
+
+    @pytest.mark.parametrize("dtype", [True, False])
+    @pytest.mark.parametrize("convert_axes", [True, False])
+    def test_frame_from_json_missing_data(self, orient, convert_axes, dtype):
+        num_df = DataFrame([[1, 2], [4, 5, 6]])
+
+        result = read_json(
+            StringIO(num_df.to_json(orient=orient)),
+            orient=orient,
+            convert_axes=convert_axes,
+            dtype=dtype,
+        )
+        assert np.isnan(result.iloc[0, 2])
+
+        obj_df = DataFrame([["1", "2"], ["4", "5", "6"]])
+        result = read_json(
+            StringIO(obj_df.to_json(orient=orient)),
+            orient=orient,
+            convert_axes=convert_axes,
+            dtype=dtype,
+        )
+        assert np.isnan(result.iloc[0, 2])
+
+    @pytest.mark.parametrize("dtype", [True, False])
+    def test_frame_read_json_dtype_missing_value(self, dtype):
+        # GH28501 Parse missing values using read_json with dtype=False
+        # to NaN instead of None
+        result = read_json(StringIO("[null]"), dtype=dtype)
+        expected = DataFrame([np.nan], dtype=object if not dtype else None)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("inf", [np.inf, -np.inf])
+    @pytest.mark.parametrize("dtype", [True, False])
+    def test_frame_infinity(self, inf, dtype):
+        # infinities get mapped to nulls which get mapped to NaNs during
+        # deserialisation
+        df = DataFrame([[1, 2], [4, 5, 6]])
+        df.loc[0, 2] = inf
+
+        data = StringIO(df.to_json())
+        result = read_json(data, dtype=dtype)
+        assert np.isnan(result.iloc[0, 2])
+
+    @pytest.mark.skipif(not IS64, reason="not compliant on 32-bit, xref #15865")
+    @pytest.mark.parametrize(
+        "value,precision,expected_val",
+        [
+            (0.95, 1, 1.0),
+            (1.95, 1, 2.0),
+            (-1.95, 1, -2.0),
+            (0.995, 2, 1.0),
+            (0.9995, 3, 1.0),
+            (0.99999999999999944, 15, 1.0),
+        ],
+    )
+    def test_frame_to_json_float_precision(self, value, precision, expected_val):
+        df = DataFrame([{"a_float": value}])
+        encoded = df.to_json(double_precision=precision)
+        assert encoded == f'{{"a_float":{{"0":{expected_val}}}}}'
+
+    def test_frame_to_json_except(self):
+        df = DataFrame([1, 2, 3])
+        msg = "Invalid value 'garbage' for option 'orient'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(orient="garbage")
+
+    def test_frame_empty(self):
+        df = DataFrame(columns=["jim", "joe"])
+        assert not df._is_mixed_type
+
+        data = StringIO(df.to_json())
+        result = read_json(data, dtype=dict(df.dtypes))
+        tm.assert_frame_equal(result, df, check_index_type=False)
+
+    def test_frame_empty_to_json(self):
+        # GH 7445
+        df = DataFrame({"test": []}, index=[])
+        result = df.to_json(orient="columns")
+        expected = '{"test":{}}'
+        assert result == expected
+
+    def test_frame_empty_mixedtype(self):
+        # mixed type
+        df = DataFrame(columns=["jim", "joe"])
+        df["joe"] = df["joe"].astype("i8")
+        assert df._is_mixed_type
+        data = df.to_json()
+        tm.assert_frame_equal(
+            read_json(StringIO(data), dtype=dict(df.dtypes)),
+            df,
+            check_index_type=False,
+        )
+
+    def test_frame_mixedtype_orient(self):  # GH10289
+        vals = [
+            [10, 1, "foo", 0.1, 0.01],
+            [20, 2, "bar", 0.2, 0.02],
+            [30, 3, "baz", 0.3, 0.03],
+            [40, 4, "qux", 0.4, 0.04],
+        ]
+
+        df = DataFrame(
+            vals, index=list("abcd"), columns=["1st", "2nd", "3rd", "4th", "5th"]
+        )
+
+        assert df._is_mixed_type
+        right = df.copy()
+
+        for orient in ["split", "index", "columns"]:
+            inp = StringIO(df.to_json(orient=orient))
+            left = read_json(inp, orient=orient, convert_axes=False)
+            tm.assert_frame_equal(left, right)
+
+        right.index = RangeIndex(len(df))
+        inp = StringIO(df.to_json(orient="records"))
+        left = read_json(inp, orient="records", convert_axes=False)
+        tm.assert_frame_equal(left, right)
+
+        right.columns = RangeIndex(df.shape[1])
+        inp = StringIO(df.to_json(orient="values"))
+        left = read_json(inp, orient="values", convert_axes=False)
+        tm.assert_frame_equal(left, right)
+
+    def test_v12_compat(self, datapath):
+        dti = date_range("2000-01-03", "2000-01-07")
+        # freq doesn't roundtrip
+        dti = DatetimeIndex(np.asarray(dti), freq=None)
+        df = DataFrame(
+            [
+                [1.56808523, 0.65727391, 1.81021139, -0.17251653],
+                [-0.2550111, -0.08072427, -0.03202878, -0.17581665],
+                [1.51493992, 0.11805825, 1.629455, -1.31506612],
+                [-0.02765498, 0.44679743, 0.33192641, -0.27885413],
+                [0.05951614, -2.69652057, 1.28163262, 0.34703478],
+            ],
+            columns=["A", "B", "C", "D"],
+            index=dti,
+        )
+        df["date"] = Timestamp("19920106 18:21:32.12").as_unit("ns")
+        df.iloc[3, df.columns.get_loc("date")] = Timestamp("20130101")
+        df["modified"] = df["date"]
+        df.iloc[1, df.columns.get_loc("modified")] = pd.NaT
+
+        dirpath = datapath("io", "json", "data")
+        v12_json = os.path.join(dirpath, "tsframe_v012.json")
+        df_unser = read_json(v12_json)
+        tm.assert_frame_equal(df, df_unser)
+
+        df_iso = df.drop(["modified"], axis=1)
+        v12_iso_json = os.path.join(dirpath, "tsframe_iso_v012.json")
+        df_unser_iso = read_json(v12_iso_json)
+        tm.assert_frame_equal(df_iso, df_unser_iso, check_column_type=False)
+
+    def test_blocks_compat_GH9037(self, using_infer_string):
+        index = date_range("20000101", periods=10, freq="h")
+        # freq doesn't round-trip
+        index = DatetimeIndex(list(index), freq=None)
+
+        df_mixed = DataFrame(
+            {
+                "float_1": [
+                    -0.92077639,
+                    0.77434435,
+                    1.25234727,
+                    0.61485564,
+                    -0.60316077,
+                    0.24653374,
+                    0.28668979,
+                    -2.51969012,
+                    0.95748401,
+                    -1.02970536,
+                ],
+                "int_1": [
+                    19680418,
+                    75337055,
+                    99973684,
+                    65103179,
+                    79373900,
+                    40314334,
+                    21290235,
+                    4991321,
+                    41903419,
+                    16008365,
+                ],
+                "str_1": [
+                    "78c608f1",
+                    "64a99743",
+                    "13d2ff52",
+                    "ca7f4af2",
+                    "97236474",
+                    "bde7e214",
+                    "1a6bde47",
+                    "b1190be5",
+                    "7a669144",
+                    "8d64d068",
+                ],
+                "float_2": [
+                    -0.0428278,
+                    -1.80872357,
+                    3.36042349,
+                    -0.7573685,
+                    -0.48217572,
+                    0.86229683,
+                    1.08935819,
+                    0.93898739,
+                    -0.03030452,
+                    1.43366348,
+                ],
+                "str_2": [
+                    "14f04af9",
+                    "d085da90",
+                    "4bcfac83",
+                    "81504caf",
+                    "2ffef4a9",
+                    "08e2f5c4",
+                    "07e1af03",
+                    "addbd4a7",
+                    "1f6a09ba",
+                    "4bfc4d87",
+                ],
+                "int_2": [
+                    86967717,
+                    98098830,
+                    51927505,
+                    20372254,
+                    12601730,
+                    20884027,
+                    34193846,
+                    10561746,
+                    24867120,
+                    76131025,
+                ],
+            },
+            index=index,
+        )
+
+        # JSON deserialisation always creates unicode strings
+        df_mixed.columns = df_mixed.columns.astype(
+            np.str_ if not using_infer_string else "str"
+        )
+        data = StringIO(df_mixed.to_json(orient="split"))
+        df_roundtrip = read_json(data, orient="split")
+        tm.assert_frame_equal(
+            df_mixed,
+            df_roundtrip,
+            check_index_type=True,
+            check_column_type=True,
+            by_blocks=True,
+            check_exact=True,
+        )
+
+    def test_frame_nonprintable_bytes(self):
+        # GH14256: failing column caused segfaults, if it is not the last one
+
+        class BinaryThing:
+            def __init__(self, hexed) -> None:
+                self.hexed = hexed
+                self.binary = bytes.fromhex(hexed)
+
+            def __str__(self) -> str:
+                return self.hexed
+
+        hexed = "574b4454ba8c5eb4f98a8f45"
+        binthing = BinaryThing(hexed)
+
+        # verify the proper conversion of printable content
+        df_printable = DataFrame({"A": [binthing.hexed]})
+        assert df_printable.to_json() == f'{{"A":{{"0":"{hexed}"}}}}'
+
+        # check if non-printable content throws appropriate Exception
+        df_nonprintable = DataFrame({"A": [binthing]})
+        msg = "Unsupported UTF-8 sequence length when encoding string"
+        with pytest.raises(OverflowError, match=msg):
+            df_nonprintable.to_json()
+
+        # the same with multiple columns threw segfaults
+        df_mixed = DataFrame({"A": [binthing], "B": [1]}, columns=["A", "B"])
+        with pytest.raises(OverflowError, match=msg):
+            df_mixed.to_json()
+
+        # default_handler should resolve exceptions for non-string types
+        result = df_nonprintable.to_json(default_handler=str)
+        expected = f'{{"A":{{"0":"{hexed}"}}}}'
+        assert result == expected
+        assert (
+            df_mixed.to_json(default_handler=str)
+            == f'{{"A":{{"0":"{hexed}"}},"B":{{"0":1}}}}'
+        )
+
+    def test_label_overflow(self):
+        # GH14256: buffer length not checked when writing label
+        result = DataFrame({"bar" * 100000: [1], "foo": [1337]}).to_json()
+        expected = f'{{"{"bar" * 100000}":{{"0":1}},"foo":{{"0":1337}}}}'
+        assert result == expected
+
+    def test_series_non_unique_index(self):
+        s = Series(["a", "b"], index=[1, 1])
+
+        msg = "Series index must be unique for orient='index'"
+        with pytest.raises(ValueError, match=msg):
+            s.to_json(orient="index")
+
+        tm.assert_series_equal(
+            s,
+            read_json(
+                StringIO(s.to_json(orient="split")), orient="split", typ="series"
+            ),
+        )
+        unserialized = read_json(
+            StringIO(s.to_json(orient="records")), orient="records", typ="series"
+        )
+        tm.assert_equal(s.values, unserialized.values)
+
+    def test_series_default_orient(self, string_series):
+        assert string_series.to_json() == string_series.to_json(orient="index")
+
+    def test_series_roundtrip_simple(self, orient, string_series, using_infer_string):
+        data = StringIO(string_series.to_json(orient=orient))
+        result = read_json(data, typ="series", orient=orient)
+
+        expected = string_series
+        if using_infer_string and orient in ("split", "index", "columns"):
+            # These schemas don't contain dtypes, so we infer string
+            expected.index = expected.index.astype("str")
+        if orient in ("values", "records"):
+            expected = expected.reset_index(drop=True)
+        if orient != "split":
+            expected.name = None
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [False, None])
+    def test_series_roundtrip_object(self, orient, dtype, object_series):
+        data = StringIO(object_series.to_json(orient=orient))
+        result = read_json(data, typ="series", orient=orient, dtype=dtype)
+
+        expected = object_series
+        if orient in ("values", "records"):
+            expected = expected.reset_index(drop=True)
+        if orient != "split":
+            expected.name = None
+
+        if using_string_dtype():
+            expected = expected.astype("str")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_roundtrip_empty(self, orient):
+        empty_series = Series([], index=[], dtype=np.float64)
+        data = StringIO(empty_series.to_json(orient=orient))
+        result = read_json(data, typ="series", orient=orient)
+
+        expected = empty_series.reset_index(drop=True)
+        if orient in ("split"):
+            expected.index = expected.index.astype(np.float64)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_roundtrip_timeseries(self, orient, datetime_series):
+        data = StringIO(datetime_series.to_json(orient=orient))
+        result = read_json(data, typ="series", orient=orient)
+
+        expected = datetime_series
+        if orient in ("values", "records"):
+            expected = expected.reset_index(drop=True)
+        if orient != "split":
+            expected.name = None
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [np.float64, int])
+    def test_series_roundtrip_numeric(self, orient, dtype):
+        s = Series(range(6), index=["a", "b", "c", "d", "e", "f"])
+        data = StringIO(s.to_json(orient=orient))
+        result = read_json(data, typ="series", orient=orient)
+
+        expected = s.copy()
+        if orient in ("values", "records"):
+            expected = expected.reset_index(drop=True)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_series_to_json_except(self):
+        s = Series([1, 2, 3])
+        msg = "Invalid value 'garbage' for option 'orient'"
+        with pytest.raises(ValueError, match=msg):
+            s.to_json(orient="garbage")
+
+    def test_series_from_json_precise_float(self):
+        s = Series([4.56, 4.56, 4.56])
+        result = read_json(StringIO(s.to_json()), typ="series", precise_float=True)
+        tm.assert_series_equal(result, s, check_index_type=False)
+
+    def test_series_with_dtype(self):
+        # GH 21986
+        s = Series([4.56, 4.56, 4.56])
+        result = read_json(StringIO(s.to_json()), typ="series", dtype=np.int64)
+        expected = Series([4] * 3)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype,expected",
+        [
+            (True, Series(["2000-01-01"], dtype="datetime64[ns]")),
+            (False, Series([946684800000])),
+        ],
+    )
+    def test_series_with_dtype_datetime(self, dtype, expected):
+        s = Series(["2000-01-01"], dtype="datetime64[ns]")
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            data = StringIO(s.to_json())
+        result = read_json(data, typ="series", dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_frame_from_json_precise_float(self):
+        df = DataFrame([[4.56, 4.56, 4.56], [4.56, 4.56, 4.56]])
+        result = read_json(StringIO(df.to_json()), precise_float=True)
+        tm.assert_frame_equal(result, df)
+
+    def test_typ(self):
+        s = Series(range(6), index=["a", "b", "c", "d", "e", "f"], dtype="int64")
+        result = read_json(StringIO(s.to_json()), typ="series")
+        tm.assert_series_equal(result, s)
+
+    def test_reconstruction_index(self):
+        df = DataFrame([[1, 2, 3], [4, 5, 6]])
+        result = read_json(StringIO(df.to_json()))
+        tm.assert_frame_equal(result, df)
+
+        df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]}, index=["A", "B", "C"])
+        result = read_json(StringIO(df.to_json()))
+        tm.assert_frame_equal(result, df)
+
+    def test_path(self, float_frame, int_frame, datetime_frame):
+        with tm.ensure_clean("test.json") as path:
+            for df in [float_frame, int_frame, datetime_frame]:
+                df.to_json(path)
+                read_json(path)
+
+    def test_axis_dates(self, datetime_series, datetime_frame):
+        # frame
+        json = StringIO(datetime_frame.to_json())
+        result = read_json(json)
+        tm.assert_frame_equal(result, datetime_frame)
+
+        # series
+        json = StringIO(datetime_series.to_json())
+        result = read_json(json, typ="series")
+        tm.assert_series_equal(result, datetime_series, check_names=False)
+        assert result.name is None
+
+    def test_convert_dates(self, datetime_series, datetime_frame):
+        # frame
+        df = datetime_frame
+        df["date"] = Timestamp("20130101").as_unit("ns")
+
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = StringIO(df.to_json())
+        result = read_json(json)
+        tm.assert_frame_equal(result, df)
+
+        df["foo"] = 1.0
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = StringIO(df.to_json(date_unit="ns"))
+
+        result = read_json(json, convert_dates=False)
+        expected = df.copy()
+        expected["date"] = expected["date"].values.view("i8")
+        expected["foo"] = expected["foo"].astype("int64")
+        tm.assert_frame_equal(result, expected)
+
+        # series
+        ts = Series(Timestamp("20130101").as_unit("ns"), index=datetime_series.index)
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = StringIO(ts.to_json())
+        result = read_json(json, typ="series")
+        tm.assert_series_equal(result, ts)
+
+    @pytest.mark.parametrize("date_format", ["epoch", "iso"])
+    @pytest.mark.parametrize("as_object", [True, False])
+    @pytest.mark.parametrize("date_typ", [datetime.date, datetime.datetime, Timestamp])
+    def test_date_index_and_values(self, date_format, as_object, date_typ):
+        data = [date_typ(year=2020, month=1, day=1), pd.NaT]
+        if as_object:
+            data.append("a")
+
+        ser = Series(data, index=data)
+        if not as_object:
+            ser = ser.astype("M8[ns]")
+            if isinstance(ser.index, DatetimeIndex):
+                ser.index = ser.index.as_unit("ns")
+
+        expected_warning = None
+        if date_format == "epoch":
+            expected = '{"1577836800000":1577836800000,"null":null}'
+            expected_warning = FutureWarning
+        else:
+            expected = (
+                '{"2020-01-01T00:00:00.000":"2020-01-01T00:00:00.000","null":null}'
+            )
+
+        msg = (
+            "'epoch' date format is deprecated and will be removed in a future "
+            "version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(expected_warning, match=msg):
+            result = ser.to_json(date_format=date_format)
+
+        if as_object:
+            expected = expected.replace("}", ',"a":"a"}')
+
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "infer_word",
+        [
+            "trade_time",
+            "date",
+            "datetime",
+            "sold_at",
+            "modified",
+            "timestamp",
+            "timestamps",
+        ],
+    )
+    def test_convert_dates_infer(self, infer_word):
+        # GH10747
+
+        data = [{"id": 1, infer_word: 1036713600000}, {"id": 2}]
+        expected = DataFrame(
+            [[1, Timestamp("2002-11-08")], [2, pd.NaT]], columns=["id", infer_word]
+        )
+        expected[infer_word] = expected[infer_word].astype("M8[ns]")
+
+        result = read_json(StringIO(ujson_dumps(data)))[["id", infer_word]]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "date,date_unit",
+        [
+            ("20130101 20:43:42.123", None),
+            ("20130101 20:43:42", "s"),
+            ("20130101 20:43:42.123", "ms"),
+            ("20130101 20:43:42.123456", "us"),
+            ("20130101 20:43:42.123456789", "ns"),
+        ],
+    )
+    def test_date_format_frame(self, date, date_unit, datetime_frame):
+        df = datetime_frame
+
+        df["date"] = Timestamp(date).as_unit("ns")
+        df.iloc[1, df.columns.get_loc("date")] = pd.NaT
+        df.iloc[5, df.columns.get_loc("date")] = pd.NaT
+        if date_unit:
+            json = df.to_json(date_format="iso", date_unit=date_unit)
+        else:
+            json = df.to_json(date_format="iso")
+
+        result = read_json(StringIO(json))
+        expected = df.copy()
+        tm.assert_frame_equal(result, expected)
+
+    def test_date_format_frame_raises(self, datetime_frame):
+        df = datetime_frame
+        msg = "Invalid value 'foo' for option 'date_unit'"
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(date_format="iso", date_unit="foo")
+
+    @pytest.mark.parametrize(
+        "date,date_unit",
+        [
+            ("20130101 20:43:42.123", None),
+            ("20130101 20:43:42", "s"),
+            ("20130101 20:43:42.123", "ms"),
+            ("20130101 20:43:42.123456", "us"),
+            ("20130101 20:43:42.123456789", "ns"),
+        ],
+    )
+    def test_date_format_series(self, date, date_unit, datetime_series):
+        ts = Series(Timestamp(date).as_unit("ns"), index=datetime_series.index)
+        ts.iloc[1] = pd.NaT
+        ts.iloc[5] = pd.NaT
+        if date_unit:
+            json = ts.to_json(date_format="iso", date_unit=date_unit)
+        else:
+            json = ts.to_json(date_format="iso")
+
+        result = read_json(StringIO(json), typ="series")
+        expected = ts.copy()
+        tm.assert_series_equal(result, expected)
+
+    def test_date_format_series_raises(self, datetime_series):
+        ts = Series(Timestamp("20130101 20:43:42.123"), index=datetime_series.index)
+        msg = "Invalid value 'foo' for option 'date_unit'"
+        with pytest.raises(ValueError, match=msg):
+            ts.to_json(date_format="iso", date_unit="foo")
+
+    def test_date_unit(self, unit, datetime_frame):
+        df = datetime_frame
+        df["date"] = Timestamp("20130101 20:43:42").as_unit("ns")
+        dl = df.columns.get_loc("date")
+        df.iloc[1, dl] = Timestamp("19710101 20:43:42")
+        df.iloc[2, dl] = Timestamp("21460101 20:43:42")
+        df.iloc[4, dl] = pd.NaT
+
+        msg = (
+            "'epoch' date format is deprecated and will be removed in a future "
+            "version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = df.to_json(date_format="epoch", date_unit=unit)
+
+        # force date unit
+        result = read_json(StringIO(json), date_unit=unit)
+        tm.assert_frame_equal(result, df)
+
+        # detect date unit
+        result = read_json(StringIO(json), date_unit=None)
+        tm.assert_frame_equal(result, df)
+
+    @pytest.mark.parametrize(
+        "df, warn",
+        [
+            (DataFrame({"A": ["a", "b", "c"], "B": np.arange(3)}), None),
+            (DataFrame({"A": [True, False, False]}), None),
+            (
+                DataFrame(
+                    {"A": ["a", "b", "c"], "B": pd.to_timedelta(np.arange(3), unit="D")}
+                ),
+                FutureWarning,
+            ),
+            (
+                DataFrame(
+                    {"A": pd.to_datetime(["2020-01-01", "2020-02-01", "2020-03-01"])}
+                ),
+                FutureWarning,
+            ),
+        ],
+    )
+    def test_default_epoch_date_format_deprecated(self, df, warn):
+        # GH 57063
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(warn, match=msg):
+            df.to_json()
+
+    @pytest.mark.parametrize("unit", ["s", "ms", "us"])
+    def test_iso_non_nano_datetimes(self, unit):
+        # Test that numpy datetimes
+        # in an Index or a column with non-nano resolution can be serialized
+        # correctly
+        # GH53686
+        index = DatetimeIndex(
+            [np.datetime64("2023-01-01T11:22:33.123456", unit)],
+            dtype=f"datetime64[{unit}]",
+        )
+        df = DataFrame(
+            {
+                "date": Series(
+                    [np.datetime64("2022-01-01T11:22:33.123456", unit)],
+                    dtype=f"datetime64[{unit}]",
+                    index=index,
+                ),
+                "date_obj": Series(
+                    [np.datetime64("2023-01-01T11:22:33.123456", unit)],
+                    dtype=object,
+                    index=index,
+                ),
+            },
+        )
+
+        buf = StringIO()
+        df.to_json(buf, date_format="iso", date_unit=unit)
+        buf.seek(0)
+
+        # read_json always reads datetimes in nanosecond resolution
+        # TODO: check_dtype/check_index_type should be removable
+        # once read_json gets non-nano support
+        tm.assert_frame_equal(
+            read_json(buf, convert_dates=["date", "date_obj"]),
+            df,
+            check_index_type=False,
+            check_dtype=False,
+        )
+
+    def test_weird_nested_json(self):
+        # this used to core dump the parser
+        s = r"""{
+        "status": "success",
+        "data": {
+        "posts": [
+            {
+            "id": 1,
+            "title": "A blog post",
+            "body": "Some useful content"
+            },
+            {
+            "id": 2,
+            "title": "Another blog post",
+            "body": "More content"
+            }
+           ]
+          }
+        }"""
+        read_json(StringIO(s))
+
+    def test_doc_example(self):
+        dfj2 = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=list("AB")
+        )
+        dfj2["date"] = Timestamp("20130101")
+        dfj2["ints"] = range(5)
+        dfj2["bools"] = True
+        dfj2.index = date_range("20130101", periods=5)
+
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = StringIO(dfj2.to_json())
+        result = read_json(json, dtype={"ints": np.int64, "bools": np.bool_})
+        tm.assert_frame_equal(result, result)
+
+    def test_round_trip_exception(self, datapath):
+        # GH 3867
+        path = datapath("io", "json", "data", "teams.csv")
+        df = pd.read_csv(path)
+        s = df.to_json()
+
+        result = read_json(StringIO(s))
+        res = result.reindex(index=df.index, columns=df.columns)
+        res = res.fillna(np.nan)
+        tm.assert_frame_equal(res, df)
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    @pytest.mark.parametrize(
+        "field,dtype",
+        [
+            ["created_at", pd.DatetimeTZDtype(tz="UTC")],
+            ["closed_at", "datetime64[ns]"],
+            ["updated_at", pd.DatetimeTZDtype(tz="UTC")],
+        ],
+    )
+    def test_url(self, field, dtype, httpserver):
+        data = '{"created_at": ["2023-06-23T18:21:36Z"], "closed_at": ["2023-06-23T18:21:36"], "updated_at": ["2023-06-23T18:21:36Z"]}\n'  # noqa: E501
+        httpserver.serve_content(content=data)
+        result = read_json(httpserver.url, convert_dates=True)
+        assert result[field].dtype == dtype
+
+    def test_timedelta(self):
+        converter = lambda x: pd.to_timedelta(x, unit="ms")
+
+        ser = Series([timedelta(23), timedelta(seconds=5)])
+        assert ser.dtype == "timedelta64[ns]"
+
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = read_json(StringIO(ser.to_json()), typ="series").apply(converter)
+        tm.assert_series_equal(result, ser)
+
+        ser = Series([timedelta(23), timedelta(seconds=5)], index=Index([0, 1]))
+        assert ser.dtype == "timedelta64[ns]"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = read_json(StringIO(ser.to_json()), typ="series").apply(converter)
+        tm.assert_series_equal(result, ser)
+
+        frame = DataFrame([timedelta(23), timedelta(seconds=5)])
+        assert frame[0].dtype == "timedelta64[ns]"
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            json = frame.to_json()
+        tm.assert_frame_equal(frame, read_json(StringIO(json)).apply(converter))
+
+    def test_timedelta2(self):
+        frame = DataFrame(
+            {
+                "a": [timedelta(days=23), timedelta(seconds=5)],
+                "b": [1, 2],
+                "c": date_range(start="20130101", periods=2),
+            }
+        )
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            data = StringIO(frame.to_json(date_unit="ns"))
+        result = read_json(data)
+        result["a"] = pd.to_timedelta(result.a, unit="ns")
+        result["c"] = pd.to_datetime(result.c)
+        tm.assert_frame_equal(frame, result)
+
+    def test_mixed_timedelta_datetime(self):
+        td = timedelta(23)
+        ts = Timestamp("20130101")
+        frame = DataFrame({"a": [td, ts]}, dtype=object)
+
+        expected = DataFrame(
+            {"a": [pd.Timedelta(td).as_unit("ns")._value, ts.as_unit("ns")._value]}
+        )
+        data = StringIO(frame.to_json(date_unit="ns"))
+        result = read_json(data, dtype={"a": "int64"})
+        tm.assert_frame_equal(result, expected, check_index_type=False)
+
+    @pytest.mark.parametrize("as_object", [True, False])
+    @pytest.mark.parametrize("date_format", ["iso", "epoch"])
+    @pytest.mark.parametrize("timedelta_typ", [pd.Timedelta, timedelta])
+    def test_timedelta_to_json(self, as_object, date_format, timedelta_typ):
+        # GH28156: to_json not correctly formatting Timedelta
+        data = [timedelta_typ(days=1), timedelta_typ(days=2), pd.NaT]
+        if as_object:
+            data.append("a")
+
+        ser = Series(data, index=data)
+        expected_warning = None
+        if date_format == "iso":
+            expected = (
+                '{"P1DT0H0M0S":"P1DT0H0M0S","P2DT0H0M0S":"P2DT0H0M0S","null":null}'
+            )
+        else:
+            expected_warning = FutureWarning
+            expected = '{"86400000":86400000,"172800000":172800000,"null":null}'
+
+        if as_object:
+            expected = expected.replace("}", ',"a":"a"}')
+
+        msg = (
+            "'epoch' date format is deprecated and will be removed in a future "
+            "version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(expected_warning, match=msg):
+            result = ser.to_json(date_format=date_format)
+        assert result == expected
+
+    @pytest.mark.parametrize("as_object", [True, False])
+    @pytest.mark.parametrize("timedelta_typ", [pd.Timedelta, timedelta])
+    def test_timedelta_to_json_fractional_precision(self, as_object, timedelta_typ):
+        data = [timedelta_typ(milliseconds=42)]
+        ser = Series(data, index=data)
+        warn = FutureWarning
+        if as_object:
+            ser = ser.astype(object)
+            warn = None
+
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(warn, match=msg):
+            result = ser.to_json()
+        expected = '{"42":42}'
+        assert result == expected
+
+    def test_default_handler(self):
+        value = object()
+        frame = DataFrame({"a": [7, value]})
+        expected = DataFrame({"a": [7, str(value)]})
+        result = read_json(StringIO(frame.to_json(default_handler=str)))
+        tm.assert_frame_equal(expected, result, check_index_type=False)
+
+    def test_default_handler_indirect(self):
+        def default(obj):
+            if isinstance(obj, complex):
+                return [("mathjs", "Complex"), ("re", obj.real), ("im", obj.imag)]
+            return str(obj)
+
+        df_list = [
+            9,
+            DataFrame(
+                {"a": [1, "STR", complex(4, -5)], "b": [float("nan"), None, "N/A"]},
+                columns=["a", "b"],
+            ),
+        ]
+        expected = (
+            '[9,[[1,null],["STR",null],[[["mathjs","Complex"],'
+            '["re",4.0],["im",-5.0]],"N\\/A"]]]'
+        )
+        assert (
+            ujson_dumps(df_list, default_handler=default, orient="values") == expected
+        )
+
+    def test_default_handler_numpy_unsupported_dtype(self):
+        # GH12554 to_json raises 'Unhandled numpy dtype 15'
+        df = DataFrame(
+            {"a": [1, 2.3, complex(4, -5)], "b": [float("nan"), None, complex(1.2, 0)]},
+            columns=["a", "b"],
+        )
+        expected = (
+            '[["(1+0j)","(nan+0j)"],["(2.3+0j)","(nan+0j)"],["(4-5j)","(1.2+0j)"]]'
+        )
+        assert df.to_json(default_handler=str, orient="values") == expected
+
+    def test_default_handler_raises(self):
+        msg = "raisin"
+
+        def my_handler_raises(obj):
+            raise TypeError(msg)
+
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"a": [1, 2, object()]}).to_json(
+                default_handler=my_handler_raises
+            )
+        with pytest.raises(TypeError, match=msg):
+            DataFrame({"a": [1, 2, complex(4, -5)]}).to_json(
+                default_handler=my_handler_raises
+            )
+
+    def test_categorical(self):
+        # GH4377 df.to_json segfaults with non-ndarray blocks
+        df = DataFrame({"A": ["a", "b", "c", "a", "b", "b", "a"]})
+        df["B"] = df["A"]
+        expected = df.to_json()
+
+        df["B"] = df["A"].astype("category")
+        assert expected == df.to_json()
+
+        s = df["A"]
+        sc = df["B"]
+        assert s.to_json() == sc.to_json()
+
+    def test_datetime_tz(self):
+        # GH4377 df.to_json segfaults with non-ndarray blocks
+        tz_range = date_range("20130101", periods=3, tz="US/Eastern")
+        tz_naive = tz_range.tz_convert("utc").tz_localize(None)
+
+        df = DataFrame({"A": tz_range, "B": date_range("20130101", periods=3)})
+
+        df_naive = df.copy()
+        df_naive["A"] = tz_naive
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            expected = df_naive.to_json()
+            assert expected == df.to_json()
+
+        stz = Series(tz_range)
+        s_naive = Series(tz_naive)
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert stz.to_json() == s_naive.to_json()
+
+    def test_sparse(self):
+        # GH4377 df.to_json segfaults with non-ndarray blocks
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        df.loc[:8] = np.nan
+
+        sdf = df.astype("Sparse")
+        expected = df.to_json()
+        assert expected == sdf.to_json()
+
+        s = Series(np.random.default_rng(2).standard_normal(10))
+        s.loc[:8] = np.nan
+        ss = s.astype("Sparse")
+
+        expected = s.to_json()
+        assert expected == ss.to_json()
+
+    @pytest.mark.parametrize(
+        "ts",
+        [
+            Timestamp("2013-01-10 05:00:00Z"),
+            Timestamp("2013-01-10 00:00:00", tz="US/Eastern"),
+            Timestamp("2013-01-10 00:00:00-0500"),
+        ],
+    )
+    def test_tz_is_utc(self, ts):
+        exp = '"2013-01-10T05:00:00.000Z"'
+
+        assert ujson_dumps(ts, iso_dates=True) == exp
+        dt = ts.to_pydatetime()
+        assert ujson_dumps(dt, iso_dates=True) == exp
+
+    def test_tz_is_naive(self):
+        ts = Timestamp("2013-01-10 05:00:00")
+        exp = '"2013-01-10T05:00:00.000"'
+
+        assert ujson_dumps(ts, iso_dates=True) == exp
+        dt = ts.to_pydatetime()
+        assert ujson_dumps(dt, iso_dates=True) == exp
+
+    @pytest.mark.parametrize(
+        "tz_range",
+        [
+            date_range("2013-01-01 05:00:00Z", periods=2),
+            date_range("2013-01-01 00:00:00", periods=2, tz="US/Eastern"),
+            date_range("2013-01-01 00:00:00-0500", periods=2),
+        ],
+    )
+    def test_tz_range_is_utc(self, tz_range):
+        exp = '["2013-01-01T05:00:00.000Z","2013-01-02T05:00:00.000Z"]'
+        dfexp = '{"DT":{"0":"2013-01-01T05:00:00.000Z","1":"2013-01-02T05:00:00.000Z"}}'
+
+        assert ujson_dumps(tz_range, iso_dates=True) == exp
+        dti = DatetimeIndex(tz_range)
+        # Ensure datetimes in object array are serialized correctly
+        # in addition to the normal DTI case
+        assert ujson_dumps(dti, iso_dates=True) == exp
+        assert ujson_dumps(dti.astype(object), iso_dates=True) == exp
+        df = DataFrame({"DT": dti})
+        result = ujson_dumps(df, iso_dates=True)
+        assert result == dfexp
+        assert ujson_dumps(df.astype({"DT": object}), iso_dates=True)
+
+    def test_tz_range_is_naive(self):
+        dti = date_range("2013-01-01 05:00:00", periods=2)
+
+        exp = '["2013-01-01T05:00:00.000","2013-01-02T05:00:00.000"]'
+        dfexp = '{"DT":{"0":"2013-01-01T05:00:00.000","1":"2013-01-02T05:00:00.000"}}'
+
+        # Ensure datetimes in object array are serialized correctly
+        # in addition to the normal DTI case
+        assert ujson_dumps(dti, iso_dates=True) == exp
+        assert ujson_dumps(dti.astype(object), iso_dates=True) == exp
+        df = DataFrame({"DT": dti})
+        result = ujson_dumps(df, iso_dates=True)
+        assert result == dfexp
+        assert ujson_dumps(df.astype({"DT": object}), iso_dates=True)
+
+    def test_read_inline_jsonl(self):
+        # GH9180
+
+        result = read_json(StringIO('{"a": 1, "b": 2}\n{"b":2, "a" :1}\n'), lines=True)
+        expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.single_cpu
+    @pytest.mark.network
+    @td.skip_if_not_us_locale
+    def test_read_s3_jsonl(self, s3_public_bucket_with_data, s3so):
+        # GH17200
+
+        result = read_json(
+            f"s3n://{s3_public_bucket_with_data.name}/items.jsonl",
+            lines=True,
+            storage_options=s3so,
+        )
+        expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_read_local_jsonl(self):
+        # GH17200
+        with tm.ensure_clean("tmp_items.json") as path:
+            with open(path, "w", encoding="utf-8") as infile:
+                infile.write('{"a": 1, "b": 2}\n{"b":2, "a" :1}\n')
+            result = read_json(path, lines=True)
+            expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+            tm.assert_frame_equal(result, expected)
+
+    def test_read_jsonl_unicode_chars(self):
+        # GH15132: non-ascii unicode characters
+        # \u201d == RIGHT DOUBLE QUOTATION MARK
+
+        # simulate file handle
+        json = '{"a": "foo”", "b": "bar"}\n{"a": "foo", "b": "bar"}\n'
+        json = StringIO(json)
+        result = read_json(json, lines=True)
+        expected = DataFrame([["foo\u201d", "bar"], ["foo", "bar"]], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+        # simulate string
+        json = StringIO('{"a": "foo”", "b": "bar"}\n{"a": "foo", "b": "bar"}\n')
+        result = read_json(json, lines=True)
+        expected = DataFrame([["foo\u201d", "bar"], ["foo", "bar"]], columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("bigNum", [sys.maxsize + 1, -(sys.maxsize + 2)])
+    def test_to_json_large_numbers(self, bigNum):
+        # GH34473
+        series = Series(bigNum, dtype=object, index=["articleId"])
+        json = series.to_json()
+        expected = '{"articleId":' + str(bigNum) + "}"
+        assert json == expected
+
+        df = DataFrame(bigNum, dtype=object, index=["articleId"], columns=[0])
+        json = df.to_json()
+        expected = '{"0":{"articleId":' + str(bigNum) + "}}"
+        assert json == expected
+
+    @pytest.mark.parametrize("bigNum", [-(2**63) - 1, 2**64])
+    def test_read_json_large_numbers(self, bigNum):
+        # GH20599, 26068
+        json = StringIO('{"articleId":' + str(bigNum) + "}")
+        msg = r"Value is too small|Value is too big"
+        with pytest.raises(ValueError, match=msg):
+            read_json(json)
+
+        json = StringIO('{"0":{"articleId":' + str(bigNum) + "}}")
+        with pytest.raises(ValueError, match=msg):
+            read_json(json)
+
+    def test_read_json_large_numbers2(self):
+        # GH18842
+        json = '{"articleId": "1404366058080022500245"}'
+        json = StringIO(json)
+        result = read_json(json, typ="series")
+        expected = Series(1.404366e21, index=["articleId"])
+        tm.assert_series_equal(result, expected)
+
+        json = '{"0": {"articleId": "1404366058080022500245"}}'
+        json = StringIO(json)
+        result = read_json(json)
+        expected = DataFrame(1.404366e21, index=["articleId"], columns=[0])
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_jsonl(self):
+        # GH9180
+        df = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+        result = df.to_json(orient="records", lines=True)
+        expected = '{"a":1,"b":2}\n{"a":1,"b":2}\n'
+        assert result == expected
+
+        df = DataFrame([["foo}", "bar"], ['foo"', "bar"]], columns=["a", "b"])
+        result = df.to_json(orient="records", lines=True)
+        expected = '{"a":"foo}","b":"bar"}\n{"a":"foo\\"","b":"bar"}\n'
+        assert result == expected
+        tm.assert_frame_equal(read_json(StringIO(result), lines=True), df)
+
+        # GH15096: escaped characters in columns and data
+        df = DataFrame([["foo\\", "bar"], ['foo"', "bar"]], columns=["a\\", "b"])
+        result = df.to_json(orient="records", lines=True)
+        expected = '{"a\\\\":"foo\\\\","b":"bar"}\n{"a\\\\":"foo\\"","b":"bar"}\n'
+        assert result == expected
+
+        tm.assert_frame_equal(read_json(StringIO(result), lines=True), df)
+
+    # TODO: there is a near-identical test for pytables; can we share?
+    @pytest.mark.xfail(reason="GH#13774 encoding kwarg not supported", raises=TypeError)
+    @pytest.mark.parametrize(
+        "val",
+        [
+            [b"E\xc9, 17", b"", b"a", b"b", b"c"],
+            [b"E\xc9, 17", b"a", b"b", b"c"],
+            [b"EE, 17", b"", b"a", b"b", b"c"],
+            [b"E\xc9, 17", b"\xf8\xfc", b"a", b"b", b"c"],
+            [b"", b"a", b"b", b"c"],
+            [b"\xf8\xfc", b"a", b"b", b"c"],
+            [b"A\xf8\xfc", b"", b"a", b"b", b"c"],
+            [np.nan, b"", b"b", b"c"],
+            [b"A\xf8\xfc", np.nan, b"", b"b", b"c"],
+        ],
+    )
+    @pytest.mark.parametrize("dtype", ["category", object])
+    def test_latin_encoding(self, dtype, val):
+        # GH 13774
+        ser = Series(
+            [x.decode("latin-1") if isinstance(x, bytes) else x for x in val],
+            dtype=dtype,
+        )
+        encoding = "latin-1"
+        with tm.ensure_clean("test.json") as path:
+            ser.to_json(path, encoding=encoding)
+            retr = read_json(StringIO(path), encoding=encoding)
+            tm.assert_series_equal(ser, retr, check_categorical=False)
+
+    def test_data_frame_size_after_to_json(self):
+        # GH15344
+        df = DataFrame({"a": [str(1)]})
+
+        size_before = df.memory_usage(index=True, deep=True).sum()
+        df.to_json()
+        size_after = df.memory_usage(index=True, deep=True).sum()
+
+        assert size_before == size_after
+
+    @pytest.mark.parametrize(
+        "index", [None, [1, 2], [1.0, 2.0], ["a", "b"], ["1", "2"], ["1.", "2."]]
+    )
+    @pytest.mark.parametrize("columns", [["a", "b"], ["1", "2"], ["1.", "2."]])
+    def test_from_json_to_json_table_index_and_columns(self, index, columns):
+        # GH25433 GH25435
+        expected = DataFrame([[1, 2], [3, 4]], index=index, columns=columns)
+        dfjson = expected.to_json(orient="table")
+
+        result = read_json(StringIO(dfjson), orient="table")
+        tm.assert_frame_equal(result, expected)
+
+    def test_from_json_to_json_table_dtypes(self):
+        # GH21345
+        expected = DataFrame({"a": [1, 2], "b": [3.0, 4.0], "c": ["5", "6"]})
+        dfjson = expected.to_json(orient="table")
+        result = read_json(StringIO(dfjson), orient="table")
+        tm.assert_frame_equal(result, expected)
+
+    # TODO: We are casting to string which coerces None to NaN before casting back
+    # to object, ending up with incorrect na values
+    @pytest.mark.xfail(using_string_dtype(), reason="incorrect na conversion")
+    @pytest.mark.parametrize("orient", ["split", "records", "index", "columns"])
+    def test_to_json_from_json_columns_dtypes(self, orient):
+        # GH21892 GH33205
+        expected = DataFrame.from_dict(
+            {
+                "Integer": Series([1, 2, 3], dtype="int64"),
+                "Float": Series([None, 2.0, 3.0], dtype="float64"),
+                "Object": Series([None, "", "c"], dtype="object"),
+                "Bool": Series([True, False, True], dtype="bool"),
+                "Category": Series(["a", "b", None], dtype="category"),
+                "Datetime": Series(
+                    ["2020-01-01", None, "2020-01-03"], dtype="datetime64[ns]"
+                ),
+            }
+        )
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            dfjson = expected.to_json(orient=orient)
+
+        result = read_json(
+            StringIO(dfjson),
+            orient=orient,
+            dtype={
+                "Integer": "int64",
+                "Float": "float64",
+                "Object": "object",
+                "Bool": "bool",
+                "Category": "category",
+                "Datetime": "datetime64[ns]",
+            },
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_to_json_with_index_as_a_column_name(self):
+        df = DataFrame(data={"index": [1, 2], "a": [2, 3]})
+        with pytest.raises(
+            ValueError, match="Overlapping names between the index and columns"
+        ):
+            df.to_json(orient="table")
+
+    @pytest.mark.parametrize("dtype", [True, {"b": int, "c": int}])
+    def test_read_json_table_dtype_raises(self, dtype):
+        # GH21345
+        df = DataFrame({"a": [1, 2], "b": [3.0, 4.0], "c": ["5", "6"]})
+        dfjson = df.to_json(orient="table")
+        msg = "cannot pass both dtype and orient='table'"
+        with pytest.raises(ValueError, match=msg):
+            read_json(dfjson, orient="table", dtype=dtype)
+
+    @pytest.mark.parametrize("orient", ["index", "columns", "records", "values"])
+    def test_read_json_table_empty_axes_dtype(self, orient):
+        # GH28558
+
+        expected = DataFrame()
+        result = read_json(StringIO("{}"), orient=orient, convert_axes=True)
+        tm.assert_index_equal(result.index, expected.index)
+        tm.assert_index_equal(result.columns, expected.columns)
+
+    def test_read_json_table_convert_axes_raises(self):
+        # GH25433 GH25435
+        df = DataFrame([[1, 2], [3, 4]], index=[1.0, 2.0], columns=["1.", "2."])
+        dfjson = df.to_json(orient="table")
+        msg = "cannot pass both convert_axes and orient='table'"
+        with pytest.raises(ValueError, match=msg):
+            read_json(dfjson, orient="table", convert_axes=True)
+
+    @pytest.mark.parametrize(
+        "data, expected",
+        [
+            (
+                DataFrame([[1, 2], [4, 5]], columns=["a", "b"]),
+                {"columns": ["a", "b"], "data": [[1, 2], [4, 5]]},
+            ),
+            (
+                DataFrame([[1, 2], [4, 5]], columns=["a", "b"]).rename_axis("foo"),
+                {"columns": ["a", "b"], "data": [[1, 2], [4, 5]]},
+            ),
+            (
+                DataFrame(
+                    [[1, 2], [4, 5]], columns=["a", "b"], index=[["a", "b"], ["c", "d"]]
+                ),
+                {"columns": ["a", "b"], "data": [[1, 2], [4, 5]]},
+            ),
+            (Series([1, 2, 3], name="A"), {"name": "A", "data": [1, 2, 3]}),
+            (
+                Series([1, 2, 3], name="A").rename_axis("foo"),
+                {"name": "A", "data": [1, 2, 3]},
+            ),
+            (
+                Series([1, 2], name="A", index=[["a", "b"], ["c", "d"]]),
+                {"name": "A", "data": [1, 2]},
+            ),
+        ],
+    )
+    def test_index_false_to_json_split(self, data, expected):
+        # GH 17394
+        # Testing index=False in to_json with orient='split'
+
+        result = data.to_json(orient="split", index=False)
+        result = json.loads(result)
+
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            (DataFrame([[1, 2], [4, 5]], columns=["a", "b"])),
+            (DataFrame([[1, 2], [4, 5]], columns=["a", "b"]).rename_axis("foo")),
+            (
+                DataFrame(
+                    [[1, 2], [4, 5]], columns=["a", "b"], index=[["a", "b"], ["c", "d"]]
+                )
+            ),
+            (Series([1, 2, 3], name="A")),
+            (Series([1, 2, 3], name="A").rename_axis("foo")),
+            (Series([1, 2], name="A", index=[["a", "b"], ["c", "d"]])),
+        ],
+    )
+    def test_index_false_to_json_table(self, data):
+        # GH 17394
+        # Testing index=False in to_json with orient='table'
+
+        result = data.to_json(orient="table", index=False)
+        result = json.loads(result)
+
+        expected = {
+            "schema": pd.io.json.build_table_schema(data, index=False),
+            "data": DataFrame(data).to_dict(orient="records"),
+        }
+
+        assert result == expected
+
+    @pytest.mark.parametrize("orient", ["index", "columns"])
+    def test_index_false_error_to_json(self, orient):
+        # GH 17394, 25513
+        # Testing error message from to_json with index=False
+
+        df = DataFrame([[1, 2], [4, 5]], columns=["a", "b"])
+
+        msg = (
+            "'index=False' is only valid when 'orient' is 'split', "
+            "'table', 'records', or 'values'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(orient=orient, index=False)
+
+    @pytest.mark.parametrize("orient", ["records", "values"])
+    def test_index_true_error_to_json(self, orient):
+        # GH 25513
+        # Testing error message from to_json with index=True
+
+        df = DataFrame([[1, 2], [4, 5]], columns=["a", "b"])
+
+        msg = (
+            "'index=True' is only valid when 'orient' is 'split', "
+            "'table', 'index', or 'columns'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.to_json(orient=orient, index=True)
+
+    @pytest.mark.parametrize("orient", ["split", "table"])
+    @pytest.mark.parametrize("index", [True, False])
+    def test_index_false_from_json_to_json(self, orient, index):
+        # GH25170
+        # Test index=False in from_json to_json
+        expected = DataFrame({"a": [1, 2], "b": [3, 4]})
+        dfjson = expected.to_json(orient=orient, index=index)
+        result = read_json(StringIO(dfjson), orient=orient)
+        tm.assert_frame_equal(result, expected)
+
+    def test_read_timezone_information(self):
+        # GH 25546
+        result = read_json(
+            StringIO('{"2019-01-01T11:00:00.000Z":88}'), typ="series", orient="index"
+        )
+        exp_dti = DatetimeIndex(["2019-01-01 11:00:00"], dtype="M8[ns, UTC]")
+        expected = Series([88], index=exp_dti)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "url",
+        [
+            "s3://example-fsspec/",
+            "gcs://another-fsspec/file.json",
+            "https://example-site.com/data",
+            "some-protocol://data.txt",
+        ],
+    )
+    def test_read_json_with_url_value(self, url):
+        # GH 36271
+        result = read_json(StringIO(f'{{"url":{{"0":"{url}"}}}}'))
+        expected = DataFrame({"url": [url]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "compression",
+        ["", ".gz", ".bz2", ".tar"],
+    )
+    def test_read_json_with_very_long_file_path(self, compression):
+        # GH 46718
+        long_json_path = f"{'a' * 1000}.json{compression}"
+        with pytest.raises(
+            FileNotFoundError, match=f"File {long_json_path} does not exist"
+        ):
+            # path too long for Windows is handled in file_exists() but raises in
+            # _get_data_from_filepath()
+            read_json(long_json_path)
+
+    @pytest.mark.parametrize(
+        "date_format,key", [("epoch", 86400000), ("iso", "P1DT0H0M0S")]
+    )
+    def test_timedelta_as_label(self, date_format, key):
+        df = DataFrame([[1]], columns=[pd.Timedelta("1D")])
+        expected = f'{{"{key}":{{"0":1}}}}'
+
+        expected_warning = None
+        if date_format == "epoch":
+            expected_warning = FutureWarning
+
+        msg = (
+            "'epoch' date format is deprecated and will be removed in a future "
+            "version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(expected_warning, match=msg):
+            result = df.to_json(date_format=date_format)
+
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "orient,expected",
+        [
+            ("index", "{\"('a', 'b')\":{\"('c', 'd')\":1}}"),
+            ("columns", "{\"('c', 'd')\":{\"('a', 'b')\":1}}"),
+            # TODO: the below have separate encoding procedures
+            pytest.param(
+                "split",
+                "",
+                marks=pytest.mark.xfail(
+                    reason="Produces JSON but not in a consistent manner"
+                ),
+            ),
+            pytest.param(
+                "table",
+                "",
+                marks=pytest.mark.xfail(
+                    reason="Produces JSON but not in a consistent manner"
+                ),
+            ),
+        ],
+    )
+    def test_tuple_labels(self, orient, expected):
+        # GH 20500
+        df = DataFrame([[1]], index=[("a", "b")], columns=[("c", "d")])
+        result = df.to_json(orient=orient)
+        assert result == expected
+
+    @pytest.mark.parametrize("indent", [1, 2, 4])
+    def test_to_json_indent(self, indent):
+        # GH 12004
+        df = DataFrame([["foo", "bar"], ["baz", "qux"]], columns=["a", "b"])
+
+        result = df.to_json(indent=indent)
+        spaces = " " * indent
+        expected = f"""{{
+{spaces}"a":{{
+{spaces}{spaces}"0":"foo",
+{spaces}{spaces}"1":"baz"
+{spaces}}},
+{spaces}"b":{{
+{spaces}{spaces}"0":"bar",
+{spaces}{spaces}"1":"qux"
+{spaces}}}
+}}"""
+
+        assert result == expected
+
+    @pytest.mark.skipif(
+        using_string_dtype(),
+        reason="Adjust expected when infer_string is default, no bug here, "
+        "just a complicated parametrization",
+    )
+    @pytest.mark.parametrize(
+        "orient,expected",
+        [
+            (
+                "split",
+                """{
+    "columns":[
+        "a",
+        "b"
+    ],
+    "index":[
+        0,
+        1
+    ],
+    "data":[
+        [
+            "foo",
+            "bar"
+        ],
+        [
+            "baz",
+            "qux"
+        ]
+    ]
+}""",
+            ),
+            (
+                "records",
+                """[
+    {
+        "a":"foo",
+        "b":"bar"
+    },
+    {
+        "a":"baz",
+        "b":"qux"
+    }
+]""",
+            ),
+            (
+                "index",
+                """{
+    "0":{
+        "a":"foo",
+        "b":"bar"
+    },
+    "1":{
+        "a":"baz",
+        "b":"qux"
+    }
+}""",
+            ),
+            (
+                "columns",
+                """{
+    "a":{
+        "0":"foo",
+        "1":"baz"
+    },
+    "b":{
+        "0":"bar",
+        "1":"qux"
+    }
+}""",
+            ),
+            (
+                "values",
+                """[
+    [
+        "foo",
+        "bar"
+    ],
+    [
+        "baz",
+        "qux"
+    ]
+]""",
+            ),
+            (
+                "table",
+                """{
+    "schema":{
+        "fields":[
+            {
+                "name":"index",
+                "type":"integer"
+            },
+            {
+                "name":"a",
+                "type":"string"
+            },
+            {
+                "name":"b",
+                "type":"string"
+            }
+        ],
+        "primaryKey":[
+            "index"
+        ],
+        "pandas_version":"1.4.0"
+    },
+    "data":[
+        {
+            "index":0,
+            "a":"foo",
+            "b":"bar"
+        },
+        {
+            "index":1,
+            "a":"baz",
+            "b":"qux"
+        }
+    ]
+}""",
+            ),
+        ],
+    )
+    def test_json_indent_all_orients(self, orient, expected):
+        # GH 12004
+        df = DataFrame([["foo", "bar"], ["baz", "qux"]], columns=["a", "b"])
+        result = df.to_json(orient=orient, indent=4)
+        assert result == expected
+
+    def test_json_negative_indent_raises(self):
+        with pytest.raises(ValueError, match="must be a nonnegative integer"):
+            DataFrame().to_json(indent=-1)
+
+    def test_emca_262_nan_inf_support(self):
+        # GH 12213
+        data = StringIO(
+            '["a", NaN, "NaN", Infinity, "Infinity", -Infinity, "-Infinity"]'
+        )
+        result = read_json(data)
+        expected = DataFrame(
+            ["a", None, "NaN", np.inf, "Infinity", -np.inf, "-Infinity"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_int_overflow(self):
+        # GH 30320
+        encoded_json = json.dumps([{"col": "31900441201190696999"}, {"col": "Text"}])
+        expected = DataFrame({"col": ["31900441201190696999", "Text"]})
+        result = read_json(StringIO(encoded_json))
+        tm.assert_frame_equal(result, expected)
+
+    def test_json_multiindex(self):
+        dataframe = DataFrame({"x": [1, 2, 3], "y": ["a", "b", "c"]})
+        expected = (
+            '{"(0, \'x\')":1,"(0, \'y\')":"a","(1, \'x\')":2,'
+            '"(1, \'y\')":"b","(2, \'x\')":3,"(2, \'y\')":"c"}'
+        )
+        series = dataframe.stack()
+        result = series.to_json(orient="index")
+        assert result == expected
+
+    @pytest.mark.single_cpu
+    @pytest.mark.network
+    def test_to_s3(self, s3_public_bucket, s3so):
+        # GH 28375
+        mock_bucket_name, target_file = s3_public_bucket.name, "test.json"
+        df = DataFrame({"x": [1, 2, 3], "y": [2, 4, 6]})
+        df.to_json(f"s3://{mock_bucket_name}/{target_file}", storage_options=s3so)
+        timeout = 5
+        while True:
+            if target_file in (obj.key for obj in s3_public_bucket.objects.all()):
+                break
+            time.sleep(0.1)
+            timeout -= 0.1
+            assert timeout > 0, "Timed out waiting for file to appear on moto"
+
+    def test_json_pandas_nulls(self, nulls_fixture):
+        # GH 31615
+        expected_warning = None
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        if nulls_fixture is pd.NaT:
+            expected_warning = FutureWarning
+
+        with tm.assert_produces_warning(expected_warning, match=msg):
+            result = DataFrame([[nulls_fixture]]).to_json()
+        assert result == '{"0":{"0":null}}'
+
+    def test_readjson_bool_series(self):
+        # GH31464
+        result = read_json(StringIO("[true, true, false]"), typ="series")
+        expected = Series([True, True, False])
+        tm.assert_series_equal(result, expected)
+
+    def test_to_json_multiindex_escape(self):
+        # GH 15273
+        df = DataFrame(
+            True,
+            index=date_range("2017-01-20", "2017-01-23"),
+            columns=["foo", "bar"],
+        ).stack()
+        result = df.to_json()
+        expected = (
+            "{\"(Timestamp('2017-01-20 00:00:00'), 'foo')\":true,"
+            "\"(Timestamp('2017-01-20 00:00:00'), 'bar')\":true,"
+            "\"(Timestamp('2017-01-21 00:00:00'), 'foo')\":true,"
+            "\"(Timestamp('2017-01-21 00:00:00'), 'bar')\":true,"
+            "\"(Timestamp('2017-01-22 00:00:00'), 'foo')\":true,"
+            "\"(Timestamp('2017-01-22 00:00:00'), 'bar')\":true,"
+            "\"(Timestamp('2017-01-23 00:00:00'), 'foo')\":true,"
+            "\"(Timestamp('2017-01-23 00:00:00'), 'bar')\":true}"
+        )
+        assert result == expected
+
+    def test_to_json_series_of_objects(self):
+        class _TestObject:
+            def __init__(self, a, b, _c, d) -> None:
+                self.a = a
+                self.b = b
+                self._c = _c
+                self.d = d
+
+            def e(self):
+                return 5
+
+        # JSON keys should be all non-callable non-underscore attributes, see GH-42768
+        series = Series([_TestObject(a=1, b=2, _c=3, d=4)])
+        assert json.loads(series.to_json()) == {"0": {"a": 1, "b": 2, "d": 4}}
+
+    @pytest.mark.parametrize(
+        "data,expected",
+        [
+            (
+                Series({0: -6 + 8j, 1: 0 + 1j, 2: 9 - 5j}),
+                '{"0":{"imag":8.0,"real":-6.0},'
+                '"1":{"imag":1.0,"real":0.0},'
+                '"2":{"imag":-5.0,"real":9.0}}',
+            ),
+            (
+                Series({0: -9.39 + 0.66j, 1: 3.95 + 9.32j, 2: 4.03 - 0.17j}),
+                '{"0":{"imag":0.66,"real":-9.39},'
+                '"1":{"imag":9.32,"real":3.95},'
+                '"2":{"imag":-0.17,"real":4.03}}',
+            ),
+            (
+                DataFrame([[-2 + 3j, -1 - 0j], [4 - 3j, -0 - 10j]]),
+                '{"0":{"0":{"imag":3.0,"real":-2.0},'
+                '"1":{"imag":-3.0,"real":4.0}},'
+                '"1":{"0":{"imag":0.0,"real":-1.0},'
+                '"1":{"imag":-10.0,"real":0.0}}}',
+            ),
+            (
+                DataFrame(
+                    [[-0.28 + 0.34j, -1.08 - 0.39j], [0.41 - 0.34j, -0.78 - 1.35j]]
+                ),
+                '{"0":{"0":{"imag":0.34,"real":-0.28},'
+                '"1":{"imag":-0.34,"real":0.41}},'
+                '"1":{"0":{"imag":-0.39,"real":-1.08},'
+                '"1":{"imag":-1.35,"real":-0.78}}}',
+            ),
+        ],
+    )
+    def test_complex_data_tojson(self, data, expected):
+        # GH41174
+        result = data.to_json()
+        assert result == expected
+
+    def test_json_uint64(self):
+        # GH21073
+        expected = (
+            '{"columns":["col1"],"index":[0,1],'
+            '"data":[[13342205958987758245],[12388075603347835679]]}'
+        )
+        df = DataFrame(data={"col1": [13342205958987758245, 12388075603347835679]})
+        result = df.to_json(orient="split")
+        assert result == expected
+
+    def test_read_json_dtype_backend(
+        self, string_storage, dtype_backend, orient, using_infer_string
+    ):
+        # GH#50750
+        df = DataFrame(
+            {
+                "a": Series([1, np.nan, 3], dtype="Int64"),
+                "b": Series([1, 2, 3], dtype="Int64"),
+                "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": [True, False, None],
+                "f": [True, False, True],
+                "g": ["a", "b", "c"],
+                "h": ["a", "b", None],
+            }
+        )
+
+        out = df.to_json(orient=orient)
+        with pd.option_context("mode.string_storage", string_storage):
+            result = read_json(
+                StringIO(out), dtype_backend=dtype_backend, orient=orient
+            )
+
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            string_dtype = pd.ArrowDtype(pa.string())
+        else:
+            string_dtype = pd.StringDtype(string_storage)
+
+        expected = DataFrame(
+            {
+                "a": Series([1, np.nan, 3], dtype="Int64"),
+                "b": Series([1, 2, 3], dtype="Int64"),
+                "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": Series([True, False, NA], dtype="boolean"),
+                "f": Series([True, False, True], dtype="boolean"),
+                "g": Series(["a", "b", "c"], dtype=string_dtype),
+                "h": Series(["a", "b", None], dtype=string_dtype),
+            }
+        )
+
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                    for col in expected.columns
+                }
+            )
+
+        if orient == "values":
+            expected.columns = list(range(8))
+
+        # the storage of the str columns' Index is also affected by the
+        # string_storage setting -> ignore that for checking the result
+        tm.assert_frame_equal(result, expected, check_column_type=False)
+
+    @pytest.mark.parametrize("orient", ["split", "records", "index"])
+    def test_read_json_nullable_series(self, string_storage, dtype_backend, orient):
+        # GH#50750
+        pa = pytest.importorskip("pyarrow")
+        ser = Series([1, np.nan, 3], dtype="Int64")
+
+        out = ser.to_json(orient=orient)
+        with pd.option_context("mode.string_storage", string_storage):
+            result = read_json(
+                StringIO(out), dtype_backend=dtype_backend, orient=orient, typ="series"
+            )
+
+        expected = Series([1, np.nan, 3], dtype="Int64")
+
+        if dtype_backend == "pyarrow":
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = Series(ArrowExtensionArray(pa.array(expected, from_pandas=True)))
+
+        tm.assert_series_equal(result, expected)
+
+    def test_invalid_dtype_backend(self):
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_json("test", dtype_backend="numpy")
+
+
+def test_invalid_engine():
+    # GH 48893
+    ser = Series(range(1))
+    out = ser.to_json()
+    with pytest.raises(ValueError, match="The engine type foo"):
+        read_json(out, engine="foo")
+
+
+def test_pyarrow_engine_lines_false():
+    # GH 48893
+    ser = Series(range(1))
+    out = ser.to_json()
+    with pytest.raises(ValueError, match="currently pyarrow engine only supports"):
+        read_json(out, engine="pyarrow", lines=False)
+
+
+def test_json_roundtrip_string_inference(orient):
+    df = DataFrame(
+        [["a", "b"], ["c", "d"]], index=["row 1", "row 2"], columns=["col 1", "col 2"]
+    )
+    out = df.to_json()
+    with pd.option_context("future.infer_string", True):
+        result = read_json(StringIO(out))
+    dtype = pd.StringDtype(na_value=np.nan)
+    expected = DataFrame(
+        [["a", "b"], ["c", "d"]],
+        dtype=dtype,
+        index=Index(["row 1", "row 2"], dtype=dtype),
+        columns=Index(["col 1", "col 2"], dtype=dtype),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_to_json_ea_null():
+    # GH#57224
+    df = DataFrame(
+        {
+            "a": Series([1, NA], dtype="int64[pyarrow]"),
+            "b": Series([2, NA], dtype="Int64"),
+        }
+    )
+    result = df.to_json(orient="records", lines=True)
+    expected = """{"a":1,"b":2}
+{"a":null,"b":null}
+"""
+    assert result == expected
+
+
+def test_read_json_lines_rangeindex():
+    # GH 57429
+    data = """
+{"a": 1, "b": 2}
+{"a": 3, "b": 4}
+"""
+    result = read_json(StringIO(data), lines=True).index
+    expected = RangeIndex(2)
+    tm.assert_index_equal(result, expected, exact=True)
+
+
+def test_large_number():
+    # GH#20608
+    result = read_json(
+        StringIO('["9999999999999999"]'),
+        orient="values",
+        typ="series",
+        convert_dates=False,
+    )
+    expected = Series([9999999999999999])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/io/json/test_readlines.py b/pandas/pandas/tests/io/json/test_readlines.py
new file mode 100644
index 0000000000000..d482eb5fa1a06
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_readlines.py
@@ -0,0 +1,542 @@
+from collections.abc import Iterator
+from io import StringIO
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    read_json,
+)
+import pandas._testing as tm
+
+from pandas.io.json._json import JsonReader
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture
+def lines_json_df():
+    df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+    return df.to_json(lines=True, orient="records")
+
+
+@pytest.fixture(params=["ujson", "pyarrow"])
+def engine(request):
+    if request.param == "pyarrow":
+        pytest.importorskip("pyarrow.json")
+    return request.param
+
+
+def test_read_jsonl():
+    # GH9180
+    result = read_json(StringIO('{"a": 1, "b": 2}\n{"b":2, "a" :1}\n'), lines=True)
+    expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_jsonl_engine_pyarrow(datapath, engine):
+    result = read_json(
+        datapath("io", "json", "data", "line_delimited.json"),
+        lines=True,
+        engine=engine,
+    )
+    expected = DataFrame({"a": [1, 3, 5], "b": [2, 4, 6]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_datetime(request, engine):
+    # GH33787
+    if engine == "pyarrow":
+        # GH 48893
+        reason = "Pyarrow only supports a file path as an input and line delimited json"
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    df = DataFrame(
+        [([1, 2], ["2020-03-05", "2020-04-08T09:58:49+00:00"], "hector")],
+        columns=["accounts", "date", "name"],
+    )
+    json_line = df.to_json(lines=True, orient="records")
+
+    if engine == "pyarrow":
+        result = read_json(StringIO(json_line), engine=engine)
+    else:
+        result = read_json(StringIO(json_line), engine=engine)
+    expected = DataFrame(
+        [[1, "2020-03-05", "hector"], [2, "2020-04-08T09:58:49+00:00", "hector"]],
+        columns=["accounts", "date", "name"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_jsonl_unicode_chars():
+    # GH15132: non-ascii unicode characters
+    # \u201d == RIGHT DOUBLE QUOTATION MARK
+
+    # simulate file handle
+    json = '{"a": "foo”", "b": "bar"}\n{"a": "foo", "b": "bar"}\n'
+    json = StringIO(json)
+    result = read_json(json, lines=True)
+    expected = DataFrame([["foo\u201d", "bar"], ["foo", "bar"]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+    # simulate string
+    json = '{"a": "foo”", "b": "bar"}\n{"a": "foo", "b": "bar"}\n'
+    result = read_json(StringIO(json), lines=True)
+    expected = DataFrame([["foo\u201d", "bar"], ["foo", "bar"]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_to_jsonl():
+    # GH9180
+    df = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+    result = df.to_json(orient="records", lines=True)
+    expected = '{"a":1,"b":2}\n{"a":1,"b":2}\n'
+    assert result == expected
+
+    df = DataFrame([["foo}", "bar"], ['foo"', "bar"]], columns=["a", "b"])
+    result = df.to_json(orient="records", lines=True)
+    expected = '{"a":"foo}","b":"bar"}\n{"a":"foo\\"","b":"bar"}\n'
+    assert result == expected
+    tm.assert_frame_equal(read_json(StringIO(result), lines=True), df)
+
+    # GH15096: escaped characters in columns and data
+    df = DataFrame([["foo\\", "bar"], ['foo"', "bar"]], columns=["a\\", "b"])
+    result = df.to_json(orient="records", lines=True)
+    expected = '{"a\\\\":"foo\\\\","b":"bar"}\n{"a\\\\":"foo\\"","b":"bar"}\n'
+    assert result == expected
+    tm.assert_frame_equal(read_json(StringIO(result), lines=True), df)
+
+
+def test_to_jsonl_count_new_lines():
+    # GH36888
+    df = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
+    actual_new_lines_count = df.to_json(orient="records", lines=True).count("\n")
+    expected_new_lines_count = 2
+    assert actual_new_lines_count == expected_new_lines_count
+
+
+@pytest.mark.parametrize("chunksize", [1, 1.0])
+def test_readjson_chunks(request, lines_json_df, chunksize, engine):
+    # Basic test that read_json(chunks=True) gives the same result as
+    # read_json(chunks=False)
+    # GH17048: memory usage when lines=True
+
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    unchunked = read_json(StringIO(lines_json_df), lines=True)
+    with read_json(
+        StringIO(lines_json_df), lines=True, chunksize=chunksize, engine=engine
+    ) as reader:
+        chunked = pd.concat(reader)
+
+    tm.assert_frame_equal(chunked, unchunked)
+
+
+def test_readjson_chunksize_requires_lines(lines_json_df, engine):
+    msg = "chunksize can only be passed if lines=True"
+    with pytest.raises(ValueError, match=msg):
+        with read_json(
+            StringIO(lines_json_df), lines=False, chunksize=2, engine=engine
+        ) as _:
+            pass
+
+
+def test_readjson_chunks_series(request, engine):
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason))
+
+    # Test reading line-format JSON to Series with chunksize param
+    s = pd.Series({"A": 1, "B": 2})
+
+    strio = StringIO(s.to_json(lines=True, orient="records"))
+    unchunked = read_json(strio, lines=True, typ="series", engine=engine)
+
+    strio = StringIO(s.to_json(lines=True, orient="records"))
+    with read_json(
+        strio, lines=True, typ="series", chunksize=1, engine=engine
+    ) as reader:
+        chunked = pd.concat(reader)
+
+    tm.assert_series_equal(chunked, unchunked)
+
+
+def test_readjson_each_chunk(request, lines_json_df, engine):
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    # Other tests check that the final result of read_json(chunksize=True)
+    # is correct. This checks the intermediate chunks.
+    with read_json(
+        StringIO(lines_json_df), lines=True, chunksize=2, engine=engine
+    ) as reader:
+        chunks = list(reader)
+    assert chunks[0].shape == (2, 2)
+    assert chunks[1].shape == (1, 2)
+
+
+def test_readjson_chunks_from_file(request, engine):
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    with tm.ensure_clean("test.json") as path:
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        df.to_json(path, lines=True, orient="records")
+        with read_json(path, lines=True, chunksize=1, engine=engine) as reader:
+            chunked = pd.concat(reader)
+        unchunked = read_json(path, lines=True, engine=engine)
+        tm.assert_frame_equal(unchunked, chunked)
+
+
+@pytest.mark.parametrize("chunksize", [None, 1])
+def test_readjson_chunks_closes(chunksize):
+    with tm.ensure_clean("test.json") as path:
+        df = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+        df.to_json(path, lines=True, orient="records")
+        reader = JsonReader(
+            path,
+            orient=None,
+            typ="frame",
+            dtype=True,
+            convert_axes=True,
+            convert_dates=True,
+            keep_default_dates=True,
+            precise_float=False,
+            date_unit=None,
+            encoding=None,
+            lines=True,
+            chunksize=chunksize,
+            compression=None,
+            nrows=None,
+        )
+        with reader:
+            reader.read()
+        assert reader.handles.handle.closed, (
+            f"didn't close stream with chunksize = {chunksize}"
+        )
+
+
+@pytest.mark.parametrize("chunksize", [0, -1, 2.2, "foo"])
+def test_readjson_invalid_chunksize(lines_json_df, chunksize, engine):
+    msg = r"'chunksize' must be an integer >=1"
+
+    with pytest.raises(ValueError, match=msg):
+        with read_json(
+            StringIO(lines_json_df), lines=True, chunksize=chunksize, engine=engine
+        ) as _:
+            pass
+
+
+@pytest.mark.parametrize("chunksize", [None, 1, 2])
+def test_readjson_chunks_multiple_empty_lines(chunksize):
+    j = """
+
+    {"A":1,"B":4}
+
+
+
+    {"A":2,"B":5}
+
+
+
+
+
+
+
+    {"A":3,"B":6}
+    """
+    orig = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+    test = read_json(StringIO(j), lines=True, chunksize=chunksize)
+    if chunksize is not None:
+        with test:
+            test = pd.concat(test)
+    tm.assert_frame_equal(orig, test, obj=f"chunksize: {chunksize}")
+
+
+def test_readjson_unicode(request, monkeypatch, engine):
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    with tm.ensure_clean("test.json") as path:
+        monkeypatch.setattr("locale.getpreferredencoding", lambda do_setlocale: "cp949")
+        with open(path, "w", encoding="utf-8") as f:
+            f.write('{"£©µÀÆÖÞßéöÿ":["АБВГДабвгд가"]}')
+
+        result = read_json(path, engine=engine)
+        expected = DataFrame({"£©µÀÆÖÞßéöÿ": ["АБВГДабвгд가"]})
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("nrows", [1, 2])
+def test_readjson_nrows(nrows, engine):
+    # GH 33916
+    # Test reading line-format JSON to Series with nrows param
+    jsonl = """{"a": 1, "b": 2}
+        {"a": 3, "b": 4}
+        {"a": 5, "b": 6}
+        {"a": 7, "b": 8}"""
+    result = read_json(StringIO(jsonl), lines=True, nrows=nrows)
+    expected = DataFrame({"a": [1, 3, 5, 7], "b": [2, 4, 6, 8]}).iloc[:nrows]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("nrows,chunksize", [(2, 2), (4, 2)])
+def test_readjson_nrows_chunks(request, nrows, chunksize, engine):
+    # GH 33916
+    # Test reading line-format JSON to Series with nrows and chunksize param
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    jsonl = """{"a": 1, "b": 2}
+        {"a": 3, "b": 4}
+        {"a": 5, "b": 6}
+        {"a": 7, "b": 8}"""
+
+    if engine != "pyarrow":
+        with read_json(
+            StringIO(jsonl), lines=True, nrows=nrows, chunksize=chunksize, engine=engine
+        ) as reader:
+            chunked = pd.concat(reader)
+    else:
+        with read_json(
+            jsonl, lines=True, nrows=nrows, chunksize=chunksize, engine=engine
+        ) as reader:
+            chunked = pd.concat(reader)
+    expected = DataFrame({"a": [1, 3, 5, 7], "b": [2, 4, 6, 8]}).iloc[:nrows]
+    tm.assert_frame_equal(chunked, expected)
+
+
+def test_readjson_nrows_requires_lines(engine):
+    # GH 33916
+    # Test ValueError raised if nrows is set without setting lines in read_json
+    jsonl = """{"a": 1, "b": 2}
+        {"a": 3, "b": 4}
+        {"a": 5, "b": 6}
+        {"a": 7, "b": 8}"""
+    msg = "nrows can only be passed if lines=True"
+    with pytest.raises(ValueError, match=msg):
+        read_json(jsonl, lines=False, nrows=2, engine=engine)
+
+
+def test_readjson_lines_chunks_fileurl(request, datapath, engine):
+    # GH 27135
+    # Test reading line-format JSON from file url
+    if engine == "pyarrow":
+        # GH 48893
+        reason = (
+            "Pyarrow only supports a file path as an input and line delimited json"
+            "and doesn't support chunksize parameter."
+        )
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=ValueError))
+
+    df_list_expected = [
+        DataFrame([[1, 2]], columns=["a", "b"], index=[0]),
+        DataFrame([[3, 4]], columns=["a", "b"], index=[1]),
+        DataFrame([[5, 6]], columns=["a", "b"], index=[2]),
+    ]
+    os_path = datapath("io", "json", "data", "line_delimited.json")
+    file_url = Path(os_path).as_uri()
+    with read_json(file_url, lines=True, chunksize=1, engine=engine) as url_reader:
+        for index, chuck in enumerate(url_reader):
+            tm.assert_frame_equal(chuck, df_list_expected[index])
+
+
+def test_chunksize_is_incremental():
+    # See https://github.com/pandas-dev/pandas/issues/34548
+    jsonl = (
+        """{"a": 1, "b": 2}
+        {"a": 3, "b": 4}
+        {"a": 5, "b": 6}
+        {"a": 7, "b": 8}\n"""
+        * 1000
+    )
+
+    class MyReader:
+        def __init__(self, contents) -> None:
+            self.read_count = 0
+            self.stringio = StringIO(contents)
+
+        def read(self, *args):
+            self.read_count += 1
+            return self.stringio.read(*args)
+
+        def __iter__(self) -> Iterator:
+            self.read_count += 1
+            return iter(self.stringio)
+
+    reader = MyReader(jsonl)
+    assert len(list(read_json(reader, lines=True, chunksize=100))) > 1
+    assert reader.read_count > 10
+
+
+@pytest.mark.parametrize("orient_", ["split", "index", "table"])
+def test_to_json_append_orient(orient_):
+    # GH 35849
+    # Test ValueError when orient is not 'records'
+    df = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    msg = (
+        r"mode='a' \(append\) is only supported when "
+        "lines is True and orient is 'records'"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.to_json(mode="a", orient=orient_)
+
+
+def test_to_json_append_lines():
+    # GH 35849
+    # Test ValueError when lines is not True
+    df = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    msg = (
+        r"mode='a' \(append\) is only supported when "
+        "lines is True and orient is 'records'"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.to_json(mode="a", lines=False, orient="records")
+
+
+@pytest.mark.parametrize("mode_", ["r", "x"])
+def test_to_json_append_mode(mode_):
+    # GH 35849
+    # Test ValueError when mode is not supported option
+    df = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    msg = (
+        f"mode={mode_} is not a valid option.Only 'w' and 'a' are currently supported."
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.to_json(mode=mode_, lines=False, orient="records")
+
+
+def test_to_json_append_output_consistent_columns():
+    # GH 35849
+    # Testing that resulting output reads in as expected.
+    # Testing same columns, new rows
+    df1 = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    df2 = DataFrame({"col1": [3, 4], "col2": ["c", "d"]})
+
+    expected = DataFrame({"col1": [1, 2, 3, 4], "col2": ["a", "b", "c", "d"]})
+    with tm.ensure_clean("test.json") as path:
+        # Save dataframes to the same file
+        df1.to_json(path, lines=True, orient="records")
+        df2.to_json(path, mode="a", lines=True, orient="records")
+
+        # Read path file
+        result = read_json(path, lines=True)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_to_json_append_output_inconsistent_columns():
+    # GH 35849
+    # Testing that resulting output reads in as expected.
+    # Testing one new column, one old column, new rows
+    df1 = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    df3 = DataFrame({"col2": ["e", "f"], "col3": ["!", "#"]})
+
+    expected = DataFrame(
+        {
+            "col1": [1, 2, None, None],
+            "col2": ["a", "b", "e", "f"],
+            "col3": [np.nan, np.nan, "!", "#"],
+        }
+    )
+    with tm.ensure_clean("test.json") as path:
+        # Save dataframes to the same file
+        df1.to_json(path, mode="a", lines=True, orient="records")
+        df3.to_json(path, mode="a", lines=True, orient="records")
+
+        # Read path file
+        result = read_json(path, lines=True)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_to_json_append_output_different_columns():
+    # GH 35849
+    # Testing that resulting output reads in as expected.
+    # Testing same, differing and new columns
+    df1 = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    df2 = DataFrame({"col1": [3, 4], "col2": ["c", "d"]})
+    df3 = DataFrame({"col2": ["e", "f"], "col3": ["!", "#"]})
+    df4 = DataFrame({"col4": [True, False]})
+
+    expected = DataFrame(
+        {
+            "col1": [1, 2, 3, 4, None, None, None, None],
+            "col2": ["a", "b", "c", "d", "e", "f", np.nan, np.nan],
+            "col3": [np.nan, np.nan, np.nan, np.nan, "!", "#", np.nan, np.nan],
+            "col4": [None, None, None, None, None, None, True, False],
+        }
+    ).astype({"col4": "float"})
+    with tm.ensure_clean("test.json") as path:
+        # Save dataframes to the same file
+        df1.to_json(path, mode="a", lines=True, orient="records")
+        df2.to_json(path, mode="a", lines=True, orient="records")
+        df3.to_json(path, mode="a", lines=True, orient="records")
+        df4.to_json(path, mode="a", lines=True, orient="records")
+
+        # Read path file
+        result = read_json(path, lines=True)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_to_json_append_output_different_columns_reordered():
+    # GH 35849
+    # Testing that resulting output reads in as expected.
+    # Testing specific result column order.
+    df1 = DataFrame({"col1": [1, 2], "col2": ["a", "b"]})
+    df2 = DataFrame({"col1": [3, 4], "col2": ["c", "d"]})
+    df3 = DataFrame({"col2": ["e", "f"], "col3": ["!", "#"]})
+    df4 = DataFrame({"col4": [True, False]})
+
+    # df4, df3, df2, df1 (in that order)
+    expected = DataFrame(
+        {
+            "col4": [True, False, None, None, None, None, None, None],
+            "col2": [np.nan, np.nan, "e", "f", "c", "d", "a", "b"],
+            "col3": [np.nan, np.nan, "!", "#", np.nan, np.nan, np.nan, np.nan],
+            "col1": [None, None, None, None, 3, 4, 1, 2],
+        }
+    ).astype({"col4": "float"})
+    with tm.ensure_clean("test.json") as path:
+        # Save dataframes to the same file
+        df4.to_json(path, mode="a", lines=True, orient="records")
+        df3.to_json(path, mode="a", lines=True, orient="records")
+        df2.to_json(path, mode="a", lines=True, orient="records")
+        df1.to_json(path, mode="a", lines=True, orient="records")
+
+        # Read path file
+        result = read_json(path, lines=True)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/json/test_ujson.py b/pandas/pandas/tests/io/json/test_ujson.py
new file mode 100644
index 0000000000000..d2bf9bdb139bd
--- /dev/null
+++ b/pandas/pandas/tests/io/json/test_ujson.py
@@ -0,0 +1,1043 @@
+import calendar
+import datetime
+import decimal
+import json
+import locale
+import math
+import re
+import time
+
+import dateutil
+import numpy as np
+import pytest
+
+import pandas._libs.json as ujson
+from pandas.compat import IS64
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    NaT,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def _clean_dict(d):
+    """
+    Sanitize dictionary for JSON by converting all keys to strings.
+
+    Parameters
+    ----------
+    d : dict
+        The dictionary to convert.
+
+    Returns
+    -------
+    cleaned_dict : dict
+    """
+    return {str(k): v for k, v in d.items()}
+
+
+@pytest.fixture(
+    params=[None, "split", "records", "values", "index"]  # Column indexed by default.
+)
+def orient(request):
+    return request.param
+
+
+class TestUltraJSONTests:
+    @pytest.mark.skipif(not IS64, reason="not compliant on 32-bit, xref #15865")
+    @pytest.mark.parametrize(
+        "value, double_precision",
+        [
+            ("1337.1337", 15),
+            ("0.95", 1),
+            ("0.94", 1),
+            ("1.95", 1),
+            ("-1.95", 1),
+            ("0.995", 2),
+            ("0.9995", 3),
+            ("0.99999999999999944", 15),
+        ],
+    )
+    def test_encode_decimal(self, value, double_precision):
+        sut = decimal.Decimal(value)
+        encoded = ujson.ujson_dumps(sut, double_precision=double_precision)
+        decoded = ujson.ujson_loads(encoded)
+        assert decoded == value
+
+    @pytest.mark.parametrize("ensure_ascii", [True, False])
+    def test_encode_string_conversion(self, ensure_ascii):
+        string_input = "A string \\ / \b \f \n \r \t </script> &"
+        not_html_encoded = '"A string \\\\ \\/ \\b \\f \\n \\r \\t <\\/script> &"'
+        html_encoded = (
+            '"A string \\\\ \\/ \\b \\f \\n \\r \\t \\u003c\\/script\\u003e \\u0026"'
+        )
+
+        def helper(expected_output, **encode_kwargs):
+            output = ujson.ujson_dumps(
+                string_input, ensure_ascii=ensure_ascii, **encode_kwargs
+            )
+
+            assert output == expected_output
+            assert string_input == json.loads(output)
+            assert string_input == ujson.ujson_loads(output)
+
+        # Default behavior assumes encode_html_chars=False.
+        helper(not_html_encoded)
+
+        # Make sure explicit encode_html_chars=False works.
+        helper(not_html_encoded, encode_html_chars=False)
+
+        # Make sure explicit encode_html_chars=True does the encoding.
+        helper(html_encoded, encode_html_chars=True)
+
+    @pytest.mark.parametrize(
+        "long_number", [-4342969734183514, -12345678901234.56789012, -528656961.4399388]
+    )
+    def test_double_long_numbers(self, long_number):
+        sut = {"a": long_number}
+        encoded = ujson.ujson_dumps(sut, double_precision=15)
+
+        decoded = ujson.ujson_loads(encoded)
+        assert sut == decoded
+
+    def test_encode_non_c_locale(self):
+        lc_category = locale.LC_NUMERIC
+
+        # We just need one of these locales to work.
+        for new_locale in ("it_IT.UTF-8", "Italian_Italy"):
+            if tm.can_set_locale(new_locale, lc_category):
+                with tm.set_locale(new_locale, lc_category):
+                    assert ujson.ujson_loads(ujson.ujson_dumps(4.78e60)) == 4.78e60
+                    assert ujson.ujson_loads("4.78", precise_float=True) == 4.78
+                break
+
+    def test_decimal_decode_test_precise(self):
+        sut = {"a": 4.56}
+        encoded = ujson.ujson_dumps(sut)
+        decoded = ujson.ujson_loads(encoded, precise_float=True)
+        assert sut == decoded
+
+    def test_encode_double_tiny_exponential(self):
+        num = 1e-40
+        assert num == ujson.ujson_loads(ujson.ujson_dumps(num))
+        num = 1e-100
+        assert num == ujson.ujson_loads(ujson.ujson_dumps(num))
+        num = -1e-45
+        assert num == ujson.ujson_loads(ujson.ujson_dumps(num))
+        num = -1e-145
+        assert np.allclose(num, ujson.ujson_loads(ujson.ujson_dumps(num)))
+
+    @pytest.mark.parametrize("unicode_key", ["key1", "بن"])
+    def test_encode_dict_with_unicode_keys(self, unicode_key):
+        unicode_dict = {unicode_key: "value1"}
+        assert unicode_dict == ujson.ujson_loads(ujson.ujson_dumps(unicode_dict))
+
+    @pytest.mark.parametrize("double_input", [math.pi, -math.pi])
+    def test_encode_double_conversion(self, double_input):
+        output = ujson.ujson_dumps(double_input)
+        assert round(double_input, 5) == round(json.loads(output), 5)
+        assert round(double_input, 5) == round(ujson.ujson_loads(output), 5)
+
+    def test_encode_with_decimal(self):
+        decimal_input = 1.0
+        output = ujson.ujson_dumps(decimal_input)
+
+        assert output == "1.0"
+
+    def test_encode_array_of_nested_arrays(self):
+        nested_input = [[[[]]]] * 20
+        output = ujson.ujson_dumps(nested_input)
+
+        assert nested_input == json.loads(output)
+        assert nested_input == ujson.ujson_loads(output)
+
+    def test_encode_array_of_doubles(self):
+        doubles_input = [31337.31337, 31337.31337, 31337.31337, 31337.31337] * 10
+        output = ujson.ujson_dumps(doubles_input)
+
+        assert doubles_input == json.loads(output)
+        assert doubles_input == ujson.ujson_loads(output)
+
+    def test_double_precision(self):
+        double_input = 30.012345678901234
+        output = ujson.ujson_dumps(double_input, double_precision=15)
+
+        assert double_input == json.loads(output)
+        assert double_input == ujson.ujson_loads(output)
+
+        for double_precision in (3, 9):
+            output = ujson.ujson_dumps(double_input, double_precision=double_precision)
+            rounded_input = round(double_input, double_precision)
+
+            assert rounded_input == json.loads(output)
+            assert rounded_input == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize(
+        "invalid_val",
+        [
+            20,
+            -1,
+            "9",
+            None,
+        ],
+    )
+    def test_invalid_double_precision(self, invalid_val):
+        double_input = 30.12345678901234567890
+        expected_exception = ValueError if isinstance(invalid_val, int) else TypeError
+        msg = (
+            r"Invalid value '.*' for option 'double_precision', max is '15'|"
+            r"an integer is required \(got type |"
+            r"object cannot be interpreted as an integer"
+        )
+        with pytest.raises(expected_exception, match=msg):
+            ujson.ujson_dumps(double_input, double_precision=invalid_val)
+
+    def test_encode_string_conversion2(self):
+        string_input = "A string \\ / \b \f \n \r \t"
+        output = ujson.ujson_dumps(string_input)
+
+        assert string_input == json.loads(output)
+        assert string_input == ujson.ujson_loads(output)
+        assert output == '"A string \\\\ \\/ \\b \\f \\n \\r \\t"'
+
+    @pytest.mark.parametrize(
+        "unicode_input",
+        ["Räksmörgås اسامة بن محمد بن عوض بن لادن", "\xe6\x97\xa5\xd1\x88"],
+    )
+    def test_encode_unicode_conversion(self, unicode_input):
+        enc = ujson.ujson_dumps(unicode_input)
+        dec = ujson.ujson_loads(enc)
+
+        assert enc == json.dumps(unicode_input)
+        assert dec == json.loads(enc)
+
+    def test_encode_control_escaping(self):
+        escaped_input = "\x19"
+        enc = ujson.ujson_dumps(escaped_input)
+        dec = ujson.ujson_loads(enc)
+
+        assert escaped_input == dec
+        assert enc == json.dumps(escaped_input)
+
+    def test_encode_unicode_surrogate_pair(self):
+        surrogate_input = "\xf0\x90\x8d\x86"
+        enc = ujson.ujson_dumps(surrogate_input)
+        dec = ujson.ujson_loads(enc)
+
+        assert enc == json.dumps(surrogate_input)
+        assert dec == json.loads(enc)
+
+    def test_encode_unicode_4bytes_utf8(self):
+        four_bytes_input = "\xf0\x91\x80\xb0TRAILINGNORMAL"
+        enc = ujson.ujson_dumps(four_bytes_input)
+        dec = ujson.ujson_loads(enc)
+
+        assert enc == json.dumps(four_bytes_input)
+        assert dec == json.loads(enc)
+
+    def test_encode_unicode_4bytes_utf8highest(self):
+        four_bytes_input = "\xf3\xbf\xbf\xbfTRAILINGNORMAL"
+        enc = ujson.ujson_dumps(four_bytes_input)
+
+        dec = ujson.ujson_loads(enc)
+
+        assert enc == json.dumps(four_bytes_input)
+        assert dec == json.loads(enc)
+
+    def test_encode_unicode_error(self):
+        string = "'\udac0'"
+        msg = (
+            r"'utf-8' codec can't encode character '\\udac0' "
+            r"in position 1: surrogates not allowed"
+        )
+        with pytest.raises(UnicodeEncodeError, match=msg):
+            ujson.ujson_dumps([string])
+
+    def test_encode_array_in_array(self):
+        arr_in_arr_input = [[[[]]]]
+        output = ujson.ujson_dumps(arr_in_arr_input)
+
+        assert arr_in_arr_input == json.loads(output)
+        assert output == json.dumps(arr_in_arr_input)
+        assert arr_in_arr_input == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize(
+        "num_input",
+        [
+            31337,
+            -31337,  # Negative number.
+            -9223372036854775808,  # Large negative number.
+        ],
+    )
+    def test_encode_num_conversion(self, num_input):
+        output = ujson.ujson_dumps(num_input)
+        assert num_input == json.loads(output)
+        assert output == json.dumps(num_input)
+        assert num_input == ujson.ujson_loads(output)
+
+    def test_encode_list_conversion(self):
+        list_input = [1, 2, 3, 4]
+        output = ujson.ujson_dumps(list_input)
+
+        assert list_input == json.loads(output)
+        assert list_input == ujson.ujson_loads(output)
+
+    def test_encode_dict_conversion(self):
+        dict_input = {"k1": 1, "k2": 2, "k3": 3, "k4": 4}
+        output = ujson.ujson_dumps(dict_input)
+
+        assert dict_input == json.loads(output)
+        assert dict_input == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize("builtin_value", [None, True, False])
+    def test_encode_builtin_values_conversion(self, builtin_value):
+        output = ujson.ujson_dumps(builtin_value)
+        assert builtin_value == json.loads(output)
+        assert output == json.dumps(builtin_value)
+        assert builtin_value == ujson.ujson_loads(output)
+
+    def test_encode_datetime_conversion(self):
+        datetime_input = datetime.datetime.fromtimestamp(time.time())
+        output = ujson.ujson_dumps(datetime_input, date_unit="s")
+        expected = calendar.timegm(datetime_input.utctimetuple())
+
+        assert int(expected) == json.loads(output)
+        assert int(expected) == ujson.ujson_loads(output)
+
+    def test_encode_date_conversion(self):
+        date_input = datetime.date.fromtimestamp(time.time())
+        output = ujson.ujson_dumps(date_input, date_unit="s")
+
+        tup = (date_input.year, date_input.month, date_input.day, 0, 0, 0)
+        expected = calendar.timegm(tup)
+
+        assert int(expected) == json.loads(output)
+        assert int(expected) == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize(
+        "test",
+        [datetime.time(), datetime.time(1, 2, 3), datetime.time(10, 12, 15, 343243)],
+    )
+    def test_encode_time_conversion_basic(self, test):
+        output = ujson.ujson_dumps(test)
+        expected = f'"{test.isoformat()}"'
+        assert expected == output
+
+    def test_encode_time_conversion_pytz(self):
+        # see gh-11473: to_json segfaults with timezone-aware datetimes
+        pytz = pytest.importorskip("pytz")
+        test = datetime.time(10, 12, 15, 343243, pytz.utc)
+        output = ujson.ujson_dumps(test)
+        expected = f'"{test.isoformat()}"'
+        assert expected == output
+
+    def test_encode_time_conversion_dateutil(self):
+        # see gh-11473: to_json segfaults with timezone-aware datetimes
+        test = datetime.time(10, 12, 15, 343243, dateutil.tz.tzutc())
+        output = ujson.ujson_dumps(test)
+        expected = f'"{test.isoformat()}"'
+        assert expected == output
+
+    @pytest.mark.parametrize(
+        "decoded_input", [NaT, np.datetime64("NaT"), np.nan, np.inf, -np.inf]
+    )
+    def test_encode_as_null(self, decoded_input):
+        assert ujson.ujson_dumps(decoded_input) == "null", "Expected null"
+
+    def test_datetime_units(self):
+        val = datetime.datetime(2013, 8, 17, 21, 17, 12, 215504)
+        stamp = Timestamp(val).as_unit("ns")
+
+        roundtrip = ujson.ujson_loads(ujson.ujson_dumps(val, date_unit="s"))
+        assert roundtrip == stamp._value // 10**9
+
+        roundtrip = ujson.ujson_loads(ujson.ujson_dumps(val, date_unit="ms"))
+        assert roundtrip == stamp._value // 10**6
+
+        roundtrip = ujson.ujson_loads(ujson.ujson_dumps(val, date_unit="us"))
+        assert roundtrip == stamp._value // 10**3
+
+        roundtrip = ujson.ujson_loads(ujson.ujson_dumps(val, date_unit="ns"))
+        assert roundtrip == stamp._value
+
+        msg = "Invalid value 'foo' for option 'date_unit'"
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_dumps(val, date_unit="foo")
+
+    def test_encode_to_utf8(self):
+        unencoded = "\xe6\x97\xa5\xd1\x88"
+
+        enc = ujson.ujson_dumps(unencoded, ensure_ascii=False)
+        dec = ujson.ujson_loads(enc)
+
+        assert enc == json.dumps(unencoded, ensure_ascii=False)
+        assert dec == json.loads(enc)
+
+    def test_decode_from_unicode(self):
+        unicode_input = '{"obj": 31337}'
+
+        dec1 = ujson.ujson_loads(unicode_input)
+        dec2 = ujson.ujson_loads(str(unicode_input))
+
+        assert dec1 == dec2
+
+    def test_encode_recursion_max(self):
+        # 8 is the max recursion depth
+
+        class O2:
+            member = 0
+
+        class O1:
+            member = 0
+
+        decoded_input = O1()
+        decoded_input.member = O2()
+        decoded_input.member.member = decoded_input
+
+        with pytest.raises(OverflowError, match="Maximum recursion level reached"):
+            ujson.ujson_dumps(decoded_input)
+
+    def test_decode_jibberish(self):
+        jibberish = "fdsa sda v9sa fdsa"
+        msg = "Unexpected character found when decoding 'false'"
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_loads(jibberish)
+
+    @pytest.mark.parametrize(
+        "broken_json",
+        [
+            "[",  # Broken array start.
+            "{",  # Broken object start.
+            "]",  # Broken array end.
+            "}",  # Broken object end.
+        ],
+    )
+    def test_decode_broken_json(self, broken_json):
+        msg = "Expected object or value"
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_loads(broken_json)
+
+    @pytest.mark.parametrize("too_big_char", ["[", "{"])
+    def test_decode_depth_too_big(self, too_big_char):
+        with pytest.raises(ValueError, match="Reached object decoding depth limit"):
+            ujson.ujson_loads(too_big_char * (1024 * 1024))
+
+    @pytest.mark.parametrize(
+        "bad_string",
+        [
+            '"TESTING',  # Unterminated.
+            '"TESTING\\"',  # Unterminated escape.
+            "tru",  # Broken True.
+            "fa",  # Broken False.
+            "n",  # Broken None.
+        ],
+    )
+    def test_decode_bad_string(self, bad_string):
+        msg = (
+            "Unexpected character found when decoding|"
+            "Unmatched ''\"' when when decoding 'string'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_loads(bad_string)
+
+    @pytest.mark.parametrize(
+        "broken_json, err_msg",
+        [
+            (
+                '{{1337:""}}',
+                "Key name of object must be 'string' when decoding 'object'",
+            ),
+            ('{{"key":"}', "Unmatched ''\"' when when decoding 'string'"),
+            ("[[[true", "Unexpected character found when decoding array value (2)"),
+        ],
+    )
+    def test_decode_broken_json_leak(self, broken_json, err_msg):
+        for _ in range(1000):
+            with pytest.raises(ValueError, match=re.escape(err_msg)):
+                ujson.ujson_loads(broken_json)
+
+    @pytest.mark.parametrize(
+        "invalid_dict",
+        [
+            "{{{{31337}}}}",  # No key.
+            '{{{{"key":}}}}',  # No value.
+            '{{{{"key"}}}}',  # No colon or value.
+        ],
+    )
+    def test_decode_invalid_dict(self, invalid_dict):
+        msg = (
+            "Key name of object must be 'string' when decoding 'object'|"
+            "No ':' found when decoding object value|"
+            "Expected object or value"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_loads(invalid_dict)
+
+    @pytest.mark.parametrize("numeric_int_as_str", ["31337", "-31337"])
+    def test_decode_numeric_int(self, numeric_int_as_str):
+        assert int(numeric_int_as_str) == ujson.ujson_loads(numeric_int_as_str)
+
+    def test_encode_null_character(self):
+        wrapped_input = "31337 \x00 1337"
+        output = ujson.ujson_dumps(wrapped_input)
+
+        assert wrapped_input == json.loads(output)
+        assert output == json.dumps(wrapped_input)
+        assert wrapped_input == ujson.ujson_loads(output)
+
+        alone_input = "\x00"
+        output = ujson.ujson_dumps(alone_input)
+
+        assert alone_input == json.loads(output)
+        assert output == json.dumps(alone_input)
+        assert alone_input == ujson.ujson_loads(output)
+        assert '"  \\u0000\\r\\n "' == ujson.ujson_dumps("  \u0000\r\n ")
+
+    def test_decode_null_character(self):
+        wrapped_input = '"31337 \\u0000 31337"'
+        assert ujson.ujson_loads(wrapped_input) == json.loads(wrapped_input)
+
+    def test_encode_list_long_conversion(self):
+        long_input = [
+            9223372036854775807,
+            9223372036854775807,
+            9223372036854775807,
+            9223372036854775807,
+            9223372036854775807,
+            9223372036854775807,
+        ]
+        output = ujson.ujson_dumps(long_input)
+
+        assert long_input == json.loads(output)
+        assert long_input == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize("long_input", [9223372036854775807, 18446744073709551615])
+    def test_encode_long_conversion(self, long_input):
+        output = ujson.ujson_dumps(long_input)
+
+        assert long_input == json.loads(output)
+        assert output == json.dumps(long_input)
+        assert long_input == ujson.ujson_loads(output)
+
+    @pytest.mark.parametrize("bigNum", [2**64, -(2**63) - 1])
+    def test_dumps_ints_larger_than_maxsize(self, bigNum):
+        encoding = ujson.ujson_dumps(bigNum)
+        assert str(bigNum) == encoding
+
+        with pytest.raises(
+            ValueError,
+            match="Value is too big|Value is too small",
+        ):
+            assert ujson.ujson_loads(encoding) == bigNum
+
+    @pytest.mark.parametrize(
+        "int_exp", ["1337E40", "1.337E40", "1337E+9", "1.337e+40", "1.337E-4"]
+    )
+    def test_decode_numeric_int_exp(self, int_exp):
+        assert ujson.ujson_loads(int_exp) == json.loads(int_exp)
+
+    def test_loads_non_str_bytes_raises(self):
+        msg = "a bytes-like object is required, not 'NoneType'"
+        with pytest.raises(TypeError, match=msg):
+            ujson.ujson_loads(None)
+
+    @pytest.mark.parametrize("val", [3590016419, 2**31, 2**32, (2**32) - 1])
+    def test_decode_number_with_32bit_sign_bit(self, val):
+        # Test that numbers that fit within 32 bits but would have the
+        # sign bit set (2**31 <= x < 2**32) are decoded properly.
+        doc = f'{{"id": {val}}}'
+        assert ujson.ujson_loads(doc)["id"] == val
+
+    def test_encode_big_escape(self):
+        # Make sure no Exception is raised.
+        for _ in range(10):
+            base = "\u00e5".encode()
+            escape_input = base * 1024 * 1024 * 2
+            ujson.ujson_dumps(escape_input)
+
+    def test_decode_big_escape(self):
+        # Make sure no Exception is raised.
+        for _ in range(10):
+            base = "\u00e5".encode()
+            quote = b'"'
+
+            escape_input = quote + (base * 1024 * 1024 * 2) + quote
+            ujson.ujson_loads(escape_input)
+
+    def test_to_dict(self):
+        d = {"key": 31337}
+
+        class DictTest:
+            def toDict(self):
+                return d
+
+        o = DictTest()
+        output = ujson.ujson_dumps(o)
+
+        dec = ujson.ujson_loads(output)
+        assert dec == d
+
+    def test_default_handler(self):
+        class _TestObject:
+            def __init__(self, val) -> None:
+                self.val = val
+
+            @property
+            def recursive_attr(self):
+                return _TestObject("recursive_attr")
+
+            def __str__(self) -> str:
+                return str(self.val)
+
+        msg = "Maximum recursion level reached"
+        with pytest.raises(OverflowError, match=msg):
+            ujson.ujson_dumps(_TestObject("foo"))
+        assert '"foo"' == ujson.ujson_dumps(_TestObject("foo"), default_handler=str)
+
+        def my_handler(_):
+            return "foobar"
+
+        assert '"foobar"' == ujson.ujson_dumps(
+            _TestObject("foo"), default_handler=my_handler
+        )
+
+        def my_handler_raises(_):
+            raise TypeError("I raise for anything")
+
+        with pytest.raises(TypeError, match="I raise for anything"):
+            ujson.ujson_dumps(_TestObject("foo"), default_handler=my_handler_raises)
+
+        def my_int_handler(_):
+            return 42
+
+        assert (
+            ujson.ujson_loads(
+                ujson.ujson_dumps(_TestObject("foo"), default_handler=my_int_handler)
+            )
+            == 42
+        )
+
+        def my_obj_handler(_):
+            return datetime.datetime(2013, 2, 3)
+
+        assert ujson.ujson_loads(
+            ujson.ujson_dumps(datetime.datetime(2013, 2, 3))
+        ) == ujson.ujson_loads(
+            ujson.ujson_dumps(_TestObject("foo"), default_handler=my_obj_handler)
+        )
+
+        obj_list = [_TestObject("foo"), _TestObject("bar")]
+        assert json.loads(json.dumps(obj_list, default=str)) == ujson.ujson_loads(
+            ujson.ujson_dumps(obj_list, default_handler=str)
+        )
+
+    def test_encode_object(self):
+        class _TestObject:
+            def __init__(self, a, b, _c, d) -> None:
+                self.a = a
+                self.b = b
+                self._c = _c
+                self.d = d
+
+            def e(self):
+                return 5
+
+        # JSON keys should be all non-callable non-underscore attributes, see GH-42768
+        test_object = _TestObject(a=1, b=2, _c=3, d=4)
+        assert ujson.ujson_loads(ujson.ujson_dumps(test_object)) == {
+            "a": 1,
+            "b": 2,
+            "d": 4,
+        }
+
+    def test_ujson__name__(self):
+        # GH 52898
+        assert ujson.__name__ == "pandas._libs.json"
+
+
+class TestNumpyJSONTests:
+    @pytest.mark.parametrize("bool_input", [True, False])
+    def test_bool(self, bool_input):
+        b = bool(bool_input)
+        assert ujson.ujson_loads(ujson.ujson_dumps(b)) == b
+
+    def test_bool_array(self):
+        bool_array = np.array(
+            [True, False, True, True, False, True, False, False], dtype=bool
+        )
+        output = np.array(ujson.ujson_loads(ujson.ujson_dumps(bool_array)), dtype=bool)
+        tm.assert_numpy_array_equal(bool_array, output)
+
+    def test_int(self, any_int_numpy_dtype):
+        klass = np.dtype(any_int_numpy_dtype).type
+        num = klass(1)
+
+        assert klass(ujson.ujson_loads(ujson.ujson_dumps(num))) == num
+
+    def test_int_array(self, any_int_numpy_dtype):
+        arr = np.arange(100, dtype=int)
+        arr_input = arr.astype(any_int_numpy_dtype)
+
+        arr_output = np.array(
+            ujson.ujson_loads(ujson.ujson_dumps(arr_input)), dtype=any_int_numpy_dtype
+        )
+        tm.assert_numpy_array_equal(arr_input, arr_output)
+
+    def test_int_max(self, any_int_numpy_dtype):
+        if any_int_numpy_dtype in ("int64", "uint64") and not IS64:
+            pytest.skip("Cannot test 64-bit integer on 32-bit platform")
+
+        klass = np.dtype(any_int_numpy_dtype).type
+
+        # uint64 max will always overflow,
+        # as it's encoded to signed.
+        if any_int_numpy_dtype == "uint64":
+            num = np.iinfo("int64").max
+        else:
+            num = np.iinfo(any_int_numpy_dtype).max
+
+        assert klass(ujson.ujson_loads(ujson.ujson_dumps(num))) == num
+
+    def test_float(self, float_numpy_dtype):
+        klass = np.dtype(float_numpy_dtype).type
+        num = klass(256.2013)
+
+        assert klass(ujson.ujson_loads(ujson.ujson_dumps(num))) == num
+
+    def test_float_array(self, float_numpy_dtype):
+        arr = np.arange(12.5, 185.72, 1.7322, dtype=float)
+        float_input = arr.astype(float_numpy_dtype)
+
+        float_output = np.array(
+            ujson.ujson_loads(ujson.ujson_dumps(float_input, double_precision=15)),
+            dtype=float_numpy_dtype,
+        )
+        tm.assert_almost_equal(float_input, float_output)
+
+    def test_float_max(self, float_numpy_dtype):
+        klass = np.dtype(float_numpy_dtype).type
+        num = klass(np.finfo(float_numpy_dtype).max / 10)
+
+        tm.assert_almost_equal(
+            klass(ujson.ujson_loads(ujson.ujson_dumps(num, double_precision=15))), num
+        )
+
+    def test_array_basic(self):
+        arr = np.arange(96)
+        arr = arr.reshape((2, 2, 2, 2, 3, 2))
+
+        tm.assert_numpy_array_equal(
+            np.array(ujson.ujson_loads(ujson.ujson_dumps(arr))), arr
+        )
+
+    @pytest.mark.parametrize("shape", [(10, 10), (5, 5, 4), (100, 1)])
+    def test_array_reshaped(self, shape):
+        arr = np.arange(100)
+        arr = arr.reshape(shape)
+
+        tm.assert_numpy_array_equal(
+            np.array(ujson.ujson_loads(ujson.ujson_dumps(arr))), arr
+        )
+
+    def test_array_list(self):
+        arr_list = [
+            "a",
+            [],
+            {},
+            {},
+            [],
+            42,
+            97.8,
+            ["a", "b"],
+            {"key": "val"},
+        ]
+        arr = np.array(arr_list, dtype=object)
+        result = np.array(ujson.ujson_loads(ujson.ujson_dumps(arr)), dtype=object)
+        tm.assert_numpy_array_equal(result, arr)
+
+    def test_array_float(self):
+        dtype = np.float32
+
+        arr = np.arange(100.202, 200.202, 1, dtype=dtype)
+        arr = arr.reshape((5, 5, 4))
+
+        arr_out = np.array(ujson.ujson_loads(ujson.ujson_dumps(arr)), dtype=dtype)
+        tm.assert_almost_equal(arr, arr_out)
+
+    def test_0d_array(self):
+        # gh-18878
+        msg = re.escape(
+            "array(1) (numpy-scalar) is not JSON serializable at the moment"
+        )
+        with pytest.raises(TypeError, match=msg):
+            ujson.ujson_dumps(np.array(1))
+
+    def test_array_long_double(self):
+        msg = re.compile(
+            "1234.5.* \\(numpy-scalar\\) is not JSON serializable at the moment"
+        )
+        with pytest.raises(TypeError, match=msg):
+            ujson.ujson_dumps(np.longdouble(1234.5))
+
+
+class TestPandasJSONTests:
+    def test_dataframe(self, orient):
+        dtype = np.int64
+
+        df = DataFrame(
+            [[1, 2, 3], [4, 5, 6]],
+            index=["a", "b"],
+            columns=["x", "y", "z"],
+            dtype=dtype,
+        )
+        encode_kwargs = {} if orient is None else {"orient": orient}
+        assert (df.dtypes == dtype).all()
+
+        output = ujson.ujson_loads(ujson.ujson_dumps(df, **encode_kwargs))
+        assert (df.dtypes == dtype).all()
+
+        # Ensure proper DataFrame initialization.
+        if orient == "split":
+            dec = _clean_dict(output)
+            output = DataFrame(**dec)
+        else:
+            output = DataFrame(output)
+
+        # Corrections to enable DataFrame comparison.
+        if orient == "values":
+            df.columns = [0, 1, 2]
+            df.index = [0, 1]
+        elif orient == "records":
+            df.index = [0, 1]
+        elif orient == "index":
+            df = df.transpose()
+
+        assert (df.dtypes == dtype).all()
+        tm.assert_frame_equal(output, df)
+
+    def test_dataframe_nested(self, orient):
+        df = DataFrame(
+            [[1, 2, 3], [4, 5, 6]], index=["a", "b"], columns=["x", "y", "z"]
+        )
+
+        nested = {"df1": df, "df2": df.copy()}
+        kwargs = {} if orient is None else {"orient": orient}
+
+        exp = {
+            "df1": ujson.ujson_loads(ujson.ujson_dumps(df, **kwargs)),
+            "df2": ujson.ujson_loads(ujson.ujson_dumps(df, **kwargs)),
+        }
+        assert ujson.ujson_loads(ujson.ujson_dumps(nested, **kwargs)) == exp
+
+    def test_series(self, orient):
+        dtype = np.int64
+        s = Series(
+            [10, 20, 30, 40, 50, 60],
+            name="series",
+            index=[6, 7, 8, 9, 10, 15],
+            dtype=dtype,
+        ).sort_values()
+        assert s.dtype == dtype
+
+        encode_kwargs = {} if orient is None else {"orient": orient}
+
+        output = ujson.ujson_loads(ujson.ujson_dumps(s, **encode_kwargs))
+        assert s.dtype == dtype
+
+        if orient == "split":
+            dec = _clean_dict(output)
+            output = Series(**dec)
+        else:
+            output = Series(output)
+
+        if orient in (None, "index"):
+            s.name = None
+            output = output.sort_values()
+            s.index = ["6", "7", "8", "9", "10", "15"]
+        elif orient in ("records", "values"):
+            s.name = None
+            s.index = [0, 1, 2, 3, 4, 5]
+
+        assert s.dtype == dtype
+        tm.assert_series_equal(output, s)
+
+    def test_series_nested(self, orient):
+        s = Series(
+            [10, 20, 30, 40, 50, 60], name="series", index=[6, 7, 8, 9, 10, 15]
+        ).sort_values()
+        nested = {"s1": s, "s2": s.copy()}
+        kwargs = {} if orient is None else {"orient": orient}
+
+        exp = {
+            "s1": ujson.ujson_loads(ujson.ujson_dumps(s, **kwargs)),
+            "s2": ujson.ujson_loads(ujson.ujson_dumps(s, **kwargs)),
+        }
+        assert ujson.ujson_loads(ujson.ujson_dumps(nested, **kwargs)) == exp
+
+    def test_index(self):
+        i = Index([23, 45, 18, 98, 43, 11], name="index")
+
+        # Column indexed.
+        output = Index(ujson.ujson_loads(ujson.ujson_dumps(i)), name="index")
+        tm.assert_index_equal(i, output)
+
+        dec = _clean_dict(ujson.ujson_loads(ujson.ujson_dumps(i, orient="split")))
+        output = Index(**dec)
+
+        tm.assert_index_equal(i, output)
+        assert i.name == output.name
+
+        tm.assert_index_equal(i, output)
+        assert i.name == output.name
+
+        output = Index(
+            ujson.ujson_loads(ujson.ujson_dumps(i, orient="values")), name="index"
+        )
+        tm.assert_index_equal(i, output)
+
+        output = Index(
+            ujson.ujson_loads(ujson.ujson_dumps(i, orient="records")), name="index"
+        )
+        tm.assert_index_equal(i, output)
+
+        output = Index(
+            ujson.ujson_loads(ujson.ujson_dumps(i, orient="index")), name="index"
+        )
+        tm.assert_index_equal(i, output)
+
+    def test_datetime_index(self):
+        date_unit = "ns"
+
+        # freq doesn't round-trip
+        rng = DatetimeIndex(list(date_range("1/1/2000", periods=20)), freq=None)
+        encoded = ujson.ujson_dumps(rng, date_unit=date_unit)
+
+        decoded = DatetimeIndex(np.array(ujson.ujson_loads(encoded)))
+        tm.assert_index_equal(rng, decoded)
+
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+        decoded = Series(ujson.ujson_loads(ujson.ujson_dumps(ts, date_unit=date_unit)))
+
+        idx_values = decoded.index.values.astype(np.int64)
+        decoded.index = DatetimeIndex(idx_values)
+        tm.assert_series_equal(ts, decoded)
+
+    @pytest.mark.parametrize(
+        "invalid_arr",
+        [
+            "[31337,]",  # Trailing comma.
+            "[,31337]",  # Leading comma.
+            "[]]",  # Unmatched bracket.
+            "[,]",  # Only comma.
+        ],
+    )
+    def test_decode_invalid_array(self, invalid_arr):
+        msg = (
+            "Expected object or value|Trailing data|"
+            "Unexpected character found when decoding array value"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ujson.ujson_loads(invalid_arr)
+
+    @pytest.mark.parametrize("arr", [[], [31337]])
+    def test_decode_array(self, arr):
+        assert arr == ujson.ujson_loads(str(arr))
+
+    @pytest.mark.parametrize("extreme_num", [9223372036854775807, -9223372036854775808])
+    def test_decode_extreme_numbers(self, extreme_num):
+        assert extreme_num == ujson.ujson_loads(str(extreme_num))
+
+    @pytest.mark.parametrize("too_extreme_num", [f"{2**64}", f"{-(2**63) - 1}"])
+    def test_decode_too_extreme_numbers(self, too_extreme_num):
+        with pytest.raises(
+            ValueError,
+            match="Value is too big|Value is too small",
+        ):
+            ujson.ujson_loads(too_extreme_num)
+
+    def test_decode_with_trailing_whitespaces(self):
+        assert {} == ujson.ujson_loads("{}\n\t ")
+
+    def test_decode_with_trailing_non_whitespaces(self):
+        with pytest.raises(ValueError, match="Trailing data"):
+            ujson.ujson_loads("{}\n\t a")
+
+    @pytest.mark.parametrize("value", [f"{2**64}", f"{-(2**63) - 1}"])
+    def test_decode_array_with_big_int(self, value):
+        with pytest.raises(
+            ValueError,
+            match="Value is too big|Value is too small",
+        ):
+            ujson.ujson_loads(value)
+
+    @pytest.mark.parametrize(
+        "float_number",
+        [
+            1.1234567893,
+            1.234567893,
+            1.34567893,
+            1.4567893,
+            1.567893,
+            1.67893,
+            1.7893,
+            1.893,
+            1.3,
+        ],
+    )
+    @pytest.mark.parametrize("sign", [-1, 1])
+    def test_decode_floating_point(self, sign, float_number):
+        float_number *= sign
+        tm.assert_almost_equal(
+            float_number, ujson.ujson_loads(str(float_number)), rtol=1e-15
+        )
+
+    def test_encode_big_set(self):
+        s = set(range(100000))
+        # Make sure no Exception is raised.
+        ujson.ujson_dumps(s)
+
+    def test_encode_empty_set(self):
+        assert "[]" == ujson.ujson_dumps(set())
+
+    def test_encode_set(self):
+        s = {1, 2, 3, 4, 5, 6, 7, 8, 9}
+        enc = ujson.ujson_dumps(s)
+        dec = ujson.ujson_loads(enc)
+
+        for v in dec:
+            assert v in s
+
+    @pytest.mark.parametrize(
+        "td",
+        [
+            Timedelta(days=366),
+            Timedelta(days=-1),
+            Timedelta(hours=13, minutes=5, seconds=5),
+            Timedelta(hours=13, minutes=20, seconds=30),
+            Timedelta(days=-1, nanoseconds=5),
+            Timedelta(nanoseconds=1),
+            Timedelta(microseconds=1, nanoseconds=1),
+            Timedelta(milliseconds=1, microseconds=1, nanoseconds=1),
+            Timedelta(milliseconds=999, microseconds=999, nanoseconds=999),
+        ],
+    )
+    def test_encode_timedelta_iso(self, td):
+        # GH 28256
+        result = ujson.ujson_dumps(td, iso_dates=True)
+        expected = f'"{td.isoformat()}"'
+
+        assert result == expected
+
+    def test_encode_periodindex(self):
+        # GH 46683
+        p = PeriodIndex(["2022-04-06", "2022-04-07"], freq="D")
+        df = DataFrame(index=p)
+        assert df.to_json() == "{}"
diff --git a/pandas/pandas/tests/io/parser/__init__.py b/pandas/pandas/tests/io/parser/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/parser/common/__init__.py b/pandas/pandas/tests/io/parser/common/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/parser/common/test_chunksize.py b/pandas/pandas/tests/io/parser/common/test_chunksize.py
new file mode 100644
index 0000000000000..65ad7273666e5
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_chunksize.py
@@ -0,0 +1,383 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas._libs import parsers as libparsers
+from pandas.errors import DtypeWarning
+
+from pandas import (
+    DataFrame,
+    concat,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.mark.parametrize("index_col", [0, "index"])
+def test_read_chunksize_with_index(all_parsers, index_col):
+    parser = all_parsers
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+
+    expected = DataFrame(
+        [
+            ["foo", 2, 3, 4, 5],
+            ["bar", 7, 8, 9, 10],
+            ["baz", 12, 13, 14, 15],
+            ["qux", 12, 13, 14, 15],
+            ["foo2", 12, 13, 14, 15],
+            ["bar2", 12, 13, 14, 15],
+        ],
+        columns=["index", "A", "B", "C", "D"],
+    )
+    expected = expected.set_index("index")
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            with parser.read_csv(StringIO(data), index_col=0, chunksize=2) as reader:
+                list(reader)
+        return
+
+    with parser.read_csv(StringIO(data), index_col=0, chunksize=2) as reader:
+        chunks = list(reader)
+    tm.assert_frame_equal(chunks[0], expected[:2])
+    tm.assert_frame_equal(chunks[1], expected[2:4])
+    tm.assert_frame_equal(chunks[2], expected[4:])
+
+
+@pytest.mark.parametrize("chunksize", [1.3, "foo", 0])
+def test_read_chunksize_bad(all_parsers, chunksize):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    msg = r"'chunksize' must be an integer >=1"
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+
+    with pytest.raises(ValueError, match=msg):
+        with parser.read_csv(StringIO(data), chunksize=chunksize) as _:
+            pass
+
+
+@pytest.mark.parametrize("chunksize", [2, 8])
+def test_read_chunksize_and_nrows(all_parsers, chunksize):
+    # see gh-15755
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    kwargs = {"index_col": 0, "nrows": 5}
+
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+        return
+
+    expected = parser.read_csv(StringIO(data), **kwargs)
+    with parser.read_csv(StringIO(data), chunksize=chunksize, **kwargs) as reader:
+        tm.assert_frame_equal(concat(reader), expected)
+
+
+def test_read_chunksize_and_nrows_changing_size(all_parsers):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    kwargs = {"index_col": 0, "nrows": 5}
+
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+        return
+
+    expected = parser.read_csv(StringIO(data), **kwargs)
+    with parser.read_csv(StringIO(data), chunksize=8, **kwargs) as reader:
+        tm.assert_frame_equal(reader.get_chunk(size=2), expected.iloc[:2])
+        tm.assert_frame_equal(reader.get_chunk(size=4), expected.iloc[2:5])
+
+        with pytest.raises(StopIteration, match=""):
+            reader.get_chunk(size=3)
+
+
+def test_get_chunk_passed_chunksize(all_parsers):
+    parser = all_parsers
+    data = """A,B,C
+1,2,3
+4,5,6
+7,8,9
+1,2,3"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            with parser.read_csv(StringIO(data), chunksize=2) as reader:
+                reader.get_chunk()
+        return
+
+    with parser.read_csv(StringIO(data), chunksize=2) as reader:
+        result = reader.get_chunk()
+
+    expected = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kwargs", [{}, {"index_col": 0}])
+def test_read_chunksize_compat(all_parsers, kwargs):
+    # see gh-12185
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), **kwargs)
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            with parser.read_csv(StringIO(data), chunksize=2, **kwargs) as reader:
+                concat(reader)
+        return
+
+    with parser.read_csv(StringIO(data), chunksize=2, **kwargs) as reader:
+        via_reader = concat(reader)
+    tm.assert_frame_equal(via_reader, result)
+
+
+def test_read_chunksize_jagged_names(all_parsers):
+    # see gh-23509
+    parser = all_parsers
+    data = "\n".join(["0"] * 7 + [",".join(["0"] * 10)])
+
+    expected = DataFrame([[0] + [np.nan] * 9] * 7 + [[0] * 10])
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            with parser.read_csv(
+                StringIO(data), names=range(10), chunksize=4
+            ) as reader:
+                concat(reader)
+        return
+
+    with parser.read_csv(StringIO(data), names=range(10), chunksize=4) as reader:
+        result = concat(reader)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_chunk_begins_with_newline_whitespace(all_parsers):
+    # see gh-10022
+    parser = all_parsers
+    data = "\n hello\nworld\n"
+
+    result = parser.read_csv(StringIO(data), header=None)
+    expected = DataFrame([" hello", "world"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.slow
+def test_chunks_have_consistent_numerical_type(all_parsers, monkeypatch):
+    # mainly an issue with the C parser
+    heuristic = 2**3
+    parser = all_parsers
+    integers = [str(i) for i in range(heuristic - 1)]
+    data = "a\n" + "\n".join(integers + ["1.0", "2.0"] + integers)
+
+    # Coercions should work without warnings.
+    with monkeypatch.context() as m:
+        m.setattr(libparsers, "DEFAULT_BUFFER_HEURISTIC", heuristic)
+        result = parser.read_csv(StringIO(data))
+
+    assert type(result.a[0]) is np.float64
+    assert result.a.dtype == float
+
+
+def test_warn_if_chunks_have_mismatched_type(all_parsers, using_infer_string):
+    warning_type = None
+    parser = all_parsers
+    size = 10000
+
+    # see gh-3866: if chunks are different types and can't
+    # be coerced using numerical types, then issue warning.
+    if parser.engine == "c" and parser.low_memory:
+        warning_type = DtypeWarning
+        # Use larger size to hit warning path
+        size = 499999
+
+    integers = [str(i) for i in range(size)]
+    data = "a\n" + "\n".join(integers + ["a", "b"] + integers)
+
+    buf = StringIO(data)
+
+    if parser.engine == "pyarrow":
+        df = parser.read_csv(
+            buf,
+        )
+    else:
+        df = parser.read_csv_check_warnings(
+            warning_type,
+            r"Columns \(0: a\) have mixed types. "
+            "Specify dtype option on import or set low_memory=False.",
+            buf,
+        )
+    if parser.engine == "c" and parser.low_memory:
+        assert df.a.dtype == object
+    elif using_infer_string:
+        assert df.a.dtype == "str"
+    else:
+        assert df.a.dtype == object
+
+
+@pytest.mark.parametrize("iterator", [True, False])
+def test_empty_with_nrows_chunksize(all_parsers, iterator):
+    # see gh-9535
+    parser = all_parsers
+    expected = DataFrame(columns=["foo", "bar"])
+
+    nrows = 10
+    data = StringIO("foo,bar\n")
+
+    if parser.engine == "pyarrow":
+        msg = (
+            "The '(nrows|chunksize)' option is not supported with the 'pyarrow' engine"
+        )
+        with pytest.raises(ValueError, match=msg):
+            if iterator:
+                with parser.read_csv(data, chunksize=nrows) as reader:
+                    next(iter(reader))
+            else:
+                parser.read_csv(data, nrows=nrows)
+        return
+
+    if iterator:
+        with parser.read_csv(data, chunksize=nrows) as reader:
+            result = next(iter(reader))
+    else:
+        result = parser.read_csv(data, nrows=nrows)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_memory_growth_chunksize(all_parsers):
+    # see gh-24805
+    #
+    # Let's just make sure that we don't crash
+    # as we iteratively process all chunks.
+    parser = all_parsers
+
+    with tm.ensure_clean() as path:
+        with open(path, "w", encoding="utf-8") as f:
+            for i in range(1000):
+                f.write(str(i) + "\n")
+
+        if parser.engine == "pyarrow":
+            msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                with parser.read_csv(path, chunksize=20) as result:
+                    for _ in result:
+                        pass
+            return
+
+        with parser.read_csv(path, chunksize=20) as result:
+            for _ in result:
+                pass
+
+
+def test_chunksize_with_usecols_second_block_shorter(all_parsers):
+    # GH#21211
+    parser = all_parsers
+    data = """1,2,3,4
+5,6,7,8
+9,10,11
+"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                names=["a", "b"],
+                chunksize=2,
+                usecols=[0, 1],
+                header=None,
+            )
+        return
+
+    result_chunks = parser.read_csv(
+        StringIO(data),
+        names=["a", "b"],
+        chunksize=2,
+        usecols=[0, 1],
+        header=None,
+    )
+
+    expected_frames = [
+        DataFrame({"a": [1, 5], "b": [2, 6]}),
+        DataFrame({"a": [9], "b": [10]}, index=[2]),
+    ]
+
+    for i, result in enumerate(result_chunks):
+        tm.assert_frame_equal(result, expected_frames[i])
+
+
+def test_chunksize_second_block_shorter(all_parsers):
+    # GH#21211
+    parser = all_parsers
+    data = """a,b,c,d
+1,2,3,4
+5,6,7,8
+9,10,11
+"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), chunksize=2)
+        return
+
+    result_chunks = parser.read_csv(StringIO(data), chunksize=2)
+
+    expected_frames = [
+        DataFrame({"a": [1, 5], "b": [2, 6], "c": [3, 7], "d": [4, 8]}),
+        DataFrame({"a": [9], "b": [10], "c": [11], "d": [np.nan]}, index=[2]),
+    ]
+
+    for i, result in enumerate(result_chunks):
+        tm.assert_frame_equal(result, expected_frames[i])
diff --git a/pandas/pandas/tests/io/parser/common/test_common_basic.py b/pandas/pandas/tests/io/parser/common/test_common_basic.py
new file mode 100644
index 0000000000000..3680273f5e98a
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_common_basic.py
@@ -0,0 +1,832 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from datetime import datetime
+from inspect import signature
+from io import StringIO
+import os
+from pathlib import Path
+import sys
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import HAS_PYARROW
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+    ParserWarning,
+)
+
+from pandas import (
+    DataFrame,
+    Index,
+    compat,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_read_csv_local(all_parsers, csv1):
+    prefix = "file:///" if compat.is_platform_windows() else "file://"
+    parser = all_parsers
+
+    fname = prefix + str(os.path.abspath(csv1))
+    result = parser.read_csv(fname, index_col=0, parse_dates=True)
+
+    expected = DataFrame(
+        [
+            [0.980269, 3.685731, -0.364216805298, -1.159738],
+            [1.047916, -0.041232, -0.16181208307, 0.212549],
+            [0.498581, 0.731168, -0.537677223318, 1.346270],
+            [1.120202, 1.567621, 0.00364077397681, 0.675253],
+            [-0.487094, 0.571455, -1.6116394093, 0.103469],
+            [0.836649, 0.246462, 0.588542635376, 1.062782],
+            [-0.157161, 1.340307, 1.1957779562, -1.097007],
+        ],
+        columns=["A", "B", "C", "D"],
+        index=Index(
+            [
+                datetime(2000, 1, 3),
+                datetime(2000, 1, 4),
+                datetime(2000, 1, 5),
+                datetime(2000, 1, 6),
+                datetime(2000, 1, 7),
+                datetime(2000, 1, 10),
+                datetime(2000, 1, 11),
+            ],
+            dtype="M8[s]",
+            name="index",
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_1000_sep(all_parsers):
+    parser = all_parsers
+    data = """A|B|C
+1|2,334|5
+10|13|10.
+"""
+    expected = DataFrame({"A": [1, 10], "B": [2334, 13], "C": [5, 10.0]})
+
+    if parser.engine == "pyarrow":
+        msg = "The 'thousands' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), sep="|", thousands=",")
+        return
+
+    result = parser.read_csv(StringIO(data), sep="|", thousands=",")
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_unnamed_columns(all_parsers):
+    data = """A,B,C,,
+1,2,3,4,5
+6,7,8,9,10
+11,12,13,14,15
+"""
+    parser = all_parsers
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]],
+        dtype=np.int64,
+        columns=["A", "B", "C", "Unnamed: 3", "Unnamed: 4"],
+    )
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_csv_mixed_type(all_parsers):
+    data = """A,B,C
+a,1,2
+b,3,4
+c,4,5
+"""
+    parser = all_parsers
+    expected = DataFrame({"A": ["a", "b", "c"], "B": [1, 3, 4], "C": [2, 4, 5]})
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_low_memory_no_rows_with_index(all_parsers):
+    # see gh-21141
+    parser = all_parsers
+
+    if not parser.low_memory:
+        pytest.skip("This is a low-memory specific test")
+
+    data = """A,B,C
+1,1,1,2
+2,2,3,4
+3,3,4,5
+"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), low_memory=True, index_col=0, nrows=0)
+        return
+
+    result = parser.read_csv(StringIO(data), low_memory=True, index_col=0, nrows=0)
+    expected = DataFrame(columns=["A", "B", "C"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_dataframe(all_parsers, csv1):
+    parser = all_parsers
+    result = parser.read_csv(csv1, index_col=0, parse_dates=True)
+    expected = DataFrame(
+        [
+            [0.980269, 3.685731, -0.364216805298, -1.159738],
+            [1.047916, -0.041232, -0.16181208307, 0.212549],
+            [0.498581, 0.731168, -0.537677223318, 1.346270],
+            [1.120202, 1.567621, 0.00364077397681, 0.675253],
+            [-0.487094, 0.571455, -1.6116394093, 0.103469],
+            [0.836649, 0.246462, 0.588542635376, 1.062782],
+            [-0.157161, 1.340307, 1.1957779562, -1.097007],
+        ],
+        columns=["A", "B", "C", "D"],
+        index=Index(
+            [
+                datetime(2000, 1, 3),
+                datetime(2000, 1, 4),
+                datetime(2000, 1, 5),
+                datetime(2000, 1, 6),
+                datetime(2000, 1, 7),
+                datetime(2000, 1, 10),
+                datetime(2000, 1, 11),
+            ],
+            dtype="M8[s]",
+            name="index",
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("nrows", [3, 3.0])
+def test_read_nrows(all_parsers, nrows):
+    # see gh-10476
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    expected = DataFrame(
+        [["foo", 2, 3, 4, 5], ["bar", 7, 8, 9, 10], ["baz", 12, 13, 14, 15]],
+        columns=["index", "A", "B", "C", "D"],
+    )
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), nrows=nrows)
+        return
+
+    result = parser.read_csv(StringIO(data), nrows=nrows)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("nrows", [1.2, "foo", -1])
+def test_read_nrows_bad(all_parsers, nrows):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    msg = r"'nrows' must be an integer >=0"
+    parser = all_parsers
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), nrows=nrows)
+
+
+def test_nrows_skipfooter_errors(all_parsers):
+    msg = "'skipfooter' not supported with 'nrows'"
+    data = "a\n1\n2\n3\n4\n5\n6"
+    parser = all_parsers
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), skipfooter=1, nrows=5)
+
+
+@skip_pyarrow
+def test_missing_trailing_delimiters(all_parsers):
+    parser = all_parsers
+    data = """A,B,C,D
+1,2,3,4
+1,3,3,
+1,4,5"""
+
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [[1, 2, 3, 4], [1, 3, 3, np.nan], [1, 4, 5, np.nan]],
+        columns=["A", "B", "C", "D"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skip_initial_space(all_parsers):
+    data = (
+        '"09-Apr-2012", "01:10:18.300", 2456026.548822908, 12849, '
+        "1.00361,  1.12551, 330.65659, 0355626618.16711,  73.48821, "
+        "314.11625,  1917.09447,   179.71425,  80.000, 240.000, -350,  "
+        "70.06056, 344.98370, 1,   1, -0.689265, -0.692787,  "
+        "0.212036,    14.7674,   41.605,   -9999.0,   -9999.0,   "
+        "-9999.0,   -9999.0,   -9999.0,  -9999.0, 000, 012, 128"
+    )
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The 'skipinitialspace' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                names=list(range(33)),
+                header=None,
+                na_values=["-9999.0"],
+                skipinitialspace=True,
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data),
+        names=list(range(33)),
+        header=None,
+        na_values=["-9999.0"],
+        skipinitialspace=True,
+    )
+    expected = DataFrame(
+        [
+            [
+                "09-Apr-2012",
+                "01:10:18.300",
+                2456026.548822908,
+                12849,
+                1.00361,
+                1.12551,
+                330.65659,
+                355626618.16711,
+                73.48821,
+                314.11625,
+                1917.09447,
+                179.71425,
+                80.0,
+                240.0,
+                -350,
+                70.06056,
+                344.9837,
+                1,
+                1,
+                -0.689265,
+                -0.692787,
+                0.212036,
+                14.7674,
+                41.605,
+                np.nan,
+                np.nan,
+                np.nan,
+                np.nan,
+                np.nan,
+                np.nan,
+                0,
+                12,
+                128,
+            ]
+        ]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_trailing_delimiters(all_parsers):
+    # see gh-2442
+    data = """A,B,C
+1,2,3,
+4,5,6,
+7,8,9,"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=False)
+
+    expected = DataFrame({"A": [1, 4, 7], "B": [2, 5, 8], "C": [3, 6, 9]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_escapechar(all_parsers):
+    # https://stackoverflow.com/questions/13824840/feature-request-for-
+    # pandas-read-csv
+    data = '''SEARCH_TERM,ACTUAL_URL
+"bra tv board","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
+"tv p\xc3\xa5 hjul","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
+"SLAGBORD, \\"Bergslagen\\", IKEA:s 1700-tals series","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"'''
+
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO(data), escapechar="\\", quotechar='"', encoding="utf-8"
+    )
+
+    assert result["SEARCH_TERM"][2] == 'SLAGBORD, "Bergslagen", IKEA:s 1700-tals series'
+
+    tm.assert_index_equal(result.columns, Index(["SEARCH_TERM", "ACTUAL_URL"]))
+
+
+def test_ignore_leading_whitespace(all_parsers):
+    # see gh-3374, gh-6607
+    parser = all_parsers
+    data = " a b c\n 1 2 3\n 4 5 6\n 7 8 9"
+
+    if parser.engine == "pyarrow":
+        msg = "the 'pyarrow' engine does not support regex separators"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), sep=r"\s+")
+        return
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+
+    expected = DataFrame({"a": [1, 4, 7], "b": [2, 5, 8], "c": [3, 6, 9]})
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+@pytest.mark.parametrize("usecols", [None, [0, 1], ["a", "b"]])
+def test_uneven_lines_with_usecols(all_parsers, usecols):
+    # see gh-12203
+    parser = all_parsers
+    data = r"""a,b,c
+0,1,2
+3,4,5,6,7
+8,9,10"""
+
+    if usecols is None:
+        # Make sure that an error is still raised
+        # when the "usecols" parameter is not provided.
+        msg = r"Expected \d+ fields in line \d+, saw \d+"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data))
+    else:
+        expected = DataFrame({"a": [0, 3, 8], "b": [1, 4, 9]})
+
+        result = parser.read_csv(StringIO(data), usecols=usecols)
+        tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        # First, check to see that the response of parser when faced with no
+        # provided columns raises the correct error, with or without usecols.
+        ("", {}, None),
+        ("", {"usecols": ["X"]}, None),
+        (
+            ",,",
+            {"names": ["Dummy", "X", "Dummy_2"], "usecols": ["X"]},
+            DataFrame(columns=["X"], index=[0], dtype=np.float64),
+        ),
+        (
+            "",
+            {"names": ["Dummy", "X", "Dummy_2"], "usecols": ["X"]},
+            DataFrame(columns=["X"]),
+        ),
+    ],
+)
+def test_read_empty_with_usecols(all_parsers, data, kwargs, expected):
+    # see gh-12493
+    parser = all_parsers
+
+    if expected is None:
+        msg = "No columns to parse from file"
+        with pytest.raises(EmptyDataError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+    else:
+        result = parser.read_csv(StringIO(data), **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,expected_data",
+    [
+        # gh-8661, gh-8679: this should ignore six lines, including
+        # lines with trailing whitespace and blank lines.
+        (
+            {
+                "header": None,
+                "sep": r"\s+",
+                "skiprows": [0, 1, 2, 3, 5, 6],
+                "skip_blank_lines": True,
+            },
+            [[1.0, 2.0, 4.0], [5.1, np.nan, 10.0]],
+        ),
+        # gh-8983: test skipping set of rows after a row with trailing spaces.
+        (
+            {
+                "sep": r"\s+",
+                "skiprows": [1, 2, 3, 5, 6],
+                "skip_blank_lines": True,
+            },
+            {"A": [1.0, 5.1], "B": [2.0, np.nan], "C": [4.0, 10]},
+        ),
+    ],
+)
+def test_trailing_spaces(all_parsers, kwargs, expected_data):
+    data = "A B C  \nrandom line with trailing spaces    \nskip\n1,2,3\n1,2.,4.\nrandom line with trailing tabs\t\t\t\n   \n5.1,NaN,10.0\n"  # noqa: E501
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        with pytest.raises(ValueError, match="the 'pyarrow' engine does not support"):
+            parser.read_csv(StringIO(data.replace(",", "  ")), **kwargs)
+        return
+    expected = DataFrame(expected_data)
+    result = parser.read_csv(StringIO(data.replace(",", "  ")), **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_filepath_or_buffer(all_parsers):
+    # see gh-43366
+    parser = all_parsers
+
+    with pytest.raises(TypeError, match="Expected file path name or file-like"):
+        parser.read_csv(filepath_or_buffer=b"input")
+
+
+def test_single_char_leading_whitespace(all_parsers):
+    # see gh-9710
+    parser = all_parsers
+    data = """\
+MyColumn
+a
+b
+a
+b\n"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'skipinitialspace' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                skipinitialspace=True,
+            )
+        return
+    expected = DataFrame({"MyColumn": list("abab")})
+    result = parser.read_csv(StringIO(data), skipinitialspace=True, sep=r"\s+")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "sep,skip_blank_lines,exp_data",
+    [
+        (",", True, [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0], [-70.0, 0.4, 1.0]]),
+        (r"\s+", True, [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0], [-70.0, 0.4, 1.0]]),
+        (
+            ",",
+            False,
+            [
+                [1.0, 2.0, 4.0],
+                [np.nan, np.nan, np.nan],
+                [np.nan, np.nan, np.nan],
+                [5.0, np.nan, 10.0],
+                [np.nan, np.nan, np.nan],
+                [-70.0, 0.4, 1.0],
+            ],
+        ),
+    ],
+)
+def test_empty_lines(all_parsers, sep, skip_blank_lines, exp_data, request):
+    parser = all_parsers
+    data = """\
+A,B,C
+1,2.,4.
+
+
+5.,NaN,10.0
+
+-70,.4,1
+"""
+
+    if sep == r"\s+":
+        data = data.replace(",", "  ")
+
+        if parser.engine == "pyarrow":
+            msg = "the 'pyarrow' engine does not support regex separators"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(
+                    StringIO(data), sep=sep, skip_blank_lines=skip_blank_lines
+                )
+            return
+
+    result = parser.read_csv(StringIO(data), sep=sep, skip_blank_lines=skip_blank_lines)
+    expected = DataFrame(exp_data, columns=["A", "B", "C"])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_whitespace_lines(all_parsers):
+    parser = all_parsers
+    data = """
+
+\t  \t\t
+\t
+A,B,C
+\t    1,2.,4.
+5.,NaN,10.0
+"""
+    expected = DataFrame([[1, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"])
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        (
+            """   A   B   C   D
+a   1   2   3   4
+b   1   2   3   4
+c   1   2   3   4
+""",
+            DataFrame(
+                [[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]],
+                columns=["A", "B", "C", "D"],
+                index=["a", "b", "c"],
+            ),
+        ),
+        (
+            "    a b c\n1 2 3 \n4 5  6\n 7 8 9",
+            DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=["a", "b", "c"]),
+        ),
+    ],
+)
+def test_whitespace_regex_separator(all_parsers, data, expected):
+    # see gh-6607
+    parser = all_parsers
+    if parser.engine == "pyarrow":
+        msg = "the 'pyarrow' engine does not support regex separators"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), sep=r"\s+")
+        return
+
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sub_character(all_parsers, csv_dir_path):
+    # see gh-16893
+    filename = os.path.join(csv_dir_path, "sub_char.csv")
+    expected = DataFrame([[1, 2, 3]], columns=["a", "\x1ab", "c"])
+
+    parser = all_parsers
+    result = parser.read_csv(filename)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("filename", ["sé-es-vé.csv", "ru-sй.csv", "中文文件名.csv"])
+def test_filename_with_special_chars(all_parsers, filename):
+    # see gh-15086.
+    parser = all_parsers
+    df = DataFrame({"a": [1, 2, 3]})
+
+    with tm.ensure_clean(filename) as path:
+        df.to_csv(path, index=False)
+
+        result = parser.read_csv(path)
+        tm.assert_frame_equal(result, df)
+
+
+def test_read_table_same_signature_as_read_csv(all_parsers):
+    # GH-34976
+    parser = all_parsers
+
+    table_sign = signature(parser.read_table)
+    csv_sign = signature(parser.read_csv)
+
+    assert table_sign.parameters.keys() == csv_sign.parameters.keys()
+    assert table_sign.return_annotation == csv_sign.return_annotation
+
+    for key, csv_param in csv_sign.parameters.items():
+        table_param = table_sign.parameters[key]
+        if key == "sep":
+            assert csv_param.default == ","
+            assert table_param.default == "\t"
+            assert table_param.annotation == csv_param.annotation
+            assert table_param.kind == csv_param.kind
+            continue
+
+        assert table_param == csv_param
+
+
+def test_read_table_equivalency_to_read_csv(all_parsers):
+    # see gh-21948
+    # As of 0.25.0, read_table is undeprecated
+    parser = all_parsers
+    data = "a\tb\n1\t2\n3\t4"
+    expected = parser.read_csv(StringIO(data), sep="\t")
+    result = parser.read_table(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("read_func", ["read_csv", "read_table"])
+def test_read_csv_and_table_sys_setprofile(all_parsers, read_func):
+    # GH#41069
+    parser = all_parsers
+    data = "a b\n0 1"
+
+    sys.setprofile(lambda *a, **k: None)
+    result = getattr(parser, read_func)(StringIO(data))
+    sys.setprofile(None)
+
+    expected = DataFrame({"a b": ["0 1"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_first_row_bom(all_parsers):
+    # see gh-26545
+    parser = all_parsers
+    data = '''\ufeff"Head1"\t"Head2"\t"Head3"'''
+
+    result = parser.read_csv(StringIO(data), delimiter="\t")
+    expected = DataFrame(columns=["Head1", "Head2", "Head3"])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_first_row_bom_unquoted(all_parsers):
+    # see gh-36343
+    parser = all_parsers
+    data = """\ufeffHead1\tHead2\tHead3"""
+
+    result = parser.read_csv(StringIO(data), delimiter="\t")
+    expected = DataFrame(columns=["Head1", "Head2", "Head3"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("nrows", range(1, 6))
+def test_blank_lines_between_header_and_data_rows(all_parsers, nrows):
+    # GH 28071
+    ref = DataFrame(
+        [[np.nan, np.nan], [np.nan, np.nan], [1, 2], [np.nan, np.nan], [3, 4]],
+        columns=list("ab"),
+    )
+    csv = "\nheader\n\na,b\n\n\n1,2\n\n3,4"
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The 'nrows' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(csv), header=3, nrows=nrows, skip_blank_lines=False
+            )
+        return
+
+    df = parser.read_csv(StringIO(csv), header=3, nrows=nrows, skip_blank_lines=False)
+    tm.assert_frame_equal(df, ref[:nrows])
+
+
+@skip_pyarrow
+def test_no_header_two_extra_columns(all_parsers):
+    # GH 26218
+    column_names = ["one", "two", "three"]
+    ref = DataFrame([["foo", "bar", "baz"]], columns=column_names)
+    stream = StringIO("foo,bar,baz,bam,blah")
+    parser = all_parsers
+    df = parser.read_csv_check_warnings(
+        ParserWarning,
+        "Length of header or names does not match length of data. "
+        "This leads to a loss of data with index_col=False.",
+        stream,
+        header=None,
+        names=column_names,
+        index_col=False,
+    )
+    tm.assert_frame_equal(df, ref)
+
+
+def test_read_csv_names_not_accepting_sets(all_parsers):
+    # GH 34946
+    data = """\
+    1,2,3
+    4,5,6\n"""
+    parser = all_parsers
+    with pytest.raises(ValueError, match="Names should be an ordered collection."):
+        parser.read_csv(StringIO(data), names=set("QAZ"))
+
+
+def test_read_csv_delimiter_and_sep_no_default(all_parsers):
+    # GH#39823
+    f = StringIO("a,b\n1,2")
+    parser = all_parsers
+    msg = "Specified a sep and a delimiter; you can only specify one."
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(f, sep=" ", delimiter=".")
+
+
+@pytest.mark.parametrize("kwargs", [{"delimiter": "\n"}, {"sep": "\n"}])
+def test_read_csv_line_break_as_separator(kwargs, all_parsers):
+    # GH#43528
+    parser = all_parsers
+    data = """a,b,c
+1,2,3
+    """
+    msg = (
+        r"Specified \\n as separator or delimiter. This forces the python engine "
+        r"which does not accept a line terminator. Hence it is not allowed to use "
+        r"the line terminator as separator."
+    )
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), **kwargs)
+
+
+@skip_pyarrow
+def test_dict_keys_as_names(all_parsers):
+    # GH: 36928
+    data = "1,2"
+
+    keys = {"a": int, "b": int}.keys()
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data), names=keys)
+    expected = DataFrame({"a": [1], "b": [2]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.xfail(using_string_dtype() and HAS_PYARROW, reason="TODO(infer_string)")
+@xfail_pyarrow  # UnicodeDecodeError: 'utf-8' codec can't decode byte 0xed in position 0
+def test_encoding_surrogatepass(all_parsers):
+    # GH39017
+    parser = all_parsers
+    content = b"\xed\xbd\xbf"
+    decoded = content.decode("utf-8", errors="surrogatepass")
+    expected = DataFrame({decoded: [decoded]}, index=[decoded * 2])
+    expected.index.name = decoded * 2
+
+    with tm.ensure_clean() as path:
+        Path(path).write_bytes(
+            content * 2 + b"," + content + b"\n" + content * 2 + b"," + content
+        )
+        df = parser.read_csv(path, encoding_errors="surrogatepass", index_col=0)
+        tm.assert_frame_equal(df, expected)
+        with pytest.raises(UnicodeDecodeError, match="'utf-8' codec can't decode byte"):
+            parser.read_csv(path)
+
+
+def test_malformed_second_line(all_parsers):
+    # see GH14782
+    parser = all_parsers
+    data = "\na\nb\n"
+    result = parser.read_csv(StringIO(data), skip_blank_lines=False, header=1)
+    expected = DataFrame({"a": ["b"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_short_single_line(all_parsers):
+    # GH 47566
+    parser = all_parsers
+    columns = ["a", "b", "c"]
+    data = "1,2"
+    result = parser.read_csv(StringIO(data), header=None, names=columns)
+    expected = DataFrame({"a": [1], "b": [2], "c": [np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Length mismatch: Expected axis has 2 elements
+def test_short_multi_line(all_parsers):
+    # GH 47566
+    parser = all_parsers
+    columns = ["a", "b", "c"]
+    data = "1,2\n1,2"
+    result = parser.read_csv(StringIO(data), header=None, names=columns)
+    expected = DataFrame({"a": [1, 1], "b": [2, 2], "c": [np.nan, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_seek(all_parsers):
+    # GH48646
+    parser = all_parsers
+    prefix = "### DATA\n"
+    content = "nkey,value\ntables,rectangular\n"
+    with tm.ensure_clean() as path:
+        Path(path).write_text(prefix + content, encoding="utf-8")
+        with open(path, encoding="utf-8") as file:
+            file.readline()
+            actual = parser.read_csv(file)
+        expected = parser.read_csv(StringIO(content))
+    tm.assert_frame_equal(actual, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_data_list.py b/pandas/pandas/tests/io/parser/common/test_data_list.py
new file mode 100644
index 0000000000000..bf9293ddd841d
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_data_list.py
@@ -0,0 +1,92 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+import csv
+from io import StringIO
+
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+from pandas.io.parsers import TextParser
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+
+@xfail_pyarrow
+def test_read_data_list(all_parsers):
+    parser = all_parsers
+    kwargs = {"index_col": 0}
+    data = "A,B,C\nfoo,1,2,3\nbar,4,5,6"
+
+    data_list = [["A", "B", "C"], ["foo", "1", "2", "3"], ["bar", "4", "5", "6"]]
+    expected = parser.read_csv(StringIO(data), **kwargs)
+
+    with TextParser(data_list, chunksize=2, **kwargs) as parser:
+        result = parser.read()
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reader_list(all_parsers):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    kwargs = {"index_col": 0}
+
+    lines = list(csv.reader(StringIO(data)))
+    with TextParser(lines, chunksize=2, **kwargs) as reader:
+        chunks = list(reader)
+
+    expected = parser.read_csv(StringIO(data), **kwargs)
+
+    tm.assert_frame_equal(chunks[0], expected[:2])
+    tm.assert_frame_equal(chunks[1], expected[2:4])
+    tm.assert_frame_equal(chunks[2], expected[4:])
+
+
+def test_reader_list_skiprows(all_parsers):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    kwargs = {"index_col": 0}
+
+    lines = list(csv.reader(StringIO(data)))
+    with TextParser(lines, chunksize=2, skiprows=[1], **kwargs) as reader:
+        chunks = list(reader)
+
+    expected = parser.read_csv(StringIO(data), **kwargs)
+
+    tm.assert_frame_equal(chunks[0], expected[1:3])
+
+
+def test_read_csv_parse_simple_list(all_parsers):
+    parser = all_parsers
+    data = """foo
+bar baz
+qux foo
+foo
+bar"""
+
+    result = parser.read_csv(StringIO(data), header=None)
+    expected = DataFrame(["foo", "bar baz", "qux foo", "foo", "bar"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_decimal.py b/pandas/pandas/tests/io/parser/common/test_decimal.py
new file mode 100644
index 0000000000000..eb6c97097e5fb
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_decimal.py
@@ -0,0 +1,73 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.mark.parametrize(
+    "data,thousands,decimal",
+    [
+        (
+            """A|B|C
+1|2,334.01|5
+10|13|10.
+""",
+            ",",
+            ".",
+        ),
+        (
+            """A|B|C
+1|2.334,01|5
+10|13|10,
+""",
+            ".",
+            ",",
+        ),
+    ],
+)
+def test_1000_sep_with_decimal(all_parsers, data, thousands, decimal):
+    parser = all_parsers
+    expected = DataFrame({"A": [1, 10], "B": [2334.01, 13], "C": [5, 10.0]})
+
+    if parser.engine == "pyarrow":
+        msg = "The 'thousands' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data), sep="|", thousands=thousands, decimal=decimal
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data), sep="|", thousands=thousands, decimal=decimal
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_euro_decimal_format(all_parsers):
+    parser = all_parsers
+    data = """Id;Number1;Number2;Text1;Text2;Number3
+1;1521,1541;187101,9543;ABC;poi;4,738797819
+2;121,12;14897,76;DEF;uyt;0,377320872
+3;878,158;108013,434;GHI;rez;2,735694704"""
+
+    result = parser.read_csv(StringIO(data), sep=";", decimal=",")
+    expected = DataFrame(
+        [
+            [1, 1521.1541, 187101.9543, "ABC", "poi", 4.738797819],
+            [2, 121.12, 14897.76, "DEF", "uyt", 0.377320872],
+            [3, 878.158, 108013.434, "GHI", "rez", 2.735694704],
+        ],
+        columns=["Id", "Number1", "Number2", "Text1", "Text2", "Number3"],
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_file_buffer_url.py b/pandas/pandas/tests/io/parser/common/test_file_buffer_url.py
new file mode 100644
index 0000000000000..cef57318195ec
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_file_buffer_url.py
@@ -0,0 +1,468 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import (
+    BytesIO,
+    StringIO,
+)
+import os
+import platform
+from urllib.error import URLError
+import uuid
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+)
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_url(all_parsers, csv_dir_path, httpserver):
+    parser = all_parsers
+    kwargs = {"sep": "\t"}
+
+    local_path = os.path.join(csv_dir_path, "salaries.csv")
+    with open(local_path, encoding="utf-8") as f:
+        httpserver.serve_content(content=f.read())
+
+    url_result = parser.read_csv(httpserver.url, **kwargs)
+
+    local_result = parser.read_csv(local_path, **kwargs)
+    tm.assert_frame_equal(url_result, local_result)
+
+
+@pytest.mark.slow
+def test_local_file(all_parsers, csv_dir_path):
+    parser = all_parsers
+    kwargs = {"sep": "\t"}
+
+    local_path = os.path.join(csv_dir_path, "salaries.csv")
+    local_result = parser.read_csv(local_path, **kwargs)
+    url = "file://localhost/" + local_path
+
+    try:
+        url_result = parser.read_csv(url, **kwargs)
+        tm.assert_frame_equal(url_result, local_result)
+    except URLError:
+        # Fails on some systems.
+        pytest.skip("Failing on: " + " ".join(platform.uname()))
+
+
+@xfail_pyarrow  # AssertionError: DataFrame.index are different
+def test_path_path_lib(all_parsers):
+    parser = all_parsers
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    result = tm.round_trip_pathlib(df.to_csv, lambda p: parser.read_csv(p, index_col=0))
+    tm.assert_frame_equal(df, result)
+
+
+@pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+def test_nonexistent_path(all_parsers):
+    # gh-2428: pls no segfault
+    # gh-14086: raise more helpful FileNotFoundError
+    # GH#29233 "File foo" instead of "File b'foo'"
+    parser = all_parsers
+    path = f"{uuid.uuid4()}.csv"
+
+    msg = r"\[Errno 2\]"
+    with pytest.raises(FileNotFoundError, match=msg) as e:
+        parser.read_csv(path)
+    assert path == e.value.filename
+
+
+@pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+@td.skip_if_windows  # os.chmod does not work in windows
+def test_no_permission(all_parsers):
+    # GH 23784
+    parser = all_parsers
+
+    msg = r"\[Errno 13\]"
+    with tm.ensure_clean() as path:
+        os.chmod(path, 0)  # make file unreadable
+
+        # verify that this process cannot open the file (not running as sudo)
+        try:
+            with open(path, encoding="utf-8"):
+                pass
+            pytest.skip("Running as sudo.")
+        except PermissionError:
+            pass
+
+        with pytest.raises(PermissionError, match=msg) as e:
+            parser.read_csv(path)
+        assert path == e.value.filename
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected,msg",
+    [
+        # gh-10728: WHITESPACE_LINE
+        (
+            "a,b,c\n4,5,6\n ",
+            {},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # gh-10548: EAT_LINE_COMMENT
+        (
+            "a,b,c\n4,5,6\n#comment",
+            {"comment": "#"},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # EAT_CRNL_NOP
+        (
+            "a,b,c\n4,5,6\n\r",
+            {},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # EAT_COMMENT
+        (
+            "a,b,c\n4,5,6#comment",
+            {"comment": "#"},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # SKIP_LINE
+        (
+            "a,b,c\n4,5,6\nskipme",
+            {"skiprows": [2]},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # EAT_LINE_COMMENT
+        (
+            "a,b,c\n4,5,6\n#comment",
+            {"comment": "#", "skip_blank_lines": False},
+            DataFrame([[4, 5, 6]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # IN_FIELD
+        (
+            "a,b,c\n4,5,6\n ",
+            {"skip_blank_lines": False},
+            DataFrame([["4", 5, 6], [" ", None, None]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # EAT_CRNL
+        (
+            "a,b,c\n4,5,6\n\r",
+            {"skip_blank_lines": False},
+            DataFrame([[4, 5, 6], [None, None, None]], columns=["a", "b", "c"]),
+            None,
+        ),
+        # ESCAPED_CHAR
+        (
+            "a,b,c\n4,5,6\n\\",
+            {"escapechar": "\\"},
+            None,
+            "(EOF following escape character)|(unexpected end of data)",
+        ),
+        # ESCAPE_IN_QUOTED_FIELD
+        (
+            'a,b,c\n4,5,6\n"\\',
+            {"escapechar": "\\"},
+            None,
+            "(EOF inside string starting at row 2)|(unexpected end of data)",
+        ),
+        # IN_QUOTED_FIELD
+        (
+            'a,b,c\n4,5,6\n"',
+            {"escapechar": "\\"},
+            None,
+            "(EOF inside string starting at row 2)|(unexpected end of data)",
+        ),
+    ],
+    ids=[
+        "whitespace-line",
+        "eat-line-comment",
+        "eat-crnl-nop",
+        "eat-comment",
+        "skip-line",
+        "eat-line-comment",
+        "in-field",
+        "eat-crnl",
+        "escaped-char",
+        "escape-in-quoted-field",
+        "in-quoted-field",
+    ],
+)
+def test_eof_states(all_parsers, data, kwargs, expected, msg, request):
+    # see gh-10728, gh-10548
+    parser = all_parsers
+
+    if parser.engine == "pyarrow" and "comment" in kwargs:
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+        return
+
+    if parser.engine == "pyarrow" and "\r" not in data:
+        # pandas.errors.ParserError: CSV parse error: Expected 3 columns, got 1:
+        # ValueError: skiprows argument must be an integer when using engine='pyarrow'
+        # AssertionError: Regex pattern did not match.
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    if expected is None:
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+    else:
+        result = parser.read_csv(StringIO(data), **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_temporary_file(all_parsers, temp_file):
+    # see gh-13398
+    parser = all_parsers
+    data = "0 0"
+
+    with temp_file.open(mode="w+", encoding="utf-8") as new_file:
+        new_file.write(data)
+        new_file.flush()
+        new_file.seek(0)
+
+        if parser.engine == "pyarrow":
+            msg = "the 'pyarrow' engine does not support regex separators"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(new_file, sep=r"\s+", header=None)
+            return
+
+        result = parser.read_csv(new_file, sep=r"\s+", header=None)
+
+        expected = DataFrame([[0, 0]])
+        tm.assert_frame_equal(result, expected)
+
+
+def test_internal_eof_byte(all_parsers):
+    # see gh-5500
+    parser = all_parsers
+    data = "a,b\n1\x1a,2"
+
+    expected = DataFrame([["1\x1a", 2]], columns=["a", "b"])
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_internal_eof_byte_to_file(all_parsers):
+    # see gh-16559
+    parser = all_parsers
+    data = b'c1,c2\r\n"test \x1a    test", test\r\n'
+    expected = DataFrame([["test \x1a    test", " test"]], columns=["c1", "c2"])
+    path = f"__{uuid.uuid4()}__.csv"
+
+    with tm.ensure_clean(path) as path:
+        with open(path, "wb") as f:
+            f.write(data)
+
+        result = parser.read_csv(path)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_file_handle_string_io(all_parsers):
+    # gh-14418
+    #
+    # Don't close user provided file handles.
+    parser = all_parsers
+    data = "a,b\n1,2"
+
+    fh = StringIO(data)
+    parser.read_csv(fh)
+    assert not fh.closed
+
+
+def test_file_handles_with_open(all_parsers, csv1):
+    # gh-14418
+    #
+    # Don't close user provided file handles.
+    parser = all_parsers
+
+    for mode in ["r", "rb"]:
+        with open(csv1, mode, encoding="utf-8" if mode == "r" else None) as f:
+            parser.read_csv(f)
+            assert not f.closed
+
+
+def test_invalid_file_buffer_class(all_parsers):
+    # see gh-15337
+    class InvalidBuffer:
+        pass
+
+    parser = all_parsers
+    msg = "Invalid file path or buffer object type"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(InvalidBuffer())
+
+
+def test_invalid_file_buffer_mock(all_parsers):
+    # see gh-15337
+    parser = all_parsers
+    msg = "Invalid file path or buffer object type"
+
+    class Foo:
+        pass
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(Foo())
+
+
+def test_valid_file_buffer_seems_invalid(all_parsers):
+    # gh-16135: we want to ensure that "tell" and "seek"
+    # aren't actually being used when we call `read_csv`
+    #
+    # Thus, while the object may look "invalid" (these
+    # methods are attributes of the `StringIO` class),
+    # it is still a valid file-object for our purposes.
+    class NoSeekTellBuffer(StringIO):
+        def tell(self):
+            raise AttributeError("No tell method")
+
+        def seek(self, pos, whence=0):
+            raise AttributeError("No seek method")
+
+    data = "a\n1"
+    parser = all_parsers
+    expected = DataFrame({"a": [1]})
+
+    result = parser.read_csv(NoSeekTellBuffer(data))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("io_class", [StringIO, BytesIO])
+@pytest.mark.parametrize("encoding", [None, "utf-8"])
+def test_read_csv_file_handle(all_parsers, io_class, encoding):
+    """
+    Test whether read_csv does not close user-provided file handles.
+
+    GH 36980
+    """
+    parser = all_parsers
+    expected = DataFrame({"a": [1], "b": [2]})
+
+    content = "a,b\n1,2"
+    handle = io_class(content.encode("utf-8") if io_class == BytesIO else content)
+
+    tm.assert_frame_equal(parser.read_csv(handle, encoding=encoding), expected)
+    assert not handle.closed
+
+
+def test_memory_map_compression(all_parsers, compression):
+    """
+    Support memory map for compressed files.
+
+    GH 37621
+    """
+    parser = all_parsers
+    expected = DataFrame({"a": [1], "b": [2]})
+
+    with tm.ensure_clean() as path:
+        expected.to_csv(path, index=False, compression=compression)
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(path, memory_map=True, compression=compression)
+            return
+
+        result = parser.read_csv(path, memory_map=True, compression=compression)
+
+    tm.assert_frame_equal(
+        result,
+        expected,
+    )
+
+
+def test_context_manager(all_parsers, datapath):
+    # make sure that opened files are closed
+    parser = all_parsers
+
+    path = datapath("io", "data", "csv", "iris.csv")
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(path, chunksize=1)
+        return
+
+    reader = parser.read_csv(path, chunksize=1)
+    assert not reader.handles.handle.closed
+    try:
+        with reader:
+            next(reader)
+            raise AssertionError
+    except AssertionError:
+        assert reader.handles.handle.closed
+
+
+def test_context_manageri_user_provided(all_parsers, datapath):
+    # make sure that user-provided handles are not closed
+    parser = all_parsers
+
+    with open(datapath("io", "data", "csv", "iris.csv"), encoding="utf-8") as path:
+        if parser.engine == "pyarrow":
+            msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(path, chunksize=1)
+            return
+
+        reader = parser.read_csv(path, chunksize=1)
+        assert not reader.handles.handle.closed
+        try:
+            with reader:
+                next(reader)
+                raise AssertionError
+        except AssertionError:
+            assert not reader.handles.handle.closed
+
+
+@skip_pyarrow  # ParserError: Empty CSV file
+def test_file_descriptor_leak(all_parsers):
+    # GH 31488
+    parser = all_parsers
+    with tm.ensure_clean() as path:
+        with pytest.raises(EmptyDataError, match="No columns to parse from file"):
+            parser.read_csv(path)
+
+
+def test_memory_map(all_parsers, csv_dir_path):
+    mmap_file = os.path.join(csv_dir_path, "test_mmap.csv")
+    parser = all_parsers
+
+    expected = DataFrame(
+        {"a": [1, 2, 3], "b": ["one", "two", "three"], "c": ["I", "II", "III"]}
+    )
+
+    if parser.engine == "pyarrow":
+        msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(mmap_file, memory_map=True)
+        return
+
+    result = parser.read_csv(mmap_file, memory_map=True)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_float.py b/pandas/pandas/tests/io/parser/common/test_float.py
new file mode 100644
index 0000000000000..4e0b61577f9e7
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_float.py
@@ -0,0 +1,80 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_linux
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@skip_pyarrow  # ParserError: CSV parse error: Empty CSV file or block
+def test_float_parser(all_parsers):
+    # see gh-9565
+    parser = all_parsers
+    data = "45e-1,4.5,45.,inf,-inf"
+    result = parser.read_csv(StringIO(data), header=None)
+
+    expected = DataFrame([[float(s) for s in data.split(",")]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_scientific_no_exponent(all_parsers_all_precisions):
+    # see gh-12215
+    df = DataFrame.from_dict({"w": ["2e"], "x": ["3E"], "y": ["42e"], "z": ["632E"]})
+    data = df.to_csv(index=False)
+    parser, precision = all_parsers_all_precisions
+
+    df_roundtrip = parser.read_csv(StringIO(data), float_precision=precision)
+    tm.assert_frame_equal(df_roundtrip, df)
+
+
+@pytest.mark.parametrize(
+    "neg_exp",
+    [
+        -617,
+        -100000,
+        pytest.param(-99999999999999999, marks=pytest.mark.skip_ubsan),
+    ],
+)
+def test_very_negative_exponent(all_parsers_all_precisions, neg_exp):
+    # GH#38753
+    parser, precision = all_parsers_all_precisions
+
+    data = f"data\n10E{neg_exp}"
+    result = parser.read_csv(StringIO(data), float_precision=precision)
+    expected = DataFrame({"data": [0.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.skip_ubsan
+@xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
+@pytest.mark.parametrize("exp", [999999999999999999, -999999999999999999])
+def test_too_many_exponent_digits(all_parsers_all_precisions, exp, request):
+    # GH#38753
+    parser, precision = all_parsers_all_precisions
+    data = f"data\n10E{exp}"
+    result = parser.read_csv(StringIO(data), float_precision=precision)
+    if precision == "round_trip":
+        if exp == 999999999999999999 and is_platform_linux():
+            mark = pytest.mark.xfail(reason="GH38794, on Linux gives object result")
+            request.applymarker(mark)
+
+        value = np.inf if exp > 0 else 0.0
+        expected = DataFrame({"data": [value]})
+    else:
+        expected = DataFrame({"data": [f"10E{exp}"]})
+
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_index.py b/pandas/pandas/tests/io/parser/common/test_index.py
new file mode 100644
index 0000000000000..cfa8785b24bde
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_index.py
@@ -0,0 +1,307 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from datetime import datetime
+from io import StringIO
+import os
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        (
+            """foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+""",
+            {"index_col": 0, "names": ["index", "A", "B", "C", "D"]},
+            DataFrame(
+                [
+                    [2, 3, 4, 5],
+                    [7, 8, 9, 10],
+                    [12, 13, 14, 15],
+                    [12, 13, 14, 15],
+                    [12, 13, 14, 15],
+                    [12, 13, 14, 15],
+                ],
+                index=Index(["foo", "bar", "baz", "qux", "foo2", "bar2"], name="index"),
+                columns=["A", "B", "C", "D"],
+            ),
+        ),
+        (
+            """foo,one,2,3,4,5
+foo,two,7,8,9,10
+foo,three,12,13,14,15
+bar,one,12,13,14,15
+bar,two,12,13,14,15
+""",
+            {"index_col": [0, 1], "names": ["index1", "index2", "A", "B", "C", "D"]},
+            DataFrame(
+                [
+                    [2, 3, 4, 5],
+                    [7, 8, 9, 10],
+                    [12, 13, 14, 15],
+                    [12, 13, 14, 15],
+                    [12, 13, 14, 15],
+                ],
+                index=MultiIndex.from_tuples(
+                    [
+                        ("foo", "one"),
+                        ("foo", "two"),
+                        ("foo", "three"),
+                        ("bar", "one"),
+                        ("bar", "two"),
+                    ],
+                    names=["index1", "index2"],
+                ),
+                columns=["A", "B", "C", "D"],
+            ),
+        ),
+    ],
+)
+def test_pass_names_with_index(all_parsers, data, kwargs, expected):
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_col", [[0, 1], [1, 0]])
+def test_multi_index_no_level_names(
+    request, all_parsers, index_col, using_infer_string
+):
+    data = """index1,index2,A,B,C,D
+foo,one,2,3,4,5
+foo,two,7,8,9,10
+foo,three,12,13,14,15
+bar,one,12,13,14,15
+bar,two,12,13,14,15
+"""
+    headless_data = "\n".join(data.split("\n")[1:])
+
+    names = ["A", "B", "C", "D"]
+    parser = all_parsers
+
+    result = parser.read_csv(
+        StringIO(headless_data), index_col=index_col, header=None, names=names
+    )
+    expected = parser.read_csv(StringIO(data), index_col=index_col)
+
+    # No index names in headless data.
+    expected.index.names = [None] * 2
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_multi_index_no_level_names_implicit(all_parsers):
+    parser = all_parsers
+    data = """A,B,C,D
+foo,one,2,3,4,5
+foo,two,7,8,9,10
+foo,three,12,13,14,15
+bar,one,12,13,14,15
+bar,two,12,13,14,15
+"""
+
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [
+            [2, 3, 4, 5],
+            [7, 8, 9, 10],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+        ],
+        columns=["A", "B", "C", "D"],
+        index=MultiIndex.from_tuples(
+            [
+                ("foo", "one"),
+                ("foo", "two"),
+                ("foo", "three"),
+                ("bar", "one"),
+                ("bar", "two"),
+            ]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize(
+    "data,columns,header",
+    [
+        ("a,b", ["a", "b"], [0]),
+        (
+            "a,b\nc,d",
+            MultiIndex.from_tuples([("a", "c"), ("b", "d")]),
+            [0, 1],
+        ),
+    ],
+)
+@pytest.mark.parametrize("round_trip", [True, False])
+def test_multi_index_blank_df(all_parsers, data, columns, header, round_trip):
+    # see gh-14545
+    parser = all_parsers
+    expected = DataFrame(columns=columns)
+    data = expected.to_csv(index=False) if round_trip else data
+
+    result = parser.read_csv(StringIO(data), header=header)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame.columns are different
+def test_no_unnamed_index(all_parsers):
+    parser = all_parsers
+    data = """ id c0 c1 c2
+0 1 0 a b
+1 2 0 c d
+2 2 2 e f
+"""
+    result = parser.read_csv(StringIO(data), sep=" ")
+    expected = DataFrame(
+        [[0, 1, 0, "a", "b"], [1, 2, 0, "c", "d"], [2, 2, 2, "e", "f"]],
+        columns=["Unnamed: 0", "id", "c0", "c1", "c2"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_duplicate_index_explicit(all_parsers):
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo,12,13,14,15
+bar,12,13,14,15
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=0)
+
+    expected = DataFrame(
+        [
+            [2, 3, 4, 5],
+            [7, 8, 9, 10],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+        ],
+        columns=["A", "B", "C", "D"],
+        index=Index(["foo", "bar", "baz", "qux", "foo", "bar"], name="index"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_read_duplicate_index_implicit(all_parsers):
+    data = """A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo,12,13,14,15
+bar,12,13,14,15
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame(
+        [
+            [2, 3, 4, 5],
+            [7, 8, 9, 10],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+            [12, 13, 14, 15],
+        ],
+        columns=["A", "B", "C", "D"],
+        index=Index(["foo", "bar", "baz", "qux", "foo", "bar"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_read_csv_no_index_name(all_parsers, csv_dir_path):
+    parser = all_parsers
+    csv2 = os.path.join(csv_dir_path, "test2.csv")
+    result = parser.read_csv(csv2, index_col=0, parse_dates=True)
+
+    expected = DataFrame(
+        [
+            [0.980269, 3.685731, -0.364216805298, -1.159738, "foo"],
+            [1.047916, -0.041232, -0.16181208307, 0.212549, "bar"],
+            [0.498581, 0.731168, -0.537677223318, 1.346270, "baz"],
+            [1.120202, 1.567621, 0.00364077397681, 0.675253, "qux"],
+            [-0.487094, 0.571455, -1.6116394093, 0.103469, "foo2"],
+        ],
+        columns=["A", "B", "C", "D", "E"],
+        index=Index(
+            [
+                datetime(2000, 1, 3),
+                datetime(2000, 1, 4),
+                datetime(2000, 1, 5),
+                datetime(2000, 1, 6),
+                datetime(2000, 1, 7),
+            ],
+            dtype="M8[s]",
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+def test_empty_with_index(all_parsers):
+    # see gh-10184
+    data = "x,y"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=0)
+
+    expected = DataFrame(columns=["y"], index=Index([], name="x"))
+    tm.assert_frame_equal(result, expected)
+
+
+# CSV parse error: Empty CSV file or block: cannot infer number of columns
+@skip_pyarrow
+def test_empty_with_multi_index(all_parsers):
+    # see gh-10467
+    data = "x,y,z"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=["x", "y"])
+
+    expected = DataFrame(
+        columns=["z"], index=MultiIndex.from_arrays([[]] * 2, names=["x", "y"])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# CSV parse error: Empty CSV file or block: cannot infer number of columns
+@skip_pyarrow
+def test_empty_with_reversed_multi_index(all_parsers):
+    data = "x,y,z"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=[1, 0])
+
+    expected = DataFrame(
+        columns=["z"], index=MultiIndex.from_arrays([[]] * 2, names=["y", "x"])
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_inf.py b/pandas/pandas/tests/io/parser/common/test_inf.py
new file mode 100644
index 0000000000000..657aa3278a442
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_inf.py
@@ -0,0 +1,59 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+
+@xfail_pyarrow  # AssertionError: DataFrame.index are different
+@pytest.mark.parametrize("na_filter", [True, False])
+def test_inf_parsing(all_parsers, na_filter):
+    parser = all_parsers
+    data = """\
+,A
+a,inf
+b,-inf
+c,+Inf
+d,-Inf
+e,INF
+f,-INF
+g,+INf
+h,-INf
+i,inF
+j,-inF"""
+    expected = DataFrame(
+        {"A": [float("inf"), float("-inf")] * 5},
+        index=["a", "b", "c", "d", "e", "f", "g", "h", "i", "j"],
+    )
+    result = parser.read_csv(StringIO(data), index_col=0, na_filter=na_filter)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame.index are different
+@pytest.mark.parametrize("na_filter", [True, False])
+def test_infinity_parsing(all_parsers, na_filter):
+    parser = all_parsers
+    data = """\
+,A
+a,Infinity
+b,-Infinity
+c,+Infinity
+"""
+    expected = DataFrame(
+        {"A": [float("infinity"), float("-infinity"), float("+infinity")]},
+        index=["a", "b", "c"],
+    )
+    result = parser.read_csv(StringIO(data), index_col=0, na_filter=na_filter)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_ints.py b/pandas/pandas/tests/io/parser/common/test_ints.py
new file mode 100644
index 0000000000000..9322e8d54f5b8
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_ints.py
@@ -0,0 +1,230 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_int_conversion(all_parsers):
+    data = """A,B
+1.0,1
+2.0,2
+3.0,3
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame([[1.0, 1], [2.0, 2], [3.0, 3]], columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        (
+            "A,B\nTrue,1\nFalse,2\nTrue,3",
+            {},
+            [[True, 1], [False, 2], [True, 3]],
+        ),
+        (
+            "A,B\nYES,1\nno,2\nyes,3\nNo,3\nYes,3",
+            {"true_values": ["yes", "Yes", "YES"], "false_values": ["no", "NO", "No"]},
+            [[True, 1], [False, 2], [True, 3], [False, 3], [True, 3]],
+        ),
+        (
+            "A,B\nTRUE,1\nFALSE,2\nTRUE,3",
+            {},
+            [[True, 1], [False, 2], [True, 3]],
+        ),
+        (
+            "A,B\nfoo,bar\nbar,foo",
+            {"true_values": ["foo"], "false_values": ["bar"]},
+            [[True, False], [False, True]],
+        ),
+    ],
+)
+def test_parse_bool(all_parsers, data, kwargs, expected):
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), **kwargs)
+    expected = DataFrame(expected, columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_integers_above_fp_precision(all_parsers):
+    data = """Numbers
+17007000002000191
+17007000002000191
+17007000002000191
+17007000002000191
+17007000002000192
+17007000002000192
+17007000002000192
+17007000002000192
+17007000002000192
+17007000002000194"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        {
+            "Numbers": [
+                17007000002000191,
+                17007000002000191,
+                17007000002000191,
+                17007000002000191,
+                17007000002000192,
+                17007000002000192,
+                17007000002000192,
+                17007000002000192,
+                17007000002000192,
+                17007000002000194,
+            ]
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("sep", [" ", r"\s+"])
+def test_integer_overflow_bug(all_parsers, sep):
+    # see gh-2601
+    data = "65248E10 11\n55555E55 22\n"
+    parser = all_parsers
+    if parser.engine == "pyarrow" and sep != " ":
+        msg = "the 'pyarrow' engine does not support regex separators"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), header=None, sep=sep)
+        return
+
+    result = parser.read_csv(StringIO(data), header=None, sep=sep)
+    expected = DataFrame([[6.5248e14, 11], [5.5555e59, 22]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_int64_min_issues(all_parsers):
+    # see gh-2599
+    parser = all_parsers
+    data = "A,B\n0,0\n0,"
+    result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame({"A": [0, 0], "B": [0, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conv", [None, np.int64, np.uint64])
+def test_int64_overflow(all_parsers, conv, request):
+    data = """ID
+00013007854817840016671868
+00013007854817840016749251
+00013007854817840016754630
+00013007854817840016781876
+00013007854817840017028824
+00013007854817840017963235
+00013007854817840018860166"""
+    parser = all_parsers
+
+    if conv is None:
+        # 13007854817840016671868 > UINT64_MAX, so this
+        # will overflow and return object as the dtype.
+        if parser.engine == "pyarrow":
+            mark = pytest.mark.xfail(reason="parses to float64")
+            request.applymarker(mark)
+
+        result = parser.read_csv(StringIO(data))
+        expected = DataFrame(
+            [
+                "00013007854817840016671868",
+                "00013007854817840016749251",
+                "00013007854817840016754630",
+                "00013007854817840016781876",
+                "00013007854817840017028824",
+                "00013007854817840017963235",
+                "00013007854817840018860166",
+            ],
+            columns=["ID"],
+        )
+        tm.assert_frame_equal(result, expected)
+    else:
+        # 13007854817840016671868 > UINT64_MAX, so attempts
+        # to cast to either int64 or uint64 will result in
+        # an OverflowError being raised.
+        msg = "|".join(
+            [
+                "Python int too large to convert to C long",
+                "long too big to convert",
+                "int too big to convert",
+            ]
+        )
+        err = OverflowError
+        if parser.engine == "pyarrow":
+            err = ValueError
+            msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+
+        with pytest.raises(err, match=msg):
+            parser.read_csv(StringIO(data), converters={"ID": conv})
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+@pytest.mark.parametrize(
+    "val", [np.iinfo(np.uint64).max, np.iinfo(np.int64).max, np.iinfo(np.int64).min]
+)
+def test_int64_uint64_range(all_parsers, val):
+    # These numbers fall right inside the int64-uint64
+    # range, so they should be parsed as string.
+    parser = all_parsers
+    result = parser.read_csv(StringIO(str(val)), header=None)
+
+    expected = DataFrame([val])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+@pytest.mark.parametrize(
+    "val", [np.iinfo(np.uint64).max + 1, np.iinfo(np.int64).min - 1]
+)
+def test_outside_int64_uint64_range(all_parsers, val):
+    # These numbers fall just outside the int64-uint64
+    # range, so they should be parsed as string.
+    parser = all_parsers
+    result = parser.read_csv(StringIO(str(val)), header=None)
+
+    expected = DataFrame([str(val)])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # gets float64 dtype instead of object
+@pytest.mark.parametrize("exp_data", [[str(-1), str(2**63)], [str(2**63), str(-1)]])
+def test_numeric_range_too_wide(all_parsers, exp_data):
+    # No numerical dtype can hold both negative and uint64
+    # values, so they should be cast as string.
+    parser = all_parsers
+    data = "\n".join(exp_data)
+    expected = DataFrame(exp_data)
+
+    result = parser.read_csv(StringIO(data), header=None)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_integer_precision(all_parsers):
+    # Gh 7072
+    s = """1,1;0;0;0;1;1;3844;3844;3844;1;1;1;1;1;1;0;0;1;1;0;0,,,4321583677327450765
+5,1;0;0;0;1;1;843;843;843;1;1;1;1;1;1;0;0;1;1;0;0,64.0,;,4321113141090630389"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(s), header=None)[4]
+    expected = Series([4321583677327450765, 4321113141090630389], name=4)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_iterator.py b/pandas/pandas/tests/io/parser/common/test_iterator.py
new file mode 100644
index 0000000000000..668aab05b9fa4
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_iterator.py
@@ -0,0 +1,160 @@
+"""
+Tests that work on both the Python and C engines but do not have a
+specific classification into the other test modules.
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    concat,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+def test_iterator(all_parsers):
+    # see gh-6607
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+    parser = all_parsers
+    kwargs = {"index_col": 0}
+
+    expected = parser.read_csv(StringIO(data), **kwargs)
+
+    if parser.engine == "pyarrow":
+        msg = "The 'iterator' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), iterator=True, **kwargs)
+        return
+
+    with parser.read_csv(StringIO(data), iterator=True, **kwargs) as reader:
+        first_chunk = reader.read(3)
+        tm.assert_frame_equal(first_chunk, expected[:3])
+
+        last_chunk = reader.read(5)
+    tm.assert_frame_equal(last_chunk, expected[3:])
+
+
+def test_iterator2(all_parsers):
+    parser = all_parsers
+    data = """A,B,C
+foo,1,2,3
+bar,4,5,6
+baz,7,8,9
+"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'iterator' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), iterator=True)
+        return
+
+    with parser.read_csv(StringIO(data), iterator=True) as reader:
+        result = list(reader)
+
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        index=["foo", "bar", "baz"],
+        columns=["A", "B", "C"],
+    )
+    tm.assert_frame_equal(result[0], expected)
+
+
+def test_iterator_stop_on_chunksize(all_parsers):
+    # gh-3967: stopping iteration when chunksize is specified
+    parser = all_parsers
+    data = """A,B,C
+foo,1,2,3
+bar,4,5,6
+baz,7,8,9
+"""
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), chunksize=1)
+        return
+
+    with parser.read_csv(StringIO(data), chunksize=1) as reader:
+        result = list(reader)
+
+    assert len(result) == 3
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        index=["foo", "bar", "baz"],
+        columns=["A", "B", "C"],
+    )
+    tm.assert_frame_equal(concat(result), expected)
+
+
+def test_nrows_iterator_without_chunksize(all_parsers):
+    # GH 59079
+    parser = all_parsers
+    data = """A,B,C
+foo,1,2,3
+bar,4,5,6
+baz,7,8,9
+"""
+    if parser.engine == "pyarrow":
+        msg = "The 'iterator' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), iterator=True, nrows=2)
+        return
+
+    with parser.read_csv(StringIO(data), iterator=True, nrows=2) as reader:
+        result = reader.get_chunk()
+
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6]],
+        index=["foo", "bar"],
+        columns=["A", "B", "C"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs", [{"iterator": True, "chunksize": 1}, {"iterator": True}, {"chunksize": 1}]
+)
+def test_iterator_skipfooter_errors(all_parsers, kwargs):
+    msg = "'skipfooter' not supported for iteration"
+    parser = all_parsers
+    data = "a\n1\n2"
+
+    if parser.engine == "pyarrow":
+        msg = (
+            "The '(chunksize|iterator)' option is not supported with the "
+            "'pyarrow' engine"
+        )
+
+    with pytest.raises(ValueError, match=msg):
+        with parser.read_csv(StringIO(data), skipfooter=1, **kwargs) as _:
+            pass
+
+
+def test_iteration_open_handle(all_parsers):
+    parser = all_parsers
+    kwargs = {"header": None}
+
+    with tm.ensure_clean() as path:
+        with open(path, "w", encoding="utf-8") as f:
+            f.write("AAA\nBBB\nCCC\nDDD\nEEE\nFFF\nGGG")
+
+        with open(path, encoding="utf-8") as f:
+            for line in f:
+                if "CCC" in line:
+                    break
+
+            result = parser.read_csv(f, **kwargs)
+            expected = DataFrame({0: ["DDD", "EEE", "FFF", "GGG"]})
+            tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/common/test_read_errors.py b/pandas/pandas/tests/io/parser/common/test_read_errors.py
new file mode 100644
index 0000000000000..a73327beea8bb
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/common/test_read_errors.py
@@ -0,0 +1,319 @@
+"""
+Tests that work on the Python, C and PyArrow engines but do not have a
+specific classification into the other test modules.
+"""
+
+import codecs
+import csv
+from io import StringIO
+import os
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+from pandas.compat import PY311
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+    ParserWarning,
+)
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_empty_decimal_marker(all_parsers):
+    data = """A|B|C
+1|2,334|5
+10|13|10.
+"""
+    # Parsers support only length-1 decimals
+    msg = "Only length-1 decimal markers supported"
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = (
+            "only single character unicode strings can be "
+            "converted to Py_UCS4, got length 0"
+        )
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), decimal="")
+
+
+def test_bad_stream_exception(all_parsers, csv_dir_path):
+    # see gh-13652
+    #
+    # This test validates that both the Python engine and C engine will
+    # raise UnicodeDecodeError instead of C engine raising ParserError
+    # and swallowing the exception that caused read to fail.
+    path = os.path.join(csv_dir_path, "sauron.SHIFT_JIS.csv")
+    codec = codecs.lookup("utf-8")
+    utf8 = codecs.lookup("utf-8")
+    parser = all_parsers
+    msg = "'utf-8' codec can't decode byte"
+
+    # Stream must be binary UTF8.
+    with (
+        open(path, "rb") as handle,
+        codecs.StreamRecoder(
+            handle, utf8.encode, utf8.decode, codec.streamreader, codec.streamwriter
+        ) as stream,
+    ):
+        with pytest.raises(UnicodeDecodeError, match=msg):
+            parser.read_csv(stream)
+
+
+def test_malformed(all_parsers):
+    # see gh-6607
+    parser = all_parsers
+    data = """ignore
+A,B,C
+1,2,3 # comment
+1,2,3,4,5
+2,3,4
+"""
+    msg = "Expected 3 fields in line 4, saw 5"
+    err = ParserError
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        err = ValueError
+    with pytest.raises(err, match=msg):
+        parser.read_csv(StringIO(data), header=1, comment="#")
+
+
+@pytest.mark.parametrize("nrows", [5, 3, None])
+def test_malformed_chunks(all_parsers, nrows):
+    data = """ignore
+A,B,C
+skip
+1,2,3
+3,5,10 # comment
+1,2,3,4,5
+2,3,4
+"""
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The 'iterator' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                header=1,
+                comment="#",
+                iterator=True,
+                chunksize=1,
+                skiprows=[2],
+            )
+        return
+
+    msg = "Expected 3 fields in line 6, saw 5"
+    with parser.read_csv(
+        StringIO(data), header=1, comment="#", iterator=True, chunksize=1, skiprows=[2]
+    ) as reader:
+        with pytest.raises(ParserError, match=msg):
+            reader.read(nrows)
+
+
+@xfail_pyarrow  # does not raise
+def test_catch_too_many_names(all_parsers):
+    # see gh-5156
+    data = """\
+1,2,3
+4,,6
+7,8,9
+10,11,12\n"""
+    parser = all_parsers
+    msg = (
+        "Too many columns specified: expected 4 and found 3"
+        if parser.engine == "c"
+        else "Number of passed names did not match number of header fields in the file"
+    )
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), header=0, names=["a", "b", "c", "d"])
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+@pytest.mark.parametrize("nrows", [0, 1, 2, 3, 4, 5])
+def test_raise_on_no_columns(all_parsers, nrows):
+    parser = all_parsers
+    data = "\n" * nrows
+
+    msg = "No columns to parse from file"
+    with pytest.raises(EmptyDataError, match=msg):
+        parser.read_csv(StringIO(data))
+
+
+def test_unexpected_keyword_parameter_exception(all_parsers):
+    # GH-34976
+    parser = all_parsers
+
+    msg = "{}\\(\\) got an unexpected keyword argument 'foo'"
+    with pytest.raises(TypeError, match=msg.format("read_csv")):
+        parser.read_csv("foo.csv", foo=1)
+    with pytest.raises(TypeError, match=msg.format("read_table")):
+        parser.read_table("foo.tsv", foo=1)
+
+
+def test_suppress_error_output(all_parsers):
+    # see gh-15925
+    parser = all_parsers
+    data = "a\n1\n1,2,3\n4\n5,6,7"
+    expected = DataFrame({"a": [1, 4]})
+
+    result = parser.read_csv(StringIO(data), on_bad_lines="skip")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_error_bad_lines(all_parsers):
+    # see gh-15925
+    parser = all_parsers
+    data = "a\n1\n1,2,3\n4\n5,6,7"
+
+    msg = "Expected 1 fields in line 3, saw 3"
+
+    if parser.engine == "pyarrow":
+        # "CSV parse error: Expected 1 columns, got 3: 1,2,3"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    with pytest.raises(ParserError, match=msg):
+        parser.read_csv(StringIO(data), on_bad_lines="error")
+
+
+def test_warn_bad_lines(all_parsers):
+    # see gh-15925
+    parser = all_parsers
+    data = "a\n1\n1,2,3\n4\n5,6,7"
+    expected = DataFrame({"a": [1, 4]})
+    match_msg = "Skipping line"
+
+    expected_warning = ParserWarning
+    if parser.engine == "pyarrow":
+        match_msg = "Expected 1 columns, but found 3: 1,2,3"
+
+    with tm.assert_produces_warning(
+        expected_warning, match=match_msg, check_stacklevel=False
+    ):
+        result = parser.read_csv(StringIO(data), on_bad_lines="warn")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_wrong_num_columns(all_parsers):
+    # Too few columns.
+    data = """A,B,C,D,E,F
+1,2,3,4,5,6
+6,7,8,9,10,11,12
+11,12,13,14,15,16
+"""
+    parser = all_parsers
+    msg = "Expected 6 fields in line 3, saw 7"
+
+    if parser.engine == "pyarrow":
+        # Expected 6 columns, got 7: 6,7,8,9,10,11,12
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    with pytest.raises(ParserError, match=msg):
+        parser.read_csv(StringIO(data))
+
+
+def test_null_byte_char(request, all_parsers):
+    # see gh-2741
+    data = "\x00,foo"
+    names = ["a", "b"]
+    parser = all_parsers
+
+    if parser.engine == "c" or (parser.engine == "python" and PY311):
+        if parser.engine == "python" and PY311:
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="In Python 3.11, this is read as an empty character not null"
+                )
+            )
+        expected = DataFrame([[np.nan, "foo"]], columns=names)
+        out = parser.read_csv(StringIO(data), names=names)
+        tm.assert_frame_equal(out, expected)
+    else:
+        if parser.engine == "pyarrow":
+            # CSV parse error: Empty CSV file or block: "
+            # cannot infer number of columns"
+            pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+        else:
+            msg = "NULL byte detected"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), names=names)
+
+
+@pytest.mark.filterwarnings("always::ResourceWarning")
+def test_open_file(all_parsers):
+    # GH 39024
+    parser = all_parsers
+
+    msg = "Could not determine delimiter"
+    err = csv.Error
+    if parser.engine == "c":
+        msg = "object of type 'NoneType' has no len"
+        err = TypeError
+    elif parser.engine == "pyarrow":
+        msg = "'utf-8' codec can't decode byte 0xe4"
+        err = ValueError
+
+    with tm.ensure_clean() as path:
+        file = Path(path)
+        file.write_bytes(b"\xe4\na\n1")
+
+        with tm.assert_produces_warning(None):
+            # should not trigger a ResourceWarning
+            with pytest.raises(err, match=msg):
+                parser.read_csv(file, sep=None, encoding_errors="replace")
+
+
+def test_invalid_on_bad_line(all_parsers):
+    parser = all_parsers
+    data = "a\n1\n1,2,3\n4\n5,6,7"
+    with pytest.raises(ValueError, match="Argument abc is invalid for on_bad_lines"):
+        parser.read_csv(StringIO(data), on_bad_lines="abc")
+
+
+def test_bad_header_uniform_error(all_parsers):
+    parser = all_parsers
+    data = "+++123456789...\ncol1,col2,col3,col4\n1,2,3,4\n"
+    msg = "Expected 2 fields in line 2, saw 4"
+    if parser.engine == "c":
+        msg = (
+            "Could not construct index. Requested to use 1 "
+            "number of columns, but 3 left to parse."
+        )
+    elif parser.engine == "pyarrow":
+        # "CSV parse error: Expected 1 columns, got 4: col1,col2,col3,col4"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    with pytest.raises(ParserError, match=msg):
+        parser.read_csv(StringIO(data), index_col=0, on_bad_lines="error")
+
+
+def test_on_bad_lines_warn_correct_formatting(all_parsers):
+    # see gh-15925
+    parser = all_parsers
+    data = """1,2
+a,b
+a,b,c
+a,b,d
+a,b
+"""
+    expected = DataFrame({"1": "a", "2": ["b"] * 2})
+    match_msg = "Skipping line"
+
+    expected_warning = ParserWarning
+    if parser.engine == "pyarrow":
+        match_msg = "Expected 2 columns, but found 3: a,b,c"
+
+    with tm.assert_produces_warning(
+        expected_warning, match=match_msg, check_stacklevel=False
+    ):
+        result = parser.read_csv(StringIO(data), on_bad_lines="warn")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/conftest.py b/pandas/pandas/tests/io/parser/conftest.py
new file mode 100644
index 0000000000000..90f77a7024235
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/conftest.py
@@ -0,0 +1,337 @@
+from __future__ import annotations
+
+import os
+
+import pytest
+
+from pandas.compat import HAS_PYARROW
+from pandas.compat._optional import VERSIONS
+
+from pandas import (
+    read_csv,
+    read_table,
+)
+import pandas._testing as tm
+
+
+class BaseParser:
+    engine: str | None = None
+    low_memory = True
+    float_precision_choices: list[str | None] = []
+
+    def update_kwargs(self, kwargs):
+        kwargs = kwargs.copy()
+        kwargs.update({"engine": self.engine, "low_memory": self.low_memory})
+
+        return kwargs
+
+    def read_csv(self, *args, **kwargs):
+        kwargs = self.update_kwargs(kwargs)
+        return read_csv(*args, **kwargs)
+
+    def read_csv_check_warnings(
+        self,
+        warn_type: type[Warning],
+        warn_msg: str,
+        *args,
+        raise_on_extra_warnings=True,
+        check_stacklevel: bool = True,
+        **kwargs,
+    ):
+        # We need to check the stacklevel here instead of in the tests
+        # since this is where read_csv is called and where the warning
+        # should point to.
+        kwargs = self.update_kwargs(kwargs)
+        with tm.assert_produces_warning(
+            warn_type,
+            match=warn_msg,
+            raise_on_extra_warnings=raise_on_extra_warnings,
+            check_stacklevel=check_stacklevel,
+        ):
+            return read_csv(*args, **kwargs)
+
+    def read_table(self, *args, **kwargs):
+        kwargs = self.update_kwargs(kwargs)
+        return read_table(*args, **kwargs)
+
+    def read_table_check_warnings(
+        self,
+        warn_type: type[Warning],
+        warn_msg: str,
+        *args,
+        raise_on_extra_warnings=True,
+        **kwargs,
+    ):
+        # We need to check the stacklevel here instead of in the tests
+        # since this is where read_table is called and where the warning
+        # should point to.
+        kwargs = self.update_kwargs(kwargs)
+        with tm.assert_produces_warning(
+            warn_type, match=warn_msg, raise_on_extra_warnings=raise_on_extra_warnings
+        ):
+            return read_table(*args, **kwargs)
+
+
+class CParser(BaseParser):
+    engine = "c"
+    float_precision_choices = [None, "high", "round_trip"]
+
+
+class CParserHighMemory(CParser):
+    low_memory = False
+
+
+class CParserLowMemory(CParser):
+    low_memory = True
+
+
+class PythonParser(BaseParser):
+    engine = "python"
+    float_precision_choices = [None]
+
+
+class PyArrowParser(BaseParser):
+    engine = "pyarrow"
+    float_precision_choices = [None]
+
+
+@pytest.fixture
+def csv_dir_path(datapath):
+    """
+    The directory path to the data files needed for parser tests.
+    """
+    return datapath("io", "parser", "data")
+
+
+@pytest.fixture
+def csv1(datapath):
+    """
+    The path to the data file "test1.csv" needed for parser tests.
+    """
+    return os.path.join(datapath("io", "data", "csv"), "test1.csv")
+
+
+_cParserHighMemory = CParserHighMemory
+_cParserLowMemory = CParserLowMemory
+_pythonParser = PythonParser
+_pyarrowParser = PyArrowParser
+
+_py_parsers_only = [_pythonParser]
+_c_parsers_only = [_cParserHighMemory, _cParserLowMemory]
+_pyarrow_parsers_only = [
+    pytest.param(
+        _pyarrowParser,
+        marks=[
+            pytest.mark.single_cpu,
+            pytest.mark.skipif(not HAS_PYARROW, reason="pyarrow is not installed"),
+        ],
+    )
+]
+
+_all_parsers = [*_c_parsers_only, *_py_parsers_only, *_pyarrow_parsers_only]
+
+_py_parser_ids = ["python"]
+_c_parser_ids = ["c_high", "c_low"]
+_pyarrow_parsers_ids = ["pyarrow"]
+
+_all_parser_ids = [*_c_parser_ids, *_py_parser_ids, *_pyarrow_parsers_ids]
+
+
+@pytest.fixture(params=_all_parsers, ids=_all_parser_ids)
+def all_parsers(request):
+    """
+    Fixture all of the CSV parsers.
+    """
+    parser = request.param()
+    if parser.engine == "pyarrow":
+        pytest.importorskip("pyarrow", VERSIONS["pyarrow"])
+        # Try finding a way to disable threads all together
+        # for more stable CI runs
+        import pyarrow
+
+        pyarrow.set_cpu_count(1)
+    return parser
+
+
+@pytest.fixture(params=_c_parsers_only, ids=_c_parser_ids)
+def c_parser_only(request):
+    """
+    Fixture all of the CSV parsers using the C engine.
+    """
+    return request.param()
+
+
+@pytest.fixture(params=_py_parsers_only, ids=_py_parser_ids)
+def python_parser_only(request):
+    """
+    Fixture all of the CSV parsers using the Python engine.
+    """
+    return request.param()
+
+
+@pytest.fixture(params=_pyarrow_parsers_only, ids=_pyarrow_parsers_ids)
+def pyarrow_parser_only(request):
+    """
+    Fixture all of the CSV parsers using the Pyarrow engine.
+    """
+    return request.param()
+
+
+def _get_all_parser_float_precision_combinations():
+    """
+    Return all allowable parser and float precision
+    combinations and corresponding ids.
+    """
+    params = []
+    ids = []
+    for parser, parser_id in zip(_all_parsers, _all_parser_ids):
+        if hasattr(parser, "values"):
+            # Wrapped in pytest.param, get the actual parser back
+            parser = parser.values[0]
+        for precision in parser.float_precision_choices:
+            # Re-wrap in pytest.param for pyarrow
+            mark = (
+                [
+                    pytest.mark.single_cpu,
+                    pytest.mark.skipif(
+                        not HAS_PYARROW, reason="pyarrow is not installed"
+                    ),
+                ]
+                if parser.engine == "pyarrow"
+                else ()
+            )
+            param = pytest.param((parser(), precision), marks=mark)
+            params.append(param)
+            ids.append(f"{parser_id}-{precision}")
+
+    return {"params": params, "ids": ids}
+
+
+@pytest.fixture(
+    params=_get_all_parser_float_precision_combinations()["params"],
+    ids=_get_all_parser_float_precision_combinations()["ids"],
+)
+def all_parsers_all_precisions(request):
+    """
+    Fixture for all allowable combinations of parser
+    and float precision
+    """
+    return request.param
+
+
+_utf_values = [8, 16, 32]
+
+_encoding_seps = ["", "-", "_"]
+_encoding_prefixes = ["utf", "UTF"]
+
+_encoding_fmts = [
+    f"{prefix}{sep}{{0}}" for sep in _encoding_seps for prefix in _encoding_prefixes
+]
+
+
+@pytest.fixture(params=_utf_values)
+def utf_value(request):
+    """
+    Fixture for all possible integer values for a UTF encoding.
+    """
+    return request.param
+
+
+@pytest.fixture(params=_encoding_fmts)
+def encoding_fmt(request):
+    """
+    Fixture for all possible string formats of a UTF encoding.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ("-1,0", -1.0),
+        ("-1,2e0", -1.2),
+        ("-1e0", -1.0),
+        ("+1e0", 1.0),
+        ("+1e+0", 1.0),
+        ("+1e-1", 0.1),
+        ("+,1e1", 1.0),
+        ("+1,e0", 1.0),
+        ("-,1e1", -1.0),
+        ("-1,e0", -1.0),
+        ("0,1", 0.1),
+        ("1,", 1.0),
+        (",1", 0.1),
+        ("-,1", -0.1),
+        ("1_,", 1.0),
+        ("1_234,56", 1234.56),
+        ("1_234,56e0", 1234.56),
+        # negative cases; must not parse as float
+        ("_", "_"),
+        ("-_", "-_"),
+        ("-_1", "-_1"),
+        ("-_1e0", "-_1e0"),
+        ("_1", "_1"),
+        ("_1,", "_1,"),
+        ("_1,_", "_1,_"),
+        ("_1e0", "_1e0"),
+        ("1,2e_1", "1,2e_1"),
+        ("1,2e1_0", "1,2e1_0"),
+        ("1,_2", "1,_2"),
+        (",1__2", ",1__2"),
+        (",1e", ",1e"),
+        ("-,1e", "-,1e"),
+        ("1_000,000_000", "1_000,000_000"),
+        ("1,e1_2", "1,e1_2"),
+        ("e11,2", "e11,2"),
+        ("1e11,2", "1e11,2"),
+        ("1,2,2", "1,2,2"),
+        ("1,2_1", "1,2_1"),
+        ("1,2e-10e1", "1,2e-10e1"),
+        ("--1,2", "--1,2"),
+        ("1a_2,1", "1a_2,1"),
+        ("1,2E-1", 0.12),
+        ("1,2E1", 12.0),
+    ]
+)
+def numeric_decimal(request):
+    """
+    Fixture for all numeric formats which should get recognized. The first entry
+    represents the value to read while the second represents the expected result.
+    """
+    return request.param
+
+
+@pytest.fixture
+def pyarrow_xfail(request):
+    """
+    Fixture that xfails a test if the engine is pyarrow.
+
+    Use if failure is do to unsupported keywords or inconsistent results.
+    """
+    if "all_parsers" in request.fixturenames:
+        parser = request.getfixturevalue("all_parsers")
+    elif "all_parsers_all_precisions" in request.fixturenames:
+        # Return value is tuple of (engine, precision)
+        parser = request.getfixturevalue("all_parsers_all_precisions")[0]
+    else:
+        return
+    if parser.engine == "pyarrow":
+        mark = pytest.mark.xfail(reason="pyarrow doesn't support this.")
+        request.applymarker(mark)
+
+
+@pytest.fixture
+def pyarrow_skip(request):
+    """
+    Fixture that skips a test if the engine is pyarrow.
+
+    Use if failure is do a parsing failure from pyarrow.csv.read_csv
+    """
+    if "all_parsers" in request.fixturenames:
+        parser = request.getfixturevalue("all_parsers")
+    elif "all_parsers_all_precisions" in request.fixturenames:
+        # Return value is tuple of (engine, precision)
+        parser = request.getfixturevalue("all_parsers_all_precisions")[0]
+    else:
+        return
+    if parser.engine == "pyarrow":
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
diff --git a/pandas/pandas/tests/io/parser/data/items.jsonl b/pandas/pandas/tests/io/parser/data/items.jsonl
new file mode 100644
index 0000000000000..f784d37befa82
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/items.jsonl
@@ -0,0 +1,2 @@
+{"a": 1, "b": 2}
+{"b":2, "a" :1}
diff --git a/pandas/pandas/tests/io/parser/data/salaries.csv b/pandas/pandas/tests/io/parser/data/salaries.csv
new file mode 100644
index 0000000000000..85631704ff6e0
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/salaries.csv
@@ -0,0 +1,47 @@
+S	X	E	M
+13876 	1 	1 	1
+11608 	1 	3 	0
+18701 	1 	3 	1
+11283 	1 	2 	0
+11767 	1 	3 	0
+20872 	2 	2 	1
+11772 	2 	2 	0
+10535 	2 	1 	0
+12195 	2 	3 	0
+12313 	3 	2 	0
+14975 	3 	1 	1
+21371 	3 	2 	1
+19800 	3 	3 	1
+11417 	4 	1 	0
+20263 	4 	3 	1
+13231 	4 	3 	0
+12884 	4 	2 	0
+13245 	5 	2 	0
+13677 	5 	3 	0
+15965 	5 	1 	1
+12336 	6 	1 	0
+21352 	6 	3 	1
+13839 	6 	2 	0
+22884 	6 	2 	1
+16978 	7 	1 	1
+14803 	8 	2 	0
+17404 	8 	1 	1
+22184 	8 	3 	1
+13548 	8 	1 	0
+14467 	10 	1 	0
+15942 	10 	2 	0
+23174 	10 	3 	1
+23780 	10 	2 	1
+25410 	11 	2 	1
+14861 	11 	1 	0
+16882 	12 	2 	0
+24170 	12 	3 	1
+15990 	13 	1 	0
+26330 	13 	2 	1
+17949 	14 	2 	0
+25685 	15 	3 	1
+27837 	16 	2 	1
+18838 	16 	2 	0
+17483 	16 	1 	0
+19207 	17 	2 	0
+19346 	20 	1 	0
diff --git a/pandas/pandas/tests/io/parser/data/salaries.csv.bz2 b/pandas/pandas/tests/io/parser/data/salaries.csv.bz2
new file mode 100644
index 0000000000000..a68b4e62bf34a
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/salaries.csv.bz2 differ
diff --git a/pandas/pandas/tests/io/parser/data/salaries.csv.gz b/pandas/pandas/tests/io/parser/data/salaries.csv.gz
new file mode 100644
index 0000000000000..629de9703d345
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/salaries.csv.gz differ
diff --git a/pandas/pandas/tests/io/parser/data/salaries.csv.xz b/pandas/pandas/tests/io/parser/data/salaries.csv.xz
new file mode 100644
index 0000000000000..40df8e8f936dc
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/salaries.csv.xz differ
diff --git a/pandas/pandas/tests/io/parser/data/salaries.csv.zst b/pandas/pandas/tests/io/parser/data/salaries.csv.zst
new file mode 100644
index 0000000000000..20c9ed8a7e39f
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/salaries.csv.zst differ
diff --git a/pandas/pandas/tests/io/parser/data/sauron.SHIFT_JIS.csv b/pandas/pandas/tests/io/parser/data/sauron.SHIFT_JIS.csv
new file mode 100644
index 0000000000000..218ddf333ef52
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/sauron.SHIFT_JIS.csv
@@ -0,0 +1,14 @@
+num, text
+1,�T�E�����iSauron�A�A�C�k�A�̑n���̎� - ��O�I3019�N3��25���j�́AJ�ER�ER�E�g�[���L���̒����𕑑�Ƃ��������w�z�r�b�g�̖`���x�w�w�֕���x�w�V���}�����̕���x�̓o��l���B
+2,�w�z�r�b�g�̖`���x�Ɍ��y�̂���u���l����Ȃ��t�v�i�f��w�z�r�b�g�V���[�Y�x�̎����ł́u���l�����i�l�N���}���T�[�j�v�j�Ƃ͔ނ̂��Ƃł���B
+3,���̑��҂ł���w�w�֕���x�ɂ����Ắu��‚̎w�ցithe One Ring�j�v�̍���A�u�����iDark Lord�j�v�A�u���̎ҁithe One�j[1]�v�Ƃ��ēo�ꂷ��B�O�j�ɂ�����w�V���}�����̕���x�ł́A����̖��������S�X�̍ł��͂��鑤�߂ł������B
+4,�T�E�����͌����A�A���_�i�n���j�̑n����S�����V�g�I�푰�A�C�k�A�̈���ł��������A�僁���R�[���̔��t�ɉ��S���đ—����A�A���_�ɊQ���Ȃ����݂ƂȂ����B
+5,�u�T�E�����v�Ƃ̓N�E�F�����Łu�g�̖т̂悾�‚��́v�Ƃ����Ӗ��ł���A�V���_�����œ��l�̈Ӗ��ł��閼�O�u�S���T�E�A�v�ƌĂ΂�邱�Ƃ�����B
+6,�����́A�T�E����������A���݌������G���t�ɂ�閼�ł���A�w�w�֕���x�쒆�ɂ����ăA���S�����́u����i�T�E�����j�͎����̖{���̖��͎g��Ȃ����A��������ɏ���������ɏo�����肷�邱�Ƃ������Ȃ��v�Ɣ������Ă���B
+7,���̂ق��A���I�ɃG���t�ɑ΂��Ď��̂����Ƃ���閼�ɁA�u�A���i�^�[���i������N�j�v�A�u�A���^�m�i���M�ȍ׍H�t�j�v�A�u�A�E�����f�B���i�A�E���̉��l�j�v������B
+8,���I�̍��̃T�E�����́A���݂ɕϐg����\�͂������Ă����B
+9,���̔\�͂��g���Ό��ڗ킵�����h�ȊO���𑕂����Ƃ�A�܂�����ȘT��z����������Ƃ����������ɕς��邱�Ƃ��ł��A�G���t���狰���ꂽ�B
+10,���I�Ɉ�‚̎w�ւ����グ���T�E�����́A���̗͂̎w�ւŐ�����鎖���₻�̏��L�҂��x�z�ł���悤�ɂȂ����B
+11,�܂��A���̂��łтĂ��w�ւ�������艽�x�ł��h�邱�Ƃ��ł����B
+12,�������k�[���m�[���v���̍ۂɔ��������̂�j�󂳂ꂽ��́A��x�Ɣ������ϐg���邱�Ƃ͂ł��Ȃ��Ȃ�A���̈��ӂ̋�̂悤�Ȍ�������낵���p�����Ƃ�Ȃ��Ȃ����Ƃ����B
+13,�܂����΂��΁u�܂Ԃ��̂Ȃ��΂ɉ����ꂽ�ځv�Ƃ������S�ە\���ő�����ꂽ�B
diff --git a/pandas/pandas/tests/io/parser/data/sub_char.csv b/pandas/pandas/tests/io/parser/data/sub_char.csv
new file mode 100644
index 0000000000000..ff1fa777832c7
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/sub_char.csv
@@ -0,0 +1,2 @@
+a,"b",c
+1,2,3
\ No newline at end of file
diff --git a/pandas/pandas/tests/io/parser/data/tar_csv.tar b/pandas/pandas/tests/io/parser/data/tar_csv.tar
new file mode 100644
index 0000000000000..d1819550e0a00
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/tar_csv.tar differ
diff --git a/pandas/pandas/tests/io/parser/data/tar_csv.tar.gz b/pandas/pandas/tests/io/parser/data/tar_csv.tar.gz
new file mode 100644
index 0000000000000..80505d345f1e2
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/tar_csv.tar.gz differ
diff --git a/pandas/pandas/tests/io/parser/data/test2.csv b/pandas/pandas/tests/io/parser/data/test2.csv
new file mode 100644
index 0000000000000..6f9141152001a
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/test2.csv
@@ -0,0 +1,6 @@
+A,B,C,D,E
+2000-01-03 00:00:00,0.980268513777,3.68573087906,-0.364216805298,-1.15973806169,foo
+2000-01-04 00:00:00,1.04791624281,-0.0412318367011,-0.16181208307,0.212549316967,bar
+2000-01-05 00:00:00,0.498580885705,0.731167677815,-0.537677223318,1.34627041952,baz
+2000-01-06 00:00:00,1.12020151869,1.56762092543,0.00364077397681,0.67525259227,qux
+2000-01-07 00:00:00,-0.487094399463,0.571454623474,-1.6116394093,0.103468562917,foo2
diff --git a/pandas/pandas/tests/io/parser/data/test_mmap.csv b/pandas/pandas/tests/io/parser/data/test_mmap.csv
new file mode 100644
index 0000000000000..2885fc2bfbd69
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/test_mmap.csv
@@ -0,0 +1,4 @@
+a,b,c
+1,one,I
+2,two,II
+3,three,III
diff --git a/pandas/pandas/tests/io/parser/data/unicode_series.csv b/pandas/pandas/tests/io/parser/data/unicode_series.csv
new file mode 100644
index 0000000000000..2485e149edb06
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/data/unicode_series.csv
@@ -0,0 +1,18 @@
+1617,King of New York (1990)
+1618,All Things Fair (1996)
+1619,"Sixth Man, The (1997)"
+1620,Butterfly Kiss (1995)
+1621,"Paris, France (1993)"
+1622,"C�r�monie, La (1995)"
+1623,Hush (1998)
+1624,Nightwatch (1997)
+1625,Nobody Loves Me (Keiner liebt mich) (1994)
+1626,"Wife, The (1995)"
+1627,Lamerica (1994)
+1628,Nico Icon (1995)
+1629,"Silence of the Palace, The (Saimt el Qusur) (1994)"
+1630,"Slingshot, The (1993)"
+1631,Land and Freedom (Tierra y libertad) (1995)
+1632,� k�ldum klaka (Cold Fever) (1994)
+1633,Etz Hadomim Tafus (Under the Domin Tree) (1994)
+1634,Two Friends (1986)
diff --git a/pandas/pandas/tests/io/parser/data/utf16_ex.txt b/pandas/pandas/tests/io/parser/data/utf16_ex.txt
new file mode 100644
index 0000000000000..f0b452a2bd5ff
Binary files /dev/null and b/pandas/pandas/tests/io/parser/data/utf16_ex.txt differ
diff --git a/pandas/pandas/tests/io/parser/dtypes/__init__.py b/pandas/pandas/tests/io/parser/dtypes/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/parser/dtypes/test_categorical.py b/pandas/pandas/tests/io/parser/dtypes/test_categorical.py
new file mode 100644
index 0000000000000..15cbac54ff8d9
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/dtypes/test_categorical.py
@@ -0,0 +1,335 @@
+"""
+Tests dtype specification during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+import os
+
+import numpy as np
+import pytest
+
+from pandas._libs import parsers as libparsers
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Timestamp,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+
+@xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "category",
+        CategoricalDtype(),
+        {"a": "category", "b": "category", "c": CategoricalDtype()},
+    ],
+)
+def test_categorical_dtype(all_parsers, dtype):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b,c
+1,a,3.4
+1,a,3.4
+2,b,4.5"""
+    expected = DataFrame(
+        {
+            "a": Categorical(["1", "1", "2"]),
+            "b": Categorical(["a", "a", "b"]),
+            "c": Categorical(["3.4", "3.4", "4.5"]),
+        }
+    )
+    actual = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.parametrize("dtype", [{"b": "category"}, {1: "category"}])
+def test_categorical_dtype_single(all_parsers, dtype, request):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b,c
+1,a,3.4
+1,a,3.4
+2,b,4.5"""
+    expected = DataFrame(
+        {"a": [1, 1, 2], "b": Categorical(["a", "a", "b"]), "c": [3.4, 3.4, 4.5]}
+    )
+    if parser.engine == "pyarrow":
+        mark = pytest.mark.xfail(
+            strict=False,
+            reason="Flaky test sometimes gives object dtype instead of Categorical",
+        )
+        request.applymarker(mark)
+
+    actual = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(actual, expected)
+
+
+@xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
+def test_categorical_dtype_unsorted(all_parsers):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b,c
+1,b,3.4
+1,b,3.4
+2,a,4.5"""
+    expected = DataFrame(
+        {
+            "a": Categorical(["1", "1", "2"]),
+            "b": Categorical(["b", "b", "a"]),
+            "c": Categorical(["3.4", "3.4", "4.5"]),
+        }
+    )
+    actual = parser.read_csv(StringIO(data), dtype="category")
+    tm.assert_frame_equal(actual, expected)
+
+
+@xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
+def test_categorical_dtype_missing(all_parsers):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b,c
+1,b,3.4
+1,nan,3.4
+2,a,4.5"""
+    expected = DataFrame(
+        {
+            "a": Categorical(["1", "1", "2"]),
+            "b": Categorical(["b", np.nan, "a"]),
+            "c": Categorical(["3.4", "3.4", "4.5"]),
+        }
+    )
+    actual = parser.read_csv(StringIO(data), dtype="category")
+    tm.assert_frame_equal(actual, expected)
+
+
+@xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
+@pytest.mark.slow
+def test_categorical_dtype_high_cardinality_numeric(all_parsers, monkeypatch):
+    # see gh-18186
+    # was an issue with C parser, due to DEFAULT_BUFFER_HEURISTIC
+    parser = all_parsers
+    heuristic = 2**5
+    data = np.sort([str(i) for i in range(heuristic + 1)])
+    expected = DataFrame({"a": Categorical(data, ordered=True)})
+    with monkeypatch.context() as m:
+        m.setattr(libparsers, "DEFAULT_BUFFER_HEURISTIC", heuristic)
+        actual = parser.read_csv(StringIO("a\n" + "\n".join(data)), dtype="category")
+    actual["a"] = actual["a"].cat.reorder_categories(
+        np.sort(actual.a.cat.categories), ordered=True
+    )
+    tm.assert_frame_equal(actual, expected)
+
+
+def test_categorical_dtype_utf16(all_parsers, csv_dir_path):
+    # see gh-10153
+    pth = os.path.join(csv_dir_path, "utf16_ex.txt")
+    parser = all_parsers
+    encoding = "utf-16"
+    sep = "\t"
+
+    expected = parser.read_csv(pth, sep=sep, encoding=encoding)
+    expected = expected.apply(Categorical)
+
+    actual = parser.read_csv(pth, sep=sep, encoding=encoding, dtype="category")
+    tm.assert_frame_equal(actual, expected)
+
+
+def test_categorical_dtype_chunksize_infer_categories(all_parsers):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b
+1,a
+1,b
+1,b
+2,c"""
+    expecteds = [
+        DataFrame({"a": [1, 1], "b": Categorical(["a", "b"])}),
+        DataFrame({"a": [1, 2], "b": Categorical(["b", "c"])}, index=[2, 3]),
+    ]
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), dtype={"b": "category"}, chunksize=2)
+        return
+
+    with parser.read_csv(
+        StringIO(data), dtype={"b": "category"}, chunksize=2
+    ) as actuals:
+        for actual, expected in zip(actuals, expecteds):
+            tm.assert_frame_equal(actual, expected)
+
+
+def test_categorical_dtype_chunksize_explicit_categories(all_parsers):
+    # see gh-10153
+    parser = all_parsers
+    data = """a,b
+1,a
+1,b
+1,b
+2,c"""
+    cats = ["a", "b", "c"]
+    expecteds = [
+        DataFrame({"a": [1, 1], "b": Categorical(["a", "b"], categories=cats)}),
+        DataFrame(
+            {"a": [1, 2], "b": Categorical(["b", "c"], categories=cats)},
+            index=[2, 3],
+        ),
+    ]
+    dtype = CategoricalDtype(cats)
+
+    if parser.engine == "pyarrow":
+        msg = "The 'chunksize' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), dtype={"b": dtype}, chunksize=2)
+        return
+
+    with parser.read_csv(StringIO(data), dtype={"b": dtype}, chunksize=2) as actuals:
+        for actual, expected in zip(actuals, expecteds):
+            tm.assert_frame_equal(actual, expected)
+
+
+def test_categorical_dtype_latin1(all_parsers, csv_dir_path):
+    # see gh-10153
+    pth = os.path.join(csv_dir_path, "unicode_series.csv")
+    parser = all_parsers
+    encoding = "latin-1"
+
+    expected = parser.read_csv(pth, header=None, encoding=encoding)
+    expected[1] = Categorical(expected[1])
+
+    actual = parser.read_csv(pth, header=None, encoding=encoding, dtype={1: "category"})
+    tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.parametrize("ordered", [False, True])
+@pytest.mark.parametrize(
+    "categories",
+    [["a", "b", "c"], ["a", "c", "b"], ["a", "b", "c", "d"], ["c", "b", "a"]],
+)
+def test_categorical_category_dtype(all_parsers, categories, ordered):
+    parser = all_parsers
+    data = """a,b
+1,a
+1,b
+1,b
+2,c"""
+    expected = DataFrame(
+        {
+            "a": [1, 1, 1, 2],
+            "b": Categorical(
+                ["a", "b", "b", "c"], categories=categories, ordered=ordered
+            ),
+        }
+    )
+
+    dtype = {"b": CategoricalDtype(categories=categories, ordered=ordered)}
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_category_dtype_unsorted(all_parsers):
+    parser = all_parsers
+    data = """a,b
+1,a
+1,b
+1,b
+2,c"""
+    dtype = CategoricalDtype(["c", "b", "a"])
+    expected = DataFrame(
+        {
+            "a": [1, 1, 1, 2],
+            "b": Categorical(["a", "b", "b", "c"], categories=["c", "b", "a"]),
+        }
+    )
+
+    result = parser.read_csv(StringIO(data), dtype={"b": dtype})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_coerces_numeric(all_parsers):
+    parser = all_parsers
+    dtype = {"b": CategoricalDtype([1, 2, 3])}
+
+    data = "b\n1\n1\n2\n3"
+    expected = DataFrame({"b": Categorical([1, 1, 2, 3])})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_coerces_datetime(all_parsers):
+    parser = all_parsers
+    dti = pd.DatetimeIndex(["2017-01-01", "2018-01-01", "2019-01-01"], freq=None)
+    dtype = {"b": CategoricalDtype(dti)}
+
+    data = "b\n2017-01-01\n2018-01-01\n2019-01-01"
+    expected = DataFrame({"b": Categorical(dtype["b"].categories)})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_coerces_timestamp(all_parsers):
+    parser = all_parsers
+    dtype = {"b": CategoricalDtype([Timestamp("2014")])}
+
+    data = "b\n2014-01-01\n2014-01-01"
+    expected = DataFrame({"b": Categorical([Timestamp("2014")] * 2)})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_coerces_timedelta(all_parsers):
+    parser = all_parsers
+    dtype = {"b": CategoricalDtype(pd.to_timedelta(["1h", "2h", "3h"]))}
+
+    data = "b\n1h\n2h\n3h"
+    expected = DataFrame({"b": Categorical(dtype["b"].categories)})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        "b\nTrue\nFalse\nNA\nFalse",
+        "b\ntrue\nfalse\nNA\nfalse",
+        "b\nTRUE\nFALSE\nNA\nFALSE",
+        "b\nTrue\nFalse\nNA\nFALSE",
+    ],
+)
+def test_categorical_dtype_coerces_boolean(all_parsers, data):
+    # see gh-20498
+    parser = all_parsers
+    dtype = {"b": CategoricalDtype([False, True])}
+    expected = DataFrame({"b": Categorical([True, False, None, False])})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_unexpected_categories(all_parsers):
+    parser = all_parsers
+    dtype = {"b": CategoricalDtype(["a", "b", "d", "e"])}
+
+    data = "b\nd\na\nc\nd"  # Unexpected c
+    expected = DataFrame({"b": Categorical(list("dacd"), dtype=dtype["b"])})
+
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/dtypes/test_dtypes_basic.py b/pandas/pandas/tests/io/parser/dtypes/test_dtypes_basic.py
new file mode 100644
index 0000000000000..75b7cf0d42cb8
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/dtypes/test_dtypes_basic.py
@@ -0,0 +1,642 @@
+"""
+Tests dtype specification during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from collections import defaultdict
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas.errors import ParserWarning
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.arrays import IntegerArray
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+
+@pytest.mark.parametrize("dtype", [str, object])
+@pytest.mark.parametrize("check_orig", [True, False])
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtype_all_columns(all_parsers, dtype, check_orig, using_infer_string):
+    # see gh-3795, gh-6607
+    parser = all_parsers
+
+    df = DataFrame(
+        np.random.default_rng(2).random((5, 2)).round(4),
+        columns=list("AB"),
+        index=["1A", "1B", "1C", "1D", "1E"],
+    )
+
+    with tm.ensure_clean("__passing_str_as_dtype__.csv") as path:
+        df.to_csv(path)
+
+        result = parser.read_csv(path, dtype=dtype, index_col=0)
+
+        if check_orig:
+            expected = df.copy()
+            result = result.astype(float)
+        elif using_infer_string and dtype is str:
+            expected = df.astype(str)
+        else:
+            expected = df.astype(str).astype(object)
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtype_per_column(all_parsers):
+    parser = all_parsers
+    data = """\
+one,two
+1,2.5
+2,3.5
+3,4.5
+4,5.5"""
+    expected = DataFrame(
+        [[1, "2.5"], [2, "3.5"], [3, "4.5"], [4, "5.5"]], columns=["one", "two"]
+    )
+    expected["one"] = expected["one"].astype(np.float64)
+
+    result = parser.read_csv(StringIO(data), dtype={"one": np.float64, 1: str})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_invalid_dtype_per_column(all_parsers):
+    parser = all_parsers
+    data = """\
+one,two
+1,2.5
+2,3.5
+3,4.5
+4,5.5"""
+
+    with pytest.raises(TypeError, match="data type [\"']foo[\"'] not understood"):
+        parser.read_csv(StringIO(data), dtype={"one": "foo", 1: "int"})
+
+
+def test_raise_on_passed_int_dtype_with_nas(all_parsers):
+    # see gh-2631
+    parser = all_parsers
+    data = """YEAR, DOY, a
+2001,106380451,10
+2001,,11
+2001,106380451,67"""
+
+    if parser.engine == "c":
+        msg = "Integer column has NA values"
+    elif parser.engine == "pyarrow":
+        msg = "The 'skipinitialspace' option is not supported with the 'pyarrow' engine"
+    else:
+        msg = "Unable to convert column DOY"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), dtype={"DOY": np.int64}, skipinitialspace=True)
+
+
+def test_dtype_with_converters(all_parsers):
+    parser = all_parsers
+    data = """a,b
+1.1,2.2
+1.2,2.3"""
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data), dtype={"a": "i8"}, converters={"a": lambda x: str(x)}
+            )
+        return
+
+    # Dtype spec ignored if converted specified.
+    result = parser.read_csv_check_warnings(
+        ParserWarning,
+        "Both a converter and dtype were specified for column a "
+        "- only the converter will be used.",
+        StringIO(data),
+        dtype={"a": "i8"},
+        converters={"a": lambda x: str(x)},
+    )
+    expected = DataFrame({"a": ["1.1", "1.2"], "b": [2.2, 2.3]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_numeric_dtype(all_parsers, any_real_numpy_dtype):
+    data = "0\n1"
+    parser = all_parsers
+    expected = DataFrame([0, 1], dtype=any_real_numpy_dtype)
+
+    result = parser.read_csv(StringIO(data), header=None, dtype=any_real_numpy_dtype)
+    tm.assert_frame_equal(expected, result, check_column_type=False)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_boolean_dtype(all_parsers):
+    parser = all_parsers
+    data = "\n".join(
+        [
+            "a",
+            "True",
+            "TRUE",
+            "true",
+            "1",
+            "1.0",
+            "False",
+            "FALSE",
+            "false",
+            "0",
+            "0.0",
+            "NaN",
+            "nan",
+            "NA",
+            "null",
+            "NULL",
+        ]
+    )
+
+    result = parser.read_csv(StringIO(data), dtype="boolean")
+    expected = DataFrame(
+        {
+            "a": pd.array(
+                [
+                    True,
+                    True,
+                    True,
+                    True,
+                    True,
+                    False,
+                    False,
+                    False,
+                    False,
+                    False,
+                    None,
+                    None,
+                    None,
+                    None,
+                    None,
+                ],
+                dtype="boolean",
+            )
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_delimiter_with_usecols_and_parse_dates(all_parsers):
+    # GH#35873
+    result = all_parsers.read_csv(
+        StringIO('"dump","-9,1","-9,1",20101010'),
+        engine="python",
+        names=["col", "col1", "col2", "col3"],
+        usecols=["col1", "col2", "col3"],
+        parse_dates=["col3"],
+        decimal=",",
+    )
+    expected = DataFrame(
+        {"col1": [-9.1], "col2": [-9.1], "col3": [Timestamp("2010-10-10")]}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("thousands", ["_", None])
+def test_decimal_and_exponential(
+    request, python_parser_only, numeric_decimal, thousands
+):
+    # GH#31920
+    decimal_number_check(request, python_parser_only, numeric_decimal, thousands, None)
+
+
+@pytest.mark.parametrize("thousands", ["_", None])
+@pytest.mark.parametrize("float_precision", [None, "legacy", "high", "round_trip"])
+def test_1000_sep_decimal_float_precision(
+    request, c_parser_only, numeric_decimal, float_precision, thousands
+):
+    # test decimal and thousand sep handling in across 'float_precision'
+    # parsers
+    decimal_number_check(
+        request, c_parser_only, numeric_decimal, thousands, float_precision
+    )
+    text, value = numeric_decimal
+    text = " " + text + " "
+    if isinstance(value, str):  # the negative cases (parse as text)
+        value = " " + value + " "
+    decimal_number_check(
+        request, c_parser_only, (text, value), thousands, float_precision
+    )
+
+
+def decimal_number_check(request, parser, numeric_decimal, thousands, float_precision):
+    # GH#31920
+    value = numeric_decimal[0]
+    if thousands is None and value in ("1_,", "1_234,56", "1_234,56e0"):
+        request.applymarker(
+            pytest.mark.xfail(reason=f"thousands={thousands} and sep is in {value}")
+        )
+    df = parser.read_csv(
+        StringIO(value),
+        float_precision=float_precision,
+        sep="|",
+        thousands=thousands,
+        decimal=",",
+        header=None,
+    )
+    val = df.iloc[0, 0]
+    assert val == numeric_decimal[1]
+
+
+@pytest.mark.parametrize("float_precision", [None, "legacy", "high", "round_trip"])
+def test_skip_whitespace(c_parser_only, float_precision):
+    DATA = """id\tnum\t
+1\t1.2 \t
+1\t 2.1\t
+2\t 1\t
+2\t 1.2 \t
+"""
+    df = c_parser_only.read_csv(
+        StringIO(DATA),
+        float_precision=float_precision,
+        sep="\t",
+        header=0,
+        dtype={1: np.float64},
+    )
+    tm.assert_series_equal(df.iloc[:, 1], pd.Series([1.2, 2.1, 1.0, 1.2], name="num"))
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_true_values_cast_to_bool(all_parsers):
+    # GH#34655
+    text = """a,b
+yes,xxx
+no,yyy
+1,zzz
+0,aaa
+    """
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO(text),
+        true_values=["yes"],
+        false_values=["no"],
+        dtype={"a": "boolean"},
+    )
+    expected = DataFrame(
+        {"a": [True, False, True, False], "b": ["xxx", "yyy", "zzz", "aaa"]}
+    )
+    expected["a"] = expected["a"].astype("boolean")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+@pytest.mark.parametrize("dtypes, exp_value", [({}, "1"), ({"a.1": "int64"}, 1)])
+def test_dtype_mangle_dup_cols(all_parsers, dtypes, exp_value):
+    # GH#35211
+    parser = all_parsers
+    data = """a,a\n1,1"""
+    dtype_dict = {"a": str, **dtypes}
+    # GH#42462
+    dtype_dict_copy = dtype_dict.copy()
+    result = parser.read_csv(StringIO(data), dtype=dtype_dict)
+    expected = DataFrame({"a": ["1"], "a.1": [exp_value]})
+    assert dtype_dict == dtype_dict_copy, "dtype dict changed"
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtype_mangle_dup_cols_single_dtype(all_parsers):
+    # GH#42022
+    parser = all_parsers
+    data = """a,a\n1,1"""
+    result = parser.read_csv(StringIO(data), dtype=str)
+    expected = DataFrame({"a": ["1"], "a.1": ["1"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtype_multi_index(all_parsers):
+    # GH 42446
+    parser = all_parsers
+    data = "A,B,B\nX,Y,Z\n1,2,3"
+
+    result = parser.read_csv(
+        StringIO(data),
+        header=list(range(2)),
+        dtype={
+            ("A", "X"): np.int32,
+            ("B", "Y"): np.int32,
+            ("B", "Z"): np.float32,
+        },
+    )
+
+    expected = DataFrame(
+        {
+            ("A", "X"): np.int32([1]),
+            ("B", "Y"): np.int32([2]),
+            ("B", "Z"): np.float32([3]),
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nullable_int_dtype(all_parsers, any_int_ea_dtype):
+    # GH 25472
+    parser = all_parsers
+    dtype = any_int_ea_dtype
+
+    data = """a,b,c
+,3,5
+1,,6
+2,4,"""
+    expected = DataFrame(
+        {
+            "a": pd.array([pd.NA, 1, 2], dtype=dtype),
+            "b": pd.array([3, pd.NA, 4], dtype=dtype),
+            "c": pd.array([5, 6, pd.NA], dtype=dtype),
+        }
+    )
+    actual = parser.read_csv(StringIO(data), dtype=dtype)
+    tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+@pytest.mark.parametrize("default", ["float", "float64"])
+def test_dtypes_defaultdict(all_parsers, default):
+    # GH#41574
+    data = """a,b
+1,2
+"""
+    dtype = defaultdict(lambda: default, a="int64")
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    expected = DataFrame({"a": [1], "b": 2.0})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtypes_defaultdict_mangle_dup_cols(all_parsers):
+    # GH#41574
+    data = """a,b,a,b,b.1
+1,2,3,4,5
+"""
+    dtype = defaultdict(lambda: "float64", a="int64")
+    dtype["b.1"] = "int64"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), dtype=dtype)
+    expected = DataFrame({"a": [1], "b": [2.0], "a.1": [3], "b.2": [4.0], "b.1": [5]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_dtypes_defaultdict_invalid(all_parsers):
+    # GH#41574
+    data = """a,b
+1,2
+"""
+    dtype = defaultdict(lambda: "invalid_dtype", a="int64")
+    parser = all_parsers
+    with pytest.raises(TypeError, match="not understood"):
+        parser.read_csv(StringIO(data), dtype=dtype)
+
+
+def test_dtype_backend(all_parsers):
+    # GH#36712
+
+    parser = all_parsers
+
+    data = """a,b,c,d,e,f,g,h,i,j
+1,2.5,True,a,,,,,12-31-2019,
+3,4.5,False,b,6,7.5,True,a,12-31-2019,
+"""
+    result = parser.read_csv(
+        StringIO(data), dtype_backend="numpy_nullable", parse_dates=["i"]
+    )
+    expected = DataFrame(
+        {
+            "a": pd.Series([1, 3], dtype="Int64"),
+            "b": pd.Series([2.5, 4.5], dtype="Float64"),
+            "c": pd.Series([True, False], dtype="boolean"),
+            "d": pd.Series(["a", "b"], dtype="string"),
+            "e": pd.Series([pd.NA, 6], dtype="Int64"),
+            "f": pd.Series([pd.NA, 7.5], dtype="Float64"),
+            "g": pd.Series([pd.NA, True], dtype="boolean"),
+            "h": pd.Series([pd.NA, "a"], dtype="string"),
+            "i": pd.Series([Timestamp("2019-12-31")] * 2),
+            "j": pd.Series([pd.NA, pd.NA], dtype="Int64"),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dtype_backend_and_dtype(all_parsers):
+    # GH#36712
+
+    parser = all_parsers
+
+    data = """a,b
+1,2.5
+,
+"""
+    result = parser.read_csv(
+        StringIO(data), dtype_backend="numpy_nullable", dtype="float64"
+    )
+    expected = DataFrame({"a": [1.0, np.nan], "b": [2.5, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dtype_backend_string(all_parsers, string_storage):
+    # GH#36712
+    with pd.option_context("mode.string_storage", string_storage):
+        parser = all_parsers
+
+        data = """a,b
+a,x
+b,
+"""
+        result = parser.read_csv(StringIO(data), dtype_backend="numpy_nullable")
+
+        expected = DataFrame(
+            {
+                "a": pd.array(["a", "b"], dtype=pd.StringDtype(string_storage)),
+                "b": pd.array(["x", pd.NA], dtype=pd.StringDtype(string_storage)),
+            },
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dtype_backend_ea_dtype_specified(all_parsers):
+    # GH#491496
+    data = """a,b
+1,2
+"""
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO(data), dtype="Int64", dtype_backend="numpy_nullable"
+    )
+    expected = DataFrame({"a": [1], "b": 2}, dtype="Int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dtype_backend_pyarrow(all_parsers, request):
+    # GH#36712
+    pa = pytest.importorskip("pyarrow")
+    parser = all_parsers
+
+    data = """a,b,c,d,e,f,g,h,i,j
+1,2.5,True,a,,,,,12-31-2019,
+3,4.5,False,b,6,7.5,True,a,12-31-2019,
+"""
+    result = parser.read_csv(StringIO(data), dtype_backend="pyarrow", parse_dates=["i"])
+    expected = DataFrame(
+        {
+            "a": pd.Series([1, 3], dtype="int64[pyarrow]"),
+            "b": pd.Series([2.5, 4.5], dtype="float64[pyarrow]"),
+            "c": pd.Series([True, False], dtype="bool[pyarrow]"),
+            "d": pd.Series(["a", "b"], dtype=pd.ArrowDtype(pa.string())),
+            "e": pd.Series([pd.NA, 6], dtype="int64[pyarrow]"),
+            "f": pd.Series([pd.NA, 7.5], dtype="float64[pyarrow]"),
+            "g": pd.Series([pd.NA, True], dtype="bool[pyarrow]"),
+            "h": pd.Series(
+                [pd.NA, "a"],
+                dtype=pd.ArrowDtype(pa.string()),
+            ),
+            "i": pd.Series([Timestamp("2019-12-31")] * 2),
+            "j": pd.Series([pd.NA, pd.NA], dtype="null[pyarrow]"),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# pyarrow engine failing:
+# https://github.com/pandas-dev/pandas/issues/56136
+@pytest.mark.usefixtures("pyarrow_xfail")
+def test_ea_int_avoid_overflow(all_parsers):
+    # GH#32134
+    parser = all_parsers
+    data = """a,b
+1,1
+,1
+1582218195625938945,1
+"""
+    result = parser.read_csv(StringIO(data), dtype={"a": "Int64"})
+    expected = DataFrame(
+        {
+            "a": IntegerArray(
+                np.array([1, 1, 1582218195625938945]), np.array([False, True, False])
+            ),
+            "b": 1,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_string_inference(all_parsers):
+    # GH#54430
+    dtype = pd.StringDtype(na_value=np.nan)
+
+    data = """a,b
+x,1
+y,2
+,3"""
+    parser = all_parsers
+    with pd.option_context("future.infer_string", True):
+        result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame(
+        {"a": pd.Series(["x", "y", None], dtype=dtype), "b": [1, 2, 3]},
+        columns=pd.Index(["a", "b"], dtype=dtype),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["O", object, "object", np.object_, str, np.str_])
+def test_string_inference_object_dtype(all_parsers, dtype, using_infer_string):
+    # GH#56047
+    data = """a,b
+x,a
+y,a
+z,a"""
+    parser = all_parsers
+    with pd.option_context("future.infer_string", True):
+        result = parser.read_csv(StringIO(data), dtype=dtype)
+
+    expected_dtype = pd.StringDtype(na_value=np.nan) if dtype is str else object
+    expected = DataFrame(
+        {
+            "a": pd.Series(["x", "y", "z"], dtype=expected_dtype),
+            "b": pd.Series(["a", "a", "a"], dtype=expected_dtype),
+        },
+        columns=pd.Index(["a", "b"], dtype=pd.StringDtype(na_value=np.nan)),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    with pd.option_context("future.infer_string", True):
+        result = parser.read_csv(StringIO(data), dtype={"a": dtype})
+
+    expected = DataFrame(
+        {
+            "a": pd.Series(["x", "y", "z"], dtype=expected_dtype),
+            "b": pd.Series(["a", "a", "a"], dtype=pd.StringDtype(na_value=np.nan)),
+        },
+        columns=pd.Index(["a", "b"], dtype=pd.StringDtype(na_value=np.nan)),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+def test_accurate_parsing_of_large_integers(all_parsers):
+    # GH#52505
+    data = """SYMBOL,MOMENT,ID,ID_DEAL
+AAPL,20230301181139587,1925036343869802844,
+AAPL,20230301181139587,2023552585717889863,2023552585717263358
+NVDA,20230301181139587,2023552585717889863,2023552585717263359
+AMC,20230301181139587,2023552585717889863,2023552585717263360
+AMZN,20230301181139587,2023552585717889759,2023552585717263360
+MSFT,20230301181139587,2023552585717889863,2023552585717263361
+NVDA,20230301181139587,2023552585717889827,2023552585717263361"""
+    orders = all_parsers.read_csv(StringIO(data), dtype={"ID_DEAL": pd.Int64Dtype()})
+    assert len(orders.loc[orders["ID_DEAL"] == 2023552585717263358, "ID_DEAL"]) == 1
+    assert len(orders.loc[orders["ID_DEAL"] == 2023552585717263359, "ID_DEAL"]) == 1
+    assert len(orders.loc[orders["ID_DEAL"] == 2023552585717263360, "ID_DEAL"]) == 2
+    assert len(orders.loc[orders["ID_DEAL"] == 2023552585717263361, "ID_DEAL"]) == 2
+
+
+def test_dtypes_with_usecols(all_parsers):
+    # GH#54868
+
+    parser = all_parsers
+    data = """a,b,c
+1,2,3
+4,5,6"""
+
+    result = parser.read_csv(StringIO(data), usecols=["a", "c"], dtype={"a": object})
+    if parser.engine == "pyarrow":
+        values = [1, 4]
+    else:
+        values = ["1", "4"]
+    expected = DataFrame({"a": pd.Series(values, dtype=object), "c": [3, 6]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_index_col_with_dtype_no_rangeindex(all_parsers):
+    data = StringIO("345.5,519.5,0\n519.5,726.5,1")
+    result = all_parsers.read_csv(
+        data,
+        header=None,
+        names=["start", "stop", "bin_id"],
+        dtype={"start": np.float32, "stop": np.float32, "bin_id": np.uint32},
+        index_col="bin_id",
+    ).index
+    expected = pd.Index([0, 1], dtype=np.uint32, name="bin_id")
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/dtypes/test_empty.py b/pandas/pandas/tests/io/parser/dtypes/test_empty.py
new file mode 100644
index 0000000000000..ebc61e7f0ca2b
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/dtypes/test_empty.py
@@ -0,0 +1,181 @@
+"""
+Tests dtype specification during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_dtype_all_columns_empty(all_parsers):
+    # see gh-12048
+    parser = all_parsers
+    result = parser.read_csv(StringIO("A,B"), dtype=str)
+
+    expected = DataFrame({"A": [], "B": []}, dtype=str)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_pass_dtype(all_parsers):
+    parser = all_parsers
+
+    data = "one,two"
+    result = parser.read_csv(StringIO(data), dtype={"one": "u1"})
+
+    expected = DataFrame(
+        {"one": np.empty(0, dtype="u1"), "two": np.empty(0, dtype=object)},
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_index_pass_dtype(all_parsers):
+    parser = all_parsers
+
+    data = "one,two"
+    result = parser.read_csv(
+        StringIO(data), index_col=["one"], dtype={"one": "u1", 1: "f"}
+    )
+
+    expected = DataFrame(
+        {"two": np.empty(0, dtype="f")}, index=Index([], dtype="u1", name="one")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_multi_index_pass_dtype(all_parsers):
+    parser = all_parsers
+
+    data = "one,two,three"
+    result = parser.read_csv(
+        StringIO(data), index_col=["one", "two"], dtype={"one": "u1", 1: "f8"}
+    )
+
+    exp_idx = MultiIndex.from_arrays(
+        [np.empty(0, dtype="u1"), np.empty(0, dtype=np.float64)],
+        names=["one", "two"],
+    )
+    expected = DataFrame({"three": np.empty(0, dtype=object)}, index=exp_idx)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_mangled_column_pass_dtype_by_names(all_parsers):
+    parser = all_parsers
+
+    data = "one,one"
+    result = parser.read_csv(StringIO(data), dtype={"one": "u1", "one.1": "f"})
+
+    expected = DataFrame(
+        {"one": np.empty(0, dtype="u1"), "one.1": np.empty(0, dtype="f")},
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_mangled_column_pass_dtype_by_indexes(all_parsers):
+    parser = all_parsers
+
+    data = "one,one"
+    result = parser.read_csv(StringIO(data), dtype={0: "u1", 1: "f"})
+
+    expected = DataFrame(
+        {"one": np.empty(0, dtype="u1"), "one.1": np.empty(0, dtype="f")},
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_dup_column_pass_dtype_by_indexes(all_parsers):
+    # see gh-9424
+    parser = all_parsers
+    expected = concat(
+        [Series([], name="one", dtype="u1"), Series([], name="one.1", dtype="f")],
+        axis=1,
+    )
+
+    data = "one,one"
+    result = parser.read_csv(StringIO(data), dtype={0: "u1", 1: "f"})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_empty_with_dup_column_pass_dtype_by_indexes_raises(all_parsers):
+    # see gh-9424
+    parser = all_parsers
+    expected = concat(
+        [Series([], name="one", dtype="u1"), Series([], name="one.1", dtype="f")],
+        axis=1,
+    )
+    expected.index = expected.index.astype(object)
+
+    with pytest.raises(ValueError, match="Duplicate names"):
+        parser.read_csv(StringIO(""), names=["one", "one"], dtype={0: "u1", 1: "f"})
+
+
+@pytest.mark.parametrize(
+    "dtype,expected",
+    [
+        (np.float64, DataFrame(columns=["a", "b"], dtype=np.float64)),
+        (
+            "category",
+            DataFrame({"a": Categorical([]), "b": Categorical([])}),
+        ),
+        (
+            {"a": "category", "b": "category"},
+            DataFrame({"a": Categorical([]), "b": Categorical([])}),
+        ),
+        ("datetime64[ns]", DataFrame(columns=["a", "b"], dtype="datetime64[ns]")),
+        (
+            "timedelta64[ns]",
+            DataFrame(
+                {
+                    "a": Series([], dtype="timedelta64[ns]"),
+                    "b": Series([], dtype="timedelta64[ns]"),
+                },
+            ),
+        ),
+        (
+            {"a": np.int64, "b": np.int32},
+            DataFrame(
+                {"a": Series([], dtype=np.int64), "b": Series([], dtype=np.int32)},
+            ),
+        ),
+        (
+            {0: np.int64, 1: np.int32},
+            DataFrame(
+                {"a": Series([], dtype=np.int64), "b": Series([], dtype=np.int32)},
+            ),
+        ),
+        (
+            {"a": np.int64, 1: np.int32},
+            DataFrame(
+                {"a": Series([], dtype=np.int64), "b": Series([], dtype=np.int32)},
+            ),
+        ),
+    ],
+)
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_dtype(all_parsers, dtype, expected):
+    # see gh-14712
+    parser = all_parsers
+    data = "a,b"
+
+    result = parser.read_csv(StringIO(data), header=0, dtype=dtype)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_c_parser_only.py b/pandas/pandas/tests/io/parser/test_c_parser_only.py
new file mode 100644
index 0000000000000..11a30a26f91ef
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_c_parser_only.py
@@ -0,0 +1,647 @@
+"""
+Tests that apply specifically to the CParser. Unless specifically stated
+as a CParser-specific issue, the goal is to eventually move as many of
+these tests out of this module as soon as the Python parser can accept
+further arguments when parsing.
+"""
+
+from decimal import Decimal
+from io import (
+    BytesIO,
+    StringIO,
+    TextIOWrapper,
+)
+import mmap
+import os
+import tarfile
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+from pandas.compat.numpy import np_version_gte1p24
+from pandas.errors import (
+    ParserError,
+    ParserWarning,
+)
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    concat,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "malformed",
+    ["1\r1\r1\r 1\r 1\r", "1\r1\r1\r 1\r 1\r11\r", "1\r1\r1\r 1\r 1\r11\r1\r"],
+    ids=["words pointer", "stream pointer", "lines pointer"],
+)
+def test_buffer_overflow(c_parser_only, malformed):
+    # see gh-9205: test certain malformed input files that cause
+    # buffer overflows in tokenizer.c
+    msg = "Buffer overflow caught - possible malformed input file."
+    parser = c_parser_only
+
+    with pytest.raises(ParserError, match=msg):
+        parser.read_csv(StringIO(malformed))
+
+
+def test_delim_whitespace_custom_terminator(c_parser_only):
+    # See gh-12912
+    data = "a b c~1 2 3~4 5 6~7 8 9"
+    parser = c_parser_only
+
+    df = parser.read_csv(StringIO(data), lineterminator="~", sep=r"\s+")
+    expected = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=["a", "b", "c"])
+    tm.assert_frame_equal(df, expected)
+
+
+def test_dtype_and_names_error(c_parser_only):
+    # see gh-8833: passing both dtype and names
+    # resulting in an error reporting issue
+    parser = c_parser_only
+    data = """
+1.0 1
+2.0 2
+3.0 3
+"""
+    # base cases
+    result = parser.read_csv(StringIO(data), sep=r"\s+", header=None)
+    expected = DataFrame([[1.0, 1], [2.0, 2], [3.0, 3]])
+    tm.assert_frame_equal(result, expected)
+
+    result = parser.read_csv(StringIO(data), sep=r"\s+", header=None, names=["a", "b"])
+    expected = DataFrame([[1.0, 1], [2.0, 2], [3.0, 3]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+    # fallback casting
+    result = parser.read_csv(
+        StringIO(data), sep=r"\s+", header=None, names=["a", "b"], dtype={"a": np.int32}
+    )
+    expected = DataFrame([[1, 1], [2, 2], [3, 3]], columns=["a", "b"])
+    expected["a"] = expected["a"].astype(np.int32)
+    tm.assert_frame_equal(result, expected)
+
+    data = """
+1.0 1
+nan 2
+3.0 3
+"""
+    # fallback casting, but not castable
+    warning = RuntimeWarning if np_version_gte1p24 else None
+    if not WASM:  # no fp exception support in wasm
+        with pytest.raises(ValueError, match="cannot safely convert"):
+            with tm.assert_produces_warning(warning, check_stacklevel=False):
+                parser.read_csv(
+                    StringIO(data),
+                    sep=r"\s+",
+                    header=None,
+                    names=["a", "b"],
+                    dtype={"a": np.int32},
+                )
+
+
+@pytest.mark.parametrize(
+    "match,kwargs",
+    [
+        # For each of these cases, all of the dtypes are valid, just unsupported.
+        (
+            (
+                "the dtype datetime64 is not supported for parsing, "
+                "pass this column using parse_dates instead"
+            ),
+            {"dtype": {"A": "datetime64", "B": "float64"}},
+        ),
+        (
+            (
+                "the dtype datetime64 is not supported for parsing, "
+                "pass this column using parse_dates instead"
+            ),
+            {"dtype": {"A": "datetime64", "B": "float64"}, "parse_dates": ["B"]},
+        ),
+        (
+            "the dtype timedelta64 is not supported for parsing",
+            {"dtype": {"A": "timedelta64", "B": "float64"}},
+        ),
+        (
+            f"the dtype {tm.ENDIAN}U8 is not supported for parsing",
+            {"dtype": {"A": "U8"}},
+        ),
+    ],
+    ids=["dt64-0", "dt64-1", "td64", f"{tm.ENDIAN}U8"],
+)
+def test_unsupported_dtype(c_parser_only, match, kwargs):
+    parser = c_parser_only
+    df = DataFrame(
+        np.random.default_rng(2).random((5, 2)),
+        columns=list("AB"),
+        index=["1A", "1B", "1C", "1D", "1E"],
+    )
+
+    with tm.ensure_clean("__unsupported_dtype__.csv") as path:
+        df.to_csv(path)
+
+        with pytest.raises(TypeError, match=match):
+            parser.read_csv(path, index_col=0, **kwargs)
+
+
+@td.skip_if_32bit
+@pytest.mark.slow
+# test numbers between 1 and 2
+@pytest.mark.parametrize("num", np.linspace(1.0, 2.0, num=21))
+def test_precise_conversion(c_parser_only, num):
+    parser = c_parser_only
+
+    normal_errors = []
+    precise_errors = []
+
+    def error(val: float, actual_val: Decimal) -> Decimal:
+        return abs(Decimal(f"{val:.100}") - actual_val)
+
+    # 25 decimal digits of precision
+    text = f"a\n{num:.25}"
+
+    normal_val = float(
+        parser.read_csv(StringIO(text), float_precision="legacy")["a"][0]
+    )
+    precise_val = float(parser.read_csv(StringIO(text), float_precision="high")["a"][0])
+    roundtrip_val = float(
+        parser.read_csv(StringIO(text), float_precision="round_trip")["a"][0]
+    )
+    actual_val = Decimal(text[2:])
+
+    normal_errors.append(error(normal_val, actual_val))
+    precise_errors.append(error(precise_val, actual_val))
+
+    # round-trip should match float()
+    assert roundtrip_val == float(text[2:])
+
+    assert sum(precise_errors) <= sum(normal_errors)
+    assert max(precise_errors) <= max(normal_errors)
+
+
+def test_usecols_dtypes(c_parser_only, using_infer_string):
+    parser = c_parser_only
+    data = """\
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+
+    result = parser.read_csv(
+        StringIO(data),
+        usecols=(0, 1, 2),
+        names=("a", "b", "c"),
+        header=None,
+        converters={"a": str},
+        dtype={"b": int, "c": float},
+    )
+    result2 = parser.read_csv(
+        StringIO(data),
+        usecols=(0, 2),
+        names=("a", "b", "c"),
+        header=None,
+        converters={"a": str},
+        dtype={"b": int, "c": float},
+    )
+
+    if using_infer_string:
+        assert (result.dtypes == ["string", int, float]).all()
+        assert (result2.dtypes == ["string", float]).all()
+    else:
+        assert (result.dtypes == [object, int, float]).all()
+        assert (result2.dtypes == [object, float]).all()
+
+
+def test_disable_bool_parsing(c_parser_only):
+    # see gh-2090
+
+    parser = c_parser_only
+    data = """A,B,C
+Yes,No,Yes
+No,Yes,Yes
+Yes,,Yes
+No,No,No"""
+
+    result = parser.read_csv(StringIO(data), dtype=object)
+    assert (result.dtypes == object).all()
+
+    result = parser.read_csv(StringIO(data), dtype=object, na_filter=False)
+    assert result["B"][2] == ""
+
+
+def test_custom_lineterminator(c_parser_only):
+    parser = c_parser_only
+    data = "a,b,c~1,2,3~4,5,6"
+
+    result = parser.read_csv(StringIO(data), lineterminator="~")
+    expected = parser.read_csv(StringIO(data.replace("~", "\n")))
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_ragged_csv(c_parser_only):
+    parser = c_parser_only
+    data = """1,2,3
+1,2,3,4
+1,2,3,4,5
+1,2
+1,2,3,4"""
+
+    nice_data = """1,2,3,,
+1,2,3,4,
+1,2,3,4,5
+1,2,,,
+1,2,3,4,"""
+    result = parser.read_csv(
+        StringIO(data), header=None, names=["a", "b", "c", "d", "e"]
+    )
+
+    expected = parser.read_csv(
+        StringIO(nice_data), header=None, names=["a", "b", "c", "d", "e"]
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+    # too many columns, cause segfault if not careful
+    data = "1,2\n3,4,5"
+
+    result = parser.read_csv(StringIO(data), header=None, names=range(50))
+    expected = parser.read_csv(StringIO(data), header=None, names=range(3)).reindex(
+        columns=range(50)
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_tokenize_CR_with_quoting(c_parser_only):
+    # see gh-3453
+    parser = c_parser_only
+    data = ' a,b,c\r"a,b","e,d","f,f"'
+
+    result = parser.read_csv(StringIO(data), header=None)
+    expected = parser.read_csv(StringIO(data.replace("\r", "\n")), header=None)
+    tm.assert_frame_equal(result, expected)
+
+    result = parser.read_csv(StringIO(data))
+    expected = parser.read_csv(StringIO(data.replace("\r", "\n")))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize("count", [3 * 2**n for n in range(6)])
+def test_grow_boundary_at_cap(c_parser_only, count):
+    # See gh-12494
+    #
+    # Cause of error was that the C parser
+    # was not increasing the buffer size when
+    # the desired space would fill the buffer
+    # to capacity, which would later cause a
+    # buffer overflow error when checking the
+    # EOF terminator of the CSV stream.
+    # 3 * 2^n commas was observed to break the parser
+    parser = c_parser_only
+
+    with StringIO("," * count) as s:
+        expected = DataFrame(columns=[f"Unnamed: {i}" for i in range(count + 1)])
+        df = parser.read_csv(s)
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize("encoding", [None, "utf-8"])
+def test_parse_trim_buffers(c_parser_only, encoding):
+    # This test is part of a bugfix for gh-13703. It attempts to
+    # to stress the system memory allocator, to cause it to move the
+    # stream buffer and either let the OS reclaim the region, or let
+    # other memory requests of parser otherwise modify the contents
+    # of memory space, where it was formally located.
+    # This test is designed to cause a `segfault` with unpatched
+    # `tokenizer.c`. Sometimes the test fails on `segfault`, other
+    # times it fails due to memory corruption, which causes the
+    # loaded DataFrame to differ from the expected one.
+
+    # Also force 'utf-8' encoding, so that `_string_convert` would take
+    # a different execution branch.
+
+    parser = c_parser_only
+
+    # Generate a large mixed-type CSV file on-the-fly (one record is
+    # approx 1.5KiB).
+    record_ = (
+        """9999-9,99:99,,,,ZZ,ZZ,,,ZZZ-ZZZZ,.Z-ZZZZ,-9.99,,,9.99,Z"""
+        """ZZZZ,,-99,9,ZZZ-ZZZZ,ZZ-ZZZZ,,9.99,ZZZ-ZZZZZ,ZZZ-ZZZZZ,"""
+        """ZZZ-ZZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,9"""
+        """99,ZZZ-ZZZZ,,ZZ-ZZZZ,,,,,ZZZZ,ZZZ-ZZZZZ,ZZZ-ZZZZ,,,9,9,"""
+        """9,9,99,99,999,999,ZZZZZ,ZZZ-ZZZZZ,ZZZ-ZZZZ,9,ZZ-ZZZZ,9."""
+        """99,ZZ-ZZZZ,ZZ-ZZZZ,,,,ZZZZ,,,ZZ,ZZ,,,,,,,,,,,,,9,,,999."""
+        """99,999.99,,,ZZZZZ,,,Z9,,,,,,,ZZZ,ZZZ,,,,,,,,,,,ZZZZZ,ZZ"""
+        """ZZZ,ZZZ-ZZZZZZ,ZZZ-ZZZZZZ,ZZ-ZZZZ,ZZ-ZZZZ,ZZ-ZZZZ,ZZ-ZZ"""
+        """ZZ,,,999999,999999,ZZZ,ZZZ,,,ZZZ,ZZZ,999.99,999.99,,,,Z"""
+        """ZZ-ZZZ,ZZZ-ZZZ,-9.99,-9.99,9,9,,99,,9.99,9.99,9,9,9.99,"""
+        """9.99,,,,9.99,9.99,,99,,99,9.99,9.99,,,ZZZ,ZZZ,,999.99,,"""
+        """999.99,ZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,,,ZZZZZ,ZZZZZ,ZZZ,ZZZ,9,9,"""
+        """,,,,,ZZZ-ZZZZ,ZZZ999Z,,,999.99,,999.99,ZZZ-ZZZZ,,,9.999"""
+        """,9.999,9.999,9.999,-9.999,-9.999,-9.999,-9.999,9.999,9."""
+        """999,9.999,9.999,9.999,9.999,9.999,9.999,99999,ZZZ-ZZZZ,"""
+        """,9.99,ZZZ,,,,,,,,ZZZ,,,,,9,,,,9,,,,,,,,,,ZZZ-ZZZZ,ZZZ-Z"""
+        """ZZZ,,ZZZZZ,ZZZZZ,ZZZZZ,ZZZZZ,,,9.99,,ZZ-ZZZZ,ZZ-ZZZZ,ZZ"""
+        """,999,,,,ZZ-ZZZZ,ZZZ,ZZZ,ZZZ-ZZZZ,ZZZ-ZZZZ,,,99.99,99.99"""
+        """,,,9.99,9.99,9.99,9.99,ZZZ-ZZZZ,,,ZZZ-ZZZZZ,,,,,-9.99,-"""
+        """9.99,-9.99,-9.99,,,,,,,,,ZZZ-ZZZZ,,9,9.99,9.99,99ZZ,,-9"""
+        """.99,-9.99,ZZZ-ZZZZ,,,,,,,ZZZ-ZZZZ,9.99,9.99,9999,,,,,,,"""
+        """,,,-9.9,Z/Z-ZZZZ,999.99,9.99,,999.99,ZZ-ZZZZ,ZZ-ZZZZ,9."""
+        """99,9.99,9.99,9.99,9.99,9.99,,ZZZ-ZZZZZ,ZZZ-ZZZZZ,ZZZ-ZZ"""
+        """ZZZ,ZZZ-ZZZZZ,ZZZ-ZZZZZ,ZZZ,ZZZ,ZZZ,ZZZ,9.99,,,-9.99,ZZ"""
+        """-ZZZZ,-999.99,,-9999,,999.99,,,,999.99,99.99,,,ZZ-ZZZZZ"""
+        """ZZZ,ZZ-ZZZZ-ZZZZZZZ,,,,ZZ-ZZ-ZZZZZZZZ,ZZZZZZZZ,ZZZ-ZZZZ"""
+        """,9999,999.99,ZZZ-ZZZZ,-9.99,-9.99,ZZZ-ZZZZ,99:99:99,,99"""
+        """,99,,9.99,,-99.99,,,,,,9.99,ZZZ-ZZZZ,-9.99,-9.99,9.99,9"""
+        """.99,,ZZZ,,,,,,,ZZZ,ZZZ,,,,,"""
+    )
+
+    # Set the number of lines so that a call to `parser_trim_buffers`
+    # is triggered: after a couple of full chunks are consumed a
+    # relatively small 'residual' chunk would cause reallocation
+    # within the parser.
+    chunksize, n_lines = 128, 2 * 128 + 15
+    csv_data = "\n".join([record_] * n_lines) + "\n"
+
+    # We will use StringIO to load the CSV from this text buffer.
+    # pd.read_csv() will iterate over the file in chunks and will
+    # finally read a residual chunk of really small size.
+
+    # Generate the expected output: manually create the dataframe
+    # by splitting by comma and repeating the `n_lines` times.
+    row = tuple(val_ if val_ else np.nan for val_ in record_.split(","))
+    expected = DataFrame(
+        [row for _ in range(n_lines)], dtype=object, columns=None, index=None
+    )
+
+    # Iterate over the CSV file in chunks of `chunksize` lines
+    with parser.read_csv(
+        StringIO(csv_data),
+        header=None,
+        dtype=object,
+        chunksize=chunksize,
+        encoding=encoding,
+    ) as chunks_:
+        result = concat(chunks_, axis=0, ignore_index=True)
+
+    # Check for data corruption if there was no segfault
+    tm.assert_frame_equal(result, expected)
+
+
+def test_internal_null_byte(c_parser_only):
+    # see gh-14012
+    #
+    # The null byte ('\x00') should not be used as a
+    # true line terminator, escape character, or comment
+    # character, only as a placeholder to indicate that
+    # none was specified.
+    #
+    # This test should be moved to test_common.py ONLY when
+    # Python's csv class supports parsing '\x00'.
+    parser = c_parser_only
+
+    names = ["a", "b", "c"]
+    data = "1,2,3\n4,\x00,6\n7,8,9"
+    expected = DataFrame([[1, 2.0, 3], [4, np.nan, 6], [7, 8, 9]], columns=names)
+
+    result = parser.read_csv(StringIO(data), names=names)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_nrows_large(c_parser_only):
+    # gh-7626 - Read only nrows of data in for large inputs (>262144b)
+    parser = c_parser_only
+    header_narrow = "\t".join(["COL_HEADER_" + str(i) for i in range(10)]) + "\n"
+    data_narrow = "\t".join(["somedatasomedatasomedata1" for _ in range(10)]) + "\n"
+    header_wide = "\t".join(["COL_HEADER_" + str(i) for i in range(15)]) + "\n"
+    data_wide = "\t".join(["somedatasomedatasomedata2" for _ in range(15)]) + "\n"
+    test_input = header_narrow + data_narrow * 1050 + header_wide + data_wide * 2
+
+    df = parser.read_csv(StringIO(test_input), sep="\t", nrows=1010)
+
+    assert df.size == 1010 * 10
+
+
+def test_float_precision_round_trip_with_text(c_parser_only):
+    # see gh-15140
+    parser = c_parser_only
+    df = parser.read_csv(StringIO("a"), header=None, float_precision="round_trip")
+    tm.assert_frame_equal(df, DataFrame({0: ["a"]}))
+
+
+def test_large_difference_in_columns(c_parser_only):
+    # see gh-14125
+    parser = c_parser_only
+
+    count = 10000
+    large_row = ("X," * count)[:-1] + "\n"
+    normal_row = "XXXXXX XXXXXX,111111111111111\n"
+    test_input = (large_row + normal_row * 6)[:-1]
+
+    result = parser.read_csv(StringIO(test_input), header=None, usecols=[0])
+    rows = test_input.split("\n")
+
+    expected = DataFrame([row.split(",")[0] for row in rows])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_data_after_quote(c_parser_only):
+    # see gh-15910
+    parser = c_parser_only
+
+    data = 'a\n1\n"b"a'
+    result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame({"a": ["1", "ba"]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comment_whitespace_delimited(c_parser_only):
+    parser = c_parser_only
+    test_input = """\
+1 2
+2 2 3
+3 2 3 # 3 fields
+4 2 3# 3 fields
+5 2 # 2 fields
+6 2# 2 fields
+7 # 1 field, NaN
+8# 1 field, NaN
+9 2 3 # skipped line
+# comment"""
+    with tm.assert_produces_warning(
+        ParserWarning, match="Skipping line", check_stacklevel=False
+    ):
+        df = parser.read_csv(
+            StringIO(test_input),
+            comment="#",
+            header=None,
+            delimiter="\\s+",
+            skiprows=0,
+            on_bad_lines="warn",
+        )
+    expected = DataFrame([[1, 2], [5, 2], [6, 2], [7, np.nan], [8, np.nan]])
+    tm.assert_frame_equal(df, expected)
+
+
+def test_file_like_no_next(c_parser_only):
+    # gh-16530: the file-like need not have a "next" or "__next__"
+    # attribute despite having an "__iter__" attribute.
+    #
+    # NOTE: This is only true for the C engine, not Python engine.
+    class NoNextBuffer(StringIO):
+        def __next__(self):
+            raise AttributeError("No next method")
+
+        next = __next__
+
+    parser = c_parser_only
+    data = "a\n1"
+
+    expected = DataFrame({"a": [1]})
+    result = parser.read_csv(NoNextBuffer(data))
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_buffer_rd_bytes_bad_unicode(c_parser_only):
+    # see gh-22748
+    t = BytesIO(b"\xb0")
+    t = TextIOWrapper(t, encoding="UTF-8", errors="surrogateescape")
+    msg = "'utf-8' codec can't encode character"
+    with pytest.raises(UnicodeError, match=msg):
+        c_parser_only.read_csv(t, encoding="UTF-8")
+
+
+@pytest.mark.parametrize("tar_suffix", [".tar", ".tar.gz"])
+def test_read_tarfile(c_parser_only, csv_dir_path, tar_suffix):
+    # see gh-16530
+    #
+    # Unfortunately, Python's CSV library can't handle
+    # tarfile objects (expects string, not bytes when
+    # iterating through a file-like).
+    parser = c_parser_only
+    tar_path = os.path.join(csv_dir_path, "tar_csv" + tar_suffix)
+
+    with tarfile.open(tar_path, "r") as tar:
+        data_file = tar.extractfile("tar_data.csv")
+
+        out = parser.read_csv(data_file)
+        expected = DataFrame({"a": [1]})
+        tm.assert_frame_equal(out, expected)
+
+
+def test_chunk_whitespace_on_boundary(c_parser_only):
+    # see gh-9735: this issue is C parser-specific (bug when
+    # parsing whitespace and characters at chunk boundary)
+    #
+    # This test case has a field too large for the Python parser / CSV library.
+    parser = c_parser_only
+
+    chunk1 = "a" * (1024 * 256 - 2) + "\na"
+    chunk2 = "\n a"
+    result = parser.read_csv(StringIO(chunk1 + chunk2), header=None)
+
+    expected = DataFrame(["a" * (1024 * 256 - 2), "a", " a"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+def test_file_handles_mmap(c_parser_only, csv1):
+    # gh-14418
+    #
+    # Don't close user provided file handles.
+    parser = c_parser_only
+
+    with open(csv1, encoding="utf-8") as f:
+        with mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ) as m:
+            parser.read_csv(m)
+            assert not m.closed
+
+
+def test_file_binary_mode(c_parser_only):
+    # see gh-23779
+    parser = c_parser_only
+    expected = DataFrame([[1, 2, 3], [4, 5, 6]])
+
+    with tm.ensure_clean() as path:
+        with open(path, "w", encoding="utf-8") as f:
+            f.write("1,2,3\n4,5,6")
+
+        with open(path, "rb") as f:
+            result = parser.read_csv(f, header=None)
+            tm.assert_frame_equal(result, expected)
+
+
+def test_unix_style_breaks(c_parser_only):
+    # GH 11020
+    parser = c_parser_only
+    with tm.ensure_clean() as path:
+        with open(path, "w", newline="\n", encoding="utf-8") as f:
+            f.write("blah\n\ncol_1,col_2,col_3\n\n")
+        result = parser.read_csv(path, skiprows=2, encoding="utf-8", engine="c")
+    expected = DataFrame(columns=["col_1", "col_2", "col_3"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("float_precision", [None, "legacy", "high", "round_trip"])
+@pytest.mark.parametrize(
+    "data,thousands,decimal",
+    [
+        (
+            """A|B|C
+1|2,334.01|5
+10|13|10.
+""",
+            ",",
+            ".",
+        ),
+        (
+            """A|B|C
+1|2.334,01|5
+10|13|10,
+""",
+            ".",
+            ",",
+        ),
+    ],
+)
+def test_1000_sep_with_decimal(
+    c_parser_only, data, thousands, decimal, float_precision
+):
+    parser = c_parser_only
+    expected = DataFrame({"A": [1, 10], "B": [2334.01, 13], "C": [5, 10.0]})
+
+    result = parser.read_csv(
+        StringIO(data),
+        sep="|",
+        thousands=thousands,
+        decimal=decimal,
+        float_precision=float_precision,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_float_precision_options(c_parser_only):
+    # GH 17154, 36228
+    parser = c_parser_only
+    s = "foo\n243.164\n"
+    df = parser.read_csv(StringIO(s))
+    df2 = parser.read_csv(StringIO(s), float_precision="high")
+
+    tm.assert_frame_equal(df, df2)
+
+    df3 = parser.read_csv(StringIO(s), float_precision="legacy")
+
+    assert not df.iloc[0, 0] == df3.iloc[0, 0]
+
+    msg = "Unrecognized float_precision option: junk"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(s), float_precision="junk")
diff --git a/pandas/pandas/tests/io/parser/test_comment.py b/pandas/pandas/tests/io/parser/test_comment.py
new file mode 100644
index 0000000000000..ba27b170aecdc
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_comment.py
@@ -0,0 +1,214 @@
+"""
+Tests that comments are properly handled during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("na_values", [None, ["NaN"]])
+def test_comment(all_parsers, na_values):
+    parser = all_parsers
+    data = """A,B,C
+1,2.,4.#hello world
+5.,NaN,10.0
+"""
+    expected = DataFrame(
+        [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"]
+    )
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", na_values=na_values)
+        return
+    result = parser.read_csv(StringIO(data), comment="#", na_values=na_values)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("read_kwargs", [{}, {"lineterminator": "*"}, {"sep": r"\s+"}])
+def test_line_comment(all_parsers, read_kwargs):
+    parser = all_parsers
+    data = """# empty
+A,B,C
+1,2.,4.#hello world
+#ignore this line
+5.,NaN,10.0
+"""
+    if read_kwargs.get("sep"):
+        data = data.replace(",", " ")
+    elif read_kwargs.get("lineterminator"):
+        data = data.replace("\n", read_kwargs.get("lineterminator"))
+
+    read_kwargs["comment"] = "#"
+    if parser.engine == "pyarrow":
+        if "lineterminator" in read_kwargs:
+            msg = (
+                "The 'lineterminator' option is not supported with the 'pyarrow' engine"
+            )
+        else:
+            msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **read_kwargs)
+        return
+    elif parser.engine == "python" and read_kwargs.get("lineterminator"):
+        msg = r"Custom line terminators not supported in python parser \(yet\)"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **read_kwargs)
+        return
+
+    result = parser.read_csv(StringIO(data), **read_kwargs)
+    expected = DataFrame(
+        [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comment_skiprows(all_parsers):
+    parser = all_parsers
+    data = """# empty
+random line
+# second empty line
+1,2,3
+A,B,C
+1,2.,4.
+5.,NaN,10.0
+"""
+    # This should ignore the first four lines (including comments).
+    expected = DataFrame(
+        [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"]
+    )
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", skiprows=4)
+        return
+
+    result = parser.read_csv(StringIO(data), comment="#", skiprows=4)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comment_header(all_parsers):
+    parser = all_parsers
+    data = """# empty
+# second empty line
+1,2,3
+A,B,C
+1,2.,4.
+5.,NaN,10.0
+"""
+    # Header should begin at the second non-comment line.
+    expected = DataFrame(
+        [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"]
+    )
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", header=1)
+        return
+    result = parser.read_csv(StringIO(data), comment="#", header=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comment_skiprows_header(all_parsers):
+    parser = all_parsers
+    data = """# empty
+# second empty line
+# third empty line
+X,Y,Z
+1,2,3
+A,B,C
+1,2.,4.
+5.,NaN,10.0
+"""
+    # Skiprows should skip the first 4 lines (including comments),
+    # while header should start from the second non-commented line,
+    # starting with line 5.
+    expected = DataFrame(
+        [[1.0, 2.0, 4.0], [5.0, np.nan, 10.0]], columns=["A", "B", "C"]
+    )
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", skiprows=4, header=1)
+        return
+
+    result = parser.read_csv(StringIO(data), comment="#", skiprows=4, header=1)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("comment_char", ["#", "~", "&", "^", "*", "@"])
+def test_custom_comment_char(all_parsers, comment_char):
+    parser = all_parsers
+    data = "a,b,c\n1,2,3#ignore this!\n4,5,6#ignorethistoo"
+
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data.replace("#", comment_char)), comment=comment_char
+            )
+        return
+    result = parser.read_csv(
+        StringIO(data.replace("#", comment_char)), comment=comment_char
+    )
+
+    expected = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["a", "b", "c"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("header", ["infer", None])
+def test_comment_first_line(all_parsers, header):
+    # see gh-4623
+    parser = all_parsers
+    data = "# notes\na,b,c\n# more notes\n1,2,3"
+
+    if header is None:
+        expected = DataFrame({0: ["a", "1"], 1: ["b", "2"], 2: ["c", "3"]})
+    else:
+        expected = DataFrame([[1, 2, 3]], columns=["a", "b", "c"])
+
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", header=header)
+        return
+    result = parser.read_csv(StringIO(data), comment="#", header=header)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_comment_char_in_default_value(all_parsers, request):
+    # GH#34002
+    if all_parsers.engine == "c":
+        reason = "see gh-34002: works on the python engine but not the c engine"
+        # NA value containing comment char is interpreted as comment
+        request.applymarker(pytest.mark.xfail(reason=reason, raises=AssertionError))
+    parser = all_parsers
+
+    data = (
+        "# this is a comment\n"
+        "col1,col2,col3,col4\n"
+        "1,2,3,4#inline comment\n"
+        "4,5#,6,10\n"
+        "7,8,#N/A,11\n"
+    )
+    if parser.engine == "pyarrow":
+        msg = "The 'comment' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), comment="#", na_values="#N/A")
+        return
+    result = parser.read_csv(StringIO(data), comment="#", na_values="#N/A")
+    expected = DataFrame(
+        {
+            "col1": [1, 4, 7],
+            "col2": [2, 5, 8],
+            "col3": [3.0, np.nan, np.nan],
+            "col4": [4.0, np.nan, 11.0],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_compression.py b/pandas/pandas/tests/io/parser/test_compression.py
new file mode 100644
index 0000000000000..191d0de50b12f
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_compression.py
@@ -0,0 +1,211 @@
+"""
+Tests compressed data parsing functionality for all
+of the parsers defined in parsers.py
+"""
+
+import os
+from pathlib import Path
+import tarfile
+import zipfile
+
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture(params=[True, False])
+def buffer(request):
+    return request.param
+
+
+@pytest.fixture
+def parser_and_data(all_parsers, csv1):
+    parser = all_parsers
+
+    with open(csv1, "rb") as f:
+        data = f.read()
+    expected = parser.read_csv(csv1)
+
+    return parser, data, expected
+
+
+@pytest.mark.parametrize("compression", ["zip", "infer", "zip2"])
+def test_zip(parser_and_data, compression):
+    parser, data, expected = parser_and_data
+
+    with tm.ensure_clean("test_file.zip") as path:
+        with zipfile.ZipFile(path, mode="w") as tmp:
+            tmp.writestr("test_file", data)
+
+        if compression == "zip2":
+            with open(path, "rb") as f:
+                result = parser.read_csv(f, compression="zip")
+        else:
+            result = parser.read_csv(path, compression=compression)
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("compression", ["zip", "infer"])
+def test_zip_error_multiple_files(parser_and_data, compression):
+    parser, data, expected = parser_and_data
+
+    with tm.ensure_clean("combined_zip.zip") as path:
+        inner_file_names = ["test_file", "second_file"]
+
+        with zipfile.ZipFile(path, mode="w") as tmp:
+            for file_name in inner_file_names:
+                tmp.writestr(file_name, data)
+
+        with pytest.raises(ValueError, match="Multiple files"):
+            parser.read_csv(path, compression=compression)
+
+
+def test_zip_error_no_files(parser_and_data):
+    parser, _, _ = parser_and_data
+
+    with tm.ensure_clean() as path:
+        with zipfile.ZipFile(path, mode="w"):
+            pass
+
+        with pytest.raises(ValueError, match="Zero files"):
+            parser.read_csv(path, compression="zip")
+
+
+def test_zip_error_invalid_zip(parser_and_data):
+    parser, _, _ = parser_and_data
+
+    with tm.ensure_clean() as path:
+        with open(path, "rb") as f:
+            with pytest.raises(zipfile.BadZipFile, match="File is not a zip file"):
+                parser.read_csv(f, compression="zip")
+
+
+@pytest.mark.parametrize("filename", [None, "test.{ext}"])
+def test_compression(
+    request,
+    parser_and_data,
+    compression_only,
+    buffer,
+    filename,
+    compression_to_extension,
+):
+    parser, data, expected = parser_and_data
+    compress_type = compression_only
+
+    ext = compression_to_extension[compress_type]
+    filename = filename if filename is None else filename.format(ext=ext)
+
+    if filename and buffer:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Cannot deduce compression from buffer of compressed data."
+            )
+        )
+
+    with tm.ensure_clean(filename=filename) as path:
+        tm.write_to_compressed(compress_type, path, data)
+        compression = "infer" if filename else compress_type
+
+        if buffer:
+            with open(path, "rb") as f:
+                result = parser.read_csv(f, compression=compression)
+        else:
+            result = parser.read_csv(path, compression=compression)
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("ext", [None, "gz", "bz2"])
+def test_infer_compression(all_parsers, csv1, buffer, ext):
+    # see gh-9770
+    parser = all_parsers
+    kwargs = {"index_col": 0, "parse_dates": True}
+
+    expected = parser.read_csv(csv1, **kwargs)
+    kwargs["compression"] = "infer"
+
+    if buffer:
+        with open(csv1, encoding="utf-8") as f:
+            result = parser.read_csv(f, **kwargs)
+    else:
+        ext = "." + ext if ext else ""
+        result = parser.read_csv(csv1 + ext, **kwargs)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compression_utf_encoding(all_parsers, csv_dir_path, utf_value, encoding_fmt):
+    # see gh-18071, gh-24130
+    parser = all_parsers
+    encoding = encoding_fmt.format(utf_value)
+    path = os.path.join(csv_dir_path, f"utf{utf_value}_ex_small.zip")
+
+    result = parser.read_csv(path, encoding=encoding, compression="zip", sep="\t")
+    expected = DataFrame(
+        {
+            "Country": ["Venezuela", "Venezuela"],
+            "Twitter": ["Hugo Chávez Frías", "Henrique Capriles R."],
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("invalid_compression", ["sfark", "bz3", "zipper"])
+def test_invalid_compression(all_parsers, invalid_compression):
+    parser = all_parsers
+    compress_kwargs = {"compression": invalid_compression}
+
+    msg = f"Unrecognized compression type: {invalid_compression}"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv("test_file.zip", **compress_kwargs)
+
+
+def test_compression_tar_archive(all_parsers, csv_dir_path):
+    parser = all_parsers
+    path = os.path.join(csv_dir_path, "tar_csv.tar.gz")
+    df = parser.read_csv(path)
+    assert list(df.columns) == ["a"]
+
+
+def test_ignore_compression_extension(all_parsers):
+    parser = all_parsers
+    df = DataFrame({"a": [0, 1]})
+    with tm.ensure_clean("test.csv") as path_csv:
+        with tm.ensure_clean("test.csv.zip") as path_zip:
+            # make sure to create un-compressed file with zip extension
+            df.to_csv(path_csv, index=False)
+            Path(path_zip).write_text(
+                Path(path_csv).read_text(encoding="utf-8"), encoding="utf-8"
+            )
+
+            tm.assert_frame_equal(parser.read_csv(path_zip, compression=None), df)
+
+
+def test_writes_tar_gz(all_parsers):
+    parser = all_parsers
+    data = DataFrame(
+        {
+            "Country": ["Venezuela", "Venezuela"],
+            "Twitter": ["Hugo Chávez Frías", "Henrique Capriles R."],
+        }
+    )
+    with tm.ensure_clean("test.tar.gz") as tar_path:
+        data.to_csv(tar_path, index=False)
+
+        # test that read_csv infers .tar.gz to gzip:
+        tm.assert_frame_equal(parser.read_csv(tar_path), data)
+
+        # test that file is indeed gzipped:
+        with tarfile.open(tar_path, "r:gz") as tar:
+            result = parser.read_csv(
+                tar.extractfile(tar.getnames()[0]), compression="infer"
+            )
+            tm.assert_frame_equal(result, data)
diff --git a/pandas/pandas/tests/io/parser/test_concatenate_chunks.py b/pandas/pandas/tests/io/parser/test_concatenate_chunks.py
new file mode 100644
index 0000000000000..3e418faeff8e0
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_concatenate_chunks.py
@@ -0,0 +1,38 @@
+import numpy as np
+import pytest
+
+from pandas.errors import DtypeWarning
+
+import pandas._testing as tm
+from pandas.core.arrays import ArrowExtensionArray
+
+from pandas.io.parsers.c_parser_wrapper import _concatenate_chunks
+
+
+def test_concatenate_chunks_pyarrow():
+    # GH#51876
+    pa = pytest.importorskip("pyarrow")
+    chunks = [
+        {0: ArrowExtensionArray(pa.array([1.5, 2.5]))},
+        {0: ArrowExtensionArray(pa.array([1, 2]))},
+    ]
+    result = _concatenate_chunks(chunks, ["column_0", "column_1"])
+    expected = ArrowExtensionArray(pa.array([1.5, 2.5, 1.0, 2.0]))
+    tm.assert_extension_array_equal(result[0], expected)
+
+
+def test_concatenate_chunks_pyarrow_strings():
+    # GH#51876
+    pa = pytest.importorskip("pyarrow")
+    chunks = [
+        {0: ArrowExtensionArray(pa.array([1.5, 2.5]))},
+        {0: ArrowExtensionArray(pa.array(["a", "b"]))},
+    ]
+    with tm.assert_produces_warning(
+        DtypeWarning, match="Columns \\(0: column_0\\) have mixed types"
+    ):
+        result = _concatenate_chunks(chunks, ["column_0", "column_1"])
+    expected = np.concatenate(
+        [np.array([1.5, 2.5], dtype=object), np.array(["a", "b"])]
+    )
+    tm.assert_numpy_array_equal(result[0], expected)
diff --git a/pandas/pandas/tests/io/parser/test_converters.py b/pandas/pandas/tests/io/parser/test_converters.py
new file mode 100644
index 0000000000000..c6ba2213033ea
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_converters.py
@@ -0,0 +1,265 @@
+"""
+Tests column conversion functionality during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+from dateutil.parser import parse
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+
+def test_converters_type_must_be_dict(all_parsers):
+    parser = all_parsers
+    data = """index,A,B,C,D
+foo,2,3,4,5
+"""
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), converters=0)
+        return
+    with pytest.raises(TypeError, match="Type converters.+"):
+        parser.read_csv(StringIO(data), converters=0)
+
+
+@pytest.mark.parametrize("column", [3, "D"])
+@pytest.mark.parametrize(
+    "converter",
+    [parse, lambda x: int(x.split("/")[2])],  # Produce integer.
+)
+def test_converters(all_parsers, column, converter):
+    parser = all_parsers
+    data = """A,B,C,D
+a,1,2,01/01/2009
+b,3,4,01/02/2009
+c,4,5,01/03/2009
+"""
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), converters={column: converter})
+        return
+
+    result = parser.read_csv(StringIO(data), converters={column: converter})
+
+    expected = parser.read_csv(StringIO(data))
+    expected["D"] = expected["D"].map(converter)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_converters_no_implicit_conv(all_parsers):
+    # see gh-2184
+    parser = all_parsers
+    data = """000102,1.2,A\n001245,2,B"""
+
+    converters = {0: lambda x: x.strip()}
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), header=None, converters=converters)
+        return
+
+    result = parser.read_csv(StringIO(data), header=None, converters=converters)
+
+    # Column 0 should not be casted to numeric and should remain as object.
+    expected = DataFrame([["000102", 1.2, "A"], ["001245", 2, "B"]])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_converters_euro_decimal_format(all_parsers):
+    # see gh-583
+    converters = {}
+    parser = all_parsers
+
+    data = """Id;Number1;Number2;Text1;Text2;Number3
+1;1521,1541;187101,9543;ABC;poi;4,7387
+2;121,12;14897,76;DEF;uyt;0,3773
+3;878,158;108013,434;GHI;rez;2,7356"""
+    converters["Number1"] = converters["Number2"] = converters["Number3"] = (
+        lambda x: float(x.replace(",", "."))
+    )
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), sep=";", converters=converters)
+        return
+
+    result = parser.read_csv(StringIO(data), sep=";", converters=converters)
+    expected = DataFrame(
+        [
+            [1, 1521.1541, 187101.9543, "ABC", "poi", 4.7387],
+            [2, 121.12, 14897.76, "DEF", "uyt", 0.3773],
+            [3, 878.158, 108013.434, "GHI", "rez", 2.7356],
+        ],
+        columns=["Id", "Number1", "Number2", "Text1", "Text2", "Number3"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_converters_corner_with_nans(all_parsers):
+    parser = all_parsers
+    data = """id,score,days
+1,2,12
+2,2-5,
+3,,14+
+4,6-12,2"""
+
+    # Example converters.
+    def convert_days(x):
+        x = x.strip()
+
+        if not x:
+            return np.nan
+
+        is_plus = x.endswith("+")
+
+        if is_plus:
+            x = int(x[:-1]) + 1
+        else:
+            x = int(x)
+
+        return x
+
+    def convert_days_sentinel(x):
+        x = x.strip()
+
+        if not x:
+            return np.nan
+
+        is_plus = x.endswith("+")
+
+        if is_plus:
+            x = int(x[:-1]) + 1
+        else:
+            x = int(x)
+
+        return x
+
+    def convert_score(x):
+        x = x.strip()
+
+        if not x:
+            return np.nan
+
+        if x.find("-") > 0:
+            val_min, val_max = map(int, x.split("-"))
+            val = 0.5 * (val_min + val_max)
+        else:
+            val = float(x)
+
+        return val
+
+    results = []
+
+    for day_converter in [convert_days, convert_days_sentinel]:
+        if parser.engine == "pyarrow":
+            msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(
+                    StringIO(data),
+                    converters={"score": convert_score, "days": day_converter},
+                    na_values=["", None],
+                )
+            continue
+
+        result = parser.read_csv(
+            StringIO(data),
+            converters={"score": convert_score, "days": day_converter},
+            na_values=["", None],
+        )
+        assert pd.isna(result["days"][1])
+        results.append(result)
+
+    if parser.engine != "pyarrow":
+        tm.assert_frame_equal(results[0], results[1])
+
+
+@pytest.mark.parametrize("conv_f", [lambda x: x, str])
+def test_converter_index_col_bug(all_parsers, conv_f):
+    # see gh-1835 , GH#40589
+    parser = all_parsers
+    data = "A;B\n1;2\n3;4"
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data), sep=";", index_col="A", converters={"A": conv_f}
+            )
+        return
+
+    rs = parser.read_csv(
+        StringIO(data), sep=";", index_col="A", converters={"A": conv_f}
+    )
+
+    xp = DataFrame({"B": [2, 4]}, index=Index(["1", "3"], name="A"))
+    tm.assert_frame_equal(rs, xp)
+
+
+def test_converter_identity_object(all_parsers):
+    # GH#40589
+    parser = all_parsers
+    data = "A,B\n1,2\n3,4"
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), converters={"A": lambda x: x})
+        return
+
+    rs = parser.read_csv(StringIO(data), converters={"A": lambda x: x})
+
+    xp = DataFrame({"A": ["1", "3"], "B": [2, 4]})
+    tm.assert_frame_equal(rs, xp)
+
+
+def test_converter_multi_index(all_parsers):
+    # GH 42446
+    parser = all_parsers
+    data = "A,B,B\nX,Y,Z\n1,2,3"
+
+    if parser.engine == "pyarrow":
+        msg = "The 'converters' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                header=list(range(2)),
+                converters={
+                    ("A", "X"): np.int32,
+                    ("B", "Y"): np.int32,
+                    ("B", "Z"): np.float32,
+                },
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data),
+        header=list(range(2)),
+        converters={
+            ("A", "X"): np.int32,
+            ("B", "Y"): np.int32,
+            ("B", "Z"): np.float32,
+        },
+    )
+
+    expected = DataFrame(
+        {
+            ("A", "X"): np.int32([1]),
+            ("B", "Y"): np.int32([2]),
+            ("B", "Z"): np.float32([3]),
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_dialect.py b/pandas/pandas/tests/io/parser/test_dialect.py
new file mode 100644
index 0000000000000..803114723bc74
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_dialect.py
@@ -0,0 +1,195 @@
+"""
+Tests that dialects are properly handled during parsing
+for all of the parsers defined in parsers.py
+"""
+
+import csv
+from io import StringIO
+
+import pytest
+
+from pandas.errors import ParserWarning
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture
+def custom_dialect():
+    dialect_name = "weird"
+    dialect_kwargs = {
+        "doublequote": False,
+        "escapechar": "~",
+        "delimiter": ":",
+        "skipinitialspace": False,
+        "quotechar": "`",
+        "quoting": 3,
+    }
+    return dialect_name, dialect_kwargs
+
+
+def test_dialect(all_parsers):
+    parser = all_parsers
+    data = """\
+label1,label2,label3
+index1,"a,c,e
+index2,b,d,f
+"""
+
+    dia = csv.excel()
+    dia.quoting = csv.QUOTE_NONE
+
+    if parser.engine == "pyarrow":
+        msg = "The 'dialect' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), dialect=dia)
+        return
+
+    df = parser.read_csv(StringIO(data), dialect=dia)
+
+    data = """\
+label1,label2,label3
+index1,a,c,e
+index2,b,d,f
+"""
+    exp = parser.read_csv(StringIO(data))
+    exp.replace("a", '"a', inplace=True)
+    tm.assert_frame_equal(df, exp)
+
+
+def test_dialect_str(all_parsers):
+    dialect_name = "mydialect"
+    parser = all_parsers
+    data = """\
+fruit:vegetable
+apple:broccoli
+pear:tomato
+"""
+    exp = DataFrame({"fruit": ["apple", "pear"], "vegetable": ["broccoli", "tomato"]})
+
+    with tm.with_csv_dialect(dialect_name, delimiter=":"):
+        if parser.engine == "pyarrow":
+            msg = "The 'dialect' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(StringIO(data), dialect=dialect_name)
+            return
+
+        df = parser.read_csv(StringIO(data), dialect=dialect_name)
+        tm.assert_frame_equal(df, exp)
+
+
+def test_invalid_dialect(all_parsers):
+    class InvalidDialect:
+        pass
+
+    data = "a\n1"
+    parser = all_parsers
+    msg = "Invalid dialect"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), dialect=InvalidDialect)
+
+
+@pytest.mark.parametrize(
+    "arg",
+    [None, "doublequote", "escapechar", "skipinitialspace", "quotechar", "quoting"],
+)
+@pytest.mark.parametrize("value", ["dialect", "default", "other"])
+def test_dialect_conflict_except_delimiter(all_parsers, custom_dialect, arg, value):
+    # see gh-23761.
+    dialect_name, dialect_kwargs = custom_dialect
+    parser = all_parsers
+
+    expected = DataFrame({"a": [1], "b": [2]})
+    data = "a:b\n1:2"
+
+    warning_klass = None
+    kwds = {}
+
+    # arg=None tests when we pass in the dialect without any other arguments.
+    if arg is not None:
+        if value == "dialect":  # No conflict --> no warning.
+            kwds[arg] = dialect_kwargs[arg]
+        elif value == "default":  # Default --> no warning.
+            from pandas.io.parsers.base_parser import parser_defaults
+
+            kwds[arg] = parser_defaults[arg]
+        else:  # Non-default + conflict with dialect --> warning.
+            warning_klass = ParserWarning
+            kwds[arg] = "blah"
+
+    with tm.with_csv_dialect(dialect_name, **dialect_kwargs):
+        if parser.engine == "pyarrow":
+            msg = "The 'dialect' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv_check_warnings(
+                    # No warning bc we raise
+                    None,
+                    "Conflicting values for",
+                    StringIO(data),
+                    dialect=dialect_name,
+                    **kwds,
+                )
+            return
+        result = parser.read_csv_check_warnings(
+            warning_klass,
+            "Conflicting values for",
+            StringIO(data),
+            dialect=dialect_name,
+            **kwds,
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,warning_klass",
+    [
+        ({"sep": ","}, None),  # sep is default --> sep_override=True
+        ({"sep": "."}, ParserWarning),  # sep isn't default --> sep_override=False
+        ({"delimiter": ":"}, None),  # No conflict
+        ({"delimiter": None}, None),  # Default arguments --> sep_override=True
+        ({"delimiter": ","}, ParserWarning),  # Conflict
+        ({"delimiter": "."}, ParserWarning),  # Conflict
+    ],
+    ids=[
+        "sep-override-true",
+        "sep-override-false",
+        "delimiter-no-conflict",
+        "delimiter-default-arg",
+        "delimiter-conflict",
+        "delimiter-conflict2",
+    ],
+)
+def test_dialect_conflict_delimiter(all_parsers, custom_dialect, kwargs, warning_klass):
+    # see gh-23761.
+    dialect_name, dialect_kwargs = custom_dialect
+    parser = all_parsers
+
+    expected = DataFrame({"a": [1], "b": [2]})
+    data = "a:b\n1:2"
+
+    with tm.with_csv_dialect(dialect_name, **dialect_kwargs):
+        if parser.engine == "pyarrow":
+            msg = "The 'dialect' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv_check_warnings(
+                    # no warning bc we raise
+                    None,
+                    "Conflicting values for 'delimiter'",
+                    StringIO(data),
+                    dialect=dialect_name,
+                    **kwargs,
+                )
+            return
+        result = parser.read_csv_check_warnings(
+            warning_klass,
+            "Conflicting values for 'delimiter'",
+            StringIO(data),
+            dialect=dialect_name,
+            **kwargs,
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_encoding.py b/pandas/pandas/tests/io/parser/test_encoding.py
new file mode 100644
index 0000000000000..5df8c3d27bf84
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_encoding.py
@@ -0,0 +1,339 @@
+"""
+Tests encoding functionality during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import (
+    BytesIO,
+    TextIOWrapper,
+)
+import os
+import tempfile
+import uuid
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    read_csv,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_bytes_io_input(all_parsers):
+    encoding = "cp1255"
+    parser = all_parsers
+
+    data = BytesIO("שלום:1234\n562:123".encode(encoding))
+    result = parser.read_csv(data, sep=":", encoding=encoding)
+
+    expected = DataFrame([[562, 123]], columns=["שלום", "1234"])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_read_csv_unicode(all_parsers):
+    parser = all_parsers
+    data = BytesIO("\u0141aski, Jan;1".encode())
+
+    result = parser.read_csv(data, sep=";", encoding="utf-8", header=None)
+    expected = DataFrame([["\u0141aski, Jan", 1]])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow
+@pytest.mark.parametrize("sep", [",", "\t"])
+@pytest.mark.parametrize("encoding", ["utf-16", "utf-16le", "utf-16be"])
+def test_utf16_bom_skiprows(all_parsers, sep, encoding):
+    # see gh-2298
+    parser = all_parsers
+    data = """skip this
+skip this too
+A,B,C
+1,2,3
+4,5,6""".replace(",", sep)
+    path = f"__{uuid.uuid4()}__.csv"
+    kwargs = {"sep": sep, "skiprows": 2}
+    utf8 = "utf-8"
+
+    with tm.ensure_clean(path) as path:
+        bytes_data = data.encode(encoding)
+
+        with open(path, "wb") as f:
+            f.write(bytes_data)
+
+        with TextIOWrapper(BytesIO(data.encode(utf8)), encoding=utf8) as bytes_buffer:
+            result = parser.read_csv(path, encoding=encoding, **kwargs)
+            expected = parser.read_csv(bytes_buffer, encoding=utf8, **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_utf16_example(all_parsers, csv_dir_path):
+    path = os.path.join(csv_dir_path, "utf16_ex.txt")
+    parser = all_parsers
+    result = parser.read_csv(path, encoding="utf-16", sep="\t")
+    assert len(result) == 50
+
+
+def test_unicode_encoding(all_parsers, csv_dir_path):
+    path = os.path.join(csv_dir_path, "unicode_series.csv")
+    parser = all_parsers
+
+    result = parser.read_csv(path, header=None, encoding="latin-1")
+    result = result.set_index(0)
+    got = result[1][1632]
+
+    expected = "\xc1 k\xf6ldum klaka (Cold Fever) (1994)"
+    assert got == expected
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        # Basic test
+        ("a\n1", {}, [1]),
+        # "Regular" quoting
+        ('"a"\n1', {"quotechar": '"'}, [1]),
+        # Test in a data row instead of header
+        ("b\n1", {"names": ["a"]}, ["b", "1"]),
+        # Test in empty data row with skipping
+        ("\n1", {"names": ["a"], "skip_blank_lines": True}, [1]),
+        # Test in empty data row without skipping
+        (
+            "\n1",
+            {"names": ["a"], "skip_blank_lines": False},
+            [np.nan, 1],
+        ),
+    ],
+)
+def test_utf8_bom(all_parsers, data, kwargs, expected):
+    # see gh-4793
+    parser = all_parsers
+    bom = "\ufeff"
+    utf8 = "utf-8"
+
+    def _encode_data_with_bom(_data):
+        bom_data = (bom + _data).encode(utf8)
+        return BytesIO(bom_data)
+
+    if (
+        parser.engine == "pyarrow"
+        and data == "\n1"
+        and kwargs.get("skip_blank_lines", True)
+    ):
+        # CSV parse error: Empty CSV file or block: cannot infer number of columns
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    result = parser.read_csv(_encode_data_with_bom(data), encoding=utf8, **kwargs)
+    expected = DataFrame({"a": expected})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_utf_aliases(all_parsers, utf_value, encoding_fmt):
+    # see gh-13549
+    expected = DataFrame({"mb_num": [4.8], "multibyte": ["test"]})
+    parser = all_parsers
+
+    encoding = encoding_fmt.format(utf_value)
+    data = "mb_num,multibyte\n4.8,test".encode(encoding)
+
+    result = parser.read_csv(BytesIO(data), encoding=encoding)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "file_path,encoding",
+    [
+        (("io", "data", "csv", "test1.csv"), "utf-8"),
+        (("io", "parser", "data", "unicode_series.csv"), "latin-1"),
+        (("io", "parser", "data", "sauron.SHIFT_JIS.csv"), "shiftjis"),
+    ],
+)
+def test_binary_mode_file_buffers(all_parsers, file_path, encoding, datapath):
+    # gh-23779: Python csv engine shouldn't error on files opened in binary.
+    # gh-31575: Python csv engine shouldn't error on files opened in raw binary.
+    parser = all_parsers
+
+    fpath = datapath(*file_path)
+    expected = parser.read_csv(fpath, encoding=encoding)
+
+    with open(fpath, encoding=encoding) as fa:
+        result = parser.read_csv(fa)
+        assert not fa.closed
+    tm.assert_frame_equal(expected, result)
+
+    with open(fpath, mode="rb") as fb:
+        result = parser.read_csv(fb, encoding=encoding)
+        assert not fb.closed
+    tm.assert_frame_equal(expected, result)
+
+    with open(fpath, mode="rb", buffering=0) as fb:
+        result = parser.read_csv(fb, encoding=encoding)
+        assert not fb.closed
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize("pass_encoding", [True, False])
+def test_encoding_temp_file(
+    all_parsers, utf_value, encoding_fmt, pass_encoding, temp_file
+):
+    # see gh-24130
+    parser = all_parsers
+    encoding = encoding_fmt.format(utf_value)
+
+    if parser.engine == "pyarrow" and pass_encoding is True and utf_value in [16, 32]:
+        # FIXME: this is bad!
+        pytest.skip("These cases freeze")
+
+    expected = DataFrame({"foo": ["bar"]})
+
+    with temp_file.open(mode="w+", encoding=encoding) as f:
+        f.write("foo\nbar")
+        f.seek(0)
+
+        result = parser.read_csv(f, encoding=encoding if pass_encoding else None)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_encoding_named_temp_file(all_parsers):
+    # see gh-31819
+    parser = all_parsers
+    encoding = "shift-jis"
+
+    title = "てすと"
+    data = "こむ"
+
+    expected = DataFrame({title: [data]})
+
+    with tempfile.NamedTemporaryFile() as f:
+        f.write(f"{title}\n{data}".encode(encoding))
+
+        f.seek(0)
+
+        result = parser.read_csv(f, encoding=encoding)
+        tm.assert_frame_equal(result, expected)
+        assert not f.closed
+
+
+@pytest.mark.parametrize(
+    "encoding", ["utf-8", "utf-16", "utf-16-be", "utf-16-le", "utf-32"]
+)
+def test_parse_encoded_special_characters(encoding):
+    # GH16218 Verify parsing of data with encoded special characters
+    # Data contains a Unicode 'FULLWIDTH COLON' (U+FF1A) at position (0,"a")
+    data = "a\tb\n：foo\t0\nbar\t1\nbaz\t2"  # noqa: RUF001
+    encoded_data = BytesIO(data.encode(encoding))
+    result = read_csv(encoded_data, delimiter="\t", encoding=encoding)
+
+    expected = DataFrame(
+        data=[["：foo", 0], ["bar", 1], ["baz", 2]],  # noqa: RUF001
+        columns=["a", "b"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("encoding", ["utf-8", None, "utf-16", "cp1255", "latin-1"])
+def test_encoding_memory_map(all_parsers, encoding):
+    # GH40986
+    parser = all_parsers
+    expected = DataFrame(
+        {
+            "name": ["Raphael", "Donatello", "Miguel Angel", "Leonardo"],
+            "mask": ["red", "purple", "orange", "blue"],
+            "weapon": ["sai", "bo staff", "nunchunk", "katana"],
+        }
+    )
+    with tm.ensure_clean() as file:
+        expected.to_csv(file, index=False, encoding=encoding)
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(file, encoding=encoding, memory_map=True)
+            return
+
+        df = parser.read_csv(file, encoding=encoding, memory_map=True)
+    tm.assert_frame_equal(df, expected)
+
+
+def test_chunk_splits_multibyte_char(all_parsers):
+    """
+    Chunk splits a multibyte character with memory_map=True
+
+    GH 43540
+    """
+    parser = all_parsers
+    # DEFAULT_CHUNKSIZE = 262144, defined in parsers.pyx
+    df = DataFrame(data=["a" * 127] * 2048)
+
+    # Put two-bytes utf-8 encoded character "ą" at the end of chunk
+    # utf-8 encoding of "ą" is b'\xc4\x85'
+    df.iloc[2047] = "a" * 127 + "ą"
+    with tm.ensure_clean("bug-gh43540.csv") as fname:
+        df.to_csv(fname, index=False, header=False, encoding="utf-8")
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(fname, header=None, memory_map=True)
+            return
+
+        dfr = parser.read_csv(fname, header=None, memory_map=True)
+    tm.assert_frame_equal(dfr, df)
+
+
+def test_readcsv_memmap_utf8(all_parsers):
+    """
+    GH 43787
+
+    Test correct handling of UTF-8 chars when memory_map=True and encoding is UTF-8
+    """
+    lines = []
+    line_length = 128
+    start_char = " "
+    end_char = "\U00010080"
+    # This for loop creates a list of 128-char strings
+    # consisting of consecutive Unicode chars
+    for lnum in range(ord(start_char), ord(end_char), line_length):
+        line = "".join([chr(c) for c in range(lnum, lnum + 0x80)]) + "\n"
+        try:
+            line.encode("utf-8")
+        except UnicodeEncodeError:
+            continue
+        lines.append(line)
+    parser = all_parsers
+    df = DataFrame(lines)
+    with tm.ensure_clean("utf8test.csv") as fname:
+        df.to_csv(fname, index=False, header=False, encoding="utf-8")
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(fname, header=None, memory_map=True, encoding="utf-8")
+            return
+
+        dfr = parser.read_csv(fname, header=None, memory_map=True, encoding="utf-8")
+    tm.assert_frame_equal(df, dfr)
+
+
+@pytest.mark.usefixtures("pyarrow_xfail")
+@pytest.mark.parametrize("mode", ["w+b", "w+t"])
+def test_not_readable(all_parsers, mode):
+    # GH43439
+    parser = all_parsers
+    content = b"abcd"
+    if "t" in mode:
+        content = "abcd"
+    with tempfile.SpooledTemporaryFile(mode=mode, encoding="utf-8") as handle:
+        handle.write(content)
+        handle.seek(0)
+        df = parser.read_csv(handle)
+    expected = DataFrame([], columns=["abcd"])
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/io/parser/test_header.py b/pandas/pandas/tests/io/parser/test_header.py
new file mode 100644
index 0000000000000..d333aef723de2
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_header.py
@@ -0,0 +1,727 @@
+"""
+Tests that the file header is properly handled or inferred
+during parsing for all of the parsers defined in parsers.py
+"""
+
+from collections import namedtuple
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas.errors import ParserError
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_read_with_bad_header(all_parsers):
+    parser = all_parsers
+    msg = r"but only \d+ lines in file"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(",,"), header=[10])
+
+
+def test_negative_header(all_parsers):
+    # see gh-27779
+    parser = all_parsers
+    data = """1,2,3,4,5
+6,7,8,9,10
+11,12,13,14,15
+"""
+    with pytest.raises(
+        ValueError,
+        match="Passing negative integer to header is invalid. "
+        "For no header, use header=None instead",
+    ):
+        parser.read_csv(StringIO(data), header=-1)
+
+
+@pytest.mark.parametrize("header", [([-1, 2, 4]), ([-5, 0])])
+def test_negative_multi_index_header(all_parsers, header):
+    # see gh-27779
+    parser = all_parsers
+    data = """1,2,3,4,5
+        6,7,8,9,10
+        11,12,13,14,15
+        """
+    with pytest.raises(
+        ValueError, match="cannot specify multi-index header with negative integers"
+    ):
+        parser.read_csv(StringIO(data), header=header)
+
+
+@pytest.mark.parametrize("header", [True, False])
+def test_bool_header_arg(all_parsers, header):
+    # see gh-6114
+    parser = all_parsers
+    data = """\
+MyColumn
+a
+b
+a
+b"""
+    msg = "Passing a bool to header is invalid"
+    with pytest.raises(TypeError, match=msg):
+        parser.read_csv(StringIO(data), header=header)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame are different
+def test_header_with_index_col(all_parsers):
+    parser = all_parsers
+    data = """foo,1,2,3
+bar,4,5,6
+baz,7,8,9
+"""
+    names = ["A", "B", "C"]
+    result = parser.read_csv(StringIO(data), names=names)
+
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        index=["foo", "bar", "baz"],
+        columns=["A", "B", "C"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_header_not_first_line(all_parsers):
+    parser = all_parsers
+    data = """got,to,ignore,this,line
+got,to,ignore,this,line
+index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+"""
+    data2 = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+"""
+
+    result = parser.read_csv(StringIO(data), header=2, index_col=0)
+    expected = parser.read_csv(StringIO(data2), header=0, index_col=0)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_multi_index(all_parsers):
+    parser = all_parsers
+
+    data = """\
+C0,,C_l0_g0,C_l0_g1,C_l0_g2
+
+C1,,C_l1_g0,C_l1_g1,C_l1_g2
+C2,,C_l2_g0,C_l2_g1,C_l2_g2
+C3,,C_l3_g0,C_l3_g1,C_l3_g2
+R0,R1,,,
+R_l0_g0,R_l1_g0,R0C0,R0C1,R0C2
+R_l0_g1,R_l1_g1,R1C0,R1C1,R1C2
+R_l0_g2,R_l1_g2,R2C0,R2C1,R2C2
+R_l0_g3,R_l1_g3,R3C0,R3C1,R3C2
+R_l0_g4,R_l1_g4,R4C0,R4C1,R4C2
+"""
+    result = parser.read_csv(StringIO(data), header=[0, 1, 2, 3], index_col=[0, 1])
+    data_gen_f = lambda r, c: f"R{r}C{c}"
+
+    data = [[data_gen_f(r, c) for c in range(3)] for r in range(5)]
+    index = MultiIndex.from_arrays(
+        [[f"R_l0_g{i}" for i in range(5)], [f"R_l1_g{i}" for i in range(5)]],
+        names=["R0", "R1"],
+    )
+    columns = MultiIndex.from_arrays(
+        [
+            [f"C_l0_g{i}" for i in range(3)],
+            [f"C_l1_g{i}" for i in range(3)],
+            [f"C_l2_g{i}" for i in range(3)],
+            [f"C_l3_g{i}" for i in range(3)],
+        ],
+        names=["C0", "C1", "C2", "C3"],
+    )
+    expected = DataFrame(data, columns=columns, index=index)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,msg",
+    [
+        (
+            {"index_col": ["foo", "bar"]},
+            (
+                "index_col must only contain "
+                "integers of column positions when specifying "
+                "a multi-index header"
+            ),
+        ),
+        (
+            {"index_col": [0, 1], "names": ["foo", "bar"]},
+            ("cannot specify names when specifying a multi-index header"),
+        ),
+        (
+            {"index_col": [0, 1], "usecols": ["foo", "bar"]},
+            ("cannot specify usecols when specifying a multi-index header"),
+        ),
+    ],
+)
+def test_header_multi_index_invalid(all_parsers, kwargs, msg):
+    data = """\
+C0,,C_l0_g0,C_l0_g1,C_l0_g2
+
+C1,,C_l1_g0,C_l1_g1,C_l1_g2
+C2,,C_l2_g0,C_l2_g1,C_l2_g2
+C3,,C_l3_g0,C_l3_g1,C_l3_g2
+R0,R1,,,
+R_l0_g0,R_l1_g0,R0C0,R0C1,R0C2
+R_l0_g1,R_l1_g1,R1C0,R1C1,R1C2
+R_l0_g2,R_l1_g2,R2C0,R2C1,R2C2
+R_l0_g3,R_l1_g3,R3C0,R3C1,R3C2
+R_l0_g4,R_l1_g4,R4C0,R4C1,R4C2
+"""
+    parser = all_parsers
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), header=[0, 1, 2, 3], **kwargs)
+
+
+_TestTuple = namedtuple("_TestTuple", ["first", "second"])
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"header": [0, 1]},
+        {
+            "skiprows": 3,
+            "names": [
+                ("a", "q"),
+                ("a", "r"),
+                ("a", "s"),
+                ("b", "t"),
+                ("c", "u"),
+                ("c", "v"),
+            ],
+        },
+        {
+            "skiprows": 3,
+            "names": [
+                _TestTuple("a", "q"),
+                _TestTuple("a", "r"),
+                _TestTuple("a", "s"),
+                _TestTuple("b", "t"),
+                _TestTuple("c", "u"),
+                _TestTuple("c", "v"),
+            ],
+        },
+    ],
+)
+def test_header_multi_index_common_format1(all_parsers, kwargs):
+    parser = all_parsers
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5, 6], [7, 8, 9, 10, 11, 12]],
+        index=["one", "two"],
+        columns=MultiIndex.from_tuples(
+            [("a", "q"), ("a", "r"), ("a", "s"), ("b", "t"), ("c", "u"), ("c", "v")]
+        ),
+    )
+    data = """,a,a,a,b,c,c
+,q,r,s,t,u,v
+,,,,,,
+one,1,2,3,4,5,6
+two,7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), index_col=0, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"header": [0, 1]},
+        {
+            "skiprows": 2,
+            "names": [
+                ("a", "q"),
+                ("a", "r"),
+                ("a", "s"),
+                ("b", "t"),
+                ("c", "u"),
+                ("c", "v"),
+            ],
+        },
+        {
+            "skiprows": 2,
+            "names": [
+                _TestTuple("a", "q"),
+                _TestTuple("a", "r"),
+                _TestTuple("a", "s"),
+                _TestTuple("b", "t"),
+                _TestTuple("c", "u"),
+                _TestTuple("c", "v"),
+            ],
+        },
+    ],
+)
+def test_header_multi_index_common_format2(all_parsers, kwargs):
+    parser = all_parsers
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5, 6], [7, 8, 9, 10, 11, 12]],
+        index=["one", "two"],
+        columns=MultiIndex.from_tuples(
+            [("a", "q"), ("a", "r"), ("a", "s"), ("b", "t"), ("c", "u"), ("c", "v")]
+        ),
+    )
+    data = """,a,a,a,b,c,c
+,q,r,s,t,u,v
+one,1,2,3,4,5,6
+two,7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), index_col=0, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"header": [0, 1]},
+        {
+            "skiprows": 2,
+            "names": [
+                ("a", "q"),
+                ("a", "r"),
+                ("a", "s"),
+                ("b", "t"),
+                ("c", "u"),
+                ("c", "v"),
+            ],
+        },
+        {
+            "skiprows": 2,
+            "names": [
+                _TestTuple("a", "q"),
+                _TestTuple("a", "r"),
+                _TestTuple("a", "s"),
+                _TestTuple("b", "t"),
+                _TestTuple("c", "u"),
+                _TestTuple("c", "v"),
+            ],
+        },
+    ],
+)
+def test_header_multi_index_common_format3(all_parsers, kwargs):
+    parser = all_parsers
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5, 6], [7, 8, 9, 10, 11, 12]],
+        index=["one", "two"],
+        columns=MultiIndex.from_tuples(
+            [("a", "q"), ("a", "r"), ("a", "s"), ("b", "t"), ("c", "u"), ("c", "v")]
+        ),
+    )
+    expected = expected.reset_index(drop=True)
+    data = """a,a,a,b,c,c
+q,r,s,t,u,v
+1,2,3,4,5,6
+7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), index_col=None, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_multi_index_common_format_malformed1(all_parsers):
+    parser = all_parsers
+    expected = DataFrame(
+        np.array([[2, 3, 4, 5, 6], [8, 9, 10, 11, 12]], dtype="int64"),
+        index=Index([1, 7]),
+        columns=MultiIndex(
+            levels=[["a", "b", "c"], ["r", "s", "t", "u", "v"]],
+            codes=[[0, 0, 1, 2, 2], [0, 1, 2, 3, 4]],
+            names=["a", "q"],
+        ),
+    )
+    data = """a,a,a,b,c,c
+q,r,s,t,u,v
+1,2,3,4,5,6
+7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), header=[0, 1], index_col=0)
+    tm.assert_frame_equal(expected, result)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_multi_index_common_format_malformed2(all_parsers):
+    parser = all_parsers
+    expected = DataFrame(
+        np.array([[2, 3, 4, 5, 6], [8, 9, 10, 11, 12]], dtype="int64"),
+        index=range(1, 13, 6),
+        columns=MultiIndex(
+            levels=[["a", "b", "c"], ["r", "s", "t", "u", "v"]],
+            codes=[[0, 0, 1, 2, 2], [0, 1, 2, 3, 4]],
+            names=[None, "q"],
+        ),
+    )
+
+    data = """,a,a,b,c,c
+q,r,s,t,u,v
+1,2,3,4,5,6
+7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), header=[0, 1], index_col=0)
+    tm.assert_frame_equal(expected, result)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_multi_index_common_format_malformed3(all_parsers):
+    parser = all_parsers
+    expected = DataFrame(
+        np.array([[3, 4, 5, 6], [9, 10, 11, 12]], dtype="int64"),
+        index=MultiIndex(levels=[[1, 7], [2, 8]], codes=[[0, 1], [0, 1]]),
+        columns=MultiIndex(
+            levels=[["a", "b", "c"], ["s", "t", "u", "v"]],
+            codes=[[0, 1, 2, 2], [0, 1, 2, 3]],
+            names=[None, "q"],
+        ),
+    )
+    data = """,a,a,b,c,c
+q,r,s,t,u,v
+1,2,3,4,5,6
+7,8,9,10,11,12"""
+
+    result = parser.read_csv(StringIO(data), header=[0, 1], index_col=[0, 1])
+    tm.assert_frame_equal(expected, result)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_multi_index_blank_line(all_parsers):
+    # GH 40442
+    parser = all_parsers
+    data = [[None, None], [1, 2], [3, 4]]
+    columns = MultiIndex.from_tuples([("a", "A"), ("b", "B")])
+    expected = DataFrame(data, columns=columns)
+    data = "a,b\nA,B\n,\n1,2\n3,4"
+    result = parser.read_csv(StringIO(data), header=[0, 1])
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize(
+    "data,header", [("1,2,3\n4,5,6", None), ("foo,bar,baz\n1,2,3\n4,5,6", 0)]
+)
+def test_header_names_backward_compat(all_parsers, data, header, request):
+    # see gh-2539
+    parser = all_parsers
+
+    if parser.engine == "pyarrow" and header is not None:
+        mark = pytest.mark.xfail(reason="DataFrame.columns are different")
+        request.applymarker(mark)
+
+    expected = parser.read_csv(StringIO("1,2,3\n4,5,6"), names=["a", "b", "c"])
+
+    result = parser.read_csv(StringIO(data), names=["a", "b", "c"], header=header)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block: cannot infer
+@pytest.mark.parametrize("kwargs", [{}, {"index_col": False}])
+def test_read_only_header_no_rows(all_parsers, kwargs):
+    # See gh-7773
+    parser = all_parsers
+    expected = DataFrame(columns=["a", "b", "c"])
+
+    result = parser.read_csv(StringIO("a,b,c"), **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,names",
+    [
+        ({}, [0, 1, 2, 3, 4]),
+        (
+            {"names": ["foo", "bar", "baz", "quux", "panda"]},
+            ["foo", "bar", "baz", "quux", "panda"],
+        ),
+    ],
+)
+def test_no_header(all_parsers, kwargs, names):
+    parser = all_parsers
+    data = """1,2,3,4,5
+6,7,8,9,10
+11,12,13,14,15
+"""
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]], columns=names
+    )
+    result = parser.read_csv(StringIO(data), header=None, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("header", [["a", "b"], "string_header"])
+def test_non_int_header(all_parsers, header):
+    # see gh-16338
+    msg = "header must be integer or list of integers"
+    data = """1,2\n3,4"""
+    parser = all_parsers
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), header=header)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_singleton_header(all_parsers):
+    # see gh-7757
+    data = """a,b,c\n0,1,2\n1,2,3"""
+    parser = all_parsers
+
+    expected = DataFrame({"a": [0, 1], "b": [1, 2], "c": [2, 3]})
+    result = parser.read_csv(StringIO(data), header=[0])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        (
+            "A,A,A,B\none,one,one,two\n0,40,34,0.1",
+            DataFrame(
+                [[0, 40, 34, 0.1]],
+                columns=MultiIndex.from_tuples(
+                    [("A", "one"), ("A", "one.1"), ("A", "one.2"), ("B", "two")]
+                ),
+            ),
+        ),
+        (
+            "A,A,A,B\none,one,one.1,two\n0,40,34,0.1",
+            DataFrame(
+                [[0, 40, 34, 0.1]],
+                columns=MultiIndex.from_tuples(
+                    [("A", "one"), ("A", "one.1"), ("A", "one.1.1"), ("B", "two")]
+                ),
+            ),
+        ),
+        (
+            "A,A,A,B,B\none,one,one.1,two,two\n0,40,34,0.1,0.1",
+            DataFrame(
+                [[0, 40, 34, 0.1, 0.1]],
+                columns=MultiIndex.from_tuples(
+                    [
+                        ("A", "one"),
+                        ("A", "one.1"),
+                        ("A", "one.1.1"),
+                        ("B", "two"),
+                        ("B", "two.1"),
+                    ]
+                ),
+            ),
+        ),
+    ],
+)
+def test_mangles_multi_index(all_parsers, data, expected):
+    # see gh-18062
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data), header=[0, 1])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+@pytest.mark.parametrize("index_col", [None, [0]])
+@pytest.mark.parametrize(
+    "columns", [None, (["", "Unnamed"]), (["Unnamed", ""]), (["Unnamed", "NotUnnamed"])]
+)
+def test_multi_index_unnamed(all_parsers, index_col, columns):
+    # see gh-23687
+    #
+    # When specifying a multi-index header, make sure that
+    # we don't error just because one of the rows in our header
+    # has ALL column names containing the string "Unnamed". The
+    # correct condition to check is whether the row contains
+    # ALL columns that did not have names (and instead were given
+    # placeholder ones).
+    parser = all_parsers
+    header = [0, 1]
+
+    if index_col is None:
+        data = ",".join(columns or ["", ""]) + "\n0,1\n2,3\n4,5\n"
+    else:
+        data = ",".join([""] + (columns or ["", ""])) + "\n,0,1\n0,2,3\n1,4,5\n"
+
+    result = parser.read_csv(StringIO(data), header=header, index_col=index_col)
+    exp_columns = []
+
+    if columns is None:
+        columns = ["", "", ""]
+
+    for i, col in enumerate(columns):
+        if not col:  # Unnamed.
+            col = f"Unnamed: {i if index_col is None else i + 1}_level_0"
+
+        exp_columns.append(col)
+
+    columns = MultiIndex.from_tuples(zip(exp_columns, ["0", "1"]))
+    expected = DataFrame([[2, 3], [4, 5]], columns=columns)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Expected 2 columns, got 3
+def test_names_longer_than_header_but_equal_with_data_rows(all_parsers):
+    # GH#38453
+    parser = all_parsers
+    data = """a, b
+1,2,3
+5,6,4
+"""
+    result = parser.read_csv(StringIO(data), header=0, names=["A", "B", "C"])
+    expected = DataFrame({"A": [1, 5], "B": [2, 6], "C": [3, 4]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_read_csv_multiindex_columns(all_parsers):
+    # GH#6051
+    parser = all_parsers
+
+    s1 = "Male, Male, Male, Female, Female\nR, R, L, R, R\n.86, .67, .88, .78, .81"
+    s2 = (
+        "Male, Male, Male, Female, Female\n"
+        "R, R, L, R, R\n"
+        ".86, .67, .88, .78, .81\n"
+        ".86, .67, .88, .78, .82"
+    )
+
+    mi = MultiIndex.from_tuples(
+        [
+            ("Male", "R"),
+            (" Male", " R"),
+            (" Male", " L"),
+            (" Female", " R"),
+            (" Female", " R.1"),
+        ]
+    )
+    expected = DataFrame(
+        [[0.86, 0.67, 0.88, 0.78, 0.81], [0.86, 0.67, 0.88, 0.78, 0.82]], columns=mi
+    )
+
+    df1 = parser.read_csv(StringIO(s1), header=[0, 1])
+    tm.assert_frame_equal(df1, expected.iloc[:1])
+    df2 = parser.read_csv(StringIO(s2), header=[0, 1])
+    tm.assert_frame_equal(df2, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_read_csv_multi_header_length_check(all_parsers):
+    # GH#43102
+    parser = all_parsers
+
+    case = """row11,row12,row13
+row21,row22, row23
+row31,row32
+"""
+
+    with pytest.raises(
+        ParserError, match="Header rows must have an equal number of columns."
+    ):
+        parser.read_csv(StringIO(case), header=[0, 2])
+
+
+@skip_pyarrow  # CSV parse error: Expected 3 columns, got 2
+def test_header_none_and_implicit_index(all_parsers):
+    # GH#22144
+    parser = all_parsers
+    data = "x,1,5\ny,2\nz,3\n"
+    result = parser.read_csv(StringIO(data), names=["a", "b"], header=None)
+    expected = DataFrame(
+        {"a": [1, 2, 3], "b": [5, np.nan, np.nan]}, index=["x", "y", "z"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # regex mismatch "CSV parse error: Expected 2 columns, got "
+def test_header_none_and_implicit_index_in_second_row(all_parsers):
+    # GH#22144
+    parser = all_parsers
+    data = "x,1\ny,2,5\nz,3\n"
+    with pytest.raises(ParserError, match="Expected 2 fields in line 2, saw 3"):
+        parser.read_csv(StringIO(data), names=["a", "b"], header=None)
+
+
+def test_header_none_and_on_bad_lines_skip(all_parsers):
+    # GH#22144
+    parser = all_parsers
+    data = "x,1\ny,2,5\nz,3\n"
+    result = parser.read_csv(
+        StringIO(data), names=["a", "b"], header=None, on_bad_lines="skip"
+    )
+    expected = DataFrame({"a": ["x", "z"], "b": [1, 3]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_missing_rows(all_parsers):
+    # GH#47400
+    parser = all_parsers
+    data = """a,b
+1,2
+"""
+    msg = r"Passed header=\[0,1,2\], len of 3, but only 2 lines in file"
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), header=[0, 1, 2])
+
+
+# ValueError: the 'pyarrow' engine does not support regex separators
+@xfail_pyarrow
+def test_header_multiple_whitespaces(all_parsers):
+    # GH#54931
+    parser = all_parsers
+    data = """aa    bb(1,1)   cc(1,1)
+                0  2  3.5"""
+
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+    expected = DataFrame({"aa": [0], "bb(1,1)": 2, "cc(1,1)": 3.5})
+    tm.assert_frame_equal(result, expected)
+
+
+# ValueError: the 'pyarrow' engine does not support regex separators
+@xfail_pyarrow
+def test_header_delim_whitespace(all_parsers):
+    # GH#54918
+    parser = all_parsers
+    data = """a,b
+1,2
+3,4
+    """
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+    expected = DataFrame({"a,b": ["1,2", "3,4"]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_no_header_pyarrow(pyarrow_parser_only):
+    parser = pyarrow_parser_only
+    data = """
+a,i,x
+b,j,y
+"""
+    result = parser.read_csv(
+        StringIO(data),
+        header=None,
+        usecols=[0, 1],
+        dtype="string[pyarrow]",
+        dtype_backend="pyarrow",
+        engine="pyarrow",
+    )
+    expected = DataFrame([["a", "i"], ["b", "j"]], dtype="string[pyarrow]")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_index_col.py b/pandas/pandas/tests/io/parser/test_index_col.py
new file mode 100644
index 0000000000000..9977e2b8e1a1d
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_index_col.py
@@ -0,0 +1,377 @@
+"""
+Tests that the specified index column (a.k.a "index_col")
+is properly handled or inferred during parsing for all of
+the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@pytest.mark.parametrize("with_header", [True, False])
+def test_index_col_named(all_parsers, with_header):
+    parser = all_parsers
+    no_header = """\
+KORD1,19990127, 19:00:00, 18:56:00, 0.8100, 2.8100, 7.2000, 0.0000, 280.0000
+KORD2,19990127, 20:00:00, 19:56:00, 0.0100, 2.2100, 7.2000, 0.0000, 260.0000
+KORD3,19990127, 21:00:00, 20:56:00, -0.5900, 2.2100, 5.7000, 0.0000, 280.0000
+KORD4,19990127, 21:00:00, 21:18:00, -0.9900, 2.0100, 3.6000, 0.0000, 270.0000
+KORD5,19990127, 22:00:00, 21:56:00, -0.5900, 1.7100, 5.1000, 0.0000, 290.0000
+KORD6,19990127, 23:00:00, 22:56:00, -0.5900, 1.7100, 4.6000, 0.0000, 280.0000"""
+    header = "ID,date,NominalTime,ActualTime,TDew,TAir,Windspeed,Precip,WindDir\n"
+
+    if with_header:
+        data = header + no_header
+
+        result = parser.read_csv(StringIO(data), index_col="ID")
+        expected = parser.read_csv(StringIO(data), header=0).set_index("ID")
+        tm.assert_frame_equal(result, expected)
+    else:
+        data = no_header
+        msg = "Index ID invalid"
+
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), index_col="ID")
+
+
+def test_index_col_named2(all_parsers):
+    parser = all_parsers
+    data = """\
+1,2,3,4,hello
+5,6,7,8,world
+9,10,11,12,foo
+"""
+
+    expected = DataFrame(
+        {"a": [1, 5, 9], "b": [2, 6, 10], "c": [3, 7, 11], "d": [4, 8, 12]},
+        index=Index(["hello", "world", "foo"], name="message"),
+    )
+    names = ["a", "b", "c", "d", "message"]
+
+    result = parser.read_csv(StringIO(data), names=names, index_col=["message"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_index_col_is_true(all_parsers):
+    # see gh-9798
+    data = "a,b\n1,2"
+    parser = all_parsers
+
+    msg = "The value of index_col couldn't be 'True'"
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), index_col=True)
+
+
+@skip_pyarrow  # CSV parse error: Expected 3 columns, got 4
+def test_infer_index_col(all_parsers):
+    data = """A,B,C
+foo,1,2,3
+bar,4,5,6
+baz,7,8,9
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data))
+
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        index=["foo", "bar", "baz"],
+        columns=["A", "B", "C"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+@pytest.mark.parametrize(
+    "index_col,kwargs",
+    [
+        (None, {"columns": ["x", "y", "z"]}),
+        (False, {"columns": ["x", "y", "z"]}),
+        (0, {"columns": ["y", "z"], "index": Index([], name="x")}),
+        (1, {"columns": ["x", "z"], "index": Index([], name="y")}),
+        ("x", {"columns": ["y", "z"], "index": Index([], name="x")}),
+        ("y", {"columns": ["x", "z"], "index": Index([], name="y")}),
+        (
+            [0, 1],
+            {
+                "columns": ["z"],
+                "index": MultiIndex.from_arrays([[]] * 2, names=["x", "y"]),
+            },
+        ),
+        (
+            ["x", "y"],
+            {
+                "columns": ["z"],
+                "index": MultiIndex.from_arrays([[]] * 2, names=["x", "y"]),
+            },
+        ),
+        (
+            [1, 0],
+            {
+                "columns": ["z"],
+                "index": MultiIndex.from_arrays([[]] * 2, names=["y", "x"]),
+            },
+        ),
+        (
+            ["y", "x"],
+            {
+                "columns": ["z"],
+                "index": MultiIndex.from_arrays([[]] * 2, names=["y", "x"]),
+            },
+        ),
+    ],
+)
+def test_index_col_empty_data(all_parsers, index_col, kwargs):
+    data = "x,y,z"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=index_col)
+
+    expected = DataFrame(**kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_empty_with_index_col_false(all_parsers):
+    # see gh-10413
+    data = "x,y"
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col=False)
+
+    expected = DataFrame(columns=["x", "y"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index_names",
+    [
+        ["", ""],
+        ["foo", ""],
+        ["", "bar"],
+        ["foo", "bar"],
+        ["NotReallyUnnamed", "Unnamed: 0"],
+    ],
+)
+def test_multi_index_naming(all_parsers, index_names, request):
+    parser = all_parsers
+
+    if parser.engine == "pyarrow" and "" in index_names:
+        mark = pytest.mark.xfail(reason="One case raises, others are wrong")
+        request.applymarker(mark)
+
+    # We don't want empty index names being replaced with "Unnamed: 0"
+    data = ",".join(index_names + ["col\na,c,1\na,d,2\nb,c,3\nb,d,4"])
+    result = parser.read_csv(StringIO(data), index_col=[0, 1])
+
+    expected = DataFrame(
+        {"col": [1, 2, 3, 4]}, index=MultiIndex.from_product([["a", "b"], ["c", "d"]])
+    )
+    expected.index.names = [name if name else None for name in index_names]
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_multi_index_naming_not_all_at_beginning(all_parsers):
+    parser = all_parsers
+    data = ",Unnamed: 2,\na,c,1\na,d,2\nb,c,3\nb,d,4"
+    result = parser.read_csv(StringIO(data), index_col=[0, 2])
+
+    expected = DataFrame(
+        {"Unnamed: 2": ["c", "d", "c", "d"]},
+        index=MultiIndex(
+            levels=[["a", "b"], [1, 2, 3, 4]], codes=[[0, 0, 1, 1], [0, 1, 2, 3]]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_no_multi_index_level_names_empty(all_parsers):
+    # GH 10984
+    parser = all_parsers
+    midx = MultiIndex.from_tuples([("A", 1, 2), ("A", 1, 2), ("B", 1, 2)])
+    expected = DataFrame(
+        np.random.default_rng(2).standard_normal((3, 3)),
+        index=midx,
+        columns=["x", "y", "z"],
+    )
+    with tm.ensure_clean() as path:
+        expected.to_csv(path)
+        result = parser.read_csv(path, index_col=[0, 1, 2])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_header_with_index_col(all_parsers):
+    # GH 33476
+    parser = all_parsers
+    data = """
+I11,A,A
+I12,B,B
+I2,1,3
+"""
+    midx = MultiIndex.from_tuples([("A", "B"), ("A", "B.1")], names=["I11", "I12"])
+    idx = Index(["I2"])
+    expected = DataFrame([[1, 3]], index=idx, columns=midx)
+
+    result = parser.read_csv(StringIO(data), index_col=0, header=[0, 1])
+    tm.assert_frame_equal(result, expected)
+
+    col_idx = Index(["A", "A.1"])
+    idx = Index(["I12", "I2"], name="I11")
+    expected = DataFrame([["B", "B"], ["1", "3"]], index=idx, columns=col_idx)
+
+    result = parser.read_csv(StringIO(data), index_col="I11", header=0)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.slow
+def test_index_col_large_csv(all_parsers, monkeypatch):
+    # https://github.com/pandas-dev/pandas/issues/37094
+    parser = all_parsers
+
+    ARR_LEN = 100
+    df = DataFrame(
+        {
+            "a": range(ARR_LEN + 1),
+            "b": np.random.default_rng(2).standard_normal(ARR_LEN + 1),
+        }
+    )
+
+    with tm.ensure_clean() as path:
+        df.to_csv(path, index=False)
+        with monkeypatch.context() as m:
+            m.setattr("pandas.core.algorithms._MINIMUM_COMP_ARR_LEN", ARR_LEN)
+            result = parser.read_csv(path, index_col=[0])
+
+    tm.assert_frame_equal(result, df.set_index("a"))
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_index_col_multiindex_columns_no_data(all_parsers):
+    # GH#38292
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO("a0,a1,a2\nb0,b1,b2\n"), header=[0, 1], index_col=0
+    )
+    expected = DataFrame(
+        [],
+        index=Index([]),
+        columns=MultiIndex.from_arrays(
+            [["a1", "a2"], ["b1", "b2"]], names=["a0", "b0"]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_index_col_header_no_data(all_parsers):
+    # GH#38292
+    parser = all_parsers
+    result = parser.read_csv(StringIO("a0,a1,a2\n"), header=[0], index_col=0)
+    expected = DataFrame(
+        [],
+        columns=["a1", "a2"],
+        index=Index([], name="a0"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_multiindex_columns_no_data(all_parsers):
+    # GH#38292
+    parser = all_parsers
+    result = parser.read_csv(StringIO("a0,a1,a2\nb0,b1,b2\n"), header=[0, 1])
+    expected = DataFrame(
+        [], columns=MultiIndex.from_arrays([["a0", "a1", "a2"], ["b0", "b1", "b2"]])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_multiindex_columns_index_col_with_data(all_parsers):
+    # GH#38292
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO("a0,a1,a2\nb0,b1,b2\ndata,data,data"), header=[0, 1], index_col=0
+    )
+    expected = DataFrame(
+        [["data", "data"]],
+        columns=MultiIndex.from_arrays(
+            [["a1", "a2"], ["b1", "b2"]], names=["a0", "b0"]
+        ),
+        index=Index(["data"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Empty CSV file or block
+def test_infer_types_boolean_sum(all_parsers):
+    # GH#44079
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO("0,1"),
+        names=["a", "b"],
+        index_col=["a"],
+        dtype={"a": "UInt8"},
+    )
+    expected = DataFrame(
+        data={
+            "a": [
+                0,
+            ],
+            "b": [1],
+        }
+    ).set_index("a")
+    # Not checking index type now, because the C parser will return a
+    # index column of dtype 'object', and the Python parser will return a
+    # index column of dtype 'int64'.
+    tm.assert_frame_equal(result, expected, check_index_type=False)
+
+
+@pytest.mark.parametrize("dtype, val", [(object, "01"), ("int64", 1)])
+def test_specify_dtype_for_index_col(all_parsers, dtype, val, request):
+    # GH#9435
+    data = "a,b\n01,2"
+    parser = all_parsers
+    if dtype == object and parser.engine == "pyarrow":
+        request.applymarker(
+            pytest.mark.xfail(reason="Cannot disable type-inference for pyarrow engine")
+        )
+    result = parser.read_csv(StringIO(data), index_col="a", dtype={"a": dtype})
+    expected = DataFrame({"b": [2]}, index=Index([val], name="a", dtype=dtype))
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_multiindex_columns_not_leading_index_col(all_parsers):
+    # GH#38549
+    parser = all_parsers
+    data = """a,b,c,d
+e,f,g,h
+x,y,1,2
+"""
+    result = parser.read_csv(
+        StringIO(data),
+        header=[0, 1],
+        index_col=1,
+    )
+    cols = MultiIndex.from_tuples(
+        [("a", "e"), ("c", "g"), ("d", "h")], names=["b", "f"]
+    )
+    expected = DataFrame([["x", 1, 2]], columns=cols, index=["y"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_mangle_dupes.py b/pandas/pandas/tests/io/parser/test_mangle_dupes.py
new file mode 100644
index 0000000000000..55c8bbc4bb9e1
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_mangle_dupes.py
@@ -0,0 +1,183 @@
+"""
+Tests that duplicate columns are handled appropriately when parsed by the
+CSV engine. In general, the expected result is that they are either thoroughly
+de-duplicated (if mangling requested) or ignored otherwise.
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_basic(all_parsers):
+    parser = all_parsers
+
+    data = "a,a,b,b,b\n1,2,3,4,5"
+    result = parser.read_csv(StringIO(data), sep=",")
+
+    expected = DataFrame([[1, 2, 3, 4, 5]], columns=["a", "a.1", "b", "b.1", "b.2"])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_basic_names(all_parsers):
+    # See gh-7160
+    parser = all_parsers
+
+    data = "a,b,a\n0,1,2\n3,4,5"
+    expected = DataFrame([[0, 1, 2], [3, 4, 5]], columns=["a", "b", "a.1"])
+
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_basic_names_raise(all_parsers):
+    # See gh-7160
+    parser = all_parsers
+
+    data = "0,1,2\n3,4,5"
+    with pytest.raises(ValueError, match="Duplicate names"):
+        parser.read_csv(StringIO(data), names=["a", "b", "a"])
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        ("a,a,a.1\n1,2,3", DataFrame([[1, 2, 3]], columns=["a", "a.2", "a.1"])),
+        (
+            "a,a,a.1,a.1.1,a.1.1.1,a.1.1.1.1\n1,2,3,4,5,6",
+            DataFrame(
+                [[1, 2, 3, 4, 5, 6]],
+                columns=["a", "a.2", "a.1", "a.1.1", "a.1.1.1", "a.1.1.1.1"],
+            ),
+        ),
+        (
+            "a,a,a.3,a.1,a.2,a,a\n1,2,3,4,5,6,7",
+            DataFrame(
+                [[1, 2, 3, 4, 5, 6, 7]],
+                columns=["a", "a.4", "a.3", "a.1", "a.2", "a.5", "a.6"],
+            ),
+        ),
+    ],
+)
+def test_thorough_mangle_columns(all_parsers, data, expected):
+    # see gh-17060
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,names,expected",
+    [
+        (
+            "a,b,b\n1,2,3",
+            ["a.1", "a.1", "a.1.1"],
+            DataFrame(
+                [["a", "b", "b"], ["1", "2", "3"]], columns=["a.1", "a.1.1", "a.1.1.1"]
+            ),
+        ),
+        (
+            "a,b,c,d,e,f\n1,2,3,4,5,6",
+            ["a", "a", "a.1", "a.1.1", "a.1.1.1", "a.1.1.1.1"],
+            DataFrame(
+                [["a", "b", "c", "d", "e", "f"], ["1", "2", "3", "4", "5", "6"]],
+                columns=["a", "a.1", "a.1.1", "a.1.1.1", "a.1.1.1.1", "a.1.1.1.1.1"],
+            ),
+        ),
+        (
+            "a,b,c,d,e,f,g\n1,2,3,4,5,6,7",
+            ["a", "a", "a.3", "a.1", "a.2", "a", "a"],
+            DataFrame(
+                [
+                    ["a", "b", "c", "d", "e", "f", "g"],
+                    ["1", "2", "3", "4", "5", "6", "7"],
+                ],
+                columns=["a", "a.1", "a.3", "a.1.1", "a.2", "a.2.1", "a.3.1"],
+            ),
+        ),
+    ],
+)
+def test_thorough_mangle_names(all_parsers, data, names, expected):
+    # see gh-17095
+    parser = all_parsers
+
+    with pytest.raises(ValueError, match="Duplicate names"):
+        parser.read_csv(StringIO(data), names=names)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame.columns are different
+def test_mangled_unnamed_placeholders(all_parsers):
+    # xref gh-13017
+    orig_key = "0"
+    parser = all_parsers
+
+    orig_value = [1, 2, 3]
+    df = DataFrame({orig_key: orig_value})
+
+    # This test recursively updates `df`.
+    for i in range(3):
+        expected = DataFrame(columns=Index([], dtype="str"))
+
+        for j in range(i + 1):
+            col_name = "Unnamed: 0" + f".{1 * j}" * min(j, 1)
+            expected.insert(loc=0, column=col_name, value=[0, 1, 2])
+
+        expected[orig_key] = orig_value
+        df = parser.read_csv(StringIO(df.to_csv()))
+
+        tm.assert_frame_equal(df, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_mangle_dupe_cols_already_exists(all_parsers):
+    # GH#14704
+    parser = all_parsers
+
+    data = "a,a,a.1,a,a.3,a.1,a.1.1\n1,2,3,4,5,6,7"
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [[1, 2, 3, 4, 5, 6, 7]],
+        columns=["a", "a.2", "a.1", "a.4", "a.3", "a.1.2", "a.1.1"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: Found non-unique column index
+def test_mangle_dupe_cols_already_exists_unnamed_col(all_parsers):
+    # GH#14704
+    parser = all_parsers
+
+    data = ",Unnamed: 0,,Unnamed: 2\n1,2,3,4"
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [[1, 2, 3, 4]],
+        columns=["Unnamed: 0.1", "Unnamed: 0", "Unnamed: 2.1", "Unnamed: 2"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("usecol, engine", [([0, 1, 1], "python"), ([0, 1, 1], "c")])
+def test_mangle_cols_names(all_parsers, usecol, engine):
+    # GH 11823
+    parser = all_parsers
+    data = "1,2,3"
+    names = ["A", "A", "B"]
+    with pytest.raises(ValueError, match="Duplicate names"):
+        parser.read_csv(StringIO(data), names=names, usecols=usecol, engine=engine)
diff --git a/pandas/pandas/tests/io/parser/test_multi_thread.py b/pandas/pandas/tests/io/parser/test_multi_thread.py
new file mode 100644
index 0000000000000..348c19ac0f0c6
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_multi_thread.py
@@ -0,0 +1,159 @@
+"""
+Tests multithreading behaviour for reading and
+parsing files for each parser defined in parsers.py
+"""
+
+from contextlib import ExitStack
+from io import BytesIO
+from multiprocessing.pool import ThreadPool
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.util.version import Version
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+
+# We'll probably always skip these for pyarrow
+# Maybe we'll add our own tests for pyarrow too
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.slow,
+]
+
+
+@pytest.mark.filterwarnings("ignore:Passing a BlockManager:DeprecationWarning")
+def test_multi_thread_string_io_read_csv(all_parsers, request):
+    # see gh-11786
+    parser = all_parsers
+    if parser.engine == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+        if Version(pa.__version__) < Version("16.0"):
+            request.applymarker(
+                pytest.mark.xfail(reason="# ValueError: Found non-unique column index")
+            )
+    max_row_range = 100
+    num_files = 10
+
+    bytes_to_df = (
+        "\n".join([f"{i:d},{i:d},{i:d}" for i in range(max_row_range)]).encode()
+        for _ in range(num_files)
+    )
+
+    # Read all files in many threads.
+    with ExitStack() as stack:
+        files = [stack.enter_context(BytesIO(b)) for b in bytes_to_df]
+
+        pool = stack.enter_context(ThreadPool(8))
+
+        results = pool.map(parser.read_csv, files)
+        first_result = results[0]
+
+        for result in results:
+            tm.assert_frame_equal(first_result, result)
+
+
+def _generate_multi_thread_dataframe(parser, path, num_rows, num_tasks):
+    """
+    Generate a DataFrame via multi-thread.
+
+    Parameters
+    ----------
+    parser : BaseParser
+        The parser object to use for reading the data.
+    path : str
+        The location of the CSV file to read.
+    num_rows : int
+        The number of rows to read per task.
+    num_tasks : int
+        The number of tasks to use for reading this DataFrame.
+
+    Returns
+    -------
+    df : DataFrame
+    """
+
+    def reader(arg):
+        """
+        Create a reader for part of the CSV.
+
+        Parameters
+        ----------
+        arg : tuple
+            A tuple of the following:
+
+            * start : int
+                The starting row to start for parsing CSV
+            * nrows : int
+                The number of rows to read.
+
+        Returns
+        -------
+        df : DataFrame
+        """
+        start, nrows = arg
+
+        if not start:
+            return parser.read_csv(
+                path, index_col=0, header=0, nrows=nrows, parse_dates=["date"]
+            )
+
+        return parser.read_csv(
+            path,
+            index_col=0,
+            header=None,
+            skiprows=int(start) + 1,
+            nrows=nrows,
+            parse_dates=[9],
+        )
+
+    tasks = [
+        (num_rows * i // num_tasks, num_rows // num_tasks) for i in range(num_tasks)
+    ]
+
+    with ThreadPool(processes=num_tasks) as pool:
+        results = pool.map(reader, tasks)
+
+    header = results[0].columns
+
+    for r in results[1:]:
+        r.columns = header
+
+    final_dataframe = pd.concat(results)
+    return final_dataframe
+
+
+@xfail_pyarrow  # ValueError: The 'nrows' option is not supported
+def test_multi_thread_path_multipart_read_csv(all_parsers):
+    # see gh-11786
+    num_tasks = 4
+    num_rows = 48
+
+    parser = all_parsers
+    file_name = "__thread_pool_reader__.csv"
+    df = DataFrame(
+        {
+            "a": np.random.default_rng(2).random(num_rows),
+            "b": np.random.default_rng(2).random(num_rows),
+            "c": np.random.default_rng(2).random(num_rows),
+            "d": np.random.default_rng(2).random(num_rows),
+            "e": np.random.default_rng(2).random(num_rows),
+            "foo": ["foo"] * num_rows,
+            "bar": ["bar"] * num_rows,
+            "baz": ["baz"] * num_rows,
+            "date": pd.date_range("20000101 09:00:00", periods=num_rows, freq="s"),
+            "int": np.arange(num_rows, dtype="int64"),
+        }
+    )
+
+    with tm.ensure_clean(file_name) as path:
+        df.to_csv(path)
+
+        result = _generate_multi_thread_dataframe(parser, path, num_rows, num_tasks)
+
+    expected = df[:]
+    expected["date"] = expected["date"].astype("M8[s]")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_na_values.py b/pandas/pandas/tests/io/parser/test_na_values.py
new file mode 100644
index 0000000000000..3a68d38cc0bde
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_na_values.py
@@ -0,0 +1,833 @@
+"""
+Tests that NA values are properly handled during
+parsing for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas._libs.parsers import STR_NA_VALUES
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_string_nas(all_parsers):
+    parser = all_parsers
+    data = """A,B,C
+a,b,c
+d,,f
+,g,h
+"""
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [["a", "b", "c"], ["d", np.nan, "f"], [np.nan, "g", "h"]],
+        columns=["A", "B", "C"],
+    )
+    if parser.engine == "pyarrow":
+        expected.loc[2, "A"] = None
+        expected.loc[1, "B"] = None
+    tm.assert_frame_equal(result, expected)
+
+
+def test_detect_string_na(all_parsers):
+    parser = all_parsers
+    data = """A,B
+foo,bar
+NA,baz
+NaN,nan
+"""
+    expected = DataFrame(
+        [["foo", "bar"], [np.nan, "baz"], [np.nan, np.nan]], columns=["A", "B"]
+    )
+    if parser.engine == "pyarrow":
+        expected.loc[[1, 2], "A"] = None
+        expected.loc[2, "B"] = None
+    result = parser.read_csv(StringIO(data))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na_values",
+    [
+        ["-999.0", "-999"],
+        [-999, -999.0],
+        [-999.0, -999],
+        ["-999.0"],
+        ["-999"],
+        [-999.0],
+        [-999],
+    ],
+)
+@pytest.mark.parametrize(
+    "data",
+    [
+        """A,B
+-999,1.2
+2,-999
+3,4.5
+""",
+        """A,B
+-999,1.200
+2,-999.000
+3,4.500
+""",
+    ],
+)
+def test_non_string_na_values(all_parsers, data, na_values, request):
+    # see gh-3611: with an odd float format, we can't match
+    # the string "999.0" exactly but still need float matching
+    parser = all_parsers
+    expected = DataFrame([[np.nan, 1.2], [2.0, np.nan], [3.0, 4.5]], columns=["A", "B"])
+
+    if parser.engine == "pyarrow" and not all(isinstance(x, str) for x in na_values):
+        msg = "The 'pyarrow' engine requires all na_values to be strings"
+        with pytest.raises(TypeError, match=msg):
+            parser.read_csv(StringIO(data), na_values=na_values)
+        return
+    elif parser.engine == "pyarrow" and "-999.000" in data:
+        # bc the pyarrow engine does not include the float-ified version
+        #  of "-999" -> -999, it does not match the entry with the trailing
+        #  zeros, so "-999.000" is not treated as null.
+        mark = pytest.mark.xfail(
+            reason="pyarrow engined does not recognize equivalent floats"
+        )
+        request.applymarker(mark)
+
+    result = parser.read_csv(StringIO(data), na_values=na_values)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_default_na_values(all_parsers):
+    _NA_VALUES = {
+        "-1.#IND",
+        "1.#QNAN",
+        "1.#IND",
+        "-1.#QNAN",
+        "#N/A",
+        "N/A",
+        "n/a",
+        "NA",
+        "<NA>",
+        "#NA",
+        "NULL",
+        "null",
+        "NaN",
+        "nan",
+        "-NaN",
+        "-nan",
+        "#N/A N/A",
+        "",
+        "None",
+    }
+    assert _NA_VALUES == STR_NA_VALUES
+
+    parser = all_parsers
+    nv = len(_NA_VALUES)
+
+    def f(i, v):
+        if i == 0:
+            buf = ""
+        elif i > 0:
+            buf = "".join([","] * i)
+
+        buf = f"{buf}{v}"
+
+        if i < nv - 1:
+            joined = "".join([","] * (nv - i - 1))
+            buf = f"{buf}{joined}"
+
+        return buf
+
+    data = StringIO("\n".join([f(i, v) for i, v in enumerate(_NA_VALUES)]))
+    expected = DataFrame(np.nan, columns=range(nv), index=range(nv))
+
+    result = parser.read_csv(data, header=None)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("na_values", ["baz", ["baz"]])
+def test_custom_na_values(all_parsers, na_values):
+    parser = all_parsers
+    data = """A,B,C
+ignore,this,row
+1,NA,3
+-1.#IND,5,baz
+7,8,NaN
+"""
+    expected = DataFrame(
+        [[1.0, np.nan, 3], [np.nan, 5, np.nan], [7, 8, np.nan]], columns=["A", "B", "C"]
+    )
+    if parser.engine == "pyarrow":
+        msg = "skiprows argument must be an integer when using engine='pyarrow'"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), na_values=na_values, skiprows=[1])
+        return
+
+    result = parser.read_csv(StringIO(data), na_values=na_values, skiprows=[1])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_bool_na_values(all_parsers):
+    data = """A,B,C
+True,False,True
+NA,True,False
+False,NA,True"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        {
+            "A": np.array([True, np.nan, False], dtype=object),
+            "B": np.array([False, True, np.nan], dtype=object),
+            "C": [True, False, True],
+        }
+    )
+    if parser.engine == "pyarrow":
+        expected.loc[1, "A"] = None
+        expected.loc[2, "B"] = None
+    tm.assert_frame_equal(result, expected)
+
+
+def test_na_value_dict(all_parsers):
+    data = """A,B,C
+foo,bar,NA
+bar,foo,foo
+foo,bar,NA
+bar,foo,foo"""
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), na_values={"A": ["foo"], "B": ["bar"]})
+        return
+
+    df = parser.read_csv(StringIO(data), na_values={"A": ["foo"], "B": ["bar"]})
+    expected = DataFrame(
+        {
+            "A": [np.nan, "bar", np.nan, "bar"],
+            "B": [np.nan, "foo", np.nan, "foo"],
+            "C": [np.nan, "foo", np.nan, "foo"],
+        }
+    )
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize(
+    "index_col,expected",
+    [
+        (
+            [0],
+            DataFrame({"b": [np.nan], "c": [1], "d": [5]}, index=Index([0], name="a")),
+        ),
+        (
+            [0, 2],
+            DataFrame(
+                {"b": [np.nan], "d": [5]},
+                index=MultiIndex.from_tuples([(0, 1)], names=["a", "c"]),
+            ),
+        ),
+        (
+            ["a", "c"],
+            DataFrame(
+                {"b": [np.nan], "d": [5]},
+                index=MultiIndex.from_tuples([(0, 1)], names=["a", "c"]),
+            ),
+        ),
+    ],
+)
+def test_na_value_dict_multi_index(all_parsers, index_col, expected):
+    data = """\
+a,b,c,d
+0,NA,1,5
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), na_values=set(), index_col=index_col)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,expected",
+    [
+        (
+            {},
+            {
+                "A": ["a", "b", np.nan, "d", "e", np.nan, "g"],
+                "B": [1, 2, 3, 4, 5, 6, 7],
+                "C": ["one", "two", "three", np.nan, "five", np.nan, "seven"],
+            },
+        ),
+        (
+            {"na_values": {"A": [], "C": []}, "keep_default_na": False},
+            {
+                "A": ["a", "b", "", "d", "e", "nan", "g"],
+                "B": [1, 2, 3, 4, 5, 6, 7],
+                "C": ["one", "two", "three", "nan", "five", "", "seven"],
+            },
+        ),
+        (
+            {"na_values": ["a"], "keep_default_na": False},
+            {
+                "A": [np.nan, "b", "", "d", "e", "nan", "g"],
+                "B": [1, 2, 3, 4, 5, 6, 7],
+                "C": ["one", "two", "three", "nan", "five", "", "seven"],
+            },
+        ),
+        (
+            {"na_values": {"A": [], "C": []}},
+            {
+                "A": ["a", "b", np.nan, "d", "e", np.nan, "g"],
+                "B": [1, 2, 3, 4, 5, 6, 7],
+                "C": ["one", "two", "three", np.nan, "five", np.nan, "seven"],
+            },
+        ),
+    ],
+)
+def test_na_values_keep_default(
+    all_parsers, kwargs, expected, request, using_infer_string
+):
+    data = """\
+A,B,C
+a,1,one
+b,2,two
+,3,three
+d,4,nan
+e,5,five
+nan,6,
+g,7,seven
+"""
+    parser = all_parsers
+    if parser.engine == "pyarrow":
+        if "na_values" in kwargs and isinstance(kwargs["na_values"], dict):
+            msg = "The pyarrow engine doesn't support passing a dict for na_values"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(StringIO(data), **kwargs)
+            return
+        if not using_infer_string or "na_values" in kwargs:
+            mark = pytest.mark.xfail()
+            request.applymarker(mark)
+
+    result = parser.read_csv(StringIO(data), **kwargs)
+    expected = DataFrame(expected)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_na_values_no_keep_default(all_parsers):
+    # see gh-4318: passing na_values=None and
+    # keep_default_na=False yields 'None" as a na_value
+    data = """\
+A,B,C
+a,1,None
+b,2,two
+,3,None
+d,4,nan
+e,5,five
+nan,6,
+g,7,seven
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), keep_default_na=False)
+
+    expected = DataFrame(
+        {
+            "A": ["a", "b", "", "d", "e", "nan", "g"],
+            "B": [1, 2, 3, 4, 5, 6, 7],
+            "C": ["None", "two", "None", "nan", "five", "", "seven"],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_keep_default_na_dict_na_values(all_parsers):
+    # see gh-19227
+    data = "a,b\n,2"
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data), na_values={"b": ["2"]}, keep_default_na=False
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data), na_values={"b": ["2"]}, keep_default_na=False
+    )
+    expected = DataFrame({"a": [""], "b": [np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_keep_default_na_dict_na_scalar_values(all_parsers):
+    # see gh-19227
+    #
+    # Scalar values shouldn't cause the parsing to crash or fail.
+    data = "a,b\n1,2"
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), na_values={"b": 2}, keep_default_na=False)
+        return
+
+    df = parser.read_csv(StringIO(data), na_values={"b": 2}, keep_default_na=False)
+    expected = DataFrame({"a": [1], "b": [np.nan]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("col_zero_na_values", [113125, "113125"])
+def test_no_keep_default_na_dict_na_values_diff_reprs(all_parsers, col_zero_na_values):
+    # see gh-19227
+    data = """\
+113125,"blah","/blaha",kjsdkj,412.166,225.874,214.008
+729639,"qwer","",asdfkj,466.681,,252.373
+"""
+    parser = all_parsers
+    expected = DataFrame(
+        {
+            0: [np.nan, 729639.0],
+            1: [np.nan, "qwer"],
+            2: ["/blaha", np.nan],
+            3: ["kjsdkj", "asdfkj"],
+            4: [412.166, 466.681],
+            5: ["225.874", ""],
+            6: [np.nan, 252.373],
+        }
+    )
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                header=None,
+                keep_default_na=False,
+                na_values={2: "", 6: "214.008", 1: "blah", 0: col_zero_na_values},
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data),
+        header=None,
+        keep_default_na=False,
+        na_values={2: "", 6: "214.008", 1: "blah", 0: col_zero_na_values},
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na_filter,row_data",
+    [
+        (True, [[1, "A"], [np.nan, np.nan], [3, "C"]]),
+        (False, [["1", "A"], ["nan", "B"], ["3", "C"]]),
+    ],
+)
+def test_na_values_na_filter_override(
+    request, all_parsers, na_filter, row_data, using_infer_string
+):
+    parser = all_parsers
+    if parser.engine == "pyarrow":
+        # mismatched dtypes in both cases, FutureWarning in the True case
+        if not (using_infer_string and na_filter):
+            mark = pytest.mark.xfail(reason="pyarrow doesn't support this.")
+            request.applymarker(mark)
+    data = """\
+A,B
+1,A
+nan,B
+3,C
+"""
+    result = parser.read_csv(StringIO(data), na_values=["B"], na_filter=na_filter)
+
+    expected = DataFrame(row_data, columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Expected 8 columns, got 5:
+def test_na_trailing_columns(all_parsers):
+    parser = all_parsers
+    data = """Date,Currency,Symbol,Type,Units,UnitPrice,Cost,Tax
+2012-03-14,USD,AAPL,BUY,1000
+2012-05-12,USD,SBUX,SELL,500"""
+
+    # Trailing columns should be all NaN.
+    result = parser.read_csv(StringIO(data))
+    expected = DataFrame(
+        [
+            ["2012-03-14", "USD", "AAPL", "BUY", 1000, np.nan, np.nan, np.nan],
+            ["2012-05-12", "USD", "SBUX", "SELL", 500, np.nan, np.nan, np.nan],
+        ],
+        columns=[
+            "Date",
+            "Currency",
+            "Symbol",
+            "Type",
+            "Units",
+            "UnitPrice",
+            "Cost",
+            "Tax",
+        ],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na_values,row_data",
+    [
+        (1, [[np.nan, 2.0], [2.0, np.nan]]),
+        ({"a": 2, "b": 1}, [[1.0, 2.0], [np.nan, np.nan]]),
+    ],
+)
+def test_na_values_scalar(all_parsers, na_values, row_data):
+    # see gh-12224
+    parser = all_parsers
+    names = ["a", "b"]
+    data = "1,2\n2,1"
+
+    if parser.engine == "pyarrow" and isinstance(na_values, dict):
+        if isinstance(na_values, dict):
+            err = ValueError
+            msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        else:
+            err = TypeError
+            msg = "The 'pyarrow' engine requires all na_values to be strings"
+        with pytest.raises(err, match=msg):
+            parser.read_csv(StringIO(data), names=names, na_values=na_values)
+        return
+    elif parser.engine == "pyarrow":
+        msg = "The 'pyarrow' engine requires all na_values to be strings"
+        with pytest.raises(TypeError, match=msg):
+            parser.read_csv(StringIO(data), names=names, na_values=na_values)
+        return
+
+    result = parser.read_csv(StringIO(data), names=names, na_values=na_values)
+    expected = DataFrame(row_data, columns=names)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_na_values_dict_aliasing(all_parsers):
+    parser = all_parsers
+    na_values = {"a": 2, "b": 1}
+    na_values_copy = na_values.copy()
+
+    names = ["a", "b"]
+    data = "1,2\n2,1"
+
+    expected = DataFrame([[1.0, 2.0], [np.nan, np.nan]], columns=names)
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), names=names, na_values=na_values)
+        return
+
+    result = parser.read_csv(StringIO(data), names=names, na_values=na_values)
+
+    tm.assert_frame_equal(result, expected)
+    tm.assert_dict_equal(na_values, na_values_copy)
+
+
+def test_na_values_dict_null_column_name(all_parsers):
+    # see gh-57547
+    parser = all_parsers
+    data = ",x,y\n\nMA,1,2\nNA,2,1\nOA,,3"
+    names = [None, "x", "y"]
+    na_values = {name: STR_NA_VALUES for name in names}
+    dtype = {None: "object", "x": "float64", "y": "float64"}
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                index_col=0,
+                header=0,
+                dtype=dtype,
+                names=names,
+                na_values=na_values,
+                keep_default_na=False,
+            )
+        return
+
+    expected = DataFrame(
+        {"x": [1.0, 2.0, np.nan], "y": [2.0, 1.0, 3.0]},
+        index=Index(["MA", "NA", "OA"], dtype=object),
+    )
+
+    result = parser.read_csv(
+        StringIO(data),
+        index_col=0,
+        header=0,
+        dtype=dtype,
+        names=names,
+        na_values=na_values,
+        keep_default_na=False,
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_na_values_dict_col_index(all_parsers):
+    # see gh-14203
+    data = "a\nfoo\n1"
+    parser = all_parsers
+    na_values = {0: "foo"}
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), na_values=na_values)
+        return
+
+    result = parser.read_csv(StringIO(data), na_values=na_values)
+    expected = DataFrame({"a": [np.nan, 1]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        (
+            str(2**63) + "\n" + str(2**63 + 1),
+            {"na_values": [2**63]},
+            [str(2**63), str(2**63 + 1)],
+        ),
+        (str(2**63) + ",1" + "\n,2", {}, [[str(2**63), 1], ["", 2]]),
+        (str(2**63) + "\n1", {"na_values": [2**63]}, [np.nan, 1]),
+    ],
+)
+def test_na_values_uint64(all_parsers, data, kwargs, expected, request):
+    # see gh-14983
+    parser = all_parsers
+
+    if parser.engine == "pyarrow" and "na_values" in kwargs:
+        msg = "The 'pyarrow' engine requires all na_values to be strings"
+        with pytest.raises(TypeError, match=msg):
+            parser.read_csv(StringIO(data), header=None, **kwargs)
+        return
+    elif parser.engine == "pyarrow":
+        mark = pytest.mark.xfail(reason="Returns float64 instead of object")
+        request.applymarker(mark)
+
+    result = parser.read_csv(StringIO(data), header=None, **kwargs)
+    expected = DataFrame(expected)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_empty_na_values_no_default_with_index(all_parsers):
+    # see gh-15835
+    data = "a,1\nb,2"
+    parser = all_parsers
+    expected = DataFrame({"1": [2]}, index=Index(["b"], name="a"))
+
+    result = parser.read_csv(StringIO(data), index_col=0, keep_default_na=False)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na_filter,index_data", [(False, ["", "5"]), (True, [np.nan, 5.0])]
+)
+def test_no_na_filter_on_index(all_parsers, na_filter, index_data, request):
+    # see gh-5239
+    #
+    # Don't parse NA-values in index unless na_filter=True
+    parser = all_parsers
+    data = "a,b,c\n1,,3\n4,5,6"
+
+    if parser.engine == "pyarrow" and na_filter is False:
+        mark = pytest.mark.xfail(reason="mismatched index result")
+        request.applymarker(mark)
+
+    expected = DataFrame({"a": [1, 4], "c": [3, 6]}, index=Index(index_data, name="b"))
+    result = parser.read_csv(StringIO(data), index_col=[1], na_filter=na_filter)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_inf_na_values_with_int_index(all_parsers):
+    # see gh-17128
+    parser = all_parsers
+    data = "idx,col1,col2\n1,3,4\n2,inf,-inf"
+
+    # Don't fail with OverflowError with inf's and integer index column.
+    out = parser.read_csv(StringIO(data), index_col=[0], na_values=["inf", "-inf"])
+    expected = DataFrame(
+        {"col1": [3, np.nan], "col2": [4, np.nan]}, index=Index([1, 2], name="idx")
+    )
+    tm.assert_frame_equal(out, expected)
+
+
+@xfail_pyarrow  # mismatched shape
+@pytest.mark.parametrize("na_filter", [True, False])
+def test_na_values_with_dtype_str_and_na_filter(all_parsers, na_filter):
+    # see gh-20377
+    parser = all_parsers
+    data = "a,b,c\n1,,3\n4,5,6"
+
+    # na_filter=True --> missing value becomes NaN.
+    # na_filter=False --> missing value remains empty string.
+    empty = np.nan if na_filter else ""
+    expected = DataFrame({"a": ["1", "4"], "b": [empty, "5"], "c": ["3", "6"]})
+
+    result = parser.read_csv(StringIO(data), na_filter=na_filter, dtype=str)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # mismatched exception message
+@pytest.mark.parametrize(
+    "data, na_values",
+    [
+        ("false,1\n,1\ntrue", None),
+        ("false,1\nnull,1\ntrue", None),
+        ("false,1\nnan,1\ntrue", None),
+        ("false,1\nfoo,1\ntrue", "foo"),
+        ("false,1\nfoo,1\ntrue", ["foo"]),
+        ("false,1\nfoo,1\ntrue", {"a": "foo"}),
+    ],
+)
+def test_cast_NA_to_bool_raises_error(all_parsers, data, na_values):
+    parser = all_parsers
+    msg = "|".join(
+        [
+            "Bool column has NA values in column [0a]",
+            "cannot safely convert passed user dtype of "
+            "bool for object dtyped data in column 0",
+        ]
+    )
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(
+            StringIO(data),
+            header=None,
+            names=["a", "b"],
+            dtype={"a": "bool"},
+            na_values=na_values,
+        )
+
+
+# TODO: this test isn't about the na_values keyword, it is about the empty entries
+#  being returned with NaN entries, whereas the pyarrow engine returns "nan"
+@xfail_pyarrow  # mismatched shapes
+def test_str_nan_dropped(all_parsers):
+    # see gh-21131
+    parser = all_parsers
+
+    data = """File: small.csv,,
+10010010233,0123,654
+foo,,bar
+01001000155,4530,898"""
+
+    result = parser.read_csv(
+        StringIO(data),
+        header=None,
+        names=["col1", "col2", "col3"],
+        dtype={"col1": str, "col2": str, "col3": str},
+    ).dropna()
+
+    expected = DataFrame(
+        {
+            "col1": ["10010010233", "01001000155"],
+            "col2": ["0123", "4530"],
+            "col3": ["654", "898"],
+        },
+        index=[1, 3],
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nan_multi_index(all_parsers):
+    # GH 42446
+    parser = all_parsers
+    data = "A,B,B\nX,Y,Z\n1,2,inf"
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine doesn't support passing a dict for na_values"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data), header=list(range(2)), na_values={("B", "Z"): "inf"}
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(data), header=list(range(2)), na_values={("B", "Z"): "inf"}
+    )
+
+    expected = DataFrame(
+        {
+            ("A", "X"): [1],
+            ("B", "Y"): [2],
+            ("B", "Z"): [np.nan],
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # Failed: DID NOT RAISE <class 'ValueError'>; it casts the NaN to False
+def test_bool_and_nan_to_bool(all_parsers):
+    # GH#42808
+    parser = all_parsers
+    data = """0
+NaN
+True
+False
+"""
+    with pytest.raises(ValueError, match="NA values"):
+        parser.read_csv(StringIO(data), dtype="bool")
+
+
+def test_bool_and_nan_to_int(all_parsers):
+    # GH#42808
+    parser = all_parsers
+    data = """0
+NaN
+True
+False
+"""
+    with pytest.raises(ValueError, match="convert|NoneType"):
+        parser.read_csv(StringIO(data), dtype="int")
+
+
+def test_bool_and_nan_to_float(all_parsers):
+    # GH#42808
+    parser = all_parsers
+    data = """0
+NaN
+True
+False
+"""
+    result = parser.read_csv(StringIO(data), dtype="float")
+    expected = DataFrame.from_dict({"0": [np.nan, 1.0, 0.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+@pytest.mark.parametrize(
+    "na_values",
+    [[-99.0, -99], [-99, -99.0]],
+)
+def test_na_values_dict_without_dtype(all_parsers, na_values):
+    parser = all_parsers
+    data = """A
+-99
+-99
+-99.0
+-99.0"""
+
+    result = parser.read_csv(StringIO(data), na_values=na_values)
+    expected = DataFrame({"A": [np.nan, np.nan, np.nan, np.nan]})
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_network.py b/pandas/pandas/tests/io/parser/test_network.py
new file mode 100644
index 0000000000000..836ac71d8e865
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_network.py
@@ -0,0 +1,329 @@
+"""
+Tests parsers ability to read and parse non-local files
+and hence require a network connection to be read.
+"""
+
+from io import BytesIO
+import logging
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+from pandas.io.feather_format import read_feather
+from pandas.io.parsers import read_csv
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+@pytest.mark.parametrize("mode", ["explicit", "infer"])
+@pytest.mark.parametrize("engine", ["python", "c"])
+def test_compressed_urls(
+    httpserver,
+    datapath,
+    salaries_table,
+    mode,
+    engine,
+    compression_only,
+    compression_to_extension,
+):
+    # test reading compressed urls with various engines and
+    # extension inference
+    if compression_only == "tar":
+        pytest.skip("TODO: Add tar salaries.csv to pandas/io/parsers/data")
+
+    extension = compression_to_extension[compression_only]
+    with open(datapath("io", "parser", "data", "salaries.csv" + extension), "rb") as f:
+        httpserver.serve_content(content=f.read())
+
+    url = httpserver.url + "/salaries.csv" + extension
+
+    if mode != "explicit":
+        compression_only = mode
+
+    url_table = read_csv(url, sep="\t", compression=compression_only, engine=engine)
+    tm.assert_frame_equal(url_table, salaries_table)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_url_encoding_csv(httpserver, datapath):
+    """
+    read_csv should honor the requested encoding for URLs.
+
+    GH 10424
+    """
+    with open(datapath("io", "parser", "data", "unicode_series.csv"), "rb") as f:
+        httpserver.serve_content(content=f.read())
+        df = read_csv(httpserver.url, encoding="latin-1", header=None)
+    assert df.loc[15, 1] == "Á köldum klaka (Cold Fever) (1994)"
+
+
+@pytest.fixture
+def tips_df(datapath):
+    """DataFrame with the tips dataset."""
+    return read_csv(datapath("io", "data", "csv", "tips.csv"))
+
+
+@pytest.mark.single_cpu
+@pytest.mark.network
+@pytest.mark.usefixtures("s3_resource")
+@td.skip_if_not_us_locale()
+class TestS3:
+    def test_parse_public_s3_bucket(self, s3_public_bucket_with_data, tips_df, s3so):
+        # more of an integration test due to the not-public contents portion
+        # can probably mock this though.
+        pytest.importorskip("s3fs")
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            df = read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                compression=comp,
+                storage_options=s3so,
+            )
+            assert isinstance(df, DataFrame)
+            assert not df.empty
+            tm.assert_frame_equal(df, tips_df)
+
+    def test_parse_private_s3_bucket(self, s3_private_bucket_with_data, tips_df, s3so):
+        # Read public file from bucket with not-public contents
+        pytest.importorskip("s3fs")
+        df = read_csv(
+            f"s3://{s3_private_bucket_with_data.name}/tips.csv", storage_options=s3so
+        )
+        assert isinstance(df, DataFrame)
+        assert not df.empty
+        tm.assert_frame_equal(df, tips_df)
+
+    def test_parse_public_s3n_bucket(self, s3_public_bucket_with_data, tips_df, s3so):
+        # Read from AWS s3 as "s3n" URL
+        df = read_csv(
+            f"s3n://{s3_public_bucket_with_data.name}/tips.csv",
+            nrows=10,
+            storage_options=s3so,
+        )
+        assert isinstance(df, DataFrame)
+        assert not df.empty
+        tm.assert_frame_equal(tips_df.iloc[:10], df)
+
+    def test_parse_public_s3a_bucket(self, s3_public_bucket_with_data, tips_df, s3so):
+        # Read from AWS s3 as "s3a" URL
+        df = read_csv(
+            f"s3a://{s3_public_bucket_with_data.name}/tips.csv",
+            nrows=10,
+            storage_options=s3so,
+        )
+        assert isinstance(df, DataFrame)
+        assert not df.empty
+        tm.assert_frame_equal(tips_df.iloc[:10], df)
+
+    def test_parse_public_s3_bucket_nrows(
+        self, s3_public_bucket_with_data, tips_df, s3so
+    ):
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            df = read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                nrows=10,
+                compression=comp,
+                storage_options=s3so,
+            )
+            assert isinstance(df, DataFrame)
+            assert not df.empty
+            tm.assert_frame_equal(tips_df.iloc[:10], df)
+
+    def test_parse_public_s3_bucket_chunked(
+        self, s3_public_bucket_with_data, tips_df, s3so
+    ):
+        # Read with a chunksize
+        chunksize = 5
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            with read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                chunksize=chunksize,
+                compression=comp,
+                storage_options=s3so,
+            ) as df_reader:
+                assert df_reader.chunksize == chunksize
+                for i_chunk in [0, 1, 2]:
+                    # Read a couple of chunks and make sure we see them
+                    # properly.
+                    df = df_reader.get_chunk()
+                    assert isinstance(df, DataFrame)
+                    assert not df.empty
+                    true_df = tips_df.iloc[
+                        chunksize * i_chunk : chunksize * (i_chunk + 1)
+                    ]
+                    tm.assert_frame_equal(true_df, df)
+
+    def test_parse_public_s3_bucket_chunked_python(
+        self, s3_public_bucket_with_data, tips_df, s3so
+    ):
+        # Read with a chunksize using the Python parser
+        chunksize = 5
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            with read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                chunksize=chunksize,
+                compression=comp,
+                engine="python",
+                storage_options=s3so,
+            ) as df_reader:
+                assert df_reader.chunksize == chunksize
+                for i_chunk in [0, 1, 2]:
+                    # Read a couple of chunks and make sure we see them properly.
+                    df = df_reader.get_chunk()
+                    assert isinstance(df, DataFrame)
+                    assert not df.empty
+                    true_df = tips_df.iloc[
+                        chunksize * i_chunk : chunksize * (i_chunk + 1)
+                    ]
+                    tm.assert_frame_equal(true_df, df)
+
+    def test_parse_public_s3_bucket_python(
+        self, s3_public_bucket_with_data, tips_df, s3so
+    ):
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            df = read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                engine="python",
+                compression=comp,
+                storage_options=s3so,
+            )
+            assert isinstance(df, DataFrame)
+            assert not df.empty
+            tm.assert_frame_equal(df, tips_df)
+
+    def test_infer_s3_compression(self, s3_public_bucket_with_data, tips_df, s3so):
+        for ext in ["", ".gz", ".bz2"]:
+            df = read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                engine="python",
+                compression="infer",
+                storage_options=s3so,
+            )
+            assert isinstance(df, DataFrame)
+            assert not df.empty
+            tm.assert_frame_equal(df, tips_df)
+
+    def test_parse_public_s3_bucket_nrows_python(
+        self, s3_public_bucket_with_data, tips_df, s3so
+    ):
+        for ext, comp in [("", None), (".gz", "gzip"), (".bz2", "bz2")]:
+            df = read_csv(
+                f"s3://{s3_public_bucket_with_data.name}/tips.csv" + ext,
+                engine="python",
+                nrows=10,
+                compression=comp,
+                storage_options=s3so,
+            )
+            assert isinstance(df, DataFrame)
+            assert not df.empty
+            tm.assert_frame_equal(tips_df.iloc[:10], df)
+
+    def test_read_s3_fails(self, s3so):
+        msg = "The specified bucket does not exist"
+        with pytest.raises(OSError, match=msg):
+            read_csv("s3://nyqpug/asdf.csv", storage_options=s3so)
+
+    def test_read_s3_fails_private(self, s3_private_bucket, s3so):
+        msg = "The specified bucket does not exist"
+        # Receive a permission error when trying to read a private bucket.
+        # It's irrelevant here that this isn't actually a table.
+        with pytest.raises(OSError, match=msg):
+            read_csv(f"s3://{s3_private_bucket.name}/file.csv")
+
+    @pytest.mark.xfail(reason="GH#39155 s3fs upgrade", strict=False)
+    def test_write_s3_csv_fails(self, tips_df, s3so):
+        # GH 32486
+        # Attempting to write to an invalid S3 path should raise
+        import botocore
+
+        # GH 34087
+        # https://boto3.amazonaws.com/v1/documentation/api/latest/guide/error-handling.html
+        # Catch a ClientError since AWS Service Errors are defined dynamically
+        error = (FileNotFoundError, botocore.exceptions.ClientError)
+
+        with pytest.raises(error, match="The specified bucket does not exist"):
+            tips_df.to_csv(
+                "s3://an_s3_bucket_data_doesnt_exit/not_real.csv", storage_options=s3so
+            )
+
+    @pytest.mark.xfail(reason="GH#39155 s3fs upgrade", strict=False)
+    def test_write_s3_parquet_fails(self, tips_df, s3so):
+        # GH 27679
+        # Attempting to write to an invalid S3 path should raise
+        pytest.importorskip("pyarrow")
+        import botocore
+
+        # GH 34087
+        # https://boto3.amazonaws.com/v1/documentation/api/latest/guide/error-handling.html
+        # Catch a ClientError since AWS Service Errors are defined dynamically
+        error = (FileNotFoundError, botocore.exceptions.ClientError)
+
+        with pytest.raises(error, match="The specified bucket does not exist"):
+            tips_df.to_parquet(
+                "s3://an_s3_bucket_data_doesnt_exit/not_real.parquet",
+                storage_options=s3so,
+            )
+
+    @pytest.mark.single_cpu
+    def test_read_csv_handles_boto_s3_object(
+        self, s3_public_bucket_with_data, tips_file
+    ):
+        # see gh-16135
+
+        s3_object = s3_public_bucket_with_data.Object("tips.csv")
+
+        with BytesIO(s3_object.get()["Body"].read()) as buffer:
+            result = read_csv(buffer, encoding="utf8")
+        assert isinstance(result, DataFrame)
+        assert not result.empty
+
+        expected = read_csv(tips_file)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.single_cpu
+    def test_read_csv_chunked_download(self, s3_public_bucket, caplog, s3so):
+        # 8 MB, S3FS uses 5MB chunks
+        df = DataFrame(np.zeros((100000, 4)), columns=list("abcd"))
+        with BytesIO(df.to_csv().encode("utf-8")) as buf:
+            s3_public_bucket.put_object(Key="large-file.csv", Body=buf)
+            uri = f"{s3_public_bucket.name}/large-file.csv"
+            match_re = re.compile(rf"^Fetch: {uri}, 0-(?P<stop>\d+)$")
+            with caplog.at_level(logging.DEBUG, logger="s3fs"):
+                read_csv(
+                    f"s3://{uri}",
+                    nrows=5,
+                    storage_options=s3so,
+                )
+                for log in caplog.messages:
+                    if match := re.match(match_re, log):
+                        # Less than 8 MB
+                        assert int(match.group("stop")) < 8000000
+
+    def test_read_s3_with_hash_in_key(self, s3_public_bucket_with_data, tips_df, s3so):
+        # GH 25945
+        result = read_csv(
+            f"s3://{s3_public_bucket_with_data.name}/tips#1.csv", storage_options=s3so
+        )
+        tm.assert_frame_equal(tips_df, result)
+
+    def test_read_feather_s3_file_path(
+        self, s3_public_bucket_with_data, feather_file, s3so
+    ):
+        # GH 29055
+        pytest.importorskip("pyarrow")
+        expected = read_feather(feather_file)
+        res = read_feather(
+            f"s3://{s3_public_bucket_with_data.name}/simple_dataset.feather",
+            storage_options=s3so,
+        )
+        tm.assert_frame_equal(expected, res)
diff --git a/pandas/pandas/tests/io/parser/test_parse_dates.py b/pandas/pandas/tests/io/parser/test_parse_dates.py
new file mode 100644
index 0000000000000..9a15d9bc84a2e
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_parse_dates.py
@@ -0,0 +1,859 @@
+"""
+Tests date parsing functionality for all of the
+parsers defined in parsers.py
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+from io import StringIO
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import date_range
+from pandas.core.tools.datetimes import start_caching_at
+
+from pandas.io.parsers import read_csv
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+def test_date_col_as_index_col(all_parsers):
+    data = """\
+KORD,19990127 19:00:00, 18:56:00, 0.8100, 2.8100, 7.2000, 0.0000, 280.0000
+KORD,19990127 20:00:00, 19:56:00, 0.0100, 2.2100, 7.2000, 0.0000, 260.0000
+KORD,19990127 21:00:00, 20:56:00, -0.5900, 2.2100, 5.7000, 0.0000, 280.0000
+KORD,19990127 21:00:00, 21:18:00, -0.9900, 2.0100, 3.6000, 0.0000, 270.0000
+KORD,19990127 22:00:00, 21:56:00, -0.5900, 1.7100, 5.1000, 0.0000, 290.0000
+"""
+    parser = all_parsers
+    kwds = {
+        "header": None,
+        "parse_dates": [1],
+        "index_col": 1,
+        "names": ["X0", "X1", "X2", "X3", "X4", "X5", "X6", "X7"],
+    }
+    result = parser.read_csv(StringIO(data), **kwds)
+
+    index = Index(
+        [
+            datetime(1999, 1, 27, 19, 0),
+            datetime(1999, 1, 27, 20, 0),
+            datetime(1999, 1, 27, 21, 0),
+            datetime(1999, 1, 27, 21, 0),
+            datetime(1999, 1, 27, 22, 0),
+        ],
+        dtype="M8[s]",
+        name="X1",
+    )
+    expected = DataFrame(
+        [
+            ["KORD", " 18:56:00", 0.81, 2.81, 7.2, 0.0, 280.0],
+            ["KORD", " 19:56:00", 0.01, 2.21, 7.2, 0.0, 260.0],
+            ["KORD", " 20:56:00", -0.59, 2.21, 5.7, 0.0, 280.0],
+            ["KORD", " 21:18:00", -0.99, 2.01, 3.6, 0.0, 270.0],
+            ["KORD", " 21:56:00", -0.59, 1.71, 5.1, 0.0, 290.0],
+        ],
+        columns=["X0", "X2", "X3", "X4", "X5", "X6", "X7"],
+        index=index,
+    )
+    if parser.engine == "pyarrow":
+        # https://github.com/pandas-dev/pandas/issues/44231
+        # pyarrow 6.0 starts to infer time type
+        expected["X2"] = pd.to_datetime("1970-01-01" + expected["X2"]).dt.time
+
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+def test_nat_parse(all_parsers):
+    # see gh-3062
+    parser = all_parsers
+    df = DataFrame(
+        {
+            "A": np.arange(10, dtype="float64"),
+            "B": Timestamp("20010101"),
+        }
+    )
+    df.iloc[3:6, :] = np.nan
+
+    with tm.ensure_clean("__nat_parse_.csv") as path:
+        df.to_csv(path)
+
+        result = parser.read_csv(path, index_col=0, parse_dates=["B"])
+        tm.assert_frame_equal(result, df)
+
+
+@skip_pyarrow
+def test_parse_dates_implicit_first_col(all_parsers):
+    data = """A,B,C
+20090101,a,1,2
+20090102,b,3,4
+20090103,c,4,5
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), parse_dates=True)
+
+    expected = parser.read_csv(StringIO(data), index_col=0, parse_dates=True)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+def test_parse_dates_string(all_parsers):
+    data = """date,A,B,C
+20090101,a,1,2
+20090102,b,3,4
+20090103,c,4,5
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), index_col="date", parse_dates=["date"])
+    # freq doesn't round-trip
+    index = date_range("1/1/2009", periods=3, name="date", unit="s")._with_freq(None)
+
+    expected = DataFrame(
+        {"A": ["a", "b", "c"], "B": [1, 3, 4], "C": [2, 4, 5]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+@pytest.mark.parametrize("parse_dates", [[0, 2], ["a", "c"]])
+def test_parse_dates_column_list(all_parsers, parse_dates):
+    data = "a,b,c\n01/01/2010,1,15/02/2010"
+    parser = all_parsers
+
+    expected = DataFrame(
+        {"a": [datetime(2010, 1, 1)], "b": [1], "c": [datetime(2010, 2, 15)]}
+    )
+    expected["a"] = expected["a"].astype("M8[s]")
+    expected["c"] = expected["c"].astype("M8[s]")
+    expected = expected.set_index(["a", "b"])
+
+    result = parser.read_csv(
+        StringIO(data), index_col=[0, 1], parse_dates=parse_dates, dayfirst=True
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+@pytest.mark.parametrize("index_col", [[0, 1], [1, 0]])
+def test_multi_index_parse_dates(all_parsers, index_col):
+    data = """index1,index2,A,B,C
+20090101,one,a,1,2
+20090101,two,b,3,4
+20090101,three,c,4,5
+20090102,one,a,1,2
+20090102,two,b,3,4
+20090102,three,c,4,5
+20090103,one,a,1,2
+20090103,two,b,3,4
+20090103,three,c,4,5
+"""
+    parser = all_parsers
+    dti = date_range("2009-01-01", periods=3, freq="D", unit="s")
+    index = MultiIndex.from_product(
+        [
+            dti,
+            ("one", "two", "three"),
+        ],
+        names=["index1", "index2"],
+    )
+
+    # Out of order.
+    if index_col == [1, 0]:
+        index = index.swaplevel(0, 1)
+
+    expected = DataFrame(
+        [
+            ["a", 1, 2],
+            ["b", 3, 4],
+            ["c", 4, 5],
+            ["a", 1, 2],
+            ["b", 3, 4],
+            ["c", 4, 5],
+            ["a", 1, 2],
+            ["b", 3, 4],
+            ["c", 4, 5],
+        ],
+        columns=["A", "B", "C"],
+        index=index,
+    )
+    result = parser.read_csv_check_warnings(
+        UserWarning,
+        "Could not infer format",
+        StringIO(data),
+        index_col=index_col,
+        parse_dates=True,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_tz_aware(all_parsers):
+    # See gh-1693
+    parser = all_parsers
+    data = "Date,x\n2012-06-13T01:39:00Z,0.5"
+
+    result = parser.read_csv(StringIO(data), index_col=0, parse_dates=True)
+    expected = DataFrame(
+        {"x": [0.5]}, index=Index([Timestamp("2012-06-13 01:39:00+00:00")], name="Date")
+    )
+    if parser.engine == "pyarrow":
+        pytz = pytest.importorskip("pytz")
+        expected_tz = pytz.utc
+    else:
+        expected_tz = timezone.utc
+    tm.assert_frame_equal(result, expected)
+    assert result.index.tz is expected_tz
+
+
+@pytest.mark.parametrize("kwargs", [{}, {"index_col": "C"}])
+def test_read_with_parse_dates_scalar_non_bool(all_parsers, kwargs):
+    # see gh-5636
+    parser = all_parsers
+    msg = "Only booleans and lists are accepted for the 'parse_dates' parameter"
+    data = """A,B,C
+    1,2,2003-11-1"""
+
+    with pytest.raises(TypeError, match=msg):
+        parser.read_csv(StringIO(data), parse_dates="C", **kwargs)
+
+
+@pytest.mark.parametrize("parse_dates", [(1,), np.array([4, 5]), {1, 3}])
+def test_read_with_parse_dates_invalid_type(all_parsers, parse_dates):
+    parser = all_parsers
+    msg = "Only booleans and lists are accepted for the 'parse_dates' parameter"
+    data = """A,B,C
+    1,2,2003-11-1"""
+
+    with pytest.raises(TypeError, match=msg):
+        parser.read_csv(StringIO(data), parse_dates=parse_dates)
+
+
+@pytest.mark.parametrize("value", ["nan", ""])
+def test_bad_date_parse(all_parsers, cache, value):
+    # if we have an invalid date make sure that we handle this with
+    # and w/o the cache properly
+    parser = all_parsers
+    s = StringIO((f"{value},\n") * (start_caching_at + 1))
+
+    parser.read_csv(
+        s,
+        header=None,
+        names=["foo", "bar"],
+        parse_dates=["foo"],
+        cache_dates=cache,
+    )
+
+
+@pytest.mark.parametrize("value", ["0"])
+def test_bad_date_parse_with_warning(all_parsers, cache, value):
+    # if we have an invalid date make sure that we handle this with
+    # and w/o the cache properly.
+    parser = all_parsers
+    s = StringIO((f"{value},\n") * 50000)
+
+    if parser.engine == "pyarrow":
+        # pyarrow reads "0" as 0 (of type int64), and so
+        # pandas doesn't try to guess the datetime format
+        # TODO: parse dates directly in pyarrow, see
+        # https://github.com/pandas-dev/pandas/issues/48017
+        warn = None
+    elif cache:
+        # Note: warning is not raised if 'cache_dates', because here there is only a
+        # single unique date and hence no risk of inconsistent parsing.
+        warn = None
+    else:
+        warn = UserWarning
+    parser.read_csv_check_warnings(
+        warn,
+        "Could not infer format",
+        s,
+        header=None,
+        names=["foo", "bar"],
+        parse_dates=["foo"],
+        cache_dates=cache,
+        raise_on_extra_warnings=False,
+    )
+
+
+def test_parse_dates_empty_string(all_parsers):
+    # see gh-2263
+    parser = all_parsers
+    data = "Date,test\n2012-01-01,1\n,2"
+    result = parser.read_csv(StringIO(data), parse_dates=["Date"], na_filter=False)
+
+    expected = DataFrame(
+        [[datetime(2012, 1, 1), 1], [pd.NaT, 2]], columns=["Date", "test"]
+    )
+    expected["Date"] = expected["Date"].astype("M8[s]")
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        (
+            "a\n04.15.2016",
+            {"parse_dates": ["a"]},
+            DataFrame([datetime(2016, 4, 15)], columns=["a"], dtype="M8[s]"),
+        ),
+        (
+            "a\n04.15.2016",
+            {"parse_dates": True, "index_col": 0},
+            DataFrame(
+                index=DatetimeIndex(["2016-04-15"], dtype="M8[s]", name="a"), columns=[]
+            ),
+        ),
+        (
+            "a,b\n04.15.2016,09.16.2013",
+            {"parse_dates": ["a", "b"]},
+            DataFrame(
+                [[datetime(2016, 4, 15), datetime(2013, 9, 16)]],
+                dtype="M8[s]",
+                columns=["a", "b"],
+            ),
+        ),
+        (
+            "a,b\n04.15.2016,09.16.2013",
+            {"parse_dates": True, "index_col": [0, 1]},
+            DataFrame(
+                index=MultiIndex.from_tuples(
+                    [
+                        (
+                            Timestamp(2016, 4, 15).as_unit("s"),
+                            Timestamp(2013, 9, 16).as_unit("s"),
+                        )
+                    ],
+                    names=["a", "b"],
+                ),
+                columns=[],
+            ),
+        ),
+    ],
+)
+def test_parse_dates_no_convert_thousands(all_parsers, data, kwargs, expected):
+    # see gh-14066
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data), thousands=".", **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_date_column_with_empty_string(all_parsers):
+    # see gh-6428
+    parser = all_parsers
+    data = "case,opdate\n7,10/18/2006\n7,10/18/2008\n621, "
+    result = parser.read_csv(StringIO(data), parse_dates=["opdate"])
+
+    expected_data = [[7, "10/18/2006"], [7, "10/18/2008"], [621, " "]]
+    expected = DataFrame(expected_data, columns=["case", "opdate"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        (
+            "a\n135217135789158401\n1352171357E+5",
+            [135217135789158401, 135217135700000],
+        ),
+        (
+            "a\n99999999999\n123456789012345\n1234E+0",
+            [99999999999, 123456789012345, 1234],
+        ),
+    ],
+)
+@pytest.mark.parametrize("parse_dates", [True, False])
+def test_parse_date_float(all_parsers, data, expected, parse_dates):
+    # see gh-2697
+    #
+    # Date parsing should fail, so we leave the data untouched
+    # (i.e. float precision should remain unchanged).
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data), parse_dates=parse_dates)
+    expected = DataFrame({"a": expected}, dtype="float64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_timezone(all_parsers):
+    # see gh-22256
+    parser = all_parsers
+    data = """dt,val
+              2018-01-04 09:01:00+09:00,23350
+              2018-01-04 09:02:00+09:00,23400
+              2018-01-04 09:03:00+09:00,23400
+              2018-01-04 09:04:00+09:00,23400
+              2018-01-04 09:05:00+09:00,23400"""
+    result = parser.read_csv(StringIO(data), parse_dates=["dt"])
+
+    dti = date_range(
+        start="2018-01-04 09:01:00",
+        end="2018-01-04 09:05:00",
+        freq="1min",
+        tz=timezone(timedelta(minutes=540)),
+        unit="s",
+    )._with_freq(None)
+    expected_data = {"dt": dti, "val": [23350, 23400, 23400, 23400, 23400]}
+
+    expected = DataFrame(expected_data)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # pandas.errors.ParserError: CSV parse error
+@pytest.mark.parametrize(
+    "date_string",
+    ["32/32/2019", "02/30/2019", "13/13/2019", "13/2019", "a3/11/2018", "10/11/2o17"],
+)
+def test_invalid_parse_delimited_date(all_parsers, date_string):
+    parser = all_parsers
+    expected = DataFrame({0: [date_string]}, dtype="str")
+    result = parser.read_csv(
+        StringIO(date_string),
+        header=None,
+        parse_dates=[0],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "date_string,dayfirst,expected",
+    [
+        # %d/%m/%Y; month > 12 thus replacement
+        ("13/02/2019", True, datetime(2019, 2, 13)),
+        # %m/%d/%Y; day > 12 thus there will be no replacement
+        ("02/13/2019", False, datetime(2019, 2, 13)),
+        # %d/%m/%Y; dayfirst==True thus replacement
+        ("04/02/2019", True, datetime(2019, 2, 4)),
+    ],
+)
+def test_parse_delimited_date_swap_no_warning(
+    all_parsers, date_string, dayfirst, expected, request
+):
+    parser = all_parsers
+    expected = DataFrame({0: [expected]}, dtype="datetime64[s]")
+    if parser.engine == "pyarrow":
+        if not dayfirst:
+            # "CSV parse error: Empty CSV file or block"
+            pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+        msg = "The 'dayfirst' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(date_string), header=None, dayfirst=dayfirst, parse_dates=[0]
+            )
+        return
+
+    result = parser.read_csv(
+        StringIO(date_string), header=None, dayfirst=dayfirst, parse_dates=[0]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# ArrowInvalid: CSV parse error: Empty CSV file or block: cannot infer number of columns
+@skip_pyarrow
+@pytest.mark.parametrize(
+    "date_string,dayfirst,expected",
+    [
+        # %d/%m/%Y; month > 12
+        ("13/02/2019", False, datetime(2019, 2, 13)),
+        # %m/%d/%Y; day > 12
+        ("02/13/2019", True, datetime(2019, 2, 13)),
+    ],
+)
+def test_parse_delimited_date_swap_with_warning(
+    all_parsers, date_string, dayfirst, expected
+):
+    parser = all_parsers
+    expected = DataFrame({0: [expected]}, dtype="datetime64[s]")
+    warning_msg = (
+        "Parsing dates in .* format when dayfirst=.* was specified. "
+        "Pass `dayfirst=.*` or specify a format to silence this warning."
+    )
+    result = parser.read_csv_check_warnings(
+        UserWarning,
+        warning_msg,
+        StringIO(date_string),
+        header=None,
+        dayfirst=dayfirst,
+        parse_dates=[0],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_multiple_delimited_dates_with_swap_warnings():
+    # GH46210
+    with pytest.raises(
+        ValueError,
+        match=(
+            r'^time data "31/05/2000" doesn\'t match format "%m/%d/%Y". '
+            r"You might want to try:"
+        ),
+    ):
+        pd.to_datetime(["01/01/2000", "31/05/2000", "31/05/2001", "01/02/2000"])
+
+
+# ArrowKeyError: Column 'fdate1' in include_columns does not exist in CSV file
+@skip_pyarrow
+@pytest.mark.parametrize(
+    "names, usecols, parse_dates, missing_cols",
+    [
+        (None, ["val"], ["date", "time"], "date, time"),
+        (None, ["val"], [0, "time"], "time"),
+        (["date1", "time1", "temperature"], None, ["date", "time"], "date, time"),
+        (
+            ["date1", "time1", "temperature"],
+            ["date1", "temperature"],
+            ["date1", "time"],
+            "time",
+        ),
+    ],
+)
+def test_missing_parse_dates_column_raises(
+    all_parsers, names, usecols, parse_dates, missing_cols
+):
+    # gh-31251 column names provided in parse_dates could be missing.
+    parser = all_parsers
+    content = StringIO("date,time,val\n2020-01-31,04:20:32,32\n")
+    msg = f"Missing column provided to 'parse_dates': '{missing_cols}'"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(
+            content, sep=",", names=names, usecols=usecols, parse_dates=parse_dates
+        )
+
+
+@xfail_pyarrow  # mismatched shape
+def test_date_parser_and_names(all_parsers):
+    # GH#33699
+    parser = all_parsers
+    data = StringIO("""x,y\n1,2""")
+    warn = UserWarning
+    if parser.engine == "pyarrow":
+        # DeprecationWarning for passing a Manager object
+        warn = (UserWarning, DeprecationWarning)
+    result = parser.read_csv_check_warnings(
+        warn,
+        "Could not infer format",
+        data,
+        parse_dates=["B"],
+        names=["B"],
+    )
+    expected = DataFrame({"B": ["y", "2"]}, index=["x", "1"])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # TypeError: an integer is required
+def test_date_parser_multiindex_columns(all_parsers):
+    parser = all_parsers
+    data = """a,b
+1,2
+2019-12-31,6"""
+    result = parser.read_csv(StringIO(data), parse_dates=[("a", "1")], header=[0, 1])
+    expected = DataFrame({("a", "1"): Timestamp("2019-12-31"), ("b", "2"): [6]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_date_parser_usecols_thousands(all_parsers):
+    # GH#39365
+    data = """A,B,C
+    1,3,20-09-01-01
+    2,4,20-09-01-01
+    """
+
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        # DeprecationWarning for passing a Manager object
+        msg = "The 'thousands' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                parse_dates=[1],
+                usecols=[1, 2],
+                thousands="-",
+            )
+        return
+
+    result = parser.read_csv_check_warnings(
+        UserWarning,
+        "Could not infer format",
+        StringIO(data),
+        parse_dates=[1],
+        usecols=[1, 2],
+        thousands="-",
+    )
+    expected = DataFrame({"B": [3, 4], "C": [Timestamp("20-09-2001 01:00:00")] * 2})
+    expected["C"] = expected["C"].astype("M8[s]")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dayfirst_warnings():
+    # GH 12585
+
+    # CASE 1: valid input
+    input = "date\n31/12/2014\n10/03/2011"
+    expected = DatetimeIndex(
+        ["2014-12-31", "2011-03-10"], dtype="datetime64[s]", freq=None, name="date"
+    )
+    warning_msg = (
+        "Parsing dates in .* format when dayfirst=.* was specified. "
+        "Pass `dayfirst=.*` or specify a format to silence this warning."
+    )
+
+    # A. dayfirst arg correct, no warning
+    res1 = read_csv(
+        StringIO(input), parse_dates=["date"], dayfirst=True, index_col="date"
+    ).index
+    tm.assert_index_equal(expected, res1)
+
+    # B. dayfirst arg incorrect, warning
+    with tm.assert_produces_warning(UserWarning, match=warning_msg):
+        res2 = read_csv(
+            StringIO(input), parse_dates=["date"], dayfirst=False, index_col="date"
+        ).index
+    tm.assert_index_equal(expected, res2)
+
+    # CASE 2: invalid input
+    # cannot consistently process with single format
+    # return to user unaltered
+
+    # first in DD/MM/YYYY, second in MM/DD/YYYY
+    input = "date\n31/12/2014\n03/30/2011"
+    expected = Index(["31/12/2014", "03/30/2011"], dtype="str", name="date")
+
+    # A. use dayfirst=True
+    res5 = read_csv(
+        StringIO(input), parse_dates=["date"], dayfirst=True, index_col="date"
+    ).index
+    tm.assert_index_equal(expected, res5)
+
+    # B. use dayfirst=False
+    with tm.assert_produces_warning(UserWarning, match=warning_msg):
+        res6 = read_csv(
+            StringIO(input), parse_dates=["date"], dayfirst=False, index_col="date"
+        ).index
+    tm.assert_index_equal(expected, res6)
+
+
+@pytest.mark.parametrize(
+    "date_string, dayfirst",
+    [
+        pytest.param(
+            "31/1/2014",
+            False,
+            id="second date is single-digit",
+        ),
+        pytest.param(
+            "1/31/2014",
+            True,
+            id="first date is single-digit",
+        ),
+    ],
+)
+def test_dayfirst_warnings_no_leading_zero(date_string, dayfirst):
+    # GH47880
+    initial_value = f"date\n{date_string}"
+    expected = DatetimeIndex(
+        ["2014-01-31"], dtype="datetime64[s]", freq=None, name="date"
+    )
+    warning_msg = (
+        "Parsing dates in .* format when dayfirst=.* was specified. "
+        "Pass `dayfirst=.*` or specify a format to silence this warning."
+    )
+    with tm.assert_produces_warning(UserWarning, match=warning_msg):
+        res = read_csv(
+            StringIO(initial_value),
+            parse_dates=["date"],
+            index_col="date",
+            dayfirst=dayfirst,
+        ).index
+    tm.assert_index_equal(expected, res)
+
+
+@skip_pyarrow  # CSV parse error: Expected 3 columns, got 4
+def test_infer_first_column_as_index(all_parsers):
+    # GH#11019
+    parser = all_parsers
+    data = "a,b,c\n1970-01-01,2,3,4"
+    result = parser.read_csv(
+        StringIO(data),
+        parse_dates=["a"],
+    )
+    expected = DataFrame({"a": "2", "b": 3, "c": 4}, index=["1970-01-01"])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # pyarrow engine doesn't support passing a dict for na_values
+def test_replace_nans_before_parsing_dates(all_parsers):
+    # GH#26203
+    parser = all_parsers
+    data = """Test
+2012-10-01
+0
+2015-05-15
+#
+2017-09-09
+"""
+    result = parser.read_csv(
+        StringIO(data),
+        na_values={"Test": ["#", "0"]},
+        parse_dates=["Test"],
+        date_format="%Y-%m-%d",
+    )
+    expected = DataFrame(
+        {
+            "Test": [
+                Timestamp("2012-10-01"),
+                pd.NaT,
+                Timestamp("2015-05-15"),
+                pd.NaT,
+                Timestamp("2017-09-09"),
+            ]
+        },
+        dtype="M8[s]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # string[python] instead of dt64[ns]
+def test_parse_dates_and_string_dtype(all_parsers):
+    # GH#34066
+    parser = all_parsers
+    data = """a,b
+1,2019-12-31
+"""
+    result = parser.read_csv(StringIO(data), dtype="string", parse_dates=["b"])
+    expected = DataFrame({"a": ["1"], "b": [Timestamp("2019-12-31")]})
+    expected["a"] = expected["a"].astype("string")
+    expected["b"] = expected["b"].astype("M8[s]")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_dot_separated_dates(all_parsers):
+    # https://github.com/pandas-dev/pandas/issues/2586
+    parser = all_parsers
+    data = """a,b
+27.03.2003 14:55:00.000,1
+03.08.2003 15:20:00.000,2"""
+    if parser.engine == "pyarrow":
+        expected_index = Index(
+            ["27.03.2003 14:55:00.000", "03.08.2003 15:20:00.000"],
+            dtype="str",
+            name="a",
+        )
+        warn = None
+    else:
+        expected_index = DatetimeIndex(
+            ["2003-03-27 14:55:00", "2003-08-03 15:20:00"],
+            dtype="datetime64[ms]",
+            name="a",
+        )
+        warn = UserWarning
+    msg = r"when dayfirst=False \(the default\) was specified"
+    result = parser.read_csv_check_warnings(
+        warn,
+        msg,
+        StringIO(data),
+        parse_dates=True,
+        index_col=0,
+        raise_on_extra_warnings=False,
+    )
+    expected = DataFrame({"b": [1, 2]}, index=expected_index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_dates_dict_format(all_parsers):
+    # GH#51240
+    parser = all_parsers
+    data = """a,b
+2019-12-31,31-12-2019
+2020-12-31,31-12-2020"""
+
+    result = parser.read_csv(
+        StringIO(data),
+        date_format={"a": "%Y-%m-%d", "b": "%d-%m-%Y"},
+        parse_dates=["a", "b"],
+    )
+    expected = DataFrame(
+        {
+            "a": [Timestamp("2019-12-31"), Timestamp("2020-12-31")],
+            "b": [Timestamp("2019-12-31"), Timestamp("2020-12-31")],
+        },
+        dtype="M8[s]",
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # object dtype index
+def test_parse_dates_dict_format_index(all_parsers):
+    # GH#51240
+    parser = all_parsers
+    data = """a,b
+2019-12-31,31-12-2019
+2020-12-31,31-12-2020"""
+
+    result = parser.read_csv(
+        StringIO(data), date_format={"a": "%Y-%m-%d"}, parse_dates=True, index_col=0
+    )
+    expected = DataFrame(
+        {
+            "b": ["31-12-2019", "31-12-2020"],
+        },
+        index=Index([Timestamp("2019-12-31"), Timestamp("2020-12-31")], name="a"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_parse_dates_arrow_engine(all_parsers):
+    # GH#53295
+    parser = all_parsers
+    data = """a,b
+2000-01-01 00:00:00,1
+2000-01-01 00:00:01,1"""
+
+    result = parser.read_csv(StringIO(data), parse_dates=["a"])
+    expected = DataFrame(
+        {
+            "a": [
+                Timestamp("2000-01-01 00:00:00"),
+                Timestamp("2000-01-01 00:00:01"),
+            ],
+            "b": 1,
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # object dtype index
+def test_from_csv_with_mixed_offsets(all_parsers):
+    parser = all_parsers
+    data = "a\n2020-01-01T00:00:00+01:00\n2020-01-01T00:00:00+00:00"
+    result = parser.read_csv(StringIO(data), parse_dates=["a"])["a"]
+    expected = Series(
+        [
+            "2020-01-01T00:00:00+01:00",
+            "2020-01-01T00:00:00+00:00",
+        ],
+        name="a",
+        index=[0, 1],
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_python_parser_only.py b/pandas/pandas/tests/io/parser/test_python_parser_only.py
new file mode 100644
index 0000000000000..a5bb151e84f47
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_python_parser_only.py
@@ -0,0 +1,564 @@
+"""
+Tests that apply specifically to the Python parser. Unless specifically
+stated as a Python-specific issue, the goal is to eventually move as many of
+these tests out of this module as soon as the C parser can accept further
+arguments when parsing.
+"""
+
+from __future__ import annotations
+
+import csv
+from io import (
+    BytesIO,
+    StringIO,
+    TextIOWrapper,
+)
+from typing import TYPE_CHECKING
+
+import numpy as np
+import pytest
+
+from pandas.errors import (
+    ParserError,
+    ParserWarning,
+)
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+if TYPE_CHECKING:
+    from collections.abc import Iterator
+
+
+def test_default_separator(python_parser_only):
+    # see gh-17333
+    #
+    # csv.Sniffer in Python treats "o" as separator.
+    data = "aob\n1o2\n3o4"
+    parser = python_parser_only
+    expected = DataFrame({"a": [1, 3], "b": [2, 4]})
+
+    result = parser.read_csv(StringIO(data), sep=None)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("skipfooter", ["foo", 1.5, True])
+def test_invalid_skipfooter_non_int(python_parser_only, skipfooter):
+    # see gh-15925 (comment)
+    data = "a\n1\n2"
+    parser = python_parser_only
+    msg = "skipfooter must be an integer"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), skipfooter=skipfooter)
+
+
+def test_invalid_skipfooter_negative(python_parser_only):
+    # see gh-15925 (comment)
+    data = "a\n1\n2"
+    parser = python_parser_only
+    msg = "skipfooter cannot be negative"
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), skipfooter=-1)
+
+
+@pytest.mark.parametrize("kwargs", [{"sep": None}, {"delimiter": "|"}])
+def test_sniff_delimiter(python_parser_only, kwargs):
+    data = """index|A|B|C
+foo|1|2|3
+bar|4|5|6
+baz|7|8|9
+"""
+    parser = python_parser_only
+    result = parser.read_csv(StringIO(data), index_col=0, **kwargs)
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        columns=["A", "B", "C"],
+        index=Index(["foo", "bar", "baz"], name="index"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sniff_delimiter_comment(python_parser_only):
+    data = """# comment line
+index|A|B|C
+# comment line
+foo|1|2|3 # ignore | this
+bar|4|5|6
+baz|7|8|9
+"""
+    parser = python_parser_only
+    result = parser.read_csv(StringIO(data), index_col=0, sep=None, comment="#")
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        columns=["A", "B", "C"],
+        index=Index(["foo", "bar", "baz"], name="index"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("encoding", [None, "utf-8"])
+def test_sniff_delimiter_encoding(python_parser_only, encoding):
+    parser = python_parser_only
+    data = """ignore this
+ignore this too
+index|A|B|C
+foo|1|2|3
+bar|4|5|6
+baz|7|8|9
+"""
+
+    if encoding is not None:
+        data = data.encode(encoding)
+        data = BytesIO(data)
+        data = TextIOWrapper(data, encoding=encoding)
+    else:
+        data = StringIO(data)
+
+    result = parser.read_csv(data, index_col=0, sep=None, skiprows=2, encoding=encoding)
+    expected = DataFrame(
+        [[1, 2, 3], [4, 5, 6], [7, 8, 9]],
+        columns=["A", "B", "C"],
+        index=Index(["foo", "bar", "baz"], name="index"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_single_line(python_parser_only):
+    # see gh-6607: sniff separator
+    parser = python_parser_only
+    result = parser.read_csv(StringIO("1,2"), names=["a", "b"], header=None, sep=None)
+
+    expected = DataFrame({"a": [1], "b": [2]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kwargs", [{"skipfooter": 2}, {"nrows": 3}])
+def test_skipfooter(python_parser_only, kwargs):
+    # see gh-6607
+    data = """A,B,C
+1,2,3
+4,5,6
+7,8,9
+want to skip this
+also also skip this
+"""
+    parser = python_parser_only
+    result = parser.read_csv(StringIO(data), **kwargs)
+
+    expected = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=["A", "B", "C"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "compression,klass", [("gzip", "GzipFile"), ("bz2", "BZ2File")]
+)
+def test_decompression_regex_sep(python_parser_only, csv1, compression, klass):
+    # see gh-6607
+    parser = python_parser_only
+
+    with open(csv1, "rb") as f:
+        data = f.read()
+
+    data = data.replace(b",", b"::")
+    expected = parser.read_csv(csv1)
+
+    module = pytest.importorskip(compression)
+    klass = getattr(module, klass)
+
+    with tm.ensure_clean() as path:
+        with klass(path, mode="wb") as tmp:
+            tmp.write(data)
+
+        result = parser.read_csv(path, sep="::", compression=compression)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_buglet_4x_multi_index(python_parser_only):
+    # see gh-6607
+    data = """                      A       B       C       D        E
+one two three   four
+a   b   10.0032 5    -0.5109 -2.3358 -0.4645  0.05076  0.3640
+a   q   20      4     0.4473  1.4152  0.2834  1.00661  0.1744
+x   q   30      3    -0.6662 -0.5243 -0.3580  0.89145  2.5838"""
+    parser = python_parser_only
+
+    expected = DataFrame(
+        [
+            [-0.5109, -2.3358, -0.4645, 0.05076, 0.3640],
+            [0.4473, 1.4152, 0.2834, 1.00661, 0.1744],
+            [-0.6662, -0.5243, -0.3580, 0.89145, 2.5838],
+        ],
+        columns=["A", "B", "C", "D", "E"],
+        index=MultiIndex.from_tuples(
+            [("a", "b", 10.0032, 5), ("a", "q", 20, 4), ("x", "q", 30, 3)],
+            names=["one", "two", "three", "four"],
+        ),
+    )
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_csv_buglet_4x_multi_index2(python_parser_only):
+    # see gh-6893
+    data = "      A B C\na b c\n1 3 7 0 3 6\n3 1 4 1 5 9"
+    parser = python_parser_only
+
+    expected = DataFrame.from_records(
+        [(1, 3, 7, 0, 3, 6), (3, 1, 4, 1, 5, 9)],
+        columns=list("abcABC"),
+        index=list("abc"),
+    )
+    result = parser.read_csv(StringIO(data), sep=r"\s+")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("add_footer", [True, False])
+def test_skipfooter_with_decimal(python_parser_only, add_footer):
+    # see gh-6971
+    data = "1#2\n3#4"
+    parser = python_parser_only
+    expected = DataFrame({"a": [1.2, 3.4]})
+
+    if add_footer:
+        # The stray footer line should not mess with the
+        # casting of the first two lines if we skip it.
+        kwargs = {"skipfooter": 1}
+        data += "\nFooter"
+    else:
+        kwargs = {}
+
+    result = parser.read_csv(StringIO(data), names=["a"], decimal="#", **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "sep", ["::", "#####", "!!!", "123", "#1!c5", "%!c!d", "@@#4:2", "_!pd#_"]
+)
+@pytest.mark.parametrize(
+    "encoding", ["utf-16", "utf-16-be", "utf-16-le", "utf-32", "cp037"]
+)
+def test_encoding_non_utf8_multichar_sep(python_parser_only, sep, encoding):
+    # see gh-3404
+    expected = DataFrame({"a": [1], "b": [2]})
+    parser = python_parser_only
+
+    data = "1" + sep + "2"
+    encoded_data = data.encode(encoding)
+
+    result = parser.read_csv(
+        BytesIO(encoded_data), sep=sep, names=["a", "b"], encoding=encoding
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("quoting", [csv.QUOTE_MINIMAL, csv.QUOTE_NONE])
+def test_multi_char_sep_quotes(python_parser_only, quoting):
+    # see gh-13374
+    kwargs = {"sep": ",,"}
+    parser = python_parser_only
+
+    data = 'a,,b\n1,,a\n2,,"2,,b"'
+
+    if quoting == csv.QUOTE_NONE:
+        msg = "Expected 2 fields in line 3, saw 3"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), quoting=quoting, **kwargs)
+    else:
+        msg = "ignored when a multi-char delimiter is used"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), quoting=quoting, **kwargs)
+
+
+def test_none_delimiter(python_parser_only):
+    # see gh-13374 and gh-17465
+    parser = python_parser_only
+    data = "a,b,c\n0,1,2\n3,4,5,6\n7,8,9"
+    expected = DataFrame({"a": [0, 7], "b": [1, 8], "c": [2, 9]})
+
+    # We expect the third line in the data to be
+    # skipped because it is malformed, but we do
+    # not expect any errors to occur.
+    with tm.assert_produces_warning(
+        ParserWarning, match="Skipping line 3", check_stacklevel=False
+    ):
+        result = parser.read_csv(
+            StringIO(data), header=0, sep=None, on_bad_lines="warn"
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("data", ['a\n1\n"b"a', 'a,b,c\ncat,foo,bar\ndog,foo,"baz'])
+@pytest.mark.parametrize("skipfooter", [0, 1])
+def test_skipfooter_bad_row(python_parser_only, data, skipfooter):
+    # see gh-13879 and gh-15910
+    parser = python_parser_only
+    if skipfooter:
+        msg = "parsing errors in the skipped footer rows"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), skipfooter=skipfooter)
+    else:
+        msg = "unexpected end of data|expected after"
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data), skipfooter=skipfooter)
+
+
+def test_malformed_skipfooter(python_parser_only):
+    parser = python_parser_only
+    data = """ignore
+A,B,C
+1,2,3 # comment
+1,2,3,4,5
+2,3,4
+footer
+"""
+    msg = "Expected 3 fields in line 4, saw 5"
+    with pytest.raises(ParserError, match=msg):
+        parser.read_csv(StringIO(data), header=1, comment="#", skipfooter=1)
+
+
+def test_python_engine_file_no_next(python_parser_only):
+    parser = python_parser_only
+
+    class NoNextBuffer:
+        def __init__(self, csv_data) -> None:
+            self.data = csv_data
+
+        def __iter__(self) -> Iterator:
+            return self.data.__iter__()
+
+        def read(self):
+            return self.data
+
+        def readline(self):
+            return self.data
+
+    parser.read_csv(NoNextBuffer("a\n1"))
+
+
+@pytest.mark.parametrize("bad_line_func", [lambda x: ["2", "3"], lambda x: x[:2]])
+def test_on_bad_lines_callable(python_parser_only, bad_line_func):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2
+2,3,4,5,6
+3,4
+"""
+    bad_sio = StringIO(data)
+    result = parser.read_csv(bad_sio, on_bad_lines=bad_line_func)
+    expected = DataFrame({"a": [1, 2, 3], "b": [2, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_on_bad_lines_callable_write_to_external_list(python_parser_only):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2
+2,3,4,5,6
+3,4
+"""
+    bad_sio = StringIO(data)
+    lst = []
+
+    def bad_line_func(bad_line: list[str]) -> list[str]:
+        lst.append(bad_line)
+        return ["2", "3"]
+
+    result = parser.read_csv(bad_sio, on_bad_lines=bad_line_func)
+    expected = DataFrame({"a": [1, 2, 3], "b": [2, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+    assert lst == [["2", "3", "4", "5", "6"]]
+
+
+@pytest.mark.parametrize("bad_line_func", [lambda x: ["foo", "bar"], lambda x: x[:2]])
+@pytest.mark.parametrize("sep", [",", "111"])
+def test_on_bad_lines_callable_iterator_true(python_parser_only, bad_line_func, sep):
+    # GH 5686
+    # iterator=True has a separate code path than iterator=False
+    parser = python_parser_only
+    data = f"""
+0{sep}1
+hi{sep}there
+foo{sep}bar{sep}baz
+good{sep}bye
+"""
+    bad_sio = StringIO(data)
+    result_iter = parser.read_csv(
+        bad_sio, on_bad_lines=bad_line_func, chunksize=1, iterator=True, sep=sep
+    )
+    expecteds = [
+        {"0": "hi", "1": "there"},
+        {"0": "foo", "1": "bar"},
+        {"0": "good", "1": "bye"},
+    ]
+    for i, (result, expected) in enumerate(zip(result_iter, expecteds)):
+        expected = DataFrame(expected, index=range(i, i + 1))
+        tm.assert_frame_equal(result, expected)
+
+
+def test_on_bad_lines_callable_dont_swallow_errors(python_parser_only):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2
+2,3,4,5,6
+3,4
+"""
+    bad_sio = StringIO(data)
+    msg = "This function is buggy."
+
+    def bad_line_func(bad_line):
+        raise ValueError(msg)
+
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(bad_sio, on_bad_lines=bad_line_func)
+
+
+def test_on_bad_lines_callable_not_expected_length(python_parser_only):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2
+2,3,4,5,6
+3,4
+"""
+    bad_sio = StringIO(data)
+
+    result = parser.read_csv_check_warnings(
+        ParserWarning, "Length of header or names", bad_sio, on_bad_lines=lambda x: x
+    )
+    expected = DataFrame({"a": [1, 2, 3], "b": [2, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_on_bad_lines_callable_returns_none(python_parser_only):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2
+2,3,4,5,6
+3,4
+"""
+    bad_sio = StringIO(data)
+
+    result = parser.read_csv(bad_sio, on_bad_lines=lambda x: None)
+    expected = DataFrame({"a": [1, 3], "b": [2, 4]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_on_bad_lines_index_col_inferred(python_parser_only):
+    # GH 5686
+    parser = python_parser_only
+    data = """a,b
+1,2,3
+4,5,6
+"""
+    bad_sio = StringIO(data)
+
+    result = parser.read_csv(bad_sio, on_bad_lines=lambda x: ["99", "99"])
+    expected = DataFrame({"a": [2, 5], "b": [3, 6]}, index=[1, 4])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_index_col_false_and_header_none(python_parser_only):
+    # GH#46955
+    parser = python_parser_only
+    data = """
+0.5,0.03
+0.1,0.2,0.3,2
+"""
+    result = parser.read_csv_check_warnings(
+        ParserWarning,
+        "Length of header",
+        StringIO(data),
+        sep=",",
+        header=None,
+        index_col=False,
+    )
+    expected = DataFrame({0: [0.5, 0.1], 1: [0.03, 0.2]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_header_int_do_not_infer_multiindex_names_on_different_line(python_parser_only):
+    # GH#46569
+    parser = python_parser_only
+    data = StringIO("a\na,b\nc,d,e\nf,g,h")
+    result = parser.read_csv_check_warnings(
+        ParserWarning, "Length of header", data, engine="python", index_col=False
+    )
+    expected = DataFrame({"a": ["a", "c", "f"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype", [{"a": object}, {"a": str, "b": np.int64, "c": np.int64}]
+)
+def test_no_thousand_convert_with_dot_for_non_numeric_cols(python_parser_only, dtype):
+    # GH#50270
+    parser = python_parser_only
+    data = """\
+a;b;c
+0000.7995;16.000;0
+3.03.001.00514;0;4.000
+4923.600.041;23.000;131"""
+    result = parser.read_csv(
+        StringIO(data),
+        sep=";",
+        dtype=dtype,
+        thousands=".",
+    )
+    expected = DataFrame(
+        {
+            "a": ["0000.7995", "3.03.001.00514", "4923.600.041"],
+            "b": [16000, 0, 23000],
+            "c": [0, 4000, 131],
+        }
+    )
+    if dtype["a"] == object:
+        expected["a"] = expected["a"].astype(object)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype,expected",
+    [
+        (
+            {"a": str, "b": np.float64, "c": np.int64},
+            {
+                "b": [16000.1, 0, 23000],
+                "c": [0, 4001, 131],
+            },
+        ),
+        (
+            str,
+            {
+                "b": ["16,000.1", "0", "23,000"],
+                "c": ["0", "4,001", "131"],
+            },
+        ),
+    ],
+)
+def test_no_thousand_convert_for_non_numeric_cols(python_parser_only, dtype, expected):
+    # GH#50270
+    parser = python_parser_only
+    data = """a;b;c
+0000,7995;16,000.1;0
+3,03,001,00514;0;4,001
+4923,600,041;23,000;131
+"""
+    result = parser.read_csv(
+        StringIO(data),
+        sep=";",
+        dtype=dtype,
+        thousands=",",
+    )
+    expected = DataFrame(expected)
+    expected.insert(0, "a", ["0000,7995", "3,03,001,00514", "4923,600,041"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_quoting.py b/pandas/pandas/tests/io/parser/test_quoting.py
new file mode 100644
index 0000000000000..a70b7e3389c1b
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_quoting.py
@@ -0,0 +1,183 @@
+"""
+Tests that quoting specifications are properly handled
+during parsing for all of the parsers defined in parsers.py
+"""
+
+import csv
+from io import StringIO
+
+import pytest
+
+from pandas.compat import PY311
+from pandas.errors import ParserError
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+
+@pytest.mark.parametrize(
+    "kwargs,msg",
+    [
+        ({"quotechar": "foo"}, '"quotechar" must be a(n)? 1-character string'),
+        (
+            {"quotechar": None, "quoting": csv.QUOTE_MINIMAL},
+            "quotechar must be set if quoting enabled",
+        ),
+        ({"quotechar": 2}, '"quotechar" must be string( or None)?, not int'),
+    ],
+)
+@skip_pyarrow  # ParserError: CSV parse error: Empty CSV file or block
+def test_bad_quote_char(all_parsers, kwargs, msg):
+    data = "1,2,3"
+    parser = all_parsers
+
+    with pytest.raises(TypeError, match=msg):
+        parser.read_csv(StringIO(data), **kwargs)
+
+
+@pytest.mark.parametrize(
+    "quoting,msg",
+    [
+        ("foo", '"quoting" must be an integer|Argument'),
+        (10, 'bad "quoting" value'),  # quoting must be in the range [0, 3]
+    ],
+)
+@xfail_pyarrow  # ValueError: The 'quoting' option is not supported
+def test_bad_quoting(all_parsers, quoting, msg):
+    data = "1,2,3"
+    parser = all_parsers
+
+    with pytest.raises(TypeError, match=msg):
+        parser.read_csv(StringIO(data), quoting=quoting)
+
+
+def test_quote_char_basic(all_parsers):
+    parser = all_parsers
+    data = 'a,b,c\n1,2,"cat"'
+    expected = DataFrame([[1, 2, "cat"]], columns=["a", "b", "c"])
+
+    result = parser.read_csv(StringIO(data), quotechar='"')
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("quote_char", ["~", "*", "%", "$", "@", "P"])
+def test_quote_char_various(all_parsers, quote_char):
+    parser = all_parsers
+    expected = DataFrame([[1, 2, "cat"]], columns=["a", "b", "c"])
+
+    data = 'a,b,c\n1,2,"cat"'
+    new_data = data.replace('"', quote_char)
+
+    result = parser.read_csv(StringIO(new_data), quotechar=quote_char)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: The 'quoting' option is not supported
+@pytest.mark.parametrize("quoting", [csv.QUOTE_MINIMAL, csv.QUOTE_NONE])
+@pytest.mark.parametrize("quote_char", ["", None])
+def test_null_quote_char(all_parsers, quoting, quote_char):
+    kwargs = {"quotechar": quote_char, "quoting": quoting}
+    data = "a,b,c\n1,2,3"
+    parser = all_parsers
+
+    if quoting != csv.QUOTE_NONE:
+        # Sanity checking.
+        msg = (
+            '"quotechar" must be a 1-character string'
+            if PY311 and all_parsers.engine == "python" and quote_char == ""
+            else "quotechar must be set if quoting enabled"
+        )
+
+        with pytest.raises(TypeError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+    elif not (PY311 and all_parsers.engine == "python"):
+        # Python 3.11+ doesn't support null/blank quote chars in their csv parsers
+        expected = DataFrame([[1, 2, 3]], columns=["a", "b", "c"])
+        result = parser.read_csv(StringIO(data), **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,exp_data",
+    [
+        ({}, [[1, 2, "foo"]]),  # Test default.
+        # QUOTE_MINIMAL only applies to CSV writing, so no effect on reading.
+        ({"quotechar": '"', "quoting": csv.QUOTE_MINIMAL}, [[1, 2, "foo"]]),
+        # QUOTE_MINIMAL only applies to CSV writing, so no effect on reading.
+        ({"quotechar": '"', "quoting": csv.QUOTE_ALL}, [[1, 2, "foo"]]),
+        # QUOTE_NONE tells the reader to do no special handling
+        # of quote characters and leave them alone.
+        ({"quotechar": '"', "quoting": csv.QUOTE_NONE}, [[1, 2, '"foo"']]),
+        # QUOTE_NONNUMERIC tells the reader to cast
+        # all non-quoted fields to float
+        ({"quotechar": '"', "quoting": csv.QUOTE_NONNUMERIC}, [[1.0, 2.0, "foo"]]),
+    ],
+)
+@xfail_pyarrow  # ValueError: The 'quoting' option is not supported
+def test_quoting_various(all_parsers, kwargs, exp_data):
+    data = '1,2,"foo"'
+    parser = all_parsers
+    columns = ["a", "b", "c"]
+
+    result = parser.read_csv(StringIO(data), names=columns, **kwargs)
+    expected = DataFrame(exp_data, columns=columns)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "doublequote,exp_data", [(True, [[3, '4 " 5']]), (False, [[3, '4 " 5"']])]
+)
+def test_double_quote(all_parsers, doublequote, exp_data, request):
+    parser = all_parsers
+    data = 'a,b\n3,"4 "" 5"'
+
+    if parser.engine == "pyarrow" and not doublequote:
+        mark = pytest.mark.xfail(reason="Mismatched result")
+        request.applymarker(mark)
+
+    result = parser.read_csv(StringIO(data), quotechar='"', doublequote=doublequote)
+    expected = DataFrame(exp_data, columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("quotechar", ['"', "\u0001"])
+def test_quotechar_unicode(all_parsers, quotechar):
+    # see gh-14477
+    data = "a\n1"
+    parser = all_parsers
+    expected = DataFrame({"a": [1]})
+
+    result = parser.read_csv(StringIO(data), quotechar=quotechar)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("balanced", [True, False])
+def test_unbalanced_quoting(all_parsers, balanced, request):
+    # see gh-22789.
+    parser = all_parsers
+    data = 'a,b,c\n1,2,"3'
+
+    if parser.engine == "pyarrow" and not balanced:
+        mark = pytest.mark.xfail(reason="Mismatched result")
+        request.applymarker(mark)
+
+    if balanced:
+        # Re-balance the quoting and read in without errors.
+        expected = DataFrame([[1, 2, 3]], columns=["a", "b", "c"])
+        result = parser.read_csv(StringIO(data + '"'))
+        tm.assert_frame_equal(result, expected)
+    else:
+        msg = (
+            "EOF inside string starting at row 1"
+            if parser.engine == "c"
+            else "unexpected end of data"
+        )
+
+        with pytest.raises(ParserError, match=msg):
+            parser.read_csv(StringIO(data))
diff --git a/pandas/pandas/tests/io/parser/test_read_fwf.py b/pandas/pandas/tests/io/parser/test_read_fwf.py
new file mode 100644
index 0000000000000..6243185294894
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_read_fwf.py
@@ -0,0 +1,1007 @@
+"""
+Tests the 'read_fwf' function in parsers.py. This
+test suite is independent of the others because the
+engine is set to 'python-fwf' internally.
+"""
+
+from io import (
+    BytesIO,
+    StringIO,
+)
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+from pandas.errors import EmptyDataError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+)
+import pandas._testing as tm
+
+from pandas.io.common import urlopen
+from pandas.io.parsers import (
+    read_csv,
+    read_fwf,
+)
+
+
+def test_basic():
+    data = """\
+A         B            C            D
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    result = read_fwf(StringIO(data))
+    expected = DataFrame(
+        [
+            [201158, 360.242940, 149.910199, 11950.7],
+            [201159, 444.953632, 166.985655, 11788.4],
+            [201160, 364.136849, 183.628767, 11806.2],
+            [201161, 413.836124, 184.375703, 11916.8],
+            [201162, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_colspecs():
+    data = """\
+A   B     C            D            E
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    colspecs = [(0, 4), (4, 8), (8, 20), (21, 33), (34, 43)]
+    result = read_fwf(StringIO(data), colspecs=colspecs)
+
+    expected = DataFrame(
+        [
+            [2011, 58, 360.242940, 149.910199, 11950.7],
+            [2011, 59, 444.953632, 166.985655, 11788.4],
+            [2011, 60, 364.136849, 183.628767, 11806.2],
+            [2011, 61, 413.836124, 184.375703, 11916.8],
+            [2011, 62, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D", "E"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_widths():
+    data = """\
+A    B    C            D            E
+2011 58   360.242940   149.910199   11950.7
+2011 59   444.953632   166.985655   11788.4
+2011 60   364.136849   183.628767   11806.2
+2011 61   413.836124   184.375703   11916.8
+2011 62   502.953953   173.237159   12468.3
+"""
+    result = read_fwf(StringIO(data), widths=[5, 5, 13, 13, 7])
+
+    expected = DataFrame(
+        [
+            [2011, 58, 360.242940, 149.910199, 11950.7],
+            [2011, 59, 444.953632, 166.985655, 11788.4],
+            [2011, 60, 364.136849, 183.628767, 11806.2],
+            [2011, 61, 413.836124, 184.375703, 11916.8],
+            [2011, 62, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D", "E"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_non_space_filler():
+    # From Thomas Kluyver:
+    #
+    # Apparently, some non-space filler characters can be seen, this is
+    # supported by specifying the 'delimiter' character:
+    #
+    # http://publib.boulder.ibm.com/infocenter/dmndhelp/v6r1mx/index.jsp?topic=/com.ibm.wbit.612.help.config.doc/topics/rfixwidth.html
+    data = """\
+A~~~~B~~~~C~~~~~~~~~~~~D~~~~~~~~~~~~E
+201158~~~~360.242940~~~149.910199~~~11950.7
+201159~~~~444.953632~~~166.985655~~~11788.4
+201160~~~~364.136849~~~183.628767~~~11806.2
+201161~~~~413.836124~~~184.375703~~~11916.8
+201162~~~~502.953953~~~173.237159~~~12468.3
+"""
+    colspecs = [(0, 4), (4, 8), (8, 20), (21, 33), (34, 43)]
+    result = read_fwf(StringIO(data), colspecs=colspecs, delimiter="~")
+
+    expected = DataFrame(
+        [
+            [2011, 58, 360.242940, 149.910199, 11950.7],
+            [2011, 59, 444.953632, 166.985655, 11788.4],
+            [2011, 60, 364.136849, 183.628767, 11806.2],
+            [2011, 61, 413.836124, 184.375703, 11916.8],
+            [2011, 62, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D", "E"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_over_specified():
+    data = """\
+A   B     C            D            E
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    colspecs = [(0, 4), (4, 8), (8, 20), (21, 33), (34, 43)]
+
+    with pytest.raises(ValueError, match="must specify only one of"):
+        read_fwf(StringIO(data), colspecs=colspecs, widths=[6, 10, 10, 7])
+
+
+def test_under_specified():
+    data = """\
+A   B     C            D            E
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    with pytest.raises(ValueError, match="Must specify either"):
+        read_fwf(StringIO(data), colspecs=None, widths=None)
+
+
+def test_read_csv_compat():
+    csv_data = """\
+A,B,C,D,E
+2011,58,360.242940,149.910199,11950.7
+2011,59,444.953632,166.985655,11788.4
+2011,60,364.136849,183.628767,11806.2
+2011,61,413.836124,184.375703,11916.8
+2011,62,502.953953,173.237159,12468.3
+"""
+    expected = read_csv(StringIO(csv_data), engine="python")
+
+    fwf_data = """\
+A   B     C            D            E
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    colspecs = [(0, 4), (4, 8), (8, 20), (21, 33), (34, 43)]
+    result = read_fwf(StringIO(fwf_data), colspecs=colspecs)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_bytes_io_input():
+    data = BytesIO("שלום\nשלום".encode())  # noqa: RUF001
+    result = read_fwf(data, widths=[2, 2], encoding="utf8")
+    expected = DataFrame([["של", "ום"]], columns=["של", "ום"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_fwf_colspecs_is_list_or_tuple():
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+
+    msg = "column specifications must be a list or tuple.+"
+
+    with pytest.raises(TypeError, match=msg):
+        read_fwf(StringIO(data), colspecs={"a": 1}, delimiter=",")
+
+
+def test_fwf_colspecs_is_list_or_tuple_of_two_element_tuples():
+    data = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+
+    msg = "Each column specification must be.+"
+
+    with pytest.raises(TypeError, match=msg):
+        read_fwf(StringIO(data), colspecs=[("a", 1)])
+
+
+@pytest.mark.parametrize(
+    "colspecs,exp_data",
+    [
+        ([(0, 3), (3, None)], [[123, 456], [456, 789]]),
+        ([(None, 3), (3, 6)], [[123, 456], [456, 789]]),
+        ([(0, None), (3, None)], [[123456, 456], [456789, 789]]),
+        ([(None, None), (3, 6)], [[123456, 456], [456789, 789]]),
+    ],
+)
+def test_fwf_colspecs_none(colspecs, exp_data):
+    # see gh-7079
+    data = """\
+123456
+456789
+"""
+    expected = DataFrame(exp_data)
+
+    result = read_fwf(StringIO(data), colspecs=colspecs, header=None)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "infer_nrows,exp_data",
+    [
+        # infer_nrows --> colspec == [(2, 3), (5, 6)]
+        (1, [[1, 2], [3, 8]]),
+        # infer_nrows > number of rows
+        (10, [[1, 2], [123, 98]]),
+    ],
+)
+def test_fwf_colspecs_infer_nrows(infer_nrows, exp_data):
+    # see gh-15138
+    data = """\
+  1  2
+123 98
+"""
+    expected = DataFrame(exp_data)
+
+    result = read_fwf(StringIO(data), infer_nrows=infer_nrows, header=None)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_fwf_regression():
+    # see gh-3594
+    #
+    # Turns out "T060" is parsable as a datetime slice!
+    tz_list = [1, 10, 20, 30, 60, 80, 100]
+    widths = [16] + [8] * len(tz_list)
+    names = ["SST"] + [f"T{z:03d}" for z in tz_list[1:]]
+
+    data = """  2009164202000   9.5403  9.4105  8.6571  7.8372  6.0612  5.8843  5.5192
+2009164203000   9.5435  9.2010  8.6167  7.8176  6.0804  5.8728  5.4869
+2009164204000   9.5873  9.1326  8.4694  7.5889  6.0422  5.8526  5.4657
+2009164205000   9.5810  9.0896  8.4009  7.4652  6.0322  5.8189  5.4379
+2009164210000   9.6034  9.0897  8.3822  7.4905  6.0908  5.7904  5.4039
+"""
+    expected = DataFrame(
+        [
+            [9.5403, 9.4105, 8.6571, 7.8372, 6.0612, 5.8843, 5.5192],
+            [9.5435, 9.2010, 8.6167, 7.8176, 6.0804, 5.8728, 5.4869],
+            [9.5873, 9.1326, 8.4694, 7.5889, 6.0422, 5.8526, 5.4657],
+            [9.5810, 9.0896, 8.4009, 7.4652, 6.0322, 5.8189, 5.4379],
+            [9.6034, 9.0897, 8.3822, 7.4905, 6.0908, 5.7904, 5.4039],
+        ],
+        index=DatetimeIndex(
+            [
+                "2009-06-13 20:20:00",
+                "2009-06-13 20:30:00",
+                "2009-06-13 20:40:00",
+                "2009-06-13 20:50:00",
+                "2009-06-13 21:00:00",
+            ],
+            dtype="M8[us]",
+        ),
+        columns=["SST", "T010", "T020", "T030", "T060", "T080", "T100"],
+    )
+    result = read_fwf(
+        StringIO(data),
+        index_col=0,
+        header=None,
+        names=names,
+        widths=widths,
+        parse_dates=True,
+        date_format="%Y%j%H%M%S",
+    )
+    expected.index = expected.index.astype("M8[s]")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_fwf_for_uint8():
+    data = """1421302965.213420    PRI=3 PGN=0xef00      DST=0x17 SRC=0x28    04 154 00 00 00 00 00 127
+1421302964.226776    PRI=6 PGN=0xf002               SRC=0x47    243 00 00 255 247 00 00 71"""  # noqa: E501
+    df = read_fwf(
+        StringIO(data),
+        colspecs=[(0, 17), (25, 26), (33, 37), (49, 51), (58, 62), (63, 1000)],
+        names=["time", "pri", "pgn", "dst", "src", "data"],
+        converters={
+            "pgn": lambda x: int(x, 16),
+            "src": lambda x: int(x, 16),
+            "dst": lambda x: int(x, 16),
+            "data": lambda x: len(x.split(" ")),
+        },
+    )
+
+    expected = DataFrame(
+        [
+            [1421302965.213420, 3, 61184, 23, 40, 8],
+            [1421302964.226776, 6, 61442, None, 71, 8],
+        ],
+        columns=["time", "pri", "pgn", "dst", "src", "data"],
+    )
+    expected["dst"] = expected["dst"].astype(object)
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("comment", ["#", "~", "!"])
+def test_fwf_comment(comment):
+    data = """\
+  1   2.   4  #hello world
+  5  NaN  10.0
+"""
+    data = data.replace("#", comment)
+
+    colspecs = [(0, 3), (4, 9), (9, 25)]
+    expected = DataFrame([[1, 2.0, 4], [5, np.nan, 10.0]])
+
+    result = read_fwf(StringIO(data), colspecs=colspecs, header=None, comment=comment)
+    tm.assert_almost_equal(result, expected)
+
+
+def test_fwf_skip_blank_lines():
+    data = """
+
+A         B            C            D
+
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+
+
+201162    502.953953   173.237159   12468.3
+
+"""
+    result = read_fwf(StringIO(data), skip_blank_lines=True)
+    expected = DataFrame(
+        [
+            [201158, 360.242940, 149.910199, 11950.7],
+            [201159, 444.953632, 166.985655, 11788.4],
+            [201162, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+    data = """\
+A         B            C            D
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+
+
+201162    502.953953   173.237159   12468.3
+"""
+    result = read_fwf(StringIO(data), skip_blank_lines=False)
+    expected = DataFrame(
+        [
+            [201158, 360.242940, 149.910199, 11950.7],
+            [201159, 444.953632, 166.985655, 11788.4],
+            [np.nan, np.nan, np.nan, np.nan],
+            [np.nan, np.nan, np.nan, np.nan],
+            [201162, 502.953953, 173.237159, 12468.3],
+        ],
+        columns=["A", "B", "C", "D"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("thousands", [",", "#", "~"])
+def test_fwf_thousands(thousands):
+    data = """\
+ 1 2,334.0    5
+10   13     10.
+"""
+    data = data.replace(",", thousands)
+
+    colspecs = [(0, 3), (3, 11), (12, 16)]
+    expected = DataFrame([[1, 2334.0, 5], [10, 13, 10.0]])
+
+    result = read_fwf(
+        StringIO(data), header=None, colspecs=colspecs, thousands=thousands
+    )
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize("header", [True, False])
+def test_bool_header_arg(header):
+    # see gh-6114
+    data = """\
+MyColumn
+   a
+   b
+   a
+   b"""
+
+    msg = "Passing a bool to header is invalid"
+    with pytest.raises(TypeError, match=msg):
+        read_fwf(StringIO(data), header=header)
+
+
+def test_full_file():
+    # File with all values.
+    test = """index                             A    B    C
+2000-01-03T00:00:00  0.980268513777    3  foo
+2000-01-04T00:00:00  1.04791624281    -4  bar
+2000-01-05T00:00:00  0.498580885705   73  baz
+2000-01-06T00:00:00  1.12020151869     1  foo
+2000-01-07T00:00:00  0.487094399463    0  bar
+2000-01-10T00:00:00  0.836648671666    2  baz
+2000-01-11T00:00:00  0.157160753327   34  foo"""
+    colspecs = ((0, 19), (21, 35), (38, 40), (42, 45))
+    expected = read_fwf(StringIO(test), colspecs=colspecs)
+
+    result = read_fwf(StringIO(test))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_full_file_with_missing():
+    # File with missing values.
+    test = """index                             A    B    C
+2000-01-03T00:00:00  0.980268513777    3  foo
+2000-01-04T00:00:00  1.04791624281    -4  bar
+                     0.498580885705   73  baz
+2000-01-06T00:00:00  1.12020151869     1  foo
+2000-01-07T00:00:00                    0  bar
+2000-01-10T00:00:00  0.836648671666    2  baz
+                                      34"""
+    colspecs = ((0, 19), (21, 35), (38, 40), (42, 45))
+    expected = read_fwf(StringIO(test), colspecs=colspecs)
+
+    result = read_fwf(StringIO(test))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_full_file_with_spaces():
+    # File with spaces in columns.
+    test = """
+Account                 Name  Balance     CreditLimit   AccountCreated
+101     Keanu Reeves          9315.45     10000.00           1/17/1998
+312     Gerard Butler         90.00       1000.00             8/6/2003
+868     Jennifer Love Hewitt  0           17000.00           5/25/1985
+761     Jada Pinkett-Smith    49654.87    100000.00          12/5/2006
+317     Bill Murray           789.65      5000.00             2/5/2007
+""".strip("\r\n")
+    colspecs = ((0, 7), (8, 28), (30, 38), (42, 53), (56, 70))
+    expected = read_fwf(StringIO(test), colspecs=colspecs)
+
+    result = read_fwf(StringIO(test))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_full_file_with_spaces_and_missing():
+    # File with spaces and missing values in columns.
+    test = """
+Account               Name    Balance     CreditLimit   AccountCreated
+101                           10000.00                       1/17/1998
+312     Gerard Butler         90.00       1000.00             8/6/2003
+868                                                          5/25/1985
+761     Jada Pinkett-Smith    49654.87    100000.00          12/5/2006
+317     Bill Murray           789.65
+""".strip("\r\n")
+    colspecs = ((0, 7), (8, 28), (30, 38), (42, 53), (56, 70))
+    expected = read_fwf(StringIO(test), colspecs=colspecs)
+
+    result = read_fwf(StringIO(test))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_messed_up_data():
+    # Completely messed up file.
+    test = """
+   Account          Name             Balance     Credit Limit   Account Created
+       101                           10000.00                       1/17/1998
+       312     Gerard Butler         90.00       1000.00
+
+       761     Jada Pinkett-Smith    49654.87    100000.00          12/5/2006
+  317          Bill Murray           789.65
+""".strip("\r\n")
+    colspecs = ((2, 10), (15, 33), (37, 45), (49, 61), (64, 79))
+    expected = read_fwf(StringIO(test), colspecs=colspecs)
+
+    result = read_fwf(StringIO(test))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_multiple_delimiters():
+    test = r"""
+col1~~~~~col2  col3++++++++++++++++++col4
+~~22.....11.0+++foo~~~~~~~~~~Keanu Reeves
+  33+++122.33\\\bar.........Gerard Butler
+++44~~~~12.01   baz~~Jennifer Love Hewitt
+~~55       11+++foo++++Jada Pinkett-Smith
+..66++++++.03~~~bar           Bill Murray
+""".strip("\r\n")
+    delimiter = " +~.\\"
+    colspecs = ((0, 4), (7, 13), (15, 19), (21, 41))
+    expected = read_fwf(StringIO(test), colspecs=colspecs, delimiter=delimiter)
+
+    result = read_fwf(StringIO(test), delimiter=delimiter)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_variable_width_unicode():
+    data = """
+שלום שלום
+ום   שלל
+של   ום
+""".strip("\r\n")
+    encoding = "utf8"
+    kwargs = {"header": None, "encoding": encoding}
+
+    expected = read_fwf(
+        BytesIO(data.encode(encoding)), colspecs=[(0, 4), (5, 9)], **kwargs
+    )
+    result = read_fwf(BytesIO(data.encode(encoding)), **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [{}, {"a": "float64", "b": str, "c": "int32"}])
+def test_dtype(dtype):
+    data = """ a    b    c
+1    2    3.2
+3    4    5.2
+"""
+    colspecs = [(0, 5), (5, 10), (10, None)]
+    result = read_fwf(StringIO(data), colspecs=colspecs, dtype=dtype)
+
+    expected = DataFrame(
+        {"a": [1, 3], "b": [2, 4], "c": [3.2, 5.2]}, columns=["a", "b", "c"]
+    )
+
+    for col, dt in dtype.items():
+        expected[col] = expected[col].astype(dt)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skiprows_inference():
+    # see gh-11256
+    data = """
+Text contained in the file header
+
+DataCol1   DataCol2
+     0.0        1.0
+   101.6      956.1
+""".strip()
+    skiprows = 2
+
+    expected = read_csv(StringIO(data), skiprows=skiprows, sep=r"\s+")
+    result = read_fwf(StringIO(data), skiprows=skiprows)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skiprows_by_index_inference():
+    data = """
+To be skipped
+Not  To  Be  Skipped
+Once more to be skipped
+123  34   8      123
+456  78   9      456
+""".strip()
+    skiprows = [0, 2]
+
+    expected = read_csv(StringIO(data), skiprows=skiprows, sep=r"\s+")
+    result = read_fwf(StringIO(data), skiprows=skiprows)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skiprows_inference_empty():
+    data = """
+AA   BBB  C
+12   345  6
+78   901  2
+""".strip()
+
+    msg = "No rows from which to infer column width"
+    with pytest.raises(EmptyDataError, match=msg):
+        read_fwf(StringIO(data), skiprows=3)
+
+
+def test_whitespace_preservation():
+    # see gh-16772
+    header = None
+    csv_data = """
+ a ,bbb
+ cc,dd """
+
+    fwf_data = """
+ a bbb
+ ccdd """
+    result = read_fwf(
+        StringIO(fwf_data), widths=[3, 3], header=header, skiprows=[0], delimiter="\n\t"
+    )
+    expected = read_csv(StringIO(csv_data), header=header)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_default_delimiter():
+    header = None
+    csv_data = """
+a,bbb
+cc,dd"""
+
+    fwf_data = """
+a \tbbb
+cc\tdd """
+    result = read_fwf(StringIO(fwf_data), widths=[3, 3], header=header, skiprows=[0])
+    expected = read_csv(StringIO(csv_data), header=header)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("infer", [True, False])
+def test_fwf_compression(compression_only, infer, compression_to_extension):
+    data = """1111111111
+    2222222222
+    3333333333""".strip()
+
+    compression = compression_only
+    extension = compression_to_extension[compression]
+
+    kwargs = {"widths": [5, 5], "names": ["one", "two"]}
+    expected = read_fwf(StringIO(data), **kwargs)
+
+    data = bytes(data, encoding="utf-8")
+
+    with tm.ensure_clean(filename="tmp." + extension) as path:
+        tm.write_to_compressed(compression, path, data)
+
+        if infer is not None:
+            kwargs["compression"] = "infer" if infer else compression
+
+        result = read_fwf(path, **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_binary_mode():
+    """
+    read_fwf supports opening files in binary mode.
+
+    GH 18035.
+    """
+    data = """aaa aaa aaa
+bba bab b a"""
+    df_reference = DataFrame(
+        [["bba", "bab", "b a"]], columns=["aaa", "aaa.1", "aaa.2"], index=[0]
+    )
+    with tm.ensure_clean() as path:
+        Path(path).write_text(data, encoding="utf-8")
+        with open(path, "rb") as file:
+            df = read_fwf(file)
+            file.seek(0)
+            tm.assert_frame_equal(df, df_reference)
+
+
+@pytest.mark.parametrize("memory_map", [True, False])
+def test_encoding_mmap(memory_map):
+    """
+    encoding should be working, even when using a memory-mapped file.
+
+    GH 23254.
+    """
+    encoding = "iso8859_1"
+    with tm.ensure_clean() as path:
+        Path(path).write_bytes(" 1 A Ä 2\n".encode(encoding))
+        df = read_fwf(
+            path,
+            header=None,
+            widths=[2, 2, 2, 2],
+            encoding=encoding,
+            memory_map=memory_map,
+        )
+    df_reference = DataFrame([[1, "A", "Ä", 2]])
+    tm.assert_frame_equal(df, df_reference)
+
+
+@pytest.mark.parametrize(
+    "colspecs, names, widths, index_col",
+    [
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("abcde"),
+            None,
+            None,
+        ),
+        (
+            None,
+            list("abcde"),
+            [6] * 4,
+            None,
+        ),
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("abcde"),
+            None,
+            True,
+        ),
+        (
+            None,
+            list("abcde"),
+            [6] * 4,
+            False,
+        ),
+        (
+            None,
+            list("abcde"),
+            [6] * 4,
+            True,
+        ),
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("abcde"),
+            None,
+            False,
+        ),
+    ],
+)
+def test_len_colspecs_len_names(colspecs, names, widths, index_col):
+    # GH#40830
+    data = """col1  col2  col3  col4
+    bab   ba    2"""
+    msg = "Length of colspecs must match length of names"
+    with pytest.raises(ValueError, match=msg):
+        read_fwf(
+            StringIO(data),
+            colspecs=colspecs,
+            names=names,
+            widths=widths,
+            index_col=index_col,
+        )
+
+
+@pytest.mark.parametrize(
+    "colspecs, names, widths, index_col, expected",
+    [
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("abc"),
+            None,
+            0,
+            DataFrame(
+                index=["col1", "ba"],
+                columns=["a", "b", "c"],
+                data=[["col2", "col3", "col4"], ["b   ba", "2", np.nan]],
+            ),
+        ),
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("ab"),
+            None,
+            [0, 1],
+            DataFrame(
+                index=[["col1", "ba"], ["col2", "b   ba"]],
+                columns=["a", "b"],
+                data=[["col3", "col4"], ["2", np.nan]],
+            ),
+        ),
+        (
+            [(0, 6), (6, 12), (12, 18), (18, None)],
+            list("a"),
+            None,
+            [0, 1, 2],
+            DataFrame(
+                index=[["col1", "ba"], ["col2", "b   ba"], ["col3", "2"]],
+                columns=["a"],
+                data=[["col4"], [np.nan]],
+            ),
+        ),
+        (
+            None,
+            list("abc"),
+            [6] * 4,
+            0,
+            DataFrame(
+                index=["col1", "ba"],
+                columns=["a", "b", "c"],
+                data=[["col2", "col3", "col4"], ["b   ba", "2", np.nan]],
+            ),
+        ),
+        (
+            None,
+            list("ab"),
+            [6] * 4,
+            [0, 1],
+            DataFrame(
+                index=[["col1", "ba"], ["col2", "b   ba"]],
+                columns=["a", "b"],
+                data=[["col3", "col4"], ["2", np.nan]],
+            ),
+        ),
+        (
+            None,
+            list("a"),
+            [6] * 4,
+            [0, 1, 2],
+            DataFrame(
+                index=[["col1", "ba"], ["col2", "b   ba"], ["col3", "2"]],
+                columns=["a"],
+                data=[["col4"], [np.nan]],
+            ),
+        ),
+    ],
+)
+def test_len_colspecs_len_names_with_index_col(
+    colspecs, names, widths, index_col, expected
+):
+    # GH#40830
+    data = """col1  col2  col3  col4
+    bab   ba    2"""
+    result = read_fwf(
+        StringIO(data),
+        colspecs=colspecs,
+        names=names,
+        widths=widths,
+        index_col=index_col,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_colspecs_with_comment():
+    # GH 14135
+    result = read_fwf(
+        StringIO("#\nA1K\n"), colspecs=[(1, 2), (2, 3)], comment="#", header=None
+    )
+    expected = DataFrame([[1, "K"]], columns=[0, 1])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skip_rows_and_n_rows():
+    # GH#44021
+    data = """a\tb
+1\t a
+2\t b
+3\t c
+4\t d
+5\t e
+6\t f
+    """
+    result = read_fwf(StringIO(data), nrows=4, skiprows=[2, 4])
+    expected = DataFrame({"a": [1, 3, 5, 6], "b": ["a", "c", "e", "f"]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_skiprows_with_iterator():
+    # GH#10261, GH#56323
+    data = """0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+    """
+    df_iter = read_fwf(
+        StringIO(data),
+        colspecs=[(0, 2)],
+        names=["a"],
+        iterator=True,
+        chunksize=2,
+        skiprows=[0, 1, 2, 6, 9],
+    )
+    expected_frames = [
+        DataFrame({"a": [3, 4]}),
+        DataFrame({"a": [5, 7]}, index=[2, 3]),
+        DataFrame({"a": [8]}, index=[4]),
+    ]
+    for i, result in enumerate(df_iter):
+        tm.assert_frame_equal(result, expected_frames[i])
+
+
+def test_names_and_infer_colspecs():
+    # GH#45337
+    data = """X   Y   Z
+      959.0    345   22.2
+    """
+    result = read_fwf(StringIO(data), skiprows=1, usecols=[0, 2], names=["a", "b"])
+    expected = DataFrame({"a": [959.0], "b": 22.2})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_widths_and_usecols():
+    # GH#46580
+    data = """0  1    n -0.4100.1
+0  2    p  0.2 90.1
+0  3    n -0.3140.4"""
+    result = read_fwf(
+        StringIO(data),
+        header=None,
+        usecols=(0, 1, 3),
+        widths=(3, 5, 1, 5, 5),
+        index_col=False,
+        names=("c0", "c1", "c3"),
+    )
+    expected = DataFrame(
+        {
+            "c0": 0,
+            "c1": [1, 2, 3],
+            "c3": [-0.4, 0.2, -0.3],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dtype_backend(string_storage, dtype_backend):
+    # GH#50289
+    data = """a  b    c      d  e     f  g    h  i
+1  2.5  True  a
+3  4.5  False b  True  6  7.5  a"""
+    with pd.option_context("mode.string_storage", string_storage):
+        result = read_fwf(StringIO(data), dtype_backend=dtype_backend)
+
+    if dtype_backend == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+        string_dtype = pd.ArrowDtype(pa.string())
+    else:
+        string_dtype = pd.StringDtype(string_storage)
+
+    expected = DataFrame(
+        {
+            "a": pd.Series([1, 3], dtype="Int64"),
+            "b": pd.Series([2.5, 4.5], dtype="Float64"),
+            "c": pd.Series([True, False], dtype="boolean"),
+            "d": pd.Series(["a", "b"], dtype=string_dtype),
+            "e": pd.Series([pd.NA, True], dtype="boolean"),
+            "f": pd.Series([pd.NA, 6], dtype="Int64"),
+            "g": pd.Series([pd.NA, 7.5], dtype="Float64"),
+            "h": pd.Series([None, "a"], dtype=string_dtype),
+            "i": pd.Series([pd.NA, pd.NA], dtype="Int64"),
+        }
+    )
+    if dtype_backend == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+        from pandas.arrays import ArrowExtensionArray
+
+        expected = DataFrame(
+            {
+                col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                for col in expected.columns
+            }
+        )
+        expected["i"] = ArrowExtensionArray(pa.array([None, None]))
+
+    # the storage of the str columns' Index is also affected by the
+    # string_storage setting -> ignore that for checking the result
+    tm.assert_frame_equal(result, expected, check_column_type=False)
+
+
+def test_invalid_dtype_backend():
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        read_fwf("test", dtype_backend="numpy")
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_url_urlopen(httpserver):
+    data = """\
+A         B            C            D
+201158    360.242940   149.910199   11950.7
+201159    444.953632   166.985655   11788.4
+201160    364.136849   183.628767   11806.2
+201161    413.836124   184.375703   11916.8
+201162    502.953953   173.237159   12468.3
+"""
+    httpserver.serve_content(content=data)
+    expected = pd.Index(list("ABCD"))
+    with urlopen(httpserver.url) as f:
+        result = read_fwf(f).columns
+
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/test_skiprows.py b/pandas/pandas/tests/io/parser/test_skiprows.py
new file mode 100644
index 0000000000000..99642ee4befc6
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_skiprows.py
@@ -0,0 +1,336 @@
+"""
+Tests that skipped rows are properly handled during
+parsing for all of the parsers defined in parsers.py
+"""
+
+from datetime import datetime
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas.errors import EmptyDataError
+
+from pandas import (
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+@pytest.mark.parametrize("skiprows", [list(range(6)), 6])
+def test_skip_rows_bug(all_parsers, skiprows):
+    # see gh-505
+    parser = all_parsers
+    text = """#foo,a,b,c
+#foo,a,b,c
+#foo,a,b,c
+#foo,a,b,c
+#foo,a,b,c
+#foo,a,b,c
+1/1/2000,1.,2.,3.
+1/2/2000,4,5,6
+1/3/2000,7,8,9
+"""
+    result = parser.read_csv(
+        StringIO(text), skiprows=skiprows, header=None, index_col=0, parse_dates=True
+    )
+    index = Index(
+        [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+        dtype="M8[s]",
+        name=0,
+    )
+
+    expected = DataFrame(
+        np.arange(1.0, 10.0).reshape((3, 3)), columns=[1, 2, 3], index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_deep_skip_rows(all_parsers):
+    # see gh-4382
+    parser = all_parsers
+    data = "a,b,c\n" + "\n".join(
+        [",".join([str(i), str(i + 1), str(i + 2)]) for i in range(10)]
+    )
+    condensed_data = "a,b,c\n" + "\n".join(
+        [",".join([str(i), str(i + 1), str(i + 2)]) for i in [0, 1, 2, 3, 4, 6, 8, 9]]
+    )
+
+    result = parser.read_csv(StringIO(data), skiprows=[6, 8])
+    condensed_result = parser.read_csv(StringIO(condensed_data))
+    tm.assert_frame_equal(result, condensed_result)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame are different
+def test_skip_rows_blank(all_parsers):
+    # see gh-9832
+    parser = all_parsers
+    text = """#foo,a,b,c
+#foo,a,b,c
+
+#foo,a,b,c
+#foo,a,b,c
+
+1/1/2000,1.,2.,3.
+1/2/2000,4,5,6
+1/3/2000,7,8,9
+"""
+    data = parser.read_csv(
+        StringIO(text), skiprows=6, header=None, index_col=0, parse_dates=True
+    )
+    index = Index(
+        [datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)],
+        dtype="M8[s]",
+        name=0,
+    )
+
+    expected = DataFrame(
+        np.arange(1.0, 10.0).reshape((3, 3)), columns=[1, 2, 3], index=index
+    )
+    tm.assert_frame_equal(data, expected)
+
+
+@pytest.mark.parametrize(
+    "data,kwargs,expected",
+    [
+        (
+            """id,text,num_lines
+1,"line 11
+line 12",2
+2,"line 21
+line 22",2
+3,"line 31",1""",
+            {"skiprows": [1]},
+            DataFrame(
+                [[2, "line 21\nline 22", 2], [3, "line 31", 1]],
+                columns=["id", "text", "num_lines"],
+            ),
+        ),
+        (
+            "a,b,c\n~a\n b~,~e\n d~,~f\n f~\n1,2,~12\n 13\n 14~",
+            {"quotechar": "~", "skiprows": [2]},
+            DataFrame([["a\n b", "e\n d", "f\n f"]], columns=["a", "b", "c"]),
+        ),
+        (
+            (
+                "Text,url\n~example\n "
+                "sentence\n one~,url1\n~"
+                "example\n sentence\n two~,url2\n~"
+                "example\n sentence\n three~,url3"
+            ),
+            {"quotechar": "~", "skiprows": [1, 3]},
+            DataFrame([["example\n sentence\n two", "url2"]], columns=["Text", "url"]),
+        ),
+    ],
+)
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_row_with_newline(all_parsers, data, kwargs, expected):
+    # see gh-12775 and gh-10911
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_row_with_quote(all_parsers):
+    # see gh-12775 and gh-10911
+    parser = all_parsers
+    data = """id,text,num_lines
+1,"line '11' line 12",2
+2,"line '21' line 22",2
+3,"line '31' line 32",1"""
+
+    exp_data = [[2, "line '21' line 22", 2], [3, "line '31' line 32", 1]]
+    expected = DataFrame(exp_data, columns=["id", "text", "num_lines"])
+
+    result = parser.read_csv(StringIO(data), skiprows=[1])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,exp_data",
+    [
+        (
+            """id,text,num_lines
+1,"line \n'11' line 12",2
+2,"line \n'21' line 22",2
+3,"line \n'31' line 32",1""",
+            [[2, "line \n'21' line 22", 2], [3, "line \n'31' line 32", 1]],
+        ),
+        (
+            """id,text,num_lines
+1,"line '11\n' line 12",2
+2,"line '21\n' line 22",2
+3,"line '31\n' line 32",1""",
+            [[2, "line '21\n' line 22", 2], [3, "line '31\n' line 32", 1]],
+        ),
+        (
+            """id,text,num_lines
+1,"line '11\n' \r\tline 12",2
+2,"line '21\n' \r\tline 22",2
+3,"line '31\n' \r\tline 32",1""",
+            [[2, "line '21\n' \r\tline 22", 2], [3, "line '31\n' \r\tline 32", 1]],
+        ),
+    ],
+)
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_row_with_newline_and_quote(all_parsers, data, exp_data):
+    # see gh-12775 and gh-10911
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), skiprows=[1])
+
+    expected = DataFrame(exp_data, columns=["id", "text", "num_lines"])
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: the 'pyarrow' engine does not support regex separators
+@pytest.mark.parametrize(
+    "lineterminator",
+    ["\n", "\r\n", "\r"],  # "LF"  # "CRLF"  # "CR"
+)
+def test_skiprows_lineterminator(all_parsers, lineterminator, request):
+    # see gh-9079
+    parser = all_parsers
+    data = "\n".join(
+        [
+            "SMOSMANIA ThetaProbe-ML2X ",
+            "2007/01/01 01:00   0.2140 U M ",
+            "2007/01/01 02:00   0.2141 M O ",
+            "2007/01/01 04:00   0.2142 D M ",
+        ]
+    )
+    expected = DataFrame(
+        [
+            ["2007/01/01", "01:00", 0.2140, "U", "M"],
+            ["2007/01/01", "02:00", 0.2141, "M", "O"],
+            ["2007/01/01", "04:00", 0.2142, "D", "M"],
+        ],
+        columns=["date", "time", "var", "flag", "oflag"],
+    )
+
+    if parser.engine == "python" and lineterminator == "\r":
+        mark = pytest.mark.xfail(reason="'CR' not respect with the Python parser yet")
+        request.applymarker(mark)
+
+    data = data.replace("\n", lineterminator)
+
+    result = parser.read_csv(
+        StringIO(data),
+        skiprows=1,
+        sep=r"\s+",
+        names=["date", "time", "var", "flag", "oflag"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # AssertionError: DataFrame are different
+def test_skiprows_infield_quote(all_parsers):
+    # see gh-14459
+    parser = all_parsers
+    data = 'a"\nb"\na\n1'
+    expected = DataFrame({"a": [1]})
+
+    result = parser.read_csv(StringIO(data), skiprows=2)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+@pytest.mark.parametrize(
+    "kwargs,expected",
+    [
+        ({}, "1"),
+        ({"header": 0, "names": ["foo"]}, "foo"),
+    ],
+)
+def test_skip_rows_callable(all_parsers, kwargs, expected):
+    parser = all_parsers
+    data = "a\n1\n2\n3\n4\n5"
+
+    result = parser.read_csv(StringIO(data), skiprows=lambda x: x % 2 == 0, **kwargs)
+    expected = DataFrame({expected: [3, 5]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_rows_callable_not_in(all_parsers):
+    parser = all_parsers
+    data = "0,a\n1,b\n2,c\n3,d\n4,e"
+    expected = DataFrame([[1, "b"], [3, "d"]])
+
+    result = parser.read_csv(
+        StringIO(data), header=None, skiprows=lambda x: x not in [1, 3]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_rows_skip_all(all_parsers):
+    parser = all_parsers
+    data = "a\n1\n2\n3\n4\n5"
+    msg = "No columns to parse from file"
+
+    with pytest.raises(EmptyDataError, match=msg):
+        parser.read_csv(StringIO(data), skiprows=lambda x: True)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_rows_bad_callable(all_parsers):
+    msg = "by zero"
+    parser = all_parsers
+    data = "a\n1\n2\n3\n4\n5"
+
+    with pytest.raises(ZeroDivisionError, match=msg):
+        parser.read_csv(StringIO(data), skiprows=lambda x: 1 / 0)
+
+
+@xfail_pyarrow  # ValueError: skiprows argument must be an integer
+def test_skip_rows_and_n_rows(all_parsers):
+    # GH#44021
+    data = """a,b
+1,a
+2,b
+3,c
+4,d
+5,e
+6,f
+7,g
+8,h
+"""
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), nrows=5, skiprows=[2, 4, 6])
+    expected = DataFrame({"a": [1, 3, 5, 7, 8], "b": ["a", "c", "e", "g", "h"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow
+def test_skip_rows_with_chunks(all_parsers):
+    # GH 55677
+    data = """col_a
+10
+20
+30
+40
+50
+60
+70
+80
+90
+100
+"""
+    parser = all_parsers
+    reader = parser.read_csv(
+        StringIO(data), engine=parser, skiprows=lambda x: x in [1, 4, 5], chunksize=4
+    )
+    df1 = next(reader)
+    df2 = next(reader)
+
+    tm.assert_frame_equal(df1, DataFrame({"col_a": [20, 30, 60, 70]}))
+    tm.assert_frame_equal(df2, DataFrame({"col_a": [80, 90, 100]}, index=[4, 5, 6]))
diff --git a/pandas/pandas/tests/io/parser/test_textreader.py b/pandas/pandas/tests/io/parser/test_textreader.py
new file mode 100644
index 0000000000000..eeb783f1957b7
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_textreader.py
@@ -0,0 +1,356 @@
+"""
+Tests the TextReader class in parsers.pyx, which
+is integral to the C engine in parsers.py
+"""
+
+from io import (
+    BytesIO,
+    StringIO,
+)
+
+import numpy as np
+import pytest
+
+import pandas._libs.parsers as parser
+from pandas._libs.parsers import TextReader
+from pandas.errors import ParserWarning
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+from pandas.io.parsers import (
+    TextFileReader,
+    read_csv,
+)
+from pandas.io.parsers.c_parser_wrapper import ensure_dtype_objs
+
+
+class TestTextReader:
+    @pytest.fixture
+    def csv_path(self, datapath):
+        return datapath("io", "data", "csv", "test1.csv")
+
+    def test_file_handle(self, csv_path):
+        with open(csv_path, "rb") as f:
+            reader = TextReader(f)
+            reader.read()
+
+    def test_file_handle_mmap(self, csv_path):
+        # this was never using memory_map=True
+        with open(csv_path, "rb") as f:
+            reader = TextReader(f, header=None)
+            reader.read()
+
+    def test_StringIO(self, csv_path):
+        with open(csv_path, "rb") as f:
+            text = f.read()
+        src = BytesIO(text)
+        reader = TextReader(src, header=None)
+        reader.read()
+
+    def test_encoding_mismatch_warning(self, csv_path):
+        # GH-57954
+        with open(csv_path, encoding="UTF-8") as f:
+            msg = "latin1 is different from the encoding"
+            with pytest.raises(ValueError, match=msg):
+                read_csv(f, encoding="latin1")
+
+    def test_string_factorize(self):
+        # should this be optional?
+        data = "a\nb\na\nb\na"
+        reader = TextReader(StringIO(data), header=None)
+        result = reader.read()
+        assert len(set(map(id, result[0]))) == 2
+
+    def test_skipinitialspace(self):
+        data = "a,   b\na,   b\na,   b\na,   b"
+
+        reader = TextReader(StringIO(data), skipinitialspace=True, header=None)
+        result = reader.read()
+
+        tm.assert_numpy_array_equal(
+            result[0], np.array(["a", "a", "a", "a"], dtype=np.object_)
+        )
+        tm.assert_numpy_array_equal(
+            result[1], np.array(["b", "b", "b", "b"], dtype=np.object_)
+        )
+
+    def test_parse_booleans(self):
+        data = "True\nFalse\nTrue\nTrue"
+
+        reader = TextReader(StringIO(data), header=None)
+        result = reader.read()
+
+        assert result[0].dtype == np.bool_
+
+    def test_delimit_whitespace(self):
+        data = 'a  b\na\t\t "b"\n"a"\t \t b'
+
+        reader = TextReader(StringIO(data), delim_whitespace=True, header=None)
+        result = reader.read()
+
+        tm.assert_numpy_array_equal(
+            result[0], np.array(["a", "a", "a"], dtype=np.object_)
+        )
+        tm.assert_numpy_array_equal(
+            result[1], np.array(["b", "b", "b"], dtype=np.object_)
+        )
+
+    def test_embedded_newline(self):
+        data = 'a\n"hello\nthere"\nthis'
+
+        reader = TextReader(StringIO(data), header=None)
+        result = reader.read()
+
+        expected = np.array(["a", "hello\nthere", "this"], dtype=np.object_)
+        tm.assert_numpy_array_equal(result[0], expected)
+
+    def test_euro_decimal(self):
+        data = "12345,67\n345,678"
+
+        reader = TextReader(StringIO(data), delimiter=":", decimal=",", header=None)
+        result = reader.read()
+
+        expected = np.array([12345.67, 345.678])
+        tm.assert_almost_equal(result[0], expected)
+
+    def test_integer_thousands(self):
+        data = "123,456\n12,500"
+
+        reader = TextReader(StringIO(data), delimiter=":", thousands=",", header=None)
+        result = reader.read()
+
+        expected = np.array([123456, 12500], dtype=np.int64)
+        tm.assert_almost_equal(result[0], expected)
+
+    def test_integer_thousands_alt(self):
+        data = "123.456\n12.500"
+
+        reader = TextFileReader(
+            StringIO(data), delimiter=":", thousands=".", header=None
+        )
+        result = reader.read()
+
+        expected = DataFrame([123456, 12500])
+        tm.assert_frame_equal(result, expected)
+
+    def test_skip_bad_lines(self):
+        # too many lines, see #2430 for why
+        data = "a:b:c\nd:e:f\ng:h:i\nj:k:l:m\nl:m:n\no:p:q:r"
+
+        reader = TextReader(StringIO(data), delimiter=":", header=None)
+        msg = r"Error tokenizing data\. C error: Expected 3 fields in line 4, saw 4"
+        with pytest.raises(parser.ParserError, match=msg):
+            reader.read()
+
+        reader = TextReader(
+            StringIO(data),
+            delimiter=":",
+            header=None,
+            on_bad_lines=2,  # Skip
+        )
+        result = reader.read()
+        expected = {
+            0: np.array(["a", "d", "g", "l"], dtype=object),
+            1: np.array(["b", "e", "h", "m"], dtype=object),
+            2: np.array(["c", "f", "i", "n"], dtype=object),
+        }
+        assert_array_dicts_equal(result, expected)
+
+        with tm.assert_produces_warning(ParserWarning, match="Skipping line"):
+            reader = TextReader(
+                StringIO(data),
+                delimiter=":",
+                header=None,
+                on_bad_lines=1,  # Warn
+            )
+            reader.read()
+
+    def test_header_not_enough_lines(self):
+        data = "skip this\nskip this\na,b,c\n1,2,3\n4,5,6"
+
+        reader = TextReader(StringIO(data), delimiter=",", header=2)
+        header = reader.header
+        expected = [["a", "b", "c"]]
+        assert header == expected
+
+        recs = reader.read()
+        expected = {
+            0: np.array([1, 4], dtype=np.int64),
+            1: np.array([2, 5], dtype=np.int64),
+            2: np.array([3, 6], dtype=np.int64),
+        }
+        assert_array_dicts_equal(recs, expected)
+
+    def test_escapechar(self):
+        data = '\\"hello world"\n\\"hello world"\n\\"hello world"'
+
+        reader = TextReader(StringIO(data), delimiter=",", header=None, escapechar="\\")
+        result = reader.read()
+        expected = {0: np.array(['"hello world"'] * 3, dtype=object)}
+        assert_array_dicts_equal(result, expected)
+
+    def test_eof_has_eol(self):
+        # handling of new line at EOF
+        pass
+
+    def test_na_substitution(self):
+        pass
+
+    def test_numpy_string_dtype(self):
+        data = """\
+a,1
+aa,2
+aaa,3
+aaaa,4
+aaaaa,5"""
+
+        def _make_reader(**kwds):
+            if "dtype" in kwds:
+                kwds["dtype"] = ensure_dtype_objs(kwds["dtype"])
+            return TextReader(StringIO(data), delimiter=",", header=None, **kwds)
+
+        reader = _make_reader(dtype="S5,i4")
+        result = reader.read()
+
+        assert result[0].dtype == "S5"
+
+        ex_values = np.array(["a", "aa", "aaa", "aaaa", "aaaaa"], dtype="S5")
+        assert (result[0] == ex_values).all()
+        assert result[1].dtype == "i4"
+
+        reader = _make_reader(dtype="S4")
+        result = reader.read()
+        assert result[0].dtype == "S4"
+        ex_values = np.array(["a", "aa", "aaa", "aaaa", "aaaa"], dtype="S4")
+        assert (result[0] == ex_values).all()
+        assert result[1].dtype == "S4"
+
+    def test_pass_dtype(self):
+        data = """\
+one,two
+1,a
+2,b
+3,c
+4,d"""
+
+        def _make_reader(**kwds):
+            if "dtype" in kwds:
+                kwds["dtype"] = ensure_dtype_objs(kwds["dtype"])
+            return TextReader(StringIO(data), delimiter=",", **kwds)
+
+        reader = _make_reader(dtype={"one": "u1", 1: "S1"})
+        result = reader.read()
+        assert result[0].dtype == "u1"
+        assert result[1].dtype == "S1"
+
+        reader = _make_reader(dtype={"one": np.uint8, 1: object})
+        result = reader.read()
+        assert result[0].dtype == "u1"
+        assert result[1].dtype == "O"
+
+        reader = _make_reader(dtype={"one": np.dtype("u1"), 1: np.dtype("O")})
+        result = reader.read()
+        assert result[0].dtype == "u1"
+        assert result[1].dtype == "O"
+
+    def test_usecols(self):
+        data = """\
+a,b,c
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+
+        def _make_reader(**kwds):
+            return TextReader(StringIO(data), delimiter=",", **kwds)
+
+        reader = _make_reader(usecols=(1, 2))
+        result = reader.read()
+
+        exp = _make_reader().read()
+        assert len(result) == 2
+        assert (result[1] == exp[1]).all()
+        assert (result[2] == exp[2]).all()
+
+    @pytest.mark.parametrize(
+        "text, kwargs",
+        [
+            ("a,b,c\r1,2,3\r4,5,6\r7,8,9\r10,11,12", {"delimiter": ","}),
+            (
+                "a  b  c\r1  2  3\r4  5  6\r7  8  9\r10  11  12",
+                {"delim_whitespace": True},
+            ),
+            ("a,b,c\r1,2,3\r4,5,6\r,88,9\r10,11,12", {"delimiter": ","}),
+            (
+                (
+                    "A,B,C,D,E,F,G,H,I,J,K,L,M,N,O\r"
+                    "AAAAA,BBBBB,0,0,0,0,0,0,0,0,0,0,0,0,0\r"
+                    ",BBBBB,0,0,0,0,0,0,0,0,0,0,0,0,0"
+                ),
+                {"delimiter": ","},
+            ),
+            ("A  B  C\r  2  3\r4  5  6", {"delim_whitespace": True}),
+            ("A B C\r2 3\r4 5 6", {"delim_whitespace": True}),
+        ],
+    )
+    def test_cr_delimited(self, text, kwargs):
+        nice_text = text.replace("\r", "\r\n")
+        result = TextReader(StringIO(text), **kwargs).read()
+        expected = TextReader(StringIO(nice_text), **kwargs).read()
+        assert_array_dicts_equal(result, expected)
+
+    def test_empty_field_eof(self):
+        data = "a,b,c\n1,2,3\n4,,"
+
+        result = TextReader(StringIO(data), delimiter=",").read()
+
+        expected = {
+            0: np.array([1, 4], dtype=np.int64),
+            1: np.array(["2", ""], dtype=object),
+            2: np.array(["3", ""], dtype=object),
+        }
+        assert_array_dicts_equal(result, expected)
+
+    @pytest.mark.parametrize("repeat", range(10))
+    def test_empty_field_eof_mem_access_bug(self, repeat):
+        # GH5664
+        a = DataFrame([["b"], [np.nan]], columns=["a"], index=["a", "c"])
+        b = DataFrame([[1, 1, 1, 0], [1, 1, 1, 0]], columns=list("abcd"), index=[1, 1])
+        c = DataFrame(
+            [
+                [1, 2, 3, 4],
+                [6, np.nan, np.nan, np.nan],
+                [8, 9, 10, 11],
+                [13, 14, np.nan, np.nan],
+            ],
+            columns=list("abcd"),
+            index=[0, 5, 7, 12],
+        )
+
+        df = read_csv(StringIO("a,b\nc\n"), skiprows=0, names=["a"], engine="c")
+        tm.assert_frame_equal(df, a)
+
+        df = read_csv(
+            StringIO("1,1,1,1,0\n" * 2 + "\n" * 2), names=list("abcd"), engine="c"
+        )
+        tm.assert_frame_equal(df, b)
+
+        df = read_csv(
+            StringIO("0,1,2,3,4\n5,6\n7,8,9,10,11\n12,13,14"),
+            names=list("abcd"),
+            engine="c",
+        )
+        tm.assert_frame_equal(df, c)
+
+    def test_empty_csv_input(self):
+        # GH14867
+        with read_csv(
+            StringIO(), chunksize=20, header=None, names=["a", "b", "c"]
+        ) as df:
+            assert isinstance(df, TextFileReader)
+
+
+def assert_array_dicts_equal(left, right):
+    for k, v in left.items():
+        tm.assert_numpy_array_equal(np.asarray(v), np.asarray(right[k]))
diff --git a/pandas/pandas/tests/io/parser/test_unsupported.py b/pandas/pandas/tests/io/parser/test_unsupported.py
new file mode 100644
index 0000000000000..07f84466e3ac2
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_unsupported.py
@@ -0,0 +1,207 @@
+"""
+Tests that features that are currently unsupported in
+either the Python or C parser are actually enforced
+and are clearly communicated to the user.
+
+Ultimately, the goal is to remove test cases from this
+test suite as new feature support is added to the parsers.
+"""
+
+from io import StringIO
+import os
+from pathlib import Path
+
+import pytest
+
+from pandas.errors import ParserError
+
+import pandas._testing as tm
+
+from pandas.io.parsers import read_csv
+import pandas.io.parsers.readers as parsers
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture(params=["python", "python-fwf"], ids=lambda val: val)
+def python_engine(request):
+    return request.param
+
+
+class TestUnsupportedFeatures:
+    def test_mangle_dupe_cols_false(self):
+        # see gh-12935
+        data = "a b c\n1 2 3"
+
+        for engine in ("c", "python"):
+            with pytest.raises(TypeError, match="unexpected keyword"):
+                read_csv(StringIO(data), engine=engine, mangle_dupe_cols=True)
+
+    def test_c_engine(self):
+        # see gh-6607
+        data = "a b c\n1 2 3"
+        msg = "does not support"
+
+        # specify C engine with unsupported options (raise)
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), engine="c", sep=r"\s")
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), engine="c", sep="\t", quotechar=chr(128))
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), engine="c", skipfooter=1)
+
+        # specify C-unsupported options without python-unsupported options
+        with tm.assert_produces_warning(parsers.ParserWarning):
+            read_csv(StringIO(data), sep=r"\s")
+        with tm.assert_produces_warning(parsers.ParserWarning):
+            read_csv(StringIO(data), sep="\t", quotechar=chr(128))
+        with tm.assert_produces_warning(parsers.ParserWarning):
+            read_csv(StringIO(data), skipfooter=1)
+
+        text = """                      A       B       C       D        E
+one two three   four
+a   b   10.0032 5    -0.5109 -2.3358 -0.4645  0.05076  0.3640
+a   q   20      4     0.4473  1.4152  0.2834  1.00661  0.1744
+x   q   30      3    -0.6662 -0.5243 -0.3580  0.89145  2.5838"""
+        msg = "Error tokenizing data"
+
+        with pytest.raises(ParserError, match=msg):
+            read_csv(StringIO(text), sep="\\s+")
+        with pytest.raises(ParserError, match=msg):
+            read_csv(StringIO(text), engine="c", sep="\\s+")
+
+        msg = "Only length-1 thousands markers supported"
+        data = """A|B|C
+1|2,334|5
+10|13|10.
+"""
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), thousands=",,")
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), thousands="")
+
+        msg = "Only length-1 line terminators supported"
+        data = "a,b,c~~1,2,3~~4,5,6"
+        with pytest.raises(ValueError, match=msg):
+            read_csv(StringIO(data), lineterminator="~~")
+
+    def test_python_engine(self, python_engine):
+        from pandas.io.parsers.readers import _python_unsupported as py_unsupported
+
+        data = """1,2,3,,
+1,2,3,4,
+1,2,3,4,5
+1,2,,,
+1,2,3,4,"""
+
+        for default in py_unsupported:
+            msg = (
+                f"The {default!r} option is not "
+                f"supported with the {python_engine!r} engine"
+            )
+
+            kwargs = {default: object()}
+            with pytest.raises(ValueError, match=msg):
+                read_csv(StringIO(data), engine=python_engine, **kwargs)
+
+    def test_python_engine_file_no_iter(self, python_engine):
+        # see gh-16530
+        class NoNextBuffer:
+            def __init__(self, csv_data) -> None:
+                self.data = csv_data
+
+            def __next__(self):
+                return self.data.__next__()
+
+            def read(self):
+                return self.data
+
+            def readline(self):
+                return self.data
+
+        data = "a\n1"
+        msg = "'NoNextBuffer' object is not iterable|argument 1 must be an iterator"
+
+        with pytest.raises(TypeError, match=msg):
+            read_csv(NoNextBuffer(data), engine=python_engine)
+
+    def test_pyarrow_engine(self):
+        from pandas.io.parsers.readers import _pyarrow_unsupported as pa_unsupported
+
+        data = """1,2,3,,
+        1,2,3,4,
+        1,2,3,4,5
+        1,2,,,
+        1,2,3,4,"""
+
+        for default in pa_unsupported:
+            msg = f"The {default!r} option is not supported with the 'pyarrow' engine"
+            kwargs = {default: object()}
+            default_needs_bool = {"warn_bad_lines", "error_bad_lines"}
+            if default == "dialect":
+                kwargs[default] = "excel"  # test a random dialect
+            elif default in default_needs_bool:
+                kwargs[default] = True
+            elif default == "on_bad_lines":
+                kwargs[default] = "warn"
+
+            with pytest.raises(ValueError, match=msg):
+                read_csv(StringIO(data), engine="pyarrow", **kwargs)
+
+    def test_on_bad_lines_callable_python_or_pyarrow(self, all_parsers):
+        # GH 5686
+        # GH 54643
+        sio = StringIO("a,b\n1,2")
+        bad_lines_func = lambda x: x
+        parser = all_parsers
+        if all_parsers.engine not in ["python", "pyarrow"]:
+            msg = (
+                "on_bad_line can only be a callable "
+                "function if engine='python' or 'pyarrow'"
+            )
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(sio, on_bad_lines=bad_lines_func)
+        else:
+            parser.read_csv(sio, on_bad_lines=bad_lines_func)
+
+
+def test_close_file_handle_on_invalid_usecols(all_parsers):
+    # GH 45384
+    parser = all_parsers
+
+    error = ValueError
+    if parser.engine == "pyarrow":
+        # Raises pyarrow.lib.ArrowKeyError
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    with tm.ensure_clean("test.csv") as fname:
+        Path(fname).write_text("col1,col2\na,b\n1,2", encoding="utf-8")
+        with tm.assert_produces_warning(False):
+            with pytest.raises(error, match="col3"):
+                parser.read_csv(fname, usecols=["col1", "col2", "col3"])
+        # unlink fails on windows if file handles still point to it
+        os.unlink(fname)
+
+
+def test_invalid_file_inputs(request, all_parsers):
+    # GH#45957
+    parser = all_parsers
+    if parser.engine == "python":
+        request.applymarker(
+            pytest.mark.xfail(reason=f"{parser.engine} engine supports lists.")
+        )
+
+    with pytest.raises(ValueError, match="Invalid"):
+        parser.read_csv([])
+
+
+def test_invalid_dtype_backend(all_parsers):
+    parser = all_parsers
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv("test", dtype_backend="numpy")
diff --git a/pandas/pandas/tests/io/parser/test_upcast.py b/pandas/pandas/tests/io/parser/test_upcast.py
new file mode 100644
index 0000000000000..bc4c4c2e24e9c
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/test_upcast.py
@@ -0,0 +1,102 @@
+import numpy as np
+import pytest
+
+from pandas._libs.parsers import (
+    _maybe_upcast,
+    na_values,
+)
+
+import pandas as pd
+from pandas import NA
+import pandas._testing as tm
+from pandas.core.arrays import (
+    ArrowStringArray,
+    BooleanArray,
+    FloatingArray,
+    IntegerArray,
+    StringArray,
+)
+
+
+def test_maybe_upcast(any_real_numpy_dtype):
+    # GH#36712
+
+    dtype = np.dtype(any_real_numpy_dtype)
+    na_value = na_values[dtype]
+    arr = np.array([1, 2, na_value], dtype=dtype)
+    result = _maybe_upcast(arr, use_dtype_backend=True)
+
+    expected_mask = np.array([False, False, True])
+    if issubclass(dtype.type, np.integer):
+        expected = IntegerArray(arr, mask=expected_mask)
+    else:
+        expected = FloatingArray(arr, mask=expected_mask)
+
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_maybe_upcast_no_na(any_real_numpy_dtype):
+    # GH#36712
+    arr = np.array([1, 2, 3], dtype=any_real_numpy_dtype)
+    result = _maybe_upcast(arr, use_dtype_backend=True)
+
+    expected_mask = np.array([False, False, False])
+    if issubclass(np.dtype(any_real_numpy_dtype).type, np.integer):
+        expected = IntegerArray(arr, mask=expected_mask)
+    else:
+        expected = FloatingArray(arr, mask=expected_mask)
+
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_maybe_upcaste_bool():
+    # GH#36712
+    dtype = np.bool_
+    na_value = na_values[dtype]
+    arr = np.array([True, False, na_value], dtype="uint8").view(dtype)
+    result = _maybe_upcast(arr, use_dtype_backend=True)
+
+    expected_mask = np.array([False, False, True])
+    expected = BooleanArray(arr, mask=expected_mask)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_maybe_upcaste_bool_no_nan():
+    # GH#36712
+    dtype = np.bool_
+    arr = np.array([True, False, False], dtype="uint8").view(dtype)
+    result = _maybe_upcast(arr, use_dtype_backend=True)
+
+    expected_mask = np.array([False, False, False])
+    expected = BooleanArray(arr, mask=expected_mask)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_maybe_upcaste_all_nan():
+    # GH#36712
+    dtype = np.int64
+    na_value = na_values[dtype]
+    arr = np.array([na_value, na_value], dtype=dtype)
+    result = _maybe_upcast(arr, use_dtype_backend=True)
+
+    expected_mask = np.array([True, True])
+    expected = IntegerArray(arr, mask=expected_mask)
+    tm.assert_extension_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("val", [na_values[np.object_], "c"])
+def test_maybe_upcast_object(val, string_storage):
+    # GH#36712
+    pa = pytest.importorskip("pyarrow")
+
+    with pd.option_context("mode.string_storage", string_storage):
+        arr = np.array(["a", "b", val], dtype=np.object_)
+        result = _maybe_upcast(arr, use_dtype_backend=True)
+
+        if string_storage == "python":
+            exp_val = "c" if val == "c" else NA
+            expected = StringArray(np.array(["a", "b", exp_val], dtype=np.object_))
+        else:
+            exp_val = "c" if val == "c" else None
+            expected = ArrowStringArray(pa.array(["a", "b", exp_val]))
+        tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/usecols/__init__.py b/pandas/pandas/tests/io/parser/usecols/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/parser/usecols/test_parse_dates.py b/pandas/pandas/tests/io/parser/usecols/test_parse_dates.py
new file mode 100644
index 0000000000000..cc54f2487aa60
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/usecols/test_parse_dates.py
@@ -0,0 +1,87 @@
+"""
+Tests the usecols functionality during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Timestamp,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+_msg_pyarrow_requires_names = (
+    "The pyarrow engine does not allow 'usecols' to be integer column "
+    "positions. Pass a list of string column names instead."
+)
+
+
+@skip_pyarrow  # pyarrow.lib.ArrowKeyError: Column 'fdate' in include_columns
+def test_usecols_with_parse_dates2(all_parsers):
+    # see gh-13604
+    parser = all_parsers
+    data = """2008-02-07 09:40,1032.43
+2008-02-07 09:50,1042.54
+2008-02-07 10:00,1051.65"""
+
+    names = ["date", "values"]
+    usecols = names[:]
+    parse_dates = [0]
+
+    index = Index(
+        [
+            Timestamp("2008-02-07 09:40"),
+            Timestamp("2008-02-07 09:50"),
+            Timestamp("2008-02-07 10:00"),
+        ],
+        name="date",
+    )
+    cols = {"values": [1032.43, 1042.54, 1051.65]}
+    expected = DataFrame(cols, index=index)
+
+    result = parser.read_csv(
+        StringIO(data),
+        parse_dates=parse_dates,
+        index_col=0,
+        usecols=usecols,
+        header=None,
+        names=names,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_with_parse_dates3(all_parsers):
+    # see gh-14792
+    parser = all_parsers
+    data = """a,b,c,d,e,f,g,h,i,j
+2016/09/21,1,1,2,3,4,5,6,7,8"""
+
+    usecols = list("abcdefghij")
+    parse_dates = [0]
+
+    cols = {
+        "a": Timestamp("2016-09-21"),
+        "b": [1],
+        "c": [1],
+        "d": [2],
+        "e": [3],
+        "f": [4],
+        "g": [5],
+        "h": [6],
+        "i": [7],
+        "j": [8],
+    }
+    expected = DataFrame(cols, columns=usecols)
+
+    result = parser.read_csv(StringIO(data), usecols=usecols, parse_dates=parse_dates)
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/usecols/test_strings.py b/pandas/pandas/tests/io/parser/usecols/test_strings.py
new file mode 100644
index 0000000000000..1538bd4e805f7
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/usecols/test_strings.py
@@ -0,0 +1,96 @@
+"""
+Tests the usecols functionality during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+def test_usecols_with_unicode_strings(all_parsers):
+    # see gh-13219
+    data = """AAA,BBB,CCC,DDD
+0.056674973,8,True,a
+2.613230982,2,False,b
+3.568935038,7,False,a"""
+    parser = all_parsers
+
+    exp_data = {
+        "AAA": {
+            0: 0.056674972999999997,
+            1: 2.6132309819999997,
+            2: 3.5689350380000002,
+        },
+        "BBB": {0: 8, 1: 2, 2: 7},
+    }
+    expected = DataFrame(exp_data)
+
+    result = parser.read_csv(StringIO(data), usecols=["AAA", "BBB"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_with_single_byte_unicode_strings(all_parsers):
+    # see gh-13219
+    data = """A,B,C,D
+0.056674973,8,True,a
+2.613230982,2,False,b
+3.568935038,7,False,a"""
+    parser = all_parsers
+
+    exp_data = {
+        "A": {
+            0: 0.056674972999999997,
+            1: 2.6132309819999997,
+            2: 3.5689350380000002,
+        },
+        "B": {0: 8, 1: 2, 2: 7},
+    }
+    expected = DataFrame(exp_data)
+
+    result = parser.read_csv(StringIO(data), usecols=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("usecols", [["AAA", b"BBB"], [b"AAA", "BBB"]])
+def test_usecols_with_mixed_encoding_strings(all_parsers, usecols):
+    data = """AAA,BBB,CCC,DDD
+0.056674973,8,True,a
+2.613230982,2,False,b
+3.568935038,7,False,a"""
+    parser = all_parsers
+    _msg_validate_usecols_arg = (
+        "'usecols' must either be list-like "
+        "of all strings, all unicode, all "
+        "integers or a callable."
+    )
+    with pytest.raises(ValueError, match=_msg_validate_usecols_arg):
+        parser.read_csv(StringIO(data), usecols=usecols)
+
+
+def test_usecols_with_multi_byte_characters(all_parsers):
+    data = """あああ,いい,ううう,ええええ
+0.056674973,8,True,a
+2.613230982,2,False,b
+3.568935038,7,False,a"""
+    parser = all_parsers
+
+    exp_data = {
+        "あああ": {
+            0: 0.056674972999999997,
+            1: 2.6132309819999997,
+            2: 3.5689350380000002,
+        },
+        "いい": {0: 8, 1: 2, 2: 7},
+    }
+    expected = DataFrame(exp_data)
+
+    result = parser.read_csv(StringIO(data), usecols=["あああ", "いい"])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/parser/usecols/test_usecols_basic.py b/pandas/pandas/tests/io/parser/usecols/test_usecols_basic.py
new file mode 100644
index 0000000000000..82b42beb38ae0
--- /dev/null
+++ b/pandas/pandas/tests/io/parser/usecols/test_usecols_basic.py
@@ -0,0 +1,547 @@
+"""
+Tests the usecols functionality during parsing
+for all of the parsers defined in parsers.py
+"""
+
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas.errors import ParserError
+
+from pandas import (
+    DataFrame,
+    Index,
+    array,
+)
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+_msg_validate_usecols_arg = (
+    "'usecols' must either be list-like "
+    "of all strings, all unicode, all "
+    "integers or a callable."
+)
+_msg_validate_usecols_names = (
+    "Usecols do not match columns, columns expected but not found: {0}"
+)
+_msg_pyarrow_requires_names = (
+    "The pyarrow engine does not allow 'usecols' to be integer column "
+    "positions. Pass a list of string column names instead."
+)
+
+xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
+skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame is deprecated:DeprecationWarning"
+)
+
+
+def test_raise_on_mixed_dtype_usecols(all_parsers):
+    # See gh-12678
+    data = """a,b,c
+        1000,2000,3000
+        4000,5000,6000
+        """
+    usecols = [0, "b", 2]
+    parser = all_parsers
+
+    with pytest.raises(ValueError, match=_msg_validate_usecols_arg):
+        parser.read_csv(StringIO(data), usecols=usecols)
+
+
+@pytest.mark.parametrize("usecols", [(1, 2), ("b", "c")])
+def test_usecols(all_parsers, usecols, request):
+    data = """\
+a,b,c
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+    parser = all_parsers
+    if parser.engine == "pyarrow" and isinstance(usecols[0], int):
+        with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
+            parser.read_csv(StringIO(data), usecols=usecols)
+        return
+
+    result = parser.read_csv(StringIO(data), usecols=usecols)
+
+    expected = DataFrame([[2, 3], [5, 6], [8, 9], [11, 12]], columns=["b", "c"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_with_names(all_parsers):
+    data = """\
+a,b,c
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+    parser = all_parsers
+    names = ["foo", "bar"]
+
+    if parser.engine == "pyarrow":
+        with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
+            parser.read_csv(StringIO(data), names=names, usecols=[1, 2], header=0)
+        return
+
+    result = parser.read_csv(StringIO(data), names=names, usecols=[1, 2], header=0)
+
+    expected = DataFrame([[2, 3], [5, 6], [8, 9], [11, 12]], columns=names)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "names,usecols", [(["b", "c"], [1, 2]), (["a", "b", "c"], ["b", "c"])]
+)
+def test_usecols_relative_to_names(all_parsers, names, usecols):
+    data = """\
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+    parser = all_parsers
+    if parser.engine == "pyarrow" and not isinstance(usecols[0], int):
+        # ArrowKeyError: Column 'fb' in include_columns does not exist
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    result = parser.read_csv(StringIO(data), names=names, header=None, usecols=usecols)
+
+    expected = DataFrame([[2, 3], [5, 6], [8, 9], [11, 12]], columns=["b", "c"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_relative_to_names2(all_parsers):
+    # see gh-5766
+    data = """\
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+    parser = all_parsers
+
+    result = parser.read_csv(
+        StringIO(data), names=["a", "b"], header=None, usecols=[0, 1]
+    )
+
+    expected = DataFrame([[1, 2], [4, 5], [7, 8], [10, 11]], columns=["a", "b"])
+    tm.assert_frame_equal(result, expected)
+
+
+# regex mismatch: "Length mismatch: Expected axis has 1 elements"
+@xfail_pyarrow
+def test_usecols_name_length_conflict(all_parsers):
+    data = """\
+1,2,3
+4,5,6
+7,8,9
+10,11,12"""
+    parser = all_parsers
+    msg = "Number of passed names did not match number of header fields in the file"
+    with pytest.raises(ValueError, match=msg):
+        parser.read_csv(StringIO(data), names=["a", "b"], header=None, usecols=[1])
+
+
+def test_usecols_single_string(all_parsers):
+    # see gh-20558
+    parser = all_parsers
+    data = """foo, bar, baz
+1000, 2000, 3000
+4000, 5000, 6000"""
+
+    with pytest.raises(ValueError, match=_msg_validate_usecols_arg):
+        parser.read_csv(StringIO(data), usecols="foo")
+
+
+# ArrowKeyError: Column 'a' in include_columns does not exist in CSV file
+@skip_pyarrow
+@pytest.mark.parametrize(
+    "data", ["a,b,c,d\n1,2,3,4\n5,6,7,8", "a,b,c,d\n1,2,3,4,\n5,6,7,8,"]
+)
+def test_usecols_index_col_false(all_parsers, data):
+    # see gh-9082
+    parser = all_parsers
+    usecols = ["a", "c", "d"]
+    expected = DataFrame({"a": [1, 5], "c": [3, 7], "d": [4, 8]})
+
+    result = parser.read_csv(StringIO(data), usecols=usecols, index_col=False)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("index_col", ["b", 0])
+@pytest.mark.parametrize("usecols", [["b", "c"], [1, 2]])
+def test_usecols_index_col_conflict(all_parsers, usecols, index_col, request):
+    # see gh-4201: test that index_col as integer reflects usecols
+    parser = all_parsers
+    data = "a,b,c,d\nA,a,1,one\nB,b,2,two"
+
+    if parser.engine == "pyarrow" and isinstance(usecols[0], int):
+        with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
+            parser.read_csv(StringIO(data), usecols=usecols, index_col=index_col)
+        return
+
+    expected = DataFrame({"c": [1, 2]}, index=Index(["a", "b"], name="b"))
+
+    result = parser.read_csv(StringIO(data), usecols=usecols, index_col=index_col)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_index_col_conflict2(all_parsers):
+    # see gh-4201: test that index_col as integer reflects usecols
+    parser = all_parsers
+    data = "a,b,c,d\nA,a,1,one\nB,b,2,two"
+
+    expected = DataFrame({"b": ["a", "b"], "c": [1, 2], "d": ("one", "two")})
+    expected = expected.set_index(["b", "c"])
+
+    result = parser.read_csv(
+        StringIO(data), usecols=["b", "c", "d"], index_col=["b", "c"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Expected 3 columns, got 4
+def test_usecols_implicit_index_col(all_parsers):
+    # see gh-2654
+    parser = all_parsers
+    data = "a,b,c\n4,apple,bat,5.7\n8,orange,cow,10"
+
+    result = parser.read_csv(StringIO(data), usecols=["a", "b"])
+    expected = DataFrame({"a": ["apple", "orange"], "b": ["bat", "cow"]}, index=[4, 8])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_index_col_middle(all_parsers):
+    # GH#9098
+    parser = all_parsers
+    data = """a,b,c,d
+1,2,3,4
+"""
+    result = parser.read_csv(StringIO(data), usecols=["b", "c", "d"], index_col="c")
+    expected = DataFrame({"b": [2], "d": [4]}, index=Index([3], name="c"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_index_col_end(all_parsers):
+    # GH#9098
+    parser = all_parsers
+    data = """a,b,c,d
+1,2,3,4
+"""
+    result = parser.read_csv(StringIO(data), usecols=["b", "c", "d"], index_col="d")
+    expected = DataFrame({"b": [2], "c": [3]}, index=Index([4], name="d"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_regex_sep(all_parsers):
+    # see gh-2733
+    parser = all_parsers
+    data = "a  b  c\n4  apple  bat  5.7\n8  orange  cow  10"
+
+    if parser.engine == "pyarrow":
+        msg = "the 'pyarrow' engine does not support regex separators"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), sep=r"\s+", usecols=("a", "b"))
+        return
+
+    result = parser.read_csv(StringIO(data), sep=r"\s+", usecols=("a", "b"))
+
+    expected = DataFrame({"a": ["apple", "orange"], "b": ["bat", "cow"]}, index=[4, 8])
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # Column 'a' in include_columns does not exist in CSV file
+def test_usecols_with_whitespace(all_parsers):
+    parser = all_parsers
+    data = "a  b  c\n4  apple  bat  5.7\n8  orange  cow  10"
+
+    result = parser.read_csv(StringIO(data), sep=r"\s+", usecols=("a", "b"))
+    expected = DataFrame({"a": ["apple", "orange"], "b": ["bat", "cow"]}, index=[4, 8])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "usecols,expected",
+    [
+        # Column selection by index.
+        ([0, 1], DataFrame(data=[[1000, 2000], [4000, 5000]], columns=["2", "0"])),
+        # Column selection by name.
+        (
+            ["0", "1"],
+            DataFrame(data=[[2000, 3000], [5000, 6000]], columns=["0", "1"]),
+        ),
+    ],
+)
+def test_usecols_with_integer_like_header(all_parsers, usecols, expected, request):
+    parser = all_parsers
+    data = """2,0,1
+1000,2000,3000
+4000,5000,6000"""
+
+    if parser.engine == "pyarrow" and isinstance(usecols[0], int):
+        with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
+            parser.read_csv(StringIO(data), usecols=usecols)
+        return
+
+    result = parser.read_csv(StringIO(data), usecols=usecols)
+    tm.assert_frame_equal(result, expected)
+
+
+@xfail_pyarrow  # mismatched shape
+def test_empty_usecols(all_parsers):
+    data = "a,b,c\n1,2,3\n4,5,6"
+    expected = DataFrame(columns=Index([]))
+    parser = all_parsers
+
+    result = parser.read_csv(StringIO(data), usecols=set())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_np_array_usecols(all_parsers):
+    # see gh-12546
+    parser = all_parsers
+    data = "a,b,c\n1,2,3"
+    usecols = np.array(["a", "b"])
+
+    expected = DataFrame([[1, 2]], columns=usecols)
+    result = parser.read_csv(StringIO(data), usecols=usecols)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "usecols,expected",
+    [
+        (
+            lambda x: x.upper() in ["AAA", "BBB", "DDD"],
+            DataFrame(
+                {
+                    "AaA": {
+                        0: 0.056674972999999997,
+                        1: 2.6132309819999997,
+                        2: 3.5689350380000002,
+                    },
+                    "bBb": {0: 8, 1: 2, 2: 7},
+                    "ddd": {0: "a", 1: "b", 2: "a"},
+                }
+            ),
+        ),
+        (lambda x: False, DataFrame(columns=Index([]))),
+    ],
+)
+def test_callable_usecols(all_parsers, usecols, expected):
+    # see gh-14154
+    data = """AaA,bBb,CCC,ddd
+0.056674973,8,True,a
+2.613230982,2,False,b
+3.568935038,7,False,a"""
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine does not allow 'usecols' to be a callable"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), usecols=usecols)
+        return
+
+    result = parser.read_csv(StringIO(data), usecols=usecols)
+    tm.assert_frame_equal(result, expected)
+
+
+# ArrowKeyError: Column 'fa' in include_columns does not exist in CSV file
+@skip_pyarrow
+@pytest.mark.parametrize("usecols", [["a", "c"], lambda x: x in ["a", "c"]])
+def test_incomplete_first_row(all_parsers, usecols):
+    # see gh-6710
+    data = "1,2\n1,2,3"
+    parser = all_parsers
+    names = ["a", "b", "c"]
+    expected = DataFrame({"a": [1, 1], "c": [np.nan, 3]})
+
+    result = parser.read_csv(StringIO(data), names=names, usecols=usecols)
+    tm.assert_frame_equal(result, expected)
+
+
+@skip_pyarrow  # CSV parse error: Expected 3 columns, got 4
+@pytest.mark.parametrize(
+    "data,usecols,kwargs,expected",
+    [
+        # see gh-8985
+        (
+            "19,29,39\n" * 2 + "10,20,30,40",
+            [0, 1, 2],
+            {"header": None},
+            [[19, 29, 39], [19, 29, 39], [10, 20, 30]],
+        ),
+        # see gh-9549
+        (
+            ("A,B,C\n1,2,3\n3,4,5\n1,2,4,5,1,6\n1,2,3,,,1,\n1,2,3\n5,6,7"),
+            ["A", "B", "C"],
+            {},
+            {
+                "A": [1, 3, 1, 1, 1, 5],
+                "B": [2, 4, 2, 2, 2, 6],
+                "C": [3, 5, 4, 3, 3, 7],
+            },
+        ),
+    ],
+)
+def test_uneven_length_cols(all_parsers, data, usecols, kwargs, expected):
+    # see gh-8985
+    parser = all_parsers
+    result = parser.read_csv(StringIO(data), usecols=usecols, **kwargs)
+    expected = DataFrame(expected)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "usecols,kwargs,expected,msg",
+    [
+        (
+            ["a", "b", "c", "d"],
+            {},
+            DataFrame({"a": [1, 5], "b": [2, 6], "c": [3, 7], "d": [4, 8]}),
+            None,
+        ),
+        (
+            ["a", "b", "c", "f"],
+            {},
+            None,
+            _msg_validate_usecols_names.format(r"\['f'\]"),
+        ),
+        (["a", "b", "f"], {}, None, _msg_validate_usecols_names.format(r"\['f'\]")),
+        (
+            ["a", "b", "f", "g"],
+            {},
+            None,
+            _msg_validate_usecols_names.format(r"\[('f', 'g'|'g', 'f')\]"),
+        ),
+        # see gh-14671
+        (
+            None,
+            {"header": 0, "names": ["A", "B", "C", "D"]},
+            DataFrame({"A": [1, 5], "B": [2, 6], "C": [3, 7], "D": [4, 8]}),
+            None,
+        ),
+        (
+            ["A", "B", "C", "f"],
+            {"header": 0, "names": ["A", "B", "C", "D"]},
+            None,
+            _msg_validate_usecols_names.format(r"\['f'\]"),
+        ),
+        (
+            ["A", "B", "f"],
+            {"names": ["A", "B", "C", "D"]},
+            None,
+            _msg_validate_usecols_names.format(r"\['f'\]"),
+        ),
+    ],
+)
+def test_raises_on_usecols_names_mismatch(
+    all_parsers, usecols, kwargs, expected, msg, request
+):
+    data = "a,b,c,d\n1,2,3,4\n5,6,7,8"
+    kwargs.update(usecols=usecols)
+    parser = all_parsers
+
+    if parser.engine == "pyarrow" and not (
+        usecols is not None and expected is not None
+    ):
+        # everything but the first case
+        # ArrowKeyError: Column 'f' in include_columns does not exist in CSV file
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    if expected is None:
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO(data), **kwargs)
+    else:
+        result = parser.read_csv(StringIO(data), **kwargs)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("usecols", [["A", "C"], [0, 2]])
+def test_usecols_subset_names_mismatch_orig_columns(all_parsers, usecols, request):
+    data = "a,b,c,d\n1,2,3,4\n5,6,7,8"
+    names = ["A", "B", "C", "D"]
+    parser = all_parsers
+
+    if parser.engine == "pyarrow":
+        if isinstance(usecols[0], int):
+            with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
+                parser.read_csv(StringIO(data), header=0, names=names, usecols=usecols)
+            return
+        # "pyarrow.lib.ArrowKeyError: Column 'A' in include_columns does not exist"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
+
+    result = parser.read_csv(StringIO(data), header=0, names=names, usecols=usecols)
+    expected = DataFrame({"A": [1, 5], "C": [3, 7]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("names", [None, ["a", "b"]])
+def test_usecols_indices_out_of_bounds(all_parsers, names):
+    # GH#25623 & GH 41130; enforced in 2.0
+    parser = all_parsers
+    data = """
+a,b
+1,2
+    """
+
+    err = ParserError
+    msg = "Defining usecols with out-of-bounds"
+    if parser.engine == "pyarrow":
+        err = ValueError
+        msg = _msg_pyarrow_requires_names
+
+    with pytest.raises(err, match=msg):
+        parser.read_csv(StringIO(data), usecols=[0, 2], names=names, header=0)
+
+
+def test_usecols_additional_columns(all_parsers):
+    # GH#46997
+    parser = all_parsers
+    usecols = lambda header: header.strip() in ["a", "b", "c"]
+
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine does not allow 'usecols' to be a callable"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO("a,b\nx,y,z"), index_col=False, usecols=usecols)
+        return
+    result = parser.read_csv(StringIO("a,b\nx,y,z"), index_col=False, usecols=usecols)
+    expected = DataFrame({"a": ["x"], "b": "y"})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_additional_columns_integer_columns(all_parsers):
+    # GH#46997
+    parser = all_parsers
+    usecols = lambda header: header.strip() in ["0", "1"]
+    if parser.engine == "pyarrow":
+        msg = "The pyarrow engine does not allow 'usecols' to be a callable"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(StringIO("0,1\nx,y,z"), index_col=False, usecols=usecols)
+        return
+    result = parser.read_csv(StringIO("0,1\nx,y,z"), index_col=False, usecols=usecols)
+    expected = DataFrame({"0": ["x"], "1": "y"})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_usecols_dtype(all_parsers):
+    parser = all_parsers
+    data = """
+col1,col2,col3
+a,1,x
+b,2,y
+"""
+    result = parser.read_csv(
+        StringIO(data),
+        usecols=["col1", "col2"],
+        dtype={"col1": "string", "col2": "uint8", "col3": "string"},
+    )
+    expected = DataFrame(
+        {"col1": array(["a", "b"]), "col2": np.array([1, 2], dtype="uint8")}
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/pytables/__init__.py b/pandas/pandas/tests/io/pytables/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/pytables/common.py b/pandas/pandas/tests/io/pytables/common.py
new file mode 100644
index 0000000000000..62582b212eb38
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/common.py
@@ -0,0 +1,50 @@
+from collections.abc import Generator
+from contextlib import contextmanager
+import pathlib
+import tempfile
+
+import pytest
+
+from pandas.io.pytables import HDFStore
+
+tables = pytest.importorskip("tables")
+# set these parameters so we don't have file sharing
+tables.parameters.MAX_NUMEXPR_THREADS = 1
+tables.parameters.MAX_BLOSC_THREADS = 1
+tables.parameters.MAX_THREADS = 1
+
+
+def safe_close(store):
+    try:
+        if store is not None:
+            store.close()
+    except OSError:
+        pass
+
+
+# contextmanager to ensure the file cleanup
+@contextmanager
+def ensure_clean_store(
+    path, mode="a", complevel=None, complib=None, fletcher32=False
+) -> Generator[HDFStore, None, None]:
+    with tempfile.TemporaryDirectory() as tmpdirname:
+        tmp_path = pathlib.Path(tmpdirname, path)
+        with HDFStore(
+            tmp_path,
+            mode=mode,
+            complevel=complevel,
+            complib=complib,
+            fletcher32=fletcher32,
+        ) as store:
+            yield store
+
+
+def _maybe_remove(store, key):
+    """
+    For tests using tables, try removing the table to be sure there is
+    no content from previous tests using the same table name.
+    """
+    try:
+        store.remove(key)
+    except (ValueError, KeyError):
+        pass
diff --git a/pandas/pandas/tests/io/pytables/conftest.py b/pandas/pandas/tests/io/pytables/conftest.py
new file mode 100644
index 0000000000000..466e4ae8bb99c
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/conftest.py
@@ -0,0 +1,9 @@
+import uuid
+
+import pytest
+
+
+@pytest.fixture
+def setup_path():
+    """Fixture for setup path"""
+    return f"tmp.__{uuid.uuid4()}__.h5"
diff --git a/pandas/pandas/tests/io/pytables/test_append.py b/pandas/pandas/tests/io/pytables/test_append.py
new file mode 100644
index 0000000000000..479f2468a86ab
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_append.py
@@ -0,0 +1,1023 @@
+import datetime
+from datetime import timedelta
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas.compat import PY312
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    _testing as tm,
+    concat,
+    date_range,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+pytestmark = [pytest.mark.single_cpu]
+
+tables = pytest.importorskip("tables")
+
+
+@pytest.mark.filterwarnings("ignore::tables.NaturalNameWarning")
+def test_append(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # this is allowed by almost always don't want to do it
+        # tables.NaturalNameWarning):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=20, freq="B"),
+        )
+        _maybe_remove(store, "df1")
+        store.append("df1", df[:10])
+        store.append("df1", df[10:])
+        tm.assert_frame_equal(store["df1"], df)
+
+        _maybe_remove(store, "df2")
+        store.put("df2", df[:10], format="table")
+        store.append("df2", df[10:])
+        tm.assert_frame_equal(store["df2"], df)
+
+        _maybe_remove(store, "df3")
+        store.append("/df3", df[:10])
+        store.append("/df3", df[10:])
+        tm.assert_frame_equal(store["df3"], df)
+
+        # this is allowed by almost always don't want to do it
+        # tables.NaturalNameWarning
+        _maybe_remove(store, "/df3 foo")
+        store.append("/df3 foo", df[:10])
+        store.append("/df3 foo", df[10:])
+        tm.assert_frame_equal(store["df3 foo"], df)
+
+        # dtype issues - mizxed type in a single object column
+        df = DataFrame(data=[[1, 2], [0, 1], [1, 2], [0, 0]])
+        df["mixed_column"] = "testing"
+        df.loc[2, "mixed_column"] = np.nan
+        _maybe_remove(store, "df")
+        store.append("df", df)
+        tm.assert_frame_equal(store["df"], df)
+
+        # uints - test storage of uints
+        uint_data = DataFrame(
+            {
+                "u08": Series(
+                    np.random.default_rng(2).integers(0, high=255, size=5),
+                    dtype=np.uint8,
+                ),
+                "u16": Series(
+                    np.random.default_rng(2).integers(0, high=65535, size=5),
+                    dtype=np.uint16,
+                ),
+                "u32": Series(
+                    np.random.default_rng(2).integers(0, high=2**30, size=5),
+                    dtype=np.uint32,
+                ),
+                "u64": Series(
+                    [2**58, 2**59, 2**60, 2**61, 2**62],
+                    dtype=np.uint64,
+                ),
+            },
+            index=np.arange(5),
+        )
+        _maybe_remove(store, "uints")
+        store.append("uints", uint_data)
+        tm.assert_frame_equal(store["uints"], uint_data, check_index_type=True)
+
+        # uints - test storage of uints in indexable columns
+        _maybe_remove(store, "uints")
+        # 64-bit indices not yet supported
+        store.append("uints", uint_data, data_columns=["u08", "u16", "u32"])
+        tm.assert_frame_equal(store["uints"], uint_data, check_index_type=True)
+
+
+def test_append_series(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # basic
+        ss = Series(range(20), dtype=np.float64, index=[f"i_{i}" for i in range(20)])
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        ns = Series(np.arange(100))
+
+        store.append("ss", ss)
+        result = store["ss"]
+        tm.assert_series_equal(result, ss)
+        assert result.name is None
+
+        store.append("ts", ts)
+        result = store["ts"]
+        tm.assert_series_equal(result, ts)
+        assert result.name is None
+
+        ns.name = "foo"
+        store.append("ns", ns)
+        result = store["ns"]
+        tm.assert_series_equal(result, ns)
+        assert result.name == ns.name
+
+        # select on the values
+        expected = ns[ns > 60]
+        result = store.select("ns", "foo>60")
+        tm.assert_series_equal(result, expected)
+
+        # select on the index and values
+        expected = ns[(ns > 70) & (ns.index < 90)]
+        # Reading/writing RangeIndex info is not supported yet
+        expected.index = Index(expected.index._data)
+        result = store.select("ns", "foo>70 and index<90")
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        # multi-index
+        mi = DataFrame(np.random.default_rng(2).standard_normal((5, 1)), columns=["A"])
+        mi["B"] = np.arange(len(mi))
+        mi["C"] = "foo"
+        mi.loc[3:5, "C"] = "bar"
+        mi.set_index(["C", "B"], inplace=True)
+        s = mi.stack()
+        s.index = s.index.droplevel(2)
+        store.append("mi", s)
+        tm.assert_series_equal(store["mi"], s, check_index_type=True)
+
+
+def test_append_some_nans(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            {
+                "A": Series(np.random.default_rng(2).standard_normal(20)).astype(
+                    "int32"
+                ),
+                "A1": np.random.default_rng(2).standard_normal(20),
+                "A2": np.random.default_rng(2).standard_normal(20),
+                "B": "foo",
+                "C": "bar",
+                "D": Timestamp("2001-01-01").as_unit("ns"),
+                "E": Timestamp("2001-01-02").as_unit("ns"),
+            },
+            index=np.arange(20),
+        )
+        # some nans
+        _maybe_remove(store, "df1")
+        df.loc[0:15, ["A1", "B", "D", "E"]] = np.nan
+        store.append("df1", df[:10])
+        store.append("df1", df[10:])
+        tm.assert_frame_equal(store["df1"], df, check_index_type=True)
+
+        # first column
+        df1 = df.copy()
+        df1["A1"] = np.nan
+        _maybe_remove(store, "df1")
+        store.append("df1", df1[:10])
+        store.append("df1", df1[10:])
+        tm.assert_frame_equal(store["df1"], df1, check_index_type=True)
+
+        # 2nd column
+        df2 = df.copy()
+        df2["A2"] = np.nan
+        _maybe_remove(store, "df2")
+        store.append("df2", df2[:10])
+        store.append("df2", df2[10:])
+        tm.assert_frame_equal(store["df2"], df2, check_index_type=True)
+
+        # datetimes
+        df3 = df.copy()
+        df3["E"] = np.nan
+        _maybe_remove(store, "df3")
+        store.append("df3", df3[:10])
+        store.append("df3", df3[10:])
+        tm.assert_frame_equal(store["df3"], df3, check_index_type=True)
+
+
+def test_append_all_nans(setup_path, using_infer_string):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            {
+                "A1": np.random.default_rng(2).standard_normal(20),
+                "A2": np.random.default_rng(2).standard_normal(20),
+            },
+            index=np.arange(20),
+        )
+        df.loc[0:15, :] = np.nan
+
+        # nan some entire rows (dropna=True)
+        _maybe_remove(store, "df")
+        store.append("df", df[:10], dropna=True)
+        store.append("df", df[10:], dropna=True)
+        tm.assert_frame_equal(store["df"], df[-4:], check_index_type=True)
+
+        # nan some entire rows (dropna=False)
+        _maybe_remove(store, "df2")
+        store.append("df2", df[:10], dropna=False)
+        store.append("df2", df[10:], dropna=False)
+        tm.assert_frame_equal(store["df2"], df, check_index_type=True)
+
+        # tests the option io.hdf.dropna_table
+        with pd.option_context("io.hdf.dropna_table", False):
+            _maybe_remove(store, "df3")
+            store.append("df3", df[:10])
+            store.append("df3", df[10:])
+            tm.assert_frame_equal(store["df3"], df)
+
+        with pd.option_context("io.hdf.dropna_table", True):
+            _maybe_remove(store, "df4")
+            store.append("df4", df[:10])
+            store.append("df4", df[10:])
+            tm.assert_frame_equal(store["df4"], df[-4:])
+
+            # nan some entire rows (string are still written!)
+            df = DataFrame(
+                {
+                    "A1": np.random.default_rng(2).standard_normal(20),
+                    "A2": np.random.default_rng(2).standard_normal(20),
+                    "B": "foo",
+                    "C": "bar",
+                },
+                index=np.arange(20),
+            )
+
+            df.loc[0:15, :] = np.nan
+
+            _maybe_remove(store, "df")
+            store.append("df", df[:10], dropna=True)
+            store.append("df", df[10:], dropna=True)
+            result = store["df"]
+            expected = df
+            if using_infer_string:
+                # TODO: Test is incorrect when not using_infer_string.
+                #       Should take the last 4 rows uncondiationally.
+                expected = expected[-4:]
+            tm.assert_frame_equal(result, expected, check_index_type=True)
+
+            _maybe_remove(store, "df2")
+            store.append("df2", df[:10], dropna=False)
+            store.append("df2", df[10:], dropna=False)
+            tm.assert_frame_equal(store["df2"], df, check_index_type=True)
+
+            # nan some entire rows (but since we have dates they are still
+            # written!)
+            df = DataFrame(
+                {
+                    "A1": np.random.default_rng(2).standard_normal(20),
+                    "A2": np.random.default_rng(2).standard_normal(20),
+                    "B": "foo",
+                    "C": "bar",
+                    "D": Timestamp("2001-01-01").as_unit("ns"),
+                    "E": Timestamp("2001-01-02").as_unit("ns"),
+                },
+                index=np.arange(20),
+            )
+
+            df.loc[0:15, :] = np.nan
+
+            _maybe_remove(store, "df")
+            store.append("df", df[:10], dropna=True)
+            store.append("df", df[10:], dropna=True)
+            tm.assert_frame_equal(store["df"], df, check_index_type=True)
+
+            _maybe_remove(store, "df2")
+            store.append("df2", df[:10], dropna=False)
+            store.append("df2", df[10:], dropna=False)
+            tm.assert_frame_equal(store["df2"], df, check_index_type=True)
+
+
+def test_append_frame_column_oriented(setup_path, request):
+    with ensure_clean_store(setup_path) as store:
+        # column oriented
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df.index = df.index._with_freq(None)  # freq doesn't round-trip
+
+        _maybe_remove(store, "df1")
+        store.append("df1", df.iloc[:, :2], axes=["columns"])
+        store.append("df1", df.iloc[:, 2:])
+        tm.assert_frame_equal(store["df1"], df)
+
+        result = store.select("df1", "columns=A")
+        expected = df.reindex(columns=["A"])
+        tm.assert_frame_equal(expected, result)
+
+        # selection on the non-indexable
+        request.applymarker(
+            pytest.mark.xfail(
+                PY312,
+                reason="AST change in PY312",
+                raises=ValueError,
+            )
+        )
+        result = store.select("df1", ("columns=A", "index=df.index[0:4]"))
+        expected = df.reindex(columns=["A"], index=df.index[0:4])
+        tm.assert_frame_equal(expected, result)
+
+        # this isn't supported
+        msg = re.escape(
+            "passing a filterable condition to a non-table indexer "
+            "[Filter: Not Initialized]"
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.select("df1", "columns=A and index>df.index[4]")
+
+
+def test_append_with_different_block_ordering(setup_path):
+    # GH 4096; using same frames, but different block orderings
+    with ensure_clean_store(setup_path) as store:
+        for i in range(10):
+            df = DataFrame(
+                np.random.default_rng(2).standard_normal((10, 2)), columns=list("AB")
+            )
+            df["index"] = range(10)
+            df["index"] += i * 10
+            df["int64"] = Series([1] * len(df), dtype="int64")
+            df["int16"] = Series([1] * len(df), dtype="int16")
+
+            if i % 2 == 0:
+                del df["int64"]
+                df["int64"] = Series([1] * len(df), dtype="int64")
+            if i % 3 == 0:
+                a = df.pop("A")
+                df["A"] = a
+
+            df.set_index("index", inplace=True)
+
+            store.append("df", df)
+
+    # test a different ordering but with more fields (like invalid
+    # combinations)
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            columns=list("AB"),
+            dtype="float64",
+        )
+        df["int64"] = Series([1] * len(df), dtype="int64")
+        df["int16"] = Series([1] * len(df), dtype="int16")
+        store.append("df", df)
+
+        # store additional fields in different blocks
+        df["int16_2"] = Series([1] * len(df), dtype="int16")
+        msg = re.escape(
+            "cannot match existing table structure for [int16] on appending data"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+        # store multiple additional fields in different blocks
+        df["float_3"] = Series([1.0] * len(df), dtype="float64")
+        msg = re.escape(
+            "cannot match existing table structure for [A,B] on appending data"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+
+def test_append_with_strings(setup_path):
+    with ensure_clean_store(setup_path) as store:
+
+        def check_col(key, name, size):
+            assert (
+                getattr(store.get_storer(key).table.description, name).itemsize == size
+            )
+
+        # avoid truncation on elements
+        df = DataFrame([[123, "asdqwerty"], [345, "dggnhebbsdfbdfb"]])
+        store.append("df_big", df)
+        tm.assert_frame_equal(store.select("df_big"), df)
+        check_col("df_big", "values_block_1", 15)
+
+        # appending smaller string ok
+        df2 = DataFrame([[124, "asdqy"], [346, "dggnhefbdfb"]])
+        store.append("df_big", df2)
+        expected = concat([df, df2])
+        tm.assert_frame_equal(store.select("df_big"), expected)
+        check_col("df_big", "values_block_1", 15)
+
+        # avoid truncation on elements
+        df = DataFrame([[123, "asdqwerty"], [345, "dggnhebbsdfbdfb"]])
+        store.append("df_big2", df, min_itemsize={"values": 50})
+        tm.assert_frame_equal(store.select("df_big2"), df)
+        check_col("df_big2", "values_block_1", 50)
+
+        # bigger string on next append
+        store.append("df_new", df)
+        df_new = DataFrame([[124, "abcdefqhij"], [346, "abcdefghijklmnopqrtsuvwxyz"]])
+        msg = (
+            r"Trying to store a string with len \[26\] in "
+            r"\[values_block_1\] column but\n"
+            r"this column has a limit of \[15\]!\n"
+            "Consider using min_itemsize to preset the sizes on these "
+            "columns"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df_new", df_new)
+
+        # min_itemsize on Series index (GH 11412)
+        df = DataFrame(
+            {
+                "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+                "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+                "C": Index(["foo1", "foo2", "foo3", "foo4", "foo5"]),
+                "D": date_range("20130101", periods=5),
+            }
+        ).set_index("C")
+        store.append("ss", df["B"], min_itemsize={"index": 4})
+        tm.assert_series_equal(store.select("ss"), df["B"])
+
+        # same as above, with data_columns=True
+        store.append("ss2", df["B"], data_columns=True, min_itemsize={"index": 4})
+        tm.assert_series_equal(store.select("ss2"), df["B"])
+
+        # min_itemsize in index without appending (GH 10381)
+        store.put("ss3", df, format="table", min_itemsize={"index": 6})
+        # just make sure there is a longer string:
+        df2 = df.copy().reset_index().assign(C="longer").set_index("C")
+        store.append("ss3", df2)
+        tm.assert_frame_equal(store.select("ss3"), concat([df, df2]))
+
+        # same as above, with a Series
+        store.put("ss4", df["B"], format="table", min_itemsize={"index": 6})
+        store.append("ss4", df2["B"])
+        tm.assert_series_equal(store.select("ss4"), concat([df["B"], df2["B"]]))
+
+        # with nans
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df["string"] = "foo"
+        df.loc[df.index[1:4], "string"] = np.nan
+        df["string2"] = "bar"
+        df.loc[df.index[4:8], "string2"] = np.nan
+        df["string3"] = "bah"
+        df.loc[df.index[1:], "string3"] = np.nan
+        store.append("df", df)
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame({"A": "foo", "B": "bar"}, index=range(10))
+
+        # a min_itemsize that creates a data_column
+        _maybe_remove(store, "df")
+        store.append("df", df, min_itemsize={"A": 200})
+        check_col("df", "A", 200)
+        assert store.get_storer("df").data_columns == ["A"]
+
+        # a min_itemsize that creates a data_column2
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["B"], min_itemsize={"A": 200})
+        check_col("df", "A", 200)
+        assert store.get_storer("df").data_columns == ["B", "A"]
+
+        # a min_itemsize that creates a data_column2
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["B"], min_itemsize={"values": 200})
+        check_col("df", "B", 200)
+        check_col("df", "values_block_0", 200)
+        assert store.get_storer("df").data_columns == ["B"]
+
+        # infer the .typ on subsequent appends
+        _maybe_remove(store, "df")
+        store.append("df", df[:5], min_itemsize=200)
+        store.append("df", df[5:], min_itemsize=200)
+        tm.assert_frame_equal(store["df"], df)
+
+        # invalid min_itemsize keys
+        df = DataFrame(["foo", "foo", "foo", "barh", "barh", "barh"], columns=["A"])
+        _maybe_remove(store, "df")
+        msg = re.escape(
+            "min_itemsize has the key [foo] which is not an axis or data_column"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df, min_itemsize={"foo": 20, "foobar": 20})
+
+
+def test_append_with_empty_string(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # with all empty strings (GH 12242)
+        df = DataFrame({"x": ["a", "b", "c", "d", "e", "f", ""]})
+        store.append("df", df[:-1], min_itemsize={"x": 1})
+        store.append("df", df[-1:], min_itemsize={"x": 1})
+        tm.assert_frame_equal(store.select("df"), df)
+
+
+def test_append_with_data_columns(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df.iloc[0, df.columns.get_loc("B")] = 1.0
+        _maybe_remove(store, "df")
+        store.append("df", df[:2], data_columns=["B"])
+        store.append("df", df[2:])
+        tm.assert_frame_equal(store["df"], df)
+
+        # check that we have indices created
+        assert store._handle.root.df.table.cols.index.is_indexed is True
+        assert store._handle.root.df.table.cols.B.is_indexed is True
+
+        # data column searching
+        result = store.select("df", "B>0")
+        expected = df[df.B > 0]
+        tm.assert_frame_equal(result, expected)
+
+        # data column searching (with an indexable and a data_columns)
+        result = store.select("df", "B>0 and index>df.index[3]")
+        df_new = df.reindex(index=df.index[4:])
+        expected = df_new[df_new.B > 0]
+        tm.assert_frame_equal(result, expected)
+
+        # data column selection with a string data_column
+        df_new = df.copy()
+        df_new["string"] = "foo"
+        df_new.loc[df_new.index[1:4], "string"] = np.nan
+        df_new.loc[df_new.index[5:6], "string"] = "bar"
+        _maybe_remove(store, "df")
+        store.append("df", df_new, data_columns=["string"])
+        result = store.select("df", "string='foo'")
+        expected = df_new[df_new.string == "foo"]
+        tm.assert_frame_equal(result, expected)
+
+        # using min_itemsize and a data column
+        def check_col(key, name, size):
+            assert (
+                getattr(store.get_storer(key).table.description, name).itemsize == size
+            )
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+        store.append("df", df_new, data_columns=["string"], min_itemsize={"string": 30})
+        check_col("df", "string", 30)
+        _maybe_remove(store, "df")
+        store.append("df", df_new, data_columns=["string"], min_itemsize=30)
+        check_col("df", "string", 30)
+        _maybe_remove(store, "df")
+        store.append("df", df_new, data_columns=["string"], min_itemsize={"values": 30})
+        check_col("df", "string", 30)
+
+    with ensure_clean_store(setup_path) as store:
+        df_new["string2"] = "foobarbah"
+        df_new["string_block1"] = "foobarbah1"
+        df_new["string_block2"] = "foobarbah2"
+        _maybe_remove(store, "df")
+        store.append(
+            "df",
+            df_new,
+            data_columns=["string", "string2"],
+            min_itemsize={"string": 30, "string2": 40, "values": 50},
+        )
+        check_col("df", "string", 30)
+        check_col("df", "string2", 40)
+        check_col("df", "values_block_1", 50)
+
+    with ensure_clean_store(setup_path) as store:
+        # multiple data columns
+        df_new = df.copy()
+        df_new.iloc[0, df_new.columns.get_loc("A")] = 1.0
+        df_new.iloc[0, df_new.columns.get_loc("B")] = -1.0
+        df_new["string"] = "foo"
+
+        sl = df_new.columns.get_loc("string")
+        df_new.iloc[1:4, sl] = np.nan
+        df_new.iloc[5:6, sl] = "bar"
+
+        df_new["string2"] = "foo"
+        sl = df_new.columns.get_loc("string2")
+        df_new.iloc[2:5, sl] = np.nan
+        df_new.iloc[7:8, sl] = "bar"
+        _maybe_remove(store, "df")
+        store.append("df", df_new, data_columns=["A", "B", "string", "string2"])
+        result = store.select("df", "string='foo' and string2='foo' and A>0 and B<0")
+        expected = df_new[
+            (df_new.string == "foo")
+            & (df_new.string2 == "foo")
+            & (df_new.A > 0)
+            & (df_new.B < 0)
+        ]
+        tm.assert_frame_equal(result, expected, check_freq=False)
+        # FIXME: 2020-05-07 freq check randomly fails in the CI
+
+        # yield an empty frame
+        result = store.select("df", "string='foo' and string2='cool'")
+        expected = df_new[(df_new.string == "foo") & (df_new.string2 == "cool")]
+        tm.assert_frame_equal(result, expected)
+
+    with ensure_clean_store(setup_path) as store:
+        # doc example
+        df_dc = df.copy()
+        df_dc["string"] = "foo"
+        df_dc.loc[df_dc.index[4:6], "string"] = np.nan
+        df_dc.loc[df_dc.index[7:9], "string"] = "bar"
+        df_dc["string2"] = "cool"
+        df_dc["datetime"] = Timestamp("20010102").as_unit("ns")
+        df_dc.loc[df_dc.index[3:5], ["A", "B", "datetime"]] = np.nan
+
+        _maybe_remove(store, "df_dc")
+        store.append(
+            "df_dc", df_dc, data_columns=["B", "C", "string", "string2", "datetime"]
+        )
+        result = store.select("df_dc", "B>0")
+
+        expected = df_dc[df_dc.B > 0]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df_dc", ["B > 0", "C > 0", "string == foo"])
+        expected = df_dc[(df_dc.B > 0) & (df_dc.C > 0) & (df_dc.string == "foo")]
+        tm.assert_frame_equal(result, expected, check_freq=False)
+        # FIXME: 2020-12-07 intermittent build failures here with freq of
+        #  None instead of BDay(4)
+
+    with ensure_clean_store(setup_path) as store:
+        # doc example part 2
+
+        index = date_range("1/1/2000", periods=8)
+        df_dc = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 3)),
+            index=index,
+            columns=["A", "B", "C"],
+        )
+        df_dc["string"] = "foo"
+        df_dc.loc[df_dc.index[4:6], "string"] = np.nan
+        df_dc.loc[df_dc.index[7:9], "string"] = "bar"
+        df_dc[["B", "C"]] = df_dc[["B", "C"]].abs()
+        df_dc["string2"] = "cool"
+
+        # on-disk operations
+        store.append("df_dc", df_dc, data_columns=["B", "C", "string", "string2"])
+
+        result = store.select("df_dc", "B>0")
+        expected = df_dc[df_dc.B > 0]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df_dc", ["B > 0", "C > 0", 'string == "foo"'])
+        expected = df_dc[(df_dc.B > 0) & (df_dc.C > 0) & (df_dc.string == "foo")]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_append_hierarchical(tmp_path, setup_path, multiindex_dataframe_random_data):
+    df = multiindex_dataframe_random_data
+    df.columns.name = None
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("mi", df)
+        result = store.select("mi")
+        tm.assert_frame_equal(result, df)
+
+        # GH 3748
+        result = store.select("mi", columns=["A", "B"])
+        expected = df.reindex(columns=["A", "B"])
+        tm.assert_frame_equal(result, expected)
+
+    path = tmp_path / "test.hdf"
+    df.to_hdf(path, key="df", format="table")
+    result = read_hdf(path, "df", columns=["A", "B"])
+    expected = df.reindex(columns=["A", "B"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_append_misc(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        store.append("df", df, chunksize=1)
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+        store.append("df1", df, expectedrows=10)
+        result = store.select("df1")
+        tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("chunksize", [10, 200, 1000])
+def test_append_misc_chunksize(setup_path, chunksize):
+    # more chunksize in append tests
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    df["string"] = "foo"
+    df["float322"] = 1.0
+    df["float322"] = df["float322"].astype("float32")
+    df["bool"] = df["float322"] > 0
+    df["time1"] = Timestamp("20130101").as_unit("ns")
+    df["time2"] = Timestamp("20130102").as_unit("ns")
+    with ensure_clean_store(setup_path, mode="w") as store:
+        store.append("obj", df, chunksize=chunksize)
+        result = store.select("obj")
+        tm.assert_frame_equal(result, df)
+
+
+def test_append_misc_empty_frame(setup_path):
+    # empty frame, GH4273
+    with ensure_clean_store(setup_path) as store:
+        # 0 len
+        df_empty = DataFrame(columns=list("ABC"))
+        store.append("df", df_empty)
+        with pytest.raises(KeyError, match="'No object named df in the file'"):
+            store.select("df")
+
+        # repeated append of 0/non-zero frames
+        df = DataFrame(np.random.default_rng(2).random((10, 3)), columns=list("ABC"))
+        store.append("df", df)
+        tm.assert_frame_equal(store.select("df"), df)
+        store.append("df", df_empty)
+        tm.assert_frame_equal(store.select("df"), df)
+
+        # store
+        df = DataFrame(columns=list("ABC"))
+        store.put("df2", df)
+        tm.assert_frame_equal(store.select("df2"), df)
+
+
+def test_append_raise(setup_path, using_infer_string):
+    with ensure_clean_store(setup_path) as store:
+        # test append with invalid input to get good error messages
+
+        # list in column
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        df["invalid"] = [["a"]] * len(df)
+        assert df.dtypes["invalid"] == np.object_
+        msg = re.escape(
+            """Cannot serialize the column [invalid]
+because its data contents are not [string] but [mixed] object dtype"""
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", df)
+
+        # multiple invalid columns
+        df["invalid2"] = [["a"]] * len(df)
+        df["invalid3"] = [["a"]] * len(df)
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", df)
+
+        # datetime with embedded nans as object
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        s = Series(datetime.datetime(2001, 1, 2), index=df.index)
+        s = s.astype(object)
+        s[0:5] = np.nan
+        df["invalid"] = s
+        assert df.dtypes["invalid"] == np.object_
+        msg = "too many timezones in this block, create separate data columns"
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", df)
+
+        # directly ndarray
+        msg = "value must be None, Series, or DataFrame"
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", np.arange(10))
+
+        # series directly
+        msg = re.escape(
+            "cannot properly create the storer for: "
+            "[group->df,value-><class 'pandas.Series'>]"
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", Series(np.arange(10)))
+
+        # appending an incompatible table
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        store.append("df", df)
+
+        df["foo"] = "foo"
+        msg = re.escape(
+            "invalid combination of [non_index_axes] on appending data "
+            "[(1, ['A', 'B', 'C', 'D', 'foo'])] vs current table "
+            "[(1, ['A', 'B', 'C', 'D'])]"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+        # incompatible type (GH 41897)
+        _maybe_remove(store, "df")
+        df["foo"] = Timestamp("20130101")
+        store.append("df", df)
+        df["foo"] = "bar"
+        msg = re.escape(
+            "Cannot serialize the column [foo] "
+            "because its data contents are not [string] "
+            "but [datetime64[s]] object dtype"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+
+def test_append_with_timedelta(setup_path):
+    # GH 3577
+    # append timedelta
+
+    ts = Timestamp("20130101").as_unit("ns")
+    df = DataFrame(
+        {
+            "A": ts,
+            "B": [ts + timedelta(days=i, seconds=10) for i in range(10)],
+        }
+    )
+    df["C"] = df["A"] - df["B"]
+    df.loc[3:5, "C"] = np.nan
+
+    with ensure_clean_store(setup_path) as store:
+        # table
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=True)
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+        result = store.select("df", where="C<100000")
+        tm.assert_frame_equal(result, df)
+
+        result = store.select("df", where="C<pd.Timedelta('-3D')")
+        tm.assert_frame_equal(result, df.iloc[3:])
+
+        result = store.select("df", "C<'-3D'")
+        tm.assert_frame_equal(result, df.iloc[3:])
+
+        # a bit hacky here as we don't really deal with the NaT properly
+
+        result = store.select("df", "C<'-500000s'")
+        result = result.dropna(subset=["C"])
+        tm.assert_frame_equal(result, df.iloc[6:])
+
+        result = store.select("df", "C<'-3.5D'")
+        result = result.iloc[1:]
+        tm.assert_frame_equal(result, df.iloc[4:])
+
+        # fixed
+        _maybe_remove(store, "df2")
+        store.put("df2", df)
+        result = store.select("df2")
+        tm.assert_frame_equal(result, df)
+
+
+def test_append_to_multiple(setup_path):
+    df1 = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df2 = df1.copy().rename(columns="{}_2".format)
+    df2["foo"] = "bar"
+    df = concat([df1, df2], axis=1)
+
+    with ensure_clean_store(setup_path) as store:
+        # exceptions
+        msg = "append_to_multiple requires a selector that is in passed dict"
+        with pytest.raises(ValueError, match=msg):
+            store.append_to_multiple(
+                {"df1": ["A", "B"], "df2": None}, df, selector="df3"
+            )
+
+        with pytest.raises(ValueError, match=msg):
+            store.append_to_multiple({"df1": None, "df2": None}, df, selector="df3")
+
+        msg = (
+            "append_to_multiple must have a dictionary specified as the way to "
+            "split the value"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append_to_multiple("df1", df, "df1")
+
+        # regular operation
+        store.append_to_multiple({"df1": ["A", "B"], "df2": None}, df, selector="df1")
+        result = store.select_as_multiple(
+            ["df1", "df2"], where=["A>0", "B>0"], selector="df1"
+        )
+        expected = df[(df.A > 0) & (df.B > 0)]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_append_to_multiple_dropna(setup_path):
+    df1 = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df2 = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    ).rename(columns="{}_2".format)
+    df1.iloc[1, df1.columns.get_indexer(["A", "B"])] = np.nan
+    df = concat([df1, df2], axis=1)
+
+    with ensure_clean_store(setup_path) as store:
+        # dropna=True should guarantee rows are synchronized
+        store.append_to_multiple(
+            {"df1": ["A", "B"], "df2": None}, df, selector="df1", dropna=True
+        )
+        result = store.select_as_multiple(["df1", "df2"])
+        expected = df.dropna()
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+        tm.assert_index_equal(store.select("df1").index, store.select("df2").index)
+
+
+def test_append_to_multiple_dropna_false(setup_path):
+    df1 = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df2 = df1.copy().rename(columns="{}_2".format)
+    df1.iloc[1, df1.columns.get_indexer(["A", "B"])] = np.nan
+    df = concat([df1, df2], axis=1)
+
+    with (
+        ensure_clean_store(setup_path) as store,
+        pd.option_context("io.hdf.dropna_table", True),
+    ):
+        # dropna=False shouldn't synchronize row indexes
+        store.append_to_multiple(
+            {"df1a": ["A", "B"], "df2a": None}, df, selector="df1a", dropna=False
+        )
+
+        msg = "all tables must have exactly the same nrows!"
+        with pytest.raises(ValueError, match=msg):
+            store.select_as_multiple(["df1a", "df2a"])
+
+        assert not store.select("df1a").index.equals(store.select("df2a").index)
+
+
+def test_append_to_multiple_min_itemsize(setup_path):
+    # GH 11238
+    df = DataFrame(
+        {
+            "IX": np.arange(1, 21),
+            "Num": np.arange(1, 21),
+            "BigNum": np.arange(1, 21) * 88,
+            "Str": ["a" for _ in range(20)],
+            "LongStr": ["abcde" for _ in range(20)],
+        }
+    )
+    expected = df.iloc[[0]]
+    # Reading/writing RangeIndex info is not supported yet
+    expected.index = Index(list(range(len(expected.index))))
+
+    with ensure_clean_store(setup_path) as store:
+        store.append_to_multiple(
+            {
+                "index": ["IX"],
+                "nums": ["Num", "BigNum"],
+                "strs": ["Str", "LongStr"],
+            },
+            df.iloc[[0]],
+            "index",
+            min_itemsize={"Str": 10, "LongStr": 100, "Num": 2},
+        )
+        result = store.select_as_multiple(["index", "nums", "strs"])
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+
+def test_append_string_nan_rep(setup_path):
+    # GH 16300
+    df = DataFrame({"A": "a", "B": "foo"}, index=np.arange(10))
+    df_nan = df.copy()
+    df_nan.loc[0:4, :] = np.nan
+    msg = "NaN representation is too large for existing column size"
+
+    with ensure_clean_store(setup_path) as store:
+        # string column too small
+        store.append("sa", df["A"])
+        with pytest.raises(ValueError, match=msg):
+            store.append("sa", df_nan["A"])
+
+        # nan_rep too big
+        store.append("sb", df["B"], nan_rep="bars")
+        with pytest.raises(ValueError, match=msg):
+            store.append("sb", df_nan["B"])
+
+        # smaller modified nan_rep
+        store.append("sc", df["A"], nan_rep="n")
+        store.append("sc", df_nan["A"])
+        result = store["sc"]
+        expected = concat([df["A"], df_nan["A"]])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/io/pytables/test_categorical.py b/pandas/pandas/tests/io/pytables/test_categorical.py
new file mode 100644
index 0000000000000..ed2616b24cd71
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_categorical.py
@@ -0,0 +1,208 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+    _testing as tm,
+    concat,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_categorical(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # Basic
+        _maybe_remove(store, "s")
+        s = Series(
+            Categorical(
+                ["a", "b", "b", "a", "a", "c"],
+                categories=["a", "b", "c", "d"],
+                ordered=False,
+            )
+        )
+        store.append("s", s, format="table")
+        result = store.select("s")
+        tm.assert_series_equal(s, result)
+
+        _maybe_remove(store, "s_ordered")
+        s = Series(
+            Categorical(
+                ["a", "b", "b", "a", "a", "c"],
+                categories=["a", "b", "c", "d"],
+                ordered=True,
+            )
+        )
+        store.append("s_ordered", s, format="table")
+        result = store.select("s_ordered")
+        tm.assert_series_equal(s, result)
+
+        _maybe_remove(store, "df")
+        df = DataFrame({"s": s, "vals": [1, 2, 3, 4, 5, 6]})
+        store.append("df", df, format="table")
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+        # Dtypes
+        _maybe_remove(store, "si")
+        s = Series([1, 1, 2, 2, 3, 4, 5]).astype("category")
+        store.append("si", s)
+        result = store.select("si")
+        tm.assert_series_equal(result, s)
+
+        _maybe_remove(store, "si2")
+        s = Series([1, 1, np.nan, 2, 3, 4, 5]).astype("category")
+        store.append("si2", s)
+        result = store.select("si2")
+        tm.assert_series_equal(result, s)
+
+        # Multiple
+        _maybe_remove(store, "df2")
+        df2 = df.copy()
+        df2["s2"] = Series(list("abcdefg")).astype("category")
+        store.append("df2", df2)
+        result = store.select("df2")
+        tm.assert_frame_equal(result, df2)
+
+        # Make sure the metadata is OK
+        info = store.info()
+        assert "/df2   " in info
+        # df2._mgr.blocks[0] and df2._mgr.blocks[2] are Categorical
+        assert "/df2/meta/values_block_0/meta" in info
+        assert "/df2/meta/values_block_2/meta" in info
+
+        # unordered
+        _maybe_remove(store, "s2")
+        s = Series(
+            Categorical(
+                ["a", "b", "b", "a", "a", "c"],
+                categories=["a", "b", "c", "d"],
+                ordered=False,
+            )
+        )
+        store.append("s2", s, format="table")
+        result = store.select("s2")
+        tm.assert_series_equal(result, s)
+
+        # Query
+        _maybe_remove(store, "df3")
+        store.append("df3", df, data_columns=["s"])
+        expected = df[df.s.isin(["b", "c"])]
+        result = store.select("df3", where=['s in ["b","c"]'])
+        tm.assert_frame_equal(result, expected)
+
+        expected = df[df.s.isin(["b", "c"])]
+        result = store.select("df3", where=['s = ["b","c"]'])
+        tm.assert_frame_equal(result, expected)
+
+        expected = df[df.s.isin(["d"])]
+        result = store.select("df3", where=['s in ["d"]'])
+        tm.assert_frame_equal(result, expected)
+
+        expected = df[df.s.isin(["f"])]
+        result = store.select("df3", where=['s in ["f"]'])
+        tm.assert_frame_equal(result, expected)
+
+        # Appending with same categories is ok
+        store.append("df3", df)
+
+        df = concat([df, df])
+        expected = df[df.s.isin(["b", "c"])]
+        result = store.select("df3", where=['s in ["b","c"]'])
+        tm.assert_frame_equal(result, expected)
+
+        # Appending must have the same categories
+        df3 = df.copy()
+        df3["s"] = df3["s"].cat.remove_unused_categories()
+
+        msg = "cannot append a categorical with different categories to the existing"
+        with pytest.raises(ValueError, match=msg):
+            store.append("df3", df3)
+
+        # Remove, and make sure meta data is removed (its a recursive
+        # removal so should be).
+        result = store.select("df3/meta/s/meta")
+        assert result is not None
+        store.remove("df3")
+
+        with pytest.raises(
+            KeyError, match="'No object named df3/meta/s/meta in the file'"
+        ):
+            store.select("df3/meta/s/meta")
+
+
+def test_categorical_conversion(tmp_path, setup_path):
+    # GH13322
+    # Check that read_hdf with categorical columns doesn't return rows if
+    # where criteria isn't met.
+    obsids = ["ESP_012345_6789", "ESP_987654_3210"]
+    imgids = ["APF00006np", "APF0001imm"]
+    data = [4.3, 9.8]
+
+    # Test without categories
+    df = DataFrame({"obsids": obsids, "imgids": imgids, "data": data})
+
+    # We are expecting an empty DataFrame matching types of df
+    expected = df.iloc[[], :]
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", data_columns=True)
+    result = read_hdf(path, "df", where="obsids=B")
+    tm.assert_frame_equal(result, expected)
+
+    # Test with categories
+    df.obsids = df.obsids.astype("category")
+    df.imgids = df.imgids.astype("category")
+
+    # We are expecting an empty DataFrame matching types of df
+    expected = df.iloc[[], :]
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", data_columns=True)
+    result = read_hdf(path, "df", where="obsids=B")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_categorical_nan_only_columns(tmp_path, setup_path):
+    # GH18413
+    # Check that read_hdf with categorical columns with NaN-only values can
+    # be read back.
+    df = DataFrame(
+        {
+            "a": ["a", "b", "c", np.nan],
+            "b": [np.nan, np.nan, np.nan, np.nan],
+            "c": [1, 2, 3, 4],
+            "d": Series([None] * 4, dtype=object),
+        }
+    )
+    df["a"] = df.a.astype("category")
+    df["b"] = df.b.astype("category")
+    df["d"] = df.b.astype("category")
+    expected = df
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", data_columns=True)
+    result = read_hdf(path, "df")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("where, expected", [["q", []], ["a", ["a"]]])
+def test_convert_value(tmp_path, setup_path, where: str, expected):
+    # GH39420
+    # Check that read_hdf with categorical columns can filter by where condition.
+    df = DataFrame({"col": ["a", "b", "s"]})
+    df.col = df.col.astype("category")
+    max_widths = {"col": 1}
+    categorical_values = sorted(df.col.unique())
+    expected = DataFrame({"col": expected})
+    expected.col = expected.col.astype("category")
+    expected.col = expected.col.cat.set_categories(categorical_values)
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", min_itemsize=max_widths)
+    result = read_hdf(path, where=f'col=="{where}"')
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/pytables/test_compat.py b/pandas/pandas/tests/io/pytables/test_compat.py
new file mode 100644
index 0000000000000..b78a503a2b8a3
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_compat.py
@@ -0,0 +1,75 @@
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+tables = pytest.importorskip("tables")
+
+
+@pytest.fixture
+def pytables_hdf5_file(tmp_path):
+    """
+    Use PyTables to create a simple HDF5 file.
+    """
+    table_schema = {
+        "c0": tables.Time64Col(pos=0),
+        "c1": tables.StringCol(5, pos=1),
+        "c2": tables.Int64Col(pos=2),
+    }
+
+    t0 = 1_561_105_000.0
+
+    testsamples = [
+        {"c0": t0, "c1": "aaaaa", "c2": 1},
+        {"c0": t0 + 1, "c1": "bbbbb", "c2": 2},
+        {"c0": t0 + 2, "c1": "ccccc", "c2": 10**5},
+        {"c0": t0 + 3, "c1": "ddddd", "c2": 4_294_967_295},
+    ]
+
+    objname = "pandas_test_timeseries"
+
+    path = tmp_path / "written_with_pytables.h5"
+    with tables.open_file(path, mode="w") as f:
+        t = f.create_table("/", name=objname, description=table_schema)
+        for sample in testsamples:
+            for key, value in sample.items():
+                t.row[key] = value
+            t.row.append()
+
+    return path, objname, pd.DataFrame(testsamples)
+
+
+class TestReadPyTablesHDF5:
+    """
+    A group of tests which covers reading HDF5 files written by plain PyTables
+    (not written by pandas).
+
+    Was introduced for regression-testing issue 11188.
+    """
+
+    def test_read_complete(self, pytables_hdf5_file):
+        path, objname, df = pytables_hdf5_file
+        result = pd.read_hdf(path, key=objname)
+        expected = df
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_read_with_start(self, pytables_hdf5_file):
+        path, objname, df = pytables_hdf5_file
+        # This is a regression test for pandas-dev/pandas/issues/11188
+        result = pd.read_hdf(path, key=objname, start=1)
+        expected = df[1:].reset_index(drop=True)
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_read_with_stop(self, pytables_hdf5_file):
+        path, objname, df = pytables_hdf5_file
+        # This is a regression test for pandas-dev/pandas/issues/11188
+        result = pd.read_hdf(path, key=objname, stop=1)
+        expected = df[:1].reset_index(drop=True)
+        tm.assert_frame_equal(result, expected, check_index_type=True)
+
+    def test_read_with_startstop(self, pytables_hdf5_file):
+        path, objname, df = pytables_hdf5_file
+        # This is a regression test for pandas-dev/pandas/issues/11188
+        result = pd.read_hdf(path, key=objname, start=1, stop=2)
+        expected = df[1:2].reset_index(drop=True)
+        tm.assert_frame_equal(result, expected, check_index_type=True)
diff --git a/pandas/pandas/tests/io/pytables/test_complex.py b/pandas/pandas/tests/io/pytables/test_complex.py
new file mode 100644
index 0000000000000..c5cac5a5caf09
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_complex.py
@@ -0,0 +1,195 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.tests.io.pytables.common import ensure_clean_store
+
+from pandas.io.pytables import read_hdf
+
+
+def test_complex_fixed(tmp_path, setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)).astype(np.complex64),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df")
+    reread = read_hdf(path, "df")
+    tm.assert_frame_equal(df, reread)
+
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)).astype(np.complex128),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df")
+    reread = read_hdf(path, "df")
+    tm.assert_frame_equal(df, reread)
+
+
+def test_complex_table(tmp_path, setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)).astype(np.complex64),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table")
+    reread = read_hdf(path, key="df")
+    tm.assert_frame_equal(df, reread)
+
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)).astype(np.complex128),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", mode="w")
+    reread = read_hdf(path, "df")
+    tm.assert_frame_equal(df, reread)
+
+
+def test_complex_mixed_fixed(tmp_path, setup_path):
+    complex64 = np.array(
+        [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex64
+    )
+    complex128 = np.array(
+        [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex128
+    )
+    df = DataFrame(
+        {
+            "A": [1, 2, 3, 4],
+            "B": ["a", "b", "c", "d"],
+            "C": complex64,
+            "D": complex128,
+            "E": [1.0, 2.0, 3.0, 4.0],
+        },
+        index=list("abcd"),
+    )
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df")
+    reread = read_hdf(path, "df")
+    tm.assert_frame_equal(df, reread)
+
+
+def test_complex_mixed_table(tmp_path, setup_path):
+    complex64 = np.array(
+        [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex64
+    )
+    complex128 = np.array(
+        [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex128
+    )
+    df = DataFrame(
+        {
+            "A": [1, 2, 3, 4],
+            "B": ["a", "b", "c", "d"],
+            "C": complex64,
+            "D": complex128,
+            "E": [1.0, 2.0, 3.0, 4.0],
+        },
+        index=list("abcd"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df, data_columns=["A", "B"])
+        result = store.select("df", where="A>2")
+        tm.assert_frame_equal(df.loc[df.A > 2], result)
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table")
+    reread = read_hdf(path, "df")
+    tm.assert_frame_equal(df, reread)
+
+
+def test_complex_across_dimensions_fixed(tmp_path, setup_path):
+    complex128 = np.array([1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j])
+    s = Series(complex128, index=list("abcd"))
+    df = DataFrame({"A": s, "B": s})
+
+    objs = [s, df]
+    comps = [tm.assert_series_equal, tm.assert_frame_equal]
+    for obj, comp in zip(objs, comps):
+        path = tmp_path / setup_path
+        obj.to_hdf(path, key="obj", format="fixed")
+        reread = read_hdf(path, "obj")
+        comp(obj, reread)
+
+
+def test_complex_across_dimensions(tmp_path, setup_path):
+    complex128 = np.array([1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j])
+    s = Series(complex128, index=list("abcd"))
+    df = DataFrame({"A": s, "B": s})
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="obj", format="table")
+    reread = read_hdf(path, "obj")
+    tm.assert_frame_equal(df, reread)
+
+
+def test_complex_indexing_error(setup_path):
+    complex128 = np.array(
+        [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex128
+    )
+    df = DataFrame(
+        {"A": [1, 2, 3, 4], "B": ["a", "b", "c", "d"], "C": complex128},
+        index=list("abcd"),
+    )
+
+    msg = (
+        "Columns containing complex values can be stored "
+        "but cannot be indexed when using table format. "
+        "Either use fixed format, set index=False, "
+        "or do not include the columns containing complex "
+        "values to data_columns when initializing the table."
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        with pytest.raises(TypeError, match=msg):
+            store.append("df", df, data_columns=["C"])
+
+
+def test_complex_series_error(tmp_path, setup_path):
+    complex128 = np.array([1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j])
+    s = Series(complex128, index=list("abcd"))
+
+    msg = (
+        "Columns containing complex values can be stored "
+        "but cannot be indexed when using table format. "
+        "Either use fixed format, set index=False, "
+        "or do not include the columns containing complex "
+        "values to data_columns when initializing the table."
+    )
+
+    path = tmp_path / setup_path
+    with pytest.raises(TypeError, match=msg):
+        s.to_hdf(path, key="obj", format="t")
+
+    path = tmp_path / setup_path
+    s.to_hdf(path, key="obj", format="t", index=False)
+    reread = read_hdf(path, "obj")
+    tm.assert_series_equal(s, reread)
+
+
+def test_complex_append(setup_path):
+    df = DataFrame(
+        {
+            "a": np.random.default_rng(2).standard_normal(100).astype(np.complex128),
+            "b": np.random.default_rng(2).standard_normal(100),
+        }
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df, data_columns=["b"])
+        store.append("df", df)
+        result = store.select("df")
+        tm.assert_frame_equal(pd.concat([df, df], axis=0), result)
diff --git a/pandas/pandas/tests/io/pytables/test_errors.py b/pandas/pandas/tests/io/pytables/test_errors.py
new file mode 100644
index 0000000000000..b28101c09820f
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_errors.py
@@ -0,0 +1,256 @@
+import datetime
+from io import BytesIO
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalIndex,
+    DataFrame,
+    HDFStore,
+    Index,
+    MultiIndex,
+    _testing as tm,
+    date_range,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import ensure_clean_store
+
+from pandas.io.pytables import (
+    Term,
+    _maybe_adjust_name,
+)
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_pass_spec_to_storer(setup_path):
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df)
+        msg = (
+            "cannot pass a column specification when reading a Fixed format "
+            "store. this store must be selected in its entirety"
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", columns=["A"])
+        msg = (
+            "cannot pass a where specification when reading from a Fixed "
+            "format store. this store must be selected in its entirety"
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", where=[("columns=A")])
+
+
+def test_table_index_incompatible_dtypes(setup_path):
+    df1 = DataFrame({"a": [1, 2, 3]})
+    df2 = DataFrame({"a": [4, 5, 6]}, index=date_range("1/1/2000", periods=3))
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("frame", df1, format="table")
+        msg = re.escape("incompatible kind in col [integer - datetime64[ns]]")
+        with pytest.raises(TypeError, match=msg):
+            store.put("frame", df2, format="table", append=True)
+
+
+def test_unimplemented_dtypes_table_columns(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        dtypes = [("date", datetime.date(2001, 1, 2))]
+
+        # currently not supported dtypes ####
+        for n, f in dtypes:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+            df[n] = f
+            msg = re.escape(f"[{n}] is not implemented as a table column")
+            with pytest.raises(TypeError, match=msg):
+                store.append(f"df1_{n}", df)
+
+    # frame
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+    df["obj1"] = "foo"
+    df["obj2"] = "bar"
+    df["datetime1"] = datetime.date(2001, 1, 2)
+    df = df._consolidate()
+
+    with ensure_clean_store(setup_path) as store:
+        # this fails because we have a date in the object block......
+        msg = "|".join(
+            [
+                re.escape(
+                    "Cannot serialize the column [datetime1]\nbecause its data "
+                    "contents are not [string] but [date] object dtype"
+                ),
+                re.escape("[date] is not implemented as a table column"),
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.append("df_unimplemented", df)
+
+
+def test_invalid_terms(tmp_path, setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df["string"] = "foo"
+        df.loc[df.index[0:4], "string"] = "bar"
+
+        store.put("df", df, format="table")
+
+        # some invalid terms
+        msg = re.escape("__init__() missing 1 required positional argument: 'where'")
+        with pytest.raises(TypeError, match=msg):
+            Term()
+
+        # more invalid
+        msg = re.escape(
+            "cannot process expression [df.index[3]], "
+            "[2000-01-06 00:00:00] is not a valid condition"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.select("df", "df.index[3]")
+
+        msg = "invalid syntax"
+        with pytest.raises(SyntaxError, match=msg):
+            store.select("df", "index>")
+
+    # from the docs
+    path = tmp_path / setup_path
+    dfq = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=list("ABCD"),
+        index=date_range("20130101", periods=10),
+    )
+    dfq.to_hdf(path, key="dfq", format="table", data_columns=True)
+
+    # check ok
+    read_hdf(path, "dfq", where="index>Timestamp('20130104') & columns=['A', 'B']")
+    read_hdf(path, "dfq", where="A>0 or C>0")
+
+    # catch the invalid reference
+    path = tmp_path / setup_path
+    dfq = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=list("ABCD"),
+        index=date_range("20130101", periods=10),
+    )
+    dfq.to_hdf(path, key="dfq", format="table")
+
+    msg = (
+        r"The passed where expression: A>0 or C>0\n\s*"
+        r"contains an invalid variable reference\n\s*"
+        r"all of the variable references must be a reference to\n\s*"
+        r"an axis \(e.g. 'index' or 'columns'\), or a data_column\n\s*"
+        r"The currently defined references are: index,columns\n"
+    )
+    with pytest.raises(ValueError, match=msg):
+        read_hdf(path, "dfq", where="A>0 or C>0")
+
+
+def test_append_with_diff_col_name_types_raises_value_error(setup_path):
+    df = DataFrame(np.random.default_rng(2).standard_normal((10, 1)))
+    df2 = DataFrame({"a": np.random.default_rng(2).standard_normal(10)})
+    df3 = DataFrame({(1, 2): np.random.default_rng(2).standard_normal(10)})
+    df4 = DataFrame({("1", 2): np.random.default_rng(2).standard_normal(10)})
+    df5 = DataFrame({("1", 2, object): np.random.default_rng(2).standard_normal(10)})
+
+    with ensure_clean_store(setup_path) as store:
+        name = "df_diff_valerror"
+        store.append(name, df)
+
+        for d in (df2, df3, df4, df5):
+            msg = re.escape(
+                "cannot match existing table structure for [0] on appending data"
+            )
+            with pytest.raises(ValueError, match=msg):
+                store.append(name, d)
+
+
+def test_invalid_complib(setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    with tm.ensure_clean(setup_path) as path:
+        msg = r"complib only supports \[.*\] compression."
+        with pytest.raises(ValueError, match=msg):
+            df.to_hdf(path, key="df", complib="foolib")
+
+
+@pytest.mark.parametrize(
+    "idx",
+    [
+        date_range("2019", freq="D", periods=3, tz="UTC"),
+        CategoricalIndex(list("abc")),
+    ],
+)
+def test_to_hdf_multiindex_extension_dtype(idx, tmp_path, setup_path):
+    # GH 7775
+    mi = MultiIndex.from_arrays([idx, idx])
+    df = DataFrame(0, index=mi, columns=["a"])
+    path = tmp_path / setup_path
+    with pytest.raises(NotImplementedError, match="Saving a MultiIndex"):
+        df.to_hdf(path, key="df")
+
+
+def test_unsuppored_hdf_file_error(datapath):
+    # GH 9539
+    data_path = datapath("io", "data", "legacy_hdf/incompatible_dataset.h5")
+    message = (
+        r"Dataset\(s\) incompatible with Pandas data types, "
+        "not table, or no datasets found in HDF5 file."
+    )
+
+    with pytest.raises(ValueError, match=message):
+        read_hdf(data_path)
+
+
+def test_read_hdf_errors(setup_path, tmp_path):
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+
+    path = tmp_path / setup_path
+    msg = r"File [\S]* does not exist"
+    with pytest.raises(OSError, match=msg):
+        read_hdf(path, "key")
+
+    df.to_hdf(path, key="df")
+    store = HDFStore(path, mode="r")
+    store.close()
+
+    msg = "The HDFStore must be open for reading."
+    with pytest.raises(OSError, match=msg):
+        read_hdf(store, "df")
+
+
+def test_read_hdf_generic_buffer_errors():
+    msg = "Support for generic buffers has not been implemented."
+    with pytest.raises(NotImplementedError, match=msg):
+        read_hdf(BytesIO(b""), "df")
+
+
+@pytest.mark.parametrize("bad_version", [(1, 2), (1,), [], "12", "123"])
+def test_maybe_adjust_name_bad_version_raises(bad_version):
+    msg = "Version is incorrect, expected sequence of 3 integers"
+    with pytest.raises(ValueError, match=msg):
+        _maybe_adjust_name("values_block_0", version=bad_version)
diff --git a/pandas/pandas/tests/io/pytables/test_file_handling.py b/pandas/pandas/tests/io/pytables/test_file_handling.py
new file mode 100644
index 0000000000000..27b5d34146f85
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_file_handling.py
@@ -0,0 +1,523 @@
+import os
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    PY311,
+    is_ci_environment,
+    is_platform_linux,
+    is_platform_little_endian,
+    is_platform_mac,
+)
+from pandas.errors import (
+    ClosedFileError,
+    PossibleDataLossError,
+)
+
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    Series,
+    _testing as tm,
+    date_range,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+    tables,
+)
+
+from pandas.io import pytables
+from pandas.io.pytables import Term
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+@pytest.mark.parametrize("mode", ["r", "r+", "a", "w"])
+def test_mode(setup_path, tmp_path, mode, using_infer_string):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    msg = r"[\S]* does not exist"
+    path = tmp_path / setup_path
+
+    # constructor
+    if mode in ["r", "r+"]:
+        with pytest.raises(OSError, match=msg):
+            HDFStore(path, mode=mode)
+
+    else:
+        with HDFStore(path, mode=mode) as store:
+            assert store._handle.mode == mode
+
+    path = tmp_path / setup_path
+
+    # context
+    if mode in ["r", "r+"]:
+        with pytest.raises(OSError, match=msg):
+            with HDFStore(path, mode=mode) as store:
+                pass
+    else:
+        with HDFStore(path, mode=mode) as store:
+            assert store._handle.mode == mode
+
+    path = tmp_path / setup_path
+
+    # conv write
+    if mode in ["r", "r+"]:
+        with pytest.raises(OSError, match=msg):
+            df.to_hdf(path, key="df", mode=mode)
+        df.to_hdf(path, key="df", mode="w")
+    else:
+        df.to_hdf(path, key="df", mode=mode)
+
+    # conv read
+    if mode in ["w"]:
+        msg = (
+            "mode w is not allowed while performing a read. "
+            r"Allowed modes are r, r\+ and a."
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_hdf(path, "df", mode=mode)
+    else:
+        result = read_hdf(path, "df", mode=mode)
+        if using_infer_string:
+            df.columns = df.columns.astype("str")
+        tm.assert_frame_equal(result, df)
+
+
+def test_default_mode(tmp_path, setup_path, using_infer_string):
+    # read_hdf uses default mode
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="w")
+    result = read_hdf(path, "df")
+    expected = df.copy()
+    if using_infer_string:
+        expected.columns = expected.columns.astype("str")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_reopen_handle(tmp_path, setup_path):
+    path = tmp_path / setup_path
+
+    store = HDFStore(path, mode="a")
+    store["a"] = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+
+    msg = (
+        r"Re-opening the file \[[\S]*\] with mode \[a\] will delete the "
+        "current file!"
+    )
+    # invalid mode change
+    with pytest.raises(PossibleDataLossError, match=msg):
+        store.open("w")
+
+    store.close()
+    assert not store.is_open
+
+    # truncation ok here
+    store.open("w")
+    assert store.is_open
+    assert len(store) == 0
+    store.close()
+    assert not store.is_open
+
+    store = HDFStore(path, mode="a")
+    store["a"] = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+
+    # reopen as read
+    store.open("r")
+    assert store.is_open
+    assert len(store) == 1
+    assert store._mode == "r"
+    store.close()
+    assert not store.is_open
+
+    # reopen as append
+    store.open("a")
+    assert store.is_open
+    assert len(store) == 1
+    assert store._mode == "a"
+    store.close()
+    assert not store.is_open
+
+    # reopen as append (again)
+    store.open("a")
+    assert store.is_open
+    assert len(store) == 1
+    assert store._mode == "a"
+    store.close()
+    assert not store.is_open
+
+
+def test_open_args(setup_path, using_infer_string):
+    with tm.ensure_clean(setup_path) as path:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+
+        # create an in memory store
+        store = HDFStore(
+            path, mode="a", driver="H5FD_CORE", driver_core_backing_store=0
+        )
+        store["df"] = df
+        store.append("df2", df)
+
+        expected = df.copy()
+        if using_infer_string:
+            expected.index = expected.index.astype("str")
+            expected.columns = expected.columns.astype("str")
+
+        tm.assert_frame_equal(store["df"], expected)
+        tm.assert_frame_equal(store["df2"], expected)
+
+        store.close()
+
+    # the file should not have actually been written
+    assert not os.path.exists(path)
+
+
+def test_flush(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = Series(range(5))
+        store.flush()
+        store.flush(fsync=True)
+
+
+def test_complibs_default_settings(tmp_path, setup_path, using_infer_string):
+    # GH15943
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+
+    # Set complevel and check if complib is automatically set to
+    # default value
+    tmpfile = tmp_path / setup_path
+    df.to_hdf(tmpfile, key="df", complevel=9)
+    result = read_hdf(tmpfile, "df")
+    expected = df.copy()
+    if using_infer_string:
+        expected.index = expected.index.astype("str")
+        expected.columns = expected.columns.astype("str")
+    tm.assert_frame_equal(result, expected)
+
+    with tables.open_file(tmpfile, mode="r") as h5file:
+        for node in h5file.walk_nodes(where="/df", classname="Leaf"):
+            assert node.filters.complevel == 9
+            assert node.filters.complib == "zlib"
+
+    # Set complib and check to see if compression is disabled
+    tmpfile = tmp_path / setup_path
+    df.to_hdf(tmpfile, key="df", complib="zlib")
+    result = read_hdf(tmpfile, "df")
+    expected = df.copy()
+    if using_infer_string:
+        expected.index = expected.index.astype("str")
+        expected.columns = expected.columns.astype("str")
+    tm.assert_frame_equal(result, expected)
+
+    with tables.open_file(tmpfile, mode="r") as h5file:
+        for node in h5file.walk_nodes(where="/df", classname="Leaf"):
+            assert node.filters.complevel == 0
+            assert node.filters.complib is None
+
+    # Check if not setting complib or complevel results in no compression
+    tmpfile = tmp_path / setup_path
+    df.to_hdf(tmpfile, key="df")
+    result = read_hdf(tmpfile, "df")
+    expected = df.copy()
+    if using_infer_string:
+        expected.index = expected.index.astype("str")
+        expected.columns = expected.columns.astype("str")
+    tm.assert_frame_equal(result, expected)
+
+    with tables.open_file(tmpfile, mode="r") as h5file:
+        for node in h5file.walk_nodes(where="/df", classname="Leaf"):
+            assert node.filters.complevel == 0
+            assert node.filters.complib is None
+
+
+def test_complibs_default_settings_override(tmp_path, setup_path):
+    # Check if file-defaults can be overridden on a per table basis
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+    tmpfile = tmp_path / setup_path
+    store = HDFStore(tmpfile)
+    store.append("dfc", df, complevel=9, complib="blosc")
+    store.append("df", df)
+    store.close()
+
+    with tables.open_file(tmpfile, mode="r") as h5file:
+        for node in h5file.walk_nodes(where="/df", classname="Leaf"):
+            assert node.filters.complevel == 0
+            assert node.filters.complib is None
+        for node in h5file.walk_nodes(where="/dfc", classname="Leaf"):
+            assert node.filters.complevel == 9
+            assert node.filters.complib == "blosc"
+
+
+@pytest.mark.parametrize("lvl", range(10))
+@pytest.mark.parametrize("lib", tables.filters.all_complibs)
+@pytest.mark.filterwarnings("ignore:object name is not a valid")
+@pytest.mark.skipif(
+    not PY311 and is_ci_environment() and is_platform_linux(),
+    reason="Segfaulting in a CI environment",
+    # with xfail, would sometimes raise UnicodeDecodeError
+    # invalid state byte
+)
+def test_complibs(tmp_path, lvl, lib, request):
+    # GH14478
+    if PY311 and is_platform_linux() and lib == "blosc2" and lvl != 0:
+        request.applymarker(
+            pytest.mark.xfail(reason=f"Fails for {lib} on Linux and PY > 3.11")
+        )
+    df = DataFrame(
+        np.ones((30, 4)), columns=list("ABCD"), index=np.arange(30).astype(np.str_)
+    )
+
+    # Remove lzo if its not available on this platform
+    if not tables.which_lib_version("lzo"):
+        pytest.skip("lzo not available")
+    # Remove bzip2 if its not available on this platform
+    if not tables.which_lib_version("bzip2"):
+        pytest.skip("bzip2 not available")
+
+    tmpfile = tmp_path / f"{lvl}_{lib}.h5"
+    gname = f"{lvl}_{lib}"
+
+    # Write and read file to see if data is consistent
+    df.to_hdf(tmpfile, key=gname, complib=lib, complevel=lvl)
+    result = read_hdf(tmpfile, gname)
+    tm.assert_frame_equal(result, df)
+
+    is_mac = is_platform_mac()
+
+    # Open file and check metadata for correct amount of compression
+    with tables.open_file(tmpfile, mode="r") as h5table:
+        for node in h5table.walk_nodes(where="/" + gname, classname="Leaf"):
+            assert node.filters.complevel == lvl
+            if lvl == 0:
+                assert node.filters.complib is None
+            elif is_mac and lib == "blosc2":
+                res = node.filters.complib
+                assert res in [lib, "blosc2:blosclz"], res
+            else:
+                assert node.filters.complib == lib
+
+
+@pytest.mark.skipif(
+    not is_platform_little_endian(), reason="reason platform is not little endian"
+)
+def test_encoding(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame({"A": "foo", "B": "bar"}, index=range(5))
+        df.loc[2, "A"] = np.nan
+        df.loc[3, "B"] = np.nan
+        _maybe_remove(store, "df")
+        store.append("df", df, encoding="ascii")
+        tm.assert_frame_equal(store["df"], df)
+
+        expected = df.reindex(columns=["A"])
+        result = store.select("df", Term("columns=A", encoding="ascii"))
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "val",
+    [
+        [b"E\xc9, 17", b"", b"a", b"b", b"c"],
+        [b"E\xc9, 17", b"a", b"b", b"c"],
+        [b"EE, 17", b"", b"a", b"b", b"c"],
+        [b"E\xc9, 17", b"\xf8\xfc", b"a", b"b", b"c"],
+        [b"", b"a", b"b", b"c"],
+        [b"\xf8\xfc", b"a", b"b", b"c"],
+        [b"A\xf8\xfc", b"", b"a", b"b", b"c"],
+        [np.nan, b"", b"b", b"c"],
+        [b"A\xf8\xfc", np.nan, b"", b"b", b"c"],
+    ],
+)
+@pytest.mark.parametrize("dtype", ["category", None])
+def test_latin_encoding(tmp_path, setup_path, dtype, val):
+    enc = "latin-1"
+    nan_rep = ""
+    key = "data"
+
+    val = [x.decode(enc) if isinstance(x, bytes) else x for x in val]
+    ser = Series(val, dtype=dtype)
+
+    store = tmp_path / setup_path
+    ser.to_hdf(store, key=key, format="table", encoding=enc, nan_rep=nan_rep)
+    retr = read_hdf(store, key)
+
+    # TODO:(3.0): once Categorical replace deprecation is enforced,
+    #  we may be able to re-simplify the construction of s_nan
+    if dtype == "category":
+        if nan_rep in ser.cat.categories:
+            s_nan = ser.cat.remove_categories([nan_rep])
+        else:
+            s_nan = ser
+    else:
+        s_nan = ser.replace(nan_rep, np.nan)
+
+    tm.assert_series_equal(s_nan, retr)
+
+
+def test_multiple_open_close(tmp_path, setup_path):
+    # gh-4409: open & close multiple times
+
+    path = tmp_path / setup_path
+
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+    df.to_hdf(path, key="df", mode="w", format="table")
+
+    # single
+    store = HDFStore(path)
+    assert "CLOSED" not in store.info()
+    assert store.is_open
+
+    store.close()
+    assert "CLOSED" in store.info()
+    assert not store.is_open
+
+    path = tmp_path / setup_path
+
+    if pytables._table_file_open_policy_is_strict:
+        # multiples
+        store1 = HDFStore(path)
+        msg = (
+            r"The file [\S]* is already opened\.  Please close it before "
+            r"reopening in write mode\."
+        )
+        with pytest.raises(ValueError, match=msg):
+            HDFStore(path)
+
+        store1.close()
+    else:
+        # multiples
+        store1 = HDFStore(path)
+        store2 = HDFStore(path)
+
+        assert "CLOSED" not in store1.info()
+        assert "CLOSED" not in store2.info()
+        assert store1.is_open
+        assert store2.is_open
+
+        store1.close()
+        assert "CLOSED" in store1.info()
+        assert not store1.is_open
+        assert "CLOSED" not in store2.info()
+        assert store2.is_open
+
+        store2.close()
+        assert "CLOSED" in store1.info()
+        assert "CLOSED" in store2.info()
+        assert not store1.is_open
+        assert not store2.is_open
+
+        # nested close
+        store = HDFStore(path, mode="w")
+        store.append("df", df)
+
+        store2 = HDFStore(path)
+        store2.append("df2", df)
+        store2.close()
+        assert "CLOSED" in store2.info()
+        assert not store2.is_open
+
+        store.close()
+        assert "CLOSED" in store.info()
+        assert not store.is_open
+
+        # double closing
+        store = HDFStore(path, mode="w")
+        store.append("df", df)
+
+        store2 = HDFStore(path)
+        store.close()
+        assert "CLOSED" in store.info()
+        assert not store.is_open
+
+        store2.close()
+        assert "CLOSED" in store2.info()
+        assert not store2.is_open
+
+    # ops on a closed store
+    path = tmp_path / setup_path
+
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+    df.to_hdf(path, key="df", mode="w", format="table")
+
+    store = HDFStore(path)
+    store.close()
+
+    msg = r"[\S]* file is not open!"
+    with pytest.raises(ClosedFileError, match=msg):
+        store.keys()
+
+    with pytest.raises(ClosedFileError, match=msg):
+        "df" in store
+
+    with pytest.raises(ClosedFileError, match=msg):
+        len(store)
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store["df"]
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.select("df")
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.get("df")
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.append("df2", df)
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.put("df3", df)
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.get_storer("df2")
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.remove("df2")
+
+    with pytest.raises(ClosedFileError, match=msg):
+        store.select("df")
+
+    msg = "'HDFStore' object has no attribute 'df'"
+    with pytest.raises(AttributeError, match=msg):
+        store.df
+
+
+def test_fspath():
+    with tm.ensure_clean("foo.h5") as path:
+        with HDFStore(path) as store:
+            assert os.fspath(store) == str(path)
diff --git a/pandas/pandas/tests/io/pytables/test_keys.py b/pandas/pandas/tests/io/pytables/test_keys.py
new file mode 100644
index 0000000000000..9c5fc8786c7c6
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_keys.py
@@ -0,0 +1,87 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    Series,
+    date_range,
+)
+from pandas.tests.io.pytables.common import (
+    ensure_clean_store,
+    tables,
+)
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_keys(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["b"] = Series(
+            range(10), dtype="float64", index=[f"i_{i}" for i in range(10)]
+        )
+        store["c"] = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+
+        assert len(store) == 3
+        expected = {"/a", "/b", "/c"}
+        assert set(store.keys()) == expected
+        assert set(store) == expected
+
+
+def test_non_pandas_keys(tmp_path, setup_path):
+    class Table1(tables.IsDescription):
+        value1 = tables.Float32Col()
+
+    class Table2(tables.IsDescription):
+        value2 = tables.Float32Col()
+
+    class Table3(tables.IsDescription):
+        value3 = tables.Float32Col()
+
+    path = tmp_path / setup_path
+    with tables.open_file(path, mode="w") as h5file:
+        group = h5file.create_group("/", "group")
+        h5file.create_table(group, "table1", Table1, "Table 1")
+        h5file.create_table(group, "table2", Table2, "Table 2")
+        h5file.create_table(group, "table3", Table3, "Table 3")
+    with HDFStore(path) as store:
+        assert len(store.keys(include="native")) == 3
+        expected = {"/group/table1", "/group/table2", "/group/table3"}
+        assert set(store.keys(include="native")) == expected
+        assert set(store.keys(include="pandas")) == set()
+        for name in expected:
+            df = store.get(name)
+            assert len(df.columns) == 1
+
+
+def test_keys_illegal_include_keyword_value(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        with pytest.raises(
+            ValueError,
+            match="`include` should be either 'pandas' or 'native' but is 'illegal'",
+        ):
+            store.keys(include="illegal")
+
+
+def test_keys_ignore_hdf_softlink(setup_path):
+    # GH 20523
+    # Puts a softlink into HDF file and rereads
+
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame({"A": range(5), "B": range(5)})
+        store.put("df", df)
+
+        assert store.keys() == ["/df"]
+
+        store._handle.create_soft_link(store._handle.root, "symlink", "df")
+
+        # Should ignore the softlink
+        assert store.keys() == ["/df"]
diff --git a/pandas/pandas/tests/io/pytables/test_put.py b/pandas/pandas/tests/io/pytables/test_put.py
new file mode 100644
index 0000000000000..c9fe6070b34c3
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_put.py
@@ -0,0 +1,419 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+    concat,
+    date_range,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+from pandas.util import _test_decorators as td
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_format_type(tmp_path, setup_path):
+    df = DataFrame({"A": [1, 2]})
+    with HDFStore(tmp_path / setup_path) as store:
+        store.put("a", df, format="fixed")
+        store.put("b", df, format="table")
+
+        assert store.get_storer("a").format_type == "fixed"
+        assert store.get_storer("b").format_type == "table"
+
+
+def test_format_kwarg_in_constructor(tmp_path, setup_path):
+    # GH 13291
+
+    msg = "format is not a defined argument for HDFStore"
+
+    with pytest.raises(ValueError, match=msg):
+        HDFStore(tmp_path / setup_path, format="table")
+
+
+def test_api_default_format(tmp_path, setup_path):
+    # default_format option
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+
+        with pd.option_context("io.hdf.default_format", "fixed"):
+            _maybe_remove(store, "df")
+            store.put("df", df)
+            assert not store.get_storer("df").is_table
+
+            msg = "Can only append to Tables"
+            with pytest.raises(ValueError, match=msg):
+                store.append("df2", df)
+
+        with pd.option_context("io.hdf.default_format", "table"):
+            _maybe_remove(store, "df")
+            store.put("df", df)
+            assert store.get_storer("df").is_table
+
+            _maybe_remove(store, "df2")
+            store.append("df2", df)
+            assert store.get_storer("df").is_table
+
+    path = tmp_path / setup_path
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    with pd.option_context("io.hdf.default_format", "fixed"):
+        df.to_hdf(path, key="df")
+        with HDFStore(path) as store:
+            assert not store.get_storer("df").is_table
+        with pytest.raises(ValueError, match=msg):
+            df.to_hdf(path, key="df2", append=True)
+
+    with pd.option_context("io.hdf.default_format", "table"):
+        df.to_hdf(path, key="df3")
+        with HDFStore(path) as store:
+            assert store.get_storer("df3").is_table
+        df.to_hdf(path, key="df4", append=True)
+        with HDFStore(path) as store:
+            assert store.get_storer("df4").is_table
+
+
+def test_put(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=20, freq="B"),
+        )
+        store["a"] = ts
+        store["b"] = df[:10]
+        store["foo/bar/bah"] = df[:10]
+        store["foo"] = df[:10]
+        store["/foo"] = df[:10]
+        store.put("c", df[:10], format="table")
+
+        # not OK, not a table
+        msg = "Can only append to Tables"
+        with pytest.raises(ValueError, match=msg):
+            store.put("b", df[10:], append=True)
+
+        # node does not currently exist, test _is_table_type returns False
+        # in this case
+        _maybe_remove(store, "f")
+        with pytest.raises(ValueError, match=msg):
+            store.put("f", df[10:], append=True)
+
+        # can't put to a table (use append instead)
+        with pytest.raises(ValueError, match=msg):
+            store.put("c", df[10:], append=True)
+
+        # overwrite table
+        store.put("c", df[:10], format="table", append=False)
+        tm.assert_frame_equal(df[:10], store["c"])
+
+
+def test_put_string_index(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        index = Index([f"I am a very long string index: {i}" for i in range(20)])
+        s = Series(np.arange(20), index=index)
+        df = DataFrame({"A": s, "B": s})
+
+        store["a"] = s
+        tm.assert_series_equal(store["a"], s)
+
+        store["b"] = df
+        tm.assert_frame_equal(store["b"], df)
+
+        # mixed length
+        index = Index(
+            ["abcdefghijklmnopqrstuvwxyz1234567890"]
+            + [f"I am a very long string index: {i}" for i in range(20)]
+        )
+        s = Series(np.arange(21), index=index)
+        df = DataFrame({"A": s, "B": s})
+        store["a"] = s
+        tm.assert_series_equal(store["a"], s)
+
+        store["b"] = df
+        tm.assert_frame_equal(store["b"], df)
+
+
+def test_put_compression(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+
+        store.put("c", df, format="table", complib="zlib")
+        tm.assert_frame_equal(store["c"], df)
+
+        # can't compress if format='fixed'
+        msg = "Compression not supported on Fixed format stores"
+        with pytest.raises(ValueError, match=msg):
+            store.put("b", df, format="fixed", complib="zlib")
+
+
+@td.skip_if_windows
+def test_put_compression_blosc(setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        # can't compress if format='fixed'
+        msg = "Compression not supported on Fixed format stores"
+        with pytest.raises(ValueError, match=msg):
+            store.put("b", df, format="fixed", complib="blosc")
+
+        store.put("c", df, format="table", complib="blosc")
+        tm.assert_frame_equal(store["c"], df)
+
+
+def test_put_datetime_ser(setup_path, performance_warning, using_infer_string):
+    # https://github.com/pandas-dev/pandas/pull/60663
+    ser = Series(3 * [Timestamp("20010102").as_unit("ns")])
+    with ensure_clean_store(setup_path) as store:
+        store.put("ser", ser)
+        expected = ser.copy()
+        result = store.get("ser")
+        tm.assert_series_equal(result, expected)
+
+
+def test_put_mixed_type(setup_path, performance_warning, using_infer_string):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df["obj1"] = "foo"
+    df["obj2"] = "bar"
+    df["bool1"] = df["A"] > 0
+    df["bool2"] = df["B"] > 0
+    df["bool3"] = True
+    df["int1"] = 1
+    df["int2"] = 2
+    df["timestamp1"] = Timestamp("20010102").as_unit("ns")
+    df["timestamp2"] = Timestamp("20010103").as_unit("ns")
+    df["datetime1"] = Timestamp("20010102").as_unit("ns")
+    df["datetime2"] = Timestamp("20010103").as_unit("ns")
+    df.loc[df.index[3:6], ["obj1"]] = np.nan
+    df = df._consolidate()
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+
+        warning = None if using_infer_string else performance_warning
+        with tm.assert_produces_warning(warning):
+            store.put("df", df)
+
+        expected = store.get("df")
+        tm.assert_frame_equal(expected, df)
+
+
+def test_put_str_frame(setup_path, performance_warning, string_dtype_arguments):
+    # https://github.com/pandas-dev/pandas/pull/60663
+    dtype = pd.StringDtype(*string_dtype_arguments)
+    df = DataFrame({"a": pd.array(["x", pd.NA, "y"], dtype=dtype)})
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+
+        store.put("df", df)
+        expected_dtype = "str" if dtype.na_value is np.nan else "string"
+        expected = df.astype(expected_dtype)
+        result = store.get("df")
+        tm.assert_frame_equal(result, expected)
+
+
+def test_put_str_series(setup_path, performance_warning, string_dtype_arguments):
+    # https://github.com/pandas-dev/pandas/pull/60663
+    dtype = pd.StringDtype(*string_dtype_arguments)
+    ser = Series(["x", pd.NA, "y"], dtype=dtype)
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+
+        store.put("ser", ser)
+        expected_dtype = "str" if dtype.na_value is np.nan else "string"
+        expected = ser.astype(expected_dtype)
+        result = store.get("ser")
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("format", ["table", "fixed"])
+@pytest.mark.parametrize(
+    "index",
+    [
+        Index([str(i) for i in range(10)]),
+        Index(np.arange(10, dtype=float)),
+        Index(np.arange(10)),
+        date_range("2020-01-01", periods=10),
+        pd.period_range("2020-01-01", periods=10),
+    ],
+)
+def test_store_index_types(setup_path, format, index):
+    # GH5386
+    # test storing various index types
+
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            columns=list("AB"),
+            index=index,
+        )
+        _maybe_remove(store, "df")
+        store.put("df", df, format=format)
+        tm.assert_frame_equal(df, store["df"])
+
+
+def test_column_multiindex(setup_path, using_infer_string):
+    # GH 4710
+    # recreate multi-indexes properly
+
+    index = MultiIndex.from_tuples(
+        [("A", "a"), ("A", "b"), ("B", "a"), ("B", "b")], names=["first", "second"]
+    )
+    df = DataFrame(np.arange(12).reshape(3, 4), columns=index)
+    expected = df.set_axis(df.index.to_numpy())
+
+    with ensure_clean_store(setup_path) as store:
+        if using_infer_string:
+            # TODO(infer_string) make this work for string dtype
+            msg = "Saving a MultiIndex with an extension dtype is not supported."
+            with pytest.raises(NotImplementedError, match=msg):
+                store.put("df", df)
+            return
+        store.put("df", df)
+        tm.assert_frame_equal(
+            store["df"], expected, check_index_type=True, check_column_type=True
+        )
+
+        store.put("df1", df, format="table")
+        tm.assert_frame_equal(
+            store["df1"], expected, check_index_type=True, check_column_type=True
+        )
+
+        msg = re.escape("cannot use a multi-index on axis [1] with data_columns ['A']")
+        with pytest.raises(ValueError, match=msg):
+            store.put("df2", df, format="table", data_columns=["A"])
+        msg = re.escape("cannot use a multi-index on axis [1] with data_columns True")
+        with pytest.raises(ValueError, match=msg):
+            store.put("df3", df, format="table", data_columns=True)
+
+    # appending multi-column on existing table (see GH 6167)
+    with ensure_clean_store(setup_path) as store:
+        store.append("df2", df)
+        store.append("df2", df)
+
+        tm.assert_frame_equal(store["df2"], concat((df, df)))
+
+    # non_index_axes name
+    df = DataFrame(np.arange(12).reshape(3, 4), columns=Index(list("ABCD"), name="foo"))
+    expected = df.set_axis(df.index.to_numpy())
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("df1", df, format="table")
+        tm.assert_frame_equal(
+            store["df1"], expected, check_index_type=True, check_column_type=True
+        )
+
+
+def test_store_multiindex(setup_path):
+    # validate multi-index names
+    # GH 5527
+    with ensure_clean_store(setup_path) as store:
+
+        def make_index(names=None):
+            dti = date_range("2013-12-01", "2013-12-02")
+            mi = MultiIndex.from_product([dti, range(2), range(3)], names=names)
+            return mi
+
+        # no names
+        _maybe_remove(store, "df")
+        df = DataFrame(np.zeros((12, 2)), columns=["a", "b"], index=make_index())
+        store.append("df", df)
+        tm.assert_frame_equal(store.select("df"), df)
+
+        # partial names
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            np.zeros((12, 2)),
+            columns=["a", "b"],
+            index=make_index(["date", None, None]),
+        )
+        store.append("df", df)
+        tm.assert_frame_equal(store.select("df"), df)
+
+        # series
+        _maybe_remove(store, "ser")
+        ser = Series(np.zeros(12), index=make_index(["date", None, None]))
+        store.append("ser", ser)
+        xp = Series(np.zeros(12), index=make_index(["date", "level_1", "level_2"]))
+        tm.assert_series_equal(store.select("ser"), xp)
+
+        # dup with column
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            np.zeros((12, 2)),
+            columns=["a", "b"],
+            index=make_index(["date", "a", "t"]),
+        )
+        msg = "duplicate names/columns in the multi-index when storing as a table"
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+        # dup within level
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            np.zeros((12, 2)),
+            columns=["a", "b"],
+            index=make_index(["date", "date", "date"]),
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df", df)
+
+        # fully names
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            np.zeros((12, 2)),
+            columns=["a", "b"],
+            index=make_index(["date", "s", "t"]),
+        )
+        store.append("df", df)
+        tm.assert_frame_equal(store.select("df"), df)
+
+
+@pytest.mark.parametrize("format", ["fixed", "table"])
+def test_store_periodindex(tmp_path, setup_path, format):
+    # GH 7796
+    # test of PeriodIndex in HDFStore
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 1)),
+        index=pd.period_range("20220101", freq="M", periods=5),
+    )
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="w", format=format)
+    expected = pd.read_hdf(path, "df")
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/io/pytables/test_pytables_missing.py b/pandas/pandas/tests/io/pytables/test_pytables_missing.py
new file mode 100644
index 0000000000000..8d9d3afc4ad6f
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_pytables_missing.py
@@ -0,0 +1,14 @@
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+
+@td.skip_if_installed("tables")
+def test_pytables_raises():
+    df = pd.DataFrame({"A": [1, 2]})
+    with pytest.raises(ImportError, match="tables"):
+        with tm.ensure_clean("foo.h5") as path:
+            df.to_hdf(path, key="df")
diff --git a/pandas/pandas/tests/io/pytables/test_read.py b/pandas/pandas/tests/io/pytables/test_read.py
new file mode 100644
index 0000000000000..7a3a7339e7809
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_read.py
@@ -0,0 +1,335 @@
+from contextlib import closing
+from pathlib import Path
+import re
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_windows
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    Series,
+    _testing as tm,
+    date_range,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+from pandas.io.pytables import TableIterator
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_read_missing_key_close_store(tmp_path, setup_path):
+    # GH 25766
+    path = tmp_path / setup_path
+    df = DataFrame({"a": range(2), "b": range(2)})
+    df.to_hdf(path, key="k1")
+
+    with pytest.raises(KeyError, match="'No object named k2 in the file'"):
+        read_hdf(path, "k2")
+
+    # smoke test to test that file is properly closed after
+    # read with KeyError before another write
+    df.to_hdf(path, key="k2")
+
+
+def test_read_index_error_close_store(tmp_path, setup_path):
+    # GH 25766
+    path = tmp_path / setup_path
+    df = DataFrame({"A": [], "B": []}, index=[])
+    df.to_hdf(path, key="k1")
+
+    with pytest.raises(IndexError, match=r"list index out of range"):
+        read_hdf(path, "k1", stop=0)
+
+    # smoke test to test that file is properly closed after
+    # read with IndexError before another write
+    df.to_hdf(path, key="k1")
+
+
+def test_read_missing_key_opened_store(tmp_path, setup_path):
+    # GH 28699
+    path = tmp_path / setup_path
+    df = DataFrame({"a": range(2), "b": range(2)})
+    df.to_hdf(path, key="k1")
+
+    with HDFStore(path, "r") as store:
+        with pytest.raises(KeyError, match="'No object named k2 in the file'"):
+            read_hdf(store, "k2")
+
+        # Test that the file is still open after a KeyError and that we can
+        # still read from it.
+        read_hdf(store, "k1")
+
+
+def test_read_column(setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+
+        # GH 17912
+        # HDFStore.select_column should raise a KeyError
+        # exception if the key is not a valid store
+        with pytest.raises(KeyError, match="No object named df in the file"):
+            store.select_column("df", "index")
+
+        store.append("df", df)
+        # error
+        with pytest.raises(
+            KeyError, match=re.escape("'column [foo] not found in the table'")
+        ):
+            store.select_column("df", "foo")
+
+        msg = re.escape("select_column() got an unexpected keyword argument 'where'")
+        with pytest.raises(TypeError, match=msg):
+            store.select_column("df", "index", where=["index>5"])
+
+        # valid
+        result = store.select_column("df", "index")
+        tm.assert_almost_equal(result.values, Series(df.index).values)
+        assert isinstance(result, Series)
+
+        # not a data indexable column
+        msg = re.escape(
+            "column [values_block_0] can not be extracted individually; "
+            "it is not data indexable"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.select_column("df", "values_block_0")
+
+        # a data column
+        df2 = df.copy()
+        df2["string"] = "foo"
+        store.append("df2", df2, data_columns=["string"])
+        result = store.select_column("df2", "string")
+        tm.assert_almost_equal(result.values, df2["string"].values)
+
+        # a data column with NaNs, result excludes the NaNs
+        df3 = df.copy()
+        df3["string"] = "foo"
+        df3.loc[df3.index[4:6], "string"] = np.nan
+        store.append("df3", df3, data_columns=["string"])
+        result = store.select_column("df3", "string")
+        tm.assert_almost_equal(result.values, df3["string"].values)
+
+        # start/stop
+        result = store.select_column("df3", "string", start=2)
+        tm.assert_almost_equal(result.values, df3["string"].values[2:])
+
+        result = store.select_column("df3", "string", start=-2)
+        tm.assert_almost_equal(result.values, df3["string"].values[-2:])
+
+        result = store.select_column("df3", "string", stop=2)
+        tm.assert_almost_equal(result.values, df3["string"].values[:2])
+
+        result = store.select_column("df3", "string", stop=-2)
+        tm.assert_almost_equal(result.values, df3["string"].values[:-2])
+
+        result = store.select_column("df3", "string", start=2, stop=-2)
+        tm.assert_almost_equal(result.values, df3["string"].values[2:-2])
+
+        result = store.select_column("df3", "string", start=-2, stop=2)
+        tm.assert_almost_equal(result.values, df3["string"].values[-2:2])
+
+        # GH 10392 - make sure column name is preserved
+        df4 = DataFrame({"A": np.random.default_rng(2).standard_normal(10), "B": "foo"})
+        store.append("df4", df4, data_columns=True)
+        expected = df4["B"]
+        result = store.select_column("df4", "B")
+        tm.assert_series_equal(result, expected)
+
+
+def test_pytables_native_read(datapath):
+    with ensure_clean_store(
+        datapath("io", "data", "legacy_hdf/pytables_native.h5"), mode="r"
+    ) as store:
+        d2 = store["detector/readout"]
+    assert isinstance(d2, DataFrame)
+
+
+@pytest.mark.skipif(is_platform_windows(), reason="native2 read fails oddly on windows")
+def test_pytables_native2_read(datapath):
+    with ensure_clean_store(
+        datapath("io", "data", "legacy_hdf", "pytables_native2.h5"), mode="r"
+    ) as store:
+        str(store)
+        d1 = store["detector"]
+    assert isinstance(d1, DataFrame)
+
+
+def test_read_hdf_open_store(tmp_path, setup_path, using_infer_string):
+    # GH10330
+    # No check for non-string path_or-buf, and no test of open store
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    df.index.name = "letters"
+    df = df.set_index(keys="E", append=True)
+
+    path = tmp_path / setup_path
+    if using_infer_string:
+        # TODO(infer_string) make this work for string dtype
+        msg = "Saving a MultiIndex with an extension dtype is not supported."
+        with pytest.raises(NotImplementedError, match=msg):
+            df.to_hdf(path, key="df", mode="w")
+        return
+    df.to_hdf(path, key="df", mode="w")
+    direct = read_hdf(path, "df")
+    with HDFStore(path, mode="r") as store:
+        indirect = read_hdf(store, "df")
+        tm.assert_frame_equal(direct, indirect)
+        assert store.is_open
+
+
+def test_read_hdf_index_not_view(tmp_path, setup_path):
+    # GH 37441
+    # Ensure that the index of the DataFrame is not a view
+    # into the original recarray that pytables reads in
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=[0, 1, 2, 3],
+        columns=list("ABCDE"),
+    )
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="w", format="table")
+
+    df2 = read_hdf(path, "df")
+    assert df2.index._data.base is None
+    tm.assert_frame_equal(df, df2)
+
+
+def test_read_hdf_iterator(tmp_path, setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    df.index.name = "letters"
+    df = df.set_index(keys="E", append=True)
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="w", format="t")
+    direct = read_hdf(path, "df")
+    iterator = read_hdf(path, "df", iterator=True)
+    with closing(iterator.store):
+        assert isinstance(iterator, TableIterator)
+        indirect = next(iterator.__iter__())
+    tm.assert_frame_equal(direct, indirect)
+
+
+def test_read_nokey(tmp_path, setup_path):
+    # GH10443
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+
+    # Categorical dtype not supported for "fixed" format. So no need
+    # to test with that dtype in the dataframe here.
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="a")
+    reread = read_hdf(path)
+    tm.assert_frame_equal(df, reread)
+    df.to_hdf(path, key="df2", mode="a")
+
+    msg = "key must be provided when HDF5 file contains multiple datasets."
+    with pytest.raises(ValueError, match=msg):
+        read_hdf(path)
+
+
+def test_read_nokey_table(tmp_path, setup_path):
+    # GH13231
+    df = DataFrame({"i": range(5), "c": Series(list("abacd"), dtype="category")})
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", mode="a", format="table")
+    reread = read_hdf(path)
+    tm.assert_frame_equal(df, reread)
+    df.to_hdf(path, key="df2", mode="a", format="table")
+
+    msg = "key must be provided when HDF5 file contains multiple datasets."
+    with pytest.raises(ValueError, match=msg):
+        read_hdf(path)
+
+
+def test_read_nokey_empty(tmp_path, setup_path):
+    path = tmp_path / setup_path
+    store = HDFStore(path)
+    store.close()
+    msg = re.escape(
+        "Dataset(s) incompatible with Pandas data types, not table, or no "
+        "datasets found in HDF5 file."
+    )
+    with pytest.raises(ValueError, match=msg):
+        read_hdf(path)
+
+
+def test_read_from_pathlib_path(tmp_path, setup_path):
+    # GH11773
+    expected = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    filename = tmp_path / setup_path
+    path_obj = Path(filename)
+
+    expected.to_hdf(path_obj, key="df", mode="a")
+    actual = read_hdf(path_obj, key="df")
+
+    tm.assert_frame_equal(expected, actual)
+
+
+@pytest.mark.parametrize("format", ["fixed", "table"])
+def test_read_hdf_series_mode_r(tmp_path, format, setup_path):
+    # GH 16583
+    # Tests that reading a Series saved to an HDF file
+    # still works if a mode='r' argument is supplied
+    series = Series(range(10), dtype=np.float64)
+    path = tmp_path / setup_path
+    series.to_hdf(path, key="data", format=format)
+    result = read_hdf(path, key="data", mode="r")
+    tm.assert_series_equal(result, series)
+
+
+def test_read_infer_string(tmp_path, setup_path):
+    # GH#54431
+    df = DataFrame({"a": ["a", "b", None]})
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="data", format="table")
+    with pd.option_context("future.infer_string", True):
+        result = read_hdf(path, key="data", mode="r")
+    expected = DataFrame(
+        {"a": ["a", "b", None]},
+        dtype=pd.StringDtype(na_value=np.nan),
+        columns=Index(["a"], dtype=pd.StringDtype(na_value=np.nan)),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_hdfstore_read_datetime64_unit_s(tmp_path, setup_path):
+    # GH 59004
+    df_s = DataFrame(["2001-01-01", "2002-02-02"], dtype="datetime64[s]")
+    path = tmp_path / setup_path
+    with HDFStore(path, mode="w") as store:
+        store.put("df_s", df_s)
+    with HDFStore(path, mode="r") as store:
+        df_fromstore = store.get("df_s")
+    tm.assert_frame_equal(df_s, df_fromstore)
diff --git a/pandas/pandas/tests/io/pytables/test_retain_attributes.py b/pandas/pandas/tests/io/pytables/test_retain_attributes.py
new file mode 100644
index 0000000000000..6284b826c3cf0
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_retain_attributes.py
@@ -0,0 +1,92 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    _testing as tm,
+    date_range,
+    errors,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+pytestmark = pytest.mark.single_cpu
+
+
+def test_retain_index_attributes(setup_path, unit):
+    # GH 3499, losing frequency info on index recreation
+    dti = date_range("2000-1-1", periods=3, freq="h", unit=unit)
+    df = DataFrame({"A": Series(range(3), index=dti)})
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "data")
+        store.put("data", df, format="table")
+
+        result = store.get("data")
+        tm.assert_frame_equal(df, result)
+
+        for attr in ["freq", "tz", "name"]:
+            for idx in ["index", "columns"]:
+                assert getattr(getattr(df, idx), attr, None) == getattr(
+                    getattr(result, idx), attr, None
+                )
+
+        dti2 = date_range("2002-1-1", periods=3, freq="D", unit=unit)
+        # try to append a table with a different frequency
+        with tm.assert_produces_warning(errors.AttributeConflictWarning):
+            df2 = DataFrame({"A": Series(range(3), index=dti2)})
+            store.append("data", df2)
+
+        assert store.get_storer("data").info["index"]["freq"] is None
+
+        # this is ok
+        _maybe_remove(store, "df2")
+        dti3 = DatetimeIndex(
+            ["2001-01-01", "2001-01-02", "2002-01-01"], dtype=f"M8[{unit}]"
+        )
+        df2 = DataFrame(
+            {
+                "A": Series(
+                    range(3),
+                    index=dti3,
+                )
+            }
+        )
+        store.append("df2", df2)
+        dti4 = date_range("2002-1-1", periods=3, freq="D", unit=unit)
+        df3 = DataFrame({"A": Series(range(3), index=dti4)})
+        store.append("df2", df3)
+
+
+def test_retain_index_attributes2(tmp_path, setup_path):
+    path = tmp_path / setup_path
+
+    with tm.assert_produces_warning(errors.AttributeConflictWarning):
+        df = DataFrame(
+            {"A": Series(range(3), index=date_range("2000-1-1", periods=3, freq="h"))}
+        )
+        df.to_hdf(path, key="data", mode="w", append=True)
+        df2 = DataFrame(
+            {"A": Series(range(3), index=date_range("2002-1-1", periods=3, freq="D"))}
+        )
+
+        df2.to_hdf(path, key="data", append=True)
+
+        idx = date_range("2000-1-1", periods=3, freq="h")
+        idx.name = "foo"
+        df = DataFrame({"A": Series(range(3), index=idx)})
+        df.to_hdf(path, key="data", mode="w", append=True)
+
+    assert read_hdf(path, key="data").index.name == "foo"
+
+    with tm.assert_produces_warning(errors.AttributeConflictWarning):
+        idx2 = date_range("2001-1-1", periods=3, freq="h")
+        idx2.name = "bar"
+        df2 = DataFrame({"A": Series(range(3), index=idx2)})
+        df2.to_hdf(path, key="data", append=True)
+
+    assert read_hdf(path, "data").index.name is None
diff --git a/pandas/pandas/tests/io/pytables/test_round_trip.py b/pandas/pandas/tests/io/pytables/test_round_trip.py
new file mode 100644
index 0000000000000..409b92d2ddde1
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_round_trip.py
@@ -0,0 +1,591 @@
+import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas.compat import is_platform_windows
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Series,
+    _testing as tm,
+    bdate_range,
+    date_range,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+from pandas.util import _test_decorators as td
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_conv_read_write():
+    with tm.ensure_clean() as path:
+
+        def roundtrip(key, obj, **kwargs):
+            obj.to_hdf(path, key=key, **kwargs)
+            return read_hdf(path, key)
+
+        o = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        tm.assert_series_equal(o, roundtrip("series", o))
+
+        o = Series(range(10), dtype="float64", index=[f"i_{i}" for i in range(10)])
+        tm.assert_series_equal(o, roundtrip("string_series", o))
+
+        o = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        tm.assert_frame_equal(o, roundtrip("frame", o))
+
+        # table
+        df = DataFrame({"A": range(5), "B": range(5)})
+        df.to_hdf(path, key="table", append=True)
+        result = read_hdf(path, "table", where=["index>2"])
+        tm.assert_frame_equal(df[df.index > 2], result)
+
+
+def test_long_strings(setup_path):
+    # GH6166
+    data = ["a" * 50] * 10
+    df = DataFrame({"a": data}, index=data)
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df, data_columns=["a"])
+
+        result = store.select("df")
+        tm.assert_frame_equal(df, result)
+
+
+def test_api(tmp_path, setup_path):
+    # GH4584
+    # API issue when to_hdf doesn't accept append AND format args
+    path = tmp_path / setup_path
+
+    df = DataFrame(range(20))
+    df.iloc[:10].to_hdf(path, key="df", append=True, format="table")
+    df.iloc[10:].to_hdf(path, key="df", append=True, format="table")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    # append to False
+    df.iloc[:10].to_hdf(path, key="df", append=False, format="table")
+    df.iloc[10:].to_hdf(path, key="df", append=True, format="table")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+
+def test_api_append(tmp_path, setup_path):
+    path = tmp_path / setup_path
+
+    df = DataFrame(range(20))
+    df.iloc[:10].to_hdf(path, key="df", append=True)
+    df.iloc[10:].to_hdf(path, key="df", append=True, format="table")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    # append to False
+    df.iloc[:10].to_hdf(path, key="df", append=False, format="table")
+    df.iloc[10:].to_hdf(path, key="df", append=True)
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+
+def test_api_2(tmp_path, setup_path):
+    path = tmp_path / setup_path
+
+    df = DataFrame(range(20))
+    df.to_hdf(path, key="df", append=False, format="fixed")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    df.to_hdf(path, key="df", append=False, format="f")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    df.to_hdf(path, key="df", append=False)
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    df.to_hdf(path, key="df")
+    tm.assert_frame_equal(read_hdf(path, "df"), df)
+
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(range(20))
+
+        _maybe_remove(store, "df")
+        store.append("df", df.iloc[:10], append=True, format="table")
+        store.append("df", df.iloc[10:], append=True, format="table")
+        tm.assert_frame_equal(store.select("df"), df)
+
+        # append to False
+        _maybe_remove(store, "df")
+        store.append("df", df.iloc[:10], append=False, format="table")
+        store.append("df", df.iloc[10:], append=True, format="table")
+        tm.assert_frame_equal(store.select("df"), df)
+
+        # formats
+        _maybe_remove(store, "df")
+        store.append("df", df.iloc[:10], append=False, format="table")
+        store.append("df", df.iloc[10:], append=True, format="table")
+        tm.assert_frame_equal(store.select("df"), df)
+
+        _maybe_remove(store, "df")
+        store.append("df", df.iloc[:10], append=False, format="table")
+        store.append("df", df.iloc[10:], append=True, format=None)
+        tm.assert_frame_equal(store.select("df"), df)
+
+
+def test_api_invalid(tmp_path, setup_path):
+    path = tmp_path / setup_path
+    # Invalid.
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    msg = "Can only append to Tables"
+
+    with pytest.raises(ValueError, match=msg):
+        df.to_hdf(path, key="df", append=True, format="f")
+
+    with pytest.raises(ValueError, match=msg):
+        df.to_hdf(path, key="df", append=True, format="fixed")
+
+    msg = r"invalid HDFStore format specified \[foo\]"
+
+    with pytest.raises(TypeError, match=msg):
+        df.to_hdf(path, key="df", append=True, format="foo")
+
+    with pytest.raises(TypeError, match=msg):
+        df.to_hdf(path, key="df", append=False, format="foo")
+
+    # File path doesn't exist
+    path = ""
+    msg = f"File {path} does not exist"
+
+    with pytest.raises(FileNotFoundError, match=msg):
+        read_hdf(path, "df")
+
+
+def test_get(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        left = store.get("a")
+        right = store["a"]
+        tm.assert_series_equal(left, right)
+
+        left = store.get("/a")
+        right = store["/a"]
+        tm.assert_series_equal(left, right)
+
+        with pytest.raises(KeyError, match="'No object named b in the file'"):
+            store.get("b")
+
+
+def test_put_integer(setup_path):
+    # non-date, non-string index
+    df = DataFrame(np.random.default_rng(2).standard_normal((50, 100)))
+    _check_roundtrip(df, tm.assert_frame_equal, setup_path)
+
+
+def test_table_values_dtypes_roundtrip(setup_path, using_infer_string):
+    with ensure_clean_store(setup_path) as store:
+        df1 = DataFrame({"a": [1, 2, 3]}, dtype="f8")
+        store.append("df_f8", df1)
+        tm.assert_series_equal(df1.dtypes, store["df_f8"].dtypes)
+
+        df2 = DataFrame({"a": [1, 2, 3]}, dtype="i8")
+        store.append("df_i8", df2)
+        tm.assert_series_equal(df2.dtypes, store["df_i8"].dtypes)
+
+        # incompatible dtype
+        msg = re.escape(
+            "Cannot serialize the column [a] "
+            "because its data contents are not [float] "
+            "but [integer] object dtype"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df_i8", df1)
+
+        # check creation/storage/retrieval of float32 (a bit hacky to
+        # actually create them thought)
+        df1 = DataFrame(np.array([[1], [2], [3]], dtype="f4"), columns=["A"])
+        store.append("df_f4", df1)
+        tm.assert_series_equal(df1.dtypes, store["df_f4"].dtypes)
+        assert df1.dtypes.iloc[0] == "float32"
+
+        # check with mixed dtypes
+        df1 = DataFrame(
+            {
+                c: Series(np.random.default_rng(2).integers(5), dtype=c)
+                for c in ["float32", "float64", "int32", "int64", "int16", "int8"]
+            }
+        )
+        df1["string"] = "foo"
+        df1["float322"] = 1.0
+        df1["float322"] = df1["float322"].astype("float32")
+        df1["bool"] = df1["float32"] > 0
+        df1["time_s_1"] = Timestamp("20130101")
+        df1["time_s_2"] = Timestamp("20130101 00:00:00")
+        df1["time_ms"] = Timestamp("20130101 00:00:00.000")
+        df1["time_ns"] = Timestamp("20130102 00:00:00.000000000")
+
+        store.append("df_mixed_dtypes1", df1)
+        result = store.select("df_mixed_dtypes1").dtypes.value_counts()
+        result.index = [str(i) for i in result.index]
+        str_dtype = "str" if using_infer_string else "object"
+        expected = Series(
+            {
+                "float32": 2,
+                "float64": 1,
+                "int32": 1,
+                "bool": 1,
+                "int16": 1,
+                "int8": 1,
+                "int64": 1,
+                str_dtype: 1,
+                "datetime64[s]": 2,
+                "datetime64[ms]": 1,
+                "datetime64[ns]": 1,
+            },
+            name="count",
+        )
+        result = result.sort_index()
+        expected = expected.sort_index()
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+def test_series(setup_path):
+    s = Series(range(10), dtype="float64", index=[f"i_{i}" for i in range(10)])
+    _check_roundtrip(s, tm.assert_series_equal, path=setup_path)
+
+    ts = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+    _check_roundtrip(ts, tm.assert_series_equal, path=setup_path)
+
+    ts2 = Series(ts.index, Index(ts.index))
+    _check_roundtrip(ts2, tm.assert_series_equal, path=setup_path)
+
+    ts3 = Series(ts.values, Index(np.asarray(ts.index)))
+    _check_roundtrip(
+        ts3, tm.assert_series_equal, path=setup_path, check_index_type=False
+    )
+
+
+def test_float_index(setup_path):
+    # GH #454
+    index = np.random.default_rng(2).standard_normal(10)
+    s = Series(np.random.default_rng(2).standard_normal(10), index=index)
+    _check_roundtrip(s, tm.assert_series_equal, path=setup_path)
+
+
+def test_tuple_index(setup_path, performance_warning):
+    # GH #492
+    col = np.arange(10)
+    idx = [(0.0, 1.0), (2.0, 3.0), (4.0, 5.0)]
+    data = np.random.default_rng(2).standard_normal(30).reshape((3, 10))
+    DF = DataFrame(data, index=idx, columns=col)
+
+    with tm.assert_produces_warning(performance_warning):
+        _check_roundtrip(DF, tm.assert_frame_equal, path=setup_path)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.PerformanceWarning")
+def test_index_types(setup_path):
+    values = np.random.default_rng(2).standard_normal(2)
+
+    func = lambda lhs, rhs: tm.assert_series_equal(lhs, rhs, check_index_type=True)
+
+    ser = Series(values, [0, "y"])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [datetime.datetime.today(), 0])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, ["y", 0])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [datetime.date.today(), "a"])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [0, "y"])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [datetime.datetime.today(), 0])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, ["y", 0])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [datetime.date.today(), "a"])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [1.23, "b"])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [1, 1.53])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser = Series(values, [1, 5])
+    _check_roundtrip(ser, func, path=setup_path)
+
+    dti = DatetimeIndex(["2012-01-01", "2012-01-02"], dtype="M8[ns]")
+    ser = Series(values, index=dti)
+    _check_roundtrip(ser, func, path=setup_path)
+
+    ser.index = ser.index.as_unit("s")
+    _check_roundtrip(ser, func, path=setup_path)
+
+
+def test_timeseries_preepoch(setup_path, request):
+    dr = bdate_range("1/1/1940", "1/1/1960")
+    ts = Series(np.random.default_rng(2).standard_normal(len(dr)), index=dr)
+    try:
+        _check_roundtrip(ts, tm.assert_series_equal, path=setup_path)
+    except OverflowError:
+        if is_platform_windows():
+            request.applymarker(
+                pytest.mark.xfail("known failure on some windows platforms")
+            )
+        raise
+
+
+@pytest.mark.parametrize(
+    "compression", [False, pytest.param(True, marks=td.skip_if_windows)]
+)
+def test_frame(compression, setup_path):
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    # put in some random NAs
+    df.iloc[0, 0] = np.nan
+    df.iloc[5, 3] = np.nan
+
+    _check_roundtrip_table(
+        df, tm.assert_frame_equal, path=setup_path, compression=compression
+    )
+    _check_roundtrip(
+        df, tm.assert_frame_equal, path=setup_path, compression=compression
+    )
+
+    tdf = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    _check_roundtrip(
+        tdf, tm.assert_frame_equal, path=setup_path, compression=compression
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        # not consolidated
+        df["foo"] = np.random.default_rng(2).standard_normal(len(df))
+        store["df"] = df
+        recons = store["df"]
+        assert recons._mgr.is_consolidated()
+
+    # empty
+    df2 = df[:0]
+    # Prevent df2 from having index with inferred_type as string
+    df2.index = Index([])
+    _check_roundtrip(df2[:0], tm.assert_frame_equal, path=setup_path)
+
+
+def test_empty_series_frame(setup_path):
+    s0 = Series(dtype=object)
+    s1 = Series(name="myseries", dtype=object)
+    df0 = DataFrame()
+    df1 = DataFrame(index=["a", "b", "c"])
+    df2 = DataFrame(columns=["d", "e", "f"])
+
+    _check_roundtrip(s0, tm.assert_series_equal, path=setup_path)
+    _check_roundtrip(s1, tm.assert_series_equal, path=setup_path)
+    _check_roundtrip(df0, tm.assert_frame_equal, path=setup_path)
+    _check_roundtrip(df1, tm.assert_frame_equal, path=setup_path)
+    _check_roundtrip(df2, tm.assert_frame_equal, path=setup_path)
+
+
+@pytest.mark.parametrize("dtype", [np.int64, np.float64, object, "m8[ns]", "M8[ns]"])
+def test_empty_series(dtype, setup_path):
+    s = Series(dtype=dtype)
+    _check_roundtrip(s, tm.assert_series_equal, path=setup_path)
+
+
+def test_can_serialize_dates(setup_path):
+    rng = [x.date() for x in bdate_range("1/1/2000", "1/30/2000")]
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+    )
+
+    _check_roundtrip(frame, tm.assert_frame_equal, path=setup_path)
+
+
+def test_store_hierarchical(
+    setup_path, using_infer_string, multiindex_dataframe_random_data
+):
+    frame = multiindex_dataframe_random_data
+
+    if using_infer_string:
+        # TODO(infer_string) make this work for string dtype
+        msg = "Saving a MultiIndex with an extension dtype is not supported."
+        with pytest.raises(NotImplementedError, match=msg):
+            _check_roundtrip(frame, tm.assert_frame_equal, path=setup_path)
+        return
+    _check_roundtrip(frame, tm.assert_frame_equal, path=setup_path)
+    _check_roundtrip(frame.T, tm.assert_frame_equal, path=setup_path)
+    _check_roundtrip(frame["A"], tm.assert_series_equal, path=setup_path)
+
+    # check that the names are stored
+    with ensure_clean_store(setup_path) as store:
+        store["frame"] = frame
+        recons = store["frame"]
+        tm.assert_frame_equal(recons, frame)
+
+
+@pytest.mark.parametrize(
+    "compression", [False, pytest.param(True, marks=td.skip_if_windows)]
+)
+def test_store_mixed(compression, setup_path):
+    def _make_one():
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        df["obj1"] = "foo"
+        df["obj2"] = "bar"
+        df["bool1"] = df["A"] > 0
+        df["bool2"] = df["B"] > 0
+        df["int1"] = 1
+        df["int2"] = 2
+        return df._consolidate()
+
+    df1 = _make_one()
+    df2 = _make_one()
+
+    _check_roundtrip(df1, tm.assert_frame_equal, path=setup_path)
+    _check_roundtrip(df2, tm.assert_frame_equal, path=setup_path)
+
+    with ensure_clean_store(setup_path) as store:
+        store["obj"] = df1
+        tm.assert_frame_equal(store["obj"], df1)
+        store["obj"] = df2
+        tm.assert_frame_equal(store["obj"], df2)
+
+    # check that can store Series of all of these types
+    _check_roundtrip(
+        df1["obj1"],
+        tm.assert_series_equal,
+        path=setup_path,
+        compression=compression,
+    )
+    _check_roundtrip(
+        df1["bool1"],
+        tm.assert_series_equal,
+        path=setup_path,
+        compression=compression,
+    )
+    _check_roundtrip(
+        df1["int1"],
+        tm.assert_series_equal,
+        path=setup_path,
+        compression=compression,
+    )
+
+
+def _check_roundtrip(obj, comparator, path, compression=False, **kwargs):
+    options = {}
+    if compression:
+        options["complib"] = "blosc"
+
+    with ensure_clean_store(path, "w", **options) as store:
+        store["obj"] = obj
+        retrieved = store["obj"]
+        comparator(retrieved, obj, **kwargs)
+
+
+def _check_roundtrip_table(obj, comparator, path, compression=False):
+    options = {}
+    if compression:
+        options["complib"] = "blosc"
+
+    with ensure_clean_store(path, "w", **options) as store:
+        store.put("obj", obj, format="table")
+        retrieved = store["obj"]
+
+        comparator(retrieved, obj)
+
+
+def test_unicode_index(setup_path):
+    unicode_values = ["\u03c3", "\u03c3\u03c3"]
+
+    s = Series(
+        np.random.default_rng(2).standard_normal(len(unicode_values)),
+        unicode_values,
+    )
+    _check_roundtrip(s, tm.assert_series_equal, path=setup_path)
+
+
+def test_unicode_longer_encoded(setup_path):
+    # GH 11234
+    char = "\u0394"
+    df = DataFrame({"A": [char]})
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df, format="table", encoding="utf-8")
+        result = store.get("df")
+        tm.assert_frame_equal(result, df)
+
+    df = DataFrame({"A": ["a", char], "B": ["b", "b"]})
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df, format="table", encoding="utf-8")
+        result = store.get("df")
+        tm.assert_frame_equal(result, df)
+
+
+def test_store_datetime_mixed(setup_path):
+    df = DataFrame({"a": [1, 2, 3], "b": [1.0, 2.0, 3.0], "c": ["a", "b", "c"]})
+    ts = Series(
+        np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+    )
+    df["d"] = ts.index[:3]
+    _check_roundtrip(df, tm.assert_frame_equal, path=setup_path)
+
+
+def test_round_trip_equals(tmp_path, setup_path):
+    # GH 9330
+    df = DataFrame({"B": [1, 2], "A": ["x", "y"]})
+
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table")
+    other = read_hdf(path, "df")
+    tm.assert_frame_equal(df, other)
+    assert df.equals(other)
+    assert other.equals(df)
+
+
+def test_infer_string_columns(tmp_path, setup_path):
+    # GH#
+    pytest.importorskip("pyarrow")
+    path = tmp_path / setup_path
+    with pd.option_context("future.infer_string", True):
+        df = DataFrame(1, columns=list("ABCD"), index=list(range(10))).set_index(
+            ["A", "B"]
+        )
+        expected = df.copy()
+        df.to_hdf(path, key="df", format="table")
+
+        result = read_hdf(path, "df")
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/pytables/test_select.py b/pandas/pandas/tests/io/pytables/test_select.py
new file mode 100644
index 0000000000000..5e76aae28c147
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_select.py
@@ -0,0 +1,1061 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas.compat import PY312
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    HDFStore,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+    bdate_range,
+    concat,
+    date_range,
+    isna,
+    read_hdf,
+)
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+from pandas.io.pytables import Term
+
+pytestmark = [pytest.mark.single_cpu]
+
+
+def test_select_columns_in_where(setup_path):
+    # GH 6169
+    # recreate multi-indexes when columns is passed
+    # in the `where` argument
+    index = MultiIndex(
+        levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],
+        codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
+        names=["foo_name", "bar_name"],
+    )
+
+    # With a DataFrame
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 3)),
+        index=index,
+        columns=["A", "B", "C"],
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df, format="table")
+        expected = df[["A"]]
+
+        tm.assert_frame_equal(store.select("df", columns=["A"]), expected)
+
+        tm.assert_frame_equal(store.select("df", where="columns=['A']"), expected)
+
+    # With a Series
+    s = Series(np.random.default_rng(2).standard_normal(10), index=index, name="A")
+    with ensure_clean_store(setup_path) as store:
+        store.put("s", s, format="table")
+        tm.assert_series_equal(store.select("s", where="columns=['A']"), s)
+
+
+def test_select_with_dups(setup_path):
+    # single dtypes
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)), columns=["A", "A", "B", "B"]
+    )
+    df.index = date_range("20130101 9:30", periods=10, freq="min")
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df)
+
+        result = store.select("df")
+        expected = df
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+        result = store.select("df", columns=df.columns)
+        expected = df
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+        result = store.select("df", columns=["A"])
+        expected = df.loc[:, ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+    # dups across dtypes
+    df = concat(
+        [
+            DataFrame(
+                np.random.default_rng(2).standard_normal((10, 4)),
+                columns=["A", "A", "B", "B"],
+            ),
+            DataFrame(
+                np.random.default_rng(2).integers(0, 10, size=20).reshape(10, 2),
+                columns=["A", "C"],
+            ),
+        ],
+        axis=1,
+    )
+    df.index = date_range("20130101 9:30", periods=10, freq="min")
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df)
+
+        result = store.select("df")
+        expected = df
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+        result = store.select("df", columns=df.columns)
+        expected = df
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+        expected = df.loc[:, ["A"]]
+        result = store.select("df", columns=["A"])
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+        expected = df.loc[:, ["B", "A"]]
+        result = store.select("df", columns=["B", "A"])
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+    # duplicates on both index and columns
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df)
+        store.append("df", df)
+
+        expected = df.loc[:, ["B", "A"]]
+        expected = concat([expected, expected])
+        result = store.select("df", columns=["B", "A"])
+        tm.assert_frame_equal(result, expected, by_blocks=True)
+
+
+def test_select(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # select with columns=
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", df)
+        result = store.select("df", columns=["A", "B"])
+        expected = df.reindex(columns=["A", "B"])
+        tm.assert_frame_equal(expected, result)
+
+        # equivalently
+        result = store.select("df", [("columns=['A', 'B']")])
+        expected = df.reindex(columns=["A", "B"])
+        tm.assert_frame_equal(expected, result)
+
+        # with a data column
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["A"])
+        result = store.select("df", ["A > 0"], columns=["A", "B"])
+        expected = df[df.A > 0].reindex(columns=["A", "B"])
+        tm.assert_frame_equal(expected, result)
+
+        # all a data columns
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=True)
+        result = store.select("df", ["A > 0"], columns=["A", "B"])
+        expected = df[df.A > 0].reindex(columns=["A", "B"])
+        tm.assert_frame_equal(expected, result)
+
+        # with a data column, but different columns
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["A"])
+        result = store.select("df", ["A > 0"], columns=["C", "D"])
+        expected = df[df.A > 0].reindex(columns=["C", "D"])
+        tm.assert_frame_equal(expected, result)
+
+
+def test_select_dtypes(setup_path, request):
+    with ensure_clean_store(setup_path) as store:
+        # with a Timestamp data column (GH #2637)
+        df = DataFrame(
+            {
+                "ts": bdate_range("2012-01-01", periods=300),
+                "A": np.random.default_rng(2).standard_normal(300),
+            }
+        )
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["ts", "A"])
+
+        result = store.select("df", "ts>=Timestamp('2012-02-01')")
+        expected = df[df.ts >= Timestamp("2012-02-01")]
+        tm.assert_frame_equal(expected, result)
+
+        # bool columns (GH #2849)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 2)), columns=["A", "B"]
+        )
+        df["object"] = "foo"
+        df.loc[4:5, "object"] = "bar"
+        df["boolv"] = df["A"] > 0
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=True)
+
+        expected = df[df.boolv == True].reindex(columns=["A", "boolv"])  # noqa: E712
+        for v in [True, "true", 1]:
+            result = store.select("df", f"boolv == {v}", columns=["A", "boolv"])
+            tm.assert_frame_equal(expected, result)
+
+        expected = df[df.boolv == False].reindex(columns=["A", "boolv"])  # noqa: E712
+        for v in [False, "false", 0]:
+            result = store.select("df", f"boolv == {v}", columns=["A", "boolv"])
+            tm.assert_frame_equal(expected, result)
+
+        # integer index
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).random(20),
+                "B": np.random.default_rng(2).random(20),
+            }
+        )
+        _maybe_remove(store, "df_int")
+        store.append("df_int", df)
+        result = store.select("df_int", "index<10 and columns=['A']")
+        expected = df.reindex(index=list(df.index)[0:10], columns=["A"])
+        tm.assert_frame_equal(expected, result)
+
+        # float index
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).random(20),
+                "B": np.random.default_rng(2).random(20),
+                "index": np.arange(20, dtype="f8"),
+            }
+        )
+        _maybe_remove(store, "df_float")
+        store.append("df_float", df)
+        result = store.select("df_float", "index<10.0 and columns=['A']")
+        expected = df.reindex(index=list(df.index)[0:10], columns=["A"])
+        tm.assert_frame_equal(expected, result)
+
+    with ensure_clean_store(setup_path) as store:
+        # floats w/o NaN
+        df = DataFrame({"cols": range(11), "values": range(11)}, dtype="float64")
+        df["cols"] = (df["cols"] + 10).apply(str)
+
+        store.append("df1", df, data_columns=True)
+        result = store.select("df1", where="values>2.0")
+        expected = df[df["values"] > 2.0]
+        tm.assert_frame_equal(expected, result)
+
+        # floats with NaN
+        df.iloc[0] = np.nan
+        expected = df[df["values"] > 2.0]
+
+        store.append("df2", df, data_columns=True, index=False)
+        result = store.select("df2", where="values>2.0")
+        tm.assert_frame_equal(expected, result)
+
+        # https://github.com/PyTables/PyTables/issues/282
+        # bug in selection when 0th row has a np.nan and an index
+        # store.append('df3',df,data_columns=True)
+        # result = store.select(
+        #    'df3', where='values>2.0')
+        # tm.assert_frame_equal(expected, result)
+
+        # not in first position float with NaN ok too
+        df = DataFrame({"cols": range(11), "values": range(11)}, dtype="float64")
+        df["cols"] = (df["cols"] + 10).apply(str)
+
+        df.iloc[1] = np.nan
+        expected = df[df["values"] > 2.0]
+
+        store.append("df4", df, data_columns=True)
+        result = store.select("df4", where="values>2.0")
+        tm.assert_frame_equal(expected, result)
+
+    # test selection with comparison against numpy scalar
+    # GH 11283
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+
+        expected = df[df["A"] > 0]
+
+        store.append("df", df, data_columns=True)
+        request.applymarker(
+            pytest.mark.xfail(
+                PY312,
+                reason="AST change in PY312",
+                raises=ValueError,
+            )
+        )
+        np_zero = np.float64(0)  # noqa: F841
+        result = store.select("df", where=["A>np_zero"])
+        tm.assert_frame_equal(expected, result)
+
+
+def test_select_with_many_inputs(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            {
+                "ts": bdate_range("2012-01-01", periods=300),
+                "A": np.random.default_rng(2).standard_normal(300),
+                "B": range(300),
+                "users": ["a"] * 50
+                + ["b"] * 50
+                + ["c"] * 100
+                + [f"a{i:03d}" for i in range(100)],
+            }
+        )
+        _maybe_remove(store, "df")
+        store.append("df", df, data_columns=["ts", "A", "B", "users"])
+
+        # regular select
+        result = store.select("df", "ts>=Timestamp('2012-02-01')")
+        expected = df[df.ts >= Timestamp("2012-02-01")]
+        tm.assert_frame_equal(expected, result)
+
+        # small selector
+        result = store.select("df", "ts>=Timestamp('2012-02-01') & users=['a','b','c']")
+        expected = df[
+            (df.ts >= Timestamp("2012-02-01")) & df.users.isin(["a", "b", "c"])
+        ]
+        tm.assert_frame_equal(expected, result)
+
+        # big selector along the columns
+        selector = ["a", "b", "c"] + [f"a{i:03d}" for i in range(60)]
+        result = store.select("df", "ts>=Timestamp('2012-02-01') and users=selector")
+        expected = df[(df.ts >= Timestamp("2012-02-01")) & df.users.isin(selector)]
+        tm.assert_frame_equal(expected, result)
+
+        selector = range(100, 200)
+        result = store.select("df", "B=selector")
+        expected = df[df.B.isin(selector)]
+        tm.assert_frame_equal(expected, result)
+        assert len(result) == 100
+
+        # big selector along the index
+        selector = Index(df.ts[0:100].values)
+        result = store.select("df", "ts=selector")
+        expected = df[df.ts.isin(selector.values)]
+        tm.assert_frame_equal(expected, result)
+        assert len(result) == 100
+
+
+def test_select_iterator(tmp_path, setup_path):
+    # single table
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", df)
+
+        expected = store.select("df")
+
+        results = list(store.select("df", iterator=True))
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+        results = list(store.select("df", chunksize=2))
+        assert len(results) == 5
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+        results = list(store.select("df", chunksize=2))
+        result = concat(results)
+        tm.assert_frame_equal(result, expected)
+
+    path = tmp_path / setup_path
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df.to_hdf(path, key="df_non_table")
+
+    msg = "can only use an iterator or chunksize on a table"
+    with pytest.raises(TypeError, match=msg):
+        read_hdf(path, "df_non_table", chunksize=2)
+
+    with pytest.raises(TypeError, match=msg):
+        read_hdf(path, "df_non_table", iterator=True)
+
+    path = tmp_path / setup_path
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df.to_hdf(path, key="df", format="table")
+
+    results = list(read_hdf(path, "df", chunksize=2))
+    result = concat(results)
+
+    assert len(results) == 5
+    tm.assert_frame_equal(result, df)
+    tm.assert_frame_equal(result, read_hdf(path, "df"))
+
+    # multiple
+
+    with ensure_clean_store(setup_path) as store:
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        store.append("df1", df1, data_columns=True)
+        df2 = df1.copy().rename(columns="{}_2".format)
+        df2["foo"] = "bar"
+        store.append("df2", df2)
+
+        df = concat([df1, df2], axis=1)
+
+        # full selection
+        expected = store.select_as_multiple(["df1", "df2"], selector="df1")
+        results = list(
+            store.select_as_multiple(["df1", "df2"], selector="df1", chunksize=2)
+        )
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+
+def test_select_iterator_complete_8014(setup_path):
+    # GH 8014
+    # using iterator and where clause
+    chunksize = 1e4
+
+    # no iterator
+    with ensure_clean_store(setup_path) as store:
+        expected = DataFrame(
+            np.random.default_rng(2).standard_normal((100064, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=100064, freq="s"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", expected)
+
+        beg_dt = expected.index[0]
+        end_dt = expected.index[-1]
+
+        # select w/o iteration and no where clause works
+        result = store.select("df")
+        tm.assert_frame_equal(expected, result)
+
+        # select w/o iterator and where clause, single term, begin
+        # of range, works
+        where = f"index >= '{beg_dt}'"
+        result = store.select("df", where=where)
+        tm.assert_frame_equal(expected, result)
+
+        # select w/o iterator and where clause, single term, end
+        # of range, works
+        where = f"index <= '{end_dt}'"
+        result = store.select("df", where=where)
+        tm.assert_frame_equal(expected, result)
+
+        # select w/o iterator and where clause, inclusive range,
+        # works
+        where = f"index >= '{beg_dt}' & index <= '{end_dt}'"
+        result = store.select("df", where=where)
+        tm.assert_frame_equal(expected, result)
+
+    # with iterator, full range
+    with ensure_clean_store(setup_path) as store:
+        expected = DataFrame(
+            np.random.default_rng(2).standard_normal((100064, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=100064, freq="s"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", expected)
+
+        beg_dt = expected.index[0]
+        end_dt = expected.index[-1]
+
+        # select w/iterator and no where clause works
+        results = list(store.select("df", chunksize=chunksize))
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+        # select w/iterator and where clause, single term, begin of range
+        where = f"index >= '{beg_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+        # select w/iterator and where clause, single term, end of range
+        where = f"index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+        # select w/iterator and where clause, inclusive range
+        where = f"index >= '{beg_dt}' & index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        tm.assert_frame_equal(expected, result)
+
+
+def test_select_iterator_non_complete_8014(setup_path):
+    # GH 8014
+    # using iterator and where clause
+    chunksize = 1e4
+
+    # with iterator, non complete range
+    with ensure_clean_store(setup_path) as store:
+        expected = DataFrame(
+            np.random.default_rng(2).standard_normal((100064, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=100064, freq="s"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", expected)
+
+        beg_dt = expected.index[1]
+        end_dt = expected.index[-2]
+
+        # select w/iterator and where clause, single term, begin of range
+        where = f"index >= '{beg_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        rexpected = expected[expected.index >= beg_dt]
+        tm.assert_frame_equal(rexpected, result)
+
+        # select w/iterator and where clause, single term, end of range
+        where = f"index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        rexpected = expected[expected.index <= end_dt]
+        tm.assert_frame_equal(rexpected, result)
+
+        # select w/iterator and where clause, inclusive range
+        where = f"index >= '{beg_dt}' & index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        rexpected = expected[(expected.index >= beg_dt) & (expected.index <= end_dt)]
+        tm.assert_frame_equal(rexpected, result)
+
+    # with iterator, empty where
+    with ensure_clean_store(setup_path) as store:
+        expected = DataFrame(
+            np.random.default_rng(2).standard_normal((100064, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=100064, freq="s"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", expected)
+
+        end_dt = expected.index[-1]
+
+        # select w/iterator and where clause, single term, begin of range
+        where = f"index > '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        assert 0 == len(results)
+
+
+def test_select_iterator_many_empty_frames(setup_path):
+    # GH 8014
+    # using iterator and where clause can return many empty
+    # frames.
+    chunksize = 10_000
+
+    # with iterator, range limited to the first chunk
+    with ensure_clean_store(setup_path) as store:
+        expected = DataFrame(
+            np.random.default_rng(2).standard_normal((100064, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=100064, freq="s"),
+        )
+        _maybe_remove(store, "df")
+        store.append("df", expected)
+
+        beg_dt = expected.index[0]
+        end_dt = expected.index[chunksize - 1]
+
+        # select w/iterator and where clause, single term, begin of range
+        where = f"index >= '{beg_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+        result = concat(results)
+        rexpected = expected[expected.index >= beg_dt]
+        tm.assert_frame_equal(rexpected, result)
+
+        # select w/iterator and where clause, single term, end of range
+        where = f"index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+
+        assert len(results) == 1
+        result = concat(results)
+        rexpected = expected[expected.index <= end_dt]
+        tm.assert_frame_equal(rexpected, result)
+
+        # select w/iterator and where clause, inclusive range
+        where = f"index >= '{beg_dt}' & index <= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+
+        # should be 1, is 10
+        assert len(results) == 1
+        result = concat(results)
+        rexpected = expected[(expected.index >= beg_dt) & (expected.index <= end_dt)]
+        tm.assert_frame_equal(rexpected, result)
+
+        # select w/iterator and where clause which selects
+        # *nothing*.
+        #
+        # To be consistent with Python idiom I suggest this should
+        # return [] e.g. `for e in []: print True` never prints
+        # True.
+
+        where = f"index <= '{beg_dt}' & index >= '{end_dt}'"
+        results = list(store.select("df", where=where, chunksize=chunksize))
+
+        # should be []
+        assert len(results) == 0
+
+
+def test_frame_select(setup_path, request):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("frame", df, format="table")
+        date = df.index[len(df) // 2]
+
+        crit1 = Term("index>=date")
+        assert crit1.env.scope["date"] == date
+
+        crit2 = "columns=['A', 'D']"
+        crit3 = "columns=A"
+
+        request.applymarker(
+            pytest.mark.xfail(
+                PY312,
+                reason="AST change in PY312",
+                raises=TypeError,
+            )
+        )
+        result = store.select("frame", [crit1, crit2])
+        expected = df.loc[date:, ["A", "D"]]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("frame", [crit3])
+        expected = df.loc[:, ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+        # invalid terms
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        store.append("df_time", df)
+        msg = "day is out of range for month: 0"
+        with pytest.raises(ValueError, match=msg):
+            store.select("df_time", "index>0")
+
+        # can't select if not written as table
+        # store['frame'] = df
+        # with pytest.raises(ValueError):
+        #     store.select('frame', [crit1, crit2])
+
+
+def test_frame_select_complex(setup_path):
+    # select via complex criteria
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df["string"] = "foo"
+    df.loc[df.index[0:4], "string"] = "bar"
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df, format="table", data_columns=["string"])
+
+        # empty
+        result = store.select("df", 'index>df.index[3] & string="bar"')
+        expected = df.loc[(df.index > df.index[3]) & (df.string == "bar")]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", 'index>df.index[3] & string="foo"')
+        expected = df.loc[(df.index > df.index[3]) & (df.string == "foo")]
+        tm.assert_frame_equal(result, expected)
+
+        # or
+        result = store.select("df", 'index>df.index[3] | string="bar"')
+        expected = df.loc[(df.index > df.index[3]) | (df.string == "bar")]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select(
+            "df", '(index>df.index[3] & index<=df.index[6]) | string="bar"'
+        )
+        expected = df.loc[
+            ((df.index > df.index[3]) & (df.index <= df.index[6]))
+            | (df.string == "bar")
+        ]
+        tm.assert_frame_equal(result, expected)
+
+        # invert
+        result = store.select("df", 'string!="bar"')
+        expected = df.loc[df.string != "bar"]
+        tm.assert_frame_equal(result, expected)
+
+        # invert not implemented in numexpr :(
+        msg = "cannot use an invert condition when passing to numexpr"
+        with pytest.raises(NotImplementedError, match=msg):
+            store.select("df", '~(string="bar")')
+
+        # invert ok for filters
+        result = store.select("df", "~(columns=['A','B'])")
+        expected = df.loc[:, df.columns.difference(["A", "B"])]
+        tm.assert_frame_equal(result, expected)
+
+        # in
+        result = store.select("df", "index>df.index[3] & columns in ['A','B']")
+        expected = df.loc[df.index > df.index[3]].reindex(columns=["A", "B"])
+        tm.assert_frame_equal(result, expected)
+
+
+def test_frame_select_complex2(tmp_path):
+    pp = tmp_path / "params.hdf"
+    hh = tmp_path / "hist.hdf"
+
+    # use non-trivial selection criteria
+    params = DataFrame({"A": [1, 1, 2, 2, 3]})
+    params.to_hdf(pp, key="df", mode="w", format="table", data_columns=["A"])
+
+    selection = read_hdf(pp, "df", where="A=[2,3]")
+    hist = DataFrame(
+        np.random.default_rng(2).standard_normal((25, 1)),
+        columns=["data"],
+        index=MultiIndex.from_tuples(
+            [(i, j) for i in range(5) for j in range(5)], names=["l1", "l2"]
+        ),
+    )
+
+    hist.to_hdf(hh, key="df", mode="w", format="table")
+
+    expected = read_hdf(hh, "df", where="l1=[2, 3, 4]")
+
+    # scope with list like
+    l0 = selection.index.tolist()  # noqa: F841
+    with HDFStore(hh) as store:
+        result = store.select("df", where="l1=l0")
+        tm.assert_frame_equal(result, expected)
+
+    result = read_hdf(hh, "df", where="l1=l0")
+    tm.assert_frame_equal(result, expected)
+
+    # index
+    index = selection.index  # noqa: F841
+    result = read_hdf(hh, "df", where="l1=index")
+    tm.assert_frame_equal(result, expected)
+
+    result = read_hdf(hh, "df", where="l1=selection.index")
+    tm.assert_frame_equal(result, expected)
+
+    result = read_hdf(hh, "df", where="l1=selection.index.tolist()")
+    tm.assert_frame_equal(result, expected)
+
+    result = read_hdf(hh, "df", where="l1=list(selection.index)")
+    tm.assert_frame_equal(result, expected)
+
+    # scope with index
+    with HDFStore(hh) as store:
+        result = store.select("df", where="l1=index")
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", where="l1=selection.index")
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", where="l1=selection.index.tolist()")
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", where="l1=list(selection.index)")
+        tm.assert_frame_equal(result, expected)
+
+
+def test_invalid_filtering(setup_path):
+    # can't use more than one filter (atm)
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df, format="table")
+
+        msg = "unable to collapse Joint Filters"
+        # not implemented
+        with pytest.raises(NotImplementedError, match=msg):
+            store.select("df", "columns=['A'] | columns=['B']")
+
+        # in theory we could deal with this
+        with pytest.raises(NotImplementedError, match=msg):
+            store.select("df", "columns=['A','B'] & columns=['C']")
+
+
+def test_string_select(setup_path):
+    # GH 2973
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+
+        # test string ==/!=
+        df["x"] = "none"
+        df.loc[df.index[2:7], "x"] = ""
+
+        store.append("df", df, data_columns=["x"])
+
+        result = store.select("df", "x=none")
+        expected = df[df.x == "none"]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", "x!=none")
+        expected = df[df.x != "none"]
+        tm.assert_frame_equal(result, expected)
+
+        df2 = df.copy()
+        df2.loc[df2.x == "", "x"] = np.nan
+
+        store.append("df2", df2, data_columns=["x"])
+        result = store.select("df2", "x!=none")
+        expected = df2[isna(df2.x)]
+        tm.assert_frame_equal(result, expected)
+
+        # int ==/!=
+        df["int"] = 1
+        df.loc[df.index[2:7], "int"] = 2
+
+        store.append("df3", df, data_columns=["int"])
+
+        result = store.select("df3", "int=2")
+        expected = df[df.int == 2]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df3", "int!=2")
+        expected = df[df.int != 2]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_select_as_multiple(setup_path):
+    df1 = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    df2 = df1.copy().rename(columns="{}_2".format)
+    df2["foo"] = "bar"
+
+    with ensure_clean_store(setup_path) as store:
+        msg = "keys must be a list/tuple"
+        # no tables stored
+        with pytest.raises(TypeError, match=msg):
+            store.select_as_multiple(None, where=["A>0", "B>0"], selector="df1")
+
+        store.append("df1", df1, data_columns=["A", "B"])
+        store.append("df2", df2)
+
+        # exceptions
+        with pytest.raises(TypeError, match=msg):
+            store.select_as_multiple(None, where=["A>0", "B>0"], selector="df1")
+
+        with pytest.raises(TypeError, match=msg):
+            store.select_as_multiple([None], where=["A>0", "B>0"], selector="df1")
+
+        msg = "'No object named df3 in the file'"
+        with pytest.raises(KeyError, match=msg):
+            store.select_as_multiple(
+                ["df1", "df3"], where=["A>0", "B>0"], selector="df1"
+            )
+
+        with pytest.raises(KeyError, match=msg):
+            store.select_as_multiple(["df3"], where=["A>0", "B>0"], selector="df1")
+
+        with pytest.raises(KeyError, match="'No object named df4 in the file'"):
+            store.select_as_multiple(
+                ["df1", "df2"], where=["A>0", "B>0"], selector="df4"
+            )
+
+        # default select
+        result = store.select("df1", ["A>0", "B>0"])
+        expected = store.select_as_multiple(
+            ["df1"], where=["A>0", "B>0"], selector="df1"
+        )
+        tm.assert_frame_equal(result, expected)
+        expected = store.select_as_multiple("df1", where=["A>0", "B>0"], selector="df1")
+        tm.assert_frame_equal(result, expected)
+
+        # multiple
+        result = store.select_as_multiple(
+            ["df1", "df2"], where=["A>0", "B>0"], selector="df1"
+        )
+        expected = concat([df1, df2], axis=1)
+        expected = expected[(expected.A > 0) & (expected.B > 0)]
+        tm.assert_frame_equal(result, expected, check_freq=False)
+        # FIXME: 2021-01-20 this is failing with freq None vs 4B on some builds
+
+        # multiple (diff selector)
+        result = store.select_as_multiple(
+            ["df1", "df2"], where="index>df2.index[4]", selector="df2"
+        )
+        expected = concat([df1, df2], axis=1)
+        expected = expected[5:]
+        tm.assert_frame_equal(result, expected)
+
+        # test exception for diff rows
+        df3 = df1.copy().head(2)
+        store.append("df3", df3)
+        msg = "all tables must have exactly the same nrows!"
+        with pytest.raises(ValueError, match=msg):
+            store.select_as_multiple(
+                ["df1", "df3"], where=["A>0", "B>0"], selector="df1"
+            )
+
+
+def test_nan_selection_bug_4858(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame({"cols": range(6), "values": range(6)}, dtype="float64")
+        df["cols"] = (df["cols"] + 10).apply(str)
+        df.iloc[0] = np.nan
+
+        expected = DataFrame(
+            {"cols": ["13.0", "14.0", "15.0"], "values": [3.0, 4.0, 5.0]},
+            index=[3, 4, 5],
+        )
+
+        # write w/o the index on that particular column
+        store.append("df", df, data_columns=True, index=["cols"])
+        result = store.select("df", where="values>2.0")
+        tm.assert_frame_equal(result, expected)
+
+
+def test_query_with_nested_special_character(setup_path):
+    df = DataFrame(
+        {
+            "a": ["a", "a", "c", "b", "test & test", "c", "b", "e"],
+            "b": [1, 2, 3, 4, 5, 6, 7, 8],
+        }
+    )
+    expected = df[df.a == "test & test"]
+    with ensure_clean_store(setup_path) as store:
+        store.append("test", df, format="table", data_columns=True)
+        result = store.select("test", 'a = "test & test"')
+    tm.assert_frame_equal(expected, result)
+
+
+def test_query_long_float_literal(setup_path):
+    # GH 14241
+    df = DataFrame({"A": [1000000000.0009, 1000000000.0011, 1000000000.0015]})
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("test", df, format="table", data_columns=True)
+
+        cutoff = 1000000000.0006
+        result = store.select("test", f"A < {cutoff:.4f}")
+        assert result.empty
+
+        cutoff = 1000000000.0010
+        result = store.select("test", f"A > {cutoff:.4f}")
+        expected = df.loc[[1, 2], :]
+        tm.assert_frame_equal(expected, result)
+
+        exact = 1000000000.0011
+        result = store.select("test", f"A == {exact:.4f}")
+        expected = df.loc[[1], :]
+        tm.assert_frame_equal(expected, result)
+
+
+def test_query_compare_column_type(setup_path):
+    # GH 15492
+    df = DataFrame(
+        {
+            "date": ["2014-01-01", "2014-01-02"],
+            "real_date": date_range("2014-01-01", periods=2),
+            "float": [1.1, 1.2],
+            "int": [1, 2],
+        },
+        columns=["date", "real_date", "float", "int"],
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("test", df, format="table", data_columns=True)
+
+        ts = Timestamp("2014-01-01")  # noqa: F841
+        result = store.select("test", where="real_date > ts")
+        expected = df.loc[[1], :]
+        tm.assert_frame_equal(expected, result)
+
+        for op in ["<", ">", "=="]:
+            # non strings to string column always fail
+            for v in [2.1, True, Timestamp("2014-01-01"), pd.Timedelta(1, "s")]:
+                query = f"date {op} v"
+                msg = f"Cannot compare {v} of type {type(v)} to string column"
+                with pytest.raises(TypeError, match=msg):
+                    store.select("test", where=query)
+
+            # strings to other columns must be convertible to type
+            v = "a"
+            for col in ["int", "float", "real_date"]:
+                query = f"{col} {op} v"
+                if col == "real_date":
+                    msg = 'Given date string "a" not likely a datetime'
+                else:
+                    msg = "could not convert string to"
+                with pytest.raises(ValueError, match=msg):
+                    store.select("test", where=query)
+
+            for v, col in zip(
+                ["1", "1.1", "2014-01-01"], ["int", "float", "real_date"]
+            ):
+                query = f"{col} {op} v"
+                result = store.select("test", where=query)
+
+                if op == "==":
+                    expected = df.loc[[0], :]
+                elif op == ">":
+                    expected = df.loc[[1], :]
+                else:
+                    expected = df.loc[[], :]
+                tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize("where", ["", (), (None,), [], [None]])
+def test_select_empty_where(tmp_path, where):
+    # GH26610
+
+    df = DataFrame([1, 2, 3])
+    path = tmp_path / "empty_where.h5"
+    with HDFStore(path) as store:
+        store.put("df", df, "t")
+        result = read_hdf(store, "df", where=where)
+        tm.assert_frame_equal(result, df)
+
+
+def test_select_large_integer(tmp_path):
+    path = tmp_path / "large_int.h5"
+
+    df = DataFrame(
+        zip(
+            ["a", "b", "c", "d"],
+            [-9223372036854775801, -9223372036854775802, -9223372036854775803, 123],
+        ),
+        columns=["x", "y"],
+    )
+    with HDFStore(path) as s:
+        s.append("data", df, data_columns=True, index=False)
+        result = s.select("data", where="y==-9223372036854775801").get("y").get(0)
+    expected = df["y"][0]
+
+    assert expected == result
diff --git a/pandas/pandas/tests/io/pytables/test_store.py b/pandas/pandas/tests/io/pytables/test_store.py
new file mode 100644
index 0000000000000..bb2058c050f2a
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_store.py
@@ -0,0 +1,1108 @@
+import contextlib
+import datetime as dt
+import hashlib
+import tempfile
+import time
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import PY312
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.conftest import has_pyarrow
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+from pandas.io.pytables import (
+    HDFStore,
+    read_hdf,
+)
+
+pytestmark = [pytest.mark.single_cpu]
+
+tables = pytest.importorskip("tables")
+
+
+def test_context(setup_path):
+    with tm.ensure_clean(setup_path) as path:
+        try:
+            with HDFStore(path) as tbl:
+                raise ValueError("blah")
+        except ValueError:
+            pass
+    with tm.ensure_clean(setup_path) as path:
+        with HDFStore(path) as tbl:
+            tbl["a"] = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+            assert len(tbl) == 1
+            assert type(tbl["a"]) == DataFrame
+
+
+def test_no_track_times(tmp_path, setup_path):
+    # GH 32682
+    # enables to set track_times (see `pytables` `create_table` documentation)
+
+    def checksum(filename, hash_factory=hashlib.md5, chunk_num_blocks=128):
+        h = hash_factory()
+        with open(filename, "rb") as f:
+            for chunk in iter(lambda: f.read(chunk_num_blocks * h.block_size), b""):
+                h.update(chunk)
+        return h.digest()
+
+    def create_h5_and_return_checksum(tmp_path, track_times):
+        path = tmp_path / setup_path
+        df = DataFrame({"a": [1]})
+
+        with HDFStore(path, mode="w") as hdf:
+            hdf.put(
+                "table",
+                df,
+                format="table",
+                data_columns=True,
+                index=None,
+                track_times=track_times,
+            )
+
+        return checksum(path)
+
+    checksum_0_tt_false = create_h5_and_return_checksum(tmp_path, track_times=False)
+    checksum_0_tt_true = create_h5_and_return_checksum(tmp_path, track_times=True)
+
+    # sleep is necessary to create h5 with different creation time
+    time.sleep(1)
+
+    checksum_1_tt_false = create_h5_and_return_checksum(tmp_path, track_times=False)
+    checksum_1_tt_true = create_h5_and_return_checksum(tmp_path, track_times=True)
+
+    # checksums are the same if track_time = False
+    assert checksum_0_tt_false == checksum_1_tt_false
+
+    # checksums are NOT same if track_time = True
+    assert checksum_0_tt_true != checksum_1_tt_true
+
+
+def test_iter_empty(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # GH 12221
+        assert list(store) == []
+
+
+def test_repr(setup_path, performance_warning, using_infer_string):
+    with ensure_clean_store(setup_path) as store:
+        repr(store)
+        store.info()
+        store["a"] = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["b"] = Series(
+            range(10), dtype="float64", index=[f"i_{i}" for i in range(10)]
+        )
+        store["c"] = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        df["obj1"] = "foo"
+        df["obj2"] = "bar"
+        df["bool1"] = df["A"] > 0
+        df["bool2"] = df["B"] > 0
+        df["bool3"] = True
+        df["int1"] = 1
+        df["int2"] = 2
+        df["timestamp1"] = Timestamp("20010102")
+        df["timestamp2"] = Timestamp("20010103")
+        df["datetime1"] = dt.datetime(2001, 1, 2, 0, 0)
+        df["datetime2"] = dt.datetime(2001, 1, 3, 0, 0)
+        df.loc[df.index[3:6], ["obj1"]] = np.nan
+        df = df._consolidate()
+
+        warning = None if using_infer_string else performance_warning
+        msg = "cannot\nmap directly to c-types .* dtype='object'"
+        with tm.assert_produces_warning(warning, match=msg):
+            store["df"] = df
+
+        # make a random group in hdf space
+        store._handle.create_group(store._handle.root, "bah")
+
+        assert store.filename in repr(store)
+        assert store.filename in str(store)
+        store.info()
+
+    # storers
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        store.append("df", df)
+
+        s = store.get_storer("df")
+        repr(s)
+        str(s)
+
+
+def test_contains(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["b"] = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        store["foo/bar"] = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        assert "a" in store
+        assert "b" in store
+        assert "c" not in store
+        assert "foo/bar" in store
+        assert "/foo/bar" in store
+        assert "/foo/b" not in store
+        assert "bar" not in store
+
+        # gh-2694: tables.NaturalNameWarning
+        with tm.assert_produces_warning(
+            tables.NaturalNameWarning, check_stacklevel=False
+        ):
+            store["node())"] = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+        assert "node())" in store
+
+
+def test_versioning(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["b"] = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=20, freq="B"),
+        )
+        _maybe_remove(store, "df1")
+        store.append("df1", df[:10])
+        store.append("df1", df[10:])
+        assert store.root.a._v_attrs.pandas_version == "0.15.2"
+        assert store.root.b._v_attrs.pandas_version == "0.15.2"
+        assert store.root.df1._v_attrs.pandas_version == "0.15.2"
+
+        # write a file and wipe its versioning
+        _maybe_remove(store, "df2")
+        store.append("df2", df)
+
+        # this is an error because its table_type is appendable, but no
+        # version info
+        store.get_node("df2")._v_attrs.pandas_version = None
+
+        msg = "'NoneType' object has no attribute 'startswith'"
+
+        with pytest.raises(Exception, match=msg):
+            store.select("df2")
+
+
+@pytest.mark.parametrize(
+    "where, expected",
+    [
+        (
+            "/",
+            {
+                "": ({"first_group", "second_group"}, set()),
+                "/first_group": (set(), {"df1", "df2"}),
+                "/second_group": ({"third_group"}, {"df3", "s1"}),
+                "/second_group/third_group": (set(), {"df4"}),
+            },
+        ),
+        (
+            "/second_group",
+            {
+                "/second_group": ({"third_group"}, {"df3", "s1"}),
+                "/second_group/third_group": (set(), {"df4"}),
+            },
+        ),
+    ],
+)
+def test_walk(where, expected):
+    # GH10143
+    objs = {
+        "df1": DataFrame([1, 2, 3]),
+        "df2": DataFrame([4, 5, 6]),
+        "df3": DataFrame([6, 7, 8]),
+        "df4": DataFrame([9, 10, 11]),
+        "s1": Series([10, 9, 8]),
+        # Next 3 items aren't pandas objects and should be ignored
+        "a1": np.array([[1, 2, 3], [4, 5, 6]]),
+        "tb1": np.array([(1, 2, 3), (4, 5, 6)], dtype="i,i,i"),
+        "tb2": np.array([(7, 8, 9), (10, 11, 12)], dtype="i,i,i"),
+    }
+
+    with ensure_clean_store("walk_groups.hdf", mode="w") as store:
+        store.put("/first_group/df1", objs["df1"])
+        store.put("/first_group/df2", objs["df2"])
+        store.put("/second_group/df3", objs["df3"])
+        store.put("/second_group/s1", objs["s1"])
+        store.put("/second_group/third_group/df4", objs["df4"])
+        # Create non-pandas objects
+        store._handle.create_array("/first_group", "a1", objs["a1"])
+        store._handle.create_table("/first_group", "tb1", obj=objs["tb1"])
+        store._handle.create_table("/second_group", "tb2", obj=objs["tb2"])
+
+        assert len(list(store.walk(where=where))) == len(expected)
+        for path, groups, leaves in store.walk(where=where):
+            assert path in expected
+            expected_groups, expected_frames = expected[path]
+            assert expected_groups == set(groups)
+            assert expected_frames == set(leaves)
+            for leaf in leaves:
+                frame_path = "/".join([path, leaf])
+                obj = store.get(frame_path)
+                if "df" in leaf:
+                    tm.assert_frame_equal(obj, objs[leaf])
+                else:
+                    tm.assert_series_equal(obj, objs[leaf])
+
+
+def test_getattr(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        s = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["a"] = s
+
+        # test attribute access
+        result = store.a
+        tm.assert_series_equal(result, s)
+        result = store.a
+        tm.assert_series_equal(result, s)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        store["df"] = df
+        result = store.df
+        tm.assert_frame_equal(result, df)
+
+        # errors
+        for x in ["d", "mode", "path", "handle", "complib"]:
+            msg = f"'HDFStore' object has no attribute '{x}'"
+            with pytest.raises(AttributeError, match=msg):
+                getattr(store, x)
+
+        # not stores
+        for x in ["mode", "path", "handle", "complib"]:
+            getattr(store, f"_{x}")
+
+
+def test_store_dropna(tmp_path, setup_path):
+    df_with_missing = DataFrame(
+        {"col1": [0.0, np.nan, 2.0], "col2": [1.0, np.nan, np.nan]},
+        index=list("abc"),
+    )
+    df_without_missing = DataFrame(
+        {"col1": [0.0, 2.0], "col2": [1.0, np.nan]}, index=list("ac")
+    )
+
+    # # Test to make sure defaults are to not drop.
+    # # Corresponding to Issue 9382
+    path = tmp_path / setup_path
+    df_with_missing.to_hdf(path, key="df", format="table")
+    reloaded = read_hdf(path, "df")
+    tm.assert_frame_equal(df_with_missing, reloaded)
+
+    path = tmp_path / setup_path
+    df_with_missing.to_hdf(path, key="df", format="table", dropna=False)
+    reloaded = read_hdf(path, "df")
+    tm.assert_frame_equal(df_with_missing, reloaded)
+
+    path = tmp_path / setup_path
+    df_with_missing.to_hdf(path, key="df", format="table", dropna=True)
+    reloaded = read_hdf(path, "df")
+    tm.assert_frame_equal(df_without_missing, reloaded)
+
+
+def test_to_hdf_with_min_itemsize(tmp_path, setup_path):
+    path = tmp_path / setup_path
+
+    # min_itemsize in index with to_hdf (GH 10381)
+    df = DataFrame(
+        {
+            "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+            "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+            "C": Index(["foo1", "foo2", "foo3", "foo4", "foo5"]),
+            "D": date_range("20130101", periods=5),
+        }
+    ).set_index("C")
+    df.to_hdf(path, key="ss3", format="table", min_itemsize={"index": 6})
+    # just make sure there is a longer string:
+    df2 = df.copy().reset_index().assign(C="longer").set_index("C")
+    df2.to_hdf(path, key="ss3", append=True, format="table")
+    tm.assert_frame_equal(read_hdf(path, "ss3"), concat([df, df2]))
+
+    # same as above, with a Series
+    df["B"].to_hdf(path, key="ss4", format="table", min_itemsize={"index": 6})
+    df2["B"].to_hdf(path, key="ss4", append=True, format="table")
+    tm.assert_series_equal(read_hdf(path, "ss4"), concat([df["B"], df2["B"]]))
+
+
+@pytest.mark.xfail(
+    using_string_dtype() and has_pyarrow,
+    reason="TODO(infer_string): can't encode '\ud800': surrogates not allowed",
+)
+@pytest.mark.parametrize("format", ["fixed", "table"])
+def test_to_hdf_errors(tmp_path, format, setup_path):
+    data = ["\ud800foo"]
+    ser = Series(data, index=Index(data))
+    path = tmp_path / setup_path
+    # GH 20835
+    ser.to_hdf(path, key="table", format=format, errors="surrogatepass")
+
+    result = read_hdf(path, "table", errors="surrogatepass")
+    tm.assert_series_equal(result, ser)
+
+
+def test_create_table_index(setup_path):
+    with ensure_clean_store(setup_path) as store:
+
+        def col(t, column):
+            return getattr(store.get_storer(t).table.cols, column)
+
+        # data columns
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df["string"] = "foo"
+        df["string2"] = "bar"
+        store.append("f", df, data_columns=["string", "string2"])
+        assert col("f", "index").is_indexed is True
+        assert col("f", "string").is_indexed is True
+        assert col("f", "string2").is_indexed is True
+
+        # specify index=columns
+        store.append("f2", df, index=["string"], data_columns=["string", "string2"])
+        assert col("f2", "index").is_indexed is False
+        assert col("f2", "string").is_indexed is True
+        assert col("f2", "string2").is_indexed is False
+
+        # try to index a non-table
+        _maybe_remove(store, "f2")
+        store.put("f2", df)
+        msg = "cannot create table index on a Fixed format store"
+        with pytest.raises(TypeError, match=msg):
+            store.create_table_index("f2")
+
+
+def test_create_table_index_data_columns_argument(setup_path):
+    # GH 28156
+
+    with ensure_clean_store(setup_path) as store:
+
+        def col(t, column):
+            return getattr(store.get_storer(t).table.cols, column)
+
+        # data columns
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df["string"] = "foo"
+        df["string2"] = "bar"
+        store.append("f", df, data_columns=["string"])
+        assert col("f", "index").is_indexed is True
+        assert col("f", "string").is_indexed is True
+
+        msg = "'Cols' object has no attribute 'string2'"
+        with pytest.raises(AttributeError, match=msg):
+            col("f", "string2").is_indexed
+
+        # try to index a col which isn't a data_column
+        msg = (
+            "column string2 is not a data_column.\n"
+            "In order to read column string2 you must reload the dataframe \n"
+            "into HDFStore and include string2 with the data_columns argument."
+        )
+        with pytest.raises(AttributeError, match=msg):
+            store.create_table_index("f", columns=["string2"])
+
+
+def test_mi_data_columns(setup_path):
+    # GH 14435
+    idx = MultiIndex.from_arrays(
+        [date_range("2000-01-01", periods=5), range(5)], names=["date", "id"]
+    )
+    df = DataFrame({"a": [1.1, 1.2, 1.3, 1.4, 1.5]}, index=idx)
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df", df, data_columns=True)
+
+        actual = store.select("df", where="id == 1")
+        expected = df.iloc[[1], :]
+        tm.assert_frame_equal(actual, expected)
+
+
+def test_table_mixed_dtypes(setup_path):
+    # frame
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    df["obj1"] = "foo"
+    df["obj2"] = "bar"
+    df["bool1"] = df["A"] > 0
+    df["bool2"] = df["B"] > 0
+    df["bool3"] = True
+    df["int1"] = 1
+    df["int2"] = 2
+    df["timestamp1"] = Timestamp("20010102").as_unit("ns")
+    df["timestamp2"] = Timestamp("20010103").as_unit("ns")
+    df["datetime1"] = Timestamp("20010102").as_unit("ns")
+    df["datetime2"] = Timestamp("20010103").as_unit("ns")
+    df.loc[df.index[3:6], ["obj1"]] = np.nan
+    df = df._consolidate()
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("df1_mixed", df)
+        tm.assert_frame_equal(store.select("df1_mixed"), df)
+
+
+def test_calendar_roundtrip_issue(setup_path):
+    # 8591
+    # doc example from tseries holiday section
+    weekmask_egypt = "Sun Mon Tue Wed Thu"
+    holidays = [
+        "2012-05-01",
+        dt.datetime(2013, 5, 1),
+        np.datetime64("2014-05-01"),
+    ]
+    bday_egypt = pd.offsets.CustomBusinessDay(
+        holidays=holidays, weekmask=weekmask_egypt
+    )
+    mydt = dt.datetime(2013, 4, 30)
+    dts = date_range(mydt, periods=5, freq=bday_egypt)
+
+    s = Series(dts.weekday, dts).map(Series("Mon Tue Wed Thu Fri Sat Sun".split()))
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("fixed", s)
+        result = store.select("fixed")
+        tm.assert_series_equal(result, s)
+
+        store.append("table", s)
+        result = store.select("table")
+        tm.assert_series_equal(result, s)
+
+
+def test_remove(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        store["a"] = ts
+        store["b"] = df
+        _maybe_remove(store, "a")
+        assert len(store) == 1
+        tm.assert_frame_equal(df, store["b"])
+
+        _maybe_remove(store, "b")
+        assert len(store) == 0
+
+        # nonexistence
+        with pytest.raises(
+            KeyError, match="'No object named a_nonexistent_store in the file'"
+        ):
+            store.remove("a_nonexistent_store")
+
+        # pathing
+        store["a"] = ts
+        store["b/foo"] = df
+        _maybe_remove(store, "foo")
+        _maybe_remove(store, "b/foo")
+        assert len(store) == 1
+
+        store["a"] = ts
+        store["b/foo"] = df
+        _maybe_remove(store, "b")
+        assert len(store) == 1
+
+        # __delitem__
+        store["a"] = ts
+        store["b"] = df
+        del store["a"]
+        del store["b"]
+        assert len(store) == 0
+
+
+def test_same_name_scoping(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 2)),
+            index=date_range("20130101", periods=20),
+        )
+        store.put("df", df, format="table")
+        expected = df[df.index > Timestamp("20130105")]
+
+        result = store.select("df", "index>datetime.datetime(2013,1,5)")
+        tm.assert_frame_equal(result, expected)
+
+        # changes what 'datetime' points to in the namespace where
+        #  'select' does the lookup
+
+        # technically an error, but allow it
+        result = store.select("df", "index>datetime.datetime(2013,1,5)")
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", "index>datetime(2013,1,5)")
+        tm.assert_frame_equal(result, expected)
+
+
+def test_store_index_name(setup_path):
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    df.index.name = "foo"
+
+    with ensure_clean_store(setup_path) as store:
+        store["frame"] = df
+        recons = store["frame"]
+        tm.assert_frame_equal(recons, df)
+
+
+@pytest.mark.parametrize("tz", [None, "US/Pacific"])
+@pytest.mark.parametrize("table_format", ["table", "fixed"])
+def test_store_index_name_numpy_str(tmp_path, table_format, setup_path, unit, tz):
+    # GH #13492
+    idx = (
+        DatetimeIndex(
+            [dt.date(2000, 1, 1), dt.date(2000, 1, 2)],
+            name="cols\u05d2",
+        )
+        .tz_localize(tz)
+        .as_unit(unit)
+    )
+    idx1 = (
+        DatetimeIndex(
+            [dt.date(2010, 1, 1), dt.date(2010, 1, 2)],
+            name="rows\u05d0",
+        )
+        .as_unit(unit)
+        .tz_localize(tz)
+    )
+    df = DataFrame(np.arange(4).reshape(2, 2), columns=idx, index=idx1)
+
+    # This used to fail, returning numpy strings instead of python strings.
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format=table_format)
+    df2 = read_hdf(path, "df")
+
+    tm.assert_frame_equal(df, df2, check_names=True)
+
+    assert isinstance(df2.index.name, str)
+    assert isinstance(df2.columns.name, str)
+
+
+def test_store_series_name(setup_path):
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    series = df["A"]
+
+    with ensure_clean_store(setup_path) as store:
+        store["series"] = series
+        recons = store["series"]
+        tm.assert_series_equal(recons, series)
+
+
+def test_overwrite_node(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        store["a"] = ts
+
+        tm.assert_series_equal(store["a"], ts)
+
+
+def test_coordinates(setup_path):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+        store.append("df", df)
+
+        # all
+        c = store.select_as_coordinates("df")
+        assert (c.values == np.arange(len(df.index))).all()
+
+        # get coordinates back & test vs frame
+        _maybe_remove(store, "df")
+
+        df = DataFrame({"A": range(5), "B": range(5)})
+        store.append("df", df)
+        c = store.select_as_coordinates("df", ["index<3"])
+        assert (c.values == np.arange(3)).all()
+        result = store.select("df", where=c)
+        expected = df.loc[0:2, :]
+        tm.assert_frame_equal(result, expected)
+
+        c = store.select_as_coordinates("df", ["index>=3", "index<=4"])
+        assert (c.values == np.arange(2) + 3).all()
+        result = store.select("df", where=c)
+        expected = df.loc[3:4, :]
+        tm.assert_frame_equal(result, expected)
+        assert isinstance(c, Index)
+
+        # multiple tables
+        _maybe_remove(store, "df1")
+        _maybe_remove(store, "df2")
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD")),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df2 = df1.copy().rename(columns="{}_2".format)
+        store.append("df1", df1, data_columns=["A", "B"])
+        store.append("df2", df2)
+
+        c = store.select_as_coordinates("df1", ["A>0", "B>0"])
+        df1_result = store.select("df1", c)
+        df2_result = store.select("df2", c)
+        result = concat([df1_result, df2_result], axis=1)
+
+        expected = concat([df1, df2], axis=1)
+        expected = expected[(expected.A > 0) & (expected.B > 0)]
+        tm.assert_frame_equal(result, expected, check_freq=False)
+        # FIXME: 2021-01-18 on some (mostly windows) builds we get freq=None
+        #  but expect freq="18B"
+
+    # pass array/mask as the coordinates
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 2)),
+            index=date_range("20000101", periods=1000),
+        )
+        store.append("df", df)
+        c = store.select_column("df", "index")
+        where = c[DatetimeIndex(c).month == 5].index
+        expected = df.iloc[where]
+
+        # locations
+        result = store.select("df", where=where)
+        tm.assert_frame_equal(result, expected)
+
+        # boolean
+        result = store.select("df", where=where)
+        tm.assert_frame_equal(result, expected)
+
+        # invalid
+        msg = (
+            "where must be passed as a string, PyTablesExpr, "
+            "or list-like of PyTablesExpr"
+        )
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", where=np.arange(len(df), dtype="float64"))
+
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", where=np.arange(len(df) + 1))
+
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", where=np.arange(len(df)), start=5)
+
+        with pytest.raises(TypeError, match=msg):
+            store.select("df", where=np.arange(len(df)), start=5, stop=10)
+
+        # selection with filter
+        selection = date_range("20000101", periods=500)
+        result = store.select("df", where="index in selection")
+        expected = df[df.index.isin(selection)]
+        tm.assert_frame_equal(result, expected)
+
+        # list
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        store.append("df2", df)
+        result = store.select("df2", where=[0, 3, 5])
+        expected = df.iloc[[0, 3, 5]]
+        tm.assert_frame_equal(result, expected)
+
+        # boolean
+        where = [True] * 10
+        where[-2] = False
+        result = store.select("df2", where=where)
+        expected = df.loc[where]
+        tm.assert_frame_equal(result, expected)
+
+        # start/stop
+        result = store.select("df2", start=5, stop=10)
+        expected = df[5:10]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_start_stop_table(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # table
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).random(20),
+                "B": np.random.default_rng(2).random(20),
+            }
+        )
+        store.append("df", df)
+
+        result = store.select("df", "columns=['A']", start=0, stop=5)
+        expected = df.loc[0:4, ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+        # out of range
+        result = store.select("df", "columns=['A']", start=30, stop=40)
+        assert len(result) == 0
+        expected = df.loc[30:40, ["A"]]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_start_stop_multiple(setup_path):
+    # GH 16209
+    with ensure_clean_store(setup_path) as store:
+        df = DataFrame({"foo": [1, 2], "bar": [1, 2]})
+
+        store.append_to_multiple(
+            {"selector": ["foo"], "data": None}, df, selector="selector"
+        )
+        result = store.select_as_multiple(
+            ["selector", "data"], selector="selector", start=0, stop=1
+        )
+        expected = df.loc[[0], ["foo", "bar"]]
+        tm.assert_frame_equal(result, expected)
+
+
+def test_start_stop_fixed(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # fixed, GH 8287
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).random(20),
+                "B": np.random.default_rng(2).random(20),
+            },
+            index=date_range("20130101", periods=20),
+        )
+        store.put("df", df)
+
+        result = store.select("df", start=0, stop=5)
+        expected = df.iloc[0:5, :]
+        tm.assert_frame_equal(result, expected)
+
+        result = store.select("df", start=5, stop=10)
+        expected = df.iloc[5:10, :]
+        tm.assert_frame_equal(result, expected)
+
+        # out of range
+        result = store.select("df", start=30, stop=40)
+        expected = df.iloc[30:40, :]
+        tm.assert_frame_equal(result, expected)
+
+        # series
+        s = df.A
+        store.put("s", s)
+        result = store.select("s", start=0, stop=5)
+        expected = s.iloc[0:5]
+        tm.assert_series_equal(result, expected)
+
+        result = store.select("s", start=5, stop=10)
+        expected = s.iloc[5:10]
+        tm.assert_series_equal(result, expected)
+
+        # sparse; not implemented
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD")),
+            index=Index([f"i-{i}" for i in range(30)]),
+        )
+        df.iloc[3:5, 1:3] = np.nan
+        df.iloc[8:10, -2] = np.nan
+
+
+def test_select_filter_corner(setup_path, request):
+    df = DataFrame(np.random.default_rng(2).standard_normal((50, 100)))
+    df.index = [f"{c:3d}" for c in df.index]
+    df.columns = [f"{c:3d}" for c in df.columns]
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("frame", df, format="table")
+
+        request.applymarker(
+            pytest.mark.xfail(
+                PY312,
+                reason="AST change in PY312",
+                raises=ValueError,
+            )
+        )
+        crit = "columns=df.columns[:75]"
+        result = store.select("frame", [crit])
+        tm.assert_frame_equal(result, df.loc[:, df.columns[:75]])
+
+        crit = "columns=df.columns[:75:2]"
+        result = store.select("frame", [crit])
+        tm.assert_frame_equal(result, df.loc[:, df.columns[:75:2]])
+
+
+def test_path_pathlib():
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    result = tm.round_trip_pathlib(
+        lambda p: df.to_hdf(p, key="df"), lambda p: read_hdf(p, "df")
+    )
+    tm.assert_frame_equal(df, result)
+
+
+@pytest.mark.parametrize("start, stop", [(0, 2), (1, 2), (None, None)])
+def test_contiguous_mixed_data_table(start, stop, setup_path):
+    # GH 17021
+    df = DataFrame(
+        {
+            "a": Series([20111010, 20111011, 20111012]),
+            "b": Series(["ab", "cd", "ab"]),
+        }
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("test_dataset", df)
+
+        result = store.select("test_dataset", start=start, stop=stop)
+        tm.assert_frame_equal(df[start:stop], result)
+
+
+def test_path_pathlib_hdfstore():
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    def writer(path):
+        with HDFStore(path) as store:
+            df.to_hdf(store, key="df")
+
+    def reader(path):
+        with HDFStore(path) as store:
+            return read_hdf(store, "df")
+
+    result = tm.round_trip_pathlib(writer, reader)
+    tm.assert_frame_equal(df, result)
+
+
+def test_pickle_path_localpath():
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+    result = tm.round_trip_pathlib(
+        lambda p: df.to_hdf(p, key="df"), lambda p: read_hdf(p, "df")
+    )
+    tm.assert_frame_equal(df, result)
+
+
+@pytest.mark.parametrize("propindexes", [True, False])
+def test_copy(propindexes):
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    with tm.ensure_clean() as path:
+        with HDFStore(path) as st:
+            st.append("df", df, data_columns=["A"])
+        with tempfile.NamedTemporaryFile() as new_f:
+            with HDFStore(path) as store:
+                with contextlib.closing(
+                    store.copy(new_f.name, keys=None, propindexes=propindexes)
+                ) as tstore:
+                    # check keys
+                    keys = store.keys()
+                    assert set(keys) == set(tstore.keys())
+                    # check indices & nrows
+                    for k in tstore.keys():
+                        if tstore.get_storer(k).is_table:
+                            new_t = tstore.get_storer(k)
+                            orig_t = store.get_storer(k)
+
+                            assert orig_t.nrows == new_t.nrows
+
+                            # check propindixes
+                            if propindexes:
+                                for a in orig_t.axes:
+                                    if a.is_indexed:
+                                        assert new_t[a.name].is_indexed
+
+
+def test_duplicate_column_name(tmp_path, setup_path):
+    df = DataFrame(columns=["a", "a"], data=[[0, 0]])
+
+    path = tmp_path / setup_path
+    msg = "Columns index has to be unique for fixed format"
+    with pytest.raises(ValueError, match=msg):
+        df.to_hdf(path, key="df", format="fixed")
+
+    df.to_hdf(path, key="df", format="table")
+    other = read_hdf(path, "df")
+
+    tm.assert_frame_equal(df, other)
+    assert df.equals(other)
+    assert other.equals(df)
+
+
+def test_preserve_timedeltaindex_type(setup_path):
+    # GH9635
+    df = DataFrame(np.random.default_rng(2).normal(size=(10, 5)))
+    df.index = timedelta_range(start="0s", periods=10, freq="1s", name="example")
+
+    with ensure_clean_store(setup_path) as store:
+        store["df"] = df
+        tm.assert_frame_equal(store["df"], df)
+
+
+def test_columns_multiindex_modified(tmp_path, setup_path):
+    # BUG: 7212
+
+    df = DataFrame(
+        np.random.default_rng(2).random((4, 5)),
+        index=list("abcd"),
+        columns=list("ABCDE"),
+    )
+    df.index.name = "letters"
+    df = df.set_index(keys="E", append=True)
+
+    data_columns = df.index.names + df.columns.tolist()
+    path = tmp_path / setup_path
+    df.to_hdf(
+        path,
+        key="df",
+        mode="a",
+        append=True,
+        data_columns=data_columns,
+        index=False,
+    )
+    cols2load = list("BCD")
+    cols2load_original = list(cols2load)
+    # GH#10055 make sure read_hdf call does not alter cols2load inplace
+    read_hdf(path, "df", columns=cols2load)
+    assert cols2load_original == cols2load
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+@pytest.mark.parametrize(
+    "columns",
+    [
+        Index([0, 1], dtype=np.int64),
+        Index([0.0, 1.0], dtype=np.float64),
+        date_range("2020-01-01", periods=2),
+        timedelta_range("1 day", periods=2),
+        period_range("2020-01-01", periods=2, freq="D"),
+    ],
+)
+def test_to_hdf_with_object_column_names_should_fail(tmp_path, setup_path, columns):
+    # GH9057
+    df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)), columns=columns)
+    path = tmp_path / setup_path
+    msg = "cannot have non-object label DataIndexableCol"
+    with pytest.raises(ValueError, match=msg):
+        df.to_hdf(path, key="df", format="table", data_columns=True)
+
+
+@pytest.mark.parametrize("dtype", [None, "category"])
+def test_to_hdf_with_object_column_names_should_run(tmp_path, setup_path, dtype):
+    # GH9057
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 2)),
+        columns=Index(["a", "b"], dtype=dtype),
+    )
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="df", format="table", data_columns=True)
+    result = read_hdf(path, "df", where=f"index = [{df.index[0]}]")
+    assert len(result)
+
+
+def test_hdfstore_strides(setup_path):
+    # GH22073
+    df = DataFrame({"a": [1, 2, 3, 4], "b": [5, 6, 7, 8]})
+    with ensure_clean_store(setup_path) as store:
+        store.put("df", df)
+        assert df["a"].values.strides == store["df"]["a"].values.strides
+
+
+def test_store_bool_index(tmp_path, setup_path):
+    # GH#48667
+    df = DataFrame([[1]], columns=[True], index=Index([False], dtype="bool"))
+    expected = df.copy()
+
+    # # Test to make sure defaults are to not drop.
+    # # Corresponding to Issue 9382
+    path = tmp_path / setup_path
+    df.to_hdf(path, key="a")
+    result = read_hdf(path, "a")
+    tm.assert_frame_equal(expected, result)
diff --git a/pandas/pandas/tests/io/pytables/test_subclass.py b/pandas/pandas/tests/io/pytables/test_subclass.py
new file mode 100644
index 0000000000000..03622faa2b5a8
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_subclass.py
@@ -0,0 +1,52 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+from pandas.io.pytables import (
+    HDFStore,
+    read_hdf,
+)
+
+pytest.importorskip("tables")
+
+
+class TestHDFStoreSubclass:
+    # GH 33748
+    def test_supported_for_subclass_dataframe(self, tmp_path):
+        data = {"a": [1, 2], "b": [3, 4]}
+        sdf = tm.SubclassedDataFrame(data, dtype=np.intp)
+
+        expected = DataFrame(data, dtype=np.intp)
+
+        path = tmp_path / "temp.h5"
+        sdf.to_hdf(path, key="df")
+        result = read_hdf(path, "df")
+        tm.assert_frame_equal(result, expected)
+
+        path = tmp_path / "temp.h5"
+        with HDFStore(path) as store:
+            store.put("df", sdf)
+        result = read_hdf(path, "df")
+        tm.assert_frame_equal(result, expected)
+
+    def test_supported_for_subclass_series(self, tmp_path):
+        data = [1, 2, 3]
+        sser = tm.SubclassedSeries(data, dtype=np.intp)
+
+        expected = Series(data, dtype=np.intp)
+
+        path = tmp_path / "temp.h5"
+        sser.to_hdf(path, key="ser")
+        result = read_hdf(path, "ser")
+        tm.assert_series_equal(result, expected)
+
+        path = tmp_path / "temp.h5"
+        with HDFStore(path) as store:
+            store.put("ser", sser)
+        result = read_hdf(path, "ser")
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/io/pytables/test_time_series.py b/pandas/pandas/tests/io/pytables/test_time_series.py
new file mode 100644
index 0000000000000..726dd0d420347
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_time_series.py
@@ -0,0 +1,72 @@
+import datetime
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    _testing as tm,
+    date_range,
+    period_range,
+)
+from pandas.tests.io.pytables.common import ensure_clean_store
+
+pytestmark = pytest.mark.single_cpu
+
+
+@pytest.mark.parametrize("unit", ["us", "ns"])
+def test_store_datetime_fractional_secs(setup_path, unit):
+    dt = datetime.datetime(2012, 1, 2, 3, 4, 5, 123456)
+    dti = DatetimeIndex([dt], dtype=f"M8[{unit}]")
+    series = Series([0], index=dti)
+    with ensure_clean_store(setup_path) as store:
+        store["a"] = series
+        assert store["a"].index[0] == dt
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_tseries_indices_series(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        idx = date_range("2020-01-01", periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        store["a"] = ser
+        result = store["a"]
+
+        tm.assert_series_equal(result, ser)
+        assert result.index.freq == ser.index.freq
+        tm.assert_class_equal(result.index, ser.index, obj="series index")
+
+        idx = period_range("2020-01-01", periods=10, freq="D")
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        store["a"] = ser
+        result = store["a"]
+
+        tm.assert_series_equal(result, ser)
+        assert result.index.freq == ser.index.freq
+        tm.assert_class_equal(result.index, ser.index, obj="series index")
+
+
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_tseries_indices_frame(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        idx = date_range("2020-01-01", periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)), index=idx
+        )
+        store["a"] = df
+        result = store["a"]
+
+        tm.assert_frame_equal(result, df)
+        assert result.index.freq == df.index.freq
+        tm.assert_class_equal(result.index, df.index, obj="dataframe index")
+
+        idx = period_range("2020-01-01", periods=10, freq="D")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(idx), 3)), idx)
+        store["a"] = df
+        result = store["a"]
+
+        tm.assert_frame_equal(result, df)
+        assert result.index.freq == df.index.freq
+        tm.assert_class_equal(result.index, df.index, obj="dataframe index")
diff --git a/pandas/pandas/tests/io/pytables/test_timezones.py b/pandas/pandas/tests/io/pytables/test_timezones.py
new file mode 100644
index 0000000000000..9192804e49bd1
--- /dev/null
+++ b/pandas/pandas/tests/io/pytables/test_timezones.py
@@ -0,0 +1,347 @@
+from datetime import (
+    date,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.timezones import maybe_get_tz
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.io.pytables.common import (
+    _maybe_remove,
+    ensure_clean_store,
+)
+
+
+def _compare_with_tz(a, b):
+    tm.assert_frame_equal(a, b)
+
+    # compare the zones on each element
+    for c in a.columns:
+        for i in a.index:
+            a_e = a.loc[i, c]
+            b_e = b.loc[i, c]
+            if not (a_e == b_e and a_e.tz == b_e.tz):
+                raise AssertionError(f"invalid tz comparison [{a_e}] [{b_e}]")
+
+
+# use maybe_get_tz instead of dateutil.tz.gettz to handle the windows
+# filename issues.
+gettz_dateutil = lambda x: maybe_get_tz("dateutil/" + x)
+gettz_pytz = lambda x: x
+
+
+@pytest.mark.parametrize("gettz", [gettz_dateutil, gettz_pytz])
+def test_append_with_timezones(setup_path, gettz):
+    # as columns
+
+    # Single-tzinfo, no DST transition
+    df_est = DataFrame(
+        {
+            "A": [
+                Timestamp("20130102 2:00:00", tz=gettz("US/Eastern")).as_unit("ns")
+                + timedelta(hours=1) * i
+                for i in range(5)
+            ]
+        }
+    )
+
+    # frame with all columns having same tzinfo, but different sides
+    #  of DST transition
+    df_crosses_dst = DataFrame(
+        {
+            "A": Timestamp("20130102", tz=gettz("US/Eastern")).as_unit("ns"),
+            "B": Timestamp("20130603", tz=gettz("US/Eastern")).as_unit("ns"),
+        },
+        index=range(5),
+    )
+
+    df_mixed_tz = DataFrame(
+        {
+            "A": Timestamp("20130102", tz=gettz("US/Eastern")).as_unit("ns"),
+            "B": Timestamp("20130102", tz=gettz("EET")).as_unit("ns"),
+        },
+        index=range(5),
+    )
+
+    df_different_tz = DataFrame(
+        {
+            "A": Timestamp("20130102", tz=gettz("US/Eastern")).as_unit("ns"),
+            "B": Timestamp("20130102", tz=gettz("CET")).as_unit("ns"),
+        },
+        index=range(5),
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df_tz")
+        store.append("df_tz", df_est, data_columns=["A"])
+        result = store["df_tz"]
+        _compare_with_tz(result, df_est)
+        tm.assert_frame_equal(result, df_est)
+
+        # select with tz aware
+        expected = df_est[df_est.A >= df_est.A[3]]
+        result = store.select("df_tz", where="A>=df_est.A[3]")
+        _compare_with_tz(result, expected)
+
+        # ensure we include dates in DST and STD time here.
+        _maybe_remove(store, "df_tz")
+        store.append("df_tz", df_crosses_dst)
+        result = store["df_tz"]
+        _compare_with_tz(result, df_crosses_dst)
+        tm.assert_frame_equal(result, df_crosses_dst)
+
+        msg = (
+            r"invalid info for \[values_block_1\] for \[tz\], "
+            r"existing_value \[(dateutil/.*)?(US/Eastern|America/New_York)\] "
+            r"conflicts with new value \[(dateutil/.*)?EET\]"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df_tz", df_mixed_tz)
+
+        # this is ok
+        _maybe_remove(store, "df_tz")
+        store.append("df_tz", df_mixed_tz, data_columns=["A", "B"])
+        result = store["df_tz"]
+        _compare_with_tz(result, df_mixed_tz)
+        tm.assert_frame_equal(result, df_mixed_tz)
+
+        # can't append with diff timezone
+        msg = (
+            r"invalid info for \[B\] for \[tz\], "
+            r"existing_value \[(dateutil/.*)?EET\] "
+            r"conflicts with new value \[(dateutil/.*)?CET\]"
+        )
+        with pytest.raises(ValueError, match=msg):
+            store.append("df_tz", df_different_tz)
+
+
+@pytest.mark.parametrize("gettz", [gettz_dateutil, gettz_pytz])
+def test_append_with_timezones_as_index(setup_path, gettz):
+    # GH#4098 example
+
+    dti = date_range("2000-1-1", periods=3, freq="h", tz=gettz("US/Eastern"))
+    dti = dti._with_freq(None)  # freq doesn't round-trip
+
+    df = DataFrame({"A": Series(range(3), index=dti)})
+
+    with ensure_clean_store(setup_path) as store:
+        _maybe_remove(store, "df")
+        store.put("df", df)
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+        _maybe_remove(store, "df")
+        store.append("df", df)
+        result = store.select("df")
+        tm.assert_frame_equal(result, df)
+
+
+def test_roundtrip_tz_aware_index(setup_path, unit):
+    # GH 17618
+    ts = Timestamp("2000-01-01 01:00:00", tz="US/Eastern")
+    dti = DatetimeIndex([ts]).as_unit(unit)
+    df = DataFrame(data=[0], index=dti)
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("frame", df, format="fixed")
+        recons = store["frame"]
+        tm.assert_frame_equal(recons, df)
+
+    value = recons.index[0]._value
+    denom = {"ns": 1, "us": 1000, "ms": 10**6, "s": 10**9}[unit]
+    assert value == 946706400000000000 // denom
+
+
+def test_store_index_name_with_tz(setup_path):
+    # GH 13884
+    df = DataFrame({"A": [1, 2]})
+    df.index = DatetimeIndex([1234567890123456787, 1234567890123456788])
+    df.index = df.index.tz_localize("UTC")
+    df.index.name = "foo"
+
+    with ensure_clean_store(setup_path) as store:
+        store.put("frame", df, format="table")
+        recons = store["frame"]
+        tm.assert_frame_equal(recons, df)
+
+
+def test_tseries_select_index_column(setup_path):
+    # GH7777
+    # selecting a UTC datetimeindex column did
+    # not preserve UTC tzinfo set before storing
+
+    # check that no tz still works
+    rng = date_range("1/1/2000", "1/30/2000")
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("frame", frame)
+        result = store.select_column("frame", "index")
+        assert rng.tz == DatetimeIndex(result.values).tz
+
+    # check utc
+    rng = date_range("1/1/2000", "1/30/2000", tz="UTC")
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("frame", frame)
+        result = store.select_column("frame", "index")
+        assert rng.tz == result.dt.tz
+
+    # double check non-utc
+    rng = date_range("1/1/2000", "1/30/2000", tz="US/Eastern")
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store.append("frame", frame)
+        result = store.select_column("frame", "index")
+        assert rng.tz == result.dt.tz
+
+
+def test_timezones_fixed_format_frame_non_empty(setup_path):
+    with ensure_clean_store(setup_path) as store:
+        # index
+        rng = date_range("1/1/2000", "1/30/2000", tz="US/Eastern")
+        rng = rng._with_freq(None)  # freq doesn't round-trip
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+        )
+        store["df"] = df
+        result = store["df"]
+        tm.assert_frame_equal(result, df)
+
+        # as data
+        # GH11411
+        _maybe_remove(store, "df")
+        df = DataFrame(
+            {
+                "A": rng,
+                "B": rng.tz_convert("UTC").tz_localize(None),
+                "C": rng.tz_convert("CET"),
+                "D": range(len(rng)),
+            },
+            index=rng,
+        )
+        store["df"] = df
+        result = store["df"]
+        tm.assert_frame_equal(result, df)
+
+
+def test_timezones_fixed_format_empty(setup_path, tz_aware_fixture, frame_or_series):
+    # GH 20594
+
+    dtype = pd.DatetimeTZDtype(tz=tz_aware_fixture)
+
+    obj = Series(dtype=dtype, name="A")
+    if frame_or_series is DataFrame:
+        obj = obj.to_frame()
+
+    with ensure_clean_store(setup_path) as store:
+        store["obj"] = obj
+        result = store["obj"]
+        tm.assert_equal(result, obj)
+
+
+def test_timezones_fixed_format_series_nonempty(setup_path, tz_aware_fixture):
+    # GH 20594
+
+    dtype = pd.DatetimeTZDtype(tz=tz_aware_fixture)
+
+    with ensure_clean_store(setup_path) as store:
+        s = Series([0], dtype=dtype)
+        store["s"] = s
+        result = store["s"]
+        tm.assert_series_equal(result, s)
+
+
+def test_fixed_offset_tz(setup_path):
+    rng = date_range("1/1/2000 00:00:00-07:00", "1/30/2000 00:00:00-07:00")
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((len(rng), 4)), index=rng
+    )
+
+    with ensure_clean_store(setup_path) as store:
+        store["frame"] = frame
+        recons = store["frame"]
+        tm.assert_index_equal(recons.index, rng)
+        assert rng.tz == recons.index.tz
+
+
+@td.skip_if_windows
+def test_store_timezone(setup_path):
+    # GH2852
+    # issue storing datetime.date with a timezone as it resets when read
+    # back in a new timezone
+
+    # original method
+    with ensure_clean_store(setup_path) as store:
+        today = date(2013, 9, 10)
+        df = DataFrame([1, 2, 3], index=[today, today, today])
+        store["obj1"] = df
+        result = store["obj1"]
+        tm.assert_frame_equal(result, df)
+
+    # with tz setting
+    with ensure_clean_store(setup_path) as store:
+        with tm.set_timezone("EST5EDT"):
+            today = date(2013, 9, 10)
+            df = DataFrame([1, 2, 3], index=[today, today, today])
+            store["obj1"] = df
+
+        with tm.set_timezone("CST6CDT"):
+            result = store["obj1"]
+
+        tm.assert_frame_equal(result, df)
+
+
+def test_dst_transitions(setup_path):
+    # make sure we are not failing on transitions
+    with ensure_clean_store(setup_path) as store:
+        times = date_range(
+            "2013-10-26 23:00",
+            "2013-10-27 01:00",
+            tz="Europe/London",
+            freq="h",
+            ambiguous="infer",
+        )
+        times = times._with_freq(None)  # freq doesn't round-trip
+
+        for i in [times, times + pd.Timedelta("10min")]:
+            _maybe_remove(store, "df")
+            df = DataFrame({"A": range(len(i)), "B": i}, index=i)
+            store.append("df", df)
+            result = store.select("df")
+            tm.assert_frame_equal(result, df)
+
+
+def test_read_with_where_tz_aware_index(tmp_path, setup_path):
+    # GH 11926
+    periods = 10
+    dts = date_range("20151201", periods=periods, freq="D", tz="UTC")
+    mi = pd.MultiIndex.from_arrays([dts, range(periods)], names=["DATE", "NO"])
+    expected = DataFrame({"MYCOL": 0}, index=mi)
+
+    key = "mykey"
+    path = tmp_path / setup_path
+    with pd.HDFStore(path) as store:
+        store.append(key, expected, format="table", append=True)
+    result = pd.read_hdf(path, key, where="DATE > 20151130")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/sas/__init__.py b/pandas/pandas/tests/io/sas/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/sas/data/0x00controlbyte.sas7bdat.bz2 b/pandas/pandas/tests/io/sas/data/0x00controlbyte.sas7bdat.bz2
new file mode 100644
index 0000000000000..ef980fb907694
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/0x00controlbyte.sas7bdat.bz2 differ
diff --git a/pandas/pandas/tests/io/sas/data/0x40controlbyte.csv b/pandas/pandas/tests/io/sas/data/0x40controlbyte.csv
new file mode 100644
index 0000000000000..e81f5cc3904b7
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/0x40controlbyte.csv
@@ -0,0 +1,2 @@
+long_string_field1,long_string_field2,long_string_field3
+00000000000000000000000000000000000000000000000000,11111111111111111111111111111111111111111111111111,aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
diff --git a/pandas/pandas/tests/io/sas/data/0x40controlbyte.sas7bdat b/pandas/pandas/tests/io/sas/data/0x40controlbyte.sas7bdat
new file mode 100644
index 0000000000000..013542e282e2f
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/0x40controlbyte.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/DEMO_G.csv b/pandas/pandas/tests/io/sas/data/DEMO_G.csv
new file mode 100644
index 0000000000000..db2158a532100
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/DEMO_G.csv
@@ -0,0 +1,9757 @@
+"SEQN","SDDSRVYR","RIDSTATR","RIAGENDR","RIDAGEYR","RIDAGEMN","RIDRETH1","RIDRETH3","RIDEXMON","RIDEXAGY","RIDEXAGM","DMQMILIZ","DMQADFC","DMDBORN4","DMDCITZN","DMDYRSUS","DMDEDUC3","DMDEDUC2","DMDMARTL","RIDEXPRG","SIALANG","SIAPROXY","SIAINTRP","FIALANG","FIAPROXY","FIAINTRP","MIALANG","MIAPROXY","MIAINTRP","AIALANGA","WTINT2YR","WTMEC2YR","SDMVPSU","SDMVSTRA","INDHHIN2","INDFMIN2","INDFMPIR","DMDHHSIZ","DMDFMSIZ","DMDHHSZA","DMDHHSZB","DMDHHSZE","DMDHRGND","DMDHRAGE","DMDHRBR4","DMDHREDU","DMDHRMAR","DMDHSEDU"
+62161,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,1,2,1,2,2,1,2,2,1,102641.406474,104236.582554,1,91,14,14,3.15,5,5,0,1,0,2,50,1,5,1,5
+62162,7,2,2,3,NA,1,1,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16116.35401,3,92,4,4,0.6,6,6,2,2,0,2,24,1,3,6,NA
+62163,7,2,1,14,NA,5,6,2,14,177,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7397.684828,7869.485117,3,90,15,15,4.07,5,5,0,2,1,1,42,1,5,1,4
+62164,7,2,2,44,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,127351.373299,127965.226204,1,94,8,8,1.67,5,5,1,2,0,1,52,1,4,1,4
+62165,7,2,2,14,NA,4,4,2,14,179,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12209.74498,13384.042162,2,90,4,4,0.57,5,5,1,2,0,2,33,2,2,77,NA
+62166,7,2,1,9,NA,3,3,2,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,60593.636684,64068.123183,1,91,77,77,NA,6,6,0,4,0,1,44,1,5,1,5
+62167,7,2,1,0,11,5,6,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5024.464768,5303.683185,2,92,99,77,NA,7,4,3,3,1,1,61,2,1,1,3
+62168,7,2,1,6,NA,5,7,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5897.024603,6245.043868,2,103,14,14,3.48,5,5,0,2,1,1,43,1,4,1,5
+62169,7,2,1,21,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,1,NA,NA,NA,1,2,2,1,14391.77847,14783.600953,1,92,2,2,0.33,5,5,0,1,0,1,51,2,1,4,NA
+62170,7,2,1,15,NA,5,7,1,15,181,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7794.52699,8291.636582,3,91,15,15,5,4,4,0,2,0,1,38,2,5,1,5
+62171,7,2,1,14,NA,1,1,1,14,175,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,22768.423624,22886.980387,3,92,9,9,2.46,4,4,0,2,0,1,43,2,3,1,4
+62172,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,26960.774346,27122.911908,2,96,5,5,2.02,1,1,0,0,0,2,43,1,3,5,NA
+62173,7,2,2,2,NA,1,1,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11401.934012,12203.058423,1,95,13,13,NA,5,5,3,0,0,2,33,2,1,1,2
+62174,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,24912.668432,27335.895242,3,90,10,10,4.3,2,2,0,0,2,2,80,1,4,1,5
+62175,7,2,1,5,NA,3,3,1,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26980.605125,30440.534478,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+62176,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,53830.599426,54203.155016,1,99,15,15,5,5,5,3,0,0,2,34,1,5,1,5
+62177,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,7879.750437,7851.284287,2,92,99,77,NA,7,4,3,3,1,1,61,2,1,1,3
+62178,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,12291.154515,13189.875012,1,95,1,1,0.05,1,1,0,0,1,1,80,1,3,2,NA
+62179,7,2,1,55,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16590.074977,17115.36835,1,92,15,15,5,4,4,0,2,0,1,55,1,5,1,5
+62180,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20457.614917,22616.494827,1,97,5,5,0.87,4,4,2,0,0,1,35,1,5,1,5
+62181,7,2,1,9,NA,1,1,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,13822.148996,14860.201344,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+62182,7,1,1,75,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,63069.107216,0,1,90,12,12,NA,2,2,0,0,2,1,75,1,5,1,4
+62183,7,2,1,6,NA,4,4,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10138.00454,10211.52145,1,100,14,14,3.6,4,4,1,1,0,1,41,1,4,1,5
+62184,7,2,1,26,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15600.678771,15236.26157,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+62185,7,2,1,16,NA,1,1,1,16,201,NA,NA,1,1,NA,10,NA,NA,NA,2,2,2,2,2,2,NA,NA,NA,NA,18635.323223,19040.145288,2,103,77,77,NA,5,5,0,2,0,2,45,2,4,5,NA
+62186,7,2,2,17,NA,4,4,1,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11224.041366,11682.365019,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+62187,7,2,2,9,NA,1,1,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,10118.363218,11093.371216,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+62188,7,2,1,2,NA,2,2,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11793.948458,12167.27893,1,100,5,5,0.78,6,5,1,2,0,2,40,2,1,5,NA
+62189,7,2,2,30,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,17983.231494,18657.922524,1,94,10,10,3.04,4,4,2,0,0,2,30,1,4,1,5
+62190,7,2,2,15,NA,1,1,1,15,189,NA,NA,2,2,4,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20419.465237,21192.774678,2,102,6,3,0.54,6,4,0,4,0,2,43,2,1,5,NA
+62191,7,2,1,70,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,8661.769277,9105.621464,1,96,14,3,0.9,2,1,0,0,2,2,71,NA,NA,3,NA
+62192,7,2,2,11,NA,1,1,2,11,141,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14087.469432,14372.406512,2,97,7,7,2.05,3,3,0,2,0,2,45,1,5,2,NA
+62193,7,2,1,17,NA,4,4,2,17,209,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11386.695644,11908.648036,2,99,6,6,1.18,5,5,0,3,0,2,38,1,2,5,NA
+62194,7,2,2,9,NA,3,3,2,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,36778.822436,37956.900598,1,99,15,15,5,5,5,0,3,0,2,43,1,5,1,5
+62195,7,2,1,35,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,15842.100091,16468.990083,2,90,8,6,2.39,4,1,1,1,0,2,21,1,5,6,NA
+62196,7,2,1,1,17,3,3,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,35830.680775,40425.523018,2,95,9,9,2.22,5,5,1,0,0,1,55,1,4,1,5
+62197,7,2,2,16,NA,4,4,1,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20015.720245,20424.69013,2,102,15,15,5,4,4,0,2,0,1,44,1,3,1,1
+62198,7,2,1,7,NA,3,3,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30754.608672,32212.772113,1,98,3,3,0.61,4,4,0,2,0,2,32,1,3,6,NA
+62199,7,2,1,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,127000.852889,130891.431194,2,92,15,15,5,2,1,0,0,0,1,57,1,5,6,NA
+62200,7,2,1,42,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,20543.822351,1,97,15,15,4.07,5,5,0,3,0,1,42,2,5,1,5
+62201,7,1,2,58,NA,5,6,NA,NA,NA,2,NA,2,2,2,NA,2,2,NA,1,2,1,1,2,2,NA,NA,NA,NA,19442.276314,0,1,95,6,6,1.34,4,4,0,2,0,2,32,2,3,2,NA
+62202,7,2,1,36,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,41155.167164,40844.556107,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+62203,7,2,1,8,NA,4,4,2,8,99,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,11324.954668,11525.075349,1,93,12,12,NA,3,3,0,1,0,2,49,2,3,1,3
+62204,7,1,2,9,NA,5,6,NA,NA,NA,NA,NA,2,1,3,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5585.02957,0,2,103,15,15,5,4,4,0,2,0,1,48,2,5,1,5
+62205,7,2,1,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,103663.693967,105583.964573,1,90,15,15,5,4,4,0,1,0,2,53,1,5,1,5
+62206,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,67177.369911,67642.297015,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+62207,7,2,1,0,0,4,4,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6988.613752,7730.080563,1,100,5,5,1.05,3,3,1,0,0,2,35,1,4,6,NA
+62208,7,2,1,38,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,4,1,NA,2,2,2,1,2,2,1,2,2,2,41241.224595,41216.943466,2,102,7,7,1.53,5,5,1,2,0,2,37,2,4,1,4
+62209,7,2,2,62,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,13473.304889,14578.166065,2,96,6,6,1.11,6,6,0,2,1,1,40,2,2,1,2
+62210,7,2,1,15,NA,3,3,2,15,188,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,100370.520459,102294.664852,1,90,15,15,5,4,4,0,2,1,2,52,1,5,1,NA
+62211,7,1,2,63,NA,1,1,NA,NA,NA,2,NA,2,2,77,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,16352.915834,0,3,92,9,9,2.22,5,5,1,0,2,1,66,2,1,1,1
+62212,7,1,2,8,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9465.598219,0,2,91,12,12,NA,3,3,0,1,0,1,56,1,4,1,4
+62213,7,2,1,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5422.375171,5488.513785,2,99,4,4,0.94,3,3,1,0,0,1,48,2,3,6,NA
+62214,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,2,1,2,2,1,2,2,1,2,2,1,18723.98095,21433.166124,2,95,7,7,1.41,5,5,2,0,0,2,53,1,3,3,NA
+62215,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,129559.2554,1,98,13,13,NA,2,2,0,0,2,2,80,1,2,2,NA
+62216,7,2,1,0,6,3,3,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19442.488468,19106.370101,2,92,15,15,5,3,3,1,0,0,2,31,2,5,1,5
+62217,7,2,2,77,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,NA,15730.58404,17568.357111,2,98,4,4,0.97,3,3,0,1,1,2,77,1,1,5,NA
+62218,7,2,2,38,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,1,2,2,1,39534.635218,41046.564195,2,102,14,14,4.05,3,3,0,1,0,1,18,1,2,NA,NA
+62219,7,2,2,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11582.174418,11621.723611,2,102,4,4,0.44,7,7,1,3,0,1,48,1,9,1,9
+62220,7,2,2,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,33193.038733,33538.564898,1,96,9,9,4.92,1,1,0,0,0,2,31,1,5,5,NA
+62221,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,25818.768393,25697.012112,1,100,14,14,3.6,4,4,1,1,0,1,41,1,4,1,5
+62222,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,89973.129052,97483.088567,1,93,15,15,5,2,2,0,0,0,2,30,1,5,1,5
+62223,7,2,1,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10133.862484,12107.78442,3,90,77,77,NA,3,3,0,1,0,1,54,2,3,1,3
+62224,7,2,2,29,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,1,2,2,2,2,1,2,2,2,2,2,2,43986.779369,46449.619953,1,93,9,9,2.46,4,4,0,2,0,1,35,2,1,1,1
+62225,7,2,2,13,NA,1,1,1,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24481.187693,25761.611656,1,92,15,15,4.99,4,4,0,2,0,2,43,1,4,1,4
+62226,7,2,1,80,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,1,NA,2,2,2,2,2,2,2,2,1,NA,13654.270555,13892.295449,2,93,9,9,3.64,2,2,0,0,2,2,79,2,2,1,4
+62227,7,2,1,19,NA,3,3,2,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,68701.580401,72973.564721,2,94,15,15,5,5,5,0,2,0,1,53,1,5,1,5
+62228,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,144353.133634,144729.952645,1,91,14,14,3.15,5,5,0,1,0,2,50,1,5,1,5
+62229,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,72426.980535,75980.145152,1,98,6,6,1.31,3,3,1,0,0,1,30,1,5,1,5
+62230,7,2,1,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,12946.818038,13703.92145,1,101,5,3,1.07,2,1,0,0,2,2,70,1,4,2,NA
+62231,7,2,2,48,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,43535.993088,52686.708499,2,102,15,15,3.92,5,5,0,0,0,1,19,1,4,NA,NA
+62232,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,115926.402585,118970.086068,1,101,14,14,3.3,4,4,0,2,0,2,42,1,4,1,3
+62233,7,2,2,63,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,10999.00871,11900.968385,2,98,6,6,0.78,7,7,1,3,1,2,63,1,2,4,NA
+62234,7,2,1,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14860.312419,14555.816778,3,90,15,15,4.89,5,5,0,0,0,2,57,2,3,1,3
+62235,7,2,2,2,NA,3,3,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49862.013993,55032.245455,2,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+62236,7,2,1,61,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9048.959172,9513.611982,1,93,15,15,5,5,5,1,0,1,1,61,2,4,1,4
+62237,7,2,2,58,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140431.173819,139253.659476,1,91,7,7,2.64,2,2,0,0,1,2,58,1,5,1,4
+62238,7,2,2,0,4,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20786.668002,20803.881706,1,101,8,8,1.85,5,5,3,0,0,2,31,1,2,1,2
+62239,7,2,2,22,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,3,6,2,2,2,2,2,2,2,2,2,2,2,39426.061521,39254.343691,2,94,14,1,0.09,5,1,0,0,0,1,24,2,4,5,NA
+62240,7,2,2,14,NA,2,2,2,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26657.121865,28092.602686,1,97,14,14,3.25,4,4,0,2,0,1,45,1,3,6,NA
+62241,7,2,1,2,NA,2,2,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9237.934626,9897.780242,2,90,14,14,3.45,4,4,1,1,0,2,34,2,5,6,NA
+62242,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,129608.716834,132839.525748,1,100,6,6,1.98,2,2,0,0,0,1,50,1,5,4,NA
+62243,7,2,2,1,16,3,3,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49862.013993,51432.868823,2,91,15,15,5,4,4,2,0,0,1,34,1,5,1,5
+62244,7,2,1,1,17,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14505.670599,14525.271178,2,102,15,15,5,4,4,2,0,0,1,32,1,5,1,5
+62245,7,2,1,8,NA,5,6,2,8,97,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10810.913614,11522.32071,1,97,15,15,5,4,4,1,1,0,1,44,2,5,1,5
+62246,7,2,2,0,1,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7210.38482,7501.352858,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+62247,7,1,2,0,6,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5158.857524,0,1,97,7,7,1.74,4,4,2,0,0,1,34,1,5,1,5
+62248,7,2,1,65,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,9048.959172,9881.360452,1,93,8,8,1.2,7,7,1,1,1,1,24,2,2,5,NA
+62249,7,2,2,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,59001.303336,59528.815212,2,101,4,4,1.38,1,1,0,0,0,2,26,1,5,5,NA
+62250,7,2,1,34,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,51543.062078,52965.08671,3,92,15,15,5,3,3,1,0,0,1,34,1,5,1,5
+62251,7,2,2,51,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16672.82247,17074.238333,1,92,12,12,NA,4,4,0,0,0,1,59,2,3,1,4
+62252,7,2,2,11,NA,1,1,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17053.854294,17379.519997,3,92,15,8,2.62,4,3,1,1,0,1,30,1,2,6,NA
+62253,7,2,1,18,NA,3,3,2,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,68701.580401,72973.564721,2,94,14,14,5,3,3,0,0,0,1,42,1,5,1,5
+62254,7,2,1,14,NA,4,4,2,14,179,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15381.581315,15476.088016,1,93,12,12,NA,5,4,0,2,0,1,32,1,2,5,NA
+62255,7,2,1,65,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,29731.886612,29889.869541,1,95,3,3,0.87,2,2,0,0,2,2,65,1,2,1,3
+62256,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,7911.357817,9487.389486,2,94,5,5,1.63,2,2,0,0,2,2,79,1,3,1,1
+62257,7,2,2,2,NA,1,1,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,NA,NA,NA,NA,NA,NA,NA,9469.751474,10135.116584,1,102,NA,NA,NA,5,5,1,2,0,1,39,NA,NA,1,NA
+62258,7,2,1,47,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,43470.92932,42833.060444,1,92,6,6,0.93,5,5,0,2,0,1,47,2,1,1,1
+62259,7,2,1,61,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,7736.56115,7645.782314,1,99,4,2,0.87,2,1,0,0,1,2,59,1,3,3,NA
+62260,7,2,1,10,NA,3,3,2,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,57305.501166,60591.443404,1,94,15,15,5,5,5,0,3,0,1,46,1,3,1,5
+62261,7,2,1,47,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,2,2,2,2,33029.272844,33366.323953,2,93,14,14,3.25,4,4,0,2,0,2,46,2,5,1,4
+62262,7,2,2,0,10,2,2,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7367.430495,7616.752822,2,91,5,5,0.74,5,5,1,1,0,1,35,2,1,1,2
+62263,7,2,1,2,NA,3,3,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,28617.223132,33531.791823,1,101,3,3,0.61,4,4,1,2,0,1,38,1,2,4,NA
+62264,7,2,1,77,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,52448.388619,55291.886181,1,99,7,7,2.72,2,2,0,0,2,1,77,1,2,1,3
+62265,7,2,1,52,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19478.845078,19591.5023,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+62266,7,2,1,64,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,29999.543427,30837.648113,2,91,12,5,1.79,3,1,0,0,1,1,52,1,4,3,NA
+62267,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,34430.806911,35615.742482,2,95,6,6,1.08,4,4,1,1,0,1,39,1,4,1,4
+62268,7,2,2,15,NA,1,1,1,16,192,NA,NA,1,1,NA,9,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,16734.618372,17498.916137,2,96,5,5,0.78,5,5,0,2,0,1,37,2,1,5,NA
+62269,7,2,1,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,14718.123234,15404.517346,2,101,1,1,0.1,6,6,1,2,1,2,27,1,2,1,2
+62270,7,2,2,33,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,5,5,2,2,2,2,1,2,2,2,2,2,2,31765.061314,30907.973228,2,93,4,4,0.56,5,5,0,0,0,2,49,2,2,5,NA
+62271,7,2,1,7,NA,4,4,2,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6971.452972,8307.996458,2,99,7,7,1.63,4,4,0,2,0,1,53,1,3,3,NA
+62272,7,2,1,9,NA,2,2,1,9,111,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,11102.340512,11169.626274,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+62273,7,2,2,15,NA,5,6,1,15,184,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7588.544207,7881.983727,3,91,14,14,4.03,4,4,0,2,0,1,51,2,4,1,5
+62274,7,2,1,2,NA,4,4,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5103.735747,5455.59,2,95,3,3,0.38,5,5,2,2,0,2,37,1,4,3,NA
+62275,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,14442.406316,14518.77693,2,99,15,15,5,1,1,0,0,0,2,41,2,5,5,NA
+62276,7,2,1,9,NA,3,3,2,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,63059.776443,67841.511058,1,99,15,15,5,4,4,1,1,0,2,42,1,5,1,5
+62277,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,139343.551779,140618.30879,2,98,9,9,5,1,1,0,0,0,2,55,1,4,3,NA
+62278,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,94,7,7,2.51,2,2,0,0,2,2,72,1,4,1,1
+62279,7,2,1,80,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,9456.784526,9941.375417,2,99,13,13,NA,3,3,0,0,1,1,80,1,2,2,NA
+62280,7,2,2,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,16966.723528,16502.38463,2,95,6,6,1.7,2,2,0,0,0,2,54,1,4,2,NA
+62281,7,2,2,13,NA,4,4,2,13,162,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11372.489138,11738.962427,1,99,6,6,0.96,5,5,1,2,0,2,35,1,4,1,2
+62282,7,2,2,57,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,21219.116826,20638.40003,1,102,6,6,1.7,2,2,0,0,1,2,80,NA,NA,2,NA
+62283,7,2,1,2,NA,1,1,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14457.854197,15490.547336,1,92,10,10,2.93,4,4,1,0,0,2,55,1,4,1,4
+62284,7,2,2,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,117778.281347,120111.495123,1,94,8,8,2.41,3,3,0,0,3,1,63,1,4,1,5
+62285,7,2,2,17,NA,3,3,2,17,206,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,103007.696238,104941.393061,1,101,6,6,1.31,3,3,0,1,0,1,51,1,4,1,4
+62286,7,2,1,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,12556.207754,12615.022145,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+62287,7,2,2,73,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,12183.823561,12607.778632,2,100,6,6,2.11,2,2,0,0,2,1,79,1,3,1,4
+62288,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,22634.531479,23281.741513,1,103,9,6,2.24,3,1,0,0,0,1,27,1,5,5,NA
+62289,7,2,1,71,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,13555.672819,15969.746949,1,94,3,3,1.07,1,1,0,0,1,1,71,1,2,3,NA
+62290,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7618.827213,7957.141798,2,101,1,1,0,3,3,2,0,0,1,22,1,3,5,NA
+62291,7,2,1,56,NA,5,7,1,NA,NA,1,1,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20111.196953,20038.543847,1,92,14,14,5,2,2,0,0,0,1,56,2,4,1,4
+62292,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,36829.543424,39682.305625,1,93,15,15,5,4,3,0,0,3,1,80,1,5,2,NA
+62293,7,2,1,67,NA,3,3,2,NA,NA,1,1,2,1,9,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,28478.57859,28745.66908,1,101,3,3,1.16,1,1,0,0,1,1,67,2,4,3,NA
+62294,7,2,1,9,NA,2,2,2,9,117,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,9390.522479,10327.334743,2,90,3,3,0.54,4,4,1,2,0,2,33,2,1,4,NA
+62295,7,2,2,69,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,12994.252166,2,90,2,2,0.79,1,1,0,0,1,2,69,2,2,2,NA
+62296,7,2,1,19,NA,4,4,2,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13251.602554,13277.802057,1,96,15,15,5,2,2,0,0,0,2,51,1,5,5,NA
+62297,7,2,1,43,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18402.969014,18977.095614,2,95,10,10,3.67,3,3,0,1,0,1,43,1,4,1,4
+62298,7,2,1,15,NA,3,3,2,16,192,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,93665.036597,95017.313859,1,91,14,14,3.8,4,4,0,2,0,1,50,NA,NA,1,5
+62299,7,2,1,8,NA,5,6,2,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6631.058951,7488.793181,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+62300,7,2,2,6,NA,4,4,2,6,83,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6929.51414,7399.559927,2,99,14,14,4.09,3,3,0,2,0,2,37,1,5,5,NA
+62301,7,2,1,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16907.09145,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+62302,7,2,2,51,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,5,NA,2,2,2,1,2,2,2,2,2,2,19676.781212,20033.616894,2,103,77,77,NA,4,4,0,0,0,2,46,2,1,4,NA
+62303,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,61501.13451,65355.920739,2,101,6,6,1.62,3,3,0,0,2,1,80,1,3,1,3
+62304,7,2,2,6,NA,1,1,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9872.244853,10573.701398,2,90,6,6,0.96,5,5,1,1,0,1,39,2,2,1,NA
+62305,7,2,1,10,NA,1,1,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13285.093011,14158.005149,2,94,9,9,2.37,5,5,0,1,0,1,48,2,4,1,2
+62306,7,2,1,0,9,2,2,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4884.343512,4884.433539,3,90,12,12,NA,3,3,1,0,0,1,40,2,5,1,4
+62307,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,7828.117773,8460.915547,2,90,4,4,1.2,2,2,0,0,2,2,80,1,3,2,NA
+62308,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,83112.549416,87237.362828,1,97,7,7,2.64,2,2,0,0,2,1,79,1,4,1,3
+62309,7,2,2,47,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,2,NA,2,2,2,2,2,2,2,2,2,2,40337.933888,40548.292782,3,92,4,4,0.46,7,7,1,2,0,2,31,2,2,1,1
+62310,7,2,2,8,NA,4,4,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7632.404654,7765.546414,2,101,1,1,0.1,6,6,1,2,1,2,27,1,2,1,2
+62311,7,2,1,18,NA,3,3,2,19,228,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,68148.957861,67253.324127,1,93,15,15,3.92,5,5,0,1,0,2,54,1,5,1,5
+62312,7,2,2,63,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,1,1,2,2,1,2,2,2,16352.915834,17178.789759,3,92,8,8,1.85,5,5,1,0,2,1,66,2,1,1,1
+62313,7,2,1,7,NA,1,1,1,7,93,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12577.115885,12876.06354,2,96,5,5,0.76,5,5,1,2,0,1,44,2,1,1,3
+62314,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,7611.107768,7639.752345,2,97,3,3,1.33,1,1,0,0,1,1,61,1,3,5,NA
+62315,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,12842.559946,13803.13967,2,95,99,99,NA,2,2,0,1,1,2,80,1,2,2,NA
+62316,7,2,1,39,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,31674.692905,34832.561132,2,100,4,4,0.81,4,4,0,2,0,2,37,1,2,1,2
+62317,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18872.772727,18824.29246,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+62318,7,2,2,0,10,3,3,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8114.787453,8336.925581,1,91,2,2,0.27,5,5,2,2,0,2,42,1,4,3,NA
+62319,7,2,2,4,NA,4,4,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11231.392369,2,100,1,1,0.08,5,5,1,2,0,2,19,1,3,NA,NA
+62320,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,122483.259869,124909.680309,3,91,7,7,2.89,2,2,0,0,2,2,64,1,4,1,4
+62321,7,2,2,8,NA,5,6,1,8,100,NA,NA,2,1,3,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6412.057856,6847.117735,3,91,14,14,4.03,4,4,0,2,0,1,50,1,5,1,5
+62322,7,2,2,4,NA,4,4,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10295.166918,11223.20055,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+62323,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7618.827213,7957.141798,2,101,1,1,0,3,3,2,0,0,1,22,1,3,5,NA
+62324,7,2,2,60,NA,1,1,2,NA,NA,2,NA,2,2,77,NA,4,4,NA,2,2,2,1,2,2,1,2,2,2,11469.456138,12167.81965,1,90,3,3,0.23,7,7,3,1,1,2,35,2,2,5,NA
+62325,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,30932.175051,32888.281888,2,101,1,1,0.1,2,2,0,0,0,1,56,1,3,6,NA
+62326,7,2,1,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,153565.050575,157855.235487,1,97,10,10,4.3,2,2,0,0,1,2,56,1,5,1,5
+62327,7,2,2,19,NA,4,4,2,19,238,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12209.74498,12386.615954,2,90,6,6,1.12,4,4,0,1,1,1,63,2,1,1,1
+62328,7,2,2,9,NA,4,4,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7899.813226,8111.155436,2,97,15,15,5,4,4,0,2,0,1,47,NA,NA,6,NA
+62329,7,2,1,2,NA,2,2,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11793.948458,12167.27893,1,100,3,3,0.39,5,5,1,2,0,1,32,2,1,6,NA
+62330,7,2,2,33,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,21097.664283,22267.583876,2,96,14,14,5,2,2,0,0,0,2,33,2,5,1,5
+62331,7,2,1,9,NA,5,7,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8246.426933,8701.859504,1,99,14,14,3.94,4,4,1,1,0,1,43,1,4,1,5
+62332,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,31962.323978,33458.586112,1,100,4,4,1.06,3,2,0,0,0,1,22,1,4,6,NA
+62333,7,2,2,45,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,13046.228603,13115.21631,2,92,12,77,NA,7,2,0,0,2,1,53,2,3,1,3
+62334,7,2,1,9,NA,3,3,1,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19590.665143,20475.022602,3,91,5,5,0.87,4,4,0,2,0,2,38,1,2,3,NA
+62335,7,2,1,14,NA,2,2,2,14,171,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13921.972975,14228.387356,3,90,14,14,3.69,4,4,0,2,0,2,49,1,4,1,4
+62336,7,2,1,55,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,4,6,NA,2,2,2,2,2,2,2,2,2,2,24211.824535,24947.165851,2,93,10,10,3.67,3,3,0,0,0,2,56,2,4,6,NA
+62337,7,2,2,10,NA,1,1,2,10,123,NA,NA,2,2,3,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15166.167659,15537.272247,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+62338,7,1,2,2,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12260.86913,0,3,92,5,5,0.68,6,6,3,0,0,2,19,1,4,NA,NA
+62339,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,71402.235366,74200.659205,2,100,15,15,5,4,4,1,1,0,1,29,1,4,1,4
+62340,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,130826.463813,1,101,6,6,1.28,4,4,2,0,0,1,44,1,4,1,4
+62341,7,2,1,8,NA,1,1,2,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13285.093011,13365.60735,2,94,7,7,2.16,3,3,0,1,0,2,28,1,2,1,1
+62342,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,78565.115731,1,101,8,8,1.85,5,5,0,3,0,1,41,1,3,1,4
+62343,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17615.311737,2,97,5,5,0.76,5,5,0,0,0,2,50,1,4,5,NA
+62344,7,1,2,32,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,NA,NA,NA,NA,NA,NA,NA,26426.249254,0,1,99,NA,NA,NA,1,1,0,0,0,2,32,1,5,5,NA
+62345,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,64324.554311,66806.652381,2,103,15,15,5,4,4,2,0,0,1,36,2,4,1,5
+62346,7,2,1,19,NA,5,7,1,19,239,2,NA,2,2,4,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6558.308393,6849.983523,2,92,99,1,0.22,4,1,0,0,0,1,19,1,4,NA,NA
+62347,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,25812.913537,27430.822937,2,95,3,3,1.21,1,1,0,0,1,2,71,1,3,2,NA
+62348,7,2,2,19,NA,4,4,2,19,236,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11419.859653,11886.179371,2,99,7,7,1.19,6,6,1,3,0,2,38,1,3,5,NA
+62349,7,2,2,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,122473.120523,125156.324631,3,91,8,5,1.5,3,2,0,0,0,1,23,1,4,1,4
+62350,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,96255.674553,99423.043377,2,94,7,7,2.72,2,2,0,1,0,2,43,1,3,3,NA
+62351,7,2,1,7,NA,3,3,2,7,89,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25604.034863,26336.174292,1,95,5,5,0.92,5,5,1,2,0,2,30,1,4,1,4
+62352,7,2,2,3,NA,4,4,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,10601.396934,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+62353,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,39587.338944,43437.954199,2,91,15,15,5,2,2,0,0,2,2,79,1,5,1,5
+62354,7,2,2,11,NA,3,3,2,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,80369.555824,79030.006233,1,97,8,8,2.72,3,3,0,2,0,2,43,1,1,3,NA
+62355,7,2,2,16,NA,5,6,1,16,197,NA,NA,2,1,3,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7176.749123,7692.153155,3,91,7,7,1.79,4,4,0,1,0,2,45,2,2,1,3
+62356,7,2,2,14,NA,1,1,1,14,179,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26325.414456,27702.295836,3,92,12,12,NA,6,6,1,3,0,2,33,1,5,1,4
+62357,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,94433.586146,101591.315487,1,91,15,15,5,3,3,1,0,0,1,36,1,4,1,5
+62358,7,2,2,0,3,1,1,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9767.083234,9561.595244,3,92,7,7,1.99,3,3,1,0,0,1,40,1,4,1,4
+62359,7,2,2,54,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22224.73066,23825.197607,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+62360,7,2,1,17,NA,4,4,1,17,210,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11429.628358,11541.555232,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+62361,7,2,1,7,NA,4,4,1,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13423.881856,14179.490667,2,101,1,1,0.21,4,4,1,2,0,2,26,1,3,5,NA
+62362,7,2,2,11,NA,4,4,2,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.490652,8919.477637,2,97,6,6,1,6,6,1,2,2,2,60,1,2,2,NA
+62363,7,2,2,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,134897.594057,143557.343892,2,102,6,6,1.43,4,4,0,1,1,1,67,NA,NA,1,NA
+62364,7,2,2,4,NA,1,1,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14899.363418,16449.297045,1,100,8,8,2.17,4,4,1,1,0,2,40,2,2,1,2
+62365,7,2,2,37,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21097.664283,21336.137518,2,96,8,8,3.4,2,2,0,0,0,1,46,2,4,1,4
+62366,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,37318.801462,40047.525745,1,98,6,6,1.65,2,2,0,0,2,1,80,1,3,1,3
+62367,7,2,1,50,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,10133.862484,10097.253197,3,90,15,15,3.23,6,6,0,2,0,1,50,2,2,1,2
+62368,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,23176.790531,24820.797478,1,98,3,3,0.81,2,2,0,0,1,2,80,2,1,2,NA
+62369,7,2,2,8,NA,2,2,1,8,100,NA,NA,2,2,3,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15352.601806,15663.127806,2,102,7,7,1.53,5,5,1,2,0,2,37,2,4,1,4
+62370,7,2,1,21,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,13843.66558,14262.333267,1,90,15,15,3.7,5,5,0,0,0,1,54,NA,NA,1,NA
+62371,7,2,1,32,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22403.911395,23863.238542,2,96,14,14,5,2,2,0,0,0,2,33,2,5,1,5
+62372,7,2,2,55,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,1,1,2,2,1,2,2,NA,17991.883465,18643.986017,2,102,8,8,2.01,4,4,0,0,0,1,59,2,4,1,4
+62373,7,2,1,46,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,37324.655911,36776.974122,1,102,6,6,1.34,4,4,0,1,0,2,48,2,3,1,1
+62374,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,33530.36206,1,101,4,4,0.78,4,4,1,2,0,2,31,1,4,3,NA
+62375,7,2,1,24,NA,1,1,1,NA,NA,2,NA,2,2,99,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,52698.05363,59566.360508,3,92,4,4,0.65,4,4,2,0,0,2,20,1,3,5,NA
+62376,7,2,1,6,NA,4,4,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7730.47951,9212.541007,1,99,5,5,0.84,5,5,2,1,0,1,35,1,3,1,2
+62377,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35965.834545,38220.111719,1,95,3,3,1.24,1,1,0,0,1,2,78,1,2,2,NA
+62378,7,2,2,68,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,3,NA,2,2,2,1,2,2,2,2,2,2,10979.149658,11879.480817,1,96,15,8,4.66,2,1,0,0,2,2,68,2,1,3,NA
+62379,7,2,1,40,NA,5,7,1,NA,NA,2,NA,2,1,5,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,24837.95225,25237.159887,1,102,10,10,3.22,4,4,0,0,2,2,29,2,5,5,NA
+62380,7,2,1,16,NA,1,1,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,15506.325263,15592.022118,1,103,8,8,2,5,5,0,1,0,2,45,2,3,1,2
+62381,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,1,2,1,2,2,1,2,2,NA,15408.94893,15514.368855,1,99,7,2,0.74,3,1,0,0,2,1,70,1,2,1,4
+62382,7,2,2,1,21,4,4,2,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5743.559235,6261.298887,2,99,2,2,0.2,7,7,1,2,1,1,63,1,1,2,NA
+62383,7,2,1,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,91704.59836,93129.802737,1,93,15,4,1.38,6,1,0,0,0,1,23,1,5,5,NA
+62384,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,27444.308565,27682.940425,1,101,7,7,2.31,2,2,0,0,1,2,69,1,4,2,NA
+62385,7,2,1,0,10,4,4,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6699.703488,6998.255943,2,96,6,6,1.32,5,5,1,3,0,2,30,1,4,3,NA
+62386,7,2,1,35,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,21163.049914,22845.160537,1,97,8,8,2.51,3,3,0,1,0,1,35,1,3,1,4
+62387,7,2,1,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17597.975048,2,97,3,3,1.29,1,1,0,0,0,1,23,1,3,5,NA
+62388,7,2,2,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,49644.076348,54110.530387,1,97,3,3,1.12,1,1,0,0,1,2,69,1,3,2,NA
+62389,7,2,2,16,NA,5,7,2,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,15,15,5,6,6,0,3,0,1,47,1,5,1,5
+62390,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,14095.963218,14697.88566,2,92,15,8,4.59,2,1,0,0,0,2,25,2,5,5,NA
+62391,7,2,2,40,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,34954.173075,36222.975959,2,98,15,15,5,3,3,0,1,0,1,38,1,4,1,3
+62392,7,2,2,1,12,1,1,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,12260.86913,12783.275371,3,92,6,6,0.96,5,5,2,1,0,2,26,2,1,1,1
+62393,7,2,1,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,48709.670361,2,98,7,7,1.53,5,5,0,0,0,2,48,1,3,5,NA
+62394,7,2,1,1,14,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,1,1,0.13,4,4,2,0,0,2,52,1,2,4,NA
+62395,7,2,2,1,16,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12871.484115,13281.030392,2,102,8,8,2.24,4,4,1,1,0,1,35,2,3,1,1
+62396,7,2,1,10,NA,1,1,1,10,123,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,10658.399025,10827.062436,1,102,5,5,0.62,7,7,1,3,0,1,49,2,2,1,1
+62397,7,2,2,70,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,1,2,2,2,2,2,NA,17077.396628,18362.705174,2,96,13,13,NA,6,6,0,2,1,1,43,2,4,1,3
+62398,7,2,1,9,NA,3,3,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19926.440922,21167.339982,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+62399,7,2,2,53,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,24870.513993,25871.320138,2,98,5,5,1.63,2,2,0,0,0,2,53,2,1,1,1
+62400,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,27463.558155,27680.636989,2,100,4,4,1.22,2,2,0,1,0,2,31,1,4,5,NA
+62401,7,2,2,15,NA,3,3,1,15,187,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,93740.540203,99321.165816,1,100,8,8,1.95,4,4,0,2,1,2,49,1,5,6,NA
+62402,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,1,4,NA,1,2,2,1,2,2,1,2,2,NA,14259.601244,14208.087437,1,93,6,6,1.15,5,5,1,0,2,2,70,NA,NA,1,NA
+62403,7,2,2,3,NA,4,4,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8192.75936,8404.22838,2,99,77,77,NA,3,3,1,0,0,1,38,2,5,1,5
+62404,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,22009.438408,22692.209281,2,99,6,6,2.95,1,1,0,0,0,1,44,1,3,5,NA
+62405,7,2,1,5,NA,2,2,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14716.463544,14890.443704,2,96,2,2,0.27,6,6,1,3,0,1,34,NA,NA,1,NA
+62406,7,2,2,10,NA,2,2,2,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19936.606751,20692.800636,1,94,7,7,1.74,4,4,0,2,0,1,44,1,5,1,5
+62407,7,2,1,49,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,28680.660607,29290.218786,1,94,3,3,1.1,1,1,0,0,0,1,49,1,4,3,NA
+62408,7,1,1,58,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,174520.785302,0,1,95,7,7,2.86,2,2,0,0,1,1,58,1,4,1,3
+62409,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,22419.63376,27869.657009,1,94,3,3,1.16,1,1,0,0,1,2,80,1,3,2,NA
+62410,7,2,1,5,NA,4,4,2,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8158.702829,8662.344233,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+62411,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15790.702799,17527.166432,2,90,7,7,3.21,1,1,0,0,0,2,51,1,4,5,NA
+62412,7,2,1,16,NA,1,1,2,16,199,NA,NA,2,2,4,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,21718.29328,22050.264992,2,94,6,6,1.5,4,4,0,2,0,1,44,2,2,1,2
+62413,7,2,1,65,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,131445.986898,130104.237054,1,101,5,5,1.6,2,2,0,0,1,1,65,1,4,6,NA
+62414,7,2,1,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,1,2,1,2,2,1,1,2,NA,17206.320427,17681.811932,1,96,2,2,0.4,3,3,0,0,0,2,56,1,3,3,NA
+62415,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,129336.409693,136474.939567,1,95,15,15,5,3,3,1,0,0,1,26,1,3,1,4
+62416,7,2,2,19,NA,2,2,2,19,235,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13824.001771,14551.102227,2,90,10,10,2.91,4,4,0,1,0,2,51,2,4,1,1
+62417,7,2,2,26,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,45207.136555,46085.238422,1,94,14,8,4.66,2,1,0,0,0,2,26,1,4,5,NA
+62418,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,46965.818538,52174.805072,1,101,6,6,1.98,2,2,0,0,1,1,80,1,1,2,NA
+62419,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,42993.150248,46481.579201,1,95,5,5,0.65,6,6,1,0,2,1,80,1,3,1,4
+62420,7,1,1,24,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,59682.963348,0,2,102,77,77,NA,4,4,0,1,0,1,47,1,2,1,3
+62421,7,2,1,23,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,15171.949804,3,91,6,1,0,2,1,0,0,0,1,24,1,5,5,NA
+62422,7,2,2,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,63504.762752,64182.95787,1,91,15,15,5,1,1,0,0,1,2,72,1,4,2,NA
+62423,7,1,2,8,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13192.206605,0,2,102,14,14,3.8,4,4,0,2,0,1,47,1,4,1,4
+62424,7,2,1,20,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,2,5,NA,2,2,2,2,2,2,2,2,2,2,41713.173502,43905.698512,1,90,99,3,0.9,3,1,0,0,0,1,41,NA,NA,99,NA
+62425,7,2,2,30,NA,3,3,2,NA,NA,2,NA,2,2,1,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,81658.419251,86275.41601,2,99,15,15,5,2,2,0,0,0,1,30,NA,NA,1,5
+62426,7,2,1,9,NA,1,1,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8828.580268,8822.70874,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+62427,7,2,1,15,NA,4,4,2,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11462.850569,11753.30805,2,100,4,4,0.69,5,5,0,3,0,1,38,1,3,6,NA
+62428,7,2,2,12,NA,4,4,1,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16598.888685,16325.283055,2,102,15,15,4.2,5,5,1,2,0,2,29,NA,NA,1,NA
+62429,7,2,1,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,25964.952645,26121.35248,2,96,14,14,5,2,2,0,1,0,1,47,1,5,5,NA
+62430,7,2,1,27,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,49006.291777,50482.76018,1,92,9,9,2.93,3,3,0,1,0,2,30,1,5,1,5
+62431,7,2,1,54,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,25969.864445,29542.265764,1,97,2,2,0.81,1,1,0,0,0,1,54,1,2,5,NA
+62432,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,75294.690128,81171.633513,1,90,15,15,5,4,4,0,2,0,1,37,1,5,1,5
+62433,7,2,1,10,NA,4,4,2,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,8579.422451,10224.240431,1,99,NA,NA,NA,4,4,1,1,0,1,42,NA,NA,1,NA
+62434,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,9113.905743,10210.592308,3,90,14,14,5,2,2,0,0,2,1,60,NA,NA,1,3
+62435,7,2,2,0,7,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16547.193167,16560.896108,1,97,12,12,NA,5,5,3,0,0,2,33,1,5,1,5
+62436,7,2,1,4,NA,1,1,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11776.305841,11467.607256,1,90,4,4,0.47,7,7,1,1,0,2,50,2,1,1,1
+62437,7,2,1,39,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,6,NA,2,2,2,1,2,2,1,2,2,2,33948.23667,35734.088319,2,97,4,4,0.67,4,4,0,2,0,1,39,2,2,6,NA
+62438,7,2,1,0,0,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6339.587912,6622.092882,1,91,7,7,1.49,5,5,3,0,0,2,38,2,4,1,4
+62439,7,2,1,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60942.568495,64726.722108,1,94,15,15,5,2,2,0,0,2,1,75,1,4,1,3
+62440,7,2,1,22,NA,1,1,2,NA,NA,2,NA,2,1,4,NA,4,1,NA,1,2,2,2,2,2,1,2,2,2,35669.2076,36294.041819,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+62441,7,2,1,68,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,1,2,2,1,2,2,2,8609.250304,11228.904188,2,90,2,2,0.31,4,4,0,0,2,1,68,2,1,1,NA
+62442,7,2,1,0,8,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6981.458608,7088.61417,2,92,9,9,2.71,4,4,1,0,0,1,43,1,2,1,4
+62443,7,2,2,4,NA,4,4,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8464.796876,9108.215863,2,93,8,8,1.67,5,5,1,1,0,2,31,1,4,5,NA
+62444,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,21503.272394,20634.528185,2,100,1,1,0,2,2,0,0,0,2,55,1,5,3,NA
+62445,7,2,2,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,19866.025076,19772.340772,2,95,6,6,1.08,4,4,1,0,0,2,42,1,4,4,NA
+62446,7,2,2,4,NA,2,2,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,17065.756351,2,98,1,1,0,3,1,2,0,0,2,27,1,3,6,NA
+62447,7,2,2,19,NA,1,1,1,19,237,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17490.464019,17826.708822,1,102,6,6,1.43,3,3,0,1,0,2,39,1,4,3,NA
+62448,7,2,1,13,NA,1,1,1,13,162,NA,NA,2,2,4,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18583.729819,18727.0393,2,96,6,6,1.11,5,5,0,3,0,2,32,2,3,1,2
+62449,7,2,2,38,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,32097.38481,34606.180038,2,100,14,14,3.36,4,4,1,1,0,1,45,2,5,1,2
+62450,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19282.792088,20545.343351,2,97,7,7,1.92,3,3,0,1,0,1,57,1,4,1,4
+62451,7,2,2,0,4,3,3,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,NA,2,1,2,2,NA,NA,NA,NA,9142.358181,9688.944551,2,98,1,1,0.23,2,2,1,0,0,2,20,1,3,6,NA
+62452,7,2,1,39,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,5,NA,2,2,2,2,2,2,2,2,2,2,34564.442715,35621.093093,2,90,4,4,0.81,3,3,0,0,0,1,39,2,3,5,NA
+62453,7,2,2,2,NA,4,4,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7329.363701,7518.547014,1,103,6,6,1.57,3,3,1,0,0,2,25,1,4,5,NA
+62454,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,7278.790659,7582.574348,2,93,5,5,1.32,2,2,0,0,1,1,49,2,2,1,4
+62455,7,2,1,18,NA,4,4,2,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7946.436474,8100.311671,3,90,15,15,5,5,5,1,0,1,1,38,2,3,1,4
+62456,7,2,1,17,NA,4,4,1,17,214,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13731.625553,14242.275028,1,100,14,14,3.93,3,3,0,1,0,2,47,1,5,4,NA
+62457,7,2,1,69,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,7736.56115,7645.782314,1,99,15,15,5,2,2,0,0,2,2,63,2,5,1,5
+62458,7,2,1,13,NA,5,6,2,13,162,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,77,77,NA,3,3,0,1,0,1,54,2,3,1,3
+62459,7,2,1,4,NA,3,3,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,76935.850725,86801.923297,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+62460,7,2,1,41,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,109679.354704,1,98,15,15,4.34,4,4,1,1,0,1,41,1,5,1,5
+62461,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,19362.260448,24981.28051,1,96,6,6,1.98,2,2,0,1,0,2,29,1,3,5,NA
+62462,7,2,2,11,NA,3,3,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,79732.314338,79120.693768,1,100,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+62463,7,2,1,7,NA,1,1,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,19774.151841,21259.203461,2,94,5,5,0.65,6,6,0,3,0,1,44,2,1,1,1
+62464,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,22824.336433,1,95,10,7,3.21,5,1,1,2,0,1,32,1,3,6,NA
+62465,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20084.755052,20430.629009,1,99,14,14,4.21,4,4,0,2,0,2,44,1,5,1,5
+62466,7,2,1,4,NA,4,4,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7629.74403,8207.121237,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+62467,7,2,1,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,117075.881463,118577.582598,1,94,5,5,2.15,1,1,0,0,1,1,65,1,5,1,NA
+62468,7,2,2,0,8,2,2,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6787.112205,6644.319314,2,94,5,5,0.89,4,4,2,0,0,2,35,2,4,1,2
+62469,7,2,1,9,NA,4,4,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,13521.226684,2,101,4,4,0.76,4,4,1,1,0,1,28,1,2,1,4
+62470,7,2,1,4,NA,1,1,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12870.245769,12532.871153,1,102,6,6,0.8,7,7,3,3,0,2,34,2,3,1,1
+62471,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,45482.078305,46877.423634,2,96,99,99,NA,4,4,0,1,1,2,51,1,2,1,4
+62472,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,27988.858556,28110.553782,1,97,6,6,2.69,2,1,0,0,0,1,31,1,5,6,NA
+62473,7,2,2,17,NA,4,4,1,17,209,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,6,6,1.9,2,2,0,1,0,2,42,1,5,5,NA
+62474,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,31347.021791,31681.790338,1,98,6,3,1.13,2,1,0,0,2,1,69,1,4,6,NA
+62475,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,36750.682715,37321.834378,2,97,6,1,0.05,2,1,0,0,0,1,22,1,3,6,NA
+62476,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,28680.660607,28606.985896,3,91,6,3,1.29,2,1,0,0,0,1,40,1,3,5,NA
+62477,7,2,1,11,NA,3,3,1,11,139,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21839.599095,22875.076583,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+62478,7,2,1,5,NA,4,4,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8275.349856,8376.287205,2,95,1,1,0.03,2,2,1,0,0,1,24,1,3,5,NA
+62479,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,11608.998717,11632.703325,3,90,10,10,2.41,5,5,1,2,0,1,44,2,4,1,5
+62480,7,2,1,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,110055.244549,1,98,14,14,4.12,4,4,0,2,0,2,36,1,5,1,3
+62481,7,2,2,39,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,4,2,2,2,2,1,2,2,1,2,2,1,27127.983961,27837.333201,2,90,10,10,3.13,4,4,1,2,0,2,39,1,5,4,NA
+62482,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,31552.004994,1,95,10,6,1.34,5,4,1,2,0,1,32,1,3,6,NA
+62483,7,2,1,11,NA,3,3,2,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25604.034863,26336.174292,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+62484,7,2,2,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,27585.470618,26830.521849,2,102,6,6,1.22,5,5,0,2,0,2,42,1,4,1,4
+62485,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,99,1,0.28,2,1,0,0,0,1,21,1,4,5,NA
+62486,7,1,2,14,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,117872.104347,0,2,101,8,8,1.72,5,5,0,3,0,1,37,1,3,1,3
+62487,7,2,1,65,NA,2,2,2,NA,NA,1,1,2,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,10655.434864,11610.446696,1,90,14,14,5,2,2,0,0,2,1,65,2,2,1,3
+62488,7,1,1,9,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,41165.805324,0,2,100,15,15,5,4,4,1,1,0,1,29,1,4,1,4
+62489,7,2,2,1,22,4,4,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7281.670423,7938.059494,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+62490,7,2,1,0,6,3,3,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15379.505002,15960.887694,1,102,14,14,3,6,6,1,2,0,1,44,1,4,1,5
+62491,7,2,2,61,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,2,1,2,1,3,18695.172864,19350.637044,2,102,15,15,5,5,5,1,0,2,1,30,1,4,1,5
+62492,7,2,1,6,NA,5,7,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8227.856305,8510.505183,3,91,10,10,2.48,5,5,2,1,0,2,27,1,2,1,4
+62493,7,2,2,0,7,5,7,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6570.947402,7236.391986,2,102,7,7,2.65,2,2,1,0,0,2,42,1,4,5,NA
+62494,7,2,1,28,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,4,1,NA,1,2,2,1,2,2,2,2,2,2,37970.860743,39407.668593,2,92,15,7,3.67,2,1,0,0,0,1,28,2,4,1,NA
+62495,7,2,1,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,86361.036512,88802.863313,2,93,10,10,3.61,3,3,0,0,2,1,75,1,4,1,4
+62496,7,2,1,4,NA,1,1,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,18754.85406,18780.196288,2,102,5,5,0.89,4,4,2,0,0,1,33,2,9,1,2
+62497,7,2,1,12,NA,3,3,2,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61479.689958,61628.148021,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+62498,7,2,2,13,NA,1,1,1,13,159,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20347.899985,21072.708732,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+62499,7,2,2,1,14,5,7,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9679.976914,10817.322533,1,95,4,4,0.97,3,3,2,0,0,2,22,1,4,5,NA
+62500,7,2,2,52,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,1,96,12,12,NA,2,2,0,0,0,1,46,1,4,1,5
+62501,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,78018.093799,1,101,14,14,3.15,5,5,2,1,0,1,35,1,4,1,5
+62502,7,2,1,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,14739.896059,14808.93904,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+62503,7,2,1,5,NA,3,3,2,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22669.354731,25576.419472,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+62504,7,2,2,30,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,1,6,2,2,2,2,2,2,2,2,2,1,2,38161.026403,37131.361742,1,100,3,3,0.39,5,5,1,2,0,1,32,2,1,6,NA
+62505,7,2,2,3,NA,1,1,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13196.707564,13241.769839,2,96,77,77,NA,7,7,3,2,0,2,33,2,2,6,NA
+62506,7,2,2,46,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,29105.053716,30161.539189,2,93,4,4,0.84,3,3,0,1,0,2,46,2,3,1,2
+62507,7,2,2,17,NA,1,1,1,17,207,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18581.167701,19284.858627,2,96,3,3,0.24,7,7,2,3,1,2,40,1,3,3,NA
+62508,7,2,1,3,NA,3,3,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58332.578536,68350.303275,1,91,15,15,5,3,3,1,0,0,1,36,1,4,1,5
+62509,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,30131.691064,30623.993423,2,95,6,6,1.36,3,3,0,0,2,2,60,1,5,1,4
+62510,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,62763.524469,78020.806152,1,92,7,7,2.64,2,2,0,0,2,1,80,1,3,1,3
+62511,7,2,1,9,NA,3,3,2,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25019.74954,25881.463934,1,101,3,3,0.44,5,5,0,3,0,1,35,1,3,1,4
+62512,7,2,1,39,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17605.619977,18470.699991,3,91,15,15,5,4,4,1,1,0,1,39,2,5,1,5
+62513,7,2,2,46,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,29470.67209,35523.062266,1,90,5,5,1,4,4,0,2,0,1,40,2,2,1,1
+62514,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19633.637051,20188.770976,2,95,6,6,1.19,4,4,0,1,0,1,44,1,3,1,2
+62515,7,2,2,42,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,5,2,2,2,2,2,2,2,2,2,2,2,31235.666551,33485.036978,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+62516,7,2,1,10,NA,2,2,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18107.947773,18538.358733,1,97,14,14,3.25,4,4,0,2,0,1,45,1,3,6,NA
+62517,7,2,1,80,NA,3,3,2,NA,NA,2,NA,2,1,8,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,11483.380372,15883.29652,1,90,77,77,NA,2,2,0,0,2,1,80,2,1,1,1
+62518,7,1,1,47,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16651.742047,0,1,100,15,15,5,3,3,0,0,0,1,47,2,5,1,5
+62519,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,17801.655316,19397.911862,2,95,7,2,0.72,3,1,0,0,0,2,56,1,3,2,NA
+62520,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11017.136221,10826.673882,1,91,3,3,0.89,2,2,1,0,0,2,23,1,3,3,NA
+62521,7,2,2,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,178239.535384,176744.998212,1,102,14,14,5,2,2,0,0,1,2,59,1,5,3,NA
+62522,7,2,1,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,24930.322327,25494.618871,2,94,9,9,5,1,1,0,0,0,1,54,1,4,3,NA
+62523,7,2,1,33,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23872.904125,23622.849234,1,91,6,6,0.99,5,5,3,0,0,2,33,2,3,1,4
+62524,7,2,1,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,91704.59836,93129.802737,1,93,15,7,3.67,6,1,0,0,0,1,23,1,5,5,NA
+62525,7,2,1,15,NA,5,6,2,15,189,NA,NA,2,2,3,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10725.405063,11099.246363,2,94,3,3,0.68,3,2,0,1,0,2,45,2,4,3,NA
+62526,7,2,2,62,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,8811.978874,9348.531299,3,90,12,12,NA,3,3,0,0,1,2,62,2,3,1,5
+62527,7,2,1,79,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,NA,9541.516696,10104.465045,1,100,77,77,NA,1,1,0,0,1,1,79,1,2,5,NA
+62528,7,2,1,40,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,6,NA,2,2,2,2,2,2,2,2,1,NA,25035.846455,24668.484001,2,99,99,99,NA,5,3,0,1,0,1,40,2,1,6,NA
+62529,7,2,1,59,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15760.921402,16568.510632,3,90,12,12,NA,3,3,0,0,0,1,59,2,4,1,4
+62530,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,141316.739364,156265.607046,2,94,5,5,0.89,4,4,0,2,0,2,51,1,2,3,NA
+62531,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,20149.292081,21423.504707,2,99,2,2,0.4,2,2,0,0,0,2,52,1,3,1,3
+62532,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,77,NA,1,2,2,1,2,2,1,2,2,1,11523.037911,11989.050618,2,97,13,13,NA,1,1,0,0,1,2,64,1,3,77,NA
+62533,7,2,2,1,17,1,1,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10089.038145,2,94,8,8,2.7,3,3,1,0,0,2,23,2,4,1,3
+62534,7,2,1,13,NA,5,7,1,13,159,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11068.745625,11332.552011,1,103,3,3,0.37,5,5,1,2,0,2,30,1,4,5,NA
+62535,7,2,2,5,NA,3,3,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,69665.012606,74034.381659,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+62536,7,2,1,46,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,14238.597316,14347.374047,2,92,15,9,5,2,1,0,0,0,1,40,NA,NA,77,NA
+62537,7,2,2,10,NA,3,3,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15442.305642,15459.941839,2,94,7,7,1.62,5,5,0,3,0,1,30,1,2,1,9
+62538,7,2,2,1,16,4,4,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6802.424856,6978.006983,2,99,15,15,5,4,4,2,0,0,1,34,1,5,1,5
+62539,7,2,1,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11218.730451,2,101,2,2,0.92,1,1,0,0,1,1,60,1,2,2,NA
+62540,7,2,2,0,3,3,3,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22478.08837,21867.533571,3,91,10,10,3,4,4,2,0,0,1,32,1,4,1,5
+62541,7,2,2,13,NA,3,3,1,13,163,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,68399.970058,69290.143431,1,102,14,14,4.05,3,3,0,2,0,2,34,1,4,3,NA
+62542,7,2,2,11,NA,2,2,1,11,137,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14414.529053,14932.182215,2,96,2,2,0.27,6,6,1,3,0,1,34,NA,NA,1,NA
+62543,7,2,2,1,12,2,2,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11981.824297,12142.965635,2,91,2,2,0.26,4,4,1,1,0,2,42,2,4,3,NA
+62544,7,2,1,42,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19320.837782,1,99,14,14,4.86,3,3,0,1,0,1,42,1,5,1,5
+62545,7,2,1,8,NA,2,2,2,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15673.230419,15680.173615,1,97,15,15,4.52,6,6,0,4,0,2,41,1,5,1,5
+62546,7,2,1,64,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,13473.930124,13736.530204,1,92,6,6,2.69,1,1,0,0,1,1,64,1,4,2,NA
+62547,7,2,1,9,NA,1,1,2,9,113,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13484.595524,13562.00827,1,97,3,3,0.5,5,5,0,2,0,1,56,2,2,6,NA
+62548,7,2,1,57,NA,2,2,1,NA,NA,1,2,2,1,8,NA,3,5,NA,1,2,2,1,2,2,2,2,2,1,25422.415762,25922.20279,2,100,7,7,2.64,2,2,0,0,1,1,57,2,3,5,NA
+62549,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,10579.097347,10866.833801,2,92,8,8,2.01,4,4,0,0,0,1,53,2,3,1,3
+62550,7,2,2,9,NA,5,7,1,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8292.876947,8466.609309,1,102,15,15,5,4,4,0,2,0,2,40,2,5,1,4
+62551,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,86578.861495,89840.962333,2,101,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+62552,7,2,2,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,139800.409559,140918.856987,1,100,15,15,4.07,5,5,0,2,0,2,41,1,5,1,4
+62553,7,2,1,8,NA,5,6,2,8,100,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,4945.578914,5359.127104,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+62554,7,2,2,15,NA,5,6,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12546.119668,12876.82216,2,102,14,14,4.86,3,3,0,1,0,1,55,NA,NA,1,5
+62555,7,2,1,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,23471.353577,23548.945779,1,100,7,7,1.83,3,3,0,1,0,2,40,1,4,6,NA
+62556,7,2,2,31,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,17978.142628,18308.053591,1,91,10,10,3.78,3,3,1,0,0,1,35,2,5,1,5
+62557,7,2,1,10,NA,1,1,2,10,122,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,10591.186197,11647.778431,1,90,4,4,0.46,7,7,2,3,0,2,34,2,1,6,NA
+62558,7,2,2,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,20084.755052,20500.648257,1,99,14,14,3.67,4,4,1,0,0,2,49,1,3,1,3
+62559,7,2,2,9,NA,1,1,2,9,109,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,2,2,2,1,18668.602894,22428.786321,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+62560,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12435.659176,12095.32474,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+62561,7,2,2,10,NA,3,3,1,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,50248.753555,51858.292831,2,100,8,8,2.91,3,3,0,2,0,2,48,1,5,1,NA
+62562,7,2,1,4,NA,1,1,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17865.135763,18076.339981,3,92,1,1,0,5,5,3,0,0,1,26,1,2,1,2
+62563,7,2,2,72,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,7,77,NA,1,2,2,1,2,2,1,2,2,NA,13638.730054,13839.452317,1,99,12,12,NA,1,1,0,0,1,2,72,1,7,77,NA
+62564,7,2,2,13,NA,4,4,2,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,11743.959438,2,95,1,1,0.17,2,2,0,1,0,2,49,1,3,99,NA
+62565,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15533.829525,2,101,7,5,1.84,2,1,0,0,0,1,27,2,5,5,NA
+62566,7,2,2,2,NA,4,4,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7734.994032,8078.467013,2,95,1,1,0.18,4,4,2,1,0,2,38,1,2,5,NA
+62567,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,22991.276372,25575.167764,1,99,99,99,NA,1,1,0,0,1,2,80,1,1,3,NA
+62568,7,2,2,1,13,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10867.770426,11631.363355,1,100,4,4,0.56,5,5,2,0,0,2,25,2,2,1,2
+62569,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,45973.010529,51521.910389,1,94,8,8,3.47,2,2,0,0,1,2,80,1,1,2,NA
+62570,7,2,1,6,NA,4,4,2,6,72,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8655.162127,8675.53524,2,101,2,2,0.22,4,4,1,1,0,2,41,1,2,4,NA
+62571,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,85420.170155,86011.35365,1,91,14,14,3.06,5,5,2,0,0,2,30,1,5,1,5
+62572,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,47566.45715,54102.989916,1,90,7,7,3.22,1,1,0,0,1,2,80,1,5,3,NA
+62573,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,30442.30641,30472.199404,1,95,6,3,1.1,2,1,0,0,0,2,48,1,4,3,NA
+62574,7,2,2,13,NA,3,3,2,13,166,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,132630.209478,136814.3405,1,97,9,9,2.88,3,3,0,1,0,2,33,1,3,5,NA
+62575,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,129336.409693,130993.13082,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+62576,7,2,1,66,NA,1,1,1,NA,NA,2,NA,2,1,77,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,14488.953694,14771.336069,3,92,9,9,2.22,5,5,1,0,2,1,66,2,1,1,1
+62577,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,72551.269339,73326.076054,1,92,6,6,2.04,2,2,0,0,2,2,71,1,4,1,1
+62578,7,2,2,6,NA,3,3,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,101,7,7,1.55,5,5,1,2,0,2,31,1,4,1,2
+62579,7,2,1,78,NA,1,1,1,NA,NA,1,2,2,1,9,NA,4,1,NA,2,2,2,1,2,2,1,2,2,NA,12782.642018,13005.47245,2,96,5,5,1.08,3,3,0,0,2,1,53,2,4,3,NA
+62580,7,2,2,14,NA,3,3,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38400.791741,38900.548719,1,92,4,4,0.5,6,6,0,3,0,2,41,1,4,1,NA
+62581,7,2,2,10,NA,3,3,1,10,128,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,53370.126656,53491.559571,2,98,15,15,5,3,3,0,1,0,1,48,1,4,1,5
+62582,7,1,2,10,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23170.920553,0,1,94,4,4,1.26,2,2,0,1,0,1,40,1,4,3,NA
+62583,7,2,2,17,NA,3,3,1,17,212,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,93740.540203,99321.165816,1,100,8,8,1.95,4,4,0,2,1,2,49,1,5,6,NA
+62584,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,29738.952706,31253.462984,2,94,8,8,3.4,2,2,0,0,0,1,22,1,3,5,NA
+62585,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,14859.685983,15314.825218,1,95,3,3,0.98,1,1,0,0,1,2,80,1,3,2,NA
+62586,7,2,1,70,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,68266.732554,71967.824097,1,95,5,5,2.02,1,1,0,0,1,1,70,1,3,5,NA
+62587,7,2,1,73,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,NA,15565.766387,16476.020536,1,101,3,3,1.07,2,1,0,0,1,1,73,1,4,6,NA
+62588,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,137038.746155,146293.19093,2,101,3,1,0.18,2,1,0,0,0,1,21,1,4,5,NA
+62589,7,2,2,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,61388.874846,64435.558603,1,90,7,7,1.89,3,3,0,0,1,2,75,1,4,3,NA
+62590,7,2,1,11,NA,4,4,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8483.005475,8584.70346,1,96,7,7,1.69,4,4,0,1,0,2,19,2,4,NA,NA
+62591,7,2,1,10,NA,3,3,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,54897.892683,57357.850008,2,98,14,14,3.58,4,4,0,2,0,1,36,1,3,1,4
+62592,7,2,1,0,5,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9665.908305,9498.805759,1,95,5,5,1.08,3,3,1,0,0,2,22,1,3,6,NA
+62593,7,2,2,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,64463.340883,65151.773075,1,91,7,7,2.92,2,2,0,0,1,2,47,1,5,5,NA
+62594,7,2,1,0,10,4,4,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5707.173345,5961.496632,1,90,4,4,0.67,5,5,3,0,0,2,32,2,3,3,NA
+62595,7,2,2,28,NA,5,7,1,NA,NA,2,NA,2,1,5,NA,4,1,3,1,2,2,1,2,2,1,2,2,1,13851.686232,14954.517784,3,91,14,14,4.32,3,3,1,0,0,1,31,2,3,1,4
+62596,7,2,2,52,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11428.6913,11489.125563,1,103,5,5,0.65,6,6,0,0,1,2,26,2,4,5,NA
+62597,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,12170.020379,2,97,6,6,1,6,6,1,2,2,2,60,1,2,2,NA
+62598,7,2,1,18,NA,4,4,1,18,223,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11666.009872,12200.765691,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+62599,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17521.481386,17644.217766,2,97,5,5,0.76,5,5,0,0,0,2,50,1,4,5,NA
+62600,7,2,1,10,NA,2,2,1,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10738.959181,10632.988891,2,93,14,14,3.25,4,4,0,2,0,2,46,2,5,1,4
+62601,7,2,1,65,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,9636.209947,9711.587411,2,97,14,14,5,2,2,0,0,1,2,52,1,5,1,3
+62602,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,60071.993203,65791.533402,1,95,8,8,4.59,1,1,0,0,1,2,80,1,4,2,NA
+62603,7,2,2,38,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,39561.667842,42653.886229,2,98,6,6,1.25,4,4,1,0,1,1,46,1,2,6,NA
+62604,7,2,2,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,69886.968852,80170.534309,1,102,4,4,1.42,1,1,0,0,1,2,73,1,3,5,NA
+62605,7,1,1,1,22,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7025.964374,0,2,91,4,4,0.65,5,5,1,3,0,1,43,2,3,5,NA
+62606,7,2,1,2,NA,4,4,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5930.749873,6003.089312,2,93,6,6,1.98,2,2,1,0,0,2,51,1,2,3,NA
+62607,7,2,1,4,NA,4,4,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,11070.778245,1,100,1,1,0.04,4,4,1,1,0,2,51,1,3,3,NA
+62608,7,1,2,67,NA,2,2,NA,NA,NA,2,NA,2,1,NA,NA,2,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,7278.790659,0,2,93,NA,NA,NA,2,2,0,0,2,1,68,NA,NA,1,2
+62609,7,2,1,9,NA,3,3,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,NA,2,1,2,2,1,2,2,1,25604.034863,26336.174292,1,95,5,5,0.92,5,5,1,2,0,2,30,1,4,1,4
+62610,7,2,2,12,NA,2,2,1,12,146,NA,NA,2,2,1,4,NA,NA,NA,2,1,2,2,2,2,2,2,1,2,18006.855255,21837.57166,1,100,3,3,0.39,5,5,1,2,0,1,32,2,1,6,NA
+62611,7,2,1,45,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,11725.7448,11815.324423,2,95,3,3,1.16,1,1,0,0,0,1,45,2,2,4,NA
+62612,7,2,2,38,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21619.283038,22049.338941,2,102,9,9,2.68,4,4,1,1,0,2,38,2,5,1,2
+62613,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,28394.357741,26997.683446,1,100,3,3,0.66,2,2,0,1,0,2,29,1,4,1,NA
+62614,7,2,1,12,NA,3,3,2,12,154,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58479.782556,60361.370686,1,99,15,15,4.47,4,4,0,2,0,2,43,1,5,1,5
+62615,7,2,2,0,11,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3774.069255,3925.050204,2,99,5,5,0.78,5,5,2,2,0,2,30,1,3,5,NA
+62616,7,2,2,37,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,1,1,2,1,2,1,1,2,2,1,2,1,NA,15542.93857,16609.63545,3,91,4,4,0.69,5,5,0,2,0,1,45,2,4,1,1
+62617,7,2,1,8,NA,5,7,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8439.403196,8726.876152,2,103,6,6,1.11,5,5,1,2,0,2,36,1,4,5,NA
+62618,7,2,1,10,NA,5,7,2,10,129,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7923.925927,8546.646915,1,91,15,15,5,5,5,0,3,0,1,40,1,5,1,5
+62619,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,24023.719783,25884.561742,2,96,7,7,3.21,2,1,0,0,1,1,80,1,5,5,NA
+62620,7,2,2,7,NA,2,2,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18660.103977,19985.967776,2,91,6,6,0.93,5,5,1,2,0,1,34,1,2,1,3
+62621,7,2,1,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,23242.990557,23753.836506,1,100,8,8,1.95,4,4,0,2,0,2,42,1,4,1,4
+62622,7,2,2,15,NA,5,6,2,16,192,NA,NA,2,1,4,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10963.340882,11440.408955,2,91,15,15,5,4,4,0,2,1,2,56,1,5,1,5
+62623,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,39653.38014,40869.906755,2,95,6,6,1.57,3,3,0,0,2,1,80,1,2,1,4
+62624,7,2,2,40,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,18255.735511,19354.37213,2,91,14,14,3.47,4,4,1,1,0,2,40,2,5,1,NA
+62625,7,2,2,60,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15207.312407,15896.113669,1,92,15,15,5,3,3,0,0,1,1,57,1,3,1,4
+62626,7,2,1,13,NA,1,1,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,17314.330266,19031.527568,1,90,4,4,0.46,7,7,2,3,0,2,34,2,1,6,NA
+62627,7,2,2,2,NA,1,1,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10379.318342,2,94,9,9,2.29,5,5,2,1,0,2,33,2,3,1,1
+62628,7,2,2,1,19,3,3,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16717.071084,17765.560854,2,94,8,8,1.6,6,6,3,1,0,2,32,1,4,1,4
+62629,7,2,1,11,NA,5,6,2,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6262.834446,6755.01452,3,90,12,12,NA,5,5,1,2,0,1,37,2,5,1,5
+62630,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,149153.684635,157450.342804,1,101,3,3,1.25,1,1,0,0,1,2,68,1,2,2,NA
+62631,7,2,2,37,NA,3,3,2,NA,NA,2,NA,2,1,4,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,79987.938327,1,101,6,6,1.28,4,4,2,0,0,1,44,1,4,1,4
+62632,7,2,1,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,14424.961621,14817.68753,2,102,5,5,1.08,3,3,0,1,0,2,46,2,1,5,NA
+62633,7,2,2,61,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12176.672371,12646.648452,2,92,15,15,5,2,2,0,0,2,1,61,1,5,1,5
+62634,7,2,1,68,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,114174.703171,113009.251918,2,97,8,8,3.06,2,2,0,0,2,1,68,1,2,1,4
+62635,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,1,2,1,2,1,1,2,2,NA,NA,NA,NA,13851.686232,14678.247845,3,91,5,5,1.08,3,3,1,0,0,1,29,2,5,1,5
+62636,7,2,1,17,NA,4,4,1,17,215,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12147.046136,12703.852077,2,100,4,4,0.86,3,3,0,2,0,2,36,1,3,5,NA
+62637,7,2,2,3,NA,4,4,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13130.790087,13857.266639,2,102,15,15,4.2,5,5,1,2,0,2,29,NA,NA,1,NA
+62638,7,2,2,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,13050.526646,2,101,13,3,0.64,5,4,0,3,1,2,62,1,1,2,NA
+62639,7,2,2,26,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,6,2,2,2,2,1,2,2,2,2,2,2,43708.837271,43518.466069,1,96,9,9,3.97,2,2,0,0,0,1,28,2,2,6,NA
+62640,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,26546.087356,29693.438127,1,102,6,6,1.12,4,4,2,0,0,2,31,1,3,6,NA
+62641,7,2,1,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10276.786905,10602.092524,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+62642,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23607.684366,1,95,1,1,0.11,1,1,0,0,0,1,30,1,3,5,NA
+62643,7,2,2,2,24,4,4,2,2,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5687.793894,6161.690011,2,90,99,99,NA,5,5,1,0,0,1,20,1,3,5,NA
+62644,7,2,1,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,11761.359913,11893.852645,1,95,5,5,1.79,1,1,0,0,1,1,61,1,4,2,NA
+62645,7,2,2,12,NA,5,6,1,12,153,NA,NA,2,2,2,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6391.016092,6494.382638,1,100,99,99,NA,6,6,0,1,0,1,53,2,2,1,3
+62646,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,17164.211773,16694.46809,2,99,8,8,3.57,2,2,0,0,1,1,67,1,2,1,2
+62647,7,2,1,0,0,2,2,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8143.831412,8143.981517,2,102,5,5,0.89,4,4,2,0,0,1,45,1,3,1,4
+62648,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,2,NA,1,2,1,1,2,2,1,2,1,NA,19516.411172,20124.53654,2,102,10,10,4.76,2,2,0,0,1,2,49,2,5,5,NA
+62649,7,2,1,76,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,9334.084938,9516.918553,2,96,8,8,2.59,3,3,0,0,2,1,76,1,3,1,4
+62650,7,2,1,71,NA,4,4,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,7110.382011,7529.893703,2,93,6,6,1.85,2,2,0,0,2,1,71,2,1,1,2
+62651,7,2,1,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,152467.08796,155918.172532,1,94,6,6,2.69,1,1,0,0,0,1,55,1,3,3,NA
+62652,7,2,1,46,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,34205.013302,40575.476948,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+62653,7,1,1,49,NA,5,6,NA,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,13323.689579,0,2,92,6,6,1.3,4,4,0,1,0,2,48,2,3,1,3
+62654,7,1,1,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30553.632981,0,1,98,5,2,0.42,4,3,2,0,0,1,24,1,3,6,NA
+62655,7,2,1,1,23,4,4,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5930.749873,6003.089312,2,93,9,9,2.6,4,4,1,1,0,1,37,1,4,1,4
+62656,7,2,2,0,9,1,1,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6787.112205,6969.467907,2,94,4,4,0.73,5,5,2,1,0,1,35,2,1,6,NA
+62657,7,2,2,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11401.934012,11887.743086,1,95,13,13,NA,5,5,1,2,0,2,34,2,1,1,1
+62658,7,2,1,68,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,28559.076421,28710.827523,1,92,6,6,2.3,1,1,0,0,1,1,68,1,3,3,NA
+62659,7,2,1,18,NA,5,6,1,18,219,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,5526.901806,5811.556543,2,103,7,7,1.65,4,4,0,1,1,2,55,2,3,1,NA
+62660,7,2,2,7,NA,5,6,1,7,88,NA,NA,2,1,2,1,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,7932.110938,8947.143287,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+62661,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,84910.063417,90031.493834,2,96,14,7,3.67,3,1,0,0,0,1,30,2,5,5,NA
+62662,7,2,2,11,NA,1,1,1,11,138,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16664.698857,2,98,7,7,1.97,4,4,0,1,0,1,40,1,3,1,3
+62663,7,1,2,3,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12984.581505,0,2,93,15,15,5,4,4,2,0,0,1,34,1,5,1,5
+62664,7,2,1,64,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,25436.904729,25763.177128,1,94,8,8,3.3,2,2,0,0,2,1,64,1,5,1,3
+62665,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,13504.560081,14155.716256,1,102,7,7,1.8,5,4,1,0,2,1,47,1,3,5,NA
+62666,7,2,1,3,NA,5,6,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8838.066777,9670.229845,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+62667,7,2,1,6,NA,3,3,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53915.042746,57006.573447,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+62668,7,2,1,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,25120.174741,24762.106831,2,102,6,6,1.22,5,5,0,2,0,2,42,1,4,1,4
+62669,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,22249.799905,2,97,6,6,1.35,3,3,0,0,0,2,54,1,3,6,NA
+62670,7,2,1,13,NA,3,3,2,14,169,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,107087.58296,107224.801212,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+62671,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,7973.883342,8036.257566,1,100,77,77,NA,1,1,0,0,1,1,69,1,4,3,NA
+62672,7,2,2,13,NA,4,4,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11078.202838,10933.134393,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+62673,7,2,1,34,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,34959.343013,34695.493801,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+62674,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,NA,28989.076324,29596.20575,1,93,6,6,1.16,4,4,2,0,0,2,33,1,5,1,4
+62675,7,2,1,0,9,3,3,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16921.794985,16629.253929,1,99,15,15,5,4,4,1,1,0,2,27,1,5,1,5
+62676,7,2,2,17,NA,3,3,2,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,111142.989658,114235.009147,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+62677,7,2,2,30,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,56677.205755,57971.218225,2,102,10,10,4.76,2,2,0,0,0,1,27,1,4,1,4
+62678,7,2,1,68,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,1,2,2,2,2,9235.951997,9462.879133,2,103,5,5,1.79,2,1,0,0,2,1,68,2,1,1,NA
+62679,7,2,2,7,NA,1,1,1,7,90,NA,NA,2,2,2,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15897.166957,16345.216522,2,102,6,3,0.54,6,4,0,4,0,2,43,2,1,5,NA
+62680,7,2,1,68,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7581.139559,8116.967146,3,90,14,14,4.96,2,2,0,0,2,1,68,1,3,1,5
+62681,7,2,2,5,NA,1,1,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15457.736897,15510.51982,2,98,3,3,0.33,7,7,2,3,0,1,40,2,1,1,1
+62682,7,2,2,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,44636.780791,48652.73082,2,91,3,3,0.86,2,2,0,0,1,2,62,1,3,2,NA
+62683,7,2,2,55,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16829.599526,17568.396647,1,90,99,99,NA,3,3,0,1,1,1,60,1,5,1,5
+62684,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,82612.6713,83351.284903,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+62685,7,2,1,36,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,5,NA,1,2,1,1,2,1,1,2,1,1,12869.386353,13443.315161,2,92,7,7,1.89,3,3,0,0,1,1,36,2,3,5,NA
+62686,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,38007.846215,47247.232001,1,94,7,7,2.31,2,2,0,0,1,2,80,1,3,2,NA
+62687,7,2,1,10,NA,4,4,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8483.005475,8584.70346,1,96,8,8,2,4,4,1,2,0,2,40,1,4,5,NA
+62688,7,2,1,24,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,10141.381563,13212.383706,3,90,6,6,1.98,2,2,0,0,0,2,47,2,5,2,NA
+62689,7,2,1,44,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,116464.874823,116611.129855,2,94,15,15,5,5,5,0,3,0,1,44,1,5,1,4
+62690,7,2,2,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,17568.277926,19062.73336,1,90,99,99,NA,2,2,0,1,0,2,26,1,4,5,NA
+62691,7,2,1,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,19387.603198,2,100,7,7,1.79,4,4,0,1,0,2,51,1,3,3,NA
+62692,7,1,2,50,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17521.481386,0,2,97,7,7,3.49,1,1,0,0,0,2,50,1,4,5,NA
+62693,7,2,2,69,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,9570.416297,10279.373686,2,98,5,5,2.2,1,1,0,0,1,2,69,1,2,2,NA
+62694,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,21219.562726,22239.20882,2,91,3,3,0.86,2,2,0,0,1,2,62,1,3,2,NA
+62695,7,2,2,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,2,1,2,2,1,2,2,1,2,2,1,19130.246369,19734.446951,2,101,2,2,0.22,4,4,1,1,0,2,41,1,2,4,NA
+62696,7,2,1,13,NA,4,4,2,13,159,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,12116.15399,2,95,6,6,1.3,4,4,1,1,0,1,38,1,4,1,4
+62697,7,2,1,9,NA,2,2,2,9,119,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12477.812875,12553.43469,2,94,6,6,1.43,5,4,2,1,0,2,23,2,3,6,NA
+62698,7,1,1,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,38666.703155,0,2,94,77,77,NA,2,2,0,0,2,2,80,1,5,1,5
+62699,7,2,1,1,23,4,4,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4912.962876,5115.96715,2,90,8,8,1.67,6,6,2,2,0,2,35,2,5,3,NA
+62700,7,2,1,9,NA,4,4,1,9,119,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9022.8939,9530.778148,2,100,3,3,0.31,7,7,3,2,0,2,28,1,3,1,3
+62701,7,2,2,29,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,30253.427014,31766.413225,2,90,8,8,2.24,4,4,1,1,0,2,29,1,4,6,NA
+62702,7,1,2,12,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,55,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,20322.312754,0,2,91,6,6,1.3,4,4,1,1,0,2,27,1,4,6,NA
+62703,7,2,2,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7785.57328,8700.336584,1,96,8,8,2.17,4,4,1,1,0,2,41,1,3,1,3
+62704,7,2,1,80,NA,5,7,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,9071.406562,10878.533019,1,94,5,5,1.59,2,2,0,0,2,2,73,1,3,1,3
+62705,7,2,2,30,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,35353.005268,34399.106917,2,91,15,15,5,2,2,0,0,0,2,30,1,4,1,4
+62706,7,2,2,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,42150.923371,53077.558906,1,97,9,9,3.7,2,2,0,0,0,1,25,NA,NA,5,NA
+62707,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,25511.973394,26011.568993,1,98,6,6,0.97,7,7,1,2,0,1,49,1,2,1,2
+62708,7,2,1,52,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,13567.923118,14187.110343,3,91,6,6,1.34,4,4,0,2,0,1,52,2,3,1,1
+62709,7,2,2,10,NA,3,3,2,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,25563.654853,1,95,6,6,1.09,5,5,0,3,0,1,31,1,4,1,4
+62710,7,2,1,41,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,36225.654087,36329.951191,2,93,5,5,0.87,4,4,1,1,0,1,41,2,5,1,3
+62711,7,2,1,1,21,3,3,2,NA,23,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18848.609643,22085.568952,1,91,6,6,2.04,2,2,1,0,0,2,31,1,5,4,NA
+62712,7,2,2,4,NA,2,2,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16087.525226,17761.059567,2,91,4,4,0.69,4,4,2,0,0,2,21,1,3,6,NA
+62713,7,2,1,15,NA,3,3,2,15,183,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17395.533107,17264.495057,1,99,13,13,NA,4,4,0,2,0,1,55,NA,NA,1,4
+62714,7,2,1,75,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,1,1,NA,2,2,2,2,2,2,1,2,2,NA,14200.083364,15006.095575,3,92,3,3,0.68,2,2,0,0,2,1,75,1,1,1,1
+62715,7,2,1,10,NA,3,3,2,11,132,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,47596.305609,49235.587472,1,99,15,15,3.82,5,5,0,3,0,1,57,1,5,1,5
+62716,7,2,2,42,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,11762.034222,12708.069605,3,90,12,12,NA,5,5,1,2,0,1,37,2,5,1,5
+62717,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,32252.129527,31981.695662,1,99,6,4,1.38,2,1,0,0,0,2,50,1,3,6,NA
+62718,7,2,1,63,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,6038.685119,6085.921613,1,96,6,6,2.51,1,1,0,0,1,1,63,1,4,5,NA
+62719,7,2,1,74,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,4,NA,2,2,2,1,2,2,2,2,2,NA,14646.929358,15478.305035,1,96,15,8,4.66,2,1,0,0,2,2,68,2,1,3,NA
+62720,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,151766.599459,151581.541662,3,91,15,15,5,2,2,0,0,0,2,57,1,5,1,5
+62721,7,2,2,2,NA,3,3,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,33127.913744,33563.225688,1,99,15,15,5,4,4,2,0,0,1,33,1,5,1,5
+62722,7,2,1,1,20,1,1,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,6,6,1.21,4,4,1,0,0,2,49,2,2,6,NA
+62723,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,95214.22557,102645.939709,1,97,15,15,5,3,3,1,0,0,2,31,1,5,1,5
+62724,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,187291.098551,189225.431861,2,91,15,15,5,2,2,0,0,1,2,57,1,5,1,5
+62725,7,2,2,3,NA,5,6,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,2,NA,NA,NA,NA,5060.292252,5371.954509,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+62726,7,1,2,1,18,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39493.444219,0,2,103,15,15,5,4,4,2,0,0,1,36,2,4,1,5
+62727,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,43807.995099,48928.4376,1,95,6,6,1.65,2,2,0,0,2,2,80,1,4,1,4
+62728,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,32569.109434,37398.26507,1,96,77,77,NA,1,1,0,0,0,2,32,1,5,5,NA
+62729,7,2,2,1,14,2,2,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6962.449638,7451.646327,3,90,12,12,NA,5,5,1,1,0,1,35,2,5,3,NA
+62730,7,2,1,0,3,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7493.544598,7825.810309,1,98,6,6,0.97,7,7,1,2,0,1,49,1,2,1,2
+62731,7,2,1,19,NA,1,1,2,19,238,2,NA,2,2,2,12,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,21718.29328,21806.698186,2,94,NA,13,NA,3,2,0,0,0,1,44,2,2,6,NA
+62732,7,2,2,71,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,2,2,2,2,2,2,NA,18067.031882,19590.328656,2,94,2,2,0.56,2,2,0,0,2,2,71,2,2,1,1
+62733,7,1,1,5,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8253.750998,0,1,90,15,15,5,4,4,2,0,0,1,36,1,5,1,5
+62734,7,2,1,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18783.626894,19376.125899,2,99,2,2,0.53,2,2,0,0,1,2,69,1,4,2,NA
+62735,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,105261.096488,129835.348564,1,97,15,15,5,4,4,1,1,0,2,33,1,5,1,3
+62736,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,17036.134307,17206.306314,2,90,6,6,1.4,3,3,0,1,0,2,40,1,4,5,NA
+62737,7,2,1,58,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,10579.097347,10676.360686,2,92,3,3,0.7,3,3,0,0,0,1,58,2,1,1,1
+62738,7,2,1,6,NA,3,3,1,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20333.209695,22709.426802,1,103,6,6,1.11,5,5,1,1,1,1,29,1,3,1,3
+62739,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,60135.660369,64822.591837,1,90,8,8,2.49,3,3,0,0,2,1,74,1,5,1,2
+62740,7,2,2,19,NA,4,4,2,19,232,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,8,8,2.7,3,3,0,0,1,2,72,1,2,3,NA
+62741,7,2,2,0,3,4,4,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3861.876549,4122.902052,1,96,14,14,2.91,6,6,1,2,1,2,32,1,5,1,4
+62742,7,2,1,70,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,1,2,2,NA,12962.876803,13431.341497,2,90,6,6,1.7,2,2,0,0,2,1,70,2,1,1,4
+62743,7,2,1,0,5,3,3,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7467.727239,7549.772257,3,91,4,4,0.65,5,5,2,2,0,2,27,2,2,3,NA
+62744,7,2,1,37,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,43108.74283,45119.517495,2,91,9,9,2.6,4,4,1,1,0,2,31,2,4,1,5
+62745,7,2,2,71,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,12183.823561,13095.132048,2,100,5,5,1.63,2,2,0,0,2,1,71,NA,NA,1,2
+62746,7,2,2,38,NA,2,2,2,NA,NA,2,NA,2,1,4,NA,4,1,2,2,2,2,2,2,2,NA,NA,NA,NA,26897.674477,26781.441446,2,99,77,77,NA,4,4,1,1,1,2,38,2,4,1,4
+62747,7,2,1,52,NA,4,4,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19064.767778,19872.734197,2,93,14,14,2.78,5,5,0,0,0,1,52,2,4,1,2
+62748,7,2,1,13,NA,2,2,1,13,164,NA,NA,2,2,1,7,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,15674.964193,15512.049412,2,93,9,9,1.49,7,7,0,3,0,2,41,2,5,1,5
+62749,7,2,1,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,22039.963414,22735.178265,1,96,15,15,5,2,2,0,0,0,1,48,1,2,1,3
+62750,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,15890.289377,17041.022437,2,94,3,3,0.68,3,2,0,1,0,2,45,2,4,3,NA
+62751,7,2,1,8,NA,3,3,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21839.599095,22875.076583,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+62752,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,59510.728426,61449.871516,1,99,15,15,5,5,5,3,0,0,2,34,1,5,1,5
+62753,7,2,2,16,NA,4,4,1,16,203,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18421.460617,19740.37466,1,92,6,6,1.3,4,4,0,1,1,1,25,1,1,1,3
+62754,7,2,2,0,7,3,3,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25216.508684,24531.572308,1,101,14,14,3.38,4,4,2,0,0,1,32,1,4,1,5
+62755,7,2,2,19,NA,4,4,2,19,235,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,3,3,0.66,3,3,2,0,0,2,19,1,3,NA,NA
+62756,7,2,1,7,NA,3,3,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25604.034863,28346.339599,1,95,7,7,1.66,5,5,0,3,0,1,34,1,2,1,4
+62757,7,2,1,54,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,105571.343095,105442.613847,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+62758,7,1,2,9,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7722.982971,0,1,93,15,15,5,3,3,0,2,0,2,40,2,5,4,NA
+62759,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,18395.989535,18290.733411,2,93,9,9,2.07,5,5,0,1,0,1,55,NA,NA,5,NA
+62760,7,2,2,10,NA,2,2,2,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9872.244853,10573.701398,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+62761,7,2,1,68,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11992.012141,12562.386395,2,102,6,6,1.97,2,2,0,0,2,2,67,1,3,1,3
+62762,7,2,2,1,13,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12260.86913,13122.344171,3,92,10,10,2.26,6,6,1,3,0,1,34,1,4,1,1
+62763,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,110931.964733,116248.869259,1,94,15,15,5,2,2,0,0,0,1,29,1,4,1,5
+62764,7,2,2,51,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,22969.116046,23881.325913,2,93,1,1,0.36,1,1,0,0,0,2,51,1,3,5,NA
+62765,7,2,2,29,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16844.740449,17564.040518,3,91,7,7,2.45,2,2,0,0,0,2,29,2,5,1,5
+62766,7,2,1,1,22,5,6,2,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5891.941477,6363.138629,1,90,6,6,0.92,6,6,2,0,2,2,30,2,5,1,5
+62767,7,1,2,40,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,NA,NA,NA,NA,NA,NA,NA,29650.79971,0,2,90,NA,NA,NA,1,1,0,0,0,2,40,1,4,5,NA
+62768,7,2,2,38,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,3,4,2,2,2,2,1,2,2,2,2,2,2,42456.72357,41311.151976,1,95,4,4,0.68,5,5,0,1,0,2,38,2,3,4,NA
+62769,7,2,2,13,NA,3,3,2,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74165.041171,76338.869017,2,94,5,5,0.89,4,4,0,2,0,2,51,1,2,3,NA
+62770,7,2,1,1,15,3,3,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51373.484735,55257.405076,3,91,14,14,4.32,3,3,1,0,0,1,28,1,5,1,5
+62771,7,2,1,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,21863.450986,22291.478781,2,99,2,2,0.75,1,1,0,0,0,1,30,1,2,5,NA
+62772,7,2,2,46,NA,1,1,2,NA,NA,2,NA,2,2,1,NA,5,5,NA,2,2,2,1,2,2,2,2,2,2,29995.384937,31080.287607,1,93,15,15,5,3,2,0,1,0,2,46,2,5,5,NA
+62773,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,27693.650341,27908.388498,3,91,1,1,0.19,3,3,0,2,0,2,50,1,4,3,NA
+62774,7,1,1,80,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11524.806252,0,1,94,3,3,0.39,6,6,1,0,2,1,80,1,4,1,3
+62775,7,1,2,69,NA,2,2,NA,NA,NA,2,NA,2,1,6,NA,1,3,NA,2,2,2,2,2,2,NA,NA,NA,NA,13676.984152,0,2,91,6,6,1.36,3,3,0,0,1,2,48,2,1,5,NA
+62776,7,2,1,9,NA,4,4,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10410.106675,10617.081899,2,96,6,6,1.62,3,3,0,2,0,2,31,1,3,5,NA
+62777,7,2,1,28,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18927.052732,18722.519921,2,92,15,5,1.93,4,1,0,0,0,1,28,1,5,5,NA
+62778,7,2,2,4,NA,3,3,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78365.406271,80834.23308,2,91,14,14,4.12,4,4,2,0,0,1,35,1,5,1,5
+62779,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,81291.35274,86533.223356,2,95,12,99,NA,2,1,0,0,0,2,40,1,4,6,NA
+62780,7,2,2,5,NA,3,3,1,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,83950.915901,92655.852423,1,100,15,15,5,4,4,1,1,0,1,40,1,5,1,5
+62781,7,2,2,3,NA,1,1,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10493.785765,10940.90082,1,103,5,5,0.71,6,6,2,2,0,2,31,2,2,1,2
+62782,7,2,2,0,3,5,6,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7997.590655,8653.844084,3,91,15,15,5,4,4,2,0,0,1,36,2,5,1,5
+62783,7,2,2,6,NA,2,2,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,15897.166957,16345.216522,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+62784,7,2,1,31,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,46646.692487,3,92,7,7,1.49,5,5,0,2,1,2,62,1,4,2,NA
+62785,7,2,1,4,NA,4,4,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8005.528865,8827.404644,1,99,14,14,3.8,4,4,1,1,0,1,48,2,5,1,5
+62786,7,2,1,7,NA,1,1,1,8,96,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,16605.106492,16594.063114,1,92,4,4,0.74,4,4,1,1,0,1,42,2,3,1,4
+62787,7,2,2,21,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,11739.283384,12800.154398,1,96,15,15,5,5,5,0,0,0,1,58,2,5,1,5
+62788,7,2,1,35,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,6,NA,2,2,2,1,2,2,2,2,2,2,34887.439952,36924.956604,2,94,4,4,0.73,5,5,2,1,0,1,35,2,1,6,NA
+62789,7,2,2,26,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,16929.836231,18556.498474,2,101,6,5,2.2,2,1,0,0,0,2,26,2,4,6,NA
+62790,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7631.175557,7797.231767,1,93,1,1,0.02,5,5,0,4,0,2,36,NA,NA,5,NA
+62791,7,2,1,14,NA,3,3,1,14,173,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24231.355333,25219.610889,2,100,5,5,1.3,3,3,0,1,0,2,46,1,3,2,NA
+62792,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,23128.736624,26524.976891,1,98,3,3,1.09,1,1,0,0,0,2,52,1,4,3,NA
+62793,7,2,1,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,87347.444626,90661.31559,1,95,15,15,5,2,2,0,0,2,1,72,1,3,1,4
+62794,7,2,2,45,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,19757.142379,19558.756904,1,96,8,8,2.17,4,4,0,2,0,2,45,2,5,4,NA
+62795,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,35522.958395,36075.029683,1,102,13,13,NA,2,1,0,0,0,1,22,1,4,5,NA
+62796,7,2,2,17,NA,5,6,2,17,211,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,6937.463063,7785.963668,3,90,NA,NA,NA,4,4,0,1,0,1,56,NA,NA,1,NA
+62797,7,2,2,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16929.836231,20416.10433,2,101,7,5,1.84,2,1,0,0,0,2,21,NA,NA,5,NA
+62798,7,2,1,50,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,2,2,2,2,27240.276328,27670.554468,1,90,6,6,1.57,3,3,0,0,0,1,50,2,2,1,2
+62799,7,2,2,0,2,4,4,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3800.155777,4057.009552,2,93,3,3,0.36,6,6,1,1,2,2,69,2,3,1,3
+62800,7,2,2,15,NA,3,3,2,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,103298.858809,106172.653059,1,95,9,9,2.66,4,4,0,2,0,1,45,1,3,1,3
+62801,7,2,2,19,NA,3,3,2,19,229,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,23572.289725,23898.02545,2,95,3,3,0.63,3,3,0,0,0,2,44,1,2,4,NA
+62802,7,2,1,6,NA,3,3,1,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64054.596883,66498.597232,1,92,6,6,1.57,3,3,0,1,0,1,29,1,4,6,NA
+62803,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,71525.773464,1,98,14,14,3.15,5,5,0,3,0,1,34,1,4,1,4
+62804,7,2,1,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,97936.972803,98855.488943,2,95,14,14,5,2,2,0,0,2,2,66,1,5,1,5
+62805,7,2,2,41,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,17059.906216,21044.744151,3,90,9,9,4.1,2,2,0,1,0,2,41,2,5,5,NA
+62806,7,2,1,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,23775.734331,24513.29876,2,96,4,2,0.76,2,1,0,0,0,1,53,1,2,6,NA
+62807,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,9,9,4.15,2,2,0,0,2,2,64,1,4,1,5
+62808,7,2,2,44,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,2,2,2,2,1,2,2,1,2,2,2,31334.47528,32083.194015,2,96,5,5,0.78,5,5,0,2,0,1,37,2,1,5,NA
+62809,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,27004.065356,27080.034561,1,96,1,1,0.23,1,1,0,0,0,2,51,1,3,2,NA
+62810,7,2,2,15,NA,3,3,1,15,190,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,105891.533689,107879.366927,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+62811,7,2,2,40,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,19258.69251,18731.627868,2,99,12,6,2.69,6,1,0,0,0,2,57,1,5,2,NA
+62812,7,2,1,50,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,37557.946192,37006.841228,2,102,NA,2,0.46,2,1,0,0,1,1,50,2,1,6,NA
+62813,7,2,1,71,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,NA,13555.672819,14397.395601,1,94,3,3,1.16,1,1,0,0,1,1,71,1,4,5,NA
+62814,7,2,1,72,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,8601.453077,8734.617022,2,101,5,5,1.7,2,2,0,0,1,1,72,1,2,2,NA
+62815,7,2,1,4,NA,2,2,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13504.027725,13522.274849,2,93,8,8,2.17,4,4,2,0,0,2,30,1,4,1,4
+62816,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,1,1,2,1,1,2,1,3,14385.653726,15564.966804,2,101,7,4,1.38,2,1,0,0,0,1,23,2,4,5,NA
+62817,7,2,1,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,19692.655418,21458.476044,2,99,6,3,0.94,3,2,0,0,0,1,41,1,3,6,NA
+62818,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15096.339697,2,100,7,7,1.34,5,5,0,2,0,2,53,1,4,4,NA
+62819,7,2,2,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140431.173819,141856.211353,1,91,8,8,2.17,4,4,0,0,0,1,59,1,4,1,5
+62820,7,2,2,53,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,1,2,1,1,2,2,1,2,1,NA,21760.356138,22156.586691,1,92,4,4,0.76,4,4,0,0,0,2,53,2,1,1,1
+62821,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,16058.142925,17073.637087,2,95,7,6,1.88,3,2,0,0,0,2,56,1,3,2,NA
+62822,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,114838.671743,118127.710351,1,91,8,4,1.38,2,1,0,0,0,1,33,1,5,6,NA
+62823,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,20393.886544,21026.540202,2,97,1,1,0.18,1,1,0,0,0,1,46,1,2,4,NA
+62824,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,29670.405171,29084.152013,2,101,14,14,4.03,4,4,0,1,0,2,40,1,5,1,5
+62825,7,2,2,3,NA,1,1,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9468.006743,10197.085534,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+62826,7,1,2,80,NA,5,7,NA,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,36240.721121,0,1,101,3,3,1.12,1,1,0,0,1,2,80,1,1,2,NA
+62827,7,2,2,0,6,5,7,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25216.508684,26724.106581,1,101,6,6,1.28,4,4,2,0,0,1,44,1,4,1,4
+62828,7,2,2,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,3,1,2,2,1,2,2,NA,NA,NA,NA,60324.348827,61156.73477,1,95,7,2,0.78,2,1,0,0,0,1,26,1,3,6,NA
+62829,7,2,1,0,3,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8429.177912,8429.333277,1,95,15,15,5,3,3,1,0,0,1,41,1,5,1,5
+62830,7,2,1,14,NA,3,3,2,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,88448.252445,87285.839384,1,101,14,14,5,3,3,0,1,0,2,36,1,5,1,5
+62831,7,2,2,4,NA,1,1,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10024.946819,10059.178608,2,93,77,77,NA,7,7,3,1,0,2,43,2,1,1,9
+62832,7,2,2,15,NA,1,1,1,15,189,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,16427.640886,17012.806816,1,102,3,3,0.54,3,3,0,2,0,2,42,2,2,3,NA
+62833,7,2,2,18,NA,3,3,2,18,219,2,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,31627.471241,33789.841724,1,95,1,1,0.21,4,4,1,0,1,2,75,1,1,2,NA
+62834,7,2,1,6,NA,1,1,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11159.151566,11226.781631,1,102,6,6,0.8,7,7,3,3,0,2,34,2,3,1,1
+62835,7,2,2,66,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,12139.361543,12607.897559,2,103,14,14,3.48,5,5,0,2,1,1,43,1,4,1,5
+62836,7,2,1,25,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,5,NA,2,2,2,2,2,2,NA,NA,NA,NA,39899.764102,44211.181291,2,93,4,4,0.82,4,4,0,0,0,1,51,2,3,1,3
+62837,7,2,2,18,NA,3,3,2,18,221,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,31716.869763,32646.512808,1,95,77,77,NA,3,3,0,0,0,2,41,1,2,5,NA
+62838,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,19258.69251,19091.502506,2,99,12,3,1.16,6,1,0,0,0,2,57,1,5,2,NA
+62839,7,2,2,3,NA,4,4,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8833.042831,9629.276723,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+62840,7,2,1,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,1,2,1,116001.539479,120724.7298,2,101,5,5,1.27,3,3,0,0,1,2,55,1,3,1,1
+62841,7,2,1,52,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,28867.753408,2,101,8,8,2.43,3,3,0,1,0,1,52,1,2,1,5
+62842,7,2,1,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9885.965005,10900.89298,2,97,3,3,0.82,2,2,1,0,0,2,22,1,4,5,NA
+62843,7,1,1,14,NA,2,2,NA,NA,NA,NA,NA,2,1,77,8,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,26616.794908,0,2,91,1,1,0.17,4,4,0,1,0,1,49,2,1,6,NA
+62844,7,2,1,7,NA,5,6,2,7,91,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4852.395137,5120.382571,1,99,14,14,2.66,7,7,3,1,0,1,35,1,5,1,5
+62845,7,2,1,27,NA,4,4,1,NA,NA,2,NA,2,2,6,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18831.340773,22141.643614,2,93,14,14,2.78,5,5,0,0,0,1,52,2,4,1,2
+62846,7,1,2,80,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,12535.973802,0,2,100,5,2,0.41,3,2,0,0,2,2,80,1,1,2,NA
+62847,7,2,1,44,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,33514.643153,35491.262929,1,90,15,15,5,4,4,0,2,0,2,43,1,5,1,5
+62848,7,2,1,38,NA,4,4,1,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,22185.88687,25293.081507,1,98,7,4,1.61,2,1,0,0,0,1,38,2,5,5,NA
+62849,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,19075.861607,19193.852226,1,96,15,15,5,4,4,0,1,0,1,42,2,4,1,4
+62850,7,2,2,1,14,5,7,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6257.554806,6809.629993,2,92,15,15,5,3,3,1,0,0,2,37,1,5,1,5
+62851,7,2,1,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,94547.245282,99362.23578,2,91,5,5,1.39,2,2,0,0,0,1,58,1,4,3,NA
+62852,7,2,2,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8602.397328,9613.133098,2,99,13,13,NA,3,3,1,1,0,2,36,NA,NA,5,NA
+62853,7,2,1,16,NA,5,6,1,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,5526.901806,5811.556543,2,103,7,7,1.65,4,4,0,1,1,2,55,2,3,1,NA
+62854,7,2,1,8,NA,4,4,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13239.363106,13690.337986,3,92,6,6,0.93,5,5,2,1,0,2,37,1,5,1,3
+62855,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,147198.558516,148052.525251,2,91,14,14,5,1,1,0,0,0,2,47,1,4,3,NA
+62856,7,2,2,2,NA,1,1,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10575.997988,11390.397051,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+62857,7,2,2,0,5,4,4,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3659.311381,3880.295282,1,99,8,8,1.99,5,5,1,0,0,1,55,1,5,1,2
+62858,7,2,1,17,NA,4,4,2,17,208,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11763.86086,12201.333419,1,99,14,14,4.86,3,3,0,1,1,2,56,1,5,1,5
+62859,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16058.142925,18803.458224,2,95,7,7,1.41,5,5,2,0,0,2,53,1,3,3,NA
+62860,7,2,1,55,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,16499.662173,16948.4296,3,91,6,6,1.81,3,3,0,1,0,2,47,2,3,1,3
+62861,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,60071.993203,66823.228048,1,95,5,5,1.88,1,1,0,0,1,2,80,1,3,1,NA
+62862,7,2,1,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9661.66953,9494.640263,2,95,4,4,0.78,4,4,2,0,0,1,27,1,4,1,4
+62863,7,2,2,1,13,1,1,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,14781.923489,3,92,5,5,0.87,4,4,2,0,0,2,28,1,3,1,3
+62864,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,32041.645327,32334.772002,1,94,1,1,0.32,1,1,0,0,0,2,57,1,4,5,NA
+62865,7,1,1,17,NA,5,6,NA,NA,NA,2,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,10346.302718,0,2,91,4,4,0.65,5,5,1,3,0,1,43,2,3,5,NA
+62866,7,2,2,5,NA,3,3,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,31376.988784,34630.493438,1,101,1,1,0.1,4,4,1,1,0,2,52,1,4,3,NA
+62867,7,2,1,12,NA,3,3,2,12,152,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21603.199906,21630.881484,2,95,5,5,1.05,3,3,0,1,0,1,43,1,3,1,2
+62868,7,2,1,27,NA,3,3,2,NA,NA,2,NA,2,1,2,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23002.111021,24919.46038,2,97,6,6,1.16,4,4,1,1,0,1,27,2,4,1,3
+62869,7,2,1,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6936.347746,7364.53247,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+62870,7,2,1,13,NA,1,1,1,13,163,NA,NA,2,2,4,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,24228.858782,24791.594558,2,102,4,4,0.57,5,5,0,3,0,1,41,2,1,1,2
+62871,7,2,1,48,NA,5,7,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12490.95898,12886.461592,2,92,15,15,4.59,4,4,0,2,0,2,45,2,5,1,5
+62872,7,2,2,17,NA,3,3,1,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,78268.283998,80112.547303,2,98,15,15,5,4,4,0,2,0,2,46,1,4,1,NA
+62873,7,1,1,15,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10879.348751,0,2,100,10,10,2.33,6,6,0,2,2,2,35,1,2,5,NA
+62874,7,2,2,33,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,3,6,2,2,2,2,1,2,2,NA,NA,NA,NA,33355.357535,33078.896803,1,90,6,6,1.11,5,5,1,2,0,1,30,2,1,6,NA
+62875,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,98217.832546,101788.674762,1,99,8,8,2.81,3,3,0,1,0,1,19,1,4,NA,NA
+62876,7,2,2,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,63504.762752,64182.95787,1,91,7,7,3.67,1,1,0,0,1,2,74,1,4,2,NA
+62877,7,2,1,8,NA,4,4,1,8,105,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,14179.490667,2,101,1,1,0.15,3,3,0,2,0,2,58,1,3,5,NA
+62878,7,2,1,15,NA,4,4,1,15,187,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16463.950838,16583.858395,2,96,5,5,0.67,6,6,1,2,1,1,34,1,4,1,4
+62879,7,2,2,14,NA,4,4,2,14,171,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,18028.02401,2,101,8,8,2.43,3,3,0,1,0,1,52,1,2,1,5
+62880,7,2,2,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,117778.281347,120111.495123,1,94,7,7,1.52,4,4,0,2,2,1,61,2,1,1,5
+62881,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,134213.669088,137679.44344,1,93,15,15,4.59,4,4,0,1,0,1,57,1,5,1,5
+62882,7,2,1,59,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11690.444016,12008.407525,3,90,77,77,NA,4,4,0,0,0,1,59,2,2,1,2
+62883,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,43807.995099,48928.4376,1,101,7,7,2.31,2,2,0,0,2,2,80,1,4,1,2
+62884,7,2,2,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,1,2,2,1,23725.035562,23797.982183,1,94,3,1,0.13,2,1,0,0,0,2,36,1,3,5,NA
+62885,7,2,1,60,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,9392.289727,9639.798184,1,90,15,15,5,3,3,0,0,1,1,60,2,4,2,NA
+62886,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15576.264148,16963.869897,1,103,8,8,1.95,4,4,0,1,0,2,48,1,5,1,5
+62887,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,11982.125462,2,101,1,1,0.2,2,2,0,0,1,2,55,1,4,4,NA
+62888,7,2,1,6,NA,4,4,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10410.106675,10390.534388,2,96,4,4,0.57,5,5,0,3,0,2,26,1,2,5,NA
+62889,7,2,1,2,NA,2,2,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8914.157023,9196.329398,1,99,8,8,1.91,5,5,2,0,1,2,38,2,4,1,4
+62890,7,2,2,16,NA,5,6,1,16,193,NA,NA,2,1,5,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8210.014908,8564.874659,2,92,15,15,5,4,4,0,2,1,2,59,1,5,1,5
+62891,7,2,2,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,13141.36586,14124.295248,1,96,12,12,NA,1,1,0,0,1,2,74,1,2,3,NA
+62892,7,2,1,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18225.504252,18491.00599,2,95,4,4,0.79,3,3,1,0,0,1,50,1,4,6,NA
+62893,7,2,2,21,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,2,4,2,2,2,2,2,2,2,NA,NA,NA,NA,44119.608456,46238.164206,2,98,6,6,1.21,4,4,1,0,0,2,49,2,2,6,NA
+62894,7,2,2,9,NA,4,4,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7620.214819,8204.982373,2,99,7,7,1.19,6,6,1,3,0,2,38,1,3,5,NA
+62895,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,78240.016337,81914.532146,1,93,15,15,5,3,3,1,0,0,1,33,1,5,1,3
+62896,7,2,2,5,NA,3,3,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53166.229434,56500.79967,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+62897,7,2,1,5,NA,2,2,2,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17458.543556,17482.134163,2,91,10,10,3.04,4,4,1,1,0,2,31,2,5,1,5
+62898,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,22473.275629,22284.590536,2,97,7,7,2.38,2,2,0,0,0,1,25,1,2,1,2
+62899,7,2,2,45,NA,2,2,2,NA,NA,2,NA,2,2,77,NA,3,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,23377.708086,24402.18571,3,90,77,77,NA,4,1,0,0,0,1,45,2,3,3,NA
+62900,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8949.073879,2,95,6,6,2.69,1,1,0,0,1,2,68,1,3,2,NA
+62901,7,2,2,64,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11704.708696,12115.082932,1,101,15,15,5,3,3,0,0,1,1,58,2,4,1,5
+62902,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,35126.205635,36772.568368,1,95,5,5,1.08,3,3,0,1,0,1,53,1,4,1,4
+62903,7,2,1,7,NA,5,6,2,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,9720.482616,10818.978947,2,91,12,12,NA,7,6,0,4,2,2,72,2,1,2,NA
+62904,7,2,1,2,NA,4,4,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6390.839385,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+62905,7,2,2,1,20,5,6,2,NA,20,NA,NA,2,2,1,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4735.146545,4730.145135,3,90,77,77,NA,3,3,1,0,0,1,35,2,3,1,5
+62906,7,2,2,0,0,1,1,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6885.817041,7118.841014,2,96,2,2,0.27,5,5,1,2,0,2,26,1,2,1,2
+62907,7,2,2,16,NA,2,2,2,16,200,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,5,5,0.8,5,5,0,3,0,2,40,2,1,5,NA
+62908,7,2,1,21,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,5,NA,2,2,2,1,2,2,2,2,2,2,30581.409928,30998.258142,3,90,7,7,1.48,5,5,0,1,0,1,43,2,1,6,NA
+62909,7,2,2,31,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,24598.096547,23934.388227,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+62910,7,2,2,7,NA,4,4,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12120.418061,13050.526646,2,101,10,10,2.91,4,4,0,1,0,2,51,1,2,5,NA
+62911,7,2,1,16,NA,1,1,1,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,24902.864049,25283.513039,1,94,8,8,1.85,5,5,0,2,0,1,44,2,1,6,NA
+62912,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,100742.761147,101756.213818,2,100,8,8,4.59,1,1,0,0,0,1,59,1,3,3,NA
+62913,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,NA,NA,NA,NA,19463.737283,25688.523518,3,91,4,4,1.09,2,2,0,1,0,2,40,1,5,3,NA
+62914,7,2,2,4,NA,3,3,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27528.66901,27890.404993,1,95,1,1,0.31,2,2,1,0,0,2,23,1,3,3,NA
+62915,7,1,1,35,NA,1,1,NA,NA,NA,2,NA,2,2,3,NA,4,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,39073.76885,0,2,99,99,3,0.66,4,2,0,0,0,1,35,2,4,1,2
+62916,7,2,1,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37911.437415,38922.114085,2,103,4,4,0.79,3,3,0,0,0,2,42,2,2,5,NA
+62917,7,2,1,13,NA,3,3,2,13,159,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,69216.263169,70543.167563,1,91,14,14,3.15,5,5,0,1,0,2,50,1,5,1,5
+62918,7,2,1,25,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,18259.989639,18151.137608,2,95,6,6,1.08,4,4,1,0,0,2,42,1,4,4,NA
+62919,7,2,2,34,NA,3,3,2,NA,NA,2,NA,2,2,99,NA,1,1,2,1,2,1,1,2,1,NA,NA,NA,NA,20717.382973,22040.621269,2,97,1,1,0.21,4,4,2,0,0,2,34,2,1,1,2
+62920,7,2,1,3,NA,2,2,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12920.2211,12937.679363,2,90,14,14,3.58,4,4,1,1,0,1,37,1,3,1,4
+62921,7,2,2,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8147.287486,8590.325322,2,90,8,8,1.67,6,6,2,2,0,2,35,2,5,3,NA
+62922,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,89807.047643,92234.66957,3,91,6,6,2.57,1,1,0,0,0,2,31,1,4,5,NA
+62923,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,20333.447198,19976.945976,1,98,12,4,1.34,4,1,0,0,0,1,21,1,4,6,NA
+62924,7,2,2,79,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,16361.152596,17584.911061,1,100,3,3,0.91,1,1,0,0,1,2,79,1,1,2,NA
+62925,7,2,2,18,NA,3,3,2,18,226,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22699.241957,23364.572192,2,95,2,2,0.46,3,3,0,0,0,2,48,1,2,1,2
+62926,7,2,2,1,21,1,1,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11414.885224,11901.246118,1,94,6,6,1.11,5,5,1,2,0,2,41,2,1,1,1
+62927,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,138075.879417,141933.339512,2,91,15,15,5,3,3,0,0,2,2,62,1,4,1,4
+62928,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,15206.604563,14790.435937,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+62929,7,2,1,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9221.19173,9293.322792,2,95,77,77,NA,2,2,0,0,2,1,68,1,4,1,4
+62930,7,2,1,23,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,4,5,NA,2,2,2,1,2,2,1,2,2,1,42077.383821,44160.926598,1,102,4,4,0.67,4,4,0,1,0,1,23,2,4,5,NA
+62931,7,2,1,15,NA,5,6,1,15,187,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11171.449402,11936.033174,1,92,14,14,3.47,4,4,0,2,0,2,53,2,4,1,4
+62932,7,1,2,2,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,35285.563011,0,2,100,15,15,5,4,4,1,1,0,1,29,1,4,1,4
+62933,7,2,2,17,NA,5,6,2,17,207,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,9,9,1.78,6,6,0,1,1,1,45,2,3,1,3
+62934,7,2,2,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,39030.893313,44151.142675,1,91,7,7,2.2,3,3,0,0,1,2,60,1,2,2,NA
+62935,7,2,1,51,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,2,2,2,2,21139.303536,21566.542108,2,103,12,12,NA,4,4,0,1,0,2,50,2,3,1,4
+62936,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,32985.054579,1,95,5,5,1.03,4,4,0,2,0,1,33,1,3,1,3
+62937,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10598.2543,1,99,5,5,1.63,2,2,0,0,2,1,64,NA,NA,1,4
+62938,7,2,2,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,36053.766709,37679.125242,1,100,6,6,1.13,4,4,0,3,0,2,32,1,3,5,NA
+62939,7,2,1,38,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,41155.167164,49173.131788,1,102,7,7,1.41,5,5,0,2,2,1,72,1,4,1,3
+62940,7,2,1,62,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9814.165625,10072.791483,1,90,15,15,4.34,4,4,0,0,1,1,62,2,5,1,3
+62941,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,13728.308948,13352.59776,2,90,99,4,1.16,5,2,0,0,1,1,43,NA,NA,1,NA
+62942,7,1,1,48,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,91761.294148,0,2,92,15,15,5,3,3,1,0,0,1,48,1,5,1,5
+62943,7,2,1,15,NA,2,2,1,15,189,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18042.255087,18115.696579,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+62944,7,2,2,17,NA,4,4,2,17,211,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13775.846051,13737.029624,2,97,7,7,1.06,7,7,1,2,0,2,40,1,4,5,NA
+62945,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19100.40225,19426.955733,1,96,15,15,4.9,4,4,0,1,0,1,47,1,3,1,5
+62946,7,2,1,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9699.683862,9681.447258,2,101,2,2,0.27,6,6,0,3,0,2,45,1,2,5,NA
+62947,7,1,2,68,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,1,2,NA,2,2,2,1,2,2,NA,NA,NA,NA,11469.456138,0,1,90,4,4,0.79,3,3,0,0,1,2,68,1,1,2,NA
+62948,7,2,1,79,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,55509.98679,58756.097166,1,101,15,15,5,3,3,0,0,2,1,79,1,4,1,4
+62949,7,2,2,8,NA,4,4,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9139.784234,9384.299154,2,100,14,14,3.89,4,4,0,2,0,1,38,1,3,1,4
+62950,7,2,2,16,NA,3,3,2,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,91539.042546,96988.607103,1,101,14,14,3.3,4,4,0,2,0,2,42,1,4,1,3
+62951,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,15190.604914,17388.543575,1,96,15,15,5,6,6,1,1,1,2,44,1,3,1,3
+62952,7,2,1,20,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,10547.036825,10866.006067,3,90,15,15,3.7,5,5,0,0,0,1,56,2,3,1,3
+62953,7,2,1,44,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17249.311662,18125.470447,2,91,14,14,3.47,4,4,1,1,0,2,36,2,3,1,5
+62954,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18523.956321,18091.254104,2,99,2,2,0.72,1,1,0,0,0,1,28,1,5,5,NA
+62955,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,1,2,2,1,76827.086279,81170.917712,2,99,15,15,5,2,1,0,0,0,2,31,1,5,1,NA
+62956,7,2,2,0,7,5,6,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,4444.744532,4630.927855,2,90,3,3,0.65,5,3,1,2,0,1,44,2,5,1,5
+62957,7,2,2,3,NA,5,6,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5307.591514,5454.53335,2,102,3,3,0.38,5,5,3,0,0,2,30,2,2,1,4
+62958,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,65611.003966,67854.891478,1,94,15,15,4.95,4,4,0,0,2,1,72,1,3,1,3
+62959,7,2,2,13,NA,5,6,1,13,166,NA,NA,2,1,4,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8210.014908,8564.874659,2,92,15,15,5,4,4,0,2,1,2,59,1,5,1,5
+62960,7,2,2,3,NA,4,4,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9727.363166,10466.751274,2,97,6,6,1,6,6,1,2,2,2,60,1,2,2,NA
+62961,7,2,1,0,6,4,4,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7882.953266,8719.306286,2,96,5,5,1.5,2,2,1,0,0,2,22,1,3,5,NA
+62962,7,2,1,2,NA,4,4,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8009.966208,8254.622305,2,96,2,2,0.44,3,3,2,0,0,2,22,1,2,5,NA
+62963,7,2,2,6,NA,5,7,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22371.648216,23759.450609,1,95,6,6,1.15,5,5,2,1,0,1,29,1,4,6,NA
+62964,7,1,1,22,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,13171.824182,0,1,100,NA,NA,NA,4,4,0,0,1,1,21,NA,NA,5,NA
+62965,7,2,2,3,NA,5,7,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,29364.157365,32408.949946,1,102,2,2,0.49,3,3,1,0,0,1,20,1,2,6,NA
+62966,7,2,2,21,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,24185.894115,24390.02655,1,93,3,3,0.7,3,3,1,0,0,1,23,2,4,1,2
+62967,7,2,1,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,27572.205373,28895.579143,1,100,15,15,5,3,3,0,0,0,2,33,1,5,1,4
+62968,7,2,1,5,NA,2,2,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13610.050526,14582.186889,2,100,3,3,0.76,3,3,1,0,0,2,31,2,1,6,NA
+62969,7,2,1,14,NA,4,4,2,14,178,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16890.963304,17218.04077,1,91,10,10,2.56,5,5,0,3,0,1,51,2,5,1,4
+62970,7,1,1,6,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12577.115885,0,2,96,3,3,0.54,4,4,1,1,0,1,29,1,2,1,2
+62971,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,175544.769665,178639.415011,1,91,15,15,5,2,2,0,0,0,1,53,1,5,1,5
+62972,7,2,1,3,NA,2,2,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17458.543556,18705.569422,2,91,4,4,0.84,3,3,1,0,0,2,21,1,4,1,2
+62973,7,2,1,10,NA,1,1,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17882.621856,18280.794545,3,92,12,12,NA,6,6,1,3,0,2,33,1,5,1,4
+62974,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,86986.68246,91060.136414,2,94,9,9,5,1,1,0,0,0,1,30,1,5,5,NA
+62975,7,2,2,10,NA,4,4,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7397.129324,7630.740597,1,96,12,12,NA,5,5,1,2,0,2,35,1,5,1,4
+62976,7,2,2,18,NA,3,3,2,18,221,2,NA,2,2,5,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,82089.722105,85880.903702,3,91,14,14,5,2,2,0,0,0,2,48,2,5,3,NA
+62977,7,2,1,25,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,50326.347525,51411.397032,1,93,12,12,NA,3,1,0,0,0,1,24,NA,NA,5,NA
+62978,7,2,2,18,NA,1,1,1,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21062.314667,23105.758741,1,91,14,14,3.9,4,4,0,1,0,1,41,1,2,1,4
+62979,7,2,2,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,53370.063988,54427.336747,3,92,4,4,1.13,2,2,0,0,2,1,64,1,3,1,4
+62980,7,1,1,39,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,19384.896286,0,1,94,5,5,1.05,3,3,0,1,0,1,39,1,4,6,NA
+62981,7,2,1,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,25120.174741,25387.314373,2,102,15,15,4.47,4,4,1,1,0,1,32,1,5,1,4
+62982,7,2,2,66,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,49644.076348,54110.530387,1,97,3,3,1.16,1,1,0,0,1,2,66,1,3,2,NA
+62983,7,2,1,7,NA,4,4,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7730.47951,9212.541007,1,99,6,6,1.3,5,5,1,2,0,1,34,1,2,1,3
+62984,7,2,2,5,NA,2,2,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8825.559072,9505.166523,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+62985,7,2,1,68,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11138.867102,11750.838884,1,102,7,7,2.72,2,2,0,0,2,2,67,NA,NA,1,5
+62986,7,2,1,14,NA,3,3,1,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,69211.537407,68936.770872,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+62987,7,2,1,10,NA,2,2,1,10,125,NA,NA,2,1,3,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13395.612951,13401.547169,2,102,7,7,1.53,5,5,1,2,0,2,37,2,4,1,4
+62988,7,2,1,10,NA,2,2,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11036.458246,11087.266596,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+62989,7,2,2,1,19,5,6,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8173.816615,8894.95474,1,97,15,15,5,4,4,2,0,0,2,35,2,4,1,4
+62990,7,2,1,79,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,79754.311902,90534.066519,1,97,9,9,3.97,2,2,0,0,2,1,79,1,3,1,3
+62991,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,129608.716834,132839.525748,1,100,8,8,2.97,2,2,0,0,0,1,24,1,5,1,5
+62992,7,2,2,9,NA,1,1,2,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18668.602894,18957.049554,1,101,4,4,0.84,3,3,0,1,0,1,42,1,4,1,4
+62993,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,74501.766669,80316.820348,1,90,14,14,3.93,3,3,1,0,0,1,35,1,2,1,5
+62994,7,2,2,49,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,3,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,18255.735511,18352.270791,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+62995,7,2,1,18,NA,4,4,2,18,220,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11386.695644,11908.648036,2,99,6,6,1.18,5,5,0,3,0,2,38,1,2,5,NA
+62996,7,2,2,9,NA,2,2,2,9,114,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13257.060167,14867.908577,2,90,8,8,1.72,5,5,1,2,0,1,20,2,1,1,2
+62997,7,2,1,60,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,29552.786683,29829.95179,2,96,3,3,0.95,2,2,0,0,2,2,62,1,4,1,4
+62998,7,2,1,19,NA,5,6,1,20,NA,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6558.308393,6849.983523,2,92,99,99,NA,4,1,0,0,0,1,19,1,4,NA,NA
+62999,7,2,2,9,NA,5,7,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6137.256046,6434.751859,1,103,15,15,3.7,5,5,0,2,1,1,55,1,5,1,5
+63000,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11862.436765,12619.924744,1,90,2,2,0.45,1,1,0,0,1,2,60,1,5,3,NA
+63001,7,2,1,5,NA,4,4,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9624.976734,10219.132066,2,96,3,3,0.38,5,5,1,2,0,2,30,1,3,5,NA
+63002,7,2,2,54,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,21901.662873,22693.823819,2,93,8,8,2.49,3,3,0,0,0,1,52,2,2,1,4
+63003,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,151649.038926,151259.483772,1,101,13,13,NA,2,2,0,0,0,1,40,1,2,1,3
+63004,7,2,2,38,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,3,1,2,1,1,2,2,1,2,2,NA,13047.751375,13287.186355,2,92,15,15,5,3,3,1,0,0,1,39,2,5,1,5
+63005,7,2,1,19,NA,1,1,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,32326.52031,34735.818434,2,94,5,5,0.65,6,6,0,3,0,1,44,2,1,1,1
+63006,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,123771.419917,127443.198085,2,98,10,10,3.78,3,3,0,0,0,1,53,1,3,1,4
+63007,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,47098.572584,49762.054694,1,95,7,7,2.38,2,2,0,0,2,1,80,1,3,1,4
+63008,7,2,1,38,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15051.024912,15121.525249,3,90,15,15,5,5,5,1,0,1,1,38,2,3,1,4
+63009,7,2,1,6,NA,4,4,2,7,84,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8483.005475,8584.70346,1,96,10,10,2.95,4,4,0,1,0,2,34,2,3,1,5
+63010,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,12535.973802,12972.182488,2,100,7,7,2.37,3,3,0,1,1,2,45,1,5,1,NA
+63011,7,2,2,16,NA,3,3,2,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,122091.956024,127137.394167,2,91,15,15,5,3,3,0,1,0,1,52,1,5,1,5
+63012,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,10,10,4.76,2,2,0,0,2,2,64,1,4,1,5
+63013,7,2,1,65,NA,5,6,1,NA,NA,1,1,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11138.867102,11750.838884,1,102,14,14,5,2,2,0,0,2,1,65,2,5,1,5
+63014,7,2,2,11,NA,4,4,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7814.742747,7978.458393,1,98,8,8,2.97,2,2,0,1,0,2,40,1,4,3,NA
+63015,7,2,1,65,NA,5,7,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,117343.481374,118848.614941,1,91,7,7,2.64,2,2,0,0,1,2,58,1,5,1,4
+63016,7,2,2,14,NA,2,2,2,14,170,NA,NA,2,2,3,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12680.621719,14355.413798,2,90,3,3,0.46,5,5,0,2,2,1,75,2,1,1,2
+63017,7,2,1,71,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,12511.526803,13352.932153,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+63018,7,2,1,10,NA,3,3,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,27154.222487,28089.451048,2,101,4,4,0.73,5,5,1,2,0,1,40,1,5,1,5
+63019,7,2,2,65,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,3,NA,2,2,2,2,2,2,1,2,2,2,10235.0654,10662.230581,2,93,3,3,0.87,2,2,0,0,1,2,65,2,4,3,NA
+63020,7,2,1,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13533.281742,15943.839793,1,100,6,6,1.11,5,5,0,2,1,1,38,2,2,1,1
+63021,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,3,5,NA,1,2,1,1,2,1,NA,NA,NA,NA,19524.115198,21455.7569,1,93,7,7,2.38,2,2,0,0,0,2,46,2,3,5,NA
+63022,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,92467.370896,92229.84126,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+63023,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,124820.027137,126759.886775,1,91,6,3,0.92,2,1,0,0,0,2,24,1,4,6,NA
+63024,7,2,2,78,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,12692.401451,13134.05328,2,95,12,12,NA,3,3,0,0,2,1,65,1,4,1,4
+63025,7,2,2,79,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,14226.966896,14722.016314,2,100,5,5,0.95,4,4,0,0,1,2,53,1,3,5,NA
+63026,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,53830.599426,56471.452044,1,99,15,15,5,3,3,1,0,0,2,31,1,5,1,5
+63027,7,2,1,15,NA,1,1,1,15,181,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18242.832494,18521.680571,1,102,6,6,1.03,6,6,0,4,0,1,34,2,2,1,1
+63028,7,2,2,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14414.529053,14932.182215,2,96,8,8,1.33,7,7,2,1,1,1,62,2,1,1,1
+63029,7,2,2,2,NA,5,7,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20272.436775,20911.100412,2,94,77,77,NA,4,4,2,0,0,2,23,1,2,6,NA
+63030,7,2,1,21,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,42165.369652,45330.590147,2,102,3,3,0.45,4,4,2,0,0,1,21,2,2,6,NA
+63031,7,2,2,77,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,15183.616455,20169.750633,1,90,6,6,1.78,2,2,0,0,1,2,77,1,1,2,NA
+63032,7,2,1,14,NA,4,4,2,14,172,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11085.075029,11001.247749,1,93,1,1,0.02,5,5,0,4,0,2,36,NA,NA,5,NA
+63033,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,153755.392794,158780.137655,1,91,15,15,5,4,4,0,1,0,1,45,1,5,1,5
+63034,7,2,2,5,NA,3,3,2,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,92,4,4,0.61,5,5,1,2,0,1,34,1,3,6,NA
+63035,7,2,1,7,NA,5,7,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8017.552697,8398.795399,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+63036,7,2,2,74,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,12895.665603,13811.504793,1,93,2,2,0.55,1,1,0,0,1,2,74,2,1,3,NA
+63037,7,2,2,41,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,4,2,2,2,2,1,2,2,2,2,2,2,34501.569761,34681.492512,1,93,5,5,1.36,2,2,0,0,0,2,41,2,1,4,NA
+63038,7,2,2,2,NA,5,7,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7871.443574,8074.618887,1,96,6,6,1.49,3,3,1,0,0,2,24,1,4,5,NA
+63039,7,2,2,2,NA,1,1,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10582.75863,3,92,6,6,1,6,6,1,1,0,1,42,2,1,1,4
+63040,7,2,2,54,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,2,1,NA,1,2,1,NA,NA,NA,1,2,1,3,15268.129241,15348.866231,2,96,NA,NA,NA,4,4,0,0,1,1,67,2,3,1,2
+63041,7,2,1,35,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14376.416432,15539.834252,2,103,15,15,5,4,4,0,3,0,1,35,1,5,5,NA
+63042,7,2,2,12,NA,4,4,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13841.239638,13875.328502,1,96,15,15,4.52,6,6,0,4,0,1,46,1,4,1,4
+63043,7,2,2,20,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,11609.8069,12151.984044,2,103,77,77,NA,5,5,0,2,0,2,39,2,5,1,5
+63044,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,79316.940936,94360.361124,1,93,10,10,5,1,1,0,0,0,1,35,1,5,5,NA
+63045,7,2,2,53,NA,1,1,1,NA,NA,2,NA,2,2,8,NA,2,4,NA,1,2,2,2,2,2,1,2,2,1,18295.488967,18390.898385,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+63046,7,2,1,1,NA,3,3,2,NA,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39241.111643,42207.80453,1,99,6,6,1.73,3,3,1,1,0,2,42,1,5,3,NA
+63047,7,2,2,6,NA,3,3,2,6,81,NA,NA,2,1,2,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14287.66096,14320.169639,2,97,6,6,1.16,4,4,1,1,0,1,27,2,4,1,3
+63048,7,2,1,29,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15875.103787,17176.519619,1,93,10,10,4.76,2,2,0,0,0,1,29,2,5,1,5
+63049,7,2,1,19,NA,2,2,1,19,233,2,NA,2,1,5,15,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,30061.88611,30025.949584,1,92,6,6,0.93,5,5,0,2,0,1,47,2,1,1,1
+63050,7,2,1,24,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,12364.328404,12075.509308,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+63051,7,2,1,30,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22403.911395,23911.071969,2,96,14,14,5,2,2,0,0,0,1,30,2,5,1,5
+63052,7,2,2,37,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,3,2,1,2,2,1,2,2,NA,NA,NA,NA,19741.154437,19781.46423,1,100,7,7,2.51,2,2,0,1,0,2,37,2,5,3,NA
+63053,7,2,2,8,NA,1,1,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12789.411811,13121.359297,1,102,3,3,0.54,3,3,0,2,0,2,42,2,2,3,NA
+63054,7,2,2,11,NA,1,1,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14469.11917,14745.425968,2,92,15,15,3.37,7,7,0,4,0,1,42,2,3,1,1
+63055,7,2,2,2,NA,3,3,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25257.21318,27876.153487,1,98,2,2,0.31,3,3,1,0,0,1,45,NA,NA,1,NA
+63056,7,2,1,21,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15196.92397,16091.373745,2,101,7,3,1.1,3,1,0,0,0,1,21,2,4,5,NA
+63057,7,2,1,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,43892.276067,44574.41704,1,98,6,6,2.6,2,1,0,0,0,1,29,1,4,5,NA
+63058,7,2,2,13,NA,5,6,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6135.470395,6372.721366,1,91,15,15,5,6,6,0,2,2,1,50,2,5,1,5
+63059,7,2,1,7,NA,4,4,1,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10469.725162,11059.049224,1,100,6,6,1.13,4,4,0,3,0,2,32,1,3,5,NA
+63060,7,2,2,7,NA,1,1,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13064.573334,13201.924867,1,103,5,5,0.71,6,6,2,2,0,2,31,2,2,1,2
+63061,7,2,2,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,1,1,2,2,1,2,2,NA,NA,NA,NA,22326.231285,22166.696692,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+63062,7,2,1,4,NA,4,4,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10690.995725,11500.031681,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+63063,7,2,2,7,NA,3,3,1,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,60557.637514,61692.695301,3,92,6,6,1.17,4,4,0,2,0,2,30,1,2,1,4
+63064,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,7233.371983,2,95,3,3,1.07,1,1,0,0,1,1,61,1,1,4,NA
+63065,7,2,1,11,NA,1,1,1,11,143,NA,NA,2,2,3,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,13822.148996,14860.201344,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+63066,7,2,1,3,NA,3,3,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,61250.535365,69105.159981,1,90,8,8,1.67,5,5,2,1,0,2,28,1,4,1,5
+63067,7,2,2,0,5,5,6,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6956.259272,7247.645977,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+63068,7,2,1,9,NA,3,3,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,26240.82746,27691.552263,1,94,4,4,0.56,5,5,1,2,0,1,34,1,2,3,NA
+63069,7,2,1,20,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15533.829525,2,101,6,2,0.46,2,1,0,0,0,1,20,2,4,5,NA
+63070,7,2,2,70,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,1,5,NA,2,2,2,1,2,2,2,2,2,NA,21122.17432,23417.039872,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+63071,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,20450.752436,20266.431761,1,98,6,6,1.72,2,2,0,0,0,2,53,1,5,2,NA
+63072,7,2,2,12,NA,3,3,2,12,152,NA,NA,2,1,1,7,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,31756.649984,32677.072621,2,97,5,5,0.8,5,5,1,2,0,1,46,2,4,1,2
+63073,7,2,1,16,NA,3,3,1,16,194,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,25367.590269,25034.201904,1,94,6,6,1.26,5,5,0,2,0,2,38,1,4,1,NA
+63074,7,2,1,3,NA,4,4,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9698.338509,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+63075,7,2,2,12,NA,2,2,1,12,152,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20502.928313,21114.699878,2,98,12,12,NA,3,3,0,1,0,2,34,1,4,1,3
+63076,7,2,1,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,21168.33083,23066.473787,2,99,2,2,0.4,2,2,0,0,0,2,52,1,3,1,3
+63077,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,86986.68246,92645.868047,2,94,15,15,4.59,4,4,1,1,0,2,37,1,5,1,5
+63078,7,2,1,6,NA,1,1,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8828.580268,8822.70874,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+63079,7,2,1,17,NA,5,6,1,17,212,2,NA,2,1,5,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11171.449402,11936.033174,1,92,14,14,3.47,4,4,0,2,0,2,53,2,4,1,4
+63080,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,NA,NA,NA,NA,30505.56355,30746.687156,1,97,3,3,0.93,2,2,0,1,0,2,34,1,4,5,NA
+63081,7,2,1,62,NA,5,7,2,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19328.482066,19868.466665,3,90,7,7,2.51,2,2,0,0,2,1,62,2,3,1,4
+63082,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,175997.804296,180603.118894,1,101,4,2,0.82,2,1,0,0,1,1,63,1,2,6,NA
+63083,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,69601.920609,70345.22994,1,98,7,7,3.58,1,1,0,0,1,2,71,1,4,2,NA
+63084,7,2,1,4,NA,3,3,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,45649.909879,51503.94697,1,99,15,15,5,5,5,3,0,0,2,34,1,5,1,5
+63085,7,2,2,4,NA,4,4,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8267.490849,9012.744386,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+63086,7,2,2,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,94,15,15,5,2,2,0,0,2,1,75,1,4,1,3
+63087,7,2,1,3,NA,3,3,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,81508.607355,87670.792766,1,94,14,6,1.65,3,2,1,0,0,1,26,1,4,6,NA
+63088,7,2,1,44,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11113.602843,11073.454175,3,90,10,10,2.41,5,5,1,2,0,1,44,2,4,1,5
+63089,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,80232.348161,83992.832552,1,99,15,15,4.47,4,4,0,2,0,2,43,1,5,1,5
+63090,7,2,2,15,NA,3,3,2,15,186,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41202.729804,43105.611516,2,91,6,6,1.26,5,5,0,1,2,2,80,1,4,2,NA
+63091,7,2,1,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,NA,55996.810038,59271.38884,2,93,15,8,4.48,2,1,0,0,2,2,63,2,4,6,NA
+63092,7,2,2,17,NA,4,4,2,18,216,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12581.940435,12644.353117,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+63093,7,2,2,10,NA,3,3,1,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,1,2,1,2,2,1,21800.021754,23152.363882,3,91,7,1,0,7,1,0,4,0,1,40,1,4,1,3
+63094,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,27167.433399,29371.776665,1,94,7,7,2.58,2,2,0,0,2,2,71,1,5,1,4
+63095,7,2,2,4,NA,4,4,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11664.396755,2,100,9,9,2.46,4,4,1,1,1,2,59,1,3,1,3
+63096,7,2,1,8,NA,2,2,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13672.897697,14159.650726,2,100,14,14,3.58,4,4,0,2,0,2,40,NA,NA,1,4
+63097,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,71070.181743,74923.26272,1,95,7,7,2.45,2,2,0,0,2,2,70,1,3,1,3
+63098,7,2,2,10,NA,3,3,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60197.256541,63931.531988,2,94,14,14,2.63,6,6,1,3,0,1,39,1,4,1,4
+63099,7,2,2,3,NA,2,2,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15550.931019,16748.422137,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+63100,7,2,2,13,NA,1,1,1,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22424.988432,22856.096824,1,94,5,5,0.87,4,4,0,2,0,2,41,2,4,1,1
+63101,7,2,1,6,NA,4,4,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11060.738342,11193.339318,2,96,6,6,1.48,4,4,1,1,0,2,25,1,4,5,NA
+63102,7,2,2,19,NA,1,1,1,19,238,2,NA,2,2,1,12,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,16427.640886,17012.806816,1,102,5,5,0.98,4,4,1,1,0,2,42,2,2,6,NA
+63103,7,2,1,46,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12820.674452,13169.378628,3,90,15,15,5,3,3,0,0,0,1,46,2,3,1,3
+63104,7,2,2,6,NA,4,4,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8146.767632,8699.382682,2,100,15,15,4.97,5,5,0,2,1,2,42,1,5,1,5
+63105,7,1,1,14,NA,5,6,NA,NA,NA,NA,NA,2,1,4,9,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8915.81491,0,1,95,3,3,0.43,4,4,0,1,2,1,65,2,5,1,3
+63106,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32838.149884,1,95,3,3,0.95,1,1,0,0,0,1,50,1,2,3,NA
+63107,7,2,2,66,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,9716.805546,12994.252166,2,90,3,3,1.1,1,1,0,0,1,2,66,2,1,1,NA
+63108,7,2,1,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,26847.643051,28699.104328,1,98,4,4,0.66,4,4,2,0,0,2,22,1,4,6,NA
+63109,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,105873.555835,111215.69511,1,101,14,14,3.9,4,4,0,2,0,2,41,1,2,1,2
+63110,7,2,1,36,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22439.245437,22882.446062,2,96,14,14,3.58,4,4,2,0,0,1,36,1,4,1,5
+63111,7,2,1,8,NA,1,1,1,8,103,NA,NA,2,7,77,1,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,11399.23838,11295.714433,1,103,8,8,1.85,5,5,2,1,0,2,25,2,2,1,2
+63112,7,2,2,66,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,17221.648537,17753.661813,1,95,6,6,1.7,2,2,0,0,2,2,66,2,1,1,3
+63113,7,2,2,3,NA,1,1,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17543.203297,19368.16722,1,95,8,8,2.24,4,4,2,0,0,2,29,1,3,1,4
+63114,7,2,1,15,NA,5,6,1,15,189,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11657.164593,12175.606972,1,92,9,9,2.88,3,3,0,2,0,1,50,2,5,3,NA
+63115,7,2,1,4,NA,3,3,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,32621.433667,36804.729607,2,101,4,4,0.73,5,5,1,2,0,1,40,1,5,1,5
+63116,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,24291.664633,24483.672027,1,99,8,8,4.13,1,1,0,0,0,2,34,1,5,5,NA
+63117,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,6,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15533.829525,2,101,8,5,2.2,2,1,0,0,0,1,24,2,4,6,NA
+63118,7,2,1,16,NA,1,1,1,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30061.88611,30025.949584,1,92,7,7,1.48,5,5,0,1,0,1,42,1,5,1,4
+63119,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,84933.772167,92023.102027,2,99,15,15,5,2,2,0,0,0,2,37,1,5,1,5
+63120,7,2,2,4,NA,3,3,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,55336.504558,58807.19381,2,102,14,14,3.44,5,5,1,2,0,2,34,1,4,6,NA
+63121,7,2,1,9,NA,1,1,1,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15233.096858,16234.004804,1,94,15,15,4.37,7,7,0,4,1,1,58,1,4,1,5
+63122,7,2,2,3,NA,2,2,1,3,44,NA,NA,2,2,2,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10301.516763,10950.95816,2,93,6,6,0.93,5,5,1,2,0,1,40,2,4,1,4
+63123,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18753.573091,18904.865313,2,99,2,2,0.77,1,1,0,0,0,1,55,1,3,5,NA
+63124,7,2,2,54,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15460.72194,15832.955199,1,102,9,9,2.39,5,5,0,1,1,1,55,2,5,1,5
+63125,7,2,2,46,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22109.546782,21887.540337,1,93,7,7,1.79,4,4,0,2,0,1,53,2,4,1,4
+63126,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,111448.08155,119470.923607,1,90,15,15,5,4,4,0,2,0,1,44,1,5,1,5
+63127,7,2,1,28,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14391.77847,15024.211748,1,92,14,14,3.3,4,4,2,0,0,1,28,1,4,1,4
+63128,7,2,2,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,3,1,2,2,1,2,2,1,2,2,1,38589.695298,40356.151096,2,96,6,5,1.84,2,1,0,0,0,2,26,1,2,5,NA
+63129,7,2,1,14,NA,1,1,1,14,172,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22768.423624,23297.239555,2,98,6,6,1.4,3,3,0,1,0,1,56,1,2,1,2
+63130,7,2,2,16,NA,5,6,2,16,196,NA,NA,2,2,1,8,NA,NA,NA,1,2,1,NA,NA,NA,1,2,1,NA,10767.566937,12084.516227,2,91,99,1,0,7,3,0,4,0,1,36,2,9,1,2
+63131,7,2,2,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8480.466509,9125.076563,1,100,13,13,NA,5,5,2,0,0,2,54,1,4,5,NA
+63132,7,2,2,80,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,4,2,NA,1,2,1,1,2,2,1,2,1,NA,18698.205673,19635.336647,1,97,15,15,4.81,5,5,0,1,1,1,51,2,5,1,5
+63133,7,2,1,24,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,35669.2076,40318.090187,2,94,8,8,2.33,4,4,2,0,0,1,24,1,2,6,NA
+63134,7,2,2,66,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,3,2,NA,2,2,2,2,2,2,2,2,2,2,11851.128358,12345.740673,2,93,4,4,0.69,4,4,0,1,1,2,66,2,3,2,NA
+63135,7,2,1,15,NA,4,4,2,15,190,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,7,7,1.74,4,4,0,2,0,2,47,1,5,4,NA
+63136,7,2,1,18,NA,3,3,2,18,224,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,66448.116716,70579.976727,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+63137,7,2,1,49,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19898.840409,19922.416835,2,94,15,15,5,2,2,0,0,0,1,49,2,5,1,5
+63138,7,2,1,15,NA,3,3,2,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,76583.482739,78051.61978,3,91,4,4,0.92,3,3,0,1,0,2,53,1,4,1,4
+63139,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,24118.129224,23465.858863,2,96,3,3,0.59,3,3,1,0,0,2,25,1,4,1,NA
+63140,7,2,2,9,NA,1,1,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12307.832776,12437.228482,2,93,77,77,NA,7,7,3,1,0,2,43,2,1,1,9
+63141,7,2,2,34,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,27303.803575,30038.060995,1,96,10,10,2.95,4,4,0,1,0,2,34,2,3,1,5
+63142,7,2,1,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19064.767778,19872.734197,2,93,10,10,3.93,3,3,0,0,2,1,54,1,5,1,5
+63143,7,2,1,70,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,10498.47031,11456.643725,3,90,4,4,1.22,2,2,0,0,2,2,69,2,4,1,1
+63144,7,2,1,1,21,2,2,2,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10803.555682,11145.535856,2,94,7,7,1.34,5,5,2,1,0,1,32,2,1,1,NA
+63145,7,2,1,50,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,15044.515884,15040.577827,3,90,5,5,0.87,4,4,0,0,0,2,43,2,3,5,NA
+63146,7,2,2,6,NA,4,4,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8362.256577,9003.967662,2,100,3,3,0.38,5,5,2,1,0,2,28,1,2,5,NA
+63147,7,2,1,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,31010.243793,32025.638239,3,92,6,6,1.41,4,3,0,1,0,1,41,1,4,1,4
+63148,7,2,1,40,NA,1,1,2,NA,NA,2,NA,2,2,77,NA,3,1,NA,2,2,2,1,2,2,2,2,2,2,31640.296506,31176.023929,2,94,6,4,1.38,2,1,0,0,0,1,40,2,3,1,NA
+63149,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,145772.192378,153127.526275,1,95,9,9,2.66,4,4,0,2,0,1,45,1,3,1,3
+63150,7,2,1,19,NA,3,3,2,19,229,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,93665.036597,95017.313859,1,91,15,15,5,4,4,0,1,0,1,45,1,5,1,5
+63151,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,17212.153036,21760.360036,3,90,7,6,2.42,2,1,0,0,0,1,44,1,2,6,NA
+63152,7,2,2,62,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15277.358397,15812.991914,2,96,5,1,0,2,1,0,0,1,1,46,2,3,3,NA
+63153,7,2,1,2,NA,2,2,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13002.944731,13931.716845,2,91,2,2,0.42,3,3,1,1,0,2,27,1,3,5,NA
+63154,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,51543.062078,51154.050295,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+63155,7,2,1,18,NA,3,3,2,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,68148.957861,67253.324127,1,93,15,15,3.92,5,5,0,1,0,2,54,1,5,1,5
+63156,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,49260.413155,50771.674072,2,100,8,8,3.4,2,2,0,0,2,1,69,1,4,1,4
+63157,7,2,1,13,NA,3,3,2,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,32477.57544,32050.74555,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+63158,7,2,2,0,5,3,3,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10263.789813,10611.184403,1,94,8,8,1.39,7,7,2,0,1,2,52,1,5,2,NA
+63159,7,2,1,48,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,37724.155022,39553.56554,2,102,4,4,1.16,2,2,0,0,0,1,48,1,4,1,4
+63160,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,130577.02995,132175.980892,1,98,5,5,1.81,1,1,0,0,1,2,61,1,3,2,NA
+63161,7,2,2,37,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,4,3,2,2,2,2,1,2,2,2,2,2,2,39450.135734,38385.688196,2,93,9,9,1.94,6,6,0,3,0,2,37,NA,NA,3,NA
+63162,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19000.571418,18807.33617,1,96,8,8,4.48,1,1,0,0,0,2,58,1,4,3,NA
+63163,7,2,1,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,146181.198007,148606.927767,2,91,15,2,0.85,7,1,0,0,1,1,49,NA,NA,5,NA
+63164,7,2,1,54,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,27343.459193,27021.369249,1,90,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+63165,7,1,2,13,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16537.460749,0,2,100,1,1,0.04,4,4,0,2,0,1,34,NA,NA,6,NA
+63166,7,2,2,17,NA,4,4,2,17,205,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10848.628906,11198.221038,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+63167,7,2,1,37,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12327.761744,13424.968474,3,90,12,12,NA,5,5,1,2,0,1,37,2,5,1,5
+63168,7,2,1,10,NA,5,6,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8580.826574,9087.233306,1,92,14,14,2.42,6,6,1,3,0,1,30,1,4,6,NA
+63169,7,2,2,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,53955.606082,55487.991675,2,100,6,6,1.7,3,3,0,1,0,2,33,1,4,6,NA
+63170,7,2,2,53,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,22224.73066,22340.630739,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+63171,7,2,1,20,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,9162.933592,9412.398479,1,103,5,5,0.65,6,6,0,0,1,2,26,2,4,5,NA
+63172,7,2,2,68,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,2,NA,2,2,2,1,2,2,2,2,2,2,13057.178942,15139.165763,1,102,6,6,1.18,5,5,0,2,1,2,42,2,2,2,NA
+63173,7,2,1,18,NA,5,6,2,18,217,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,5096.872559,5447.231524,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+63174,7,2,2,47,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,29650.79971,34013.933936,2,90,14,14,5,1,1,0,0,0,2,47,1,5,5,NA
+63175,7,2,1,15,NA,3,3,1,15,184,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,31348.486193,31739.246143,1,98,5,5,0.74,5,5,0,3,0,1,35,1,2,6,NA
+63176,7,2,1,19,NA,4,4,2,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13513.882801,1,96,4,4,0.65,4,4,0,0,0,1,19,1,4,NA,NA
+63177,7,2,1,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,8355.583627,8420.94363,2,93,10,10,3.93,3,3,0,0,2,1,54,1,5,1,5
+63178,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,30543.064147,31583.211468,2,101,8,8,2.81,3,3,0,1,0,1,35,1,1,1,2
+63179,7,2,1,8,NA,2,2,1,8,100,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,13898.598114,14013.214919,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+63180,7,2,1,50,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10857.294346,10859.552755,1,91,15,15,5,6,6,0,2,2,1,50,2,5,1,5
+63181,7,2,2,67,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,10308.451947,2,90,8,8,2.01,4,4,0,0,1,2,67,2,4,2,NA
+63182,7,2,2,14,NA,4,4,2,14,171,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10671.280357,10591.403308,1,99,15,8,2.7,4,3,0,2,0,1,49,1,4,6,NA
+63183,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,20135.631696,26575.299535,1,96,4,4,1.12,2,2,0,1,0,2,44,1,2,5,NA
+63184,7,2,1,3,NA,3,3,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,86817.367332,95073.481824,2,91,15,15,5,4,4,2,0,0,1,34,1,5,1,5
+63185,7,2,1,3,NA,4,4,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,11070.778245,1,100,3,3,0.73,3,3,2,0,0,2,39,1,3,5,NA
+63186,7,2,1,11,NA,5,7,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11611.58308,11813.013945,1,98,14,14,3.16,6,6,2,2,0,1,39,1,5,1,5
+63187,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,26773.686592,29962.63786,2,98,6,6,1.7,2,2,0,0,2,2,80,1,2,1,2
+63188,7,2,2,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,12782.405209,13353.125929,1,100,3,3,1.17,1,1,0,0,1,2,60,1,4,4,NA
+63189,7,2,2,0,5,4,4,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4964.196196,5018.399368,1,93,1,1,0.16,3,3,1,1,0,2,39,1,3,5,NA
+63190,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,104765.204996,105038.68379,2,92,15,15,5,5,5,0,3,0,2,46,1,5,1,5
+63191,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,70667.985128,72578.248908,2,92,15,7,3.67,2,1,0,0,0,2,30,1,5,1,NA
+63192,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,41229.806244,46206.205733,2,103,6,6,1.82,2,2,0,0,2,1,70,1,2,5,NA
+63193,7,2,2,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,26389.420306,26882.397721,2,98,3,3,0.96,1,1,0,0,0,2,41,1,4,3,NA
+63194,7,2,2,56,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,2,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12649.084278,13204.362319,3,90,4,4,0.92,3,3,0,0,1,2,56,2,2,1,2
+63195,7,2,1,59,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,3,1,NA,2,2,2,2,2,2,1,2,1,2,24211.824535,24594.266281,2,93,8,8,2.57,3,3,0,0,1,1,59,2,3,1,3
+63196,7,2,1,39,NA,4,4,1,NA,NA,2,NA,2,1,4,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18838.303827,18900.579975,2,93,12,12,NA,4,4,1,1,0,2,27,2,4,1,4
+63197,7,2,2,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,18070.666316,17918.945427,1,99,7,7,2.38,2,2,0,0,0,2,27,1,5,1,5
+63198,7,2,1,80,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,7199.330978,7340.349591,1,93,4,4,1.56,1,1,0,0,1,1,80,2,5,2,NA
+63199,7,2,1,12,NA,3,3,1,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,27156.800586,27669.661454,1,92,5,5,1.05,3,3,1,1,0,2,35,1,4,5,NA
+63200,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,164722.990078,163821.547042,1,90,15,15,5,3,3,0,0,0,1,59,1,5,1,5
+63201,7,2,1,7,NA,1,1,1,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12665.770043,13116.669484,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+63202,7,2,1,0,3,5,7,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6371.499593,6469.293142,2,95,12,14,3.93,4,3,1,0,0,1,35,1,5,1,4
+63203,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35334.703093,40990.264786,1,101,3,3,0.9,1,1,0,0,1,2,80,1,2,2,NA
+63204,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,26757.554986,26943.613048,2,97,7,7,1.89,3,3,0,0,0,1,50,1,2,1,2
+63205,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,28216.191929,28480.821588,1,101,7,7,1.83,3,3,0,0,2,1,67,1,1,1,2
+63206,7,2,1,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,14204.262514,14001.792163,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+63207,7,2,2,8,NA,3,3,2,8,107,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,52414.628675,53163.961325,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+63208,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105261.096488,116369.237268,1,97,15,15,5,3,3,1,0,0,2,31,1,5,1,5
+63209,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,65033.706797,70202.805273,1,102,8,8,2.42,4,4,0,2,0,2,34,1,4,1,3
+63210,7,2,1,10,NA,4,4,2,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7250.311091,8640.316317,2,99,2,2,0.2,7,7,1,2,1,1,63,1,1,2,NA
+63211,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,93110.905446,112207.13845,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+63212,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,11764.405491,12074.424659,1,97,4,4,1.34,1,1,0,0,1,1,62,1,5,5,NA
+63213,7,2,1,43,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13535.287828,13486.390654,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+63214,7,2,1,37,NA,1,1,1,NA,NA,2,NA,2,1,5,NA,1,6,NA,2,2,2,1,2,2,1,2,2,2,37715.365512,38126.275234,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+63215,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,11862.334174,1,96,15,15,5,3,3,0,0,1,2,62,1,4,3,NA
+63216,7,2,1,54,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,29042.244317,28700.143679,1,90,12,12,NA,4,4,0,0,0,1,54,2,4,1,2
+63217,7,1,2,8,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16195.492817,0,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+63218,7,2,1,35,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,34887.439952,35451.076414,2,94,3,3,0.82,2,2,0,0,0,1,35,2,1,1,2
+63219,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,20698.946375,20973.485449,3,91,15,15,5,1,1,0,0,0,1,39,2,5,5,NA
+63220,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,22634.531479,23281.741513,1,103,9,2,0.81,3,1,0,0,0,1,27,1,5,5,NA
+63221,7,2,2,67,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,16352.915834,18960.412859,3,92,8,8,2.97,2,2,0,0,1,1,49,1,2,3,NA
+63222,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20360.306379,19986.123493,2,99,15,15,5,2,2,0,0,0,2,51,1,5,1,5
+63223,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,174307.982009,1,95,7,7,3.21,1,1,0,0,0,1,59,1,4,3,NA
+63224,7,2,2,31,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,29102.738194,28317.485179,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+63225,7,2,2,63,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,11862.334174,1,96,12,12,NA,5,5,2,0,1,2,63,2,5,3,NA
+63226,7,2,2,2,NA,2,2,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9267.834226,9392.475614,2,93,77,77,NA,4,4,1,1,0,2,33,2,4,1,4
+63227,7,2,2,40,NA,1,1,1,NA,NA,2,NA,99,NA,NA,NA,3,1,2,2,2,2,2,2,2,NA,NA,NA,NA,25713.328161,31117.9447,1,103,5,5,0.74,5,5,1,1,0,2,40,99,3,1,1
+63228,7,2,1,27,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,44074.735764,53571.145963,2,91,4,4,0.84,3,3,1,0,0,2,21,1,4,1,2
+63229,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,19440.793325,19514.660132,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+63230,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14455.875684,16658.375886,1,93,9,9,5,1,1,0,0,0,1,48,2,5,5,NA
+63231,7,2,2,42,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,4,5,2,2,2,2,2,2,2,2,2,1,2,29141.220673,30195.229087,2,99,99,1,0,4,1,0,0,0,2,42,2,4,5,NA
+63232,7,2,2,14,NA,4,4,1,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10128.026158,10319.790822,2,103,7,7,1.48,5,5,0,1,1,2,80,1,4,3,NA
+63233,7,2,1,32,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,23284.536512,27522.885754,1,97,5,5,1.04,4,4,1,1,0,1,32,1,3,6,NA
+63234,7,2,2,35,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,20310.041341,20837.576644,1,92,3,3,0.9,1,1,0,0,0,2,35,1,5,5,NA
+63235,7,2,2,63,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,16352.915834,17093.606152,3,92,3,3,0.95,2,2,0,0,2,2,63,1,4,1,1
+63236,7,2,1,80,NA,2,2,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,14200.083364,14896.175407,2,98,3,3,1.1,1,1,0,0,1,1,80,1,3,3,NA
+63237,7,2,1,2,NA,3,3,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49276.9767,53002.397569,2,98,15,15,5,3,3,1,0,0,1,26,1,4,1,4
+63238,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,2,2,NA,2,2,2,1,2,2,NA,NA,NA,NA,18006.276697,19962.608629,2,93,3,3,0.66,2,2,0,0,1,2,80,2,2,2,NA
+63239,7,2,2,6,NA,4,4,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9399.281543,9563.245188,2,96,4,4,0.57,5,5,0,3,0,2,26,1,2,5,NA
+63240,7,2,1,5,NA,2,2,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17306.08847,2,98,6,6,1.07,5,5,3,0,0,2,24,1,3,1,3
+63241,7,1,1,9,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37831.648033,0,1,99,77,77,NA,4,4,0,2,0,2,45,1,3,1,NA
+63242,7,2,1,59,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,25221.349696,24851.265269,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+63243,7,2,1,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13058.677088,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+63244,7,2,1,4,NA,4,4,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9228.425814,10175.8485,1,96,8,8,2,4,4,1,2,0,2,40,1,4,5,NA
+63245,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,99275.150567,100291.087043,2,95,12,5,1.79,2,1,0,0,0,2,40,1,4,6,NA
+63246,7,2,1,13,NA,4,4,2,13,167,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13416.172328,13513.882801,1,96,15,15,5,4,4,0,1,0,1,42,2,4,1,4
+63247,7,2,1,19,NA,5,7,2,19,239,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13251.602554,13277.802057,1,96,15,15,5,3,3,0,0,0,2,40,1,5,1,4
+63248,7,2,1,75,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,67447.112074,71391.281096,1,101,15,15,5,3,3,0,0,2,1,75,1,2,1,2
+63249,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,102183.259724,105037.978798,2,90,15,8,4.59,2,1,0,0,2,1,67,1,5,6,NA
+63250,7,2,1,17,NA,3,3,2,17,212,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,88198.948426,91036.751291,1,101,9,9,2.6,4,4,0,1,2,2,63,1,4,1,4
+63251,7,2,2,80,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,1,2,NA,1,2,1,1,2,2,NA,NA,NA,NA,15288.064726,15818.288224,3,91,3,3,0.54,3,3,0,0,1,1,57,2,4,1,3
+63252,7,2,1,7,NA,4,4,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11709.3276,12266.117914,1,100,8,8,3.3,2,2,0,1,0,2,44,1,4,3,NA
+63253,7,2,2,13,NA,4,4,2,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11160.089502,11615.801738,2,99,99,99,NA,5,5,0,2,0,2,20,1,3,6,NA
+63254,7,2,2,72,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,2,1,NA,2,2,2,1,2,2,1,2,2,NA,23271.708938,25108.705116,1,102,7,7,1.7,4,4,0,0,2,1,44,1,4,4,NA
+63255,7,1,1,64,NA,4,4,NA,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,6239.974623,0,1,96,77,77,NA,2,2,0,0,2,1,64,1,5,1,5
+63256,7,2,1,9,NA,4,4,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10122.702296,10196.10824,2,93,9,9,2.86,4,4,1,1,0,1,30,1,4,6,NA
+63257,7,2,2,6,NA,5,6,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6834.715094,7332.504615,2,90,5,5,1.08,3,3,0,1,0,2,29,2,4,1,5
+63258,7,2,1,1,12,3,3,2,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46703.291366,54723.864587,1,93,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+63259,7,2,2,7,NA,1,1,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10118.363218,10311.586628,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+63260,7,2,2,42,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,36276.046363,36418.32035,2,91,99,99,NA,3,3,0,0,0,1,40,NA,NA,1,4
+63261,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20434.221508,20198.461156,1,102,15,15,5,3,3,0,1,0,1,41,1,5,1,5
+63262,7,2,2,76,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,2,2,NA,2,2,2,1,2,2,2,2,2,NA,19352.965911,23161.019266,1,103,6,6,0.97,6,6,0,3,1,2,50,2,1,1,1
+63263,7,2,1,53,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,2,2,2,2,2,2,1,2,2,1,24211.824535,24594.266281,2,93,6,6,1.39,4,4,0,0,0,1,53,2,3,1,3
+63264,7,2,1,32,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,1,1,2,2,1,2,1,NA,20963.414192,21636.182301,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+63265,7,2,2,28,NA,4,4,2,NA,NA,2,NA,2,2,1,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,28823.434754,27922.831827,1,97,6,1,0,2,1,0,0,0,1,31,1,5,6,NA
+63266,7,2,1,0,10,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6575.334977,7272.954395,2,95,2,2,0.42,3,3,2,0,0,1,25,1,3,5,NA
+63267,7,2,2,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,35965.834545,38220.111719,1,101,3,3,1.12,1,1,0,0,1,2,72,1,3,2,NA
+63268,7,2,1,12,NA,2,2,1,12,154,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15506.325263,16662.012915,1,103,5,5,0.74,5,5,0,1,0,1,47,2,1,1,1
+63269,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,25840.959268,29976.982116,1,101,77,77,NA,2,2,0,0,2,2,80,1,1,2,NA
+63270,7,2,1,4,NA,1,1,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17306.08847,2,98,14,14,3.25,5,5,2,1,0,1,37,1,5,1,5
+63271,7,2,1,19,NA,5,6,1,19,236,2,NA,2,2,5,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8062.957534,8614.793403,1,92,12,12,NA,7,7,1,2,1,2,45,2,3,1,3
+63272,7,2,2,20,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,NA,NA,NA,1,2,2,1,114993.808573,116714.079488,1,98,7,NA,NA,4,1,0,0,0,2,20,1,4,5,NA
+63273,7,2,1,66,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11492.781131,11942.286414,1,100,15,15,5,2,2,0,0,2,2,64,1,3,1,3
+63274,7,2,2,5,NA,5,7,1,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11956.543345,12952.739683,1,94,2,2,0.33,5,5,1,3,0,2,37,1,4,3,NA
+63275,7,2,1,0,3,3,3,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22570.662508,23423.888442,1,92,9,9,2,6,6,1,3,0,1,33,1,4,1,4
+63276,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,18079.825103,18175.893804,2,97,3,3,0.92,1,1,0,0,1,1,67,1,1,2,NA
+63277,7,2,1,18,NA,3,3,2,18,224,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,69216.263169,70543.167563,1,91,14,14,3.15,5,5,0,1,0,2,50,1,5,1,5
+63278,7,2,2,3,NA,5,6,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7029.864692,7462.832472,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+63279,7,2,1,1,18,4,4,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5770.570361,6126.790978,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+63280,7,2,1,12,NA,5,6,1,12,150,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7558.274942,7903.176011,2,96,15,15,5,3,3,0,2,0,2,42,2,5,4,NA
+63281,7,2,1,40,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,35406.972937,35699.930106,2,98,6,6,1.21,4,4,1,0,0,2,49,2,2,6,NA
+63282,7,2,1,5,NA,5,7,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11081.601387,11216.767571,2,95,6,6,1.3,4,4,1,1,0,1,47,1,5,1,4
+63283,7,2,1,36,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19133.795531,19387.575333,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+63284,7,2,1,79,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,81415.860066,87761.355472,1,97,7,7,2.64,2,2,0,0,2,1,79,1,4,1,3
+63285,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,42992.537371,49508.460769,2,95,3,3,1.16,1,1,0,0,1,2,80,1,3,2,NA
+63286,7,2,2,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,10402.056617,1,101,4,4,0.79,3,3,1,0,0,1,41,1,3,1,3
+63287,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,18956.418923,20374.294465,2,101,99,99,NA,3,3,0,1,1,2,78,1,1,2,NA
+63288,7,2,2,58,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16033.091438,16307.204086,1,96,77,77,NA,4,4,0,0,0,1,52,2,5,1,5
+63289,7,2,1,6,NA,3,3,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47990.623546,51629.685381,1,100,15,15,4.07,5,5,0,2,0,2,41,1,5,1,4
+63290,7,2,2,16,NA,4,4,2,16,201,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11436.28571,11295.489882,1,93,1,1,0.02,5,5,0,4,0,2,36,NA,NA,5,NA
+63291,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,21799.527449,22114.217255,1,96,14,14,5,1,1,0,0,0,1,47,1,5,3,NA
+63292,7,2,2,6,NA,5,7,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12292.225251,12666.385814,1,95,14,14,3.04,6,6,0,4,0,1,56,1,5,1,4
+63293,7,2,2,61,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,9570.416297,10355.226074,2,98,5,5,1.93,1,1,0,0,1,2,61,1,2,2,NA
+63294,7,2,2,62,NA,5,7,1,NA,NA,2,NA,2,1,5,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9525.506997,9950.810696,2,93,9,9,5,1,1,0,0,1,2,62,2,4,3,NA
+63295,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,44886.08631,48362.89646,2,102,6,6,1.15,5,5,0,0,2,1,80,1,5,1,1
+63296,7,2,2,55,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16672.82247,17074.238333,1,92,15,15,4.44,5,5,0,0,1,1,65,NA,NA,1,5
+63297,7,2,1,50,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,182298.775287,1,101,14,14,5,3,3,0,1,0,2,36,1,5,1,5
+63298,7,2,2,22,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16291.99448,17037.586582,1,90,15,15,3.7,5,5,0,0,0,1,54,NA,NA,1,NA
+63299,7,2,1,17,NA,1,1,1,17,208,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22768.423624,23116.446945,2,98,9,9,2.6,4,4,0,2,0,1,30,1,2,1,2
+63300,7,2,2,5,NA,1,1,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19235.084509,21236.049482,3,92,4,4,0.65,4,4,2,0,0,2,20,1,3,5,NA
+63301,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,12433.776145,13392.195631,2,99,6,6,1.39,4,4,1,0,1,2,63,1,3,3,NA
+63302,7,2,1,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,123110.069898,131423.888983,1,101,5,5,1.45,2,2,0,0,0,2,49,1,4,3,NA
+63303,7,2,1,7,NA,1,1,2,7,85,NA,NA,2,2,3,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11546.167056,12698.029578,1,90,4,4,0.47,7,7,1,1,0,2,50,2,1,1,1
+63304,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,98760.497744,125641.994435,2,91,15,6,2.69,2,1,0,0,0,1,44,1,3,5,NA
+63305,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16905.961576,17390.157008,2,94,7,7,1.18,7,7,1,4,0,2,31,1,4,6,NA
+63306,7,2,1,37,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,34438.924452,34418.648235,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+63307,7,2,1,8,NA,3,3,1,8,104,NA,NA,1,1,NA,1,NA,NA,NA,1,NA,2,1,2,2,1,2,2,1,66868.503099,69864.859716,1,98,15,15,4.34,4,4,1,1,0,1,41,1,5,1,5
+63308,7,2,1,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,22497.998055,22497.243767,1,102,4,2,0.73,2,1,0,0,0,1,36,NA,NA,4,NA
+63309,7,2,1,0,10,5,6,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9148.090461,9898.749752,1,92,14,14,3.3,4,4,2,0,0,1,28,1,4,1,4
+63310,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,25964.813959,25539.412946,2,101,2,2,0.74,1,1,0,0,0,2,53,1,2,5,NA
+63311,7,2,1,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7610.147862,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+63312,7,2,1,74,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,65208.881742,76821.663824,3,91,7,7,2.92,2,2,0,0,2,1,74,1,3,1,3
+63313,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20297.001922,20419.260869,1,96,15,15,5,2,2,0,0,0,1,56,1,5,1,5
+63314,7,2,2,0,8,4,4,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4200.172174,4368.199293,2,93,10,10,2.26,6,6,2,0,0,1,34,1,4,1,4
+63315,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17852.668137,18502.046882,3,91,3,3,0.54,3,3,0,0,1,1,57,2,4,1,3
+63316,7,2,1,10,NA,4,4,2,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9023.469661,9333.449986,2,90,14,14,4.25,4,4,0,2,1,2,45,2,5,5,NA
+63317,7,2,1,32,NA,5,6,2,NA,NA,1,1,2,2,5,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17585.737085,18369.998342,2,91,12,5,2.2,3,1,0,0,0,1,29,NA,NA,5,NA
+63318,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17115.540769,2,97,12,6,2.75,3,1,0,0,0,1,21,NA,NA,77,NA
+63319,7,1,2,51,NA,5,6,NA,NA,NA,2,NA,2,1,7,NA,5,77,NA,1,2,2,1,2,2,NA,NA,NA,NA,18322.475193,0,3,91,12,14,5,2,1,0,0,0,2,51,2,5,77,NA
+63320,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16844.740449,17983.530016,3,91,7,7,2.72,2,2,0,0,0,2,27,2,5,1,5
+63321,7,2,1,6,NA,3,3,2,6,82,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,65374.972543,69123.624578,1,90,15,15,5,4,4,0,2,0,1,37,1,5,1,5
+63322,7,2,2,6,NA,4,4,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9460.53021,9835.43092,1,96,6,6,1.98,2,2,0,1,0,2,29,1,3,5,NA
+63323,7,2,2,48,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16833.890659,17522.038775,3,91,15,15,5,4,4,0,2,0,1,38,2,5,1,5
+63324,7,2,1,35,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,1,NA,NA,NA,NA,11715.79771,11871.189399,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+63325,7,2,2,2,NA,1,1,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,15816.39252,3,92,15,15,5,3,3,1,0,0,1,41,2,5,1,3
+63326,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,11574.200203,11257.441852,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+63327,7,2,1,14,NA,3,3,1,14,176,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25736.568804,25798.716176,3,92,5,5,1.03,4,4,0,3,0,1,55,1,4,4,NA
+63328,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9793.924718,10231.212981,2,100,10,10,2.75,5,5,1,1,1,1,27,1,3,1,5
+63329,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,27842.199551,30775.350974,1,90,15,15,5,2,2,0,0,2,1,74,1,3,1,3
+63330,7,2,1,51,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,174520.785302,174976.353865,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+63331,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,64048.78027,64245.709042,2,95,6,6,1.98,2,2,0,0,0,1,35,1,4,1,4
+63332,7,2,2,15,NA,1,1,1,15,186,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18515.058419,19360.671834,2,96,6,6,0.87,6,6,1,3,0,1,46,2,1,1,1
+63333,7,2,1,66,NA,4,4,1,NA,NA,2,NA,2,2,8,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,8062.039403,9043.239367,2,103,4,4,0.99,2,2,0,0,2,1,66,2,3,1,NA
+63334,7,2,1,9,NA,1,1,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10890.103352,10791.203193,1,102,4,4,0.61,5,5,2,2,0,2,27,2,2,5,NA
+63335,7,2,1,10,NA,4,4,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10665.048307,10877.092316,1,96,6,6,1.52,3,3,0,1,0,2,44,1,3,1,3
+63336,7,2,2,9,NA,5,6,1,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7932.110938,8316.610099,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+63337,7,2,1,75,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,8544.243039,9048.352361,2,95,12,12,NA,3,3,0,0,3,2,73,1,2,1,1
+63338,7,2,1,31,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,28996.250643,28582.932384,2,101,14,14,5,1,1,0,0,0,1,31,1,4,3,NA
+63339,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,15747.833197,16391.584661,3,91,14,14,4.03,4,4,0,2,0,1,51,2,4,1,5
+63340,7,2,1,26,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,34629.549705,1,94,5,5,0.74,5,5,1,1,0,2,24,1,3,1,4
+63341,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,18094.847125,18029.478224,3,91,14,9,2.68,6,4,0,0,2,1,48,2,1,1,1
+63342,7,2,1,10,NA,5,6,2,11,132,NA,NA,2,2,2,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8151.552109,8792.161653,1,90,14,14,3.98,3,3,0,1,0,1,33,2,5,1,5
+63343,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,129042.31855,133624.984525,2,101,14,14,4.86,3,3,0,1,0,1,53,1,4,1,5
+63344,7,2,2,26,NA,1,1,1,NA,NA,2,NA,2,2,1,NA,5,1,2,2,1,2,2,2,2,NA,NA,NA,NA,46606.430863,48004.221702,2,102,15,15,5,3,3,1,0,0,1,41,2,2,1,5
+63345,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10598.2543,1,99,6,6,1.84,2,2,0,0,2,1,69,1,3,1,4
+63346,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,51543.062078,54553.310105,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+63347,7,2,2,15,NA,1,1,1,15,189,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,26325.414456,26852.811114,3,92,7,7,1.3,5,5,1,2,0,2,33,2,2,1,1
+63348,7,1,2,10,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14716.367649,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+63349,7,2,2,48,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,20001.392001,24207.190257,1,100,99,99,NA,6,6,0,1,0,1,53,2,2,1,3
+63350,7,2,1,16,NA,3,3,1,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,103085.56884,109495.609645,1,94,15,15,5,4,4,0,1,0,1,41,1,5,1,5
+63351,7,2,2,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,26465.930618,26744.909102,2,100,14,14,4.59,3,3,1,0,0,1,30,NA,NA,1,4
+63352,7,2,1,64,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,125558.167126,127168.668278,1,94,15,15,5,4,3,0,0,1,1,33,1,2,5,NA
+63353,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,36283.627734,36608.028102,1,102,5,1,0.21,5,4,1,1,0,2,24,1,4,5,NA
+63354,7,2,2,1,17,4,4,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7881.296797,8480.363269,2,96,14,14,3.06,5,5,1,1,1,2,54,1,3,6,NA
+63355,7,2,2,9,NA,1,1,2,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17596.36942,17952.278489,1,97,4,4,0.65,4,4,0,1,0,2,45,2,2,3,NA
+63356,7,1,2,70,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,29145.675285,0,3,92,6,6,1.98,2,2,0,0,2,1,72,1,4,1,1
+63357,7,2,2,31,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,2,6,2,2,2,2,1,2,2,2,2,2,2,31460.18163,32805.207594,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+63358,7,2,1,6,NA,3,3,2,6,82,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14307.565788,15198.555052,3,91,3,3,0.76,3,3,0,1,0,2,24,1,4,6,NA
+63359,7,1,1,1,16,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50163.512784,0,1,90,15,15,5,4,4,1,1,0,2,39,1,5,1,4
+63360,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35525.197101,39677.515157,2,93,8,8,2.17,4,4,0,0,2,2,62,1,4,3,NA
+63361,7,2,1,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,8232.241159,8296.636338,2,98,6,6,2.75,1,1,0,0,1,1,67,1,5,2,NA
+63362,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,38758.039282,0,1,101,6,6,1.78,2,2,0,0,2,1,80,1,2,1,1
+63363,7,2,2,65,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,35475.142583,36054.748166,2,95,2,2,0.75,1,1,0,0,1,2,65,1,5,5,NA
+63364,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,21143.97379,23464.841482,1,97,12,12,NA,4,4,0,0,2,1,72,1,2,1,3
+63365,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,1,2,2,2,2,1,2,23200.373382,24548.135184,2,93,5,5,1.26,3,3,0,1,0,1,55,2,2,1,2
+63366,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5221.819282,5505.774343,3,90,3,3,0.37,5,5,2,2,0,2,36,2,4,4,NA
+63367,7,2,1,10,NA,4,4,1,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8714.559478,8865.734494,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+63368,7,2,2,6,NA,1,1,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14300.71869,15541.734563,2,94,9,9,2.1,5,5,1,2,0,1,31,2,4,1,4
+63369,7,2,2,10,NA,4,4,2,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8147.287486,8590.325322,2,90,8,8,2.59,3,3,0,2,0,2,35,1,4,6,NA
+63370,7,2,2,6,NA,2,2,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15148.721588,15796.67129,2,91,2,2,0.22,4,4,0,3,0,2,45,2,5,4,NA
+63371,7,2,1,13,NA,4,4,2,13,161,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10162.625863,10628.468372,2,99,5,5,0.78,5,5,2,2,0,2,30,1,3,5,NA
+63372,7,2,2,2,NA,1,1,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9955.153132,10271.907274,2,94,77,77,NA,4,4,2,0,0,2,27,2,3,1,3
+63373,7,2,2,29,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16937.04417,23964.143641,2,98,9,9,3.83,2,2,1,0,0,2,29,2,4,5,NA
+63374,7,2,1,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,16851.334496,17374.092058,2,95,4,4,1.34,1,1,0,0,0,1,57,1,1,2,NA
+63375,7,2,2,17,NA,3,3,2,17,206,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22046.14863,22741.646094,2,97,9,9,1.45,7,7,1,2,2,2,45,1,3,5,NA
+63376,7,2,1,10,NA,1,1,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14042.313177,2,98,3,3,0.54,3,3,0,2,0,2,35,1,3,5,NA
+63377,7,2,2,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,28315.055559,38147.595136,1,96,15,15,5,2,2,0,1,0,2,39,1,4,5,NA
+63378,7,2,2,18,NA,1,1,1,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,16360.434077,17687.989006,3,91,5,5,1.03,4,4,0,2,0,2,42,2,1,5,NA
+63379,7,2,1,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9983.293162,10181.782419,1,97,4,4,0.46,7,7,3,3,0,2,31,1,3,1,NA
+63380,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,134450.153638,1,101,7,7,1.74,4,4,1,0,0,1,24,NA,NA,1,4
+63381,7,2,1,3,NA,4,4,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8947.841984,9500.1974,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+63382,7,2,1,39,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,30626.581617,30979.39441,2,90,6,6,0.96,5,5,1,1,0,1,39,2,2,1,NA
+63383,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16801.552467,2,97,6,6,1.7,2,2,0,0,1,2,62,2,5,1,2
+63384,7,2,2,13,NA,5,7,2,13,161,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9508.389138,9876.066661,3,91,8,8,2.24,4,4,0,2,0,1,45,1,4,1,4
+63385,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,15369.196003,16055.414165,1,100,15,5,2.15,2,1,0,0,2,1,60,1,5,6,NA
+63386,7,2,1,7,NA,2,2,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13927.458372,13742.678011,2,98,6,6,0.78,7,7,1,3,1,2,63,1,2,4,NA
+63387,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,25989.236835,2,101,99,1,0.23,3,1,0,0,0,1,20,1,4,5,NA
+63388,7,2,2,44,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,19100.40225,19426.955733,1,96,4,4,0.65,4,4,0,0,0,1,19,1,4,NA,NA
+63389,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,31313.292063,33738.773556,1,95,6,6,1.65,2,2,0,0,2,2,80,1,4,1,4
+63390,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,151766.599459,154414.492891,3,91,7,7,1.97,4,4,0,0,1,2,77,1,5,2,NA
+63391,7,2,1,70,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,4,1,NA,2,2,2,1,2,2,1,2,2,NA,11755.776731,12180.618102,3,90,7,7,1.15,7,7,2,1,1,2,30,1,9,1,4
+63392,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,63717.895797,67578.761066,2,101,5,5,1.36,2,2,0,0,0,2,22,1,4,5,NA
+63393,7,2,2,19,NA,1,1,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20502.928313,21226.512457,2,98,8,8,2.24,4,4,0,0,0,1,58,2,1,1,3
+63394,7,2,1,16,NA,5,6,1,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,5873.088927,6275.048272,1,103,6,6,1.82,2,2,0,1,0,2,56,2,5,77,NA
+63395,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,37456.985636,43133.944105,2,98,99,99,NA,2,2,0,0,2,2,80,1,3,1,1
+63396,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,26344.362464,28384.957988,2,95,13,13,NA,2,2,0,0,2,1,80,1,5,1,3
+63397,7,2,2,51,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,2,1,NA,1,2,1,1,2,1,1,2,1,NA,17018.449206,17108.441787,2,91,99,1,0,7,3,0,4,0,1,36,2,9,1,2
+63398,7,2,2,78,NA,5,6,1,NA,NA,2,NA,2,2,8,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,11812.364238,12222.042847,1,94,4,4,1.11,2,2,0,0,1,2,37,2,4,5,NA
+63399,7,2,2,5,NA,1,1,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18078.669459,18848.958225,1,101,2,2,0.26,5,5,3,0,0,2,26,1,2,1,3
+63400,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,14883.664782,14924.287598,2,90,8,6,1.46,4,3,1,1,0,2,21,1,5,6,NA
+63401,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,163194.688032,166437.638141,1,97,8,8,3.57,2,2,0,0,0,2,49,1,3,3,NA
+63402,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16844.740449,18124.20195,3,91,14,14,5,2,2,0,0,0,2,27,2,5,1,5
+63403,7,2,1,8,NA,4,4,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10199.928366,10684.945227,1,100,9,9,2.22,5,5,1,2,0,2,40,2,4,1,4
+63404,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,20048.680628,21335.7055,2,95,2,2,0.46,3,3,0,0,0,2,48,1,2,1,2
+63405,7,2,2,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,15914.916287,16468.700526,2,99,8,8,3.47,2,2,0,0,1,2,74,1,5,2,NA
+63406,7,2,1,12,NA,4,4,2,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12462.601191,12584.643654,2,97,5,5,0.92,5,5,0,3,0,2,54,1,3,2,NA
+63407,7,2,2,16,NA,4,4,1,16,203,NA,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12531.903464,13043.632492,2,100,4,4,0.86,3,3,0,2,0,2,36,1,3,5,NA
+63408,7,2,1,8,NA,4,4,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13818.701911,14596.53451,2,101,2,2,0.38,3,3,0,2,0,2,56,1,3,2,NA
+63409,7,2,2,69,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,11009.072628,11500.615946,2,99,2,2,0.53,2,2,0,0,1,2,69,1,4,2,NA
+63410,7,2,1,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16598.645683,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+63411,7,2,1,19,NA,4,4,2,19,233,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13863.378072,13964.345562,1,96,15,15,5,4,4,0,1,0,1,56,1,4,1,5
+63412,7,2,2,40,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,2,1,2,2,2,2,2,2,2,2,2,2,2,26889.724138,34281.795458,3,90,12,12,NA,2,2,0,1,0,2,40,2,2,1,NA
+63413,7,2,2,61,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,128294.718377,130390.843426,1,93,15,15,5,4,3,0,0,3,1,80,1,5,2,NA
+63414,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,113642.287024,118593.484239,1,101,9,6,2.24,2,1,0,0,0,2,27,1,3,6,NA
+63415,7,2,1,18,NA,4,4,2,18,226,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8364.097643,8526.060452,2,90,6,6,0.96,5,5,0,1,0,1,55,1,4,6,NA
+63416,7,2,2,7,NA,4,4,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7619.934086,8032.046596,1,103,8,8,1.95,4,4,0,1,0,2,48,1,5,1,5
+63417,7,2,2,6,NA,3,3,1,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,61745.495006,63723.291316,2,101,7,7,1.88,4,4,0,2,0,2,36,1,4,1,5
+63418,7,2,2,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,26595.398371,26963.678025,1,91,7,7,2.2,3,3,0,0,1,2,60,1,2,2,NA
+63419,7,2,1,2,NA,3,3,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25107.382643,27005.542261,1,95,6,6,1.35,3,3,1,0,0,1,22,1,3,1,4
+63420,7,2,1,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,21963.854411,22169.845613,2,102,3,3,1.29,1,1,0,0,1,1,62,1,4,3,NA
+63421,7,2,1,13,NA,5,6,1,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,8915.81491,9312.338117,1,92,2,2,0.33,5,5,0,1,0,1,51,2,1,4,NA
+63422,7,2,2,64,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,12043.867275,12495.995159,1,91,1,1,0.05,2,1,0,0,2,1,72,1,1,3,NA
+63423,7,2,2,2,NA,5,6,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6097.947961,6201.918891,1,96,15,15,4.34,4,4,1,1,0,1,36,2,5,1,5
+63424,7,2,1,17,NA,1,1,1,17,206,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24902.864049,25040.491572,1,94,5,5,0.94,4,4,0,2,0,2,37,2,3,1,2
+63425,7,2,1,51,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,1,1,2,2,1,2,2,1,13567.923118,13997.525763,3,91,14,14,4.03,4,4,0,2,0,1,51,2,4,1,5
+63426,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,33802.428047,1,95,3,3,1.19,1,1,0,0,0,1,58,1,2,3,NA
+63427,7,2,2,9,NA,4,4,2,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9565.802332,9920.158544,1,96,14,14,2.58,6,6,2,2,0,1,40,2,4,1,4
+63428,7,2,2,13,NA,1,1,1,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18515.058419,19360.671834,2,96,6,6,1.12,4,4,0,3,0,1,26,1,2,77,NA
+63429,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,16386.190684,20296.970403,2,90,7,4,1.38,2,1,0,0,0,1,21,1,3,5,NA
+63430,7,2,2,4,NA,3,3,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26966.264969,27815.81123,1,98,6,6,1.31,3,3,2,0,0,2,22,1,3,5,NA
+63431,7,2,1,34,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,2,5,NA,1,2,2,1,2,1,1,2,2,1,20071.705576,20599.326983,3,91,6,6,1.12,4,4,0,0,2,1,69,2,3,1,1
+63432,7,2,1,7,NA,3,3,1,7,89,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48147.167375,49984.220445,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+63433,7,2,1,4,NA,5,7,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10276.262805,11525.953064,1,97,15,15,4.77,4,4,1,1,0,2,40,1,5,1,5
+63434,7,2,2,23,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16844.740449,17849.904413,3,91,6,3,1.1,3,1,0,0,0,2,23,1,5,5,NA
+63435,7,2,1,14,NA,3,3,2,14,177,NA,NA,2,1,1,9,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,26858.274239,26923.130249,2,97,5,5,0.8,5,5,1,2,0,1,46,2,4,1,2
+63436,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,1,2,1,2,2,1,2,2,NA,29302.509441,32727.496464,2,95,15,15,5,3,3,0,0,1,2,47,1,4,1,5
+63437,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,128575.977224,133862.807708,1,102,6,4,1.38,2,1,0,0,0,2,23,1,5,5,NA
+63438,7,2,1,64,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,6358.062034,6283.458155,1,99,15,15,5,2,2,0,0,2,1,64,1,3,1,4
+63439,7,2,1,51,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,23431.677775,25582.72205,2,93,10,10,3.4,3,3,0,1,0,1,51,1,3,1,4
+63440,7,2,1,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,NA,54654.899954,58048.628546,2,103,6,6,1.82,2,2,0,0,2,1,70,1,2,5,NA
+63441,7,2,1,3,NA,2,2,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14505.510202,14676.996441,2,94,6,6,1.43,5,4,2,1,0,2,23,2,3,6,NA
+63442,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,24719.680932,27438.0417,1,97,4,4,1.61,1,1,0,0,0,2,51,1,4,3,NA
+63443,7,2,2,8,NA,3,3,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,23750.822126,1,95,7,7,1.17,6,6,1,3,0,2,44,1,4,1,NA
+63444,7,2,2,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8317.65714,8487.710712,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+63445,7,2,2,21,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,41527.748953,47819.246487,2,97,1,1,0.01,1,1,0,0,0,2,21,1,4,5,NA
+63446,7,2,1,16,NA,4,4,1,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,6,6,1.16,4,4,0,3,0,2,36,1,4,4,NA
+63447,7,2,2,9,NA,4,4,2,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8192.839047,8748.579234,2,99,5,5,1.32,2,2,0,1,0,2,34,1,4,5,NA
+63448,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,47098.572584,49762.054694,1,95,5,5,1.43,2,2,0,0,2,1,80,1,3,1,4
+63449,7,2,1,38,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,2,1,2,2,1,2,2,2,41241.224595,43514.733172,2,102,6,6,1.12,4,4,1,1,0,1,38,2,2,1,3
+63450,7,2,1,6,NA,2,2,1,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16028.98796,15961.028504,1,100,14,14,4.71,3,3,0,1,0,1,38,1,5,1,5
+63451,7,2,1,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16088.355002,17176.258631,2,100,13,13,NA,2,2,0,0,1,2,71,NA,NA,1,NA
+63452,7,2,1,11,NA,4,4,1,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10469.725162,11059.049224,1,100,1,1,0,4,4,1,2,0,2,35,1,2,5,NA
+63453,7,2,1,48,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,1,1,2,1,2,2,2,2,46745.699003,49120.516055,2,98,13,13,NA,5,5,0,2,0,1,48,2,1,1,2
+63454,7,2,1,13,NA,5,6,2,13,159,NA,NA,2,2,1,6,NA,NA,NA,1,1,1,NA,NA,NA,1,2,1,NA,10346.302718,11892.421636,2,91,99,1,0,7,3,0,4,0,1,36,2,9,1,2
+63455,7,2,1,30,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,24134.97174,25042.42256,1,98,7,7,1.52,4,4,2,0,0,1,30,1,3,1,4
+63456,7,2,1,80,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,17609.976188,17569.292452,1,98,15,15,5,5,5,0,1,1,2,55,1,5,1,5
+63457,7,2,1,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25807.688156,26357.095043,1,100,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+63458,7,2,2,0,1,2,2,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8264.844608,8090.961959,1,93,14,14,4.75,3,3,1,0,0,2,42,1,5,1,5
+63459,7,2,2,80,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,2,2,2,2,2,2,NA,21933.218587,23583.994398,1,93,2,2,0.43,2,2,0,0,2,2,80,2,1,2,NA
+63460,7,2,1,2,NA,3,3,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47507.757497,54303.981935,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+63461,7,2,2,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,34906.069211,35828.732314,1,103,1,1,0.03,3,3,0,0,0,1,50,1,2,3,NA
+63462,7,2,2,18,NA,4,4,2,18,218,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10484.6104,10912.740054,2,99,3,3,0.56,4,4,1,0,0,2,38,1,3,5,NA
+63463,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,117419.769432,119338.215671,2,95,15,15,5,2,2,0,0,1,1,54,NA,NA,1,4
+63464,7,2,1,7,NA,1,1,1,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17882.621856,18280.794545,3,92,8,8,1.55,6,6,1,3,0,2,38,1,5,1,4
+63465,7,2,1,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16385.890285,2,100,10,10,4.42,2,2,0,0,0,2,55,1,2,1,4
+63466,7,2,1,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17602.101156,21123.738865,1,96,14,14,2.19,7,7,0,2,0,1,39,1,2,1,3
+63467,7,2,2,10,NA,4,4,1,11,132,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9139.784234,11102.911341,2,100,6,6,0.85,6,6,0,2,0,1,59,1,3,1,3
+63468,7,2,1,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,26052.523863,2,101,NA,99,NA,2,1,0,0,0,1,55,1,2,3,NA
+63469,7,2,2,14,NA,2,2,2,14,175,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,14437.97544,16344.869867,2,90,3,3,0.38,5,5,0,4,0,2,33,2,2,5,NA
+63470,7,2,1,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6580.937346,7256.559544,2,97,3,3,0.66,3,3,2,0,0,2,19,1,3,NA,NA
+63471,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,37074.886861,38850.492434,1,101,3,3,1.07,2,1,0,0,1,2,65,1,3,3,NA
+63472,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,137368.929197,138597.101074,1,91,6,5,1.93,2,1,0,0,0,2,24,1,4,6,NA
+63473,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,31313.292063,33738.773556,1,101,7,7,2.31,2,2,0,0,2,2,80,1,4,1,2
+63474,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,74761.468834,80588.325624,2,91,7,7,1.89,3,3,0,0,2,2,69,NA,NA,1,4
+63475,7,2,1,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,111236.95715,110951.212312,2,92,15,6,2.75,2,1,0,0,0,1,41,1,4,5,NA
+63476,7,2,2,15,NA,1,1,1,15,187,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21818.047789,22469.060223,2,102,5,5,0.89,4,4,1,2,0,2,36,2,5,3,NA
+63477,7,2,2,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10389.292229,10657.457488,2,95,6,6,1.65,2,2,1,0,0,2,27,2,4,5,NA
+63478,7,2,1,62,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,22184.040999,24922.402846,2,94,3,3,1.01,1,1,0,0,1,1,62,1,3,3,NA
+63479,7,2,2,19,NA,4,4,1,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,1,1,0.16,4,1,0,0,0,2,21,1,4,5,NA
+63480,7,2,2,9,NA,1,1,1,9,117,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,10870.942302,10985.231575,2,103,5,5,0.89,5,5,1,3,0,2,34,2,1,99,NA
+63481,7,2,2,48,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,33737.181071,40665.783943,2,91,4,4,0.76,4,4,1,0,0,2,25,2,4,77,NA
+63482,7,2,2,43,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,22513.236051,21981.065205,1,91,10,10,2.56,5,5,0,3,0,1,51,2,5,1,4
+63483,7,2,1,2,NA,1,1,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14810.484787,14830.497242,1,92,10,10,3.04,4,4,2,0,0,2,37,2,5,1,5
+63484,7,2,1,69,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,95193.424282,96086.209653,1,100,5,5,2.15,1,1,0,0,1,1,69,1,5,3,NA
+63485,7,2,1,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18668.895165,18966.875509,1,96,13,13,NA,5,5,1,1,0,1,42,1,3,5,NA
+63486,7,2,2,11,NA,2,2,2,11,134,NA,NA,2,1,4,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15583.587534,16935.930722,1,93,15,15,5,4,4,0,2,0,1,50,1,5,1,5
+63487,7,2,1,17,NA,4,4,1,17,208,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12972.249316,12937.057154,1,100,15,15,3.7,5,5,0,3,0,1,51,1,5,1,5
+63488,7,2,2,6,NA,4,4,1,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9309.947844,9654.826166,2,98,14,14,3.36,4,4,0,2,0,1,37,1,4,1,4
+63489,7,2,1,32,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,85455.763827,87903.260844,2,92,10,10,4.89,2,2,0,0,0,2,34,2,5,1,5
+63490,7,2,2,11,NA,3,3,1,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58636.069237,60514.266203,1,100,6,6,1.18,5,5,2,2,0,2,40,1,5,3,NA
+63491,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,146951.437687,151959.13798,1,98,15,15,5,3,3,0,0,0,1,56,1,5,1,5
+63492,7,1,1,69,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,7101.739553,0,2,100,3,3,0.92,1,1,0,0,1,1,69,1,2,3,NA
+63493,7,2,2,65,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,2,1,2,1,3,14388.729229,15167.035829,1,90,2,2,0.33,2,2,0,0,2,2,65,2,3,1,5
+63494,7,2,2,0,2,4,4,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5099.47985,5303.483608,1,100,9,9,1.78,6,6,1,1,0,2,45,2,3,1,3
+63495,7,2,2,14,NA,5,7,1,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11419.438939,11635.655291,2,103,7,7,1.55,5,5,2,2,0,2,31,1,4,3,NA
+63496,7,2,1,58,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,124170.603852,125419.736085,2,98,6,6,2.57,1,1,0,0,0,1,58,1,3,3,NA
+63497,7,2,2,1,21,2,2,2,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10047.902864,11093.154402,1,90,3,3,0.43,4,4,2,0,0,1,31,1,3,6,NA
+63498,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,20131.904783,23186.488493,2,98,6,6,3.01,1,1,0,0,1,2,70,1,3,2,NA
+63499,7,2,2,33,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,1,6,2,2,2,2,2,2,2,2,2,2,2,38218.668882,37878.487888,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+63500,7,2,1,53,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,15001.445077,15740.201709,2,99,7,7,1.63,4,4,0,2,0,1,53,1,3,3,NA
+63501,7,1,2,4,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13366.393396,0,3,92,8,8,2.01,4,4,1,0,0,2,49,2,5,4,NA
+63502,7,2,1,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,11245.778093,11672.431186,1,98,5,5,1.59,2,2,0,0,2,1,73,1,2,1,5
+63503,7,2,2,3,NA,3,3,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,52983.056893,55223.723565,3,91,15,15,5,6,6,1,3,0,2,40,1,5,1,5
+63504,7,2,1,54,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,2,33162.406014,34847.152396,3,92,10,10,4.3,5,2,2,1,1,2,68,1,3,1,1
+63505,7,2,2,0,1,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23207.538828,22577.170535,1,94,9,9,2.88,3,3,1,0,0,2,28,1,4,1,4
+63506,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,2,2,2,1,2,2,2,9716.805546,10308.451947,2,90,5,5,1.19,3,3,1,0,1,2,60,2,1,4,NA
+63507,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,9015.10987,2,95,3,3,1.31,1,1,0,0,1,2,61,1,2,2,NA
+63508,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,21340.150623,22501.62641,2,95,4,4,1.19,2,2,0,0,0,2,34,1,4,4,NA
+63509,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,3,2,0.54,4,1,0,0,0,1,20,1,4,5,NA
+63510,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,199029.909449,1,91,15,15,5,2,2,0,0,0,2,56,1,4,1,4
+63511,7,2,1,1,17,3,3,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24594.444896,28818.16319,1,98,4,4,0.66,4,4,2,0,0,2,22,1,4,6,NA
+63512,7,2,1,64,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,11019.434708,12461.093856,3,91,4,4,0.81,3,3,0,0,1,1,64,2,1,1,1
+63513,7,1,1,17,NA,4,4,NA,NA,NA,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,15360.522091,0,1,98,15,15,5,4,4,0,2,0,2,50,1,5,1,5
+63514,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,13399.379222,14870.161708,2,90,4,4,1.2,2,2,0,0,2,2,80,1,3,2,NA
+63515,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,220233.315202,222507.874115,1,98,14,14,5,2,2,0,0,0,1,59,1,3,1,4
+63516,7,2,2,54,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,22632.809716,22750.837894,1,103,3,3,0.79,2,2,0,0,0,2,54,2,1,4,NA
+63517,7,2,1,10,NA,5,7,1,10,123,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7101.095162,7528.512272,2,96,15,15,5,3,3,0,2,0,2,42,2,5,4,NA
+63518,7,2,2,3,NA,1,1,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13366.393396,13935.903375,2,94,6,6,1.15,5,5,1,2,0,1,33,1,2,1,2
+63519,7,2,1,54,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,16117.991297,16620.832307,1,96,9,9,2.78,4,4,0,2,0,1,54,2,5,4,NA
+63520,7,2,2,0,0,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21671.775435,21083.121886,1,98,15,15,4.77,4,4,2,0,0,1,35,1,4,1,5
+63521,7,2,2,15,NA,5,6,2,15,183,NA,NA,2,1,4,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8424.942002,9162.575745,2,94,15,15,5,3,3,0,1,0,1,50,1,2,1,4
+63522,7,2,1,6,NA,4,4,2,6,82,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8579.422451,8987.382625,1,99,3,3,0.75,2,2,0,1,0,2,41,1,5,77,NA
+63523,7,2,2,22,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,128171.594518,135937.909642,1,93,14,7,3.95,2,1,0,0,0,2,26,1,5,5,NA
+63524,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,24072.616069,27614.797496,1,100,4,4,1.16,2,2,0,0,2,1,73,1,3,1,3
+63525,7,1,2,52,NA,5,6,NA,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12649.084278,0,3,90,77,77,NA,3,3,0,0,0,1,56,NA,NA,1,5
+63526,7,2,2,9,NA,3,3,1,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,51531.402068,50847.086518,1,94,14,14,2.96,5,5,0,3,0,2,39,1,4,1,3
+63527,7,2,2,11,NA,1,1,1,11,140,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12215.503444,12532.555214,1,102,5,5,0.92,5,5,0,3,0,2,39,2,3,1,3
+63528,7,2,2,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,43717.124878,44415.60038,2,101,6,6,1.54,3,3,0,1,0,2,34,1,4,1,3
+63529,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,31962.323978,33458.586112,1,100,4,4,1.06,3,2,0,0,0,1,22,1,4,6,NA
+63530,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,28483.117096,31921.002992,1,94,3,3,0.39,6,6,1,0,2,1,80,1,4,1,3
+63531,7,2,1,32,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,1,2,2,1,2,2,2,54969.430704,54554.559034,1,100,NA,13,NA,2,1,0,0,0,1,32,2,1,6,NA
+63532,7,2,1,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,19752.546483,23704.421997,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+63533,7,2,2,32,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,3,1,2,2,1,2,2,1,2,2,1,41791.57979,40949.069487,2,102,15,12,NA,5,4,0,3,0,1,42,2,4,6,NA
+63534,7,2,1,39,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,43108.74283,42783.38752,2,91,8,8,1.85,5,5,0,2,1,1,39,2,3,1,4
+63535,7,2,2,3,NA,2,2,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9793.360626,10410.766193,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+63536,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12105.109893,11778.058179,1,95,6,6,0.83,7,6,2,1,0,1,43,1,4,1,4
+63537,7,2,1,1,19,5,6,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5492.796032,5818.035599,2,100,4,4,0.5,6,6,2,1,0,1,30,2,4,1,3
+63538,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16519.058735,17573.900053,1,97,77,77,NA,3,3,0,0,3,2,62,1,5,1,NA
+63539,7,2,1,14,NA,1,1,1,14,173,NA,NA,2,2,2,8,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,30061.88611,30025.949584,1,92,5,5,0.87,4,4,0,2,0,1,42,2,1,1,4
+63540,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,27944.169354,28022.783308,2,94,3,3,1.24,1,1,0,0,0,2,57,1,2,3,NA
+63541,7,2,1,70,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,11755.776731,15354.199056,3,90,12,12,NA,6,6,0,0,2,1,70,2,1,1,1
+63542,7,2,2,2,NA,3,3,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14718.567269,15747.930582,2,97,3,3,0.43,6,6,1,3,0,2,36,2,4,3,NA
+63543,7,2,2,74,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,13446.397433,13865.38304,1,93,13,13,NA,2,2,0,0,2,1,76,2,5,1,5
+63544,7,2,1,63,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,4,1,NA,2,2,2,2,2,2,1,2,2,2,9357.880765,9682.593718,2,90,10,10,2,7,7,0,3,1,1,63,2,4,1,NA
+63545,7,2,2,56,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19319.908753,18964.846355,1,96,15,15,5,2,2,0,0,0,1,56,1,5,1,5
+63546,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,30212.098573,35047.752261,2,98,3,3,1.1,1,1,0,0,1,2,80,1,1,2,NA
+63547,7,2,1,0,4,4,4,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5825.739712,6022.371944,1,99,2,2,0.43,3,3,2,0,0,2,26,1,4,5,NA
+63548,7,2,2,0,10,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9189.424641,8939.81946,1,98,6,6,1.65,2,2,1,0,0,2,24,1,3,4,NA
+63549,7,2,2,5,NA,4,4,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9404.475744,9924.78953,1,96,14,14,2.58,6,6,2,2,0,1,40,2,4,1,4
+63550,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,14834.928147,19579.353906,1,103,2,2,0.53,2,2,0,1,0,2,51,1,3,5,NA
+63551,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,109027.902363,130159.069706,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+63552,7,2,1,50,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23431.677775,23884.31716,2,93,8,8,2,4,4,1,1,0,1,50,2,4,1,4
+63553,7,2,1,76,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,8155.039267,8917.162886,3,90,99,99,NA,2,2,0,0,1,1,76,2,3,3,NA
+63554,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,8895.639271,8965.223759,1,96,15,15,5,2,2,0,0,2,1,61,1,5,1,5
+63555,7,2,1,6,NA,2,2,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13898.598114,14013.214919,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+63556,7,2,1,64,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9082.311855,9548.677283,1,96,7,7,2.64,2,2,0,0,2,1,64,2,5,1,5
+63557,7,2,1,3,NA,5,7,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8981.553859,10073.795326,3,91,15,15,5,3,3,1,0,0,2,39,2,5,1,5
+63558,7,2,2,18,NA,3,3,2,19,228,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,99340.784743,108314.70799,2,94,12,12,NA,5,5,1,1,0,1,37,1,4,1,3
+63559,7,2,1,41,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,20543.822351,1,97,15,15,4.77,4,4,1,1,0,1,41,2,4,1,5
+63560,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,116165.700541,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+63561,7,1,2,65,NA,2,2,NA,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,10938.831279,0,1,93,6,6,1.55,3,3,0,0,3,1,61,2,4,1,1
+63562,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,43892.276067,44574.41704,1,98,6,1,0,2,1,0,0,0,1,29,1,4,5,NA
+63563,7,2,1,16,NA,4,4,2,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10817.360862,11313.215634,2,99,99,99,NA,5,5,0,2,0,2,20,1,3,6,NA
+63564,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,65706.229298,0,1,101,10,10,4.3,2,2,0,0,2,1,80,1,2,1,4
+63565,7,2,2,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12280.58152,13558.091553,2,92,14,14,4.03,4,4,1,1,1,2,30,1,5,4,NA
+63566,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,14809.997435,16069.288743,2,101,13,3,0.64,5,4,0,3,1,2,62,1,1,2,NA
+63567,7,2,2,9,NA,5,7,2,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10699.45895,11230.540406,1,97,8,8,2.51,3,3,0,2,0,2,39,2,4,2,NA
+63568,7,2,1,19,NA,4,4,1,19,235,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12792.875152,12871.476461,2,93,99,99,NA,7,6,1,0,0,1,19,1,3,NA,NA
+63569,7,2,2,49,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,2,2,2,1,2,2,NA,NA,NA,NA,25778.164795,26907.837256,2,90,77,77,NA,3,3,0,0,0,2,49,1,2,5,NA
+63570,7,1,1,71,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,17780.584319,0,1,97,3,3,1.16,1,1,0,0,1,1,71,1,2,3,NA
+63571,7,2,1,36,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,12031.037329,12793.158379,1,99,10,10,3.99,3,3,0,1,0,1,36,2,2,6,NA
+63572,7,2,2,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,81857.569857,83650.947594,2,92,15,7,3.67,4,1,0,0,0,1,28,1,5,5,NA
+63573,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,22154.886709,21259.816918,1,96,7,7,2.31,2,2,0,0,1,2,62,1,3,3,NA
+63574,7,2,1,6,NA,5,6,2,6,78,NA,NA,1,1,NA,1,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5399.929397,5851.470272,3,91,15,15,5,3,3,0,1,0,2,40,2,5,1,5
+63575,7,2,1,19,NA,5,6,2,19,233,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6999.347953,7445.743611,3,90,9,9,2.6,4,4,0,0,1,1,62,2,4,1,5
+63576,7,2,1,46,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,39244.104141,38781.831565,1,90,15,15,5,4,4,1,1,0,2,40,1,5,6,NA
+63577,7,2,2,5,NA,1,1,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12714.663639,13256.404976,1,102,4,4,0.5,6,6,2,2,0,1,25,1,2,1,3
+63578,7,2,2,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,17754.413992,17268.517878,2,99,6,6,1.11,5,5,1,2,0,2,41,1,2,5,NA
+63579,7,2,1,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,120735.071461,122283.708051,1,94,8,8,2.41,3,3,0,0,3,1,63,1,4,1,5
+63580,7,2,1,9,NA,3,3,1,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24713.905595,25821.327941,1,98,4,4,0.67,5,5,1,2,0,1,29,1,4,1,3
+63581,7,2,2,14,NA,5,6,1,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,5760.953091,5914.685125,2,92,8,8,1.91,5,5,0,2,1,2,47,2,1,1,3
+63582,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,33821.324425,34164.529253,3,92,4,3,0.52,5,4,0,0,0,2,57,1,4,1,2
+63583,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,38321.717684,40488.857872,1,95,15,15,5,2,2,0,0,2,2,80,1,4,1,4
+63584,7,2,2,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,33506.462855,33906.169633,2,96,3,3,0.38,5,5,1,2,0,2,30,1,3,5,NA
+63585,7,2,2,8,NA,4,4,2,8,99,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7282.523598,8846.731146,2,99,5,5,0.76,5,5,0,2,0,1,51,1,2,1,2
+63586,7,2,1,4,NA,4,4,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11342.022131,11810.675983,2,102,4,4,0.53,6,6,2,2,0,2,27,1,2,1,2
+63587,7,2,1,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,69063.138927,71041.142737,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+63588,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,41229.806244,46206.205733,1,103,6,6,1.98,2,2,0,0,2,1,80,1,5,1,4
+63589,7,2,1,80,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,2,1,2,1,1,2,2,NA,12882.868646,13921.030537,2,92,77,77,NA,5,5,0,1,2,2,80,NA,NA,1,1
+63590,7,2,1,21,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21495.752024,21074.130073,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+63591,7,2,2,19,NA,5,6,1,19,236,2,NA,2,2,1,13,NA,NA,NA,1,2,1,1,2,1,1,2,2,NA,6145.01663,6308.997467,2,101,12,1,0,5,1,0,0,0,2,19,2,3,NA,NA
+63592,7,2,2,2,NA,4,4,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6276.300496,6799.229858,2,99,1,1,0.1,6,6,2,3,0,2,31,1,2,5,NA
+63593,7,2,1,77,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,73144.179917,78844.986335,2,91,77,77,NA,2,2,0,0,2,2,70,1,3,1,4
+63594,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5256.453918,5473.651288,1,99,5,5,0.84,5,5,2,1,0,1,35,1,3,1,2
+63595,7,2,2,27,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,1,2,1,2,1,1,2,1,NA,NA,NA,NA,8044.549611,8388.06604,2,92,10,6,1.12,7,4,1,1,1,2,27,2,3,1,3
+63596,7,2,1,74,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,15470.980685,17519.370244,1,91,7,7,3.58,1,1,0,0,1,1,74,1,2,1,NA
+63597,7,2,1,22,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,14385.653726,15564.966804,2,101,14,8,4.59,2,1,0,0,0,1,22,2,4,5,NA
+63598,7,2,2,18,NA,2,2,2,18,223,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16189.692833,16766.382859,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+63599,7,1,2,76,NA,2,2,NA,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,18241.877822,0,2,93,7,7,2.31,2,2,0,0,2,1,80,2,1,1,1
+63600,7,2,2,10,NA,3,3,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,51093.739991,50940.240166,2,102,14,14,3.44,5,5,1,2,0,2,34,1,4,6,NA
+63601,7,2,1,5,NA,3,3,1,6,73,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,50644.672852,56040.796491,2,100,NA,NA,NA,5,5,1,2,0,1,36,NA,NA,3,NA
+63602,7,2,1,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,18838.303827,19546.605226,2,93,5,5,1.05,3,3,1,0,0,2,29,1,3,5,NA
+63603,7,2,1,11,NA,1,1,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13533.281742,14131.961026,1,100,7,7,1.74,4,4,0,2,0,2,39,2,1,1,3
+63604,7,2,2,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11414.885224,11778.085279,1,94,7,7,1.56,4,4,2,0,0,1,21,1,4,1,4
+63605,7,2,1,7,NA,4,4,2,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9502.15317,9484.287948,2,97,3,3,0.4,6,6,2,3,0,2,25,1,2,5,NA
+63606,7,2,1,6,NA,3,3,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24713.905595,25821.327941,1,98,4,4,0.67,5,5,1,2,0,1,29,1,4,1,3
+63607,7,2,1,41,NA,1,1,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,36273.943099,35741.678901,1,90,15,15,4.77,4,4,1,1,0,2,41,1,5,1,2
+63608,7,2,2,71,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,1,4,NA,2,2,2,1,2,2,2,2,1,NA,12972.932238,13949.32349,2,93,6,6,0.98,5,5,0,2,1,1,48,2,5,1,5
+63609,7,2,1,66,NA,5,7,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,23761.280382,24066.059994,2,103,5,5,1.2,3,3,0,0,2,1,66,2,2,1,2
+63610,7,2,1,16,NA,4,4,2,16,195,NA,NA,2,2,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11523.287163,11546.069645,1,96,8,8,2.17,4,4,0,2,0,2,45,2,5,4,NA
+63611,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,1,2,1,2,2,1,2,2,NA,20944.990388,22567.358502,2,95,6,6,1.57,3,3,0,0,2,1,80,1,2,1,4
+63612,7,1,1,5,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8006.699948,0,1,90,15,15,5,3,3,1,0,0,1,42,1,5,1,5
+63613,7,2,1,30,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,40219.06806,40845.122132,1,93,9,9,3.98,3,2,0,0,0,2,27,1,3,1,4
+63614,7,2,2,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32144.824104,32969.273278,2,97,2,2,0.77,1,1,0,0,0,2,58,1,2,3,NA
+63615,7,2,2,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,128171.594518,135937.909642,1,93,14,9,5,2,1,0,0,0,2,26,1,5,5,NA
+63616,7,2,1,9,NA,5,6,2,9,115,NA,NA,2,1,3,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6026.102306,6805.584778,2,95,6,6,1.08,4,4,1,1,0,1,39,1,4,1,4
+63617,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,90303.174138,94631.352733,2,92,15,8,4.59,2,1,0,0,0,2,29,1,5,1,NA
+63618,7,2,2,2,NA,5,6,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6024.192029,6555.678695,2,103,15,15,5,3,3,1,0,0,1,35,1,9,1,5
+63619,7,2,1,1,21,3,3,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53040.637714,62149.552868,3,91,8,8,1.95,4,4,2,0,0,2,30,1,5,1,4
+63620,7,2,1,61,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14113.631895,14272.623174,1,92,9,9,3.97,2,2,0,0,1,2,59,1,5,1,4
+63621,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,129559.2554,1,98,14,14,4.96,2,2,0,0,2,1,71,1,5,1,5
+63622,7,2,1,18,NA,3,3,2,18,222,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,88448.252445,87285.839384,1,95,9,9,3.24,3,3,0,0,0,2,42,1,4,3,NA
+63623,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,20611.860643,21777.486671,1,99,5,5,2.02,1,1,0,0,1,1,80,1,5,2,NA
+63624,7,2,2,7,NA,1,1,1,7,90,NA,NA,2,2,3,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13616.85154,14798.521446,1,102,6,6,1.73,3,3,0,1,0,1,30,2,5,1,4
+63625,7,2,2,23,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,50915.06085,54078.415833,3,92,2,2,0.4,3,3,1,0,0,1,21,1,2,6,NA
+63626,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,58826.425292,59007.297053,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+63627,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,56180.550638,62747.143757,2,98,8,8,3.4,2,2,0,0,2,1,80,1,4,1,4
+63628,7,1,1,61,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,125558.167126,0,3,91,12,5,2.02,2,1,0,0,1,1,61,1,3,3,NA
+63629,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,15542.93857,15828.161907,3,91,7,7,2.16,3,3,1,0,1,2,36,2,5,1,NA
+63630,7,2,2,11,NA,3,3,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41020.258665,40851.726244,1,103,15,15,5,2,2,0,1,0,1,49,1,5,77,NA
+63631,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85610.546667,92292.669073,2,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+63632,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,65891.955175,73297.270697,1,95,7,7,2.72,2,2,0,0,2,1,80,1,3,1,3
+63633,7,2,1,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13232.135,13189.930654,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+63634,7,2,1,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8890.779467,9258.147648,3,91,1,1,0.07,6,6,2,3,0,2,30,1,2,3,NA
+63635,7,2,1,79,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,1,2,NA,55241.951079,58472.387275,2,93,7,7,2.31,2,2,0,0,2,2,79,1,3,1,5
+63636,7,2,1,17,NA,4,4,2,17,204,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11093.269222,11505.803928,2,99,3,3,0.44,5,5,1,1,0,2,53,1,4,1,3
+63637,7,2,1,7,NA,3,3,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,74331.764009,78594.005469,2,91,15,15,5,3,3,0,2,0,1,44,2,5,3,NA
+63638,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,125007.681473,127469.032295,1,92,10,10,3.4,3,3,0,0,0,1,56,1,4,1,5
+63639,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16970.447459,16981.184092,1,99,8,8,1.99,5,5,1,0,0,1,55,1,5,1,2
+63640,7,2,1,0,1,3,3,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6420.20398,6662.903304,2,97,3,3,0.33,6,6,2,0,0,2,22,2,4,1,3
+63641,7,2,1,12,NA,3,3,1,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,81241.067732,81008.266507,2,102,14,14,3.44,5,5,1,2,0,2,34,1,4,6,NA
+63642,7,2,1,25,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,20759.115219,21340.685849,1,93,2,2,0.32,3,3,0,1,0,1,25,1,3,6,NA
+63643,7,2,2,18,NA,3,3,1,18,223,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,71832.578284,75129.017547,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+63644,7,2,2,2,NA,4,4,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8010.447273,8217.210509,2,93,7,7,2.58,2,2,1,0,0,2,32,1,5,5,NA
+63645,7,2,1,15,NA,5,7,2,15,185,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13693.32264,13720.395459,1,96,5,5,1.45,2,2,0,1,0,2,41,1,4,3,NA
+63646,7,1,1,9,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13822.148996,0,2,94,5,5,0.65,6,6,0,2,0,1,53,NA,NA,6,NA
+63647,7,2,2,2,NA,4,4,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5158.14785,5587.91487,3,90,99,99,NA,5,5,1,1,1,2,63,1,3,2,NA
+63648,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,71525.773464,1,101,7,7,1.55,5,5,1,2,0,2,31,1,4,1,2
+63649,7,2,1,1,21,3,3,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,38636.988369,42311.269327,1,91,15,15,4.59,4,4,2,0,0,1,35,1,5,1,5
+63650,7,2,2,62,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,7863.861726,8261.011593,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+63651,7,2,2,76,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,1,1,2,2,1,2,1,NA,13446.397433,13865.38304,1,93,3,3,0.82,2,2,0,0,2,1,80,2,5,1,1
+63652,7,2,1,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,1,2,1,2,2,1,1,2,NA,16995.648055,19983.260181,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+63653,7,2,2,19,NA,2,2,1,19,229,2,NA,2,2,3,12,NA,NA,NA,2,2,2,1,2,2,2,2,1,2,12712.538972,12956.930724,2,93,4,4,0.56,5,5,0,0,0,2,49,2,2,5,NA
+63654,7,2,1,11,NA,2,2,2,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9807.589376,10028.771,2,90,7,7,1.66,4,4,0,3,0,2,34,1,5,3,NA
+63655,7,2,2,17,NA,3,3,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,97212.131473,101229.4133,1,91,15,15,5,3,3,0,1,0,1,52,NA,NA,1,5
+63656,7,2,2,16,NA,4,4,2,16,203,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7825.55935,7975.961244,3,90,3,3,0.37,5,5,2,2,0,2,36,2,4,4,NA
+63657,7,2,1,30,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21280.199633,23174.199432,1,97,14,14,4.5,3,3,1,0,0,1,30,1,5,1,5
+63658,7,2,1,10,NA,3,3,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71470.369236,87471.572436,2,101,8,8,2.81,3,3,0,2,0,1,48,1,3,3,NA
+63659,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,36810.071228,39501.597518,1,95,14,14,5,1,1,0,0,1,1,80,1,3,2,NA
+63660,7,2,1,11,NA,3,3,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20080.431771,20654.625485,1,94,4,4,1,3,3,0,1,0,2,41,1,4,5,NA
+63661,7,2,2,46,NA,3,3,1,NA,NA,2,NA,2,1,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,30747.096519,32457.841726,2,91,2,2,0.44,3,3,0,1,0,1,46,2,3,1,4
+63662,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,33707.673642,36318.620408,2,95,10,10,4.3,2,2,0,0,2,1,80,1,5,1,4
+63663,7,2,2,42,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,4,2,2,2,2,1,2,2,NA,NA,NA,NA,32208.300114,32376.263659,1,102,4,4,0.67,4,4,0,1,0,1,23,2,4,5,NA
+63664,7,2,1,58,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,18916.732604,18939.145414,2,91,9,9,5,1,1,0,0,0,1,58,1,5,3,NA
+63665,7,2,1,56,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,5,NA,2,2,2,1,2,2,2,2,1,2,23431.677775,27795.676997,2,93,14,14,2.91,6,6,2,0,1,2,74,NA,NA,2,NA
+63666,7,2,2,79,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,30880.887565,31700.101057,1,98,3,3,1.12,1,1,0,0,1,2,79,1,3,2,NA
+63667,7,2,2,2,NA,3,3,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,36506.435703,36986.142564,1,98,9,9,2.6,4,4,1,1,0,1,31,1,4,1,5
+63668,7,2,1,4,NA,1,1,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14716.463544,14890.443704,2,96,5,5,0.76,5,5,1,2,0,1,44,2,1,1,3
+63669,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,87811.478076,90326.44396,1,90,8,8,1.67,5,5,2,1,0,2,28,1,4,1,5
+63670,7,2,2,8,NA,4,4,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7239.045424,7432.710238,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+63671,7,1,2,14,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5268.765205,0,1,99,9,9,2.68,4,4,0,2,0,1,43,2,3,1,NA
+63672,7,2,1,9,NA,2,2,2,9,112,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,11567.189008,12721.148711,2,90,2,2,0.32,3,3,0,1,0,1,53,NA,NA,1,1
+63673,7,2,1,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,126463.833097,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+63674,7,2,1,5,NA,5,6,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8253.750998,8652.000238,1,90,7,7,1.52,4,4,2,0,0,2,30,2,4,6,NA
+63675,7,2,1,12,NA,3,3,2,12,155,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,101055.977672,102514.960722,1,90,15,15,5,4,4,0,2,0,1,44,1,5,1,5
+63676,7,2,1,2,NA,4,4,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5447.377416,6006.624362,2,90,15,15,5,4,4,1,1,0,1,53,2,5,1,5
+63677,7,2,2,2,NA,1,1,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13065.99844,14425.21291,1,95,6,6,1.37,3,3,1,1,0,2,28,1,4,5,NA
+63678,7,2,2,22,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,2,2,2,1,2,2,1,32537.532358,33640.063825,2,90,1,1,0.22,3,3,0,1,0,2,48,2,2,5,NA
+63679,7,2,1,0,11,3,3,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20370.716701,21140.779334,1,91,15,15,4.2,5,5,3,0,0,2,32,1,4,1,5
+63680,7,2,1,54,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22446.308035,23586.645568,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+63681,7,2,2,5,NA,3,3,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,59190.129038,61693.294338,2,91,15,15,5,5,5,2,1,0,2,40,1,5,1,5
+63682,7,2,1,14,NA,3,3,2,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,77169.155154,77548.231618,1,101,14,14,3.9,4,4,0,2,0,2,41,1,2,1,2
+63683,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,29181.414565,29257.589647,1,94,3,3,0.39,6,6,1,0,2,1,80,1,4,1,3
+63684,7,2,2,54,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,17364.980275,22918.553497,2,93,7,7,2.78,2,2,0,1,0,2,54,1,4,3,NA
+63685,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,12437.053229,13000.378144,3,91,5,5,0.89,4,4,0,2,2,1,61,2,3,1,4
+63686,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,1,NA,1,2,2,NA,NA,NA,1,2,2,1,12237.578196,12748.319173,2,93,8,8,2.17,4,4,0,0,3,1,80,2,2,1,2
+63687,7,2,1,69,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11761.359913,11893.852645,1,92,14,14,5,2,2,0,0,2,1,69,1,5,1,5
+63688,7,2,1,2,NA,5,6,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8253.750998,8652.000238,1,90,7,7,1.52,4,4,2,0,0,2,30,2,4,6,NA
+63689,7,2,2,66,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,93265.413087,97732.118213,2,98,15,15,5,2,2,0,0,2,2,66,1,3,1,1
+63690,7,2,1,29,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12971.121387,14034.473347,1,93,15,15,5,2,2,0,0,0,1,29,2,5,1,5
+63691,7,2,1,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,2,2,2,1,2,2,1,40899.412129,44211.773156,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+63692,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,30259.77569,2,101,4,2,0.6,2,1,0,0,0,2,22,1,4,5,NA
+63693,7,2,2,11,NA,1,1,1,11,137,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,16986.005478,17421.337631,2,102,8,8,1.91,5,5,1,2,0,1,36,2,1,1,4
+63694,7,1,1,6,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46081.129115,0,2,95,15,15,4.63,5,5,1,2,0,1,32,1,4,1,4
+63695,7,1,1,56,NA,4,4,NA,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,15599.953109,0,1,99,12,12,NA,1,1,0,0,0,1,56,1,3,5,NA
+63696,7,2,1,15,NA,1,1,1,15,191,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,18635.323223,18738.312717,1,103,5,5,0.74,5,5,1,1,0,2,40,99,3,1,1
+63697,7,2,1,41,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17602.101156,18151.242681,1,96,10,10,2.95,4,4,0,1,0,2,34,2,3,1,5
+63698,7,2,1,42,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,2,2,2,2,2,2,1,2,1,NA,34128.967046,33628.177065,2,93,3,3,0.87,2,2,0,0,1,2,65,2,4,3,NA
+63699,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,18462.756377,18592.086302,1,99,6,6,1.46,3,3,0,0,1,2,80,NA,NA,2,NA
+63700,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,17602.101156,21123.738865,1,96,14,14,2.19,7,7,0,2,0,1,39,1,2,1,3
+63701,7,2,1,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,145790.86143,146040.776498,1,90,15,15,5,4,4,0,1,0,2,53,1,5,1,5
+63702,7,2,2,11,NA,3,3,1,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,56803.692074,58623.195384,1,100,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+63703,7,2,2,2,24,4,4,1,2,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,7791.142847,2,100,3,3,0.38,5,5,2,1,0,2,28,1,2,5,NA
+63704,7,2,2,75,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,14534.533756,15040.28544,1,96,12,12,NA,3,3,0,0,2,1,77,NA,NA,6,NA
+63705,7,2,1,8,NA,2,2,2,8,101,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,9807.589376,10786.008845,2,90,3,3,0.38,5,5,0,4,0,2,33,2,2,5,NA
+63706,7,2,1,13,NA,3,3,1,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28732.336483,28618.270464,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+63707,7,2,2,63,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,3,3,NA,1,2,2,1,2,2,1,2,2,2,9048.366564,9718.651548,2,93,9,9,4.92,1,1,0,0,1,2,63,2,3,3,NA
+63708,7,2,1,63,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13133.86792,13808.275718,2,102,8,8,1.72,5,5,0,2,1,1,63,2,5,1,5
+63709,7,2,2,16,NA,4,4,2,16,202,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10484.6104,10912.740054,2,99,1,1,0.07,4,4,1,1,0,2,24,1,2,5,NA
+63710,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,94644.050918,104631.781207,2,91,15,15,5,5,5,1,2,0,1,37,1,4,6,NA
+63711,7,1,2,7,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16442.636525,0,1,98,8,8,2.97,2,2,0,1,0,2,38,1,5,5,NA
+63712,7,2,1,2,NA,3,3,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21071.650532,24690.382982,2,97,6,6,1.16,4,4,1,1,0,1,27,2,4,1,3
+63713,7,2,1,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,127000.852889,126845.993399,2,92,15,15,5,2,1,0,0,0,1,52,1,5,6,NA
+63714,7,2,1,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,87954.465296,88343.63479,2,97,15,15,4.97,5,5,1,0,0,1,48,1,4,1,3
+63715,7,2,2,8,NA,4,4,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,13050.526646,2,101,3,3,0.65,3,3,0,1,0,2,54,1,3,5,NA
+63716,7,2,1,2,NA,2,2,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8331.647763,8430.145707,2,90,3,3,0.46,5,5,1,3,0,2,35,2,1,4,NA
+63717,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,37070.062227,37637.166459,1,92,1,1,0.03,2,2,0,1,0,2,50,1,4,3,NA
+63718,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,30543.825378,35038.368904,2,91,2,2,0.86,1,1,0,0,0,2,50,2,3,2,NA
+63719,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,34629.549705,1,94,5,5,0.74,5,5,1,1,0,2,24,1,3,1,4
+63720,7,2,1,5,NA,1,1,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17973.290893,2,98,1,1,0.19,3,3,2,0,0,2,31,1,4,2,NA
+63721,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,NA,NA,NA,1,2,2,1,118761.81384,121363.708734,1,94,8,8,2.43,3,3,0,0,0,2,46,1,3,1,NA
+63722,7,1,1,35,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,17371.064048,0,1,96,NA,NA,NA,4,4,1,1,0,2,37,NA,NA,1,4
+63723,7,2,2,62,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,14204.126838,15256.34028,3,92,5,5,1.56,2,2,0,0,1,1,58,1,3,1,2
+63724,7,2,1,33,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,51543.062078,51154.050295,3,92,3,3,0.52,5,5,2,1,0,2,29,2,1,1,3
+63725,7,2,1,5,NA,3,3,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30883.231636,34843.624635,1,95,7,7,1.17,6,6,1,3,0,2,44,1,4,1,NA
+63726,7,2,2,12,NA,2,2,2,12,147,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21449.694498,22089.715913,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+63727,7,2,2,4,NA,5,6,1,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5879.523197,6241.641704,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+63728,7,2,2,11,NA,1,1,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16986.005478,17733.622754,2,102,7,7,1.53,5,5,1,2,0,1,36,1,2,1,3
+63729,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,160904.289966,160708.090054,1,97,15,15,3.89,5,5,0,2,0,1,50,1,4,6,NA
+63730,7,2,1,14,NA,3,3,1,14,170,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71934.689876,71392.8162,1,100,15,15,5,5,5,0,3,0,1,47,1,5,1,5
+63731,7,2,1,51,NA,2,2,2,NA,NA,2,NA,2,2,77,NA,1,3,NA,2,2,2,2,2,2,NA,NA,NA,NA,24172.845721,24635.602694,2,99,1,1,0.03,2,2,0,0,0,1,51,2,1,3,NA
+63732,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,1,NA,1,2,2,1,2,2,1,2,2,2,5185.036848,5649.75477,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+63733,7,2,2,58,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,23297.357076,24139.998866,2,93,9,9,2.6,4,4,0,0,0,2,58,2,4,4,NA
+63734,7,2,1,43,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,32895.997285,1,95,2,2,0.6,1,1,0,0,0,1,43,1,4,3,NA
+63735,7,2,1,29,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,4,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,14385.653726,15533.829525,2,101,5,3,1.1,2,1,0,0,0,1,29,2,4,1,NA
+63736,7,2,1,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,30631.476666,31783.189579,2,101,6,6,1.54,3,3,0,1,0,2,34,1,4,1,3
+63737,7,2,1,8,NA,4,4,1,8,104,NA,NA,2,2,2,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11443.206518,12087.326286,1,100,9,9,1.78,6,6,1,1,0,2,45,2,3,1,3
+63738,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19260.892847,19199.459573,2,97,5,5,0.76,5,5,0,0,0,2,50,1,4,5,NA
+63739,7,2,2,71,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,17087.605763,2,90,2,2,0.31,5,5,0,2,1,2,71,1,2,2,NA
+63740,7,2,1,8,NA,3,3,2,8,100,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26349.460983,27598.763847,2,97,4,4,1.09,2,2,0,1,0,2,35,1,2,4,NA
+63741,7,2,1,72,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,NA,21815.897449,24704.366996,3,92,5,4,1.39,2,1,0,0,2,2,63,NA,NA,3,NA
+63742,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,1,2,1,2,2,1,1,2,NA,87944.345504,91474.433025,2,98,10,10,3.78,3,3,0,0,0,1,53,1,3,1,4
+63743,7,2,2,9,NA,1,1,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16371.237244,2,98,14,14,4.05,3,3,0,1,0,1,38,1,3,1,4
+63744,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,202246.928022,1,91,15,15,5,3,3,0,1,0,1,52,NA,NA,1,5
+63745,7,2,1,0,4,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,1,2,NA,NA,NA,NA,7284.164858,7692.652584,2,98,5,5,0.76,5,5,3,1,0,2,27,1,3,4,NA
+63746,7,2,1,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,28542.421068,29427.85637,2,101,2,2,0.73,1,1,0,0,0,1,49,1,3,5,NA
+63747,7,2,1,4,NA,5,6,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5953.107662,6129.253313,3,90,5,5,1.15,3,3,1,0,0,1,32,2,3,1,1
+63748,7,2,2,11,NA,2,2,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12307.832776,13375.906079,2,93,10,10,2.26,6,6,0,4,0,1,34,1,4,1,3
+63749,7,2,2,43,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,1,1,2,1,1,2,1,3,9145.761989,9194.124252,2,92,10,8,2.01,7,4,1,1,1,2,27,2,3,1,3
+63750,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,12,1,0.27,4,1,0,0,0,1,21,1,4,6,NA
+63751,7,2,2,5,NA,1,1,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10493.785765,11155.348431,2,103,77,77,NA,5,5,1,2,0,2,30,1,2,1,2
+63752,7,2,1,60,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,25436.904729,28576.794771,1,94,3,3,1.25,1,1,0,0,1,1,60,1,3,3,NA
+63753,7,2,1,33,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,6,NA,2,2,2,2,2,2,2,2,2,2,34438.924452,34835.654759,1,100,8,3,0.68,6,3,1,0,0,1,33,2,3,6,NA
+63754,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,99120.116925,102640.158973,2,98,15,15,5,4,4,0,2,0,2,46,1,4,1,NA
+63755,7,2,1,2,NA,3,3,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46257.816906,49754.984306,2,94,15,15,5,3,3,1,0,0,1,34,1,5,1,5
+63756,7,2,2,2,NA,4,4,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8754.193667,9543.319886,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+63757,7,2,2,5,NA,2,2,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8825.559072,9505.166523,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+63758,7,2,1,28,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,3,11182.713216,11928.362428,2,90,99,4,1.38,3,1,0,0,2,1,60,NA,NA,1,NA
+63759,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,90969.330762,95023.624595,1,99,14,14,4.86,3,3,0,1,0,1,56,1,5,1,5
+63760,7,2,1,16,NA,4,4,2,16,195,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13756.125082,1,96,7,7,1.39,5,5,0,2,2,1,69,2,2,1,2
+63761,7,2,2,21,NA,2,2,1,NA,NA,2,NA,2,1,2,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,40149.982555,39975.11173,2,103,7,7,2.64,2,2,0,0,0,2,21,2,3,1,3
+63762,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,34128.92969,36973.759081,1,92,5,5,1.15,3,3,1,0,0,1,23,1,4,1,4
+63763,7,2,2,17,NA,5,6,2,17,205,2,NA,2,1,4,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7649.811754,7945.620413,2,90,6,6,1.24,4,4,0,1,0,1,57,2,5,1,3
+63764,7,2,2,14,NA,1,1,2,14,172,NA,NA,2,2,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18368.872199,18995.639955,2,94,12,12,NA,4,4,0,2,0,1,47,2,2,1,2
+63765,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,1,1,2,NA,24952.423186,27719.90048,2,98,14,14,4.64,3,3,0,0,1,1,49,1,3,1,9
+63766,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,163102.567998,164787.083347,1,99,15,15,5,2,2,0,0,0,1,54,1,3,1,4
+63767,7,2,1,66,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,6,NA,2,2,2,1,2,2,1,2,2,2,11019.434708,11290.181976,3,91,4,4,1.38,2,1,0,0,1,1,66,2,1,6,NA
+63768,7,2,1,64,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,8609.250304,9380.869295,2,90,12,12,NA,3,3,0,0,2,2,61,2,3,4,NA
+63769,7,2,1,44,NA,2,2,2,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,2,1,2,2,2,2,2,2,37883.328863,38044.589557,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+63770,7,2,2,11,NA,4,4,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10422.423011,10751.577119,1,98,3,3,0.86,2,2,0,1,0,2,34,1,4,5,NA
+63771,7,2,1,43,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,32719.762791,34306.488277,1,101,7,7,2.71,2,2,0,0,0,1,43,1,4,1,4
+63772,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,136843.962333,135447.112041,1,95,14,14,5,2,2,0,0,2,2,65,1,4,1,4
+63773,7,2,1,56,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,31302.008691,30991.200711,2,91,9,9,5,1,1,0,0,0,1,56,1,5,5,NA
+63774,7,2,1,8,NA,3,3,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,62291.152254,66151.751642,3,92,6,6,1.17,4,4,0,2,0,2,30,1,2,1,4
+63775,7,2,2,1,22,5,6,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4686.494502,4910.642079,1,90,14,14,3.33,5,5,1,2,0,1,41,1,5,1,5
+63776,7,2,1,7,NA,1,1,2,7,87,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,16288.924956,19190.320193,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+63777,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,30442.30641,30653.986309,1,95,77,77,NA,3,3,0,0,0,2,41,1,2,5,NA
+63778,7,2,2,14,NA,4,4,1,14,171,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10972.028338,11002.546815,1,103,2,2,0.53,2,2,0,1,0,2,51,1,3,5,NA
+63779,7,2,1,4,NA,4,4,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9488.600894,10206.6462,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+63780,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,35126.205635,35295.519265,1,95,7,7,1.17,6,6,1,3,0,2,44,1,4,1,NA
+63781,7,2,2,63,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,4,6,NA,1,2,2,1,2,2,2,2,2,2,10680.068244,11125.805824,2,93,15,8,4.48,2,1,0,0,2,2,63,2,4,6,NA
+63782,7,2,2,1,19,5,7,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5899.406975,6347.827712,1,103,3,3,0.37,5,5,1,2,0,2,30,1,4,5,NA
+63783,7,2,2,10,NA,4,4,1,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8930.342072,9169.253826,1,100,15,15,3.7,5,5,0,3,0,1,51,1,5,1,5
+63784,7,2,2,40,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,4,1,2,2,2,2,2,2,2,NA,NA,NA,NA,34954.173075,36023.601038,3,92,9,9,2.46,4,4,0,2,0,1,43,2,3,1,4
+63785,7,1,2,32,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,3,1,2,2,1,2,2,NA,NA,NA,NA,86578.861495,0,2,101,8,8,1.72,5,5,0,3,0,1,37,1,3,1,3
+63786,7,2,1,8,NA,5,6,2,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6276.493588,6689.515302,1,98,15,15,5,4,4,1,1,0,1,40,NA,NA,1,5
+63787,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18404.681357,22792.166915,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+63788,7,2,2,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,123493.356057,126566.961964,1,102,3,3,0.92,1,1,0,0,0,2,40,1,3,3,NA
+63789,7,1,1,29,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,84850.826285,0,3,90,15,15,5,2,1,0,0,0,1,29,1,5,1,NA
+63790,7,2,1,11,NA,4,4,2,11,135,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11277.594097,11665.009628,1,90,5,5,0.74,5,5,0,2,0,2,18,1,4,NA,NA
+63791,7,2,2,0,8,1,1,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6700.950086,6971.360383,2,96,3,3,0.24,7,7,2,3,1,2,40,1,3,3,NA
+63792,7,2,2,2,NA,5,6,2,2,35,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6075.554662,6069.137471,1,93,14,14,4.86,3,3,1,0,0,1,31,2,5,1,5
+63793,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,33202.24577,45456.634755,1,99,4,4,1.02,2,2,0,1,0,2,27,1,4,3,NA
+63794,7,2,1,4,NA,3,3,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78946.660075,92504.536814,3,91,15,15,5,4,4,1,1,0,2,41,1,5,1,5
+63795,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,152281.176016,1,94,10,7,3.76,2,1,0,0,0,1,59,1,4,6,NA
+63796,7,2,2,36,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,18901.436377,19274.920746,3,91,15,10,5,2,1,0,0,0,2,36,1,5,6,NA
+63797,7,2,1,14,NA,3,3,2,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,127790.772987,128099.355719,1,97,15,15,3.89,5,5,0,2,0,1,50,1,4,6,NA
+63798,7,2,2,2,NA,3,3,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50865.409844,56139.684211,1,101,7,7,1.74,4,4,1,0,0,1,24,NA,NA,1,4
+63799,7,2,1,69,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,10288.337394,12038.909334,3,91,6,6,1.12,4,4,0,0,2,1,69,2,3,1,1
+63800,7,2,2,3,NA,5,6,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5022.237557,1,91,2,2,0.32,3,3,1,1,0,2,28,1,4,77,NA
+63801,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10440.656902,1,99,6,6,2.3,1,1,0,0,1,2,64,1,5,3,NA
+63802,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,18462.756377,18043.935864,1,99,7,7,3.4,1,1,0,0,0,2,48,1,5,3,NA
+63803,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,6,2,2,2,2,2,2,2,2,2,2,2,41791.57979,45058.092516,2,102,7,7,1.79,4,4,0,2,0,1,40,2,2,6,NA
+63804,7,2,1,61,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,5729.786464,5774.606655,2,95,14,14,4.58,3,3,0,1,1,1,61,1,4,1,5
+63805,7,2,2,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27769.056387,28050.845299,1,94,7,7,1.65,5,4,0,0,0,1,46,1,4,1,4
+63806,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,38226.070503,39676.327756,1,99,12,12,NA,2,2,0,0,2,1,70,1,5,1,5
+63807,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,10798.193871,11318.855823,2,95,2,2,0.67,2,2,0,0,1,2,61,1,3,3,NA
+63808,7,1,2,13,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23477.73925,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+63809,7,2,2,10,NA,3,3,2,10,126,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55626.447796,54931.333538,1,95,9,9,2.13,6,6,0,4,0,2,44,1,1,1,1
+63810,7,2,1,16,NA,5,6,1,16,198,NA,NA,2,2,1,8,NA,NA,NA,1,2,1,1,2,2,1,2,1,NA,7558.274942,7964.822305,2,96,4,4,0.92,3,3,0,1,1,2,41,2,2,1,2
+63811,7,1,2,4,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12129.691938,0,1,97,7,7,1.74,4,4,2,0,0,1,34,1,5,1,5
+63812,7,2,2,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,30932.175051,30085.634936,2,101,3,3,0.68,2,2,0,0,0,1,58,1,1,6,NA
+63813,7,2,1,67,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,6016.509449,6063.572478,1,93,4,4,1.38,1,1,0,0,1,1,67,2,5,2,NA
+63814,7,2,1,17,NA,3,3,2,17,208,2,NA,2,1,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,57357.902168,67351.058423,1,99,77,77,NA,4,4,1,1,0,1,31,2,3,1,3
+63815,7,2,1,8,NA,4,4,2,8,98,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.422451,9062.344401,1,99,5,5,1.13,3,3,1,1,0,2,30,1,1,4,NA
+63816,7,2,1,5,NA,4,4,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7431.820906,7738.904727,1,99,6,6,1.3,5,5,1,2,0,1,34,1,2,1,3
+63817,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,19075.861607,18553.800612,1,96,15,15,5,2,2,0,1,0,2,47,1,5,3,NA
+63818,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,NA,NA,NA,NA,35126.205635,36772.568368,1,95,1,1,0.12,3,3,0,2,0,2,40,1,5,3,NA
+63819,7,2,2,0,11,1,1,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7579.106293,8082.789623,1,90,15,15,4.77,4,4,1,1,0,2,41,1,5,1,2
+63820,7,2,2,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15561.208623,2,100,5,5,1.08,3,3,0,0,0,1,38,1,2,5,NA
+63821,7,2,1,4,NA,4,4,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10510.490567,10944.785425,2,98,6,6,1,5,5,2,1,0,2,31,1,4,6,NA
+63822,7,2,1,13,NA,4,4,2,13,160,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13416.172328,13756.125082,1,96,14,14,2.19,7,7,0,2,0,1,39,1,2,1,3
+63823,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,104488.914565,109862.797404,1,98,3,1,0.27,4,1,0,0,0,1,20,1,4,5,NA
+63824,7,1,1,34,NA,5,6,NA,NA,NA,2,NA,2,1,5,NA,3,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,16974.834956,0,1,93,9,9,1.77,7,7,0,2,0,2,56,NA,NA,5,NA
+63825,7,2,1,31,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,20729.794283,21506.407763,3,91,14,14,5,1,1,0,0,0,1,31,1,5,5,NA
+63826,7,1,1,9,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,55902.987481,0,1,103,15,15,5,2,2,0,1,0,2,52,2,5,3,NA
+63827,7,2,2,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,29756.667529,30535.306876,2,98,4,4,1.19,2,2,0,0,0,2,52,1,3,1,NA
+63828,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,10,10,4.76,2,2,0,0,2,2,64,1,4,1,5
+63829,7,2,2,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,80418.665565,83274.564983,2,100,10,10,3.13,4,4,0,2,0,1,45,1,4,1,4
+63830,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,1,2,2,2,2,1,NA,11225.5556,12146.092964,2,93,8,8,2.62,3,3,0,0,2,2,64,2,1,1,1
+63831,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,20000.263815,19016.481945,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+63832,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,133542.212862,137914.897649,3,91,8,8,1.95,4,4,2,0,0,2,30,1,5,1,4
+63833,7,2,1,9,NA,2,2,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,13742.678011,2,98,6,6,0.78,7,7,1,3,1,2,63,1,2,4,NA
+63834,7,2,2,52,NA,4,4,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,15776.849485,18474.07459,2,93,14,14,2.78,5,5,0,0,0,1,52,2,4,1,2
+63835,7,2,2,76,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,1,2,1,2,2,1,1,2,NA,12535.973802,12972.182488,2,100,15,15,4.97,5,5,0,2,1,2,42,1,5,1,5
+63836,7,2,1,16,NA,4,4,2,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12462.601191,12584.643654,2,97,5,5,0.92,5,5,0,3,0,2,54,1,3,2,NA
+63837,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,63504.762752,64182.95787,1,91,10,10,5,1,1,0,0,1,2,70,1,4,2,NA
+63838,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,25812.913537,27430.822937,2,95,4,4,1.34,1,1,0,0,1,2,78,1,3,2,NA
+63839,7,2,1,75,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,9257.537917,9400.859081,1,99,10,10,2.71,5,5,1,1,2,1,75,1,1,1,3
+63840,7,2,1,10,NA,3,3,2,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,57057.523607,61384.115788,1,98,10,10,3.04,4,4,0,2,0,2,47,1,4,1,3
+63841,7,2,1,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,59682.963348,63721.06052,2,102,10,10,4.76,2,2,0,0,0,1,27,1,4,1,4
+63842,7,2,1,0,7,5,6,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8577.614982,8519.98112,2,91,14,14,4.19,3,3,1,0,0,2,31,1,5,1,5
+63843,7,2,2,13,NA,2,2,2,13,159,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,17848.732433,20206.102312,2,91,99,99,NA,6,6,1,3,0,2,20,2,2,5,NA
+63844,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,28974.537912,30613.083607,1,99,6,6,1.7,2,2,0,0,2,2,77,1,3,1,5
+63845,7,2,1,43,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,29247.173902,1,94,3,3,1.21,1,1,0,0,0,1,43,1,4,3,NA
+63846,7,2,1,3,NA,3,3,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,38749.197676,45403.777431,1,92,6,6,1.24,4,4,1,1,0,1,30,1,3,3,NA
+63847,7,2,1,58,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,155918.172532,3,91,6,6,2.24,1,1,0,0,0,1,58,1,3,5,NA
+63848,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,16214.132654,16029.957596,3,90,15,15,4.89,5,5,0,0,0,2,57,2,3,1,3
+63849,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,71279.707602,73321.195079,1,103,14,14,2.96,5,5,1,2,0,1,34,1,4,1,5
+63850,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,140084.621651,140222.178533,1,102,10,10,4.76,2,2,0,0,0,1,23,1,5,5,NA
+63851,7,2,1,9,NA,4,4,1,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,14179.490667,2,101,1,1,0.27,3,3,0,2,0,2,36,1,3,5,NA
+63852,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,158509.005274,161135.953834,1,100,15,15,5,4,4,0,1,0,1,50,1,4,1,4
+63853,7,2,2,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,77299.255327,77536.924917,2,92,15,15,5,2,2,0,0,0,2,36,1,5,1,5
+63854,7,2,2,5,NA,1,1,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14812.229505,15746.041025,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+63855,7,2,1,8,NA,2,2,2,8,97,NA,NA,2,1,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14700.46789,14706.980156,1,97,15,15,5,4,4,1,1,0,2,43,1,5,1,5
+63856,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,2,1,0.13,4,1,0,0,0,2,19,1,4,NA,NA
+63857,7,1,1,4,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8199.080864,0,1,90,4,4,0.67,5,5,3,0,0,2,32,2,3,3,NA
+63858,7,2,2,20,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,2,6,1,2,2,2,2,2,2,2,2,2,2,42621.881199,51089.379491,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+63859,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,17648.758915,17930.45653,2,99,3,3,0.44,5,5,1,1,0,2,53,1,4,1,3
+63860,7,2,2,50,NA,5,7,2,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,120670.257812,123358.620194,3,90,9,9,3.14,3,3,0,0,0,1,56,2,3,1,3
+63861,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,138075.879417,141933.339512,2,91,15,15,5,2,2,0,0,1,2,57,1,5,1,5
+63862,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,NA,43004.283229,49692.210403,1,97,NA,77,NA,2,1,0,0,2,1,80,1,3,6,NA
+63863,7,2,1,6,NA,4,4,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9502.15317,9484.287948,2,97,13,13,NA,6,6,2,2,0,2,24,1,2,6,NA
+63864,7,2,1,3,NA,1,1,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14993.478359,14600.446315,1,97,6,3,0.45,7,6,2,1,0,1,29,2,2,1,1
+63865,7,2,1,2,NA,2,2,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11853.772636,11869.789876,2,96,6,6,1.12,4,4,2,0,0,1,27,2,2,6,NA
+63866,7,2,2,0,3,2,2,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7367.430495,7419.943227,2,91,2,2,0.19,5,5,3,0,0,1,24,2,1,1,3
+63867,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,28930.832375,30957.619671,1,92,5,5,1.05,3,3,1,1,0,2,35,1,4,5,NA
+63868,7,2,1,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,131445.986898,130104.237054,1,95,10,10,5,1,1,0,0,1,1,69,1,5,3,NA
+63869,7,1,1,36,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,86986.68246,0,2,94,15,15,5,2,2,0,0,0,1,36,1,2,1,4
+63870,7,2,2,11,NA,4,4,2,11,140,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8184.286585,8442.757341,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+63871,7,2,2,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18490.479848,18604.849683,2,100,7,7,2.37,3,3,0,1,1,2,45,1,5,1,NA
+63872,7,2,2,18,NA,5,6,1,18,218,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,6370.316505,6989.133332,2,92,7,7,1.17,6,6,0,1,1,1,78,2,1,1,3
+63873,7,2,1,80,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,8763.51401,9280.560292,2,100,5,3,1.29,3,1,0,0,2,2,80,1,1,2,NA
+63874,7,2,1,3,NA,1,1,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14505.510202,14964.673559,2,94,7,7,1.23,6,6,2,1,0,1,33,2,1,6,NA
+63875,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,10266.896689,11139.888992,1,96,3,3,1.25,1,1,0,0,1,2,64,1,2,2,NA
+63876,7,2,2,9,NA,4,4,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7321.387082,7717.353133,1,103,3,3,0.52,3,3,0,2,0,2,45,1,4,5,NA
+63877,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,29672.504425,2,101,12,1,0.32,4,1,0,0,0,2,20,1,4,5,NA
+63878,7,2,1,8,NA,3,3,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25457.327997,26428.651236,3,92,6,6,1.41,4,3,0,1,0,1,41,1,4,1,4
+63879,7,2,1,76,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,38226.070503,39676.327756,1,99,7,7,3.31,1,1,0,0,1,1,76,1,5,2,NA
+63880,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53541.401974,54961.757487,1,99,7,7,2.72,2,2,0,0,2,1,77,1,2,1,3
+63881,7,2,2,25,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,50197.83865,52206.228199,1,93,12,6,2.75,4,1,0,0,0,2,25,2,5,5,NA
+63882,7,2,1,43,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15271.238172,15754.77307,3,91,15,15,5,3,3,0,1,0,2,44,2,5,1,5
+63883,7,1,2,80,NA,2,2,NA,NA,NA,2,NA,2,1,8,NA,1,2,NA,2,2,2,2,2,2,NA,NA,NA,NA,17318.187297,0,2,90,4,4,0.57,5,5,1,0,2,2,80,2,1,2,NA
+63884,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,7867.418849,7869.055339,1,103,77,77,NA,6,6,0,2,2,1,70,NA,NA,1,1
+63885,7,2,1,19,NA,4,4,1,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,1,1,0.32,2,1,0,0,0,1,19,1,4,NA,NA
+63886,7,2,2,7,NA,5,6,2,7,95,NA,NA,2,1,2,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11068.40581,11617.80042,1,97,15,15,5,3,3,0,1,0,2,40,1,5,1,5
+63887,7,2,2,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,94637.019439,107051.932277,1,94,7,7,3.13,1,1,0,0,1,2,64,1,3,2,NA
+63888,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,20948.005175,20851.734447,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+63889,7,2,1,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,29425.409675,1,94,1,1,0.36,1,1,0,0,0,1,48,1,3,5,NA
+63890,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,9716.805546,10308.451947,2,90,3,3,0.68,2,2,0,0,1,1,21,2,4,5,NA
+63891,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,1,96,9,9,4.13,2,2,0,0,1,2,55,1,4,5,NA
+63892,7,2,2,10,NA,3,3,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49543.011606,52616.361829,1,94,15,15,5,6,6,0,4,0,1,38,1,5,1,4
+63893,7,1,1,6,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12722.361971,0,2,93,8,8,2.62,3,3,0,1,0,1,43,2,4,6,NA
+63894,7,2,2,19,NA,4,4,2,19,231,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13775.846051,13737.029624,2,97,7,7,1.06,7,7,1,2,0,2,40,1,4,5,NA
+63895,7,2,1,37,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18094.858384,18452.252422,1,96,15,15,5,3,3,0,1,0,2,37,1,4,1,4
+63896,7,2,2,9,NA,5,6,2,9,118,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9620.269705,10320.938187,2,91,14,14,3.47,4,4,1,1,0,2,40,2,5,1,NA
+63897,7,2,2,7,NA,2,2,2,7,95,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13338.830817,13909.366846,1,90,6,6,1.68,3,3,1,1,0,2,36,2,4,4,NA
+63898,7,2,2,51,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,NA,NA,NA,1,2,2,1,24870.513993,25000.211608,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+63899,7,2,2,75,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,1,3,NA,2,2,2,2,2,2,NA,NA,NA,NA,18006.276697,20109.914446,2,93,3,3,1.14,1,1,0,0,1,2,75,2,1,3,NA
+63900,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,1,1,1,1,2,1,1,2,1,NA,NA,NA,NA,14138.631841,14167.501749,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+63901,7,2,1,12,NA,2,2,1,12,155,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20560.901695,20875.182272,2,96,2,2,0.27,6,6,1,3,0,1,34,NA,NA,1,NA
+63902,7,2,1,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,143397.643998,144742.519652,2,98,10,10,4.76,2,2,0,0,2,1,68,1,4,1,NA
+63903,7,2,1,19,NA,4,4,1,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13731.625553,14242.275028,1,100,14,14,3.93,3,3,0,1,0,2,47,1,5,4,NA
+63904,7,2,2,42,NA,3,3,2,NA,NA,2,NA,2,2,6,NA,4,4,3,1,2,2,1,2,2,NA,NA,NA,NA,113005.700223,113116.666806,1,99,10,10,5,1,1,0,0,0,2,42,2,4,4,NA
+63905,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,1,1,2,1,NA,NA,NA,NA,14979.624397,15387.451243,1,102,3,2,0.73,2,1,0,0,0,1,24,2,4,5,NA
+63906,7,2,1,8,NA,4,4,2,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11324.954668,11525.075349,1,93,4,4,1.03,3,3,1,1,0,2,35,2,3,4,NA
+63907,7,2,1,60,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,11492.781131,11582.681279,1,100,15,14,5,2,1,0,0,2,1,60,1,5,6,NA
+63908,7,2,1,55,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,22446.308035,22116.943066,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+63909,7,2,2,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,43972.04458,44690.47559,2,96,3,3,0.95,2,2,0,0,2,2,62,1,4,1,4
+63910,7,1,2,41,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,134076.17118,0,1,98,15,15,5,3,3,0,0,0,2,41,1,5,1,NA
+63911,7,2,2,67,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10083.559248,10533.779383,1,96,9,9,4.92,1,1,0,0,1,2,67,1,4,3,NA
+63912,7,2,1,28,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,11477.00145,11789.46785,2,100,12,5,1.88,5,1,0,0,0,1,22,NA,NA,5,NA
+63913,7,2,1,18,NA,2,2,2,18,223,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14081.782012,14391.713696,2,90,2,2,0.25,5,5,0,1,0,2,41,2,4,1,NA
+63914,7,2,2,7,NA,1,1,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16535.37648,2,98,9,9,2.6,4,4,0,2,0,1,30,1,2,1,2
+63915,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,20048.680628,20209.076405,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+63916,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,19788.748292,21708.160855,1,94,4,4,0.59,5,5,0,3,0,1,34,1,2,1,4
+63917,7,2,1,71,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,14055.866747,14300.892374,2,93,5,5,1.32,2,2,0,0,2,1,71,2,4,1,1
+63918,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,157853.944421,1,95,12,12,NA,3,3,0,0,0,1,49,1,5,1,NA
+63919,7,2,1,2,NA,1,1,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11160.282155,11957.436882,2,97,2,2,0.27,3,3,2,0,0,2,19,1,3,NA,NA
+63920,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,1,1,2,2,1,2,2,1,2,2,1,104934.725755,109561.954446,2,98,10,10,4.42,2,2,0,0,0,1,31,1,4,1,3
+63921,7,2,1,3,NA,4,4,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8402.233801,8920.908511,2,101,1,1,0.1,6,6,1,2,1,2,27,1,2,1,2
+63922,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,5,2,1,2,2,1,2,2,1,2,2,3,14767.290807,18767.43866,2,90,5,5,1.43,2,2,0,1,0,2,38,2,4,5,NA
+63923,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13728.308948,13391.816234,2,90,15,15,5,4,4,0,0,0,1,57,2,5,1,5
+63924,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,2,1,2,2,1,2,2,1,2,2,1,18730.678506,18151.501185,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+63925,7,1,2,32,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,6,3,1,2,2,1,2,2,NA,NA,NA,NA,18933.643351,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+63926,7,2,1,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21317.283165,21013.422419,2,96,5,5,0.67,6,6,1,2,1,1,34,1,4,1,4
+63927,7,2,1,37,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,2,2,1,1,2,1,2,2,2,2,34997.800447,34733.660984,2,96,3,3,0.46,5,5,1,2,0,1,37,1,1,1,2
+63928,7,2,2,42,NA,4,4,2,NA,NA,2,NA,2,1,8,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,22754.478969,25596.207975,1,96,7,7,3.58,1,1,0,0,0,2,42,2,5,5,NA
+63929,7,2,2,34,NA,2,2,2,NA,NA,2,NA,2,1,99,NA,4,1,1,1,2,2,2,2,2,1,2,2,1,28899.648059,29089.927914,1,96,15,15,5,4,4,0,2,0,1,36,2,3,1,4
+63930,7,2,1,4,NA,5,7,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7666.445036,8598.756895,1,96,6,6,1.34,3,3,1,0,0,2,42,2,4,6,NA
+63931,7,1,2,32,NA,5,6,NA,NA,NA,2,NA,2,1,4,NA,3,2,3,1,2,2,1,2,2,NA,NA,NA,NA,20584.427267,0,1,95,6,6,1.34,4,4,0,2,0,2,32,2,3,2,NA
+63932,7,2,1,1,13,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15546.999135,15290.852049,2,98,10,10,3.82,3,3,1,0,0,2,33,1,4,1,2
+63933,7,2,2,6,NA,3,3,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19880.837381,19632.404233,1,95,2,2,0.22,4,4,2,1,0,2,22,1,2,5,NA
+63934,7,2,1,18,NA,2,2,2,18,227,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26628.917773,31355.430562,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+63935,7,2,1,17,NA,3,3,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20991.942732,21394.366957,1,91,3,3,0.62,3,3,0,1,0,2,55,1,4,4,NA
+63936,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,110819.46221,119829.247006,2,96,10,7,3.67,2,1,0,0,0,1,35,1,5,5,NA
+63937,7,2,2,15,NA,4,4,1,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12744.907234,12578.013387,2,93,7,7,2.78,2,2,0,1,0,2,54,1,4,3,NA
+63938,7,2,2,12,NA,4,4,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12208.965913,12602.394285,2,100,15,15,4.97,5,5,0,2,1,2,42,1,5,1,5
+63939,7,2,1,66,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9221.19173,9293.322792,2,95,5,5,1.32,2,2,0,0,2,1,66,1,2,1,4
+63940,7,2,1,15,NA,2,2,2,15,183,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,16033.31661,16386.200415,2,90,10,10,3.13,4,4,1,2,0,2,39,1,5,4,NA
+63941,7,2,2,53,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14680.520497,15280.64157,3,91,9,9,3.24,3,3,0,1,1,1,64,2,2,1,5
+63942,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,145772.192378,148626.818504,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+63943,7,2,2,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,137368.929197,138597.101074,1,91,8,7,3.21,2,1,0,0,0,1,33,1,5,6,NA
+63944,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39031.957066,39875.03294,1,95,77,77,NA,3,3,0,0,0,2,41,1,2,5,NA
+63945,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,87611.353116,90505.825807,2,92,10,10,5,1,1,0,0,0,2,29,1,5,5,NA
+63946,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,15867.258653,19649.848899,3,91,6,6,1.22,5,5,1,2,0,2,37,1,4,1,2
+63947,7,2,1,1,16,4,4,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6454.74783,2,100,6,6,1.51,3,3,1,0,0,1,29,1,3,1,4
+63948,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,24351.694017,24510.535836,1,100,12,14,5,2,1,0,0,0,2,50,1,3,6,NA
+63949,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18295.488967,19299.909712,2,103,12,12,NA,4,4,0,1,0,2,50,2,3,1,4
+63950,7,2,2,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,22531.325817,21485.967466,2,91,15,15,5,4,4,0,0,1,1,61,NA,NA,1,2
+63951,7,2,1,14,NA,4,4,2,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11806.79775,12091.578602,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+63952,7,2,2,10,NA,3,3,1,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,63812.58912,64724.869865,3,91,15,15,5,2,2,0,1,0,1,42,1,5,2,NA
+63953,7,2,2,31,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,39561.667842,38764.112139,3,92,7,7,1.49,5,5,0,2,1,2,62,1,4,2,NA
+63954,7,2,1,41,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,17249.311662,20609.214634,2,91,6,6,1.57,3,3,0,1,0,1,41,2,3,1,3
+63955,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,128294.718377,130390.843426,1,93,15,12,NA,4,1,0,0,3,1,80,1,5,2,NA
+63956,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,87648.893016,94490.113574,2,91,15,15,5,4,4,2,0,0,2,33,1,5,1,5
+63957,7,2,2,59,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19676.781212,23812.59189,1,103,7,7,1.33,5,5,0,1,1,1,28,1,2,1,3
+63958,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,2,NA,2,2,2,1,2,2,2,2,1,2,11851.128358,12729.036357,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+63959,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,28129.563738,1,101,6,6,1.17,4,4,0,1,0,1,41,1,3,6,NA
+63960,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25651.892165,25126.271361,1,97,6,6,1.41,3,3,0,1,0,2,51,1,4,5,NA
+63961,7,2,2,0,9,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6699.512423,6879.514499,3,91,15,14,4.03,5,4,2,0,0,1,42,2,4,1,5
+63962,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,54095.581484,0,2,91,6,6,2.24,1,1,0,0,1,2,80,1,3,2,NA
+63963,7,2,1,72,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,83244.241244,88112.194036,2,102,15,15,5,2,2,0,0,2,1,72,1,5,1,NA
+63964,7,2,1,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,85786.890667,86591.455494,2,96,15,15,5,3,2,0,0,1,1,62,1,5,1,2
+63965,7,2,1,72,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,19115.0887,22519.216384,2,98,6,6,1.26,4,4,0,1,2,2,63,1,3,1,3
+63966,7,2,2,34,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,26426.249254,26363.744856,1,99,7,1,0.24,2,1,0,0,0,2,38,1,5,6,NA
+63967,7,2,2,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,16929.836231,18556.498474,2,101,8,5,1.84,2,1,0,0,0,2,27,2,5,5,NA
+63968,7,2,2,61,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,3,NA,2,2,2,1,2,2,2,2,2,2,10310.145764,11073.900839,1,100,13,13,NA,4,4,0,1,1,1,26,2,3,5,NA
+63969,7,2,1,8,NA,4,4,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9173.220503,9355.60378,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+63970,7,2,2,9,NA,5,6,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10235.530383,11121.127441,1,100,7,7,2.51,2,2,0,1,0,2,37,2,5,3,NA
+63971,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,3,1,2,1,2,1,1,2,2,1,2,1,NA,21470.176619,21617.60214,1,98,7,7,2.71,2,2,0,0,0,1,43,1,3,1,3
+63972,7,2,1,15,NA,3,3,2,15,189,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74123.161671,78732.269389,1,93,15,15,4.59,4,4,0,1,0,1,57,1,5,1,5
+63973,7,2,2,15,NA,5,6,2,15,190,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12224.9472,12547.183925,1,97,14,14,2.72,7,7,0,2,0,1,40,1,5,1,5
+63974,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,139662.869107,140583.8015,1,100,15,15,5,3,3,0,0,0,1,53,1,5,1,4
+63975,7,2,2,6,NA,4,4,1,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9309.947844,9472.353128,2,98,2,2,0.31,4,4,2,1,0,2,27,1,2,4,NA
+63976,7,2,1,1,13,4,4,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8022.100831,8575.148723,2,102,7,7,1.53,5,5,1,3,0,2,36,1,5,5,NA
+63977,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,18070.666316,17713.108755,1,99,6,6,1.46,3,3,0,0,1,2,80,NA,NA,2,NA
+63978,7,2,1,12,NA,2,2,1,12,146,NA,NA,2,2,1,6,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15674.964193,15512.049412,2,93,9,9,1.49,7,7,0,3,0,2,41,2,5,1,5
+63979,7,2,1,8,NA,3,3,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,65374.972543,69123.624578,1,90,15,15,5,4,4,0,2,0,1,37,1,5,1,5
+63980,7,2,2,6,NA,4,4,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7814.742747,8095.069503,2,95,7,7,2.64,2,2,0,1,0,1,48,1,2,77,NA
+63981,7,2,2,6,NA,3,3,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,43528.185872,43349.34959,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+63982,7,2,2,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,41983.304745,42669.243011,1,101,7,7,2.31,3,2,0,0,1,2,69,1,4,2,NA
+63983,7,2,1,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,94698.084211,97410.286035,1,101,5,5,0.89,5,5,1,2,0,1,31,1,2,1,1
+63984,7,2,1,24,NA,5,7,2,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,33942.93341,34470.44913,3,91,6,6,2.04,2,2,0,0,0,1,24,2,4,5,NA
+63985,7,2,1,5,NA,2,2,2,5,70,NA,NA,2,2,2,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12403.412256,12420.172189,2,90,2,2,0.32,4,4,1,2,0,2,34,2,1,77,NA
+63986,7,2,1,75,NA,1,1,2,NA,NA,2,NA,2,1,9,NA,1,1,NA,2,2,2,1,2,2,2,1,2,NA,16607.774808,17550.450209,1,101,3,3,0.93,2,2,0,0,2,1,75,2,1,1,1
+63987,7,2,1,0,1,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6238.040326,6448.58866,1,93,10,10,2.91,4,4,2,0,0,2,27,1,5,1,4
+63988,7,2,1,4,NA,1,1,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16775.083123,16797.750213,2,98,2,2,0.27,4,4,2,0,0,2,20,2,2,6,NA
+63989,7,2,2,53,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,31785.728924,32871.467684,1,101,3,3,0.92,2,2,0,1,0,2,53,1,5,3,NA
+63990,7,2,1,17,NA,5,6,2,17,204,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,13,13,NA,3,3,0,2,0,2,41,2,3,4,NA
+63991,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,22707.329726,23247.668083,2,97,2,2,0.27,4,4,0,2,0,1,51,1,2,4,NA
+63992,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19696.958879,19993.739669,2,95,1,1,0.4,1,1,0,0,0,1,44,1,4,5,NA
+63993,7,2,1,8,NA,1,1,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14007.413517,3,92,7,7,1.41,5,5,1,2,0,1,40,1,3,1,4
+63994,7,2,2,9,NA,4,4,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,13050.526646,2,101,1,1,0.21,4,4,1,2,0,2,26,1,3,5,NA
+63995,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,16851.334496,16987.280633,2,95,1,1,0,1,1,0,0,0,1,56,1,2,2,NA
+63996,7,2,1,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,19478.738759,23024.340691,1,90,7,7,2.1,3,3,0,1,0,1,35,1,3,6,NA
+63997,7,2,2,79,NA,3,3,2,NA,NA,2,NA,2,2,6,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,27102.582153,29013.853259,1,93,13,13,NA,1,1,0,0,1,2,79,2,1,2,NA
+63998,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,29448.834066,32687.247837,1,97,9,9,5,1,1,0,0,0,2,48,1,5,5,NA
+63999,7,2,2,44,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,14254.989855,14385.318229,1,103,15,15,3.7,5,5,0,2,1,1,55,1,5,1,5
+64000,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,17879.023129,20112.694262,2,90,NA,NA,NA,1,1,0,0,0,1,31,1,3,5,NA
+64001,7,2,1,7,NA,4,4,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13423.881856,14062.200951,2,101,5,5,0.89,4,4,1,1,1,2,38,1,4,77,NA
+64002,7,2,1,80,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,31039.556381,33443.834712,1,96,15,15,5,2,2,0,0,2,2,80,1,5,1,5
+64003,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,138322.767578,143283.052066,1,95,14,14,5,2,2,0,0,0,1,54,1,4,1,3
+64004,7,2,2,0,10,1,1,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6331.376224,6501.48724,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+64005,7,2,2,12,NA,1,1,1,12,152,NA,NA,2,2,3,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18515.058419,19512.090709,2,96,5,5,0.76,5,5,1,2,0,1,44,2,1,1,3
+64006,7,2,2,9,NA,5,7,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19608.444487,19549.535258,2,98,3,3,0.38,5,5,0,4,0,2,39,1,4,5,NA
+64007,7,1,1,0,7,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7530.249163,0,2,102,12,12,NA,6,6,1,0,0,2,53,1,4,1,1
+64008,7,2,1,4,NA,4,4,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8123.332359,8458.989525,1,96,14,14,2.58,6,6,2,2,0,1,40,2,4,1,4
+64009,7,2,1,32,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18666.270458,19921.991676,2,93,77,77,NA,2,2,0,0,0,2,28,2,5,1,5
+64010,7,1,1,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13345.162299,0,2,102,14,14,4.86,3,3,1,0,0,1,30,1,5,1,5
+64011,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,27118.033816,33710.198359,1,94,5,5,1.39,2,2,0,0,2,1,60,1,1,5,NA
+64012,7,2,2,16,NA,3,3,2,17,205,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27482.814969,29965.467716,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+64013,7,2,1,80,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,8606.120537,9222.025464,1,93,2,2,0.69,1,1,0,0,1,1,80,2,5,2,NA
+64014,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,26141.606772,25260.281919,1,92,9,6,2.24,3,1,0,0,0,2,21,NA,NA,5,NA
+64015,7,2,1,32,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,45691.377881,46792.745533,3,91,14,14,5,2,2,0,0,0,1,32,1,4,6,NA
+64016,7,2,2,0,8,2,2,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5234.186769,5271.49441,2,90,3,3,0.46,5,5,3,0,0,2,22,1,2,5,NA
+64017,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,128585.755579,132183.659765,2,97,9,9,4.08,2,2,0,0,0,2,24,1,4,1,3
+64018,7,2,1,35,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18289.793332,18322.51979,1,99,10,10,3.13,4,4,0,2,0,1,35,1,4,1,5
+64019,7,2,2,10,NA,2,2,2,10,123,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13154.122703,13626.512275,1,93,9,9,2.46,4,4,0,2,0,1,35,2,1,1,1
+64020,7,2,2,17,NA,5,6,2,17,209,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11225.761275,11659.847432,2,91,8,8,2.34,4,4,0,2,0,1,56,2,5,1,5
+64021,7,2,2,62,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13057.178942,13648.591881,1,102,10,10,4.42,2,2,0,0,2,2,62,1,5,1,4
+64022,7,2,2,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,14516.765165,13843.248577,3,90,15,15,4.34,4,4,0,0,1,1,65,1,3,1,4
+64023,7,2,1,80,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,14200.083364,15006.095575,2,98,99,99,NA,1,1,0,0,1,1,80,1,1,2,NA
+64024,7,2,2,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,94637.019439,96511.799704,1,94,4,4,1.71,1,1,0,0,1,2,62,1,4,3,NA
+64025,7,2,1,31,NA,3,3,1,NA,NA,2,NA,2,1,7,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,88274.617148,90802.847573,2,92,15,14,5,2,1,0,0,0,2,29,1,5,6,NA
+64026,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17207.598344,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+64027,7,2,2,5,NA,3,3,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25864.922734,27487.163063,1,95,7,2,0.35,5,4,1,2,0,1,26,1,4,6,NA
+64028,7,2,1,11,NA,2,2,2,11,142,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,9390.522479,9862.720437,2,90,1,1,0.14,2,2,0,1,0,2,36,1,3,3,NA
+64029,7,2,1,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,39915.513053,41236.907607,2,98,7,7,1.53,5,5,0,0,0,2,48,1,3,5,NA
+64030,7,2,1,6,NA,2,2,1,6,78,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14821.597351,14686.992722,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+64031,7,2,2,19,NA,1,1,1,19,233,2,NA,1,1,NA,13,NA,NA,NA,2,2,2,2,2,2,1,2,2,NA,16781.078148,17658.768186,2,103,77,77,NA,5,5,0,2,0,2,45,2,4,5,NA
+64032,7,2,1,46,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,36319.60275,35786.668567,1,103,8,8,4.48,1,1,0,0,0,1,46,1,4,3,NA
+64033,7,2,2,64,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,1,NA,1,2,1,1,2,2,1,2,1,3,14102.354333,14646.654863,3,91,7,7,2.16,3,3,1,0,1,2,36,2,5,1,NA
+64034,7,2,2,62,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,13473.304889,14578.166065,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+64035,7,2,1,8,NA,2,2,1,8,98,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,NA,15039.041447,15306.786696,3,91,5,5,0.89,4,4,0,2,2,1,61,2,3,1,4
+64036,7,2,1,69,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,28953.774534,28658.225591,1,98,6,5,1.93,2,1,0,0,2,1,69,1,4,6,NA
+64037,7,2,1,28,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,12538.868652,13740.340257,2,95,15,15,5,4,1,0,0,0,1,29,NA,NA,99,NA
+64038,7,2,1,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8325.511113,9180.237397,2,99,6,6,1.13,4,4,1,1,0,1,33,1,3,6,NA
+64039,7,2,2,0,4,4,4,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3689.415307,3729.699375,2,99,13,13,NA,4,4,1,0,0,2,50,1,2,1,9
+64040,7,2,1,31,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19185.622388,19060.093906,2,95,5,1,0,3,1,0,0,0,1,31,1,4,5,NA
+64041,7,2,1,9,NA,4,4,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10122.702296,10692.492997,2,100,77,77,NA,4,4,0,1,1,2,28,1,3,5,NA
+64042,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16905.961576,17944.772858,2,94,4,4,1.16,2,2,0,0,0,1,39,1,2,1,5
+64043,7,2,1,14,NA,4,4,2,14,177,NA,NA,2,2,4,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11523.287163,11546.069645,1,96,8,8,2.17,4,4,0,2,0,2,45,2,5,4,NA
+64044,7,2,2,8,NA,1,1,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,10870.942302,11918.469007,1,103,10,10,2.82,4,4,0,2,0,1,41,2,1,1,1
+64045,7,2,2,4,NA,3,3,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,52159.241328,54365.068635,1,100,6,6,1.18,5,5,2,2,0,2,40,1,5,3,NA
+64046,7,2,2,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,38500.994547,38409.930489,2,91,2,2,0.81,1,1,0,0,0,2,39,1,2,6,NA
+64047,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,58826.425292,61096.366503,1,102,8,8,2.42,4,4,0,2,0,2,34,1,4,1,3
+64048,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,150482.742079,150600.121666,2,98,8,8,3.67,2,2,0,0,0,2,54,1,4,3,NA
+64049,7,2,2,16,NA,1,1,1,16,195,NA,NA,2,2,4,10,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,15690.47168,16407.081535,1,102,5,5,0.62,7,7,1,3,0,1,49,2,2,1,1
+64050,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,35469.911999,45030.280097,2,91,3,3,0.86,2,2,0,0,0,2,41,1,4,1,3
+64051,7,2,1,21,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,52571.533452,58252.213979,1,93,4,4,0.92,3,3,0,0,1,1,60,NA,NA,1,4
+64052,7,2,1,60,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,10004.038848,10557.19122,2,102,8,8,4.13,1,1,0,0,1,1,60,2,1,4,NA
+64053,7,2,1,8,NA,1,1,2,8,102,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13285.093011,13346.253362,2,94,6,6,1.15,5,5,1,2,0,1,33,1,2,1,2
+64054,7,2,2,19,NA,4,4,2,19,233,2,NA,1,1,NA,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10484.6104,10912.740054,2,99,3,3,0.56,4,4,1,0,0,2,38,1,3,5,NA
+64055,7,2,2,59,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,16033.091438,16414.611167,1,96,4,4,1.02,3,3,0,0,0,2,59,1,3,2,NA
+64056,7,2,1,18,NA,3,3,2,18,221,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,34847.322278,35515.35986,2,91,6,6,1.26,5,5,0,1,2,2,80,1,4,2,NA
+64057,7,2,2,11,NA,3,3,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38662.840486,39387.514697,2,92,15,15,5,5,5,0,3,0,2,46,1,5,1,5
+64058,7,2,2,50,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15499.383981,15868.203732,2,99,5,5,1.26,3,3,1,0,0,2,50,2,3,5,NA
+64059,7,2,2,68,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,12886.419545,13879.729659,2,97,1,1,0.22,1,1,0,0,1,2,68,1,3,2,NA
+64060,7,2,2,2,NA,3,3,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27529.278041,30383.810541,1,92,5,5,1.05,3,3,1,1,0,2,35,1,4,5,NA
+64061,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,35126.205635,35329.988956,1,95,1,1,0.17,2,2,0,1,0,2,47,1,2,3,NA
+64062,7,2,2,10,NA,4,4,1,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9309.947844,9654.826166,2,98,14,14,3.36,4,4,0,2,0,1,37,1,4,1,4
+64063,7,2,2,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32161.692893,32052.891523,1,99,2,2,0.78,1,1,0,0,1,2,62,1,4,3,NA
+64064,7,2,2,46,NA,5,6,1,NA,NA,2,NA,2,2,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17668.324663,18093.708295,1,102,14,14,4.86,3,3,0,1,0,1,42,1,4,1,5
+64065,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,106195.998472,105923.202819,1,93,15,15,5,2,2,0,0,0,1,46,1,4,1,5
+64066,7,2,2,11,NA,2,2,1,11,142,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,15929.068964,16501.112136,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+64067,7,2,1,2,NA,2,2,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,12915.99793,13324.846215,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+64068,7,2,1,10,NA,1,1,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12231.897958,12180.037346,1,100,7,7,1.3,5,5,0,3,0,1,43,2,2,1,4
+64069,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,74050.783686,75666.677003,1,94,6,6,1.31,3,3,0,0,0,2,46,1,5,6,NA
+64070,7,2,2,7,NA,3,3,1,7,89,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,101,7,7,1.55,5,5,1,2,0,2,31,1,4,1,2
+64071,7,2,1,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,11074.62349,11161.252677,2,101,5,5,0.89,4,4,1,1,1,2,38,1,4,77,NA
+64072,7,2,2,50,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,1,2,2,1,2,2,2,21828.388259,21942.221447,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+64073,7,2,1,13,NA,4,4,1,13,159,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12972.249316,12937.057154,1,100,15,15,3.7,5,5,0,3,0,1,51,1,5,1,5
+64074,7,1,1,50,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16287.780872,0,2,100,6,6,2.04,2,2,0,0,0,2,50,1,2,1,3
+64075,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,35074.640491,1,94,2,2,0.42,3,3,0,0,0,2,52,1,4,1,1
+64076,7,2,2,1,21,3,3,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26071.535804,27174.107547,1,93,10,10,2.48,5,5,2,1,0,1,40,2,5,1,5
+64077,7,2,1,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,30489.729187,30783.113246,3,92,12,12,NA,3,2,0,0,0,1,45,1,3,1,3
+64078,7,2,1,17,NA,4,4,1,17,206,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,14,14,4.03,4,4,0,1,0,2,40,1,5,1,5
+64079,7,2,2,25,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,41328.871311,42415.982935,2,93,9,9,2.6,4,4,0,0,0,2,58,2,4,4,NA
+64080,7,2,1,0,3,2,2,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7004.278208,7004.40731,1,93,5,5,1.3,3,3,1,1,0,2,28,2,4,5,NA
+64081,7,2,1,11,NA,4,4,2,11,143,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12176.538896,12387.770281,2,97,2,2,0.3,4,4,0,2,0,1,42,1,2,6,NA
+64082,7,2,1,64,NA,3,3,2,NA,NA,2,NA,2,2,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,167711.394252,169284.299948,1,101,10,10,4.63,2,2,0,0,1,1,64,2,3,1,4
+64083,7,2,2,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,30634.82105,30961.983694,1,95,7,7,2.16,3,3,0,0,1,1,45,1,3,1,4
+64084,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,33810.591598,40546.042621,3,91,13,13,NA,3,3,0,1,1,1,80,1,3,2,NA
+64085,7,2,2,13,NA,5,6,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10164.140113,10557.174707,3,91,15,15,5,4,4,0,2,0,1,44,2,5,1,5
+64086,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18040.833018,19102.925896,1,96,9,9,4.1,2,2,0,0,0,2,45,2,5,1,5
+64087,7,2,1,18,NA,1,1,2,18,225,2,NA,2,2,5,14,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,20398.562455,20186.553847,3,92,8,8,2.01,4,4,1,0,0,2,49,2,5,4,NA
+64088,7,2,2,9,NA,3,3,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24832.585001,24642.096114,2,101,3,3,0.6,3,3,0,2,0,1,39,1,4,4,NA
+64089,7,2,2,54,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16567.527819,16240.172481,1,96,15,15,5,3,3,0,0,0,1,55,1,4,1,5
+64090,7,2,1,18,NA,4,4,2,18,218,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,14,14,3.47,4,4,0,0,0,2,45,1,4,1,4
+64091,7,2,1,53,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12274.966216,12700.371312,3,90,5,5,1.05,3,3,0,0,0,1,53,2,1,1,2
+64092,7,2,1,18,NA,3,3,2,18,220,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,71458.892941,71654.216305,2,94,3,3,0.54,4,4,0,1,0,2,48,1,3,1,3
+64093,7,2,2,8,NA,5,6,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,1,1,2,1,1,2,2,NA,10387.513218,10891.034934,1,92,15,15,5,4,4,1,1,0,1,38,2,5,1,5
+64094,7,2,1,0,8,4,4,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7882.953266,8719.306286,2,96,7,7,1.79,4,4,2,0,0,2,49,1,3,1,3
+64095,7,2,1,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,113559.363135,118892.640077,2,102,14,14,4.93,3,3,0,1,0,1,37,1,5,1,5
+64096,7,2,1,10,NA,5,6,2,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,8246.426933,8701.859504,1,99,7,5,1.84,2,1,0,1,0,1,47,2,4,5,NA
+64097,7,2,2,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9139.784234,9841.162002,2,100,2,2,0.33,5,5,0,4,0,2,27,1,3,5,NA
+64098,7,2,1,54,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19374.410926,19463.695548,1,96,4,4,0.65,4,4,0,0,0,1,19,1,4,NA,NA
+64099,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,90234.245265,93002.542962,2,101,7,7,3.9,1,1,0,0,1,2,71,1,5,3,NA
+64100,7,1,1,5,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58116.402634,0,1,97,15,15,5,5,5,2,0,1,1,43,1,5,1,5
+64101,7,2,2,0,5,4,4,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3861.876549,4016.370213,1,96,8,8,1.61,6,6,3,0,0,1,33,2,5,1,4
+64102,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,9240.841805,10026.588852,2,95,13,13,NA,2,2,0,0,1,1,56,1,9,1,1
+64103,7,2,2,12,NA,5,6,1,12,148,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9505.147857,9693.556382,1,92,14,14,2.42,6,6,1,3,0,1,30,1,4,6,NA
+64104,7,2,2,4,NA,4,4,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8340.858072,8802.325948,2,103,6,6,1.11,5,5,1,2,0,2,36,1,4,5,NA
+64105,7,2,2,8,NA,1,1,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14042.378151,15260.975241,1,102,14,14,4.03,4,4,0,2,0,1,30,1,5,6,NA
+64106,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,68184.631488,70137.474393,2,95,4,4,1.34,1,1,0,0,0,1,34,1,4,5,NA
+64107,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,65891.955175,72165.622258,1,95,7,7,2.38,2,2,0,0,2,1,80,1,3,1,4
+64108,7,2,2,2,NA,3,3,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18127.266286,20006.89679,2,95,6,6,1.08,4,4,1,1,0,1,39,1,4,1,4
+64109,7,1,2,45,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,80232.348161,0,1,99,77,77,NA,4,4,0,2,0,2,45,1,3,1,NA
+64110,7,2,2,6,NA,4,4,2,6,81,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7282.523598,7776.514874,2,99,3,3,0.93,2,2,0,1,0,2,27,1,4,3,NA
+64111,7,2,2,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,28047.519499,31779.09658,2,90,1,1,0.14,1,1,0,0,0,2,33,1,2,5,NA
+64112,7,2,1,14,NA,2,2,1,14,179,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18635.323223,19040.145288,2,103,12,12,NA,4,4,0,1,0,2,50,2,3,1,4
+64113,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,38808.357492,1,95,3,3,0.96,1,1,0,0,1,2,80,1,4,2,NA
+64114,7,2,1,9,NA,5,6,1,9,111,NA,NA,2,1,3,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5769.526317,6088.16495,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+64115,7,2,1,8,NA,4,4,2,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7370.805738,7721.29497,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+64116,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,35276.751365,1,94,7,7,1.65,5,4,0,0,0,1,46,1,4,1,4
+64117,7,2,2,16,NA,3,3,2,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,60694.411883,64307.712903,1,99,15,15,4.47,4,4,0,2,0,2,43,1,5,1,5
+64118,7,2,2,0,3,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7782.311913,7756.671982,1,94,7,7,1.56,4,4,2,0,0,1,21,1,4,1,4
+64119,7,2,2,6,NA,3,3,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,98,15,15,5,5,5,0,3,0,2,41,1,5,6,NA
+64120,7,2,1,77,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,4,NA,2,2,2,2,2,2,1,2,2,NA,15301.031416,15853.994551,1,90,99,99,NA,5,5,0,2,1,2,50,2,4,4,NA
+64121,7,2,2,34,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,27127.983961,27837.333201,2,90,7,7,1.66,4,4,0,3,0,2,34,1,5,3,NA
+64122,7,2,2,67,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,1,2,2,2,15876.871857,17178.834476,2,102,NA,13,NA,2,1,0,0,1,1,50,2,1,6,NA
+64123,7,2,2,1,21,1,1,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12871.484115,14210.463876,2,102,3,3,0.45,4,4,2,0,0,1,21,2,2,6,NA
+64124,7,2,1,6,NA,4,4,1,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9261.557132,9782.87535,2,100,8,8,1.8,5,5,0,3,0,2,43,1,3,1,3
+64125,7,1,1,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,55061.722794,0,2,95,15,15,4.63,5,5,1,2,0,1,32,1,4,1,4
+64126,7,2,2,7,NA,1,1,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,19059.339877,19259.716076,1,95,13,13,NA,5,5,1,2,0,2,34,2,1,1,1
+64127,7,2,2,69,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18695.172864,19350.637044,2,102,7,7,1.68,5,5,0,0,3,1,70,2,4,1,4
+64128,7,2,1,59,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,124170.603852,126212.047685,1,92,6,6,2.39,1,1,0,0,0,1,59,1,4,3,NA
+64129,7,2,2,7,NA,4,4,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7588.605543,8170.947433,1,99,4,4,0.53,7,7,3,1,0,2,26,1,1,5,NA
+64130,7,2,1,69,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,2,1,NA,1,2,1,1,2,1,1,2,1,1,13977.762704,14745.704028,1,92,77,77,NA,4,4,0,0,2,1,59,2,5,1,5
+64131,7,2,2,79,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,36067.495928,37024.300659,1,95,5,5,1.36,2,2,0,0,1,2,79,1,3,2,NA
+64132,7,2,2,41,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20303.639991,21936.687597,1,97,15,15,4.07,5,5,0,3,0,1,42,2,5,1,5
+64133,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,57151.56429,59304.909373,2,95,8,8,2.17,4,4,1,1,0,1,43,1,4,1,5
+64134,7,2,2,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,29670.405171,29084.152013,2,101,10,10,3.89,3,3,0,1,0,2,49,1,4,1,3
+64135,7,2,2,23,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,11206.329484,11903.241195,2,103,15,15,5,3,3,0,0,0,2,52,2,4,1,5
+64136,7,2,1,9,NA,1,1,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12624.747427,12616.351223,2,92,15,15,3.37,7,7,0,4,0,1,42,2,3,1,1
+64137,7,2,2,72,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,1,2,NA,1,2,1,1,2,1,1,2,1,NA,10125.392704,11197.313648,2,91,12,12,NA,7,6,0,4,2,2,72,2,1,2,NA
+64138,7,2,1,36,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,20071.705576,20337.925283,3,91,14,14,5,2,1,0,0,0,1,36,1,5,6,NA
+64139,7,2,2,32,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,6,2,2,2,2,1,2,2,NA,NA,NA,NA,34401.268523,35501.848089,2,97,4,4,0.67,4,4,0,2,0,1,39,2,2,6,NA
+64140,7,2,2,16,NA,4,4,1,16,192,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12531.903464,12935.738352,2,100,5,5,1.07,4,4,0,1,0,2,36,1,3,5,NA
+64141,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,26344.362464,29482.177939,2,95,4,4,1.46,1,1,0,0,1,1,80,1,1,2,NA
+64142,7,2,1,73,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,14537.797299,15440.503889,3,91,77,77,NA,2,2,0,0,2,1,73,1,5,1,5
+64143,7,2,1,58,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,25737.628192,27045.174332,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+64144,7,2,2,2,NA,5,7,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9254.729183,9958.191837,2,98,15,15,4.17,6,6,1,1,0,2,40,1,4,1,4
+64145,7,2,1,2,NA,1,1,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,1,1,0.19,3,3,2,0,0,2,31,1,4,2,NA
+64146,7,2,2,47,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20303.639991,21936.687597,1,97,15,15,5,6,6,0,3,0,1,47,1,5,1,5
+64147,7,2,2,34,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,6,2,2,2,2,2,2,2,1,2,2,2,45655.090694,44423.220436,3,92,3,3,0.51,5,5,1,2,0,2,34,2,1,6,NA
+64148,7,2,1,2,NA,4,4,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6580.937346,6661.20735,2,101,7,7,2.16,3,3,1,0,1,2,64,1,3,2,NA
+64149,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,108504.032354,117548.425574,1,92,6,6,1.57,3,3,0,1,0,1,29,1,4,6,NA
+64150,7,2,1,8,NA,4,4,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12839.999482,13062.740187,2,96,8,8,1.72,5,5,0,3,0,1,39,1,5,1,4
+64151,7,2,1,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,124820.027137,131239.54259,1,91,10,10,3.51,3,3,0,0,1,1,21,1,4,5,NA
+64152,7,2,1,14,NA,4,4,1,14,173,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16261.995423,16421.244198,2,96,5,5,1.13,3,3,1,1,0,2,31,1,3,5,NA
+64153,7,2,2,11,NA,4,4,2,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6979.847147,8479.043057,1,99,7,7,1.53,5,5,0,3,0,1,39,1,3,1,3
+64154,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,19675.36066,19726.184442,2,96,14,14,4.26,3,3,0,0,0,1,20,1,4,5,NA
+64155,7,2,1,11,NA,4,4,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11543.27965,11681.665542,1,98,1,1,0.03,3,3,0,2,0,2,38,1,4,5,NA
+64156,7,2,2,8,NA,5,6,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5142.091355,5516.602825,3,91,15,15,5,3,3,0,1,0,2,44,2,5,1,5
+64157,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,136832.42119,141624.96623,2,91,12,5,2.03,2,1,0,0,2,2,68,1,4,6,NA
+64158,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17095.959542,2,97,7,7,3.67,1,1,0,0,0,1,27,1,4,1,NA
+64159,7,2,1,75,NA,1,1,2,NA,NA,2,NA,2,2,9,NA,1,1,NA,2,2,2,1,2,2,2,2,2,NA,17161.371047,18002.626237,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+64160,7,2,1,13,NA,5,6,1,13,167,NA,NA,2,1,2,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11171.449402,11936.033174,1,92,7,7,1.56,4,4,0,2,0,2,38,2,4,6,NA
+64161,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,1,96,15,15,5,2,2,0,0,0,2,51,1,5,5,NA
+64162,7,2,1,18,NA,3,3,2,18,222,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,35639.306876,36153.844915,2,91,99,99,NA,3,3,0,0,0,1,40,NA,NA,1,4
+64163,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12262.33683,12778.149281,2,100,15,15,5,3,3,0,1,0,1,48,2,5,1,5
+64164,7,2,1,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19440.793325,20264.695737,2,95,15,10,3.67,5,3,0,0,0,1,47,1,5,1,3
+64165,7,2,2,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,20495.125801,20710.596821,2,98,3,3,0.4,6,6,1,2,0,2,29,2,1,4,NA
+64166,7,2,1,7,NA,2,2,2,7,92,NA,NA,2,2,3,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10490.055059,10676.813132,3,90,4,4,0.63,5,5,0,3,0,1,45,2,4,1,4
+64167,7,2,2,21,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16929.836231,17652.770039,2,101,1,1,0.14,2,1,0,0,0,2,21,2,4,5,NA
+64168,7,2,1,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,96844.935107,100539.350089,2,98,10,10,4.42,2,2,0,0,0,1,31,1,4,1,3
+64169,7,2,1,17,NA,3,3,1,17,207,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,29202.587704,29754.083616,3,92,6,6,0.74,7,7,2,1,0,2,46,1,2,1,4
+64170,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,19337.563688,19353.557903,1,96,6,6,1.21,4,4,2,0,0,1,24,1,4,1,3
+64171,7,2,1,11,NA,3,3,2,11,135,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25604.034863,28346.339599,1,95,10,6,1.34,5,4,1,2,0,1,32,1,3,6,NA
+64172,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,10455.739374,11701.098067,1,101,3,3,0.86,2,2,0,0,2,2,80,1,2,1,1
+64173,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,17430.913214,17517.807291,1,90,5,5,1.06,4,4,0,2,0,1,53,1,3,1,4
+64174,7,2,2,18,NA,2,2,2,18,222,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16189.692833,16766.382859,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+64175,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,39416.921733,1,95,2,2,0.83,1,1,0,0,1,2,80,1,2,2,NA
+64176,7,2,1,72,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,10160.645851,10359.669924,1,96,9,9,3.97,2,2,0,0,2,1,72,1,4,1,5
+64177,7,2,1,2,NA,3,3,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,22388.62869,26233.531918,2,97,1,1,0.21,4,4,2,0,0,2,34,2,1,1,2
+64178,7,2,1,7,NA,5,6,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6765.693703,7164.978412,2,92,15,15,5,3,3,0,1,0,1,45,2,4,1,4
+64179,7,2,1,52,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,25118.469449,25612.281104,2,98,5,5,1.63,2,2,0,0,0,2,53,2,1,1,1
+64180,7,2,2,31,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,4,1,2,2,2,2,1,2,2,NA,NA,NA,NA,38184.257672,39182.709093,2,91,9,9,2.6,4,4,1,1,0,2,31,2,4,1,5
+64181,7,2,2,56,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,23409.465713,23644.84324,2,97,14,9,5,2,1,0,0,0,2,56,1,3,6,NA
+64182,7,2,2,67,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,90638.027317,92118.903864,2,100,6,6,2.04,2,2,0,0,2,1,74,1,4,1,3
+64183,7,2,2,18,NA,1,1,1,18,223,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27070.679378,27847.54398,1,92,7,7,1.48,5,5,0,1,0,1,42,1,5,1,4
+64184,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,36438.408188,36780.151097,1,101,3,3,0.66,2,2,0,0,1,1,67,1,2,3,NA
+64185,7,2,2,14,NA,5,6,2,14,170,NA,NA,2,1,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,7,7,1.48,5,5,1,2,0,1,40,2,5,1,4
+64186,7,2,1,63,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,1,2,2,1,2,2,1,9068.437099,9499.757798,2,93,9,9,3.97,2,2,0,0,1,2,57,2,3,1,1
+64187,7,2,1,6,NA,3,3,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,66003.625505,68276.88064,1,101,14,14,3.15,5,5,2,1,0,1,35,1,4,1,5
+64188,7,1,1,65,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20187.172902,0,2,100,4,4,1.29,2,2,0,0,2,1,65,1,3,1,3
+64189,7,2,1,2,NA,5,6,2,2,33,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6208.192797,6963.167461,1,91,10,10,3.22,4,4,1,1,0,1,38,2,5,1,5
+64190,7,2,1,0,0,1,1,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7284.164858,7284.299119,2,98,4,4,0.67,4,4,1,0,0,2,40,1,3,3,NA
+64191,7,2,1,72,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,3,1,NA,2,2,2,1,2,2,2,2,1,NA,14159.984279,14406.82491,2,93,4,4,0.99,2,2,0,0,2,1,72,2,3,1,4
+64192,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,118336.754883,2,92,15,15,5,6,6,2,0,0,1,18,1,4,NA,NA
+64193,7,2,1,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16058.989596,16964.610342,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+64194,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10598.2543,1,99,15,15,5,2,2,0,0,2,1,67,1,5,1,5
+64195,7,2,1,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19008.083201,19399.303007,2,97,12,12,NA,3,3,0,0,0,1,33,1,4,5,NA
+64196,7,2,2,9,NA,5,6,1,9,115,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9211.293091,9657.799007,1,92,14,14,3.47,4,4,0,2,0,1,37,1,5,1,5
+64197,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,1,2,1,2,2,1,1,2,NA,7360.303891,7756.453804,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+64198,7,2,2,76,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,14275.35524,2,90,3,3,0.92,1,1,0,0,1,2,76,1,3,2,NA
+64199,7,2,2,2,NA,2,2,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8512.480199,8626.96297,2,90,3,3,0.68,2,2,1,0,0,2,23,1,1,5,NA
+64200,7,2,2,3,NA,2,2,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15233.991221,16407.076559,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+64201,7,2,1,20,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,9177.295801,9548.31812,2,92,77,77,NA,4,4,0,0,1,1,20,1,2,5,NA
+64202,7,2,1,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9298.536372,9483.411197,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+64203,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,15262.313583,16475.41377,1,101,2,2,0.47,3,3,1,0,0,1,35,1,2,6,NA
+64204,7,2,2,0,7,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16547.193167,16560.896108,1,97,12,12,NA,5,5,3,0,0,2,33,1,5,1,5
+64205,7,2,2,1,12,4,4,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7556.585831,7751.634132,1,96,77,77,NA,3,3,1,0,0,2,39,1,7,5,NA
+64206,7,2,2,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,12886.419545,13879.729659,2,101,3,3,1.19,1,1,0,0,1,2,69,1,2,2,NA
+64207,7,2,2,28,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,2,1,2,1,2,2,1,2,2,1,2,1,NA,45798.520132,47350.399021,2,91,14,14,4.71,3,3,0,0,0,1,28,2,1,1,2
+64208,7,2,1,15,NA,1,1,1,15,182,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,14432.845547,14415.592261,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+64209,7,2,1,11,NA,2,2,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14820.807433,14943.02937,2,102,5,5,0.89,4,4,0,3,0,2,44,2,2,4,NA
+64210,7,2,1,3,NA,4,4,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8414.846149,8934.299427,1,90,7,7,1.3,5,5,1,2,1,2,62,1,2,2,NA
+64211,7,2,2,1,20,1,1,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11582.174418,11621.723611,2,102,8,8,1.28,7,7,1,3,0,1,39,2,1,1,3
+64212,7,2,2,8,NA,5,7,1,8,107,NA,NA,2,1,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8128.755281,8532.236543,1,98,9,9,2.39,4,4,0,2,0,2,48,1,5,1,5
+64213,7,2,1,37,NA,1,1,2,NA,NA,2,NA,2,2,77,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,38610.51698,2,94,77,77,NA,3,3,0,1,0,2,42,2,2,6,NA
+64214,7,2,1,18,NA,1,1,1,18,227,2,NA,2,2,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,2,24228.858782,24599.205366,2,102,5,5,1.3,3,3,0,0,0,1,42,NA,NA,1,NA
+64215,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,16494.288293,17728.004854,2,101,2,2,0.77,1,1,0,0,1,2,80,1,1,2,NA
+64216,7,2,2,2,24,1,1,1,2,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,12710.400836,2,98,7,7,2.16,3,3,1,0,0,2,26,1,3,1,3
+64217,7,2,1,53,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,1,2,2,2,2,2,2,2,20592.227875,20667.110628,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+64218,7,2,1,53,NA,4,4,2,NA,NA,1,2,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,14699.320127,15747.67742,2,90,15,15,5,4,4,1,1,0,1,53,2,5,1,5
+64219,7,2,2,80,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,1,5,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,13555.744544,2,90,14,14,4.25,4,4,0,2,1,2,45,2,5,5,NA
+64220,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,99381.891022,101559.2006,2,92,15,6,2.3,4,1,0,0,0,1,27,NA,NA,5,NA
+64221,7,2,1,19,NA,1,1,1,19,239,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,20634.3158,20949.718537,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+64222,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105902.251482,110971.270582,3,91,10,10,3.51,3,3,0,1,0,1,39,1,5,1,3
+64223,7,2,2,0,1,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20239.765739,19690.008753,2,94,9,9,2.51,4,4,2,0,0,1,30,2,4,1,4
+64224,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,33292.204257,2,101,2,2,0.64,1,1,0,0,0,2,22,1,4,5,NA
+64225,7,2,1,8,NA,3,3,2,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,50571.965712,52018.055374,2,95,15,15,4.63,5,5,1,2,0,2,36,1,5,1,3
+64226,7,2,1,73,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,73758.836891,83728.230941,2,91,3,3,1.25,1,1,0,0,1,1,73,1,3,5,NA
+64227,7,2,2,59,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,4,5,NA,2,2,2,2,2,2,1,2,1,2,24553.193015,24681.235828,2,93,3,3,0.9,1,1,0,0,0,2,59,2,4,5,NA
+64228,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,146181.198007,184923.777793,2,91,15,15,5,2,1,0,0,0,1,44,1,3,5,NA
+64229,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,32609.153024,32528.629595,1,98,3,3,0.73,3,3,0,0,0,1,52,1,4,1,3
+64230,7,2,2,2,NA,4,4,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8296.101892,8926.698178,2,96,7,7,1.49,5,5,2,1,0,1,51,1,5,1,3
+64231,7,2,1,68,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7820.017944,8252.409456,2,98,7,7,2.1,3,3,0,0,1,1,68,1,2,1,9
+64232,7,2,1,0,6,3,3,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23912.171644,25599.009772,1,94,9,9,3.14,3,3,1,0,0,1,28,1,5,1,5
+64233,7,2,1,14,NA,4,4,1,15,180,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13731.625553,14361.066702,1,100,6,6,1.13,4,4,0,3,0,2,32,1,3,5,NA
+64234,7,2,1,29,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,59682.963348,63262.110969,2,102,15,15,3.92,5,5,0,0,0,1,19,1,4,NA,NA
+64235,7,2,2,56,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,2,2,2,2,22224.73066,22874.201587,2,94,12,12,NA,2,2,0,0,0,1,46,1,3,1,1
+64236,7,2,1,5,NA,1,1,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11388.091908,11522.723578,1,103,8,8,1.85,5,5,2,1,0,2,25,2,2,1,2
+64237,7,2,2,16,NA,3,3,2,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,38400.791741,38900.548719,1,92,4,4,0.5,6,6,0,3,0,2,41,1,4,1,NA
+64238,7,2,1,0,2,5,6,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5024.464768,5303.683185,2,92,99,77,NA,7,3,3,3,1,1,61,2,1,1,3
+64239,7,2,1,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5396.442999,5408.183432,1,99,4,4,0.53,7,7,3,1,0,2,26,1,1,5,NA
+64240,7,2,1,19,NA,1,1,1,19,233,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,23389.620035,23146.524434,1,94,10,10,2.94,4,4,0,2,0,2,52,1,5,2,NA
+64241,7,2,2,0,9,3,3,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6423.658168,6554.989142,2,96,3,3,0.53,5,5,3,0,0,2,26,1,4,1,4
+64242,7,2,2,16,NA,3,3,1,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,82145.180523,84080.796514,1,98,15,15,5,5,5,0,3,0,2,44,1,5,1,5
+64243,7,2,2,7,NA,3,3,2,7,85,NA,NA,2,1,1,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,52455.416493,51758.830416,1,93,15,15,5,4,4,0,2,0,2,42,1,5,1,NA
+64244,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,30888.99461,30919.326232,1,99,2,2,0.61,2,2,0,0,0,1,46,1,5,5,NA
+64245,7,2,1,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,120041.937453,119895.563372,1,98,9,9,2.88,6,3,1,3,0,1,51,1,2,1,3
+64246,7,2,1,14,NA,2,2,2,14,172,NA,NA,2,1,2,7,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,26616.794908,27940.371621,2,91,7,7,2.64,2,2,0,1,0,1,33,2,3,1,NA
+64247,7,2,1,42,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,43470.92932,47764.563168,1,92,5,5,0.87,4,4,0,2,0,1,42,2,1,1,4
+64248,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,67727.881967,67936.122757,2,99,15,15,5,1,1,0,0,0,2,38,1,5,5,NA
+64249,7,1,1,11,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57057.523607,0,1,101,15,15,5,4,4,0,2,0,2,40,1,4,1,3
+64250,7,2,2,3,NA,1,1,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13366.393396,13791.68675,2,94,7,7,1.79,4,4,1,1,0,2,32,1,4,1,4
+64251,7,1,1,4,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7232.559351,0,2,103,14,14,3.86,4,4,2,0,0,2,37,2,5,1,NA
+64252,7,2,1,61,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,3,1,NA,2,2,2,1,2,2,1,2,2,2,11019.434708,11234.197915,3,91,5,5,0.89,4,4,0,2,2,1,61,2,3,1,4
+64253,7,2,1,13,NA,5,7,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5388.361335,5758.757236,3,91,15,15,4.47,4,4,0,3,0,2,44,2,5,1,NA
+64254,7,2,1,0,8,1,1,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6910.953528,6911.08091,2,96,4,4,0.81,3,3,1,0,0,2,37,2,5,1,3
+64255,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,1,2,2,3,13859.220514,14321.474311,1,98,12,14,5,3,1,0,0,0,1,24,1,4,5,NA
+64256,7,2,2,67,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,11523.037911,11989.050618,2,97,6,6,2.31,2,2,0,0,1,2,67,1,3,2,NA
+64257,7,2,2,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,53047.531099,53521.812192,1,99,2,1,0.18,2,1,0,0,0,2,24,1,5,5,NA
+64258,7,2,2,13,NA,5,6,1,13,163,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12918.122917,13476.480232,1,92,10,10,2.82,4,4,0,2,0,2,48,2,5,1,5
+64259,7,2,2,35,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,2,1,2,2,2,2,1,2,2,2,2,2,2,35353.005268,36099.35979,2,94,6,6,1.34,4,4,0,2,0,1,37,2,4,1,2
+64260,7,2,2,17,NA,4,4,2,17,214,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11438.127668,11466.298034,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+64261,7,2,1,67,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,1,1,2,2,1,2,1,NA,13639.016686,14339.363246,2,102,15,15,5,5,5,1,0,2,1,30,1,4,1,5
+64262,7,2,2,6,NA,4,4,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10225.297464,10592.094499,1,93,2,2,0.32,3,3,0,1,0,1,25,1,3,6,NA
+64263,7,2,2,62,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,2,2,2,2,9203.46153,9587.572248,2,97,6,6,1.7,2,2,0,0,1,2,62,2,5,1,2
+64264,7,2,1,14,NA,4,4,2,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8364.097643,8526.060452,2,90,6,6,0.96,5,5,0,1,0,1,55,1,4,6,NA
+64265,7,2,1,48,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,27087.386131,27212.215064,1,101,6,6,1.43,4,4,0,1,2,2,72,1,2,1,NA
+64266,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,90274.900586,93718.683628,2,99,15,15,5,2,1,0,0,0,1,31,1,5,6,NA
+64267,7,2,2,10,NA,4,4,2,10,129,NA,NA,2,2,3,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7136.421849,7524.490259,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+64268,7,2,2,11,NA,4,4,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12476.900945,13434.365665,2,101,2,2,0.38,3,3,0,2,0,2,56,1,3,2,NA
+64269,7,1,1,61,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9611.684527,0,1,98,7,7,1.52,4,4,0,1,2,1,61,1,4,1,2
+64270,7,2,2,47,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27934.372045,28667.797605,1,101,5,5,1.15,3,3,0,1,0,1,49,1,3,1,4
+64271,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,45549.853584,49025.946555,3,91,3,3,0.76,3,3,0,1,0,2,24,1,4,6,NA
+64272,7,2,2,19,NA,4,4,2,19,239,2,NA,2,2,2,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12209.74498,12386.615954,2,90,6,6,1.34,4,4,1,0,0,1,38,2,4,6,NA
+64273,7,2,1,5,NA,4,4,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9242.457181,9624.356732,2,97,5,5,0.76,5,5,1,2,0,1,32,1,4,6,NA
+64274,7,2,1,0,3,1,1,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,7757.493251,7980.837709,2,98,3,3,0.4,6,6,1,2,0,2,29,2,1,4,NA
+64275,7,1,1,62,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,4,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,10288.337394,0,3,91,8,8,4.13,1,1,0,0,1,1,62,2,4,4,NA
+64276,7,2,1,50,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,23857.322871,23507.253435,1,103,1,1,0.03,3,3,0,0,0,1,50,1,2,3,NA
+64277,7,2,2,48,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,36924.381422,39366.592716,2,102,7,7,1.89,3,3,0,1,0,1,41,2,2,6,NA
+64278,7,2,2,18,NA,2,2,2,18,226,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13824.001771,14551.102227,2,90,5,5,1.19,3,3,0,0,0,2,50,2,4,4,NA
+64279,7,2,1,11,NA,3,3,2,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,52386.010737,55389.865547,3,91,14,14,3.4,4,4,0,2,0,1,40,1,4,1,4
+64280,7,2,1,3,NA,4,4,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7014.280192,7545.08247,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+64281,7,2,1,44,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,34000.722046,33736.711001,2,93,14,14,3.52,5,5,1,2,0,1,44,1,5,1,5
+64282,7,2,1,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6074.284591,6449.253661,2,99,3,3,0.32,6,6,2,1,1,2,59,1,4,1,NA
+64283,7,2,1,5,NA,4,4,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9035.668246,2,100,3,3,0.31,7,7,3,2,0,2,28,1,3,1,3
+64284,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,38808.357492,1,101,1,1,0.18,1,1,0,0,1,2,80,1,4,2,NA
+64285,7,2,1,7,NA,1,1,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16334.967392,17408.274993,1,95,5,5,1.5,2,2,0,1,0,1,47,1,4,2,NA
+64286,7,2,2,56,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16109.317112,16194.502254,1,103,15,15,5,2,2,0,0,1,2,56,2,5,1,5
+64287,7,2,1,7,NA,3,3,2,7,89,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24713.905595,26080.214484,1,98,6,6,0.97,7,7,1,2,0,1,49,1,2,1,2
+64288,7,2,2,41,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,31809.199771,32336.174545,2,103,9,9,3.97,2,2,0,0,1,1,61,1,5,1,3
+64289,7,2,2,0,3,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23207.538828,22577.170535,1,94,6,6,1.21,4,4,2,0,0,1,27,1,4,1,3
+64290,7,2,2,65,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10598.2543,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+64291,7,2,1,50,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,22238.49412,25006.164622,2,90,4,4,1.38,1,1,0,0,0,1,50,1,3,3,NA
+64292,7,2,1,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,21924.03349,22615.591521,3,92,2,2,0.65,2,2,0,0,1,2,80,NA,NA,2,NA
+64293,7,2,2,4,NA,3,3,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53166.229434,56500.79967,1,101,7,7,1.55,5,5,1,2,0,2,31,1,4,1,2
+64294,7,2,1,19,NA,2,2,2,19,234,2,NA,2,1,4,15,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,17149.002427,17445.048827,3,90,77,77,NA,4,3,0,0,0,1,45,2,3,3,NA
+64295,7,2,1,10,NA,3,3,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,NA,NA,NA,1,2,2,1,69959.431608,85622.357277,2,100,NA,NA,NA,5,5,1,2,0,1,36,NA,NA,3,NA
+64296,7,2,2,40,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,33716.655399,36032.83001,1,100,8,8,2.17,4,4,1,1,0,2,40,2,2,1,2
+64297,7,1,2,9,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15002.25457,0,1,91,7,7,2.72,2,2,0,1,0,2,31,1,2,5,NA
+64298,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,10991.458112,11482.21496,2,99,10,10,4.76,2,2,0,0,2,1,80,1,2,2,NA
+64299,7,2,1,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,95606.814645,96503.477057,2,98,15,15,5,2,2,0,0,1,2,52,1,4,1,4
+64300,7,1,2,6,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11035.074625,0,2,100,1,1,0.04,4,4,0,2,0,1,34,NA,NA,6,NA
+64301,7,2,2,73,NA,5,7,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,12813.709202,13258.116643,2,103,12,12,NA,2,2,0,0,2,1,73,2,5,1,4
+64302,7,2,1,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,136880.768184,139080.782402,1,94,15,15,4.95,4,4,0,0,2,1,72,1,3,1,3
+64303,7,1,2,77,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,81395.297807,0,1,92,4,4,1.61,1,1,0,0,1,2,77,1,3,2,NA
+64304,7,2,2,0,5,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8359.077295,8583.668456,3,92,5,5,0.81,5,5,3,0,0,2,23,1,4,5,NA
+64305,7,2,2,41,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,14831.744106,14468.205243,3,90,10,10,3.67,3,3,0,1,0,2,52,2,3,5,NA
+64306,7,2,2,2,NA,4,4,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7826.463896,8746.031629,1,96,6,6,1.21,4,4,2,0,0,1,24,1,4,1,3
+64307,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,102855.726146,106395.504024,1,101,7,7,1.74,4,4,1,0,0,1,24,NA,NA,1,4
+64308,7,2,2,31,NA,2,2,1,NA,NA,2,NA,2,2,77,NA,1,6,2,2,2,2,1,2,2,2,2,2,2,32097.38481,33124.257579,2,100,3,3,0.76,3,3,1,0,0,2,31,2,1,6,NA
+64309,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53541.401974,54961.757487,1,99,13,13,NA,2,2,0,0,2,1,80,1,2,1,3
+64310,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,163194.688032,170363.320524,1,97,15,15,5,2,2,0,0,1,2,49,1,5,1,3
+64311,7,2,1,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17636.923031,18334.929229,2,97,15,14,5,2,1,0,0,0,1,54,1,4,5,NA
+64312,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,42993.150248,46481.579201,1,92,7,7,2.64,2,2,0,0,2,2,68,1,4,1,4
+64313,7,2,1,65,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,3,6,NA,2,2,2,2,2,2,2,2,2,2,14067.170863,14736.245577,2,102,14,8,4.59,2,1,0,0,2,1,65,2,3,6,NA
+64314,7,2,1,6,NA,1,1,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13285.093011,13495.322435,2,94,6,6,1.5,4,4,0,2,0,1,44,2,2,1,2
+64315,7,2,2,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,147501.330925,150787.451601,1,101,6,6,1.31,3,3,0,0,1,2,80,1,1,2,NA
+64316,7,2,2,12,NA,4,4,1,12,155,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,6,6,1.16,4,4,0,3,0,2,36,1,4,4,NA
+64317,7,2,2,18,NA,3,3,1,18,222,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,32725.110579,34226.88515,2,92,3,2,0.64,2,1,0,0,0,2,18,1,4,NA,NA
+64318,7,2,2,14,NA,4,4,2,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13790.410898,14648.625582,1,93,7,7,1.79,4,4,0,2,0,1,53,2,4,1,4
+64319,7,2,2,14,NA,5,6,2,14,174,NA,NA,2,1,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8675.8731,8857.654394,1,96,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+64320,7,2,1,14,NA,4,4,2,14,171,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,18413.211403,2,101,13,3,0.64,5,4,0,3,1,2,62,1,1,2,NA
+64321,7,2,1,12,NA,3,3,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,90320.117415,93226.168932,1,100,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+64322,7,2,1,8,NA,3,3,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41342.668304,42524.849071,1,102,8,8,2.42,4,4,0,2,0,2,34,1,4,1,3
+64323,7,2,1,3,NA,1,1,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17865.135763,18076.339981,3,92,1,1,0,5,5,3,0,0,1,26,1,2,1,2
+64324,7,2,2,28,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,2,2,2,2,2,2,2,2,2,2,2,38364.674202,40512.730459,2,97,13,13,NA,3,3,0,1,0,1,28,2,2,1,1
+64325,7,2,1,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,1,2,1,2,2,1,2,2,NA,108408.375382,111473.582646,2,98,14,14,4.16,3,3,0,0,0,1,49,1,5,1,4
+64326,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,90303.174138,94631.352733,2,92,14,9,5,2,1,0,0,0,1,29,1,5,6,NA
+64327,7,2,1,9,NA,3,3,2,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30554.050397,31755.009616,1,97,3,3,0.93,2,2,0,1,0,2,34,1,4,5,NA
+64328,7,2,1,78,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,13523.396212,13874.942566,2,94,2,2,0.56,2,2,0,0,2,2,71,2,2,1,1
+64329,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,34049.658577,35975.208577,1,91,9,9,4.27,2,2,0,0,2,1,80,1,4,1,3
+64330,7,2,1,43,NA,5,7,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,129656.862785,1,101,15,15,5,4,4,0,2,0,1,43,1,4,1,5
+64331,7,2,1,16,NA,3,3,1,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,59545.101745,58762.542429,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+64332,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,17579.006909,17715.920575,2,103,2,1,0.43,2,1,0,0,0,1,20,1,3,6,NA
+64333,7,2,1,46,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,3,NA,1,2,2,1,2,2,1,2,2,3,20197.354438,20825.605262,2,96,5,5,1.84,2,1,0,0,1,1,46,2,3,3,NA
+64334,7,2,1,6,NA,2,2,2,6,80,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11477.372934,11658.996164,1,93,9,9,2.46,4,4,0,2,0,1,35,2,1,1,1
+64335,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,NA,NA,NA,1,2,2,1,43813.24867,46884.227528,1,98,NA,NA,NA,4,4,0,2,0,1,31,NA,NA,1,2
+64336,7,2,1,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,19869.518693,21796.765593,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+64337,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,42894.724338,43561.362107,1,98,3,3,0.93,2,2,0,0,0,1,21,1,4,5,NA
+64338,7,1,2,46,NA,1,1,NA,NA,NA,2,NA,2,2,6,NA,1,6,NA,2,2,2,1,2,2,NA,NA,NA,NA,40337.933888,0,2,94,5,5,0.65,6,6,0,2,0,1,53,NA,NA,6,NA
+64339,7,2,1,6,NA,2,2,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12604.990052,13551.632975,1,103,7,7,1.03,7,7,0,3,0,1,50,2,1,1,1
+64340,7,1,1,80,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9443.12441,0,1,99,77,77,NA,2,2,0,0,2,1,80,1,5,1,4
+64341,7,2,2,51,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,2,1,NA,2,2,2,1,2,2,2,2,1,2,23200.373382,24548.135184,2,93,7,7,2.1,3,3,0,1,0,2,51,2,2,1,NA
+64342,7,2,1,12,NA,3,3,2,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,103364.662244,103614.262032,1,98,14,14,3.15,5,5,0,3,0,1,34,1,4,1,4
+64343,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,84821.217572,87190.888968,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+64344,7,2,1,37,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,28213.419452,30402.035877,1,100,14,5,1.79,2,1,0,0,0,1,37,1,2,6,NA
+64345,7,2,1,21,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21763.209029,22385.504537,2,92,4,1,0.18,4,1,0,0,0,1,21,1,4,5,NA
+64346,7,2,1,3,NA,5,7,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8402.098771,9423.873047,3,91,14,14,4.32,3,3,1,0,0,1,31,2,3,1,4
+64347,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32838.149884,1,95,1,1,0,1,1,0,0,0,1,50,1,3,5,NA
+64348,7,2,2,13,NA,3,3,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,105891.533689,107879.366927,1,101,8,8,1.85,5,5,0,3,0,1,41,1,3,1,4
+64349,7,2,2,24,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,NA,NA,NA,1,2,2,1,11475.373333,11965.391974,2,92,12,NA,NA,7,1,0,0,2,1,53,2,3,1,3
+64350,7,2,1,12,NA,4,4,2,12,145,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,12116.15399,2,95,7,7,1.55,5,5,0,3,0,1,30,1,4,1,4
+64351,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18353.275855,19744.388768,1,91,4,4,0.81,4,4,1,1,0,1,32,1,4,6,NA
+64352,7,2,2,2,NA,4,4,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6085.457443,6422.142601,1,99,7,7,1.53,5,5,2,0,0,2,37,1,4,1,3
+64353,7,2,1,0,11,2,2,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7973.737542,8118.415368,1,91,7,7,1.66,4,4,2,0,0,1,32,2,5,1,4
+64354,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,1,2,NA,2,2,2,1,2,2,1,2,2,NA,21122.17432,23417.039872,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+64355,7,1,2,27,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,4,6,3,1,2,2,2,2,2,NA,NA,NA,NA,42583.505439,0,2,91,6,6,1.3,4,4,1,1,0,2,27,1,4,6,NA
+64356,7,2,1,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,41236.907607,2,98,7,7,1.33,6,6,0,3,0,1,31,1,3,6,NA
+64357,7,1,2,1,22,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9670.78354,0,2,93,15,15,5,4,4,2,0,0,1,34,1,5,1,5
+64358,7,2,1,5,NA,4,4,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,8005.528865,8250.049428,1,99,NA,NA,NA,4,4,1,1,0,1,42,NA,NA,1,NA
+64359,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,25815.880139,26556.735732,2,101,99,2,0.73,2,1,0,0,0,1,24,1,4,5,NA
+64360,7,2,2,54,NA,1,1,1,NA,NA,2,NA,2,2,99,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,21939.007884,22053.417944,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+64361,7,2,2,16,NA,5,6,2,16,199,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8301.172828,8662.397073,1,91,15,15,5,3,3,0,1,0,2,47,2,5,1,5
+64362,7,2,1,68,NA,4,4,2,NA,NA,2,NA,2,2,8,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,8491.292032,8823.40317,1,96,6,6,2.06,2,2,0,0,2,1,68,2,3,1,2
+64363,7,2,1,22,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,17701.114982,3,91,14,14,3.53,5,5,0,1,1,1,69,1,4,3,NA
+64364,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,38976.947883,43357.400647,1,99,12,12,NA,1,1,0,0,1,2,80,1,3,3,NA
+64365,7,2,2,5,NA,2,2,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14852.990935,14903.708855,3,92,7,7,0.93,7,7,1,3,0,2,20,1,3,1,1
+64366,7,2,1,4,NA,4,4,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10234.881417,10547.495238,2,97,2,2,0.33,4,4,2,1,0,2,34,1,2,5,NA
+64367,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20039.469886,21581.359058,1,97,15,15,5,4,4,1,0,0,2,39,2,5,1,5
+64368,7,2,1,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,8497.912951,8999.288803,2,100,2,2,0.77,1,1,0,0,1,1,70,1,1,3,NA
+64369,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,17824.805721,17767.952896,2,95,4,2,0.55,2,1,0,0,0,2,47,1,3,4,NA
+64370,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25890.025676,30408.154535,2,93,3,3,1.25,1,1,0,0,0,1,25,1,4,5,NA
+64371,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,1,96,15,9,4.92,2,1,0,0,0,2,55,1,4,5,NA
+64372,7,2,2,19,NA,4,4,2,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13176.946531,13078.313981,2,99,9,9,1.78,6,6,1,1,0,1,46,1,3,6,NA
+64373,7,2,2,6,NA,3,3,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,41790.228676,44382.642913,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+64374,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10700.939986,11250.682738,1,99,15,15,4.47,4,4,0,2,0,2,52,2,5,1,5
+64375,7,2,2,14,NA,2,2,2,14,175,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,5,5,0.8,5,5,0,3,0,2,40,2,1,5,NA
+64376,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,34296.120763,1,95,3,3,0.65,3,3,1,0,0,1,58,1,3,3,NA
+64377,7,2,2,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,19060.786733,19867.822877,2,97,5,5,0.92,5,5,0,3,0,2,54,1,3,2,NA
+64378,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,115391.113177,120921.742093,1,102,14,7,3.31,2,1,0,0,0,1,27,1,4,5,NA
+64379,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,30320.430859,2,101,4,3,1.15,2,1,0,0,0,2,22,1,4,5,NA
+64380,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,114993.808573,116714.079488,1,98,7,3,0.9,4,1,0,0,0,2,20,NA,NA,5,NA
+64381,7,2,2,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,148725.079159,151155.002758,2,94,15,15,5,2,2,0,0,2,1,63,1,5,1,5
+64382,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,6,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15533.829525,2,101,6,1,0.22,2,1,0,0,0,2,26,2,4,6,NA
+64383,7,2,2,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,19658.218913,18995.471706,1,99,13,13,NA,4,4,1,0,0,2,26,1,4,4,NA
+64384,7,2,1,0,5,3,3,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24315.386145,24328.135897,1,102,6,6,1.23,4,4,2,0,0,2,25,1,5,1,5
+64385,7,2,2,13,NA,5,6,1,13,160,NA,NA,2,1,3,6,NA,NA,NA,1,1,2,1,2,1,1,2,2,1,8878.081187,9740.503961,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+64386,7,2,2,9,NA,1,1,1,10,121,NA,NA,2,2,3,2,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,16397.644545,16823.243233,1,91,7,7,2.1,3,3,0,1,0,2,29,2,3,6,NA
+64387,7,2,1,71,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,5,1,NA,1,2,1,1,2,1,1,2,1,NA,9748.579573,10638.31203,3,90,9,9,2.93,3,3,0,0,2,1,71,2,5,1,4
+64388,7,2,1,69,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,8232.241159,8296.636338,2,98,5,5,2.2,1,1,0,0,1,1,69,1,4,3,NA
+64389,7,2,1,10,NA,1,1,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,9485.228453,10197.575258,1,103,10,10,2.82,4,4,0,2,0,1,41,2,1,1,1
+64390,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18494.181242,18934.264494,2,99,4,4,0.78,4,4,0,2,0,2,45,1,3,5,NA
+64391,7,2,2,26,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,36731.516901,42296.380247,1,99,10,6,2.75,2,1,0,0,0,1,31,1,4,6,NA
+64392,7,2,1,10,NA,3,3,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,46228.073505,48314.890943,2,94,5,5,0.89,4,4,0,2,0,2,51,1,2,3,NA
+64393,7,2,2,16,NA,4,4,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14166.687432,14623.20249,1,100,15,15,3.87,6,6,1,3,0,2,39,1,4,1,4
+64394,7,2,2,2,NA,1,1,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8082.096138,8109.693752,1,90,4,4,0.46,7,7,2,3,0,2,34,2,1,6,NA
+64395,7,2,1,6,NA,3,3,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58159.917125,59822.981903,3,92,15,15,5,4,4,0,2,0,2,38,1,5,1,5
+64396,7,2,2,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,162038.12624,162493.980108,1,94,77,77,NA,1,1,0,0,0,2,51,1,3,3,NA
+64397,7,2,2,54,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15714.689474,15797.787849,2,99,15,15,5,1,1,0,0,0,2,54,1,5,5,NA
+64398,7,2,1,4,NA,4,4,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14467.4421,15952.720644,2,101,5,5,1.19,3,3,1,0,0,2,32,1,4,1,3
+64399,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,26763.110196,36351.891896,2,94,3,3,0.95,2,2,0,1,0,2,45,1,5,3,NA
+64400,7,2,2,61,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,11623.354795,12210.370942,1,94,6,6,1.98,2,2,0,0,2,1,65,2,1,1,1
+64401,7,2,1,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,29537.209125,29019.340657,1,95,12,3,1.19,4,1,0,0,0,1,29,1,3,6,NA
+64402,7,2,1,48,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,5,1,NA,2,2,2,2,2,2,1,2,2,2,27776.016947,27368.446715,2,90,6,6,1.15,5,5,0,2,0,2,47,2,1,1,5
+64403,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,22685.373982,24056.024455,1,98,4,4,1,3,3,0,1,1,1,65,1,2,1,NA
+64404,7,2,1,33,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,99283.360764,105383.970589,3,91,8,8,3.4,2,2,0,0,0,1,33,2,5,1,4
+64405,7,2,1,78,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,11550.158096,12480.92393,2,92,3,3,0.4,6,6,0,1,2,1,78,2,1,1,1
+64406,7,2,1,18,NA,5,6,2,18,219,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6999.347953,8045.308483,3,90,6,6,1.3,4,4,0,0,0,1,55,2,1,1,1
+64407,7,2,1,54,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,17801.655316,18085.793361,2,95,6,6,0.97,6,6,2,1,0,1,54,1,3,6,NA
+64408,7,2,2,15,NA,4,4,2,15,184,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13124.737024,13087.755238,2,101,5,5,1.08,3,3,0,1,1,2,62,1,4,2,NA
+64409,7,2,1,7,NA,4,4,2,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7886.48532,7943.675078,1,99,8,8,1.76,5,5,0,2,1,1,37,1,4,1,3
+64410,7,2,2,13,NA,2,2,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16360.434077,17220.943159,1,94,14,14,3.4,5,5,0,3,0,2,41,1,4,1,4
+64411,7,2,1,37,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18745.208345,19348.815401,1,92,7,7,2.1,3,3,0,0,2,1,37,2,5,5,NA
+64412,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,27553.988486,28260.692091,2,98,3,3,0.93,1,1,0,0,0,2,54,1,3,3,NA
+64413,7,2,2,48,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,2,5,NA,2,2,2,2,2,2,1,2,2,2,25778.164795,26907.837256,2,90,1,1,0.22,3,3,0,1,0,2,48,2,2,5,NA
+64414,7,2,2,1,23,3,3,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17516.090261,18614.694278,1,98,2,2,0.36,5,5,3,0,0,1,25,1,3,1,3
+64415,7,2,2,55,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18525.3583,18971.375925,1,92,14,14,3.9,4,4,0,0,0,2,55,2,5,1,5
+64416,7,2,1,6,NA,5,6,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11522.32071,1,97,7,7,1.48,5,5,1,2,0,1,40,2,5,1,4
+64417,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,135384.418391,141422.067719,1,98,15,15,5,3,3,0,0,0,1,48,1,4,1,4
+64418,7,2,2,17,NA,3,3,1,17,211,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,80091.55101,82618.226829,1,98,6,6,1.11,5,5,0,2,1,2,37,1,1,1,1
+64419,7,2,2,63,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,16352.915834,17178.789759,3,92,5,5,1.26,3,3,0,0,2,1,76,2,1,1,1
+64420,7,2,2,54,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,17928.412795,17976.980061,1,90,15,15,4.34,4,4,0,0,1,1,62,2,5,1,3
+64421,7,2,2,2,NA,4,4,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8754.193667,9419.610037,2,98,8,8,1.8,5,5,2,1,0,1,32,1,4,1,5
+64422,7,2,1,16,NA,2,2,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,15351.501195,16874.029415,2,90,1,1,0.22,3,3,0,1,0,2,48,2,2,5,NA
+64423,7,2,2,12,NA,4,4,1,12,151,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12826.497818,12890.123626,2,96,3,3,0.38,5,5,1,2,0,2,30,1,3,5,NA
+64424,7,2,1,0,6,3,3,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25469.602005,25029.287945,2,91,15,15,5,3,3,1,0,0,1,35,1,5,1,5
+64425,7,2,2,7,NA,5,6,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7722.982971,8635.133042,1,93,8,8,2.24,4,4,0,2,0,1,44,2,5,1,4
+64426,7,2,2,0,1,4,4,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4412.61884,4541.653777,1,93,7,7,1.97,4,4,1,2,0,2,33,1,4,3,NA
+64427,7,2,1,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105412.227726,111169.022023,2,101,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+64428,7,2,1,7,NA,5,6,1,7,86,NA,NA,2,1,2,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9101.455527,9700.372536,2,102,8,8,1.72,5,5,0,2,1,1,63,2,5,1,5
+64429,7,2,1,4,NA,5,6,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6185.185728,6767.562311,1,97,14,14,2.29,7,7,1,2,2,1,40,2,1,1,1
+64430,7,2,1,65,NA,5,7,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,7415.090784,7473.093956,2,95,12,12,NA,3,3,0,0,2,1,65,1,4,1,4
+64431,7,2,2,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,42468.064168,43146.583227,2,101,5,5,1.19,3,3,1,0,0,2,32,1,4,1,3
+64432,7,2,2,12,NA,4,4,1,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16598.888685,16552.117725,2,102,5,5,0.76,5,5,1,3,0,2,30,1,4,4,NA
+64433,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,1,2,NA,6882.452425,7271.66356,1,99,6,6,1.12,4,4,0,0,2,1,51,1,4,3,NA
+64434,7,2,2,7,NA,3,3,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,56541.371473,55790.525616,2,94,3,3,0.82,2,2,0,1,0,2,38,1,5,3,NA
+64435,7,2,1,16,NA,5,7,2,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9136.388281,9719.077424,3,91,8,8,2.24,4,4,0,2,0,1,45,1,4,1,4
+64436,7,2,2,34,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,27127.983961,27837.333201,2,90,14,14,3.45,4,4,1,1,0,2,34,2,5,6,NA
+64437,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,5,2,1,2,2,1,2,2,1,2,2,1,19520.240895,25185.107635,2,95,2,2,0.33,2,2,1,0,0,2,21,1,1,5,NA
+64438,7,2,2,44,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,25774.834017,34018.000767,1,100,8,8,3.3,2,2,0,1,0,2,44,1,4,3,NA
+64439,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,148039.171547,149225.244125,2,98,15,15,5,2,2,0,0,0,2,52,1,3,1,4
+64440,7,2,1,40,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,24506.160629,24417.630409,1,92,77,77,NA,2,2,0,0,1,1,40,2,5,5,NA
+64441,7,2,1,0,1,4,4,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7867.225786,8217.805417,3,92,6,6,0.93,5,5,2,1,0,2,37,1,5,1,3
+64442,7,1,1,61,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,8390.796678,0,3,92,7,7,2.45,2,2,0,0,1,2,55,1,1,1,1
+64443,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7567.81858,8250.001805,2,97,7,7,1.06,7,7,1,2,0,2,40,1,4,5,NA
+64444,7,2,1,18,NA,3,3,2,18,223,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,74123.161671,78732.269389,1,93,15,15,4.59,4,4,0,1,0,1,57,1,5,1,5
+64445,7,1,2,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11353.600279,0,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+64446,7,2,1,47,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,104871.724763,1,98,7,7,1.66,5,5,2,1,0,2,37,1,5,1,3
+64447,7,2,2,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,27934.372045,33663.467631,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+64448,7,2,1,0,0,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22215.410216,21831.35406,2,98,14,14,4.32,3,3,1,0,0,1,26,1,4,6,NA
+64449,7,2,2,18,NA,5,6,1,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10312.363668,10584.186706,2,102,10,10,3.62,3,3,0,0,0,1,51,2,5,1,5
+64450,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,9845.894687,11807.30791,2,94,14,14,3.58,4,4,1,0,1,1,80,1,3,2,NA
+64451,7,2,1,32,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18464.039593,21965.968723,1,101,6,6,2.51,2,1,0,0,0,1,32,1,4,6,NA
+64452,7,2,1,0,10,2,2,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7284.164858,7284.299119,1,92,6,6,1.67,3,3,1,0,0,1,27,1,3,6,NA
+64453,7,2,2,5,NA,4,4,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15449.848607,17265.123353,2,102,1,1,0.16,3,3,1,0,1,2,63,1,2,4,NA
+64454,7,2,2,4,NA,5,7,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13130.790087,14128.876607,2,102,14,14,3.25,5,5,1,1,0,2,32,1,4,1,3
+64455,7,2,1,6,NA,4,4,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12399.014378,12645.533348,2,96,6,6,1.35,3,3,1,1,0,2,25,1,3,5,NA
+64456,7,2,2,31,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,32097.38481,32019.876048,2,100,14,14,3.58,4,4,1,1,0,1,33,1,4,1,5
+64457,7,1,2,14,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20322.312754,0,2,91,6,6,0.93,5,5,1,2,0,2,50,2,1,5,NA
+64458,7,2,1,4,NA,2,2,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17533.373205,2,98,6,6,0.78,7,7,1,3,1,2,63,1,2,4,NA
+64459,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,129553.609875,132943.624501,2,95,10,10,4.49,2,2,0,0,1,1,62,NA,NA,1,3
+64460,7,2,1,10,NA,3,3,2,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,46081.129115,48091.57352,2,95,5,5,0.87,4,4,1,1,0,1,34,1,3,6,NA
+64461,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,24919.497762,25578.256222,1,101,1,1,0.1,4,4,1,1,0,2,52,1,4,3,NA
+64462,7,2,1,12,NA,4,4,1,13,157,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12147.046136,12703.852077,2,100,4,4,0.85,4,4,0,2,0,2,39,1,3,6,NA
+64463,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,52209.836905,57180.809983,1,98,13,13,NA,2,2,0,0,2,2,80,1,2,2,NA
+64464,7,2,1,0,7,5,6,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7997.590655,8653.844084,3,91,15,15,5,4,4,2,0,0,2,33,2,5,1,5
+64465,7,2,2,32,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,2,2,2,1,2,2,1,2,2,1,38184.257672,37153.966183,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+64466,7,2,2,23,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,18723.98095,18882.01405,2,95,1,1,0.03,3,3,1,0,0,1,23,1,3,6,NA
+64467,7,2,1,4,NA,4,4,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10510.490567,11159.309176,1,92,77,77,NA,5,5,1,2,0,2,41,1,3,5,NA
+64468,7,2,2,12,NA,5,7,2,12,147,NA,NA,2,1,4,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8301.172828,8622.168806,1,94,15,15,4.2,5,5,1,2,0,1,47,1,5,1,5
+64469,7,2,1,47,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,17787.524589,17730.790674,2,100,8,8,4.96,1,1,0,0,0,1,47,1,5,5,NA
+64470,7,2,2,20,NA,5,7,2,NA,NA,2,NA,2,2,3,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,55392.206282,55887.450369,3,91,6,6,2.04,2,2,0,0,0,1,24,2,4,5,NA
+64471,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,61994.231866,64386.409549,2,92,15,15,5,3,3,1,0,0,1,48,1,5,1,5
+64472,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,31460.11615,32394.500316,1,97,3,3,0.73,3,3,0,0,0,2,50,1,4,1,3
+64473,7,2,1,79,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11869.786782,12478.026352,2,101,6,6,2.04,2,2,0,0,2,2,74,1,3,1,3
+64474,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,23594.223469,25276.966347,1,97,15,15,5,4,4,0,1,0,1,40,1,4,1,4
+64475,7,2,1,62,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,1,NA,2,2,2,2,2,2,1,2,2,1,8197.936864,8932.691175,3,90,8,8,3.21,2,2,0,0,2,2,80,2,3,2,NA
+64476,7,2,2,58,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,1,4,NA,2,2,2,2,2,2,NA,NA,NA,NA,26272.35217,26409.360251,2,102,4,4,0.65,5,5,1,0,0,2,58,2,1,4,NA
+64477,7,2,1,49,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,19895.710643,19832.252596,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+64478,7,2,1,10,NA,4,4,2,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8655.162127,8668.652185,2,95,6,6,1.36,3,3,0,1,1,2,62,1,4,5,NA
+64479,7,2,2,17,NA,1,1,1,17,205,2,NA,2,2,5,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,20117.170449,20520.192725,1,94,2,2,0.27,5,5,0,4,0,2,47,2,1,4,NA
+64480,7,2,1,4,NA,5,6,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10276.262805,11525.953064,1,97,15,15,5,4,4,2,0,0,1,40,2,5,1,5
+64481,7,2,2,19,NA,3,3,2,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,35205.804094,37301.699975,2,101,2,1,0.32,3,1,0,0,0,2,19,NA,NA,NA,NA
+64482,7,2,1,2,NA,5,6,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10498.222836,11004.769407,1,100,15,15,5,4,4,2,0,0,1,39,2,5,1,5
+64483,7,2,2,5,NA,1,1,1,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16462.187772,17500.018071,3,92,5,5,0.81,5,5,3,0,0,2,23,1,4,5,NA
+64484,7,2,2,26,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,15265.136017,16765.388266,1,93,15,15,5,2,2,0,0,0,1,29,2,5,1,5
+64485,7,2,1,32,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,2,2,2,2,37658.482129,37374.26166,1,100,3,3,0.39,5,5,1,2,0,1,32,2,1,6,NA
+64486,7,2,1,9,NA,4,4,1,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11443.206518,11526.18825,1,100,12,12,NA,5,5,0,3,0,1,39,1,5,1,3
+64487,7,2,1,7,NA,3,3,1,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18812.486733,19704.440218,1,98,6,6,0.81,6,6,0,4,0,2,34,NA,NA,1,2
+64488,7,2,2,39,NA,4,4,2,NA,NA,2,NA,2,2,5,NA,3,4,2,1,2,2,1,2,2,NA,NA,NA,NA,35601.750356,47964.629837,1,93,4,4,1.09,2,2,1,0,0,2,39,2,3,4,NA
+64489,7,2,2,65,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,9991.888445,10532.363744,3,90,9,9,2.93,3,3,0,0,2,1,71,2,5,1,4
+64490,7,2,1,8,NA,1,1,1,8,104,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17882.621856,18280.794545,3,92,12,12,NA,6,6,1,3,0,2,33,1,5,1,4
+64491,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,87028.709207,89819.695129,1,98,8,8,2.46,3,3,1,0,0,1,29,1,4,6,NA
+64492,7,2,1,0,5,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20854.290322,20493.76497,2,94,14,14,4.05,3,3,1,0,0,2,37,1,5,1,5
+64493,7,2,1,17,NA,4,4,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13513.882801,1,96,15,15,4.9,4,4,0,1,0,1,47,1,3,1,5
+64494,7,2,2,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,28546.083602,28843.236835,2,95,9,9,4.1,2,2,0,0,0,2,19,1,4,NA,NA
+64495,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,166234.629208,1,95,10,10,4.76,2,2,0,0,0,1,53,1,2,1,3
+64496,7,1,1,30,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,25120.174741,0,2,102,14,14,4.86,3,3,1,0,0,1,30,1,5,1,5
+64497,7,2,1,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,25495.045769,28642.112646,1,94,5,5,1.39,2,2,0,0,2,1,60,1,1,5,NA
+64498,7,2,2,47,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15747.833197,16088.766331,3,91,15,15,5,3,3,0,1,0,1,47,2,5,1,5
+64499,7,2,1,63,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10004.038848,10164.290775,2,102,9,9,3.24,3,3,0,0,1,1,63,1,4,1,4
+64500,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,105593.259211,109935.078529,2,92,15,9,5,2,1,0,0,0,2,29,1,5,6,NA
+64501,7,2,1,66,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,1,NA,2,2,2,2,2,2,2,2,1,2,9068.437099,9213.701881,2,93,15,15,4.84,6,6,1,1,2,1,66,2,4,1,3
+64502,7,2,1,14,NA,4,4,2,14,177,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10366.393886,10567.129261,1,90,9,9,1.65,7,7,0,4,0,1,36,1,4,1,4
+64503,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,13555.744544,2,90,9,9,3.02,3,3,0,0,2,2,70,1,4,1,2
+64504,7,2,2,57,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,16165.962054,15723.538094,2,99,12,4,1.34,6,1,0,0,0,2,57,1,5,2,NA
+64505,7,2,1,18,NA,4,4,1,18,218,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13731.625553,14361.066702,1,100,1,1,0.04,4,4,1,1,0,2,51,1,3,3,NA
+64506,7,2,1,9,NA,3,3,2,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21087.869274,22686.932895,1,101,4,4,1.16,2,2,0,1,0,2,51,1,4,3,NA
+64507,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,46023.826844,47769.922841,1,91,9,9,3.24,3,3,0,0,1,1,70,1,1,2,NA
+64508,7,2,1,14,NA,4,4,2,14,171,NA,NA,2,2,2,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9987.254512,10059.992052,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+64509,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,36283.627734,36608.028102,1,102,5,4,1.74,5,1,1,1,0,2,24,1,4,5,NA
+64510,7,2,2,10,NA,1,1,1,10,123,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13028.403003,13395.598637,2,96,6,6,1.11,5,5,0,3,0,2,32,2,3,1,2
+64511,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,9,9,4.15,2,2,0,0,2,2,64,1,4,1,5
+64512,7,2,1,7,NA,4,4,2,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7886.48532,7943.675078,1,99,14,14,3.8,4,4,1,1,0,1,48,2,5,1,5
+64513,7,1,2,80,NA,2,2,NA,NA,NA,2,NA,2,1,9,NA,1,2,NA,2,2,2,2,2,2,NA,NA,NA,NA,17318.187297,0,2,90,77,77,NA,1,1,0,0,1,2,80,2,1,2,NA
+64514,7,2,2,31,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,6,2,2,2,2,2,2,2,1,2,2,1,29176.121289,29479.238307,2,90,4,4,0.81,3,3,0,0,0,1,39,2,3,5,NA
+64515,7,2,2,41,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,NA,NA,NA,NA,36924.381422,38397.364281,2,102,7,7,2.16,3,3,0,2,0,2,41,1,5,3,NA
+64516,7,2,2,2,NA,3,3,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37236.564417,41097.653079,2,95,7,7,2.91,2,2,1,0,0,1,32,1,5,2,NA
+64517,7,2,2,4,NA,3,3,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51483.624552,52160.136719,1,91,8,8,3.57,2,2,1,0,0,2,33,1,5,3,NA
+64518,7,2,1,14,NA,1,1,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,24805.109913,25184.264697,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+64519,7,1,2,1,18,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37915.354974,0,2,94,12,12,NA,5,5,1,0,2,2,39,1,5,3,NA
+64520,7,2,2,64,NA,2,2,2,NA,NA,2,NA,2,1,NA,NA,3,1,NA,1,1,2,1,2,2,1,2,2,NA,12026.225854,13435.778737,1,90,14,14,5,2,2,0,0,2,1,65,2,2,1,3
+64521,7,2,1,15,NA,4,4,2,15,182,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,5,5,1.18,3,3,0,1,0,2,55,1,4,5,NA
+64522,7,2,1,13,NA,5,7,1,13,163,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5535.564809,5914.423708,2,92,15,15,4.59,4,4,0,2,0,2,45,2,5,1,5
+64523,7,2,2,8,NA,4,4,2,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10888.493631,11627.08662,1,101,1,1,0.21,3,3,0,2,0,2,32,1,4,5,NA
+64524,7,2,1,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25372.225163,26225.545107,2,93,6,2,0.72,4,1,0,0,1,1,69,NA,NA,1,NA
+64525,7,2,2,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,45673.879644,47551.913663,1,94,5,5,1.47,2,2,0,0,0,1,24,1,4,1,4
+64526,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,139800.409559,143196.722449,1,100,8,8,1.95,4,4,0,2,1,2,49,1,5,6,NA
+64527,7,1,2,42,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,NA,NA,NA,NA,NA,NA,NA,29650.79971,0,2,90,NA,NA,NA,2,2,0,0,0,1,40,NA,NA,1,5
+64528,7,2,2,7,NA,4,4,2,7,89,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8579.490652,8729.153641,2,97,13,13,NA,6,6,2,2,0,2,24,1,2,6,NA
+64529,7,2,1,18,NA,3,3,1,18,219,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,112494.70924,119489.826838,1,94,5,5,1.04,4,4,1,1,0,1,18,1,2,NA,NA
+64530,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,181786.280703,1,95,10,10,4.76,2,2,0,0,0,1,53,1,2,1,3
+64531,7,2,2,2,NA,5,6,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5248.47937,5499.505789,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+64532,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,14340.013944,13947.561844,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+64533,7,2,1,13,NA,4,4,2,13,157,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12418.38217,12987.625827,2,99,6,6,1.57,3,3,0,2,0,2,31,1,3,77,NA
+64534,7,2,2,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,3,1,2,2,1,2,2,1,2,2,1,19658.218913,19871.634013,1,99,2,2,0.31,4,4,1,0,1,2,67,1,3,3,NA
+64535,7,2,2,7,NA,4,4,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6491.294105,6989.429429,2,99,6,6,1.11,5,5,1,2,0,2,41,1,2,5,NA
+64536,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,68852.695662,70965.028494,1,101,15,15,5,3,3,0,0,2,1,75,1,2,1,2
+64537,7,1,1,7,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8749.381325,0,2,100,15,15,5,4,3,0,2,0,1,42,1,3,5,NA
+64538,7,2,2,11,NA,1,1,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,16986.005478,17733.622754,2,102,6,6,1,6,6,1,3,0,1,35,2,3,1,3
+64539,7,2,2,7,NA,3,3,2,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,80369.555824,80552.420473,1,97,15,15,4.07,5,5,0,3,0,1,36,1,5,1,5
+64540,7,2,1,69,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,1,2,1,2,2,1,2,2,NA,11992.012141,12184.108862,2,102,99,99,NA,2,2,0,0,2,2,67,2,5,1,5
+64541,7,2,2,0,10,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27492.029044,26745.284482,1,97,10,10,2.95,4,4,2,0,0,1,28,1,5,1,4
+64542,7,2,2,5,NA,1,1,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13366.393396,13791.68675,2,94,77,77,NA,4,4,2,0,0,1,26,2,2,1,4
+64543,7,2,2,0,7,3,3,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12997.670925,13008.434462,1,99,10,10,2.48,5,5,2,1,0,1,33,1,5,1,5
+64544,7,2,1,69,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,7688.593855,7926.541218,2,100,14,14,4.59,3,3,0,0,2,1,69,2,5,1,5
+64545,7,2,1,15,NA,2,2,1,15,184,NA,NA,2,2,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17555.907575,18699.509115,2,93,8,8,2,4,4,1,1,0,1,50,2,4,1,4
+64546,7,1,2,6,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14300.71869,0,2,94,4,4,0.72,4,4,1,1,0,1,30,2,1,1,3
+64547,7,2,1,2,NA,1,1,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10803.555682,10931.27688,2,94,9,9,2.1,5,5,1,2,0,1,31,2,4,1,4
+64548,7,2,2,59,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,17950.494975,18045.41616,1,92,2,2,0.24,5,5,0,2,0,1,35,2,4,1,3
+64549,7,2,1,44,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,42123.732489,42281.342391,1,102,7,7,1.7,4,4,0,0,2,1,44,1,4,4,NA
+64550,7,2,2,62,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,3,12449.239856,12885.71781,1,96,7,7,2.64,2,2,0,0,2,1,64,2,5,1,5
+64551,7,2,2,10,NA,2,2,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14414.529053,14504.545241,2,96,14,14,3.36,4,4,1,1,0,2,28,1,2,6,NA
+64552,7,2,1,76,NA,2,2,2,NA,NA,1,2,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,13725.398968,14221.420409,2,90,7,7,2.52,2,2,0,0,2,2,71,1,4,1,4
+64553,7,2,2,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,13422.679076,14021.98734,2,92,6,6,2.31,2,2,0,0,2,2,65,1,4,5,NA
+64554,7,2,2,1,17,3,3,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,43715.161102,44289.593043,1,98,15,15,5,3,3,1,0,0,1,33,1,5,1,5
+64555,7,2,2,1,20,3,3,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20272.436775,20538.823422,2,94,77,77,NA,2,2,1,0,0,1,24,1,4,77,NA
+64556,7,2,2,0,4,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9767.083234,10416.172562,3,92,2,2,0.4,3,3,1,0,0,1,21,1,2,6,NA
+64557,7,2,1,3,NA,4,4,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11324.865632,11670.771889,2,96,5,5,1.13,3,3,1,1,0,2,31,1,3,5,NA
+64558,7,2,1,62,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11764.405491,12074.424659,1,97,14,14,3.93,3,3,0,1,2,2,63,1,4,1,4
+64559,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,72551.269339,74475.92188,1,95,7,7,2.45,2,2,0,0,2,2,70,1,3,1,3
+64560,7,1,2,13,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,21282.11341,0,1,90,6,6,1.44,3,3,0,1,0,1,44,2,1,1,3
+64561,7,2,1,6,NA,2,2,2,6,83,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8927.426533,9245.241527,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+64562,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,23235.97926,24326.785264,2,95,6,6,1.19,4,4,0,1,0,1,44,1,3,1,2
+64563,7,2,2,0,7,2,2,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7849.042868,8059.930457,2,98,6,6,1.07,5,5,3,0,0,2,24,1,3,1,3
+64564,7,2,2,54,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,19302.748944,20091.820859,3,92,15,15,5,3,3,0,1,0,1,55,2,5,1,4
+64565,7,1,1,54,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,20286.317046,0,2,99,99,99,NA,1,1,0,0,0,1,54,1,3,5,NA
+64566,7,2,1,7,NA,4,4,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8294.996898,8761.908989,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+64567,7,2,1,37,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,38835.309696,39871.800261,1,100,15,15,4.34,4,4,2,0,0,2,35,1,5,1,5
+64568,7,2,2,5,NA,1,1,1,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10493.785765,10940.90082,1,103,5,5,0.71,6,6,2,2,0,2,31,2,2,1,2
+64569,7,2,1,14,NA,4,4,1,14,177,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9188.45337,9407.445906,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+64570,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,49428.481088,50236.061312,1,101,2,2,0.72,1,1,0,0,1,2,65,1,4,2,NA
+64571,7,2,1,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,124091.929364,132249.317489,1,95,15,15,5,3,3,1,0,0,1,26,1,3,1,4
+64572,7,2,1,50,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15693.68983,16353.841376,1,91,77,77,NA,4,4,0,2,0,1,50,2,5,1,5
+64573,7,2,1,7,NA,4,4,1,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9022.8939,9451.941588,2,100,5,5,0.88,5,5,2,1,0,2,30,1,4,6,NA
+64574,7,2,1,17,NA,4,4,2,17,207,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11834.781205,12063.950506,2,90,6,6,1.12,4,4,0,1,1,1,63,2,1,1,1
+64575,7,2,2,6,NA,2,2,2,6,72,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13895.77426,15584.232002,2,91,99,99,NA,6,6,1,3,0,2,20,2,2,5,NA
+64576,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,38666.703155,42427.769217,2,94,15,15,5,1,1,0,0,1,1,80,1,5,2,NA
+64577,7,2,1,0,4,3,3,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12008.205501,11800.609728,1,92,4,4,1.22,2,2,1,0,0,2,30,1,4,5,NA
+64578,7,2,1,1,12,1,1,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12567.081957,12360.030976,1,97,7,7,2.31,2,2,1,0,0,1,22,1,4,5,NA
+64579,7,2,2,3,NA,4,4,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10683.855206,11939.151167,1,96,3,3,0.43,4,4,1,1,0,2,39,2,4,1,3
+64580,7,2,1,7,NA,1,1,1,7,88,NA,NA,2,7,77,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11367.678664,11216.859778,2,96,77,77,NA,7,7,3,2,0,2,33,2,2,6,NA
+64581,7,2,2,31,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,14179.938483,16215.179647,3,90,10,10,4.63,2,2,0,0,0,2,31,1,5,1,4
+64582,7,2,1,59,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,1,1,2,2,1,2,2,1,16628.326744,17078.450103,2,102,8,8,2.01,4,4,0,0,0,1,59,2,4,1,4
+64583,7,2,1,16,NA,5,6,2,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8305.829479,8595.334823,1,93,15,15,5,3,3,0,2,0,2,48,2,5,3,NA
+64584,7,2,1,41,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12471.410981,12474.00514,2,103,77,77,NA,5,5,0,2,0,2,39,2,5,1,5
+64585,7,2,2,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,3,1,2,2,1,2,2,NA,NA,NA,NA,25189.042335,25118.259708,1,100,8,8,1.95,4,4,0,2,0,2,42,1,4,1,4
+64586,7,2,2,25,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,1,1,2,2,2,2,2,2,2,NA,NA,NA,NA,50177.882654,49959.335916,1,100,4,4,0.78,4,4,0,0,1,1,33,2,1,1,1
+64587,7,2,1,19,NA,4,4,1,19,239,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,2,1,0.18,2,1,0,0,0,1,19,NA,NA,NA,NA
+64588,7,2,2,3,NA,5,6,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8171.700571,8311.029296,1,92,7,7,1.65,4,4,2,0,0,1,24,1,4,1,3
+64589,7,2,1,13,NA,2,2,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14081.782012,14782.02856,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+64590,7,2,1,6,NA,4,4,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13423.881856,14062.200951,2,97,7,7,1.38,5,5,0,1,1,2,79,1,5,5,NA
+64591,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,20075.522681,20011.491118,3,92,4,4,1.22,2,2,0,0,0,1,51,1,2,1,3
+64592,7,2,2,29,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,6,2,2,2,1,1,2,1,1,2,2,NA,45207.136555,47234.971464,1,91,7,7,2.1,3,3,0,1,0,2,29,2,3,6,NA
+64593,7,2,1,47,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19686.178677,19752.591494,2,94,15,15,5,5,5,0,2,1,1,47,2,5,1,5
+64594,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21399.459455,20960.973241,1,90,15,15,4.34,4,4,0,0,1,1,62,2,5,1,3
+64595,7,2,1,4,NA,3,3,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,27813.351627,32589.867729,2,97,1,1,0.21,4,4,2,0,0,2,34,2,1,1,2
+64596,7,2,2,12,NA,1,1,1,12,150,NA,NA,2,2,4,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20419.465237,21192.774678,2,102,6,3,0.54,6,4,0,4,0,2,43,2,1,5,NA
+64597,7,2,2,16,NA,3,3,2,16,203,NA,NA,1,1,NA,11,NA,NA,NA,1,2,1,1,2,1,1,2,2,1,23708.623398,23934.71279,2,97,2,2,0.38,4,4,0,2,2,2,64,2,1,1,NA
+64598,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,98514.948291,98181.677236,1,99,15,15,5,2,2,0,0,2,1,62,1,5,1,4
+64599,7,2,1,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,29756.291619,30015.261296,2,101,8,8,2.43,3,3,0,1,0,1,35,1,4,6,NA
+64600,7,2,2,14,NA,4,4,2,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11478.437608,11446.094627,1,96,10,10,3.04,4,4,0,1,0,2,43,1,5,1,4
+64601,7,2,1,6,NA,5,6,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8556.838894,9269.764719,2,95,15,15,5,3,3,0,1,0,2,34,2,5,1,NA
+64602,7,2,2,59,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,22224.73066,22340.630739,2,94,14,14,5,1,1,0,0,0,2,59,1,4,2,NA
+64603,7,2,2,40,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,20303.639991,21936.687597,1,97,15,15,4.77,4,4,1,1,0,2,40,1,5,1,5
+64604,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,2,3,NA,2,2,2,1,2,2,2,2,2,2,23200.373382,24548.135184,2,93,5,5,0.89,4,4,0,2,0,1,42,NA,NA,6,NA
+64605,7,2,2,0,6,3,3,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21671.775435,21083.121886,1,98,6,6,1.34,4,4,2,0,0,2,25,1,3,1,4
+64606,7,2,2,59,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,23644.678283,24188.238174,2,93,5,5,1.32,2,2,0,0,0,2,59,2,4,2,NA
+64607,7,2,2,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,14447.262849,14051.875602,3,90,15,15,5,5,5,1,0,1,1,38,2,3,1,4
+64608,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,104371.641146,109271.336207,1,90,15,15,5,3,3,0,0,0,1,59,1,5,1,5
+64609,7,2,1,64,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,8385.499814,8451.09383,2,92,12,12,NA,2,1,0,0,1,2,58,1,3,5,NA
+64610,7,2,1,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,141773.363283,145979.183499,1,97,15,15,5,4,4,0,0,1,1,67,NA,NA,2,NA
+64611,7,2,1,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17206.320427,17271.69718,2,100,14,14,3.06,5,5,1,0,0,1,50,1,5,1,5
+64612,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9013.911777,9371.113947,1,97,4,4,0.46,7,7,3,3,0,2,31,1,3,1,NA
+64613,7,2,1,25,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,49741.714519,50662.492276,2,91,7,7,3.76,1,1,0,0,0,1,25,1,5,5,NA
+64614,7,2,1,7,NA,3,3,1,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,71470.369236,87471.572436,2,101,8,8,2.81,3,3,0,2,0,1,48,1,3,3,NA
+64615,7,2,2,14,NA,3,3,2,14,179,NA,NA,1,1,NA,9,NA,NA,NA,1,1,1,NA,NA,NA,1,2,2,1,23708.623398,23934.71279,2,97,2,2,0.38,4,4,0,2,2,2,64,2,1,1,NA
+64616,7,2,1,21,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,32384.468407,33958.290597,2,100,14,14,3.58,4,4,0,1,0,1,46,2,5,1,5
+64617,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,9053.837749,9198.868668,2,99,77,77,NA,4,4,1,1,1,2,38,2,4,1,4
+64618,7,2,1,1,21,1,1,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8481.734412,8750.218705,1,103,5,5,0.74,5,5,1,1,0,2,40,99,3,1,1
+64619,7,2,1,16,NA,4,4,2,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13969.457688,14694.403205,1,90,14,14,3.25,4,4,0,2,0,2,33,2,3,1,3
+64620,7,2,1,36,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14221.330587,14963.406508,3,90,12,12,NA,4,4,0,0,1,1,62,2,4,3,NA
+64621,7,2,1,38,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,39053.240727,39737.089727,1,100,14,14,4.71,3,3,0,1,0,1,38,1,5,1,5
+64622,7,2,1,13,NA,1,1,2,13,165,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20398.562455,20710.36162,2,94,13,13,NA,5,5,0,3,0,1,32,2,2,1,1
+64623,7,2,1,2,NA,2,2,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8488.185756,9094.478438,2,99,77,77,NA,4,4,1,1,1,2,38,2,4,1,4
+64624,7,2,1,1,15,1,1,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10078.126531,10467.748183,2,97,NA,2,0.36,6,4,3,1,0,2,25,1,4,6,NA
+64625,7,2,1,69,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,2,2,2,1,8491.292032,8823.40317,1,96,7,7,1.39,5,5,0,2,2,1,69,2,2,1,2
+64626,7,2,1,0,4,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9081.731172,9425.042696,1,101,5,5,0.74,6,6,1,3,0,1,38,1,4,1,4
+64627,7,2,2,17,NA,5,6,1,18,216,2,NA,2,2,5,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9269.277563,9333.801929,2,102,5,5,1.08,3,3,0,1,0,2,46,2,1,5,NA
+64628,7,2,2,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27524.153939,27114.967247,3,92,12,12,NA,3,2,0,0,0,1,45,1,3,1,3
+64629,7,2,1,14,NA,3,3,2,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,119111.433099,126517.990149,2,94,7,7,2.38,2,2,0,1,0,1,39,1,4,3,NA
+64630,7,2,1,0,3,3,3,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20370.716701,21140.779334,1,91,15,15,5,5,5,3,0,0,1,45,1,5,1,5
+64631,7,2,2,1,17,5,6,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7029.864692,7462.832472,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+64632,7,2,1,21,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,51858.383222,54820.448996,2,91,10,10,4.63,2,2,0,0,0,1,55,2,3,3,NA
+64633,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,110095.623149,1,94,10,10,4.3,2,2,0,0,0,1,27,1,5,1,5
+64634,7,2,2,74,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,16361.152596,17584.911061,1,100,1,1,0.36,1,1,0,0,1,2,74,1,2,2,NA
+64635,7,2,1,29,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19510.698389,19365.075974,1,102,7,7,1.9,4,4,1,1,0,1,29,1,4,1,3
+64636,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,2,1,0.02,2,1,0,0,0,1,20,1,4,5,NA
+64637,7,2,1,77,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,1,2,2,2,2,2,NA,16591.871479,16938.772121,3,91,77,77,NA,4,4,0,0,2,1,54,1,4,1,2
+64638,7,2,1,8,NA,4,4,1,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10410.106675,10617.081899,2,96,5,5,1.08,3,3,0,1,0,2,41,1,3,1,NA
+64639,7,2,1,58,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,25963.141347,26172.595903,1,92,9,9,3.7,2,2,0,0,0,1,58,1,2,1,4
+64640,7,2,1,80,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,1,1,NA,2,2,2,2,2,2,1,2,2,NA,18949.267372,20024.848444,2,98,4,4,1.15,2,2,0,0,2,1,80,1,1,1,1
+64641,7,2,1,4,NA,2,2,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17533.373205,2,98,8,8,1.48,7,7,3,0,0,1,26,1,3,1,3
+64642,7,2,1,15,NA,2,2,1,15,183,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,24585.624844,25010.051033,3,91,5,5,0.89,4,4,0,2,2,1,61,2,3,1,4
+64643,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22560.963402,22680.264479,1,98,6,6,1.65,2,2,0,1,0,2,52,2,5,1,NA
+64644,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,93796.829073,95761.252302,1,100,6,6,1.78,3,3,1,1,0,2,35,1,5,4,NA
+64645,7,2,1,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5942.817425,6352.518414,2,97,5,5,0.84,5,5,2,1,0,2,27,1,3,1,3
+64646,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,32860.812334,35527.111698,1,91,7,7,2.68,2,2,0,0,2,1,80,1,4,1,4
+64647,7,2,1,4,NA,2,2,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13504.027725,14468.59111,2,93,8,8,2,4,4,1,1,0,1,50,2,4,1,4
+64648,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,37080.526463,37974.333878,1,103,15,15,5,2,2,0,0,0,1,36,2,5,1,5
+64649,7,2,2,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,2,NA,NA,NA,NA,15326.318384,16292.539752,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+64650,7,2,2,10,NA,3,3,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22371.648216,23759.450609,1,95,5,5,1.08,3,3,0,1,0,1,53,1,4,1,4
+64651,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,28454.541725,29098.608791,3,91,3,3,1.24,1,1,0,0,0,1,58,1,4,5,NA
+64652,7,2,2,10,NA,5,6,1,10,125,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7905.246022,8589.222558,1,100,3,3,0.73,2,2,0,1,0,2,38,2,5,3,NA
+64653,7,2,2,1,18,4,4,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9287.278834,9993.215624,2,102,5,3,0.63,5,4,2,1,0,1,24,1,4,6,NA
+64654,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,151649.038926,165350.159672,1,101,7,7,2.31,2,2,0,0,0,1,44,1,3,1,2
+64655,7,2,2,9,NA,1,1,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15510.382876,18634.445505,1,100,13,13,NA,4,4,1,1,0,1,28,2,1,1,1
+64656,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,14314.968597,14503.241546,1,96,15,15,5,4,3,0,0,1,2,54,2,4,1,5
+64657,7,2,2,4,NA,3,3,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,71121.657525,76575.027332,1,92,9,4,1,7,3,2,1,0,1,45,1,4,2,NA
+64658,7,2,2,51,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,1,2,25483.560748,27318.705448,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+64659,7,2,2,11,NA,4,4,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7646.649777,8233.445923,2,99,4,4,1,3,3,0,1,0,2,38,1,3,5,NA
+64660,7,2,2,4,NA,2,2,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13366.393396,14209.054666,2,94,7,7,1.34,5,5,2,1,0,1,32,2,1,1,NA
+64661,7,2,1,29,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,31312.870743,38059.589922,2,90,8,8,2.24,4,4,1,1,0,2,29,1,4,6,NA
+64662,7,2,1,10,NA,3,3,1,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,66003.625505,68276.88064,1,101,8,8,1.85,5,5,0,3,0,1,41,1,3,1,4
+64663,7,2,2,33,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21522.871343,21589.047039,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+64664,7,2,2,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,40760.712736,41570.695949,1,98,4,4,1.26,2,2,0,0,1,2,80,1,4,2,NA
+64665,7,2,2,6,NA,3,3,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24070.467912,23750.822126,1,92,4,4,0.5,6,6,0,3,0,2,41,1,4,1,NA
+64666,7,2,2,11,NA,4,4,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10332.067017,10608.478853,1,100,12,12,NA,5,5,0,3,0,1,39,1,5,1,3
+64667,7,2,1,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,26052.523863,2,101,3,3,0.68,2,2,0,0,0,1,58,1,1,6,NA
+64668,7,1,1,21,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,52698.05363,0,3,92,2,2,0.4,3,3,1,0,0,1,21,1,2,6,NA
+64669,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,35475.142583,37174.132552,2,95,6,6,2.95,1,1,0,0,1,2,60,1,3,3,NA
+64670,7,2,2,2,NA,3,3,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,55441.113009,61189.845638,1,92,6,6,1.62,3,3,1,0,0,2,26,1,5,1,5
+64671,7,2,2,13,NA,3,3,1,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,117872.104347,120084.841085,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+64672,7,2,1,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16715.026676,2,100,7,7,1.38,5,5,1,0,0,2,45,1,2,3,NA
+64673,7,2,1,12,NA,5,6,1,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5388.361335,5758.757236,3,91,6,6,1.22,5,5,1,2,0,2,37,1,4,1,2
+64674,7,2,2,39,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,1,2,2,2,2,2,2,2,NA,NA,NA,NA,38161.026403,39476.245254,1,100,7,7,1.74,4,4,0,2,0,2,39,2,1,1,3
+64675,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,159455.333031,1,95,15,15,5,4,4,0,2,0,2,42,1,5,1,5
+64676,7,2,2,56,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,4,6,NA,2,2,2,2,2,2,1,2,1,2,19969.163208,20990.960204,2,93,10,10,3.67,3,3,0,0,0,2,56,2,4,6,NA
+64677,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19210.136544,18684.400763,1,96,12,12,NA,1,1,0,0,0,2,56,1,4,3,NA
+64678,7,2,2,0,4,4,4,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7333.056164,7413.124496,2,101,5,5,1.08,3,3,1,0,0,1,31,1,4,6,NA
+64679,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,17978.142628,18905.276914,1,91,7,7,1.57,4,4,0,3,0,2,38,2,2,3,NA
+64680,7,2,2,35,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,21116.677917,21908.928919,1,92,14,14,3.47,4,4,0,2,0,1,37,1,5,1,5
+64681,7,2,1,15,NA,3,3,2,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,68148.957861,67253.324127,1,93,7,7,2.16,3,3,0,1,0,2,50,1,5,3,NA
+64682,7,2,1,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,133542.212862,160359.69013,1,94,14,14,2.96,5,5,0,3,0,2,39,1,4,1,3
+64683,7,1,2,26,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,6,3,1,2,2,1,2,2,NA,NA,NA,NA,109522.19868,0,1,90,9,1,0,2,1,0,0,0,2,26,1,5,6,NA
+64684,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,1,2,1,2,2,1,2,2,1,16995.648055,19983.260181,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+64685,7,2,1,12,NA,2,2,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13752.835112,15116.811136,3,90,7,7,1.48,5,5,0,1,0,1,43,2,1,6,NA
+64686,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6226.488588,6570.976462,2,99,6,6,1.03,6,6,3,0,0,1,33,1,3,6,NA
+64687,7,2,2,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,20247.768461,19251.812222,2,93,4,4,0.56,5,5,2,1,0,1,27,1,2,6,NA
+64688,7,2,1,1,18,1,1,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11853.772636,11869.789876,2,96,4,4,0.81,4,4,1,1,0,1,36,2,1,6,NA
+64689,7,2,2,42,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,2,5,2,2,2,2,1,2,2,2,2,2,2,27654.660303,28156.17371,2,103,4,4,0.79,3,3,0,0,0,2,42,2,2,5,NA
+64690,7,2,2,11,NA,4,4,2,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8147.287486,8590.325322,2,90,2,2,0.38,4,4,1,2,0,2,32,1,4,5,NA
+64691,7,2,1,0,10,2,2,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,2,2,1,NA,NA,NA,NA,5357.080288,5657.49929,1,103,12,12,NA,6,6,2,1,0,2,27,2,2,1,3
+64692,7,1,2,59,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,13728.308948,0,2,90,10,10,3.51,3,3,0,0,1,2,59,1,4,1,NA
+64693,7,1,2,46,NA,2,2,NA,NA,NA,2,NA,2,2,4,NA,2,4,NA,2,2,2,2,2,2,NA,NA,NA,NA,23968.560941,0,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+64694,7,2,1,23,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,44887.234615,46900.157613,2,93,9,9,2.6,4,4,0,0,0,2,58,2,4,4,NA
+64695,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,11608.998717,12267.891619,3,90,14,14,3.47,4,4,1,1,0,2,38,2,5,1,5
+64696,7,2,1,11,NA,4,4,2,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.422451,8641.637123,1,99,14,14,4.86,3,3,0,1,0,1,42,1,5,1,5
+64697,7,2,1,8,NA,4,4,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11443.206518,11987.342507,1,100,6,6,1.65,2,2,0,1,0,2,42,1,4,5,NA
+64698,7,2,1,13,NA,3,3,1,13,166,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23685.067681,25157.846578,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+64699,7,2,1,16,NA,4,4,1,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18123.994406,21384.995218,2,101,8,8,2.81,3,3,0,1,0,1,35,1,1,1,2
+64700,7,1,1,24,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,108410.783716,0,1,91,8,8,2.51,3,3,1,0,0,2,24,1,4,1,3
+64701,7,2,1,57,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,3,NA,2,2,2,1,2,2,2,2,1,2,27131.416371,32184.468102,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+64702,7,2,2,4,NA,3,3,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,31376.988784,34630.493438,1,101,7,7,1.82,4,4,2,0,0,2,27,1,2,1,3
+64703,7,2,2,22,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,32537.532358,33640.063825,2,90,5,5,1.19,3,3,0,0,0,2,50,2,4,4,NA
+64704,7,2,1,62,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,11937.570805,12597.633777,2,96,8,8,1.33,7,7,2,1,1,1,62,2,1,1,1
+64705,7,2,2,72,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,13242.661868,13437.555164,1,99,5,5,1.69,2,2,0,0,2,1,79,1,1,1,4
+64706,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20332.409486,19927.598485,2,93,6,6,1.47,3,3,0,0,0,2,47,1,4,5,NA
+64707,7,2,1,10,NA,1,1,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,17882.621856,17720.218058,3,92,4,4,0.67,4,4,0,3,0,2,36,2,1,5,NA
+64708,7,2,2,11,NA,1,1,2,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15002.25457,15643.939501,1,91,6,6,1.35,3,3,0,2,0,2,38,1,4,3,NA
+64709,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9221.19173,9652.555988,2,95,3,3,0.75,2,2,0,0,2,1,60,1,2,1,2
+64710,7,2,1,3,NA,5,7,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8981.553859,10073.795326,3,91,15,15,4.47,4,4,2,0,0,1,33,1,5,1,5
+64711,7,2,2,35,NA,5,7,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,18346.384255,19018.034723,2,96,15,15,5,4,4,1,1,0,2,35,2,5,1,5
+64712,7,2,1,6,NA,1,1,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14820.807433,14827.373008,2,102,5,5,0.89,4,4,1,2,0,2,36,2,5,3,NA
+64713,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22224.73066,22340.630739,2,94,15,15,5,2,2,0,0,1,1,77,1,5,1,4
+64714,7,2,2,17,NA,4,4,2,17,204,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13661.047334,13622.554377,2,97,5,5,0.76,5,5,1,1,0,2,47,1,4,5,NA
+64715,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,4,2,0.83,2,1,0,0,0,1,24,1,4,5,NA
+64716,7,2,1,1,19,3,3,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46257.816906,54201.886729,2,94,14,14,3.36,4,4,2,0,0,1,31,1,3,1,5
+64717,7,2,1,1,13,4,4,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,11283.202594,2,101,5,5,0.89,4,4,1,1,1,2,38,1,4,77,NA
+64718,7,2,1,18,NA,4,4,1,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14848.504688,14956.646891,1,98,2,2,0.45,2,1,0,0,0,1,19,1,4,NA,NA
+64719,7,2,2,39,NA,5,6,2,NA,NA,1,2,2,1,7,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,18480.909695,18518.646173,1,93,9,9,5,1,1,0,0,0,2,39,2,5,3,NA
+64720,7,2,1,29,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,21548.249649,21044.90278,1,100,8,8,1.61,6,6,1,3,0,1,29,1,5,6,NA
+64721,7,2,1,56,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,17206.320427,17151.440282,2,100,5,5,1.08,3,3,0,0,0,1,38,1,2,5,NA
+64722,7,2,2,33,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,1,1,2,2,2,2,2,2,2,1,2,2,NA,38218.668882,37187.448906,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+64723,7,2,2,6,NA,4,4,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8292.876947,8466.609309,1,102,1,1,0,5,5,0,3,0,2,41,1,4,1,4
+64724,7,2,2,3,NA,3,3,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50908.326714,52512.144621,1,99,15,15,5,4,4,1,1,0,2,42,1,5,1,5
+64725,7,2,2,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,101005.054511,105662.333954,2,91,15,8,4.48,2,1,0,0,0,2,55,1,5,6,NA
+64726,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18486.334367,18427.371573,2,95,8,8,1.85,5,5,1,2,0,1,55,1,2,1,3
+64727,7,2,1,50,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,1,2,22446.308035,22401.210337,2,94,77,77,NA,4,4,0,0,0,1,28,2,1,3,NA
+64728,7,2,1,23,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,3,5,NA,2,2,2,1,2,2,2,2,2,2,38474.772527,42632.210531,2,93,7,5,1.79,3,1,0,0,0,1,25,2,4,5,NA
+64729,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,2,NA,1,2,1,NA,NA,NA,NA,NA,NA,NA,14314.616082,14811.078253,2,92,NA,NA,NA,2,1,0,0,1,2,58,2,4,5,NA
+64730,7,2,1,9,NA,2,2,2,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9390.522479,10327.334743,2,90,1,1,0.19,4,4,0,1,0,2,44,1,2,5,NA
+64731,7,2,2,34,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,36904.965687,36815.84758,2,93,9,9,3.14,3,3,0,2,0,2,34,2,5,3,NA
+64732,7,2,1,17,NA,4,4,1,17,215,2,NA,2,1,4,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16261.995423,16380.43213,2,96,6,6,1.7,2,2,0,1,0,1,25,2,4,5,NA
+64733,7,2,2,30,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,51753.495707,50822.503524,1,100,8,8,3.17,2,2,0,0,0,1,31,2,4,1,5
+64734,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12331.419303,12882.003985,2,95,5,5,1.32,2,2,0,0,2,1,66,1,2,1,4
+64735,7,2,2,6,NA,3,3,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49543.011606,52616.361829,1,94,15,15,5,6,6,0,4,0,1,38,1,5,1,4
+64736,7,2,1,6,NA,5,6,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8376.521179,8870.870608,1,92,12,12,NA,4,4,1,1,0,1,33,2,4,1,4
+64737,7,2,1,19,NA,4,4,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,2,2,0.73,2,1,0,0,0,1,20,1,4,5,NA
+64738,7,2,1,15,NA,3,3,2,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71458.892941,70519.759062,2,91,15,15,5,4,4,0,2,0,2,48,1,5,1,5
+64739,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,25026.664203,25574.319716,1,97,15,15,5,4,4,0,1,0,1,40,1,4,1,4
+64740,7,2,1,45,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,2,1,2,1,3,16903.372311,16842.307696,1,97,9,9,1.78,6,6,0,1,1,1,45,2,3,1,3
+64741,7,2,2,44,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,25189.042335,25070.255719,1,100,14,14,5,3,3,0,1,1,2,44,1,4,5,NA
+64742,7,2,1,56,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16419.180608,2,100,4,4,1.74,1,1,0,0,0,1,56,1,3,3,NA
+64743,7,2,1,10,NA,4,4,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8017.552697,8398.795399,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+64744,7,2,2,5,NA,3,3,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,29420.847299,31676.739048,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+64745,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20820.576198,20899.685598,2,96,7,7,1.49,5,5,2,1,0,1,51,1,5,1,3
+64746,7,2,1,29,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18523.956321,18232.779525,2,99,15,15,4.34,4,4,0,0,0,1,59,2,4,1,5
+64747,7,1,1,5,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9787.724348,0,1,96,10,10,3.51,3,3,1,0,0,1,25,1,4,1,5
+64748,7,2,2,0,4,4,4,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5576.185193,5739.245437,2,96,6,6,1.35,3,3,1,1,0,2,25,1,3,5,NA
+64749,7,2,1,2,NA,5,6,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9239.758777,10363.400393,2,91,14,14,3.47,4,4,1,1,0,2,36,2,3,1,5
+64750,7,2,1,18,NA,1,1,1,18,226,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,32326.52031,34735.818434,2,98,13,13,NA,5,5,0,2,0,1,48,2,1,1,2
+64751,7,2,1,9,NA,3,3,2,9,118,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61920.455064,68552.408117,2,94,12,12,NA,5,5,1,1,0,1,37,1,4,1,3
+64752,7,2,2,62,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16352.915834,17093.606152,3,92,8,8,2.97,2,2,0,0,2,2,62,1,4,1,4
+64753,7,2,2,68,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,9716.805546,12994.252166,2,90,77,77,NA,1,1,0,0,1,2,68,2,1,4,NA
+64754,7,2,1,36,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,14204.262514,14001.792163,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+64755,7,2,1,3,NA,1,1,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14415.164987,15444.808928,1,103,77,77,NA,4,4,1,0,1,1,20,1,3,6,NA
+64756,7,2,1,72,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,2,2,2,1,2,2,2,2,2,NA,13654.270555,14323.606702,2,93,10,10,3.04,4,4,0,0,2,1,72,2,3,1,2
+64757,7,2,2,9,NA,5,7,2,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,67349.005531,67425.922822,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+64758,7,2,1,33,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20071.705576,20986.552878,3,91,14,14,5,2,2,0,0,0,2,27,2,5,1,5
+64759,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,1,2,1,2,2,1,1,2,NA,52279.532372,58506.425352,2,101,5,5,1.63,2,2,0,0,2,1,80,1,1,1,1
+64760,7,2,1,4,NA,3,3,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,77702.196479,85091.48102,1,95,9,9,2.68,4,4,2,0,0,2,27,1,4,1,4
+64761,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,35276.751365,1,94,7,7,1.65,5,4,0,0,0,1,46,1,4,1,4
+64762,7,2,1,17,NA,4,4,1,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,10,10,3.89,3,3,0,1,0,2,49,1,4,1,3
+64763,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,17701.770315,2,95,5,2,0.63,3,1,0,0,0,1,31,1,4,5,NA
+64764,7,2,1,36,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,24555.036413,24478.369844,2,94,8,8,3.4,2,2,0,0,0,1,36,1,2,1,5
+64765,7,2,1,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,78529.577822,80778.70531,1,101,7,7,1.55,5,5,1,2,0,2,31,1,4,1,2
+64766,7,2,2,3,NA,1,1,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11859.546176,12772.78418,1,102,5,5,0.62,7,7,1,3,0,1,49,2,2,1,1
+64767,7,2,2,7,NA,2,2,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15510.382876,16516.802792,1,100,5,5,0.78,6,5,1,2,0,2,40,2,1,5,NA
+64768,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,38616.556866,1,94,6,6,1.21,4,4,2,0,0,1,27,1,2,1,2
+64769,7,2,1,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19093.004278,19397.753966,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+64770,7,2,1,11,NA,5,6,2,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5855.595238,6345.238788,1,91,15,15,3.25,7,7,1,2,0,2,31,1,5,1,5
+64771,7,2,1,64,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,11937.570805,12505.355675,2,96,3,3,0.24,7,7,2,3,1,2,40,1,3,3,NA
+64772,7,2,1,50,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,2,NA,2,2,2,2,2,2,NA,NA,NA,NA,31872.125984,32482.275435,2,96,4,4,1.56,1,1,0,0,0,1,50,2,1,2,NA
+64773,7,2,2,16,NA,1,1,1,16,203,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19836.724169,20428.617367,1,98,15,15,5,5,5,0,1,1,2,55,1,5,1,5
+64774,7,2,2,5,NA,1,1,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16648.051651,2,98,5,5,0.59,7,7,2,1,2,2,71,1,2,1,1
+64775,7,2,2,6,NA,5,6,1,7,84,NA,NA,1,1,NA,1,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,6402.995556,7222.354698,2,92,7,7,1.61,4,4,0,2,0,1,51,2,3,1,3
+64776,7,2,2,29,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,2,2,2,2,2,2,2,NA,NA,NA,NA,50915.06085,54061.945895,3,92,3,3,0.52,5,5,2,1,0,2,29,2,1,1,3
+64777,7,2,2,52,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15790.702799,16017.473382,2,90,14,14,5,2,2,0,0,1,2,52,2,5,1,NA
+64778,7,1,1,22,NA,5,6,NA,NA,NA,2,NA,2,1,99,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,13367.406737,0,1,92,15,15,4.44,5,5,0,0,1,1,65,NA,NA,1,5
+64779,7,2,2,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,26465.930618,27659.054945,2,100,7,7,1.79,4,4,0,1,0,2,51,1,3,3,NA
+64780,7,2,2,18,NA,4,4,1,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13697.127402,14256.437303,2,100,4,4,1.16,2,2,0,0,1,2,18,1,2,NA,NA
+64781,7,2,1,15,NA,2,2,2,15,185,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,19820.949231,20163.121943,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+64782,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,29250.663295,29731.562834,1,94,2,2,0.81,1,1,0,0,0,1,50,1,4,3,NA
+64783,7,2,1,0,1,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4461.618312,4741.503802,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+64784,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,20135.920214,21074.978603,2,97,5,5,0.76,5,5,1,1,0,2,47,1,4,5,NA
+64785,7,2,1,76,NA,1,1,1,NA,NA,2,NA,2,1,99,NA,1,1,NA,2,2,2,2,2,2,2,2,2,NA,21815.897449,22383.010849,3,92,5,5,1.26,3,3,0,0,2,1,76,2,1,1,1
+64786,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,113128.964375,115356.428049,1,94,15,15,4.95,4,4,0,0,2,1,72,1,3,1,3
+64787,7,2,1,10,NA,2,2,2,10,129,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,9390.522479,10327.334743,2,90,1,1,0.06,3,3,0,2,0,2,32,2,2,77,NA
+64788,7,2,1,5,NA,4,4,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12996.965152,13155.4937,1,101,6,6,1.72,2,2,1,0,0,2,29,1,4,5,NA
+64789,7,2,1,67,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,7736.56115,7645.782314,1,99,15,15,5,2,2,0,0,2,1,67,1,5,1,5
+64790,7,2,1,2,NA,3,3,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46703.291366,50234.137371,1,93,15,15,5,3,3,1,0,0,1,39,NA,NA,1,NA
+64791,7,2,2,26,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,6,2,2,2,2,NA,NA,NA,NA,NA,NA,NA,31196.446669,31060.57243,2,99,12,3,0.52,5,3,0,1,0,1,30,2,2,4,NA
+64792,7,2,2,40,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,35815.777398,37277.17285,1,91,14,14,3.9,4,4,0,1,0,1,41,1,2,1,4
+64793,7,2,2,14,NA,2,2,2,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23968.380373,25111.880337,2,91,14,14,4.19,3,3,0,1,0,1,55,2,3,1,5
+64794,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,20443.961017,19891.057859,2,99,6,6,1.13,4,4,1,1,0,1,33,1,3,6,NA
+64795,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,9772.038079,10324.658831,1,95,3,3,0.78,3,3,0,0,2,1,80,1,4,1,3
+64796,7,2,1,1,15,5,6,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5492.796032,5818.035599,2,100,4,4,0.5,6,6,2,1,0,1,30,2,4,1,3
+64797,7,2,1,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21055.122405,22540.415493,1,93,15,15,5,3,3,1,0,0,1,33,1,5,1,3
+64798,7,2,2,6,NA,3,3,2,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,62595.719575,66478.781231,1,95,6,6,1.35,3,3,0,1,0,1,42,1,4,1,4
+64799,7,2,1,28,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,38502.344077,41392.592866,1,100,13,13,NA,4,4,1,1,0,1,28,2,1,1,1
+64800,7,2,2,52,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,12449.932013,12483.65832,3,90,10,10,3.67,3,3,0,1,0,2,52,2,3,5,NA
+64801,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,29167.119125,32047.917266,1,101,6,6,1.52,3,3,0,1,1,1,62,1,2,1,3
+64802,7,2,1,42,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,24874.121286,25386.96888,1,102,8,8,4.78,1,1,0,0,0,1,42,1,3,5,NA
+64803,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19440.629552,20463.14283,1,93,6,6,1.16,4,4,2,0,0,2,33,1,5,1,4
+64804,7,2,2,9,NA,3,3,2,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,73810.484644,74865.697805,1,94,10,10,2.67,5,5,0,3,0,1,40,1,5,1,2
+64805,7,2,1,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,19008.083201,19380.210545,2,101,2,2,0.73,1,1,0,0,0,1,38,1,2,5,NA
+64806,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,40157.007559,43267.510233,2,98,8,8,3.4,2,2,0,0,2,1,80,1,4,1,4
+64807,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,66503.043118,67175.236045,2,98,14,14,3.25,5,5,2,1,0,1,37,1,5,1,5
+64808,7,2,1,44,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16651.742047,16721.380189,1,100,15,15,5,4,4,0,1,0,1,44,2,5,1,5
+64809,7,2,2,26,NA,4,4,2,NA,NA,2,NA,2,1,2,NA,4,1,3,1,2,2,1,2,2,NA,NA,NA,NA,20146.149642,19228.641128,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+64810,7,1,1,28,NA,5,6,NA,NA,NA,2,NA,2,2,1,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,11321.172632,0,1,96,3,3,0.54,4,4,0,1,0,1,28,2,5,5,NA
+64811,7,2,2,16,NA,5,6,1,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8663.507259,9422.028242,1,102,9,9,2.39,5,5,0,1,1,1,55,2,5,1,5
+64812,7,2,2,17,NA,4,4,2,17,215,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16317.790515,16843.623891,1,101,6,6,1.43,4,4,0,1,2,2,72,1,2,1,NA
+64813,7,2,2,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19944.119297,19741.288195,2,98,2,2,0.52,1,1,0,0,0,2,55,1,4,3,NA
+64814,7,2,2,1,22,4,4,2,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,7960.486003,2,95,9,9,1.78,6,6,2,0,0,2,48,1,3,1,2
+64815,7,2,2,11,NA,1,1,1,11,137,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15352.601806,16028.326912,2,102,4,4,0.61,5,5,0,3,0,1,34,2,3,1,3
+64816,7,2,1,21,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,21763.209029,22385.504537,2,92,4,1,0,4,1,0,0,0,1,21,1,4,5,NA
+64817,7,2,1,1,20,4,4,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6454.74783,2,100,14,1,0,3,1,1,0,1,1,62,1,5,1,5
+64818,7,2,1,8,NA,5,6,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,7030.880662,7583.419516,3,91,14,14,3.58,4,4,1,1,0,1,39,2,5,1,5
+64819,7,2,2,2,NA,4,4,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,8250.113235,2,100,9,9,3.24,3,3,1,0,0,1,32,1,3,1,4
+64820,7,2,1,7,NA,1,1,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19774.151841,21259.203461,3,92,7,7,1.65,4,4,1,1,0,1,27,1,3,1,3
+64821,7,2,1,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,11281.447612,11722.687225,2,92,8,8,4.59,1,1,0,0,1,1,62,1,3,4,NA
+64822,7,2,1,53,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,29181.414565,30161.390695,1,94,2,2,0.63,2,1,0,0,0,1,53,1,3,6,NA
+64823,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,23799.136983,23147.811167,1,92,15,15,4.44,5,5,0,3,0,2,43,1,5,6,NA
+64824,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,33364.852884,34296.975056,1,97,2,2,0.76,1,1,0,0,1,1,62,1,5,3,NA
+64825,7,2,2,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,150482.742079,150600.121666,2,98,15,15,5,2,2,0,0,1,2,52,1,4,1,4
+64826,7,2,1,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7800.5318,8105.625946,2,96,3,3,1.01,1,1,0,0,1,1,61,1,3,3,NA
+64827,7,2,2,8,NA,4,4,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11334.095519,11753.9565,2,96,4,4,0.65,5,5,0,3,0,1,30,1,4,1,2
+64828,7,2,2,58,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,35505.352567,36934.11174,3,92,5,5,1.39,2,2,0,0,0,1,56,2,1,1,1
+64829,7,2,2,11,NA,2,2,2,11,134,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12985.951695,13541.394178,1,90,15,15,5,4,4,0,2,0,2,43,1,5,1,5
+64830,7,2,1,15,NA,1,1,1,15,181,NA,NA,1,1,NA,9,NA,NA,NA,2,1,1,2,2,1,1,2,2,1,15506.325263,15357.347514,2,103,2,2,0.42,3,3,0,2,0,2,51,2,2,5,NA
+64831,7,2,1,24,NA,5,7,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,10559.047286,11105.574109,2,94,5,5,2.2,2,1,0,0,0,1,23,NA,NA,5,NA
+64832,7,2,1,32,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,3,6,NA,1,2,2,1,2,1,2,2,2,2,40003.013263,41663.413523,1,91,7,7,2.1,3,3,0,1,0,2,29,2,3,6,NA
+64833,7,2,2,37,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,4,1,2,2,2,2,1,2,2,2,2,2,2,42456.72357,44069.75885,1,92,5,5,0.87,4,4,0,2,0,1,42,2,1,1,4
+64834,7,2,1,0,3,2,2,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,5271.338128,5219.242818,2,99,12,6,0.9,7,6,1,3,0,2,20,2,2,5,NA
+64835,7,2,1,39,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,1,2,1,2,2,1,2,2,NA,25120.174741,26325.85923,2,102,3,3,0.76,3,3,0,1,1,2,66,1,2,3,NA
+64836,7,2,2,0,2,2,2,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5803.542897,5681.443153,2,90,5,5,0.94,4,4,2,1,0,2,33,1,3,3,NA
+64837,7,2,2,11,NA,5,6,1,11,143,NA,NA,2,1,3,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5710.045626,5965.417422,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+64838,7,2,2,18,NA,4,4,1,18,218,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,10128.026158,10853.158449,2,103,2,1,0.43,2,1,0,0,0,1,20,1,3,6,NA
+64839,7,2,1,37,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,39017.692919,2,98,14,14,3.25,5,5,2,1,0,1,37,1,5,1,5
+64840,7,2,1,10,NA,4,4,2,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8655.162127,8668.652185,2,95,6,6,0.86,6,6,0,4,0,2,32,1,4,6,NA
+64841,7,2,1,8,NA,1,1,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,12622.023337,12821.760042,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+64842,7,2,1,2,NA,1,1,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13968.423539,13738.28453,2,102,7,7,1.57,4,4,2,0,0,2,34,2,2,1,5
+64843,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,12764.594076,12653.781126,2,90,6,6,0.96,5,5,0,1,0,1,55,1,4,6,NA
+64844,7,2,1,4,NA,4,4,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7431.820906,7890.591472,1,99,6,6,0.96,5,5,1,2,0,2,35,1,4,1,2
+64845,7,2,1,0,3,1,1,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6298.658963,6480.002411,2,94,9,9,2.29,5,5,2,1,0,2,33,2,3,1,1
+64846,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,12977.791943,13288.853946,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+64847,7,2,2,11,NA,4,4,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11784.279565,12220.817194,2,96,8,8,2.46,4,4,0,1,0,1,46,1,4,1,4
+64848,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,9535.353518,9981.414158,2,97,4,4,1.02,2,2,0,0,2,2,80,1,1,2,NA
+64849,7,2,1,0,5,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20370.716701,20998.185689,1,91,14,14,3.06,5,5,2,0,0,2,30,1,5,1,5
+64850,7,2,1,19,NA,3,3,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,99014.954203,98621.869511,1,92,10,10,3.4,3,3,0,0,0,1,56,1,4,1,5
+64851,7,2,2,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,20212.648666,20286.466668,2,95,4,3,1.1,2,1,0,0,0,2,47,1,3,4,NA
+64852,7,2,1,2,NA,4,4,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8009.966208,8832.297541,2,96,7,7,1.79,4,4,2,0,0,2,49,1,3,1,3
+64853,7,2,1,0,11,1,1,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,6999.189812,7438.261811,2,98,6,6,0.59,7,7,2,2,1,2,52,2,1,1,1
+64854,7,2,2,69,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,10561.745159,14124.187136,2,90,6,6,1.73,2,2,0,0,2,2,69,1,2,1,2
+64855,7,1,1,8,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9720.482616,0,2,91,4,4,0.65,5,5,1,3,0,1,43,2,3,5,NA
+64856,7,2,1,34,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,1,5,NA,2,2,2,2,2,2,NA,NA,NA,NA,30626.581617,30395.433124,2,90,6,6,0.96,5,5,1,1,0,1,39,2,2,1,NA
+64857,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,189736.955264,189051.893543,1,98,4,4,1.34,1,1,0,0,0,2,54,1,5,3,NA
+64858,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,36148.01469,38158.744241,1,101,3,3,0.66,2,2,0,0,1,2,65,1,2,3,NA
+64859,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,133853.800452,138128.446231,1,94,8,8,2.43,3,3,0,0,0,2,46,1,3,1,NA
+64860,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,23605.367258,23717.233847,2,97,7,7,1.89,3,3,0,0,0,1,50,1,2,1,2
+64861,7,2,2,26,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,NA,NA,NA,NA,19266.853809,20089.582382,2,102,8,8,1.72,5,5,0,2,1,1,63,2,5,1,5
+64862,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,22911.854766,1,95,5,5,0.92,5,5,1,2,0,2,30,1,4,1,4
+64863,7,1,2,34,NA,3,3,NA,NA,NA,2,NA,2,2,2,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,81658.419251,0,2,99,15,15,5,2,1,0,0,0,2,34,2,5,5,NA
+64864,7,2,1,12,NA,4,4,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,18413.211403,2,101,3,3,0.54,3,3,0,2,0,2,36,1,3,5,NA
+64865,7,2,2,2,NA,4,4,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6677.461889,6849.81851,2,90,5,1,0,3,1,1,1,0,2,48,1,5,5,NA
+64866,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,17794.144581,16918.878285,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+64867,7,2,1,78,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,12856.852517,14031.201295,2,98,5,5,1.24,3,3,0,0,1,2,58,1,2,5,NA
+64868,7,2,1,5,NA,1,1,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17973.290893,3,92,3,3,0.54,4,4,3,0,0,2,22,1,3,5,NA
+64869,7,2,2,11,NA,5,6,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7751.448366,8179.286646,1,96,15,15,4.34,4,4,1,1,0,1,36,2,5,1,5
+64870,7,2,2,80,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,2,NA,1,1,1,1,2,2,1,2,1,NA,18698.205673,19635.336647,1,97,9,9,1.78,6,6,0,1,1,1,45,2,3,1,3
+64871,7,2,2,49,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,141912.982157,152670.471787,1,97,14,14,3.36,4,4,0,2,0,2,49,1,5,1,5
+64872,7,2,2,2,NA,1,1,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12871.484115,13281.030392,2,102,5,5,0.89,4,4,1,2,0,2,36,2,5,3,NA
+64873,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,111065.717962,115110.76456,2,91,15,15,5,4,4,0,2,0,1,53,1,5,1,4
+64874,7,1,1,80,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,8168.05279,0,2,98,3,3,0.56,4,4,0,0,2,2,79,1,1,1,1
+64875,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,99606.074294,100835.770786,1,93,15,15,5,2,2,0,0,0,2,34,1,5,1,5
+64876,7,2,1,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,28680.660607,28606.985896,3,91,6,3,1.33,2,1,0,0,0,1,40,1,3,5,NA
+64877,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19260.892847,19199.459573,2,97,1,1,0.09,2,1,0,0,0,1,46,1,3,5,NA
+64878,7,2,1,12,NA,3,3,2,12,150,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19197.94644,18945.641357,1,98,6,6,1.73,3,3,0,1,0,2,39,1,4,1,1
+64879,7,2,2,8,NA,4,4,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9153.600624,9398.485171,1,100,8,8,1.61,6,6,1,3,0,1,29,1,5,6,NA
+64880,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,153565.050575,157855.235487,1,97,15,15,5,2,2,0,0,1,2,49,1,5,1,3
+64881,7,2,1,33,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15014.015332,15213.152707,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+64882,7,2,1,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,160284.827389,161788.081948,2,102,14,14,5,2,2,0,0,2,2,65,1,5,1,5
+64883,7,2,1,18,NA,2,2,2,18,218,2,NA,2,2,1,10,NA,NA,NA,2,2,2,1,2,2,2,2,2,2,13752.835112,15116.811136,3,90,7,7,1.48,5,5,0,1,0,1,43,2,1,6,NA
+64884,7,2,1,8,NA,4,4,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7908.091112,7920.416775,2,93,8,8,1.67,5,5,1,1,0,2,31,1,4,5,NA
+64885,7,2,1,0,7,4,4,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6275.847063,6684.003207,2,100,10,10,2.75,5,5,1,1,1,1,27,1,3,1,5
+64886,7,2,2,59,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,23476.411085,23902.152524,3,91,4,4,0.81,3,3,0,0,1,1,64,2,1,1,1
+64887,7,2,1,13,NA,3,3,2,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,32477.57544,34497.087834,1,95,7,7,1.66,5,5,0,3,0,1,34,1,2,1,4
+64888,7,2,2,2,NA,4,4,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8072.012312,8280.364646,3,91,8,8,2.24,4,4,2,0,0,1,39,2,3,1,3
+64889,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,20933.43123,22068.787492,2,95,6,6,1.19,4,4,0,1,0,1,44,1,3,1,2
+64890,7,2,1,29,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,15173.157442,14818.726738,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+64891,7,2,2,11,NA,2,2,1,11,138,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,14446.475406,15838.541371,1,103,7,7,1.03,7,7,0,3,0,1,50,2,1,1,1
+64892,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,17171.250195,2,100,5,5,1.08,3,3,0,1,0,2,50,1,4,3,NA
+64893,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,76027.409363,76707.146234,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+64894,7,2,2,6,NA,2,2,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15583.587534,16935.930722,1,93,14,14,3.06,5,5,0,2,0,1,46,2,1,1,4
+64895,7,2,1,67,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7380.991723,2,100,14,14,3.58,4,4,0,1,1,2,55,1,5,1,4
+64896,7,2,2,0,0,4,4,2,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4706.214557,5182.816301,1,96,6,6,1.21,4,4,2,0,0,1,24,1,4,1,3
+64897,7,2,1,42,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,12254.763576,14641.804715,2,90,4,4,0.81,3,3,0,1,0,2,41,2,2,6,NA
+64898,7,2,1,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,17036.36313,16982.025069,1,99,2,2,0.48,2,2,0,0,0,2,44,1,3,1,3
+64899,7,2,2,13,NA,5,6,1,13,157,NA,NA,2,2,2,7,NA,NA,NA,1,1,1,1,2,1,1,2,1,NA,8868.843265,9252.179416,2,92,8,8,2.43,3,3,0,1,1,2,58,NA,5,1,5
+64900,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,42894.724338,43561.362107,1,98,3,3,0.93,2,2,0,0,0,1,21,1,4,5,NA
+64901,7,2,2,18,NA,4,4,1,18,216,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10143.901078,10335.966318,2,95,2,2,0.67,2,2,0,0,1,2,61,1,3,3,NA
+64902,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,21490.337905,2,97,3,2,0.83,2,1,0,0,0,1,30,1,4,5,NA
+64903,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23530.130278,1,95,5,5,1.84,3,1,0,0,0,1,35,1,3,6,NA
+64904,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,34802.051557,36349.694696,1,95,6,4,1.7,2,1,0,0,0,1,53,1,2,6,NA
+64905,7,2,2,1,12,3,3,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,41219.757096,41761.398587,1,93,15,15,5,3,3,1,0,0,1,35,1,5,1,5
+64906,7,2,1,80,NA,5,7,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,1,NA,13312.080419,14392.211707,1,102,10,10,3.22,4,4,0,0,2,2,29,2,5,5,NA
+64907,7,1,1,41,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,126789.52929,0,1,101,15,15,5,4,4,0,2,0,2,40,1,4,1,3
+64908,7,2,2,60,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,13057.178942,13648.591881,1,102,6,6,2.94,1,1,0,0,1,2,60,1,4,3,NA
+64909,7,2,1,74,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,12468.859946,13204.521198,2,96,7,7,1.33,6,6,0,3,1,1,74,1,1,1,NA
+64910,7,2,2,15,NA,3,3,2,15,182,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71837.483011,75155.181458,1,91,14,14,3.06,5,5,0,3,0,2,46,1,5,1,5
+64911,7,2,2,15,NA,5,6,2,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5956.736263,6113.749553,2,100,15,15,5,3,3,0,1,0,1,48,2,5,1,5
+64912,7,2,2,9,NA,4,4,2,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8226.044997,8398.377256,2,95,1,1,0,4,4,2,1,0,2,27,1,4,5,NA
+64913,7,2,1,12,NA,3,3,2,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26824.630008,26493.112081,1,95,1,1,0.17,2,2,0,1,0,2,47,1,2,3,NA
+64914,7,2,1,31,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,3,17165.91562,17929.203991,2,96,15,7,3.67,3,1,0,0,0,1,31,2,5,1,NA
+64915,7,2,1,4,NA,4,4,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8626.000813,9278.769273,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+64916,7,2,1,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15787.238477,16796.740962,2,98,4,3,1.01,2,1,0,0,0,1,46,NA,NA,77,NA
+64917,7,2,1,5,NA,5,6,2,5,67,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8197.191196,8439.736714,1,93,7,7,2.32,3,3,1,0,0,1,36,2,5,1,5
+64918,7,2,1,1,16,3,3,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24594.444896,28818.16319,1,98,5,5,1.05,3,3,1,0,0,1,24,1,3,1,3
+64919,7,2,1,10,NA,5,6,2,10,127,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,9720.482616,10818.978947,2,91,12,12,NA,7,6,0,4,2,2,72,2,1,2,NA
+64920,7,2,2,8,NA,4,4,2,8,103,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8395.64512,8728.346696,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+64921,7,2,2,16,NA,4,4,2,16,202,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,5,5,0.92,5,5,0,3,0,2,54,1,3,2,NA
+64922,7,2,1,9,NA,5,6,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8376.521179,8870.870608,1,92,15,15,5,4,4,0,2,0,1,55,1,5,1,5
+64923,7,2,2,64,NA,4,4,2,NA,NA,2,NA,2,2,5,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9113.905743,12222.082956,3,90,8,8,3.3,2,2,0,0,1,2,64,2,2,1,NA
+64924,7,2,2,10,NA,5,7,1,10,129,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,12942.575479,2,101,7,7,1.88,4,4,0,2,0,2,36,1,4,1,5
+64925,7,2,2,0,4,2,2,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7534.765808,7789.750955,1,98,3,3,0.37,5,5,2,1,0,1,44,2,4,1,3
+64926,7,2,1,14,NA,4,4,2,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13352.786991,13379.186543,2,97,7,7,1.06,7,7,1,2,0,2,40,1,4,5,NA
+64927,7,2,2,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,7498.263904,2,100,3,3,0.39,7,7,3,3,0,2,30,1,2,5,NA
+64928,7,2,1,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,1,2,NA,14073.595908,15970.731967,1,92,6,6,1.3,4,4,0,1,1,1,25,1,1,1,3
+64929,7,2,1,19,NA,4,4,2,19,228,2,NA,1,1,NA,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10791.290138,11351.304534,3,91,2,2,0.25,4,4,0,2,0,2,35,1,3,5,NA
+64930,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,13219.758853,13936.470139,2,97,4,4,1.29,2,2,0,0,2,1,74,1,3,3,NA
+64931,7,2,2,59,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26668.458882,27266.525973,2,102,6,6,2.04,2,2,0,0,1,1,64,1,3,1,4
+64932,7,2,1,6,NA,3,3,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,86609.650237,90490.601456,2,102,14,14,4.93,3,3,0,1,0,1,37,1,5,1,5
+64933,7,2,2,41,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,31301.615973,33816.788039,1,90,15,15,4.77,4,4,1,1,0,2,41,1,5,1,2
+64934,7,2,2,16,NA,5,6,2,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,5607.756021,6010.481547,3,91,6,6,1.81,3,3,0,1,0,2,47,2,3,1,3
+64935,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,73820.224093,75002.443492,2,95,9,9,2.22,5,5,1,0,0,1,55,1,4,1,5
+64936,7,2,2,6,NA,2,2,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17458.526997,17628.076413,1,94,2,2,0.26,4,4,2,1,0,2,25,1,4,5,NA
+64937,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,37456.985636,43452.233213,2,98,6,6,1.7,2,2,0,0,2,2,80,1,2,1,2
+64938,7,2,1,40,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,27938.918808,28342.234566,1,100,7,7,3.4,1,1,0,0,0,1,40,1,5,5,NA
+64939,7,2,1,19,NA,4,4,2,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,1,1,0.04,2,1,0,0,0,1,19,1,4,NA,NA
+64940,7,2,1,13,NA,3,3,2,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,91343.020442,91081.271456,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+64941,7,2,2,2,NA,1,1,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,12710.400836,2,98,2,2,0.27,4,4,2,1,0,2,32,2,2,5,NA
+64942,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,7101.739553,7433.956549,2,100,3,3,1.19,1,1,0,0,1,1,69,1,2,2,NA
+64943,7,2,1,1,20,5,6,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7638.291796,7864.300119,1,93,14,14,4.86,3,3,1,0,0,1,30,2,5,1,5
+64944,7,2,2,18,NA,4,4,2,18,225,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15259.822784,15553.106132,1,90,5,5,0.74,5,5,0,2,0,2,18,1,4,NA,NA
+64945,7,1,1,37,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,105412.227726,0,2,101,8,8,1.72,5,5,0,3,0,1,37,1,3,1,3
+64946,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,5866.209951,5912.09729,2,95,5,5,1.05,3,3,0,0,2,1,60,1,1,1,4
+64947,7,2,1,6,NA,5,6,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7491.978458,7984.984584,2,100,15,15,5,4,4,1,1,0,1,36,2,5,1,5
+64948,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8679.600111,2,95,5,5,1.92,1,1,0,0,1,2,61,1,4,2,NA
+64949,7,2,2,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,110941.813533,112262.474161,1,97,15,15,4.77,4,4,0,0,0,1,56,1,4,1,4
+64950,7,2,2,6,NA,2,2,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11210.83392,12007.401767,2,90,6,6,1.34,4,4,1,2,0,2,36,2,3,77,NA
+64951,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,107912.705683,113084.898879,3,91,4,4,1.02,2,2,0,0,0,1,22,1,5,1,5
+64952,7,2,2,8,NA,4,4,2,8,102,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7120.704736,7311.203604,1,99,8,8,1.76,5,5,0,2,1,1,37,1,4,1,3
+64953,7,2,2,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,30863.871606,31440.435679,1,101,3,3,1.12,1,1,0,0,0,2,44,1,4,1,NA
+64954,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,1,3,NA,1,2,1,1,2,2,NA,NA,NA,NA,10964.859884,11022.841433,3,90,7,7,1.82,4,4,1,0,0,2,54,2,1,3,NA
+64955,7,2,1,3,NA,4,4,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11342.022131,11810.675983,2,102,15,15,3.82,5,5,1,2,0,1,34,1,3,1,4
+64956,7,2,1,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,27572.205373,28421.693812,1,100,15,15,5,3,3,0,0,0,2,33,1,5,1,4
+64957,7,1,2,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11623.462338,0,1,100,3,3,0.73,3,3,1,1,0,2,32,1,3,5,NA
+64958,7,2,2,8,NA,1,1,1,8,107,NA,NA,1,1,NA,3,NA,NA,NA,2,1,1,1,2,2,1,2,2,1,16217.354723,16799.750785,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+64959,7,2,2,6,NA,2,2,2,6,82,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17962.96045,18326.284291,1,97,15,15,4.52,6,6,0,4,0,2,41,1,5,1,5
+64960,7,2,2,6,NA,3,3,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13450.606713,13706.553337,3,91,5,5,1.07,4,4,0,2,0,2,36,1,5,1,4
+64961,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,141773.363283,142016.391545,1,97,15,15,4.77,4,4,0,0,0,1,56,1,4,1,4
+64962,7,2,2,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,39483.840194,39476.565757,2,102,14,14,3.25,5,5,1,1,0,2,32,1,4,1,3
+64963,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,20666.038321,20600.123288,1,96,14,14,5,1,1,0,0,0,1,55,1,5,5,NA
+64964,7,2,1,80,NA,3,3,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,2,1,2,1,NA,35161.248998,42165.766203,3,91,15,15,3.33,6,6,0,2,2,1,80,2,3,1,3
+64965,7,2,1,26,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,71402.235366,73020.331081,2,97,15,15,4.97,5,5,1,0,0,1,48,1,4,1,3
+64966,7,2,1,6,NA,3,3,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57305.501166,60591.443404,1,94,15,15,5,3,3,0,1,0,1,46,1,5,1,5
+64967,7,1,2,58,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,24905.670199,0,2,101,1,1,0.15,3,3,0,2,0,2,58,1,3,5,NA
+64968,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16567.527819,16240.172481,1,96,15,15,5,4,4,0,1,0,1,56,1,4,1,5
+64969,7,2,1,18,NA,3,3,1,18,224,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24849.884345,24917.808193,3,91,5,5,0.87,4,4,0,2,0,2,38,1,2,3,NA
+64970,7,2,1,12,NA,2,2,1,12,149,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15626.107676,15463.700675,2,93,9,9,1.94,6,6,0,3,0,2,37,NA,NA,3,NA
+64971,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,6358.062034,6381.990744,1,99,8,8,3.4,2,2,0,0,1,1,60,1,3,3,NA
+64972,7,2,1,61,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7289.557268,8219.094942,3,90,1,1,0,2,2,0,0,1,1,61,2,3,1,3
+64973,7,2,1,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,35869.019314,36329.101667,3,92,4,4,1.13,2,2,0,0,2,1,64,1,3,1,4
+64974,7,2,2,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,27585.470618,27304.927199,2,102,7,7,2.65,2,2,1,0,0,2,42,1,4,5,NA
+64975,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,23016.024704,24471.525432,1,96,7,7,2.78,2,2,0,0,0,1,48,1,2,6,NA
+64976,7,2,1,69,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7380.991723,2,100,10,10,2.33,6,6,0,2,2,2,35,1,2,5,NA
+64977,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15732.436891,17625.53878,1,97,2,2,0.72,1,1,0,0,1,2,63,1,3,5,NA
+64978,7,2,2,56,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,24548.478709,25991.798892,1,93,5,5,1.43,2,2,0,0,1,1,60,2,5,1,4
+64979,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18822.975485,18307.835382,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+64980,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,33147.414266,37148.276517,2,94,99,99,NA,1,1,0,0,1,2,80,1,4,2,NA
+64981,7,2,1,46,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,1,1,2,2,1,19436.026093,20775.484483,2,97,5,5,0.8,5,5,1,2,0,1,46,2,4,1,2
+64982,7,2,1,16,NA,3,3,2,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,66448.116716,66629.743791,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+64983,7,2,2,76,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,17836.372654,19170.472552,2,96,6,6,1.77,2,2,0,0,2,1,77,1,2,1,2
+64984,7,2,2,17,NA,3,3,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,76360.13568,92040.479005,1,99,NA,NA,NA,4,4,0,1,0,1,50,NA,NA,1,NA
+64985,7,2,1,65,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,7410.50521,7700.344649,2,96,6,6,2.75,1,1,0,0,1,1,65,1,2,2,NA
+64986,7,1,2,28,NA,5,6,NA,NA,NA,2,NA,2,1,4,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,16937.04417,0,1,95,3,3,0.43,4,4,0,1,2,1,65,2,5,1,3
+64987,7,2,1,17,NA,4,4,1,17,210,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18413.211403,2,101,1,1,0.27,3,3,0,2,0,2,36,1,3,5,NA
+64988,7,2,2,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8897.78821,9699.858422,1,96,7,7,1,7,7,2,1,1,2,53,1,4,1,3
+64989,7,2,1,11,NA,2,2,2,11,138,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13217.721247,13515.808466,2,91,8,8,2.97,3,2,0,2,0,2,33,2,4,5,NA
+64990,7,1,1,2,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7390.005875,0,1,91,15,15,5,5,5,2,1,0,1,40,1,5,1,5
+64991,7,2,2,10,NA,5,6,2,10,128,NA,NA,1,1,NA,5,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6198.268014,6618.822195,3,90,15,15,3.23,6,6,0,2,0,1,50,2,2,1,2
+64992,7,2,1,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,15176.622228,16064.120023,1,101,3,3,1.12,1,1,0,0,1,1,76,1,5,2,NA
+64993,7,1,1,57,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,16287.780872,0,2,100,15,14,5,2,1,0,0,0,1,57,1,5,6,NA
+64994,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,106239.028397,111331.003215,2,92,10,5,1.96,2,1,0,0,0,2,27,2,5,1,NA
+64995,7,2,2,68,NA,3,3,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,35700.429895,37686.262589,1,93,2,2,0.41,2,2,0,0,2,2,68,2,1,1,1
+64996,7,2,1,0,6,3,3,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19091.539058,20438.314938,2,95,8,8,2.17,4,4,1,1,0,1,43,1,4,1,5
+64997,7,2,2,15,NA,2,2,2,15,184,NA,NA,1,1,NA,66,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13824.001771,15649.805677,2,90,2,2,0.3,3,3,0,1,0,2,51,2,2,5,NA
+64998,7,2,1,54,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,144858.963765,145378.942103,2,97,9,9,5,1,1,0,0,0,1,54,1,5,5,NA
+64999,7,2,1,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,19434.054198,2,100,5,5,1.63,3,2,0,1,0,2,50,1,2,5,NA
+65000,7,2,1,4,NA,2,2,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12403.412256,12420.172189,2,90,3,3,0.54,4,4,1,2,0,2,33,2,1,4,NA
+65001,7,2,2,13,NA,5,7,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8777.486283,9116.900706,1,91,15,15,5,5,5,0,3,0,1,40,1,5,1,5
+65002,7,2,1,68,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,NA,NA,NA,1,2,2,1,140267.560043,144186.249663,2,91,12,NA,NA,2,1,0,0,2,2,68,1,4,6,NA
+65003,7,2,1,9,NA,3,3,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17738.769196,18719.457495,1,98,4,4,1,3,3,0,1,1,1,65,1,2,1,NA
+65004,7,2,2,25,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,2,2,2,2,2,2,2,NA,NA,NA,NA,42557.597671,44940.417695,1,100,13,13,NA,4,4,1,1,0,1,28,2,1,1,1
+65005,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,73204.401173,96856.987278,2,91,12,12,NA,2,2,0,0,2,2,71,1,2,1,NA
+65006,7,2,2,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,25221.447441,25652.65048,1,100,14,14,3.93,3,3,0,1,0,2,47,1,5,4,NA
+65007,7,2,2,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,18801.993237,20649.846581,2,96,14,14,5,2,2,0,0,0,1,30,2,5,1,5
+65008,7,1,1,80,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11740.600601,0,3,91,13,13,NA,3,3,0,0,2,2,80,1,3,1,1
+65009,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21491.090123,20964.324934,1,97,6,6,1.41,3,3,0,1,0,2,51,1,4,5,NA
+65010,7,2,1,6,NA,4,4,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10469.725162,10967.571132,1,100,15,15,3.87,6,6,1,3,0,2,39,1,4,1,4
+65011,7,2,1,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18289.793332,19321.216647,1,99,6,6,0.96,5,5,1,2,0,2,35,1,4,1,2
+65012,7,2,2,68,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,2,2,NA,1,2,1,1,2,1,1,2,1,3,12312.400687,12851.340698,1,103,2,2,0.45,1,1,0,0,1,2,68,2,2,2,NA
+65013,7,2,2,7,NA,2,2,2,7,87,NA,NA,2,1,1,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9872.244853,10294.506103,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+65014,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,166234.629208,1,95,9,9,4.01,2,2,0,0,1,1,60,1,3,1,3
+65015,7,2,1,19,NA,4,4,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,99,1,0.09,4,1,0,0,0,1,18,2,4,NA,NA
+65016,7,2,1,11,NA,3,3,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22651.436723,23515.701302,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+65017,7,2,1,4,NA,4,4,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10859.969359,11974.89205,1,101,2,2,0.47,3,3,1,0,0,1,35,1,2,6,NA
+65018,7,2,2,14,NA,2,2,2,14,176,NA,NA,2,1,1,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18254.385443,19735.623608,2,91,12,12,NA,5,5,0,1,1,2,43,2,3,6,NA
+65019,7,2,2,9,NA,4,4,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11076.064101,11269.278002,2,102,15,15,4.2,5,5,1,2,0,2,29,NA,NA,1,NA
+65020,7,2,1,62,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,8609.250304,11228.904188,2,90,2,2,0.73,1,1,0,0,1,1,62,2,1,4,NA
+65021,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,96524.78948,102108.384344,1,97,7,7,1.87,4,4,1,1,0,1,35,1,2,1,4
+65022,7,2,2,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,1,2,2,1,2,2,1,13358.458751,14155.588359,2,92,15,15,5,3,1,0,0,0,2,33,NA,NA,5,NA
+65023,7,2,2,13,NA,4,4,2,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,6,6,0.86,6,6,0,4,0,2,32,1,4,6,NA
+65024,7,2,1,9,NA,2,2,1,9,109,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16281.509392,17100.217351,2,91,6,6,0.93,5,5,1,2,0,1,34,1,2,1,3
+65025,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,22249.799905,2,97,5,5,1.08,3,3,0,0,0,1,38,1,4,5,NA
+65026,7,2,2,0,11,4,4,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5058.520291,5113.753371,1,100,7,7,1.65,4,4,2,0,0,2,24,1,4,1,3
+65027,7,2,2,23,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,4,5,2,2,2,2,1,2,2,2,2,2,2,35424.746838,35270.456492,2,93,4,4,0.56,5,5,0,0,0,2,49,2,2,5,NA
+65028,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,15682.233511,17550.107712,1,101,3,3,0.95,2,2,0,0,2,1,80,1,2,1,NA
+65029,7,2,2,6,NA,2,2,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19936.606751,20692.800636,1,94,7,7,1.74,4,4,0,2,0,1,44,1,5,1,5
+65030,7,2,2,5,NA,4,4,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8041.669581,8652.926173,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+65031,7,2,1,16,NA,2,2,1,16,199,NA,NA,2,2,1,8,NA,NA,NA,2,2,2,2,2,2,1,2,2,2,22124.028915,23090.48094,1,100,5,1,0,6,1,1,2,0,2,40,2,1,5,NA
+65032,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,72094.421241,74006.954437,2,95,6,6,1.95,2,2,0,0,2,1,80,1,1,1,3
+65033,7,2,2,7,NA,4,4,1,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9453.111053,10094.338553,1,100,8,8,1.95,4,4,0,2,0,2,42,1,4,1,4
+65034,7,2,2,16,NA,4,4,2,17,204,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12161.822321,12177.990347,2,95,5,5,1.05,3,3,0,1,1,1,63,1,2,1,3
+65035,7,1,1,19,NA,1,1,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,18042.255087,0,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+65036,7,2,2,49,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,14814.740715,14686.129879,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+65037,7,1,1,18,NA,3,3,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,26749.020961,0,1,101,4,4,0.79,3,3,0,0,0,2,47,1,5,3,NA
+65038,7,2,1,13,NA,2,2,1,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21898.969807,22012.999264,2,102,15,12,NA,5,4,0,3,0,1,42,2,4,6,NA
+65039,7,2,2,6,NA,5,6,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8067.514021,8614.897055,1,90,14,14,3.33,5,5,1,2,0,1,41,1,5,1,5
+65040,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18710.568399,18832.614147,2,92,12,12,NA,2,1,0,0,1,2,58,1,3,5,NA
+65041,7,2,2,11,NA,5,6,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8290.163782,8692.019106,1,92,14,14,3.69,4,4,0,2,0,1,47,2,4,4,NA
+65042,7,2,1,14,NA,5,7,2,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15861.375368,15958.829994,1,95,7,7,1.57,4,4,0,2,0,1,39,1,3,1,3
+65043,7,2,2,7,NA,1,1,1,7,92,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15841.451259,16252.614023,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+65044,7,2,2,15,NA,5,6,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10142.281747,10534.471105,3,91,15,15,5,4,4,0,2,0,1,44,2,5,1,5
+65045,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,NA,10479.637868,11291.375185,1,92,2,2,0.87,1,1,0,0,1,1,80,1,4,5,NA
+65046,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,13704.691347,15557.045236,1,90,15,15,5,2,2,0,0,2,1,74,1,3,1,3
+65047,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,140129.484883,1,101,14,14,3.3,4,4,0,2,0,2,42,1,4,1,3
+65048,7,2,1,61,NA,1,1,1,NA,NA,2,NA,2,2,9,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,10596.142548,10802.656005,1,94,7,7,1.52,4,4,0,2,2,1,61,2,1,1,5
+65049,7,2,1,24,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,37911.437415,38428.199561,2,103,7,7,2.64,2,2,0,0,0,2,21,2,3,1,3
+65050,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,39565.288792,43332.356399,1,99,10,10,4.89,2,2,0,0,2,2,63,1,5,5,NA
+65051,7,2,2,80,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,1,2,NA,2,2,2,1,2,2,1,2,2,NA,17318.187297,23904.945555,2,90,3,3,0.92,1,1,0,0,1,2,80,2,1,2,NA
+65052,7,2,1,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,109309.268477,109028.475475,3,91,15,15,5,2,2,0,1,0,1,42,1,5,2,NA
+65053,7,2,2,0,4,1,1,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7565.515117,7406.345541,1,100,6,6,1.57,3,3,1,0,0,1,39,1,3,1,5
+65054,7,2,1,10,NA,3,3,1,10,127,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,32654.748828,36152.216038,1,94,5,5,0.87,4,4,0,1,0,1,40,1,5,1,5
+65055,7,1,1,33,NA,5,6,NA,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,2,1,2,1,NA,NA,NA,NA,15014.015332,0,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+65056,7,2,1,76,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,8992.410435,9337.125921,2,90,9,9,3.02,3,3,0,0,2,2,70,1,4,1,2
+65057,7,2,2,3,NA,5,6,2,3,44,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,3864.413878,4049.242646,1,99,6,6,1.07,6,6,2,1,2,1,44,2,5,4,NA
+65058,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,23338.32018,25560.39495,1,99,77,77,NA,2,2,0,0,2,1,80,1,5,1,4
+65059,7,2,2,0,1,4,4,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4128.726485,4173.807313,2,93,6,6,1.08,4,4,2,0,0,1,25,1,3,6,NA
+65060,7,2,1,0,6,4,4,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6275.847063,6487.671462,2,93,9,9,2.86,4,4,1,1,0,1,30,1,4,6,NA
+65061,7,2,1,38,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19091.246741,19659.303383,1,91,10,10,3.22,4,4,1,1,0,1,38,2,5,1,5
+65062,7,2,1,10,NA,4,4,2,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8836.464036,8942.399437,1,96,15,15,5,5,5,0,3,0,2,47,1,5,1,5
+65063,7,2,1,28,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19137.192683,18862.909907,1,96,14,14,5,2,2,0,0,0,2,47,2,4,5,NA
+65064,7,2,1,7,NA,1,1,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13665.416457,13541.311863,1,94,2,2,0.27,5,5,0,4,0,2,47,2,1,4,NA
+65065,7,2,2,34,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,77299.255327,80009.772488,2,92,10,10,4.89,2,2,0,0,0,2,34,2,5,1,5
+65066,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,33913.235726,1,95,7,7,2.16,3,3,0,0,1,1,45,1,3,1,4
+65067,7,2,2,42,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,118432.300444,124617.656629,2,93,9,9,3.77,2,2,0,0,0,2,42,2,4,1,5
+65068,7,2,1,0,8,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5572.446681,5657.975893,1,99,10,10,3.51,3,3,1,0,0,1,27,1,4,1,4
+65069,7,2,1,62,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,11397.665899,11580.241952,1,101,3,3,0.41,5,5,0,2,1,2,36,2,4,4,NA
+65070,7,2,1,3,NA,2,2,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11186.576454,11318.825779,2,90,3,3,0.46,5,5,3,0,0,2,22,1,2,5,NA
+65071,7,2,2,79,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,55250.133125,58713.117248,1,91,14,14,3.25,4,4,0,0,1,1,50,1,2,1,3
+65072,7,2,2,27,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,48991.867549,51378.468052,1,93,14,14,5,2,2,0,0,0,2,59,2,5,5,NA
+65073,7,2,2,4,NA,3,3,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,52402.97499,54053.880354,1,91,15,15,4.59,4,4,2,0,0,1,35,1,5,1,5
+65074,7,2,1,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,21924.03349,22615.591521,2,98,10,10,4.76,2,2,0,0,0,1,42,1,2,6,NA
+65075,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108335.731552,113413.419502,1,98,15,15,4.34,4,4,0,2,0,1,51,1,5,1,5
+65076,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16859.368198,16397.96736,1,99,8,8,1.99,5,5,1,0,0,1,55,1,5,1,2
+65077,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,5,1,NA,2,2,2,2,2,2,2,2,2,2,9068.437099,9213.701881,2,93,77,77,NA,2,2,0,0,1,2,50,NA,NA,1,5
+65078,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,51218.356583,59416.206008,2,98,2,2,0.77,1,1,0,0,1,2,80,1,1,2,NA
+65079,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,29733.812317,30162.048345,1,101,4,4,0.99,2,2,0,1,0,2,35,1,3,5,NA
+65080,7,2,1,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,7005.52791,7142.750358,2,99,6,6,2.24,1,1,0,0,1,1,74,1,4,2,NA
+65081,7,2,1,8,NA,2,2,1,8,101,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,12577.115885,12876.06354,2,96,6,6,1.25,4,4,1,1,0,1,31,2,3,1,3
+65082,7,2,2,14,NA,1,1,1,14,173,NA,NA,1,1,NA,8,NA,NA,NA,2,1,1,1,2,2,1,2,2,1,20830.737445,21782.111761,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+65083,7,2,2,0,2,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7798.643057,7634.568742,1,102,6,6,1.73,3,3,1,0,0,2,24,2,5,1,5
+65084,7,2,2,26,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18299.168537,17535.171386,2,100,5,5,0.95,4,4,0,0,1,2,53,1,3,5,NA
+65085,7,2,1,51,NA,4,4,1,NA,NA,2,NA,2,2,5,NA,3,1,NA,1,2,1,1,2,1,1,2,1,NA,18061.358948,18950.803173,2,93,4,4,0.69,4,4,0,0,0,1,23,2,3,5,NA
+65086,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,98850.665857,102087.416196,1,103,15,15,5,3,3,0,1,0,1,46,1,5,1,5
+65087,7,2,2,74,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,2,NA,2,2,2,1,2,2,1,2,2,NA,17318.187297,19970.708273,2,90,7,7,1.57,4,4,0,0,1,2,20,1,2,5,NA
+65088,7,2,2,37,NA,3,3,2,NA,NA,2,NA,2,2,1,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,95214.22557,102645.939709,1,97,15,1,0.4,5,1,1,1,0,2,38,NA,NA,1,5
+65089,7,2,2,3,NA,4,4,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9630.497627,10362.522874,1,96,13,13,NA,5,5,1,1,0,1,42,1,3,5,NA
+65090,7,2,2,18,NA,5,7,1,18,218,2,NA,2,2,2,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6391.016092,6494.382638,1,100,99,99,NA,6,6,0,1,0,1,53,2,2,1,3
+65091,7,2,1,33,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23952.480361,1,95,5,5,1.03,4,4,0,2,0,1,33,1,3,1,3
+65092,7,2,1,8,NA,2,2,2,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8613.834494,9046.977066,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+65093,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,6,6,1.67,3,3,0,0,0,2,22,1,4,5,NA
+65094,7,2,1,10,NA,4,4,1,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,14062.200951,2,101,6,6,0.96,5,5,0,4,0,2,36,1,4,4,NA
+65095,7,2,2,41,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,4,1,1,2,2,2,2,2,2,2,2,1,2,33716.655399,34130.121207,1,100,7,7,1.3,5,5,0,3,0,1,43,2,2,1,4
+65096,7,2,2,7,NA,1,1,1,8,96,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15962.145468,16535.37648,2,98,3,3,0.33,7,7,2,3,0,1,40,2,1,1,1
+65097,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,28280.669788,29879.976389,1,99,14,14,5,2,2,0,0,2,2,79,1,3,1,5
+65098,7,2,2,68,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,12230.621635,1,96,7,7,1.39,5,5,0,2,2,1,69,2,2,1,2
+65099,7,2,2,1,18,4,4,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5987.067673,6442.151602,2,93,4,4,0.56,5,5,2,1,0,1,27,1,2,6,NA
+65100,7,2,2,2,NA,5,6,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5520.445386,5614.569811,2,100,15,15,5,4,4,1,1,0,1,41,2,5,1,5
+65101,7,2,1,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,10310.165525,10913.082896,1,93,3,3,0.92,1,1,0,0,1,1,75,1,5,3,NA
+65102,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,17884.885732,18259.455804,2,95,3,3,0.52,3,3,1,0,0,1,37,1,4,1,4
+65103,7,2,1,12,NA,4,4,2,12,149,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10453.50644,10655.928666,1,90,14,14,2.96,5,5,1,2,0,1,31,1,5,1,4
+65104,7,2,2,8,NA,3,3,2,9,108,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,80369.555824,80552.420473,1,97,15,15,3.89,5,5,0,2,0,1,50,1,4,6,NA
+65105,7,2,2,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,24654.107413,23441.410215,1,100,10,10,2.59,5,5,0,1,0,2,40,1,5,1,NA
+65106,7,2,1,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,20880.649884,21300.208451,1,96,14,6,2.69,2,1,0,0,0,2,29,1,5,6,NA
+65107,7,2,1,14,NA,1,1,2,14,174,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,20398.562455,21727.336109,2,94,5,5,0.67,6,6,1,3,0,1,37,2,3,1,4
+65108,7,2,2,8,NA,3,3,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55469.656717,57246.428971,1,98,9,9,2.15,5,5,0,3,0,2,32,1,3,1,4
+65109,7,2,1,3,NA,5,6,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10273.602479,11522.969217,2,98,9,9,3.83,2,2,1,0,0,2,29,2,4,5,NA
+65110,7,2,2,23,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,16929.836231,18593.694637,2,101,4,2,0.55,2,1,0,0,0,2,23,2,4,5,NA
+65111,7,2,1,14,NA,3,3,2,14,177,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18645.590959,18737.183314,2,97,9,9,1.45,7,7,1,2,2,2,45,1,3,5,NA
+65112,7,2,2,60,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,12592.413049,13033.910742,1,93,8,8,4.13,1,1,0,0,1,2,60,2,4,3,NA
+65113,7,2,1,43,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,11558.146957,12085.615817,3,90,6,6,1.3,4,4,0,2,0,2,37,2,4,1,3
+65114,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,27375.405353,30621.081524,3,91,5,5,0.87,4,4,0,2,0,2,38,1,2,3,NA
+65115,7,2,2,19,NA,4,4,1,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,2,1,0.23,4,1,0,0,0,2,19,1,4,NA,NA
+65116,7,2,1,0,10,2,2,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4884.343512,4884.433539,3,90,7,7,2.1,3,3,1,0,0,1,34,2,3,1,2
+65117,7,2,1,16,NA,4,4,2,16,203,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11351.725436,12116.15399,2,95,14,14,4.58,3,3,0,1,1,1,61,1,4,1,5
+65118,7,2,1,8,NA,2,2,2,8,99,NA,NA,2,1,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10872.115681,10826.020253,1,99,77,77,NA,3,3,0,1,0,1,52,1,5,1,5
+65119,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,66503.043118,66707.517935,2,98,8,8,1.8,5,5,2,1,0,1,32,1,4,1,5
+65120,7,2,1,8,NA,2,2,1,8,100,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14570.588291,14577.04302,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+65121,7,2,1,3,NA,2,2,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,15745.774489,15931.923312,2,91,2,2,0.19,5,5,3,0,0,1,24,2,1,1,3
+65122,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114993.808573,120122.111367,1,98,8,8,2.62,3,3,0,0,0,1,50,NA,NA,3,NA
+65123,7,2,2,13,NA,4,4,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19222.393687,19147.000359,1,92,9,7,1.74,7,4,2,1,0,1,45,1,4,2,NA
+65124,7,2,2,28,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,30253.427014,31766.413225,2,90,6,6,1.62,3,3,1,0,0,2,28,1,5,1,4
+65125,7,2,2,40,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,2,3,2,2,2,2,2,2,2,2,2,2,2,25713.328161,27565.019075,2,103,5,5,0.65,6,6,1,0,1,2,61,2,1,2,NA
+65126,7,2,2,63,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10598.2543,1,99,15,15,5,2,2,0,0,2,2,63,2,5,1,5
+65127,7,1,2,17,NA,2,2,NA,NA,NA,2,NA,2,1,3,12,NA,NA,NA,2,2,2,1,2,2,NA,NA,NA,NA,26657.121865,0,1,97,15,15,5,4,4,0,2,0,1,51,1,5,1,NA
+65128,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26503.609729,2,101,1,1,0.09,2,1,0,0,0,1,23,1,5,5,NA
+65129,7,2,2,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,23759.65375,25262.180522,3,92,4,4,1.22,2,2,0,0,0,1,51,1,2,1,3
+65130,7,1,2,38,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,95214.22557,0,1,97,15,15,5,5,5,2,0,1,1,43,1,5,1,5
+65131,7,2,1,24,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39031.957066,40864.300461,1,101,5,5,1.24,3,3,0,0,1,2,61,1,4,1,3
+65132,7,2,2,33,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,6,2,2,2,2,2,2,2,2,2,2,2,38737.690941,37692.46666,1,97,4,4,0.72,5,5,2,1,0,2,33,2,1,6,NA
+65133,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,99275.150567,104459.98751,2,95,3,3,0.93,2,2,0,0,0,1,45,1,4,1,5
+65134,7,2,2,2,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6689.921427,7247.312894,1,96,15,15,4.81,5,5,1,1,0,2,35,1,5,1,5
+65135,7,2,1,55,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16499.662173,16440.055989,3,91,15,15,5,2,2,0,0,0,2,46,1,5,1,5
+65136,7,2,1,7,NA,4,4,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10311.165779,10291.779514,2,98,3,3,0.54,3,3,1,1,0,2,29,1,2,1,NA
+65137,7,2,2,8,NA,3,3,2,8,103,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,48532.852397,48387.04619,1,98,14,14,3.9,4,4,0,3,0,2,31,1,4,1,NA
+65138,7,2,2,0,11,5,7,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4867.693222,4920.842696,2,97,2,2,0.43,4,4,3,0,0,2,25,1,4,5,NA
+65139,7,2,1,61,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,14488.953694,16384.525762,3,92,4,1,0.11,7,1,0,3,1,1,61,1,3,3,NA
+65140,7,2,2,11,NA,5,6,2,11,136,NA,NA,2,2,4,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6198.268014,6649.703491,3,90,14,14,3.47,4,4,1,1,0,2,38,2,5,1,5
+65141,7,2,1,21,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,39084.166385,39616.913732,1,97,4,4,0.65,4,4,0,1,0,2,45,2,2,3,NA
+65142,7,2,1,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,18076.30431,2,95,1,1,0.03,3,3,1,0,0,1,23,1,3,6,NA
+65143,7,2,1,62,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,10555.964373,11135.911352,2,92,6,5,2.2,3,1,0,0,2,1,80,NA,NA,2,NA
+65144,7,2,1,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,87072.438071,86848.766907,1,102,8,8,1.91,5,5,1,2,0,2,38,1,5,1,4
+65145,7,2,2,57,NA,1,1,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,28713.036659,29755.292291,1,97,10,10,3.59,3,3,0,0,0,2,57,1,3,1,NA
+65146,7,2,1,14,NA,2,2,2,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15991.147079,16786.340864,2,90,6,6,1.34,4,4,1,2,0,2,36,2,3,77,NA
+65147,7,1,2,36,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,97803.500399,0,1,95,14,14,3.8,4,4,1,1,0,1,36,1,4,1,5
+65148,7,2,2,12,NA,1,1,1,12,145,NA,NA,2,2,3,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18515.058419,19067.51542,2,96,5,5,0.89,4,4,1,1,0,2,36,2,4,6,NA
+65149,7,2,1,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,19508.889464,19600.270833,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+65150,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,25964.813959,25700.75257,2,101,7,7,2.52,2,2,0,0,0,2,53,1,4,1,2
+65151,7,2,2,26,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,44119.608456,48674.237788,2,98,7,7,2.16,3,3,1,0,0,2,26,1,3,1,3
+65152,7,2,1,13,NA,4,4,2,13,162,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16890.963304,17218.04077,1,91,10,10,2.56,5,5,0,3,0,1,51,2,5,1,4
+65153,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,40859.270352,44024.169267,2,101,14,14,5,1,1,0,0,1,1,80,1,5,5,NA
+65154,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,4,1,0.4,4,1,0,0,0,1,22,1,5,5,NA
+65155,7,2,2,8,NA,5,6,1,8,100,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6273.782555,6554.366499,2,94,15,15,5,5,5,0,2,1,1,47,2,5,1,5
+65156,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,31053.116279,31557.850777,2,97,5,5,1.84,2,1,0,0,0,2,45,1,2,4,NA
+65157,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,51644.110977,57877.507418,1,102,6,6,2.48,1,1,0,0,1,2,80,1,4,3,NA
+65158,7,2,2,4,NA,5,6,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,6683.092466,6676.033575,3,91,13,13,NA,3,3,1,0,0,2,41,2,1,1,3
+65159,7,2,2,0,4,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,11331.751026,1,95,1,1,0.21,4,4,1,0,1,2,75,1,1,2,NA
+65160,7,2,2,17,NA,4,4,2,17,211,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12209.74498,12781.910285,2,90,5,5,1.08,3,3,1,1,0,2,23,1,2,5,NA
+65161,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,18344.917534,19717.054857,2,101,3,3,0.9,2,2,0,0,1,1,57,1,2,5,NA
+65162,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,83819.702285,89179.486894,1,99,9,9,5,1,1,0,0,0,1,32,1,3,5,NA
+65163,7,2,1,70,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,69210.206708,72962.448849,1,98,12,12,NA,2,2,0,0,2,1,70,1,2,1,3
+65164,7,2,1,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,20953.00978,22152.088593,2,97,2,2,0.3,4,4,0,2,0,1,42,1,2,6,NA
+65165,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10561.526228,11375.629073,2,95,7,7,2.31,2,2,0,0,2,2,63,1,3,1,3
+65166,7,2,1,17,NA,4,4,2,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12462.601191,12584.643654,2,97,4,4,0.81,4,4,1,1,0,2,51,1,3,4,NA
+65167,7,2,2,17,NA,4,4,2,17,208,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9608.885901,9632.551122,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+65168,7,2,1,16,NA,2,2,1,17,204,NA,NA,2,2,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15626.107676,15463.700675,2,93,9,9,1.94,6,6,0,3,0,2,37,NA,NA,3,NA
+65169,7,2,2,59,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12649.084278,13261.522577,3,90,1,1,0,2,2,0,0,1,1,61,2,3,1,3
+65170,7,2,1,4,NA,5,6,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7580.437211,8294.186048,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+65171,7,2,2,13,NA,3,3,2,13,157,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,99340.784743,108314.70799,2,91,15,15,5,4,4,0,2,0,1,53,1,5,1,4
+65172,7,2,2,34,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,89807.047643,91549.325071,3,91,7,7,1.74,4,4,0,2,0,1,45,2,2,1,4
+65173,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,NA,NA,NA,1,2,2,1,12969.776823,13038.360257,2,90,3,3,0.65,5,3,1,2,0,1,44,2,5,1,5
+65174,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,85610.546667,92292.669073,2,91,14,14,4.19,3,3,0,1,0,2,31,1,4,1,3
+65175,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,16049.689005,2,100,5,5,1.63,3,2,0,1,0,2,50,1,2,5,NA
+65176,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19130.246369,18737.157388,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+65177,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,120604.496044,126250.391527,2,92,10,10,3.51,3,3,0,0,0,1,24,1,4,5,NA
+65178,7,2,2,0,10,3,3,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7569.605114,7776.819156,2,102,6,6,1.01,5,5,2,0,0,2,18,1,3,NA,NA
+65179,7,2,1,48,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22433.902709,22559.455212,1,100,15,15,5,2,2,0,0,0,1,48,2,5,1,5
+65180,7,2,2,14,NA,4,4,2,14,172,NA,NA,2,2,3,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10694.834447,10900.381844,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+65181,7,1,1,3,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58618.419318,0,2,94,14,14,2.87,5,5,2,1,0,1,37,1,3,1,4
+65182,7,2,1,45,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,2,6,NA,2,2,2,1,2,2,2,2,1,2,34128.967046,33628.177065,2,93,3,3,0.43,4,4,0,0,0,1,45,2,2,6,NA
+65183,7,2,1,62,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,8623.181934,10090.425838,2,103,4,4,1.34,3,1,0,0,1,1,62,2,3,1,NA
+65184,7,2,2,50,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11446.604914,11655.034159,2,92,77,77,NA,4,4,0,0,0,1,27,2,2,5,NA
+65185,7,2,1,6,NA,5,7,2,6,80,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9502.15317,9616.069144,2,100,9,9,2.46,4,4,0,2,0,2,36,2,4,1,3
+65186,7,2,2,0,2,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4418.651245,4685.491283,2,95,7,7,1.41,5,5,2,0,0,2,53,1,3,3,NA
+65187,7,2,1,11,NA,4,4,1,11,141,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9418.975084,9571.535533,2,93,5,5,0.92,4,4,1,1,0,1,27,2,2,5,NA
+65188,7,2,2,59,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14068.752668,14242.311751,1,93,15,15,5,5,5,1,0,1,1,61,2,4,1,4
+65189,7,2,1,63,NA,2,2,1,NA,NA,2,NA,77,NA,NA,NA,3,77,NA,1,2,2,1,2,2,1,2,2,1,9250.428657,10460.65091,2,92,99,99,NA,1,1,0,0,1,1,63,77,3,77,NA
+65190,7,2,2,62,NA,3,3,2,NA,NA,2,NA,2,1,7,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,44636.780791,48652.73082,2,91,6,6,1.78,3,3,0,0,1,2,62,2,3,3,NA
+65191,7,2,2,30,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,13124.243834,13151.042447,1,93,15,6,2.3,6,1,0,0,0,1,34,2,5,5,NA
+65192,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16677.13986,16814.569762,1,97,15,15,5,4,4,0,0,0,1,51,2,5,1,5
+65193,7,2,2,3,NA,4,4,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16749.124902,17181.448234,2,101,4,4,1.16,2,2,1,0,0,2,28,1,4,5,NA
+65194,7,2,1,31,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13936.822202,14121.672299,1,93,15,5,1.84,6,1,0,0,0,1,34,2,5,5,NA
+65195,7,2,1,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13232.135,13189.930654,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+65196,7,2,2,1,22,3,3,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,41956.615141,45173.707461,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+65197,7,2,1,6,NA,5,7,1,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15279.821652,15716.743409,1,102,6,6,1.34,4,4,2,1,0,2,27,1,4,3,NA
+65198,7,2,1,32,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,34624.133414,2,94,7,7,1.34,5,5,2,1,0,1,32,2,1,1,NA
+65199,7,2,2,51,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17852.668137,18282.490034,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+65200,7,2,1,10,NA,5,6,2,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7832.194045,8303.616222,1,96,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+65201,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,56666.803206,58405.284869,1,101,15,15,5,3,3,0,0,2,1,79,1,4,1,4
+65202,7,2,1,45,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,52941.648658,52164.811575,2,102,5,5,1.08,3,3,0,0,0,1,55,2,1,5,NA
+65203,7,2,2,7,NA,3,3,2,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,70999.269152,75403.63644,1,93,15,15,5,4,4,0,2,0,1,51,1,3,1,5
+65204,7,2,1,15,NA,4,4,2,15,188,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12195.546462,12314.973739,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+65205,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,171848.340208,2,91,7,7,1.61,4,4,0,0,3,1,65,1,3,6,NA
+65206,7,2,2,6,NA,5,6,2,6,75,NA,NA,2,1,2,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10029.642884,10710.156898,2,91,14,14,3.69,4,4,1,1,0,1,53,1,4,1,5
+65207,7,2,2,11,NA,2,2,2,11,134,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,15583.587534,16935.930722,1,93,14,14,3.06,5,5,0,2,0,1,46,2,1,1,4
+65208,7,2,2,5,NA,3,3,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,85197.465687,87881.529962,1,92,15,15,5,4,4,2,0,0,2,46,1,5,1,5
+65209,7,2,2,71,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,28297.225791,31371.640749,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+65210,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,19333.302971,19445.46438,1,99,3,3,0.88,2,2,0,1,0,2,48,1,1,3,NA
+65211,7,2,2,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6293.31819,6860.614572,1,96,7,7,1,7,7,2,1,1,2,53,1,4,1,3
+65212,7,2,1,14,NA,5,6,1,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7291.654281,7756.692286,3,91,9,9,3.24,3,3,0,1,1,1,64,2,2,1,5
+65213,7,2,2,64,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15369.196003,16553.883322,1,100,15,15,5,2,2,0,0,2,2,64,1,3,1,3
+65214,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,74517.751389,77393.175383,2,94,10,10,2.91,4,4,0,2,0,2,38,1,4,1,4
+65215,7,2,1,1,14,3,3,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37534.976354,40372.682594,1,99,14,14,4.03,4,4,1,1,0,1,40,2,4,1,5
+65216,7,1,2,1,22,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9955.153132,0,2,94,4,4,0.72,4,4,1,1,0,1,30,2,1,1,3
+65217,7,2,2,22,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,60461.427433,62051.800421,2,92,2,2,0.4,3,3,0,0,0,1,50,2,4,1,4
+65218,7,2,1,74,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,9748.579573,13410.055152,3,90,13,3,0.79,3,2,0,0,2,1,74,2,1,1,NA
+65219,7,2,1,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6936.347746,7414.543244,1,98,10,10,2.59,5,5,2,1,0,1,45,1,5,1,5
+65220,7,2,1,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,1,2,NA,NA,NA,NA,9850.66662,10431.854777,2,98,3,3,1.1,1,1,0,0,1,1,80,1,1,2,NA
+65221,7,2,2,2,NA,3,3,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,45222.58045,46647.274385,1,98,15,15,4.56,4,4,2,0,0,2,33,1,4,1,4
+65222,7,2,2,7,NA,1,1,2,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12958.860039,13685.571339,1,99,15,15,5,3,3,0,1,0,2,34,1,4,1,4
+65223,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,99,99,NA,2,1,0,0,0,1,21,1,4,5,NA
+65224,7,1,1,58,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,16851.334496,0,2,95,3,3,0.95,2,2,0,0,1,2,65,1,2,3,NA
+65225,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16411.593279,16969.580129,2,103,15,15,5,2,2,0,0,0,2,39,2,5,5,NA
+65226,7,2,2,3,NA,3,3,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,71938.505792,79397.866058,3,91,8,8,1.95,4,4,2,0,0,2,30,1,5,1,4
+65227,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,19831.16853,19892.1428,1,103,3,3,0.37,5,5,1,2,0,2,30,1,4,5,NA
+65228,7,2,2,44,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,16237.004863,17542.96783,1,90,15,15,5,5,5,0,2,0,1,47,2,5,1,5
+65229,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,109181.566304,112638.448741,2,91,14,14,3.93,3,3,0,0,2,1,70,NA,NA,1,NA
+65230,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,25128.435397,27949.568291,2,94,5,5,1.3,3,3,0,1,0,1,43,1,3,6,NA
+65231,7,2,1,31,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,34997.800447,36560.756135,2,96,6,6,1.25,4,4,1,1,0,1,31,2,3,1,3
+65232,7,2,2,0,4,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8775.375504,8590.751882,2,102,8,8,2.24,4,4,1,1,0,1,40,1,3,1,3
+65233,7,2,1,30,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,47932.213152,49254.617856,1,92,14,14,3.9,4,4,2,0,0,2,29,1,4,1,4
+65234,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,1,2,1,2,2,NA,NA,NA,NA,26773.686592,29743.160504,2,98,99,99,NA,2,2,0,0,2,2,80,1,3,1,1
+65235,7,2,2,5,NA,2,2,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12541.254112,13845.88109,2,100,14,14,3.36,4,4,1,1,0,1,45,2,5,1,2
+65236,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,12980.451176,13849.150613,1,96,7,7,3.58,1,1,0,0,1,2,73,1,2,3,NA
+65237,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,53574.557692,55073.604958,1,90,4,1,0.27,2,1,0,0,0,2,25,1,5,6,NA
+65238,7,2,1,19,NA,1,1,1,19,228,2,NA,2,2,4,15,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,29234.272259,29869.33924,3,92,4,4,0.46,7,7,1,2,0,2,31,2,2,1,1
+65239,7,2,2,60,NA,4,4,2,NA,NA,2,NA,2,2,8,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,15732.436891,16737.04767,1,97,3,3,0.9,1,1,0,0,1,2,60,2,4,2,NA
+65240,7,2,1,73,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,1,1,2,1,1,2,1,NA,11066.512629,11628.392246,1,99,9,9,4.08,2,2,0,0,2,1,73,2,5,1,5
+65241,7,2,2,42,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,15147.212851,15611.778104,1,96,6,6,1.34,3,3,1,0,0,2,42,2,4,6,NA
+65242,7,2,1,5,NA,1,1,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,17865.135763,18076.339981,3,92,6,6,0.96,5,5,2,1,0,2,26,2,1,1,1
+65243,7,2,2,35,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,49102.007191,48391.207475,3,92,10,10,3.77,3,3,0,1,0,2,52,1,4,6,NA
+65244,7,2,2,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18490.479848,19120.175119,2,100,4,4,0.91,3,3,0,0,0,2,49,1,2,1,2
+65245,7,1,2,2,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9008.172545,0,1,97,15,15,5,3,3,1,0,0,1,28,1,3,6,NA
+65246,7,2,1,3,NA,3,3,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,95119.048777,111454.284969,1,100,6,6,1.78,3,3,1,1,0,2,35,1,5,4,NA
+65247,7,2,2,8,NA,4,4,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10221.991799,10600.655937,1,98,6,6,1.36,3,3,0,2,0,1,35,1,5,3,NA
+65248,7,2,2,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,14,14,5,2,2,0,0,2,1,68,1,4,1,4
+65249,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,53401.089129,55537.639072,2,101,99,2,0.73,3,1,0,0,0,2,22,1,4,5,NA
+65250,7,2,1,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,29413.309667,30038.439043,1,94,3,3,0.9,1,1,0,0,0,1,41,1,3,5,NA
+65251,7,2,2,10,NA,4,4,2,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9565.802332,9920.158544,1,96,14,14,2.58,6,6,2,2,0,1,40,2,4,1,4
+65252,7,2,2,6,NA,4,4,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9089.153301,9279.567325,1,93,1,1,0,2,2,0,1,0,2,38,1,4,3,NA
+65253,7,2,2,3,NA,5,6,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4919.743181,5055.947351,1,97,7,7,1.48,5,5,1,2,0,1,40,2,5,1,4
+65254,7,2,2,2,NA,4,4,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5899.406975,6225.79867,2,103,7,7,1.55,5,5,2,2,0,2,31,1,4,3,NA
+65255,7,2,1,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,24022.668025,24108.898567,1,99,10,10,5,1,1,0,0,0,1,52,1,5,3,NA
+65256,7,2,2,36,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85610.546667,92292.669073,2,91,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+65257,7,2,1,13,NA,3,3,2,13,163,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30100.326038,35344.542616,1,101,6,6,1.52,3,3,0,1,1,1,62,1,2,1,3
+65258,7,2,2,16,NA,1,1,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21818.047789,22469.060223,2,102,7,7,2.16,3,3,0,2,0,2,41,1,5,3,NA
+65259,7,2,2,5,NA,3,3,2,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,23388.579221,24855.504015,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+65260,7,2,2,58,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13354.133365,13424.749254,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+65261,7,2,2,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,30932.175051,34795.188535,2,101,7,7,3.21,1,1,0,0,0,2,41,1,5,5,NA
+65262,7,2,1,35,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,4,5,NA,1,2,2,1,2,1,1,2,2,1,20463.181443,21839.784829,1,93,7,7,2.38,2,2,0,0,0,2,46,2,3,5,NA
+65263,7,2,2,13,NA,2,2,1,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20460.442471,21235.303768,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+65264,7,2,1,40,NA,5,7,2,NA,NA,2,NA,2,2,3,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,26449.318243,30106.409363,1,90,77,77,NA,2,2,0,0,0,1,40,2,2,1,5
+65265,7,2,1,56,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,6,NA,2,2,2,1,2,2,NA,NA,NA,NA,24595.31055,24234.412281,1,97,3,3,0.5,5,5,0,2,0,1,56,2,2,6,NA
+65266,7,2,2,55,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,24870.513993,25871.320138,3,92,7,7,2.45,2,2,0,0,1,2,55,1,1,1,1
+65267,7,2,2,37,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,14553.261149,15099.267285,2,92,15,15,5,3,3,1,0,0,2,37,1,5,1,5
+65268,7,2,2,20,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,2,1,2,2,1,2,2,1,2,2,3,16929.836231,17652.770039,2,101,12,8,4.82,2,1,0,0,0,2,21,NA,NA,5,NA
+65269,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22233.683089,22579.222386,1,102,5,1,0.21,5,4,1,1,0,2,24,1,4,5,NA
+65270,7,2,1,24,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,34629.549705,1,94,5,5,1.47,2,2,0,0,0,1,24,1,4,1,4
+65271,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,36449.806171,39411.375474,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+65272,7,2,2,1,19,4,4,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7618.827213,8513.998744,2,97,2,2,0.34,2,2,1,0,0,2,20,1,3,5,NA
+65273,7,2,1,5,NA,1,1,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14098.200143,14544.470341,1,102,4,4,0.61,5,5,2,2,0,2,27,2,2,5,NA
+65274,7,1,1,25,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,15435.176007,0,2,99,4,4,0.78,4,4,0,2,0,2,45,1,3,5,NA
+65275,7,1,1,80,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,20611.860643,0,1,99,7,7,3.9,1,1,0,0,1,1,80,1,4,2,NA
+65276,7,2,2,11,NA,4,4,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11593.230877,12022.691286,2,96,8,8,1.72,5,5,0,3,0,1,39,1,5,1,4
+65277,7,2,2,2,NA,3,3,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,44626.8774,45213.2896,1,92,14,14,4.59,3,3,1,0,0,1,31,1,4,1,5
+65278,7,2,2,13,NA,1,1,2,13,157,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,18368.872199,20151.001259,2,94,6,6,1.34,4,4,0,2,0,1,37,2,4,1,2
+65279,7,2,2,6,NA,5,7,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19880.837381,19932.405693,1,95,3,3,0.45,4,4,2,1,0,2,26,1,3,4,NA
+65280,7,2,1,62,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,3,3,NA,2,2,2,1,2,2,NA,NA,NA,NA,14488.953694,16384.525762,3,92,1,1,0,6,6,0,3,1,1,62,2,3,3,NA
+65281,7,2,2,8,NA,3,3,2,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28912.545431,29050.529973,1,98,3,3,0.61,4,4,0,2,0,2,32,1,3,6,NA
+65282,7,2,1,2,NA,1,1,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12387.68972,12183.594345,1,94,6,6,1.78,3,3,1,0,0,2,31,2,5,1,1
+65283,7,2,2,31,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,29102.738194,28317.485179,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+65284,7,2,2,6,NA,5,6,2,6,77,NA,NA,2,2,1,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5795.227223,6217.308223,1,91,14,14,4.32,3,3,0,2,0,2,37,2,5,1,NA
+65285,7,2,1,44,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,2,1,2,2,1,2,2,2,34153.424332,35777.45113,1,100,8,8,2.17,4,4,1,1,0,2,40,2,2,1,2
+65286,7,2,1,30,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,1,1,2,2,1,20228.224986,20656.862173,2,102,15,15,3.82,5,5,0,1,2,1,60,2,2,1,1
+65287,7,2,1,1,15,5,7,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26781.430467,29328.277521,1,102,6,6,1.34,4,4,2,1,0,2,27,1,4,3,NA
+65288,7,2,2,41,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,37512.060155,37707.682357,1,92,14,14,5,2,2,0,1,0,2,41,1,5,3,NA
+65289,7,2,2,73,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,17474.843163,17732.021804,1,98,6,6,1.57,3,3,0,0,2,1,66,2,2,1,4
+65290,7,2,2,0,8,4,4,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4439.36229,4707.453058,2,100,5,5,0.88,5,5,2,1,0,2,30,1,4,6,NA
+65291,7,2,1,29,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,35669.2076,37942.468331,2,94,9,9,2.51,4,4,2,0,0,2,34,2,3,1,3
+65292,7,2,1,3,NA,5,6,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7769.861689,8391.242047,1,101,8,8,1.81,5,5,2,0,1,2,37,2,4,1,2
+65293,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,10038.743429,10812.549167,1,96,6,6,1.83,2,2,0,0,1,2,64,1,3,2,NA
+65294,7,2,2,15,NA,4,4,2,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11133.192774,11654.909373,3,91,2,2,0.25,4,4,0,2,0,2,35,1,3,5,NA
+65295,7,2,2,17,NA,1,1,1,17,212,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18674.868029,19048.995585,2,92,8,8,1.42,7,7,0,4,0,2,37,1,1,6,NA
+65296,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,38666.703155,42427.769217,2,94,9,9,3.97,2,2,0,0,2,1,80,1,5,1,5
+65297,7,2,1,72,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,1,1,2,2,NA,12375.169389,14759.280437,2,92,4,4,1.14,2,2,0,0,2,1,72,2,3,1,3
+65298,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140932.152825,151077.472396,1,97,14,14,3.36,4,4,0,2,0,2,49,1,5,1,5
+65299,7,2,1,29,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,94547.245282,99078.839738,2,94,15,10,5,2,1,0,0,0,1,29,1,4,6,NA
+65300,7,2,1,1,16,3,3,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50563.666669,55372.144903,1,95,9,9,2.68,4,4,2,0,0,2,27,1,4,1,4
+65301,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10604.379638,1,99,8,8,4.13,1,1,0,0,1,2,62,1,2,3,NA
+65302,7,2,1,11,NA,3,3,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21767.839187,23123.409471,1,101,5,5,1.28,3,3,0,2,0,2,44,1,5,3,NA
+65303,7,2,2,19,NA,5,6,2,19,239,2,NA,2,2,2,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7588.544207,7881.983727,3,91,4,4,0.69,5,5,0,2,0,1,45,2,4,1,1
+65304,7,2,1,43,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,30596.964094,35273.693641,2,102,14,14,5,1,1,0,0,0,1,43,1,4,3,NA
+65305,7,2,2,16,NA,5,7,2,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6386.337576,6452.948976,1,93,15,15,4.59,4,4,0,2,0,2,45,1,5,1,5
+65306,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,30802.731848,31491.045952,1,100,6,6,1.18,5,5,1,2,0,2,30,1,3,1,4
+65307,7,2,1,3,NA,3,3,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27608.061058,31148.453885,1,94,6,6,1.18,5,5,1,2,0,1,30,1,3,1,3
+65308,7,2,2,54,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,2,2,1,2,25483.560748,25945.701567,1,94,12,3,1.07,4,1,0,0,1,2,37,NA,NA,6,NA
+65309,7,1,1,77,NA,1,1,NA,NA,NA,1,1,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,14200.083364,0,2,98,3,3,1.1,1,1,0,0,1,1,77,1,2,2,NA
+65310,7,2,1,40,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,28721.124083,1,101,6,6,1.21,4,4,0,2,0,2,33,1,2,6,NA
+65311,7,2,2,7,NA,1,1,1,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15510.382876,16516.802792,1,100,8,8,2.17,4,4,1,1,0,2,40,2,2,1,2
+65312,7,2,1,2,NA,5,7,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26781.430467,29328.277521,1,102,6,6,1.34,4,4,2,1,0,2,27,1,4,3,NA
+65313,7,2,1,3,NA,4,4,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6946.177172,7233.19413,2,90,4,4,0.57,5,5,1,2,0,2,33,2,2,77,NA
+65314,7,2,2,16,NA,4,4,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12208.965913,12707.508076,2,100,6,6,0.99,5,5,0,3,0,2,40,1,3,1,3
+65315,7,2,1,7,NA,2,2,2,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12049.155217,12263.670489,2,90,14,14,3.58,4,4,1,1,0,1,37,1,3,1,4
+65316,7,2,1,54,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,26127.59163,27234.881102,1,100,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+65317,7,2,2,26,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,15267.012422,14516.051827,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+65318,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,51433.469947,56330.524019,1,98,5,5,1.97,1,1,0,0,1,2,80,1,4,2,NA
+65319,7,2,1,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11849.199361,13959.374637,1,100,4,4,1.16,2,2,0,0,2,1,73,1,3,1,3
+65320,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,109181.566304,114466.270601,2,91,15,3,0.98,7,1,0,0,1,1,49,NA,NA,5,NA
+65321,7,2,1,7,NA,2,2,1,7,90,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14820.807433,15173.085782,2,102,6,6,1.12,4,4,1,1,0,1,38,2,2,1,3
+65322,7,2,2,23,NA,2,2,2,NA,NA,2,NA,2,1,4,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,30253.427014,31766.413225,2,90,6,6,1.35,3,3,1,0,0,1,31,1,3,1,4
+65323,7,2,1,0,11,2,2,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6266.985765,6267.101277,1,93,15,15,5,3,3,1,0,0,2,32,2,5,1,5
+65324,7,2,2,0,3,1,1,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7011.218293,6988.118839,2,97,8,8,2.01,4,4,2,0,0,2,24,1,4,1,1
+65325,7,2,2,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,72443.10981,74665.593008,2,102,7,7,2.72,2,2,0,0,1,2,73,1,4,2,NA
+65326,7,2,2,44,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,1,1,2,1,2,2,1,2,2,NA,NA,NA,NA,12544.244874,13448.569073,1,102,5,5,0.92,5,5,1,2,0,2,44,2,1,1,2
+65327,7,2,1,38,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,22007.065507,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+65328,7,2,1,63,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,4,5,NA,2,2,2,2,2,2,2,2,2,2,10903.483462,11078.143343,1,93,2,2,0.43,2,2,0,0,2,2,80,2,1,2,NA
+65329,7,2,1,21,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,43287.255521,44006.168052,1,90,6,6,1.57,3,3,0,0,0,1,50,2,2,1,2
+65330,7,2,2,0,9,3,3,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8814.114917,8994.318513,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+65331,7,2,2,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16741.034883,16282.872546,2,95,2,2,0.81,1,1,0,0,0,2,54,1,4,5,NA
+65332,7,2,1,7,NA,1,1,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14880.007592,15211.324369,3,92,2,2,0.47,3,3,1,1,0,2,33,1,4,5,NA
+65333,7,2,2,2,NA,5,6,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,7029.864692,7462.832472,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+65334,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,2,3,NA,2,2,2,2,2,2,2,2,2,2,13676.984152,18290.186018,2,91,2,2,0.75,1,1,0,0,1,2,60,2,2,3,NA
+65335,7,2,1,55,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14288.99518,14503.461604,1,102,9,9,2.39,5,5,0,1,1,1,55,2,5,1,5
+65336,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,124037.944076,128061.498173,1,100,15,15,5,4,4,0,1,0,1,50,1,4,1,4
+65337,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,28478.57859,28745.66908,1,98,3,3,1.25,1,1,0,0,1,1,63,1,4,5,NA
+65338,7,2,2,22,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16239.242782,17454.103497,2,101,99,1,0.22,3,1,0,0,0,2,22,1,4,5,NA
+65339,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,4,6,NA,2,2,2,1,2,2,2,2,2,2,9911.173561,10104.337319,2,92,15,10,5,2,1,0,0,1,2,54,2,5,3,NA
+65340,7,2,2,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,71351.478679,72494.161504,2,97,15,15,4.97,5,5,1,0,0,1,48,1,4,1,3
+65341,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,116464.874823,118336.754883,2,94,14,14,5,3,3,0,0,0,1,42,1,5,1,5
+65342,7,2,2,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6306.491784,7047.470907,2,90,5,5,1.08,3,3,1,1,0,2,23,1,2,5,NA
+65343,7,2,2,4,NA,3,3,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,40290.055403,42817.035801,1,99,15,15,5,5,5,3,0,0,2,34,1,5,1,5
+65344,7,2,1,60,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,9745.303498,10208.81789,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+65345,7,2,2,1,13,4,4,2,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7734.994032,8078.467013,2,95,1,1,0,4,4,2,1,0,2,27,1,4,5,NA
+65346,7,2,2,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18490.479848,18110.536811,2,100,14,14,3.06,5,5,1,0,0,1,50,1,5,1,5
+65347,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,17983.231494,18657.922524,3,91,10,10,3.67,3,3,1,0,0,2,30,2,4,1,4
+65348,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,99120.116925,101722.54619,2,98,15,15,5,3,3,0,1,0,1,56,1,5,1,5
+65349,7,2,1,19,NA,2,2,1,19,231,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,17555.907575,18699.509115,2,93,3,3,0.52,5,5,0,2,0,1,41,2,4,1,4
+65350,7,2,1,6,NA,4,4,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11321.293503,11517.688596,1,98,6,6,1.36,3,3,0,2,0,1,35,1,5,3,NA
+65351,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,21171.267805,24852.610883,1,97,5,5,1.79,1,1,0,0,0,1,36,1,3,3,NA
+65352,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11240.290931,11199.684593,2,92,12,77,NA,7,2,0,0,2,1,53,2,3,1,3
+65353,7,2,2,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,18490.479848,18110.536811,2,100,15,15,4.97,5,5,0,2,1,2,42,1,5,1,5
+65354,7,2,1,13,NA,3,3,2,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,127790.772987,128099.355719,1,97,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+65355,7,2,1,37,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19799.565045,19517.338151,1,90,5,5,0.74,5,5,0,2,0,2,18,1,4,NA,NA
+65356,7,2,1,8,NA,3,3,1,8,103,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24713.905595,26080.214484,1,98,5,5,0.74,5,5,0,3,0,1,35,1,2,6,NA
+65357,7,2,2,7,NA,4,4,1,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8504.389189,8822.229593,2,93,5,5,0.74,5,5,1,2,0,2,28,1,2,6,NA
+65358,7,2,2,45,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,115926.402585,116948.004168,1,91,10,10,3.8,3,3,0,0,0,1,45,NA,NA,1,4
+65359,7,2,1,14,NA,2,2,2,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25622.388537,25620.13505,1,97,15,15,4.52,6,6,0,4,0,2,41,1,5,1,5
+65360,7,2,2,17,NA,3,3,2,17,205,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,111142.989658,114400.666347,1,95,NA,NA,NA,5,5,0,2,0,2,37,1,3,1,NA
+65361,7,2,1,11,NA,4,4,1,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8447.917731,9106.89532,1,102,15,15,5,3,3,0,1,0,1,41,1,5,1,5
+65362,7,2,2,11,NA,2,2,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,15148.721588,15796.67129,2,91,2,2,0.22,4,4,0,3,0,2,45,2,5,4,NA
+65363,7,2,1,4,NA,1,1,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,16797.750213,2,98,14,14,3.91,4,4,1,1,0,1,36,2,3,1,5
+65364,7,2,2,8,NA,1,1,2,8,98,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,9872.244853,10294.506103,2,90,6,6,1.15,5,5,0,2,0,2,47,2,1,1,5
+65365,7,2,2,17,NA,4,4,1,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11078.202838,10933.134393,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+65366,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,1,2,2,2,2,2,2,6449.12882,6755.867756,2,93,8,8,2.97,2,2,0,0,1,1,60,2,3,1,3
+65367,7,2,1,10,NA,2,2,2,10,127,NA,NA,2,2,2,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,16281.509392,17100.217351,2,91,4,4,0.43,7,7,0,1,1,1,41,2,1,4,NA
+65368,7,2,1,23,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27681.749998,29590.732776,2,98,3,3,0.54,3,3,1,0,0,1,23,1,3,1,2
+65369,7,2,1,3,NA,3,3,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,71085.311148,83293.12191,2,94,14,14,3.36,4,4,2,0,0,1,31,1,3,1,5
+65370,7,2,2,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,112992.533921,121615.449738,1,94,6,6,1.57,3,3,0,1,0,2,28,1,4,1,4
+65371,7,2,1,10,NA,4,4,1,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12552.995979,12770.757918,2,96,4,4,0.65,5,5,0,3,0,1,30,1,4,1,2
+65372,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,4,3,1,2,2,1,2,2,1,2,2,1,27367.658704,37250.756368,1,91,6,6,2.04,2,2,1,0,0,2,31,1,5,4,NA
+65373,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,154825.466557,160454.355913,1,91,15,15,5,4,4,0,1,0,1,45,1,5,1,5
+65374,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,32415.779658,36870.322177,2,91,4,4,1.02,2,2,0,0,2,1,80,1,4,1,2
+65375,7,2,2,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,150482.742079,154420.406935,2,98,15,15,5,2,2,0,0,0,2,52,1,3,1,4
+65376,7,1,1,2,24,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,0,2,98,5,5,1.63,2,2,1,0,0,2,31,1,1,3,NA
+65377,7,2,1,19,NA,4,4,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13276.485807,13770.209619,1,96,4,4,1.29,2,2,0,0,0,2,48,1,4,3,NA
+65378,7,2,2,3,NA,5,6,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5022.237557,1,91,15,15,5,4,4,1,1,0,1,43,2,5,1,5
+65379,7,2,2,12,NA,3,3,1,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39616.634313,40360.331753,2,101,3,3,0.6,3,3,0,2,0,1,39,1,4,4,NA
+65380,7,2,2,17,NA,5,6,1,17,205,2,NA,2,2,1,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9884.956101,9995.989819,1,94,99,1,0.32,6,1,0,3,0,2,45,NA,NA,6,NA
+65381,7,2,2,9,NA,4,4,1,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9453.111053,10094.338553,1,100,6,6,1.39,4,4,0,3,0,2,29,1,4,5,NA
+65382,7,1,1,4,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6946.177172,0,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+65383,7,2,1,15,NA,1,1,2,16,193,NA,NA,2,2,2,9,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,20398.562455,20202.582308,2,94,77,77,NA,3,3,0,1,0,2,42,2,2,6,NA
+65384,7,2,2,72,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,20267.042371,20898.557176,2,102,7,7,1.68,5,5,0,0,3,1,70,2,4,1,4
+65385,7,2,2,12,NA,5,6,2,12,149,NA,NA,2,1,99,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11975.458482,12414.350614,1,97,14,14,2.29,7,7,1,2,2,1,40,2,1,1,1
+65386,7,2,1,6,NA,1,1,1,6,83,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,19735.224235,20204.314213,2,102,8,8,2.24,4,4,1,1,0,1,35,2,3,1,1
+65387,7,1,2,34,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,3,3,1,2,2,1,2,2,NA,NA,NA,NA,71034.153987,0,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+65388,7,2,1,33,NA,4,4,2,NA,NA,2,NA,2,1,3,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,23454.081246,24179.640416,1,91,99,99,NA,3,3,1,0,0,1,33,2,5,5,NA
+65389,7,2,1,32,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,13593.59406,16516.502952,2,90,77,77,NA,4,3,1,0,0,2,30,2,2,5,NA
+65390,7,2,2,2,NA,5,6,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6796.893869,7396.552454,2,93,15,15,5,3,3,1,0,0,1,41,2,5,1,5
+65391,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,11389.795225,13662.741278,2,92,8,8,1.91,5,5,0,2,1,2,47,2,1,1,3
+65392,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,17635.020067,18381.689068,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+65393,7,2,1,78,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,15844.527135,16445.651972,1,101,4,4,1.22,2,2,0,0,2,2,77,1,4,1,3
+65394,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,28175.708676,28476.609384,1,91,4,4,1.7,1,1,0,0,1,2,71,1,4,2,NA
+65395,7,2,1,77,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,14537.797299,15440.503889,3,91,2,2,0.83,1,1,0,0,1,1,77,1,5,5,NA
+65396,7,2,1,54,NA,5,6,1,NA,NA,2,NA,2,7,4,NA,5,3,NA,1,2,1,1,2,1,1,2,1,NA,12983.367248,13800.313434,2,101,77,77,NA,4,1,0,0,0,1,39,NA,NA,5,NA
+65397,7,1,1,65,NA,5,6,NA,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12579.986433,0,1,95,3,3,0.43,4,4,0,1,2,1,65,2,5,1,3
+65398,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,9,NA,4,2,NA,1,2,2,1,2,2,1,1,2,NA,18693.365067,19341.691824,1,92,12,12,NA,1,1,0,0,1,2,80,2,4,2,NA
+65399,7,2,1,59,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18753.573091,18839.996658,2,99,15,15,4.34,4,4,0,0,0,1,59,2,4,1,5
+65400,7,2,1,16,NA,1,1,1,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22768.423624,22886.980387,3,92,7,7,1.41,5,5,1,2,0,1,40,1,3,1,4
+65401,7,2,1,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,31291.360507,35666.138476,2,91,4,4,1.4,1,1,0,0,0,1,55,1,4,5,NA
+65402,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,99283.360764,105284.943086,1,94,15,15,5,4,3,0,0,1,1,33,1,2,5,NA
+65403,7,2,1,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,31291.360507,35666.138476,2,91,3,3,1.16,1,1,0,0,0,1,52,1,1,4,NA
+65404,7,2,2,27,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,5,5,2,1,2,2,1,2,2,1,2,2,3,16929.836231,18593.694637,2,101,8,6,2.85,2,1,0,0,0,2,27,2,5,5,NA
+65405,7,2,1,17,NA,1,1,1,17,211,2,NA,1,1,NA,12,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,18635.323223,19040.145288,1,103,6,6,1.57,3,3,0,1,0,2,50,2,3,4,NA
+65406,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,3,1,0.22,4,1,0,0,0,1,20,1,4,5,NA
+65407,7,2,1,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18790.641284,22550.069226,2,100,6,6,0.99,5,5,0,3,0,2,40,1,3,1,3
+65408,7,2,2,19,NA,2,2,2,19,233,2,NA,2,2,4,10,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,12680.621719,14355.413798,2,90,3,3,0.46,5,5,0,2,2,1,75,2,1,1,2
+65409,7,2,1,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7701.757497,8492.447039,2,90,2,2,0.38,4,4,1,2,0,2,32,1,4,5,NA
+65410,7,2,2,71,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,2,NA,1,2,1,1,2,2,1,2,1,NA,13838.805939,14318.765975,1,103,15,15,3.7,5,5,0,2,1,1,55,1,5,1,5
+65411,7,2,2,16,NA,5,6,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8391.252153,8615.174176,1,92,8,8,2.3,4,4,0,1,0,2,41,NA,NA,1,3
+65412,7,2,1,41,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18898.73153,19305.900504,1,100,8,8,2.62,3,3,0,1,0,1,41,2,5,1,5
+65413,7,2,2,1,17,2,2,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8214.128831,8731.974448,1,99,8,8,1.91,5,5,2,0,1,2,38,2,4,1,4
+65414,7,2,2,6,NA,5,6,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8290.163782,8692.019106,1,92,15,15,5,4,4,0,2,0,1,55,1,5,1,5
+65415,7,2,2,1,22,3,3,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14349.194039,14956.025029,1,103,6,6,1.11,5,5,1,1,1,1,29,1,3,1,3
+65416,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18088.213601,18580.673153,1,98,12,5,2.15,2,1,0,0,0,1,23,2,5,5,NA
+65417,7,2,2,75,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,NA,29145.675285,31446.34406,3,92,10,10,2.82,4,4,0,1,1,1,36,1,3,1,5
+65418,7,2,2,7,NA,4,4,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7282.523598,7776.514874,2,99,6,6,1.11,5,5,0,4,0,2,34,1,4,5,NA
+65419,7,2,1,19,NA,3,3,1,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,33505.877646,33253.482687,2,101,5,3,1.1,2,1,0,0,0,1,19,1,4,NA,NA
+65420,7,2,2,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,22306.465066,21209.244743,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+65421,7,2,2,70,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,27103.875166,29389.100868,1,95,4,4,0.65,4,4,0,0,1,2,70,1,1,2,NA
+65422,7,2,1,7,NA,4,4,1,7,94,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9535.948,10279.797103,2,93,4,4,0.56,5,5,2,1,0,1,27,1,2,6,NA
+65423,7,2,1,50,NA,1,1,2,NA,NA,2,NA,2,2,7,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,21506.056514,21429.79108,2,95,6,6,2.24,1,1,0,0,0,1,50,2,1,4,NA
+65424,7,2,1,2,NA,5,6,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8821.037035,9893.758146,1,102,15,15,4.59,4,4,1,1,0,1,35,1,5,1,5
+65425,7,2,2,2,NA,5,7,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21940.920626,22868.807624,1,95,6,6,1.15,5,5,2,1,0,1,29,1,4,6,NA
+65426,7,2,2,4,NA,3,3,2,4,58,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,40064.649797,41758.993816,1,93,10,10,2.48,5,5,2,1,0,1,40,2,5,1,5
+65427,7,2,2,64,NA,3,3,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,36910.075958,37362.049791,2,97,2,2,0.38,4,4,0,2,2,2,64,2,1,1,NA
+65428,7,2,2,8,NA,5,6,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9007.62445,9504.796896,2,102,15,15,3.92,5,5,1,2,0,1,34,2,5,1,5
+65429,7,2,1,17,NA,2,2,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16033.31661,16386.200415,2,90,5,5,0.8,5,5,0,3,0,2,40,2,1,5,NA
+65430,7,2,2,34,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,1,2,2,2,2,2,2,2,2,2,2,2,35353.005268,36099.35979,2,94,9,9,2.51,4,4,2,0,0,2,34,2,3,1,3
+65431,7,2,1,73,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,NA,8497.912951,8664.367847,2,100,4,4,1.47,1,1,0,0,1,1,73,1,4,4,NA
+65432,7,1,2,70,NA,4,4,NA,NA,NA,2,NA,2,2,6,NA,4,2,NA,1,2,1,1,2,1,NA,NA,NA,NA,12344.929687,0,1,93,2,2,0.78,1,1,0,0,1,2,70,2,4,2,NA
+65433,7,2,2,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,34073.955911,34748.968506,2,92,14,14,4.03,4,4,1,1,1,2,30,1,5,4,NA
+65434,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,3,15494.204758,16562.601332,2,95,15,15,5,3,3,0,1,0,2,34,2,5,1,NA
+65435,7,2,2,4,NA,4,4,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8340.858072,8802.325948,2,103,7,7,1.55,5,5,2,2,0,2,31,1,4,3,NA
+65436,7,2,1,28,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15196.92397,16091.373745,2,101,7,3,1.1,3,1,0,0,0,1,21,2,4,5,NA
+65437,7,1,1,73,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,14077.240235,0,1,97,5,5,1.84,1,1,0,0,1,1,73,1,4,2,NA
+65438,7,2,2,3,NA,4,4,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8833.042831,9629.276723,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+65439,7,2,2,15,NA,2,2,2,15,180,NA,NA,2,1,3,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16189.692833,16766.382859,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+65440,7,2,1,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7380.991723,2,100,8,8,2.67,3,3,0,0,1,1,61,1,3,1,4
+65441,7,2,2,10,NA,3,3,1,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22308.590534,22137.463018,1,95,7,2,0.35,5,4,1,2,0,1,26,1,4,6,NA
+65442,7,2,1,2,NA,4,4,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5942.817425,6188.375425,2,97,13,13,NA,6,6,2,2,0,2,24,1,2,6,NA
+65443,7,2,2,37,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,1,2,1,2,1,1,2,1,1,2,1,3,17978.142628,18308.053591,1,91,14,14,4.32,3,3,0,2,0,2,37,2,5,1,NA
+65444,7,2,1,4,NA,1,1,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12870.245769,13452.024092,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+65445,7,2,1,1,22,2,2,1,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8742.6471,9019.39043,2,100,99,99,NA,6,6,1,1,2,1,37,2,3,1,3
+65446,7,2,1,65,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,8168.47072,10568.054811,2,90,6,6,1.73,2,2,0,0,2,2,69,1,2,1,2
+65447,7,2,1,44,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18841.134943,20530.600583,2,95,14,14,5,2,2,0,0,0,1,44,1,3,1,4
+65448,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10764.448363,1,99,13,13,NA,3,3,0,0,2,1,67,1,2,1,2
+65449,7,2,2,2,NA,4,4,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7630.869112,8210.900286,2,95,6,6,0.97,6,6,2,2,0,1,37,1,3,1,4
+65450,7,2,2,49,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,34954.173075,37266.073044,2,98,6,6,1.21,4,4,1,0,0,2,49,2,2,6,NA
+65451,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,87322.7576,94138.522454,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+65452,7,2,2,16,NA,4,4,2,16,203,NA,NA,2,2,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10694.834447,10900.381844,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+65453,7,2,2,67,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,12823.794396,13219.948623,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+65454,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,11329.574648,11407.369464,1,99,10,10,4.76,2,2,1,0,0,2,36,2,5,3,NA
+65455,7,2,2,14,NA,3,3,2,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,101168.631125,108085.531312,1,99,6,6,1.52,4,4,0,2,0,2,43,1,3,5,NA
+65456,7,2,2,12,NA,1,1,2,13,156,NA,NA,2,2,3,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,16594.391299,17940.930507,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+65457,7,2,1,25,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13367.406737,13987.969236,1,92,15,15,4.44,5,5,0,0,1,1,65,NA,NA,1,5
+65458,7,2,1,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,152458.779794,153888.636628,2,94,15,15,5,2,2,0,0,2,1,63,1,5,1,5
+65459,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,7537.914984,2,95,1,1,0.26,2,2,1,0,0,2,20,1,2,77,NA
+65460,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,2,2,NA,2,2,2,1,2,2,1,2,2,1,7608.893707,8232.851326,2,100,5,5,2.11,1,1,0,0,1,2,64,2,2,2,NA
+65461,7,2,1,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,23775.734331,24432.76962,2,96,5,5,1.08,3,3,0,0,0,1,50,1,3,1,4
+65462,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,149120.841984,149985.960772,1,98,9,9,5,1,1,0,0,0,2,49,1,5,3,NA
+65463,7,2,2,6,NA,4,4,2,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7814.742747,8587.431439,2,95,7,7,1.55,5,5,0,3,0,1,30,1,4,1,4
+65464,7,2,2,29,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,18601.70604,17686.717051,1,90,14,14,2.96,5,5,1,2,0,1,31,1,5,1,4
+65465,7,2,2,7,NA,5,6,1,7,89,NA,NA,2,1,3,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11141.059365,11755.98595,1,98,10,10,1.89,7,7,3,2,0,1,50,1,5,1,5
+65466,7,2,1,7,NA,4,4,1,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12221.817539,12310.445459,2,100,10,10,2.75,5,5,1,1,1,1,27,1,3,1,5
+65467,7,2,1,0,6,5,6,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9148.090461,9086.623516,1,92,6,6,1.47,3,3,1,0,0,2,32,2,3,1,3
+65468,7,2,1,7,NA,4,4,2,7,85,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8655.162127,8808.105516,2,95,4,4,0.65,4,4,1,2,0,2,27,1,3,5,NA
+65469,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,180467.973037,1,95,77,77,NA,3,3,0,0,2,1,80,NA,NA,1,NA
+65470,7,2,1,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14321.363328,13945.949794,2,96,77,77,NA,7,7,3,2,0,2,33,2,2,6,NA
+65471,7,1,2,80,NA,5,6,NA,NA,NA,2,NA,2,2,7,NA,5,2,NA,1,1,1,1,2,1,NA,NA,NA,NA,10831.995402,0,3,90,4,4,0.92,3,3,0,0,1,2,56,2,2,1,2
+65472,7,2,1,3,NA,2,2,1,3,38,NA,NA,2,2,2,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13504.027725,14468.59111,2,93,7,7,1.52,4,4,1,1,0,1,44,2,4,1,NA
+65473,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,9015.10987,2,95,2,2,0.81,1,1,0,0,1,2,60,1,2,2,NA
+65474,7,2,2,63,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10644.199654,11088.440242,2,98,7,7,2.72,2,2,0,0,2,1,63,1,4,1,4
+65475,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,15520.204559,18173.553405,1,96,15,15,5,6,6,1,1,1,2,44,1,3,1,3
+65476,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,38151.592121,38761.146216,1,101,1,1,0.21,3,3,0,2,0,2,32,1,4,5,NA
+65477,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,3,3,NA,1,2,2,1,2,2,1,2,2,2,26609.95229,27262.621072,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+65478,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,9680.216878,10112.428208,1,96,10,10,5,1,1,0,0,1,2,61,1,4,2,NA
+65479,7,2,1,51,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21283.272729,21364.140181,1,100,15,15,3.7,5,5,0,3,0,1,51,1,5,1,5
+65480,7,2,1,10,NA,4,4,2,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8988.052978,9133.633722,2,95,8,8,1.85,5,5,1,2,0,1,55,1,2,1,3
+65481,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,12782.405209,13767.697696,1,100,2,2,0.54,1,1,0,0,1,2,62,1,3,2,NA
+65482,7,1,1,78,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,81678.388584,0,1,92,9,9,3.97,2,2,0,0,2,2,67,1,3,1,5
+65483,7,1,2,52,NA,3,3,NA,NA,NA,2,NA,2,1,8,NA,5,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,119276.206757,0,1,103,15,15,5,2,2,0,1,0,2,52,2,5,3,NA
+65484,7,2,1,1,14,5,7,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6328.720733,6765.026101,1,91,15,15,4.01,5,5,1,2,0,2,34,1,5,1,5
+65485,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,30407.489838,32596.441724,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+65486,7,2,2,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,10346.035773,10598.2543,1,99,7,6,1.84,3,2,0,0,2,1,70,1,2,1,4
+65487,7,2,2,1,18,5,6,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8165.183174,1,97,14,14,4.45,3,3,1,0,0,2,35,2,5,1,5
+65488,7,2,1,38,NA,3,3,2,NA,NA,2,NA,2,1,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15425.274863,16733.462521,1,90,7,7,2.1,3,3,1,0,0,2,40,2,5,1,4
+65489,7,2,1,18,NA,4,4,2,18,216,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10558.572325,11042.564542,2,99,4,4,0.41,7,7,0,2,0,2,36,1,3,5,NA
+65490,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,1,1,2,2,1,2,2,NA,NA,NA,NA,22225.098465,28674.927734,2,97,1,1,0.27,2,2,1,0,0,2,20,1,2,5,NA
+65491,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,24654.107413,23891.769846,1,100,5,5,0.85,5,5,0,2,0,2,54,1,2,2,NA
+65492,7,2,2,65,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,11342.714991,11512.583626,1,99,6,6,1.07,6,6,2,1,2,1,44,2,5,4,NA
+65493,7,2,1,9,NA,5,7,2,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9720.482616,10484.39038,2,91,15,15,3.7,5,5,1,2,0,1,50,NA,NA,1,5
+65494,7,2,1,3,NA,3,3,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,88790.489317,95503.197037,2,91,15,15,5,3,3,1,0,0,1,40,1,4,6,NA
+65495,7,2,2,5,NA,1,1,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11859.546176,12772.78418,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+65496,7,2,1,5,NA,1,1,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16613.593777,17139.487274,1,92,14,14,3.15,5,5,1,2,0,1,34,1,4,1,4
+65497,7,2,1,42,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,116464.874823,129540.216925,2,94,3,3,0.54,4,4,0,1,0,2,48,1,3,1,3
+65498,7,2,2,39,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,4,2,1,2,2,1,2,2,1,2,2,1,16369.013285,16419.342576,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+65499,7,2,1,19,NA,3,3,2,19,231,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19277.921327,19518.221274,2,97,7,7,1.89,3,3,0,0,0,1,50,1,2,1,2
+65500,7,2,1,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,17787.524589,17931.023329,2,100,8,8,4.82,1,1,0,0,0,1,46,1,3,3,NA
+65501,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,53634.754806,54437.113731,1,98,5,1,0,3,1,0,0,0,1,32,1,5,5,NA
+65502,7,2,1,1,16,4,4,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9438.902193,9727.203655,2,102,4,4,0.72,4,4,2,0,0,1,48,1,3,1,3
+65503,7,2,2,44,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,1,2,2,2,2,2,2,2,2,2,2,2,33767.584626,34595.8104,2,102,6,4,1.02,6,2,0,4,0,2,43,2,1,5,NA
+65504,7,2,1,7,NA,1,1,1,7,86,NA,NA,2,2,2,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13870.762641,14200.45921,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+65505,7,2,1,55,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17839.845235,18404.710221,3,92,15,15,5,3,3,0,1,0,1,55,2,5,1,4
+65506,7,2,1,50,NA,1,1,1,NA,NA,2,NA,99,NA,NA,NA,2,5,NA,2,2,2,1,2,2,1,2,2,2,30839.213846,30386.695922,1,95,4,4,0.68,5,5,0,1,0,2,38,2,3,4,NA
+65507,7,2,2,13,NA,2,2,2,13,158,NA,NA,2,1,3,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15442.648697,16179.397194,3,90,4,4,0.63,5,5,0,3,0,1,45,2,4,1,4
+65508,7,2,1,31,NA,2,2,2,NA,NA,1,1,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,32770.630289,33645.258285,2,95,14,14,4.45,3,3,1,0,0,2,29,1,5,1,5
+65509,7,2,2,10,NA,4,4,2,10,130,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9235.947079,9749.775473,1,96,7,7,1.39,5,5,0,2,2,1,69,2,2,1,2
+65510,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,44238.530111,44780.242841,1,95,3,3,0.87,2,2,0,0,2,2,65,1,2,1,3
+65511,7,2,1,7,NA,1,1,1,7,86,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14820.807433,14905.891142,2,102,7,7,1.53,5,5,0,3,0,1,43,2,2,1,4
+65512,7,2,2,39,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,3,1,1,2,2,2,2,2,2,2,2,2,2,35353.005268,38359.582487,2,94,4,4,0.81,3,3,0,1,0,1,49,2,3,1,3
+65513,7,2,1,32,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,47932.213152,49921.729758,1,92,10,10,3.04,4,4,1,1,0,1,32,1,3,1,2
+65514,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,43425.032766,48500.713195,1,96,15,15,5,2,2,0,0,2,2,80,1,5,1,5
+65515,7,2,2,14,NA,4,4,2,14,173,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17185.945299,17434.901758,2,91,2,2,0.26,4,4,0,1,0,1,20,1,3,5,NA
+65516,7,2,1,0,9,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,7527.69611,7949.840792,1,92,99,99,NA,5,5,1,0,0,1,46,2,3,1,3
+65517,7,2,2,25,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,4,1,2,1,2,2,1,2,2,2,2,2,2,39426.061521,41875.604468,2,94,8,8,2.7,3,3,1,0,0,1,27,1,3,1,4
+65518,7,2,1,45,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16052.801806,16001.600848,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+65519,7,2,1,13,NA,5,6,2,13,165,NA,NA,2,2,1,6,NA,NA,NA,1,1,1,1,2,1,1,2,1,1,10346.302718,11892.421636,2,91,99,99,NA,7,4,0,4,0,1,36,2,9,1,2
+65520,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,44274.069981,46208.106563,2,101,7,7,2.58,2,2,0,0,0,2,36,1,5,1,3
+65521,7,2,1,17,NA,3,3,1,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26457.098276,26381.28385,2,98,3,3,0.38,5,5,0,4,0,2,39,1,4,5,NA
+65522,7,2,2,71,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,1,2,NA,11550.700389,12323.715655,2,93,2,2,0.54,1,1,0,0,1,2,71,1,2,2,NA
+65523,7,2,2,11,NA,4,4,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8362.256577,9003.967662,2,100,8,8,1.8,5,5,0,3,0,2,43,1,3,1,3
+65524,7,2,1,63,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,1,1,2,1,2,2,1,2,10596.142548,10856.489537,1,94,7,7,1.48,5,5,0,0,1,2,52,2,1,1,1
+65525,7,2,2,69,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,11224.764345,11466.33634,1,101,5,5,1.84,1,1,0,0,1,2,69,1,5,2,NA
+65526,7,2,1,4,NA,4,4,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12032.669734,13267.985975,2,96,6,6,1.48,4,4,1,1,0,2,25,1,4,5,NA
+65527,7,2,1,11,NA,4,4,2,11,142,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7730.47951,9212.541007,1,99,7,7,1.53,5,5,0,3,0,1,39,1,3,1,3
+65528,7,2,1,0,0,4,4,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,5360.999096,5929.781839,2,90,8,6,1.46,4,3,1,1,0,2,21,1,5,6,NA
+65529,7,2,2,1,18,5,7,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22159.470641,22857.583467,1,95,3,3,0.45,4,4,2,1,0,2,26,1,3,4,NA
+65530,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,136880.768184,145646.064425,1,94,5,5,1.04,4,4,1,1,0,1,18,1,2,NA,NA
+65531,7,2,1,11,NA,1,1,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11881.117946,11763.87702,1,102,15,15,4.47,4,4,0,2,0,2,30,1,4,1,4
+65532,7,2,1,18,NA,4,4,2,18,224,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11107.552941,11616.709768,2,99,4,4,1,3,3,0,1,0,2,38,1,3,5,NA
+65533,7,2,1,4,NA,4,4,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11302.271106,11769.28244,3,92,6,6,0.93,5,5,2,1,0,2,37,1,5,1,3
+65534,7,2,2,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,27381.645976,28428.806364,2,95,15,1,0.18,5,1,0,0,0,1,47,1,5,1,3
+65535,7,2,2,0,10,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8383.207272,8612.693254,1,91,6,6,1.03,5,5,3,0,0,2,37,1,5,6,NA
+65536,7,2,2,61,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,15876.871857,17178.834476,2,102,4,4,0.67,4,4,0,0,2,2,20,1,1,NA,NA
+65537,7,2,1,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,31707.924183,33437.457992,1,97,3,3,0.73,3,3,0,0,0,2,50,1,4,1,3
+65538,7,2,2,9,NA,3,3,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,47428.012402,47285.525443,2,98,10,10,4.42,2,2,0,1,0,2,41,1,4,3,NA
+65539,7,2,1,27,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13367.406737,13987.969236,1,92,15,15,4.44,5,5,0,0,1,1,65,NA,NA,1,5
+65540,7,2,1,9,NA,3,3,1,9,115,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74331.764009,78594.005469,2,91,9,9,2.6,4,4,0,2,0,1,53,1,2,1,5
+65541,7,2,2,9,NA,4,4,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,12444.673315,2,97,7,7,1.72,5,5,1,2,0,1,32,1,4,1,4
+65542,7,2,1,2,NA,3,3,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37959.146468,40828.920669,1,94,15,15,5,3,3,1,0,0,1,35,1,5,1,5
+65543,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,108373.053289,112114.199141,3,92,15,15,5,3,3,0,0,0,1,56,NA,NA,1,4
+65544,7,2,2,19,NA,4,4,2,19,234,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19113.842115,19208.656277,1,97,1,1,0.09,4,4,0,1,0,2,44,2,2,1,3
+65545,7,2,1,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,116462.885745,124577.022727,1,93,15,15,5,2,2,0,0,0,1,28,1,5,6,NA
+65546,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,25052.373156,24265.266353,1,92,6,6,1.31,3,3,0,0,1,2,80,1,3,4,NA
+65547,7,2,2,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,24832.705563,25238.431199,2,100,2,2,0.73,1,1,0,0,1,2,63,1,4,2,NA
+65548,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,99120.116925,103753.216485,2,98,14,14,4.16,3,3,0,0,0,1,49,1,5,1,4
+65549,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,12291.154515,13189.875012,1,92,5,5,1.84,1,1,0,0,1,1,80,1,3,2,NA
+65550,7,2,1,31,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,34997.800447,36838.864542,2,96,14,14,3.36,4,4,1,1,0,2,28,1,2,6,NA
+65551,7,2,2,8,NA,3,3,2,9,108,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60197.256541,63931.531988,2,94,14,14,2.63,6,6,1,3,0,1,39,1,4,1,4
+65552,7,2,2,13,NA,3,3,1,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,37065.780724,38522.476457,1,98,5,5,0.74,5,5,0,3,0,1,35,1,2,6,NA
+65553,7,2,2,42,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,2,4,2,2,2,2,2,2,2,NA,NA,NA,NA,32229.130119,41089.02125,2,90,13,2,0.46,2,1,0,0,1,2,80,NA,NA,77,NA
+65554,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,34973.413194,1,95,5,5,1.19,3,3,1,1,0,1,47,1,2,3,NA
+65555,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,53634.754806,56026.668478,1,98,5,2,0.45,3,1,0,0,0,1,32,1,5,5,NA
+65556,7,2,1,5,NA,4,4,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10234.881417,10547.495238,2,97,2,2,0.33,4,4,2,1,0,2,34,1,2,5,NA
+65557,7,2,2,30,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,5,5,2,2,2,2,2,2,2,NA,NA,NA,NA,27973.581456,27218.795462,2,99,99,99,NA,5,3,0,1,0,1,40,2,1,6,NA
+65558,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,42489.109693,44345.173355,1,100,14,9,5,2,1,0,0,0,1,37,1,2,6,NA
+65559,7,2,1,8,NA,2,2,2,8,101,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,9807.589376,10786.008845,2,90,3,3,0.38,5,5,0,4,0,2,33,2,2,5,NA
+65560,7,2,1,4,NA,2,2,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11852.624029,11657.344215,2,99,99,99,NA,3,3,1,0,0,1,35,2,2,1,2
+65561,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,1,2,NA,60148.377616,67408.231176,1,92,8,8,2.17,4,4,0,1,2,2,80,1,3,2,NA
+65562,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,9548.871949,9975.218868,2,95,6,6,1.36,3,3,0,1,1,2,62,1,4,5,NA
+65563,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,39260.973419,41280.172031,1,91,4,2,0.55,3,1,0,0,0,2,22,1,5,6,NA
+65564,7,2,2,36,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,19738.81952,19597.773568,3,90,3,3,0.37,5,5,2,2,0,2,36,2,4,4,NA
+65565,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,113500.095101,1,91,15,6,2.69,3,1,0,0,0,1,27,1,3,6,NA
+65566,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85444.349063,88741.398785,3,91,8,8,1.95,4,4,2,0,0,2,30,1,5,1,4
+65567,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,32375.321924,34954.784356,1,95,1,1,0.21,4,4,1,0,1,2,75,1,1,2,NA
+65568,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,2,2,2,2,1,2,2,NA,NA,NA,NA,45655.090694,44423.220436,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+65569,7,2,2,16,NA,2,2,2,16,197,NA,NA,1,1,NA,9,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,20678.81116,21093.085203,1,93,5,5,0.84,5,5,1,2,0,2,52,2,1,3,NA
+65570,7,2,1,16,NA,3,3,2,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,35593.481513,36275.823276,1,91,5,5,1.36,2,2,0,1,0,2,49,1,5,3,NA
+65571,7,2,2,74,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,2,NA,1,2,1,1,2,1,1,2,1,NA,13446.397433,14295.858749,1,93,3,3,0.65,3,3,0,0,3,2,74,2,1,2,NA
+65572,7,2,2,79,NA,5,6,1,NA,NA,2,NA,2,1,9,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,15984.880532,16163.124545,1,98,9,9,4.92,1,1,0,0,1,2,79,2,5,5,NA
+65573,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,22165.906062,22485.884772,1,96,7,7,3.58,1,1,0,0,0,1,44,1,4,3,NA
+65574,7,2,1,0,3,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8637.740206,8903.804193,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+65575,7,2,1,28,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,1,2,2,2,35669.2076,38557.984895,2,94,14,4,1.74,5,1,0,0,0,1,24,2,4,5,NA
+65576,7,2,2,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,63207.045171,66153.523399,2,102,10,6,2.3,2,1,0,0,0,2,27,1,4,6,NA
+65577,7,2,1,54,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15044.515884,15040.577827,3,90,15,15,5,3,3,0,0,0,2,55,2,4,1,4
+65578,7,2,2,1,18,1,1,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11901.705423,12408.80856,2,102,8,8,1.91,5,5,1,2,0,1,36,2,1,1,4
+65579,7,2,2,3,NA,5,6,2,3,47,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5380.83825,5375.15484,2,95,15,15,5,3,3,1,0,0,1,53,1,5,1,2
+65580,7,2,2,19,NA,2,2,1,19,233,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17338.975742,19021.185309,2,93,8,8,2.49,3,3,0,0,0,1,52,2,2,1,4
+65581,7,2,1,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,28542.421068,31101.791288,2,101,7,7,2.58,2,2,0,0,0,2,36,1,5,1,3
+65582,7,2,2,7,NA,5,6,2,7,95,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5481.728631,5853.665419,1,91,15,15,5,6,6,0,2,2,1,50,2,5,1,5
+65583,7,2,2,16,NA,4,4,2,16,202,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10694.834447,10849.760352,1,90,9,9,1.65,7,7,0,4,0,1,36,1,4,1,4
+65584,7,2,1,11,NA,4,4,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.422451,8987.382625,1,99,6,6,2.18,2,2,0,1,0,2,31,1,4,5,NA
+65585,7,2,2,49,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,3,21956.01693,22136.94803,2,91,8,8,2.34,4,4,0,2,0,1,56,2,5,1,5
+65586,7,2,2,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,32501.623429,34582.098932,2,97,3,3,0.46,5,5,0,3,0,1,40,1,2,1,3
+65587,7,2,1,78,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,62466.132707,66344.89016,1,94,6,6,1.91,2,2,0,0,2,1,78,1,4,1,3
+65588,7,2,2,15,NA,3,3,2,15,184,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74165.041171,75130.242541,2,94,10,10,3.51,3,3,0,2,0,2,39,2,4,3,NA
+65589,7,2,1,69,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,2,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,10288.337394,10907.668198,3,91,14,4,1.02,6,2,0,0,2,1,48,2,1,1,1
+65590,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,18795.138328,22008.372948,2,93,9,9,2.07,5,5,0,1,0,1,55,NA,NA,5,NA
+65591,7,2,2,5,NA,2,2,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11429.37307,11583.084593,2,90,15,15,5,4,4,1,1,0,1,49,1,4,1,4
+65592,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,26010.201422,30173.312639,2,95,3,3,1.21,1,1,0,0,1,2,80,1,2,2,NA
+65593,7,2,2,51,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,2,2,2,2,20178.078974,20795.430208,2,100,14,14,3.58,4,4,0,1,0,1,46,2,5,1,5
+65594,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10604.379638,1,99,4,4,1.74,1,1,0,0,1,2,60,1,3,4,NA
+65595,7,1,2,66,NA,2,2,NA,NA,NA,2,NA,2,2,4,NA,4,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,10235.0654,0,2,93,6,6,1.3,4,4,0,0,2,2,36,2,4,1,4
+65596,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11896.060838,12427.20723,1,96,15,15,5,2,2,0,0,2,1,61,1,5,1,5
+65597,7,2,2,19,NA,4,4,1,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,2,2,0.46,1,1,0,0,0,2,19,1,4,NA,NA
+65598,7,1,1,53,NA,3,3,NA,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20980.882737,0,2,93,6,6,1.48,4,4,0,1,0,1,53,2,2,1,3
+65599,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12127.194143,12676.977658,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+65600,7,2,2,19,NA,1,1,2,19,232,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20937.26435,21936.153747,1,90,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+65601,7,2,1,15,NA,4,4,2,15,190,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13366.904548,14267.035861,1,93,7,7,1.79,4,4,0,2,0,1,53,2,4,1,4
+65602,7,2,1,19,NA,4,4,2,20,NA,2,NA,2,2,3,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12548.434193,12514.39179,2,99,3,1,0.31,4,1,0,0,0,1,19,2,4,NA,NA
+65603,7,2,2,63,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,1,1,2,1,NA,13656.521422,13861.04165,1,91,4,4,1.33,2,2,0,0,2,1,65,2,4,1,3
+65604,7,2,1,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9668.679611,9969.996889,1,102,15,15,3.82,5,5,1,1,0,1,29,1,4,1,4
+65605,7,2,1,68,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10274.998921,11619.264446,1,91,8,8,3.57,2,2,0,0,2,1,68,1,3,1,2
+65606,7,2,2,48,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,18215.139307,19150.911381,1,91,77,77,NA,4,4,0,2,0,1,50,2,5,1,5
+65607,7,2,2,5,NA,4,4,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8873.727797,9364.677323,2,99,5,5,0.78,5,5,2,2,0,2,30,1,3,5,NA
+65608,7,2,1,5,NA,2,2,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15745.774489,16244.197679,2,91,4,4,0.67,5,4,2,0,2,2,66,2,1,1,NA
+65609,7,2,2,1,19,3,3,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25257.21318,27876.153487,1,101,7,7,1.82,4,4,2,0,0,2,27,1,2,1,3
+65610,7,2,1,9,NA,4,4,1,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8714.559478,8865.734494,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+65611,7,2,2,19,NA,2,2,2,19,232,2,NA,2,2,3,12,NA,NA,NA,2,2,2,2,2,2,1,2,2,2,12680.621719,13709.581084,2,90,99,99,NA,5,5,1,1,0,2,40,2,3,1,1
+65612,7,2,1,5,NA,4,4,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,7991.632447,8596.395387,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+65613,7,2,2,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,122472.877524,125201.399718,2,94,15,15,5,3,3,0,1,1,1,63,1,5,1,3
+65614,7,2,2,10,NA,1,1,1,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15859.146626,16025.878294,1,92,1,1,0,3,1,0,2,0,2,43,1,2,4,NA
+65615,7,1,1,80,NA,3,3,NA,NA,NA,2,NA,2,1,9,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,38666.703155,0,2,94,2,2,0.81,1,1,0,0,1,1,80,2,1,2,NA
+65616,7,2,1,0,3,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9570.577309,9932.368407,1,95,6,6,1,6,6,3,0,0,2,23,1,4,6,NA
+65617,7,2,1,41,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21571.318341,21852.480517,1,102,15,15,5,3,3,0,1,0,1,41,1,5,1,5
+65618,7,2,2,2,NA,2,2,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9955.153132,10990.75621,2,94,9,9,2.51,4,4,2,0,0,2,34,2,3,1,3
+65619,7,2,1,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,50647.682308,53126.874664,2,96,8,8,3.67,2,2,0,0,0,1,58,1,3,3,NA
+65620,7,2,2,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,110849.032294,115406.960266,2,92,14,6,2.75,2,1,0,0,0,2,25,1,5,5,NA
+65621,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,8748.603912,9090.779024,1,96,14,14,5,2,2,0,0,2,2,61,1,3,1,3
+65622,7,2,1,7,NA,4,4,1,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15130.588085,15393.064565,2,102,15,15,3.82,5,5,1,2,0,1,34,1,3,1,4
+65623,7,2,2,25,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18626.118419,17709.928623,1,93,15,15,5,5,5,0,1,0,2,25,1,5,5,NA
+65624,7,2,2,11,NA,4,4,1,11,135,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,12444.673315,2,97,7,7,1.72,5,5,1,2,0,1,32,1,4,1,4
+65625,7,2,1,23,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14690.01297,15450.354874,3,91,2,2,0.73,1,1,0,0,0,1,23,1,5,5,NA
+65626,7,2,1,11,NA,4,4,2,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8017.552697,8398.795399,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+65627,7,2,1,16,NA,4,4,2,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15970.206483,16086.518081,2,97,2,2,0.3,4,4,0,2,0,1,42,1,2,6,NA
+65628,7,2,1,56,NA,1,1,2,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,25381.488007,25009.053797,2,99,6,6,2.33,1,1,0,0,0,1,56,1,4,2,NA
+65629,7,2,2,51,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,2,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,13281.538492,13864.580434,3,90,5,5,1.05,3,3,0,0,0,1,53,2,1,1,2
+65630,7,2,1,4,NA,1,1,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16214.341036,15789.305867,1,98,9,9,2,7,6,3,2,0,2,32,1,4,1,4
+65631,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17014.040717,2,97,2,2,0.49,2,2,0,0,0,1,24,1,4,6,NA
+65632,7,2,2,42,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,34639.996543,35471.414202,1,102,7,7,1.7,4,4,0,0,2,1,44,1,4,4,NA
+65633,7,2,1,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,31699.998102,36617.55692,1,98,1,1,0.04,1,1,0,0,0,1,57,1,4,5,NA
+65634,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,9113.905743,9695.883475,3,90,15,15,5,5,5,0,1,1,2,61,1,5,2,NA
+65635,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,22225.098465,22243.480978,2,97,4,4,0.81,4,4,1,1,0,2,51,1,3,4,NA
+65636,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,19323.414782,18794.578845,2,91,15,15,5,4,4,0,0,1,1,61,NA,NA,1,2
+65637,7,2,1,66,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,6358.062034,6283.458155,1,99,7,7,3.49,1,1,0,0,1,1,66,1,4,3,NA
+65638,7,2,1,12,NA,5,7,2,12,153,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21852.102821,22875.024578,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+65639,7,1,2,14,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,91539.042546,0,1,101,15,15,5,4,4,0,2,0,2,40,1,4,1,3
+65640,7,2,2,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,12751.545122,13835.806616,2,101,1,1,0.08,2,2,0,0,2,2,80,1,1,2,NA
+65641,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,74517.751389,83352.706958,2,94,5,5,0.89,4,4,0,2,0,2,51,1,2,3,NA
+65642,7,2,2,41,NA,3,3,2,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,96255.674553,98150.909956,2,94,14,14,5,3,3,0,0,0,1,42,1,5,1,5
+65643,7,2,2,38,NA,3,3,2,NA,NA,2,NA,2,2,4,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,95214.22557,102645.939709,1,97,15,15,5,5,4,1,1,0,2,38,NA,NA,1,5
+65644,7,2,2,19,NA,4,4,2,19,233,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18905.695776,2,101,1,1,0.23,2,1,0,0,0,2,19,1,3,NA,NA
+65645,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,38666.703155,42427.769217,2,91,7,7,1.61,4,4,0,0,3,1,65,1,3,6,NA
+65646,7,2,1,8,NA,3,3,1,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18596.465852,19407.799286,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+65647,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,52701.331723,59294.433957,1,94,5,5,1.04,4,4,0,2,0,2,29,1,3,1,3
+65648,7,2,1,61,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7101.739553,7379.502561,2,100,4,4,1.74,1,1,0,0,1,1,61,1,3,3,NA
+65649,7,2,2,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,26428.874559,27145.22781,2,99,14,14,4.09,3,3,0,2,0,2,37,1,5,5,NA
+65650,7,2,2,79,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,20101.581228,20798.747999,3,92,77,77,NA,1,1,0,0,1,2,79,1,4,2,NA
+65651,7,2,1,6,NA,5,7,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7923.925927,8546.646915,1,91,15,15,5,5,5,0,3,0,1,40,1,5,1,5
+65652,7,1,2,1,20,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24749.659974,0,1,94,1,1,0.08,7,7,2,4,0,1,31,1,2,1,4
+65653,7,2,2,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,33506.462855,33906.169633,2,96,4,4,0.65,5,5,0,3,0,1,30,1,4,1,2
+65654,7,2,1,10,NA,5,6,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6850.601671,7624.777305,3,91,6,6,1.34,4,4,0,2,0,1,52,2,3,1,1
+65655,7,2,2,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16741.034883,16786.385595,2,95,12,12,NA,3,3,0,0,0,2,29,2,5,5,NA
+65656,7,2,2,2,NA,3,3,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,45341.612978,50043.120503,3,91,14,14,5,3,3,1,0,0,2,30,1,4,1,5
+65657,7,2,1,15,NA,1,1,1,16,192,NA,NA,1,1,NA,9,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,29234.272259,28953.402615,2,98,7,7,2.25,3,3,0,1,0,2,51,2,1,1,1
+65658,7,2,2,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,29448.834066,28780.798614,1,97,9,9,4.92,1,1,0,0,0,2,49,1,5,5,NA
+65659,7,2,2,71,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,12344.929687,13268.288363,1,93,2,2,0.74,1,1,0,0,1,2,71,2,2,2,NA
+65660,7,2,2,73,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,36122.708948,47794.076653,1,97,2,2,0.75,1,1,0,0,1,2,73,2,2,2,NA
+65661,7,2,2,38,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,3,3,2,2,2,2,1,2,2,2,2,2,2,35678.162793,37949.435224,2,99,3,3,0.52,3,3,0,0,1,2,38,2,3,3,NA
+65662,7,2,2,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,33506.462855,35016.985063,2,96,6,6,1.62,3,3,0,2,0,2,31,1,3,5,NA
+65663,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,3,1,2,2,1,2,2,1,2,2,1,52701.331723,53172.517558,1,94,5,5,0.74,5,5,1,1,0,2,24,1,3,1,4
+65664,7,2,2,13,NA,3,3,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71837.483011,75155.181458,1,94,15,15,5,3,3,0,1,0,2,43,1,5,1,5
+65665,7,2,1,19,NA,1,1,1,19,233,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15626.107676,15463.700675,2,93,9,9,1.94,6,6,0,3,0,2,37,NA,NA,3,NA
+65666,7,2,2,3,NA,2,2,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16648.051651,2,98,8,8,1.48,7,7,3,0,0,1,26,1,3,1,3
+65667,7,2,1,8,NA,4,4,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10410.106675,10534.907593,2,96,6,6,1.32,5,5,1,3,0,2,30,1,4,3,NA
+65668,7,2,2,8,NA,5,7,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11541.681354,12101.149927,1,92,15,15,5,4,4,0,2,0,1,41,2,5,1,5
+65669,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,159007.057971,1,95,14,8,4.41,2,1,0,0,0,1,47,1,4,3,NA
+65670,7,2,1,13,NA,1,1,1,14,169,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,17424.208904,17520.505085,1,102,5,5,0.92,5,5,0,3,0,2,39,2,3,1,3
+65671,7,2,2,2,NA,4,4,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7444.013422,7636.156038,1,90,5,5,1.32,2,2,1,0,0,2,27,2,3,5,NA
+65672,7,2,2,22,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,33292.204257,2,101,3,3,0.92,1,1,0,0,0,2,22,1,4,5,NA
+65673,7,2,1,2,NA,3,3,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24594.444896,28818.16319,1,98,1,1,0.16,3,3,1,0,0,1,28,1,2,6,NA
+65674,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,17285.492411,17948.070756,1,102,7,7,1.8,5,4,1,0,2,1,47,1,3,5,NA
+65675,7,2,1,66,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11568.876339,11794.347884,1,102,10,10,4.42,2,2,0,0,2,2,62,1,5,1,4
+65676,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,75000.659691,1,98,9,9,2.6,4,4,1,1,0,2,35,1,2,1,NA
+65677,7,2,2,4,NA,1,1,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,2,NA,2,2,2,2,NA,NA,NA,NA,13193.876261,13756.035699,3,91,7,7,1.23,6,6,2,2,0,1,36,2,1,1,1
+65678,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,12697.263698,13996.985089,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+65679,7,2,1,69,NA,5,7,2,NA,NA,2,NA,2,1,7,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,10298.198778,10826.998489,1,95,2,2,0.72,1,1,0,0,1,1,69,2,4,3,NA
+65680,7,2,2,44,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16162.24986,16822.942152,3,91,15,15,5,3,3,0,1,0,2,44,2,5,1,5
+65681,7,2,2,4,NA,1,1,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16462.187772,17163.602129,3,92,8,8,1.55,6,6,1,3,0,2,38,1,5,1,4
+65682,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,143785.115498,143265.966876,1,99,10,10,5,1,1,0,0,0,2,57,1,5,3,NA
+65683,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,1,1,2,2,1,7879.750437,7851.284287,2,92,10,8,2.01,7,4,1,1,1,2,27,2,3,1,3
+65684,7,2,1,14,NA,1,1,1,14,179,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18242.832494,18343.652859,1,102,2,2,0.52,3,3,0,2,0,2,36,2,3,4,NA
+65685,7,2,2,29,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,51746.15111,54067.303468,1,95,8,8,2.24,4,4,2,0,0,2,29,1,3,1,4
+65686,7,2,2,9,NA,4,4,1,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10085.683644,10404.203071,2,102,6,6,1.22,5,5,0,2,0,2,42,1,4,1,4
+65687,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,23128.736624,23721.941544,2,101,2,2,0.79,1,1,0,0,0,2,55,1,2,3,NA
+65688,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,187291.098551,186266.152024,2,91,15,15,5,3,3,0,0,0,2,54,1,4,1,4
+65689,7,2,1,19,NA,4,4,2,19,234,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13969.457688,14694.403205,1,90,6,6,1.57,3,3,0,1,0,2,36,1,3,5,NA
+65690,7,2,1,16,NA,3,3,2,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,107087.58296,107224.801212,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+65691,7,2,2,16,NA,5,6,1,16,199,NA,NA,2,2,2,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8391.252153,8753.945484,1,92,12,12,NA,7,7,1,2,1,2,45,2,3,1,3
+65692,7,2,1,14,NA,3,3,2,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26824.630008,26472.092796,1,95,6,6,1.3,4,4,0,3,0,2,46,1,4,3,NA
+65693,7,2,1,7,NA,3,3,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,66868.503099,69864.859716,1,98,15,15,5,4,4,1,1,0,1,40,1,4,1,5
+65694,7,2,2,5,NA,2,2,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13366.393396,14756.857003,2,94,9,9,2.51,4,4,2,0,0,2,34,2,3,1,3
+65695,7,1,1,9,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8943.919305,0,1,92,5,5,0.63,7,7,0,4,1,1,60,NA,NA,1,NA
+65696,7,2,2,8,NA,4,4,2,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6996.144083,7533.021082,2,99,5,5,0.78,5,5,2,2,0,2,30,1,3,5,NA
+65697,7,2,1,16,NA,3,3,2,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26749.020961,27839.957403,1,101,6,6,1.21,4,4,0,2,0,2,33,1,2,6,NA
+65698,7,2,1,8,NA,4,4,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11094.855304,11787.231447,1,90,7,7,2.1,3,3,0,1,0,1,35,1,3,6,NA
+65699,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,154000.877343,1,94,6,6,2.24,1,1,0,0,0,1,50,1,3,3,NA
+65700,7,2,2,18,NA,4,4,1,18,220,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14166.687432,14623.20249,1,100,13,13,NA,5,5,2,0,0,2,54,1,4,5,NA
+65701,7,2,2,17,NA,4,4,2,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+65702,7,2,2,62,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,10999.00871,11458.057418,3,92,7,7,1.49,5,5,0,2,1,2,62,1,4,2,NA
+65703,7,2,2,29,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,39426.061521,39893.228552,3,92,8,8,2.01,4,4,1,0,0,2,49,2,5,4,NA
+65704,7,2,2,7,NA,3,3,2,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,82043.921571,82230.595899,1,97,14,14,3.9,4,4,0,2,0,1,47,1,3,1,5
+65705,7,2,2,63,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,15207.312407,15975.329738,1,92,4,4,0.99,2,2,0,0,1,1,26,1,1,5,NA
+65706,7,2,2,11,NA,3,3,2,11,143,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,24097.958076,1,95,6,3,0.45,6,4,1,2,0,1,28,1,2,1,2
+65707,7,2,1,75,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,62703.997496,73870.695024,1,94,10,10,3.04,4,4,0,0,2,1,75,1,3,1,3
+65708,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35334.703093,40990.264786,1,101,3,3,1.25,1,1,0,0,1,2,80,1,2,2,NA
+65709,7,2,2,17,NA,3,3,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,97212.131473,101229.4133,1,91,14,14,3.8,4,4,0,2,0,1,50,NA,NA,1,5
+65710,7,2,2,17,NA,4,4,2,17,210,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11608.900361,11982.991894,1,99,8,8,2.59,3,3,0,2,0,2,46,1,4,2,NA
+65711,7,2,1,46,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11113.602843,11073.454175,3,90,77,77,NA,5,5,0,2,0,1,46,2,3,1,3
+65712,7,2,1,75,NA,2,2,2,NA,NA,1,1,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,14090.083482,14599.284236,2,90,6,6,2.24,2,2,0,0,2,1,75,2,4,1,2
+65713,7,2,1,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,14078.198261,14841.449973,1,91,1,1,0.05,2,1,0,0,2,1,72,1,1,3,NA
+65714,7,2,1,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,64271.597433,68262.462992,1,94,15,15,4.95,4,4,0,0,2,1,72,1,3,1,3
+65715,7,2,1,12,NA,5,7,2,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60818.973858,62775.825513,1,99,15,15,5,4,4,0,2,0,2,44,1,5,1,5
+65716,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,42559.487719,42454.393427,1,98,6,6,1.57,3,3,0,0,0,1,58,1,3,1,3
+65717,7,2,1,10,NA,4,4,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11443.206518,12087.326286,1,100,13,13,NA,3,3,0,1,0,2,52,2,3,1,NA
+65718,7,2,2,11,NA,5,7,2,11,137,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9227.090107,9736.375878,2,100,15,15,4.83,4,4,1,1,0,1,43,2,5,1,5
+65719,7,2,1,68,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,2,1,NA,1,2,2,1,2,2,1,2,2,3,11145.558675,12081.671552,2,96,4,4,0.92,3,3,0,1,1,2,41,2,2,1,2
+65720,7,2,1,5,NA,2,2,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12403.412256,13289.361067,2,90,6,6,1.62,3,3,1,0,0,2,28,1,5,1,4
+65721,7,2,2,6,NA,1,1,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14538.945634,14691.797657,2,92,8,8,1.42,7,7,0,4,0,2,37,1,1,6,NA
+65722,7,2,2,39,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,37237.570046,37476.359157,2,97,4,1,0,2,1,0,0,0,1,41,1,2,6,NA
+65723,7,1,1,73,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15200.416613,0,1,98,4,4,1.19,2,2,0,0,2,2,72,1,3,1,4
+65724,7,1,2,4,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,13366.393396,0,2,94,14,7,2.72,3,2,1,0,0,1,25,2,3,6,NA
+65725,7,2,2,8,NA,3,3,2,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,70999.269152,75403.63644,1,93,15,15,5,4,4,0,2,0,1,51,1,3,1,5
+65726,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15196.92397,16091.373745,2,101,7,3,1.1,3,1,0,0,0,1,21,2,4,5,NA
+65727,7,2,2,3,NA,1,1,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16432.24332,3,92,7,7,1.41,5,5,1,2,0,1,40,1,3,1,4
+65728,7,2,2,0,1,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4581.358327,5045.31579,2,101,4,4,0.85,4,4,1,0,1,2,61,1,4,3,NA
+65729,7,2,1,9,NA,3,3,1,9,113,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23188.935049,25672.571973,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+65730,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16181.169973,16457.814273,2,100,14,14,4.86,3,3,0,0,0,2,52,1,5,1,3
+65731,7,2,1,17,NA,2,2,1,17,208,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18206.126374,18306.74388,2,93,4,4,0.84,3,3,0,1,0,2,46,2,3,1,2
+65732,7,2,1,2,NA,1,1,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10803.555682,10625.559962,2,94,6,6,1.32,4,4,2,0,0,1,29,1,3,1,3
+65733,7,1,2,29,NA,5,6,NA,NA,NA,2,NA,2,2,2,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,12636.996393,0,3,90,12,12,NA,3,3,0,0,0,2,29,2,5,1,5
+65734,7,2,2,14,NA,4,4,2,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13841.239638,13875.328502,1,96,15,15,4.52,6,6,0,4,0,1,46,1,4,1,4
+65735,7,2,2,69,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,2,NA,2,2,2,1,2,2,1,2,2,1,9104.567599,9484.550932,2,93,6,6,2.69,1,1,0,0,1,2,69,2,4,2,NA
+65736,7,2,1,71,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,2,1,2,1,NA,16439.475505,17615.98168,1,100,5,5,1.18,3,3,0,0,2,2,34,2,5,5,NA
+65737,7,2,2,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,33274.645943,34015.957912,2,95,14,14,3.34,4,4,0,0,0,1,43,1,3,1,3
+65738,7,2,1,35,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18025.279948,19114.89054,1,97,4,3,0.93,3,2,0,0,0,1,35,1,3,1,4
+65739,7,2,1,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,58205.324383,61609.052582,2,98,8,8,3.06,2,2,0,0,2,1,71,1,4,1,3
+65740,7,2,1,68,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10098.443227,10212.20306,1,102,9,9,5,1,1,0,0,1,1,68,1,4,3,NA
+65741,7,2,1,26,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,1,2,2,2,38560.502118,39812.43256,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+65742,7,2,1,30,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,3,5,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,35069.154585,2,94,15,15,5,3,3,0,0,0,1,41,2,3,1,NA
+65743,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,25569.682712,26728.878308,1,97,15,15,5,4,4,0,1,0,1,40,1,4,1,4
+65744,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,54095.581484,0,2,94,77,77,NA,2,2,0,0,2,2,80,1,5,1,5
+65745,7,2,1,27,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,69082.166945,70155.790389,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+65746,7,2,1,19,NA,3,3,1,19,234,2,NA,2,1,4,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,103364.662244,103614.262032,1,98,3,1,0.09,4,1,0,0,0,1,20,1,4,5,NA
+65747,7,2,2,6,NA,4,4,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7646.649777,8233.445923,2,99,3,3,0.52,3,3,0,1,0,2,23,1,3,1,3
+65748,7,2,1,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,7907.095293,7814.315452,2,97,5,5,1.3,3,3,0,1,1,1,64,1,4,3,NA
+65749,7,2,2,2,NA,2,2,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9534.013652,9566.569071,2,97,12,6,0.89,7,7,3,0,0,2,26,2,1,6,NA
+65750,7,2,1,14,NA,3,3,1,14,173,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,79147.85144,78921.048453,1,98,15,15,5,5,5,0,3,0,2,44,1,5,1,5
+65751,7,2,1,54,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,3,1,NA,2,2,2,1,2,2,2,2,2,2,22446.308035,22366.708253,2,94,1,1,0.03,2,2,0,0,0,2,48,2,1,1,3
+65752,7,2,1,21,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15564.966804,2,101,6,99,NA,2,1,0,0,0,1,20,2,4,5,NA
+65753,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,27738.890335,33497.780693,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+65754,7,2,2,50,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,5,NA,2,2,2,1,2,2,1,2,2,2,28701.155283,32172.514581,3,92,13,13,NA,4,4,0,2,0,2,50,2,1,5,NA
+65755,7,1,2,64,NA,1,1,NA,NA,NA,2,NA,2,2,4,NA,1,2,NA,2,2,2,NA,NA,NA,NA,NA,NA,NA,16352.915834,0,3,92,NA,NA,NA,5,5,2,1,1,2,64,2,1,2,NA
+65756,7,1,1,12,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7964.334824,0,1,95,6,6,1.34,4,4,0,2,0,2,32,2,3,2,NA
+65757,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,21857.756498,21879.219845,2,97,9,9,1.45,7,7,1,2,2,2,45,1,3,5,NA
+65758,7,2,2,67,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,12403.522912,13236.989116,1,93,3,3,0.65,3,3,0,0,3,2,74,2,1,2,NA
+65759,7,2,1,11,NA,3,3,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53915.042746,57006.573447,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+65760,7,2,1,41,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,37402.70356,37784.383858,2,102,14,14,3.8,4,4,2,0,0,2,41,2,4,1,5
+65761,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,67671.21344,73410.277865,3,90,15,15,5,3,3,1,0,0,1,31,1,5,1,5
+65762,7,2,2,11,NA,5,6,2,11,140,NA,NA,2,2,2,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6834.715094,7298.452383,2,90,3,1,0,5,1,1,2,0,1,44,2,5,1,5
+65763,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8949.073879,2,95,6,6,2.69,1,1,0,0,1,2,68,1,3,2,NA
+65764,7,2,2,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,63767.913736,65948.76781,1,94,6,6,1.91,2,2,0,0,2,1,78,1,4,1,3
+65765,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,128575.977224,134245.696628,1,102,6,4,1.42,2,1,0,0,0,2,23,1,5,5,NA
+65766,7,1,2,67,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,36706.470429,0,1,103,3,3,0.98,1,1,0,0,1,2,67,1,5,4,NA
+65767,7,2,1,1,12,3,3,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,40111.361732,44385.174851,1,98,7,7,1.48,5,5,1,1,0,1,46,1,3,1,3
+65768,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,38954.135779,39992.589933,1,91,5,5,1.3,3,3,0,1,0,2,50,1,4,2,NA
+65769,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,15051.024912,14836.484643,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+65770,7,2,1,40,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,37402.70356,40836.297693,2,102,7,7,1.79,4,4,0,2,0,1,40,2,2,6,NA
+65771,7,2,1,55,NA,5,7,1,NA,NA,1,2,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16575.817424,16910.393297,1,94,8,8,2.97,2,2,0,0,1,2,74,2,3,1,4
+65772,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,20131.904783,20903.590301,1,92,6,6,1.31,3,3,0,0,1,2,80,1,3,4,NA
+65773,7,2,1,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22937.913723,23181.846174,1,100,14,14,3.47,4,4,2,0,0,1,34,1,5,1,5
+65774,7,2,1,24,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,37911.437415,38428.199561,2,103,6,6,1.98,2,2,0,0,0,1,24,1,2,6,NA
+65775,7,2,2,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,18494.181242,18934.264494,2,99,9,9,2.43,4,4,0,2,0,2,49,1,3,3,NA
+65776,7,2,1,2,NA,4,4,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5853.073657,6453.970807,2,95,2,2,0.41,3,3,2,0,0,2,19,1,2,NA,NA
+65777,7,2,2,11,NA,4,4,2,12,144,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6655.097829,7106.52929,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+65778,7,2,1,11,NA,4,4,2,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,14179.490667,2,101,13,3,0.64,5,4,0,3,1,2,62,1,1,2,NA
+65779,7,2,2,2,NA,5,7,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8579.211947,8800.655964,1,97,9,9,2.6,4,4,1,1,0,2,45,1,4,1,4
+65780,7,2,2,7,NA,3,3,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24070.467912,25563.654853,1,95,10,6,1.34,5,4,1,2,0,1,32,1,3,6,NA
+65781,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,15091.588227,15171.391678,1,93,6,6,1.15,5,5,1,0,2,2,70,NA,NA,1,NA
+65782,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22565.644629,22012.541181,1,97,15,15,5,6,6,0,1,1,2,53,1,4,1,NA
+65783,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,124820.027137,131239.54259,1,91,10,10,3.78,3,3,0,0,2,1,62,1,5,1,5
+65784,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,17622.141982,17298.28124,2,95,7,7,2.58,2,2,0,0,0,2,57,1,4,3,NA
+65785,7,2,1,0,8,1,1,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,2,NA,2,2,2,2,NA,NA,NA,NA,6217.36354,6396.366428,3,91,7,7,1.23,6,6,2,2,0,1,36,2,1,1,1
+65786,7,2,1,8,NA,3,3,1,8,104,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49922.147265,51884.389986,1,98,7,7,1.66,5,5,2,1,0,2,37,1,5,1,3
+65787,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,86986.68246,91060.136414,2,94,7,7,2.38,2,2,0,1,0,1,39,1,4,3,NA
+65788,7,2,2,5,NA,5,6,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7143.995395,7265.801592,3,91,15,15,5,4,4,1,1,0,1,40,2,5,1,5
+65789,7,2,1,5,NA,4,4,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7430.518669,8193.361877,2,99,1,1,0.07,4,4,1,1,0,2,24,1,2,5,NA
+65790,7,2,2,5,NA,2,2,2,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11429.37307,11793.034101,2,90,2,2,0.49,3,3,2,0,0,2,26,1,4,1,NA
+65791,7,2,1,24,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19775.099106,19464.255682,1,96,15,15,5,3,3,0,0,0,1,55,1,4,1,5
+65792,7,2,1,0,6,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6999.189812,6930.018579,3,92,4,4,0.43,7,7,2,2,0,2,36,1,2,6,NA
+65793,7,2,2,16,NA,3,3,2,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25543.162298,26253.778018,2,95,6,6,1.19,4,4,0,1,0,1,44,1,3,1,2
+65794,7,2,1,9,NA,5,6,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5062.43381,5485.75338,3,91,6,6,1.22,5,5,1,2,0,2,37,1,4,1,2
+65795,7,2,1,9,NA,4,4,1,9,111,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14792.384662,15048.994189,2,102,5,3,0.63,5,4,2,1,0,1,24,1,4,6,NA
+65796,7,2,1,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,28479.463721,2,101,1,1,0.1,2,2,0,0,0,1,56,1,3,6,NA
+65797,7,2,2,52,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19735.328357,19287.640039,1,90,8,8,4.48,1,1,0,0,0,2,52,2,4,3,NA
+65798,7,2,2,26,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,3,1,2,1,2,2,2,2,2,1,2,2,1,42621.881199,43251.535191,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+65799,7,2,1,21,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15572.117537,16108.231355,2,94,15,15,4.44,5,5,0,1,1,2,74,1,5,2,NA
+65800,7,2,2,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,153418.292529,156457.542801,1,92,15,15,5,3,3,0,0,2,2,61,1,5,1,5
+65801,7,2,2,9,NA,4,4,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.490652,8919.477637,2,97,6,6,1,6,6,1,2,2,2,60,1,2,2,NA
+65802,7,1,2,9,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7710.907339,0,2,99,6,6,0.94,7,7,0,4,0,2,32,1,3,1,3
+65803,7,2,2,15,NA,3,3,2,15,189,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38400.791741,41869.717003,1,95,7,7,1.66,5,5,0,3,0,1,34,1,2,1,4
+65804,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,185033.990584,192737.162475,1,97,14,14,4.96,2,2,0,0,0,1,59,1,4,1,5
+65805,7,2,2,11,NA,3,3,2,11,139,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19824.800404,20630.46787,1,101,6,6,1.21,4,4,0,2,0,2,33,1,2,6,NA
+65806,7,2,1,5,NA,1,1,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16632.464801,17158.955649,3,91,3,3,0.39,6,6,1,1,0,1,39,2,1,6,NA
+65807,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108408.375382,108975.782069,2,98,14,14,4.16,3,3,0,0,0,1,49,1,5,1,4
+65808,7,2,1,65,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7610.147862,2,100,9,9,2.46,4,4,1,1,1,2,59,1,3,1,3
+65809,7,2,1,0,8,3,3,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22795.485282,22401.400888,1,92,9,9,3.24,3,3,1,0,0,1,30,1,5,6,NA
+65810,7,2,2,2,NA,3,3,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,44572.321124,49194.06017,1,98,8,8,2.46,3,3,1,0,0,1,29,1,4,6,NA
+65811,7,2,2,11,NA,1,1,2,11,136,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13085.954443,13661.915968,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+65812,7,2,1,30,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14894.763298,16100.128566,2,92,6,6,1.35,3,3,1,0,0,2,32,1,5,1,5
+65813,7,2,1,11,NA,1,1,2,11,138,NA,NA,2,1,3,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,13285.093011,14158.005149,2,94,7,7,1.57,4,4,0,2,0,1,30,2,3,1,4
+65814,7,2,1,11,NA,4,4,1,11,134,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13213.679978,13442.903072,2,96,7,7,1.49,5,5,2,1,0,1,51,1,5,1,3
+65815,7,1,2,65,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,10033.449917,0,2,95,3,3,0.95,2,2,0,0,1,2,65,1,2,3,NA
+65816,7,2,1,6,NA,3,3,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23796.980721,24835.204126,1,92,4,4,0.61,5,5,1,2,0,1,34,1,3,6,NA
+65817,7,2,2,19,NA,3,3,2,19,235,2,NA,2,1,5,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,77001.138762,79232.910414,1,98,15,15,5,3,3,0,0,0,1,56,1,5,1,5
+65818,7,2,1,17,NA,5,6,2,17,211,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6852.192992,7202.81544,2,100,15,15,5,3,3,0,1,0,1,58,2,5,1,5
+65819,7,2,1,9,NA,1,1,1,9,108,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11159.151566,11335.739115,1,102,6,6,1.03,6,6,0,4,0,1,34,2,2,1,1
+65820,7,2,1,77,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,2,2,NA,1,2,1,1,2,1,1,2,1,NA,9681.885604,10696.882276,1,93,2,2,0.74,1,1,0,0,1,1,77,2,2,2,NA
+65821,7,2,2,5,NA,5,6,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8171.700571,8311.029296,1,92,15,15,5,4,4,1,1,0,1,38,2,5,1,5
+65822,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,88617.795432,89088.673687,1,91,5,5,2.2,1,1,0,0,1,1,63,1,2,2,NA
+65823,7,2,1,37,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,2,2,2,1,2,2,1,46446.757775,47702.387865,1,102,1,1,0.33,2,2,0,0,0,1,37,1,2,4,NA
+65824,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,104228.438447,104101.346681,2,95,10,10,5,1,1,0,0,0,1,57,1,5,5,NA
+65825,7,2,2,8,NA,4,4,1,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7619.934086,8032.046596,2,103,7,7,1.55,5,5,2,2,0,2,31,1,4,3,NA
+65826,7,2,1,15,NA,1,1,1,16,192,NA,NA,2,2,4,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,29234.272259,28953.402615,3,92,13,13,NA,4,4,0,2,0,2,50,2,1,5,NA
+65827,7,2,1,42,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,4,6,NA,1,2,2,1,2,2,1,2,2,2,37402.70356,37112.276585,2,102,15,14,5,5,1,0,3,0,1,42,2,4,6,NA
+65828,7,2,2,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,118761.81384,123645.102237,3,91,7,4,1.74,2,1,0,0,0,1,23,NA,NA,6,NA
+65829,7,2,2,3,NA,3,3,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,69494.442609,71683.798234,1,98,15,15,4.56,4,4,2,0,0,2,33,1,4,1,4
+65830,7,2,1,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15546.999135,15568.006794,3,92,5,5,0.87,4,4,2,0,0,2,28,1,3,1,3
+65831,7,2,2,19,NA,4,4,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12531.903464,13043.632492,2,100,1,1,0.08,5,5,1,2,0,2,19,1,3,NA,NA
+65832,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,1,2,1,2,2,1,1,2,NA,15852.523312,16085.826144,2,97,6,6,2.04,2,2,0,0,2,2,80,1,3,2,NA
+65833,7,2,2,1,13,1,1,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11901.705423,12652.027961,2,102,4,4,0.65,5,5,1,0,0,2,58,2,1,4,NA
+65834,7,2,1,6,NA,4,4,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10229.206765,10406.656985,1,96,15,15,4.52,6,6,0,4,0,1,46,1,4,1,4
+65835,7,2,1,18,NA,5,6,2,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,77,77,NA,5,5,0,2,0,1,46,2,3,1,3
+65836,7,2,2,18,NA,5,7,2,18,216,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,58352.457563,59447.870488,1,99,77,77,NA,3,3,0,0,0,1,42,1,4,6,NA
+65837,7,2,1,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,11212.469396,11300.176858,1,100,15,99,NA,5,2,0,0,3,2,50,1,4,6,NA
+65838,7,2,1,63,NA,2,2,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11061.174348,11020.288023,1,95,7,7,2.65,2,2,0,0,1,1,63,1,4,1,NA
+65839,7,2,1,4,NA,1,1,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14318.290734,14771.527769,3,91,15,14,4.03,5,4,2,0,0,1,42,2,4,1,5
+65840,7,2,1,21,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,12395.607222,12971.055335,1,102,9,9,2.39,5,5,0,1,1,1,55,2,5,1,5
+65841,7,2,1,63,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,85242.614785,86042.075041,1,93,15,15,5,1,1,0,0,1,1,63,1,5,5,NA
+65842,7,2,1,0,3,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9064.168162,9064.335231,3,92,14,14,4.71,3,3,1,0,0,1,29,1,5,1,5
+65843,7,2,1,9,NA,1,1,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12477.812875,13297.681754,2,94,5,5,0.67,6,6,1,3,0,1,37,2,3,1,4
+65844,7,2,1,69,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,2,1,2,1,2,2,2,1,2,NA,12118.033999,12538.522644,2,91,3,3,0.66,2,2,0,0,1,1,69,2,5,1,1
+65845,7,2,1,13,NA,5,6,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7350.524832,7855.79983,2,90,8,8,2.59,3,3,0,1,0,2,41,2,4,1,4
+65846,7,2,2,61,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,7278.790659,7582.574348,2,93,15,15,5,4,4,0,0,1,1,57,2,5,1,4
+65847,7,2,1,3,NA,2,2,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14004.176901,13773.448831,2,93,3,3,0.66,2,2,1,0,0,2,31,2,3,5,NA
+65848,7,1,1,11,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7803.43213,0,1,93,15,15,5,3,3,0,2,0,2,40,2,5,4,NA
+65849,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,17753.406962,18715.910687,1,101,4,4,1.29,2,2,0,0,2,2,74,1,3,1,3
+65850,7,2,1,58,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,25118.469449,25612.281104,3,92,5,5,1.56,2,2,0,0,1,1,58,1,3,1,2
+65851,7,2,2,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,50915.06085,56529.78018,3,92,4,4,0.65,4,4,2,0,0,2,20,1,3,5,NA
+65852,7,2,2,29,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,40212.914634,41314.674899,2,92,15,10,5,2,1,0,0,0,2,29,1,5,6,NA
+65853,7,2,2,14,NA,3,3,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,97212.131473,101229.4133,1,91,14,14,4.03,4,4,0,2,0,1,52,1,4,1,5
+65854,7,2,1,56,NA,5,7,2,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11690.444016,12008.407525,3,90,9,9,3.14,3,3,0,0,0,1,56,2,3,1,3
+65855,7,2,1,8,NA,3,3,2,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23796.980721,24835.204126,1,95,5,5,1.03,4,4,0,2,0,1,33,1,3,1,3
+65856,7,2,1,63,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,8060.574062,8123.626407,1,93,7,7,1.61,4,4,0,0,1,2,27,2,3,5,NA
+65857,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,97050.018399,96721.703127,1,99,14,14,5,1,1,0,0,1,2,64,1,5,3,NA
+65858,7,2,2,5,NA,1,1,2,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13366.393396,14209.054666,2,94,7,7,1.23,6,6,2,1,0,1,33,2,1,6,NA
+65859,7,2,2,0,10,3,3,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25770.951107,25070.954843,2,91,15,15,4.34,4,4,2,0,0,1,39,1,5,1,5
+65860,7,2,2,5,NA,4,4,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11142.946789,11989.935044,2,96,5,5,0.67,6,6,1,2,1,1,34,1,4,1,4
+65861,7,2,1,8,NA,2,2,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,10248.861635,10307.698588,1,96,15,15,5,4,4,0,2,0,1,36,2,3,1,4
+65862,7,2,1,54,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,27227.937106,27194.736507,1,101,3,3,1.1,1,1,0,0,0,1,54,1,4,3,NA
+65863,7,2,2,33,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,15817.041096,15849.338176,2,92,15,8,4.59,2,1,0,0,0,2,25,2,5,5,NA
+65864,7,2,1,0,1,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13392.309303,13898.571164,1,99,14,14,3.44,5,5,3,0,0,1,30,1,4,1,5
+65865,7,2,2,34,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,NA,NA,NA,NA,39561.667842,39822.148122,2,98,12,12,NA,3,3,0,1,0,2,34,1,4,1,3
+65866,7,2,2,6,NA,4,4,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9209.511624,9916.240201,1,100,5,5,0.85,5,5,0,2,0,2,54,1,2,2,NA
+65867,7,2,2,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,37934.469637,38685.959862,1,94,6,6,1.98,2,2,0,0,1,2,60,1,4,3,NA
+65868,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,NA,53069.077479,57205.244401,2,90,15,15,5,2,1,0,0,2,1,67,1,5,6,NA
+65869,7,2,2,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,10252.529496,2,100,4,4,1.16,2,2,0,0,1,2,18,1,2,NA,NA
+65870,7,2,2,1,12,5,6,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,4488.195848,4764.622817,2,92,99,2,0.31,7,4,3,3,1,1,61,2,1,1,3
+65871,7,2,2,6,NA,1,1,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18721.371751,18903.185351,1,95,6,6,1.37,3,3,1,1,0,2,28,1,4,5,NA
+65872,7,2,1,1,15,1,1,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10666.404974,11148.562293,2,96,8,8,1.33,7,7,2,1,1,1,62,2,1,1,1
+65873,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,18018.210636,21650.053885,2,91,14,14,3.47,4,4,1,1,0,2,36,2,3,1,5
+65874,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,127315.335607,127929.014806,1,91,10,10,2.77,5,5,0,3,0,1,43,1,5,1,5
+65875,7,2,2,4,NA,1,1,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16648.051651,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+65876,7,2,1,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21288.18311,20984.737161,1,102,1,1,0,5,5,0,3,0,2,41,1,4,1,4
+65877,7,2,1,76,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,1,1,2,2,1,2,2,NA,10123.333237,10321.62644,1,93,2,2,0.54,2,2,0,0,2,1,76,2,4,1,1
+65878,7,2,1,29,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,1,NA,NA,NA,NA,9177.295801,9603.338468,2,92,7,7,1.89,3,3,0,0,1,1,36,2,3,5,NA
+65879,7,2,2,32,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,45655.090694,45995.96139,3,92,15,15,5,3,3,1,0,0,1,34,1,5,1,5
+65880,7,2,2,26,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,53638.260635,61764.513393,2,96,4,4,1.47,1,1,0,0,0,2,26,1,4,5,NA
+65881,7,2,1,1,13,4,4,2,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6992.24593,7710.094516,2,97,2,2,0.27,3,3,1,0,0,2,21,1,3,6,NA
+65882,7,2,2,11,NA,2,2,2,11,140,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10762.400563,11222.736057,2,90,12,12,NA,4,4,0,2,0,2,38,2,4,1,4
+65883,7,2,1,79,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,7199.330978,7568.243922,1,93,2,2,0.64,1,1,0,0,1,1,79,2,3,4,NA
+65884,7,2,1,2,NA,5,6,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6672.795321,7206.439819,2,102,15,15,3.92,5,5,1,2,0,1,34,2,5,1,5
+65885,7,2,1,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,21941.544332,21717.573526,1,99,4,4,1.16,2,2,0,0,2,2,63,1,5,6,NA
+65886,7,2,2,2,NA,5,7,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8692.478172,9353.203477,1,100,6,6,1.18,5,5,1,2,0,2,30,1,3,1,4
+65887,7,2,1,0,1,1,1,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,5875.721278,6205.225081,2,97,6,6,1.1,5,5,2,1,0,2,29,2,2,6,NA
+65888,7,2,2,11,NA,4,4,1,11,134,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9139.784234,9759.758014,2,100,14,14,3.58,4,4,0,1,1,2,55,1,5,1,4
+65889,7,2,1,30,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,2,2,2,1,2,2,1,2,2,1,37080.526463,37488.446325,1,103,6,6,1.57,3,3,0,1,0,2,50,2,3,4,NA
+65890,7,2,1,3,NA,2,2,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15915.595287,17052.411707,2,96,14,14,3.36,4,4,1,1,0,2,28,1,2,6,NA
+65891,7,2,1,3,NA,4,4,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8961.738846,9235.465849,1,93,10,10,2.91,4,4,2,0,0,2,27,1,5,1,4
+65892,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,25840.959268,28861.347345,1,101,3,3,0.86,2,2,0,0,1,2,80,1,1,2,NA
+65893,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16058.142925,16013.018639,2,95,5,5,1.18,3,3,0,1,0,2,55,1,4,5,NA
+65894,7,2,1,12,NA,4,4,1,12,150,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13653.432599,13787.136512,2,96,6,6,1.62,3,3,0,2,0,2,31,1,3,5,NA
+65895,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,26922.241442,27209.395492,1,90,7,7,1.55,5,5,0,3,0,1,51,2,3,1,2
+65896,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,89807.047643,92868.804281,1,94,15,1,0.44,4,1,0,0,1,1,33,1,2,5,NA
+65897,7,2,1,13,NA,1,1,1,14,168,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23389.620035,23146.524434,1,94,10,10,2.94,4,4,0,2,0,2,52,1,5,2,NA
+65898,7,2,1,66,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,5950.975297,5997.525697,2,99,6,6,1.11,5,5,0,0,1,1,66,2,4,1,4
+65899,7,2,2,12,NA,4,4,2,12,147,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17426.123122,17678.558809,1,91,10,10,2.56,5,5,0,3,0,1,51,2,5,1,4
+65900,7,2,2,18,NA,3,3,1,18,218,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,76992.7514,81576.336255,1,100,15,15,4.07,5,5,0,2,0,2,41,1,5,1,4
+65901,7,2,1,22,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,11164.358128,11468.312676,2,93,1,1,0.09,1,1,0,0,0,1,22,1,4,5,NA
+65902,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,1,1,2,2,1,2,2,1,2,2,1,18723.98095,21433.166124,2,95,7,7,1.41,5,5,2,0,0,2,53,1,3,3,NA
+65903,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4912.962876,5216.242891,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+65904,7,2,2,4,NA,2,2,1,4,48,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13201.625134,13621.675873,2,92,15,15,5,4,4,1,1,0,2,47,1,5,1,5
+65905,7,2,1,54,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,1,1,NA,1,2,1,1,2,2,1,2,1,NA,15728.666463,15671.845555,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+65906,7,2,2,7,NA,2,2,1,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15009.847173,16312.401192,2,93,15,15,4.34,4,4,0,2,0,1,33,2,5,1,5
+65907,7,1,2,47,NA,2,2,NA,NA,NA,2,NA,2,7,77,NA,3,1,NA,2,1,2,1,2,2,NA,NA,NA,NA,23968.560941,0,2,90,99,99,NA,4,4,0,2,0,1,39,NA,NA,4,NA
+65908,7,2,1,17,NA,4,4,2,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18526.8521,18563.481203,1,97,6,6,1.41,3,3,0,1,0,2,51,1,4,5,NA
+65909,7,2,1,35,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19091.246741,19659.303383,1,91,10,10,3.78,3,3,1,0,0,1,35,2,5,1,5
+65910,7,2,1,36,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13601.994691,14414.229968,2,100,15,15,5,4,4,1,1,0,1,36,2,5,1,5
+65911,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,38833.86357,44548.098777,3,92,5,5,1.59,2,2,0,0,2,1,64,1,3,1,3
+65912,7,2,1,18,NA,5,6,1,18,220,2,NA,2,2,1,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9948.022697,10353.284725,2,101,99,99,NA,4,1,0,0,0,1,18,2,4,NA,NA
+65913,7,2,2,19,NA,4,4,2,19,239,2,NA,2,1,4,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11478.437608,11446.094627,1,96,7,7,1.69,4,4,0,1,0,2,19,2,4,NA,NA
+65914,7,2,2,79,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,17087.605763,2,90,2,2,0.87,1,1,0,0,1,2,79,1,2,2,NA
+65915,7,2,1,37,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,30626.581617,36447.669921,2,90,3,3,0.58,4,4,0,2,0,2,36,2,3,1,3
+65916,7,2,1,9,NA,4,4,2,9,117,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9738.268689,9808.886868,2,99,9,9,1.78,6,6,1,1,0,1,46,1,3,6,NA
+65917,7,2,2,6,NA,5,7,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18348.624469,20109.577944,2,90,77,77,NA,6,6,0,4,0,2,41,NA,NA,4,NA
+65918,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,2,77,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,15792.287872,16329.218615,1,103,12,12,NA,2,2,0,0,0,1,34,2,5,1,5
+65919,7,2,2,73,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,1,2,NA,1,2,1,1,2,2,1,2,1,NA,15825.056964,16318.16087,2,94,15,15,5,3,3,0,0,1,2,40,2,5,1,5
+65920,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,31552.004994,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+65921,7,2,2,41,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,19711.72366,19172.260632,1,96,5,5,1.45,2,2,0,1,0,2,41,1,4,3,NA
+65922,7,2,1,10,NA,4,4,2,10,121,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8861.237416,9017.822314,2,95,2,2,0.26,3,3,0,2,0,2,31,1,3,5,NA
+65923,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,138075.879417,141933.339512,2,91,14,14,5,1,1,0,0,1,1,62,1,5,3,NA
+65924,7,2,2,19,NA,4,4,1,19,238,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15006.987722,14964.702287,1,98,2,1,0.21,4,1,0,0,0,2,19,1,4,NA,NA
+65925,7,2,1,23,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,29738.952706,30589.306008,2,94,6,3,0.68,3,2,0,0,0,1,26,1,4,5,NA
+65926,7,2,2,14,NA,4,4,1,14,170,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11791.755593,11820.79689,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+65927,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,111378.575836,111001.787784,2,97,8,8,3.06,2,2,0,0,2,1,68,1,2,1,4
+65928,7,2,2,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,17884.885732,18357.680967,1,98,6,6,1.73,3,3,0,1,0,2,39,1,4,1,1
+65929,7,2,1,64,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,35869.019314,40296.63257,3,92,5,5,1.59,2,2,0,0,2,1,64,1,3,1,3
+65930,7,2,2,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,59902.67172,68981.485025,1,101,3,3,1.3,1,1,0,0,1,2,80,1,3,2,NA
+65931,7,2,1,22,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,2,5,NA,1,2,2,1,2,2,1,2,2,NA,10141.381563,10817.596093,3,90,12,12,NA,4,4,0,0,1,1,62,2,4,3,NA
+65932,7,2,1,17,NA,2,2,1,17,215,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18242.832494,18053.229058,1,102,77,77,NA,6,6,0,2,1,2,37,1,4,1,4
+65933,7,2,1,11,NA,3,3,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,54897.892683,57357.850008,2,98,15,15,5,3,3,0,1,0,1,56,1,5,1,5
+65934,7,2,1,3,NA,3,3,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,75551.990912,88526.885345,1,90,14,14,3.93,3,3,1,0,0,1,35,1,2,1,5
+65935,7,2,1,18,NA,4,4,2,18,223,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12530.944806,12773.594654,2,90,8,8,1.67,6,6,1,1,0,1,52,1,3,1,5
+65936,7,2,1,7,NA,3,3,2,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64387.576177,71283.768751,1,95,8,8,2.7,3,3,0,1,2,1,69,1,5,1,3
+65937,7,2,2,61,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,3,4,NA,2,2,2,2,2,2,2,2,2,2,9716.805546,10855.679158,2,90,12,12,NA,3,3,0,0,2,2,61,2,3,4,NA
+65938,7,2,1,0,2,1,1,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9064.168162,9064.335231,3,92,4,4,0.89,3,3,1,0,0,2,24,2,4,1,3
+65939,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17207.598344,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+65940,7,2,2,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,13851.686232,14498.558979,3,91,15,3,0.92,5,1,2,0,0,1,42,2,4,1,5
+65941,7,2,1,6,NA,2,2,1,6,74,NA,NA,2,1,3,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10212.603023,10439.995832,2,92,15,15,5,4,4,1,1,0,2,47,1,5,1,5
+65942,7,2,2,3,NA,1,1,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19235.084509,19847.108514,3,92,15,8,2.62,4,3,1,1,0,1,30,1,2,6,NA
+65943,7,2,1,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8009.966208,8254.622305,2,96,3,3,0.54,4,4,2,1,0,2,25,1,4,2,NA
+65944,7,2,2,64,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,10308.451947,2,90,9,9,5,1,1,0,0,1,2,64,2,4,3,NA
+65945,7,2,2,2,NA,1,1,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13464.808163,14038.51136,1,101,2,2,0.26,5,5,3,0,0,2,26,1,2,1,3
+65946,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,15415.338508,15856.841729,1,93,4,4,1.65,1,1,0,0,0,1,34,1,5,4,NA
+65947,7,2,2,3,NA,5,6,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5485.572703,5823.427887,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+65948,7,2,1,5,NA,4,4,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9388.597537,2,100,3,3,0.38,5,5,2,1,0,2,28,1,2,5,NA
+65949,7,1,1,48,NA,2,2,NA,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,35393.002863,0,2,93,4,4,0.84,3,3,0,1,0,2,46,2,3,1,2
+65950,7,2,1,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,23471.353577,23675.625296,1,100,14,6,1.85,3,2,1,0,0,1,33,1,5,5,NA
+65951,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,102085.27025,105869.935801,1,99,14,14,4.86,3,3,0,1,0,1,56,1,5,1,5
+65952,7,2,1,17,NA,2,2,1,17,212,2,NA,2,2,1,13,NA,NA,NA,2,2,2,1,2,2,2,2,2,2,24228.858782,24791.594558,2,102,6,6,1,6,6,1,3,0,1,35,2,3,1,3
+65953,7,2,2,24,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,5,1,2,2,2,2,2,2,1,2,2,1,36169.442288,36681.102865,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+65954,7,2,2,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10440.656902,1,99,12,12,NA,1,1,0,0,1,2,66,1,5,1,NA
+65955,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,29016.444561,32518.702589,3,91,4,4,1.16,2,2,0,0,2,1,80,1,5,1,5
+65956,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,86986.68246,89821.123051,2,94,12,12,NA,5,5,1,1,0,1,37,1,4,1,3
+65957,7,2,2,47,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,30039.01139,30195.662279,2,95,8,8,3.44,2,2,0,0,0,2,24,1,4,5,NA
+65958,7,2,2,13,NA,3,3,1,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,34778.638222,37920.357213,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+65959,7,2,1,13,NA,1,1,1,13,159,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,1,1,2,2,1,20560.901695,20875.182272,2,96,3,3,0.46,5,5,1,2,0,1,37,1,1,1,2
+65960,7,2,2,0,4,4,4,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4137.127382,4386.966342,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+65961,7,2,2,0,0,3,3,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17773.146728,17119.196492,2,98,99,99,NA,7,7,1,1,1,1,19,1,3,NA,NA
+65962,7,2,2,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,16966.723528,16502.38463,2,92,2,2,0.57,2,2,0,0,0,2,56,1,3,2,NA
+65963,7,2,2,60,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,3,2,NA,1,2,1,NA,NA,NA,1,2,1,3,7575.470578,7867.856697,2,92,12,NA,NA,7,1,0,0,2,1,53,2,3,1,3
+65964,7,2,1,13,NA,4,4,2,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18526.8521,18563.481203,1,97,7,7,1.74,4,4,0,3,0,2,32,1,4,5,NA
+65965,7,2,1,6,NA,2,2,1,6,77,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8635.515602,8675.270774,2,93,4,4,0.56,5,5,0,2,0,1,37,NA,NA,1,1
+65966,7,2,1,19,NA,4,4,2,19,230,2,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16658.161391,16980.73087,2,91,2,2,0.26,4,4,0,1,0,1,20,1,3,5,NA
+65967,7,2,2,36,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,35353.005268,35616.958386,2,94,8,8,2.01,4,4,1,1,0,1,44,2,4,1,4
+65968,7,2,2,8,NA,2,2,1,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12307.832776,12427.36095,2,93,14,14,3.25,4,4,0,2,0,2,46,2,5,1,4
+65969,7,2,1,14,NA,1,1,2,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20398.562455,20202.582308,2,94,14,14,4.03,4,4,0,2,0,2,33,2,2,1,NA
+65970,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,20901.316439,20413.081322,2,91,14,14,3.42,5,5,1,0,0,2,28,NA,NA,1,NA
+65971,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,80069.373466,86138.346669,1,98,6,6,1.31,3,3,1,0,0,1,30,1,5,1,5
+65972,7,2,2,1,17,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,11667.597427,3,92,6,6,1.7,2,2,1,0,0,2,20,1,3,4,NA
+65973,7,2,2,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9287.93089,10379.213194,2,90,7,7,1.61,4,4,1,1,1,2,65,1,3,2,NA
+65974,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,10,10,4.49,2,2,0,0,2,1,60,1,5,1,4
+65975,7,2,2,30,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,5,4,2,2,2,2,2,2,2,1,2,2,1,19486.670926,20613.901312,2,90,5,5,1.19,3,3,1,0,1,2,60,2,1,4,NA
+65976,7,2,1,13,NA,4,4,1,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13523.666124,13550.403519,2,98,3,3,0.88,2,2,0,1,0,2,38,1,4,5,NA
+65977,7,2,1,3,NA,2,2,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11992.176902,12455.795985,1,96,10,10,2.59,5,5,1,0,0,1,32,2,4,1,2
+65978,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,145224.318312,145047.237897,1,98,6,6,2.86,1,1,0,0,0,1,57,1,4,5,NA
+65979,7,2,2,2,NA,5,7,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6443.362832,6553.223138,1,98,14,14,3.9,4,4,2,0,0,1,39,1,5,1,4
+65980,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,13076.210481,14300.146605,2,97,5,5,1.19,3,3,0,1,0,2,41,1,3,1,2
+65981,7,2,1,8,NA,2,2,1,8,102,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,13742.678011,2,98,7,7,2.16,3,3,0,1,0,2,51,1,1,1,2
+65982,7,2,2,4,NA,3,3,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78165.891242,86270.974005,1,101,6,6,1.28,4,4,2,0,0,1,44,1,4,1,4
+65983,7,2,1,9,NA,4,4,2,9,112,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,8017.552697,8398.795399,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+65984,7,2,2,16,NA,5,6,2,16,196,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11225.761275,11659.847432,2,91,8,8,2.34,4,4,0,2,0,1,56,2,5,1,5
+65985,7,2,2,7,NA,5,6,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5139.061504,5511.809249,2,103,77,77,NA,5,5,0,2,0,2,39,2,5,1,5
+65986,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,82864.936499,86441.088971,1,99,15,15,5,4,4,0,2,0,2,44,1,5,1,5
+65987,7,2,2,29,NA,5,7,1,NA,NA,2,NA,2,1,6,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,47970.616628,49304.059008,1,102,10,10,3.22,4,4,0,0,2,2,29,2,5,5,NA
+65988,7,2,1,20,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21763.209029,22385.504537,2,92,4,1,0.23,4,1,0,0,0,1,21,1,4,5,NA
+65989,7,2,1,1,18,2,2,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8326.330571,8424.765655,2,93,3,3,0.37,5,5,3,0,0,1,28,2,1,6,NA
+65990,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,1,1,2,2,1,2,1,NA,17232.67865,17489.495288,1,100,99,99,NA,6,6,0,1,0,1,53,2,2,1,3
+65991,7,2,2,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,118761.81384,121363.708734,3,91,15,6,2.75,3,1,0,0,0,2,26,1,4,5,NA
+65992,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11699.431733,12601.246533,1,96,14,14,5,2,2,0,0,2,2,61,1,3,1,3
+65993,7,2,2,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,149765.47604,152732.36321,1,92,8,8,2.17,4,4,0,1,2,2,80,1,3,2,NA
+65994,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7903.072331,7964.892648,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+65995,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,21756.194073,0,1,93,2,2,0.59,1,1,0,0,1,2,80,1,3,2,NA
+65996,7,2,2,9,NA,3,3,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20242.832328,19974.015949,2,93,6,6,2.05,2,2,0,1,0,2,30,1,4,3,NA
+65997,7,2,2,17,NA,4,4,1,17,209,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13620.028009,14467.639304,2,93,9,9,2.07,5,5,0,1,0,1,55,NA,NA,5,NA
+65998,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,7869.59899,8220.968168,2,100,3,3,1.1,1,1,0,0,1,2,67,1,4,5,NA
+65999,7,2,1,69,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,3,NA,2,2,2,1,2,2,1,2,2,2,9404.30514,9554.950099,2,93,6,6,1.72,2,2,0,0,2,1,69,2,4,3,NA
+66000,7,2,2,54,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,24287.448154,24414.10513,2,92,15,10,5,2,1,0,0,1,2,54,2,5,3,NA
+66001,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,100988.137931,99957.29004,1,99,15,15,5,2,2,0,0,2,1,62,1,5,1,4
+66002,7,2,1,11,NA,1,1,1,11,143,NA,NA,2,7,77,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11367.678664,11216.859778,2,96,77,77,NA,7,7,3,2,0,2,33,2,2,6,NA
+66003,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,4,2,0.64,2,1,0,0,0,1,24,1,4,5,NA
+66004,7,2,1,2,NA,1,1,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13305.770449,13463.073191,3,92,6,6,1.06,5,5,2,0,0,2,54,2,1,77,NA
+66005,7,2,2,18,NA,2,2,2,18,219,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,6,6,1.21,4,4,0,0,0,2,59,2,1,6,NA
+66006,7,2,2,18,NA,3,3,2,18,218,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,36586.371708,44099.282264,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+66007,7,2,2,9,NA,3,3,2,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,23750.822126,1,95,7,7,1.17,6,6,1,3,0,2,44,1,4,1,NA
+66008,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,154825.466557,160454.355913,1,91,15,14,5,3,2,0,0,2,2,73,1,4,3,NA
+66009,7,2,2,0,6,4,4,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4581.358327,5045.31579,2,97,3,3,0.66,3,3,2,0,0,2,19,1,3,NA,NA
+66010,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,27934.372045,28826.461363,1,101,6,6,1.17,4,4,0,1,0,1,41,1,3,6,NA
+66011,7,2,2,12,NA,4,4,2,13,156,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10848.628906,11198.221038,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+66012,7,2,1,2,NA,1,1,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11160.282155,11175.362327,2,97,8,8,2.01,4,4,2,0,0,2,24,1,4,1,1
+66013,7,2,1,23,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,35338.972549,35820.670014,2,90,2,2,0.25,5,5,0,1,0,2,41,2,4,1,NA
+66014,7,2,2,5,NA,4,4,2,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10389.292229,10964.092156,2,95,1,1,0.09,5,5,3,1,0,2,31,1,2,1,NA
+66015,7,2,1,64,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,7926.298973,8333.304577,2,96,8,8,3.06,2,2,0,0,1,1,64,2,5,1,5
+66016,7,2,2,4,NA,5,6,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7349.373527,7474.681451,2,91,7,7,1.56,4,4,1,1,0,2,37,2,5,1,5
+66017,7,2,2,11,NA,3,3,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,48532.852397,48387.04619,1,98,15,15,5,5,5,0,3,0,2,41,1,5,6,NA
+66018,7,2,2,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10788.880391,11608.95565,2,95,6,6,0.97,6,6,2,2,0,1,37,1,3,1,4
+66019,7,2,1,12,NA,4,4,1,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16089.133406,15900.017206,2,102,3,3,0.76,3,3,0,1,1,2,66,1,2,3,NA
+66020,7,2,2,9,NA,1,1,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,NA,15225.935813,15373.8033,2,94,7,7,1.79,4,4,1,1,0,2,32,1,4,1,4
+66021,7,2,1,33,NA,5,7,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,25461.575888,25330.616054,1,93,15,15,5,2,2,0,0,0,2,34,1,5,1,5
+66022,7,2,2,35,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,28958.579549,28751.652767,1,96,12,12,NA,5,5,1,2,0,2,35,1,5,1,4
+66023,7,2,1,2,NA,5,6,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10123.286306,10611.741928,1,100,10,10,3.04,4,4,2,0,0,1,30,2,5,1,5
+66024,7,2,2,11,NA,1,1,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15841.451259,16252.614023,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+66025,7,2,1,8,NA,4,4,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9468.438225,10001.401449,2,99,6,6,1.57,3,3,0,2,0,2,31,1,3,77,NA
+66026,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,65748.417123,66365.048213,2,99,15,15,5,1,1,0,0,1,1,65,1,5,3,NA
+66027,7,2,2,40,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,5,1,3,1,2,2,1,2,2,1,2,2,1,31235.666551,32148.463094,2,94,6,6,2.94,1,1,0,0,0,2,40,2,5,1,NA
+66028,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,96499.801072,98371.919459,1,92,14,14,3.93,3,3,1,0,0,1,20,1,4,1,4
+66029,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,112153.23206,1,91,8,8,2.17,4,4,0,0,0,1,59,1,4,1,5
+66030,7,2,1,13,NA,1,1,1,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,29234.272259,29395.83763,3,92,10,1,0,5,1,2,1,1,2,68,1,3,1,1
+66031,7,2,2,56,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,19183.115011,19284.554221,1,92,4,4,1.75,1,1,0,0,0,2,56,1,5,5,NA
+66032,7,2,1,36,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,30626.581617,32055.135325,2,90,14,14,3.45,4,4,1,1,0,2,34,2,5,6,NA
+66033,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,64581.191728,67036.946354,2,94,99,99,NA,2,2,0,0,0,2,37,1,2,1,4
+66034,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,26847.643051,29218.940922,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+66035,7,2,1,0,5,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4461.618312,4741.503802,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+66036,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,94,9,9,3.97,2,2,0,0,2,1,74,1,4,1,4
+66037,7,2,2,14,NA,4,4,2,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,12440.215685,1,98,10,10,3.77,3,3,0,1,0,1,48,NA,NA,1,4
+66038,7,2,2,2,NA,1,1,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10271.907274,2,94,77,77,NA,4,4,2,0,0,1,26,2,2,1,4
+66039,7,2,1,8,NA,3,3,2,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15653.970322,16628.805574,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+66040,7,2,2,24,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,NA,NA,NA,1,2,2,1,40149.982555,41250.018596,2,103,6,6,1.98,2,2,0,0,0,1,24,1,2,6,NA
+66041,7,2,1,58,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,31872.125984,32498.709833,2,96,8,8,3.67,2,2,0,0,0,1,58,1,3,3,NA
+66042,7,2,1,15,NA,2,2,2,15,186,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18909.886675,20785.327762,2,90,8,8,1.72,5,5,1,2,0,1,20,2,1,1,2
+66043,7,2,2,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,166028.936087,173980.461474,2,101,12,12,NA,1,1,0,0,1,2,64,1,3,1,NA
+66044,7,2,1,14,NA,5,6,1,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12013.02156,13498.577364,3,92,12,12,NA,5,5,0,2,0,1,47,1,3,1,3
+66045,7,2,1,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,22506.522055,23661.545768,1,98,4,4,0.75,4,4,0,1,0,2,48,1,2,1,3
+66046,7,2,2,0,0,1,1,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9767.083234,10416.172562,3,92,8,8,2,4,4,2,0,0,1,30,1,4,1,4
+66047,7,2,2,0,1,4,4,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3594.351671,3811.412689,2,99,6,6,1.11,5,5,1,2,0,2,41,1,2,5,NA
+66048,7,2,2,6,NA,5,6,1,6,72,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10387.513218,11140.943027,1,92,77,77,NA,4,4,1,1,0,2,40,2,4,1,4
+66049,7,2,1,73,NA,2,2,2,NA,NA,2,NA,2,2,8,NA,5,1,NA,2,2,2,2,2,2,2,2,2,NA,12343.565567,12789.649007,3,90,10,10,3.04,4,4,0,0,2,2,80,2,1,3,NA
+66050,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,10072.885959,10423.387676,2,100,2,2,0.78,1,1,0,0,1,2,80,1,4,2,NA
+66051,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,30451.148148,30372.925419,2,101,4,4,0.73,5,5,1,2,0,1,40,1,5,1,5
+66052,7,2,2,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,138322.767578,145546.941919,1,101,7,7,2.31,2,2,0,1,0,2,43,1,4,3,NA
+66053,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,133492.667054,133041.068157,1,95,14,14,5,2,2,0,0,2,2,65,1,4,1,4
+66054,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,95214.22557,100722.009256,1,97,15,15,5,4,4,1,1,0,2,33,1,5,1,3
+66055,7,2,1,4,NA,4,4,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7311.663111,7763.016267,2,93,4,4,0.56,5,5,2,1,0,1,27,1,2,6,NA
+66056,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,23325.73926,22943.576186,1,101,10,10,5,1,1,0,0,0,2,57,1,3,3,NA
+66057,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,24004.078396,24148.434886,2,99,2,2,0.81,1,1,0,0,0,2,48,1,2,5,NA
+66058,7,2,2,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,64463.340883,65151.773075,1,91,8,8,3.4,2,2,0,0,2,2,74,1,4,1,2
+66059,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,11019.434708,11234.197915,3,91,15,15,4.47,4,4,0,1,2,2,79,1,4,3,NA
+66060,7,2,2,8,NA,2,2,2,8,102,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15148.721588,15796.67129,2,91,8,1,0,3,1,0,2,0,2,33,2,4,5,NA
+66061,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,22295.761589,23482.073935,3,92,4,4,1.34,1,1,0,0,0,1,34,1,1,6,NA
+66062,7,2,1,7,NA,3,3,2,8,96,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23729.905536,29042.69522,1,101,3,3,0.61,4,4,1,2,0,1,38,1,2,4,NA
+66063,7,2,1,0,0,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4944.997189,5222.307037,2,103,13,13,NA,5,5,2,1,0,1,32,2,2,1,2
+66064,7,2,2,3,NA,1,1,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,21465.084926,22148.063325,1,92,4,4,0.74,4,4,1,1,0,1,42,2,3,1,4
+66065,7,2,2,2,NA,5,6,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8894.95474,1,97,14,14,4.5,3,3,1,0,0,1,30,1,5,1,5
+66066,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,34718.816885,38024.444986,1,91,15,15,5,3,3,0,0,2,1,60,1,5,1,3
+66067,7,2,2,7,NA,4,4,2,7,95,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10994.192555,11605.838362,2,97,3,3,0.46,5,5,0,3,0,1,40,1,2,1,3
+66068,7,2,1,76,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,6945.351825,7467.825932,1,103,5,5,0.65,6,6,0,0,1,2,26,2,4,5,NA
+66069,7,1,2,55,NA,2,2,NA,NA,NA,2,NA,2,1,7,NA,2,4,NA,2,2,2,2,2,2,NA,NA,NA,NA,24902.414229,0,1,90,77,77,NA,2,2,0,0,0,2,55,2,2,4,NA
+66070,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,12174.606153,16326.594828,1,90,7,7,1.3,5,5,1,2,1,2,62,1,2,2,NA
+66071,7,2,2,65,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11862.436765,13289.857171,1,90,7,7,3.85,1,1,0,0,1,2,65,1,3,1,NA
+66072,7,1,1,77,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,38226.070503,0,1,99,NA,NA,NA,2,2,0,0,2,1,77,1,3,1,NA
+66073,7,2,1,80,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,2,NA,2,2,2,1,2,2,2,2,2,NA,8388.502332,8563.888023,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+66074,7,1,1,42,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,18790.641284,0,2,100,15,15,5,4,3,0,2,0,1,42,1,3,5,NA
+66075,7,2,1,69,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,3,12579.986433,13271.133625,1,92,7,7,2.1,3,3,0,0,2,1,37,2,5,5,NA
+66076,7,2,1,17,NA,2,2,1,17,213,2,NA,2,2,3,11,NA,NA,NA,2,2,2,1,2,2,2,2,2,2,22721.243258,23068.545411,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+66077,7,2,1,63,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22234.74698,22519.946154,2,94,15,15,5,2,2,0,0,1,1,63,1,5,1,4
+66078,7,2,1,3,NA,3,3,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,32718.806351,38337.758985,1,94,6,6,1.21,4,4,2,0,0,1,27,1,2,1,2
+66079,7,2,1,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,26946.665552,2,101,4,4,1.65,1,1,0,0,0,1,54,1,3,5,NA
+66080,7,2,1,8,NA,4,4,2,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10651.061092,11315.742189,1,90,6,6,1.57,3,3,0,1,0,2,36,1,3,5,NA
+66081,7,1,1,8,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14610.93355,0,2,102,14,14,3.8,4,4,0,2,0,1,47,1,4,1,4
+66082,7,2,2,9,NA,4,4,1,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10332.067017,12551.283611,1,100,9,9,2.6,4,4,0,1,0,1,45,1,3,1,3
+66083,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,36038.266622,37278.43838,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+66084,7,2,2,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,73204.401173,80916.420945,2,91,7,7,3.13,1,1,0,0,1,2,76,1,3,3,NA
+66085,7,2,2,5,NA,5,7,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27969.562936,30869.748923,3,91,7,7,1.57,4,4,2,0,0,2,29,2,3,1,3
+66086,7,1,2,80,NA,2,2,NA,NA,NA,2,NA,2,1,4,NA,1,2,NA,2,1,2,1,2,2,NA,NA,NA,NA,10430.111347,0,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+66087,7,2,2,18,NA,2,2,2,18,223,2,NA,2,1,5,12,NA,NA,NA,1,2,2,2,2,1,1,2,2,1,19458.108146,21037.021471,2,91,4,4,0.94,3,3,0,1,0,2,50,2,1,4,NA
+66088,7,2,1,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5662.231921,5835.178913,1,99,5,5,1.13,3,3,1,1,0,2,30,1,1,4,NA
+66089,7,2,1,34,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17802.31438,18635.96967,1,102,3,3,0.92,1,1,0,0,0,1,34,1,4,5,NA
+66090,7,2,2,25,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,NA,NA,NA,NA,50915.06085,51518.363256,3,92,7,7,2.1,3,3,1,1,0,2,25,1,4,5,NA
+66091,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,33092.243754,1,95,3,3,0.76,3,3,0,0,1,1,41,1,2,1,4
+66092,7,2,2,68,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17243.546687,17909.086027,1,92,9,9,3.64,2,2,0,0,2,1,77,1,5,1,4
+66093,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29010.447112,32633.46257,1,100,5,5,2.15,1,1,0,0,0,2,41,1,4,5,NA
+66094,7,2,1,25,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,5,5,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15564.966804,2,101,8,6,2.8,2,1,0,0,0,1,24,2,5,5,NA
+66095,7,2,2,72,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,64368.917314,67563.498341,1,90,8,8,3.3,2,2,0,0,2,1,77,2,2,1,5
+66096,7,2,2,26,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,34906.069211,35828.732314,1,103,1,1,0.03,3,3,0,0,0,1,50,1,2,3,NA
+66097,7,2,2,11,NA,5,7,2,11,134,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16195.492817,16951.263709,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+66098,7,2,1,29,NA,4,4,2,NA,NA,1,1,2,1,3,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15920.644312,16085.139317,1,96,7,7,2.23,3,3,1,0,0,1,29,2,5,1,5
+66099,7,2,1,15,NA,5,6,1,16,192,NA,NA,2,2,2,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10346.302718,11006.15627,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+66100,7,2,2,16,NA,4,4,2,16,203,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,11743.959438,2,95,4,4,0.84,3,3,0,1,0,2,40,1,3,3,NA
+66101,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,48943.054903,67007.111083,2,98,1,1,0.23,2,2,1,0,0,2,20,1,3,6,NA
+66102,7,2,2,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11846.491502,12833.518532,2,101,3,3,0.46,5,5,1,2,0,1,34,1,2,6,NA
+66103,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,101419.325386,103801.228657,1,100,6,6,1.31,3,3,0,0,2,1,65,1,5,1,5
+66104,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,32467.087681,32793.115,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+66105,7,2,1,6,NA,3,3,2,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,68423.584566,71489.623895,1,98,15,15,3.7,5,5,2,1,0,1,34,1,5,1,5
+66106,7,2,2,2,NA,2,2,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7660.112391,8143.030995,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+66107,7,2,1,16,NA,4,4,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13756.125082,1,96,14,14,2.19,7,7,0,2,0,1,39,1,2,1,3
+66108,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,1,1,2,2,1,2,1,3,20111.196953,20296.097622,1,92,4,4,0.76,4,4,0,0,0,2,53,2,1,1,1
+66109,7,2,1,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,15561.574634,17154.493542,1,98,6,6,0.81,6,6,0,4,0,2,34,NA,NA,1,2
+66110,7,2,1,63,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105578.507837,104500.803223,1,99,15,15,5,2,2,0,0,2,1,63,1,5,1,NA
+66111,7,2,2,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,26311.803433,25091.046888,1,97,15,15,5,6,6,0,1,1,2,53,1,4,1,NA
+66112,7,1,2,33,NA,4,4,NA,NA,NA,2,NA,2,2,2,NA,3,5,3,1,2,1,1,2,1,NA,NA,NA,NA,29610.008111,0,2,93,1,1,0.2,2,2,1,0,0,2,33,2,3,5,NA
+66113,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,105141.812429,109679.354704,1,98,15,15,5,4,4,1,1,0,1,40,1,4,1,5
+66114,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,21306.824647,21522.187676,2,98,3,3,0.38,5,5,0,4,0,2,39,1,4,5,NA
+66115,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,202246.928022,1,91,10,10,3.51,3,3,0,0,1,1,21,1,4,5,NA
+66116,7,2,1,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,28542.421068,28714.346279,2,101,5,5,1.36,2,2,0,0,1,1,79,1,4,2,NA
+66117,7,2,2,16,NA,3,3,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,65698.166619,66553.178083,2,95,15,15,5,2,2,0,1,0,2,52,1,5,3,NA
+66118,7,2,1,10,NA,1,1,1,10,126,NA,NA,2,2,2,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13581.60325,14182.420159,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+66119,7,2,1,14,NA,3,3,1,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,59545.101745,58762.542429,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+66120,7,2,2,3,NA,4,4,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8451.169331,8918.740308,2,99,13,13,NA,5,5,2,0,0,2,21,1,3,5,NA
+66121,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,146181.198007,146233.940601,2,91,15,15,5,3,2,0,0,1,1,47,1,5,5,NA
+66122,7,2,1,2,NA,3,3,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47507.757497,52569.646909,1,101,8,8,1.85,5,5,3,0,0,2,31,1,2,1,2
+66123,7,2,2,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,35250.588028,36394.049804,2,95,4,4,0.79,3,3,1,0,0,1,50,1,4,6,NA
+66124,7,2,2,37,NA,1,1,1,NA,NA,2,NA,2,1,5,NA,1,1,2,2,2,2,1,2,2,1,2,2,2,38218.668882,37878.487888,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+66125,7,2,2,46,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,37411.196266,38309.127338,1,102,14,14,4.59,3,3,0,0,1,2,46,1,4,1,1
+66126,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12751.545122,13320.887973,2,101,9,9,4.08,2,2,0,0,2,2,67,1,5,1,3
+66127,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,11982.125462,2,97,7,7,2.65,2,2,0,1,1,2,61,1,4,3,NA
+66128,7,2,2,1,14,2,2,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8230.248767,8808.52375,2,100,15,15,4.26,5,5,1,1,0,1,54,1,4,1,5
+66129,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,97050.018399,96721.703127,1,99,12,12,NA,1,1,0,0,1,2,60,1,5,3,NA
+66130,7,2,2,0,1,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24719.075551,24047.650564,2,91,10,10,3.4,3,3,1,0,0,2,32,1,5,1,5
+66131,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,78018.093799,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+66132,7,2,1,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,29738.952706,30589.306008,2,94,6,3,0.68,3,2,0,0,0,1,26,1,4,5,NA
+66133,7,2,1,40,NA,2,2,1,NA,NA,1,1,2,1,7,NA,3,6,NA,1,2,2,2,2,2,1,2,2,1,39096.402803,46193.238956,2,91,14,7,3.94,3,1,0,1,0,1,40,2,3,6,NA
+66134,7,2,1,4,NA,1,1,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14505.510202,15541.607306,2,94,8,8,2.7,3,3,1,0,0,1,27,1,3,1,4
+66135,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,NA,NA,NA,1,2,2,1,12861.670836,12561.234288,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+66136,7,2,1,27,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,76458.986322,98479.173389,3,90,9,9,3.64,2,2,0,0,0,1,27,2,4,1,5
+66137,7,2,1,20,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,12135.693667,13397.909622,1,90,15,15,5,5,5,0,2,0,1,47,2,5,1,5
+66138,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16579.134475,17201.146893,3,91,15,15,5,4,4,1,1,0,1,39,2,5,1,5
+66139,7,2,1,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,21571.318341,25887.074025,1,102,7,7,1.8,5,4,1,0,2,1,47,1,3,5,NA
+66140,7,2,2,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,11879.290971,12794.969668,2,96,77,77,NA,1,1,0,0,1,2,63,1,3,5,NA
+66141,7,2,1,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,1,2,1,2,2,1,2,2,NA,69210.206708,71835.969778,1,98,14,14,4.96,2,2,0,0,2,1,71,1,5,1,5
+66142,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,27271.751091,27454.884133,2,101,6,2,0.46,3,1,0,0,0,1,21,1,4,5,NA
+66143,7,2,2,2,24,2,2,1,2,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9611.087345,9740.345081,2,93,9,9,2.6,4,4,2,0,0,2,20,1,5,1,3
+66144,7,2,2,4,NA,1,1,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19235.084509,21236.049482,3,92,7,7,1.65,4,4,1,1,0,1,27,1,3,1,3
+66145,7,2,2,5,NA,4,4,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13130.790087,14128.876607,2,102,5,3,0.63,5,4,2,1,0,1,24,1,4,6,NA
+66146,7,2,2,6,NA,3,3,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21910.300386,22800.721264,1,99,3,3,0.88,2,2,0,1,0,2,48,1,1,3,NA
+66147,7,2,2,57,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,2,NA,1,2,1,1,2,1,1,2,2,NA,13697.309651,14098.608838,1,103,2,2,0.53,2,2,0,0,0,1,29,2,4,5,NA
+66148,7,2,1,0,6,4,4,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7434.326058,7971.593122,1,100,8,8,1.7,5,5,2,0,0,2,26,1,3,5,NA
+66149,7,2,1,18,NA,4,4,2,18,227,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,14,14,4.03,4,4,0,1,0,2,40,1,5,1,5
+66150,7,2,2,72,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,1,2,1,2,2,1,2,2,NA,17029.680467,17622.254657,1,101,6,6,1.43,4,4,0,1,2,2,72,1,2,1,NA
+66151,7,2,2,74,NA,4,4,2,NA,NA,2,NA,2,1,9,NA,2,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,16291.303936,17381.577792,1,93,2,2,0.61,2,2,0,0,1,2,45,2,3,5,NA
+66152,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,13093.362366,12735.028157,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+66153,7,2,1,78,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,15176.622228,16064.120023,1,101,4,4,1.11,2,2,0,0,1,1,78,1,4,2,NA
+66154,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,68074.313029,74840.450144,1,101,3,3,1.16,1,1,0,0,1,1,74,1,2,2,NA
+66155,7,2,2,19,NA,4,4,1,19,230,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11078.202838,10933.134393,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+66156,7,2,2,15,NA,4,4,2,15,189,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,6,6,0.86,6,6,0,4,0,2,32,1,4,6,NA
+66157,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,47098.572584,50542.386793,1,95,7,7,2.72,2,2,0,0,2,1,80,1,3,1,3
+66158,7,2,1,11,NA,5,6,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9175.735601,9896.833095,3,91,15,15,5,4,4,1,1,0,1,40,2,5,1,5
+66159,7,2,2,6,NA,5,6,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9007.62445,9504.796896,2,102,8,8,1.72,5,5,0,2,1,1,63,2,5,1,5
+66160,7,2,2,36,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,NA,NA,NA,NA,27627.442402,28606.852107,1,99,14,14,5,2,2,0,0,1,2,64,1,3,3,NA
+66161,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,14979.892428,16276.055879,1,92,3,3,0.98,2,1,0,0,1,1,53,NA,NA,5,NA
+66162,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,2,2,2,2,2,2,2,2,2,2,10614.141896,11057.12801,2,93,15,15,5,2,2,0,0,2,1,60,2,5,1,5
+66163,7,2,1,12,NA,1,1,1,12,147,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18242.832494,18521.680571,1,102,6,6,1.03,6,6,0,4,0,1,34,2,2,1,1
+66164,7,2,2,6,NA,4,4,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8292.876947,8466.609309,1,102,1,1,0,5,5,0,3,0,2,41,1,4,1,4
+66165,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,26847.643051,28699.104328,1,98,5,5,1.05,3,3,1,0,0,1,24,1,3,1,3
+66166,7,2,2,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,8211.623817,2,101,2,2,0.22,4,4,1,1,0,2,41,1,2,4,NA
+66167,7,2,2,23,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,11717.478547,12264.683949,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+66168,7,2,2,8,NA,3,3,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,53931.048037,53769.024148,2,96,7,7,2.38,2,2,0,1,0,2,30,1,4,3,NA
+66169,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105295.506815,106646.104223,2,94,15,15,5,3,3,0,1,1,1,63,1,5,1,3
+66170,7,1,1,2,24,5,7,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25107.382643,0,1,101,13,13,NA,3,3,1,0,0,2,20,1,2,6,NA
+66171,7,2,2,6,NA,4,4,2,6,80,NA,NA,2,1,3,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9235.947079,9578.084112,1,96,9,9,2.78,4,4,0,2,0,1,54,2,5,4,NA
+66172,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,53634.754806,54437.113731,1,98,3,3,0.73,3,3,0,0,0,1,52,1,4,1,3
+66173,7,2,2,32,NA,1,1,1,NA,NA,2,NA,2,1,99,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,39561.667842,40925.159766,2,98,2,2,0.27,4,4,2,1,0,2,32,2,2,5,NA
+66174,7,2,2,4,NA,4,4,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11231.392369,2,100,5,5,0.88,5,5,2,1,0,2,30,1,4,6,NA
+66175,7,2,1,13,NA,3,3,2,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30943.024697,32205.010072,1,98,3,3,0.5,5,5,0,3,0,2,56,1,3,3,NA
+66176,7,2,1,0,5,3,3,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30452.785006,32601.018115,3,92,15,15,5,4,4,2,0,0,1,38,1,5,1,5
+66177,7,2,1,44,NA,4,4,1,NA,NA,2,NA,2,2,1,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,22480.961662,23102.216297,1,98,6,6,1.57,3,3,0,0,2,1,66,2,2,1,4
+66178,7,2,1,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10420.275705,10827.833191,2,96,14,14,3.06,5,5,1,1,1,2,54,1,3,6,NA
+66179,7,2,2,23,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,1,NA,NA,NA,1,2,2,1,16937.04417,17660.28577,1,92,2,2,0.33,5,5,0,1,0,1,51,2,1,4,NA
+66180,7,2,2,15,NA,5,6,2,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,7284.336217,7807.466703,3,90,2,2,0.45,3,3,0,1,0,1,55,2,1,1,3
+66181,7,2,2,11,NA,3,3,2,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58907.362493,59749.51689,1,91,14,14,4.03,4,4,0,2,0,1,52,1,4,1,5
+66182,7,2,1,7,NA,3,3,2,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25019.74954,25881.463934,1,101,3,3,0.44,5,5,0,3,0,1,35,1,3,1,4
+66183,7,2,2,14,NA,4,4,2,14,173,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14143.201945,13910.07431,2,97,2,2,0.33,4,4,2,1,0,2,34,1,2,5,NA
+66184,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25807.688156,25402.018879,1,100,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+66185,7,2,2,2,NA,3,3,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23729.271287,24041.081904,1,94,3,3,0.65,3,3,1,0,0,2,20,1,4,6,NA
+66186,7,2,2,45,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,1,1,2,1,2,2,2,2,40337.933888,43242.784835,2,98,13,13,NA,5,5,0,2,0,1,48,2,1,1,2
+66187,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,26847.643051,28699.104328,1,98,5,3,1.07,4,1,2,0,0,1,24,1,3,6,NA
+66188,7,2,1,7,NA,2,2,1,7,88,NA,NA,2,1,2,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11778.213296,12552.114189,2,93,5,5,0.87,4,4,1,1,0,1,41,2,5,1,3
+66189,7,2,1,47,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,31640.296506,32716.968709,2,94,12,12,NA,4,4,0,2,0,1,47,2,2,1,2
+66190,7,2,2,1,18,5,7,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7280.108168,7886.672866,3,91,14,14,3.17,6,6,1,3,0,1,39,1,4,1,5
+66191,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,24382.996387,23715.691726,1,92,3,3,0.98,2,2,0,0,1,1,70,1,2,1,2
+66192,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,32785.873783,35446.092694,1,94,9,9,3.8,2,2,0,0,2,1,80,1,4,1,4
+66193,7,2,2,25,NA,4,4,2,NA,NA,2,NA,2,2,1,NA,4,1,2,1,2,1,1,2,2,NA,NA,NA,NA,21640.010524,20942.104189,1,93,6,6,1.35,3,3,0,1,0,1,32,2,4,1,4
+66194,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,149804.257477,155051.140087,1,97,15,15,5,3,3,0,1,2,2,63,1,5,1,NA
+66195,7,2,2,9,NA,4,4,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7619.934086,8032.046596,2,103,6,6,1.3,4,4,1,1,0,2,26,1,4,1,3
+66196,7,2,2,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,NA,NA,NA,NA,36053.766709,35796.14047,1,100,5,5,0.85,5,5,0,2,0,2,54,1,2,2,NA
+66197,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,10536.096735,10975.825598,2,93,5,5,1.32,2,2,0,0,2,1,71,2,4,1,1
+66198,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,11762.034222,11824.231183,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+66199,7,2,2,58,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,2,2,2,1,2,2,1,19676.781212,20033.616894,2,103,12,12,NA,3,3,0,0,1,1,60,2,2,1,2
+66200,7,2,1,10,NA,1,1,1,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12577.115885,12876.06354,2,96,7,7,1.79,4,4,0,2,0,1,43,2,3,1,2
+66201,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,159455.333031,1,101,8,8,2.24,4,4,0,1,0,1,45,1,4,1,NA
+66202,7,2,1,61,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,5201.567667,6086.620135,2,92,99,77,NA,7,3,3,3,1,1,61,2,1,1,3
+66203,7,2,1,17,NA,5,7,2,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13748.325141,13775.506705,1,97,15,15,5,6,6,0,1,1,2,53,1,4,1,NA
+66204,7,2,2,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,36053.766709,37031.002154,1,100,14,14,3.47,4,4,2,0,0,1,34,1,5,1,5
+66205,7,2,2,28,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,112991.277498,114001.49655,1,93,14,14,5,1,1,0,0,0,2,28,1,5,3,NA
+66206,7,2,1,62,NA,4,4,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,7973.883342,8036.257566,1,100,9,9,2.88,3,3,0,0,1,1,62,2,5,1,4
+66207,7,2,1,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,7663.797586,8115.960731,1,96,7,7,3.58,1,1,0,0,1,1,80,1,1,2,NA
+66208,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,35958.875702,38183.099815,1,98,8,8,2.62,4,3,0,0,0,1,53,1,2,1,3
+66209,7,2,1,15,NA,4,4,1,16,192,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13731.625553,14242.275028,1,100,14,14,5,3,3,0,1,1,2,44,1,4,5,NA
+66210,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9113.905743,10210.592308,3,90,3,3,0.78,3,3,0,1,2,1,80,2,3,1,2
+66211,7,2,2,8,NA,5,7,1,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7877.267255,8558.823016,2,94,15,15,4.44,5,5,0,1,1,2,74,1,5,2,NA
+66212,7,2,1,8,NA,1,1,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,16747.549238,17012.570163,2,96,4,4,0.81,4,4,1,1,0,1,36,2,1,6,NA
+66213,7,2,1,18,NA,5,7,1,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,93641.872053,93270.118243,2,98,10,10,3.04,4,4,0,0,0,1,55,1,4,1,4
+66214,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,1,2,NA,30548.472436,31485.669458,1,101,7,7,1.83,3,3,0,0,2,1,67,1,1,1,2
+66215,7,2,1,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,28739.023279,1,94,3,3,1.16,2,1,0,0,0,2,36,1,3,5,NA
+66216,7,2,2,8,NA,5,6,1,8,105,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9620.269705,11100.236566,2,91,6,6,1.26,5,5,0,2,0,2,47,2,1,1,1
+66217,7,2,2,51,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,1,1,2,1,2,2,2,2,28701.155283,32172.514581,2,98,7,7,2.25,3,3,0,1,0,2,51,2,1,1,1
+66218,7,2,2,43,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,3,1,2,2,1,2,2,1,2,2,1,34666.955582,39600.756419,1,97,6,6,2.24,2,1,0,0,0,2,43,1,4,3,NA
+66219,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,138322.767578,143283.052066,1,101,7,7,2.64,2,2,0,0,0,1,57,1,2,1,2
+66220,7,2,1,19,NA,3,3,2,20,NA,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,56223.281913,55799.760014,1,99,8,8,2.81,3,3,0,1,0,1,19,1,4,NA,NA
+66221,7,2,2,59,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11446.604914,11655.034159,2,92,3,3,0.7,3,3,0,0,0,1,58,2,1,1,1
+66222,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16382.74577,17320.281883,2,99,7,7,1.63,4,4,0,2,0,1,53,1,3,3,NA
+66223,7,2,2,9,NA,3,3,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,43528.185872,44344.05334,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+66224,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,42992.537371,49508.460769,2,95,2,2,0.54,1,1,0,0,1,2,80,1,3,2,NA
+66225,7,2,2,0,4,4,4,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4439.36229,4707.453058,2,100,14,14,3.06,5,5,1,0,0,1,50,1,5,1,5
+66226,7,1,1,2,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11853.772636,0,2,96,3,3,0.54,4,4,1,1,0,1,29,1,2,1,2
+66227,7,2,1,66,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,6357.471593,6706.752466,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+66228,7,2,1,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,115391.113177,120921.742093,1,102,14,8,4.41,2,1,0,0,0,1,27,1,4,5,NA
+66229,7,2,1,71,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,51668.474366,53628.722404,1,99,15,15,5,2,2,0,0,2,2,68,1,3,1,5
+66230,7,2,2,64,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,3,4,NA,2,2,2,2,2,2,1,2,1,2,13676.984152,14509.761798,2,91,8,8,1.85,5,5,0,2,1,1,39,2,3,1,4
+66231,7,2,1,0,4,1,1,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5323.166653,5657.098649,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+66232,7,2,2,17,NA,5,7,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,15,15,5,6,6,0,3,0,1,47,1,5,1,5
+66233,7,2,2,25,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,16937.04417,18839.574737,1,92,14,14,3.3,4,4,2,0,0,1,28,1,4,1,4
+66234,7,2,1,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108504.032354,111983.725775,1,92,6,6,1.62,3,3,1,0,0,2,26,1,5,1,5
+66235,7,2,2,33,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,17978.142628,18308.053591,1,91,5,5,0.89,4,4,2,0,0,1,39,1,4,1,5
+66236,7,2,1,3,NA,1,1,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20874.345556,20902.551716,3,92,12,12,NA,4,4,2,0,0,1,30,1,3,1,4
+66237,7,2,1,52,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,1,2,2,2,2,1,2,26554.904329,26165.252043,2,93,8,8,2.49,3,3,0,0,0,1,52,2,2,1,4
+66238,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,7869.59899,8538.749518,2,100,6,6,2.39,1,1,0,0,1,2,61,1,2,2,NA
+66239,7,2,1,80,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,14765.705646,15876.477147,3,91,14,14,5,1,1,0,0,1,1,80,1,5,2,NA
+66240,7,2,1,47,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,18194.001026,18423.321012,1,92,14,14,3.69,4,4,0,2,0,1,47,2,4,4,NA
+66241,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,4,4,NA,2,2,2,1,2,2,1,2,2,2,18341.621382,19145.40337,2,90,5,5,1.19,3,3,0,0,0,2,50,2,4,4,NA
+66242,7,2,2,14,NA,1,1,1,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16781.078148,17378.8338,2,103,2,2,0.22,7,7,0,3,0,2,39,2,1,5,NA
+66243,7,2,2,39,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,26465.930618,27183.288271,2,97,15,15,5,4,4,0,2,0,1,47,NA,NA,6,NA
+66244,7,2,2,3,NA,1,1,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13366.393396,14756.857003,2,94,4,4,0.81,4,4,2,0,0,1,26,2,2,1,2
+66245,7,2,1,47,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18094.847125,18029.478224,3,91,10,10,4.42,2,2,0,0,0,1,47,2,2,1,NA
+66246,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,NA,NA,NA,1,2,2,1,152858.509804,163392.947883,1,95,NA,NA,NA,5,5,0,2,0,2,37,1,3,1,NA
+66247,7,2,1,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,9850.66662,10431.854777,2,98,2,2,0.75,1,1,0,0,1,1,70,1,2,2,NA
+66248,7,1,1,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,81508.607355,0,1,91,8,8,2.51,3,3,1,0,0,2,24,1,4,1,3
+66249,7,2,2,27,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18801.993237,20649.846581,2,96,8,6,2.39,2,1,0,0,0,1,26,2,5,5,NA
+66250,7,2,2,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,1,2,2,1,55628.505329,57174.814545,3,92,4,3,0.52,5,4,0,0,0,2,57,1,4,1,2
+66251,7,2,1,33,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,53303.690379,52901.390578,1,100,4,4,0.78,4,4,0,0,1,1,33,2,1,1,1
+66252,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11017.136221,11794.318892,1,101,1,1,0,2,2,1,0,0,2,32,1,3,3,NA
+66253,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,80709.488833,1,101,9,9,2.6,4,4,0,2,0,2,38,1,4,1,4
+66254,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,1,2,NA,2,5,NA,2,2,2,2,2,2,2,2,2,2,10235.0654,10662.230581,2,93,7,7,2.31,2,2,0,0,1,2,40,2,4,5,NA
+66255,7,2,1,12,NA,1,1,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,32326.52031,34735.818434,2,98,13,13,NA,5,5,0,2,0,1,48,2,1,1,2
+66256,7,2,2,49,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,5,4,NA,2,2,2,1,2,2,2,2,2,2,31235.666551,31909.585756,3,92,8,8,2.01,4,4,1,0,0,2,49,2,5,4,NA
+66257,7,2,1,15,NA,4,4,2,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9509.185342,11220.147089,2,99,2,2,0.2,7,7,1,2,1,1,63,1,1,2,NA
+66258,7,2,1,14,NA,5,6,1,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7721.59541,8116.704638,2,96,8,8,2.7,3,3,0,1,0,1,49,2,5,1,5
+66259,7,2,1,45,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,3,16898.996353,17358.625105,1,92,8,8,2.3,4,4,0,1,0,2,41,NA,NA,1,3
+66260,7,2,2,4,NA,5,6,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5373.703942,1,91,10,10,3.78,3,3,1,0,0,1,35,2,5,1,5
+66261,7,2,1,52,NA,3,3,2,NA,NA,2,NA,2,2,7,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,169896.236244,2,91,15,3,0.9,7,1,0,0,1,1,49,NA,NA,5,NA
+66262,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,52209.836905,0,1,98,6,6,1.95,2,2,0,0,2,1,80,1,3,1,3
+66263,7,2,1,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,105401.67423,106753.633419,2,92,14,14,5,1,1,0,0,1,1,65,1,5,5,NA
+66264,7,2,2,64,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,16352.915834,17178.789759,3,92,5,5,1.26,3,3,0,0,1,1,59,2,1,1,1
+66265,7,2,1,16,NA,4,4,2,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11462.850569,11753.30805,2,100,4,4,0.69,5,5,0,3,0,1,38,1,3,6,NA
+66266,7,2,2,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,25052.373156,28677.21758,1,95,9,9,2.3,5,5,2,1,0,1,28,1,3,1,3
+66267,7,2,1,0,9,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17196.879565,17385.815001,2,94,14,14,2.63,6,6,1,3,0,1,39,1,4,1,4
+66268,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,167183.296853,2,91,15,15,5,3,3,0,0,0,2,54,1,4,1,4
+66269,7,2,1,19,NA,3,3,2,19,229,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,88582.269653,92728.90632,1,91,15,15,5,3,3,0,0,0,2,57,1,3,1,3
+66270,7,2,1,76,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12101.489198,12967.543396,1,93,13,13,NA,2,2,0,0,2,1,76,2,5,1,5
+66271,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,14337.499427,15385.847212,1,95,3,3,0.82,2,2,0,0,2,2,80,1,3,1,2
+66272,7,2,2,29,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,4,1,3,1,2,2,1,2,2,1,2,2,1,11408.12687,12544.799584,2,90,5,5,1.08,3,3,0,1,0,2,29,2,4,1,5
+66273,7,2,2,13,NA,2,2,2,13,161,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13824.001771,14551.102227,2,90,8,8,2.7,3,3,0,1,0,2,31,1,5,4,NA
+66274,7,2,1,5,NA,1,1,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11776.305841,11467.607256,1,90,4,4,0.46,7,7,2,3,0,2,34,2,1,6,NA
+66275,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18626.118419,18025.412332,1,93,7,7,1.79,4,4,0,0,0,2,37,2,4,6,NA
+66276,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26595.398371,26621.513872,1,94,3,3,0.79,2,2,0,0,0,1,51,NA,NA,1,4
+66277,7,2,2,19,NA,5,7,1,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9200.381964,9723.240452,2,101,2,1,0,4,1,0,0,0,2,19,1,4,NA,NA
+66278,7,2,2,36,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,6,2,2,2,2,1,2,2,2,2,2,2,26494.281052,25779.409691,3,90,7,7,1.48,5,5,0,1,0,1,43,2,1,6,NA
+66279,7,2,1,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,8074.676044,8165.637965,2,92,12,12,NA,1,1,0,0,1,1,62,1,4,5,NA
+66280,7,2,2,69,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,19159.496319,19898.984475,1,92,77,77,NA,4,4,0,0,2,1,59,2,5,1,5
+66281,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,175997.804296,174522.064022,1,101,7,7,2.31,2,2,0,0,0,1,44,1,3,1,2
+66282,7,2,2,1,16,5,7,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7143.995395,7265.801592,3,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+66283,7,2,2,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,32620.612627,33141.797376,2,98,14,14,3.36,4,4,0,2,0,1,37,1,4,1,4
+66284,7,2,1,12,NA,4,4,1,12,148,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13523.666124,13550.403519,2,98,8,8,2.43,3,3,0,2,0,2,31,1,4,1,NA
+66285,7,2,1,51,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17493.087649,18732.743426,1,97,15,15,4.81,5,5,0,1,1,1,51,2,5,1,5
+66286,7,2,1,11,NA,4,4,1,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8714.559478,8865.734494,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+66287,7,2,1,47,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15341.871992,16429.080873,1,90,15,15,5,5,5,0,2,0,1,47,2,5,1,5
+66288,7,2,2,4,NA,5,7,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7143.995395,7265.801592,3,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+66289,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,118219.690774,2,94,7,7,3.21,1,1,0,0,0,1,41,1,5,5,NA
+66290,7,2,2,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,162373.29329,165026.206299,2,102,9,9,4.08,2,2,0,0,2,1,70,1,5,1,5
+66291,7,2,2,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,26426.249254,26363.744856,1,99,8,8,4.59,1,1,0,0,0,2,33,1,5,5,NA
+66292,7,2,2,9,NA,2,2,1,9,110,NA,NA,2,2,2,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,9897.093392,10755.962815,2,93,6,6,0.93,5,5,1,2,0,1,40,2,4,1,4
+66293,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,106709.949255,125265.093735,1,97,7,7,3.76,1,1,0,0,0,1,33,1,3,5,NA
+66294,7,2,1,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,53149.251154,56449.488341,2,91,12,12,NA,2,2,0,0,2,1,76,1,5,1,2
+66295,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,46927.876098,52201.899738,1,91,7,7,4.02,1,1,0,0,1,2,80,1,3,2,NA
+66296,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,24213.568265,24151.368594,1,99,3,3,1.1,1,1,0,0,0,1,43,1,5,3,NA
+66297,7,2,1,68,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,6358.062034,6381.990744,1,99,7,7,3.03,2,2,0,0,2,1,68,1,3,1,3
+66298,7,2,2,61,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,3,NA,2,2,2,2,2,2,1,2,2,1,8725.210615,9089.36131,2,93,8,8,3.3,2,2,0,0,1,2,61,2,5,3,NA
+66299,7,2,2,22,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,5,2,1,2,1,1,2,1,1,2,2,3,11232.759507,11862.956868,1,103,4,4,0.82,3,3,0,0,0,1,48,2,4,1,1
+66300,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,27738.890335,29287.813784,2,101,3,3,0.59,4,3,0,2,0,1,39,1,1,6,NA
+66301,7,2,1,12,NA,4,4,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,7,7,2.16,3,3,0,1,0,2,44,1,4,6,NA
+66302,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,34922.33705,36456.366148,2,97,5,1,0,2,1,0,0,0,2,45,1,2,4,NA
+66303,7,2,2,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28103.729242,2,101,8,8,2.43,3,3,0,1,0,1,35,1,4,6,NA
+66304,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,114838.671743,120356.919439,1,91,15,15,5,2,1,0,0,0,1,30,1,5,6,NA
+66305,7,2,2,13,NA,4,4,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,12440.215685,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+66306,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,16058.142925,15728.179634,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+66307,7,2,1,44,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,25454.275893,28114.898546,2,102,15,15,5,4,4,0,2,0,1,44,1,3,1,1
+66308,7,2,1,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,32461.799549,34082.06298,1,101,3,3,0.86,2,2,0,0,1,2,80,1,1,2,NA
+66309,7,2,2,30,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,26546.087356,27263.66872,1,102,4,4,1.02,2,2,0,1,0,2,30,1,2,5,NA
+66310,7,2,2,53,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12649.084278,13204.362319,3,90,3,3,0.75,3,3,0,0,0,1,55,2,4,1,3
+66311,7,2,1,39,NA,5,7,1,NA,NA,1,1,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13833.910151,14953.425412,2,103,15,15,5,3,3,0,1,0,2,37,2,5,1,5
+66312,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,28280.669788,29879.976389,1,99,14,14,5,2,2,0,0,2,1,80,1,4,1,4
+66313,7,2,2,1,17,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,15816.39252,3,92,4,4,0.65,4,4,2,0,0,2,20,1,3,5,NA
+66314,7,1,1,2,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13968.423539,0,2,102,3,3,0.7,3,3,1,0,0,1,28,2,2,1,2
+66315,7,2,1,2,NA,1,1,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13281.490378,14230.158403,1,98,8,8,1.95,4,4,1,1,1,2,59,1,3,1,1
+66316,7,2,2,0,8,3,3,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11902.236733,11578.945551,2,97,5,5,1.24,3,3,1,0,0,2,27,1,3,1,3
+66317,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,126435.397679,126594.17356,1,91,10,10,2.77,5,5,0,3,0,1,43,1,5,1,5
+66318,7,2,2,6,NA,4,4,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7746.357421,7953.594257,1,99,6,6,1.34,4,4,0,2,0,1,40,1,4,1,4
+66319,7,1,2,76,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,76025.86855,0,2,98,14,14,5,2,2,0,0,2,1,78,1,5,1,3
+66320,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,14519.343202,14452.61667,2,90,6,6,0.96,5,5,0,1,0,1,55,1,4,6,NA
+66321,7,2,2,71,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,56693.798097,58632.718368,1,103,12,6,2.51,3,1,0,0,1,2,71,1,5,3,NA
+66322,7,1,1,56,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17796.910402,0,2,103,8,8,4.83,1,1,0,0,0,1,56,1,5,5,NA
+66323,7,2,2,49,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,34954.173075,37355.359293,2,98,5,5,1.07,4,4,0,1,0,1,53,2,1,1,1
+66324,7,2,1,1,13,4,4,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5853.073657,6453.970807,2,95,2,2,0.41,3,3,2,0,0,2,19,1,2,NA,NA
+66325,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,18670.751147,19670.682386,1,96,10,10,2.95,4,4,0,1,0,2,34,2,3,1,5
+66326,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16429.782444,20771.252762,3,90,14,14,5,2,2,0,0,0,1,47,1,2,5,NA
+66327,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,2,2,2,2,2,2,2,2,1,2,9404.30514,9554.950099,2,93,15,15,5,2,2,0,0,2,1,60,2,5,1,5
+66328,7,2,2,12,NA,5,6,2,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,15,15,4.81,5,5,0,1,1,1,51,2,5,1,5
+66329,7,2,1,6,NA,5,7,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7482.593572,7851.213843,1,101,4,4,0.78,4,4,1,2,0,2,31,1,4,3,NA
+66330,7,2,2,2,NA,3,3,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,55441.113009,56169.627907,1,92,9,9,3.04,3,3,1,0,0,1,48,1,3,1,4
+66331,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,36067.495928,36452.676481,1,95,2,2,0.83,1,1,0,0,1,2,77,1,4,2,NA
+66332,7,2,1,19,NA,4,4,1,19,238,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14848.504688,14956.646891,1,98,2,1,0.36,2,1,0,0,0,1,19,1,4,NA,NA
+66333,7,2,1,6,NA,2,2,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10738.959181,10632.988891,2,93,9,9,3.14,3,3,0,2,0,2,34,2,5,3,NA
+66334,7,2,1,37,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,1,1,2,2,1,2,2,2,40003.013263,42339.292617,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+66335,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,15262.313583,15991.170182,1,98,5,5,0.74,5,5,0,3,0,1,35,1,2,6,NA
+66336,7,2,1,71,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,43078.255639,45597.38383,1,93,12,12,NA,1,1,0,0,1,1,71,1,5,5,NA
+66337,7,2,2,9,NA,5,6,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3972.173341,4164.719478,1,103,77,77,NA,6,6,0,2,2,1,70,NA,NA,1,1
+66338,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,71034.153987,76359.172033,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+66339,7,2,1,18,NA,1,1,2,18,220,2,NA,2,2,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26704.187335,28443.712885,1,95,9,9,2.46,4,4,0,0,0,1,42,2,2,1,3
+66340,7,2,1,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8143.831412,8871.858164,2,102,10,10,3.04,4,4,2,0,0,2,31,2,2,1,NA
+66341,7,2,1,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,142634.419514,148973.13954,1,100,15,15,4.63,5,5,0,0,0,1,51,1,5,1,3
+66342,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,29733.812317,40471.383048,1,101,1,1,0,2,2,1,0,0,2,32,1,3,3,NA
+66343,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114993.808573,116714.079488,1,98,7,1,0.27,4,1,0,0,0,2,20,1,4,5,NA
+66344,7,2,2,7,NA,4,4,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8610.003992,8918.857984,2,93,6,6,1.15,5,5,3,1,0,1,29,1,3,5,NA
+66345,7,1,2,25,NA,5,6,NA,NA,NA,2,NA,2,2,2,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,16021.911789,0,1,91,9,9,2.97,3,3,1,0,0,1,31,2,5,1,5
+66346,7,2,2,17,NA,5,7,1,17,207,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8111.892084,8425.568811,3,91,15,15,5,4,4,0,2,0,1,38,2,5,1,5
+66347,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,29265.714405,31405.602677,1,95,12,12,NA,3,3,0,0,2,2,56,1,3,5,NA
+66348,7,2,2,13,NA,5,7,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8164.811332,8249.972742,1,93,15,15,5,5,5,1,2,0,2,40,1,5,1,5
+66349,7,2,2,24,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,10550.607376,12189.917123,2,100,6,6,1.62,3,3,1,0,0,1,32,1,3,1,3
+66350,7,2,1,14,NA,1,1,2,14,171,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,19483.203262,19562.510174,2,94,7,7,1.23,6,6,2,1,0,1,33,2,1,6,NA
+66351,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,53612.939104,58717.503545,1,95,15,15,5,2,2,0,0,2,2,80,1,4,1,4
+66352,7,2,2,18,NA,2,2,1,18,224,2,NA,2,2,3,12,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,12970.724558,13432.752316,2,103,5,5,0.65,6,6,1,0,1,2,61,2,1,2,NA
+66353,7,2,1,5,NA,1,1,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,14716.463544,14890.443704,2,96,4,4,0.69,5,5,2,0,0,2,57,2,1,4,NA
+66354,7,2,2,6,NA,1,1,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,20753.369981,21498.66298,1,97,6,3,0.45,7,6,2,1,0,1,29,2,2,1,1
+66355,7,2,1,62,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,1,1,2,2,1,2,1,NA,7576.466116,7992.718813,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+66356,7,1,1,0,1,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8390.245917,0,1,92,7,7,1.83,3,3,1,1,0,2,28,1,3,5,NA
+66357,7,1,1,52,NA,1,1,NA,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,1,1,2,1,NA,NA,NA,NA,26478.915067,0,1,102,99,99,NA,5,5,0,2,1,1,52,2,1,1,1
+66358,7,2,1,64,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,131818.641085,133509.443669,1,102,15,15,5,2,2,0,0,2,2,63,1,4,1,4
+66359,7,2,2,10,NA,3,3,2,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,72263.168327,73296.260667,2,91,15,15,5,3,3,0,2,0,1,44,2,5,3,NA
+66360,7,2,2,6,NA,1,1,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10118.363218,10311.586628,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+66361,7,2,1,4,NA,2,2,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,17306.08847,2,98,6,6,1.07,5,5,3,0,0,2,24,1,3,1,3
+66362,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,36187.441698,36468.041201,1,92,5,5,1.05,3,3,0,0,0,2,55,1,4,4,NA
+66363,7,2,2,4,NA,2,2,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16921.080098,17459.476859,1,91,7,7,1.66,4,4,2,0,0,1,32,2,5,1,4
+66364,7,1,2,2,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10806.348349,0,2,91,6,6,0.93,5,5,1,2,0,2,50,2,1,5,NA
+66365,7,2,1,4,NA,3,3,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23328.816247,26320.448782,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+66366,7,2,2,26,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,1,1,2,2,1,2,2,1,2,2,1,26388.213487,25090.218308,2,96,6,2,0.64,2,1,0,0,0,2,26,1,2,5,NA
+66367,7,2,2,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9399.281543,9696.12347,2,96,6,6,1.32,5,5,1,3,0,2,30,1,4,3,NA
+66368,7,2,1,7,NA,1,1,2,7,91,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10591.186197,11647.778431,1,90,4,4,0.46,7,7,2,3,0,2,34,2,1,6,NA
+66369,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19943.783243,19397.969296,2,95,9,9,5,1,1,0,0,0,2,48,1,4,3,NA
+66370,7,2,1,2,NA,1,1,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12413.685227,13300.367814,1,102,5,5,0.98,4,4,1,1,0,2,42,2,2,6,NA
+66371,7,2,1,77,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,10244.849903,10769.823368,2,99,6,6,1.98,2,2,0,0,2,1,77,1,3,1,3
+66372,7,2,1,5,NA,4,4,2,5,67,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7246.488236,7794.862746,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+66373,7,2,2,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,38724.771087,39491.917359,1,94,3,3,0.86,2,2,0,0,2,1,68,1,4,1,2
+66374,7,2,2,3,NA,1,1,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9468.006743,10197.085534,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+66375,7,2,1,31,NA,3,3,2,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,59510.728426,64240.841943,1,99,77,77,NA,4,4,1,1,0,1,31,2,3,1,3
+66376,7,2,1,10,NA,3,3,2,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23796.980721,24835.204126,1,95,5,5,1.03,4,4,0,2,0,1,33,1,3,1,3
+66377,7,2,1,43,NA,3,3,2,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,20650.777724,21846.156397,2,97,1,1,0.21,4,4,2,0,0,2,34,2,1,1,2
+66378,7,2,2,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,162969.911481,163428.386694,1,101,6,6,2.42,1,1,0,0,0,2,59,1,3,3,NA
+66379,7,2,2,1,22,2,2,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11346.120897,12219.822862,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+66380,7,2,2,0,0,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,29555.41642,28752.625674,3,92,14,14,5,3,3,1,0,0,2,36,1,5,6,NA
+66381,7,2,1,66,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,5854.547456,5900.343568,2,99,3,3,1.24,1,1,0,0,1,1,66,1,4,2,NA
+66382,7,2,2,40,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,40337.933888,43255.958723,3,92,10,10,2.82,4,4,0,1,1,1,36,1,3,1,5
+66383,7,2,2,3,NA,5,6,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6184.163292,6565.044513,1,91,12,2,0.48,7,2,2,2,0,2,54,NA,NA,1,NA
+66384,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,17246.716674,18154.82322,1,99,6,6,1.12,4,4,0,2,0,1,39,1,3,1,3
+66385,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,37453.906829,43130.398676,2,100,3,3,0.65,3,3,0,0,2,2,62,1,3,3,NA
+66386,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,4,4,NA,2,2,2,2,2,2,NA,NA,NA,NA,17465.336559,18230.717255,3,90,10,10,3.04,4,4,0,0,2,2,80,2,1,3,NA
+66387,7,2,2,2,NA,1,1,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,11414.885224,12293.882349,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+66388,7,2,2,7,NA,4,4,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7814.742747,8106.808519,2,95,1,1,0.25,3,3,1,1,0,2,26,1,2,5,NA
+66389,7,2,1,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,145342.530024,151393.302038,1,99,12,12,NA,2,2,0,0,0,1,55,1,5,1,4
+66390,7,2,2,9,NA,1,1,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17053.854294,17778.314301,3,92,5,5,0.95,4,4,0,2,0,2,51,2,5,1,1
+66391,7,2,2,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,122655.643802,126707.897291,3,91,15,15,5,2,2,0,0,0,2,28,1,5,1,5
+66392,7,2,1,14,NA,5,6,2,14,170,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8305.829479,8595.334823,1,93,15,15,5,3,3,0,2,0,2,48,2,5,3,NA
+66393,7,2,2,43,NA,5,7,1,NA,NA,2,NA,2,1,7,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20577.778251,21052.363481,1,92,14,5,2.06,2,1,0,0,0,2,43,2,4,5,NA
+66394,7,2,2,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7051.674614,7594.685042,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+66395,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,12969.776823,15634.514863,2,90,4,4,0.81,3,3,0,1,0,2,41,2,2,6,NA
+66396,7,2,2,46,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,19299.941018,19401.997997,2,102,5,5,1.08,3,3,0,1,0,2,46,2,1,5,NA
+66397,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,14699.320127,14652.436145,2,90,6,6,0.96,5,5,0,1,0,1,55,1,4,6,NA
+66398,7,2,1,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,1,2,1,1,2,1,2,2,NA,17113.447343,18123.138698,2,101,77,77,NA,2,2,0,0,2,1,70,1,2,5,NA
+66399,7,2,1,61,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,6815.198656,6994.794831,3,90,15,15,5,2,2,0,0,2,1,61,1,5,1,5
+66400,7,2,2,11,NA,5,7,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10699.45895,11230.540406,1,97,8,8,2.51,3,3,0,2,0,2,39,2,4,2,NA
+66401,7,2,2,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14414.529053,14783.957167,2,96,1,1,0.06,5,5,2,1,0,1,27,2,3,1,4
+66402,7,2,1,60,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,15,6,2.3,3,1,0,0,2,2,73,1,4,3,NA
+66403,7,2,2,5,NA,1,1,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,17887.570772,19748.358154,1,92,4,4,0.74,4,4,1,1,0,1,51,2,1,1,1
+66404,7,2,2,35,NA,3,3,2,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,59095.849383,59277.549533,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+66405,7,2,1,58,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,15601.50288,15551.741356,2,99,99,3,0.9,7,1,1,0,1,1,60,NA,NA,1,NA
+66406,7,2,1,10,NA,5,6,2,10,120,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7923.925927,8546.646915,1,91,77,77,NA,4,4,0,2,0,1,50,2,5,1,5
+66407,7,2,2,9,NA,2,2,2,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,15148.721588,15796.67129,2,91,2,2,0.22,4,4,0,3,0,2,45,2,5,4,NA
+66408,7,2,2,0,6,4,4,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3756.765605,4035.952682,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+66409,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,31552.004994,1,95,6,6,1.09,5,5,0,3,0,1,31,1,4,1,4
+66410,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,27356.080541,28721.124083,1,101,6,6,1.31,3,3,0,2,0,1,43,1,3,4,NA
+66411,7,2,2,11,NA,4,4,2,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7814.742747,8587.431439,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+66412,7,2,2,41,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,4,1,2,2,2,2,2,2,2,NA,NA,NA,NA,32606.880052,32776.922157,2,93,6,6,0.93,5,5,1,2,0,1,40,2,4,1,4
+66413,7,2,2,4,NA,5,7,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11412.589323,11707.167617,3,91,6,6,1.75,3,3,1,1,1,1,63,1,5,4,NA
+66414,7,1,2,4,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8937.555666,0,2,99,15,15,5,3,3,1,0,0,2,34,1,5,1,5
+66415,7,2,1,1,20,2,2,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,15546.999135,16657.487532,3,92,NA,NA,NA,4,4,1,0,1,1,67,NA,NA,77,NA
+66416,7,2,2,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,131590.908764,1,98,8,8,3.3,2,2,0,0,1,1,50,NA,NA,1,3
+66417,7,2,2,5,NA,4,4,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12819.235788,13388.4749,1,98,5,5,1.26,3,3,1,1,0,2,27,1,5,5,NA
+66418,7,2,2,26,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,1,3,1,2,2,1,2,2,1,2,2,1,16844.740449,17564.040518,1,94,10,10,4.76,2,2,0,0,0,1,58,1,5,1,4
+66419,7,2,2,7,NA,1,1,1,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16664.698857,1,92,6,6,1.12,4,4,0,2,0,1,20,1,2,1,2
+66420,7,2,1,12,NA,5,6,1,12,152,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6747.259224,7022.128717,2,94,15,15,5,5,5,0,2,1,1,47,2,5,1,5
+66421,7,2,1,73,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,12146.092906,13015.339209,1,96,9,9,4.23,2,2,0,0,2,2,71,2,5,1,5
+66422,7,2,2,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,29248.061332,30978.18811,1,102,7,7,1.57,4,4,0,2,0,2,33,1,4,1,4
+66423,7,2,1,7,NA,5,6,2,7,91,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6620.63733,7182.792631,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+66424,7,2,2,14,NA,1,1,1,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18674.868029,19048.995585,2,92,8,8,1.42,7,7,0,4,0,2,37,1,1,6,NA
+66425,7,2,1,79,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9257.537917,9255.717695,1,99,5,5,1.69,2,2,0,0,2,1,79,1,1,1,4
+66426,7,2,2,2,NA,1,1,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10990.75621,2,94,14,14,3.58,4,4,1,0,1,1,80,1,3,2,NA
+66427,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,1,8746.76562,8644.133303,2,97,15,15,5,3,3,0,0,3,2,80,1,3,2,NA
+66428,7,2,2,1,15,5,7,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7483.230909,7475.326886,2,96,15,15,5,3,3,1,0,0,1,39,1,4,1,4
+66429,7,2,2,69,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,2,2,2,2,2,2,1,2,2,2,9716.805546,10308.451947,2,90,2,2,0.73,1,1,0,0,1,2,69,2,4,3,NA
+66430,7,2,1,11,NA,1,1,1,11,142,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11036.458246,10927.552283,1,102,13,13,NA,6,6,1,2,0,2,36,2,4,6,NA
+66431,7,2,2,23,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,85402.868381,88914.492385,2,94,15,6,2.94,2,1,0,0,0,1,29,1,4,6,NA
+66432,7,2,2,9,NA,3,3,2,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17668.551726,18764.602322,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+66433,7,2,2,11,NA,3,3,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16950.724686,16668.200588,2,98,5,5,0.8,5,5,1,3,0,2,37,NA,NA,4,NA
+66434,7,2,2,17,NA,2,2,2,17,210,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17848.732433,18700.272027,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+66435,7,2,1,3,NA,2,2,2,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15372.687224,15393.459347,1,90,6,6,1.68,3,3,1,1,0,2,36,2,4,4,NA
+66436,7,2,2,74,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,1,1,2,2,1,2,1,NA,18266.489157,18835.667339,1,100,5,5,1.18,3,3,0,0,2,2,34,2,5,5,NA
+66437,7,2,1,9,NA,5,7,1,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8439.403196,8726.876152,1,103,3,3,0.37,5,5,1,2,0,2,30,1,4,5,NA
+66438,7,2,1,44,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,4,3,NA,2,2,2,2,2,2,2,2,1,2,35393.002863,35146.477046,2,93,3,3,0.58,4,4,0,1,1,1,65,2,1,3,NA
+66439,7,2,2,48,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,34954.173075,42301.098402,2,98,7,7,1.53,5,5,0,0,0,2,48,1,3,5,NA
+66440,7,2,1,8,NA,2,2,2,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12477.812875,13297.681754,2,94,8,8,2.01,4,4,1,1,0,1,44,2,4,1,4
+66441,7,2,1,74,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,11550.158096,12419.035396,2,94,8,8,1.8,6,6,0,1,2,1,74,2,5,1,5
+66442,7,2,1,48,NA,3,3,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,79677.823556,83116.431703,1,99,15,15,4.47,4,4,0,2,0,2,52,2,5,1,5
+66443,7,2,2,38,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,5,2,2,2,1,2,2,1,2,2,2,2,45655.090694,60280.136423,3,92,4,4,1.12,2,2,0,1,0,2,38,2,1,5,NA
+66444,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,21969.841763,21746.409123,1,96,15,15,5,1,1,0,0,0,2,43,1,5,5,NA
+66445,7,2,2,11,NA,1,1,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17053.854294,17496.484818,2,98,8,8,2.26,4,4,0,1,0,2,43,1,3,1,2
+66446,7,2,1,4,NA,4,4,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7466.078788,7482.321895,1,96,3,2,0.16,7,6,1,4,0,2,32,1,2,5,NA
+66447,7,2,1,27,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16598.645683,2,100,5,5,0.88,5,5,2,1,0,2,30,1,4,6,NA
+66448,7,2,1,44,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,11113.602843,11073.454175,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+66449,7,2,1,62,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,7289.557268,7804.776102,3,90,77,77,NA,2,1,0,0,1,1,62,2,5,3,NA
+66450,7,2,2,7,NA,3,3,2,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,74327.830279,75742.187091,2,91,6,6,1.34,4,4,1,2,0,2,33,1,4,3,NA
+66451,7,2,2,52,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,24902.414229,25152.803056,1,90,9,9,5,1,1,0,0,0,2,52,2,3,5,NA
+66452,7,2,1,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6992.24593,7205.816818,2,97,3,1,0.44,3,1,2,0,0,2,20,1,2,5,NA
+66453,7,2,2,22,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16021.911789,17903.386129,1,91,12,2,0.48,7,2,2,2,0,2,54,NA,NA,1,NA
+66454,7,2,1,13,NA,4,4,1,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19401.033367,19892.636678,2,102,15,15,5,4,4,0,2,0,1,44,1,3,1,1
+66455,7,2,2,2,NA,1,1,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9132.13761,9422.704955,2,103,9,9,2.6,4,4,2,0,0,2,35,1,4,1,3
+66456,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,12842.559946,13702.031196,2,101,7,7,2.64,2,2,0,0,1,2,70,1,3,4,NA
+66457,7,2,1,70,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,16066.082986,17530.388953,1,90,3,3,0.79,2,2,0,0,1,1,70,2,2,1,NA
+66458,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,20137.117895,21220.947807,1,97,3,3,1.16,1,1,0,0,1,2,70,1,4,3,NA
+66459,7,2,2,15,NA,1,1,1,15,191,NA,NA,2,2,4,9,NA,NA,NA,2,1,1,2,2,1,1,2,2,1,13963.420591,15453.171136,2,103,13,13,NA,3,3,0,1,0,1,47,2,1,1,1
+66460,7,2,1,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,83081.99261,85461.503598,2,92,15,15,5,2,2,0,0,0,1,37,1,5,1,5
+66461,7,2,1,0,11,5,7,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6751.596541,7305.608189,2,92,6,6,1.35,3,3,1,0,0,2,32,1,5,1,5
+66462,7,2,1,57,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,28017.200708,28243.226144,2,102,8,8,4.87,1,1,0,0,0,1,57,1,3,3,NA
+66463,7,1,2,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,10683.855206,0,1,96,NA,NA,NA,4,4,1,1,0,2,37,NA,NA,1,4
+66464,7,2,1,21,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37911.437415,38922.114085,2,103,4,4,0.79,3,3,0,0,0,2,42,2,2,5,NA
+66465,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,74517.751389,74746.868777,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+66466,7,2,1,6,NA,1,1,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,2,1,1,1,2,2,NA,NA,NA,NA,14150.136224,14374.05449,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+66467,7,2,2,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,25511.973394,27084.821435,1,98,4,4,0.75,4,4,0,1,0,2,48,1,2,1,3
+66468,7,2,1,9,NA,5,7,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24259.597356,25770.339373,3,91,1,1,0.19,3,3,0,2,0,2,50,1,4,3,NA
+66469,7,2,1,18,NA,3,3,2,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26749.020961,27839.957403,1,101,1,1,0.08,3,3,1,0,0,2,19,1,2,NA,NA
+66470,7,2,2,31,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,3,3,3,1,2,2,1,2,2,1,2,2,1,35353.005268,35220.675513,2,94,7,7,3.21,1,1,0,0,0,2,31,2,3,3,NA
+66471,7,2,1,10,NA,4,4,2,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7886.48532,7943.675078,1,99,10,10,3.13,4,4,0,2,0,1,35,1,4,1,5
+66472,7,2,1,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,113655.661464,115113.492348,1,103,15,15,5,2,2,0,0,1,1,61,1,4,1,5
+66473,7,2,1,56,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,144757.450849,145711.976748,1,101,15,15,5,3,3,0,0,2,1,75,1,2,1,2
+66474,7,2,2,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,33331.144292,34526.406365,1,91,1,1,0.2,2,2,0,0,0,1,44,1,2,1,2
+66475,7,2,2,61,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,12403.522912,12838.398002,1,93,12,12,NA,4,4,0,0,2,1,66,2,4,1,2
+66476,7,2,1,60,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,131445.986898,130104.237054,1,101,9,9,5,1,1,0,0,1,1,60,1,4,3,NA
+66477,7,1,2,32,NA,2,2,NA,NA,NA,2,NA,2,7,77,NA,2,77,3,2,2,2,2,2,2,NA,NA,NA,NA,27127.983961,0,2,90,77,77,NA,2,2,0,1,0,2,32,2,2,77,NA
+66478,7,2,1,14,NA,1,1,1,14,172,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22768.423624,22944.003607,2,98,14,14,2.87,5,5,0,3,0,2,34,1,2,1,2
+66479,7,2,2,2,NA,3,3,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46813.021927,48287.821192,1,91,8,8,2.17,4,4,2,0,0,2,28,1,4,1,5
+66480,7,2,2,13,NA,1,1,1,13,162,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,21818.047789,22992.945413,2,102,7,7,1.79,4,4,0,2,0,1,40,2,2,6,NA
+66481,7,2,1,12,NA,2,2,2,12,152,NA,NA,2,2,2,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14081.782012,15478.382268,2,90,3,3,0.46,5,5,0,2,2,1,75,2,1,1,2
+66482,7,2,2,9,NA,1,1,1,9,109,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,16986.005478,17733.622754,2,102,6,6,1,6,6,1,3,0,1,35,2,3,1,3
+66483,7,1,2,13,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,55,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12531.903464,0,2,100,2,2,0.25,4,4,2,1,0,2,39,1,2,5,NA
+66484,7,2,1,6,NA,3,3,2,6,80,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23796.980721,25112.597396,1,95,5,5,1.19,3,3,1,1,0,1,47,1,2,3,NA
+66485,7,2,2,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,27338.92927,27355.916385,1,96,15,15,5,3,3,0,0,1,2,62,1,4,3,NA
+66486,7,2,1,1,23,2,2,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13002.944731,13931.716845,2,91,4,4,0.76,4,4,1,0,0,2,25,2,4,77,NA
+66487,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,12556.207754,12615.022145,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+66488,7,2,2,10,NA,5,7,2,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9620.269705,10273.007637,2,91,12,12,NA,4,4,0,2,0,1,40,1,5,1,5
+66489,7,2,1,4,NA,3,3,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,81508.607355,89259.872051,1,91,8,8,2.17,4,4,2,0,0,2,28,1,4,1,5
+66490,7,1,2,57,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,42475.69088,0,1,97,9,5,1.97,2,1,0,0,1,1,61,1,5,3,NA
+66491,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,22138.245498,30132.880506,2,93,6,6,2.05,2,2,0,1,0,2,30,1,4,3,NA
+66492,7,2,1,13,NA,2,2,2,13,157,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,1,2,NA,19186.411189,19608.692738,1,90,6,6,0.81,6,6,0,3,0,2,45,1,4,1,2
+66493,7,2,2,19,NA,4,4,2,19,238,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16363.914542,16162.453075,1,101,13,13,NA,3,3,1,0,0,2,19,1,2,NA,NA
+66494,7,2,1,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,32719.762791,34375.255008,1,101,6,6,1.65,2,2,0,0,0,1,47,1,4,1,2
+66495,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,2,NA,2,2,2,2,2,2,2,2,2,NA,18241.877822,20223.807162,2,93,8,8,2.57,3,3,0,0,1,1,59,2,3,1,3
+66496,7,2,2,2,NA,5,7,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7734.994032,7934.647352,2,95,2,2,0.56,2,2,1,0,0,2,22,1,3,5,NA
+66497,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,17794.144581,16918.878285,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+66498,7,2,2,14,NA,3,3,1,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,113907.203714,124196.980527,1,94,8,8,1.67,5,5,1,2,0,1,52,1,4,1,4
+66499,7,2,1,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,114943.792804,135002.90259,2,101,1,1,0.37,1,1,0,0,0,1,21,1,4,5,NA
+66500,7,2,1,15,NA,5,6,2,15,189,NA,NA,2,2,2,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7291.654281,7756.692286,3,91,4,4,0.69,5,5,0,2,0,1,45,2,4,1,1
+66501,7,2,1,0,6,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7237.392208,7237.525607,1,102,6,6,1.43,4,4,2,0,0,1,39,2,3,1,3
+66502,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,25212.845431,2,101,2,2,0.46,1,1,0,0,0,1,20,1,4,5,NA
+66503,7,2,1,50,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16117.991297,17802.733496,1,96,15,15,5,4,4,1,1,0,1,50,1,3,1,4
+66504,7,2,1,70,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,53028.045147,56320.756196,2,94,8,8,3.23,2,2,0,0,1,1,70,1,4,3,NA
+66505,7,2,1,34,NA,1,1,1,NA,NA,2,NA,2,1,5,NA,3,1,NA,2,2,2,1,2,2,1,2,2,2,41155.167164,40844.556107,1,102,5,5,0.92,5,5,0,3,0,2,39,2,3,1,3
+66506,7,2,2,25,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,39833.117645,41470.989238,2,94,4,4,1.16,2,2,0,0,0,1,39,1,2,1,5
+66507,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,53028.045147,62471.592039,2,94,8,8,4.59,1,1,0,0,1,1,74,1,3,3,NA
+66508,7,2,2,13,NA,4,4,2,13,159,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16363.914542,16149.629998,1,95,12,12,NA,2,2,0,1,0,2,41,1,4,5,NA
+66509,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,51433.805604,53192.834838,1,94,9,9,5,1,1,0,0,1,2,70,1,4,2,NA
+66510,7,2,1,11,NA,4,4,1,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10199.928366,10684.945227,1,100,9,9,2.22,5,5,1,2,0,2,40,2,4,1,4
+66511,7,2,2,52,NA,2,2,2,NA,NA,2,NA,2,1,4,NA,4,5,NA,2,2,2,1,2,2,2,2,2,2,27285.659216,28272.553853,1,93,4,4,0.92,3,3,0,0,1,1,60,NA,NA,1,4
+66512,7,2,2,8,NA,1,1,1,8,103,NA,NA,2,2,2,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,21075.336925,21477.798111,1,92,5,5,0.87,4,4,0,2,0,1,42,2,1,1,4
+66513,7,2,1,69,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,2,NA,1,2,1,1,2,1,1,2,1,3,7289.557268,7804.776102,3,90,77,77,NA,4,4,0,0,2,1,69,2,5,2,NA
+66514,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,38589.695298,44311.547931,2,96,8,8,4.59,1,1,0,0,0,2,36,1,5,5,NA
+66515,7,2,2,70,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,14786.399759,15527.475853,1,90,15,15,5,1,1,0,0,1,2,70,2,5,3,NA
+66516,7,2,1,6,NA,1,1,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13927.458372,14007.413517,2,98,14,14,3.91,4,4,1,1,0,1,36,2,3,1,5
+66517,7,2,1,2,NA,4,4,1,2,27,NA,NA,2,1,1,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7366.629832,7382.656579,1,98,10,10,1.89,7,7,3,2,0,1,50,1,5,1,5
+66518,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,10634.832821,11430.282078,2,95,6,6,1.65,2,2,0,0,1,2,80,1,1,2,NA
+66519,7,2,2,9,NA,4,4,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11116.391625,11717.604707,1,92,5,5,0.95,4,4,0,2,0,2,33,1,4,5,NA
+66520,7,2,2,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,21143.964074,22787.585902,1,100,8,8,2.36,3,3,1,0,0,2,37,1,3,4,NA
+66521,7,2,2,8,NA,3,3,2,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15442.305642,15459.941839,2,94,7,7,1.62,5,5,0,3,0,1,30,1,2,1,9
+66522,7,2,2,46,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,154664.071516,157709.344695,3,92,15,15,5,3,3,0,1,0,1,45,1,5,1,5
+66523,7,2,2,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10440.656902,1,99,9,9,5,1,1,0,0,1,2,66,1,5,2,NA
+66524,7,2,1,51,NA,3,3,2,NA,NA,2,NA,2,2,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,22746.832388,23155.578286,1,90,7,7,1.55,5,5,0,3,0,1,51,2,3,1,2
+66525,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,151649.038926,156529.976221,1,101,6,6,1.98,2,2,0,0,1,1,80,1,1,2,NA
+66526,7,2,1,0,7,3,3,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11414.210614,12219.404134,1,93,6,6,1.16,4,4,2,0,0,2,33,1,5,1,4
+66527,7,2,2,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,14534.533756,15040.28544,1,96,14,14,5,2,2,0,0,1,2,74,1,5,5,NA
+66528,7,2,2,13,NA,4,4,2,13,163,NA,NA,2,2,2,6,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,14105.705847,14983.542191,1,93,5,5,0.64,7,7,0,2,1,1,21,2,4,5,NA
+66529,7,2,1,0,0,4,4,2,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5360.999096,5443.282885,2,90,6,6,1.4,3,3,1,1,0,2,33,1,4,5,NA
+66530,7,2,2,3,NA,5,6,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7143.995395,7774.277146,3,91,15,15,5,4,4,2,0,0,2,33,2,5,1,5
+66531,7,2,2,57,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,24217.957803,24014.890408,2,94,15,15,5,2,2,0,0,1,1,63,1,5,1,4
+66532,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17501.079959,17489.07354,1,90,9,9,2.6,4,4,0,1,0,2,49,2,2,1,5
+66533,7,2,1,79,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,17113.447343,17448.661073,2,101,5,5,1.36,2,2,0,0,1,1,79,1,4,2,NA
+66534,7,2,1,3,NA,4,4,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,9588.632035,2,100,1,1,0.06,3,3,1,1,0,2,30,1,4,5,NA
+66535,7,2,1,23,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11743.336654,12063.054146,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+66536,7,2,2,52,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,2,2,2,1,2,2,1,2,2,1,19132.515566,20057.4089,3,90,77,77,NA,2,2,0,0,1,1,60,2,5,1,5
+66537,7,2,1,15,NA,3,3,1,15,182,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39918.077466,39803.689785,1,98,9,9,4.03,2,2,0,1,0,2,49,1,5,3,NA
+66538,7,2,1,69,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,92913.469106,93784.871586,2,100,8,8,3.4,2,2,0,0,2,1,69,1,4,1,4
+66539,7,2,2,9,NA,3,3,2,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,43531.157975,45562.561553,1,91,7,7,2.05,3,3,0,1,0,2,32,1,3,1,NA
+66540,7,2,2,0,4,4,4,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5264.07765,5655.281842,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+66541,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,15783.902141,16840.218806,2,95,5,5,1.45,2,2,0,0,2,1,72,1,3,1,3
+66542,7,2,1,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,8883.099096,8952.58549,2,96,8,8,2.7,3,3,0,0,2,2,52,1,3,1,4
+66543,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18724.594661,18534.166088,2,99,15,15,5,2,2,0,0,0,1,57,1,4,1,5
+66544,7,2,2,37,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,2,1,2,2,1,2,2,1,2,2,1,36053.766709,39664.262913,1,100,8,8,2.36,3,3,1,0,0,2,37,1,3,4,NA
+66545,7,2,2,6,NA,4,4,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7814.742747,8106.808519,2,95,1,1,0.09,5,5,3,1,0,2,31,1,2,1,NA
+66546,7,2,2,2,NA,2,2,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8867.166874,8986.419761,2,90,6,6,1.12,4,4,1,1,0,2,35,2,4,1,3
+66547,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140586.349952,144513.945685,2,91,15,15,5,2,2,0,0,2,1,65,1,5,1,5
+66548,7,1,2,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,68579.013834,0,1,92,4,4,1.29,2,2,1,0,0,2,24,1,4,3,NA
+66549,7,2,2,9,NA,1,1,1,9,117,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14414.529053,14932.182215,2,96,6,6,1.12,4,4,0,3,0,1,26,1,2,77,NA
+66550,7,2,2,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,30932.175051,30363.701754,2,101,6,6,1.9,2,2,0,1,0,2,42,1,5,5,NA
+66551,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,33109.285968,36756.097052,2,98,3,3,0.75,2,2,0,0,0,1,22,1,2,5,NA
+66552,7,2,1,45,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,34825.476578,34314.466393,2,93,6,6,2.69,1,1,0,0,0,1,45,2,4,1,NA
+66553,7,2,1,61,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,2,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,6967.511455,7138.703058,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+66554,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,167711.394252,169284.299948,1,101,3,3,0.73,2,2,0,0,1,1,60,1,3,1,5
+66555,7,2,2,1,16,4,4,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6398.740074,6885.115697,1,93,6,6,0.83,6,6,3,1,0,1,37,NA,NA,1,3
+66556,7,2,1,60,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,10,10,3.51,3,3,0,0,1,1,21,1,4,5,NA
+66557,7,2,1,13,NA,1,1,2,13,159,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21399.234084,21870.218555,1,91,6,6,1.35,3,3,0,2,0,2,38,1,4,3,NA
+66558,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,159097.269198,162641.73095,1,97,15,15,5,3,3,0,0,1,1,64,1,3,1,3
+66559,7,2,1,47,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,38543.316567,38089.298859,1,90,15,15,5,6,6,0,4,0,2,48,1,5,1,5
+66560,7,2,2,5,NA,2,2,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13366.393396,14756.857003,2,94,8,8,2.01,4,4,1,1,0,1,44,2,4,1,4
+66561,7,2,2,42,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,3,2,2,2,2,1,2,2,2,2,2,2,32208.300114,36103.843028,1,102,3,3,0.54,3,3,0,2,0,2,42,2,2,3,NA
+66562,7,2,2,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,97925.559493,98801.080755,1,94,10,10,4.3,2,2,0,0,0,1,27,1,5,1,5
+66563,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,15226.654411,15756.489509,1,96,7,7,2.64,2,2,0,0,2,1,68,NA,NA,1,4
+66564,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,105873.555835,105977.518969,1,98,9,9,2.88,6,3,1,3,0,1,51,1,2,1,3
+66565,7,2,2,66,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,37934.469637,38685.959862,1,94,3,3,0.8,2,2,0,0,1,2,66,1,4,3,NA
+66566,7,2,1,17,NA,5,6,2,17,204,2,NA,2,1,4,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9099.599144,9679.941995,1,90,15,15,5,3,3,0,1,0,1,39,2,5,1,NA
+66567,7,2,2,2,NA,1,1,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7677.363986,8161.370187,2,90,6,6,0.96,5,5,1,1,0,1,39,2,2,1,NA
+66568,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,126314.769628,126413.297707,2,95,15,15,5,2,2,0,1,0,2,52,1,5,3,NA
+66569,7,2,2,62,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,12994.252166,2,90,2,2,0.64,1,1,0,0,1,2,62,2,2,3,NA
+66570,7,2,2,28,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,15665.259727,17204.835995,2,93,77,77,NA,2,2,0,0,0,2,28,2,5,1,5
+66571,7,2,1,13,NA,3,3,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61475.445782,62636.42019,2,92,15,15,5,5,5,0,3,0,2,46,1,5,1,5
+66572,7,1,2,2,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8474.492088,0,1,102,9,9,3.24,3,3,1,0,0,1,40,1,2,1,4
+66573,7,2,1,63,NA,1,1,2,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,9145.939054,9651.644596,1,101,5,5,0.87,4,4,0,0,2,1,63,2,1,1,NA
+66574,7,2,1,6,NA,2,2,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14820.807433,14905.891142,2,102,15,15,4.47,4,4,1,1,0,1,32,1,5,1,4
+66575,7,2,1,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,107347.639721,114331.46987,3,92,15,15,5,4,4,2,0,0,1,38,1,5,1,5
+66576,7,2,1,15,NA,4,4,2,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11252.388648,11670.840662,1,99,6,6,1.35,3,3,1,1,0,2,42,1,4,4,NA
+66577,7,2,2,11,NA,4,4,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7868.372593,8159.849106,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+66578,7,2,2,18,NA,2,2,1,18,223,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16781.078148,17658.768186,2,103,12,12,NA,4,4,0,1,0,2,50,2,3,1,4
+66579,7,2,1,13,NA,5,6,2,13,162,NA,NA,2,1,3,6,NA,NA,NA,1,1,1,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,99,99,NA,4,4,0,1,0,1,40,2,3,6,NA
+66580,7,2,1,1,19,4,4,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7084.261593,7811.556831,1,93,3,3,0.7,3,3,1,0,0,1,23,2,4,1,2
+66581,7,2,2,17,NA,5,6,1,17,209,2,NA,2,1,5,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7534.59005,7613.178065,1,102,15,15,5,3,3,0,1,0,2,49,1,5,1,5
+66582,7,2,1,55,NA,5,6,2,NA,NA,2,NA,2,1,9,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,15215.995952,15234.024078,1,90,10,10,5,1,1,0,0,0,1,55,2,5,2,NA
+66583,7,2,1,64,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9611.684527,9576.156071,1,98,12,6,1.98,3,2,0,0,1,1,64,1,4,1,3
+66584,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,122586.881737,126968.15893,1,90,7,7,1.89,3,3,0,0,1,2,75,1,4,3,NA
+66585,7,2,1,40,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15271.238172,15463.719206,3,91,15,15,5,3,3,1,0,0,1,40,2,5,1,5
+66586,7,2,1,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17265.374984,16862.072044,2,93,8,8,1.67,5,5,1,1,0,2,31,1,4,5,NA
+66587,7,2,1,47,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,36610.126289,37722.018472,1,96,8,8,2.17,4,4,0,0,2,1,80,NA,NA,1,NA
+66588,7,2,2,4,NA,1,1,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,9468.006743,10197.085534,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+66589,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,136372.590634,138491.991585,1,98,8,8,2.97,2,2,0,0,0,1,23,1,3,1,5
+66590,7,2,2,7,NA,5,6,2,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,6347.955472,6765.486426,1,93,8,8,1.2,7,7,1,1,1,1,24,2,2,5,NA
+66591,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,80355.847074,82974.223453,1,98,15,15,3.7,5,5,2,1,0,1,34,1,5,1,5
+66592,7,2,2,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7588.605543,8103.359102,1,99,14,14,4.21,4,4,0,2,0,2,44,1,5,1,5
+66593,7,2,1,13,NA,1,1,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20638.769105,20614.097145,2,92,15,15,3.37,7,7,0,4,0,1,42,2,3,1,1
+66594,7,2,1,61,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,2,2,2,2,2,2,1,2,2,1,9430.93681,9879.499027,2,98,3,3,1.19,1,1,0,0,1,1,61,1,3,3,NA
+66595,7,2,1,16,NA,5,7,2,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11834.781205,12120.23702,2,90,5,5,1.43,2,2,0,1,0,2,38,2,4,5,NA
+66596,7,2,2,7,NA,5,6,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7231.974056,7722.665474,3,91,15,15,5,4,4,1,1,0,1,39,2,5,1,5
+66597,7,2,2,54,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,30236.240945,31437.061964,2,96,7,7,4.04,1,1,0,0,0,2,54,1,2,4,NA
+66598,7,2,1,37,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,69324.74001,85851.040558,2,95,15,15,4.63,5,5,1,2,0,2,36,1,5,1,3
+66599,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,66503.043118,73915.916489,2,98,7,7,1.61,4,4,1,1,0,1,43,NA,NA,6,NA
+66600,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10325.56467,10786.590089,1,96,14,14,5,3,2,0,0,1,2,64,1,4,3,NA
+66601,7,2,2,1,20,5,6,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3600.631978,3772.844478,3,90,10,10,2.41,5,5,1,2,0,1,44,2,4,1,5
+66602,7,2,2,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,105873.555835,110835.841786,1,98,15,15,4.34,4,4,1,1,0,1,41,1,5,1,5
+66603,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,2,1,2,2,1,2,2,1,2,2,1,24919.497762,30093.052062,1,98,3,3,0.5,5,5,0,3,0,2,56,1,3,3,NA
+66604,7,2,2,1,19,5,7,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8892.555701,9937.38353,1,98,7,7,1.52,4,4,2,0,0,1,30,1,3,1,4
+66605,7,2,2,11,NA,4,4,2,11,142,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9122.654131,9410.759793,1,96,15,15,5,2,2,0,1,0,2,47,1,5,3,NA
+66606,7,2,2,0,5,3,3,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20261.925369,19711.566478,2,103,15,15,5,3,3,1,0,0,1,31,1,5,1,5
+66607,7,1,1,67,NA,2,2,NA,NA,NA,2,NA,2,2,8,NA,3,1,NA,1,2,2,2,2,2,NA,NA,NA,NA,7379.175826,0,1,96,14,14,5,2,2,0,0,1,1,67,2,3,1,NA
+66608,7,2,2,18,NA,5,6,2,18,219,2,NA,2,2,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8501.305782,9245.625446,1,93,7,7,1.64,5,5,0,2,0,1,47,2,5,1,1
+66609,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,26792.151435,27749.707744,1,97,3,3,0.73,3,3,0,0,0,2,50,1,4,1,3
+66610,7,2,2,3,NA,5,6,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7483.230909,7475.326886,2,96,14,14,4.32,3,3,1,0,0,2,33,2,5,1,5
+66611,7,2,2,15,NA,2,2,1,15,189,NA,NA,2,2,3,10,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,18556.092615,19088.608674,1,103,6,6,0.93,5,5,0,1,0,1,39,2,3,1,3
+66612,7,2,1,64,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10739.724778,10911.761454,2,99,15,15,5,2,2,0,0,2,1,64,2,5,1,4
+66613,7,2,1,9,NA,5,6,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9928.619925,10708.884665,1,91,12,12,NA,4,4,0,2,0,1,43,2,5,1,5
+66614,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,126334.218747,128844.689973,1,95,9,9,4.08,2,2,0,0,0,1,51,NA,NA,1,3
+66615,7,2,2,16,NA,4,4,2,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14085.742306,14155.614592,1,96,8,8,2.17,4,4,1,1,0,2,41,1,3,1,3
+66616,7,2,1,5,NA,1,1,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,10274.893484,10739.35317,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+66617,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12754.235811,13157.031697,1,97,7,7,1.74,4,4,0,3,0,2,32,1,4,5,NA
+66618,7,2,2,11,NA,5,6,2,12,145,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10699.45895,11230.540406,1,97,14,14,2.87,5,5,0,3,0,2,40,2,5,1,5
+66619,7,2,2,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,72551.269339,74475.92188,1,95,6,6,2.04,2,2,0,0,2,2,72,1,3,1,NA
+66620,7,2,1,64,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,157269.814945,159287.072954,1,92,15,15,5,3,3,0,0,2,2,61,1,5,1,5
+66621,7,2,1,0,2,3,3,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9570.577309,9932.368407,1,95,1,1,0.1,5,5,2,1,0,1,35,1,9,1,3
+66622,7,2,1,9,NA,5,7,1,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41342.668304,42524.849071,1,102,8,8,1.91,5,5,1,2,0,2,38,1,5,1,4
+66623,7,2,1,62,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,7289.557268,7804.776102,3,90,12,12,NA,4,4,0,0,1,1,62,2,4,3,NA
+66624,7,2,2,6,NA,4,4,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8504.389189,8682.552645,2,93,6,6,1.72,2,2,0,1,0,2,29,1,4,3,NA
+66625,7,2,1,4,NA,5,6,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7812.10492,8436.863602,1,95,4,4,0.62,5,5,2,0,2,2,29,2,3,5,NA
+66626,7,2,1,6,NA,5,6,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8095.170809,8542.249827,1,100,15,15,5,3,3,0,1,0,1,38,1,5,1,5
+66627,7,2,1,17,NA,4,4,2,17,206,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11023.237662,11433.168046,1,99,6,6,0.96,5,5,1,2,0,2,35,1,4,1,2
+66628,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,133853.800452,133985.238945,3,91,8,8,3.4,2,2,0,0,0,1,33,2,5,1,4
+66629,7,1,2,0,11,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,5263.118333,0,3,90,4,4,0.74,4,4,1,1,0,1,32,2,4,1,2
+66630,7,2,1,58,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,14897.510053,15360.906039,1,101,14,14,5,2,2,0,0,0,1,58,2,3,1,1
+66631,7,2,2,69,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10604.379638,1,99,3,3,1.01,1,1,0,0,1,2,69,1,2,3,NA
+66632,7,2,1,58,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,181786.280703,1,95,8,8,3.47,2,2,0,0,0,1,58,1,2,1,4
+66633,7,2,1,41,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19564.598892,20184.074944,3,92,15,15,5,3,3,1,0,0,1,41,2,5,1,3
+66634,7,2,1,9,NA,5,6,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10810.913614,11522.32071,1,97,15,15,4.07,5,5,0,3,0,1,42,2,5,1,5
+66635,7,2,1,78,NA,5,6,2,NA,NA,2,NA,2,2,8,NA,5,1,NA,1,2,1,1,2,1,1,2,2,NA,8637.841003,9256.016034,1,96,9,9,3.97,2,2,0,0,1,1,78,2,5,1,5
+66636,7,2,2,19,NA,4,4,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16537.460749,16413.673958,2,100,8,8,2.67,3,3,0,0,1,1,61,1,3,1,4
+66637,7,2,2,1,16,4,4,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,7498.263904,2,100,3,3,0.39,7,7,3,3,0,2,30,1,2,5,NA
+66638,7,2,2,9,NA,1,1,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15225.935813,15598.502882,2,94,77,77,NA,5,5,1,1,0,1,41,2,2,1,2
+66639,7,2,1,10,NA,1,1,1,10,130,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,8828.580268,9491.612368,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+66640,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,85420.170155,90189.982607,1,91,7,7,2.05,3,3,0,1,0,2,32,1,3,1,NA
+66641,7,2,2,46,NA,3,3,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21224.321717,21245.16306,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+66642,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,126717.185106,132638.866244,2,91,14,14,4.19,3,3,0,1,0,2,31,1,4,1,3
+66643,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,15875.759889,17102.194972,1,96,6,6,1.34,3,3,1,0,0,2,42,2,4,6,NA
+66644,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,32647.160346,32536.716665,1,99,4,4,1.16,2,2,0,0,2,2,63,1,5,6,NA
+66645,7,2,1,13,NA,5,6,2,13,157,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,13,13,NA,3,3,0,2,0,2,41,2,3,4,NA
+66646,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,81416.374938,85457.008484,1,97,15,15,5,1,1,0,0,1,2,70,1,4,2,NA
+66647,7,2,1,19,NA,3,3,1,19,229,2,NA,2,1,99,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,110478.18082,109645.964567,2,101,3,3,1.15,2,1,0,0,0,1,20,2,4,5,NA
+66648,7,2,2,7,NA,4,4,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10654.944111,10939.993817,1,100,4,4,0.99,2,2,0,1,0,1,36,2,4,4,NA
+66649,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15096.339697,2,100,7,7,1.34,5,5,0,2,0,2,53,1,4,4,NA
+66650,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16088.355002,16260.034828,1,96,1,1,0.2,2,2,0,0,0,1,25,1,2,5,NA
+66651,7,2,1,10,NA,2,2,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19735.224235,20204.314213,2,102,8,8,1.91,5,5,0,3,0,1,39,1,3,1,3
+66652,7,2,1,17,NA,4,4,2,17,215,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9313.795042,9381.627752,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+66653,7,2,1,42,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,19436.026093,20405.866121,2,97,5,5,1.08,3,3,0,1,0,2,45,1,4,6,NA
+66654,7,2,1,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,28542.421068,28714.346279,2,101,5,5,1.5,2,2,0,0,0,1,47,1,4,3,NA
+66655,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,3,6,2,1,2,2,1,2,2,NA,NA,NA,NA,11608.998717,11632.703325,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+66656,7,1,2,41,NA,2,2,NA,NA,NA,2,NA,2,1,4,NA,2,3,3,2,2,2,1,2,2,NA,NA,NA,NA,34999.007145,0,1,90,1,1,0.32,2,2,0,1,0,2,41,2,2,3,NA
+66657,7,2,2,16,NA,3,3,2,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,58352.457563,59447.870488,1,99,8,8,2.81,3,3,0,1,0,1,19,1,4,NA,NA
+66658,7,2,2,14,NA,5,6,2,14,171,NA,NA,2,1,4,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10963.340882,11440.408955,2,91,15,15,5,4,4,0,2,1,2,56,1,5,1,5
+66659,7,2,1,7,NA,1,1,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18107.947773,18211.90239,1,97,10,10,2.32,6,6,0,4,0,1,42,1,4,1,4
+66660,7,2,1,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18289.793332,19167.642325,1,99,2,2,0.31,4,4,1,0,1,2,67,1,3,3,NA
+66661,7,2,1,68,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20250.944593,20440.871021,2,95,6,6,1.36,3,3,0,0,2,2,60,1,5,1,4
+66662,7,2,1,36,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,18955.147124,22251.142446,1,90,2,2,0.63,2,2,0,0,1,1,36,1,4,5,NA
+66663,7,2,1,62,NA,5,6,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9355.567184,9645.104203,2,97,15,15,5,2,2,0,0,1,2,58,2,5,1,5
+66664,7,2,2,0,8,4,4,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4439.36229,4769.277094,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+66665,7,2,2,27,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,116919.871791,121727.422593,2,92,10,8,4.59,2,1,0,0,0,2,27,2,5,1,NA
+66666,7,2,1,1,17,3,3,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,28617.223132,30780.732491,1,101,3,3,0.63,3,3,1,0,0,2,47,1,1,3,NA
+66667,7,2,2,11,NA,3,3,2,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,80369.555824,80128.103294,1,97,15,15,5,4,4,1,1,0,2,33,1,5,1,3
+66668,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,99831.393624,101809.075589,1,100,15,15,5,2,2,0,0,2,1,79,1,5,1,4
+66669,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,137143.403965,141946.841153,2,91,15,15,5,2,2,0,0,2,1,65,1,5,1,5
+66670,7,2,1,53,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,1,1,2,2,1,2,1,3,13537.092442,13488.188749,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+66671,7,2,1,46,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,28726.575428,29019.719154,2,100,14,14,3.58,4,4,0,1,0,1,46,2,5,1,5
+66672,7,2,1,60,NA,1,1,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,2,2,2,2,8961.035147,9135.681188,2,94,15,15,5,2,2,0,0,1,1,60,1,5,1,5
+66673,7,2,2,25,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,50915.06085,50887.683329,3,92,9,9,2.22,5,5,1,0,2,1,66,2,1,1,1
+66674,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,35334.703093,40990.264786,1,101,3,3,0.9,1,1,0,0,1,2,80,1,1,3,NA
+66675,7,2,1,4,NA,2,2,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,18754.85406,20094.472571,2,102,6,6,1.12,4,4,1,1,0,1,38,2,2,1,3
+66676,7,2,2,10,NA,4,4,1,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,1,2,1,2,2,1,8362.256577,9003.967662,2,100,5,1,0,3,1,0,1,0,2,50,1,2,5,NA
+66677,7,2,2,18,NA,2,2,2,18,219,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,18368.872199,19023.186366,2,91,10,10,2.95,4,4,0,1,0,2,18,1,3,NA,NA
+66678,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,137038.746155,146586.432966,2,101,4,3,0.92,2,1,0,0,0,1,21,1,4,5,NA
+66679,7,2,1,30,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21280.199633,23174.199432,1,97,15,15,4.84,6,6,2,0,0,1,53,NA,NA,1,NA
+66680,7,2,2,56,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15892.013862,19604.056525,3,90,12,12,NA,2,2,0,0,0,2,56,2,4,1,5
+66681,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,1,1,2,2,1,2,2,1,2,2,1,26999.643202,25671.572746,2,102,5,3,0.63,5,4,2,1,0,1,24,1,4,6,NA
+66682,7,2,2,6,NA,2,2,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14580.994397,14937.78024,2,93,4,4,0.97,3,3,0,1,0,1,38,2,3,1,3
+66683,7,2,1,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,NA,27609.8026,29259.766266,2,101,10,10,2.91,4,4,0,1,0,2,51,1,2,5,NA
+66684,7,2,1,16,NA,5,6,1,16,193,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7669.677276,8194.5967,2,92,15,15,4.59,4,4,0,2,0,2,48,1,5,1,5
+66685,7,2,1,29,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,33592.259589,37970.446859,1,103,13,13,NA,4,4,2,0,0,2,27,2,2,6,NA
+66686,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,130229.183917,130070.387937,2,103,77,77,NA,2,1,0,0,0,1,50,1,5,5,NA
+66687,7,2,2,53,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,22969.116046,23497.145655,2,93,10,10,5,1,1,0,0,0,2,53,2,5,5,NA
+66688,7,2,2,19,NA,5,7,1,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,15318.952876,15275.788411,1,98,2,1,0.18,4,1,0,0,0,2,20,NA,NA,5,NA
+66689,7,1,2,71,NA,1,1,NA,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,22927.30561,0,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+66690,7,2,1,65,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,14067.170863,14844.985608,2,102,4,4,0.67,4,4,0,0,2,2,20,1,1,NA,NA
+66691,7,2,1,1,20,4,4,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,10655.495346,2,97,7,7,1.72,5,5,1,2,0,1,32,1,4,1,4
+66692,7,2,2,1,17,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11582.174418,12474.053558,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+66693,7,2,1,18,NA,5,6,2,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11506.937395,12095.740214,1,97,9,9,1.78,6,6,0,1,1,1,45,2,3,1,3
+66694,7,2,1,1,13,2,2,2,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10487.35151,11236.440262,2,95,14,14,4.45,3,3,1,0,0,2,29,1,5,1,5
+66695,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,21102.441005,21214.570635,1,98,2,2,0.67,1,1,0,0,1,1,60,1,3,3,NA
+66696,7,2,1,11,NA,3,3,2,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,19926.440922,21167.339982,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+66697,7,2,2,13,NA,3,3,1,13,161,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61800.07471,62604.355485,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+66698,7,2,1,21,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,100807.076657,102674.431164,1,97,15,15,5,4,4,0,0,1,1,60,1,5,1,5
+66699,7,2,1,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105412.227726,111169.022023,2,101,7,7,1.88,4,4,0,2,0,2,36,1,4,1,5
+66700,7,2,1,10,NA,4,4,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7005.767895,7244.406666,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+66701,7,2,1,75,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,13928.293734,14290.365569,2,92,5,5,1.36,2,2,0,0,2,1,75,2,1,1,1
+66702,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,18533.049642,19445.722605,1,99,13,13,NA,4,4,1,0,0,2,26,1,4,4,NA
+66703,7,2,1,2,NA,5,7,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8375.75457,9394.325143,2,95,15,15,5,3,3,1,0,0,1,50,1,5,1,NA
+66704,7,1,2,28,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,NA,NA,NA,NA,15967.106149,0,2,103,1,1,0.04,2,2,0,1,0,2,28,1,4,5,NA
+66705,7,2,2,3,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9727.363166,10265.541082,2,97,3,3,0.4,6,6,2,3,0,2,25,1,2,5,NA
+66706,7,2,1,8,NA,5,6,2,8,102,NA,NA,2,1,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7923.925927,8546.646915,1,94,7,7,1.79,4,4,0,1,0,1,59,2,4,1,4
+66707,7,2,1,13,NA,1,1,1,13,161,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24902.864049,25040.491572,1,94,5,5,0.94,4,4,0,2,0,2,37,2,3,1,2
+66708,7,2,2,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20842.960361,20272.538073,3,91,12,12,NA,2,2,0,0,0,1,52,1,5,1,5
+66709,7,1,1,57,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19478.845078,0,2,95,2,2,0.46,3,3,0,0,0,2,48,1,2,1,2
+66710,7,2,1,66,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9082.311855,9548.677283,1,96,7,7,1.83,3,3,0,0,1,1,66,2,5,1,3
+66711,7,1,2,11,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5885.567617,0,1,93,9,9,1.77,7,7,0,2,0,2,56,NA,NA,5,NA
+66712,7,2,1,31,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14221.330587,14963.406508,3,90,12,12,NA,4,4,0,0,1,1,62,2,4,3,NA
+66713,7,2,2,60,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,35354.773526,36055.159401,2,103,5,5,1.2,3,3,0,0,2,1,66,2,2,1,2
+66714,7,2,2,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,54951.692454,55140.36098,2,102,15,15,3.92,5,5,0,0,0,1,19,1,4,NA,NA
+66715,7,2,1,64,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,9535.353518,9609.942052,2,101,9,9,4.08,2,2,0,0,2,2,67,1,5,1,3
+66716,7,2,1,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21317.283165,21013.422419,2,96,4,4,0.65,5,5,0,3,0,1,30,1,4,1,2
+66717,7,2,2,23,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,6,3,1,2,2,1,2,2,1,2,2,1,35710.33222,42634.350272,1,90,3,3,0.43,4,4,2,0,0,1,31,1,3,6,NA
+66718,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,36950.912612,36817.49814,1,97,3,3,1.16,1,1,0,0,0,2,56,1,3,3,NA
+66719,7,2,2,9,NA,3,3,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55469.656717,57246.428971,1,101,9,9,2.6,4,4,0,2,0,2,38,1,4,1,4
+66720,7,2,1,58,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15044.515884,14996.530888,3,90,15,15,4.89,5,5,0,0,0,2,57,2,3,1,3
+66721,7,2,2,47,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12977.791943,13644.504126,2,100,15,15,5,3,3,0,1,0,1,48,2,5,1,5
+66722,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,110070.015649,114078.798472,3,91,15,15,5,4,4,1,1,0,2,41,1,5,1,5
+66723,7,2,1,14,NA,4,4,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,18260.901254,2,101,6,6,0.96,5,5,0,4,0,2,36,1,4,4,NA
+66724,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23630.941535,1,95,6,6,1.35,3,3,1,0,0,1,31,1,5,1,5
+66725,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,124263.643735,125259.229556,2,95,8,8,3.53,2,2,0,0,0,1,57,1,4,1,4
+66726,7,2,1,18,NA,5,7,2,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21852.102821,22875.024578,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+66727,7,2,1,11,NA,5,7,2,11,135,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8246.426933,9050.014459,1,99,10,10,3.51,3,3,0,1,0,2,44,1,3,1,5
+66728,7,2,2,71,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,1,1,2,1,NA,13750.49328,15781.175386,2,92,4,4,1.14,2,2,0,0,2,1,72,2,3,1,3
+66729,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,137143.403965,141946.841153,2,91,15,15,5,2,2,0,0,2,1,68,1,5,1,5
+66730,7,2,1,9,NA,5,7,1,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11661.909323,12350.149476,1,92,15,15,5,4,4,0,2,0,1,41,2,5,1,5
+66731,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,117009.557742,134978.408148,2,99,14,14,5,1,1,0,0,0,1,46,1,5,5,NA
+66732,7,2,1,16,NA,5,6,1,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9099.087557,9564.682199,2,102,12,12,NA,3,3,0,1,0,1,57,2,5,1,5
+66733,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20131.904783,20903.590301,1,92,15,15,5,2,2,0,0,2,2,80,1,5,1,NA
+66734,7,2,2,16,NA,5,6,1,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,5760.953091,5914.685125,2,92,8,8,1.91,5,5,0,2,1,2,47,2,1,1,3
+66735,7,2,1,24,NA,5,7,1,NA,NA,2,NA,2,1,4,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,27271.751091,27454.884133,2,101,6,4,1.74,3,1,0,0,0,1,21,1,4,5,NA
+66736,7,2,2,55,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16178.54595,16391.328992,1,93,15,15,5,3,3,0,0,1,1,63,1,5,1,5
+66737,7,2,2,14,NA,2,2,2,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,15087.58237,15620.049499,1,96,15,15,5,4,4,0,2,0,1,36,2,3,1,4
+66738,7,2,1,59,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,27240.276328,26969.79863,1,90,15,15,5,2,2,0,0,1,2,63,1,5,1,5
+66739,7,2,1,9,NA,5,6,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5438.768263,5891.907372,2,92,15,15,5,4,4,0,2,0,2,41,1,5,1,5
+66740,7,2,2,0,7,3,3,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20899.681083,21326.972731,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+66741,7,2,2,73,NA,3,3,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,19901.857177,20582.498016,2,94,1,1,0.08,1,1,0,0,1,2,73,2,4,3,NA
+66742,7,2,1,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7431.820906,7738.904727,1,99,7,7,1.53,5,5,2,0,0,2,37,1,4,1,3
+66743,7,2,2,68,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,1,4,NA,2,2,2,2,2,2,2,2,1,2,8725.210615,9440.710379,2,93,4,4,0.94,3,3,0,1,1,2,68,2,1,4,NA
+66744,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,10479.637868,11727.843054,2,98,77,77,NA,1,1,0,0,1,1,80,1,2,2,NA
+66745,7,2,2,16,NA,1,1,1,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26325.414456,27702.295836,3,92,10,10,3.77,3,3,0,1,0,2,52,1,4,6,NA
+66746,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,97925.559493,100647.602701,1,94,6,6,1.31,3,3,0,0,0,2,46,1,5,6,NA
+66747,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,157587.268357,1,94,9,9,3.97,2,2,0,0,0,1,59,1,3,5,NA
+66748,7,2,1,73,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,9662.124837,9885.036833,2,92,3,3,1.01,1,1,0,0,1,1,73,1,4,3,NA
+66749,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,16741.034883,16841.712651,2,95,2,2,0.7,1,1,0,0,0,2,56,1,3,4,NA
+66750,7,2,2,73,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,3,NA,2,2,2,1,2,2,2,2,2,NA,20922.745102,23195.943223,1,93,15,15,2.96,7,7,0,1,1,2,18,1,2,NA,NA
+66751,7,2,2,9,NA,5,6,2,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,7722.982971,8635.133042,1,93,8,8,2.24,4,4,0,2,0,1,44,2,5,1,4
+66752,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21491.090123,20964.324934,1,97,14,14,5,3,3,0,0,0,2,51,1,5,1,4
+66753,7,2,2,2,NA,3,3,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20820.221848,22126.060024,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+66754,7,2,1,47,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,19371.546331,19651.186711,2,95,12,12,NA,2,2,0,1,0,1,47,1,4,2,NA
+66755,7,2,2,33,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,6,3,2,2,2,2,2,2,1,2,2,2,27127.983961,26396.013964,2,90,6,6,0.96,5,5,1,1,0,1,39,2,2,1,NA
+66756,7,1,1,11,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11567.189008,0,2,90,14,14,3.06,5,5,1,2,0,1,42,1,4,1,5
+66757,7,1,2,59,NA,5,6,NA,NA,NA,2,NA,2,1,99,NA,5,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12649.084278,0,3,90,77,77,NA,2,2,0,0,1,1,65,2,3,1,5
+66758,7,2,2,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,60071.993203,65791.533402,1,95,99,99,NA,1,1,0,0,1,2,80,1,4,2,NA
+66759,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,12449.932013,12144.773422,3,90,7,7,2.23,3,3,0,0,0,2,51,1,4,3,NA
+66760,7,1,2,32,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,1,3,1,2,2,1,2,2,NA,NA,NA,NA,25241.487585,0,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+66761,7,2,1,2,NA,2,2,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10057.661774,10071.252045,2,93,12,77,NA,3,1,1,1,0,2,43,1,5,3,NA
+66762,7,2,1,16,NA,4,4,2,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11252.388648,11221.862262,1,99,7,7,1.89,3,3,0,1,0,1,50,1,5,1,2
+66763,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,54095.581484,61529.339683,2,91,7,7,1.61,4,4,0,0,3,1,65,1,3,6,NA
+66764,7,2,1,2,NA,1,1,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12005.116852,11690.420322,3,92,14,14,2.29,7,7,2,0,0,2,50,2,1,1,9
+66765,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,1,2,1,1,2,NA,NA,NA,NA,40859.270352,45390.978638,2,101,8,8,4.22,1,1,0,0,1,1,80,1,3,2,NA
+66766,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,94698.084211,103561.845332,1,101,14,14,4.5,3,3,0,1,0,1,39,1,2,1,5
+66767,7,2,2,1,21,1,1,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12260.86913,13033.834526,3,92,5,5,0.81,5,5,3,0,0,2,23,1,4,5,NA
+66768,7,2,2,7,NA,4,4,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8282.497467,8610.715242,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+66769,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18101.423817,17611.874146,2,99,9,9,2.43,4,4,0,2,0,2,49,1,3,3,NA
+66770,7,2,2,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,59974.233027,60510.443557,2,97,6,6,2.85,2,1,0,0,0,1,22,1,3,6,NA
+66771,7,2,2,31,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,37439.743351,39040.415193,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+66772,7,2,1,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7907.070371,7770.374311,2,95,2,2,0.26,3,3,1,0,0,2,54,1,3,2,NA
+66773,7,1,1,24,NA,4,4,NA,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,16052.583048,0,2,99,77,77,NA,3,3,0,1,0,2,46,NA,NA,77,NA
+66774,7,2,2,17,NA,1,1,1,18,216,2,NA,2,2,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26325.414456,26852.811114,3,92,13,13,NA,4,4,0,2,0,2,50,2,1,5,NA
+66775,7,2,1,7,NA,1,1,2,7,95,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15039.041447,16539.35823,3,91,7,7,1.23,6,6,2,2,0,1,36,2,1,1,1
+66776,7,2,2,2,NA,4,4,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,8250.113235,2,100,14,14,4.59,3,3,1,0,0,1,30,NA,NA,1,4
+66777,7,2,1,39,NA,2,2,1,NA,NA,2,NA,2,1,3,NA,1,1,NA,2,2,2,1,2,2,2,2,1,2,37631.514869,37347.497931,2,93,3,3,0.43,4,4,0,0,0,1,45,2,2,6,NA
+66778,7,2,1,11,NA,4,4,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14125.862146,14370.909235,1,97,15,15,5,4,4,0,1,0,1,40,1,4,1,4
+66779,7,2,1,14,NA,5,6,1,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,1,1,2,2,1,6121.087833,6878.034577,2,92,3,3,0.4,6,6,0,1,2,1,78,2,1,1,1
+66780,7,2,1,0,2,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,5588.585837,5533.355248,1,102,6,6,0.8,7,7,3,3,0,2,34,2,3,1,1
+66781,7,2,1,68,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11243.205109,11633.337706,1,90,4,4,1.12,2,2,0,0,2,1,68,2,4,1,1
+66782,7,2,2,19,NA,4,4,2,19,236,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10848.628906,11198.221038,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+66783,7,2,2,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,11331.751026,1,101,4,4,0.78,4,4,1,2,0,2,32,1,3,3,NA
+66784,7,2,1,9,NA,4,4,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9261.557132,9701.953496,2,100,7,7,1.34,5,5,0,2,0,2,53,1,4,4,NA
+66785,7,2,2,60,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,1,1,2,1,1,2,2,NA,18002.759054,18633.946783,2,102,15,15,3.82,5,5,0,1,2,1,60,2,2,1,1
+66786,7,2,1,13,NA,5,6,2,13,163,NA,NA,2,1,3,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7046.884472,7540.251296,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+66787,7,2,2,8,NA,2,2,1,8,99,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12307.832776,12608.996083,2,93,5,5,0.89,4,4,0,2,0,1,42,NA,NA,6,NA
+66788,7,2,1,31,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,39049.131011,2,98,99,99,NA,4,4,1,0,1,2,68,1,9,2,NA
+66789,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,30505.56355,32886.600907,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+66790,7,2,1,4,NA,4,4,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,11070.778245,1,100,3,3,0.73,3,3,2,0,0,2,39,1,3,5,NA
+66791,7,2,2,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,12,77,NA,4,1,0,0,0,1,21,1,4,6,NA
+66792,7,2,2,4,NA,2,2,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11429.37307,12618.334582,2,90,6,6,1.35,3,3,1,0,0,1,31,1,3,1,4
+66793,7,2,1,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,16851.334496,17382.882475,2,101,7,7,2.64,2,2,0,0,1,2,70,1,3,4,NA
+66794,7,2,1,10,NA,5,6,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6850.601671,7388.972861,3,91,14,14,4.03,4,4,0,2,0,1,51,2,4,1,5
+66795,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,34810.007519,37700.123082,1,94,3,3,0.37,5,5,0,3,0,2,29,1,4,4,NA
+66796,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,79632.40863,82653.579173,1,95,9,9,4.08,2,2,0,0,2,2,65,1,5,1,3
+66797,7,2,2,30,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,27303.803575,27793.15685,1,96,7,7,1.52,4,4,0,2,0,2,30,2,4,1,5
+66798,7,2,2,56,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,4,1,NA,1,2,2,1,2,2,2,2,2,2,34082.505027,35315.234687,2,92,2,2,0.4,3,3,0,0,0,1,50,2,4,1,4
+66799,7,2,1,6,NA,4,4,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10665.048307,10792.905577,1,96,8,8,2.78,3,3,0,2,0,2,34,1,4,3,NA
+66800,7,2,1,11,NA,1,1,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,13742.678011,2,98,4,4,0.75,4,4,0,2,0,2,33,1,2,5,NA
+66801,7,2,1,7,NA,3,3,2,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57057.523607,61384.115788,1,101,14,14,4.21,4,4,1,1,0,2,37,1,5,1,5
+66802,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,26763.110196,28578.287397,2,94,5,5,1.3,3,3,0,1,0,1,43,1,3,6,NA
+66803,7,2,2,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,1,1,2,2,1,2,2,NA,NA,NA,NA,43813.24867,44924.427845,1,98,4,4,0.66,4,4,2,0,0,2,22,1,4,6,NA
+66804,7,2,1,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,12824.368043,13581.00423,2,101,3,3,1.13,1,1,0,0,1,1,74,1,1,2,NA
+66805,7,2,2,2,NA,4,4,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,8048.181894,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+66806,7,1,2,29,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,14698.806915,0,2,99,15,15,5,2,2,0,0,0,1,30,NA,NA,1,5
+66807,7,2,1,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,1,1,2,2,1,9177.295801,9603.338468,2,92,3,3,0.45,4,4,0,0,1,1,64,2,1,1,1
+66808,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,5950.162866,6182.885443,2,95,5,5,1.05,3,3,0,1,1,1,63,1,2,1,3
+66809,7,1,1,68,NA,2,2,NA,NA,NA,2,NA,2,1,6,NA,3,2,NA,2,2,2,1,2,2,NA,NA,NA,NA,9693.846555,0,2,93,5,5,1.04,4,4,0,1,1,1,68,2,3,2,NA
+66810,7,2,1,14,NA,4,4,1,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16147.713323,16532.569027,1,92,NA,1,0.18,4,3,0,2,0,2,56,1,4,4,NA
+66811,7,2,1,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,9177.295801,9548.31812,2,92,77,77,NA,4,4,0,0,1,1,20,1,2,5,NA
+66812,7,2,1,15,NA,4,4,1,15,186,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10569.808278,10821.723266,2,92,10,10,4.76,2,2,0,1,0,2,40,1,5,4,NA
+66813,7,2,1,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8655.162127,9330.305761,2,95,10,10,3.67,3,3,0,1,0,1,43,1,4,1,4
+66814,7,1,1,34,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,23284.536512,0,1,97,7,7,1.74,4,4,2,0,0,1,34,1,5,1,5
+66815,7,1,1,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58116.402634,0,1,97,15,15,5,5,5,2,0,1,1,43,1,5,1,5
+66816,7,2,1,0,8,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,4461.618312,4741.503802,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+66817,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,22820.082463,26177.958982,1,94,2,2,0.83,1,1,0,0,1,2,72,1,3,2,NA
+66818,7,2,2,8,NA,3,3,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39976.778207,40067.73727,2,100,15,15,5,3,3,0,1,0,1,38,1,4,1,4
+66819,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,102928.893739,104528.537732,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+66820,7,1,2,11,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7710.907339,0,2,99,6,6,0.94,7,7,0,4,0,2,32,1,3,1,3
+66821,7,2,2,30,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,3,1,2,2,2,2,1,2,2,2,2,2,2,41791.57979,45058.092516,2,102,6,6,1.12,4,4,1,1,0,1,38,2,2,1,3
+66822,7,2,1,19,NA,5,6,1,19,230,2,NA,2,1,3,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8431.995003,8807.000735,1,95,3,3,0.71,3,3,0,0,0,1,57,2,2,1,4
+66823,7,2,2,5,NA,5,6,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8894.95474,1,97,15,15,5,4,4,1,1,0,1,44,2,5,1,5
+66824,7,2,2,14,NA,2,2,1,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18728.878486,20545.934985,2,93,10,10,3.4,3,3,0,1,0,1,51,1,3,1,4
+66825,7,2,1,8,NA,2,2,2,8,107,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11546.167056,12698.029578,1,90,7,7,1.56,4,4,1,1,0,2,37,1,2,77,NA
+66826,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11355.3308,11862.334174,1,96,9,9,3.97,2,2,0,0,2,1,72,1,4,1,5
+66827,7,2,1,11,NA,5,6,2,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9175.735601,9896.833095,3,91,15,15,5,4,4,0,2,0,1,44,2,5,1,5
+66828,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,116839.212439,120894.200639,2,97,9,9,4.08,2,2,0,0,0,2,24,1,4,1,3
+66829,7,2,1,8,NA,5,6,2,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6262.834446,6786.531245,3,90,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+66830,7,2,2,12,NA,5,6,1,12,154,NA,NA,2,1,4,6,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,5668.184078,6218.795289,2,92,99,2,0.31,7,4,3,3,1,1,61,2,1,1,3
+66831,7,2,1,0,0,1,1,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7222.23638,7675.3005,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+66832,7,2,1,75,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,7608.031426,7606.53553,1,99,5,5,1.84,1,1,0,0,1,1,75,1,5,2,NA
+66833,7,2,2,39,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,3,1,2,2,2,2,1,2,2,1,2,2,2,41067.133288,41493.788594,2,91,3,3,0.73,3,3,0,0,0,2,22,2,2,5,NA
+66834,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14366.887416,15220.688498,3,90,15,15,5,5,5,0,1,1,2,61,1,5,2,NA
+66835,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,15800.216306,15663.050285,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+66836,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,35522.958395,37225.553122,1,102,13,1,0,2,1,0,0,0,1,22,1,4,5,NA
+66837,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16489.238752,16479.116235,1,96,6,6,1.21,4,4,2,0,0,1,24,1,4,1,3
+66838,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,152978.138087,155477.549686,1,101,7,7,3.13,1,1,0,0,1,2,65,1,4,2,NA
+66839,7,2,2,25,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,11696.173591,12195.620792,1,93,15,1,0.41,6,1,0,0,0,1,34,2,5,5,NA
+66840,7,2,2,10,NA,3,3,1,10,123,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61059.578287,60876.138307,1,98,15,15,4.34,4,4,0,2,0,2,52,1,3,1,5
+66841,7,2,1,3,NA,5,7,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26980.605125,29855.353305,1,94,2,2,0.3,5,5,1,2,0,1,23,1,1,6,NA
+66842,7,1,2,0,9,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3861.876549,0,1,96,15,15,4.81,5,5,1,2,0,1,33,1,5,1,3
+66843,7,2,1,5,NA,5,7,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10859.969359,11974.89205,1,98,7,7,1.52,4,4,2,0,0,1,30,1,3,1,4
+66844,7,2,1,18,NA,4,4,1,18,221,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16147.713323,16270.487634,1,92,77,77,NA,5,5,1,2,0,2,41,1,3,5,NA
+66845,7,2,1,1,15,3,3,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47507.757497,53600.040598,1,101,14,14,3.15,5,5,2,1,0,1,35,1,4,1,5
+66846,7,2,1,2,NA,4,4,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,7031.647494,2,100,8,8,2.7,3,3,1,0,0,2,41,1,4,1,3
+66847,7,2,1,31,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16505.268729,17550.815541,3,91,14,14,4.32,3,3,1,0,0,1,31,2,3,1,4
+66848,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,141821.601523,151144.430607,1,90,15,15,5,4,4,0,2,1,2,52,1,5,1,NA
+66849,7,2,1,0,11,5,7,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7317.981812,7846.84086,1,97,15,15,3.7,5,5,1,1,0,2,21,1,4,5,NA
+66850,7,2,2,71,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19603.448069,21893.680196,2,98,5,5,0.59,7,7,2,1,2,2,71,1,2,1,1
+66851,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,143866.32507,144700.959982,1,97,15,15,5,1,1,0,0,0,2,46,1,4,2,NA
+66852,7,2,1,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8569.438441,9098.434789,2,99,7,7,1.19,6,6,1,3,0,2,38,1,3,5,NA
+66853,7,2,1,2,NA,2,2,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11030.983881,11818.903113,2,93,7,7,1.56,4,4,1,1,0,1,35,2,4,1,4
+66854,7,2,2,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,26021.868354,25835.926172,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+66855,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17606.560681,17743.688949,2,95,6,6,1.7,2,2,0,0,0,2,54,1,4,2,NA
+66856,7,2,2,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12871.484115,13281.030392,2,102,5,5,0.89,4,4,1,1,0,2,28,2,2,1,2
+66857,7,2,2,15,NA,5,6,2,15,181,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6486.356303,6952.179218,1,91,7,7,1.57,4,4,0,3,0,2,38,2,2,3,NA
+66858,7,2,2,6,NA,4,4,1,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8362.256577,8929.488766,2,100,7,7,2.37,3,3,0,1,1,2,45,1,5,1,NA
+66859,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6441.733603,6839.385512,1,91,6,6,0.99,5,5,3,0,0,2,33,2,3,1,4
+66860,7,2,1,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,51965.135941,60918.873896,2,101,2,1,0.37,2,1,0,0,0,1,21,1,4,5,NA
+66861,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,17478.873252,2,95,6,6,1.36,3,3,0,1,1,2,62,1,4,5,NA
+66862,7,2,1,10,NA,4,4,2,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9699.683862,9681.447258,2,101,2,2,0.27,6,6,0,3,0,2,45,1,2,5,NA
+66863,7,2,1,77,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,70642.066815,74773.070249,2,96,12,12,NA,2,2,0,0,2,1,77,1,5,1,4
+66864,7,2,1,16,NA,4,4,1,16,196,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11666.009872,12200.765691,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+66865,7,2,2,15,NA,1,1,1,15,184,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22424.988432,22856.096824,1,94,5,5,0.87,4,4,0,2,0,2,41,2,4,1,1
+66866,7,2,1,5,NA,4,4,1,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8385.172131,8641.288503,2,93,5,5,1.04,4,4,1,1,0,1,29,1,3,6,NA
+66867,7,1,1,18,NA,1,1,NA,NA,NA,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,22721.243258,0,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+66868,7,2,1,8,NA,5,6,2,8,100,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9928.619925,10708.884665,1,91,15,15,5,4,4,1,1,0,1,43,2,5,1,5
+66869,7,2,2,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,24719.680932,24496.884814,1,97,15,15,5,2,2,0,0,0,1,59,1,5,1,5
+66870,7,2,2,12,NA,1,1,1,12,153,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22424.988432,24600.637711,1,94,15,15,4.37,7,7,0,4,1,1,58,1,4,1,5
+66871,7,2,1,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,32375.321924,32878.474997,1,95,7,6,2.6,5,1,1,2,0,1,26,1,4,6,NA
+66872,7,2,1,37,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22424.102245,23525.946014,1,92,14,14,3.47,4,4,0,2,0,1,37,1,5,1,5
+66873,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,163102.567998,168673.834909,1,99,12,12,NA,2,2,0,0,0,1,55,1,5,1,4
+66874,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,53955.606082,55085.619185,2,100,15,15,5,4,4,1,1,0,1,29,1,4,1,4
+66875,7,2,2,69,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,10760.495249,11342.54557,3,90,4,4,1.22,2,2,0,0,2,2,69,2,4,1,1
+66876,7,2,2,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,134694.414609,139412.076132,2,91,15,15,5,3,3,0,0,2,2,62,1,4,1,4
+66877,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,56180.550638,64695.241485,2,98,5,5,1.63,2,2,0,0,2,1,80,1,3,1,3
+66878,7,2,2,15,NA,1,1,1,15,180,NA,NA,2,2,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21062.314667,21780.985882,3,91,3,3,0.39,6,6,1,1,0,1,39,2,1,6,NA
+66879,7,2,2,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,139800.409559,144765.126463,1,100,15,15,4.56,4,4,0,2,0,2,42,1,4,1,3
+66880,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18452.546861,18969.498532,1,93,12,12,NA,3,3,0,1,0,2,48,1,3,5,NA
+66881,7,2,2,40,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,20303.639991,20411.004471,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+66882,7,2,2,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,50922.951701,51648.728485,1,92,15,15,5,3,3,0,0,1,1,57,1,3,1,4
+66883,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,109181.566304,117126.933999,2,91,7,7,2.31,2,2,0,0,0,2,58,1,5,3,NA
+66884,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,20691.085146,20754.703375,2,91,4,4,1.22,2,2,0,0,0,1,53,1,4,1,4
+66885,7,2,1,76,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,6856.239408,7260.756717,2,95,5,5,0.87,4,4,0,0,2,2,77,1,2,1,1
+66886,7,2,1,5,NA,2,2,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14318.290734,14771.527769,3,91,6,6,0.83,6,6,1,3,0,1,37,1,4,1,4
+66887,7,1,2,6,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9122.654131,0,1,96,9,9,2.18,5,5,1,1,0,1,26,1,4,1,4
+66888,7,2,2,1,18,4,4,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7281.670423,7938.059494,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+66889,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,19541.667675,22573.12843,2,95,1,1,0.36,1,1,0,0,0,1,59,1,5,5,NA
+66890,7,2,1,14,NA,1,1,2,14,178,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,22044.437334,22214.434256,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+66891,7,2,2,15,NA,3,3,2,15,191,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,132630.209478,136474.31291,1,97,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+66892,7,2,2,1,16,3,3,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46281.468826,49830.1764,1,92,9,4,1,7,3,2,1,0,1,45,1,4,2,NA
+66893,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,168185.448935,169079.116739,1,95,9,9,4.01,2,2,0,0,1,1,60,1,3,1,3
+66894,7,1,2,1,12,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,0,1,91,9,9,2.97,3,3,1,0,0,1,31,2,5,1,5
+66895,7,2,1,10,NA,2,2,2,10,121,NA,NA,2,2,3,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10490.055059,10676.813132,3,90,4,4,0.63,5,5,0,3,0,1,45,2,4,1,4
+66896,7,2,1,64,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,8886.717016,9234.294003,2,101,3,3,0.78,3,3,0,0,1,1,64,1,2,1,3
+66897,7,2,2,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,18097.801029,17642.065425,2,100,9,9,3.24,3,3,1,0,0,1,32,1,3,1,4
+66898,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,24056.374863,23398.009058,2,99,6,6,2.6,1,1,0,0,0,2,45,1,5,5,NA
+66899,7,2,2,11,NA,1,1,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,22662.992756,24846.804377,2,98,13,13,NA,5,5,0,2,0,1,48,2,1,1,2
+66900,7,2,2,10,NA,4,4,1,10,121,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,13050.526646,2,101,1,1,0.15,3,3,0,2,0,2,58,1,3,5,NA
+66901,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11449.435533,12331.980143,2,99,6,6,1.98,2,2,0,0,2,1,77,1,3,1,3
+66902,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,25691.564623,26900.812132,1,90,3,3,0.63,3,3,1,1,0,2,32,1,4,5,NA
+66903,7,2,2,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,74517.751389,74746.868777,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+66904,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,11507.810748,11238.9991,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+66905,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18801.993237,20649.846581,2,96,8,5,2.2,2,1,0,0,0,2,25,2,5,5,NA
+66906,7,2,2,11,NA,4,4,2,11,135,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10888.493631,11627.08662,1,101,1,1,0.21,3,3,0,2,0,2,32,1,4,5,NA
+66907,7,2,2,37,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,11105.558187,11128.234812,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+66908,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,17824.805721,17767.952896,2,92,2,2,0.57,2,2,0,0,0,2,56,1,3,2,NA
+66909,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,52280.406546,52747.829022,1,95,3,1,0.28,2,1,0,0,0,1,27,1,3,5,NA
+66910,7,1,2,8,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14465.997114,0,2,96,3,3,0.24,7,7,2,3,1,2,40,1,3,3,NA
+66911,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,1,22697.846242,22639.540149,2,100,4,4,1.47,2,1,0,0,0,2,38,1,3,2,NA
+66912,7,2,2,14,NA,5,7,2,14,172,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6386.337576,6452.948976,1,93,15,15,4.59,4,4,0,2,0,2,45,1,5,1,5
+66913,7,2,1,9,NA,2,2,2,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8516.07921,8708.134582,3,90,14,14,3.69,4,4,0,2,0,2,49,1,4,1,4
+66914,7,2,2,51,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12162.017722,12226.329772,2,92,12,9,5,7,1,0,0,2,1,53,2,3,1,3
+66915,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32461.799549,32422.216998,1,101,4,4,0.99,2,2,0,0,0,2,51,1,5,1,2
+66916,7,2,1,1,23,4,4,1,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6859.539552,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+66917,7,2,1,7,NA,1,1,1,7,89,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16747.549238,17012.570163,2,96,7,7,1.34,5,5,0,2,0,1,24,2,2,5,NA
+66918,7,2,1,72,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,32916.648979,34960.567692,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+66919,7,2,2,13,NA,1,1,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,27070.679378,27847.54398,1,92,14,14,3.15,5,5,1,2,0,1,34,1,4,1,4
+66920,7,2,1,77,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,17113.447343,18123.138698,2,101,6,6,2.04,2,2,0,0,2,1,77,1,1,1,2
+66921,7,2,2,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,38679.12951,52110.643741,1,98,3,3,0.86,2,2,0,1,0,2,34,1,4,5,NA
+66922,7,2,2,12,NA,4,4,2,12,149,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12180.874919,12414.982997,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+66923,7,2,2,47,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17623.300255,17855.084956,1,100,15,15,5,3,3,0,0,0,1,47,2,5,1,5
+66924,7,2,1,36,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,2,2,2,NA,NA,NA,NA,28519.067294,31362.329476,1,96,15,15,5,4,4,0,2,0,1,36,2,3,1,4
+66925,7,2,2,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,46764.716703,54568.74353,1,99,2,2,0.46,1,1,0,0,0,2,23,1,5,5,NA
+66926,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,175544.769665,178639.415011,1,91,15,15,5,2,1,0,0,0,1,51,1,3,1,NA
+66927,7,2,1,20,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,51965.135941,52673.460841,2,101,1,1,0.11,2,1,0,0,0,1,19,NA,NA,NA,NA
+66928,7,2,2,64,NA,5,6,1,NA,NA,2,NA,2,1,1,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,7118.69664,7393.452915,1,103,77,77,NA,6,6,0,2,2,1,70,NA,NA,1,1
+66929,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12331.419303,12882.003985,2,95,14,14,5,2,2,0,0,2,1,73,1,5,1,4
+66930,7,2,1,41,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,31640.296506,31576.726829,2,94,15,15,5,3,3,0,0,0,1,41,2,3,1,NA
+66931,7,2,1,16,NA,3,3,2,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,68148.957861,67253.324127,1,93,15,15,3.92,5,5,0,1,0,2,54,1,5,1,5
+66932,7,2,1,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,10420.275705,10501.786379,2,96,7,7,2.58,2,2,0,0,1,2,55,1,4,3,NA
+66933,7,2,1,34,NA,5,6,1,NA,NA,2,NA,2,7,77,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16770.056318,16992.484825,1,103,12,12,NA,2,2,0,0,0,1,34,2,5,1,5
+66934,7,2,2,12,NA,5,6,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7952.410952,8162.028124,2,98,9,9,2.29,5,5,0,2,0,1,36,1,4,1,4
+66935,7,2,1,1,19,5,7,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6370.584728,7024.611386,2,94,10,10,3.51,3,3,1,0,0,2,30,2,5,1,NA
+66936,7,2,1,28,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,2,1,2,1,1,2,2,3,14313.345971,14892.009983,3,91,6,6,1.12,4,4,0,0,2,1,69,2,3,1,1
+66937,7,2,1,37,NA,3,3,1,NA,NA,2,NA,2,2,6,NA,2,5,NA,1,2,2,2,2,2,1,2,2,1,63557.943986,66218.577849,2,103,12,12,NA,3,3,0,0,1,1,60,2,2,1,2
+66938,7,2,2,62,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,NA,NA,NA,1,2,1,NA,17248.011865,22917.917651,1,97,14,14,2.29,7,7,1,2,2,1,40,2,1,1,1
+66939,7,1,2,15,NA,2,2,NA,NA,NA,NA,NA,2,2,3,10,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,18053.382334,0,1,90,10,10,3.67,3,3,0,1,0,2,40,2,2,77,NA
+66940,7,2,2,14,NA,1,1,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20117.170449,20520.192725,1,94,2,2,0.27,5,5,0,4,0,2,47,2,1,4,NA
+66941,7,2,1,25,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,42077.383821,43345.097117,1,102,14,14,4.32,3,3,1,0,0,1,25,1,4,1,4
+66942,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,2,1,2,2,1,2,2,1,2,2,1,20562.749362,20582.941069,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+66943,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,27934.372045,29727.618806,1,101,7,2,0.67,3,1,0,0,1,2,69,1,4,2,NA
+66944,7,2,1,72,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,79754.311902,85970.307401,1,97,6,6,2.15,2,2,0,0,2,1,72,1,5,1,NA
+66945,7,2,2,57,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,24646.971819,24775.50368,1,102,6,6,1.48,3,3,0,0,1,2,57,2,1,1,4
+66946,7,2,1,32,NA,5,7,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,21271.615575,1,94,6,6,1.33,4,4,2,0,0,2,29,1,2,1,4
+66947,7,2,1,51,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,29883.483388,30470.971906,1,102,9,9,3.74,2,2,0,0,0,2,45,1,4,1,2
+66948,7,2,1,14,NA,5,6,1,15,180,NA,NA,2,1,3,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6140.97758,6391.148404,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+66949,7,2,2,54,NA,5,7,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21760.356138,21875.42365,1,92,14,14,5,2,2,0,0,0,1,56,2,4,1,4
+66950,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,111065.717962,123974.675785,2,94,3,3,0.54,4,4,0,1,0,2,48,1,3,1,3
+66951,7,2,2,8,NA,1,1,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,2,2,2,2,17053.854294,17496.484818,3,92,7,7,0.93,7,7,1,3,0,2,20,1,3,1,1
+66952,7,2,2,4,NA,3,3,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28409.87702,1,101,6,6,0.87,6,6,2,2,0,2,23,1,4,6,NA
+66953,7,2,1,10,NA,1,1,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13837.588743,14876.800627,1,92,8,8,2.62,3,3,0,1,0,1,41,2,3,1,9
+66954,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,81857.569857,83650.947594,2,92,15,10,5,4,1,0,0,0,1,28,1,5,5,NA
+66955,7,2,1,38,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,39049.131011,2,98,15,15,5,3,3,0,1,0,1,38,1,4,1,3
+66956,7,2,2,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,41018.498876,43893.586815,2,98,3,3,0.54,3,3,1,0,0,1,23,1,3,1,2
+66957,7,2,1,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,136361.754972,134969.826866,1,98,6,6,1.75,2,2,0,0,2,1,62,1,4,1,3
+66958,7,2,1,52,NA,3,3,1,NA,NA,2,NA,2,2,5,NA,3,1,NA,2,2,2,1,2,2,2,2,2,2,24127.240234,24937.486066,2,93,6,6,1.65,2,2,0,0,0,1,52,2,3,1,5
+66959,7,2,2,43,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,5,2,2,2,2,2,2,2,1,2,2,2,33767.584626,33943.679708,2,102,6,3,0.54,6,4,0,4,0,2,43,2,1,5,NA
+66960,7,2,1,1,20,5,6,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8674.760516,9729.693026,1,94,10,10,3.04,4,4,2,0,0,2,30,1,4,1,5
+66961,7,2,1,36,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19144.719218,19850.422567,1,101,10,10,3.67,3,3,1,0,0,1,36,2,5,1,5
+66962,7,2,1,1,22,4,4,1,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5930.749873,6003.089312,2,93,7,7,1.83,3,3,1,0,0,2,34,2,3,1,3
+66963,7,1,2,70,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,64383.454106,0,1,90,12,12,NA,2,2,0,0,2,1,75,1,5,1,4
+66964,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,100264.332334,101647.830414,1,99,7,7,4.09,1,1,0,0,0,2,27,1,5,5,NA
+66965,7,2,1,2,NA,3,3,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21718.301271,24503.405153,1,94,5,5,0.74,5,5,1,1,0,2,24,1,3,1,4
+66966,7,2,2,14,NA,1,1,2,14,170,NA,NA,2,2,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19850.979841,20551.555585,1,97,3,3,0.5,5,5,0,2,0,1,56,2,2,6,NA
+66967,7,2,1,33,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,2,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,39073.76885,38778.866765,2,99,99,3,0.79,4,2,0,0,0,2,42,2,4,5,NA
+66968,7,2,2,11,NA,1,1,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21075.336925,21477.798111,1,92,14,14,3.15,5,5,1,2,0,1,34,1,4,1,4
+66969,7,2,1,10,NA,1,1,1,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8828.580268,8822.70874,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+66970,7,2,1,18,NA,1,1,2,18,222,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26704.187335,27112.370182,1,95,8,5,1.36,3,2,0,0,0,1,50,1,9,6,NA
+66971,7,2,2,8,NA,5,6,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9620.269705,10273.007637,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+66972,7,1,1,1,13,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,0,2,100,2,2,0.25,4,4,2,1,0,2,39,1,2,5,NA
+66973,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,111065.717962,111174.779569,2,94,10,10,4.42,2,2,0,0,0,2,49,1,4,1,1
+66974,7,2,1,72,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11034.04089,11599.454609,2,95,5,5,1.45,2,2,0,0,2,1,72,1,3,1,3
+66975,7,2,1,17,NA,2,2,1,17,205,2,NA,2,1,4,9,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,18206.126374,19392.083527,2,93,3,3,0.58,4,4,0,1,1,1,65,2,1,3,NA
+66976,7,2,1,24,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,35338.972549,37196.45724,2,90,6,6,1.21,4,4,0,0,0,2,59,2,1,6,NA
+66977,7,2,2,10,NA,4,4,2,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.490652,8919.477637,2,97,4,4,0.57,5,5,1,3,0,2,33,1,3,5,NA
+66978,7,2,1,7,NA,5,6,2,7,86,NA,NA,2,2,3,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9928.619925,10758.848877,1,91,14,14,3.69,4,4,1,1,0,2,29,2,5,1,5
+66979,7,2,2,16,NA,5,6,2,16,200,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,7588.544207,8133.521636,3,91,6,6,1.34,4,4,0,2,0,1,52,2,3,1,1
+66980,7,2,1,7,NA,5,6,2,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9928.619925,10758.848877,1,91,2,2,0.32,3,3,1,1,0,2,28,1,4,77,NA
+66981,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,27921.272721,1,101,4,4,0.84,3,3,0,1,0,1,42,1,4,1,4
+66982,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,21924.837493,22597.469895,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+66983,7,2,2,9,NA,4,4,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,8362.256577,9003.967662,2,100,1,1,0.08,5,5,1,2,0,2,19,1,3,NA,NA
+66984,7,2,1,75,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,53663.609619,56801.747638,2,93,10,10,3.61,3,3,0,0,2,1,75,1,4,1,4
+66985,7,2,2,77,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,3,2,NA,1,2,2,1,2,2,2,2,1,NA,18241.877822,19614.829564,2,93,3,3,0.66,2,2,0,0,2,2,69,2,4,3,NA
+66986,7,2,1,9,NA,5,6,1,9,116,NA,NA,1,1,NA,4,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6254.093492,6623.185563,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+66987,7,2,1,0,8,5,6,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5507.575232,5738.278852,1,97,15,15,4.84,6,6,2,0,0,1,53,NA,NA,1,NA
+66988,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,19870.689174,19326.875638,1,96,3,3,0.93,2,2,0,1,0,2,40,1,5,5,NA
+66989,7,2,2,1,14,1,1,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,11879.078775,2,98,2,2,0.35,3,3,2,0,0,2,20,1,4,5,NA
+66990,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17527.360027,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+66991,7,2,1,24,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,14313.345971,14703.033095,3,91,10,10,4.42,2,2,0,0,0,1,47,2,2,1,NA
+66992,7,2,1,7,NA,3,3,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,42986.51011,44926.997612,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+66993,7,2,2,38,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,2,6,2,2,2,2,2,2,2,NA,NA,NA,NA,32982.479382,32092.542799,2,99,99,3,0.79,4,2,0,0,0,2,42,2,4,5,NA
+66994,7,2,1,52,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,25264.512558,25333.955786,2,93,15,15,5,2,2,0,0,0,1,52,2,5,1,5
+66995,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,76827.086279,81170.917712,2,99,15,15,5,2,2,0,0,0,2,37,1,5,1,5
+66996,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,25651.892165,25827.388922,1,97,14,14,5,3,3,0,0,0,2,51,1,5,1,4
+66997,7,2,2,15,NA,4,4,1,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19770.196972,20174.150218,2,102,10,10,3.78,3,3,0,1,0,1,33,1,3,5,NA
+66998,7,2,2,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10771.85499,12037.490455,2,97,4,4,0.81,4,4,1,1,0,2,51,1,3,4,NA
+66999,7,2,2,15,NA,4,4,1,15,191,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13801.622751,14246.373765,1,100,10,10,2.59,5,5,0,1,0,2,40,1,5,1,NA
+67000,7,2,1,49,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19320.837782,1,99,15,8,4.59,4,1,0,2,0,1,49,1,4,6,NA
+67001,7,2,1,76,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,63147.363155,65543.108264,1,91,8,8,3.4,2,2,0,0,2,2,74,1,4,1,2
+67002,7,2,1,39,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,21317.283165,22287.043754,2,96,5,5,1.08,3,3,0,1,0,2,41,1,3,1,NA
+67003,7,2,1,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,105401.67423,106753.633419,2,92,15,15,5,1,1,0,0,1,1,68,1,5,3,NA
+67004,7,2,2,4,NA,1,1,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18078.669459,18848.958225,1,101,2,2,0.26,5,5,3,0,0,2,26,1,2,1,3
+67005,7,2,1,19,NA,4,4,2,19,239,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+67006,7,2,2,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,7906.792868,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+67007,7,2,2,40,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,19247.348846,19527.962249,2,101,9,9,3.74,2,2,0,0,0,2,40,2,4,1,2
+67008,7,2,2,17,NA,4,4,1,17,210,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16659.324602,16705.662179,1,92,77,77,NA,5,5,1,2,0,2,41,1,3,5,NA
+67009,7,2,2,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,90299.161173,96625.186997,1,92,6,6,1.57,3,3,0,1,0,1,29,1,4,6,NA
+67010,7,2,2,3,NA,2,2,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11009.851855,11478.955249,1,93,13,3,0.54,6,3,2,1,0,1,23,NA,NA,5,NA
+67011,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,7869.59899,8538.749518,2,100,2,2,0.72,1,1,0,0,1,2,67,1,1,4,NA
+67012,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,14138.631841,14167.501749,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+67013,7,2,1,11,NA,4,4,2,11,132,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9699.683862,9867.948233,2,97,14,14,3.91,4,4,1,1,0,1,38,1,4,1,5
+67014,7,2,2,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18696.216547,18746.863759,2,100,5,5,0.95,4,4,0,0,1,2,53,1,3,5,NA
+67015,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,14516.765165,13843.248577,3,90,7,7,2.23,3,3,0,0,0,2,51,1,4,3,NA
+67016,7,2,2,11,NA,3,3,2,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,98,14,14,3.9,4,4,0,3,0,2,31,1,4,1,NA
+67017,7,2,1,11,NA,3,3,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,82670.203859,85919.643529,1,97,15,15,5,5,5,0,3,0,2,47,2,5,1,5
+67018,7,1,1,5,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,17458.543556,0,2,91,6,6,1.3,4,4,1,1,0,2,27,1,4,6,NA
+67019,7,2,1,6,NA,5,7,1,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7101.095162,7528.512272,2,96,15,15,5,4,4,1,1,0,2,35,2,5,1,5
+67020,7,2,1,78,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,74473.849242,78828.927467,2,98,14,14,5,2,2,0,0,2,1,78,1,5,1,3
+67021,7,2,1,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15599.953109,16099.73574,1,99,7,7,1.89,3,3,0,1,0,1,50,1,5,1,2
+67022,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,35813.944922,38588.040359,1,100,8,8,2.97,2,2,0,0,1,2,54,1,4,3,NA
+67023,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,8746.76562,8644.133303,2,97,15,15,5,3,3,0,0,3,2,80,1,3,2,NA
+67024,7,2,2,12,NA,3,3,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71303.309206,77744.474592,2,94,15,15,5,5,5,0,2,0,1,53,1,5,1,5
+67025,7,2,2,78,NA,2,2,2,NA,NA,2,NA,2,1,NA,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,17318.187297,23904.945555,2,90,2,2,0.89,1,1,0,0,1,2,78,2,1,1,NA
+67026,7,2,1,40,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,35406.972937,38657.357615,2,98,12,12,NA,3,3,0,1,0,2,34,1,4,1,3
+67027,7,2,2,15,NA,1,1,2,16,192,NA,NA,2,1,4,10,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18166.909002,19929.443801,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+67028,7,2,2,6,NA,1,1,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10118.363218,10311.586628,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+67029,7,2,2,2,NA,5,6,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5373.703942,1,91,5,5,0.89,4,4,2,0,0,1,39,1,4,1,5
+67030,7,2,1,6,NA,4,4,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8714.559478,8865.734494,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+67031,7,2,2,18,NA,4,4,2,18,219,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,5,5,0.92,5,5,0,3,0,2,54,1,3,2,NA
+67032,7,2,2,32,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,21351.921325,21766.153521,2,92,6,6,1.35,3,3,1,0,0,2,32,1,5,1,5
+67033,7,2,1,3,NA,4,4,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10615.626123,11270.93481,1,100,8,8,1.61,6,6,1,3,0,1,29,1,5,6,NA
+67034,7,2,2,37,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,2,1,2,2,1,2,2,2,2,2,2,35353.005268,34399.106917,2,94,12,12,NA,4,4,0,2,0,1,47,2,2,1,2
+67035,7,2,2,3,NA,1,1,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13196.707564,14212.912965,2,96,8,8,1.33,7,7,2,1,1,1,62,2,1,1,1
+67036,7,2,2,16,NA,2,2,2,16,202,NA,NA,1,1,NA,11,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,12536.713942,13553.99602,3,90,12,12,NA,2,2,0,1,0,2,40,2,2,1,NA
+67037,7,2,2,66,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,13676.984152,18290.186018,2,91,4,4,0.67,5,4,2,0,2,2,66,2,1,1,NA
+67038,7,2,1,11,NA,3,3,2,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13713.949705,16784.308768,1,99,6,6,1.12,4,4,0,2,0,1,39,1,3,1,3
+67039,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,63630.652053,80950.199865,3,90,9,9,3.64,2,2,0,0,0,1,27,2,4,1,5
+67040,7,2,2,19,NA,1,1,2,19,232,2,NA,1,1,NA,15,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,18368.872199,18722.003928,2,94,9,1,0.14,3,1,0,0,0,2,48,2,2,6,NA
+67041,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,42468.064168,44382.588967,2,101,3,3,0.54,3,3,0,2,0,2,36,1,3,5,NA
+67042,7,2,2,9,NA,5,6,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8022.905544,9049.556422,2,92,6,6,1.34,4,4,1,1,0,1,40,2,3,1,3
+67043,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,17703.992112,2,95,15,15,5,3,3,0,0,0,1,59,NA,NA,6,NA
+67044,7,2,2,19,NA,5,7,2,19,231,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11975.458482,12291.118947,1,97,15,15,5,6,6,0,3,0,1,47,1,5,1,5
+67045,7,2,2,1,13,3,3,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19168.30863,20370.539296,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+67046,7,2,1,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,7514.993062,7713.03043,2,90,7,7,2.38,2,2,0,0,1,1,66,1,4,5,NA
+67047,7,2,2,54,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,35630.227837,35816.036445,2,102,8,8,4.59,1,1,0,0,0,2,54,2,5,5,NA
+67048,7,2,2,12,NA,5,7,1,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12427.929278,12265.18624,1,102,15,15,5,4,4,0,2,0,2,40,2,5,1,4
+67049,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,10072.885959,10826.303506,2,100,13,13,NA,1,1,0,0,1,2,78,1,2,2,NA
+67050,7,2,2,8,NA,1,1,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,13231.432201,13813.796743,2,97,13,13,NA,3,3,0,1,0,1,28,2,2,1,1
+67051,7,2,2,37,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,13379.422066,13458.148951,2,103,14,14,3.48,5,5,0,2,1,1,43,1,4,1,5
+67052,7,2,1,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,28585.875492,29232.915291,1,94,2,2,0.75,1,1,0,0,0,1,56,1,2,3,NA
+67053,7,2,1,67,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,1,2,2,1,2,1,2,8017.002318,8145.424458,2,93,7,7,1.74,4,4,0,0,1,2,44,2,3,1,4
+67054,7,2,2,8,NA,1,1,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17458.526997,18973.577367,1,94,6,6,0.97,6,6,1,3,0,1,40,1,3,1,4
+67055,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,28244.319428,29210.271354,1,94,4,4,1.24,2,2,0,0,2,2,71,1,2,1,4
+67056,7,2,1,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,11551.71585,11723.869749,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+67057,7,2,1,0,2,4,4,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6423.674397,6522.268756,1,96,14,14,4.19,3,3,1,0,0,1,44,1,4,6,NA
+67058,7,2,1,42,NA,4,4,2,NA,NA,2,NA,2,2,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17602.101156,18151.242681,1,96,15,15,5,4,4,0,1,0,1,42,2,4,1,4
+67059,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,NA,NA,NA,NA,53799.276134,54280.27834,1,94,3,3,0.37,5,5,0,3,0,2,29,1,4,4,NA
+67060,7,2,2,16,NA,4,4,2,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11478.437608,11446.094627,1,96,6,6,1.21,4,4,0,2,0,2,41,1,4,4,NA
+67061,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,36955.110877,37344.656251,1,100,14,14,4.45,3,3,1,0,0,1,33,1,4,1,4
+67062,7,2,2,1,14,3,3,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46813.021927,47428.160802,1,94,8,8,2.62,3,3,1,0,0,2,26,1,4,5,NA
+67063,7,2,1,15,NA,3,3,1,15,188,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,84860.612577,85065.529717,2,98,15,15,5,4,4,0,2,0,1,48,1,3,1,4
+67064,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,94433.586146,97510.682342,1,91,14,14,3.06,5,5,2,0,0,2,30,1,5,1,5
+67065,7,2,1,36,NA,4,4,1,NA,NA,2,NA,2,2,4,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,23654.723526,31640.766843,1,100,4,4,0.99,2,2,0,1,0,1,36,2,4,4,NA
+67066,7,2,1,65,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26087.95381,26422.577042,1,94,4,4,1.24,2,2,0,0,2,2,71,1,2,1,4
+67067,7,2,1,23,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,3,6,NA,2,2,2,NA,NA,NA,2,2,1,2,38474.772527,42632.210531,2,93,3,3,0.43,4,4,0,0,0,1,45,2,2,6,NA
+67068,7,2,1,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,17879.023129,20112.694262,2,90,2,2,0.87,1,1,0,0,0,1,32,1,5,5,NA
+67069,7,2,1,19,NA,3,3,2,19,235,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,69216.263169,70543.167563,1,91,15,15,5,3,3,0,0,0,2,50,1,4,1,4
+67070,7,2,2,64,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,14809.997435,15471.247979,2,95,4,4,0.76,4,4,0,1,2,1,80,1,1,2,NA
+67071,7,2,2,16,NA,3,3,2,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,135393.338842,139317.527763,1,97,14,14,3.9,4,4,0,2,0,1,47,1,3,1,5
+67072,7,2,1,11,NA,2,2,1,11,133,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11971.590477,11920.833512,2,100,14,14,3.58,4,4,0,1,0,1,46,2,5,1,5
+67073,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,42894.724338,43561.362107,1,98,6,3,0.9,2,1,0,0,0,1,24,1,5,5,NA
+67074,7,2,2,37,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,4,1,2,2,2,2,1,2,2,NA,NA,NA,NA,41791.57979,40949.069487,2,102,7,7,1.53,5,5,1,2,0,2,37,2,4,1,4
+67075,7,2,2,14,NA,3,3,1,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,114661.989457,116814.464785,2,101,14,14,4.86,3,3,0,1,0,1,53,1,4,1,5
+67076,7,2,2,10,NA,3,3,1,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23170.920553,22863.220381,3,91,2,2,0.33,4,4,0,3,0,2,31,1,4,4,NA
+67077,7,1,2,4,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5878.066175,0,1,91,15,15,5,5,5,2,1,0,1,40,1,5,1,5
+67078,7,2,2,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,28175.708676,28476.609384,1,94,3,3,0.92,1,1,0,0,1,2,75,1,4,1,NA
+67079,7,2,1,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,20397.472757,21204.731618,1,90,6,6,2.15,2,2,0,0,1,2,79,NA,NA,77,NA
+67080,7,2,2,11,NA,3,3,2,11,140,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14287.66096,14320.169639,2,97,7,7,1.92,3,3,0,1,0,1,57,1,4,1,4
+67081,7,2,2,50,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,22916.776688,24567.079865,1,102,5,5,0.62,7,7,1,3,0,1,49,2,2,1,1
+67082,7,2,1,18,NA,1,1,1,18,224,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22768.423624,22766.421142,2,98,5,5,1.61,2,2,0,0,0,2,44,2,4,3,NA
+67083,7,2,1,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,111284.977736,116509.217288,3,91,15,15,5,4,4,0,0,1,1,60,NA,NA,4,NA
+67084,7,2,2,5,NA,3,3,2,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27528.66901,30383.138359,1,95,5,5,1.19,3,3,1,1,0,1,47,1,2,3,NA
+67085,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,46498.414015,0,1,90,6,6,1.65,2,2,0,0,1,1,55,1,4,77,NA
+67086,7,2,1,8,NA,1,1,1,9,108,NA,NA,2,2,3,2,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,17882.621856,17720.218058,2,98,3,3,0.4,6,6,1,2,0,2,29,2,1,4,NA
+67087,7,2,2,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,15267.012422,14516.051827,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+67088,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,51218.356583,59416.206008,2,98,12,12,NA,2,2,0,0,1,2,80,1,2,2,NA
+67089,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,24654.107413,23987.341621,1,100,6,6,1.39,4,4,0,3,0,2,29,1,4,5,NA
+67090,7,2,2,0,4,2,2,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6752.982789,6610.907942,1,93,12,7,2,4,3,1,0,0,1,43,2,1,1,5
+67091,7,2,2,12,NA,2,2,1,12,149,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15809.066118,17342.845429,2,93,3,3,0.52,5,5,0,2,0,1,41,2,4,1,4
+67092,7,2,1,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,NA,63054.867183,67969.319596,1,90,14,8,4.03,2,1,0,0,2,1,76,1,5,6,NA
+67093,7,2,1,10,NA,1,1,1,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,16288.924956,16868.808147,1,101,3,3,0.41,5,5,0,2,1,2,36,2,4,4,NA
+67094,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,6038.685119,6085.921613,1,96,15,15,5,2,2,0,0,2,1,60,1,5,1,3
+67095,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,28585.875492,28917.983006,1,91,5,5,1.2,3,3,0,0,1,1,58,1,2,1,2
+67096,7,2,1,11,NA,1,1,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11399.23838,11295.714433,1,103,5,5,0.71,6,6,2,2,0,2,31,2,2,1,2
+67097,7,2,2,7,NA,1,1,2,7,85,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15225.935813,15373.8033,2,94,7,7,1.88,4,4,0,2,0,2,28,1,4,4,NA
+67098,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,9793.924718,10548.859356,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+67099,7,2,1,7,NA,2,2,1,7,94,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12231.897958,14410.652577,1,100,3,3,0.39,5,5,1,2,0,1,32,2,1,6,NA
+67100,7,2,2,29,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,20247.768461,26123.76716,2,93,6,6,1.72,2,2,0,1,0,2,29,1,4,3,NA
+67101,7,2,2,8,NA,1,1,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20495.125801,21365.773499,3,92,4,4,0.46,7,7,1,2,0,2,31,2,2,1,1
+67102,7,2,2,13,NA,3,3,2,13,162,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,133500.800632,137370.136946,1,98,5,5,1.39,2,2,0,1,0,1,46,1,4,3,NA
+67103,7,2,2,3,NA,2,2,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11429.37307,12618.334582,2,90,5,5,1.19,3,3,1,0,1,2,60,2,1,4,NA
+67104,7,2,1,8,NA,4,4,2,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9699.683862,9892.534359,2,97,5,5,1.08,3,3,1,1,0,2,27,1,3,5,NA
+67105,7,2,2,43,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,34954.173075,36984.739503,2,98,2,2,0.35,3,3,0,1,0,2,43,1,3,1,3
+67106,7,2,1,21,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,35338.972549,36508.861705,2,90,2,2,0.25,5,5,0,1,0,2,41,2,4,1,NA
+67107,7,2,1,77,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,9662.124837,10964.58979,2,95,2,2,0.6,1,1,0,0,1,1,77,1,3,3,NA
+67108,7,2,1,6,NA,1,1,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11881.117946,11935.814841,1,102,6,6,0.96,5,5,0,2,0,1,32,2,2,1,3
+67109,7,2,2,29,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,47348.206546,48810.979214,1,92,14,14,3.9,4,4,2,0,0,2,29,1,4,1,4
+67110,7,2,2,7,NA,4,4,1,7,90,NA,NA,2,1,3,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7140.212598,7289.797115,2,93,12,12,NA,4,4,1,1,0,2,27,2,4,1,4
+67111,7,2,1,11,NA,1,1,1,11,141,NA,NA,2,2,2,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,13870.762641,14200.45921,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+67112,7,2,2,64,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9581.703872,9981.600708,2,96,15,15,5,2,2,0,0,2,1,66,1,5,1,5
+67113,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,183805.817006,191979.640064,1,93,15,15,5,2,2,0,0,0,1,58,2,5,1,5
+67114,7,2,1,6,NA,5,6,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7241.695104,8178.415743,1,102,15,15,4.59,4,4,1,1,0,1,35,1,5,1,5
+67115,7,2,1,59,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,28247.128624,28689.709737,2,93,9,9,2.6,4,4,0,0,0,2,58,2,4,4,NA
+67116,7,2,2,31,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,43270.789623,48694.023669,2,99,10,10,5,2,1,0,0,0,2,31,1,5,5,NA
+67117,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,47973.37979,50072.031476,1,101,2,2,0.22,4,4,1,0,0,2,25,1,4,6,NA
+67118,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,88056.562286,93150.302158,2,91,6,6,1.34,4,4,1,2,0,2,33,1,4,3,NA
+67119,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,110369.721342,112542.856633,1,94,6,6,1.31,3,3,0,0,0,2,46,1,5,6,NA
+67120,7,1,2,30,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,19305.389921,0,1,93,10,10,5,1,1,0,0,0,2,30,1,5,5,NA
+67121,7,2,1,39,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19091.246741,19659.303383,1,91,5,5,0.89,4,4,2,0,0,1,39,1,4,1,5
+67122,7,2,1,17,NA,4,4,1,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12296.68269,12194.010773,1,102,7,7,2.72,2,2,0,1,0,2,49,1,4,5,NA
+67123,7,2,2,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,32501.623429,43787.968887,2,97,5,5,1.63,2,2,1,0,0,2,33,1,3,5,NA
+67124,7,2,2,8,NA,3,3,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23170.920553,22863.220381,1,94,4,4,1.06,2,2,0,1,0,2,26,1,4,5,NA
+67125,7,2,1,2,NA,2,2,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8331.647763,8595.381151,2,90,15,15,4.2,5,5,1,0,0,2,50,NA,NA,6,NA
+67126,7,2,2,22,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16239.242782,17454.103497,2,101,99,99,NA,3,1,0,0,0,2,22,1,4,5,NA
+67127,7,2,1,18,NA,4,4,1,19,228,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,14650.502937,14679.468181,2,102,14,14,3.25,5,5,1,1,0,2,32,1,4,1,3
+67128,7,2,2,47,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,18811.641937,20880.310604,2,90,3,3,1.04,1,1,0,0,0,2,47,2,4,5,NA
+67129,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18723.98095,17927.802584,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+67130,7,2,2,10,NA,2,2,1,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15352.601806,15746.071667,2,102,15,12,NA,5,4,0,3,0,1,42,2,4,6,NA
+67131,7,2,1,8,NA,3,3,1,8,103,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18834.739821,20852.023287,1,94,5,5,1.2,3,3,0,1,0,1,49,1,4,5,NA
+67132,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,36598.587714,38321.205688,1,97,7,7,1.74,4,4,0,3,0,2,32,1,4,5,NA
+67133,7,2,1,33,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32856.012738,32608.038024,2,103,77,77,NA,5,5,1,2,0,2,30,1,2,1,2
+67134,7,2,1,29,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,40333.47326,40883.249623,2,99,9,9,4.92,1,1,0,0,0,1,29,1,5,5,NA
+67135,7,2,2,58,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,1,2,NA,1,2,1,1,2,2,NA,NA,NA,NA,17277.829496,17117.256291,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+67136,7,2,2,11,NA,5,7,2,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18348.624469,20109.577944,2,90,77,77,NA,6,6,0,4,0,2,41,NA,NA,4,NA
+67137,7,2,1,67,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,8479.54701,8948.405799,2,98,6,6,2.04,2,2,0,0,2,2,66,1,1,1,1
+67138,7,2,2,8,NA,2,2,2,8,102,NA,NA,2,1,3,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15583.587534,16935.930722,1,93,15,15,5,4,4,0,2,0,1,50,1,5,1,5
+67139,7,2,2,0,1,1,1,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8907.98929,8720.575629,1,95,1,1,0,3,3,1,0,0,2,21,1,2,6,NA
+67140,7,1,2,12,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5268.765205,0,1,99,9,9,2.68,4,4,0,2,0,1,43,2,3,1,NA
+67141,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,72114.232586,74326.626155,2,96,12,12,NA,2,2,0,0,2,1,77,1,5,1,4
+67142,7,2,1,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6992.24593,7710.094516,2,97,1,1,0.33,2,2,1,0,0,2,24,1,2,5,NA
+67143,7,2,1,33,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,3,5,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,34624.133414,2,94,14,8,3.06,5,2,0,0,0,1,24,2,4,5,NA
+67144,7,2,1,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,13629.653758,14595.525499,1,90,15,15,5,5,5,0,3,0,2,46,2,4,1,5
+67145,7,2,2,0,9,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3320.8855,3567.679794,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+67146,7,2,2,7,NA,3,3,2,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24070.467912,25563.654853,1,95,6,6,1.09,5,5,0,3,0,1,31,1,4,1,4
+67147,7,2,1,1,22,4,4,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6347.215153,6541.084247,2,95,6,6,1.3,4,4,1,1,0,1,38,1,4,1,4
+67148,7,2,2,68,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,3,2,NA,2,2,2,2,2,2,1,2,2,2,9716.805546,10855.679158,2,90,2,2,0.64,1,1,0,0,1,2,68,2,3,2,NA
+67149,7,2,2,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,9943.806181,2,100,6,6,2.31,2,2,0,0,2,2,65,1,4,1,2
+67150,7,2,1,0,7,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6246.568228,6909.306675,2,95,1,1,0.13,2,2,1,0,0,2,22,1,3,5,NA
+67151,7,2,1,5,NA,4,4,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,10162.494907,1,100,1,1,0.26,2,2,1,0,0,2,28,1,3,5,NA
+67152,7,2,2,10,NA,1,1,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,14414.529053,14932.182215,2,96,3,3,0.46,5,5,1,2,0,1,37,1,1,1,2
+67153,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17538.193115,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+67154,7,2,1,4,NA,5,7,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10276.262805,11525.953064,1,97,15,15,4.77,4,4,1,1,0,1,41,2,4,1,5
+67155,7,2,2,2,NA,5,6,1,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7128.631964,7757.55827,3,91,15,15,5,4,4,1,1,0,1,39,2,5,1,5
+67156,7,2,2,10,NA,2,2,1,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12307.832776,13375.906079,2,93,10,10,2.26,6,6,0,4,0,1,34,1,4,1,3
+67157,7,2,2,18,NA,5,6,2,18,222,2,NA,2,1,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6937.463063,7435.682574,3,90,99,99,NA,4,4,0,1,0,1,40,2,3,6,NA
+67158,7,2,1,14,NA,2,2,1,14,178,NA,NA,2,2,3,9,NA,NA,NA,2,1,2,1,2,2,1,2,1,2,14117.225999,14174.690553,2,93,4,4,0.56,5,5,0,2,0,1,37,NA,NA,1,1
+67159,7,2,1,80,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,1,1,2,1,1,2,2,NA,11550.158096,12419.035396,2,92,5,5,1.08,3,3,0,0,2,1,46,NA,NA,5,NA
+67160,7,2,2,15,NA,4,4,2,15,185,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12209.74498,12444.407921,2,90,14,14,4.25,4,4,0,2,1,2,45,2,5,5,NA
+67161,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,48113.658256,0,1,90,5,5,2.15,1,1,0,0,1,2,80,1,3,2,NA
+67162,7,2,1,4,NA,5,6,2,4,54,NA,NA,2,2,1,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8402.098771,9423.873047,3,91,5,5,1.08,3,3,1,0,0,1,29,2,5,1,5
+67163,7,2,1,4,NA,1,1,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14505.510202,14964.673559,2,94,4,4,0.73,5,5,2,1,0,1,35,2,1,6,NA
+67164,7,2,2,13,NA,5,6,1,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10142.281747,10534.471105,3,91,15,15,5,4,4,0,2,0,1,44,2,5,1,5
+67165,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,35126.205635,35160.69801,1,101,2,2,0.66,1,1,0,0,0,2,46,1,4,3,NA
+67166,7,2,2,9,NA,2,2,2,9,116,NA,NA,2,1,3,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,16099.075547,16268.329655,1,93,5,5,0.84,5,5,1,2,0,2,52,2,1,3,NA
+67167,7,2,1,8,NA,2,2,2,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9390.522479,10327.334743,2,90,3,3,0.7,3,3,1,1,0,2,25,1,1,1,NA
+67168,7,2,2,11,NA,3,3,1,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18800.96526,18551.296274,1,94,3,3,0.37,5,5,0,3,0,2,29,1,4,4,NA
+67169,7,2,1,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,27572.205373,27179.185736,1,100,7,7,3.58,1,1,0,0,0,1,30,1,5,5,NA
+67170,7,2,1,13,NA,2,2,2,13,163,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16033.31661,17623.465787,2,90,5,5,0.76,5,5,0,4,0,2,32,1,2,3,NA
+67171,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,19060.786733,19662.793534,2,97,4,4,0.81,4,4,1,0,0,2,51,1,2,4,NA
+67172,7,2,1,5,NA,1,1,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19885.456648,19912.326587,1,92,10,10,3.04,4,4,2,0,0,2,37,2,5,1,5
+67173,7,2,2,21,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,1,2,2,1,19463.594852,21649.931723,3,92,6,6,1.6,3,3,0,0,1,2,77,1,3,99,NA
+67174,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,1,1,2,2,1,2,2,1,2,2,1,100264.332334,116996.29597,1,99,7,7,3.63,1,1,0,0,0,2,25,1,5,5,NA
+67175,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,31540.022655,31610.470938,2,94,NA,77,NA,2,1,0,0,0,1,49,1,3,3,NA
+67176,7,2,1,17,NA,4,4,2,17,206,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11834.781205,12120.23702,2,90,8,8,2.59,3,3,0,2,0,2,35,1,4,6,NA
+67177,7,2,2,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,66567.821082,69187.508251,1,94,3,3,0.88,2,2,0,0,1,2,74,1,2,2,NA
+67178,7,2,1,4,NA,1,1,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17865.135763,18430.646082,3,92,3,3,0.51,5,5,1,2,0,2,34,2,1,6,NA
+67179,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,NA,NA,NA,NA,30275.274308,28786.080652,2,101,2,2,0.51,1,1,0,0,0,2,20,1,4,5,NA
+67180,7,2,2,78,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,14534.533756,15507.238104,1,96,6,6,2.06,2,2,0,0,2,1,68,2,3,1,2
+67181,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,28721.124083,1,101,7,7,1.82,4,4,2,0,0,2,27,1,2,1,3
+67182,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,19130.246369,19076.4893,1,98,10,10,3.77,3,3,0,1,0,1,48,NA,NA,1,4
+67183,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21143.964074,20709.49715,1,100,15,15,3.7,5,5,0,3,0,1,51,1,5,1,5
+67184,7,2,1,42,NA,1,1,2,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,31640.296506,31176.023929,2,94,15,9,5,2,1,0,0,0,1,42,1,4,6,NA
+67185,7,2,2,51,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,1,1,2,2,NA,24178.173487,24760.288626,1,92,7,7,2.1,3,3,0,0,0,1,24,2,4,5,NA
+67186,7,2,1,6,NA,2,2,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11971.590477,11920.833512,2,100,14,14,3.58,4,4,1,1,0,1,33,1,4,1,5
+67187,7,2,2,15,NA,1,1,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12996.753859,13369.730018,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+67188,7,2,2,13,NA,4,4,1,13,161,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14166.687432,14623.20249,1,100,6,6,1.39,4,4,0,3,0,2,29,1,4,5,NA
+67189,7,2,2,70,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,3,2,NA,1,2,1,1,2,1,1,2,2,NA,12813.709202,14706.046404,2,103,4,4,1.43,1,1,0,0,1,2,70,2,3,2,NA
+67190,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,79987.938327,1,101,14,14,4.5,3,3,0,1,0,1,39,1,2,1,5
+67191,7,2,2,33,NA,3,3,1,NA,NA,2,NA,2,7,2,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,29733.812317,32603.623062,1,101,6,6,1.65,2,2,0,0,0,1,47,1,4,1,2
+67192,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126117.5094,130777.186989,2,98,8,8,2.7,3,3,0,0,1,2,71,NA,NA,2,NA
+67193,7,2,1,37,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18289.793332,19321.216647,1,99,8,8,1.76,5,5,0,2,1,1,37,1,4,1,3
+67194,7,2,1,51,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21762.20618,21775.974398,1,91,10,10,2.56,5,5,0,3,0,1,51,2,5,1,4
+67195,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,133542.212862,137840.368471,1,94,9,9,3.97,2,2,0,0,0,1,49,1,3,1,3
+67196,7,2,2,5,NA,1,1,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11859.546176,12772.78418,1,102,8,8,1.33,7,7,1,4,0,2,32,1,3,1,2
+67197,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,66253.989241,66961.54455,1,101,15,15,5,2,2,0,0,2,2,73,1,4,1,4
+67198,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32461.799549,32721.879701,1,101,3,3,0.66,2,2,0,0,1,1,67,1,2,3,NA
+67199,7,2,1,41,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15574.704266,15518.439557,2,92,15,15,5,2,2,0,0,0,2,36,1,5,1,5
+67200,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,97074.465027,101633.534325,1,93,15,8,4.41,3,1,0,0,0,1,30,1,5,5,NA
+67201,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,20953.00978,22383.074498,2,97,3,3,0.46,5,5,0,3,0,1,40,1,2,1,3
+67202,7,2,1,11,NA,5,6,2,11,133,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8423.270513,9127.622487,1,90,15,15,5,5,5,0,3,0,2,46,2,4,1,5
+67203,7,2,1,33,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17879.023129,20112.694262,2,90,5,5,2.02,1,1,0,0,0,1,33,2,3,5,NA
+67204,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,20333.447198,20233.229833,1,98,6,6,1.72,2,2,0,0,0,2,53,1,5,2,NA
+67205,7,2,1,5,NA,5,7,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8100.706553,8491.571292,1,98,14,14,3.9,4,4,2,0,0,1,39,1,5,1,4
+67206,7,2,1,8,NA,5,7,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21087.869274,22876.159598,1,101,5,5,1.26,3,3,0,1,1,2,42,1,3,5,NA
+67207,7,2,2,10,NA,4,4,2,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7136.421849,7489.546444,1,90,9,9,1.65,7,7,0,4,0,1,36,1,4,1,4
+67208,7,2,1,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7117.971973,7173.650989,2,100,6,6,2.31,2,2,0,0,2,2,65,1,4,1,2
+67209,7,2,2,80,NA,3,3,1,NA,NA,2,NA,2,1,9,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,51644.110977,57877.507418,1,102,5,5,1.84,1,1,0,0,1,2,80,2,5,2,NA
+67210,7,2,2,63,NA,4,4,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,2,1,1,1,NA,10585.751811,11058.394307,2,93,6,6,1.13,4,4,0,0,2,1,60,2,3,1,3
+67211,7,2,1,3,NA,4,4,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8973.990262,9895.291699,2,95,3,3,0.43,4,4,2,0,0,2,23,1,2,5,NA
+67212,7,1,2,67,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,125706.50355,0,3,91,14,14,5,2,1,0,0,1,2,67,1,5,5,NA
+67213,7,2,1,4,NA,4,4,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,10259.663981,2,97,1,1,0.27,2,2,1,0,0,2,20,1,2,5,NA
+67214,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,99283.360764,103932.649449,3,91,14,14,5,1,1,0,0,0,1,39,1,5,5,NA
+67215,7,2,1,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,2,2,2,1,2,2,1,40899.412129,44211.773156,1,94,5,5,0.57,7,7,2,1,0,1,58,2,1,1,1
+67216,7,2,2,0,7,1,1,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6744.839779,7193.080406,1,103,5,5,1.02,4,4,2,0,0,1,25,1,2,1,4
+67217,7,2,1,41,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,37402.70356,39087.586144,2,102,7,7,1.89,3,3,0,1,0,1,41,2,2,6,NA
+67218,7,2,1,0,8,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13392.309303,13804.825901,1,99,15,15,5,5,5,3,0,0,2,34,1,5,1,5
+67219,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,61212.189461,64753.083655,3,90,15,15,5,3,3,1,0,0,1,31,1,5,1,5
+67220,7,2,1,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,106105.629898,111686.437619,1,100,14,9,4.92,3,1,1,0,0,1,33,1,5,5,NA
+67221,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,17293.36291,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+67222,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,62796.637452,70136.543572,2,102,6,6,1.15,5,5,0,0,2,1,80,1,5,1,1
+67223,7,2,1,40,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18898.73153,22670.159935,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+67224,7,2,1,2,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6760.288309,6966.774293,1,96,99,99,NA,4,4,1,1,0,2,35,2,3,1,3
+67225,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,8318.523894,9127.657215,3,90,3,3,1.07,1,1,0,0,1,1,80,1,5,2,NA
+67226,7,2,2,43,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,14831.744106,14871.922635,3,90,5,5,0.87,4,4,0,0,0,2,43,2,3,5,NA
+67227,7,2,2,17,NA,4,4,2,17,208,2,NA,2,2,4,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11072.776368,11285.58755,3,90,9,9,4.1,2,2,0,1,0,2,41,2,5,5,NA
+67228,7,2,1,41,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,14823.307875,15546.691363,1,96,8,8,2.62,3,3,0,1,0,1,41,2,3,1,5
+67229,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,138075.879417,141933.339512,2,91,7,7,1.61,4,4,0,0,3,1,65,1,3,6,NA
+67230,7,2,1,59,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,25969.864445,26997.658596,1,97,15,15,5,2,2,0,0,0,1,59,1,5,1,5
+67231,7,2,1,18,NA,4,4,1,18,219,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16147.713323,16532.569027,1,92,5,5,0.95,4,4,0,2,0,2,33,1,4,5,NA
+67232,7,2,2,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,35933.117795,36141.582386,1,98,1,1,0.18,1,1,0,0,0,2,48,1,5,1,NA
+67233,7,2,2,7,NA,4,4,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8757.841043,9034.425418,2,97,2,2,0.49,3,3,0,1,0,2,27,1,4,5,NA
+67234,7,2,1,12,NA,1,1,1,12,149,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20705.832333,21431.576705,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+67235,7,2,1,6,NA,4,4,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14125.862146,14295.209168,1,97,5,5,0.91,4,4,0,3,0,2,44,1,4,5,NA
+67236,7,2,2,48,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,43535.993088,43763.029589,2,102,77,77,NA,3,3,0,0,0,2,48,2,2,1,3
+67237,7,2,2,51,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,23953.14388,24078.057488,1,90,12,12,NA,4,4,0,0,0,1,54,2,4,1,2
+67238,7,2,2,15,NA,5,6,1,15,189,NA,NA,2,1,4,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7382.152016,8028.485773,1,94,14,14,2.78,6,5,0,2,1,1,61,1,4,1,5
+67239,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,1,4,NA,2,2,2,2,2,1,NA,NA,NA,NA,24004.6026,28934.42641,2,91,4,4,0.94,3,3,0,1,0,2,50,2,1,4,NA
+67240,7,2,2,5,NA,1,1,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19235.084509,21236.049482,3,92,8,8,2,4,4,2,0,0,1,30,1,4,1,4
+67241,7,1,1,80,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,99,NA,1,1,2,1,2,2,NA,NA,NA,NA,14359.447628,0,1,97,4,4,0.81,3,3,0,0,2,1,80,1,3,99,NA
+67242,7,2,1,2,NA,4,4,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8009.966208,8832.297541,2,96,6,6,1.35,3,3,1,1,0,2,23,1,2,5,NA
+67243,7,2,2,68,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,7869.59899,8538.749518,1,96,3,3,0.68,2,2,0,0,1,2,68,1,2,3,NA
+67244,7,2,2,13,NA,3,3,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28650.637159,30356.286382,2,100,2,2,0.38,3,3,0,2,0,2,35,1,4,5,NA
+67245,7,2,1,27,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,2,5,NA,1,2,2,1,2,1,1,2,2,1,9177.295801,9548.31812,2,92,77,77,NA,4,4,0,0,0,1,27,2,2,5,NA
+67246,7,2,1,15,NA,3,3,2,15,185,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,68701.580401,72973.564721,2,94,15,15,5,5,5,0,2,0,1,53,1,5,1,5
+67247,7,2,1,8,NA,5,6,2,8,102,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8151.552109,8792.161653,1,90,14,14,3.33,5,5,1,2,0,1,41,1,5,1,5
+67248,7,2,2,14,NA,1,1,1,14,174,NA,NA,2,2,4,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15690.47168,16407.081535,1,102,5,5,0.62,7,7,1,3,0,1,49,2,2,1,1
+67249,7,2,2,58,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16663.192678,16751.306683,2,92,NA,4,1.65,2,1,0,0,1,2,58,2,4,5,NA
+67250,7,2,2,43,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,42424.587753,44729.774642,2,102,4,4,1.38,2,1,0,0,0,2,43,1,5,6,NA
+67251,7,2,2,7,NA,4,4,2,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,1,2,NA,NA,NA,NA,7814.742747,7984.793423,2,95,2,2,0.26,4,4,0,2,0,2,44,NA,NA,5,NA
+67252,7,2,2,24,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16844.740449,17983.530016,3,91,3,3,1.1,1,1,0,0,0,2,24,1,5,5,NA
+67253,7,2,2,10,NA,3,3,2,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,44941.581862,44344.776378,2,94,10,10,2.91,4,4,0,2,0,2,38,1,4,1,4
+67254,7,2,1,8,NA,1,1,1,9,108,NA,NA,2,2,2,2,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,13870.762641,13985.149896,2,102,6,4,1.02,6,2,0,4,0,2,43,2,1,5,NA
+67255,7,2,2,38,NA,3,3,2,NA,NA,2,NA,2,1,4,NA,3,1,2,1,2,2,1,2,2,1,1,2,NA,53830.599426,56664.989337,1,99,77,77,NA,4,4,1,1,0,1,31,2,3,1,3
+67256,7,2,1,10,NA,1,1,1,11,132,NA,NA,2,2,3,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15456.524105,16140.282926,1,100,99,99,NA,6,6,0,1,0,2,22,2,3,1,3
+67257,7,2,2,9,NA,5,6,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8556.454265,9137.012028,1,90,15,15,5,5,5,0,2,0,1,47,2,5,1,5
+67258,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,54256.870337,62240.534341,2,94,6,6,2.04,2,2,0,0,2,1,75,1,3,1,3
+67259,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,90081.78624,90194.909743,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+67260,7,2,1,3,NA,3,3,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27542.701065,31074.71228,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+67261,7,2,1,57,NA,3,3,2,NA,NA,2,NA,2,1,99,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,20160.790176,25538.533486,3,90,77,99,NA,2,1,0,0,1,1,62,2,5,3,NA
+67262,7,1,2,68,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,1,1,2,1,NA,NA,NA,NA,9793.924718,0,2,100,2,2,0.39,3,3,0,0,1,2,45,1,2,5,NA
+67263,7,2,2,3,NA,1,1,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15979.952759,16987.381382,2,102,6,6,1.03,5,5,1,1,0,1,37,1,2,1,2
+67264,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,81062.798322,86779.337922,1,101,4,4,1.61,1,1,0,0,1,2,77,1,2,2,NA
+67265,7,2,2,0,5,4,4,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4681.626184,4732.744026,1,96,6,6,1.31,4,3,1,0,0,2,27,2,4,1,3
+67266,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,51543.062078,51154.050295,3,92,7,7,1.3,5,5,1,2,0,2,33,2,2,1,1
+67267,7,1,1,10,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10524.718724,0,1,96,77,77,NA,2,2,0,1,0,2,35,1,3,3,NA
+67268,7,2,2,18,NA,3,3,2,18,217,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,100004.689924,107460.712122,1,90,12,12,NA,2,2,0,0,0,2,45,2,3,4,NA
+67269,7,2,1,59,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21762.20618,22687.256739,1,91,15,15,5,4,4,0,1,0,2,51,2,4,1,5
+67270,7,2,2,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,34874.122648,36144.248604,3,92,6,6,1.41,4,3,0,1,0,1,41,1,4,1,4
+67271,7,2,2,4,NA,4,4,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8833.042831,9629.276723,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+67272,7,2,2,11,NA,3,3,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18417.272091,18465.044105,1,94,5,5,1.04,4,4,0,2,0,2,29,1,3,1,3
+67273,7,2,1,44,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,31740.385214,31855.807763,2,96,5,5,0.78,5,5,0,2,0,1,37,2,1,5,NA
+67274,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,19150.604366,19251.871661,3,91,14,9,2.68,6,4,0,0,2,1,48,2,1,1,1
+67275,7,2,1,77,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,16603.347511,18119.902269,2,98,3,3,0.88,2,2,0,0,2,1,77,1,1,1,3
+67276,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,3,NA,2,2,2,2,2,2,1,2,2,2,9716.805546,10855.679158,2,90,7,7,1.34,5,5,0,1,2,1,61,2,1,4,NA
+67277,7,2,2,4,NA,5,6,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5307.591514,5454.53335,2,102,3,3,0.38,5,5,3,0,0,2,30,2,2,1,4
+67278,7,2,1,2,NA,2,2,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,1,1,0.14,3,2,2,0,0,2,27,1,3,6,NA
+67279,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,21765.629914,22790.091994,2,90,2,2,0.38,4,4,1,2,0,2,32,1,4,5,NA
+67280,7,2,1,27,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37911.437415,38428.199561,2,103,12,6,2.24,2,1,0,0,0,1,27,1,4,5,NA
+67281,7,2,2,42,NA,1,1,2,NA,NA,2,NA,2,2,77,NA,2,6,2,2,2,2,2,2,2,2,2,2,2,31235.666551,35013.570975,2,94,77,77,NA,3,3,0,1,0,2,42,2,2,6,NA
+67282,7,2,2,8,NA,1,1,1,8,102,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,13028.403003,13395.598637,2,96,6,6,1.11,5,5,0,3,0,2,32,2,3,1,2
+67283,7,2,2,56,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12441.719186,12605.354834,1,96,15,15,5,5,5,0,0,0,1,58,2,5,1,5
+67284,7,2,2,12,NA,5,6,1,12,147,NA,NA,2,2,1,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6370.316505,6496.587233,2,94,8,8,1.8,6,6,0,1,2,1,74,2,5,1,5
+67285,7,2,2,70,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,22166.463682,23916.215243,3,91,6,6,2.89,1,1,0,0,1,2,70,2,5,2,NA
+67286,7,2,2,6,NA,4,4,2,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6655.097829,7106.52929,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+67287,7,2,2,10,NA,5,6,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7932.110938,8316.610099,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+67288,7,2,1,24,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,17446.328105,17921.312041,1,92,15,15,5,2,2,0,0,0,2,49,1,5,3,NA
+67289,7,2,1,5,NA,5,6,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7042.228842,7898.630137,2,103,14,14,3.47,4,4,1,0,1,1,47,2,5,1,5
+67290,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,34099.599202,35276.751365,1,94,7,1,0.09,5,1,0,0,0,1,46,1,4,1,4
+67291,7,2,2,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,35160.494695,36239.184013,1,100,4,4,1.16,2,2,0,0,2,1,74,1,3,1,5
+67292,7,2,1,8,NA,4,4,1,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9022.8939,9530.778148,2,100,6,6,0.99,5,5,0,3,0,2,40,1,3,1,3
+67293,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,132969.642582,137509.365878,2,91,15,15,5,4,4,0,2,0,1,53,1,5,1,4
+67294,7,1,2,44,NA,2,2,NA,NA,NA,2,NA,2,1,5,NA,3,1,3,1,2,2,1,2,2,NA,NA,NA,NA,32606.880052,0,2,93,6,6,1.48,4,4,0,1,0,1,53,2,2,1,3
+67295,7,2,2,4,NA,5,6,2,4,52,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,4932.827461,1,91,14,14,4.32,3,3,1,0,0,1,34,2,5,1,3
+67296,7,2,2,15,NA,4,4,2,15,182,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12265.19283,12230.632995,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+67297,7,2,1,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,123110.069898,136669.986833,1,101,8,4,1.34,2,1,0,0,0,2,20,1,4,6,NA
+67298,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,17377.366864,16840.035727,2,99,6,6,1.11,5,5,1,2,0,2,41,1,2,5,NA
+67299,7,2,2,7,NA,2,2,2,7,84,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15148.721588,16225.089736,2,91,2,2,0.42,3,3,1,1,0,2,27,1,3,5,NA
+67300,7,2,1,72,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,71709.98031,77299.00621,2,91,9,9,2.6,4,4,0,0,2,1,72,1,4,1,5
+67301,7,1,2,18,NA,3,3,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,45824.798023,0,1,97,1,1,0.09,2,1,0,0,0,1,24,1,3,6,NA
+67302,7,2,2,3,NA,4,4,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16749.124902,18717.057678,2,97,1,1,0.13,4,4,2,0,1,2,62,1,2,4,NA
+67303,7,2,1,8,NA,3,3,2,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,82670.203859,85919.643529,1,97,15,15,5,5,5,0,3,0,2,47,2,5,1,5
+67304,7,2,2,12,NA,2,2,1,12,151,NA,NA,2,2,2,5,NA,NA,NA,2,1,2,2,2,2,1,2,1,2,12712.538972,13945.896408,2,93,6,6,0.93,5,5,1,2,0,1,40,2,4,1,4
+67305,7,2,2,20,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,3,1,0.22,3,1,0,0,0,2,20,2,4,5,NA
+67306,7,2,1,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,19186.370211,20284.349464,2,97,2,2,0.27,4,4,0,2,0,1,51,1,2,4,NA
+67307,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,21022.682584,21611.637105,1,100,3,3,0.43,4,4,2,0,0,1,20,1,3,6,NA
+67308,7,2,1,21,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,10131.732483,10407.573345,2,103,12,3,1.07,5,1,0,1,0,2,47,NA,NA,3,NA
+67309,7,2,1,72,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,2,2,1,2,2,2,2,2,2,NA,15435.262366,15704.333986,2,93,9,9,3.14,3,3,0,0,2,1,43,NA,NA,5,NA
+67310,7,2,1,46,NA,2,2,1,NA,NA,2,NA,2,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,49060.272708,48961.70392,1,92,14,5,1.88,3,1,0,1,0,2,39,2,4,6,NA
+67311,7,2,2,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,50915.06085,52488.027702,3,92,8,8,2,4,4,2,0,0,1,30,1,4,1,4
+67312,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,160001.813639,1,95,9,9,2.66,4,4,0,2,0,1,45,1,3,1,3
+67313,7,2,1,11,NA,1,1,1,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17882.621856,18280.794545,3,92,8,8,1.55,6,6,1,3,0,2,38,1,5,1,4
+67314,7,2,1,62,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,40561.050563,40941.458273,2,101,3,3,0.98,2,2,0,0,2,1,62,1,4,1,3
+67315,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15890.452621,2,100,7,7,1.79,4,4,0,1,0,2,51,1,3,3,NA
+67316,7,2,2,48,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,32208.300114,33976.532975,1,102,6,6,1.34,4,4,0,1,0,2,48,2,3,1,1
+67317,7,2,1,0,10,4,4,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6275.847063,6372.172481,2,100,14,14,3.58,4,4,1,1,0,2,31,1,5,1,4
+67318,7,2,1,26,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,12364.328404,12075.509308,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+67319,7,2,1,6,NA,4,4,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8065.706636,8519.712345,2,99,1,1,0.03,3,3,0,1,0,2,56,1,3,5,NA
+67320,7,2,1,19,NA,2,2,2,19,230,2,NA,1,1,NA,12,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,14081.782012,14782.02856,2,90,7,7,1.34,5,5,0,1,2,1,61,2,1,4,NA
+67321,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,24930.322327,24899.923342,2,91,4,4,1.22,2,2,0,0,0,1,53,1,4,1,4
+67322,7,2,1,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,NA,NA,NA,1,2,2,1,20308.910079,25155.898326,1,90,NA,NA,NA,2,2,0,0,0,2,40,NA,NA,5,NA
+67323,7,2,2,10,NA,1,1,1,10,122,NA,NA,2,2,3,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20495.125801,21027.07407,3,92,3,3,0.52,5,5,2,1,0,2,29,2,1,1,3
+67324,7,2,1,7,NA,5,6,2,7,95,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7193.657603,7758.988679,2,90,14,14,4.32,3,3,0,1,0,1,48,2,4,1,5
+67325,7,2,2,15,NA,1,1,1,15,182,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16781.078148,17658.768186,2,103,77,77,NA,5,5,0,2,0,2,45,2,4,5,NA
+67326,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,24126.537834,25148.058527,2,97,5,5,0.76,5,5,1,1,0,2,47,1,4,5,NA
+67327,7,2,1,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53149.251154,62614.383127,2,94,6,6,2.04,2,2,0,0,2,1,75,1,3,1,3
+67328,7,2,2,6,NA,5,6,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9007.62445,9504.796896,2,102,15,15,3.92,5,5,1,2,0,1,34,2,5,1,5
+67329,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,13934.943848,14557.124186,2,96,3,3,0.47,4,4,1,0,1,2,61,1,4,3,NA
+67330,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,108408.375382,114958.410267,2,98,6,6,1.25,4,4,1,0,1,1,46,1,2,6,NA
+67331,7,2,2,15,NA,4,4,2,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14417.957956,14453.467189,1,96,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+67332,7,2,2,45,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,1,1,2,1,1,2,1,3,13749.025709,14311.068456,1,103,7,7,1.89,3,3,0,1,0,2,45,2,5,5,NA
+67333,7,2,1,42,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17602.101156,17988.968846,1,96,7,7,1.52,4,4,0,2,0,2,30,2,4,1,5
+67334,7,1,2,32,NA,1,1,NA,NA,NA,2,NA,2,2,3,NA,3,1,3,2,2,2,2,2,2,NA,NA,NA,NA,35353.005268,0,2,94,4,4,0.72,4,4,1,1,0,1,30,2,1,1,3
+67335,7,2,2,30,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,26074.980194,26960.560427,1,93,12,2,0.69,4,1,0,0,0,2,25,2,5,5,NA
+67336,7,2,2,7,NA,5,6,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9054.387575,9459.329211,2,100,14,14,4.03,4,4,0,2,0,1,48,2,5,1,5
+67337,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,64463.340883,65151.773075,1,91,15,14,5,3,2,0,0,2,2,73,1,4,3,NA
+67338,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15971.744676,15898.343344,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+67339,7,2,1,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,86445.94283,91501.124576,2,102,9,9,4.08,2,2,0,0,2,1,70,1,5,1,5
+67340,7,2,2,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,15521.115746,15096.339697,2,100,5,5,1.07,4,4,0,1,0,2,36,1,3,5,NA
+67341,7,2,1,74,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,14472.233702,14665.217659,1,98,8,8,3.3,2,2,0,0,2,1,74,1,1,1,3
+67342,7,2,2,48,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,12770.403552,13002.937616,1,103,4,4,0.82,3,3,0,0,0,1,48,2,4,1,1
+67343,7,2,1,58,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,29903.342494,29807.964765,1,95,6,4,1.74,2,1,0,0,0,2,48,1,4,3,NA
+67344,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,2,2,2,1,2,2,1,2,2,1,9691.985299,9847.238527,1,93,5,5,1.43,2,2,0,0,1,1,60,2,5,1,4
+67345,7,2,2,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6555.231326,6846.316857,1,99,6,6,1.35,3,3,1,1,0,2,42,1,4,4,NA
+67346,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,40859.270352,45390.978638,2,101,12,12,NA,1,1,0,0,1,1,80,1,3,3,NA
+67347,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,38321.717684,40488.857872,1,92,6,6,2.04,2,2,0,0,2,2,71,1,4,1,1
+67348,7,2,1,16,NA,5,6,2,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6666.045669,7317.485505,3,90,77,77,NA,5,5,0,2,0,1,46,2,3,1,3
+67349,7,2,1,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23160.426103,23797.470648,2,98,10,10,3.78,3,3,0,0,0,2,46,1,4,1,4
+67350,7,2,1,2,NA,4,4,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8431.543376,9297.155306,2,96,14,14,3.58,4,4,2,0,0,1,36,1,4,1,5
+67351,7,2,1,10,NA,5,7,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20857.728985,21677.56405,2,98,3,3,0.38,5,5,0,4,0,2,39,1,4,5,NA
+67352,7,2,2,40,NA,3,3,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,25784.374025,27218.997103,1,90,7,7,2.1,3,3,1,0,0,2,40,2,5,1,4
+67353,7,2,1,61,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,8883.460927,9371.520203,2,92,15,15,5,2,2,0,0,2,1,61,1,5,1,5
+67354,7,1,2,18,NA,5,6,NA,NA,NA,2,NA,2,1,4,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,9758.609905,0,3,91,14,2,0.83,2,1,0,0,1,2,67,1,5,5,NA
+67355,7,2,2,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,32455.694722,32314.335903,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+67356,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,80418.665565,96911.831087,2,97,15,15,4.97,5,5,1,0,0,1,48,1,4,1,3
+67357,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,36154.496177,37370.445999,2,101,1,1,0.14,3,1,0,0,0,1,23,1,4,5,NA
+67358,7,2,1,7,NA,4,4,2,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9554.261994,9836.582126,2,90,8,8,1.67,6,6,1,1,0,1,52,1,3,1,5
+67359,7,2,1,74,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,59027.291951,62692.518829,1,103,15,15,5,2,2,0,0,2,1,74,1,5,1,5
+67360,7,2,1,10,NA,1,1,1,10,131,NA,NA,2,2,4,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11159.151566,11226.781631,1,102,5,5,0.98,4,4,1,1,0,2,42,2,2,6,NA
+67361,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,32185.399379,40770.619238,2,91,4,4,1.19,2,2,0,0,0,1,59,1,1,1,3
+67362,7,2,2,47,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,13219.753222,13289.658517,2,92,8,8,1.91,5,5,0,2,1,2,47,2,1,1,3
+67363,7,2,1,61,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,4,NA,2,2,2,2,2,2,1,2,2,2,8609.250304,9380.869295,2,90,7,7,1.34,5,5,0,1,2,1,61,2,1,4,NA
+67364,7,2,2,19,NA,3,3,1,19,231,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22916.593117,23588.294062,1,102,4,4,0.97,3,3,0,1,0,2,19,1,2,NA,NA
+67365,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16033.091438,16307.204086,1,96,15,15,5,3,3,0,0,1,2,62,1,4,3,NA
+67366,7,2,2,5,NA,3,3,1,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64434.313673,68475.614831,1,101,5,5,0.89,5,5,1,2,0,1,31,1,2,1,1
+67367,7,2,2,32,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,1,2,2,2,2,2,2,2,NA,NA,NA,NA,35464.8385,41544.401264,2,96,6,6,1.11,5,5,0,3,0,2,32,2,3,1,2
+67368,7,2,2,23,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,3,5,2,2,2,2,2,2,2,1,2,2,1,41646.508203,41998.067158,3,91,4,4,0.81,3,3,0,0,1,1,64,2,1,1,1
+67369,7,2,2,3,NA,4,4,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9445.424546,10163.382174,2,99,6,6,1.15,5,5,1,2,0,2,34,1,4,77,NA
+67370,7,2,1,41,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,1,1,2,1,1,2,2,3,19184.316833,20543.822351,1,97,15,15,5,4,4,2,0,0,2,35,2,4,1,4
+67371,7,2,1,13,NA,4,4,1,13,163,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9188.45337,9407.445906,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+67372,7,2,2,0,8,1,1,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6588.463421,6449.849868,2,95,12,6,1.98,4,2,1,0,0,2,25,1,4,77,NA
+67373,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,10746.81867,11542.921605,2,100,99,99,NA,6,6,1,1,2,1,37,2,3,1,3
+67374,7,1,1,10,NA,5,6,NA,NA,NA,NA,NA,2,1,3,4,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,8014.738552,0,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+67375,7,2,2,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,18490.479848,18318.636758,2,100,7,7,1.38,5,5,1,0,0,2,45,1,2,3,NA
+67376,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,38254.514187,37933.750372,1,98,2,2,0.72,1,1,0,0,0,2,57,1,4,3,NA
+67377,7,2,1,0,2,3,3,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9570.577309,9932.368407,1,95,2,2,0.18,6,6,1,2,2,2,69,1,2,1,2
+67378,7,2,1,9,NA,2,2,2,9,108,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,1,NA,NA,NA,NA,13217.721247,13882.368076,2,91,4,4,0.94,3,3,0,1,0,2,50,2,1,4,NA
+67379,7,2,2,12,NA,2,2,1,12,148,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,23968.380373,25913.276293,2,91,14,7,2.91,3,2,0,1,0,1,40,2,3,6,NA
+67380,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16165.962054,15723.538094,2,99,12,5,1.88,6,1,0,0,0,2,57,1,5,2,NA
+67381,7,2,1,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32680.7991,1,95,2,2,0.74,1,1,0,0,0,1,50,1,4,3,NA
+67382,7,2,2,18,NA,4,4,1,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18905.695776,2,101,4,4,0.86,3,3,0,1,0,2,18,1,2,NA,NA
+67383,7,2,2,13,NA,4,4,2,13,165,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10401.399206,10736.579564,2,99,9,9,2.43,4,4,0,2,0,2,49,1,3,3,NA
+67384,7,2,1,19,NA,3,3,2,20,NA,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,99249.131685,98501.502244,1,101,14,14,3.25,4,4,0,1,0,1,48,1,4,1,2
+67385,7,2,1,47,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18116.816149,18405.984579,2,90,99,4,1.16,5,2,0,0,1,1,43,NA,NA,1,NA
+67386,7,2,2,46,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,1,2,2,NA,NA,NA,NA,27654.660303,28156.17371,2,103,77,77,NA,4,4,0,0,0,2,46,2,1,4,NA
+67387,7,1,1,74,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,8601.453077,0,2,97,4,4,0.53,7,7,0,2,2,1,74,1,2,1,2
+67388,7,2,1,35,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,2,2,2,2,54095.173154,55399.110059,1,92,10,10,4.63,2,2,0,0,0,1,35,2,5,1,4
+67389,7,2,1,62,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,2,NA,2,2,2,2,2,2,1,2,2,2,10585.549682,13806.559103,1,90,13,13,NA,2,2,0,0,1,1,62,2,1,2,NA
+67390,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,121463.203322,122574.38549,2,95,9,9,4.92,1,1,0,0,0,2,56,1,5,3,NA
+67391,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,73849.700988,79023.338367,3,91,14,14,5,3,3,1,0,0,2,30,1,4,1,5
+67392,7,2,1,41,NA,2,2,1,NA,NA,2,NA,2,2,1,NA,5,1,NA,2,2,2,2,2,2,2,2,2,2,33029.272844,32544.619181,2,93,9,9,1.49,7,7,0,3,0,2,41,2,5,1,5
+67393,7,2,2,29,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,5,2,1,2,2,2,2,2,1,2,2,1,35424.746838,35546.372478,2,93,10,10,3.67,3,3,0,0,0,2,56,2,4,6,NA
+67394,7,2,2,8,NA,5,7,2,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8746.306586,9472.024073,1,91,3,3,0.66,4,4,1,2,0,2,33,1,3,5,NA
+67395,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16741.034883,16850.233711,2,95,3,3,0.66,2,2,0,0,0,2,55,1,2,1,NA
+67396,7,2,1,4,NA,1,1,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16118.799226,16847.423074,3,92,4,4,0.46,7,7,1,2,0,2,31,2,2,1,1
+67397,7,2,1,20,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,12364.328404,12075.509308,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+67398,7,2,2,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5847.515059,6374.625564,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+67399,7,2,2,55,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11428.6913,11636.794359,2,103,7,7,1.65,4,4,0,1,1,2,55,2,3,1,NA
+67400,7,2,1,61,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,9355.567184,9645.104203,2,100,14,14,3.06,5,5,1,0,1,2,31,2,5,1,5
+67401,7,2,1,5,NA,1,1,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17865.135763,18076.339981,3,92,1,1,0,5,5,3,0,0,1,26,1,2,1,2
+67402,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,16181.169973,15883.791498,2,100,7,7,2.72,2,2,0,0,0,2,59,1,4,3,NA
+67403,7,2,1,15,NA,1,1,1,15,191,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,24325.638415,24460.075857,3,92,5,5,1.05,3,3,1,1,0,2,38,2,3,5,NA
+67404,7,1,2,63,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,18002.759054,0,2,102,5,5,1.36,2,2,0,0,2,1,70,2,5,1,4
+67405,7,2,1,5,NA,1,1,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11388.091908,11522.723578,2,103,5,5,0.89,5,5,1,3,0,2,34,2,1,99,NA
+67406,7,2,1,48,NA,1,1,2,NA,NA,2,NA,2,2,2,NA,5,5,NA,2,2,2,1,2,2,1,2,2,2,36475.684097,37047.191703,1,93,15,15,5,3,2,0,1,0,2,46,2,5,5,NA
+67407,7,2,1,64,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,153565.050575,166235.02721,1,97,15,15,5,3,3,0,0,1,1,64,1,3,1,3
+67408,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,14647.161643,15007.506342,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+67409,7,2,2,45,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,3,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,123440.03559,156711.394658,1,90,12,12,NA,2,2,0,0,0,2,45,2,3,4,NA
+67410,7,2,2,18,NA,1,1,2,18,220,2,NA,2,2,2,11,NA,NA,NA,2,2,2,1,2,2,2,2,2,2,19557.287652,19949.093379,2,94,99,99,NA,3,3,1,0,0,2,18,2,2,NA,NA
+67411,7,2,1,43,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,6,NA,2,2,2,1,2,2,2,2,2,2,27127.17615,26729.126657,3,90,7,7,1.48,5,5,0,1,0,1,43,2,1,6,NA
+67412,7,2,1,77,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,1,1,1,1,2,1,2,1,NA,17775.981268,18879.758897,1,92,4,4,1.1,2,2,0,0,2,2,69,2,4,1,4
+67413,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,32485.015568,36087.933902,2,101,6,6,1.62,3,3,0,0,2,1,80,1,3,1,3
+67414,7,2,2,77,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,13045.741791,13445.321496,1,99,3,3,1.07,1,1,0,0,1,2,77,1,2,2,NA
+67415,7,1,1,5,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,9016.053035,0,2,100,NA,NA,NA,3,3,1,0,0,1,29,NA,NA,1,NA
+67416,7,2,1,4,NA,4,4,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,9417.927146,2,97,2,2,0.45,3,3,1,0,0,1,24,1,2,1,3
+67417,7,2,2,58,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19329.435508,19798.346555,2,97,15,15,5,2,2,0,0,1,2,58,2,5,1,5
+67418,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,39416.921733,1,95,2,2,0.87,1,1,0,0,1,2,80,1,2,2,NA
+67419,7,2,1,4,NA,4,4,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10291.033925,11347.547701,1,100,14,14,4.45,3,3,1,0,0,1,33,1,4,1,4
+67420,7,2,1,22,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,49741.714519,52356.235161,2,91,14,14,4.71,3,3,0,0,0,1,28,2,1,1,2
+67421,7,2,2,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16664.698857,2,98,9,9,3.08,3,3,0,1,0,1,35,2,1,1,3
+67422,7,2,1,19,NA,3,3,2,19,238,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,83180.831932,83381.692822,2,100,14,14,3.06,5,5,0,0,0,1,55,1,5,1,5
+67423,7,2,2,3,NA,3,3,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25881.48305,27865.988236,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+67424,7,2,2,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,42468.064168,46720.900929,2,101,3,3,0.65,3,3,0,1,0,2,54,1,3,5,NA
+67425,7,2,1,37,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,2,2,2,1,2,2,1,2,2,1,41241.224595,41716.316195,2,102,6,6,1.03,5,5,1,1,0,1,37,1,2,1,2
+67426,7,1,2,24,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,3,1,2,2,1,2,2,NA,NA,NA,NA,112992.533921,0,1,91,8,8,2.51,3,3,1,0,0,2,24,1,4,1,3
+67427,7,2,1,24,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,14977.822328,3,91,15,15,5,3,3,0,0,2,2,61,1,5,1,5
+67428,7,1,1,72,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,68460.271241,0,1,102,9,9,4.08,2,2,0,0,1,1,44,1,3,3,NA
+67429,7,2,2,65,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,2,6,NA,2,2,2,1,2,2,1,2,2,2,9716.805546,12994.252166,2,90,4,4,1.47,1,1,0,0,1,2,65,2,2,6,NA
+67430,7,2,1,8,NA,5,6,2,8,99,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10105.916709,10900.114748,1,91,6,6,1.25,4,4,1,1,0,1,26,2,4,6,NA
+67431,7,2,1,61,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,1,NA,NA,NA,1,2,1,3,5526.665646,5830.301859,2,92,12,NA,NA,7,1,0,0,2,1,53,2,3,1,3
+67432,7,2,2,10,NA,4,4,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11195.065587,11638.702248,2,96,2,2,0.43,3,3,0,2,0,2,50,1,2,5,NA
+67433,7,2,1,65,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,5,3,NA,2,2,2,NA,NA,NA,2,2,2,2,13473.930124,13736.530204,2,98,4,4,0.48,6,6,2,0,2,2,65,2,1,2,NA
+67434,7,2,1,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,184986.088848,184760.524604,1,92,15,15,5,2,2,0,0,0,2,56,1,4,1,5
+67435,7,2,1,80,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,1,NA,12101.489198,12967.543396,1,93,3,3,0.82,2,2,0,0,2,1,80,2,5,1,1
+67436,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,19783.396474,19410.189178,2,92,14,4,1.38,2,1,0,0,0,1,29,1,5,6,NA
+67437,7,2,2,0,2,3,3,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16423.151355,15977.062092,1,99,9,9,3.24,3,3,1,0,0,1,29,1,5,1,5
+67438,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,6943.972932,7268.809657,2,92,5,5,1.32,2,2,0,0,1,2,51,1,2,1,2
+67439,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21619.283038,21663.427812,2,102,8,8,1.72,5,5,0,2,1,1,63,2,5,1,5
+67440,7,1,2,61,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,16913.219568,0,2,98,77,77,NA,1,1,0,0,1,2,61,1,4,2,NA
+67441,7,2,1,80,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,36829.543424,39682.305625,1,93,15,15,5,2,2,0,0,2,2,75,1,5,1,5
+67442,7,2,2,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,28187.999204,29187.28099,1,99,7,7,2.72,2,2,0,0,0,2,39,1,5,5,NA
+67443,7,2,1,4,NA,1,1,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16169.123686,17324.04909,1,100,13,13,NA,4,4,1,1,0,1,28,2,1,1,1
+67444,7,2,1,3,NA,5,6,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9459.753034,9916.192713,2,91,10,10,3.04,4,4,1,1,0,1,37,2,5,1,5
+67445,7,2,2,38,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,2,1,2,1,2,2,1,2,2,2,2,2,2,32097.38481,33203.6204,2,100,4,4,0.81,4,4,0,2,0,1,56,1,4,1,2
+67446,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19835.707828,24564.335286,1,94,6,6,1.18,5,5,1,2,0,1,30,1,3,1,3
+67447,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,140431.173819,139253.659476,1,94,8,8,3.47,2,2,0,0,1,2,80,1,1,2,NA
+67448,7,2,1,4,NA,4,4,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7629.74403,7646.343205,1,99,4,4,0.53,7,7,3,1,0,2,26,1,1,5,NA
+67449,7,2,2,48,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,111065.717962,115110.76456,2,91,15,15,5,4,4,0,2,0,2,48,1,5,1,5
+67450,7,2,1,55,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,166693.693828,2,91,15,6,2.24,7,1,0,0,1,1,49,NA,NA,5,NA
+67451,7,2,2,18,NA,1,1,2,19,228,2,NA,2,2,3,12,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,16594.391299,17940.930507,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+67452,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,134694.414609,139412.076132,2,91,15,10,5,3,1,0,0,1,1,47,1,5,5,NA
+67453,7,2,1,0,1,3,3,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23912.171644,23498.78217,1,94,9,9,2.39,4,4,2,0,0,2,30,2,4,1,3
+67454,7,2,2,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53103.835249,56432.29308,2,92,6,6,2.02,2,2,0,0,2,2,76,1,3,1,NA
+67455,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,87101.243392,95057.097172,1,101,6,6,1.65,2,2,0,0,2,1,70,1,3,1,1
+67456,7,2,1,7,NA,4,4,2,7,93,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9699.683862,9681.447258,2,100,13,13,NA,4,4,0,2,0,2,28,1,2,6,NA
+67457,7,2,1,49,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,31010.243793,30930.584862,3,92,6,6,0.74,7,7,2,1,0,2,46,1,2,1,4
+67458,7,2,1,9,NA,5,6,1,10,121,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7921.96624,8581.996701,1,92,7,7,2.31,2,2,0,1,0,2,26,1,4,5,NA
+67459,7,2,2,24,NA,4,4,1,NA,NA,2,NA,2,2,4,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,17427.779559,17080.798702,2,93,6,6,1.13,4,4,0,0,2,1,60,2,3,1,3
+67460,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,44341.489734,49693.466543,2,103,5,5,1.79,1,1,0,0,1,2,80,1,4,2,NA
+67461,7,2,2,0,11,3,3,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,11331.751026,1,101,1,1,0.08,3,3,1,0,0,2,19,1,2,NA,NA
+67462,7,2,2,38,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,3,2,1,2,1,1,2,1,2,2,2,NA,35464.8385,46825.562506,2,96,3,3,0.95,2,2,0,1,0,2,38,2,3,3,NA
+67463,7,2,1,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,124091.929364,131797.624562,1,95,15,15,5,3,3,1,0,0,1,28,1,5,1,5
+67464,7,2,1,10,NA,3,3,1,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41165.805324,42583.611962,2,100,10,10,3.13,4,4,0,2,0,1,45,1,4,1,4
+67465,7,1,2,58,NA,4,4,NA,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15519.323571,0,2,99,10,10,3.67,3,3,0,0,1,1,64,2,4,1,5
+67466,7,2,1,55,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,23431.677775,24302.344664,2,93,5,5,1.26,3,3,0,1,0,1,55,2,2,1,2
+67467,7,1,1,80,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,39280.460347,0,2,91,4,4,1.39,1,1,0,0,1,1,80,1,3,2,NA
+67468,7,2,2,17,NA,5,6,1,17,212,2,NA,2,1,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9416.499309,9568.799193,1,100,15,15,5,4,4,0,1,0,1,44,2,5,1,5
+67469,7,2,1,23,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,14703.033095,3,91,7,7,2.58,2,2,0,0,0,2,55,2,5,3,NA
+67470,7,2,2,37,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,22326.231285,22166.696692,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+67471,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,155208.037885,157114.920604,2,102,8,8,4.32,1,1,0,0,0,2,45,1,3,4,NA
+67472,7,2,1,2,NA,3,3,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,40111.361732,45255.148434,1,98,15,15,3.7,5,5,2,1,0,1,34,1,5,1,5
+67473,7,2,2,79,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,29145.675285,31446.34406,3,92,7,7,2.31,2,2,0,0,2,1,77,1,4,1,4
+67474,7,2,1,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18423.855809,18161.238545,1,96,12,12,NA,5,5,1,2,0,2,35,1,5,1,4
+67475,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16793.056025,2,100,10,10,5,1,1,0,0,0,1,52,1,3,3,NA
+67476,7,2,1,6,NA,3,3,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57218.802967,58901.688698,1,95,9,9,2.13,6,6,0,4,0,2,44,1,1,1,1
+67477,7,2,2,57,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12449.932013,12109.206963,3,90,15,15,4.89,5,5,0,0,0,2,57,2,3,1,3
+67478,7,2,1,2,NA,3,3,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37959.146468,40828.920669,1,94,99,99,NA,3,3,1,0,0,1,31,1,4,6,NA
+67479,7,2,2,5,NA,4,4,2,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10771.85499,11250.180007,2,97,2,2,0.38,3,3,1,1,0,2,27,1,2,5,NA
+67480,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,6503.357568,6816.933235,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+67481,7,2,2,69,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10235.0654,10662.230581,2,93,3,3,0.66,2,2,0,0,2,2,69,2,4,3,NA
+67482,7,2,2,14,NA,2,2,1,14,173,NA,NA,1,1,NA,11,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18515.058419,19360.671834,2,96,2,2,0.27,6,6,1,3,0,1,34,NA,NA,1,NA
+67483,7,2,2,64,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,2,2,2,2,2,2,2,2,2,2,9570.416297,10355.226074,1,92,2,2,0.66,1,1,0,0,1,2,64,1,1,3,NA
+67484,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,NA,NA,NA,NA,14289.513581,13586.634632,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+67485,7,2,1,1,22,3,3,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22261.258803,25115.990281,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+67486,7,2,2,41,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,36924.381422,38054.088192,2,102,14,14,3.8,4,4,2,0,0,2,41,2,4,1,5
+67487,7,2,2,75,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,16494.288293,17728.004854,2,101,3,3,0.92,1,1,0,0,1,2,75,1,2,2,NA
+67488,7,2,1,29,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,3,12241.196357,12961.680012,2,96,15,7,3.21,3,1,0,0,0,1,31,2,5,1,NA
+67489,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,24158.28574,25084.308361,1,92,14,9,3.97,3,2,0,0,2,1,51,1,4,5,NA
+67490,7,2,1,34,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,1,2,2,2,2,2,2,41241.224595,41716.316195,2,102,4,4,0.61,5,5,0,3,0,1,34,2,3,1,3
+67491,7,2,1,2,NA,1,1,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13968.423539,13738.28453,2,102,3,3,0.54,4,4,1,1,0,1,28,2,2,6,NA
+67492,7,2,2,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85444.349063,87754.040572,1,94,14,14,5,2,2,0,0,0,1,50,1,3,1,5
+67493,7,2,1,76,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,8517.336599,8684.171961,2,100,9,9,4.35,2,2,0,0,2,2,79,1,5,1,3
+67494,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,21265.137164,1,94,4,4,0.56,5,5,1,2,0,1,34,1,2,3,NA
+67495,7,2,1,11,NA,2,2,1,11,142,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,13898.598114,14013.214919,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+67496,7,2,1,2,NA,1,1,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13305.770449,13463.073191,3,92,4,4,0.59,5,5,2,1,0,1,20,2,1,1,3
+67497,7,2,2,6,NA,5,6,1,6,74,NA,NA,2,2,2,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5064.232234,5712.276567,2,92,99,77,NA,7,4,3,3,1,1,61,2,1,1,3
+67498,7,2,2,16,NA,4,4,1,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14166.687432,14745.171396,1,100,1,1,0.04,4,4,1,1,0,2,51,1,3,3,NA
+67499,7,2,1,73,NA,2,2,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,13345.06742,13314.236759,1,99,15,15,5,2,2,0,0,2,1,73,1,4,1,5
+67500,7,2,2,9,NA,5,7,2,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,18348.624469,20109.577944,2,90,77,77,NA,6,6,0,4,0,2,41,NA,NA,4,NA
+67501,7,2,2,62,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,164066.603708,163511.57474,1,95,10,10,4.76,2,2,0,0,2,1,65,1,4,1,4
+67502,7,2,1,56,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10976.758665,11141.51106,2,103,15,15,5,3,3,0,0,0,2,52,2,4,1,5
+67503,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,154825.466557,154977.498295,1,94,12,7,3.67,2,1,0,0,0,2,45,1,5,5,NA
+67504,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,1,1,2,2,1,2,2,1,2,2,1,27532.825087,28471.490951,1,101,6,6,1.3,4,4,0,2,0,1,43,2,1,1,3
+67505,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,163604.158578,166277.181905,1,101,10,10,4.63,2,2,0,0,2,1,62,1,5,1,5
+67506,7,2,2,6,NA,3,3,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21658.223225,22668.915202,1,90,7,7,1.55,5,5,0,3,0,1,51,2,3,1,2
+67507,7,2,1,28,NA,4,4,1,NA,NA,2,NA,2,2,4,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17840.217575,17960.016751,2,93,6,6,1.13,4,4,0,0,2,1,60,2,3,1,3
+67508,7,2,1,18,NA,5,7,2,19,228,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13547.769361,13874.542578,1,91,9,9,4.35,2,2,0,0,0,2,40,1,4,3,NA
+67509,7,2,1,79,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,68266.732554,70856.701199,1,95,7,7,3.67,1,1,0,0,1,1,79,1,5,5,NA
+67510,7,2,1,4,NA,4,4,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7311.663111,7613.781996,2,93,6,6,1.15,5,5,3,1,0,1,29,1,3,5,NA
+67511,7,1,2,21,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,3,6,3,1,2,2,1,2,2,NA,NA,NA,NA,44119.608456,0,1,92,6,6,1.51,3,3,0,0,0,1,46,1,3,3,NA
+67512,7,2,1,3,NA,4,4,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,10259.663981,2,97,13,13,NA,4,4,1,0,1,2,45,1,2,5,NA
+67513,7,2,1,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,20286.317046,20915.633712,2,99,2,2,0.5,2,2,0,0,0,1,53,1,3,1,NA
+67514,7,2,1,6,NA,1,1,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11036.458246,10936.228943,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+67515,7,2,2,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,18174.62037,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+67516,7,2,2,47,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20773.573942,20827.234538,2,93,6,6,1.47,3,3,0,0,0,2,47,1,4,5,NA
+67517,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,23660.610934,25318.185975,3,91,5,5,1.07,4,4,0,2,0,2,36,1,5,1,4
+67518,7,2,2,5,NA,1,1,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11009.851855,11478.955249,1,93,13,3,0.54,6,3,2,1,0,1,23,NA,NA,5,NA
+67519,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,6612.194774,6921.511029,2,99,6,6,1.98,2,2,0,0,1,1,63,1,2,2,NA
+67520,7,2,1,79,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11245.778093,11855.469704,1,98,4,4,1.19,2,2,0,0,2,1,79,1,3,1,3
+67521,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,81062.798322,86779.337922,1,101,4,4,1.48,1,1,0,0,1,2,76,1,2,3,NA
+67522,7,2,2,26,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,63207.045171,63513.398112,2,102,15,15,5,2,2,0,0,0,1,41,2,4,6,NA
+67523,7,1,1,54,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,31699.998102,0,1,98,5,5,2.15,1,1,0,0,0,1,54,1,3,3,NA
+67524,7,2,1,6,NA,2,2,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9807.589376,10786.008845,2,90,5,5,0.76,5,5,0,4,0,2,32,1,2,3,NA
+67525,7,2,1,67,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,1,1,2,2,1,2,1,3,10497.261485,11239.197739,1,90,2,2,0.33,2,2,0,0,2,2,65,2,3,1,5
+67526,7,2,1,33,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,34887.439952,36924.956604,2,94,7,7,1.23,6,6,2,1,0,1,33,2,1,6,NA
+67527,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,17399.722435,17053.300185,2,99,6,3,1.27,3,1,0,0,0,1,41,1,3,6,NA
+67528,7,2,1,2,NA,3,3,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,38457.282261,42554.855377,1,100,6,6,1.18,5,5,2,2,0,2,40,1,5,3,NA
+67529,7,2,2,11,NA,3,3,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,80369.555824,80552.420473,1,97,15,15,4.07,5,5,0,3,0,1,36,1,5,1,5
+67530,7,2,1,52,NA,1,1,2,NA,NA,2,NA,2,2,7,NA,5,4,NA,2,2,2,2,2,2,1,2,2,2,22446.308035,22508.005013,2,94,4,4,1.38,1,1,0,0,0,1,52,2,5,4,NA
+67531,7,2,2,1,15,4,4,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6880.068407,7403.030352,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+67532,7,2,2,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,88916.414931,94489.544183,1,101,6,6,1.65,2,2,0,0,2,1,70,1,3,1,1
+67533,7,2,1,0,4,1,1,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6298.658963,6298.775059,2,94,7,7,1.57,4,4,2,0,0,2,27,1,4,1,2
+67534,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,31070.966983,31355.213599,1,102,2,2,0.66,1,1,0,0,0,2,56,1,4,2,NA
+67535,7,2,2,63,NA,4,4,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9113.905743,9695.883475,3,90,9,9,3.14,3,3,0,1,1,2,63,2,4,3,NA
+67536,7,2,2,14,NA,3,3,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,91797.787708,92992.465808,1,95,15,15,5,4,4,0,2,0,2,42,1,5,1,5
+67537,7,2,1,33,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19638.748956,20390.674215,2,99,15,15,5,2,1,0,0,0,2,31,1,5,1,NA
+67538,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,25645.251384,25368.119207,1,92,6,6,1.3,4,4,0,1,1,1,25,1,1,1,3
+67539,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,2,1,2,2,1,2,2,1,2,2,1,102855.726146,103775.32646,1,101,5,5,0.89,5,5,1,2,0,1,31,1,2,1,1
+67540,7,2,2,12,NA,1,1,2,12,155,NA,NA,2,2,4,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13824.001771,14945.739837,2,90,3,3,0.58,4,4,0,2,0,2,36,2,3,1,3
+67541,7,2,1,11,NA,5,6,2,11,143,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10810.913614,11522.32071,1,97,15,15,4.07,5,5,0,3,0,1,42,2,5,1,5
+67542,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,3,1,2,1,2,1,1,2,1,NA,NA,NA,NA,11032.714892,11055.242776,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+67543,7,2,2,16,NA,2,2,2,16,194,NA,NA,1,1,NA,9,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,12680.621719,13709.581084,2,90,7,7,1.34,5,5,0,1,2,1,61,2,1,4,NA
+67544,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,17521.481386,17363.561951,2,101,1,1,0.2,2,2,0,0,1,2,55,1,4,4,NA
+67545,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,124604.441853,125718.490512,1,92,9,4,1,7,3,2,1,0,1,45,1,4,2,NA
+67546,7,2,2,30,NA,2,2,1,NA,NA,2,NA,2,1,3,NA,4,1,2,2,2,2,1,2,2,NA,NA,NA,NA,40476.413979,40301.503085,2,93,7,7,1.83,3,3,1,0,0,1,40,2,5,1,4
+67547,7,2,1,8,NA,3,3,2,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,16549.21669,17579.802509,1,91,12,8,2.15,6,4,1,1,0,2,29,1,4,6,NA
+67548,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,12532.471503,12837.991566,2,97,4,4,0.84,3,3,1,0,1,2,68,1,4,2,NA
+67549,7,2,1,25,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,36391.614555,41134.650764,1,103,5,5,1.02,4,4,2,0,0,1,25,1,2,1,4
+67550,7,2,2,13,NA,4,4,1,13,159,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15483.914568,16116.18632,1,100,4,4,1.23,2,2,0,1,0,2,47,1,2,4,NA
+67551,7,2,2,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,17427.779559,17328.063231,2,93,12,12,NA,4,4,0,0,2,1,72,1,2,1,4
+67552,7,2,2,16,NA,3,3,1,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,76992.7514,81576.336255,1,100,15,15,4.07,5,5,0,2,0,2,41,1,5,1,4
+67553,7,2,2,9,NA,4,4,2,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7388.61397,8001.678046,3,90,2,2,0.39,2,2,0,1,0,2,32,1,3,5,NA
+67554,7,1,2,61,NA,4,4,NA,NA,NA,2,NA,2,1,99,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9024.164523,0,2,93,15,15,5,1,1,0,0,1,2,61,2,5,1,NA
+67555,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,31785.728924,32871.467684,1,95,4,4,1.22,2,2,0,0,1,1,79,1,1,1,3
+67556,7,2,1,14,NA,4,4,2,14,178,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13251.602554,13095.83949,1,96,9,9,3.14,3,3,0,2,0,2,39,NA,NA,3,NA
+67557,7,2,2,15,NA,4,4,2,15,187,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9728.028593,10125.263908,2,99,6,6,1.11,5,5,1,2,0,2,41,1,2,5,NA
+67558,7,2,2,0,7,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4476.23918,4525.114437,1,90,3,3,0.43,4,4,2,0,0,1,46,1,3,1,4
+67559,7,2,1,9,NA,3,3,2,9,118,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30554.050397,31755.009616,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+67560,7,2,1,68,NA,5,6,1,NA,NA,2,NA,2,2,8,NA,1,5,NA,1,2,2,2,2,2,1,2,2,1,13527.667075,14341.996991,2,98,6,6,0.59,7,7,2,2,1,2,52,2,1,1,1
+67561,7,2,2,12,NA,4,4,1,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14086.017075,14046.326674,2,96,5,5,1.07,4,4,0,3,0,2,46,1,4,3,NA
+67562,7,2,1,2,NA,4,4,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6347.215153,6609.482921,2,95,1,1,0.09,5,5,3,1,0,2,31,1,2,1,NA
+67563,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,39416.921733,1,92,2,2,0.64,1,1,0,0,1,2,80,1,3,2,NA
+67564,7,2,2,43,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,37512.060155,39042.669235,1,92,15,15,4.99,4,4,0,2,0,2,43,1,4,1,4
+67565,7,2,1,10,NA,3,3,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58429.74688,62860.392832,1,100,15,15,4.56,4,4,0,2,0,2,42,1,4,1,3
+67566,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,30796.941649,33729.162412,1,98,4,4,1.26,2,2,0,0,1,2,80,1,4,2,NA
+67567,7,2,1,69,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9618.837002,9727.194029,3,91,14,14,3.53,5,5,0,1,1,1,69,1,4,3,NA
+67568,7,2,2,7,NA,3,3,2,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22933.149195,22757.230737,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+67569,7,2,1,7,NA,3,3,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23188.935049,25672.571973,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+67570,7,2,2,6,NA,2,2,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13720.541363,13932.536034,2,100,14,14,3.36,4,4,1,1,0,1,45,2,5,1,2
+67571,7,2,2,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,28747.860416,30588.052168,1,99,6,6,0.96,5,5,1,2,0,2,35,1,4,1,2
+67572,7,2,2,1,23,3,3,1,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,59617.625426,65799.423913,3,92,15,15,5,4,4,2,0,0,1,38,1,5,1,5
+67573,7,2,2,80,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,2,2,2,2,2,2,2,2,2,NA,25315.905293,28273.512517,2,98,4,4,1.15,2,2,0,0,2,1,80,1,1,1,1
+67574,7,2,2,12,NA,4,4,2,12,147,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11608.900361,11982.991894,1,99,8,8,2.59,3,3,0,2,0,2,46,1,4,2,NA
+67575,7,2,2,6,NA,1,1,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17458.526997,17642.073401,1,94,6,6,1.11,5,5,1,2,0,2,41,2,1,1,1
+67576,7,2,2,8,NA,1,1,2,8,97,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,15225.935813,15621.122955,2,94,77,77,NA,6,6,0,3,0,2,58,1,3,1,9
+67577,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,118933.861968,118531.514768,1,91,8,8,4.82,1,1,0,0,1,2,60,1,4,3,NA
+67578,7,2,2,10,NA,3,3,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,73810.484644,75214.997077,1,94,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+67579,7,2,1,10,NA,1,1,1,10,121,NA,NA,1,1,NA,2,NA,NA,NA,2,1,1,2,2,1,1,2,2,1,14880.007592,14744.872501,3,92,4,4,1.12,2,2,0,1,0,2,38,2,1,5,NA
+67580,7,2,2,67,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,37934.469637,42910.885175,1,94,3,3,0.95,2,2,0,0,1,2,67,1,2,2,NA
+67581,7,2,2,19,NA,2,2,2,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,5,5,0.8,5,5,0,3,0,2,40,2,1,5,NA
+67582,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,NA,22306.465066,22494.734848,2,98,3,3,0.54,3,3,1,1,0,2,29,1,2,1,NA
+67583,7,1,1,6,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14456.751936,0,1,90,15,15,5,5,5,1,1,0,1,32,2,1,1,4
+67584,7,2,1,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19008.083201,19380.210545,2,97,7,7,3.67,1,1,0,0,0,1,38,1,3,5,NA
+67585,7,2,2,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16181.169973,16286.716901,2,100,10,10,4.42,2,2,0,0,0,2,55,1,2,1,4
+67586,7,2,2,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,17622.141982,17728.11858,2,95,2,2,0.75,1,1,0,0,0,2,54,1,2,5,NA
+67587,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,33147.414266,37148.276517,2,94,7,7,3.49,1,1,0,0,1,2,80,1,5,2,NA
+67588,7,2,1,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7814.415054,8296.803376,1,93,6,6,0.83,6,6,3,1,0,1,37,NA,NA,1,3
+67589,7,2,1,79,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,8517.336599,8684.171961,2,100,6,6,2.11,2,2,0,0,2,1,79,1,3,1,4
+67590,7,2,1,31,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,36150.793885,43021.850084,1,90,3,3,0.43,4,4,2,0,0,1,31,1,3,6,NA
+67591,7,2,1,40,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11113.602843,11678.105413,3,90,14,14,3.47,4,4,1,1,0,2,38,2,5,1,5
+67592,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17810.743272,17839.416552,1,99,15,15,5,2,2,0,0,0,2,46,1,5,1,5
+67593,7,2,1,4,NA,4,4,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9035.668246,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+67594,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,100867.796959,104079.392539,2,99,15,15,5,3,3,1,0,0,1,43,1,5,1,5
+67595,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25128.435397,25532.311821,2,94,6,5,2.11,3,1,0,0,0,1,26,1,4,5,NA
+67596,7,2,1,63,NA,5,7,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9314.746352,9205.449493,1,91,15,15,5,3,3,0,0,1,1,63,2,5,1,5
+67597,7,2,2,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,21077.085361,20500.255628,2,99,12,12,NA,1,1,0,0,0,2,41,1,5,5,NA
+67598,7,2,2,80,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,3,NA,2,2,2,2,2,2,2,2,2,NA,16490.79781,18057.97943,3,90,10,10,3.04,4,4,0,0,2,2,80,2,1,3,NA
+67599,7,2,2,4,NA,4,4,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13130.790087,13857.266639,2,102,5,5,0.76,5,5,1,3,0,2,30,1,4,4,NA
+67600,7,2,2,15,NA,3,3,2,15,181,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,34552.42011,37128.535405,1,90,7,7,1.55,5,5,0,3,0,1,51,2,3,1,2
+67601,7,1,2,8,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7842.234542,0,1,91,15,15,5,5,5,2,1,0,1,40,1,5,1,5
+67602,7,2,1,43,NA,1,1,1,NA,NA,1,2,2,2,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,31740.385214,32919.784432,2,96,7,7,1.79,4,4,0,2,0,1,43,2,3,1,2
+67603,7,2,2,62,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17243.546687,17909.086027,1,92,14,14,5,2,2,0,0,2,2,62,1,4,1,4
+67604,7,2,2,24,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,13820.210756,14410.358083,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+67605,7,2,2,19,NA,1,1,1,19,230,2,NA,2,7,77,10,NA,NA,NA,2,2,2,2,2,2,2,2,2,1,18581.167701,19429.800436,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+67606,7,2,1,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,28542.421068,29427.85637,2,101,2,2,0.74,1,1,0,0,0,1,48,1,2,5,NA
+67607,7,2,2,63,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,3,1,NA,1,2,1,1,2,1,1,2,1,1,17248.011865,19146.123826,1,97,14,14,4.96,2,2,0,0,1,1,50,1,3,1,3
+67608,7,2,1,4,NA,5,6,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10273.602479,11522.969217,1,92,15,15,5,3,3,1,0,0,1,50,1,4,1,4
+67609,7,2,2,6,NA,4,4,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9399.281543,9563.245188,2,96,6,6,1.35,3,3,1,1,0,2,23,1,2,5,NA
+67610,7,2,1,2,NA,4,4,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,5652.403121,6080.146007,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+67611,7,2,2,14,NA,3,3,2,14,175,NA,NA,2,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,76360.13568,77793.595093,1,99,15,15,5,3,3,0,1,0,1,50,1,4,1,4
+67612,7,2,1,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17882.621856,18280.794545,3,92,7,7,2.1,3,3,1,1,0,2,25,1,4,5,NA
+67613,7,2,2,21,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,11739.283384,12617.501315,1,96,7,7,1.83,3,3,0,0,1,1,66,2,5,1,3
+67614,7,2,1,53,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,25118.469449,26312.875816,2,98,5,5,1.07,4,4,0,1,0,1,53,2,1,1,1
+67615,7,2,2,17,NA,5,7,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12821.227649,2,100,9,9,2.46,4,4,0,2,0,2,36,2,4,1,3
+67616,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,167711.394252,175827.980982,1,101,4,2,0.81,2,1,0,0,1,1,63,1,2,6,NA
+67617,7,2,2,13,NA,2,2,1,13,162,NA,NA,2,2,4,6,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,15809.066118,17342.845429,2,93,7,7,1.52,4,4,1,1,0,1,44,2,4,1,NA
+67618,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,39565.288792,43332.356399,1,99,14,14,5,2,2,0,0,2,1,80,1,4,1,4
+67619,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,120041.937453,120355.294586,1,98,15,15,5,5,5,0,1,0,1,53,1,5,1,5
+67620,7,2,1,54,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,28017.200708,32299.614806,2,102,3,3,0.92,1,1,0,0,0,1,54,1,4,3,NA
+67621,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,43813.24867,44204.968982,1,98,4,4,0.67,5,5,1,2,0,1,29,1,4,1,3
+67622,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,18246.235208,18261.326779,2,99,1,1,0.07,4,4,1,1,0,2,24,1,2,5,NA
+67623,7,2,1,8,NA,3,3,2,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,74331.764009,78960.699825,2,91,14,14,4.19,3,3,0,1,0,2,31,1,4,1,3
+67624,7,2,1,30,NA,3,3,2,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,99283.360764,104566.027434,3,91,1,1,0.09,1,1,0,0,0,1,30,2,5,5,NA
+67625,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,25359.946386,29418.979823,2,95,2,2,0.88,1,1,0,0,1,2,80,1,1,2,NA
+67626,7,2,1,1,15,2,2,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13305.770449,13726.956753,3,92,9,9,2.22,5,5,1,0,2,1,66,2,1,1,1
+67627,7,2,1,25,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,23064.54207,22591.937434,2,91,15,15,5,4,4,0,0,1,1,61,NA,NA,1,2
+67628,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,53634.754806,54437.113731,1,98,99,99,NA,3,2,0,0,0,2,22,1,4,5,NA
+67629,7,2,2,70,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,18698.205673,19635.336647,1,97,12,12,NA,5,5,0,1,1,2,40,2,9,1,NA
+67630,7,2,1,12,NA,3,3,1,12,147,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,26749.020961,27839.957403,1,98,6,6,1.11,5,5,1,2,0,2,32,1,2,1,2
+67631,7,2,1,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,22007.065507,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+67632,7,2,2,51,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,24646.971819,25093.941214,1,102,7,7,1.89,3,3,0,0,0,1,53,2,1,1,1
+67633,7,2,2,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7944.571197,8483.470772,2,99,15,15,5,3,3,0,1,0,2,48,1,5,4,NA
+67634,7,2,2,8,NA,4,4,2,8,104,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8184.286585,8442.757341,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+67635,7,2,1,65,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,10298.484448,10551.517983,1,94,6,6,1.98,2,2,0,0,2,1,65,2,1,1,1
+67636,7,1,1,7,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17227.332856,0,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+67637,7,1,2,9,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7489.549692,0,2,100,10,10,2.33,6,6,0,2,2,2,35,1,2,5,NA
+67638,7,2,2,19,NA,4,4,1,19,232,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11437.338051,11296.529267,2,95,2,2,0.41,3,3,2,0,0,2,19,1,2,NA,NA
+67639,7,2,1,0,2,4,4,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6476.58432,6575.990772,2,97,3,3,0.75,2,2,1,0,0,2,22,1,4,4,NA
+67640,7,2,1,58,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27595.50738,27352.795749,1,98,6,6,1.57,3,3,0,0,0,1,58,1,3,1,3
+67641,7,2,1,5,NA,1,1,2,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16074.564008,16583.394837,1,97,4,4,0.72,5,5,2,1,0,2,33,2,1,6,NA
+67642,7,2,2,24,NA,1,1,2,NA,NA,2,NA,2,1,4,NA,2,6,2,1,2,2,1,2,2,NA,NA,NA,NA,39426.061521,43773.817687,2,94,8,8,2.33,4,4,2,0,0,1,24,1,2,6,NA
+67643,7,2,1,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,51472.063256,53424.859661,2,95,15,15,5,2,2,0,0,2,1,76,1,4,1,5
+67644,7,2,1,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,23005.210125,23019.76475,1,95,14,14,3.04,6,6,0,4,0,1,56,1,5,1,4
+67645,7,2,1,15,NA,4,4,2,15,187,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11386.695644,11908.648036,2,99,6,6,1.18,5,5,0,3,0,2,38,1,2,5,NA
+67646,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11113.602843,11678.105413,3,90,5,5,0.93,4,4,1,0,0,1,48,2,4,1,NA
+67647,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,16494.288293,17598.146607,2,95,3,3,1.1,1,1,0,0,1,2,70,1,3,4,NA
+67648,7,1,2,2,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6662.138091,0,2,95,14,14,3.58,4,4,1,0,0,1,39,1,3,1,3
+67649,7,2,2,41,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,2,1,2,1,2,1,1,2,2,1,2,1,3,16378.162652,17998.555773,2,96,4,4,0.92,3,3,0,1,1,2,41,2,2,1,2
+67650,7,2,2,11,NA,3,3,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16950.724686,16899.799988,2,98,2,2,0.34,2,2,0,1,0,2,30,1,4,4,NA
+67651,7,2,2,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20430.250572,20111.803159,2,97,5,5,0.76,5,5,0,0,0,2,50,1,4,5,NA
+67652,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,108408.375382,118202.807654,1,92,6,6,1.51,3,3,0,0,0,1,46,1,3,3,NA
+67653,7,2,2,39,NA,3,3,2,NA,NA,2,NA,2,1,7,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,74517.751389,77393.175383,2,94,10,10,3.51,3,3,0,2,0,2,39,2,4,3,NA
+67654,7,2,2,16,NA,1,1,1,16,193,NA,NA,1,1,NA,8,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,20460.442471,21235.303768,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+67655,7,2,2,8,NA,4,4,2,8,99,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9326.540969,10100.402919,1,91,8,8,1.76,5,5,0,3,0,2,42,1,3,6,NA
+67656,7,2,1,53,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15599.953109,16217.343339,1,99,14,1,0.23,2,1,0,0,0,2,51,1,5,1,NA
+67657,7,2,1,1,20,3,3,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,32956.344514,37182.588634,2,97,15,15,4.97,5,5,1,0,0,1,48,1,4,1,3
+67658,7,2,1,77,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,52501.717941,55571.910913,2,102,8,8,4.13,1,1,0,0,1,1,77,1,5,2,NA
+67659,7,2,2,15,NA,5,6,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8272.446168,8996.728356,1,102,15,15,3.82,5,5,1,1,0,1,29,1,4,1,4
+67660,7,2,1,16,NA,5,7,2,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9136.388281,9764.423514,3,91,15,15,4.47,4,4,0,1,0,2,45,2,5,1,5
+67661,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,13408.721263,15005.802731,2,98,4,4,1.22,2,2,0,0,2,1,80,1,1,1,1
+67662,7,2,2,11,NA,5,6,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,5064.232234,5431.552443,2,92,10,8,2.01,7,4,1,1,1,2,27,2,3,1,3
+67663,7,2,1,3,NA,4,4,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9319.143734,9894.41982,2,95,8,8,1.85,5,5,1,2,0,1,55,1,2,1,3
+67664,7,2,1,11,NA,1,1,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11399.23838,11468.323492,2,103,77,77,NA,5,5,1,2,0,2,30,1,2,1,2
+67665,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,19498.713386,19633.661812,2,97,5,5,0.84,5,5,0,2,0,2,33,1,4,1,3
+67666,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,10816.310126,11253.74147,1,99,14,14,5,2,2,0,0,1,2,64,1,3,3,NA
+67667,7,1,1,35,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,79582.971419,0,2,99,15,15,5,1,1,0,0,0,1,35,1,5,5,NA
+67668,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,26302.518039,27109.453365,1,101,5,3,0.98,2,1,0,0,2,2,70,1,4,2,NA
+67669,7,2,2,2,NA,5,6,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8818.200077,8968.551716,2,102,6,6,1.52,4,4,2,0,0,1,30,2,4,1,4
+67670,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15497.844354,15146.281728,1,99,7,7,3.21,1,1,0,0,0,2,52,1,4,1,NA
+67671,7,2,2,5,NA,3,3,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,59321.981286,63042.638464,1,94,8,8,1.67,5,5,1,2,0,1,52,1,4,1,4
+67672,7,2,1,12,NA,3,3,2,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,27702.556686,27838.63939,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+67673,7,2,2,13,NA,4,4,1,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14166.687432,14623.20249,1,100,15,15,3.87,6,6,1,3,0,2,39,1,4,1,4
+67674,7,2,2,0,4,4,4,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4964.196196,5466.923926,1,93,4,4,1.09,2,2,1,0,0,2,39,2,3,4,NA
+67675,7,2,1,66,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,11924.931067,12067.158788,1,98,8,8,1.95,4,4,1,1,1,2,59,1,3,1,1
+67676,7,1,1,6,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10651.061092,0,1,90,99,99,NA,2,2,0,1,0,2,26,1,4,5,NA
+67677,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17342.255821,20631.425049,1,93,3,3,1.29,1,1,0,0,0,1,36,1,4,5,NA
+67678,7,2,2,6,NA,3,3,1,7,85,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22308.590534,22137.463018,1,95,7,2,0.35,5,4,1,2,0,1,26,1,4,6,NA
+67679,7,2,1,9,NA,1,1,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14820.807433,14905.891142,2,102,7,7,1.53,5,5,0,3,0,1,43,2,2,1,4
+67680,7,2,2,10,NA,3,3,2,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,63555.637461,64464.244761,1,90,15,15,5,4,4,0,2,0,1,44,1,5,1,5
+67681,7,2,2,33,NA,4,4,2,NA,NA,2,NA,2,1,3,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,28747.860416,29850.909164,1,99,14,14,3.8,4,4,1,1,0,1,48,2,5,1,5
+67682,7,2,2,13,NA,1,1,2,13,157,NA,NA,2,2,3,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,19850.979841,20602.760063,1,97,4,4,0.72,5,5,2,1,0,2,33,2,1,6,NA
+67683,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,31785.728924,32633.084965,1,95,7,7,2.16,3,3,0,0,1,1,45,1,3,1,4
+67684,7,2,1,8,NA,2,2,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14820.807433,14943.02937,2,102,5,5,0.89,4,4,0,3,0,2,44,2,2,4,NA
+67685,7,2,1,9,NA,5,6,2,9,113,NA,NA,2,2,2,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9720.482616,10533.307174,2,91,14,14,3.47,4,4,1,1,0,2,36,2,3,1,5
+67686,7,2,1,12,NA,1,1,1,12,147,NA,NA,2,2,3,5,NA,NA,NA,2,1,1,1,2,2,1,2,2,1,32326.52031,34735.818434,3,92,7,7,1.41,5,5,1,2,0,1,20,2,1,1,1
+67687,7,2,2,8,NA,5,7,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,10422.423011,10751.577119,1,98,5,5,1.26,3,3,1,1,0,2,27,1,5,5,NA
+67688,7,2,2,0,10,5,7,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8303.503441,8530.807571,1,94,7,7,1.23,6,6,3,1,0,1,32,1,4,1,4
+67689,7,2,2,18,NA,1,1,2,18,219,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,19557.287652,21454.715555,2,94,9,9,2.37,5,5,0,1,0,1,48,2,4,1,2
+67690,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20480.890987,19920.377678,2,93,3,3,1.07,1,1,0,0,0,2,40,1,4,5,NA
+67691,7,2,1,18,NA,3,3,1,19,228,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,34443.919252,35848.685683,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+67692,7,2,2,12,NA,2,2,2,12,153,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23072.919059,23761.374659,1,97,15,15,4.52,6,6,0,4,0,2,41,1,5,1,5
+67693,7,1,2,2,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12871.484115,0,2,102,15,15,4.47,4,4,1,1,0,1,32,1,5,1,4
+67694,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,8890.956373,9238.69917,2,99,12,12,NA,2,1,0,0,1,2,46,1,3,2,NA
+67695,7,1,2,55,NA,2,2,NA,NA,NA,2,NA,2,1,5,NA,4,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,23200.373382,0,2,93,3,3,0.63,3,3,0,1,0,1,53,2,2,1,4
+67696,7,2,2,67,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18695.172864,19350.637044,2,102,7,7,1.68,5,5,0,0,3,1,70,2,4,1,4
+67697,7,2,2,3,NA,5,7,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8818.200077,9596.189182,2,102,15,15,5,3,3,1,0,0,2,34,1,5,1,5
+67698,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,129559.2554,1,98,9,9,3.97,2,2,0,0,2,1,63,1,5,1,4
+67699,7,2,1,32,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,20254.971305,2,100,9,9,3.24,3,3,1,0,0,1,32,1,3,1,4
+67700,7,2,1,69,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9883.238978,10581.776717,1,90,4,4,0.78,4,4,0,0,1,1,69,2,4,1,3
+67701,7,2,1,54,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17869.988703,18923.921245,3,90,14,14,3.69,4,4,0,2,0,2,49,1,4,1,4
+67702,7,1,1,8,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8613.834494,0,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+67703,7,2,1,72,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,21815.897449,22272.021332,3,92,6,6,1.98,2,2,0,0,2,1,72,1,4,1,1
+67704,7,2,2,10,NA,4,4,2,10,128,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10221.991799,10790.677347,1,98,15,15,3.7,5,5,0,3,0,1,37,1,2,1,2
+67705,7,1,1,78,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,12173.814732,0,2,92,3,3,0.98,1,1,0,0,1,1,78,1,1,2,NA
+67706,7,2,1,7,NA,5,6,1,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9307.245755,9856.522898,1,92,14,14,3.47,4,4,0,2,0,1,37,1,5,1,5
+67707,7,2,1,46,NA,3,3,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27340.471576,32700.198254,2,91,2,2,0.44,3,3,0,1,0,1,46,2,3,1,4
+67708,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,22685.373982,23047.569967,1,98,6,6,0.97,7,7,1,2,0,1,49,1,2,1,2
+67709,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,9235.951997,10444.280286,2,103,12,12,NA,3,3,0,0,1,1,60,2,2,1,2
+67710,7,2,1,50,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,22238.49412,21912.1785,2,90,7,7,1.34,5,5,0,1,2,1,61,2,1,4,NA
+67711,7,2,1,17,NA,4,4,2,17,214,2,NA,2,1,5,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,77,77,NA,7,7,0,3,1,2,43,77,5,5,NA
+67712,7,2,1,15,NA,4,4,1,15,181,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13653.432599,13680.426553,2,96,5,5,1.07,4,4,0,3,0,2,46,1,4,3,NA
+67713,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,167711.394252,169284.299948,1,101,10,10,4.63,2,2,0,0,2,1,62,1,5,1,5
+67714,7,2,1,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,43188.300631,46554.36669,2,91,15,2,0.86,7,1,0,0,1,1,49,NA,NA,5,NA
+67715,7,2,2,3,NA,5,7,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12382.393854,12702.004448,1,97,10,10,2.82,4,4,1,1,0,2,24,1,3,5,NA
+67716,7,2,1,60,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12585.09113,12786.688219,1,92,5,5,1.41,2,2,0,0,2,1,60,1,2,1,4
+67717,7,2,1,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,10717.375231,11218.730451,2,101,3,3,1.1,1,1,0,0,1,1,66,1,1,2,NA
+67718,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,138322.767578,145546.941919,1,101,5,5,1.45,2,2,0,0,0,2,49,1,4,3,NA
+67719,7,2,1,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16214.132654,20271.082789,3,90,14,14,5,2,2,0,0,0,1,47,1,2,5,NA
+67720,7,2,1,64,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12279.19827,12375.250021,2,102,14,14,5,2,2,0,0,2,1,64,1,4,1,3
+67721,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,49191.372812,55403.452354,3,91,3,3,1.1,1,1,0,0,1,2,80,1,1,2,NA
+67722,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,4,1,0.4,4,1,0,0,0,1,22,1,5,5,NA
+67723,7,1,2,19,NA,2,2,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,21850.358234,0,2,93,99,99,NA,3,3,0,0,0,2,54,2,3,4,NA
+67724,7,2,1,1,14,4,4,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9438.902193,9727.203655,2,102,4,4,0.81,3,3,2,0,0,2,23,1,4,5,NA
+67725,7,1,2,68,NA,2,2,NA,NA,NA,2,NA,2,1,5,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,10561.745159,0,2,90,77,77,NA,2,2,0,0,2,2,68,2,1,1,NA
+67726,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,2,2,2,2,22466.936477,23282.464675,1,90,6,6,1.57,3,3,0,0,0,1,50,2,2,1,2
+67727,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,29733.812317,31414.972893,1,101,5,5,1.45,2,2,0,0,0,1,41,NA,NA,1,4
+67728,7,2,1,66,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,74792.980299,75494.43714,2,101,99,2,0.67,7,1,0,0,1,1,55,NA,NA,77,NA
+67729,7,2,2,2,NA,2,2,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11412.410776,11898.666241,1,97,3,3,0.44,5,5,2,2,0,2,26,1,4,4,NA
+67730,7,2,1,14,NA,3,3,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,102027.743122,101259.182802,1,101,8,8,1.85,5,5,0,3,0,1,41,1,3,1,4
+67731,7,2,1,0,11,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13392.309303,13030.480476,1,99,15,15,3.45,7,7,1,4,0,1,42,1,5,1,5
+67732,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,7233.371983,2,95,1,1,0,1,1,0,0,1,1,62,1,2,2,NA
+67733,7,2,2,18,NA,2,2,1,18,217,2,NA,1,1,NA,11,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,15809.066118,16348.490248,2,93,6,6,1.39,4,4,0,0,0,1,53,2,3,1,3
+67734,7,2,2,19,NA,4,4,2,19,237,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18905.695776,2,101,2,2,0.48,2,2,1,0,0,2,19,1,3,NA,NA
+67735,7,2,1,16,NA,5,6,2,16,195,NA,NA,2,2,3,9,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,7350.524832,8448.96413,2,90,3,3,0.54,4,4,0,1,0,2,52,NA,1,1,2
+67736,7,2,2,5,NA,1,1,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13366.393396,13935.903375,2,94,5,5,0.67,6,6,1,3,0,1,37,2,3,1,4
+67737,7,2,2,25,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,10800.351372,11406.289115,2,92,15,15,5,3,1,0,0,0,2,25,1,5,5,NA
+67738,7,2,1,6,NA,2,2,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10738.959181,10632.988891,2,93,9,9,3.14,3,3,0,2,0,2,34,2,5,3,NA
+67739,7,2,2,17,NA,5,6,2,17,214,2,NA,2,1,5,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11975.458482,12226.374363,1,97,15,15,5,3,3,0,1,2,2,63,1,5,1,NA
+67740,7,2,1,52,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,4,NA,2,2,2,1,2,2,NA,NA,NA,NA,21084.369131,21009.599098,2,95,4,4,1.12,2,2,0,1,0,1,52,2,3,4,NA
+67741,7,2,2,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,NA,NA,NA,NA,105039.256649,105978.379214,1,90,8,8,1.67,5,5,2,1,0,2,28,1,4,1,5
+67742,7,2,1,5,NA,1,1,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17865.135763,18430.646082,3,92,3,3,0.52,5,5,2,1,0,2,29,2,1,1,3
+67743,7,2,1,51,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16117.991297,16911.733026,1,96,3,3,0.43,4,4,1,1,0,2,39,2,4,1,3
+67744,7,2,1,11,NA,1,1,1,11,143,NA,NA,2,2,3,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,13822.148996,14860.201344,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+67745,7,2,2,72,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,29534.722322,30544.805853,3,91,2,2,0.83,1,1,0,0,1,2,72,1,4,3,NA
+67746,7,2,1,10,NA,3,3,2,10,128,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39772.343364,41554.529575,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+67747,7,2,2,2,NA,1,1,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8141.317022,8654.572344,1,103,14,14,2.96,5,5,1,2,0,1,34,1,4,1,5
+67748,7,2,2,4,NA,3,3,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26966.264969,27815.81123,1,98,6,6,1.31,3,3,2,0,0,2,22,1,3,5,NA
+67749,7,2,1,1,19,5,6,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6389.003009,6767.308805,3,91,14,14,3.06,5,5,3,0,0,1,34,2,5,1,5
+67750,7,2,2,2,NA,4,4,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7016.981922,7550.350821,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+67751,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,43813.24867,47373.102231,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+67752,7,2,1,14,NA,5,7,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11506.937395,12095.740214,1,97,15,15,5,6,6,0,3,0,1,47,1,5,1,5
+67753,7,1,2,6,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,12789.411811,0,1,102,99,99,NA,5,5,0,2,1,1,52,2,1,1,1
+67754,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,14010.52688,14536.170828,2,103,15,15,5,3,3,1,0,0,1,35,1,9,1,5
+67755,7,2,1,11,NA,3,3,1,11,138,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74331.764009,78594.005469,2,91,9,9,2.6,4,4,0,2,0,1,53,1,2,1,5
+67756,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,17583.693727,17958.024135,2,95,15,7,3.13,5,1,0,0,0,1,47,1,5,1,3
+67757,7,2,1,0,4,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6910.953528,6911.08091,2,96,4,4,0.97,3,3,1,0,0,2,37,2,2,1,1
+67758,7,2,2,7,NA,4,4,2,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5758.0032,6529.524105,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+67759,7,2,1,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17341.8035,17546.820808,2,102,7,7,1.53,5,5,1,2,0,1,36,1,2,1,3
+67760,7,2,1,17,NA,5,6,2,18,216,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7057.977053,7861.480262,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+67761,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,27271.751091,29115.882263,2,101,6,6,1.67,3,3,0,0,0,2,22,1,4,5,NA
+67762,7,2,1,53,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,21016.438271,22051.40744,1,93,12,12,NA,3,3,0,1,0,2,49,2,3,1,3
+67763,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,97803.500399,99700.910854,1,95,14,14,3.8,4,4,0,2,0,2,37,1,5,1,5
+67764,7,2,1,76,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,2,2,2,1,2,2,NA,12962.876803,16930.790311,2,90,6,5,1.84,2,1,0,0,2,1,76,2,1,2,NA
+67765,7,2,1,2,NA,2,2,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10803.555682,10931.27688,2,94,6,6,1.43,5,4,2,1,0,2,23,2,3,6,NA
+67766,7,2,2,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,37934.469637,38685.959862,1,94,4,4,1.26,2,2,0,0,1,2,41,1,4,5,NA
+67767,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,29167.119125,34820.124106,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+67768,7,2,2,14,NA,3,3,2,14,174,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,103298.858809,112630.333619,1,101,8,8,2.24,4,4,0,1,0,1,45,1,4,1,NA
+67769,7,2,2,54,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,1,2,1,1,2,2,2,2,2,2,18295.488967,18390.898385,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+67770,7,2,2,0,10,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4418.651245,4685.491283,2,95,6,6,0.97,6,6,2,1,0,1,54,1,3,6,NA
+67771,7,2,1,13,NA,4,4,2,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,6,6,1.51,3,3,0,1,1,1,65,1,2,1,4
+67772,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53541.401974,54113.194007,1,99,9,9,3.74,2,2,0,0,2,2,73,1,3,1,4
+67773,7,2,2,60,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17243.546687,17909.086027,1,92,8,8,3.44,2,2,0,0,2,1,71,1,3,1,5
+67774,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16905.961576,17390.157008,2,94,8,8,1.6,6,6,3,1,0,2,32,1,4,1,4
+67775,7,2,2,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,93589.96496,99393.853006,1,93,15,15,5,3,3,1,0,0,1,33,1,5,1,3
+67776,7,2,1,54,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,1,2,1,1,2,2,2,2,2,2,21139.303536,20829.116843,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+67777,7,1,1,2,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5547.430651,0,1,96,9,9,2.18,5,5,1,1,0,1,26,1,4,1,4
+67778,7,2,1,31,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17802.31438,18749.539496,1,102,3,3,0.82,2,2,0,0,0,1,31,1,4,1,4
+67779,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,207644.101324,209788.639452,1,97,10,10,4.3,2,2,0,0,1,2,56,1,5,1,5
+67780,7,2,2,7,NA,1,1,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16061.826248,2,98,4,4,0.75,4,4,0,2,0,2,33,1,2,5,NA
+67781,7,2,2,41,NA,2,2,1,NA,NA,2,NA,2,2,1,NA,5,1,2,2,2,2,2,2,2,2,2,1,2,32606.880052,32776.922157,2,93,9,9,1.49,7,7,0,3,0,2,41,2,5,1,5
+67782,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19440.793325,19514.660132,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+67783,7,2,2,13,NA,5,6,2,13,162,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11975.458482,12226.374363,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+67784,7,2,2,18,NA,1,1,1,18,216,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15690.47168,16249.379042,1,102,5,5,0.86,5,5,2,0,0,2,21,2,2,5,NA
+67785,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,27156.758177,27240.256208,1,95,4,4,1.21,2,2,0,0,0,1,46,1,2,1,2
+67786,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,153199.936938,1,95,3,3,1.33,1,1,0,0,0,1,45,1,2,5,NA
+67787,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,24881.621161,24802.260383,2,94,15,15,5,1,1,0,0,0,1,44,1,5,3,NA
+67788,7,2,1,62,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,2,2,2,2,2,2,1,2,2,2,9871.376767,9834.888389,2,97,4,4,1.29,2,2,0,0,2,2,70,1,3,1,4
+67789,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,1,1,2,2,1,2,2,1,2,2,1,18723.98095,18882.01405,2,101,2,2,0.22,4,4,1,1,0,2,41,1,2,4,NA
+67790,7,2,1,14,NA,4,4,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10065.83895,10440.165744,2,99,14,14,4.09,3,3,0,2,0,2,37,1,5,5,NA
+67791,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,32489.451884,36954.118364,2,91,3,3,1.16,1,1,0,0,1,2,80,1,4,2,NA
+67792,7,1,1,10,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8943.919305,0,1,92,5,5,0.63,7,7,0,4,1,1,60,NA,NA,1,NA
+67793,7,2,1,39,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,37658.482129,41412.915504,1,100,7,7,1.74,4,4,0,2,0,2,39,2,1,1,3
+67794,7,2,1,51,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,25139.75955,26335.178282,2,99,77,77,NA,3,3,0,0,0,2,57,2,2,1,1
+67795,7,2,1,41,NA,5,6,2,NA,NA,2,NA,2,2,99,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,11113.602843,11073.454175,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+67796,7,2,2,1,20,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7108.64371,7648.979929,1,90,14,14,2.96,5,5,1,2,0,1,31,1,5,1,4
+67797,7,2,1,45,NA,2,2,2,NA,NA,2,NA,2,1,77,NA,3,3,NA,2,2,2,2,2,2,2,2,2,2,28428.303115,29087.004066,3,90,77,77,NA,4,3,0,0,0,1,45,2,3,3,NA
+67798,7,2,1,51,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,24923.937021,24558.216666,2,93,7,7,2.58,2,2,0,0,0,1,51,1,4,1,4
+67799,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,27945.726298,28031.650144,1,94,6,6,1.26,5,5,0,2,0,2,38,1,4,1,NA
+67800,7,2,1,0,8,2,2,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5863.789973,5970.184293,2,93,8,8,2.17,4,4,2,0,0,2,30,1,4,1,4
+67801,7,2,2,11,NA,3,3,1,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17482.620882,17502.587262,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+67802,7,2,1,4,NA,5,7,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8890.779467,9258.147648,3,91,10,10,2.48,5,5,2,1,0,2,27,1,2,1,4
+67803,7,2,1,6,NA,5,7,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18834.739821,19876.018883,1,94,2,2,0.3,5,5,1,2,0,1,23,1,1,6,NA
+67804,7,2,2,4,NA,5,6,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4437.828549,4650.082829,1,102,5,5,0.92,5,5,1,2,0,2,44,2,1,1,2
+67805,7,2,1,6,NA,4,4,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7687.032802,8119.723568,2,99,4,4,0.78,4,4,0,2,0,2,45,1,3,5,NA
+67806,7,2,2,16,NA,2,2,2,16,198,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,20937.26435,23702.551829,1,90,5,5,1,4,4,0,2,0,1,40,2,2,1,1
+67807,7,2,2,3,NA,3,3,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,76623.952798,84569.150751,2,91,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+67808,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,1,2,NA,40827.983435,45600.111014,1,98,8,8,2.62,3,3,0,0,3,1,68,1,4,1,4
+67809,7,2,1,13,NA,1,1,1,13,164,NA,NA,2,2,3,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,32326.52031,34735.818434,2,94,5,5,0.65,6,6,0,3,0,1,44,2,1,1,1
+67810,7,2,1,0,6,2,2,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5420.429995,5420.529903,3,90,6,6,1.3,4,4,2,0,0,1,20,2,5,6,NA
+67811,7,2,1,2,NA,4,4,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6454.74783,2,100,15,15,5,3,3,1,0,0,2,35,1,5,1,2
+67812,7,2,1,66,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,1,5,NA,2,2,2,1,2,2,2,2,2,2,12118.033999,12538.522644,2,91,99,99,NA,3,3,0,0,1,2,31,2,5,5,NA
+67813,7,2,1,67,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,8561.661692,8628.633702,2,99,15,14,5,2,1,0,0,2,2,65,NA,NA,6,NA
+67814,7,2,2,19,NA,4,4,1,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,1,1,0.06,1,1,0,0,0,2,19,1,4,NA,NA
+67815,7,2,2,0,4,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20786.668002,20803.881706,1,101,8,8,1.85,5,5,3,0,0,2,31,1,2,1,2
+67816,7,2,1,14,NA,4,4,2,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13969.457688,14306.402061,1,90,3,3,0.63,3,3,1,1,0,2,32,1,4,5,NA
+67817,7,2,1,55,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,35958.875702,36129.285962,1,98,8,1,0,4,1,0,0,0,1,53,1,2,1,3
+67818,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16007.367503,15264.693262,2,90,15,15,5,4,4,0,0,0,1,57,2,5,1,5
+67819,7,2,2,79,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,44856.466004,46390.551109,2,94,3,3,1.16,1,1,0,0,1,2,79,1,4,2,NA
+67820,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,19130.246369,19040.031855,2,95,7,7,1.74,4,4,0,2,0,2,47,1,5,4,NA
+67821,7,2,1,7,NA,4,4,1,8,96,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12399.014378,12375.702699,2,96,2,2,0.31,4,4,0,2,0,2,30,NA,NA,6,NA
+67822,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,38369.898602,44511.264162,2,95,3,3,0.95,2,2,0,0,2,2,80,1,1,1,2
+67823,7,2,2,15,NA,3,3,1,15,189,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,92934.523767,97199.343719,1,92,8,8,2.17,4,4,0,1,2,2,80,1,3,2,NA
+67824,7,2,1,40,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15851.563418,15794.298537,2,95,15,15,5,1,1,0,0,0,1,40,2,5,5,NA
+67825,7,2,2,11,NA,4,4,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11116.391625,11531.852198,1,92,77,77,NA,5,5,1,2,0,2,41,1,3,5,NA
+67826,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,122833.610417,128308.021903,1,98,15,15,4.34,4,4,0,2,0,1,51,1,5,1,5
+67827,7,2,1,53,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27609.8026,27737.039026,2,101,8,8,2.7,3,3,0,1,0,1,53,1,4,1,2
+67828,7,2,1,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11136.552864,2,101,4,4,1.22,2,2,0,0,2,1,67,1,1,1,2
+67829,7,2,1,19,NA,3,3,1,19,229,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,73465.215193,72911.812289,1,100,15,15,4.63,5,5,0,0,0,1,51,1,5,1,3
+67830,7,2,1,0,5,2,2,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,5733.655482,6014.350383,1,90,10,10,2.44,5,5,1,0,0,2,56,2,1,1,1
+67831,7,2,2,0,1,3,3,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16608.743852,16157.613485,1,94,15,15,5,4,4,2,0,0,1,37,1,5,1,4
+67832,7,2,2,71,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,17836.372654,18457.016952,2,96,5,5,0.67,6,6,1,2,1,1,34,1,4,1,4
+67833,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,12526.25131,13442.161918,2,95,5,5,1.32,2,2,0,0,2,2,78,1,2,1,2
+67834,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,92613.517882,94092.287712,3,91,8,6,2.75,3,1,0,0,0,1,23,1,4,1,4
+67835,7,2,2,9,NA,4,4,2,9,114,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7405.320425,7603.433563,2,99,6,6,1.13,4,4,1,1,0,1,33,1,3,6,NA
+67836,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,112153.23206,1,91,8,8,2.17,4,4,0,0,0,1,59,1,4,1,5
+67837,7,2,1,8,NA,3,3,2,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49922.147265,52289.098209,1,98,9,9,2.6,4,4,1,1,0,2,35,1,2,1,NA
+67838,7,2,1,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,11650.457723,11613.298191,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+67839,7,2,1,61,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9313.036488,9824.696739,1,103,15,15,5,2,2,0,0,1,2,56,2,5,1,5
+67840,7,2,1,28,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,115954.75973,120569.27907,1,92,5,5,1.97,1,1,0,0,0,1,28,1,4,3,NA
+67841,7,2,1,58,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11498.794569,11546.882913,1,96,15,15,5,5,5,0,0,0,1,58,2,5,1,5
+67842,7,2,2,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,16420.864787,17686.616768,2,102,1,1,0.16,3,3,1,0,1,2,63,1,2,4,NA
+67843,7,2,1,3,NA,3,3,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24655.003412,27816.70307,1,98,2,2,0.36,5,5,3,0,0,1,25,1,3,1,3
+67844,7,2,1,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,111284.977736,114467.051739,3,91,8,5,1.5,3,2,0,0,0,1,23,1,4,1,4
+67845,7,2,1,49,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18116.816149,22904.074887,2,90,8,7,3.21,2,1,0,0,0,1,51,1,2,5,NA
+67846,7,2,1,15,NA,5,7,2,15,183,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60818.973858,62775.825513,1,99,15,15,5,4,4,0,2,0,2,44,1,5,1,5
+67847,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,16186.470589,15743.485359,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+67848,7,2,1,66,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,12585.09113,13280.957381,2,98,4,4,1.34,2,2,0,0,2,1,66,1,1,1,1
+67849,7,2,2,0,11,5,7,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5882.73073,5843.204067,3,90,15,15,5,3,3,1,0,0,2,32,2,5,1,5
+67850,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10440.656902,1,99,10,10,5,1,1,0,0,1,2,64,1,5,3,NA
+67851,7,2,2,5,NA,3,3,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51483.624552,53105.566664,1,94,15,15,5,4,4,2,0,0,1,51,2,5,1,5
+67852,7,2,1,15,NA,5,6,1,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12412.721558,13262.259083,1,92,10,10,2.82,4,4,0,2,0,2,48,2,5,1,5
+67853,7,2,1,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19923.530941,23644.216851,2,95,6,6,0.97,6,6,2,2,0,1,37,1,3,1,4
+67854,7,2,1,80,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12101.489198,13370.144062,1,93,3,3,0.65,3,3,0,0,3,2,74,2,1,2,NA
+67855,7,2,1,58,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11498.794569,11546.882913,2,100,15,15,5,3,3,0,1,0,1,58,2,5,1,5
+67856,7,2,2,0,8,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4523.857227,4573.252417,2,95,6,6,1.57,3,3,1,0,0,1,29,1,3,1,4
+67857,7,2,2,25,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16021.911789,17471.931528,1,94,7,7,1.79,4,4,0,1,0,1,59,2,4,1,4
+67858,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,NA,10049.7347,10691.470743,2,90,5,5,1.98,1,1,0,0,1,2,68,1,4,5,NA
+67859,7,2,1,18,NA,4,4,1,18,219,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12792.875152,12871.476461,2,93,99,99,NA,7,6,1,0,0,1,19,1,3,NA,NA
+67860,7,2,2,14,NA,3,3,2,14,174,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23708.623398,24267.27809,2,97,5,5,1.08,3,3,0,1,0,2,45,1,4,6,NA
+67861,7,2,1,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,85786.890667,86591.455494,2,96,9,9,5,1,1,0,0,1,1,60,1,5,5,NA
+67862,7,2,1,1,20,3,3,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23524.870375,27564.905667,1,90,7,7,2.1,3,3,1,0,0,2,40,2,5,1,4
+67863,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,202246.928022,1,91,15,15,5,3,3,0,0,0,2,50,1,4,1,4
+67864,7,2,2,22,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,15735.580263,16455.708445,1,97,15,15,5,4,4,0,0,0,1,51,2,5,1,5
+67865,7,2,1,59,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17121.370948,17378.349074,1,92,14,14,3.9,4,4,0,0,0,2,55,2,5,1,5
+67866,7,2,2,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,13101.965339,13686.954077,1,100,6,6,2.24,1,1,0,0,1,2,60,1,5,3,NA
+67867,7,2,2,11,NA,2,2,1,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,13720.541363,14610.824098,2,100,4,4,0.81,4,4,0,2,0,2,37,1,2,1,2
+67868,7,2,2,13,NA,1,1,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23979.296993,25562.604106,1,101,3,3,0.41,5,5,0,2,1,2,36,2,4,4,NA
+67869,7,2,2,4,NA,2,2,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,2,2,1,NA,NA,NA,NA,12675.140707,13078.439708,1,96,5,5,0.94,4,4,2,0,0,1,32,2,3,1,4
+67870,7,2,2,20,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,1,1,2,2,1,10345.80475,11593.940235,3,90,15,15,3.23,6,6,0,2,0,1,50,2,2,1,2
+67871,7,2,2,6,NA,2,2,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15148.721588,15796.67129,2,91,10,10,3.04,4,4,1,1,0,2,31,2,5,1,5
+67872,7,2,2,78,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,1,1,NA,2,2,2,1,2,2,2,2,2,NA,20441.922878,23572.887355,1,90,4,4,0.94,3,3,0,0,2,2,78,2,1,1,2
+67873,7,2,2,50,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,20388.95294,20876.134898,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+67874,7,2,1,69,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,138075.879417,141933.339512,2,94,10,10,4.3,2,2,0,0,2,1,69,1,4,1,5
+67875,7,2,2,9,NA,4,4,1,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11195.065587,11548.620784,2,96,15,9,3.74,3,2,0,1,0,2,38,1,5,6,NA
+67876,7,2,2,5,NA,4,4,2,5,63,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11353.600279,12216.600549,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+67877,7,2,2,79,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,33602.625,34751.830256,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+67878,7,2,1,9,NA,4,4,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8714.559478,8887.823591,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+67879,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,139800.409559,142430.529735,1,100,15,15,5,4,4,0,1,0,1,50,1,4,1,4
+67880,7,2,1,13,NA,5,6,1,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9737.713978,10134.40846,2,100,14,14,4.03,4,4,0,2,0,1,48,2,5,1,5
+67881,7,2,2,17,NA,3,3,1,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,31282.308091,32019.424199,2,98,3,3,0.38,5,5,0,4,0,2,39,1,4,5,NA
+67882,7,2,1,4,NA,3,3,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,86817.367332,101726.917198,2,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+67883,7,2,1,5,NA,5,7,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12996.965152,13393.943951,1,101,2,2,0.33,4,4,2,1,0,2,26,1,4,5,NA
+67884,7,2,2,10,NA,2,2,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16986.005478,17464.743129,2,102,5,5,0.89,4,4,0,3,0,2,44,2,2,4,NA
+67885,7,2,2,80,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,3,2,NA,1,2,1,1,2,1,1,2,1,NA,12344.929687,13171.097693,1,93,1,1,0.28,1,1,0,0,1,2,80,2,3,2,NA
+67886,7,2,2,3,NA,1,1,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12714.663639,13516.237728,1,102,5,5,0.86,5,5,2,0,0,2,21,2,2,5,NA
+67887,7,2,2,67,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17243.546687,17909.086027,1,92,14,14,5,2,2,0,0,2,1,62,1,4,1,4
+67888,7,2,1,8,NA,3,3,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,27154.222487,28089.451048,2,101,4,4,0.73,5,5,1,2,0,1,40,1,5,1,5
+67889,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,30863.871606,31440.435679,1,101,4,4,1.43,1,1,0,0,0,2,42,1,5,5,NA
+67890,7,2,2,7,NA,1,1,2,7,88,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14300.71869,14814.284705,2,94,13,13,NA,5,5,0,3,0,1,32,2,2,1,1
+67891,7,2,2,29,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,17204.739153,20747.6165,1,93,9,9,5,1,1,0,0,0,2,29,1,5,5,NA
+67892,7,2,2,10,NA,3,3,1,10,123,NA,NA,2,2,3,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,13450.606713,14078.286101,3,91,4,4,0.65,5,5,2,2,0,2,27,2,2,3,NA
+67893,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,83819.702285,99717.125794,1,99,14,14,5,1,1,0,0,0,1,34,1,5,5,NA
+67894,7,2,1,80,NA,5,7,1,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,38080.409655,40864.822189,2,95,6,6,1.95,2,2,0,0,2,1,80,1,1,1,3
+67895,7,2,1,8,NA,4,4,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9261.557132,9782.87535,2,100,7,7,1.79,4,4,0,1,0,2,51,1,3,3,NA
+67896,7,2,1,17,NA,4,4,2,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15381.581315,15476.088016,1,93,12,12,NA,5,4,0,2,0,1,32,1,2,5,NA
+67897,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,60257.408263,62318.203523,1,103,15,15,5,2,2,0,0,2,1,74,1,5,1,5
+67898,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,12170.020379,2,97,6,6,1,6,6,1,2,2,2,60,1,2,2,NA
+67899,7,2,1,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,NA,7663.797586,8115.960731,1,96,3,3,1.25,1,1,0,0,1,1,80,1,2,4,NA
+67900,7,1,1,24,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,13859.220514,0,1,98,12,2,0.54,3,1,0,0,0,1,24,1,4,5,NA
+67901,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12222.312509,13136.499838,2,95,12,12,NA,3,3,0,0,3,2,73,1,2,1,1
+67902,7,2,1,21,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,4,NA,2,2,2,2,2,2,2,2,2,2,39915.513053,44831.646235,2,98,5,5,1.07,4,4,0,1,0,1,53,2,1,1,1
+67903,7,2,2,3,NA,4,4,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10771.85499,11049.894844,2,97,1,1,0.2,2,2,1,0,0,2,26,1,4,5,NA
+67904,7,2,1,1,21,2,2,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12373.634774,12852.001067,1,92,10,10,2.63,5,5,2,1,0,2,26,1,4,1,4
+67905,7,2,2,10,NA,5,7,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5836.229473,6119.1334,2,103,14,14,3.48,5,5,0,2,1,1,43,1,4,1,5
+67906,7,2,2,8,NA,3,3,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,25563.654853,1,95,15,15,3.62,7,7,2,4,0,1,59,1,5,1,2
+67907,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,74517.751389,89988.433696,2,94,12,12,NA,5,5,1,1,0,1,37,1,4,1,3
+67908,7,2,1,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,23235.97926,24675.846787,2,95,6,6,1.36,3,3,0,0,2,2,60,1,5,1,4
+67909,7,2,1,0,6,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,6218.697738,6608.80804,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+67910,7,2,1,80,NA,2,2,1,NA,NA,1,1,99,NA,NA,NA,9,1,NA,1,1,2,1,1,2,1,1,2,NA,13906.496347,14197.251523,2,103,99,99,NA,6,1,0,0,3,1,80,NA,NA,2,NA
+67911,7,2,1,15,NA,3,3,1,15,186,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,81936.967747,80860.127913,1,94,14,14,2.96,5,5,0,3,0,2,39,1,4,1,3
+67912,7,2,1,65,NA,2,2,1,NA,NA,1,2,2,1,7,NA,5,1,NA,2,2,2,1,2,2,1,2,2,1,14067.170863,14292.508976,2,102,15,15,5,3,3,0,0,2,1,36,2,5,5,NA
+67913,7,2,1,11,NA,1,1,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13665.416457,13541.311863,1,94,2,2,0.27,5,5,0,4,0,2,47,2,1,4,NA
+67914,7,2,1,8,NA,3,3,1,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23188.935049,25672.571973,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+67915,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15976.466658,17286.192057,2,96,8,5,2.2,2,1,0,0,0,2,25,2,5,5,NA
+67916,7,2,2,39,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,4,1,2,2,2,2,2,2,2,NA,NA,NA,NA,38184.257672,39931.153329,2,91,8,8,1.85,5,5,0,2,1,1,39,2,3,1,4
+67917,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,22634.531479,23281.741513,1,103,9,6,2.6,3,1,0,0,0,1,27,1,5,5,NA
+67918,7,2,2,19,NA,4,4,2,19,232,2,NA,2,2,3,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10694.834447,10900.381844,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+67919,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,186864.831109,211461.930376,1,97,3,3,1.3,1,1,0,0,0,2,59,1,4,3,NA
+67920,7,2,2,1,19,2,2,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13464.808163,13893.232889,1,101,15,15,4.99,4,4,2,0,0,1,31,1,4,1,4
+67921,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,51644.110977,57877.507418,1,102,14,14,5,2,2,0,0,1,2,59,1,5,3,NA
+67922,7,2,2,3,NA,5,6,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7483.230909,7475.326886,2,96,15,15,5,3,3,1,0,0,1,34,2,5,1,5
+67923,7,2,2,11,NA,4,4,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8115.309776,8917.717263,2,95,6,6,0.97,6,6,2,2,0,1,37,1,3,1,4
+67924,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19150.604366,19611.675453,3,91,15,15,4.47,4,4,0,1,0,2,45,2,5,1,5
+67925,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,12449.932013,12144.773422,3,90,7,7,2.23,3,3,0,0,0,2,51,1,4,3,NA
+67926,7,2,2,40,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,15819.48188,17091.862943,1,90,15,15,5,3,3,1,0,0,1,42,1,5,1,5
+67927,7,2,1,32,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15137.053737,17738.779486,2,99,15,15,5,1,1,0,0,0,1,32,1,5,5,NA
+67928,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,64581.191728,64779.757488,2,94,15,15,5,2,2,0,0,0,1,36,1,2,1,4
+67929,7,1,2,20,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,NA,NA,NA,NA,18723.98095,0,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+67930,7,2,1,59,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,27595.50738,28151.492876,1,98,10,10,3.51,3,3,0,0,0,1,59,2,5,1,4
+67931,7,2,1,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5991.543696,6284.863687,1,102,5,5,0.86,5,5,2,0,0,2,21,2,2,5,NA
+67932,7,2,1,9,NA,3,3,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,48147.167375,49984.220445,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+67933,7,2,1,12,NA,3,3,1,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,124064.997098,122531.713298,1,95,7,7,2.54,2,2,0,1,0,2,37,1,1,5,NA
+67934,7,2,2,16,NA,3,3,1,16,194,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,113907.203714,124196.980527,1,94,7,7,1.52,4,4,0,2,2,1,61,2,1,1,5
+67935,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,74265.831326,76685.765974,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+67936,7,2,1,13,NA,5,6,2,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,10346.302718,11357.395963,2,91,12,12,NA,7,6,0,4,2,2,72,2,1,2,NA
+67937,7,1,2,74,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15730.58404,0,2,98,77,77,NA,4,4,0,0,2,1,71,NA,NA,1,1
+67938,7,2,2,27,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16026.268721,19326.470131,1,93,14,14,5,1,1,0,0,0,2,27,1,5,5,NA
+67939,7,2,1,72,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,49653.587757,54188.961403,2,98,15,15,5,2,2,0,0,2,2,66,1,3,1,1
+67940,7,2,1,66,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22184.040999,22468.589781,2,94,8,8,3.06,2,2,0,0,2,1,66,1,4,1,4
+67941,7,2,2,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,2,1,2,2,1,2,2,1,2,2,1,35313.648114,35629.376203,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+67942,7,2,1,1,12,2,2,2,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8331.647763,8595.381151,2,90,99,99,NA,5,5,1,1,0,2,40,2,3,1,1
+67943,7,2,1,5,NA,5,7,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6176.631821,6670.596558,1,93,15,15,5,5,5,1,2,0,2,40,1,5,1,5
+67944,7,2,2,10,NA,4,4,2,10,131,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11467.793741,13004.375485,2,91,6,6,0.78,7,7,1,4,0,2,38,2,2,77,NA
+67945,7,2,1,6,NA,3,3,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,38712.032122,41122.784973,3,91,15,15,5,6,6,1,3,0,2,40,1,5,1,5
+67946,7,2,1,3,NA,1,1,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,18754.85406,18780.196288,2,102,5,5,0.89,4,4,2,0,0,1,33,2,9,1,2
+67947,7,2,1,79,NA,4,4,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,2,1,2,1,1,2,2,NA,8074.739647,8488.510868,2,93,13,13,NA,3,3,0,0,2,2,63,2,2,1,3
+67948,7,2,1,27,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,4,5,NA,2,2,2,NA,NA,NA,1,2,2,1,44549.73661,46547.524849,2,93,4,4,0.69,4,4,0,1,1,2,66,2,3,2,NA
+67949,7,2,2,10,NA,2,2,2,10,129,NA,NA,2,2,1,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,10762.400563,12070.126078,2,90,3,3,0.46,5,5,1,3,0,2,35,2,1,4,NA
+67950,7,2,2,4,NA,4,4,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11914.048896,13313.886049,1,100,14,14,3.6,4,4,1,1,0,1,41,1,4,1,5
+67951,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,18174.62037,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+67952,7,2,2,65,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,12994.252166,2,90,2,2,0.55,1,1,0,0,1,2,65,2,1,5,NA
+67953,7,2,1,0,9,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5081.270665,5448.486113,2,99,77,77,NA,6,6,1,1,0,2,42,1,3,1,3
+67954,7,2,1,15,NA,2,2,2,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21399.234084,21768.65302,1,91,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+67955,7,2,2,12,NA,3,3,2,12,147,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71303.309206,77744.474592,2,94,15,15,5,5,5,0,3,0,1,44,1,5,1,4
+67956,7,2,2,3,NA,5,6,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4369.011217,4638.097632,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+67957,7,2,1,50,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,2,3,NA,1,2,2,NA,NA,NA,1,2,2,1,27609.8026,27521.740194,2,101,99,99,NA,3,3,0,1,1,2,78,1,1,2,NA
+67958,7,2,2,43,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,3,1,2,2,1,2,2,NA,NA,NA,NA,9131.449129,11051.567129,1,103,77,77,NA,6,6,0,2,2,1,70,NA,NA,1,1
+67959,7,2,1,13,NA,4,4,1,13,162,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16147.713323,16532.569027,1,92,5,5,0.95,4,4,0,2,0,2,33,1,4,5,NA
+67960,7,2,2,40,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,5,1,2,2,2,2,2,2,2,2,2,2,2,31334.47528,32058.983464,2,96,7,7,1.57,4,4,0,2,0,1,40,2,2,1,5
+67961,7,2,1,6,NA,2,2,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15882.795076,15889.831109,1,97,3,3,0.44,5,5,2,2,0,2,26,1,4,4,NA
+67962,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,2,1,2,2,1,2,2,1,2,2,1,16339.124275,15891.961277,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+67963,7,2,2,11,NA,4,4,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8146.767632,8771.942314,2,100,6,6,0.99,5,5,0,3,0,2,40,1,3,1,3
+67964,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,NA,NA,NA,NA,31335.13799,33684.15133,1,95,6,3,0.45,6,4,1,2,0,1,28,1,2,1,2
+67965,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,52448.388619,54438.22579,1,99,8,8,3.4,2,2,0,0,2,1,74,1,5,1,4
+67966,7,2,1,20,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,101419.325386,103801.228657,1,100,15,15,4.63,5,5,0,0,0,1,51,1,5,1,3
+67967,7,2,1,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,158509.005274,159425.752864,1,100,15,15,4.07,5,5,0,2,0,2,41,1,5,1,4
+67968,7,2,2,7,NA,5,6,2,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6878.565723,7345.278289,3,90,14,14,4.71,3,3,0,1,0,1,43,1,5,1,5
+67969,7,2,2,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,118269.59683,119479.91378,2,98,10,10,4.55,2,2,0,0,1,2,66,1,5,2,NA
+67970,7,2,1,72,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,48254.793439,51076.634961,2,100,99,99,NA,2,2,0,0,2,1,72,1,4,1,4
+67971,7,2,2,11,NA,1,1,1,11,140,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12215.503444,12532.555214,1,102,5,5,0.92,5,5,0,3,0,2,39,2,3,1,3
+67972,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,84910.063417,90031.493834,2,96,14,4,1.38,3,1,0,0,0,1,30,2,5,5,NA
+67973,7,2,1,6,NA,5,6,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11522.32071,1,97,15,15,4.07,5,5,0,3,0,1,42,2,5,1,5
+67974,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37814.382501,44413.458724,1,98,3,3,1.03,1,1,0,0,0,1,27,1,4,5,NA
+67975,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,33731.056243,36109.766611,2,98,4,4,1.22,2,2,0,0,2,1,80,1,1,1,1
+67976,7,1,1,34,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,99134.771466,0,2,99,15,15,5,2,2,0,0,0,2,34,NA,NA,1,5
+67977,7,2,2,56,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,17152.714252,16747.256128,1,102,7,1,0.33,5,1,1,0,2,1,47,1,3,5,NA
+67978,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,49428.481088,50236.061312,1,101,1,1,0.13,1,1,0,0,1,2,60,1,4,3,NA
+67979,7,2,1,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,8232.241159,8296.636338,2,98,7,7,2.72,2,2,0,0,1,1,65,1,4,3,NA
+67980,7,2,2,8,NA,4,4,1,8,105,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8362.256577,8929.488766,2,100,7,7,1.34,5,5,0,2,0,2,53,1,4,4,NA
+67981,7,2,1,19,NA,1,1,2,19,238,2,NA,2,1,4,15,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,20174.283097,21488.447069,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+67982,7,1,1,42,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,148374.146816,0,2,96,10,6,2.66,2,1,0,0,0,1,35,1,5,5,NA
+67983,7,2,2,47,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,13568.706187,17300.043727,3,90,6,6,1.98,2,2,0,0,0,2,47,2,5,2,NA
+67984,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,85040.351078,89341.219459,1,93,7,7,2.16,3,3,0,1,0,2,50,1,5,3,NA
+67985,7,2,1,13,NA,3,3,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58479.782556,60361.370686,1,99,15,15,5,5,5,0,3,0,2,43,1,5,1,5
+67986,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,8411.959333,8787.544319,1,96,9,9,3.97,2,2,0,0,1,1,28,1,5,5,NA
+67987,7,2,2,6,NA,4,4,2,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7446.876055,7952.015759,2,99,6,6,1.15,5,5,1,2,0,2,34,1,4,77,NA
+67988,7,2,1,14,NA,3,3,2,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,58479.782556,60361.370686,1,99,15,15,5,5,5,0,3,0,2,43,1,5,1,5
+67989,7,2,2,22,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,NA,NA,NA,NA,55392.206282,57222.234271,3,91,3,3,1.29,1,1,0,0,0,2,22,1,3,5,NA
+67990,7,2,2,12,NA,5,6,2,12,148,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6937.463063,7205.726089,3,90,10,10,2.41,5,5,1,2,0,1,44,2,4,1,5
+67991,7,2,1,57,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11690.444016,12095.591726,3,90,4,4,0.92,3,3,0,0,1,2,56,2,2,1,2
+67992,7,2,1,41,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,52941.648658,52646.945429,2,102,15,15,5,2,2,0,0,0,1,41,2,4,6,NA
+67993,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,5,NA,2,2,2,2,2,2,1,2,2,NA,8609.250304,11228.904188,2,90,4,4,0.57,5,5,1,0,2,2,80,2,1,2,NA
+67994,7,2,1,12,NA,3,3,2,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71458.892941,71654.216305,2,94,3,3,0.54,4,4,0,1,0,2,48,1,3,1,3
+67995,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,58072.588922,61468.555023,1,101,14,14,5,2,2,0,0,2,2,70,1,5,1,2
+67996,7,2,1,9,NA,1,1,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,2,2,2,2,11367.678664,11547.566042,2,96,5,5,0.78,5,5,0,2,0,1,37,2,1,5,NA
+67997,7,1,2,25,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,11934.941038,0,3,90,8,8,1.85,5,5,0,0,1,2,25,1,5,5,NA
+67998,7,2,2,52,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,17991.883465,18228.515837,2,102,10,10,3.62,3,3,0,0,0,1,51,2,5,1,5
+67999,7,2,2,52,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,24004.6026,25447.359355,2,91,14,14,4.19,3,3,0,1,0,1,55,2,3,1,5
+68000,7,2,1,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21941.544332,21717.573526,1,99,10,10,2.58,5,5,0,1,2,1,65,1,5,1,3
+68001,7,2,1,79,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,NA,8992.410435,11769.853451,2,90,4,4,1.43,1,1,0,0,1,1,79,1,2,4,NA
+68002,7,2,2,11,NA,3,3,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55469.656717,57246.428971,1,98,9,9,2.15,5,5,0,3,0,2,32,1,3,1,4
+68003,7,2,1,72,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,15262.668516,16210.38509,1,102,7,7,1.41,5,5,0,2,2,1,72,1,4,1,3
+68004,7,2,2,33,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,36002.138941,36147.943555,1,103,15,15,5,2,2,0,0,0,1,36,2,5,1,5
+68005,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,99,2,0.55,2,1,0,0,0,1,21,NA,NA,5,NA
+68006,7,2,1,13,NA,4,4,2,13,167,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15970.206483,16374.875978,2,97,3,3,0.46,5,5,0,3,0,1,40,1,2,1,3
+68007,7,2,2,33,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23038.68441,23821.989403,1,92,6,6,1.92,2,2,0,0,0,2,33,2,4,5,NA
+68008,7,2,1,80,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,3,2,NA,2,2,2,2,2,2,NA,NA,NA,NA,9710.399795,10186.406299,2,93,99,99,NA,1,1,0,0,1,1,80,2,3,2,NA
+68009,7,2,2,14,NA,4,4,1,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13697.127402,13594.601175,2,100,10,7,2.05,4,3,0,2,0,1,20,1,4,6,NA
+68010,7,2,1,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11218.730451,2,97,3,3,1.29,1,1,0,0,1,1,63,1,2,2,NA
+68011,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,47973.37979,48741.666001,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+68012,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,12537.004291,13376.026635,2,98,77,77,NA,2,2,0,0,2,2,70,1,3,1,2
+68013,7,2,2,60,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14067.755128,14924.32642,2,91,8,8,4.3,1,1,0,0,1,2,60,2,5,5,NA
+68014,7,2,1,16,NA,3,3,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30100.326038,29873.584609,1,101,6,6,1.17,4,4,0,1,0,1,41,1,3,6,NA
+68015,7,2,2,9,NA,3,3,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19824.800404,20630.46787,1,101,4,4,0.78,4,4,1,2,0,2,32,1,3,3,NA
+68016,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,159097.269198,158226.612798,1,97,15,15,4.77,4,4,0,0,0,1,56,1,4,1,4
+68017,7,2,2,68,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16352.915834,18960.412859,3,92,10,10,4.3,5,2,2,1,1,2,68,1,3,1,1
+68018,7,2,1,1,22,5,6,2,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5891.941477,6363.138629,1,90,6,6,0.92,6,6,2,0,2,2,30,2,5,1,5
+68019,7,2,1,12,NA,3,3,1,12,149,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71934.689876,71392.8162,1,100,15,15,5,5,5,0,3,0,1,47,1,5,1,5
+68020,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,35334.703093,40990.264786,1,101,3,3,1.25,1,1,0,0,1,2,80,1,2,2,NA
+68021,7,2,1,34,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,20803.970543,21211.256549,1,93,5,5,1.84,1,1,0,0,0,1,34,2,2,5,NA
+68022,7,2,1,58,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,37426.314738,36877.141263,3,92,15,15,5,2,2,0,0,0,1,58,1,4,1,4
+68023,7,2,1,25,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,5,NA,1,2,2,NA,NA,NA,1,2,2,1,9177.295801,9548.31812,2,92,77,77,NA,4,4,0,0,0,1,27,2,2,5,NA
+68024,7,2,2,26,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,2,2,2,1,2,2,NA,NA,NA,NA,30253.427014,34374.819903,2,90,6,6,2.01,2,2,0,1,0,2,26,1,3,5,NA
+68025,7,2,2,19,NA,2,2,1,19,238,2,NA,2,2,3,15,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,18556.092615,19088.608674,1,103,6,6,0.93,5,5,0,1,0,1,39,2,3,1,3
+68026,7,2,2,1,19,5,7,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6683.092466,7272.710878,3,91,8,8,2.7,3,3,1,0,0,1,31,1,5,1,5
+68027,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,18490.479848,18604.849683,2,100,8,8,2.7,3,3,1,0,0,2,41,1,4,1,3
+68028,7,2,2,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,14994.337564,16150.13007,1,100,9,9,3.97,2,2,0,0,2,1,70,NA,NA,1,3
+68029,7,2,1,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,7973.883342,8285.757609,1,100,3,3,0.9,1,1,0,0,1,1,60,1,3,5,NA
+68030,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,89234.06428,94763.645369,1,93,15,15,5,4,3,0,0,3,1,80,1,5,2,NA
+68031,7,2,1,2,NA,1,1,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10276.786905,10398.280565,1,102,4,4,0.5,6,6,2,2,0,1,25,1,2,1,3
+68032,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,36401.782557,40656.559204,2,100,6,6,2.01,2,2,0,0,2,1,80,1,5,1,5
+68033,7,2,1,2,NA,3,3,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49276.9767,53002.397569,2,98,8,8,2.42,4,4,2,0,0,2,31,1,4,1,2
+68034,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,11190.39853,11690.037855,2,99,99,99,NA,2,2,0,0,2,2,62,1,5,3,NA
+68035,7,2,2,78,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,55793.89654,57702.040299,2,103,12,12,NA,1,1,0,0,1,2,78,1,4,2,NA
+68036,7,2,1,3,NA,1,1,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14505.510202,14266.522334,2,94,5,5,1.3,3,3,1,0,0,2,38,2,1,1,4
+68037,7,2,1,13,NA,5,6,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6666.045669,7124.269577,3,90,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+68038,7,2,1,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7238.674935,7295.298127,1,96,15,15,5,3,3,0,0,2,2,44,1,5,5,NA
+68039,7,2,2,0,1,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7849.042868,8059.930457,2,98,14,14,3.25,5,5,2,1,0,1,37,1,5,1,5
+68040,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,29738.952706,30201.133298,2,94,8,8,3.4,2,2,0,0,0,1,22,1,3,5,NA
+68041,7,2,1,63,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,6441.264127,6693.19464,1,96,15,15,5,2,2,0,0,2,1,63,1,3,1,3
+68042,7,2,1,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,32375.321924,34954.784356,1,98,2,2,0.31,3,3,1,0,0,1,45,NA,NA,1,NA
+68043,7,2,2,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,79351.077678,83837.616505,2,98,9,9,4.01,2,2,0,0,0,1,27,1,5,1,4
+68044,7,2,2,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,187291.098551,189225.431861,2,91,7,7,2.31,2,2,0,0,0,2,58,1,5,3,NA
+68045,7,2,2,13,NA,4,4,2,13,162,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10671.280357,10591.403308,1,99,10,10,3.13,4,4,0,2,0,1,35,1,4,1,5
+68046,7,2,2,65,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,149575.283913,151406.873597,1,95,5,5,1.84,1,1,0,0,1,2,65,1,3,3,NA
+68047,7,2,1,7,NA,5,6,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11461.625841,1,97,15,15,5,3,3,0,1,0,2,36,2,5,1,5
+68048,7,2,1,10,NA,4,4,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9699.683862,9815.96792,2,100,1,1,0.06,3,3,1,1,0,2,30,1,4,5,NA
+68049,7,2,2,6,NA,5,6,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9227.090107,9736.375878,2,100,15,15,5,4,4,1,1,0,1,41,2,5,1,5
+68050,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,21143.97379,22281.995197,1,97,77,77,NA,3,3,0,0,3,2,62,1,5,1,NA
+68051,7,1,1,51,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,24595.31055,0,1,97,15,15,5,4,4,0,2,0,1,51,1,5,1,NA
+68052,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18404.681357,22792.166915,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+68053,7,2,1,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,18754.85406,20094.472571,2,102,10,10,3.04,4,4,2,0,0,2,31,2,2,1,NA
+68054,7,2,2,8,NA,3,3,2,8,98,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,43531.157975,44359.496292,1,91,7,7,1.88,4,4,1,2,0,2,43,1,5,4,NA
+68055,7,2,1,0,1,5,6,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5300.922196,5735.896742,3,90,7,7,1.82,4,4,1,0,0,2,54,2,1,3,NA
+68056,7,2,2,6,NA,4,4,2,6,83,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7136.421849,7489.546444,1,90,9,9,1.65,7,7,0,4,0,1,36,1,4,1,4
+68057,7,1,1,6,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,66000.998272,0,1,90,15,15,5,4,4,1,1,0,2,39,1,5,1,4
+68058,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,123771.419917,129271.862102,2,98,8,8,2.7,3,3,0,0,1,2,71,NA,NA,2,NA
+68059,7,1,1,29,NA,1,1,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,35782.041084,0,2,96,3,3,0.54,4,4,1,1,0,1,29,1,2,1,2
+68060,7,2,1,18,NA,3,3,1,18,217,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,69211.537407,74163.290023,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+68061,7,2,2,35,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21619.283038,21815.3204,2,102,15,15,3.92,5,5,1,2,0,1,34,2,5,1,5
+68062,7,2,1,28,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,59682.963348,63721.06052,2,102,10,7,3.67,2,1,0,0,0,2,27,1,4,6,NA
+68063,7,2,1,51,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,27227.937106,28128.07977,1,101,1,1,0.27,1,1,0,0,0,1,51,1,3,3,NA
+68064,7,2,2,80,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,17057.278872,18910.504828,2,98,2,2,0.82,1,1,0,0,1,2,80,1,3,2,NA
+68065,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114993.808573,116714.079488,1,98,7,3,0.9,4,1,0,0,0,2,20,1,4,5,NA
+68066,7,2,2,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,9943.806181,2,100,14,14,5,2,2,0,0,2,2,63,1,5,1,4
+68067,7,2,2,2,NA,5,6,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8818.200077,9596.189182,2,102,9,9,2.68,4,4,1,1,0,2,38,2,5,1,2
+68068,7,2,2,79,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,74947.363106,78666.944453,2,91,15,15,5,2,2,0,0,2,2,79,1,5,1,5
+68069,7,2,2,66,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,3,5,NA,2,2,2,2,2,2,NA,NA,NA,NA,10136.963678,11753.317783,2,94,2,2,0.41,2,2,0,1,1,2,66,2,3,5,NA
+68070,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19692.655418,21458.476044,2,99,13,13,NA,3,3,0,0,1,1,80,1,2,2,NA
+68071,7,1,1,27,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,11977.649578,0,2,92,15,10,5,3,1,0,0,0,1,29,1,5,5,NA
+68072,7,2,1,14,NA,3,3,1,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,75412.415858,75196.316989,2,98,15,15,5,4,4,0,2,0,2,46,1,4,1,NA
+68073,7,2,1,58,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,26124.541799,26375.922323,1,94,4,4,1.34,4,1,0,0,0,1,58,1,4,6,NA
+68074,7,2,1,1,16,5,7,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8361.302835,9219.703624,1,101,13,13,NA,3,3,1,0,0,2,19,1,2,NA,NA
+68075,7,2,1,59,NA,5,6,2,NA,NA,1,1,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15693.68983,16353.841376,1,94,7,7,1.79,4,4,0,1,0,1,59,2,4,1,4
+68076,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,101896.080785,107317.764998,3,91,6,6,2.3,1,1,0,0,0,1,30,1,4,3,NA
+68077,7,2,2,72,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,54256.870337,56112.448006,2,91,12,12,NA,2,2,0,0,2,1,76,1,5,1,2
+68078,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,2,99,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,13392.164122,13834.479479,3,90,15,15,5,3,3,0,0,0,1,46,2,3,1,3
+68079,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,37307.740846,37344.375392,1,91,5,5,1.36,2,2,0,1,0,2,49,1,5,3,NA
+68080,7,2,2,39,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,19741.154437,19920.161476,1,100,5,5,0.74,6,6,0,3,0,1,40,2,3,1,4
+68081,7,2,2,1,21,4,4,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,7573.246603,2,100,8,8,2.33,4,4,1,0,0,2,50,1,4,3,NA
+68082,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22570.662508,23423.888442,1,95,12,12,NA,5,5,1,1,0,2,46,1,4,1,4
+68083,7,2,1,10,NA,2,2,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12577.115885,12649.318947,2,96,7,7,1.57,4,4,0,2,0,1,40,2,2,1,5
+68084,7,2,1,20,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,1,1,0.14,2,1,0,0,0,1,20,1,4,5,NA
+68085,7,2,2,16,NA,1,1,2,16,194,NA,NA,2,2,4,9,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,18368.872199,19023.186366,2,94,4,4,0.63,6,6,1,2,0,2,36,2,3,1,1
+68086,7,2,2,27,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,1,2,2,2,2,2,2,2,1,2,2,2,50915.06085,50693.303376,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+68087,7,2,1,12,NA,4,4,2,12,154,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13742.402649,13769.572504,1,96,15,15,5,2,2,0,1,0,2,39,1,4,5,NA
+68088,7,2,2,10,NA,5,7,1,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5139.061504,5511.809249,2,103,15,15,5,3,3,0,1,0,2,37,2,5,1,5
+68089,7,2,1,46,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18335.522037,18907.54446,1,96,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+68090,7,2,1,13,NA,3,3,1,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19381.771783,19434.749274,1,102,4,4,0.97,3,3,0,1,0,2,19,1,2,NA,NA
+68091,7,2,2,74,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,2,2,2,NA,17057.278872,19050.04708,2,98,13,13,NA,1,1,0,0,1,2,74,1,1,2,NA
+68092,7,2,2,11,NA,5,7,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6137.256046,6434.751859,1,103,15,15,3.7,5,5,0,2,1,1,55,1,5,1,5
+68093,7,2,1,47,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18186.25816,18751.952628,2,101,9,9,3.74,2,2,0,0,0,2,40,2,4,1,2
+68094,7,2,1,3,NA,2,2,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14716.463544,15182.304505,2,96,1,1,0.06,5,5,2,1,0,1,27,2,3,1,4
+68095,7,2,2,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,35126.205635,35295.519265,1,92,4,4,0.5,6,6,0,3,0,2,41,1,4,1,NA
+68096,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,28280.669788,29879.976389,1,99,77,77,NA,2,2,0,0,2,2,80,1,4,1,4
+68097,7,2,2,8,NA,4,4,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9399.281543,9696.12347,2,96,3,3,0.54,4,4,2,1,0,2,25,1,4,2,NA
+68098,7,2,2,9,NA,3,3,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12892.545573,15136.637829,1,99,6,6,1.12,4,4,0,2,0,1,39,1,3,1,3
+68099,7,2,2,35,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,1,1,2,2,2,2,2,2,2,1,2,2,2,34898.504426,33956.869463,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+68100,7,2,2,65,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,1,2,1,1,2,2,1,2,1,NA,11313.630983,11818.772505,1,93,2,2,0.54,2,2,0,0,2,1,76,2,4,1,1
+68101,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,11740.600601,12693.223305,3,91,4,4,1.16,2,2,0,0,2,1,80,1,5,1,5
+68102,7,2,1,75,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,1,2,2,NA,12962.876803,16930.790311,2,90,3,3,0.46,5,5,0,2,2,1,75,2,1,1,2
+68103,7,2,2,0,3,4,4,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4099.350341,4514.494508,2,99,6,6,1.39,4,4,1,0,1,2,63,1,3,3,NA
+68104,7,2,1,18,NA,5,7,1,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14831.338089,14976.576871,1,98,3,3,0.43,4,4,0,1,0,2,39,1,2,5,NA
+68105,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16675.763807,16960.86402,2,95,12,12,NA,3,3,0,0,2,1,65,1,4,1,4
+68106,7,2,2,72,NA,3,3,1,NA,NA,2,NA,2,1,9,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,69886.968852,72277.093789,1,102,7,7,2.86,2,2,0,0,2,1,73,1,5,1,3
+68107,7,2,2,8,NA,3,3,1,8,105,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,64166.795496,75335.746424,1,101,5,5,0.89,5,5,1,2,0,1,31,1,2,1,1
+68108,7,2,1,2,NA,4,4,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6314.243514,6704.025386,2,97,5,5,0.76,5,5,1,1,0,2,47,1,4,5,NA
+68109,7,2,1,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18790.641284,20754.743742,2,100,8,8,2.7,3,3,1,0,0,2,41,1,4,1,3
+68110,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,10,10,4.49,2,2,0,0,2,1,60,1,5,1,4
+68111,7,2,2,12,NA,1,1,1,12,150,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17490.464019,17826.708822,1,102,15,15,4.47,4,4,0,2,0,2,30,1,4,1,4
+68112,7,2,1,52,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17950.706927,18011.264999,2,94,15,15,4.44,5,5,0,1,1,2,74,1,5,2,NA
+68113,7,2,2,0,6,1,1,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7565.515117,7406.345541,1,100,1,1,0.09,4,4,2,0,0,2,28,2,2,1,2
+68114,7,2,2,16,NA,3,3,2,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,71303.309206,77744.474592,2,94,15,15,5,3,3,0,1,1,1,63,1,5,1,3
+68115,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,44066.377955,48262.000685,2,97,5,5,2.11,1,1,0,0,1,2,80,1,4,2,NA
+68116,7,2,1,8,NA,1,1,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14042.313177,2,98,3,3,0.54,3,3,0,2,0,2,35,1,3,5,NA
+68117,7,2,1,28,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,23022.732862,27069.827485,2,102,14,14,3.25,5,5,1,1,0,2,32,1,4,1,3
+68118,7,2,2,18,NA,4,4,2,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,10154.02528,10568.655787,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+68119,7,2,1,30,NA,4,4,1,NA,NA,2,NA,2,2,3,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,30195.362375,34424.306144,2,102,8,8,4.59,1,1,0,0,0,1,30,2,4,5,NA
+68120,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,26388.213487,26374.704746,2,96,6,3,1.01,2,1,0,0,0,1,25,NA,NA,6,NA
+68121,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,62513.129282,66168.769092,2,95,15,15,5,2,2,0,0,2,1,70,1,5,1,5
+68122,7,2,1,70,NA,3,3,2,NA,NA,1,9,1,1,NA,NA,4,1,NA,1,1,2,1,2,2,NA,NA,NA,NA,68074.313029,72055.159478,1,101,9,9,4.13,2,2,0,0,2,1,70,1,4,1,3
+68123,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,1,2,1,1,2,1,2,2,NA,15437.181938,16772.913227,2,98,3,3,0.97,1,1,0,0,1,1,80,1,1,5,NA
+68124,7,2,1,22,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,14385.653726,15564.966804,2,101,14,6,2.3,2,1,0,0,0,1,22,2,4,5,NA
+68125,7,2,1,15,NA,4,4,2,15,190,NA,NA,2,1,2,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13969.457688,14694.403205,1,90,14,14,3.25,4,4,0,2,0,2,33,2,3,1,3
+68126,7,2,1,6,NA,3,3,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22394.780012,23678.914206,1,91,4,4,0.81,4,4,1,1,0,1,32,1,4,6,NA
+68127,7,2,2,50,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,11446.604914,11862.967437,2,92,8,8,2.01,4,4,0,0,0,1,53,2,3,1,3
+68128,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25786.235831,27127.570278,1,93,3,3,0.75,2,2,0,0,1,2,80,1,1,2,NA
+68129,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,16494.288293,17728.004854,2,97,2,2,0.84,1,1,0,0,1,2,78,1,2,2,NA
+68130,7,2,1,7,NA,3,3,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57057.523607,61384.115788,1,98,9,9,2.15,5,5,0,3,0,2,32,1,3,1,4
+68131,7,2,2,39,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,34898.504426,34194.957212,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+68132,7,2,1,5,NA,1,1,2,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14505.510202,14964.673559,2,94,4,4,0.63,6,6,1,2,0,2,36,2,3,1,1
+68133,7,2,1,44,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,33029.272844,33302.556931,2,93,7,7,1.52,4,4,1,1,0,1,44,2,4,1,NA
+68134,7,2,1,32,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,4,6,NA,2,2,2,2,2,2,1,2,2,2,41155.167164,40844.556107,1,102,13,13,NA,6,6,1,2,0,2,36,2,4,6,NA
+68135,7,2,1,50,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16590.074977,17115.36835,1,92,15,15,5,3,3,1,0,0,1,50,1,4,1,4
+68136,7,2,1,15,NA,2,2,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19570.996814,20425.923806,2,100,4,4,0.81,4,4,0,2,0,2,37,1,2,1,2
+68137,7,2,2,68,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,2,4,NA,1,2,1,1,2,1,1,2,1,NA,9991.888445,10532.363744,3,90,8,4,1.72,3,1,0,0,1,2,68,2,2,4,NA
+68138,7,2,2,5,NA,3,3,2,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78165.891242,86270.974005,1,101,14,14,4.21,4,4,1,1,0,2,37,1,5,1,5
+68139,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19210.136544,19335.440888,1,96,15,15,5,2,2,0,0,0,1,48,1,2,1,3
+68140,7,2,1,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19008.083201,18737.138245,2,97,1,1,0,2,2,0,0,1,2,63,1,4,5,NA
+68141,7,2,2,4,NA,4,4,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10544.002566,11345.464378,1,91,6,6,0.99,5,5,3,0,0,2,33,2,3,1,4
+68142,7,2,1,36,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15732.465299,16321.860662,2,92,14,6,2.75,2,1,0,0,0,1,48,NA,NA,5,NA
+68143,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,4,NA,2,2,2,1,2,2,2,2,1,NA,16623.349489,17874.484754,2,93,6,6,1.41,3,3,0,1,1,2,80,2,4,4,NA
+68144,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,15083.375446,14341.447668,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+68145,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,9221.19173,9581.850682,2,95,6,6,1.65,2,2,0,0,2,1,62,1,1,1,3
+68146,7,2,2,6,NA,5,7,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22371.648216,22397.198208,1,92,3,3,0.46,5,5,2,1,0,1,30,1,3,1,2
+68147,7,2,1,75,NA,4,4,1,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,9662.124837,9885.036833,2,92,3,3,1.1,1,1,0,0,1,1,75,2,4,3,NA
+68148,7,2,1,9,NA,1,1,1,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14258.502552,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+68149,7,2,1,8,NA,1,1,1,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11399.23838,11468.323492,2,103,77,77,NA,5,5,1,2,0,2,30,1,2,1,2
+68150,7,2,1,13,NA,2,2,1,13,167,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,19820.949231,20163.121943,2,91,8,8,1.85,5,5,0,2,1,1,39,2,3,1,4
+68151,7,2,1,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16088.355002,15889.986896,2,100,3,3,0.92,1,1,0,0,0,1,29,1,2,5,NA
+68152,7,2,1,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16737.888813,2,100,14,14,4.86,3,3,0,0,0,2,52,1,5,1,3
+68153,7,2,1,14,NA,5,6,1,14,170,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6015.001712,6229.601178,1,102,5,5,0.92,5,5,1,2,0,2,44,2,1,1,2
+68154,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15599.953109,15843.647059,1,99,15,15,5,2,2,0,0,0,1,56,1,4,1,4
+68155,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,25189.042335,24499.678113,1,100,10,10,2.59,5,5,0,1,0,2,40,1,5,1,NA
+68156,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,13934.943848,14557.124186,2,96,12,10,5,2,1,0,0,1,1,53,1,4,3,NA
+68157,7,2,2,24,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,14756.436992,15386.562828,2,95,14,8,4.59,2,1,0,0,0,2,24,1,4,6,NA
+68158,7,2,1,7,NA,1,1,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8828.580268,9491.612368,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+68159,7,2,1,66,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12579.986433,13271.133625,1,92,6,6,2.15,2,2,0,0,2,2,61,2,4,1,5
+68160,7,2,1,11,NA,3,3,2,11,143,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24713.905595,26080.214484,1,98,6,6,0.97,7,7,1,2,0,1,49,1,2,1,2
+68161,7,2,1,10,NA,4,4,1,10,131,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8894.789377,9317.745565,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+68162,7,2,1,0,9,3,3,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17196.879565,17846.963436,2,94,99,99,NA,6,6,2,0,0,2,26,1,4,1,NA
+68163,7,2,2,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20729.303307,20166.249394,1,90,14,14,4.98,3,3,1,0,0,2,33,2,5,1,5
+68164,7,2,2,14,NA,1,1,1,14,169,NA,NA,2,2,3,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,26325.414456,26852.811114,3,92,4,4,0.67,4,4,0,3,0,2,36,2,1,5,NA
+68165,7,2,2,0,5,2,2,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6850.346072,7305.598313,1,90,3,3,0.43,4,4,2,0,0,1,31,1,3,6,NA
+68166,7,2,1,6,NA,5,6,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6631.058951,7488.793181,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+68167,7,1,1,5,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27873.065855,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+68168,7,2,2,2,NA,5,7,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6247.52442,6810.692829,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+68169,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,55695.737629,59946.103917,1,92,5,5,1.15,3,3,1,0,0,1,23,1,4,1,4
+68170,7,2,1,37,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22856.432972,22931.992355,2,98,14,14,3.36,4,4,0,2,0,1,37,1,4,1,4
+68171,7,2,2,58,NA,1,1,1,NA,NA,2,NA,2,2,8,NA,1,4,NA,2,2,2,2,2,2,NA,NA,NA,NA,28349.668436,29490.478086,1,102,1,1,0.33,2,2,0,0,0,1,37,1,2,4,NA
+68172,7,2,2,9,NA,3,3,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,50928.148447,51656.230004,3,91,14,14,3.4,4,4,0,2,0,1,40,1,4,1,4
+68173,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,22824.336433,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+68174,7,2,1,5,NA,3,3,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30883.231636,34843.624635,1,95,6,3,0.45,6,4,1,2,0,1,28,1,2,1,2
+68175,7,2,1,4,NA,2,2,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16169.123686,17324.04909,1,100,10,10,2.91,4,4,1,1,0,1,32,1,5,1,5
+68176,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,17164.211773,17267.434455,2,99,6,6,2.24,1,1,0,0,0,2,58,1,2,5,NA
+68177,7,2,1,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,32476.100686,32780.682518,1,100,4,4,1.19,2,2,0,0,1,1,62,1,5,1,5
+68178,7,2,1,18,NA,5,6,2,18,223,2,NA,2,1,3,15,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,9923.450213,10376.279545,2,100,5,5,0.89,4,4,0,1,0,2,40,2,3,1,3
+68179,7,2,1,17,NA,4,4,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,NA,9178.515376,10829.980579,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+68180,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19799.565045,19517.338151,1,90,14,14,2.96,5,5,1,2,0,1,31,1,5,1,4
+68181,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,15498.463997,16338.715638,1,98,5,5,1.39,2,2,0,0,2,2,71,1,3,1,3
+68182,7,2,1,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,138834.18124,153441.395439,1,100,15,15,4.56,4,4,0,2,0,2,42,1,4,1,3
+68183,7,2,2,28,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,23293.719218,24697.897932,1,92,7,7,1.83,3,3,1,1,0,2,28,1,3,5,NA
+68184,7,2,1,7,NA,4,4,1,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12267.215138,12414.279886,2,102,5,5,0.76,5,5,1,3,0,2,30,1,4,4,NA
+68185,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26847.643051,27369.434392,1,98,4,4,0.67,5,5,1,2,0,1,29,1,4,1,3
+68186,7,2,2,17,NA,1,1,1,17,206,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20370.629189,20955.218165,1,95,15,15,5,3,3,0,1,0,1,50,1,3,1,4
+68187,7,2,1,3,NA,4,4,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7431.820906,7890.591472,1,99,10,10,2.71,5,5,1,1,2,1,75,1,1,1,3
+68188,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,64463.340883,66173.435468,1,91,14,14,5,2,2,0,0,2,2,70,1,2,1,NA
+68189,7,2,2,10,NA,4,4,2,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9326.540969,10100.402919,1,91,8,8,1.76,5,5,0,3,0,2,42,1,3,6,NA
+68190,7,2,2,1,12,1,1,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11438.839305,12628.78556,1,102,14,14,4.32,3,3,1,0,0,1,25,1,4,1,4
+68191,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,29167.119125,34820.124106,1,98,3,3,0.5,5,5,0,3,0,2,56,1,3,3,NA
+68192,7,2,2,5,NA,1,1,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14852.990935,14903.708855,2,94,6,6,0.8,7,7,1,3,0,2,36,2,3,1,1
+68193,7,2,1,17,NA,3,3,2,17,208,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,71458.892941,70519.759062,2,94,10,10,3.51,3,3,0,2,0,2,39,2,4,3,NA
+68194,7,2,1,15,NA,4,4,2,15,180,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10745.098568,10766.342508,1,96,12,12,NA,5,5,1,2,0,2,35,1,5,1,4
+68195,7,2,1,74,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,17306.956173,18319.031536,1,100,4,4,1.16,2,2,0,0,2,1,74,1,3,1,5
+68196,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,125040.051934,125530.45672,1,97,15,15,5,4,4,0,0,1,1,67,NA,NA,2,NA
+68197,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,196252.38045,199712.076327,1,98,14,14,5,2,2,0,0,0,1,59,1,3,1,4
+68198,7,2,1,32,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,17371.064048,17243.514155,1,96,12,12,NA,5,5,2,0,1,2,63,2,5,3,NA
+68199,7,2,2,30,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,29148.354549,29365.982413,2,92,14,14,4.03,4,4,1,1,1,2,30,1,5,4,NA
+68200,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,31155.769617,31636.819209,1,92,14,9,3.97,3,2,0,0,2,1,51,1,4,5,NA
+68201,7,2,2,12,NA,1,1,1,12,153,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,19719.98657,20781.904006,2,102,6,6,1.03,5,5,1,1,0,1,37,1,2,1,2
+68202,7,2,1,9,NA,3,3,2,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,75923.373594,80651.425295,1,94,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+68203,7,2,2,78,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,72551.269339,73326.076054,1,95,4,4,1.12,2,2,0,0,2,2,78,1,4,1,2
+68204,7,2,2,18,NA,3,3,2,18,226,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24820.424867,25966.716212,1,91,3,3,0.62,3,3,0,1,0,2,55,1,4,4,NA
+68205,7,2,1,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15601.50288,15850.523572,2,99,2,2,0.2,7,7,1,2,1,1,63,1,1,2,NA
+68206,7,2,2,35,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,17978.142628,18308.053591,1,91,10,10,3.22,4,4,1,1,0,1,38,2,5,1,5
+68207,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,22013.270774,22643.742168,1,101,3,3,0.66,2,2,0,0,1,2,65,1,2,3,NA
+68208,7,2,1,16,NA,5,7,2,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,5388.361335,5758.757236,3,91,15,15,4.47,4,4,0,3,0,2,44,2,5,1,NA
+68209,7,2,1,15,NA,2,2,1,15,189,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20798.348063,20598.526957,2,93,12,12,NA,3,1,1,1,0,2,43,1,5,3,NA
+68210,7,2,1,27,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,9177.295801,9548.31812,2,92,15,8,4.59,3,1,0,0,0,2,25,1,5,5,NA
+68211,7,2,1,2,NA,3,3,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37959.146468,40828.920669,1,91,6,6,1.62,3,3,1,0,0,1,30,1,4,1,4
+68212,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,1,1,2,NA,16321.652472,17542.456462,2,97,4,4,1.02,2,2,0,0,2,2,80,1,1,2,NA
+68213,7,2,2,41,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,4,2,2,2,2,1,2,2,NA,NA,NA,NA,40880.818805,41857.641766,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+68214,7,2,2,3,NA,5,6,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6858.963321,7464.098007,3,91,15,15,5,3,3,1,0,0,1,40,2,5,1,5
+68215,7,2,2,14,NA,4,4,1,14,178,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13895.342981,13964.270792,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+68216,7,1,1,60,NA,5,6,NA,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11992.891463,0,1,100,15,15,5,2,2,0,0,1,1,60,2,5,1,NA
+68217,7,2,2,19,NA,4,4,2,19,234,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13841.239638,13875.328502,1,96,15,15,4.9,4,4,0,1,0,1,47,1,3,1,5
+68218,7,2,1,1,14,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13968.423539,13738.28453,2,102,6,6,1.62,3,3,1,0,0,1,20,2,4,1,4
+68219,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,25270.027986,25977.972711,1,100,9,6,2.24,3,1,0,0,0,1,25,NA,NA,5,NA
+68220,7,2,1,14,NA,4,4,2,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11834.781205,13035.394237,2,90,2,2,0.31,5,5,0,2,1,2,71,1,2,2,NA
+68221,7,2,1,3,NA,4,4,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8385.172131,8641.288503,2,93,6,2,0.46,3,2,1,1,0,2,31,2,3,3,NA
+68222,7,2,1,36,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,38746.026117,2,98,14,14,2.87,5,5,0,3,0,2,34,1,2,1,2
+68223,7,2,2,9,NA,3,3,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25527.806244,26565.240294,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+68224,7,2,2,11,NA,2,2,2,11,134,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15166.167659,15710.813593,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+68225,7,2,1,2,NA,5,6,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6389.003009,6946.807658,3,91,6,6,1.15,5,5,1,0,0,1,55,2,5,1,5
+68226,7,2,2,38,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,20039.469886,21581.359058,1,97,15,15,5,4,4,1,1,0,1,44,2,5,1,5
+68227,7,2,2,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,25638.18991,26248.270338,2,101,2,2,0.38,3,3,0,2,0,2,56,1,3,2,NA
+68228,7,2,1,11,NA,4,4,2,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6377.235034,7094.374704,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+68229,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,97803.500399,102601.600974,1,101,14,14,5,3,3,0,1,0,2,36,1,5,1,5
+68230,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,59001.303336,68847.35366,2,101,2,2,0.46,1,1,0,0,0,2,22,1,4,5,NA
+68231,7,2,1,13,NA,5,6,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10592.643259,11320.781443,2,91,77,77,NA,3,3,0,1,0,2,43,2,5,1,5
+68232,7,2,2,9,NA,1,1,1,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13064.573334,13403.6626,2,103,2,2,0.22,7,7,0,3,0,2,39,2,1,5,NA
+68233,7,2,2,5,NA,3,3,2,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51483.624552,53105.566664,1,91,15,15,5,4,4,1,1,0,1,38,1,5,1,5
+68234,7,2,1,68,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,4,5,NA,1,2,2,1,2,2,2,2,2,2,12845.115724,13050.878075,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+68235,7,2,1,79,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19473.412374,20529.16646,1,95,4,4,1.22,2,2,0,0,1,1,79,1,1,1,3
+68236,7,2,2,41,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,26595.398371,26621.513872,1,94,4,4,1.26,2,2,0,0,1,2,41,1,4,5,NA
+68237,7,2,1,63,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,120735.071461,122283.708051,1,94,8,8,2.41,3,3,0,0,3,1,63,1,4,1,5
+68238,7,2,2,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10771.85499,11250.180007,2,97,5,5,1.08,3,3,1,1,0,2,27,1,3,5,NA
+68239,7,2,2,7,NA,1,1,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20753.369981,21285.255435,1,97,10,10,2.32,6,6,0,4,0,1,42,1,4,1,4
+68240,7,2,1,19,NA,1,1,2,19,239,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,25099.482648,25097.275151,1,97,4,4,0.65,4,4,0,1,0,2,45,2,2,3,NA
+68241,7,2,1,9,NA,1,1,1,9,114,NA,NA,2,2,3,3,NA,NA,NA,2,1,1,1,2,2,1,2,2,1,19774.151841,21259.203461,3,92,7,7,1.41,5,5,1,2,0,1,20,2,1,1,1
+68242,7,2,1,7,NA,1,1,2,7,85,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9390.522479,9862.720437,2,90,3,3,0.58,4,4,0,2,0,2,36,2,3,1,3
+68243,7,2,1,13,NA,1,1,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22621.505951,23112.921348,1,92,14,14,5,2,2,0,1,0,2,41,1,5,3,NA
+68244,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,29108.403608,1,94,2,2,0.42,3,3,0,0,0,2,52,1,4,1,1
+68245,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,18246.235208,18400.236064,2,99,1,1,0.07,4,4,1,1,0,2,24,1,2,5,NA
+68246,7,2,1,9,NA,5,6,2,9,116,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5377.477719,5731.340117,2,100,4,4,0.5,6,6,2,1,0,1,30,2,4,1,3
+68247,7,2,2,80,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,12833.793728,13278.897744,2,92,77,77,NA,2,2,0,0,2,2,80,1,3,1,4
+68248,7,2,1,15,NA,4,4,1,15,190,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12147.046136,12703.852077,2,100,8,8,1.8,5,5,0,3,0,2,43,1,3,1,3
+68249,7,2,1,60,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,117075.881463,118577.582598,1,94,9,9,3.97,2,2,0,0,1,1,60,1,4,1,4
+68250,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,207644.101324,214736.820497,1,97,15,15,5,2,2,0,0,0,1,50,1,3,6,NA
+68251,7,2,2,34,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,39561.667842,39233.767064,2,98,14,14,2.87,5,5,0,3,0,2,34,1,2,1,2
+68252,7,2,2,45,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,15206.604563,15895.992431,2,90,14,14,4.25,4,4,0,2,1,2,45,2,5,5,NA
+68253,7,2,2,8,NA,3,3,2,8,102,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,63165.080986,62975.31549,1,98,15,15,5,3,3,0,1,0,2,43,1,5,1,5
+68254,7,2,2,55,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15499.383981,15426.291906,2,99,5,5,1.26,3,3,1,0,0,2,50,2,3,5,NA
+68255,7,2,1,60,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,6800.602543,7011.068263,1,99,6,6,1.62,3,3,0,0,1,1,60,2,5,1,3
+68256,7,2,2,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22790.461787,22849.332243,2,98,10,10,3.78,3,3,0,0,0,2,46,1,4,1,4
+68257,7,2,1,12,NA,5,6,1,12,155,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10346.302718,11892.421636,2,91,6,6,1.26,5,5,0,2,0,2,47,2,1,1,1
+68258,7,2,1,3,NA,2,2,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12403.412256,13289.361067,2,90,10,10,3.13,4,4,1,2,0,2,39,1,5,4,NA
+68259,7,2,2,4,NA,5,7,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,95,10,6,1.34,5,4,1,2,0,1,32,1,3,6,NA
+68260,7,2,1,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11218.730451,2,95,3,3,1.29,1,1,0,0,1,1,61,1,2,3,NA
+68261,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,14971.827573,15192.169136,2,97,15,15,5,3,3,0,0,3,2,80,1,3,2,NA
+68262,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,10576.529742,11391.789088,2,99,2,2,0.72,1,1,0,0,1,2,64,1,3,5,NA
+68263,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,10,10,3.78,3,3,0,0,2,1,62,1,5,1,5
+68264,7,2,1,9,NA,1,1,2,9,111,NA,NA,2,2,3,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11917.887108,12700.965428,2,94,7,7,1.34,5,5,2,1,0,1,32,2,1,1,NA
+68265,7,2,1,11,NA,1,1,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11881.117946,11953.12349,1,102,6,6,1.18,5,5,0,2,1,2,42,2,2,2,NA
+68266,7,2,2,11,NA,4,4,2,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8757.841043,9104.895674,2,97,7,7,2.16,3,3,0,1,0,2,31,1,3,6,NA
+68267,7,2,1,3,NA,5,6,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6389.003009,6899.95174,3,91,6,6,1.22,5,5,1,2,0,2,37,1,4,1,2
+68268,7,1,2,6,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10332.067017,0,1,100,3,3,0.73,3,3,1,1,0,2,32,1,3,5,NA
+68269,7,2,2,2,NA,1,1,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13065.99844,14425.21291,1,95,8,8,2.24,4,4,2,0,0,2,29,1,3,1,4
+68270,7,2,2,48,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,6,NA,2,2,2,2,2,2,2,2,2,2,31235.666551,31802.121006,2,94,9,9,4.21,3,2,0,0,0,2,48,2,2,6,NA
+68271,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,127198.447143,126871.700524,1,101,8,8,4.25,1,1,0,0,0,1,43,1,4,5,NA
+68272,7,2,1,1,18,5,7,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26337.322319,30860.353051,1,94,6,6,1.33,4,4,2,0,0,2,29,1,2,1,4
+68273,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,59510.728426,64021.429591,1,99,5,5,0.89,4,4,2,0,0,2,31,1,4,1,5
+68274,7,2,1,12,NA,5,6,1,12,147,NA,NA,2,2,3,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8530.740143,8910.137481,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+68275,7,2,1,71,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,7608.031426,7606.53553,1,99,14,14,5,2,2,0,0,2,1,71,1,4,1,3
+68276,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,31909.283101,36294.223393,2,91,6,6,1.26,5,5,0,1,2,2,80,1,4,2,NA
+68277,7,2,1,61,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,4,1,NA,2,2,2,1,2,2,1,2,2,2,11568.876339,11794.347884,1,102,6,6,1.48,3,3,0,0,1,2,57,2,1,1,4
+68278,7,2,2,0,0,1,1,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7896.364456,7653.697165,2,102,15,15,2.43,7,7,3,2,0,1,28,2,5,1,4
+68279,7,2,1,20,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14424.961621,15273.975784,2,102,8,8,2.01,4,4,0,0,0,1,59,2,4,1,4
+68280,7,2,2,15,NA,3,3,2,15,184,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,91539.042546,96988.607103,1,101,9,9,2.6,4,4,0,2,0,2,38,1,4,1,4
+68281,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17115.540769,2,97,12,6,2.75,3,1,0,0,0,1,21,NA,NA,77,NA
+68282,7,2,2,54,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,13728.308948,13865.439464,2,90,6,6,1.12,4,4,0,1,1,1,63,2,1,1,1
+68283,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,2,1,2,2,1,2,2,3,13155.047649,13224.610785,3,90,15,15,3.7,5,5,0,0,0,1,56,2,3,1,3
+68284,7,2,1,8,NA,1,1,1,8,106,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14007.413517,2,98,15,15,4.97,5,5,0,3,0,1,39,1,5,1,5
+68285,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,110369.721342,114327.60352,1,94,9,9,3.97,2,2,0,0,0,1,49,1,3,1,3
+68286,7,2,2,16,NA,3,3,2,17,205,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,111142.989658,114400.666347,1,95,NA,NA,NA,5,5,0,2,0,2,37,1,3,1,NA
+68287,7,2,1,12,NA,1,1,1,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22768.423624,22944.003607,1,92,5,5,1.24,3,3,0,1,0,2,29,2,3,6,NA
+68288,7,2,1,28,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,34708.958385,37032.750996,2,97,13,13,NA,3,3,0,1,0,1,28,2,2,1,1
+68289,7,2,1,9,NA,4,4,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10199.928366,10774.065998,1,100,5,5,0.85,5,5,0,2,0,2,54,1,2,2,NA
+68290,7,2,2,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,81416.374938,89993.518978,1,97,9,9,3.97,2,2,0,0,2,1,79,1,3,1,3
+68291,7,2,1,69,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,2,2,2,2,2,2,2,2,1,2,6687.985443,7006.085081,2,93,4,4,0.99,2,2,0,0,2,1,69,2,3,1,1
+68292,7,2,1,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12005.116852,11690.420322,3,92,14,14,2.29,7,7,2,0,0,2,50,2,1,1,9
+68293,7,2,1,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,191792.099076,199776.618579,1,98,6,6,1.98,2,2,0,0,0,1,54,1,4,1,3
+68294,7,2,2,11,NA,3,3,2,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55469.656717,57246.428971,1,101,14,14,3.3,4,4,0,2,0,2,42,1,4,1,3
+68295,7,1,2,20,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,6,3,1,2,2,1,2,2,NA,NA,NA,NA,60324.348827,0,1,101,13,13,NA,3,3,1,0,0,2,20,1,2,6,NA
+68296,7,2,2,69,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,46357.185832,48577.343045,2,98,3,3,0.68,2,2,0,0,2,1,80,1,1,1,3
+68297,7,2,1,5,NA,5,6,1,5,63,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10273.602479,11522.969217,1,92,14,14,3.3,4,4,2,0,0,1,28,1,4,1,4
+68298,7,2,1,1,23,2,2,2,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12567.081957,12360.030976,1,97,15,15,5,4,4,1,1,0,1,42,1,5,1,5
+68299,7,2,2,12,NA,4,4,2,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19113.842115,19208.656277,1,97,1,1,0.09,4,4,0,1,0,2,44,2,2,1,3
+68300,7,2,1,16,NA,4,4,2,16,193,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11023.237662,11433.168046,1,99,14,14,4.05,3,3,0,1,0,2,52,1,4,4,NA
+68301,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,169896.236244,2,91,15,2,0.63,7,1,0,0,1,1,49,NA,NA,5,NA
+68302,7,2,1,17,NA,4,4,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11023.237662,11433.168046,1,99,15,15,5,4,4,0,2,0,2,46,1,5,1,5
+68303,7,2,2,16,NA,1,1,2,16,198,NA,NA,2,2,1,10,NA,NA,NA,1,2,2,1,2,2,2,2,2,2,16896.101801,17220.920308,1,93,15,1,0,3,1,0,1,0,2,46,2,5,5,NA
+68304,7,2,1,19,NA,1,1,1,19,232,2,NA,2,2,3,11,NA,NA,NA,2,2,2,2,2,2,1,2,2,2,25268.119938,26371.916437,1,100,99,99,NA,6,6,0,1,0,2,22,2,3,1,3
+68305,7,2,2,17,NA,5,6,1,17,205,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,10081.858636,10347.605805,2,102,15,15,3.82,5,5,0,1,2,1,60,2,2,1,1
+68306,7,2,1,17,NA,3,3,2,17,212,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,88448.252445,87285.839384,1,95,8,1,0.09,4,1,0,1,0,2,57,1,5,5,NA
+68307,7,2,1,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6347.215153,6424.634366,2,95,7,7,1.83,3,3,1,0,0,1,33,1,3,6,NA
+68308,7,2,1,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8361.170561,8618.715512,1,94,4,4,0.56,5,5,1,2,0,1,34,1,2,3,NA
+68309,7,2,1,57,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16499.662173,16747.308945,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+68310,7,2,2,47,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20303.639991,21936.687597,1,97,10,10,3.67,3,3,0,1,0,1,47,1,5,1,5
+68311,7,2,2,33,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,3,1,1,1,2,2,1,2,2,1,2,2,1,18018.210636,18055.00232,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+68312,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,62471.259637,62663.338056,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+68313,7,2,1,5,NA,3,3,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,67064.292398,74209.90502,2,91,15,15,5,5,5,2,1,0,2,40,1,5,1,5
+68314,7,2,1,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16995.30979,18681.886374,2,103,6,6,1.3,4,4,1,1,0,2,26,1,4,1,3
+68315,7,2,2,10,NA,1,1,1,10,121,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15510.382876,18634.445505,1,100,6,6,1.11,5,5,0,2,1,1,38,2,2,1,1
+68316,7,2,2,70,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,13204.205822,14198.003586,1,98,9,9,2.49,4,4,0,1,2,2,70,1,4,2,NA
+68317,7,2,2,19,NA,3,3,1,19,237,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,36241.268921,38398.808798,2,101,2,1,0.28,2,1,0,0,0,2,21,1,4,5,NA
+68318,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,18693.365067,19341.691824,1,92,12,12,NA,7,7,1,2,1,2,45,2,3,1,3
+68319,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,90708.718111,91145.947869,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+68320,7,2,1,0,1,5,7,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6420.539566,6377.399312,2,99,15,15,5,3,3,1,0,0,1,35,2,5,1,5
+68321,7,2,1,80,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,28658.58153,30983.915333,2,94,8,8,3.4,2,2,0,0,2,1,80,1,3,1,5
+68322,7,2,1,32,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,30626.581617,32055.135325,2,90,6,6,1.62,3,3,1,0,0,2,28,1,5,1,4
+68323,7,2,2,16,NA,5,7,1,16,202,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,23414.779924,23738.339087,1,102,5,5,1.27,3,3,0,2,0,2,38,1,2,3,NA
+68324,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,90179.087212,91826.282843,2,95,14,14,5,2,2,0,0,0,1,39,1,4,1,5
+68325,7,2,1,2,NA,4,4,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5662.231921,6243.536586,1,99,2,2,0.31,4,4,1,0,1,2,67,1,3,3,NA
+68326,7,2,1,12,NA,4,4,2,12,152,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,NA,NA,NA,1,2,2,1,10366.393886,10567.129261,1,90,9,9,1.65,7,7,0,4,0,1,36,1,4,1,4
+68327,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,26019.443397,28034.871168,2,100,6,6,2.01,2,2,0,0,2,1,80,1,5,1,5
+68328,7,2,1,5,NA,3,3,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,56162.95725,63365.162813,2,95,15,15,4.63,5,5,1,2,0,2,36,1,5,1,3
+68329,7,2,2,15,NA,1,1,1,15,188,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18515.058419,19512.090709,2,96,7,7,1.79,4,4,0,2,0,1,43,2,3,1,2
+68330,7,2,2,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,133492.667054,133041.068157,1,95,6,6,1.94,2,2,0,0,2,2,69,1,2,1,4
+68331,7,2,1,66,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14488.953694,14771.336069,3,92,14,14,5,2,2,0,0,2,1,66,1,4,1,4
+68332,7,2,2,17,NA,3,3,2,17,205,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,123622.182994,127245.633335,2,94,7,7,2.72,2,2,0,1,0,2,43,1,3,3,NA
+68333,7,2,1,2,NA,3,3,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,43818.485,49773.383799,2,91,10,10,2.5,5,5,1,0,0,1,57,1,9,1,3
+68334,7,2,2,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,26682.998353,27885.908424,2,99,3,3,0.56,4,4,1,0,0,2,38,1,3,5,NA
+68335,7,1,2,14,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7969.467397,0,2,103,15,15,4.34,4,4,0,2,0,1,48,2,5,1,5
+68336,7,2,2,1,22,1,1,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8832.868731,8863.029978,1,102,6,6,0.8,7,7,3,3,0,2,34,2,3,1,1
+68337,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,4,1,NA,2,2,2,2,2,2,1,2,2,2,41241.224595,42341.927552,2,102,10,10,3.04,4,4,2,0,0,2,31,2,2,1,NA
+68338,7,2,1,51,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,2,2,2,2,2,2,2,2,1,2,25108.558777,27044.812277,2,93,4,4,0.82,4,4,0,0,0,1,51,2,3,1,3
+68339,7,2,1,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,18790.641284,20754.743742,2,100,4,4,0.85,4,4,0,2,0,2,39,1,3,6,NA
+68340,7,2,1,10,NA,3,3,2,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21087.869274,22876.159598,1,101,4,4,0.78,4,4,1,2,0,2,32,1,3,3,NA
+68341,7,2,2,64,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,2,2,2,1,10235.0654,10993.258671,2,93,10,10,3.04,4,4,0,0,2,1,72,2,3,1,2
+68342,7,2,1,59,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,1,6,NA,1,2,1,1,2,1,NA,NA,NA,NA,16499.662173,16440.055989,3,91,6,4,1.38,3,1,0,0,0,1,59,2,1,6,NA
+68343,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,17206.320427,17285.613667,1,96,2,2,0.4,3,3,0,0,0,2,56,1,3,3,NA
+68344,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,152858.509804,152465.847768,1,95,14,7,3.67,2,1,0,0,0,1,47,1,4,3,NA
+68345,7,2,2,18,NA,4,4,1,18,223,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,2,1,0.09,4,1,0,0,0,2,19,1,4,NA,NA
+68346,7,2,1,0,4,4,4,2,NA,4,NA,NA,2,2,99,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6327.979262,6425.104835,1,90,13,13,NA,3,3,2,0,0,2,21,2,4,5,NA
+68347,7,2,2,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20899.681083,21326.972731,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+68348,7,2,1,32,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,51543.062078,54569.929737,3,92,14,14,3.25,4,4,2,0,0,2,33,1,5,1,5
+68349,7,1,2,12,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24869.937631,0,2,93,6,6,1.48,4,4,0,1,0,1,53,2,2,1,3
+68350,7,2,2,12,NA,3,3,1,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,93740.540203,99321.165816,1,100,15,15,4.56,4,4,0,2,0,2,42,1,4,1,3
+68351,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,90303.174138,92885.29608,2,92,15,15,5,4,1,0,0,0,1,27,NA,NA,5,NA
+68352,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,30548.472436,32702.747345,1,95,1,1,0.21,4,4,1,0,1,2,75,1,1,2,NA
+68353,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,94698.084211,100770.654731,1,101,14,14,4.21,4,4,1,1,0,2,37,1,5,1,5
+68354,7,2,1,45,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,1,1,2,2,1,2,1,NA,14879.667962,14825.914121,3,91,4,4,0.69,5,5,0,2,0,1,45,2,4,1,1
+68355,7,2,2,19,NA,4,4,2,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,1,1,0.05,1,1,0,0,0,2,19,1,4,NA,NA
+68356,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,15616.794261,15846.627851,2,97,7,7,3.49,1,1,0,0,1,2,73,1,4,3,NA
+68357,7,2,1,1,12,3,3,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26337.322319,28328.46741,1,94,3,3,0.93,2,2,1,0,0,2,25,1,5,3,NA
+68358,7,2,2,0,5,1,1,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6787.112205,6644.319314,2,94,3,3,0.54,3,3,1,0,0,2,21,1,4,1,3
+68359,7,1,2,11,NA,2,2,NA,NA,NA,NA,NA,2,1,4,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15821.530831,0,2,91,6,6,0.93,5,5,1,2,0,2,50,2,1,5,NA
+68360,7,2,1,13,NA,5,6,1,13,166,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12013.02156,13498.577364,3,92,12,12,NA,5,5,0,2,0,1,47,1,3,1,3
+68361,7,2,1,8,NA,1,1,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,13898.598114,14118.535925,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+68362,7,2,1,2,NA,4,4,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8009.966208,8254.622305,2,96,13,13,NA,4,4,1,1,0,2,40,1,3,77,NA
+68363,7,2,1,1,21,5,7,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6929.441295,7483.610581,2,102,15,15,5,5,5,1,0,2,1,30,1,4,1,5
+68364,7,2,1,63,NA,4,4,2,NA,NA,2,NA,2,2,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,7514.993062,7713.03043,2,90,6,6,1.12,4,4,0,1,1,1,63,2,1,1,1
+68365,7,2,1,64,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,22184.040999,22468.589781,2,94,6,6,2.66,1,1,0,0,1,1,64,1,4,3,NA
+68366,7,2,1,3,NA,4,4,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11920.911192,13144.753907,2,96,3,3,0.59,3,3,1,0,0,2,25,1,4,1,NA
+68367,7,2,1,78,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,10160.645851,10681.304475,1,96,8,8,3.14,2,2,0,0,2,2,64,1,3,1,2
+68368,7,2,1,17,NA,5,6,2,17,207,2,NA,2,1,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9136.388281,9764.423514,3,91,9,9,4.08,2,2,0,1,0,2,54,2,5,1,NA
+68369,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,39565.288792,43332.356399,1,99,77,77,NA,2,2,0,0,2,2,80,1,4,1,4
+68370,7,2,2,10,NA,3,3,2,10,129,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24070.467912,23750.822126,1,95,7,7,1.17,6,6,1,3,0,2,44,1,4,1,NA
+68371,7,2,1,10,NA,3,3,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18216.94614,18737.854061,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+68372,7,2,2,2,NA,5,6,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4962.240532,5199.576603,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+68373,7,2,2,8,NA,4,4,2,8,97,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11467.793741,13004.375485,2,91,6,6,0.78,7,7,1,4,0,2,38,2,2,77,NA
+68374,7,2,1,45,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20696.713928,21333.342141,2,102,15,15,5,3,3,1,0,0,2,34,1,5,1,5
+68375,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,29167.119125,29596.13582,1,101,1,1,0.1,4,4,1,1,0,2,52,1,4,3,NA
+68376,7,2,2,61,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,13057.178942,13648.591881,1,102,77,77,NA,2,2,0,0,2,1,68,2,4,1,1
+68377,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,NA,8386.962371,8877.414759,1,93,3,3,0.93,1,1,0,0,1,1,74,1,4,4,NA
+68378,7,2,1,0,3,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9419.940635,10084.452218,2,98,3,3,0.54,3,3,1,0,0,1,23,1,3,1,2
+68379,7,2,1,1,16,4,4,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6580.937346,6661.20735,2,97,4,4,0.97,3,3,1,0,0,1,38,1,3,6,NA
+68380,7,2,2,43,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,17623.300255,17855.084956,1,100,15,15,5,4,4,0,1,0,1,44,2,5,1,5
+68381,7,2,1,9,NA,3,3,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21087.869274,22876.159598,1,101,4,4,0.99,2,2,0,1,0,2,35,1,3,5,NA
+68382,7,2,1,0,11,3,3,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9081.731172,9425.042696,1,95,8,8,1.45,6,6,1,1,2,1,69,1,1,1,3
+68383,7,2,2,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21905.560663,22505.214105,1,95,5,5,0.76,5,5,2,1,0,2,27,1,4,6,NA
+68384,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,61710.107686,62261.838119,1,98,2,2,0.63,1,1,0,0,0,2,23,1,4,5,NA
+68385,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,22912.222762,29561.458771,2,96,5,5,1.5,2,2,1,0,0,2,22,1,3,5,NA
+68386,7,1,1,30,NA,1,1,NA,NA,NA,2,NA,2,2,77,NA,3,5,NA,2,2,2,2,2,2,NA,NA,NA,NA,53303.690379,0,1,100,4,4,0.78,4,4,0,0,1,1,33,2,1,1,1
+68387,7,2,2,70,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,1,1,2,1,1,2,2,NA,12296.397953,12433.512478,1,99,9,9,4.08,2,2,0,0,2,1,73,2,5,1,5
+68388,7,2,2,2,NA,3,3,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47602.397711,49102.065502,1,102,6,6,1.23,4,4,2,0,0,2,25,1,5,1,5
+68389,7,2,2,15,NA,4,4,1,15,191,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15483.914568,16116.18632,1,100,6,6,2.12,2,2,0,1,0,2,44,1,3,3,NA
+68390,7,1,2,25,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,2,1,3,1,2,2,1,2,2,NA,NA,NA,NA,39550.779175,0,2,96,3,3,0.54,4,4,1,1,0,1,29,1,2,1,2
+68391,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,1,2,1,2,2,NA,NA,NA,NA,60163.952904,0,1,97,12,99,NA,2,1,0,0,2,2,65,1,2,2,NA
+68392,7,1,2,61,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,14994.337564,0,1,100,10,10,4.63,2,2,0,0,1,1,58,1,5,1,4
+68393,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,2,1,0.11,4,1,0,0,0,2,19,1,4,NA,NA
+68394,7,2,2,58,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,139343.551779,138175.157138,2,98,14,14,5,1,1,0,0,0,2,58,1,5,5,NA
+68395,7,2,2,0,10,1,1,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9179.885292,9149.640857,1,101,5,5,1.23,3,3,2,0,0,2,24,1,2,5,NA
+68396,7,2,2,0,6,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6700.950086,6495.019697,2,96,3,3,0.34,7,7,3,1,0,2,49,2,1,4,NA
+68397,7,1,2,55,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,141316.739364,0,2,94,8,8,1.33,7,7,1,2,1,1,34,NA,NA,6,NA
+68398,7,2,2,6,NA,4,4,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11116.391625,11531.852198,1,92,7,7,1.83,3,3,1,1,0,2,28,1,3,5,NA
+68399,7,2,2,5,NA,5,6,1,5,64,NA,NA,2,2,1,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,6858.963321,7464.098007,3,91,14,14,3.58,4,4,1,1,0,1,39,2,5,1,5
+68400,7,2,1,8,NA,4,4,2,9,108,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7687.032802,8052.55895,2,99,9,9,2.43,4,4,0,2,0,2,49,1,3,3,NA
+68401,7,2,1,43,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,5,NA,2,2,2,2,2,2,1,2,2,2,31347.36219,35248.66815,2,90,2,2,0.46,1,1,0,0,0,1,43,2,1,5,NA
+68402,7,2,2,1,19,5,7,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3842.794137,4026.588867,1,93,15,15,5,5,5,1,0,1,1,61,2,4,1,4
+68403,7,2,2,36,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,2,2,2,2,2,2,2,NA,NA,NA,NA,35425.867861,35132.246023,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+68404,7,2,2,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,11879.290971,12409.688606,2,96,3,3,1.1,1,1,0,0,1,2,65,1,4,5,NA
+68405,7,2,1,1,20,3,3,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22170.765881,25013.89276,1,94,3,3,0.39,6,6,1,0,2,1,80,1,4,1,3
+68406,7,2,2,17,NA,5,6,2,17,215,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12054.065975,12306.62888,1,98,6,6,1.65,2,2,0,1,0,2,52,2,5,1,NA
+68407,7,2,2,80,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,2,1,2,2,1,2,NA,NA,NA,NA,16878.750567,18850.661665,2,98,3,3,0.98,2,2,0,0,1,2,80,1,1,2,NA
+68408,7,2,2,16,NA,3,3,1,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,91539.042546,96988.607103,1,101,7,7,2.31,2,2,0,1,0,2,43,1,4,3,NA
+68409,7,2,1,3,NA,1,1,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,20874.345556,20902.551716,3,92,5,5,1.05,3,3,1,1,0,2,38,2,3,5,NA
+68410,7,2,1,16,NA,4,4,2,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13863.378072,13964.345562,1,96,15,15,5,4,4,0,1,0,1,56,1,4,1,5
+68411,7,2,2,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,89807.047643,92234.66957,3,91,9,7,3.67,2,1,0,0,0,1,41,NA,NA,3,NA
+68412,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,22013.270774,24073.71767,1,101,5,5,1.05,3,3,0,0,1,2,55,1,4,1,NA
+68413,7,2,2,3,NA,1,1,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15326.318384,16920.666785,1,94,6,6,1.3,4,4,2,0,0,1,24,2,1,1,4
+68414,7,1,1,77,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19418.523783,0,1,101,4,4,0.99,2,2,0,0,2,1,77,1,3,1,3
+68415,7,2,2,43,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,2,1,2,2,1,2,2,1,2,2,1,49900.868115,50161.096898,3,92,7,7,2.78,2,2,0,0,0,1,24,1,3,5,NA
+68416,7,2,1,63,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,3,8623.181934,10090.425838,2,103,3,3,0.63,3,3,0,0,1,1,63,2,3,1,NA
+68417,7,2,1,19,NA,1,1,2,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,14081.782012,14391.713696,2,90,6,6,1.15,5,5,0,2,0,2,47,2,1,1,5
+68418,7,2,2,70,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,2,2,2,2,2,2,NA,20621.25319,22173.285594,2,93,9,9,3.14,3,3,0,0,2,1,43,NA,NA,5,NA
+68419,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,81416.374938,85457.008484,1,97,5,5,1.79,1,1,0,0,1,2,70,1,4,2,NA
+68420,7,2,1,22,NA,4,4,1,NA,NA,2,NA,2,1,4,NA,4,5,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,18831.340773,18391.458344,2,93,9,9,2.07,5,5,0,1,0,1,55,NA,NA,5,NA
+68421,7,1,2,15,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12327.773112,0,2,95,12,12,NA,2,2,0,1,0,1,47,1,4,2,NA
+68422,7,2,2,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,53370.063988,54427.336747,3,92,13,13,NA,2,2,0,0,2,2,60,1,4,1,NA
+68423,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,154825.466557,157902.106304,1,91,15,15,5,2,2,0,0,0,1,44,NA,NA,1,5
+68424,7,2,1,72,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,1,1,1,2,2,1,2,1,NA,16828.011748,23148.456034,1,97,14,14,2.29,7,7,1,2,2,1,40,2,1,1,1
+68425,7,2,1,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16385.890285,2,100,12,12,NA,2,2,0,0,1,1,56,1,5,1,4
+68426,7,2,2,19,NA,3,3,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,35205.804094,37301.699975,2,101,99,2,0.55,3,1,0,0,0,2,19,1,4,NA,NA
+68427,7,2,2,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,30275.274308,30259.77569,2,101,3,1,0.18,2,1,0,0,0,2,25,1,4,5,NA
+68428,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,2,1,2,2,1,2,2,1,2,2,1,24842.055317,24866.449113,2,95,3,3,0.63,3,3,0,0,0,2,44,1,2,4,NA
+68429,7,2,1,27,NA,3,3,2,NA,NA,2,NA,2,1,4,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,118671.226879,152848.538678,2,91,15,9,5,2,1,0,0,0,2,26,1,5,5,NA
+68430,7,2,2,2,NA,5,6,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6365.363193,6669.808387,1,92,14,14,2.42,6,6,1,3,0,1,30,1,4,6,NA
+68431,7,2,1,3,NA,4,4,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7076.684446,7163.001252,2,99,2,2,0.22,4,4,1,1,0,1,18,1,2,NA,NA
+68432,7,2,1,80,NA,2,2,2,NA,NA,1,2,2,1,9,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11755.776731,12827.229805,3,90,3,3,0.78,3,3,0,1,2,1,80,2,3,1,2
+68433,7,2,2,6,NA,1,1,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13085.05107,13640.91346,2,92,14,14,4.03,4,4,1,1,1,2,30,1,5,4,NA
+68434,7,2,2,67,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,11576.638399,12468.988052,1,101,3,3,1.12,1,1,0,0,1,2,67,1,2,5,NA
+68435,7,2,2,8,NA,3,3,1,8,101,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,25826.784831,26196.011011,2,91,2,2,0.44,3,3,0,1,0,1,46,2,3,1,4
+68436,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,203724.329642,1,91,15,12,NA,2,1,0,0,0,1,51,1,3,1,NA
+68437,7,2,1,4,NA,4,4,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7354.77583,7808.790358,2,90,8,8,1.67,6,6,1,1,0,1,52,1,3,1,5
+68438,7,2,1,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,25964.952645,26783.974907,2,96,5,5,1.36,2,2,0,0,0,2,51,1,4,1,3
+68439,7,2,1,14,NA,5,7,2,14,174,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8676.361974,9272.775003,1,90,77,77,NA,4,4,0,2,0,2,51,1,5,1,5
+68440,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,116165.700541,2,94,99,99,NA,2,2,0,0,0,2,37,1,2,1,4
+68441,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22631.175755,22215.258702,2,96,5,5,1.36,2,2,0,0,0,2,51,1,4,1,3
+68442,7,2,2,42,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,5,2,2,2,2,2,2,2,1,2,2,2,30678.628571,30838.615009,3,91,5,5,1.03,4,4,0,2,0,2,42,2,1,5,NA
+68443,7,2,1,73,NA,4,4,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,6725.306794,6857.040344,2,99,77,77,NA,1,1,0,0,1,1,73,2,5,1,NA
+68444,7,2,1,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19537.368697,21142.881814,2,96,3,3,0.47,4,4,1,0,1,2,61,1,4,3,NA
+68445,7,2,1,70,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,4,1,NA,2,2,2,2,2,2,1,2,1,NA,9710.399795,9879.67408,2,93,12,12,NA,5,5,1,0,2,1,70,2,4,1,5
+68446,7,2,2,68,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11838.431472,12295.352662,2,92,7,7,2.72,2,2,0,0,2,2,68,1,5,1,NA
+68447,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,49568.196121,49400.509437,1,95,2,2,0.78,1,1,0,0,1,2,68,1,4,3,NA
+68448,7,2,2,28,NA,5,7,1,NA,NA,2,NA,2,1,3,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18100.072824,19323.729203,1,92,7,7,3.22,1,1,0,0,0,2,28,2,5,5,NA
+68449,7,2,1,12,NA,2,2,2,12,145,NA,NA,2,1,4,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15594.049321,16140.624477,2,99,77,77,NA,4,4,1,1,1,2,38,2,4,1,4
+68450,7,2,2,47,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,21481.050033,21698.613985,2,91,8,7,3.31,2,1,0,0,0,1,55,1,2,6,NA
+68451,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16715.026676,2,100,7,7,1.38,5,5,1,0,0,2,45,1,2,3,NA
+68452,7,2,2,69,NA,2,2,2,NA,NA,2,NA,2,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,9716.805546,12994.252166,2,90,4,4,1.47,1,1,0,0,1,2,69,2,2,3,NA
+68453,7,2,2,79,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,56393.181686,58123.268911,2,93,7,7,2.31,2,2,0,0,2,2,79,1,3,1,5
+68454,7,1,1,28,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,103001.073873,0,2,97,7,7,3.21,1,1,0,0,0,1,28,1,4,5,NA
+68455,7,2,1,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39031.957066,39638.562591,1,95,3,2,0.74,2,1,0,0,0,1,27,1,3,5,NA
+68456,7,2,1,3,NA,2,2,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11179.43727,11311.602194,2,93,3,3,0.37,5,5,3,0,0,1,28,2,1,6,NA
+68457,7,2,1,0,8,3,3,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,28319.415322,27829.834184,1,92,9,9,2.6,4,4,2,0,0,2,32,1,3,1,5
+68458,7,2,1,44,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,29797.944087,29360.705192,1,103,5,5,0.74,5,5,1,1,0,2,40,99,3,1,1
+68459,7,2,2,16,NA,4,4,2,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10460.187371,12283.198438,1,99,7,7,1.53,5,5,0,3,0,1,39,1,3,1,3
+68460,7,2,1,4,NA,2,2,1,4,55,NA,NA,2,1,2,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11179.43727,11533.315753,2,93,14,14,2.91,6,6,2,0,1,2,74,NA,NA,2,NA
+68461,7,2,1,6,NA,3,3,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,63513.013195,68899.034203,2,101,7,7,1.57,4,4,0,2,0,2,28,1,3,6,NA
+68462,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,29483.108693,29407.372658,1,94,4,4,0.79,3,3,0,1,0,1,49,1,2,3,NA
+68463,7,2,2,0,7,3,3,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19676.563141,20852.949038,2,99,15,15,5,3,3,1,0,0,1,43,1,5,1,5
+68464,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,42101.975168,41959.54633,1,101,5,5,1.24,3,3,0,0,1,2,61,1,4,1,3
+68465,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,2,NA,1,2,1,1,2,1,1,2,1,NA,13689.379977,14234.742701,2,92,2,2,0.89,1,1,0,0,1,2,80,2,1,2,NA
+68466,7,2,1,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,25624.148584,27611.905955,2,96,12,12,NA,2,1,0,0,0,1,33,1,3,3,NA
+68467,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,139800.409559,144765.126463,1,100,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+68468,7,2,1,61,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,9115.676792,11889.427964,2,90,6,6,1.7,2,2,0,0,2,1,61,2,1,1,2
+68469,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,29738.952706,30201.133298,2,94,3,3,1.01,1,1,0,0,0,1,25,1,4,5,NA
+68470,7,2,1,56,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,2,1,2,2,1,2,2,3,12158.061776,12114.139928,3,90,15,15,3.7,5,5,0,0,0,1,56,2,3,1,3
+68471,7,2,2,15,NA,4,4,2,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11600.051433,12143.645666,3,90,14,14,2.97,5,5,0,2,1,1,73,2,3,2,NA
+68472,7,2,1,9,NA,3,3,2,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18324.386573,20287.008963,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+68473,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17811.496463,19015.919888,2,99,NA,77,NA,3,2,0,0,1,1,63,1,3,5,NA
+68474,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108014.727405,114941.235527,2,94,14,14,4.71,3,3,1,0,0,1,35,1,5,1,5
+68475,7,2,2,12,NA,3,3,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,36586.371708,39087.782259,1,98,3,3,0.5,5,5,0,3,0,2,56,1,3,3,NA
+68476,7,2,2,9,NA,4,4,1,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9139.784234,11102.911341,2,100,3,3,0.63,4,4,0,1,0,1,51,1,2,77,NA
+68477,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16386.190684,20296.970403,2,90,7,5,1.84,2,1,0,0,0,1,21,1,3,5,NA
+68478,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,120278.381994,2,94,10,10,2.91,4,4,0,2,0,2,38,1,4,1,4
+68479,7,2,2,3,NA,3,3,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+68480,7,2,1,6,NA,1,1,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12477.812875,12553.43469,2,94,7,7,1.17,6,6,1,2,0,2,30,2,3,6,NA
+68481,7,2,2,19,NA,3,3,2,19,239,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,100004.689924,104137.373956,1,90,15,15,5,3,3,0,0,1,1,66,NA,NA,1,4
+68482,7,2,2,77,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,15211.382198,16229.383727,2,101,3,3,0.65,3,3,0,0,1,2,77,1,3,2,NA
+68483,7,2,2,14,NA,1,1,1,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16427.640886,18021.433684,1,102,7,7,1.41,5,5,0,2,2,1,72,1,4,1,3
+68484,7,2,1,47,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,10846.102759,10858.953375,1,99,7,5,1.84,2,1,0,1,0,1,47,2,4,5,NA
+68485,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,122483.259869,124909.680309,3,91,6,6,2.3,1,1,0,0,1,2,64,1,5,3,NA
+68486,7,2,1,3,NA,4,4,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10690.995725,11500.031681,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+68487,7,2,1,19,NA,4,4,2,19,229,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,10817.360862,11219.635132,2,99,6,6,1.15,5,5,1,2,0,2,34,1,4,77,NA
+68488,7,2,2,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,16741.034883,16282.872546,2,95,4,4,1.61,1,1,0,0,0,2,58,1,5,5,NA
+68489,7,2,2,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7746.357421,8271.811673,1,99,7,7,2.52,2,2,0,1,0,2,40,1,4,3,NA
+68490,7,2,2,4,NA,5,6,1,4,59,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5601.441711,5595.525293,1,103,5,5,1.26,3,3,1,0,0,1,40,2,5,1,5
+68491,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,35161.248998,42165.766203,3,91,7,7,2.45,2,2,0,0,2,1,80,1,2,1,2
+68492,7,2,1,0,1,1,1,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8519.229538,8673.80493,1,101,15,15,4.99,4,4,2,0,0,1,31,1,4,1,4
+68493,7,2,1,29,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11770.082057,12287.307332,3,91,5,5,1.08,3,3,1,0,0,1,29,2,5,1,5
+68494,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,2,1,2,2,1,2,2,1,2,2,1,20626.479002,22617.279591,2,100,4,1,0,2,1,0,0,0,2,38,1,3,2,NA
+68495,7,2,2,78,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,27275.632708,28985.222851,2,96,2,2,0.87,1,1,0,0,1,2,78,1,3,2,NA
+68496,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,17188.361186,17721.573899,2,95,2,2,0.75,1,1,0,0,0,1,56,1,2,3,NA
+68497,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,59613.972918,59797.266126,2,100,5,5,2.2,1,1,0,0,0,2,34,1,5,3,NA
+68498,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9547.901254,10157.592212,3,90,15,15,5,5,5,1,0,1,1,38,2,3,1,4
+68499,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,175544.769665,182852.89468,1,91,15,15,5,2,2,0,0,0,2,56,1,4,1,4
+68500,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,12863.404053,13555.744544,2,90,6,6,2.75,1,1,0,0,1,2,80,1,5,2,NA
+68501,7,2,1,52,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,152865.087613,1,94,8,8,1.67,5,5,1,2,0,1,52,1,4,1,4
+68502,7,2,1,27,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,21873.767586,21362.816924,2,99,77,4,1.43,2,1,0,0,0,1,27,2,5,5,NA
+68503,7,2,2,3,NA,5,7,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,92,3,3,0.46,5,5,2,1,0,1,30,1,3,1,2
+68504,7,2,2,6,NA,3,3,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,98,14,14,4.12,4,4,0,2,0,2,36,1,5,1,3
+68505,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,151766.599459,156268.87702,3,91,7,7,1.97,4,4,0,0,1,2,77,1,5,2,NA
+68506,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,32189.450686,32533.215899,1,101,4,4,1.22,2,2,0,0,2,2,77,1,4,1,3
+68507,7,2,2,49,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,25189.111847,25556.352159,2,102,10,10,4.76,2,2,0,0,1,2,49,2,5,5,NA
+68508,7,2,2,31,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,1,2,2,1,15542.93857,15828.161907,3,91,8,8,2.7,3,3,1,0,0,1,31,1,5,1,5
+68509,7,2,2,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,14140.449064,14449.882828,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+68510,7,2,1,65,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,8886.717016,8956.231711,2,101,6,6,1.51,3,3,0,1,1,1,65,1,2,1,4
+68511,7,2,2,9,NA,5,7,1,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8128.755281,8415.30848,1,98,4,4,0.94,3,3,1,1,0,2,28,1,2,77,NA
+68512,7,2,2,57,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,20734.495277,22158.856945,2,99,77,77,NA,3,3,0,0,0,2,57,2,2,1,1
+68513,7,2,2,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,54370.517487,54186.584794,1,95,6,6,2.04,2,2,0,0,2,1,71,1,3,1,4
+68514,7,2,1,3,NA,5,7,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7629.74403,8207.121237,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+68515,7,2,1,13,NA,5,7,1,13,163,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,63907.082519,63067.196763,1,102,8,8,1.91,5,5,1,2,0,2,38,1,5,1,4
+68516,7,2,1,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,28213.419452,32164.786649,1,100,7,7,3.13,1,1,0,0,0,1,30,1,5,5,NA
+68517,7,2,1,28,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,1,1,2,2,NA,NA,NA,NA,49741.714519,52356.235161,2,91,14,14,4.71,3,3,0,0,0,1,28,2,1,1,2
+68518,7,2,2,25,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,19495.209875,21153.580858,2,90,7,7,3.31,1,1,0,0,0,2,25,1,5,5,NA
+68519,7,2,1,5,NA,3,3,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58332.578536,63879.860857,1,91,7,7,1.88,4,4,1,2,0,2,43,1,5,4,NA
+68520,7,2,2,17,NA,3,3,2,17,207,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,29885.567338,31265.78413,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+68521,7,2,2,28,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,47348.206546,48810.979214,1,92,10,10,2.93,4,4,1,0,0,2,55,1,4,1,4
+68522,7,2,1,12,NA,4,4,2,12,154,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10817.360862,11219.635132,2,99,6,6,1.15,5,5,1,2,0,2,34,1,4,77,NA
+68523,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,11879.290971,12889.384857,2,96,2,2,0.83,1,1,0,0,1,2,62,1,2,5,NA
+68524,7,2,1,11,NA,4,4,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9184.716222,9199.031651,1,102,7,7,1.57,4,4,0,2,0,2,33,1,4,1,4
+68525,7,2,2,39,NA,1,1,1,NA,NA,2,NA,2,1,5,NA,3,1,2,2,2,2,1,2,2,2,2,2,2,36453.846815,35470.245447,1,102,5,5,0.92,5,5,0,3,0,2,39,2,3,1,3
+68526,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,15799.067708,16817.831279,2,95,5,5,1.52,2,2,0,0,0,2,58,1,2,2,NA
+68527,7,2,2,8,NA,3,3,1,8,107,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,48532.852397,48387.04619,1,98,15,15,5,5,5,0,3,0,2,41,1,5,6,NA
+68528,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,90234.245265,93002.542962,2,101,6,6,2.75,1,1,0,0,1,2,70,1,4,3,NA
+68529,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,42992.537371,48017.666119,2,95,5,5,1.79,1,1,0,0,1,2,80,1,4,2,NA
+68530,7,2,1,13,NA,2,2,2,13,166,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13921.972975,14228.387356,3,90,14,14,3.93,3,3,0,1,0,2,36,2,3,1,4
+68531,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,6038.685119,6085.921613,1,96,15,15,5,3,3,0,0,1,1,60,1,4,1,4
+68532,7,2,1,3,NA,1,1,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,17341.8035,17890.748046,2,102,6,6,1,6,6,1,3,0,1,35,2,3,1,3
+68533,7,2,2,9,NA,5,6,2,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6198.268014,6830.048804,3,90,77,77,NA,5,5,0,2,0,1,46,2,3,1,3
+68534,7,2,2,8,NA,2,2,1,8,99,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,15897.166957,16345.216522,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+68535,7,2,2,0,7,1,1,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6787.112205,7016.795893,2,94,13,2,0.36,5,4,1,1,0,1,25,2,4,1,4
+68536,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8949.073879,2,95,7,7,3.22,1,1,0,0,1,2,60,1,3,3,NA
+68537,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,95087.883412,95380.247139,1,100,15,15,5,2,2,0,0,0,1,46,1,4,1,5
+68538,7,2,1,72,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,9472.03363,9657.569347,2,93,12,12,NA,4,4,0,0,2,1,72,1,2,1,4
+68539,7,1,1,72,NA,2,2,NA,NA,NA,2,NA,2,1,9,NA,2,5,NA,2,2,2,1,2,2,NA,NA,NA,NA,12962.876803,0,2,90,3,3,0.92,1,1,0,0,1,1,72,2,2,5,NA
+68540,7,2,1,9,NA,3,3,2,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14307.565788,15198.555052,3,91,5,5,1.07,4,4,0,2,0,2,36,1,5,1,4
+68541,7,2,1,12,NA,4,4,2,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,77,77,NA,7,7,0,3,1,2,43,77,5,5,NA
+68542,7,2,2,4,NA,3,3,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21341.740728,22680.288414,1,98,4,4,0.67,5,5,1,2,0,1,29,1,4,1,3
+68543,7,2,2,13,NA,1,1,1,13,160,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,21819.210646,22256.331163,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+68544,7,2,1,66,NA,5,6,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,8623.181934,9096.941425,2,103,6,6,2.28,1,1,0,0,1,1,66,1,4,5,NA
+68545,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,8219.195224,8540.664122,2,99,2,2,0.2,7,7,1,2,1,1,63,1,1,2,NA
+68546,7,2,2,6,NA,4,4,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10888.493631,13227.224662,1,101,13,13,NA,5,5,0,1,0,1,53,1,3,1,3
+68547,7,2,2,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,60914.616471,70137.240791,2,91,2,2,0.58,1,1,0,0,0,2,21,1,5,5,NA
+68548,7,2,2,66,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,124544.616504,126579.471046,1,100,6,6,2.82,1,1,0,0,1,2,66,1,5,2,NA
+68549,7,2,1,67,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,2,2,2,1,2,2,2,2,2,1,9430.93681,9952.400706,1,92,3,3,0.88,2,2,0,0,2,1,67,1,1,1,1
+68550,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,1,2,1,2,2,NA,NA,NA,NA,37019.065541,0,1,96,15,15,5,3,2,0,0,2,2,57,1,5,6,NA
+68551,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,11532.104822,12394.666968,2,99,4,4,1.61,1,1,0,0,1,2,80,1,1,2,NA
+68552,7,2,2,0,2,4,4,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4439.36229,4769.277094,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+68553,7,2,2,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,131590.908764,1,98,6,6,1.11,5,5,0,2,1,2,37,1,1,1,1
+68554,7,2,2,11,NA,1,1,1,11,132,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17053.854294,17496.484818,2,94,6,6,0.8,7,7,1,3,0,2,36,2,3,1,1
+68555,7,2,2,6,NA,3,3,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22308.590534,22137.463018,1,101,5,5,0.71,6,6,1,1,1,1,63,1,2,1,5
+68556,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,24902.414229,28566.820481,1,90,15,15,5,2,2,0,0,0,2,50,2,4,3,NA
+68557,7,2,1,71,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,15176.622228,16562.859469,1,101,3,3,0.98,1,1,0,0,1,1,71,1,3,3,NA
+68558,7,2,2,62,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,17243.546687,17998.333464,1,92,1,1,0.26,2,2,0,0,2,1,63,2,1,1,1
+68559,7,2,1,80,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,34540.440658,39912.090511,1,90,6,6,2.69,1,1,0,0,1,1,80,2,3,2,NA
+68560,7,2,2,28,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,51746.15111,53882.638481,1,95,6,6,1.37,3,3,1,1,0,2,28,1,4,5,NA
+68561,7,2,2,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,145813.864843,148196.223981,1,100,6,6,1.31,3,3,0,0,2,1,65,1,5,1,5
+68562,7,2,1,3,NA,4,4,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9126.025,2,100,6,6,1.52,4,4,1,2,0,2,39,1,4,3,NA
+68563,7,2,2,17,NA,4,4,1,17,208,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13452.957635,13490.376767,2,93,4,4,1.29,2,2,0,1,0,2,56,2,3,4,NA
+68564,7,2,2,35,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,4,1,2,1,2,1,1,2,1,NA,NA,NA,NA,20039.469886,21581.359058,1,97,15,15,5,4,4,2,0,0,2,35,2,4,1,4
+68565,7,2,2,65,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14102.354333,14646.654863,3,91,15,15,5,2,2,0,0,2,2,65,2,5,1,5
+68566,7,2,1,13,NA,4,4,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13468.614146,13600.508187,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+68567,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,24217.957803,24014.890408,2,94,4,1,0.09,2,1,0,0,0,2,51,1,2,6,NA
+68568,7,2,1,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,114410.004231,116188.079068,1,97,15,15,3.89,5,5,0,2,0,1,50,1,4,6,NA
+68569,7,2,2,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,19977.261965,20728.031273,1,92,NA,1,0.18,4,3,0,2,0,2,56,1,4,4,NA
+68570,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,23465.604153,23669.640656,2,98,3,3,1.19,1,1,0,0,0,1,49,1,3,3,NA
+68571,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10562.541357,10524.383438,1,103,5,5,0.65,6,6,0,0,1,2,26,2,4,5,NA
+68572,7,2,1,42,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,117613.757473,123317.366869,2,93,9,9,3.77,2,2,0,0,0,2,42,2,4,1,5
+68573,7,2,1,64,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,10288.337394,10853.580889,3,91,9,9,3.24,3,3,0,1,1,1,64,2,2,1,5
+68574,7,2,1,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,40776.102186,2,98,7,7,1.53,5,5,0,0,0,2,48,1,3,5,NA
+68575,7,2,2,42,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,33565.905936,33977.52311,2,93,14,14,3.52,5,5,1,2,0,1,44,1,5,1,5
+68576,7,2,2,1,20,1,1,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9439.743884,10034.856293,1,90,6,6,1.11,5,5,1,2,0,1,30,2,1,6,NA
+68577,7,2,1,19,NA,2,2,2,19,239,2,NA,1,1,NA,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22228.448129,22318.929635,1,93,6,6,1.78,3,3,0,0,0,1,19,1,3,NA,NA
+68578,7,2,1,8,NA,2,2,1,8,100,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13665.416457,13530.568366,1,94,5,5,0.74,5,5,1,1,0,2,24,1,3,1,4
+68579,7,2,1,6,NA,1,1,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11159.151566,11335.739115,1,102,6,6,1.03,6,6,0,4,0,1,34,2,2,1,1
+68580,7,2,2,47,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,46405.063078,46647.061535,2,98,10,10,4.43,2,2,0,0,1,2,47,1,4,3,NA
+68581,7,2,1,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18790.641284,18862.037788,2,100,15,15,4.97,5,5,0,2,1,2,42,1,5,1,5
+68582,7,2,2,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,19807.824556,20146.472612,1,96,9,9,3.35,3,3,0,0,0,2,42,1,4,5,NA
+68583,7,2,1,11,NA,3,3,2,11,135,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,42986.51011,44926.997612,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+68584,7,2,1,18,NA,3,3,2,18,221,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,88198.948426,91036.751291,1,101,9,9,2.6,4,4,0,1,2,2,63,1,4,1,4
+68585,7,2,2,1,17,1,1,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11901.705423,12652.027961,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+68586,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,1,1,0.05,4,1,0,0,0,2,21,1,4,5,NA
+68587,7,2,2,2,NA,4,4,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6854.479526,6961.787344,2,91,6,6,0.78,7,7,1,4,0,2,38,2,2,77,NA
+68588,7,2,2,14,NA,1,1,1,14,174,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,16360.434077,17687.989006,3,91,5,5,1.03,4,4,0,2,0,2,42,2,1,5,NA
+68589,7,2,1,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16117.991297,16472.240486,1,96,10,10,3.04,4,4,0,1,0,2,43,1,5,1,4
+68590,7,2,1,1,16,4,4,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5942.817425,6188.375425,2,97,13,13,NA,6,6,2,2,0,2,24,1,2,6,NA
+68591,7,2,1,67,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,8088.223021,7993.317874,1,99,14,14,4.86,3,3,0,1,1,2,56,1,5,1,5
+68592,7,2,1,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,24793.720953,26072.340199,1,92,14,14,3.25,4,4,0,2,0,1,34,1,4,6,NA
+68593,7,2,2,63,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14745.12656,14965.949752,1,98,10,10,5,1,1,0,0,1,2,63,2,5,5,NA
+68594,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,12291.154515,13189.875012,1,95,3,3,0.92,1,1,0,0,1,1,80,1,3,2,NA
+68595,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,36676.329165,37617.002202,1,95,4,4,1.56,1,1,0,0,0,2,50,1,2,4,NA
+68596,7,2,2,6,NA,3,3,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17958.854782,18534.102526,2,100,2,2,0.38,3,3,0,2,0,2,35,1,4,5,NA
+68597,7,2,2,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,NA,49260.413155,50771.674072,2,100,14,5,2.2,2,1,0,0,2,1,71,1,2,6,NA
+68598,7,2,1,14,NA,4,4,1,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11429.628358,11512.87076,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+68599,7,2,2,2,NA,1,1,1,2,32,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13322.479814,13501.65138,1,92,5,5,1.15,3,3,1,0,0,2,27,1,3,1,3
+68600,7,2,2,7,NA,3,3,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24070.467912,24097.958076,1,95,6,3,0.45,6,4,1,2,0,1,28,1,2,1,2
+68601,7,2,1,26,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9956.598907,10266.888978,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+68602,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,34997.800447,35379.102239,2,96,6,6,1.11,5,5,0,3,0,2,32,2,3,1,2
+68603,7,2,1,19,NA,4,4,2,19,234,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12721.673656,12814.326071,2,97,2,2,0.49,2,2,0,0,0,1,24,1,4,6,NA
+68604,7,2,2,2,NA,4,4,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5403.773938,5854.006054,3,90,10,10,2.59,5,5,1,2,0,2,32,1,4,1,4
+68605,7,2,2,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,23003.908463,24458.642976,2,99,12,1,0.12,2,1,0,0,1,2,46,1,3,2,NA
+68606,7,2,2,62,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,2,2,2,1,2,2,2,9330.330573,12477.420452,3,90,6,6,1.04,5,5,0,1,1,2,50,2,1,1,2
+68607,7,2,2,0,4,5,6,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5507.575232,5813.640954,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+68608,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,2,1,9,NA,5,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,41252.836432,0,1,100,15,15,5,3,1,0,0,3,2,69,NA,NA,5,NA
+68609,7,2,1,4,NA,1,1,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14505.510202,14676.996441,2,94,7,7,1.17,6,6,1,2,0,2,30,2,3,6,NA
+68610,7,2,1,6,NA,4,4,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8439.403196,8726.876152,2,103,6,6,1.11,5,5,1,2,0,2,36,1,4,5,NA
+68611,7,2,1,8,NA,5,6,2,8,96,NA,NA,2,2,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9928.619925,10708.884665,1,91,12,12,NA,4,4,0,2,0,1,43,2,5,1,5
+68612,7,2,2,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,24713.281483,24036.937705,1,92,10,10,4.3,2,2,0,0,0,2,55,1,4,1,5
+68613,7,2,2,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,166028.936087,168741.576297,2,101,8,8,4.41,1,1,0,0,1,2,61,1,4,2,NA
+68614,7,2,1,73,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,2,1,NA,2,2,2,1,2,2,2,2,2,NA,15301.031416,16695.608527,1,90,4,4,0.94,3,3,0,0,2,2,78,2,1,1,2
+68615,7,2,1,0,0,2,2,1,NA,0,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7530.249163,7898.897498,2,102,14,14,3.25,5,5,2,0,0,1,27,1,5,1,5
+68616,7,2,2,17,NA,4,4,2,17,206,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11956.705842,11923.015236,1,96,15,15,5,5,5,0,3,0,2,47,1,5,1,5
+68617,7,2,2,2,NA,5,6,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5022.237557,1,91,14,14,3.69,4,4,1,1,0,2,29,2,5,1,5
+68618,7,2,2,13,NA,5,6,2,13,158,NA,NA,2,2,2,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7649.811754,7945.620413,2,90,3,1,0,5,1,1,2,0,1,44,2,5,1,5
+68619,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16942.073558,18043.178849,1,99,6,6,1.12,4,4,2,0,0,2,29,1,2,1,4
+68620,7,2,1,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16555.478822,16319.493771,2,96,3,3,0.47,6,6,0,4,0,1,36,1,4,1,4
+68621,7,2,2,14,NA,4,4,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,6,6,1.16,4,4,0,3,0,2,36,1,4,4,NA
+68622,7,2,2,70,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,64463.340883,65151.773075,1,91,7,7,2.72,2,2,0,0,2,1,70,NA,NA,1,4
+68623,7,2,1,28,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11279.598253,11667.930056,1,93,15,15,5,5,5,1,0,1,1,61,2,4,1,4
+68624,7,2,1,45,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,34205.013302,34329.398365,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+68625,7,2,1,24,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,2,2,2,1,2,2,1,2,2,1,47487.549895,49034.44781,1,102,77,77,NA,3,3,0,0,1,1,61,2,4,1,1
+68626,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,54781.947924,56462.604085,2,93,10,10,3.61,3,3,0,0,2,1,75,1,4,1,4
+68627,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,74668.489197,78374.230241,2,91,77,77,NA,2,2,0,0,2,2,70,1,3,1,4
+68628,7,2,2,17,NA,4,4,2,17,206,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+68629,7,2,2,4,NA,4,4,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11378.89676,1,96,12,12,NA,7,7,1,0,1,2,59,1,3,1,1
+68630,7,2,1,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,69063.138927,75761.928286,1,92,10,10,2.1,6,6,1,1,0,2,29,1,4,1,2
+68631,7,2,1,12,NA,5,6,2,12,149,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10618.845283,10988.972369,1,93,14,14,5,2,2,0,1,0,2,46,2,5,3,NA
+68632,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,134694.414609,139412.076132,2,94,15,15,5,2,2,0,0,2,1,64,1,5,1,5
+68633,7,2,2,38,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,3,1,2,1,2,2,1,2,2,1,2,2,2,46400.322077,46544.459997,1,95,9,9,2.46,4,4,0,0,0,1,42,2,2,1,3
+68634,7,2,2,15,NA,4,4,1,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17373.928778,17416.718116,2,96,15,15,5,3,3,0,1,0,1,55,1,5,1,4
+68635,7,2,2,58,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19183.115011,19284.554221,1,92,14,14,5,2,2,0,0,0,1,40,1,5,1,4
+68636,7,2,1,1,17,5,6,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6319.053383,6506.026949,1,91,15,15,5,3,3,1,0,0,1,39,2,5,1,5
+68637,7,2,2,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,28059.790487,28146.065043,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+68638,7,2,1,18,NA,4,4,2,18,222,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11538.004005,11560.815583,1,96,9,9,3.35,3,3,0,0,0,2,42,1,4,5,NA
+68639,7,2,2,19,NA,3,3,2,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,123622.182994,134789.559864,2,94,15,15,3.82,5,5,0,0,0,1,50,1,5,1,5
+68640,7,2,2,28,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,12412.338679,12980.380959,3,90,15,15,3.7,5,5,0,0,0,1,56,2,3,1,3
+68641,7,2,2,14,NA,1,1,1,14,176,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,20460.442471,21235.303768,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+68642,7,2,1,16,NA,4,4,1,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16147.713323,16532.569027,1,92,NA,1,0.18,4,3,0,2,0,2,56,1,4,4,NA
+68643,7,2,1,17,NA,1,1,1,17,207,2,NA,1,1,NA,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,32326.52031,34735.818434,2,94,5,5,0.65,6,6,0,3,0,1,44,2,1,1,1
+68644,7,2,2,1,13,2,2,2,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9889.733317,10584.607254,1,91,5,5,0.8,5,5,2,1,0,2,31,2,3,1,1
+68645,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,48731.095364,50226.117333,2,93,6,6,2.51,1,1,0,0,1,2,75,1,5,3,NA
+68646,7,2,2,45,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,34226.159502,34404.646012,1,97,4,4,0.65,4,4,0,1,0,2,45,2,2,3,NA
+68647,7,2,1,54,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,16851.334496,17374.092058,2,95,3,3,1.07,1,1,0,0,0,1,54,1,2,5,NA
+68648,7,2,2,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85420.170155,86011.35365,1,91,14,14,3.06,5,5,2,0,0,2,30,1,5,1,5
+68649,7,2,2,0,8,1,1,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6787.112205,7238.162123,2,94,8,8,2.33,4,4,2,0,0,1,24,1,2,6,NA
+68650,7,2,1,3,NA,5,7,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6538.646646,7333.807608,2,100,6,6,1.62,3,3,1,0,0,1,32,1,3,1,3
+68651,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,87808.984693,88078.968218,1,93,15,15,5,1,1,0,0,0,2,32,1,5,5,NA
+68652,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,78529.577822,86568.047729,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+68653,7,2,1,9,NA,1,1,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11159.151566,11210.524757,1,102,4,4,0.5,6,6,2,2,0,1,25,1,2,1,3
+68654,7,2,2,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5687.793894,6161.690011,2,90,7,7,1.53,5,5,2,1,0,1,35,2,4,1,4
+68655,7,2,1,19,NA,5,6,1,19,231,2,NA,2,2,1,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9948.022697,10353.284725,2,101,5,5,1.84,2,1,0,0,0,1,19,2,4,NA,NA
+68656,7,2,1,10,NA,4,4,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8183.206265,8693.880504,3,90,3,3,0.78,3,3,0,1,2,1,80,2,3,1,2
+68657,7,2,1,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,22705.790734,22474.018796,1,95,3,3,0.76,3,3,0,0,1,1,41,1,2,1,4
+68658,7,2,2,17,NA,5,6,2,17,212,2,NA,2,1,99,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10998.124667,11476.706351,1,91,15,15,5,4,4,0,2,0,2,55,1,5,1,5
+68659,7,2,2,15,NA,3,3,1,15,186,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,85039.920663,86146.650284,1,94,14,14,2.96,5,5,0,3,0,2,39,1,4,1,3
+68660,7,2,1,24,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,1,5,NA,2,2,2,1,2,2,2,2,1,2,38474.772527,38999.213364,2,93,4,4,0.56,5,5,0,0,0,2,49,2,2,5,NA
+68661,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10991.458112,11482.21496,2,99,15,15,5,2,2,0,0,2,1,60,1,5,1,5
+68662,7,2,2,37,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,16369.916397,16518.354031,1,101,8,8,1.81,5,5,2,0,1,2,37,2,4,1,2
+68663,7,2,2,48,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,14101.070104,15807.720539,2,92,6,6,1.3,4,4,0,1,0,2,48,2,3,1,3
+68664,7,2,1,6,NA,2,2,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13953.558291,13815.866863,2,93,15,15,5,3,3,0,1,0,2,36,2,3,1,5
+68665,7,2,2,51,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,3,3,NA,2,2,2,2,2,2,2,2,1,2,23122.188977,23961.502212,2,93,4,4,0.69,4,4,0,1,1,2,66,2,3,2,NA
+68666,7,2,1,15,NA,5,6,1,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9587.464696,10013.85895,3,92,15,15,5,3,3,0,1,0,1,55,2,5,1,4
+68667,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,1,1,2,1,1,2,2,3,22345.774392,22681.166648,1,92,7,7,2.1,3,3,0,0,0,1,24,2,4,5,NA
+68668,7,2,2,8,NA,1,1,1,8,104,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15841.451259,16252.614023,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+68669,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,27140.404673,27611.034752,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+68670,7,2,2,5,NA,5,7,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8894.95474,1,97,15,15,5,4,4,1,0,0,2,39,2,5,1,5
+68671,7,2,1,6,NA,4,4,2,6,81,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7504.213986,7926.614174,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+68672,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,133495.735893,134864.21511,3,92,12,3,0.98,3,1,0,0,0,1,45,1,3,1,3
+68673,7,2,1,62,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,10004.038848,10164.290775,2,102,5,5,1.3,3,3,0,0,1,1,56,1,2,5,NA
+68674,7,2,2,0,5,5,7,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8377.359485,8321.07116,2,96,12,12,NA,4,4,1,0,0,2,20,1,5,1,5
+68675,7,2,2,73,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,18481.000773,19056.863071,1,100,10,10,4.63,2,2,0,0,2,1,78,2,5,1,5
+68676,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,19060.786733,19514.352798,2,97,4,4,0.81,4,4,1,1,0,2,51,1,3,4,NA
+68677,7,2,2,2,NA,5,7,1,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8474.492088,9470.199674,1,102,7,7,1.9,4,4,1,1,0,1,29,1,4,1,3
+68678,7,2,1,43,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,27776.016947,27368.446715,2,90,99,99,NA,5,5,1,1,0,2,40,2,3,1,1
+68679,7,2,1,43,NA,5,7,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,168806.614822,176992.791466,2,101,5,5,1.36,2,2,0,0,0,2,47,1,2,1,4
+68680,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,53634.754806,54437.113731,1,98,12,2,0.45,3,1,0,0,0,2,22,NA,NA,5,NA
+68681,7,2,1,66,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,1,2,NA,7973.883342,8285.757609,1,100,5,5,2.06,1,1,0,0,1,1,66,1,3,3,NA
+68682,7,2,2,79,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,98976.420245,105956.226489,2,101,5,5,1.63,2,2,0,0,2,1,80,1,1,1,1
+68683,7,2,2,20,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,2,2,2,1,2,2,1,2,2,2,32537.532358,33640.063825,2,90,8,8,2.01,4,4,0,0,1,2,67,2,4,2,NA
+68684,7,2,2,6,NA,3,3,2,6,72,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23521.178662,23340.749474,1,101,3,3,0.44,5,5,0,3,0,1,35,1,3,1,4
+68685,7,2,2,12,NA,4,4,2,12,154,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11905.948878,13980.928588,2,99,5,5,1.32,2,2,0,1,0,1,46,2,2,5,NA
+68686,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,18801.993237,20608.537144,2,96,7,7,2.45,2,2,0,0,0,1,24,2,5,5,NA
+68687,7,2,2,45,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,39843.937983,40051.72073,1,102,9,9,3.74,2,2,0,0,0,2,45,1,4,1,2
+68688,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,94644.050918,104631.781207,2,91,15,15,5,4,4,2,0,0,1,35,1,5,1,5
+68689,7,2,2,70,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,1,2,2,NA,18607.968221,20176.873294,2,92,5,5,1.36,2,2,0,0,2,1,75,2,1,1,1
+68690,7,2,2,29,NA,5,7,2,NA,NA,2,NA,2,1,6,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,45549.853584,51248.283427,3,91,7,7,1.57,4,4,2,0,0,2,29,2,3,1,3
+68691,7,2,1,7,NA,3,3,2,7,90,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,66868.503099,69864.859716,1,98,14,14,3.15,5,5,0,3,0,1,34,1,4,1,4
+68692,7,2,2,4,NA,5,6,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5081.860978,5222.553414,3,91,14,14,3.06,5,5,3,0,0,1,34,2,5,1,5
+68693,7,2,1,28,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,124091.929364,126020.47345,1,95,15,15,4.14,7,7,2,2,0,1,28,1,4,5,NA
+68694,7,2,1,57,NA,4,4,2,NA,NA,1,1,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16589.308426,16584.966003,2,90,15,15,5,4,4,0,0,0,1,57,2,5,1,5
+68695,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,25815.880139,26556.735732,2,101,1,1,0.23,2,1,0,0,0,1,20,1,4,5,NA
+68696,7,2,1,44,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,5,3,NA,2,2,2,2,2,2,2,2,2,2,35393.002863,36467.929589,2,93,3,3,0.58,4,4,0,1,1,1,65,2,1,3,NA
+68697,7,2,1,10,NA,1,1,2,10,130,NA,NA,2,2,3,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12477.812875,12675.267527,2,94,4,4,0.73,5,5,2,1,0,1,35,2,1,6,NA
+68698,7,2,1,8,NA,1,1,1,8,107,NA,NA,2,2,2,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13870.762641,14200.45921,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+68699,7,2,2,18,NA,4,4,1,19,228,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13801.622751,14246.373765,1,100,10,10,2.59,5,5,0,1,0,2,40,1,5,1,NA
+68700,7,2,2,4,NA,1,1,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12261.362755,13536.873522,1,103,13,13,NA,4,4,2,0,0,2,27,2,2,6,NA
+68701,7,2,2,27,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,2,6,2,2,2,2,2,2,2,2,2,1,2,32455.694722,36034.785338,1,103,13,13,NA,4,4,2,0,0,2,27,2,2,6,NA
+68702,7,2,1,56,NA,4,4,2,NA,NA,2,NA,2,2,7,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,20066.354173,20688.84721,1,96,6,6,2.24,1,1,0,0,0,1,56,2,3,3,NA
+68703,7,2,1,63,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,9235.951997,9415.956027,1,103,15,15,5,2,2,0,0,1,1,63,1,5,3,NA
+68704,7,2,1,7,NA,3,3,2,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18324.386573,20287.008963,2,95,7,7,1.13,6,6,0,3,1,1,52,1,4,1,4
+68705,7,2,1,6,NA,4,4,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12267.215138,12414.279886,2,102,5,5,0.76,5,5,1,3,0,2,30,1,4,4,NA
+68706,7,2,1,26,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,21022.682584,20897.3615,1,100,5,5,1.05,3,3,1,0,0,2,35,1,4,6,NA
+68707,7,2,1,54,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,166897.201244,167183.296853,2,91,15,15,5,4,4,0,0,0,1,54,1,5,1,NA
+68708,7,2,2,61,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,1,2,NA,2,2,2,2,2,2,1,2,2,2,6278.072933,7279.121105,2,103,5,5,0.65,6,6,1,0,1,2,61,2,1,2,NA
+68709,7,2,2,30,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,2,5,3,1,2,2,1,2,2,NA,NA,NA,NA,12801.027411,15381.268364,2,90,77,77,NA,4,3,1,0,0,2,30,2,2,5,NA
+68710,7,2,1,40,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,2,2,2,2,2,2,2,2,1,2,33029.272844,32544.619181,2,93,6,6,0.93,5,5,1,2,0,1,40,2,4,1,4
+68711,7,2,1,26,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,12084.002684,12472.760771,1,99,9,9,5,1,1,0,0,0,1,26,1,5,5,NA
+68712,7,2,1,0,2,5,7,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7882.953266,8003.945501,2,96,7,7,1.79,4,4,1,0,0,2,30,1,4,6,NA
+68713,7,2,2,22,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,45798.520132,47350.399021,2,91,3,3,0.73,3,3,0,0,0,2,22,2,2,5,NA
+68714,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,1,96,15,9,5,2,1,0,0,0,2,55,1,4,5,NA
+68715,7,2,1,54,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,145342.530024,147904.745834,1,99,15,15,5,2,2,0,0,0,1,54,1,3,1,4
+68716,7,2,2,12,NA,1,1,1,12,151,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,24481.187693,24971.637581,1,95,13,13,NA,5,5,1,2,0,2,34,2,1,1,1
+68717,7,2,1,80,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,9835.055907,10027.702409,2,99,10,10,4.76,2,2,0,0,2,1,80,1,2,2,NA
+68718,7,2,2,6,NA,1,1,2,6,79,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13231.432201,13706.598104,2,97,4,4,0.67,4,4,0,2,0,1,39,2,2,6,NA
+68719,7,2,1,2,NA,5,6,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6185.185728,6679.834548,1,97,15,15,4.84,6,6,2,0,0,1,53,NA,NA,1,NA
+68720,7,2,2,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,215228.012584,222579.783434,1,98,6,6,1.98,2,2,0,0,0,1,54,1,4,1,3
+68721,7,2,2,39,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,36053.766709,37679.125242,1,100,3,3,0.73,3,3,2,0,0,2,39,1,3,5,NA
+68722,7,2,1,40,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,31740.385214,33170.19689,2,96,7,7,1.57,4,4,0,2,0,1,40,2,2,1,5
+68723,7,2,2,9,NA,1,1,1,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10118.363218,10311.586628,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+68724,7,2,2,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,35160.335949,35576.957572,1,103,5,5,1.02,4,4,2,0,0,1,25,1,2,1,4
+68725,7,2,2,0,9,3,3,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27484.911887,26738.360643,1,92,14,14,3.9,4,4,2,0,0,2,28,1,3,1,3
+68726,7,2,2,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27769.056387,27890.133764,1,94,2,2,0.42,3,3,0,0,0,2,52,1,4,1,1
+68727,7,2,2,9,NA,5,6,2,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6198.268014,6618.822195,3,90,12,12,NA,5,5,1,2,0,1,37,2,5,1,5
+68728,7,2,2,9,NA,2,2,1,9,117,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12307.832776,12608.996083,2,93,5,5,0.89,4,4,0,2,0,1,42,NA,NA,6,NA
+68729,7,2,1,8,NA,1,1,1,8,100,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12577.115885,12776.141954,2,96,6,6,1.11,6,6,0,2,1,1,40,2,2,1,2
+68730,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,9260.072397,9783.740861,1,98,3,3,1.07,1,1,0,0,1,1,80,1,4,3,NA
+68731,7,2,2,1,16,1,1,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,15816.39252,3,92,14,14,3.25,4,4,2,0,0,2,33,1,5,1,5
+68732,7,2,2,33,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,42456.72357,41469.55672,1,92,14,14,3.15,5,5,1,2,0,1,34,1,4,1,4
+68733,7,2,1,54,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,37557.946192,44552.872015,2,102,15,15,3.92,5,5,0,0,0,1,19,1,4,NA,NA
+68734,7,2,2,1,17,3,3,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27529.278041,30383.810541,1,92,5,5,1.15,3,3,1,0,0,1,23,1,4,1,4
+68735,7,2,1,11,NA,4,4,2,11,136,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8453.214884,8599.856249,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+68736,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,28499.00321,31184.438144,2,98,3,3,0.75,2,2,0,0,0,1,22,1,2,5,NA
+68737,7,2,1,42,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,2,1,2,2,1,2,2,2,41527.444056,43502.112286,1,95,9,9,2.46,4,4,0,0,0,1,42,2,2,1,3
+68738,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,120220.968534,126279.549,1,98,6,6,1.78,2,2,0,0,0,2,48,1,4,5,NA
+68739,7,2,1,31,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,14204.262514,14001.792163,1,99,6,6,0.6,7,7,2,1,1,2,69,1,3,2,NA
+68740,7,2,1,64,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,101090.822284,102038.91726,2,102,14,14,5,1,1,0,0,1,1,64,1,4,3,NA
+68741,7,2,2,31,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,3,2,1,2,2,1,2,2,NA,NA,NA,NA,39561.667842,38928.923531,2,98,5,5,1.63,2,2,1,0,0,2,31,1,1,3,NA
+68742,7,2,2,3,NA,2,2,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15358.480588,15565.033952,1,102,5,5,1.36,2,2,1,0,0,2,21,1,3,5,NA
+68743,7,2,1,9,NA,5,7,2,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5171.245544,5675.166629,1,99,15,15,4.47,4,4,0,2,0,2,52,2,5,1,5
+68744,7,2,1,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,22235.85388,1,94,5,5,1.04,4,4,0,2,0,2,29,1,3,1,3
+68745,7,2,2,61,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,31362.042024,31983.331006,1,94,4,4,1.4,1,1,0,0,1,2,61,1,4,3,NA
+68746,7,2,2,60,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,2,1,2,1,NA,17243.546687,17909.086027,1,92,7,7,2.1,3,3,0,0,2,1,37,2,5,5,NA
+68747,7,2,2,10,NA,2,2,1,11,132,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16061.826248,2,98,6,6,0.78,7,7,1,3,1,2,63,1,2,4,NA
+68748,7,2,1,70,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,18239.933451,19545.291055,2,102,7,7,1.68,5,5,0,0,3,1,70,2,4,1,4
+68749,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,15842.721579,15352.843701,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+68750,7,2,2,16,NA,4,4,2,16,200,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12209.74498,12781.910285,2,90,3,3,0.95,2,2,0,1,0,2,49,1,3,5,NA
+68751,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,83226.861139,86401.777435,2,95,14,14,5,2,2,0,0,0,1,39,1,4,1,5
+68752,7,2,1,45,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,46745.699003,51362.782618,2,98,8,8,2.26,4,4,0,1,0,2,43,1,3,1,2
+68753,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,127315.335607,133282.596235,1,94,15,15,5,3,3,0,1,0,2,43,1,5,1,5
+68754,7,2,1,57,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,143642.364525,143888.596634,1,90,14,14,3.25,4,4,0,0,1,2,77,1,3,3,NA
+68755,7,2,2,2,NA,1,1,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,11879.078775,2,98,2,2,0.35,3,3,2,0,0,2,20,1,4,5,NA
+68756,7,2,2,4,NA,1,1,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10024.946819,10059.178608,2,93,77,77,NA,7,7,3,1,0,2,43,2,1,1,9
+68757,7,2,2,0,2,5,6,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9150.459338,9901.313011,1,97,15,15,5,4,4,2,0,0,1,40,2,5,1,5
+68758,7,2,1,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,1,2,NA,17703.366452,19320.397271,1,101,4,4,1.22,2,2,0,0,2,2,77,1,3,1,2
+68759,7,1,2,15,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15809.066118,0,2,93,3,3,0.63,3,3,0,1,0,1,53,2,2,1,4
+68760,7,2,1,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37326.925742,39362.952436,2,91,6,1,0.31,4,1,0,0,0,1,25,NA,NA,5,NA
+68761,7,2,2,4,NA,4,4,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9268.093277,10357.044767,1,99,3,3,0.54,3,3,1,0,0,2,29,1,4,1,4
+68762,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,60664.751082,63005.627956,1,102,14,14,4.05,3,3,0,2,0,2,34,1,4,3,NA
+68763,7,2,1,26,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,5,NA,1,2,2,2,2,2,1,2,2,1,35338.972549,37196.45724,2,90,4,4,0.81,3,3,0,0,0,1,39,2,3,5,NA
+68764,7,2,1,3,NA,4,4,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9572.619158,2,100,7,7,1.38,5,5,1,0,0,2,45,1,2,3,NA
+68765,7,2,1,25,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,4,6,NA,1,2,2,1,2,2,1,2,2,2,41258.226616,42207.135518,2,99,15,1,0.27,2,1,0,0,0,1,31,1,5,6,NA
+68766,7,2,1,6,NA,4,4,2,6,77,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10464.577474,10672.635694,1,96,99,99,NA,4,4,1,1,0,2,35,2,3,1,3
+68767,7,2,2,68,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,9680.216878,10112.428208,1,96,99,99,NA,2,2,0,0,1,2,68,2,4,5,NA
+68768,7,2,2,77,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,2,2,2,1,2,2,NA,17318.187297,23904.945555,2,90,2,2,0.64,1,1,0,0,1,2,77,2,1,4,NA
+68769,7,2,2,56,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,14680.520497,15280.64157,3,91,15,15,5,3,3,0,1,1,2,56,2,5,1,NA
+68770,7,2,2,39,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,6,2,2,2,2,1,2,2,1,2,1,2,29102.738194,30199.222574,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+68771,7,2,2,24,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,23845.8146,23474.128607,1,98,15,15,5,4,4,1,0,0,2,24,1,4,1,NA
+68772,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,1,2,1,2,2,NA,NA,NA,NA,11289.606124,12205.635507,1,94,8,8,1.39,7,7,2,0,1,2,52,1,5,2,NA
+68773,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,163102.567998,164787.083347,1,99,9,9,4.35,2,2,0,0,1,1,66,1,2,1,3
+68774,7,2,1,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,43104.257334,46015.156479,2,101,5,5,1.36,2,2,0,0,0,2,22,1,4,5,NA
+68775,7,2,2,18,NA,5,6,2,18,217,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6226.949288,6291.898221,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+68776,7,2,1,8,NA,3,3,1,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,42634.626688,43853.750605,1,102,14,14,4.05,3,3,0,2,0,2,34,1,4,3,NA
+68777,7,2,1,17,NA,3,3,2,17,210,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30100.326038,35344.542616,1,101,6,6,1.31,3,3,0,2,0,1,43,1,3,4,NA
+68778,7,2,1,0,8,1,1,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7284.164858,7692.652584,2,98,4,4,0.65,5,5,3,0,0,2,23,1,4,5,NA
+68779,7,2,1,16,NA,1,1,1,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27378.670648,27376.262696,2,96,9,9,4.21,2,2,0,1,0,2,50,1,4,4,NA
+68780,7,2,1,78,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,16632.531676,17822.854094,1,100,10,10,4.63,2,2,0,0,2,1,78,2,5,1,5
+68781,7,2,1,34,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13936.822202,14121.672299,1,93,15,6,2.3,6,1,0,0,0,1,34,2,5,5,NA
+68782,7,2,2,0,8,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4543.931297,4593.545672,1,96,14,14,3.34,4,4,1,1,0,2,43,1,5,77,NA
+68783,7,2,1,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,32375.321924,36431.315201,1,101,2,2,0.22,4,4,1,0,0,2,25,1,4,6,NA
+68784,7,2,2,18,NA,1,1,2,19,228,2,NA,2,2,3,12,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,16594.391299,17940.930507,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+68785,7,2,1,0,10,4,4,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6275.847063,6372.172481,2,100,1,1,0,4,4,2,0,0,2,23,1,4,5,NA
+68786,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,33707.673642,36318.620408,2,95,15,15,5,2,2,0,0,2,1,80,1,5,1,5
+68787,7,2,1,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,149474.480274,150876.348421,1,100,6,6,1.31,3,3,0,0,2,1,65,1,5,1,5
+68788,7,2,1,9,NA,1,1,1,9,113,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14258.502552,1,92,6,6,1.12,4,4,0,2,0,1,20,1,2,1,2
+68789,7,2,1,16,NA,3,3,1,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,126777.533816,125822.536881,1,100,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+68790,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,38954.135779,38857.944364,1,91,3,3,0.62,3,3,0,1,0,2,55,1,4,4,NA
+68791,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19130.246369,18752.254654,2,95,14,14,3.47,4,4,0,0,0,2,45,1,4,1,4
+68792,7,2,2,5,NA,1,1,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,16648.051651,2,98,5,5,0.59,7,7,2,1,2,2,71,1,2,1,1
+68793,7,2,1,73,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,68460.271241,72711.227332,1,102,7,7,2.86,2,2,0,0,2,1,73,1,5,1,3
+68794,7,2,1,74,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,87101.243392,92194.745773,1,101,14,14,4.96,2,2,0,0,2,1,74,1,4,1,3
+68795,7,1,2,4,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13366.393396,0,2,94,7,7,1.18,7,7,1,4,0,2,31,1,4,6,NA
+68796,7,2,1,4,NA,2,2,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14505.510202,15161.207968,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+68797,7,2,1,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,17138.242525,20070.705691,1,90,6,6,1.57,3,3,1,0,0,2,25,1,3,6,NA
+68798,7,2,1,0,6,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7757.493251,8192.524698,3,92,13,13,NA,6,6,1,2,0,1,53,1,9,1,3
+68799,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,46023.826844,48519.015788,1,91,99,99,NA,1,1,0,0,1,1,70,1,2,3,NA
+68800,7,2,2,1,22,1,1,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11512.764389,11552.07657,2,98,7,7,1.26,7,7,1,2,0,1,43,1,2,1,1
+68801,7,2,2,3,NA,4,4,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10295.166918,11223.20055,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+68802,7,2,2,40,NA,4,4,1,NA,NA,2,NA,2,1,4,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,25189.042335,24671.456999,1,100,9,9,2.22,5,5,1,2,0,2,40,2,4,1,4
+68803,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18070.666316,17788.997869,1,99,14,14,4.05,3,3,0,1,0,2,52,1,4,4,NA
+68804,7,1,1,31,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19788.748292,0,1,94,1,1,0.08,7,7,2,4,0,1,31,1,2,1,4
+68805,7,2,1,0,4,3,3,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17113.022485,16817.175522,1,91,15,15,5,3,3,1,0,0,1,36,1,5,1,5
+68806,7,2,2,3,NA,5,6,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8171.700571,8311.029296,1,92,7,7,1.65,4,4,2,0,0,1,24,1,4,1,3
+68807,7,2,2,35,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,15388.376971,15419.798755,2,99,15,15,5,2,2,0,0,0,2,35,1,5,1,5
+68808,7,2,1,58,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,18032.281818,17974.767241,2,99,1,1,0.4,1,1,0,0,0,1,58,1,4,3,NA
+68809,7,2,2,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,7906.792868,2,95,6,6,0.97,6,6,2,1,0,1,54,1,3,6,NA
+68810,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,12291.154515,13189.875012,1,95,2,2,0.66,1,1,0,0,1,1,80,1,3,5,NA
+68811,7,2,1,0,7,1,1,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6999.189812,6930.018579,3,92,12,12,NA,7,7,2,1,0,2,30,1,2,1,9
+68812,7,2,1,42,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,19115.012147,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+68813,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,52701.331723,59294.433957,3,91,3,3,0.73,3,3,1,0,0,2,24,1,2,1,3
+68814,7,2,1,32,NA,4,4,2,NA,NA,2,NA,2,1,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20133.630718,20200.188991,1,93,6,6,1.35,3,3,0,1,0,1,32,2,4,1,4
+68815,7,2,2,10,NA,5,6,2,10,123,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6412.057856,6879.064185,3,91,8,8,2.81,3,3,0,2,0,2,31,1,4,3,NA
+68816,7,2,2,58,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,21097.069797,24201.517172,2,90,2,2,0.83,1,1,0,0,0,2,58,1,3,3,NA
+68817,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18353.275855,20231.959782,1,91,12,8,2.15,6,4,1,1,0,2,29,1,4,6,NA
+68818,7,2,1,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7410.50521,7700.344649,2,96,6,6,2.75,1,1,0,0,1,1,62,1,3,3,NA
+68819,7,2,1,47,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,52941.648658,53777.896935,2,102,77,77,NA,4,4,0,1,0,1,47,1,2,1,3
+68820,7,2,2,2,NA,2,2,1,2,31,NA,NA,2,1,2,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10164.721409,11222.125211,2,93,5,5,0.87,4,4,1,1,0,1,41,2,5,1,3
+68821,7,2,1,2,NA,4,4,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9438.902193,9727.203655,2,102,4,4,0.72,4,4,2,0,0,1,48,1,3,1,3
+68822,7,2,1,3,NA,2,2,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15745.774489,16244.197679,2,91,4,4,0.67,5,4,2,0,2,2,66,2,1,1,NA
+68823,7,2,1,64,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7897.700656,8206.595267,2,95,7,7,2.31,2,2,0,0,2,2,63,1,3,1,3
+68824,7,2,2,80,NA,2,2,2,NA,NA,1,2,2,1,9,NA,2,1,NA,2,1,2,1,2,2,1,1,2,NA,18824.116627,20613.042179,2,90,6,6,2.24,2,2,0,0,2,1,75,2,4,1,2
+68825,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,1,2,1,1,2,1,1,2,2,3,16590.074977,17628.593761,1,92,2,2,0.33,5,5,0,1,0,1,51,2,1,4,NA
+68826,7,2,1,7,NA,4,4,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10409.90361,10390.331705,1,96,3,2,0.16,7,6,1,4,0,2,32,1,2,5,NA
+68827,7,2,1,43,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17046.256569,17889.989739,1,96,9,9,4.1,2,2,0,0,0,2,45,2,5,1,5
+68828,7,2,1,24,NA,5,6,2,NA,NA,2,NA,2,1,99,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,10141.381563,10448.082757,3,90,15,15,5,3,3,0,0,0,1,46,2,3,1,3
+68829,7,2,2,0,6,2,2,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7446.889459,7422.354625,2,96,6,6,1.12,4,4,2,0,0,1,27,2,2,6,NA
+68830,7,2,2,39,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,13027.332011,13103.987136,2,103,77,77,NA,5,5,0,2,0,2,39,2,5,1,5
+68831,7,2,2,79,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,16612.938931,17191.011953,2,99,14,14,4.96,2,2,0,0,2,2,79,1,5,1,NA
+68832,7,2,1,15,NA,5,6,2,15,189,NA,NA,2,2,2,8,NA,NA,NA,1,1,1,1,2,1,1,2,1,NA,6666.045669,7091.184391,3,90,6,6,1.31,3,3,0,1,0,1,49,2,3,1,4
+68833,7,2,1,41,NA,2,2,2,NA,NA,2,NA,77,NA,NA,NA,3,5,NA,2,2,2,2,2,2,2,2,1,2,25035.846455,25142.418345,2,99,99,4,1.61,5,1,0,1,0,1,40,2,1,6,NA
+68834,7,2,1,7,NA,4,4,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14832.155253,15089.454697,1,97,15,15,5,4,4,1,1,0,1,35,1,5,1,5
+68835,7,1,1,49,NA,2,2,NA,NA,NA,2,NA,2,1,77,NA,1,6,NA,2,2,2,1,2,2,NA,NA,NA,NA,39096.402803,0,2,91,1,1,0.17,4,4,0,1,0,1,49,2,1,6,NA
+68836,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,4299.99963,4413.31452,3,90,15,15,4.2,6,6,1,0,2,1,60,1,5,1,4
+68837,7,2,2,35,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,22326.231285,22166.696692,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+68838,7,2,2,7,NA,1,1,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12789.411811,13121.359297,1,102,2,2,0.52,3,3,0,2,0,2,36,2,3,4,NA
+68839,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,44041.763669,47615.276255,1,92,10,10,3.67,3,3,0,0,2,1,52,1,4,77,NA
+68840,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19117.284298,19018.955883,2,95,7,7,2.58,2,2,0,0,0,2,57,1,4,3,NA
+68841,7,2,1,49,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21158.364877,21090.879552,1,96,7,7,2.75,2,2,0,0,0,1,49,1,4,1,5
+68842,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,29670.405171,29530.485322,2,95,4,4,0.76,4,4,0,1,2,1,80,1,1,2,NA
+68843,7,2,1,52,NA,1,1,2,NA,NA,2,NA,2,2,2,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,22446.308035,24663.335125,2,94,4,2,0.54,4,2,0,0,0,1,46,NA,NA,1,NA
+68844,7,2,1,8,NA,3,3,1,8,100,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,85148.03778,90558.20429,1,98,7,7,1.48,5,5,1,1,0,1,46,1,3,1,3
+68845,7,2,2,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,36067.495928,37024.300659,1,95,3,3,1.1,1,1,0,0,1,2,75,1,3,2,NA
+68846,7,2,2,49,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,23742.825358,24088.979786,1,100,15,15,5,2,2,0,0,0,1,48,2,5,1,5
+68847,7,2,1,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,2,2,2,1,2,2,1,59682.963348,63262.110969,2,102,9,9,3.24,3,3,0,0,0,1,54,2,4,1,4
+68848,7,2,2,4,NA,3,3,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20589.729566,21881.111334,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+68849,7,2,1,5,NA,4,4,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,10346.695479,1,100,3,3,0.52,3,3,1,1,0,2,25,1,3,5,NA
+68850,7,2,1,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,52698.05363,59566.360508,3,92,7,7,1.65,4,4,1,1,0,1,27,1,3,1,3
+68851,7,2,2,13,NA,1,1,1,13,157,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18515.058419,19360.671834,2,96,6,6,0.87,6,6,1,3,0,1,46,2,1,1,1
+68852,7,2,1,7,NA,5,6,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9720.482616,10533.307174,2,91,7,7,1.56,4,4,1,1,0,2,37,2,5,1,5
+68853,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,27428.65319,29015.297283,2,91,2,2,0.26,4,4,0,1,0,1,20,1,3,5,NA
+68854,7,2,1,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,NA,10037.214229,10115.728475,2,92,6,6,2.31,2,2,0,0,2,2,65,1,4,5,NA
+68855,7,2,1,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,71070.181743,73766.510327,1,95,6,6,1.94,2,2,0,0,2,2,69,1,2,1,4
+68856,7,2,2,14,NA,5,7,1,14,175,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15711.746539,15750.442176,1,98,14,14,3.16,6,6,2,2,0,1,39,1,5,1,5
+68857,7,2,2,12,NA,3,3,2,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,91797.787708,92992.465808,1,95,14,14,3.8,4,4,0,2,0,2,37,1,5,1,5
+68858,7,2,2,12,NA,2,2,2,12,148,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,8,8,2.24,4,4,1,1,0,2,29,1,4,6,NA
+68859,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,85444.349063,86035.699897,1,94,14,14,2.96,5,5,0,3,0,2,39,1,4,1,3
+68860,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23248.471636,1,95,7,7,1.66,5,5,0,3,0,1,34,1,2,1,4
+68861,7,2,1,8,NA,1,1,1,8,100,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11159.151566,11057.808004,1,102,2,2,0.31,4,4,1,2,0,2,25,1,2,4,NA
+68862,7,2,1,3,NA,5,7,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8269.653573,9275.321266,1,101,4,4,0.78,4,4,1,2,0,2,31,1,4,3,NA
+68863,7,2,1,4,NA,5,6,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6319.053383,6623.952114,1,91,6,6,1.25,4,4,1,1,0,1,26,2,4,6,NA
+68864,7,2,2,57,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,1,2,2,2,2,1,2,29695.385784,37858.742479,2,91,3,3,0.66,2,2,0,0,1,1,69,2,5,1,1
+68865,7,2,1,62,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,2,1,2,1,1,2,2,1,8460.109732,8526.287371,1,93,5,5,0.64,7,7,0,2,1,1,21,2,4,5,NA
+68866,7,2,1,22,NA,2,2,2,NA,NA,2,NA,2,1,2,NA,4,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,31312.870743,31739.689334,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+68867,7,2,1,12,NA,3,3,1,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,79897.505873,82468.209663,2,100,6,6,1.7,3,3,0,1,0,2,33,1,4,6,NA
+68868,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,47157.788417,52669.767304,2,95,10,10,4.3,2,2,0,0,2,1,80,1,5,1,4
+68869,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,156728.884529,156630.502182,1,100,15,15,5,3,3,0,0,0,1,53,1,5,1,4
+68870,7,2,2,0,10,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4581.358327,4631.381361,2,97,5,5,1.3,3,3,1,0,0,2,46,1,3,1,3
+68871,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,83239.679734,92023.913528,1,90,15,15,5,4,4,0,2,0,1,37,1,5,1,5
+68872,7,2,1,34,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,39040.678458,40784.183756,2,98,2,2,0.35,3,3,0,1,0,2,43,1,3,1,3
+68873,7,2,2,24,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16131.725974,16820.579069,1,99,2,2,0.66,2,1,0,0,0,2,24,1,5,5,NA
+68874,7,2,1,55,NA,2,2,2,NA,NA,2,NA,2,2,8,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,32634.009061,33534.701149,2,91,10,10,4.63,2,2,0,0,0,1,55,2,3,3,NA
+68875,7,2,1,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,174520.785302,181786.280703,1,95,7,7,2.86,2,2,0,0,0,2,50,1,3,1,3
+68876,7,2,2,78,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,81416.374938,89993.518978,1,97,15,15,5,2,2,0,0,2,1,78,1,3,1,3
+68877,7,2,1,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19387.053587,1,99,14,14,3.67,4,4,1,0,0,2,49,1,3,1,3
+68878,7,2,1,2,NA,2,2,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10803.555682,10625.559962,2,94,1,1,0.16,2,2,1,0,0,2,25,2,3,1,NA
+68879,7,2,1,2,NA,3,3,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21718.301271,24669.801903,3,91,6,6,1,6,6,1,1,0,2,39,1,1,3,NA
+68880,7,2,1,15,NA,5,7,1,15,184,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,59545.101745,58762.542429,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+68881,7,2,2,35,NA,3,3,2,NA,NA,2,NA,2,1,1,NA,2,1,2,1,2,1,1,2,1,NA,NA,NA,NA,19498.713386,20960.418705,2,97,5,5,0.8,5,5,1,2,0,1,46,2,4,1,2
+68882,7,2,2,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16313.554006,16926.636803,1,103,8,8,1.95,4,4,0,1,0,2,48,1,5,1,5
+68883,7,1,1,53,NA,2,2,NA,NA,NA,2,NA,2,2,7,NA,2,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,23431.677775,0,2,93,3,3,0.63,3,3,0,1,0,1,53,2,2,1,4
+68884,7,2,2,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,24905.670199,26841.707501,2,101,3,3,0.65,3,3,0,1,0,2,54,1,3,5,NA
+68885,7,2,2,11,NA,1,1,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12481.057866,12612.274737,1,102,4,4,0.61,5,5,2,2,0,2,27,2,2,5,NA
+68886,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,83441.642088,87352.545854,1,99,15,15,5,4,4,0,2,0,2,44,1,5,1,5
+68887,7,2,1,1,19,1,1,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,3,92,3,3,0.54,4,4,3,0,0,2,22,1,3,5,NA
+68888,7,2,1,74,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,NA,9257.537917,9255.717695,1,99,9,9,5,2,1,0,0,2,2,61,1,4,6,NA
+68889,7,2,2,3,NA,4,4,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8674.686381,9059.886471,2,99,77,77,NA,4,4,1,1,0,2,47,1,2,77,NA
+68890,7,2,2,8,NA,1,1,1,8,101,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,22618.378147,23255.85994,2,102,5,5,0.89,4,4,1,1,0,2,28,2,2,1,2
+68891,7,2,2,3,NA,1,1,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15308.9293,16274.054403,2,98,4,4,0.48,6,6,2,0,2,2,65,2,1,2,NA
+68892,7,2,1,75,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,8431.365248,9201.488687,1,96,10,10,3.04,4,4,0,0,2,1,56,1,3,3,NA
+68893,7,2,1,15,NA,4,4,2,15,188,NA,NA,1,1,NA,11,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11023.237662,11433.168046,1,99,15,15,5,4,4,0,2,0,2,46,1,5,1,5
+68894,7,2,1,36,NA,5,7,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,65033.706797,66896.305599,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+68895,7,2,2,74,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,12813.709202,13258.116643,2,103,14,14,3.47,4,4,1,0,1,1,47,2,5,1,5
+68896,7,2,1,4,NA,4,4,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8402.233801,8749.415211,2,97,3,3,0.4,6,6,2,3,0,2,25,1,2,5,NA
+68897,7,2,2,1,15,5,6,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6184.163292,6116.466669,1,91,15,15,3.25,7,7,1,2,0,2,31,1,5,1,5
+68898,7,2,1,3,NA,2,2,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13891.873932,13527.718832,2,97,12,6,0.89,7,7,3,0,0,2,26,2,1,6,NA
+68899,7,2,1,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18790.641284,22550.069226,2,100,14,14,5,7,1,1,3,1,2,62,1,3,5,NA
+68900,7,2,1,7,NA,3,3,2,7,93,NA,NA,2,2,2,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,49047.972936,54301.226542,1,93,10,10,2.48,5,5,2,1,0,1,40,2,5,1,5
+68901,7,2,2,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,112011.120236,112121.110185,2,92,15,15,5,2,1,0,0,0,1,52,1,5,6,NA
+68902,7,2,2,78,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,12493.113557,12927.830854,2,99,77,77,NA,1,1,0,0,1,2,78,1,4,2,NA
+68903,7,1,2,75,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,89167.746947,0,1,95,5,5,1.43,2,2,0,0,2,1,80,1,3,1,4
+68904,7,2,1,6,NA,4,4,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7758.863533,8127.805304,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+68905,7,2,2,1,21,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7618.827213,7815.48207,2,97,6,6,1.7,2,2,1,0,0,2,20,1,4,5,NA
+68906,7,2,2,16,NA,1,1,2,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,12680.621719,13347.583861,2,90,6,6,1.15,5,5,0,2,0,2,47,2,1,1,5
+68907,7,2,2,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,1,2,2,1,119412.590109,122441.098321,1,92,6,6,1.62,3,3,1,0,0,2,26,1,5,1,5
+68908,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,16727.646581,2,100,9,9,2.46,4,4,1,1,1,2,59,1,3,1,3
+68909,7,2,1,16,NA,4,4,2,16,202,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13513.882801,1,96,15,15,5,4,4,1,1,0,1,50,1,3,1,4
+68910,7,2,1,74,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,2,1,NA,2,2,2,1,2,2,1,2,2,NA,17419.161186,17783.358691,1,102,7,7,1.7,4,4,0,0,2,1,44,1,4,4,NA
+68911,7,2,1,13,NA,4,4,1,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,95,4,4,0.76,4,4,0,1,2,1,80,1,1,2,NA
+68912,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,2,NA,1,1,2,1,2,2,1,2,1,NA,9509.326077,9839.130289,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+68913,7,2,2,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,17260.508485,17587.688781,2,101,1,1,0.1,6,6,1,2,1,2,27,1,2,1,2
+68914,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,38093.351998,47908.278929,2,91,4,4,1.19,2,2,0,0,0,1,59,1,1,1,3
+68915,7,2,2,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,80418.665565,83274.564983,2,100,8,8,2.91,3,3,0,2,0,2,48,1,5,1,NA
+68916,7,2,1,56,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,37426.314738,38162.090079,3,92,5,5,1.39,2,2,0,0,0,1,56,2,1,1,1
+68917,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,65331.560138,67202.689613,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+68918,7,2,1,55,NA,1,1,2,NA,NA,2,NA,2,2,2,NA,1,5,NA,2,2,2,2,2,2,2,2,2,2,22446.308035,22401.210337,2,94,77,77,NA,4,4,0,0,0,1,28,2,1,3,NA
+68919,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,71525.773464,1,98,7,7,1.66,5,5,2,1,0,2,37,1,5,1,3
+68920,7,2,2,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17805.920521,18110.342663,2,93,12,12,NA,4,4,0,0,2,1,72,1,2,1,4
+68921,7,2,2,8,NA,3,3,1,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,45129.675368,44994.093254,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+68922,7,2,2,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,88916.414931,91644.282892,1,101,14,14,4.96,2,2,0,0,2,1,74,1,4,1,3
+68923,7,2,1,13,NA,5,6,2,13,157,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11506.937395,12032.024805,1,97,7,7,1.48,5,5,0,1,0,2,46,2,4,1,NA
+68924,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,4,1,0.18,4,1,0,0,0,1,22,1,5,5,NA
+68925,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,8286.407536,8892.303775,1,91,3,3,0.98,1,1,0,0,1,1,80,1,3,2,NA
+68926,7,2,2,3,NA,5,6,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6537.38756,6940.023785,1,92,12,12,NA,7,7,1,2,1,2,45,2,3,1,3
+68927,7,2,1,11,NA,4,4,2,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10669.652248,10854.743056,2,97,5,5,0.76,5,5,1,2,0,1,32,1,4,6,NA
+68928,7,2,1,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,19260.892847,19291.900704,2,97,13,13,NA,1,1,0,0,0,1,41,1,2,5,NA
+68929,7,2,2,19,NA,2,2,2,19,235,2,NA,2,7,77,13,NA,NA,NA,2,2,2,2,2,2,2,2,1,2,23968.380373,25913.276293,2,91,4,4,0.43,7,7,0,1,1,1,41,2,1,4,NA
+68930,7,2,2,6,NA,4,4,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8031.102104,8647.40073,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+68931,7,2,1,22,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14424.961621,15273.975784,2,102,8,8,2.01,4,4,0,0,0,1,59,2,4,1,4
+68932,7,2,1,4,NA,4,4,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,11070.778245,1,100,8,8,2.36,3,3,1,0,0,2,37,1,3,4,NA
+68933,7,2,1,72,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,8517.336599,8684.171961,2,100,14,14,5,2,2,0,0,2,2,63,1,5,1,4
+68934,7,2,2,47,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,4,NA,2,2,2,2,2,2,1,2,1,2,35815.777398,36002.553631,1,94,2,2,0.27,5,5,0,4,0,2,47,2,1,4,NA
+68935,7,2,1,46,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,31233.526521,31366.480522,1,100,8,4,1.43,6,1,1,0,0,1,33,2,3,6,NA
+68936,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,26465.930618,28724.649216,2,100,4,4,0.86,3,3,0,2,0,2,36,1,3,5,NA
+68937,7,2,1,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19185.622388,22768.505116,2,95,15,10,3.67,5,3,0,0,0,1,47,1,5,1,3
+68938,7,2,1,9,NA,1,1,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,NA,11316.846999,12445.832216,3,91,4,4,1.02,2,2,0,1,0,2,24,2,1,5,NA
+68939,7,2,1,7,NA,1,1,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14820.807433,14905.891142,2,102,8,8,1.91,5,5,1,2,0,1,36,2,1,1,4
+68940,7,2,2,2,NA,2,2,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11981.824297,12363.063282,2,91,2,2,0.22,4,4,1,1,0,2,48,2,9,5,NA
+68941,7,2,1,0,10,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6780.545196,6884.61702,1,96,14,14,3.47,4,4,1,1,0,2,27,1,3,6,NA
+68942,7,2,2,14,NA,4,4,1,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12531.903464,12935.738352,2,100,5,5,1.08,3,3,0,1,0,2,50,1,4,3,NA
+68943,7,2,2,0,2,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4099.350341,4144.110416,2,99,7,7,1.74,4,4,2,0,0,2,56,1,2,5,NA
+68944,7,2,2,79,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,20341.088581,21872.034752,1,93,2,2,0.64,1,1,0,0,1,2,79,1,4,3,NA
+68945,7,2,2,27,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,17568.277926,19062.73336,1,90,2,2,0.56,2,2,1,0,0,2,27,1,3,5,NA
+68946,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,82743.954233,85642.054874,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+68947,7,2,2,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,10033.449917,10806.84762,2,95,13,13,NA,2,2,0,0,2,2,80,1,1,1,NA
+68948,7,2,2,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,86578.861495,89037.775585,2,101,7,7,1.88,4,4,0,2,0,2,36,1,4,1,5
+68949,7,2,1,63,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,9145.939054,9651.644596,1,101,5,5,0.87,4,4,0,0,2,1,63,2,1,1,NA
+68950,7,2,2,38,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,39561.667842,38494.210933,2,98,7,7,1.33,6,6,0,3,0,1,31,1,3,6,NA
+68951,7,2,1,4,NA,1,1,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16775.083123,16973.400581,3,92,4,4,0.6,6,6,2,2,0,2,24,1,3,6,NA
+68952,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,56397.521122,64147.609457,2,91,7,7,2.51,2,2,0,0,1,2,80,1,5,2,NA
+68953,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,24919.497762,27697.191423,1,101,4,4,0.78,4,4,1,2,0,2,32,1,3,3,NA
+68954,7,2,2,1,22,4,4,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10927.526143,11412.763748,2,102,2,2,0.36,4,4,1,2,0,2,36,1,3,5,NA
+68955,7,2,1,59,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32680.7991,1,95,15,15,3.62,7,7,2,4,0,1,59,1,5,1,2
+68956,7,2,2,14,NA,3,3,1,14,177,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28004.718304,30534.517049,1,94,7,7,1.21,6,6,2,2,0,1,31,1,2,6,NA
+68957,7,2,1,12,NA,4,4,2,13,156,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,6,6,1.85,2,2,0,1,0,2,48,1,4,5,NA
+68958,7,2,1,12,NA,1,1,1,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30061.88611,32302.400867,1,92,10,10,3.04,4,4,1,1,0,1,32,1,3,1,2
+68959,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,17777.254625,18901.462969,2,99,NA,77,NA,3,1,0,0,1,1,63,1,3,5,NA
+68960,7,2,2,10,NA,2,2,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,15897.166957,16345.216522,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+68961,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,117419.769432,119338.215671,2,95,15,15,5,2,2,0,0,2,1,70,1,5,1,5
+68962,7,2,1,14,NA,5,6,2,14,177,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11506.937395,12032.024805,1,97,15,15,5,3,3,0,1,0,2,45,2,5,1,5
+68963,7,2,2,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6306.491784,7047.470907,2,90,6,6,1.34,4,4,1,0,0,1,38,2,4,6,NA
+68964,7,2,1,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19130.866571,19325.235628,2,99,6,6,2.95,1,1,0,0,0,1,20,1,3,5,NA
+68965,7,2,2,2,NA,3,3,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20820.221848,22126.060024,1,101,5,5,0.89,5,5,1,0,0,1,25,1,2,77,NA
+68966,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,11982.125462,2,97,1,1,0,2,2,0,0,1,2,63,1,4,5,NA
+68967,7,2,1,20,NA,4,4,1,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26503.609729,2,101,3,1,0,2,1,0,0,0,1,20,2,4,5,NA
+68968,7,2,1,12,NA,1,1,1,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22768.423624,22944.003607,3,92,4,4,0.81,3,3,0,2,0,2,31,1,3,1,NA
+68969,7,2,1,0,1,1,1,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7040.676479,6971.095244,1,98,9,1,0.04,7,1,3,2,0,2,32,1,4,1,4
+68970,7,2,1,11,NA,1,1,2,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13285.093011,13153.997889,2,94,7,7,1.88,4,4,0,2,0,2,28,1,4,4,NA
+68971,7,2,2,21,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,42583.505439,50840.190326,2,91,4,4,0.69,4,4,2,0,0,2,21,1,3,6,NA
+68972,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,11771.283932,12488.02543,1,90,9,9,2.07,5,5,0,0,1,1,46,2,4,1,3
+68973,7,2,1,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6837.992772,6921.398172,1,96,15,15,4.34,4,4,1,1,0,1,39,2,5,1,5
+68974,7,2,1,10,NA,4,4,1,10,123,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10311.165779,10434.780551,2,98,8,8,2.43,3,3,0,2,0,2,31,1,4,1,NA
+68975,7,2,2,3,NA,3,3,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,85197.465687,87881.529962,1,92,15,15,5,4,4,2,0,0,2,46,1,5,1,5
+68976,7,2,1,44,NA,5,7,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12254.763576,12210.492386,2,90,3,3,0.65,5,3,1,2,0,1,44,2,5,1,5
+68977,7,2,1,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,1,2,2,NA,13251.84987,13511.423639,2,95,4,4,0.76,4,4,0,1,2,1,80,1,1,2,NA
+68978,7,2,1,39,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,113559.363135,120841.424294,2,102,15,15,5,4,4,0,2,0,1,39,1,4,1,5
+68979,7,2,2,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,74518.988514,1,98,14,14,4.12,4,4,0,2,0,2,36,1,5,1,3
+68980,7,2,1,20,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,2,2,2,1,2,2,1,35669.2076,36620.108921,2,94,77,77,NA,4,4,0,0,0,1,28,2,1,3,NA
+68981,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,40870.323556,45647.400007,2,99,15,15,5,1,1,0,0,1,2,80,1,5,3,NA
+68982,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,3,1,1,1,2,2,1,2,2,1,2,2,1,17858.942687,18621.550977,1,97,15,15,4.84,6,6,2,0,0,1,53,NA,NA,1,NA
+68983,7,2,1,67,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11568.876339,11794.347884,1,102,6,6,2.75,1,1,0,0,1,1,67,1,5,3,NA
+68984,7,2,2,0,9,3,3,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21671.775435,21083.121886,1,101,9,9,2.39,4,4,1,0,0,2,57,1,2,77,NA
+68985,7,2,2,16,NA,4,4,1,16,201,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13895.342981,13964.270792,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+68986,7,2,2,18,NA,4,4,2,19,228,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,21329.621847,2,101,12,12,NA,4,4,0,0,0,1,57,1,3,1,3
+68987,7,2,1,9,NA,2,2,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15882.795076,15889.831109,1,97,3,3,0.44,5,5,2,2,0,2,26,1,4,4,NA
+68988,7,2,1,72,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,NA,8497.912951,8999.288803,2,100,2,2,0.73,1,1,0,0,1,1,72,1,1,4,NA
+68989,7,2,1,34,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,13601.994691,14463.629992,2,100,9,9,2.46,4,4,0,2,0,2,36,2,4,1,3
+68990,7,2,1,53,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22446.308035,22116.943066,2,91,10,10,2.95,4,4,0,1,0,2,18,1,3,NA,NA
+68991,7,2,1,40,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,34152.149953,40351.498105,1,90,5,5,1,4,4,0,2,0,1,40,2,2,1,1
+68992,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,17533.239798,17053.396696,2,99,3,3,0.44,5,5,1,1,0,2,53,1,4,1,3
+68993,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,19005.010125,19043.816772,1,92,2,2,0.24,5,5,0,2,0,1,35,2,4,1,3
+68994,7,2,1,19,NA,3,3,1,19,233,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,74191.50646,74370.660378,1,98,8,8,2.62,3,3,0,0,0,1,50,NA,NA,3,NA
+68995,7,2,1,75,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,60942.568495,64726.722108,1,94,14,14,5,2,2,0,0,2,2,72,1,5,1,5
+68996,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,NA,1,2,2,1,2,2,1,2,2,NA,14385.653726,15564.966804,2,101,7,5,1.84,2,1,0,0,0,1,23,2,4,5,NA
+68997,7,2,1,7,NA,2,2,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13599.766245,14493.354345,2,93,3,3,0.48,4,4,1,1,0,1,49,2,3,1,4
+68998,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,15206.604563,15895.992431,2,90,15,15,5,4,4,1,1,0,1,53,2,5,1,5
+68999,7,2,1,2,NA,4,4,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8022.100831,8353.575096,2,102,4,4,0.53,6,6,2,2,0,2,27,1,2,1,2
+69000,7,2,1,13,NA,3,3,2,13,163,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,26824.630008,26472.092796,1,95,1,1,0.12,3,3,0,2,0,2,40,1,5,3,NA
+69001,7,2,1,68,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,6038.685119,6274.869975,1,96,6,6,1.57,3,3,0,0,1,1,42,1,3,NA,NA
+69002,7,2,1,8,NA,1,1,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14007.413517,2,98,15,15,4.97,5,5,0,3,0,1,39,1,5,1,5
+69003,7,2,1,7,NA,2,2,2,7,88,NA,NA,2,2,3,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9390.522479,10327.334743,2,90,2,2,0.32,4,4,1,2,0,2,34,2,1,77,NA
+69004,7,2,1,16,NA,1,1,2,16,200,NA,NA,1,1,NA,9,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,18120.499457,19917.650851,1,90,1,1,0.02,5,5,0,1,0,2,39,2,1,1,2
+69005,7,2,2,77,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60723.387753,62586.321432,1,96,14,14,5,2,2,0,0,2,1,74,1,5,1,4
+69006,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,95214.22557,95506.977757,1,97,15,15,3.89,5,5,0,2,0,1,50,1,4,6,NA
+69007,7,2,1,55,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,2,4,NA,2,2,2,NA,NA,NA,2,2,1,2,24211.824535,26078.926124,2,93,3,3,0.43,4,4,0,0,0,1,45,2,2,6,NA
+69008,7,1,2,65,NA,2,2,NA,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9325.158469,0,1,93,15,15,5,2,2,0,0,2,2,65,2,4,1,5
+69009,7,2,1,56,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18416.819037,18519.889723,2,96,8,8,3.67,2,2,0,0,0,1,56,2,5,1,5
+69010,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,13799.782578,14328.748539,2,103,7,7,1.48,5,5,0,1,1,2,80,1,4,3,NA
+69011,7,2,2,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,170311.62251,172039.874492,1,94,15,15,5,4,3,0,0,1,1,33,1,2,5,NA
+69012,7,2,1,1,14,2,2,2,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8331.647763,8430.145707,2,90,8,8,1.72,5,5,1,2,0,1,20,2,1,1,2
+69013,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,18876.596264,19790.278241,1,101,3,3,0.44,5,5,0,3,0,1,35,1,3,1,4
+69014,7,2,1,78,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60942.568495,64726.722108,1,91,10,10,4.3,2,2,0,0,2,1,78,1,4,1,4
+69015,7,2,1,0,1,2,2,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,7328.080818,7255.659238,2,102,15,15,2.43,7,7,3,2,0,1,28,2,5,1,4
+69016,7,2,2,13,NA,4,4,1,13,159,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14166.687432,14745.171396,1,100,1,1,0,4,4,1,2,0,2,35,1,2,5,NA
+69017,7,2,1,58,NA,4,4,2,NA,NA,2,NA,2,2,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16695.743237,21107.492082,3,90,12,12,NA,2,2,0,0,0,2,56,2,4,1,5
+69018,7,2,1,39,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,16058.989596,16964.610342,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+69019,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,138322.767578,151356.795097,1,101,13,13,NA,2,2,0,0,0,1,40,1,2,1,3
+69020,7,2,1,42,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,35551.992846,35164.643493,3,91,15,14,4.03,5,4,2,0,0,1,42,2,4,1,5
+69021,7,2,1,18,NA,1,1,1,18,220,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27186.265479,27336.512389,1,95,4,4,0.68,5,5,0,1,0,2,38,2,3,4,NA
+69022,7,2,2,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,42424.587753,44729.774642,2,102,4,4,1.16,2,2,0,0,0,1,48,1,4,1,4
+69023,7,2,2,7,NA,5,7,1,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14676.903347,14493.499136,1,102,5,5,1.27,3,3,0,2,0,2,38,1,2,3,NA
+69024,7,2,1,65,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11764.405491,12074.424659,1,97,4,4,1.34,1,1,0,0,1,1,65,1,5,3,NA
+69025,7,2,1,60,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,11937.570805,12597.633777,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+69026,7,2,2,1,19,1,1,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9469.751474,10066.755659,1,102,13,13,NA,6,6,1,2,0,2,36,2,4,6,NA
+69027,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,96536.931937,99301.799305,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+69028,7,2,1,0,5,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7757.493251,8192.524698,3,92,8,8,1.45,6,6,2,0,0,2,58,2,5,1,9
+69029,7,2,2,6,NA,4,4,2,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7631.175557,7797.231767,1,93,1,1,0.02,5,5,0,4,0,2,36,NA,NA,5,NA
+69030,7,1,1,36,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,22188.836739,0,1,101,13,13,NA,3,3,1,0,0,2,20,1,2,6,NA
+69031,7,2,1,22,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,5,NA,1,2,2,1,2,2,1,2,2,3,15196.92397,17064.888752,2,101,9,6,2.3,4,1,0,0,0,1,22,2,3,5,NA
+69032,7,2,2,64,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,2,NA,2,2,2,1,2,2,NA,NA,NA,NA,6287.91334,6572.718606,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+69033,7,2,1,2,NA,3,3,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50064.977404,56868.769772,1,95,14,14,2.98,5,5,1,2,0,1,33,1,4,1,5
+69034,7,2,1,33,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,2,6,NA,2,2,2,1,2,2,2,2,1,2,36418.962534,36838.502529,2,93,4,4,0.56,5,5,0,2,0,1,37,NA,NA,1,1
+69035,7,2,1,68,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,14,14,5,2,2,0,0,2,1,68,1,4,1,4
+69036,7,2,1,76,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,8764.234393,9213.337166,2,98,77,77,NA,2,2,0,0,2,2,70,1,3,1,2
+69037,7,2,1,41,NA,1,1,1,NA,NA,2,NA,2,1,4,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,37402.70356,39181.236579,2,102,15,15,5,3,3,1,0,0,1,41,2,2,1,5
+69038,7,2,2,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,99381.891022,103468.309204,2,92,15,15,5,2,1,0,0,0,2,29,1,5,1,NA
+69039,7,1,1,39,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,83081.99261,0,2,92,15,15,5,2,1,0,0,0,1,41,1,4,5,NA
+69040,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,92399.914697,92845.296279,2,92,15,15,5,5,5,0,3,0,2,46,1,5,1,5
+69041,7,2,1,8,NA,4,4,1,8,101,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10311.165779,10516.173846,2,98,4,4,1.29,2,2,0,1,0,2,27,1,2,5,NA
+69042,7,2,1,33,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,3,6,NA,2,2,2,1,2,2,1,2,2,1,43108.74283,43605.347908,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+69043,7,2,1,5,NA,5,6,2,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4526.77644,4888.797036,3,90,12,12,NA,5,5,1,2,0,1,37,2,5,1,5
+69044,7,2,2,0,8,2,2,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,7135.777159,7112.267328,1,90,7,7,1.56,4,4,1,1,0,2,37,1,2,77,NA
+69045,7,2,2,65,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,4,1,NA,2,2,2,1,2,2,1,2,2,2,13676.984152,14509.761798,2,91,8,8,3.3,2,2,0,0,2,1,65,NA,NA,1,4
+69046,7,2,1,3,NA,5,6,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8402.098771,9423.873047,3,91,7,7,2.16,3,3,1,0,1,2,36,2,5,1,NA
+69047,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,98514.948291,98181.677236,1,99,15,15,5,2,2,0,0,2,1,73,1,4,1,5
+69048,7,2,2,11,NA,3,3,1,11,143,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18800.96526,18551.296274,1,94,6,6,1.26,5,5,0,2,0,2,38,1,4,1,NA
+69049,7,2,1,71,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,1,2,1,2,2,1,1,2,NA,10776.442569,10987.528707,1,96,7,7,1,7,7,2,1,1,2,53,1,4,1,3
+69050,7,2,2,1,22,3,3,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39879.891371,42381.146401,1,91,14,14,3.06,5,5,2,0,0,2,30,1,5,1,5
+69051,7,2,2,5,NA,3,3,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51483.624552,53105.566664,1,94,15,15,5,4,4,2,0,0,1,51,2,5,1,5
+69052,7,1,2,40,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,3,1,2,2,1,2,2,NA,NA,NA,NA,115926.402585,0,1,101,15,15,5,4,4,0,2,0,2,40,1,4,1,3
+69053,7,2,2,43,NA,2,2,2,NA,NA,2,NA,2,1,1,NA,3,6,2,2,2,2,2,2,2,2,2,2,2,34503.935186,34683.870273,2,91,12,12,NA,5,5,0,1,1,2,43,2,3,6,NA
+69054,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19633.637051,21699.358471,2,95,5,5,1.05,3,3,0,1,0,1,43,1,3,1,2
+69055,7,2,2,28,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,53638.260635,56251.206542,2,96,7,7,2.38,2,2,0,0,0,1,29,1,3,1,4
+69056,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13593.59406,14525.996816,2,90,5,5,1.08,3,3,0,1,0,2,29,2,4,1,5
+69057,7,2,2,39,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,4,1,2,2,2,2,1,2,2,NA,NA,NA,NA,29148.354549,28361.870708,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+69058,7,2,1,43,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,6,NA,2,2,1,1,2,2,NA,NA,NA,NA,37324.655911,39220.856657,1,102,5,5,0.86,5,5,2,0,0,2,21,2,2,5,NA
+69059,7,2,1,2,NA,4,4,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5662.231921,6243.536586,1,99,13,13,NA,4,4,1,0,0,2,26,1,4,4,NA
+69060,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,137038.746155,146586.432966,2,101,3,3,0.92,2,1,0,0,0,1,21,1,4,5,NA
+69061,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,26105.275101,26256.723843,1,98,9,9,4.03,2,2,0,1,0,2,49,1,5,3,NA
+69062,7,2,2,70,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,2,NA,1,2,1,1,2,1,1,2,1,NA,10831.995402,11374.881633,3,90,77,77,NA,4,4,0,0,2,1,69,2,5,2,NA
+69063,7,2,1,2,NA,1,1,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10803.555682,11221.222518,2,91,6,6,1,6,6,1,3,0,2,35,2,2,1,1
+69064,7,2,2,6,NA,4,4,1,6,81,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11195.065587,11390.355382,2,96,2,2,0.31,4,4,0,2,0,2,30,NA,NA,6,NA
+69065,7,2,1,28,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,15171.949804,3,91,15,6,2.75,2,1,0,0,0,2,30,NA,NA,6,NA
+69066,7,2,1,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60942.568495,64726.722108,1,94,9,9,3.97,2,2,0,0,2,1,74,1,4,1,4
+69067,7,2,1,53,NA,5,6,1,NA,NA,2,NA,2,2,77,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16590.074977,17115.36835,1,92,12,12,NA,4,4,0,1,0,1,53,2,5,1,4
+69068,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,16614.865368,17238.21833,3,91,15,15,5,3,3,1,0,0,2,39,2,5,1,5
+69069,7,2,1,11,NA,4,4,1,11,136,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10269.191209,10473.364732,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+69070,7,2,2,4,NA,5,6,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6683.092466,6676.033575,3,91,6,6,1.34,4,4,1,1,0,1,36,2,4,1,NA
+69071,7,2,1,4,NA,5,6,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7580.437211,8294.186048,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+69072,7,1,2,0,3,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6440.205112,0,2,103,14,14,3.86,4,4,2,0,0,2,37,2,5,1,NA
+69073,7,2,2,3,NA,5,6,1,3,44,NA,NA,2,2,1,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8052.137959,8043.633038,1,100,7,7,2.2,3,3,1,0,0,2,28,2,2,1,3
+69074,7,2,1,19,NA,1,1,2,19,237,2,NA,2,2,4,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26704.187335,28443.712885,1,95,9,9,2.46,4,4,0,0,0,1,42,2,2,1,3
+69075,7,2,1,2,NA,4,4,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5163.141902,5376.483607,2,99,13,13,NA,5,5,2,0,0,2,21,1,3,5,NA
+69076,7,2,1,18,NA,3,3,1,18,220,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24446.632088,24751.360191,1,94,4,4,0.79,3,3,0,1,0,1,49,1,2,3,NA
+69077,7,2,2,58,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,35630.227837,37045.269032,2,102,6,6,2.75,1,1,0,0,0,2,58,1,2,3,NA
+69078,7,2,2,59,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,27617.038547,29182.278408,1,93,14,14,5,2,2,0,0,0,2,59,2,5,5,NA
+69079,7,2,1,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,18683.628164,19333.389409,1,98,5,4,1.7,3,1,0,0,0,1,32,1,5,5,NA
+69080,7,2,1,16,NA,4,4,1,16,202,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18413.211403,2,101,4,4,0.86,3,3,0,1,0,2,18,1,2,NA,NA
+69081,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,15992.133387,15205.505219,2,95,15,15,3.85,7,7,0,3,1,2,62,1,4,2,NA
+69082,7,2,1,8,NA,1,1,2,8,97,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14134.674028,14215.818763,1,96,8,8,2.62,3,3,0,1,0,1,41,2,3,1,5
+69083,7,2,1,14,NA,3,3,2,14,171,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20051.633575,19994.174398,2,97,5,5,0.84,5,5,0,2,0,2,33,1,4,1,3
+69084,7,2,1,2,NA,4,4,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,10545.226242,2,101,1,1,0.21,4,4,1,2,0,2,26,1,3,5,NA
+69085,7,2,2,49,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,2,5,NA,2,2,2,1,2,2,1,2,2,2,28065.587512,28753.959014,2,93,4,4,0.56,5,5,0,0,0,2,49,2,2,5,NA
+69086,7,2,2,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,104934.725755,111516.131691,2,98,10,10,4.42,2,2,0,0,0,1,25,1,5,1,5
+69087,7,2,2,15,NA,3,3,2,15,189,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,36586.371708,39087.782259,1,98,3,3,0.5,5,5,0,3,0,2,56,1,3,3,NA
+69088,7,2,2,67,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,10614.141896,11057.12801,2,93,4,4,0.99,2,2,0,0,2,1,72,2,3,1,4
+69089,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,12170.074646,12490.78413,1,97,9,7,3.36,2,1,0,0,1,1,61,1,5,3,NA
+69090,7,2,1,2,NA,3,3,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20538.767297,24065.985233,2,95,3,3,0.52,3,3,1,0,0,1,37,1,4,1,4
+69091,7,2,1,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,132969.642582,135979.403496,2,94,8,8,4.59,1,1,0,0,0,1,52,1,3,3,NA
+69092,7,2,2,0,4,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16423.151355,17405.028297,1,99,15,15,5,3,3,1,0,0,2,31,1,5,1,5
+69093,7,2,1,16,NA,5,6,2,16,201,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9099.599144,9725.105491,1,90,9,9,2.6,4,4,0,1,0,2,49,2,2,1,5
+69094,7,2,1,3,NA,3,3,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21357.821814,24096.698658,2,96,3,3,0.53,5,5,3,0,0,2,26,1,4,1,4
+69095,7,2,1,1,14,1,1,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12387.68972,13272.515505,1,94,6,6,1.3,4,4,2,0,0,1,24,2,1,1,4
+69096,7,1,1,12,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13731.625553,0,1,100,8,8,1.95,4,4,0,2,0,2,42,1,4,1,4
+69097,7,2,1,31,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,2,2,2,1,2,2,2,2,2,2,36946.697686,39295.803447,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+69098,7,2,1,14,NA,1,1,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18242.832494,18343.652859,1,102,6,6,1.34,4,4,0,1,0,2,48,2,3,1,1
+69099,7,2,1,6,NA,4,4,2,6,82,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9399.447563,9986.021526,2,90,7,7,1.61,4,4,1,1,1,2,65,1,3,2,NA
+69100,7,1,2,77,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,14125.505993,0,1,99,14,14,5,2,2,0,0,2,1,79,NA,NA,1,5
+69101,7,2,1,58,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,25042.846308,24675.381142,1,93,14,6,2.94,5,1,1,0,0,2,22,2,1,6,NA
+69102,7,2,2,61,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,1,NA,1,2,2,1,2,2,2,2,2,2,5852.076897,6537.978856,2,90,7,7,0.89,7,7,1,3,3,1,60,2,3,1,3
+69103,7,2,1,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,16995.648055,16598.645683,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+69104,7,2,1,5,NA,1,1,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14993.478359,14600.446315,1,97,6,3,0.63,7,3,2,1,0,1,29,2,2,1,1
+69105,7,2,1,55,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,1,5,NA,2,2,2,2,2,2,1,2,2,2,37557.946192,37006.841228,2,102,5,5,1.08,3,3,0,0,0,1,55,2,1,5,NA
+69106,7,2,2,0,8,5,7,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7434.184594,7384.233557,1,93,15,15,5,3,3,1,0,0,2,34,1,5,1,5
+69107,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,7323.703412,7380.991723,2,100,4,4,0.97,3,3,0,0,3,2,80,1,5,2,NA
+69108,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,22154.880695,2,97,5,5,1.08,3,3,0,0,0,1,38,1,4,5,NA
+69109,7,2,1,1,19,1,1,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11793.948458,12249.904005,1,100,5,5,0.65,6,6,1,2,0,1,32,2,2,1,2
+69110,7,2,2,56,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,24870.513993,25282.537357,2,98,8,8,3.06,2,2,0,1,0,2,56,1,3,3,NA
+69111,7,2,1,9,NA,3,3,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,40950.556053,42121.524492,2,95,15,15,4.77,4,4,0,2,0,2,36,1,4,1,5
+69112,7,2,2,23,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,16929.836231,18556.498474,2,101,4,3,1.1,2,1,0,0,0,2,23,2,4,5,NA
+69113,7,2,2,8,NA,3,3,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,51531.402068,50847.086518,1,94,6,6,1.57,3,3,0,1,0,2,28,1,4,1,4
+69114,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,NA,NA,NA,NA,NA,NA,NA,84761.905629,91116.013503,1,95,NA,NA,NA,5,5,0,2,0,2,37,1,3,1,NA
+69115,7,2,1,79,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,8517.336599,8684.171961,2,100,4,4,1.02,2,2,0,0,1,1,79,1,1,2,NA
+69116,7,2,1,14,NA,3,3,2,14,170,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,88198.948426,91036.751291,1,98,10,10,3.04,4,4,0,2,0,2,47,1,4,1,3
+69117,7,2,1,28,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21399.459455,20960.973241,1,90,15,15,4.34,4,4,0,0,1,1,62,2,5,1,3
+69118,7,2,1,11,NA,5,6,2,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,1,1,2,2,NA,4852.395137,5120.382571,1,99,6,6,1.07,6,6,2,1,2,1,44,2,5,4,NA
+69119,7,2,1,16,NA,4,4,1,16,203,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15008.400374,15566.530345,1,100,7,7,2.78,2,2,0,1,0,2,37,2,4,5,NA
+69120,7,2,2,39,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,27303.803575,27591.614113,1,96,3,3,0.43,4,4,1,1,0,2,39,2,4,1,3
+69121,7,2,2,5,NA,5,7,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,92,3,3,0.46,5,5,2,1,0,1,30,1,3,1,2
+69122,7,2,1,16,NA,4,4,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10732.729133,10991.603368,3,90,10,10,3.67,3,3,0,1,0,2,52,2,3,5,NA
+69123,7,2,2,63,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,128590.415432,131137.819973,1,102,15,15,5,2,2,0,0,2,2,63,1,4,1,4
+69124,7,2,2,19,NA,5,6,2,20,NA,2,NA,2,2,2,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10353.070387,10520.518116,2,101,5,3,1.1,2,1,0,0,0,1,29,2,4,1,NA
+69125,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,133853.800452,137185.265763,3,91,5,5,1.93,1,1,0,0,0,2,49,1,3,3,NA
+69126,7,2,1,3,NA,3,3,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,77607.205684,90935.051698,1,98,6,6,1.31,3,3,1,0,0,1,30,1,5,1,5
+69127,7,2,1,8,NA,3,3,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,54897.892683,58386.014643,2,98,7,7,1.61,4,4,1,1,0,1,43,NA,NA,6,NA
+69128,7,2,1,8,NA,4,4,1,8,97,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9184.716222,9199.031651,1,102,1,1,0,5,5,0,3,0,2,41,1,4,1,4
+69129,7,2,1,18,NA,2,2,1,18,225,2,NA,2,2,2,66,NA,NA,NA,2,2,2,2,2,2,1,2,2,2,22721.243258,22896.459408,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+69130,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,31540.022655,39899.113022,2,91,3,3,0.86,2,2,0,0,0,2,41,1,4,1,3
+69131,7,2,2,25,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,22941.698794,22487.759922,2,96,5,5,1.08,3,3,0,0,0,1,50,1,3,1,4
+69132,7,2,1,6,NA,4,4,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11277.594097,11665.009628,1,90,5,5,0.74,5,5,0,2,0,2,18,1,4,NA,NA
+69133,7,2,2,1,20,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7623.406054,7742.751523,1,97,4,4,0.46,7,7,3,3,0,2,31,1,3,1,NA
+69134,7,2,2,2,NA,4,4,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5687.793894,6002.477845,2,90,8,5,1.36,5,2,2,0,0,2,25,2,3,5,NA
+69135,7,1,1,40,NA,5,6,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17210.953493,0,1,91,15,15,5,5,5,2,1,0,1,40,1,5,1,5
+69136,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,4,6,2,2,2,2,2,2,2,1,2,2,2,36453.846815,35470.245447,1,102,13,13,NA,6,6,1,2,0,2,36,2,4,6,NA
+69137,7,2,1,56,NA,4,4,2,NA,NA,2,NA,2,2,8,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15760.921402,15756.795819,3,90,15,15,5,4,4,0,0,0,1,56,2,4,1,5
+69138,7,2,2,16,NA,3,3,1,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71832.578284,73443.760792,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+69139,7,2,2,17,NA,2,2,1,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,2,2,2,NA,NA,NA,NA,27070.679378,27847.54398,1,92,6,6,0.93,5,5,0,2,0,1,47,2,1,1,1
+69140,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,44862.41765,53799.51702,1,92,7,7,2.64,2,2,0,0,2,1,80,1,3,1,3
+69141,7,2,1,80,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,8497.912951,8999.288803,1,96,6,6,1.82,2,2,0,0,2,1,80,1,1,1,9
+69142,7,2,1,72,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,38226.070503,39676.327756,1,99,9,9,5,1,1,0,0,1,1,72,1,5,2,NA
+69143,7,2,2,10,NA,4,4,1,10,130,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7489.549692,8064.290136,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+69144,7,2,2,71,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,14936.784897,17875.873101,2,94,5,5,2.2,1,1,0,0,1,2,71,1,3,2,NA
+69145,7,2,1,18,NA,1,1,1,18,217,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15469.666055,15805.71937,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+69146,7,2,1,41,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,28361.113525,28406.771707,1,97,8,8,4.7,1,1,0,0,0,1,41,1,3,3,NA
+69147,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,1,1,2,2,1,17605.619977,18470.699991,3,91,15,15,5,3,3,0,1,0,2,40,2,5,1,5
+69148,7,2,1,5,NA,3,3,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,73006.119819,83450.013654,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+69149,7,2,1,62,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7791.74373,7821.068134,1,91,7,7,3.54,1,1,0,0,1,1,62,1,3,3,NA
+69150,7,2,2,6,NA,4,4,2,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8757.841043,9104.895674,2,101,5,5,1.03,4,4,1,1,0,2,31,1,3,5,NA
+69151,7,2,1,57,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,119599.307776,123574.267321,1,93,15,15,4.59,4,4,0,1,0,1,57,1,5,1,5
+69152,7,2,1,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17743.990528,2,97,3,3,0.82,2,2,0,0,0,1,24,1,3,5,NA
+69153,7,2,1,10,NA,4,4,2,10,124,NA,NA,2,1,3,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10229.206765,10406.656985,1,96,9,9,2.78,4,4,0,2,0,1,54,2,5,4,NA
+69154,7,2,1,66,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,29999.543427,32474.673757,2,91,2,2,0.56,1,1,0,0,1,1,66,1,3,3,NA
+69155,7,2,2,9,NA,3,3,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74327.830279,75742.187091,2,91,6,6,1.34,4,4,1,2,0,2,33,1,4,3,NA
+69156,7,2,1,12,NA,3,3,1,12,155,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,84860.612577,85065.529717,2,98,15,15,5,4,4,0,2,0,1,48,1,3,1,4
+69157,7,2,2,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,1,2,2,1,112960.559471,113970.503883,1,94,99,99,NA,5,5,1,1,0,2,21,1,3,5,NA
+69158,7,2,1,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32719.762791,33772.872733,1,91,1,1,0.2,2,2,0,0,0,1,44,1,2,1,2
+69159,7,2,2,22,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,5,5,2,2,2,2,2,2,2,1,2,2,1,35424.746838,36356.556801,2,93,NA,3,0.98,4,1,0,0,0,1,28,NA,NA,4,NA
+69160,7,2,1,17,NA,1,1,1,17,208,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,27186.265479,27776.842846,1,92,15,15,4.99,4,4,0,2,0,2,43,1,4,1,4
+69161,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,22233.683089,25178.838093,1,102,2,2,0.49,3,3,1,0,0,1,20,1,2,6,NA
+69162,7,2,1,72,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,4,5,NA,2,2,2,2,2,2,2,2,2,NA,12250.041239,12463.587232,2,100,1,1,0,1,1,0,0,1,1,72,2,4,5,NA
+69163,7,2,2,10,NA,5,7,1,10,123,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9007.62445,9504.796896,2,102,15,15,5,4,4,0,2,0,1,39,1,4,1,5
+69164,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,51433.469947,56330.524019,1,98,7,7,3.8,1,1,0,0,1,2,80,1,5,2,NA
+69165,7,2,1,72,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,18239.933451,19545.291055,2,102,8,8,3.67,2,2,0,0,2,1,72,2,5,1,NA
+69166,7,2,1,60,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,6038.685119,6085.921613,1,96,15,15,5,3,3,0,0,1,1,60,2,5,1,5
+69167,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,27681.279,30792.259764,1,91,3,3,1.29,1,1,0,0,1,2,80,1,2,2,NA
+69168,7,2,1,75,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,4,NA,1,2,2,1,2,2,1,2,2,NA,8902.117734,9243.371936,1,97,15,15,5,6,6,0,1,1,2,53,1,4,1,NA
+69169,7,2,1,10,NA,4,4,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10122.702296,10604.04698,2,100,3,3,0.42,5,5,0,1,0,2,51,1,4,5,NA
+69170,7,2,2,51,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22224.73066,22340.630739,2,91,10,10,2.95,4,4,0,1,0,2,18,1,3,NA,NA
+69171,7,2,2,40,NA,2,2,1,NA,NA,2,NA,2,1,2,NA,4,5,2,2,2,2,2,2,2,2,2,1,2,32281.860274,32750.128834,2,93,7,7,2.31,2,2,0,0,1,2,40,2,4,5,NA
+69172,7,2,2,65,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9570.416297,9969.842041,2,98,6,6,2.75,1,1,0,0,1,2,65,1,4,3,NA
+69173,7,2,1,14,NA,4,4,1,14,171,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14848.504688,14877.861397,1,98,8,8,1.95,4,4,0,2,0,2,31,1,2,1,5
+69174,7,2,2,12,NA,4,4,1,12,152,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11838.873374,11750.256617,2,100,15,15,4.47,4,4,0,2,0,1,39,NA,NA,1,5
+69175,7,2,2,32,NA,2,2,2,NA,NA,1,1,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,27127.983961,26396.013964,2,90,15,15,4.2,5,5,1,0,0,2,50,NA,NA,6,NA
+69176,7,2,2,57,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,26668.458882,26828.838508,2,102,5,5,1.84,1,1,0,0,0,2,57,1,3,6,NA
+69177,7,2,1,66,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,8,8,3.4,2,2,0,0,2,1,66,1,5,1,4
+69178,7,2,2,28,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,1,1,2,2,1,10800.351372,11472.015663,2,92,3,3,0.45,4,4,0,0,1,1,64,2,1,1,1
+69179,7,1,2,14,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15847.19263,0,1,98,15,15,5,4,4,0,2,0,2,50,1,5,1,5
+69180,7,2,2,1,16,4,4,2,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,7906.792868,2,95,10,10,2.32,6,6,1,2,0,1,44,1,4,1,4
+69181,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,5,2,1,2,2,1,2,2,1,2,2,1,90299.161173,91599.679037,1,95,7,7,2.54,2,2,0,1,0,2,37,1,1,5,NA
+69182,7,2,2,8,NA,2,2,2,8,103,NA,NA,2,2,1,3,NA,NA,NA,2,1,2,1,2,2,2,2,2,2,10762.400563,12070.126078,2,90,3,3,0.46,5,5,1,3,0,2,35,2,1,4,NA
+69183,7,2,1,6,NA,1,1,1,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10658.399025,10722.994279,1,102,8,8,1.33,7,7,1,4,0,2,32,1,3,1,2
+69184,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,34082.896551,1,92,4,4,1.16,2,2,0,0,1,1,56,1,2,1,3
+69185,7,2,1,60,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,2,1,2,2,2,2,1,2,6449.12882,6552.435629,2,93,7,7,1.83,3,3,0,1,1,1,60,2,2,1,4
+69186,7,1,1,5,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,0,2,100,2,2,0.25,4,4,2,1,0,2,39,1,2,5,NA
+69187,7,2,2,22,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,1,5,2,2,2,2,2,2,2,NA,NA,NA,NA,31196.446669,31060.57243,2,99,99,1,0,5,1,0,1,0,1,40,2,1,6,NA
+69188,7,2,2,8,NA,5,7,2,8,100,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9620.269705,10273.007637,2,91,12,12,NA,4,4,0,2,0,1,40,1,5,1,5
+69189,7,2,1,3,NA,4,4,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,10162.494907,1,100,15,15,5,3,3,1,0,0,2,34,1,5,1,5
+69190,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,9,2,NA,1,1,2,1,2,2,1,1,2,NA,14971.827573,15430.401607,2,97,13,13,NA,4,4,1,0,1,2,45,1,2,5,NA
+69191,7,2,1,0,11,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7757.493251,8192.524698,3,92,1,1,0.18,5,5,2,2,0,2,31,2,1,5,NA
+69192,7,2,2,0,2,5,7,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9288.555962,9478.459453,1,95,6,6,0.96,5,5,1,0,1,2,69,1,1,2,NA
+69193,7,2,2,12,NA,2,2,2,12,147,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,19480.517135,20370.224649,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+69194,7,2,2,6,NA,4,4,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10332.067017,11124.939356,1,100,3,3,0.52,3,3,1,1,0,2,25,1,3,5,NA
+69195,7,2,2,10,NA,3,3,2,10,130,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14935.786363,14737.445343,2,94,7,7,1.18,7,7,1,4,0,2,31,1,4,6,NA
+69196,7,2,1,27,NA,5,7,1,NA,NA,2,NA,2,2,3,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,32866.0119,33575.905633,1,98,99,99,NA,3,1,0,0,0,2,22,1,4,5,NA
+69197,7,2,2,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8832.868731,8863.029978,1,102,2,2,0.19,7,7,2,2,0,1,48,2,9,1,9
+69198,7,2,2,13,NA,1,1,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,18368.872199,20151.001259,2,94,5,5,0.67,6,6,1,3,0,1,37,2,3,1,4
+69199,7,2,1,60,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,2,4,NA,1,2,1,1,2,1,1,2,1,3,6814.038157,7682.939373,2,91,12,13,NA,7,1,0,4,2,2,72,2,1,2,NA
+69200,7,2,1,34,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,22957.82581,23425.428596,2,102,15,15,3.92,5,5,1,2,0,1,34,2,5,1,5
+69201,7,2,2,44,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20834.783335,21091.811515,2,94,15,15,5,5,5,0,2,1,1,47,2,5,1,5
+69202,7,2,1,9,NA,1,1,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,17882.621856,17720.218058,3,92,7,7,1.3,5,5,1,2,0,2,33,2,2,1,1
+69203,7,2,2,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8679.600111,2,95,9,9,4.92,1,1,0,0,1,2,66,1,5,3,NA
+69204,7,2,2,16,NA,3,3,1,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30430.369428,31312.351655,3,92,5,5,1.03,4,4,0,3,0,1,55,1,4,4,NA
+69205,7,2,2,47,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,14872.222375,15689.845848,2,92,7,7,1.61,4,4,0,2,0,1,51,2,3,1,3
+69206,7,1,1,80,NA,2,2,NA,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,13654.270555,0,2,93,7,7,2.31,2,2,0,0,2,1,80,2,1,1,1
+69207,7,2,2,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,27381.645976,29718.515739,2,95,5,4,1.52,3,1,0,0,0,1,31,1,4,5,NA
+69208,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,1,2,NA,25840.959268,29976.982116,1,101,3,3,0.86,2,2,0,0,2,2,80,1,2,1,1
+69209,7,1,2,47,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,35469.911999,0,2,91,6,6,2.78,1,1,0,0,0,2,47,1,4,3,NA
+69210,7,2,1,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26556.735732,2,101,1,1,0.09,2,1,0,0,0,1,23,1,5,5,NA
+69211,7,2,1,0,2,5,6,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8377.359485,8321.07116,2,96,4,4,0.97,3,3,1,0,0,1,31,2,5,1,5
+69212,7,2,1,38,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,2,4,NA,2,2,2,1,2,2,1,2,2,2,41241.224595,41716.316195,2,102,6,6,1.03,5,5,1,1,0,1,37,1,2,1,2
+69213,7,2,2,47,NA,3,3,1,NA,NA,2,NA,2,2,7,NA,5,1,NA,2,2,2,1,2,2,2,2,2,2,27349.117265,27628.99562,2,93,6,6,1.65,2,2,0,0,0,1,52,2,3,1,5
+69214,7,2,2,0,1,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4137.127382,4416.757179,2,97,1,1,0.12,5,5,1,2,0,2,24,1,3,5,NA
+69215,7,2,2,9,NA,1,1,1,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19194.228411,19883.529671,2,96,7,7,1.34,5,5,0,2,0,1,24,2,2,5,NA
+69216,7,2,1,35,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22937.913723,23137.542903,1,100,3,3,0.43,4,4,0,2,0,1,35,1,4,1,3
+69217,7,2,2,16,NA,5,7,2,16,193,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,31716.869763,32129.640601,1,101,3,3,0.92,2,2,0,1,0,2,53,1,5,3,NA
+69218,7,2,2,2,NA,1,1,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9132.13761,10082.124987,1,103,13,13,NA,4,4,2,0,0,2,27,2,2,6,NA
+69219,7,2,1,4,NA,2,2,1,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15915.595287,17052.411707,2,96,6,6,1.25,4,4,1,1,0,1,31,2,3,1,3
+69220,7,2,2,21,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,6,1,1,2,2,1,2,2,1,2,2,1,52280.406546,52747.829022,1,92,1,1,0,2,1,0,0,0,1,30,1,3,6,NA
+69221,7,2,2,51,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17950.494975,18684.288461,1,92,15,15,5,4,4,0,2,0,1,55,1,5,1,5
+69222,7,2,1,6,NA,3,3,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47178.298856,52231.302145,2,95,8,8,2.17,4,4,1,1,0,1,43,1,4,1,5
+69223,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,100722.771154,105550.354971,2,92,14,8,4.13,2,1,0,0,0,2,25,1,5,5,NA
+69224,7,2,2,59,NA,2,2,2,NA,NA,2,NA,2,1,4,NA,1,6,NA,2,2,2,1,2,2,2,2,2,2,18341.621382,19145.40337,2,90,6,6,1.21,4,4,0,0,0,2,59,2,1,6,NA
+69225,7,2,2,17,NA,1,1,2,17,209,2,NA,2,2,4,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17544.592739,19246.751055,3,92,6,6,1,6,6,1,1,0,1,42,2,1,1,4
+69226,7,2,2,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7588.605543,8103.359102,1,99,14,14,4.21,4,4,0,2,0,2,44,1,5,1,5
+69227,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14994.337564,15663.818698,1,100,15,15,5,2,2,0,0,1,2,48,1,5,5,NA
+69228,7,2,1,17,NA,3,3,2,17,215,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,61479.689958,61628.148021,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+69229,7,2,1,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,93164.281583,96130.599912,2,95,8,8,2.17,4,4,1,1,0,1,43,1,4,1,5
+69230,7,2,2,8,NA,1,1,1,8,102,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16412.026403,3,92,4,4,0.81,3,3,0,2,0,2,31,1,3,1,NA
+69231,7,2,1,10,NA,4,4,2,10,131,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12701.072484,14129.347101,2,91,6,6,0.78,7,7,1,4,0,2,38,2,2,77,NA
+69232,7,2,1,0,6,2,2,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,6837.213048,7220.636221,2,91,1,1,0.02,5,5,1,2,0,2,27,2,3,1,2
+69233,7,2,2,2,NA,3,3,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26167.521666,28880.852575,1,93,6,6,1.16,4,4,2,0,0,2,33,1,5,1,4
+69234,7,2,1,8,NA,4,4,2,8,107,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12176.538896,12609.8265,2,97,2,2,0.27,4,4,0,2,0,1,51,1,2,4,NA
+69235,7,2,2,3,NA,5,7,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27969.562936,30869.748923,3,91,7,7,1.57,4,4,2,0,0,2,29,2,3,1,3
+69236,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18417.587863,18945.077049,2,97,4,4,1.35,2,2,0,0,0,2,27,1,4,1,4
+69237,7,2,2,49,NA,2,2,2,NA,NA,1,2,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,46417.079566,46883.794079,1,97,15,15,5,2,2,0,0,0,2,49,2,5,1,4
+69238,7,2,2,5,NA,3,3,2,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,54058.97777,57449.540919,1,90,8,8,1.67,5,5,2,1,0,2,28,1,4,1,5
+69239,7,2,2,19,NA,5,6,1,19,239,2,NA,2,2,1,15,NA,NA,NA,1,2,1,1,2,1,1,2,1,NA,9200.381964,9723.240452,2,101,2,1,0.32,2,1,0,0,0,2,20,NA,NA,5,NA
+69240,7,2,2,7,NA,5,7,1,7,89,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8504.389189,9345.267202,2,93,6,6,1.95,2,2,0,1,0,1,30,1,4,5,NA
+69241,7,2,1,36,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,19332.808712,19683.545891,1,90,15,15,5,1,1,0,0,0,1,36,1,5,5,NA
+69242,7,1,1,25,NA,2,2,NA,NA,NA,2,NA,2,7,77,NA,2,5,NA,2,2,2,1,2,2,NA,NA,NA,NA,44074.735764,0,2,91,1,1,0.17,4,4,0,1,0,1,49,2,1,6,NA
+69243,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105483.169118,109229.181554,1,93,15,15,4.59,4,4,0,2,0,2,45,1,5,1,5
+69244,7,1,2,10,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7899.813226,0,2,100,15,15,5,4,3,0,2,0,1,42,1,3,5,NA
+69245,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,21398.47235,20994.797177,1,103,14,8,4.21,2,1,0,0,0,2,29,1,5,5,NA
+69246,7,2,1,19,NA,5,7,2,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7672.605662,7940.039559,1,93,14,14,5,2,2,0,0,0,2,53,2,5,3,NA
+69247,7,2,2,80,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,12344.929687,12774.490696,1,93,2,2,0.83,1,1,0,0,1,2,80,1,4,3,NA
+69248,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,3,1,2,1,2,1,NA,NA,NA,1,2,1,NA,18255.735511,22006.513457,2,91,6,6,1.57,3,3,0,1,0,1,41,2,3,1,3
+69249,7,2,1,34,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,1,2,1,2,2,1,1,2,NA,19923.530941,20672.635795,2,95,6,6,1.08,4,4,1,0,0,2,42,1,4,4,NA
+69250,7,2,2,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,17622.141982,18017.336291,2,95,8,8,2.97,2,2,0,0,0,2,56,1,5,2,NA
+69251,7,2,2,5,NA,1,1,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14665.744588,16191.375588,2,96,5,5,0.89,4,4,1,1,0,2,36,2,4,6,NA
+69252,7,2,1,11,NA,1,1,1,11,140,NA,NA,2,2,3,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14820.807433,15173.085782,2,102,7,7,1.79,4,4,0,2,0,1,40,2,2,6,NA
+69253,7,2,2,45,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,3,NA,2,2,2,2,2,2,1,2,2,2,36457.299109,41822.014095,1,90,5,5,1.79,1,1,0,0,0,2,45,2,2,3,NA
+69254,7,2,1,4,NA,1,1,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15835.271712,16336.527884,1,100,8,3,0.68,6,3,1,0,0,1,33,2,3,6,NA
+69255,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,42993.150248,46481.579201,1,92,6,6,1.98,2,2,0,0,2,1,80,1,5,1,4
+69256,7,2,1,56,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,34082.896551,1,101,3,3,0.88,2,2,0,0,0,1,56,1,2,1,4
+69257,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,24344.236685,23945.277508,2,96,6,6,1.48,4,4,1,1,0,2,25,1,4,5,NA
+69258,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21621.615608,21991.273471,1,100,15,15,5,4,4,0,0,0,1,54,1,5,1,5
+69259,7,2,1,30,NA,5,7,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,84910.063417,90031.493834,2,96,14,6,2.75,3,1,0,0,0,1,30,2,5,5,NA
+69260,7,2,2,3,NA,5,6,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4882.863582,4877.706148,1,99,7,7,2.89,2,2,1,0,0,2,35,2,5,1,NA
+69261,7,2,2,51,NA,3,3,1,NA,NA,2,NA,2,1,8,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,27769.056387,28050.845299,1,94,99,99,NA,3,3,0,0,2,1,74,NA,NA,1,NA
+69262,7,2,2,2,NA,1,1,1,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12871.484115,13281.030392,2,102,14,14,3.58,4,4,2,0,0,1,25,2,3,1,1
+69263,7,2,1,73,NA,1,1,1,NA,NA,2,NA,2,1,9,NA,1,1,NA,1,2,2,1,2,2,2,2,2,NA,14673.422679,15506.302145,2,98,5,5,0.59,7,7,2,1,2,2,71,1,2,1,1
+69264,7,2,1,69,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11212.469396,11300.176858,1,100,15,15,5,3,3,0,0,1,1,69,1,5,1,4
+69265,7,2,2,11,NA,1,1,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17053.854294,17233.146259,3,92,1,1,0.26,2,2,0,1,0,2,45,1,2,3,NA
+69266,7,2,1,27,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,17265.374984,17186.028516,2,93,4,4,0.56,5,5,2,1,0,1,27,1,2,6,NA
+69267,7,2,1,29,NA,3,3,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,94165.180484,100725.804227,2,99,14,14,5,2,2,0,0,0,2,28,1,5,1,5
+69268,7,2,2,11,NA,4,4,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11076.064101,11269.278002,2,102,4,4,0.53,6,6,2,2,0,2,27,1,2,1,2
+69269,7,2,1,16,NA,3,3,2,16,195,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,91343.020442,91081.271456,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+69270,7,2,1,66,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,9048.959172,9513.611982,1,93,12,12,NA,4,4,0,0,2,1,66,2,4,1,2
+69271,7,2,1,1,21,1,1,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12569.23571,12239.751631,2,102,15,15,2.43,7,7,3,2,0,1,28,2,5,1,4
+69272,7,2,2,43,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,1,6,2,2,2,2,1,2,2,NA,NA,NA,NA,32606.880052,32776.922157,2,93,4,4,0.56,5,5,0,2,0,1,37,NA,NA,1,1
+69273,7,2,1,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8890.779467,9258.147648,3,91,1,1,0.07,6,6,2,3,0,2,30,1,2,3,NA
+69274,7,2,2,21,NA,4,4,2,NA,NA,2,NA,2,1,3,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18446.691823,17539.327743,1,90,4,4,0.58,6,6,0,3,0,2,21,2,5,5,NA
+69275,7,2,1,13,NA,4,4,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,11265.8817,2,95,99,99,NA,2,2,0,1,1,2,80,1,2,2,NA
+69276,7,2,1,16,NA,1,1,1,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22768.423624,22886.980387,2,98,15,15,5,3,3,0,1,0,1,38,1,4,1,3
+69277,7,2,1,24,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,15171.949804,3,91,6,6,2.94,2,1,0,0,0,1,24,1,5,5,NA
+69278,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,13549.492282,14020.967918,2,93,6,6,2.28,2,2,0,0,1,2,78,1,5,3,NA
+69279,7,2,2,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,126314.769628,126413.297707,2,95,15,15,5,2,2,0,0,1,1,61,1,5,1,5
+69280,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,22419.63376,27869.657009,1,94,3,3,1.29,1,1,0,0,1,2,80,1,3,2,NA
+69281,7,2,1,3,NA,3,3,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,76114.759421,89186.300704,1,98,15,15,5,4,4,1,1,0,1,40,1,4,1,5
+69282,7,2,2,19,NA,3,3,2,19,236,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,31627.471241,33789.841724,1,101,1,1,0.08,3,3,1,0,0,2,19,1,2,NA,NA
+69283,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,10298.451562,11068.740647,1,93,99,1,0.32,2,1,0,0,1,1,50,NA,NA,5,NA
+69284,7,2,1,53,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,26651.800212,26887.922292,2,92,3,3,0.92,1,1,0,0,0,1,53,2,4,3,NA
+69285,7,2,2,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,37537.641876,38018.476349,2,97,5,5,0.84,5,5,0,2,0,2,33,1,4,1,3
+69286,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,19096.438335,22378.694988,1,93,15,15,5,1,1,0,0,0,1,39,2,5,5,NA
+69287,7,2,1,54,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,173139.914798,174527.091604,1,95,14,14,5,2,2,0,0,0,1,54,1,4,1,3
+69288,7,2,2,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,10252.529496,2,100,3,3,0.75,2,2,0,0,2,1,67,1,3,1,2
+69289,7,2,2,5,NA,2,2,1,5,69,NA,NA,2,1,3,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15979.952759,16987.381382,2,102,7,7,1.53,5,5,1,2,0,2,37,2,4,1,4
+69290,7,2,2,1,21,1,1,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9955.153132,10990.75621,2,94,4,4,0.81,4,4,2,0,0,1,26,2,2,1,2
+69291,7,2,1,0,9,5,6,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,4326.150649,4507.366261,1,99,6,6,1.07,6,6,2,1,2,1,44,2,5,4,NA
+69292,7,2,1,12,NA,3,3,1,12,147,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,109750.935844,116575.440818,1,94,7,7,1.52,4,4,0,2,2,1,61,2,1,1,5
+69293,7,2,1,9,NA,3,3,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23188.935049,25672.571973,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+69294,7,2,1,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6946.177172,7374.968672,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+69295,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,30658.109612,1,94,5,5,1.2,3,3,0,1,0,1,49,1,4,5,NA
+69296,7,2,2,6,NA,5,6,2,6,78,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9849.323746,10517.603056,2,91,10,10,3.04,4,4,1,1,0,1,37,2,5,1,5
+69297,7,2,2,7,NA,2,2,1,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13498.913367,14670.348602,2,93,7,7,1.56,4,4,1,1,0,1,35,2,4,1,4
+69298,7,2,1,1,20,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6429.93791,7090.057975,1,90,3,3,0.63,3,3,1,1,0,2,32,1,4,5,NA
+69299,7,2,1,72,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,14781.102246,16162.46005,1,97,12,12,NA,4,4,0,0,2,1,72,1,2,1,3
+69300,7,2,1,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,29159.620934,2,101,1,1,0.33,1,1,0,0,0,1,23,1,4,5,NA
+69301,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,99831.393624,101809.075589,2,94,8,8,3.4,2,2,0,0,2,2,64,1,4,1,2
+69302,7,2,2,14,NA,4,4,1,14,176,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17373.928778,17416.718116,2,96,8,8,1.72,5,5,0,3,0,1,39,1,5,1,4
+69303,7,2,2,12,NA,1,1,1,12,154,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,29109.978779,29945.36646,3,92,10,10,2.82,4,4,0,1,1,1,36,1,3,1,5
+69304,7,2,1,11,NA,3,3,2,11,138,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,44777.275016,47565.734765,1,91,14,14,3.06,5,5,0,3,0,2,46,1,5,1,5
+69305,7,2,1,14,NA,4,4,2,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12462.601191,12584.643654,2,97,4,4,0.57,5,5,1,3,0,2,33,1,3,5,NA
+69306,7,2,2,0,5,4,4,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7123.540273,7201.320939,2,101,1,1,0.16,3,3,2,0,0,2,21,1,2,5,NA
+69307,7,2,1,58,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,21168.33083,21825.00909,2,99,5,5,1.79,1,1,0,0,0,1,58,1,2,3,NA
+69308,7,2,1,12,NA,4,4,2,13,156,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11351.725436,11256.943498,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+69309,7,2,2,45,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,27149.46917,27176.128744,1,94,2,1,0,2,1,0,0,0,1,53,1,3,6,NA
+69310,7,2,2,18,NA,2,2,2,18,224,2,NA,2,2,2,66,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,16896.101801,17234.594007,1,93,14,8,2.01,5,4,1,0,0,2,22,2,1,6,NA
+69311,7,2,2,16,NA,1,1,1,16,198,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,20502.928313,21226.512457,3,92,9,9,2.46,4,4,0,2,0,1,43,2,3,1,4
+69312,7,1,2,6,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13257.060167,0,2,90,14,14,3.06,5,5,1,2,0,1,42,1,4,1,5
+69313,7,2,1,1,14,4,4,2,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6580.937346,6661.20735,2,97,5,5,1.08,3,3,1,0,0,1,28,1,3,5,NA
+69314,7,2,1,50,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,16117.991297,17081.203544,1,96,7,7,1.69,4,4,0,1,0,2,19,2,4,NA,NA
+69315,7,1,1,11,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15653.970322,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+69316,7,2,2,17,NA,5,6,2,17,214,2,NA,2,2,2,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7588.544207,7881.983727,3,91,4,4,0.69,5,5,0,2,0,1,45,2,4,1,1
+69317,7,2,2,3,NA,4,4,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10447.689164,11241.830083,1,102,7,7,1.8,5,4,1,0,2,1,47,1,3,5,NA
+69318,7,2,1,4,NA,4,4,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11324.865632,11670.771889,2,96,2,2,0.44,3,3,2,0,0,2,22,1,2,5,NA
+69319,7,2,1,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18790.641284,19715.999527,2,100,15,1,0.05,4,1,0,2,0,1,42,1,3,5,NA
+69320,7,2,2,58,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,4,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,29087.427528,29444.125326,1,101,3,3,0.41,5,5,0,2,1,2,36,2,4,4,NA
+69321,7,2,1,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21390.017054,21518.859775,2,99,15,15,5,2,2,0,0,0,2,51,1,5,1,5
+69322,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,24460.137584,1,92,4,4,0.61,5,5,1,2,0,1,34,1,3,6,NA
+69323,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,7903.072331,8212.17704,2,99,NA,77,NA,3,2,0,0,1,1,63,1,3,5,NA
+69324,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,153755.392794,158780.137655,1,91,15,15,5,3,3,0,0,0,2,50,1,4,1,4
+69325,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,20039.469886,21176.851739,1,97,15,15,5,3,3,1,0,0,1,40,1,3,1,5
+69326,7,2,2,2,NA,5,6,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6213.806926,6511.003426,1,95,4,4,0.62,5,5,2,0,2,2,29,2,3,5,NA
+69327,7,2,2,10,NA,3,3,2,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55626.447796,54887.751745,1,95,15,15,5,4,4,0,2,0,2,42,1,5,1,5
+69328,7,2,2,71,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,13495.958207,13916.488113,1,96,9,9,4.23,2,2,0,0,2,2,71,2,5,1,5
+69329,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,42559.487719,42202.626733,1,98,3,2,0.62,2,1,0,0,0,2,50,1,2,3,NA
+69330,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,NA,NA,NA,NA,23614.167119,30467.110717,2,97,1,1,0.33,2,2,1,0,0,2,24,1,2,5,NA
+69331,7,2,2,13,NA,4,4,2,13,166,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10848.628906,11198.221038,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+69332,7,2,1,15,NA,4,4,1,15,186,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17606.165994,20773.995354,2,101,9,9,2.46,4,4,0,2,0,1,42,1,3,1,3
+69333,7,2,2,76,NA,5,7,1,NA,NA,2,NA,2,1,5,NA,1,1,NA,1,2,2,1,2,2,1,2,1,NA,33487.945981,34633.22923,1,102,10,10,3.22,4,4,0,0,2,2,29,2,5,5,NA
+69334,7,2,2,58,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,24938.6524,26291.237881,2,98,5,5,1.24,3,3,0,0,1,2,58,1,2,5,NA
+69335,7,2,1,54,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32838.149884,1,95,5,5,1.84,1,1,0,0,0,1,54,1,4,3,NA
+69336,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,19401.570044,22254.979865,1,92,6,6,1.24,4,4,1,1,0,1,30,1,3,3,NA
+69337,7,2,2,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,NA,NA,NA,NA,22912.222762,23105.605227,2,96,6,6,1.35,3,3,1,1,0,2,23,1,2,5,NA
+69338,7,2,2,13,NA,1,1,1,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15690.47168,16249.379042,1,102,8,8,1.33,7,7,1,4,0,2,32,1,3,1,2
+69339,7,2,1,22,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,1,6,NA,2,2,2,2,2,2,1,2,2,2,38560.502118,39812.43256,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+69340,7,2,1,14,NA,5,6,2,14,179,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6666.045669,7091.184391,3,90,10,10,2.41,5,5,1,2,0,1,44,2,4,1,5
+69341,7,2,1,6,NA,1,1,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11159.151566,11057.808004,1,102,2,2,0.31,4,4,1,2,0,2,25,1,2,4,NA
+69342,7,2,1,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,16645.008535,17770.553389,2,95,2,2,0.4,2,2,0,0,0,2,43,1,1,1,NA
+69343,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,118671.226879,120869.497653,2,91,15,15,5,3,3,0,0,0,2,54,1,4,1,4
+69344,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26091.362984,2,101,5,5,1.5,2,2,0,0,0,1,47,1,4,3,NA
+69345,7,2,1,11,NA,3,3,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18216.94614,18737.854061,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+69346,7,2,1,18,NA,2,2,1,18,226,2,NA,2,2,4,13,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,15506.325263,16662.012915,1,103,5,5,0.74,5,5,0,1,0,1,47,2,1,1,1
+69347,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,20137.063643,20085.335673,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+69348,7,2,2,13,NA,3,3,1,13,166,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,31627.471241,33789.841724,1,98,6,6,1.11,5,5,1,2,0,2,32,1,2,1,2
+69349,7,2,1,74,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,59483.758768,62962.247193,1,96,14,14,5,2,2,0,0,2,1,74,1,5,1,4
+69350,7,1,2,77,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,12888.823036,0,2,99,7,7,3.13,1,1,0,0,1,2,77,1,5,2,NA
+69351,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,66503.043118,69008.73133,2,98,7,7,1.94,3,3,1,0,0,2,31,1,4,1,NA
+69352,7,2,1,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9064.168162,9228.63108,3,92,12,12,NA,4,4,2,0,0,1,30,1,3,1,4
+69353,7,2,2,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7871.443574,8074.618887,1,96,9,9,2.88,3,3,1,0,0,1,27,1,3,1,5
+69354,7,2,1,42,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,34205.013302,34329.398365,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+69355,7,2,2,27,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,NA,NA,NA,NA,44119.608456,46238.164206,2,98,1,1,0.14,3,2,2,0,0,2,27,1,3,6,NA
+69356,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,11982.125462,2,101,5,5,1.08,3,3,0,1,1,2,62,1,4,2,NA
+69357,7,2,1,8,NA,4,4,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,8894.789377,9317.745565,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+69358,7,2,1,11,NA,3,3,1,11,134,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41342.668304,42524.849071,1,102,8,8,2.42,4,4,0,2,0,2,34,1,4,1,3
+69359,7,2,1,8,NA,4,4,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9022.8939,9530.778148,2,100,3,3,0.31,7,7,3,2,0,2,28,1,3,1,3
+69360,7,2,2,0,0,5,7,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6280.554922,6238.355238,2,92,10,10,3.78,3,3,1,0,0,1,35,1,4,6,NA
+69361,7,2,1,66,NA,4,4,1,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,10209.064769,10089.274206,1,98,6,6,1.57,3,3,0,0,2,1,66,2,2,1,4
+69362,7,2,1,5,NA,4,4,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8892.687359,9565.637195,2,96,12,10,2.17,7,6,2,3,0,1,29,1,4,3,NA
+69363,7,2,2,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,44119.608456,46238.164206,3,92,3,3,0.54,4,4,3,0,0,2,22,1,3,5,NA
+69364,7,2,2,11,NA,1,1,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,10118.363218,11093.371216,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+69365,7,2,2,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,185476.232403,190233.31946,1,95,4,4,1.47,1,1,0,0,0,2,59,1,1,5,NA
+69366,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,89167.746947,90120.008289,1,95,15,15,5,2,2,0,0,2,1,72,1,3,1,4
+69367,7,2,1,2,NA,3,3,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46257.816906,54201.886729,2,94,15,15,4.59,4,4,1,1,0,2,37,1,5,1,5
+69368,7,2,1,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16851.334496,16952.838472,2,95,14,14,5,2,2,0,0,0,1,52,1,4,1,NA
+69369,7,2,1,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,7233.371983,2,95,2,2,0.63,1,1,0,0,1,1,66,1,1,3,NA
+69370,7,2,1,17,NA,1,1,1,18,216,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20560.901695,21038.44606,2,96,10,10,3.78,3,3,0,1,0,1,42,1,3,1,3
+69371,7,2,1,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,22650.334558,22327.472235,1,93,12,12,NA,5,1,0,2,0,1,32,1,2,5,NA
+69372,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,23022.732862,22484.94345,2,102,5,3,0.92,5,1,2,1,0,1,24,1,4,6,NA
+69373,7,2,2,20,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114492.825466,119073.624167,2,101,3,2,0.73,3,1,0,0,0,2,20,2,4,5,NA
+69374,7,1,2,52,NA,2,2,NA,NA,NA,2,NA,2,2,5,NA,1,6,NA,2,2,2,1,2,2,NA,NA,NA,NA,24004.6026,0,2,91,1,1,0.17,4,4,0,1,0,1,49,2,1,6,NA
+69375,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,148501.534483,154474.828602,2,91,15,15,5,2,2,0,0,1,2,66,1,5,3,NA
+69376,7,2,1,19,NA,3,3,1,19,236,2,NA,2,1,5,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,41421.091412,40909.179995,1,94,3,3,0.93,2,2,0,0,0,2,41,2,2,3,NA
+69377,7,1,1,70,NA,5,6,NA,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,17564.38036,0,2,102,5,5,1.36,2,2,0,0,2,1,70,2,5,1,4
+69378,7,2,2,20,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,3,5,2,2,2,2,1,2,2,2,2,2,2,39788.68078,43305.780488,2,99,3,3,0.52,3,3,0,0,1,2,38,2,3,3,NA
+69379,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,22824.336433,1,92,1,1,0,2,1,0,0,0,1,30,1,3,6,NA
+69380,7,2,1,20,NA,2,2,1,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,42587.888675,2,98,1,1,0.13,4,4,2,0,0,2,52,1,2,4,NA
+69381,7,2,1,6,NA,2,2,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13837.588743,14145.695126,1,92,9,9,2.93,3,3,0,1,0,2,30,1,5,1,5
+69382,7,2,1,6,NA,5,6,2,6,79,NA,NA,2,2,1,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,9720.482616,11328.640802,2,91,99,99,NA,7,4,0,4,0,1,36,2,9,1,2
+69383,7,2,2,10,NA,5,7,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5010.242859,5375.15147,3,91,15,15,4.47,4,4,0,3,0,2,44,2,5,1,NA
+69384,7,2,1,7,NA,4,4,2,7,90,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10665.048307,10877.092316,1,96,4,4,1.12,2,2,0,1,0,2,44,1,2,5,NA
+69385,7,2,1,12,NA,5,6,2,12,154,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6666.045669,7091.184391,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+69386,7,2,1,52,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19541.667675,19738.253068,2,95,3,3,1.21,1,1,0,0,0,1,52,1,3,5,NA
+69387,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,71632.519751,73567.29218,2,100,8,8,4.13,1,1,0,0,0,1,25,1,3,5,NA
+69388,7,2,1,61,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,11568.876339,13082.418262,1,102,4,4,0.78,4,4,0,1,1,1,61,1,3,1,1
+69389,7,2,2,22,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,53638.260635,55237.528848,2,96,5,1,0.37,2,1,0,0,0,2,22,1,3,6,NA
+69390,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,96485.877517,101155.423058,2,95,9,9,3.97,2,2,0,0,1,1,62,1,2,1,2
+69391,7,2,2,52,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,30868.065568,31427.853631,3,92,10,10,3.77,3,3,0,1,0,2,52,1,4,6,NA
+69392,7,2,1,35,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,13147.594977,13793.622874,1,102,15,15,4.59,4,4,1,1,0,1,35,1,5,1,5
+69393,7,2,1,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23511.361566,23589.086028,1,100,14,14,4.45,3,3,1,0,0,1,33,1,4,1,4
+69394,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,39061.30289,2,101,4,4,1.22,2,2,1,0,0,2,25,1,4,5,NA
+69395,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20813.587171,20953.352876,2,93,6,6,1.47,3,3,0,0,0,2,47,1,4,5,NA
+69396,7,2,1,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8439.71412,8914.772112,2,99,7,7,1.19,6,6,1,3,0,2,38,1,3,5,NA
+69397,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,30152.053647,30291.005728,2,101,14,14,4.03,4,4,0,1,0,2,40,1,5,1,5
+69398,7,2,1,3,NA,4,4,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9388.597537,2,100,99,99,NA,6,6,2,1,0,2,44,1,3,1,4
+69399,7,1,1,58,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,25583.266805,0,1,100,10,10,4.63,2,2,0,0,1,1,58,1,5,1,4
+69400,7,2,1,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,NA,NA,NA,1,2,2,1,21600.805431,21945.582953,2,96,3,3,0.38,5,5,1,2,0,2,30,1,3,5,NA
+69401,7,2,1,0,0,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4724.065742,4677.378988,2,93,77,77,NA,7,7,3,1,0,2,43,2,1,1,9
+69402,7,2,2,59,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14262.400197,16582.236492,1,93,14,14,5,2,2,0,0,0,1,36,2,5,1,NA
+69403,7,2,1,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,115789.798768,119024.647099,1,90,6,6,2.86,1,1,0,0,1,1,61,1,5,3,NA
+69404,7,2,1,57,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20536.772571,20857.586819,1,90,15,15,5,2,2,0,0,0,1,57,2,4,1,5
+69405,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,152467.08796,155127.202914,1,91,8,8,2.17,4,4,0,0,0,1,59,1,4,1,5
+69406,7,2,1,16,NA,5,6,2,16,197,NA,NA,2,2,2,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8168.705487,9098.657657,1,93,7,7,1.64,5,5,0,2,0,1,47,2,5,1,1
+69407,7,2,1,14,NA,5,6,1,14,173,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,1,1,2,2,1,6121.087833,6878.034577,2,92,7,7,1.17,6,6,0,1,1,1,78,2,1,1,3
+69408,7,2,2,16,NA,1,1,1,16,202,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,2,2,1,1,2,2,1,13963.420591,14243.160208,2,103,2,2,0.42,3,3,0,2,0,2,51,2,2,5,NA
+69409,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,21605.242088,21013.957961,3,91,14,14,5,1,1,0,0,0,2,50,1,5,5,NA
+69410,7,2,2,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23409.362971,23343.581342,2,96,15,15,5,3,3,0,1,0,1,55,1,5,1,4
+69411,7,2,1,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,68266.732554,70856.701199,1,95,7,7,2.72,2,2,0,0,2,1,75,1,5,1,5
+69412,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,38833.86357,40362.1181,3,92,3,3,0.93,2,2,0,0,1,2,75,1,1,3,NA
+69413,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15477.500537,2,100,5,5,1.08,3,3,0,1,0,2,50,1,4,3,NA
+69414,7,2,2,4,NA,3,3,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,51483.624552,56822.002088,1,91,4,4,1.29,2,2,1,0,0,2,26,1,4,5,NA
+69415,7,2,2,15,NA,1,1,2,15,182,NA,NA,2,2,4,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18368.872199,18995.639955,2,94,12,12,NA,4,4,0,2,0,1,47,2,2,1,2
+69416,7,2,1,2,NA,5,6,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8100.706553,9085.828699,1,98,15,15,5,4,4,1,1,0,1,40,NA,NA,1,5
+69417,7,2,1,27,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,124091.929364,126503.690206,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+69418,7,2,1,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,60803.589953,64579.113865,1,91,15,15,5,2,2,0,0,2,1,71,1,5,1,4
+69419,7,2,1,57,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,25963.141347,25880.331023,1,92,10,10,4.3,2,2,0,0,0,2,55,1,4,1,5
+69420,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,NA,NA,NA,1,2,2,NA,60163.952904,72064.333099,1,97,NA,NA,NA,2,1,0,0,2,1,80,1,3,6,NA
+69421,7,2,2,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,149765.47604,152732.36321,1,92,7,7,2.64,2,2,0,0,2,2,68,1,4,1,4
+69422,7,2,2,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,38477.954475,38315.213578,1,97,15,15,5,6,6,0,1,1,2,53,1,4,1,NA
+69423,7,2,1,6,NA,3,3,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58429.74688,62860.392832,1,100,6,6,1.78,3,3,1,1,0,2,35,1,5,4,NA
+69424,7,2,1,4,NA,4,4,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7359.751824,8115.329859,2,99,5,5,1.26,3,3,1,0,0,2,50,2,3,5,NA
+69425,7,2,2,66,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15268.186173,15857.48365,1,102,14,14,5,2,2,0,0,2,1,65,2,5,1,5
+69426,7,2,2,14,NA,5,6,2,14,175,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11975.458482,12226.374363,1,97,14,14,2.87,5,5,0,3,0,2,40,2,5,1,5
+69427,7,2,1,44,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,18774.610796,1,99,3,3,0.54,3,3,1,0,0,2,29,1,4,1,4
+69428,7,2,2,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,33331.144292,35071.927834,1,101,7,7,2.71,2,2,0,0,0,1,43,1,4,1,4
+69429,7,2,2,18,NA,1,1,1,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24481.187693,25353.22751,1,95,4,4,0.68,5,5,0,1,0,2,38,2,3,4,NA
+69430,7,2,2,67,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,1,1,NA,1,2,1,1,2,1,1,2,2,NA,14102.354333,14719.644428,3,91,14,4,1.02,6,2,0,0,2,1,48,2,1,1,1
+69431,7,2,2,79,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,12183.823561,12607.778632,2,100,9,9,4.35,2,2,0,0,2,2,79,1,5,1,3
+69432,7,2,1,0,5,4,4,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5782.580039,5871.334502,2,99,7,7,1.65,4,4,2,1,0,2,29,1,3,4,NA
+69433,7,2,1,53,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,29883.483388,29823.443389,1,102,7,7,1.89,3,3,0,0,0,1,53,2,1,1,1
+69434,7,2,2,43,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,40880.818805,42051.054889,1,101,4,4,0.84,3,3,0,1,0,1,42,1,4,1,4
+69435,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,47157.788417,52669.767304,2,95,15,15,5,2,2,0,0,2,1,80,1,5,1,5
+69436,7,2,1,46,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,28353.771379,28938.361413,1,92,8,8,4.66,1,1,0,0,0,1,46,1,4,5,NA
+69437,7,2,2,6,NA,3,3,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19880.837381,19616.828143,1,95,6,6,1.3,4,4,0,3,0,2,46,1,4,3,NA
+69438,7,2,1,16,NA,3,3,2,16,201,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,29222.804528,30590.757203,1,90,7,7,1.55,5,5,0,3,0,1,51,2,3,1,2
+69439,7,2,1,67,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7903.072331,8272.775405,2,99,8,8,3.57,2,2,0,0,1,1,67,1,2,1,2
+69440,7,2,1,3,NA,5,6,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6940.398869,7784.416682,2,100,15,15,5,4,4,1,1,0,1,36,2,5,1,5
+69441,7,2,2,9,NA,3,3,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,62419.284609,61940.470984,1,101,14,14,4.5,3,3,0,1,0,1,39,1,2,1,5
+69442,7,2,1,19,NA,4,4,1,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,2,1,0.18,4,1,0,0,0,1,19,1,4,NA,NA
+69443,7,2,2,29,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,11378.639129,11910.020755,1,102,15,15,3.82,5,5,1,1,0,1,29,1,4,1,4
+69444,7,2,2,2,NA,4,4,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7618.827213,8513.998744,2,97,5,5,1.63,2,2,1,0,0,2,33,1,3,5,NA
+69445,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,18490.479848,17984.43936,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+69446,7,2,1,25,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,5,NA,2,2,2,1,2,2,1,2,2,2,38560.502118,39812.43256,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+69447,7,2,2,13,NA,4,4,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16659.324602,16974.753267,1,92,15,15,4.44,5,5,0,3,0,2,43,1,5,6,NA
+69448,7,2,1,3,NA,4,4,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8892.687359,9565.637195,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+69449,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19260.892847,19877.962631,2,97,12,12,NA,3,3,0,0,0,1,33,1,4,5,NA
+69450,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,93796.829073,96460.739647,1,100,15,15,5,4,4,1,1,0,1,40,1,5,1,5
+69451,7,2,1,67,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7736.56115,7765.677873,1,99,13,13,NA,3,3,0,0,2,1,67,1,2,1,2
+69452,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,175633.860713,177416.12057,3,91,15,15,5,4,4,0,0,1,1,60,NA,NA,4,NA
+69453,7,2,2,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,171038.159272,172773.783849,3,92,15,15,5,3,3,0,0,0,1,56,NA,NA,1,4
+69454,7,2,2,13,NA,4,4,2,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,6,6,0.96,5,5,0,4,0,2,36,1,4,4,NA
+69455,7,2,2,80,NA,3,3,1,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,1,1,2,2,1,2,1,NA,49191.372812,61149.379277,3,91,15,15,3.33,6,6,0,2,2,1,80,2,3,1,3
+69456,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,21897.080981,21827.239495,3,91,12,12,NA,2,2,0,0,0,1,52,1,5,1,5
+69457,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,13408.721263,14895.884708,2,98,3,3,0.68,2,2,0,0,2,1,80,1,1,1,3
+69458,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,167711.394252,174540.034956,1,101,7,7,2.31,2,2,0,0,1,1,60,1,3,1,3
+69459,7,2,2,62,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,15207.312407,15896.113669,2,98,10,10,4.43,2,2,0,0,1,2,47,1,4,3,NA
+69460,7,2,1,50,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,23259.140149,22917.848121,1,90,10,10,2.44,5,5,1,0,0,2,56,2,1,1,1
+69461,7,2,2,13,NA,5,6,2,13,160,NA,NA,1,1,NA,7,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6937.463063,7205.726089,3,90,15,15,3.23,6,6,0,2,0,1,50,2,2,1,2
+69462,7,2,2,36,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,42468.064168,42460.239922,2,101,6,6,0.96,5,5,0,4,0,2,36,1,4,4,NA
+69463,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,27771.028362,29170.559675,3,92,4,4,1.16,2,2,0,0,0,2,20,1,2,1,2
+69464,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105261.096488,116369.237268,1,97,15,15,5,4,4,0,2,0,1,39,1,5,1,NA
+69465,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,134213.669088,137679.44344,1,93,7,7,2.16,3,3,0,1,0,2,50,1,5,3,NA
+69466,7,2,2,8,NA,5,6,1,8,99,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,1,2,1,1,6189.617175,6489.651129,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+69467,7,1,1,3,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,77702.196479,0,1,95,14,14,3.8,4,4,1,1,0,1,36,1,4,1,5
+69468,7,2,1,69,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,7736.56115,7645.782314,1,99,6,6,1.84,2,2,0,0,2,1,69,1,3,1,4
+69469,7,2,1,0,2,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8519.229538,8996.978293,1,101,4,4,0.57,5,5,1,2,0,1,28,2,1,1,1
+69470,7,2,2,4,NA,1,1,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,16462.187772,17500.018071,3,92,7,7,1.3,5,5,1,2,0,2,33,2,2,1,1
+69471,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,94,14,14,5,2,2,0,0,2,2,72,1,5,1,5
+69472,7,2,1,38,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18544.003944,19241.239961,2,100,14,14,3.76,4,4,1,1,0,1,38,1,3,1,NA
+69473,7,2,1,79,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,8456.543684,8955.478754,2,92,7,7,3.13,1,1,0,0,1,1,79,1,1,2,NA
+69474,7,2,2,66,NA,2,2,1,NA,NA,2,NA,2,2,5,NA,3,5,NA,2,2,2,1,2,2,1,2,2,2,11570.073931,12052.956309,2,93,7,7,1.74,4,4,1,0,1,2,24,NA,NA,4,NA
+69475,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20430.250572,20111.803159,2,97,5,5,0.76,5,5,0,0,0,2,50,1,4,5,NA
+69476,7,2,2,0,5,4,4,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4584.111679,4588.281921,1,97,4,4,0.46,7,7,3,3,0,2,31,1,3,1,NA
+69477,7,2,1,0,5,1,1,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6390.260233,6748.618798,2,96,6,6,1,6,6,1,3,0,2,32,2,1,6,NA
+69478,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,34874.122648,37228.07844,3,92,6,6,0.74,7,7,2,1,0,2,46,1,2,1,4
+69479,7,2,2,47,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,25713.328161,27565.019075,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+69480,7,2,2,1,13,4,4,1,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8480.466509,9125.076563,1,100,9,9,2.22,5,5,1,2,0,2,40,2,4,1,4
+69481,7,2,2,0,2,1,1,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6787.112205,7016.795893,2,94,7,7,1.33,6,6,2,0,0,2,32,2,2,1,2
+69482,7,2,2,3,NA,1,1,1,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17282.036547,19079.832121,2,102,3,3,0.45,4,4,2,0,0,1,21,2,2,6,NA
+69483,7,2,1,4,NA,5,6,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10273.602479,10769.310961,1,92,77,77,NA,4,4,1,1,0,2,40,2,4,1,4
+69484,7,2,1,17,NA,1,1,1,17,210,2,NA,2,2,5,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,19996.544021,19901.195571,1,100,7,7,1.3,5,5,0,3,0,1,43,2,2,1,4
+69485,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22329.779038,21943.745693,1,98,14,14,3.93,3,3,1,0,0,1,36,1,5,1,5
+69486,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11679.736252,12242.903977,1,103,8,8,4.66,1,1,0,0,1,2,67,1,5,3,NA
+69487,7,2,1,21,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14852.674152,15542.18804,1,92,14,14,3.9,4,4,0,0,0,2,55,2,5,1,5
+69488,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,33784.29429,34197.992084,1,95,5,5,1.3,3,3,0,0,1,2,19,1,3,NA,NA
+69489,7,2,2,62,NA,5,6,2,NA,NA,2,NA,2,2,7,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,12941.046565,13394.767563,1,93,9,9,5,1,1,0,0,1,2,62,2,4,5,NA
+69490,7,2,1,0,3,3,3,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22795.485282,22401.400888,1,95,9,9,2.46,4,4,2,0,0,2,25,1,5,1,5
+69491,7,2,2,67,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,2,NA,2,2,2,2,2,2,2,2,2,2,10490.803278,11351.088215,1,93,2,2,0.77,1,1,0,0,1,2,67,2,1,2,NA
+69492,7,2,2,15,NA,4,4,2,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,18905.695776,2,101,3,3,0.54,3,3,0,2,0,2,36,1,3,5,NA
+69493,7,2,2,17,NA,4,4,2,17,209,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,7,7,1.74,4,4,0,2,0,2,47,1,5,4,NA
+69494,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,97803.500399,98104.213748,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+69495,7,2,2,22,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,3,5,2,1,2,2,1,2,2,1,2,2,3,16239.242782,18510.062544,2,101,9,3,1.1,4,1,0,0,0,1,22,2,3,5,NA
+69496,7,2,2,9,NA,4,4,2,9,116,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7631.175557,7916.380746,1,93,12,12,NA,3,3,0,1,0,2,48,1,3,5,NA
+69497,7,2,1,26,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,13749.764747,14666.581717,1,90,4,4,0.78,4,4,0,0,1,1,69,2,4,1,3
+69498,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,59510.728426,64021.429591,1,99,15,15,5,3,3,1,0,0,2,31,1,5,1,5
+69499,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,129098.803996,133683.475932,1,93,15,15,5,4,4,0,2,0,2,42,1,5,1,NA
+69500,7,2,1,19,NA,3,3,2,19,234,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,119111.433099,126517.990149,2,94,15,15,3.82,5,5,0,0,0,1,50,1,5,1,5
+69501,7,2,2,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,11331.751026,1,101,2,2,0.22,4,4,1,0,0,2,25,1,4,6,NA
+69502,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,143785.115498,143265.966876,1,99,8,8,4.87,1,1,0,0,0,2,50,1,5,5,NA
+69503,7,2,2,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,35933.117795,41598.834451,1,98,2,2,0.72,1,1,0,0,0,2,44,1,4,3,NA
+69504,7,2,1,54,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13002.519536,13402.475336,1,93,15,15,4.59,4,4,0,2,0,2,45,1,5,1,5
+69505,7,2,1,60,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,27043.072221,28144.17465,2,101,5,4,1.47,2,1,0,0,1,2,49,1,3,6,NA
+69506,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,12256.510142,13615.881666,1,95,4,4,1.26,2,2,0,0,2,1,80,1,1,1,3
+69507,7,2,2,8,NA,5,6,2,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10699.45895,11290.011566,1,97,10,10,3.67,3,3,0,1,0,1,47,1,5,1,5
+69508,7,2,1,37,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,36924.956604,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+69509,7,2,1,62,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,1,2,2,NA,8609.250304,11228.904188,2,90,4,4,1.02,2,2,0,0,1,1,62,2,1,1,1
+69510,7,2,1,19,NA,5,6,1,19,234,2,NA,2,1,4,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6558.308393,6849.983523,2,92,99,1,0.28,4,1,0,0,0,1,19,1,4,NA,NA
+69511,7,2,2,65,NA,4,4,1,NA,NA,2,NA,2,2,3,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,13998.494211,15077.525188,2,102,4,4,1.38,1,1,0,0,1,2,65,2,2,4,NA
+69512,7,2,1,17,NA,3,3,2,17,213,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17395.533107,17264.495057,1,99,13,13,NA,4,4,0,2,0,1,55,NA,NA,1,4
+69513,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19320.837782,1,99,15,15,5,4,4,0,2,0,2,46,1,5,1,5
+69514,7,2,1,57,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20181.678047,20475.67653,1,96,14,14,3.99,4,4,0,0,0,2,53,1,3,1,4
+69515,7,2,1,63,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,6612.194774,6663.917441,2,99,7,7,3.55,2,1,0,0,1,1,63,1,4,3,NA
+69516,7,2,1,74,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,52448.388619,54438.22579,1,99,9,9,3.74,2,2,0,0,2,2,73,1,3,1,4
+69517,7,2,1,0,7,4,4,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6246.568228,6652.820193,2,95,7,7,1.41,5,5,2,0,0,2,53,1,3,3,NA
+69518,7,2,2,2,NA,4,4,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6226.488588,6570.976462,2,99,6,6,1.03,6,6,3,0,0,1,33,1,3,6,NA
+69519,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,19520.240895,25185.107635,2,95,1,1,0.13,2,2,1,0,0,2,22,1,3,5,NA
+69520,7,2,1,22,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,101419.325386,103801.228657,1,100,15,15,4.63,5,5,0,0,0,1,51,1,5,1,3
+69521,7,2,1,74,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,48254.793439,51076.634961,2,100,6,6,2.04,2,2,0,0,2,1,74,1,4,1,3
+69522,7,2,2,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,52701.331723,53172.517558,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+69523,7,2,1,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,128128.531162,128588.454794,1,99,8,8,4.78,1,1,0,0,0,1,56,1,3,3,NA
+69524,7,2,2,61,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,9943.806181,2,100,12,12,NA,2,2,0,0,1,1,56,1,5,1,4
+69525,7,2,2,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,18462.756377,17957.474615,1,99,2,2,0.48,2,2,0,0,0,2,44,1,3,1,3
+69526,7,2,1,19,NA,3,3,2,19,239,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22313.526699,26201.091455,1,99,5,2,0.73,4,1,0,0,0,1,19,1,3,NA,NA
+69527,7,2,1,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,39031.957066,39638.562591,1,95,6,4,1.47,2,1,0,0,0,2,28,1,2,6,NA
+69528,7,2,1,80,NA,1,1,1,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,14673.422679,14929.213705,2,98,15,15,4.56,4,4,0,0,3,1,80,1,1,1,NA
+69529,7,2,1,7,NA,3,3,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,82670.203859,85919.643529,1,97,14,14,3.36,4,4,0,2,0,2,49,1,5,1,5
+69530,7,2,2,9,NA,3,3,2,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,46660.163959,46766.329713,1,98,15,15,5,5,5,0,1,0,1,53,1,5,1,5
+69531,7,2,1,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,26847.643051,28917.411142,1,98,1,1,0.16,3,3,1,0,0,1,28,1,2,6,NA
+69532,7,2,2,47,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,2,1,2,1,NA,NA,NA,NA,19150.604366,19847.194665,3,91,6,6,1.81,3,3,0,1,0,2,47,2,3,1,3
+69533,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114993.808573,116714.079488,1,98,7,4,1.34,4,1,0,0,0,2,20,NA,NA,5,NA
+69534,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,21503.272394,20634.528185,1,96,14,10,5,2,1,0,0,0,2,29,1,5,6,NA
+69535,7,2,1,16,NA,5,7,1,16,199,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,59545.101745,58762.542429,1,102,8,8,1.6,7,7,0,4,0,2,39,1,4,1,4
+69536,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126314.769628,126413.297707,2,95,8,8,3.53,2,2,0,0,0,1,57,1,4,1,4
+69537,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140932.152825,151077.472396,1,97,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+69538,7,2,2,19,NA,2,2,2,20,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,26657.121865,27597.898247,1,97,15,15,5,3,3,0,0,0,1,45,1,4,1,3
+69539,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,27199.141352,29187.923269,1,98,7,7,3.58,1,1,0,0,1,1,80,1,2,2,NA
+69540,7,2,2,45,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,24625.307946,26182.57746,1,93,2,2,0.61,2,2,0,0,1,2,45,2,3,5,NA
+69541,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,34952.089182,39581.970787,1,94,4,4,1,3,3,0,1,0,2,41,1,4,5,NA
+69542,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,3,1,2,2,1,2,2,1,2,2,1,35126.205635,36772.568368,1,95,3,3,0.65,3,3,1,0,0,1,58,1,3,3,NA
+69543,7,2,2,9,NA,4,4,1,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9139.784234,9759.758014,2,100,14,14,4.86,3,3,0,1,0,2,41,1,4,3,NA
+69544,7,1,2,16,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,10,NA,NA,NA,2,2,2,1,2,2,NA,NA,NA,NA,20347.899985,0,2,94,5,5,0.65,6,6,0,2,0,1,53,NA,NA,6,NA
+69545,7,2,2,17,NA,3,3,1,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30430.369428,31312.351655,3,92,5,5,1.03,4,4,0,3,0,1,55,1,4,4,NA
+69546,7,2,1,26,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15976.466658,17286.192057,2,96,8,6,2.39,2,1,0,0,0,1,26,2,5,5,NA
+69547,7,2,1,73,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,8543.37447,9341.78969,3,90,14,14,2.97,5,5,0,2,1,1,73,2,3,2,NA
+69548,7,2,2,7,NA,4,4,2,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7710.907339,8302.634544,2,99,77,77,NA,4,4,1,1,0,2,47,1,2,77,NA
+69549,7,2,2,66,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10429.953328,10902.368494,2,94,15,15,3.42,6,6,0,1,2,1,40,1,3,1,4
+69550,7,2,1,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,16287.780872,16385.890285,1,96,9,6,2.3,2,1,0,0,0,1,58,1,4,6,NA
+69551,7,2,2,12,NA,4,4,1,12,148,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11838.873374,11750.256617,2,97,15,15,5,4,4,0,2,0,1,47,NA,NA,6,NA
+69552,7,2,2,64,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,10999.00871,11813.793381,2,98,6,6,1.25,4,4,1,0,1,1,46,1,2,6,NA
+69553,7,2,2,70,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,2,2,2,NA,15705.521724,21678.922586,3,90,12,12,NA,6,6,0,0,2,1,70,2,1,1,1
+69554,7,2,1,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9344.689579,9327.120408,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+69555,7,2,2,11,NA,4,4,2,11,136,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7282.523598,7776.514874,2,99,6,6,1.11,5,5,0,4,0,2,34,1,4,5,NA
+69556,7,2,1,59,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17738.246838,17681.670095,2,95,2,2,0.54,1,1,0,0,0,1,59,1,4,1,NA
+69557,7,2,2,0,1,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8775.375504,8746.463784,2,102,14,14,3.58,4,4,2,0,0,1,25,2,3,1,1
+69558,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,47973.37979,51335.952518,1,101,7,7,1.82,4,4,2,0,0,2,27,1,2,1,3
+69559,7,2,2,8,NA,1,1,2,8,100,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14300.71869,15541.734563,2,94,9,9,2.1,5,5,1,2,0,1,31,2,4,1,4
+69560,7,2,1,26,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,2,5,NA,2,2,2,2,2,2,1,2,2,1,59682.963348,66132.07802,2,102,5,5,1.08,3,3,0,0,0,1,55,2,1,5,NA
+69561,7,2,2,19,NA,4,4,1,20,NA,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11224.041366,11682.365019,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+69562,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,1,2,NA,27199.141352,29187.923269,1,98,77,77,NA,2,2,0,0,2,1,80,1,1,1,3
+69563,7,2,2,5,NA,5,6,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8818.200077,8968.551716,2,102,6,6,1.52,4,4,2,0,0,1,30,2,4,1,4
+69564,7,2,1,37,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18743.758781,18547.428727,1,98,15,15,5,6,6,3,0,0,1,37,2,5,1,4
+69565,7,2,2,6,NA,2,2,2,6,72,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10276.40638,10435.186077,2,99,99,99,NA,5,3,0,1,0,1,40,2,1,6,NA
+69566,7,2,1,80,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,2,2,2,2,2,1,NA,13654.270555,14429.301836,2,93,12,12,NA,2,2,0,0,2,2,79,NA,NA,1,1
+69567,7,2,2,10,NA,3,3,1,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18800.96526,18551.296274,1,94,3,3,0.37,5,5,0,3,0,2,29,1,4,4,NA
+69568,7,2,2,11,NA,4,4,1,11,137,NA,NA,2,2,2,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8504.389189,8822.229593,2,93,6,6,2.24,3,1,1,1,0,2,31,2,3,3,NA
+69569,7,2,2,61,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,17243.546687,17909.086027,1,92,6,6,2.15,2,2,0,0,2,2,61,2,4,1,5
+69570,7,1,1,8,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,73471.277275,0,2,101,8,8,1.72,5,5,0,3,0,1,37,1,3,1,3
+69571,7,2,1,19,NA,5,6,2,20,NA,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11506.937395,12032.024805,1,97,7,7,1.48,5,5,0,1,0,2,46,2,4,1,NA
+69572,7,2,2,42,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,4,4,2,1,2,1,1,2,1,NA,NA,NA,NA,13568.706187,14164.354411,3,90,8,4,1.61,3,1,0,0,1,2,68,2,2,4,NA
+69573,7,2,2,53,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,15790.702799,16017.473382,2,90,2,2,0.48,2,2,0,0,1,2,53,2,3,1,3
+69574,7,2,2,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,163194.688032,166437.638141,1,97,14,14,4.96,2,2,0,0,0,1,59,1,4,1,5
+69575,7,2,2,12,NA,4,4,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11990.226944,12023.57752,2,95,2,2,0.26,3,3,0,2,0,2,31,1,3,5,NA
+69576,7,2,1,2,NA,4,4,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7325.198118,7777.386797,1,100,13,13,NA,5,5,2,0,0,2,54,1,4,5,NA
+69577,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,18061.358948,18207.066814,2,93,3,3,1.07,1,1,0,0,0,1,52,1,3,4,NA
+69578,7,2,2,6,NA,1,1,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16412.026403,2,98,2,2,0.35,3,3,0,1,0,2,43,1,3,1,3
+69579,7,2,2,4,NA,5,6,1,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6443.362832,6436.557151,1,98,15,15,5,3,3,1,0,0,2,37,2,5,1,5
+69580,7,2,1,23,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20623.434727,20610.774284,1,93,3,3,0.7,3,3,1,0,0,1,23,2,4,1,2
+69581,7,2,1,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,16386.190684,16050.428992,2,90,15,15,5,4,4,0,0,0,1,57,2,5,1,5
+69582,7,2,2,30,NA,2,2,2,NA,NA,2,NA,2,7,77,NA,1,2,2,2,2,2,2,2,2,NA,NA,NA,NA,35353.005268,36684.976775,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+69583,7,2,2,4,NA,5,6,2,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4962.240532,5199.576603,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+69584,7,2,2,0,6,5,7,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14926.308861,15231.47558,1,102,8,8,1.91,5,5,1,2,0,2,38,1,5,1,4
+69585,7,1,2,4,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53166.229434,0,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+69586,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5255.532858,5319.636438,2,99,1,1,0.1,3,3,1,1,0,2,28,1,4,5,NA
+69587,7,2,1,77,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,16823.6553,18089.239033,1,92,9,9,3.64,2,2,0,0,2,1,77,1,5,1,4
+69588,7,2,2,10,NA,3,3,2,10,127,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,66519.408962,66670.760406,1,98,15,15,4.34,4,4,0,2,0,1,51,1,5,1,5
+69589,7,2,1,0,2,3,3,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17418.612172,17117.482228,1,91,14,14,5,3,3,1,0,0,2,30,1,5,1,5
+69590,7,2,2,19,NA,4,4,2,19,235,2,NA,1,1,NA,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,4,4,0.81,4,4,1,0,0,2,51,1,2,4,NA
+69591,7,2,2,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,122655.643802,123752.268878,3,91,15,15,5,2,2,0,0,0,1,29,1,4,1,4
+69592,7,1,1,16,NA,1,1,NA,NA,NA,NA,NA,2,2,3,10,NA,NA,NA,1,2,2,1,2,1,NA,NA,NA,NA,18242.832494,0,1,102,99,99,NA,5,5,0,2,1,1,52,2,1,1,1
+69593,7,2,2,1,18,4,4,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6306.491784,6469.273037,2,90,9,9,3.18,3,3,1,0,0,2,38,2,4,5,NA
+69594,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,22911.854766,1,95,6,6,1.09,5,5,0,3,0,1,31,1,4,1,4
+69595,7,2,2,3,NA,3,3,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,68579.013834,75690.025736,1,95,15,15,5,3,3,1,0,0,1,26,1,3,1,4
+69596,7,2,1,64,NA,4,4,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,8219.195224,8283.488354,2,99,10,10,3.67,3,3,0,0,1,1,64,2,4,1,5
+69597,7,2,2,2,NA,5,6,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6075.554662,6069.137471,1,93,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+69598,7,2,1,34,NA,3,3,2,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,87891.395784,90408.650551,2,99,15,15,5,1,1,0,0,0,1,34,2,5,1,NA
+69599,7,2,1,21,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,5,NA,2,2,2,1,2,2,1,2,2,1,42077.383821,44759.048785,1,102,4,4,0.67,4,4,0,1,0,1,23,2,4,5,NA
+69600,7,2,1,65,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,6815.198656,6994.794831,3,90,15,15,4.34,4,4,0,0,1,1,65,1,3,1,4
+69601,7,2,1,17,NA,3,3,2,17,205,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,93665.036597,95017.313859,1,91,15,15,5,4,4,0,1,0,1,45,1,5,1,5
+69602,7,2,1,49,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,1,2,2,1,2,1,1,2,NA,35406.972937,36103.049421,2,98,3,3,0.98,2,2,0,0,1,2,80,1,1,2,NA
+69603,7,2,1,26,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,35669.2076,40318.090187,2,94,4,4,0.81,4,4,2,0,0,1,26,2,2,1,2
+69604,7,2,2,1,15,1,1,1,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,14326.094268,14518.763259,3,92,4,4,0.65,4,4,2,0,0,2,24,2,2,1,2
+69605,7,2,1,12,NA,2,2,1,12,147,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14532.438529,15479.089621,2,93,14,14,3.52,5,5,1,2,0,1,44,1,5,1,5
+69606,7,2,2,11,NA,4,4,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10332.067017,11124.939356,1,100,12,12,NA,3,3,0,1,0,2,52,1,3,1,3
+69607,7,2,1,40,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,34025.01199,1,95,2,2,0.46,1,1,0,0,0,1,40,1,1,3,NA
+69608,7,2,1,75,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,9662.124837,10964.58979,2,92,4,4,1.43,1,1,0,0,1,1,75,1,3,3,NA
+69609,7,2,1,4,NA,4,4,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9885.965005,10187.921537,2,97,3,1,0.44,3,1,2,0,0,2,20,1,2,5,NA
+69610,7,2,1,20,NA,5,7,1,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14385.653726,15040.55165,2,101,99,2,0.46,4,1,0,0,0,1,18,NA,NA,NA,NA
+69611,7,2,2,8,NA,5,7,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9007.62445,9504.796896,2,102,15,15,5,4,4,0,2,0,1,39,1,4,1,5
+69612,7,2,2,15,NA,3,3,1,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74658.856536,76060.378937,1,100,15,15,5,5,5,0,3,0,1,47,1,5,1,5
+69613,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,11739.283384,12336.327776,1,96,15,15,5,5,5,0,0,0,1,58,2,5,1,5
+69614,7,1,1,3,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,81535.075159,0,3,91,8,8,3.4,2,2,1,0,0,2,31,1,3,5,NA
+69615,7,2,2,12,NA,1,1,1,12,150,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18515.058419,19360.671834,2,96,6,6,1.12,4,4,0,3,0,1,26,1,2,77,NA
+69616,7,2,2,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,44672.331977,48301.986589,1,98,2,2,0.36,5,5,3,0,0,1,25,1,3,1,3
+69617,7,2,2,25,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,97925.559493,100647.602701,1,91,15,6,2.69,3,1,0,0,0,1,27,1,3,6,NA
+69618,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,127865.461733,133989.246473,1,101,4,4,1.26,2,2,0,0,2,1,62,1,3,1,3
+69619,7,2,2,41,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,19075.861607,18884.31701,1,96,6,6,1.21,4,4,0,2,0,2,41,1,4,4,NA
+69620,7,2,1,49,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,31347.36219,32073.699502,2,90,6,6,1.21,4,4,0,0,0,2,59,2,1,6,NA
+69621,7,2,1,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,30631.476666,32101.68524,2,101,8,8,2.81,3,3,0,1,0,1,35,1,1,1,2
+69622,7,2,1,58,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,27140.404673,28346.53309,1,95,5,5,1.1,3,3,0,0,1,2,63,1,4,1,5
+69623,7,2,1,4,NA,1,1,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19411.993395,20798.549942,1,92,14,14,3.9,4,4,2,0,0,2,29,1,4,1,4
+69624,7,2,1,67,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,3,1,NA,1,2,1,NA,NA,NA,1,2,1,3,11145.558675,11717.869264,2,96,NA,NA,NA,4,4,0,0,1,1,67,2,3,1,2
+69625,7,2,1,29,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,111450.827181,113182.912698,3,91,15,15,5,2,2,0,0,0,1,29,1,4,1,4
+69626,7,2,1,53,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,161135.312305,165002.304429,2,101,14,14,4.86,3,3,0,1,0,1,53,1,4,1,5
+69627,7,2,2,13,NA,4,4,2,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13621.250519,13659.137754,1,93,7,7,2.72,2,2,0,1,0,2,45,1,3,3,NA
+69628,7,2,1,9,NA,1,1,1,9,116,NA,NA,2,2,3,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,14821.597351,14686.992722,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+69629,7,2,2,16,NA,3,3,1,16,197,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,39616.634313,47751.80094,2,101,3,3,0.59,4,3,0,2,0,1,39,1,1,6,NA
+69630,7,2,1,23,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,33592.259589,34050.148008,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+69631,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,1,2,1,2,2,1,2,2,NA,29183.224814,31443.714435,1,95,7,7,2.3,3,3,0,0,2,1,80,1,4,1,2
+69632,7,2,1,44,NA,2,2,2,NA,NA,1,2,2,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,31640.296506,32279.766727,2,94,8,8,2.01,4,4,1,1,0,1,44,2,4,1,4
+69633,7,2,1,38,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,42890.643983,44071.237251,1,98,15,15,5,5,5,2,1,0,1,38,1,4,1,5
+69634,7,2,2,14,NA,2,2,2,14,174,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22139.315046,23195.552704,3,91,6,6,0.83,6,6,1,3,0,1,37,1,4,1,4
+69635,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,175997.804296,180603.118894,1,101,7,7,2.31,2,2,0,0,1,1,60,1,3,1,3
+69636,7,2,1,2,NA,2,2,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10136.626471,10860.663695,2,100,14,14,3.58,4,4,1,1,0,1,33,1,4,1,5
+69637,7,2,2,58,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,189736.955264,194603.321172,1,98,6,6,2.6,1,1,0,0,0,2,58,1,2,3,NA
+69638,7,2,2,5,NA,3,3,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,53166.229434,56500.79967,1,98,7,7,1.66,5,5,2,1,0,2,37,1,5,1,3
+69639,7,1,1,48,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,20480.361635,0,2,99,6,6,2.69,1,1,0,0,0,1,48,1,3,5,NA
+69640,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,14883.664782,16969.371761,2,90,5,5,1.08,3,3,1,1,0,2,23,1,2,5,NA
+69641,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,1,2,1,2,2,NA,NA,NA,NA,109290.289961,113677.204054,1,100,15,15,5,3,3,0,0,0,1,53,1,5,1,4
+69642,7,2,1,19,NA,4,4,1,19,232,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,1,1,0.05,1,1,0,0,0,1,19,1,4,NA,NA
+69643,7,2,2,0,10,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7565.515117,8068.295221,1,100,3,3,0.43,4,4,2,0,0,1,20,1,3,6,NA
+69644,7,2,2,64,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11838.431472,12295.352662,2,94,8,8,1.8,6,6,0,1,2,1,74,2,5,1,5
+69645,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,49260.413155,50771.674072,2,100,99,99,NA,2,2,0,0,2,1,72,1,4,1,4
+69646,7,2,1,9,NA,3,3,2,10,120,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,82670.203859,86374.629723,1,97,15,15,4.07,5,5,0,3,0,1,36,1,5,1,5
+69647,7,2,1,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,113559.363135,116811.761686,2,102,14,14,3.44,5,5,1,2,0,2,34,1,4,6,NA
+69648,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,32471.790416,34987.007487,2,96,15,15,5,2,2,0,0,2,2,80,2,5,1,5
+69649,7,2,1,27,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,1,2,2,3,14385.653726,15533.829525,2,101,7,5,1.84,2,1,0,0,0,1,27,2,5,5,NA
+69650,7,2,1,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,36244.629058,36151.524069,1,98,3,2,0.72,2,1,0,0,0,2,50,1,2,3,NA
+69651,7,2,1,15,NA,4,4,1,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18120.648572,18071.489418,1,100,3,3,0.43,4,4,0,2,0,1,35,1,4,1,3
+69652,7,2,2,74,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,12833.793728,13278.897744,2,92,5,5,1.08,3,3,0,0,2,1,46,NA,NA,5,NA
+69653,7,2,1,28,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,111887.48668,1,94,9,9,3.14,3,3,1,0,0,1,28,1,5,1,5
+69654,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,23484.626749,23642.328802,2,96,14,14,4.26,3,3,0,0,0,1,20,1,4,5,NA
+69655,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,22419.63376,27869.657009,1,94,2,2,0.72,1,1,0,0,1,2,80,1,3,2,NA
+69656,7,2,1,62,NA,4,4,2,NA,NA,2,NA,2,2,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,7514.993062,8122.478925,2,90,14,14,4.59,3,3,0,0,1,2,56,2,3,1,1
+69657,7,2,2,4,NA,5,6,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6899.969666,7508.722153,1,94,10,10,3.04,4,4,2,0,0,2,30,1,4,1,5
+69658,7,2,1,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,11034.04089,11250.172797,2,95,14,14,5,2,2,0,0,2,1,73,1,5,1,4
+69659,7,2,2,12,NA,3,3,2,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,42712.452836,43268.322789,2,94,3,3,0.95,2,2,0,1,0,2,45,1,5,3,NA
+69660,7,2,1,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,11764.405491,12715.399053,1,97,NA,NA,NA,7,7,0,3,1,2,47,NA,NA,1,NA
+69661,7,2,2,57,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11446.604914,11507.133903,2,92,3,3,0.45,4,4,0,0,1,1,64,2,1,1,1
+69662,7,2,2,7,NA,1,1,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14300.71869,14671.891945,2,94,7,7,1.17,6,6,1,2,0,2,30,2,3,6,NA
+69663,7,2,1,3,NA,3,3,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78932.667512,89054.807173,3,92,14,14,2.74,6,6,2,2,0,1,35,1,5,1,4
+69664,7,2,1,3,NA,5,6,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5492.796032,5932.072263,2,100,14,14,3.06,5,5,1,0,1,2,31,2,5,1,5
+69665,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,16058.142925,15728.179634,2,101,7,7,1.3,5,5,2,0,1,2,50,1,4,1,3
+69666,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,69047.315371,69784.701831,1,95,8,8,2.7,3,3,0,1,2,1,69,1,5,1,3
+69667,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,8,8,3.4,2,2,0,0,2,1,66,1,5,1,4
+69668,7,2,1,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,120367.559207,121496.443862,2,95,15,15,5,2,2,0,0,1,1,61,1,5,1,5
+69669,7,2,2,66,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,9680.216878,10112.428208,1,96,15,15,5,1,1,0,0,1,2,66,1,5,2,NA
+69670,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60148.377616,67408.231176,1,92,6,6,1.98,2,2,0,0,2,1,80,1,5,1,4
+69671,7,2,2,58,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,1,4,NA,2,2,2,2,2,2,2,2,1,2,20130.149569,20235.126587,1,90,99,99,NA,6,6,0,3,0,2,58,2,1,4,NA
+69672,7,2,1,69,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,5201.567667,5487.342926,2,92,10,6,1.12,7,4,1,1,1,2,27,2,3,1,3
+69673,7,2,2,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6247.52442,6810.692829,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+69674,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,7736.56115,7645.782314,1,99,8,8,1.76,5,5,0,2,1,1,37,1,4,1,3
+69675,7,2,2,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,24977.658912,25631.246482,2,95,5,5,1.52,2,2,0,0,0,2,58,1,2,2,NA
+69676,7,2,1,5,NA,2,2,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16632.464801,16654.939181,1,94,2,2,0.26,4,4,2,1,0,2,25,1,4,5,NA
+69677,7,2,1,13,NA,4,4,1,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13653.432599,13680.426553,2,96,5,5,1.07,4,4,0,3,0,2,46,1,4,3,NA
+69678,7,2,2,1,18,1,1,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13464.808163,14865.509546,1,101,9,9,2.88,3,3,1,0,0,1,36,2,2,1,4
+69679,7,2,1,26,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17095.959542,2,100,8,8,3.06,2,2,0,0,0,1,26,1,5,1,5
+69680,7,2,1,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,16058.989596,16124.245459,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+69681,7,2,1,4,NA,4,4,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7299.892005,7750.518524,2,99,9,9,1.78,6,6,1,1,0,1,46,1,3,6,NA
+69682,7,2,1,67,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,2,NA,2,2,2,2,2,2,1,2,2,NA,8609.250304,11228.904188,2,90,4,4,1.02,2,2,0,0,1,1,67,2,1,2,NA
+69683,7,2,1,8,NA,4,4,2,9,108,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8017.552697,8398.795399,1,99,10,10,2.07,7,7,2,3,1,2,35,1,5,4,NA
+69684,7,2,1,65,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,1,3,NA,2,2,2,2,2,2,2,2,2,2,9404.30514,9554.950099,2,93,3,3,0.58,4,4,0,1,1,1,65,2,1,3,NA
+69685,7,2,2,17,NA,1,1,1,17,205,2,NA,2,2,3,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24654.448233,25524.546766,2,96,6,6,1.12,4,4,0,1,0,1,57,2,1,1,4
+69686,7,2,1,5,NA,1,1,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17048.276421,17071.31266,1,94,5,5,0.87,4,4,1,1,0,1,35,2,1,1,1
+69687,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,141912.982157,152670.471787,1,97,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+69688,7,2,1,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,17036.36313,17063.789746,1,99,14,14,5,1,1,0,0,0,1,42,1,4,4,NA
+69689,7,2,1,51,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15413.22404,15402.649971,1,97,15,15,5,4,4,0,0,0,1,51,2,5,1,5
+69690,7,2,2,2,NA,3,3,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,37915.354974,40293.39486,2,91,15,15,5,5,5,1,2,0,1,37,1,4,6,NA
+69691,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,24919.497762,30093.052062,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+69692,7,2,1,50,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,29903.342494,29985.789069,1,95,12,5,1.84,4,1,0,0,0,1,29,1,3,6,NA
+69693,7,2,2,37,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,43463.010321,44104.658285,1,98,6,6,1.98,2,2,0,1,0,2,37,1,4,3,NA
+69694,7,2,1,15,NA,4,4,2,15,191,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9757.309092,9852.859452,1,96,15,15,5,6,6,1,1,1,2,44,1,3,1,3
+69695,7,2,1,80,NA,4,4,2,NA,NA,2,NA,2,1,9,NA,9,3,NA,1,2,2,1,2,2,1,2,2,NA,8992.410435,9337.125921,2,90,4,4,1.54,1,1,0,0,1,1,80,2,9,3,NA
+69696,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32144.824104,32969.273278,2,97,3,3,1.1,1,1,0,0,0,2,56,1,2,3,NA
+69697,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,12,2,0.54,4,1,0,0,0,1,21,1,4,6,NA
+69698,7,2,1,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,20851.046913,21183.857295,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+69699,7,1,1,11,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10229.206765,0,1,96,14,14,2.96,5,5,0,3,0,1,46,NA,NA,1,5
+69700,7,2,2,27,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,NA,NA,NA,NA,42583.505439,50840.190326,2,91,2,2,0.42,3,3,1,1,0,2,27,1,3,5,NA
+69701,7,2,1,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17636.923031,17665.316481,2,101,6,6,2.57,1,1,0,0,0,1,52,1,3,5,NA
+69702,7,2,1,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,97539.00155,100726.587936,1,99,77,77,NA,3,3,0,0,0,1,42,1,4,6,NA
+69703,7,2,1,0,3,4,4,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7441.732415,8231.273488,2,98,3,3,0.54,3,3,1,1,0,2,29,1,2,1,NA
+69704,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,109181.566304,114466.270601,2,91,15,3,1.07,7,1,0,0,1,1,49,NA,NA,5,NA
+69705,7,2,1,42,NA,4,4,1,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17221.349323,18312.731752,1,103,8,8,1.95,4,4,0,1,0,2,48,1,5,1,5
+69706,7,2,2,9,NA,5,6,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,8290.163782,8692.019106,1,92,2,2,0.24,5,5,0,2,0,1,35,2,4,1,3
+69707,7,2,2,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12736.176535,13301.727395,1,97,15,15,5,4,4,1,1,0,1,35,1,5,1,5
+69708,7,2,1,14,NA,1,1,1,14,179,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15469.666055,15805.71937,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+69709,7,2,2,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,128065.93322,127632.692688,1,98,4,4,1.52,1,1,0,0,1,2,62,1,4,5,NA
+69710,7,2,2,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,51744.846443,52360.821424,1,97,4,3,0.93,3,2,0,0,0,1,35,1,3,1,4
+69711,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16675.763807,16530.996917,2,95,8,8,1.85,5,5,1,2,0,1,55,1,2,1,3
+69712,7,2,1,31,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,23631.037942,24059.753898,1,91,9,9,5,1,1,0,0,0,1,31,2,5,5,NA
+69713,7,2,1,17,NA,2,2,2,17,207,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24585.624844,25010.051033,3,91,6,6,0.83,6,6,1,3,0,1,37,1,4,1,4
+69714,7,2,1,18,NA,4,4,1,18,226,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,99,99,NA,4,1,0,0,0,1,18,1,4,NA,NA
+69715,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,15598.269927,16258.702668,1,99,4,4,0.41,7,7,2,4,0,2,43,1,4,4,NA
+69716,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126687.376548,128757.240271,1,101,9,9,2.6,4,4,0,1,2,2,63,1,4,1,4
+69717,7,2,1,3,NA,1,1,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19476.194601,19502.51153,2,102,15,15,5,4,4,2,0,0,1,32,1,5,1,5
+69718,7,2,2,15,NA,3,3,1,15,182,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,82923.224369,87859.866176,2,100,8,8,2.91,3,3,0,2,0,2,48,1,5,1,NA
+69719,7,2,1,3,NA,3,3,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,31190.854587,34157.026846,1,95,2,2,0.22,4,4,2,1,0,2,22,1,2,5,NA
+69720,7,2,2,3,NA,4,4,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,10901.48845,2,100,5,5,0.94,4,4,2,0,0,2,33,1,4,6,NA
+69721,7,2,2,22,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,14500.122872,13786.884381,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+69722,7,2,1,7,NA,4,4,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7758.863533,8127.805304,2,99,3,3,0.42,6,6,1,2,0,2,43,1,4,6,NA
+69723,7,2,2,9,NA,3,3,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22371.648216,22397.198208,1,92,4,4,0.61,5,5,1,2,0,1,34,1,3,6,NA
+69724,7,2,1,58,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,28585.875492,28917.983006,1,91,5,5,1.2,3,3,0,0,1,1,58,1,2,1,2
+69725,7,2,1,5,NA,3,3,1,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,73006.119819,82368.252941,1,101,14,14,3.15,5,5,2,1,0,1,35,1,4,1,5
+69726,7,1,1,53,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,30582.99679,0,1,97,4,3,1.07,2,1,0,0,1,2,66,NA,NA,5,NA
+69727,7,2,1,4,NA,5,6,1,4,55,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7042.228842,7250.600343,2,103,9,9,3.14,3,3,1,0,0,1,32,2,5,1,5
+69728,7,2,1,35,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20071.705576,20986.552878,3,91,14,14,5,2,2,0,0,0,1,35,2,5,1,5
+69729,7,2,2,3,NA,4,4,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10389.292229,11178.99433,2,101,7,7,1.3,5,5,2,0,1,2,50,1,4,1,3
+69730,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,11194.560864,11943.741966,2,95,5,5,1.88,1,1,0,0,1,2,70,1,3,3,NA
+69731,7,2,2,19,NA,4,4,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,99,1,0.02,2,1,0,0,0,2,19,1,4,NA,NA
+69732,7,2,1,33,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,5,1,NA,1,2,1,1,2,1,1,2,1,3,12327.761744,13173.339372,3,90,5,5,0.93,4,4,1,0,0,1,48,2,4,1,NA
+69733,7,1,1,67,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,7117.971973,0,2,100,3,3,0.75,2,2,0,0,2,1,67,1,3,1,2
+69734,7,2,1,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,74110.989124,87044.270957,2,95,8,8,4.48,1,1,0,0,0,1,26,1,5,5,NA
+69735,7,2,1,20,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14949.232836,15463.9021,1,100,99,99,NA,6,6,0,1,0,1,53,2,2,1,3
+69736,7,2,2,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,1,1,0.14,4,1,0,0,0,2,21,1,4,5,NA
+69737,7,2,1,31,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,58170.292683,59516.479078,3,92,8,8,2.36,3,3,0,0,0,1,34,NA,NA,5,NA
+69738,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,34802.051557,35120.431265,1,101,4,4,1.16,2,2,0,1,0,2,51,1,4,3,NA
+69739,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,17260.508485,17587.688781,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+69740,7,2,2,66,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,95538.505899,97099.44822,2,95,14,14,5,2,2,0,0,2,2,66,1,5,1,5
+69741,7,2,1,10,NA,1,1,1,10,129,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,17882.621856,17720.218058,3,92,3,3,0.51,5,5,1,2,0,2,34,2,1,6,NA
+69742,7,2,1,4,NA,5,6,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6138.820061,6885.358348,1,99,10,10,4.76,2,2,1,0,0,2,36,2,5,3,NA
+69743,7,2,1,46,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,29276.135189,1,94,7,7,1.65,5,4,0,0,0,1,46,1,4,1,4
+69744,7,1,1,12,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13416.172328,0,1,96,14,14,2.96,5,5,0,3,0,1,46,NA,NA,1,5
+69745,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,27005.036261,27122.782392,1,102,1,1,0,3,1,0,0,1,2,50,1,3,3,NA
+69746,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,115926.402585,118970.086068,1,101,15,15,5,4,4,0,2,0,1,43,1,4,1,5
+69747,7,2,2,34,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,NA,NA,NA,1,2,2,1,21229.081867,21886.919672,3,90,5,5,0.87,4,4,0,0,0,2,43,2,3,5,NA
+69748,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,162404.084268,1,95,7,7,2.86,2,2,0,0,0,2,50,1,3,1,3
+69749,7,2,1,12,NA,2,2,2,12,151,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16033.31661,17623.465787,2,90,5,5,0.76,5,5,0,4,0,2,32,1,2,3,NA
+69750,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,79677.823556,83116.431703,1,99,15,15,4.47,4,4,0,2,0,2,43,1,5,1,5
+69751,7,2,2,0,5,1,1,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6787.112205,6644.319314,2,94,8,8,2.43,3,3,1,0,0,1,24,2,4,1,4
+69752,7,2,2,79,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,81062.798322,83549.725067,1,101,4,4,1.75,1,1,0,0,1,2,79,1,5,2,NA
+69753,7,2,2,63,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14204.126838,14796.942631,1,92,5,5,1.41,2,2,0,0,2,1,60,1,2,1,4
+69754,7,2,1,80,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,10160.645851,10359.669924,1,96,77,77,NA,7,7,0,3,1,2,43,77,5,5,NA
+69755,7,2,2,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,156359.467613,158914.124584,2,102,14,14,5,2,2,0,0,2,2,65,1,5,1,5
+69756,7,2,1,2,NA,4,4,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4778.010882,5139.584556,2,93,99,99,NA,7,6,1,0,0,1,19,1,3,NA,NA
+69757,7,2,1,19,NA,5,7,2,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,10,10,3.04,4,4,0,1,0,2,43,1,5,1,4
+69758,7,2,2,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,21127.557839,23583.17383,1,102,4,4,0.97,3,3,0,1,0,2,19,1,2,NA,NA
+69759,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,32838.149884,1,95,1,1,0.18,1,1,0,0,0,1,51,1,4,4,NA
+69760,7,2,1,5,NA,4,4,2,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9446.305539,9734.833129,2,95,1,1,0,4,4,2,1,0,2,27,1,4,5,NA
+69761,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,14859.685983,15314.825218,1,95,3,3,0.96,1,1,0,0,1,2,80,1,2,2,NA
+69762,7,2,1,15,NA,1,1,1,15,190,NA,NA,2,2,3,9,NA,NA,NA,2,1,2,1,2,2,2,2,2,2,24230.15013,23997.357824,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+69763,7,2,2,6,NA,4,4,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7282.523598,8846.731146,2,99,5,5,0.76,5,5,0,2,0,1,51,1,2,1,2
+69764,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17328.248994,2,100,14,14,3.06,5,5,1,0,0,1,50,1,5,1,5
+69765,7,2,2,65,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,3,NA,2,2,2,1,2,2,1,2,1,2,11495.911371,12347.505602,2,99,3,3,0.52,3,3,0,0,1,2,38,2,3,3,NA
+69766,7,2,2,4,NA,2,2,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15979.952759,16987.381382,2,102,14,14,3.25,5,5,2,0,0,1,27,1,5,1,5
+69767,7,2,1,56,NA,5,7,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,151766.599459,154414.492891,3,91,7,7,1.97,4,4,0,0,1,2,77,1,5,2,NA
+69768,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,23041.888919,1,98,3,2,0.81,4,1,0,0,0,2,21,NA,NA,5,NA
+69769,7,2,2,0,4,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21671.775435,21083.121886,1,98,7,7,1.83,3,3,1,0,0,2,26,1,5,1,4
+69770,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,13495.651715,13078.348337,2,99,NA,77,NA,7,7,1,0,1,2,51,1,2,1,3
+69771,7,2,1,3,NA,5,6,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,10276.262805,11525.953064,1,97,15,15,5,4,4,2,0,0,2,35,2,4,1,4
+69772,7,2,2,3,NA,4,4,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11671.9972,12190.293097,1,93,4,4,1.03,3,3,1,1,0,2,35,2,3,4,NA
+69773,7,2,1,40,NA,4,4,1,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23242.990557,23331.304109,1,100,9,9,2.22,5,5,1,2,0,2,40,2,4,1,4
+69774,7,2,1,54,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,20167.650721,25203.089486,3,90,12,7,3.58,2,1,0,0,1,1,71,2,2,1,NA
+69775,7,2,1,8,NA,2,2,2,8,101,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15039.041447,15306.786696,3,91,6,6,0.83,6,6,1,3,0,1,37,1,4,1,4
+69776,7,2,1,57,NA,1,1,2,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22446.308035,22116.943066,2,94,7,7,1.33,6,6,0,1,0,1,55,2,2,1,1
+69777,7,2,2,60,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10003.290711,10218.574955,1,99,10,10,2.58,5,5,0,1,2,1,65,1,5,1,3
+69778,7,2,1,42,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,3,6,NA,2,2,2,2,2,2,2,2,2,2,31640.296506,31576.726829,2,94,9,9,4.21,3,2,0,0,0,2,48,2,2,6,NA
+69779,7,2,1,13,NA,3,3,2,13,166,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,77169.155154,76948.022242,1,98,14,14,3.9,4,4,0,3,0,2,31,1,4,1,NA
+69780,7,2,1,9,NA,3,3,2,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,66868.503099,69864.859716,1,98,14,14,3.15,5,5,0,3,0,1,34,1,4,1,4
+69781,7,2,1,0,10,3,3,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26076.0211,27915.503824,2,91,15,15,5,4,4,2,0,0,2,33,1,5,1,5
+69782,7,2,1,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,20075.522681,20698.300141,2,98,77,77,NA,1,1,0,0,0,1,52,1,3,5,NA
+69783,7,2,2,66,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13023.675419,13480.293434,2,96,2,2,0.55,1,1,0,0,1,2,66,2,5,5,NA
+69784,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11218.730451,2,101,2,2,0.64,1,1,0,0,1,1,69,1,2,5,NA
+69785,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,22419.63376,25250.873067,1,94,3,3,1.08,1,1,0,0,1,2,80,1,2,2,NA
+69786,7,2,2,1,22,3,3,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46813.021927,47428.160802,1,94,15,15,4.77,4,4,2,0,0,1,48,1,4,1,5
+69787,7,2,2,57,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,22969.116046,23893.408664,2,93,9,9,3.97,2,2,0,0,1,2,57,2,3,1,1
+69788,7,2,1,14,NA,2,2,1,14,169,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,30061.88611,30025.949584,1,92,6,6,0.93,5,5,0,2,0,1,47,2,1,1,1
+69789,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,16181.169973,15916.061734,2,100,99,99,NA,1,1,0,0,0,2,53,1,2,3,NA
+69790,7,2,1,12,NA,5,6,1,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6886.223812,7737.788899,2,92,7,7,1.61,4,4,0,2,0,1,51,2,3,1,3
+69791,7,1,2,63,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,30036.804187,0,2,100,4,4,1.29,2,2,0,0,2,1,65,1,3,1,3
+69792,7,2,1,5,NA,4,4,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11324.865632,11670.771889,2,96,3,3,0.54,4,4,2,1,0,2,25,1,4,2,NA
+69793,7,2,1,35,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20181.692021,20449.370526,1,92,2,2,0.24,5,5,0,2,0,1,35,2,4,1,3
+69794,7,2,1,9,NA,2,2,1,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8889.501355,9473.596146,2,93,14,14,3.52,5,5,1,2,0,1,44,1,5,1,5
+69795,7,1,2,80,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,NA,NA,NA,NA,16321.652472,0,2,101,1,1,0.08,2,2,0,0,2,2,80,1,1,2,NA
+69796,7,2,2,3,NA,4,4,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11990.08101,12901.460915,1,100,15,15,3.87,6,6,1,3,0,2,39,1,4,1,4
+69797,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,74517.751389,77393.175383,2,94,14,14,3.36,4,4,2,0,0,1,31,1,3,1,5
+69798,7,2,2,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,65591.951555,69705.816211,2,101,2,2,0.46,2,1,0,0,0,2,21,1,4,5,NA
+69799,7,2,2,8,NA,4,4,2,8,96,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7814.742747,8587.431439,2,95,7,7,1.55,5,5,0,3,0,1,30,1,4,1,4
+69800,7,2,1,10,NA,5,6,2,10,131,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6262.834446,6755.01452,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+69801,7,2,2,3,NA,5,6,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3864.413878,3822.111022,1,99,14,14,2.66,7,7,3,1,0,1,35,1,5,1,5
+69802,7,2,2,17,NA,3,3,1,17,207,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,36586.371708,39087.782259,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+69803,7,2,1,44,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,20543.822351,1,97,15,15,5,4,4,1,1,0,1,44,2,5,1,5
+69804,7,2,1,6,NA,3,3,1,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47373.769078,49181.31317,1,103,15,15,5,3,3,0,1,0,1,46,1,5,1,5
+69805,7,2,1,1,19,5,6,1,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6389.003009,6767.308805,3,91,14,14,3.06,5,5,3,0,0,1,34,2,5,1,5
+69806,7,2,1,75,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,8763.51401,8935.17143,2,100,6,6,1.62,3,3,0,0,2,1,75,1,5,1,NA
+69807,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,10083.559248,10533.779383,1,96,14,14,4.06,3,3,0,0,1,2,61,1,4,5,NA
+69808,7,1,1,80,NA,4,4,NA,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,1,2,1,2,2,NA,NA,NA,NA,7041.644998,0,2,100,13,13,NA,2,2,0,0,2,2,77,1,3,1,4
+69809,7,2,1,4,NA,4,4,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8973.990262,9344.796725,2,95,1,1,0.09,5,5,3,1,0,2,31,1,2,1,NA
+69810,7,2,2,14,NA,2,2,1,14,178,NA,NA,2,2,2,9,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15809.066118,16348.490248,2,93,5,5,1.26,3,3,0,1,0,1,55,2,2,1,2
+69811,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,124170.603852,124019.195449,2,98,6,6,2.75,1,1,0,0,0,1,59,1,4,4,NA
+69812,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,47636.23298,52171.746701,1,91,9,9,4.27,2,2,0,0,2,1,80,1,4,1,3
+69813,7,2,2,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,28351.482206,28369.098474,1,96,9,9,3.35,3,3,0,0,0,2,42,1,4,5,NA
+69814,7,2,2,15,NA,2,2,2,15,187,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18368.872199,19023.186366,2,91,10,10,2.95,4,4,0,1,0,2,18,1,3,NA,NA
+69815,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,15205.189409,15734.277598,2,96,7,7,2.72,2,2,0,0,1,2,70,1,4,2,NA
+69816,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,20000.263815,22894.1152,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+69817,7,2,1,33,NA,5,7,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16045.513116,17473.610556,1,90,6,6,0.92,6,6,2,0,2,2,30,2,5,1,5
+69818,7,2,2,19,NA,5,6,1,20,NA,2,NA,1,1,NA,15,NA,NA,NA,1,2,1,NA,NA,NA,1,2,2,1,9278.834813,9462.75742,1,92,2,2,0.33,5,5,0,1,0,1,51,2,1,4,NA
+69819,7,2,2,11,NA,3,3,1,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71426.628275,70878.719996,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+69820,7,2,2,6,NA,1,1,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16664.698857,2,98,6,6,0.63,7,7,2,2,1,1,60,1,3,1,2
+69821,7,2,2,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,7869.59899,8476.202883,2,100,1,1,0.14,1,1,0,0,1,2,66,1,3,2,NA
+69822,7,2,1,64,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,8636.698123,9111.200197,2,92,3,3,0.45,4,4,0,0,1,1,64,2,1,1,1
+69823,7,2,2,55,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,18441.731082,18102.807884,2,100,1,1,0,2,2,0,0,0,2,55,1,5,3,NA
+69824,7,2,1,64,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,8275.049402,8531.146449,1,99,6,6,1.07,6,6,2,1,2,1,44,2,5,4,NA
+69825,7,2,2,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,139800.409559,140918.856987,1,100,15,15,5,5,5,0,3,0,1,47,1,5,1,5
+69826,7,2,1,3,NA,3,3,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,20717.313101,23374.052614,2,97,5,5,0.8,5,5,1,2,0,1,46,2,4,1,2
+69827,7,2,1,30,NA,2,2,2,NA,NA,2,NA,2,2,4,NA,2,4,NA,2,2,2,2,2,2,1,2,1,2,27605.196104,27396.850962,2,99,12,77,NA,5,1,0,1,0,1,30,2,2,4,NA
+69828,7,2,2,18,NA,4,4,2,19,228,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,13176.946531,13078.313981,2,99,6,3,0.94,3,2,0,0,0,1,41,1,3,6,NA
+69829,7,2,1,18,NA,4,4,1,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19844.605924,19989.134915,2,102,14,14,4.05,3,3,0,1,0,1,18,1,2,NA,NA
+69830,7,2,2,4,NA,2,2,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13301.733639,13724.969642,2,93,3,3,0.48,4,4,1,1,0,1,49,2,3,1,4
+69831,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,2,NA,1,2,1,1,2,1,1,2,1,NA,13689.379977,14234.742701,2,92,2,2,0.88,1,1,0,0,1,2,80,2,1,2,NA
+69832,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,7074.645577,7351.348885,2,95,7,7,3.31,1,1,0,0,1,1,61,1,3,2,NA
+69833,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,29534.722322,30544.805853,3,91,77,77,NA,2,2,0,0,2,1,73,1,5,1,5
+69834,7,1,2,37,NA,5,6,NA,NA,NA,2,NA,2,1,5,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,13379.422066,0,2,103,14,14,3.86,4,4,2,0,0,2,37,2,5,1,NA
+69835,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,35334.703093,40990.264786,1,101,2,2,0.63,1,1,0,0,1,2,80,1,1,2,NA
+69836,7,2,1,1,21,3,3,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,35290.29852,40086.22326,1,102,14,14,2.87,5,5,3,0,0,1,35,1,5,1,5
+69837,7,2,2,9,NA,1,1,1,9,114,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15962.145468,16535.37648,2,98,3,3,0.33,7,7,2,3,0,1,40,2,1,1,1
+69838,7,2,1,31,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,30626.581617,36447.669921,2,90,6,6,1.35,3,3,1,0,0,1,31,1,3,1,4
+69839,7,2,2,8,NA,1,1,1,8,97,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,15841.451259,17367.937444,3,92,4,4,0.55,6,6,0,4,0,1,36,2,1,1,3
+69840,7,2,1,67,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,6964.172059,2,95,6,4,1.5,2,1,0,0,1,1,67,1,4,3,NA
+69841,7,2,1,16,NA,1,1,2,16,199,NA,NA,2,2,3,10,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,21633.039913,21721.097793,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+69842,7,2,1,37,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,34887.439952,35849.951626,2,94,6,6,1.34,4,4,0,2,0,1,37,2,4,1,2
+69843,7,2,1,40,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,20543.822351,1,97,15,15,5,4,4,2,0,0,1,40,2,5,1,5
+69844,7,2,1,6,NA,1,1,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11770.89642,11957.164501,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+69845,7,2,2,15,NA,3,3,2,15,186,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,29885.567338,31265.78413,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+69846,7,2,1,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,100807.076657,102674.431164,1,97,15,15,4.77,4,4,0,0,0,1,56,1,4,1,4
+69847,7,2,1,80,NA,5,6,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,14603.225127,15344.674112,1,98,15,15,5,2,2,0,0,2,1,80,2,5,1,NA
+69848,7,2,1,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,15417.485683,15656.172471,1,92,3,3,0.98,2,2,0,0,1,1,70,1,2,1,2
+69849,7,2,1,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,9257.537917,9255.717695,1,99,7,6,1.84,3,2,0,0,2,1,70,1,2,1,4
+69850,7,2,2,10,NA,1,1,2,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,15225.935813,16547.241996,2,94,7,7,1.57,4,4,0,2,0,1,30,2,3,1,4
+69851,7,2,2,31,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,18901.436377,19544.076782,3,91,14,14,5,2,2,0,0,0,1,35,2,5,1,5
+69852,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,35434.580514,39416.921733,1,95,4,4,1.47,1,1,0,0,1,2,80,1,3,1,NA
+69853,7,2,1,10,NA,4,4,2,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8227.856305,9153.104022,3,91,1,1,0.07,6,6,2,3,0,2,30,1,2,3,NA
+69854,7,2,1,65,NA,5,7,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,168185.448935,166468.676858,1,95,10,10,4.76,2,2,0,0,2,1,65,1,4,1,4
+69855,7,2,2,41,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,11762.034222,12708.069605,3,90,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+69856,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,78240.016337,83257.203545,1,93,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+69857,7,2,1,16,NA,5,7,2,16,203,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21852.102821,22875.024578,3,91,5,5,0.65,7,7,0,4,0,2,39,1,3,4,NA
+69858,7,2,1,65,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9655.271429,9541.978936,1,95,9,9,4.21,2,2,0,0,1,1,65,1,5,1,4
+69859,7,2,2,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9503.429019,9925.429486,1,91,3,3,0.66,4,4,1,2,0,2,33,1,3,5,NA
+69860,7,2,2,18,NA,4,4,1,18,222,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11224.041366,11682.365019,1,96,5,5,0.53,7,7,2,2,0,2,38,1,9,6,NA
+69861,7,2,2,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,15852.523312,16085.826144,2,97,6,6,2.04,2,2,0,0,2,2,80,1,3,2,NA
+69862,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,75772.894935,76005.871473,1,94,15,15,5,2,2,0,0,0,1,29,1,4,1,5
+69863,7,2,1,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,28337.446334,29138.677291,2,95,14,14,3.34,4,4,0,0,0,1,43,1,3,1,3
+69864,7,2,1,52,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,25583.266805,26390.249407,1,100,12,99,NA,2,1,0,0,0,2,50,1,3,6,NA
+69865,7,2,2,4,NA,3,3,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,28877.220658,31871.522389,1,94,6,6,1.21,4,4,2,0,0,1,27,1,2,1,2
+69866,7,2,1,63,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,1,1,NA,1,2,1,1,2,1,1,2,1,3,12579.986433,13337.268472,1,92,1,1,0.26,2,2,0,0,2,1,63,2,1,1,1
+69867,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17521.481386,17363.561951,2,97,15,12,NA,2,1,0,0,0,1,54,1,4,5,NA
+69868,7,2,1,40,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,6,NA,2,2,2,2,2,2,1,2,2,2,31740.385214,32919.784432,2,96,5,5,0.89,4,4,1,1,0,2,36,2,4,6,NA
+69869,7,2,2,36,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,1,6,2,2,2,2,1,2,2,NA,NA,NA,NA,38218.668882,37878.487888,2,102,5,5,0.59,7,7,1,3,0,1,37,2,1,6,NA
+69870,7,2,1,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,131818.641085,133509.443669,1,102,15,15,5,2,2,0,0,2,2,68,1,4,1,5
+69871,7,2,2,1,22,1,1,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12602.647442,12772.138116,1,91,99,99,NA,4,4,2,1,0,2,36,2,2,4,NA
+69872,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,105412.227726,110446.221907,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+69873,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,20135.920214,20700.031916,2,97,2,2,0.27,3,3,1,0,0,2,21,1,3,6,NA
+69874,7,2,2,3,NA,3,3,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78448.332626,86582.701959,2,91,15,15,5,4,4,2,0,0,2,33,1,5,1,5
+69875,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,1,1,2,1,1,2,1,3,11762.034222,11824.231183,3,90,15,15,3.23,6,6,0,2,0,1,50,2,2,1,2
+69876,7,2,2,36,NA,1,1,2,NA,NA,2,NA,2,1,3,NA,3,1,2,2,2,2,2,2,2,2,2,2,2,35353.005268,34399.106917,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+69877,7,2,2,38,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,NA,NA,NA,NA,19257.047323,21540.196698,1,102,5,5,1.27,3,3,0,2,0,2,38,1,2,3,NA
+69878,7,2,2,15,NA,5,7,1,15,189,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15844.005605,16354.571478,1,100,7,7,1.83,3,3,0,1,0,2,40,1,4,6,NA
+69879,7,2,1,22,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,104488.914565,106745.836574,1,98,4,2,0.56,4,1,0,0,0,1,22,1,5,5,NA
+69880,7,2,2,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,102720.446375,103036.277703,2,102,14,14,3.44,5,5,1,2,0,2,34,1,4,6,NA
+69881,7,2,1,12,NA,2,2,1,12,155,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18242.832494,18053.229058,1,102,77,77,NA,6,6,0,2,1,2,37,1,4,1,4
+69882,7,2,2,14,NA,4,4,1,14,179,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11791.755593,11850.248564,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+69883,7,2,2,14,NA,1,1,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,23979.296993,29080.570096,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+69884,7,2,1,56,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18916.732604,18903.754992,2,91,8,8,2.34,4,4,0,2,0,1,56,2,5,1,5
+69885,7,2,2,42,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,19075.861607,19022.257361,1,96,9,9,2.78,4,4,0,2,0,1,54,2,5,4,NA
+69886,7,2,2,5,NA,2,2,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,15979.952759,16987.381382,2,102,4,4,0.57,6,6,2,3,0,2,26,2,3,1,NA
+69887,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,17260.508485,16474.419641,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+69888,7,2,1,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,19940.089683,1,94,3,3,0.8,2,2,0,0,1,2,66,1,4,3,NA
+69889,7,2,1,13,NA,4,4,2,13,163,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13975.179508,14076.961252,1,96,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+69890,7,2,1,5,NA,3,3,1,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,76114.759421,89186.300704,1,98,15,15,4.34,4,4,1,1,0,1,41,1,5,1,5
+69891,7,2,1,79,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,8992.410435,11769.853451,2,90,10,10,4.76,2,2,0,0,1,1,79,1,2,2,NA
+69892,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,8521.670488,9178.537312,2,95,4,4,1.34,1,1,0,0,1,2,63,1,3,2,NA
+69893,7,2,1,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,32866.0119,33575.905633,1,98,12,4,1.52,3,1,0,0,0,2,22,NA,NA,5,NA
+69894,7,2,2,3,NA,3,3,2,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27257.164734,28966.726071,1,95,4,4,0.65,6,6,2,2,0,2,36,1,4,6,NA
+69895,7,1,1,17,NA,3,3,NA,NA,NA,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,68701.580401,0,2,94,14,14,2.87,5,5,2,1,0,1,37,1,3,1,4
+69896,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,146181.198007,148606.927767,2,91,15,6,2.86,7,1,0,0,1,1,49,NA,NA,5,NA
+69897,7,2,2,1,16,1,1,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,15816.39252,3,92,7,7,2.1,3,3,1,1,0,2,25,1,4,5,NA
+69898,7,2,1,71,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,19473.412374,20212.213332,1,95,6,6,2.04,2,2,0,0,2,1,71,1,3,1,4
+69899,7,2,2,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,30664.033268,30444.919975,2,102,5,5,0.67,6,6,0,4,0,2,33,1,2,6,NA
+69900,7,2,2,40,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,1,2,1,2,2,1,2,2,1,2,2,NA,31334.47528,33406.965231,2,96,7,7,1.79,4,4,0,2,0,1,43,2,3,1,2
+69901,7,2,1,17,NA,1,1,1,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20638.769105,20614.097145,2,92,15,15,3.37,7,7,0,4,0,1,42,2,3,1,1
+69902,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,19130.246369,18964.171442,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+69903,7,2,1,3,NA,1,1,1,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,2,NA,NA,NA,NA,17865.135763,18076.339981,3,92,7,7,1.41,5,5,1,2,0,1,20,2,1,1,1
+69904,7,2,2,16,NA,2,2,2,16,198,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,7,7,1.66,4,4,0,3,0,2,34,1,5,3,NA
+69905,7,2,1,65,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,1,1,2,1,1,2,1,3,9596.331248,10123.555774,2,92,8,8,2.43,3,3,0,1,1,2,58,NA,5,1,5
+69906,7,2,1,42,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,41410.39303,40802.759347,1,101,12,3,1.07,3,1,0,0,0,1,41,2,1,4,NA
+69907,7,2,1,8,NA,1,1,2,8,100,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,11417.89405,11689.28796,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+69908,7,1,2,0,9,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4090.43871,0,2,99,6,6,1.73,3,3,1,1,1,2,60,1,4,3,NA
+69909,7,2,2,15,NA,5,6,1,15,185,NA,NA,2,1,2,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11626.310625,12128.832209,1,92,7,7,1.56,4,4,0,2,0,2,38,2,4,6,NA
+69910,7,2,1,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,102928.893739,104528.537732,1,101,6,6,0.97,7,7,2,1,0,1,43,1,2,1,NA
+69911,7,2,2,80,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,2,NA,1,1,2,2,2,2,NA,NA,NA,NA,16490.79781,19016.592593,3,90,8,8,3.21,2,2,0,0,2,2,80,2,3,2,NA
+69912,7,2,2,8,NA,4,4,2,8,103,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6388.247052,6617.403229,1,93,6,6,0.83,6,6,3,1,0,1,37,NA,NA,1,3
+69913,7,2,2,64,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,14102.354333,16246.228018,3,91,6,6,1.12,4,4,0,0,2,1,69,2,3,1,1
+69914,7,2,1,11,NA,3,3,2,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,46081.129115,51016.6207,2,95,9,9,2.6,4,4,0,2,0,1,42,1,4,1,4
+69915,7,2,1,6,NA,4,4,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,7970.311962,8205.827749,1,90,14,14,2.96,5,5,1,2,0,1,31,1,5,1,4
+69916,7,2,2,22,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,2,2,2,1,2,2,1,35710.33222,35554.798024,1,90,10,10,2.44,5,5,1,0,0,2,56,2,1,1,1
+69917,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,72686.111056,73189.163544,1,98,15,15,3.7,5,5,2,1,0,1,34,1,5,1,5
+69918,7,2,2,11,NA,4,4,2,11,140,NA,NA,2,2,3,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7659.302568,7901.193624,1,96,3,3,0.43,4,4,1,1,0,2,39,2,4,1,3
+69919,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,83819.702285,99717.125794,1,99,12,12,NA,1,1,0,0,0,1,31,1,4,5,NA
+69920,7,2,2,9,NA,5,7,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22308.590534,22137.463018,1,101,5,5,1.15,3,3,0,1,0,1,49,1,3,1,4
+69921,7,2,1,75,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,10980.245706,10978.086763,1,91,12,6,2.39,2,1,0,0,2,2,73,NA,NA,77,NA
+69922,7,2,1,42,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,31640.296506,33247.715312,3,92,6,6,1,6,6,1,1,0,1,42,2,1,1,4
+69923,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,169546.363168,172535.263747,2,91,15,15,5,2,2,0,0,1,2,60,1,5,1,5
+69924,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,27426.222967,30692.896319,2,95,3,3,0.95,2,2,0,0,2,2,80,1,1,1,2
+69925,7,2,2,80,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,NA,17563.671235,18809.520935,1,99,3,3,1.3,1,1,0,0,1,2,80,1,1,5,NA
+69926,7,2,2,11,NA,2,2,1,11,132,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16699.231378,17036.994683,1,98,3,3,0.4,7,7,2,3,0,2,31,2,5,1,2
+69927,7,2,2,2,NA,2,2,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8645.395449,9311.129497,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+69928,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,9015.10987,2,95,1,1,0.36,1,1,0,0,1,2,63,1,1,2,NA
+69929,7,2,1,6,NA,5,6,2,6,83,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11461.625841,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+69930,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,112992.533921,115858.21849,1,94,9,9,3.14,3,3,1,0,0,1,28,1,5,1,5
+69931,7,2,2,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,129336.409693,136474.939567,1,95,15,15,5,3,3,1,0,0,1,28,1,5,1,5
+69932,7,2,2,21,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,38177.662675,41078.639454,1,103,3,3,0.37,5,5,1,2,0,2,30,1,4,5,NA
+69933,7,2,2,6,NA,3,3,1,6,79,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20531.504977,20629.491172,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+69934,7,2,2,56,NA,3,3,1,NA,NA,2,NA,2,2,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,153680.330235,152391.721814,1,103,15,15,5,2,2,0,0,1,1,61,1,4,1,5
+69935,7,2,1,80,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,8543.37447,9341.78969,3,90,9,9,3.65,2,2,0,0,2,1,80,1,2,1,3
+69936,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15072.065618,15224.718219,1,99,15,15,5,2,2,0,0,1,2,54,2,5,1,5
+69937,7,2,1,19,NA,3,3,1,19,233,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,23403.89977,23518.866268,1,98,4,4,0.75,4,4,0,1,0,2,48,1,2,1,3
+69938,7,2,2,16,NA,4,4,1,16,203,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11791.755593,11850.248564,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+69939,7,2,2,0,5,2,2,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7192.863376,7041.533956,1,90,3,3,0.52,3,3,2,0,0,2,20,1,4,5,NA
+69940,7,2,1,61,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,3,1,NA,1,2,2,1,2,2,1,2,2,2,6638.80908,6954.569749,2,93,15,15,5,2,2,0,0,2,1,61,2,3,1,3
+69941,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,12098.219371,13033.848361,2,100,14,14,3.06,5,5,1,0,1,2,31,2,5,1,5
+69942,7,1,1,61,NA,2,2,NA,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,9691.985299,0,1,93,6,6,1.55,3,3,0,0,3,1,61,2,4,1,1
+69943,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,109679.354704,1,98,15,15,5,4,4,2,0,0,1,40,1,5,1,NA
+69944,7,2,2,18,NA,4,4,2,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,NA,11072.776368,11285.58755,3,90,5,5,0.87,4,4,0,0,0,2,43,2,3,5,NA
+69945,7,2,1,2,NA,5,7,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8100.706553,8340.3972,1,98,15,15,5,4,4,2,0,0,2,35,1,5,1,4
+69946,7,2,1,6,NA,3,3,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,42986.51011,44926.997612,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+69947,7,2,1,7,NA,2,2,2,7,88,NA,NA,2,2,1,0,NA,NA,NA,2,1,2,NA,NA,NA,NA,NA,NA,NA,8966.477743,10563.593323,2,99,12,3,0.52,5,3,0,1,0,1,30,2,2,4,NA
+69948,7,2,1,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,27739.528889,28855.317795,1,90,4,3,1.07,2,1,0,0,0,2,25,1,5,6,NA
+69949,7,2,2,12,NA,3,3,2,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38400.791741,41869.717003,1,95,6,6,1.09,5,5,0,3,0,1,31,1,4,1,4
+69950,7,2,2,10,NA,4,4,2,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8115.309776,8285.322178,2,95,7,7,2.78,2,2,0,1,0,2,32,1,4,5,NA
+69951,7,2,1,4,NA,5,6,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6185.185728,6767.562311,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+69952,7,2,1,4,NA,3,3,2,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,77702.196479,91046.355703,1,95,15,15,5,3,3,1,0,0,1,28,1,5,1,5
+69953,7,2,1,8,NA,2,2,2,8,106,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13804.767816,14116.094161,2,91,9,9,2.6,4,4,1,1,0,2,31,2,4,1,5
+69954,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,1,1,2,2,1,2,2,NA,NA,NA,NA,15309.176013,16540.512525,1,90,14,14,3.33,5,5,1,2,0,1,41,1,5,1,5
+69955,7,2,1,10,NA,5,7,2,10,121,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7482.593572,7851.213843,1,101,4,4,0.78,4,4,1,2,0,2,31,1,4,3,NA
+69956,7,2,1,3,NA,3,3,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,33334.752566,38103.457131,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+69957,7,2,1,32,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,37658.482129,38663.56389,1,100,10,10,2.91,4,4,1,1,0,1,32,1,5,1,5
+69958,7,2,1,12,NA,3,3,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71231.747774,75661.062324,2,95,9,9,2.6,4,4,0,2,0,1,42,1,4,1,4
+69959,7,2,2,51,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,2,4,NA,2,2,2,2,2,2,2,2,2,2,29695.385784,34065.080879,2,91,1,1,0,1,1,0,0,0,2,51,2,2,4,NA
+69960,7,1,2,14,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,116754.883807,0,1,94,5,5,1.04,4,4,1,1,0,1,18,1,2,NA,NA
+69961,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,204553.029374,204712.584881,1,100,15,15,5,2,2,0,0,0,2,56,1,5,1,NA
+69962,7,2,1,69,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11992.012141,12562.386395,2,102,4,4,1.09,2,2,0,0,2,2,68,1,3,1,3
+69963,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,29672.504425,2,101,12,4,1.79,4,1,0,0,0,2,20,1,4,5,NA
+69964,7,2,2,75,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,70652.532816,72526.815349,1,98,6,6,1.65,2,2,0,0,2,1,80,1,3,1,3
+69965,7,2,2,27,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16181.486766,17520.496205,1,90,4,4,0.78,4,4,0,0,1,1,69,2,4,1,3
+69966,7,2,1,3,NA,4,4,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13345.162299,13752.776418,2,102,4,4,0.81,3,3,2,0,0,2,23,1,4,5,NA
+69967,7,2,1,8,NA,3,3,1,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28126.350642,29877.890829,2,91,5,5,1.2,3,3,0,1,0,2,40,1,5,1,5
+69968,7,2,1,16,NA,3,3,1,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,113571.164423,112715.649169,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+69969,7,2,1,11,NA,1,1,1,11,138,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14007.413517,2,98,1,1,0.18,4,4,0,2,0,1,29,1,4,6,NA
+69970,7,2,2,13,NA,5,6,2,13,162,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9029.632215,9422.555256,1,90,77,77,NA,4,4,0,2,0,2,51,1,5,1,5
+69971,7,2,2,2,NA,3,3,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22159.470641,22450.653549,1,95,3,3,0.7,3,3,1,0,0,1,25,1,4,1,4
+69972,7,2,1,79,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,53149.251154,56449.488341,1,100,15,15,5,2,2,0,0,2,1,79,1,5,1,4
+69973,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,26637.81974,28307.564263,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+69974,7,2,1,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,15408.94893,15514.368855,1,99,13,13,NA,3,3,0,0,2,1,67,1,2,1,2
+69975,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,60071.993203,65791.533402,1,95,14,14,5,1,1,0,0,1,2,80,1,4,1,NA
+69976,7,2,1,61,NA,2,2,2,NA,NA,2,NA,2,1,99,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,10585.549682,10952.862023,1,90,12,12,NA,3,3,0,0,1,1,35,2,4,5,NA
+69977,7,2,2,12,NA,2,2,2,12,154,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,13824.001771,15649.805677,2,90,3,3,0.54,4,4,1,2,0,2,33,2,1,4,NA
+69978,7,2,2,13,NA,4,4,1,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18698.220599,19300.762251,2,101,6,6,1.54,3,3,0,1,0,2,34,1,4,1,3
+69979,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,NA,NA,NA,NA,26465.930618,28724.649216,2,100,5,5,1.07,4,4,0,1,0,2,36,1,3,5,NA
+69980,7,2,2,0,2,2,2,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4481.392842,4662.235074,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+69981,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15991.70237,16076.675652,1,99,3,3,1.29,1,1,0,0,1,1,62,1,3,5,NA
+69982,7,2,2,8,NA,3,3,2,8,98,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,41790.228676,44382.642913,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+69983,7,2,1,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,19940.089683,1,94,3,3,0.95,2,2,0,0,1,2,67,1,2,2,NA
+69984,7,2,2,16,NA,1,1,1,16,200,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27070.679378,29958.836982,1,92,4,4,0.74,4,4,1,1,0,1,51,2,1,1,1
+69985,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,19901.857177,20582.498016,3,92,3,3,1.29,1,1,0,0,1,2,73,1,4,2,NA
+69986,7,2,1,14,NA,1,1,1,14,170,NA,NA,2,2,4,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,25525.43565,25915.600865,1,94,5,5,0.87,4,4,1,1,0,1,35,2,1,1,1
+69987,7,2,2,3,NA,1,1,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,17065.756351,3,92,3,3,0.54,4,4,3,0,0,2,22,1,3,5,NA
+69988,7,2,1,14,NA,1,1,1,14,175,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18242.832494,19431.180704,1,102,7,7,1.41,5,5,0,2,2,1,72,1,4,1,3
+69989,7,2,2,75,NA,2,2,2,NA,NA,2,NA,2,1,99,NA,1,6,NA,2,2,2,2,2,2,1,2,2,NA,17318.187297,19970.708273,2,90,4,4,1.02,2,2,0,0,2,2,75,2,1,6,NA
+69990,7,2,2,23,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,18412.14266,22203.276692,2,90,2,2,0.67,2,2,0,0,1,2,64,1,5,5,NA
+69991,7,2,2,56,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,27407.957254,27308.998449,2,90,3,3,1.29,1,1,0,0,0,2,56,1,3,3,NA
+69992,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,150082.940829,151424.784638,1,98,8,8,2.97,2,2,0,0,0,1,23,1,3,1,5
+69993,7,2,2,40,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,29650.79971,37801.89955,2,90,4,4,1.29,2,2,0,0,0,2,40,1,2,5,NA
+69994,7,2,2,21,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,71351.478679,74215.528244,2,100,10,10,3.13,4,4,0,0,1,2,53,1,2,1,2
+69995,7,2,1,18,NA,1,1,2,18,226,2,NA,1,1,NA,11,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,18120.499457,19917.650851,1,90,1,1,0.02,5,5,0,1,0,2,39,2,1,1,2
+69996,7,2,2,40,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,18490.479848,21651.629766,2,100,6,6,0.99,5,5,0,3,0,2,40,1,3,1,3
+69997,7,2,1,46,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,25123.480232,25043.348036,1,92,15,15,4.44,5,5,0,3,0,2,43,1,5,6,NA
+69998,7,1,1,17,NA,4,4,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,11125.932433,0,1,96,NA,NA,NA,4,4,1,1,0,2,37,NA,NA,1,4
+69999,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,20247.768461,20264.515513,2,93,5,5,1.05,3,3,1,0,0,2,29,1,3,5,NA
+70000,7,2,2,59,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,19100.364335,18787.428747,1,96,4,4,1.34,1,1,0,0,0,2,59,2,3,3,NA
+70001,7,1,2,61,NA,5,6,NA,NA,NA,2,NA,2,1,4,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17243.546687,0,1,95,3,3,0.43,4,4,0,1,2,1,65,2,5,1,3
+70002,7,2,2,9,NA,2,2,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,19084.197249,19808.05931,2,91,15,15,5,4,4,0,2,0,1,45,2,5,1,4
+70003,7,2,1,3,NA,5,6,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5953.107662,6677.061573,3,90,14,14,3.47,4,4,1,1,0,2,38,2,5,1,5
+70004,7,2,1,12,NA,3,3,2,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,99249.131685,98501.502244,1,101,15,15,5,4,4,0,2,0,1,43,1,4,1,5
+70005,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,1,2,1,2,2,NA,NA,NA,NA,33146.291352,36370.393263,1,90,5,5,1.05,3,3,0,0,3,2,60,1,5,77,NA
+70006,7,2,1,29,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,5,5,NA,1,2,2,NA,NA,NA,1,2,2,1,38474.772527,40200.135096,2,93,NA,NA,NA,4,2,0,0,0,1,28,NA,NA,4,NA
+70007,7,2,2,4,NA,4,4,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10389.292229,10850.629517,2,95,1,1,0.25,3,3,1,1,0,2,26,1,2,5,NA
+70008,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,165029.101567,1,95,8,6,2.04,4,2,0,1,0,2,57,1,5,5,NA
+70009,7,2,1,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,74929.366953,78437.058525,1,93,15,9,5,3,1,0,0,0,1,26,1,5,5,NA
+70010,7,2,2,58,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12441.719186,13672.655562,1,96,7,7,1.83,3,3,0,0,1,1,66,2,5,1,3
+70011,7,2,1,14,NA,5,6,2,14,170,NA,NA,2,2,3,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8168.705487,9098.657657,1,93,7,7,1.64,5,5,0,2,0,1,47,2,5,1,1
+70012,7,2,1,0,5,4,4,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5719.440362,6091.410026,1,96,15,15,5,6,6,1,1,1,2,44,1,3,1,3
+70013,7,2,1,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,7526.944058,7674.379869,1,96,15,15,5,2,2,0,0,2,1,70,1,5,1,4
+70014,7,2,1,32,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,85120.619542,87558.517853,1,93,15,15,5,1,1,0,0,0,1,32,1,4,5,NA
+70015,7,2,1,57,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,16851.334496,19427.051933,2,95,4,4,1.66,1,1,0,0,0,1,57,1,4,3,NA
+70016,7,2,2,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,94,15,15,5,2,2,0,0,2,1,73,NA,NA,1,5
+70017,7,2,2,14,NA,4,4,2,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12209.74498,13384.042162,2,90,4,4,0.57,5,5,1,2,0,2,33,2,2,77,NA
+70018,7,2,1,28,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27911.790319,31609.08818,3,91,7,7,1.57,4,4,2,0,0,2,29,2,3,1,3
+70019,7,2,1,38,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18544.003944,18279.674065,2,100,5,5,1.08,3,3,0,0,0,1,38,1,2,5,NA
+70020,7,2,2,21,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,4,6,3,1,2,2,1,2,2,1,2,2,1,16929.836231,18556.498474,2,101,8,5,2.2,2,1,0,0,0,1,24,2,4,6,NA
+70021,7,2,2,10,NA,3,3,1,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,45129.675368,44994.093254,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+70022,7,2,2,2,NA,5,6,2,3,36,NA,NA,2,2,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8313.181418,1,97,15,15,4.34,4,4,2,0,0,1,35,2,5,1,5
+70023,7,2,2,9,NA,4,4,2,10,120,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7659.302568,7901.193624,1,96,77,77,NA,7,7,0,3,1,2,43,77,5,5,NA
+70024,7,2,2,9,NA,4,4,2,9,109,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7268.721126,7826.51541,2,99,4,4,0.41,7,7,0,2,0,2,36,1,3,5,NA
+70025,7,2,2,16,NA,2,2,2,16,196,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,7,7,1.66,4,4,0,3,0,2,34,1,5,3,NA
+70026,7,2,1,31,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17554.053699,17463.765652,2,100,7,7,2.72,2,2,0,0,0,2,59,1,4,3,NA
+70027,7,2,1,7,NA,1,1,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10658.399025,10722.994279,1,102,8,8,1.33,7,7,1,4,0,2,32,1,3,1,2
+70028,7,2,2,2,NA,3,3,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22159.470641,22857.583467,1,95,2,2,0.22,4,4,2,1,0,2,22,1,2,5,NA
+70029,7,2,1,16,NA,4,4,1,16,198,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,18413.211403,2,101,99,99,NA,3,3,0,1,1,2,78,1,1,2,NA
+70030,7,1,2,61,NA,5,7,NA,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,10346.035773,0,1,99,9,1,0,2,1,0,0,2,2,61,1,4,6,NA
+70031,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10192.188896,10440.656902,1,99,7,7,3.31,1,1,0,0,1,2,63,1,4,3,NA
+70032,7,2,1,80,NA,5,6,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,11550.158096,12419.035396,2,92,77,77,NA,2,2,0,0,2,2,80,1,3,1,4
+70033,7,2,2,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,160743.928829,166234.629208,1,95,12,12,NA,2,2,0,0,0,2,53,1,4,1,NA
+70034,7,2,2,4,NA,4,4,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9689.370244,10119.627445,1,99,2,2,0.43,3,3,2,0,0,2,26,1,4,5,NA
+70035,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,91836.529686,93336.987794,1,93,10,10,5,1,1,0,0,1,2,65,1,5,3,NA
+70036,7,2,2,7,NA,2,2,1,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16765.041162,17104.135554,1,98,4,4,0.94,3,3,0,1,0,2,35,2,5,1,5
+70037,7,2,2,1,23,2,2,2,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8563.768225,8593.010585,1,93,6,6,0.74,7,7,1,2,0,1,53,2,2,1,2
+70038,7,1,1,20,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,44074.735764,0,2,91,4,4,0.69,4,4,2,0,0,2,21,1,3,6,NA
+70039,7,2,2,9,NA,3,3,2,9,109,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,NA,NA,NA,1,2,2,1,22933.149195,26924.921202,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+70040,7,2,1,3,NA,5,7,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12996.965152,13393.943951,1,101,5,5,1.23,3,3,2,0,0,2,24,1,2,5,NA
+70041,7,1,2,80,NA,5,6,NA,NA,NA,2,NA,2,1,7,NA,1,2,NA,1,2,1,1,2,1,NA,NA,NA,NA,13689.379977,0,2,92,2,2,0.64,1,1,0,0,1,2,80,2,1,2,NA
+70042,7,2,2,2,NA,3,3,2,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16635.553691,18360.507343,1,91,4,4,0.81,4,4,1,1,0,1,32,1,4,6,NA
+70043,7,2,2,12,NA,3,3,2,12,150,NA,NA,2,1,4,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,74165.041171,75130.242541,2,91,15,15,5,4,4,0,2,0,2,48,1,5,1,5
+70044,7,2,1,9,NA,4,4,2,9,117,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9023.469661,9333.449986,2,90,8,6,1.46,4,3,1,1,0,2,21,1,5,6,NA
+70045,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,70772.230318,73438.772102,1,93,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+70046,7,2,1,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,19541.667675,19517.839388,2,95,3,3,1.16,1,1,0,0,0,1,55,1,4,4,NA
+70047,7,2,2,19,NA,4,4,1,19,233,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11838.873374,11750.256617,2,100,14,14,3.06,5,5,1,0,0,1,50,1,5,1,5
+70048,7,2,2,11,NA,4,4,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7891.243393,8330.261165,2,100,4,4,0.69,5,5,0,3,0,1,38,1,3,6,NA
+70049,7,2,1,41,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,1,2,1,1,2,2,1,2,2,NA,20696.713928,21871.839567,2,102,9,9,2.68,4,4,1,1,0,2,38,2,5,1,2
+70050,7,2,2,49,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,29010.447112,29720.490657,1,100,9,9,3.64,2,2,0,0,0,2,49,1,4,5,NA
+70051,7,2,1,10,NA,4,4,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11185.87189,11566.898318,1,92,7,7,2.25,3,3,0,2,0,2,35,1,4,77,NA
+70052,7,2,2,28,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,11696.173591,12195.620792,1,93,15,5,1.84,6,1,0,0,0,1,34,2,5,5,NA
+70053,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,32866.0119,33575.905633,1,98,99,99,NA,3,2,0,0,0,2,22,1,4,5,NA
+70054,7,2,1,38,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,6,NA,2,2,2,1,2,2,NA,NA,NA,NA,32856.012738,34774.888791,2,103,77,77,NA,7,7,0,4,0,1,38,2,1,6,NA
+70055,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,115926.402585,120043.284447,1,98,10,10,3.04,4,4,0,2,0,2,47,1,4,1,3
+70056,7,2,1,3,NA,5,7,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11821.601823,13035.249088,1,95,4,4,0.97,3,3,2,0,0,2,22,1,4,5,NA
+70057,7,2,2,7,NA,1,1,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15352.601806,16028.326912,2,102,4,4,0.61,5,5,0,3,0,1,34,2,3,1,3
+70058,7,2,1,38,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20071.705576,20986.552878,3,91,7,7,2.45,2,2,0,0,0,2,29,2,5,1,5
+70059,7,2,2,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,25812.913537,27633.237354,2,95,3,3,1.27,1,1,0,0,1,2,74,1,2,2,NA
+70060,7,2,1,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5447.377416,5513.820992,2,90,4,4,0.97,3,3,1,0,0,2,23,2,3,5,NA
+70061,7,2,1,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,7101.739553,7433.956549,2,100,2,2,0.83,1,1,0,0,1,1,65,1,2,5,NA
+70062,7,2,2,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,69726.261922,71865.394906,1,93,15,15,5,2,2,0,0,2,2,75,1,5,1,5
+70063,7,2,2,43,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,1,2,1,2,2,1,2,2,1,2,2,1,27585.470618,28524.896774,2,102,15,15,5,4,4,0,2,0,1,44,1,3,1,1
+70064,7,2,2,38,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,2,1,2,2,1,2,2,1,2,2,1,23725.035562,24103.855713,1,91,7,7,2.2,3,3,0,0,1,2,60,1,2,2,NA
+70065,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,74518.988514,1,98,14,14,3.9,4,4,0,3,0,2,31,1,4,1,NA
+70066,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,19260.892847,19199.459573,2,97,8,5,2.02,2,1,0,0,0,1,47,1,4,3,NA
+70067,7,2,1,5,NA,3,3,1,5,64,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21357.821814,24096.698658,2,96,3,3,0.53,5,5,3,0,0,2,26,1,4,1,4
+70068,7,2,2,8,NA,4,4,2,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8147.287486,8590.325322,2,90,2,2,0.38,4,4,1,2,0,2,32,1,4,5,NA
+70069,7,2,2,50,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,4,6,NA,2,2,2,1,2,2,2,2,2,2,24352.519425,24479.515743,1,97,3,3,0.5,5,5,0,2,0,1,56,2,2,6,NA
+70070,7,2,2,2,NA,4,4,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9273.23044,9512.588252,2,96,NA,3,0.65,4,3,1,1,0,1,21,1,2,6,NA
+70071,7,2,1,0,0,1,1,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,7757.493251,7980.837709,3,92,4,4,0.59,5,5,2,1,0,1,20,2,1,1,3
+70072,7,2,1,53,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,26659.265986,2,101,7,7,2.16,3,3,0,1,0,2,44,1,4,6,NA
+70073,7,2,1,19,NA,3,3,2,19,239,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30943.024697,36334.060093,1,101,1,1,0.08,6,6,0,1,0,1,51,1,2,5,NA
+70074,7,2,1,16,NA,1,1,1,16,195,NA,NA,1,1,NA,9,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,19996.544021,19901.195571,1,100,7,7,1.3,5,5,0,3,0,1,43,2,2,1,4
+70075,7,2,2,31,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,1,2,1,2,1,1,2,1,NA,NA,NA,NA,11608.998717,13948.968232,3,90,5,5,0.93,4,4,1,0,0,1,48,2,4,1,NA
+70076,7,2,1,14,NA,5,6,2,14,172,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6232.587755,6661.015771,1,91,15,15,3.25,7,7,1,2,0,2,31,1,5,1,5
+70077,7,2,2,58,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,16119.136275,16768.999275,1,101,14,14,5,2,2,0,0,0,1,58,2,3,1,1
+70078,7,2,2,33,NA,1,1,2,NA,NA,2,NA,2,2,2,NA,2,1,2,2,2,2,2,2,2,2,2,2,2,32982.479382,35082.144563,2,99,99,3,0.66,4,2,0,0,0,1,35,2,4,1,2
+70079,7,2,2,1,21,2,2,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9267.834226,9392.475614,2,93,15,15,4.51,4,4,1,1,0,1,40,1,4,1,5
+70080,7,2,2,32,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16614.865368,17238.21833,3,91,15,15,4.47,4,4,2,0,0,1,33,1,5,1,5
+70081,7,2,2,19,NA,4,4,2,19,236,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,10154.02528,10568.655787,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+70082,7,2,2,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,21398.47235,20994.797177,1,103,14,8,4.39,2,1,0,0,0,2,29,1,5,5,NA
+70083,7,1,2,24,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,3,3,1,2,2,1,2,2,NA,NA,NA,NA,129336.409693,0,1,92,4,4,1.29,2,2,1,0,0,2,24,1,4,3,NA
+70084,7,2,1,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7117.971973,7450.948316,2,100,3,3,0.68,2,2,0,0,2,1,66,1,2,1,2
+70085,7,2,2,6,NA,3,3,2,7,84,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18732.936406,19089.39853,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+70086,7,2,1,18,NA,1,1,1,18,223,2,NA,2,2,4,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,23389.620035,23484.828122,3,91,3,3,0.39,6,6,1,1,0,1,39,2,1,6,NA
+70087,7,2,2,7,NA,4,4,1,7,95,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7899.813226,8111.155436,2,100,15,15,4.47,4,4,0,2,0,1,39,NA,NA,1,5
+70088,7,2,1,47,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17452.049284,17389.002535,1,93,7,7,1.64,5,5,0,2,0,1,47,2,5,1,1
+70089,7,2,1,0,11,1,1,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7222.23638,7150.860834,1,94,4,4,0.32,7,7,3,2,0,2,28,2,2,1,9
+70090,7,2,2,7,NA,3,3,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39810.933651,39282.261038,2,95,15,15,4.77,4,4,0,2,0,2,36,1,4,1,5
+70091,7,2,2,4,NA,3,3,2,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,78813.208592,86985.412201,2,91,6,6,1.34,4,4,1,2,0,2,33,1,4,3,NA
+70092,7,2,1,24,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,120604.496044,126250.391527,2,92,10,10,3.51,3,3,0,0,0,1,24,1,4,5,NA
+70093,7,2,1,14,NA,4,4,2,14,177,NA,NA,2,2,3,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14853.30651,14966.239185,1,93,14,14,5,2,2,0,1,0,2,52,2,3,1,NA
+70094,7,2,2,36,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,60859.550805,63845.233776,2,100,15,15,5,3,3,0,1,0,1,38,1,4,1,4
+70095,7,2,2,59,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,10420.55184,10475.655119,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+70096,7,2,2,77,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,66567.821082,68844.431611,3,91,7,7,1.97,4,4,0,0,1,2,77,1,5,2,NA
+70097,7,2,2,7,NA,3,3,2,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,67046.323141,68322.122844,1,90,8,8,1.67,5,5,2,1,0,2,28,1,4,1,5
+70098,7,2,2,0,9,2,2,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5234.186769,5411.317673,2,90,14,14,3.08,6,6,1,1,1,2,60,2,5,2,NA
+70099,7,2,1,17,NA,3,3,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,100370.520459,102294.664852,1,90,15,15,5,4,4,0,1,0,2,53,1,5,1,5
+70100,7,2,2,11,NA,4,4,1,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9399.281543,9696.12347,2,96,6,6,1.32,5,5,1,3,0,2,30,1,4,3,NA
+70101,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,22109.546782,29180.501371,1,93,7,7,2.72,2,2,0,1,0,2,45,1,3,3,NA
+70102,7,2,1,68,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,92308.775062,93174.50633,1,100,6,6,2.24,1,1,0,0,1,1,68,1,4,2,NA
+70103,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,125443.355152,127362.409324,2,103,10,10,5,1,1,0,0,0,2,56,1,4,3,NA
+70104,7,2,2,34,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16614.865368,17238.21833,3,91,15,15,5,4,4,2,0,0,1,36,2,5,1,5
+70105,7,2,1,25,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,NA,52698.05363,53416.368765,3,92,5,5,0.81,5,5,3,0,0,2,23,1,4,5,NA
+70106,7,2,1,18,NA,3,3,2,19,228,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30100.326038,29873.584609,1,101,6,6,1.17,4,4,0,1,0,1,41,1,3,6,NA
+70107,7,2,2,0,10,4,4,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,4358.100179,4799.448128,2,93,NA,NA,NA,4,4,1,0,1,1,63,NA,NA,6,NA
+70108,7,2,2,11,NA,3,3,2,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49164.586897,48511.701666,2,95,15,15,4.63,5,5,1,2,0,2,36,1,5,1,3
+70109,7,2,1,57,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,27595.50738,28151.492876,1,98,15,15,5,5,5,0,1,1,2,55,1,5,1,5
+70110,7,2,2,0,11,4,4,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5576.185193,5739.245437,2,96,5,5,1.24,3,3,2,0,0,1,29,1,3,5,NA
+70111,7,2,1,56,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12517.592486,15021.976783,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+70112,7,2,2,9,NA,3,3,2,9,110,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,55626.447796,54887.751745,1,95,14,14,3.8,4,4,0,2,0,2,37,1,5,1,5
+70113,7,1,1,28,NA,1,1,NA,NA,NA,2,NA,2,1,77,NA,1,3,NA,2,2,2,2,2,2,NA,NA,NA,NA,35669.2076,0,2,94,77,77,NA,4,4,0,0,0,1,28,2,1,3,NA
+70114,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,14221.330587,18275.989104,3,90,12,12,NA,2,2,0,0,0,1,39,2,3,1,NA
+70115,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,4,2,2,2,2,2,2,2,NA,NA,NA,NA,36453.846815,39554.044508,1,102,2,2,0.52,3,3,0,2,0,2,36,2,3,4,NA
+70116,7,2,1,0,4,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23841.007913,24575.370543,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+70117,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,79677.823556,79777.881901,1,99,15,15,5,5,5,0,3,0,2,43,1,5,1,5
+70118,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,54095.581484,61529.339683,2,91,5,5,1.84,1,1,0,0,1,2,80,1,4,2,NA
+70119,7,2,1,19,NA,3,3,1,19,229,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,110478.18082,109645.964567,2,101,1,1,0.11,2,1,0,0,0,1,19,1,4,NA,NA
+70120,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,53541.401974,54113.194007,1,99,8,8,3.4,2,2,0,0,2,1,74,1,5,1,4
+70121,7,2,2,33,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,22758.541444,23956.866949,1,101,2,2,0.47,3,3,1,0,0,1,35,1,2,6,NA
+70122,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,86578.861495,87453.974807,2,101,10,10,2.33,6,6,1,3,0,1,39,1,2,1,4
+70123,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11762.034222,11824.231183,3,90,77,77,NA,5,5,0,2,0,1,46,2,3,1,3
+70124,7,2,2,16,NA,4,4,1,17,205,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,12531.903464,13043.632492,2,100,4,4,0.85,4,4,0,2,0,2,39,1,3,6,NA
+70125,7,2,1,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,28996.250643,29563.919546,2,101,4,4,1.52,1,1,0,0,0,1,30,1,3,5,NA
+70126,7,2,1,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6992.24593,7077.532809,2,97,1,1,0.33,2,2,1,0,0,2,29,1,3,5,NA
+70127,7,2,2,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,112960.559471,154652.396986,1,91,4,4,1.29,2,2,1,0,0,2,26,1,4,5,NA
+70128,7,1,1,24,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,41383.258526,0,1,97,1,1,0.22,2,1,0,0,0,1,24,1,3,6,NA
+70129,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,19384.896286,19940.089683,1,94,3,3,0.39,6,6,2,2,0,2,25,1,4,1,2
+70130,7,2,1,42,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,21600.805431,23858.641894,2,96,7,7,1.79,4,4,2,0,0,2,49,1,3,1,3
+70131,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12449.932013,12144.773422,3,90,15,15,4.34,4,4,0,0,1,1,65,1,3,1,4
+70132,7,2,1,62,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11761.359913,11893.852645,1,92,9,9,4.23,2,2,0,0,1,1,62,1,3,1,4
+70133,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,21490.337905,2,97,3,1,0.09,2,1,0,0,0,1,30,1,4,5,NA
+70134,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,138322.767578,139738.299504,1,101,10,10,4.63,2,2,0,0,1,1,64,2,3,1,4
+70135,7,2,1,2,NA,1,1,1,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8481.734412,8582.006703,1,103,8,8,1.85,5,5,2,1,0,2,25,2,2,1,2
+70136,7,2,1,39,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,17643.563124,18168.544198,3,91,15,15,5,4,4,2,0,0,2,33,2,5,1,5
+70137,7,2,1,17,NA,2,2,2,17,210,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14081.782012,14391.713696,2,90,2,2,0.25,5,5,0,1,0,2,41,2,4,1,NA
+70138,7,2,1,6,NA,5,7,2,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21147.476454,21752.181979,1,95,5,5,0.89,4,4,0,1,0,2,42,1,4,6,NA
+70139,7,2,2,10,NA,4,4,1,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9453.111053,10178.533204,1,100,6,6,1.13,4,4,0,3,0,2,32,1,3,5,NA
+70140,7,2,1,3,NA,3,3,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,60239.023202,67963.933877,1,98,7,7,1.66,5,5,2,1,0,2,37,1,5,1,3
+70141,7,2,1,0,11,2,2,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4854.394074,5092.044151,2,93,15,15,4.84,6,6,1,1,2,1,66,2,4,1,3
+70142,7,2,2,69,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,105434.067206,107522.739372,2,103,9,9,4.92,1,1,0,0,1,2,69,1,4,2,NA
+70143,7,2,2,40,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,13510.18419,14596.8255,2,90,14,14,4.32,3,3,0,1,0,1,48,2,4,1,5
+70144,7,2,1,12,NA,1,1,1,12,149,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18987.489596,19405.392998,3,91,6,6,2.04,2,2,0,1,0,2,51,1,4,4,NA
+70145,7,2,2,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,29010.447112,28216.500124,1,100,6,6,2.51,1,1,0,0,0,2,49,1,4,3,NA
+70146,7,2,2,12,NA,5,7,2,12,152,NA,NA,2,1,4,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10767.566937,11183.935292,2,91,15,15,3.7,5,5,1,2,0,1,50,NA,NA,1,5
+70147,7,2,1,14,NA,5,6,1,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8915.81491,9312.338117,1,92,14,14,3.69,4,4,0,2,0,1,47,2,4,4,NA
+70148,7,2,1,30,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,24460.137584,1,92,3,3,0.46,5,5,2,1,0,1,30,1,3,1,2
+70149,7,2,1,13,NA,5,6,2,13,160,NA,NA,2,1,4,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8965.57404,9581.867503,1,90,15,15,5,5,5,0,3,0,2,46,2,4,1,5
+70150,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,76374.321112,79950.814336,2,99,14,14,5,1,1,0,0,0,1,39,1,5,5,NA
+70151,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,64679.499599,80402.538623,3,92,6,6,2.69,1,1,0,0,1,2,80,1,3,2,NA
+70152,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,161992.272945,167997.905035,1,91,14,14,3.8,4,4,0,2,0,1,50,NA,NA,1,5
+70153,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,24930.322327,25181.116543,2,94,3,3,0.92,1,1,0,0,0,1,57,1,2,3,NA
+70154,7,2,1,0,1,5,6,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5941.765349,6071.648885,2,102,3,3,0.38,5,5,3,0,0,2,30,2,2,1,4
+70155,7,2,1,18,NA,5,7,1,18,219,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13276.485807,15665.287637,2,100,99,99,NA,3,3,0,0,0,1,46,1,9,3,NA
+70156,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,34853.379657,39642.894617,3,90,10,10,4.3,2,2,0,0,2,2,80,1,4,1,5
+70157,7,2,2,1,22,4,4,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9273.23044,9685.008923,2,96,5,5,1.24,3,3,2,0,0,1,29,1,3,5,NA
+70158,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,24905.670199,24973.138542,2,101,8,8,2.7,3,3,0,1,0,1,53,1,4,1,2
+70159,7,2,1,13,NA,4,4,2,13,162,NA,NA,2,1,4,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,7,7,1.52,4,4,0,2,0,2,30,2,4,1,5
+70160,7,2,2,61,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,10288.382343,13758.61994,2,90,6,6,1.7,2,2,0,0,2,1,61,2,1,1,2
+70161,7,2,1,34,NA,1,1,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,37715.365512,38149.83984,2,102,7,7,1.33,6,6,1,3,0,1,34,2,2,1,1
+70162,7,2,1,34,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21288.18311,21326.274673,1,102,7,7,1.57,4,4,0,2,0,2,33,1,4,1,4
+70163,7,2,2,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,163605.682975,162710.354378,1,90,15,15,5,4,4,0,1,0,2,53,1,5,1,5
+70164,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,23884.62129,24647.381072,2,98,5,5,1.24,3,3,0,0,1,2,58,1,2,5,NA
+70165,7,2,2,18,NA,2,2,2,19,228,2,NA,1,1,NA,13,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,15442.648697,16179.397194,3,90,77,77,NA,4,3,0,0,0,1,45,2,3,3,NA
+70166,7,2,1,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,137038.746155,146586.432966,2,101,4,2,0.66,2,1,0,0,0,1,21,1,4,5,NA
+70167,7,2,2,47,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,153972.608815,154123.803084,2,101,5,5,1.36,2,2,0,0,0,2,47,1,2,1,4
+70168,7,2,1,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,64901.456576,67363.75016,1,101,15,15,5,2,2,0,0,2,2,73,1,4,1,4
+70169,7,2,2,23,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,60324.348827,60863.689632,1,95,6,6,0.96,5,5,1,0,1,2,69,1,1,2,NA
+70170,7,2,2,59,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,15497.844354,15073.705188,1,99,4,2,0.74,2,1,0,0,1,2,59,1,3,3,NA
+70171,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,27532.825087,28260.668065,1,101,6,6,1.3,4,4,2,0,0,2,30,1,4,6,NA
+70172,7,2,2,11,NA,5,7,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5147.116597,5520.448595,2,92,15,15,4.59,4,4,0,2,0,2,45,2,5,1,5
+70173,7,2,2,2,NA,2,2,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9653.164181,10331.416262,1,90,15,15,4.44,5,5,2,1,0,1,36,1,3,1,4
+70174,7,1,1,54,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,24930.322327,0,1,100,1,1,0.01,1,1,0,0,0,1,54,1,2,3,NA
+70175,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,11699.431733,12221.798846,1,96,7,7,2.31,2,2,0,0,1,2,62,1,3,3,NA
+70176,7,2,2,23,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,41953.42893,43056.968872,2,99,4,1,0.22,4,1,0,0,0,2,21,NA,NA,5,NA
+70177,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,45403.540522,46903.566528,2,92,9,6,2.75,2,1,0,0,0,1,26,99,9,6,NA
+70178,7,1,2,6,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,62595.719575,0,1,95,14,14,3.8,4,4,1,1,0,1,36,1,4,1,5
+70179,7,2,1,4,NA,4,4,1,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10690.995725,11017.541084,2,98,2,2,0.31,4,4,2,1,0,2,27,1,2,4,NA
+70180,7,2,1,4,NA,4,4,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10510.490567,10944.785425,2,98,6,6,1,5,5,2,1,0,2,31,1,4,6,NA
+70181,7,2,1,6,NA,1,1,2,6,72,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,13285.093011,13365.60735,2,94,77,77,NA,6,6,0,3,0,2,58,1,3,1,9
+70182,7,1,1,14,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10346.302718,0,2,91,4,4,0.65,5,5,1,3,0,1,43,2,3,5,NA
+70183,7,2,2,37,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,2,1,2,2,2,2,1,2,2,1,2,2,2,41791.57979,40663.955172,2,102,6,6,1.03,5,5,1,1,0,1,37,1,2,1,2
+70184,7,2,1,16,NA,4,4,1,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16147.713323,16532.569027,1,92,15,15,4.44,5,5,0,3,0,2,43,1,5,6,NA
+70185,7,2,2,4,NA,2,2,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13647.772496,15067.50708,2,93,7,7,1.83,3,3,1,0,0,1,40,2,5,1,4
+70186,7,2,1,48,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19320.837782,1,99,14,14,3.8,4,4,1,1,0,1,48,2,5,1,5
+70187,7,2,1,15,NA,5,6,2,15,180,NA,NA,2,1,99,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10567.840237,11294.273462,1,91,15,15,5,4,4,0,2,0,2,55,1,5,1,5
+70188,7,1,1,34,NA,5,6,NA,NA,NA,2,NA,2,2,4,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15057.879445,0,3,90,12,12,NA,3,3,0,0,0,2,29,2,5,1,5
+70189,7,2,2,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,5,3,2,1,2,2,2,2,NA,NA,NA,NA,32455.694722,32314.335903,1,103,5,5,0.74,5,5,1,1,0,2,40,99,3,1,1
+70190,7,2,1,9,NA,1,1,1,9,116,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14042.313177,3,92,4,4,0.6,6,6,2,2,0,2,24,1,3,6,NA
+70191,7,2,2,52,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,3,NA,2,2,2,1,2,2,NA,NA,NA,NA,30346.899457,30505.155956,1,93,5,5,0.84,5,5,1,2,0,2,52,2,1,3,NA
+70192,7,2,2,22,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,17397.027021,18573.153951,3,91,7,7,3.21,1,1,0,0,0,2,22,1,5,5,NA
+70193,7,2,2,17,NA,2,2,2,17,208,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17277.331962,18186.067085,1,90,6,6,0.81,6,6,0,3,0,2,45,1,4,1,2
+70194,7,2,2,29,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,13440.945024,14349.620822,1,103,15,8,4.03,2,1,0,0,0,2,29,2,5,6,NA
+70195,7,2,1,79,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,16888.31509,17277.940364,1,92,14,5,2.15,3,1,0,0,2,1,51,1,4,5,NA
+70196,7,2,1,37,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,5,NA,2,2,2,1,2,2,2,2,2,2,34997.800447,35379.102239,2,96,5,5,0.78,5,5,0,2,0,1,37,2,1,5,NA
+70197,7,2,1,0,6,2,2,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5385.874932,5867.351136,2,90,8,8,2.24,4,4,1,1,0,2,29,1,4,6,NA
+70198,7,2,2,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,11862.334174,1,96,15,15,5,2,2,0,0,2,2,62,1,4,1,2
+70199,7,2,1,7,NA,3,3,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,60593.636684,64068.123183,1,91,10,10,2.77,5,5,0,3,0,1,43,1,5,1,5
+70200,7,2,1,52,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,27551.752692,27682.321328,1,98,3,3,0.73,3,3,0,0,0,1,52,1,4,1,3
+70201,7,1,2,26,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,130601.953362,0,2,91,15,9,5,2,1,0,0,0,2,26,1,5,5,NA
+70202,7,2,1,22,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,9956.598907,10266.888978,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+70203,7,2,1,46,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12262.33683,12218.03828,2,100,4,4,0.44,7,7,1,2,2,1,71,2,1,1,1
+70204,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,18097.801029,17328.248994,2,100,7,7,1.38,5,5,1,0,0,2,45,1,2,3,NA
+70205,7,2,2,7,NA,3,3,1,7,90,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,25527.806244,26565.240294,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+70206,7,2,2,34,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,12808.938247,13423.491647,2,100,15,15,5,4,4,1,1,0,1,36,2,5,1,5
+70207,7,2,2,8,NA,3,3,1,8,101,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,63195.899182,62711.128206,1,100,15,15,5,4,4,1,1,0,1,40,1,5,1,5
+70208,7,2,1,2,NA,3,3,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,29706.069213,32871.212026,1,99,10,10,2.48,5,5,2,1,0,1,33,1,5,1,5
+70209,7,2,2,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,27945.726298,28031.650144,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+70210,7,2,1,47,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,41527.444056,55523.849773,1,95,5,5,1.5,2,2,0,1,0,1,47,1,4,2,NA
+70211,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,38758.039282,44961.529423,1,98,6,6,1.9,2,2,0,0,2,1,80,1,1,1,2
+70212,7,2,2,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,20048.680628,21335.7055,2,95,5,5,1.05,3,3,0,1,0,1,43,1,3,1,2
+70213,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,94698.084211,97410.286035,1,101,14,14,3.15,5,5,2,1,0,1,35,1,4,1,5
+70214,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,25212.845431,2,101,1,1,0.42,1,1,0,0,0,1,21,1,4,5,NA
+70215,7,2,2,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,1,2,NA,44484.790948,49854.064081,1,103,15,15,5,3,2,0,0,3,2,63,1,5,77,NA
+70216,7,2,2,0,8,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3941.796129,4340.984667,1,99,14,14,3.67,4,4,1,0,0,2,49,1,3,1,3
+70217,7,2,1,40,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19158.860679,19089.647955,1,100,15,15,5,2,2,0,0,0,1,40,2,5,1,5
+70218,7,2,1,0,11,1,1,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6298.658963,6412.943673,2,94,5,5,1.07,4,4,2,0,0,1,37,2,1,1,1
+70219,7,2,2,69,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,42101.975168,42617.525207,1,95,6,6,0.96,5,5,1,0,1,2,69,1,1,2,NA
+70220,7,2,2,31,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,1,2,2,2,2,2,2,2,2,2,2,2,41791.57979,43231.925575,2,102,10,10,3.04,4,4,2,0,0,2,31,2,2,1,NA
+70221,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9122.654131,9281.792212,1,96,9,9,3.14,3,3,0,2,0,2,39,NA,NA,3,NA
+70222,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,88617.795432,87713.219218,1,91,15,15,5,1,1,0,0,1,1,60,1,5,1,NA
+70223,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,146181.198007,145805.68861,2,91,15,2,0.79,7,1,0,0,1,1,49,NA,NA,5,NA
+70224,7,2,2,19,NA,3,3,1,19,234,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,39616.634313,40360.331753,2,101,5,3,1.1,2,1,0,0,0,1,19,1,4,NA,NA
+70225,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105262.341027,112370.874556,2,98,10,10,4.42,2,2,0,0,0,1,25,1,5,1,5
+70226,7,2,2,70,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,NA,12849.335508,13810.42202,1,96,4,4,1.18,2,2,0,0,1,2,70,1,1,5,NA
+70227,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15819.48188,17091.862943,1,90,15,15,5,5,5,0,3,0,2,46,2,4,1,5
+70228,7,2,1,67,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7410.50521,7700.344649,2,96,3,3,0.82,2,2,0,1,1,1,67,1,3,3,NA
+70229,7,2,2,30,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,3,1,2,2,2,2,1,2,2,1,2,2,1,41791.57979,41445.196731,2,102,4,4,0.61,5,5,0,3,0,1,34,2,3,1,3
+70230,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,15976.466658,17286.192057,2,96,7,7,2.45,2,2,0,0,0,1,24,2,5,5,NA
+70231,7,2,2,37,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,21619.283038,22210.398171,2,102,15,15,5,4,4,0,2,0,1,39,1,4,1,5
+70232,7,2,2,18,NA,5,6,2,18,217,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12224.9472,12547.183925,1,97,14,14,2.72,7,7,0,2,0,1,40,1,5,1,5
+70233,7,2,2,12,NA,1,1,2,12,148,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,15591.526146,16856.688627,1,90,6,6,1.11,5,5,1,2,0,1,30,2,1,6,NA
+70234,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,1,2,1,1,2,1,1,2,NA,28559.076421,28710.827523,1,95,2,2,0.87,1,1,0,0,1,1,63,1,1,5,NA
+70235,7,2,1,61,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,2,NA,2,2,2,1,2,2,1,2,2,2,8609.250304,11228.904188,2,90,3,3,1.01,1,1,0,0,1,1,61,2,1,2,NA
+70236,7,2,2,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,11862.334174,1,96,15,15,5,2,2,0,0,2,1,68,1,3,1,4
+70237,7,2,1,13,NA,3,3,2,14,168,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,68701.580401,72973.564721,2,94,15,15,5,5,5,0,3,0,1,44,1,5,1,4
+70238,7,2,2,25,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,32455.694722,32314.335903,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+70239,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,19075.861607,19022.257361,1,96,10,10,3.04,4,4,0,1,0,2,43,1,5,1,4
+70240,7,2,2,4,NA,4,4,1,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11015.482909,2,100,99,99,NA,6,6,2,1,0,2,44,1,3,1,4
+70241,7,2,1,13,NA,4,4,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11800.231369,11823.561392,1,96,7,7,1,7,7,2,1,1,2,53,1,4,1,3
+70242,7,2,1,1,22,4,4,2,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5799.129348,6038.750138,1,90,4,4,0.67,5,5,3,0,0,2,32,2,3,3,NA
+70243,7,1,2,51,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,1,3,NA,1,2,1,1,2,1,NA,NA,NA,NA,21018.496735,0,1,91,9,9,2.68,4,4,0,0,1,1,20,NA,NA,5,NA
+70244,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,9772.038079,10486.56257,1,95,77,77,NA,3,3,0,0,2,1,80,1,1,1,3
+70245,7,2,2,1,16,4,4,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6398.740074,6885.115697,1,93,6,6,0.83,6,6,3,1,0,1,37,NA,NA,1,3
+70246,7,2,2,72,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,14971.827573,15430.401607,2,97,8,8,2.7,3,3,0,0,1,2,72,1,2,3,NA
+70247,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17852.668137,17947.072019,3,91,9,9,4.08,2,2,0,1,0,2,54,2,5,1,NA
+70248,7,2,2,67,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,7278.790659,7875.678599,2,93,3,3,0.66,2,2,0,0,1,1,54,2,1,1,1
+70249,7,2,1,7,NA,3,3,2,8,96,NA,NA,2,1,3,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,38712.032122,41122.784973,3,91,15,15,5,4,4,1,1,0,2,41,1,5,1,5
+70250,7,2,2,0,5,2,2,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10218.497734,10897.586635,2,101,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+70251,7,2,2,43,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,43535.993088,45116.30779,2,102,77,77,NA,4,4,0,1,0,1,47,1,2,1,3
+70252,7,2,1,40,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,14227.821648,16178.124803,3,90,2,2,0.45,1,1,0,0,0,1,40,2,2,4,NA
+70253,7,2,1,0,8,5,7,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6246.568228,6909.306675,2,95,1,1,0.03,3,3,1,0,0,1,23,1,3,6,NA
+70254,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,115926.402585,128579.517653,1,101,14,14,3.25,4,4,0,1,0,1,48,1,4,1,2
+70255,7,2,2,38,NA,1,1,1,NA,NA,2,NA,2,2,99,NA,1,6,2,2,2,2,1,2,2,2,2,2,2,53370.792448,51930.736348,1,100,NA,13,NA,2,1,0,0,0,1,32,2,1,6,NA
+70256,7,2,2,16,NA,5,6,1,16,198,NA,NA,2,1,4,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9098.177005,9288.806625,1,98,14,14,3.9,4,4,0,1,0,2,52,1,5,1,5
+70257,7,2,2,1,20,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5687.793894,6002.477845,2,90,8,6,1.67,5,3,2,0,0,2,25,2,3,5,NA
+70258,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,61710.107686,72008.199278,1,98,3,3,0.9,1,1,0,0,0,2,24,1,5,5,NA
+70259,7,2,1,65,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,1,1,7514.993062,8122.478925,2,90,2,2,0.48,2,2,0,0,1,2,53,2,3,1,3
+70260,7,2,2,0,11,2,2,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,6235.764746,6104.57156,2,92,NA,NA,NA,4,4,2,0,0,1,40,NA,NA,1,NA
+70261,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,37814.382501,38402.065392,1,98,6,4,1.34,2,1,0,0,0,2,22,1,4,6,NA
+70262,7,2,1,7,NA,2,2,1,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10738.959181,11444.574702,2,93,10,10,2.26,6,6,0,4,0,1,34,1,4,1,3
+70263,7,2,2,49,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,18522.193415,19195.925691,1,90,9,9,2.6,4,4,0,1,0,2,49,2,2,1,5
+70264,7,2,2,23,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,NA,NA,NA,1,2,2,1,11475.373333,11965.391974,2,92,12,NA,NA,7,1,0,0,2,1,53,2,3,1,3
+70265,7,2,2,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,19378.199212,21493.287569,2,100,5,5,1.3,3,3,0,1,0,2,46,1,3,2,NA
+70266,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16420.864787,17154.038836,2,102,14,14,5,2,2,0,0,2,1,64,1,4,1,3
+70267,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,20891.980831,21490.337905,2,97,12,14,5,2,1,0,0,0,1,53,NA,NA,5,NA
+70268,7,2,1,46,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,30549.358294,31009.655884,2,101,5,5,1.93,1,1,0,0,0,1,46,1,4,5,NA
+70269,7,2,1,0,9,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8212.233655,8522.67606,2,91,2,2,0.33,5,5,1,1,0,2,48,1,4,3,NA
+70270,7,2,2,2,NA,4,4,2,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6084.391121,6799.274602,2,99,3,3,0.56,4,4,1,0,0,2,38,1,3,5,NA
+70271,7,1,1,34,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17879.023129,0,2,90,7,7,3.49,1,1,0,0,0,1,34,1,3,5,NA
+70272,7,2,2,1,16,4,4,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6811.556363,7329.310628,1,96,12,12,NA,5,5,1,2,0,2,35,1,5,1,4
+70273,7,2,2,8,NA,1,1,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16371.237244,2,98,1,1,0.18,4,4,0,2,0,1,29,1,4,6,NA
+70274,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,30212.098573,33743.401779,2,98,3,3,1.24,1,1,0,0,1,2,80,1,4,2,NA
+70275,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,26999.643202,25671.572746,2,102,6,6,1.22,5,5,0,2,0,2,42,1,4,1,4
+70276,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,164908.075108,169304.25724,1,98,15,15,5,3,3,0,0,0,1,56,1,5,1,5
+70277,7,2,2,0,9,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6021.973783,5836.909368,1,102,2,2,0.19,7,7,2,2,0,1,48,2,9,1,9
+70278,7,2,2,36,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,2,1,3,1,2,2,1,2,2,NA,NA,NA,NA,21765.629914,21610.101161,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+70279,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,17521.481386,17626.852628,2,97,99,99,NA,1,1,0,0,0,2,53,1,3,3,NA
+70280,7,2,1,18,NA,2,2,1,18,221,2,NA,2,2,3,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20606.470013,20581.836663,1,103,6,6,0.93,5,5,0,1,0,1,39,2,3,1,3
+70281,7,2,2,30,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,20012.879008,20053.743639,1,100,7,7,3.13,1,1,0,0,0,2,30,1,4,5,NA
+70282,7,2,1,24,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,2,5,NA,2,2,2,2,2,2,1,2,2,1,35669.2076,36155.406423,2,94,7,7,1.23,6,6,2,1,0,1,33,2,1,6,NA
+70283,7,2,1,11,NA,1,1,2,11,141,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13484.595524,13595.798197,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+70284,7,2,2,4,NA,5,6,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3864.413878,3822.111022,1,99,14,14,2.66,7,7,3,1,0,1,35,1,5,1,5
+70285,7,2,2,37,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,28958.579549,29304.033516,1,96,13,13,NA,5,5,1,1,0,1,42,1,3,5,NA
+70286,7,2,1,18,NA,3,3,2,18,221,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,71231.747774,75661.062324,2,95,9,9,2.22,5,5,1,0,0,1,55,1,4,1,5
+70287,7,2,2,46,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18844.090639,18943.737107,2,96,15,15,5,2,2,0,0,0,1,45,1,5,1,4
+70288,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,2,1,2,2,1,2,2,1,2,2,1,71034.153987,76359.172033,1,98,6,6,1.11,5,5,0,2,1,2,37,1,1,1,1
+70289,7,2,1,8,NA,3,3,2,8,99,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,46228.073505,47549.950918,2,94,10,10,2.91,4,4,0,2,0,2,38,1,4,1,4
+70290,7,2,1,60,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,10880.024478,11166.738171,2,91,9,9,3.14,3,3,0,1,2,1,60,1,4,1,3
+70291,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,35313.648114,35629.376203,2,95,6,6,0.9,6,6,1,1,0,1,49,1,1,1,1
+70292,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19130.246369,22400.771377,2,95,15,10,3.67,5,3,0,0,0,1,47,1,5,1,3
+70293,7,2,2,51,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,17149.727778,17902.578105,1,90,4,4,0.78,4,4,0,0,1,1,69,2,4,1,3
+70294,7,2,1,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,8219.195224,8283.488354,2,99,15,15,5,2,2,0,0,2,1,60,1,5,1,5
+70295,7,2,1,47,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,21158.364877,21743.069946,1,96,15,15,4.9,4,4,0,1,0,1,47,1,3,1,5
+70296,7,2,2,0,4,3,3,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17674.326293,18731.006164,3,90,15,15,5,3,3,1,0,0,1,31,1,5,1,5
+70297,7,2,1,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,60942.568495,64726.722108,1,94,7,7,2.51,2,2,0,0,2,2,72,1,4,1,1
+70298,7,2,2,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,22707.329726,23560.698233,2,97,2,2,0.3,4,4,0,2,0,1,42,1,2,6,NA
+70299,7,2,1,62,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,7192.368251,7796.453586,1,93,4,4,1.74,1,1,0,0,1,1,62,2,3,2,NA
+70300,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,27466.648066,27396.091902,1,99,2,2,0.61,2,2,0,0,0,1,46,1,5,5,NA
+70301,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,21143.964074,21647.100945,1,100,1,1,0.04,4,4,1,1,0,2,51,1,3,3,NA
+70302,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,42894.724338,43561.362107,1,98,6,4,1.34,2,1,0,0,0,1,24,1,5,5,NA
+70303,7,2,1,56,NA,2,2,2,NA,NA,1,1,2,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,34813.426994,34467.752983,1,97,15,15,5,1,1,0,0,0,1,56,2,4,3,NA
+70304,7,2,2,65,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,122483.259869,138551.168646,3,91,7,7,3.49,1,1,0,0,1,2,65,1,3,2,NA
+70305,7,2,1,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23912.171644,23498.78217,1,91,15,15,5,3,3,1,0,0,1,33,1,5,1,5
+70306,7,2,1,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25645.251384,25514.298117,1,92,6,6,1.31,3,3,0,0,1,2,80,1,3,4,NA
+70307,7,2,1,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,47487.549895,49034.44781,1,102,6,6,1.48,3,3,0,0,1,2,57,2,1,1,4
+70308,7,2,1,5,NA,1,1,1,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20874.345556,20902.551716,3,92,2,2,0.47,3,3,1,1,0,2,33,1,4,5,NA
+70309,7,2,1,2,NA,1,1,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12413.685227,12209.161559,1,102,6,6,1.46,3,3,1,1,0,2,28,2,4,3,NA
+70310,7,1,2,11,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,1,2,NA,NA,NA,NA,9139.784234,0,2,100,6,6,1.18,5,5,0,2,1,2,70,1,2,77,NA
+70311,7,2,1,7,NA,1,1,1,8,96,NA,NA,2,2,2,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13581.60325,14182.420159,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+70312,7,2,2,73,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,1,4,NA,2,2,2,1,2,2,1,2,2,NA,17318.187297,23904.945555,2,90,2,2,0.73,1,1,0,0,1,2,73,2,1,4,NA
+70313,7,2,2,37,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,26465.930618,27183.288271,2,100,15,15,4.47,4,4,0,2,0,1,39,NA,NA,1,5
+70314,7,2,1,5,NA,3,3,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58618.419318,66135.507563,2,94,12,12,NA,5,5,1,1,0,1,37,1,4,1,3
+70315,7,2,2,5,NA,5,6,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4938.043177,5373.703942,1,91,5,5,0.89,4,4,2,0,0,1,39,1,4,1,5
+70316,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4667.634102,4955.769831,3,90,15,15,5,5,5,1,0,1,1,38,2,3,1,4
+70317,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,12813.519628,13398.278814,2,92,7,7,1.61,4,4,0,2,0,1,51,2,3,1,3
+70318,7,2,2,23,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,85402.868381,86166.428247,2,94,77,77,NA,2,2,0,0,0,2,23,1,3,6,NA
+70319,7,2,1,55,NA,5,6,1,NA,NA,2,NA,2,2,7,NA,1,4,NA,1,2,2,1,2,2,1,2,2,1,16499.662173,17012.89406,3,91,6,6,1.77,2,2,0,0,0,2,57,2,1,1,1
+70320,7,2,1,4,NA,2,2,1,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16288.754504,16020.386509,2,92,15,15,5,3,3,1,0,0,2,48,2,5,1,5
+70321,7,2,1,63,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,2,2,2,1,2,2,1,2,2,2,9115.676792,9431.98541,2,90,5,5,1.84,1,1,0,0,1,1,63,2,4,3,NA
+70322,7,2,1,16,NA,1,1,2,16,200,NA,NA,2,2,3,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,21633.039913,21721.097793,2,94,7,7,1.04,7,7,0,3,0,1,37,2,1,1,3
+70323,7,2,2,7,NA,4,4,2,7,89,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8757.841043,8910.615224,2,97,2,2,0.38,3,3,1,1,0,2,27,1,2,5,NA
+70324,7,2,2,14,NA,3,3,2,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,109773.944307,112955.590462,1,98,15,15,4.34,4,4,0,2,0,1,51,1,5,1,5
+70325,7,2,2,6,NA,1,1,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14679.337943,15307.210901,1,90,15,15,4.77,4,4,1,1,0,2,41,1,5,1,2
+70326,7,2,1,35,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,22188.836739,23952.480361,1,95,5,5,0.89,4,4,0,1,0,2,42,1,4,6,NA
+70327,7,2,2,62,NA,2,2,2,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,2,12121.359422,12627.249955,2,99,15,15,5,2,2,0,0,2,1,64,2,5,1,4
+70328,7,2,1,27,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,3,1,NA,2,2,2,1,2,2,1,2,2,2,35782.041084,36966.59793,2,96,1,1,0.06,5,5,2,1,0,1,27,2,3,1,4
+70329,7,2,2,60,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,9716.805546,12994.252166,2,90,3,3,0.46,5,5,0,2,2,1,75,2,1,1,2
+70330,7,2,1,77,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,8286.514589,8909.879599,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+70331,7,2,1,52,NA,1,1,1,NA,NA,2,NA,2,2,7,NA,1,1,NA,2,2,1,2,2,1,NA,NA,NA,NA,33162.406014,36437.864606,3,92,4,4,0.66,4,4,0,1,0,1,52,2,1,1,1
+70332,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,10676.164039,11456.797578,2,97,4,4,1.29,2,2,0,0,2,1,74,1,3,3,NA
+70333,7,2,2,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,48470.428632,49964.788237,1,98,15,15,5,5,5,0,1,1,2,55,1,5,1,5
+70334,7,2,1,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10229.206765,10406.656985,1,96,15,15,4.52,6,6,0,4,0,1,46,1,4,1,4
+70335,7,2,2,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,38954.135779,44674.189003,1,91,3,3,1.15,1,1,0,0,0,2,53,1,4,3,NA
+70336,7,2,2,74,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,2,2,2,NA,15730.58404,17568.357111,2,98,6,6,1.62,3,3,0,0,1,2,74,1,1,2,NA
+70337,7,2,1,38,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19788.748292,24506.181083,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+70338,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,42468.064168,42915.721814,2,101,6,6,1.16,4,4,0,3,0,2,36,1,4,4,NA
+70339,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,33248.548181,34134.900613,1,98,4,4,0.89,3,3,0,0,1,2,55,1,5,1,NA
+70340,7,2,2,16,NA,5,6,2,16,197,NA,NA,1,1,NA,8,NA,NA,NA,1,2,1,1,2,1,1,2,2,1,10767.566937,11540.84845,2,91,12,12,NA,7,6,0,4,2,2,72,2,1,2,NA
+70341,7,2,1,62,NA,5,6,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12579.986433,13271.133625,1,92,14,14,5,2,2,0,0,2,1,62,1,4,1,4
+70342,7,2,2,61,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,2,2,NA,1,2,1,1,2,1,1,2,1,3,11838.431472,12295.352662,2,92,7,7,1.89,3,3,0,0,1,1,36,2,3,5,NA
+70343,7,2,1,11,NA,5,6,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7179.084455,7651.500738,2,98,9,9,2.29,5,5,0,2,0,1,36,1,4,1,4
+70344,7,2,1,1,18,4,4,1,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6376.965739,6390.839385,2,100,3,3,0.31,7,7,3,2,0,2,28,1,3,1,3
+70345,7,2,1,5,NA,1,1,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14321.363328,13945.949794,2,96,77,77,NA,7,7,3,2,0,2,33,2,2,6,NA
+70346,7,2,2,11,NA,1,1,1,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16412.026403,2,98,14,14,2.87,5,5,0,3,0,2,34,1,2,1,2
+70347,7,2,1,70,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,10025.317761,10256.608873,2,103,15,15,5,3,3,0,0,1,2,55,1,4,1,5
+70348,7,2,2,9,NA,3,3,1,9,118,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22933.149195,23865.138008,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+70349,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,12535.973802,12972.182488,2,100,4,4,0.97,3,3,0,0,3,2,80,1,5,2,NA
+70350,7,2,2,18,NA,3,3,1,18,221,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,71832.578284,79011.982474,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+70351,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,38799.676345,43482.761402,1,94,6,6,2.24,1,1,0,0,1,2,80,1,4,2,NA
+70352,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,29167.119125,29652.805867,1,101,2,2,0.22,4,4,1,0,0,2,25,1,4,6,NA
+70353,7,2,1,6,NA,4,4,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8227.856305,9153.104022,3,91,1,1,0.07,6,6,2,3,0,2,30,1,2,3,NA
+70354,7,2,1,17,NA,4,4,2,17,207,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,8,8,2,4,4,1,2,0,2,40,1,4,5,NA
+70355,7,2,1,65,NA,5,6,2,NA,NA,2,NA,2,1,4,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,9963.081107,10271.419526,1,91,4,4,1.33,2,2,0,0,2,1,65,2,4,1,3
+70356,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,5,2,1,2,2,1,2,2,1,2,2,1,67534.233567,86418.733625,1,97,3,3,0.83,2,2,0,0,0,2,25,1,1,5,NA
+70357,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,12601.697316,12716.910215,1,91,15,15,5,6,6,0,2,2,1,50,2,5,1,5
+70358,7,2,2,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,154825.466557,155723.680504,1,91,8,8,4.41,1,1,0,0,0,2,44,1,4,3,NA
+70359,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,46965.818538,50603.721684,1,95,14,14,5,2,2,0,0,2,1,80,1,4,1,NA
+70360,7,2,2,8,NA,4,4,2,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8147.287486,8823.301862,2,90,5,5,1.63,2,2,0,1,0,2,38,2,3,5,NA
+70361,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,1,1,2,NA,50879.818001,58591.132961,1,101,6,6,1.31,3,3,0,0,1,2,80,1,1,2,NA
+70362,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,8308.628726,8679.600111,2,95,5,5,1.96,1,1,0,0,1,2,61,1,4,2,NA
+70363,7,2,2,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,122483.259869,124909.680309,3,91,15,15,5,2,2,0,0,2,2,62,1,5,1,5
+70364,7,2,1,73,NA,5,7,1,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11532.082424,12399.599956,2,103,12,12,NA,2,2,0,0,2,1,73,2,5,1,4
+70365,7,2,2,7,NA,5,6,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8290.163782,8692.019106,1,92,2,2,0.24,5,5,0,2,0,1,35,2,4,1,3
+70366,7,2,2,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,19693.606802,19976.427063,1,99,12,12,NA,2,2,0,0,0,1,28,1,5,5,NA
+70367,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,56938.907307,57514.428801,2,95,15,15,4.63,5,5,1,2,0,2,36,1,5,1,3
+70368,7,2,1,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19779.547388,20214.272069,2,96,15,15,5,3,3,0,1,0,1,55,1,5,1,4
+70369,7,2,2,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,60324.348827,64927.943097,1,95,4,4,0.97,3,3,2,0,0,2,22,1,4,5,NA
+70370,7,2,1,6,NA,4,4,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8439.71412,8914.772112,2,99,7,7,1.19,6,6,1,3,0,2,38,1,3,5,NA
+70371,7,2,2,2,NA,1,1,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11582.174418,12474.053558,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+70372,7,2,1,61,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,4,1,NA,2,2,2,1,2,2,1,2,2,1,11568.876339,11794.347884,1,102,77,77,NA,3,3,0,0,1,1,61,2,4,1,1
+70373,7,2,1,5,NA,1,1,2,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14505.510202,14676.996441,2,94,77,77,NA,5,5,1,1,0,1,41,2,2,1,2
+70374,7,2,1,4,NA,3,3,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,33334.752566,38103.457131,3,92,6,6,0.74,7,7,2,1,0,2,46,1,2,1,4
+70375,7,2,2,19,NA,1,1,1,19,231,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21062.314667,23105.758741,1,94,6,6,1.3,4,4,2,0,0,1,24,2,1,1,4
+70376,7,2,1,17,NA,1,1,1,17,215,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14432.845547,14415.592261,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+70377,7,2,1,3,NA,2,2,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12403.412256,12420.172189,2,90,2,2,0.49,3,3,2,0,0,2,26,1,4,1,NA
+70378,7,2,1,53,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,18681.278463,18953.419735,1,93,7,7,1.79,4,4,0,2,0,1,53,2,4,1,4
+70379,7,2,2,2,NA,2,2,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11443.671453,12634.120381,1,100,15,15,4.34,4,4,2,0,0,2,35,1,5,1,5
+70380,7,2,1,9,NA,4,4,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7730.47951,9212.541007,1,99,6,6,1.3,5,5,1,2,0,1,34,1,2,1,3
+70381,7,2,1,24,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,39031.957066,39638.562591,1,95,6,6,0.96,5,5,1,0,1,2,69,1,1,2,NA
+70382,7,2,1,1,22,1,1,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,10960.671575,11090.250219,2,96,4,4,0.69,5,5,2,0,0,2,57,2,1,4,NA
+70383,7,2,2,4,NA,3,3,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,67177.961189,74143.696839,1,98,9,9,2.6,4,4,1,1,0,2,35,1,2,1,NA
+70384,7,2,1,25,NA,4,4,1,NA,NA,2,NA,2,1,4,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,23484.626749,24110.252242,2,96,6,6,1.7,2,2,0,1,0,1,25,2,4,5,NA
+70385,7,2,2,57,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,15002.194143,18506.393496,3,90,7,2,0.77,2,1,0,0,0,1,44,1,2,6,NA
+70386,7,2,1,38,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17554.053699,17499.245836,1,96,2,2,0.73,1,1,0,0,0,1,38,1,3,5,NA
+70387,7,2,2,60,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,3,6,NA,2,2,2,2,2,2,2,2,2,2,15876.871857,17052.998921,2,102,14,7,3.67,2,1,0,0,2,1,65,2,3,6,NA
+70388,7,2,1,12,NA,3,3,2,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,56223.281913,55799.760014,1,99,14,14,4.86,3,3,0,1,0,1,56,1,5,1,5
+70389,7,2,2,56,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,3,4,NA,2,2,2,2,2,2,2,2,1,2,20734.495277,22784.232134,2,99,99,99,NA,4,1,0,0,0,2,42,2,4,5,NA
+70390,7,2,2,8,NA,1,1,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10118.363218,10311.586628,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+70391,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,27681.279,30792.259764,1,91,3,3,1.29,1,1,0,0,1,2,80,1,3,5,NA
+70392,7,2,1,0,10,2,2,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5610.286388,5712.081065,2,90,6,6,1.34,4,4,1,2,0,2,36,2,3,77,NA
+70393,7,2,1,9,NA,4,4,1,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9418.975084,9571.535533,2,93,5,5,1.04,4,4,1,1,0,1,29,1,3,6,NA
+70394,7,2,1,47,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13883.119445,13832.965705,1,103,15,15,5,1,1,0,0,0,1,47,1,5,5,NA
+70395,7,1,1,36,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,114168.79702,0,1,95,14,14,3.8,4,4,1,1,0,1,36,1,4,1,5
+70396,7,2,1,6,NA,5,7,1,6,77,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8930.369586,9194.254241,3,91,14,14,3.8,4,4,0,2,0,1,47,1,5,1,5
+70397,7,2,1,19,NA,1,1,1,19,234,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21898.969807,22012.999264,2,102,15,15,3.92,5,5,0,0,0,1,19,1,4,NA,NA
+70398,7,2,2,57,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,2,4,NA,2,2,2,1,2,2,1,2,2,NA,21097.069797,24201.517172,2,90,6,6,1.7,2,2,0,0,0,2,57,2,2,4,NA
+70399,7,2,1,24,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14379.41014,15757.242021,1,94,7,3,0.9,4,1,0,0,0,1,24,NA,NA,5,NA
+70400,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,18723.98095,17927.802584,2,101,7,7,1.3,5,5,2,0,1,2,50,1,4,1,3
+70401,7,2,1,47,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,14879.667962,15815.934168,3,91,15,15,5,3,3,0,1,0,1,47,2,5,1,5
+70402,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,8796.577101,9772.20689,2,95,5,5,1.32,2,2,0,0,2,1,80,1,3,1,3
+70403,7,2,1,13,NA,1,1,1,13,159,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,19242.904841,19537.039374,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+70404,7,2,2,7,NA,2,2,2,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20753.369981,21666.802981,1,97,5,5,1.04,4,4,1,1,0,1,32,1,3,6,NA
+70405,7,2,2,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,118611.064701,118209.809508,1,91,15,15,5,2,2,0,0,2,1,63,1,4,1,5
+70406,7,2,2,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,31454.59168,33587.963223,1,94,4,4,1,3,3,0,1,0,2,41,1,4,5,NA
+70407,7,1,2,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9492.170663,0,1,90,4,4,0.67,5,5,3,0,0,2,32,2,3,3,NA
+70408,7,2,1,33,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,20803.970543,21557.298406,1,93,14,14,5,2,2,0,0,1,1,33,2,5,5,NA
+70409,7,2,1,44,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,130374.456976,1,101,6,6,1.31,3,3,0,0,1,2,80,1,1,2,NA
+70410,7,2,1,6,NA,2,2,1,6,80,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9741.585979,9800.624884,2,103,12,2,0.59,5,2,0,1,0,2,47,NA,NA,3,NA
+70411,7,2,2,23,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,50915.06085,50693.303376,3,92,5,5,0.81,5,5,3,0,0,2,23,1,4,5,NA
+70412,7,2,1,14,NA,1,1,2,15,180,NA,NA,1,1,NA,10,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,20398.562455,21727.336109,2,94,6,6,1.34,4,4,0,2,0,1,37,2,4,1,2
+70413,7,2,1,51,NA,5,6,2,NA,NA,2,NA,2,1,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9889.944368,10305.962662,1,99,10,10,3.51,3,3,0,1,0,2,44,1,3,1,5
+70414,7,2,2,25,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,128171.594518,136210.394705,1,93,15,15,5,2,2,0,0,0,1,28,1,5,6,NA
+70415,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,52752.276869,57774.89643,1,98,12,12,NA,1,1,0,0,1,2,80,1,5,2,NA
+70416,7,2,1,25,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,4,5,NA,1,2,2,1,2,2,1,2,1,2,38474.772527,40200.135096,2,93,7,2,0.81,3,1,0,0,0,1,25,2,4,5,NA
+70417,7,2,1,80,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,2,2,2,1,2,2,1,2,2,NA,14200.083364,15006.095575,3,92,4,4,1.65,1,1,0,0,1,1,80,1,1,3,NA
+70418,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,47566.45715,0,1,90,6,6,2.86,1,1,0,0,1,2,80,1,4,5,NA
+70419,7,2,1,2,NA,2,2,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,2,2,1,NA,NA,NA,NA,10244.841997,10258.685193,1,96,5,5,0.94,4,4,2,0,0,1,32,2,3,1,4
+70420,7,2,2,55,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,2,NA,2,2,2,2,2,2,NA,NA,NA,NA,24004.6026,24129.784561,2,91,4,4,0.43,7,7,0,1,1,1,41,2,1,4,NA
+70421,7,2,2,51,NA,2,2,2,NA,NA,2,NA,2,2,7,NA,2,3,NA,1,2,2,1,2,2,1,2,2,2,17054.056149,20556.446647,2,90,8,8,2.7,3,3,0,1,0,2,31,1,5,4,NA
+70422,7,2,2,9,NA,2,2,2,9,114,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15148.721588,16989.423356,2,91,2,2,0.22,4,4,1,1,0,2,48,2,9,5,NA
+70423,7,2,1,52,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32461.799549,39033.401297,1,101,5,5,0.89,5,5,1,0,0,1,25,1,2,77,NA
+70424,7,2,1,28,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,1,2,2,1,2,2,1,41271.869706,41834.437137,1,96,9,9,3.97,2,2,0,0,0,1,28,2,2,6,NA
+70425,7,2,1,77,NA,3,3,2,NA,NA,1,2,2,1,9,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,63054.867183,67969.319596,1,90,8,8,3.3,2,2,0,0,2,1,77,2,2,1,5
+70426,7,2,1,66,NA,1,1,1,NA,NA,2,NA,2,1,8,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,14488.953694,14844.946966,3,92,8,8,1.85,5,5,1,0,2,1,66,2,1,1,1
+70427,7,2,2,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,75152.05379,75383.121447,2,92,15,15,5,2,2,0,0,0,1,37,1,5,1,5
+70428,7,2,2,0,4,3,3,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10025.884543,10365.226844,3,92,6,6,0.74,7,7,2,1,0,2,46,1,2,1,4
+70429,7,2,2,44,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,134587.275919,139488.984678,1,92,8,8,2.17,4,4,0,1,2,2,80,1,3,2,NA
+70430,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,97705.030285,101386.34066,2,94,14,14,4.71,3,3,1,0,0,1,35,1,5,1,5
+70431,7,2,2,20,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,2,5,2,1,2,2,2,2,2,1,2,2,1,32455.694722,32314.335903,2,103,5,5,0.65,6,6,1,0,1,2,61,2,1,2,NA
+70432,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,145772.192378,152604.524555,1,95,15,15,5,4,4,0,2,0,2,42,1,5,1,5
+70433,7,2,1,16,NA,1,1,1,16,199,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,22768.423624,22944.003607,2,98,14,14,2.87,5,5,0,3,0,2,34,1,2,1,2
+70434,7,2,2,0,2,3,3,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18869.566209,19997.704869,2,92,15,15,5,3,3,1,0,0,1,48,1,5,1,5
+70435,7,2,2,17,NA,2,2,1,17,214,2,NA,2,1,4,14,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,14984.257718,15414.270546,2,92,10,7,2.38,3,2,0,1,0,1,29,2,4,6,NA
+70436,7,2,2,8,NA,4,4,1,8,103,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11116.391625,11717.604707,2,98,6,6,1,5,5,2,1,0,2,31,1,4,6,NA
+70437,7,2,1,43,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,2,2,2,2,2,2,2,34153.424332,40514.279864,1,100,7,7,1.3,5,5,0,3,0,1,43,2,2,1,4
+70438,7,2,1,42,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,138834.18124,143254.613293,1,100,15,15,5,4,4,0,2,0,2,47,1,5,1,5
+70439,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,76027.409363,76707.146234,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+70440,7,2,1,14,NA,4,4,2,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10081.970102,10544.115448,2,99,4,4,0.78,4,4,0,2,0,2,45,1,3,5,NA
+70441,7,2,2,2,NA,2,2,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8134.473424,8243.872451,2,99,5,5,1.26,3,3,1,0,0,1,24,2,2,1,1
+70442,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,41309.21018,41341.432201,1,100,4,4,1.19,2,2,0,0,1,1,62,1,5,1,5
+70443,7,2,2,52,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,33018.025291,33777.066037,1,92,8,8,3.3,2,2,0,0,0,1,40,2,5,1,5
+70444,7,2,1,13,NA,1,1,1,13,160,NA,NA,2,2,3,6,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,27378.670648,27589.802811,2,96,3,3,0.95,2,2,0,1,0,2,38,2,3,3,NA
+70445,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,52814.190351,53378.216173,1,98,12,12,NA,2,2,0,0,2,1,65,1,4,1,3
+70446,7,2,2,45,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,4,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,31235.666551,31398.557894,3,92,6,6,1,6,6,1,1,0,1,42,2,1,1,4
+70447,7,2,1,29,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,23484.626749,23642.328802,2,96,14,14,4.26,3,3,0,0,0,1,20,1,4,5,NA
+70448,7,2,1,29,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,3,4,NA,2,2,2,1,2,2,2,2,2,2,40078.999044,48909.175469,3,91,6,6,0.89,7,7,1,1,0,1,59,2,1,1,1
+70449,7,2,2,12,NA,2,2,1,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20128.183753,21457.209224,2,100,14,14,3.58,4,4,0,2,0,2,40,NA,NA,1,4
+70450,7,2,1,15,NA,3,3,1,15,190,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,63633.689496,63154.346062,2,100,10,10,3.13,4,4,0,2,0,1,45,1,4,1,4
+70451,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,17286.767396,17743.751095,2,100,2,2,0.38,3,3,0,2,0,2,35,1,4,5,NA
+70452,7,2,1,17,NA,5,6,2,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,6232.587755,6841.667882,1,91,7,7,1.57,4,4,0,3,0,2,38,2,2,3,NA
+70453,7,2,2,1,22,4,4,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8221.160724,9187.103228,1,100,8,8,2.36,3,3,1,0,1,2,60,1,3,3,NA
+70454,7,2,2,2,NA,2,2,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8200.025088,8716.981559,1,93,14,8,2.01,5,4,1,0,0,2,22,2,1,6,NA
+70455,7,2,1,69,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,4526.005699,5855.572964,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+70456,7,2,1,0,4,1,1,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7594.427215,7594.567194,1,98,15,15,5,3,3,1,0,0,1,38,1,4,1,5
+70457,7,2,2,40,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,1,1,2,1,1,2,2,3,16797.688743,17484.356971,3,91,15,15,5,3,3,0,1,0,2,40,2,5,1,5
+70458,7,2,2,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,26465.930618,26765.106529,2,100,10,10,2.33,6,6,0,2,2,2,35,1,2,5,NA
+70459,7,2,2,4,NA,4,4,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9727.363166,10466.751274,2,97,6,6,1.02,6,6,1,2,0,1,37,1,3,1,3
+70460,7,2,2,5,NA,4,4,2,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9156.613403,9392.96113,2,99,7,5,1.59,3,2,1,0,0,2,23,1,4,5,NA
+70461,7,1,2,5,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13196.707564,0,2,96,3,3,0.24,7,7,2,3,1,2,40,1,3,3,NA
+70462,7,2,1,8,NA,2,2,2,8,107,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13232.943476,13442.347662,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+70463,7,2,2,7,NA,3,3,2,8,96,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,35359.674949,35088.43355,1,99,10,10,2.48,5,5,2,1,0,1,33,1,5,1,5
+70464,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,11523.037911,11803.949722,2,97,4,4,1.65,1,1,0,0,1,2,60,1,4,5,NA
+70465,7,2,1,4,NA,5,6,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9239.758777,10363.400393,2,91,14,14,3.47,4,4,1,1,0,2,40,2,5,1,NA
+70466,7,2,2,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,1,1,2,2,1,2,2,NA,NA,NA,NA,43813.24867,46530.283453,1,98,1,1,0.16,3,3,1,0,0,1,28,1,2,6,NA
+70467,7,2,1,12,NA,2,2,2,12,146,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21399.234084,21768.65302,1,91,15,15,5,4,4,0,2,0,1,49,1,5,1,5
+70468,7,2,2,0,8,5,7,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5820.770257,5884.32621,1,95,2,2,0.47,3,3,2,0,0,2,24,1,4,5,NA
+70469,7,2,2,0,4,2,2,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4807.611315,4659.866106,1,103,4,4,0.54,7,7,2,2,0,2,35,2,1,6,NA
+70470,7,2,2,43,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,15552.65085,16188.423507,2,92,15,15,5,3,3,0,1,0,1,45,2,4,1,4
+70471,7,2,2,16,NA,5,7,2,16,198,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,12440.215685,2,95,6,6,1.31,3,3,0,1,0,1,39,1,3,1,4
+70472,7,2,1,12,NA,3,3,2,13,157,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20152.417898,19887.568869,2,94,7,7,1.18,7,7,1,4,0,2,31,1,4,6,NA
+70473,7,2,2,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,126117.5094,130777.186989,2,98,10,10,3.78,3,3,0,0,0,1,53,1,3,1,4
+70474,7,2,1,44,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,18116.816149,22904.074887,2,90,8,8,3.4,2,2,0,0,0,1,44,1,3,6,NA
+70475,7,2,1,16,NA,5,6,2,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,6269.080236,6655.935643,1,93,NA,NA,NA,3,3,0,1,0,1,53,NA,NA,1,NA
+70476,7,2,1,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16905.961576,20936.167849,2,94,7,7,1.62,5,5,0,3,0,1,30,1,2,1,9
+70477,7,2,1,9,NA,2,2,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15353.368573,15360.170072,1,97,7,7,1.74,4,4,0,3,0,2,32,1,4,5,NA
+70478,7,2,2,34,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,17430.913214,17550.603029,1,90,15,15,5,2,2,0,0,0,2,34,2,5,1,5
+70479,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,17397.027021,18573.153951,3,91,2,2,0.73,1,1,0,0,0,2,25,2,4,5,NA
+70480,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,136832.42119,141624.96623,2,91,15,15,5,2,2,0,0,1,2,60,1,5,1,5
+70481,7,2,1,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,32946.178621,33458.203523,2,96,3,3,0.53,5,5,3,0,0,2,26,1,4,1,4
+70482,7,2,1,17,NA,5,7,2,17,207,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11712.675291,11995.185902,3,91,15,15,4.47,4,4,0,1,2,2,79,1,4,3,NA
+70483,7,2,1,20,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15740.336751,16168.874324,2,91,10,10,2.59,5,5,0,1,0,1,48,NA,NA,6,NA
+70484,7,2,2,19,NA,1,1,2,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19557.287652,19933.266032,2,94,5,5,1.08,3,3,0,1,0,2,37,2,2,4,NA
+70485,7,2,1,34,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14790.183811,15030.369455,3,90,14,14,4.71,3,3,0,0,2,1,64,NA,NA,1,NA
+70486,7,2,1,10,NA,2,2,2,10,130,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13285.093011,13365.60735,2,94,2,2,0.41,2,2,0,1,1,2,66,2,3,5,NA
+70487,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,12935.870815,12581.846775,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+70488,7,2,2,6,NA,4,4,2,6,83,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,1,2,NA,NA,NA,NA,7814.742747,7984.793423,2,95,2,2,0.26,4,4,0,2,0,2,44,NA,NA,5,NA
+70489,7,2,1,40,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,5,1,NA,2,2,2,1,2,2,2,2,1,2,36225.654087,36329.951191,2,93,7,7,1.83,3,3,1,0,0,1,40,2,5,1,4
+70490,7,1,1,43,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,140932.152825,0,1,97,15,15,5,5,5,2,0,1,1,43,1,5,1,5
+70491,7,2,2,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8833.042831,8971.325329,1,99,4,4,0.53,7,7,3,1,0,2,26,1,1,5,NA
+70492,7,2,2,0,0,4,4,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3516.231705,3777.543312,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+70493,7,2,1,52,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,19541.667675,20187.705923,2,95,2,2,0.67,1,1,0,0,0,1,52,1,2,5,NA
+70494,7,2,1,25,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,44831.646235,2,98,7,7,1.97,4,4,0,1,0,1,40,1,3,1,3
+70495,7,2,1,68,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26026.192556,26360.023593,1,94,3,3,0.86,2,2,0,0,2,1,68,1,4,1,2
+70496,7,1,2,30,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,39483.840194,0,2,102,14,14,4.86,3,3,1,0,0,1,30,1,5,1,5
+70497,7,2,2,3,NA,4,4,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,10437.988787,11664.396755,2,100,NA,NA,NA,4,4,1,0,0,2,38,NA,NA,77,NA
+70498,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,98514.948291,98181.677236,1,99,10,10,4.89,2,2,0,0,2,2,63,1,5,5,NA
+70499,7,2,2,47,NA,1,1,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,2,2,2,2,2,2,1,2,2,2,23968.560941,24093.554947,2,90,6,6,1.15,5,5,0,2,0,2,47,2,1,1,5
+70500,7,2,1,15,NA,1,1,1,15,183,NA,NA,1,1,NA,9,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,24902.864049,25283.513039,1,94,8,8,1.85,5,5,0,2,0,1,44,2,1,6,NA
+70501,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105873.555835,106383.882408,1,98,15,15,5,5,5,0,1,0,1,53,1,5,1,5
+70502,7,2,1,71,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,NA,48254.793439,51076.634961,2,100,14,10,5,2,1,0,0,2,1,71,1,2,6,NA
+70503,7,2,2,18,NA,4,4,1,18,224,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,99,1,0.09,4,1,0,0,0,2,18,1,4,NA,NA
+70504,7,2,1,59,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,19671.580166,20292.088214,2,99,7,7,2.31,2,2,0,0,1,1,62,1,3,5,NA
+70505,7,2,2,40,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,3,1,2,2,2,2,2,2,2,1,2,2,2,23968.560941,24556.443674,2,90,99,99,NA,5,5,1,1,0,2,40,2,3,1,1
+70506,7,2,2,9,NA,4,4,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9153.600624,9398.485171,1,100,8,8,1.61,6,6,1,3,0,1,29,1,5,6,NA
+70507,7,2,2,80,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,19066.820866,21138.377975,2,99,6,6,2.69,1,1,0,0,1,2,80,2,3,5,NA
+70508,7,1,1,0,3,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6246.568228,0,2,95,9,9,1.81,6,6,1,1,0,2,56,1,4,3,NA
+70509,7,2,1,8,NA,2,2,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13395.612951,13472.514863,2,102,15,12,NA,5,4,0,3,0,1,42,2,4,6,NA
+70510,7,2,1,6,NA,5,6,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6905.911299,7686.337387,2,90,4,4,0.81,3,3,0,1,0,2,41,2,2,6,NA
+70511,7,2,1,36,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17643.563124,18510.507534,3,91,15,15,5,4,4,2,0,0,1,36,2,5,1,5
+70512,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,2,1,2,2,1,2,2,1,2,2,3,16929.836231,20416.10433,2,101,5,3,0.92,2,1,0,0,0,2,20,NA,NA,5,NA
+70513,7,1,1,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15682.233511,0,1,101,6,6,1.78,2,2,0,0,2,1,80,1,2,1,1
+70514,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,52160.006964,54246.901852,1,100,4,1,0,3,1,0,0,0,1,22,1,4,6,NA
+70515,7,1,1,5,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7843.21745,0,1,96,12,12,NA,5,5,2,0,1,2,63,2,5,3,NA
+70516,7,2,1,11,NA,4,4,2,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12176.538896,12609.8265,2,97,3,3,0.46,5,5,0,3,0,1,40,1,2,1,3
+70517,7,2,1,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19440.793325,20264.695737,2,95,14,14,3.47,4,4,0,0,0,2,45,1,4,1,4
+70518,7,2,2,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,7032.269464,7566.800387,2,99,3,3,0.44,5,5,1,1,0,2,53,1,4,1,3
+70519,7,2,1,5,NA,5,6,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8981.553859,10073.795326,3,91,15,15,5,4,4,2,0,0,1,36,2,5,1,5
+70520,7,2,1,67,NA,1,1,1,NA,NA,1,1,2,1,8,NA,3,1,NA,1,2,2,1,2,2,2,2,2,2,11992.012141,12562.386395,2,102,5,5,1.36,2,2,0,0,2,2,66,2,2,1,3
+70521,7,2,1,53,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19084.761772,19687.829911,1,91,15,15,5,4,4,0,2,0,2,52,1,5,1,5
+70522,7,2,2,21,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,41953.42893,43056.968872,2,99,4,1,0.08,4,1,0,0,0,2,21,NA,NA,5,NA
+70523,7,2,1,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,86986.68246,93900.677523,2,94,14,14,3.36,4,4,2,0,0,1,31,1,3,1,5
+70524,7,2,1,0,6,3,3,2,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17113.022485,16817.175522,1,91,15,15,5,3,3,1,0,0,1,41,1,5,1,5
+70525,7,2,1,63,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,7410.50521,7700.344649,2,96,2,2,0.71,1,1,0,0,1,1,63,1,3,3,NA
+70526,7,2,2,10,NA,5,7,1,10,131,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8063.227462,8462.212273,3,91,14,14,3.8,4,4,0,2,0,1,47,1,5,1,5
+70527,7,2,1,63,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,1,2,2,1,1,2,2,2,9235.951997,9462.879133,2,103,7,7,1.89,3,3,0,0,1,1,63,2,2,1,9
+70528,7,2,1,15,NA,4,4,1,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19163.050164,19302.61536,2,102,8,8,3.1,2,2,0,1,0,1,40,1,4,3,NA
+70529,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,36244.629058,37702.525304,1,98,3,3,0.73,2,2,0,0,0,1,49,1,2,1,2
+70530,7,2,2,1,17,4,4,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6765.566493,7560.484444,2,103,6,6,1.3,4,4,1,1,0,2,26,1,4,1,3
+70531,7,2,1,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,30152.053647,30985.296608,2,101,10,10,3.89,3,3,0,1,0,2,49,1,4,1,3
+70532,7,2,2,9,NA,1,1,1,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20495.125801,21365.773499,3,92,8,8,1.55,6,6,1,3,0,2,38,1,5,1,4
+70533,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,196284.082305,1,91,7,7,3.31,1,1,0,0,0,2,51,1,5,3,NA
+70534,7,2,2,48,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,2,6,NA,2,2,2,NA,NA,NA,2,2,2,2,46417.079566,59177.283457,1,97,NA,NA,NA,2,1,0,0,0,1,30,NA,NA,6,NA
+70535,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,28813.038041,29206.765831,1,94,5,5,0.87,4,4,0,1,0,1,40,1,5,1,5
+70536,7,2,1,30,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,18838.303827,25198.281368,2,93,6,6,1.95,2,2,0,1,0,1,30,1,4,5,NA
+70537,7,2,1,26,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37326.925742,39362.952436,2,91,6,77,NA,4,1,0,0,0,1,25,NA,NA,5,NA
+70538,7,2,2,11,NA,1,1,1,11,134,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,19059.339877,19554.022514,1,95,4,4,0.68,5,5,0,1,0,2,38,2,3,4,NA
+70539,7,1,1,22,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,13205.475858,0,1,102,1,1,0,2,1,0,0,0,1,18,NA,NA,NA,NA
+70540,7,2,1,18,NA,3,3,1,18,227,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,105662.708061,105798.100528,1,94,12,12,NA,2,2,0,0,0,1,18,1,3,NA,NA
+70541,7,2,2,13,NA,2,2,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19836.724169,20587.964512,1,98,4,4,0.75,4,4,0,1,0,2,48,1,2,1,3
+70542,7,2,2,14,NA,4,4,2,14,177,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11956.705842,11923.015236,1,96,15,15,5,5,5,0,3,0,2,47,1,5,1,5
+70543,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,1,2,NA,11684.182359,12558.119387,2,98,6,6,1.35,3,3,0,0,2,1,79,NA,NA,1,2
+70544,7,2,1,77,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,7608.031426,7606.53553,1,99,6,6,1.34,4,4,0,1,1,1,77,1,3,1,5
+70545,7,2,1,31,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32856.012738,39257.112915,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+70546,7,2,2,0,8,4,4,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4418.651245,4866.131244,2,95,3,3,0.43,4,4,2,0,0,2,23,1,2,5,NA
+70547,7,2,2,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,23204.787354,24686.898455,1,98,3,3,0.43,4,4,0,1,0,2,39,1,2,5,NA
+70548,7,2,1,34,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,3,4,NA,2,2,2,1,2,2,2,2,2,2,54095.173154,53686.899782,1,92,7,7,1.48,5,5,0,1,0,1,42,1,5,1,4
+70549,7,2,2,4,NA,1,1,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12468.876716,13254.955592,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+70550,7,2,1,76,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,14073.595908,14398.283636,1,92,14,14,5,3,3,0,0,2,1,76,1,4,1,4
+70551,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13209.159597,13903.918018,2,95,3,3,0.52,3,3,1,0,0,1,37,1,4,1,4
+70552,7,2,1,3,NA,4,4,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9969.040341,10992.497123,2,97,4,4,0.84,3,3,1,0,1,2,68,1,4,2,NA
+70553,7,1,1,2,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6347.215153,0,2,95,1,1,0.09,7,7,2,4,1,2,60,1,3,5,NA
+70554,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,71034.153987,74518.988514,1,98,15,15,5,4,4,1,1,0,1,40,1,4,1,5
+70555,7,2,2,22,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,16239.242782,17706.77205,2,101,12,12,NA,4,4,0,0,0,1,57,NA,NA,1,NA
+70556,7,2,1,2,NA,4,4,2,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5125.369941,5513.229844,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+70557,7,2,1,1,22,4,4,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5547.430651,5776.651218,1,96,8,8,1.61,6,6,3,0,0,1,33,2,5,1,4
+70558,7,2,1,14,NA,4,4,2,14,172,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,6,6,1.21,4,4,0,2,0,2,41,1,4,4,NA
+70559,7,2,2,1,22,4,4,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8221.160724,8586.221973,1,100,1,1,0,4,4,1,2,0,2,35,1,2,5,NA
+70560,7,1,1,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7843.21745,0,1,96,12,12,NA,5,5,2,0,1,2,63,2,5,3,NA
+70561,7,2,1,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,18533.049642,19320.837782,1,99,14,14,4.21,4,4,0,2,0,2,44,1,5,1,5
+70562,7,2,1,52,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,1,2,2,1,20958.247774,21169.083688,2,103,3,1,0.36,2,1,0,0,1,1,52,1,2,6,NA
+70563,7,2,1,0,6,5,7,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6395.296043,7073.813977,2,95,4,4,0.79,3,3,1,0,0,1,50,1,4,6,NA
+70564,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,8796.577101,9772.20689,2,95,3,3,1.03,1,1,0,0,1,1,80,1,3,2,NA
+70565,7,2,2,76,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,18693.365067,19341.691824,1,92,7,7,3.58,1,1,0,0,1,2,76,1,4,3,NA
+70566,7,2,2,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,28033.589927,29297.386717,2,95,2,2,0.26,3,3,0,2,0,2,31,1,3,5,NA
+70567,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,28280.669788,30348.532296,1,99,13,13,NA,2,2,0,0,2,1,80,1,2,1,3
+70568,7,2,1,67,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,8561.661692,8628.633702,2,99,99,99,NA,3,3,0,1,1,1,67,1,4,2,NA
+70569,7,2,1,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22570.662508,23265.895325,1,95,12,12,NA,6,6,2,0,0,2,42,1,2,1,5
+70570,7,2,1,62,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,8621.533452,8958.738589,2,99,7,7,2.31,2,2,0,0,1,1,62,1,3,5,NA
+70571,7,2,1,0,11,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5707.173345,6119.621796,1,90,2,2,0.22,5,5,1,1,2,1,44,2,4,5,NA
+70572,7,2,2,2,NA,2,2,2,2,25,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10806.348349,11487.615985,2,91,99,99,NA,6,6,1,3,0,2,20,2,2,5,NA
+70573,7,2,1,24,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,59473.789098,60284.462735,3,92,7,7,2.78,2,2,0,0,0,1,24,1,3,5,NA
+70574,7,2,2,36,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,42468.064168,44382.588967,2,101,1,1,0.27,3,3,0,2,0,2,36,1,3,5,NA
+70575,7,2,2,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,149804.257477,155051.140087,1,97,9,9,4.92,1,1,0,0,1,2,69,1,4,3,NA
+70576,7,1,1,31,NA,5,6,NA,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19091.246741,0,1,91,9,9,2.97,3,3,1,0,0,1,31,2,5,1,5
+70577,7,2,2,80,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,4,77,NA,1,2,1,1,2,1,1,2,1,NA,14314.616082,14811.078253,2,92,77,77,NA,1,1,0,0,1,2,80,2,4,77,NA
+70578,7,2,1,16,NA,4,4,2,16,194,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12721.673656,12746.825446,2,97,7,7,2.65,2,2,0,1,1,2,61,1,4,3,NA
+70579,7,2,2,4,NA,3,3,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,85219.527381,94056.007225,1,97,15,15,5,4,4,1,1,0,2,33,1,5,1,3
+70580,7,2,1,58,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,19581.573846,19613.097963,1,90,6,6,2.12,2,2,0,0,0,2,55,1,3,1,3
+70581,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,128226.88751,129797.060324,1,95,7,7,2.86,2,2,0,0,1,1,59,1,2,1,2
+70582,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,4,6,2,2,2,2,2,2,2,2,2,2,2,35464.8385,35311.584148,2,96,5,5,0.89,4,4,1,1,0,2,36,2,4,6,NA
+70583,7,2,1,14,NA,2,2,1,14,170,NA,NA,2,2,4,8,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,22721.243258,22896.459408,2,102,8,8,1.09,7,7,1,3,0,2,33,2,1,6,NA
+70584,7,2,2,36,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20039.469886,21581.359058,1,97,15,15,5,4,4,2,0,0,1,40,2,5,1,5
+70585,7,2,1,5,NA,5,7,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6138.820061,6435.022409,1,99,14,14,3.94,4,4,1,1,0,1,43,1,4,1,5
+70586,7,2,2,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,24905.670199,25753.835465,2,101,10,10,2.91,4,4,0,1,0,2,51,1,2,5,NA
+70587,7,2,2,50,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,2,5,NA,2,2,2,2,2,2,NA,NA,NA,NA,17054.056149,17142.991601,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+70588,7,2,1,0,8,4,4,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5836.706781,6455.960401,2,93,12,12,NA,4,4,1,1,0,2,27,2,4,1,4
+70589,7,2,2,10,NA,4,4,2,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7889.400564,8100.464206,2,99,99,99,NA,3,3,0,1,1,1,67,1,4,2,NA
+70590,7,2,2,51,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,27551.607232,31605.843936,1,90,12,9,5,2,1,0,0,0,2,51,1,3,3,NA
+70591,7,1,2,65,NA,2,2,NA,NA,NA,2,NA,2,1,8,NA,1,3,NA,2,2,2,2,2,2,NA,NA,NA,NA,9716.805546,0,2,90,4,4,1.65,1,1,0,0,1,2,65,2,1,3,NA
+70592,7,2,1,2,NA,4,4,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7561.636249,7792.598574,2,98,2,2,0.31,4,4,2,1,0,2,27,1,2,4,NA
+70593,7,2,1,80,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,7608.031426,7725.81565,1,99,5,5,1.84,1,1,0,0,1,1,80,1,3,2,NA
+70594,7,2,2,3,NA,4,4,1,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11729.417673,12620.984256,2,96,7,7,1.49,5,5,2,1,0,1,51,1,5,1,3
+70595,7,2,2,60,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,13057.178942,13648.591881,1,102,77,77,NA,6,6,0,2,1,2,37,1,4,1,4
+70596,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,85420.170155,89610.762168,1,91,15,15,5,3,3,1,0,0,1,36,1,4,1,5
+70597,7,2,2,49,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,35126.205635,35814.074718,1,95,6,6,0.96,5,5,1,0,1,2,69,1,1,2,NA
+70598,7,2,2,28,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,12426.473257,13266.565637,2,92,15,14,5,2,1,0,0,0,1,27,1,5,6,NA
+70599,7,2,1,54,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,36134.528328,36424.033955,2,101,6,6,2.04,2,2,0,0,0,2,51,1,5,1,4
+70600,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,2,7,NA,1,2,NA,2,2,2,1,2,2,2,2,2,NA,18142.687884,20262.262283,2,93,5,5,1.07,3,3,0,0,1,2,44,2,1,1,1
+70601,7,2,1,0,2,1,1,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,9064.168162,9064.335231,3,92,3,3,0.48,4,4,2,0,0,1,21,2,1,1,2
+70602,7,2,2,10,NA,3,3,2,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60197.256541,63931.531988,2,91,15,15,5,4,4,0,2,0,1,53,1,5,1,4
+70603,7,2,1,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,127000.852889,130891.431194,2,92,15,15,5,2,1,0,0,0,1,57,1,5,6,NA
+70604,7,2,1,39,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,13855.593765,14267.864487,2,92,15,15,5,3,3,1,0,0,1,39,2,5,1,5
+70605,7,2,1,56,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20020.224623,20868.683381,1,96,15,15,5,4,4,0,1,0,1,56,1,4,1,5
+70606,7,2,1,80,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,16325.758272,16610.353254,2,93,8,8,2.17,4,4,0,0,3,1,80,2,2,1,2
+70607,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19870.689174,19814.851418,1,96,15,15,5,4,4,0,2,0,1,46,1,5,1,5
+70608,7,1,2,68,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,54217.266426,0,1,101,4,4,0.99,2,2,0,0,2,1,77,1,3,1,3
+70609,7,2,2,1,23,1,1,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8832.868731,9513.039058,1,102,13,13,NA,7,7,3,1,2,2,62,2,1,1,2
+70610,7,2,2,1,21,1,1,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11414.885224,11453.86318,1,94,4,4,0.5,7,7,2,3,0,2,32,1,4,6,NA
+70611,7,2,2,12,NA,3,3,2,12,150,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,101168.631125,108085.531312,1,99,6,6,1.52,4,4,0,2,0,2,43,1,3,5,NA
+70612,7,2,2,5,NA,2,2,1,5,60,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13647.772496,14082.018801,2,93,7,7,1.99,3,3,1,1,0,2,38,1,3,4,NA
+70613,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,1,1,NA,1,2,1,1,2,2,1,2,1,3,12649.084278,13204.362319,3,90,8,3,1.25,3,1,0,0,1,2,68,2,2,4,NA
+70614,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,1,1,2,2,1,2,2,1,2,2,1,22225.098465,28674.927734,2,97,2,2,0.34,2,2,1,0,0,2,20,1,3,5,NA
+70615,7,2,1,8,NA,1,1,2,8,106,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,10591.186197,11123.758959,1,90,6,6,1.11,5,5,1,2,0,1,30,2,1,6,NA
+70616,7,2,1,17,NA,5,6,2,17,214,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8869.170018,9434.816848,1,90,99,99,NA,3,3,0,1,1,1,60,1,5,1,5
+70617,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,21857.756498,23340.233721,2,97,5,5,1.08,3,3,0,1,0,2,45,1,4,6,NA
+70618,7,2,1,13,NA,1,1,1,13,161,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,22768.423624,23297.239555,2,98,5,5,1.07,4,4,0,1,0,1,53,2,1,1,1
+70619,7,2,2,4,NA,1,1,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,16462.187772,17163.602129,3,92,12,12,NA,6,6,1,3,0,2,33,1,5,1,4
+70620,7,2,1,9,NA,1,1,2,9,113,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12477.812875,12364.493687,2,94,14,14,4.03,4,4,0,2,0,2,33,2,2,1,NA
+70621,7,2,1,10,NA,5,7,1,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9375.648038,9390.261055,1,102,6,6,2.11,2,2,0,1,0,2,32,1,4,99,NA
+70622,7,2,1,50,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,32274.162059,31800.588482,1,102,14,14,4.59,3,3,0,0,1,2,46,1,4,1,1
+70623,7,2,1,54,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,33691.382998,1,101,5,5,1.24,3,3,0,0,1,2,61,1,4,1,3
+70624,7,2,1,42,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,44926.921381,45636.571752,2,96,10,10,3.78,3,3,0,1,0,1,42,1,3,1,3
+70625,7,2,2,0,7,1,1,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5289.715679,5503.176989,1,90,3,3,0.23,7,7,3,1,1,2,35,2,2,5,NA
+70626,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,13076.210481,13450.719873,2,97,5,5,0.84,5,5,0,2,0,2,33,1,4,1,3
+70627,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,15497.844354,15606.405341,1,99,14,14,4.05,3,3,0,1,0,2,52,1,4,4,NA
+70628,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,47566.45715,54102.989916,1,90,7,7,3.22,1,1,0,0,1,2,80,1,4,2,NA
+70629,7,2,2,28,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,16131.725974,16820.579069,1,99,14,7,3.4,2,1,0,0,0,2,28,2,5,1,NA
+70630,7,2,1,11,NA,5,7,2,11,142,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,28060.491868,29807.930767,1,91,4,4,0.76,4,4,0,2,0,2,44,1,4,6,NA
+70631,7,1,1,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,46965.818538,0,1,101,10,10,4.3,2,2,0,0,2,1,80,1,2,1,4
+70632,7,2,2,0,0,3,3,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22123.758463,21522.828051,1,95,7,7,1.57,4,4,1,1,0,1,37,1,3,1,5
+70633,7,2,1,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,123771.419917,127885.04228,2,98,15,15,5,3,3,0,1,0,1,56,1,5,1,5
+70634,7,2,2,9,NA,1,1,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16664.698857,2,98,7,7,1.33,6,6,0,3,0,1,31,1,3,6,NA
+70635,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,129999.035519,131590.908764,1,98,12,12,NA,2,2,0,0,2,1,70,1,2,1,3
+70636,7,2,1,12,NA,3,3,2,12,147,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,93665.036597,95017.313859,1,91,10,10,2.77,5,5,0,3,0,1,43,1,5,1,5
+70637,7,2,1,8,NA,4,4,2,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8483.005475,8584.70346,1,96,7,7,1.52,4,4,0,2,0,2,30,2,4,1,5
+70638,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,32720.69734,34178.98356,1,95,5,5,1.08,3,3,0,1,0,1,53,1,4,1,4
+70639,7,2,2,34,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,2,2,2,2,1,2,2,1,2,2,2,40536.844796,42064.124353,3,91,7,7,1.42,6,6,1,3,0,1,37,2,1,1,1
+70640,7,2,1,6,NA,5,7,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11522.32071,1,97,15,15,4.77,4,4,1,1,0,1,41,2,4,1,5
+70641,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,7233.371983,2,95,2,2,0.75,1,1,0,0,1,1,63,1,1,5,NA
+70642,7,2,1,12,NA,3,3,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30943.024697,30709.935341,1,95,6,6,0.81,6,6,2,2,0,1,30,1,3,1,4
+70643,7,2,1,7,NA,3,3,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39769.597728,42234.385729,2,92,15,15,5,5,5,0,3,0,2,46,1,5,1,5
+70644,7,2,1,41,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,130826.463813,1,101,9,9,2.6,4,4,0,2,0,2,38,1,4,1,4
+70645,7,2,1,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11543.27965,11681.665542,1,98,1,1,0.03,3,3,0,2,0,2,38,1,4,5,NA
+70646,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,54095.581484,61529.339683,2,94,14,14,5,1,1,0,0,1,2,80,1,4,2,NA
+70647,7,2,2,38,NA,5,7,1,NA,NA,1,2,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,58826.425292,59007.297053,1,102,8,8,1.91,5,5,1,2,0,2,38,1,5,1,4
+70648,7,2,1,9,NA,1,1,1,9,113,NA,NA,1,1,NA,2,NA,NA,NA,2,NA,2,1,2,2,1,2,2,1,13395.612951,13714.015605,2,102,4,4,0.61,5,5,0,3,0,1,34,2,3,1,3
+70649,7,2,1,63,NA,2,2,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,9745.303498,9901.410815,2,98,15,15,4.56,4,4,0,0,3,1,80,1,1,1,NA
+70650,7,2,1,2,NA,3,3,2,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,43818.485,49437.664471,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+70651,7,2,2,3,NA,5,7,1,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10292.958129,10862.428258,3,91,10,10,2.48,5,5,2,1,0,2,27,1,2,1,4
+70652,7,2,2,16,NA,3,3,2,16,199,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,31716.869763,32155.152073,1,95,NA,NA,NA,2,2,0,1,0,2,51,NA,NA,3,NA
+70653,7,2,1,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,137286.989143,140204.788291,3,91,14,14,5,1,1,0,0,0,1,43,1,5,5,NA
+70654,7,2,2,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,52280.406546,52747.829022,1,95,6,4,1.47,2,1,0,0,0,2,28,1,2,6,NA
+70655,7,2,2,17,NA,3,3,2,17,206,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,103298.858809,106172.653059,1,95,9,9,2.66,4,4,0,2,0,1,45,1,3,1,3
+70656,7,2,2,2,NA,1,1,1,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12260.86913,13033.834526,3,92,8,8,1.85,5,5,1,0,2,1,66,2,1,1,1
+70657,7,2,1,0,8,3,3,2,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,22595.741117,22205.109866,1,93,15,15,5,3,2,1,0,0,2,35,1,5,6,NA
+70658,7,2,1,16,NA,1,1,1,16,200,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,27186.265479,27776.842846,1,92,14,9,3.77,3,2,0,1,0,2,39,2,4,6,NA
+70659,7,2,2,0,5,1,1,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,6787.112205,7238.162123,2,94,99,99,NA,3,3,1,0,0,2,18,2,2,NA,NA
+70660,7,2,2,19,NA,1,1,1,19,230,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12996.753859,13369.730018,2,103,10,10,1.63,7,7,1,4,0,1,31,NA,NA,1,4
+70661,7,2,2,65,NA,2,2,1,NA,NA,2,NA,2,1,2,NA,4,1,NA,2,2,2,1,2,2,2,2,2,2,15876.871857,16539.500436,2,102,15,15,5,3,3,0,0,2,1,36,2,5,5,NA
+70662,7,2,1,11,NA,3,3,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,42986.51011,47590.55439,2,94,7,7,1.17,6,6,0,3,0,1,40,1,3,1,5
+70663,7,2,2,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,3,3,1,2,2,1,2,2,1,2,2,1,23740.353448,26106.078779,2,99,6,6,2.6,1,1,0,0,0,2,29,1,5,3,NA
+70664,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,39561.827371,40611.330327,1,92,4,4,1.16,2,2,0,0,1,1,56,1,2,1,3
+70665,7,2,2,13,NA,4,4,2,13,164,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14241.099758,14200.972378,1,96,15,15,5,3,3,0,1,0,2,37,1,4,1,4
+70666,7,2,2,3,NA,1,1,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16462.187772,17500.018071,3,92,3,3,0.52,5,5,2,1,0,2,29,2,1,1,3
+70667,7,2,1,5,NA,3,3,1,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,73403.03993,82816.073154,1,92,8,8,1.45,6,6,1,3,0,1,36,1,3,1,4
+70668,7,2,2,41,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,1,1,2,2,2,2,2,2,2,NA,NA,NA,NA,37512.060155,37707.682357,1,92,6,6,0.93,5,5,0,2,0,1,47,2,1,1,1
+70669,7,2,2,3,NA,1,1,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13366.393396,14756.857003,2,94,8,8,2.33,4,4,2,0,0,1,24,1,2,6,NA
+70670,7,2,2,56,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,20130.149569,20235.126587,1,90,10,10,2.44,5,5,1,0,0,2,56,2,1,1,1
+70671,7,2,2,5,NA,4,4,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10683.855206,11939.151167,1,96,15,15,5,4,4,1,1,0,1,50,1,3,1,4
+70672,7,2,2,1,19,2,2,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9534.013652,9566.569071,2,97,12,6,0.89,7,7,3,0,0,2,26,2,1,6,NA
+70673,7,2,2,14,NA,5,7,2,14,173,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13124.737024,13189.842246,2,97,5,5,1.19,3,3,0,1,0,2,41,1,3,1,2
+70674,7,2,1,67,NA,3,3,2,NA,NA,2,NA,2,2,6,NA,5,2,NA,1,2,2,1,2,1,1,2,2,NA,21313.15842,21908.589414,2,90,2,2,0.87,1,1,0,0,1,1,67,2,5,2,NA
+70675,7,2,1,65,NA,5,6,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12579.986433,13271.133625,1,92,14,14,5,2,2,0,0,2,2,62,1,4,1,4
+70676,7,2,2,75,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,44856.466004,46390.551109,2,94,10,10,4.69,2,2,0,0,1,2,59,1,3,3,NA
+70677,7,2,1,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,11977.649578,12533.694613,2,92,15,14,5,3,1,0,0,0,1,29,1,5,5,NA
+70678,7,2,1,21,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,37235.048974,39044.599947,2,93,6,6,1.48,4,4,0,1,0,1,53,2,2,1,3
+70679,7,2,1,4,NA,3,3,2,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,97310.423154,111231.170223,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+70680,7,2,2,2,NA,4,4,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,6543.852411,7133.732612,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+70681,7,2,1,61,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,125558.167126,127168.668278,3,91,15,15,5,2,2,0,0,2,2,62,1,5,1,5
+70682,7,2,2,22,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,24654.107413,23441.410215,1,100,10,10,2.59,5,5,0,1,0,2,40,1,5,1,NA
+70683,7,2,2,0,1,1,1,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9767.083234,9561.595244,3,92,8,8,2.51,3,3,1,0,0,2,25,1,4,5,NA
+70684,7,2,1,58,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,13189.987508,14124.702096,1,90,77,77,NA,4,4,0,2,0,2,51,1,5,1,5
+70685,7,2,2,5,NA,2,2,2,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13474.744062,14324.236113,1,93,5,5,0.84,5,5,1,2,0,2,52,2,1,3,NA
+70686,7,2,2,48,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,29010.447112,28477.291441,1,100,15,15,5,2,2,0,0,1,2,48,1,5,5,NA
+70687,7,2,1,59,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,169076.731628,169683.640829,1,98,14,14,5,1,1,0,0,0,1,59,1,5,2,NA
+70688,7,2,2,13,NA,4,4,1,13,163,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16598.888685,16325.283055,2,102,4,4,0.53,6,6,2,2,0,2,27,1,2,1,2
+70689,7,2,1,56,NA,5,6,2,NA,NA,2,NA,2,2,99,NA,5,5,NA,1,2,1,1,2,1,1,2,1,3,11690.444016,11682.423918,3,90,77,77,NA,4,4,0,0,2,1,69,2,5,2,NA
+70690,7,2,2,13,NA,5,6,1,13,160,NA,NA,2,1,3,6,NA,NA,NA,1,1,2,1,2,1,1,2,2,NA,8878.081187,9740.503961,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+70691,7,2,2,14,NA,4,4,1,14,172,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15114.677795,15072.08892,2,102,6,6,1.22,5,5,0,2,0,2,42,1,4,1,4
+70692,7,2,1,3,NA,4,4,2,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,10259.663981,2,97,4,4,0.81,4,4,1,0,0,2,51,1,2,4,NA
+70693,7,2,1,5,NA,3,3,2,5,71,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,31190.854587,36547.404966,1,95,3,3,0.65,3,3,1,0,0,1,58,1,3,3,NA
+70694,7,2,1,8,NA,5,6,2,8,106,NA,NA,2,2,2,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,7923.925927,8546.646915,1,91,10,10,3.22,4,4,1,1,0,1,38,2,5,1,5
+70695,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,4,2,2,1,2,2,1,2,2,NA,NA,NA,NA,20039.469886,21176.851739,1,97,8,8,2.51,3,3,0,2,0,2,39,2,4,2,NA
+70696,7,2,2,47,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,3,1,NA,2,2,2,2,2,2,2,2,1,2,29105.053716,31982.273564,2,93,4,4,0.82,4,4,0,0,0,1,51,2,3,1,3
+70697,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,82917.179795,86800.065133,2,95,7,7,2.91,2,2,1,0,0,1,32,1,5,2,NA
+70698,7,2,1,1,16,5,6,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8674.760516,9729.693026,3,91,10,10,3.67,3,3,1,0,0,2,30,2,4,1,4
+70699,7,2,1,0,7,2,2,1,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4724.065742,4677.378988,2,93,77,77,NA,7,7,3,0,1,2,40,2,4,1,NA
+70700,7,2,1,0,2,2,2,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4854.394074,4994.155976,2,93,3,3,0.37,5,5,3,0,0,1,28,2,1,6,NA
+70701,7,2,1,24,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14782.636003,15547.771991,3,91,10,9,5,2,1,0,0,0,1,24,1,4,5,NA
+70702,7,2,1,26,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27681.749998,28652.049948,2,98,8,8,1.48,7,7,3,0,0,1,26,1,3,1,3
+70703,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,166234.629208,1,95,8,8,3.47,2,2,0,0,0,1,58,1,2,1,4
+70704,7,2,2,35,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,36453.846815,42702.950408,1,102,14,14,2.83,6,6,1,2,0,1,36,1,2,1,3
+70705,7,2,2,0,3,5,7,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5820.770257,5884.32621,1,95,1,1,0.13,3,3,1,0,0,1,31,1,4,1,4
+70706,7,2,2,48,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,1,1,NA,2,2,2,1,2,2,1,2,2,2,34639.996543,34820.641178,1,102,77,77,NA,3,3,0,0,1,1,61,2,4,1,1
+70707,7,2,2,10,NA,5,7,2,11,132,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8746.306586,9472.024073,1,91,3,3,0.66,4,4,1,2,0,2,33,1,3,5,NA
+70708,7,2,1,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,173139.914798,178954.334223,1,101,7,7,2.64,2,2,0,0,0,1,57,1,2,1,2
+70709,7,2,1,9,NA,4,4,2,9,111,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8988.052978,9133.633722,2,95,8,8,1.85,5,5,1,2,0,1,55,1,2,1,3
+70710,7,2,2,57,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,29087.427528,30575.794656,1,101,3,3,1.07,3,1,0,1,0,2,57,1,4,3,NA
+70711,7,2,2,12,NA,5,7,2,12,151,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,47404.963472,48778.930863,1,97,8,8,2.51,3,3,0,1,0,1,35,1,3,1,4
+70712,7,2,1,2,NA,5,6,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10123.286306,10611.741928,1,100,10,10,3.04,4,4,2,0,0,1,30,2,5,1,5
+70713,7,2,2,30,NA,5,6,2,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,15109.988743,16272.590756,1,90,6,6,0.92,6,6,2,0,2,2,30,2,5,1,5
+70714,7,2,2,1,21,5,6,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4576.510691,4795.397917,1,102,15,15,3.82,5,5,1,1,0,1,29,1,4,1,4
+70715,7,2,2,79,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,53541.401974,54113.194007,1,99,14,14,5,2,2,0,0,2,2,79,1,3,1,5
+70716,7,2,2,43,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,99538.625493,103163.851943,1,103,15,15,5,3,3,0,1,0,1,46,1,5,1,5
+70717,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,24480.331443,26311.37682,2,101,6,6,2.04,2,2,0,0,2,1,77,1,1,1,2
+70718,7,2,1,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,13555.672819,14397.395601,1,94,4,4,1.43,1,1,0,0,1,1,75,1,4,3,NA
+70719,7,1,1,12,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,58479.782556,0,1,99,77,77,NA,4,4,0,2,0,2,45,1,3,1,NA
+70720,7,2,2,7,NA,5,6,2,7,86,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8371.948513,8981.698617,1,90,15,15,5,3,3,0,1,0,2,34,2,5,1,5
+70721,7,2,2,79,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,2,2,2,2,2,1,NA,18241.877822,19614.829564,2,93,9,9,3.64,2,2,0,0,2,2,79,2,2,1,4
+70722,7,2,2,5,NA,4,4,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10936.097083,11421.715282,2,95,1,1,0.18,4,4,2,1,0,2,38,1,2,5,NA
+70723,7,2,1,32,NA,3,3,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,86204.990126,88673.945369,2,103,14,14,5,2,2,0,0,0,1,32,2,5,1,NA
+70724,7,2,1,5,NA,2,2,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,15527.966893,15272.133376,1,90,9,9,2.6,4,4,1,1,0,1,41,1,5,1,4
+70725,7,2,2,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11173.849134,11672.749546,2,93,10,10,3.93,3,3,0,0,2,1,54,1,5,1,5
+70726,7,2,1,5,NA,4,4,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9310.029529,9330.284312,1,97,4,4,0.46,7,7,3,3,0,2,31,1,3,1,NA
+70727,7,2,2,44,NA,4,4,2,NA,NA,2,NA,2,1,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,18488.34481,18123.036332,2,99,15,15,4.34,4,4,0,0,0,1,59,2,4,1,5
+70728,7,2,2,42,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,2,1,2,2,1,2,2,NA,NA,NA,NA,24056.374863,24201.045855,2,99,1,1,0,1,1,0,0,0,2,42,1,3,5,NA
+70729,7,2,1,80,NA,5,7,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,26344.362464,28384.957988,2,95,7,7,2.55,2,2,0,0,2,1,80,1,4,1,2
+70730,7,2,1,55,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,1,1,2,1,1,2,2,2,33162.406014,36437.864606,2,98,7,7,2.25,3,3,0,1,0,2,51,2,1,1,1
+70731,7,2,1,6,NA,5,6,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8580.826574,9087.233306,1,92,14,14,2.42,6,6,1,3,0,1,30,1,4,6,NA
+70732,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,157061.455656,156201.940202,1,90,7,7,1.89,3,3,0,0,1,2,75,1,4,3,NA
+70733,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,10634.832821,11430.282078,2,95,3,3,0.92,1,1,0,0,1,2,80,1,1,2,NA
+70734,7,2,1,7,NA,3,3,1,7,92,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50961.223155,56419.394274,1,94,15,15,5,6,6,0,4,0,1,38,1,5,1,4
+70735,7,2,2,80,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,10072.885959,10826.303506,2,100,3,3,1.1,1,1,0,0,1,2,80,1,2,2,NA
+70736,7,2,2,32,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,31816.953817,32705.969127,1,91,15,15,5,3,3,0,0,1,1,63,2,5,1,5
+70737,7,2,1,28,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,4,5,NA,2,2,2,2,2,2,2,2,1,2,38474.772527,40200.135096,2,93,NA,1,0.22,4,1,0,0,0,1,28,NA,NA,4,NA
+70738,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,1,1,2,1,2,1,1,2,2,1,2,1,NA,18229.985416,18267.209527,1,93,7,7,1.64,5,5,0,2,0,1,47,2,5,1,1
+70739,7,2,1,63,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,22641.791224,22854.140555,1,101,5,5,0.71,6,6,1,1,1,1,63,1,2,1,5
+70740,7,2,2,7,NA,4,4,2,7,91,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9795.701448,10340.670956,1,96,13,13,NA,5,5,1,1,0,1,42,1,3,5,NA
+70741,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,18723.98095,18120.118724,2,95,7,7,1.74,4,4,0,2,0,2,47,1,5,4,NA
+70742,7,2,1,10,NA,2,2,2,10,124,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11316.846999,12445.832216,3,91,13,4,0.81,4,3,0,2,0,1,57,NA,NA,1,NA
+70743,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,105376.658808,1,98,5,5,1.39,2,2,0,1,0,1,46,1,4,3,NA
+70744,7,1,2,65,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,10135.841003,0,2,99,2,2,0.67,1,1,0,0,1,2,65,1,2,5,NA
+70745,7,2,1,51,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,17206.320427,18234.571296,2,100,4,4,0.91,3,3,0,0,0,2,49,1,2,1,2
+70746,7,2,1,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,16589.308426,20972.932517,2,90,8,NA,NA,2,1,0,0,0,1,51,1,2,5,NA
+70747,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,14359.447628,16592.595881,1,97,3,3,1.16,1,1,0,0,1,1,80,1,3,2,NA
+70748,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,31335.13799,31552.004994,1,95,7,7,1.66,5,5,0,3,0,1,34,1,2,1,4
+70749,7,2,1,2,NA,2,2,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8571.222647,8842.539637,2,93,14,14,3.52,5,5,1,2,0,1,44,1,5,1,5
+70750,7,2,2,26,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16844.740449,17564.040518,3,91,1,1,0.11,1,1,0,0,0,2,26,1,5,5,NA
+70751,7,2,1,27,NA,1,1,1,NA,NA,2,NA,2,7,3,NA,3,4,NA,2,2,2,2,2,2,1,2,2,2,35210.641637,36376.282426,1,100,8,3,0.9,6,1,1,0,0,1,33,2,3,6,NA
+70752,7,1,1,40,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,18872.772727,0,1,91,6,6,1.13,6,6,1,3,0,1,40,1,4,6,NA
+70753,7,2,1,30,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,1,2,2,2,2,2,2,37657.076964,37372.8671,1,90,6,6,1.11,5,5,1,2,0,1,30,2,1,6,NA
+70754,7,2,2,15,NA,3,3,1,15,184,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,77001.138762,80653.195108,1,98,9,9,2.88,6,3,1,3,0,1,51,1,2,1,3
+70755,7,2,1,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,141736.66089,144209.560558,1,92,10,10,3.4,3,3,0,0,0,1,56,1,4,1,5
+70756,7,2,1,78,NA,3,3,2,NA,NA,2,NA,2,2,3,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,12786.51051,14057.405205,1,93,2,2,0.41,2,2,0,0,2,2,68,2,1,1,1
+70757,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,74874.432638,77018.874888,2,99,15,15,5,2,2,0,0,0,2,35,2,5,1,5
+70758,7,2,1,21,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,41120.308343,45563.612914,1,93,5,5,1.36,2,2,0,0,0,2,41,2,1,4,NA
+70759,7,2,1,41,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,1,4,NA,2,2,2,1,2,2,2,2,2,2,41410.39303,40802.759347,1,101,12,7,3.58,3,1,0,0,0,1,41,2,1,4,NA
+70760,7,2,1,7,NA,5,6,2,7,87,NA,NA,2,1,2,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6681.315407,7240.00549,3,91,15,15,4.33,7,6,1,3,0,2,40,1,5,1,5
+70761,7,2,1,46,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,4,1,NA,2,2,2,1,2,2,2,2,1,2,33629.261806,33135.804229,2,103,12,8,4.48,2,1,0,0,0,1,46,2,4,1,NA
+70762,7,2,2,34,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,23408.914544,24510.72255,2,90,7,7,1.61,4,4,1,1,0,2,34,1,5,1,5
+70763,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,54095.581484,61529.339683,2,91,6,6,1.88,2,2,0,0,1,2,80,1,5,2,NA
+70764,7,1,2,2,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,34597.213785,0,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+70765,7,2,2,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,44238.530111,44780.242841,1,95,6,6,2.04,2,2,0,0,2,2,67,1,1,2,NA
+70766,7,2,2,45,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16354.688935,16518.054061,2,90,9,9,3.02,3,3,0,0,2,2,70,1,4,1,2
+70767,7,2,2,18,NA,1,1,2,19,228,2,NA,2,2,4,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19850.979841,20551.555585,1,97,3,3,0.5,5,5,0,2,0,1,56,2,2,6,NA
+70768,7,2,2,9,NA,3,3,2,9,113,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,73410.202745,74807.098361,2,91,15,15,5,5,5,2,1,0,2,40,1,5,1,5
+70769,7,2,2,21,NA,1,1,1,NA,NA,2,NA,2,2,6,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,40653.73216,40476.667284,1,102,5,5,0.86,5,5,2,0,0,2,21,2,2,5,NA
+70770,7,2,1,7,NA,1,1,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18107.947773,18211.90239,1,97,10,10,2.32,6,6,0,4,0,1,42,1,4,1,4
+70771,7,2,1,17,NA,5,6,2,17,213,2,NA,2,1,5,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8434.080857,8971.979116,1,91,77,77,NA,4,4,0,2,0,1,50,2,5,1,5
+70772,7,2,1,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,50383.817476,53512.338469,1,94,15,15,5,2,2,0,0,1,2,56,1,3,1,4
+70773,7,2,2,33,NA,2,2,1,NA,NA,2,NA,2,1,5,NA,5,1,1,1,2,2,1,2,2,1,2,2,1,38161.026403,38068.87516,1,100,10,10,2.91,4,4,1,1,0,1,32,1,5,1,5
+70774,7,2,2,49,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18811.641937,19081.796284,2,90,3,3,0.95,2,2,0,1,0,2,49,1,3,5,NA
+70775,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,4,NA,5,1,NA,1,2,2,1,2,2,1,2,2,2,7807.558918,8078.476616,3,90,77,77,NA,2,2,0,0,1,1,60,2,5,1,5
+70776,7,2,1,60,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,1,1,2,1,1,2,1,NA,13133.86792,13808.275718,2,102,15,15,3.82,5,5,0,1,2,1,60,2,2,1,1
+70777,7,2,2,0,10,5,7,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8582.213422,8500.433306,2,94,77,77,NA,4,4,2,0,0,2,23,1,2,6,NA
+70778,7,2,2,3,NA,3,3,2,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,54402.653374,57814.771756,1,98,15,15,3.7,5,5,2,1,0,1,34,1,5,1,5
+70779,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,32785.873783,35446.092694,1,94,6,6,1.82,2,2,0,0,2,1,80,1,4,1,3
+70780,7,2,2,11,NA,3,3,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49777.284509,49627.739683,1,98,15,15,5,5,5,0,3,0,2,44,1,5,1,5
+70781,7,2,1,19,NA,1,1,1,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20638.769105,20614.097145,2,92,15,15,3.37,7,7,0,4,0,1,42,2,3,1,1
+70782,7,2,1,64,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,138075.879417,141933.339512,2,94,15,15,5,2,2,0,0,2,1,64,1,5,1,5
+70783,7,2,2,37,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,13027.332011,13925.626381,2,103,15,15,5,3,3,0,1,0,2,37,2,5,1,5
+70784,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,196995.351093,203724.329642,1,91,15,15,5,2,2,0,0,0,1,53,1,5,1,5
+70785,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,20434.221508,19874.985431,1,102,1,1,0,5,5,0,3,0,2,41,1,4,1,4
+70786,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,153565.050575,157855.235487,1,97,15,15,5,4,4,0,0,1,1,60,1,5,1,5
+70787,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,164066.603708,163511.57474,1,95,7,7,2.86,2,2,0,0,1,1,58,1,4,1,3
+70788,7,2,1,34,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11793.151047,11949.568707,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+70789,7,2,1,5,NA,3,3,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,27873.065855,31447.442273,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+70790,7,2,1,49,NA,3,3,2,NA,NA,2,NA,2,2,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,120278.381994,2,91,15,15,5,4,4,0,2,0,2,48,1,5,1,5
+70791,7,2,2,45,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20303.639991,21936.687597,1,97,15,15,5,3,3,0,1,0,2,45,2,5,1,5
+70792,7,2,1,30,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,6,NA,2,2,2,2,2,2,NA,NA,NA,NA,27605.196104,27396.850962,2,99,12,3,0.52,5,3,0,1,0,1,30,2,2,4,NA
+70793,7,2,1,66,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,100988.137931,101524.74706,1,99,9,9,4.35,2,2,0,0,1,1,66,1,2,1,3
+70794,7,2,2,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,9518.80186,10328.183799,2,100,3,3,0.68,2,2,0,0,2,1,66,1,2,1,2
+70795,7,2,1,12,NA,4,4,1,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11429.628358,11541.555232,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+70796,7,2,2,10,NA,5,6,1,10,125,NA,NA,2,2,2,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8281.624869,8692.694009,2,102,9,9,2.68,4,4,1,1,0,2,38,2,5,1,2
+70797,7,2,2,33,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,16614.865368,17760.536638,3,91,15,15,5,4,4,2,0,0,2,33,2,5,1,5
+70798,7,2,2,53,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,15693.813435,15736.327271,2,100,5,5,0.95,4,4,0,0,1,2,53,1,3,5,NA
+70799,7,2,2,4,NA,5,7,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7143.995395,7774.277146,3,91,15,15,4.47,4,4,2,0,0,1,33,1,5,1,5
+70800,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,NA,NA,NA,1,2,2,1,23484.626749,23642.328802,2,96,NA,NA,NA,3,3,0,0,0,2,50,NA,NA,1,NA
+70801,7,2,2,28,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,103632.167909,110131.87866,2,99,14,14,5,2,2,0,0,0,2,28,1,5,1,5
+70802,7,2,2,5,NA,3,3,1,5,67,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21760.206232,23124.999952,1,98,2,2,0.36,5,5,3,0,0,1,25,1,3,1,3
+70803,7,2,1,7,NA,3,3,1,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15279.821652,15716.743409,1,102,5,1,0.21,5,4,1,1,0,2,24,1,4,5,NA
+70804,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12037.168211,11707.739526,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+70805,7,2,2,55,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12649.084278,13109.186175,3,90,77,77,NA,4,4,0,0,0,1,59,2,2,1,2
+70806,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,126334.218747,127212.680922,1,95,9,9,3.24,3,3,0,0,0,2,42,1,4,3,NA
+70807,7,2,2,22,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,46606.430863,46728.795934,2,102,14,14,3.25,5,5,2,0,0,1,27,1,5,1,5
+70808,7,2,1,77,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,7225.866295,7367.404674,1,96,15,15,5,4,4,0,0,2,1,77,1,5,1,3
+70809,7,2,1,0,1,5,7,1,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9289.230808,9431.807424,2,102,6,6,1.65,3,3,1,0,0,2,21,1,4,1,5
+70810,7,2,2,3,NA,5,6,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,4735.146545,5152.906676,3,90,5,5,0.93,4,4,1,0,0,1,48,2,4,1,NA
+70811,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,2,1,2,2,NA,NA,NA,NA,14138.631841,14167.501749,3,92,8,8,0.91,7,7,3,3,1,1,61,NA,NA,1,4
+70812,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25645.251384,25046.202404,1,92,14,14,4.96,2,2,0,0,0,1,25,1,4,5,NA
+70813,7,2,2,7,NA,2,2,1,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15992.05837,16407.130123,2,93,7,7,1.99,3,3,1,1,0,2,38,1,3,4,NA
+70814,7,2,2,7,NA,3,3,1,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,48532.852397,48387.04619,1,98,14,14,4.12,4,4,0,2,0,2,36,1,5,1,3
+70815,7,1,2,9,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6758.20346,0,2,103,1,1,0.04,2,2,0,1,0,2,28,1,4,5,NA
+70816,7,2,2,30,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,26465.930618,26276.815338,2,100,5,5,0.88,5,5,2,1,0,2,30,1,4,6,NA
+70817,7,2,2,21,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,32537.532358,33640.063825,2,90,8,8,2.01,4,4,0,0,1,2,67,2,4,2,NA
+70818,7,2,1,0,1,2,2,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4997.17037,5277.406032,2,93,15,15,4.84,6,6,1,0,0,1,50,1,4,1,2
+70819,7,2,2,0,2,5,6,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9150.459338,9253.718962,1,97,15,15,4.34,4,4,2,0,0,1,35,2,5,1,5
+70820,7,2,2,13,NA,2,2,2,13,160,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20099.773776,21058.707597,1,90,15,15,5,3,3,0,1,0,1,55,2,3,1,5
+70821,7,2,1,63,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,112072.982256,113124.075015,2,98,14,14,5,2,2,0,0,1,1,63,1,4,1,4
+70822,7,2,2,0,3,4,4,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3484.44779,3522.493854,2,99,1,1,0.03,3,3,1,0,0,2,19,1,3,NA,NA
+70823,7,2,1,9,NA,4,4,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10103.729975,10224.857915,1,96,15,15,5,4,4,0,2,0,2,39,1,4,6,NA
+70824,7,2,1,10,NA,1,1,1,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8828.580268,8822.70874,1,103,7,7,0.51,7,7,3,4,0,1,54,2,1,1,1
+70825,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,129971.22201,129637.352703,1,93,15,15,5,1,1,0,0,0,1,47,1,5,3,NA
+70826,7,2,1,60,NA,1,1,1,NA,NA,2,NA,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14488.953694,14771.336069,3,92,8,8,2.97,2,2,0,0,2,2,62,1,4,1,4
+70827,7,2,1,7,NA,3,3,2,8,96,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,23729.905536,29042.69522,1,101,3,3,0.61,4,4,1,2,0,1,38,1,2,4,NA
+70828,7,2,1,26,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,94547.245282,96016.627954,2,94,77,77,NA,2,2,0,0,0,2,23,1,3,6,NA
+70829,7,2,2,13,NA,4,4,1,13,159,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18163.985724,21329.621847,2,101,9,9,2.46,4,4,0,2,0,1,42,1,3,1,3
+70830,7,2,2,0,10,4,4,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4484.450821,4755.264461,1,91,6,6,0.99,5,5,3,0,0,2,33,2,3,1,4
+70831,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,2,1,2,1,1,2,1,1,2,2,1,19005.010125,19718.036027,1,92,12,12,NA,4,4,1,1,0,1,33,2,4,1,4
+70832,7,2,2,31,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,3,1,2,2,1,2,2,1,2,2,1,53830.599426,56471.452044,1,99,5,5,0.89,4,4,2,0,0,2,31,1,4,1,5
+70833,7,2,2,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,36053.766709,36629.803511,1,100,5,5,1.05,3,3,1,0,0,2,35,1,4,6,NA
+70834,7,2,1,9,NA,3,3,2,9,115,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,53956.99815,55499.881764,1,93,15,15,5,4,4,0,2,0,2,42,1,5,1,NA
+70835,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,20439.261207,20630.95379,2,95,4,2,0.81,2,1,0,0,1,1,62,1,5,3,NA
+70836,7,2,1,6,NA,1,1,1,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19735.224235,20204.314213,2,102,6,6,1.35,3,3,0,1,0,1,37,2,1,1,3
+70837,7,2,1,60,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,1,NA,2,2,2,1,2,2,2,2,2,2,8489.547987,8893.33507,2,96,5,5,0.89,5,4,0,0,1,1,22,2,3,6,NA
+70838,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,6,6,1.67,3,3,0,0,0,2,22,1,4,5,NA
+70839,7,2,2,13,NA,3,3,2,13,167,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,132630.209478,133894.993306,1,97,8,8,2.72,3,3,0,2,0,2,43,1,1,3,NA
+70840,7,2,2,10,NA,4,4,1,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12474.59761,12808.328162,1,100,3,3,0.43,4,4,0,2,0,1,35,1,4,1,3
+70841,7,2,1,11,NA,3,3,2,11,142,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21147.476454,21752.181979,1,95,1,1,0.12,3,3,0,2,0,2,40,1,5,3,NA
+70842,7,2,1,32,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,16945.481829,17170.237079,1,99,77,77,NA,1,1,0,0,0,1,32,2,5,5,NA
+70843,7,2,2,0,11,3,3,2,NA,13,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8383.207272,8554.601009,1,91,12,1,0.07,6,1,1,1,0,2,29,1,4,6,NA
+70844,7,2,1,19,NA,4,4,2,19,235,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10817.360862,11313.215634,2,99,99,99,NA,5,5,0,2,0,2,20,1,3,6,NA
+70845,7,2,2,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,1,1,2,2,1,2,2,1,2,2,1,20000.263815,19016.481945,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+70846,7,2,1,9,NA,2,2,2,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,11807.210833,12400.930816,1,90,7,7,2.1,3,3,0,2,0,2,37,1,3,5,NA
+70847,7,2,2,19,NA,4,4,2,19,232,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11711.384457,11743.959438,2,95,3,3,0.43,4,4,2,0,0,2,23,1,2,5,NA
+70848,7,2,1,80,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,6994.319065,7352.726651,2,99,3,3,0.9,1,1,0,0,1,1,80,1,3,2,NA
+70849,7,2,2,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,37102.230755,38432.724809,2,101,5,2,0.73,2,1,0,0,1,2,49,1,3,6,NA
+70850,7,2,2,28,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,17858.942687,20013.475913,1,97,14,14,4.5,3,3,1,0,0,1,30,1,5,1,5
+70851,7,2,1,36,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,73127.351507,77018.309866,1,103,9,9,3.64,2,2,0,0,0,1,36,1,5,1,5
+70852,7,2,1,14,NA,4,4,1,14,179,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10879.348751,11378.044973,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+70853,7,2,2,19,NA,4,4,1,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12531.903464,12935.738352,2,100,5,5,1.07,4,4,0,1,0,2,36,1,3,5,NA
+70854,7,1,2,53,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,16033.091438,0,1,96,8,8,2.62,3,3,0,0,0,2,53,1,4,4,NA
+70855,7,2,1,9,NA,1,1,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14820.807433,15173.085782,2,102,7,7,1.53,5,5,1,2,0,1,36,1,2,1,3
+70856,7,2,1,1,16,4,4,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,11283.202594,2,97,1,1,0.13,4,4,2,0,1,2,62,1,2,4,NA
+70857,7,2,1,9,NA,3,3,1,9,119,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,26240.82746,27691.552263,1,94,4,4,0.56,5,5,1,2,0,1,34,1,2,3,NA
+70858,7,2,2,9,NA,3,3,2,10,120,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,58907.362493,59749.51689,1,91,10,10,2.77,5,5,0,3,0,1,43,1,5,1,5
+70859,7,2,2,0,6,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4581.358327,4715.327587,2,100,6,6,1.34,4,4,1,2,0,2,31,1,4,5,NA
+70860,7,2,2,44,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,2,1,2,1,2,2,1,2,2,NA,NA,NA,NA,19075.861607,19725.492117,1,96,7,7,1.69,4,4,0,1,0,2,19,2,4,NA,NA
+70861,7,2,2,16,NA,2,2,2,16,201,NA,NA,2,2,2,10,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,17848.732433,20206.102312,2,91,99,99,NA,6,6,1,3,0,2,20,2,2,5,NA
+70862,7,2,1,75,NA,5,6,2,NA,NA,2,NA,2,1,9,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,10818.545624,11805.931644,3,90,4,4,0.99,2,2,0,0,2,1,75,2,4,1,NA
+70863,7,2,2,77,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,3,NA,1,2,1,1,2,1,1,2,1,NA,11462.813869,11819.991623,1,93,2,2,0.87,1,1,0,0,1,2,77,2,4,3,NA
+70864,7,2,1,1,12,2,2,2,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12567.081957,13464.72134,1,97,5,5,1.04,4,4,1,1,0,1,32,1,3,6,NA
+70865,7,2,1,45,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,116464.874823,119757.87667,2,94,9,9,3.35,3,3,0,0,1,2,52,1,2,1,3
+70866,7,2,2,16,NA,3,3,2,16,203,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,80091.55101,82618.226829,1,101,14,14,3.9,4,4,0,2,0,2,41,1,2,1,2
+70867,7,2,2,29,NA,1,1,2,NA,NA,2,NA,2,1,3,NA,4,6,1,2,2,2,2,2,2,2,2,2,2,39426.061521,39404.861746,2,94,7,7,1.34,5,5,2,1,0,1,32,2,1,1,NA
+70868,7,2,1,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,32980.717958,32895.997285,1,95,4,4,1.21,2,2,0,0,0,1,46,1,2,1,2
+70869,7,2,1,38,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,1,2,2,1,2,2,2,37715.365512,45063.17837,2,102,7,7,1.04,7,7,1,2,0,2,37,2,1,1,2
+70870,7,2,1,7,NA,3,3,2,7,86,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,46228.073505,47549.950918,2,94,15,15,4.59,4,4,1,1,0,2,37,1,5,1,5
+70871,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11565.978374,12485.281365,1,99,6,6,1.12,4,4,0,2,0,1,39,1,3,1,3
+70872,7,1,2,23,NA,5,6,NA,NA,NA,2,NA,2,2,3,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,16844.740449,0,3,91,7,7,3.9,2,1,0,0,0,2,21,NA,NA,5,NA
+70873,7,2,2,6,NA,1,1,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16371.237244,2,98,14,14,3.25,5,5,2,1,0,1,37,1,5,1,5
+70874,7,2,1,73,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,6725.306794,7069.929472,2,99,7,7,1.52,4,4,0,0,2,1,73,1,3,2,NA
+70875,7,2,1,15,NA,3,3,1,15,191,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71757.969058,71552.342229,1,98,10,10,2.2,6,6,1,3,0,2,31,1,4,6,NA
+70876,7,2,1,3,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9090.947573,10024.256259,1,90,6,6,1.57,3,3,1,0,0,2,25,1,3,6,NA
+70877,7,2,1,56,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,2,2,2,2,2,2,2,22157.736644,21832.606015,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+70878,7,2,1,25,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,19163.076248,19068.627285,2,97,7,7,2.38,2,2,0,0,0,1,25,1,2,1,2
+70879,7,2,2,10,NA,5,6,2,10,122,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,9620.269705,10600.850342,2,91,6,6,1.57,3,3,0,1,0,1,41,2,3,1,3
+70880,7,2,2,0,1,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8114.787453,8336.925581,1,94,6,6,0.87,6,6,2,2,0,2,24,1,4,6,NA
+70881,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,112935.983109,114691.149811,1,102,10,10,4.76,2,2,0,0,0,1,23,1,5,5,NA
+70882,7,2,2,8,NA,4,4,1,8,98,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12120.418061,12942.575479,2,101,6,6,0.96,5,5,0,4,0,2,36,1,4,4,NA
+70883,7,2,1,31,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,19117.284298,19849.243802,2,95,8,8,2.97,2,2,0,0,0,2,56,1,5,2,NA
+70884,7,2,1,14,NA,4,4,2,14,175,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12721.673656,12746.825446,2,100,6,6,1.34,4,4,1,2,0,2,31,1,4,5,NA
+70885,7,2,1,8,NA,1,1,1,8,106,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11399.23838,11468.323492,2,103,2,2,0.22,7,7,0,3,0,2,39,2,1,5,NA
+70886,7,2,2,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,29040.300396,28462.118402,2,101,1,1,0.11,4,1,0,0,0,2,21,1,4,5,NA
+70887,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,26668.458882,26178.344406,2,102,10,10,4.62,2,2,0,0,0,2,59,1,5,1,NA
+70888,7,1,1,0,9,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17196.879565,0,2,94,14,14,2.87,5,5,2,1,0,1,37,1,3,1,4
+70889,7,2,2,18,NA,1,1,1,18,220,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16781.078148,17378.8338,2,103,2,2,0.22,7,7,0,3,0,2,39,2,1,5,NA
+70890,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,16181.169973,15883.791498,1,96,9,7,3.21,2,1,0,0,0,1,58,1,4,6,NA
+70891,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,4,2,1,2,2,1,2,2,NA,NA,NA,NA,14599.561708,14200.006402,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+70892,7,2,1,6,NA,4,4,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8453.214884,8437.321796,2,101,1,1,0.1,6,6,1,2,1,2,27,1,2,1,2
+70893,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,97101.614214,108614.152251,3,92,6,6,1.17,4,4,0,2,0,2,30,1,2,1,4
+70894,7,2,1,70,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,53149.251154,56449.488341,2,94,8,8,3.4,2,2,0,0,2,2,64,1,4,1,2
+70895,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,NA,27602.403077,28546.401551,1,94,2,2,0.55,2,2,0,0,2,2,75,1,1,3,NA
+70896,7,2,1,15,NA,4,4,1,15,191,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13276.485807,13240.468288,2,100,10,7,2.05,4,3,0,2,0,1,20,1,4,6,NA
+70897,7,2,1,52,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17801.655316,19397.911862,2,95,7,6,1.88,3,2,0,0,0,2,56,1,3,2,NA
+70898,7,2,1,71,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,2,1,NA,2,2,2,2,2,2,1,2,2,NA,11755.776731,12827.229805,3,90,12,13,NA,2,1,0,0,1,1,71,2,2,1,NA
+70899,7,2,2,66,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,136832.42119,141624.96623,2,91,15,15,5,2,2,0,0,1,2,66,1,5,3,NA
+70900,7,2,1,59,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,19093.004278,19032.106491,2,99,2,2,0.19,6,6,0,1,0,1,59,1,2,5,NA
+70901,7,2,2,0,11,4,4,1,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4943.553921,5444.191192,1,100,14,14,3.47,4,4,2,0,0,1,34,1,5,1,5
+70902,7,2,1,11,NA,4,4,2,11,140,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8930.369586,9237.151674,3,91,4,4,1.09,2,2,0,1,0,2,40,1,5,3,NA
+70903,7,2,2,18,NA,1,1,2,19,228,2,NA,2,2,4,13,NA,NA,NA,1,2,2,1,2,2,2,2,2,2,17544.592739,19246.751055,3,92,6,6,1,6,6,1,1,0,1,42,2,1,1,4
+70904,7,2,1,19,NA,2,2,2,19,233,2,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15351.501195,16874.029415,2,90,4,4,1.29,2,2,0,0,0,2,40,1,2,5,NA
+70905,7,2,2,1,16,1,1,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13322.479814,14708.37523,1,92,14,14,3.9,4,4,2,0,0,2,29,1,4,1,4
+70906,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,78529.577822,84771.37367,1,101,14,14,3.9,4,4,0,2,0,2,41,1,2,1,2
+70907,7,2,1,2,NA,4,4,2,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8064.791396,8892.751276,1,91,99,99,NA,3,3,1,0,0,1,33,2,5,5,NA
+70908,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,74379.731652,76506.539917,2,92,15,7,3.67,4,1,0,0,0,1,28,1,5,5,NA
+70909,7,1,2,13,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6587.470541,0,1,93,9,9,1.77,7,7,0,2,0,2,56,NA,NA,5,NA
+70910,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,18404.681357,20079.255367,1,101,3,3,0.61,4,4,1,2,0,1,38,1,2,4,NA
+70911,7,2,1,59,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,146786.506338,147038.128142,1,90,15,15,5,3,3,0,0,0,1,59,1,5,1,5
+70912,7,2,1,67,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,5,2,NA,2,2,2,1,2,2,2,2,2,2,7290.319536,7407.101145,2,93,10,10,5,1,1,0,0,1,1,67,2,5,2,NA
+70913,7,2,1,28,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15507.441472,18423.236589,1,100,9,9,4.92,1,1,0,0,0,1,28,1,5,5,NA
+70914,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,20568.024192,20733.954282,2,92,1,1,0.02,1,1,0,0,0,1,40,1,2,5,NA
+70915,7,1,1,80,NA,3,3,NA,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,37318.801462,0,1,98,6,6,1.95,2,2,0,0,2,1,80,1,3,1,3
+70916,7,2,2,78,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,1,2,NA,2,2,2,2,2,2,1,2,2,NA,17318.187297,23904.945555,2,90,6,3,0.92,2,1,0,0,2,1,76,2,1,2,NA
+70917,7,2,1,48,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,22165.906062,24153.500588,1,96,7,7,2.78,2,2,0,0,0,1,48,1,2,6,NA
+70918,7,2,2,7,NA,3,3,2,7,89,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21910.300386,22612.118571,1,99,4,4,1.02,2,2,0,1,0,2,27,1,4,3,NA
+70919,7,2,2,7,NA,3,3,1,7,89,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64166.795496,75335.746424,1,101,5,5,0.89,5,5,1,2,0,1,31,1,2,1,1
+70920,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20303.639991,21936.687597,1,97,7,7,1.48,5,5,0,1,0,2,46,2,4,1,NA
+70921,7,2,2,29,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,13976.989712,14629.714123,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+70922,7,2,2,2,NA,5,7,2,2,28,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5588.504831,5743.223814,2,91,15,15,3.7,5,5,1,2,0,1,50,NA,NA,1,5
+70923,7,2,1,63,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,89103.220446,89938.887954,2,98,7,7,2.16,3,3,0,0,1,2,55,1,3,1,4
+70924,7,2,1,0,10,5,6,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5528.06179,5490.918184,1,91,14,14,4.48,3,3,1,0,0,1,31,2,5,1,5
+70925,7,2,1,57,NA,5,6,1,NA,NA,2,NA,2,2,7,NA,3,1,NA,1,2,2,1,2,2,1,2,2,3,18416.819037,20059.719948,2,96,5,5,1.36,2,2,0,1,0,1,57,2,3,1,NA
+70926,7,1,2,15,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,66,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17848.732433,0,2,91,8,8,1.85,5,5,0,2,1,1,39,2,3,1,4
+70927,7,2,1,63,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,9946.027787,10495.972559,2,93,8,8,2.62,3,3,0,0,2,2,64,2,1,1,1
+70928,7,2,2,6,NA,4,4,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,7814.742747,8095.069503,2,95,6,6,0.97,6,6,2,1,0,1,54,1,3,6,NA
+70929,7,2,2,29,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,17350.142808,18091.024448,1,92,14,14,2.42,6,6,1,3,0,1,30,1,4,6,NA
+70930,7,2,1,6,NA,2,2,2,6,79,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9807.589376,10028.771,2,90,10,10,3.13,4,4,1,2,0,2,39,1,5,4,NA
+70931,7,2,2,44,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,36945.168658,38286.242768,2,96,10,10,3.78,3,3,0,1,0,1,42,1,3,1,3
+70932,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,2,1,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,18396.558383,19503.669771,1,90,15,15,5,4,4,1,1,0,2,43,2,5,1,5
+70933,7,2,2,46,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19150.604366,19251.871661,3,91,15,15,5,2,2,0,0,0,2,46,1,5,1,5
+70934,7,2,1,0,4,4,4,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5089.464235,5316.261144,2,93,6,6,1.15,5,5,3,1,0,1,29,1,3,5,NA
+70935,7,2,2,46,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,20084.755052,20430.629009,1,99,15,15,5,4,4,0,2,0,2,46,1,5,1,5
+70936,7,2,1,27,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,2,2,2,2,35669.2076,40318.090187,2,94,8,8,2.7,3,3,1,0,0,1,27,1,3,1,4
+70937,7,2,2,12,NA,4,4,2,12,154,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,4,4,0.57,5,5,1,3,0,2,33,1,3,5,NA
+70938,7,2,1,49,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,52756.101074,53793.249362,3,92,8,8,2.97,2,2,0,0,1,1,49,1,2,3,NA
+70939,7,2,2,28,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,6,2,1,2,2,1,2,2,1,2,2,1,39550.779175,41765.24601,2,96,14,14,3.36,4,4,1,1,0,2,28,1,2,6,NA
+70940,7,2,1,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,131567.279219,134004.437006,1,97,15,15,5,3,3,0,0,0,1,53,NA,NA,1,4
+70941,7,2,1,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,87954.465296,90754.904886,2,100,10,10,3.13,4,4,0,2,0,1,45,1,4,1,4
+70942,7,2,2,16,NA,1,1,1,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15690.47168,16249.379042,1,102,8,8,1.33,7,7,1,4,0,2,32,1,3,1,2
+70943,7,2,1,8,NA,1,1,2,8,105,NA,NA,1,1,NA,3,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,10449.970381,10975.442173,1,90,2,1,0.17,5,2,0,1,0,1,25,2,2,6,NA
+70944,7,2,2,0,0,1,1,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7798.643057,7634.568742,1,102,8,8,2.06,4,4,2,0,0,1,28,1,2,1,5
+70945,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,33558.731068,36590.680559,1,101,3,1,0,2,1,0,0,1,2,65,1,3,3,NA
+70946,7,1,1,2,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24594.444896,0,1,98,5,2,0.42,4,3,2,0,0,1,24,1,3,6,NA
+70947,7,2,2,4,NA,3,3,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,59430.588937,63158.057955,2,91,15,15,4.2,6,6,2,0,2,1,63,1,1,1,3
+70948,7,2,2,19,NA,4,4,1,19,231,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18905.695776,2,101,1,1,0.05,2,1,0,0,0,2,19,1,3,NA,NA
+70949,7,2,2,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,128590.415432,131137.819973,1,102,15,15,5,2,2,0,0,2,2,68,1,4,1,5
+70950,7,2,1,50,NA,2,2,1,NA,NA,2,NA,2,2,6,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,34422.302712,34961.637568,2,92,2,2,0.4,3,3,0,0,0,1,50,2,4,1,4
+70951,7,2,1,7,NA,4,4,2,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8050.419807,8503.565046,2,99,4,4,0.41,7,7,0,2,0,2,36,1,3,5,NA
+70952,7,1,2,64,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19125.640099,0,2,95,77,77,NA,2,2,0,0,2,1,64,NA,NA,1,3
+70953,7,1,2,9,NA,2,2,NA,NA,NA,NA,NA,2,1,2,4,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12789.411811,0,1,102,7,7,1.9,4,4,1,1,0,1,29,1,4,1,3
+70954,7,2,1,61,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,28559.076421,28710.827523,1,95,5,5,2.2,1,1,0,0,1,1,61,1,3,3,NA
+70955,7,2,1,14,NA,5,6,1,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8915.81491,9312.338117,1,92,12,12,NA,4,4,0,1,0,1,53,2,5,1,4
+70956,7,2,1,15,NA,1,1,2,15,188,NA,NA,1,1,NA,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20398.562455,20481.595361,2,94,4,4,0.81,3,3,0,1,0,1,49,2,3,1,3
+70957,7,2,1,12,NA,1,1,2,12,154,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,20398.562455,20710.36162,2,94,13,13,NA,5,5,0,3,0,1,32,2,2,1,1
+70958,7,2,1,1,15,2,2,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9237.934626,9085.733547,2,90,5,5,0.89,4,4,2,0,0,2,26,2,2,4,NA
+70959,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,44296.789734,46174.107604,2,102,4,1,0,2,1,0,0,0,2,43,1,5,6,NA
+70960,7,2,2,8,NA,4,4,2,8,100,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7120.704736,7311.203604,1,99,15,8,2.7,4,3,0,2,0,1,49,1,4,6,NA
+70961,7,2,2,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,NA,55881.349278,57792.483909,2,92,9,9,5,1,1,0,0,1,2,74,1,5,5,NA
+70962,7,2,1,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,NA,NA,NA,1,2,2,1,14321.1466,14549.731107,1,96,15,15,5,6,6,1,1,1,2,44,1,3,1,3
+70963,7,2,2,17,NA,2,2,1,17,209,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,18556.092615,20535.833102,1,103,7,7,1.03,7,7,0,3,0,1,50,2,1,1,1
+70964,7,2,2,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,27738.890335,27824.17823,2,101,4,4,0.73,5,5,1,2,0,1,40,1,5,1,5
+70965,7,2,1,5,NA,2,2,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11161.439403,11514.748174,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+70966,7,2,1,20,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,14313.345971,15054.191928,3,91,5,3,0.92,2,1,0,0,0,1,20,1,4,6,NA
+70967,7,2,1,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,22723.53676,24747.854656,1,92,NA,99,NA,4,1,0,2,0,2,56,1,4,4,NA
+70968,7,2,2,8,NA,3,3,1,9,108,NA,NA,2,2,3,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,13450.606713,14078.286101,3,91,4,4,0.65,5,5,2,2,0,2,27,2,2,3,NA
+70969,7,2,1,45,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,31233.526521,31366.480522,1,100,8,3,1.03,6,1,1,0,0,1,33,2,3,6,NA
+70970,7,2,1,31,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,15867.258653,18200.873488,3,91,3,3,0.76,3,3,0,1,0,2,24,1,4,6,NA
+70971,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,20486.987447,21605.827026,2,101,3,3,0.6,3,3,0,2,0,1,39,1,4,4,NA
+70972,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,5950.975297,5997.525697,2,99,77,77,NA,3,3,0,1,1,1,61,1,4,3,NA
+70973,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,18404.681357,19079.901799,1,101,7,7,1.83,3,3,0,0,2,1,67,1,1,1,2
+70974,7,2,1,49,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,30234.306382,1,101,5,5,1.15,3,3,0,1,0,1,49,1,3,1,4
+70975,7,2,2,22,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,118761.81384,123645.102237,3,91,4,4,1.02,2,2,0,0,0,1,22,1,5,1,5
+70976,7,2,1,17,NA,2,2,1,17,211,2,NA,2,1,4,11,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,20327.892981,20116.618863,2,93,4,4,0.69,4,4,0,1,1,2,66,2,3,2,NA
+70977,7,2,1,3,NA,3,3,1,4,48,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,98584.614475,106037.774242,2,101,6,6,1.35,3,3,1,0,0,1,42,1,4,6,NA
+70978,7,2,1,15,NA,4,4,2,15,183,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11834.781205,13035.394237,2,90,2,2,0.31,5,5,0,2,1,2,71,1,2,2,NA
+70979,7,2,1,46,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,18230.629439,18292.13185,1,93,15,15,5,5,5,1,2,0,2,40,1,5,1,5
+70980,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,114168.79702,117438.65007,1,95,8,8,1.28,7,7,1,4,0,1,32,1,3,1,3
+70981,7,2,1,50,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,13525.38183,13953.637486,2,95,15,15,5,3,3,1,0,0,1,50,1,5,1,NA
+70982,7,2,1,44,NA,2,2,2,NA,NA,2,NA,2,2,1,NA,2,1,NA,2,2,2,2,2,2,2,2,1,2,39096.402803,46193.238956,2,91,4,4,0.76,4,4,1,0,0,2,25,2,4,77,NA
+70983,7,2,2,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,71746.365541,83232.546386,1,93,15,15,5,2,2,0,0,0,2,39,1,5,1,5
+70984,7,2,1,73,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,96955.880155,102625.661518,2,101,15,15,5,2,2,0,0,2,1,73,1,5,1,5
+70985,7,2,1,74,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,8601.453077,8734.617022,2,101,5,5,2.02,1,1,0,0,1,1,74,1,2,3,NA
+70986,7,2,2,4,NA,4,4,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9268.093277,9507.318489,1,99,14,14,4.05,3,3,1,0,0,2,32,1,5,1,4
+70987,7,2,2,22,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,130601.953362,132156.64994,2,91,15,15,5,4,4,0,0,0,1,54,1,5,1,NA
+70988,7,2,2,1,14,4,4,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8754.193667,9419.610037,2,98,8,8,1.8,5,5,2,1,0,1,32,1,4,1,5
+70989,7,2,1,16,NA,3,3,1,16,193,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,30943.024697,32205.010072,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+70990,7,2,2,6,NA,1,1,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15048.411882,15335.781033,1,103,14,14,2.96,5,5,1,2,0,1,34,1,4,1,5
+70991,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,25520.996455,28503.986081,1,93,3,3,0.75,2,2,0,0,1,2,80,1,1,2,NA
+70992,7,2,2,1,14,2,2,2,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11412.410776,11898.666241,1,97,3,3,0.44,5,5,2,2,0,2,26,1,4,4,NA
+70993,7,2,1,38,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,15399.847485,15604.102319,1,103,8,8,4.3,1,1,0,0,0,1,38,1,5,5,NA
+70994,7,2,2,13,NA,2,2,2,13,159,NA,NA,2,2,3,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,12680.621719,13709.581084,2,90,99,99,NA,5,5,1,1,0,2,40,2,3,1,1
+70995,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,2,1,2,2,1,2,2,1,2,2,1,31335.13799,32015.561405,1,95,6,4,1.26,6,2,1,2,0,1,28,1,2,1,2
+70996,7,2,1,41,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,49072.976757,48585.714967,1,97,15,15,5,2,2,0,0,0,1,41,1,5,1,5
+70997,7,2,2,62,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,9570.416297,10279.373686,2,98,2,2,0.83,1,1,0,0,1,2,62,1,3,3,NA
+70998,7,2,2,6,NA,2,2,2,6,73,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10762.400563,11527.105697,2,90,6,6,2.01,2,2,0,1,0,2,26,1,3,5,NA
+70999,7,2,2,50,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,161992.272945,161249.908749,1,91,14,14,3.15,5,5,0,1,0,2,50,1,5,1,5
+71000,7,2,2,0,6,5,6,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,7065.624287,7145.357301,2,92,6,6,1.34,4,4,1,1,0,1,40,2,3,1,3
+71001,7,2,1,6,NA,4,4,2,6,74,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7386.979006,7738.237294,2,99,5,5,0.65,6,6,2,1,0,2,53,1,4,3,NA
+71002,7,2,1,13,NA,1,1,1,13,165,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,24228.858782,24226.727855,2,102,7,7,2.16,3,3,0,2,0,2,41,1,5,3,NA
+71003,7,2,1,78,NA,3,3,1,NA,NA,2,NA,2,1,7,NA,1,5,NA,1,2,1,1,2,1,1,2,1,NA,12184.871688,13005.96831,2,103,2,2,0.45,1,1,0,0,1,1,78,2,1,5,NA
+71004,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,14809.997435,16069.288743,2,97,1,1,0.13,4,4,2,0,1,2,62,1,2,4,NA
+71005,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,12230.621635,1,96,8,8,3.14,2,2,0,0,2,2,64,1,3,1,2
+71006,7,2,2,4,NA,1,1,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13366.393396,13546.155547,2,94,5,5,1.3,3,3,1,1,0,2,34,2,2,5,NA
+71007,7,2,2,1,16,1,1,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,15816.39252,3,92,15,15,5,3,3,1,0,0,1,34,1,5,1,5
+71008,7,2,1,50,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,25969.864445,25963.066565,1,97,14,14,5,3,3,0,0,0,2,51,1,5,1,4
+71009,7,1,2,4,NA,5,7,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15046.736747,0,1,101,6,6,1.07,5,5,2,1,0,1,30,1,3,1,3
+71010,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,43813.24867,47156.814318,1,98,5,2,0.42,4,3,2,0,0,1,24,1,3,6,NA
+71011,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,175997.804296,176135.086149,1,101,3,3,0.73,2,2,0,0,1,1,60,1,3,1,5
+71012,7,2,2,24,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,144794.522788,148845.957701,1,97,8,8,3.57,2,2,0,0,0,2,49,1,3,3,NA
+71013,7,2,2,5,NA,4,4,2,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9986.420822,10886.623262,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+71014,7,2,2,16,NA,5,7,2,16,199,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11747.462633,12227.159724,2,99,6,6,1.18,5,5,0,3,0,2,38,1,2,5,NA
+71015,7,2,2,49,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,23342.391629,23465.824854,1,92,15,15,5,2,2,0,0,0,2,49,1,5,3,NA
+71016,7,2,2,5,NA,1,1,1,5,62,NA,NA,2,2,1,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,17282.036547,19079.832121,2,102,15,15,5,3,3,1,0,0,1,41,2,2,1,5
+71017,7,2,2,10,NA,3,3,2,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15442.305642,16400.253331,2,94,5,5,1.3,3,3,0,1,0,1,43,1,3,6,NA
+71018,7,2,1,5,NA,3,3,2,5,62,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57680.74785,67586.530677,1,99,5,5,0.89,4,4,2,0,0,2,31,1,4,1,5
+71019,7,2,2,10,NA,1,1,1,10,130,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,15962.145468,16061.826248,2,98,2,2,0.27,4,4,2,1,0,2,32,2,2,5,NA
+71020,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,30212.098573,34791.026267,1,92,3,3,1.01,1,1,0,0,1,2,80,1,3,2,NA
+71021,7,2,1,64,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,6011.560142,6320.246286,2,99,3,3,1.29,1,1,0,0,1,1,64,2,4,3,NA
+71022,7,2,1,11,NA,3,3,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19590.665143,20475.022602,3,91,5,5,0.87,4,4,0,2,0,2,38,1,2,3,NA
+71023,7,2,1,10,NA,1,1,1,10,122,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13822.148996,13905.918164,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+71024,7,2,1,7,NA,3,3,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,47596.305609,49235.587472,1,99,14,14,5,3,3,0,1,0,1,35,1,5,1,5
+71025,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,7514.993062,9722.610426,2,90,1,1,0.28,1,1,0,0,1,1,61,1,2,5,NA
+71026,7,2,1,14,NA,1,1,1,14,176,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19570.996814,19477.677478,2,100,4,4,0.81,4,4,0,2,0,1,56,1,4,1,2
+71027,7,2,1,10,NA,1,1,1,10,129,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13927.458372,14147.852881,2,98,3,3,0.33,7,7,2,3,0,1,40,2,1,1,1
+71028,7,2,1,2,NA,1,1,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10736.26749,11503.13577,1,103,5,5,1.02,4,4,2,0,0,1,25,1,2,1,4
+71029,7,2,2,25,NA,5,6,2,NA,NA,2,NA,2,1,3,NA,3,1,2,1,2,1,1,2,2,NA,NA,NA,NA,13820.210756,14687.211885,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+71030,7,2,2,13,NA,2,2,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14437.97544,15197.369043,2,90,14,14,3.45,4,4,1,1,0,2,34,2,5,6,NA
+71031,7,1,1,27,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,34087.731449,0,3,92,4,3,0.52,5,4,0,0,0,2,57,1,4,1,2
+71032,7,2,2,12,NA,1,1,1,13,156,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,1,2,2,2,22424.988432,23223.784759,1,94,7,7,2.1,3,3,0,1,0,2,48,2,1,1,3
+71033,7,1,1,26,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17579.006909,0,2,103,15,15,5,1,1,0,0,0,1,26,1,3,5,NA
+71034,7,2,1,11,NA,4,4,2,11,134,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8451.853606,8465.026785,1,93,7,7,1.97,4,4,1,2,0,2,33,1,4,3,NA
+71035,7,2,1,7,NA,5,6,2,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10810.913614,11461.625841,1,97,14,14,2.87,5,5,0,3,0,2,40,2,5,1,5
+71036,7,2,1,29,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,105262.341027,110623.017272,2,98,8,8,3.67,2,2,0,0,0,2,54,1,4,3,NA
+71037,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15497.844354,15358.163767,1,99,15,15,5,2,2,0,0,0,1,56,1,4,1,4
+71038,7,2,1,60,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,2,1,NA,2,2,2,2,2,2,1,2,2,2,8609.250304,11228.904188,2,90,1,1,0.27,2,2,0,0,1,1,60,2,2,1,1
+71039,7,2,2,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,81385.450947,85814.999376,1,93,15,15,5,2,2,0,0,0,2,30,1,5,1,5
+71040,7,2,1,63,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,36541.405352,36735.571242,1,101,4,4,1.22,2,2,0,0,1,1,63,1,2,1,2
+71041,7,2,1,9,NA,5,6,2,9,112,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10810.913614,11461.625841,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+71042,7,2,1,38,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,3,13147.594977,14207.241654,1,102,5,5,0.92,5,5,1,2,0,2,44,2,1,1,2
+71043,7,2,2,9,NA,4,4,1,9,110,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13192.206605,13608.834197,2,102,9,9,2.68,4,4,0,2,0,2,32,1,4,1,NA
+71044,7,2,1,0,2,5,7,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9570.577309,9675.725524,1,95,6,6,1.15,5,5,2,1,0,1,29,1,4,6,NA
+71045,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,NA,10947.445281,13128.299802,1,94,4,4,1.43,1,1,0,0,1,1,80,1,3,4,NA
+71046,7,2,1,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,144353.133634,150786.620773,1,91,14,14,4.03,4,4,0,2,0,1,52,1,4,1,5
+71047,7,2,2,52,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,NA,16058.142925,17306.419461,2,95,5,5,1.18,3,3,0,1,0,2,55,1,4,5,NA
+71048,7,2,2,52,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,NA,NA,NA,NA,24870.513993,26515.471818,2,98,1,1,0.13,4,4,2,0,0,2,52,1,2,4,NA
+71049,7,2,2,20,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,57438.356061,66140.327263,2,101,1,1,0.08,1,1,0,0,0,2,20,1,4,5,NA
+71050,7,2,2,0,4,2,2,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,4481.392842,4343.672823,2,90,12,12,NA,7,7,2,2,0,1,39,2,1,1,1
+71051,7,2,1,78,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,79754.311902,90534.066519,1,97,15,15,5,2,2,0,0,2,1,78,1,3,1,3
+71052,7,2,2,18,NA,1,1,1,18,217,2,NA,2,2,3,10,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,19993.78712,21491.713499,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+71053,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,9680.216878,10112.428208,1,96,12,10,5,2,1,0,0,1,2,61,1,4,3,NA
+71054,7,2,2,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,140431.173819,139253.659476,1,94,10,6,2.42,2,1,0,0,0,1,59,1,4,6,NA
+71055,7,2,2,0,4,5,6,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6580.402774,6856.045444,1,94,99,99,NA,5,5,1,1,0,2,21,1,3,5,NA
+71056,7,1,1,59,NA,5,6,NA,NA,NA,2,NA,2,1,7,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,15409.233853,0,1,92,12,12,NA,4,4,0,0,0,1,59,2,3,1,4
+71057,7,2,2,62,NA,4,4,2,NA,NA,1,2,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,11190.39853,11690.037855,2,99,99,99,NA,2,2,0,0,2,2,62,1,5,3,NA
+71058,7,2,2,19,NA,3,3,2,19,233,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,41202.729804,43105.611516,2,91,6,6,1.26,5,5,0,1,2,2,80,1,4,2,NA
+71059,7,2,1,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,27227.937106,27501.844882,1,101,3,3,0.96,2,2,0,0,0,1,53,1,1,1,2
+71060,7,2,2,31,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,2,2,1,2,2,1,2,2,1,2,2,1,21306.824647,21870.080421,2,98,1,1,0.19,3,3,2,0,0,2,31,1,4,2,NA
+71061,7,2,1,54,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,2,2,2,2,2,2,1,2,2,2,37557.946192,37296.340716,2,102,9,9,3.24,3,3,0,0,0,1,54,2,4,1,4
+71062,7,2,1,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,31291.360507,32453.566842,2,91,12,3,1.25,3,1,0,0,1,1,52,1,4,3,NA
+71063,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,NA,NA,NA,NA,85610.546667,92292.669073,2,91,15,15,5,5,5,1,2,0,1,37,1,4,6,NA
+71064,7,2,2,26,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,10783.449126,11243.921386,1,103,5,5,0.65,6,6,0,0,1,2,26,2,4,5,NA
+71065,7,2,2,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,38739.556504,38769.774143,2,101,6,6,2.04,2,2,0,0,0,2,51,1,5,1,4
+71066,7,2,2,71,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,5,NA,2,2,2,2,2,2,1,2,2,NA,18697.677461,20882.090197,1,93,6,6,2.66,1,1,0,0,1,2,71,2,1,5,NA
+71067,7,2,2,4,NA,1,1,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14812.229505,15746.041025,1,97,4,4,0.72,5,5,2,1,0,2,33,2,1,6,NA
+71068,7,2,1,45,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,108839.502248,114346.356011,2,95,3,3,0.93,2,2,0,0,0,1,45,1,4,1,5
+71069,7,2,2,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,133022.268903,132572.26134,1,98,6,6,1.75,2,2,0,0,2,1,62,1,4,1,3
+71070,7,2,2,79,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,2,2,NA,1,2,1,1,2,1,1,2,1,NA,12344.929687,13268.288363,1,93,2,2,0.73,1,1,0,0,1,2,79,2,2,2,NA
+71071,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,14939.690418,15327.735621,2,92,8,8,3.54,2,2,0,0,0,1,30,1,5,1,5
+71072,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,4,NA,2,2,2,1,2,2,1,2,2,1,19676.781212,20391.029456,1,103,6,6,1.57,3,3,0,1,0,2,50,2,3,4,NA
+71073,7,2,2,18,NA,4,4,2,18,226,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12857.456314,12921.235691,2,97,8,8,2.7,3,3,0,0,1,2,72,1,2,3,NA
+71074,7,1,2,1,14,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6565.095533,0,1,90,15,15,5,4,4,2,0,0,1,36,1,5,1,5
+71075,7,2,2,3,NA,3,3,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50908.326714,56187.051166,1,99,5,5,0.89,4,4,2,0,0,2,31,1,4,1,5
+71076,7,2,2,63,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15732.436891,16737.04767,1,97,14,14,3.93,3,3,0,1,2,2,63,1,4,1,4
+71077,7,2,2,13,NA,1,1,1,13,158,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,2,2,2,1,2,2,1,12970.724558,13432.752316,2,103,8,8,1.29,7,7,3,1,0,2,53,2,2,4,NA
+71078,7,2,2,22,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23614.167119,30467.110717,2,97,3,3,0.82,2,2,1,0,0,2,22,1,4,5,NA
+71079,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,78529.577822,81088.445647,1,98,14,14,3.15,5,5,0,3,0,1,34,1,4,1,4
+71080,7,1,1,22,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,88006.905716,0,1,92,6,6,1.51,3,3,0,0,0,1,46,1,3,3,NA
+71081,7,2,1,4,NA,4,4,2,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7431.820906,7738.904727,1,99,5,5,0.84,5,5,2,1,0,1,35,1,3,1,2
+71082,7,2,1,78,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,11550.158096,13775.328408,2,92,7,7,1.17,6,6,0,1,1,1,78,2,1,1,3
+71083,7,2,2,62,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,9497.094386,9921.129492,2,100,15,15,5,2,2,0,0,1,1,51,1,5,1,3
+71084,7,2,1,55,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,23205.758348,23717.246496,3,92,5,5,1.03,4,4,0,3,0,1,55,1,4,4,NA
+71085,7,2,1,1,15,4,4,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6837.992772,6921.398172,1,96,12,12,NA,2,2,1,0,0,2,32,1,5,5,NA
+71086,7,2,1,24,NA,5,6,1,NA,NA,1,2,2,2,3,NA,4,5,NA,1,2,2,1,2,2,1,2,2,3,14979.624397,15387.451243,1,102,3,1,0.28,2,1,0,0,0,1,24,2,4,5,NA
+71087,7,2,2,65,NA,1,1,1,NA,NA,2,NA,2,2,8,NA,1,2,NA,2,2,2,2,2,2,2,2,2,2,15207.312407,15896.113669,2,98,4,4,0.48,6,6,2,0,2,2,65,2,1,2,NA
+71088,7,2,2,13,NA,2,2,2,13,160,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,16680.090857,17557.412914,1,90,15,15,5,4,4,0,2,0,2,43,1,5,1,5
+71089,7,2,2,32,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,3,4,2,2,2,2,2,2,2,1,2,2,2,49274.703023,52411.531429,2,102,77,77,NA,3,3,0,0,0,2,48,2,2,1,3
+71090,7,2,1,49,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11113.602843,11901.173467,3,90,15,15,5,4,4,0,2,0,2,41,2,5,1,5
+71091,7,2,2,39,NA,2,2,1,NA,NA,2,NA,2,1,NA,NA,4,6,2,1,2,2,1,2,2,NA,NA,NA,NA,45662.166351,45001.161705,1,92,14,9,3.77,3,2,0,1,0,2,39,2,4,6,NA
+71092,7,2,2,71,NA,1,1,1,NA,NA,2,NA,2,2,77,NA,1,2,NA,2,2,2,2,2,2,2,2,2,NA,20127.084548,24087.460037,1,103,77,77,NA,4,4,1,0,1,1,20,1,3,6,NA
+71093,7,2,1,50,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,185033.990584,188295.9196,1,97,15,15,5,2,2,0,0,0,1,50,1,3,6,NA
+71094,7,2,2,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,140431.173819,139253.659476,1,94,9,9,3.97,2,2,0,0,0,1,59,1,3,5,NA
+71095,7,2,1,31,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,81642.217375,85347.318656,3,91,14,14,5,3,3,1,0,0,2,30,1,4,1,5
+71096,7,2,2,13,NA,1,1,1,13,160,NA,NA,1,1,NA,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,18515.058419,19360.671834,2,96,6,6,1.11,6,6,0,2,1,1,40,2,2,1,2
+71097,7,2,1,21,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,25989.236835,2,101,99,1,0.21,3,1,0,0,0,1,20,1,4,5,NA
+71098,7,2,2,19,NA,4,4,1,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,2,1,0.32,4,1,0,0,0,2,19,1,4,NA,NA
+71099,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,47566.45715,54102.989916,1,90,9,9,5,1,1,0,0,1,2,80,1,5,3,NA
+71100,7,2,2,80,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,12020.872946,12623.344251,3,90,15,15,4.07,5,5,0,2,1,1,42,1,5,1,4
+71101,7,2,1,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,19541.667675,20187.705923,2,95,2,2,0.75,1,1,0,0,0,1,51,1,2,5,NA
+71102,7,2,1,19,NA,5,6,1,20,NA,2,NA,2,2,4,15,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,5904.602463,6308.718622,2,92,6,6,1.3,4,4,0,1,0,2,48,2,3,1,3
+71103,7,2,2,57,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,32709.179605,34590.259766,1,94,6,6,1.26,5,5,0,2,0,2,38,1,4,1,NA
+71104,7,2,2,78,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,13141.36586,14020.834298,1,96,6,6,2.46,1,1,0,0,1,2,78,1,2,2,NA
+71105,7,2,2,66,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,3,NA,2,2,2,2,2,2,2,2,1,2,8725.210615,9371.55685,2,93,14,14,2.43,7,7,1,1,1,2,66,2,2,3,NA
+71106,7,2,1,1,19,3,3,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,31199.621619,36557.677601,1,97,5,5,0.87,4,4,2,0,0,1,35,1,5,1,5
+71107,7,2,2,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,2,1,2,2,1,2,2,1,2,2,1,64581.191728,65990.445093,2,94,15,10,5,3,1,0,0,0,1,26,NA,NA,77,NA
+71108,7,2,2,20,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23845.8146,22808.13483,1,98,2,1,0.04,4,1,0,0,0,2,19,1,4,NA,NA
+71109,7,2,2,0,9,1,1,1,NA,9,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8775.375504,9358.559086,2,102,14,14,3.8,4,4,2,0,0,2,41,2,4,1,5
+71110,7,2,1,2,NA,2,2,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13002.944731,13931.716845,2,91,4,4,0.69,4,4,2,0,0,2,21,1,3,6,NA
+71111,7,2,2,75,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,18069.116427,18623.459107,2,100,2,2,0.62,1,1,0,0,1,2,75,1,4,2,NA
+71112,7,2,2,55,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,22631.175755,22215.258702,2,96,7,7,2.58,2,2,0,0,1,2,55,1,4,3,NA
+71113,7,2,2,26,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,12412.338679,12980.380959,3,90,15,15,3.7,5,5,0,0,0,1,56,2,3,1,3
+71114,7,2,2,20,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,132307.499792,154386.779868,1,98,1,1,0.31,1,1,0,0,0,2,20,1,4,5,NA
+71115,7,2,2,12,NA,4,4,2,12,150,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11600.051433,12143.645666,3,90,14,14,2.97,5,5,0,2,1,1,73,2,3,2,NA
+71116,7,2,2,7,NA,4,4,2,7,92,NA,NA,2,2,2,0,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,9412.419416,10343.079616,1,93,5,5,0.64,7,7,0,2,1,1,21,2,4,5,NA
+71117,7,2,2,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,25241.487585,26857.231387,2,97,2,2,0.21,7,7,2,3,0,2,32,1,4,5,NA
+71118,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,34802.051557,36349.694696,1,101,4,4,0.99,2,2,0,0,0,2,51,1,5,1,2
+71119,7,2,1,36,NA,3,3,1,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,71111.990643,75738.399435,2,103,15,15,5,4,4,2,0,0,1,36,2,4,1,5
+71120,7,2,1,69,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,28559.076421,28710.827523,1,95,2,2,0.88,1,1,0,0,1,1,69,1,3,3,NA
+71121,7,2,2,0,11,5,6,2,NA,12,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4326.150649,4254.537398,1,99,14,14,2.66,7,7,3,1,0,1,35,1,5,1,5
+71122,7,2,1,63,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,15,15,5,2,2,0,0,2,1,63,1,4,1,5
+71123,7,2,1,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,23824.065321,24171.124621,1,100,14,14,3.6,4,4,1,1,0,1,41,1,4,1,5
+71124,7,1,1,0,10,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,4944.997189,0,1,103,13,13,NA,6,6,1,2,0,2,40,2,3,3,NA
+71125,7,2,2,9,NA,5,6,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,5064.232234,5712.276567,2,92,99,2,0.31,7,4,3,3,1,1,61,2,1,1,3
+71126,7,2,1,5,NA,1,1,1,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,14321.363328,14968.737037,2,96,6,6,0.77,7,7,2,1,0,1,53,2,1,1,1
+71127,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,59282.809425,61101.54758,1,101,14,14,5,2,2,0,0,2,2,70,1,5,1,2
+71128,7,2,1,6,NA,3,3,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19998.804,20570.663593,1,94,3,3,0.37,5,5,0,3,0,2,29,1,4,4,NA
+71129,7,2,1,16,NA,3,3,1,16,200,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,69211.537407,74163.290023,1,92,14,14,3.16,6,6,1,1,0,1,49,1,1,1,3
+71130,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,19130.568715,18607.010517,2,98,5,5,0.59,7,7,3,0,0,2,50,1,5,4,NA
+71131,7,2,1,13,NA,3,3,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,71934.689876,71392.8162,1,100,15,15,5,4,4,0,1,0,1,50,1,4,1,4
+71132,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,134694.414609,139412.076132,2,94,10,10,4.3,2,2,0,0,2,1,69,1,4,1,5
+71133,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,82820.100138,94804.847305,1,90,7,7,3.13,1,1,0,0,0,2,36,1,5,5,NA
+71134,7,2,1,3,NA,3,3,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,57680.74785,67586.530677,1,99,77,77,NA,4,4,1,1,0,1,31,2,3,1,3
+71135,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,33731.056243,35845.261325,2,98,3,3,0.88,2,2,0,0,2,1,77,1,1,1,3
+71136,7,2,1,21,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,9956.598907,12313.186759,1,95,5,5,0.73,6,6,1,0,1,1,62,2,3,1,NA
+71137,7,2,1,61,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,NA,14488.953694,14771.336069,3,92,1,1,0.24,1,1,0,0,1,1,61,1,4,5,NA
+71138,7,2,1,0,7,4,4,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10070.421465,10410.321555,2,101,4,4,0.76,4,4,1,1,0,1,28,1,2,1,4
+71139,7,2,1,7,NA,1,1,1,7,90,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11159.151566,11210.524757,1,102,4,4,0.5,6,6,2,2,0,1,25,1,2,1,3
+71140,7,2,1,26,NA,1,1,2,NA,NA,2,NA,2,2,2,NA,4,5,NA,2,2,2,2,2,2,2,2,2,2,35669.2076,36620.108921,2,94,15,15,5,3,3,0,0,0,1,41,2,3,1,NA
+71141,7,2,2,78,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,20131.904783,20903.590301,1,92,8,8,2.51,3,3,0,0,1,2,78,1,5,2,NA
+71142,7,1,1,8,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,1,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8943.919305,0,1,92,5,5,0.63,7,7,0,4,1,1,60,NA,NA,1,NA
+71143,7,2,2,53,NA,4,4,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19548.136896,19371.951409,1,90,15,15,5,2,2,0,0,0,1,57,2,4,1,5
+71144,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,22758.541444,24464.617143,1,98,7,7,1.03,7,7,0,4,0,2,20,1,3,5,NA
+71145,7,2,2,15,NA,1,1,1,15,191,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21062.314667,21467.226385,1,94,10,10,2.94,4,4,0,2,0,2,52,1,5,2,NA
+71146,7,2,1,34,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,130589.073376,141489.20172,2,102,8,6,2.57,2,1,0,0,0,1,34,1,4,5,NA
+71147,7,2,2,11,NA,4,4,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8757.841043,8910.615224,2,100,13,13,NA,4,4,0,2,0,2,28,1,2,6,NA
+71148,7,2,1,0,0,4,4,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6433.441277,6898.375971,2,97,1,1,0,5,5,3,0,0,2,23,1,2,1,3
+71149,7,2,1,12,NA,5,7,2,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5504.178541,5957.59682,1,99,15,15,4.47,4,4,0,2,0,2,52,2,5,1,5
+71150,7,2,1,8,NA,4,4,2,8,100,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14125.862146,14295.209168,1,97,5,5,0.91,4,4,0,3,0,2,44,1,4,5,NA
+71151,7,2,2,8,NA,4,4,1,8,102,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11334.095519,11753.9565,2,96,5,5,0.67,6,6,1,2,1,1,34,1,4,1,4
+71152,7,2,1,80,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,1,2,1,2,2,1,2,2,NA,18812.694081,19206.027387,1,102,14,14,4.59,3,3,0,0,1,2,46,1,4,1,1
+71153,7,2,2,7,NA,4,4,2,7,94,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6655.097829,7106.52929,2,99,15,15,4.9,7,7,1,4,0,2,53,1,5,1,5
+71154,7,2,2,11,NA,5,6,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10699.45895,11230.540406,1,97,15,15,2.33,7,7,2,4,0,2,40,2,5,1,4
+71155,7,2,1,41,NA,3,3,2,NA,NA,2,NA,2,2,5,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,117676.178911,119449.245993,2,99,14,14,5,1,1,0,0,0,1,41,2,5,5,NA
+71156,7,2,2,16,NA,1,1,2,16,194,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,21120.105603,21750.292661,1,98,6,6,1.98,2,2,0,1,0,2,37,1,4,3,NA
+71157,7,2,2,26,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,15967.106149,16195.487754,2,103,6,6,1.3,4,4,1,1,0,2,26,1,4,1,3
+71158,7,2,1,39,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,5,1,NA,1,2,1,1,2,1,1,2,2,3,16939.617906,17443.653216,3,91,14,14,3.58,4,4,1,1,0,1,39,2,5,1,5
+71159,7,2,2,37,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,97101.614214,100848.484003,3,92,15,15,5,4,4,2,0,0,1,38,1,5,1,5
+71160,7,2,2,3,NA,5,6,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7483.230909,7475.326886,2,96,7,7,2.27,3,3,1,0,0,1,34,2,5,1,5
+71161,7,2,1,45,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,18178.365056,18317.23991,1,94,6,6,2.05,2,2,0,1,0,1,45,1,2,5,NA
+71162,7,2,1,13,NA,5,7,1,13,164,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14831.338089,14976.576871,1,98,3,3,0.43,4,4,0,1,0,2,39,1,2,5,NA
+71163,7,2,1,11,NA,3,3,1,11,135,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,49922.147265,52289.098209,1,98,6,6,1.11,5,5,0,2,1,2,37,1,1,1,1
+71164,7,2,1,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,129656.862785,1,101,14,14,3.25,4,4,0,1,0,1,48,1,4,1,2
+71165,7,2,2,0,4,4,4,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4128.726485,4546.845595,2,93,5,5,1.05,3,3,1,0,0,2,29,1,3,5,NA
+71166,7,2,1,55,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,NA,32720.69734,33802.428047,1,95,2,2,0.67,1,1,0,0,0,1,55,1,3,3,NA
+71167,7,2,2,19,NA,5,7,2,19,231,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,91797.787708,94488.443358,1,95,5,5,1.3,3,3,0,0,1,2,19,1,3,NA,NA
+71168,7,1,1,79,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,7,1,NA,1,2,2,NA,NA,NA,NA,NA,NA,NA,74473.849242,0,2,98,NA,NA,NA,2,2,0,0,2,1,79,1,7,1,NA
+71169,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,121588.761604,120347.630506,1,91,10,10,3.78,3,3,0,0,2,1,62,1,5,1,5
+71170,7,1,1,23,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,106185.516032,0,2,96,5,5,2.02,1,1,0,0,0,1,23,1,4,5,NA
+71171,7,2,2,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12331.419303,13281.948926,2,95,6,6,1.65,2,2,0,0,2,1,62,1,1,1,3
+71172,7,2,2,42,NA,4,4,1,NA,NA,2,NA,2,2,5,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,25774.834017,25516.02368,1,100,5,5,1.05,3,3,1,0,0,2,42,2,5,1,NA
+71173,7,2,2,5,NA,1,1,1,5,69,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,1,1,2,1,NA,NA,NA,NA,13560.780585,14415.696661,2,96,3,3,0.46,5,5,1,2,0,1,37,1,1,1,2
+71174,7,2,1,11,NA,3,3,1,11,141,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19272.757026,20113.595939,1,94,6,6,1.18,5,5,1,2,0,1,30,1,3,1,3
+71175,7,2,1,73,NA,2,2,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,16928.800715,17540.589716,1,90,9,9,3.97,2,2,0,0,1,1,73,1,4,1,5
+71176,7,2,1,62,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,7611.107768,7521.80097,2,97,4,4,1.47,1,1,0,0,1,1,62,1,4,3,NA
+71177,7,1,2,4,NA,2,2,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10308.095307,0,2,90,14,14,3.06,5,5,1,2,0,1,42,1,4,1,5
+71178,7,2,2,4,NA,2,2,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10308.095307,10747.298537,2,90,3,3,0.46,5,5,3,0,0,2,22,1,2,5,NA
+71179,7,1,2,32,NA,4,4,NA,NA,NA,2,NA,2,1,2,NA,5,1,3,1,2,2,1,2,2,NA,NA,NA,NA,27303.803575,0,1,96,7,7,2.23,3,3,1,0,0,1,29,2,5,1,5
+71180,7,2,1,57,NA,4,4,2,NA,NA,2,NA,2,2,2,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,12517.592486,12477.667198,1,96,10,10,1.8,7,7,1,1,0,1,57,2,1,1,3
+71181,7,2,1,19,NA,4,4,1,19,235,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,17606.165994,17558.40257,2,101,6,1,0.23,3,1,0,0,0,1,21,1,4,5,NA
+71182,7,2,2,47,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,20084.755052,20500.648257,1,99,15,8,2.7,4,3,0,2,0,1,49,1,4,6,NA
+71183,7,2,1,38,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,107347.639721,117795.71535,3,92,6,6,1.17,4,4,0,2,0,2,30,1,2,1,4
+71184,7,2,1,64,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,12579.986433,13271.133625,1,92,14,14,5,1,1,0,0,1,1,64,1,5,3,NA
+71185,7,2,1,29,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,50647.682308,56120.478778,2,96,7,7,2.38,2,2,0,0,0,1,29,1,3,1,4
+71186,7,2,1,1,17,1,1,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,6,6,1.25,4,4,1,0,1,1,46,1,2,6,NA
+71187,7,2,2,63,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,3,NA,1,2,2,1,2,2,1,2,2,1,9609.522554,10094.833809,3,91,2,2,0.72,1,1,0,0,1,2,63,1,1,3,NA
+71188,7,2,2,1,14,4,4,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,7710.515686,2,100,5,5,0.94,4,4,2,0,0,2,33,1,4,6,NA
+71189,7,2,1,51,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,144353.133634,150786.620773,1,94,15,15,5,3,3,0,1,0,2,43,1,5,1,5
+71190,7,2,2,12,NA,4,4,1,12,150,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12327.773112,12176.001893,2,95,1,1,0.18,4,4,2,1,0,2,38,1,2,5,NA
+71191,7,2,2,15,NA,4,4,2,16,192,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11711.384457,11558.024533,2,95,8,8,1.61,6,6,1,3,0,2,48,1,3,5,NA
+71192,7,2,1,1,23,2,2,2,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13002.944731,13931.716845,2,91,9,9,2.6,4,4,1,1,0,2,31,2,4,1,5
+71193,7,2,2,7,NA,3,3,2,7,85,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,19880.837381,19616.828143,1,101,6,6,0.87,6,6,2,2,0,2,23,1,4,6,NA
+71194,7,2,2,61,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,11696.973403,11982.125462,2,101,4,4,0.85,4,4,1,0,1,2,61,1,4,3,NA
+71195,7,2,2,36,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,16663.714937,17273.763859,2,93,15,15,5,3,3,1,0,0,1,41,2,5,1,5
+71196,7,2,2,18,NA,5,6,1,18,223,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,1,1,2,2,1,5668.184078,5913.178809,2,92,10,8,2.01,7,4,1,1,1,2,27,2,3,1,3
+71197,7,2,1,36,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15867.258653,16587.349339,3,91,5,5,1.07,4,4,0,2,0,2,36,1,5,1,4
+71198,7,2,2,11,NA,4,4,1,11,138,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11195.065587,11726.17681,2,96,7,7,1.33,6,6,0,3,1,1,74,1,1,1,NA
+71199,7,2,2,49,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21857.756498,23340.233721,2,97,7,7,1.92,3,3,0,1,0,1,57,1,4,1,4
+71200,7,2,1,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,151766.599459,151581.541662,3,91,15,15,5,1,1,0,0,0,1,51,1,5,5,NA
+71201,7,2,2,57,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,17852.668137,18572.420627,3,91,6,6,1.77,2,2,0,0,0,2,57,2,1,1,1
+71202,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,65706.229298,75664.626338,1,101,7,7,2.64,2,2,0,0,2,1,80,1,1,1,3
+71203,7,2,2,63,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,10391.563983,10953.658293,3,90,14,14,4.96,2,2,0,0,2,1,68,1,3,1,5
+71204,7,2,2,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,46606.430863,46728.795934,2,102,14,14,3.25,5,5,2,0,0,1,27,1,5,1,5
+71205,7,2,2,12,NA,3,3,2,12,146,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,132630.209478,135755.422092,1,97,14,14,3.36,4,4,0,2,0,2,49,1,5,1,5
+71206,7,2,1,29,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,9938.495618,10209.075521,1,93,15,6,2.3,6,1,0,0,0,1,34,2,5,5,NA
+71207,7,1,1,7,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10229.206765,0,1,96,14,14,2.96,5,5,0,3,0,1,46,NA,NA,1,5
+71208,7,2,1,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16905.961576,17805.493573,2,94,10,10,4.76,2,2,0,0,0,1,32,1,5,1,5
+71209,7,2,2,2,NA,1,1,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,10099.930724,10530.26456,2,96,6,6,0.87,6,6,1,3,0,1,46,2,1,1,1
+71210,7,2,1,0,0,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12909.068713,12685.898973,2,92,1,1,0.22,3,3,1,0,0,1,22,1,4,6,NA
+71211,7,2,2,32,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23038.68441,23821.989403,1,92,6,6,1.92,2,2,0,0,0,2,33,2,4,5,NA
+71212,7,2,2,65,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,49568.196121,50175.172049,1,101,3,3,1.24,1,1,0,0,1,2,65,1,2,2,NA
+71213,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,65897.669973,68436.247366,1,99,77,77,NA,3,3,0,0,0,1,42,1,4,6,NA
+71214,7,2,2,11,NA,1,1,1,11,138,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,14414.529053,14932.182215,2,96,6,6,0.87,6,6,1,3,0,1,46,2,1,1,1
+71215,7,2,1,60,NA,3,3,2,NA,NA,2,NA,2,1,77,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,19328.482066,19868.466665,3,90,77,77,NA,1,1,0,0,1,1,60,2,5,5,NA
+71216,7,2,1,46,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,20197.354438,20310.389983,2,96,8,8,3.4,2,2,0,0,0,1,46,2,4,1,4
+71217,7,2,1,28,NA,1,1,1,NA,NA,2,NA,2,7,5,NA,3,5,NA,2,2,2,2,2,2,1,2,2,1,47487.549895,48753.51505,1,102,7,7,1.89,3,3,0,0,0,1,53,2,1,1,1
+71218,7,2,1,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,87891.395784,92007.216086,2,99,15,15,5,1,1,0,0,0,1,33,1,5,5,NA
+71219,7,2,1,60,NA,4,4,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7915.816068,7977.736071,2,93,6,6,1.13,4,4,0,0,2,1,60,2,3,1,3
+71220,7,2,1,4,NA,4,4,1,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11920.911192,13144.753907,2,96,14,14,3.58,4,4,2,0,0,1,36,1,4,1,5
+71221,7,2,1,1,20,1,1,1,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13305.770449,13726.956753,3,92,5,5,0.68,6,6,3,0,0,2,19,1,4,NA,NA
+71222,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,191547.714427,193526.009858,2,91,15,10,5,2,1,0,0,0,2,55,1,5,6,NA
+71223,7,2,2,25,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,30275.274308,30320.430859,2,101,3,2,0.82,2,1,0,0,0,2,25,1,4,5,NA
+71224,7,2,2,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,44636.780791,46200.180609,2,91,6,6,1.26,5,5,0,1,2,2,80,1,4,2,NA
+71225,7,2,1,0,1,1,1,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6298.658963,6412.943673,2,94,77,77,NA,4,4,2,0,0,2,27,2,3,1,3
+71226,7,2,2,0,7,3,3,1,NA,7,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7272.728291,7471.815482,1,98,13,13,NA,5,5,2,1,0,1,31,1,3,1,3
+71227,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13209.159597,13930.540942,2,95,6,6,1.08,4,4,1,1,0,1,39,1,4,1,4
+71228,7,2,2,49,NA,1,1,2,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,31273.071103,31436.157508,1,90,12,12,NA,5,5,0,1,0,2,49,2,1,1,NA
+71229,7,2,1,48,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,1,1,2,1,1,2,1,3,12066.381592,12177.318898,1,103,4,4,0.82,3,3,0,0,0,1,48,2,4,1,1
+71230,7,1,1,30,NA,1,1,NA,NA,NA,2,NA,2,2,4,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,34887.439952,0,2,94,4,4,0.72,4,4,1,1,0,1,30,2,1,1,3
+71231,7,2,1,52,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,17796.910402,18163.842064,1,103,2,2,0.63,1,1,0,0,0,1,52,1,3,4,NA
+71232,7,2,2,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,32041.645327,32131.786508,1,94,2,2,0.73,1,1,0,0,0,2,55,1,3,3,NA
+71233,7,2,1,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12267.215138,12511.113657,2,102,2,2,0.36,4,4,1,2,0,2,36,1,3,5,NA
+71234,7,2,2,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,154825.466557,157902.106304,1,91,7,7,2.92,2,2,0,0,1,2,47,1,5,5,NA
+71235,7,2,1,9,NA,3,3,1,9,115,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38712.032122,41122.784973,3,91,15,15,5,6,6,1,3,0,2,40,1,5,1,5
+71236,7,2,2,2,NA,4,4,2,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7338.212404,7527.624117,1,93,7,7,2.16,3,3,1,0,0,2,28,2,2,1,5
+71237,7,2,2,43,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,35469.911999,36868.395052,2,91,12,13,NA,3,1,0,0,1,1,52,1,4,3,NA
+71238,7,2,2,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,20186.483875,20343.011032,1,99,13,13,NA,4,4,0,2,0,1,55,NA,NA,1,4
+71239,7,2,1,48,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,1,148090.195644,148420.972197,1,98,5,5,1.79,1,1,0,0,0,1,48,1,1,2,NA
+71240,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,6512.729795,7017.196255,1,93,2,2,0.69,1,1,0,0,1,1,80,1,5,2,NA
+71241,7,2,2,0,9,4,4,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4588.59937,4638.701468,2,95,10,10,3.82,3,3,1,0,0,1,25,1,4,1,4
+71242,7,2,1,48,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,19436.026093,19746.342802,2,97,9,9,1.45,7,7,1,2,2,2,45,1,3,5,NA
+71243,7,2,1,17,NA,4,4,1,18,216,2,NA,2,1,5,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19163.050164,19648.623212,2,102,14,14,4.48,3,3,0,2,0,1,41,1,3,4,NA
+71244,7,2,2,64,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,118437.327505,122585.585802,1,90,15,15,5,1,1,0,0,1,2,64,1,5,3,NA
+71245,7,2,1,62,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,7654.035132,8195.014907,3,90,9,9,2.6,4,4,0,0,1,1,62,2,4,1,5
+71246,7,2,1,14,NA,1,1,1,14,178,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23389.620035,24913.232835,1,91,14,14,3.9,4,4,0,1,0,1,41,1,2,1,4
+71247,7,2,1,13,NA,4,4,1,13,162,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,NA,NA,NA,1,2,2,1,14650.502937,14679.468181,2,102,14,14,3.25,5,5,1,1,0,2,32,1,4,1,3
+71248,7,2,1,4,NA,5,6,1,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10123.286306,10422.822776,1,100,15,15,5,3,3,1,0,0,2,28,2,5,1,5
+71249,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,17420.978407,17335.115714,2,97,3,3,0.82,2,2,0,0,0,1,24,1,3,5,NA
+71250,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,20486.987447,22351.0228,2,101,3,1,0,4,1,0,2,0,1,39,1,1,6,NA
+71251,7,1,1,60,NA,4,4,NA,NA,NA,2,NA,2,1,99,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,12848.137255,0,1,101,15,15,5,2,2,0,0,2,1,60,2,5,1,NA
+71252,7,2,2,41,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,4,1,2,2,2,2,1,2,2,1,2,2,2,25778.164795,25912.595732,2,90,2,2,0.25,5,5,0,1,0,2,41,2,4,1,NA
+71253,7,1,2,47,NA,3,3,NA,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,125799.478905,0,1,100,15,15,4.63,5,5,0,0,0,1,51,1,5,1,3
+71254,7,2,2,50,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10964.859884,13217.672684,3,90,77,77,NA,3,3,0,1,0,1,54,2,3,1,3
+71255,7,2,2,2,NA,3,3,1,2,35,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14950.402914,15888.087777,1,102,5,1,0.21,5,4,1,1,0,2,24,1,4,5,NA
+71256,7,2,2,7,NA,4,4,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7428.928475,8424.338415,3,91,1,1,0.07,6,6,2,3,0,2,30,1,2,3,NA
+71257,7,2,1,7,NA,4,4,1,7,90,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8714.559478,8887.823591,2,96,7,7,1.04,7,7,0,4,0,2,37,1,3,3,NA
+71258,7,2,2,19,NA,2,2,2,19,232,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,15442.648697,17482.235263,3,90,10,8,4.3,2,1,0,0,0,1,33,1,3,6,NA
+71259,7,2,2,23,NA,5,6,2,NA,NA,1,2,1,1,NA,NA,3,5,2,1,2,2,1,2,2,1,2,2,1,16844.740449,19586.242825,3,91,99,1,0.28,3,1,0,0,0,2,23,1,3,5,NA
+71260,7,2,2,14,NA,1,1,1,14,171,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20347.899985,21072.708732,3,92,15,15,3.15,7,7,0,4,0,2,35,2,3,3,NA
+71261,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,1,3,NA,1,5,NA,2,2,2,1,2,2,1,2,2,2,24004.6026,24129.784561,2,91,6,6,0.93,5,5,1,2,0,2,50,2,1,5,NA
+71262,7,2,1,15,NA,3,3,2,15,181,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,61479.689958,61628.148021,2,100,15,15,4.5,6,6,0,4,0,1,45,1,5,1,5
+71263,7,2,1,47,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,22685.373982,23664.393091,1,98,1,1,0,4,4,0,3,0,1,47,1,5,3,NA
+71264,7,2,2,20,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,30253.427014,30121.660039,2,90,15,15,4.2,5,5,1,0,0,2,50,NA,NA,6,NA
+71265,7,2,2,15,NA,4,4,2,15,187,NA,NA,1,1,NA,10,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7825.55935,7975.961244,3,90,15,15,5,5,5,0,1,1,2,61,1,5,2,NA
+71266,7,2,2,68,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,117131.838201,119452.245796,1,94,14,14,5,2,2,0,0,2,1,74,1,4,1,4
+71267,7,2,2,41,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,20001.392001,20614.835005,1,100,8,8,2.62,3,3,0,1,0,1,41,2,5,1,5
+71268,7,2,1,5,NA,3,3,2,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,26500.001625,29898.299515,2,91,7,7,1.29,6,6,2,2,0,1,33,2,3,6,NA
+71269,7,1,1,33,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,12556.207754,0,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+71270,7,2,2,15,NA,2,2,2,15,185,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17381.663677,18792.085501,1,90,7,7,2.1,3,3,0,2,0,2,37,1,3,5,NA
+71271,7,2,1,57,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,31872.125984,31404.451666,2,96,6,6,1.12,4,4,0,1,0,1,57,2,1,1,4
+71272,7,2,1,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,77440.662578,92128.223811,2,93,8,8,3.3,2,2,0,0,0,2,26,1,4,1,5
+71273,7,2,1,11,NA,1,1,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,13933.628196,2,98,6,6,1.9,2,2,0,1,0,2,34,1,4,3,NA
+71274,7,2,2,28,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,6,2,1,2,2,1,2,2,1,2,2,1,49473.624024,50829.110777,3,91,14,14,5,2,2,0,0,0,1,32,1,4,6,NA
+71275,7,2,2,12,NA,4,4,1,12,148,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,20473.346601,20523.769333,2,102,15,15,3.82,5,5,1,2,0,1,34,1,3,1,4
+71276,7,2,2,0,5,3,3,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17070.097848,16606.436082,1,101,7,7,1.79,4,4,1,0,0,2,30,1,4,1,4
+71277,7,2,2,4,NA,1,1,1,4,52,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15457.736897,17065.756351,2,98,2,2,0.27,4,4,2,1,0,2,32,2,2,5,NA
+71278,7,2,1,69,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7101.739553,7379.502561,1,96,99,99,NA,1,1,0,0,1,1,69,1,2,1,NA
+71279,7,2,1,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,27356.080541,27285.808403,1,98,6,6,1.11,5,5,1,2,0,2,32,1,2,1,2
+71280,7,2,2,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,95778.224595,101717.815372,1,98,8,8,2.46,3,3,1,0,0,1,29,1,4,6,NA
+71281,7,2,1,2,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7667.509064,7761.032365,2,91,2,2,0.41,3,3,1,1,0,2,30,1,3,5,NA
+71282,7,2,2,25,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,12426.473257,13266.565637,2,92,14,7,3.4,2,1,0,0,0,1,26,NA,NA,5,NA
+71283,7,2,2,0,1,2,2,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,8168.833251,8141.91987,2,91,5,5,1.05,3,3,2,0,0,2,26,2,3,4,NA
+71284,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,10049.7347,10691.470743,2,90,2,2,0.67,2,2,0,0,1,2,64,1,5,5,NA
+71285,7,2,1,10,NA,1,1,1,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13927.458372,14258.502552,2,98,5,5,1.07,4,4,0,1,0,1,52,2,1,1,3
+71286,7,2,2,56,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,22969.116046,23497.145655,2,93,6,6,2.69,1,1,0,0,0,2,56,2,4,3,NA
+71287,7,1,2,3,NA,3,3,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,26966.264969,0,1,98,4,4,0.66,4,4,2,0,0,2,22,1,4,6,NA
+71288,7,2,1,68,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,140586.349952,144513.945685,2,91,15,15,5,2,2,0,0,2,1,68,1,5,1,5
+71289,7,2,2,32,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,27375.405353,27906.494528,1,94,5,5,0.95,4,4,1,2,0,2,32,1,4,3,NA
+71290,7,2,2,10,NA,2,2,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12737.089292,13281.887291,1,94,14,14,3.4,5,5,0,3,0,2,41,1,4,1,4
+71291,7,2,1,50,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,182279.501656,183739.903765,1,100,6,6,1.98,2,2,0,0,0,1,50,1,5,4,NA
+71292,7,2,2,36,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,28747.860416,30448.398533,1,99,10,10,3.13,4,4,0,2,0,1,35,1,4,1,5
+71293,7,2,1,3,NA,2,2,2,3,41,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,13353.801759,13870.061432,1,93,6,6,0.99,5,5,1,0,0,1,28,2,2,5,NA
+71294,7,2,1,16,NA,4,4,1,16,201,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19163.050164,19648.623212,2,102,14,14,4.48,3,3,0,2,0,1,41,1,3,4,NA
+71295,7,2,1,25,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,110218.970616,129453.540645,1,93,7,7,3.21,1,1,0,0,0,1,25,1,4,5,NA
+71296,7,2,2,32,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,27367.658704,28808.672115,1,91,4,4,0.81,4,4,1,1,0,1,32,1,4,6,NA
+71297,7,2,1,19,NA,4,4,1,19,230,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12147.046136,12703.852077,2,100,4,4,0.91,3,3,0,0,0,2,49,1,2,1,2
+71298,7,2,2,7,NA,5,6,2,7,90,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10699.45895,11403.206057,1,97,14,14,2.29,7,7,1,2,2,1,40,2,1,1,1
+71299,7,2,1,2,NA,4,4,1,2,26,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,11283.202594,2,101,4,4,1.22,2,2,1,0,0,2,25,1,4,5,NA
+71300,7,2,2,35,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,12381.032047,12845.540957,1,102,15,15,4.59,4,4,1,1,0,1,35,1,5,1,5
+71301,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,44417.237618,45651.906489,1,102,1,1,0.24,3,2,0,0,1,2,50,1,3,3,NA
+71302,7,2,2,56,NA,4,4,1,NA,NA,2,NA,2,2,6,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,18146.994087,18265.36374,2,93,4,4,1.29,2,2,0,1,0,2,56,2,3,4,NA
+71303,7,2,2,4,NA,4,4,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9707.610673,10138.677842,2,93,5,5,0.92,4,4,1,1,0,1,27,2,2,5,NA
+71304,7,2,1,1,19,1,1,2,NA,19,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,11972.170914,12351.142913,1,97,8,8,1.45,6,6,2,2,0,2,36,2,2,1,1
+71305,7,2,1,7,NA,1,1,1,7,89,NA,NA,1,1,NA,0,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,14820.807433,15055.338709,2,102,7,7,1.89,3,3,0,1,0,1,41,2,2,6,NA
+71306,7,2,2,52,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,12441.719186,12605.354834,2,100,15,15,5,3,3,0,1,0,1,58,2,5,1,5
+71307,7,2,1,58,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,132969.642582,132807.505002,2,91,5,5,1.39,2,2,0,0,0,1,58,1,4,3,NA
+71308,7,2,2,69,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,8493.33267,8821.144988,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+71309,7,2,2,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,NA,25964.813959,27606.791932,2,101,NA,77,NA,2,1,0,0,0,1,55,1,2,3,NA
+71310,7,2,2,67,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,11879.290971,12409.688606,2,96,6,6,2.04,2,2,0,0,1,2,67,1,4,2,NA
+71311,7,2,2,80,NA,2,2,1,NA,NA,2,NA,2,1,9,NA,2,1,NA,2,2,2,1,2,2,2,2,2,NA,21810.940874,23452.513609,2,93,8,8,2.17,4,4,0,0,3,1,80,2,2,1,2
+71312,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15214.43631,2,100,8,8,2.67,3,3,0,0,1,1,61,1,3,1,4
+71313,7,2,1,52,NA,4,4,2,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,19374.410926,19463.695548,1,96,77,77,NA,4,4,0,0,0,1,52,2,5,1,5
+71314,7,2,1,51,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,30839.213846,30386.695922,1,92,4,4,0.74,4,4,1,1,0,1,51,2,1,1,1
+71315,7,2,1,44,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,105141.812429,112387.793787,1,98,6,6,1.11,5,5,0,2,1,2,37,1,1,1,1
+71316,7,2,1,57,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,100418.893093,106630.269812,2,100,10,10,3.13,4,4,0,0,1,2,53,1,2,1,2
+71317,7,2,1,0,1,4,4,2,NA,2,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5848.582979,6109.207769,2,97,4,4,0.57,5,5,1,3,0,2,33,1,3,5,NA
+71318,7,2,2,2,NA,4,4,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6984.912286,7371.360851,2,93,10,10,2.26,6,6,2,0,0,1,34,1,4,1,4
+71319,7,2,2,2,NA,3,3,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,33502.281129,33942.512387,1,91,15,15,5,4,4,1,1,0,1,29,1,4,6,NA
+71320,7,1,1,2,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,10273.602479,0,1,92,12,12,NA,4,4,1,1,0,1,33,2,4,1,4
+71321,7,2,2,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,24713.281483,24036.937705,1,92,14,14,4.96,2,2,0,0,0,1,25,1,4,5,NA
+71322,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,20949.647122,21006.826162,2,91,4,4,0.84,3,3,1,0,0,2,21,1,4,1,2
+71323,7,2,1,9,NA,5,6,2,9,117,NA,NA,2,2,1,3,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,5855.595238,6345.238788,1,91,14,14,4.32,3,3,0,2,0,2,37,2,5,1,NA
+71324,7,2,1,35,NA,2,2,2,NA,NA,2,NA,2,2,5,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,38923.140296,40970.697433,1,93,9,9,2.46,4,4,0,2,0,1,35,2,1,1,1
+71325,7,2,2,3,NA,3,3,2,3,42,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24600.429016,26143.360672,1,91,6,6,1.07,6,6,3,1,0,2,27,1,4,6,NA
+71326,7,2,1,6,NA,4,4,2,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8655.162127,8668.652185,2,95,6,6,0.86,6,6,0,4,0,2,32,1,4,6,NA
+71327,7,2,1,3,NA,1,1,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16775.083123,16797.750213,2,98,2,2,0.27,4,4,2,0,0,2,20,2,2,6,NA
+71328,7,2,2,54,NA,2,2,1,NA,NA,2,NA,2,1,4,NA,4,3,NA,2,2,2,1,2,2,2,2,2,2,26226.50904,26977.257389,2,93,7,7,1.74,4,4,1,0,1,2,24,NA,NA,4,NA
+71329,7,2,1,8,NA,3,3,2,8,100,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,63513.013195,68899.034203,2,101,7,7,1.57,4,4,0,2,0,2,28,1,3,6,NA
+71330,7,2,1,39,NA,5,6,1,NA,NA,2,NA,2,1,3,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,9585.652672,9712.791327,2,92,10,6,1.12,7,4,1,1,1,2,27,2,3,1,3
+71331,7,2,2,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,9767.083234,9561.595244,3,92,10,10,2.82,4,4,2,0,0,1,26,2,3,1,3
+71332,7,2,2,10,NA,1,1,1,10,129,NA,NA,1,1,NA,4,NA,NA,NA,2,1,1,1,2,1,1,2,2,1,20495.125801,20710.596821,3,92,6,6,0.96,5,5,2,1,0,2,26,2,1,1,1
+71333,7,2,2,30,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,52521.960331,51104.807504,1,92,10,10,4.63,2,2,0,0,0,1,35,2,5,1,4
+71334,7,2,1,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,15682.233511,17550.107712,1,98,6,6,1.9,2,2,0,0,2,1,80,1,1,1,2
+71335,7,1,1,61,NA,2,2,NA,NA,NA,2,NA,2,1,6,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,6449.12882,0,2,93,77,77,NA,3,3,0,0,3,1,61,2,1,1,1
+71336,7,2,1,45,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14695.245859,15160.543049,2,92,15,15,5,3,3,0,1,0,1,45,2,4,1,4
+71337,7,2,1,7,NA,3,3,2,7,87,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64268.100156,67147.933534,1,98,15,15,4.07,5,5,0,3,0,1,38,1,2,1,4
+71338,7,2,2,8,NA,4,4,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11195.065587,11638.702248,2,96,2,2,0.43,3,3,0,2,0,2,50,1,2,5,NA
+71339,7,2,2,69,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,39030.893313,39804.103933,1,91,5,5,1.2,3,3,0,0,1,1,58,1,2,1,2
+71340,7,2,2,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,24919.497762,26574.647337,1,95,5,5,1.1,3,3,0,0,1,2,63,1,4,1,5
+71341,7,2,1,58,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,14897.510053,14959.811939,1,101,15,15,5,3,3,0,0,1,1,58,2,4,1,5
+71342,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,46965.818538,52174.805072,1,101,7,7,2.64,2,2,0,0,2,1,80,1,1,1,3
+71343,7,2,2,29,NA,3,3,2,NA,NA,2,NA,2,1,6,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,105134.853452,122679.602456,2,99,14,14,5,1,1,0,0,0,2,29,2,5,5,NA
+71344,7,2,1,65,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,28953.774534,29107.622897,1,98,4,4,1,3,3,0,1,1,1,65,1,2,1,NA
+71345,7,1,1,60,NA,1,1,NA,NA,NA,2,NA,2,2,77,NA,2,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,12845.115724,0,1,100,4,4,0.78,4,4,0,0,1,1,33,2,1,1,1
+71346,7,2,2,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,NA,44856.466004,46390.551109,2,94,12,9,5,2,1,0,0,1,2,74,1,5,3,NA
+71347,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,32461.799549,32422.216998,1,95,6,2,0.81,2,1,0,0,0,1,53,1,2,6,NA
+71348,7,2,1,56,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,25118.469449,26051.81361,2,98,6,6,1.4,3,3,0,1,0,1,56,1,2,1,2
+71349,7,1,1,78,NA,5,6,NA,NA,NA,2,NA,2,1,6,NA,1,2,NA,1,2,1,1,2,2,NA,NA,NA,NA,9748.579573,0,3,90,8,8,1.85,5,5,0,0,1,2,25,1,5,5,NA
+71350,7,2,2,6,NA,4,4,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9502.76472,9802.874785,1,96,3,3,0.93,2,2,0,1,0,2,40,1,5,5,NA
+71351,7,2,2,52,NA,5,6,1,NA,NA,2,NA,2,1,8,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,11876.875273,12162.823626,2,103,15,15,5,3,3,0,0,0,2,52,2,4,1,5
+71352,7,2,2,10,NA,5,6,2,10,130,NA,NA,2,1,3,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9227.090107,9685.088635,2,100,5,5,0.89,4,4,0,1,0,2,40,2,3,1,3
+71353,7,2,2,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,36067.495928,37024.300659,1,95,3,3,1.29,1,1,0,0,1,2,76,1,3,2,NA
+71354,7,2,1,50,NA,1,1,1,NA,NA,1,1,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,37557.946192,37666.079092,2,102,7,7,3.49,4,1,0,0,1,1,63,NA,NA,5,NA
+71355,7,2,2,20,NA,5,6,1,NA,NA,2,NA,2,1,4,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,16026.446928,16774.880842,1,94,14,14,2.78,6,5,0,2,1,1,61,1,4,1,5
+71356,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,129867.826953,131085.811251,1,101,9,9,2.6,4,4,0,1,2,2,63,1,4,1,4
+71357,7,2,1,29,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15408.94893,15333.002917,1,99,7,7,2.38,2,2,0,0,0,2,27,1,5,1,5
+71358,7,2,2,1,22,2,2,1,NA,23,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,7815.664341,8308.38945,2,103,5,5,0.65,6,6,1,0,1,2,61,2,1,2,NA
+71359,7,2,1,41,NA,2,2,2,NA,NA,2,NA,2,7,77,NA,1,1,NA,2,2,2,2,2,2,NA,NA,NA,NA,31640.296506,33247.715312,2,94,1,1,0.01,7,7,1,3,0,1,41,2,1,1,1
+71360,7,2,2,25,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,2,1,2,2,2,2,2,2,2,NA,NA,NA,NA,39426.061521,43773.817687,2,94,4,4,0.81,4,4,2,0,0,1,26,2,2,1,2
+71361,7,2,2,55,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,17852.668137,17947.072019,3,91,7,7,2.58,2,2,0,0,0,2,55,2,5,3,NA
+71362,7,2,1,23,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,25815.880139,26503.609729,2,101,2,1,0.05,2,1,0,0,0,1,24,1,4,5,NA
+71363,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,2,1,2,2,NA,NA,NA,1,2,2,1,16165.066826,16696.869727,2,96,NA,NA,NA,4,4,0,0,1,1,67,2,3,1,2
+71364,7,2,1,13,NA,4,4,1,13,157,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11666.009872,12200.765691,2,100,14,4,0.43,7,7,1,3,1,2,62,1,3,5,NA
+71365,7,2,1,22,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21763.209029,22385.504537,2,92,4,4,1.38,4,1,0,0,0,1,21,1,4,5,NA
+71366,7,2,2,20,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,114492.825466,119073.624167,2,101,3,1,0.06,3,1,0,0,0,2,20,2,4,5,NA
+71367,7,2,1,7,NA,2,2,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13533.281742,13475.903543,1,100,10,10,2.91,4,4,1,1,0,1,32,1,5,1,5
+71368,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,82363.088109,88909.58429,1,90,14,14,3.93,3,3,1,0,0,1,35,1,2,1,5
+71369,7,2,2,3,NA,5,6,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3842.794137,4026.588867,1,93,6,6,1.15,5,5,1,0,2,2,70,NA,NA,1,NA
+71370,7,2,1,9,NA,3,3,2,9,114,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,21087.869274,22686.932895,1,101,1,1,0.1,4,4,1,1,0,2,52,1,4,3,NA
+71371,7,2,2,63,NA,5,7,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,28759.144922,29766.431764,3,90,7,7,2.51,2,2,0,0,2,1,62,2,3,1,4
+71372,7,2,1,37,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,1,6,NA,2,2,2,2,2,2,2,2,2,2,51543.062078,51154.050295,3,92,3,3,0.51,5,5,1,2,0,2,34,2,1,6,NA
+71373,7,2,1,1,15,1,1,1,NA,15,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9925.934568,9762.398236,2,92,5,5,1.08,3,3,1,0,0,2,30,2,3,1,2
+71374,7,2,2,52,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,12059.495297,12169.751033,1,102,15,15,3.82,5,5,1,1,0,1,29,1,4,1,4
+71375,7,2,1,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,70819.746399,73506.573727,1,98,6,6,1.98,2,2,0,0,2,2,70,1,4,1,5
+71376,7,2,1,3,NA,1,1,1,3,39,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18754.85406,20094.472571,2,102,14,14,3.8,4,4,2,0,0,2,41,2,4,1,5
+71377,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,27199.141352,28737.286192,1,98,77,77,NA,2,2,0,0,2,1,80,1,5,1,5
+71378,7,2,1,5,NA,3,3,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39061.897809,44649.899401,1,95,15,15,3.62,7,7,2,4,0,1,59,1,5,1,2
+71379,7,2,2,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,27351.473517,27893.312251,2,94,2,2,0.83,1,1,0,0,1,2,60,1,5,5,NA
+71380,7,1,2,80,NA,4,4,NA,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,15914.916287,0,2,99,5,5,1.59,2,2,0,0,1,2,80,1,3,2,NA
+71381,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,1,2,2,3,12765.378725,13670.003215,2,90,14,14,4.32,3,3,0,1,0,1,48,2,4,1,5
+71382,7,2,2,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,54095.581484,61529.339683,2,94,9,9,3.97,2,2,0,0,2,1,80,1,5,1,5
+71383,7,2,2,3,NA,5,6,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,2,NA,NA,NA,NA,5060.292252,5371.954509,1,90,8,8,1.43,7,7,2,0,0,1,23,2,4,1,3
+71384,7,2,2,7,NA,1,1,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15962.145468,16285.000268,3,92,7,7,1.49,5,5,0,2,1,2,62,1,4,2,NA
+71385,7,2,2,6,NA,3,3,2,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,28723.999699,28637.704803,1,97,6,6,1.03,6,6,2,2,0,2,38,1,5,1,4
+71386,7,2,2,1,23,1,1,1,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14326.094268,14518.763259,3,92,14,14,3.47,4,4,1,1,0,1,44,1,3,1,5
+71387,7,2,2,3,NA,4,4,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10437.988787,11378.89676,2,97,NA,99,NA,7,6,2,1,1,2,56,1,3,5,NA
+71388,7,2,1,42,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,49060.272708,48340.388834,1,92,7,7,1.48,5,5,0,1,0,1,42,1,5,1,4
+71389,7,2,2,1,21,2,2,1,NA,22,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11443.671453,12634.120381,1,100,15,15,4.34,4,4,2,0,0,2,35,1,5,1,5
+71390,7,2,1,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,30626.145273,33111.125584,2,103,7,7,2.64,2,2,0,0,1,1,54,1,5,5,NA
+71391,7,2,1,11,NA,3,3,1,11,137,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,38712.032122,41122.784973,3,91,15,15,5,6,6,1,3,0,2,40,1,5,1,5
+71392,7,2,2,43,NA,2,2,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,35485.230645,37305.51435,1,94,7,7,1.74,4,4,0,2,0,1,44,1,5,1,5
+71393,7,2,1,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,108410.783716,113500.095101,1,91,15,7,3.58,3,1,0,0,0,1,27,1,3,6,NA
+71394,7,1,1,78,NA,5,6,NA,NA,NA,2,NA,2,1,5,NA,3,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,12320.168636,0,2,92,4,4,1.09,2,2,0,0,2,1,78,2,3,1,NA
+71395,7,2,1,30,NA,2,2,2,NA,NA,2,NA,2,2,6,NA,1,4,NA,2,2,2,2,2,2,2,2,2,2,27605.196104,27396.850962,2,99,12,7,3.31,5,1,0,1,0,1,30,2,2,4,NA
+71396,7,2,2,3,NA,4,4,2,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10936.097083,12221.030117,2,95,2,2,0.42,3,3,2,0,0,1,25,1,3,5,NA
+71397,7,2,1,24,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,4,NA,1,2,2,1,2,2,1,2,2,1,23484.626749,26526.417464,2,96,14,14,5,1,1,0,0,0,1,24,1,4,4,NA
+71398,7,2,1,13,NA,4,4,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11788.330261,12582.159913,2,95,6,6,0.97,6,6,2,2,0,1,37,1,3,1,4
+71399,7,2,1,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,103893.364631,104494.238956,2,95,9,9,2.22,5,5,1,0,0,1,55,1,4,1,5
+71400,7,1,1,18,NA,5,6,NA,NA,NA,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,7669.677276,0,2,95,3,3,0.99,2,2,0,0,0,1,18,1,3,NA,NA
+71401,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,31785.728924,32871.467684,1,101,4,4,1.22,2,2,0,0,1,1,63,1,2,1,2
+71402,7,2,1,0,3,3,3,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21055.122405,20691.125102,1,93,15,15,4.47,4,4,2,0,0,1,31,1,5,1,5
+71403,7,2,2,30,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,31335.13799,31552.004994,1,95,5,5,0.92,5,5,1,2,0,2,30,1,4,1,4
+71404,7,2,1,18,NA,4,4,1,18,219,2,NA,1,1,NA,66,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13276.485807,15665.287637,2,100,99,99,NA,3,3,0,0,0,1,46,1,9,3,NA
+71405,7,2,1,63,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,6910.118936,7233.371983,2,95,4,4,1.77,1,1,0,0,1,1,63,1,2,3,NA
+71406,7,2,2,8,NA,4,4,1,8,106,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8558.832746,8898.001102,2,96,3,3,0.38,5,5,1,2,0,2,30,1,3,5,NA
+71407,7,2,2,49,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,20418.635225,20918.390367,2,95,14,14,5,2,2,0,0,0,1,44,1,3,1,4
+71408,7,2,2,19,NA,2,2,2,19,238,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,16995.403684,17639.03984,2,97,2,2,0.27,3,3,2,0,0,2,19,1,3,NA,NA
+71409,7,2,1,41,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,116464.874823,116165.700541,2,94,14,14,2.83,6,6,0,4,0,2,38,1,2,1,2
+71410,7,2,2,15,NA,3,3,2,15,182,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,103007.696238,104941.393061,1,101,15,15,5,4,4,0,2,0,1,43,1,4,1,5
+71411,7,2,1,8,NA,4,4,1,8,103,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,NA,9261.557132,9782.87535,2,100,1,1,0.08,5,5,1,2,0,2,19,1,3,NA,NA
+71412,7,2,2,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,35269.9609,35902.08788,2,96,14,14,3.58,4,4,2,0,0,1,36,1,4,1,5
+71413,7,2,1,40,NA,2,2,1,NA,NA,2,NA,2,2,77,NA,4,6,NA,2,2,2,1,2,2,1,2,2,2,28726.575428,28964.259011,2,100,3,3,0.76,3,3,1,0,0,2,31,2,1,6,NA
+71414,7,2,1,17,NA,1,1,1,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24228.858782,24355.020132,2,102,77,77,NA,4,4,0,1,0,1,47,1,2,1,3
+71415,7,2,1,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,78529.577822,86568.047729,1,98,7,7,1,7,7,2,2,0,2,34,1,4,3,NA
+71416,7,2,2,39,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,16411.593279,16445.104385,2,103,15,15,5,2,2,0,0,0,2,39,2,5,5,NA
+71417,7,2,1,64,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,12279.19827,12375.250021,2,102,6,6,2.04,2,2,0,0,1,1,64,1,3,1,4
+71418,7,2,2,7,NA,5,6,2,7,90,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10699.45895,11230.540406,1,97,15,15,4.77,4,4,1,1,0,2,40,1,5,1,5
+71419,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,NA,NA,NA,NA,74517.751389,77393.175383,2,94,15,15,4.59,4,4,1,1,0,2,37,1,5,1,5
+71420,7,2,2,54,NA,1,1,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,2,2,2,1,2,2,2,2,2,2,23953.14388,24078.057488,1,90,12,12,NA,4,4,0,0,0,1,54,2,4,1,2
+71421,7,2,1,3,NA,3,3,1,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,32621.433667,37288.094364,2,101,3,3,0.3,7,7,1,2,0,2,50,1,2,4,NA
+71422,7,2,2,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,77778.949308,80709.488833,1,101,14,14,4.21,4,4,1,1,0,2,37,1,5,1,5
+71423,7,2,1,27,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,12540.575493,12881.998163,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+71424,7,2,1,19,NA,3,3,1,19,236,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,110478.18082,109645.964567,2,101,1,1,0.09,2,1,0,0,0,1,19,1,4,NA,NA
+71425,7,2,2,13,NA,3,3,2,13,167,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,35589.982806,42898.287648,1,101,6,6,1.31,3,3,0,2,0,1,43,1,3,4,NA
+71426,7,2,1,3,NA,4,4,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9016.053035,9035.668246,2,100,3,3,0.31,7,7,3,2,0,2,28,1,3,1,3
+71427,7,2,2,39,NA,5,6,1,NA,NA,2,NA,2,2,1,NA,5,1,2,1,2,1,1,2,1,NA,NA,NA,NA,15951.963269,17051.924396,3,91,14,14,3.58,4,4,1,1,0,1,39,2,5,1,5
+71428,7,2,1,35,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,99741.609501,104268.100538,1,94,6,6,1.57,3,3,0,1,0,2,28,1,4,1,4
+71429,7,2,2,9,NA,4,4,2,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8579.490652,8729.153641,2,97,3,3,0.4,6,6,2,3,0,2,25,1,2,5,NA
+71430,7,2,2,65,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,1,10468.473646,11728.156896,2,90,7,7,1.61,4,4,1,1,1,2,65,1,3,2,NA
+71431,7,2,2,2,NA,4,4,1,2,32,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7382.686927,8048.181894,2,100,3,3,0.27,7,7,2,1,0,2,41,1,2,5,NA
+71432,7,2,2,13,NA,1,1,1,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,19329.26314,19883.967368,1,103,14,14,2.96,5,5,1,2,0,1,34,1,4,1,5
+71433,7,2,1,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,19020.83925,19736.003798,2,99,6,6,1.13,4,4,1,1,0,1,33,1,3,6,NA
+71434,7,2,1,17,NA,5,6,2,17,214,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9202.202094,9789.088627,1,90,15,15,5,5,5,0,2,0,1,47,2,5,1,5
+71435,7,2,1,4,NA,5,6,2,4,49,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6564.381324,7362.671865,2,90,77,77,NA,4,3,1,0,0,2,30,2,2,5,NA
+71436,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,11355.3308,11862.334174,1,96,9,9,4.13,2,2,0,0,1,2,55,1,4,5,NA
+71437,7,1,2,3,NA,4,4,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,9707.610673,0,2,93,1,1,0.2,2,2,1,0,0,2,33,2,3,5,NA
+71438,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,13549.492282,14020.967918,2,93,12,12,NA,4,4,0,0,2,1,72,1,2,1,4
+71439,7,2,2,3,NA,3,3,2,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,64475.568487,69419.338511,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+71440,7,2,1,10,NA,1,1,1,10,124,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12665.770043,13116.669484,1,100,9,9,2.02,6,6,0,3,1,2,39,1,4,1,5
+71441,7,2,2,12,NA,2,2,2,12,153,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,14437.97544,16344.869867,2,90,3,3,0.38,5,5,0,4,0,2,33,2,2,5,NA
+71442,7,2,1,1,12,3,3,1,NA,12,NA,NA,2,2,1,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,21380.815125,22997.240081,1,103,15,15,5,3,3,1,0,0,1,26,2,5,1,5
+71443,7,2,1,7,NA,1,1,2,7,94,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,12477.812875,12553.43469,2,94,4,4,0.63,6,6,1,2,0,2,36,2,3,1,1
+71444,7,2,2,0,3,5,7,2,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6399.50062,6586.636465,1,101,2,2,0.33,4,4,2,1,0,2,26,1,4,5,NA
+71445,7,2,1,2,NA,2,2,1,2,30,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,12005.116852,12547.788444,3,92,6,6,0.86,7,7,1,4,0,2,36,2,1,1,1
+71446,7,2,2,11,NA,1,1,1,11,135,NA,NA,2,2,3,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,16986.005478,17733.622754,2,102,4,4,0.57,5,5,0,3,0,1,41,2,1,1,2
+71447,7,2,2,1,19,4,4,2,NA,20,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7348.24433,8211.623817,2,95,2,2,0.33,2,2,1,0,0,2,21,1,1,5,NA
+71448,7,2,2,6,NA,4,4,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9453.111053,10094.338553,1,100,6,6,1.39,4,4,0,3,0,2,29,1,4,5,NA
+71449,7,2,2,40,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,3,2,1,2,2,1,2,2,1,2,2,1,128187.954928,128313.829811,1,99,12,7,3.81,2,1,0,0,0,2,40,1,4,3,NA
+71450,7,2,2,34,NA,3,3,1,NA,NA,2,NA,2,2,2,NA,4,1,2,2,2,2,2,2,2,NA,NA,NA,NA,66218.994273,67969.524045,2,93,7,7,1.52,4,4,1,1,0,1,44,2,4,1,NA
+71451,7,2,2,51,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,24217.957803,24014.890408,2,94,4,4,1.56,2,1,0,0,0,2,51,1,2,6,NA
+71452,7,2,1,36,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,96897.480702,107123.014093,2,91,15,15,5,4,4,2,0,0,2,33,1,5,1,5
+71453,7,2,1,14,NA,5,6,1,14,171,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11657.164593,12175.606972,1,92,9,9,2.88,3,3,0,2,0,1,50,2,5,3,NA
+71454,7,2,2,8,NA,3,3,1,8,106,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,18417.272091,18465.044105,1,94,5,5,1.04,4,4,0,2,0,2,29,1,3,1,3
+71455,7,2,1,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,NA,13838.08267,16302.45001,1,94,3,3,1.16,1,1,0,0,1,1,70,1,3,5,NA
+71456,7,2,1,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,59473.789098,72577.011806,3,92,4,1,0.27,5,1,0,0,0,2,57,1,4,1,2
+71457,7,2,2,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,23325.73926,22687.369215,1,101,2,2,0.68,1,1,0,0,0,2,53,1,4,2,NA
+71458,7,2,1,5,NA,5,6,1,5,66,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,1,NA,NA,NA,NA,8838.066777,9670.229845,3,92,77,77,NA,7,7,2,4,1,1,62,NA,NA,1,NA
+71459,7,2,2,14,NA,3,3,2,14,176,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,93307.64449,95903.481223,2,94,15,15,5,3,3,0,1,0,1,49,1,3,1,1
+71460,7,2,2,47,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,19075.861607,19022.257361,1,96,15,15,5,4,4,1,1,0,1,50,1,3,1,4
+71461,7,2,1,7,NA,4,4,1,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13423.881856,13521.226684,2,101,8,8,2.43,3,3,0,1,0,1,35,1,4,6,NA
+71462,7,2,1,7,NA,4,4,2,7,91,NA,NA,2,1,1,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10424.657432,11237.830079,1,93,6,6,1.35,3,3,0,1,0,1,32,2,4,1,4
+71463,7,2,1,0,4,1,1,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6298.658963,6651.880632,2,94,15,15,3.33,6,6,1,2,0,2,20,1,3,1,NA
+71464,7,2,2,3,NA,1,1,2,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13366.393396,13791.68675,2,94,5,5,1.07,4,4,2,0,0,1,37,2,1,1,1
+71465,7,2,2,26,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,18070.666316,19871.407536,1,99,6,6,2.75,1,1,0,0,0,2,26,2,5,4,NA
+71466,7,2,1,2,NA,2,2,1,2,33,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,11727.291842,11865.933582,2,91,2,2,0.19,5,5,3,0,0,1,24,2,1,1,3
+71467,7,2,1,48,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,141134.499671,148176.982862,2,101,8,8,2.81,3,3,0,2,0,1,48,1,3,3,NA
+71468,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,17191.889135,17295.278265,2,93,2,2,0.75,1,1,0,0,0,2,50,1,3,3,NA
+71469,7,2,1,12,NA,4,4,2,12,155,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9313.795042,9381.627752,1,96,77,77,NA,7,7,1,3,0,1,56,1,3,1,4
+71470,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,52209.836905,58077.47757,1,98,8,8,3.48,2,2,0,0,2,1,80,1,3,1,2
+71471,7,2,1,10,NA,3,3,2,10,123,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,44777.275016,47565.734765,1,91,7,7,1.88,4,4,1,2,0,2,43,1,5,4,NA
+71472,7,2,1,32,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,16058.989596,15830.0816,2,97,3,3,0.33,6,6,2,0,0,2,32,1,2,1,3
+71473,7,2,2,38,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,34007.843648,34076.908664,2,98,10,10,4.76,2,2,0,0,0,1,42,1,2,6,NA
+71474,7,2,1,25,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,18509.151292,19027.688747,2,95,2,2,0.42,3,3,2,0,0,1,25,1,3,5,NA
+71475,7,2,2,12,NA,3,3,1,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39616.634313,47751.80094,2,101,3,3,0.59,4,3,0,2,0,1,39,1,1,6,NA
+71476,7,2,2,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,35375.447072,38057.027004,1,97,8,8,2.51,3,3,0,1,0,1,35,1,3,1,4
+71477,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,12842.559946,13702.031196,2,95,13,13,NA,2,2,0,0,2,2,80,1,1,1,NA
+71478,7,2,1,43,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,6,NA,1,2,2,1,2,2,1,2,2,1,35406.972937,35535.728875,2,98,7,7,1.53,5,5,0,0,0,2,48,1,3,5,NA
+71479,7,2,2,15,NA,5,6,2,15,190,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,12224.9472,12547.183925,1,97,14,14,2.72,7,7,0,2,0,1,40,1,5,1,5
+71480,7,2,2,5,NA,4,4,2,5,65,NA,NA,2,1,2,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10544.002566,11127.362026,1,91,7,7,1.49,5,5,3,0,0,2,38,2,4,1,4
+71481,7,2,2,26,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,50915.06085,56529.78018,3,92,7,7,1.65,4,4,1,1,0,1,27,1,3,1,3
+71482,7,2,2,40,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,19075.861607,18884.31701,1,96,8,8,2,4,4,1,2,0,2,40,1,4,5,NA
+71483,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,49191.372812,61149.379277,3,91,7,7,2.45,2,2,0,0,2,1,80,1,2,1,2
+71484,7,1,1,39,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,113827.590402,0,2,91,7,7,1.89,3,3,0,0,2,2,69,NA,NA,1,4
+71485,7,1,1,18,NA,5,6,NA,NA,NA,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,6666.045669,0,3,90,8,8,1.85,5,5,0,0,1,2,25,1,5,5,NA
+71486,7,2,2,11,NA,4,4,2,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8549.049441,9128.952244,2,99,6,6,1.73,3,3,1,1,1,2,60,1,4,3,NA
+71487,7,2,2,13,NA,5,6,1,13,159,NA,NA,2,2,4,7,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,6145.01663,6410.621394,2,92,6,6,1.3,4,4,0,1,0,2,48,2,3,1,3
+71488,7,2,2,41,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,23409.362971,23966.406748,2,96,5,5,1.08,3,3,0,1,0,2,41,1,3,1,NA
+71489,7,2,2,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,23545.494186,22594.243077,2,99,8,8,3.47,2,2,0,0,1,2,74,1,5,2,NA
+71490,7,2,2,11,NA,1,1,1,11,133,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13490.527555,13974.998066,2,96,5,5,0.68,6,6,0,3,2,1,60,2,1,1,1
+71491,7,2,1,12,NA,3,3,1,13,156,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,23588.786918,23619.012758,1,94,7,7,1.29,6,6,1,3,0,1,38,1,3,1,2
+71492,7,2,2,51,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,32041.645327,31772.976371,3,91,7,7,1.1,7,7,0,4,0,1,40,1,4,1,3
+71493,7,2,1,14,NA,1,1,1,15,180,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22768.423624,22944.003607,2,98,8,8,3.06,2,2,0,1,0,2,56,1,3,3,NA
+71494,7,2,1,29,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,17191.065623,18600.361918,1,100,5,5,1.79,2,1,0,0,0,1,27,NA,NA,5,NA
+71495,7,2,2,28,NA,2,2,2,NA,NA,2,NA,2,1,6,NA,3,6,2,1,2,2,1,2,2,1,2,2,1,43200.700326,51577.05558,1,97,5,5,1.04,4,4,1,1,0,1,32,1,3,6,NA
+71496,7,2,1,65,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,1,4,NA,2,2,2,2,2,2,1,2,2,2,6564.099986,6927.048144,2,95,12,3,1.07,4,1,0,0,1,1,65,2,1,4,NA
+71497,7,2,2,71,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,NA,27602.403077,28546.401551,1,94,2,2,0.55,2,2,0,0,2,2,75,1,1,3,NA
+71498,7,2,2,5,NA,4,4,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10389.292229,10850.629517,2,95,4,4,0.65,4,4,1,2,0,2,27,1,3,5,NA
+71499,7,2,1,16,NA,4,4,2,16,197,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11834.781205,12063.950506,2,90,15,15,5,4,4,1,1,0,1,53,2,5,1,5
+71500,7,2,1,6,NA,1,1,1,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13927.458372,14007.413517,2,98,15,15,4.97,5,5,0,3,0,1,39,1,5,1,5
+71501,7,2,2,2,NA,3,3,2,2,29,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,50995.241584,52601.79766,2,91,14,14,4.12,4,4,2,0,0,1,35,1,5,1,5
+71502,7,2,1,16,NA,1,1,1,16,193,NA,NA,1,1,NA,8,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,19570.996814,19477.677478,2,100,4,4,0.81,4,4,0,2,0,1,56,1,4,1,2
+71503,7,2,1,30,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21200.922169,24170.169783,2,99,9,9,4.92,1,1,0,0,0,1,30,1,4,5,NA
+71504,7,2,2,16,NA,1,1,1,16,200,NA,NA,2,2,4,10,NA,NA,NA,2,2,2,1,2,2,1,2,2,1,22753.900764,24458.613732,1,100,2,2,0.35,4,4,0,1,0,2,40,2,2,5,NA
+71505,7,2,1,33,NA,2,2,2,NA,NA,2,NA,2,1,3,NA,3,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,43108.74283,42989.380022,2,91,7,7,2.64,2,2,0,1,0,1,33,2,3,1,NA
+71506,7,2,2,0,2,5,7,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9148.090461,9086.623516,1,92,9,9,3.14,3,3,1,0,0,2,30,1,5,1,5
+71507,7,2,1,1,15,1,1,2,NA,16,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11160.282155,11957.436882,2,97,2,2,0.27,3,3,2,0,0,2,19,1,3,NA,NA
+71508,7,2,1,1,20,4,4,2,NA,21,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10232.679671,11283.202594,2,101,2,2,0.48,2,2,1,0,0,2,19,1,3,NA,NA
+71509,7,2,2,60,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,3,2,NA,1,2,1,1,2,1,1,2,1,3,12627.660237,14547.347393,2,101,99,99,NA,2,2,0,0,1,2,60,2,3,2,NA
+71510,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,1,2,1,2,2,1,2,2,NA,46372.402053,52744.849017,1,90,5,5,1.05,3,3,0,0,3,2,60,1,5,77,NA
+71511,7,2,1,64,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,1,NA,1,2,2,NA,NA,NA,1,2,2,1,11842.156092,12450.236093,2,96,NA,NA,NA,3,3,0,0,2,1,64,2,4,1,NA
+71512,7,2,1,18,NA,2,2,1,18,222,2,NA,2,2,2,13,NA,NA,NA,2,2,2,2,2,2,2,2,2,2,18206.126374,18280.234909,2,93,4,4,0.82,4,4,0,0,0,1,51,2,3,1,3
+71513,7,2,1,18,NA,2,2,1,18,217,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,2,2,2,1,2,2,1,17555.907575,17627.369376,2,93,6,6,1.39,4,4,0,0,0,1,53,2,3,1,3
+71514,7,2,2,1,15,2,2,2,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9955.153132,10089.038145,2,94,77,77,NA,4,4,2,0,0,2,32,2,4,1,4
+71515,7,2,1,55,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27735.830485,32770.478954,2,91,14,14,4.19,3,3,0,1,0,1,55,2,3,1,5
+71516,7,2,1,23,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,79607.896323,82775.96099,2,92,6,6,2.75,1,1,0,0,0,1,23,1,5,5,NA
+71517,7,2,2,56,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,15521.115746,15563.161752,1,96,2,2,0.4,3,3,0,0,0,2,56,1,3,3,NA
+71518,7,2,2,6,NA,5,6,1,6,75,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5139.061504,5511.809249,2,103,77,77,NA,5,5,0,2,0,2,39,2,5,1,5
+71519,7,2,1,12,NA,4,4,2,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11085.075029,10992.519505,1,93,7,7,1.97,4,4,1,2,0,2,33,1,4,3,NA
+71520,7,2,2,2,NA,2,2,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10099.930724,10736.663476,2,96,1,1,0.06,5,5,2,1,0,1,27,2,3,1,4
+71521,7,2,1,18,NA,4,4,2,18,222,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,13416.172328,13513.882801,1,96,15,15,5,4,4,0,1,0,1,42,2,4,1,4
+71522,7,2,2,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,41983.304745,42669.243011,1,95,5,5,1.1,3,3,0,0,1,2,63,1,4,1,5
+71523,7,2,1,2,NA,4,4,2,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6441.733603,6707.906886,1,91,7,7,1.49,5,5,3,0,0,2,38,2,4,1,4
+71524,7,2,2,16,NA,5,6,1,16,199,NA,NA,2,2,2,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8391.252153,8753.945484,1,92,12,12,NA,7,7,1,2,1,2,45,2,3,1,3
+71525,7,2,2,70,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,72295.614975,73067.691445,1,98,6,6,1.98,2,2,0,0,2,2,70,1,4,1,5
+71526,7,2,2,39,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,17978.142628,18083.929198,1,91,15,15,5,5,5,0,3,0,1,40,1,5,1,5
+71527,7,2,2,54,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,32032.578233,33220.140308,1,91,4,4,0.86,3,3,1,0,0,2,54,1,4,3,NA
+71528,7,2,2,30,NA,4,4,2,NA,NA,2,NA,2,1,4,NA,5,5,2,1,2,2,1,2,2,NA,NA,NA,NA,26462.874679,26479.317455,2,99,15,15,4.34,4,4,0,0,0,1,59,2,4,1,5
+71529,7,2,1,33,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,26972.787983,26588.312559,2,96,10,10,4.76,2,2,0,0,0,2,26,1,5,1,4
+71530,7,2,1,68,NA,1,1,1,NA,NA,1,1,2,1,9,NA,4,1,NA,2,2,2,1,2,2,1,2,2,2,11568.876339,11794.347884,1,102,77,77,NA,2,2,0,0,2,1,68,2,4,1,1
+71531,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,109181.566304,112308.582503,2,91,15,2,0.72,7,1,0,0,1,1,49,NA,NA,5,NA
+71532,7,2,1,21,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,42077.383821,44759.048785,1,102,6,6,1.34,4,4,0,1,0,2,48,2,3,1,1
+71533,7,2,2,38,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,2,1,2,2,1,2,2,NA,NA,NA,NA,31646.000316,42635.226522,1,93,1,1,0,2,2,0,1,0,2,38,1,4,3,NA
+71534,7,2,1,76,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,15844.527135,16703.540642,1,95,6,6,2.04,2,2,0,0,2,2,67,1,1,2,NA
+71535,7,2,2,46,NA,2,2,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,32606.880052,33604.492246,2,93,14,14,3.25,4,4,0,2,0,2,46,2,5,1,4
+71536,7,2,2,44,NA,3,3,2,NA,NA,2,NA,2,1,5,NA,5,3,1,1,2,2,1,2,2,NA,NA,NA,NA,26763.110196,27429.212861,1,100,3,3,0.92,1,1,0,0,0,2,44,2,5,3,NA
+71537,7,2,1,11,NA,4,4,1,11,135,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13423.881856,14179.490667,2,101,4,4,1.02,2,2,0,1,0,2,30,1,2,77,NA
+71538,7,2,2,54,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,1,21143.964074,20565.30403,1,100,5,5,0.85,5,5,0,2,0,2,54,1,2,2,NA
+71539,7,2,2,16,NA,5,6,1,16,195,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,4445.888073,4534.01329,1,103,77,77,NA,6,6,0,2,2,1,70,NA,NA,1,1
+71540,7,2,2,57,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,17364.980275,17214.230108,2,93,8,8,1.67,5,5,1,1,0,2,31,1,4,5,NA
+71541,7,2,2,8,NA,1,1,2,8,104,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,1,2,2,NA,15225.935813,15373.8033,2,94,5,5,1.08,3,3,0,1,0,2,37,2,2,4,NA
+71542,7,2,2,67,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,99831.393624,101809.075589,2,94,15,15,5,2,2,0,0,2,2,67,1,5,1,NA
+71543,7,2,2,10,NA,1,1,1,11,132,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,1,12789.411811,13899.276526,1,102,4,4,0.67,4,4,0,1,0,1,23,2,4,5,NA
+71544,7,2,1,49,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,140129.484883,1,98,10,10,3.04,4,4,0,2,0,2,47,1,4,1,3
+71545,7,2,1,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,39915.513053,42616.161803,2,98,9,9,4.01,2,2,0,0,0,1,27,1,5,1,4
+71546,7,2,1,9,NA,4,4,2,9,117,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6377.235034,7094.374704,2,90,6,6,0.84,6,6,1,3,1,2,43,1,2,5,NA
+71547,7,2,2,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,NA,NA,NA,NA,61710.107686,63275.181919,1,98,5,3,0.9,3,1,0,0,0,2,21,1,4,5,NA
+71548,7,2,2,13,NA,1,1,1,14,168,NA,NA,1,1,NA,7,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,26325.414456,26852.811114,3,92,3,3,0.51,5,5,1,2,0,2,34,2,1,6,NA
+71549,7,2,1,80,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,7912.12564,8378.94011,2,99,5,5,1.88,1,1,0,0,1,1,80,1,1,2,NA
+71550,7,2,2,0,3,3,3,2,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10692.488346,10402.056617,1,98,4,4,1.01,3,3,1,0,0,1,23,1,2,6,NA
+71551,7,2,1,3,NA,3,3,2,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,39061.897809,44649.899401,1,95,15,15,3.62,7,7,2,4,0,1,59,1,5,1,2
+71552,7,2,1,65,NA,2,2,2,NA,NA,2,NA,2,1,9,NA,2,6,NA,2,2,2,2,2,2,1,2,2,2,8609.250304,9380.869295,2,90,4,4,1.02,2,2,0,0,2,2,75,2,1,6,NA
+71553,7,2,1,2,NA,4,4,2,3,36,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4455.468915,4730.507566,3,90,3,3,0.37,5,5,2,2,0,2,36,2,4,4,NA
+71554,7,2,2,33,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,25558.459243,25637.043034,1,101,3,3,0.44,5,5,0,3,0,1,35,1,3,1,4
+71555,7,2,1,60,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,14488.953694,14771.336069,3,92,6,6,1.65,2,2,0,0,1,1,60,1,3,1,4
+71556,7,2,1,0,2,1,1,1,NA,3,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6876.075003,7261.67751,1,100,4,4,0.56,5,5,3,0,0,2,28,2,2,6,NA
+71557,7,2,1,6,NA,5,7,1,6,76,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13614.165673,13762.118902,1,95,14,14,3.04,6,6,0,4,0,1,56,1,5,1,4
+71558,7,2,2,12,NA,4,4,2,12,145,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12581.940435,12644.353117,2,97,6,6,0.92,7,7,1,4,0,2,29,1,3,5,NA
+71559,7,2,2,51,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,2,5,NA,2,2,2,2,2,2,1,2,2,2,17054.056149,20556.446647,2,90,2,2,0.3,3,3,0,1,0,2,51,2,2,5,NA
+71560,7,2,1,80,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,14979.892428,17964.055881,1,92,5,5,1.45,2,2,0,0,2,1,80,1,2,1,3
+71561,7,2,2,40,NA,5,6,1,NA,NA,2,NA,2,2,4,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,12544.244874,13057.037707,1,102,15,15,5,4,4,0,2,0,2,40,2,5,1,4
+71562,7,2,2,0,8,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4543.931297,5004.098497,1,96,7,7,2.23,3,3,1,0,0,1,29,2,5,1,5
+71563,7,2,1,4,NA,2,2,1,4,58,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,15745.774489,15931.923312,2,91,6,6,0.93,5,5,1,2,0,1,34,1,2,1,3
+71564,7,2,1,25,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,27374.067425,29791.861332,1,98,2,2,0.36,5,5,3,0,0,1,25,1,3,1,3
+71565,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,142639.05444,145244.474771,1,100,8,8,2.97,2,2,0,0,0,1,24,1,5,1,5
+71566,7,2,1,48,NA,5,6,2,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19184.316833,20543.822351,1,97,15,15,5,3,3,0,1,0,2,45,2,5,1,5
+71567,7,2,1,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,6,NA,1,2,2,1,2,2,1,2,2,1,114838.671743,123195.718665,1,91,15,3,0.92,2,1,0,0,0,1,30,1,5,6,NA
+71568,7,2,1,13,NA,1,1,1,13,159,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,NA,NA,NA,1,2,2,1,22768.423624,22766.421142,3,92,7,7,1.49,5,5,0,2,1,2,62,1,4,2,NA
+71569,7,2,1,56,NA,1,1,2,NA,NA,2,NA,2,2,77,NA,1,1,NA,2,2,2,1,2,2,2,2,2,2,22446.308035,22116.943066,2,94,6,4,1.38,2,1,0,0,0,1,40,2,3,1,NA
+71570,7,2,1,3,NA,2,2,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,17458.543556,17482.134163,2,91,5,5,1.05,3,3,2,0,0,2,26,2,3,4,NA
+71571,7,2,1,60,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,10717.375231,11136.552864,2,101,14,14,5,2,2,0,0,1,1,60,1,3,1,3
+71572,7,2,1,8,NA,3,3,1,8,104,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,54897.892683,57357.850008,2,98,14,14,3.58,4,4,0,2,0,1,36,1,3,1,4
+71573,7,2,2,11,NA,3,3,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,39810.933651,39282.261038,2,95,15,15,4.34,4,4,0,2,0,2,37,1,4,1,4
+71574,7,2,2,31,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,3,1,2,2,2,2,2,2,2,2,2,2,2,35464.8385,36323.14329,2,96,6,6,1.25,4,4,1,1,0,1,31,2,3,1,3
+71575,7,2,2,66,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,16420.864787,17817.127844,2,102,3,3,0.76,3,3,0,1,1,2,66,1,2,3,NA
+71576,7,2,2,0,9,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8359.077295,8583.668456,3,92,5,5,0.68,6,6,3,0,0,2,19,1,4,NA,NA
+71577,7,2,1,47,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17787.524589,17894.667705,2,100,4,4,1.02,2,2,0,0,1,1,79,1,1,2,NA
+71578,7,2,1,41,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,126789.52929,152250.956405,1,101,8,8,1.85,5,5,0,3,0,1,41,1,3,1,4
+71579,7,2,1,0,9,1,1,1,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,6910.953528,6911.08091,2,96,3,3,0.59,3,3,1,0,0,2,28,2,1,77,NA
+71580,7,2,2,50,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,19969.163208,20694.024637,2,93,8,8,2.57,3,3,0,0,1,1,59,2,3,1,3
+71581,7,2,1,36,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,51543.062078,57043.574316,3,92,10,10,2.82,4,4,0,1,1,1,36,1,3,1,5
+71582,7,2,1,29,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,3,6,NA,2,2,2,2,2,2,2,2,1,2,39084.166385,40378.039385,1,97,4,4,0.72,5,5,2,1,0,2,33,2,1,6,NA
+71583,7,2,1,53,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19374.410926,19463.695548,1,96,12,12,NA,3,3,0,0,2,1,77,NA,NA,6,NA
+71584,7,2,2,58,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21171.165283,21533.12194,1,100,15,15,5,3,3,0,0,1,1,69,1,5,1,4
+71585,7,2,2,39,NA,2,2,1,NA,NA,2,NA,2,2,4,NA,4,1,2,1,2,2,1,2,2,2,2,2,2,36904.965687,36815.84758,2,93,8,8,2,4,4,1,1,0,1,50,2,4,1,4
+71586,7,2,2,7,NA,5,7,1,7,87,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7252.281896,7644.510498,2,92,6,6,1.7,2,2,0,1,0,2,32,2,4,3,NA
+71587,7,2,2,9,NA,1,1,1,9,116,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17053.854294,17379.519997,3,92,8,8,2.17,4,4,0,2,0,1,40,2,2,1,4
+71588,7,2,2,12,NA,3,3,2,12,152,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,86497.469217,90071.968674,1,91,14,14,2.44,7,7,2,4,0,1,33,1,5,1,5
+71589,7,2,1,4,NA,3,3,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,30883.231636,34843.624635,1,95,5,5,0.92,5,5,1,2,0,2,30,1,4,1,4
+71590,7,2,1,0,3,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5777.934266,5882.770795,2,103,9,9,2.6,4,4,2,0,0,2,35,1,4,1,3
+71591,7,2,2,55,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,38954.135779,39992.589933,1,91,4,3,0.82,3,2,0,0,0,2,22,1,5,6,NA
+71592,7,2,2,8,NA,1,1,2,8,103,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,1,2,2,2,2,2,2,14300.71869,14671.891945,2,94,9,9,2.29,5,5,2,1,0,2,33,2,3,1,1
+71593,7,2,1,45,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,26215.80546,26132.189255,1,92,9,7,1.74,7,4,2,1,0,1,45,1,4,2,NA
+71594,7,2,1,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,91704.59836,93129.802737,1,93,15,14,5,6,1,0,0,0,1,23,1,5,5,NA
+71595,7,2,2,56,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,207590.346266,205849.702763,1,92,15,15,5,2,2,0,0,0,2,56,1,4,1,5
+71596,7,2,2,19,NA,1,1,1,19,237,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,NA,NA,NA,NA,26325.414456,27702.295836,3,92,5,5,0.68,6,6,3,0,0,2,19,1,4,NA,NA
+71597,7,2,2,13,NA,3,3,2,13,158,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,29885.567338,31265.78413,1,94,7,7,0.94,7,7,1,4,0,2,46,2,5,1,5
+71598,7,2,2,27,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,1,1,2,2,1,2,2,1,2,2,1,77792.702831,79765.659277,1,102,14,14,4.32,3,3,1,0,0,1,25,1,4,1,4
+71599,7,2,1,59,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,105379.391148,106074.259408,1,93,15,15,3.92,5,5,0,1,0,2,54,1,5,1,5
+71600,7,2,1,1,14,1,1,1,NA,14,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,12493.910388,13386.323284,2,98,1,1,0.13,4,4,2,0,0,2,52,1,2,4,NA
+71601,7,2,1,6,NA,5,7,2,6,81,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9502.15317,9616.069144,2,97,5,5,0.84,5,5,0,2,0,2,33,1,4,1,3
+71602,7,2,2,9,NA,3,3,1,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17685.703081,17770.107744,1,98,6,6,0.81,6,6,0,4,0,2,34,NA,NA,1,2
+71603,7,2,2,16,NA,4,4,2,16,201,NA,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10484.6104,10912.740054,2,99,13,13,NA,3,3,1,1,0,2,36,NA,NA,5,NA
+71604,7,2,1,49,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,31640.296506,34765.415986,2,94,4,4,0.81,3,3,0,1,0,1,49,2,3,1,3
+71605,7,2,2,38,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,64581.191728,64779.757488,2,94,9,5,1.84,2,1,0,0,0,2,38,1,4,6,NA
+71606,7,2,1,13,NA,5,6,1,13,163,NA,NA,2,1,3,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9048.093498,9416.694257,1,100,8,8,2.62,3,3,0,1,0,1,41,2,5,1,5
+71607,7,2,2,8,NA,1,1,1,8,104,NA,NA,2,2,2,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,NA,16986.005478,17733.622754,2,102,4,4,0.57,5,5,0,3,0,1,41,2,1,1,2
+71608,7,2,1,10,NA,4,4,1,10,125,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10138.00454,10211.52145,1,100,8,8,1.61,6,6,1,3,0,1,29,1,5,6,NA
+71609,7,2,2,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,21640.010524,20970.872814,1,93,6,6,0.83,6,6,3,1,0,1,37,NA,NA,1,3
+71610,7,2,1,16,NA,1,1,1,16,200,NA,NA,2,2,2,8,NA,NA,NA,2,2,2,2,2,2,2,2,1,2,22203.024273,23172.92708,1,100,99,99,NA,7,7,2,3,0,2,35,2,1,1,NA
+71611,7,2,2,15,NA,3,3,2,15,191,NA,NA,1,1,NA,9,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,103007.696238,104941.393061,1,101,14,14,3.25,4,4,0,1,0,1,48,1,4,1,2
+71612,7,2,2,4,NA,2,2,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11429.37307,11583.084593,2,90,2,2,0.54,2,2,1,0,0,2,32,2,2,4,NA
+71613,7,2,1,75,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,12824.368043,13581.00423,2,101,9,9,4.08,2,2,0,0,1,1,75,1,1,2,NA
+71614,7,2,1,4,NA,4,4,1,4,50,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10040.033098,11070.778245,1,100,14,14,3.47,4,4,2,0,0,1,34,1,5,1,5
+71615,7,2,1,41,NA,5,6,1,NA,NA,2,NA,2,1,7,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15952.616949,16443.31736,2,93,15,15,5,3,3,1,0,0,1,41,2,5,1,5
+71616,7,2,2,3,NA,2,2,2,3,46,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,13490.92482,14894.343037,1,90,2,2,0.56,2,2,1,0,0,2,27,1,3,5,NA
+71617,7,2,1,2,NA,2,2,2,2,27,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9237.934626,9250.417252,2,90,3,3,0.7,3,3,1,1,0,2,25,1,1,1,NA
+71618,7,2,1,4,NA,4,4,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7604.054172,7918.254694,2,99,6,6,1.03,6,6,3,0,0,1,33,1,3,6,NA
+71619,7,2,1,55,NA,5,6,2,NA,NA,2,NA,2,1,2,NA,4,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11690.444016,12095.591726,3,90,3,3,0.75,3,3,0,0,0,1,55,2,4,1,3
+71620,7,2,1,65,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,1,NA,1,2,1,1,2,1,1,2,1,3,7289.557268,7804.776102,3,90,77,77,NA,2,2,0,0,1,1,65,2,3,1,5
+71621,7,2,2,12,NA,1,1,1,12,152,NA,NA,2,2,4,6,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,26325.414456,27702.295836,3,92,4,4,0.46,7,7,1,2,0,2,31,2,2,1,1
+71622,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,40157.007559,44610.827766,2,98,5,5,1.63,2,2,0,0,2,1,80,1,3,1,3
+71623,7,2,1,12,NA,4,4,2,12,144,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8364.097643,9212.617312,2,90,6,6,1.03,6,6,3,1,0,1,45,2,2,1,2
+71624,7,2,2,18,NA,5,6,1,18,219,2,NA,2,2,2,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,10767.566937,11183.935292,2,91,15,15,4.63,7,7,1,2,0,1,36,2,4,1,3
+71625,7,2,1,39,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,18353.275855,20139.588201,1,91,4,4,0.76,4,4,0,2,0,2,44,1,4,6,NA
+71626,7,2,1,59,NA,2,2,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,1,2,2,1,2,2,2,2,2,2,30839.213846,31462.493685,1,92,10,10,2.93,4,4,1,0,0,2,55,1,4,1,4
+71627,7,2,1,19,NA,5,6,2,19,236,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,9099.599144,9725.105491,1,90,9,9,2.6,4,4,0,1,0,2,49,2,2,1,5
+71628,7,2,2,53,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,3,4,NA,1,2,2,1,2,2,1,2,2,1,28349.668436,29178.127765,1,102,5,5,1.56,2,2,0,0,0,2,53,1,3,4,NA
+71629,7,2,2,16,NA,4,4,1,16,193,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,20473.346601,20523.769333,2,102,14,14,4.05,3,3,0,1,0,1,18,1,2,NA,NA
+71630,7,2,1,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6121.794646,6421.491158,1,102,4,4,0.61,5,5,2,2,0,2,27,2,2,5,NA
+71631,7,2,1,22,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,37970.860743,38488.432875,2,92,NA,3,0.92,5,1,0,0,0,1,22,1,5,5,NA
+71632,7,2,1,18,NA,4,4,1,18,223,2,NA,1,1,NA,12,NA,NA,NA,1,2,2,NA,NA,NA,1,2,2,1,13276.485807,13885.0639,2,100,NA,NA,NA,4,4,1,0,0,2,38,NA,NA,77,NA
+71633,7,2,2,64,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,1,2,1,1,2,1,1,2,1,NA,11838.431472,12356.62484,2,92,3,3,0.4,6,6,0,1,2,1,78,2,1,1,1
+71634,7,2,2,21,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,36850.868456,37875.214466,2,103,5,5,1.2,3,3,0,0,2,1,66,2,2,1,2
+71635,7,2,2,42,NA,4,4,2,NA,NA,2,NA,2,1,8,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,22513.236051,21981.065205,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+71636,7,2,1,9,NA,4,4,1,9,111,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10410.106675,10390.534388,2,96,4,4,0.57,5,5,0,3,0,2,26,1,2,5,NA
+71637,7,2,1,14,NA,3,3,1,14,173,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,95925.820151,96188.020601,1,92,8,8,3.17,2,2,0,1,0,2,35,1,2,77,NA
+71638,7,2,1,23,NA,4,4,1,NA,NA,1,2,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,25815.880139,26503.609729,2,101,99,13,NA,2,1,0,0,0,1,24,1,4,5,NA
+71639,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,22109.546782,22635.660435,1,93,12,12,NA,3,3,0,1,0,2,48,1,3,5,NA
+71640,7,2,2,51,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,15497.844354,15720.409191,1,99,14,14,5,2,1,0,0,0,2,51,1,5,1,NA
+71641,7,2,2,7,NA,3,3,2,7,95,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,73810.484644,74865.697805,1,91,15,15,5,4,4,1,1,0,1,38,1,5,1,5
+71642,7,1,2,3,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,15358.480588,0,1,102,2,2,0.31,4,4,1,2,0,2,25,1,2,4,NA
+71643,7,2,1,31,NA,2,2,1,NA,NA,1,2,2,1,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,53303.690379,53390.332272,1,100,8,8,3.17,2,2,0,0,0,1,31,2,4,1,5
+71644,7,2,2,80,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,45973.010529,51521.910389,1,91,7,7,2.68,2,2,0,0,2,1,80,1,4,1,4
+71645,7,2,1,7,NA,3,3,2,7,93,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,44777.275016,47565.734765,1,91,14,14,3.06,5,5,0,3,0,2,46,1,5,1,5
+71646,7,2,1,3,NA,1,1,1,3,44,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,16775.083123,16335.348923,2,98,3,3,0.33,7,7,2,3,0,1,40,2,1,1,1
+71647,7,2,2,80,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,16076.770364,16942.062244,1,90,14,14,5,2,2,0,0,1,1,58,1,5,5,NA
+71648,7,2,1,7,NA,1,1,1,7,91,NA,NA,1,1,NA,1,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,13533.281742,14131.961026,1,100,7,7,1.74,4,4,0,2,0,2,39,2,1,1,3
+71649,7,2,1,12,NA,5,6,2,12,154,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7057.977053,7861.480262,2,92,12,12,NA,7,7,2,4,0,1,54,2,2,1,5
+71650,7,2,1,60,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,7140.457792,7196.312698,2,101,7,7,1.3,5,5,2,0,1,2,50,1,4,1,3
+71651,7,2,2,46,NA,5,6,2,NA,NA,2,NA,2,2,6,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,19294.309,26319.879121,1,93,14,14,5,2,2,0,1,0,2,46,2,5,3,NA
+71652,7,2,1,5,NA,5,6,2,5,68,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,4858.407579,4952.635908,1,99,14,14,2.66,7,7,3,1,0,1,35,1,5,1,5
+71653,7,2,2,10,NA,5,6,1,10,124,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6779.975678,7239.998882,3,91,15,15,5,3,3,0,1,0,1,47,2,5,1,5
+71654,7,2,2,17,NA,3,3,1,17,213,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,49337.766842,51601.905979,2,103,15,15,3.44,7,7,0,1,2,2,79,1,3,2,NA
+71655,7,2,1,65,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,7323.703412,7380.991723,2,97,NA,99,NA,7,1,2,1,1,2,56,1,3,5,NA
+71656,7,2,1,58,NA,3,3,2,NA,NA,2,NA,2,1,9,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,163791.427718,171966.737289,1,93,15,15,5,2,2,0,0,0,1,58,2,5,1,5
+71657,7,2,2,49,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,2,2,2,1,2,2,1,2,2,1,30105.942022,31646.425257,1,96,8,8,2.17,4,4,0,0,2,1,80,NA,NA,1,NA
+71658,7,2,1,25,NA,1,1,2,NA,NA,2,NA,2,2,3,NA,2,6,NA,2,2,2,2,2,2,1,2,1,2,37987.561992,39245.132137,1,90,2,1,0.13,5,3,0,1,0,1,25,2,2,6,NA
+71659,7,2,2,33,NA,4,4,2,NA,NA,2,NA,2,2,3,NA,3,1,2,1,2,2,1,2,2,NA,NA,NA,NA,32799.237043,32564.866412,1,91,6,6,0.99,5,5,3,0,0,2,33,2,3,1,4
+71660,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,4,3,1,2,2,1,2,2,1,2,2,1,19498.713386,19652.836095,2,97,4,4,1.09,2,2,0,1,0,2,35,1,2,4,NA
+71661,7,2,1,8,NA,5,6,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,5651.170392,5934.27615,1,102,5,5,0.92,5,5,1,2,0,2,44,2,1,1,2
+71662,7,2,1,74,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,79733.665046,84684.628619,1,92,9,9,3.8,2,2,0,0,2,2,66,1,4,1,5
+71663,7,2,2,9,NA,4,4,1,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,1,1,2,1,1,2,2,1,11195.065587,11548.620784,2,96,77,77,NA,7,3,1,4,0,2,32,NA,NA,1,NA
+71664,7,2,1,4,NA,4,4,2,4,55,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6299.35077,6688.213305,3,90,3,3,0.37,5,5,2,2,0,2,36,2,4,4,NA
+71665,7,2,2,7,NA,4,4,1,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11334.095519,11753.9565,2,96,4,4,0.65,5,5,0,3,0,1,30,1,4,1,2
+71666,7,2,1,61,NA,4,4,2,NA,NA,1,1,1,1,NA,NA,5,6,NA,1,2,2,1,2,2,1,2,2,1,7017.945976,7072.842557,2,99,1,1,0.36,3,1,1,0,1,2,55,1,4,6,NA
+71667,7,2,2,9,NA,1,1,2,9,118,NA,NA,1,1,NA,4,NA,NA,NA,2,1,2,1,2,2,1,2,2,2,13231.432201,13706.598104,2,97,4,4,0.67,4,4,0,2,0,1,39,2,2,6,NA
+71668,7,2,2,12,NA,4,4,1,12,154,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12531.903464,13043.632492,2,100,9,9,2.46,4,4,1,1,1,2,59,1,3,1,3
+71669,7,2,1,0,7,5,6,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6918.643947,7208.455005,1,101,8,8,1.81,5,5,2,0,1,2,37,2,4,1,2
+71670,7,2,1,24,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14860.312419,14555.816778,3,90,15,15,4.34,4,4,0,0,1,1,65,1,3,1,4
+71671,7,2,1,65,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,125558.167126,127168.668278,3,91,7,7,2.89,2,2,0,0,2,2,64,1,4,1,4
+71672,7,2,2,18,NA,4,4,2,18,227,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11372.489138,11836.874523,1,99,14,14,3.67,4,4,1,0,0,2,49,1,3,1,3
+71673,7,2,2,9,NA,5,7,2,9,112,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7294.840846,7578.01532,1,93,15,15,5,5,5,1,2,0,2,40,1,5,1,5
+71674,7,2,1,10,NA,4,4,2,10,127,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,9502.15317,9484.287948,2,97,3,3,0.4,6,6,2,3,0,2,25,1,2,5,NA
+71675,7,2,2,73,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,66567.821082,76362.988288,3,91,7,7,2.92,2,2,0,0,2,1,74,1,3,1,3
+71676,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,37318.801462,40047.525745,1,98,8,8,3.48,2,2,0,0,2,1,80,1,3,1,2
+71677,7,2,1,35,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,29756.291619,30875.098449,2,101,5,5,1.19,3,3,1,0,0,2,32,1,4,1,3
+71678,7,2,2,70,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,98976.420245,102012.919253,2,101,15,15,5,2,2,0,0,2,1,73,1,5,1,5
+71679,7,2,2,27,NA,1,1,1,NA,NA,2,NA,2,2,5,NA,2,5,2,1,2,2,1,2,2,NA,NA,NA,NA,41537.508946,44104.799715,1,102,4,4,0.61,5,5,2,2,0,2,27,2,2,5,NA
+71680,7,2,1,55,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,28575.926692,28541.082411,3,92,4,3,0.52,5,4,0,0,0,2,57,1,4,1,2
+71681,7,2,2,35,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,1,1,2,2,1,2,2,NA,NA,NA,NA,68146.615091,70082.037538,2,99,15,15,5,3,3,1,0,0,1,43,1,5,1,5
+71682,7,2,1,69,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,7101.739553,7379.502561,2,100,5,5,1.56,2,2,0,0,2,1,69,1,2,1,3
+71683,7,2,2,1,17,4,4,1,NA,18,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8412.350508,9400.756791,1,100,5,5,1.05,3,3,1,0,0,2,42,2,5,1,NA
+71684,7,2,1,0,3,3,3,1,NA,4,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24509.975935,26238.985016,1,94,5,5,1.04,4,4,1,1,0,1,18,1,2,NA,NA
+71685,7,2,2,11,NA,4,4,2,11,140,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10994.192555,11605.838362,2,97,2,2,0.27,4,4,0,2,0,1,51,1,2,4,NA
+71686,7,2,1,31,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16505.268729,16996.380056,3,91,8,8,2.7,3,3,1,0,0,1,31,1,5,1,5
+71687,7,2,2,13,NA,5,6,2,13,157,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,6486.356303,6952.179218,1,91,7,7,1.57,4,4,0,3,0,2,38,2,2,3,NA
+71688,7,2,1,51,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,16628.326744,17325.680634,2,102,10,10,3.62,3,3,0,0,0,1,51,2,5,1,5
+71689,7,2,2,17,NA,4,4,1,17,210,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16659.324602,16974.753267,1,92,15,15,4.44,5,5,0,3,0,2,43,1,5,6,NA
+71690,7,2,1,2,NA,2,2,1,2,31,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,9382.177239,9806.283152,2,93,6,6,0.64,7,7,2,1,3,2,60,2,3,2,NA
+71691,7,2,2,76,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,62212.598767,64340.261278,1,91,10,10,4.3,2,2,0,0,2,1,78,1,4,1,4
+71692,7,2,1,8,NA,3,3,1,8,104,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22651.436723,23515.701302,3,92,7,7,0.81,7,7,2,4,0,1,40,NA,NA,1,4
+71693,7,2,2,4,NA,5,6,2,4,53,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7497.268293,8158.717681,1,101,10,10,3.67,3,3,1,0,0,1,36,2,5,1,5
+71694,7,2,1,0,6,5,6,1,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5024.464768,5303.683185,2,92,10,6,1.12,7,4,1,1,1,2,27,2,3,1,3
+71695,7,2,2,54,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,1,1,NA,1,2,2,1,2,1,1,2,2,1,14680.520497,14758.150245,3,91,6,6,1.34,4,4,0,2,0,1,52,2,3,1,1
+71696,7,1,2,58,NA,5,6,NA,NA,NA,2,NA,2,2,4,NA,1,2,NA,1,2,1,1,2,1,NA,NA,NA,NA,12649.084278,0,3,90,12,12,NA,4,4,0,0,0,2,58,2,1,2,NA
+71697,7,2,1,74,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,NA,15176.622228,16685.078243,1,101,4,4,1.16,2,2,0,0,2,1,74,1,1,1,2
+71698,7,2,2,46,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,127315.335607,133282.596235,1,91,14,14,4.03,4,4,0,2,0,1,52,1,4,1,5
+71699,7,2,2,15,NA,2,2,1,15,184,NA,NA,2,1,4,9,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,15809.066118,17342.845429,2,93,3,3,0.52,5,5,0,2,0,1,41,2,4,1,4
+71700,7,2,2,1,16,5,6,1,NA,17,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5610.22155,6105.185511,2,92,15,15,5,3,3,1,0,0,1,39,2,5,1,5
+71701,7,2,2,17,NA,3,3,1,17,210,2,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,113907.203714,124196.980527,1,94,8,8,1.67,5,5,1,2,0,1,52,1,4,1,4
+71702,7,2,2,70,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,2,NA,1,2,2,1,2,2,1,2,2,NA,12669.609493,13517.506264,2,98,99,99,NA,1,1,0,0,1,2,70,1,3,2,NA
+71703,7,2,1,53,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,NA,NA,NA,NA,26240.791307,26334.983911,1,90,4,4,1.34,1,1,0,0,0,1,53,1,4,6,NA
+71704,7,2,1,77,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,71070.181743,73766.510327,1,95,4,4,1.12,2,2,0,0,2,2,78,1,4,1,2
+71705,7,2,1,67,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,14488.953694,14771.336069,3,92,3,3,0.95,2,2,0,0,2,2,63,1,4,1,1
+71706,7,2,2,30,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,4,2,1,2,2,1,2,2,1,2,2,1,25423.189953,34604.094756,2,98,2,2,0.34,2,2,0,1,0,2,30,1,4,4,NA
+71707,7,2,2,7,NA,4,4,1,7,91,NA,NA,1,1,NA,2,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11195.065587,11638.702248,2,96,13,13,NA,4,4,1,1,0,2,40,1,3,77,NA
+71708,7,2,1,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,6612.194774,6663.917441,2,99,5,5,1.88,1,1,0,0,1,1,64,1,5,5,NA
+71709,7,2,2,45,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,40760.712736,41570.695949,1,98,3,3,0.73,2,2,0,0,0,1,49,1,2,1,2
+71710,7,2,1,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,8491.292032,8557.713588,1,96,15,15,5,2,2,0,0,2,1,68,1,3,1,4
+71711,7,2,1,3,NA,4,4,1,3,43,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10273.522239,11328.238204,1,100,14,6,1.85,3,2,1,0,0,1,33,1,5,5,NA
+71712,7,2,2,53,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,19927.035776,20217.558558,2,96,8,8,3.67,2,2,0,0,0,1,56,2,5,1,5
+71713,7,2,1,17,NA,5,6,2,17,213,2,NA,2,1,4,12,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,8965.57404,9581.867503,1,90,15,15,5,5,5,0,3,0,2,46,2,4,1,5
+71714,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,29470.441678,31861.649147,1,103,6,6,1.98,2,2,0,0,2,1,80,1,5,1,4
+71715,7,2,2,47,NA,4,4,2,NA,NA,2,NA,2,2,6,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,21969.841763,21566.078744,1,96,14,14,5,2,2,0,0,0,2,47,2,4,5,NA
+71716,7,2,2,21,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,14516.765165,13843.248577,3,90,15,15,4.89,5,5,0,0,0,2,57,2,3,1,3
+71717,7,2,2,27,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,5,2,1,2,2,1,2,2,1,2,2,1,23255.074483,22360.893796,2,98,4,4,1.29,2,2,0,1,0,2,27,1,2,5,NA
+71718,7,2,2,68,NA,4,4,2,NA,NA,2,NA,2,2,4,NA,4,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,9644.668529,10075.292662,1,93,3,3,1.1,1,1,0,0,1,2,68,2,4,1,NA
+71719,7,2,1,59,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,1,5,NA,1,2,2,1,2,2,1,2,2,1,26135.885159,26946.665552,2,101,2,2,0.74,1,1,0,0,0,1,59,1,1,5,NA
+71720,7,1,1,80,NA,3,3,NA,NA,NA,1,2,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,6882.452425,0,1,99,3,3,1.19,1,1,0,0,1,1,80,1,5,2,NA
+71721,7,2,2,11,NA,3,3,2,11,138,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,73810.484644,74865.697805,1,94,10,10,2.67,5,5,0,3,0,1,40,1,5,1,2
+71722,7,2,2,48,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,4,NA,1,2,2,1,2,2,1,2,2,1,19258.69251,18731.627868,2,99,12,3,1.16,6,1,0,0,0,2,57,1,5,2,NA
+71723,7,2,2,5,NA,5,6,2,6,72,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5520.445386,5614.569811,2,100,15,15,4.83,4,4,1,1,0,1,43,2,5,1,5
+71724,7,2,1,54,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,37557.946192,39249.822004,2,102,99,3,1.29,3,1,0,0,1,1,61,NA,NA,5,NA
+71725,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,9177.295801,9548.31812,2,92,15,3,0.92,3,1,0,0,0,2,25,1,5,5,NA
+71726,7,2,2,5,NA,4,4,2,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9268.093277,9507.318489,1,99,3,3,0.82,2,2,1,0,0,2,29,1,4,4,NA
+71727,7,1,2,11,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,5,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,8851.712341,0,1,92,5,5,0.63,7,7,0,4,1,1,60,NA,NA,1,NA
+71728,7,2,2,61,NA,2,2,1,NA,NA,2,NA,2,1,3,NA,5,1,NA,2,2,2,1,2,2,1,2,2,2,11291.029617,11762.265953,2,102,14,14,5,2,2,0,0,2,1,68,1,4,1,5
+71729,7,2,1,33,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,16214.132654,20271.082789,3,90,8,8,3.3,2,2,0,0,1,2,64,2,2,1,NA
+71730,7,2,2,40,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,1,3,2,2,2,2,2,2,2,1,2,2,2,33737.181071,33913.1176,2,91,99,99,NA,6,6,1,3,0,2,20,2,2,5,NA
+71731,7,1,2,24,NA,5,6,NA,NA,NA,2,NA,2,1,4,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,15265.136017,0,1,93,15,5,1.84,3,1,0,0,0,2,24,2,5,5,NA
+71732,7,2,1,63,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,1,6,NA,2,2,2,2,2,2,1,2,2,2,10596.142548,10856.489537,1,94,12,3,1.07,4,1,0,0,1,2,37,NA,NA,6,NA
+71733,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,1,2,NA,1,2,2,1,2,2,1,2,2,NA,24816.484998,26631.048773,1,91,3,3,1.33,1,1,0,0,1,1,80,1,1,2,NA
+71734,7,2,1,5,NA,4,4,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7311.663111,7613.781996,2,93,6,6,1.15,5,5,3,1,0,1,29,1,3,5,NA
+71735,7,2,2,67,NA,2,2,1,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,14204.126838,15368.918135,2,98,4,4,1.34,2,2,0,0,2,1,66,1,1,1,1
+71736,7,2,1,74,NA,3,3,1,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,1,2,1,2,2,1,2,2,NA,62359.83753,66231.994715,1,94,14,14,5,2,2,0,0,2,1,74,1,4,1,4
+71737,7,2,2,36,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,1,5,2,2,2,2,2,2,2,2,2,2,2,45655.090694,49537.803198,3,92,4,4,0.67,4,4,0,3,0,2,36,2,1,5,NA
+71738,7,1,2,80,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,NA,NA,NA,NA,47254.485787,0,1,90,8,8,4.21,1,1,0,0,1,2,80,1,4,2,NA
+71739,7,2,1,80,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,NA,13116.035678,14391.817492,2,91,4,4,1.02,2,2,0,0,2,1,80,1,4,1,2
+71740,7,2,2,20,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,NA,NA,NA,1,2,2,1,20443.961017,19438.354372,2,99,3,3,0.44,5,5,1,1,0,2,53,1,4,1,3
+71741,7,2,1,63,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,5,3,NA,2,2,2,1,2,2,2,2,2,2,6449.12882,6552.435629,2,93,3,3,0.9,1,1,0,0,1,1,63,2,5,3,NA
+71742,7,2,1,0,10,5,6,2,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,5441.212985,5743.59084,1,93,8,8,1.2,7,7,1,1,1,1,24,2,2,5,NA
+71743,7,2,2,0,4,1,1,1,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,2,2,2,NA,NA,NA,NA,8359.077295,8418.658174,3,92,6,6,1.06,5,5,2,0,0,2,54,2,1,77,NA
+71744,7,2,2,40,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,144726.059187,145423.66071,2,98,15,15,4.17,6,6,1,1,0,2,40,1,4,1,4
+71745,7,2,2,21,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,5,2,1,2,2,1,2,2,1,2,2,1,130601.953362,132156.64994,2,91,15,15,5,4,4,0,0,0,1,54,1,5,1,NA
+71746,7,2,2,18,NA,3,3,1,18,219,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,32725.110579,34226.88515,2,92,3,1,0.28,2,1,0,0,0,2,18,1,4,NA,NA
+71747,7,2,1,7,NA,5,7,1,7,88,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11800.806722,11886.381633,1,100,9,9,2.78,4,4,0,2,0,1,39,1,5,1,5
+71748,7,2,2,36,NA,1,1,2,NA,NA,2,NA,2,2,4,NA,3,6,2,2,2,2,2,2,2,2,2,1,2,32910.619435,32637.844845,1,90,2,1,0.17,5,2,0,1,0,1,25,2,2,6,NA
+71749,7,2,2,0,4,1,1,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7011.218293,6863.710472,2,97,6,6,1.84,2,2,1,0,0,2,27,1,4,5,NA
+71750,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,NA,19552.617734,21912.600615,2,92,4,4,1.12,2,2,0,0,1,2,80,1,5,2,NA
+71751,7,2,2,42,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,2,1,2,2,1,2,2,1,2,2,1,110070.015649,117535.388004,3,91,14,14,3.4,4,4,0,2,0,1,40,1,4,1,4
+71752,7,2,1,36,NA,1,1,1,NA,NA,2,NA,2,2,4,NA,2,1,NA,1,2,2,1,2,2,2,2,2,2,45660.210546,48177.324946,1,101,9,9,2.88,3,3,1,0,0,1,36,2,2,1,4
+71753,7,2,2,80,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,NA,33147.414266,37148.276517,2,94,14,14,5,2,2,0,0,1,2,80,1,4,2,NA
+71754,7,2,1,0,0,3,3,1,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20173.318836,21596.406788,2,103,15,15,5,4,4,2,0,0,1,36,2,4,1,5
+71755,7,2,1,22,NA,2,2,1,NA,NA,2,NA,2,2,3,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,38474.772527,42632.210531,2,93,7,3,0.9,3,1,0,0,0,1,25,2,4,5,NA
+71756,7,2,1,63,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,133262.624177,131902.330801,1,98,9,9,3.97,2,2,0,0,2,1,63,1,5,1,4
+71757,7,2,1,23,NA,5,6,1,NA,NA,2,NA,2,2,3,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,13254.861315,15747.113823,1,101,8,8,4.48,1,1,0,0,0,1,23,2,5,5,NA
+71758,7,2,2,24,NA,2,2,2,NA,NA,2,NA,2,1,3,NA,4,6,2,2,2,2,2,2,2,1,2,2,2,30253.427014,30121.660039,2,90,6,6,0.66,7,7,2,2,0,2,24,2,4,6,NA
+71759,7,2,1,16,NA,4,4,1,17,204,NA,NA,1,1,NA,10,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,17606.165994,17558.40257,2,101,8,8,2.7,3,3,0,1,0,1,53,1,4,1,2
+71760,7,2,2,60,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,14994.337564,16150.13007,1,100,8,8,2.36,3,3,1,0,1,2,60,1,3,3,NA
+71761,7,2,1,38,NA,1,1,1,NA,NA,2,NA,2,7,77,NA,2,1,NA,2,2,2,1,2,2,NA,NA,NA,NA,32907.512068,32659.148672,2,92,12,12,NA,7,7,0,1,2,2,64,2,1,2,NA
+71762,7,2,1,12,NA,3,3,1,12,155,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,30943.024697,32205.010072,1,101,4,4,0.58,6,6,0,4,0,2,41,1,3,5,NA
+71763,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,109309.268477,116308.99629,3,91,14,14,3.4,4,4,0,2,0,1,40,1,4,1,4
+71764,7,2,1,33,NA,5,6,1,NA,NA,2,NA,2,1,5,NA,4,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,20181.692021,21173.351412,1,92,12,12,NA,4,4,1,1,0,1,33,2,4,1,4
+71765,7,2,1,39,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,19144.719218,23354.203452,1,101,8,8,1.81,5,5,2,0,1,2,37,2,4,1,2
+71766,7,2,2,38,NA,2,2,1,NA,NA,2,NA,2,1,2,NA,3,1,3,2,2,2,2,2,2,NA,NA,NA,NA,40476.413979,41456.006766,2,93,5,5,0.87,4,4,1,1,0,1,41,2,5,1,3
+71767,7,2,1,38,NA,1,1,2,NA,NA,2,NA,1,1,NA,NA,5,3,NA,1,2,2,1,2,2,1,2,2,1,39133.405322,40210.578118,1,91,12,8,4.59,5,1,0,2,0,1,38,1,5,3,NA
+71768,7,2,2,7,NA,3,3,2,7,88,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,24070.467912,25563.654853,1,95,15,15,3.62,7,7,2,4,0,1,59,1,5,1,2
+71769,7,2,1,24,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,4,5,NA,2,2,2,2,2,2,2,2,2,2,35669.2076,36155.406423,2,94,14,8,3.06,5,2,0,0,0,1,24,2,4,5,NA
+71770,7,2,2,64,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,10687.292889,11164.469085,2,99,3,3,0.75,2,2,0,0,1,2,64,1,4,3,NA
+71771,7,2,1,3,NA,1,1,2,3,47,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18363.136017,19674.774959,1,91,4,4,0.86,3,3,1,0,0,2,54,1,4,3,NA
+71772,7,2,2,52,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,NA,NA,NA,NA,37518.850453,37204.255172,1,98,3,3,1.18,1,1,0,0,0,2,52,1,4,3,NA
+71773,7,2,1,53,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,4,3,NA,1,2,2,1,2,2,1,2,2,1,23775.734331,23918.947415,2,96,12,99,NA,2,1,0,0,1,1,53,1,4,3,NA
+71774,7,1,1,2,NA,1,1,NA,NA,NA,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,10412.950357,0,1,90,15,15,5,5,5,1,1,0,1,32,2,1,1,4
+71775,7,2,1,62,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,28216.191929,29365.059825,1,101,6,6,1.52,3,3,0,1,1,1,62,1,2,1,3
+71776,7,2,2,0,10,1,1,1,NA,11,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,5328.475628,5508.797375,1,103,13,13,NA,6,6,1,2,0,1,42,2,9,1,9
+71777,7,2,2,77,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,9807.668192,10541.248354,2,95,5,5,0.87,4,4,0,0,2,2,77,1,2,1,1
+71778,7,2,1,20,NA,2,2,2,NA,NA,2,NA,2,1,3,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,43798.832177,56306.768819,1,90,3,3,0.79,2,2,0,0,1,1,70,2,2,1,NA
+71779,7,2,1,19,NA,4,4,2,19,239,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11125.932433,11147.929312,1,96,6,6,1.21,4,4,0,2,0,2,41,1,4,4,NA
+71780,7,2,1,10,NA,4,4,2,10,128,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,14125.862146,14370.909235,1,97,15,15,5,4,4,0,2,0,1,48,1,5,1,5
+71781,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,17152.714252,16683.285229,1,102,12,12,NA,7,7,3,2,0,2,52,1,4,5,NA
+71782,7,2,1,5,NA,5,6,1,5,70,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,10498.222836,11004.769407,1,100,15,15,5,4,4,2,0,0,1,39,2,5,1,5
+71783,7,1,1,12,NA,5,6,NA,NA,NA,NA,NA,2,1,3,5,NA,NA,NA,1,1,1,1,2,1,NA,NA,NA,NA,6174.283826,0,2,92,77,77,NA,5,5,0,1,2,2,80,NA,NA,1,1
+71784,7,2,2,24,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,41018.498876,42180.331054,2,98,6,6,1.07,5,5,3,0,0,2,24,1,3,1,3
+71785,7,2,1,64,NA,4,4,1,NA,NA,1,1,1,1,NA,NA,2,3,NA,1,2,2,1,2,2,1,2,2,1,7101.739553,7433.956549,2,100,77,77,NA,1,1,0,0,1,1,64,1,2,3,NA
+71786,7,2,2,42,NA,5,7,1,NA,NA,2,NA,2,1,7,NA,5,4,2,1,2,2,1,2,2,1,2,2,1,18588.235275,19331.159141,2,96,15,15,5,3,3,0,2,0,2,42,2,5,4,NA
+71787,7,2,2,50,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,16859.368198,17492.963343,1,99,13,13,NA,4,4,1,0,0,2,26,1,4,4,NA
+71788,7,2,1,6,NA,1,1,1,6,82,NA,NA,1,1,NA,0,NA,NA,NA,2,1,1,2,2,1,NA,NA,NA,NA,14880.007592,15997.505807,3,92,4,4,0.66,4,4,0,1,0,1,52,2,1,1,1
+71789,7,2,2,50,NA,5,7,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,13149.417265,14101.663559,2,103,15,15,5,3,3,0,0,0,2,50,1,4,1,4
+71790,7,2,2,11,NA,5,7,1,11,139,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,10332.067017,11032.916236,1,100,9,9,2.22,5,5,0,3,0,2,52,1,4,1,4
+71791,7,2,1,18,NA,1,1,1,18,221,2,NA,1,1,NA,12,NA,NA,NA,2,2,2,1,2,2,1,2,2,2,32262.881978,32260.04446,2,102,8,8,2.01,4,4,0,0,0,2,48,2,4,3,NA
+71792,7,2,1,34,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,69548.324117,71540.223881,2,95,6,6,2.69,1,1,0,0,0,1,34,1,5,5,NA
+71793,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,170311.62251,168883.560827,3,91,15,15,5,2,2,0,0,0,2,57,1,5,1,5
+71794,7,2,1,4,NA,4,4,2,4,57,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8973.990262,9527.959827,2,101,7,7,1.3,5,5,2,0,1,2,50,1,4,1,3
+71795,7,2,2,8,NA,5,7,1,9,108,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,17706.623308,17924.842987,1,94,2,2,0.3,5,5,1,2,0,1,23,1,1,6,NA
+71796,7,2,2,45,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,2,NA,1,2,2,1,2,2,1,2,2,1,135834.019526,140154.312816,2,92,10,10,3.51,3,3,0,0,0,1,24,1,4,5,NA
+71797,7,2,1,80,NA,3,3,1,NA,NA,1,1,1,1,NA,NA,3,2,NA,1,2,2,NA,NA,NA,1,2,2,NA,26771.485907,29740.715741,2,100,NA,NA,NA,4,4,0,0,2,2,51,NA,NA,1,NA
+71798,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,8491.292032,8557.713588,1,96,15,15,5,2,2,0,0,2,2,62,1,4,1,2
+71799,7,2,1,59,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,151766.599459,151581.541662,3,91,14,1,0.41,2,1,0,0,0,1,47,NA,NA,5,NA
+71800,7,2,2,33,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,2,1,2,2,2,2,1,2,2,1,2,2,1,45655.090694,44423.220436,3,92,7,7,1.3,5,5,1,2,0,2,33,2,2,1,1
+71801,7,2,2,4,NA,1,1,1,4,56,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17887.570772,19748.358154,1,92,10,10,3.04,4,4,1,1,0,1,32,1,3,1,2
+71802,7,2,1,53,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,NA,1,2,2,1,2,2,1,2,2,1,27250.100481,27216.872857,1,99,6,3,0.92,2,1,0,0,0,2,50,1,3,6,NA
+71803,7,2,2,52,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,16063.886913,15924.431895,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+71804,7,2,2,9,NA,1,1,1,9,110,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,13616.85154,13950.045541,1,102,6,6,0.96,5,5,0,2,0,1,32,2,2,1,3
+71805,7,2,1,22,NA,5,6,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,1,2,2,1,14313.345971,14977.822328,3,91,7,7,1.33,6,6,0,0,2,2,51,2,5,1,5
+71806,7,2,2,12,NA,4,4,1,12,147,NA,NA,1,1,NA,6,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12412.374308,12473.945857,2,96,4,4,0.57,6,6,0,3,0,2,29,1,3,4,NA
+71807,7,2,1,12,NA,2,2,1,12,147,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,2,2,2,2,2,2,2,20560.901695,20667.963739,2,96,7,7,1.57,4,4,0,2,0,1,40,2,2,1,5
+71808,7,2,1,12,NA,1,1,1,12,147,NA,NA,1,1,NA,5,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,24228.858782,24355.020132,2,102,7,7,1.53,5,5,0,3,0,1,43,2,2,1,4
+71809,7,2,1,5,NA,4,4,1,5,65,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,14467.4421,14643.906573,2,101,4,4,1.22,2,2,1,0,0,2,31,1,4,77,NA
+71810,7,2,1,33,NA,2,2,2,NA,NA,2,NA,2,2,2,NA,3,1,NA,2,2,2,2,2,2,2,2,2,2,34887.439952,34624.133414,2,94,14,4,1.47,5,1,0,0,0,1,24,2,4,5,NA
+71811,7,2,2,53,NA,5,6,2,NA,NA,2,NA,2,1,99,NA,1,3,NA,1,2,1,1,2,2,1,1,1,3,14813.321679,14891.653673,2,94,77,77,NA,6,6,2,0,0,2,18,1,3,NA,NA
+71812,7,2,1,3,NA,2,2,1,3,38,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,11179.43727,11533.315753,2,93,14,14,2.91,6,6,2,0,1,2,74,NA,NA,2,NA
+71813,7,2,1,36,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17091.959624,18613.193853,1,90,15,15,5,4,4,2,0,0,1,36,1,5,1,5
+71814,7,2,1,3,NA,1,1,1,3,45,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,20874.345556,22365.354744,3,92,14,14,3.25,4,4,2,0,0,2,33,1,5,1,5
+71815,7,2,1,0,0,2,2,2,NA,1,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,2,2,2,NA,NA,NA,NA,4857.494088,5095.295928,2,90,4,4,0.57,5,5,1,0,2,2,80,2,1,2,NA
+71816,7,2,1,2,NA,1,1,1,2,34,NA,NA,1,1,NA,NA,NA,NA,NA,2,1,2,1,2,2,NA,NA,NA,NA,14457.854197,14219.651494,1,92,6,6,1.34,4,4,1,0,0,1,25,2,3,1,3
+71817,7,2,1,8,NA,4,4,1,8,103,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8294.996898,8761.908989,2,100,8,8,1.1,7,7,3,3,0,2,58,1,3,5,NA
+71818,7,2,1,8,NA,1,1,1,8,105,NA,NA,1,1,NA,2,NA,NA,NA,2,1,2,2,2,2,1,2,2,1,11036.458246,10927.552283,1,102,13,13,NA,6,6,1,2,0,2,36,2,4,6,NA
+71819,7,2,2,25,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,4,1,2,2,2,2,1,2,2,1,2,1,2,51600.2972,52672.714014,1,101,9,9,2.88,3,3,1,0,0,1,36,2,2,1,4
+71820,7,2,1,47,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,138834.18124,139448.476692,1,100,15,15,5,5,5,0,3,0,1,47,1,5,1,5
+71821,7,2,2,50,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,NA,NA,NA,NA,16181.169973,16286.716901,2,100,6,6,2.04,2,2,0,0,0,2,50,1,2,1,3
+71822,7,2,2,19,NA,4,4,1,19,229,2,NA,1,1,NA,15,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,18163.985724,18749.311901,2,101,2,1,0.32,2,1,0,0,0,2,19,1,4,NA,NA
+71823,7,2,2,4,NA,3,3,1,4,51,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,87448.870731,96516.512941,2,94,14,14,4.71,3,3,1,0,0,1,35,1,5,1,5
+71824,7,2,1,48,NA,2,2,2,NA,NA,2,NA,2,2,3,NA,5,1,NA,2,2,2,1,2,2,2,2,2,2,44123.284536,45145.647683,2,91,3,3,0.73,3,3,0,0,0,2,22,2,2,5,NA
+71825,7,2,2,53,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,102322.335011,107871.941405,2,100,10,10,3.13,4,4,0,0,1,2,53,1,2,1,2
+71826,7,2,2,77,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,2,2,NA,1,2,2,1,2,2,1,2,2,NA,30880.887565,31700.101057,1,98,1,1,0.04,1,1,0,0,1,2,77,1,2,2,NA
+71827,7,2,1,14,NA,5,6,1,14,169,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7669.677276,8194.5967,2,92,15,15,4.59,4,4,0,2,0,2,48,1,5,1,5
+71828,7,2,2,49,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,34954.173075,38933.293222,2,98,7,7,1.97,4,4,0,1,0,1,40,1,3,1,3
+71829,7,2,1,57,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,4,NA,1,2,2,1,2,2,1,2,2,1,23857.322871,24256.958876,2,103,2,2,0.46,2,2,0,0,0,1,57,1,2,4,NA
+71830,7,2,2,34,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,3,6,2,2,2,2,2,2,2,NA,NA,NA,NA,34898.504426,34609.253559,1,100,8,3,0.68,6,3,1,0,0,1,33,2,3,6,NA
+71831,7,2,2,41,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,3,1,2,1,2,2,1,2,2,1,2,2,1,33767.584626,40865.10406,2,102,14,14,2.44,7,7,0,2,1,2,71,1,3,3,NA
+71832,7,2,2,49,NA,1,1,2,NA,NA,2,NA,2,1,6,NA,1,3,NA,2,2,2,1,2,2,2,2,2,2,40880.818805,41857.641766,1,101,5,5,0.51,7,7,0,3,2,1,75,2,1,1,1
+71833,7,2,2,30,NA,5,6,2,NA,NA,2,NA,2,2,2,NA,5,1,2,1,2,1,1,2,2,1,2,2,1,20963.809917,22126.306943,1,93,10,10,4.76,2,2,0,0,0,1,29,2,5,1,5
+71834,7,2,2,0,5,4,4,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3861.876549,4016.370213,1,96,8,8,1.61,6,6,3,0,0,1,33,2,5,1,4
+71835,7,2,2,17,NA,4,4,2,17,207,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,11133.192774,11654.909373,3,91,2,2,0.25,4,4,0,2,0,2,35,1,3,5,NA
+71836,7,2,1,65,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,6036.688933,6083.909812,2,99,15,15,5,2,2,0,0,2,1,65,1,4,1,5
+71837,7,2,2,13,NA,2,2,2,13,161,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13824.001771,14551.102227,2,90,12,12,NA,4,4,0,2,0,2,38,2,4,1,4
+71838,7,2,2,10,NA,4,4,2,10,126,NA,NA,1,1,NA,4,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,7814.742747,8106.808519,2,95,4,4,0.65,4,4,1,2,0,2,27,1,3,5,NA
+71839,7,2,2,57,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,160743.928829,165029.101567,1,95,8,6,2.04,4,2,0,1,0,2,57,1,5,5,NA
+71840,7,2,2,11,NA,2,2,2,11,134,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,13450.921832,14026.252281,1,90,6,6,0.81,6,6,0,3,0,2,45,1,4,1,2
+71841,7,2,1,45,NA,2,2,1,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,28726.575428,29281.498535,2,100,14,14,3.36,4,4,1,1,0,1,45,2,5,1,2
+71842,7,2,1,6,NA,2,2,2,7,84,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,18107.947773,18211.90239,1,97,7,7,2.2,3,3,0,1,0,1,34,2,3,1,4
+71843,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,11507.810748,11454.924313,2,99,13,13,NA,6,6,2,1,0,2,31,1,4,6,NA
+71844,7,2,2,43,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,1,1,2,1,2,2,1,2,2,1,2,2,1,19943.783243,19397.969296,2,95,2,2,0.4,2,2,0,0,0,2,43,1,1,1,NA
+71845,7,2,1,8,NA,5,7,2,8,97,NA,NA,1,1,NA,1,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,8246.426933,10295.491756,1,99,10,10,3.99,3,3,0,1,0,1,36,2,2,6,NA
+71846,7,2,1,72,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,NA,15429.860615,16015.253437,1,98,12,12,NA,3,3,0,0,2,1,72,1,5,1,3
+71847,7,2,1,37,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,94644.050918,104631.781207,2,91,15,15,5,3,3,1,0,0,1,37,1,5,1,5
+71848,7,2,1,26,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,10559.047286,11105.574109,2,92,8,6,2.75,2,1,0,0,0,1,26,2,5,5,NA
+71849,7,2,2,56,NA,2,2,2,NA,NA,2,NA,2,1,7,NA,3,4,NA,2,2,2,2,2,2,1,2,2,2,21097.069797,24201.517172,2,90,3,3,1.1,1,1,0,0,0,2,56,2,3,4,NA
+71850,7,2,2,6,NA,4,4,1,7,84,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8437.316833,8584.499702,2,96,4,4,0.4,7,7,3,2,0,2,25,1,2,5,NA
+71851,7,2,1,17,NA,1,1,1,17,213,2,NA,2,2,3,11,NA,NA,NA,2,2,2,2,2,2,1,2,2,1,29234.272259,28953.402615,3,92,4,4,0.67,4,4,0,3,0,2,36,2,1,5,NA
+71852,7,2,1,19,NA,5,7,1,19,232,2,NA,1,1,NA,13,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24231.355333,25219.610889,2,100,5,5,1.3,3,3,0,1,0,2,46,1,3,2,NA
+71853,7,2,1,14,NA,4,4,1,14,169,NA,NA,1,1,NA,7,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,12046.265019,11945.683979,1,102,7,7,1.57,4,4,0,2,0,2,33,1,4,1,4
+71854,7,2,2,0,7,4,4,2,NA,8,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3756.765605,3760.183196,1,99,4,4,0.53,7,7,3,1,0,2,26,1,1,5,NA
+71855,7,2,2,62,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,1,60351.409248,61337.452183,2,101,3,3,0.98,2,2,0,0,2,1,62,1,4,1,3
+71856,7,2,2,32,NA,5,6,2,NA,NA,2,NA,2,2,3,NA,5,1,2,1,2,2,1,2,2,1,2,2,1,16369.916397,16695.550655,1,101,10,10,3.67,3,3,1,0,0,1,36,2,5,1,5
+71857,7,2,1,39,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,78529.577822,81088.445647,1,98,15,15,5,5,5,0,3,0,2,41,1,5,6,NA
+71858,7,2,1,27,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,NA,NA,NA,1,2,2,1,33510.443506,34234.256724,1,98,4,4,0.89,3,3,0,0,1,2,55,1,5,1,NA
+71859,7,2,2,0,9,4,4,2,NA,10,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,NA,NA,NA,NA,NA,NA,NA,3516.231705,3777.543312,2,99,2,2,0.19,7,7,3,1,0,2,43,1,2,4,NA
+71860,7,2,1,28,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,16436.212192,16355.203111,1,99,12,12,NA,2,2,0,0,0,1,28,1,5,5,NA
+71861,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,1,2,2,1,11862.436765,15907.964192,1,90,3,3,1.07,1,1,0,0,1,2,68,1,2,5,NA
+71862,7,2,2,54,NA,5,6,1,NA,NA,2,NA,2,2,6,NA,1,4,NA,1,1,2,1,2,2,NA,NA,NA,NA,10489.35334,11245.538838,3,91,14,14,2.5,6,6,1,1,1,2,37,2,2,1,5
+71863,7,2,1,38,NA,3,3,2,NA,NA,1,2,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,67281.364849,72381.052566,2,100,15,15,5,3,3,0,1,0,1,38,1,4,1,4
+71864,7,2,1,18,NA,2,2,2,18,221,2,NA,1,1,NA,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,16033.31661,16386.200415,2,90,8,8,2.01,4,4,0,0,1,2,67,2,4,2,NA
+71865,7,2,2,1,23,4,4,2,NA,24,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6247.52442,6810.692829,1,99,7,7,1.06,7,7,3,1,0,1,38,1,4,6,NA
+71866,7,2,1,32,NA,1,1,1,NA,NA,2,NA,2,2,3,NA,4,3,NA,2,2,2,1,2,2,2,2,2,2,37080.526463,37974.333878,2,103,12,77,NA,2,1,0,0,0,1,46,2,4,1,NA
+71867,7,2,2,26,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,5,5,2,1,2,2,1,2,2,1,2,2,1,11696.173591,12753.148801,1,93,10,10,3.09,4,4,0,0,1,2,41,NA,NA,1,NA
+71868,7,2,1,43,NA,1,1,1,NA,NA,2,NA,2,1,6,NA,3,1,NA,2,2,2,2,2,2,1,2,2,2,35406.972937,38657.357615,3,92,9,9,2.46,4,4,0,2,0,1,43,2,3,1,4
+71869,7,2,1,69,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,130234.912635,128905.524968,1,95,8,8,2.7,3,3,0,1,2,1,69,1,5,1,3
+71870,7,2,2,41,NA,3,3,1,NA,NA,2,NA,2,1,5,NA,2,3,2,1,2,2,1,2,2,1,2,2,1,26595.398371,26621.513872,1,94,3,3,0.93,2,2,0,0,0,2,41,2,2,3,NA
+71871,7,2,2,43,NA,5,6,2,NA,NA,2,NA,2,1,6,NA,3,3,2,1,2,1,1,2,1,1,2,1,NA,18255.735511,18352.270791,2,91,12,12,NA,5,5,1,1,0,1,39,NA,NA,1,NA
+71872,7,2,1,77,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,NA,12468.859946,13204.521198,2,96,6,6,1.77,2,2,0,0,2,1,77,1,2,1,2
+71873,7,2,2,20,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,4,6,2,1,2,2,1,2,2,1,2,2,1,122726.905904,130163.310429,1,101,8,6,2.69,2,1,0,0,0,2,20,1,4,6,NA
+71874,7,2,1,24,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,14385.653726,15564.966804,2,101,8,4,1.7,2,1,0,0,0,1,24,2,5,5,NA
+71875,7,2,1,42,NA,4,4,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,23766.023166,23781.059132,1,91,15,15,5,6,6,1,2,0,2,42,2,5,1,5
+71876,7,2,1,14,NA,1,1,1,14,177,NA,NA,1,1,NA,8,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,22621.505951,24307.488585,1,92,4,4,0.65,4,4,0,1,0,1,47,2,2,1,3
+71877,7,2,1,72,NA,2,2,1,NA,NA,2,NA,2,2,2,NA,2,1,NA,2,2,2,2,2,2,1,2,1,NA,14336.984082,15039.787037,2,100,99,99,NA,6,6,1,1,2,1,37,2,3,1,3
+71878,7,2,1,35,NA,2,2,1,NA,NA,2,NA,2,1,3,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,39943.378263,40565.140919,2,93,7,7,1.56,4,4,1,1,0,1,35,2,4,1,4
+71879,7,2,1,24,NA,4,4,1,NA,NA,2,NA,1,1,NA,NA,4,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,25815.880139,26556.735732,2,101,2,2,0.51,2,1,0,0,0,1,24,1,4,5,NA
+71880,7,2,1,46,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,21158.364877,21285.812242,1,96,12,12,NA,2,2,0,0,0,1,46,1,4,1,5
+71881,7,2,1,20,NA,5,6,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,13370.868197,14159.618266,1,97,15,15,5,4,4,0,0,0,1,51,2,5,1,5
+71882,7,2,2,60,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,2,1,NA,1,2,2,1,2,2,1,2,2,1,12331.419303,13379.957577,2,95,3,3,0.75,2,2,0,0,2,1,60,1,2,1,2
+71883,7,2,1,42,NA,5,6,2,NA,NA,2,NA,2,2,5,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,17210.953493,17934.928395,1,91,77,77,NA,4,4,1,1,0,1,42,2,5,1,5
+71884,7,2,1,49,NA,5,6,1,NA,NA,2,NA,2,2,5,NA,1,1,NA,1,2,1,1,2,1,NA,NA,NA,NA,11601.882948,11708.549693,2,92,77,77,NA,4,4,0,0,0,1,27,2,2,5,NA
+71885,7,2,1,35,NA,1,1,2,NA,NA,2,NA,2,2,5,NA,2,6,NA,2,2,2,1,2,2,2,2,2,2,31045.881083,33019.807379,2,97,4,4,0.6,6,6,2,2,0,1,35,2,2,6,NA
+71886,7,2,1,61,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,5,NA,1,2,2,1,2,2,1,2,2,1,6612.194774,6870.810722,2,99,6,6,2.28,1,1,0,0,1,1,61,1,3,5,NA
+71887,7,2,2,68,NA,2,2,2,NA,NA,2,NA,2,1,8,NA,1,1,NA,1,2,2,NA,NA,NA,1,2,2,1,12689.611047,13462.268549,1,90,4,4,1.12,2,2,0,0,2,1,68,2,4,1,1
+71888,7,2,2,67,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,3,3,NA,1,2,2,1,2,2,1,2,2,1,10346.035773,10764.448363,1,99,2,2,0.31,4,4,1,0,1,2,67,1,3,3,NA
+71889,7,2,2,5,NA,5,7,1,5,61,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7483.230909,8143.441843,2,96,15,15,5,4,4,1,1,0,2,35,2,5,1,5
+71890,7,1,2,31,NA,3,3,NA,NA,NA,2,NA,1,1,NA,NA,3,5,3,1,2,2,1,2,2,NA,NA,NA,NA,76271.00266,0,3,91,8,8,3.4,2,2,1,0,0,2,31,1,3,5,NA
+71891,7,2,2,54,NA,1,1,1,NA,NA,2,NA,2,1,7,NA,4,1,NA,2,2,2,2,2,2,2,2,2,2,30976.631116,32561.666417,2,102,9,9,3.24,3,3,0,0,0,1,54,2,4,1,4
+71892,7,2,1,0,3,4,4,2,NA,5,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,6327.979262,6425.104835,1,90,6,6,1.92,2,2,1,0,0,2,51,2,1,5,NA
+71893,7,2,2,3,NA,5,7,2,3,40,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,8173.816615,8894.95474,1,97,15,15,5,3,3,1,0,0,1,40,1,3,1,5
+71894,7,2,1,11,NA,2,2,2,11,143,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,11113.498573,11364.131177,1,94,14,14,3.4,5,5,0,3,0,2,41,1,4,1,4
+71895,7,2,1,31,NA,5,6,1,NA,NA,2,NA,2,2,2,NA,5,1,NA,1,2,2,1,2,2,1,2,2,3,17165.91562,17929.203991,2,96,15,6,2.3,3,1,0,0,0,1,31,2,5,1,NA
+71896,7,2,1,4,NA,4,4,2,4,59,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,9304.437652,9588.632035,2,97,14,14,3.91,4,4,1,1,0,1,38,1,4,1,5
+71897,7,2,2,68,NA,4,4,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,12331.419303,12882.003985,2,95,77,77,NA,2,2,0,0,2,1,68,1,4,1,4
+71898,7,2,2,65,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,15207.312407,15896.113669,1,92,9,9,3.97,2,2,0,0,2,2,65,1,4,1,4
+71899,7,2,2,3,NA,3,3,1,3,37,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,66833.888628,73763.947124,2,98,7,7,1.61,4,4,1,1,0,1,43,NA,NA,6,NA
+71900,7,1,2,26,NA,2,2,NA,NA,NA,2,NA,1,1,NA,NA,5,5,3,1,2,2,1,2,2,NA,NA,NA,NA,55675.708832,0,1,93,3,2,0.46,2,1,0,0,0,2,26,1,5,5,NA
+71901,7,2,2,48,NA,5,7,2,NA,NA,2,NA,1,1,NA,NA,4,1,NA,1,2,2,1,2,2,1,2,2,1,30442.30641,31779.541767,1,101,3,3,0.88,2,2,0,0,0,1,56,1,2,1,4
+71902,7,2,2,67,NA,5,6,2,NA,NA,2,NA,2,1,8,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,11494.286549,11937.924974,2,94,8,8,3.4,2,2,0,0,2,1,80,1,3,1,5
+71903,7,2,2,6,NA,4,4,1,6,73,NA,NA,1,1,NA,0,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,11076.064101,11514.984987,2,102,2,2,0.36,4,4,1,2,0,2,36,1,3,5,NA
+71904,7,2,2,30,NA,5,6,1,NA,NA,2,NA,2,1,6,NA,4,5,2,1,2,2,1,2,1,NA,NA,NA,NA,11032.714892,11055.242776,2,92,5,5,0.64,7,7,1,2,1,1,66,2,1,1,3
+71905,7,1,2,8,NA,5,6,NA,NA,NA,NA,NA,1,1,NA,3,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,7405.452085,0,1,95,6,6,1.34,4,4,0,2,0,2,32,2,3,2,NA
+71906,7,2,2,4,NA,5,6,2,4,54,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,3788.132941,4021.443194,3,90,77,77,NA,7,7,1,2,0,1,41,2,3,6,NA
+71907,7,2,1,80,NA,3,3,2,NA,NA,1,1,1,1,NA,NA,3,1,NA,1,2,2,1,2,2,1,2,2,NA,47098.572584,50542.386793,1,95,9,9,4.08,2,2,0,0,2,1,80,1,3,1,NA
+71908,7,2,2,66,NA,3,3,1,NA,NA,2,NA,1,1,NA,NA,5,2,NA,1,2,2,1,2,2,1,2,2,1,93265.413087,94789.216803,2,98,10,10,4.55,2,2,0,0,1,2,66,1,5,2,NA
+71909,7,2,1,28,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,2,5,NA,1,2,2,1,2,2,NA,NA,NA,NA,37911.437415,38428.199561,2,103,2,2,0.46,2,2,0,0,0,1,57,1,2,4,NA
+71910,7,2,2,0,5,3,3,2,NA,6,NA,NA,1,1,NA,NA,NA,NA,NA,1,1,2,1,2,2,NA,NA,NA,NA,17151.556324,17621.071345,1,98,14,14,3.37,5,5,1,2,0,1,27,1,5,1,5
+71911,7,2,1,27,NA,1,1,1,NA,NA,2,NA,1,1,NA,NA,5,1,NA,1,2,2,1,2,2,1,2,2,1,42165.369652,43039.787442,2,102,14,14,3.25,5,5,2,0,0,1,27,1,5,1,5
+71912,7,2,1,40,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,1,1,NA,1,2,2,1,2,2,1,2,2,1,19633.637051,20770.138122,1,98,6,6,1.73,3,3,0,1,0,2,39,1,4,1,1
+71913,7,2,2,18,NA,5,6,1,18,226,2,NA,2,1,4,11,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,7382.152016,8028.485773,1,94,14,14,2.78,6,5,0,2,1,1,61,1,4,1,5
+71914,7,2,2,10,NA,3,3,2,10,131,NA,NA,1,1,NA,5,NA,NA,NA,1,1,2,1,2,2,1,2,2,1,60197.256541,63931.531988,2,94,14,14,2.63,6,6,1,3,0,1,39,1,4,1,4
+71915,7,2,1,60,NA,3,3,2,NA,NA,2,NA,1,1,NA,NA,5,5,NA,1,2,2,1,2,2,1,2,2,1,88961.259215,91446.591982,3,90,10,10,5,1,1,0,0,1,1,60,1,5,5,NA
+71916,7,2,1,16,NA,3,3,1,16,198,NA,NA,1,1,NA,9,NA,NA,NA,1,2,2,1,2,2,1,2,2,1,24446.632088,24751.360191,1,94,4,4,0.79,3,3,0,1,0,1,49,1,2,3,NA
diff --git a/pandas/pandas/tests/io/sas/data/DEMO_PUF.cpt b/pandas/pandas/tests/io/sas/data/DEMO_PUF.cpt
new file mode 100644
index 0000000000000..d74b6a70d2812
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/DEMO_PUF.cpt differ
diff --git a/pandas/pandas/tests/io/sas/data/DRXFCD_G.csv b/pandas/pandas/tests/io/sas/data/DRXFCD_G.csv
new file mode 100644
index 0000000000000..3fceacd1273bd
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/DRXFCD_G.csv
@@ -0,0 +1,7619 @@
+"DRXFDCD","DRXFCSD","DRXFCLD"
+1.1e+07,"MILK, HUMAN","Milk, human"
+11100000,"MILK, NFS","Milk, NFS"
+11111000,"MILK, COW'S, FLUID, WHOLE","Milk, cow's, fluid, whole"
+11111100,"MILK, COW'S, FLUID, WHOLE, LOW SODIUM","Milk, cow's, fluid, whole, low-sodium"
+11111150,"MILK, CALCIUM FORTIFIED, WHOLE, COW'S, FLUID","Milk, calcium fortified, cow's, fluid, whole"
+11111160,"MILK, CALCIUM FORTIFIED, COW'S, FLUID, 1% FAT","Milk, calcium fortified, cow's, fluid, 1% fat"
+11111170,"MILK, CALCIUM FORTIFIED, SKIM/NONFAT, COW, FLUID","Milk, calcium fortified, cow's, fluid, skim or nonfat"
+11112110,"MILK, COW'S, FLUID, 2% FAT","Milk, cow's, fluid, 2% fat"
+11112120,"MILK, COW'S, FLUID, ACIDOPHILUS, 1% FAT","Milk, cow's, fluid, acidophilus, 1% fat"
+11112130,"MILK, COW'S, FLUID, ACIDOPHILUS, 2% FAT","Milk, cow's, fluid, acidophilus, 2% fat"
+11112210,"MILK, COW'S, FLUID, 1% FAT","Milk, cow's, fluid, 1% fat"
+11113000,"MILK, COW'S, FLUID, SKIM OR NONFAT","Milk, cow's, fluid, skim or nonfat, 0.5% or less butterfat"
+11114300,"MILK, LOW LACTOSE, 1% FAT","Milk, cow's, fluid, lactose reduced, 1% fat"
+11114310,"MILK, LOW LACTOSE, 1% FAT, FORTIFIED WITH CALCIUM","Milk, cow's, fluid, lactose reduced, 1% fat, fortified with calcium"
+11114320,"MILK, LOW LACTOSE, NONFAT","Milk, cow's, fluid, lactose reduced, nonfat"
+11114321,"MILK, LOW LACTOSE, NONFAT, W/ CALCIUM","Milk, cow's, fluid, lactose reduced, nonfat, fortified with calcium"
+11114330,"MILK, COW'S FL LACTOSE REDUCED 2% FAT (LACTAID)","Milk, cow's, fluid, lactose reduced, 2% fat"
+11114350,"MILK, COW'S, FLUID, LACTOSE REDUCED, WHOLE","Milk, cow's, fluid, lactose reduced, whole"
+11115000,"BUTTERMILK, FLUID (INCLUDE KEFIR MILK)","Buttermilk, fluid, nonfat"
+11115100,"BUTTERMILK, FLUID, 1% FAT","Buttermilk, fluid, 1% fat"
+11115200,"BUTTERMILK, FLUID, 2% FAT","Buttermilk, fluid, 2% fat"
+11115300,"BUTTERMILK, FLUID, WHOLE","Buttermilk, fluid, whole"
+11116000,"MILK, GOAT'S, FLUID, WHOLE","Milk, goat's, fluid, whole"
+11120000,"MILK, DRY, RECONSTITUTED, NFS","Milk, dry, reconstituted, NFS"
+11121100,"MILK, DRY, RECONSTITUTED, WHOLE","Milk, dry, reconstituted, whole"
+11121210,"MILK, DRY, RECONSTITUTED, LOWFAT","Milk, dry, reconstituted, lowfat"
+11121300,"MILK, DRY, RECONSTITUTED, NONFAT","Milk, dry, reconstituted, nonfat"
+11210050,"MILK, EVAPORATED, NS AS TO FAT CONTENT","Milk, evaporated, NS as to fat content (formerly NS as to dilution, used in coffee or tea, assume undiluted)"
+11211050,"MILK, EVAPORATED, WHOLE","Milk, evaporated, whole (formerly NS as to dilution, used in coffee or tea)"
+11211400,"MILK, EVAPORATED, 2% FAT","Milk, evaporated, 2% fat (formerly NS as to dilution)"
+11212050,"MILK, EVAPORATED, SKIM","Milk, evaporated, skim (formerly NS as to dilution, used in coffee or tea)"
+11220000,"MILK, CONDENSED, SWEETENED","Milk, condensed, sweetened (formerly NS as to dilution)"
+11320000,"MILK, SOY, READY-TO-DRINK, NOT BABY","Milk, soy, ready-to-drink, not baby's"
+11320100,"MILK, SOY, LIGHT, READY-TO-DRINK, NOT BABY'S","Milk, soy, light, ready-to-drink, not baby's"
+11320200,"MILK, SOY, NONFAT, READY-TO-DRINK, NOT BABY'S","Milk, soy, nonfat, ready-to-drink, not baby's"
+11321000,"MILK, SOY, READY-TO-DRINK, NOT BABY'S, CHOCOLATE","Milk, soy, ready-to-drink, not baby's, chocolate"
+11321100,"MILK, SOY, LIGHT, READY-TO-DRINK, NOT BABY'S, CHOCOLATE","Milk, soy, light, ready-to-drink, not baby's, chocolate"
+11321200,"MILK, SOY, NONFAT, READY-TO-DRINK, NOT BABY'S, CHOCOLATE","Milk, soy, nonfat, ready-to-drink, not baby's, chocolate"
+11340000,"MILK,IMITATION,FLUID,NONSOY,SWEETENED,NOT CHOCOLATE","Milk, imitation, fluid, non-soy, sweetened, flavors other than chocolate"
+11350000,"MILK, ALMOND, READY-TO-DRINK","Milk, almond, ready-to-drink"
+11350010,"MILK, ALMOND, READY-TO-DRINK, CHOCOLATE","Milk, almond, ready-to-drink, chocolate"
+11410000,"YOGURT, NS AS TO TYPE OF MILK/FLAVOR","Yogurt, NS as to type of milk or flavor"
+11411010,"YOGURT, PLAIN, NS AS TO TYPE OF MILK","Yogurt, plain, NS as to type of milk"
+11411100,"YOGURT, PLAIN, WHOLE MILK","Yogurt, plain, whole milk"
+11411200,"YOGURT, PLAIN, LOWFAT MILK","Yogurt, plain, lowfat milk"
+11411300,"YOGURT, PLAIN, NONFAT MILK","Yogurt, plain, nonfat milk"
+11420000,"YOGURT, VANILLA, LEMON, COFFEE, NS AS TO MILK TYPE","Yogurt, vanilla, lemon, or coffee flavor, NS as to type of milk"
+11421000,"YOGURT, VANILLA, LEMON, COFFEE, WHOLE MILK","Yogurt, vanilla, lemon, or coffee flavor, whole milk"
+11422000,"YOGURT, VANILLA, LEMON, COFFEE, LOWFAT MILK","Yogurt, vanilla, lemon, maple, or coffee flavor, lowfat milk"
+11422100,"YOGURT, VANILLA, LEMON, COFFEE, LOWFAT MILK, LOW CAL SWTNR","Yogurt, vanilla, lemon, maple, or coffee flavor, lowfat milk, sweetened with low calorie sweetener"
+11423000,"YOGURT, VANILLA, LEMON, COFFEE, NONFAT MILK","Yogurt, vanilla, lemon, maple, or coffee flavor, nonfat milk"
+11424000,"YOGURT, VANILLA, LEMON, COFFEE, NONFAT MILK, LOW CAL SWEET","Yogurt, vanilla, lemon, maple, or coffee flavor, nonfat milk, sweetened with low calorie sweetener"
+11425000,"YOGURT, CHOCOLATE, NS AS TO TYPE OF MILK","Yogurt, chocolate, NS as to type of milk"
+11426000,"YOGURT, CHOCOLATE, WHOLE MILK","Yogurt, chocolate, whole milk"
+11427000,"YOGURT, CHOCOLATE, NONFAT MILK","Yogurt, chocolate, nonfat milk"
+11430000,"YOGURT, FRUIT VARIETY, NS AS TO MILK TYPE","Yogurt, fruit variety, NS as to type  of milk"
+11431000,"YOGURT, FRUIT VARIETY, WHOLE MILK","Yogurt, fruit variety, whole milk"
+11432000,"YOGURT, FRUIT VARIETY, LOWFAT MILK","Yogurt, fruit variety, lowfat milk"
+11432500,"YOGURT, FRUIT VARIETY, LOWFAT MILK, W/ LOW CAL SWEETENER","Yogurt, fruit variety, lowfat milk, sweetened with low-calorie sweetener"
+11433000,"YOGURT, FRUIT VARIETY, NONFAT MILK","Yogurt, fruit variety, nonfat milk"
+11433500,"YOGURT, FRUITED, NONFAT MILK, LOW CAL SWEETENER","Yogurt, fruit variety, nonfat milk, sweetened with low-calorie sweetener"
+11446000,"FRUIT AND LOWFAT YOGURT PARFAIT","Fruit and lowfat yogurt parfait"
+11459990,"YOGURT, FROZEN, NS AS TO FLAVOR, NS TO TYPE OF MILK","Yogurt, frozen, NS as to flavor, NS as to type of milk"
+11460000,"YOGURT, FROZEN, NOT CHOCOLATE, TYPE OF MILK NS","Yogurt, frozen, flavors other than chocolate, NS as to type of milk"
+11460100,"YOGURT, FROZEN, CHOCOLATE, TYPE OF MILK NS","Yogurt, frozen, chocolate, NS as to type of milk"
+11460150,"YOGURT, FROZEN, NS AS TO FLAVOR, LOWFAT MILK","Yogurt, frozen, NS as to flavor, lowfat milk"
+11460160,"YOGURT, FROZEN, CHOCOLATE, LOWFAT MILK","Yogurt, frozen, chocolate, lowfat milk"
+11460170,"YOGURT, FROZEN, NOT CHOCOLATE, LOWFAT MILK","Yogurt, frozen, flavors other than chocolate, lowfat milk"
+11460190,"YOGURT, FROZEN, NS AS TO FLAVOR, NONFAT MILK","Yogurt, frozen, NS as to flavor, nonfat milk"
+11460200,"YOGURT, FROZEN, CHOCOLATE, NONFAT MILK","Yogurt, frozen, chocolate, nonfat milk"
+11460250,"YOGURT,FROZEN,NOT CHOCOLATE,W/ SORBET/SORBET-COATED","Yogurt, frozen, flavors other than chocolate, with sorbet or sorbet-coated"
+11460300,"YOGURT, FROZEN, NOT CHOCOLATE, NONFAT MILK","Yogurt, frozen, flavors other than chocolate, nonfat milk"
+11460400,"YOGURT,FRZ,CHOCOLATE,NONFAT MILK,W/ LOW-CAL SWEET","Yogurt, frozen, chocolate, nonfat milk, with low-calorie sweetener"
+11460410,"YOGURT,FRZ,NOT CHOC,NONFAT MILK,W/ LOW-CAL SWEET","Yogurt, frozen, flavors other than chocolate, nonfat milk, with low-calorie sweetener"
+11460420,"YOGURT, FROZEN, NS AS TO FLAVOR, WHOLE MILK","Yogurt, frozen, NS as to flavor, whole milk"
+11460430,"YOGURT, FROZEN, CHOCOLATE, WHOLE MILK","Yogurt, frozen, chocolate, whole milk"
+11460440,"YOGURT, FROZEN, NOT CHOCOLATE, WHOLE MILK","Yogurt, frozen, flavors other than chocolate, whole milk"
+11461000,"YOGURT, FROZEN, CHOCOLATE-COATED","Yogurt, frozen, chocolate-coated"
+11461200,"YOGURT, FROZEN, SANDWICH","Yogurt, frozen, sandwich"
+11461250,"YOGURT, FROZEN, CONE, CHOCOLATE","Yogurt, frozen, cone, chocolate"
+11461260,"YOGURT, FROZEN, CONE, NOT CHOCOLATE","Yogurt, frozen, cone, flavors other than chocolate"
+11461270,"YOGURT, FROZEN, CONE, NOT CHOCOLATE, LOWFAT MILK","Yogurt, frozen, cone, flavors other than chocolate, lowfat milk"
+11461280,"YOGURT, FROZ, CONE, CHOCOLATE, LOWFAT MILK","Yogurt, frozen, cone, chocolate, lowfat milk"
+11480010,"YOGURT, WHOLE MILK, BF","Yogurt, whole milk, baby food"
+11480020,"YOGURT, WHOLE MILK, BF, W/FRUIT& MULTIGRAIN CEREAL,NFS","Yogurt, whole milk, baby food, with fruit and multigrain cereal puree, NFS"
+11480030,"YOGURT, WHOLE MILK, BF, W/FRUIT&MULTIGRAIN CEREAL + IRON","Yogurt, whole milk, baby food, with fruit and multigrain cereal puree, plus iron"
+11480040,"YOGURT, WHOLE MILK, BF, W/FRUIT&MULTIGRAIN CEREAL + DHA","Yogurt, whole milk, baby food, with fruit and multigrain cereal puree, plus DHA"
+11511000,"MILK, CHOCOLATE, NFS","Milk, chocolate, NFS"
+11511100,"MILK, CHOCOLATE, WHOLE MILK BASED","Milk, chocolate, whole milk-based"
+11511200,"MILK, CHOCOLATE, RED FAT, 2%","Milk, chocolate, reduced fat milk-based, 2% (formerly ""lowfat"")"
+11511300,"MILK, CHOCOLATE, SKIM MILK BASED","Milk, chocolate, skim milk-based"
+11511400,"MILK, CHOCOLATE, LOWFAT MILK BASED","Milk, chocolate, lowfat milk-based"
+11512000,"COCOA,HOT CHOCOLATE,NOT FROM DRY MIX, W/WHOLE MILK","Cocoa, hot chocolate, not from dry mix, made with whole milk"
+11512500,"HOT CHOCOLATE, P.R., MADE W/ WHOLE MILK","Hot chocolate, Puerto Rican style, made with whole milk"
+11512510,"HOT CHOCOLATE, P.R., MADE W/ LOW FAT MILK","Hot chocolate, Puerto Rican style, made with low fat milk"
+11513000,"COCOA & SUGAR MIXTURE, MILK ADDED, NS TYPE MILK","Cocoa and sugar mixture, milk added, NS as to type of milk"
+11513100,"COCOA & SUGAR MIXTURE, WHOLE MILK ADDED","Cocoa and sugar mixture, whole milk added"
+11513150,"COCOA & SUGAR MIXTURE, REDUCED FAT MILK ADDED","Cocoa and sugar mixture, reduced fat milk added"
+11513200,"COCOA & SUGAR MIXTURE, LOWFAT MILK ADDED","Cocoa and sugar mixture, lowfat milk added"
+11513300,"COCOA & SUGAR MIXTURE, SKIM MILK ADDED","Cocoa and sugar mixture, skim milk added"
+11513350,"COCOA AND SUGAR MIXTURE, REDUCED SUGAR, MILK ADDED, NS TYPE","Cocoa and sugar mixture, reduced sugar, milk added, NS as to type of milk"
+11513355,"COCOA AND SUGAR MIXTURE, REDUCED SUGAR, WHOLE MILK ADDED","Cocoa and sugar mixture, reduced sugar, whole milk added"
+11513360,"COCOA AND SUGAR MIXTURE, REDUCED SUGAR, REDUCED FAT MILK ADD","Cocoa and sugar mixture, reduced sugar, reduced fat milk added"
+11513365,"COCOA AND SUGAR MIXTURE, REDUCED SUGAR, LOWFAT MILK ADDED","Cocoa and sugar mixture, reduced sugar, lowfat milk added"
+11513370,"COCOA AND SUGAR MIXTURE, REDUCED SUGAR, SKIM MILK ADDED","Cocoa and sugar mixture, reduced sugar, skim milk added"
+11513400,"CHOCOLATE SYRUP, MILK ADDED, NS AS TO TYPE OF MILK","Chocolate syrup, milk added, NS as to type of milk"
+11513500,"CHOCOLATE SYRUP, WHOLE MILK ADDED","Chocolate syrup, whole milk added"
+11513550,"CHOCOLATE SYRUP, RED FAT MILK ADDED","Chocolate syrup, reduced fat milk added"
+11513600,"CHOCOLATE SYRUP, LOWFAT MILK ADDED","Chocolate syrup, lowfat milk added"
+11513700,"CHOCOLATE SYRUP, SKIM MILK ADDED","Chocolate syrup, skim milk added"
+11514100,"COCOA, SUGAR, & DRY MILK MIXTURE, WATER ADDED","Cocoa, sugar, and dry milk mixture, water added"
+11514300,"COCOA W/ NF DRY MILK, LO CAL SWEETENER, WATER ADDED","Cocoa with nonfat dry milk and low calorie sweetener, mixture, water added"
+11514500,"COCOA W/ WHEY, LO CAL SWEETNR, FORTIFD, WATER ADDED","Cocoa, whey, and low calorie sweetener, mixture, fortified, water added"
+11515100,"COCOA & SUGAR W/ MILK, FORTIFIED, PUERTO RICAN","Cocoa and sugar mixture fortified with vitamins and minerals, milk added, NS as to type of milk, Puerto Rican style"
+11516000,"COCOA, WHEY, LO CAL SWEETNER MIX, LOWFAT MILK ADDED","Cocoa, whey, and low-calorie sweetener mixture, lowfat milk added"
+11518050,"MILK BEV W/NF DRY MILK, LO CAL SWEET,WATER,NOT CHOC","Milk beverage with nonfat dry milk and low calorie sweetener, water added, flavors other than chocolate"
+11519000,"MILK BEVERAGE, NOT CHOCOLATE, W/ WHOLE MILK","Milk beverage, made with whole milk, flavors other than chocolate"
+11519040,"MILK, FLAVORS OTHER THAN CHOCOLATE, NFS","Milk, flavors other than chocolate, NFS"
+11519050,"MILK, NOT CHOCOLATE, WHOLE MILK BASED","Milk, flavors other than chocolate, whole milk-based"
+11519105,"MILK, FLAVORS OTHER THAN CHOCOLATE, REDUCED FAT MILK-BASED","Milk, flavors other than chocolate, reduced fat milk-based"
+11519200,"MILK, FLAVORS OTHER THAN CHOCOLATE, LOWFAT MILK-BASED","Milk, flavors other than chocolate, lowfat milk-based"
+11519205,"MILK, FLAVORS OTHER THAN CHOCOLATE, SKIM-MILK BASED","Milk, flavors other than chocolate, skim-milk based"
+11520000,"MILK, MALTED, UNFORTIFIED, FLAVOR NS","Milk, malted, unfortified, NS as to flavor, made with milk"
+11521000,"MILK, MALTED, UNFORTIFIED, CHOCOLATE FLAVOR","Milk, malted, unfortified, chocolate, made with milk"
+11522000,"MILK, MALTED, UNFORTIFIED, NATURAL FLAVOR","Milk, malted, unfortified, natural flavor, made with milk"
+11525000,"MILK,MALTED,FORTIFIED,NATURAL FLAVOR (INCL OVALTINE","Milk, malted, fortified, natural flavor, made with milk"
+11526000,"MILK, MALTED, FORTIFIED, CHOCOLATE (INCL OVALTINE)","Milk, malted, fortified, chocolate, made with milk"
+11527000,"MILK, MALTED, FORTIFIED, (INCL OVALTINE)","Milk, malted, fortified, NS as to flavor, made with milk"
+11531000,"EGGNOG, MADE W/ WHOLE MILK (INCLUDE EGG NOG, NFS)","Eggnog, made with whole milk"
+11531500,"EGGNOG, MADE W/ 2% REDUCED FAT MILK","Eggnog, made with 2% reduced fat milk (formerly eggnog, made with ""2% lowfat"" milk)"
+11541000,"MILK SHAKE, NS AS TO FLAVOR OR TYPE","Milk shake, NS as to flavor or type"
+11541100,"MILK SHAKE,HOMEMADE/ FOUNTAIN-TYPE, NS AS TO FLAVOR","Milk shake, homemade or fountain-type, NS as to flavor"
+11541110,"MILK SHAKE, HOMEMADE OR FOUNTAIN-TYPE, CHOCOLATE","Milk shake, homemade or fountain-type, chocolate"
+11541120,"MILK SHAKE, HOMEMADE/FOUNTAIN-TYPE, NOT CHOCOLATE","Milk shake, homemade or fountain-type, flavors other than chocolate"
+11541400,"MILK SHAKE WITH MALT (INCL MALTED MILK W/ICE CREAM)","Milk shake with malt"
+11541500,"MILK SHAKE, MADE W/ SKIM MILK, CHOCOLATE","Milk shake, made with skim milk, chocolate"
+11541510,"MILK SHAKE,MADE W/ SKIM MILK, NOT CHOCOLATE","Milk shake, made with skim milk, flavors other than chocolate"
+11542000,"CARRY-OUT MILK SHAKE, NS AS TO FLAVOR","Carry-out milk shake, NS as to flavor"
+11542100,"CARRY-OUT MILK SHAKE, CHOCOLATE","Carry-out milk shake, chocolate"
+11542200,"CARRY-OUT MILK SHAKE, NOT CHOCOLATE","Carry-out milk shake, flavors other than chocolate"
+11551050,"MILK FRUIT DRINK (INCL LICUADO)","Milk fruit drink"
+11552200,"ORANGE JULIUS","Orange Julius"
+11553000,"FRUIT SMOOTHIE DRINK, W/ FRUIT OR JUICE & DAIRY PRODUCTS","Fruit smoothie drink, made with fruit or fruit juice and dairy products"
+11553100,"FRUIT SMOOTHIE DRINK, NFS","Fruit smoothie drink, NFS"
+11560000,"CHOC-FLAVORED DRINK, WHEY-&MILK-BASED(INCL YOO-HOO)","Chocolate-flavored drink, whey- and milk-based"
+11560020,"MILK DRINK, WHEY&MILK-BASE, NOT CHOC (INCL YOO-HOO)","Flavored milk drink, whey- and milk-based, flavors other than chocolate"
+11561000,"CAFE CON LECHE","Cafe con leche"
+11561010,"CAFE CON LECHE PREPARED W/ SUGAR","Cafe con leche prepared with sugar"
+11710000,"INFANT FORMULA, NFS","Infant formula, NFS"
+11710050,"SIMILAC EXPERT CARE ALIMENTUM, INFANT FORMULA, NS AS TO FORM","Similac Expert Care Alimentum, infant formula, NS as to form"
+11710051,"SIMILAC EXPERT CARE ALIMENTUM, INFANT FORMULA, READY-TO-FEED","Similac Expert Care Alimentum, infant formula, ready-to-feed"
+11710053,"SIMILAC EXPERT CARE ALIMENTUM,INF FORM,PREP FR PDR,WATER NFS","Similac Expert Care Alimentum, infant formula, prepared from powder, made with water, NFS"
+11710054,"SIMILAC EXPERT CARE ALIMENTUM,INF FORM,PREP FR PDR,TAP WATER","Similac Expert Care Alimentum, infant formula, prepared from powder, made with tap water"
+11710055,"SIMILAC EXPERT CARE ALIMENTUM, INF FORM, FR PDR, BTL WATER","Similac Expert Care Alimentum, infant formula, prepared from powder, made with plain bottled water"
+11710056,"SIMILAC EXPERT CARE ALIMENTUM,INF FORM,PREP FR PDR,BABY WATR","Similac Expert Care Alimentum, infant formula, prepared from powder, made with baby water"
+11710350,"SIMILAC ADVANCE, INFANT FORMULA, NS AS TO FORM","Similac Advance, infant formula, NS as to form"
+11710351,"SIMILAC ADVANCE, INFANT FORMULA, READY-TO-FEED","Similac Advance, infant formula, ready-to-feed"
+11710352,"SIMILAC ADVANCE, INF FORMULA, PREP FRM LIQ CONC, W/WATER,NFS","Similac Advance, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710353,"SIMILAC ADVANCE, INFANT FORMULA, PREP FRM PDR, W/WATER NFS","Similac Advance, infant formula, prepared from powder, made with water, NFS"
+11710354,"SIMILAC ADVANCE, INF FORMULA, PREP FRM LIQ CONC, W/TAP WATER","Similac Advance, infant formula, prepared from liquid concentrate, made with tap water"
+11710355,"SIMILAC ADVANCE, INF FORMULA, PREP FRM LIQ CONC, W/BOT WATER","Similac Advance, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710356,"SIMILAC ADVANCE, INF FORMULA, PREP FR LIQ CONC, W/ BABY WATR","Similac Advance, infant formula, prepared from liquid concentrate, made with baby water"
+11710357,"SIMILAC ADVANCE, INFANT FORMULA, PREP FRM PDR, W/TAP WATER","Similac Advance, infant formula, prepared from powder, made with tap water"
+11710358,"SIMILAC ADVANCE, INFANT FORMULA, PREP FRM PDR, W/BOT WATER","Similac Advance, infant formula, prepared from powder, made with plain bottled water"
+11710359,"SIMILAC ADVANCE, INFANT FORMULA, PREP FRM PDR, W/BABY WATER","Similac Advance, infant formula, prepared from powder, made with baby water"
+11710360,"SIMILAC ADVANCE ORGANIC, INFANT FORMULA, NS AS TO FORM","Similac Advance Organic, infant formula, NS as to form"
+11710361,"SIMILAC ADVANCE ORGANIC, INFANT FORMULA, READY-TO-FEED","Similac Advance Organic, infant formula, ready-to-feed"
+11710363,"SIMILAC ADVANCE ORGANIC,INF FORM,PREP FR PDR,W/WATER,NFS","Similac Advance Organic, infant formula, prepared from powder, made with water, NFS"
+11710367,"SIMILAC ADVANCE ORGANIC,INF FORM,PREP FR PDR,W/TAP WATER","Similac Advance Organic, infant formula, prepared from powder, made with tap water"
+11710368,"SIMILAC ADVANCE ORGANIC,INF FORM,PREP FR PDR,W/BOT WATER","Similac Advance Organic, infant formula, prepared from powder, made with plain bottled water"
+11710369,"SIMILAC ADVANCE ORGANIC,INF FORM,PREP FR PDR,W/BABY WATER","Similac Advance Organic, infant formula, prepared from powder, made with baby water"
+11710370,"SIMILAC SENSITIVE, INFANT FORMULA, NS AS TO FORM","Similac Sensitive, infant formula, NS as to form"
+11710371,"SIMILAC SENSITIVE, INFANT FORMULA, READY-TO-FEED","Similac Sensitive, infant formula, ready-to-feed"
+11710372,"SIMILAC SENSITIVE, INF FORM, PREP FRM LIQ CONC, W/WATER,NFS","Similac Sensitive, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710373,"SIMILAC SENSITIVE, INF FORM, PREP FRM PDR, W/ WATER,NFS","Similac Sensitive, infant formula, prepared from powder, made with water, NFS"
+11710374,"SIMILAC SENSITIVE, INF FORM, PREP FRM LIQ CONC, W/TAP WATER","Similac Sensitive, infant formula, prepared from liquid concentrate, made with tap water"
+11710375,"SIMILAC SENSITIVE, INF FORM, PREP FRM LIQ CONC, W/BOT WATER","Similac Sensitive, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710376,"SIMILAC SENSITIVE, INF FORM, PREP FRM LIQ CONC, W/BABY WATER","Similac Sensitive, infant formula, prepared from liquid concentrate, made with baby water"
+11710377,"SIMILAC SENSITIVE, INF FORM, PREP FRM PDR, W/TAP  WATER","Similac Sensitive, infant formula, prepared from powder, made with tap water"
+11710378,"SIMILAC SENSITIVE, INF FORM, PREP FRM PDR, W/ BOT WATER","Similac Sensitive, infant formula, prepared from powder, made with plain bottled water"
+11710379,"SIMILAC SENSITIVE, INF FORM, PREP FRM PDR, W/ BABY WATER","Similac Sensitive, infant formula, prepared from powder, made with baby water"
+11710380,"SIMILAC SENSITIVE FOR SPIT-UP, INFANT FORMULA, NS AS TO FORM","Similac Sensitive for Spit-Up, infant formula, NS as to form"
+11710381,"SIMILAC SENSITIVE FOR SPIT-UP, INFANT FORMULA, READY-TO-FEED","Similac Sensitive for Spit-Up, infant formula, ready-to-feed"
+11710383,"SIMILAC SENSITIVE SPIT-UP,INF FORM, FR PDR, W/ WATER, NFS","Similac Sensitive for Spit-Up, infant formula, prepared from powder, made with water, NFS"
+11710387,"SIMILAC SENSITIVE SPIT-UP,INF FORM,PREP FR PDR,W/TAP WATER","Similac Sensitive for Spit-Up, infant formula, prepared from powder, made with tap water"
+11710388,"SIMILAC SENSITIVE SPIT-UP,INF FORM,PREP FR PDR,W/BOT WATER","Similac Sensitive for Spit-Up, infant formula, prepared from powder, made with plain bottled water"
+11710389,"SIMILAC SENSITIVE SPIT-UP,INF FORM,PREP FR PDR,W/BABY WATER","Similac Sensitive for Spit-Up, infant formula, prepared from powder, made with baby water"
+11710470,"SIMILAC EXPERT CARE NEOSURE, INFANT FORMULA, NS AS TO FORM","Similac Expert Care NeoSure, infant formula, NS as to form"
+11710471,"SIMILAC EXPERT CARE NEOSURE, INFANT FORMULA, READY-TO-FEED","Similac Expert Care NeoSure, infant formula, ready-to-feed"
+11710473,"SIMILAC EXPERT CARE NEOSURE,INF FORM,PREP FR PDR,W/WATER,NFS","Similac Expert Care NeoSure, infant formula, prepared from powder, made with water, NFS"
+11710477,"SIMILAC EXPERT CARE NEOSURE,INF FORM,PREP FR PDR,W/TAP WATER","Similac Expert Care NeoSure, infant formula, prepared from powder, made with tap water"
+11710478,"SIMILAC EXPERT CARE NEOSURE,INF FORM,PREP FR PDR,W/BOT WATER","Similac Expert Care NeoSure, infant formula, prepared from powder, made with plain bottled water"
+11710479,"SIMILAC EXPERT CARE NEOSURE,INF FORM,PREP FR PDR,W/BABY WAT","Similac Expert Care NeoSure, infant formula, prepared from powder, made with baby water"
+11710480,"SIMILAC GO AND GROW, INFANT FORMULA, NS AS TO FORM","Similac Go and Grow, infant formula, NS as to form"
+11710481,"SIMILAC GO AND GROW,INF FORM,PREP FR PDR,W/WATER,NFS","Similac Go and Grow, infant formula, prepared from powder, made with water, NFS"
+11710482,"SIMILAC GO AND GROW,INF FORM,PREP FR PDR,W/TAP WATER","Similac Go and Grow, infant formula, prepared from powder, made with tap water"
+11710483,"SIMILAC GO AND GROW,INF FORM,PREP FR PDR,W/BOT WATER","Similac Go and Grow, infant formula, prepared from powder, made with plain bottled water"
+11710484,"SIMILAC GO AND GROW,INF FORM,PREP FR PDR,W/BABY WATER","Similac Go and Grow, infant formula, prepared from powder, made with baby water"
+11710620,"ENFAMIL PREMIUM NEWBORN, INFANT FORMULA, NS AS TO FORM","Enfamil PREMIUM Newborn, infant formula, NS as to form"
+11710621,"ENFAMIL PREMIUM NEWBORN, INFANT FORMULA, READY-TO-FEED","Enfamil PREMIUM Newborn, infant formula, ready-to-feed"
+11710626,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRO PDR, WATER NFS","Enfamil PREMIUM Newborn, infant formula, prepared from powder, made with water, NFS"
+11710627,"ENFAMIL PREMIUM NEWBORN, INFANT FORMULA, PREP FRM PDR,TAP","Enfamil PREMIUM Newborn, infant formula, prepared from powder, made with tap water"
+11710628,"ENFAMIL PREMIUM NEWBORN, INF FORM, PREP FRM PDR,BOTTLE WATER","Enfamil PREMIUM Newborn, infant formula, prepared from powder, made with plain bottled water"
+11710629,"ENFAMIL PREMIUM NEWBORN, INFANT FORMULA, PREP FRM PDR, BABY","Enfamil PREMIUM Newborn, infant formula, prepared from powder, made with baby water"
+11710630,"ENFAMIL PREMIUM INFANT, INFANT FORMULA, NS AS TO FORM","Enfamil PREMIUM Infant, infant formula, NS as to form"
+11710631,"ENFAMIL PREMIUM INFANT, INFANT FORMULA, READY-TO-FEED","Enfamil PREMIUM Infant, infant formula, ready-to-feed"
+11710632,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM LIQ CONC,WATER NF","Enfamil PREMIUM Infant, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710633,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM LIQ CONC,TAP WATE","Enfamil PREMIUM Infant, infant formula, prepared from liquid concentrate, made with tap water"
+11710634,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM LIQ CONC,BOT WATE","Enfamil PREMIUM Infant, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710635,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM LIQ CONC, BABY","Enfamil PREMIUM Infant, infant formula, prepared from liquid concentrate, made with baby water"
+11710636,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM PDR, WATER NFS","Enfamil PREMIUM Infant, infant formula, prepared from powder, made with water, NFS"
+11710637,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM PDR, TAP WATER","Enfamil PREMIUM Infant, infant formula, prepared from powder, made with tap water"
+11710638,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM PDR, BOT WATER","Enfamil PREMIUM Infant, infant formula, prepared from powder, made with plain bottled water"
+11710639,"ENFAMIL PREMIUM INFANT, INF FORM, PREP FRM PDR, BABY WATER","Enfamil PREMIUM Infant, infant formula, prepared from powder, made with baby water"
+11710640,"ENFAMIL PREMIUM LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil PREMIUM LIPIL, infant formula, NS as to form"
+11710642,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR LIQ CONC,W/WATER,NFS","Enfamil PREMIUM LIPIL, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710643,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR PDR,W/WATER,NFS","Enfamil PREMIUM LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710644,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR LIQ CONC,W/TAP WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from liquid concentrate, made with tap water"
+11710645,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR LIQ CONC,W/BOT WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710646,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR LIQ CONC,W/BABY WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from liquid concentrate, made with baby water"
+11710647,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR PDR,W/TAP WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from powder, made with tap water"
+11710648,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR PDR,W/BOT WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710649,"ENFAMIL PREMIUM LIPIL,INF FORM,PREP FR PDR,W/BABY WATER","Enfamil PREMIUM LIPIL, infant formula, prepared from powder, made with baby water"
+11710650,"ENFAMIL LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil LIPIL, infant formula, NS as to form"
+11710651,"ENFAMIL LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil LIPIL, infant formula, ready-to-feed"
+11710652,"ENFAMIL LIPIL, INFANT FORMULA, PREP FR LIQ CONC, W/WATER,NFS","Enfamil LIPIL, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710653,"ENFAMIL LIPIL, INFANT FORMULA, PREP FRM PDR, W/WATER,NFS","Enfamil LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710654,"ENFAMIL LIPIL, INFANT FORMULA, PREP FR LIQ CONC, W/TAP WATER","Enfamil LIPIL, infant formula, prepared from liquid concentrate, made with tap water"
+11710655,"ENFAMIL LIPIL, INFANT FORMULA, PREP FR LIQ CONC, W/BOT WATER","Enfamil LIPIL, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710656,"ENFAMIL LIPIL, INFANT FORMULA, PREP FR LIQ CONC, W/BABY WAT","Enfamil LIPIL, infant formula, prepared from liquid concentrate, made with baby water"
+11710657,"ENFAMIL LIPIL, INFANT FORMULA, PREP FRM PDR, W/TAP WATER","Enfamil LIPIL, infant formula, prepared from powder, made with tap water"
+11710658,"ENFAMIL LIPIL, INFANT FORMULA, PREP FRM PDR, W/BOT WATER","Enfamil LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710659,"ENFAMIL LIPIL, INFANT FORMULA, PREP FRM PDR, W/BABY WATER","Enfamil LIPIL, infant formula, prepared from powder, made with baby water"
+11710660,"ENFAMIL A.R. LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil A.R. Lipil, infant formula, NS as to form"
+11710661,"ENFAMIL A.R. LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil A.R. Lipil, infant formula, ready-to-feed"
+11710663,"ENFAMIL A.R., INFANT FORMULA, PREP FR PDR, W/WATER, NFS","Enfamil A.R. LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710664,"ENFAMIL A.R., INFANT FORMULA, PREP FR PDR, W/TAP WATER","Enfamil A.R. LIPIL, infant formula, prepared from powder, made with tap water"
+11710665,"ENFAMIL ENFACARE LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil EnfaCare LIPIL, infant formula, NS as to form"
+11710666,"ENFAMIL ENFACARE LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil EnfaCare LIPIL, infant formula, ready-to-feed"
+11710667,"ENFAMIL ENFACARE LIPIL, INF FORM, PREP FR PDR, W/ WATER,NFS","Enfamil EnfaCare LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710668,"ENFAMIL A.R., INFANT FORMULA, PREP FR PDR, W/BOT WATER","Enfamil A.R. LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710669,"ENFAMIL A.R., INFANT FORMULA, PREP FR PDR, W/BABY WATER","Enfamil A.R. LIPIL, infant formula, prepared from powder, made with baby water"
+11710670,"ENFAMIL GENTLEASE LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil Gentlease LIPIL, infant formula, NS as to form"
+11710671,"ENFAMIL GENTLEASE LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil Gentlease LIPIL, infant formula, ready-to-feed"
+11710673,"ENFAMIL GENTLEASE LIPIL, INF FORM, PREP FRM PDR, W/WATER,NFS","Enfamil Gentlease LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710674,"ENFAMIL ENFACARE LIPIL, INF FORM, PREP FR PDR, W/ TAP WATER","Enfamil EnfaCare LIPIL, infant formula, prepared from powder, made with tap water"
+11710675,"ENFAMIL ENFACARE LIPIL, INF FORM, PREP FR PDR, W/ BOT WATER","Enfamil EnfaCare LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710676,"ENFAMIL ENFACARE LIPIL, INF FORM, PREP FR PDR, W/BABY  WATER","Enfamil EnfaCare LIPIL, infant formula, prepared from powder, made with baby water"
+11710677,"ENFAMIL GENTLEASE LIPIL, INF FORM, PREP FRM PDR, W/TAP WATER","Enfamil Gentlease LIPIL, infant formula, prepared from powder, made with tap water"
+11710678,"ENFAMIL GENTLEASE LIPIL, INF FORM, PREP FRM PDR, W/BOT WATER","Enfamil Gentlease LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710679,"ENFAMIL GENTLEASE LIPIL, INF FORM, PREP FRM PDR, W/ BABY WAT","Enfamil Gentlease LIPIL, infant formula, prepared from powder, made with baby water"
+11710680,"ENFAMIL ENFAGROW PREM NEXT STEP LIPIL , INF FORMULA, NS FORM","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, NS as to form"
+11710681,"ENFAMIL ENFAGROW PREM NEXT STEP, INF FORMULA, RTF","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, ready-to-feed"
+11710683,"ENFAMIL ENFAGROW PREM NEXT STEP,INF FORMULA,PDR,W/WATER,NFS","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710687,"ENFAMIL ENFAGROW PREM NEXT STEP,INF FORMULA,PDR,W/TAP WATER","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, prepared from powder, made with tap water"
+11710688,"ENFAMIL ENFAGROW PREM NEXT STEP,INF FORMULA,PDR,W/BOT WATER","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710689,"ENFAMIL ENFAGROW PREM NEXT STEP,INF FORMULA,PDR,W/BABY WATER","Enfamil Enfagrow PREMIUM Next Step LIPIL, infant formula, prepared from powder, made with baby water"
+11710690,"ENFAMIL GENTLEASE NEXT STEP LIPIL, INFANT FORMULA, NS FORM","Enfamil Gentlease Next Step LIPIL, infant formula, NS as to form"
+11710693,"ENFAMIL GENTLEASE NEXT STEP,INF FORM,PREP FR PDR,W/WATER,NFS","Enfamil Gentlease Next Step LIPIL, infant formula, prepared from powder, made with water, NFS"
+11710697,"ENFAMIL GENTLEASE NEXT STEP,INF FORM,PREP FR PDR,W/TAP WATER","Enfamil Gentlease Next Step LIPIL, infant formula, prepared from powder, made with tap water"
+11710698,"ENFAMIL GENTLEASE NEXT STEP,INF FORM,PREP FR PDR,W/BOT WATER","Enfamil Gentlease Next Step LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11710699,"ENFAMIL GENTLEASE NEXT STEP,INF FORM,PREP FR PDR,W/BABY WAT","Enfamil Gentlease Next Step LIPIL, infant formula, prepared from powder, made with baby water"
+11710800,"PEDIASURE, INFANT FORMULA, NS AS TO FORM","Pediasure, infant formula, NS as to form"
+11710801,"PEDIASURE,INFANT FORMULA, READY-TO-FEED","Pediasure, infant formula, ready-to-feed"
+11710805,"PEDIASURE FIBER, INFANT FORMULA, NS AS TO FORM","Pediasure Fiber, infant formula, NS as to form"
+11710806,"PEDIASURE FIBER, INFANT FORMULA, READY-TO-FEED","Pediasure Fiber, infant formula, ready-to-feed"
+11710910,"GERBER GOOD START  GENTLE PLUS, INF FORM, NS FORM","Gerber Good Start  Gentle Plus, infant formula, NS as to form"
+11710911,"GERBER GOOD START GENTLE PLUS, INFANT FORMULA, READY-TO-FEED","Gerber Good Start Gentle Plus, infant formula, ready-to-feed"
+11710912,"GERBER GOOD START GENTLE PLUS,PREP FRM LIQ CONC,W/WATER,NFS","Gerber Good Start Gentle Plus, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710913,"GERBER GOOD START GENTLE PLUS, PREP FRM PDR,W/WATER,NFS","Gerber Good Start Gentle Plus, infant formula, prepared from powder, made with water, NFS"
+11710914,"GERBER GOOD START GENTLE PLUS,PREP FRM LIQ CONC,W/TAP WATER","Gerber Good Start Gentle Plus, infant formula, prepared from liquid concentrate, made with tap water"
+11710915,"GERBER GOOD START GENTLE PLUS,PREP FRM LIQ CONC,W/BOT WATER","Gerber Good Start Gentle Plus, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710916,"GERBER GOOD START GENTLE PLUS,PREP FRM LIQ CONC,W/BABY WATER","Gerber Good Start Gentle Plus, infant formula, prepared from liquid concentrate, made with baby water"
+11710917,"GERBER GOOD START GENTLE PLUS, PREP FRM PDR,W/TAP WATER","Gerber Good Start Gentle Plus, infant formula, prepared from powder, made with tap water"
+11710918,"GERBER GOOD START GENTLE PLUS, PREP FRM PDR,W/BOT WATER","Gerber Good Start Gentle Plus, infant formula, prepared from powder, made with plain bottled water"
+11710919,"GERBER GOOD START GENTLE PLUS, PREP FRM PDR,W/BABY WATER","Gerber Good Start Gentle Plus, infant formula, prepared from powder, made with baby water"
+11710920,"GERBER GOOD START PROTECT PLUS, INFANT FORMULA, NS FORM","Gerber Good Start Protect Plus, infant formula, NS as to form"
+11710923,"GERBER GOOD START PROTECT PLUS, PREP FRM PDR,W/WATER,NFS","Gerber Good Start Protect Plus, infant formula, prepared from powder, made with water, NFS"
+11710927,"GERBER GOOD START PROTECT PLUS, PREP FRM PDR,W/TAP WATER","Gerber Good Start Protect Plus, infant formula, prepared from powder, made with tap water"
+11710928,"GERBER GOOD START PROTECT PLUS, PREP FRM PDR,W/BOT WATER","Gerber Good Start Protect Plus, infant formula, prepared from powder, made with plain bottled water"
+11710929,"GERBER GOOD START PROTECT PLUS, PREP FRM PDR,W/BABY WATER","Gerber Good Start Protect Plus, infant formula, prepared from powder, made with baby water"
+11710930,"GERBER GOOD START 2 GENTLE PLUS, INFANT FORMULA, NS  FORM","Gerber Good Start 2 Gentle Plus, infant formula, NS as to form"
+11710933,"GERBER GOOD START 2 GENTLE PLUS, PREP FRM PDR,W/WATER,NFS","Gerber Good Start 2 Gentle Plus, infant formula, prepared from powder, made with water, NFS"
+11710937,"GERBER GOOD START 2 GENTLE PLUS, PREP FRM PDR,W/TAP WATER","Gerber Good Start 2 Gentle Plus, infant formula, prepared from powder, made with tap water"
+11710938,"GERBER GOOD START 2 GENTLE PLUS, PREP FRM PDR,W/BOT WATER","Gerber Good Start 2 Gentle Plus, infant formula, prepared from powder, made with plain bottled water"
+11710939,"GERBER GOOD START 2 GENTLE PLUS, PREP FRM PDR,W/BABY WATER","Gerber Good Start 2 Gentle Plus, infant formula, prepared from powder, made with baby water"
+11710940,"GERBER GOOD START 2 PROTECT PLUS, INFANT FORMULA, NS FORM","Gerber Good Start 2 Protect Plus, infant formula, NS as to form"
+11710943,"GERBER GOOD START 2 PROTECT PLUS, PREP FRM PDR,W/WATER,NFS","Gerber Good Start 2 Protect Plus, infant formula, prepared from powder, made with water, NFS"
+11710947,"GERBER GOOD START 2 PROTECT PLUS, PREP FRM PDR,W/TAP WATER","Gerber Good Start 2 Protect Plus, infant formula, prepared from powder, made with tap water"
+11710948,"GERBER GOOD START 2 PROTECT PLUS, PREP FRM PDR,W/BOT WATER","Gerber Good Start 2 Protect Plus, infant formula, prepared from powder, made with plain bottled water"
+11710949,"GERBER GOOD START 2 PROTECT PLUS, PREP FRM PDR,W/BABY WATER","Gerber Good Start 2 Protect Plus, infant formula, prepared from powder, made with baby water"
+11710960,"AMERICA'S STORE BRAND, INFANT FORMULA, NS AS TO FORM","America's Store Brand, infant formula, NS as to form"
+11710961,"AMERICA'S STORE BRAND,INF FORM,PREP FRM LIQ CONC,W/WATER,NFS","America's Store Brand, infant formula, prepared from liquid concentrate, made with water, NFS"
+11710962,"AMERICA'S STORE BRAND,INF FORM,PREP FRM PDR,W/ WATER, NFS","America's Store Brand, infant formula, prepared from powder, made with water, NFS"
+11710963,"AMERICA'S STORE BRAND, INFANT FORMULA, READY-TO-FEED","America's Store Brand, infant formula, ready-to-feed"
+11710964,"AMERICA'S STORE BRAND,INF FORM,PREP FRM LIQ CONC,W/TAP WATER","America's Store Brand, infant formula, prepared from liquid concentrate, made with tap water"
+11710965,"AMERICA'S STORE BRAND,INF FORM,PREP FRM LIQ CONC,W/BOT WATER","America's Store Brand, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11710966,"AMERICA'S STORE BRAND,INF FORM,PREP FRM LIQ CONC,W/BABY WATR","America's Store Brand, infant formula, prepared from liquid concentrate, made with baby water"
+11710967,"AMERICA'S STORE BRAND,INF FORM,PREP FRM PDR,W/ TAP WATER","America's Store Brand, infant formula, prepared from powder, made with tap water"
+11710968,"AMERICA'S STORE BRAND,INF FORM,PREP FRM PDR,W/ BOT WATER","America's Store Brand, infant formula, prepared from powder, made with plain bottled water"
+11710969,"AMERICA'S STORE BRAND,INF FORM,PREP FRM PDR,W/ BABY WATER","America's Store Brand, infant formula, prepared from powder, made with baby water"
+11720310,"ENFAMIL PROSOBEE LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil ProSobee LIPIL, infant formula, NS as to form"
+11720311,"ENFAMIL PROSOBEE LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil ProSobee Lipil, infant formula, ready-to-feed"
+11720312,"ENFAMIL PROSOBEE LIPIL,INF FORM, FR LIQ CONC,W/ WATER,NFS","Enfamil ProSobee LIPIL, infant formula, prepared from liquid concentrate, made with water, NFS"
+11720313,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR  PDR, W/WATER, NFS","Enfamil ProSobee LIPIL, infant formula, prepared from powder, made with water, NFS"
+11720314,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR LIQ CONC,W/ TAP WATE","Enfamil ProSobee LIPIL, infant formula, prepared from liquid concentrate, made with tap water"
+11720315,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR LIQ CONC,W/ BOT WATE","Enfamil ProSobee LIPIL, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11720316,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR LIQ CONC,W/ BABY WAT","Enfamil ProSobee LIPIL, infant formula, prepared from liquid concentrate, made with baby water"
+11720317,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR  PDR, W/TAP WATER","Enfamil ProSobee LIPIL, infant formula, prepared from powder, made with tap water"
+11720318,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR  PDR, W/BOT WATER","Enfamil ProSobee LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11720319,"ENFAMIL PROSOBEE LIPIL,INF FORM,PREP FR  PDR, W/BABY WATER","Enfamil ProSobee LIPIL, infant formula, prepared from powder, made with baby water"
+11720320,"ENFAMIL ENFAGROW SOY NEXT STEP LIPIL, INF FORMULA, NS FORM","Enfamil Enfagrow Soy Next Step LIPIL, infant formula, NS as to form"
+11720323,"ENFAGROW SOY NEXT STEP LIPIL, INF FOR,FR PDR,W/WATER,NFS","Enfamil Enfagrow Soy Next Step LIPIL, infant formula, prepared from powder, made with water, NFS"
+11720327,"ENFAGROW SOY NEXT STEP LIPIL,INF FORM,PREP FR PDR,TAP WATER","Enfamil Enfagrow Soy Next Step LIPIL, infant formula, prepared from powder, made with tap water"
+11720328,"ENFAGROW SOY NEXT STEP LIPIL,INF FORM,PREP FR PDR, BOT WATER","Enfamil Enfagrow Soy Next Step LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11720329,"ENFAGROW SOY NEXT STEP LIPIL,INF FORM,PREP FR PDR,BABY WATER","Enfamil Enfagrow Soy Next Step LIPIL, infant formula, prepared from powder, made with baby water"
+11720410,"SIMILAC SENSITIVE ISOMIL SOY, INFANT FORMULA, NS AS TO FORM","Similac Sensitive Isomil Soy, infant formula, NS as to form"
+11720411,"SIMILAC SENSITIVE ISOMIL SOY, INFANT FORMULA, READY-TO-FEED","Similac Sensitive Isomil Soy, infant formula, ready-to-feed"
+11720412,"SIMILAC SENSITIVE ISOMIL SOY, PREP FR LIQ CONC,W/WATER,NFS","Similac Sensitive Isomil Soy, infant formula, prepared from liquid concentrate, made with water, NFS"
+11720413,"SIMILAC SENSITIVE ISOMIL SOY,INF FORM,FR PDR,W/WATER,NFS","Similac Sensitive Isomil Soy, infant formula, prepared from powder, made with water, NFS"
+11720414,"SIMILAC SENSITIVE ISOMIL SOY,PREP  FR LIQ CONC,W/TAP WATER","Similac Sensitive Isomil Soy, infant formula, prepared from liquid concentrate, made with tap water"
+11720415,"SIMILAC SENSITIVE ISOMIL SOY,PREP  FR LIQ CONC,W/BOT WATER","Similac Sensitive Isomil Soy, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11720416,"SIMILAC SENSITIVE ISOMIL SOY,PREP  FR LIQ CONC,W/BABY WATER","Similac Sensitive Isomil Soy, infant formula, prepared from liquid concentrate, made with baby water"
+11720417,"SIMILAC SENSITIVE ISOMIL SOY,INF FORM,PREP FR PDR+TAP WATER","Similac Sensitive Isomil Soy, infant formula, prepared from powder, made with tap water"
+11720418,"SIMILAC SENSITIVE ISOMIL SOY,INF FORM,PREP FR PDR,W/BOT WATE","Similac Sensitive Isomil Soy, infant formula, prepared from powder, made with plain bottled water"
+11720419,"SIMILAC SENSITIVE ISOMIL SOY,INF FORM,PREP FR PDR,W/BABY WAT","Similac Sensitive Isomil Soy, infant formula, prepared from powder, made with baby water"
+11720430,"SIMILAC EXPERT CARE FOR DIARRHEA, INFANT FORMULA, NS FORM","Similac Expert Care for Diarrhea, infant formula, NS as to form"
+11720431,"SIMILAC EXPERT CARE FOR DIARRHEA, INFANT FORMULA, RTF","Similac Expert Care for Diarrhea, infant formula, ready-to-feed"
+11720440,"SIMILAC GO AND GROW SOY, INFANT FORMULA, NS AS TO FORM","Similac Go and Grow Soy, infant formula, NS as to form"
+11720443,"SIMILAC GO AND GROW SOY, INF FORM, PREP FR PDR, WATER, NFS","Similac Go and Grow Soy, infant formula, prepared from powder, made with water, NFS"
+11720447,"SIMILAC GO AND GROW SOY,INF FORM,PREP FR PDR,TAP WATER","Similac Go and Grow Soy, infant formula, prepared from powder, made with tap water"
+11720448,"SIMILAC GO AND GROW SOY,INF FORM,PREP FR PDR,BOT WATER","Similac Go and Grow Soy, infant formula, prepared from powder, made with plain bottled water"
+11720449,"SIMILAC GO AND GROW SOY,INF FORM,PREP FR PDR,BABY WATER","Similac Go and Grow Soy, infant formula, prepared from powder, made with baby water"
+11720610,"GERBER GOOD START SOY PLUS, INFANT FORMULA, NS AS TO FORM","Gerber Good Start Soy Plus, infant formula, NS as to form"
+11720611,"GERBER GOOD START SOY PLUS, INFANT FORMULA, READY-TO-FEED","Gerber Good Start Soy Plus, infant formula, ready-to-feed"
+11720612,"GERBER GOOD START SOY PLUS,INF FORM,PREP FR LIQ CONC,W/WATER","Gerber Good Start Soy Plus, infant formula, prepared from liquid concentrate, made with water, NFS"
+11720613,"GERBER GOOD START SOY PLUS,INF FORM, PREP FR PDR,W/WATER,NFS","Gerber Good Start Soy Plus, infant formula, prepared from powder, made with water, NFS"
+11720614,"GERBER GOOD START SOY PLUS,INF FORM,FR LIQ CONC,W/TAP WATER","Gerber Good Start Soy Plus, infant formula, prepared from liquid concentrate, made with tap water"
+11720615,"GERBER GOOD START SOY PLUS,INF FORM,FR LIQ CONC,W/BOT WATER","Gerber Good Start Soy Plus, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11720616,"GERBER GOOD START SOY PLUS,INF FORM,FR LIQ CONC,W/BABY WTR","Gerber Good Start Soy Plus, infant formula, prepared from liquid concentrate, made with baby water"
+11720617,"GERBER GOOD START SOY PLUS,INF FORM, PREP FR PDR,W/TAP WATER","Gerber Good Start Soy Plus, infant formula, prepared from powder, made with tap water"
+11720618,"GERBER GOOD START SOY PLUS,INF FORM, PREP FR PDR,W/BOT WATER","Gerber Good Start Soy Plus, infant formula, prepared from powder, made with plain bottled water"
+11720619,"GERBER GOOD START SOY PLUS,INF FORM, PREP FR PDR,W/BABY WATE","Gerber Good Start Soy Plus, infant formula, prepared from powder, made with baby water"
+11720620,"GERBER GOOD START 2 SOY PLUS, INFANT FORMULA, NS AS TO FORM","Gerber Good Start 2 Soy Plus, infant formula, NS as to form"
+11720623,"GERBER GOOD START 2 SOY PLUS, INF FORM,PREP FR PDR,WATER,NFS","Gerber Good Start 2 Soy Plus, infant formula, prepared from powder, made with water, NFS"
+11720627,"GERBER GOOD START 2 SOY PLUS,INF FORM,PREP FR PDR,TAP WATER","Gerber Good Start 2 Soy Plus, infant formula, prepared from powder, made with tap water"
+11720628,"GERBER GOOD START 2 SOY PLUS,INF FORM,PREP FR PDR,BOT WATER","Gerber Good Start 2 Soy Plus, infant formula, prepared from powder, made with plain bottled water"
+11720629,"GERBER GOOD START 2 SOY PLUS,INF FORM,PREP FR PDR,BABY WATER","Gerber Good Start 2 Soy Plus, infant formula, prepared from powder, made with baby water"
+11720800,"AMERICA'S STORE BRAND SOY, INFANT FORMULA, NS AS TO FORM","America's Store Brand Soy, infant formula, NS as to form"
+11720801,"AMERICA'S STORE BRAND SOY, INFANT FORMULA, READY-TO-FEED","America's Store Brand Soy, infant formula, ready-to-feed"
+11720802,"AMERICA'S STORE BRAND SOY,INF FORM, FR LIQ CONC,W/WATER NFS","America's Store Brand Soy, infant formula, prepared from liquid concentrate, made with water, NFS"
+11720803,"AMERICA'S STORE BRAND SOY,INF FORM,FR PDR,W/WATER,NFS","America's Store Brand Soy, infant formula, prepared from powder, made with water, NFS"
+11720804,"AMERICA'S STORE BRAND SOY,INF FORM, FR LIQ CONC,W/TAP WATER","America's Store Brand Soy, infant formula, prepared from liquid concentrate, made with tap water"
+11720805,"AMERICA'S STORE BRAND SOY,INF FORM,FR LIQ CONC,W/BOT WATER","America's Store Brand Soy, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11720806,"AMERICA'S STORE BRAND SOY,INF FORM,FR LIQ CONC,W/BABY WATER","America's Store Brand Soy, infant formula, prepared from liquid concentrate, made with baby water"
+11720807,"AMERICA'S STORE BRAND SOY,INF FORM,PREP FR PDR,W/TAP WATER","America's Store Brand Soy, infant formula, prepared from powder, made with tap water"
+11720808,"AMERICA'S STORE BRAND SOY,INF FORM,PREP FR PDR,W/BOT WATER","America's Store Brand Soy, infant formula, prepared from powder, made with plain bottled water"
+11720809,"AMERICA'S STORE BRAND SOY,INF FORM,PREP FR PDR,W/BABY WATER","America's Store Brand Soy, infant formula, prepared from powder, made with baby water"
+11740310,"ENFAMIL NUTRAMIGEN LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil Nutramigen LIPIL, infant formula, NS as to form"
+11740311,"ENFAMIL NUTRAMIGEN LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil Nutramigen LIPIL, infant formula, ready-to-feed"
+11740312,"ENFAMIL NUTRAMIGEN LIPIL,INF FORM,PREP FR LIQ CONC,W/WAT,NFS","Enfamil Nutramigen LIPIL, infant formula, prepared from liquid concentrate, made with water, NFS"
+11740313,"ENFAMIL NUTRAMIGEN LIPIL, INF FORM, PREP FR PDR, W/WATER,NFS","Enfamil Nutramigen LIPIL, infant formula, prepared from powder, made with water, NFS"
+11740314,"ENFAMIL NUTRAMIGEN LIPIL,INF FORM,FR LIQ CONC,W/TAP WATER","Enfamil Nutramigen LIPIL, infant formula, prepared from liquid concentrate, made with tap water"
+11740315,"ENFAMIL NUTRAMIGEN LIPIL,INF FORM,FR LIQ CONC,W/BOT WATER","Enfamil Nutramigen LIPIL, infant formula, prepared from liquid concentrate, made with plain bottled water"
+11740316,"ENFAMIL NUTRAMIGEN LIPIL,INF FORM,FR LIQ CONC,W/BABY WATER","Enfamil Nutramigen LIPIL, infant formula, prepared from liquid concentrate, made with baby water"
+11740317,"ENFAMIL NUTRAMIGEN LIPIL, INF FORM, PREP FR PDR, W/TAP WATER","Enfamil Nutramigen LIPIL, infant formula, prepared from powder, made with tap water"
+11740318,"ENFAMIL NUTRAMIGEN LIPIL, INF FORM, PREP FR PDR, W/BOT WATER","Enfamil Nutramigen LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11740319,"ENFAMIL NUTRAMIGEN LIPIL, INF FORM, PREP FR PDR, W/BABY WATR","Enfamil Nutramigen LIPIL, infant formula, prepared from powder, made with baby water"
+11740320,"ENFAMIL NUTRAMIGEN AA LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil Nutramigen AA LIPIL, infant formula, NS as to form"
+11740323,"ENFAMIL NUTRAMIGEN AA LIPIL,INF FORM,PREP FR PDR,W/WATER,NFS","Enfamil Nutramigen AA LIPIL, infant formula, prepared from powder, made with water, NFS"
+11740327,"ENFAMIL NUTRAMIGEN AA LIPIL,INF FORM,PREP FR PDR,W/TAP WATER","Enfamil Nutramigen AA LIPIL, infant formula, prepared from powder, made with tap water"
+11740328,"ENFAMIL NUTRAMIGEN AA LIPIL,INF FORM,PREP FR PDR,W/BOT WATER","Enfamil Nutramigen AA LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11740329,"ENFAMIL NUTRAMIGEN AA LIPIL,INF FORM,PREP FR PDR,W/BABY  WAT","Enfamil Nutramigen AA LIPIL, infant formula, prepared from powder, made with baby water"
+11740400,"ENFAMIL PREGESTIMIL LIPIL, INFANT FORMULA, NS AS TO FORM","Enfamil Pregestimil LIPIL, infant formula, NS as to form"
+11740401,"ENFAMIL PREGESTIMIL LIPIL, INFANT FORMULA, READY-TO-FEED","Enfamil Pregestimil LIPIL, infant formula, ready-to-feed"
+11740403,"ENFAMIL PREGESTIMIL LIPIL,INF FORM,PREP FR PDR, W/WATER,NFS","Enfamil Pregestimil LIPIL, infant formula, prepared from powder, made with water, NFS"
+11740407,"ENFMAIL PREGESTIMIL LIPIL,INF FORM,PREP FR PDR, W/TAP WATER","Enfmail Pregestimil LIPIL, infant formula, prepared from powder, made with tap water"
+11740408,"ENFMAIL PREGESTIMIL LIPIL,INF FORM,PREP FR PDR, W/BOT WATER","Enfmail Pregestimil LIPIL, infant formula, prepared from powder, made with plain bottled water"
+11740409,"ENFMAIL PREGESTIMIL LIPIL,INF FORM,PREP FR PDR, W/BABY WATER","Enfmail Pregestimil LIPIL, infant formula, prepared from powder, made with baby water"
+11740510,"ENFAMIL PREMATURE LIPIL 20, LOW IRON, INF FORMULA, NS FORM","Enfamil Premature LIPIL 20, low iron, infant formula, NS as to form"
+11740511,"ENFAMIL PREMATURE LIPIL 20, LOW IRON, INFANT FORMULA, RTF","Enfamil Premature LIPIL 20, low iron, infant formula, ready-to-feed"
+11740520,"ENFAMIL PREMATURE LIPIL 20, W/IRON, INFANT FORMULA, NS FORM","Enfamil Premature LIPIL 20, with iron, infant formula, NS as to form"
+11740521,"ENFAMIL PREMATURE LIPIL 20, WITH IRON, INFANT FORMULA, RTF","Enfamil Premature LIPIL 20, with iron, infant formula, ready-to-feed"
+11740540,"ENFAMIL PREMATURE LIPIL 24, LOW IRON, INF FORMULA, NS FORM","Enfamil Premature LIPIL 24, low iron, infant formula, NS as to form"
+11740541,"ENFAMIL PREMATURE LIPIL 24, LOW IRON, INFANT FORMULA, RTF","Enfamil Premature LIPIL 24, low iron, infant formula, ready-to-feed"
+11740550,"ENFAMIL PREMATURE LIPIL 24, W/IRON, INFANT FORMULA, NS FORM","Enfamil Premature LIPIL 24, with iron, infant formula, NS as to form"
+11740551,"ENFAMIL PREMATURE LIPIL 24, WITH IRON, INFANT FORMULA, RTF","Enfamil Premature LIPIL 24, with iron, infant formula, ready-to-feed"
+11810000,"MILK, DRY, NOT RECONSTITUTED, NS AS TO FAT","Milk, dry, not reconstituted, NS as to whole, lowfat, or nonfat"
+11811000,"MILK, DRY, WHOLE, NOT RECONSTITUTED","Milk, dry, whole, not reconstituted"
+11812000,"MILK, DRY, LOWFAT, NOT RECONSTITUTED","Milk, dry, lowfat, not reconstituted"
+11813000,"MILK, DRY, NONFAT, NOT RECONSTITUTED","Milk, dry, nonfat, not reconstituted"
+11825000,"WHEY, SWEET, DRY","Whey, sweet, dry"
+11830100,"COCOA W/DRY MILK & SUGAR, DRY MIX, NOT RECONST","Cocoa (or chocolate) with dry milk and sugar, dry mix, not reconstituted"
+11830110,"COCOA POWDER W/ NFD MILK, LOW CAL SWEETENER, DRY","Cocoa powder with nonfat dry milk and low calorie sweetener, dry mix, not reconstituted"
+11830120,"COCOA W/ WHEY, LO CAL SWEETENER, FORTIFIED, DRY MIX","Cocoa, whey, and low calorie sweetener, fortified, dry mix, not reconstituted"
+11830140,"CHOCOLATE, INST, DRY MIX, FORTIFD, NOT RECONST,P.R.","Chocolate, instant, dry mix, fortified with vitamins and minerals, not reconstituted, Puerto Rican style"
+11830150,"COCOA POWDER, NOT RECONSTITUTED (NO DRY MILK)","Cocoa powder, not reconstituted (no dry milk)"
+11830160,"COCOA-FLAVORED BEVERAGE POWDER W/ SUGAR, DRY MIX","Cocoa (or chocolate) flavored beverage powder with sugar, dry mix, not reconstituted"
+11830165,"COCOA FLAV BEV PDR W/ RED SUGAR,DRY MIX,NOT RECONSTITUTED","Cocoa (or chocolate) flavored beverage powder with reduced sugar, dry mix, not reconstituted"
+11830170,"COCOA FLAV BEV PDR W/ LOW CAL SWTNR,DRY MIX,NOT RECONSTITUTD","Cocoa (or chocolate) flavored beverage powder with low-calorie sweetener, dry mix, not reconstituted"
+11830200,"MILK, MALTED, DRY, UNFORTIFD, NOT RECONST, NOT CHOC","Milk, malted, dry mix, unfortified, not reconstituted, flavors other than chocolate"
+11830210,"MILK, MALTED, DRY, FORTIFD, NOT RECONST, NOT CHOC","Milk, malted, dry mix, fortified, not reconstituted, flavors other than chocolate"
+11830250,"MILK, MALTED, DRY, UNFORTIFIED, NOT RECONST, CHOC","Milk, malted, dry mix, unfortified, not reconstituted, chocolate"
+11830260,"MILK, MALTED, DRY, FORTIFIED, NOT RECONST, CHOC","Milk, malted, dry mix, fortified, not reconstituted, chocolate"
+11830400,"MILK BEV POWDER, DRY, NOT RECONST, NOT CHOC","Milk beverage, powder, dry mix, not reconstituted, flavors other than chocolate"
+11830450,"MILK BEV MIX, W/ SUGAR,EGG WHITE, NOT RECONSTITUTED","Milk beverage with sugar, dry milk, and egg white powder, dry mix, not reconstituted"
+11830500,"MILK BEV POWDER W/ NFD MILK, LOW CAL, DRY, CHOC","Milk beverage, powder, with nonfat dry milk and low calorie sweetener, dry mix, not reconstituted, chocolate"
+11830550,"MILK BEV POWDER W/ NFD MILK, LOW CAL, DRY, NOT CHOC","Milk beverage, powder, with nonfat dry milk and low calorie sweetener, dry mix, not reconstituted, flavors other than chocolate"
+12100100,"CREAM, FLUID, NS AS TO LIGHT, HEAVY OR HALF & HALF","Cream, NS as to light, heavy, or half and half"
+12110100,"CREAM, LIGHT, FLUID (INCL COFFEE CRM, TABLE CREAM)","Cream, light, fluid"
+12110300,"CREAM, LIGHT, WHIPPED, UNSWEETENED","Cream, light, whipped, unsweetened"
+12120100,"CREAM, HALF & HALF","Cream, half and half"
+12120105,"CREAM, HALF & HALF, LOW FAT","Cream, half and half, low fat"
+12120110,"CREAM, HALF & HALF, FAT FREE","Cream, half and half, fat free"
+12130100,"CREAM, HEAVY, FLUID","Cream, heavy, fluid"
+12130200,"CREAM, HEAVY, WHIPPED, UNSWEETENED","Cream, heavy, whipped, unsweetened"
+12140000,"CREAM, HEAVY, WHIPPED, SWEETENED","Cream, heavy, whipped, sweetened"
+12140100,"CREAM, WHIPPED, PRESSURIZED CONTAINER","Cream, whipped, pressurized container"
+12140105,"CREAM, WHIPPED, PRESSURIZED CONTAINER, LIGHT","Cream, whipped, pressurized container, light"
+12140110,"WHIPPED TOPPING, DAIRY BASED, FAT FREE, PRESSURIZED CONTAINR","Whipped topping, dairy based, fat free, pressurized container"
+12200100,"CREAM SUBSTITUTE, NS AS TO FROZEN,LIQUID OR POWDER","Cream substitute, NS as to frozen, liquid, or powdered"
+12210100,"CREAM SUBSTITUTE, FROZEN","Cream substitute, frozen"
+12210200,"CREAM SUBSTITUTE, LIQUID (INCLUDE COFFEE WHITNER)","Cream substitute, liquid"
+12210210,"CREAM SUBSTITUTE, FLAVORED, LIQUID","Cream substitute, flavored, liquid"
+12210250,"CREAM SUBSTITUTE, LIGHT, LIQUID","Cream substitute, light, liquid"
+12210255,"CREAM SUBSTITUTE, LIGHT, FLAVORED, LIQUID","Cream substitute, light, flavored, liquid"
+12210260,"CREAM SUBSTITUTE, FAT FREE, LIQUID","Cream substitute, fat free, liquid"
+12210270,"CREAM SUBSTITUTE, FAT FREE, FLAVORED, LIQUID","Cream substitute, fat free, flavored, liquid"
+12210305,"CREAM SUBSTITUTE, SUGAR FREE, LIQUID","Cream substitute, sugar free, liquid"
+12210310,"CREAM SUBSTITUTE, SUGAR FREE, FLAVORED, LIQUID","Cream substitute, sugar free, flavored, liquid"
+12210400,"CREAM SUBSTITUTE, POWDERED","Cream substitute, powdered"
+12210410,"CREAM SUBST, LIGHT, POWDERED (INCL COFFEE MATE, CRE","Cream substitute, light, powdered"
+12210420,"CREAM SUBSTITUTE, FLAVORED, POWDERED","Cream substitute, flavored, powdered"
+12210430,"CREAM SUBSTITUTE, FAT FREE, POWDER","Cream substitute, fat free, powder"
+12210440,"CREAM SUBSTITUTE, FAT FREE, FLAVORED, POWDER","Cream substitute, fat free, flavored, powder"
+12210500,"CREAM SUBSTITUTE, SUGAR FREE, POWDER","Cream substitute, sugar free, powder"
+12210505,"CREAM SUBSTITUTE, SUGAR FREE, FLAVORED, POWDER","Cream substitute, sugar free, flavored, powder"
+12220000,"WHIPPED TOPPING, NONDAIRY, NS AS TO CND/FRZ/POWDER","Whipped topping, nondairy, NS as to canned, frozen, or made from powdered mix"
+12220100,"WHIPPED TOPPING, NONDAIRY, PRESSURIZED CAN","Whipped topping, nondairy, pressurized can"
+12220200,"WHIPPED TOPPING, NONDAIRY, FROZEN (INCL COOL WHIP)","Whipped topping, nondairy, frozen"
+12220250,"WHIPPED TOPPING, NONDAIRY, FZN, LOWFAT (INCL COOL)","Whipped topping, nondairy, frozen, lowfat"
+12220270,"WHIPPED TOPPING, NONDAIRY, FROZEN, FAT FREE","Whipped topping, nondairy, frozen, fat free"
+12220280,"WHIPPED TOPPING, NONDAIRY, FROZEN, SUGAR FREE","Whipped topping, nondairy, frozen, sugar free"
+12220300,"WHIPPED CREAM SUBST, NONDAIRY, FROM POWDERED MIX","Whipped cream substitute, nondairy, made from powdered mix"
+12220400,"WHIP CREAM SUB,NONDAIRY,LOWFAT,LO SUGAR,FROM MIX","Whipped cream substitute, nondairy, lowfat, low sugar, made from powdered mix"
+12310100,"SOUR CREAM (INCL W/ CHIVES)","Sour cream"
+12310200,"SOUR CREAM, HALF & HALF","Sour cream, half and half"
+12310300,"SOUR CREAM, REDUCED FAT","Sour cream, reduced fat"
+12310350,"SOUR CREAM, LIGHT","Sour cream, light"
+12310370,"SOUR CREAM, FAT FREE","Sour cream, fat free"
+12320100,"SOUR CREAM, IMITATION","Sour cream, imitation (nondairy)"
+12320200,"SOUR CREAM, FILLED, SOUR DRESSING, NONBUTTERFAT","Sour cream, filled, sour dressing, nonbutterfat"
+12350000,"DIP, SOUR CREAM BASE (INCLUDE BUTTERMILK-TYPE DIP)","Dip, sour cream base"
+12350020,"DIP, SOUR CREAM BASE, REDUCED CALORIE","Dip, sour cream base, reduced calorie"
+12350100,"SPINACH DIP","Spinach dip"
+12350110,"SPINACH AND ARTICHOKE DIP","Spinach and artichoke dip"
+13110000,"ICE CREAM, NFS","Ice cream, NFS"
+13110100,"ICE CREAM, REGULAR, NOT CHOCOLATE","Ice cream, regular, flavors other than chocolate"
+13110110,"ICE CREAM, REGULAR, CHOCOLATE","Ice cream, regular, chocolate"
+13110120,"ICE CREAM, RICH, FLAVORS OTHER THAN CHOCOLATE","Ice cream, rich, flavors other than chocolate"
+13110130,"ICE CREAM, RICH, CHOCOLATE","Ice cream, rich, chocolate"
+13110140,"ICE CREAM, RICH, NS AS TO FLAVOR","Ice cream, rich, NS as to flavor"
+13110200,"ICE CREAM, SOFT SERVE, NOT CHOCOLATE","Ice cream, soft serve, flavors other than chocolate"
+13110210,"ICE CREAM, SOFT SERVE, CHOCOLATE","Ice cream, soft serve, chocolate"
+13110220,"ICE CREAM, SOFT SERVE, NS AS TO FLAVOR","Ice cream, soft serve, NS as to flavor"
+13110310,"ICE CREAM, NO SUGAR ADDED, NS AS TO FLAVOR","Ice cream, no sugar added, NS as to flavor"
+13110320,"ICE CREAM, NO SUGAR ADDED, FLAVORS OTHER THAN CHOCOLATE","Ice cream, no sugar added, flavors other than chocolate"
+13110330,"ICE CREAM, NO SUGAR ADDED, CHOCOLATE","Ice cream, no sugar added, chocolate"
+13120050,"ICE CREAM BAR OR STICK, NOT CHOC- OR CAKE-COVERED","Ice cream bar or stick, not chocolate covered or cake covered"
+13120100,"ICE CREAM BAR/STICK, CHOCOLATE COVERED","Ice cream bar or stick, chocolate covered"
+13120110,"ICE CREAM BAR, CHOCOLATE/CARAMEL COVERED, W/ NUTS","Ice cream bar or stick, chocolate or caramel covered, with nuts"
+13120120,"ICE CREAM BAR,RICH CHOC ICE CREAM,THICK CHOC COVER","Ice cream bar or stick, rich chocolate ice cream, thick chocolate covering"
+13120121,"ICE CREAM BAR,RICH ICE CREAM,THICK CHOC COVER","Ice cream bar or stick, rich ice cream, thick chocolate covering"
+13120130,"ICE CREAM BAR/STICK,RICH ICE CREAM,CHOC COVER,W/NUT","Ice cream bar or stick, rich ice cream, chocolate covered, with nuts"
+13120140,"ICE CREAM BAR/STICK, CHOC ICE CREAM, CHOC COVER","Ice cream bar or stick, chocolate ice cream, chocolate covered"
+13120300,"ICE CREAM BAR, CAKE-COVERED","Ice cream bar, cake covered"
+13120310,"ICE CREAM BAR, STICK OR NUGGET, WITH CRUNCH COATING","Ice cream bar, stick or nugget, with crunch coating"
+13120400,"ICE CREAM BAR/STICK W/ FRUIT","Ice cream bar or stick with fruit"
+13120500,"ICE CREAM SANDWICH","Ice cream sandwich"
+13120550,"ICE CREAM COOKIE SANDWICH (INCLUDE CHIPWICH)","Ice cream cookie sandwich"
+13120700,"ICE CREAM CONE, W/ NUTS, NOT CHOCOLATE","Ice cream cone with nuts, flavors other than chocolate"
+13120710,"ICE CREAM CONE, CHOC-COVERED, W/ NUTS, NOT CHOC","Ice cream cone, chocolate covered, with nuts, flavors other than chocolate"
+13120720,"ICE CREAM CONE, CHOC-COVERED OR DIPPED, NOT CHOC","Ice cream cone, chocolate covered or dipped, flavors other than chocolate"
+13120730,"ICE CREAM CONE, NO TOPPING, NOT CHOCOLATE","Ice cream cone, no topping, flavors other than chocolate"
+13120740,"ICE CREAM CONE, NO TOPPING, NS AS TO FLAVOR","Ice cream cone, no topping, NS as to flavor"
+13120750,"ICE CREAM CONE, W/NUTS, CHOCOLATE ICE CREAM","Ice cream cone with nuts, chocolate ice cream"
+13120760,"ICE CREAM CONE, CHOC-COVERED, CHOC ICE CREAM","Ice cream cone, chocolate covered or dipped, chocolate ice cream"
+13120770,"ICE CREAM CONE, NO TOPPING, CHOCOLATE ICE CREAM","Ice cream cone, no topping, chocolate ice cream"
+13120780,"ICE CREAM CONE, CHOC-COVERED, W/NUT, CHOC ICE CREAM","Ice cream cone, chocolate covered, with nuts, chocolate ice cream"
+13120790,"ICE CREAM SUNDAE CONE (INCL DRUMSTICK, ALL FLAVORS)","Ice cream sundae cone"
+13120800,"ICE CREAM SODA, NOT CHOCOLATE","Ice cream soda, flavors other than chocolate"
+13120810,"ICE CREAM SODA, CHOCOLATE","Ice cream soda, chocolate"
+13121000,"ICE CREAM SUNDAE, TOPPING NS, W/ WHIPPED CREAM","Ice cream sundae, NS as to topping, with whipped cream"
+13121100,"ICE CREAM SUNDAE, FRUIT TOPPING, W/ WHIPPED CREAM","Ice cream sundae, fruit topping, with whipped cream"
+13121200,"ICE CREAM SUNDAE, PREPACKAGED, NOT CHOCOLATE","Ice cream sundae, prepackaged type, flavors other than chocolate"
+13121300,"ICE CREAM SUNDAE,CHOCOLATE TOPPING,W/ WHIPPED CREAM","Ice cream sundae, chocolate or fudge topping, with whipped cream"
+13121400,"ICE CREAM SUNDAE, NOT FRUIT/ CHOC TOP,W/ WHIP CREAM","Ice cream sundae, not fruit or chocolate topping, with whipped cream"
+13121500,"ICE CREAM SUNDAE, FUDGE TOPPING, W/ CAKE","Ice cream sundae, fudge topping, with cake, with whipped cream"
+13122100,"ICE CREAM PIE, NO CRUST","Ice cream pie, no crust"
+13122500,"ICE CREAM PIE,COOKIE CRUST,FUDGE TOPPING,WHIP CREAM","Ice cream pie, with cookie crust, fudge topping, and whipped cream"
+13126000,"ICE CREAM, FRIED","Ice cream, fried"
+13127000,"DIPPIN' DOTS, ICE CREAM, FLAVORS OTHER THAN CHOCOLATE","Dippin' Dots, flash frozen ice cream snacks, flavors other than chocolate"
+13127010,"DIPPIN' DOTS, ICE CREAM, CHOCOLATE","Dippin' Dots, flash frozen ice cream snacks, chocolate"
+13130100,"LT ICE CREAM, NS FLAV ( ICE MILK)","Light ice cream, NS as to flavor (formerly ice milk)"
+13130300,"LIGHT ICE CREAM,NOT CHOCOLATE (FORMERLY ICE MILK)","Light ice cream, flavors other than chocolate (formerly ice milk)"
+13130310,"LIGHT ICE CREAM,CHOCOLATE (FORMERLY ICE MILK)","Light ice cream, chocolate (formerly ice milk)"
+13130320,"LIGHT ICE CREAM, NO SUGAR ADDED, NS AS TO FLAVOR","Light ice cream, no sugar added, NS as to flavor"
+13130330,"LIGHT ICE CREAM, NO SUGAR ADDED, NOT CHOCOLATE","Light ice cream, no sugar added, flavors other than chocolate"
+13130340,"LIGHT ICE CREAM, NO SUGAR ADDED, CHOCOLATE","Light ice cream, no sugar added, chocolate"
+13130590,"LIGHT ICE CREAM,SOFT SERVE, NS FLAVOR (FORMERLY ICE MILK)","Light ice cream, soft serve, NS as to flavor (formerly ice milk)"
+13130600,"LIGHT ICE CREAM,SOFT SERVE, NOT CHOC (FORMERLY ICE MILK)","Light ice cream, soft serve, flavors other than chocolate (formerly ice milk)"
+13130610,"LIGHT ICE CREAM,SOFT SERVE CHOC (TASTEE FRZ, DAIRY QUEEN)","Light ice cream, soft serve, chocolate (formerly ice milk)"
+13130620,"LIGHT ICE CREAM,SOFT SERVE CONE,NOT CHOC (DAIRY QUEEN)","Light ice cream, soft serve cone, flavors other than chocolate (formerly ice milk)"
+13130630,"LIGHT ICE CREAM,SOFT SERVE CONE, CHOC (FORMERLY ICE MILK)","Light ice cream, soft serve cone, chocolate (formerly ice milk)"
+13130640,"LIGHT ICE CREAM,SOFT SERVE CONE, NS FLAV(FORMERLY ICE MILK)","Light ice cream, soft serve cone, NS as to flavor (formerly ice milk)"
+13130700,"LIGHT ICE CREAM, SOFT SERVE, BLENDED W/ CANDY OR COOKIES","Light ice cream, soft serve, blended with candy or cookies"
+13135000,"ICE CREAM SANDWICH, MADE W/ LIGHT ICE CREAM, NOT CHOCOLATE","Ice cream sandwich, made with light ice cream, flavors other than chocolate"
+13135010,"ICE CREAM SANDWICH, MADE W/ LIGHT CHOCOLATE ICE CREAM","Ice cream sandwich, made with light chocolate ice cream"
+13136000,"ICE CREAM SANDWICH, MADE W/ LIGHT, NO SUGAR ADDED ICE CREAM","Ice cream sandwich, made with light, no sugar added ice cream"
+13140100,"LIGHT ICE CREAM,BAR/STICK, CHOC-COATED (FORMERLY ICE MILK)","Light ice cream, bar or stick, chocolate-coated (formerly ice milk)"
+13140110,"LIGHT ICE CREAM,BAR, CHOC COVERED,W/NUTS (FORMERLY ICE MILK)","Light ice cream, bar or stick, chocolate covered, with nuts (formerly ice milk)"
+13140450,"LIGHT ICE CREAM,CONE, NFS (FORMERLY ICE MILK)","Light ice cream, cone, NFS (formerly ice milk)"
+13140500,"LIGHT ICE CREAM,CONE, NOT CHOCOLATE (FORMERLY ICE MILK)","Light ice cream, cone, flavors other than chocolate (formerly ice milk)"
+13140550,"LIGHT ICE CREAM,CONE, CHOCOLATE (FORMERLY ICE MILK)","Light ice cream, cone, chocolate (formerly ice milk)"
+13140570,"LIGHT ICE CREAM, NO SUGAR ADDED, CONE, NS AS TO FLAVOR","Light ice cream, no sugar added, cone, NS as to flavor"
+13140575,"LIGHT ICE CREAM, NO SUGAR ADDED, CONE, NOT CHOC","Light ice cream, no sugar added, cone, flavors other than chocolate"
+13140580,"LIGHT ICE CREAM, NO SUGAR ADDED, CONE, CHOCOLATE","Light ice cream, no sugar added, cone, chocolate"
+13140600,"LIGHT ICE CREAM,SUNDAE,SOFT SERVE,CHOC/FUDGE TOP (ICE MILK)","Light ice cream, sundae, soft serve, chocolate or fudge topping, with whipped cream (formerly ice milk)"
+13140630,"LIGHT ICE CREAM,SUNDAE,SOFT SERVE,FRUIT TOPPING (ICE MILK)","Light ice cream, sundae, soft serve, fruit topping, with whipped cream (formerly ice milk)"
+13140650,"LIGHT ICE CREAM,SUNDAE,SOFT SERVE,NOT FRUIT/CHOC TOPPING","Light ice cream, sundae, soft serve, not fruit or chocolate topping, with whipped cream (formerly ice milk)"
+13140660,"LIGHT ICE CREAM,SUNDAE,CHOC / FUDGE TOP (W/O WHIP CREAM)","Light ice cream, sundae, soft serve, chocolate or fudge topping (without whipped cream) (formerly ice milk)"
+13140670,"LIGHT ICE CREAM,SUNDAE,FRUIT TOP (W/O WHIP CREAM)(ICE MILK)","Light ice cream, sundae, soft serve, fruit topping (without whipped cream) (formerly ice milk)"
+13140680,"LIGHT ICE CREAM,SUNDAE,NO FRUIT/CHOC TOP (W/O WHIP CREAM)","Light ice cream, sundae, soft serve, not fruit or chocolate topping (without whipped cream) (formerly ice milk)"
+13140700,"LIGHT ICE CREAM,CREAMSICLE OR DREAMSICLE (FORMERLY ICE MILK)","Light ice cream, creamsicle or dreamsicle (formerly ice milk)"
+13140710,"LIGHT ICE CREAM, CREAMSICLE OR DREAMSICLE, NO SUGAR ADDED","Light ice cream, creamsicle or dreamsicle, no sugar added"
+13140900,"LIGHT ICE CREAM,FUDGESICLE (FORMERLY ICE MILK)","Light ice cream, fudgesicle (formerly ice milk)"
+13142000,"MILK DESSERT BAR/STICK, FROZEN, W/ COCONUT","Milk dessert bar or stick, frozen, with coconut"
+13150000,"SHERBET, ALL FLAVORS","Sherbet, all flavors"
+13160150,"FAT FREE ICE CREAM, NO SUGAR ADD, CHOC","Fat free ice cream, no sugar added, chocolate"
+13160160,"FAT FREE ICE CREAM, NO SUGAR ADD, FLAVORS OTHER THAN CHOC","Fat free ice cream, no sugar added, flavors other than chocolate"
+13160400,"FAT FREE ICE CREAM, FLAVORS OTHER THAN CHOC","Fat free ice cream, flavors other than chocolate"
+13160410,"FAT FREE ICE CREAM, CHOC","Fat free ice cream, chocolate"
+13160420,"FAT FREE ICE CREAM, NS AS TO FLAVOR","Fat free ice cream, NS as to flavor"
+13161000,"MILK DESSERT BAR, FROZEN, MADE FROM LOWFAT MILK","Milk dessert bar, frozen, made from lowfat milk"
+13161500,"MILK DESSERT SANDWICH BAR, FROZEN, DIETARY","Milk dessert sandwich bar, frozen, made from lowfat milk"
+13161520,"MILK DESSERT SANDWICH BAR,FRZ,W/LOW-CAL SWEET,LOFAT","Milk dessert sandwich bar, frozen, with low-calorie sweetener, made from lowfat milk"
+13161600,"MILK DES BAR, FROZEN, LOFAT MILK&LO CAL SWEETENER","Milk dessert bar, frozen, made from lowfat milk and low calorie sweetener"
+13161630,"LIGHT ICE CREAM,BAR/STICK, W/ LOW-CAL SWEETENER, CHOC COAT","Light ice cream, bar or stick, with low-calorie sweetener, chocolate-coated (formerly ice milk)"
+13170000,"BAKED ALASKA","Baked Alaska"
+13200110,"PUDDING, NFS","Pudding, NFS"
+13210110,"PUDDING, BREAD (INCLUDE W/ RAISINS)","Pudding, bread"
+13210150,"PUERTO RICAN BREAD PUDDING MADE W/ EVAP MILK & RUM","Puerto Rican bread pudding made with evaporated milk and rum (Budin de pan)"
+13210160,"DIPLOMAT PUDDING, P.R. (BUDIN DIPLOMATICO)","Diplomat pudding, Puerto Rican style (Budin Diplomatico)"
+13210180,"PUDDING, MEXICAN BREAD (CAPIROTADA)","Pudding, Mexican bread (Capirotada)"
+13210190,"PUDDING, MEXICAN BREAD (CAPIROTADA), LOWER FAT","Pudding, Mexican bread (Capirotada), lower fat"
+13210220,"PUDDING, CHOCOLATE, NS AS TO FROM DRY MIX/RTE","Pudding, chocolate, NS as to from dry mix or ready-to-eat"
+13210250,"PUDDING, CHOC, LO CAL, W/ART SWTNER, NS DRY/RTE","Pudding, chocolate, low calorie, containing artificial sweetener, NS as to from dry mix or ready-to-eat"
+13210260,"RICE FLOUR CREAM, P.R.STYLE (MANJAR BLANCO)","Rice flour cream, Puerto Rican style (manjar blanco)"
+13210270,"CUSTARD, P.R. (MAICENA, NATILLA)","Custard, Puerto Rican style (Maicena, Natilla)"
+13210280,"PUDDING, NOT CHOC, NS FROM DRY OR RTE","Pudding, flavors other than chocolate, NS as to from dry mix or ready-to-eat"
+13210290,"PUDDING, NOT CHOC, LO CAL, W/ART SWTNER, NS DRY MIX OR RTE","Pudding, flavors other than chocolate, low calorie, containing artificial sweetener, NS as to from dry mix or ready-to-eat"
+13210300,"CUSTARD","Custard"
+13210350,"FLAN","Flan"
+13210410,"PUDDING, RICE","Pudding, rice"
+13210450,"PUDDING, RICE FLOUR, W/ NUTS (INDIAN DESSERT)","Pudding, rice flour, with nuts (Indian dessert)"
+13210500,"PUDDING, TAPIOCA,MADE FROM HOME RECIPE, MADE W/ MILK","Pudding, tapioca, made from home recipe, made with milk"
+13210520,"PUDDING, TAPIOCA,MADE FROM DRY MIX,MADE W/ MILK","Pudding, tapioca, made from dry mix, made with milk"
+13210530,"PUDDING, TAPIOCA,CHOCOLATE,MADE W/ MILK","Pudding, tapioca, chocolate, made with milk"
+13210610,"PUDDING, COCONUT","Pudding, coconut"
+13210710,"PUDDING, INDIAN (MILK, MOLASSES, CORNMEAL-BASED)","Pudding, Indian (milk, molasses and cornmeal-based pudding)"
+13210750,"PUDDING, PUMPKIN","Pudding, pumpkin"
+13210810,"P.R. PUMPKIN PUDDING (FLAN DE CALABAZA)","Puerto Rican pumpkin pudding (Flan de calabaza)"
+13210820,"FRESH CORN CUSTARD, PUERTO RICAN STYLE","Fresh corn custard, Puerto Rican style (Mazamorra, Mundo Nuevo)"
+13220110,"PUDDING,NOT CHOCOLATE,PREPARED FROM DRY MIX,MILK ADDED","Pudding, flavors other than chocolate, prepared from dry mix, milk added"
+13220120,"PUDDING,CHOCOLATE,PREPARED FROM DRY MIX,MILK ADDED","Pudding, chocolate, prepared from dry mix, milk added"
+13220210,"PUDDING,NOT CHOC,FROM DRY,LOW CAL,ARTIFICIAL SWEET,W/MILK","Pudding, flavors other than chocolate, prepared from dry mix, low calorie, containing artificial sweetener, milk added"
+13220220,"PUDDING,CHOC,FROM DRY,LOW CAL,ARTIFICIAL SWEET,MILK ADDED","Pudding, chocolate, prepared from dry mix, low calorie, containing artificial sweetener, milk added"
+13220230,"PUDDING, RTE, CHOCOLATE, RED FAT","Pudding, ready-to-eat, chocolate, reduced fat"
+13220235,"PUDDING, RTE, CHOCOLATE, FAT FREE","Pudding, ready-to-eat, chocolate, fat free"
+13220240,"PUDDING, RTE, FLAVORS OTHER THAN CHOCOLATE, RED FAT","Pudding, ready-to-eat, flavors other than chocolate, reduced fat"
+13220245,"PUDDING, RTE, FLAVORS OTHER THAN CHOCOLATE, FAT FREE","Pudding, ready-to-eat, flavors other than chocolate, fat free"
+13230110,"PUDDING, RTE, FLAVORS OTHER THAN CHOCOLATE","Pudding, ready-to-eat, flavors other than chocolate"
+13230120,"PUDDING, RTE, LOW CAL, W/ARTIFICIAL SWTNR, NOT CHOC","Pudding, ready-to-eat, low calorie, containing artificial sweetener, flavors other than chocolate"
+13230130,"PUDDING, RTE, CHOCOLATE","Pudding, ready-to-eat, chocolate"
+13230140,"PUDDING,RTE, LO CAL/ W ART SWTNER, CHOC","Pudding, ready-to-eat, low calorie, containing artificial sweetener, chocolate"
+13230200,"PUDDING, RTE, CHOC  & NON-CHOC  FLAVORS COMBINED","Pudding, ready-to-eat, chocolate and non-chocolate flavors combined"
+13230500,"PUDDING, READY-TO-EAT, TAPIOCA","Pudding, ready-to-eat, tapioca"
+13230510,"PUDDING, READY-TO-EAT, TAPIOCA, FAT FREE","Pudding, ready-to-eat, tapioca, fat free"
+13241000,"PUDDING, W/ FRUIT & VANILLA WAFERS","Pudding, with fruit and vanilla wafers"
+13250000,"MOUSSE, CHOCOLATE","Mousse, chocolate"
+13250100,"MOUSSE, NOT CHOCOLATE","Mousse, not chocolate"
+13250200,"MOUSSE,CHOCOLATE,LOW FAT,REDUCED CAL,DRY MIX","Mousse, chocolate, lowfat, reduced calorie, prepared from dry mix, water added"
+13252100,"COCONUT CUSTARD, P.R. (FLAN DE COCO)","Coconut custard, Puerto Rican style (Flan de coco)"
+13252200,"MILK DESSERT OR MILK CANDY, P.R. (DULCE DE LECHE)","Milk dessert or milk candy, Puerto Rican style (Dulce de leche)"
+13252500,"BARFI/BURFI,INDIAN DESSERT,FROM MILK/CREAM/RICOTTA","Barfi or Burfi, Indian dessert, made from milk and/or cream and/or Ricotta cheese"
+13252600,"TIRAMISU","Tiramisu"
+13310000,"CUSTARD PUDDING, NOT CHOC, BABY, NS AS TO STR OR JR","Custard pudding, flavor other than chocolate, baby food, NS as to strained or junior"
+13311000,"CUSTARD PUDDING, BABY, NOT CHOCOLATE, STRAINED","Custard pudding, baby food, flavor other than chocolate, strained"
+13312000,"CUSTARD PUDDING, BABY, NOT CHOCOLATE, JUNIOR","Custard pudding, baby food, flavor other than chocolate, junior"
+13411000,"WHITE SAUCE, MILK SAUCE","White sauce, milk sauce"
+13412000,"MILK GRAVY, QUICK GRAVY","Milk gravy, quick gravy"
+14010000,"CHEESE, NFS","Cheese, NFS"
+14101010,"CHEESE, BLUE OR ROQUEFORT","Cheese, Blue or Roquefort"
+14102010,"CHEESE, BRICK","Cheese, Brick"
+14103010,"CHEESE, CAMEMBERT","Cheese, Camembert"
+14103020,"CHEESE, BRIE","Cheese, Brie"
+14104100,"CHEESE, CHEDDAR","Cheese, Cheddar"
+14104110,"CHEESE, CHEDDAR, REDUCED FAT","Cheese, Cheddar, reduced fat"
+14104115,"CHEESE, CHEDDAR, NONFAT OR FAT FREE","Cheese, Cheddar, nonfat or fat free"
+14104200,"CHEESE, COLBY","Cheese, Colby"
+14104250,"CHEESE, COLBY JACK","Cheese, Colby Jack"
+14104400,"CHEESE, FETA (INCLUDE GOAT CHEESE)","Cheese, Feta"
+14104600,"CHEESE, FONTINA","Cheese, Fontina"
+14104700,"CHEESE, GOAT","Cheese, goat"
+14105010,"CHEESE, GOUDA OR EDAM","Cheese, Gouda or Edam"
+14105200,"CHEESE, GRUYERE","Cheese, Gruyere"
+14106010,"CHEESE, LIMBURGER","Cheese, Limburger"
+14106200,"CHEESE, MONTEREY","Cheese, Monterey"
+14106500,"CHEESE, MONTEREY, REDUCED FAT","Cheese, Monterey, reduced fat"
+14107010,"CHEESE, MOZZARELLA, NFS (INCLUDE PIZZA CHEESE)","Cheese, Mozzarella, NFS"
+14107020,"CHEESE, MOZZARELLA, WHOLE MILK","Cheese, Mozzarella, whole milk"
+14107030,"CHEESE, MOZZARELLA, PART SKIM (INCL ""LOWFAT"")","Cheese, Mozzarella, part skim"
+14107040,"CHEESE, MOZZARELLA, REDUCED SODIUM","Cheese, Mozzarella, reduced sodium"
+14107060,"CHEESE, MOZZARELLA, NONFAT OR FAT FREE","Cheese, Mozzarella, nonfat or fat free"
+14107200,"CHEESE, MUENSTER","Cheese, Muenster"
+14107250,"CHEESE, MUENSTER, REDUCED FAT","Cheese, Muenster, reduced fat"
+14108010,"CHEESE, PARMESAN, DRY, GRATED (INCLUDE ROMANO)","Cheese, Parmesan, dry grated"
+14108015,"CHEESE, PARMESAN, DRY GRATED, REDUCED FAT","Cheese, Parmesan, dry grated, reduced fat"
+14108020,"CHEESE, PARMESAN, HARD (INCLUDE ROMANO)","Cheese, Parmesan, hard"
+14108060,"CHEESE, PARMESAN, DRY GRATED, FAT FREE","Cheese, Parmesan, dry grated, fat free"
+14108200,"CHEESE, PORT DU SALUT","Cheese, Port du Salut"
+14108400,"CHEESE, PROVOLONE","Cheese, Provolone"
+14108420,"CHEESE, PROVOLONE, REDUCED FAT","Cheese, provolone, reduced fat"
+14109010,"CHEESE, SWISS","Cheese, Swiss"
+14109020,"CHEESE, SWISS, REDUCED SODIUM","Cheese, Swiss, reduced sodium"
+14109030,"CHEESE, SWISS, REDUCED FAT","Cheese, Swiss, reduced fat"
+14109040,"CHEESE, SWISS, NONFAT OR FAT FREE","Cheese, Swiss, nonfat or fat free"
+14110010,"CHEESE, CHEDDAR, REDUCED SODIUM","Cheese, Cheddar, reduced sodium"
+14120010,"CHEESE, MEXICAN BLEND","Cheese, Mexican blend"
+14120020,"CHEESE, MEXICAN BLEND, REDUCED FAT","Cheese, Mexican blend, reduced fat"
+14131000,"QUESO ANEJO (AGED MEXICAN CHEESE)","Queso Anejo (aged Mexican cheese)"
+14131500,"QUESO ASADERO (INCL OAXACAN-STYLE STRING CHEESE)","Queso Asadero"
+14132000,"QUESO CHIHUAHUA (INCL MENNONITE CHEESE)","Queso Chihuahua"
+14133000,"QUESO FRESCO (HISPANIC-STYLE FARMER CHEESE)","Queso Fresco"
+14134000,"QUESO COTIJA","Queso cotija"
+14200100,"CHEESE, COTTAGE, NFS","Cheese, cottage, NFS"
+14201010,"CHEESE, COTTAGE, CREAMED","Cheese, cottage, creamed, large or small curd"
+14201200,"COTTAGE CHEESE, FARMER'S","Cottage cheese, farmer's"
+14201500,"CHEESE, RICOTTA","Cheese, Ricotta"
+14202010,"CHEESE, COTTAGE, W/ FRUIT","Cheese, cottage, with fruit"
+14202020,"CHEESE, COTTAGE, W/ VEGETABLES","Cheese, cottage, with vegetables"
+14203010,"CHEESE, COTTAGE, DRY CURD","Cheese, cottage, dry curd"
+14203020,"CHEESE, COTTAGE, SALTED, DRY CURD","Cheese, cottage, salted, dry curd"
+14203510,"P.R. WHITE CHEESE (QUESO DEL PAIS, BLANCO)","Puerto Rican white cheese (queso del pais, blanco)"
+14204010,"CHEESE, COTTAGE, LOWFAT","Cheese, cottage, lowfat (1-2% fat)"
+14204020,"CHEESE, COTTAGE, LOWFAT, W/ FRUIT","Cheese, cottage, lowfat, with fruit"
+14204030,"CHEESE, COTTAGE, LOWFAT, W/ VEGETABLES","Cheese, cottage, lowfat, with vegetables"
+14206010,"CHEESE, COTTAGE, LOWFAT, LOW SODIUM","Cheese, cottage, lowfat, low sodium"
+14207010,"CHEESE, COTTAGE, LOWFAT, LACTOSE REDUCED","Cheese, cottage, lowfat, lactose reduced"
+14301010,"CHEESE, CREAM","Cheese, cream"
+14303010,"CHEESE, CREAM, LIGHT/LITE (FORMERLY CALLED CR CHEESE LOWFAT)","Cheese, cream, light or lite (formerly called Cream Cheese Lowfat)"
+14410100,"CHEESE, AMERICAN AND SWISS BLENDS","Cheese, American and Swiss blends"
+14410110,"CHEESE, AMERICAN","Cheese, American"
+14410120,"CHEESE, AMERICAN, REDUCED FAT","Cheese, American, reduced fat"
+14410130,"CHEESE, AMERICAN, NONFAT OR FAT FREE","Cheese, American, nonfat or fat free"
+14410210,"CHEESE, AMERICAN, REDUCED SODIUM","Cheese, American, reduced sodium"
+14410330,"CHEESE SPREAD, AMERICAN OR CHEDDAR CHEESE BASE, REDUCED FAT","Cheese spread, American or Cheddar cheese base, reduced fat"
+14410380,"CHEESE, PROCESSED CREAM CHEESE PRODUCT, NONFAT","Cheese, processed cream cheese product, nonfat or fat free"
+14410500,"CHEESE, PROCESSED, CHEESE FOOD","Cheese, processed cheese food"
+14410600,"CHEESE, PROCESSED, W/VEGETABLES(INCL PEPPER CHEESE)","Cheese, processed, with vegetables"
+14410620,"CHEESE,  WITH WINE","Cheese,  with wine"
+14420100,"CHEESE SPREAD, AMERICAN OR CHEDDAR CHEESE BASE","Cheese spread, American or Cheddar cheese base"
+14420160,"CHEESE SPREAD, SWISS CHEESE BASE","Cheese spread, Swiss cheese base"
+14420200,"CHEESE SPRD, CREAM CHEESE, REG","Cheese spread, cream cheese, regular"
+14420210,"CHEESE SPREAD, CREAM CHEESE, LIGHT OR LITE","Cheese spread, cream cheese, light or lite"
+14420300,"CHEESE SPREAD, PRESSURIZED CAN","Cheese spread, pressurized can"
+14502000,"IMITATION CHEESE","Imitation cheese"
+14610200,"COTTAGE CHEESE, W/ GELATIN DESSERT","Cheese, cottage cheese, with gelatin dessert"
+14610210,"COTTAGE CHEESE, W/ GELATIN DESSERT & FRUIT","Cheese, cottage cheese, with gelatin dessert and fruit"
+14610250,"COTTAGE CHEESE W/ GELATIN DESSERT & VEGETABLES","Cheese, cottage cheese, with gelatin dessert and vegetables"
+14610520,"CHEESE W/ NUTS (INCL CHEESE BALL)","Cheese with nuts"
+14620100,"DIP, CREAM CHEESE BASE","Dip, cream cheese base"
+14620120,"SHRIMP DIP, CREAM CHEESE BASE (INCL CLAM DIP)","Shrimp dip, cream cheese base"
+14620150,"DIP, CHEESE W/ CHILI PEPPER (CHILI CON QUESO)","Dip, cheese with chili pepper (chili con queso)"
+14620200,"DIP, CHEESE BASE OTHER THAN CREAM CHEESE","Dip, cheese base other than cream cheese"
+14620300,"TOPPING FROM CHEESE PIZZA","Topping from cheese pizza"
+14620310,"TOPPING FROM VEGETABLE PIZZA","Topping from vegetable pizza"
+14620320,"TOPPING FROM MEAT PIZZA","Topping from meat pizza"
+14620330,"TOPPING FROM MEAT AND VEGETABLE PIZZA","Topping from meat and vegetable pizza"
+14630100,"CHEESE FONDUE","Cheese fondue"
+14630200,"CHEESE SOUFFLE","Cheese souffle"
+14630300,"WELSH RAREBIT","Welsh rarebit"
+14640000,"CHEESE SANDWICH","Cheese sandwich"
+14640100,"CHEESE SANDWICH, GRILLED","Cheese sandwich, grilled"
+14640200,"CHEESE SANDWICH, HOAGIE","Cheese sandwich, hoagie"
+14650100,"CHEESE SAUCE","Cheese sauce"
+14650150,"CHEESE SAUCE MADE W/ LOWFAT CHEESE","Cheese sauce made with lowfat cheese"
+14650160,"ALFREDO SAUCE","Alfredo sauce"
+14660200,"CHEESE, NUGGETS, FRIED (INCL BANQUET BRAND)","Cheese, nuggets or pieces, breaded, baked, or fried"
+14670000,"MOZZARELLA CHEESE, TOMATO, BASIL, W/ OIL, VINEGAR","Mozzarella cheese, tomato, and basil, with oil and vinegar dressing"
+14710100,"CHEDDAR CHEESE SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Cheddar cheese soup, home recipe, canned or ready-to-serve"
+14710200,"BEER CHEESE SOUP, MADE WITH MILK","Beer cheese soup, made with milk"
+2e+07,"MEAT, NFS","Meat, NFS"
+20000070,"MEAT, BABY, NS AS TO TYPE, NS AS TO STR OR JR","Meat, baby food, NS as to type, NS as to strained or junior"
+20000090,"MEAT STICKS, BABY, NS AS TO TYPE OF MEAT","Meat sticks, baby food, NS as to type of meat"
+20000200,"GROUND MEAT,NFS","Ground meat, NFS"
+21000100,"BEEF, NS AS TO CUT, COOKED, NS AS TO FAT","Beef, NS as to cut, cooked, NS as to fat eaten"
+21000110,"BEEF, NS AS TO CUT, COOKED, LEAN & FAT","Beef, NS as to cut, cooked, lean and fat eaten"
+21000120,"BEEF, NS AS TO CUT, COOKED, LEAN ONLY","Beef, NS as to cut, cooked, lean only eaten"
+21001000,"STEAK, NS AS TO TYPE OF MEAT, COOKED, NS AS TO FAT","Steak, NS as to type of meat, cooked, NS as to fat eaten"
+21001010,"STEAK, NS AS TO TYPE OF MEAT, COOKED, LEAN & FAT","Steak, NS as to type of meat, cooked, lean and fat eaten"
+21001020,"STEAK, NS AS TO TYPE OF MEAT, COOKED, LEAN ONLY","Steak, NS as to type of meat, cooked, lean only eaten"
+21002000,"BEEF, PICKLED","Beef, pickled"
+21003000,"BEEF, NS AS TO CUT, FRIED, NS AS TO FAT EATEN","Beef, NS as to cut, fried, NS to fat eaten"
+21101000,"BEEF STEAK, NS AS TO COOKING METHOD, NS AS TO FAT","Beef steak, NS as to cooking method, NS as to fat eaten"
+21101010,"BEEF STEAK, NS AS TO COOKING METHOD, LEAN & FAT","Beef steak, NS as to cooking method, lean and fat eaten"
+21101020,"BEEF STEAK, NS AS TO COOKING METHOD, LEAN ONLY","Beef steak, NS as to cooking method, lean only eaten"
+21101110,"BEEF STEAK, BROILED OR BAKED, NS AS TO FAT","Beef steak, broiled or baked, NS as to fat eaten"
+21101120,"BEEF STEAK, BROILED OR BAKED, LEAN & FAT","Beef steak, broiled or baked, lean and fat eaten"
+21101130,"BEEF STEAK, BROILED OR BAKED, LEAN ONLY","Beef steak, broiled or baked, lean only eaten"
+21102110,"BEEF STEAK, FRIED, NS AS TO FAT","Beef steak, fried, NS as to fat eaten"
+21102120,"BEEF STEAK, FRIED, LEAN & FAT","Beef steak, fried, lean and fat eaten"
+21102130,"BEEF STEAK, FRIED, LEAN ONLY","Beef steak, fried, lean only eaten"
+21103110,"BEEF STEAK,BREADED/FLOURED,BAKED/FRIED,NS AS TO FAT","Beef steak, breaded or floured, baked or fried, NS as to fat eaten"
+21103120,"BEEF STEAK, BREADED/FLOURED,BAKED/FRIED, LEAN & FAT","Beef steak, breaded or floured, baked or fried, lean and fat eaten"
+21103130,"BEEF STEAK, BREADED/FLOURED, BAKED/FRIED, LEAN ONLY","Beef steak, breaded or floured, baked or fried, lean only eaten"
+21104110,"BEEF STEAK, BATTERED, FRIED, NS AS TO FAT","Beef steak, battered, fried, NS as to fat eaten"
+21104120,"BEEF STEAK, BATTERED, FRIED, LEAN & FAT","Beef steak, battered, fried, lean and fat eaten"
+21104130,"BEEF STEAK, BATTERED, FRIED, LEAN ONLY","Beef steak, battered, fried, lean only eaten"
+21105110,"BEEF STEAK, BRAISED, NS AS TO FAT","Beef steak, braised, NS as to fat eaten"
+21105120,"BEEF STEAK, BRAISED, LEAN & FAT","Beef steak, braised, lean and fat eaten"
+21105130,"BEEF STEAK, BRAISED, LEAN ONLY","Beef steak, braised, lean only eaten"
+21301000,"BEEF, OXTAILS, COOKED","Beef, oxtails, cooked"
+21302000,"BEEF, NECK BONES, COOKED","Beef, neck bones, cooked"
+21304000,"BEEF, SHORTRIBS, COOKED, NS AS TO FAT","Beef, shortribs, cooked, NS as to fat eaten"
+21304110,"BEEF, SHORTRIBS, COOKED, LEAN & FAT","Beef, shortribs, cooked, lean and fat eaten"
+21304120,"BEEF, SHORTRIBS, COOKED, LEAN ONLY","Beef, shortribs, cooked, lean only eaten"
+21304200,"BEEF, SHORTRIBS, BBQ, W/ SAUCE, NS AS TO FAT","Beef, shortribs, barbecued, with sauce, NS as to fat eaten"
+21304210,"BEEF, SHORTRIBS, BBQ, W/ SAUCE, LEAN & FAT","Beef, shortribs, barbecued, with sauce, lean and fat eaten"
+21304220,"BEEF, SHORTRIBS, BBQ, W/ SAUCE, LEAN ONLY","Beef, shortribs, barbecued, with sauce, lean only eaten"
+21305000,"BEEF, COW HEAD, COOKED","Beef, cow head, cooked"
+21401000,"BEEF, ROAST, ROASTED, NS AS TO FAT","Beef, roast, roasted, NS as to fat eaten"
+21401110,"BEEF, ROAST, ROASTED, LEAN & FAT","Beef, roast, roasted, lean and fat eaten"
+21401120,"BEEF, ROAST, ROASTED, LEAN ONLY","Beef, roast, roasted, lean only eaten"
+21401400,"BEEF, ROAST, CANNED","Beef, roast, canned"
+21407000,"BEEF, POT ROAST, BRAISED OR BOILED, NS AS TO FAT","Beef, pot roast, braised or boiled, NS as to fat eaten"
+21407110,"BEEF, POT ROAST, BRAISED OR BOILED, LEAN & FAT","Beef, pot roast, braised or boiled, lean and fat eaten"
+21407120,"BEEF, POT ROAST, BRAISED OR BOILED, LEAN ONLY","Beef, pot roast, braised or boiled, lean only eaten"
+21410000,"BEEF, STEW MEAT, COOKED, NS AS TO FAT","Beef, stew meat, cooked, NS as to fat eaten"
+21410110,"BEEF, STEW MEAT, COOKED, LEAN & FAT","Beef, stew meat, cooked, lean and fat eaten"
+21410120,"BEEF, STEW MEAT, COOKED, LEAN ONLY","Beef, stew meat, cooked, lean only eaten"
+21416000,"CORNED BEEF, COOKED, NS AS TO FAT","Corned beef, cooked, NS as to fat eaten"
+21416110,"CORNED BEEF, COOKED, LEAN & FAT","Corned beef, cooked, lean and fat eaten"
+21416120,"CORNED BEEF, COOKED, LEAN ONLY","Corned beef, cooked, lean only eaten"
+21416150,"CORNED BEEF, CANNED, READY TO EAT","Corned beef, canned, ready-to-eat"
+21417100,"BEEF BRISKET, COOKED, NS AS TO FAT","Beef brisket, cooked, NS as to fat eaten"
+21417110,"BEEF BRISKET, COOKED, LEAN & FAT","Beef brisket, cooked, lean and fat eaten"
+21417120,"BEEF BRISKET, COOKED, LEAN ONLY","Beef brisket, cooked, lean only eaten"
+21420100,"BEEF, SANDWICH STEAK (FLAKED,FORMED, THINLY SLICED)","Beef, sandwich steak (flaked, formed, thinly sliced)"
+21500000,"GROUND BEEF, RAW","Ground beef, raw"
+21500100,"GROUND BEEF OR PATTY, NS AS TO %LEAN","Ground beef or patty, cooked, NS as to percent lean (formerly NS as to regular, lean, or extra lean)"
+21500200,"GROUND BEEF OR PATTY, BREADED, COOKED","Ground beef or patty, breaded, cooked"
+21500300,"GROUND BEEF PATTY, COOKED (FOR FAST FOOD SANDWICHES)","Ground beef patty, cooked (for fast food sandwiches)"
+21501000,"GROUND BEEF, LESS THAN 80% LEAN, COOKED","Ground beef, less than 80% lean, cooked (formerly regular)"
+21501200,"GROUND BEEF, 80% - 84% LEAN, COOKED","Ground beef, 80% - 84% lean, cooked (formerly lean)"
+21501300,"GROUND BEEF, 85% - 89% LEAN, COOKED","Ground beef, 85% - 89% lean, cooked (formerly extra lean)"
+21501350,"GROUND BEEF, 90% - 94% LEAN, COOKED","Ground beef, 90% - 94% lean, cooked"
+21501360,"GROUND BEEF, 95% OR MORE LEAN, COOKED","Ground beef, 95% or more lean, cooked"
+21540100,"GROUND BEEF W/ TEXTURED VEGETABLE PROTEIN, COOKED","Ground beef with textured vegetable protein, cooked"
+21601000,"BEEF, BACON, COOKED","Beef, bacon, cooked"
+21601500,"BEEF BACON, FORMED, LEAN MEAT ADDED (INCL SIZZLEAN)","Beef, bacon, formed, lean meat added, cooked"
+21602000,"BEEF, DRIED, CHIPPED, UNCOOKED","Beef, dried, chipped, uncooked"
+21602010,"BEEF, DRIED, CHIPPED, COOKED IN FAT","Beef, dried, chipped, cooked in fat"
+21602100,"BEEF JERKY","Beef jerky"
+21603000,"BEEF, PASTRAMI (BEEF, SMOKED, SPICED)","Beef, pastrami (beef, smoked, spiced)"
+21701000,"BEEF, BABY, NS AS TO STRAINED OR JUNIOR","Beef, baby food, NS as to strained or junior"
+21701010,"BEEF, BABY, STRAINED","Beef, baby food, strained"
+21701020,"BEEF, BABY, JUNIOR","Beef, baby food, junior"
+22000100,"PORK, NS AS TO CUT, COOKED, NS AS TO FAT EATEN","Pork, NS as to cut, cooked, NS as to fat eaten"
+22000110,"PORK, NS AS TO CUT, COOKED, LEAN & FAT EATEN","Pork, NS as to cut, cooked, lean and fat eaten"
+22000120,"PORK, NS AS TO CUT, COOKED, LEAN ONLY EATEN","Pork, NS as to cut, cooked, lean only eaten"
+22000200,"PORK, NS AS TO CUT, FRIED, NS AS TO FAT EATEN","Pork, NS as to cut, fried, NS as to fat eaten"
+22000210,"PORK, NS AS TO CUT, FRIED, LEAN & FAT EATEN","Pork, NS as to cut, fried, lean and fat eaten"
+22000220,"PORK, NS AS TO CUT, FRIED, LEAN ONLY EATEN","Pork, NS as to cut, fried, lean only eaten"
+22000300,"PORK, NS AS TO CUT, BREADED, FRIED, NS AS TO FAT","Pork, NS as to cut, breaded or floured, fried, NS as to fat eaten"
+22000310,"PORK, NS AS TO CUT, BREADED, FRIED, FAT EATEN","Pork, NS as to cut, breaded or floured, fried, lean and fat eaten"
+22000320,"PORK, NS AS TO CUT, BREADED, FRIED, LEAN ONLY","Pork, NS as to cut, breaded or floured, fried, lean only eaten"
+22001000,"PORK, PICKLED, NS AS TO CUT","Pork, pickled, NS as to cut"
+22002000,"PORK, GROUND OR PATTY, COOKED","Pork, ground or patty, cooked"
+22002100,"PORK, GROUND, GROUND OR PATTY, BREADED, COOKED","Pork, ground or patty, breaded, cooked"
+22002800,"PORK JERKY","Pork jerky"
+22101000,"PORK CHOP, NS AS TO COOKING METHOD, NS AS TO FAT","Pork chop, NS as to cooking method, NS as to fat eaten"
+22101010,"PORK CHOP, NS AS TO COOKING METHOD, LEAN & FAT","Pork chop, NS as to cooking method, lean and fat eaten"
+22101020,"PORK CHOP, NS AS TO COOKING METHOD, LEAN ONLY","Pork chop, NS as to cooking method, lean only eaten"
+22101100,"PORK CHOP, BROILED OR BAKED, NS AS TO FAT","Pork chop, broiled or baked, NS as to fat eaten"
+22101110,"PORK CHOP, BROILED OR BAKED, LEAN & FAT","Pork chop, broiled or baked, lean and fat eaten"
+22101120,"PORK CHOP, BROILED OR BAKED, LEAN ONLY","Pork chop, broiled or baked, lean only eaten"
+22101130,"PORK CHOP, BREADED, BROILED OR BAKED, NS AS TO FAT","Pork chop, breaded or floured, broiled or baked, NS as to fat eaten"
+22101140,"PORK CHOP, BREADED, BROILED OR BAKED, LEAN & FAT","Pork chop, breaded or floured, broiled or baked, lean and fat eaten"
+22101150,"PORK CHOP, BREADED, BROILED OR BAKED, LEAN ONLY","Pork chop, breaded or floured, broiled or baked, lean only eaten"
+22101200,"PORK CHOP, FRIED, NS AS TO FAT","Pork chop, fried, NS as to fat eaten"
+22101210,"PORK CHOP, FRIED, LEAN & FAT","Pork chop, fried, lean and fat eaten"
+22101220,"PORK CHOP, FRIED, LEAN ONLY","Pork chop, fried, lean only eaten"
+22101300,"PORK CHOP, BREADED, FRIED, NS AS TO FAT","Pork chop, breaded or floured, fried, NS as to fat eaten"
+22101310,"PORK CHOP, BREADED, FRIED, LEAN & FAT","Pork chop, breaded or floured, fried, lean and fat eaten"
+22101320,"PORK CHOP, BREADED, FRIED, LEAN ONLY","Pork chop, breaded or floured, fried, lean only eaten"
+22101400,"PORK CHOP, BATTERED, FRIED, NS AS TO FAT","Pork chop, battered, fried, NS as to fat eaten"
+22101410,"PORK CHOP, BATTERED, FRIED, LEAN & FAT","Pork chop, battered, fried, lean and fat eaten"
+22101420,"PORK CHOP, BATTERED, FRIED, LEAN ONLY","Pork chop, battered, fried, lean only eaten"
+22101500,"PORK CHOP, STEWED, NS AS TO FAT EATEN","Pork chop, stewed, NS as to fat eaten"
+22101510,"PORK CHOP, STEWED, LEAN & FAT EATEN","Pork chop, stewed, lean and fat eaten"
+22101520,"PORK CHOP, STEWED, LEAN ONLY EATEN","Pork chop, stewed, lean only eaten"
+22107000,"PORK CHOP, SMOKED OR CURED, COOKED, NS AS TO FAT","Pork chop, smoked or cured, cooked, NS as to fat eaten"
+22107010,"PORK CHOP, SMOKED OR CURED, COOKED, LEAN & FAT","Pork chop, smoked or cured, cooked, lean and fat eaten"
+22107020,"PORK CHOP, SMOKED OR CURED, COOKED, LEAN ONLY","Pork chop, smoked or cured, cooked, lean only eaten"
+22201000,"PORK STEAK, NS AS TO COOKING METHOD, NS AS TO FAT","Pork steak or cutlet, NS as to cooking method, NS as to fat eaten"
+22201010,"PORK STEAK, NS AS TO COOKING METHOD, LEAN & FAT","Pork steak or cutlet, NS as to cooking method, lean and fat eaten"
+22201020,"PORK STEAK, NS AS TO COOKING METHOD, LEAN ONLY","Pork steak or cutlet, NS as to cooking method, lean only eaten"
+22201050,"PORK STEAK OR CUTLET, BATTERED, FRIED, NS AS TO FAT","Pork steak or cutlet, battered, fried, NS as to fat eaten"
+22201060,"PORK STEAK OR CUTLET, BATTERED, FRIED, LEAN & FAT","Pork steak or cutlet, battered, fried, lean and fat eaten"
+22201070,"PORK STEAK OR CUTLET, BATTERED, FRIED, LEAN ONLY","Pork steak or cutlet, battered, fried, lean only eaten"
+22201100,"PORK STEAK OR CUTLET, BROILED OR BAKD, NS AS TO FAT","Pork steak or cutlet, broiled or baked, NS as to fat eaten"
+22201110,"PORK STEAK OR CUTLET, BROILED OR BAKED, LEAN & FAT","Pork steak or cutlet, broiled or baked, lean and fat eaten"
+22201120,"PORK STEAK OR CUTLET, BROILED OR BAKED, LEAN ONLY","Pork steak or cutlet, broiled or baked, lean only eaten"
+22201200,"PORK STEAK OR CUTLET, FRIED, NS AS TO FAT","Pork steak or cutlet, fried, NS as to fat eaten"
+22201210,"PORK STEAK OR CUTLET, FRIED, LEAN & FAT","Pork steak or cutlet, fried, lean and fat eaten"
+22201220,"PORK STEAK OR CUTLET, FRIED, LEAN ONLY","Pork steak or cutlet, fried, lean only eaten"
+22201300,"PORK CUTLET, BREADED, BROILED/BAKED, NS AS TO FAT","Pork steak or cutlet, breaded or floured, broiled or baked, NS as to fat eaten"
+22201310,"PORK CUTLET, BREADED, BROILED/BAKED, LEAN & FAT","Pork steak or cutlet, breaded or floured, broiled or baked, lean and fat eaten"
+22201320,"PORK CUTLET, BREADED, BROILED/BAKED, LEAN ONLY","Pork steak or cutlet, breaded or floured, broiled or baked, lean only eaten"
+22201400,"PORK STEAK OR CUTLET, BREADED, FRIED, NS AS TO FAT","Pork steak or cutlet, breaded or floured, fried, NS as to fat eaten"
+22201410,"PORK STEAK OR CUTLET, BREADED, FRIED, LEAN & FAT","Pork steak or cutlet, breaded or floured, fried, lean and fat eaten"
+22201420,"PORK STEAK OR CUTLET, BREADED, FRIED, LEAN ONLY","Pork steak or cutlet, breaded or floured, fried, lean only eaten"
+22210300,"PORK, TENDERLOIN, COOKED, NS AS TO METHOD","Pork, tenderloin, cooked, NS as to cooking method"
+22210310,"PORK, TENDERLOIN, BREADED, FRIED","Pork, tenderloin, breaded, fried"
+22210350,"PORK, TENDERLOIN, BRAISED","Pork, tenderloin, braised"
+22210400,"PORK, TENDERLOIN, BAKED","Pork, tenderloin, baked"
+22210450,"PORK, TENDERLOIN, BATTERED, FRIED","Pork, tenderloin, battered, fried"
+22300120,"HAM, FRIED, NS AS TO FAT","Ham, fried, NS as to fat eaten"
+22300130,"HAM, FRIED, LEAN & FAT","Ham, fried, lean and fat eaten"
+22300140,"HAM, FRIED, LEAN ONLY","Ham, fried, lean only eaten"
+22300150,"HAM, BREADED, FRIED, NS AS TO FAT","Ham, breaded or floured, fried, NS as to fat eaten"
+22300160,"HAM, BREADED, FRIED, LEAN & FAT","Ham, breaded or floured, fried, lean and fat eaten"
+22300170,"HAM, BREADED, FRIED, LEAN ONLY","Ham, breaded or floured, fried, lean only eaten"
+22301000,"HAM, FRESH, COOKED, NS AS TO FAT","Ham, fresh, cooked, NS as to fat eaten"
+22301110,"HAM, FRESH, COOKED, LEAN & FAT","Ham, fresh, cooked, lean and fat eaten"
+22301120,"HAM, FRESH, COOKED, LEAN ONLY","Ham, fresh, cooked, lean only eaten"
+22311000,"HAM, SMOKED OR CURED, COOKED, NS AS TO FAT","Ham, smoked or cured, cooked, NS as to fat eaten"
+22311010,"HAM, SMOKED OR CURED, COOKED, LEAN & FAT","Ham, smoked or cured, cooked, lean and fat eaten"
+22311020,"HAM, SMOKED OR CURED, COOKED, LEAN ONLY","Ham, smoked or cured, cooked, lean only eaten"
+22311200,"HAM, SMOKED OR CURED, LOW NA, NS AS TO FAT","Ham, smoked or cured, low sodium, cooked, NS as to fat eaten"
+22311210,"HAM, SMOKED OR CURED, LOW NA, LEAN & FAT","Ham, smoked or cured, low sodium, cooked, lean and fat eaten"
+22311220,"HAM, SMOKED OR CURED, LOW NA, LEAN ONLY","Ham, smoked or cured, low sodium, cooked, lean only eaten"
+22311450,"HAM, PROSCIUTTO","Ham, prosciutto"
+22311500,"HAM, SMOKED OR CURED, CANNED, NS AS TO FAT EATEN","Ham, smoked or cured, canned, NS as to fat eaten"
+22311510,"HAM, SMOKED OR CURED, CANNED, LEAN & FAT EATEN","Ham, smoked or cured, canned, lean and fat eaten"
+22311520,"HAM, SMOKED OR CURED, CANNED, LEAN ONLY EATEN","Ham, smoked or cured, canned, lean only eaten"
+22321110,"HAM, SMOKED OR CURED, GROUND PATTY","Ham, smoked or cured, ground patty"
+22400100,"PORK ROAST, NS AS TO CUT, NS AS TO FAT","Pork roast, NS as to cut, cooked, NS as to fat eaten"
+22400110,"PORK ROAST, NS AS TO CUT, COOKED, LEAN & FAT","Pork roast, NS as to cut, cooked, lean and fat eaten"
+22400120,"PORK ROAST, NS AS TO CUT, COOKED, LEAN ONLY","Pork roast, NS as to cut, cooked, lean only eaten"
+22401000,"PORK ROAST, LOIN, COOKED, NS AS TO FAT","Pork roast, loin, cooked, NS as to fat eaten"
+22401010,"PORK ROAST, LOIN, COOKED, LEAN & FAT","Pork roast, loin, cooked, lean and fat eaten"
+22401020,"PORK ROAST, LOIN, COOKED, LEAN ONLY","Pork roast, loin, cooked, lean only eaten"
+22402510,"FRIED PORK CHUNKS, P.R. (CARNE DE CERDO FRITA)","Fried pork chunks, Puerto Rican style (Carne de cerdo frita, masitas fritas)"
+22411000,"PORK ROAST, SHOULDER, COOKED, NS AS TO FAT","Pork roast, shoulder, cooked, NS as to fat eaten"
+22411010,"PORK ROAST, SHOULDER, COOKED, LEAN & FAT","Pork roast, shoulder, cooked, lean and fat eaten"
+22411020,"PORK ROAST, SHOULDER, COOKED, LEAN ONLY","Pork roast, shoulder, cooked, lean only eaten"
+22421000,"PORK ROAST, SMOKED OR CURED, COOKED, NS AS TO FAT","Pork roast, smoked or cured, cooked, NS as to fat eaten"
+22421010,"PORK ROAST, SMOKED OR CURED, COOKED, LEAN & FAT","Pork roast, smoked or cured, cooked, lean and fat eaten"
+22421020,"PORK ROAST, SMOKED OR CURED, COOKED, LEAN ONLY","Pork roast, smoked or cured, cooked, lean only eaten"
+22431000,"PORK ROLL, CURED, FRIED","Pork roll, cured, fried"
+22501010,"BACON, CANADIAN, COOKED","Canadian bacon, cooked"
+22600100,"BACON, NS AS TO TYPE OF MEAT, COOKED","Bacon, NS as to type of meat, cooked"
+22600200,"PORK BACON, NS AS TO FRESH/SMOKED/CURED, COOKED","Pork bacon, NS as to fresh, smoked or cured, cooked"
+22601000,"PORK BACON, SMOKED OR CURED, COOKED","Pork bacon, smoked or cured, cooked"
+22601040,"BACON OR SIDE PORK, FRESH, COOKED","Bacon or side pork, fresh, cooked"
+22602010,"PORK BACON, SMOKED OR CURED, LOWER SODIUM","Pork bacon, smoked or cured, lower sodium"
+22605010,"BACON, FORMED, LEAN MEAT ADDED, COOKED","Pork bacon, formed, lean meat added, cooked"
+22621000,"SALT PORK, COOKED","Salt pork, cooked"
+22621100,"FAT BACK, COOKED (INCLUDE HOG JOWL)","Fat back, cooked"
+22701000,"PORK, SPARERIBS, COOKED, NS AS TO FAT EATEN","Pork, spareribs, cooked, NS as to fat eaten"
+22701010,"PORK, SPARERIBS, COOKED, LEAN & FAT","Pork, spareribs, cooked, lean and fat eaten"
+22701020,"PORK, SPARERIBS, COOKED, LEAN ONLY","Pork, spareribs, cooked, lean only eaten"
+22701030,"PORK, SPARERIBS, BBQ, W/ SAUCE, NS FAT EATEN","Pork, spareribs, barbecued, with sauce, NS as to fat eaten"
+22701040,"PORK, SPARERIBS, BBQ, W/ SAUCE, LEAN & FAT EATEN","Pork, spareribs, barbecued, with sauce, lean and fat eaten"
+22701050,"PORK, SPARERIBS, BBQ, W/ SAUCE, LEAN ONLY EATEN","Pork, spareribs, barbecued, with sauce, lean only eaten"
+22704010,"PORK, CRACKLINGS, COOKED","Pork, cracklings, cooked"
+22705010,"PORK, EARS, TAIL, HEAD, SNOUT, MISC PARTS, COOKED","Pork ears, tail, head, snout, miscellaneous parts, cooked"
+22706010,"PORK, NECK BONES, COOKED","Pork, neck bones, cooked"
+22707010,"PORK, PIG'S FEET, COOKED","Pork, pig's feet, cooked"
+22707020,"PORK, PIG'S FEET, PICKLED","Pork, pig's feet, pickled"
+22708010,"PORK, PIG'S HOCKS, COOKED","Pork, pig's hocks, cooked"
+22709010,"PORK SKIN, RINDS, DEEP-FRIED","Pork skin, rinds, deep-fried"
+22709110,"PORK SKIN, BOILED","Pork skin, boiled"
+22810010,"HAM, BABY, STRAINED","Ham, baby food, strained"
+22820000,"MEAT STICK, BABY FOOD","Meat stick, baby food"
+23000100,"LAMB, NS AS TO CUT, COOKED","Lamb, NS as to cut, cooked"
+23101000,"LAMB CHOP, COOKED, NS AS TO CUT & FAT","Lamb chop, NS as to cut, cooked, NS as to fat eaten"
+23101010,"LAMB CHOP, NS AS TO CUT, COOKED, LEAN & FAT","Lamb chop, NS as to cut, cooked, lean and fat eaten"
+23101020,"LAMB CHOP, NS AS TO CUT, COOKED, LEAN ONLY","Lamb chop, NS as to cut, cooked, lean only eaten"
+23104000,"LAMB, LOIN CHOP, COOKED, NS AS TO FAT","Lamb, loin chop, cooked, NS as to fat eaten"
+23104010,"LAMB, LOIN CHOP, COOKED, LEAN & FAT","Lamb, loin chop, cooked, lean and fat eaten"
+23104020,"LAMB, LOIN CHOP, COOKED, LEAN ONLY","Lamb, loin chop, cooked, lean only eaten"
+23107000,"LAMB, SHOULDER CHOP, COOKED, NS AS TO FAT","Lamb, shoulder chop, cooked, NS as to fat eaten"
+23107010,"LAMB, SHOULDER CHOP, COOKED, LEAN & FAT","Lamb, shoulder chop, cooked, lean and fat eaten"
+23107020,"LAMB, SHOULDER CHOP, COOKED, LEAN ONLY","Lamb, shoulder chop, cooked, lean only eaten"
+23110000,"LAMB, RIBS, COOKED, LEAN ONLY","Lamb, ribs, cooked, lean only eaten"
+23110010,"LAMB, RIBS, COOKED, NS AS TO FAT","Lamb, ribs, cooked, NS as to fat eaten"
+23110050,"LAMB, RIBS, COOKED, LEAN & FAT","Lamb, ribs, cooked, lean and fat eaten"
+23111010,"LAMB HOCKS, COOKED","Lamb hocks, cooked"
+23120100,"LAMB, ROAST, COOKED, NS AS TO FAT EATEN","Lamb, roast, cooked, NS as to fat eaten"
+23120110,"LAMB, ROAST, COOKED, LEAN & FAT EATEN","Lamb, roast, cooked, lean and fat eaten"
+23120120,"LAMB, ROAST, COOKED, LEAN ONLY EATEN","Lamb, roast, cooked, lean only eaten"
+23132000,"LAMB, GROUND OR PATTY, COOKED","Lamb, ground or patty, cooked"
+23150100,"GOAT, BOILED","Goat, boiled"
+23150200,"GOAT, FRIED","Goat, fried"
+23150250,"GOAT, BAKED","Goat, baked"
+23150270,"GOAT HEAD, COOKED","Goat head, cooked"
+23150300,"GOAT RIBS, COOKED","Goat ribs, cooked"
+23200100,"VEAL, COOKED, NS AS TO CUT & FAT","Veal, NS as to cut, cooked, NS as to fat eaten"
+23200110,"VEAL, NS AS TO CUT, COOKED, LEAN & FAT","Veal, NS as to cut, cooked, lean and fat eaten"
+23200120,"VEAL, NS AS TO CUT, COOKED, LEAN ONLY","Veal, NS as to cut, cooked, lean only eaten"
+23201010,"VEAL CHOP, NS AS TO COOKING METHOD, NS AS TO FAT","Veal chop, NS as to cooking method, NS as to fat eaten"
+23201020,"VEAL CHOP, NS AS TO COOKING METHOD, LEAN & FAT","Veal chop, NS as to cooking method, lean and fat eaten"
+23201030,"VEAL CHOP, NS AS TO COOKING METHOD, LEAN ONLY","Veal chop, NS as to cooking method, lean only eaten"
+23203010,"VEAL CHOP, FRIED, NS AS TO FAT","Veal chop, fried, NS as to fat eaten"
+23203020,"VEAL CHOP, FRIED, LEAN & FAT","Veal chop, fried, lean and fat eaten"
+23203030,"VEAL CHOP, FRIED, LEAN ONLY","Veal chop, fried, lean only eaten"
+23203100,"VEAL CHOP, BROILED, NS AS TO FAT","Veal chop, broiled, NS as to fat eaten"
+23203110,"VEAL CHOP, BROILED, LEAN & FAT","Veal chop, broiled, lean and fat eaten"
+23203120,"VEAL CHOP, BROILED, LEAN ONLY","Veal chop, broiled, lean only eaten"
+23204010,"VEAL CUTLET, NS AS TO COOKING METHOD, NS AS TO FAT","Veal cutlet or steak, NS as to cooking method, NS as to fat eaten"
+23204020,"VEAL CUTLET, NS AS TO COOKING METHOD, LEAN & FAT","Veal cutlet or steak, NS as to cooking method, lean and fat eaten"
+23204030,"VEAL CUTLET, NS AS TO COOKING METHOD, LEAN ONLY","Veal cutlet or steak, NS as to cooking method, lean only eaten"
+23204200,"VEAL CUTLET OR STEAK, BROILED, NS AS TO FAT","Veal cutlet or steak, broiled, NS as to fat eaten"
+23204210,"VEAL CUTLET OR STEAK, BROILED, LEAN & FAT","Veal cutlet or steak, broiled, lean and fat eaten"
+23204220,"VEAL CUTLET OR STEAK, BROILED, LEAN ONLY","Veal cutlet or steak, broiled, lean only eaten"
+23205010,"VEAL CUTLET OR STEAK, FRIED, NS AS TO FAT","Veal cutlet or steak, fried, NS as to fat eaten"
+23205020,"VEAL CUTLET OR STEAK, FRIED, LEAN & FAT","Veal cutlet or steak, fried, lean and fat eaten"
+23205030,"VEAL CUTLET OR STEAK, FRIED, LEAN ONLY","Veal cutlet or steak, fried, lean only eaten"
+23210010,"VEAL, ROASTED, NS AS TO FAT","Veal, roasted, NS as to fat eaten"
+23210020,"VEAL, ROASTED, LEAN & FAT","Veal, roasted, lean and fat eaten"
+23210030,"VEAL, ROASTED, LEAN ONLY","Veal, roasted, lean only eaten"
+23220010,"VEAL, GROUND OR PATTY, COOKED","Veal, ground or patty, cooked"
+23220020,"MOCK CHICKEN LEGS, COOKED","Mock chicken legs, cooked"
+23220030,"VEAL PATTY, BREADED, COOKED","Veal patty, breaded, cooked"
+23310000,"RABBIT, NS AS TO DOMESTIC OR WILD, COOKED","Rabbit, NS as to domestic or wild, cooked"
+23311100,"RABBIT, DOMESTIC, NS AS TO COOKING METHOD","Rabbit, domestic, NS as to cooking method"
+23311120,"RABBIT, NS AS TO DOMESTIC OR WILD, BREADED, FRIED","Rabbit, NS as to domestic or wild, breaded, fried"
+23311200,"RABBIT, WILD, COOKED","Rabbit, wild, cooked"
+23321000,"VENISON/DEER, NFS","Venison/deer, NFS"
+23321050,"VENISON/DEER, CURED","Venison/deer, cured"
+23321100,"VENISON/DEER, ROASTED (INCLUDE ROAST ANTELOPE)","Venison/deer, roasted"
+23321200,"VENISON/DEER STEAK, COOKED, NS AS TO METHOD","Venison/deer steak, cooked, NS as to cooking method"
+23321250,"VENISON/DEER STEAK, BREADED OR FLOURED, COOKED","Venison/deer steak, breaded or floured, cooked, NS as to cooking method"
+23321900,"VENISON/DEER JERKY","Venison/deer jerky"
+23322100,"DEER BOLOGNA","Deer bologna"
+23322300,"DEER CHOP, COOKED (INCLUDE VENISON CHOP)","Deer chop, cooked"
+23322350,"VENISON/DEER RIBS, COOKED","Venison/deer ribs, cooked"
+23322400,"VENISON/DEER, STEWED","Venison/deer, stewed"
+23323100,"MOOSE, COOKED","Moose, cooked"
+23323500,"BEAR, COOKED","Bear, cooked"
+23324100,"CARIBOU, COOKED","Caribou, cooked"
+23326100,"BISON, COOKED","Bison, cooked"
+23331100,"GROUND HOG, COOKED","Ground hog, cooked"
+23332100,"OPOSSUM, COOKED","Opossum, cooked"
+23333100,"SQUIRREL, COOKED","Squirrel, cooked"
+23334100,"BEAVER, COOKED","Beaver, cooked"
+23335100,"RACCOON, COOKED","Raccoon, cooked"
+23340100,"ARMADILLO, COOKED","Armadillo, cooked"
+23345100,"WILD PIG, SMOKED","Wild pig, smoked"
+23350100,"OSTRICH, COOKED","Ostrich, cooked"
+23410010,"LAMB, BABY, STRAINED","Lamb, baby food, strained"
+23420010,"VEAL, BABY, STRAINED","Veal, baby food, strained"
+24100000,"CHICKEN, NS AS TO PART, NS METHOD, SKIN","Chicken, NS as to part and cooking method, NS as to skin eaten"
+24100010,"CHICKEN, NS AS TO PART, NS METHOD, W/ SKIN","Chicken, NS as to part and cooking method, skin eaten"
+24100020,"CHICKEN, NS AS TO PART, NS METHOD, W/O SKIN","Chicken, NS as to part and cooking method, skin not eaten"
+24102000,"CHICKEN, NS PART, ROASTED/BROILED/BAKED, NS SKIN","Chicken, NS as to part, roasted, broiled, or baked, NS as to skin eaten"
+24102010,"CHICKEN, NS PART, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, NS as to part, roasted, broiled, or baked, skin eaten"
+24102020,"CHICKEN, NS PART, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, NS as to part, roasted, broiled, or baked, skin not eaten"
+24103000,"CHICKEN, STEWED, NS PART, NS SKIN","Chicken, NS as to part, stewed, NS as to skin eaten"
+24103010,"CHICKEN, STEWED, NS PART, W/ SKIN","Chicken, NS as to part, stewed, skin eaten"
+24103020,"CHICKEN, STEWED, NS PART, W/O SKIN","Chicken, NS as to part, stewed, skin not eaten"
+24104000,"CHICKEN, FRIED, NO COATING, NS PART, NS SKIN","Chicken, NS as to part, fried, no coating, NS as to skin eaten"
+24104010,"CHICKEN, FRIED, NO COATING, NS PART, W/ SKIN","Chicken, NS as to part, fried, no coating, skin eaten"
+24104020,"CHICKEN, FRIED, NO COATING, NS PART, W/O SKIN","Chicken, NS as to part, fried, no coating, skin not eaten"
+24107000,"CHICKEN, COATED, BKD/FRD, PPD W/ SKIN, NS SKIN EATEN","Chicken, NS as to part, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24107010,"CHICKEN, COATED, BKD/FRD, PPD W/ SKIN, SKIN EATEN","Chicken, NS as to part, coated, baked or fried, prepared with skin, skin/coating eaten"
+24107020,"CHICKEN, COATED, BKD/FRD, PPD W/ SKIN, SKIN NOT EATEN","Chicken, NS as to part, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24107040,"CHICKEN, NS PART,COATED,BKD/FRD,PREP SKINLESS,NS COAT EATEN","Chicken, NS as to part, coated, baked or fried, prepared skinless, NS as to coating eaten"
+24107050,"CHICKEN, NS PART,COATED,BKD/FRD,PREP SKINLESS,COATING EATEN","Chicken, NS as to part, coated, baked or fried, prepared skinless, coating eaten"
+24107060,"CHICKEN, NS PART,COATED,BKD/FRD,PREP SKINLESS,COAT NOT EATEN","Chicken, NS as to part, coated, baked or fried, prepared skinless, coating not eaten"
+24120100,"CHICKEN, BREAST, NFS","Chicken, breast, NS as to cooking method, NS as to skin eaten"
+24120110,"CHICKEN, BREAST, NS AS TO COOKING METHOD, W/SKIN","Chicken, breast, NS as to cooking method, skin eaten"
+24120120,"CHICKEN, BREAST, NS AS TO COOKING METHOD, W/O SKIN","Chicken, breast, NS as to cooking method, skin not eaten"
+24122100,"CHICKEN, BREAST, ROASTED/BROILED/BAKED, NS SKIN","Chicken, breast, roasted, broiled, or baked, NS as to skin eaten"
+24122110,"CHICKEN, BREAST, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, breast, roasted, broiled, or baked, skin eaten"
+24122120,"CHICKEN, BREAST, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, breast, roasted, broiled, or baked, skin not eaten"
+24123100,"CHICKEN, BREAST, STEWED, NS AS TO SKIN","Chicken, breast, stewed, NS as to skin eaten"
+24123110,"CHICKEN, BREAST, STEWED, W/ SKIN","Chicken, breast, stewed, skin eaten"
+24123120,"CHICKEN, BREAST, STEWED, W/O SKIN","Chicken, breast, stewed, skin not eaten"
+24124100,"CHICKEN, BREAST, FRIED, NO COATING, NS AS TO SKIN","Chicken, breast, fried, no coating, NS as to skin eaten"
+24124110,"CHICKEN, BREAST, FRIED, NO COATING, W/ SKIN","Chicken, breast, fried, no coating, skin eaten"
+24124120,"CHICKEN, BREAST, FRIED, NO COATING, W/O SKIN","Chicken, breast, fried, no coating, skin not eaten"
+24127100,"CHICKEN, BREAST,COATED,BKD/FRD,PPD W/ SKIN,NS SKIN EATEN","Chicken, breast, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24127110,"CHICKEN, BREAST,COATED,BKD/FRD,PPD W/ SKIN, SKIN EATEN","Chicken, breast, coated, baked or fried, prepared with skin, skin/coating eaten"
+24127120,"CHICKEN, BREAST,COATED,BKD/FRD,PPD W/ SKIN, SKIN NOT EATEN","Chicken, breast, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24127125,"CHIC, BREAST, FF, COATED/BAKED/FRIED, PREP SKIN,NS SKIN EATE","Chicken, breast, from fast food, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24127130,"CHIC, BREAST, FF, COATED, BAKED/FRIED, PREP SKIN,SKIN EATEN","Chicken, breast, from fast food, coated, baked or fried, prepared with skin, skin/coating eaten"
+24127135,"CHICK,BREAST,FF,COATED, BAKED/ FRIED,PREP SKIN,NO SKIN EATEN","Chicken, breast, from fast food, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24127140,"CHICKEN,BREAST,COATED,BKD/FRD,PPD SKINLESS,NS COAT EATEN","Chicken, breast, coated, baked or fried, prepared skinless, NS as to coating eaten"
+24127150,"CHICKEN,BREAST,COATED,BKD/FRD,PPD SKINLESS,COAT EATEN","Chicken, breast, coated, baked or fried, prepared skinless, coating eaten"
+24127160,"CHICKEN,BREAST,COATED,BKD/FRD,PPD SKINLESS,COAT NOT EATEN","Chicken, breast, coated, baked or fried, prepared skinless, coating not eaten"
+24130200,"CHICKEN, LEG, NFS","Chicken, leg (drumstick and thigh), NS as to cooking method, NS as to skin eaten"
+24130210,"CHICKEN, LEG, NS AS TO COOKING METHOD, W/ SKIN","Chicken, leg (drumstick and thigh), NS as to cooking method, skin eaten"
+24130220,"CHICKEN, LEG, NS AS TO COOKING METHOD, W/O SKIN","Chicken, leg (drumstick and thigh), NS as to cooking method, skin not eaten"
+24132200,"CHICKEN, LEG, ROASTED/BROILED/BAKED, NS SKIN","Chicken, leg (drumstick and thigh), roasted, broiled, or baked, NS as to skin eaten"
+24132210,"CHICKEN, LEG, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, leg (drumstick and thigh), roasted, broiled, or baked, skin eaten"
+24132220,"CHICKEN, LEG, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, leg (drumstick and thigh), roasted, broiled, or baked, skin not eaten"
+24133200,"CHICKEN, LEG, STEWED, NS AS TO SKIN","Chicken, leg (drumstick and thigh), stewed, NS as to skin eaten"
+24133210,"CHICKEN, LEG, STEWED, W/ SKIN","Chicken, leg (drumstick and thigh), stewed, skin eaten"
+24133220,"CHICKEN, LEG, STEWED, W/O SKIN","Chicken, leg (drumstick and thigh), stewed, skin not eaten"
+24134200,"CHICKEN, LEG, FRIED, NO COATING, NS AS TO SKIN","Chicken, leg (drumstick and thigh), fried, no coating, NS as to skin eaten"
+24134210,"CHICKEN, LEG, FRIED, NO COATING, W/ SKIN","Chicken, leg (drumstick and thigh), fried, no coating, skin eaten"
+24134220,"CHICKEN, LEG, FRIED, NO COATING, W/O SKIN","Chicken, leg (drumstick and thigh), fried, no coating, skin not eaten"
+24137200,"CHICKEN,LEG,COATED,BKD/FRD,PPD W/SKIN,NS SKIN EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24137210,"CHICKEN, LEG,COATED,BKD/FRD,PPD W/ SKIN, SKIN EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared with skin, skin/coating eaten"
+24137220,"CHICKEN, LEG,COATED,BKD/FRD,PPD W/ SKIN, SKIN NOT EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared with skin, skin/coating not eaten"
+24137240,"CHICKEN,LEG,COATED,BKD/FRD,PPD SKINLESS,NS COAT EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared skinless, NS as to coating eaten"
+24137250,"CHICKEN,LEG,COATED,BKD/FRD,PPD SKINLESS,COAT EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared skinless, coating eaten"
+24137260,"CHICKEN,LEG,COATED,BKD/FRD,PPD SKINLESS,COAT NOT EATEN","Chicken, leg (drumstick and thigh), coated, baked or fried, prepared skinless, coating not eaten"
+24140200,"CHICKEN, DRUMSTICK, NFS","Chicken, drumstick, NS as to cooking method, NS as to skin eaten"
+24140210,"CHICKEN, DRUMSTICK, NS AS TO COOKING METHOD,W/ SKIN","Chicken, drumstick, NS as to cooking method, skin eaten"
+24140220,"CHICKEN, DRUMSTICK, NS COOKING METHOD, W/O SKIN","Chicken, drumstick, NS as to cooking method, skin not eaten"
+24142200,"CHICKEN, DRUMSTICK, ROASTED/BROILED/BAKED, NS SKIN","Chicken, drumstick, roasted, broiled, or baked, NS as to skin eaten"
+24142210,"CHICKEN, DRUMSTICK, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, drumstick, roasted, broiled, or baked, skin eaten"
+24142220,"CHICKEN, DRUMSTICK, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, drumstick, roasted, broiled, or baked, skin not eaten"
+24143200,"CHICKEN, DRUMSTICK, STEWED, NS AS TO SKIN","Chicken, drumstick, stewed, NS as to skin eaten"
+24143210,"CHICKEN, DRUMSTICK, STEWED, W/ SKIN","Chicken, drumstick, stewed, skin eaten"
+24143220,"CHICKEN, DRUMSTICK, STEWED, W/O SKIN","Chicken, drumstick, stewed, skin not eaten"
+24144200,"CHICKEN, DRUMSTICK, FRIED, NO COATING,NS AS TO SKIN","Chicken, drumstick, fried, no coating, NS as to skin eaten"
+24144210,"CHICKEN, DRUMSTICK, FRIED, NO COATING, W/ SKIN","Chicken, drumstick, fried, no coating, skin eaten"
+24144220,"CHICKEN, DRUMSTICK, FRIED, NO COATING, W/O SKIN","Chicken, drumstick, fried, no coating, skin not eaten"
+24147200,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD W/SKIN,NS SKIN EAT","Chicken, drumstick, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24147210,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD W/SKIN, SKIN EAT","Chicken, drumstick, coated, baked or fried, prepared with skin, skin/coating eaten"
+24147220,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD W/SKIN, SKIN NOT EAT","Chicken, drumstick, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24147225,"CHICK,DRUMSTICK,FF,COATED,BAKED/FRIED,PREP SKIN,NS SKIN EAT","Chicken, drumstick, from fast food, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24147230,"CHIC, DRUMSTICK,FF,COATED, BAKED/FRIED,PREP SKIN,SKIN EATEN","Chicken, drumstick, from fast food, coated, baked or fried, prepared with skin, skin/coating eaten"
+24147235,"CHICK,DRUMSTICK,FF,COATED,BAKED/FRIED,PREP SKIN,SKIN NOT EAT","Chicken, drumstick, from fast food, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24147240,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD SKINLESS,NS COAT EAT","Chicken, drumstick, coated, baked or fried, prepared skinless, NS as to coating eaten"
+24147250,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD SKINLESS, COAT EAT","Chicken, drumstick, coated, baked or fried, prepared skinless, coating eaten"
+24147260,"CHICKEN,DRUMSTICK,COATED,BKD/FRD,PPD SKINLESS, COAT NOT EAT","Chicken, drumstick, coated, baked or fried, prepared skinless, coating not eaten"
+24150200,"CHICKEN, THIGH, NFS","Chicken, thigh, NS as to cooking method, NS as to skin eaten"
+24150210,"CHICKEN, THIGH, NS AS TO COOKING METHOD, W/ SKIN","Chicken, thigh, NS as to cooking method, skin eaten"
+24150220,"CHICKEN, THIGH, NS AS TO COOKING METHOD, W/O SKIN","Chicken, thigh, NS as to cooking method, skin not eaten"
+24152200,"CHICKEN, THIGH, ROASTED/BROILED/BAKED,  NS SKIN","Chicken, thigh, roasted, broiled, or baked,  NS as to skin eaten"
+24152210,"CHICKEN, THIGH, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, thigh, roasted, broiled, or baked, skin eaten"
+24152220,"CHICKEN, THIGH, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, thigh, roasted, broiled, or baked, skin not eaten"
+24153200,"CHICKEN, THIGH, STEWED, NS AS TO SKIN","Chicken, thigh, stewed, NS as to skin eaten"
+24153210,"CHICKEN, THIGH, STEWED, W/ SKIN","Chicken, thigh, stewed, skin eaten"
+24153220,"CHICKEN, THIGH, STEWED, W/O SKIN","Chicken, thigh, stewed, skin not eaten"
+24154200,"CHICKEN, THIGH, FRIED, NO COATING, NS AS TO SKIN","Chicken, thigh, fried, no coating, NS as to skin eaten"
+24154210,"CHICKEN, THIGH, FRIED, NO COATING, W/ SKIN","Chicken, thigh, fried, no coating, skin eaten"
+24154220,"CHICKEN, THIGH, FRIED, NO COATING, W/O SKIN","Chicken, thigh, fried, no coating, skin not eaten"
+24157200,"CHICKEN,THIGH,COATED,BKD/FRD,PPD W/SKIN,NS SKIN EATEN","Chicken, thigh, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24157210,"CHICKEN,THIGH,COATED,BKD/FRD,PPD W/SKIN, SKIN EATEN","Chicken, thigh, coated, baked or fried, prepared with skin, skin/coating eaten"
+24157220,"CHICKEN,THIGH,COATED,BKD/FRD,PPD W/SKIN, SKIN NOT EATEN","Chicken, thigh, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24157225,"CHIC, THIGH, FF, COATED, BAKED/ FRIED, PREP SKIN,NS SKIN EAT","Chicken, thigh, from fast food, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24157230,"CHICK, THIGH, FF, COATED, BAKED OR FRIED, PREP SKIN,SKIN EAT","Chicken, thigh, from fast food, coated, baked or fried, prepared with skin, skin/coating eaten"
+24157235,"CHICK,THIGH,FF,COATED,BAKED/BROILED,PREP SKIN,SKIN NOT EATEN","Chicken, thigh, from fast food, coated, baked or broiled, prepared with skin, skin/coating not eaten"
+24157240,"CHICKEN,THIGH,COATED,BKD/FRD,PPD SKINLESS,NS COAT EATEN","Chicken, thigh, coated, baked or fried, prepared skinless, NS as to coating eaten"
+24157250,"CHICKEN,THIGH,COATED,BKD/FRD,PPD SKINLESS, COAT EATEN","Chicken, thigh, coated, baked or fried, prepared skinless, coating eaten"
+24157260,"CHICKEN,THIGH,COATED,BKD/FRD,PPD SKINLESS, COAT NOT EATEN","Chicken, thigh, coated, baked or fried, prepared skinless, coating not eaten"
+24160100,"CHICKEN, WING, NFS","Chicken, wing, NS as to cooking method, NS as to skin eaten"
+24160110,"CHICKEN, WING, NS AS TO COOKING METHOD, W/ SKIN","Chicken, wing, NS as to cooking method, skin eaten"
+24160120,"CHICKEN, WING, NS AS TO COOKING METHOD, W/O SKIN","Chicken, wing, NS as to cooking method, skin not eaten"
+24162100,"CHICKEN, WING, ROASTED/BROILED/BAKED, NS SKIN","Chicken, wing, roasted, broiled, or baked, NS as to skin eaten"
+24162110,"CHICKEN, WING, ROASTED/BROILED/BAKED, W/ SKIN","Chicken, wing, roasted, broiled, or baked, skin eaten"
+24162120,"CHICKEN, WING, ROASTED/BROILED/BAKED, W/O SKIN","Chicken, wing, roasted, broiled, or baked, skin not eaten"
+24163100,"CHICKEN, WING, STEWED, NS AS TO SKIN","Chicken, wing, stewed, NS as to skin eaten"
+24163110,"CHICKEN, WING, STEWED, W/ SKIN","Chicken, wing, stewed, skin eaten"
+24163120,"CHICKEN, WING, STEWED, W/O SKIN","Chicken, wing, stewed, skin not eaten"
+24164100,"CHICKEN, WING, FRIED, NO COATING, NS AS TO SKIN","Chicken, wing, fried, no coating, NS as to skin eaten"
+24164110,"CHICKEN, WING, FRIED, NO COATING, W/ SKIN","Chicken, wing, fried, no coating, skin eaten"
+24164120,"CHICKEN, WING, FRIED, NO COATING, W/O SKIN","Chicken, wing, fried, no coating, skin not eaten"
+24167100,"CHICKEN,WING,COATED,BKD/FRD,PPD W/SKIN,NS SKIN EATEN","Chicken, wing, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24167110,"CHICKEN,WING,COATED,BKD/FRD,PPD W/SKIN, SKIN EATEN","Chicken, wing, coated, baked or fried, prepared with skin, skin/coating eaten"
+24167120,"CHICKEN,WING,COATED,BKD/FRD,PPD W/SKIN, SKIN NOT EATEN","Chicken, wing, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24167125,"CHIC, WING, FF, COATED, BAKED/FRIED, PREP SKIN,NS SKIN EATEN","Chicken, wing, from fast food, coated, baked or fried, prepared with skin, NS as to skin/coating eaten"
+24167130,"CHIC, WING, FF, COATED, BAKED/FRIED, PREP SKIN, SKIN EATEN","Chicken, wing, from fast food, coated, baked or fried, prepared with skin, skin/coating eaten"
+24167135,"CHIC, WING, FF, COATED,BAKED/FRIED, PREP SKIN, SKIN NO EATEN","Chicken, wing, from fast food, coated, baked or fried, prepared with skin, skin/coating not eaten"
+24170200,"CHICKEN, BACK","Chicken, back"
+24180200,"CHICKEN, NECK OR RIBS, NFS","Chicken, neck or ribs"
+24198340,"CHICKEN TAIL","Chicken, tail"
+24198440,"CHICKEN SKIN","Chicken skin"
+24198500,"CHICKEN FEET","Chicken feet"
+24198570,"CHICKEN, CANNED, MEAT ONLY, LIGHT & DARK MEAT","Chicken, canned, meat only"
+24198670,"CHICKEN ROLL, ROASTED, LIGHT & DARK MEAT","Chicken, chicken roll, roasted"
+24198690,"CHICKEN PATTY, FILLET, OR TENDERS, BREADED, COOKED, FAST FD","Chicken patty, fillet, or tenders, breaded, cooked, from fast food / restaurant"
+24198700,"CHICKEN PATTY/FILLET/TENDERS, BREADED, COOKED","Chicken patty, fillet, or tenders, breaded, cooked"
+24198710,"CHICKEN  PATTY W/ CHEESE, BREADED, COOKED","Chicken patty with cheese, breaded, cooked"
+24198720,"CHICKEN, GROUND","Chicken, ground"
+24198730,"CHICKEN NUGGETS, FROM FAST FOOD / RESTAURANT","Chicken nuggets, from fast food / restaurant"
+24198740,"CHICKEN NUGGETS","Chicken nuggets"
+24198840,"FRIED CHICKEN CHUNKS, P. R. (CHICHARRONES DE POLLO)","Fried chicken chunks, Puerto Rican style (Chicharrones de pollo)"
+24201000,"TURKEY, NFS","Turkey, NFS"
+24201010,"TURKEY, LIGHT MEAT, COOKED, NS AS TO SKIN","Turkey, light meat, cooked, NS as to skin eaten"
+24201020,"TURKEY, LIGHT MEAT, COOKED, W/O SKIN","Turkey, light meat, cooked, skin not eaten"
+24201030,"TURKEY, LIGHT MEAT, COOKED, W/ SKIN","Turkey, light meat, cooked, skin eaten"
+24201050,"TURKEY, LIGHT, BREADED, BAKED/FRIED, NS AS TO SKIN","Turkey, light meat, breaded, baked or fried, NS as to skin eaten"
+24201060,"TURKEY, LIGHT MEAT, BREADED, BAKED/FRIED, W/O SKIN","Turkey, light meat, breaded, baked or fried, skin not eaten"
+24201070,"TURKEY, LIGHT MEAT, BREADED, BAKED/FRIED, W/ SKIN","Turkey, light meat, breaded, baked or fried, skin eaten"
+24201110,"TURKEY, LIGHT MEAT, ROASTED, NS AS TO SKIN","Turkey, light meat, roasted, NS as to skin eaten"
+24201120,"TURKEY, LIGHT MEAT, ROASTED, W/O SKIN","Turkey, light meat, roasted, skin not eaten"
+24201130,"TURKEY, LIGHT MEAT, ROASTED, W/ SKIN","Turkey, light meat, roasted, skin eaten"
+24201210,"TURKEY, DARK MEAT, ROASTED, NS AS TO SKIN","Turkey, dark meat, roasted, NS as to skin eaten"
+24201220,"TURKEY, DARK MEAT, ROASTED, W/O SKIN","Turkey, dark meat, roasted, skin not eaten"
+24201230,"TURKEY, DARK MEAT, ROASTED, W/ SKIN","Turkey, dark meat, roasted, skin eaten"
+24201310,"TURKEY, LIGHT & DARK MEAT, ROASTED, NS AS TO SKIN","Turkey, light and dark meat, roasted, NS as to skin eaten"
+24201320,"TURKEY, LIGHT & DARK MEAT, ROASTED, W/O SKIN","Turkey, light and dark meat, roasted, skin not eaten"
+24201330,"TURKEY, LIGHT & DARK MEAT, ROASTED, W/ SKIN","Turkey, light and dark meat, roasted, skin eaten"
+24201350,"TURKEY, LT/DK MEAT, BATTERED, FRIED, NS AS TO SKIN","Turkey, light or dark meat, battered, fried, NS as to skin eaten"
+24201360,"TURKEY, LIGHT/DARK MEAT, BATTERED, FRIED, W/O SKIN","Turkey, light or dark meat, battered, fried, skin not eaten"
+24201370,"TURKEY, LIGHT/DARK MEAT, BATTERED, FRIED, W/ SKIN","Turkey, light or dark meat, battered, fried, skin eaten"
+24201400,"TURKEY, LIGHT/DARK MEAT, STEWED, NS AS TO SKIN","Turkey, light or dark meat, stewed, NS as to skin eaten"
+24201410,"TURKEY, LIGHT/DARK MEAT, STEWED, W/O SKIN","Turkey, light or dark meat, stewed, skin not eaten"
+24201420,"TURKEY, LIGHT/DARK MEAT, STEWED, W/ SKIN","Turkey light or dark meat, stewed, skin eaten"
+24201500,"TURKEY, SMOKED, NS AS TO SKIN","Turkey, light or dark meat, smoked, cooked, NS as to skin eaten"
+24201510,"TURKEY, SMOKED, SKIN EATEN","Turkey, light or dark meat, smoked, cooked, skin eaten"
+24201520,"TURKEY, SMOKED, SKIN NOT EATEN","Turkey, light or dark meat, smoked, cooked, skin not eaten"
+24202000,"TURKEY, DRUMSTICK, COOKED, NS AS TO SKIN","Turkey, drumstick, cooked, NS as to skin eaten"
+24202010,"TURKEY, DRUMSTICK, COOKED, W/O SKIN","Turkey, drumstick, cooked, skin not eaten"
+24202020,"TURKEY, DRUMSTICK, COOKED, W/ SKIN","Turkey, drumstick, cooked, skin eaten"
+24202050,"TURKEY, DRUMSTICK, ROASTED, NS AS TO SKIN","Turkey, drumstick, roasted, NS as to skin eaten"
+24202060,"TURKEY, DRUMSTICK, ROASTED, W/O SKIN","Turkey, drumstick, roasted, skin not eaten"
+24202070,"TURKEY, DRUMSTICK, ROASTED, W/ SKIN","Turkey, drumstick, roasted, skin eaten"
+24202120,"TURKEY, DRUMSTICK, SMOKED, SKIN EATEN","Turkey, drumstick, smoked, cooked, skin eaten"
+24202450,"TURKEY, THIGH, COOKED, NS AS TO SKIN","Turkey, thigh, cooked, NS as to skin eaten"
+24202460,"TURKEY, THIGH, COOKED, W/ SKIN","Turkey, thigh, cooked, skin eaten"
+24202500,"TURKEY, THIGH, COOKED, W/O SKIN","Turkey, thigh, cooked, skin not eaten"
+24202600,"TURKEY, NECK, COOKED","Turkey, neck, cooked"
+24203000,"TURKEY, WING, COOKED, NS AS TO SKIN","Turkey, wing, cooked, NS as to skin eaten"
+24203010,"TURKEY, WING, COOKED, W/O SKIN","Turkey, wing, cooked, skin not eaten"
+24203020,"TURKEY, WING, COOKED, W/ SKIN","Turkey, wing, cooked, skin eaten"
+24203120,"TURKEY, WING, SMOKED, COOKED, SKIN EATEN","Turkey, wing, smoked, cooked, skin eaten"
+24204000,"TURKEY, ROLLED ROAST, LIGHT OR DARK MEAT, COOKED","Turkey, rolled roast, light or dark meat, cooked"
+24205000,"TURKEY, TAIL, COOKED","Turkey, tail, cooked"
+24205100,"TURKEY, BACK, COOKED","Turkey, back, cooked"
+24206000,"TURKEY, CANNED","Turkey, canned"
+24207000,"TURKEY, GROUND","Turkey, ground"
+24208000,"TURKEY NUGGETS","Turkey, nuggets"
+24208500,"TURKEY BACON, COOKED","Turkey bacon, cooked"
+24300100,"DUCK, COOKED, NS AS TO SKIN","Duck, cooked, NS as to skin eaten"
+24300110,"DUCK, COOKED, W/ SKIN","Duck, cooked, skin eaten"
+24300120,"DUCK, COOKED, W/O SKIN","Duck, cooked, skin not eaten"
+24301000,"DUCK, ROASTED, NS AS TO SKIN","Duck, roasted, NS as to skin eaten"
+24301010,"DUCK, ROASTED, W/ SKIN","Duck, roasted, skin eaten"
+24301020,"DUCK, ROASTED, W/O SKIN","Duck, roasted, skin not eaten"
+24301210,"DUCK, BATTERED, FRIED","Duck, battered, fried"
+24302010,"DUCK, PRESSED, CHINESE","Duck, pressed, Chinese"
+24311010,"GOOSE, WILD, ROASTED","Goose, wild, roasted"
+24400000,"CORNISH GAME HEN, COOKED, NS AS TO SKIN","Cornish game hen, cooked, NS as to skin eaten"
+24400010,"CORNISH GAME HEN, COOKED, W/ SKIN","Cornish game hen, cooked, skin eaten"
+24400020,"CORNISH GAME HEN, COOKED, W/O SKIN","Cornish game hen, cooked, skin not eaten"
+24401000,"CORNISH GAME HEN, ROASTED, NS AS TO SKIN","Cornish game hen, roasted, NS as to skin eaten"
+24401010,"CORNISH GAME HEN, ROASTED, W/ SKIN","Cornish game hen, roasted, skin eaten"
+24401020,"CORNISH GAME HEN, ROASTED, W/O SKIN","Cornish game hen, roasted, skin not eaten"
+24402100,"DOVE, COOKED, NS AS TO COOKING METHOD","Dove, cooked, NS as to cooking method"
+24402110,"DOVE, FRIED","Dove, fried"
+24403100,"QUAIL, COOKED","Quail, cooked"
+24404100,"PHEASANT, COOKED","Pheasant, cooked"
+24701000,"CHICKEN, BABY, NS AS TO STRAINED OR JUNIOR","Chicken, baby food, NS as to strained or junior"
+24701010,"CHICKEN, BABY, STRAINED","Chicken, baby food, strained"
+24701020,"CHICKEN, BABY, JUNIOR","Chicken, baby food, junior"
+24703000,"TURKEY, BABY, NS AS TO STRAINED OR JUNIOR","Turkey, baby food, NS as to strained or junior"
+24703010,"TURKEY, BABY, STRAINED","Turkey, baby food, strained"
+24703020,"TURKEY, BABY, JUNIOR","Turkey, baby food, junior"
+24705010,"CHICKEN STICK, BABY FOOD","Chicken stick, baby food"
+24706010,"TURKEY STICK, BABY FOOD","Turkey stick, baby food"
+25110120,"BEEF LIVER, BRAISED","Beef liver, braised"
+25110140,"BEEF LIVER, FRIED","Beef liver, fried"
+25110420,"CHICKEN LIVER, BRAISED","Chicken liver, braised"
+25110450,"CHICKEN LIVER, FRIED","Chicken liver, fried"
+25112200,"LIVER PASTE OR PATE, CHICKEN (INCLUDE PATE, NFS)","Liver paste or pate, chicken"
+25120000,"HEART, COOKED","Heart, cooked"
+25130000,"KIDNEY, COOKED","Kidney, cooked"
+25140110,"SWEETBREADS, COOKED","Sweetbreads, cooked"
+25150000,"BRAINS, COOKED","Brains, cooked"
+25160000,"TONGUE, COOKED","Tongue, cooked"
+25160110,"TONGUE, SMOKED,CURED OR PICKLED, COOKED","Tongue, smoked, cured, or pickled, cooked"
+25160130,"TONGUE POT ROAST, P.R. (LENGUA AL CALDERO)","Tongue pot roast, Puerto Rican style (Lengua al caldero)"
+25170110,"TRIPE, COOKED","Tripe, cooked"
+25170210,"CHITTERLINGS, COOKED","Chitterlings, cooked"
+25170310,"HOG MAWS (STOMACH) COOKED","Hog maws (stomach), cooked"
+25170420,"GIZZARD, COOKED","Gizzard, cooked"
+25210110,"FRANKFURTER, WIENER OR HOT DOG, NFS","Frankfurter, wiener, or hot dog, NFS"
+25210150,"FRANKFURTER OR HOT DOG, CHEESE-FILLED","Frankfurter or hot dog, cheese-filled"
+25210210,"FRANKFURTER OR HOT DOG, BEEF","Frankfurter or hot dog, beef"
+25210220,"FRANKFURTER OR HOT DOG, BEEF & PORK","Frankfurter or hot dog, beef and pork"
+25210240,"FRANFURTER/HOT DOG, BEEF & PORK, LIGHT","Frankfurter or hot dog, beef and pork, reduced fat or light"
+25210250,"FRANKFURTER OR HOT DOG, MEAT & POULTRY, FAT FREE","Frankfurter or hot dog, meat and poultry, fat free"
+25210280,"FRANKFURTER OR HOT DOG, MEAT & POULTRY","Frankfurter or hot dog, meat and poultry"
+25210290,"FRANKFURTER OR HOT DOG, MEAT & POULTRY, LIGHT","Frankfurter or hot dog, meat and poultry, reduced fat or light"
+25210310,"FRANKFURTER OR HOT DOG, CHICKEN","Frankfurter or hot dog, chicken"
+25210410,"FRANKFURTER OR HOT DOG, TURKEY","Frankfurter or hot dog, turkey"
+25210620,"FRANKFURTER OR HOT DOG, BEEF, REDUCED FAT OR LIGHT","Frankfurter or hot dog, beef, reduced fat or light"
+25210750,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, NFS","Frankfurter or hot dog, reduced fat or light, NFS"
+25220010,"COLD CUT, NFS","Cold cut, NFS"
+25220105,"BEEF SAUSAGE","Beef sausage"
+25220106,"BEEF SAUSAGE, REDUCED FAT","Beef sausage, reduced fat"
+25220150,"BEEF SAUSAGE WITH CHEESE","Beef sausage with cheese"
+25220210,"BLOOD SAUSAGE","Blood sausage"
+25220350,"BRATWURST","Bratwurst"
+25220360,"BRATWURST W/ CHEESE","Bratwurst, with cheese"
+25220390,"BOLOGNA, BEEF, LOW FAT","Bologna, beef, lowfat"
+25220400,"BOLOGNA, PORK AND BEEF","Bologna, pork and beef"
+25220410,"BOLOGNA, NFS","Bologna, NFS"
+25220420,"BOLOGNA, LEBANON","Bologna, Lebanon"
+25220430,"BOLOGNA, BEEF","Bologna, beef"
+25220440,"BOLOGNA, TURKEY","Bologna, turkey"
+25220450,"BOLOGNA RING, SMOKED","Bologna ring, smoked"
+25220460,"BOLOGNA, PORK","Bologna, pork"
+25220470,"BOLOGNA, BEEF, LOWER SODIUM","Bologna, beef, lower sodium"
+25220480,"BOLOGNA, CHICKEN, BEEF, & PORK","Bologna, chicken, beef, and pork"
+25220490,"BOLOGNA, W/ CHEESE","Bologna, with cheese"
+25220500,"BOLOGNA, BEEF & PORK, LOWFAT","Bologna, beef and pork, lowfat"
+25220510,"CAPICOLA","Capicola"
+25220650,"TURKEY OR CHICKEN AND BEEF SAUSAGE","Turkey or chicken and beef sausage"
+25220710,"CHORIZO","Chorizo"
+25220910,"HEAD CHEESE","Head cheese"
+25221110,"KNOCKWURST","Knockwurst"
+25221210,"MORTADELLA","Mortadella"
+25221250,"PEPPERONI","Pepperoni"
+25221310,"POLISH SAUSAGE","Polish sausage"
+25221350,"ITALIAN SAUSAGE","Italian sausage"
+25221400,"SAUSAGE (NOT COLD CUT), NFS","Sausage (not cold cut), NFS"
+25221405,"PORK SAUSAGE","Pork sausage"
+25221406,"PORK SAUSAGE, REDUCED FAT","Pork sausage, reduced fat"
+25221450,"PORK SAUSAGE RICE LINKS","Pork sausage rice links"
+25221460,"PORK & BEEF SAUSAGE","Pork and beef sausage"
+25221500,"SALAMI, NFS","Salami, NFS"
+25221510,"SALAMI, SOFT, COOKED","Salami, soft, cooked"
+25221520,"SALAMI, DRY OR HARD","Salami, dry or hard"
+25221530,"SALAMI, BEEF","Salami, beef"
+25221610,"SCRAPPLE, COOKED","Scrapple, cooked"
+25221710,"SOUSE","Souse"
+25221810,"THURINGER (INCLUDE SUMMER SAUSAGE)","Thuringer"
+25221830,"TURKEY OR CHICKEN SAUSAGE","Turkey or chicken sausage"
+25221860,"TURKEY OR CHICKEN SAUSAGE, REDUCED FAT","Turkey or chicken sausage, reduced fat"
+25221870,"TURKEY AND PORK SAUSAGE","Turkey or chicken and pork sausage"
+25221880,"TURKEY OR CHICKEN, PORK, AND BEEF SAUSAGE, REDUCED FAT","Turkey or chicken, pork, and beef sausage, reduced fat"
+25221910,"VIENNA SAUSAGE, CANNED","Vienna sausage, canned"
+25221950,"PICKLED SAUSAGE","Pickled sausage"
+25230110,"LUNCHEON MEAT, NFS","Luncheon meat, NFS"
+25230210,"HAM, SLICED, PREPACKAGED OR DELI, LUNCHEON MEAT","Ham, sliced, prepackaged or deli, luncheon meat"
+25230220,"HAM, SLICED, LOW SALT, PREPACKAGED/DELI, LUNCH MEAT","Ham, sliced, low salt, prepackaged or deli, luncheon meat"
+25230230,"HAM, SLICED, EXTRA LEAN, PREPACKAGED/DELI","Ham, sliced, extra lean, prepackaged or deli, luncheon meat"
+25230235,"HAM, SLICED, EXTRA LEAN, LOWER SODIUM, PREPACKAGED OR DELI","Ham, sliced, extra lean, lower sodium, prepackaged or deli, luncheon meat"
+25230310,"CHICKEN/TURKEY LOAF, PREPACK/DELI, LUNCHEON MEAT","Chicken or turkey loaf, prepackaged or deli, luncheon meat"
+25230410,"HAM LOAF, LUNCHEON MEAT","Ham loaf, luncheon meat"
+25230430,"HAM & CHEESE LOAF","Ham and cheese loaf"
+25230450,"HONEY LOAF","Honey loaf"
+25230510,"HAM,LUNCH MEAT,CHOP,MINCED,PRESSD,MINCED,NOT CANNED","Ham, luncheon meat, chopped, minced, pressed, spiced, not canned"
+25230520,"HAM, LUNCHEON MEAT, CHOPPED, SPICED,LOWFAT, NOT CAN","Ham, luncheon meat, chopped, minced, pressed, spiced, lowfat, not canned"
+25230530,"HAM/PORK , LUNCHEON MEAT, CHOPPED, CAN (INCL SPAM)","Ham and pork, luncheon meat, chopped, minced, pressed, spiced, canned"
+25230540,"HAM, PORK & CHICKEN, LUNCHEON MEAT, CHOPPED, CANNED","Ham, pork and chicken, luncheon meat, chopped, minced, pressed, spiced, canned"
+25230550,"HAM, PORK & CHICKEN, LUNCHEON MEAT, CHOPPED, CAN, RED SODIUM","Ham, pork, and chicken, luncheon meat, chopped, minced, pressed, spiced, canned, reduced sodium"
+25230560,"LIVERWURST","Liverwurst"
+25230610,"LUNCHEON LOAF (OLIVE, PICKLE OR PIMIENTO)","Luncheon loaf (olive, pickle, or pimiento)"
+25230710,"SANDWICH LOAF, LUNCHEON MEAT","Sandwich loaf, luncheon meat"
+25230790,"TURKEY HAM, SLICED, XTRA LEAN, PKG'D, DELI","Turkey ham, sliced, extra lean, prepackaged or deli, luncheon meat"
+25230800,"TURKEY HAM","Turkey ham"
+25230810,"VEAL LOAF","Veal loaf"
+25230820,"TURKEY PASTRAMI","Turkey pastrami"
+25230840,"TURKEY SALAMI","Turkey salami"
+25230900,"TURKEY OR CHICKEN BREAST, PKG'D/DELI, LUNCHEON MEAT","Turkey or chicken breast, prepackaged or deli, luncheon meat"
+25230905,"TURKEY/CHICKEN BREAST, LOW SALT, PREPACK/DELI, LUNCHEON MEAT","Turkey or chicken breast, low salt, prepackaged or deli, luncheon meat"
+25231110,"BEEF, SLICED, PREPACKAGED/DELI, LUNCHEON MEAT","Beef, sliced, prepackaged or deli, luncheon meat"
+25231150,"CORNED BEEF, PRESSED","Corned beef, pressed"
+25240000,"MEAT SPREAD OR POTTED MEAT, NFS","Meat spread or potted meat, NFS"
+25240110,"CHICKEN SALAD SPREAD","Chicken salad spread"
+25240210,"HAM, DEVILED OR POTTED","Ham, deviled or potted"
+25240220,"HAM SALAD SPREAD","Ham salad spread"
+25240310,"ROAST BEEF SPREAD","Roast beef spread"
+25240320,"CORNED BEEF SPREAD","Corned beef spread"
+26100100,"FISH, NS AS TO TYPE, RAW","Fish, NS as to type, raw"
+26100110,"FISH, COOKED, NS AS TO TYPE & COOKING METHOD","Fish, NS as to type, cooked, NS as to cooking method"
+26100120,"FISH, NS AS TO TYPE, BAKED OR BROILED, MADE WITH OIL","Fish, NS as to type, baked or broiled, made with oil"
+26100121,"FISH, NS AS TO TYPE, BAKED OR BROILED, MADE WITH BUTTER","Fish, NS as to type, baked or broiled, made with butter"
+26100122,"FISH, NS AS TO TYPE, BAKED OR BROILED, MADE WITH MARGARINE","Fish, NS as to type, baked or broiled, made with margarine"
+26100123,"FISH, NS AS TO TYPE, BAKED OR BROILED, MADE WITHOUT FAT","Fish, NS as to type, baked or broiled, made without fat"
+26100124,"FISH, NS AS TO TYPE, BAKED OR BROILED, MADE W/COOKING SPRAY","Fish, NS as to type, baked or broiled, made with cooking spray"
+26100130,"FISH, NS AS TO TYPE, COATED, BAKED, MADE WITH OIL","Fish, NS as to type, coated, baked or broiled, made with oil"
+26100131,"FISH, NS AS TO TYPE, COATED, BAKED OR BROILED,W/ BUTTER","Fish, NS as to type, coated, baked or broiled, made with butter"
+26100132,"FISH, NS AS TO TYPE, COATED, BAKED OR BROILED, W/MARGARINE","Fish, NS as to type, coated, baked or broiled, made with margarine"
+26100133,"FISH, NS AS TO TYPE, COATED, BAKED OR BROILED, W/OUT FAT","Fish, NS as to type, coated, baked or broiled, made without fat"
+26100134,"FISH, NS AS TO TYPE, COATED, BAKED/ BROILED, W/COOKING SPRAY","Fish, NS as to type, coated, baked or broiled, made with cooking spray"
+26100140,"FISH, NS AS TO TYPE, COATED, FRIED, MADE WITH OIL","Fish, NS as to type, coated, fried, made with oil"
+26100141,"FISH, NS AS TO TYPE, COATED, FRIED, MADE WITH BUTTER","Fish, NS as to type, coated, fried, made with butter"
+26100142,"FISH, NS AS TO TYPE, COATED, FRIED, MADE WITH MARGARINE","Fish, NS as to type, coated, fried, made with margarine"
+26100143,"FISH, NS AS TO TYPE, COATED, FRIED, MADE WITHOUT FAT","Fish, NS as to type, coated, fried, made without fat"
+26100144,"FISH, NS AS TO TYPE, COATED, MADE WITH COOKING SPRAY","Fish, NS as to type, coated, fried, made with cooking spray"
+26100160,"FISH, NS AS TO TYPE, STEAMED","Fish, NS as to type, steamed"
+26100170,"FISH, NS AS TO TYPE, DRIED","Fish, NS as to type, dried"
+26100180,"FISH, NS AS TO TYPE, CANNED","Fish, NS as to type, canned"
+26100190,"FISH, NS AS TO TYPE, SMOKED","Fish, NS as to type, smoked"
+26100200,"FISH, NS AS TO TYPE, FROM FAST FOOD","Fish, NS as to type, from fast food"
+26100260,"FISH STICK, PATTY OR NUGGET FROM FAST FOOD","Fish stick, patty or nugget from fast food"
+26100270,"FISH STICK, PATTY OR NUGGET FROM RESTAURANT, HOME, OR OTHER","Fish stick, patty or nugget from restaurant, home, or other place"
+26101110,"ANCHOVY, COOKED, NS AS TO COOKING METHOD","Anchovy, cooked, NS as to cooking method"
+26101180,"ANCHOVY, CANNED","Anchovy, canned"
+26103110,"BARRACUDA, COOKED, NS AS TO COOKING METHOD","Barracuda, cooked, NS as to cooking method"
+26103120,"BARRACUDA, BAKED OR BROILED, FAT ADDED IN COOKING","Barracuda, baked or broiled, fat added in cooking"
+26103121,"BARRACUDA, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Barracuda, baked or broiled, fat not added in cooking"
+26103130,"BARRACUDA, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Barracuda, coated, baked or broiled, fat added in cooking"
+26103131,"BARRACUDA, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKIN","Barracuda, coated, baked or broiled, fat not added in cooking"
+26103140,"BARRACUDA, COATED, FRIED","Barracuda, coated, fried"
+26103160,"BARRACUDA, STEAMED OR POACHED","Barracuda, steamed or poached"
+26105110,"CARP, COOKED, NS AS TO COOKING METHOD","Carp, cooked, NS as to cooking method"
+26105120,"CARP, BAKED OR BROILED, FAT ADDED IN COOKING","Carp, baked or broiled, fat added in cooking"
+26105121,"CARP, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Carp, baked or broiled, fat not added in cooking"
+26105130,"CARP, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Carp, coated, baked or broiled, fat added in cooking"
+26105131,"CARP, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Carp, coated, baked or broiled, fat not added in cooking"
+26105140,"CARP, COATED, FRIED","Carp, coated, fried"
+26105160,"CARP, STEAMED OR POACHED","Carp, steamed or poached"
+26105190,"CARP, SMOKED","Carp, smoked"
+26107110,"CATFISH, COOKED, NS AS TO COOKING METHOD","Catfish, cooked, NS as to cooking method"
+26107120,"CATFISH, BAKED OR BROILED, MADE WITH OIL","Catfish, baked or broiled, made with oil"
+26107121,"CATFISH, BAKED OR BROILED, MADE WITH BUTTER","Catfish, baked or broiled, made with butter"
+26107122,"CATFISH, BAKED OR BROILED, MADE WITH MARGARINE","Catfish, baked or broiled, made with margarine"
+26107123,"CATFISH, BAKED OR BROILED, MADE WITHOUT FAT","Catfish, baked or broiled, made without fat"
+26107124,"CATFISH, BAKED OR BROILED, MADE WITH COOKING SPRAY","Catfish, baked or broiled, made with cooking spray"
+26107130,"CATFISH, COATED, BAKED OR BROILED, MADE WITH OIL","Catfish, coated, baked or broiled, made with oil"
+26107131,"CATFISH, COATED, BAKED OR BROILED, MADE WITH BUTTER","Catfish, coated, baked or broiled, made with butter"
+26107132,"CATFISH, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Catfish, coated, baked or broiled, made with margarine"
+26107133,"CATFISH, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Catfish, coated, baked or broiled, made without fat"
+26107134,"CATFISH, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Catfish, coated, baked or broiled, made with cooking spray"
+26107140,"CATFISH, COATED, FRIED, MADE WITH OIL","Catfish, coated, fried, made with oil"
+26107141,"CATFISH, COATED, FRIED, MADE WITH BUTTER","Catfish, coated, fried, made with butter"
+26107142,"CATFISH, COATED, FRIED, MADE WITH MARGARINE","Catfish, coated, fried, made with margarine"
+26107143,"CATFISH, COATED, FRIED, MADE WITHOUT FAT","Catfish, coated, fried, made without fat"
+26107144,"CATFISH, COATED, FRIED, MADE WITH COOKING SPRAY","Catfish, coated, fried, made with cooking spray"
+26107160,"CATFISH, STEAMED OR POACHED","Catfish, steamed or poached"
+26109110,"COD, COOKED, NS AS TO COOKING METHOD","Cod, cooked, NS as to cooking method"
+26109120,"COD, BAKED OR BROILED, MADE WITH OIL","Cod, baked or broiled, made with oil"
+26109121,"COD, BAKED OR BROILED, MADE WITH BUTTER","Cod, baked or broiled, made with butter"
+26109122,"COD, BAKED OR BROILED, MADE WITH MARGARINE","Cod, baked or broiled, made with margarine"
+26109123,"COD, BAKED OR BROILED, MADE WITHOUT FAT","Cod, baked or broiled, made without fat"
+26109124,"COD, BAKED OR BROILED, MADE WITH COOKING SPRAY","Cod, baked or broiled, made with cooking spray"
+26109130,"COD, COATED, BAKED OR BROILED, MADE WITH OIL","Cod, coated, baked or broiled, made with oil"
+26109131,"COD, COATED, BAKED OR BROILED, MADE WITH BUTTER","Cod, coated, baked or broiled, made with butter"
+26109132,"COD, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Cod, coated, baked or broiled, made with margarine"
+26109133,"COD, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Cod, coated, baked or broiled, made without fat"
+26109134,"COD, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Cod, coated, baked or broiled, made with cooking spray"
+26109140,"COD, COATED, FRIED, MADE WITH OIL","Cod, coated, fried, made with oil"
+26109141,"COD, COATED, FRIED, MADE WITH BUTTER","Cod, coated, fried, made with butter"
+26109142,"COD, COATED, FRIED, MADE WITH MARGARINE","Cod, coated, fried, made with margarine"
+26109143,"COD, COATED, FRIED, MADE WITHOUT FAT","Cod, coated, fried, made without fat"
+26109144,"COD, COATED, FRIED, MADE WITH COOKING SPRAY","Cod, coated, fried, made with cooking spray"
+26109160,"COD, STEAMED OR POACHED","Cod, steamed or poached"
+26109170,"COD, DRIED, SALTED","Cod, dried, salted"
+26109180,"COD, DRIED, SALTED, SALT REMOVED IN WATER","Cod, dried, salted, salt removed in water"
+26109190,"COD, SMOKED","Cod, smoked"
+26111110,"CROAKER, COOKED, NS AS TO COOKING METHOD","Croaker, cooked, NS as to cooking method"
+26111120,"CROAKER, BAKED OR BROILED, FAT ADDED IN COOKING","Croaker, baked or broiled, fat added in cooking"
+26111121,"CROAKER, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Croaker, baked or broiled, fat not added in cooking"
+26111130,"CROAKER, COATED, BAKED, FAT ADDED IN COOKING","Croaker, coated, baked or broiled, fat added in cooking"
+26111131,"CROAKER, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Croaker, coated, baked or broiled, fat not added in cooking"
+26111140,"CROAKER, COATED, FRIED","Croaker, coated, fried"
+26111160,"CROAKER, STEAMED OR POACHED","Croaker, steamed or poached"
+26113110,"EEL, COOKED, NS AS TO COOKING METHOD","Eel, cooked, NS as to cooking method"
+26113160,"EEL, STEAMED OR POACHED","Eel, steamed or poached"
+26113190,"EEL, SMOKED","Eel, smoked"
+26115000,"FLOUNDER, RAW","Flounder, raw"
+26115110,"FLOUNDER, COOKED, NS AS TO COOKING METHOD","Flounder, cooked, NS as to cooking method"
+26115120,"FLOUNDER, BAKED OR BROILED, MADE WTIH OIL","Flounder, baked or broiled, made with oil"
+26115121,"FLOUNDER, BAKED OR BROILED, MADE WITH BUTTER","Flounder, baked or broiled, made with butter"
+26115122,"FLOUNDER, BAKED OR BROILED, MADE WITH MARGARINE","Flounder, baked or broiled, made with margarine"
+26115123,"FLOUNDER, BAKED OR BROILED, MADE WITHOUT FAT","Flounder, baked or broiled, made without fat"
+26115124,"FLOUNDER, BAKED OR BROILED, MADE WITH COOKING SPRAY","Flounder, baked or broiled, made with cooking spray"
+26115130,"FLOUNDER, COATED, BAKED OR BROILED, MADE WITH OIL","Flounder, coated, baked or broiled, made with oil"
+26115131,"FLOUNDER, COATED, BAKED OR BROILED, MADE WITH BUTTER","Flounder, coated, baked or broiled, made with butter"
+26115132,"FLOUNDER, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Flounder, coated, baked or broiled, made with margarine"
+26115133,"FLOUNDER, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Flounder, coated, baked or broiled, made without fat"
+26115134,"FLOUNDER, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Flounder, coated, baked or broiled, made with cooking spray"
+26115140,"FLOUNDER, COATED, FRIED, MADE WITH OIL","Flounder, coated, fried, made with oil"
+26115141,"FLOUNDER, COATED, FRIED, MADE WITH BUTTER","Flounder, coated, fried, made with butter"
+26115142,"FLOUNDER, COATED, FRIED, MADE WITH MARGARINE","Flounder, coated, fried, made with margarine"
+26115143,"FLOUNDER, COATED, FRIED, MADE WITHOUT FAT","Flounder, coated, fried, made without fat"
+26115144,"FLOUNDER, COATED, FRIED, MADE WITH COOKING SPRAY","Flounder, coated, fried, made with cooking spray"
+26115160,"FLOUNDER, STEAMED OR POACHED","Flounder, steamed or poached"
+26115190,"FLOUNDER, SMOKED","Flounder, smoked"
+26117110,"HADDOCK, COOKED, NS AS TO COOKING METHOD","Haddock, cooked, NS as to cooking method"
+26117120,"HADDOCK, BAKED OR BROILED, FAT ADDED IN COOKING","Haddock, baked or broiled, fat added in cooking"
+26117121,"HADDOCK, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Haddock, baked or broiled, fat not added in cooking"
+26117130,"HADDOCK, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Haddock, coated, baked or broiled, fat added in cooking"
+26117131,"HADDOCK, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Haddock, coated, baked or broiled, fat not added in cooking"
+26117140,"HADDOCK, COATED, FRIED","Haddock, coated, fried"
+26117160,"HADDOCK, STEAMED OR POACHED","Haddock, steamed or poached"
+26117190,"HADDOCK, SMOKED","Haddock, smoked"
+26118000,"HALIBUT, RAW","Halibut, raw"
+26118010,"HALIBUT, COOKED, NS AS TO COOKING METHOD","Halibut, cooked, NS as to cooking method"
+26118020,"HALIBUT, BAKED OR BROILED, MADE WITH OIL","Halibut, baked or broiled, made with oil"
+26118021,"HALIBUT, BAKED OR BROILED, MADE WITH BUTTER","Halibut, baked or broiled, made with butter"
+26118022,"HALIBUT, BAKED OR BROILED, MADE WITH MARGARINE","Halibut, baked or broiled, made with margarine"
+26118023,"HALIBUT, BAKED OR BROILED, MADE WITHOUT FAT","Halibut, baked or broiled, made without fat"
+26118024,"HALIBUT, BAKED OR BROILED, MADE WITH COOKING SPRAY","Halibut, baked or broiled, made with cooking spray"
+26118030,"HALIBUT, COATED, BAKED OR BROILED, MADE WITH OIL","Halibut, coated, baked or broiled, made with oil"
+26118031,"HALIBUT, COATED, BAKED OR BROILED, MADE WITH BUTTER","Halibut, coated, baked or broiled, made with butter"
+26118032,"HALIBUT, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Halibut, coated, baked or broiled, made with margarine"
+26118033,"HALIBUT, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Halibut, coated, baked or broiled, made without fat"
+26118034,"HALIBUT, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Halibut, coated, baked or broiled, made with cooking spray"
+26118040,"HALIBUT, COATED, FRIED, MADE WITH OIL","Halibut, coated, fried, made with oil"
+26118041,"HALIBUT, COATED, FRIED, MADE WITH BUTTER","Halibut, coated, fried, made with butter"
+26118042,"HALIBUT, COATED, FRIED, MADE WITH MARGARINE","Halibut, coated, fried, made with margarine"
+26118043,"HALIBUT, COATED, FRIED, MADE WITHOUT FAT","Halibut, coated, fried, made without fat"
+26118044,"HALIBUT, COATED, FRIED, MADE WITH COOKING SPRAY","Halibut, coated, fried, made with cooking spray"
+26118050,"HALIBUT, STEAMED OR POACHED","Halibut, steamed or poached"
+26118060,"HALIBUT, SMOKED","Halibut, smoked"
+26119100,"HERRING, RAW","Herring, raw"
+26119110,"HERRING, COOKED, NS AS TO COOKING METHOD","Herring, cooked, NS as to cooking method"
+26119120,"HERRING, BAKED OR BROILED, FAT ADDED IN COOKING","Herring, baked or broiled, fat added in cooking"
+26119121,"HERRING, BAKED OR BROILED, FAT NOT USED IN PREPARATION","Herring, baked or broiled, fat not added in cooking"
+26119130,"HERRING, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Herring, coated, baked or broiled, fat added in cooking"
+26119131,"HERRING, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Herring, coated, baked or broiled, fat not added in cooking"
+26119140,"HERRING, COATED, FRIED","Herring, coated, fried"
+26119160,"HERRING, PICKLED, IN CREAM SAUCE","Herring, pickled, in cream sauce"
+26119170,"HERRING, DRIED, SALTED","Herring, dried, salted"
+26119180,"HERRING, PICKLED","Herring, pickled"
+26119190,"HERRING, SMOKED, KIPPERED","Herring, smoked, kippered"
+26121100,"MACKEREL, RAW","Mackerel, raw"
+26121110,"MACKEREL, COOKED, NS AS TO COOKING METHOD","Mackerel, cooked, NS as to cooking method"
+26121120,"MACKEREL, BAKED OR BROILED, FAT ADDED IN COOKING","Mackerel, baked or broiled, fat added in cooking"
+26121121,"MACKEREL, BAKED OR BROILED, FAT NOT USED IN PREPARATION","Mackerel, baked or broiled, fat not added in cooking"
+26121131,"MACKEREL, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Mackerel, coated, baked or broiled, fat added in cooking"
+26121132,"MACKEREL, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Mackerel, coated, baked or broiled, fat not added in cooking"
+26121140,"MACKEREL, COATED, FRIED","Mackerel, coated, fried"
+26121160,"MACKEREL, PICKLED","Mackerel, pickled"
+26121180,"MACKEREL, CANNED","Mackerel, canned"
+26121190,"MACKEREL, SMOKED","Mackerel, smoked"
+26123100,"MULLET, RAW","Mullet, raw"
+26123110,"MULLET, COOKED, NS AS TO COOKING METHOD","Mullet, cooked, NS as to cooking method"
+26123120,"MULLET, BAKED OR BROILED, FAT USED IN PREPARATION","Mullet, baked or broiled, fat added in cooking"
+26123121,"MULLET, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Mullet, baked or broiled, fat not added in cooking"
+26123130,"MULLET, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Mullet, coated, baked or broiled, fat added in cooking"
+26123131,"MULLET, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Mullet, coated, baked or broiled, fat not added in cooking"
+26123140,"MULLET, COATED, FRIED","Mullet, coated, fried"
+26123160,"MULLET, STEAMED OR POACHED","Mullet, steamed or poached"
+26125100,"OCEAN PERCH, RAW","Ocean perch, raw"
+26125110,"OCEAN PERCH, COOKED, NS AS TO COOKING METHOD","Ocean perch, cooked, NS as to cooking method"
+26125120,"OCEAN PERCH, BAKED OR BROILED, FAT USED IN PREPARATION","Ocean perch, baked or broiled, fat added in cooking"
+26125121,"OCEAN PERCH, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Ocean perch, baked or broiled, fat not added in cooking"
+26125130,"OCEAN PERCH, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Ocean perch, coated, baked or broiled, fat added in cooking"
+26125131,"OCEAN PERCH, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOK","Ocean perch, coated, baked or broiled, fat not added in cooking"
+26125140,"OCEAN PERCH, COATED, FRIED","Ocean perch, coated, fried"
+26125160,"OCEAN PERCH, STEAMED OR POACHED","Ocean perch, steamed or poached"
+26127110,"PERCH, COOKED, NS AS TO COOKING METHOD","Perch, cooked, NS as to cooking method"
+26127120,"PERCH, BAKED OR BROILED, MADE WITH OIL","Perch, baked or broiled, made with oil"
+26127121,"PERCH, BAKED OR BROILED, MADE WITH BUTTER","Perch, baked or broiled, made with butter"
+26127122,"PERCH, BAKED OR BROILED, MADE WITH MARGARINE","Perch, baked or broiled, made with margarine"
+26127123,"PERCH, BAKED OR BROILED, MADE WITHOUT FAT","Perch, baked or broiled, made without fat"
+26127124,"PERCH, BAKED OR BROILED, MADE WITH COOKING SPRAY","Perch, baked or broiled, made with cooking spray"
+26127130,"PERCH, COATED, BAKED OR BROILED, MADE WITH OIL","Perch, coated, baked or broiled, made with oil"
+26127131,"PERCH, COATED, BAKED OR BROILED, MADE WITH BUTTER","Perch, coated, baked or broiled, made with butter"
+26127132,"PERCH, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Perch, coated, baked or broiled, made with margarine"
+26127133,"PERCH, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Perch, coated, baked or broiled, made without fat"
+26127134,"PERCH, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Perch, coated, baked or broiled, made with cooking spray"
+26127140,"PERCH, COATED, FRIED","Perch, coated, fried, made with oil"
+26127141,"PERCH, COATED, FRIED, MADE WITH BUTTER","Perch, coated, fried, made with butter"
+26127142,"PERCH, COATED, FRIED, MADE WITH MARGARINE","Perch, coated, fried, made with margarine"
+26127143,"PERCH, COATED, FRIED, MADE WITHOUT FAT","Perch, coated, fried, made without fat"
+26127144,"PERCH, COATED, FRIED, MADE WITH COOKING SPRAY","Perch, coated, fried, made with cooking spray"
+26127160,"PERCH, STEAMED OR POACHED","Perch, steamed or poached"
+26129110,"PIKE, COOKED, NS AS TO COOKING METHOD","Pike, cooked, NS as to cooking method"
+26129120,"PIKE, BAKED OR BROILED, FAT ADDED IN COOKING","Pike, baked or broiled, fat added in cooking"
+26129121,"PIKE, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Pike, baked or broiled, fat not added in cooking"
+26129130,"PIKE, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Pike, coated, baked or broiled, fat added in cooking"
+26129131,"PIKE, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Pike, coated, baked or broiled, fat not added in cooking"
+26129140,"PIKE, COATED, FRIED","Pike, coated, fried"
+26129160,"PIKE, STEAMED OR POACHED","Pike, steamed or poached"
+26131100,"POMPANO, RAW","Pompano, raw"
+26131110,"POMPANO, COOKED, NS AS TO COOKING METHOD","Pompano, cooked, NS as to cooking method"
+26131120,"POMPANO, BAKED OR BROILED, FAT ADDED IN COOKING","Pompano, baked or broiled, fat added in cooking"
+26131121,"POMPANO, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Pompano, baked or broiled, fat not added in cooking"
+26131130,"POMPANO, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Pompano, coated, baked or broiled, fat added in cooking"
+26131131,"POMPANO, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Pompano, coated, baked or broiled, fat not added in cooking"
+26131140,"POMPANO, COATED, FRIED","Pompano, coated, fried"
+26131160,"POMPANO, STEAMED OR POACHED","Pompano, steamed or poached"
+26131190,"POMPANO, SMOKED","Pompano, smoked"
+26133100,"PORGY, RAW","Porgy, raw"
+26133110,"PORGY, COOKED, NS AS TO COOKING METHOD","Porgy, cooked, NS as to cooking method"
+26133120,"PORGY, BAKED OR BROILED, FAT ADDED IN COOKING","Porgy, baked or broiled, fat added in cooking"
+26133121,"PORGY, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Porgy, baked or broiled, fat not added in cooking"
+26133130,"PORGY, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Porgy, coated, baked or broiled, fat added in cooking"
+26133131,"PORGY, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Porgy, coated, baked or broiled, fat not added in cooking"
+26133140,"PORGY, COATED, FRIED","Porgy, coated, fried"
+26133160,"PORGY, STEAMED OR POACHED","Porgy, steamed or poached"
+26135110,"RAY, COOKED, NS AS TO COOKING METHOD","Ray, cooked, NS as to cooking method"
+26135120,"RAY, BAKED OR BROILED, FAT ADDED IN COOKING","Ray, baked or broiled, fat added in cooking"
+26135121,"RAY, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Ray, baked or broiled, fat not added in cooking"
+26135130,"RAY, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Ray, coated, baked or broiled, fat added in cooking"
+26135131,"RAY, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Ray, coated, baked or broiled, fat not added in cooking"
+26135140,"RAY, COATED, FRIED","Ray, coated, fried"
+26135160,"RAY, STEAMED OR POACHED","Ray, steamed or poached"
+26137100,"SALMON, RAW","Salmon, raw"
+26137110,"SALMON, COOKED, NS AS TO COOKING METHOD","Salmon, cooked, NS as to cooking method"
+26137120,"SALMON, BAKED OR BROILED, MADE WITH OIL","Salmon, baked or broiled, made with oil"
+26137121,"SALMON, BAKED OR BROILED, MADE WITH BUTTER","Salmon, baked or broiled, made with butter"
+26137122,"SALMON, BAKED OR BROILED, MADE WITH MARGARINE","Salmon, baked or broiled, made with margarine"
+26137123,"SALMON, BAKED OR BROILED, MADE WITHOUT FAT","Salmon, baked or broiled, made without fat"
+26137124,"SALMON, BAKED OR BROILED, MADE WITH COOKING SPRAY","Salmon, baked or broiled, made with cooking spray"
+26137130,"SALMON, COATED, BAKED OR BROILED, MADE WITH OIL","Salmon, coated, baked or broiled, made with oil"
+26137131,"SALMON, COATED, BAKED OR BROILED, MADE WITH BUTTER","Salmon, coated, baked or broiled, made with butter"
+26137132,"SALMON, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Salmon, coated, baked or broiled, made with margarine"
+26137133,"SALMON, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Salmon, coated, baked or broiled, made without fat"
+26137134,"SALMON, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Salmon, coated, baked or broiled, made with cooking spray"
+26137140,"SALMON, COATED, FRIED, MADE WITH OIL","Salmon, coated, fried, made with oil"
+26137141,"SALMON, COATED, FRIED, MADE WITH BUTTER","Salmon, coated, fried, made with butter"
+26137142,"SALMON, COATED, FRIED, MADE WITH MARGARINE","Salmon, coated, fried, made with margarine"
+26137143,"SALMON, COATED, FRIED, MADE WITHOUT FAT","Salmon, coated, fried, made without fat"
+26137144,"SALMON, COATED, FRIED, MADE WITH COOKING SPRAY","Salmon, coated, fried, made with cooking spray"
+26137160,"SALMON, STEAMED OR POACHED","Salmon, steamed or poached"
+26137170,"SALMON, DRIED","Salmon, dried"
+26137180,"SALMON, CANNED","Salmon, canned"
+26137190,"SALMON, SMOKED (INCLUDE LOX)","Salmon, smoked"
+26139110,"SARDINES, COOKED","Sardines, cooked"
+26139170,"SARDINE, DRIED","Sardines, dried"
+26139180,"SARDINES, CANNED IN OIL","Sardines, canned in oil"
+26139190,"SARDINES, SKINLESS, BONELESS, PACKED IN WATER","Sardines, skinless, boneless, packed in water"
+26141110,"SEA BASS, COOKED, NS AS TO COOKING METHOD","Sea bass, cooked, NS as to cooking method"
+26141120,"SEA BASS, BAKED OR BROILED, FAT ADDED IN COOKING","Sea bass, baked or broiled, fat added in cooking"
+26141121,"SEA BASS, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Sea bass, baked or broiled, fat not added in cooking"
+26141130,"SEA BASS, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Sea bass, coated, baked or broiled, fat added in cooking"
+26141131,"SEA BASS, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Sea bass, coated, baked or broiled, fat not added in cooking"
+26141140,"SEA BASS, COATED, FRIED","Sea bass, coated, fried"
+26141160,"SEA BASS, STEAMED OR POACHED","Sea bass, steamed or poached"
+26141180,"SEA BASS, PICKLED (MERO EN ESCABECHE)","Sea bass, pickled (Mero en escabeche)"
+26143110,"SHARK, COOKED, NS AS TO COOKING METHOD","Shark, cooked, NS as to cooking method"
+26143120,"SHARK, BAKED OR BROILED, FAT ADDED IN COOKING","Shark, baked or broiled, fat added in cooking"
+26143121,"SHARK, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Shark, baked or broiled, fat not added in cooking"
+26143130,"SHARK, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Shark, coated, baked or broiled, fat added in cooking"
+26143131,"SHARK, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Shark, coated, baked or broiled, fat not added in cooking"
+26143140,"SHARK, COATED, FRIED","Shark, coated, fried"
+26143160,"SHARK, STEAMED OR POACHED","Shark, steamed or poached"
+26147110,"STURGEON, COOKED, NS AS TO COOKING METHOD","Sturgeon, cooked, NS as to cooking method"
+26147120,"STURGEON, BAKED OR BROILED, FAT ADDED IN COOKING","Sturgeon, baked or broiled, fat added in cooking"
+26147130,"STURGEON, STEAMED","Sturgeon, steamed"
+26147140,"STURGEON, COATED, FRIED","Sturgeon, coated, fried"
+26147190,"STURGEON, SMOKED","Sturgeon, smoked"
+26149110,"SWORDFISH, COOKED, NS AS TO COOKING METHOD","Swordfish, cooked, NS as to cooking method"
+26149120,"SWORDFISH, BAKED OR BROILED, FAT ADDED IN COOKING","Swordfish, baked or broiled, fat added in cooking"
+26149121,"SWORDFISH, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Swordfish, baked or broiled, fat not added in cooking"
+26149130,"SWORDFISH, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Swordfish, coated, baked or broiled, fat added in cooking"
+26149131,"SWORDFISH, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKIN","Swordfish, coated, baked or broiled, fat not added in cooking"
+26149140,"SWORDFISH, COATED, FRIED","Swordfish, coated, fried"
+26149160,"SWORDFISH, STEAMED OR POACHED","Swordfish, steamed or poached"
+26151110,"TROUT, COOKED, NS AS TO COOKING METHOD","Trout, cooked, NS as to cooking method"
+26151120,"TROUT, BAKED OR BROILED, MADE WITH OIL","Trout, baked or broiled, made with oil"
+26151121,"TROUT, BAKED OR BROILED, MADE WITH BUTTER","Trout, baked or broiled, made with butter"
+26151122,"TROUT, BAKED OR BROILED, MADE WITH MARGARINE","Trout, baked or broiled, made with margarine"
+26151123,"TROUT, BAKED OR BROILED, MADE WITHOUT FAT","Trout, baked or broiled, made without fat"
+26151124,"TROUT, BAKED OR BROILED, MADE WITH COOKING SPRAY","Trout, baked or broiled, made with cooking spray"
+26151130,"TROUT, COATED, BAKED OR BROILED, MADE WITH OIL","Trout, coated, baked or broiled, made with oil"
+26151131,"TROUT, COATED, BAKED OR BROILED, MADE WITH BUTTER","Trout, coated, baked or broiled, made with butter"
+26151132,"TROUT, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Trout, coated, baked or broiled, made with margarine"
+26151133,"TROUT, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Trout, coated, baked or broiled, made without fat"
+26151134,"TROUT, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Trout, coated, baked or broiled, made with cooking spray"
+26151140,"TROUT, COATED, FRIED, MADE WITH OIL","Trout, coated, fried, made with oil"
+26151141,"TROUT, COATED, FRIED, MADE WITH BUTTER","Trout, coated, fried, made with butter"
+26151142,"TROUT, COATED, FRIED, MADE WITH MARGARINE","Trout, coated, fried, made with margarine"
+26151143,"TROUT, COATED, FRIED, MADE WITHOUT FAT","Trout, coated, fried, made without fat"
+26151144,"TROUT, COATED, FRIED, MADE WITH COOKING SPRAY","Trout, coated, fried, made with cooking spray"
+26151160,"TROUT, STEAMED OR POACHED","Trout, steamed or poached"
+26151190,"TROUT, SMOKED","Trout, smoked"
+26153100,"TUNA, FRESH, RAW","Tuna, fresh, raw"
+26153110,"TUNA, FRESH, COOKED, NS AS TO COOKING METHOD","Tuna, fresh, cooked, NS as to cooking method"
+26153120,"TUNA, FRESH, BAKED OR BROILED, FAT ADDED IN COOKING","Tuna, fresh, baked or broiled, fat added in cooking"
+26153122,"TUNA, FRESH, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Tuna, fresh, baked or broiled, fat not added in cooking"
+26153130,"TUNA, FRESH, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Tuna, fresh, coated, baked or broiled, fat added in cooking"
+26153131,"TUNA, FRESH, COATED, BAKED OR BROILED, FAT NOT ADDED","Tuna, fresh, coated, baked or broiled, fat not added"
+26153140,"TUNA, FRESH, COATED, FRIED","Tuna, fresh, coated, fried"
+26153160,"TUNA, FRESH, STEAMED OR POACHED","Tuna, fresh, steamed or poached"
+26153170,"TUNA, FRESH, DRIED","Tuna, fresh, dried"
+26153190,"TUNA, FRESH, SMOKED","Tuna, fresh, smoked"
+26155110,"TUNA, CANNED, NS AS TO OIL OR WATER PACK","Tuna, canned, NS as to oil or water pack"
+26155180,"TUNA, CANNED, OIL PACK","Tuna, canned, oil pack"
+26155190,"TUNA, CANNED, WATER PACK","Tuna, canned, water pack"
+26157110,"WHITING, COOKED, NS AS TO COOKING METHOD","Whiting, cooked, NS as to cooking method"
+26157120,"WHITING, BAKED OR BROILED, MADE WITH OIL","Whiting, baked or broiled, made with oil"
+26157121,"WHITING, BAKED OR BROILED, MADE WITH BUTTER","Whiting, baked or broiled, made with butter"
+26157122,"WHITING, BAKED OR BROILED, MADE WITH MARGARINE","Whiting, baked or broiled, made with margarine"
+26157123,"WHITING, BAKED OR BROILED, MADE WITHOUT FAT","Whiting, baked or broiled, made without fat"
+26157124,"WHITING, BAKED OR BROILED, MADE WITH COOKING SPRAY","Whiting, baked or broiled, made with cooking spray"
+26157130,"WHITING, COATED, BAKED OR BROILED, MADE WITH OIL","Whiting, coated, baked or broiled, made with oil"
+26157131,"WHITING, COATED, BAKED OR BROILED, MADE WITH BUTTER","Whiting, coated, baked or broiled, made with butter"
+26157132,"WHITING, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Whiting, coated, baked or broiled, made with margarine"
+26157133,"WHITING, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Whiting, coated, baked or broiled, made without fat"
+26157134,"WHITING, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Whiting, coated, baked or broiled, made with cooking spray"
+26157140,"WHITING, COATED, FRIED, MADE WITH OIL","Whiting, coated, fried, made with oil"
+26157141,"WHITING, COATED, FRIED, MADE WITH BUTTER","Whiting, coated, fried, made with butter"
+26157142,"WHITING, COATED, FRIED, MADE WITH MARGARINE","Whiting, coated, fried, made with margarine"
+26157143,"WHITING, COATED, FRIED, MADE WITHOUT FAT","Whiting, coated, fried, made without fat"
+26157144,"WHITING, COATED, FRIED, MADE WITH COOKING SPRAY","Whiting, coated, fried, made with cooking spray"
+26157160,"WHITING, STEAMED OR POACHED","Whiting, steamed or poached"
+26158000,"TILAPIA, COOKED, NS AS TO COOKING METHOD","Tilapia, cooked, NS as to cooking method"
+26158010,"TILAPIA, BAKED OR BROILED, MADE WITH OIL","Tilapia, baked or broiled, made with oil"
+26158011,"TILAPIA, BAKED OR BROILED, MADE WITH BUTTER","Tilapia, baked or broiled, made with butter"
+26158012,"TILAPIA, BAKED OR BROILED, MADE WITH MARGARINE","Tilapia, baked or broiled, made with margarine"
+26158013,"TILAPIA, BAKED OR BROILED, MADE WITHOUT FAT","Tilapia, baked or broiled, made without fat"
+26158014,"TILAPIA, BAKED OR BROILED, MADE WITH COOKING SPRAY","Tilapia, baked or broiled, made with cooking spray"
+26158020,"TILAPIA, COATED, BAKED OR BROILED, MADE  WITH OIL","Tilapia, coated, baked or broiled, made  with oil"
+26158021,"TILAPIA, COATED, BAKED OR BROILED, MADE WITH BUTTER","Tilapia, coated, baked or broiled, made with butter"
+26158022,"TILAPIA, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Tilapia, coated, baked or broiled, made with margarine"
+26158023,"TILAPIA, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Tilapia, coated, baked or broiled, made without fat"
+26158024,"TILAPIA, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Tilapia, coated, baked or broiled, made with cooking spray"
+26158030,"TILAPIA, COATED, FRIED, MADE WITH OIL","Tilapia, coated, fried, made with oil"
+26158031,"TILAPIA, COATED, FRIED, MADE WITH BUTTER","Tilapia, coated, fried, made with butter"
+26158032,"TILAPIA, COATED, FRIED, MADE WITH MARGARINE","Tilapia, coated, fried, made with margarine"
+26158033,"TILAPIA, COATED, FRIED, MADE WITHOUT FAT","Tilapia, coated, fried, made without fat"
+26158034,"TILAPIA, COATED, FRIED, MADE WITH COOKING SPRAY","Tilapia, coated, fried, made with cooking spray"
+26158050,"TILAPIA, STEAMED OR POACHED","Tilapia, steamed or poached"
+26203110,"FROG LEGS, NS AS TO COOKING METHOD","Frog legs, NS as to cooking method"
+26203160,"FROG LEGS, STEAMED","Frog legs, steamed"
+26205110,"OCTOPUS, COOKED, NS AS TO COOKING METHOD","Octopus, cooked, NS as to cooking method"
+26205160,"OCTOPUS, STEAMED","Octopus, steamed"
+26205170,"OCTOPUS, DRIED","Octopus, dried"
+26205180,"OCTOPUS, DRIED, BOILED","Octopus, dried, boiled"
+26205190,"OCTOPUS, SMOKED","Octopus, smoked"
+26207110,"ROE, SHAD, COOKED (INCL COD ROE)","Roe, shad, cooked"
+26209100,"ROE, HERRING","Roe, herring"
+26211100,"ROE, STURGEON (INCLUDE CAVIAR)","Roe, sturgeon"
+26213100,"SQUID, RAW","Squid, raw"
+26213120,"SQUID, BAKED, BROILED, FAT ADDED IN COOKING","Squid, baked or broiled, fat added in cooking"
+26213121,"SQUID, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Squid, baked or broiled, fat not added in cooking"
+26213130,"SQUID, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Squid, coated, baked or broiled, fat added in cooking"
+26213131,"SQUID, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Squid, coated, baked or broiled, fat not added in cooking"
+26213140,"SQUID, COATED, FRIED","Squid, coated, fried"
+26213160,"SQUID, STEAMED OR BOILED","Squid, steamed or boiled"
+26213170,"SQUID, DRIED","Squid, dried"
+26213180,"SQUID, PICKLED","Squid, pickled"
+26213190,"SQUID, CANNED","Squid, canned"
+26215120,"TURTLE, COOKED, NS AS TO METHOD","Turtle (terrapin), cooked, NS as to cooking method"
+26301110,"ABALONE, COOKED, NS AS TO COOKING METHOD","Abalone, cooked, NS as to cooking method"
+26301140,"ABALONE, FLOURED OR BREADED, FRIED","Abalone, floured or breaded, fried"
+26301160,"ABALONE, STEAMED OR POACHED","Abalone, steamed or poached"
+26303100,"CLAMS, RAW","Clams, raw"
+26303110,"CLAMS, COOKED, NS AS TO COOKING METHOD","Clams, cooked, NS as to cooking method"
+26303120,"CLAMS, BAKED OR BROILED, FAT ADDED IN COOKING","Clams, baked or broiled, fat added in cooking"
+26303121,"CLAMS, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Clams, baked or broiled, fat not added in cooking"
+26303130,"CLAMS, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Clams, coated, baked or broiled, fat added in cooking"
+26303131,"CLAMS ,BAKED OR BROILED, FAT NOT ADDED IN COOKING","Clams, coated, baked or broiled, fat not added in cooking"
+26303140,"CLAMS,COATED, FRIED","Clams, coated, fried"
+26303160,"CLAMS, STEAMED OR BOILED","Clams, steamed or boiled"
+26303180,"CLAMS, CANNED","Clams, canned"
+26303190,"CLAMS, SMOKED, IN OIL","Clams, smoked, in oil"
+26305110,"CRAB, COOKED, NS AS TO COOKING METHOD","Crab, cooked, NS as to cooking method"
+26305120,"CRAB, BAKED OR BROILED, FAT ADDED IN COOKING","Crab, baked or broiled, fat added in cooking"
+26305121,"CRAB, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Crab, baked or broiled, fat not added in cooking"
+26305130,"CRAB, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Crab, coated, baked or broiled, fat added in cooking"
+26305131,"CRAB, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Crab, coated, baked or broiled, fat not added in cooking"
+26305160,"CRAB, HARD SHELL, STEAMED","Crab, hard shell, steamed"
+26305180,"CRAB, CANNED","Crab, canned"
+26307140,"CRAB, SOFT SHELL, COATED, FRIED","Crab, soft shell, coated, fried"
+26309140,"CRAYFISH, COATED, FRIED","Crayfish, coated, fried"
+26309160,"CRAYFISH, BOILED OR STEAMED","Crayfish, boiled or steamed"
+26311110,"LOBSTER, COOKED, NS AS TO METHOD","Lobster, cooked, NS as to cooking method"
+26311120,"LOBSTER, BAKED OR BROILED, FAT ADDED IN COOKING","Lobster, baked or broiled, fat added in cooking"
+26311121,"LOBSTER, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Lobster, baked or broiled, fat not added in cooking"
+26311140,"LOBSTER, COATED, FRIED","Lobster, coated, fried"
+26311160,"LOBSTER, STEAMED OR BOILED","Lobster, steamed or boiled"
+26311170,"LOBSTER, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Lobster, coated, baked or broiled, fat added in cooking"
+26311171,"LOBSTER, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Lobster, coated, baked or broiled, fat not added in cooking"
+26311180,"LOBSTER, CANNED","Lobster, canned"
+26313100,"MUSSELS, RAW","Mussels, raw"
+26313110,"MUSSELS, COOKED, NS AS TO COOKING METHOD","Mussels, cooked, NS as to cooking method"
+26313160,"MUSSELS, STEAMED","Mussels, steamed or poached"
+26315100,"OYSTERS, RAW","Oysters, raw"
+26315110,"OYSTERS, COOKED, NS AS TO COOKING METHOD","Oysters, cooked, NS as to cooking method"
+26315120,"OYSTERS, BAKED OR BROILED, FAT ADDED IN COOKING","Oysters, baked or broiled, fat added in cooking"
+26315121,"OYSTERS, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Oysters, baked or broiled, fat not added in cooking"
+26315130,"OYSTERS, STEAMED","Oysters, steamed"
+26315140,"OYSTERS, COATED, FRIED","Oysters, coated, fried"
+26315160,"OYSTERS, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Oysters, coated, baked or broiled, fat added in cooking"
+26315161,"OYSTERS, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Oysters, coated, baked or broiled, fat not added in cooking"
+26315180,"OYSTERS, CANNED","Oysters, canned"
+26315190,"OYSTERS, SMOKED","Oysters, smoked"
+26317110,"SCALLOPS, COOKED, NS AS TO COOKING METHOD","Scallops, cooked, NS as to cooking method"
+26317120,"SCALLOPS, BAKED OR BROILED, FAT ADDED IN COOKING","Scallops, baked or broiled, fat added in cooking"
+26317121,"SCALLOPS, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Scallops, baked or broiled, fat not added in cooking"
+26317130,"SCALLOPS, STEAMED OR BOILED","Scallops, steamed or boiled"
+26317140,"SCALLOPS, COATED, FRIED","Scallops, coated, fried"
+26317160,"SCALLOPS, COATED, BAKED OR BROILED, FAT ADDED IN COOKING","Scallops, coated, baked or broiled, fat added in cooking"
+26317161,"SCALLOPS, COATED, BAKED OR BROILED, FAT NOT ADDED IN COOKING","Scallops, coated, baked or broiled, fat not added in cooking"
+26319110,"SHRIMP, COOKED, NS AS TO COOKING METHOD","Shrimp, cooked, NS as to cooking method"
+26319120,"SHRIMP, BAKED OR BROILED, MADE WITH OIL","Shrimp, baked or broiled, made with oil"
+26319121,"SHRIMP, BAKED OR BROILED, MADE WITH BUTTER","Shrimp, baked or broiled, made with butter"
+26319122,"SHRIMP, BAKED OR BROILED, MADE WITH MARGARINE","Shrimp, baked or broiled, made with margarine"
+26319123,"SHRIMP, BAKED OR BROILED, MADE WITHOUT FAT","Shrimp, baked or broiled, made without fat"
+26319124,"SHRIMP, BAKED OR BROILED, MADE WITH COOKING SPRAY","Shrimp, baked or broiled, made with cooking spray"
+26319130,"SHRIMP, STEAMED OR BOILED","Shrimp, steamed or boiled"
+26319140,"SHRIMP, COATED, FRIED, MADE WITH OIL","Shrimp, coated, fried, made with oil"
+26319141,"SHRIMP, COATED, FRIED, MADE WITH BUTTER","Shrimp, coated, fried, made with butter"
+26319142,"SHRIMP, COATED, FRIED, MADE WITH MARGARINE","Shrimp, coated, fried, made with margarine"
+26319143,"SHRIMP, COATED, FRIED, MADE WITHOUT FAT","Shrimp, coated, fried, made without fat"
+26319144,"SHRIMP, COATED, FRIED, MADE WITH COOKING SPRAY","Shrimp, coated, fried, made with cooking spray"
+26319145,"SHRIMP, COATED, FRIED, FROM FAST FOOD / RESTAURANT","Shrimp, coated, fried, from fast food / restaurant"
+26319160,"SHIRMP, COATED, BAKED OR BROILED, MADE WITH OIL","Shrimp, coated, baked or broiled, made with oil"
+26319161,"SHRIMP, COATED, BAKED OR BROILED, MADE WITH BUTTER","Shrimp, coated, baked or broiled, made with butter"
+26319162,"SHRIMP, COATED, BAKED OR BROILED, MADE WITH MARGARINE","Shrimp, coated, baked or broiled, made with margarine"
+26319163,"SHRIMP, COATED, BAKED OR BROILED, MADE WITHOUT FAT","Shrimp, coated, baked or broiled, made without fat"
+26319164,"SHRIMP, COATED, BAKED OR BROILED, MADE WITH COOKING SPRAY","Shrimp, coated, baked or broiled, made with cooking spray"
+26319170,"SHRIMP, DRIED","Shrimp, dried"
+26319180,"SHRIMP, CANNED","Shrimp, canned"
+26321110,"SNAILS, COOKED, NS AS TO METHOD","Snails, cooked, NS as to cooking method"
+27111000,"BEEF W/ TOMATO-BASED SAUCE (MIXTURE)","Beef with tomato-based sauce (mixture)"
+27111050,"SPAGHETTI SAUCE W/ BEEF/MEAT, HOMEMADE-STYLE","Spaghetti sauce with beef or meat other than lamb or mutton, homemade-style"
+27111100,"BEEF GOULASH","Beef goulash"
+27111200,"BEEF BURGUNDY (BEEF BOURGUIGNONNE)","Beef burgundy (beef bourguignonne)"
+27111300,"MEXICAN BEEF STEW, NO POTATOES, TOMATO SAUCE","Mexican style beef stew, no potatoes, tomato-based sauce (mixture) (Carne guisada sin papas)"
+27111310,"MEXICAN BEEF STEW, NO POTATOES, W/ CHILI PEPPERS, TOMATO SCE","Mexican style beef stew, no potatoes, with chili peppers, tomato-based sauce (mixture) (Carne guisada con chile)"
+27111400,"CHILI CON CARNE, NS AS TO BEANS","Chili con carne, NS as to beans"
+27111410,"CHILI CON CARNE W/ BEANS","Chili con carne with beans"
+27111420,"CHILI CON CARNE W/O BEANS","Chili con carne without beans"
+27111430,"CHILI CON CARNE, NS AS TO BEANS, W/ CHEESE","Chili con carne, NS as to beans, with cheese"
+27111440,"CHILI CON CARNE W/ BEANS & CHEESE","Chili con carne with beans and cheese"
+27111500,"BEEF SLOPPY JOE (NO BUN)","Beef sloppy joe (no bun)"
+27112000,"BEEF W/ GRAVY (MIXTURE) (INCLUDE COUNTRY STYLE)","Beef with gravy (mixture)"
+27112010,"SALISBURY STEAK W/ GRAVY (MIXTURE)","Salisbury steak with gravy (mixture)"
+27113000,"BEEF W/ CREAM OR WHITE SAUCE (MIXTURE)","Beef with cream or white sauce (mixture)"
+27113100,"BEEF STROGANOFF","Beef stroganoff"
+27113200,"CREAMED CHIPPED OR DRIED BEEF","Creamed chipped or dried beef"
+27113300,"SWEDISH MEATBALLS W/ CREAM OR WHITE SAUCE (MIXTURE)","Swedish meatballs with cream or white sauce (mixture)"
+27114000,"BEEF W/ (MUSHROOM) SOUP (MIXTURE)","Beef with (mushroom) soup (mixture)"
+27115000,"BEEF W/ SOY-BASED SAUCE (MIXTURE)","Beef with soy-based sauce (mixture)"
+27115100,"STEAK TERIYAKI W/ SAUCE (MIXTURE)","Steak teriyaki with sauce (mixture)"
+27116100,"BEEF CURRY","Beef curry"
+27116200,"BEEF W/ BARBECUE SAUCE (MIXTURE)","Beef with barbecue sauce (mixture)"
+27116300,"BEEF W/ SWEET & SOUR SAUCE (MIXTURE)","Beef with sweet and sour sauce (mixture)"
+27116350,"STEWED, SEASONED GROUND BEEF, MEXICAN","Stewed, seasoned, ground beef, Mexican style (Picadillo de carne de rez)"
+27116400,"STEAK TARTARE (RAW GROUND BEEF & EGG)","Steak tartare (raw ground beef and egg)"
+27118110,"MEATBALLS, P. R. (ALBONDIGAS GUISADAS)","Meatballs, Puerto Rican style (Albondigas guisadas)"
+27118120,"STEWED,SEASONED GROUND BEEF,PUERTO RICAN STYLE","Stewed seasoned ground beef, Puerto Rican style (Picadillo guisado, picadillo de carne)"
+27118130,"STEWED DRIED BEEF, P.R. (TASAJO GUISADO)","Stewed dried beef, Puerto Rican style (Tasajo guisado, carne cecina guisada)"
+27118140,"STUFFED POT ROAST, P.R.,NFS(ASSUME GRAVY,STUFFING)","Stuffed pot roast, Puerto Rican style, NFS (assume with gravy and stuffing)"
+27118180,"BEEF STEW, P.R., MEAT W/ GRAVY (POTATO SEPARATE)","Puerto Rican style beef stew, meat with gravy (potatoes reported separately)"
+27120020,"HAM/PORK W/ GRAVY (MIXTURE)","Ham or pork with gravy (mixture)"
+27120030,"HAM/PORK W/ BARBECUE SAUCE","Ham or pork with barbecue sauce (mixture)"
+27120060,"SWEET & SOUR PORK","Sweet and sour pork"
+27120080,"HAM STROGANOFF (INCL HAM W/ CREAM OR WHITE SAUCE)","Ham stroganoff"
+27120090,"HAM/PORK W/ (MUSHROOM) SOUP-BASE SAUCE (MIXTURE)","Ham or pork with (mushroom) soup (mixture)"
+27120100,"HAM/PORK W/ TOMATO-BASED SAUCE (MIXTURE)","Ham or pork with tomato-based sauce (mixture)"
+27120110,"SAUSAGE W/ TOMATO-BASED SAUCE (MIXTURE)","Sausage with tomato-based sauce (mixture)"
+27120120,"SAUSAGE GRAVY","Sausage gravy"
+27120130,"MEXICAN STYLE PORK STEW,NO POT,TOM-BASE SCE(MIXTUR","Mexican style pork stew, no potatoes, tomato-based sauce (mixture) (cerdo guisado sin papas)"
+27120150,"PORK OR HAM W/ SOY-BASED SAUCE (MIXTURE)","Pork or ham with soy-based sauce (mixture)"
+27120160,"PORK CURRY","Pork curry"
+27120210,"FRANKFURTER /HOT DOG,W/CHILI,NO BUN (INCL CHILI DOG,NO BUN)","Frankfurter or hot dog, with chili, no bun"
+27120250,"FRANKFURTERS/HOT DOGS W/ TOM-BASED SCE (MIXTURE)","Frankfurters or hot dogs with tomato-based sauce (mixture)"
+27121000,"PORK W/ CHILE & TOM (MIXTURE) (PUERCO CON CHILE)","Pork with chili and tomatoes (mixture) (Puerco con chile)"
+27121010,"STEWED PORK, P.R.","Stewed pork, Puerto Rican style"
+27121410,"CHILI CON CARNE W/ BEANS, MADE W/ PORK","Chili con carne with beans, made with pork"
+27130010,"LAMB W/ GRAVY (MIXTURE)","Lamb or mutton with gravy (mixture)"
+27130040,"SPAGHETTI SAUCE W/ LAMB, HOMEMADE-STYLE","Spaghetti sauce with lamb or mutton, homemade-style"
+27130050,"LAMB GOULASH","Lamb or mutton goulash"
+27130100,"LAMB OR MUTTON CURRY","Lamb or mutton curry"
+27133010,"STEWED GOAT, P.R. (CABRITO EN FRICASE)","Stewed goat, Puerto Rican style (Cabrito en fricase, chilindron de chivo)"
+27135010,"VEAL W/ GRAVY (MIXTURE)","Veal with gravy (mixture)"
+27135020,"VEAL SCALLOPINI","Veal scallopini"
+27135030,"VEAL W/ CREAM SAUCE (INCLUDE VEAL PAPRIKASH)","Veal with cream sauce (mixture)"
+27135040,"VEAL W/ BUTTER SAUCE","Veal with butter sauce (mixture)"
+27135050,"VEAL MARSALA","Veal Marsala"
+27135110,"VEAL PARMIGIANA","Veal parmigiana"
+27135150,"VEAL CORDON BLEU","Veal cordon bleu"
+27136050,"VENISON/DEER W/ TOMATO-BASED SAUCE (MIXTURE)","Venison/deer with tomato-based sauce (mixture)"
+27136080,"VENISON/DEER W/ GRAVY","Venison/deer with gravy (mixture)"
+27136100,"CHILI CON CARNE W/ VENISON/DEER & BEANS","Chili con carne with venison/deer and beans"
+27141000,"CHICKEN CACCIATORE (INCLUDE CHICKEN W/TOMATO SAUCE)","Chicken or turkey cacciatore"
+27141030,"SPAGHETTI SAUCE W/ POULTRY, HOMEMADE","Spaghetti sauce with poultry, home-made style"
+27141050,"STEWED CHICKEN W/ TOMATO SAUCE, MEXICAN STYLE","Stewed chicken with tomato-based sauce, Mexican style (mixture) (Pollo guisado con tomate)"
+27141500,"CHILI CON CARNE W/ CHICKEN & BEANS","Chili con carne with chicken or turkey and beans"
+27142000,"CHICKEN W/ GRAVY (MIXTURE)","Chicken with gravy (mixture)"
+27142100,"CHICKEN FRICASSEE","Chicken or turkey fricassee"
+27142200,"TURKEY W/ GRAVY (MIXTURE)","Turkey with gravy (mixture)"
+27143000,"CHICKEN OR TURKEY W/ CREAM SAUCE (MIXTURE)","Chicken or turkey with cream sauce (mixture)"
+27144000,"CHICKEN W/ (MUSHROOM) SOUP-BASED SAUCE (MIXTURE)","Chicken or turkey with (mushroom) soup (mixture)"
+27145000,"CHICKEN TERIYAKI","Chicken or turkey teriyaki (chicken or turkey with soy-based sauce)"
+27146000,"CHICKEN OR TURKEY W/ BBQ SAUCE, SKIN EATEN","Chicken or turkey with barbecue sauce, skin eaten"
+27146010,"CHICKEN OR TURKEY W/ BBQ SAUCE, SKIN NOT EATEN","Chicken or turkey with barbecue sauce, skin not eaten"
+27146050,"CHICKEN WING W/ HOT PEPPER SCE (INCL BUFFALO WING)","Chicken wing with hot pepper sauce"
+27146100,"SWEET & SOUR CHICKEN","Sweet and sour chicken or turkey"
+27146110,"SWEET AND SOUR CHICKEN OR TURKEY, WITHOUT VEGETABLES","Sweet and sour chicken or turkey, without vegetables"
+27146150,"CHICKEN CURRY","Chicken curry"
+27146160,"CHICKEN WITH MOLE SAUCE","Chicken with mole sauce"
+27146200,"CHICKEN W/ CHEESE SAUCE (MIXTURE)","Chicken or turkey with cheese sauce (mixture)"
+27146250,"CHICKEN CORDON BLEU","Chicken or turkey cordon bleu"
+27146300,"CHICKEN PARMIGIANA","Chicken or turkey parmigiana"
+27146350,"ORANGE CHICKEN","Orange chicken"
+27146360,"SESAME CHICKEN","Sesame chicken"
+27146400,"CHICKEN KIEV","Chicken kiev"
+27148010,"STUFFED CHICKEN, DRUMSTICK OR BREAST, P.R.","Stuffed chicken, drumstick or breast, Puerto Rican style (Muslo de pollo o pechuga rellena)"
+27150010,"FISH W/ CREAM OR WHITE SAUCE, NOT TUNA OR LOBSTER","Fish with cream or white sauce, not tuna or lobster (mixture)"
+27150020,"CRAB, DEVILED","Crab, deviled"
+27150030,"CRAB IMPERIAL (INCLUDE STUFFED CRAB)","Crab imperial"
+27150050,"FISH TIMBALE OR MOUSSE","Fish timbale or mousse"
+27150060,"LOBSTER NEWBURG (INCLUDE LOBSTER THERMIDOR)","Lobster newburg"
+27150070,"LOBSTER W/ BUTTER SAUCE (INCLUDE LOBSTER NORFOLK)","Lobster with butter sauce (mixture)"
+27150100,"SHRIMP CURRY","Shrimp curry"
+27150110,"SHRIMP COCKTAIL (SHRIMP W/ COCKTAIL SAUCE)","Shrimp cocktail (shrimp with cocktail sauce)"
+27150120,"TUNA W/ CREAM OR WHITE SAUCE (MIXTURE)","Tuna with cream or white sauce (mixture)"
+27150130,"SEAFOOD NEWBURG (INCLUDE CRABMEAT THERMIDOR)","Seafood newburg"
+27150140,"CLAM SAUCE, WHITE","Clam sauce, white"
+27150160,"SHRIMP W/ LOBSTER SAUCE (MIXTURE)","Shrimp with lobster sauce (mixture)"
+27150170,"SWEET & SOUR SHRIMP","Sweet and sour shrimp"
+27150190,"LOBSTER SAUCE (BROTH-BASED)","Lobster sauce (broth-based)"
+27150200,"OYSTER SCE (WHITE SCE-BASED)","Oyster sauce (white sauce-based)"
+27150210,"FISH SAUCE (BAGOONG)","Fish sauce (bagoong)"
+27150230,"SHRIMP SCAMPI","Shrimp scampi"
+27150250,"FISH MOOCHIM (KOREAN STYLE), DRIED FISH W/ SOY SCE","Fish moochim (Korean style), dried fish with soy sauce"
+27150310,"FISH W/ TOMATO-BASED SAUCE (MIXTURE)","Fish with tomato-based sauce (mixture)"
+27150320,"FISH CURRY","Fish curry"
+27150330,"MUSSELS W/ TOMATO-BASED SAUCE (MIXTURE)","Mussels with tomato-based sauce (mixture)"
+27150350,"SARDINES W/ TOMATO-BASED SAUCE (MIXTURE)","Sardines with tomato-based sauce (mixture)"
+27150370,"SARDINES W/ MUSTARD SAUCE (MIXTURE)","Sardines with mustard sauce (mixture)"
+27150410,"SHRIMP TERIYAKI","Shrimp teriyaki (shrimp with soy-based sauce) (mixture)"
+27150510,"SCALLOPS W/ CHEESE SAUCE (MIXTURE)","Scallops with cheese sauce (mixture)"
+27151030,"MARINATED FISH (CEVICHE)","Marinated fish (Ceviche)"
+27151040,"CRABS IN TOMATO-BASED SAUCE, PUERTO RICAN STYLE","Crabs in tomato-based sauce, Puerto Rican style (mixture) (Salmorejo de jueyes)"
+27151050,"SHRIMP IN GARLIC SAUCE, P.R. (CAMARONES AL AJILLO)","Shrimp in garlic sauce, Puerto Rican style (mixture) (Camarones al ajillo)"
+27151070,"STEWED CODFISH, PUERTO RICAN STYLE, NO POTATOES","Stewed codfish, Puerto Rican style, no potatoes (potatoes reported separately)"
+27160010,"MEAT W/ BARBECUE SAUCE, NS AS TO TYPE OF MEAT","Meat with barbecue sauce, NS as to type of meat (mixture)"
+27160100,"MEATBALLS, NS AS TO TYPE OF MEAT, W/ SAUCE","Meatballs, NS as to type of meat, with sauce (mixture)"
+27161010,"MEAT LOAF, P.R. (ALBONDIGON)","Puerto Rican style meat loaf (Albondigon)"
+27162010,"MEAT W/ TOMATO-BASED SAUCE","Meat with tomato-based sauce (mixture)"
+27162050,"SPAGHETTI SAUCE W/ COMBINATION OF MEATS, HOMEMADE","Spaghetti sauce with combination of meats, homemade-style"
+27162060,"SPAGHETTI SAUCE W/ MEAT & VEGETABLES, HOMEMADE-STYLE","Spaghetti sauce with meat and vegetables, homemade-style"
+27162500,"STEWED SEASONED GROUND BEEF & PORK, MEXICAN","Stewed, seasoned, ground beef and pork, Mexican style (Picadillo de carne de rez y puerco)"
+27163010,"MEAT W/ GRAVY, NS AS TO TYPE OF MEAT (MIXTURE)","Meat with gravy, NS as to type of meat (mixture)"
+27211000,"BEEF & POTATOES, NO SAUCE (MIXTURE)","Beef and potatoes, no sauce (mixture)"
+27211100,"BEEF STEW W/ POTATOES, TOMATO-BASED SAUCE","Beef stew with potatoes, tomato-based sauce (mixture)"
+27211110,"MEXICAN BEEF STEW W/POT,TOM SCE (CARNE GUISADA CON)","Mexican style beef stew with potatoes, tomato-based sauce (mixture) (Carne guisada con papas)"
+27211150,"BEEF GOULASH W/ POTATOES (INCL BEEF GOULASH, NFS)","Beef goulash with potatoes"
+27211190,"BEEF & POTATOES W/ CREAM, WHITE, MUSHROOM SOUP SCE (MIXTURE)","Beef and potatoes with cream sauce, white sauce or mushroom soup-based sauce (mixture)"
+27211200,"BEEF STEW W/ POTATOES, GRAVY","Beef stew with potatoes, gravy"
+27211300,"BEEF (ROAST) HASH","Beef (roast) hash"
+27211400,"CORNED BEEF HASH","Corned beef hash"
+27211500,"BEEF & POTATOES W/ CHEESE SAUCE (MIXTURE)","Beef and potatoes with cheese sauce (mixture)"
+27211550,"STEWED SEASONED GROUND BEEF W/ POTATOES, MEXICAN","Stewed, seasoned, ground beef with potatoes, Mexican style (Picadillo de carne de rez con papas)"
+27212000,"BEEF & NOODLES, NO SAUCE","Beef and noodles, no sauce (mixture)"
+27212050,"BEEF & MACARONI WITH CHEESE SAUCE (MIXTURE)","Beef and macaroni with cheese sauce (mixture)"
+27212100,"BEEF & NOODLES W/ TOMATO-BASED SAUCE (MIXTURE)","Beef and noodles with tomato-based sauce (mixture)"
+27212120,"CHILI CON CARNE W/ BEANS & MACARONI","Chili con carne with beans and macaroni"
+27212150,"BEEF GOULASH W/ NOODLES","Beef goulash with noodles"
+27212200,"BEEF & NOODLES W/ GRAVY (MIXTURE)","Beef and noodles with gravy (mixture)"
+27212300,"BEEF & NOODLES W/ CREAM OR WHITE SAUCE (MIXTURE)","Beef and noodles with cream or white sauce (mixture)"
+27212350,"BEEF STROGANOFF W/ NOODLES","Beef stroganoff with noodles"
+27212400,"BEEF & NOODLES W/ (MUSHROOM) SOUP (MIXTURE)","Beef and noodles with (mushroom) soup (mixture)"
+27212500,"BEEF AND NOODLES WITH SOY-BASED SAUCE (MIXTURE)","Beef and noodles with soy-based sauce (mixture)"
+27213000,"BEEF & RICE, NO SAUCE (MIXTURE)","Beef and rice, no sauce (mixture)"
+27213010,"BIRYANI WITH MEAT","Biryani with meat"
+27213100,"BEEF & RICE W/ TOMATO-BASED SAUCE (MIXTURE)","Beef and rice with tomato-based sauce (mixture)"
+27213120,"PORCUPINE BALLS W/ TOMATO-BASED SAUCE (MIXTURE)","Porcupine balls with tomato-based sauce (mixture)"
+27213150,"CHILI CON CARNE W/ BEANS & RICE","Chili con carne with beans and rice"
+27213200,"BEEF & RICE W/ GRAVY (MIXTURE)","Beef and rice with gravy (mixture)"
+27213300,"BEEF & RICE W/ CREAM SAUCE (MIXTURE)","Beef and rice with cream sauce (mixture)"
+27213400,"BEEF & RICE W/ (MUSHROOM) SOUP (MIXTURE)","Beef and rice with (mushroom) soup (mixture)"
+27213420,"PORCUPINE BALLS W/ (MUSHROOM) SOUP (MIXTURE)","Porcupine balls with (mushroom) soup (mixture)"
+27213500,"BEEF & RICE W/ SOY-BASED SAUCE (MIXTURE)","Beef and rice with soy-based sauce (mixture)"
+27213600,"BEEF AND RICE WITH CHEESE SAUCE (MIXTURE)","Beef and rice with cheese sauce (mixture)"
+27214100,"MEAT LOAF MADE W/ BEEF","Meat loaf made with beef"
+27214110,"MEAT LOAF W/ BEEF, W/ TOMATO SAUCE","Meat loaf made with beef, with tomato-based sauce"
+27214300,"BEEF WELLINGTON","Beef wellington"
+27214500,"CORNED BEEF PATTY","Corned beef patty"
+27214600,"CREAMED DRIED BEEF ON TOAST","Creamed dried beef on toast"
+27218110,"STUFFED POT ROAST (LARDED MEAT) W/ POTATOES, P.R.","Puerto Rican style stuffed pot roast (larded meat) with potatoes (Carne mechada con papas boliche)"
+27218210,"BEEF STEW, P.R. W/ POTATOES (CARNE GUISADA CON PAPAS)","Puerto Rican style beef stew with potatoes (Carne guisada con papas)"
+27218310,"STEWED CORNED BEEF, P.R. (""CORNED BEEF"" GUISADO)","Stewed corned beef, Puerto Rican style (""Corned beef"" guisado)"
+27220010,"MEAT LOAF MADE W/ HAM (NOT LUNCHEON MEAT)","Meat loaf made with ham (not luncheon meat)"
+27220020,"HAM & NOODLES W/ CREAM OR WHITE SAUCE (MIXTURE)","Ham and noodles with cream or white sauce (mixture)"
+27220030,"HAM & RICE W/ (MUSHROOM) SOUP (MIXTURE)","Ham and rice with (mushroom) soup (mixture)"
+27220050,"HAM OR PORK  W/ STUFFING","Ham or pork with stuffing (mixture)"
+27220080,"HAM CROQUETTE","Ham croquette"
+27220110,"PORK & RICE W/ TOMATO-BASED SAUCE (MIXTURE)","Pork and rice with tomato-based sauce (mixture)"
+27220120,"SAUSAGE & RICE W/ TOMATO-BASED SAUCE (MIXTURE)","Sausage and rice with tomato-based sauce (mixture)"
+27220150,"SAUSAGE & RICE W/ (MUSHROOM) SOUP (MIXTURE)","Sausage and rice with (mushroom) soup (mixture)"
+27220170,"SAUSAGE & RICE W/ CHEESE SAUCE (MIXTURE)","Sausage and rice with cheese sauce (mixture)"
+27220190,"SAUSAGE & NOODLES W/ CREAM OR WHITE SAUCE (MIXTURE)","Sausage and noodles with cream or white sauce (mixture)"
+27220210,"HAM & NOODLES, NO SAUCE (MIXTURE)","Ham and noodles, no sauce (mixture)"
+27220310,"HAM & RICE, NO SAUCE (MIXTURE)","Ham or pork and rice, no sauce (mixture)"
+27220510,"HAM/PORK & POTATOES W/ GRAVY (MIXTURE)","Ham or pork and potatoes with gravy (mixture)"
+27220520,"PORK & POTATOES W/ CHEESE SAUCE","Ham or pork and potatoes with cheese sauce (mixture)"
+27221100,"STEWED PIG'S FEET, P.R. (PATITAS DE CERDO GUISADAS)","Stewed pig's feet, Puerto Rican style (Patitas de cerdo guisadas)"
+27221110,"STUFFED PORK ROAST, P.R.","Stuffed pork roast, Puerto Rican style"
+27221150,"MEXICAN STYLE PORK STEW W/POT,TOM-BASE SCE(MIXTURE)","Mexican style pork stew, with potatoes, tomato-based sauce (mixture) (cerdo guisado con papas)"
+27230010,"LAMB LOAF","Lamb or mutton loaf"
+27231000,"LAMB OR MUTTON & POTATOES W/ GRAVY (MIXTURE)","Lamb or mutton and potatoes with gravy (mixture)"
+27232000,"LAMB & POTATOES W/ TOMATO-BASED SAUCE (MIXTURE)","Lamb or mutton and potatoes with tomato-based sauce (mixture)"
+27233000,"LAMB OR MUTTON & NOODLES W/ GRAVY (MIXTURE)","Lamb or mutton and noodles with gravy (mixture)"
+27235000,"MEAT LOAF MADE WITH VENISON/DEER","Meat loaf made with venison/deer"
+27235750,"VEAL & NOODLES W/ CREAM/WHITE SCE (MIXTURE)","Veal and noodles with cream or white sauce (mixture)"
+27236000,"VENISON/DEER & NOODLE MIXTURE W/ CREAM/WHITE SAUCE","Venison/deer and noodles with cream or white sauce (mixture)"
+27241000,"CHICKEN OR TURKEY HASH","Chicken or turkey hash"
+27241010,"CHICKEN OR TURKEY & POTATOES W/ GRAVY (MIXTURE)","Chicken or turkey and potatoes with gravy (mixture)"
+27242000,"CHICKEN OR TURKEY & NOODLES, NO SAUCE (MIXTURE)","Chicken or turkey and noodles, no sauce (mixture)"
+27242200,"CHICKEN OR TURKEY & NOODLES W/ GRAVY (MIXTURE)","Chicken or turkey and noodles with gravy (mixture)"
+27242250,"CHICKEN OR TURKEY & NOODLES W/ (MUSHROOM) SOUP","Chicken or turkey and noodles with (mushroom) soup (mixture)"
+27242300,"CHICKEN OR TURKEY & NOODLES W/ CREAM OR WHITE SAUCE","Chicken or turkey and noodles with cream or white sauce (mixture)"
+27242310,"CHICKEN & NOODLES W/ CHEESE SAUCE","Chicken or turkey and noodles with cheese sauce (mixture)"
+27242350,"CHICKEN OR TURKEY TETRAZZINI","Chicken or turkey tetrazzini"
+27242400,"CHICKEN & NOODLES, TOMATO-BASED SAUCE (MIXTURE)","Chicken or turkey and noodles, tomato-based sauce (mixture)"
+27242500,"CHICKEN OR TURKEY AND NOODLES WITH SOY-BASED SAUCE (MIXTURE)","Chicken or turkey and noodles with soy-based sauce (mixture)"
+27243000,"CHICKEN & RICE, NO SAUCE (MIXTURE)","Chicken or turkey and rice, no sauce (mixture)"
+27243100,"BIRYANI WITH CHICKEN","Biryani with chicken"
+27243300,"CHICKEN & RICE W/ CREAM SAUCE (MIXTURE)","Chicken or turkey and rice with cream sauce (mixture)"
+27243400,"CHICKEN & RICE W/ (MUSHROOM) SOUP-BASED SAUCE","Chicken or turkey and rice with (mushroom) soup (mixture)"
+27243500,"CHICKEN & RICE W/ TOMATO-BASED SAUCE (MIXTURE)","Chicken or turkey and rice with tomato-based sauce (mixture)"
+27243600,"CHICKEN & RICE W/ SOY-BASED SAUCE (MIXTURE)","Chicken or turkey and rice with soy-based sauce (mixture)"
+27243700,"CHICKEN IN CHEESE SCE W/ SPANISH RICE","Chicken in cheese sauce with Spanish rice"
+27246100,"CHICKEN W/ DUMPLINGS (MIXTURE)","Chicken or turkey with dumplings (mixture)"
+27246200,"CHICKEN W/ STUFFING (MIXTURE)","Chicken or turkey with stuffing (mixture)"
+27246300,"CHICKEN OR TURKEY CAKE, PATTY OR CROQUETTE","Chicken or turkey cake, patty, or croquette"
+27246400,"CHICKEN SOUFFLE","Chicken or turkey souffle"
+27246500,"MEAT LOAF MADE W/ CHICKEN OR TURKEY","Meat loaf made with chicken or turkey"
+27246505,"MEAT LOAF W/ CHICKEN OR TURKEY, W/ TOMATO SAUCE","Meat loaf made with chicken or turkey, with tomato-based sauce"
+27250020,"CLAMS, STUFFED","Clams, stuffed"
+27250030,"CODFISH BALL OR CAKE","Codfish ball or cake"
+27250040,"CRAB CAKE","Crab cake"
+27250050,"FISH CAKE OR PATTY, NS AS TO FISH","Fish cake or patty, NS as to fish"
+27250060,"GEFILTE FISH","Gefilte fish"
+27250070,"SALMON CAKE OR PATTY (INCLUDE SALMON CROQUETTE)","Salmon cake or patty"
+27250080,"SALMON LOAF","Salmon loaf"
+27250110,"SCALLOPS & NOODLES W/ CHEESE SAUCE (MIXTURE)","Scallops and noodles with cheese sauce (mixture)"
+27250120,"SHRIMP AND NOODLES, NO SAUCE (MIXTURE)","Shrimp and noodles, no sauce (mixture)"
+27250122,"SHRIMP AND NOODLES WITH GRAVY (MIXTURE)","Shrimp and noodles with gravy (mixture)"
+27250124,"SHRIMP AND NOODLES WITH (MUSHROOM) SOUP (MIXTURE)","Shrimp and noodles with (mushroom) soup (mixture)"
+27250126,"SHRIMP AND NOODLES WITH CREAM OR WHITE SAUCE (MIXTURE)","Shrimp and noodles with cream or white sauce (mixture)"
+27250128,"SHRIMP AND NOODLES WITH SOY-BASED SAUCE (MIXTURE)","Shrimp and noodles with soy-based sauce (mixture)"
+27250130,"SHRIMP & NOODLES W/ CHEESE SAUCE","Shrimp and noodles with cheese sauce (mixture)"
+27250132,"SHRIMP AND NOODLES WITH TOMATO SAUCE (MIXTURE)","Shrimp and noodles with tomato sauce (mixture)"
+27250150,"TUNA LOAF","Tuna loaf"
+27250160,"TUNA CAKE OR PATTY","Tuna cake or patty"
+27250210,"CLAM CAKE OR PATTY (INCLUDE DEVILED)","Clam cake or patty"
+27250220,"OYSTER FRITTER","Oyster fritter"
+27250250,"FLOUNDER W/CRAB STUFFING","Flounder with crab stuffing"
+27250260,"LOBSTER W/ BREAD STUFFING,BAKED","Lobster with bread stuffing, baked"
+27250270,"CLAMS, CASINO","Clams Casino"
+27250300,"MACKEREL CAKE OR PATTY","Mackerel cake or patty"
+27250310,"HADDOCK CAKE OR PATTY","Haddock cake or patty"
+27250400,"SHRIMP CAKE OR PATTY (INCL SHRIMP BURGER OR STICK)","Shrimp cake or patty"
+27250410,"SHRIMP W/ CRAB STUFFING","Shrimp with crab stuffing"
+27250450,"SHRIMP TOAST, FRIED","Shrimp toast, fried"
+27250520,"SEAFOOD, RESTRUCTURED (INCL IMITATION CRABMEAT)","Seafood restructured"
+27250550,"SEAFOOD SOUFFLE","Seafood souffle"
+27250610,"TUNA NOODLE CASSEROLE W/ CREAM OR WHITE SAUCE","Tuna noodle casserole with cream or white sauce"
+27250630,"TUNA NOODLE CASSEROLE W/ (MUSHROOM) SOUP","Tuna noodle casserole with (mushroom) soup"
+27250710,"TUNA & RICE W/ (MUSHROOM) SOUP (MIXTURE)","Tuna and rice with (mushroom) soup (mixture)"
+27250810,"FISH & RICE W/ TOMATO-BASED SAUCE","Fish and rice with tomato-based sauce"
+27250820,"FISH & RICE W/ CREAM SAUCE","Fish and rice with cream sauce"
+27250830,"FISH & RICE W/ (MUSHROOM) SOUP","Fish and rice with (mushroom) soup"
+27250900,"FISH & NOODLES W/ (MUSHROOM) SOUP (MIXTURE)","Fish and noodles with (mushroom) soup"
+27250950,"SHELLFISH & NOODLES, TOMATO-BASED SAUCE","Shellfish mixture and noodles, tomato-based sauce (mixture)"
+27251010,"STEWED SALMON, P.R. (SALMON GUISADO)","Stewed salmon, Puerto Rican style (Salmon guisado)"
+27260010,"MEATLOAF, NS AS TO TYPE OF MEAT","Meat loaf, NS as to type of meat"
+27260050,"MEATBALLS, W/ BREADING, W/ GRAVY","Meatballs, with breading, NS as to type of meat, with gravy"
+27260080,"MEAT LOAF MADE W/ BEEF & PORK","Meat loaf made with beef and pork"
+27260090,"MEAT LOAF W/ BEEF, VEAL & PORK","Meat loaf made with beef, veal and pork"
+27260100,"MEAT LOAF W/ BEEF & PORK, W/ TOMATO SAUCE","Meat loaf made with beef and pork, with tomato-based sauce"
+27260110,"HASH, NS AS TO TYPE OF MEAT","Hash, NS as to type of meat"
+27260500,"VIENNA SAUSAGES STEWED W/ POTATOES, P.R.","Vienna sausages stewed with potatoes, Puerto Rican style (Salchichas guisadas)"
+27260510,"LIVER DUMPLING","Liver dumpling"
+27261000,"BREADED BRAINS, P.R. (SESOS REBOSADOS)","Breaded brains, Puerto Rican style (Sesos rebosados)"
+27261500,"STEWED SEASONED GROUND BEEF&PORK,W/POT, MEXICAN","Stewed, seasoned, ground beef and pork, with potatoes, Mexican style (Picadillo de carne de rez y puerco con papas)"
+27311110,"BEEF, POTATOES, & VEG (W/ CAR/DK GREEN), NO SAUCE","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27311120,"BEEF, POTATOES, & VEG (NO CAR/DK GREEN), NO SAUCE","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27311210,"CORNED BEEF, POT & VEG(W/ CAR/DK GREEN), NO SAUCE","Corned beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27311220,"CORNED BEEF, POTATO & VEG (NO CAR/DK GRN), NO SAUCE","Corned beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27311310,"BEEF STEW W/ POT & VEG(W/ CAR/DK GRN), TOMATO SAUCE","Beef stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27311320,"BEEF STEW W/ POT & VEG (NO CAR/DK GREEN), TOM SAUCE","Beef stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce"
+27311410,"BEEF STEW W/ POT & VEG (W/ CAR, DK GREEN), GRAVY","Beef stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy"
+27311420,"BEEF STEW W/ POT & VEG (NO CAR, DK GREEN), GRAVY","Beef stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy"
+27311510,"SHEPHERD'S PIE W/ BEEF","Shepherd's pie with beef"
+27311600,"BEEF, POT, & VEG (INCL CAR/DK GRN), GRAVY","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27311605,"BEEF, POT, & VEG (NO CAR/DK GREEN), GRAVY","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27311610,"BEEF, POT & VEG (INCL CAR/DK GRN), CR/SOUP-BASED SAUCE","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27311620,"BEEF, POT & VEG (NO CAR/DK GRN), CR/SOUP-BASED SAUCE","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27311625,"BEEF, POT, & VEG (INCL CAR/DK GRN), TOMATO-BASED SAUCE","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27311630,"BEEF, POT, & VEG (NO CAR/DK GREEN), TOMATO-BASED SAUCE","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27311635,"BEEF, POT, & VEG (INCL CAR/DK GRN), CHEESE SAUCE","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), cheese sauce (mixture)"
+27311640,"BEEF, POT, & VEG (NO CAR/DK GREEN), CHEESE SAUCE","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27311645,"BEEF, POT, & VEG (INCL CAR/DK GRN), SOY-BASED SAUCE","Beef, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), soy-based sauce (mixture)"
+27311650,"BEEF, POT, & VEG (NO CAR/DK GREEN), SOY-BASED SAUCE","Beef, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), soy-based sauce (mixture)"
+27313010,"BEEF, NOODLES & VEG (W/ CARROTS/DK GREEN), NO SAUCE","Beef, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27313020,"BEEF, NOODLES & VEG (NO CARROTS/DK GREEN), NO SAUCE","Beef, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27313110,"BEEF CHOW MEIN OR CHOP SUEY W/ NOODLES","Beef chow mein or chop suey with noodles"
+27313150,"BEEF, NOODLES & VEG (W/ CAR/DK GREEN), SOY SAUCE","Beef, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), soy-based sauce (mixture)"
+27313160,"BEEF, NOODLES & VEG (NO CAR/DK GREEN), SOY SAUCE","Beef, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), soy-based sauce (mixture)"
+27313210,"BEEF, NOODLES & VEG (W/ CAR/DK GREEN), TOMATO SAUCE","Beef, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27313220,"BEEF, NOODLES & VEG (NO CAR/DK GREEN), TOMATO SAUCE","Beef, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27313310,"BEEF, NOODLES, VEG(INCL CARROTS/DK GREEN), SOUP","Beef, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), (mushroom) soup (mixture)"
+27313320,"BEEF, NOODLES, VEG (NO CARROTS/DK GREEN), SOUP","Beef, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), (mushroom) soup (mixture)"
+27313410,"BEEF, NOODLES, & VEG (INCL CAR/DK GRN), GRAVY","Beef, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27313420,"BEEF, NOODLES, & VEG (NO CAR/DK GRN), GRAVY","Beef, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27315010,"BEEF, RICE & VEG (W/ CARROTS/DK GREEN), NO SAUCE","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27315020,"BEEF, RICE & VEG (NO CARROTS/DK GREEN), NO SAUCE","Beef, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27315210,"BEEF, RICE & VEG (W/ CAR/DK GREEN), TOMATO SAUCE","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27315220,"BEEF, RICE & VEG (NO CAR/DK GREEN), TOMATO SAUCE","Beef, rice, and vegetables (excluding carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27315250,"STUFFED CABBAGE ROLLS W/ BEEF AND RICE","Stuffed cabbage rolls with beef and rice"
+27315270,"STUFFED GRAPE LEAVES W/ BEEF & RICE","Stuffed grape leaves with beef and rice"
+27315310,"BEEF, RICE & VEGETABLES (W/ CARROTS/DK GREEN), SOUP","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), (mushroom) soup (mixture)"
+27315320,"BEEF, RICE & VEGETABLES (NO CARROTS/DK GREEN), SOUP","Beef, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), (mushroom) soup (mixture)"
+27315330,"BEEF, RICE & VEG (INCL CAR/DK GRN), CHEESE SAUCE","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), cheese sauce (mixture)"
+27315340,"BEEF, RICE & VEG (NO CAR/DK GRN), CHEESE SAUCE","Beef, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27315410,"BEEF, RICE & VEG (INCL CAR/DK GRN), GRAVY, MIXTURE","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27315420,"BEEF, RICE & VEG (NO CAR/DK GRN), GRAVY, MIXTURE","Beef, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27315510,"BEEF, RICE & VEG (INCL CAR/DK GRN), SOY-BASED SAUCE","Beef, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), soy-based sauce (mixture)"
+27315520,"BEEF, RICE & VEG (NO CAR/DK GRN), SOY-BASED SAUCE","Beef, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), soy-based sauce (mixture)"
+27317010,"BEEF POT PIE (INCLUDE GREEK MEAT PIE)","Beef pot pie"
+27317100,"BEEF, DUMPLINGS & VEG (INCL CAR/DK GRN), GRAVY","Beef, dumplings, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27317110,"BEEF, DUMPLINGS & VEG (NO CAR/DK GRN), GRAVY","Beef, dumplings, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27319010,"STUFFED GREEN PEPPER, P.R. (PIMIENTO RELLENO)","Stuffed green pepper, Puerto Rican style (Pimiento relleno)"
+27320020,"HAM POT PIE","Ham pot pie"
+27320025,"HAM OR PORK, NOODLES, VEG (NO CAR, BROC, DK GRN)NO SAUCE","Ham or pork, noodles and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27320027,"HAM OR PORK, NOODLES, VEG (INCL CAR, BROC, DARK GREEN)NO SAU","Ham or pork, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27320030,"HAM/PORK, NOODLES & VEG (NO CAR/DK GR), CHEESE SCE","Ham or pork, noodles and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27320040,"PORK, POTATOES & VEG (W/ CAR, DK GREEN), NO SAUCE","Pork, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27320070,"PORK, NOODLES, VEG (INCL CAR/DK GRN), TOMATO SAUCE","Ham or pork, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27320080,"SAUSAGE, NOODLES, VEG (NO CAR/DK GRN), TOMATO SAUCE","Sausage, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce"
+27320090,"SAUSAGE, NOODLES, VEG (W/ CAR/DK GRN), TOMATO SAUCE","Sausage, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27320100,"PORK, POTATOES & VEG (W/ CAR, DK GREEN), TOMATO SCE","Pork, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27320110,"PORK, POTATOES & VEG (NO CAR, DK GREEN), TOMATO SCE","Pork, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27320120,"SAUSAGE, POT, & VEG (INCL CAR/BROC/DK GREEN), GRAVY","Sausage, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27320130,"SAUSAGE, POT, & VEG (NO CAR/BROC/DK GREEN), GRAVY","Sausage, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27320140,"PORK, POT, & VEG (INCL CAR/DK GRN), GRAVY, MIXTURE","Pork, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27320150,"PORK, POT, & VEG (NO CAR/DK GRN), GRAVY, MIXTURE","Pork, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27320210,"PORK, POTATOES & VEG (NO CAR, DK GREEN), NO SAUCE","Pork, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27320310,"PORK CHOW MEIN OR CHOP SUEY W/ NOODLES","Pork chow mein or chop suey with noodles"
+27320320,"PORK, RICE & VEG (INCL CAR/DK GRN), SOY-BASED SAUCE","Pork, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), soy-based sauce (mixture)"
+27320330,"PORK, RICE & VEG (NO CAR/DK GRN), SOY-BASED SAUCE","Pork, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), soy-based sauce (mixture)"
+27320340,"PORK, RICE & VEG (INCL CAR/DK GRN), TOMATO SAUCE","Pork, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27320350,"PORK, RICE & VEG (NO CAR/DK GRN), TOMATO SAUCE","Pork, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27320410,"HAM, POTATOES & VEG (NO CARROTS/DK GREEN), NO SAUCE","Ham, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27320450,"HAM, POTATOES & VEG (W/ CARROTS/DK GREEN), NO SAUCE","Ham, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27320500,"SWEET & SOUR PORK W/ RICE","Sweet and sour pork with rice"
+27330010,"SHEPHERD'S PIE W/ LAMB","Shepherd's pie with lamb"
+27330030,"LAMB STEW W/ POT & VEG (INCL CAR/DK GREEN), GRAVY","Lamb or mutton stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy"
+27330050,"LAMB, RICE & VEGETABLES (NO CARROT/DK GREEN), GRAVY","Lamb or mutton, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27330060,"LAMB, RICE & VEG (INCL CAR/DK GRN), TOMATO SAUCE","Lamb or mutton, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27330080,"LAMB, RICE, & VEGETABLES (INCL CAR, DK GRN), GRAVY","Lamb or mutton, rice, and vegetables  (including carrots, broccoli, and/or dark-green leafy), gravy"
+27330110,"LAMB STEW W/ POT & VEG (NO CAR/DK GREEN), GRAVY","Lamb or mutton stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy"
+27330170,"STUFFED GRAPE LEAVES W/ LAMB & RICE","Stuffed grape leaves with lamb and rice"
+27330210,"LAMB STEW W/ POT & VEG (INCL CAR/DK GRN), TOM SAUCE","Lamb or mutton stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27330220,"LAMB STEW W/ POT & VEG (NO CAR/DK GRN), TOMATO SCE","Lamb or mutton stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce"
+27331150,"VEAL FRICASSEE, P.R. (TERNERA EN FRICASE)","Veal fricassee, Puerto Rican style (ternera en fricase)"
+27332100,"VEAL STEW W/ POT, VEG (INCL CAR/DK GRN) TOM SAUCE","Veal stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27332110,"VEAL STEW W/ POT, VEG (NO CAR,DK GRN), TOMATO SAUCE","Veal stew with potatoes and vegetables (excluding carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27335100,"RABBIT STEW W/ POTATOES & VEGETABLES","Rabbit stew with potatoes and vegetables"
+27335500,"STEWED RABBIT, P.R. (FRICASE DE CONEJO)","Stewed rabbit, Puerto Rican style (Fricase de conejo)"
+27336100,"VENISON/DEER STEW W/ POTATO & VEG(W/ CAR/DK GRN),TOM SCE","Venison/deer stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27336150,"VENISON/DEER STEW W/ POTATO & VEG(NO CAR/DK GRN),TOM SCE","Venison/deer stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce"
+27336200,"VENISON/DEER, POTATOES & VEG (INCL CAR/DK GRN), GRAVY","Venison/deer, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27336250,"VENISON/DEER, POTATOES & VEG (NO CAR/DK GRN), GRAVY","Venison/deer, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27336300,"VENISON/DEER, NOODLES & VEG (INCL CAR/DK GRN),TOM SAUCE","Venison/deer, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27336310,"VENISON/DEER, NOODLES & VEG (NO CAR/DK GRN), TOM SAUCE","Venison/deer, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27341000,"CHICKEN OR TURKEY, POTATOES, CORN, AND CHEESE, WITH GRAVY","Chicken or turkey, potatoes, corn, and cheese, with gravy"
+27341010,"CHICKEN, POT & VEG (INCL CAR/DK GRN), NO SAUCE","Chicken or turkey, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27341020,"CHICKEN, POT & VEG (NO CAR/DK GRN), NO SAUCE","Chicken or turkey, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27341025,"CHICKEN, POT & VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27341030,"CHICKEN, POT & VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27341035,"CHICKEN, POT & VEG (INCL CAR/DK GRN), CREAM/SOUP-BASED SAUCE","Chicken or turkey, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27341040,"CHICKEN, POT & VEG (NO CAR/DK GRN), CREAM/SOUP-BASED SAUCE","Chicken or turkey, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27341045,"CHICKEN, POT & VEG (INCL CAR/DK GRN), CHEESE SAUCE","Chicken or turkey, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), cheese sauce (mixture)"
+27341050,"CHICKEN, POT & VEG (NO CAR/DK GRN), CHEESE SAUCE","Chicken or turkey, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27341055,"CHICKEN, POT & VEG (INCL CAR/DK GRN), TOMATO-BASED SAUCE","Chicken or turkey, potatoes, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27341060,"CHICKEN, POT & VEG (NO CAR/DK GRN), TOMATO-BASED SAUCE","Chicken or turkey, potatoes, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27341310,"CHICKEN STEW W/ POT, VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy"
+27341320,"CHICKEN STEW W/ POT & VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy"
+27341510,"CHICKEN STEW W/ POT & VEG(INCL CAR/DK GRN), TOM SCE","Chicken or turkey stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce"
+27341520,"CHICKEN STEW W/ POT & VEG(NO CAR/DK GRN), TOM SAUCE","Chicken or turkey stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato- based sauce"
+27343010,"CHICKEN, NOODLES & VEG (INCL CAR/DK GRN), NO SAUCE","Chicken or turkey, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27343020,"CHICKEN, NOODLES & VEG (NO CAR/DK GRN), NO SAUCE","Chicken or turkey, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27343410,"CHICKEN, NOODLES & VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27343420,"CHICKEN, NOODLES & VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27343470,"CHICKEN OR TURKEY, NOODLES, AND VEGETABLES (INCLUDING CARROT","Chicken or turkey, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27343480,"CHIX, NDL, VEG(NO CAR/DK GRN), CR/SOUP-BASED SAUCE","Chicken or turkey, noodles, and vegetables (excluding carrots, broccoli, and/or dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27343510,"CHICKEN, NOODLES, VEG (INCL CAR/DK GRN), TOMATO SCE","Chicken or turkey, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27343520,"CHICKEN, NOODLES, VEG (NO CAR/DK GRN), TOMATO SAUCE","Chicken or turkey, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27343910,"CHICKEN CHOW MEIN/CHOP SUEY W/ NOODLES","Chicken or turkey chow mein or chop suey with noodles"
+27343950,"CHICKEN, NOODLES & VEG(INCL CAR/DK GRN), CHEESE SCE","Chicken or turkey, noodles, and vegetables (including carrots, broccoli, and/or dark-green leafy), cheese sauce (mixture)"
+27343960,"CHICKEN, NOODLES & VEG(NO CAR/DK GRN), CHEESE SAUCE","Chicken or turkey, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27345010,"CHICKEN, RICE & VEG (INCL CAR/DK GRN), NO SAUCE","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27345020,"CHICKEN, RICE & VEG (NO CAR/DK GRN), NO SAUCE","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), no sauce (mixture)"
+27345210,"CHICKEN, RICE & VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27345220,"CHICKEN, RICE & VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27345230,"CHICKEN OR TURKEY, RICE, CORN, AND CHEESE WITH GRAVY","Chicken or turkey, rice, corn, and cheese, with gravy"
+27345310,"CHICKEN, RICE & VEG (INCL CAR/DK GRN), SOY SAUCE","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), soy-based sauce (mixture)"
+27345320,"CHICKEN, RICE & VEG (NO CAR/DK GRN), SOY SAUCE","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), soy-based sauce (mixture)"
+27345410,"CHIX, RICE, & VEG(INCL CAR/DK GRN), CR/SOUP-BASED SAU","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27345420,"CHIX, RICE, AND VEG(NO CAR/DK GRN), CR/SOUP-BASED SAU","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), cream sauce, white sauce, or mushroom soup-based sauce (mixture)"
+27345440,"CHICKEN, RICE & VEG (INCL CAR/DK GRN), CHEESE SAUCE","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), cheese sauce (mixture)"
+27345450,"CHICKEN, RICE, VEG (NO CAR/DK GRN), CHEESE SAUCE","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27345510,"CHICKEN, RICE & VEG (INCL CAR/DK GRN), TOMATO SAUCE","Chicken or turkey, rice, and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-based sauce (mixture)"
+27345520,"CHICKEN, RICE & VEG (NO CAR/DK GRN), TOMATO SAUCE","Chicken or turkey, rice, and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-based sauce (mixture)"
+27347100,"CHICKEN OR TURKEY POT PIE","Chicken or turkey pot pie"
+27347200,"CHICKEN, STUFFING & VEG (INCL CAR/DK GRN), NO SAUCE","Chicken or turkey, stuffing, and vegetables (including carrots, broccoli, and/or dark-green leafy), no sauce (mixture)"
+27347210,"CHICKEN, STUFFING, VEG (NO CAR/DK GRN), NO SAUCE","Chicken or turkey,stuffing, and vegetables (excluding carrots, broccoli, and dark green leafy), no sauce (mixture)"
+27347220,"CHICKEN, STUFFING & VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey, stuffing, and vegetables (including carrots, broccoli, and/or dark-green leafy), gravy (mixture)"
+27347230,"CHICKEN, STUFFING & VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey, stuffing, and vegetables (excluding carrots, broccoli, and dark-green leafy), gravy (mixture)"
+27347240,"CHICKEN, DUMPLINGS, VEG (INCL CAR/DK GRN), GRAVY","Chicken or turkey, dumplings, and vegetables (including carrots, broccoli, and/or dark green leafy), gravy (mixture)"
+27347250,"CHICKEN, DUMPLINGS, VEG (NO CAR/DK GRN), GRAVY","Chicken or turkey, dumplings, and vegetables (excluding carrots, broccoli, and dark green leafy), gravy (mixture)"
+27348100,"CHICKEN FRICASSEE, P.R. (FRICASE DE POLLO)","Chicken fricassee, Puerto Rican style (Fricase de pollo)"
+27350020,"PAELLA WITH SEAFOOD","Paella with seafood"
+27350030,"SEAFOOD STEW W/ POT & VEG (NO CAR/DK GREEN),TOM SCE","Seafood stew with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy), tomato-base sauce"
+27350050,"SHRIMP CHOW MEIN OR CHOP SUEY W/ NOODLES","Shrimp chow mein or chop suey with noodles"
+27350060,"SHRIMP CREOLE W/ RICE (INCLUDE SHRIMP JAMBALAYA)","Shrimp creole, with rice"
+27350070,"TUNA POT PIE","Tuna pot pie"
+27350080,"TUNA NOODLE CASSEROLE W/ VEG, CREAM OR WHITE SAUCE","Tuna noodle casserole with vegetables, cream or white sauce"
+27350090,"FISH, NOODLES, VEG (INCL CAR/DK GRN), CHEESE SAUCE","Fish, noodles, and vegetables (including carrots, broccoli, and/or dark green leafy), cheese sauce (mixture)"
+27350100,"FISH, NOODLES, VEG (NO CAR/DK GRN), CHEESE SAUCE","Fish, noodles, and vegetables (excluding carrots, broccoli, and dark-green leafy), cheese sauce (mixture)"
+27350110,"BOUILLABAISSE","Bouillabaisse"
+27350200,"OYSTER PIE (INCLUDE OYSTER POT PIE)","Oyster pie"
+27350310,"SEAFOOD STEW W/ POT & VEG (W/ CAR/DK GREEN),TOM SCE","Seafood stew with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy), tomato-base sauce"
+27350410,"TUNA NOODLE CASSEROLE W/ VEG & (MUSHROOM) SOUP","Tuna noodle casserole with vegetables and (mushroom) soup"
+27351010,"CODFISH W/ STARCHY VEG, P.R. (SERENATA DE BACALAO)","Codfish with starchy vegetables, Puerto Rican style (Serenata de bacalao) (mixture)"
+27351020,"CODFISH SALAD, P.R. (GAZPACHO DE BACALAO)","Codfish salad, Puerto Rican style (Gazpacho de bacalao)"
+27351030,"STEWED CODFISH, P.R. (BACALAO GUISADO)","Stewed codfish, Puerto Rican style (Bacalao guisado)"
+27351040,"BISCAYNE CODFISH, P.R. (BACALAO A LA VIZCAINA)","Biscayne codfish, Puerto Rican style (Bacalao a la Vizcaina)"
+27351050,"CODFISH SALAD, P.R. (ENSALADA DE BACALAO)","Codfish salad, Puerto Rican style (Ensalada de bacalao)"
+27360000,"STEW, NFS","Stew, NFS"
+27360010,"GOULASH, NFS","Goulash, NFS"
+27360050,"MEAT PIE, NFS","Meat pie, NFS"
+27360080,"CHOW MEIN, NS AS TO TYPE OF MEAT, W/ NOODLES","Chow mein or chop suey, NS as to type of meat, with noodles"
+27360090,"PAELLA, NFS","Paella, NFS"
+27360100,"BRUNSWICK STEW","Brunswick stew"
+27360120,"CHOW MEIN/CHOP SUEY,VARIOUS MEATS, W/ NOODLES","Chow mein or chop suey, various types of meat, with noodles"
+27361010,"STEWED VARIETY MEATS (MOSTLY LIVER), P.R.(GANDINGA)","Stewed variety meats, Puerto Rican style (mostly liver) (Gandinga)"
+27362000,"STEWED TRIPE W/ POTATOES, P.R. (MONDONGO)","Stewed tripe, Puerto Rican style, with potatoes (Mondongo)"
+27363000,"GUMBO W/ RICE (NEW ORLEANS TYPE)","Gumbo with rice (New Orleans type with shellfish, pork, and/or poultry, tomatoes, okra, rice)"
+27363100,"JAMBALAYA W/ MEAT & RICE","Jambalaya with meat and rice"
+27410210,"BEEF & VEG (W/ CAR/DK GREEN, NO POTATO), NO SAUCE","Beef and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), no sauce (mixture)"
+27410220,"BEEF & VEG (NO CAR/DK GREEN, NO POTATO), NO SAUCE","Beef and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), no sauce (mixture)"
+27410250,"BEEF SHISH KABOB W/ VEGETABLES, EXCLUDING POTATOES","Beef shish kabob with vegetables, excluding potatoes"
+27411100,"BEEF & VEG(W/ CAR/DK GREEN, NO POTATO), TOMATO SCE","Beef with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27411120,"SWISS STEAK","Swiss steak"
+27411150,"BEEF ROLL,STUFFED W/VEG/MEAT MIXTURE,TOM-BASE SAUCE","Beef rolls, stuffed with vegetables or meat mixture, tomato-based sauce"
+27411200,"BEEF W/ VEG (NO CAR/DK GREEN, NO POTATO),TOMATO SCE","Beef with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27414100,"BEEF W/ VEG (INCL CAR/DK GRN, NO POT), SOUP","Beef with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), (mushroom) soup (mixture)"
+27414200,"BEEF W/ VEG (NO CAR/DK GRN, NO POT), SOUP","Beef with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), (mushroom) soup (mixture)"
+27415100,"BEEF & VEG (W/ CAR/DK GREEN, NO POTATO), SOY SAUCE","Beef and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27415110,"BEEF AND BROCCOLI","Beef and broccoli"
+27415120,"BEEF, TOFU & VEG(W/ CAR/DK GRN, NO POTATO),SOY SCE","Beef, tofu, and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27415130,"SZECHUAN BEEF","Szechuan beef"
+27415140,"HUNAN BEEF","Hunan beef"
+27415150,"BEEF, CHOW MEIN OR CHOP SUEY, NO NOODLES","Beef chow mein or chop suey, no noodles"
+27415170,"KUNG PAO BEEF","Kung Pao beef"
+27415200,"BEEF & VEG (NO CAR/DK GREEN, NO POTATO), SOY SAUCE","Beef and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27415220,"BEEF, TOFU & VEG(NO CAR/DK GRN, NO POTATO), SOY SCE","Beef, tofu, and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27416100,"BEEF & VEGETABLES, HAWAIIAN STYLE (MIXTURE)","Beef and vegetables, Hawaiian style (mixture)"
+27416150,"PEPPER STEAK","Pepper steak"
+27416200,"BEEF, GROUND, W/ EGG & ONION (MIXTURE)","Beef, ground, with egg and onion (mixture)"
+27416250,"BEEF SALAD","Beef salad"
+27416300,"BEEF TACO FILLING: BEEF, CHEESE, TOMATO, TACO SAUCE","Beef taco filling: beef, cheese, tomato, taco sauce"
+27416400,"SUKIYAKI (STIR FRIED BEEF & VEGS IN SOY SAUCE)","Sukiyaki (stir fried beef and vegetables in soy sauce)"
+27416450,"BEEF & VEG (INCL CAR/DK GRN, NO POTATOES), GRAVY","Beef and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), gravy (mixture)"
+27416500,"BEEF & VEG (NO CAR/DK GREEN, NO POT), GRAVY","Beef and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), gravy (mixture)"
+27418110,"SEASONED SHREDDED SOUP MEAT","Seasoned shredded soup meat (Ropa vieja, sopa de carne ripiada)"
+27418210,"BEEF STEW, P.R. W/ VEGETABLES, NO POTATO (CARNE A LA JUDIA)","Puerto Rican style beef stew with vegetables, excluding potatoes (Carne a la Judia)"
+27418310,"CORNED BEEF W/ TOMATO SAUCE & ONION, P.R. STYLE","Corned beef with tomato sauce and onion, Puerto Rican style (mixture)"
+27418410,"BEEF STEAK W/ ONIONS, P.R. (BIFTEC ENCEBOLLADO)","Beef steak with onions, Puerto Rican style (mixture) (Biftec encebollado)"
+27420010,"CABBAGE W/ HAM HOCKS (MIXTURE)","Cabbage with ham hocks (mixture)"
+27420020,"HAM OR PORK SALAD","Ham or pork salad"
+27420040,"FRANKFURTERS OR HOT DOGS & SAUERKRAUT (MIXTURE)","Frankfurters or hot dogs and sauerkraut (mixture)"
+27420060,"PORK & VEG (W/ CAR/DK GREEN, NO POTATO), NO SAUCE","Pork and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), no sauce (mixture)"
+27420080,"GREENS W/ HAM OR PORK (MIXTURE)","Greens with ham or pork (mixture)"
+27420100,"PORK, TOFU & VEG (W/ CAR/DK GRN,NO POTATO), SOY SCE","Pork, tofu, and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-base sauce (mixture)"
+27420110,"PORK & VEGETABLES, HAWAIIAN STYLE (MIXTURE)","Pork and vegetables, Hawaiian style (mixture)"
+27420120,"PORK & WATERCRESS W/ SOY-BASED SAUCE (MIXTURE)","Pork and watercress with soy-based sauce (mixture)"
+27420150,"KUNG PAO PORK","Kung Pao pork"
+27420160,"MOO SHU (MU SHI) PORK, W/O PANCAKE","Moo Shu (Mu Shi) Pork, without Chinese pancake"
+27420170,"PORK AND ONIONS W/ SOY-BASED SAUCE","Pork and onions with soy-based sauce (mixture)"
+27420200,"PORK HASH,HAWAIIAN--PORK,VEG(NO CAR/DK GRN),SOY SCE","Pork hash, Hawaiian style-ground pork, vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce"
+27420250,"HAM & VEG (W/ CARROT/DK GREEN, NO POTATO), NO SAUCE","Ham and vegetables (including carrots, broccoli, and/or dark- green leafy (no potatoes)), no sauce (mixture)"
+27420270,"HAM & VEG (NO CARROT/DK GREEN, NO POTATO), NO SAUCE","Ham and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), no sauce (mixture)"
+27420350,"PORK & VEG (NO CAR/DK GREEN, NO POTATO), NO SAUCE","Pork and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), no sauce (mixture)"
+27420370,"PORK,TOFU & VEG(NO CAR/DK GREEN,NO POTATO)SOY SAUCE","Pork, tofu, and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27420390,"PORK CHOW MEIN OR CHOP SUEY, NO NOODLES","Pork chow mein or chop suey, no noodles"
+27420400,"PORK & VEG (INCL CAR/DK GRN, NO POT), TOMATO SAUCE","Pork and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27420410,"PORK & VEG (NO CAR/DK GRN, NO POT), TOMATO SAUCE","Pork and vegetables (excluding carrots, broccoli, and dark- green leafy (no potatoes)), tomato-based sauce (mixture)"
+27420450,"SAUSAGE & VEG (INCL CAR/DK GRN)(NO POT), TOM SAUCE","Sausage and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27420460,"SAUSAGE & VEG (NO CAR/DK GRN/POT), TOMATO SAUCE","Sausage and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27420470,"SAUSAGE & PEPPERS, NO SAUCE","Sausage and peppers, no sauce (mixture)"
+27420500,"PORK & VEG (INCL CAR/DK GRN), SOY-BASED SAUCE","Pork and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27420510,"PORK & VEG (NO CAR/DK GRN), SOY-BASED SAUCE","Pork and vegetables (excluding carrots, broccoli, and dark- green leafy (no potatoes)), soy-based sauce (mixture)"
+27420520,"PORK SHISH KABOB WITH VEGETABLES, EXCLUDING POTATOES","Pork shish kabob with vegetables, excluding potatoes"
+27421010,"STUFFED CHRISTOPHINE, P.R. (CHAYOTE RELLENO)","Stuffed christophine, Puerto Rican style (Chayote relleno)"
+27422010,"PORK CHOPS STEWED W/VEG, P.R. (CHULETAS A LA JARD.)","Pork chop stewed with vegetables, Puerto Rican style (mixture) (Chuletas a la jardinera)"
+27430400,"LAMB STEW W/ VEG (INCL CAR/DK GRN, NO POT), GRAVY","Lamb or mutton stew with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), gravy"
+27430410,"LAMB STEW W/ VEG (NO CAR/DK GRN, NO POT), GRAVY","Lamb or mutton stew with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), gravy"
+27430500,"VEAL GOULASH W/VEG(NO CAR/DK GREEN, NO POT),TOM SCE","Veal goulash with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), tomato-base sauce"
+27430510,"VEAL GOULASH W/VEG(W/ CAR/DK GREEN, NO POT),TOM SCE","Veal goulash with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-base sauce"
+27430580,"VEAL W/ VEG (INCL CAR/DK GRN), NO POT, CREAM SAUCE","Veal with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), cream or white sauce"
+27430590,"VEAL W/ VEG (NO CAR/DK GRN), NO POT, CREAM SAUCE","Veal with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), cream or white sauce"
+27430610,"LAMB SHISH KABOB W/ VEGETABLES, EXCLUDING POTATOES","Lamb shish kabob with vegetables, excluding potatoes"
+27440110,"CHICK/TURK & VEG (W/ CAR/DK GRN, NO POT), NO SAUCE","Chicken or turkey and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), no sauce (mixture)"
+27440120,"CHICK/TURK & VEG (NO CAR/DK GRN, NO POT), NO SAUCE","Chicken or turkey and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), no sauce (mixture)"
+27440130,"CHICKEN OR TURKEY SHISH KABOB W/VEGETABLES, EXCL POTATOES","Chicken or turkey shish kabob with vegetables, excluding potatoes"
+27441120,"CHICKEN CREOLE W/O RICE","Chicken or turkey creole, without rice"
+27442110,"CHICKEN/TURKEY & VEG (W/ CAR/DK GREEN,NO POT),GRAVY","Chicken or turkey and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), gravy (mixture)"
+27442120,"CHICKEN/TURKEY & VEG(NO CAR/DK GREEN,NO POT), GRAVY","Chicken or turkey and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), gravy (mixture)"
+27443110,"CHICKEN A LA KING W/VEG(INCL CAR/DK GRN),WHITE SCE","Chicken or turkey a la king with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), cream, white, or soup-based sauce"
+27443120,"CHICKEN A LA KING W/ VEG(NO CAR/DK GRN),WHITE SAUCE","Chicken or turkey a la king with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), cream, white, or soup-based sauce"
+27443150,"CHICKEN DIVAN","Chicken or turkey divan"
+27445110,"CHICKEN & VEG (INCL CAR/DK GRN, NO POT), SOY SAUCE","Chicken or turkey and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27445120,"CHICKEN & VEG (NO CAR/DK GRN, NO POT), SOY SAUCE","Chicken or turkey and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27445125,"CHICKEN & VEG (INCL CAR/DK GRN, NO POT), TOMATO SAUCE","Chicken or turkey and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27445130,"CHICKEN & VEG (NO CAR/DK GRN, NO POT), TOMATO SAUCE","Chicken or turkey and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27445150,"GENERAL TSO CHICKEN","General Tso chicken"
+27445180,"MOO GOO GAI PAN","Moo Goo Gai Pan"
+27445220,"KUNG PAO CHICKEN","Kung pao chicken"
+27445250,"ALMOND CHICKEN","Almond chicken"
+27446100,"CHICKEN CHOW MEIN/CHOP SUEY, NO NOODLES","Chicken or turkey chow mein or chop suey, no noodles"
+27446200,"CHICKEN OR TURKEY SALAD, W/ MAYO","Chicken or turkey salad, made with mayonnaise"
+27446205,"CHICKEN/TURKEY SALAD WITH NUTS AND/OR FRUITS","Chicken or turkey salad with nuts and/or fruits"
+27446220,"CHICKEN SALAD W/ EGG","Chicken or turkey salad with egg"
+27446225,"CHICKEN OR TURKEY SALAD, W/ LT MAYO","Chicken or turkey salad, made with light mayonnaise"
+27446230,"CHICKEN OR TURKEY SALAD, W/ MAYO-TYPE DRSG","Chicken or turkey salad, made with mayonnaise-type salad dressing"
+27446235,"CHICKEN OR TURKEY SALAD, MADE W/ LT MAYO-TYPE DRSG","Chicken or turkey salad, made with light mayonnaise-type salad dressing"
+27446240,"CHICKEN OR TURKEY SALAD, W/CREAMY DRSG","Chicken or turkey salad, made with creamy dressing"
+27446245,"CHICKEN OR TURKEY SALAD, W/ LIT CREAMY DRSG","Chicken or turkey salad, made with light creamy dressing"
+27446250,"CHICKEN OR TURKEY SALAD, MADE W/ ITALIAN DRESSING","Chicken or turkey salad, made with Italian dressing"
+27446255,"CHICKEN OR TURKEY SALAD, MADE W/ LT ITALIAN DRSG","Chicken or turkey salad, made with light Italian dressing"
+27446260,"CHICKEN OR TURKEY SALAD, MADE W/ FAT FREE DRSG","Chicken or turkey salad, made with any type of fat free dressing"
+27446300,"CHICKEN GARDEN SALAD W/ TOMATO/CARROT, NO DRESSING","Chicken or turkey garden salad (chicken and/or turkey, tomato and/or carrots, other vegetables), no dressing"
+27446310,"CHICKEN GARDEN SALAD W/VEG, NO CAR/TOM, NO DRESSING","Chicken or turkey garden salad (chicken and/or turkey, other vegetables excluding tomato and carrots), no dressing"
+27446315,"CHICKEN GARDEN SALAD W/ BACON,CHEESE,TOMATO/CARROT,NO DRSG","Chicken or turkey garden salad with bacon and cheese (chicken and/or turkey, bacon, cheese, lettuce and/or greens, tomato and/or carrots, other vegetables), no dressing"
+27446320,"CHICKN(BRD,FRD)GARDEN SALAD W/ BACON,CHEESE,TOM/CAR,NO DRSG","Chicken or turkey (breaded, fried) garden salad with bacon and cheese (chicken and/or turkey, bacon, cheese, lettuce and/or greens, tomato and/or carrots, other vegetables), no dressing"
+27446330,"CHICKN GARDEN SALAD W/ CHEESE,TOM/CAR,NO DRSG","Chicken or turkey garden salad with cheese (chicken and/or turkey, cheese, lettuce and/or greens, tomato and/or carrots, other vegetables), no dressing"
+27446332,"CHICKN(BRD,FRD)GARDEN SALAD W/ CHEESE,TOM/CAR,NO DRSG","Chicken or turkey (breaded, fried) garden salad with cheese (chicken and/or turkey, cheese, lettuce and/or greens, tomato and/or carrots, other vegetables), no dressing"
+27446350,"ASIAN CHICKEN/TURKEY GARDEN SALAD, NO DRESSING","Asian chicken or turkey garden salad (chicken and/or turkey, lettuce, fruit, nuts), no dressing"
+27446355,"ASIAN CHICKEN GARDEN SALAD W/CRISPY NOODLES , NO DRESSING","Asian chicken or turkey garden salad with crispy noodles (chicken and/or turkey, lettuce, fruit, nuts, crispy noodles), no dressing"
+27446360,"CHICKEN/TURKEY CAESAR GARDEN SALAD, NO DRESSING","Chicken or turkey caesar garden salad (chicken and/or turkey, lettuce, tomato, cheese), no dressing"
+27446362,"CHICKEN/TURKEY (BREADED, FRIED) CAESAR GARDEN SALAD, NO DRSG","Chicken or turkey (breaded, fried) caesar garden salad (chicken and/or turkey, lettuce, tomatoes, cheese), no dressing"
+27446400,"CHICKEN & VEG (INCL CAR/DK GRN)(NO POT), CHEESE SCE","Chicken or turkey and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), cheese sauce (mixture)"
+27446410,"CHICKEN & VEG (NO CAR/DK GRN)(NO POT), CHEESE SAUCE","Chicken or turkey and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), cheese sauce (mixture)"
+27448020,"CHICKEN FRICASSEE W/SAUCE,NO POT,PUERTO RICAN STYLE","Chicken or turkey fricassee, with sauce, no potatoes, Puerto Rican style (potatoes reported separately)"
+27448030,"CHICKEN FRICASSEE, NO SCE OR POT,PUERTO RICAN STYLE","Chicken or turkey fricassee, no sauce, no potatoes, Puerto Rican style (sauce and potatoes reported separately)"
+27450010,"CRAB SALAD","Crab salad"
+27450020,"LOBSTER SALAD","Lobster salad"
+27450030,"SALMON SALAD","Salmon salad"
+27450040,"SHRIMP CHOW MEIN OR CHOP SUEY, NO NOODLES","Shrimp chow mein or chop suey, no noodles"
+27450060,"TUNA SALAD, W/MAYO","Tuna salad, made with mayonnaise"
+27450061,"TUNA SALAD, W/LT MAYO","Tuna salad, made with light mayonnaise"
+27450062,"TUNA SALAD, W/ MAYO-TYPE DRESSING","Tuna salad, made with mayonnaise-type salad dressing"
+27450063,"TUNA SALAD, W/ LT MAYO-TYPE DRSG","Tuna salad, made with light mayonnaise-type salad dressing"
+27450064,"TUNA SALAD, W/ CREAMY DRSG","Tuna salad, made with creamy dressing"
+27450065,"TUNA SALAD, W/ LT CREAMY DRSG","Tuna salad, made with light creamy dressing"
+27450066,"TUNA SALAD, W/ ITALIAN DRSG","Tuna salad, made with Italian dressing"
+27450067,"TUNA SALAD, W/ LT ITALIAN DRSG","Tuna salad, made with light Italian dressing"
+27450068,"TUNA SALAD, W/ ANY TYPE OF FAT FREE DRSG","Tuna salad, made with any type of fat free dressing"
+27450070,"SHRIMP SALAD","Shrimp salad"
+27450080,"SEAFOOD SALAD","Seafood salad"
+27450090,"TUNA SALAD W/ CHEESE","Tuna salad with cheese"
+27450100,"TUNA SALAD W/ EGG","Tuna salad with egg"
+27450110,"SHRIMP GARDEN SALAD W/ TOMATO/CARROT, NO DRESSING","Shrimp garden salad (shrimp, lettuce, eggs, tomato and/or carrots, other vegetables), no dressing"
+27450120,"SHRIMP GARDEN SALAD (NO TOMATO/CARROT, NO DRESSING)","Shrimp garden salad (shrimp, lettuce, eggs, vegetables excluding tomato and carrots), no dressing"
+27450130,"CRAB SALAD MADE W/ IMITATION CRAB","Crab salad made with imitation crab"
+27450150,"FISH, TOFU, & VEGETABLES, TEMPURA, HAWAIIAN","Fish, tofu, and vegetables, tempura, Hawaiian style (mixture)"
+27450180,"SEAFOOD GARDEN SALAD W/ VEG(NO TOM/CAR), NO DRESSING","Seafood garden salad with seafood, lettuce, vegetables excluding tomato and carrots, no dressing"
+27450190,"SEAFOOD GARDEN SALAD W/ TOM/CAR, NO DRESSING","Seafood garden salad with seafood, lettuce, tomato and/or carrots, other vegetables, no dressing"
+27450200,"SEAFOOD GARDEN SALAD W/ EGG, VEG, (NO CAR/TOM) NO  DRESSING","Seafood garden salad with seafood, lettuce, eggs, vegetables excluding tomato and carrots, no dressing"
+27450210,"SEAFOOD GARDEN SALAD W/EGG, TOM/CAR, NO DRESSING","Seafood garden salad with seafood, lettuce, eggs, tomato and/or carrots, other vegetables, no dressing"
+27450250,"OYSTERS ROCKEFELLER","Oysters Rockefeller"
+27450310,"LOMI SALMON","Lomi salmon"
+27450400,"SHRIMP & VEG (W/ CAR/DK GREEN, NO POT), NO SAUCE","Shrimp and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), no sauce (mixture)"
+27450405,"SHRIMP & VEG (NO CARROT/DK GREEN, NO POT), NO SAUCE","Shrimp and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), no sauce (mixture)"
+27450410,"SHRIMP & VEG (W/ CAR/DK GREEN, NO POT), SOY SAUCE","Shrimp and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27450420,"SHRIMP & VEG (NO CARROT/DK GREEN, NO POT),SOY SAUCE","Shrimp and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27450430,"SHRIMP SHISH KABOB WITH VEGETABLES, EXCLUDING POTATOES","Shrimp shish kabob with vegetables, excluding potatoes"
+27450450,"SHRIMP CREOLE, NO RICE","Shrimp creole, no rice"
+27450470,"KUNG PAO SHRIMP","Kung Pao shrimp"
+27450510,"TUNA CASSEROLE W/ VEG & SOUP, NO NOODLES","Tuna casserole with vegetables and (mushroom) soup, no noodles"
+27450600,"SHELLFISH MIXTURE & VEG (INCL CAR/DK GRN), SOY SCE","Shellfish mixture and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce"
+27450610,"SHELLFISH MIXTURE & VEG (NO CAR/DK GRN), SOY SAUCE","Shellfish mixture and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce"
+27450650,"SHELLFISH & VEG(INCL CAR/DK GRN)(NO POT),SOUP SAUCE","Shellfish mixture and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), (mushroom) soup (mixture)"
+27450660,"SHELLFISH & VEG(NO CAR/DK GRN/POT),SOUP-BASED SAUCE","Shellfish mixture and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), (mushroom) soup (mixture)"
+27450700,"FISH & VEG (INCL CAR/DK GRN, NO POT), TOMATO SAUCE","Fish and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), tomato-based sauce (mixture)"
+27450710,"FISH & VEG (NO CAR/DK GRN, NO POT), TOMATO SAUCE","Fish and vegetables (excluding carrots, broccoli, and dark- green leafy (no potatoes)), tomato-based sauce (mixture)"
+27450740,"FISH & VEGETABLES (W/ CAR/DK GRN), SOY-BASED SAUCE","Fish and vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27450750,"FISH & VEGETABLES (NO CAR/DK GRN), SOY-BASED SAUCE","Fish and vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes)), soy-based sauce (mixture)"
+27450760,"FISH SHISH KABOB WITH VEGETABLES, EXCLUDING POTATOES","Fish shish kabob with vegetables, excluding potatoes"
+27451010,"FRIED FISH W/ SAUCE, P.R. (PESCADO FRITO CON MOJO)","Fried fish with sauce, Puerto Rican style (Pescado frito con mojo)"
+27451030,"LOBSTER W/ SAUCE, P.R. (LANGOSTA A LA CRIOLLA)","Lobster with sauce, Puerto Rican style (Langosta a la criolla)"
+27451060,"OCTOPUS SALAD, P.R. (ENSALADA DE PULPO)","Octopus salad, Puerto Rican style (Ensalada de pulpo)"
+27451070,"CODFISH SALAD, P.R. (SERENATA)","Codfish salad, Puerto Rican style (Serenata)"
+27460010,"CHOW MEIN, NS AS TO TYPE OF MEAT, NO NOODLES","Chow mein or chop suey, NS as to type of meat, no noodles"
+27460100,"LAU LAU(PORK & FISH WRAPPED IN TARO/SPINACH LEAVES)","Lau lau (pork and fish wrapped in taro or spinach leaves)"
+27460490,"JULIENNE SALAD (MEAT, CHEESE, EGG, VEG) NO DRESSING","Julienne salad (meat, cheese, eggs, vegetables), no dressing"
+27460510,"ANTIPASTO W/ HAM, FISH, CHEESE, VEGETABLES","Antipasto with ham, fish, cheese, vegetables"
+27460710,"LIVERS, CHICKEN, CHOPPED, W/ EGGS & ONION (MIXTURE)","Livers, chicken, chopped, with eggs and onion (mixture)"
+27460750,"LIVER, BEEF OR CALVES, & ONIONS","Liver, beef or calves, and onions"
+27461010,"STEWED SEASONED GROUND BEEF, P.R.","Stewed seasoned ground beef, Puerto Rican style (Picadillo para relleno)"
+27462000,"STEWED CHITTERLINGS, P.R. (CUAJO GUISADO)","Stewed chitterlings, Puerto Rican style (cuajo guisado)"
+27463000,"STEWED GIZZARDS, P.R. (MOLLEJITAS GUISADAS)","Stewed gizzards, Puerto Rican style (Mollejitas guisadas)"
+27464000,"GUMBO, NO RICE (NEW ORLEANS TYPE W/MEAT, TOM, OKRA)","Gumbo, no rice (New Orleans type with shellfish, pork, and/or poultry, tomatoes, okra)"
+27500050,"SANDWICH, NFS","Sandwich, NFS"
+27500100,"MEAT SANDWICH, NFS","Meat sandwich, NFS"
+27500200,"WRAP SANDWICH, W/ MEAT, POULTRY OR FISH, VEGETABLES & CHEESE","Wrap sandwich, filled with meat, poultry, or fish, vegetables, and cheese"
+27500300,"WRAP SANDWICH, W/ MEAT, POULTRY OR FISH & VEGETABLES","Wrap sandwich, filled with meat, poultry, or fish, and vegetables"
+27510000,"BEEF SANDWICH, NFS","Beef sandwich, NFS"
+27510110,"BEEF BARBECUE SANDWICH OR SLOPPY JOE, ON BUN","Beef barbecue sandwich or Sloppy Joe, on bun"
+27510130,"BEEF BARBECUE SUBMARINE SANDWICH, ON BUN","Beef barbecue submarine sandwich, on bun"
+27510210,"CHEESEBURGER, PLAIN, ON BUN","Cheeseburger, plain, on bun"
+27510220,"CHEESEBURGER, W/ MAYO, ON BUN","Cheeseburger, with mayonnaise or salad dressing, on bun"
+27510230,"CHEESEBURGER, W/ MAYO & TOMATO/CATSUP, ON BUN","Cheeseburger, with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510240,"CHEESEBURGER, 1/4 LB MEAT, PLAIN, ON BUN","Cheeseburger, 1/4 lb meat, plain, on bun"
+27510250,"CHEESEBURGER, 1/4 LB MEAT, W/ MAYO, ON BUN","Cheeseburger, 1/4 lb meat, with mayonnaise or salad dressing, on bun"
+27510260,"CHEESEBURGER, 1/4 LB, W/ MUSHROOM SAUCE, ON BUN","Cheeseburger, 1/4 lb meat, with mushrooms in sauce, on bun"
+27510265,"DOUBLE CHEESEBURGER, PLAIN, ON MINIATURE BUN","Double cheeseburger, (2 patties, 1 oz each), plain, on miniature bun"
+27510270,"DOUBLE CHEESEBURGER, PLAIN, ON BUN","Double cheeseburger (2 patties), plain, on bun"
+27510280,"DOUBLE CHEESEBURGER, W/ MAYO, ON BUN","Double cheeseburger (2 patties), with mayonnaise or salad dressing, on bun"
+27510290,"DOUBLE CHEESEBURGER, PLAIN, ON DOUBLE-DECKER BUN","Double cheeseburger (2 patties), plain, on double-decker bun"
+27510300,"DOUBLE CHEESEBURGER, W/MAYO, ON DOUBLE-DECKER BUN","Double cheeseburger (2 patties), with mayonnaise or salad dressing, on double-decker bun"
+27510310,"CHEESEBURGER W/ TOMATO & OR CATSUP, ON BUN","Cheeseburger with tomato and/or catsup, on bun"
+27510311,"CHEESEBURGER, 1 OZ MEAT, PLAIN, ON MINI BUN","Cheeseburger, 1 oz meat, plain, on miniature bun"
+27510320,"CHEESEBURGER, 1/4 LB MEAT,W/ TOMATO/CATSUP, BUN","Cheeseburger, 1/4 lb meat, with tomato and/or catsup, on bun"
+27510330,"DOUBLE CHEESEBURGER W/TOMATO & OR CATSUP, ON BUN","Double cheeseburger (2 patties), with tomato and/or catsup, on bun"
+27510340,"DOUBLE CHEESEBURGER, W/ MAYO & TOMATO, ON BUN","Double cheeseburger (2 patties), with mayonnaise or salad dressing and tomatoes and/or catsup, on bun"
+27510350,"CHEESEBURGER, 1/4 LB MEAT, W/ MAYO & TOMATO/CATSUP, ON BUN","Cheeseburger, 1/4 lb meat, with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510355,"CHEESEBURGER, 1/3 LB MEAT, W/MAYO, TOMATO, ON BUN","Cheeseburger, 1/3 lb meat, with mayonnaise or salad dressing, tomato and/or catsup on bun"
+27510359,"CHEESEBURGER, 1/3 LB MEAT, W/MAYO, MUSHROOMS,ON BUN","Cheeseburger, 1/3 lb meat, with mayonnaise or salad dressing, and mushrooms, on bun"
+27510360,"BACON CHEESEBURGER, W/MAYO/SALAD DRSG, TOMATO/CATSUP,ON BUN","Bacon cheeseburger, with mayonnaise or salad dressing, tomato and/or catsup, on bun"
+27510370,"DOUBLE CHEESEBURGER W/ MAYONNAISE, ON BUN","Double cheeseburger (2 patties, 1/4 lb meat each), with mayonnaise or salad dressing, on bun"
+27510375,"DOUBLE CHEESEBURGER(2 PATTIES,1/4 LB EA) W/TOMATO/CATSUP/BUN","Double cheeseburger (2 patties, 1/4 lb meat each), with tomato and/or catsup, on bun"
+27510380,"TRIPLE CHEESEBURGER W/ MAYO, TOMATO, ON BUN","Triple cheeseburger (3 patties, 1/4 lb meat each), with mayonnaise or salad dressing and tomatoes and/or catsup, on bun"
+27510385,"DOUBLE BACON CHEESEBURGER (2 PATTIES), W/ TOMATO/CATSUP","Double bacon cheeseburger (2 patties), with tomato and/or catsup, on bun"
+27510390,"DOUBLE BACON CHEESEBURGER, ON BUN","Double bacon cheeseburger (2 patties, 1/4 lb meat each), on bun"
+27510400,"BACON CHEESEBURGER, 1/4 LB MEAT, W/ TOMATO, ON BUN","Bacon cheeseburger, 1/4 lb meat, with tomato and/or catsup, on bun"
+27510410,"CHILIBURGER, ON BUN (INCLUDE HAMBURGER W/ CHILI)","Chiliburger, on bun"
+27510420,"TACO BURGER, ON BUN (INCL CHILIBURGER W/ CHEESE)","Taco burger, on bun"
+27510425,"DOUBLE BACON CHEESEBURGER (2 PATTIES,1/4 LB EA), W/ MAYO/BUN","Double bacon cheeseburger (2 patties, 1/4 lb meat each), with mayonnaise or salad dressing, on bun"
+27510430,"DOUBLE BACON CHEESEBURGER, W/MAYO/DRSG,TOMATO/CATSUP,ON BUN","Double bacon cheeseburger (2 patties, 1/4 lb meat each), with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510435,"DOUBLE BACON CHEESEBURGER (2 PATTIES,1/3 LB EA) W/ MAYO/ BUN","Double bacon cheeseburger (2 patties,1/3 lb meat each), with mayonnaise or salad dressing, on bun"
+27510440,"BACON CHEESEBURGER, 1/4 LB, W/MAYO/DRSG,TOMATO/CATSUP,ON BUN","Bacon cheeseburger, 1/4 lb meat, with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510445,"BACON CHEESEBURGER, 1/3 LB MEAT, W/TOMATO +/OR CATSUP,","Bacon cheeseburger, 1/3 lb meat, with tomato and/or catsup, on bun"
+27510450,"CHEESEBURGER, 1/4 LB MEAT, W/ HAM, ON BUN","Cheeseburger, 1/4 lb meat, with ham, on bun"
+27510480,"CHEESEBURGER, W/ ONIONS, ON RYE BUN","Cheeseburger (hamburger with cheese sauce), 1/4 lb meat, with grilled onions, on rye bun"
+27510500,"HAMBURGER, PLAIN, ON BUN","Hamburger, plain, on bun"
+27510510,"HAMBURGER, W/ TOMATO & OR CATSUP, ON BUN","Hamburger, with tomato and/or catsup, on bun"
+27510520,"HAMBURGER, W/ MAYO & TOMATO/CATSUP, ON BUN","Hamburger, with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510530,"HAMBURGER, 1/4 LB MEAT, PLAIN, ON BUN","Hamburger, 1/4 lb meat, plain, on bun"
+27510540,"DOUBLE HAMBURGER W/TOMATO & OR CATSUP, ON BUN","Double hamburger (2 patties), with tomato and/or catsup, on bun"
+27510550,"DOUBLE HAMBURGER W/ MAYO & TOMATO, DBL-DECKER BUN","Double hamburger (2 patties), with mayonnaise or salad dressing and tomatoes, on double-decker bun"
+27510560,"HAMBURGER, 1/4 LB MEAT W/ MAYO & TOMATO/CATSUP, ON BUN","Hamburger, 1/4 lb meat, with mayonnaise or salad dressing, and tomato and/or catsup, on bun"
+27510570,"HAMBURGER, 2.5 OZ MEAT, W/ MAYO & TOMATO, ON BUN","Hamburger, 2-1/2 oz meat, with mayonnaise or salad dressing and tomatoes, on bun"
+27510590,"HAMBURGER, W/ MAYO, ON BUN","Hamburger, with mayonnaise or salad dressing, on bun"
+27510600,"HAMBURGER, 1 OZ MEAT,PLAIN, ON MINIATURE BUN","Hamburger, 1 oz meat, plain, on miniature bun"
+27510610,"HAMBURGER, 1 OZ MEAT, TOMATO, ON MINIATURE BUN","Hamburger, 1 oz meat, with tomato and/or catsup, on miniature bun"
+27510620,"HAMBURGER, 1/4 LB MEAT, W/ TOMATO & OR CATSUP, BUN","Hamburger, 1/4 lb meat, with tomato and/or catsup, on bun"
+27510630,"HAMBURGER, 1/4 LB MEAT, W/ MAYO, ON BUN","Hamburger, 1/4 lb meat, with mayonnaise or salad dressing, on bun"
+27510650,"DOUBLE HAMBURGER, PLAIN, ON BUN","Double hamburger (2 patties), plain, on bun"
+27510660,"DOUBLE HAMBURGER, W/ MAYO, ON BUN","Double hamburger (2 patties), with mayonnaise or salad dressing, on bun"
+27510670,"DOUBLE HAMBURGER, W/ MAYO & TOMATO, ON BUN","Double hamburger (2 patties), with mayonnaise or salad dressing and tomatoes, on bun"
+27510680,"DOUBLE HAMBURGER (1/2 LB MEAT), W/ TOM/CATSUP, BUN","Double hamburger (2 patties, 1/4 lb meat each), with tomato and/or catsup, on bun"
+27510690,"DOUBLE HAMBURGER,1/2 LB MEAT,W/MAYO&TOM/CATSUP,BUN","Double hamburger (2 patties, 1/4 lb meat each), with mayonnaise or salad dressing and tomatoes and/or catsup, on double-decker bun"
+27510700,"MEATBALL & SPAG SAU SUB SAND","Meatball and spaghetti sauce submarine sandwich"
+27510710,"PIZZABURGER (HAMBURGER, CHEESE, SAUCE), ON 1/2 BUN","Pizzaburger (hamburger, cheese, sauce) on 1/2 bun"
+27510720,"PIZZABURGER (HAMBURGER, CHEESE, SAUCE), WHOLE BUN","Pizzaburger (hamburger, cheese, sauce) on whole bun"
+27510910,"CORNED BEEF SANDWICH","Corned beef sandwich"
+27510950,"REUBEN(CORN BEEF W/ SAUERKRAUT & CHEESE) W/ SPREAD","Reuben sandwich (corned beef sandwich with sauerkraut and cheese), with spread"
+27511010,"PASTRAMI SANDWICH","Pastrami sandwich"
+27513010,"ROAST BEEF SANDWICH","Roast beef sandwich"
+27513020,"ROAST BEEF SANDWICH, W/ GRAVY","Roast beef sandwich, with gravy"
+27513040,"ROAST BEEF SUB SAND, W/ LETT, TOM, SPRD","Roast beef submarine sandwich, with lettuce, tomato and spread"
+27513041,"ROAST BEEF SUB SAND, W/ CHEESE, LETTUCE, TOMATO, SPRD","Roast beef submarine sandwich, with cheese, lettuce, tomato and spread"
+27513050,"ROAST BEEF SANDWICH W/ CHEESE","Roast beef sandwich with cheese"
+27513060,"ROAST BEEF SANDWICH W/ BACON & CHEESE SAUCE","Roast beef sandwich with bacon and cheese sauce"
+27513070,"ROAST BEEF SUBMARINE SANDWICH, ON ROLL, AU JUS","Roast beef submarine sandwich, on roll, au jus"
+27515000,"STEAK SUBMARINE SANDWICH WITH LETTUCE AND TOMATO","Steak submarine sandwich with lettuce and tomato"
+27515010,"STEAK SANDWICH, PLAIN, ON ROLL","Steak sandwich, plain, on roll"
+27515020,"STEAK , CHEESE SUB SAND, W/ LETT, TOM","Steak and cheese submarine sandwich, with lettuce and tomato"
+27515030,"STEAK & CHEESE SANDWICH, PLAIN, ON ROLL","Steak and cheese sandwich, plain, on roll"
+27515040,"STEAK & CHEESE SUBMARINE SANDWICH, PLAIN, ON ROLL","Steak and cheese submarine sandwich, plain, on roll"
+27515050,"FAJITA-STYLE BEEF SAND W/ CHEESE,PITA BRD,W/LET+TOM","Fajita-style beef sandwich with cheese, on pita bread, with lettuce and tomato"
+27515070,"STEAK & CHEESE SUB, FRIED PEP & ONIONS, ON ROLL","Steak and cheese submarine sandwich, with fried peppers and onions, on roll"
+27515080,"STEAK SANDWICH, PLAIN, ON BISCUIT","Steak sandwich, plain, on biscuit"
+27516010,"GYRO SANDWICH W/ TOMATO & SPREAD","Gyro sandwich (pita bread, beef, lamb, onion, condiments), with tomato and spread"
+27517000,"WRAP SANDWICH FILLED WITH BEEF PATTY, CHEESE, LETTUCE,SPREAD","Wrap sandwich filled with beef patty, cheese and spread and/or sauce"
+27517010,"WRAP SANDWICH FILLED WITH BEEF PATTY, CHEESE, TOMATO, SPREAD","Wrap sandwich filled with beef patty, cheese, tomato and/or catsup, and spread and/or sauce"
+27518000,"WRAP SAND W/BEEF PATTY,BAC, CHS, TOM,SPREAD","Wrap sandwich filled with beef patty, bacon, cheese, tomato and/or catsup, and spread and/or sauce"
+27520110,"BACON SANDWICH W/ SPREAD","Bacon sandwich, with spread"
+27520120,"BACON & CHEESE SANDWICH, W/ SPREAD","Bacon and cheese sandwich, with spread"
+27520130,"BACON, CHICK, & TOM CLUB SANDWICH W/ LETTUCE+SPREAD","Bacon, chicken, and tomato club sandwich, with lettuce and spread"
+27520135,"BACON, CHICKN & TOMATO CLUB SANDWICH W/CHEESE, LETTUCE &SPRD","Bacon, chicken, and tomato club sandwich, with cheese, lettuce and spread"
+27520140,"BACON & EGG SANDWICH","Bacon and egg sandwich"
+27520150,"BACON, LETTUCE, & TOMATO SANDWICH W/ SPREAD","Bacon, lettuce, and tomato sandwich with spread"
+27520160,"BACON,CHICK,&TOMATO CLUB SANDWICH,MULTIGR W/ SPREAD","Bacon, chicken, and tomato club sandwich, on multigrain roll with lettuce and spread"
+27520165,"BACON, CHICK FILLET (BRD, FRIED),& TOM CLUB W/LETTUCE & SPRD","Bacon, chicken fillet (breaded, fried), and tomato club with lettuce and spread"
+27520166,"BACON, CHICK FILLET (BRD,FRIED),&TOM CLUB W/CHS,LETTUCE&SPRD","Bacon, chicken fillet (breaded, fried), and tomato club sandwich with cheese, lettuce and spread"
+27520170,"BACON ON BISCUIT","Bacon on biscuit"
+27520250,"HAM ON BISCUIT","Ham on biscuit"
+27520300,"HAM SANDWICH W/ SPREAD","Ham sandwich, with spread"
+27520310,"HAM SANDWICH W/ LETTUCE & SPREAD","Ham sandwich with lettuce and spread"
+27520320,"HAM & CHEESE SANDWICH, W/ LETTUCE & SPREAD","Ham and cheese sandwich, with lettuce and spread"
+27520330,"HAM & EGG SANDWICH","Ham and egg sandwich"
+27520340,"HAM SALAD SANDWICH","Ham salad sandwich"
+27520350,"HAM & CHEESE SANDWICH W/ SPREAD, GRILLED","Ham and cheese sandwich, with spread, grilled"
+27520360,"HAM & CHEESE SANDWICH ON BUN W/ LETTUCE & SPREAD","Ham and cheese sandwich, on bun, with lettuce and spread"
+27520370,"HOT HAM & CHEESE SANDWICH, ON BUN","Hot ham and cheese sandwich, on bun"
+27520380,"HAM & CHEESE ON ENGLISH MUFFIN","Ham and cheese on English muffin"
+27520390,"HAM & CHEESE SUB, W/ LETTUCE, TOMATO & SPREAD","Ham and cheese submarine sandwich, with lettuce, tomato and spread"
+27520410,"CUBAN SAND, P.R STYLE(SANDWICH CUBANA), W/ SPREAD","Cuban sandwich, (Sandwich cubano), with spread"
+27520420,"MIDNIGHT SAND,P.R. STYLE (MEDIA NOCHE), W/ SPREAD","Midnight sandwich, (Media noche), with spread"
+27520500,"PORK SANDWICH, ON WHITE ROLL, W/ ONIONS, PICKLES & BBQ SAUCE","Pork sandwich, on white roll, with onions, dill pickles and barbecue sauce"
+27520510,"PORK BARBECUE SANDWICH OR SLOPPY JOE, ON BUN","Pork barbecue sandwich or Sloppy Joe, on bun"
+27520520,"PORK SANDWICH","Pork sandwich"
+27520530,"PORK SANDWICH W/ GRAVY","Pork sandwich, with gravy"
+27520540,"HAM & TOMATO CLUB SAND, W/ SPREAD","Ham and tomato club sandwich, with lettuce and spread"
+27540110,"CHICKEN SANDWICH, W/ SPREAD","Chicken sandwich, with spread"
+27540111,"CHICKEN SANDWICH, W/ CHEESE & SPREAD","Chicken sandwich, with cheese and spread"
+27540120,"CHICKEN SALAD OR CHICKEN SPREAD SANDWICH","Chicken salad or chicken spread sandwich"
+27540130,"CHICKEN BARBECUE SANDWICH","Chicken barbecue sandwich"
+27540140,"CHICKEN FILLET (BREADED, FRIED) SANDWICH","Chicken fillet (breaded, fried) sandwich"
+27540145,"CHICKEN FILLET (BREADED, FRIED) SANDWICH ON BISCUIT","Chicken fillet (breaded, fried) sandwich on biscuit"
+27540150,"CHICKEN FILLET(BR FRIED) SAND W/ LET, TOM & SPREAD","Chicken fillet (breaded, fried) sandwich with lettuce, tomato and spread"
+27540151,"CHICKEN FILLET(BRD, FRIED) SAND W/ CHEESE, LETT, TOM & SPRD","Chicken fillet (breaded, fried) sandwich with cheese, lettuce, tomato and spread"
+27540170,"CHICKEN PATTY SANDWICH, MINI, W/ SPREAD","Chicken patty sandwich, miniature, with spread"
+27540180,"CHICKEN PATTY SANDWICH ON BISCUIT","Chicken patty sandwich or biscuit"
+27540190,"CHICKEN PATTY SANDWICH W/ LETTUCE & SPREAD","Chicken patty sandwich, with lettuce and spread"
+27540200,"FAJITA-STYLE CHICKEN SANDWICH W/ CHEESE, LETTUC,TOM","Fajita-style chicken sandwich with cheese, on pita bread, with lettuce and tomato"
+27540210,"WRAP SNDWCH W/CHICK STRIPS(BREADED,FRIED),CHS,LETTUCE & SPRD","Wrap sandwich filled with chicken strips (breaded, fried), cheese, lettuce, and spread"
+27540230,"CHICKEN PATTY SAND W/ CHEES,WHEAT BUN,LET,TOM, SPRE","Chicken patty sandwich with cheese, on wheat bun, with lettuce, tomato and spread"
+27540235,"CHICKEN FILLET, BROILED, SANDWICH WITH LETTUCE, TOMATO, AND","Chicken fillet, broiled, sandwich with lettuce, tomato, and spread"
+27540240,"CHICKEN FILLET,(BROIL) SAND W/ LET, TOM, & SPREAD","Chicken fillet, (broiled), sandwich, on whole wheat roll, with lettuce, tomato and spread"
+27540250,"CHICK FILLET,BROIL,SANDWICH,W/CHEESE,WW ROLL","Chicken fillet, broiled, sandwich with cheese, on whole wheat roll, with lettuce, tomato and non-mayonnaise type spread"
+27540260,"CHICK FILLET, BROILED,SANDWICH,ON OAT BRAN BUN(LTS)","Chicken fillet, broiled, sandwich, on oat bran bun, with lettuce, tomato, spread"
+27540270,"CHICKEN FILLET,SANDWICH,W/LETT,TOM,&NON-MAYO SPREAD","Chicken fillet, broiled, sandwich, with lettuce, tomato, and non-mayonnaise type spread"
+27540280,"CHICKEN FILLET,BROILED,SANDWICH,W/CHEESE,ON BUN","Chicken fillet, broiled, sandwich with cheese, on bun, with lettuce, tomato and spread"
+27540290,"CHICKEN SUB SANDWICH, W/ LETTUCE, TOMATO & SPREAD","Chicken submarine sandwich, with lettuce, tomato, and spread"
+27540291,"CHICKEN SUB SANDWICH, W/ CHEESE, LETTUCE, TOMATO & SPREAD","Chicken submarine sandwich, with cheese, lettuce, tomato, and spread"
+27540300,"WRAP SNDWCH W/CHICK STRIPS (BROILED),CHS,LETTUCE & SPRD","Wrap sandwich filled with chicken strips (broiled), cheese, lettuce, and spread"
+27540310,"TURKEY SANDWICH W/ SPREAD","Turkey sandwich, with spread"
+27540320,"TURKEY SALAD SANDWICH","Turkey salad or turkey spread sandwich"
+27540330,"TURKEY SANDWICH W/ GRAVY","Turkey sandwich, with gravy"
+27540350,"TURKEY SUB SAND, W/ CHEESE, LETT, TOM, SPRD","Turkey submarine sandwich, with cheese, lettuce, tomato and spread"
+27541000,"TURKEY, HAM & ROAST BEEF CLUB SANDWCH W/LETT,TOM,SPRD","Turkey, ham, and roast beef club sandwich, with lettuce, tomato and spread"
+27541001,"TURKEY, HAM & ROAST BEEF CLUB SANDWCH W/CHEESE,LETT,TOM,SPRD","Turkey, ham, and roast beef club sandwich with cheese, lettuce, tomato, and spread"
+27550000,"FISH SANDWICH, ON BUN, W/ SPREAD","Fish sandwich, on bun, with spread"
+27550100,"FISH SANDWICH, ON BUN, W/ CHEESE AND SPREAD","Fish sandwich, on bun, with cheese and spread"
+27550110,"CRAB CAKE SANDWICH, ON BUN","Crab cake sandwich, on bun"
+27550510,"SARDINE SANDWICH, W/ LETTUCE & SPREAD","Sardine sandwich, with lettuce and spread"
+27550710,"TUNA SALAD SANDWICH W/ LETTUCE","Tuna salad sandwich, with lettuce"
+27550720,"TUNA SALAD SANDWICH","Tuna salad sandwich"
+27550730,"TUNA MELT SANDWICH","Tuna melt sandwich"
+27550750,"TUNA SALSUB SAND, W/ LETT & TOMATO","Tuna salad submarine sandwich, with lettuce and tomato"
+27550751,"TUNA SALAD SUB SANDWCH, W/ CHEESE, LETTUCE & TOMATO","Tuna salad submarine sandwich, with cheese, lettuce and tomato"
+27560000,"LUNCHEON MEAT SANDWICH, NFS, W/ SPREAD","Luncheon meat sandwich, NFS, with spread"
+27560110,"BOLOGNA SANDWICH, W/ SPREAD","Bologna sandwich, with spread"
+27560120,"BOLOGNA & CHEESE SANDWICH W/ SPREAD","Bologna and cheese sandwich, with spread"
+27560300,"CORN DOG (FRANKFURTER/HOT DOG W/ CORNBREAD COATING)","Corn dog (frankfurter or hot dog with cornbread coating)"
+27560350,"PIG IN A BLANKET (FRANKFURTER OR HOT DOG  WRAPPED IN DOUGH)","Pig in a blanket (frankfurter or hot dog wrapped in dough)"
+27560410,"PUERTO RICAN SANDWICH, P.R. (SANDWICH CRIOLLO)","Puerto Rican sandwich (Sandwich criollo)"
+27560500,"PEPPERONI, SALAMI SUBM SANDWICH, WITH LETTUCE, TOM, SPREAD","Pepperoni and salami submarine sandwich, with lettuce, tomato, and spread"
+27560510,"SALAMI SANDWICH W/ SPREAD","Salami sandwich, with spread"
+27560650,"SAUSAGE ON BISCUIT(INCL JIMMY DEAN SAUSAGE BISCUIT)","Sausage on biscuit"
+27560660,"SAUSAGE GRIDDLE CAKE SANDWICH","Sausage griddle cake sandwich"
+27560670,"SAUSAGE & CHEESE ON ENGLISH MUFFIN","Sausage and cheese on English muffin"
+27560705,"SAUSAGE BALLS (MADE W/ BISCUIT MIX & CHEESE)","Sausage balls (made with biscuit mix and cheese)"
+27560710,"SAUSAGE SANDWICH","Sausage sandwich"
+27560720,"SAUSAGE & SPAGH SAUCE SANDWICH","Sausage and spaghetti sauce sandwich"
+27560910,"COLD CUT SUB SANDWICH, W/ CHEESE, LETTUCE, TOMATO, SPRD","Cold cut submarine sandwich, with cheese, lettuce, tomato, and spread"
+27563010,"MEAT SPREAD OR POTTED MEAT SANDWICH","Meat spread or potted meat sandwich"
+27564000,"FRANKFURTER OR HOT DOG, NFS, PLAIN, ON BUN","Frankfurter or hot dog sandwich, NFS, plain, on bun"
+27564010,"FRANKFURTER OR HOT DOG, NFS, PLAIN, ON WHITE BREAD","Frankfurter or hot dog sandwich, NFS, plain, on white bread"
+27564020,"FRANKFURTER OR HOT DOG, NFS, PLAIN, ON WHEAT BREAD","Frankfurter or hot dog sandwich, NFS, plain, on wheat bread"
+27564030,"FRANKFURTER/HOT DOG, NFS, PLAIN, WHL WHT BREAD, NS TO 100%","Frankfurter or hot dog sandwich, NFS, plain, on whole wheat bread, NS as to 100%"
+27564040,"FRANKFURTER OR HOT DOG, NFS, PLAIN, ON WHOLE GRAIN WHITE BRE","Frankfurter or hot dog sandwich, NFS, plain, on whole grain white bread"
+27564050,"FRANKFURTER OR HOT DOG, NFS, PLAIN, ON MULTIGRAIN BREAD","Frankfurter or hot dog sandwich, NFS, plain, on multigrain bread"
+27564060,"FRANKFURTER OR HOT DOG, BEEF, PLAIN, ON BUN","Frankfurter or hot dog sandwich, beef, plain,on bun"
+27564070,"FRANKFURTER OR HOT DOG, PLAIN, WHITE BREAD","Frankfurter or hot dog sandwich, beef, plain, on white bread"
+27564080,"FRANKFURTER OR HOT DOG, PLAIN, WHEAT BREAD","Frankfurter or hot dog sandwich, beef, plain, on wheat bread"
+27564090,"FRANKFURTER/ HOT DOG, BEEF,PLAIN, WHOLE WHEAT BREAD, NS 100%","Frankfurter or hot dog sandwich, beef, plain, on whole wheat bread, NS as to 100%"
+27564100,"FRANKFURTER OR HOT DOG, PLAIN, ON WHOLE GRAIN WHITE","Frankfurter or hot dog sandwich, beef, plain, on whole grain white bread"
+27564110,"FRANKFURTER OR HOT DOG, BEEF, PLAIN, ON MULTIGRAIN BREAD","Frankfurter or hot dog sandwich, beef, plain, on multigrain bread"
+27564120,"FRANKFURTER OR HOT DOG, BEEF/PORK, PLAIN, ON BUN","Frankfurter or hot dog sandwich, beef and pork, plain, on bun"
+27564130,"FRANKFURTER OR HOT DOG, BEEF & PORK, PLAIN, ON WHITE BREAD","Frankfurter or hot dog sandwich, beef and pork, plain, on white bread"
+27564140,"FRANKFURTER OR HOT DOG, BEEF/PORK, PLAIN, ON WHEAT BREAD","Frankfurter or hot dog sandwich, beef and pork, plain, on wheat bread"
+27564150,"FRANKFURTER/HOT DOG, BEEF&PORK,PLAIN,ON WHL WHT, NS TO 100%","Frankfurter or hot dog sandwich, beef and pork, plain, on whole wheat bread, NS as to 100%"
+27564160,"FRANKFURTER OR HOT DOG, BEEF/PORK, PLAIN, WHOLE GRAIN WHITE","Frankfurter or hot dog sandwich, beef and pork, plain, on whole grain white bread"
+27564170,"FRANKFURTER OR HOT DOG, BEEF/PORK, PLAIN, MULTIGRAIN BREAD","Frankfurter or hot dog sandwich, beef and pork, plain, on multigrain bread"
+27564180,"FRANKFURTER OR HOT DOG, MEAT/POULTRY, PLAIN, ON BUN","Frankfurter or hot dog sandwich, meat and poultry, plain, on bun"
+27564190,"FRANKFURTER OR HOT DOG, MEAT AND POULTRY, PLAIN, ON WHITE BR","Frankfurter or hot dog sandwich, meat and poultry, plain, on white bread"
+27564200,"FRANKFURTER OR HOT DOG, MEAT AND POULTRY, PLAIN, ON WHEAT BR","Frankfurter or hot dog sandwich, meat and poultry, plain, on wheat bread"
+27564210,"FRANKFURTER/HOT DOG, MEAT&POULTRY,PLAIN,WHL WHT,NS TO 100%","Frankfurter or hot dog sandwich, meat and poultry, plain, on whole wheat bread, NS as to 100%"
+27564220,"FRANKFURTER OR HOT DOG, MEAT AND POULTRY, PLAIN, ON WHOLE GR","Frankfurter or hot dog sandwich, meat and poultry, plain, on whole grain white bread"
+27564230,"FRANKFURTER OR HOT DOG, MEAT AND POULTRY, PLAIN, ON MULTIGRA","Frankfurter or hot dog sandwich, meat and poultry, plain, on multigrain bread"
+27564240,"FRANKFURTER OR HOT DOG, CHICKEN AND/OR TURKEY, PLAIN, ON BUN","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on bun"
+27564250,"FRANKFURTER OR HOT DOG, CHICKEN / TURKEY, PLAIN, ON WHITE BR","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on white bread"
+27564260,"FRANKFURTER OR HOT DOG, CHICKEN AND/OR TURKEY, PLAIN, ON WHE","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on wheat bread"
+27564270,"FRANKFURTER/HOT DOG, CHICK/TURKEY,PLAIN,WHL WHT,NS TO 100%","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on whole wheat bread, NS as to 100%"
+27564280,"FRANKFURTER OR HOT DOG, CHICKEN AND/OR TURKEY, PLAIN, ON WHO","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on whole grain white bread"
+27564290,"FRANKFURTER OR HOT DOG, CHICKEN AND/OR TURKEY, PLAIN, ON MUL","Frankfurter or hot dog sandwich, chicken and/or turkey, plain, on multigrain bread"
+27564300,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, PLAIN, ON BUN","Frankfurter or hot dog sandwich, reduced fat or light, plain, on bun"
+27564310,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, PLAIN, ON WHIT","Frankfurter or hot dog sandwich, reduced fat or light, plain, on white bread"
+27564320,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, PLAIN, ON WHEA","Frankfurter or hot dog sandwich, reduced fat or light, plain, on wheat bread"
+27564330,"FRANKFURTER/HOT DOG, RED FAT/LIGHT,PLAIN,WHL WHT,NS TO 100%","Frankfurter or hot dog sandwich, reduced fat or light, plain, on whole wheat bread, NS as to 100%"
+27564340,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, PLAIN, ON WHOL","Frankfurter or hot dog sandwich, reduced fat or light, plain, on whole grain white bread"
+27564350,"FRANKFURTER OR HOT DOG, REDUCED FAT OR LIGHT, PLAIN, ON MULT","Frankfurter or hot dog sandwich, reduced fat or light, plain, on multigrain bread"
+27564360,"FRANKFURTER OR HOT DOG, FAT FREE, PLAIN, ON BUN","Frankfurter or hot dog sandwich, fat free, plain, on bun"
+27564370,"FRANKFURTER OR HOT DOG, FAT FREE, PLAIN, ON WHITE BREAD","Frankfurter or hot dog sandwich, fat free, plain, on white bread"
+27564380,"FRANKFURTER OR HOT DOG, FAT FREE, PLAIN, ON WHEAT BREAD","Frankfurter or hot dog sandwich, fat free, plain, on wheat bread"
+27564390,"FRANKFURTER/HOT DOG, FAT FREE, PLAIN, WHL WHT, NS TO 100%","Frankfurter or hot dog sandwich, fat free, plain, on whole wheat bread, NS as to 100%"
+27564400,"FRANKFURTER OR HOT DOG, FAT FREE, PLAIN, ON WHOLE GRAIN WHIT","Frankfurter or hot dog sandwich, fat free, plain, on whole grain white bread"
+27564410,"FRANKFURTER OR HOT DOG, FAT FREE, PLAIN, ON MULTIGRAIN BREAD","Frankfurter or hot dog sandwich, fat free, plain, on multigrain bread"
+27564420,"FRANKFURTER OR HOT DOG, MEATLESS, PLAIN, ON BUN","Frankfurter or hot dog sandwich, meatless, plain, on bun"
+27564430,"FRANKFURTER OR HOT DOG, MEATLESS, PLAIN, ON BREAD","Frankfurter or hot dog sandwich, meatless, plain, on bread"
+27564440,"FRANKFURTER OR HOT DOG, WITH CHILI, ON BUN","Frankfurter or hot dog sandwich, with chili, on bun"
+27564450,"FRANKFURTER OR HOT DOG, WITH CHILI, ON WHITE BREAD","Frankfurter or hot dog sandwich, with chili, on white bread"
+27564460,"FRANKFURTER OR HOT DOG, WITH CHILI, ON WHEAT BREAD","Frankfurter or hot dog sandwich, with chili, on wheat bread"
+27564470,"FRANKFURTER/HOT DOG, W/CHILI, ON WHL WHT BREAD, NS TO 100%","Frankfurter or hot dog sandwich, with chili, on whole wheat bread, NS as to 100%"
+27564480,"FRANKFURTER OR HOT DOG, WITH CHILI, ON WHOLE GRAIN WHITE BRE","Frankfurter or hot dog sandwich, with chili, on whole grain white bread"
+27564490,"FRANKFURTER OR HOT DOG, WITH CHILI, ON MULTI-GRAIN BREAD","Frankfurter or hot dog sandwich, with chili, on multi-grain bread"
+27564500,"FRANKFURTER OR HOT DOG, W/ VEGETARIAN CHILI, ON BUN","Frankfurter or hot dog sandwich, with vegetarian chili, on bun"
+27564510,"FRANKFURTER OR HOT DOG, W/ VEGETARIAN CHILI, ON WHITE BREAD","Frankfurter or hot dog sandwich, with vegetarian chili, on white bread"
+27564520,"FRANKFURTER OR HOT DOG, W/ VEGETARIAN CHILI, ON WHEAT BREAD","Frankfurter or hot dog sandwich, with vegetarian chili, on wheat bread"
+27564530,"FRANKFURTER/HOT DOG, W/MEATLESS CHILI, ON WHL WHT,NS TO 100%","Frankfurter or hot dog sandwich, with meatless chili, on whole wheat bread, NS as to 100%"
+27564540,"FRANKFURTER OR HOT DOG, W/ VEGETARIAN CHILI, ON WHOLE GRAIN","Frankfurter or hot dog sandwich, with vegetarian chili, on whole grain white bread"
+27564550,"FRANKFURTER OR HOT DOG, W/ VEGETARIAN CHILI, ON MULTIGRAIN B","Frankfurter or hot dog sandwich, with vegetarian chili, on multigrain bread"
+27564560,"FRANKFURTER OR HOT DOG, MEATLESS, ON BUN, WITH CHILI","Frankfurter or hot dog sandwich, meatless, on bun, with vegetarian chili"
+27564570,"FRANKFURTER OR HOT DOG, MEATLESS, ON BREAD, WITH CHILI","Frankfurter or hot dog sandwich, meatless, on bread, with vegetarian chili"
+27570310,"HORS D'OEUVRES, W/ SPREAD","Hors d'oeuvres, with spread"
+27601000,"BEEF STEW, BABY FOOD, TODDLER","Beef stew, baby food, toddler"
+27610100,"BEEF & EGG NOODLES, BABY, NS AS TO STR OR JR","Beef and egg noodles, baby food, NS as to strained or junior"
+27610110,"BEEF & EGG NOODLES, BABY, STRAINED","Beef and egg noodles, baby food, strained"
+27610120,"BEEF & EGG NOODLES, BABY, JUNIOR","Beef and egg noodles, baby food, junior"
+27610710,"BEEF W/ VEGETABLES, BABY, STRAINED","Beef with vegetables, baby food, strained"
+27610730,"BEEF W/ VEGETABLES, BABY FOOD, TODDLER","Beef with vegetables, baby food, toddler"
+27640050,"CHICKEN & RICE DINNER, BABY, STRAINED","Chicken and rice dinner, baby food, strained"
+27640100,"CHICKEN NOODLE DINNER, BABY, NS AS TO STR OR JR","Chicken noodle dinner, baby food, NS as to strained or junior"
+27640110,"CHICKEN NOODLE DINNER, BABY, STRAINED","Chicken noodle dinner, baby food, strained"
+27640120,"CHICKEN NOODLE DINNER, BABY, JUNIOR","Chicken noodle dinner, baby food, junior"
+27640810,"CHICKEN, NOODLES & VEGETABLES, BABY, TODDLER","Chicken, noodles, and vegetables, baby food, toddler"
+27641000,"CHICKEN STEW, BABY FOOD, TODDLER","Chicken stew, baby food, toddler"
+27642100,"TURKEY, RICE & VEGETABLES, BABY, NS AS TO STR OR JR","Turkey, rice and vegetables, baby food, NS as to strained or junior"
+27642110,"TURKEY, RICE & VEGETABLES, BABY, STRAINED","Turkey, rice and vegetables, baby food, strained"
+27642120,"TURKEY, RICE & VEGETABLES, BABY, JUNIOR","Turkey, rice and vegetables, baby food, junior"
+27642130,"TURKEY, RICE, & VEGETABLES, BABY, TODDLER","Turkey, rice, and vegetables, baby food, toddler"
+27644110,"CHICKEN SOUP, BABY","Chicken soup, baby food"
+28101000,"FROZEN DINNER, NFS","Frozen dinner, NFS"
+28110000,"BEEF DINNER, NFS (FROZEN)","Beef dinner, NFS (frozen meal)"
+28110120,"BEEF W/ POTATOES (FROZEN MEAL, LARGE MEAT PORTION)","Beef with potatoes (frozen meal, large meat portion)"
+28110150,"BEEF W/ VEGETABLE (DIET FROZEN MEAL)","Beef with vegetable (diet frozen meal)"
+28110220,"SIRLOIN, CHOPPED, W/ GRAVY, POT, VEG (FROZEN MEAL)","Sirloin, chopped, with gravy, mashed potatoes, vegetable (frozen meal)"
+28110250,"SIRLOIN TIPS W/ GRAVY, POTATOES, VEG (FROZEN MEAL)","Sirloin tips with gravy, potatoes, vegetable (frozen meal)"
+28110270,"SIRLOIN BEEF W/ GRAVY, POTATOES, VEG (FROZ MEAL)","Sirloin beef with gravy, potatoes, vegetable (frozen meal)"
+28110300,"SALISBURY STEAK DINNER, NFS (FROZEN)","Salisbury steak dinner, NFS (frozen meal)"
+28110310,"SALISBURY STEAK W/ GRAVY, POTATOES, VEG (FROZ MEAL)","Salisbury steak with gravy, potatoes, vegetable (frozen meal)"
+28110330,"SALISBURY STEAK, GRAVY, POT, VEG, DESSERT(FRZ MEAL)","Salisbury steak with gravy, whipped potatoes, vegetable, dessert (frozen meal)"
+28110340,"SALISBURY STK, GRAVY,POT,VEG,SOUP,DESSERT(FRZ MEAL)","Salisbury steak with gravy, potatoes, vegetable, soup or macaroni and cheese, dessert (frozen meal)"
+28110350,"SALISBURY STEAK, POT,VEG,DESSERT(FROZ MEAL,LG MEAT)","Salisbury steak with gravy, potatoes, vegetable, dessert (frozen meal, large meat portion)"
+28110370,"SALISBURY STEAK, GRAVY, MAC&CHEESE, VEG (FROZ MEAL)","Salisbury steak with gravy, macaroni and cheese, vegetable (frozen meal)"
+28110380,"SALISBURY STEAK W/GRAV,MACARONI & CHEESE (FRZ MEAL)","Salisbury steak with gravy, macaroni and cheese (frozen meal)"
+28110390,"SALISBURY STEAK, POT, VEG, DESSERT (DIET FZN MEAL)","Salisbury steak, potatoes, vegetable, dessert (diet frozen meal)"
+28110510,"BEEF, SLICED, W/ GRAVY, POTATOES, VEG (FROZEN MEAL)","Beef, sliced, with gravy, potatoes, vegetable (frozen meal)"
+28110620,"SHORTRIBS W/ BBQ SAUCE, POTATOES & VEG (FROZ MEAL)","Beef short ribs, boneless, with barbecue sauce, potatoes, vegetable (frozen meal)"
+28110640,"MEATBALLS, SWEDISH, IN SAUCE W/ NOODLES (FROZ MEAL)","Meatballs, Swedish, in sauce, with noodles (frozen meal)"
+28110660,"MEATBALLS,SWEDISH,W/GRAVY & NOODLES (DIET FRZ MEAL)","Meatballs, Swedish, in gravy, with noodles (diet frozen meal)"
+28113110,"SALISBURY STEAK W/ TOM SAUCE, VEG (DIET FROZ MEAL)","Salisbury steak, baked, with tomato sauce, vegetable (diet frozen meal)"
+28113140,"BEEF W/ SPAETZLE OR RICE, VEGETABLE (FROZEN MEAL)","Beef with spaetzle or rice, vegetable (frozen meal)"
+28130000,"VEAL DINNER, NFS (FROZEN)","Veal dinner, NFS (frozen meal)"
+28133110,"VEAL, BREADED, W/ SPAGHETTI, TOM SAUCE (FROZ MEAL)","Veal, breaded, with spaghetti, in tomato sauce (frozen meal)"
+28133340,"VEAL PARMIGIANA, VEG, FETTUCCINE,DESSERT(FROZ MEAL)","Veal parmigiana with vegetable, fettuccine alfredo, dessert (frozen meal)"
+28140100,"CHICKEN DINNER, NFS (FROZEN)","Chicken dinner, NFS (frozen meal)"
+28140150,"CHICKEN DIVAN (FROZEN MEAL)","Chicken divan (frozen meal)"
+28140250,"CHICKEN,GRAVY,DRESS,RICE,VEG,DESSRT(Z MEAL,LG MEAT)","Chicken, boneless, with gravy, dressing, rice, vegetable, dessert (frozen meal, large meat portion)"
+28140320,"CHICKEN & NOODLES W/ VEG, DESSERT (FROZEN MEAL)","Chicken and noodles with vegetable, dessert (frozen meal)"
+28140710,"CHICKEN, FRIED, W/ POTATOES, VEGETABLE (FROZ MEAL)","Chicken, fried, with potatoes, vegetable (frozen meal)"
+28140720,"CHICKEN PATTY, POTATOES, VEGETABLE (FROZEN MEAL)","Chicken patty, or nuggets, boneless, breaded, potatoes, vegetable (frozen meal)"
+28140730,"CHICKEN PATTY, TOM SCE, FETTUCCINE, VEG (FROZ MEAL)","Chicken patty, breaded, with tomato sauce and cheese, fettuccine alfredo, vegetable (frozen meal)"
+28140740,"CHICKEN PATTY/NUGGET,PASTA,FRUIT,DESSERT(FROZ MEAL)","Chicken patty, or nuggets, boneless, breaded, with pasta and tomato sauce, fruit, dessert (frozen meal)"
+28140810,"CHICKEN, FRIED, W/ POT, VEG, DESSERT (FROZEN MEAL)","Chicken, fried, with potatoes, vegetable, dessert (frozen meal)"
+28141010,"CHICKEN, FRIED, POT,VEG, DESSERT(FROZ MEAL,LG MEAT)","Chicken, fried, with potatoes, vegetable, dessert (frozen meal, large meat portion)"
+28141050,"CHICKEN PATTY PARMIGIANA, W/ VEG (DIET FROZ MEAL)","Chicken patty parmigiana, breaded, with vegetable (diet frozen meal)"
+28141201,"TERIYAKI CHICKEN W/ RICE & VEGETABLE (DIET FROZ MEAL)","Teriyaki chicken with rice and vegetable (diet frozen meal)"
+28141250,"CHICKEN W/ RICE-VEGETABLE MIXTURE (DIET FROZ MEAL)","Chicken with rice-vegetable mixture (diet frozen meal)"
+28141300,"CHICKEN W/RICE & VEG, REDUCED FAT&SODIUM(DIET FROZ)","Chicken with rice and vegetable, reduced fat and sodium (diet frozen meal)"
+28141600,"CHICKEN A LA KING W/ RICE (FROZEN MEAL)","Chicken a la king with rice (frozen meal)"
+28141610,"CHICKEN & VEGETABLES IN CREAM SCE (DIET FROZ MEAL)","Chicken and vegetables in cream or white sauce (diet frozen meal)"
+28141650,"CHICKEN & VEGETABLES AU GRATIN (DIET FROZEN MEAL)","Chicken and vegetables au gratin with rice-vegetable mixture (diet frozen entree)"
+28143020,"CHICKEN AND VEGETABLE W/ RICE, ASIAN (DIET FROZEN MEAL)","Chicken and vegetable entree with rice, Asian (diet frozen meal)"
+28143040,"CHICKEN CHOW MEIN W/ RICE (DIET FROZEN MEAL)","Chicken chow mein with rice (diet frozen meal)"
+28143050,"CHICK CHOWMEIN W/RICE,REDUCED FAT&SODIUM(DIET FROZ)","Chicken chow mein with rice, reduced fat and sodium (diet frozen meal)"
+28143080,"CHICKEN W/NOODLES & CHEESE SAUCE (DIET FROZEN MEAL)","Chicken with noodles and cheese sauce (diet frozen meal)"
+28143110,"CHICKEN CACCIATORE W/ NOODLES (DIET FROZEN MEAL)","Chicken cacciatore with noodles (diet frozen meal)"
+28143130,"CHICKEN & VEG ENTREE W/ NOODLES (FROZEN MEAL)","Chicken and vegetable entree with noodles (frozen meal)"
+28143150,"CHICK & VEG ENTREE W/ NOODLES, (DIET FROZEN MEAL)","Chicken and vegetable entree with noodles (diet frozen meal)"
+28143170,"CHICKEN IN CREAM SAUCE W/ NOODLES & VEG (FROZ MEAL)","Chicken in cream sauce with noodles and vegetable (frozen meal)"
+28143180,"CHICKEN,BUTTER SCE,W/POT & VEG (FRZ, DIET MEAL)","Chicken in butter sauce with potatoes and vegetable (diet frozen meal)"
+28143190,"CHICKEN, MUSHROOM SAUCE, WILD RICE, VEG (FROZ MEAL)","Chicken in mushroom sauce, white and wild rice, vegetable (frozen meal)"
+28143200,"CHICKEN IN SOY-BASED SAUCE,RICE&VEG (FROZEN MEAL)","Chicken in soy-based sauce, rice and vegetables (frozen meal)"
+28143210,"CHICKEN IN ORANGE SAUCE W/ RICE (DIET FROZEN MEAL)","Chicken in orange sauce with almond rice (diet frozen meal)"
+28143220,"CHICKEN IN BBQ SCE,W/RICE,VEG&DES,RED FAT&SODIUM,FRZ,DIET","Chicken in barbecue sauce, with rice, vegetable and dessert, reduced fat and sodium (diet frozen meal)"
+28144100,"CHICKEN & VEG W/ NOODLES & CREAM SCE (FROZEN MEAL)","Chicken and vegetable entree with noodles and cream sauce (frozen meal)"
+28145000,"TURKEY DINNER, NFS (FROZEN)","Turkey dinner, NFS (frozen meal)"
+28145100,"TURKEY W/DRESSING, GRAVY,VEG, FRUIT (DIET FRZ MEAL)","Turkey with dressing, gravy, vegetable and fruit (diet frozen meal)"
+28145110,"TURKEY W/ VEGETABLE, STUFFING (DIET FROZEN MEAL)","Turkey with vegetable, stuffing (diet frozen meal)"
+28145210,"TURKEY W/ GRAVY, DRESSING, POT, VEG (FROZEN MEAL)","Turkey with gravy, dressing, potatoes, vegetable (frozen meal)"
+28145610,"TURKEY, DRESSING,POT,VEG,DESSERT(FROZ MEAL,LG MEAT)","Turkey with gravy, dressing, potatoes, vegetable, dessert (frozen meal, large meat portion)"
+28145710,"TURKEY TETRAZZINI (FROZEN MEAL)","Turkey tetrazzini (frozen meal)"
+28150000,"FISH DINNER, NFS (FROZEN)","Fish dinner, NFS (frozen meal)"
+28150210,"HADDOCK W/ CHOPPED SPINACH (DIET FROZEN MEAL)","Haddock with chopped spinach (diet frozen meal)"
+28150220,"FLOUNDER W/ CHOPPED BROCCOLI (DIET FROZEN MEAL)","Flounder with chopped broccoli (diet frozen meal)"
+28150510,"FISH IN LEMON SAUCE W/ STARCH ITEM, VEG (FROZ MEAL)","Fish in lemon-butter sauce with starch item, vegetable (frozen meal)"
+28150650,"FISH,BREADED/FISH STICKS,W/PASTA,VEG,DES (FRZ MEAL)","Fish, breaded, or fish sticks, with pasta, vegetable and dessert (frozen meal)"
+28153010,"SHRIMP & CLAMS IN TOMATO SCE, W/ NOODLES(FROZ MEAL)","Shrimp and clams in tomato-based sauce, with noodles (frozen meal)"
+28154010,"SHRIMP & VEG IN SAUCE W/ NOODLES (DIET FROZEN MEAL)","Shrimp and vegetables in sauce with noodles (diet frozen meal)"
+28160300,"MEAT LOAF DINNER, NFS (FROZEN)","Meat loaf dinner, NFS (frozen meal)"
+28160310,"MEATLOAF W/ POTATO, VEG (FROZ MEAL)","Meat loaf with potatoes, vegetable (frozen meal)"
+28160650,"STUFFED GREEN PEPPER (FROZEN MEAL)","Stuffed green pepper (frozen meal)"
+28160710,"STUFFED CABBAGE, W/ MEAT & TOM SCE (DIET FROZ MEAL)","Stuffed cabbage, with meat and tomato sauce (diet frozen meal)"
+28310110,"BEEF BROTH, BOUILLON OR CONSOMME (INCL BROTH, NFS)","Beef, broth, bouillon, or consomme"
+28310120,"BEEF BROTH OR BOUILLON, CANNED, LOW SODIUM","Beef, broth, bouillon, or consomme, canned, low sodium"
+28310150,"OXTAIL SOUP","Oxtail soup"
+28310160,"BEEF BROTH, W/ TOMATO, HOME RECIPE","Beef broth, with tomato, home recipe"
+28310170,"BEEF BROTH, W/O TOMATO, HOME RECIPE","Beef broth, without tomato, home recipe"
+28310230,"MEATBALL SOUP, MEXICAN STYLE, HOME RECIPE (SOPA DE ALBONDIGA","Meatball soup, Mexican style, home recipe (Sopa de Albondigas)"
+28310320,"BEEF NOODLE SOUP, P.R. (SOPA DE CARNE Y FIDEOS)","Beef noodle soup, Puerto Rican style (Sopa de carne y fideos)"
+28310330,"MEAT AND RICE NOODLE SOUP, ASIAN STYLE (VIETNAMESE PHO BO)","Meat and rice noodle soup, Asian style (Vietnamese Pho Bo)"
+28310420,"BEEF & RICE SOUP, P.R.","Beef and rice soup, Puerto Rican style"
+28311010,"PEPPERPOT (TRIPE) SOUP","Pepperpot (tripe) soup"
+28311020,"MENUDO SOUP, HOME RECIPE","Menudo soup, home recipe"
+28311030,"MENUDO, CANNED","Menudo soup, canned, prepared with water or ready-to-serve"
+28315050,"BEEF VEGETABLE SOUP W/ POTATO, PASTA OR RICE, CHUNKY STYLE","Beef vegetable soup with potato, pasta, or rice, chunky style, canned, or ready-to-serve"
+28315140,"BEEF VEGETABLE SOUP, MEXICAN STYLE, HOME RECIPE, (SOPA / CAL","Beef vegetable soup, Mexican style, home recipe, (Sopa / Caldo de Res)"
+28315150,"MEAT AND CORN HOMINY SOUP, MEXICAN STYLE, HOME RECIPE (POZOL","Meat and corn hominy soup, Mexican style, home recipe (Pozole)"
+28315160,"ITALIAN WEDDING SOUP","Italian Wedding Soup"
+28317010,"BEEF STROGANOFF SOUP, CHUNKY STYLE, HOME RECIPE, CANNED OR R","Beef stroganoff soup, chunky style, home recipe, canned or ready-to-serve"
+28320130,"HAM, RICE, & POTATO SOUP, P.R.","Ham, rice, and potato soup, Puerto Rican style"
+28320140,"HAM, NOODLE & VEGETABLE SOUP, P.R.","Ham, noodle, and vegetable soup, Puerto Rican style"
+28320160,"PORK VEGETABLE SOUP W/ POTATO, PASTA, OR RICE, CHUNKY","Pork vegetable soup with potato, pasta, or rice, stew type, chunky style"
+28320300,"PORK W/VEG (NO CAR, BROC  AND/OR DK GRN)SOUP, ASIAN STYLE","Pork with vegetable (excluding carrots, broccoli and/or dark-green leafy) soup, Asian Style"
+28321130,"BACON SOUP, CREAM OF, PREPARED W/ WATER","Bacon soup, cream of, prepared with water"
+28330110,"SCOTCH BROTH (LAMB, VEGETABLES, BARLEY)","Scotch broth (lamb, vegetables, and barley)"
+28331110,"LAMB, PASTA & VEGETABLE SOUP, P.R.","Lamb, pasta, and vegetable soup, Puerto Rican style"
+28340110,"CHICKEN OR TURKEY BROTH, BOUILLON, OR CONSOMME","Chicken or turkey broth, bouillon, or consomme"
+28340120,"CHICKEN OR TURKEY BROTH, WITHOUT TOMATO, HOME RECIPE","Chicken or turkey broth, without tomato, home recipe"
+28340130,"CHICKEN OR TURKEY BROTH, WITH TOMATO, HOME RECIPE","Chicken or turkey broth, with tomato, home recipe"
+28340150,"MEXICAN STYLE CHICKEN BROTH SOUP STOCK","Mexican style chicken broth soup stock"
+28340180,"CHICKEN OR TURKEY BROTH, LESS/REDUCED SODIUM, CANNED OR RTS","Chicken or turkey broth, less or reduced sodium, canned or ready-to-serve"
+28340210,"CHICKEN RICE SOUP, P.R. (SOPA DE POLLO CON ARROZ)","Chicken rice soup, Puerto Rican style (Sopa de pollo con arroz)"
+28340220,"CHICKEN SOUP W/ NOODLES & POTATOES, P.R.","Chicken soup with noodles and potatoes, Puerto Rican style"
+28340310,"CHICKEN OR TURKEY GUMBO SOUP","Chicken or turkey gumbo soup, home recipe, canned or ready-to-serve"
+28340510,"CHICKEN OR TURKEY NOODLE SOUP, CHUNKY STYLE, CANNED OR RTS","Chicken or turkey noodle soup, chunky style, canned or ready-to-serve"
+28340550,"SWEET & SOUR SOUP","Sweet and sour soup"
+28340580,"CHICKEN OR TURKEY SOUP WITH VEGETABLES, ASIAN STYLE","Chicken or turkey soup with vegetables (broccoli, carrots, celery, potatoes and onions), Asian style"
+28340590,"CHICKEN OR TURKEY CORN SOUP WITH NOODLES, HOME RECIPE","Chicken or turkey corn soup with noodles, home recipe"
+28340600,"CHICK/TURK+VEG SOUP, CANNED","Chicken or turkey vegetable soup, canned, prepared with water or ready-to-serve"
+28340610,"CHICKEN VEGETABLE SOUP, STEW TYPE (INCL CHUNKY)","Chicken or turkey vegetable soup, stew type"
+28340630,"CHICKEN OR TURKEY VEGETABLE SOUP WITH RICE, STEW TYPE, CHUNK","Chicken or turkey vegetable soup with rice, stew type, chunky style"
+28340640,"CHICKEN OR TURKEY VEGETABLE SOUP WITH NOODLES, STEW TYPE, CH","Chicken or turkey vegetable soup with noodles, stew type, chunky style, canned or ready-to-serve"
+28340660,"CHICKEN OR TURKEY  VEGETABLE SOUP, HOME RECIPE","Chicken or turkey vegetable soup, home recipe"
+28340670,"CHICKEN OR TURKEY VEGETABLE SOUP WITH RICE, MEXICAN STYLE, H","Chicken or turkey vegetable soup with rice, Mexican style, home recipe (Sopa / Caldo de Pollo)"
+28340680,"CHICKEN OR TURKEY AND CORN HOMINY SOUP, MEXICAN STYLE, HOME","Chicken or turkey and corn hominy soup, Mexican style, home recipe (Pozole)"
+28340690,"CHICKEN OR TURKEY VEGETABLE SOUP WITH POTATO AND CHEESE, CHU","Chicken or turkey vegetable soup with potato and cheese, chunky style, canned or ready-to-serve"
+28340700,"BIRD'S NEST SOUP (CHICKEN, HAM, NOODLES)","Bird's nest soup (chicken, ham, and noodles)"
+28340750,"HOT & SOUR SOUP (INCLUDE HOT & SPICY CHINESE SOUP)","Hot and sour soup"
+28340800,"CHICKEN OR TURKEY SOUP WITH VEGETABLES AND FRUIT, ASIAN STYL","Chicken or turkey soup with vegetables and fruit, Asian Style"
+28345010,"CHICKEN/TURKEY SOUP, CM OF, CAN, RED SOD, NS W/ MILK/WATER","Chicken or turkey soup, cream of, canned, reduced sodium, NS as to made with milk or water"
+28345020,"CHICKEN/TURKEY SOUP, CM OF, CAN, RED SOD, W/ MILK","Chicken or turkey soup, cream of, canned, reduced sodium, made with milk"
+28345030,"CHICKEN/TURKEY SOUP, CM OF, CAN, RED SOD, W/ WATER","Chicken or turkey soup, cream of, canned, reduced sodium, made with water"
+28345110,"CHICKEN SOUP, CREAM OF, NS AS TO MILK OR WATER","Chicken or turkey soup, cream of, NS as to prepared with milk or water"
+28345120,"CHICKEN/TURKEY SOUP,CREAM OF, W/ MILK","Chicken or turkey soup, cream of, prepared with milk"
+28345130,"CHICKEN SOUP, CREAM OF, PREPARED W/ WATER","Chicken or turkey soup, cream of, prepared with water"
+28345160,"CHICKEN OR TURKEY MUSHROOM SOUP, CREAM OF, PREPARED WITH MIL","Chicken or turkey mushroom soup, cream of, prepared with milk"
+28345170,"DUCK SOUP","Duck soup"
+28350040,"FISH STOCK, HOME RECIPE","Fish stock, home recipe"
+28350050,"FISH CHOWDER (INCL FISHERMAN'S SOUP, SEAFOOD CHOWD)","Fish chowder"
+28350110,"CRAB SOUP, NS AS TO TOMATO-BASE OR CREAM","Crab soup, NS as to tomato-base or cream style"
+28350120,"CRAB SOUP, TOMATO BASE","Crab soup, tomato-base"
+28350210,"CLAM CHOWDER, NS AS TO MANHATTAN OR NEW ENGLAND","Clam chowder, NS as to Manhattan or New England style"
+28350220,"CLAM CHOWDER, MANHATTAN (INCLUDE CHUNKY)","Clam chowder, Manhattan"
+28350310,"TURTLE & VEGETABLE SOUP (INCLUDE SNAPPER SOUP)","Turtle and vegetable soup"
+28351110,"FISH AND VEGETABLE SOUP, NO POTATOES (SOPA DE PESCADO Y MARI","Fish and vegetable soup, no potatoes (Sopa de pescado y mariscos)"
+28351120,"FISH SOUP, WITH POTATOES (SOPA DE PESCADO Y MARISCOS)","Fish soup, with potatoes (Sopa de Pescado y Mariscos)"
+28351160,"CODFISH, RICE & VEGETABLE SOUP, P.R.","Codfish, rice, and vegetable soup, Puerto Rican style"
+28351170,"CODFISH SOUP W/ NOODLES, P.R.","Codfish soup with noodles, Puerto Rican style"
+28355110,"CLAM CHOWDER, NEW ENG, NS AS TO MILK OR WATER ADDED","Clam chowder, New England, NS as to prepared with water or milk"
+28355120,"CLAM CHOWDER, NEW ENGLAND, W/ MILK","Clam chowder, New England, prepared with milk"
+28355130,"CLAM CHOWDER, NEW ENGLAND, W/ WATER","Clam chowder, New England, prepared with water"
+28355140,"CLAM CHOWDER, NEW ENGLAND, REDUCED SODIUM, CANNED OR READY-T","Clam chowder, New England, reduced sodium, canned or ready-to-serve"
+28355210,"CRAB SOUP, CREAM OF, W/ MILK","Crab soup, cream of, prepared with milk"
+28355250,"LOBSTER BISQUE","Lobster bisque"
+28355260,"LOBSTER GUMBO","Lobster gumbo"
+28355310,"OYSTER STEW","Oyster stew"
+28355350,"SALMON SOUP, CREAM STYLE","Salmon soup, cream style"
+28355410,"SHRIMP SOUP, CREAM OF, NS AS TO MILK/WATER ADDED","Shrimp soup, cream of, NS as to prepared with milk or water"
+28355420,"SHRIMP SOUP, CREAM OF, W/ MILK","Shrimp soup, cream of, prepared with milk"
+28355430,"SHRIMP SOUP, CREAM OF, W/ WATER","Shrimp soup, cream of, prepared with water"
+28355440,"SHRIMP GUMBO","Shrimp gumbo"
+28355450,"SEAFOOD SOUP W/ POTATOES & VEGETABLES (INCL DK GREEN LEAF)","Seafood soup with potatoes and vegetables (including carrots, broccoli, and/or dark-green leafy)"
+28355460,"SEAFOOD SOUP W/ POTATOES & VEGETABLES (EXCL DK GREEN LEAF)","Seafood soup with potatoes and vegetables (excluding carrots, broccoli, and dark-green leafy)"
+28355470,"SEAFOOD SOUP W/ VEGETABLES (INCL DK GREEN LEAFY)","Seafood soup with vegetables (including carrots, broccoli, and/or dark-green leafy (no potatoes))"
+28355480,"SEAFOOD SOUP W/ VEGETABLES (EXCL DK GREEN LEAFY)","Seafood soup with vegetables (excluding carrots, broccoli, and dark-green leafy (no potatoes))"
+28360100,"MEAT BROTH, P.R. STYLE","Meat broth, Puerto Rican style"
+28360210,"SPANISH VEGETABLE SOUP, P.R. (CALDO GALLEGO)","Spanish vegetable soup, Puerto Rican style (Caldo gallego)"
+28500000,"GRAVY, POULTRY","Gravy, poultry"
+28500010,"GRAVY, MEAT/POULTRY, W/ WINE","Gravy, meat or poultry, with wine"
+28500020,"GRAVY, MEAT, W/ FRUIT (INCLUDE FRENCH SAUCE)","Gravy, meat, with fruit"
+28500030,"GRAVY, POULTRY, LOW SODIUM","Gravy, poultry, low sodium"
+28500040,"GRAVY, BEEF/MEAT (INCL GRAVY,NFS;BROWN GRAVY;SWISS STEAK GRV","Gravy, beef or meat"
+28500050,"GRAVY, GIBLET(INCL ANY POULTRY GRAVY W/PCS OF MEAT)","Gravy, giblet"
+28500060,"GRAVY, BEEF OR MEAT, LOW SODIUM","Gravy, beef or meat, low sodium"
+28500070,"GRAVY, BEEF OR MEAT, HOME RECIPE","Gravy, beef or meat, home recipe"
+28500080,"GRAVY, POULTRY, HOME RECIPE","Gravy, poultry, home recipe"
+28500100,"GRAVY, MUSHROOM","Gravy, mushroom"
+28500150,"GRAVY, REDEYE","Gravy, redeye"
+28501010,"GRAVY, BEEF/MEAT, FAT FREE","Gravy, beef or meat, fat free"
+28501110,"GRAVY, POULTRY, FAT FREE","Gravy, poultry, fat free"
+28510010,"GRAVY/SAUCE, POULTRY FROM CHICKEN FRICASSEE, P.R.","Gravy or sauce, poultry-based from Puerto Rican-style chicken fricasse"
+28510020,"GRAVY, MEAT-BASED, FROM PUERTO RICAN POT ROAST","Gravy, meat-based, from Puerto-Rican style stuffed pot roast"
+28510030,"GRAVY, MEAT-BASED, FROM PUERTO RICAN BEEF STEW","Gravy, meat-based, from Puerto-Rican style beef stew"
+28520000,"GRAVY/SAUCE,CHINESE(SOY SCE,STOCK/BOUILL,CRNSTRCH)","Gravy or sauce, Chinese (soy sauce, stock or bouillon, cornstarch)"
+28520100,"OYSTER-FLAVORED SAUCE","Oyster-flavored sauce"
+28522000,"MOLE POBLANA (SAUCE)","Mole poblano (sauce)"
+28522050,"MOLE VERDE (SAUCE)","Mole verde (sauce)"
+31101010,"EGGS, WHOLE, RAW","Egg, whole, raw"
+31102000,"EGGS, WHOLE, COOKED, NS AS TO METHOD","Egg, whole, cooked, NS as to cooking method"
+31103010,"EGG, WHOLE, BOILED OR POACHED","Egg, whole, boiled or poached"
+31105010,"EGG, WHOLE, FRIED WITHOUT FAT","Egg, whole, fried without fat"
+31105020,"EGG, WHOLE, FRIED WITH MARGARINE","Egg, whole, fried with margarine"
+31105030,"EGG, WHOLE, FRIED WITH OIL","Egg, whole, fried with oil"
+31105040,"EGG, WHOLE, FRIED WITH BUTTER","Egg, whole, fried with butter"
+31105060,"EGG, WHOLE, FRIED WITH ANIMAL FAT OR MEAT DRIPPINGS","Egg, whole, fried with animal fat or meat drippings"
+31105080,"EGG, WHOLE, FRIED WITH COOKING SPRAY","Egg, whole, fried with cooking spray"
+31105090,"EGG, WHOLE, FRIED, FROM FAST FOOD / RESTAURANT","Egg, whole, fried, from fast food / restaurant"
+31106000,"EGGS, WHOLE, BAKED, NS AS TO ADDED FAT","Egg, whole, baked, NS as to fat added in cooking"
+31106010,"EGGS, WHOLE, BAKED, NO FAT ADDED","Egg, whole, baked, fat not added in cooking"
+31106020,"EGGS, WHOLE, BAKED, FAT ADDED","Egg, whole, baked, fat added in cooking"
+31107000,"EGGS, WHOLE, PICKLED","Egg, whole, pickled"
+31108010,"EGGS, WHITE ONLY, RAW","Egg, white only, raw"
+31110010,"EGG YOLK, ONLY, RAW","Egg, yolk only, raw"
+31111000,"EGG, YOLK ONLY, COOKED, NS AS TO FAT ADDED IN COOKING","Egg, yolk only, cooked, NS as to fat added in cooking"
+31111010,"EGG, YOLK ONLY, COOKED, NO FAT ADDED","Egg, yolk only, cooked, fat not added in cooking"
+31111020,"EGG, YOLK ONLY, COOKED, FAT ADDED IN COOKING","Egg, yolk only, cooked, fat added in cooking"
+31201000,"DUCK EGG, COOKED","Duck egg, cooked"
+31202000,"GOOSE EGG, COOKED","Goose egg, cooked"
+31203000,"QUAIL EGG, CANNED","Quail egg, canned"
+32101000,"EGGS, CREAMED","Egg, creamed"
+32101500,"EGGS, BENEDICT","Egg, Benedict"
+32101530,"EGG CURRY","Egg curry"
+32102000,"EGGS, DEVILED","Egg, deviled"
+32103000,"EGG SALAD, W/ MAYO","Egg salad, made with mayonnaise"
+32103015,"EGG SALAD, W/ LT MAYO","Egg salad, made with light mayonnaise"
+32103020,"EGG SALAD, W/ MAYO-TYPE DRSG","Egg salad, made with mayonnaise-type salad dressing"
+32103025,"EGG SALAD, W/ LIGHT MAYO-TYPE DRSG","Egg salad, made with light mayonnaise-type salad dressing"
+32103030,"EGG SALAD, W/ CREAMY DRSG","Egg salad, made with creamy dressing"
+32103035,"EGG SALAD,W/ LT CREAMY DRSG","Egg salad, made with light creamy dressing"
+32103040,"EGG SALAD, W/ ITALIAN DRSG","Egg salad, made with Italian dressing"
+32103045,"EGG SALAD, W/ LT ITALIAN DRSG","Egg salad, made with light Italian dressing"
+32103050,"EGG SALAD, W/ ANY TYPE OF FAT FREE DRSG","Egg Salad, made with any type of fat free dressing"
+32105180,"HUEVOS RANCHEROS","Huevos rancheros"
+32105190,"EGG CASSEROLE W/ BREAD, CHEESE, MILK & MEAT","Egg casserole with bread, cheese, milk and meat"
+32105200,"EGG FOO YUNG, NFS","Egg foo yung (young), NFS"
+32105210,"CHICKEN EGG FOO YUNG","Chicken egg foo yung (young)"
+32105220,"PORK EGG FOO YUNG","Pork egg foo yung (young)"
+32105230,"SHRIMP EGG FOO YUNG","Shrimp egg foo yung (young)"
+32105240,"BEEF EGG FOO YUNG","Beef egg foo yung (young)"
+32105310,"RIPE PLANTAIN OMELET, P.R. (TORTILLA DE AMARILLO)","Ripe plantain omelet, Puerto Rican style (Tortilla de amarillo)"
+32105330,"SCRAMBLED EGGS W/ JERKED BEEF, P.R.","Scrambled eggs with jerked beef, Puerto Rican style (Revoltillo de tasajo)"
+32110100,"EGGS, A LA MALAGUENA, P.R.(HUEVOS A LA MALAGUENA)","Eggs a la Malaguena, Puerto Rican style (Huevos a la Malaguena)"
+32110150,"SHRIMP-EGG PATTY (TORTA DE CAMERON SECO)","Shrimp-egg patty (Torta de Cameron seco)"
+32120100,"EGG DESSERT, CUSTARD-LIKE, W/ WATER & SUGAR, P.R.","Egg dessert, custard-like, made with water and sugar, Puerto Rican style (Tocino del cielo; Heaven's delight)"
+32120200,"ZABAGLIONE","Zabaglione"
+32130000,"EGG OMELET OR SCRAMBLED EGG, MADE WITH MARGARINE","Egg omelet or scrambled egg, made with margarine"
+32130010,"EGG OMELET OR SCRAMBLED EGG, MADE WITH OIL","Egg omelet or scrambled egg, made with oil"
+32130020,"EGG OMELET OR SCRAMBLED EGG, MADE WITH BUTTER","Egg omelet or scrambled egg, made with butter"
+32130040,"EGG OMELET OR SCRAMBLED EGG, MADE W/ANIMAL FAT OR MEAT DRIP","Egg omelet or scrambled egg, made with animal fat or meat drippings"
+32130060,"EGG OMELET OR SCRAMBLED EGG, MADE WITH COOKING SPRAY","Egg omelet or scrambled egg, made with cooking spray"
+32130070,"EGG OMELET OR SCRAMBLED EGG, MADE WITHOUT FAT","Egg omelet or scrambled egg, made without fat"
+32130080,"EGG OMELET OR SCRAMBLED EGG, FROM FAST FOOD / RESTAURANT","Egg omelet or scrambled egg, from fast food / restaurant"
+32130100,"EGG OMELET OR SCRAMBLED EGG, WITH CHEESE, MADE WITH MARGARIN","Egg omelet or scrambled egg, with cheese, made with margarine"
+32130110,"EGG OMELET OR SCRAMBLED EGG, WITH CHEESE, MADE WITH OIL","Egg omelet or scrambled egg, with cheese, made with oil"
+32130120,"EGG OMELET OR SCRAMBLED EGG, WITH CHEESE, MADE WITH BUTTER","Egg omelet or scrambled egg, with cheese, made with butter"
+32130140,"EGG OMELET OR SCRAMBLED EGG, W/ CHEESE, MADE W/ ANIMAL FAT","Egg omelet or scrambled egg, with cheese, made with animal fat or meat drippings"
+32130160,"EGG OMELET OR SCRAMBLED EGG, WITH CHEESE, MADE WITH COOKING","Egg omelet or scrambled egg, with cheese, made with cooking spray"
+32130170,"EGG OMELET OR SCRAMBLED EGG, WITH CHEESE, MADE WITHOUT FAT","Egg omelet or scrambled egg, with cheese, made without fat"
+32130200,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE WITH MARGARINE","Egg omelet or scrambled egg, with meat, made with margarine"
+32130210,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE WITH OIL","Egg omelet or scrambled egg, with meat, made with oil"
+32130220,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE WITH BUTTER","Egg omelet or scrambled egg, with meat, made with butter"
+32130240,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE WITH ANIMAL FAT","Egg omelet or scrambled egg, with meat, made with animal fat or meat drippings"
+32130260,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE W/COOKING SPRAY","Egg omelet or scrambled egg, with meat, made with cooking spray"
+32130270,"EGG OMELET OR SCRAMBLED EGG, WITH MEAT, MADE WITHOUT FAT","Egg omelet or scrambled egg, with meat, made without fat"
+32130300,"EGG OMELET OR SCRAMBLED EGG, W/CHEESE & MEAT, MADE W/MARGARI","Egg omelet or scrambled egg, with cheese and meat, made with margarine"
+32130310,"EGG OMELET OR SCRAMBLED EGG, W/CHEESE & MEAT, MADE W/OIL","Egg omelet or scrambled egg, with cheese and meat, made with oil"
+32130320,"EGG OMELET OR SCRAMBLED EGG, W/ CHEESE &MEAT, MADEW/BUTTER","Egg omelet or scrambled egg, with cheese and meat, made with butter"
+32130340,"EGG OMELET OR SCR EGG, WITH CHEESE & MEAT, MADE W/ANIMAL FAT","Egg omelet or scrambled egg, with cheese and meat, made with animal fat or meat drippings"
+32130360,"EGG OMELET OR SCR EGG, W/CHEESE& MEAT, MADE W/COOKING SPRAY","Egg omelet or scrambled egg, with cheese and meat, made with cooking spray"
+32130370,"EGG OMELET OR SCRAMBLED EGG, W/CHEESE & MEAT, MADE WO/FAT","Egg omelet or scrambled egg, with cheese and meat, made without fat"
+32130400,"EGG OMELET OR SCRAMBLED EGG, WITH TOMATOES, FAT ADDED IN COO","Egg omelet or scrambled egg, with tomatoes, fat added in cooking"
+32130410,"EGG OMELET OR SCRAMBLED EGG, WITH TOMATOES, FAT NOT ADDED IN","Egg omelet or scrambled egg, with tomatoes, fat not added in cooking"
+32130420,"EGG OMELET OR SCRAMBLED EGG, WITH TOMATOES, NS AS TO FAT ADD","Egg omelet or scrambled egg, with tomatoes, NS as to fat added in cooking"
+32130430,"EGG OMELET OR SCRAMBLED EGG, W/ DARK-GREEN VEGS, FAT ADDED","Egg omelet or scrambled egg, with dark-green vegetables, fat added in cooking"
+32130440,"EGG OMELET OR SCRAMBLED EGG, W/DARK-GREEN VEGS, FAT NOT ADDE","Egg omelet or scrambled egg, with dark-green vegetables, fat not added in cooking"
+32130450,"EGG OMELET OR SCRAMBLED EGG, W/ DARK-GREEN VEGS,NS AS TO FAT","Egg omelet or scrambled egg, with dark-green vegetables, NS as to fat added in cooking"
+32130460,"EGG OMELET OR SCR EGG, W/TOMATOES & DK-GREEN VEGS, FAT ADDED","Egg omelet or scrambled egg, with tomatoes and dark-green vegetables, fat added in cooking"
+32130470,"EGG OMELET OR SCR EGG, W/ TOMATOES & DK-GRN VEGS,FAT NOT ADD","Egg omelet or scrambled egg, with tomatoes and dark-green vegetables, fat not added in cooking"
+32130480,"EGG OMELET OR SCR EGG, W/ TOMATOES & DK-GRN VEGS, NS FAT","Egg omelet or scrambled egg, with tomatoes and dark-green vegetables, NS as to fat added in cooking"
+32130490,"EGG OMELET OR SCR EGG, W/ OTHER VEGS, FAT ADDED","Egg omelet or scrambled egg, with vegetables other than dark green and/or tomatoes, fat added in cooking"
+32130500,"EGG OMELET OR SCRAMBLED EGG, WITH OTHER VEGS. FAT NOT ADDED","Egg omelet or scrambled egg, with vegetables other than dark green and/or tomatoes, fat not added in cooking"
+32130510,"EGG OMELET OR SCRAMBLED EGG, W/ OTHER VEGS, NS FAT","Egg omelet or scrambled egg, with vegetables other than dark green and/or tomatoes, NS as to fat added in cooking"
+32130600,"EGG OMELET OR SCR EGG, W/CHEESE & TOMATOES, FAT ADDED","Egg omelet or scrambled egg, with cheese and tomatoes, fat added in cooking"
+32130610,"EGG OMELET OR SCR EGG, W/CHEESE & TOMATOES, FAT NOT ADDED","Egg omelet or scrambled egg, with cheese and tomatoes, fat not added in cooking"
+32130620,"EGG OMELET OR SCR EGG, W/CHEESE & TOMATOES, NS FAT ADDED","Egg omelet or scrambled egg, with cheese and tomatoes, NS as to fat added in cooking"
+32130630,"EGG OMELET OR SCR EGG, W/ CHEESE&DK-GRN VEGS, FAT ADDED","Egg omelet or scrambled egg, with cheese and dark-green vegetables, fat added in cooking"
+32130640,"EGG OMELET OR SCR EGG, W/ CHEESE&DK-GRN VEGS, FAT NOT ADDED","Egg omelet or scrambled egg, with cheese and dark-green vegetables, fat not added in cooking"
+32130650,"EGG OMELET OR SCR EGG, W/ CHEESE&DK-GRN VEGS, NS FAT ADDED","Egg omelet or scrambled egg, with cheese and dark-green vegetables, NS as to fat added in cooking"
+32130660,"EGG OMELET OR SCR EGG, W/CHEESE, TOM & DK-GRN VEGS,FAT ADDED","Egg omelet or scrambled egg, with cheese, tomatoes, and dark-green vegetables, fat added in cooking"
+32130670,"EGG OMELET OR SCR EGG, W/CHEESE, TOM & DK-GRN VEGS,FAT NOT A","Egg omelet or scrambled egg, with cheese, tomatoes, and dark-green vegetables, fat not added in cooking"
+32130680,"EGG OMELET OR SCR EGG, W/CHEESE, TOM & DK-GRN VEGS,NS FAT AD","Egg omelet or scrambled egg, with cheese, tomatoes, and dark-green vegetables, NS as to fat added in cooking"
+32130690,"EGG OMELET OR SCR EGG, W/CHEESE & OTHER VEGS, FAT ADDED","Egg omelet or scrambled egg, with cheese and vegetables other than dark green and/or tomatoes, fat added in cooking"
+32130700,"EGG OMELET OR SCREGG, W/CHEESE & OTHER VEGS, FAT NOT ADDED","Egg omelet or scrambled egg, with cheese and vegetables other than dark green and/or tomatoes, fat not added in cooking"
+32130710,"EGG OMELET OR SCREGG, W/CHEESE & OTHER VEGS, NS FAT ADDED","Egg omelet or scrambled egg, with cheese and vegetables other than dark green and/or tomatoes, NS as to fat added in cooking"
+32130800,"EGG OMELET OR SCRAMBLED EGG, W/MEAT & TOMATOES, FAT ADDED","Egg omelet or scrambled egg, with meat and tomatoes, fat added in cooking"
+32130810,"EGG OMELET OR SCRAMBLED EGG, W/MEAT & TOMATOES, FAT NOT ADDE","Egg omelet or scrambled egg, with meat and tomatoes, fat not added in cooking"
+32130820,"EGG OMELET OR SCRAMBLED EGG, W/MEAT & TOMATOES, NS FAT ADDED","Egg omelet or scrambled egg, with meat and tomatoes, NS as to fat added in cooking"
+32130830,"EGG OMELET OR SCR EGG, W/MEAT & DK-GRN VEGS, FAT ADDED","Egg omelet or scrambled egg, with meat and dark-green vegetables, fat added in cooking"
+32130840,"EGG OMELET OR SCR EGG, W/MEAT & DK-GRN VEGS, FAT NOT ADDED","Egg omelet or scrambled egg, with meat and dark-green vegetables, fat not added in cooking"
+32130850,"EGG OMELET OR SCR EGG, W/MEAT & DK-GRN VEGS, NS FAT ADDED","Egg omelet or scrambled egg, with meat and dark-green vegetables, NS as to fat added in cooking"
+32130860,"EGG OMELET OR SCR EGG, W/MEAT,TOM & DK-GRN VEGS, FAT ADDED","Egg omelet or scrambled egg, with meat, tomatoes, and dark-green vegetables, fat added in cooking"
+32130870,"EGG OMELET OR SCR EGG, W/MEAT,TOM & DK-GRN VEGS, FAT NOT ADD","Egg omelet or scrambled egg, with meat, tomatoes, and dark-green vegetables, fat not added in cooking"
+32130880,"EGG OMELET OR SCR EGG, W/MEAT,TOM & DK-GRN VEGS, NS FAT ADDE","Egg omelet or scrambled egg, with meat, tomatoes, and dark-green vegetables, NS as to fat added in cooking"
+32130890,"EGG OMELET OR SCR EGG, W/MEAT & OTHER VEGS, FAT ADDED","Egg omelet or scrambled egg, with meat and vegetables other than dark-green and/or tomatoes, fat added in cooking"
+32130900,"EGG OMELET OR SCR EGG, W/MEAT & OTHER VEGS, FAT NOT ADDED","Egg omelet or scrambled egg, with meat and vegetables other than dark-green and/or tomatoes, fat not added in cooking"
+32130910,"EGG OMELET OR SCR EGG, W/MEAT & OTHER VEGS, NS FAT ADDED","Egg omelet or scrambled egg, with meat and vegetables other than dark-green and/or tomatoes, NS as to fat added in cooking"
+32131000,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & TOMATOES, FAT ADDED","Egg omelet or scrambled egg, with cheese, meat, and tomatoes, fat added in cooking"
+32131010,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & TOMATOES, FAT NOT A","Egg omelet or scrambled egg, with cheese, meat, and tomatoes, fat not added in cooking"
+32131020,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & TOMATOES, NS FAT AD","Egg omelet or scrambled egg, with cheese, meat, and tomatoes, NS as to fat added in cooking"
+32131030,"EGG OMELET OR SCR EGG, W/CHEESE, MEAT&DK GRN VEG,FAT ADDED","Egg omelet or scrambled egg, with cheese, meat, and dark-green vegetables, fat added in cooking"
+32131040,"EGG OMELET OR SCR EGG, W/CHEESE, MEAT&DK GRN VEG,FAT NOT ADD","Egg omelet or scrambled egg, with cheese, meat, and dark-green vegetables, fat not added in cooking"
+32131050,"EGG OMELET OR SCR EGG, W/CHEESE, MEAT&DK GRN VEG,NS FAT ADDE","Egg omelet or scrambled egg, with cheese, meat, and dark-green vegetables, NS as to fat added in cooking"
+32131060,"EGG OMELET/SCR EGG,W/CHEESE,MEAT,TOM&DK GRN VEG,FAT ADDED","Egg omelet or scrambled egg, with cheese, meat, tomatoes, and dark-green vegetables, fat added in cooking"
+32131070,"EGG OMELET/SCR EGG,W/CHEESE,MEAT,TOM&DK GRN VEG,FAT NOT ADDE","Egg omelet or scrambled egg, with cheese, meat, tomatoes, and dark-green vegetables, fat not added in cooking"
+32131080,"EGG OMELET/SCR EGG,W/CHEESE,MEAT,TOM&DK GRN VEG,NS FAT ADDED","Egg omelet or scrambled egg, with cheese, meat, tomatoes, and dark-green vegetables, NS as to fat added in cooking"
+32131090,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & OTHER VEG,FAT ADDED","Egg omelet or scrambled egg, with cheese, meat, and vegetables other than dark-green and/or tomatoes, fat added in cooking"
+32131100,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & OTHER VEG,FAT NOT A","Egg omelet or scrambled egg, with cheese, meat, and vegetables other than dark-green and/or tomatoes, fat not added in cooking"
+32131110,"EGG OMELET OR SCR EGG, W/ CHEESE, MEAT & OTHER VEG,NS FAT AD","Egg omelet or scrambled egg, with cheese, meat, and vegetables other than dark-green and/or tomatoes, NS as to fat added in cooking"
+32131200,"EGG OMELET OR SCR EGG, W/POTATOES +/OR ONIONS, FAT ADDED","Egg omelet or scrambled egg, with potatoes and/or onions, fat added in cooking"
+32131210,"EGG OMELET OR SCR EGG, W/POTATOES +/OR ONIONS, FAT NOT ADDED","Egg omelet or scrambled egg, with potatoes and/or onions, fat not added in cooking"
+32131220,"EGG OMELET OR SCR EGG, W/POTATOES +/OR ONIONS, NS FAT ADDED","Egg omelet or scrambled egg, with potatoes and/or onions, NS as to fat added in cooking"
+32201000,"FRIED EGG SANDWICH","Fried egg sandwich"
+32202000,"EGG, CHEESE, HAM, & BACON ON BUN","Egg, cheese, ham, and bacon on bun"
+32202010,"EGG, CHEESE & HAM ON ENGLISH MUFFIN","Egg, cheese, and ham on English muffin"
+32202020,"EGG, CHEESE & HAM ON BISCUIT","Egg, cheese, and ham on biscuit"
+32202025,"EGG, CHEESE & HAM ON BAGEL","Egg, cheese and ham on bagel"
+32202030,"EGG, CHEESE & SAUSAGE ON ENGLISH MUFFIN","Egg, cheese, and sausage on English muffin"
+32202035,"EGG, EXTRA CHEESE (2 SL), & EXTRA SAUSAGE (2 PATTIES) ON BUN","Egg, extra cheese (2 slices), and extra sausage (2 patties) on bun"
+32202040,"EGG, CHEESE & BEEF ON ENGLISH MUFFIN","Egg, cheese, and beef on English Muffin"
+32202045,"EGG, CHEESE & STEAK ON BAGEL","Egg, cheese, and steak on bagel"
+32202050,"EGG, CHEESE & SAUSAGE ON BISCUIT","Egg, cheese, and sausage on biscuit"
+32202055,"EGG, CHEESE & SAUSAGE GRIDDLE CAKE SANDWICH","Egg, cheese, and sausage griddle cake sandwich"
+32202060,"EGG & SAUSAGE ON BISCUIT","Egg and sausage on biscuit"
+32202070,"EGG, CHEESE & BACON ON BISCUIT","Egg, cheese, and bacon on biscuit"
+32202075,"EGG, CHEESE & BACON GRIDDLE CAKE SANDWICH","Egg, cheese, and bacon griddle cake sandwich"
+32202080,"EGG, CHEESE & BACON ON ENGLISH MUFFIN","Egg, cheese, and bacon on English muffin"
+32202085,"EGG, CHEESE & BACON ON BAGEL","Egg, cheese and bacon on bagel"
+32202090,"EGG & BACON ON BISCUIT","Egg and bacon on biscuit"
+32202110,"EGG & HAM ON BISCUIT","Egg and ham on biscuit"
+32202120,"EGG, CHEESE & SAUSAGE ON BAGEL","Egg, cheese and sausage on bagel"
+32202130,"EGG & STEAK ON BISCUIT","Egg and steak on biscuit"
+32202200,"EGG & CHEESE ON BISCUIT","Egg and cheese on biscuit"
+32203010,"EGG SALAD SANDWICH","Egg salad sandwich"
+32204010,"SCRAMBLED EGG SANDWICH","Scrambled egg sandwich"
+32300100,"EGG DROP SOUP","Egg drop soup"
+32301100,"GARLIC EGG SOUP, P.R. (SOPA DE AJO)","Garlic egg soup, Puerto Rican style (Sopa de ajo)"
+32400060,"EGG WHITE OMELET, SCRAMBLED, OR FRIED, MADE WITH MARGARINE","Egg white omelet, scrambled, or fried, made with margarine"
+32400065,"EGG WHITE OMELET, SCRAMBLED, OR FRIED, MADE WITH OIL","Egg white omelet, scrambled, or fried, made with oil"
+32400070,"EGG WHITE OMELET, SCRAMBLED, OR FRIED, MADE WITH BUTTER","Egg white omelet, scrambled, or fried, made with butter"
+32400075,"EGG WHITE OMELET, SCRAMBLED, OR FRIED, MADE WITH COOKING SPR","Egg white omelet, scrambled, or fried, made with cooking spray"
+32400080,"EGG WHITE OMELET, SCRAMBLED, OR FRIED, MADE WITHOUT FAT","Egg white omelet, scrambled, or fried, made without fat"
+32400100,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/CHEESE, FAT ADDED","Egg white, omelet, scrambled, or fried, with cheese, fat added in cooking"
+32400110,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ CHEESE, FAT NOT A","Egg white, omelet, scrambled, or fried, with cheese, fat not added in cooking"
+32400120,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, WITH CHEESE, NS FAT","Egg white, omelet, scrambled, or fried, with cheese, NS as to fat added in cooking"
+32400200,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ MEAT, FAT ADDED","Egg white, omelet, scrambled, or fried, with meat, fat added in cooking"
+32400210,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ MEAT, FAT NOT ADD","Egg white, omelet, scrambled, or fried, with meat, fat not added in cooking"
+32400220,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, WITH MEAT, NS FAT","Egg white, omelet, scrambled, or fried, with meat, NS as to fat added in cooking"
+32400300,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ VEGS, FAT ADDED","Egg white, omelet, scrambled, or fried, with vegetables, fat added in cooking"
+32400310,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ VEGS, FAT NOT ADD","Egg white, omelet, scrambled, or fried, with vegetables, fat not added in cooking"
+32400320,"EGG WHITE, OMELET, SCRAMBLED, OR FRIED, W/ VEGS, NS FAT","Egg white, omelet, scrambled, or fried, with vegetables, NS as to fat added in cooking"
+32400400,"EGG WHITE, OMELET, SCR OR FRIED, W/ CHEESE & MEAT, FAT ADDED","Egg white, omelet, scrambled, or fried, with cheese and meat, fat added in cooking"
+32400410,"EGG WHITE, OMELET, SCR OR FRIED, W/ CHEESE & MEAT, FAT NOT A","Egg white, omelet, scrambled, or fried, with cheese and meat, fat not added in cooking"
+32400420,"EGG WHITE, OMELET, SCR OR FRIED, W/ CHEESE & MEAT, NS FAT AD","Egg white, omelet, scrambled, or fried, with cheese and meat, NS as to fat added in cooking"
+32400500,"EGG WHITE, OMELET, SCR OR FRIED, W/CHEESE & VEG, FAT ADDED","Egg white, omelet, scrambled, or fried, with cheese and vegetables, fat added in cooking"
+32400510,"EGG WHITE, OMELET, SCR OR FRIED, W/CHEESE & VEG, FAT NOT ADD","Egg white, omelet, scrambled, or fried, with cheese and vegetables, fat not added in cooking"
+32400520,"EGG WHITE, OMELET, SCR OR FRIED, W/CHEESE & VEG, NS FAT ADDE","Egg white, omelet, scrambled, or fried, with cheese and vegetables, NS as to fat added in cooking"
+32400600,"EGG WHITE, OMELET, SCR OR FRIED, W/MEAT & VEG, FAT ADDED","Egg white, omelet, scrambled, or fried, with meat and vegetables, fat added in cooking"
+32400610,"EGG WHITE, OMELET, SCR OR FRIED, W/MEAT & VEG, FAT NOT ADDED","Egg white, omelet, scrambled, or fried, with meat and vegetables, fat not added in cooking"
+32400620,"EGG WHITE, OMELET, SCR OR FRIED, W/MEAT & VEG, NS FAT ADDED","Egg white, omelet, scrambled, or fried, with meat and vegetables, NS as to fat added in cooking"
+32400700,"EGG WHITE,OMELET,SCR OR FRIED,W/CHEESE, MEAT&VEG,FAT ADDED","Egg white, omelet, scrambled, or fried, with cheese, meat, and vegetables, fat added in cooking"
+32400710,"EGG WHITE,OMELET,SCR OR FRIED,W/CHEESE, MEAT&VEG,FAT NOT ADD","Egg white, omelet, scrambled, or fried, with cheese, meat, and vegetables, fat not added in cooking"
+32400720,"EGG WHITE,OMELET,SCR OR FRIED,W/CHEESE, MEAT&VEG,NS FAT ADDE","Egg white, omelet, scrambled, or fried, with cheese, meat, and vegetables, NS as to fat added in cooking"
+32401000,"MERINGUES","Meringues"
+33001000,"EGG SUB, OMELET, SCR, OR FRIED, MADE W/ MARGARINE","Egg substitute, omelet, scrambled, or fried, made with margarine"
+33001010,"EGG SUB, OMELET, SCR, OR FRIED, MADE W/ OIL","Egg substitute, omelet, scrambled, or fried, made with oil"
+33001020,"EGG SUB, OMELET, SCR, OR FRIED, MADE W/ BUTTER","Egg substitute, omelet, scrambled, or fried, made with butter"
+33001040,"EGG SUB, OMELET, SCR, OR FRIED, MADE W/ COOKING SPRAY","Egg substitute, omelet, scrambled, or fried, made with cooking spray"
+33001050,"EGG SUB, OMELET, SCR, OR FRIED, MADE WO/ FAT","Egg substitute, omelet, scrambled, or fried, made without fat"
+33001100,"EGG SUBSTITUTE, CHSE FLAV,OMELET,SCRM,FRIED,FAT ADDED","Egg substitute, cheese flavored, omelet, scrambled, or fried, fat added in cooking"
+33001110,"EGG SUBSTITUTE, CHSE FLAV,OMELET,SCRM,FRIED,NO FAT","Egg substitute, cheese flavored, omelet, scrambled, or fried, fat not added in cooking"
+33001120,"EGG SUBSTITUTE, CHSE FLAV,OMELET,SCRM,FRIED,NS AS TO FAT","Egg substitute, cheese flavored, omelet, scrambled, or fried, NS as to fat added in cooking"
+33001200,"EGG SUBSTITUTE, VEG FLAV,OMELET,SCRM,FRIED,FAT ADDED","Egg substitute, vegetable flavored, omelet, scrambled, or fried, fat added in cooking"
+33001210,"EGG SUBSTITUTE, VEG FLAV,OMELET,SCRM,FRIED,NO FAT","Egg substitute, vegetable flavored, omelet, scrambled, or fried, fat not added in cooking"
+33001220,"EGG SUBSTITUTE, VEG FLAV,OMELET,SCRM,FRIED,NS AS TO FAT","Egg substitute, vegetable flavored, omelet, scrambled, or fried, NS as to fat added in cooking"
+33401000,"EGG SUB, OMELET, SCR, OR FRIED, W/ CHEESE, FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese, fat added in cooking"
+33401010,"EGG SUB, OMELET, SCR, OR FRIED, W/ CHEESE, FAT NOT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese, fat not added in cooking"
+33401020,"EGG SUB, OMELET, SCR, OR FRIED, W/ CHEESE, NS FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese, NS as to fat added in cooking"
+33401100,"EGG SUB, OMELET, SCR, OR FRIED, W/ MEAT, FAT ADDED","Egg substitute, omelet, scrambled, or fried, with meat, fat added in cooking"
+33401110,"EGG SUB, OMELET, SCR, OR FRIED, W/ MEAT, FAT NOT ADDED","Egg substitute, omelet, scrambled, or fried, with meat, fat not added in cooking"
+33401120,"EGG SUB, OMELET, SCR, OR FRIED, W/ MEAT, NS FAT ADDED","Egg substitute, omelet, scrambled, or fried, with meat, NS as to fat added in cooking"
+33401200,"EGG SUB, OMELET, SCR OR FRIED, W/VEGS, FAT ADDED IN COOKING","Egg substitute, omelet, scrambled, or fried, with vegetables, fat added in cooking"
+33401210,"EGG SUB, OMELET, SCR OR FRIED, W/VEGS, FAT NOT ADDED IN COOK","Egg substitute, omelet, scrambled, or fried, with vegetables, fat not added in cooking"
+33401220,"EGG SUB, OMELET, SCR OR FRIED, W/VEGS, NS FAT ADDED IN COOKI","Egg substitute, omelet, scrambled, or fried, with vegetables, NS as to fat added in cooking"
+33401300,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE&MEAT, FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and meat, fat added in cooking"
+33401310,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE&MEAT, FAT NOT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and meat, fat not added in cooking"
+33401320,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE&MEAT, NS FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and meat, NS as to fat added in cooking"
+33401400,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE & VEG, FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and vegetables, fat added in cooking"
+33401410,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE & VEG, FAT NOT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and vegetables, fat not added in cooking"
+33401420,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE & VEG, NS FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese and vegetables, NS as to fat added in cooking"
+33401500,"EGG SUB, OMELET, SCR OR FRIED, W/ MEAT & VEG, FAT ADDED","Egg substitute, omelet, scrambled, or fried, with meat and vegetables, fat added in cooking"
+33401510,"EGG SUB, OMELET, SCR OR FRIED, W/ MEAT & VEG, FAT NOT ADDED","Egg substitute, omelet, scrambled, or fried, with meat and vegetables, fat not added in cooking"
+33401520,"EGG SUB, OMELET, SCR OR FRIED, W/ MEAT & VEG, NS FAT ADDED","Egg substitute, omelet, scrambled, or fried, with meat and vegetables, NS as to fat added in cooking"
+33401600,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE,MEAT&VEG,FAT ADDED","Egg substitute, omelet, scrambled, or fried, with cheese, meat, and vegetables, fat added in cooking"
+33401610,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE,MEAT&VEG,FAT NOT ADD","Egg substitute, omelet, scrambled, or fried, with cheese, meat, and vegetables, fat not added in cooking"
+33401620,"EGG SUB, OMELET, SCR OR FRIED, W/CHEESE,MEAT&VEG,NS FAT ADDE","Egg substitute, omelet, scrambled, or fried, with cheese, meat, and vegetables, NS as to fat added in cooking"
+41101000,"BEANS, DRY, COOKED, NS AS TO TYPE, NS ADDED FAT","Beans, dry, cooked, NS as to type and as to fat added in cooking"
+41101010,"BEANS, DRY, COOKED, NS AS TO TYPE, ADDED FAT","Beans, dry, cooked, NS as to type, fat added in cooking"
+41101020,"BEANS, DRY, COOKED, NS AS TO TYPE, NO FAT ADDED","Beans, dry, cooked, NS as to type, fat not added in cooking"
+41101100,"WHITE BEAN, DRY, COOKED, NS AS TO ADDED FAT","White beans, dry, cooked, NS as to fat added in cooking"
+41101110,"WHITE BEAN, DRY, COOKED, FAT ADDED","White beans, dry, cooked, fat added in cooking"
+41101120,"WHITE BEAN, DRY, COOKED, NO FAT ADDED","White beans, dry, cooked, fat not added in cooking"
+41101200,"WHITE BEANS, CANNED, LOW SODIUM, NS AS TO FAT ADDED","White beans, canned, low sodium, NS as to fat added in cooking"
+41101210,"WHITE BEANS, CANNED, LOW SODIUM, FAT ADDED IN COOKING","White beans, canned, low sodium, fat added in cooking"
+41101220,"WHITE BEANS, CANNED, LOW SODIUM, FAT NOT ADDED IN COOKING","White beans, canned, low sodium, fat not added in cooking"
+41102000,"BLACK, BROWN OR BAYO BEAN, DRY, COOKED, FAT NS","Black, brown, or Bayo beans, dry, cooked, NS as to fat added in cooking"
+41102010,"BLACK, BROWN OR BAYO BEAN, DRY, COOKED, FAT ADDED","Black, brown, or Bayo beans, dry, cooked, fat added in cooking"
+41102020,"BLACK, BROWN OR BAYO BEAN, DRY, COOKED, NO FAT","Black, brown, or Bayo beans, dry, cooked, fat not added in cooking"
+41102100,"BLACK, BROWN, OR BAYO BEANS, CANNED, LOW SODIUM, NS FAT","Black, brown, or Bayo beans, canned, low sodium, NS as to fat added in cooking"
+41102110,"BLACK, BROWN, OR BAYO BEANS, CANNED, LOW SODIUM, FAT ADDED","Black, brown, or Bayo beans, canned, low sodium, fat added in cooking"
+41102120,"BLACK, BROWN, OR BAYO BEANS, CANNED, LOW SODIUM, NO FAT","Black, brown, or Bayo beans, canned, low sodium, fat not added in cooking"
+41102200,"FAVA BEANS, COOKED, NS AS TO ADDED FAT","Fava beans, cooked, NS as to fat added in cooking"
+41102210,"FAVA BEANS, COOKED, FAT ADDED","Fava beans, cooked, fat added in cooking"
+41102220,"FAVA BEANS, COOKED, NO FAT ADDED","Fava beans, cooked, fat not added in cooking"
+41103000,"LIMA  BEANS, DRY, COOKED, NS AS TO ADDED FAT","Lima beans, dry, cooked, NS as to fat added in cooking"
+41103010,"LIMA  BEANS, DRY, COOKED, FAT ADDED","Lima beans, dry, cooked, fat added in cooking"
+41103020,"LIMA  BEANS, DRY, COOKED, NO FAT ADDED","Lima beans, dry, cooked, fat not added in cooking"
+41103050,"PINK BEANS, DRY, COOKED, NS AS TO FAT","Pink beans, dry, cooked, NS as to fat added in cooking"
+41103060,"PINK BEANS, DRY, COOKED, NO FAT ADDED","Pink beans, dry, cooked, fat not added in cooking"
+41103070,"PINK BEANS, DRY, COOKED, FAT ADDED","Pink beans, dry, cooked, fat added in cooking"
+41104000,"PINTO, CALICO/RED/MEX BEAN, DRY, COOKED, FAT NS","Pinto, calico, or red Mexican beans, dry, cooked, NS as to fat added in cooking"
+41104010,"PINTO, CALICO/RED/MEX BEAN, DRY, COOKED, FAT ADDED","Pinto, calico, or red Mexican beans, dry, cooked, fat added in cooking"
+41104020,"PINTO, CALICO/RED/MEX BEAN, DRY, COOKED, NO FAT","Pinto, calico, or red Mexican beans, dry, cooked, fat not added in cooking"
+41104100,"PINTO,CALICO, RED MEXICAN BEANS, CANNED, LOW SODIUM, NS FAT","Pinto, calico, or red Mexican beans, canned, low sodium, NS as to fat added in cooking"
+41104110,"PINTO,CALICO,RED MEX BEANS, CANNED, LOW SODIUM, FAT ADDED","Pinto, calico, or red Mexican beans, canned, low sodium, fat added in cooking"
+41104120,"PINTO,CALICO,RED MEXICAN BEANS, CANNED, LOW SODIUM, NO FAT","Pinto, calico, or red Mexican beans, canned, low sodium, fat not added in cooking"
+41106000,"RED KIDNEY BEANS, DRY, COOKED, NS AS TO ADDED FAT","Red kidney beans, dry, cooked, NS as to fat added in cooking"
+41106010,"RED KIDNEY BEANS, DRY, COOKED, FAT ADDED","Red kidney beans, dry, cooked, fat added in cooking"
+41106020,"RED KIDNEY BEANS, DRY, COOKED, NO FAT ADDED","Red kidney beans, dry, cooked, fat not added in cooking"
+41106100,"RED KIDNEY BEANS, CANNED, LOW SODIUM, FAT ADDED IN COOKING","Red kidney beans, canned, low sodium, NS as to fat added in cooking"
+41106110,"RED KIDNEY BEANS, CANNED, LOW SODIUM, FAT ADDED IN COOKING","Red kidney beans, canned, low sodium, fat added in cooking"
+41106120,"RED KIDNEY BEANS, CANNED, LOW SODIUM, FAT NOT ADDED","Red kidney beans, canned, low sodium, fat not added in cooking"
+41107000,"SOYBEANS, COOKED, FAT NOT ADDED","Soybeans, cooked, fat not added in cooking"
+41108000,"MUNG BEANS, NO FAT ADDED","Mung beans, fat not added in cooking"
+41108010,"MUNG BEANS, FAT ADDED","Mung beans, fat added in cooking"
+41108020,"MUNG BEANS, NS AS TO FAT ADDED","Mung beans, NS as to fat added in cooking"
+41109000,"MUNGO BEANS, COOKED, NO FAT ADDED","Mungo beans, cooked, fat not added in cooking"
+41201010,"BAKED BEANS, NFS","Baked beans, NFS"
+41201020,"BAKED BEANS, VEGETARIAN","Baked beans, vegetarian"
+41202020,"CHILI BEANS, BARBECUE BEANS, RANCH OR MEXICAN STYLE","Chili beans, barbecue beans, ranch style beans or Mexican- style beans"
+41202500,"BEANS AND TOMATOES, NS AS TO FAT ADDED","Beans and tomatoes, NS as to fat added in cooking"
+41202505,"BEANS AND TOMATOES, FAT NOT ADDED","Beans and tomatoes, fat not added in cooking"
+41202510,"BEANS AND TOMATOES, FAT ADDED","Beans and tomatoes, fat added in cooking"
+41203030,"BLACK BEAN SALAD","Black bean salad"
+41204020,"BOSTON BAKED BEANS","Boston baked beans"
+41205010,"REFRIED BEANS","Refried beans"
+41205015,"REFRIED BEANS, FAT NOT ADDED IN COOKING","Refried beans, fat not added in cooking"
+41205020,"REFRIED BEANS W/ CHEESE","Refried beans with cheese"
+41205030,"REFRIED BEANS W/ MEAT","Refried beans with meat"
+41205040,"REFRIED BEANS, CANNED, LOW SODIUM","Refried beans, canned, low sodium"
+41205050,"BEAN DIP, W/ REFRIED BEANS","Bean dip, made with refried beans"
+41205070,"HUMMUS","Hummus"
+41205100,"BLACK BEAN SAUCE","Black bean sauce"
+41206030,"BEANS & FRANKS","Beans and franks"
+41207030,"BEANS, DRY, COOKED, W/ GROUND BEEF","Beans, dry, cooked with ground beef"
+41208030,"PORK & BEANS","Pork and beans"
+41208100,"BEANS, DRY, COOKED, W/ PORK","Beans, dry, cooked with pork"
+41209000,"FALAFEL","Falafel"
+41210000,"BEAN CAKE, JAPANESE STYLE","Bean cake"
+41210090,"STEWED BEANS W/ PORK, TOMATOES, & CHILI PEPPERS, MEXICAN","Stewed beans with pork, tomatoes, and chili peppers, Mexican style (Frijoles a la charra)"
+41210100,"STEWED RED BEANS, P.R.","Stewed red beans, Puerto Rican style (Habichuelas coloradas guisadas)"
+41210110,"STEWED DRY LIMA BEANS, P.R","Stewed dry lima beans, Puerto Rican style"
+41210120,"STEWED WHITE BEANS, P.R.","Stewed white beans, Puerto Rican style"
+41210150,"STEWED PINK BEANS W/ WHITE POTATOES & HAM, P.R.","Stewed pink beans with white potatoes and ham, Puerto Rican style"
+41210160,"STEWED PINK BEANS W/ PIG'S FEET, P.R","Stewed pink beans with pig's feet, Puerto Rican style"
+41210170,"STEWED RED BEANS W/ PIG'S FEET, P.R.","Stewed red beans with pig's feet, Puerto Rican style"
+41210180,"STEWED WHITE BEANS W/ PIG'S FEET, P.R.","Stewed white beans with pig's feet, Puerto Rican style"
+41210190,"STEWED RED BEANS W/ PIGS FEET & POTATO, P.R.","Stewed red beans with pig's feet and potatoes, Puerto Rican style"
+41210200,"BLACK BEANS, CUBAN","Black beans, Cuban style (Habichuelas negras guisadas a la Cubana)"
+41221010,"BAKED BEANS, LOW SODIUM","Baked beans, low sodium"
+41221020,"CHILI WITH BEANS, WITHOUT MEAT","Chili with beans, without meat"
+41301000,"COWPEAS, DRY, COOKED, NS AS TO ADDED FAT","Cowpeas, dry, cooked, NS as to fat added in cooking"
+41301010,"COWPEAS, DRY, COOKED, FAT ADDED","Cowpeas, dry, cooked, fat added in cooking"
+41301020,"COWPEAS, DRY, COOKED, NO FAT ADDED","Cowpeas, dry, cooked, fat not added in cooking"
+41302000,"CHICKPEAS, DRY, COOKED, NS AS TO ADDED FAT","Chickpeas, dry, cooked, NS as to fat added in cooking"
+41302010,"CHICKPEAS, DRY, COOKED, FAT ADDED","Chickpeas, dry, cooked, fat added in cooking"
+41302020,"CHICKPEAS, DRY, COOKED, NO FAT ADDED","Chickpeas, dry, cooked, fat not added in cooking"
+41302100,"CHICKPEAS, CANNED, LOW SODIUM, NS AS TO FAT ADDED IN COOKING","Chickpeas, canned, low sodium, NS as to fat added in cooking"
+41302110,"CHICKPEAS, CANNED, LOW SODIUM, FAT ADDED IN COOKING","Chickpeas, canned, low sodium, fat added in cooking"
+41302120,"CHICKPEAS, CANNED, LOW SODIUM, FAT NOT ADDED IN COOKING","Chickpeas, canned, low sodium, fat not added in cooking"
+41303000,"GREEN/YELLOW SPLIT PEAS, DRY, COOKED, NO FAT ADDED","Green or yellow split peas, dry, cooked, fat not added in cooking"
+41303010,"GREEN OR YELLOW SPLIT PEAS, DRY, COOKED, FAT ADDED","Green or yellow split peas, dry, cooked, fat added in cooking"
+41303020,"SPLIT PEAS, DRY, COOKED, NS AS TO ADDED FAT","Green or yellow split peas, dry, cooked, NS as to fat added in cooking"
+41303500,"STEWED GREEN PEAS, PUERTO RICAN STYLE","Stewed green peas, Puerto Rican style"
+41303550,"STEWED GREEN PEAS, W/ PIG'S FEET & POTATO, P.R.","Stewed green peas with pig's feet and potatoes, Puerto Rican style"
+41304000,"WASABI PEAS","Wasabi peas"
+41304030,"PEAS, DRY, COOKED W/ PORK","Peas, dry, cooked with pork"
+41304130,"COWPEAS, DRY, COOKED W/ PORK","Cowpeas, dry, cooked with pork"
+41304980,"LENTILS, DRY, COOKED, NS AS TO ADDED FAT","Lentils, dry, cooked, NS as to fat added in cooking"
+41304990,"LENTILS, DRY, COOKED, FAT ADDED","Lentils, dry, cooked, fat added in cooking"
+41305000,"LENTILS, DRY, COOKED, NO FAT ADDED","Lentils, dry, cooked, fat not added in cooking"
+41306000,"LOAF, LENTIL","Loaf, lentil"
+41310100,"STEWED PIGEON PEAS, P.R.","Stewed pigeon peas, Puerto Rican style (Gandules guisados, Gandur, Gandules)"
+41310150,"STEWED CHICKPEAS, P.R.","Stewed chickpeas, Puerto Rican style"
+41310160,"STEWED CHICKPEAS, W/ POTATOES, P.R.","Stewed chickpeas, with potatoes, Puerto Rican style"
+41310200,"CHICKPEAS STEWED W/ PIG'S FEET, P.R.","Chickpeas stewed with pig's feet, Puerto Rican style (Garbanzos guisados con patitas de cerdo)"
+41310210,"CHICKPEAS, W/ SPANISH SAUSAGE, P.R.","Stewed chickpeas with Spanish sausages, Puerto Rican style (Garbanzos guisados con chorizos)"
+41310220,"FRIED CHICKPEAS W/ BACON, P.R.","Fried chickpeas with bacon, Puerto Rican style (Garbanzos fritos con tocineta)"
+41310310,"STEWED BLACKEYE PEAS OR COWPEAS, P.R.","Stewed blackeye peas or cowpeas, Puerto Rican style"
+41311000,"PAPAD(INDIAN APPETIZER),GRILLED OR BROILED","Papad (Indian appetizer), grilled or broiled"
+41410010,"SOY NUTS","Soy nuts"
+41410015,"SOY CHIPS","Soy chips"
+41420010,"SOYBEAN CURD","Soybean curd"
+41420050,"SOYBEAN CURD CHEESE","Soybean curd cheese"
+41420100,"MISO SAUCE (INCLUDES AE SAUCE)","Miso sauce"
+41420110,"MISO (FERMENTED SOYBEAN PASTE)","Miso (fermented soybean paste)"
+41420200,"NATTO (FERMENTED SOYBEAN PRODUCT)","Natto (fermented soybean product)"
+41420250,"HOISIN SAUCE","Hoisin sauce"
+41420300,"SOY SAUCE","Soy sauce"
+41420350,"SOY SAUCE, REDUCED SODIUM","Soy sauce, reduced sodium"
+41420380,"SOY YOGURT","Soy yogurt"
+41420400,"TERIYAKI SAUCE (INCLUDE ORIENTAL BARBECUE SAUCE)","Teriyaki sauce"
+41420410,"TERIYAKI SAUCE, REDUCED SODIUM","Teriyaki sauce, reduced sodium"
+41420450,"WORCESTERSHIRE SAUCE","Worcestershire sauce"
+41421010,"SOYBEAN CURD, DEEP-FRIED","Soybean curd, deep fried"
+41421020,"SOYBEAN CURD, BREADED, FRIED","Soybean curd, breaded, fried"
+41422010,"SOYBEAN MEAL","Soybean meal"
+41425010,"VERMICELLI, MADE FROM SOYBEANS","Vermicelli, made from soybeans"
+41440000,"TEXTURED VEGETABLE PROTEIN, DRY","Textured vegetable protein, dry"
+41480000,"TOFU FROZEN DESSERT, NOT CHOCOLATE (INCL TOFUTTI)","Tofu, frozen dessert, flavors other than chocolate"
+41480010,"TOFU FROZEN DESSERT, CHOCOLATE (INCLUDE TOFUTTI)","Tofu, frozen dessert, chocolate"
+41601010,"BEAN SOUP, NFS","Bean soup, NFS"
+41601020,"BEAN WITH BACON OR HAM SOUP, CANNED OR READY-TO-SERVE","Bean with bacon or ham soup, canned or ready-to-serve"
+41601030,"BLACK BEAN SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Black bean soup, home recipe, canned or ready-to-serve"
+41601040,"LIMA BEAN SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Lima bean soup, home recipe, canned or ready-to-serve"
+41601070,"SOYBEAN SOUP, MISO BROTH","Soybean soup, miso broth"
+41601080,"PINTO BEAN SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Pinto bean soup, home recipe, canned or ready-to-serve"
+41601090,"BEAN SOUP, WITH MACARONI, HOME RECIPE, CANNED, OR READY-TO-S","Bean soup, with macaroni, home recipe, canned, or ready-to-serve"
+41601100,"PORTUGUESE BEAN SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Portuguese bean soup, home recipe, canned or ready-to-serve"
+41601110,"BEAN AND HAM SOUP, CHUNKY STYLE, CANNED OR READY-TO-SERVE","Bean and ham soup, chunky style, canned or ready-to-serve"
+41601130,"BEAN SOUP, MIXED BEANS, HOME RECIPE, CANNED OR READY-TO-SERV","Bean soup, mixed beans, home recipe, canned or ready-to-serve"
+41601140,"BEAN SOUP, HOME RECIPE","Bean soup, home recipe"
+41601160,"BEAN & HAM SOUP, CAN, REDUCED SODIUM, W/ WATER/RTS","Bean and ham soup, canned, reduced sodium, prepared with water or ready-to-serve"
+41601180,"BEAN & HAM SOUP, HOME RECIPE","Bean and ham soup, home recipe"
+41601200,"LIQUID FROM STEWED KIDNEY BEANS, P.R.","Liquid from stewed kidney beans, Puerto Rican style"
+41602010,"PEA AND HAM SOUP, CHUNKY STYLE, CANNED OR READY-TO-SERVE","Pea and ham soup, chunky style, canned or ready-to-serve"
+41602020,"GARBANZO BEAN OR CHICKPEA SOUP, HOME RECIPE, CANNED OR READY","Garbanzo bean or chickpea soup, home recipe, canned or ready-to-serve"
+41602030,"SPLIT PEA & HAM SOUP","Split pea and ham soup"
+41602050,"SPLIT PEA SOUP","Split pea soup"
+41602070,"SPLIT PEA SOUP, CAN, REDUCED SODIUM, W/ WATER/RTS","Split pea soup, canned, reduced sodium, prepared with water or ready-to-serve"
+41602090,"SPLIT PEA & HAM SOUP, CAN, REDUCED SODIUM, W/ WATER/RTS","Split pea and ham soup, canned, reduced sodium, prepared with water or ready-to-serve"
+41603010,"LENTIL SOUP, HOME RECIPE, CANNED, OR READY-TO-SERVE","Lentil soup, home recipe, canned, or ready-to-serve"
+41610100,"WHITE BEAN SOUP, P.R.","White bean soup, Puerto Rican style (Sopon de habichuelas blancas)"
+41810200,"BACON STRIP, MEATLESS","Bacon strip, meatless"
+41810250,"BACON BITS, MEATLESS","Bacon bits, meatless"
+41810400,"BREAKFAST LINK,PATTY,/SLICE, MEATLESS","Breakfast link, pattie, or slice, meatless"
+41810600,"CHICKEN, MEATLESS NFS","Chicken, meatless, NFS"
+41810610,"CHICKEN, MEATLESS, BREADED, FRIED (INCL LOMA LINDA)","Chicken, meatless, breaded, fried"
+41811200,"FISH STICK, MEATLESS","Fish stick, meatless"
+41811400,"FRANKFURTER OR HOT DOG, MEATLESS","Frankfurter or hot dog, meatless"
+41811600,"LUNCHEON SLICE,MEATLESS-BEEF,CHICKEN,SALAM / TURKEY","Luncheon slice, meatless-beef, chicken, salami or turkey"
+41811800,"MEATBALL, MEATLESS","Meatball, meatless"
+41811850,"SCALLOPS, MEATLESS, BREADED, FRIED","Scallops, meatless, breaded, fried (made with meat substitute)"
+41811890,"VEGETARIAN BURGER OR PATTY, MEATLESS, NO BUN","Vegetarian burger or patty, meatless, no bun"
+41811950,"SWISS STEAK, W/ GRAVY, MEATLESS","Swiss steak, with gravy, meatless"
+41812000,"SANDWICH SPREAD, MEAT SUBSTITUTE TYPE","Sandwich spread, meat substitute type"
+41812400,"VEGETARIAN POT PIE","Vegetarian pot pie"
+41812450,"VEGETARIAN CHILI (MADE W/ MEAT SUBSTITUTE)","Vegetarian chili (made with meat substitute)"
+41812500,"TOFU & VEG (W/ CARROT/DK GRN, NO POTATO) W/ SOY SAUCE","Tofu and vegetables (including carrots, broccoli, and/or dark-green leafy vegetables (no potatoes)), with soy-based sauce (mixture)"
+41812510,"TOFU & VEG (NO CARROT/DK GRN, NO POTATO) W/ SOY SAUCE","Tofu and vegetables (excluding carrots, broccoli, and dark-green leafy vegetables (no potatoes)), with soy-based sauce (mixture)"
+41812600,"VEGETARIAN FILLET","Vegetarian, fillet"
+41812800,"VEGETARIAN STEW","Vegetarian stew"
+41812850,"VEGETARIAN STROGANOFF (MADE W/ MEAT SUBSTITUTE)","Vegetarian stroganoff (made with meat substitute)"
+41812900,"VEGETARIAN MEAT LOAF OR PATTIES","Vegetarian meat loaf or patties (meat loaf made with meat substitute)"
+41813000,"VEGATARIAN BOUILLON, DRY","Vegetarian bouillon, dry"
+41901020,"SOYBURGER W/ CHEESE ON BUN","Soyburger, meatless, with cheese on bun"
+42100050,"NUTS, NFS","Nuts, nfs"
+42100100,"ALMONDS, NFS","Almonds, NFS"
+42101000,"ALMONDS, UNROASTED","Almonds, unroasted"
+42101100,"ALMONDS, ROASTED","Almonds, roasted"
+42101200,"ALMONDS, DRY ROASTED","Almonds, dry roasted (assume salted)"
+42101210,"ALMONDS, DRY ROASTED, W/O SALT","Almonds, dry roasted, without salt"
+42101350,"ALMONDS, HONEY-ROASTED","Almonds, honey-roasted"
+42102000,"BRAZIL NUTS","Brazil nuts"
+42104000,"CASHEW NUTS, NFS","Cashew nuts, NFS"
+42104100,"CASHEW NUTS, ROASTED","Cashew nuts, roasted (assume salted)"
+42104110,"CASHEW NUTS, ROASTED, W/O SALT","Cashew nuts, roasted, without salt"
+42104200,"CASHEW NUTS, DRY ROASTED","Cashew nuts, dry roasted"
+42104205,"CASHEW NUTS, DRY ROASTED, WITHOUT SALT","Cashew nuts, dry roasted, without salt"
+42104500,"CASHEW NUTS, HONEY-ROASTED","Cashew nuts, honey-roasted"
+42105000,"CHESTNUTS, ROASTED","Chestnuts, roasted"
+42106000,"COCONUT MEAT, FRESH","Coconut meat, fresh"
+42106020,"COCONUT MEAT, DRIED, SWEETENED, SHREDDED","Coconut meat, dried, sweetened"
+42107000,"FILBERTS, HAZELNUTS","Filberts, hazelnuts"
+42109000,"MACADAMIA NUTS, UNROASTED","Macadamia nuts, unroasted"
+42109100,"MACADAMIA NUTS, ROASTED","Macadamia nuts, roasted"
+42110000,"MIXED NUTS, NFS","Mixed nuts, NFS"
+42110100,"MIXED NUTS, ROASTED, W/ PEANUTS","Mixed nuts, roasted, with peanuts"
+42110150,"MIXED NUTS, ROASTED, W/O PEANUTS","Mixed nuts, roasted, without peanuts"
+42110200,"MIXED NUTS, DRY ROASTED","Mixed nuts, dry roasted"
+42110300,"MIXED NUTS, HONEY-ROASTED, WITH PEANUTS","Mixed nuts, honey-roasted, with peanuts"
+42111000,"PEANUTS, NFS","Peanuts, NFS"
+42111030,"PEANUTS, BOILED","Peanuts, boiled"
+42111100,"PEANUTS, ROASTED, SALTED","Peanuts, roasted, salted"
+42111110,"PEANUTS, ROASTED, W/O SALT","Peanuts, roasted, without salt"
+42111200,"PEANUTS, DRY ROASTED, SALTED","Peanuts, dry roasted, salted"
+42111210,"PEANUTS, DRY ROASTED, W/O SALT","Peanuts, dry roasted, without salt"
+42111500,"PEANUTS, HONEY ROASTED (INCL BEERNUTS)","Peanuts, honey-roasted"
+42112000,"PECANS","Pecans"
+42113000,"PINE NUTS (PIGNOLIAS)","Pine nuts (Pignolias)"
+42114130,"PISTACHIO NUTS","Pistachio nuts"
+42116000,"WALNUTS","Walnuts"
+42116100,"WALNUTS, HONEY-ROASTED","Walnuts, honey-roasted"
+42200500,"ALMOND BUTTER","Almond butter"
+42200600,"ALMOND PASTE (MARZIPAN PASTE)","Almond paste (Marzipan paste)"
+42201000,"CASHEW BUTTER","Cashew butter"
+42202000,"PEANUT BUTTER","Peanut butter"
+42202010,"PEANUT BUTTER, LOW SODIUM","Peanut butter, low sodium"
+42202100,"PEANUT BUTTER, REDUCED SODIUM & REDUCED SUGAR","Peanut butter, reduced sodium and reduced sugar"
+42202130,"PEANUT BUTTER, REDUCED SUGAR","Peanut butter, reduced sugar"
+42202150,"PEANUT BUTTER, REDUCED FAT","Peanut butter, reduced fat"
+42202200,"PEANUT BUTTER, VITAMIN & MINERAL FORTIFIED","Peanut butter, vitamin and mineral fortified"
+42203000,"PEANUT BUTTER & JELLY","Peanut butter and jelly"
+42204050,"PEANUT SAUCE","Peanut sauce"
+42204100,"BROWN NUT GRAVY ( MEATLESS)","Brown nut gravy, meatless"
+42301010,"PEANUT BUTTER SANDWICH","Peanut butter sandwich"
+42302010,"PEANUT BUTTER & JELLY SANDWICH","Peanut butter and jelly sandwich"
+42303010,"PEANUT BUTTER & BANANA SANDWICH","Peanut butter and banana sandwich"
+42401010,"COCONUT MILK","Coconut milk (liquid expressed from grated coconut meat, water added)"
+42402010,"COCONUT CREAM, CANNED, SWEETENED (INCL COCO LOPEZ)","Coconut cream (liquid expressed from grated coconut meat), canned, sweetened"
+42403010,"COCONUT WATER (LIQUID FROM COCONUTS)","Coconut water (liquid from coconuts)"
+42404010,"COCONUT WATER, CANNED OR BOTTLED","Coconut water, canned or bottled"
+42501000,"NUT MIXTURE W/ DRIED FRUIT & SEEDS","Nut mixture with dried fruit and seeds"
+42501500,"NUT MIXTURE WITH DRIED FRUIT, SEEDS, AND CHOCOLATE","Nut mixture with dried fruit, seeds, and chocolate"
+42502000,"NUT MIXTURE W/ SEEDS","Nut mixture with seeds"
+43101000,"PUMPKIN & SQUASH SEEDS, HULLED, UNROASTED","Pumpkin and/or squash seeds, hulled, unroasted"
+43101100,"PUMPKIN & SQUASH SEEDS, HULLED, ROASTED, SALTED","Pumpkin and/or squash seeds, hulled, roasted, salted"
+43101150,"PUMPKIN & SQUASH SEEDS, HULLED, ROASTED, NO SALT","Pumpkin and/or squash seeds, hulled, roasted, without salt"
+43102000,"SUNFLOWER SEEDS, HULLED, UNROASTED, WITHOUT SALT","Sunflower seeds, hulled, unroasted, without salt"
+43102100,"SUNFLOWER SEEDS, HULLED, ROASTED, SALTED","Sunflower seeds, hulled, roasted, salted"
+43102110,"SUNFLOWER SEEDS, HULLED, ROASTED, W/O SALT","Sunflower seeds, hulled, roasted, without salt"
+43102200,"SUNFLOWER SEEDS, HULLED, DRY ROASTED","Sunflower seeds, hulled, dry roasted"
+43103000,"SESAME SEEDS (INCLUDE TOASTED)","Sesame seeds"
+43103050,"SESAME SEEDS, WHOLE SEEDS","Sesame seeds, whole seed"
+43103100,"SESAME SAUCE","Sesame sauce"
+43103200,"SESAME PASTE","Sesame paste (sesame butter made from whole seeds)"
+43103300,"SESAME BUTTER (TAHINI) FROM KERNELS","Sesame butter (tahini) (made from kernels)"
+43104000,"FLAX SEED","Flax seeds"
+43107000,"MIXED SEEDS","Mixed seeds"
+44101000,"CAROB POWDER OR FLOUR","Carob powder or flour"
+44201000,"CAROB CHIPS","Carob chips"
+44202000,"CAROB SYRUP","Carob syrup"
+50010000,"FLOUR, WHITE (INCLUDE FLOUR, NFS)","Flour, white"
+50020000,"FLOUR, WHOLE WHEAT","Flour, whole wheat"
+50030000,"BISCUIT MIX, DRY","Biscuit mix, dry"
+51000100,"BREAD, NS AS TO MAJOR FLOUR","Bread, NS as to major flour"
+51000110,"BREAD, NS AS TO MAJOR FLOUR, TOASTED","Bread, NS as to major flour, toasted"
+51000180,"BREAD, HOMEMADE/PURCH AT A BAKERY, NS AS TO FLOUR","Bread, made from home recipe or purchased at a bakery, NS as to major flour"
+51000190,"BREAD, HOMEMADE/PURCH AT A BAKERY, TOASTD,NS FLOUR","Bread, made from home recipe or purchased at a bakery, toasted, NS as to major flour"
+51000200,"ROLL, NS AS TO MAJOR FLOUR","Roll, NS as to major flour"
+51000230,"ROLL, NS AS TO MAJOR FLOUR, TOASTED","Roll, NS as to major flour, toasted"
+51000250,"ROLL, HOMEMADE/PURCH AT A BAKERY, NS AS TO FLOUR","Roll, made from home recipe or purchased at a bakery, NS as to major flour"
+51000260,"ROLL, HOMEMADE/PURCH AT A BAKERY, TOASTD,NS FLOUR","Roll, made from home recipe or purchased at a bakery, toasted, NS as to major flour"
+51000300,"ROLL, HARD, NS AS TO MAJOR FLOUR","Roll, hard, NS as to major flour"
+51000400,"ROLL, BRAN, NS AS TO TYPE OF BRAN","Roll, bran, NS as to type of bran"
+51101000,"BREAD, WHITE","Bread, white"
+51101010,"BREAD, WHITE, TOASTED","Bread, white, toasted"
+51101050,"BREAD, WHITE, HOMEMADE OR PURCHASED AT A BAKERY","Bread, white, made from home recipe or purchased at a bakery"
+51101060,"BREAD, WHITE, HOMEMADE OR PURCH AT A BAKERY TOASTED","Bread, white, made from home recipe or purchased at a bakery, toasted"
+51102010,"BREAD, WHITE W/ WHOLE WHEAT SWIRL","Bread, white with whole wheat swirl"
+51102020,"BREAD, WHITE W/ WHOLE WHEAT SWIRL, TOASTED","Bread, white with whole wheat swirl, toasted"
+51105010,"BREAD, CUBAN (INCLUDE SPANISH, PORTUGUESE)","Bread, Cuban"
+51105040,"BREAD, CUBAN, TOASTED (INCLUDE SPANISH, PORTUGUESE)","Bread, Cuban, toasted"
+51106010,"BREAD, NATIVE, WATER, P.R. (PAN CRIOLLO)","Bread, Native, water, Puerto Rican style (Pan Criollo)"
+51106020,"BREAD, NATIVE, WATER, P.R., TOASTED (PAN CRIOLLO)","Bread, Native, water, Puerto Rican style, toasted (Pan Criollo)"
+51106200,"BREAD, LARD, P.R. (PAN DE MANTECA)","Bread, lard, Puerto Rican style (Pan de manteca)"
+51106210,"BREAD, LARD, P.R., TOASTED (PAN DE MANTECA)","Bread, lard, Puerto Rican style, toasted (Pan de manteca)"
+51106300,"BREAD, CARESSED, P.R. (PAN SOBAO)","Bread, caressed, Puerto Rican style (Pan sobao)"
+51106310,"BREAD, CARESSED, P.R., TOASTED (PAN SOBAO)","Bread, caressed, Puerto Rican style, toasted (Pan sobao)"
+51107010,"BREAD, FRENCH OR VIENNA","Bread, French or Vienna"
+51107040,"BREAD, FRENCH OR VIENNA, TOASTED","Bread, French or Vienna, toasted"
+51108010,"FOCACCIA, ITALIAN FLATBREAD, PLAIN","Focaccia, Italian flatbread, plain"
+51108100,"NAAN, INDIAN FLATBREAD","Naan, Indian flatbread"
+51109010,"BREAD, ITALIAN, GRECIAN, ARMENIAN","Bread, Italian, Grecian, Armenian"
+51109040,"BREAD, ITALIAN, GRECIAN, ARMENIAN, TOASTED","Bread, Italian, Grecian, Armenian, toasted"
+51109100,"BREAD, PITA","Bread, pita"
+51109110,"BREAD, PITA, TOASTED","Bread, pita, toasted"
+51109150,"BREAD, PITA W/ FRUIT","Bread, pita with fruit"
+51109200,"BREAD, PITA W/ FRUIT, TOASTED","Bread, pita with fruit, toasted"
+51111010,"BREAD, CHEESE (INCLUDE ONION CHEESE)","Bread, cheese"
+51111040,"BREAD, CHEESE, TOASTED (INCLUDE ONION CHEESE)","Bread, cheese, toasted"
+51113010,"BREAD, CINNAMON","Bread, cinnamon"
+51113100,"BREAD, CINNAMON, TOASTED","Bread, cinnamon, toasted"
+51115010,"BREAD, CORNMEAL AND MOLASSES","Bread, cornmeal and molasses"
+51115020,"BREAD, CORNMEAL AND MOLASSES, TOASTED","Bread, cornmeal and molasses, toasted"
+51119010,"BREAD, EGG, CHALLAH","Bread, egg, Challah"
+51119040,"BREAD, EGG, CHALLAH, TOASTED","Bread, egg, Challah, toasted"
+51121010,"BREAD, GARLIC","Bread, garlic"
+51121110,"BREAD, ONION","Bread, onion"
+51121120,"BREAD, ONION, TOASTED","Bread, onion, toasted"
+51122000,"BREAD, REDUCED CALORIE/HIGH FIBER","Bread, reduced calorie and/or high fiber, white or NFS"
+51122010,"BREAD, REDUCED CALORIE/HIGH FIBER, TOASTED","Bread, reduced calorie and/or high fiber, white or NFS, toasted"
+51122100,"BREAD, REDUCED CALORIE/ HIGH FIBER, W/ FRUIT/NUTS","Bread, reduced calorie and/or high fiber, white or NFS, with fruit and/or nuts"
+51122110,"BREAD, REDUCED CALORIE/HI FIBER, W/FRUIT/NUTS,TOAST","Bread, reduced calorie and/or high fiber, white or NFS, with fruit and/or nuts, toasted"
+51122300,"BREAD, WHITE, SPECIAL FORMULA, ADDED FIBER","Bread, white, special formula, added fiber"
+51122310,"BREAD,WHITE,SPECIAL FORMULA,ADDED FIBER,TOASTED","Bread, white, special formula, added fiber, toasted"
+51123010,"BREAD, HIGH PROTEIN","Bread, high protein"
+51123020,"BREAD, HIGH PROTEIN, TOASTED","Bread, high protein, toasted"
+51127010,"BREAD, POTATO","Bread, potato"
+51127020,"BREAD, POTATO, TOASTED","Bread, potato, toasted"
+51129010,"BREAD, RAISIN","Bread, raisin"
+51129020,"BREAD, RAISIN, TOASTED","Bread, raisin, toasted"
+51130510,"BREAD, WHITE, LOW SODIUM OR NO SALT","Bread, white, low sodium or no salt"
+51130520,"BREAD, WHITE, LOW SODIUM/NO SALT, TOASTED","Bread, white, low sodium or no salt, toasted"
+51133010,"BREAD, SOUR DOUGH","Bread, sour dough"
+51133020,"BREAD, SOUR DOUGH, TOASTED","Bread, sour dough, toasted"
+51134000,"BREAD, SWEET POTATO","Bread, sweet potato"
+51134010,"BREAD, SWEET POTATO, TOASTED","Bread, sweet potato, toasted"
+51135000,"BREAD, VEGETABLE","Bread, vegetable"
+51135010,"BREAD, VEGETABLE, TOASTED","Bread, vegetable, toasted"
+51136000,"BRUSCHETTA","Bruschetta"
+51140100,"BREAD  DOUGH, FRIED","Bread, dough, fried"
+51150000,"ROLL, WHITE, SOFT","Roll, white, soft"
+51150100,"ROLL, WHITE, SOFT, TOASTED","Roll, white, soft, toasted"
+51151060,"ROLL, WHITE, SOFT, HOMEMADE/PURCH AT A BAKERY","Roll, white, soft, made from home recipe or purchased at a bakery"
+51152000,"ROLL, WHITE, SOFT, REDUCED CALORIE/ HIGH FIBER","Roll, white, soft, reduced calorie and/or high fiber"
+51152100,"ROLL, WHITE, REDUCED CALORIE/ HIGH FIBER, TOASTED","Roll, white, soft, reduced calorie and/or high fiber, toasted"
+51153000,"ROLL, WHITE, HARD","Roll, white, hard"
+51153010,"ROLL, WHITE, HARD, TOASTED","Roll, white, hard, toasted"
+51154510,"ROLL, DIET","Roll, diet"
+51154550,"ROLL, EGG BREAD","Roll, egg bread"
+51154560,"ROLL, EGG BREAD, TOASTED","Roll, egg bread, toasted"
+51154600,"ROLL, CHEESE","Roll, cheese"
+51155000,"ROLL, FRENCH OR VIENNA","Roll, French or Vienna"
+51155010,"ROLL, FRENCH OR VIENNA, TOASTED","Roll, French or Vienna, toasted"
+51156500,"ROLL, GARLIC","Roll, garlic"
+51157000,"ROLL, HOAGIE, SUBMARINE,","Roll, hoagie, submarine"
+51157010,"ROLL, HOAGIE, SUBMARINE, TOASTED","Roll, hoagie, submarine, toasted"
+51158100,"ROLL, MEXICAN, BOLILLO","Roll, Mexican, bolillo"
+51159000,"ROLL, SOUR DOUGH","Roll, sour dough"
+51160000,"ROLL, SWEET, NO FROSTING","Roll, sweet, no frosting"
+51160100,"ROLL, SWEET, CINNAMON BUN, NO FROSTING","Roll, sweet, cinnamon bun, no frosting"
+51160110,"ROLL, SWEET, CINNAMON BUN, FROSTED","Roll, sweet, cinnamon bun, frosted"
+51161000,"ROLL, SWEET, W/ FRUIT, NO FROSTING","Roll, sweet, with fruit, no frosting"
+51161020,"ROLL, SWEET, W/ FRUIT, FROSTED","Roll, sweet, with fruit, frosted"
+51161030,"ROLL, SWEET, W/ FRUIT, FROSTED, DIET","Roll, sweet, with fruit, frosted, diet"
+51161050,"ROLL, SWEET, FROSTED","Roll, sweet, frosted"
+51161250,"ROLL, SWEET, NO TOPPING, MEXICAN (PAN DULCE)","Roll, sweet, no topping, Mexican (Pan Dulce)"
+51161270,"ROLL, SWEET, SUGAR TOPPING, MEXICAN  (PAN DULCE)","Roll, sweet, sugar topping, Mexican (Pan Dulce)"
+51161280,"ROLL,SWEET,W/ RAISINS & ICING,MEXICAN (PAN DULCE)","Roll, sweet, with raisins and icing, Mexican (Pan Dulce)"
+51165000,"COFFEE CAKE, YEAST TYPE","Coffee cake, yeast type"
+51166000,"CROISSANT","Croissant"
+51166100,"CROISSANT, CHEESE","Croissant, cheese"
+51166200,"CROISSANT, CHOCOLATE","Croissant, chocolate"
+51166500,"CROISSANT, FRUIT","Croissant, fruit"
+51167000,"BRIOCHE","Brioche"
+51168000,"COFFEE BREAD, SPANISH","Bread, Spanish coffee"
+51180010,"BAGEL","Bagel"
+51180020,"BAGEL, TOASTED","Bagel, toasted"
+51180030,"BAGEL, W/ RAISINS","Bagel, with raisins"
+51180040,"BAGEL, W/ RAISINS, TOASTED","Bagel, with raisins, toasted"
+51180080,"BAGEL W/ FRUIT OTHER THAN RAISINS","Bagel, with fruit other than raisins"
+51180090,"BAGEL W/ FRUIT OTHER THAN RAISINS, TOASTED","Bagel, with fruit other than raisins, toasted"
+51182010,"BREAD, STUFFING (INCLUDE HOMEMADE; STUFFING, NFS)","Bread stuffing"
+51182020,"BREAD STUFFING W/ EGG","Bread stuffing made with egg"
+51184000,"BREAD STICK, HARD","Bread sticks, hard"
+51184010,"BREAD STICK, SOFT","Bread stick, soft"
+51184020,"BREAD STICK, NS AS TO HARD OR SOFT","Bread stick, NS as to hard or soft"
+51184030,"BREAD STICK, SOFT, PREP W/ GARLIC & PARMESAN CHEESE","Bread stick, soft, prepared with garlic and parmesan cheese"
+51184100,"BREAD STICK, HARD, LOW SODIUM","Bread stick, hard, low sodium"
+51185000,"CROUTONS","Croutons"
+51186010,"MUFFIN, ENGLISH (INCLUDE SOUR DOUGH)","Muffin, English"
+51186020,"MUFFIN, ENGLISH, TOASTED","Muffin, English, toasted"
+51186100,"MUFFIN, ENGLISH, W/ RAISINS","Muffin, English, with raisins"
+51186120,"MUFFIN, ENGLISH, W/ RAISINS, TOASTED","Muffin, English, with raisins, toasted"
+51186130,"MUFFIN, ENGLISH, CHEESE","Muffin, English, cheese"
+51186140,"MUFFIN, ENGLISH, CHEESE, TOASTED","Muffin, English, cheese, toasted"
+51186160,"MUFFIN, ENGLISH, W/ FRUIT OTHER THAN RAISINS","Muffin, English, with fruit other than raisins"
+51186180,"MUFFIN, ENGLISH, W/ FRUIT OTHER THAN RAISINS, TSTD","Muffin, English, with fruit other than raisins, toasted"
+51187000,"MELBA TOAST","Melba toast"
+51187020,"ANISETTE TOAST","Anisette toast"
+51188100,"PANNETONE (ITALIAN-STYLE SWEET BREAD)","Pannetone (Italian-style sweet bread)"
+51188500,"ZWIEBACK TOAST (INCL RUSK)","Zwieback toast"
+51201010,"BREAD, 100% WHOLE WHEAT","Bread, whole wheat, 100%"
+51201020,"BREAD, 100% WHOLE WHEAT, TOASTED","Bread, whole wheat, 100%, toasted"
+51201060,"BREAD, 100% WHOLE WHEAT, HOME-MADE","Bread, whole wheat, 100%, made from home recipe or purchased at bakery"
+51201070,"BREAD, 100% WHOLE WHEAT, HOME-MADE, TOASTED","Bread, whole wheat, 100%, made from home recipe or purchased at bakery, toasted"
+51201150,"BREAD, PITA, 100%  WHOLE WHEAT","Bread, pita, whole wheat, 100%"
+51201160,"BREAD, PITA, 100% WHOLE WHEAT, TOASTED","Bread, pita, whole wheat, 100%, toasted"
+51202000,"MUFFIN, ENGLISH, 100% WHOLE WHEAT","Muffin, English, whole wheat, 100%"
+51202020,"MUFFIN, ENGLISH, 100% WHOLE WHEAT, TOASTED","Muffin, English, whole wheat, 100%, toasted"
+51202050,"MUFFIN, ENGLISH, 100% WHOLE WHEAT, W/ RAISINS","Muffin, English, whole wheat, 100%, with raisins"
+51202060,"MUFFIN, ENGLISH, WHOLE WHEAT, W/ RAISINS, TOASTED","Muffin, English, whole wheat, 100%, with raisins, toasted"
+51207010,"BREAD, SPROUTED WHEAT","Bread, sprouted wheat"
+51207020,"BREAD, SPROUTED WHEAT, TOASTED","Bread, sprouted wheat, toasted"
+51208000,"BAGEL, 100% WHOLE WHEAT","Bagel, whole wheat, 100%"
+51208010,"BAGEL, 100% WHOLE WHEAT, TOASTED","Bagel, whole wheat, 100%, toasted"
+51208100,"BAGEL, 100% WHOLE WHEAT, W/ RAISINS","Bagel, whole wheat, 100%, with raisins"
+51208110,"BAGEL, 100% WHOLE WHEAT, W/ RAISINS, TOASTED","Bagel, whole wheat, 100%, with raisins, toasted"
+51220000,"ROLL, 100% WHOLE WHEAT","Roll, whole wheat, 100%"
+51220010,"ROLL, 100% WHOLE WHEAT, TOASTED","Roll, whole wheat, 100%, toasted"
+51220030,"ROLL, 100% WHOLE WHEAT, HOME RECIPE/BAKERY","Roll, whole wheat, 100%, made from home recipe or purchased at bakery"
+51220040,"ROLL, 100% WHOLE WHEAT, HOME RECIPE/BAKERY, TOASTED","Roll, whole wheat, 100%, made from home recipe or purchased at bakery, toasted"
+51300050,"BREAD, WHOLE GRAIN WHITE","Bread, whole grain white"
+51300060,"BREAD, WHOLE GRAIN WHITE, TOASTED","Bread, whole grain white, toasted"
+51300100,"BAGEL, WHOLE GRAIN WHITE","Bagel, whole grain white"
+51300110,"BREAD, WHOLE WHEAT, NS AS TO 100%","Bread, whole wheat, NS as to 100%"
+51300120,"BREAD, WHOLE WHEAT, NS AS TO 100%, TOASTED","Bread, whole wheat, NS as to 100%, toasted"
+51300140,"BREAD, WHOLE WHEAT, NS AS TO 100%, MADE FROM HOME RECIPE OR","Bread, whole wheat, NS as to 100%, made from home recipe or purchased at bakery"
+51300150,"BREAD, WHOLE WHEAT, NS 100%, HOME RECIPE/BAKERY, TOASTED","Bread, whole wheat, NS as to 100%, made from home recipe or purchased at bakery, toasted"
+51300175,"BREAD, CHAPPATTI OR ROTI (INDIAN BREAD), WHEAT","Bread, chappatti or roti (Indian bread), wheat"
+51300180,"BREAD, PURI OR POORI (INDIAN PUFFED BREAD), WHEAT","Bread, puri or poori (Indian puffed bread), wheat"
+51300185,"BREAD, PARATHA, (INDIAN FLAT BREAD), WHEAT","Bread, paratha, (Indian flat bread), wheat"
+51300210,"BREAD, WHOLE WHEAT,  WITH RAISINS","Bread, whole wheat, with raisins"
+51300220,"BREAD, WHOLE WHEAT, WITH RAISINS, TOASTED","Bread, whole wheat, with raisins, toasted"
+51301010,"BREAD, WHEAT OR CRACKED WHEAT","Bread, wheat or cracked wheat"
+51301020,"BREAD, WHEAT OR CRACKED WHEAT, TOASTED","Bread, wheat or cracked wheat, toasted"
+51301040,"BREAD, CRACKED WHEAT, HOME RECIPE/BAKERY","Bread, wheat or cracked wheat, made from home recipe or purchased at bakery"
+51301050,"BREAD, CRACKED WHEAT, HOME RECIPE/BAKERY, TOASTED","Bread, wheat or cracked wheat, made from home recipe or purchased at bakery, toasted"
+51301120,"BREAD, WHEAT OR CRACKED WHEAT, W/ RAISINS","Bread, wheat or cracked wheat, with raisins"
+51301130,"BREAD, WHEAT OR CRACKED WHEAT, W/ RAISINS, TOASTED","Bread, wheat or cracked wheat, with raisins, toasted"
+51301510,"BREAD, CRACKED WHEAT, REDUCED CALORIE/ HIGH FIBER","Bread, wheat or cracked wheat, reduced calorie and/or high fiber"
+51301520,"BREAD, CRACKED WHEAT, RED CALORIE/ HI FIBER, TOAST","Bread, wheat or cracked wheat, reduced calorie and/or high fiber, toasted"
+51301540,"BREAD, FRENCH OR VIENNA, WHOLE WHEAT, NS AS TO 100%","Bread, French or Vienna, whole wheat, NS as to 100%"
+51301550,"BREAD, FRENCH OR VIENNA, WHOLE WHEAT, NS AS TO 100%, TOASTED","Bread, French or Vienna, whole wheat, NS as to 100%, toasted"
+51301600,"BREAD, PITA, WHOLE WHEAT, NS AS TO 100%","Bread, pita, whole wheat, NS as to 100%"
+51301610,"BREAD, PITA, WHOLE WHEAT, NS AS TO 100%, TOASTED","Bread, pita, whole wheat, NS as to 100%, toasted"
+51301620,"BREAD, PITA, CRACKED WHEAT","Bread, pita, wheat or cracked wheat"
+51301630,"BREAD, PITA, CRACKED WHEAT, TOASTED","Bread, pita, wheat or cracked wheat, toasted"
+51301700,"BAGEL, WHEAT","Bagel, wheat"
+51301710,"BAGEL, WHEAT, TOASTED","Bagel, wheat, toasted"
+51301750,"BAGEL, WHOLE WHEAT, NS AS TO 100%","Bagel, whole wheat, NS as to 100%"
+51301760,"BAGEL, WHOLE WHEAT, NS AS TO 100%, TOASTED","Bagel, whole wheat, NS as to 100%, toasted"
+51301800,"BAGEL, WHEAT, W/ RAISINS","Bagel, wheat, with raisins"
+51301810,"BAGEL, WHEAT, W/ RAISINS, TOASTED","Bagel, wheat, with raisins, toasted"
+51301820,"BAGEL, WHEAT, W/ FRUITS & NUTS","Bagel, wheat, with fruit and nuts"
+51301830,"BAGEL, WHEAT, W/ FRUITS & NUTS, TOASTED","Bagel, wheat, with fruit and nuts, toasted"
+51301900,"BAGEL, WHEAT BRAN","Bagel, wheat bran"
+51301910,"BAGEL, WHEAT BRAN, TOASTED","Bagel, wheat bran, toasted"
+51302500,"MUFFIN, ENGLISH, WHEAT BRAN","Muffin, English, wheat bran"
+51302510,"MUFFIN, ENGLISH, WHEAT BRAN, TOASTED","Muffin, English, wheat bran, toasted"
+51302520,"MUFFIN, ENGLISH, WHEAT BRAN, W/ RAISINS","Muffin, English, wheat bran, with raisins"
+51302530,"MUFFIN, ENGLISH, WHEAT BRAN, W/ RAISINS, TOASTED","Muffin, English, wheat bran, with raisins, toasted"
+51303010,"MUFFIN, ENGLISH, WHEAT OR CRACKED WHEAT","Muffin, English, wheat or cracked wheat"
+51303020,"MUFFIN, ENGLISH, WHEAT OR CRACKED WHEAT, TOASTED","Muffin, English, wheat or cracked wheat, toasted"
+51303030,"MUFFIN, ENGLISH, WHOLE WHEAT, NS AS TO 100%","Muffin, English, whole wheat, NS as to 100%"
+51303040,"MUFFIN, ENGLISH, WHOLE WHEAT, NS AS TO 100%, TOASTED","Muffin, English, whole wheat, NS as to 100%, toasted"
+51303050,"MUFFIN, ENGLISH, WHEAT OR CRACKED WHEAT W/ RAISINS","Muffin, English, wheat or cracked wheat, with raisins"
+51303060,"MUFFIN, ENGLISH, WHEAT W/ RAISINS, TOASTED","Muffin, English, wheat or cracked wheat, with raisins, toasted"
+51303070,"MUFFIN, ENGLISH, WHOLE WHEAT, NS AS TO 100%, WITH RAISINS","Muffin, English, whole wheat, NS as to 100%, with raisins"
+51303080,"MUFFIN, ENGLISH, WHOLE WHEAT, NS 100%, W/RAISINS, TOASTED","Muffin, English, whole wheat, NS as to 100%, with raisins, toasted"
+51306000,"BREAD STICK, HARD, WHOLE WHEAT, NS AS TO 100 %","Bread stick, hard, whole wheat, NS as to 100%"
+51320010,"ROLL, WHEAT OR CRACKED WHEAT","Roll, wheat or cracked wheat"
+51320020,"ROLL, WHEAT OR CRACKED WHEAT, TOASTED","Roll, wheat or cracked wheat, toasted"
+51320040,"ROLL, CRACKED WHEAT, HOME RECIPE/BAKERY","Roll, wheat or cracked wheat, made from home recipe or purchased at bakery"
+51320050,"ROLL, CRACKED WHEAT, HOME RECIPE/BAKERY, TOASTED","Roll, wheat or cracked wheat, made from home recipe or purchased at bakery, toasted"
+51320500,"ROLL, WHOLE WHEAT, NS AS TO 100%","Roll, whole wheat, NS as to 100%"
+51320510,"ROLL, WHOLE WHEAT, NS AS TO 100%, TOASTED","Roll, whole wheat, NS as to 100%, toasted"
+51320530,"ROLL, WHOLE WHEAT, NS 100%, MADE FROM HOMEMADE/BAKERY","Roll, whole wheat, NS as to 100%, made from home recipe or purchased at bakery"
+51320540,"ROLL, WHOLE WHEAT, NS AS TO 100%, HOMEMADE/BAKERY, TOASTED","Roll, whole wheat, NS as to 100%, made from home recipe or purchased at bakery, toasted"
+51401010,"BREAD, RYE","Bread, rye"
+51401020,"BREAD, RYE, TOASTED","Bread, rye, toasted"
+51401030,"BREAD, MARBLE RYE & PUMPERNICKEL","Bread, marble rye and pumpernickel"
+51401040,"BREAD, MARBLE RYE & PUMPERNICKEL, TOASTED","Bread, marble rye and pumpernickel, toasted"
+51401060,"BREAD, RYE, REDUCED CALORIE/ HIGH FIBER (INCL LESS)","Bread, rye, reduced calorie and/or high fiber"
+51401070,"BREAD, RYE, REDUCED CALORIE/ HIGH FIBER, TOASTED","Bread, rye, reduced calorie and/or high fiber, toasted"
+51401200,"MUFFIN, ENGLISH, RYE","Muffin, English, rye"
+51401210,"MUFFIN, ENGLISH, RYE, TOASTED","Muffin, English, rye, toasted"
+51404010,"BREAD, PUMPERNICKEL","Bread, pumpernickel"
+51404020,"BREAD, PUMPERNICKEL, TOASTED","Bread, pumpernickel, toasted"
+51404500,"BAGEL, PUMPERNICKEL","Bagel, pumpernickel"
+51404510,"BAGEL, PUMPERNICKEL, TOASTED","Bagel, pumpernickel, toasted"
+51404550,"MUFFIN, ENGLISH, PUMPERNICKEL","Muffin, English, pumpernickel"
+51404560,"MUFFIN, ENGLISH, PUMPERNICKEL, TOASTED","Muffin, English, pumpernickel, toasted"
+51407010,"BREAD, BLACK","Bread, black"
+51407020,"BREAD, BLACK, TOASTED","Bread, black, toasted"
+51420000,"ROLL, RYE","Roll, rye"
+51421000,"ROLL, PUMPERNICKEL","Roll, pumpernickel"
+51421100,"ROLL, PUMPERNICKEL, TOASTED","Roll, pumpernickel, toasted"
+51501010,"BREAD, OATMEAL","Bread, oatmeal"
+51501020,"BREAD, OATMEAL, TOASTED","Bread, oatmeal, toasted"
+51501040,"BREAD, OAT BRAN","Bread, oat bran"
+51501050,"BREAD, OAT BRAN, TOASTED","Bread, oat bran, toasted"
+51501060,"BREAD, OAT BRAN, REDUCED CALORIE/ HIGH FIBER","Bread, oat bran, reduced calorie and/or high fiber"
+51501070,"BREAD, OAT BRAN REDUCED CALORIE/HI FIBER, TOASTED","Bread, oat bran, reduced calorie and/or high fiber, toasted"
+51501080,"BAGEL, OAT BRAN","Bagel, oat bran"
+51501090,"BAGEL, OAT BRAN, TOASTED","Bagel, oat bran, toasted"
+51502010,"ROLL, OATMEAL","Roll, oatmeal"
+51502020,"ROLL, OATMEAL, TOASTED","Roll, oatmeal, toasted"
+51502100,"ROLL, OAT BRAN","Roll, oat bran"
+51502110,"ROLL, OAT BRAN, TOASTED","Roll, oat bran, toasted"
+51503000,"MUFFIN, ENGLISH, OAT BRAN","Muffin, English, oat bran"
+51503010,"MUFFIN, ENGLISH, OAT BRAN, TOASTED","Muffin, English, oat bran, toasted"
+51503040,"MUFFIN, ENGLISH, OAT BRAN, WITH RAISINS","Muffin, English, oat bran, with raisins"
+51503050,"MUFFIN, ENGLISH, OAT BRAN, W/ RAISINS, TOASTED","Muffin, English, oat bran with raisins, toasted"
+51601010,"BREAD, MULTIGRAIN, TOASTED","Bread, multigrain, toasted"
+51601020,"BREAD, MULTIGRAIN","Bread, multigrain"
+51601210,"BREAD, MULTIGRAIN, W/ RAISINS","Bread, multigrain, with raisins"
+51601220,"BREAD, MULTIGRAIN, W/ RAISINS, TOASTED","Bread, multigrain, with raisins, toasted"
+51602010,"BREAD, MULTIGRAIN, REDUCED CALORIE/ HIGH FIBER","Bread, multigrain, reduced calorie and/or high fiber"
+51602020,"BREAD, MULTIGRAIN, REDUCED CALORIE/ HI FIBER, TOAST","Bread, multigrain, reduced calorie and/or high fiber, toasted"
+51620000,"ROLL, MULTIGRAIN","Roll, multigrain"
+51620010,"ROLL, MULTIGRAIN, TOASTED","Roll, multigrain, toasted"
+51630000,"BAGEL, MULTIGRAIN","Bagel, multigrain"
+51630010,"BAGEL, MULTIGRAIN, TOASTED","Bagel, multigrain, toasted"
+51630100,"BAGEL, MULTIGRAIN, W/ RAISINS","Bagel, multigrain, with raisins"
+51630110,"BAGEL, MULTIGRAIN, W/ RAISINS, TOASTED","Bagel, multigrain, with raisins, toasted"
+51630200,"MUFFIN, ENGLISH, MULTIGRAIN","Muffin, English, multigrain"
+51630210,"MUFFIN, ENGLISH, MULTIGRAIN, TOASTED","Muffin, English, multigrain, toasted"
+51801010,"BREAD, BARLEY","Bread, barley"
+51801020,"BREAD, BARLEY, TOASTED","Bread, barley, toasted"
+51804010,"BREAD, SOY","Bread, soy"
+51804020,"BREAD, SOY, TOASTED","Bread, soy, toasted"
+51805010,"BREAD, SUNFLOWER MEAL","Bread, sunflower meal"
+51805020,"BREAD, SUNFLOWER MEAL, TOASTED","Bread, sunflower meal, toasted"
+51806010,"BREAD, RICE","Bread, rice"
+51806020,"BREAD, RICE, TOASTED","Bread, rice, toasted"
+51807000,"INJERA (AMERICAN-STYLE ETHIOPIAN BREAD)","Injera (American-style Ethiopian bread)"
+51808000,"BREAD, LOW GLUTEN","Bread, low gluten"
+51808010,"BREAD, LOW GLUTEN, TOASTED","Bread, low gluten, toasted"
+52101000,"BISCUIT, BAKING POWDER OR BUTTERMILK TYPE, NFS","Biscuit, baking powder or buttermilk type, NS as to made from mix, refrigerated dough, or home recipe"
+52101030,"BISCUIT DOUGH, FRIED","Biscuit dough, fried"
+52101040,"CRUMPET","Crumpet"
+52101050,"CRUMPET, TOASTED","Crumpet, toasted"
+52101100,"BISCUIT, BAKING POWDER OR BUTTERMILK, FROM MIX","Biscuit, baking powder or buttermilk type, made from mix"
+52101150,"BISCUIT,BAKING PWR/BUTTER MILK,REFRIG DOUGH,LOWFAT","Biscuit, baking powder or buttermilk type, made from refrigerated dough, lowfat"
+52102040,"BISCUIT, BAK POWDER OR BUTTERMILK, FROM REFRG DOUGH","Biscuit, baking powder or buttermilk type, made from refrigerated dough"
+52103000,"BISCUIT, BAKING POWDER/BUTTERMILK TYPE, COMMERCIALLY BAKED","Biscuit, baking powder or buttermilk type, commercially baked"
+52104010,"BISCUIT, BAKING POWDER OR BUTTERMILK, HOMEMADE","Biscuit, baking powder or buttermilk type, made from home recipe"
+52104040,"BISCUIT, WHOLE WHEAT","Biscuit, whole wheat"
+52104100,"BISCUIT, CHEESE","Biscuit, cheese"
+52104200,"BISCUIT, CINNAMON-RAISIN","Biscuit, cinnamon-raisin"
+52105100,"SCONES","Scone"
+52105110,"SCONES, WHOLE WHEAT","Scone, whole wheat"
+52105200,"SCONE, WITH FRUIT","Scone, with fruit"
+52201000,"CORNBREAD, PREPARED FROM MIX","Cornbread, prepared from mix"
+52202060,"CORNBREAD, HOMEMADE","Cornbread, made from home recipe"
+52204000,"CORNBREAD STUFFING","Cornbread stuffing"
+52206010,"CORNBREAD MUFFIN, STICK, ROUND","Cornbread muffin, stick, round"
+52206060,"CORNBREAD MUFFIN, STICK, ROUND, HOMEMADE","Cornbread muffin, stick, round, made from home recipe"
+52207010,"CORN FLOUR PATTIES OR TARTS, FRIED","Corn flour patty or tart, fried"
+52208010,"CORN PONE,BAKED (INCL HOE CAKE)","Corn pone, baked"
+52208020,"CORN PONE FRIED","Corn pone, fried"
+52208760,"GORDITA/SOPE SHELL, PLAIN, NO FILLING","Gordita/sope shell, plain, no filling"
+52209010,"HUSH PUPPY","Hush puppy"
+52211010,"JOHNNYCAKE","Johnnycake"
+52213010,"SPOONBREAD","Spoonbread"
+52215000,"TORTILLA, NFS","Tortilla, NFS"
+52215100,"TORTILLA, CORN","Tortilla, corn"
+52215200,"TORTILLA, FLOUR (WHEAT)","Tortilla, flour (wheat)"
+52215260,"TORTILLA, WHOLE WHEAT","Tortilla, whole wheat"
+52215300,"TACO SHELL, CORN","Taco shell, corn"
+52215350,"TACO SHELL; FLOUR","Taco shell, flour"
+52220110,"CORNMEAL BREAD, DOMINICAN","Cornmeal bread, Dominican style (Arepa Dominicana)"
+52301000,"MUFFIN, NFS","Muffin, NFS"
+52302010,"MUFFIN, FRUIT","Muffin, fruit"
+52302020,"MUFFIN, FRUIT, LOW FAT","Muffin, fruit, low fat"
+52302500,"MUFFIN, CHOCOLATE CHIP","Muffin, chocolate chip"
+52302600,"MUFFIN, CHOCOLATE","Muffin, chocolate"
+52303010,"MUFFIN, WHOLE WHEAT","Muffin, whole wheat"
+52303500,"MUFFIN, WHEAT","Muffin, wheat"
+52304000,"MUFFIN, WHOLE GRAIN","Muffin, whole grain"
+52304010,"MUFFIN, WHEAT BRAN (INCLUDE W/ RAISINS & NUTS)","Muffin, wheat bran"
+52304040,"MUFFIN,BRAN,W/ FRUIT, LOWFAT","Muffin, bran with fruit, lowfat"
+52304100,"MUFFIN, OATMEAL","Muffin, oatmeal"
+52304150,"MUFFIN, OAT BRAN","Muffin, oat bran"
+52306010,"MUFFIN, PLAIN","Muffin, plain"
+52306300,"MUFFIN, CHEESE","Muffin, cheese"
+52306500,"MUFFIN, PUMPKIN, W/ RAISINS","Muffin, pumpkin"
+52306550,"MUFFIN, ZUCCHINI","Muffin, zucchini"
+52306700,"MUFFIN, CARROT (INCL W/ RAISINS/NUTS)","Muffin, carrot"
+52311010,"POPOVER","Popover"
+52401000,"BREAD, BOSTON BROWN","Bread, Boston Brown"
+52403000,"BREAD, NUT","Bread, nut"
+52404060,"BREAD, PUMPKIN (INCLUDE W/ RAISINS)","Bread, pumpkin"
+52405010,"BREAD, FRUIT","Bread, fruit"
+52407000,"BREAD, ZUCCHINI (INCL SQUASH BREAD; W/ NUTS)","Bread, zucchini"
+52408000,"BREAD, IRISH SODA","Bread, Irish soda"
+53100050,"CAKE, BATTER, CHOCOLATE, RAW","Cake batter, raw, chocolate"
+53100070,"CAKE, BATTER, RAW, NOT CHOCOLATE","Cake batter, raw, not chocolate"
+53100100,"CAKE OR CUPCAKE, NS AS TO TYPE","Cake or cupcake, NS as to type"
+53101100,"CAKE, ANGEL FOOD, W/O ICING","Cake, angel food, without icing or filling"
+53101200,"CAKE, ANGEL FOOD, W/ ICING","Cake, angel food, with icing or filling"
+53101250,"CAKE, ANGEL FOOD, W/ FRUIT & ICING/FILLING","Cake, angel food, with fruit and icing or filling"
+53102100,"CAKE OR CUPCAKE,APPLESAUCE W/O ICING","Cake or cupcake, applesauce, without icing or filling"
+53102200,"CAKE OR CUPCAKE,APPLESAUCE W/ ICING","Cake or cupcake, applesauce, with icing or filling"
+53102600,"CAKE OR CUPCAKE,BANANA, W/O ICING","Cake or cupcake, banana, without icing or filling"
+53102700,"CAKE OR CUPCAKE,BANANA, W/ ICING","Cake or cupcake, banana, with icing or filling"
+53102800,"CAKE OR CUPCAKE,BLACK FOREST (CHOC-CHERRY)","Cake or cupcake, black forest (chocolate-cherry)"
+53103000,"CAKE, BOSTON CREAM PIE","Cake, Boston cream pie"
+53104100,"CAKE OR CUPCAKE,CARROT, NO ICING","Cake or cupcake, carrot, without icing or filling"
+53104260,"CAKE OR CUPCAKE, CARROT, WITH ICING","Cake or cupcake, carrot, with icing or filling"
+53104300,"CARROT CAKE, DIET","Cake, carrot, diet"
+53104400,"CAKE OR CUPCAKE,COCONUT, W/ ICING","Cake or cupcake, coconut, with icing or filling"
+53104500,"CHEESECAKE","Cheesecake"
+53104550,"CHEESECAKE, W/ FRUIT","Cheesecake with fruit"
+53104600,"CHEESECAKE, CHOCOLATE","Cheesecake, chocolate"
+53105270,"CAKE OR CUPCAKE, CHOC, DEVIL'S FOOD OR FUDGE, W/ICING /FILL","Cake or cupcake, chocolate, devil's food or fudge, with icing or filling"
+53105275,"CAKE OR CUPCAKE, CHOCOLATE, DEVIL'S FOOD OR FUDGE, W/O ICING","Cake or cupcake, chocolate, devil's food or fudge, without icing or filling"
+53105300,"CAKE OR CUPCAKE,GERMAN CHOC, W/ ICING or FILLING","Cake or cupcake, German chocolate, with icing or filling"
+53105500,"CAKE, CHOC, W/ ICING, DIET","Cake, chocolate, with icing, diet"
+53106500,"CAKE, CREAM, W/O ICING OR TOPPING","Cake, cream, without icing or topping"
+53108200,"SNACK CAKE, CHOCOLATE, WITH ICING OR FILLING","Snack cake, chocolate, with icing or filling"
+53108220,"SNACK CAKE, CHOC, W/ICING OR FILLING, REDUCED FAT&CALORIE","Snack cake, chocolate, with icing or filling, reduced fat and calories"
+53109200,"SNACK CAKE, NOT CHOCOLATE, WITH ICING OR FILLING","Snack cake, not chocolate, with icing or filling"
+53109220,"SNACK CAKE, NOT CHOC, W/ ICING OR FILLING, RED FAT&CALS","Snack cake, not chocolate, with icing or filling, reduced fat and calories"
+53109300,"CAKE,DOBOS TORTE(NON-CHOC CAKE W/CHOC FILL & ICING)","Cake, Dobos Torte (non-chocolate layer cake with chocolate filling and icing)"
+53110000,"CAKE, FRUITCAKE, LIGHT/DARK, HOLIDAY TYPE CAKE","Cake, fruit cake, light or dark, holiday type cake"
+53111000,"CAKE OR CUPCAKE, GINGERBREAD","Cake or cupcake, gingerbread"
+53112000,"CAKE, ICE CREAM & CAKE ROLL, CHOCOLATE","Cake, ice cream and cake roll, chocolate"
+53112100,"CAKE, ICE CREAM & CAKE ROLL, NOT CHOCOLATE","Cake, ice cream and cake roll, not chocolate"
+53113000,"CAKE, JELLY ROLL","Cake, jelly roll"
+53114000,"CAKE OR CUPCAKE,LEMON, W/O ICING","Cake or cupcake, lemon, without icing or filling"
+53114100,"CAKE OR CUPCAKE,LEMON, W/ ICING","Cake or cupcake, lemon, with icing or filling"
+53115100,"CAKE OR CUPCAKE, MARBLE, W/O ICING OR FILLING","Cake or cupcake, marble, without icing or filling"
+53115200,"CAKE OR CUPCAKE, MARBLE, WITH ICING OR FILLING","Cake or cupcake, marble, with icing or filling"
+53115310,"CAKE OR CUPCAKE,NUT, W/O ICING","Cake or cupcake, nut, without icing or filling"
+53115320,"CAKE OR CUPCAKE,NUT, W/ ICING","Cake or cupcake, nut, with icing or filling"
+53115410,"CAKE OR CUPCAKE, OATMEAL","Cake or cupcake, oatmeal"
+53115450,"CAKE OR CUPCAKE, PEANUT BUTTER","Cake or cupcake, peanut butter"
+53116000,"CAKE, POUND, W/O ICING","Cake, pound, without icing or filling"
+53116020,"CAKE, POUND, W/ ICING","Cake, pound, with icing or filling"
+53116270,"CAKE, POUND, CHOCOLATE","Cake, pound, chocolate"
+53116350,"CAKE, POUND, P.R. (PONQUE)","Cake, pound, Puerto Rican style (Ponque)"
+53116390,"CAKE, POUND, REDUCED FAT, NO CHOLESTEROL","Cake, pound, reduced fat, cholesterol free"
+53116500,"CAKE OR CUPCAKE,PUMPKIN, W/O ICING","Cake or cupcake, pumpkin, without icing or filling"
+53116510,"CAKE OR CUPCAKE,PUMPKIN,W/ ICING","Cake or cupcake, pumpkin, with icing or filling"
+53116550,"CAKE OR CUPCAKE, RAISIN-NUT","Cake or cupcake, raisin-nut"
+53116570,"CAKE, RAVANI (MADE W/ FARINA)","Cake, Ravani (made with farina)"
+53116600,"CAKE, RICE FLOUR, W/O ICING","Cake, rice flour, without icing or filling"
+53116650,"CAKE, QUEZADILLA, EL SALVADORIAN STYLE","Cake, Quezadilla, El Salvadorian style"
+53117100,"CAKE OR CUPCAKE,SPICE, W/O ICING","Cake or cupcake, spice, without icing or filling"
+53117200,"CAKE OR CUPCAKE,SPICE, W/ ICING","Cake or cupcake, spice, with icing or filling"
+53118100,"CAKE, SPONGE, W/O ICING","Cake, sponge, without icing or filling"
+53118200,"CAKE, SPONGE, W/ ICING","Cake, sponge, with icing or filling"
+53118300,"CAKE, SPONGE, CHOCOLATE","Cake, sponge, chocolate"
+53118410,"RUM CAKE, WITHOUT ICING (SOPA BORRACHA)","Rum cake, without icing (Sopa Borracha)"
+53118500,"CAKE, TORTE","Cake, torte"
+53118550,"CAKE, TRES LECHE","Cake, tres leche"
+53119000,"CAKE, UPSIDE DOWN (ALL FRUITS)","Cake, upside down (all fruits)"
+53120270,"CAKE OR CUPCAKE, WHITE, WITH ICING OR FILLING","Cake or cupcake, white, with icing or filling"
+53120275,"CAKE OR CUPCAKE, WHITE, WITHOUT ICING OR FILLING","Cake or cupcake, white, without icing or filling"
+53121270,"CAKE OR CUPCAKE, YELLOW, WITH ICING OR FILLING","Cake or cupcake, yellow, with icing or filling"
+53121275,"CAKE OR CUPCAKE, YELLOW, WITHOUT ICING OR FILLING","Cake or cupcake, yellow, without icing or filling"
+53122070,"CAKE, SHORTCAKE, BISCUIT, W/ WHIPPED CREAM & FRUIT","Cake, shortcake, biscuit type, with whipped cream and fruit"
+53122080,"CAKE, SHORTCAKE, BISCUIT, W/ FRUIT","Cake, shortcake, biscuit type, with fruit"
+53123070,"CAKE, SHORTCAKE, SPONGE, W/ WHIPPED CREAM & FRUIT","Cake, shortcake, sponge type, with whipped cream and fruit"
+53123080,"CAKE, SHORTCAKE, SPONGE, W/ FRUIT","Cake, shortcake, sponge type, with fruit"
+53123500,"CAKE, SHORTCAKE, W/ WHIP TOPPING & FRUIT, DIET","Cake, shortcake, with whipped topping and fruit, diet"
+53124110,"CAKE OR CUPCAKE, ZUCCHINI","Cake or cupcake, zucchini"
+53200100,"COOKIE, BATTER OR DOUGH, RAW","Cookie, batter or dough, raw"
+53201000,"COOKIE, NFS","Cookie, NFS"
+53202000,"COOKIE, ALMOND","Cookie, almond"
+53203000,"COOKIE, APPLESAUCE","Cookie, applesauce"
+53203500,"COOKIE, BISCOTTI","Cookie, biscotti (Italian sugar cookie)"
+53204000,"COOKIE, BROWNIE, NS AS TO ICING","Cookie, brownie, NS as to icing"
+53204010,"COOKIE, BROWNIE, W/O ICING","Cookie, brownie, without icing"
+53204100,"COOKIE, BROWNIE, WITH ICING OR FILLING","Cookie, brownie, with icing or filling"
+53204840,"COOKIE, BROWNIE, REDUCED FAT, NS AS TO ICING","Cookie, brownie, reduced fat, NS as to icing"
+53204860,"COOKIE, BROWNIE, FAT FREE, NS AS TO ICING","Cookie, brownie, fat free, NS as to icing"
+53205250,"COOKIE, BUTTERSCOTCH, BROWNIE","Cookie, butterscotch, brownie"
+53205260,"COOKIE, BAR, WITH CHOCOLATE","Cookie, bar, with chocolate"
+53206000,"COOKIE, CHOCOLATE CHIP","Cookie, chocolate chip"
+53206020,"COOKIE, CHOC CHIP, HOMEMADE OR PURCHASED AT BAKERY","Cookie, chocolate chip, made from home recipe or purchased at a bakery"
+53206030,"COOKIE, CHOC CHIP, REDUCED FAT","Cookie, chocolate chip, reduced fat"
+53206100,"COOKIE, CHOCOLATE CHIP SANDWICH","Cookie, chocolate chip sandwich"
+53206500,"COOKIE, CHOCOLATE, MADE WITH RICE CEREAL","Cookie, chocolate, made with rice cereal"
+53206550,"COOKIE, CHOCOLATE, MADE W/ OATMEAL & COCONUT","Cookie, chocolate, made with oatmeal and coconut (no-bake)"
+53207000,"COOKIE, CHOCOLATE FUDGE","Cookie, chocolate or fudge"
+53207020,"COOKIE, CHOCOLATE OR FUDGE, REDUCED FAT","Cookie, chocolate or fudge, reduced fat"
+53207050,"COOKIE, CHOCOLATE, W/ CHOC FILLING/COATING, FAT FREE","Cookie, chocolate, with chocolate filling or coating, fat free"
+53208000,"COOKIE, MARSHMALLOW, CHOCOLATE-COVERED","Cookie, marshmallow, chocolate-covered"
+53208200,"COOKIE, CHOCOLATE-COVERED, MARSHMALLOW PIE","Cookie, marshmallow pie, chocolate covered"
+53209005,"COOKIE, CHOCOLATE, WITH ICING OR COATING","Cookie, chocolate, with icing or coating"
+53209010,"COOKIE, SUGAR WAFER, CHOCOLATE-COVERED","Cookie, sugar wafer, chocolate-covered"
+53209015,"COOKIE, CHOCOLATE SANDWICH","Cookie, chocolate sandwich"
+53209020,"COOKIE, CHOCOLATE SANDWICH, REDUCED FAT","Cookie, chocolate sandwich, reduced fat"
+53209100,"COOKIE, CHOCOLATE, SANDWICH, W/ EXTRA FILLING","Cookie, chocolate, sandwich, with extra filling"
+53209500,"COOKIE, CHOCOLATE & VANILLA SANDWICH","Cookie, chocolate and vanilla sandwich"
+53210000,"COOKIE, CHOCOLATE WAFER","Cookie, chocolate wafer"
+53210900,"COOKIE, GRAHAM CRACKER WITH CHOCOLATE AND MARSHMALLOW","Cookie, graham cracker with chocolate and marshmallow"
+53211000,"COOKIE, BAR, W/ CHOCOLATE, NUTS, & GRAHAM CRACKERS","Cookie bar, with chocolate, nuts, and graham crackers"
+53215500,"COOKIE, COCONUT","Cookie, coconut"
+53220000,"COOKIE, FRUIT-FILLED","Cookie, fruit-filled bar"
+53220010,"COOKIE, FRUIT-FILLED BAR, FAT FREE","Cookie, fruit-filled bar, fat free"
+53220030,"COOKIE, FIG BAR","Cookie, fig bar"
+53220040,"COOKIE, FIG BAR, FAT FREE","Cookie, fig bar, fat free"
+53222010,"COOKIE, FORTUNE","Cookie, fortune"
+53222020,"COOKIE, CONE SHELL, ICE CREAM TYPE,WAFER / CAKE","Cookie, cone shell, ice cream type, wafer or cake"
+53223000,"COOKIE, GINGERSNAPS","Cookie, gingersnaps"
+53223100,"COOKIE, GRANOLA","Cookie, granola"
+53224000,"COOKIE, LADY FINGER","Cookie, ladyfinger"
+53224250,"COOKIE, LEMON BAR","Cookie, lemon bar"
+53225000,"COOKIE, MACAROON","Cookie, macaroon"
+53226000,"COOKIE, MARSHMALLOW, W/ COCONUT","Cookie, marshmallow, with coconut"
+53226500,"COOKIE, MARSHMALLOW, W/ RICE CEREAL (NO-BAKE)","Cookie, marshmallow, with rice cereal (no-bake)"
+53226550,"COOKIE, MARSHMALLOW, W/ RICE CEREAL & CHOC CHIPS","Cookie, marshmallow, with rice cereal and chocolate chips"
+53226600,"COOKIE, MARSHMALLOW & PEANUT BUTTER, W/ OAT CEREAL (NO-BAKE)","Cookie, marshmallow and peanut butter, with oat cereal (no-bake)"
+53228000,"COOKIE, MERINGUE","Cookie, meringue"
+53230000,"COOKIE, MOLASSES","Cookie, molasses"
+53231000,"COOKIE, LEBKUCHEN","Cookie, Lebkuchen"
+53231400,"COOKIE, MULTIGRAIN, HIGH FIBER","Cookie, multigrain, high fiber"
+53233000,"COOKIE, OATMEAL","Cookie, oatmeal"
+53233010,"COOKIE, OATMEAL, W/ RAISINS OR DATES","Cookie, oatmeal, with raisins"
+53233040,"COOKIE, OATMEAL, REDUCED FAT, NS AS TO RAISINS","Cookie, oatmeal, reduced fat, NS as to raisins"
+53233050,"COOKIE, OATMEAL SANDWICH, W/ CREME FILLING","Cookie, oatmeal sandwich, with creme filling"
+53233060,"COOKIE, OATMEAL, W/ CHOCOLATE CHIPS","Cookie, oatmeal, with chocolate chips"
+53233080,"COOKIE, OATMEAL SANDWICH, W/ PEANUT BUTTER & JELLY FILLING","Cookie, oatmeal sandwich, with peanut butter and jelly filling"
+53233100,"COOKIE,OATMEAL,W/ CHOC & PEANUT BUTTER (NO-BAKE)","Cookie, oatmeal, with chocolate and peanut butter (no-bake)"
+53234000,"COOKIE, PEANUT BUTTER (INCLUDE PB WAFER)","Cookie, peanut butter"
+53234100,"COOKIE, PEANUT BUTTER, W/ CHOCOLATE (INCL NASSAU)","Cookie, peanut butter, with chocolate"
+53234250,"COOKIE, PEANUT BUTTER W/ RICE CEREAl (NO-BAKE)","Cookie, peanut butter with rice cereal (no-bake)"
+53235000,"COOKIE, PEANUT BUTTER SANDWICH","Cookie, peanut butter sandwich"
+53235500,"COOKIE, W/ PEANUT BUTTER FILLING, CHOCOLATE-COATED","Cookie, with peanut butter filling, chocolate-coated"
+53235600,"COOKIE, PFEFFERNUSSE","Cookie, Pfeffernusse"
+53236000,"COOKIE, PIZZELLE (ITALIAN STYLE WAFER)","Cookie, pizzelle (Italian style wafer)"
+53236100,"COOKIE, PUMPKIN","Cookie, pumpkin"
+53237000,"COOKIE, RAISIN","Cookie, raisin"
+53237010,"COOKIE, RAISIN SANDWICH, CREAM-FILLED","Cookie, raisin sandwich, cream-filled"
+53237500,"COOKIE, RUM BALL (NO-BAKE)","Cookie, rum ball (no-bake)"
+53238000,"COOKIE, SANDWICH TYPE, NOT CHOCOLATE OR VANILLA","Cookie, sandwich-type, not chocolate or vanilla"
+53239000,"COOKIE, SHORTBREAD","Cookie, shortbread"
+53239010,"COOKIE, SHORTBREAD, REDUCED FAT","Cookie, shortbread, reduced fat"
+53239050,"COOKIE, SHORTBREAD, WITH ICING OR FILLING","Cookie, shortbread, with icing or filling"
+53240000,"COOKIE, ANIMAL","Cookie, animal"
+53240010,"COOKIE, ANIMAL, WITH FROSTING OR ICING","Cookie, animal, with frosting or icing"
+53241500,"COOKIE, BUTTER OR SUGAR","Cookie, butter or sugar"
+53241510,"MARIE BISCUIT","Marie biscuit"
+53241600,"COOKIE, BUTTER OR SUGAR, WITH FRUIT AND/OR NUTS","Cookie, butter or sugar, with fruit and/or nuts"
+53242000,"COOKIE, SUGAR WAFER","Cookie, sugar wafer"
+53242500,"COOKIE, TOFFEE BAR","Cookie, toffee bar"
+53243000,"COOKIE, VANILLA SANDWICH","Cookie, vanilla sandwich"
+53243010,"COOKIE, VANILLA SANDWICH, EXTRA FILLING","Cookie, vanilla sandwich, extra filling"
+53243050,"COOKIE, VANILLA SANDWICH, REDUCED FAT","Cookie, vanilla sandwich, reduced fat"
+53244010,"COOKIE, BUTTER/SUGAR, W/ CHOCOLATE ICING / FILLING","Cookie, butter or sugar, with chocolate icing or filling"
+53244020,"COOKIE, BUTTER/SUGAR, W/ ICING/FILLING OTHER THAN CHOC","Cookie, butter or sugar, with icing or filling other than chocolate"
+53246000,"COOKIE, TEA, JAPANESE","Cookie, tea, Japanese"
+53247000,"COOKIE, VANILLA WAFER, NS AS TO TYPE","Cookie, vanilla wafer"
+53247050,"COOKIE, VANILLA WAFER, REDUCED FAT","Cookie, vanilla wafer, reduced fat"
+53247500,"COOKIE, VANILLA W/ CARAMEL, COCONUT, CHOC COATING","Cookie, vanilla with caramel, coconut, and chocolate coating"
+53251100,"COOKIE, RUGELACH","Cookie, rugelach"
+53260030,"COOKIE, CHOCOLATE CHIP, SUGAR FREE","Cookie, chocolate chip, sugar free"
+53260200,"COOKIE, OATMEAL, SUGAR FREE","Cookie, oatmeal, sugar free"
+53260300,"COOKIE, SANDWICH, SUGAR FREE","Cookie, sandwich, sugar free"
+53260400,"COOKIE, SUGAR OR PLAIN, SUGAR FREE","Cookie, sugar or plain, sugar free"
+53260500,"COOKIE, SUGAR WAFER, SUGAR FREE","Cookie, sugar wafer, sugar free"
+53260600,"COOKIE, PEANUT BUTTER, SUGAR FREE","Cookie, peanut butter, sugar free"
+53270100,"COOKIE, P.R. (MANTECADITOS POLVORONES)","Cookies, Puerto Rican (Mantecaditos polvorones)"
+53300100,"PIE, NFS","Pie, NFS"
+53300170,"PIE, INDIVIDUAL SIZE OR TART, NFS","Pie, individual size or tart, NFS"
+53300180,"PIE, FRIED, NFS","Pie, fried, NFS"
+53301000,"PIE, APPLE, TWO CRUST","Pie, apple, two crust"
+53301070,"PIE, APPLE, INDIVIDUAL SIZE OR TART","Pie, apple, individual size or tart"
+53301080,"PIE, APPLE, FRIED PIE","Pie, apple, fried pie"
+53301500,"PIE, APPLE, ONE CRUST (INCL W/ CRUMB TOPPING)","Pie, apple, one crust"
+53301750,"PIE, APPLE, DIET","Pie, apple, diet"
+53302000,"PIE, APRICOT, TWO CRUST","Pie, apricot, two crust"
+53302070,"PIE, APRICOT, INDIVIDUAL SIZE OR TART","Pie, apricot, individual size or tart"
+53302080,"PIE, APRICOT, FRIED","Pie, apricot, fried pie"
+53303000,"PIE, BLACKBERRY, TWO CRUST","Pie, blackberry, two crust"
+53303070,"PIE, BLACKBERRY, INDIVIDUAL SIZE OR TART","Pie, blackberry, individual size or tart"
+53303500,"PIE, BERRY NOT BLACK,BLUE,BOYSEN,RASP....,TWO CRUST","Pie, berry, not blackberry, blueberry, boysenberry, huckleberry, raspberry, or strawberry; two crust"
+53303510,"PIE, BERRY, ONE CRUST","Pie, berry, not blackberry, blueberry, boysenberry, huckleberry, raspberry, or strawberry; one crust"
+53303570,"PIE, BERRY, INDIVIDUAL SIZE OR TART","Pie, berry, not blackberry, blueberry, boysenberry, huckleberry, raspberry, or strawberry, individual size or tart"
+53304000,"PIE, BLUEBERRY, TWO CRUST","Pie, blueberry, two crust"
+53304050,"PIE, BLUEBERRY, ONE CRUST","Pie, blueberry, one crust"
+53304070,"PIE, BLUEBERRY, INDIVIDUAL SIZE OR TART","Pie, blueberry, individual size or tart"
+53305000,"PIE, CHERRY, TWO CRUST","Pie, cherry, two crust"
+53305010,"PIE, CHERRY, ONE CRUST","Pie, cherry, one crust"
+53305070,"PIE, CHERRY, INDIVIDUAL SIZE OR TART","Pie, cherry, individual size or tart"
+53305080,"PIE, CHERRY, FRIED PIE","Pie, cherry, fried pie"
+53305700,"PIE, LEMON (NOT CREAM OR MERINGUE)","Pie, lemon (not cream or meringue)"
+53305720,"PIE, LEMON (NOT CREAM OR MERINGUE), INDIVIDUAL SIZE","Pie, lemon (not cream or meringue), individual size or tart"
+53305750,"PIE, LEMON, FRIED","Pie, lemon, fried pie"
+53306000,"PIE, MINCE, TWO CRUST","Pie, mince, two crust"
+53306070,"PIE, MINCE, INDIVIDUAL SIZE OR TART","Pie, mince, individual size or tart"
+53307000,"PIE, PEACH, TWO CRUST","Pie, peach, two crust"
+53307050,"PIE, PEACH, ONE-CRUST","Pie, peach, one crust"
+53307070,"PIE, PEACH, INDIVIDUAL SIZE OR TART","Pie, peach, individual size or tart"
+53307080,"PIE, PEACH, FRIED","Pie, peach, fried pie"
+53307500,"PIE, PEAR, TWO CRUST","Pie, pear, two crust"
+53307570,"PIE, PEAR, INDIVIDUAL SIZE OR TART","Pie, pear, individual size or tart"
+53308000,"PIE, PINEAPPLE, TWO CRUST","Pie, pineapple, two crust"
+53308070,"PIE, PINEAPPLE, INDIVIDUAL SIZE OR TART","Pie, pineapple, individual size or tart"
+53308300,"PIE, PLUM, TWO CRUST","Pie, plum, two crust"
+53308500,"PIE, PRUNE, ONE CRUST","Pie, prune, one crust"
+53309000,"PIE, RAISIN, TWO CRUST","Pie, raisin, two crust"
+53309070,"PIE, RAISIN, INDIVIDUAL SIZE OR TART","Pie, raisin, individual size or tart"
+53310000,"PIE, RASPBERRY, ONE CRUST","Pie, raspberry, one crust"
+53310050,"PIE, RASPBERRY, TWO CRUST","Pie, raspberry, two crust"
+53311000,"PIE, RHUBARB, TWO CRUST","Pie, rhubarb, two crust"
+53311050,"PIE, RHUBARB, ONE CRUST","Pie, rhubarb, one crust"
+53311070,"PIE, RHUBARB, INDIVIDUAL SIZE OR TART","Pie, rhubarb, individual size or tart"
+53312000,"PIE, STRAWBERRY, ONE CRUST","Pie, strawberry, one crust"
+53313000,"PIE, STRAWBERRY-RHUBARB, TWO CRUST","Pie, strawberry-rhubarb, two crust"
+53314000,"PIE, STRAWBERRY, INDIVIDUAL SIZE OR TART","Pie, strawberry, individual size or tart"
+53340000,"PIE, APPLE-SOUR CREAM","Pie, apple-sour cream"
+53340500,"PIE, CHERRY, W/ CREAM CHEESE & SOUR CREAM","Pie, cherry, made with cream cheese and sour cream"
+53341000,"PIE, BANANA CREAM","Pie, banana cream"
+53341070,"PIE, BANANA CREAM, INDIVIDUAL SIZE OR TART","Pie, banana cream, individual size or tart"
+53341500,"PIE, BUTTERMILK","Pie, buttermilk"
+53341750,"PIE, CHESS (INCL LEMON CHESS PIE)","Pie, chess"
+53342000,"PIE, CHOCOLATE CREAM","Pie, chocolate cream"
+53342070,"PIE, CHOCOLATE CREAM, INDIVIDUAL SIZE OR TART","Pie, chocolate cream, individual size or tart"
+53343000,"PIE, COCONUT CREAM","Pie, coconut cream"
+53343070,"PIE, COCONUT CREAM, INDIVIDUAL SIZE OR TART","Pie, coconut cream, individual size or tart"
+53344000,"PIE, CUSTARD","Pie, custard"
+53344070,"PIE, CUSTARD, INDIVIDUAL SIZE OR TART","Pie, custard, individual size or tart"
+53344200,"MIXED FRUIT TART FILLED WITH CUSTARD OR CREAM CHEESE","Mixed fruit tart filled with custard or cream cheese"
+53344300,"DESSERT PIZZA","Dessert pizza"
+53345000,"PIE, LEMON CREAM","Pie, lemon cream"
+53345070,"PIE, LEMON CREAM, INDIVIDUAL SIZE OR TART","Pie, lemon cream, individual size or tart"
+53346000,"PIE, PEANUT BUTTER CREAM","Pie, peanut butter cream"
+53346500,"PIE, PINEAPPLE CREAM","Pie, pineapple cream"
+53347000,"PIE, PUMPKIN","Pie, pumpkin"
+53347070,"PIE, PUMPKIN, INDIVIDUAL SIZE OR TART","Pie, pumpkin, individual size or tart"
+53347100,"PIE, RASPBERRY CREAM","Pie, raspberry cream"
+53347500,"PIE, SOUR CREAM, RAISIN","Pie, sour cream, raisin"
+53347600,"PIE, SQUASH","Pie, squash"
+53348000,"PIE, STRAWBERRY CREAM","Pie, strawberry cream"
+53348070,"PIE, STRAWBERRY CREAM, INDIVIDUAL SIZE OR TART","Pie, strawberry cream, individual size or tart"
+53360000,"PIE, SWEET POTATO","Pie, sweet potato"
+53365000,"PIE, VANILLA CREAM","Pie, vanilla cream"
+53366000,"PIE, YOGURT, FROZEN","Pie, yogurt, frozen"
+53370000,"PIE, CHIFFON, NOT CHOCOLATE","Pie, chiffon, not chocolate"
+53371000,"PIE, CHIFFON, CHOCOLATE","Pie, chiffon, chocolate"
+53371100,"PIE, CHIFFON, W/ LIQUEUR","Pie, chiffon, with liqueur"
+53373000,"PIE, BLACK BOTTOM","Pie, black bottom"
+53381000,"PIE, LEMON MERINGUE","Pie, lemon meringue"
+53381070,"PIE, LEMON MERINGUE, INDIVIDUAL SIZE OR TART","Pie, lemon meringue, individual size or tart"
+53382000,"PIE, CHOCOLATE-MARSHMALLOW","Pie, chocolate-marshmallow"
+53385000,"PIE, PECAN","Pie, pecan"
+53385070,"PIE, PECAN, INDIVIDUAL SIZE","Pie, pecan, individual size or tart"
+53385500,"PIE, OATMEAL","Pie, oatmeal"
+53386000,"PIE, PUDDING, NOT CHOCOLATE","Pie, pudding, flavors other than chocolate"
+53386050,"PIE, PUDDING, NOT CHOC, INDIVIDUAL SIZE","Pie, pudding, flavors other than chocolate, individual size or tart"
+53386250,"PIE, PUDDING, CHOC, W/ CHOC COATING, INDIVID SIZE","Pie, pudding, chocolate, with chocolate coating, individual size"
+53386500,"PIE, PUDDING, NOT CHOC, CHOC-COATED, INDIVID SIZE","Pie, pudding, flavors other than chocolate, with chocolate coating, individual size"
+53387000,"PIE, TOLL HOUSE CHOCOLATE CHIP","Pie, Toll house chocolate chip"
+53390000,"PIE, SHOO-FLY","Pie, shoo-fly"
+53390100,"PIE, TOFU W/ FRUIT","Pie, tofu with fruit"
+53391000,"PIE SHELL","Pie shell"
+53391100,"PIE SHELL, GRAHAM CRACKER","Pie shell, graham cracker"
+53391150,"PIE SHELL, CHOCOLATE WAFER","Pie shell, chocolate wafer"
+53391200,"VANILLA WAFER DESSERT BASE","Vanilla wafer dessert base"
+53400200,"BLINTZ, CHEESE-FILLED","Blintz, cheese-filled"
+53400300,"BLINTZ, FRUIT-FILLED","Blintz, fruit-filled"
+53410100,"COBBLER, APPLE (INCLUDE FRUIT COBBLER)","Cobbler, apple"
+53410200,"COBBLER, APRICOT","Cobbler, apricot"
+53410300,"COBBLER, BERRY","Cobbler, berry"
+53410500,"COBBLER, CHERRY","Cobbler, cherry"
+53410800,"COBBLER, PEACH","Cobbler, peach"
+53410850,"COBBLER, PEAR","Cobbler, pear"
+53410860,"COBBLER, PINEAPPLE","Cobbler, pineapple"
+53410880,"COBBLER, PLUM","Cobbler, plum"
+53410900,"COBBLER, RHUBARB","Cobbler, rhubarb"
+53415100,"CRISP, APPLE, APPLE DESSERT","Crisp, apple, apple dessert"
+53415120,"FRITTER, APPLE","Fritter, apple"
+53415200,"FRITTER, BANANA","Fritter, banana"
+53415220,"FRITTER, BERRY","Fritter, berry"
+53415300,"CRISP, BLUEBERRY","Crisp, blueberry"
+53415400,"CRISP, CHERRY","Crisp, cherry"
+53415500,"CRISP, PEACH","Crisp, peach"
+53415600,"CRISP, RHUBARB","Crisp, rhubarb"
+53420000,"CREAM PUFF/ECLAIR, CUSTARD/CREAM-FILLED, NS ICING","Cream puff, eclair, custard or cream filled, NS as to icing"
+53420100,"CREAM PUFF/ECLAIR, CUSTARD/CREAM-FILLED, NOT ICED","Cream puff, eclair, custard or cream filled, not iced"
+53420200,"CREAM PUFF/ECLAIR, CUSTARD/CREAM-FILLED, ICED","Cream puff, eclair, custard or cream filled, iced"
+53420210,"CREAM PUFF/ECLAIR, CUSTARD/CREAM-FILLED, ICED, REDUCED FAT","Cream puff, eclair, custard or cream filled, iced, reduced fat"
+53420250,"CREAM PUFFS, NO FILLING OR ICING","Cream puff, no filling or icing"
+53420300,"AIR-FILLED FRITTER, W/O SYRUP, PUERTO RICAN STYLE","Air filled fritter or fried puff, without syrup, Puerto Rican style (Bunuelos de viento)"
+53420310,"WHEAT FLOUR FRITTER, W/O SYRUP","Wheat flour fritter, without syrup"
+53420400,"SOPAIPILLA W/O SYRUP OR HONEY","Sopaipilla, without syrup or honey"
+53420410,"SOPAIPILLA W/ SYRUP OR HONEY","Sopaipilla with syrup or honey"
+53430000,"CREPE, DESSERT TYPE, NS AS TO FILLING","Crepe, dessert type, NS as to filling"
+53430100,"CREPE, DESSERT TYPE, CHOCOLATE-FILLED","Crepe, dessert type, chocolate-filled"
+53430200,"CREPE, DESSERT TYPE, FRUIT-FILLED","Crepe, dessert type, fruit-filled"
+53430250,"CREPE SUZETTE","Crepe suzette"
+53430300,"CREPE, DESSERT TYPE, ICE CREAM-FILLED","Crepe, dessert type, ice cream-filled"
+53430700,"TAMALE, SWEET","Tamale, sweet"
+53430750,"TAMALE, SWEET, W/ FRUIT","Tamale, sweet, with fruit"
+53440000,"STRUDEL, APPLE (INCLUDE STRUDEL, NFS)","Strudel, apple"
+53440300,"STRUDEL, BERRY","Strudel, berry"
+53440500,"STRUDEL, CHERRY","Strudel, cherry"
+53440600,"STRUDEL, CHEESE","Strudel, cheese"
+53440700,"STRUDEL, PEACH","Strudel, peach"
+53440750,"STRUDEL, PINEAPPLE","Strudel, pineapple"
+53440800,"STRUDEL, CHEESE & FRUIT","Strudel, cheese and fruit"
+53441110,"BAKLAVA (INCLUDE KADAYIF)","Baklava"
+53441210,"BASBOUSA (SEMOLINA DESSERT DISH)","Basbousa (semolina dessert dish)"
+53450000,"TURNOVER OR DUMPLING, APPLE","Turnover or dumpling, apple"
+53450300,"TURNOVER OR DUMPLING, BERRY","Turnover or dumpling, berry"
+53450500,"TURNOVER OR DUMPLING, CHERRY","Turnover or dumpling, cherry"
+53450800,"TURNOVER OR DUMPLING, LEMON","Turnover or dumpling, lemon"
+53451000,"TURNOVER OR DUMPLING, PEACH","Turnover or dumpling, peach"
+53451500,"TURNOVER, GUAVA","Turnover, guava"
+53451750,"TURNOVER, PUMPKIN","Turnover, pumpkin"
+53452100,"PASTRY, FRUIT-FILLED","Pastry, fruit-filled"
+53452120,"PASTRY, ASIAN, MADE WITH BEAN OR LOTUS SEED PASTE FILLING","Pastry, Asian, made with bean or lotus seed paste filling (baked)"
+53452130,"PASTRY, ASIAN, MADE WITH BEAN PASTE AND SALTED EGG YOLK FILL","Pastry, Asian, made with bean paste and salted egg yolk filling (baked)"
+53452150,"PASTRY, CHINESE (INCLUDE 9-LAYER PUDDING)","Pastry, Chinese, made with rice flour"
+53452170,"PASTRY, COOKIE TYPE, FRIED(INCL POLISH PACZKI)","Pastry, cookie type, fried"
+53452200,"PASTRY, ITALIAN, W/ CHEESE (INCLUDE CANNOLI)","Pastry, Italian, with cheese"
+53452400,"PASTRY, PUFF","Pastry, puff"
+53452420,"PASTRY, PUFF, CUSTARD/CREAM FILLED, ICED/NOT ICED","Pastry, puff, custard or cream filled, iced or not iced"
+53452450,"CHEESE PASTRY PUFF","Cheese pastry puffs"
+53452500,"PASTRY, MAINLY  FLOUR & WATER, FRIED","Pastry, mainly flour and water, fried"
+53453150,"EMPANADA, MEXICAN TURNOVER, FRUIT-FILLED","Empanada, Mexican turnover, fruit-filled"
+53453170,"EMPANADA, MEXICAN TURNOVER, PUMPKIN","Empanada,  Mexican turnover, pumpkin"
+53500100,"BREAKFAST PASTRY, NFS","Breakfast pastry, NFS"
+53510000,"DANISH PASTRY, PLAIN/SPICE (INCL W/ ICING)","Danish pastry, plain or spice"
+53510100,"DANISH PASTRY, W/ FRUIT","Danish pastry, with fruit"
+53511000,"DANISH PASTRY, W/ CHEESE","Danish pastry, with cheese"
+53520000,"DOUGHNUT, NS AS TO CAKE OR YEAST","Doughnut, NS as to cake or yeast"
+53520110,"DOUGHNUT, CAKE TYPE","Doughnut, cake type"
+53520120,"DOUGHNUT, CHOCOLATE, CAKE TYPE","Doughnut, chocolate, cake type"
+53520140,"DOUGHNUT, CAKE TYPE, CHOCOLATE COVERED","Doughnut, cake type, chocolate covered"
+53520150,"DOUGHNUT, CAKE TYPE, CHOCOLATE COVERED, W/ PEANUTS","Doughnut, cake type, chocolate covered, dipped in peanuts"
+53520160,"DOUGHNUT, CHOCOLATE, CAKE TYPE, WITH CHOCOLATE ICING","Doughnut, chocolate, cake type, with chocolate icing"
+53520200,"CHURROS (INCL MEXICAN CRUELLERS)","Churros"
+53520500,"DOUGHNUT, ASIAN","Doughnut, Asian"
+53520600,"CRULLER, NFS","Cruller, NFS"
+53520700,"FRENCH CRULLER","French cruller"
+53521100,"DOUGHNUT, CHOCOLATE, RAISED OR YEAST, WITH CHOCOLATE ICING","Doughnut, chocolate, raised or yeast, with chocolate icing"
+53521110,"DOUGHNUT, RAISED / YEAST","Doughnut, raised or yeast"
+53521120,"DOUGHNUT, CHOCOLATE, RAISED OR YEAST","Doughnut, chocolate, raised or yeast"
+53521130,"DOUGHNUT, RAISED OR YEAST, CHOCOLATE COVERED","Doughnut, raised or yeast, chocolate covered"
+53521140,"DOUGHNUT, JELLY","Doughnut, jelly"
+53521210,"DOUGHNUT, CUSTARD-FILLED","Doughnut, custard-filled"
+53521220,"DOUGHNUT, CHOCOLATE CREAM-FILLED","Doughnut, chocolate cream-filled"
+53521230,"DOUGHNUT, CUSTARD-FILLED, WITH ICING","Doughnut, custard-filled, with icing"
+53521250,"DOUGHNUT, WHEAT","Doughnut, wheat"
+53521300,"DOUGHNUT, WHEAT, CHOCOLATE COVERED","Doughnut, wheat, chocolate covered"
+53530000,"BREAKFAST TART","Breakfast tart"
+53530010,"BREAKFAST TART, LOWFAT","Breakfast tart, lowfat"
+53610100,"COFFEE CAKE, CRUMB OR QUICK-BREAD TYPE","Coffee cake, crumb or quick-bread type"
+53610170,"COFFEE CAKE, CRUMB OR QUICK-BREAD TYPE, W/ FRUIT","Coffee cake, crumb or quick-bread type, with fruit"
+53610200,"COFFEECAKE, CRUMB OR QUICK-BREAD TYPE, CHEESE FILLD","Coffee cake, crumb or quick-bread type, cheese-filled"
+53710400,"FIBER ONE CHEWY BAR","Fiber One Chewy Bar"
+53710500,"KELLOGG'S NUTRI-GRAIN CEREAL BAR","Kellogg's Nutri-Grain Cereal Bar"
+53710502,"KELLOGG'S NUTRI-GRAIN YOGURT BAR","Kellogg's Nutri-Grain Yogurt Bar"
+53710504,"KELLOGG'S NUTRI-GRAIN FRUIT AND NUT BAR","Kellogg's Nutri-Grain Fruit and Nut Bar"
+53710600,"MILK 'N CEREAL BAR","Milk 'n Cereal bar"
+53710700,"KELLOGG'S SPECIAL K BAR","Kellogg's Special K bar"
+53710800,"KASHI GOLEAN CHEWY BARS","Kashi GOLEAN Chewy Bars"
+53710802,"KASHI TLC CHEWY GRANOLA BAR","Kashi TLC Chewy Granola Bar"
+53710804,"KASHI GOLEAN CRUNCHY BARS","Kashi GOLEAN Crunchy Bars"
+53710806,"KASHI TLC CRUNCHY GRANOLA BAR","Kashi TLC Crunchy Granola Bar"
+53710900,"NATURE VALLEY CHEWY TRAIL MIX GRANOLA BAR","Nature Valley Chewy Trail Mix Granola Bar"
+53710902,"NATURE VALLEY CHEWY GRANOLA BAR WITH YOGURT COATING","Nature Valley Chewy Granola Bar with Yogurt Coating"
+53710904,"NATURE VALLEY SWEET AND SALTY GRANOLA BAR","Nature Valley Sweet and Salty Granola Bar"
+53710906,"NATURE VALLEY CRUNCHY GRANOLA BAR","Nature Valley Crunchy Granola Bar"
+53711000,"QUAKER CHEWY GRANOLA BAR","Quaker Chewy Granola Bar"
+53711002,"QUAKER CHEWY 90 CALORIE GRANOLA BAR","Quaker Chewy 90 Calorie Granola Bar"
+53711004,"QUAKER CHEWY 25% LESS SUGAR GRANOLA BAR","Quaker Chewy 25% Less Sugar Granola Bar"
+53711006,"QUAKER CHEWY DIPPS GRANOLA BAR","Quaker Chewy Dipps Granola Bar"
+53711100,"QUAKER GRANOLA BITES","Quaker Granola Bites"
+53712000,"SNACK BAR, OATMEAL","Snack bar, oatmeal"
+53712100,"GRANOLA BAR, NFS","Granola bar, NFS"
+53712200,"GRANOLA BAR, LOWFAT, NFS","Granola bar, lowfat, NFS"
+53712210,"GRANOLA BAR, NONFAT","Granola bar, nonfat"
+53713000,"GRANOLA BAR, REDUCED SUGAR, NFS","Granola bar, reduced sugar, NFS"
+53713100,"GRANOLA BAR, PEANUTS , OATS, SUGAR, WHEAT GERM","Granola bar, peanuts , oats, sugar, wheat germ"
+53714200,"GRANOLA BAR, CHOCOLATE-COATED, NFS","Granola bar, chocolate-coated, NFS"
+53714210,"GRANOLA BAR, WITH COCONUT, CHOCOLATE-COATED","Granola bar, with coconut, chocolate-coated"
+53714220,"GRANOLA BAR WITH NUTS, CHOCOLATE-COATED","Granola bar with nuts, chocolate-coated"
+53714230,"GRANOLA BAR, OATS, NUTS, COATED WITH NON-CHOCOLATE COATING","Granola bar, oats, nuts, coated with non-chocolate coating"
+53714250,"GRANOLA BAR, COATED WITH NON-CHOCOLATE COATING","Granola bar, coated with non-chocolate coating"
+53714300,"GRANOLA BAR, HIGH FIBER, COATED W/ NON-CHOC YOGURT COATING","Granola bar, high fiber, coated with non-chocolate yogurt coating"
+53714400,"GRANOLA BAR, WITH RICE CEREAL","Granola bar, with rice cereal"
+53714500,"BREAKFAST BAR, NFS","Breakfast bar, NFS"
+53714510,"BREAKFAST BAR, DATE, WITH YOGURT COATING","Breakfast bar, date, with yogurt coating"
+53714520,"BREAKFAST BAR, CEREAL CRUST WITH FRUIT FILLING, LOWFAT","Breakfast bar, cereal crust with fruit filling, lowfat"
+53720100,"BALANCE ORIGINAL BAR","Balance Original Bar"
+53720200,"CLIF BAR","Clif Bar"
+53720300,"POWERBAR","PowerBar"
+53720400,"SLIM FAST ORIGINAL MEAL BAR","Slim Fast Original Meal Bar"
+53720500,"SNICKERS MARATHON PROTEIN BAR","Snickers Marathon Protein bar"
+53720510,"SNICKERS MARATHON ENERGY BAR","Snickers Marathon Energy bar"
+53720600,"SOUTH BEACH LIVING MEAL BAR","South Beach Living Meal Bar"
+53720610,"SOUTH BEACH LIVING HIGH PROTEIN BAR","South Beach Living High Protein Bar"
+53720700,"TIGER'S MILK BAR","Tiger's Milk bar"
+53720800,"ZONE PERFECT CLASSIC CRUNCH NUTRITION BAR","Zone Perfect Classic Crunch nutrition bar"
+53729000,"NUTRITION BAR OR MEAL REPLACEMENT BAR, NFS","Nutrition bar or meal replacement bar, NFS"
+53801000,"CEREAL BAR WITH FRUIT FILLING, BABY FOOD","Cereal bar with fruit filling, baby food"
+53803050,"COOKIE, FRUIT, BABY FOOD","Cookie, fruit, baby food"
+53803100,"COOKIE, BABY FOOD","Cookie, baby food"
+53803250,"COOKIE, TEETHING, BABY","Cookie, teething, baby"
+53803300,"COOKIE, RICE, BABY","Cookie, rice, baby"
+54001000,"CRACKER, NS AS TO SWEET/NONSWEET (INCL CRACKER,NFS)","Crackers, NS as to sweet or nonsweet"
+54102010,"CRACKERS, GRAHAM","Crackers, graham"
+54102020,"CRACKERS, GRAHAM, CHOCOLATE COVERED","Crackers, graham, chocolate covered"
+54102050,"CRACKERS, OATMEAL","Crackers, oatmeal"
+54102060,"CRACKERS, CUBAN","Crackers, Cuban"
+54102070,"CRACKERS, CUCA","Crackers, Cuca"
+54102080,"CRACKERS, GRAHAM, W/ RAISINS","Crackers, graham, with raisins"
+54102100,"CRACKERS, GRAHAM, LOWFAT","Crackers, graham, lowfat"
+54102110,"CRACKERS, GRAHAM, FAT FREE","Crackers, graham, fat free"
+54102200,"CRACKERS, GRAHAM, SANDWICH-TYPE, WITH FILLING","Crackers, graham, sandwich-type, with filling"
+54201010,"CRACKERS, MATZO, LOW SODIUM","Crackers, matzo, low sodium"
+54202010,"CRACKERS, SALTINE, LOW SODIUM","Crackers, saltine, low sodium"
+54203010,"CRACKERS, TOAST THINS (RYE/WHEAT/WHITE), LOW SODIUM","Crackers, toast thins (rye, wheat, white flour), low sodium"
+54204010,"CRACKER, 100% WHOLE WHEAT,LO SODIUM","Cracker, 100% whole wheat, low sodium"
+54205010,"CRACKER, SNACK, LOW SODIUM","Cracker, snack, low sodium"
+54205030,"CRACKER, CHEESE, LOW SODIUM","Cracker, cheese, low sodium"
+54205100,"CRACKER, SNACK, LOWFAT, LOW SODIUM","Cracker, snack, lowfat, low sodium"
+54206010,"PUFFED RICE CAKE W/O SALT","Puffed rice cake without salt"
+54207010,"CRISPBREAD, WHEAT, LOW SODIUM","Crispbread, wheat, low sodium"
+54210010,"CRACKER, MULTIGRAIN, LOW SODIUM","Cracker, multigrain, low sodium"
+54222000,"CRISPBREAD, RYE, LOW SODIUM","Crispbread, rye, low sodium"
+54301000,"CRACKER, SNACK","Cracker, snack"
+54301100,"CRACKER, SNACK, REDUCED FAT","Cracker, snack, reduced fat"
+54301200,"CRACKER, SNACK, FAT FREE","Cracker, snack, fat free"
+54304000,"CRACKERS, CHEESE","Cracker, cheese"
+54304100,"CRACKER, CHEESE, REDUCED FAT","Cracker, cheese, reduced fat"
+54304150,"CRACKER, CHEESE, WHOLE GRAIN","Cracker, cheese, whole grain"
+54304500,"CRACKER, HIGH FIBER, NO ADDED FAT","Cracker, high fiber, no added fat"
+54305000,"CRISPBREAD, WHEAT, NO ADDED FAT","Crispbread, wheat, no added fat"
+54307000,"CRACKERS, MATZO","Crackers, matzo"
+54308000,"CRACKERS, MILK","Crackers, milk"
+54309000,"CRACKERS, OAT BRAN (INCLUDE NABISCO OAT THINS)","Crackers, oat"
+54313000,"CRACKERS, OYSTER","Crackers, oyster"
+54318000,"CHIPS, BROWN RICE","Chips, brown rice"
+54318500,"RICE CAKE, CRACKER-TYPE","Rice cake, cracker-type"
+54319000,"CRACKERS, RICE","Crackers, rice"
+54319010,"PUFFED RICE CAKE","Puffed rice cake"
+54319020,"POPCORN CAKE (INCL PUFFED CORN & RICE CAKE)","Popcorn cake"
+54319200,"PUFFED WHEAT CAKE (INCL QUAKER)","Puffed wheat cake"
+54319500,"RICE PAPER","Rice paper"
+54322000,"CRISPBREAD, RYE, NO ADDED FAT","Crispbread, rye, no added fat"
+54325000,"CRACKERS, SALTINES","Crackers, saltine"
+54325010,"CRACKERS, SALTINE, FAT FREE","Crackers, saltine, fat free"
+54325050,"CRACKERS, SALTINE, WHOLE WHEAT","Crackers, saltine, whole wheat"
+54326000,"CRACKERS, MULTIGRAIN","Crackers, multigrain, made with whole wheat, wheat, oat, and other flours"
+54327950,"CRACKERS, CYLINDRICAL, PEANUT BUTTER-FILLED","Crackers, cylindrical, peanut-butter filled"
+54328000,"CRACKER, SANDWICH-TYPE, NFS","Crackers, sandwich-type, NFS"
+54328100,"CRACKER,SANDWICH-TYPE,PEANUT BUTTER FILLED","Cracker, sandwich-type, peanut butter filled"
+54328110,"CRACKER, SANDWICH, PEANUT BUTTER FILLED, RED FAT","Cracker, sandwich-type, peanut butter filled, reduced fat"
+54328200,"CRACKER,SANDWICH-TYPE, CHEESE-FILLED","Cracker, sandwich-type, cheese-filled"
+54334000,"CRACKERS, TOAST THINS","Crackers, toast thins (rye, pumpernickel, white flour)"
+54336000,"CRACKER, WATER BISCUIT","Crackers, water biscuits"
+54337000,"CRACKER, 100% WHOLE WHEAT","Cracker, 100% whole wheat"
+54337050,"CRACKER, 100% WHOLE WHEAT, REDUCED FAT","Cracker, 100% whole wheat, reduced fat"
+54338000,"CRACKERS, WHEAT","Crackers, wheat"
+54338100,"CRACKERS, WHEAT, REDUCED FAT","Crackers, wheat, reduced fat"
+54339000,"CRACKER CORN (INCL STONED CORN CRACKER)","Crackers, corn"
+54350000,"CRACKERS, BABY FOOD","Crackers, baby food"
+54350010,"GERBER FINGER FOODS, PUFFS, BABY FOOD","Gerber Finger Foods, Puffs, baby food"
+54360000,"CRUNCHY SNACKS, CORN BASED, BABY FOOD","Crunchy snacks, corn based, baby food"
+54401010,"SALTY SNACKS, CORN / CORNMEAL BASE, NUT /NUG, TSTD","Salty snacks, corn or cornmeal base, nuts or nuggets, toasted"
+54401020,"SALTY SNACKS, CORN OR CORNMEAL, CORN CHIPS, CHEESE","Salty snacks, corn or cornmeal base, corn chips, corn-cheese chips"
+54401050,"SALTY SNACKS, CORN OR CORNMEAL, CORN PUFFS, TWISTS","Salty snacks, corn or cornmeal base, corn puffs and twists; corn-cheese puffs and twists"
+54401080,"SALTY SNACKS, CORN OR CORNMEAL, TORTILLA CHIPS","Salty snacks, corn or cornmeal base, tortilla chips"
+54401090,"SALTY SNACKS, CORN/CORN-CHEESE CHIPS, UNSALTED","Salty snacks, corn or cornmeal base, corn chips, corn-cheese chips, unsalted"
+54401100,"SALTY SNACKS,CORN / CORNMEAL BASE,TORTILLA CHIPS LT","Salty snacks, corn or cornmeal base, tortilla chips, light (baked with less oil)"
+54401120,"SALTY SNACKS, TORTILLA CHIPS, FAT FREE, W/ OLEAN","Salty snacks, corn or cornmeal base, tortilla chips, fat free, made with Olean"
+54401150,"SALTY SNACKS,CORN/CORNMEAL BASE,TORTILLA,LOWFAT,BKD","Salty snacks, corn or cornmeal base, tortilla chips, lowfat, baked without fat"
+54401170,"SALTY SNACKS,CORN/CORNMEAL,TORTILLA,LOWFAT,BKD,NO SALT","Salty snacks, corn or cornmeal base, tortilla chips, lowfat, baked without fat, unsalted"
+54401200,"SALTY SNACKS, CORN/CORNML BASE,W/OAT BRAN,TORT CHPS","Salty snacks, corn or cornmeal base, with oat bran, tortilla chips"
+54401210,"SALTY SNACKS, CORN BASED/CHEESE PUFFS & TWISTS, LOWFAT","Salty snacks, corn based puffs and twists, cheese puffs and twists, lowfat"
+54402080,"TORTILLA CHIPS, UNSALTED","Salty snacks, corn or cornmeal base, tortilla chips, unsalted"
+54402200,"SALTY SNACK MIXTURE,MOSTLY CORN,W/PRETZELS,W/O NUTS","Salty snack mixture, mostly corn or cornmeal based, with pretzels, without nuts"
+54402300,"SALTY SNACKS, WHEAT-BASE, HIGH FIBER","Salty snacks, wheat-based, high fiber"
+54402500,"SALTY SNACKS, WHEAT-AND CORN-BASED CHIPS","Salty snacks, wheat- and corn-based chips"
+54402600,"SALTY SNACKS, MULTIGRAIN, WHOLE GRAIN, CHIPS","Salty snacks, multigrain, whole grain, chips (made with whole corn, whole wheat, rice flour, and whole oat flour)"
+54402610,"SALTY SNACKS, MULTIGRAIN& POT CHIPS(W/RICE FL,POT,CORN FL)","Salty snacks, multigrain and potato chips (made with rice flour, dried potatoes, corn flour, and wheat starch)"
+54402700,"PITA CHIPS","Pita chips"
+54403000,"POPCORN, POPPED IN OIL, UNBUTTERED","Popcorn, popped in oil, unbuttered"
+54403010,"POPCORN, AIR-POPPED (NO BUTTER OR OIL ADDED)","Popcorn, air-popped (no butter or no oil added)"
+54403020,"POPCORN, POPPED IN OIL, BUTTERED","Popcorn, popped in oil, buttered"
+54403040,"POPCORN, AIR-POPPED, BUTTERED","Popcorn, air-popped, buttered"
+54403050,"POPCORN, FLAVORED (CHEESE, BBQ, SOUR CREAM, ONION)","Popcorn, flavored"
+54403060,"POPCORN, POPPED IN OIL, LOWFAT, LOW SODIUM","Popcorn, popped in oil, lowfat, low sodium"
+54403070,"POPCORN, POPPED IN OIL, LOWFAT","Popcorn, popped in oil, lowfat"
+54403090,"POPCORN, POPPED IN OIL, UNSALTED","Popcorn, popped in oil, unsalted"
+54403110,"POPCORN, SUGAR SYRUP OR CARAMEL COATED","Popcorn, sugar syrup or caramel-coated"
+54403120,"POPCORN, SUGAR SYRUP OR CARAMEL COATED, W/ NUTS","Popcorn, sugar syrup or caramel-coated, with nuts"
+54403150,"POPCORN, SUGAR SYRUP/CARAMEL COATED, FAT FREE","Popcorn, sugar syrup or caramel-coated, fat free"
+54406010,"SNACKS, ONION-FLAVORED RINGS","Snacks, onion-flavored rings"
+54406200,"SHRIMP CHIPS","Shrimp chips (tapioca base)"
+54408000,"PRETZELS, NFS","Pretzels, NFS"
+54408010,"PRETZELS, HARD","Pretzels, hard"
+54408020,"PRETZELS, SOFT","Pretzels, soft"
+54408030,"PRETZELS, HARD, UNSALTED","Pretzel, hard, unsalted"
+54408040,"PRETZELS, SOFT, UNSALTED","Pretzels, soft, unsalted"
+54408050,"PRETZEL, OAT BRAN, HARD","Pretzel, oatbran, hard"
+54408070,"PRETZEL, HARD, MULTIGRAIN","Pretzel, hard, multigrain"
+54408100,"PRETZEL, BABY FOOD","Pretzel, baby food"
+54408200,"PRETZEL, HARD, CHOCOLATE COATED","Pretzel, hard, chocolate-coated"
+54408250,"PRETZEL, YOGURT COVERED","Pretzel, yogurt-covered"
+54408300,"PRETZELS, CHEESE-FILLED (INCL COMBOS)","Pretzels, cheese-filled"
+54412110,"WHEAT STICKS, 100% WHOLE WHEAT","Wheat sticks, 100% whole wheat"
+54420010,"MULTIGRAIN MIXTURE, PRETZELS, CEREAL &/ CRACKERS,NUTS","Multigrain mixture, pretzels, cereal and/or crackers, nuts"
+54420100,"ORIENTAL PARTY MIX, W/ PEANUTS, SESAME STICKS, ETC","Oriental party mix, with peanuts, sesame sticks, chili rice crackers and fried green peas"
+54420200,"MULTIGRAIN MIX, BREAD STICKS, SESAME NUGGETS, PRETZ","Multigrain mixture, bread sticks, sesame nuggets, pretzels, rye chips"
+54430010,"YOGURT CHIPS","Yogurt chips"
+54440010,"BAGEL CHIP","Bagel chip"
+55101000,"PANCAKES, PLAIN (INCLUDE PANCAKES, NFS)","Pancakes, plain"
+55101010,"PANCAKES, REDUCED CALORIE, HIGH FIBER","Pancakes, reduced calorie, high fiber"
+55101015,"PANCAKES, PLAIN, REDUCED FAT","Pancakes, plain, reduced fat"
+55101020,"PANCAKES, PLAIN, FAT FREE","Pancakes, plain, fat free"
+55103000,"PANCAKES, W/ FRUIT (INCLUDE BLUEBERRY PANCAKES)","Pancakes, with fruit"
+55103100,"PANCAKES W/ CHOC CHIPS","Pancakes, with chocolate chips"
+55105000,"PANCAKES, BUCKWHEAT","Pancakes, buckwheat"
+55105100,"PANCAKES, CORNMEAL","Pancakes, cornmeal"
+55105200,"PANCAKES, WHOLE WHEAT","Pancakes, whole wheat"
+55105205,"PANCAKES, WHOLE WHEAT, REDUCED FAT","Pancakes, whole wheat, reduced fat"
+55105210,"PANCAKES, WHOLE WHEAT, FAT FREE","Pancakes, whole wheat, fat free"
+55105300,"PANCAKES, SOURDOUGH","Pancakes, sour dough"
+55105400,"PANCAKES, RYE","Pancakes, rye"
+55201000,"WAFFLE, PLAIN","Waffle, plain"
+55202000,"WAFFLE, WHEAT, BRAN, OR MULTIGRAIN","Waffle, wheat, bran, or multigrain"
+55203000,"WAFFLE, FRUIT","Waffle, fruit"
+55203500,"WAFFLE, NUT & HONEY (INCL EGGO)","Waffle, nut and honey"
+55203600,"WAFFLE, CHOCOLATE CHIP","Waffle, chocolate chip"
+55204000,"WAFFLE, CORNMEAL","Waffle, cornmeal"
+55205000,"WAFFLE, 100% WHOLE WHEAT OR 100% WHOLE GRAIN","Waffle, 100% whole wheat or 100% whole grain"
+55206000,"WAFFLE, OAT BRAN","Waffle, oat bran"
+55207000,"WAFFLE, MULTI-BRAN (INCLUDE EGGO NUTRIGRAIN)","Waffle, multi-bran"
+55211000,"WAFFLE, PLAIN, FAT FREE","Waffle, plain, fat free"
+55211050,"WAFFLE, PLAIN, LOWFAT","Waffle, plain, lowfat"
+55212000,"WAFFLE, WHOLE WHEAT, LOWFAT","Waffle, whole wheat, lowfat"
+55301000,"FRENCH TOAST, PLAIN (INCLUDE ROMAN MEAL)","French toast, plain"
+55301050,"FRENCH TOAST STICKS, PLAIN","French toast sticks, plain"
+55310100,"BREAD FRITTERS, P.R.","Bread fritters, Puerto Rican style (Torrejas gallegas, Galician fritters)"
+55401000,"CREPE, PLAIN (INCLUDE FRENCH PANCAKE)","Crepe, plain"
+55501000,"FLOUR & WATER PATTY (INCLUDE CHINESE PANCAKE)","Flour and water patty"
+55502000,"FLOUR AND WATER GRAVY","Flour and water gravy"
+55610200,"DUMPLING, FRIED, PUERTO RICAN STYLE","Dumpling, fried, Puerto Rican style"
+55610300,"DUMPLING, PLAIN","Dumpling, plain"
+55701000,"CAKE MADE W/ GLUTINOUS RICE","Cake made with glutinous rice"
+55702000,"CAKE OR PANCAKE MADE W/ RICE FLOUR &/OR DRIED BEANS","Cake or pancake made with rice flour and/or dried beans"
+55702100,"DOSA (INDIAN)","Dosa (Indian), plain"
+55703000,"CAKE MADE W/ GLUTINOUS RICE & DRIED BEANS","Cake made with glutinous rice and dried beans"
+55801000,"FUNNEL CAKE WITH SUGAR","Funnel cake with sugar"
+55801010,"FUNNEL CAKE WITH SUGAR AND FRUIT","Funnel cake with sugar and fruit"
+56101000,"MACARONI, COOKED, NS AS TO ADDED FAT","Macaroni, cooked, NS as to fat added in cooking"
+56101010,"MACARONI, COOKED, NO FAT ADDED","Macaroni, cooked, fat not added in cooking"
+56101030,"MACARONI, COOKED, FAT ADDED","Macaroni, cooked, fat added in cooking"
+56102000,"MACARONI, WHOLE WHEAT, COOKED, NS AS TO ADDED FAT","Macaroni, whole wheat, cooked, NS as to fat added in cooking"
+56102010,"MACARONI, WHOLE WHEAT, NO FAT ADDED","Macaroni, whole wheat, cooked, fat not added in cooking"
+56102020,"MACARONI, WHOLE WHEAT, FAT ADDED","Macaroni, whole wheat, cooked, fat added in cooking"
+56103000,"MACARONI, SPINACH, NS AS TO ADDED FAT","Macaroni, cooked, spinach, NS as to fat added in cooking"
+56103010,"MACARONI, SPINACH, NO FAT ADDED","Macaroni, cooked, spinach, fat not added in cooking"
+56103020,"MACARONI, SPINACH, FAT ADDED","Macaroni, cooked, spinach, fat added in cooking"
+56104000,"MACARONI,CKD,VEGETABLE,NS AS TO FAT ADDED","Macaroni, cooked, vegetable, NS as to fat added in cooking"
+56104010,"MACARONI,COOKED,VEGETABLE,FAT NOT ADDED IN COOKING","Macaroni, cooked, vegetable, fat not added in cooking"
+56104020,"MACARONI,COOKED,VEGETABLE, FAT ADDED IN COOKING","Macaroni, cooked, vegetable, fat added in cooking"
+56112000,"NOODLES, COOKED, NS AS TO ADDED FAT","Noodles, cooked, NS as to fat added in cooking"
+56112010,"NOODLES, COOKED, NO FAT ADDED","Noodles, cooked, fat not added in cooking"
+56112030,"NOODLES, COOKED, FAT ADDED","Noodles, cooked, fat added in cooking"
+56113000,"NOODLES, COOKED,WHOLE WHEAT,NS AS TO FAT ADDED","Noodles, cooked, whole wheat, NS as to fat added in cooking"
+56113010,"NOODLES, WHOLE WHEAT, COOKED, NO FAT ADDED","Noodles, cooked, whole wheat, fat not added in cooking"
+56113990,"NOODLES, COOKED, SPINACH, NS AS TO FAT","Noodles, cooked, spinach, NS as to fat added in cooking"
+56114000,"NOODLES, SPINACH, COOKED, NO FAT ADDED","Noodles, cooked, spinach, fat not added in cooking"
+56114020,"NOODLES, COOKED, SPINACH, FAT ADDED","Noodles, cooked, spinach, fat added in cooking"
+56116000,"NOODLES, CHOW MEIN","Noodles, chow mein"
+56116990,"LONG RICE NOODLES(FROM MUNG BEANS),CKD,NS FAT ADDed","Long rice noodles (made from mung beans) cooked, NS as to fat added in cooking"
+56117000,"LONG RICE NOODLES, COOKED, NO FAT ADDED","Long rice noodles (made from mung beans), cooked, fat not added in cooking"
+56117010,"LONG RICE NOODLES, COOKED, FAT ADDED","Long rice noodles (made from mung beans), cooked, fat added in cooking"
+56117090,"CHOW FUN RICE NOODLES,COOKED,NS AS TO FAT ADDED","Chow fun rice noodles, cooked, NS as to fat added in cooking"
+56117100,"CHOW FUN RICE NOODLES, COOKED, NO FAT ADDED","Chow fun rice noodles, cooked, fat not added in cooking"
+56117110,"CHOW FUN RICE NOODLES, COOKED, FAT ADDED","Chow fun rice noodles, cooked, fat added in cooking"
+56130000,"SPAGHETTI, COOKED, NS AS TO ADDED FAT","Spaghetti, cooked, NS as to fat added in cooking"
+56130010,"SPAGHETTI, COOKED, NO FAT ADDED","Spaghetti, cooked, fat not added in cooking"
+56131000,"SPAGHETTI, COOKED, FAT ADDED","Spaghetti, cooked, fat added in cooking"
+56132990,"SPAGHETTI, COOKED, WHOLE WHEAT, NS AS TO ADDED FAT","Spaghetti, cooked, whole wheat, NS as to fat added in cooking"
+56133000,"SPAGHETTI, COOKED, WHOLE WHEAT, NO FAT ADDED","Spaghetti, cooked, whole wheat, fat not added in cooking"
+56133010,"SPAGHETTI, COOKED, WHOLE WHEAT, FAT ADDED","Spaghetti, cooked, whole wheat, fat added in cooking"
+56200300,"CEREAL, COOKED, NFS","Cereal, cooked, NFS"
+56200350,"CEREAL, COOKED, INSTANT, NS AS TO GRAIN","Cereal, cooked, instant, NS as to grain"
+56200390,"BARLEY, COOKED, NS AS TO FAT ADDED IN COOKING","Barley, cooked, NS as to fat added in cooking"
+56200400,"BARLEY, COOKED, NO FAT ADDED","Barley, cooked, fat not added in cooking"
+56200490,"BUCKWHEAT GROATS, COOKED, NS AS TO FAT ADDED","Buckwheat groats, cooked, NS as to fat added in cooking"
+56200500,"BUCKWHEAT GROATS, COOKED, NO FAT ADDED (INCL KASHA)","Buckwheat groats, cooked, fat not added in cooking"
+56200510,"BUCKWHEAT GROATS, COOKED, FAT ADDED","Buckwheat groats, cooked, fat added in cooking"
+56200990,"GRITS, COOKED,CORN/HOMINY, NS REG, QUICK, INST, NS FAT ADDED","Grits, cooked, corn or hominy, NS as to regular, quick, or instant, NS as to fat added in cooking"
+56201000,"GRITS, CORN OR HOMINY, NFS, NO FAT ADDED","Grits, cooked, corn or hominy, NS as to regular, quick, or instant, fat not added in cooking"
+56201010,"GRITS, CKD, CORN/HOMINY, REGULAR, NO FAT","Grits, cooked, corn or hominy, regular, fat not added in cooking"
+56201020,"GRITS, COOKED, CORN/HOMINY, REGULAR, FAT ADDED","Grits, cooked, corn or hominy, regular, fat added in cooking"
+56201030,"GRITS, COOKED, CORN/HOMINY, REGULAR, NS AS TO FAT","Grits, cooked, corn or hominy, regular, NS as to fat added in cooking"
+56201040,"GRITS, COOKED, CORN/HOMINY, FAT ADDED","Grits, cooked, corn or hominy, NS as to regular, quick, or instant, fat added in cooking"
+56201060,"GRITS,CKD,CORN/HOMINY,W/CHEESE,NS TYPE,NS FAT ADDED","Grits, cooked, corn or hominy, with cheese, NS as to regular, quick, or instant, NS as to fat added in cooking"
+56201061,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,NS TYPE,FAT NOT ADDED","Grits, cooked, corn or hominy, with cheese, NS as to regular, quick, or instant, fat not added in cooking"
+56201062,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,NS TYPE,FAT ADDED","Grits, cooked, corn or hominy, with cheese, NS as to regular, quick, or instant, fat added in cooking"
+56201070,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,REG,NS FAT ADDED","Grits, cooked, corn or hominy, with cheese, regular, NS as to fat added in cooking"
+56201071,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,REG,FAT NOT ADDED","Grits, cooked, corn or hominy, with cheese, regular, fat not added in cooking"
+56201072,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,REG,FAT ADDED","Grits, cooked, corn or hominy, with cheese, regular, fat added in cooking"
+56201080,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,QUICK,NS FAT ADDED","Grits, cooked, corn or hominy, with cheese, quick, NS as to fat added in cooking"
+56201081,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,QUICK,FAT NOT ADDED","Grits, cooked, corn or hominy, with cheese, quick, fat not added in cooking"
+56201082,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,QUICK,FAT ADDED","Grits, cooked, corn or hominy, with cheese, quick, fat added in cooking"
+56201090,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,INSTANT,NS FAT ADDED","Grits, cooked, corn or hominy, with cheese, instant, NS as to fat added in cooking"
+56201091,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,INSTANT,FAT NOT ADDED","Grits, cooked, corn or hominy, with cheese, instant, fat not added in cooking"
+56201092,"GRITS,CKD,CORN/HOMINY,W/ CHEESE,INSTANT,FAT ADDED","Grits, cooked, corn or hominy, with cheese, instant, fat added in cooking"
+56201110,"GRITS, COOKED, CORN/HOMINY, QUICK, NO FAT ADDED","Grits, cooked, corn or hominy, quick, fat not added in cooking"
+56201120,"GRITS, COOKED, CORN/HOMINY, QUICK, FAT ADDED","Grits, cooked, corn or hominy, quick, fat added in cooking"
+56201130,"GRITS, COOKED, CORN/HOMINY,QUICK,NS AS TO ADDED FAT","Grits, cooked, corn or hominy, quick, NS as to fat added in cooking"
+56201210,"GRITS, COOKED, CORN/HOMINY, INSTANT, NO FAT ADDED","Grits, cooked, corn or hominy, instant, fat not added in cooking"
+56201220,"GRITS, CORN/HOMINY, INSTANT, FAT ADDED","Grits, cooked, corn or hominy, instant, fat added in cooking"
+56201230,"GRITS, CORN/HOMINY, INSTANT, COOKED, NS AS TO FAT","Grits, cooked, corn or hominy, instant, NS as to fat added in cooking"
+56201240,"GRITS, FLAVORED, INSTANT, NO FAT ADDED","Grits, cooked, flavored, corn or hominy, instant, fat not added in cooking"
+56201250,"GRITS, FLAVORED, INSTANT, FAT ADDED","Grits, cooked, flavored, corn or hominy, instant, fat added in cooking"
+56201260,"GRITS, FLAVORED, INSTANT, NS AS TO ADDED FAT","Grits, cooked, flavored, corn or hominy, instant, NS as to fat added in cooking"
+56201300,"GRITS,COOKED,CORN/HOM,NFS,MADE W/MILK, NS AS TO FAT ADDED","Grits, cooked, corn or hominy, NS as to regular, quick, or instant, made with milk, NS as to fat added in cooking"
+56201510,"CORNMEAL MUSH, MADE W/ WATER","Cornmeal mush, made with water"
+56201520,"CORNMEAL MUSH, FRIED","Cornmeal mush, fried"
+56201530,"CORNMEAL MUSH, MADE W/ MILK","Cornmeal mush, made with milk"
+56201540,"CORNMEAL, MADE W/ MILK & SUGAR, P. R. STYLE","Cornmeal, made with milk and sugar, Puerto Rican Style (Harina de maiz)"
+56201550,"CORNMEAL DUMPLINGS","Cornmeal dumpling"
+56201560,"CORNMEAL STICKS, BOILED (INCL CORNMEAL GUANINES)","Cornmeal sticks, boiled"
+56201600,"CORNMEAL, LIME-TREATED, COOKED","Cornmeal, lime-treated, cooked (Masa harina)"
+56201700,"CORNSTARCH W/ MILK, EATEN AS CEREAL","Cornstarch with milk, eaten as a cereal (2 tbsp cornstarch in 2-1/2 cups milk)"
+56201750,"CORNSTARCH, DRY","Cornstarch, dry"
+56201800,"CORNSTARCH, HYDROLYZED, POWDER","Cornstarch, hydrolyzed powder"
+56201990,"MILLET, COOKED, NS AS TO FAT ADDED IN COOKING","Millet, cooked, NS as to fat added in cooking"
+56202000,"MILLET, COOKED, NO FAT ADDED","Millet, cooked, fat not added in cooking"
+56202100,"MILLET, COOKED, FAT ADDED IN COOKING","Millet, cooked, fat added in cooking"
+56202960,"OATMEAL,COOKED,NS AS TO REG,QUICK/INST,NS TO FAT","Oatmeal, cooked, NS as to regular, quick or instant; NS as to fat added in cooking"
+56202970,"OATMEAL, COOKED, QUICK, NS TO FAT ADDED","Oatmeal, cooked, quick (1 or 3 minutes), NS as to fat added in cooking"
+56202980,"OATMEAL, COOKED, REG, NS TO FAT ADDED","Oatmeal, cooked, regular, NS as to fat added in cooking"
+56203000,"OATMEAL, COOKED, NFS, NO FAT ADDED","Oatmeal, cooked, NS as to regular, quick or instant, fat not added in cooking"
+56203010,"OATMEAL, COOKED, REGULAR, NO FAT ADDED","Oatmeal, cooked, regular, fat not added in cooking"
+56203020,"OATMEAL,  COOKED, QUICK, NO FAT ADDED","Oatmeal, cooked, quick (1 or 3 minutes), fat not added in cooking"
+56203030,"OATMEAL, COOKED, INSTANT, NO FAT ADDED IN COOKING","Oatmeal, cooked, instant, fat not added in cooking"
+56203040,"OATMEAL, FAT ADDED IN COOKING, NFS","Oatmeal, cooked, NS as to regular, quick, or instant, fat added in cooking"
+56203050,"OATMEAL, REGULAR, FAT ADDED IN COOKING","Oatmeal, cooked, regular, fat added in cooking"
+56203060,"OATMEAL, QUICK, FAT ADDED IN COOKING","Oatmeal, cooked, quick (1 or 3 minutes), fat added in cooking"
+56203070,"OATMEAL, INSTANT, FAT ADDED","Oatmeal, cooked, instant, fat added in cooking"
+56203080,"OATMEAL, INSTANT, NS AS TO ADDED FAT","Oatmeal, cooked, instant, NS as to fat added in cooking"
+56203110,"OATMEAL, MAPLE FLAVOR, COOKED (INCL MAYPO)","Oatmeal with maple flavor, cooked"
+56203200,"OATMEAL, W/ FRUIT, COOKED","Oatmeal with fruit, cooked"
+56203210,"OATMEAL, NS TYPE, MADE W/ MILK, NO FAT ADDED","Oatmeal, NS as to regular, quick, or instant, made with milk, fat not added in cooking"
+56203211,"OATMEAL, CKD, REG, MADE W/ MILK, FAT NOT ADDED IN COOKING","Oatmeal, cooked, regular, made with milk, fat not added in cooking"
+56203212,"OATMEAL, CKD, QUICK, MADE W/ MILK, FAT NOT ADDED IN COOKING","Oatmeal, cooked, quick (1 or 3 minutes), made with milk, fat not added in cooking"
+56203213,"OATMEAL, CKD, INST, MADE W/ MILK, FAT NOT ADDED IN COOKING","Oatmeal, cooked, instant, made with milk, fat not added in cooking"
+56203220,"OATMEAL, NS TYPE, MADE W/ MILK, FAT ADDED","Oatmeal, NS as to regular, quick, or instant, made with milk, fat added in cooking"
+56203221,"OATMEAL, CKD, REG, MADE W/ MILK, FAT ADDED IN COOKING","Oatmeal, cooked, regular, made with milk, fat added in cooking"
+56203222,"OATMEAL, CKD, QUICK, MADE W/ MILK, FAT ADDED IN COOKING","Oatmeal, cooked, quick (1 or 3 minutes), made with milk, fat added in cooking"
+56203223,"OATMEAL, CKD, INST, MADE W/ MILK, FAT ADDED IN COOKING","Oatmeal, cooked, instant, made with milk, fat added in cooking"
+56203230,"OATMEAL, NS TYPE, MADE W/ MILK, NS AS TO ADDED FAT","Oatmeal, NS as to regular, quick, or instant, made with milk, NS as to fat added in cooking"
+56203231,"OATMEAL, CKD, REG, MADE W/ MILK, NS AS TO FAT ADDED","Oatmeal, cooked, regular, made with milk, NS as to fat added in cooking"
+56203232,"OATMEAL, CKD, QUICK, MADE W/ MILK, NS AS TO FAT ADDED","Oatmeal, cooked, quick (1 or 3 minutes), made with milk, NS as to fat added in cooking"
+56203233,"OATMEAL, CKD, INST, MADE W/ MILK, NS AS TO FAT ADDED","Oatmeal, cooked, instant, made with milk, NS as to fat added in cooking"
+56203540,"OATMEAL, MADE W/ MILK & SUGAR, P.R. STYLE","Oatmeal, made with milk and sugar, Puerto Rican style"
+56203600,"OATMEAL, MULTIGRAIN, COOKED, NS FAT ADDED","Oatmeal, multigrain, cooked, NS as to fat added in cooking"
+56203610,"OATMEAL, MULTIGRAIN, COOKED, FAT NOT ADDED","Oatmeal, multigrain, cooked, fat not added in cooking"
+56203620,"OATMEAL, MULTIGRAIN, COOKED, FAT ADDED","Oatmeal, multigrain, cooked, fat added in cooking"
+56204000,"QUINOA, COOKED, NS AS TO FAT ADDED IN COOKING","Quinoa, cooked, NS as to fat added in cooking"
+56204005,"QUINOA, COOKED, FAT NOT ADDED IN COOKING","Quinoa, cooked, fat not added in cooking"
+56204010,"QUINOA, COOKED, FAT ADDED IN COOKING","Quinoa, cooked, fat added in cooking"
+56205000,"RICE, COOKED, NS AS TO TYPE","Rice, cooked, NFS"
+56205002,"RICE, WHITE, COOKED, MADE WITH OIL","Rice, white, cooked, fat added in cooking, made with oil"
+56205004,"RICE, WHITE, COOKED, MADE WITH BUTTER","Rice, white, cooked, fat added in cooking, made with butter"
+56205006,"RICE, WHITE, COOKED, MADE WITH MARGARINE","Rice, white, cooked, fat added in cooking, made with margarine"
+56205008,"RICE, WHITE, COOKED, FAT NOT ADDED IN COOKING","Rice, white, cooked, fat not added in cooking"
+56205012,"RICE, BROWN, COOKED, MADE WITH OIL","Rice, brown, cooked, fat added in cooking, made with oil"
+56205014,"RICE, BROWN, COOKED, MADE WITH BUTTER","Rice, brown, cooked, fat added in cooking, made with butter"
+56205016,"RICE, BROWN, COOKED, MADE WITH MARGARINE","Rice, brown, cooked, fat added in cooking, made with margarine"
+56205018,"RICE, BROWN, COOKED, FAT NOT ADDED IN COOKING","Rice, brown, cooked, fat not added in cooking"
+56205050,"RICE, CREAM OF, COOKED, NO FAT ADDED","Rice, cream of, cooked, fat not added in cooking"
+56205060,"RICE, COOKED, W/ MILK","Rice, cooked, with milk"
+56205070,"RICE, SWEET, (RICE, COOKED, W/ HONEY)","Rice, sweet (rice, cooked, with honey)"
+56205080,"RICE, CREAMED, W/ MILK & SUGAR, PUERTO RICAN","Rice, creamed, made with milk and sugar, Puerto Rican style"
+56205090,"RICE, CREAM OF, COOKED, FAT ADDED IN COOKING","Rice, cream of, cooked, fat added in cooking"
+56205130,"YELLOW RICE, COOKED, NS AS TO FAT ADDED IN COOKING","Yellow rice, cooked, NS as to fat added in cooking"
+56205150,"YELLOW RICE, COOKED, FAT NOT ADDED IN COOKING","Yellow rice, cooked, fat not added in cooking"
+56205170,"YELLOW RICE, COOKED, FAT ADDED IN COOKING","Yellow rice, cooked, fat added in cooking"
+56205190,"RICE, WHITE, COOKED, GLUTINOUS (INCL STICKY RICE)","Rice, white, cooked, glutinous"
+56205200,"RICE, FRZ DES,NONDAIRY,NOT CHOC (INCL RICE DREAM)","Rice, frozen dessert, nondairy, flavors other than chocolate"
+56205205,"RICE, WILD, 100%, COOKED, NS AS TO FAT ADDED IN COOKING","Rice, wild, 100%, cooked, NS as to fat added in cooking"
+56205210,"RICE, WILD, 100%, COOKED, NO FAT ADDED","Rice, wild, 100%, cooked, fat not added in cooking"
+56205215,"RICE, WILD, 100%, COOKED, FAT ADDED IN COOKING","Rice, wild, 100%, cooked, fat added in cooking"
+56205230,"RICE DESSERT BAR,FRZ,NOT CHOC,NONDAIRY,CAROB,COVER","Rice dessert bar, frozen, flavors other than chocolate, nondairy, carob covered"
+56205240,"RICE DESSERT BAR,FRZ,CHOC,NONDAIRY,CHOC COVERED","Rice dessert bar, frozen, chocolate, nondairy, chocolate covered"
+56205300,"RICE, WHITE & WILD, COOKED, NO FAT ADDED","Rice, white and wild, cooked, fat not added in cooking"
+56205310,"RICE, BROWN & WILD, COOKED, NO FAT ADDED","Rice, brown and wild, cooked, fat not added in cooking"
+56205320,"RICE, WHITE & WILD, FAT ADDED","Rice, white and wild, cooked, fat added in cooking"
+56205330,"RICE, WHITE & WILD, NS AS TO ADDED FAT","Rice, white and wild, cooked, NS as to fat added in cooking"
+56205340,"RICE, BROWN & WILD, FAT ADDED","Rice, brown and wild, cooked, fat added in cooking"
+56205350,"RICE, BROWN & WILD, NS AS TO ADDED FAT","Rice, brown and wild, cooked, NS as to fat added in cooking"
+56205410,"RICE, WHITE, COOKED W/ (FAT) OIL, PUERTO RICAN STYLE","Rice, white, cooked with (fat) oil, Puerto Rican style (Arroz blanco)"
+56206970,"WHEAT, CREAM OF,COOKED,QUICK,NS AS TO ADDED FAT","Wheat, cream of, cooked, quick, NS as to fat added in cooking"
+56206980,"WHEAT, CREAM OF,COOKED,REG,NS AS TO ADDED FAT","Wheat, cream of, cooked, regular, NS as to fat added in cooking"
+56206990,"WHEAT, CREAM OF,COOKED,NS AS REG,QUICK,/INST","Wheat, cream of, cooked, NS as to regular, quick, or instant, NS as to fat added in cooking"
+56207000,"WHEAT, CREAM OF, COOKED, NFS, NO FAT ADDED","Wheat, cream of, cooked, NS as to regular, quick, or instant, fat not added in cooking"
+56207010,"WHEAT, CREAM OF, COOKED, REGULAR, NO FAT ADDED","Wheat, cream of, cooked, regular, fat not added in cooking"
+56207020,"WHEAT, CREAM OF, COOKED, QUICK, NO FAT ADDED","Wheat, cream of, cooked, quick, fat not added in cooking"
+56207030,"WHEAT, CREAM OF, COOKED, INSTANT, NO FAT ADDED","Wheat, cream of, cooked, instant, fat not added in cooking"
+56207040,"WHEAT, CREAM OF, MADE W/ MILK","Wheat, cream of, cooked, made with milk"
+56207050,"WHEAT, CREAM OF, MADE W/ MILK & SUGAR, P.R. STYLE","Wheat, cream of, cooked, made with milk and sugar, Puerto Rican style"
+56207060,"WHEAT, CREAM OF, INSTANT, COOKED, FAT ADDED","Wheat, cream of, cooked, instant, fat added in cooking"
+56207070,"WHEAT, CREAM OF, INSTANT,COOKED, NS AS TO ADDED FAT","Wheat, cream of, cooked, instant, NS as to fat added in cooking"
+56207080,"WHEAT, CREAM OF,COOKED,NS AS TO REG,QUICK, OR INST","Wheat, cream of, cooked, NS as to regular, quick, or instant, fat added in cooking"
+56207100,"WHEAT, ROLLED, COOKED, NO FAT ADDED","Wheat, rolled, cooked, fat not added in cooking"
+56207110,"BULGUR, COOKED OR CANNED, NO FAT ADDED","Bulgur, cooked or canned, fat not added in cooking"
+56207120,"BULGAR, COOKED OR CANNNED, FAT ADDED IN COOKING","Bulgur, cooked or canned, fat added in cooking"
+56207130,"BULGUR, COOKED OR CANNED, NS AS TO ADDED FAT","Bulgur, cooked or canned, NS as to fat added in cooking"
+56207140,"WHEAT ROLLED,COOKED,NS AS TO ADDED FAT","Wheat, rolled, cooked, NS as to fat added in cooking"
+56207150,"COUSCOUS, PLAIN, COOKED, FAT NOT ADDED IN COOKING","Couscous, plain, cooked, fat not added in cooking"
+56207160,"COUSCOUS, PLAIN, COOKED, NS AS TO ADDED FAT","Couscous, plain, cooked, NS as to fat added in cooking"
+56207180,"COUSCOUS, PLAIN, COOKED, FAT ADDED IN COOKING","Couscous, plain, cooked, fat added in cooking"
+56207190,"WHOLE WHEAT CEREAL, COOKED, NS AS TO ADDED FAT","Whole wheat cereal, cooked, NS as to fat added in cooking"
+56207200,"WHOLE WHEAT CEREAL, COOKED, NO FAT ADDED","Whole wheat cereal, cooked, fat not added in cooking"
+56207210,"WHOLE WHEAT CEREAL, COOKED, FAT ADDED","Whole wheat cereal, cooked, fat added in cooking"
+56207220,"WHEAT, CREAM OF, COOKED, REGULAR, FAT ADDED","Wheat, cream of, cooked, regular, fat added in cooking"
+56207230,"WHEAT, CREAM OF, COOKED,QUICK,FAT ADDED IN COOKING","Wheat, cream of, cooked, quick, fat added in cooking"
+56207300,"WHOLE WHEAT CEREAL, W/ BARLEY, COOKED, NO FAT ADDED","Whole wheat cereal, wheat and barley, cooked, fat not added in cooking"
+56207330,"WHOLE WHEAT CEREAL, WHEAT & BARLEY, FAT ADDED","Whole wheat cereal, wheat and barley, cooked, fat added in cooking"
+56207340,"WHOLE WHEAT CEREAL, WHEAT & BARLEY, ADDED FAT NS","Whole wheat cereal, wheat and barley, cooked, NS as to fat added in cooking"
+56207350,"WHEAT CEREAL, CHOC FLAVORED, COOKED W/ MILK","Wheat cereal, chocolate flavored, cooked, made with milk"
+56207360,"WHEAT CEREAL, CHOC FLAVORED, COOKED, NO FAT ADDED","Wheat cereal, chocolate flavored, cooked, fat not added in cooking"
+56207370,"WHEAT CEREAL, CHOC FLAV,COOKED,NS AS TO ADDED FAT","Wheat cereal, chocolate flavored, cooked, NS as to fat added in cooking"
+56208500,"OAT BRAN CEREAL, COOKED, NO FAT ADDED","Oat bran cereal, cooked, fat not added in cooking"
+56208510,"OAT BRAN CEREAL, COOKED, FAT ADDED","Oat bran cereal, cooked, fat added in cooking"
+56208520,"OAT BRAN CEREAL, COOKED, NS AS TO ADDED FAT","Oat bran cereal, cooked, NS as to fat added in cooking"
+56208530,"OAT BRAN CEREAL, MADE W/ MILK, NO FAT ADDED","Oat bran cereal, cooked, made with milk, fat not added in cooking"
+56208540,"OAT BRAN CEREAL, MADE W/ MILK, FAT ADDED","Oat bran cereal, cooked, made with milk, fat added in cooking"
+56208550,"OAT BRAN CEREAL, MADE W/ MILK, NS AS TO ADDED FAT","Oat bran cereal, cooked, made with milk, NS as to fat added in cooking"
+56209000,"RYE, CREAM OF, COOKED","Rye, cream of, cooked"
+56210000,"NESTUM, CEREAL","Nestum cereal"
+5.7e+07,"CEREAL, NFS","Cereal, NFS"
+57000050,"KASHI CEREAL, NS AS TO READY-TO-EAT OR COOKED","Kashi cereal, NS as to ready to eat or cooked"
+57000100,"OAT CEREAL, NFS","Oat cereal, NFS"
+57100100,"CEREAL, READY-TO-EAT, NFS","Cereal, ready-to-eat, NFS"
+57101000,"ALL-BRAN CEREAL","All-Bran"
+57102000,"ALPEN CEREAL","Alpen"
+57103000,"ALPHA-BITS CEREAL","Alpha-Bits"
+57103020,"ALPHA-BITS W/ MARSHMALLOWS CEREAL","Alpha-bits with marshmallows"
+57103050,"AMARANTH FLAKES CEREAL","Amaranth Flakes"
+57103100,"APPLE CINNAMON CHEERIOS","Apple Cinnamon Cheerios"
+57104000,"APPLE JACKS CEREAL","Apple Jacks"
+57106050,"BANANA NUT CRUNCH CEREAL (POST)","Banana Nut Crunch Cereal (Post)"
+57106060,"BANANA NUT CHEERIOS","Banana Nut Cheerios"
+57106100,"BASIC 4 (RTE CEREAL)","Basic 4"
+57106250,"BERRY BERRY KIX","Berry Berry Kix"
+57106260,"BERRY BURST CHEERIOS","Berry Burst Cheerios"
+57106530,"BLUEBERRY MORNING, POST","Blueberry Morning, Post"
+57107000,"BOOBERRY CEREAL","Booberry"
+57110000,"ALL-BRAN BRAN BUDS CEREAL, KELLOGG'S (FORMERLY BRAN BUDS)","All-Bran Bran Buds, Kellogg's (formerly Bran Buds)"
+57111000,"BRAN CHEX CEREAL","Bran Chex"
+57117000,"CAP'N CRUNCH CEREAL","Cap'n Crunch"
+57117500,"CAP'N CRUNCH'S CHRISTMAS CRUNCH CEREAL","Cap'n Crunch's Christmas Crunch"
+57119000,"CAP'N CRUNCH'S CRUNCH BERRIES CEREAL","Cap'n Crunch's Crunch Berries"
+57120000,"CAP'N CRUNCH'S PEANUT BUTTER CRUNCH CEREAL","Cap'n Crunch's Peanut Butter Crunch"
+57123000,"CHEERIOS","Cheerios"
+57124000,"CHEX CEREAL, NFS","Chex cereal, NFS"
+57124050,"CHEX CINNAMON","Chex Cinnamon"
+57124100,"CHOCOLATE CHEERIOS","Chocolate Cheerios"
+57124200,"CHOCOLATE FLAVORED FROSTED PUFFED CORN CEREAL","Chocolate flavored frosted puffed corn cereal"
+57124300,"CHOCOLATE LUCKY CHARMS","Chocolate Lucky Charms"
+57125000,"CINNAMON TOAST CRUNCH CEREAL","Cinnamon Toast Crunch"
+57125010,"CINNAMON TOAST CRUNCH REDUCED SUGAR","Cinnamon Toast Crunch Reduced Sugar"
+57125900,"HONEY NUT CLUSTERS CEREAL","Honey Nut Clusters (formerly called Clusters)"
+57126000,"COCOA KRISPIES CEREAL","Cocoa Krispies"
+57127000,"COCOA PEBBLES CEREAL","Cocoa Pebbles"
+57128000,"COCOA PUFFS CEREAL","Cocoa Puffs"
+57128005,"COCOA PUFFS, REDUCED SUGAR","Cocoa Puffs, reduced sugar"
+57130000,"COOKIE-CRISP CEREAL (INCLUDE ALL FLAVORS)","Cookie-Crisp"
+57131000,"CRUNCHY CORN BRAN CEREAL, QUAKER","Crunchy Corn Bran, Quaker"
+57132000,"CORN CHEX CEREAL","Corn Chex"
+57134000,"CORN FLAKES, NFS (INCLUDE STORE BRANDS)","Corn flakes, NFS"
+57134090,"CORN FLAKES, LOW SODIUM","Corn flakes, low sodium"
+57135000,"CORN FLAKES, KELLOGG'S","Corn flakes, Kellogg's"
+57137000,"CORN PUFFS CEREAL","Corn Puffs"
+57139000,"COUNT CHOCULA CEREAL","Count Chocula"
+57143000,"CRACKLIN' OAT BRAN CEREAL","Cracklin' Oat Bran"
+57143500,"CRANBERRY ALMOND CRUNCH, POST","Cranberry Almond Crunch, Post"
+57144000,"CRISP CRUNCH CEREAL","Crisp Crunch"
+57148000,"CRISPIX CEREAL","Crispix"
+57148500,"CRISPY BROWN RICE CEREAL","Crispy Brown Rice Cereal"
+57151000,"CRISPY RICE CEREAL","Crispy Rice"
+57201900,"DORA THE EXPLORER CEREAL","Dora the Explorer Cereal"
+57206000,"FAMILIA CEREAL","Familia"
+57206700,"FIBER ONE CEREAL","Fiber One"
+57206705,"FIBER ONE CARAMEL DELIGHT","Fiber One Caramel Delight"
+57206710,"FIBER ONE HONEY CLUSTERS","Fiber One Honey Clusters"
+57206715,"FIBER ONE RAISIN BRAN CLUSTERS","Fiber One Raisin Bran Clusters"
+57206800,"FIBER 7 FLAKES CEREAL, HEALTH VALLEY","Fiber 7 Flakes, Health Valley"
+57207000,"BRAN FLAKES CEREAL, NFS (FORMERLY 40% BRAN FLAKES, NFS)","Bran Flakes, NFS (formerly 40% Bran Flakes, NFS)"
+57208000,"ALL-BRAN COMPLETE WHEAT FLAKES, KELLOGG'S","All-Bran Complete Wheat Flakes, Kellogg's"
+57209000,"NATURAL BRAN FLAKES CEREAL, POST","Natural Bran Flakes, Post (formerly called 40% Bran Flakes, Post)"
+57211000,"FRANKENBERRY CEREAL","Frankenberry"
+57213000,"FROOT LOOPS CEREAL","Froot Loops"
+57213850,"FROSTED CHEERIOS CEREAL","Frosted Cheerios"
+57214000,"FROSTED MINI-WHEATS CEREAL (INCL ALL FLAVORS)","Frosted Mini-Wheats"
+57214100,"FROSTED WHEAT BITES","Frosted Wheat Bites"
+57215000,"FROSTY O'S CEREAL","Frosty O's"
+57216000,"FROSTED RICE CEREAL, NFS","Frosted rice, NFS"
+57218000,"FROSTED RICE KRISPIES, KELLOGG'S","Frosted Rice Krispies, Kellogg's"
+57219000,"FRUIT & FIBRE CEREAL, NFS","Fruit & Fibre (fiber), NFS"
+57221000,"FRUIT & FIBRE CEREAL, W/ DATES, RAISINS, & WALNUTS","Fruit & Fibre (fiber) with dates, raisins, and walnuts"
+57221650,"FRUIT HARVEST CEREAL, KELLOGG'S","Fruit Harvest cereal, Kellogg's"
+57221700,"FRUIT RINGS, NFS (INCLUDE STORE BRANDS)","Fruit Rings, NFS"
+57221800,"FRUIT WHIRLS CEREAL","Fruit Whirls"
+57221810,"FRUITY CHEERIOS","Fruity Cheerios"
+57223000,"FRUITY PEBBLES CEREAL","Fruity Pebbles"
+57224000,"GOLDEN GRAHAMS CEREAL","Golden Grahams"
+57227000,"GRANOLA, NFS","Granola, NFS"
+57228000,"GRANOLA, HOMEMADE","Granola, homemade"
+57229000,"GRANOLA, LOWFAT, KELLOGG'S","Granola, lowfat, Kellogg's"
+57229500,"GRANOLA W/ RAISINS, LOWFAT, KELLOGG'S","Granola with Raisins, lowfat, Kellogg's"
+57230000,"GRAPE-NUTS CEREAL","Grape-Nuts"
+57231000,"GRAPE-NUTS FLAKES","Grape-Nuts Flakes"
+57231100,"GRAPE-NUTS TRAIL MIX CRUNCH","Grape-Nuts Trail Mix Crunch"
+57231200,"GREAT GRAINS, RAISIN, DATE, & PECAN,WHOLE GRAIN CEREAL, POST","Great Grains, Raisin, Date, and Pecan Whole Grain Cereal, Post"
+57231250,"GREAT GRAINS DOUBLE PECAN WHOLE GRAIN CEREAL, POST","Great Grains Double Pecan Whole Grain Cereal, Post"
+57237100,"HONEY BUNCHES OF OATS HONEY ROASTED CEREAL","Honey Bunches of Oats Honey Roasted Cereal"
+57237200,"HONEY BUNCHES OF OATS WITH VANILLA CLUSTERS, POST","Honey Bunches of Oats with Vanilla Clusters, Post"
+57237300,"HONEY BUNCHES OF OATS W/ ALMONDS, POST","Honey Bunches of Oats with Almonds, Post"
+57237310,"HONEY BUNCHES OF OATS WITH PECAN BUNCHES","Honey Bunches of Oats with Pecan Bunches"
+57237900,"HONEY BUNCHES OF OATS JUST BUNCHES","Honey Bunches of Oats Just Bunches"
+57238000,"HONEYCOMB CEREAL, PLAIN","Honeycomb, plain"
+57239000,"HONEYCOMB CEREAL, STRAWBERRY","Honeycomb, strawberry"
+57239100,"HONEY CRUNCH CORN FLAKES CEREAL, KELLOGG'S","Honey Crunch Corn Flakes, Kellogg's"
+57240100,"HONEY NUT CHEX CEREAL","Honey Nut Chex"
+57241000,"HONEY NUT CHEERIOS","Honey Nut Cheerios"
+57241200,"HONEY NUT SHREDDED WHEAT CEREAL, POST","Honey Nut Shredded Wheat, Post"
+57243000,"HONEY SMACKS, KELLOGG'S","Honey Smacks, Kellogg's (formerly Smacks; Honey Smacks)"
+57301500,"KASHI, PUFFED","Kashi, Puffed"
+57301505,"KASHI AUTUMN WHEAT","Kashi Autumn Wheat"
+57301510,"KASHI GOLEAN","Kashi GOLEAN"
+57301511,"KASHI GOLEAN CRUNCH","Kashi GOLEAN Crunch"
+57301512,"KASHI GOLEAN CRUNCH HONEY ALMOND FLAX","Kashi GOLEAN Crunch Honey Almond Flax"
+57301520,"KASHI GOOD FRIENDS","Kashi Good Friends"
+57301530,"KASHI HEART TO HEART HONEY TOASTED OAT","Kashi Heart to Heart Honey Toasted Oat"
+57301535,"KASHI HEART TO HEART OAT FLAKES AND BLUEBERRY CLUSTERS","Kashi Heart to Heart Oat Flakes and Blueberry Clusters"
+57301540,"KASHI HONEY SUNSHINE","Kashi Honey Sunshine"
+57302100,"KING VITAMAN CEREAL","King Vitaman"
+57303100,"KIX CEREAL","Kix"
+57303105,"HONEY KIX","Honey Kix"
+57304100,"LIFE CEREAL (PLAIN & CINNAMON)","Life (plain and cinnamon)"
+57305100,"LUCKY CHARMS CEREAL","Lucky Charms"
+57305150,"FROSTED OAT CEREAL W/ MARSHMALLOWS","Frosted oat cereal with marshmallows"
+57305160,"MALT-O-MEAL BLUEBERRY MUFFIN TOPS","Malt-O-Meal Blueberry Muffin Tops"
+57305165,"MALT-O-MEAL CINNAMON TOASTERS","Malt-O-Meal Cinnamon Toasters"
+57305170,"MALT-O-MEAL COCO-ROOS CEREAL","Malt-O-Meal Coco-Roos"
+57305174,"MALT-O-MEAL COLOSSAL CRUNCH","Malt-O-Meal Colossal Crunch"
+57305175,"MALT-O-MEAL COCOA DYNO-BITES","Malt-O-Meal Cocoa Dyno-Bites"
+57305180,"MALT-O-MEAL CORN BURSTS CEREAL","Malt-O-Meal Corn Bursts"
+57305200,"MALT-O-MEAL CRISPY RICE CEREAL","Malt-O-Meal Crispy Rice"
+57305210,"MALT-O-MEAL FROSTED FLAKES","Malt-O-Meal Frosted Flakes"
+57305215,"MALT-O-MEAL FROSTED MINI SPOONERS","Malt-O-Meal Frosted Mini Spooners"
+57305300,"MALT-O-MEAL FRUITY DYNO-BITES","Malt-O-Meal Fruity Dyno-Bites"
+57305400,"MALT-O-MEAL HONEY GRAHAM SQUARES","Malt-O-Meal Honey Graham Squares"
+57305500,"MALT-O-MEAL HONEY & NUT TOASTY O'S CEREAL","Malt-O-Meal Honey and Nut Toasty O's"
+57305600,"MALT-O-MEAL MARSHMALLOW MATEYS CEREAL","Malt-O-Meal Marshmallow Mateys"
+57306100,"MALT-O-MEAL PUFFED RICE CEREAL","Malt-O-Meal Puffed Rice"
+57306120,"MALTO-O-MEAL PUFFED WHEAT CEREAL","Malt-O-Meal Puffed Wheat"
+57306130,"MALT-O-MEAL RAISIN BRAN","Malt-O-Meal Raisin Bran"
+57306500,"MALT-O-MEAL GOLDEN PUFFS CEREAL (FORMERLY SUGAR PUFFS)","Malt-O-Meal Golden Puffs (formerly Sugar Puffs)"
+57306700,"MALT-O-MEAL TOASTED OAT CEREAL","Malt-O-Meal Toasted Oat Cereal"
+57306800,"MALT-O-MEAL TOOTIE FRUITIES (RTE CEREAL)","Malt-O-meal Tootie Fruities"
+57307010,"MAPLE PECAN CRUNCH CEREAL, POST","Maple Pecan Crunch Cereal, Post"
+57307500,"MILLET, PUFFED (CEREAL)","Millet, puffed"
+57307600,"MINI-SWIRLZ CINNAMON BUN CEREAL, KELLOGG'S","Mini-Swirlz Cinnamon Bun Cereal, Kellogg's"
+57308150,"MUESLIX CEREAL, NFS","Mueslix cereal, NFS"
+57308190,"MUESLI, DRIED FRUIT&NUTS","Muesli, dried fruit and nuts (formerly Muesli with raisins, dates, and almonds)"
+57308300,"MULTI BRAN CHEX","Multi Bran Chex"
+57308400,"MULTIGRAIN CHEERIOS","MultiGrain Cheerios"
+57309100,"NATURE VALLEY GRANOLA, W/ FRUIT & NUTS","Nature Valley Granola, with fruit and nuts"
+57316200,"NUTTY NUGGETS (RALSTON)","Nutty Nuggets, Ralston Purina"
+57316300,"OAT BRAN FLAKES, HEALTH VALLEY","Oat Bran Flakes, Health Valley"
+57316380,"OAT CLUSTER CHEERIOS CRUNCH","Oat Cluster Cheerios Crunch"
+57316450,"OATMEAL CRISP W/ ALMONDS CEREAL","Oatmeal Crisp with Almonds"
+57316500,"OATMEAL CRISP, RAISIN","Oatmeal Crisp, Raisin (formerly Oatmeal Raisin Crisp)"
+57316710,"OH'S, HONEY GRAHAM CEREAL","Oh's, Honey Graham"
+57319000,"100% NATURAL CEREAL, PLAIN, QUAKER","100% Natural Cereal, plain, Quaker"
+57319500,"SUN COUNTRY 100% NATURAL GRANOLA, WITH ALMONDS","Sun Country 100% Natural Granola, with Almonds"
+57320500,"100 % NATURAL CEREAL, W/ OATS,HONEY & RAISINS,QUAKER","100 % Natural Cereal, with oats, honey and raisins, Quaker"
+57321500,"100% NATURAL WHOLEGRAIN CEREAL W/ RAISINS, LOWFAT, QUAKER","100 % Natural Wholegrain Cereal with raisins, lowfat, Quaker"
+57321700,"OPTIMUM, NATURE'S PATH","Optimum, Nature's Path"
+57321800,"OPTIMUM SLIM, NATURE'S PATH","Optimum Slim, Nature's Path"
+57321900,"ORGANIC FLAX PLUS, NATURE'S PATH","Organic Flax Plus, Nature's Path"
+57323000,"SWEET CRUNCH CEREAL, QUAKER (FORMERLY POPEYE)","Sweet Crunch, Quaker (formerly called Popeye)"
+57325000,"PRODUCT 19 CEREAL","Product 19"
+57326000,"PUFFINS CEREAL","Puffins Cereal"
+57327450,"QUAKER OAT BRAN CEREAL","Quaker Oat Bran Cereal"
+57327500,"QUAKER OATMEAL SQUARES CEREAL (FORMERLY QUAKER OAT SQUARES)","Quaker Oatmeal Squares (formerly Quaker Oat Squares)"
+57328000,"QUISP CEREAL","Quisp"
+57329000,"RAISIN BRAN CEREAL, NFS","Raisin bran, NFS"
+57330000,"RAISIN BRAN, KELLOGG'S","Raisin Bran, Kellogg's"
+57330010,"RAISIN BRAN CRUNCH, KELLOGG'S","Raisin Bran Crunch, Kellogg's"
+57331000,"RAISIN BRAN CEREAL, POST","Raisin Bran, Post"
+57332050,"RAISIN BRAN, TOTAL","Raisin Bran, Total"
+57332100,"RAISIN NUT BRAN CEREAL","Raisin Nut Bran"
+57335550,"REESE'S PEANUT BUTTER PUFFS CEREAL","Reese's Peanut Butter Puffs cereal"
+57336000,"RICE CHEX CEREAL","Rice Chex"
+57337000,"RICE FLAKES, NFS","Rice Flakes, NFS"
+57339000,"RICE KRISPIES, KELLOGG'S","Rice Krispies, Kellogg's"
+57339500,"RICE KRISPIES TREATS CEREAL, KELLOGG'S","Rice Krispies Treats Cereal, Kellogg's"
+57340000,"PUFFED RICE CEREAL","Rice, puffed"
+57341000,"SHREDDED WHEAT 'N BRAN CEREAL","Shredded Wheat'N Bran"
+57341200,"SMART START STRONG HEART ANTIOXIDANTS CEREAL, KELLOGG'S","Smart Start Strong Heart Antioxidants Cereal, Kellogg's"
+57342010,"SMORZ, KELLOGG'S","Smorz, Kellogg's"
+57344000,"SPECIAL K CEREAL","Special K"
+57344001,"SPECIAL K BLUEBERRY","Special K Blueberry"
+57344005,"SPECIAL K CHOCOLATEY DELIGHT","Special K Chocolatey Delight"
+57344007,"SPECIAL K LOW FAT GRANOLA","Special K Low Fat Granola"
+57344010,"SPECIAL K RED BERRIES","Special K Red Berries"
+57344015,"SPECIAL K FRUIT & YOGURT","Special K Fruit & Yogurt"
+57344020,"SPECIAL K VANILLA ALMOND","Special K Vanilla Almond"
+57344025,"SPECIAL K CINNAMON PECAN, KELLOGG'S","Special K Cinnamon Pecan, Kellogg's"
+57346500,"OATMEAL HONEY NUT HEAVEN, QUAKER","Oatmeal Honey Nut Heaven, Quaker (formerly Toasted Oatmeal, Honey Nut)"
+57347000,"CORN POPS CEREAL","Corn Pops"
+57348000,"FROSTED CORN FLAKES, NFS","Frosted corn flakes, NFS"
+57349000,"FROSTED FLAKES, KELLOGG'S","Frosted Flakes, Kellogg's"
+57349020,"REDUCED SUGAR FROSTED FLAKES CEREAL, KELLOGG'S","Reduced Sugar Frosted Flakes Cereal, Kellogg's"
+57355000,"GOLDEN CRISP CEREAL","Golden Crisp (Formerly called Super Golden Crisp)"
+57401100,"TOASTED OAT CEREAL","Toasted oat cereal"
+57403100,"TOASTIES, POST","Toasties, Post"
+57406100,"TOTAL CEREAL","Total"
+57406105,"TOTAL CRANBERRY CRUNCH","Total Cranberry Crunch"
+57407100,"TRIX CEREAL","Trix"
+57407110,"TRIX, REDUCED SUGAR","Trix, reduced sugar"
+57408100,"UNCLE SAM CEREAL","Uncle Sam Cereal (formerly Uncle Sam's Hi Fiber Cereal)"
+57409100,"WAFFLE CRISP CEREAL, POST","Waffle Crisp, Post"
+57410000,"WEETABIX WHOLE WHEAT CEREAL","Weetabix Whole Wheat Cereal"
+57411000,"WHEAT CHEX CEREAL","Wheat Chex"
+57412000,"WHEAT GERM CEREAL, PLAIN","Wheat germ, plain"
+57413000,"WHEAT GERM CEREAL, W/ SUGAR & HONEY","Wheat germ, with sugar and honey"
+57416000,"PUFFED WHEAT CEREAL, PLAIN","Wheat, puffed, plain"
+57416010,"WHEAT, PUFFED, PRESWEETENED W/ SUGAR","Wheat, puffed, presweetened with sugar"
+57417000,"SHREDDED WHEAT, 100%","Shredded Wheat, 100%"
+57418000,"WHEATIES CEREAL","Wheaties"
+57419000,"YOGURT BURST CHEERIOS","Yogurt Burst Cheerios"
+57601100,"WHEAT BRAN, UNPROCESSED","Wheat bran, unprocessed"
+57602100,"OATS, RAW","Oats, raw"
+57602500,"OAT BRAN, UNCOOKED","Oat bran, uncooked"
+57603100,"RICE POLISHINGS","Rice polishings"
+57603200,"RICE BRAN CEREAL, UNCOOKED","Rice bran, uncooked"
+57604100,"WHOLE WHEAT, CRACKED","Whole wheat, cracked"
+57801000,"BARLEY CEREAL, BABY, DRY, INSTANT","Barley cereal, baby food, dry, instant"
+57803000,"MIXED CEREAL, BABY, DRY, INSTANT","Mixed cereal, baby food, dry, instant"
+57804000,"OATMEAL CEREAL, BABY, DRY, INSTANT","Oatmeal cereal, baby food, dry, instant"
+57805000,"RICE CEREAL, BABY, DRY, INSTANT","Rice cereal, baby food, dry, instant"
+57805080,"RICE CEREAL W/ APPLES, BABY, DRY, INSTANT","Rice cereal with apples, baby food, dry, instant"
+57805090,"RICE CEREAL WITH MIXED FRUITS, BABY FOOD, DRY, INSTANT","Rice cereal with mixed fruits, baby food, dry, instant"
+57805100,"RICE CEREAL W/ BANANAS, BABY, DRY, INSTANT","Rice cereal with bananas, baby food, dry, instant"
+57805500,"BROWN RICE CEREAL, BABY FOOD, DRY, INSTANT","Brown rice cereal, baby food, dry, instant"
+57806000,"MIXED CEREAL W/ BANANAS, BABY, DRY, INSTANT","Mixed cereal with bananas, baby food, dry, instant"
+57806050,"MULTIGRAIN, WHOLE GRAIN CEREAL, BABY FOOD, DRY, INSTANT","Multigrain, whole grain cereal, baby food, dry, instant"
+57806100,"OATMEAL CEREAL W/ BANANAS, BABY, DRY, INSTANT","Oatmeal cereal with bananas, baby food, dry, instant"
+57806200,"OATMEAL W/ FRUIT, BABY, DRY, INSTANT, TODDLER","Oatmeal cereal with fruit, baby food, dry, instant, toddler"
+57807010,"WHOLE WHEAT CEREAL W/ APPLES, BABY, DRY, INSTANT","Whole wheat cereal with apples, baby food, dry, instant"
+57820000,"CEREAL, BABY, JARRED, NFS","Cereal, baby food, jarred, NFS"
+57820100,"RICE CEREAL, BABY FOOD, JARRED, NFS","Rice cereal, baby food, jarred, NFS"
+57822000,"MIXED CEREAL W/ APPLESAUCE & BANANAS, BABY, JARRED","Mixed cereal with applesauce and bananas, baby food, jarred"
+57823000,"OATMEAL W/ APPLESAUCE & BANANAS, BABY, JARRED","Oatmeal with applesauce and bananas, baby food, jarred"
+57824000,"RICE CEREAL, W/ APPLESAUCE & BANANAS, BABY, JARRED","Rice cereal with applesauce and bananas, baby food, jarred"
+57824500,"RICE CEREAL W/ MIXED FRUIT, BABY, JARRED","Rice cereal with mixed fruit, baby food, jarred"
+57830100,"GERBER GRADUATES FINGER SNACKS CEREAL,BABY FOOD","Gerber Graduates Finger Snacks Cereal, baby food"
+58100000,"BURRITO, TACO, OR QUESADILLA W/ EGG","Burrito, taco, or quesadilla with egg"
+58100005,"BURRITO, TACO, OR QUESADILLA W/ EGG & POTATO","Burrito, taco, or quesadilla with egg and potato"
+58100010,"BURRITO, TACO, OR QUESADILLA W/ EGG & BREAKFAST MEAT","Burrito, taco, or quesadilla with egg and breakfast meat"
+58100013,"BURRITO, TACO, OR QUESADILLA WITH EGG AND BREAKFAST MEAT, FF","Burrito, taco, or quesadilla with egg and breakfast meat, from fast food"
+58100015,"BURRITO, TACO, OR QUESADILLA W/EGG, POTATO, & BREAKFAST MEAT","Burrito, taco, or quesadilla with egg, potato, and breakfast meat"
+58100017,"BURRITO, TACO, OR QUESADILLA WITH EGG, POTATO, BRK MEAT, FF","Burrito, taco, or quesadilla with egg, potato, and breakfast meat, from fast food"
+58100020,"BURRITO, TACO, OR QUESADILLA W/EGG, BEANS,& BREAKFAST MEAT","Burrito, taco, or quesadilla with egg, beans, and breakfast meat"
+58100100,"BURRITO W/ MEAT","Burrito with meat"
+58100120,"BURRITO W/ MEAT & BEANS","Burrito with meat and beans"
+58100125,"BURRITO WITH MEAT AND BEANS, FROM FAST FOOD","Burrito with meat and beans, from fast food"
+58100135,"BURRITO W/ MEAT & SOUR CREAM","Burrito with meat and sour cream"
+58100140,"BURRITO W/ MEAT, BEANS, & SOUR CREAM","Burrito with meat, beans, and sour cream"
+58100145,"BURRITO WITH MEAT, BEANS, AND SOUR CREAM, FROM FAST FOOD","Burrito with meat, beans, and sour cream, from fast food"
+58100160,"BURRITO W/ MEAT, BEANS, & RICE","Burrito with meat, beans, and rice"
+58100165,"BURRITO W/ MEAT, BEANS, RICE, & SOUR CREAM","Burrito with meat, beans, rice, and sour cream"
+58100200,"BURRITO W/ CHICKEN","Burrito with chicken"
+58100220,"BURRITO W/ CHICKEN AND BEANS","Burrito with chicken and beans"
+58100235,"BURRITO W/ CHICKEN & SOUR CREAM","Burrito with chicken and sour cream"
+58100245,"BURRITO W/ CHICKEN, BEANS, & SOUR CREAM","Burrito with chicken, beans, and sour cream"
+58100255,"BURRITO W/ CHICKEN, BEANS, & RICE","Burrito with chicken, beans, and rice"
+58100260,"BURRITO W/ CHICKEN, BEANS, RICE, & SOUR CREAM","Burrito with chicken, beans, rice, and sour cream"
+58100300,"BURRITO W/ BEANS & RICE, MEATLESS","Burrito with beans and rice, meatless"
+58100320,"BURRITO W/ BEANS","Burrito with beans, meatless"
+58100325,"BURRITO WITH BEANS, MEATLESS, FROM FAST FOOD","Burrito with beans, meatless, from fast food"
+58100330,"BURRITO W/ BEANS, RICE, & SOUR CREAM, MEATLESS","Burrito with beans, rice, and sour cream, meatless"
+58100360,"CHILAQUILES,TORTILLA CASSEROLE W/ SALSA,CHEESE, EGG","Chilaquiles, tortilla casserole with salsa, cheese, and egg"
+58100370,"CHILAQUILES, TORTILLA CASSEROLE, NO EGG","Chilaquiles, tortilla casserole with salsa and cheese, no egg"
+58100520,"ENCHILADA W/ MEAT & BEANS, RED-CHILE OR ENCHILADA SAUCE","Enchilada with meat and beans, red-chile or enchilada sauce"
+58100525,"ENCHILADA WITH MEAT & BEANS, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada with meat and beans, green-chile or enchilada sauce"
+58100530,"ENCHILADA W/ MEAT, RED-CHILE OR ENCHILADA SAUCE","Enchilada with meat, red-chile or enchilada sauce"
+58100535,"ENCHILADA WITH MEAT, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada with meat, green-chile or enchilada sauce"
+58100620,"ENCHILADA W/ CHICKEN & BEANS, RED-CHILE OR ENCHILADA SAUCE","Enchilada with chicken and beans, red-chile or enchilada sauce"
+58100625,"ENCHILADA WITH CHIC & BEANS, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada with chicken and beans, green-chile or enchilada sauce"
+58100630,"ENCHILADA W/ CHICKEN, RED-CHILE OR ENCHILADA SAUCE","Enchilada with chicken, red-chile or enchilada sauce"
+58100635,"ENCHILADA WITH CHICKEN, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada with chicken, green-chile or enchilada sauce"
+58100720,"ENCHILADA W/ BEANS, MEATLESS, RED-CHILE OR ENCHILADA SAUCE","Enchilada with beans, meatless, red-chile or enchilada sauce"
+58100725,"ENCHILADA WITH BEANS, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada with beans, green-chile or enchilada sauce"
+58100800,"ENCHILADA, JUST CHEESE,  NO BEANS, RED-CHILE OR ENCHILADA SC","Enchilada, just cheese, meatless, no beans, red-chile or enchilada sauce"
+58100805,"ENCHILADA, JUST CHEESE, GREEN-CHILE OR ENCHILADA SAUCE","Enchilada, just cheese, meatless, no beans, green-chile or enchilada sauce"
+58101320,"TACO OR TOSTADA W/ MEAT","Taco or tostada with meat"
+58101323,"TACO OR TOSTADA WITH MEAT, FROM FAST FOOD","Taco or tostada with meat, from fast food"
+58101325,"TACO OR TOSTADA W/ MEAT & SOUR CREAM","Taco or tostada with meat and sour cream"
+58101345,"SOFT TACO W/ MEAT","Soft taco with meat"
+58101347,"SOFT TACO WITH MEAT, FROM FAST FOOD","Soft taco with meat, from fast food"
+58101350,"SOFT TACO W/ MEAT & SOUR CREAM","Soft taco with meat and sour cream"
+58101357,"SOFT TACO WITH MEAT AND SOUR CREAM, FROM FAST FOOD","Soft taco with meat and sour cream, from fast food"
+58101450,"SOFT TACO WITH CHICKEN","Soft taco with chicken"
+58101457,"SOFT TACO WITH CHICKEN, FROM FAST FOOD","Soft taco with chicken, from fast food"
+58101460,"SOFT TACO W/ CHICKEN & SOUR CREAM","Soft taco with chicken and sour cream"
+58101520,"TACO OR TOSTADA W/ CHICKEN","Taco or tostada with chicken"
+58101525,"TACO OR TOSTADA W/ CHICKEN & SOUR CREAM","Taco or tostada with chicken and sour cream"
+58101540,"TACO OR TOSTADA W/ FISH","Taco or tostada with fish"
+58101555,"SOFT TACO W/ FISH","Soft taco with fish"
+58101610,"SOFT TACO W/  BEANS","Soft taco with beans"
+58101615,"SOFT TACO W/ BEANS & SOUR CREAM","Soft taco with beans and sour cream"
+58101620,"SOFT TACO W/ MEAT & BEANS","Soft taco with meat and beans"
+58101625,"SOFT TACO W/ CHICKEN & BEANS","Soft taco with chicken and beans"
+58101630,"SOFT TACO W/ MEAT, BEANS, & SOUR CREAM","Soft taco with meat, beans, and sour cream"
+58101635,"SOFT TACO W/ CHICKEN, BEANS, & SOUR CREAM","Soft taco with chicken, beans, and sour cream"
+58101720,"TACO OR TOSTADA W/ BEANS","Taco or tostada with beans"
+58101725,"TACO OR TOSTADA W/ BEANS & SOUR CREAM","Taco or tostada with beans and sour cream"
+58101730,"TACO OR TOSTADA W/ MEAT & BEANS","Taco or tostada with meat and beans"
+58101733,"TACO OR TOSTADA WITH MEAT AND BEANS, FROM FAST FOOD","Taco or tostada with meat and beans, from fast food"
+58101735,"TACO OR TOSTADA W/ CHICKEN & BEANS","Taco or tostada with chicken and beans"
+58101745,"TACO OR TOSTADA W/ MEAT, BEANS, & SOUR CREAM","Taco or tostada with meat, beans, and sour cream"
+58101750,"TACO OR TOSTADA W/ CHICKEN, BEANS, & SOUR CREAM","Taco or tostada with chicken, beans, and sour cream"
+58101800,"GROUND BEEF W/ TOMATO SAUCE, ON A CORNBREAD CRUST","Ground beef with tomato sauce and taco seasonings on a cornbread crust"
+58101820,"MEXICAN CASSEROLE W/ BEEF & BEANS","Mexican casserole made with ground beef, beans, tomato sauce, cheese, taco seasonings, and corn chips"
+58101830,"MEXICAN CASSEROLE W/ BEEF (INCL FRITO PIE, NFS)","Mexican casserole made with ground beef, tomato sauce, cheese, taco seasonings, and corn chips"
+58101930,"TACO OR TOSTADA SALAD W/ MEAT","Taco or tostada salad with meat"
+58101935,"TACO OR TOSTADA SALAD WITH CHICKEN","Taco or tostada salad with chicken"
+58101940,"TACO OR TOSTADA SALAD, MEATLESS","Taco or tostada salad, meatless"
+58101945,"TACO SALAD W/ MEAT & SOUR CREAM","Taco or tostada salad with meat and sour cream"
+58101950,"TACO OR TOSTADA SALAD W/ CHICKEN & SOUR CREAM","Taco or tostada salad with chicken and sour cream"
+58101955,"TACO OR TOSTADA SALAD, MEATLESS W/ SOUR CREAM","Taco or tostada salad, meatless with sour cream"
+58103120,"TAMALE WITH MEAT","Tamale with meat"
+58103130,"TAMALE WITH CHICKEN","Tamale with chicken"
+58103200,"TAMALE, PLAIN, MEATLESS, NO SAUCE, PR STYLE","Tamale, plain, meatless, no sauce, Puerto Rican style or Carribean Style"
+58103210,"TAMALE, MEATLESS, W/ SAUCE, P.R. OR CARIBBEAN STYLE","Tamale, meatless, with sauce, Puerto Rican or Caribbean style"
+58103250,"TAMALE, PLAIN, MEATLESS, NO SAUCE, MEXICAN","Tamale, plain, meatless, no sauce, Mexican style"
+58103310,"TAMALE CASSEROLE W/ MEAT","Tamale casserole with meat"
+58104090,"NACHOS W/ CHEESE & SOUR CREAM","Nachos with cheese and sour cream"
+58104120,"NACHOS W/ CHEESE","Nachos with cheese"
+58104130,"NACHOS W/ MEAT & CHEESE","Nachos with meat and cheese"
+58104150,"NACHOS W/ CHICKEN & CHEESE","Nachos with chicken and cheese"
+58104160,"NACHOS W/ CHILI","Nachos with chili"
+58104180,"NACHOS W/ MEAT, CHEESE, & SOUR CREAM","Nachos with meat, cheese, and sour cream"
+58104190,"NACHOS W/ CHICKEN, CHEESE, & SOUR CREAM","Nachos with chicken, cheese, and sour cream"
+58104260,"GORDITA, SOPE, OR CHALUPA W/ BEANS","Gordita, sope, or chalupa with beans"
+58104270,"GORDITA, SOPE, OR CHALUPA W/ BEANS & SOUR CREAM","Gordita, sope, or chalupa with beans and sour cream"
+58104280,"GORDITA, SOPE, OR CHALUPA W/ MEAT & SOUR CREAM","Gordita, sope, or chalupa with meat and sour cream"
+58104290,"GORDITA, SOPE, OR CHALUPA W/ MEAT","Gordita, sope, or chalupa with meat"
+58104320,"GORDITA, SOPE, OR CHALUPA W/ CHICKEN & SOUR CREAM","Gordita, sope, or chalupa with chicken and sour cream"
+58104340,"GORDITA, SOPE, OR CHALUPA W/ CHICKEN","Gordita, sope, or chalupa with chicken"
+58104500,"CHIMICHANGA W/ MEAT","Chimichanga with meat"
+58104520,"CHIMICHANGA, MEATLESS","Chimichanga, meatless"
+58104530,"CHIMICHANGA W/ CHICKEN","Chimichanga with chicken"
+58104535,"CHIMICHANGA W/ MEAT & SOUR CREAM","Chimichanga with meat and sour cream"
+58104540,"CHIMICHANGA, MEATLESS, W/ SOUR CREAM","Chimichanga, meatless, with sour cream"
+58104550,"CHIMICHANGA W/ CHICKEN & SOUR CREAM","Chimichanga with chicken and sour cream"
+58104710,"QUESADILLA, JUST CHEESE, MEATLESS","Quesadilla, just cheese, meatless"
+58104720,"QUESADILLA, JUST CHEESE, FROM FAST FOOD","Quesadilla, just cheese, from fast food"
+58104730,"QUESADILLA W/ MEAT","Quesadilla with meat"
+58104740,"QUESADILLA W/ CHICKEN","Quesadilla with chicken"
+58104745,"QUESADILLA WITH CHICKEN, FROM FAST FOOD","Quesadilla with chicken, from fast food"
+58104750,"QUESADILLA W/ VEGETABLES","Quesadilla with vegetables"
+58104760,"QUESADILLA W/ VEGETABLES & MEAT","Quesadilla with vegetables and meat"
+58104770,"QUESADILLA W/ VEGETABLES & CHICKEN","Quesadilla with vegetables and chicken"
+58104800,"TAQUITO OR FLAUTA W/ CHEESE","Taquito or flauta with cheese"
+58104820,"TAQUITO OR FLAUTA W/ MEAT","Taquito or flauta with meat"
+58104825,"TAQUITO OR FLAUTA W/ MEAT & CHEESE","Taquito or flauta with meat and cheese"
+58104830,"TAQUITO OR FLAUTA W/ CHICKEN","Taquito or flauta with chicken"
+58104835,"TAQUITO OR FLAUTA W/ CHICKEN AND CHEESE","Taquito or flauta with chicken and cheese"
+58104900,"TAQUITO OR FLAUTA W/ EGG","Taquito or flauta with egg"
+58104905,"TAQUITO OR FLAUTA W/ EGG & BREAKFAST MEAT","Taquito or flauta with egg and breakfast meat"
+58105000,"FAJITA W/ CHICKEN & VEGETABLES","Fajita with chicken and vegetables"
+58105050,"FAJITA W/ MEAT & VEGETABLES","Fajita with meat and vegetables"
+58105075,"FAJITA W/ VEGETABLES","Fajita with vegetables"
+58105100,"PUPUSA, CHEESE-FILLED","Pupusa, cheese-filled"
+58105105,"PUPUSA, BEAN-FILLED","Pupusa, bean-filled"
+58105110,"PUPUSA, MEAT-FILLED","Pupusa, meat-filled"
+58106200,"PIZZA, CHEESE, PREP FROM FROZEN, THIN CRUST","Pizza, cheese, prepared from frozen, thin crust"
+58106205,"PIZZA, CHEESE, PREP FROM FROZEN, THICK CRUST","Pizza, cheese, prepared from frozen, thick crust"
+58106210,"PIZZA, CHEESE,FRM REST/FF, NS AS TO TYPE OF CRUST","Pizza, cheese, from restaurant or fast food, NS as to type of crust"
+58106220,"PIZZA, CHEESE, FROM RESTAURANT OR FAST FOOD, THIN CRUST","Pizza, cheese, from restaurant or fast food, thin crust"
+58106225,"PIZZA, CHEESE, FROM RESTAURANT OR FAST FOOD, REGULAR CRUST","Pizza, cheese, from restaurant or fast food, regular crust"
+58106230,"PIZZA, CHEESE, FROM RESTAURANT OR FAST FOOD, THICK CRUST","Pizza, cheese, from restaurant or fast food, thick crust"
+58106233,"PIZZA, CHEESE, STUFFED CRUST","Pizza, cheese, stuffed crust"
+58106235,"PIZZA, CHEESE, FROM SCHOOL LUNCH, THIN CRUST","Pizza, cheese, from school lunch, thin crust"
+58106236,"PIZZA, CHEESE, FROM SCHOOL LUNCH, THICK CRUST","Pizza, cheese, from school lunch, thick crust"
+58106240,"PIZZA, EXTRA CHEESE, NS AS TO TYPE OF CRUST","Pizza, extra cheese, NS as to type of crust"
+58106250,"PIZZA, EXTRA CHEESE, THIN CRUST","Pizza, extra cheese, thin crust"
+58106255,"PIZZA, EXTRA CHEESE, REGULAR CRUST","Pizza, extra cheese, regular crust"
+58106260,"PIZZA, EXTRA CHEESE, THICK CRUST","Pizza, extra cheese, thick crust"
+58106300,"PIZZA, CHEESE, W/ VEGETABLES, PREP FROM FROZEN, THIN CRUST","Pizza, cheese, with vegetables, prepared from frozen, thin crust"
+58106305,"PIZZA, CHEESE, W/ VEGETABLES, PREP FROM FROZEN, THICK CRUST","Pizza, cheese with vegetables, prepared from frozen, thick crust"
+58106310,"PIZZA, CHEESE, W/ VEG, NS AS TO TYPE OF CRUST","Pizza, cheese, with vegetables, NS as to type of crust"
+58106320,"PIZZA, CHEESE, W/ VEGETABLES, THIN CRUST","Pizza, cheese, with vegetables, thin crust"
+58106325,"PIZZA, CHEESE, W/ VEGETABLES, REGULAR CRUST","Pizza, cheese, with vegetables, regular crust"
+58106330,"PIZZA, CHEESE, W/ VEGETABLES, THICK CRUST","Pizza, cheese, with vegetables, thick crust"
+58106340,"PIZZA W/ CHEESE & EXTRA VEGETABLES, NS AS TO CRUST","Pizza, with cheese and extra vegetables, NS as to type of crust"
+58106345,"PIZZA W/ CHEESE & EXTRA VEGETABLES, THIN CRUST","Pizza with cheese and extra vegetables, thin crust"
+58106347,"PIZZA W/ CHEESE & EXTRA VEGETABLES, REGULAR CRUST","Pizza with cheese and extra vegetables, regular crust"
+58106350,"PIZZA W/ CHEESE & EXTRA VEGETABLES, THICK CRUST","Pizza with cheese and extra vegetables, thick crust"
+58106357,"PIZZA, CHEESE, W/ FRUIT, NS AS TO TYPE OF CRUST","Pizza, cheese, with fruit, NS as to type of crust"
+58106358,"PIZZA, CHEESE, W/ FRUIT, THIN CRUST","Pizza, cheese, with fruit, thin crust"
+58106359,"PIZZA, CHEESE, W/ FRUIT, REGULAR CRUST","Pizza, cheese, with fruit, regular crust"
+58106360,"PIZZA, CHEESE, W/ FRUIT, THICK CRUST","Pizza, cheese, with fruit, thick crust"
+58106500,"PIZZA W/ MEAT, PREP FROM FROZEN, THIN CRUST","Pizza with meat, prepared from frozen, thin crust"
+58106505,"PIZZA W/ MEAT, PREP FROM FROZEN, THICK CRUST","Pizza with meat, prepared from frozen, thick crust"
+58106540,"PIZZA W/ PEPPERONI,FRM REST/FF, NS AS TO TYPE OF CRUST","Pizza with pepperoni, from restaurant or fast food, NS as to type of crust"
+58106550,"PIZZA W/PEPPERONI, FROM RESTAURANT/FAST FOOD, THIN CRUST","Pizza with pepperoni, from restaurant or fast food, thin crust"
+58106555,"PIZZA W/PEPPERONI, FROM RESTAURANT/FAST FOOD, REGULAR CRUST","Pizza with pepperoni, from restaurant or fast food, regular crust"
+58106560,"PIZZA W/ PEPPERONI, FROM RESTAURANT/FAST FOOD, THICK CRUST","Pizza with pepperoni, from restaurant or fast food, thick crust"
+58106565,"PIZZA WITH PEPPERONI, STUFFED CRUST","Pizza with pepperoni, stuffed crust"
+58106570,"PIZZA WITH PEPPERONI, FROM SCHOOL LUNCH, THIN CRUST","Pizza with pepperoni, from school lunch, thin crust"
+58106580,"PIZZA WITH PEPPERONI, FROM SCHOOL LUNCH, THICK CRUST","Pizza with pepperoni, from school lunch, thick crust"
+58106610,"PIZZA W/ MEAT OTHER THAN PEPP, FRM REST/FF, NS TYPE OF CRUST","Pizza with meat other than pepperoni, from restaurant or fast food, NS as to type of crust"
+58106620,"PIZZA W/MEAT NOT PEPPERONI, FRM RESTAURANT/FF,THIN CRUST","Pizza with meat other than pepperoni, from restaurant or fast food, thin crust"
+58106625,"PIZZA W/MEAT NOT PEPPERONI, FRM RESTAURANT/FF, REG CRUST","Pizza with meat other than pepperoni, from restaurant or fast food, regular crust"
+58106630,"PIZZA W/MEAT NOT PEPPERONI, FRM RESTAURANT/FF, THICK CRUST","Pizza with meat other than pepperoni, from restaurant or fast food, thick crust"
+58106633,"PIZZA, W/MEAT NOT PEPPERONI, STUFFED CRUST","Pizza, with meat other than pepperoni, stuffed crust"
+58106635,"PIZZA, W/MEAT OTHER THAN PEPPERONI, FRM SCL LUNCH, THIN CRUS","Pizza, with meat other than pepperoni, from school lunch, thin crust"
+58106636,"PIZZA, W/MEAT OTHER THAN PEPPERONI, FRM SCL LUNCH, THICK CRU","Pizza, with meat other than pepperoni, from school lunch, thick crust"
+58106640,"PIZZA W/ EXTRA MEAT, NS AS TO TYPE OF CRUST","Pizza with extra meat, NS as to type of crust"
+58106650,"PIZZA W/ EXTRA MEAT, THIN CRUST","Pizza with extra meat, thin crust"
+58106655,"PIZZA W/ EXTRA MEAT, REGULAR CRUST","Pizza with extra meat, regular crust"
+58106660,"PIZZA W/ EXTRA MEAT, THICK CRUST","Pizza with extra meat, thick crust"
+58106700,"PIZZA W/ MEAT & VEGS, PREP FROM FROZEN, THIN CRUST","Pizza with meat and vegetables, prepared from frozen, thin crust"
+58106705,"PIZZA W/ MEAT & VEGS, PREP FROM FROZEN, THICK CRUST","Pizza with meat and vegetables, prepared from frozen, thick crust"
+58106710,"PIZZA W/ MEAT & VEG, NS AS TO TYPE OF CRUST","Pizza with meat and vegetables, NS as to type of crust"
+58106720,"PIZZA W/ MEAT & VEGETABLES, THIN CRUST","Pizza with meat and vegetables, thin crust"
+58106725,"PIZZA W/ MEAT & VEGETABLES, REGULAR CRUST","Pizza with meat and vegetables, regular crust"
+58106730,"PIZZA W/ MEAT & VEGETABLES, THICK CRUST","Pizza with meat and vegetables, thick crust"
+58106735,"PIZZA W/ EXTRA MEAT & EXTRA VEGS, NS AS TO TYPE OF CRUST","Pizza with extra meat and extra vegetables, NS as to type of crust"
+58106736,"PIZZA W/ EXTRA MEAT & EXTRA VEGS, THIN CRUST","Pizza with extra meat and extra vegetables, thin crust"
+58106737,"PIZZA W/ EXTRA MEAT & EXTRA VEGS, THICK CRUST","Pizza with extra meat and extra vegetables, thick crust"
+58106738,"PIZZA W/ EXTRA MEAT & EXTRA VEGS, REGULAR CRUST","Pizza with extra meat and extra vegetables, regular crust"
+58106740,"PIZZA W/ MEAT & FRUIT, NS AS TO TYPE OF CRUST","Pizza with meat and fruit, NS as to type of crust"
+58106750,"PIZZA W/ MEAT & FRUIT, THIN CRUST","Pizza with meat and fruit, thin crust"
+58106755,"PIZZA W/ MEAT & FRUIT, REGULAR CRUST","Pizza with meat and fruit, regular crust"
+58106760,"PIZZA W/ MEAT & FRUIT, THICK CRUST","Pizza with meat and fruit, thick crust"
+58106820,"PIZZA W/ BEANS & VEG, THIN CRUST (INCL TACO PIZZA)","Pizza with beans and vegetables, thin crust"
+58106830,"PIZZA W/ BEANS & VEG, THICK CRUST (INCL TACO PIZZA)","Pizza with beans and vegetables, thick crust"
+58107050,"PIZZA, NO CHEESE, THIN CRUST","Pizza, no cheese, thin crust"
+58107100,"PIZZA, NO CHEESE, THICK CRUST","Pizza, no cheese, thick crust"
+58107220,"WHITE PIZZA, THIN CRUST","White pizza, thin crust"
+58107230,"WHITE PIZZA, THICK CRUST","White pizza, thick crust"
+58108000,"CALZONE, W/ CHEESE, MEATLESS (INCL STROMBOLI)","Calzone, with cheese, meatless"
+58108010,"CALZONE, W/ MEAT & CHEESE (INCLUDE STROMBOLI)","Calzone, with meat and cheese"
+58108050,"PIZZA ROLLS (INCLUDE PIZZA BITES)","Pizza rolls"
+58110110,"EGG ROLL, MEATLESS","Egg roll, meatless"
+58110120,"EGG ROLL, W/ SHRIMP","Egg roll, with shrimp"
+58110130,"EGG ROLL, W/ BEEF/PORK","Egg roll, with beef and/or pork"
+58110170,"EGG ROLL, W/ CHICKEN","Egg roll, with chicken or turkey"
+58110200,"ROLL W/MEAT&/SHRIMP,VEGETABLES&RICE PAPER(NOT FRIED","Roll with meat and/or shrimp, vegetables and rice paper (not fried)"
+58111110,"WON TON (WONTON), FRIED, FILLED W/MEAT, POULTRY, OR SEAFOOD","Won ton (wonton), fried, filled with meat, poultry, or seafood"
+58111120,"WON TON (WONTON), FRIED, MEATLESS","Won ton (wonton), fried, meatless"
+58111130,"WON TON (WONTON), FRIED, FILLED WITH MEAT, POULTRY, OR SEAFO","Won ton (wonton), fried, filled with meat, poultry, or seafood, and vegetable"
+58111200,"PUFFS, FRIED, CRAB MEAT & CREAM CHEESE FILLED","Puffs, fried, crab meat and cream cheese filled"
+58112510,"DUMPLING, STEAMED, FILLED W/ MEAT OR SEAFOOD","Dumpling, steamed, filled with meat, poultry, or seafood"
+58115110,"TAMALE CASSEROLE, P.R. (TAMALES EN CAZUELA)","Tamale casserole, Puerto Rican style (Tamales en cazuela)"
+58115150,"TAMAL IN A LEAF, P.R. (TAMALES EN HOJA)","Tamal in a leaf, Puerto Rican style (Tamales en hoja)"
+58115210,"TACO W/ CRAB MEAT, P.R. (TACOS DE JUEYES)","Taco with crab meat, Puerto Rican style (Taco de jueye)"
+58116110,"MEAT TURNOVER, PUERTO RICAN STYLE","Meat turnover, Puerto Rican style (Pastelillo de carne; Empanadilla)"
+58116115,"EMPANADA, MEXICAN TURNOVER, W/ CHS & VEG","Empanada, Mexican turnover, filled with cheese and vegetables"
+58116120,"EMPANADA, MEXICAN TURNOVER, W/ MEAT & VEGS","Empanada, Mexican turnover, filled with meat and vegetables"
+58116130,"EMPANADA, MEXICAN TURNOVER, W/ CHIC & VEG","Empanada, Mexican turnover, filled with chicken and vegetables"
+58116210,"MEAT PIE, P.R. (PASTELON DE CARNE)","Meat pie, Puerto Rican style (Pastelon de carne)"
+58116310,"CHEESE TURNOVER, PUERTO RICAN STYLE","Cheese turnover, Puerto Rican style (Pastelillo de queso; Empanadilla)"
+58117110,"CORNMEAL FRITTER, P.R. (AREPA, P.R. AREPITAS)","Cornmeal fritter, Puerto Rican style (Arepa; P.R. arepita)"
+58117210,"CORNMEAL STICK, P.R. (SORULLOS / SORULLITOS DE MAIZ)","Cornmeal stick, Puerto Rican style (Sorullos / Sorullitos de maiz)"
+58117310,"KIBBY, P.R. (BEEF & BULGUR) (PLATO ARABE)","Kibby, Puerto Rican style (beef and bulgur) (Plato Arabe)"
+58117410,"CODFISH FRITTER, P.R. (BACALAITOS FRITOS)","Codfish fritter, Puerto Rican style (Bacalaitos fritos)"
+58117510,"HAYACAS, P.R. (HOMINY, PORK OR HAM, VEGETABLES)","Hayacas, Puerto Rican style (hominy, pork or ham, vegetables)"
+58118110,"CORNSTARCH COCONUT DESSERT, P.R. (TEMBLEQUE)","Cornstarch coconut dessert, Puerto Rican style (Tembleque)"
+58118210,"CORNMEAL COCONUT DESSERT, P.R.","Cornmeal coconut dessert, Puerto Rican style (Harina de maiz con coco)"
+58120110,"CREPES, FILLED W/ MEAT, FISH OR POULTRY, W/ SAUCE","Crepes, filled with meat, fish, or poultry, with sauce"
+58120120,"CREPE,FILLED W/ MEAT, FISH & POULTRY,NO SCE ON TOP","Crepe, filled with beef, pork, fish and/or poultry, no sauce on top"
+58121510,"DUMPLING, MEAT-FILLED (INCLUDE PIEROGI, PIROSHKI)","Dumpling, meat-filled"
+58121610,"DUMPLING, POTATO/CHEESE-FILLED (INCLUDE PIEROGI)","Dumpling, potato- or cheese-filled"
+58121620,"DUMPLING, VEGETABLE","Dumpling, vegetable"
+58122210,"GNOCCHI, CHEESE","Gnocchi, cheese"
+58122220,"GNOCCHI, POTATO","Gnocchi, potato"
+58122250,"KISHKE, STUFFED DERMA","Kishke, stuffed derma"
+58122310,"KNISH, POTATO (PASTRY FILLED WITH POTATO)","Knish, potato (pastry filled with potato)"
+58122320,"KNISH, CHEESE (PASTRY FILLED WITH CHEESE)","Knish, cheese (pastry filled with cheese)"
+58122330,"KNISH, MEAT (PASTRY FILLED WITH MEAT)","Knish, meat (pastry filled with meat)"
+58123110,"SWEET BREAD DOUGH, FILLED WITH MEAT, STEAMED","Sweet bread dough, filled with meat, steamed"
+58123120,"SWEET BREAD DOUGH, FILLED WITH BEAN PASTE, MEATLESS, STEAMED","Sweet bread dough, filled with bean paste, meatless, steamed"
+58124210,"PASTRY, CHEESE-FILLED","Pastry, cheese-filled"
+58124250,"SPANAKOPITTA (INCL GREEK SPINACH-CHEESE PIE)","Spanakopitta"
+58124500,"PASTRY,FILLED W/POTATOES & PEAS, FRIED","Pastry, filled with potatoes and peas, fried"
+58125110,"QUICHE W/ MEAT, POULTRY OR FISH","Quiche with meat, poultry or fish"
+58125120,"SPINACH QUICHE, MEATLESS","Spinach quiche, meatless"
+58125180,"CHEESE QUICHE, MEATLESS","Cheese quiche, meatless"
+58126000,"BIEROCK (TURNOVER W/ BEEF & CABBAGE)","Bierock (turnover filled with ground beef and cabbage mixture)"
+58126110,"TURNOVER, MEAT-FILLED, NO GRAVY","Turnover, meat-filled, no gravy"
+58126120,"TURNOVER, MEAT-FILLED, W/ GRAVY","Turnover, meat-filled, with gravy"
+58126130,"TURNOVER, MEAT- & CHEESE-FILLED, NO GRAVY","Turnover, meat- and cheese-filled, no gravy"
+58126140,"TURNOVER, MEAT- & BEAN-FILLED, NO GRAVY","Turnover, meat- and bean-filled, no gravy"
+58126150,"TURNOVER, MEAT & CHEESE, TOMATO SAUCE","Turnover, meat- and cheese-filled, tomato-based sauce"
+58126160,"TURNOVER, CHEESE-FILLED, TOMATO-BASED SAUCE","Turnover, cheese-filled, tomato-based sauce"
+58126170,"TURNOVER, MEAT & VEG (NO POTATO), NO GRAVY","Turnover, meat-and vegetable- filled (no potatoes, no gravy)"
+58126180,"TURNOVER,MEAT- POTATO- & VEGETABLE-FILLED NO GRAVY","Turnover, meat-, potato-, and vegetable-filled, no gravy"
+58126270,"TURNOVER,CHICKEN/TURKEY FILLED,NO GRAVY","Turnover, chicken- or turkey-, and cheese-filled, no gravy"
+58126280,"TURNOVER, CHICKEN/TURKEY- & VEG-FILLED, LOWER FAT","Turnover, chicken- or turkey-, and vegetable-filled, lower in fat"
+58126290,"TURNOVER, MEAT- & CHEESE-FILLED, LOWER FAT","Turnover, meat- and cheese-filled, lower in fat"
+58126300,"TURNOVER, MEAT- & CHEESE-FILLED, TOMATO SAUCE, LOWER FAT","Turnover, meat- and cheese-filled, tomato-based sauce, lower in fat"
+58126310,"TURNOVER, CHICKEN, W/ GRAVY","Turnover, chicken, with gravy"
+58126400,"TURNOVER, FILLED W/ EGG, MEAT & CHEESE","Turnover, filled with egg, meat and cheese"
+58126410,"TURNOVER, FILLED W/ EGG, MEAT & CHEESE, LOWER IN FAT","Turnover, filled with egg, meat, and cheese, lower in fat"
+58127110,"VEGETABLES IN PASTRY (INCL ALL VARIETIES)","Vegetables in pastry"
+58127150,"VEGETABLES & CHEESE IN PASTRY","Vegetables and cheese in pastry"
+58127200,"CROISSANT, FILLED W/ BROCCOLI & CHEESE","Croissant sandwich, filled with broccoli and cheese"
+58127210,"CROISSANT, FILLED W/ HAM & CHEESE","Croissant sandwich, filled with ham and cheese"
+58127220,"CROISSANT, FILLED W/CHICKEN,BROCCOLI & CHEESE SAUCE","Croissant sandwich, filled with chicken, broccoli, and cheese sauce"
+58127270,"CROISSANT W/ SAUSAGE & EGG","Croissant sandwich with sausage and egg"
+58127290,"CROISSANT W/ BACON & EGG","Croissant sandwich with bacon and egg"
+58127310,"CROISSANT W/ HAM, EGG, & CHEESE","Croissant sandwich with ham, egg, and cheese"
+58127330,"CROISSANT W/ SAUSAGE, EGG, & CHEESE","Croissant sandwich with sausage, egg, and cheese"
+58127350,"CROISSANT W/ BACON, EGG, & CHEESE","Croissant sandwich with bacon, egg, and cheese"
+58127500,"VEGETABLE SUBMARINE SANDWICH, W/ FAT FREE SPREAD","Vegetable submarine sandwich, with fat free spread"
+58128000,"BISCUIT W/ GRAVY","Biscuit with gravy"
+58128110,"CHICKEN CORNBREAD","Chicken cornbread"
+58128120,"CORNMEAL DRESSING W/ CHICKEN & VEGETABLES","Cornmeal dressing with chicken or turkey and vegetables"
+58128210,"DRESSING W/ OYSTERS","Dressing with oysters"
+58128220,"DRESSING W/ CHICKEN/TURKEY & VEGETABLES","Dressing with chicken or turkey and vegetables"
+58128250,"DRESSING W/ MEAT & VEGETABLES","Dressing with meat and vegetables"
+58130011,"LASAGNA WITH MEAT","Lasagna with meat"
+58130013,"LASAGNA W/ MEAT, CANNED","Lasagna with meat, canned"
+58130020,"LASAGNA, W/ MEAT & SPINACH","Lasagna with meat and spinach"
+58130140,"LASAGNA WITH CHICKEN OR TURKEY","Lasagna with chicken or turkey"
+58130150,"LASAGNA W/ CHIC OR TURKEY, & SPINACH","Lasagna, with chicken or turkey, and spinach"
+58130310,"LASAGNA, MEATLESS","Lasagna, meatless"
+58130320,"LASAGNA, MEATLESS, W/ VEGETABLES","Lasagna, meatless, with vegetables"
+58130610,"LASAGNA W/ MEAT, WHOLE WHEAT NOODLES","Lasagna with meat, whole wheat noodles"
+58130810,"LASAGNA, MEATLESS, WHOLE WHEAT NOODLES","Lasagna, meatless, whole wheat noodles"
+58130910,"LASAGNA W/ MEAT, SPINACH NOODLES","Lasagna with meat, spinach noodles"
+58130950,"LASAGNA, MEATLESS, SPINACH NOODLES","Lasagna, meatless, spinach noodles"
+58131100,"RAVIOLI, FILLING NS, NO SAUCE","Ravioli, NS as to filling, no sauce"
+58131110,"RAVIOLI, FILLING NS, TOMATO SAUCE","Ravioli, NS as to filling, with tomato sauce"
+58131120,"RAVIOLI, NS AS TO FILLING, WITH CREAM SAUCE","Ravioli, NS as to filling, with cream sauce"
+58131310,"RAVIOLI, MEAT-FILLED, NO SAUCE","Ravioli, meat-filled, no sauce"
+58131320,"RAVIOLI, MEAT-FILLED, W/ TOMATO OR MEAT SAUCE","Ravioli, meat-filled, with tomato sauce or meat sauce"
+58131323,"RAVIOLI, MEAT-FILLED, W/ TOMATO OR MEAT SAUCE, CANNED","Ravioli, meat-filled, with tomato sauce or meat sauce, canned"
+58131330,"RAVIOLI, MEAT-FILLED, WITH CREAM SAUCE","Ravioli, meat-filled, with cream sauce"
+58131510,"RAVIOLI, CHEESE-FILLED, NO SAUCE","Ravioli, cheese-filled, no sauce"
+58131520,"RAVIOLI, CHEESE-FILLED, W/ TOMATO SAUCE","Ravioli, cheese-filled, with tomato sauce"
+58131523,"RAVIOLI, CHEESE-FILLED, W/ TOMATO SAUCE, CANNED","Ravioli, cheese-filled, with tomato sauce, canned"
+58131530,"RAVIOLI, CHEESE-FILLED, W/ MEAT SAUCE","Ravioli, cheese-filled, with meat sauce"
+58131535,"RAVIOLI, CHEESE-FILLED, WITH CREAM SAUCE","Ravioli, cheese-filled, with cream sauce"
+58131590,"RAVIOLI, CHEESE AND SPINACH-FILLED, NO SAUCE","Ravioli, cheese and spinach-filled, no sauce"
+58131600,"RAVIOLI, CHEESE&SPINACH-FILLED, W/ CREAM SAUCE","Ravioli, cheese and spinach-filled, with cream sauce"
+58131610,"RAVIOLI, CHEESE AND SPINACH FILLED, WITH TOMATO SAUCE","Ravioli, cheese and spinach filled, with tomato sauce"
+58132110,"SPAGHETTI W/ TOMATO SAUCE, MEATLESS","Spaghetti with tomato sauce, meatless"
+58132113,"PASTA, W/ TOMATO SAUCE & CHEESE, CANNED","Pasta with tomato sauce and cheese, canned"
+58132310,"SPAGHETTI W/TOMAT SAUCE & MEAT SAUCE","Spaghetti with tomato sauce and meatballs or spaghetti with meat sauce or spaghetti with meat sauce and meatballs"
+58132313,"PASTA W/ TOMATO SAUCE & MEAT/MEATBALLS, CANNED","Pasta with tomato sauce and meat or meatballs, canned"
+58132340,"SPAGHETTI W/ TOMATO SAUCE & VEGETABLES","Spaghetti with tomato sauce and vegetables"
+58132350,"SPAGHETTI, WHOLE WHEAT, W/ TOMATO SAUCE, MEATLESS","Spaghetti with tomato sauce, meatless, whole wheat noodles"
+58132360,"SPAGHETTI, WHOLE WHEAT, W/ TOMATO & MEAT SAUCE","Spaghetti with tomato sauce and meatballs, whole wheat noodles or spaghetti with meat sauce, whole wheat noodles or spaghetti with meat sauce and meatballs, whole wheat noodles"
+58132450,"SPAGHETTI W/ TOM SAUCE, MEATLESS, SPINACH NOODLES","Spaghetti with tomato sauce, meatless, made with spinach noodles"
+58132460,"SPAGHETTI W/ TOMATO & MEAT SAUCE, SPINACH NOODLES","Spaghetti with tomato sauce and meatballs made with spinach noodles, or spaghetti with meat sauce made with spinach noodles,  or spaghetti with meat sauce and meatballs made with spinach noodles"
+58132710,"SPAGHETTI W/ TOMATO SAUCE & FRANKFURTERS/HOT DOG","Spaghetti with tomato sauce and frankfurters or hot dogs"
+58132713,"PASTA W/ TOMATO SAUCE & FRANKFURTERS/HOT DOGS, CANNED","Pasta with tomato sauce and frankfurters or hot dogs, canned"
+58132800,"SPAGHETTI W/ CLAM SAUCE, NS AS TO RED OR WHITE","Spaghetti with clam sauce, NS as to red or white"
+58132810,"SPAGHETTI W/ RED CLAM SAUCE","Spaghetti with red clam sauce"
+58132820,"SPAGHETTI W/ WHITE CLAM SAUCE","Spaghetti with white clam sauce"
+58132910,"SPAGHETTI WITH TOMATO SAUCE AND POULTRY","Spaghetti with tomato sauce and poultry"
+58133110,"MANICOTTI, CHEESE-FILLED, NO SAUCE","Manicotti, cheese-filled, no sauce"
+58133120,"MANICOTTI, CHEESE-FILLED, W/ TOMATO SAUCE, MEATLESS","Manicotti, cheese-filled, with tomato sauce, meatless"
+58133130,"MANICOTTI, CHEESE-FILLED, W/ MEAT SAUCE","Manicotti, cheese-filled, with meat sauce"
+58133140,"MANICOTTI, VEG- & CHEESE-FILLED, W/TOM SCE,MEATLESS","Manicotti, vegetable- and cheese-filled, with tomato sauce, meatless"
+58134110,"STUFFED SHELLS, CHEESE-FILLED, NO SAUCE","Stuffed shells, cheese-filled, no sauce"
+58134120,"STUFFED SHELLS, CHEESE-FILLED, W/ TOM SC, MEATLESS","Stuffed shells, cheese-filled, with tomato sauce, meatless"
+58134130,"STUFFED SHELLS, CHEESE-FILLED, W/ MEAT SAUCE","Stuffed shells, cheese-filled, with meat sauce"
+58134160,"STUFFED SHELLS, CHEESE  AND SPINACH  FILLED, NO SAUCE","Stuffed shells, cheese- and spinach- filled, no sauce"
+58134210,"STUFFED SHELLS, W/ CHICKEN, W/ TOM SCE","Stuffed shells, with chicken, with tomato sauce"
+58134310,"STUFFED SHELLS, W/ FISH &/OR SHELLFISH, W/ TOM SCE","Stuffed shells, with fish and/or shellfish, with tomato sauce"
+58134610,"TORTELLINI, MEAT-FILLED, W/ TOMATO SAUCE","Tortellini, meat-filled, with tomato sauce"
+58134613,"TORTELLINI, MEAT-FILLED, W/ TOMATO SAUCE, CANNED","Tortellini, meat-filled, with tomato sauce, canned"
+58134620,"TORTELLINI, CHEESE-FILLED, MEATLESS, W/TOMATO SAUCE","Tortellini, cheese-filled, meatless, with tomato sauce"
+58134623,"TORTELLINI,CHEESE-FILLED,MEATLESS,W/TOMATO SAUCE,CANNED","Tortellini, cheese-filled, meatless, with tomato sauce, canned"
+58134630,"TORTELLINI, CHEESE, W/ VEGETABLES & DRESSING","Tortellini, cheese-filled, meatless, with vegetables and vinaigrette dressing"
+58134640,"TORTELLINI, CHEESE-FILLED, MEATLESS, W/ VINAIGRETTE","Tortellini, cheese-filled, meatless, with vinaigrette dressing"
+58134650,"TORTELLINI, MEAT-FILLED, NO SAUCE","Tortellini, meat-filled, no sauce"
+58134660,"TORTELLINI, CHEESE-FILLED, W/ CREAM SAUCE","Tortellini, cheese-filled, with cream sauce"
+58134680,"TORTELLINI, CHEESE-FILLED, NO SAUCE","Tortellini, cheese-filled, no sauce"
+58134710,"TORTELLINI, SPINACH-FILLED, W/ TOMATO SAUCE","Tortellini, spinach-filled, with tomato sauce"
+58134720,"TORTELLINI, SPINACH-FILLED, NO SAUCE","Tortellini, spinach-filled, no sauce"
+58134810,"CANNELONI, CHEESE & SPINACH-FILLED, NO SAUCE","Cannelloni, cheese- and spinach-filled, no sauce"
+58135110,"CHOW FUN NOODLES W/ MEAT & VEGETABLES","Chow fun noodles with meat and vegetables"
+58135120,"CHOW FUN NOODLES W/ VEGETABLES, MEATLESS","Chow fun noodles with vegetables, meatless"
+58136110,"LO MEIN, NFS","Lo mein, NFS"
+58136120,"LO MEIN, MEATLESS","Lo mein, meatless"
+58136130,"LO MEIN WITH SHRIMP","Lo mein, with shrimp"
+58136140,"LO MEIN W/ PORK","Lo mein, with pork"
+58136150,"LO MEIN W/ BEEF","Lo mein, with beef"
+58136160,"LO MEIN W/ CHICKEN (INCL TURKEY)","Lo mein, with chicken"
+58137210,"PAD THAI, NFS","Pad Thai, NFS"
+58137220,"PAD THAI, MEATLESS","Pad Thai, meatless"
+58137230,"PAD THAI WITH CHICKEN","Pad Thai with chicken"
+58137240,"PAD THAI WITH SEAFOOD","Pad Thai with seafood"
+58137250,"PAD THAI WITH MEAT","Pad Thai with meat"
+58140110,"SPAGHETTI W/ CORNED BEEF, P.R.","Spaghetti with corned beef, Puerto Rican style"
+58140310,"MACARONI W/ TUNA, P.R. (MACARRONES CON ATUN)","Macaroni with tuna, Puerto Rican style (Macarrones con atun)"
+58145110,"MACARONI OR NOODLES W/ CHEESE","Macaroni or noodles with cheese"
+58145112,"MACARONI OR NOODLES WITH CHEESE, MADE FROM PACKAGED MIX","Macaroni or noodles with cheese, made from packaged mix"
+58145113,"MACARONI OR NOODLES W/ CHEESE, CANNED","Macaroni or noodles with cheese, canned"
+58145117,"MACARONI OR NOODLES WITH CHEESE, EASY MAC TYPE","Macaroni or noodles with cheese, Easy Mac type"
+58145119,"MACARONI OR NOODLES WITH CHEESE, MADE FRM RED FAT PACKAGE","Macaroni or noodles with cheese, made from reduced fat packaged mix"
+58145120,"MACARONI OR NOODLES W/ CHEESE & TUNA","Macaroni or noodles with cheese and tuna"
+58145135,"MACARONI OR NOODLES WITH CHEESE AND MEAT","Macaroni or noodles with cheese and meat"
+58145136,"MACARONI OR NOODLES W/CHEESE & MEAT, FR HAMBURGER HELPER","Macaroni or noodles with cheese and meat, prepared from Hamburger Helper mix"
+58145140,"MACARONI OR NOODLES W/ CHEESE & TOMATO","Macaroni or noodles with cheese and tomato"
+58145160,"MACARONI/NOODLES W/ CHEESE & FRANKFURTER/HOT DOG","Macaroni or noodles with cheese and frankfurters or hot dogs"
+58145170,"MACARONI OR NOODLES WITH CHEESE AND EGG","Macaroni or noodles with cheese and egg"
+58145190,"MACARONI W/ CHEESE & CHICKEN","Macaroni or noodles with cheese and chicken or turkey"
+58146100,"PASTA W/ TOMATO SAUCE, MEATLESS","Pasta with tomato sauce, meatless"
+58146110,"PASTA W/ MEAT SAUCE (INCLUDE AMER CHOP SUEY)","Pasta with meat sauce"
+58146120,"PASTA W/ CHEESE & MEAT SAUCE","Pasta with cheese and meat sauce"
+58146130,"PASTA W/ CARBONARA SAUCE","Pasta with carbonara sauce"
+58146150,"PASTA W/ CHEESE & TOMATO SAUCE, MEATLESS","Pasta with cheese and tomato sauce, meatless"
+58146160,"PASTA WITH VEGETABLES, NO SAUCE OR DRESSING","Pasta with vegetables, no sauce or dressing"
+58146200,"PASTA, MEAT-FILLED, W/ GRAVY, CANNED","Pasta, meat-filled, with gravy, canned"
+58146300,"PASTA, WHOLE WHEAT, WITH MEAT SAUCE","Pasta, whole wheat, with meat sauce"
+58146310,"PASTA, WHOLE WHEAT, W/ TOMATO SAUCE, MEATLESS","Pasta, whole wheat, with tomato sauce, meatless"
+58147100,"PASTA W/ PESTO SAUCE","Pasta with pesto sauce"
+58147110,"MACARONI OR NOODLES W/ BEANS & TOMATO SAUCE","Macaroni or noodles with beans or lentils and tomato sauce"
+58147310,"MACARONI, CREAMED","Macaroni, creamed"
+58147330,"MACARONI OR NOODLES, CREAMED, WITH CHEESE","Macaroni or noodles, creamed, with cheese"
+58147340,"MACARONI OR NOODLES, CREAMED, WITH CHEESE AND TUNA","Macaroni or noodles, creamed, with cheese and tuna"
+58147350,"MACARONI, CREAMED, W/ VEGETABLES","Macaroni, creamed, with vegetables"
+58147510,"FLAVORED PASTA (INCL LIPTON BEEF, CHICKEN FLAVORS)","Flavored pasta"
+58147520,"YAT GA MEIN WITH MEAT, FISH, OR POULTRY","Yat Ga Mein with meat, fish, or poultry"
+58148110,"MACARONI OR PASTA SALAD, W/  MAYO","Macaroni or pasta salad, made with mayonnaise"
+58148111,"MACARONI OR PASTA SALAD, W/ LT MAYO","Macaroni or pasta salad, made with light mayonnaise"
+58148112,"MACARONI OR PASTA SALAD, W/ MAYO-TYPE DRSG","Macaroni or pasta salad, made with mayonnaise-type salad dressing"
+58148113,"MACARONI OR PASTA SALAD, W/LT MAYO-TYPE DRSG","Macaroni or pasta salad, made with light mayonnaise-type salad dressing"
+58148114,"MACARONI OR PASTA SALAD, W/ ITALIAN DRSG","Macaroni or pasta salad, made with Italian dressing"
+58148115,"MACARONI OR PASTA SALAD, W/LT ITALIAN DRSG","Macaroni or pasta salad, made with light Italian dressing"
+58148116,"MACARONI OR PASTA SALAD, W/ CREAMY DRSG","Macaroni or pasta salad, made with creamy dressing"
+58148117,"MACARONI OR PASTA SALAD, W/ LT CREAMY DRSG","Macaroni or pasta salad, made with light creamy dressing"
+58148118,"MACARONI OR PASTA SALAD, W/ ANY TYPE OF FAT FREE DRSG","Macaroni or pasta salad, made with any type of fat free dressing"
+58148120,"MACARONI OR PASTA SALAD WITH EGG","Macaroni or pasta salad with egg"
+58148130,"MACARONI OR PASTA SALAD WITH TUNA","Macaroni or pasta salad with tuna"
+58148140,"MACARONI OR PASTA SALAD WITH CRAB MEAT","Macaroni or pasta salad with crab meat"
+58148150,"MACARONI OR PASTA SALAD WITH SHRIMP","Macaroni or pasta salad with shrimp"
+58148160,"MACARONI OR PASTA SALAD WITH TUNA AND EGG","Macaroni or pasta salad with tuna and egg"
+58148170,"MACARONI OR PASTA SALAD WITH CHICKEN","Macaroni or pasta salad with chicken"
+58148180,"MACARONI OR PASTA SALAD WITH CHEESE","Macaroni or pasta salad with cheese"
+58148550,"MACARONI OR PASTA SALAD W/ MEAT","Macaroni or pasta salad with meat"
+58148600,"PASTA TETRAZZINI,DRY MIX,PREPARED W/ WATER","Pasta tetrazzini, dry mix, prepared with water"
+58149110,"NOODLE PUDDING (INCLUDE KUGEL)","Noodle pudding"
+58149160,"NOODLE PUDDING,W/ MILK","Noodle pudding, with milk"
+58149210,"SOMEN SALAD W/ NOODLE, LETTUCE, EGG, FISH, PORK","Somen salad with noodles, lettuce, egg, fish, and pork"
+58150100,"BIBIMBAP (KOREAN)","Bibimbap (Korean)"
+58150110,"RICE, FRIED, MEATLESS","Rice, fried, meatless"
+58150310,"RICE, FRIED, NFS","Rice, fried, NFS"
+58150320,"RICE, FRIED, W/ CHICKEN (INCL TURKEY)","Rice, fried, with chicken"
+58150330,"RICE, FRIED, W/ PORK","Rice, fried, with pork"
+58150340,"RICE, FRIED, W/ BEEF","Rice, fried, with beef"
+58150510,"RICE, FRIED, W/ SHRIMP","Rice, fried, with shrimp"
+58150520,"DUKBOKI / TTEOKBOKKI (KOREAN)","Dukboki / Tteokbokki (Korean)"
+58151100,"SUSHI, NFS","Sushi, NFS"
+58151110,"SUSHI, NO VEG, NO SEAFOOD/FISH/SHELLFISH","Sushi, no vegetables, no seafood (no fish or shellfish)"
+58151120,"SUSHI, W/ VEG, NO SEAFOOD/FISH/SHELLFISH","Sushi, with vegetables, no seafood (no fish or shellfish)"
+58151130,"SUSHI, W/ VEG &  SEAFD","Sushi, with vegetables and seafood"
+58151140,"SUSHI, W/ VEGETABLES, ROLLED IN SEAWEED","Sushi, with vegetables, rolled in seaweed"
+58151150,"SUSHI, WITH SEAFOOD, NO VEGETABLES","Sushi, with seafood, no vegetables"
+58151160,"SUSHI, W/ EGG, NO VEG/SEAFD/FISH/SHLFISH, ROLL IN SEAWEED","Sushi, with egg, no vegetables, no seafood (no fish or shellfish), rolled in seaweed"
+58155110,"RICE W/ CHICKEN, P.R. (ARROZ CON POLLO)","Rice with chicken, Puerto Rican style (Arroz con Pollo)"
+58155210,"STUFFED RICE W/ CHICKEN, DOMINICAN STYLE","Stuffed rice with chicken, Dominican style (Arroz relleno Dominicano)"
+58155310,"PAELLA, VALENCIANA STYLE, W/ MEAT","Paella, Valenciana style, with meat (Paella Valenciana)"
+58155320,"SEAFOOD PAELLA, PUERTO RICAN STYLE","Seafood paella, Puerto Rican style"
+58155410,"SOUPY RICE W/ CHICKEN, P.R. (ASOPAO DE POLLO)","Soupy rice with chicken, Puerto Rican style (Asopao de pollo)"
+58155510,"SOUPY RICE MIXTURE W/ CHICKEN & POTATOES, P.R.STYLE","Soupy rice mixture with chicken and potatoes, Puerto Rican style"
+58155610,"RICE MEAL FRITTER, PUERTO RICAN (ALMOJAMBANA)","Rice meal fritter, Puerto Rican style (Almojabana)"
+58155810,"STEWED RICE, P.R. (ARROZ GUISADO)","Stewed rice, Puerto Rican style (arroz guisado)"
+58155910,"RICE W/ SQUID, P.R. (ARROZ CON CALAMARES)","Rice with squid, Puerto Rican style (arroz con calamares)"
+58156110,"FRIED RICE, P.R. (ARROZ FRITO)","Fried rice, Puerto Rican style (arroz frito)"
+58156210,"RICE W/ VIENNA SAUSAGE, P.R. (ARROZ CON SALCHICHAS)","Rice with vienna sausage, Puerto Rican style (arroz con salchichas)"
+58156310,"RICE W/ SPANISH SAUSAGE, P.R.","Rice with Spanish sausage, Puerto Rican style"
+58156410,"RICE W/ ONIONS, P.R. (ARROZ CON CEBOLLAS)","Rice with onions, Puerto Rican style (arroz con cebollas)"
+58156510,"SOUPY RICE, FROM P.R. ASOPAO DE POLLO","Soupy rice from Puerto Rican style Asopao de Pollo (chicken parts reported separately)"
+58156610,"PIGEON PEA ASOPAO (ASOPAO DE GRANDULES)","Pigeon pea asopao (Asopao de gandules)"
+58156710,"RICE W/ STEWED BEANS, P.R.","Rice with stewed beans, Puerto Rican style"
+58157110,"SPICEY RICE PUDDING, P.R.","Spicy rice pudding, Puerto Rican style"
+58157210,"RICE PUDDING MADE W/ COCONUT MILK, P.R.","Rice pudding made with coconut milk, Puerto Rican style"
+58160000,"BIRYANI WITH VEGETABLES","Biryani with vegetables"
+58160110,"RICE W/ BEANS","Rice with beans"
+58160120,"RICE W/ BEANS & TOMATOES","Rice with beans and tomatoes"
+58160130,"RICE W/ BEANS & CHICKEN","Rice with beans and chicken"
+58160135,"RICE W/ BEANS AND BEEF","Rice with beans and beef"
+58160140,"RICE W/ BEANS & PORK","Rice with beans and pork"
+58160150,"RED BEANS & RICE","Red beans and rice"
+58160160,"HOPPING JOHN (BLACKEYE PEAS & RICE)","Hopping John (blackeye peas and rice)"
+58160400,"RICE, WHITE, WITH CORN, NS AS TO FAT ADDED IN COOKING","Rice, white, with corn, NS as to fat added in cooking"
+58160410,"RICE, WHITE, WITH CORN, FAT NOT ADDED IN COOKING","Rice, white, with corn, fat not added in cooking"
+58160420,"RICE, WHITE, WITH CORN, FAT ADDED IN COOKING","Rice, white, with corn, fat added in cooking"
+58160430,"RICE, WHITE, WITH PEAS, NS AS TO FAT ADDED IN COOKING","Rice, white, with peas, NS as to fat added in cooking"
+58160440,"RICE, WHITE, WITH PEAS, FAT NOT ADDED IN COOKING","Rice, white, with peas, fat not added in cooking"
+58160450,"RICE, WHITE, WITH PEAS, FAT ADDED IN COOKING","Rice, white, with peas, fat added in cooking"
+58160460,"RICE, WHITE, WITH CARROTS, NS AS TO FAT ADDED IN COOKING","Rice, white, with carrots, NS as to fat added in cooking"
+58160470,"RICE, WHITE, WITH CARROTS, FAT NOT ADDED IN COOKING","Rice, white, with carrots, fat not added in cooking"
+58160480,"RICE, WHITE, WITH CARROTS, FAT ADDED IN COOKING","Rice, white, with carrots, fat added in cooking"
+58160490,"RICE, WHITE, W/ PEAS&CARROTS, NS AS TO FAT ADDED IN COOKING","Rice, white, with peas and carrots, NS as to fat added in cooking"
+58160500,"RICE, WHITE, WITH PEAS AND CARROTS, FAT NOT ADDED IN COOKING","Rice, white, with peas and carrots, fat not added in cooking"
+58160510,"RICE, WHITE, WITH PEAS AND CARROTS, FAT ADDED IN COOKING","Rice, white, with peas and carrots, fat added in cooking"
+58160520,"RICE, WHITE, W/TOMATOES/TOMATO BASED SAUCE, NS AS TO FAT","Rice, white, with tomatoes (and/or tomato based sauce), NS as to fat added in cooking"
+58160530,"RICE, WHITE, W/TOMATOES/ TOMATO BASED SAUCE, FAT NOT ADDED","Rice, white, with tomatoes (and/or tomato based sauce), fat not added in cooking"
+58160540,"RICE, WHITE, W/ TOMATOES/TOMATO BASED SAUCE, FAT ADDED","Rice, white, with tomatoes (and/or tomato based sauce), fat added in cooking"
+58160550,"RICE, WHITE, WITH DARK GREEN VEGETABLES, NS AS TO FAT ADDED","Rice, white, with dark green vegetables, NS as to fat added in cooking"
+58160560,"RICE, WHITE, WITH DARK GREEN VEGS, FAT NOT ADDED IN CO","Rice, white, with dark green vegetables, fat not added in cooking"
+58160570,"RICE, WHITE, WITH DARK GREEN VEGETABLES, FAT ADDED IN COOKIN","Rice, white, with dark green vegetables, fat added in cooking"
+58160580,"RICE, WHITE, W/ CARROTS, TOMATOES, +/OR TOM SC, NS AS TO FAT","Rice, white, with carrots and tomatoes (and/or tomato-based sauce), NS as to fat added in cooking"
+58160590,"RICE, WHITE, W/CARROTS, TOMATOES+/OR TOM SC, FAT NOT ADDED","Rice, white, with carrots and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58160600,"RICE, WHITE, W/ CARROTS, TOMATOES +/OR TOM SC, FAT ADDED","Rice, white, with carrots and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58160610,"RICE, WHITE, W/ DK GRN VEGS, TOMATOES +/OR TOM SC, NS FAT","Rice, white, with dark green vegetables and tomatoes (and/or tomato-based sauce), NS as to fat added in cooking"
+58160620,"RICE, WHITE, W/ DK GRN VEGS, TOMATOES +/OR TOM SC, FAT NOT","Rice, white, with dark green vegetables and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58160630,"RICE, WHITE, W/DK GRN VEGS, TOMATOES +/OR TOM SC , FAT ADDED","Rice, white, with dark green vegetables and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58160640,"RICE, WHITE, WITH CARROTS AND DARK GREEN VEGS, NS FAT","Rice, white, with carrots and dark green vegetables, NS as to fat added in cooking"
+58160650,"RICE, WHITE, WITH CARROTS AND DARK GREEN VEGS, FAT NOT ADDED","Rice, white, with carrots and dark green vegetables, fat not added in cooking"
+58160660,"RICE, WHITE, WITH CARROTS AND DARK GREEN VEGS, FAT ADDED","Rice, white, with carrots and dark green vegetables, fat added in cooking"
+58160670,"RICE, WHITE, W/ CARROTS, DARK GRN VEG&TOMATO/SC, NS FAT","Rice, white, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), NS as to fat added in cooking"
+58160680,"RICE, WHITE, W/ CARROTS, DARK GRN VEG,&TOMATOES,  NO FAT","Rice, white, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58160690,"RICE, WHITE, W/ CARROTS, DARK GRN VEG, & TOMATOES, W/ FAT","Rice, white, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58160700,"RICE, WHITE, WITH OTHER VEGS, NS AS TO FAT ADDED IN COOKING","Rice, white, with other vegetables, NS as to fat added in cooking"
+58160710,"RICE, WHITE, WITH OTHER VEGS, FAT NOT ADDED IN COOKING","Rice, white, with other vegetables, fat not added in cooking"
+58160720,"RICE, WHITE, WITH OTHER VEGS, FAT ADDED IN COOKING","Rice, white, with other vegetables, fat added in cooking"
+58160800,"RICE, WHITE, WITH LENTILS, NS AS TO FAT ADDED IN COOKING","Rice, white, with lentils, NS as to fat added in cooking"
+58160805,"RICE, WHITE, WITH LENTILS, FAT ADDED IN COOKING","Rice, white, with lentils, fat added in cooking"
+58160810,"RICE, WHITE, WITH LENTILS, FAT NOT ADDED IN COOKING","Rice, white, with lentils, fat not added in cooking"
+58161200,"RICE, COOKED W/ COCONUT MILK (ARROZ CON COCO)","Rice, cooked with coconut milk (Arroz con coco)"
+58161320,"RICE, BROWN, W/ BEANS","Rice, brown, with beans"
+58161325,"RICE, BROWN, W/ BEANS AND TOMATOES","Rice, brown, with beans and tomatoes"
+58161420,"RICE, BROWN, W/ CORN, NS AS TO FAT","Rice, brown, with corn, NS as to fat added in cooking"
+58161422,"RICE, BROWN, W/ CORN, FAT NOT ADDED","Rice, brown, with corn, fat not added in cooking"
+58161424,"RICE, BROWN, W/ CORN, FAT ADDED","Rice, brown, with corn, fat added in cooking"
+58161430,"RICE, BROWN, W/ PEAS, NS AS TO FAT","Rice, brown, with peas, NS as to fat added in cooking"
+58161432,"RICE, BROWN, W/ PEAS, FAT NOT ADDED","Rice, brown, with peas, fat not added in cooking"
+58161434,"RICE, BROWN, W/ PEAS, FAT ADDED","Rice, brown, with peas, fat added in cooking"
+58161435,"RICE, BROWN, WITH CARROTS, NS AS TO FAT ADDED IN COOKING","Rice, brown, with carrots, NS as to fat added in cooking"
+58161437,"RICE, BROWN, WITH CARROTS, FAT NOT ADDED IN COOKING","Rice, brown, with carrots, fat not added in cooking"
+58161439,"RICE, BROWN, WITH CARROTS, FAT ADDED IN COOKING","Rice, brown, with carrots, fat added in cooking"
+58161440,"RICE, BROWN, W/ PEAS AND CARROTS, NS AS TO FAT","Rice, brown, with peas and carrots, NS as to fat added in cooking"
+58161442,"RICE, BROWN, W/ PEAS AND CARROTS, FAT NOT ADDED","Rice, brown, with peas and carrots, fat not added in cooking"
+58161444,"RICE, BROWN, W/ PEAS AND CARROTS, FAT ADDED","Rice, brown, with peas and carrots, fat added in cooking"
+58161460,"RICE, BROWN, WITH TOMATOES AND/OR TOMATO BASED SAUCE, NS AS","Rice, brown, with tomatoes (and/or tomato based sauce), NS as to fat added in cooking"
+58161462,"RICE, BROWN, WITH TOMATOES AND/OR TOMATO BASED SAUCE, FAT NO","Rice, brown, with tomatoes (and/or tomato based sauce), fat not added in cooking"
+58161464,"RICE, BROWN, WITH TOMATOES AND/OR TOMATO BASED SAUCE, FAT AD","Rice, brown, with tomatoes (and/or tomato based sauce), fat added in cooking"
+58161470,"RICE, BROWN, WITH DARK GREEN VEGETABLES, NS AS TO FAT ADDED","Rice, brown, with dark green vegetables, NS as to fat added in cooking"
+58161472,"RICE, BROWN, WITH DARK GREEN VEGETABLES, FAT NOT ADDED IN CO","Rice, brown, with dark green vegetables, fat not added in cooking"
+58161474,"RICE, BROWN, WITH DARK GREEN VEGETABLES, FAT ADDED IN COOKIN","Rice, brown, with dark green vegetables, fat added in cooking"
+58161480,"RICE, BROWN, WITH CARROTS &TOMATOES/SC, NS AS TO FAT ADDED","Rice, brown, with carrots and tomatoes (and/or tomato-based sauce), NS as to fat added in cooking"
+58161482,"RICE, BROWN, WITH CARROTS & TOMATOES/SC, FAT NOT ADDED IN C","Rice, brown, with carrots and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58161484,"RICE, BROWN, WITH CARROTS & TOMATOES/SC, FAT ADDED IN COOKIN","Rice, brown, with carrots and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58161490,"RICE, BROWN, W/ DK GRN VEGS, TOMATOES/SC , NS AS TO FAT","Rice, brown, with dark green vegetables and tomatoes (and/or tomato-based sauce) , NS as to fat added in cooking"
+58161492,"RICE, BROWN, W/ DK GRN VEGS, TOMATOES/SC , FAT NOT ADDED","Rice, brown, with dark green vegetables and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58161494,"RICE, BROWN, W/ DK GRN VEGS, TOMATOES/SC , FAT ADDED","Rice, brown, with dark green vegetables and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58161500,"RICE, BROWN, WITH CARROTS AND DARK GREEN VEGETABLES, NS FAT","Rice, brown, with carrots and dark green vegetables, NS as to fat added in cooking"
+58161502,"RICE, BROWN, WITH CARROTS AND DARK GREEN VEGETABLES, FAT NOT","Rice, brown, with carrots and dark green vegetables, fat not added in cooking"
+58161504,"RICE, BROWN, WITH CARROTS AND DARK GREEN VEGETABLES, FAT ADD","Rice, brown, with carrots and dark green vegetables, fat added in cooking"
+58161510,"GRAPE LEAVES STUFFED W/ RICE","Grape leaves stuffed with rice"
+58161520,"RICE, BROWN, W/CARROTS, DK GRN VEGS,TOMATOES/SC, NS FAT","Rice, brown, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), NS as to fat added in cooking"
+58161522,"RICE, BROWN, W/ CARROTS, DK GRN VEGS,TOMATOES/SC, NO FAT","Rice, brown, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), fat not added in cooking"
+58161524,"RICE, BROWN, W/CARROTS, DK GRN VEGS,TOMATOES/SC, FAT ADDED","Rice, brown, with carrots, dark green vegetables, and tomatoes (and/or tomato-based sauce), fat added in cooking"
+58161530,"RICE, BROWN, W/ OTHER VEGS, NS AS TO FAT ADDED","Rice, brown, with other vegetables, NS as to fat added in cooking"
+58161532,"RICE, BROWN, WITH OTHER VEGETABLES, FAT NOT ADDED IN COOKING","Rice, brown, with other vegetables, fat not added in cooking"
+58161534,"RICE, BROWN, WITH OTHER VEGETABLES, FAT ADDED IN COOKING","Rice, brown, with other vegetables, fat added in cooking"
+58161710,"RICE CROQUETTE","Rice croquette"
+58162090,"STUFFED PEPPER W/ MEAT","Stuffed pepper, with meat"
+58162110,"STUFFED PEPPER, W/ RICE & MEAT","Stuffed pepper, with rice and meat"
+58162120,"STUFFED PEPPER, W/ RICE, MEATLESS","Stuffed pepper, with rice, meatless"
+58162130,"STUFFED TOMATO W/ RICE & MEAT","Stuffed tomato, with rice and meat"
+58162140,"STUFFED TOMATO W/ RICE, MEATLESS","Stuffed tomato, with rice, meatless"
+58162310,"RICE PILAF","Rice pilaf"
+58163130,"DIRTY RICE","Dirty rice"
+58163310,"FLAVORED RICE MIXTURE","Flavored rice mixture"
+58163330,"FLAVORED RICE MIXTURE W/ CHEESE","Flavored rice mixture with cheese"
+58163360,"FLAVORED RICE, BROWN & WILD","Flavored rice, brown and wild"
+58163380,"FLAVORED RICE&PASTA MIXTURE (INCL RICE-A-RONI)","Flavored rice and pasta mixture"
+58163400,"FLAVORED RICE & PASTA MIXTURE, REDUCED SODIUM","Flavored rice and pasta mixture, reduced sodium"
+58163410,"SPANISH RICE, FAT ADDED IN COOKING","Spanish rice, fat added in cooking"
+58163420,"SPANISH RICE, FAT NOT ADDED IN COOKING","Spanish rice, fat not added in cooking"
+58163430,"SPANISH RICE, NS AS TO FAT ADDED IN COOKING","Spanish rice, NS as to fat added in cooking"
+58163450,"SPANISH RICE W/ GROUND BEEF","Spanish rice with ground beef"
+58163510,"RICE DRESSING (INCLUDE COMBINED W/ BREAD)","Rice dressing"
+58164110,"RICE W/ RAISINS","Rice with raisins"
+58164210,"RICE DESSERT/SALAD W/ FRUIT","Rice dessert or salad with fruit"
+58164500,"RICE, WHITE, WITH CHEESE AND/OR CREAM BASED SAUCE, NS FAT","Rice, white, with cheese and/or cream based sauce, NS as to fat added in cooking"
+58164510,"RICE, WHITE, WITH CHEESE AND/OR CREAM BASED SAUCE, FAT NOT A","Rice, white, with cheese and/or cream based sauce, fat not added in cooking"
+58164520,"RICE, WHITE, WITH CHEESE AND/OR CREAM BASED SAUCE, FAT ADDED","Rice, white, with cheese and/or cream based sauce, fat added in cooking"
+58164530,"RICE, WHITE, WITH GRAVY, NS AS TO FAT ADDED IN COOKING","Rice, white, with gravy, NS as to fat added in cooking"
+58164540,"RICE, WHITE, WITH GRAVY, FAT NOT ADDED IN COOKING","Rice, white, with gravy, fat not added in cooking"
+58164550,"RICE, WHITE, WITH GRAVY, FAT ADDED IN COOKING","Rice, white, with gravy, fat added in cooking"
+58164560,"RICE, WHITE, WITH SOY BASED SAUCE, NS AS TO FAT ADDED IN COO","Rice, white, with soy-based sauce, NS as to fat added in cooking"
+58164570,"RICE, WHITE, WITH SOY BASED SAUCE, FAT NOT ADDED IN COOKING","Rice, white, with soy-based sauce, fat not added in cooking"
+58164580,"RICE, WHITE, WITH SOY BASED SAUCE, FAT ADDED IN COOKING","Rice, white, with soy-based sauce, fat added in cooking"
+58164800,"RICE, BROWN, WITH CHEESE AND/OR CREAM BASED SAUCE, NS FAT","Rice, brown, with cheese and/or cream based sauce, NS as to fat added in cooking"
+58164810,"RICE, BROWN, WITH CHEESE AND/OR CREAM BASED SAUCE, FAT NOT A","Rice, brown, with cheese and/or cream based sauce, fat not added in cooking"
+58164820,"RICE, BROWN, WITH CHEESE AND/OR CREAM BASED SAUCE, FAT ADDED","Rice, brown, with cheese and/or cream based sauce, fat added in cooking"
+58164830,"RICE, BROWN, WITH GRAVY, NS AS TO FAT ADDED IN COOKING","Rice, brown, with gravy, NS as to fat added in cooking"
+58164840,"RICE, BROWN, WITH GRAVY, FAT NOT ADDED IN COOKING","Rice, brown, with gravy, fat not added in cooking"
+58164850,"RICE, BROWN, WITH GRAVY, FAT ADDED IN COOKING","Rice, brown, with gravy, fat added in cooking"
+58164860,"RICE, BROWN, WITH SOY BASED SAUCE, NS AS TO FAT ADDED IN COO","Rice, brown, with soy-based sauce, NS as to fat added in cooking"
+58164870,"RICE, BROWN, WITH A SOY BASED SAUCE, FAT NOT ADDED IN COOKI","Rice, brown, with soy-based sauce, fat not added in cooking"
+58164880,"RICE, BROWN, WITH SOY BASED SAUCE, FAT ADDED IN COOKING","Rice, brown, with soy-based sauce, fat added in cooking"
+58165000,"RICE, WHITE, WITH VEGS, CHEESE +/OR CREAM BASED SC, NS FAT","Rice, white, with vegetables, cheese and/or cream based sauce, NS as to fat added in cooking"
+58165010,"RICE, WHITE, WITH VEGS, CHEESE +/OR CREAM BASED SC, FAT NOT","Rice, white, with vegetables, cheese and/or cream based sauce, fat not added in cooking"
+58165020,"RICE, WHITE, WITH VEGS, CHEESE +/OR CREAM BASED SC,FAT ADDED","Rice, white, with vegetables, cheese and/or cream based sauce, fat added in cooking"
+58165030,"RICE, WHITE, WITH VEGETABLES AND GRAVY, NS AS TO FAT ADDED","Rice, white, with vegetables and gravy, NS as to fat added in cooking"
+58165040,"RICE, WHITE, WITH VEGETABLES AND GRAVY, FAT NOT ADDED","Rice, white, with vegetables and gravy, fat not added in cooking"
+58165050,"RICE, WHITE, WITH VEGETABLES AND GRAVY, FAT ADDED IN COOKING","Rice, white, with vegetables and gravy, fat added in cooking"
+58165060,"RICE, WHITE, WITH VEGETABLES, SOY-BASED SAUCE, NS AS TO FAT","Rice, white, with vegetables, soy-based sauce, NS as to fat added in cooking"
+58165070,"RICE, WHITE, WITH VEGETABLES, SOY-BASED SAUCE, FAT NOT ADDED","Rice, white, with vegetables, soy-based sauce, fat not added in cooking"
+58165080,"RICE, WHITE, WITH VEGETABLES, SOY-BASED SAUCE, FAT ADDED IN","Rice, white, with vegetables, soy-based sauce, fat added in cooking"
+58165400,"RICE, BROWN, WITH VEGS, CHEESE +/OR CREAM BASED SC, NS FAT","Rice, brown, with vegetables, cheese and/or cream based sauce, NS as to fat added in cooking"
+58165410,"RICE, BROWN, WITH VEGS, CHEESE +/OR CREAM BASED SC, FAT NOT","Rice, brown, with vegetables, cheese and/or cream based sauce, fat not added in cooking"
+58165420,"RICE, BROWN, WITH VEGS, CHEESE +/OR CREAM BASED SC,FAT ADDED","Rice, brown, with vegetables, cheese and/or cream based sauce, fat added in cooking"
+58165430,"RICE, BROWN, WITH VEGETABLES AND GRAVY, NS AS TO FAT ADDED I","Rice, brown, with vegetables and gravy, NS as to fat added in cooking"
+58165440,"RICE, BROWN, WITH VEGETABLES AND GRAVY, FAT NOT ADDED","Rice, brown, with vegetables and gravy, fat not added in cooking"
+58165450,"RICE, BROWN, WITH VEGETABLES AND GRAVY, FAT ADDED IN COOKING","Rice, brown, with vegetables and gravy, fat added in cooking"
+58165460,"RICE, BROWN, WITH VEGETABLES, SOY-BASED SAUCE, NS AS TO FAT","Rice, brown, with vegetables, soy-based sauce, NS as to fat added in cooking"
+58165470,"RICE, BROWN, WITH VEGETABLES, SOY-BASED SAUCE, FAT NOT ADDED","Rice, brown, with vegetables, soy-based sauce, fat not added in cooking"
+58165480,"RICE, BROWN, WITH VEGETABLES, SOY-BASED SAUCE, FAT ADDED","Rice, brown, with vegetables, soy-based sauce, fat added in cooking"
+58174000,"UPMA (INDIAN BREAKFAST DISH)","Upma (Indian breakfast dish)"
+58175110,"TABBOULEH (INCLUDE TABBULI)","Tabbouleh (bulgar with tomatoes and parsley)"
+58200100,"WRAP SANDWICH, W/ MEAT, POULTRY OR FISH, VEGETABLES & RICE","Wrap sandwich, filled with meat, poultry, or fish, vegetables, and rice"
+58200200,"WRAP SANDWICH, W/ VEGETABLES & RICE","Wrap sandwich, filled with vegetables and rice"
+58200250,"WRAP SANDWICH, W/ VEGETABLES","Wrap sandwich, filled with vegetables"
+58200300,"WRAP SANDWICH, W/ MEAT, POULTRY, OR FISH, VEG, RICE & CHEESE","Wrap sandwich, filled with meat, poultry, or fish, vegetables, rice, and cheese"
+58301020,"LASAGNA W/ CHEESE & SAUCE (DIET FROZEN MEAL)","Lasagna with cheese and sauce (diet frozen meal)"
+58301030,"VEAL LASAGNA (DIET FROZEN MEAL) (INCL LEAN CUISINE)","Veal lasagna (diet frozen meal)"
+58301050,"LASAGNA, W/ CHEESE & MEAT SAUCE (DIET FROZEN MEAL)","Lasagna with cheese and meat sauce (diet frozen meal)"
+58301080,"LASAGNA W/CHEESE&MEAT SAU,REDUCED FAT&NA(DIET FROZ)","Lasagna with cheese and meat sauce, reduced fat and sodium (diet frozen meal)"
+58301110,"VEGETABLE LASAGNA (FROZEN MEAL)","Vegetable lasagna (frozen meal)"
+58301150,"ZUCCHINI LASAGNA (DIET FROZEN MEAL)","Zucchini lasagna (diet frozen meal)"
+58302000,"MACARONI & CHEESE (DIET FROZEN MEAL)","Macaroni and cheese (diet frozen meal)"
+58302050,"BEEF & NOODLES W/ MEAT SCE & CHEESE (DIET FRZ MEAL)","Beef and noodles with meat sauce and cheese (diet frozen meal)"
+58302060,"SPAG W/ BEEF, TOM-BASED SAUCE, LOWFAT, RED SODIUM, FRZ, DIET","Spaghetti or noodles with beef in tomato-based sauce, lowfat, reduced sodium (diet frozen meal)"
+58302080,"NOODLES W/ VEG, TOM-BASED SAUCE, FRZ, DIET","Noodles with vegetables in tomato-based sauce (diet frozen meal)"
+58303100,"RICE W/ BROC CHEESE SCE (FRZ SIDE DISH)","Rice, with broccoli, cheese sauce (frozen side dish)"
+58303200,"RICE,GREEN BEANS,WATER CHESTNUTS IN SCE (FRZ DISH)","Rice, with green beans, water chestnuts, in sherry mushroom sauce (frozen side dish)"
+58304010,"SPAGHETTI & MEATBALLS DINNER, NFS (FROZEN MEAL)","Spaghetti and meatballs dinner, NFS (frozen meal)"
+58304020,"SPAGHETTI,MEATBALLS,TOM SCE,APPLES,BREAD(FROZ MEAL)","Spaghetti and meatballs with tomato sauce, sliced apples, bread (frozen meal)"
+58304050,"SPAGHETTI W/ MEAT & MUSHROOM SAUCE (DIET FROZ MEAL)","Spaghetti with meat and mushroom sauce (diet frozen meal)"
+58304060,"SPAGHETTI W/ MEAT SAUCE (DIET FROZEN MEAL)","Spaghetti with meat sauce (diet frozen meal)"
+58304200,"RAVIOLI, CHEESE-FILLED, TOMATO SCE (DIET FROZ MEAL)","Ravioli, cheese-filled, with tomato sauce (diet frozen meal)"
+58304220,"RIGATONI W/ MEAT SCE & CHEESE (DIET FRZ MEAL)","Rigatoni with meat sauce and cheese (diet frozen meal)"
+58304230,"RAVIOLI, CHEESE-FILLED W/ VEG & FRUIT (FZN MEAL)","Ravioli, cheese-filled, with vegetable and fruit (frozen meal)"
+58304250,"MANICOTTI W/ CHEESE, TOMATO SAUCE (DIET FROZ MEAL)","Manicotti, cheese-filled, with tomato sauce (diet frozen meal)"
+58304300,"CANNELLONI, CHEESE-FILLED, TOM SCE (DIET FROZ MEAL)","Cannelloni, cheese-filled, with tomato sauce (diet frozen meal)"
+58304400,"LINGUINI W/ VEG & SEAFOOD IN SCE (DIET FROZEN MEAL)","Linguini with vegetables and seafood in white wine sauce (diet frozen meal)"
+58305250,"PASTA,W/ VEGETABLES & CHEESE SAUCE (DIET FROZ MEAL)","Pasta with vegetable and cheese sauce (diet frozen meal)"
+58306010,"BEEF ENCHILADA DINNER, NFS (FROZEN MEAL)","Beef enchilada dinner, NFS (frozen meal)"
+58306020,"BEEF ENCHILADA, GRAVY, RICE, REFRIED BEANS (FROZEN)","Beef enchilada, chili gravy, rice, refried beans (frozen meal)"
+58306070,"CHEESE ENCHILADA (FROZEN MEAL)","Cheese enchilada (frozen meal)"
+58306100,"CHICKEN ENCHILADA ( DIET FROZEN MEAL)","Chicken enchilada (diet frozen meal)"
+58306200,"CHICKEN FAJITAS (DIET FROZEN MEAL)","Chicken fajitas (diet frozen meal)"
+58306500,"CHICKEN BURRITOS (DIET FROZEN MEAL)","Chicken burritos (diet frozen meal)"
+58310210,"SAUSAGE & FRENCH TOAST (FROZEN MEAL)","Sausage and french toast (frozen meal)"
+58310310,"PANCAKE & SAUSAGE (FROZEN MEAL)","Pancakes and sausage (frozen meal)"
+58400000,"SOUP, NFS","Soup, NFS"
+58400100,"NOODLE SOUP, NFS","Noodle soup, NFS"
+58400200,"RICE SOUP, NFS","Rice soup, NFS"
+58401010,"BARLEY SOUP, HOME RECIPE, CANNED, OR READY-TO-SERVE","Barley soup, home recipe, canned, or ready-to-serve"
+58401200,"BARLEY SOUP, SWEET, WITH OR WITHOUT NUTS, ASIAN STYLE","Barley soup, sweet, with or without nuts, Asian Style"
+58402010,"BEEF NOODLE SOUP, CANNED OR READY-TO-SERVE","Beef noodle soup, canned or ready-to-serve"
+58402020,"BEEF DUMPLING SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Beef dumpling soup, home recipe, canned or ready-to-serve"
+58402030,"BEEF RICE SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Beef rice soup, home recipe, canned or ready-to-serve"
+58402100,"BEEF NOODLE SOUP, HOME RECIPE","Beef noodle soup, home recipe"
+58403010,"CHICKEN OR TURKEY NOODLE SOUP, CANNED OR READY-TO-SERVE","Chicken or turkey noodle soup, canned or ready-to-serve"
+58403040,"CHICKEN OR TURKEY NOODLE SOUP, HOME RECIPE","Chicken or turkey noodle soup, home recipe"
+58403050,"CHICKEN OR TURKEY NOODLE SOUP, CREAM OF, HOME RECIPE, CANNED","Chicken or turkey noodle soup, cream of, home recipe, canned, or ready-to-serve"
+58403060,"CHICKEN OR TURKEY NOODLE SOUP,  REDUCED SODIUM,CANNED, RTF","Chicken or turkey noodle soup, reduced sodium, canned or ready-to-serve"
+58403100,"NOODLE & POTATO SOUP, P.R.","Noodle and potato soup, Puerto Rican style"
+58404010,"CHICKEN OR TURKEY RICE SOUP, CANNED, OR READY-TO-SERVE","Chicken or turkey rice soup, canned, or ready-to-serve"
+58404030,"CHICKEN OR TURKEY  RICE SOUP, HOME RECIPE","Chicken or turkey rice soup, home recipe"
+58404040,"CHICKEN OR TURKEY RICE SOUP, REDUCED SODIUM, CANNED, PREPARE","Chicken or turkey rice soup, reduced sodium, canned, prepared with water or ready-to-serve"
+58404050,"CHICKEN OR TURKEY RICE SOUP, REDUCED SODIUM, CANNED, PREPARE","Chicken or turkey rice soup, reduced sodium, canned, prepared with milk"
+58404100,"RICE AND POTATO SOUP, P.R.","Rice and potato soup, Puerto Rican style"
+58404500,"MATZO BALL SOUP","Matzo ball soup"
+58404510,"CHICKEN OR TURKEY SOUP WITH DUMPLINGS AND POTATOES,","Chicken or turkey soup with dumplings and potatoes, home recipe, canned, or ready-to-serve"
+58404520,"CHICKEN OR TURKEY SOUP WITH DUMPLINGS, HOME RECIPE, CANNED O","Chicken or turkey soup with dumplings, home recipe, canned or ready-to-serve"
+58407010,"INSTANT SOUP, NOODLE","Instant soup, noodle"
+58407030,"SOUP, MOSTLY NOODLES","Soup, mostly noodles"
+58407035,"SOUP, MOSTLY NOODLES, REDUCED SODIUM","Soup, mostly noodles, reduced sodium"
+58407050,"INSTANT SOUP, NOODLE W/ EGG, SHRIMP OR CHICKEN","Instant soup, noodle with egg, shrimp or chicken"
+58408010,"WON TON (WONTON) SOUP","Won ton (wonton) soup"
+58408500,"NOODLE SOUP WITH VEGETABLES, ASIAN STYLE","Noodle soup with vegetables, Asian style"
+58409000,"NOODLE SOUP,W/ FISH BALL,SHRIMP,&DK GREEN LEAFY VEG","Noodle soup, with fish ball, shrimp, and dark green leafy vegetable"
+58421000,"SOPA SECA (DRY SOUP), Mexican style, NFS","Sopa seca (dry soup), Mexican style, NFS"
+58421010,"SOPA SECA DE FIDEO, MEXICAN STYLE, MADE WITH DRY NOODLES, HO","Sopa Seca de Fideo, Mexican style, made with dry noodles, home recipe"
+58421020,"SOPA DE FIDEO AGUADA, MEXICAN STYLE NOODLE SOUP, HOME RECIPE","Sopa de Fideo Aguada, Mexican style noodle soup, home recipe"
+58421060,"SOPA SECA DE ARROZ (DRY RICE SOUP), MEXICAN STYLE, HOME RECI","Sopa seca de arroz (dry rice soup), Mexican style, home recipe"
+58421080,"SOPA DE TORTILLA, MEXICAN STYLE TORTILLA SOUP, HOME RECIPE","Sopa de tortilla, Mexican style tortilla soup, home recipe"
+58503000,"MACARONI, TOMATOES & BEEF, BABY, NS STR/JR","Macaroni, tomatoes, and beef, baby food, NS as to strained or junior"
+58503010,"MACARONI, TOMATOES & BEEF, BABY, STR","Macaroni, tomatoes, and beef, baby food, strained"
+58503020,"MACARONI, TOMATOES & BEEF, BABY, JR","Macaroni, tomatoes, and beef, baby food, junior"
+58503050,"MACARONI W/ BEEF & TOM SCE, BABY FOOD, TODDLER","Macaroni with beef and tomato sauce, baby food, toddler"
+58508000,"MACARONI & CHEESE, BABY, STRAINED","Macaroni and cheese, baby food, strained"
+58508300,"MACARONI & CHEESE, BABY, TODDLER","Macaroni and cheese, baby food, toddler"
+58509020,"SPAGHETTI, TOMATO SAUCE & BEEF, BABY, JUNIOR","Spaghetti, tomato sauce, and beef, baby food, junior"
+58509100,"RAVIOLI, CHEESE-FILLED, W/ TOM SAUCE, BABY, TODDLER","Ravioli, cheese-filled, with tomato sauce, baby food, toddler"
+58509200,"MACARONI W/ VEGETABLES, BABY, STRAINED","Macaroni with vegetables, baby food, strained"
+59003000,"MEAT SUBSTITUTE,CEREAL- & VEGETABLE PROTEIN-BASED","Meat substitute, cereal- and vegetable protein-based, fried"
+61100500,"CALAMONDIN, RAW","Calamondin, raw"
+61101010,"GRAPEFRUIT, RAW (INCLUDE GRAPEFRUIT, NFS)","Grapefruit, raw"
+61101200,"GRAPEFRUIT, CANNED OR FROZEN, NS AS TO ADDED SWTNER","Grapefruit, canned or frozen, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+61101220,"GRAPEFRUIT, CANNED OR FROZEN, UNSWT, WATER PACK","Grapefruit, canned or frozen, unsweetened, water pack"
+61101230,"GRAPEFRUIT, CANNED OR FROZEN, IN LIGHT SYRUP","Grapefruit, canned or frozen, in light syrup"
+61104010,"GRAPEFRUIT & ORANGE SECTIONS, RAW","Grapefruit and orange sections, raw"
+61104200,"GRAPEFRUIT & ORANGE SEC, CKD/CND/FRZ, NS SWEETENER","Grapefruit and orange sections, cooked, canned, or frozen, NS as to added sweetener"
+61104220,"GRAPEFRUIT & ORANGE SEC, CKD/CND/FRZ, UNSWEETENED","Grapefruit and orange sections, cooked, canned, or frozen, unsweetened, water pack"
+61104230,"GRAPEFRUIT & ORANGE SEC, CKD/CND/FRZ, LIGHT SYRUP","Grapefruit and orange sections, cooked, canned, or frozen, in light syrup"
+61110010,"KUMQUAT, RAW","Kumquat, raw"
+61110230,"KUMQUAT, COOKED OR CANNED, IN SYRUP","Kumquat, cooked or canned, in syrup"
+61113010,"LEMON, RAW","Lemon, raw"
+61113500,"LEMON PIE FILLING","Lemon pie filling"
+61116010,"LIME, RAW","Lime, raw"
+61119010,"ORANGE, RAW","Orange, raw"
+61119020,"ORANGE SECTIONS, CANNED, JUICE PACK","Orange, sections, canned, juice pack"
+61119100,"ORANGE PEEL","Orange peel"
+61122300,"ORANGES, MANDARIN, CANNED OR FROZEN, SWEETENER NS","Orange, mandarin, canned or frozen, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+61122320,"ORANGES, MANDARIN, CANNED OR FROZEN, JUICE PACK","Orange, mandarin, canned or frozen, juice pack"
+61122330,"ORANGES, MANDARIN, CANNED OR FROZEN, IN LIGHT SYRUP","Orange, mandarin, canned or frozen, in light syrup"
+61122350,"ORANGES, MANDARIN, CANNED OR FROZEN, DRAINED","Orange, mandarin, canned or frozen, drained"
+61125000,"TANGELO, RAW","Tangelo, raw"
+61125010,"TANGERINE, RAW (INCLUDE MANDARIN ORANGE, SATSUMA)","Tangerine, raw"
+61201010,"GRAPEFRUIT JUICE, FRESHLY SQUEEZED","Grapefruit juice, freshly squeezed"
+61201020,"GRAPEFRUIT JUICE, NS AS TO FORM","Grapefruit juice, NS as to form"
+61201220,"GRAPEFRUIT JUICE, CANNED, BOTTLED OR IN A CARTON","Grapefruit juice, canned, bottled or in a carton"
+61201620,"GRAPEFRUIT JUICE, FROZEN (RECONSTITUTED WITH WATER)","Grapefruit juice, frozen (reconstituted with water)"
+61204000,"LEMON JUICE, NS AS TO FORM","Lemon juice, NS as to form"
+61204010,"LEMON JUICE, FRESHLY SQUEEZED","Lemon juice, freshly squeezed"
+61204200,"LEMON JUICE, CANNED OR BOTTLED","Lemon juice, canned or bottled"
+61204600,"LEMON JUICE, FROZEN","Lemon juice, frozen"
+61207000,"LIME JUICE, NS AS TO FORM","Lime juice, NS as to form"
+61207010,"LIME JUICE, FRESHLY SQUEEZED","Lime juice, freshly squeezed"
+61207200,"LIME JUICE, CANNED OR BOTTLED","Lime juice, canned or bottled"
+61207600,"LIME JUICE, FROZEN","Lime juice, frozen"
+61210000,"ORANGE JUICE, NFS","Orange juice, NFS"
+61210010,"ORANGE JUICE, FRESHLY SQUEEZED","Orange juice, freshly squeezed"
+61210220,"ORANGE JUICE, CANNED, BOTTLED OR IN A CARTON","Orange juice, canned, bottled or in a carton"
+61210250,"ORANGE JUICE, W/ CALCIUM, CAN/BOTTLED/CARTON","Orange juice, with calcium added, canned, bottled or in a carton"
+61210620,"ORANGE JUICE, FROZEN (RECONSTITUTED WITH WATER)","Orange juice, frozen (reconstituted with water)"
+61210720,"ORANGE JUICE, FROZEN, NOT RECONSTITUTED","Orange juice, frozen, not reconstituted"
+61210820,"ORANGE JUICE,FROZ, W/,CALCIUM ADDED,RECON W/WATER","Orange juice, frozen, with calcium added (reconstituted with water)"
+61213000,"TANGERINE JUICE, NFS","Tangerine juice, NFS"
+61213220,"TANGERINE JUICE, CANNED","Tangerine juice, canned"
+61213620,"TANGERINE JUICE, FROZEN (RECONSTITUTED)","Tangerine juice, frozen (reconstituted with water)"
+61213800,"FRUIT JUICE BLEND, INCL CITRUS, 100% JUICE","Fruit juice blend, including citrus, 100% juice"
+61213900,"FRUIT JUICE BLEND, INCL CITRUS, 100% JUICE, W/ CALCIUM","Fruit juice blend, including citrus, 100% juice, with calcium added"
+62101000,"FRUIT, DRIED, NFS (ASSUME UNCOOKED)","Fruit, dried, NFS (assume uncooked)"
+62101050,"FRUIT MIXTURE, DRIED","Fruit mixture, dried (mixture includes three or more of the following: apples, apricots, dates, papaya, peaches, pears, pineapples, prunes, raisins)"
+62101100,"APPLE, DRIED, UNCOOKED","Apple, dried, uncooked"
+62101150,"APPLE, DRIED, UNCOOKED, LOW SODIUM","Apple, dried, uncooked, low sodium"
+62101200,"APPLE, DRIED, COOKED, NS AS TO ADDED SWEETENER","Apple, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62101220,"APPLE, DRIED, COOKED, UNSWEETENED","Apple, dried, cooked, unsweetened"
+62101230,"APPLE, DRIED, COOKED, W/ SUGAR","Apple, dried, cooked, with sugar"
+62101300,"APPLE CHIPS","Apple chips"
+62104100,"APRICOT, DRIED, UNCOOKED","Apricot, dried, uncooked"
+62104200,"APRICOT, DRIED, COOKED, NS AS TO ADDED SWEETENER","Apricot, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62104220,"APRICOT, DRIED, COOKED, UNSWEETENED","Apricot, dried, cooked, unsweetened"
+62104230,"APRICOT, DRIED, COOKED, W/ SUGAR","Apricot, dried, cooked, with sugar"
+62105000,"BLUEBERRIES, DRIED","Blueberries, dried"
+62106000,"CHERRIES, DRIED","Cherries, dried"
+62107100,"BANANA FLAKES, DEHYDRATED","Banana flakes, dehydrated"
+62107200,"BANANA CHIPS","Banana chips"
+62108100,"CURRANTS, DRIED","Currants, dried"
+62109100,"CRANBERRIES, DRIED","Cranberries, dried"
+62110100,"DATE","Date"
+62113100,"FIG, DRIED, UNCOOKED","Fig, dried, uncooked"
+62113200,"FIG, DRIED, COOKED, NS AS TO ADDED SWEETENER","Fig, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62113220,"FIG, DRIED, COOKED, UNSWEETENED","Fig, dried, cooked, unsweetened"
+62113230,"FIG, DRIED, COOKED, W/ SUGAR","Fig, dried, cooked, with sugar"
+62114000,"LYCHEE, DRIED (LYCHEE NUTS)","Lychee, dried (lychee nuts)"
+62114050,"MANGO, DRIED","Mango, dried"
+62114110,"PAPAYA, DRIED","Papaya, dried"
+62116100,"PEACH, DRIED, UNCOOKED","Peach, dried, uncooked"
+62116200,"PEACH, DRIED, COOKED, NS AS TO ADDED SWEETENER","Peach, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62116220,"PEACH, DRIED, COOKED, UNSWEETENED","Peach, dried, cooked, unsweetened"
+62116230,"PEACH, DRIED, COOKED, W/ SUGAR","Peach, dried, cooked, with sugar"
+62119100,"PEAR, DRIED, UNCOOKED","Pear, dried, uncooked"
+62119200,"PEAR, DRIED, COOKED, NS AS TO SWEETENER","Pear, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62119220,"PEAR, DRIED, COOKED, UNSWEETENED","Pear, dried, cooked, unsweetened"
+62119230,"PEAR, DRIED, COOKED, W/ SUGAR","Pear, dried, cooked, with sugar"
+62120100,"PINEAPPLE, DRIED","Pineapple, dried"
+62121100,"PLUM, ROCK SALT, DRIED","Plum, rock salt, dried"
+62122100,"PRUNE, DRIED, UNCOOKED","Prune, dried, uncooked"
+62122200,"PRUNE, DRIED, COOKED, NS AS TO ADDED SWEETENER","Prune, dried, cooked, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+62122220,"PRUNE, DRIED, COOKED, UNSWEETENED","Prune, dried, cooked, unsweetened"
+62122230,"PRUNE, DRIED, COOKED, W/ SUGAR","Prune, dried, cooked, with sugar"
+62125100,"RAISINS (INCLUDE CINNAMON-COATED RAISINS)","Raisins"
+62125110,"RAISINS, COOKED","Raisins, cooked"
+62126000,"TAMARIND PULP, DRIED, SWEETENED (""PULPITAS"")","Tamarind pulp, dried, sweetened (""Pulpitas"")"
+63100100,"FRUIT, NS AS TO TYPE","Fruit, NS as to type"
+63101000,"APPLE, RAW","Apple, raw"
+63101110,"APPLESAUCE, STEWED APPLES, NS AS TO ADDED SWEETENER","Applesauce, stewed apples, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63101120,"APPLESAUCE, STEWED APPLES, UNSWEETENED (INCL FRESH)","Applesauce, stewed apples, unsweetened"
+63101130,"APPLESAUCE, STEWED APPLES, W/ SUGAR","Applesauce, stewed apples, with sugar"
+63101140,"APPLESAUCE, STEWED APPLES, W/ LOW CALORIE SWEETENER","Applesauce, stewed apples, sweetened with low calorie sweetener"
+63101150,"APPLESAUCE / OTHER FRUITS(INCLUDE MOTT'S FRUIT PAK)","Applesauce with other fruits"
+63101210,"APPLE, COOKED OR CANNED, W/ SYRUP","Apple, cooked or canned, with syrup"
+63101310,"APPLE, BAKED, NS AS TO ADDED SWEETENER","Apple, baked, NS as to added sweetener"
+63101320,"APPLE, BAKED, UNSWEETENED","Apple, baked, unsweetened"
+63101330,"APPLE, BAKED, W/ SUGAR","Apple, baked, with sugar"
+63101410,"APPLE RINGS, FRIED","Apple rings, fried"
+63101420,"APPLE, PICKLED (INCLUDE SPICED)","Apple, pickled"
+63101500,"APPLE, FRIED","Apple, fried"
+63103010,"APRICOT, RAW","Apricot, raw"
+63103110,"APRICOT, COOKED OR CANNED, NS AS TO ADDED SWEETENER","Apricot, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63103120,"APRICOT, COOKED OR CANNED, WATER PACK, UNSWEETENED","Apricot, cooked or canned, unsweetened, water pack"
+63103130,"APRICOT, COOKED OR CANNED, IN HEAVY SYRUP","Apricot, cooked or canned, in heavy syrup"
+63103140,"APRICOT, COOKED OR CANNED, IN LIGHT SYRUP","Apricot, cooked or canned, in light syrup"
+63103150,"APRICOT, COOKED OR CANNED, DRAINED SOLIDS","Apricot, cooked or canned, drained solids"
+63103170,"APRICOT, COOKED OR CANNED, JUICE PACK","Apricot, cooked or canned, juice pack"
+63105010,"AVOCADO, RAW","Avocado, raw"
+63107010,"BANANA, RAW","Banana, raw"
+63107050,"BANANA, WHITE, RIPE (GUINEO BLANCO MADURO)","Banana, white, ripe (guineo blanco maduro)"
+63107070,"BANANA, CHINESE, RAW (INCL CAVENDISH,DWARF,FINGER)","Banana, Chinese, raw"
+63107080,"BANANA, RED, RIPE (INCLUDE GUINEO MORADO)","Banana, red, ripe (guineo morado)"
+63107090,"BANANA, RED, FRIED","Banana, red, fried"
+63107110,"BANANA, BAKED","Banana, baked"
+63107210,"BANANA, RIPE, FRIED","Banana, ripe, fried"
+63107310,"BANANA, RIPE, BOILED","Banana, ripe, boiled"
+63107410,"BANANA, BATTER-DIPPED, FRIED","Banana, batter-dipped, fried"
+63109010,"CANTALOUPE (MUSKMELON), RAW (INCLUDE MELON, NFS)","Cantaloupe (muskmelon), raw"
+63109610,"CANTALOUPE, FROZEN (BALLS)","Cantaloupe, frozen (balls)"
+63109700,"CARAMBOLA (STARFRUIT),RAW","Carambola (starfruit), raw"
+63109750,"CARAMBOLA (STARFRUIT), COOKED, W/ SUGAR","Carambola (starfruit), cooked, with sugar"
+63110010,"CASSABA MELON, RAW","Cassaba melon, raw"
+63111010,"CHERRIES, MARASCHINO","Cherries, maraschino"
+63113010,"CHERRIES, SOUR, RED, RAW","Cherries, sour, red, raw"
+63113030,"CHERRY PIE FILLING","Cherry pie filling"
+63113050,"CHERRY PIE FILLING, LOW CALORIE","Cherry pie filling, low calorie"
+63113110,"CHERRIES, SOUR, RED, COOKED, UNSWEETENED","Cherries, sour, red, cooked, unsweetened"
+63115010,"CHERRIES, SWEET, RAW (INCLUDE CHERRIES, FRESH, NFS)","Cherries, sweet, raw (Queen Anne, Bing)"
+63115110,"CHERRIES, SWEET, COOKED/CANNED,NS AS TO ADDED SWEET","Cherries, sweet, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63115120,"CHERRIES, SWEET, COOKED, UNSWEETENED, WATER PACK","Cherries, sweet, cooked, unsweetened, water pack"
+63115130,"CHERRIES, SWEET, COOKED OR CANNED, IN HEAVY SYRUP","Cherries, sweet, cooked or canned, in heavy syrup"
+63115140,"CHERRIES, SWEET, COOKED OR CANNED, IN LIGHT SYRUP","Cherries, sweet, cooked or canned, in light syrup"
+63115150,"CHERRIES, SWEET, COOKED OR CANNED, DRAINED SOLIDS","Cherries, sweet, cooked or canned, drained solids"
+63115170,"CHERRIES, SWEET, COOKED OR CANNED, JUICE PACK","Cherries, sweet, cooked or canned, juice pack"
+63115200,"CHERRIES, FROZEN","Cherries, frozen"
+63117010,"CURRANTS, RAW","Currants, raw"
+63119010,"FIG, RAW","Fig, raw"
+63119110,"FIG, COOKED OR CANNED, NS AS TO ADDED SWEETENER","Fig, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63119120,"FIG, COOKED OR CANNED, UNSWEETENED, WATER PACK","Fig, cooked or canned, unsweetened, water pack"
+63119130,"FIG, COOKED OR CANNED, IN HEAVY SYRUP","Fig, cooked or canned, in heavy syrup"
+63119140,"FIGS, COOKED OR CANNED, IN LIGHT SYRUP","Figs, cooked or canned, in light syrup"
+63123000,"GRAPES, RAW, NS AS TO TYPE","Grapes, raw, NS as to type"
+63123010,"GRAPES, EUROPEAN TYPE,ADHERENT SKIN,RAW(INCL TOKAY)","Grapes, European type, adherent skin, raw"
+63123020,"GRAPES, AMERICAN TYPE, SLIP SKIN, RAW(INCL CONCORD)","Grapes, American type, slip skin, raw"
+63123110,"GRAPES, SEEDLESS, COOKED/CANNED, NS ADDED SWEETNER","Grapes, seedless, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63123120,"GRAPES, SEEDLESS, COOKED OR CANNED, UNSWEETENED","Grapes, seedless, cooked or canned, unsweetened, water pack"
+63123130,"GRAPES, SEEDLESS, COOKED OR CANNED, IN HEAVY SYRUP","Grapes, seedless, cooked or canned, in heavy syrup"
+63125010,"GUAVA, RAW","Guava, raw"
+63125100,"GUAVA SHELL (ASSUME CANNED IN HEAVY SYRUP)","Guava shell (assume canned in heavy syrup)"
+63126010,"JUNEBERRY, RAW","Juneberry, raw"
+63126500,"KIWI FRUIT, RAW","Kiwi fruit, raw"
+63126510,"LYCHEE, RAW","Lychee, raw"
+63126600,"LYCHEE, COOKED OR CANNED, IN SUGAR OR SYRUP","Lychee, cooked or canned, in sugar or syrup"
+63127010,"HONEYDEW MELON, RAW","Honeydew melon, raw"
+63127610,"HONEYDEW MELON, FROZEN (BALLS)","Honeydew, frozen (balls)"
+63129010,"MANGO, RAW","Mango, raw"
+63129020,"MANGO, PICKLED","Mango, pickled"
+63129030,"MANGO, COOKED","Mango, cooked"
+63131010,"NECTARINE, RAW","Nectarine, raw"
+63131110,"NECTARINE, COOKED","Nectarine, cooked"
+63133010,"PAPAYA, RAW","Papaya, raw"
+63133050,"PAPAYA, GREEN, COOKED","Papaya, green, cooked"
+63133100,"PAPAYA, COOKED OR CANNED, IN SUGAR OR SYRUP","Papaya, cooked or canned, in sugar or syrup"
+63134010,"PASSION FRUIT, RAW","Passion fruit, raw"
+63135010,"PEACH, RAW","Peach, raw"
+63135110,"PEACH, COOKED OR CANNED, NS AS TO SWEETENER","Peach, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63135120,"PEACH, COOKED OR CANNED, UNSWEETENED, WATER PACK","Peach, cooked or canned, unsweetened, water pack"
+63135130,"PEACH, COOKED OR CANNED, IN HEAVY SYRUP","Peach, cooked or canned, in heavy syrup"
+63135140,"PEACH, COOKED OR CANNED, IN LIGHT OR MEDIUM SYRUP","Peach, cooked or canned, in light or medium syrup"
+63135150,"PEACH, COOKED OR CANNED, DRAINED SOLIDS","Peach, cooked or canned, drained solids"
+63135170,"PEACH, COOKED OR CANNED, JUICE PACK","Peach, cooked or canned, juice pack"
+63135610,"PEACH, FROZEN, NS AS TO ADDED SWEETENER","Peach, frozen, NS as to added sweetener"
+63135620,"PEACH, FROZEN, UNSWEETENED","Peach, frozen, unsweetened"
+63135630,"PEACH, FROZEN, W/ SUGAR","Peach, frozen, with sugar"
+63135650,"PEACH, PICKLED","Peach, pickled"
+63135660,"PEACH, SPICED","Peach, spiced"
+63137010,"PEAR, RAW","Pear, raw"
+63137050,"PEAR, JAPANESE, RAW","Pear, Japanese, raw"
+63137110,"PEAR, COOKED OR CANNED, NS AS TO ADDED SWEETENER","Pear, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63137120,"PEAR, COOKED OR CANNED, UNSWEETENED, WATER PACK","Pear, cooked or canned, unsweetened, water pack"
+63137130,"PEAR, COOKED OR CANNED, IN HEAVY SYRUP","Pear, cooked or canned, in heavy syrup"
+63137140,"PEAR, COOKED OR CANNED, IN LIGHT SYRUP","Pear, cooked or canned, in light syrup"
+63137150,"PEAR, COOKED OR CANNED, DRAINED SOLIDS","Pear, cooked or canned, drained solids"
+63137170,"PEAR, COOKED OR CANNED, JUICE PACK","Pear, cooked or canned, juice pack"
+63139010,"PERSIMMONS, RAW","Persimmon, raw"
+63141010,"PINEAPPLE, RAW","Pineapple, raw"
+63141110,"PINEAPPLE, CANNED, NS AS TO ADDED SWEETENER","Pineapple, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63141120,"PINEAPPLE, COOKED OR CANNED, UNSWEETENED, WATERPACK","Pineapple, cooked or canned, unsweetened, waterpack"
+63141130,"PINEAPPLE, COOKED OR CANNED, IN HEAVY SYRUP","Pineapple, cooked or canned, in heavy syrup"
+63141140,"PINEAPPLE, COOKED OR CANNED, IN LIGHT SYRUP","Pineapple, cooked or canned, in light syrup"
+63141150,"PINEAPPLE, COOKED OR CANNED, DRAINED SOLIDS","Pineapple, cooked or canned, drained solids"
+63141170,"PINEAPPLE, COOKED OR CANNED, JUICE PACK","Pineapple, cooked or canned, juice pack"
+63143010,"PLUM, RAW","Plum, raw"
+63143110,"PLUM, COOKED OR CANNED, NS AS TO ADDED SWEETENER","Plum, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63143120,"PLUM, COOKED OR CANNED, UNSWEETENED, WATER PACK","Plum, cooked or canned, unsweetened, water pack"
+63143130,"PLUM, COOKED OR CANNED, IN HEAVY SYRUP","Plum, cooked or canned, in heavy syrup"
+63143140,"PLUM, COOKED OR CANNED, IN LIGHT SYRUP","Plum, cooked or canned, in light syrup"
+63143150,"PLUM, COOKED OR CANNED, DRAINED SOLIDS","Plum, cooked or canned, drained solids"
+63143170,"PLUM, COOKED OR CANNED, JUICE PACK","Plum, cooked or canned, juice pack"
+63143650,"PLUM, PICKLED","Plum, pickled"
+63145010,"POMEGRANATE, RAW","Pomegranate, raw"
+63147010,"RHUBARB, RAW","Rhubarb, raw"
+63147110,"RHUBARB, COOKED OR CANNED, NS ADDED SWEETNER","Rhubarb, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63147120,"RHUBARB, COOKED OR CANNED, UNSWEETENED","Rhubarb, cooked or canned, unsweetened"
+63147130,"RHUBARB, COOKED OR CANNED, IN HEAVY SYRUP","Rhubarb, cooked or canned, in heavy syrup"
+63147140,"RHUBARB, COOKED OR CANNED, IN LIGHT SYRUP","Rhubarb, cooked or canned, in light syrup"
+63147150,"RHUBARB, COOKED OR CANNED, DRAINED SOLIDS","Rhubarb, cooked or canned, drained solids"
+63147600,"RHUBARB, FROZEN, NS AS TO ADDED SWEETENER","Rhubarb, frozen, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63147620,"RHUBARB, FROZEN, W/ SUGAR","Rhubarb, frozen, with sugar"
+63148750,"TAMARIND, RAW","Tamarind, raw"
+63149010,"WATERMELON, RAW","Watermelon, raw"
+63200100,"BERRIES, RAW, NFS","Berries, raw, NFS"
+63200200,"BERRIES, FROZEN, NFS","Berries, frozen, NFS"
+63201010,"BLACKBERRIES, RAW","Blackberries, raw"
+63201110,"BLACKBERRIES, COOKED OR CANNED, NS ADDED SWEETNER","Blackberries, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63201130,"BLACKBERRIES, COOKED OR CANNED, IN HEAVY SYRUP","Blackberries, cooked or canned, in heavy syrup"
+63201600,"BLACKBERRIES, FROZEN","Blackberries, frozen"
+63201800,"BLACKBERRIES, FROZEN, SWEETENED, NFS","Blackberries, frozen, sweetened, NS as to type of sweetener"
+63203010,"BLUEBERRIES, RAW","Blueberries, raw"
+63203110,"BLUEBERRIES, COOKED OR CANNED, NS AS TO SWEETENER","Blueberries, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63203120,"BLUEBERRIES, COOKED OR CANNED, UNSWEET, WATER PACK","Blueberries, cooked or canned, unsweetened, water pack"
+63203125,"BLUEBERRIES, COOKED OR CANNED, IN LIGHT SYRUP","Blueberries, cooked or canned, in light syrup"
+63203130,"BLUEBERRIES, COOKED OR CANNED, IN HEAVY SYRUP","Blueberries, cooked or canned, in heavy syrup"
+63203550,"BLUEBERRIES, FROZEN, SWEETENED","Blueberries, frozen, sweetened"
+63203570,"BLUEBERRIES, FROZEN, NS SWEETENED OR UNSWEETENED","Blueberries, frozen, NS as to sweetened or unsweetened"
+63203600,"BLUEBERRIES, FROZEN, UNSWEETENED","Blueberries, frozen, unsweetened"
+63203700,"BLUEBERRY PIE FILLING","Blueberry pie filling"
+63205010,"BOYSENBERRIES, RAW","Boysenberries, raw"
+63205600,"BOYSENBERRIES, FROZEN","Boysenberries, frozen"
+63207000,"CRANBERRIES, NS AS TO RAW, COOKED OR CANNED","Cranberries, NS as to raw, cooked, or canned"
+63207010,"CRANBERRIES, RAW","Cranberries, raw"
+63207110,"CRANBERRIES, COOKED OR CANNED (INCL CRANBERRY SCE)","Cranberries, cooked or canned"
+63208000,"DEWBERRIES, RAW","Dewberries, raw"
+63214000,"HUCKLEBERRIES, RAW","Huckleberries, raw"
+63215010,"LOGANBERRIES, RAW","Loganberries, raw"
+63215600,"LOGANBERRIES, FROZEN","Loganberries, frozen"
+63217010,"MULBERRIES, RAW","Mulberries, raw"
+63219000,"RASPBERRIES, RAW, NS AS TO COLOR","Raspberries, raw, NS as to color"
+63219010,"RASPBERRIES, BLACK, RAW","Raspberries, black, raw"
+63219020,"RASPBERRIES, RED, RAW","Raspberries, red, raw"
+63219110,"RASPBERRIES, COOKED OR CANNED, NS ADDED SWEETENER","Raspberries, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63219120,"RASPBERRIES, CKD OR CND, UNSWEETENED, WATER PACK","Raspberries, cooked or canned, unsweetened, water pack"
+63219130,"RASPBERRIES, COOKED OR CANNED, IN HEAVY SYRUP","Raspberries, cooked or canned, in heavy syrup"
+63219600,"RASPBERRIES, FROZEN, NS AS TO ADDED SWEETNER","Raspberries, frozen, NS as to added sweetener"
+63219610,"RASPBERRIES, FROZEN, UNSWEETENED","Raspberries, frozen, unsweetened"
+63219620,"RASPBERRIES, FROZEN, W/ SUGAR","Raspberries, frozen, with sugar"
+63223020,"STRAWBERRIES, RAW","Strawberries, raw"
+63223030,"STRAWBERRIES, RAW, W/ SUGAR","Strawberries, raw, with sugar"
+63223110,"STRAWBERRIES, COOKED OR CANNED, NS ADDED SWEETNER","Strawberries, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63223120,"STRAWBERRIES, CKD OR CND, UNSWEETENED, WATER PACK","Strawberries, cooked or canned, unsweetened, water pack"
+63223130,"STRAWBERRIES, COOKED OR CANNED, IN SYRUP","Strawberries, cooked or canned, in syrup"
+63223600,"STRAWBERRIES, FROZEN, NS AS TO ADDED SWEETNER","Strawberries, frozen, NS as to added sweetener"
+63223610,"STRAWBERRIES, FROZEN, UNSWEETENED","Strawberries, frozen, unsweetened"
+63223620,"STRAWBERRIES, FROZEN, W/ SUGAR","Strawberries, frozen, with sugar"
+63224000,"YOUNGBERRIES, RAW","Youngberries, raw"
+63301010,"AMBROSIA","Ambrosia"
+63307010,"CRANBERRY-ORANGE RELISH, UNCOOKED","Cranberry-orange relish, uncooked"
+63307100,"CRANBERRY-RASPBERRY SAUCE","Cranberry-raspberry Sauce"
+63311000,"FRUIT SALAD, FRESH OR RAW, (EXCL CITRUS), NO DRSG","Fruit salad, fresh or raw, (excluding citrus fruits), no dressing"
+63311050,"FRUIT SALAD, FRESH OR RAW, (INCL CITRUS), NO DRSG","Fruit salad, fresh or raw, (including citrus fruits), no dressing"
+63311080,"FRUIT COCKTAIL OR MIX, FROZEN","Fruit cocktail or mix, frozen"
+63311110,"FRUIT COCKTAIL, COOKED OR CANNED, NS ADDED SWEETNER","Fruit cocktail, cooked or canned, NS as to sweetened or unsweetened; sweetened, NS as to type of sweetener"
+63311120,"FRUIT COCKTAIL, CKD OR CND, UNSWEET, WATER PACK","Fruit cocktail, cooked or canned, unsweetened, water pack"
+63311130,"FRUIT COCKTAIL, COOKED OR CANNED, IN HEAVY SYRUP","Fruit cocktail, cooked or canned, in heavy syrup"
+63311140,"FRUIT COCKTAIL, COOKED OR CANNED, IN LIGHT SYRUP","Fruit cocktail, cooked or canned, in light syrup"
+63311145,"TROPICAL FRUIT COCKTAIL, COOKED OR CANNED, IN LIGHT SYRUP","Tropical fruit cocktail, cooked or canned, in light syrup"
+63311150,"FRUIT COCKTAIL, COOKED OR CANNED, DRAINED SOLIDS","Fruit cocktail, cooked or canned, drained solids"
+63311170,"FRUIT COCKTAIL, COOKED OR CANNED, JUICE PACK","Fruit cocktail, cooked or canned, juice pack"
+63320100,"FRUIT SALAD, P.R. STYLE (ENSALADA DE FRUTA)","Fruit salad, Puerto Rican style (Mixture includes bananas, papayas, oranges, etc.) (Ensalada de frutas tropicales)"
+63401010,"APPLE SALAD W/ DRESSING (INCLUDE WALDORF SALAD)","Apple salad with dressing"
+63401015,"APPLE AND GRAPE SALAD W/ YOGURT & WALNUTS","Apple and grape salad with yogurt and walnuts"
+63401060,"APPLE, CANDIED (INCLUDE CARAMEL APPLES)","Apple, candied"
+63401070,"FRUIT, CHOCOLATE COVERED","Fruit, chocolate covered"
+63401990,"BANANA, CHOCOLATE-COVERED, W/ NUTS","Banana, chocolate-covered with nuts"
+63402010,"BANANA WHIP","Banana whip"
+63402030,"PRUNE WHIP","Prune whip"
+63402045,"FRIED DWARF BANANA, PUERTO RICAN STYLE","Fried dwarf banana, Puerto Rican style"
+63402050,"FRIED DWARF BANANA W/ CHEESE, PUERTO RICAN STYLE","Fried dwarf banana with cheese, Puerto Rican style"
+63402950,"FRUIT SALAD (NO CITRUS) W/ SALAD DRESSING","Fruit salad (excluding citrus fruits) with salad dressing or mayonnaise"
+63402960,"FRUIT SALAD (EXCLUDING CITRUS FRUITS) WITH WHIPPED CREAM","Fruit salad (excluding citrus fruits) with whipped cream"
+63402970,"FRUIT SALAD (EXCLUDING CITRUS FRUITS) WITH NONDAIRY WHIPPED","Fruit salad (excluding citrus fruits) with nondairy whipped topping"
+63402980,"FRUIT SALAD (NO CITRUS) W/ MARSHMALLOWS","Fruit salad (excluding citrus fruits) with marshmallows"
+63402990,"FRUIT SALAD (W/ CITRUS) W/ PUDDING","Fruit salad (including citrus fruits) with pudding"
+63403000,"FRUIT SALAD (NO CITRUS FRUITS) W/ PUDDING","Fruit salad (excluding citrus fruits) with pudding"
+63403010,"FRUIT SALAD (INCL CITRUS FRUITS) W/ SALAD DRESSING","Fruit salad (including citrus fruits) with salad dressing or mayonnaise"
+63403020,"FRUIT SALAD (INCLUDING CITRUS FRUIT) WITH WHIPPED CREAM","Fruit salad (including citrus fruit) with whipped cream"
+63403030,"FRUIT SALAD (INCLUDING CITRUS FRUITS) WITH NONDAIRY WHIPPED","Fruit salad (including citrus fruits) with nondairy whipped topping"
+63403040,"FRUIT SALAD W/ MARSHMALLOWS","Fruit salad (including citrus fruits) with marshmallows"
+63403100,"FRUIT DESSERT W/ CREAM & OR PUDDING & NUTS","Fruit dessert with cream and/or pudding and nuts"
+63403150,"LIME SOUFFLE (INCLUDE OTHER CITRUS FRUITS)","Lime souffle"
+63409010,"GUACAMOLE","Guacamole"
+63409020,"CHUTNEY","Chutney"
+63411010,"CRANBERRY SALAD, CONGEALED","Cranberry salad, congealed"
+63413010,"PINEAPPLE SALAD W/ DRESSING","Pineapple salad with dressing"
+63415100,"SOUP, FRUIT","Soup, fruit"
+63420100,"FRUIT JUICE BAR, FROZEN, ORANGE FLAVOR","Fruit juice bar, frozen, orange flavor"
+63420110,"FRUIT JUICE BAR, FROZEN, FLAVOR OTHER THAN ORANGE","Fruit juice bar, frozen, flavor other than orange"
+63420200,"FRUIT JUICE BAR, FROZ, LOW CAL SWEETNER, NOT ORANGE","Fruit juice bar, frozen, sweetened with low calorie sweetener, flavors other than orange"
+63430100,"SORBET, FRUIT, NONCITRUS FLAVOR","Sorbet, fruit, noncitrus flavor"
+63430110,"SORBET, FRUIT, CITRUS FLAVOR","Sorbet, fruit, citrus flavor"
+63430500,"FRUIT JUICE BAR W/ CREAM, FROZEN","Fruit juice bar with cream, frozen"
+64100100,"FRUIT JUICE, NFS","Fruit juice, NFS"
+64100110,"FRUIT JUICE BLEND, 100% JUICE","Fruit juice blend, 100% juice"
+64100200,"FRUIT JUICE BLEND, WITH CRANBERRY, 100% JUICE","Fruit juice blend, with cranberry, 100% juice"
+64101010,"APPLE CIDER (INCLUDE CIDER, NFS)","Apple cider"
+64104010,"APPLE JUICE","Apple juice"
+64104600,"BLACKBERRY JUICE (INCL BOYSENBERRY JUICE)","Blackberry juice"
+64105400,"CRANBERRY JUICE, 100%, NOT A BLEND","Cranberry juice, 100%, not a blend"
+64116020,"GRAPE JUICE","Grape juice"
+64120010,"PAPAYA JUICE","Papaya juice"
+64121000,"PASSION FRUIT JUICE","Passion fruit juice"
+64124020,"PINEAPPLE JUICE","Pineapple juice"
+64126000,"POMEGRANATE JUICE","Pomegranate juice"
+64132010,"PRUNE JUICE","Prune juice"
+64132500,"STRAWBERRY JUICE","Strawberry juice"
+64133100,"WATERMELON JUICE","Watermelon juice"
+64134000,"FRUIT SMOOTHIE DRINK, W/ FRUIT OR JUICE ONLY (NO DAIRY)","Fruit smoothie drink, made with fruit or fruit juice only (no dairy products)"
+64200100,"FRUIT NECTAR, NFS","Fruit nectar, NFS"
+64201010,"APRICOT NECTAR","Apricot nectar"
+64201500,"BANANA NECTAR","Banana nectar"
+64202010,"CANTALOUPE NECTAR","Cantaloupe nectar"
+64203020,"GUAVA NECTAR","Guava nectar"
+64204010,"MANGO NECTAR","Mango nectar"
+64205010,"PEACH NECTAR","Peach nectar"
+64210010,"PAPAYA NECTAR","Papaya nectar"
+64213010,"PASSION FRUIT NECTAR","Passion fruit nectar"
+64215010,"PEAR NECTAR","Pear nectar"
+64221010,"SOURSOP (GUANABANA) NECTAR","Soursop (Guanabana) nectar"
+64401000,"VINEGAR","Vinegar"
+67100100,"FRUIT, BABY, NFS","Fruit, baby food, NFS"
+67100110,"FRUIT BAR, WITH ADDED VITAMIN C, BABY FOOD, TODDLER","Fruit bar, with added vitamin C, baby food, toddler"
+67100200,"TROPICAL FRUIT MEDLEY, BABY FOOD, STRAINED","Tropical fruit medley, baby food, strained"
+67100300,"APPLES, BABY, TODDLER","Apples, baby food, toddler"
+67101000,"APPLE-RASPBERRY, BABY, NS AS TO STRAINED OR JUNIOR","Apple-raspberry, baby food, NS as to strained or junior"
+67101010,"APPLE-RASPBERRY, BABY, STRAINED","Apple-raspberry, baby food, strained"
+67101020,"APPLE-RASPBERRY, BABY, JUNIOR","Apple-raspberry, baby food, junior"
+67102000,"APPLESAUCE, BABY, NS AS TO STRAINED OR JUNIOR","Applesauce, baby food, NS as to strained or junior"
+67102010,"APPLESAUCE, BABY, STRAINED","Applesauce, baby food, strained"
+67102020,"APPLESAUCE, BABY, JUNIOR","Applesauce, baby food, junior"
+67104000,"APPLESAUCE & APRICOTS, BABY, NS AS TO STR OR JR","Applesauce and apricots, baby food, NS as to strained or junior"
+67104010,"APPLESAUCE & APRICOTS, BABY, STRAINED","Applesauce and apricots, baby food, strained"
+67104020,"APPLESAUCE & APRICOTS, BABY, JUNIOR","Applesauce and apricots, baby food, junior"
+67104030,"APPLESAUCE W/ BANANAS, BABY, NS STRAINED/JUNIOR","Applesauce with bananas, baby food, NS as to strained or junior"
+67104040,"APPLESAUCE W/ BANANAS, BABY, STRAINED","Applesauce with bananas, baby food, strained"
+67104060,"APPLESAUCE W/ BANANAS, BABY, JUNIOR","Applesauce with bananas, baby food, junior"
+67104070,"APPLESAUCE W/ CHERRIES, BABY, STRAINED","Applesauce with cherries, baby food, strained"
+67104080,"APPLESAUCE W/ CHERRIES, BABY, JUNIOR","Applesauce with cherries, baby food, junior"
+67104090,"APPLESAUCE W/ CHERRIES, BABY, NS STRAINED/JUNIOR","Applesauce with cherries, baby food, NS as to strained or junior"
+67105030,"BANANAS,BABY FOOD,STRAINED","Bananas, baby food, strained"
+67106010,"BANANAS W/ APPLES & PEARS, BABY, STRAINED","Bananas with apples and pears, baby food, strained"
+67106050,"BANANA WITH MIXED BERRIES, BABY FOOD, STRAINED","Banana with mixed berries, baby food, strained"
+67108000,"PEACHES, BABY, NS AS TO STRAINED OR JUNIOR","Peaches, baby food, NS as to strained or junior"
+67108010,"PEACHES, BABY, STRAINED","Peaches, baby food, strained"
+67108020,"PEACHES, BABY, JUNIOR","Peaches, baby food, junior"
+67108030,"PEACHES, BABY, TODDLER","Peaches, baby food, toddler"
+67109000,"PEARS, BABY, NS AS TO STRAINED OR JUNIOR","Pears, baby food, NS as to strained or junior"
+67109010,"PEARS, BABY, STRAINED","Pears, baby food, strained"
+67109020,"PEARS, BABY, JUNIOR","Pears, baby food, junior"
+67109030,"PEARS, BABY, TODDLER","Pears, baby food, toddler"
+67110000,"PRUNES, BABY, STRAINED","Prunes, baby food, strained"
+67113000,"APPLES & PEARS, BABY, NS AS TO STRAINED OR JUNIOR","Apples and pears, baby food, NS as to strained or junior"
+67113010,"APPLES & PEARS, BABY, STRAINED","Apples and pears, baby food, strained"
+67113020,"APPLES & PEARS, BABY, JUNIOR","Apples and pears, baby food, junior"
+67114000,"PEARS & PINEAPPLE, BABY, NS AS TO STR OR JR","Pears and pineapple, baby food, NS as to strained or junior"
+67114010,"PEARS & PINEAPPLE, BABY, STRAINED","Pears and pineapple, baby food, strained"
+67114020,"PEARS & PINEAPPLE, BABY, JUNIOR","Pears and pineapple, baby food, junior"
+67202000,"APPLE JUICE, BABY","Apple juice, baby food"
+67202010,"APPLE JUICE, W/ CALCIUM, BABY","Apple juice, with added calcium, baby food"
+67203000,"APPLE W/ OTHER FRUIT JUICE, BABY","Apple with other fruit juice, baby food"
+67203200,"APPLE-BANANA JUICE, BABY","Apple-banana juice, baby food"
+67203400,"APPLE-CHERRY JUICE, BABY","Apple-cherry juice, baby food"
+67203450,"APPLE-CRANBERRY JUICE, BABY","Apple-cranberry juice, baby food"
+67203500,"APPLE-GRAPE JUICE, BABY","Apple-grape juice, baby food"
+67203600,"APPLE-PEACH JUICE, BABY","Apple-peach juice, baby food"
+67203700,"APPLE-PRUNE JUICE, BABY","Apple-prune juice, baby food"
+67203800,"GRAPE JUICE, BABY","Grape juice, baby food"
+67204000,"MIXED FRUIT JUICE, NOT CITRUS, BABY","Mixed fruit juice, not citrus, baby food"
+67204100,"MIXED FRUIT JUICE, NOT CITRUS, W/ CALCIUM, BABY","Mixed fruit juice, not citrus, with added calcium, baby food"
+67205000,"ORANGE JUICE, BABY","Orange juice, baby food"
+67211000,"ORANGE-APPLE-BANANA JUICE, BABY","Orange-apple-banana juice, baby food"
+67212000,"PEAR JUICE, BABY FOOD","Pear juice, baby food"
+67230000,"APPLE-SWEET POTATO-JUICE,BABY FOOD","Apple-sweet potato juice, baby food"
+67230500,"ORANGE-CARROT JUICE, BABY FOOD","Orange-carrot juice, baby food"
+67250100,"BANANA JUICE W/ LOWFAT YOGURT, BABY FOOD","Banana juice with lowfat yogurt, baby food"
+67250150,"MIXED FRUIT JUICE W/ LOWFAT YOGURT, BABY FOOD","Mixed fruit juice with lowfat yogurt, baby food"
+67260000,"FRUIT JUICE DRINK, BABY, W/ HI VIT C + CA, B VITS","Fruit juice drink, baby, with high vitamin C plus added calcium and B vitamins"
+67304000,"PLUMS, BABY, NS AS TO STRAINED OR JUNIOR","Plums, baby food, NS as to strained or junior"
+67304010,"PLUMS, BABY, STRAINED","Plums, baby food, strained"
+67304020,"PLUMS, BABY, JUNIOR","Plums, baby food, junior"
+67304030,"PLUMS, BANANAS & RICE, BABY, STRAINED","Plums, bananas, and rice, baby food strained"
+67304500,"PRUNES W/ OATMEAL, BABY, STRAINED","Prunes with oatmeal, baby food, strained"
+67307000,"APRICOTS, BABY, NS AS TO STR OR JR","Apricots, baby food, NS as to strained or junior"
+67307010,"APRICOTS, BABY, STRAINED","Apricots, baby food, strained"
+67307020,"APRICOTS, BABY, JUNIOR","Apricots, baby food, junior"
+67308000,"BANANAS, BABY, NS AS TO STR OR JR","Bananas, baby food, NS as to strained or junior"
+67308020,"BANANAS, BABY, JUNIOR","Bananas, baby food, junior"
+67309000,"BANANAS & PINEAPPLE, BABY,NS AS TO STR/JR","Bananas and pineapple, baby food, NS as to strained or junior"
+67309010,"BANANAS & PINEAPPLE, BABY, STRAINED","Bananas and pineapple, baby food, strained"
+67309020,"BANANAS & PINEAPPLE, BABY, JUNIOR","Bananas and pineapple, baby food, junior"
+67309030,"BANANAS AND STRAWBERRY, BABY FOOD, JUNIOR","Bananas and strawberry, baby food, junior"
+67404000,"FRUIT DESSERT, BABY, NS AS TO STR OR JR","Fruit dessert, baby food, NS as to strained or junior"
+67404010,"FRUIT DESSERT, BABY, STRAINED","Fruit dessert, baby food, strained"
+67404020,"FRUIT DESSERT, BABY, JUNIOR","Fruit dessert, baby food, junior"
+67404050,"FRUIT SUPREME DESSERT, BABY, ALL FLAVORS","Fruit Supreme dessert, baby food"
+67404070,"APPLE YOGURT DESSERT, BABY, STRAINED","Apple yogurt dessert, baby food, strained"
+67404110,"BANANA APPLE DESSERT, BABY FOOD, STRAINED","Banana apple dessert, baby food, strained"
+67404300,"BLUEBERRY YOGURT DESSERT, BABY, STRAINED","Blueberry yogurt dessert, baby food, strained"
+67404500,"MIXED FRUIT YOGURT DESSERT, BABY, STRAINED","Mixed fruit yogurt dessert, baby food, strained"
+67404550,"CHERRY COBBLER, BABY, JUNIOR","Cherry cobbler, baby food, junior"
+67405000,"PEACH COBBLER, BABY, NS AS TO STRAINED OR JUNIOR","Peach cobbler, baby food, NS as to strained or junior"
+67405010,"PEACH COBBLER, BABY, STRAINED","Peach cobbler, baby food, strained"
+67405020,"PEACH COBBLER, BABY, JUNIOR","Peach cobbler, baby food, junior"
+67408010,"BANANA PUDDING, BABY, STRAINED","Banana pudding, baby food, strained"
+67408500,"BANANA YOGURT DESSERT, BABY, STRAINED","Banana yogurt dessert, baby food, strained"
+67410000,"CHERRY VANILLA PUDDING, BABY","Cherry vanilla pudding, baby food, strained"
+67412000,"DUTCH APPLE DESSERT, BABY, NS AS TO STR OR JR","Dutch apple dessert, baby food, NS as to strained or junior"
+67412010,"DUTCH APPLE DESSERT, BABY, STRAINED","Dutch apple dessert, baby food, strained"
+67412020,"DUTCH APPLE DESSERT, BABY, JUNIOR","Dutch apple dessert, baby food, junior"
+67413700,"PEACH YOGURT DESSERT, BABY, STRAINED","Peach yogurt dessert, baby food, strained"
+67414010,"PINEAPPLE DESSERT, BABY, STRAINED","Pineapple dessert, baby food, strained"
+67414100,"MANGO DESSERT,  BABY","Mango dessert, baby food"
+67415000,"TUTTI-FRUITTI PUDDING, BABY, NS AS TO STR OR JR","Tutti-fruitti pudding, baby food, NS as to strained or junior"
+67415010,"TUTTI-FRUTTI PUDDING, BABY, STRAINED","Tutti-fruitti pudding, baby food, strained"
+67415020,"TUTTI-FRUITTI PUDDING, BABY, JUNIOR","Tutti-fruitti pudding, baby food, junior"
+67430000,"FRUIT FLAVORED SNACK, BABY FOOD","Fruit flavored snack, baby food"
+67430500,"YOGURT AND FRUIT SNACK, BABY FOOD","Yogurt and fruit snack, baby food"
+67501000,"APPLES & CHICKEN, BABY FOOD, STRAINED","Apples and chicken, baby food, strained"
+67501100,"APPLES W/ HAM, BABY, STRAINED","Apples with ham, baby food, strained"
+67600100,"APPLES & SWEET POTATOES, BABY, STRAINED","Apples and sweet potatoes, baby food, strained"
+71000100,"WHITE POTATO, NFS","White potato, NFS"
+71001000,"WHITE POTATO, RAW, W/ OR W/O PEEL","White potato, raw, with or without peel (assume peel not eaten)"
+71050000,"WHITE POTATO, DRY, POWDERED, NOT RECONSTITUTED","White potato, dry, powdered, not reconstituted"
+71101000,"WHITE POTATO, BAKED, PEEL NOT EATEN","White potato, baked, peel not eaten"
+71101100,"WHITE POT,BAKED,PEEL EATEN,NS TO FAT ADDED IN COOK","White potato, baked, peel eaten, NS as to fat added in cooking"
+71101110,"WHITE POT,BAKED,PEEL EATEN,FAT NOT ADDED IN COOKING","White potato, baked, peel eaten, fat not added in cooking"
+71101120,"WHITE POT, BAKED,PEEL EATEN, FAT ADDED IN COOKING","White potato, baked, peel eaten, fat added in cooking"
+71101150,"WHITE POTATO SKINS, W/ ADHERING FLESH, BAKED","White potato skins, with adhering flesh, baked"
+71103000,"WHITE POTATO, BOILED, W/O PEEL, NS AS TO FAT","White potato, boiled, without peel, NS as to fat added in cooking"
+71103010,"WHITE POTATO, BOILED, W/O PEEL, FAT NOT ADDED","White potato, boiled, without peel, fat not added in cooking"
+71103020,"WHITE POTATO, BOILED, W/O PEEL, FAT ADDED","White potato, boiled, without peel, fat added in cooking"
+71103100,"WHITE POTATO, BOILED W/ PEEL, PEEL NOT EATEN, NS AS TO FAT","White potato, boiled with peel, peel not eaten, NS as to fat added in cooking"
+71103110,"WHITE POTATO, BOILED W/PEEL, PEEL NOT EATEN, FAT NOT ADDED","White potato, boiled with peel, peel not eaten, fat not added in cooking"
+71103120,"WHITE POTATO, BOILED W/ PEEL, PEEL NOT EATEN, FAT ADDED","White potato, boiled with peel, peel not eaten, fat added in cooking"
+71103200,"WHITE POTATO, CANNED, LOW SODIUM,NS AS TO ADDED FAT","White potato, boiled, without peel, canned, low sodium, NS as to fat added in cooking"
+71103210,"WHITE POTATO, CANNED, LOW SODIUM, NO FAT ADDED","White potato, boiled, without peel, canned, low sodium, fat not added in cooking"
+71103220,"WHITE POTATO, CANNED, LOW SODIUM, FAT ADDED","White potato, boiled, without peel, canned, low sodium, fat added in cooking"
+71104000,"WHITE POTATO, ROASTED, NS FAT ADDED","White potato, roasted, NS as to fat added in cooking"
+71104010,"WHITE POTATO, ROASTED, FAT NOT ADDED","White potato, roasted, fat not added in cooking"
+71104020,"WHITE POTATO, ROASTED, FAT ADDED","White potato, roasted, fat added in cooking"
+71106000,"STEWED POTATOES, P.R. (PAPAS GUISADAS)","Stewed potatoes, Puerto Rican style (Papas guisadas)"
+71106010,"POTATO ONLY FROM P.R. MIXED DISHES","Potato only from Puerto Rican mixed dishes, gravy and other components reported separately"
+71106020,"POTATO FROM PUERTO RICAN STYLE,POT ROAST, W/ GRAVY","Potato from Puerto Rican style stuffed pot roast, with gravy"
+71106050,"POTATO FROM PUERTO RICAN BEEF STEW, W/ GRAVY","Potato from Puerto Rican beef stew, with gravy"
+71106070,"POTATO FROM PUERTO RICAN CHICKEN FRICASSEE, W/ SCE","Potato from Puerto Rican chicken fricassee, with sauce"
+71201015,"WHITE POTATO CHIPS, REGULAR CUT","White potato chips, regular cut"
+71201020,"WHITE POTATO CHIPS, RUFFLED/RIPPLED/CRINKLE CUT","White potato chips, ruffled, rippled, or crinkle cut"
+71201050,"WHITE POTATO, CHIPS, REDUCED FAT","White potato, chips, reduced fat"
+71201080,"WHITE POTATO, CHIPS, FAT FREE","White potato, chips, fat free"
+71201090,"WHITE POTATO, CHIPS, FAT FREE, W/ OLEAN","White potato, chips, fat free, made with Olean"
+71201100,"WHITE POTATO, CHIPS, RESTRUCTURED","White potato, chips, restructured"
+71201200,"WHITE POTATO, CHIPS, RESTRUCTURED, RED FAT/SODIUM","White potato, chips, restructured, reduced fat and reduced sodium"
+71201210,"WHITE POTATO, CHIPS, RESTRUCTURED, FAT FREE, W/ OLEAN","White potato, chips, restructured, fat free, made with Olean"
+71201250,"WHITE POTATO, CHIPS, RESTRUCTURED, BAKED","White potato, chips, restructured, baked"
+71201300,"POTATO-BASED SNACKS, REDUCED FAT, LOW SODIUM","Potato based snacks, reduced fat, low sodium, all flavors"
+71202000,"WHITE POTATO, CHIPS, UNSALTED","White potato, chips, unsalted"
+71202100,"WHITE POTATO, CHIPS, UNSALTED, REDUCED FAT","White potato, chips, unsalted, reduced fat"
+71202500,"WHITE POTATO CHIPS, LIGHTLY SALTED","White potato chips, lightly salted"
+71204000,"POTATO PUFFS, CHEESE-FILLED","Potato puffs, cheese-filled"
+71205000,"WHITE POTATO, STICKS","White potato, sticks"
+71211000,"WHITE POTATO SKINS, CHIPS","White potato skins, chips"
+71220000,"VEGETABLE CHIPS","Vegetable chips"
+71301000,"WHITE POTATO, COOKED, W/ SAUCE, NS AS TO SAUCE","White potato, cooked, with sauce, NS as to sauce"
+71301020,"WHITE POTATO, COOKED, WITH CHEESE","White potato, cooked, with cheese"
+71301120,"WHITE POTATO, COOKED, WITH HAM AND CHEESE","White potato, cooked, with ham and cheese"
+71305010,"WHITE POTATO, SCALLOPED","White potato, scalloped"
+71305110,"WHITE POTATO, SCALLOPED, W/ HAM","White potato, scalloped, with ham"
+71401000,"WHITE POTATO, FRENCH FRIES, NS AS TO FROM FRESH/FRZ","White potato, french fries, NS as to from fresh or frozen"
+71401010,"WHITE POTATO, FRENCH FRIES, FROM FRESH, DEEP-FRIED","White potato, french fries, from fresh, deep fried"
+71401015,"WHITE POTATO, FRENCH FRIES, FROM FRESH, OVEN BAKED","White potato, french fries, from fresh, oven baked"
+71401020,"WHITE POTATO, FRENCH FRIES, FROM FROZEN, OVEN-BAKED","White potato, french fries, from frozen, oven baked"
+71401030,"WHITE POTATO, FRENCH FRIES, FRM FRZ, DEEP FRD, FF/REST","White potato, french fries, from frozen, deep fried, from fast food / restaurant"
+71401035,"WHITE POTATO, FRENCH FRIES, FR FRZN, NS AS TO FRIED OR BKD","White potato, french fries, from frozen, NS as to deep fried or oven baked"
+71402500,"WHITE POTATO, FRENCH FRIES, W/ CHEESE","White potato, french fries, with cheese"
+71402505,"WHITE POTATO, FRENCH FRIES, W/ CHEESE AND BACON","White potato, french fries, with cheese and bacon"
+71402510,"WHITE POTATO, FRENCH FRIES, W/ CHILI & CHEESE","White potato, french fries, with chili and cheese"
+71402520,"WHITE POTATO, FRENCH FRIES, W/ CHILI CON CARNE","White potato, french fries, with chili con carne"
+71403000,"WHITE POTATO, HOME FRIES","White potato, home fries"
+71403500,"WHITE POTATO, HOME FRIES, W/ GREEN/RED PEPPERS & ONIONS","White potato, home fries, with green or red peppers and onions"
+71405000,"WHITE POTATO, HASH BROWN","White potato, hash brown, NS as to from fresh, frozen, or dry mix"
+71405010,"WHITE POTATO, HASH BROWN, FROM FRESH","White potato, hash brown, from fresh"
+71405020,"WHITE POTATO, HASH BROWN, FROM FROZEN","White potato, hash brown, from frozen"
+71405030,"WHITE POTATO, HASH BROWN, FROM DRY MIX","White potato, hash brown, from dry mix"
+71405100,"WHITE POTATO, HASH BROWN W/ CHEESE","White potato, hash brown, with cheese"
+71410000,"WHITE POTATO SKINS, W/ ADHERING FLESH, FRIED","White potato skins, with adhering flesh, fried"
+71410500,"WHITE POTATO SKINS W/ FLESH, FRIED, W/ CHEESE","White potato skins, with adhering flesh, fried, with cheese"
+71411000,"POTATO SKINS W/ ADHERING FLESH, W/ CHEESE & BACON","White potato skins, with adhering flesh, fried, with cheese and bacon"
+71501000,"WHITE POTATO, MASHED, NFS","White potato, mashed, NFS"
+71501010,"WHITE POTATO, FRESH, MASHED, MADE W/ MILK","White potato, from fresh, mashed, made with milk"
+71501015,"WHITE POTATO, FRESH, MASHED, MADE W/ MILK,/ SOUR CRM/CHEZ","White potato, from fresh, mashed, made with milk, and sour cream and/or cream cheese"
+71501020,"WHITE POTATO, FRESH, MASHED, MADE W/ MILK & FAT","White potato, from fresh, mashed, made with milk and fat"
+71501025,"WHITE POTATO, FRESH, MASHED, MADE W/ MILK/SOUR CRM/CHEZ &FAT","White potato, from fresh, mashed, made with milk, and sour cream and/or cream cheese and fat"
+71501030,"WHITE POTATO, FRESH, MASHED, MADE W/ FAT","White potato, from fresh, mashed, made with fat"
+71501040,"WHITE POTATO, DRY, MASHED, MADE W/ MILK & FAT","White potato, from dry, mashed, made with milk and fat"
+71501050,"WHITE POTATO, FRESH, MASHED, MADE W/ MILK, FAT & CHEESE","White potato, from fresh, mashed, made with milk, fat and cheese"
+71501055,"WHITE POTATO, FRESH, MASHED, MADE W/ SOUR CRM/CHEZ & FAT","White potato, from fresh, mashed, made with sour cream and/or cream cheese and fat"
+71501060,"WHITE POTATO, DRY, MASHED, MADE W/ MILK, FAT & EGG","White potato, from dry, mashed, made with milk, fat and egg"
+71501070,"WHITE POTATO, DRY, MASHED, MADE W/ MILK, FAT, EGG & CHEESE","White potato, from dry, mashed, made with milk, fat, egg and cheese"
+71501080,"WHITE POTATO, FRESH, MASHED, NOT MADE W/ MILK OR FAT","White potato, from fresh, mashed, not made with milk or fat"
+71501090,"WHITE POTATO, DRY, MASHED, MADE W/ MILK, NO FAT","White potato, from dry, mashed, made with milk, no fat"
+71501200,"WHITE POTATO, COMPLETE DRY MIX, MASHED, MADE W/ WATER","White potato, from complete dry mix, mashed, made with water"
+71501300,"WHITE POTATO, DRY, MASHED, NS AS TO MILK OR FAT","White potato, from dry, mashed, NS as to milk or fat"
+71501310,"WHITE POTATO, FRESH, MASHED, NS AS TO MILK OR FAT","White potato, from fresh, mashed, NS as to milk or fat"
+71503010,"WHITE POTATO, PATTY (INCLUDE POTATO CROQUETTES)","White potato, patty"
+71505000,"WHITE POTATO, PUFFS","White potato, puffs"
+71507000,"WHITE POTATO, BAKED, STUFFED, PEEL NOT EATEN, NS TOPPING","White potato, stuffed, baked, peel not eaten, NS as to topping"
+71507005,"WHITE POTATO,BAKED,STUFF W/ BUTTER/MARG, NO PEEL","White potato, stuffed, baked, peel not eaten, stuffed with butter or margarine"
+71507010,"WHITE POTATO, BAKED, STUFFED W/SOUR CREAM, NO PEEL","White potato, stuffed, baked, peel not eaten, stuffed with sour cream"
+71507020,"WHITE POT,BAKED, STUFFED W/ CHEESE, PEEL NOT EATEN","White potato, stuffed, baked, peel not eaten, stuffed with cheese"
+71507030,"WHITE POT, BAKED, STUFFED W/ CHILI, PEEL NOT EATEN","White potato, stuffed, baked, peel not eaten, stuffed with chili"
+71507040,"WHITE POT, BAKED, STUFFED W/BROC&CHEESE SCE,NO PEEL","White potato, stuffed, baked, peel not eaten, stuffed with broccoli and cheese sauce"
+71507050,"WHITE POT, BAKED, STUFFD W/ MEAT IN CRM SC, NO PEEL","White potato, stuffed, baked, peel not eaten, stuffed with meat in cream sauce"
+71507100,"WHITE POT,BAKD,STUF W/CHIC,BROC,CHEESE,PEEL NOT EAT","White potato, stuffed, baked, peel not eaten, stuffed with chicken, broccoli and cheese sauce"
+71508000,"WHITE POTATO, BAKED, STUFFED, PEEL EATEN","White potato, stuffed, baked, peel eaten, NS as to topping"
+71508005,"WHITE POTATO,BAKED,STUFF W/ BUTTER/MARG, PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with butter or margarine"
+71508010,"WHITE POTATO, BAKED, STUFFED W/SOUR CRM, PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with sour cream"
+71508020,"WHITE POTATO, BAKED, STUFFED W/ CHEESE, PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with cheese"
+71508030,"WHITE POTATO, BAKED, STUFFED W/ CHILI, PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with chili"
+71508040,"WHITE POT, BKD, STUFD W/BROC&CHEESE SCE,PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with broccoli and cheese sauce"
+71508050,"WHITE POT,BAKED, STUFFED W/MEAT&CRM SCE,PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with meat in cream sauce"
+71508060,"WHITE POT, BAKED, STUFD W/ BACON&CHEESE, PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with bacon and cheese"
+71508070,"WHITE POT,STUFF,BAKED,NO PEEL,W/ BACON & CHEESE","White potato, stuffed, baked, peel not eaten, stuffed with bacon and cheese"
+71508100,"WHITE POT,BAKED,STUFF W/CHIC,BROC,CHEESE,PEEL EATEN","White potato, stuffed, baked, peel eaten, stuffed with chicken, broccoli and cheese sauce"
+71508120,"WHITE POT,STUFF W/HAM,BROC,&CHEESE SAUCE,BKD,W/PEEL","White potato, stuffed with ham, broccoli and cheese sauce, baked, peel eaten"
+71601010,"POTATO SALAD WITH EGG, W/ MAYO","Potato salad with egg, made with mayonnaise"
+71601015,"POTATO SALAD W/ EGG, MADE W/ LT MAYO","Potato salad with egg, made with light mayonnaise"
+71601020,"POTATO SALAD  W/ EGG, MADE W/ MAYO-TYPE DRSG","Potato salad with egg, made with mayonnaise-type salad dressing"
+71601025,"POTATO SALAD W/ EGG, MADE W/ LT MAYO-TYPE DRSG","Potato salad with egg, made with light mayonnaise-type salad dressing"
+71601030,"POTATO SALAD W/ EGG, MADE W/ CREAMY DRSG","Potato salad with egg, made with creamy dressing"
+71601035,"POTATO SALAD W/EGG, MADE W/ LT CREAMY DRSG","Potato salad with egg, made with light creamy dressing"
+71601040,"POTATO SALAD W/ EGG, MADE W/ ITALIAN DRSG","Potato salad with egg, made with Italian dressing"
+71601045,"POTATO SALAD W/ EGG, MADE W/ LT ITALIAN DRSG","Potato salad with egg, made with light Italian dressing"
+71601050,"POTATO SALAD W/ EGG, MADE W/ ANY TYPE OF FAT FREE DRSG","Potato salad with egg, made with any type of fat free dressing"
+71602010,"POTATO SALAD, GERMAN","Potato salad, German style"
+71603010,"POTATO SALAD, MADE WITH MAYONNAISE","Potato salad, made with mayonnaise"
+71603015,"POTATO SALAD, W/ LT MAYO","Potato salad, made with light mayonnaise"
+71603020,"POTATO SALAD, W/ MAYO-TYPE DRSG","Potato salad, made with mayonnaise-type salad dressing"
+71603025,"POTATO SALAD, W/ LT MAYO-TYPE DRSG","Potato salad, made with light mayonnaise-type salad dressing"
+71603030,"POTATO SALAD, W/ CREAMY DRSG","Potato salad, made with creamy dressing"
+71603035,"POTATO SALAD, W/ LT CREAMY DRSG","Potato salad, made with light creamy dressing"
+71603040,"POTATO SALAD, W/ ITALIAN DRESSING","Potato salad, made with Italian dressing"
+71603045,"POTATO SALAD, W/ LT ITALIAN DRSG","Potato salad, made with light Italian dressing"
+71603050,"POTATO SALAD, W/ ANY TYPE OF FAT FREE DRSG","Potato salad, made with any type of fat free dressing"
+71701000,"POTATO PANCAKE","Potato pancake"
+71701500,"NORWEGIAN LEFSE, POTATO & FLOUR PANCAKE","Norwegian Lefse, potato and flour pancake"
+71702000,"POTATO PUDDING","Potato pudding"
+71703000,"STEWED POTATOES, MEXICAN (PAPAS GUISADAS)","Stewed potatoes, Mexican style (Papas guisadas)"
+71703040,"STEWED POT W/TOM,MEXICAN(PAPAS GUISADAS CON TOMATE)","Stewed potatoes with tomatoes, Mexican style (Papas guisadas con tomate)"
+71704000,"STEWED POTATOES WITH TOMATOES","Stewed potatoes with tomatoes"
+71801000,"POTATO SOUP, NS AS TO MADE W/MILK OR WATER","Potato soup, NS as to made with milk or water"
+71801010,"POTATO SOUP, CREAM OF, W/ MILK","Potato soup, cream of, prepared with milk"
+71801020,"POTATO SOUP, PREPARED W/ WATER","Potato soup, prepared with water"
+71801100,"POTATO & CHEESE SOUP","Potato and cheese soup"
+71803010,"POTATO CHOWDER (INCL CORN CHOWDER)","Potato chowder"
+71851010,"PLANTAIN SOUP, P.R. (SOPA DE PLATANO)","Plantain soup, Puerto Rican style (Sopa de platano)"
+71900100,"PLANTAIN, BOILED, NS AS TO GREEN OR RIPE","Plantain, boiled, NS as to green or ripe"
+71900200,"PLANTAIN, FRIED, NS TO GREEN OR RIPE","Plantain, fried, NS as to green or ripe"
+71901010,"GREEN PLANTAIN, BOILED OR BAKED","Green plantains, boiled"
+71901110,"FRIED GREEN PLANTAIN, P.R.","Fried green plantain, Puerto Rican style"
+71905000,"RIPE PLANTAIN, RAW","Ripe plantain, raw"
+71905010,"RIPE PLANTAIN, BOILED (INCL BAKED RIPE PLANTAIN)","Ripe plantain, boiled"
+71905110,"FRIED RIPE PLANTAIN, P.R. (PLATANO MADURO FRITO)","Fried ripe plantain, Puerto Rican style (Platano maduro frito)"
+71905120,"PLANTAIN, RIPE, ROLLED IN FLOUR, FRIED","Plantain, ripe, rolled in flour, fried"
+71905210,"CANDIED RIPE PLANTAIN, P.R. (PLATANO EN ALMIBAR)","Candied ripe plantain, Puerto Rican style (Platano en almibar)"
+71905410,"PLANTAIN CHIPS","Plantain chips"
+71910110,"GREEN BANANA (COOKED IN SALT WATER)","Green banana, cooked (in salt water)"
+71910210,"GREEN BANANA, FRIED","Green banana, fried"
+71910310,"PICKLED GREEN BANANA, P.R","Pickled green bananas, Puerto Rican style (Guineos verdes en escabeche)"
+71930090,"CASSAVA (YUCA BLANCA), COOKED, NS AS TO ADDED FAT","Cassava (yuca blanca), cooked, NS as to fat added in cooking"
+71930100,"CASSAVA (YUCA BLANCA), COOKED, NO FAT ADDED","Cassava (yuca blanca), cooked, fat not added in cooking"
+71930120,"CASSAVA (YUCA BLANCA), COOKED, FAT ADDED","Cassava (yuca blanca), cooked, fat added in cooking"
+71930200,"CASABE, CASSAVA BREAD","Casabe, cassava bread"
+71931010,"CASSAVA W/ CREOLE SAUCE, P.R. (YUCA AL MAJO)","Cassava with creole sauce, Puerto Rican style (Yuca al mojo)"
+71941110,"SWEET POTATOES, WHITE, P.R., FRIED","Sweet potatoes, white, Puerto Rican, fried"
+71941120,"SWEET POTATOES, WHITE, P.R., BOILED","Sweet potatoes, white, Puerto Rican, boiled"
+71941130,"SWEET POTATOES, WHITE, P.R., ROASTED OR BAKED","Sweet potatoes, white, Puerto Rican, roasted or baked"
+71945010,"YAM, PUERTO RICAN, COOKED (NAME HERVIDO)","Yam, Puerto Rican, cooked (Name hervido)"
+71945020,"YAM BUNS, P.R. (BUNUELOS DE NAME)","Yam buns, Puerto Rican style (Bunuelos de name)"
+71950010,"TANNIER, COOKED (INCLUDE YAUTIA)","Tannier, cooked"
+71961010,"CELERIAC, COOKED (INCLUDE P.R. APIO)","Celeriac, cooked"
+71962010,"DASHEEN, BOILED (INCLUDE MALANGA)","Dasheen, boiled"
+71962020,"DASHEEN, FRIED (INCLUDE MALANGA)","Dasheen, fried"
+71962040,"TARO, BAKED","Taro, baked"
+71970110,"STARCHY VEGETABLES, P.R. STYLE, NFS (VIANDAS HERVIDAS)","Starchy vegetables, Puerto Rican style, NFS (viandas hervidas)"
+71970120,"STARCHY VEGETABLES, P.R., NO PLANTAINS","Starchy vegetables, Puerto Rican style, including tannier, white sweet potato and yam, with green or ripe plantains (viandas hervidas)"
+71970130,"STARCHY VEGETABLES, P.R., NO PLANTAINS","Starchy vegetables, Puerto Rican style, including tannier, white sweet potato and yam, no plantain (viandas hervidas)"
+71970200,"FUFU (AFRICAN)","Fufu (African)"
+71980100,"POI","Poi"
+71980200,"TARO CHIPS","Taro chips"
+72101100,"BEET GREENS, RAW","Beet greens, raw"
+72101200,"BEET GREENS, COOKED, NS AS TO ADDED FAT","Beet greens, cooked, NS as to fat added in cooking"
+72101210,"BEET GREENS, COOKED, FAT NOT ADDED","Beet greens, cooked, fat not added in cooking"
+72101220,"BEET GREENS, COOKED, FAT ADDED","Beet greens, cooked, fat added in cooking"
+72103000,"BROCCOLI RAAB, RAW","Broccoli raab, raw"
+72103010,"BROCCOLI RAAB, COOKED, NS AS TO FAT ADDED","Broccoli raab, cooked, NS as to fat added in cooking"
+72103020,"BROCCOLI RAAB, COOKED, FAT NOT ADDED","Broccoli raab, cooked, fat not added in cooking"
+72103030,"BROCCOLI RAAB, COOKED, FAT ADDED IN COOKING","Broccoli raab, cooked, fat added in cooking"
+72104100,"CHARD, RAW","Chard, raw"
+72104200,"CHARD, COOKED, NS AS TO ADDED FAT","Chard, cooked, NS as to fat added in cooking"
+72104210,"CHARD, COOKED, FAT NOT ADDED","Chard, cooked, fat not added in cooking"
+72104220,"CHARD, COOKED, FAT ADDED","Chard, cooked, fat added in cooking"
+72107100,"COLLARDS, RAW","Collards, raw"
+72107200,"COLLARDS, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Collards, cooked, NS as to form, NS as to fat added in cooking"
+72107201,"COLLARDS,COOKED,FROM FRESH,NS FAT ADDED","Collards, cooked, from fresh, NS as to fat added in cooking"
+72107202,"COLLARDS,COOKED,FROM FROZEN,NS FAT ADDED","Collards, cooked, from frozen, NS as to fat added in cooking"
+72107203,"COLLARDS,COOKED,FROM CANNED,NS FAT ADDED","Collards, cooked, from canned, NS as to fat added in cooking"
+72107210,"COLLARDS, COOKED, NS AS TO FORM, FAT NOT ADDED","Collards, cooked, NS as to form, fat not added in cooking"
+72107211,"COLLARDS,COOKED,FROM FRESH,FAT NOT ADDED","Collards, cooked, from fresh, fat not added in cooking"
+72107212,"COLLARDS,COOKED,FROM FROZEN,FAT NOT ADDED","Collards, cooked, from frozen, fat not added in cooking"
+72107213,"COLLARDS,COOKED,FROM CANNED,FAT NOT ADDED","Collards, cooked, from canned, fat not added in cooking"
+72107220,"COLLARDS, COOKED, NS AS TO FORM, FAT ADDED","Collards, cooked, NS as to form, fat added in cooking"
+72107221,"COLLARDS,COOKED,FROM FRESH,FAT ADDED","Collards, cooked, from fresh, fat added in cooking"
+72107222,"COLLARDS,COOKED,FROM FROZEN,FAT ADDED","Collards, cooked, from frozen, fat added in cooking"
+72107223,"COLLARDS,COOKED,FROM CANNED,FAT ADDED","Collards, cooked, from canned, fat added in cooking"
+72110100,"CRESS, RAW","Cress, raw"
+72110200,"CRESS, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Cress, cooked, NS as to form, NS as to fat added in cooking"
+72110201,"CRESS, COOKED, FROM FRESH, NS FAT ADDED","Cress, cooked, from fresh, NS as to fat added in cooking"
+72110203,"CRESS, COOKED, FROM CANNED, NS FAT ADDED","Cress, cooked, from canned, NS as to fat added in cooking"
+72110210,"CRESS, COOKED, NS AS TO FORM, FAT NOT ADDED","Cress, cooked, NS as to form, fat not added in cooking"
+72110211,"CRESS, COOKED, FROM FRESH, FAT NOT ADDED","Cress, cooked, from fresh, fat not added in cooking"
+72110213,"CRESS, COOKED, FROM CANNED, FAT NOT ADDED","Cress, cooked, from canned, fat not added in cooking"
+72110220,"CRESS, COOKED, NS AS TO FORM, FAT ADDED","Cress, cooked, NS as to form, fat added in cooking"
+72110221,"CRESS, COOKED, FROM FRESH, FAT ADDED","Cress, cooked, from fresh, fat added in cooking"
+72110223,"CRESS, COOKED, FROM CANNED, FAT ADDED","Cress, cooked, from canned, fat added in cooking"
+72113100,"DANDELION GREENS, RAW","Dandelion greens, raw"
+72113200,"DANDELION GREENS, COOKED, NS AS TO ADDED FAT","Dandelion greens, cooked, NS as to fat added in cooking"
+72113210,"DANDELION GREENS, COOKED, FAT NOT ADDED","Dandelion greens, cooked, fat not added in cooking"
+72113220,"DANDELION GREENS, COOKED, FAT ADDED","Dandelion greens, cooked, fat added in cooking"
+72116000,"ENDIVE, CHICORY, ESCAROLE OR ROMAINE LETTUCE, RAW","Endive, chicory, escarole, or romaine lettuce, raw"
+72116150,"CAESAR SALAD (WITH ROMAINE), NO DRESSING","Caesar salad (with romaine), no dressing"
+72116200,"ESCAROLE, COOKED, NS AS TO ADDED FAT","Escarole, cooked, NS as to fat added in cooking"
+72116210,"ESCAROLE, COOKED, FAT NOT ADDED","Escarole, cooked, fat not added in cooking"
+72116220,"ESCAROLE, COOKED, FAT ADDED","Escarole, cooked, fat added in cooking"
+72116230,"ESCAROLE, CREAMED","Escarole, creamed"
+72118200,"GREENS, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Greens, cooked, NS as to form, NS as to fat added in cooking"
+72118201,"GREENS, COOKED, FROM FRESH, NS FAT ADDED","Greens, cooked, from fresh, NS as to fat added in cooking"
+72118202,"GREENS, COOKED, FROM FROZ, NS FAT ADDED","Greens, cooked, from frozen, NS as to fat added in cooking"
+72118203,"GREENS, COOKED, FROM CANNED, NS FAT ADDED","Greens, cooked, from canned, NS as to fat added in cooking"
+72118210,"GREENS, COOKED, NS AS TO FORM, FAT NOT ADDED","Greens, cooked, NS as to form, fat not added in cooking"
+72118211,"GREENS, COOKED, FROM FRESH, FAT NOT ADDED","Greens, cooked, from fresh, fat not added in cooking"
+72118212,"GREENS, COOKED, FROM FROZEN, FAT NOT ADDED","Greens, cooked, from frozen, fat not added in cooking"
+72118213,"GREENS, COOKED, FROM CANNED, FAT NOT ADDED","Greens, cooked, from canned, fat not added in cooking"
+72118220,"GREENS, COOKED, NS AS TO FORM, FAT ADDED","Greens, cooked, NS as to form, fat added in cooking"
+72118221,"GREENS, COOKED, FROM FRESH, FAT ADDED","Greens, cooked, from fresh, fat added in cooking"
+72118222,"GREENS, COOKED, FROM FROZEN, FAT ADDED","Greens, cooked, from frozen, fat added in cooking"
+72118223,"GREENS, COOKED, FROM CANNED, FAT ADDED","Greens, cooked, from canned, fat added in cooking"
+72118300,"CHAMNAMUL (KOREAN LEAF VEGETABLE), COOKED, NS AS TO FAT","Chamnamul (Korean leaf vegetable), cooked, NS as to fat added in cooking"
+72118305,"CHAMNAMUL (KOREAN LEAF VEGETABLE), COOKED, FAT NOT ADDED","Chamnamul (Korean leaf vegetable), cooked, fat not added in cooking"
+72118310,"CHAMNAMUL (KOREAN LEAF VEGETABLE), COOKED, FAT ADDED","Chamnamul (Korean leaf vegetable), cooked, fat added in cooking"
+72119200,"KALE, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Kale, cooked, NS as to form, NS as to fat added in cooking"
+72119201,"KALE,COOKED,FROM FRESH,NS FAT ADDED","Kale, cooked, from fresh, NS as to fat added in cooking"
+72119202,"KALE,COOKED,FROM FROZEN,NS FAT ADDED","Kale, cooked, from frozen, NS as to fat added in cooking"
+72119203,"KALE,COOKED,FROM CANNED,NS FAT ADDED","Kale, cooked, from canned, NS as to fat added in cooking"
+72119210,"KALE, COOKED, NS AS TO FORM, FAT NOT ADDED","Kale, cooked, NS as to form, fat not added in cooking"
+72119211,"KALE,COOKED,FROM FRESH,FAT NOT ADDED","Kale, cooked, from fresh, fat not added in cooking"
+72119212,"KALE,COOKED,FROM FROZEN,FAT NOT ADDED","Kale, cooked, from frozen, fat not added in cooking"
+72119213,"KALE,COOKED,FROM CANNED,FAT NOT ADDED","Kale, cooked, from canned, fat not added in cooking"
+72119220,"KALE, COOKED, NS AS TO FORM, FAT ADDED","Kale, cooked, NS as to form, fat added in cooking"
+72119221,"KALE,COOKED,FROM FRESH,FAT ADDED","Kale, cooked, from fresh, fat added in cooking"
+72119222,"KALE,COOKED,FROM FROZEN,FAT ADDED","Kale, cooked, from frozen, fat added in cooking"
+72119223,"KALE,COOKED,FROM CANNED,FAT ADDED","Kale, cooked, from canned, fat added in cooking"
+72120200,"LAMBSQUARTER, COOKED, NS AS TO ADDED FAT","Lambsquarter, cooked, NS as to fat added in cooking"
+72120210,"CRESS, COOKED, FAT NOT ADDEDS","Lambsquarter, cooked, fat not added in cooking"
+72120220,"LAMBSQUARTER, COOKED, FAT ADDED","Lambsquarter, cooked, fat added in cooking"
+72121210,"MUSTARD CABBAGE,COOKED, FAT NOT ADDED IN COOKING","Mustard cabbage, cooked, fat not added in cooking"
+72122100,"MUSTARD GREENS, RAW","Mustard greens, raw"
+72122200,"MUSTARD GREENS, COOKED, NS FORM, NS FAT ADDED","Mustard greens, cooked, NS as to form, NS as to fat added in cooking"
+72122201,"MUSTARD GREENS,COOKED,FROM FRESH,NS FAT ADDED","Mustard greens, cooked, from fresh, NS as to fat added in cooking"
+72122202,"MUSTARD GREENS,COOKED,FROM FROZEN,NS FAT ADDED","Mustard greens, cooked, from frozen, NS as to fat added in cooking"
+72122203,"MUSTARD GREENS,COOKED,FROM CANNED,NS FAT ADDED","Mustard greens, cooked, from canned, NS as to fat added in cooking"
+72122210,"MUSTARD GREENS, COOKED,  NS FORM, FAT NOT ADDED","Mustard greens, cooked, NS as to form, fat not added in cooking"
+72122211,"MUSTARD GREENS,COOKED,FROM FRESH,FAT NOT ADDED","Mustard greens, cooked, from fresh, fat not added in cooking"
+72122212,"MUSTARD GREENS,COOKED,FROM FROZEN,FAT NOT ADDED","Mustard greens, cooked, from frozen, fat not added in cooking"
+72122213,"MUSTARD GREENS,COOKED,FROM CANNED,FAT NOT ADDED","Mustard greens, cooked, from canned, fat not added in cooking"
+72122220,"MUSTARD GREEN, COOKED, NS FORM, FAT ADDED","Mustard greens, cooked, NS as to form, fat added in cooking"
+72122221,"MUSTARD GREENS,COOKED,FROM FRESH,FAT ADDED","Mustard greens, cooked, from fresh, fat added in cooking"
+72122222,"MUSTARD GREENS,COOKED,FROM FROZEN,FAT ADDED","Mustard greens, cooked, from frozen, fat added in cooking"
+72122223,"MUSTARD GREENS,COOKED,FROM CANNED,FAT ADDED","Mustard greens, cooked, from canned, fat added in cooking"
+72123000,"POKE GREENS, COOKED, NS AS TO ADDED FAT","Poke greens, cooked, NS as to fat added in cooking"
+72123010,"POKE GREENS, COOKED, FAT NOT ADDED","Poke greens, cooked, fat not added in cooking"
+72123020,"POKE GREENS, COOKED, FAT ADDED","Poke greens, cooked, fat added in cooking"
+72124100,"RADICCHIO, RAW","Radicchio, raw"
+72125100,"SPINACH, RAW","Spinach, raw"
+72125200,"SPINACH, COOKED, NS FORM, NS AS TO ADDED FAT","Spinach, cooked, NS as to form, NS as to fat added in cooking"
+72125201,"SPINACH,COOKED,FROM FRESH,NS FAT ADDED","Spinach, cooked, from fresh, NS as to fat added in cooking"
+72125202,"SPINACH,COOKED,FROM FROZEN,NS FAT ADDED","Spinach, cooked, from frozen, NS as to fat added in cooking"
+72125203,"SPINACH,COOKED,FROM CANNED,NS FAT ADDED","Spinach, cooked, from canned, NS as to fat added in cooking"
+72125210,"SPINACH, COOKED, NS AS TO FORM, FAT NOT ADDED","Spinach, cooked, NS as to form, fat not added in cooking"
+72125211,"SPINACH,COOKED,FROM FRESH,FAT NOT ADDED","Spinach, cooked, from fresh, fat not added in cooking"
+72125212,"SPINACH,COOKED,FROM FROZEN,FAT NOT ADDED","Spinach, cooked, from frozen, fat not added in cooking"
+72125213,"SPINACH,COOKED,FROM CANNED,FAT NOT ADDED","Spinach, cooked, from canned, fat not added in cooking"
+72125220,"SPINACH, COOKED, NS AS TO FORM, FAT ADDED","Spinach, cooked, NS as to form, fat added in cooking"
+72125221,"SPINACH,COOKED,FROM FRESH,FAT ADDED","Spinach, cooked, from fresh, fat added in cooking"
+72125222,"SPINACH,COOKED,FROM FROZEN,FAT ADDED","Spinach, cooked, from frozen, fat added in cooking"
+72125223,"SPINACH,COOKED,FROM CANNED,FAT ADDED","Spinach, cooked, from canned, fat added in cooking"
+72125230,"SPINACH, NS AS TO FORM, CREAMED","Spinach, NS as to form, creamed"
+72125231,"SPINACH, FROM FRESH, CREAMED","Spinach, from fresh, creamed"
+72125232,"SPINACH, FROM FROZEN, CREAMED","Spinach, from frozen, creamed"
+72125233,"SPINACH, FROM CANNED, CREAMED","Spinach, from canned, creamed"
+72125240,"SPINACH SOUFFLE","Spinach souffle"
+72125250,"SPINACH, COOKED, NS AS TO FORM, W/ CHEESE SAUCE","Spinach, cooked, NS as to form, with cheese sauce"
+72125251,"SPINACH, COOKED, FROM FRESH, W/ CHEESE SAUCE","Spinach, cooked, from fresh, with cheese sauce"
+72125252,"SPINACH, COOKED, FROM FROZEN, W/ CHEESE SAUCE","Spinach, cooked, from frozen, with cheese sauce"
+72125253,"SPINACH, COOKED, FROM CANNED, W/ CHEESE SAUCE","Spinach, cooked, from canned, with cheese sauce"
+72125260,"SPINACH & CHEESE CASSEROLE","Spinach and cheese casserole"
+72125310,"SPINACH AND COTTAGE CHEESE","Palak Paneer or Saag Paneer (Indian)"
+72125500,"SPINACH & CHICK PEAS, FAT ADDED","Spinach and chickpeas, fat added"
+72126000,"TARO LEAVES, COOKED, FAT NOT ADDED IN COOKING","Taro leaves, cooked, fat not added in cooking"
+72127000,"THISTLE LEAVES, COOKED, FAT NOT ADDED IN COOKING","Thistle leaves, cooked, fat not added in cooking"
+72128200,"TURNIP GREENS, COOKED, NS FORM, NS AS TO ADDED FAT","Turnip greens, cooked, NS as to form, NS as to fat added in cooking"
+72128201,"TURNIP GREENS,COOKED,FROM FRESH,NS FAT ADDED","Turnip greens, cooked, from fresh, NS as to fat added in cooking"
+72128202,"TURNIP GREENS,COOKED,FROM FROZEN,NS FAT ADDED","Turnip greens, cooked, from frozen, NS as to fat added in cooking"
+72128203,"TURNIP GREENS,COOKED,FROM CANNED,NS FAT ADDED","Turnip greens, cooked, from canned, NS as to fat added in cooking"
+72128210,"TURNIP GREENS, NS FORM, COOKED, FAT NOT ADDED","Turnip greens, cooked, NS as to form, fat not added in cooking"
+72128211,"TURNIP GREENS,COOKED,FROM FRESH,FAT NOT ADDED","Turnip greens, cooked, from fresh, fat not added in cooking"
+72128212,"TURNIP GREENS,COOKED,FROM FROZEN,FAT NOT ADDED","Turnip greens, cooked, from frozen, fat not added in cooking"
+72128213,"TURNIP GREENS,COOKED,FROM CANNED,FAT NOT ADDED","Turnip greens, cooked, from canned, fat not added in cooking"
+72128220,"TURNIP GREENS, COOKED, NS FORM, FAT ADDED","Turnip greens, cooked, NS as to form, fat added in cooking"
+72128221,"TURNIP GREENS,COOKED,FROM FRESH,FAT ADDED","Turnip greens, cooked, from fresh, fat added in cooking"
+72128222,"TURNIP GREENS,COOKED,FROM FROZEN,FAT ADDED","Turnip greens, cooked, from frozen, fat added in cooking"
+72128223,"TURNIP GREENS,COOKED,FROM CANNED,FAT ADDED","Turnip greens, cooked, from canned, fat added in cooking"
+72128400,"TURNIP GREENS W/ ROOTS, CKD, NS FORM, NS ADDED FAT","Turnip greens with roots, cooked, NS as to form, NS as to fat added in cooking"
+72128401,"TURNIP GREENS W/ ROOTS, CKD, FROM FRESH, NS ADDED FAT","Turnip greens with roots, cooked, from fresh, NS as to fat added in cooking"
+72128402,"TURNIP GREENS W/ ROOTS, CKD, FROM FROZ, NS ADDED FAT","Turnip greens with roots, cooked, from frozen, NS as to fat added in cooking"
+72128403,"TURNIP GREENS W/ ROOTS, CKD, FROM CAN, NS ADDED FAT","Turnip greens with roots, cooked, from canned, NS as to fat added in cooking"
+72128410,"TURNIP GREENS W/ ROOTS, CKD, NS FORM, FAT NOT ADDED","Turnip greens with roots, cooked, NS as to form, fat not added in cooking"
+72128411,"TURNIP GREENS W/ ROOTS, CKD, FROM FRESH, NO FAT ADDED","Turnip greens with roots, cooked, from fresh, fat not added in cooking"
+72128412,"TURNIP GREENS W/ ROOTS, CKD, FROM FROZ, NO FAT ADDED","Turnip greens with roots, cooked, from frozen, fat not added in cooking"
+72128413,"TURNIP GREENS W/ ROOTS, CKD, FROM CAN, NO FAT ADDED","Turnip greens with roots, cooked, from canned, fat not added in cooking"
+72128420,"TURNIP GREENS W/ ROOTS, COOKED, NS FORM, FAT ADDED","Turnip greens with roots, cooked, NS as to form, fat added in cooking"
+72128421,"TURNIP GREENS W/ ROOTS, CKD, FROM FRESH, FAT ADDED","Turnip greens with roots, cooked, from fresh, fat added in cooking"
+72128422,"TURNIP GREENS W/ ROOTS, CKD, FROM FROZ, FAT ADDED","Turnip greens with roots, cooked, from frozen, fat added in cooking"
+72128423,"TURNIP GREENS W/ ROOTS, CKD, FROM CAN, FAT ADDED","Turnip greens with roots, cooked, from canned, fat added in cooking"
+72128500,"TURNIP GREENS, CANNED, LOW NA, NS AS TO ADDED FAT","Turnip greens, canned, low sodium, cooked, NS as to fat added in cooking"
+72128510,"TURNIP GREENS, CANNED, LOW SODIUM, FAT NOT ADDED","Turnip greens, canned, low sodium, cooked, fat not added in cooking"
+72128520,"TURNIP GREENS, CANNED, LOW SODIUM, FAT ADDED","Turnip greens, canned, low sodium, cooked, fat added in cooking"
+72130100,"WATERCRESS, RAW","Watercress, raw"
+72130200,"WATERCRESS, COOKED, FAT NOT ADDED IN COOKING","Watercress, cooked, fat not added in cooking"
+72132200,"BITTERMELON,HORSERADISH,JUTE,RADISH LVES,CKD,NO FAT","Bitter melon leaves, horseradish leaves, jute leaves, or radish leaves, cooked, fat not added in cooking"
+72133200,"SWEET POTATO,SQUASH,PUMPKIN LEAVES,CKD,FAT NOT ADDED","Sweet potato leaves, squash leaves, pumpkin leaves, chrysanthemum leaves, bean leaves, or swamp cabbage, cooked, fat not added in cooking"
+72201100,"BROCCOLI, RAW","Broccoli, raw"
+72201200,"BROCCOLI, CKD, NS FORM, NS FAT (INCL BROCCOLI, NFS)","Broccoli, cooked, NS as to form, NS as to fat added in cooking"
+72201201,"BROCCOLI, CKD, FROM FRESH, NS FAT (INCL BROCCOLI, NFS)","Broccoli, cooked, from fresh, NS as to fat added in cooking"
+72201202,"BROCCOLI, CKD, FROM FROZ, NS FAT (INCL BROCCOLI, NFS)","Broccoli, cooked, from frozen, NS as to fat added in cooking"
+72201210,"BROCCOLI, COOKED, NS AS TO FORM, NO FAT ADDED","Broccoli, cooked, NS as to form, fat not added in cooking"
+72201211,"BROCCOLI, COOKED, FROM FRESH, NO FAT ADDED","Broccoli, cooked, from fresh, fat not added in cooking"
+72201212,"BROCCOLI, COOKED, FROM FROZ, NO FAT ADDED","Broccoli, cooked, from frozen, fat not added in cooking"
+72201220,"BROCCOLI, COOKED, NS AS TO FORM, FAT ADDED","Broccoli, cooked, NS as to form, fat added in cooking"
+72201221,"BROCCOLI, COOKED, FROM FRESH, FAT ADDED","Broccoli, cooked, from fresh, fat added in cooking"
+72201222,"BROCCOLI, COOKED, FROM FROZ, FAT ADDED","Broccoli, cooked, from frozen, fat added in cooking"
+72201230,"BROCCOLI, COOKED, NS AS TO FORM, W/ CHEESE SAUCE","Broccoli, cooked, NS as to form, with cheese sauce"
+72201231,"BROCCOLI, COOKED, FROM FRESH, W/ CHEESE SAUCE","Broccoli, cooked, from fresh, with cheese sauce"
+72201232,"BROCCOLI, COOKED, FROM FROZEN, W/ CHEESE SAUCE","Broccoli, cooked, from frozen, with cheese sauce"
+72201240,"BROCCOLI, COOKED, NS AS TO FORM, W/ MUSHROOM SAUCE","Broccoli, cooked, NS as to form, with mushroom sauce"
+72201241,"BROCCOLI, COOKED, FROM FRESH, W/ MUSHROOM SAUCE","Broccoli, cooked, from fresh, with mushroom sauce"
+72201242,"BROCCOLI, COOKED, FROM FROZEN, W/ MUSHROOM SAUCE","Broccoli, cooked, from frozen, with mushroom sauce"
+72201250,"BROCCOLI, COOKED, NS AS TO FORM, W/ CREAM SAUCE","Broccoli, cooked, NS as to form, with cream sauce"
+72201251,"BROCCOLI, COOKED, FROM FRESH, W/ CREAM SAUCE","Broccoli, cooked, from fresh, with cream sauce"
+72201252,"BROCCOLI, COOKED, FROM FROZEN, W/ CREAM SAUCE","Broccoli, cooked, from frozen, with cream sauce"
+72202010,"BROCCOLI CASSEROLE (BROC, NOODLES, CREAM SAUCE)","Broccoli casserole (broccoli, noodles, and cream sauce)"
+72202020,"BROCCOLI CASSEROLE (BROC,RICE,CHEESE,MUSHROOM SCE)","Broccoli casserole (broccoli, rice, cheese, and mushroom sauce)"
+72202030,"BROCCOLI, BATTER-DIPPED & FRIED","Broccoli, batter-dipped and fried"
+72302000,"BROCCOLI SOUP, PREPARED WITH MILK, HOME RECIPE, CANNED OR RE","Broccoli soup, prepared with milk, home recipe, canned or ready-to-serve"
+72302020,"BROCCOLI SOUP, PREP W/ WATER","Broccoli soup, prepared with water, home recipe, canned, or ready-to-serve"
+72302100,"BROCCOLI CHEESE SOUP, PREPARED WITH MILK, HOME RECIPE, CANNE","Broccoli cheese soup, prepared with milk, home recipe, canned, or ready-to-serve"
+72306000,"WATERCRESS BROTH W/ SHRIMP","Watercress broth with shrimp"
+72307000,"SPINACH SOUP","Spinach soup"
+72308000,"DARK-GREEN LEAFY VEGETABLE SOUP WITH MEAT, ASIAN STYLE","Dark-green leafy vegetable soup with meat, Asian style"
+72308500,"DARK-GREEN LEAFY VEGETABLE SOUP, MEATLESS, ASIAN STYLE","Dark-green leafy vegetable soup, meatless, Asian style"
+73101010,"CARROTS, RAW","Carrots, raw"
+73101110,"CARROTS, RAW, SALAD (INCLUDE CARROT-RAISIN SALAD)","Carrots, raw, salad"
+73101210,"CARROTS, RAW, SALAD W/ APPLES","Carrots, raw, salad with apples"
+73102200,"CARROTS, COOKED, NS AS TO FORM, NS FAT ADDED","Carrots, cooked, NS as to form, NS as to fat added in cooking"
+73102201,"CARROTS, COOKED, FROM FRESH, NS FAT ADDED","Carrots, cooked, from fresh, NS as to fat added in cooking"
+73102202,"CARROTS, COOKED, FROM FROZEN, NS FAT ADDED","Carrots, cooked, from frozen, NS as to fat added in cooking"
+73102203,"CARROTS, COOKED, FROM CANNED, NS FAT ADDED","Carrots, cooked, from canned, NS as to fat added in cooking"
+73102210,"CARROTS, COOKED, NS AS TO FORM, FAT NOT ADDED","Carrots, cooked, NS as to form, fat not added in cooking"
+73102211,"CARROTS, COOKED, FROM FRESH, FAT NOT ADDED","Carrots, cooked, from fresh, fat not added in cooking"
+73102212,"CARROTS, COOKED, FROM FROZEN, FAT NOT ADDED","Carrots, cooked, from frozen, fat not added in cooking"
+73102213,"CARROTS, COOKED, FROM CANNED, FAT NOT ADDED","Carrots, cooked, from canned, fat not added in cooking"
+73102220,"CARROTS, COOKED, NS AS TO FORM, FAT ADDED","Carrots, cooked, NS as to form, fat added in cooking"
+73102221,"CARROTS, COOKED, FROM FRESH, FAT ADDED","Carrots, cooked, from fresh, fat added in cooking"
+73102222,"CARROTS, COOKED, FROM FROZEN, FAT ADDED","Carrots, cooked, from frozen, fat added in cooking"
+73102223,"CARROTS, COOKED, FROM CANNED, FAT ADDED","Carrots, cooked, from canned, fat added in cooking"
+73102230,"CARROTS, COOKED, NS AS TO FORM, CREAMED","Carrots, cooked, NS as to form, creamed"
+73102231,"CARROTS, COOKED, FROM FRESH, CREAMED","Carrots, cooked, from fresh, creamed"
+73102232,"CARROTS, COOKED, FROM FROZEN, CREAMED","Carrots, cooked, from frozen, creamed"
+73102233,"CARROTS, COOKED, FROM CANNED, CREAMED","Carrots, cooked, from canned, creamed"
+73102240,"CARROTS, COOKED, NS AS TO FORM, GLAZED","Carrots, cooked, NS as to form, glazed"
+73102241,"CARROTS, COOKED, FROM FRESH, GLAZED","Carrots, cooked, from fresh, glazed"
+73102242,"CARROTS, COOKED, FROM FROZEN, GLAZED","Carrots, cooked, from frozen, glazed"
+73102243,"CARROTS, COOKED, FROM CANNED, GLAZED","Carrots, cooked, from canned, glazed"
+73102250,"CARROTS, COOKED, NS AS TO FORM, W/ CHEESE SAUCE","Carrots, cooked, NS as to form, with cheese sauce"
+73102251,"CARROTS, COOKED, FROM FRESH, W/ CHEESE SAUCE","Carrots, cooked, from fresh, with cheese sauce"
+73102252,"CARROTS, COOKED, FROM FROZEN, W/ CHEESE SAUCE","Carrots, cooked, from frozen, with cheese sauce"
+73102253,"CARROTS, COOKED, FROM CANNED, W/ CHEESE SAUCE","Carrots, cooked, from canned, with cheese sauce"
+73103000,"CARROTS, CANNED, LOW SODIUM, NS AS TO ADDED FAT","Carrots, canned, low sodium, NS as to fat added in cooking"
+73103010,"CARROTS, CANNED, LOW SODIUM, NO FAT ADDED","Carrots, canned, low sodium, fat not added in cooking"
+73103020,"CARROTS, CANNED, LOW SODIUM, FAT ADDED","Carrots, canned, low sodium, fat added in cooking"
+73105010,"CARROT JUICE","Carrot juice"
+73111030,"PEAS & CARROTS, NS AS TO FORM, CREAMED","Peas and carrots, NS as to form, creamed"
+73111031,"PEAS & CARROTS, FROM FRESH, CREAMED","Peas and carrots, from fresh, creamed"
+73111032,"PEAS & CARROTS, FROM FROZEN, CREAMED","Peas and carrots, from frozen, creamed"
+73111033,"PEAS & CARROTS, FROM CANNED, CREAMED","Peas and carrots, from canned, creamed"
+73111200,"PEAS & CARROTS, COOKED, NS FORM, NS AS TO ADDED FAT","Peas and carrots, cooked, NS as to form, NS as to fat added in cooking"
+73111201,"PEAS & CARROTS, COOKED, FROM FRESH, NS FAT ADDED","Peas and carrots, cooked, from fresh, NS as to fat added in cooking"
+73111202,"PEAS & CARROTS, COOKED, FROM FROZ, NS FAT ADDED","Peas and carrots, cooked, from frozen, NS as to fat added in cooking"
+73111203,"PEAS & CARROTS, COOKED, FROM CANNED, NS FAT ADDED","Peas and carrots, cooked, from canned, NS as to fat added in cooking"
+73111210,"PEAS & CARROTS, COOKED, NS FORM, FAT NOT ADDED","Peas and carrots, cooked, NS as to form, fat not added in cooking"
+73111211,"PEAS & CARROTS, COOKED, FROM FRESH, FAT NOT ADDED","Peas and carrots, cooked, from fresh, fat not added in cooking"
+73111212,"PEAS & CARROTS, COOKED, FROM FROZ, FAT NOT ADDED","Peas and carrots, cooked, from frozen, fat not added in cooking"
+73111213,"PEAS & CARROTS, COOKED, FROM CANNED, FAT NOT ADDED","Peas and carrots, cooked, from canned, fat not added in cooking"
+73111220,"PEAS & CARROTS, COOKED, NS AS TO FORM, FAT ADDED","Peas and carrots, cooked, NS as to form, fat added in cooking"
+73111221,"PEAS & CARROTS, COOKED, FROM FRESH, FAT ADDED","Peas and carrots, cooked, from fresh, fat added in cooking"
+73111222,"PEAS & CARROTS, COOKED, FROM FROZ, FAT ADDED","Peas and carrots, cooked, from frozen, fat added in cooking"
+73111223,"PEAS & CARROTS, COOKED, FROM CANNED, FAT ADDED","Peas and carrots, cooked, from canned, fat added in cooking"
+73111250,"PEAS & CARROTS, CANNED, LOW SODIUM, NS ADDED FAT","Peas and carrots, canned, low sodium, NS as to fat added in cooking"
+73111260,"PEAS & CARROTS, CANNED, LOW SODIUM, FAT ADDED","Peas and carrots, canned, low sodium, fat added in cooking"
+73111270,"PEAS & CARROTS, CANNED, LOW SODIUM, NO FAT ADDED","Peas and carrots, canned, low sodium, fat not added in cooking"
+73111400,"CARROTS IN TOMATO SAUCE","Carrots in tomato sauce"
+73112000,"CARROT CHIPS, DRIED","Carrot chips, dried"
+73201000,"PUMPKIN, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Pumpkin, cooked, NS as to form, NS as to fat added in cooking"
+73201001,"PUMPKIN, COOKED, FROM FRESH, NS AS TO ADDED FAT","Pumpkin, cooked, from fresh, NS as to fat added in cooking"
+73201002,"PUMPKIN, COOKED, FROM FROZEN, NS AS TO ADDED FAT","Pumpkin, cooked, from frozen, NS as to fat added in cooking"
+73201003,"PUMPKIN, COOKED, FROM CANNED, NS AS TO ADDED FAT","Pumpkin, cooked, from canned, NS as to fat added in cooking"
+73201010,"PUMPKIN, COOKED, NS AS TO FORM, FAT NOT ADDED","Pumpkin, cooked, NS as to form, fat not added in cooking"
+73201011,"PUMPKIN, COOKED, FROM FRESH, FAT NOT ADDED","Pumpkin, cooked, from fresh, fat not added in cooking"
+73201012,"PUMPKIN, COOKED, FROM FROZEN, FAT NOT ADDED","Pumpkin, cooked, from frozen, fat not added in cooking"
+73201013,"PUMPKIN, COOKED, FROM CANNED, FAT NOT ADDED","Pumpkin, cooked, from canned, fat not added in cooking"
+73201020,"PUMPKIN, COOKED, NS AS TO FORM, FAT ADDED","Pumpkin, cooked, NS as to form, fat added in cooking"
+73201021,"PUMPKIN, COOKED, FROM FRESH, FAT ADDED","Pumpkin, cooked, from fresh, fat added in cooking"
+73201022,"PUMPKIN, COOKED, FROM FROZEN, FAT ADDED","Pumpkin, cooked, from frozen, fat added in cooking"
+73201023,"PUMPKIN, COOKED, FROM CANNED, FAT ADDED","Pumpkin, cooked, from canned, fat added in cooking"
+73210010,"CALABAZA (SPANISH PUMPKIN), COOKED","Calabaza (Spanish pumpkin), cooked"
+73210110,"PUMPKIN FRITTERS, P.R.","Pumpkin fritters, Puerto Rican style"
+73211110,"SWEET POTATO & PUMPKIN CASSEROLE, P.R","Sweet potato and pumpkin casserole, Puerto Rican style"
+73301000,"SQUASH, WINTER, MASHED, NS AS TO ADDED FAT/SUGAR","Squash, winter type, mashed, NS as to fat or sugar added in cooking"
+73301010,"SQUASH, WINTER, MASHED, NO FAT OR SUGAR ADDED","Squash, winter type, mashed, no fat or sugar added in cooking"
+73301020,"SQUASH, WINTER, COOKED, MASHED, FAT ADDED, NO SUGAR","Squash, winter type, mashed, fat added in cooking, no sugar added in cooking"
+73301030,"SQUASH, WINTER, COOKED, MASHED, FAT & SUGAR ADDED","Squash, winter type, mashed, fat and sugar added in cooking"
+73302010,"SQUASH, WINTER, RAW","Squash, winter type, raw"
+73303000,"SQUASH, WINTER, BAKED, NS FAT OR SUGAR ADDED","Squash, winter type, baked, NS as to fat or sugar added in cooking"
+73303010,"SQUASH, WINTER, BAKED, NO FAT OR SUGAR ADDED","Squash, winter type, baked, no fat or sugar added in cooking"
+73303020,"SQUASH, WINTER, BAKED, FAT ADDED, NO SUGAR","Squash, winter type, baked, fat added in cooking, no sugar added in cooking"
+73303030,"SQUASH, WINTER, BAKED, FAT & SUGAR ADDED","Squash, winter type, baked, fat and sugar added in cooking"
+73303040,"SQUASH, WINTER, BAKED, NO ADDED FAT, SUGAR ADDED","Squash, winter type, baked, no fat added in cooking, sugar added in cooking"
+73304010,"SQUASH, FRITTER OR CAKE","Squash fritter or cake"
+73305010,"SQUASH, WINTER, BAKED W/ CHEESE","Squash, winter, baked with cheese"
+73305020,"SQUASH, WINTER, SOUFFLE","Squash, winter, souffle"
+73401000,"SWEET POTATO, NFS","Sweet potato, NFS"
+73402000,"SWEET POTATO, BAKED, PEEL EATEN, NS AS TO ADDED FAT","Sweet potato, baked, peel eaten, NS as to fat added in cooking"
+73402010,"SWEET POTATO, BAKED, PEEL EATEN, NO FAT ADDED","Sweet potato, baked, peel eaten, fat not added in cooking"
+73402020,"SWEET POTATO, BAKED, PEEL EATEN, FAT ADDED","Sweet potato, baked, peel eaten, fat added in cooking"
+73403000,"SWEET POTATO, BAKED, PEEL NOT EATEN, NS AS TO FAT","Sweet potato, baked, peel not eaten, NS as to fat added in cooking"
+73403010,"SWEET POTATO, BAKED, PEEL NOT EATEN, FAT NOT ADDED","Sweet potato, baked, peel not eaten, fat not added in cooking"
+73403020,"SWEET POTATO, BAKED, PEEL NOT EATEN, FAT ADDED","Sweet potato, baked, peel not eaten, fat added in cooking"
+73405000,"SWEET POTATO, BOILED, W/O PEEL, NS AS TO ADDED FAT","Sweet potato, boiled, without peel, NS as to fat added in cooking"
+73405010,"SWEET POTATO, BOILED, W/O PEEL, FAT NOT ADDED","Sweet potato, boiled, without peel, fat not added in cooking"
+73405020,"SWEET POTATO, BOILED, W/O PEEL, FAT ADDED","Sweet potato, boiled, without peel, fat added in cooking"
+73405100,"SWEET POTATO, BOILED W/ PEEL, PEEL NOT EATEN, NS AS TO FAT","Sweet potato, boiled with peel, peel not eaten, NS as to fat added in cooking"
+73405110,"SWEET POTATO, BOILED W/ PEEL, PEEL NOT EATEN, FAT NOT ADDED","Sweet potato, boiled with peel, peel not eaten, fat not added in cooking"
+73405120,"SWEET POTATO, BOILED W/ PEEL, PEEL NOT EATEN, FAT ADDED","Sweet potato, boiled with peel, peel not eaten, fat added in cooking"
+73406000,"SWEET POTATO, CANDIED","Sweet potato, candied"
+73406010,"SWEET POTATO W/ FRUIT","Sweet potato with fruit"
+73407000,"SWEET POTATO, CANNED, NS AS TO SYRUP","Sweet potato, canned, NS as to syrup"
+73407010,"SWEET POTATO, CANNED, W/O SYRUP","Sweet potato, canned without syrup"
+73407020,"SWEET POTATO, CANNED IN SYRUP","Sweet potato, canned in syrup"
+73407030,"SWEET POTATO, CANNED IN SYRUP, W/ FAT ADDED","Sweet potato, canned in syrup, with fat added in cooking"
+73409000,"SWEET POTATO, CASSEROLE OR MASHED","Sweet potato, casserole or mashed"
+73410110,"SWEET POTATO, FRIED","Sweet potato, fried"
+73410210,"SWEET POTATO, CHIPS","Sweet potato, chips"
+73410300,"SWEET POTATO, FRENCH FRIES","Sweet potato, french fries"
+73421000,"SWEET POTATO, YELLOW, P.R., COOKED","Sweet potato, yellow, Puerto Rican, cooked"
+73501000,"CARROT SOUP, CREAM OF, PREPARED WITH MILK, HOME RECIPE, CANN","Carrot soup, cream of, prepared with milk, home recipe, canned or ready-to-serve"
+73501010,"CARROT WITH RICE SOUP, CREAM OF, PREPARED WITH MILK, HOME RE","Carrot with rice soup, cream of, prepared with milk, home recipe, canned or ready-to-serve"
+74101000,"TOMATOES, RAW","Tomatoes, raw"
+74102000,"TOMATOES, GREEN, RAW","Tomatoes, green, raw"
+74201000,"TOMATOES, COOKED, NS AS TO FORM, NS AS TO METHOD","Tomatoes, cooked, NS as to form, NS as to method"
+74201001,"TOMATOES, COOKED, FROM FRESH, NS AS TO METHOD","Tomatoes, cooked, from fresh, NS as to method"
+74201003,"TOMATOES, COOKED, FROM CANNED, NS AS TO METHOD","Tomatoes, cooked, from canned, NS as to method"
+74202010,"TOMATOES, NS AS TO FORM, BROILED","Tomatoes, NS as to form, broiled"
+74202011,"TOMATOES, FROM FRESH, BROILED","Tomatoes, from fresh, broiled"
+74202050,"TOMATOES, RED, NS AS TO FORM, FRIED","Tomatoes, red, NS as to form, fried"
+74202051,"TOMATOES, RED, FROM FRESH, FRIED","Tomatoes, red, from fresh, fried"
+74203010,"TOMATOES, NS AS TO FORM, SCALLOPED","Tomatoes, NS as to form, scalloped"
+74203011,"TOMATOES, FROM FRESH, SCALLOPED","Tomatoes, from fresh, scalloped"
+74204010,"TOMATOES, NS AS TO FORM, STEWED","Tomatoes, NS as to form, stewed"
+74204011,"TOMATOES, FROM FRESH, STEWED","Tomatoes, from fresh, stewed"
+74204013,"TOMATOES, FROM CANNED, STEWED","Tomatoes, from canned, stewed"
+74204500,"TOMATOES, CANNED, LOW SODIUM","Tomatoes, canned, low sodium"
+74205010,"TOMATOES, GREEN, COOKED, NS AS TO FORM (INCL FRIED)","Tomatoes, green, cooked, NS as to form"
+74205011,"TOMATOES, GREEN, COOKED, FROM FRESH (INCL FRIED)","Tomatoes, green, cooked, from fresh"
+74205020,"TOMATOES, GREEN, PICKLED","Tomato, green, pickled"
+74206000,"TOMATOES, RED, DRIED","Tomatoes, red, dried"
+74301100,"TOMATO JUICE","Tomato juice"
+74301150,"TOMATO JUICE, LOW SODIUM","Tomato juice, low sodium"
+74302000,"TOMATO JUICE COCKTAIL","Tomato juice cocktail"
+74303000,"TOMATO & VEGETABLE JUICE, MOSTLY TOMATO (INCL V-8)","Tomato and vegetable juice, mostly tomato"
+74303100,"TOMATO & VEGETABLE JUICE, MOSTLY TOMATO, LOW SODIUM","Tomato and vegetable juice, mostly tomato, low sodium"
+74304000,"TOMATO JUICE W/ CLAM OR BEEF JUICE","Tomato juice with clam or beef juice"
+74401010,"TOMATO CATSUP","Tomato catsup"
+74401110,"TOMATO CATSUP, REDUCED SODIUM","Tomato catsup, reduced sodium"
+74402010,"TOMATO CHILI SAUCE (CATSUP TYPE)","Tomato chili sauce (catsup-type)"
+74402100,"SALSA, NFS","Salsa, NFS"
+74402110,"SALSA, PICO DE GALLO","Salsa, pico de gallo"
+74402150,"SALSA, RED, COMMERCIALLY-PREPARED","Salsa, red, commercially-prepared"
+74402200,"SALSA, RED, HOMEMADE","Salsa, red, homemade"
+74402250,"ENCHILADA SAUCE, RED","Enchilada sauce, red"
+74402260,"ENCHILADA SAUCE, GREEN","Enchilada sauce, green"
+74402350,"SALSA VERDE OR SALSA, GREEN","Salsa verde or salsa, green"
+74404010,"SPAGHETTI SAUCE, MEATLESS","Spaghetti sauce, meatless"
+74404020,"SPAGHETTI SAUCE W/ VEGETABLES, HOMEMADE-STYLE","Spaghetti sauce with vegetables, homemade-style"
+74404030,"SPAGHETTI SAUCE W/ MEAT, CANNED, NO EXTRA MEAT","Spaghetti sauce with meat, canned, no extra meat added"
+74404050,"SPAGHETTI SAUCE, MEATLESS, REDUCED SODIUM","Spaghetti sauce, meatless, reduced sodium"
+74404060,"SPAGHETTI SAUCE, MEATLESS, FAT FREE","Spaghetti sauce, meatless, fat free"
+74404090,"VODKA FLAVORED PASTA SAUCE MADE WITH TOMATOES AND CREAM","Vodka flavored pasta sauce made with tomatoes and cream"
+74405010,"TOMATO RELISH (INCLUDE TOMATO PRESERVES)","Tomato relish"
+74406010,"BARBECUE SAUCE","Barbecue sauce"
+74406050,"BARBECUE SAUCE, REDUCED SODIUM","Barbecue sauce, reduced sodium"
+74406100,"STEAK SAUCE, TOMATO-BASE (INCLUDE A-1)","Steak sauce, tomato-base"
+74406500,"COCKTAIL SAUCE","Cocktail sauce"
+74410110,"PUERTO RICAN SEASONING WITH HAM","Puerto Rican seasoning with ham"
+74415110,"PUERTO RICAN SEASONING W/ HAM & TOMATO SAUCE","Puerto Rican seasoning with ham and tomato sauce"
+74420110,"PUERTO RICAN SEASONING WO/ HAM & TOMATO SAUCE","Puerto Rican seasoning without ham and tomato sauce"
+74501010,"TOMATO ASPIC","Tomato aspic"
+74503010,"TOMATO & CORN, COOKED, FAT NOT ADDED IN COOKING","Tomato and corn, cooked, fat not added in cooking"
+74504000,"TOMATO & OKRA, COOKED, NS AS TO ADDED FAT","Tomato and okra, cooked, NS as to fat added in cooking"
+74504010,"TOMATO & OKRA, COOKED, NO FAT ADDED","Tomato and okra, cooked, fat not added in cooking"
+74504020,"TOMATO & OKRA, COOKED, FAT ADDED","Tomato and okra, cooked, fat added in cooking"
+74504100,"TOMATO & ONION, COOKED, NS FAT ADDED","Tomato and onion, cooked, NS as to fat added in cooking"
+74504110,"TOMATO & ONION, COOKED, FAT NOT ADDED IN COOKING","Tomato and onion, cooked, fat not added in cooking"
+74504120,"TOMATO & ONION, COOKED, FAT ADDED","Tomato and onion, cooked, fat added in cooking"
+74504150,"TOMATO & CELERY, COOKED, FAT NOT ADDED IN COOKING","Tomato and celery, cooked, fat not added in cooking"
+74505000,"TOMATO W/ CORN & OKRA, COOKED, NS AS TO ADDED FAT","Tomato with corn and okra, cooked, NS as to fat added in cooking"
+74505010,"TOMATO W/ CORN & OKRA, COOKED, NO FAT ADDED","Tomato with corn and okra, cooked, fat not added in cooking"
+74505020,"TOMATO W/ CORN & OKRA, COOKED, FAT ADDED","Tomato with corn and okra, cooked, fat added in cooking"
+74506000,"TOMATO & CUCUMBER SALAD W/ OIL & VINEGAR","Tomato and cucumber salad made with tomato, cucumber, oil, and vinegar"
+74601000,"TOMATO SOUP, NFS","Tomato soup, NFS"
+74601010,"TOMATO SOUP, CREAM OF,PREP W/ MILK","Tomato soup, cream of, prepared with milk"
+74602010,"TOMATO SOUP, PREPARED WITH WATER, OR READY-TO-SERVE","Tomato soup, prepared with water, or ready-to-serve"
+74602050,"TOMATO SOUP, INSTANT TYPE, PREPARED W/ WATER","Tomato soup, instant type, prepared with water"
+74602200,"TOMATO SOUP, CANNED, REDUCED SODIUM, PREPARED WITH WATER, OR","Tomato soup, canned, reduced sodium, prepared with water, or ready-to-serve"
+74602300,"TOMATO SOUP, CANNED, REDUCED SODIUM, PREP W/ MILK","Tomato soup, canned, reduced sodium, prepared with milk"
+74603010,"TOMATO BEEF SOUP, PREPARED W/ WATER","Tomato beef soup, prepared with water"
+74604010,"TOMATO BEEF NOODLE SOUP, PREPARED W/ WATER","Tomato beef noodle soup, prepared with water"
+74604100,"TOMATO BEEF RICE SOUP, PREPARED W/ WATER","Tomato beef rice soup, prepared with water"
+74604500,"TOMATO NOODLE SOUP, CANNED, PREPARED WITH WATER OR READY-TO-","Tomato noodle soup, canned, prepared with water or ready-to-serve"
+74604600,"TOMATO NOODLE SOUP, CANNED, PREPARED WITH MILK","Tomato noodle soup, canned, prepared with milk"
+74605010,"TOMATO RICE SOUP, PREPARED W/ WATER","Tomato rice soup, prepared with water"
+74606010,"TOMATO VEGETABLE SOUP, PREP W/ WATER","Tomato vegetable soup, prepared with water"
+74606020,"TOMATO VEGETABLE SOUP W/NOODLES, PREPARED W/ WATER","Tomato vegetable soup with noodles, prepared with water"
+74701000,"TOMATO SANDWICH","Tomato sandwich"
+75100250,"RAW VEGETABLE, NFS","Raw vegetable, NFS"
+75100300,"SPROUTS, NFS","Sprouts, NFS"
+75100500,"ALFALFA SPROUTS, RAW","Alfalfa sprouts, raw"
+75100750,"ARTICHOKE, JERUSALEM, RAW (INCLUDE SUNCHOKE)","Artichoke, Jerusalem, raw"
+75100800,"ASPARAGUS, RAW","Asparagus, raw"
+75101000,"BEAN SPROUTS, RAW (SOYBEAN/MUNG)","Bean sprouts, raw (soybean or mung)"
+75101800,"BEANS, STRING, GREEN, RAW","Beans, string, green, raw"
+75102000,"BEANS, LIMA, RAW","Beans, lima, raw"
+75102500,"BEETS, RAW","Beets, raw"
+75102600,"BROCCOFLOWER, RAW","Broccoflower, raw"
+75102750,"BRUSSELS SPROUTS, RAW","Brussels sprouts, raw"
+75103000,"CABBAGE, GREEN, RAW","Cabbage, green, raw"
+75104000,"CABBAGE, CHINESE, RAW","Cabbage, Chinese, raw"
+75105000,"CABBAGE, RED, RAW","Cabbage, red, raw"
+75105500,"CACTUS, RAW","Cactus, raw"
+75107000,"CAULIFLOWER, RAW","Cauliflower, raw"
+75109000,"CELERY, RAW (INCLUDE CELERY, NFS)","Celery, raw"
+75109010,"FENNEL BULB, RAW","Fennel bulb, raw"
+75109400,"BASIL, RAW","Basil, raw"
+75109500,"CHIVES, RAW (INCLUDE CHIVES, NFS)","Chives, raw"
+75109550,"CILANTRO, RAW","Cilantro, raw"
+75109600,"CORN, RAW","Corn, raw"
+75111000,"CUCUMBER, RAW (INCLUDE CUCUMBER, NFS)","Cucumber, raw"
+75111200,"EGGPLANT, RAW","Eggplant, raw"
+75111500,"GARLIC, RAW","Garlic, raw"
+75111800,"JICAMA, RAW (INCLUDE YAMBEAN)","Jicama, raw"
+75112000,"KOHLRABI, RAW","Kohlrabi, raw"
+75112500,"LEEK, RAW","Leek, raw"
+75113000,"LETTUCE, RAW","Lettuce, raw"
+75113060,"LETTUCE, BOSTON, RAW","Lettuce, Boston, raw"
+75113070,"LETTUCE, MANOA","Lettuce, manoa"
+75113080,"LETTUCE, ARUGULA, RAW","Lettuce, arugula, raw"
+75114000,"MIXED SALAD GREENS, RAW","Mixed salad greens, raw"
+75115000,"MUSHROOMS, RAW","Mushrooms, raw"
+75117010,"ONIONS, YOUNG GREEN, RAW","Onions, young green, raw"
+75117020,"ONIONS, MATURE, RAW","Onions, mature, raw"
+75119000,"PARSLEY, RAW","Parsley, raw"
+75120000,"PEAS, GREEN, RAW","Peas, green, raw"
+75121000,"PEPPER, HOT CHILI, RAW (INCLUDE JALAPENO)","Pepper, hot chili, raw"
+75121400,"PEPPER, POBLANO, RAW","Pepper, poblano, raw"
+75121500,"PEPPER, SERRANO, RAW","Pepper, Serrano, raw"
+75122000,"PEPPER, RAW, NFS","Pepper, raw, NFS"
+75122100,"PEPPER, SWEET, GREEN, RAW","Pepper, sweet, green, raw"
+75122200,"PEPPER, SWEET, RED, RAW","Pepper, sweet, red, raw"
+75124000,"PEPPER, BANANA, RAW","Pepper, banana, raw"
+75125000,"RADISH, RAW","Radish, raw"
+75127000,"RUTABAGA, RAW","Rutabaga, raw"
+75127500,"SEAWEED, RAW (INCLUDE BLANCHED)","Seaweed, raw"
+75127750,"SNOWPEA (PEA POD), RAW","Snowpeas (pea pod), raw"
+75128000,"SQUASH, SUMMER, YELLOW, RAW","Squash, summer, yellow, raw"
+75128010,"SQUASH, SUMMER, GREEN, RAW (INCLUDE ZUCCHINI)","Squash, summer, green, raw"
+75129000,"TURNIP, RAW","Turnip, raw"
+75132000,"MIXED VEGETABLE JUICE (OTHER THAN TOMATO)","Mixed vegetable juice (vegetables other than tomato)"
+75132100,"CELERY JUICE","Celery juice"
+75140500,"BROCCOLI SALAD W/CAULIFLOWER,CHEESE,BACON,&DRESSING","Broccoli salad with cauliflower, cheese, bacon bits, and dressing"
+75140510,"BROCCOLI SLAW SALAD","Broccoli slaw salad"
+75140990,"CABBAGE SALAD OR COLESLAW, FROM FAST FOOD / RESTAURANT","Cabbage salad or coleslaw, from fast food / restaurant"
+75141000,"CABBAGE SALAD OR COLESLAW, MADE WITH COLESLAW DRESSING","Cabbage salad or coleslaw, made with coleslaw dressing"
+75141005,"CABBAGE SALAD OR COLESLAW, MADE W/ LIGHT COLESLAW DRESSING","Cabbage salad or coleslaw, made with light coleslaw dressing"
+75141020,"CABBAGE SALAD OR COLESLAW, W/ ITALIAN DRSG","Cabbage salad or coleslaw, made with Italian dressing"
+75141025,"CABBAGE SALAD OR COLESLAW, W/LT ITALIAN DRSG","Cabbage salad or coleslaw, made with light Italian dressing"
+75141030,"CABBAGE SALAD OR COLESLAW, W/ CREAMY DRSG","Cabbage salad or coleslaw, made with creamy dressing"
+75141035,"CABBAGE SALAD OR COLESLAW, W/ LT CREAMY DRSG","Cabbage salad or coleslaw, made with light creamy dressing"
+75141040,"CABBAGE SALAD OR COLESLAW, W/ FAT FREE DRSG","Cabbage salad or coleslaw, made with any type of fat free dressing"
+75141100,"CABBAGE SALAD OR COLESLAW, W/APPLES/RAISINS, DRESS","Cabbage salad or coleslaw with apples and/or raisins, with dressing"
+75141200,"CABBAGE SALAD OR COLESLAW, W/ PINEAPPLE, DRESSING","Cabbage salad or coleslaw with pineapple, with dressing"
+75141300,"CABBAGE, CHINESE, SALAD, W/ DRESSING","Cabbage, Chinese, salad, with dressing"
+75141500,"CELERY, STUFFED W/ CHEESE","Celery, stuffed with cheese"
+75142000,"CUCUMBER & VEGETABLE NAMASU","Cucumber and vegetable namasu"
+75142500,"CUCUMBER SALAD, MADE WITH SOUR CREAM DRESSING","Cucumber salad, made with sour cream dressing"
+75142550,"CUCUMBER SALAD, W/ ITALIAN DRSG","Cucumber salad, made with Italian dressing"
+75142600,"CUCUMBER SALAD MADE W/ CUCUMBER AND VINEGAR","Cucumber salad made with cucumber and vinegar"
+75143000,"LETTUCE SALAD W/ ASSORTED VEGETABLES","Lettuce, salad with assorted vegetables including tomatoes and/or carrots, no dressing"
+75143050,"LETTUCE SALAD, W/ ASST VEG, NO TOM OR CAR, NO DRESS","Lettuce, salad with assorted vegetables excluding tomatoes and carrots, no dressing"
+75143100,"LETTUCE SALAD, W/ AVOCADO, TOMATO/CAR, NO DRESS","Lettuce, salad with avocado, tomato, and/or carrots, with or without other vegetables, no dressing"
+75143200,"LETTUCE SALAD, W/ CHEESE, TOM/CAR, NO DRESSING","Lettuce, salad with cheese, tomato and/or carrots, with or without other vegetables, no dressing"
+75143300,"LETTUCE SALAD, W/ EGG, TOM/CAR, NO DRESSING","Lettuce, salad with egg, tomato, and/or carrots, with or without other vegetables, no dressing"
+75143350,"LETTUCE SALAD W/ EGG, CHEESE, TOM/CAR, NO DRESSING","Lettuce, salad with egg, cheese, tomato, and/or carrots, with or without other vegetables, no dressing"
+75144100,"LETTUCE, WILTED, W/ BACON DRESSING","Lettuce, wilted, with bacon dressing"
+75145000,"SEVEN-LAYER SALAD(LETTUCE, MAYO, CHEESE, EGG, PEAS)","Seven-layer salad (lettuce salad made with a combination of onion, celery, green pepper, peas, mayonnaise, cheese, eggs, and/or bacon)"
+75146000,"GREEK SALAD, NO DRESSING","Greek Salad, no dressing"
+75147000,"SPINACH SALAD, NO DRESSING","Spinach salad, no dressing"
+75148010,"COBB SALAD, NO DRESSING","Cobb salad, no dressing"
+75200100,"VEGETABLES, NS AS TO TYPE, NS AS TO ADDED FAT","Vegetables, NS as to type, cooked, NS as to fat added in cooking"
+75200110,"VEGETABLES, NS AS TO TYPE, NO FAT ADDED","Vegetables, NS as to type, cooked, fat not added in cooking"
+75200120,"VEGETABLES, NS AS TO TYPE, COOKED, FAT ADDED","Vegetables, NS as to type, cooked, fat added in cooking"
+75200600,"ALGAE, DRIED (INCLUDE SPIRULINA)","Algae, dried"
+75200700,"ALOE VERA JUICE","Aloe vera juice"
+75201000,"ARTICHOKE, GLOBE(FRENCH), CKD, NS FORM, NS FAT ADDED","Artichoke, globe (French), cooked, NS as to form, NS as to fat added in cooking"
+75201001,"ARTICHOKE, GLOBE(FRENCH), CKD, FROM FRESH, NS FAT ADDED","Artichoke, globe (French), cooked, from fresh, NS as to fat added in cooking"
+75201002,"ARTICHOKE, GLOBE(FRENCH), CKD, FROM FROZ, NS FAT ADDED","Artichoke, globe (French), cooked, from frozen, NS as to fat added in cooking"
+75201003,"ARTICHOKE, GLOBE(FRENCH), CKD, FROM CAN, NS FAT ADDED","Artichoke, globe (French), cooked, from canned, NS as to fat added in cooking"
+75201010,"ARTICHOKE,GLOBE (FRENCH),CKD,NS FORM,FAT NOT ADDED","Artichoke, globe (French), cooked, NS as to form, fat not added in cooking"
+75201011,"ARTICHOKE,GLOBE (FRENCH),CKD,FROM FRESH,FAT NOT ADDED","Artichoke, globe (French), cooked, from fresh, fat not added in cooking"
+75201012,"ARTICHOKE,GLOBE (FRENCH),CKD,FROM FROZ,FAT NOT ADDED","Artichoke, globe (French), cooked, from frozen, fat not added in cooking"
+75201013,"ARTICHOKE,GLOBE (FRENCH),CKD,FROM CAN,FAT NOT ADDED","Artichoke, globe (French), cooked, from canned, fat not added in cooking"
+75201020,"ARTICHOKE, GLOBE (FRENCH), CKD, NS FORM, FAT ADDED","Artichoke, globe (French), cooked, NS as to form, fat added in cooking"
+75201021,"ARTICHOKE, GLOBE (FRENCH), CKD, FROM FRESH, FAT ADDED","Artichoke, globe (French), cooked, from fresh, fat added in cooking"
+75201022,"ARTICHOKE, GLOBE (FRENCH), CKD, FROM FROZ, FAT ADDED","Artichoke, globe (French), cooked, from frozen, fat added in cooking"
+75201023,"ARTICHOKE, GLOBE (FRENCH), CKD, FROM CAN, FAT ADDED","Artichoke, globe (French), cooked, from canned, fat added in cooking"
+75201030,"ARTICHOKE SALAD IN OIL","Artichoke salad in oil"
+75202000,"ASPARAGUS, COOKED, NS AS TO FORM, NS FAT ADDED","Asparagus, cooked, NS as to form, NS as to fat added in cooking"
+75202001,"ASPARAGUS, COOKED, FROM FRESH, NS FAT ADDED","Asparagus, cooked, from fresh, NS as to fat added in cooking"
+75202002,"ASPARAGUS, COOKED, FROM FROZEN, NS FAT ADDED","Asparagus, cooked, from frozen, NS as to fat added in cooking"
+75202003,"ASPARAGUS, COOKED, FROM CANNED, NS FAT ADDED","Asparagus, cooked, from canned, NS as to fat added in cooking"
+75202010,"ASPARAGUS, COOKED, NS AS TO FORM, FAT NOT ADDED","Asparagus, cooked, NS as to form, fat not added in cooking"
+75202011,"ASPARAGUS, COOKED, FROM FRESH, FAT NOT ADDED","Asparagus, cooked, from fresh, fat not added in cooking"
+75202012,"ASPARAGUS, COOKED, FROM FROZEN, FAT NOT ADDED","Asparagus, cooked, from frozen, fat not added in cooking"
+75202013,"ASPARAGUS, COOKED, FROM CANNED, FAT NOT ADDED","Asparagus, cooked, from canned, fat not added in cooking"
+75202020,"ASPARAGUS, COOKED, NS AS TO FORM, FAT ADDED","Asparagus, cooked, NS as to form, fat added in cooking"
+75202021,"ASPARAGUS, COOKED, FROM FRESH, FAT ADDED","Asparagus, cooked, from fresh, fat added in cooking"
+75202022,"ASPARAGUS, COOKED, FROM FROZEN, FAT ADDED","Asparagus, cooked, from frozen, fat added in cooking"
+75202023,"ASPARAGUS, COOKED, FROM CANNED, FAT ADDED","Asparagus, cooked, from canned, fat added in cooking"
+75203000,"BAMBOO SHOOTS, COOKED, FAT NOT ADDED IN COOKING","Bamboo shoots, cooked, fat not added in cooking"
+75203020,"BAMBOO SHOOTS, COOKED, FAT ADDED IN COOKING","Bamboo shoots, cooked, fat added in cooking"
+75204000,"BEANS, LIMA, IMMATURE, COOKED, NS FORM, NS AS TO FAT","Beans, lima, immature, cooked, NS as to form, NS as to fat added in cooking"
+75204001,"BEANS, LIMA, IMMATURE, COOKED, FROM FRESH, NS FAT ADDED","Beans, lima, immature, cooked, from fresh, NS as to fat added in cooking"
+75204002,"BEANS, LIMA, IMMATURE, COOKED, FROM FROZEN, NS FAT ADDED","Beans, lima, immature, cooked, from frozen, NS as to fat added in cooking"
+75204003,"BEANS, LIMA, IMMATURE, COOKED, FROM CANNED, NS FAT ADDED","Beans, lima, immature, cooked, from canned, NS as to fat added in cooking"
+75204010,"BEANS, LIMA, IMMATURE, COOKED, NS FORM, NO FAT ADDED","Beans, lima, immature, cooked, NS as to form, fat not added in cooking"
+75204011,"BEANS, LIMA, IMMATURE, COOKED, FROM FRESH, NO FAT ADDED","Beans, lima, immature, cooked, from fresh, fat not added in cooking"
+75204012,"BEANS, LIMA, IMMATURE, COOKED, FROM FROZ, NO FAT ADDED","Beans, lima, immature, cooked, from frozen, fat not added in cooking"
+75204013,"BEANS, LIMA, IMMATURE, COOKED, FROM CAN, NO FAT ADDED","Beans, lima, immature, cooked, from canned, fat not added in cooking"
+75204020,"BEANS, LIMA, IMMATURE, COOKED, NS FORM, FAT ADDED","Beans, lima, immature, cooked, NS as to form, fat added in cooking"
+75204021,"BEANS, LIMA, IMMATURE, COOKED, FROM FRESH, FAT ADDED","Beans, lima, immature, cooked, from fresh, fat added in cooking"
+75204022,"BEANS, LIMA, IMMATURE, COOKED, FROM FROZ, FAT ADDED","Beans, lima, immature, cooked, from frozen, fat added in cooking"
+75204023,"BEANS, LIMA, IMMATURE, COOKED, FROM CAN, FAT ADDED","Beans, lima, immature, cooked, from canned, fat added in cooking"
+75204100,"BEANS, LIMA,IMMATURE,CANNED,LOW SODIUM,NS AS TO FAT","Beans, lima, immature, canned, low sodium, NS as to fat added in cooking"
+75204110,"BEANS, LIMA,IMMATURE,CANNED,LOW SODIUM,NO FAT ADDED","Beans, lima, immature, canned, low sodium, fat not added in cooking"
+75204120,"BEANS, LIMA, IMMATURE, CANNED,LOW SODIUM, FAT ADDED","Beans, lima, immature, canned, low sodium, fat added in cooking"
+75204980,"BEANS, STRING, CKD, NS FORM, NS COLOR, FAT ADDED","Beans, string, cooked, NS as to form, NS as to color, fat added in cooking"
+75204981,"BEANS, STRING, CKD, FROM FRESH, NS COLOR, FAT ADDED","Beans, string, cooked, from fresh, NS as to color, fat added in cooking"
+75204982,"BEANS, STRING, CKD, FROM FROZ, NS COLOR, FAT ADDED","Beans, string, cooked, from frozen, NS as to color, fat added in cooking"
+75204983,"BEANS, STRING, CKD, FROM CAN, NS COLOR, FAT ADDED","Beans, string, cooked, from canned, NS as to color, fat added in cooking"
+75204990,"BEANS, STRING, CKD, NS FORM, NS COLOR, NO FAT ADDED","Beans, string, cooked, NS as to form, NS as to color, fat not added in cooking"
+75204991,"BEANS, STRING, CKD, FROM FRESH, NS COLOR, NO FAT ADDED","Beans, string, cooked, from fresh, NS as to color, fat not added in cooking"
+75204992,"BEANS, STRING, CKD, FROM FROZ, NS COLOR, NO FAT ADDED","Beans, string, cooked, from frozen, NS as to color, fat not added in cooking"
+75204993,"BEANS, STRING, CKD, FROM CAN, NS COLOR, NO FAT ADDED","Beans, string, cooked, from canned, NS as to color, fat not added in cooking"
+75205000,"BEANS, STRING, CKD, NS FORM, NS COLOR, NS FAT ADDED","Beans, string, cooked, NS as to form, NS as to color, NS as to fat added in cooking"
+75205001,"BEANS, STRING, CKD, FROM FRESH, NS COLOR, NS FAT ADDED","Beans, string, cooked, from fresh, NS as to color, NS as to fat added in cooking"
+75205002,"BEANS, STRING, CKD, FROM FROZ, NS COLOR, NS FAT ADDED","Beans, string, cooked, from frozen, NS as to color, NS as to fat added in cooking"
+75205003,"BEANS, STRING, CKD, FROM CAN, NS COLOR, NS FAT ADDED","Beans, string, cooked, from canned, NS as to color, NS as to fat added in cooking"
+75205010,"BEANS, STRING, GREEN, COOKED, NS FORM, NS FAT ADDED","Beans, string, green, cooked, NS as to form, NS as to fat added in cooking"
+75205011,"BEANS, STRING, GREEN, COOKED, FROM FRESH, NS FAT ADDED","Beans, string, green, cooked, from fresh, NS as to fat added in cooking"
+75205012,"BEANS, STRING, GREEN, COOKED, FROM FROZEN, NS FAT ADDED","Beans, string, green, cooked, from frozen, NS as to fat added in cooking"
+75205013,"BEANS, STRING, GREEN, COOKED, FROM CANNED, NS FAT ADDED","Beans, string, green, cooked, from canned, NS as to fat added in cooking"
+75205020,"BEANS, STRING, GREEN, COOKED, NS FORM,FAT NOT ADDED","Beans, string, green, cooked, NS as to form, fat not added in cooking"
+75205021,"BEANS, STRING, GREEN, COOKED, FROM FRESH, FAT NOT ADDED","Beans, string, green, cooked, from fresh, fat not added in cooking"
+75205022,"BEANS, STRING, GREEN, COOKED, FROM FROZEN, FAT NOT ADDED","Beans, string, green, cooked, from frozen, fat not added in cooking"
+75205023,"BEANS, STRING, GREEN, COOKED, FROM CANNED, FAT NOT ADDED","Beans, string, green, cooked, from canned, fat not added in cooking"
+75205030,"BEANS, STRING, GREEN, COOKED, NS FORM, FAT ADDED","Beans, string, green, cooked, NS as to form, fat added in cooking"
+75205031,"BEANS, STRING, GREEN, COOKED, FROM FRESH, FAT ADDED","Beans, string, green, cooked, from fresh, fat added in cooking"
+75205032,"BEANS, STRING, GREEN, COOKED, FROM FROZEN, FAT ADDED","Beans, string, green, cooked, from frozen, fat added in cooking"
+75205033,"BEANS, STRING, GREEN, COOKED, FROM CANNED, FAT ADDED","Beans, string, green, cooked, from canned, fat added in cooking"
+75205110,"BEANS, GREEN, CANNED, LO NA, NS AS TO ADDED FAT","Beans, string, green, canned, low sodium, NS as to fat added in cooking"
+75205120,"BEANS, GREEN, CANNED, LO NA, FAT NOT ADDED","Beans, string, green, canned, low sodium, fat not added in cooking"
+75205130,"BEANS, GREEN, CANNED, LO NA, FAT ADDED","Beans, string, green, canned, low sodium, fat added in cooking"
+75206000,"BEANS, STRING, YELLOW, COOKED, NS FORM, NS ADDED FAT","Beans, string, yellow, cooked, NS as to form, NS as to fat added in cooking"
+75206001,"BEANS, STRING, YELLOW, COOKED, FROM FRESH, NS ADDED FAT","Beans, string, yellow, cooked, from fresh, NS as to fat added in cooking"
+75206002,"BEANS, STRING, YELLOW, COOKED, FROM FROZ, NS ADDED FAT","Beans, string, yellow, cooked, from frozen, NS as to fat added in cooking"
+75206003,"BEANS, STRING, YELLOW, COOKED, FROM CANNED, NS ADDED FAT","Beans, string, yellow, cooked, from canned, NS as to fat added in cooking"
+75206010,"BEANS, STRING, YELLOW, COOKED, NS FORM, NO FAT ADDED","Beans, string, yellow, cooked, NS as to form, fat not added in cooking"
+75206011,"BEANS, STRING, YELLOW, COOKED, FROM FRESH, NO FAT ADDED","Beans, string, yellow, cooked, from fresh, fat not added in cooking"
+75206012,"BEANS, STRING, YELLOW, COOKED, FROM FROZ, NO FAT ADDED","Beans, string, yellow, cooked, from frozen, fat not added in cooking"
+75206013,"BEANS, STRING, YELLOW, COOKED, FROM CANNED, NO FAT ADDED","Beans, string, yellow, cooked, from canned, fat not added in cooking"
+75206020,"BEANS, STRING, YELLOW, COOKED, NS FORM, FAT ADDED","Beans, string, yellow, cooked, NS as to form, fat added in cooking"
+75206021,"BEANS, STRING, YELLOW, COOKED, FROM FRESH, FAT ADDED","Beans, string, yellow, cooked, from fresh, fat added in cooking"
+75206022,"BEANS, STRING, YELLOW, COOKED, FROM FROZ, FAT ADDED","Beans, string, yellow, cooked, from frozen, fat added in cooking"
+75206023,"BEANS, STRING, YELLOW, COOKED, FROM CAN, FAT ADDED","Beans, string, yellow, cooked, from canned, fat added in cooking"
+75207000,"BEAN SPROUTS, COOKED, NS FORM, NS AS TO ADDED FAT","Bean sprouts, cooked, NS as to form, NS as to fat added in cooking"
+75207001,"BEAN SPROUTS, COOKED, FROM FRESH, NS  FAT ADDED","Bean sprouts, cooked, from fresh, NS as to fat added in cooking"
+75207003,"BEAN SPROUTS, COOKED, FROM CANNED, NS FAT ADDED","Bean sprouts, cooked, from canned, NS as to fat added in cooking"
+75207010,"BEAN SPROUTS, COOKED, NS AS TO FORM, FAT NOT ADDED","Bean sprouts, cooked, NS as to form, fat not added in cooking"
+75207011,"BEAN SPROUTS, COOKED, FROM FRESH, FAT NOT ADDED","Bean sprouts, cooked, from fresh, fat not added in cooking"
+75207013,"BEAN SPROUTS, COOKED, FROM CANNED, FAT NOT ADDED","Bean sprouts, cooked, from canned, fat not added in cooking"
+75207020,"BEAN SPROUTS, COOKED, NS AS TO FORM, FAT ADDED","Bean sprouts, cooked, NS as to form, fat added in cooking"
+75207021,"BEAN SPROUTS, COOKED, FROM FRESH, FAT ADDED","Bean sprouts, cooked, from fresh, fat added in cooking"
+75207023,"BEAN SPROUTS, COOKED, FROM CANNED, FAT ADDED","Bean sprouts, cooked, from canned, fat added in cooking"
+75208000,"BEETS, COOKED, NS AS TO FORM, NS AS TO FAT ADDED","Beets, cooked, NS as to form, NS as to fat added in cooking"
+75208001,"BEETS, COOKED, FROM FRESH, NS AS TO FAT ADDED","Beets, cooked, from fresh, NS as to fat added in cooking"
+75208002,"BEETS, COOKED, FROM FROZEN, NS AS TO FAT ADDED","Beets, cooked, from frozen, NS as to fat added in cooking"
+75208003,"BEETS, COOKED, FROM CANNED, NS AS TO FAT ADDED","Beets, cooked, from canned, NS as to fat added in cooking"
+75208010,"BEETS, COOKED, NS AS TO FORM, FAT NOT ADDED","Beets, cooked, NS as to form, fat not added in cooking"
+75208011,"BEETS, COOKED, FROM FRESH, FAT NOT ADDED","Beets, cooked, from fresh, fat not added in cooking"
+75208012,"BEETS, COOKED, FROM FROZEN, FAT NOT ADDED","Beets, cooked, from frozen, fat not added in cooking"
+75208013,"BEETS, COOKED, FROM CANNED, FAT NOT ADDED","Beets, cooked, from canned, fat not added in cooking"
+75208020,"BEETS, COOKED, NS AS TO FORM, FAT ADDED","Beets, cooked, NS as to form, fat added in cooking"
+75208021,"BEETS, COOKED, FROM FRESH, FAT ADDED","Beets, cooked, from fresh, fat added in cooking"
+75208022,"BEETS, COOKED, FROM FROZEN, FAT ADDED","Beets, cooked, from frozen, fat added in cooking"
+75208023,"BEETS, COOKED, FROM CANNED, FAT ADDED","Beets, cooked, from canned, fat added in cooking"
+75208100,"BEETS, CANNED, LOW SODIUM, NS AS TO ADDED FAT","Beets, canned, low sodium, NS as to fat added in cooking"
+75208110,"BEETS, CANNED, LOW SODIUM, FAT NOT ADDED","Beets, canned, low sodium, fat not added in cooking"
+75208120,"BEETS, CANNED, LOW SODIUM, FAT ADDED","Beets, canned, low sodium, fat added in cooking"
+75208290,"BITTERMELON, COOKED, NS AS TO ADDED FAT","Bitter melon, cooked, NS as to fat added in cooking"
+75208300,"BITTERMELON, COOKED, NO FAT ADDED(INCL BALSAM PEAR)","Bitter melon, cooked, fat not added in cooking"
+75208310,"BITTERMELON, COOKED, FAT ADDED (INCL BALSAM PEAR)","Bitter melon, cooked, fat added in cooking"
+75208500,"BREADFRUIT, COOKED,FAT NOT ADDED IN COOKING","Breadfruit, cooked, fat not added in cooking"
+75208520,"BREADFRUIT, FRIED","Breadfruit, fried"
+75208700,"BROCCOFLOWER,COOKED,NS AS TO FAT ADDED IN COOKING","Broccoflower, cooked, NS as to fat added in cooking"
+75208710,"BROCCOFLOWER,COOKED,FAT NOT ADDED IN COOKING","Broccoflower, cooked, fat not added in cooking"
+75208720,"BROCCOFLOWER,COOKED,FAT ADDED IN COOKING","Broccoflower, cooked, fat added in cooking"
+75209000,"BRUSSELS SPROUTS, COOKED, NS FORM, NS ADDED FAT","Brussels sprouts, cooked, NS as to form, NS as to fat added in cooking"
+75209001,"BRUSSELS SPROUTS, COOKED, FROM FRESH, NS ADDED FAT","Brussels sprouts, cooked, from fresh, NS as to fat added in cooking"
+75209002,"BRUSSELS SPROUTS, COOKED, FROM FROZ, NS ADDED FAT","Brussels sprouts, cooked, from frozen, NS as to fat added in cooking"
+75209010,"BRUSSELS SPROUTS, COOKED, NS FORM, FAT NOT ADDED","Brussels sprouts, cooked, NS as to form, fat not added in cooking"
+75209011,"BRUSSELS SPROUTS, COOKED, FROM FRESH, FAT NOT ADDED","Brussels sprouts, cooked, from fresh, fat not added in cooking"
+75209012,"BRUSSELS SPROUTS, COOKED, FROM FROZ, FAT NOT ADDED","Brussels sprouts, cooked, from frozen, fat not added in cooking"
+75209020,"BRUSSELS SPROUTS, COOKED, NS AS TO FORM, FAT ADDED","Brussels sprouts, cooked, NS as to form, fat added in cooking"
+75209021,"BRUSSELS SPROUTS, COOKED, FROM FRESH, FAT ADDED","Brussels sprouts, cooked, from fresh, fat added in cooking"
+75209022,"BRUSSELS SPROUTS, COOKED, FROM FROZ, FAT ADDED","Brussels sprouts, cooked, from frozen, fat added in cooking"
+75209500,"BURDOCK, COOKED, FAT NOT ADDED IN COOKING","Burdock, cooked, fat not added in cooking"
+75210000,"CABBAGE, CHINESE, COOKED, NS AS TO ADDED FAT","Cabbage, Chinese, cooked, NS as to fat added in cooking"
+75210010,"CABBAGE, CHINESE, COOKED, FAT NOT ADDED","Cabbage, Chinese, cooked, fat not added in cooking"
+75210020,"CABBAGE, CHINESE, COOKED, FAT ADDED","Cabbage, Chinese, cooked, fat added in cooking"
+75211010,"CABBAGE, GREEN, COOKED, NS FAT","Cabbage, green, cooked, NS as to fat added in cooking"
+75211020,"CABBAGE, GREEN, COOKED, FAT NOT ADDED","Cabbage, green, cooked, fat not added in cooking"
+75211030,"CABBAGE, GREEN, COOKED, FAT ADDED","Cabbage, green, cooked, fat added in cooking"
+75212000,"CABBAGE, RED, COOKED, NS AS TO ADDED FAT","Cabbage, red, cooked, NS as to fat added in cooking"
+75212010,"CABBAGE, RED, COOKED, FAT NOT ADDED","Cabbage, red, cooked, fat not added in cooking"
+75212020,"CABBAGE, RED, COOKED, FAT ADDED","Cabbage, red, cooked, fat added in cooking"
+75213000,"CABBAGE, SAVOY, COOKED, NS AS TO ADDED FAT","Cabbage, savoy, cooked, NS as to fat added in cooking"
+75213010,"CABBAGE, SAVOY, COOKED, FAT NOT ADDED","Cabbage, savoy, cooked, fat not added in cooking"
+75213020,"CABBAGE, SAVOY, COOKED, FAT ADDED","Cabbage, savoy, cooked, fat added in cooking"
+75213100,"CACTUS, COOKED, NS AS TO ADDED FAT","Cactus, cooked, NS as to fat added in cooking"
+75213110,"CACTUS, COOKED, FAT NOT ADDED","Cactus, cooked, fat not added in cooking"
+75213120,"CACTUS, COOKED, FAT ADDED","Cactus, cooked, fat added in cooking"
+75214000,"CAULIFLOWER, COOKED, NS FORM, NS FAT ADDED","Cauliflower, cooked, NS as to form, NS as to fat added in cooking"
+75214001,"CAULIFLOWER, COOKED, FROM FRESH, NS FAT ADDED","Cauliflower, cooked, from fresh, NS as to fat added in cooking"
+75214002,"CAULIFLOWER, COOKED, FROM FROZEN, NS FAT ADDED","Cauliflower, cooked, from frozen, NS as to fat added in cooking"
+75214003,"CAULIFLOWER, COOKED, FROM CANNED, NS FAT ADDED","Cauliflower, cooked, from canned, NS as to fat added in cooking"
+75214010,"CAULIFLOWER, COOKED, NS FORM, FAT NOT ADDED","Cauliflower, cooked, NS as to form, fat not added in cooking"
+75214011,"CAULIFLOWER, COOKED, FROM FRESH, FAT NOT ADDED","Cauliflower, cooked, from fresh, fat not added in cooking"
+75214012,"CAULIFLOWER, COOKED, FROM FROZEN, FAT NOT ADDED","Cauliflower, cooked, from frozen, fat not added in cooking"
+75214013,"CAULIFLOWER, COOKED, FROM CANNED, FAT NOT ADDED","Cauliflower, cooked, from canned, fat not added in cooking"
+75214020,"CAULIFLOWER, COOKED, NS AS TO FORM, FAT ADDED","Cauliflower, cooked, NS as to form, fat added in cooking"
+75214021,"CAULIFLOWER, COOKED, FROM FRESH, FAT ADDED","Cauliflower, cooked, from fresh, fat added in cooking"
+75214022,"CAULIFLOWER, COOKED, FROM FROZEN, FAT ADDED","Cauliflower, cooked, from frozen, fat added in cooking"
+75214023,"CAULIFLOWER, COOKED, FROM CANNED, FAT ADDED","Cauliflower, cooked, from canned, fat added in cooking"
+75215000,"CELERY, COOKED, NS AS TO ADDED FAT","Celery, cooked, NS as to fat added in cooking"
+75215010,"CELERY, COOKED, FAT NOT ADDED","Celery, cooked, fat not added in cooking"
+75215020,"CELERY, COOKED, FAT ADDED","Celery, cooked, fat added in cooking"
+75215100,"FENNEL BULB, COOKED, NS AS TO FAT ADDED","Fennel bulb, cooked, NS as to fat added in cooking"
+75215110,"FENNEL BULB, COOKED, FAT NOT ADDED IN COOKING","Fennel bulb, cooked, fat not added in cooking"
+75215120,"FENNEL BULB, COOKED, FAT ADDED IN COOKING","Fennel bulb, cooked, fat added in cooking"
+75215510,"CHRISTOPHINE, COOKED, FAT NOT ADDED IN COOKING","Christophine, cooked, fat not added in cooking"
+75216000,"CORN, COOKED, NS FORM, NS COLOR. NS FAT ADDED","Corn, cooked, NS as to form, NS as to color, NS as to fat added in cooking"
+75216001,"CORN, COOKED, FROM FRESH, NS COLOR. NS FAT ADDED","Corn, cooked, from fresh, NS as to color, NS as to fat added in cooking"
+75216002,"CORN, COOKED, FROM FROZEN, NS COLOR. NS FAT ADDED","Corn, cooked, from frozen, NS as to color, NS as to fat added in cooking"
+75216003,"CORN, COOKED, FROM CANNED, NS COLOR, NS FAT ADDED","Corn, cooked, from canned, NS as to color, NS as to fat added in cooking"
+75216010,"CORN, COOKED, NS FORM, NS COLOR, FAT NOT ADDED","Corn, cooked, NS as to form, NS as to color, fat not added in cooking"
+75216011,"CORN, COOKED, FROM FRESH, NS COLOR, FAT NOT ADDED","Corn, cooked, from fresh, NS as to color, fat not added in cooking"
+75216012,"CORN, COOKED, FROM FROZEN, NS COLOR, FAT NOT ADDED","Corn, cooked, from frozen, NS as to color, fat not added in cooking"
+75216013,"CORN, COOKED, FROM CANNED, NS COLOR, FAT NOT ADDED","Corn, cooked, from canned, NS as to color, fat not added in cooking"
+75216020,"CORN, COOKED, NS FORM, NS COLOR, FAT ADDED","Corn, cooked, NS as to form, NS as to color, fat added in cooking"
+75216021,"CORN, COOKED, FROM FRESH, NS COLOR, FAT ADDED","Corn, cooked, from fresh, NS as to color, fat added in cooking"
+75216022,"CORN, COOKED, FROM FROZEN, NS COLOR, FAT ADDED","Corn, cooked, from frozen, NS as to color, fat added in cooking"
+75216023,"CORN, COOKED, FROM CANNED, NS COLOR, FAT ADDED","Corn, cooked, from canned, NS as to color, fat added in cooking"
+75216050,"CORN, NS AS TO FORM, NS AS TO COLOR, CREAM STYLE","Corn, NS as to form, NS as to color, cream style"
+75216053,"CORN, FROM CANNED, NS AS TO COLOR, CREAM STYLE","Corn, from canned, NS as to color, cream style"
+75216070,"CORN, DRIED, COOKED","Corn, dried, cooked"
+75216100,"CORN, YELLOW, COOKED, NS FORM, NS FAT ADDED","Corn, yellow, cooked, NS as to form, NS as to fat added in cooking"
+75216101,"CORN, YELLOW, COOKED, FROM FRESH, NS FAT ADDED","Corn, yellow, cooked, from fresh, NS as to fat added in cooking"
+75216102,"CORN, YELLOW, COOKED, FROM FROZEN, NS FAT ADDED","Corn, yellow, cooked, from frozen, NS as to fat added in cooking"
+75216103,"CORN, YELLOW, COOKED, FROM CANNED, NS FAT ADDED","Corn, yellow, cooked, from canned, NS as to fat added in cooking"
+75216110,"CORN, YELLOW, COOKED, NS FORM, FAT NOT ADDED","Corn, yellow, cooked, NS as to form, fat not added in cooking"
+75216111,"CORN, YELLOW, COOKED, FROM FRESH, FAT NOT ADDED","Corn, yellow, cooked, from fresh, fat not added in cooking"
+75216112,"CORN, YELLOW, COOKED, FROM FROZEN, FAT NOT ADDED","Corn, yellow, cooked, from frozen, fat not added in cooking"
+75216113,"CORN, YELLOW, COOKED, FROM CANNED, FAT NOT ADDED","Corn, yellow, cooked, from canned, fat not added in cooking"
+75216120,"CORN, YELLOW, COOKED, NS FORM, FAT ADDED","Corn, yellow, cooked, NS as to form, fat added in cooking"
+75216121,"CORN, YELLOW, COOKED, FROM FRESH, FAT ADDED","Corn, yellow, cooked, from fresh, fat added in cooking"
+75216122,"CORN, YELLOW, COOKED, FROM FROZEN, FAT ADDED","Corn, yellow, cooked, from frozen, fat added in cooking"
+75216123,"CORN, YELLOW, COOKED, FROM CANNED, FAT ADDED","Corn, yellow, cooked, from canned, fat added in cooking"
+75216150,"CORN, YELLOW, NS AS TO FORM, CREAM STYLE","Corn, yellow, NS as to form, cream style"
+75216153,"CORN, YELLOW, FROM CANNED, CREAM STYLE","Corn, yellow, from canned, cream style"
+75216160,"CORN, YELLOW & WHITE, COOKED, NS FORM, NS FAT ADDED","Corn, yellow and white, cooked, NS as to form, NS as to fat added in cooking"
+75216161,"CORN, YELLOW & WHITE, COOKED, FROM FRESH, NS FAT ADDED","Corn, yellow and white, cooked, from fresh, NS as to fat added in cooking"
+75216162,"CORN, YELLOW & WHITE, COOKED, FROM FROZ, NS FAT ADDED","Corn, yellow and white, cooked, from frozen, NS as to fat added in cooking"
+75216163,"CORN, YELLOW & WHITE, COOKED, FROM CAN, NS FAT ADDED","Corn, yellow and white, cooked, from canned, NS as to fat added in cooking"
+75216170,"CORN, YELLOW & WHITE, COOKED, NS FORM, NO FAT ADDED","Corn, yellow and white, cooked, NS as to form, fat not added in cooking"
+75216171,"CORN, YELLOW & WHITE, COOKED, FROM FRESH, NO FAT ADDED","Corn, yellow and white, cooked, from fresh, fat not added in cooking"
+75216172,"CORN, YELLOW & WHITE, COOKED, FROM FROZ, NO FAT ADDED","Corn, yellow and white, cooked, from frozen, fat not added in cooking"
+75216173,"CORN, YELLOW & WHITE, COOKED, FROM CAN, NO FAT ADDED","Corn, yellow and white, cooked, from canned, fat not added in cooking"
+75216180,"CORN, YELLOW & WHITE, COOKED, NS FORM, FAT ADDED","Corn, yellow and white, cooked, NS as to form, fat added in cooking"
+75216181,"CORN, YELLOW & WHITE, COOKED, FROM FRESH, FAT ADDED","Corn, yellow and white, cooked, from fresh, fat added in cooking"
+75216182,"CORN, YELLOW & WHITE, COOKED, FROM FROZ, FAT ADDED","Corn, yellow and white, cooked, from frozen, fat added in cooking"
+75216183,"CORN, YELLOW & WHITE, COOKED, FROM CAN, FAT ADDED","Corn, yellow and white, cooked, from canned, fat added in cooking"
+75216190,"CORN, YELLOW, NS AS TO FORM, CREAM STYLE, FAT ADDED","Corn, yellow, NS as to form, cream style, fat added in cooking"
+75216193,"CORN, YELLOW, FROM CANNED, CREAM STYLE, FAT ADDED","Corn, yellow, from canned, cream style, fat added in cooking"
+75216200,"CORN, WHITE, COOKED, NS FORM, NS FAT ADDED","Corn, white, cooked, NS as to form, NS as to fat added in cooking"
+75216201,"CORN, WHITE, COOKED, FROM FRESH, NS FAT ADDED","Corn, white, cooked, from fresh, NS as to fat added in cooking"
+75216202,"CORN, WHITE, COOKED, FROM FROZEN, NS FAT ADDED","Corn, white, cooked, from frozen, NS as to fat added in cooking"
+75216203,"CORN, WHITE, COOKED, FROM CANNED, NS FAT ADDED","Corn, white, cooked, from canned, NS as to fat added in cooking"
+75216210,"CORN, WHITE, COOKED, NS AS TO FORM, FAT NOT ADDED","Corn, white, cooked, NS as to form, fat not added in cooking"
+75216211,"CORN, WHITE, COOKED, FROM FRESH, FAT NOT ADDED","Corn, white, cooked, from fresh, fat not added in cooking"
+75216212,"CORN, WHITE, COOKED, FROM FROZEN, FAT NOT ADDED","Corn, white, cooked, from frozen, fat not added in cooking"
+75216213,"CORN, WHITE, COOKED, FROM CANNED, FAT NOT ADDED","Corn, white, cooked, from canned, fat not added in cooking"
+75216220,"CORN, WHITE, COOKED, NS AS TO FORM, FAT ADDED","Corn, white, cooked, NS as to form, fat added in cooking"
+75216221,"CORN, WHITE, COOKED, FROM FRESH, FAT ADDED","Corn, white, cooked, from fresh, fat added in cooking"
+75216222,"CORN, WHITE, COOKED, FROM FROZEN, FAT ADDED","Corn, white, cooked, from frozen, fat added in cooking"
+75216223,"CORN, WHITE, COOKED, FROM CANNED, FAT ADDED","Corn, white, cooked, from canned, fat added in cooking"
+75216250,"CORN, WHITE, NS AS TO FORM, CREAM STYLE","Corn, white, NS as to form, cream style"
+75216253,"CORN, WHITE, FROM CANNED, CREAM STYLE","Corn, white, from canned, cream style"
+75216300,"CORN, YELLOW, CANNED, LO NA, NS AS TO ADDED FAT","Corn, yellow, canned, low sodium, NS as to fat added in cooking"
+75216310,"CORN, YELLOW, CANNED, LOW SODIUM, FAT NOT ADDED","Corn, yellow, canned, low sodium, fat not added in cooking"
+75216320,"CORN, YELLOW, CANNED, LOW SODIUM, FAT ADDED","Corn, yellow, canned, low sodium, fat added in cooking"
+75216700,"CUCUMBER, COOKED, NS AS TO ADDED FAT","Cucumber, cooked, NS as to fat added in cooking"
+75216710,"CUCUMBER, COOKED, FAT NOT ADDED","Cucumber, cooked, fat not added in cooking"
+75216720,"CUCUMBER, COOKED, FAT ADDED","Cucumber, cooked, fat added in cooking"
+75217000,"EGGPLANT, COOKED, NS AS TO ADDED FAT","Eggplant, cooked, NS as to fat added in cooking"
+75217010,"EGGPLANT, COOKED, FAT NOT ADDED","Eggplant, cooked, fat not added in cooking"
+75217020,"EGGPLANT, COOKED, FAT ADDED","Eggplant, cooked, fat added in cooking"
+75217300,"FLOWERS / BLOSSOMS OF SESBANIA/LILY/SQUASH, NO FAT","Flowers or blossoms of sesbania, squash, or lily, fat not added in cooking"
+75217400,"GARLIC, COOKED","Garlic, cooked"
+75217490,"HOMINY, COOKED, NS AS TO ADDED FAT","Hominy, cooked, NS as to fat added in cooking"
+75217500,"HOMINY, COOKED, NO FAT ADDED","Hominy, cooked, fat not added in cooking"
+75217520,"HOMINY, COOKED, FAT ADDED","Hominy, cooked, fat added in cooking"
+75218010,"KOHLRABI, COOKED,FAT NOT ADDED IN COOKING","Kohlrabi, cooked, fat not added in cooking"
+75218400,"LEEK, COOKED, NS AS TO FAT ADDED IN COOKING","Leek, cooked, NS as to fat added in cooking"
+75218500,"LOTUS ROOT, COOKED, FAT NOT ADDED IN COOKING","Lotus root, cooked, fat not added in cooking"
+75219000,"MUSHROOMS, COOKED, NS FORM, NS AS TO ADDED FAT","Mushrooms, cooked, NS as to form, NS as to fat added in cooking"
+75219001,"MUSHROOMS, COOKED, FROM FRESH, NS FAT ADDED","Mushrooms, cooked, from fresh, NS as to fat added in cooking"
+75219002,"MUSHROOMS, COOKED, FROM FROZ, NS FAT ADDED","Mushrooms, cooked, from frozen, NS as to fat added in cooking"
+75219003,"MUSHROOMS, COOKED, FROM CANNED, NS FAT ADDED","Mushrooms, cooked, from canned, NS as to fat added in cooking"
+75219010,"MUSHROOMS, COOKED, NS AS TO FORM, FAT NOT ADDED","Mushrooms, cooked, NS as to form, fat not added in cooking"
+75219011,"MUSHROOMS, COOKED, FROM FRESH, FAT NOT ADDED","Mushrooms, cooked, from fresh, fat not added in cooking"
+75219012,"MUSHROOMS, COOKED, FROM FROZ, FAT NOT ADDED","Mushrooms, cooked, from frozen, fat not added in cooking"
+75219013,"MUSHROOMS, COOKED, FROM CANNED, FAT NOT ADDED","Mushrooms, cooked, from canned, fat not added in cooking"
+75219020,"MUSHROOMS, COOKED, NS AS TO FORM, FAT ADDED","Mushrooms, cooked, NS as to form, fat added in cooking"
+75219021,"MUSHROOMS, COOKED, FROM FRESH, FAT ADDED","Mushrooms, cooked, from fresh, fat added in cooking"
+75219022,"MUSHROOMS, COOKED, FROM FROZ, FAT ADDED","Mushrooms, cooked, from frozen, fat added in cooking"
+75219023,"MUSHROOMS, COOKED, FROM CANNED, FAT ADDED","Mushrooms, cooked, from canned, fat added in cooking"
+75219100,"MUSHROOM, ASIAN, COOKED, FROM DRIED","Mushroom, Asian, cooked, from dried"
+75220000,"OKRA, COOKED, NS FORM, NS FAT ADDED","Okra, cooked, NS as to form, NS as to fat added in cooking"
+75220001,"OKRA, COOKED, FROM FRESH, NS FAT ADDED","Okra, cooked, from fresh, NS as to fat added in cooking"
+75220002,"OKRA, COOKED, FROM FROZ, NS FAT ADDED","Okra, cooked, from frozen, NS as to fat added in cooking"
+75220003,"OKRA, COOKED, FROM CANNED, NS FAT ADDED","Okra, cooked, from canned, NS as to fat added in cooking"
+75220010,"OKRA, COOKED, NS FORM, FAT NOT ADDED","Okra, cooked, NS as to form, fat not added in cooking"
+75220011,"OKRA, COOKED, FROM FRESH, FAT NOT ADDED","Okra, cooked, from fresh, fat not added in cooking"
+75220012,"OKRA, COOKED, FROM FROZ, FAT NOT ADDED","Okra, cooked, from frozen, fat not added in cooking"
+75220013,"OKRA, COOKED, FROM CANNED, FAT NOT ADDED","Okra, cooked, from canned, fat not added in cooking"
+75220020,"OKRA, COOKED, NS FORM, FAT ADDED","Okra, cooked, NS as to form, fat added in cooking"
+75220021,"OKRA, COOKED, FROM FRESH, FAT ADDED","Okra, cooked, from fresh, fat added in cooking"
+75220022,"OKRA, COOKED, FROM FROZ, FAT ADDED","Okra, cooked, from frozen, fat added in cooking"
+75220023,"OKRA, COOKED, FROM CANNED, FAT ADDED","Okra, cooked, from canned, fat added in cooking"
+75220050,"LETTUCE, COOKED, FAT NOT ADDED IN COOKING","Lettuce, cooked, fat not added in cooking"
+75220100,"LUFFA (CHINESE OKRA), COOKED, NO FAT ADDED","Luffa (Chinese okra), cooked, fat not added in cooking"
+75221000,"ONIONS, MATURE, COOKED, NS FORM, NS AS TO ADDED FAT","Onions, mature, cooked, NS as to form, NS as to fat added in cooking"
+75221001,"ONIONS, MATURE, COOKED, FROM FRESH, NS FAT ADDED","Onions, mature, cooked, from fresh, NS as to fat added in cooking"
+75221002,"ONIONS, MATURE, COOKED, FROM FROZ, NS FAT ADDED","Onions, mature, cooked, from frozen, NS as to fat added in cooking"
+75221010,"ONIONS, MATURE, COOKED, NS FORM, FAT NOT ADDED","Onions, mature, cooked, NS as to form, fat not added in cooking"
+75221011,"ONIONS, MATURE, COOKED, FROM FRESH, FAT NOT ADDED","Onions, mature, cooked, from fresh, fat not added in cooking"
+75221012,"ONIONS, MATURE, COOKED, FROM FROZ, FAT NOT ADDED","Onions, mature, cooked, from frozen, fat not added in cooking"
+75221020,"ONIONS, MATURE, COOKED, NS AS TO FORM, FAT ADDED","Onions, mature, cooked or sauteed, NS as to form, fat added in cooking"
+75221021,"ONIONS, MATURE, COOKED, FROM FRESH, FAT ADDED","Onions, mature, cooked or sauteed, from fresh, fat added in cooking"
+75221022,"ONIONS, MATURE, COOKED, FROM FROZ, FAT ADDED","Onions, mature, cooked or sauteed, from frozen, fat added in cooking"
+75221030,"ONIONS, PEARL, COOKED, NS FORM (INCL PICKLED/COCKTAIL)","Onions, pearl, cooked, NS as to form"
+75221031,"ONIONS, PEARL, COOKED, FROM FRESH (INCL PICKLED/COCKTAIL)","Onions, pearl, cooked, from fresh"
+75221032,"ONIONS, PEARL, COOKED, FROM FROZ (INCL PICKLED/COCKTAIL)","Onions, pearl, cooked, from frozen"
+75221033,"ONIONS, PEARL, COOKED, FROM CAN (INCL PICKLED/COCKTAIL)","Onions, pearl, cooked, from canned"
+75221040,"ONIONS, YOUNG GREEN, COOKED, NS FORM, NS ADDED FAT","Onion, young green, cooked, NS as to form, NS as to fat added in cooking"
+75221041,"ONIONS, YOUNG GREEN, COOKED, FROM FRESH, NS ADDED FAT","Onion, young green, cooked, from fresh, NS as to fat added in cooking"
+75221050,"ONIONS, YOUNG GREEN, COOKED, NS FORM, NO FAT ADDED","Onions, young green, cooked, NS as to form, fat not added in cooking"
+75221051,"ONIONS, YOUNG GREEN, COOKED, FROM FRESH, NO FAT ADDED","Onions, young green, cooked, from fresh, fat not added in cooking"
+75221060,"ONIONS, YOUNG GREEN, COOKED, NS FORM, FAT ADDED","Onion, young green, cooked, NS as to form, fat added in cooking"
+75221061,"ONIONS, YOUNG GREEN, COOKED, FROM FRESH, FAT ADDED","Onion, young green, cooked, from fresh, fat added in cooking"
+75221100,"ONIONS, DEHYDRATED","Onion, dehydrated"
+75221160,"PALM HEARTS, COOKED (ASSUME NO FAT ADDED)","Palm hearts, cooked (assume fat not added in cooking)"
+75221210,"PARSLEY, COOKED (ASSUME NO FAT ADDED)","Parsley, cooked (assume fat not added in cooking)"
+75222000,"PARSNIPS, COOKED, NS AS TO ADDED FAT","Parsnips, cooked, NS as to fat added in cooking"
+75222010,"PARSNIPS, COOKED, FAT NOT ADDED","Parsnips, cooked, fat not added in cooking"
+75222020,"PARSNIPS, COOKED, FAT ADDED","Parsnips, cooked, fat added in cooking"
+75223000,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,NS FORM,NS FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, NS as to form, NS as to fat added in cooking"
+75223001,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD, FROM FRESH,NS FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from fresh, NS as to fat added in cooking"
+75223002,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD, FROM FROZ,NS FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from frozen, NS as to fat added in cooking"
+75223003,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD, FROM CAN,NS FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from canned, NS as to fat added in cooking"
+75223010,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,NS FORM,NO FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, NS as to form, fat not added in cooking"
+75223011,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM FRESH,NO FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from fresh, fat not added in cooking"
+75223012,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM FROZ,NO FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from frozen, fat not added in cooking"
+75223013,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM CAN,NO FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from canned, fat not added in cooking"
+75223020,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,NS FORM,W/ FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, NS as to form, fat added in cooking"
+75223021,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM FRESH,W/ FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from fresh, fat added in cooking"
+75223022,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM FROZ,W/ FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from frozen, fat added in cooking"
+75223023,"PEAS,COW/FIELD/BLACKEYE,NOT DRY,CKD,FROM CAN,W/ FAT","Peas, cowpeas, field peas, or blackeye peas (not dried), cooked, from canned, fat added in cooking"
+75224010,"PEAS, GREEN, COOKED, NS FORM, NS AS TO ADDED FAT","Peas, green, cooked, NS as to form, NS as to fat added in cooking"
+75224011,"PEAS, GREEN, COOKED, FROM FRESH, NS FAT ADDED","Peas, green, cooked, from fresh, NS as to fat added in cooking"
+75224012,"PEAS, GREEN, COOKED, FROM FROZ, NS FAT ADDED","Peas, green, cooked, from frozen, NS as to fat added in cooking"
+75224013,"PEAS, GREEN, COOKED, FROM CANNED, NS FAT ADDED","Peas, green, cooked, from canned, NS as to fat added in cooking"
+75224020,"PEAS, GREEN, COOKED, NS AS TO FORM, FAT NOT ADDED","Peas, green, cooked, NS as to form, fat not added in cooking"
+75224021,"PEAS, GREEN, COOKED, FROM FRESH, FAT NOT ADDED","Peas, green, cooked, from fresh, fat not added in cooking"
+75224022,"PEAS, GREEN, COOKED, FROM FROZ, FAT NOT ADDED","Peas, green, cooked, from frozen, fat not added in cooking"
+75224023,"PEAS, GREEN, COOKED, FROM CANNED, FAT NOT ADDED","Peas, green, cooked, from canned, fat not added in cooking"
+75224030,"PEAS, GREEN, COOKED, NS AS TO FORM, FAT ADDED","Peas, green, cooked, NS as to form, fat added in cooking"
+75224031,"PEAS, GREEN, COOKED, FROM FRESH, FAT ADDED","Peas, green, cooked, from fresh, fat added in cooking"
+75224032,"PEAS, GREEN, COOKED, FROM FROZ, FAT ADDED","Peas, green, cooked, from frozen, fat added in cooking"
+75224033,"PEAS, GREEN, COOKED, FROM CANNED, FAT ADDED","Peas, green, cooked, from canned, fat added in cooking"
+75224110,"PEAS, GREEN, CANNED, LOW SODIUM, NS AS TO ADDED FAT","Peas, green, canned, low sodium, NS as to fat added in cooking"
+75224120,"PEAS, GREEN, CANNED, LOW SODIUM, FAT NOT ADDED","Peas, green, canned, low sodium, fat not added in cooking"
+75224130,"PEAS, GREEN, CANNED, LOW SODIUM, FAT ADDED","Peas, green, canned, low sodium, fat added in cooking"
+75225010,"PIGEON PEAS, COOKED, NS AS TO FORM, FAT NOT ADDED","Pigeon peas, cooked, NS as to form, fat not added in cooking"
+75225011,"PIGEON PEAS, COOKED, FROM FRESH, FAT NOT ADDED","Pigeon peas, cooked, from fresh, fat not added in cooking"
+75225013,"PIGEON PEAS, COOKED, FROM CANNED, FAT NOT ADDED","Pigeon peas, cooked, from canned, fat not added in cooking"
+75226000,"PEPPERS, GREEN, COOKED, NS AS TO FAT","Peppers, green, cooked, NS as to fat added in cooking"
+75226010,"PEPPERS, GREEN, COOKED, FAT NOT ADDED","Peppers, green, cooked, fat not added in cooking"
+75226020,"PEPPERS, GREEN, COOKED, FAT ADDED","Peppers, green, cooked, fat added in cooking"
+75226040,"PEPPERS, RED, COOKED, NS AS TO ADDED FAT","Peppers, red, cooked, NS as to fat added in cooking"
+75226050,"PEPPERS, RED, COOKED, FAT NOT ADDED","Peppers, red, cooked, fat not added in cooking"
+75226060,"PEPPERS, RED, COOKED, FAT ADDED","Peppers, red, cooked, fat added in cooking"
+75226090,"PEPPERS, HOT, COOKED, NS FORM, NS FAT ADDED","Peppers, hot, cooked, NS as to form, NS as to fat added in cooking"
+75226091,"PEPPERS, HOT, COOKED, FROM FRESH, NS FAT ADDED","Peppers, hot, cooked, from fresh, NS as to fat added in cooking"
+75226092,"PEPPERS, HOT, COOKED, FROM FROZ, NS FAT ADDED","Peppers, hot, cooked, from frozen, NS as to fat added in cooking"
+75226093,"PEPPERS, HOT, COOKED, FROM CANNED, NS FAT ADDED","Peppers, hot, cooked, from canned, NS as to fat added in cooking"
+75226100,"PEPPERS, HOT, COOKED, NS FORM, NO FAT ADDED","Peppers, hot, cooked, NS as to form, fat not added in cooking"
+75226101,"PEPPERS, HOT, COOKED, FROM FRESH, NO FAT ADDED","Peppers, hot, cooked, from fresh, fat not added in cooking"
+75226102,"PEPPERS, HOT, COOKED, FROM FROZ, NO FAT ADDED","Peppers, hot, cooked, from frozen, fat not added in cooking"
+75226103,"PEPPERS, HOT, COOKED, FROM CANNED, NO FAT ADDED","Peppers, hot, cooked, from canned, fat not added in cooking"
+75226110,"PEPPERS, HOT, COOKED, NS FORM, FAT ADDED","Peppers, hot, cooked, NS as to form, fat added in cooking"
+75226111,"PEPPERS, HOT, COOKED, FROM FRESH, FAT ADDED","Peppers, hot, cooked, from fresh, fat added in cooking"
+75226112,"PEPPERS, HOT, COOKED, FROM FROZ, FAT ADDED","Peppers, hot, cooked, from frozen, fat added in cooking"
+75226113,"PEPPERS, HOT, COOKED, FROM CANNED, FAT ADDED","Peppers, hot, cooked, from canned, fat added in cooking"
+75226700,"PIMIENTO","Pimiento"
+75227100,"RADISH, JAPANESE (DAIKON), COOKED, NO FAT ADDED","Radish, Japanese (daikon), cooked, fat not added in cooking"
+75227110,"RADISH, JAPANESE (DAIKON), COOKED, FAT ADDED","Radish, Japanese (daikon), cooked, fat added in cooking"
+75228000,"RUTABAGA, COOKED, NS AS TO ADDED FAT","Rutabaga, cooked, NS as to fat added in cooking"
+75228010,"RUTABAGA, COOKED, FAT NOT ADDED","Rutabaga, cooked, fat not added in cooking"
+75228020,"RUTABAGA, COOKED, FAT ADDED","Rutabaga, cooked, fat added in cooking"
+75229010,"SALSIFY (VEGETABLE OYSTER), COOKED, NO FAT ADDED","Salsify (vegetable oyster), cooked, fat not added in cooking"
+75230000,"SAUERKRAUT, NS AS TO ADDED FAT","Sauerkraut, cooked, NS as to fat added in cooking"
+75230010,"SAUERKRAUT, NO FAT ADDED","Sauerkraut, cooked, fat not added in cooking"
+75230020,"SAUERKRAUT, FAT ADDED","Sauerkraut, cooked, fat added in cooking"
+75230100,"SAUERKRAUT, CANNED, LO NA","Sauerkraut, canned, low sodium"
+75231000,"SNOWPEA(PEA POD), COOKED, NS FORM, NS AS TO FAT","Snowpea (pea pod), cooked, NS as to form, NS as to fat added in cooking"
+75231001,"SNOWPEA(PEA POD), COOKED, FROM FRESH, NS AS TO FAT","Snowpea (pea pod), cooked, from fresh, NS as to fat added in cooking"
+75231002,"SNOWPEA(PEA POD), COOKED, FROM FROZEN, NS AS TO FAT","Snowpea (pea pod), cooked, from frozen, NS as to fat added in cooking"
+75231010,"SNOWPEA(PEA POD), COOKED, NS FORM, NO FAT ADDED","Snowpea (pea pod), cooked, NS as to form, fat not added in cooking"
+75231011,"SNOWPEA(PEA POD), COOKED, FROM FRESH, NO FAT ADDED","Snowpea (pea pod), cooked, from fresh, fat not added in cooking"
+75231012,"SNOWPEA(PEA POD), COOKED, FROM FROZ, NO FAT ADDED","Snowpea (pea pod), cooked, from frozen, fat not added in cooking"
+75231020,"SNOWPEA(PEA POD), COOKED, NS AS TO FORM, FAT ADDED","Snowpea (pea pod), cooked, NS as to form, fat added in cooking"
+75231021,"SNOWPEA(PEA POD), COOKED, FROM FRESH, FAT ADDED","Snowpea (pea pod), cooked, from fresh, fat added in cooking"
+75231022,"SNOWPEA(PEA POD), COOKED, FROM FROZ, FAT ADDED","Snowpea (pea pod), cooked, from frozen, fat added in cooking"
+75232000,"SEAWEED, DRIED","Seaweed, dried"
+75232050,"SEAWEED, PREPARED W/ SOY SAUCE","Seaweed, prepared with soy sauce"
+75232100,"SEAWEED, COOKED, NS AS TO FAT ADDED IN COOKING","Seaweed, cooked, NS as to fat added in cooking"
+75232110,"SEAWEED, COOKED, FAT NOT ADDED IN COOKING","Seaweed, cooked, fat not added in cooking"
+75232120,"SEAWEED, COOKED, FAT ADDED IN COOKING","Seaweed, cooked, fat added in cooking"
+75233000,"SQUASH, SUMMER, COOKED, NS FORM, NS AS TO ADDED FAT","Squash, summer, cooked, NS as to form, NS as to fat added in cooking"
+75233001,"SQUASH, SUMMER, COOKED, FROM FRESH, NS FAT ADDED","Squash, summer, cooked, from fresh, NS as to fat added in cooking"
+75233002,"SQUASH, SUMMER, COOKED, FROM FROZ, NS FAT ADDED","Squash, summer, cooked, from frozen, NS as to fat added in cooking"
+75233003,"SQUASH, SUMMER, COOKED, FROM CANNED, NS FAT ADDED","Squash, summer, cooked, from canned, NS as to fat added in cooking"
+75233010,"SQUASH, SUMMER, COOKED, NS FORM, FAT NOT ADDED","Squash, summer, cooked, NS as to form, fat not added in cooking"
+75233011,"SQUASH, SUMMER,YELLOW OR GREEN, CKD, FRESH, FAT NOT ADDED","Squash, summer,yellow or green, cooked, from fresh, fat not added in cooking"
+75233012,"SQUASH, SUMMER, COOKED, FROM FROZ, FAT NOT ADDED","Squash, summer, cooked, from frozen, fat not added in cooking"
+75233013,"SQUASH, SUMMER, COOKED, FROM CANNED, FAT NOT ADDED","Squash, summer, cooked, from canned, fat not added in cooking"
+75233020,"SQUASH, SUMMER, COOKED, NS AS TO FORM, FAT ADDED","Squash, summer, cooked, NS as to form, fat added in cooking"
+75233021,"SQUASH, SUMMER, COOKED, FROM FRESH, FAT ADDED","Squash, summer, cooked, from fresh, fat added in cooking"
+75233022,"SQUASH, SUMMER, COOKED, FROM FROZ, FAT ADDED","Squash, summer, cooked, from frozen, fat added in cooking"
+75233023,"SQUASH, SUMMER, COOKED, FROM CANNED, FAT ADDED","Squash, summer, cooked, from canned, fat added in cooking"
+75233200,"SQUASH, SPAGHETTI, NS AS TO ADDED FAT","Squash, spaghetti, cooked, NS as to fat added in cooking"
+75233210,"SQUASH, SPAGHETTI, FAT ADDED","Squash, spaghetti, cooked, fat added in cooking"
+75233220,"SQUASH, SPAGHETTI, NO FAT ADDED","Squash, spaghetti, cooked, fat not added in cooking"
+75233510,"SEQUIN (PORTUGUESE SQUASH), COOKED, NO FAT ADDED","Sequin (Portuguese squash), cooked, fat not added in cooking"
+75234000,"TURNIP, COOKED, NS AS TO FORM, NS AS TO ADDED FAT","Turnip, cooked, NS as to form, NS as to fat added in cooking"
+75234001,"TURNIP, COOKED, FROM FRESH, NS AS TO ADDED FAT","Turnip, cooked, from fresh, NS as to fat added in cooking"
+75234002,"TURNIP, COOKED, FROM FROZ, NS AS TO ADDED FAT","Turnip, cooked, from frozen, NS as to fat added in cooking"
+75234003,"TURNIP, COOKED, FROM CAN, NS AS TO ADDED FAT","Turnip, cooked, from canned, NS as to fat added in cooking"
+75234010,"TURNIP, COOKED, NS AS TO FORM, FAT NOT ADDED","Turnip, cooked, NS as to form, fat not added in cooking"
+75234011,"TURNIP, COOKED, FROM FRESH, FAT NOT ADDED","Turnip, cooked, from fresh, fat not added in cooking"
+75234012,"TURNIP, COOKED, FROM FROZ, FAT NOT ADDED","Turnip, cooked, from frozen, fat not added in cooking"
+75234013,"TURNIP, COOKED, FROM CANNED, FAT NOT ADDED","Turnip, cooked, from canned, fat not added in cooking"
+75234020,"TURNIP, COOKED, NS AS TO FORM, FAT ADDED","Turnip, cooked, NS as to form, fat added in cooking"
+75234021,"TURNIP, COOKED, FROM FRESH, FAT ADDED","Turnip, cooked, from fresh, fat added in cooking"
+75234022,"TURNIP, COOKED, FROM FROZ, FAT ADDED","Turnip, cooked, from frozen, fat added in cooking"
+75234023,"TURNIP, COOKED, FROM CAN, FAT ADDED","Turnip, cooked, from canned, fat added in cooking"
+75235000,"WATER CHESTNUT","Water chestnut"
+75235750,"WINTER MELON, COOKED (INCL CHINESE MELON, TOGAN)","Winter melon, cooked"
+75236000,"YEAST (INCLUDE BREWER'S YEAST)","Yeast"
+75236500,"YEAST EXTRACT SPREAD (INCL VEGEMITE, MARMITE)","Yeast extract spread"
+75301100,"BEANS, LIMA, & CORN (SUCCOTASH), NS AS TO ADDED FAT","Beans, lima and corn (succotash), cooked, NS as to fat added in cooking"
+75301110,"BEANS, LIMA, & CORN (SUCCOTASH), NO FAT ADDED","Beans, lima and corn (succotash), cooked, fat not added in cooking"
+75301120,"BEANS, LIMA, & CORN (SUCCOTASH), FAT ADDED","Beans, lima and corn (succotash), cooked, fat added in cooking"
+75302010,"BEANS, STRING, GREEN, W/ TOMATOES, FAT NOT ADDED IN COOKING","Beans, string, green, with tomatoes, cooked, fat not added in cooking"
+75302020,"BEANS, STRING, GREEN, W/ ONIONS, FAT NOT ADDED IN COOKING","Beans, string, green, with onions, cooked, fat not added in cooking"
+75302030,"BEANS, STRING, GREEN, W/ CHICKPEAS, FAT NOT ADDED IN COOKING","Beans, string, green, with chickpeas, cooked, fat not added in cooking"
+75302040,"BEANS, STRING, GREEN, W/ ALMONDS,  FAT NOT ADDED IN COOKING","Beans, string, green, with almonds, cooked, fat not added in cooking"
+75302045,"BEANS, STRING, GREEN, W/ ALMONDS, FAT ADDED IN COOKING","Beans, string, green, with almonds, cooked, fat added in cooking"
+75302050,"BEANS, STRING, GREEN, & POTATOES, FAT NOT ADDED","Beans, string, green, and potatoes, cooked, fat not added in cooking"
+75302060,"BEANS, STRING, GREEN, W/ PINTO BEANS, FAT NOT ADDED","Beans, string, green, with pinto beans, cooked, fat not added in cooking"
+75302070,"BEANS, STRING, GREEN, W/ SPAETZEL, FAT NOT ADDED","Beans, string, green, with spaetzel, cooked, fat not added in cooking"
+75302080,"BEAN SALAD, YELLOW &/OR GREEN STRING BEANS","Bean salad, yellow and/or green string beans"
+75302200,"BEANS, STRING, GREEN, W/ ONIONS, NS AS TO FAT ADDED","Beans, string, green, with onions, NS as to fat added in cooking"
+75302210,"BEANS, STRING, GREEN, W/ ONIONS, FAT ADDED IN COOKING","Beans, string, green, with onions, fat added in cooking"
+75302500,"BEANS, STRING, GREEN, & POTATOES,  NS AS TO FAT ADDED","Beans, string, green, and potatoes, cooked, NS as to fat added in cooking"
+75302510,"BEANS, STRING, GREEN, & POTATOES, FAT ADDED","Beans, string, green, and potatoes, cooked, fat added in cooking"
+75303000,"CORN W/ PEPPERS, RED OR GREEN,COOKED, NS FAT ADDED","Corn with peppers, red or green, cooked, NS as to fat added in cooking"
+75303010,"CORN W/ PEPPERS, RED OR GREEN,COOKED, NO FAT ADDED","Corn with peppers, red or green, cooked, fat not added in cooking"
+75303020,"CORN W/ PEPPERS, RED OR GREEN,COOKED,FAT ADDED","Corn with peppers, red or green, cooked, fat added in cooking"
+75306010,"EGGPLANT IN TOM SCE, COOKED,NO FAT ADDED","Eggplant in tomato sauce, cooked, fat not added in cooking"
+75307000,"GREEN PEPPERS & ONIONS, COOKED,FAT ADDED IN COOKING","Green peppers and onions, cooked, fat added in cooking"
+75311000,"MIXED VEGS (CORN,LIMA,PEAS,GRBNS,CAR), NS FORM & FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, NS as to form, NS as to fat added in cooking"
+75311002,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), FROZ, NS FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from frozen, NS as to fat added in cooking"
+75311003,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), CANNED, NS FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from canned, NS as to fat added in cooking"
+75311010,"MIXED VEGS (CORN,LIMA,PEAS,GRBN,CAR), NS FORM, NO FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, NS as to form, fat not added in cooking"
+75311012,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), FROZ, NO FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from frozen, fat not added in cooking"
+75311013,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), CANNED, NO FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from canned, fat not added in cooking"
+75311020,"MIXED VEGS (CORN,LIMA,PEAS,GRBNS,CAR), NS FORM,W/FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, NS as to form, fat added in cooking"
+75311022,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), FROZ, W/ FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from frozen, fat added in cooking"
+75311023,"MIXED VEGETABLES (CORN,LIMA,PEAS,GRBNS,CAR), CANNED, W/ FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), cooked, from canned, fat added in cooking"
+75311100,"MIXED VEGETABLES, CANNED, LOW SODIUM, NS ADDED FAT","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), canned, low sodium, NS as to fat added in cooking"
+75311110,"MIXED VEGETABLES, CANNED, LOW SODIUM, NO FAT ADDED","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), canned, low sodium, fat not added in cooking"
+75311120,"MIXED VEGETABLES, CANNED, LOW SODIUM, FAT ADDED","Mixed vegetables (corn, lima beans, peas, green beans, and carrots), canned, low sodium, fat added in cooking"
+75315000,"PEAS & CORN, COOKED, NS AS TO ADDED FAT","Peas and corn, cooked, NS as to fat added in cooking"
+75315010,"PEAS & CORN, COOKED, NO FAT ADDED","Peas and corn, cooked, fat not added in cooking"
+75315020,"PEAS & CORN, COOKED, FAT ADDED","Peas and corn, cooked, fat added in cooking"
+75315100,"PEAS & ONIONS, COOKED, NS AS TO ADDED FAT","Peas and onions, cooked, NS as to fat added in cooking"
+75315110,"PEAS & ONIONS, COOKED, FAT NOT ADDED","Peas and onions, cooked, fat not added in cooking"
+75315120,"PEAS & ONIONS, COOKED, FAT ADDED","Peas and onions, cooked, fat added in cooking"
+75315200,"PEAS W/ MUSHROOMS, COOKED, NS AS TO FAT","Peas with mushrooms, cooked, NS as to fat added in cooking"
+75315210,"PEAS W/ MUSHROOMS, COOKED, NO FAT ADDED","Peas with mushrooms, cooked, fat not added in cooking"
+75315215,"PEAS W/ MUSHROOMS, COOKED, FAT ADDED","Peas with mushrooms, cooked, fat added in cooking"
+75315250,"COWPEAS W/ SNAP BEANS,COOKED, NO FAT ADDED IN COOK","Cowpeas with snap beans, cooked, fat not added in cooking"
+75315300,"PEAS & POTATOES, COOKED, NO FAT ADDED IN COOKING","Peas and potatoes, cooked, fat not added in cooking"
+75315305,"PEAS AND POTATOES, COOKED, NS AS TO FAT","Peas and potatoes, cooked, NS as to fat added in cooking"
+75315310,"PEAS AND POTATOES, COOKED, FAT ADDED","Peas and potatoes, cooked, fat added in cooking"
+75316000,"SQUASH, SUMMER, & ONIONS,COOKED, NO FAT ADDED","Squash, summer, and onions, cooked, fat not added in cooking"
+75316010,"ZUCCHINI W/ TOM SCE, CKD,NO FAT ADDED IN COOKING","Zucchini with tomato sauce, cooked, fat not added in cooking"
+75316020,"SQUASH, SUMMER, & ONIONS, COOKED, FAT ADDED","Squash, summer, and onions, cooked, fat added in cooking"
+75316050,"RATATOUILLE","Ratatouille"
+75317000,"VEGETABLES,STEWTYPE(POT,CRT,ONION,CELERY)COOK,NS FA","Vegetables, stew type (including potatoes, carrots, onions, celery) cooked, NS as to fat added in cooking"
+75317010,"VEGETABLES,STEWTYPE(POT,CRT,ONION,CELERY)COOK,W/FAT","Vegetables, stew type (including potatoes, carrots, onions, celery) cooked, fat added in cooking"
+75317020,"VEGETABLES,STEWTYPE(POT,CRT,ONION,CELERY)COOK,NO FA","Vegetables, stew type (including potatoes, carrots, onions, celery) cooked, fat not added in cooking"
+75330100,"VEG COMBINATION (INCL CAR/ DK GRN), NO SAUCE, NS FAT","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, no sauce, NS as to fat added in cooking"
+75330110,"VEG COMBINATION (INCL CAR/ DK GRN), NO SAUCE, FAT NOT ADDED","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, no sauce, fat not added in cooking"
+75330120,"VEG COMBINATION (INCL CAR/ DK GRN), NO SAUCE, FAT ADDED","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, no sauce, fat added in cooking"
+75330130,"VEG COMBINATION (NO CAR/ DK GRN), NO SAUCE, NS FAT","Vegetable combination (excluding carrots, broccoli, and dark-green leafy), cooked, no sauce, NS as to fat added in cooking"
+75330140,"VEG COMBINATION (NO CAR/ DK GRN), NO SAUCE, FAT NOT ADDED","Vegetable combination (excluding carrots, broccoli, and dark-green leafy), cooked, no sauce,  fat not added in cooking"
+75330150,"VEG COMBINATION (NO CAR/ DK GRN), NO SAUCE, FAT ADDED","Vegetable combination (excluding carrots, broccoli, and dark-green leafy), cooked, no sauce,  fat added in cooking"
+75340000,"VEG ASIAN,ORIENTAL STYLE,CKD,NS FAT ADDED IN COOKING","Vegetable combinations, Asian style, (broccoli, green pepper, water chestnut, etc) cooked, NS as to fat added in cooking"
+75340010,"VEG COMBO ASIAN STYLE, CKD, FAT, NOT ADDED","Vegetable combinations, Asian style, (broccoli, green pepper,  water chestnuts, etc), cooked, fat not added in cooking"
+75340020,"VEG COMBO, ASIAN STYLE, CKD, FAT ADDED","Vegetable combinations, Asian style, (broccoli, green pepper, water chestnuts, etc), cooked, fat added in cooking"
+75340160,"VEG & PASTA COMBOS, W/ CREAM/CHEESE SCE, COOKED","Vegetable and pasta combinations with cream or cheese sauce (broccoli, pasta, carrots, corn, zucchini, peppers, cauliflower, peas, etc.), cooked"
+75340200,"JAI, MONK'S FOOD (MSHRMS,LILY RTS,B.CURD,W.CHSTNUT)","Jai, Monk's Food (mushrooms, lily roots, bean curd, water chestnuts)"
+75340300,"PINACBET (EGGPLANT W/ TOMATO, BITTERMELON, ETC)","Pinacbet (eggplant with tomatoes, bitter melon, etc.)"
+75365000,"VEGETABLE MIXTURE, DRIED (INCL SALAD CRUNCHIES)","Vegetable mixture, dried"
+75400500,"ARTICHOKES, STUFFED","Artichokes, stuffed"
+75401010,"ASPARAGUS, NS FORM, CREAMED OR W/ CHEESE SAUCE","Asparagus, NS as to form, creamed or with cheese sauce"
+75401011,"ASPARAGUS, FROM FRESH, CREAMED OR W/ CHEESE SAUCE","Asparagus, from fresh, creamed or with cheese sauce"
+75401012,"ASPARAGUS, FROM FROZEN, CREAMED OR W/ CHEESE SAUCE","Asparagus, from frozen, creamed or with cheese sauce"
+75401013,"ASPARAGUS, FROM CANNED, CREAMED OR W/ CHEESE SAUCE","Asparagus, from canned, creamed or with cheese sauce"
+75402010,"BEANS, LIMA, IMMATURE, NS FORM, CREAMED/ CHEESE SCE","Beans, lima, immature, NS as to form, creamed or with cheese sauce"
+75402011,"BEANS, LIMA, IMMATURE, FROM FRESH, CREAMED/CHEESE SCE","Beans, lima, immature, from fresh, creamed or with cheese sauce"
+75402012,"BEANS, LIMA, IMMATURE, FROM FROZEN, CREAMED/CHEESE SCE","Beans, lima, immature, from frozen, creamed or with cheese sauce"
+75402013,"BEANS, LIMA, IMMATURE, FROM CANNED, CREAMED/CHEESE SCE","Beans, lima, immature, from canned, creamed or with cheese sauce"
+75402020,"BEANS, LIMA, IMMATURE, CKD, NS FORM, W/ MUSHROOM SCE","Beans, lima, immature, cooked, NS as to form, with mushroom sauce"
+75402021,"BEANS, LIMA, IMMATURE, CKD, FROM FRESH, W/ MUSHROOM SCE","Beans, lima, immature, cooked, from fresh, with mushroom sauce"
+75402022,"BEANS, LIMA, IMMATURE, CKD, FROM FROZ, W/ MUSHROOM SCE","Beans, lima, immature, cooked, from frozen, with mushroom sauce"
+75402023,"BEANS, LIMA, IMMATURE, CKD, FROM CAN, W/ MUSHROOM SCE","Beans, lima, immature, cooked, from canned, with mushroom sauce"
+75403010,"BEANS, STRING, GREEN, NS FORM, CREAMED/CHEESE SCE","Beans, string, green, NS as to form, creamed or with cheese sauce"
+75403011,"BEANS, STRING, GREEN, FROM FRESH, CREAMED/CHEESE SCE","Beans, string, green, from fresh, creamed or with cheese sauce"
+75403012,"BEANS, STRING, GREEN, FROM FROZEN, CREAMED/CHEESE SCE","Beans, string, green, from frozen, creamed or with cheese sauce"
+75403013,"BEANS, STRING, GREEN, FROM CANNED, CREAMED/CHEESE SCE","Beans, string, green, from canned, creamed or with cheese sauce"
+75403020,"BEANS, STRING, GREEN, CKD, NS FORM,  W/ MUSHROOM SCE","Beans, string, green, cooked, NS as to form, with mushroom sauce"
+75403021,"BEANS, STRING, GREEN, CKD, FROM FRESH, W/ MUSHROOM SCE","Beans, string, green, cooked, from fresh, with mushroom sauce"
+75403022,"BEANS, STRING, GREEN, CKD, FROM FROZ, W/ MUSHROOM SCE","Beans, string, green, cooked, from frozen, with mushroom sauce"
+75403023,"BEANS, STRING, GREEN, CKD, FROM CAN, W/ MUSHROOM SCE","Beans, string, green, cooked, from canned, with mushroom sauce"
+75403200,"BEANS, STRING, GREEN, SZECHUAN-STYLE, FAT ADDED","Beans, string, green, cooked, Szechuan-style, fat added in cooking"
+75404010,"BEANS, STRING, YELLOW, NS FORM, CREAMED/ CHEESE SCE","Beans, string, yellow, NS as to form, creamed or with cheese sauce"
+75404011,"BEANS, STRING, YELLOW, FROM FRESH, CREAMED/ CHEESE SCE","Beans, string, yellow, from fresh, creamed or with cheese sauce"
+75404012,"BEANS, STRING, YELLOW, FROM FROZ, CREAMED/ CHEESE SCE","Beans, string, yellow, from frozen, creamed or with cheese sauce"
+75404013,"BEANS, STRING, YELLOW, FROM CANNED, CREAMED/ CHEESE SCE","Beans, string, yellow, from canned, creamed or with cheese sauce"
+75405010,"BEETS WITH HARVARD SAUCE","Beets with Harvard sauce"
+75406010,"BRUSSEL SPROUTS, NS AS TO FORM, CREAMED","Brussels sprouts, NS as to form, creamed"
+75406011,"BRUSSEL SPROUTS, FROM FRESH, CREAMED","Brussels sprouts, from fresh, creamed"
+75406012,"BRUSSEL SPROUTS, FROM FROZ, CREAMED","Brussels sprouts, from frozen, creamed"
+75407010,"CABBAGE, CREAMED","Cabbage, creamed"
+75409010,"CAULIFLOWER, NS FORM, CREAMED(INCL W/ CHEESE SAUCE)","Cauliflower, NS as to form, creamed"
+75409011,"CAULIFLOWER, FROM FRESH, CREAMED(INCL W/ CHEESE SAUCE)","Cauliflower, from fresh, creamed"
+75409012,"CAULIFLOWER, FROM FROZ, CREAMED(INCL W/ CHEESE SAUCE)","Cauliflower, from frozen, creamed"
+75409013,"CAULIFLOWER, FROM CANNED, CREAMED(INCL W/ CHEESE SAUCE)","Cauliflower, from canned, creamed"
+75409020,"CAULIFLOWER, BATTER-DIPPED, FRIED","Cauliflower, batter-dipped, fried"
+75410010,"CELERY, CREAMED","Celery, creamed"
+75410500,"CHILES RELLENOS, CHEESE-FILLED","Chiles rellenos, cheese-filled (stuffed chili peppers)"
+75410530,"CHILES RELLENOS, FILLED W/ MEAT & CHEESE","Chiles rellenos, filled with meat and cheese (stuffed chili peppers)"
+75410550,"JALAPENO PEPPER, STUFFED W/ CHEESE, BATTERED, FRIED","Jalapeno pepper, stuffed with cheese, breaded or battered, fried"
+75411010,"CORN, SCALLOPED OR PUDDING (INCLUDE CORN SOUFFLE)","Corn, scalloped or pudding"
+75411020,"CORN FRITTER","Corn fritter"
+75411030,"CORN, COOKED, NS FORM, W/ CREAM SAUCE, MADE W/ MILK","Corn, cooked, NS as to form, with cream sauce, made with milk"
+75411031,"CORN, COOKED, FROM FRESH, W/ CREAM SAUCE, MADE W/ MILK","Corn, cooked, from fresh, with cream sauce, made with milk"
+75411032,"CORN, COOKED, FROM FROZ, W/ CREAM SAUCE, MADE W/ MILK","Corn, cooked, from frozen, with cream sauce, made with milk"
+75411033,"CORN, COOKED, FROM CAN, W/ CREAM SAUCE, MADE W/ MILK","Corn, cooked, from canned, with cream sauce, made with milk"
+75412010,"EGGPLANT, BATTER-DIPPED, FRIED","Eggplant, batter-dipped, fried"
+75412030,"EGGPLANT DIP (INCL BABA GHANOUSH)","Eggplant dip"
+75412060,"EGGPLANT PARMESAN CASSEROLE, REGULAR","Eggplant parmesan casserole, regular"
+75412070,"EGGPLANT W/ CHEESE & TOMATO SAUCE","Eggplant with cheese and tomato sauce"
+75413010,"KOHLRABI, CREAMED","Kohlrabi, creamed"
+75414010,"MUSHROOMS, NS AS TO FORM, CREAMED","Mushrooms, NS as to form, creamed"
+75414011,"MUSHROOMS, FROM FRESH, CREAMED","Mushrooms, from fresh, creamed"
+75414012,"MUSHROOMS, FROM FROZEN, CREAMED","Mushrooms, from frozen, creamed"
+75414013,"MUSHROOMS, FROM CANNED, CREAMED","Mushrooms, from canned, creamed"
+75414020,"MUSHROOMS, STUFFED","Mushrooms, stuffed"
+75414030,"MUSHROOM, BATTER-DIPPED, FRIED","Mushrooms, batter-dipped, fried"
+75414500,"OKRA, BATTER-DIPPED, FRIED","Okra, batter-dipped, fried"
+75415010,"ONIONS, NS AS TO FORM, CREAMED","Onions, NS as to form, creamed"
+75415011,"ONIONS, FROM FRESH, CREAMED","Onions, from fresh, creamed"
+75415020,"ONION RINGS, NS FORM, BATTER-DIPPED, BAKED/FRIED","Onion rings, NS as to form, batter-dipped, baked or fried"
+75415021,"ONION RINGS, FROM FRESH, BATTERED, BAKED/FRIED","Onion rings, from fresh, batter-dipped, baked or fried"
+75415022,"ONION RINGS, FROM FROZ, BATTERED, BAKED/FRIED","Onion rings, from frozen, batter-dipped, baked or fried"
+75416010,"PARSNIPS, CREAMED","Parsnips, creamed"
+75416500,"PEA SALAD","Pea salad"
+75416600,"PEA SALAD W/ CHEESE","Pea salad with cheese"
+75417010,"PEAS, NS AS TO FORM, CREAMED","Peas, NS as to form, creamed"
+75417011,"PEAS, FROM FRESH, CREAMED","Peas, from fresh, creamed"
+75417012,"PEAS, FROM FROZEN, CREAMED","Peas, from frozen, creamed"
+75417013,"PEAS, FROM CANNED, CREAMED","Peas, from canned, creamed"
+75417020,"PEAS, COOKED, NS AS TO FORM,  W/ MUSHROOM SAUCE","Peas, cooked, NS as to form, with mushroom sauce"
+75417021,"PEAS, COOKED, FROM FRESH,  W/ MUSHROOM SAUCE","Peas, cooked, from fresh, with mushroom sauce"
+75417022,"PEAS, COOKED, FROM FROZEN,  W/ MUSHROOM SAUCE","Peas, cooked, from frozen, with mushroom sauce"
+75417023,"PEAS, COOKED, FROM CANNED,  W/ MUSHROOM SAUCE","Peas, cooked, from canned, with mushroom sauce"
+75417030,"PEAS, COOKED, NS AS TO FORM, W/ TOMATO SAUCE","Peas, cooked, NS as to form, with tomato sauce"
+75417031,"PEAS, COOKED, FROM FRESH, W/ TOMATO SAUCE","Peas, cooked, from fresh, with tomato sauce"
+75417032,"PEAS, COOKED, FROM FROZEN, W/ TOMATO SAUCE","Peas, cooked, from frozen, with tomato sauce"
+75417033,"PEAS, COOKED, FROM CANNED, W/ TOMATO SAUCE","Peas, cooked, from canned, with tomato sauce"
+75418000,"SQUASH, SUMMER, BREADED, BAKED","Squash,summer, yellow or green, breaded or battered, baked"
+75418010,"SQUASH, SUMMER, BREADED OR BATTERED, FRIED","Squash, summer, yellow or green, breaded or battered, fried"
+75418020,"SQUASH, SUMMER, CASSEROLE, W/ TOMATO & CHEESE","Squash, summer, casserole with tomato and cheese"
+75418030,"SQUASH, SUMMER, CASSEROLE, W/ RICE & TOMATO SAUCE","Squash, summer, casserole, with rice and tomato sauce"
+75418040,"SQUASH, SUMMER, CASSEROLE, W/ CHEESE SAUCE","Squash, summer, casserole, with cheese sauce"
+75418050,"SQUASH, SUMMER, NS AS TO FORM, CREAMED","Squash, summer, NS as to form, creamed"
+75418051,"SQUASH, SUMMER, FROM FRESH, CREAMED","Squash, summer, from fresh, creamed"
+75418052,"SQUASH, SUMMER, FROM FROZEN, CREAMED","Squash, summer, from frozen, creamed"
+75418053,"SQUASH, SUMMER, FROM CANNED, CREAMED","Squash, summer, from canned, creamed"
+75418060,"SQUASH, SUMMER, SOUFFLE","Squash, summer, souffle"
+75418100,"TURNIPS, NS AS TO FORM, CREAMED","Turnips, NS as to form, creamed"
+75418101,"TURNIPS, FROM FRESH, CREAMED","Turnips, from fresh, creamed"
+75418102,"TURNIPS, FROM FROZEN, CREAMED","Turnips, from frozen, creamed"
+75418103,"TURNIPS, FROM CANNED, CREAMED","Turnips, from canned, creamed"
+75418220,"CREAMED CHRISTOPHINE, P.R. (CHAYOTE A LA CREMA)","Creamed christophine, Puerto Rican style (Chayote a la crema)"
+75439010,"VEGETABLE STEW, W/O MEAT","Vegetable stew without meat"
+75439500,"CHOW MEIN OR CHOP SUEY, MEATLESS, NO NOODLES","Chow mein or chop suey, meatless, no noodles"
+75440100,"VEG COMBINATION (INCL CAR/ DK GRN), W/ SOY-BASE SCE","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, with soy-based sauce"
+75440110,"VEG COMBINATION (NO CAR/ DK GRN), W/ SOY-BASE SAUCE","Vegetable combination (excluding carrots, broccoli, and dark-green leafy), cooked, with soy-based sauce"
+75440170,"VEGETABLE STICKS, BREADED(INCL CORN,CARROT,GR BEAN)","Vegetable sticks, breaded (including corn, carrots, and green beans)"
+75440200,"VEGETABLE TEMPURA","Vegetable tempura"
+75440300,"VEG COMBINATIONS (INCL CAR/DK GRN), W/ TOMATO SAUCE","Vegetable combinations (including carrots, broccoli, and/or dark-green leafy), cooked, with tomato sauce"
+75440310,"VEG COMBINATIONS (NO CAR/DK GRN), W/ TOMATO SAUCE","Vegetable combinations (excluding carrots, broccoli, and dark-green leafy), cooked, with tomato sauce"
+75440400,"VEGETABLE,IN CHICK-PEA FLOUR BATTER,(PAKORA),FRIED","Vegetables, dipped in chick-pea flour batter, (pakora), fried"
+75440500,"VEG COMBINATION (INCL CAR/DK GRN), W/ CHEESE SAUCE","Vegetable combinations (including carrots, broccoli, and/or dark-green leafy), cooked, with cheese sauce"
+75440510,"VEG COMBINATION (NO CAR/ DK GRN), W/ CHEESE SAUCE","Vegetable combinations (excluding carrots, broccoli, and dark-green leafy), cooked, with cheese sauce"
+75440600,"VEGETABLE CURRY","Vegetable curry"
+75450500,"VEG COMBINATION (INCL CAR, DK GRN), W/ CREAM SAUCE","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, with cream sauce"
+75450510,"VEG COMBINATION (NO CAR, DK GRN), W/ CREAM SAUCE","Vegetable combination (excluding carrots, broccoli, and dark-green leafy), cooked, with cream sauce"
+75450600,"VEG COMBINATION(INCL CAR,BROC,DK GRN)W/BUTTER SAUCE","Vegetable combination (including carrots, broccoli, and/or dark-green leafy), cooked, with butter sauce"
+75460700,"VEGETABLE COMBINATION (INCL CAR/DK GRN), W/ PASTA","Vegetable combinations (including carrots, broccoli, and/or dark-green leafy), cooked, with pasta"
+75460710,"VEGETABLE COMBINATION (NO CAR/DK GRN), W/ PASTA","Vegetable combinations (excluding carrots, broccoli, and dark-green leafy), cooked, with pasta"
+75460800,"VEGETABLE COMB(INCL CAR/DK GRN),CKD,W/ BUTTER SAUCE","Vegetable combinations (including carrots, broccoli, and/or dark-green leafy), cooked, with butter sauce and pasta"
+75460810,"VEGETABLE COMB (NO CAR/DK GRN),CKD, W/ BUTTER SAUCE","Vegetable combinations (excluding carrots, broccoli, and dark-green leafy), cooked, with butter sauce and pasta"
+75460900,"CHOW MEIN OR CHOP SUEY, MEATLESS, WITH NOODLES","Chow mein or chop suey, meatless, with noodles"
+75500110,"BEANS, STRING, GREEN, PICKLED","Beans, string, green, pickled"
+75500210,"BEETS, PICKLED (INCLUDE W/ ONIONS, BEET SALAD)","Beets, pickled"
+75500510,"CELERY, PICKLED","Celery, pickled"
+75501010,"CORN RELISH","Corn relish"
+75502010,"CAULIFLOWER, PICKLED","Cauliflower, pickled"
+75502500,"CABBAGE, FRESH, PICKLED, JAPANESE","Cabbage, fresh, pickled, Japanese style"
+75502510,"CABBAGE, RED, PICKLED (INCL SWEET & SOUR CABBAGE)","Cabbage, red, pickled"
+75502520,"CABBAGE, KIMCHI (KIM CHEE) STYLE","Cabbage, Kimchi (Kim Chee) style"
+75502550,"CABBAGE, MUSTARD, SALTED","Cabbage, mustard, salted"
+75503010,"CUCUMBER PICKLES, DILL","Cucumber pickles, dill"
+75503020,"CUCUMBER PICKLES, RELISH","Cucumber pickles, relish"
+75503030,"CUCUMBER PICKLES, SOUR","Cucumber pickles, sour"
+75503040,"CUCUMBER PICKLES, SWEET","Cucumber pickles, sweet"
+75503080,"EGGPLANT, PICKLED","Eggplant, pickled"
+75503090,"HORSERADISH","Horseradish"
+75503100,"MUSTARD PICKLES (INCL CHOW-CHOW, HOT DOG RELISH)","Mustard pickles"
+75503110,"CUCUMBER PICKLE, DILL, REDUCED SALT","Cucumber pickles, dill, reduced salt"
+75503140,"CUCUMBER PICKLE, SWEET, REDUCED SALT","Cucumber pickles, sweet, reduced salt"
+75505000,"MUSHROOMS, PICKLED","Mushrooms, pickled"
+75506010,"MUSTARD (INCL HORSERADISH MUSTARD, CHINESE MUSTARD)","Mustard"
+75506100,"MUSTARD SAUCE","Mustard sauce"
+75507000,"OKRA, PICKLED","Okra, pickled"
+75510000,"OLIVES, NFS","Olives, NFS"
+75510010,"OLIVES, GREEN","Olives, green"
+75510020,"OLIVES, BLACK","Olives, black"
+75510030,"OLIVES, GREEN, STUFFED","Olives, green, stuffed"
+75511010,"HOT PEPPER SAUCE","Hot pepper sauce"
+75511020,"PEPPERS, PICKLED","Peppers, pickled"
+75511040,"PEPPER, HOT, PICKLED","Pepper, hot, pickled"
+75511100,"PICKLES, NS AS TO VEGETABLE","Pickles, NS as to vegetable"
+75511200,"PICKLES, MIXED","Pickles, mixed"
+75512010,"RADISHES, PICKLED, HAWAIIAN","Radishes, pickled, Hawaiian style"
+75512510,"RECAITO (P.R. LITTLE CORIANDER)","Recaito (Puerto Rican little coriander)"
+75513010,"SEAWEED, PICKLED","Seaweed, pickled"
+75515000,"VEGETABLES, PICKLED, HAWAIIAN","Vegetables, pickled, Hawaiian style"
+75515010,"VEGETABLE RELISH","Vegetable relish"
+75515100,"VEGETABLES, PICKLED (INCLUDE GIARDINIERA)","Vegetables, pickled"
+75534030,"TURNIP, PICKLED","Turnip, pickled"
+75534500,"TSUKEMONO, JAPANESE PICKLES","Tsukemono, Japanese pickles"
+75535000,"ZUCCHINI, PICKLED","Zucchini, pickled"
+75600150,"SOUP, CREAM OF, NFS","Soup, cream of, NFS"
+75601000,"ASPARAGUS SOUP, CREAM OF, NS AS TO W/ MILK OR WATER","Asparagus soup, cream of, NS as to made with milk or water"
+75601010,"ASPARAGUS SOUP, CREAM OF,W/ MILK","Asparagus soup, cream of, prepared with milk"
+75601020,"ASPARAGUS SOUP, CREAM OF, PREPARED W/ WATER","Asparagus soup, cream of, prepared with water"
+75601100,"BEET SOUP (BORSCHT)","Beet soup (borscht)"
+75601200,"CABBAGE SOUP, HOME RECIPE, CANNED OR READY-TO-SERVE","Cabbage soup, home recipe, canned or ready-to-serve"
+75601210,"CABBAGE WITH MEAT SOUP, HOME RECIPE, CANNED OR READY-TO-SERV","Cabbage with meat soup, home recipe, canned or ready-to-serve"
+75603010,"CELERY SOUP, CREAM OF, PREPARED WITH MILK, HOME RECIPE, CANN","Celery soup, cream of, prepared with milk, home recipe, canned or ready-to-serve"
+75603020,"CELERY SOUP, CREAM OF, PREPARED WITH WATER, HOME RECIPE, CAN","Celery soup, cream of, prepared with water, home recipe, canned or ready-to-serve"
+75604010,"CORN SOUP, CREAM OF, W/ MILK","Corn soup, cream of, prepared with milk"
+75604020,"CORN SOUP, CREAM OF, PREPARED W/ WATER","Corn soup, cream of, prepared with water"
+75604600,"GAZPACHO","Gazpacho"
+75605010,"LEEK SOUP, CREAM OF, PREP W/ MILK","Leek soup, cream of, prepared with milk"
+75607000,"MUSHROOM SOUP, NFS","Mushroom soup, NFS"
+75607010,"MUSHROOM SOUP, CREAM OF, PREP W/ MILK","Mushroom soup, cream of, prepared with milk"
+75607020,"MUSHROOM SOUP, CREAM OF, PREPARED W/ WATER","Mushroom soup, cream of, prepared with water"
+75607040,"MUSHROOM SOUP, W/ MEAT BROTH, PREPARED W/ WATER","Mushroom soup, with meat broth, prepared with water"
+75607050,"MUSHROOM SOUP, CM OF, LOW SOD, PREP W/ WATER","Mushroom soup, cream of, low sodium, prepared with water"
+75607060,"MUSHROOM SOUP, CREAM OF, NS AS TO W/ MILK OR WATER","Mushroom soup, cream of, NS as to made with milk or water"
+75607080,"MUSHROOM W/ CHICKEN SOUP, CREAM OF, PREP W/ MILK","Mushroom with chicken soup, cream of, prepared with milk"
+75607090,"MUSHROOM SOUP, CREAM OF, CAN, RED. SOD., NS  W/ MILK/WATER","Mushroom soup, cream of, canned, reduced sodium, NS as to made with milk or water"
+75607100,"MUSHROOM SOUP, CREAM OF, CAN, RED. SODIUM, PREP W/ MILK","Mushroom soup, cream of, canned, reduced sodium, prepared with milk"
+75607130,"MUSHROOM SOUP, MADE FROM DRY MIX","Mushroom soup, made from dry mix"
+75607140,"MUSHROOM SOUP, CM OF, CAN, RED SOD, PREP W/ WATER","Mushroom soup, cream of, canned, reduced sodium, prepared with water"
+75608010,"ONION SOUP, CREAM OF, PREP W/ MILK","Onion soup, cream of, prepared with milk"
+75608100,"ONION SOUP, FRENCH","Onion soup, French"
+75608200,"ONION SOUP, MADE FROM DRY MIX","Onion soup, made from dry mix"
+75609010,"PEA SOUP, PREPARED WITH MILK","Pea soup, prepared with milk"
+75611010,"VEGETABLE SOUP, CREAM OF, PREP W/ MILK","Vegetable soup, cream of, prepared with milk"
+75612010,"ZUCCHINI SOUP, CREAM OF, PREP W/ MILK","Zucchini soup, cream of, prepared with milk"
+75646010,"SHAV SOUP","Shav soup"
+75647000,"SEAWEED SOUP","Seaweed soup"
+75649010,"VEGETABLE SOUP, CANNED, PREPARED WITH WATER OR READY-TO-SERV","Vegetable soup, canned, prepared with water or ready-to-serve"
+75649050,"VEGETABLE SOUP, MADE FROM DRY MIX","Vegetable soup, made from dry mix"
+75649070,"VEGETABLE SOUP, FROM DRY MIX, LOW SODIUM","Vegetable soup, made from dry mix, low sodium"
+75649110,"VEGETABLE SOUP, HOME RECIPE","Vegetable soup, home recipe"
+75649150,"VEGETABLE NOODLE SOUP, HOME RECIPE","Vegetable noodle soup, home recipe"
+75650990,"MINESTRONE SOUP, REDUCED SODIUM, CANNED OR READY-TO-SERVE","Minestrone soup, reduced sodium, canned or ready-to-serve"
+75651000,"MINESTRONE SOUP, HOME RECIPE","Minestrone soup, home recipe"
+75651010,"MINESTRONE SOUP, CANNED, PREPARED WITH WATER, OR READY-TO-SE","Minestrone soup, canned, prepared with water, or ready-to-serve"
+75651020,"VEGETABLE BEEF SOUP, CANNED, PREPARED WITH WATER, OR READY-T","Vegetable beef soup, canned, prepared with water, or ready-to-serve"
+75651030,"VEGETABLE BEEF NOODLE SOUP, PREPARED W/ WATER","Vegetable beef noodle soup, prepared with water"
+75651040,"VEGETABLE NOODLE SOUP, CANNED, PREPARED WITH WATER, OR READY","Vegetable noodle soup, canned, prepared with water, or ready-to-serve"
+75651070,"VEGETABLE RICE SOUP, CANNED, PREPARED WITH WATER OR READY-TO","Vegetable rice soup, canned, prepared with water or ready-to-serve"
+75651080,"VEGETABLE BEEF SOUP WITH RICE, CANNED, PREPARED WITH WATER O","Vegetable beef soup with rice, canned, prepared with water or ready-to-serve"
+75651110,"VEGETABLE CHICKEN RICE SOUP, CANNED, PREP W/WATER OR RTF","Vegetable chicken rice soup, canned, prepared with water or ready-to-serve"
+75651140,"VEGETABLE SOUP WITH CHICKEN BROTH, MEXICAN STYLE, HOME RECIP","Vegetable soup with chicken broth, Mexican style, home recipe (Sopa Ranchera)"
+75651150,"VEGETABLE NOODLE SOUP, RED SODIUM, CAN, PREP W/ WATER/RTS","Vegetable noodle soup, reduced sodium, canned, prepared with water or ready-to-serve"
+75652010,"VEGETABLE BEEF SOUP, HOME RECIPE","Vegetable beef soup, home recipe"
+75652030,"VEGETABLE BEEF SOUP, CANNED, PREPARED WITH MILK","Vegetable beef soup, canned, prepared with milk"
+75652040,"VEG BEEF SOUP W/ NOODLES, HOME RECIPE","Vegetable beef soup with noodles or pasta, home recipe"
+75652050,"VEG BEEF SOUP W/ RICE, HOME RECIPE","Vegetable beef soup with rice, home recipe"
+75654010,"VEGETARIAN VEGETABLE SOUP, PREPARED W/ WATER","Vegetarian vegetable soup, prepared with water"
+75656010,"VEGETABLE SOUP, SPANISH, STEW TYPE","Vegetable soup, Spanish style, stew type"
+75656020,"VEGETABLE SOUP, CHUNKY STYLE","Vegetable soup, chunky style"
+75656040,"VEGETABLE SOUP W/ PASTA, CHUNKY STYLE","Vegetable soup, with pasta, chunky style"
+75656060,"VEG BEEF SOUP, CHUNKY STYLE (INCL VEG W/ MEAT SOUPS","Vegetable beef soup, chunky style"
+75657000,"VEGETABLE BROTH, BOUILLON (INCL POT LIQUOR)","Vegetable broth, bouillon"
+76102010,"SPINACH, CREAMED, BABY, STRAINED","Spinach, creamed, baby food, strained"
+76102030,"BROCCOLI, CARROTS & CHEESE, BABY, JUNIOR","Broccoli, carrots and cheese, baby food, junior"
+76201000,"CARROTS, BABY, NS AS TO STRAINED OR JUNIOR","Carrots, baby food, NS as to strained or junior"
+76201010,"CARROTS, BABY, STRAINED","Carrots, baby food, strained"
+76201020,"CARROTS, BABY, JUNIOR","Carrots, baby food, junior"
+76201030,"CARROTS, BABY FOOD, TODDLER","Carrots, baby food, toddler"
+76202000,"CARROTS & PEAS, BABY, STRAINED","Carrots and peas, baby food, strained"
+76205000,"SQUASH, BABY, NS AS TO STRAINED OR JUNIOR","Squash, baby food, NS as to strained or junior"
+76205010,"SQUASH, BABY, STRAINED","Squash, baby food, strained"
+76205020,"SQUASH, BABY, JUNIOR","Squash, baby food, junior"
+76205030,"SQUASH & CORN, BABY, STRAINED","Squash and corn, baby food, strained"
+76205060,"CORN AND SWEET POTATOES, BABY FOOD, STRAINED","Corn and sweet potatoes, baby food, strained"
+76209000,"SWEET POTATOES, BABY, NS AS TO STRAINED OR JUNIOR","Sweet potatoes, baby food, NS as to strained or junior"
+76209010,"SWEET POTATOES, BABY, STRAINED","Sweet potatoes, baby food, strained"
+76209020,"SWEET POTATOES, BABY, JUNIOR","Sweet potatoes, baby food, junior"
+76401000,"BEANS, GREEN STRING, BABY, NS AS TO STR OR JR","Beans, green string, baby food, NS as to strained or junior"
+76401010,"BEANS, GREEN STRING, BABY, STRAINED","Beans, green string, baby food, strained"
+76401020,"BEANS, GREEN STRING, BABY, JUNIOR","Beans, green string, baby food, junior"
+76401060,"BEANS, GREEN STRING, BABY, TODDLER","Beans, green string, baby food, toddler"
+76402000,"GREEN BEANS & POTATOES, BABY, STRAINED","Green beans and potatoes, baby food, strained"
+76403010,"BEETS, BABY, STRAINED","Beets, baby food, strained"
+76405000,"CORN, CREAMED, BABY, NS AS TO STRAINED OR JUNIOR","Corn, creamed, baby food, NS as to strained or junior"
+76405010,"CORN, CREAMED, BABY, STRAINED","Corn, creamed, baby food, strained"
+76405020,"CORN, CREAMED, BABY, JUNIOR","Corn, creamed, baby food, junior"
+76407000,"MIXED VEG, GARDEN VEG, BABY, NS AS TO STR OR JR","Mixed vegetables, garden vegetables, baby food, NS as to strained or junior"
+76407010,"MIXED VEGETABLES, GARDEN VEGETABLES, BABY, STRAINED","Mixed vegetables, garden vegetables, baby food, strained"
+76407020,"MIXED VEGETABLES, GARDEN VEGETABLES, BABY, JUNIOR","Mixed vegetables, garden vegetables, baby food, junior"
+76409000,"PEAS, BABY, NS AS TO STRAINED OR JUNIOR","Peas, baby food, NS as to strained or junior"
+76409010,"PEAS, BABY, STRAINED","Peas, baby food, strained"
+76409020,"PEAS, BABY, JUNIOR","Peas, baby food, junior"
+76409030,"PEAS, BABY, TODDLER","Peas, baby food, toddler"
+76420000,"POTATOES, BABY, TODDLER","Potatoes, baby food, toddler"
+76501000,"VEGETABLES & RICE, BABY, STRAINED","Vegetables and rice, baby food, strained"
+76502000,"PEAS & BROWN RICE, BABY","Peas and brown rice, baby food"
+76601010,"VEGETABLE & BACON, BABY, STRAINED","Vegetable and bacon, baby food, strained"
+76602000,"CARROTS & BEEF, BABY, STRAINED","Carrots and beef, baby food, strained"
+76603000,"VEGETABLE & BEEF, BABY, NS AS TO STRAINED OR JUNIOR","Vegetable and beef, baby food, NS as to strained or junior"
+76603010,"VEGETABLE & BEEF, BABY, STRAINED","Vegetable and beef, baby food, strained"
+76603020,"VEGETABLE & BEEF, BABY, JUNIOR","Vegetable and beef, baby food, junior"
+76604000,"BROCCOLI & CHICKEN, BABY, STRAINED","Broccoli and chicken, baby food, strained"
+76604500,"SWEET POTATOES & CHICKEN, BABY, STRAINED","Sweet potatoes and chicken, baby food, strained"
+76605000,"VEGETABLE & CHICKEN, BABY, NS AS TO STR OR JR","Vegetable and chicken, baby food, NS as to strained or junior"
+76605010,"VEGETABLE & CHICKEN, BABY, STRAINED","Vegetable and chicken, baby food, strained"
+76605020,"VEGETABLE & CHICKEN, BABY, JUNIOR","Vegetable and chicken, baby food, junior"
+76607000,"VEGETABLE & HAM, BABY, NS AS TO STRAINED OR JUNIOR","Vegetable and ham, baby food, NS as to strained or junior"
+76607010,"VEGETABLE & HAM, BABY, STRAINED","Vegetable and ham, baby food, strained"
+76607020,"VEGETABLE & HAM, BABY, JUNIOR","Vegetable and ham, baby food, junior"
+76607030,"POTATOES W/ CHEESE & HAM, BABY FOOD, TODDLER","Potatoes with cheese and ham, baby food, toddler"
+76607100,"POTATOES WITH CHEESE AND BROCCOLI, BABY FOOD, TODDLER","Potatoes with cheese and broccoli, baby food, toddler"
+76609010,"VEGETABLE & LAMB, BABY, STRAINED","Vegetable and lamb, baby food, strained"
+76611000,"VEGETABLE & TURKEY, BABY, NS AS TO STR OR JR","Vegetable and turkey, baby food, NS as to strained or junior"
+76611010,"VEGETABLE & TURKEY, BABY, STRAINED","Vegetable and turkey, baby food, strained"
+76611020,"VEGETABLE & TURKEY, BABY, JUNIOR","Vegetable and turkey, baby food, junior"
+77121010,"FRIED STUFFED POTATOES, P.R. (RELLENOS DE PAPAS)","Fried stuffed potatoes, Puerto Rican style (Rellenos de papas)"
+77121110,"POTATO&HAM FRITTERS,P.R.(FRITURAS DE PAPA Y JAMON)","Potato and ham fritters, Puerto Rican style (Frituras de papa y jamon)"
+77141010,"POTATO CHICKEN PIE, P.R. (PASTELON DE POLLO)","Potato chicken pie, Puerto Rican style (Pastelon de pollo)"
+77201210,"GREEN PLANTAIN W/ CRACKLINGS, P.R. (MOFONGO)","Green plantain with cracklings, Puerto Rican style (Mofongo)"
+77205110,"RIPE PLANTAIN FRITTERS, P.R. (PIONONO)","Ripe plantain fritters, Puerto Rican style (Pionono)"
+77205610,"RIPE PLANTAIN MEAT PIE, P.R. (PINON)","Ripe plantain meat pie, Puerto Rican style (Pinon)"
+77230210,"CASSAVA PASTELES, P.R. (PASTELES DE YUCA)","Cassava Pasteles, Puerto Rican style (Pasteles de yuca)"
+77230510,"CASSAVA FRITTER STUFFED W/ CRAB, P.R. (EMPANADA DE YUCA)","Cassava fritter stuffed with crab meat, Puerto Rican style (Empanada de yuca y jueyes)"
+77250110,"STUFFED TANNIER FRITTERS, P.R. (ALCAPURRIAS)","Stuffed tannier fritters, Puerto Rican style (Alcapurrias)"
+77250710,"TANNIER FRITTERS, P.R. (FRITURAS DE YAUTIA)","Tannier fritters, Puerto Rican style (Frituras de yautia)"
+77272010,"PUERTO RICAN PASTELES (PASTELES DE MASA)","Puerto Rican pasteles (Pasteles de masa)"
+77316010,"STUFFED CABBAGE, W/ MEAT, P.R.(REPOLLO RELLENO CON CARNE)","Stuffed cabbage, with meat, Puerto Rican style (Repollo relleno con carne)"
+77316510,"STUFFED CABBAGE, W MEAT& RICE, SYRIAN DISH, P.R. STYLE","Stuffed cabbage, with meat and rice, Syrian dish, Puerto Rican style (Repollo relleno con carne y con arroz; Arabe Mihsy Melful)"
+77316600,"EGGPLANT AND MEAT CASSEROLE","Eggplant and meat casserole"
+77513010,"SPANISH STEW, P.R. (COCIDO ESPANOL)","Spanish stew, Puerto Rican style (Cocido Espanol)"
+77563010,"PUERTO RICAN STEW (SALCOCHO / SANCOCHO)","Puerto Rican stew (Salcocho / Sancocho)"
+78101000,"VEGETABLE & FRUIT JUICE BLEND,100% JUICE,W/ HIGH VIT C,+ E,A","Vegetable and fruit juice blend, 100% juice, with high vitamin C plus added vitamin E and vitamin A"
+81100000,"TABLE FAT, NFS","Table fat, NFS"
+81100500,"BUTTER, NFS","Butter, NFS"
+81101000,"BUTTER, STICK, SALTED","Butter, stick, salted"
+81101010,"BUTTER, WHIPPED, TUB, SALTED","Butter, whipped, tub, salted"
+81101020,"BUTTER, WHIPPED, STICK, SALTED","Butter, whipped, stick, salted"
+81101100,"BUTTER, STICK, UNSALTED","Butter, stick, unsalted"
+81101110,"BUTTER, WHIPPED, TUB, UNSALTED","Butter, whipped, tub, unsalted"
+81101120,"BUTTER, WHIPPED, STICK, UNSALTED","Butter, whipped, stick, unsalted"
+81101500,"LIGHT BUTTER, STICK, SALTED","Light butter, stick, salted"
+81101510,"LIGHT BUTTER, STICK, UNSALTED","Light butter, stick, unsalted"
+81101520,"LIGHT BUTTER, WHIPPED, TUB, SALTED","Light butter, whipped, tub, salted"
+81102000,"MARGARINE, NFS","Margarine, NFS"
+81102010,"MARGARINE, STICK, SALTED","Margarine, stick, salted"
+81102020,"MARGARINE, TUB, SALTED","Margarine, tub, salted"
+81103020,"MARGARINE, WHIPPED, TUB, SALTED","Margarine, whipped, tub, salted"
+81103030,"MARGARINE, STICK, UNSALTED","Margarine, stick, unsalted"
+81103040,"MARGARINE-LIKE SPREAD, STICK, SALTED","Margarine-like spread, stick, salted"
+81103041,"MARGARINE-LIKE SPREAD, MADE W/ YOGURT, STICK, SALTED","Margarine-like spread, made with yogurt, stick, salted"
+81103060,"MARGARINE, TUB, UNSALTED","Margarine, tub, unsalted"
+81103070,"MARGARINE, WHIPPED, TUB, UNSALTED","Margarine, whipped, tub, unsalted"
+81103080,"MARGARINE-LIKE SPREAD, TUB, SALTED","Margarine-like spread, tub, salted"
+81103090,"MARGARINE-LIKE SPREAD, LIQUID, SALTED","Margarine-like spread, liquid, salted"
+81103100,"MARGARINE-LIKE SPREAD, STICK, UNSALTED","Margarine-like spread, stick, unsalted"
+81103120,"MARGARINE-LIKE SPREAD, TUB, UNSALTED","Margarine-like spread, tub, unsalted"
+81103130,"MARGARINE-LIKE SPREAD, WHIPPED, TUB, SALTED","Margarine-like spread, whipped, tub, salted"
+81103140,"MARGARINE-LIKE SPREAD, TUB, SWEETENED","Margarine-like spread, tub, sweetened"
+81104010,"MARGARINE-LIKE SPREAD, RED CAL, 40% FAT, TUB, SALTED","Margarine-like spread, reduced calorie, about 40% fat, tub, salted"
+81104011,"MARGARINE-LIKE SPREAD,RED CAL,40% FAT,MADE W/ YOGURT,TUB","Margarine-like spread, reduced calorie, about 40% fat, made with yogurt, tub, salted"
+81104020,"MARGARINE-LIKE SPREAD, RED CAL, 40% FAT, STICK, SALTED","Margarine-like spread, reduced calorie, about 40% fat, stick, salted"
+81104050,"MARGARINE-LIKE SPREAD, RED CAL, 20% FAT, TUB, SALTED","Margarine-like spread, reduced calorie, about 20% fat, tub, salted"
+81104070,"MARGARINE-LIKE SPREAD, RED CAL, 20% FAT, TUB, UNSALTED","Margarine-like spread, reduced calorie, about 20% fat, tub, unsalted"
+81104100,"MARGARINE-LIKE SPREAD, FAT FREE, TUB, SALTED","Margarine-like spread, fat free, tub, salted"
+81104110,"MARGARINE-LIKE SPREAD, FAT FREE, LIQUID, SALTED","Margarine-like spread, fat free, liquid, salted"
+81104500,"VEGETABLE OIL-BUTTER SPREAD, STICK, SALTED","Vegetable oil-butter spread, stick, salted"
+81104510,"VEGETABLE OIL-BUTTER SPREAD, TUB, SALTED","Vegetable oil-butter spread, tub, salted"
+81104550,"VEGETABLE OIL-BUTTER SPREAD, RED CAL, STICK, SALTED","Vegetable oil-butter spread, reduced calorie, stick, salted"
+81104560,"VEGETABLE OIL-BUTTER SPREAD, RED CAL, TUB, SALTED","Vegetable oil-butter spread, reduced calorie, tub, salted"
+81105010,"BUTTER-MARGARINE BLEND, STICK, SALTED","Butter-margarine blend, stick, salted"
+81105020,"BUTTER-MARGARINE BLEND, TUB, SALTED","Butter-margarine blend, tub, salted"
+81105500,"BUTTER-VEG OIL BLEND","Butter-vegetable oil blend"
+81106010,"BUTTER REPLACEMENT, FAT-FREE POWDER, NOT RECONST","Butter replacement, fat-free powder"
+81201000,"ANIMAL FAT OR DRIPPINGS","Animal fat or drippings"
+81202000,"LARD","Lard"
+81203000,"SHORTENING, NS AS TO VEGETABLE OR ANIMAL","Shortening, NS as to vegetable or animal"
+81203100,"SHORTENING, VEGETABLE","Shortening, vegetable"
+81203200,"SHORTENING, ANIMAL","Shortening, animal"
+81204000,"GHEE, CLARIFIED BUTTER","Ghee, clarified butter"
+81301000,"GARLIC SAUCE","Garlic sauce"
+81301020,"LEMON-BUTTER SAUCE","Lemon-butter sauce"
+81302010,"HOLLANDAISE SAUCE","Hollandaise sauce"
+81302040,"SANDWICH SPREAD","Sandwich spread"
+81302050,"TARTAR SAUCE","Tartar sauce"
+81302060,"HORSERADISH SAUCE","Horseradish sauce"
+81302070,"PESTO SAUCE","Pesto sauce"
+81312000,"TARTAR SAUCE, REDUCED FAT/CALORIE","Tartar sauce, reduced fat/calorie"
+81322000,"HONEY BUTTER","Honey butter"
+81330210,"ADOBO FRESCO (INCL ADOBO CRIOLLO)","Adobo fresco"
+82101000,"VEGETABLE OIL, NFS (INCLUDE OIL, NFS)","Vegetable oil, NFS"
+82101300,"ALMOND OIL","Almond oil"
+82101500,"COCONUT OIL","Coconut oil"
+82102000,"CORN OIL","Corn oil"
+82102500,"CORN & CANOLA OIL","Corn and canola oil"
+82103000,"COTTONSEED OIL","Cottonseed oil"
+82103500,"FLAXSEED OIL","Flaxseed oil"
+82104000,"OLIVE OIL","Olive oil"
+82105000,"PEANUT OIL","Peanut oil"
+82105500,"RAPESEED OIL (INCL CANOLA OIL, PURITAN)","Rapeseed oil"
+82105750,"CANOLA & SOYBEAN OIL","Canola and soybean oil"
+82105800,"CANOLA, SOYBEAN & SUNFLOWER OIL","Canola, soybean and sunflower oil"
+82106000,"SAFFLOWER OIL","Safflower oil"
+82107000,"SESAME OIL","Sesame oil"
+82108000,"SOYBEAN OIL","Soybean oil"
+82108250,"SOYBEAN & SUNFLOWER OIL","Soybean and sunflower oil"
+82108500,"SUNFLOWER OIL","Sunflower oil"
+82108700,"WALNUT OIL","Walnut oil"
+82109000,"WHEAT GERM OIL","Wheat germ oil"
+83100100,"SALAD DRESSING, NFS, FOR SALADS","Salad dressing, NFS, for salads"
+83100200,"SALAD DRESSING, NFS, FOR SANDWICHES","Salad dressing, NFS, for sandwiches"
+83101000,"BLUE OR ROQUEFORT CHEESE DRESSING","Blue or roquefort cheese dressing"
+83101500,"BACON DRESSING (HOT)","Bacon dressing (hot)"
+83101600,"BACON & TOMATO DRESSING","Bacon and tomato dressing"
+83102000,"CAESAR DRESSING","Caesar dressing"
+83103000,"COLESLAW DRESSSING","Coleslaw dressing"
+83104000,"FRENCH OR CATALINA DRESSING","French or Catalina dressing"
+83105500,"HONEY MUSTARD DRESSING","Honey mustard dressing"
+83106000,"ITALIAN DRESSING, W/ VINEGAR & OIL","Italian dressing, made with vinegar and oil"
+83107000,"MAYONNAISE, REGULAR","Mayonnaise, regular"
+83108000,"MAYONNAISE, IMITATION","Mayonnaise, imitation"
+83109000,"RUSSIAN DRESSING","Russian dressing"
+83110000,"MAYONNAISE-TYPE SALAD DRESSING","Mayonnaise-type salad dressing"
+83112000,"AVOCADO DRESSING","Avocado dressing"
+83112500,"CREAMY DRESSING","Creamy dressing"
+83112950,"POPPY SEED DRESSING","Poppy seed dressing"
+83112990,"SESAME DRESSING","Sesame dressing"
+83114000,"THOUSAND ISLAND DRESSING","Thousand Island dressing"
+83115000,"YOGURT DRESSING","Yogurt dressing"
+83200100,"SALAD DRESSING, LIGHT, NFS","Salad dressing, light, NFS"
+83201000,"BLUE OR ROQUEFORT CHEESE DRESSING, LIGHT","Blue or roquefort cheese dressing, light"
+83201400,"COLESLAW DRESSING, LIGHT","Coleslaw dressing, light"
+83202020,"FRENCH OR CATALINA DRESSING, LIGHT","French or Catalina dressing, light"
+83203000,"CAESAR DRESSING, LIGHT","Caesar dressing, light"
+83204000,"MAYONNAISE, LIGHT","Mayonnaise, light"
+83204030,"MAYONNAISE, REGULAR,  WITH OLIVE OIL","Mayonnaise, reduced fat,  with olive oil"
+83204050,"MAYONNAISE-TYPE SALAD DRESSING, LIGHT","Mayonnaise-type salad dressing, light"
+83204500,"HONEY MUSTARD DRESSING, LIGHT","Honey mustard dressing, light"
+83205450,"ITALIAN DRESSING, LIGHT","Italian dressing, light"
+83206000,"RUSSIAN DRESSING, LIGHT","Russian dressing, light"
+83206500,"SESAME DRESSING, LIGHT","Sesame dressing, light"
+83207000,"THOUSAND ISLAND DRESSING, LIGHT","Thousand Island dressing, light"
+83208500,"KOREAN DRESSING OR MARINADE","Korean dressing or marinade"
+83210100,"CREAMY DRESSING, LIGHT","Creamy dressing, light"
+83300100,"BLUE OR ROQUEFORT CHEESE DRESSING, FAT FREE","Blue or roquefort cheese dressing, fat free"
+83300200,"CAESAR DRESSING, FAT FREE","Caesar dressing, fat free"
+83300300,"CREAMY DRESSING, FAT FREE","Creamy dressing, fat free"
+83300400,"FRENCH OR CATALINA DRESSING, FAT FREE","French or Catalina dressing, fat free"
+83300500,"HONEY MUSTARD DRESSING, FAT FREE","Honey mustard dressing, fat free"
+83300600,"ITALIAN DRESSING, FAT FREE","Italian dressing, fat free"
+83300700,"MAYONNAISE, FAT FREE","Mayonnaise, fat free"
+83300800,"RUSSIAN DRESSING, FAT FREE","Russian dressing, fat free"
+83300900,"SALAD DRESSING, FAT FREE, NFS","Salad dressing, fat free, NFS"
+83301000,"THOUSAND ISLAND DRESSING, FAT FREE","Thousand Island dressing, fat free"
+91101000,"SUGAR, NFS","Sugar, NFS"
+91101010,"SUGAR, WHITE, GRANULATED OR LUMP","Sugar, white, granulated or lump"
+91101020,"SUGAR, WHITE, CONFECTIONER'S, POWDERED","Sugar, white, confectioner's, powdered"
+91102010,"SUGAR, BROWN","Sugar, brown"
+91103010,"SUGAR, MAPLE","Sugar, maple"
+91104100,"SUGAR, CINNAMON","Sugar, cinnamon"
+91104200,"SUGAR, RAW","Sugar, raw"
+91105010,"FRUCTOSE SWEETENER, SUGAR SUBSTITUTE, DRY POWDER","Fructose sweetener, sugar substitute, dry powder"
+91106000,"SUGAR SUBSTITUTE, SUGAR-ASPARTAME BLEND, DRY PWD","Sugar substitute, sugar-aspartame blend, dry powder"
+91107000,"SUCRALOSE-BASED SWEETENER, SUGAR SUBSTITUTE","Sucralose-based sweetener, sugar substitute"
+91108000,"SUGAR SUB, HERBAL EXTRACT SWEETENER, POWDER","Sugar substitute, herbal extract sweetener, powder"
+91108010,"SUGAR SUB, HERBAL EXTRACT SWEETENER, LIQUID","Sugar substitute, herbal extract sweetener, liquid"
+91109000,"BLUE AGAVE LIQUID SWEETENER, SUGAR SUBSTITUTE","Blue Agave liquid sweetener, sugar substitute"
+91200000,"SUGAR SUBSTITUTE, LOW CALORIE, POWDERED, NFS","Sugar substitute, low-calorie, powdered, NFS"
+91200020,"SUGAR SUBSTITUTE, SACCHARIN-BASED, DRY POWDER","Sugar substitute, saccharin-based, dry powder"
+91200030,"BROWN SUGAR SUBSTITUTE, SACCHARIN-BASED, DRY POWDER","Brown sugar substitute, saccharin-based, dry powder"
+91200040,"SUGAR SUBSTITUTE, SACCHARIN-BASED, DRY POWDER AND TABLETS","Sugar substitute, saccharin-based, dry powder and tablets"
+91200110,"SUGAR SUBSTITUTE, SACCHARIN-BASED, LIQUID","Sugar substitute, saccharin-based, liquid"
+91201010,"SUGAR SUBSTITUTE, ASPARTAME-BASED, DRY POWDER","Sugar substitute, aspartame-based, dry powder"
+91300010,"SYRUP, NFS","Syrup, NFS"
+91300100,"PANCAKE SYRUP, NFS","Pancake syrup, NFS"
+91301020,"CANE & CORN PANCAKE SYRUP","Cane and corn pancake syrup"
+91301030,"CORN SYRUP, LIGHT OR DARK","Corn syrup, light or dark"
+91301040,"BUTTERED BLENDS SYRUP (INCL MRS BUTTERWORTH)","Buttered blends syrup"
+91301050,"FRUIT SYRUP","Fruit syrup"
+91301060,"MAPLE SYRUP(100% MAPLE)(INCLUDE MAPLE CREAM)","Maple syrup (100% maple)"
+91301080,"CHOCOLATE SYRUP, THIN TYPE","Chocolate syrup, thin type"
+91301081,"CHOCOLATE SYRUP, THIN TYPE, LIGHT","Chocolate syrup, thin type, light"
+91301082,"CHOCOLATE SYRUP, THIN TYPE, SUGAR FREE","Chocolate syrup, thin type, sugar free"
+91301090,"SORGHUM SYRUP","Sorghum syrup"
+91301100,"SUGAR (WHITE) & WATER SYRUP (INCLUDE SIMPLE SYRUP)","Sugar (white) and water syrup"
+91301120,"SUGAR, CARMELIZED","Sugar, carmelized"
+91301130,"FRUIT FLAVORED SYRUP USED FOR MILK BEVERAGES","Fruit flavored syrup used for milk beverages"
+91301200,"SUGAR (BROWN) &  WATER SYRUP","Sugar (brown) and water syrup"
+91301250,"MAPLE & CORN  &/OR  CANE PANCAKE SYRUP BLENDS","Maple and corn and/or cane pancake syrup blends (formerly Corn and maple syrup (2% maple))"
+91301510,"SYRUP, PANCAKE, REDUCED CALORIE","Syrup, pancake, reduced calorie"
+91302010,"HONEY (INCLUDE PEAR HONEY, RAW HONEY)","Honey"
+91303000,"MOLASSES","Molasses"
+91303500,"SUGAR, BROWN, LIQUID","Sugar, brown, liquid"
+91303750,"CHOCOLATE GRAVY","Chocolate gravy"
+91304010,"TOPPING, BUTTERSCOTCH OR CARAMEL","Topping, butterscotch or caramel"
+91304020,"TOPPING, CHOCOLATE, THICK, FUDGE TYPE","Topping, chocolate, thick, fudge type"
+91304030,"TOPPING, FRUIT","Topping, fruit"
+91304040,"TOPPING, MARSHMALLOW","Topping, marshmallow"
+91304050,"HARD SAUCE","Hard sauce"
+91304060,"TOPPING, NUT (WET)","Topping, nut (wet)"
+91304070,"TOPPING, PEANUT BUTTER, THICK FUDGE TYPE","Topping, peanut butter, thick, fudge type"
+91304080,"TOPPING, FRUIT, UNSWEETENED","Topping, fruit, unsweetened"
+91304090,"TOPPING, CHOC FLAVOR HAZELNUT SPREAD (INCL NUTELLA)","Topping, chocolate flavored hazelnut spread"
+91304250,"TOPPING, MILK CHOCOLATE W/ CEREAL","Topping, milk chocolate with cereal"
+91304300,"TOPPING, CHOCOLATE, HARD COATING","Topping, chocolate, hard coating"
+91305010,"ICING, CHOCOLATE","Icing, chocolate"
+91305020,"ICING, WHITE","Icing, white"
+91351010,"SYRUP, DIETETIC","Syrup, dietetic"
+91351020,"TOPPING, DIETETIC","Topping, dietetic"
+91361010,"SWEET & SOUR SAUCE (INCLUDE VIETNAMESE SAUCE)","Sweet and sour sauce"
+91361020,"FRUIT SAUCE (INCLUDE ALL FRUITS)","Fruit sauce"
+91361040,"DESSERT SAUCE","Dessert sauce"
+91361050,"DUCK SAUCE (INCLUDE CHAISNI SAUCE)","Duck sauce"
+91361070,"PLUM SAUCE, ASIAN STYLE","Plum sauce, Asian style"
+91401000,"JELLY, ALL FLAVORS","Jelly, all flavors"
+91402000,"JAM, PRESERVES, ALL FLAVORS","Jam, preserves, all flavors"
+91403000,"FRUIT BUTTER, ALL FLAVORS (INCLUDE APPLE BUTTER)","Fruit butter, all flavors"
+91404000,"MARMALADE, ALL FLAVORS","Marmalade, all flavors"
+91405000,"JELLY, DIETETIC, ALL FLAVORS,SWEETENED W/ ARTIFICIAL SWEETEN","Jelly, dietetic, all flavors, sweetened with artificial sweetener"
+91405500,"JELLY, REDUCED SUGAR, ALL FLAVORS","Jelly, reduced sugar, all flavors"
+91406000,"JAM, MARMALADES, ARTIFICIALLY SWEETENED","Jams, preserves, marmalades, dietetic, all flavors, sweetened with artificial sweetener"
+91406500,"JAM PRESERVES,MARMALADES,SWEET W/ FRUIT JUICE CONC","Jams, preserves, marmalades, sweetened with fruit juice concentrates, all flavors"
+91406600,"JAMS,PRESERVES,MARMALADES,LOW SUGAR (ALL FLAVORS)","Jams, preserves, marmalades, low sugar (all flavors)"
+91407100,"GUAVA PASTE","Guava paste"
+91407120,"SWEET POTATO PASTE","Sweet potato paste"
+91407150,"BEAN PASTE, SWEETENED","Bean paste, sweetened"
+91408100,"CHINESE PRESERVED SWEET VEGETABLE","Chinese preserved sweet vegetable"
+91500200,"GELATIN POWDER, SWEETENED, DRY","Gelatin powder, sweetened, dry"
+91501010,"GELATIN DESSERT","Gelatin dessert"
+91501015,"GELATIN SNACKS","Gelatin snacks"
+91501020,"GELATIN DESSERT W/ FRUIT","Gelatin dessert with fruit"
+91501030,"GELATIN DESSERT W/ WHIPPED CREAM","Gelatin dessert with whipped cream"
+91501040,"GELATIN DESSERT W/ FRUIT & WHIPPED CREAM","Gelatin dessert with fruit and whipped cream"
+91501050,"GELATIN DESSERT W/ CREAM CHEESE","Gelatin dessert with cream cheese"
+91501060,"GELATIN DESSERT W/ SOUR CREAM","Gelatin dessert with sour cream"
+91501070,"GELATIN DESSERT W/ FRUIT & SOUR CREAM","Gelatin dessert with fruit and sour cream"
+91501080,"GELATIN DESSERT W/ FRUIT & CREAM CHEESE","Gelatin dessert with fruit and cream cheese"
+91501090,"GELATIN DESSERT W/ FRUIT, VEGETABLES, & NUTS","Gelatin dessert with fruit, vegetable, and nuts"
+91501100,"GELATIN SALAD W/ VEGETABLES","Gelatin salad with vegetables"
+91501110,"GELATIN DESSERT W/ FRUIT & WHIPPED TOPPING","Gelatin dessert with fruit and whipped topping"
+91501120,"GELATIN DESSERT W/ FRUIT & VEGETABLES","Gelatin dessert with fruit and vegetables"
+91510100,"GELATIN POWDER, DIETETIC, DRY","Gelatin powder, dietetic, sweetened with low calorie sweetener, dry"
+91511010,"GELATIN DESSERT, DIETETIC, W/ LO CAL SWEETENER","Gelatin dessert, dietetic, sweetened with low calorie sweetener"
+91511020,"GELATIN DESSERT, DIET, W/ FRUIT, LO CAL SWEETNER","Gelatin dessert, dietetic, with fruit, sweetened with low calorie sweetener"
+91511030,"GELATIN DESSERT, DIETETIC, W/ WHIPPED TOPPING","Gelatin dessert, dietetic, with whipped topping, sweetened with low calorie sweetener"
+91511050,"GELATIN DESSERT, DIETETIC, W/ CREAM CHEESE","Gelatin dessert, dietetic, with cream cheese, sweetened with low calorie sweetener"
+91511060,"GELATIN DESSERT, DIETETIC, W/ SOUR CREAM","Gelatin dessert, dietetic, with sour cream, sweetened with low calorie sweetener"
+91511070,"GELATIN DESSERT, DIETETIC, W/ FRUIT & SOUR CREAM","Gelatin dessert, dietetic, with fruit and sour cream, sweetened with low calorie sweetener"
+91511080,"GELATIN DESSERT, DIETETIC, W/ FRUIT & CREAM CHEESE","Gelatin dessert, dietetic, with fruit and cream cheese, sweetened with low calorie sweetener"
+91511090,"GELATIN DESSERT, DIETETIC, W/ FRUIT & VEGETABLES","Gelatin dessert, dietetic, with fruit and vegetable(s), sweetened with low calorie sweetener"
+91511100,"GELATIN DESSERT, DIETETIC, W/ VEGETABLES","Gelatin salad, dietetic, with vegetables, sweetened with low calorie sweetener"
+91511110,"GELATIN DESSERT, DIETETIC, W/ FRUIT & WHIP TOPPING","Gelatin dessert, dietetic, with fruit and whipped topping, sweetened with low calorie sweetener"
+91512010,"DANISH DESSERT PUDDING","Danish dessert pudding"
+91520100,"YOOKAN, JAPANESE DESSERT MADE W/ BEAN PASTE & SUGAR","Yookan (Yokan), a Japanese dessert made with bean paste and sugar"
+91550100,"COCONUT CREAM CAKE, P.R. (BIEN ME SABE)","Coconut cream cake, Puerto Rican style (Bien me sabe, ""Tastes good to me"")"
+91550300,"PINEAPPLE CUSTARD, P.R. (FLAN DE PINA)","Pineapple custard, Puerto Rican style (Flan de pina)"
+91560100,"HAUPIA (COCONUT PUDDING)","Haupia (coconut pudding)"
+91580000,"GELATIN,FROZ,WHIPPED,ON STICK(INCL JELLO GLTN POPS)","Gelatin, frozen, whipped, on a stick"
+91601000,"ICE, FRUIT","Ice, fruit"
+91611000,"ICE POP","Ice pop"
+91611050,"ICE POP FILLED W/ ICE CREAM, ALL FLAVOR VARIETIES","Ice pop filled with ice cream, all flavor varieties"
+91611100,"ICE POP, SWEETENED W/ LOW CALORIE SWEETENER","Ice pop, sweetened with low calorie sweetener"
+91621000,"SNOW CONE","Snow cone"
+91700010,"CANDY, NFS","Candy, NFS"
+91700500,"M&M'S ALMOND CHOCOLATE CANDIES","M&M's Almond Chocolate Candies"
+91701010,"ALMONDS, CHOCOLATE-COVERED","Almonds, chocolate covered"
+91701020,"ALMONDS, SUGAR-COATED (INCL JORDAN ALMONDS)","Almonds, sugar-coated"
+91701030,"ALMONDS, YOGURT-COVERED","Almonds, yogurt-covered"
+91702010,"BUTTERSCOTCH MORSELS","Butterscotch morsels"
+91703010,"CARAMEL CANDY, CHOC-FLAVOR ROLL (INCL TOOTSIE ROLL)","Caramel, chocolate-flavored roll"
+91703020,"CARAMEL CANDY, NOT CHOCOLATE","Caramel, flavor other than chocolate"
+91703030,"CARAMEL CANDY, W/ NUTS","Caramel, with nuts"
+91703040,"CARAMEL CANDY, CHOCOLATE COVERED","Caramel candy, chocolate covered"
+91703050,"CARAMEL CANDY, W/ NUTS & CEREAL, CHOCOLATE-COVERED","Caramel with nuts and cereal, chocolate covered"
+91703060,"CARAMEL CANDY, W/ NUTS, CHOCOLATE-COVERED","Caramel with nuts, chocolate covered"
+91703070,"ROLOS CANDY","Rolo"
+91703080,"CARAMEL, ALL FLAVORS, SUGAR FREE","Caramel, all flavors, sugar free"
+91703150,"TOBLERONE,MILK CHOCOLATE W/ HONEY & ALMOND NOUGAT","Toblerone, milk chocolate with honey and almond nougat"
+91703200,"TWIX CARAMEL COOKIE BARS","TWIX Caramel Cookie Bars (formerly TWIX Cookie Bars)"
+91703250,"TWIX CHOCOLATE FUDGE COOKIE BARS","TWIX Chocolate Fudge Cookie Bars"
+91703300,"TWIX PEANUT BUTTER COOKIE BARS","TWIX Peanut Butter Cookie Bars"
+91703400,"WHATCHAMACALLIT CANDY","Whatchamacallit"
+91703500,"NUTS, CAROB-COATED","Nuts, carob-coated"
+91703600,"ESPRESSO COFFEE BEANS, CHOCOLATE-COVERED","Espresso coffee beans, chocolate-covered"
+91705010,"MILK CHOCOLATE CANDY, PLAIN","Milk chocolate candy, plain"
+91705020,"MILK CHOCOLATE CANDY, WITH CEREAL","Milk chocolate candy, with cereal"
+91705030,"KIT KAT CANDY BAR","Kit Kat"
+91705040,"CHOCOLATE, MILK, W/ NUTS, NOT ALMONDS OR PEANUTS","Chocolate, milk, with nuts, not almond or peanuts"
+91705050,"MILK CHOCOLATE CANDY, WITH FRUIT AND NUTS","Milk chocolate candy, with fruit and nuts"
+91705060,"MILK CHOCOLATE CANDY, WITH ALMONDS","Milk chocolate candy, with almonds"
+91705070,"CHOCOLATE, MILK, W/ PEANUTS (INCLUDE MR GOODBAR)","Chocolate, milk, with peanuts"
+91705090,"CHOCOLATE CANDY WITH FONDANT AND CARAMEL","Chocolate candy with fondant and caramel"
+91705200,"CHOCOLATE, SEMI-SWEET","Chocolate, semi-sweet morsel"
+91705300,"CHOCOLATE CANDY, SWEET OR DARK","Chocolate, sweet or dark"
+91705310,"CHOCOLATE, SWEET OR DARK, WITH ALMONDS","Chocolate, sweet or dark, with almonds"
+91705400,"CHOCOLATE CANDY, WHITE","Chocolate, white"
+91705410,"CHOCOLATE CANDY, WHITE, W/ ALMONDS","Chocolate, white, with almonds"
+91705420,"CHOCOLATE, WHITE, W/ CEREAL, CANDY","Chocolate, white, with cereal"
+91705430,"KIT KAT WHITE","Kit Kat White"
+91705500,"MEXICAN CHOCOLATE (TABLET)","Mexican chocolate (tablet)"
+91706000,"COCONUT CANDY, CHOCOLATE-COVERED","Coconut candy, chocolate covered"
+91706100,"COCONUT CANDY, NO CHOCOLATE COVERING","Coconut candy, no chocolate covering"
+91706400,"COCONUT CANDY, P.R. STYLE","Coconut candy, Puerto Rican style"
+91707000,"FONDANT CANDY","Fondant"
+91707010,"FONDANT CANDY, CHOCOLATE COVERED","Fondant, chocolate covered"
+91708000,"FRUIT PEEL, CANDIED","Fruit peel, candied"
+91708010,"FRUIT CANDY BAR","Date candy"
+91708020,"SOFT FRUIT CONFECTION","Soft fruit confections"
+91708030,"FRUIT LEATHER / FRUIT SNACKS CANDY","Fruit leather and fruit snacks candy"
+91708040,"FUN FRUITS CREME SUPREMES CANDY","Fun Fruits Creme Supremes"
+91708070,"TAMARIND CANDY","Tamarind candy"
+91708100,"FRUIT SNACKS CANDY W/ HI VIT C","Fruit snacks candy, with high vitamin C"
+91708150,"YOGURT COVERED FRUIT SNACKS CANDY, W/ ADDED VITAMIN C","Yogurt covered fruit snacks candy, with added vitamin C"
+91708160,"YOGURT COVERED FRUIT SNACKS CANDY ROLLS, W/ HIGH VITAMIN C","Yogurt covered fruit snacks candy rolls, with high vitamin C"
+91709000,"GUMDROPS, CHOCOLATE-COVERED","Gumdrops, chocolate covered"
+91713010,"FUDGE, CHOCOLATE, CHOCOLATE-COATED","Fudge, chocolate, chocolate-coated"
+91713020,"FUDGE, CHOCOLATE, CHOCOLATE-COATED, W/ NUTS","Fudge, chocolate, chocolate-coated, with nuts"
+91713030,"FUDGE, CHOCOLATE","Fudge, chocolate"
+91713040,"FUDGE, CHOCOLATE, W/ NUTS","Fudge, chocolate, with nuts"
+91713050,"FUDGE, PEANUT BUTTER","Fudge, peanut butter"
+91713060,"FUDGE, PEANUT BUTTER, W/ NUTS","Fudge, peanut butter, with nuts"
+91713070,"FUDGE, VANILLA","Fudge, vanilla"
+91713080,"FUDGE, VANILLA, W/ NUTS","Fudge, vanilla, with nuts"
+91713090,"FUDGE, DIVINITY","Fudge, divinity"
+91713100,"FUDGE, BROWN SUGAR (PANUCHI)","Fudge, brown sugar (penuche)"
+91715000,"FUDGE, CARAMEL AND NUT, CHOCOLATE-COATED CANDY","Fudge, caramel and nut, chocolate-coated candy"
+91715100,"SNICKERS CANDY BAR","SNICKERS Bar"
+91715200,"BABY RUTH CANDY BAR","Baby Ruth"
+91715300,"100 GRAND BAR (INCL $100,000 BAR)","100 GRAND Bar"
+91716010,"HALVAH, PLAIN","Halvah, plain"
+91716110,"HALVAH, CHOCOLATE-COVERED","Halvah, chocolate covered"
+91718000,"HONEY-COMBED HARD CANDY, PEANUT BUTTER","Honey-combed hard candy with peanut butter"
+91718050,"HONEY-COMBED CANDY, PEANUT BUTTER, CHOC-COVERED","Honey-combed hard candy with peanut butter, chocolate covered"
+91718100,"BUTTERFINGER CANDY BAR","Butterfinger"
+91718110,"BUTTERFINGER CRISP","Butterfinger Crisp"
+91718200,"JIMMIES (INCLUDE CHOCOLATE-FLAVORED SPRINKLES)","Chocolate-flavored sprinkles"
+91718300,"LADOO, ROUND BALL, ASIAN-INDIAN DESSERT","Ladoo, round ball, Asian-Indian dessert"
+91721000,"LICORICE CANDY","Licorice"
+91723000,"MARSHMALLOW","Marshmallow"
+91723010,"MARSHMALLOW, CHOCOLATE-COVERED","Marshmallow, chocolate covered"
+91723020,"MARSHMALLOW, CANDY-COATED","Marshmallow, candy-coated"
+91723050,"MARSHMALLOW, COCONUT-COATED","Marshmallow, coconut-coated"
+91726000,"NOUGAT CANDY, PLAIN","Nougat, plain"
+91726110,"NOUGAT CANDY, W/ CARAMEL, CHOCOLATE-COVERED","Nougat, with caramel, chocolate covered"
+91726130,"MILKY  WAY BAR","MILKY WAY Bar"
+91726140,"MILKY WAY MIDNIGHT BAR (FORMERLY MILKY WAY DARK BAR)","MILKY WAY MIDNIGHT Bar (formerly MILKY WAY DARK Bar)"
+91726150,"MARS ALMOND BAR (FORMERLY MARS BAR)","MARS Almond Bar (formerly MARS bar)"
+91726410,"NOUGAT CANDY, CHOCOLATE-COVERED","Nougat, chocolate covered"
+91726420,"3 MUSKETEERS BAR","3 MUSKETEERS Bar"
+91726425,"3 MUSKETEERS TRUFFLE CRISP BAR","3 Musketeers Truffle Crisp Bar"
+91727010,"NUTS, CHOCOLATE-COVERED, NOT ALMONDS OR PEANUTS","Nuts, chocolate covered, not almonds or peanuts"
+91728000,"NUT ROLL, FUDGE OR NOUGAT, CARAMEL & NUTS","Nut roll, fudge or nougat, caramel and nuts"
+91728500,"SUGARED PECANS (SUGAR & EGG WHITE COATING)","Sugared pecans (sugar and egg white coating)"
+91731000,"PEANUTS, CHOCOLATE-COVERED","Peanuts, chocolate covered"
+91731010,"M&M'S PEANUT CANDIES","M&M's Peanut Chocolate Candies"
+91731060,"M&M'S PEANUT BUTTER CHOCOLATE CANDIES","M&M's Peanut Butter Chocolate Candies"
+91731100,"PEANUTS, SUGAR-COATED","Peanuts, sugar-coated"
+91731150,"PEANUTS, YOGURT-COVERED","Peanuts, yogurt covered"
+91732000,"PEANUT CANDY BAR","Peanut bar"
+91732100,"PLANTERS PEANUT CANDY BAR","Planters Peanut Bar"
+91733000,"PEANUT BRITTLE","Peanut brittle"
+91733200,"PEANUT BAR, CHOCOLATE COVERED CANDY","Peanut Bar, chocolate covered candy"
+91734000,"PEANUT BUTTER CANDY, CHOCOLATE-COVERED","Peanut butter, chocolate covered"
+91734100,"REESE'S PEANUT BUTTER CUPS","Reese's Peanut Butter Cup"
+91734200,"REESE'S PIECES CANDY","Reese's Pieces"
+91734300,"REESE'S STICKS","Reese's Sticks"
+91734400,"REESE'S FAST BREAK","Reese's Fast Break"
+91734450,"REESE'S CRISPY CRUNCHY BAR","Reese's Crispy Crunchy Bar"
+91734500,"PEANUT BUTTER MORSELS CANDY","Peanut butter morsels"
+91735000,"PRALINES","Pralines"
+91736000,"PINEAPPLE CANDY, P.R. STYLE","Pineapple candy, Puerto Rican style"
+91739010,"RAISINS, CHOCOLATE-COVERED","Raisins, chocolate covered"
+91739600,"RAISINS, YOGURT-COVERED","Raisins, yogurt covered"
+91742010,"SESAME CRUNCH CANDY (SAHADI)","Sesame Crunch (Sahadi)"
+91745010,"GUMDROPS","Gumdrops"
+91745020,"HARD CANDY","Hard candy"
+91745040,"BUTTERSCOTCH HARD CANDY","Butterscotch hard candy"
+91745100,"SKITTLES CANDY","Skittles"
+91746010,"SUGAR-COATED CHOCOLATE DISCS CANDY","Sugar-coated chocolate discs"
+91746100,"M&M'S MILK CHOCOLATE CANDIES","M&M's Milk Chocolate Candies (formerly M&M's Plain Chocolate Candies)"
+91746120,"SIXLETS CANDY","Sixlets"
+91746150,"EASTER EGG, CANDY-COATED CHOCOLATE","Easter egg, candy coated chocolate"
+91746200,"M&M'S PRETZEL CHOCOLATE CANDIES","M&M's Pretzel Chocolate Candies"
+91750000,"TAFFY","Taffy"
+91760000,"TOFFEE, PLAIN","Toffee, plain"
+91760100,"TOFFEE, CHOCOLATE COVERED (INCL HEATH BAR, SKOR)","Toffee, chocolate covered"
+91760200,"TOFFEE, CHOCOLATE-COATED, W/ NUTS","Toffee, chocolate-coated, with nuts"
+91760500,"TRUFFLES","Truffles"
+91760700,"WAX CANDY, LIQUID FILLED","Wax candy, liquid filled"
+91770000,"DIETETIC OR  LOW CALORIE CANDY, NFS","Dietetic or low calorie candy, NFS"
+91770010,"DIETETIC OR LOW CALORIE GUMDROPS","Dietetic or low calorie gumdrops"
+91770020,"DIETETIC OR LOW CALORIE HARD CANDY","Dietetic or low calorie hard candy"
+91770030,"DIETETIC OR  LOW CALORIE CANDY, CHOCOLATE-COVERED","Dietetic or low calorie candy, chocolate covered"
+91770050,"MINTS, DIETETIC OR LOW CALORIE","Dietetic or low calorie mints"
+91800100,"CHEWING GUM, NFS","Chewing gum, NFS"
+91801000,"CHEWING GUM, SUGARED","Chewing gum, sugared"
+91802000,"CHEWING GUM, SUGARLESS","Chewing gum, sugarless"
+92100000,"COFFEE, NS AS TO TYPE","Coffee, NS as to type"
+92100500,"COFFEE, REGULAR, NS GROUND/INSTANT","Coffee, regular, NS as to ground or instant"
+92101000,"COFFEE, MADE FROM GROUND, REGULAR","Coffee, made from ground, regular"
+92101500,"COFFEE, BREWED, EQUAL PARTS REG & DECAFFEINATED","Coffee, made from ground, equal parts regular and decaffeinated"
+92101600,"COFFEE, TURKISH","Coffee, Turkish"
+92101610,"COFFEE, ESPRESSO","Coffee, espresso"
+92101630,"COFFEE, ESPRESSO, DECAFFEINATED","Coffee, espresso, decaffeinated"
+92101640,"COFFEE, MEXICAN, REG, UNSWEETENED (NO MILK)","Coffee, Mexican, regular, unsweetened (no milk; not cafe con leche)"
+92101650,"COFFEE, MEXICAN, REG, SWEETENED (NO MILK)","Coffee, Mexican, regular, sweetened (no milk; not cafe con leche)"
+92101660,"COFFEE, MEXICAN, DECAF, UNSWEETENED (NO MILK)","Coffee, Mexican, decaffeinated, unsweetened (no milk; not cafe con leche)"
+92101670,"COFFEE, MEXICAN, DECAF, SWEETENED (NO MILK)","Coffee, Mexican, decaffeinated, sweetened (no milk; not cafe con leche)"
+92101700,"COFFEE, MADE FROM GROUND, REGULAR, FLAVORED","Coffee, made from ground, regular, flavored"
+92101800,"COFFEE, CUBAN","Coffee, Cuban"
+92101900,"COFFEE, LATTE","Coffee, Latte"
+92101910,"COFFEE, LATTE, DECAFFEINATED","Coffee, Latte, decaffeinated"
+92101920,"BLENDED COFFEE BEVERAGE, REGULAR, SWEETENED","Blended coffee beverage, made with regular coffee, milk, and ice, sweetened"
+92101925,"BLENDED COFFEE BEVERAGE,  REG, SWTND, W/ WHP CRM","Blended coffee beverage, made with regular coffee, milk, and ice, sweetened, with whipped cream"
+92101930,"BLENDED COFFEE BEVERAGE, DECAF, SWEETENED","Blended coffee beverage, made with decaffeinated coffee, milk, and ice, sweetened"
+92101935,"BLENDED COFFEE BEVERAGE, DECAF, SWTND, W/ WHP CRM","Blended coffee beverage, made with decaffeinated coffee, milk, and ice, sweetened, with whipped cream"
+92101950,"COFFEE, MOCHA","Coffee, mocha"
+92101960,"COFFEE, MOCHA, MADE W/ SOY MILK","Coffee, mocha, made with soy milk"
+92103000,"COFFEE, MADE FROM POWDERED INSTANT, REGULAR","Coffee, made from powdered instant, regular"
+92104000,"COFFEE, FROM POWDER, 50% LESS CAFFEINE","Coffee, made from powdered instant, 50% less caffeine"
+92105000,"COFFEE, LIQUID CONCENTRATE, NOT RECONSTITUTED","Coffee, liquid concentrate"
+92105010,"COFFEE, MADE FROM LIQUID CONCENTRATE","Coffee, made from liquid concentrate"
+92106000,"COFFEE, ACID NEUTRALIZED, FROM POWDERED INSTANT","Coffee, acid neutralized, from powdered instant"
+92111000,"COFFEE, DECAFFEINATED, NS AS TO GROUND OR INSTANT","Coffee, decaffeinated, NS as to ground or instant"
+92111010,"COFFEE, DECAFFEINATED, MADE FROM GROUND","Coffee, decaffeinated, made from ground"
+92114000,"COFFEE, DECAFFEINATED, MADE FROM POWDERED INSTANT","Coffee, decaffeinated, made from powdered instant"
+92121000,"COFFEE, FROM POWDERED MIX,W/WHITENER&SUGAR, INSTANT","Coffee, made from powdered instant mix, with whitener and sugar, instant"
+92121010,"COFFEE, FROM POWDER, PRESWEETENED, NO WHITENER","Coffee, made from powdered instant mix, presweetened, no whitener"
+92121020,"COFFEE & COCOA (MOCHA), W/ WHITENER, PRESWEETENED","Coffee and cocoa (mocha), made from powdered instant mix, with whitener, presweetened"
+92121030,"COFFEE & COCOA, FROM MIX, W/WHITENER, LOW CAL SWEET","Coffee and cocoa (mocha), made from powdered instant mix, with whitener and low calorie sweetener"
+92121040,"COFFEE, FROM POWDER, W/ WHITENER & LO CAL SWEETENER","Coffee, made from powdered instant mix, with whitener and low calorie sweetener"
+92121050,"COFFEE&COCOA,FROM PWDR,W/WHITE&LOW CAL SWEET,DECAF","Coffee and cocoa (mocha), made from powdered instant mix, with whitener and low calorie sweetener, decaffeinated"
+92130000,"COFFEE, REG, PRESWEETENED W/SUGAR, PRE-LIGHTENED","Coffee, regular, presweetened with sugar, pre-lightened"
+92130001,"COFFEE, DECAFFEINATED, PRESWEETENED W/ SUGAR, PRE-LIGHTENED","Coffee, decaffeinated, presweetened with sugar, pre-lightened"
+92130005,"COFFEE, REGULAR, WITH LOW CALORIE SWEETENER, PRE-LIGHTENED","Coffee, regular, with low-calorie sweetener, pre-lightened"
+92130006,"COFFEE, DECAFFEINATED,W/ LOW CALORIE SWEETENER,PRE-LIGHTENED","Coffee, decaffeinated, with low-calorie sweetener, pre-lightened"
+92130010,"COFFEE, PRE-LIGHTENED, NO SUGAR","Coffee, pre-lightened, no sugar"
+92130020,"COFFEE, PRESWEETENED W/ SUGAR","Coffee, presweetened with sugar"
+92150000,"COFFEE & CHICORY, NS AS TO GROUND OR INSTANT","Coffee and chicory, NS as to ground or instant"
+92151000,"COFFEE & CHICORY, MADE FROM POWDERED INSTANT","Coffee and chicory, made from powdered instant"
+92151100,"COFFEE, DECAFFEINATED, AND CHICORY, FROM INSTANT","Coffee, decaffeinated, and chicory, made from powdered instant"
+92152000,"COFFEE & CHICORY, MADE FROM GROUND","Coffee and chicory, made from ground"
+92153000,"COFFEE, REGULAR, W/ CEREAL (INCLUDE W/ BARLEY)","Coffee, regular, with cereal"
+92153100,"COFFEE, DECAFFEINATED, W/ CEREAL (INCLUDE W/BARLEY)","Coffee, decaffeinated, with cereal"
+92161000,"CAPPUCCINO","Cappuccino"
+92161005,"CAPPUCCINO, SWEETENED","Cappuccino, sweetened"
+92162000,"CAPPUCCINO, DECAFFEINATED","Cappuccino, decaffeinated"
+92162005,"CAPPUCCINO, DECAFFEINATED, SWEETENED","Cappuccino, decaffeinated, sweetened"
+92191000,"COFFEE, DRY POWDER, NS AS TO REG OR DECAF","Coffee, dry instant powder, NS as to regular or decaffeinated"
+92191100,"COFFEE, DRY POWDER, REGULAR","Coffee, dry instant powder, regular"
+92191200,"COFFEE, DRY POWDER, DECAFFEINATED","Coffee, dry instant powder, decaffeinated"
+92191250,"COFFEE, DRY, ACID NEUTRALIZED (INCLUDE KAVA)","Coffee, dry, acid neutralized"
+92191500,"COFFEE & CHICORY, DRY POWDER","Coffee and chicory, dry instant powder"
+92191520,"COFFEE, DECAFFEINATED, AND CHICORY, DRY POWDER","Coffee, decaffeinated, and chicory, dry instant powder"
+92192000,"COFFEE & COCOA (MOCHA) MIX,W/WHITENER,PRESWEET, DRY","Coffee and cocoa (mocha) mix, dry instant powder with whitener, presweetened"
+92192030,"COFFEE & COCOA (MOCHA) MIX,W/WHITENER, LOW CAL, DRY","Coffee and cocoa (mocha) mix, dry instant powder with whitener and low calorie sweetener"
+92192040,"COFFEE&COCOA MIX,DRY,W/WHITENER&LOW CAL SWEET,DECAF","Coffee and cocoa (mocha) mix, dry instant powder, with whitener and low calorie sweetener, decaffeinated"
+92193000,"COFFEE, DRY MIX, W/ WHITENER & SUGAR","Coffee, dry instant powder, with whitener and sugar"
+92193020,"COFFEE, DRY MIX, W/ WHITENER & LOW CAL SWEETENER","Coffee, dry instant powder, with whitener and low calorie sweetener"
+92201010,"POSTUM (COFFEE SUBSTITUTE)","Postum"
+92202010,"CHICORY (COFFEE SUBSTITUTE)","Chicory"
+92203000,"CEREAL, BEVERAGE (INCLUDE PERO, BREAK AWAY)","Cereal beverage"
+92203110,"CEREAL BEVERAGE, W/BEET ROOTS,FROM POWDERED INSTANT","Cereal beverage with beet roots, from powdered instant"
+92204000,"MATE, SWEETENED BEVERAGE FROM DRIED GREEN LEAVES","Mate, sweetened beverage made from dried green leaves"
+92205000,"RICE BEVERAGE (INCL RICE TEA)","Rice beverage"
+92291300,"POSTUM, DRY POWDER","Postum, dry powder"
+92301000,"TEA, NS AS TO TYPE, UNSWEETENED","Tea, NS as to type, unsweetened"
+92301060,"TEA, NS AS TO TYPE, PRESWEETENED W/ SUGAR","Tea, NS as to type, presweetened with sugar"
+92301080,"TEA, PRESWEETENED W/ LOW CALORIE SWEETENER","Tea, NS as to type, presweetened with low calorie sweetener"
+92301100,"TEA, NS AS TO TYPE, DECAFFEINATED, UNSWEETENED","Tea, NS as to type, decaffeinated, unsweetened"
+92301130,"TEA, NS AS TO TYPE, PRESWEETENED, NS AS TO SWEETNER","Tea, NS as to type, presweetened, NS as to sweetener"
+92301160,"TEA, DECAFFEINATED, W/ SUGAR, NFS","Tea, NS as to type, decaffeinated, presweetened with sugar"
+92301180,"TEA, DECAFFEINATED, LOW CALORIE SWEETENER, NFS","Tea, NS as to type, decaffeinated, presweetened with low calorie sweetener"
+92301190,"TEA, PRESWEETENED, NS SWEETENER, DECAFFEINATED","Tea, NS as to type, decaffeinated, presweetened, NS as to sweetener"
+92302000,"TEA, LEAF, UNSWEETENED","Tea, leaf, unsweetened"
+92302200,"TEA, LEAF, PRESWEETENED W/ SUGAR","Tea, leaf, presweetened with sugar"
+92302300,"TEA, LEAF, PRESWEETENED W/ LOW CALORIE SWEETENER","Tea, leaf, presweetened with low calorie sweetener"
+92302400,"TEA, LEAF, PRESWEETENED, NS AS TO SWEETENER","Tea, leaf, presweetened, NS as to sweetener"
+92302500,"TEA, DECAFFEINATED, UNSWEETENED","Tea, leaf, decaffeinated, unsweetened"
+92302600,"TEA, LEAF, DECAFFEINATED, PRESWEETENED W/ SUGAR","Tea, leaf, decaffeinated, presweetened with sugar"
+92302700,"TEA, LEAF, DECAFFEINATED, LOW CALORIE SWEETENER","Tea, leaf, decaffeinated, presweetened with low calorie sweetener"
+92302800,"TEA, LEAF, DECAFFEINATED, PRESWEETENED, NFS","Tea, leaf, decaffeinated, presweetened, NS as to sweetener"
+92304000,"TEA, MADE FROM FROZEN CONCENTRATE, UNSWEETENED","Tea, made from frozen concentrate, unsweetened"
+92304700,"TEA, FROM FROZ CONC, DECAF, PRESWEETND, LOW CALORIE","Tea, made from frozen concentrate, decaffeinated, presweetened with low calorie sweetener"
+92305000,"TEA, MADE FROM POWDERED INSTANT, PRESWEETENED","Tea, made from powdered instant, presweetened, NS as to sweetener"
+92305010,"TEA, MADE FROM POWDERED INSTANT, UNSWEETENED","Tea, made from powdered instant, unsweetened"
+92305040,"TEA, MADE FROM POWDERED INSTANT,PRESWEETEND W/SUGAR","Tea, made from powdered instant, presweetened with sugar"
+92305050,"TEA, FROM POWDER, DECAFFEINATED, PRESWEET W/ SUGAR","Tea, made from powdered instant, decaffeinated, presweetened with sugar"
+92305090,"TEA, MADE FROM POWDERED INSTANT,W/LO CAL SWEETENER","Tea, made from powdered instant, presweetened with low calorie sweetener"
+92305110,"TEA, FROM INSTANT, DECAF, PRESWEETENED, LOW CALORIE","Tea, made from powdered instant, decaffeinated, presweetened with low calorie sweetener"
+92305180,"TEA ,MADE FROM POWDERED INSTANT, DECAF ,UNSWEET","Tea, made from powdered instant, decaffeinated, unsweetened"
+92305800,"TEA, FROM POWDER, DECAFFEINATED, PRESWEETENED","Tea, made from powdered instant, decaffeinated, presweetened, NS as to sweetener"
+92306000,"TEA, HERBAL (INCLUDE SASSAFRAS,LICORICE)","Tea, herbal"
+92306020,"TEA, HERBAL, PRESWEETENED W/ SUGAR","Tea, herbal, presweetened with sugar"
+92306030,"TEA, HERBAL, PRESWEETENED W/ LOW CAL SWEETENER","Tea, herbal, presweetened with low calorie sweetener"
+92306040,"TEA, HERBAL, PRESWEETENED, NS AS TO SWEETENER","Tea, herbal, presweetened, NS as to sweetener"
+92306050,"TEA, MADE FROM CARAWAY SEEDS","Tea, made from caraway seeds"
+92306090,"TEA, HIBISCUS","Tea, hibiscus"
+92306100,"CORN BEVERAGE(INCLUDE CORN TEA)","Corn beverage"
+92306200,"BEAN BEVERAGE (INCLUDE BEAN TEA)","Bean beverage"
+92306610,"TEA, RUSSIAN","Tea, Russian"
+92306700,"TEA, CHAMOMILE","Tea, chamomile"
+92307000,"TEA, POWDERED INSTANT, UNSWEETENED, DRY","Tea, powdered instant, unsweetened, dry"
+92307400,"TEA, POWDERED INSTANT, SWEETENED, NS SWEETENER, DRY","Tea, powdered instant, sweetened, NS as to sweetener, dry"
+92307500,"HALF AND HALF BEVERAGE, HALF ICED TEA/HALF LEMONADE","Half and Half beverage, half iced tea and half fruit juice drink (lemonade)"
+92307510,"HALF&HALF BEV, HALF ICED TEA/HALF LEMONADE,LOW CAL","Half and Half beverage, half iced tea and half fruit juice drink (lemonade), low calorie"
+92400000,"SOFT DRINK, NFS","Soft drink, NFS"
+92400100,"SOFT DRINK, NFS, SUGAR-FREE","Soft drink, NFS, sugar-free"
+92410110,"CARBONATED WATER,SWEETEND(INCL TONIC,QUININE WATER)","Carbonated water, sweetened"
+92410210,"CARBONATED WATER, UNSWEETENED (INCL CLUB SODA)","Carbonated water, unsweetened"
+92410250,"CARBONATED WATER, SWEETENED, WITH LOW-CALORIE OR NO-CALORIE","Carbonated water, sweetened, with low-calorie or no-calorie sweetener"
+92410310,"SOFT DRINK, COLA-TYPE","Soft drink, cola-type"
+92410315,"SOFT DRINK, COLA TYPE, REDUCED SUGAR","Soft drink, cola type, reduced sugar"
+92410320,"SOFT DRINK, COLA-TYPE, SUGAR-FREE","Soft drink, cola-type, sugar-free"
+92410330,"SOFT DRINK, COLA-TYPE, W/ HIGHER CAFFEINE (INCL JOLT)","Soft drink, cola-type, with higher caffeine"
+92410340,"SOFT DRINK, COLA-TYPE, DECAFFEINATED","Soft drink, cola-type, decaffeinated"
+92410350,"SOFT DRINK, COLA-TYPE, DECAFFEINATED, SUGAR-FREE","Soft drink, cola-type, decaffeinated, sugar-free"
+92410360,"SOFT DRINK, PEPPER-TYPE (INCL DR. PEPPER, MR. PIBB)","Soft drink, pepper-type"
+92410370,"SOFT DRINK, PEPPER-TYPE, SUGAR-FREE","Soft drink, pepper-type, sugar-free"
+92410390,"SOFT DRINK, PEPPER-TYPE, DECAFFEINATED","Soft drink, pepper-type, decaffeinated"
+92410400,"SOFT DRINK, PEPPER-TYPE, DECAFFEINATED, SUGAR-FREE","Soft drink, pepper-type, decaffeinated, sugar-free"
+92410410,"CREAM SODA","Cream soda"
+92410420,"CREAM SODA, SUGAR-FREE","Cream soda, sugar-free"
+92410510,"SOFT DRINK, FRUIT-FLAVORED, CAFFEINE FREE","Soft drink, fruit-flavored, caffeine free"
+92410520,"SOFT DRINK, FRUIT-FLAV, SUGAR-FREE, CAFFEINE FREE","Soft drink, fruit-flavored, sugar free, caffeine free"
+92410550,"SOFT DRINK, FRUIT-FLAVORED, W/ CAFFEINE","Soft drink, fruit flavored, caffeine containing"
+92410560,"SOFT DRINK, FRUIT-FLAVORED, W/ CAFFEINE, SUGAR-FREE","Soft drink, fruit flavored, caffeine containing, sugar-free"
+92410610,"GINGER ALE","Ginger ale"
+92410620,"GINGERALE, SUGAR-FREE","Ginger ale, sugar-free"
+92410710,"ROOT BEER","Root beer"
+92410720,"ROOT BEER, SUGAR-FREE","Root beer, sugar-free"
+92410810,"CHOCOLATE-FLAVORED SODA","Chocolate-flavored soda"
+92410820,"CHOCOLATE-FLAVORED SODA, SUGAR-FREE","Chocolate-flavored soda, sugar-free"
+92411510,"COLA W/ FRUIT OR VANILLA FLAVOR","Cola with fruit or vanilla flavor"
+92411520,"COLA W/ CHOCOLATE FLAVOR","Cola with chocolate flavor"
+92411610,"COLA W/ FRUIT OR VANILLA FLAVOR, SUGAR-FREE","Cola with fruit or vanilla flavor, sugar-free"
+92411620,"COLA W/ CHOC FLAVOR, SUGAR FREE","Cola with chocolate flavor, sugar-free"
+92417010,"SOFT DRINK, ALE TYPE (INCLUDE ALE-8)","Soft drink, ale type"
+92431000,"CARBONATED JUICE DRINK, NS AS TO TYPE OF JUICE","Carbonated juice drink, NS as to type of juice"
+92432000,"CARBONATED CITRUS JUICE DRINK","Carbonated citrus juice drink"
+92433000,"CARBONATED NONCITRUS JUICE DRINK","Carbonated noncitrus juice drink"
+92510610,"FRUIT JUICE DRINK","Fruit juice drink"
+92510650,"TAMARIND DRINK, P.R. (REFRESCO DE TAMARINDO)","Tamarind drink, Puerto Rican (Refresco de tamarindo)"
+92510720,"FRUIT PUNCH, MADE W/ FRUIT JUICE & SODA","Fruit punch, made with fruit juice and soda"
+92510730,"FRUIT PUNCH, MADE W/ SODA, FRUIT JUICE & SHERBET","Fruit punch, made with soda, fruit juice, and sherbet or ice cream"
+92511000,"LEMONADE, FROZEN CONCENTRATE, NOT RECONSTITUTED","Lemonade, frozen concentrate, not reconstituted"
+92511010,"FRUIT FLAVORED DRINK (FORMERLY LEMONADE)","Fruit flavored drink (formerly lemonade)"
+92511250,"CITRUS FRUIT JUICE DRINK, CONTAINING 40-50% JUICE","Citrus fruit juice drink, containing 40-50% juice"
+92512040,"FROZEN DAIQUIRI MIX, CONCENTRATE, NOT RECONSTITUTED","Frozen daiquiri mix, frozen concentrate, not reconstituted"
+92512050,"FROZEN DAIQUIRI MIX, FROM FROZ CONC, RECONSTITUTED","Frozen daiquiri mix, from frozen concentrate, reconstituted"
+92512090,"PINA COLADA, NONALCOHOLIC","Pina Colada, nonalcoholic"
+92512110,"MARGARITA MIX, NONALCOHOLIC","Margarita mix, nonalcoholic"
+92513000,"FRUIT FLAVORED FROZEN DRINK","Fruit flavored frozen drink"
+92530410,"FRUIT FLAVORED DRINK, WITH HIGH VITAMIN C","Fruit flavored drink, with high vitamin C"
+92530510,"CRANBERRY JUICE DRINK OR COCKTAIL, W/ HIGH VIT C","Cranberry juice drink or cocktail, with high vitamin C"
+92530610,"FRUIT JUICE DRINK, WITH HIGH VITAMIN C","Fruit juice drink, with high vitamin C"
+92530950,"VEGETABLE & FRUIT JUICE DRINK, W/ HI VIT C","Vegetable and fruit juice drink, with high vitamin C"
+92531030,"FRUIT JUICE DRINK, W/ VIT B1) & HI VIT C","Fruit juice drink, with thiamin (vitamin B1) and high vitamin C"
+92541010,"FRUIT FLAVORED DRINK, MADE FROM POWDERED MIX","Fruit flavored drink, made from powdered mix"
+92542000,"FRUIT FLAVORED DRINK, MADE FROM POWDERED MIX, W/ HI VIT C","Fruit flavored drink, made from powdered mix,with high vitamin C"
+92550030,"FRUIT JUICE DRINK, LOW CALORIE, W/ HIGH VITAMIN C","Fruit juice drink, low calorie, with high vitamin C"
+92550040,"FRUIT JUICE DRINK, LOW CALORIE","Fruit juice drink, low calorie"
+92550110,"CRANBERRY JUICE DRINK OR COCKTAIL, LOW CAL, W/ HIGH VIT C","Cranberry juice drink or cocktail, low calorie, with high vitamin C"
+92550350,"LIGHT ORANGE JC BEVERAGE, 40-50% JC, LOWER SUGAR & CALORIES","Light orange juice beverage, 40-50% juice, lower sugar and calories, with artificial sweetener"
+92550400,"VEGETABLE & FRUIT JUICE DRINK, LOW CAL, W/ HIGH VIT C","Vegetable and fruit juice drink, low calorie, with high vitamin C"
+92550405,"VEGETABLE & FRUIT JUICE DRINK, LOW CAL, W/ HIGH VIT C,+E,A","Vegetable and fruit juice drink, low calorie, with high vitamin C plus added vitamin E and vitamin A"
+92550610,"FRUIT FLAVORED DRINK, LOW CAL, W/ HIGH VIT C","Fruit flavored drink, low calorie, with high vitamin C"
+92550620,"FRUIT FLAVORED DRINK, LOW CALORIE","Fruit flavored drink, low calorie"
+92552000,"FRUIT FLAV DRINK, MADE FROM PWDR, LOW CAL, W/ HI VIT C","Fruit flavored drink, made from powdered mix, low calorie, with high vitamin C"
+92552010,"FRUIT FLAVORED DRINK, MADE FROM PWDR, LOW CALORIE","Fruit flavored drink, made from powdered mix, low calorie"
+92552020,"FRUIT JUICE DRINK, REDUCED SUGAR, W/ VIT B1 & HI VIT C","Fruit juice drink, reduced sugar, with thiamin (vitamin B1) and high vitamin C"
+92552030,"FRUIT JUICE DRINK, REDUCED SUGAR, WITH VITAMIN E","Fruit juice drink, reduced sugar, with vitamin E"
+92582100,"FRUIT JUICE DRINK, WITH HIGH VITAMIN C, PLUS ADDED CALCIUM","Fruit juice drink, with high vitamin C, plus added calcium"
+92582110,"FRUIT JUICE DRINK, W/ VIT B1, HI VIT C + CALCIUM","Fruit juice drink, with thiamin (vitamin B1) and high vitamin C plus calcium"
+92610010,"HORCHATA BEVERAGE, MADE W/ ALMONDS","Horchata beverage, made with almonds or other nuts and seeds"
+92610110,"COCONUT BEVERAGE, P.R.","Coconut beverage, Puerto Rican"
+92611010,"OATMEAL BEVERAGE, P.R.","Oatmeal beverage, Puerto Rican"
+92611100,"OATMEAL BEVERAGE W/ MILK","Oatmeal beverage with milk (Atole de avena)"
+92611510,"HORCHATA BEVERAGE, MADE W/ RICE","Horchata beverage, made with rice"
+92611600,"HORCHATA BEVERAGE, NFS","Horchata beverage, NFS"
+92612010,"SUGAR CANE BEVERAGE, P.R.","Sugar cane beverage, Puerto Rican"
+92613010,"ATOLE (CORNMEAL BEVERAGE)","Atole (corn meal beverage)"
+92613510,"CORN BEV W/ CHOC & MILK(CHAMPURRADO,ATOLE DE CHOC)","Corn beverage with chocolate and milk (Champurrado, Atole de Chocolate)"
+92801000,"NONALCOHOLIC WINE","Wine, nonalcoholic"
+92802000,"WINE, LIGHT, NONALCOHOLIC","Wine, light, nonalcoholic"
+92803000,"NONALCOHOLIC MALT BEVERAGE","Nonalcoholic malt beverage"
+92804000,"SHIRLEY TEMPLE","Shirley Temple"
+92900100,"TANG, DRY CONCENTRATE","Tang, dry concentrate"
+92900110,"FRUIT-FLAV BEVERAGE, DRY CONC, W/ SUGAR, NOT RECONSTITUTED","Fruit-flavored beverage, dry concentrate, with sugar, not reconstituted"
+92900200,"FRUIT-FLAV BEV, DRY CONC,LO CAL(INCL CRYSTAL LIGHT)","Fruit-flavored beverage, dry concentrate, low calorie, not reconstituted"
+92900300,"FRUIT-FLAV THIRST QUENCH BEV, DRY CONC (GATORADE)","Fruit-flavored thirst quencher beverage, dry concentrate, not reconstituted"
+93101000,"BEER","Beer"
+93102000,"BEER, LITE","Beer, lite"
+93106000,"ALCOHOLIC MALT BEVERAGE, SWEETENED","Alcoholic malt beverage, sweetened"
+93201000,"CORDIAL OR LIQUEUR","Cordial or liqueur"
+93301000,"COCKTAIL, NFS","Cocktail, NFS"
+93301010,"ALEXANDER","Alexander"
+93301020,"BACARDI COCKTAIL","Bacardi cocktail"
+93301030,"BLOODY MARY","Bloody Mary"
+93301031,"CANADIAN CLUB & SODA","Canadian Club and soda"
+93301032,"CAPE COD","Cape Cod"
+93301040,"DAIQUIRI","Daiquiri"
+93301050,"GIMLET","Gimlet"
+93301060,"GIN & TONIC","Gin and Tonic"
+93301070,"GRASSHOPPER","Grasshopper"
+93301080,"HIGH BALL","High ball"
+93301085,"KAMIKAZE","Kamikaze"
+93301090,"MANHATTAN","Manhattan"
+93301100,"MARGARITA","Margarita"
+93301110,"MARTINI","Martini"
+93301115,"MIMOSA","Mimosa"
+93301120,"MINT JULEP","Mint julep"
+93301125,"MOJITO","Mojito"
+93301130,"OLD FASHIONED","Old fashioned"
+93301135,"ROB ROY","Rob Roy"
+93301136,"RUSTY NAIL","Rusty Nail"
+93301139,"SALTY DOG","Salty Dog"
+93301140,"SCREWDRIVER (INCLUDE HARVEY WALLBANGER, SLO-SCREW)","Screwdriver"
+93301141,"SEABREEZE","Seabreeze"
+93301142,"SEVEN AND SEVEN","Seven and Seven"
+93301150,"TOM COLLINS (INCLUDE VODKA COLLINS)","Tom Collins"
+93301160,"WHISKEY SOUR(INCL SCOTCH,VODKA,APRICOT,BRANDY SOUR)","Whiskey sour"
+93301170,"BOURBON & SODA (INCLUDE SCOTCH & SODA, RUM & SODA)","Bourbon and soda"
+93301180,"MIXED DRINKS (FOR RECIPE MODIFICATIONS)","Mixed Drinks (for recipe modifications)"
+93301190,"RUM & COLA","Rum and cola"
+93301200,"PINA COLADA","Pina Colada"
+93301220,"COQUITO, P.R. (COCONUT, RUM)","Coquito, Puerto Rican (coconut, rum)"
+93301230,"SLOE GIN FIZZ","Sloe gin fizz"
+93301240,"BLACK RUSSIAN","Black Russian"
+93301250,"WHITE RUSSIAN","White Russian"
+93301270,"FRUIT PUNCH, ALCOHOLIC","Fruit punch, alcoholic"
+93301280,"SINGAPORE SLING","Singapore Sling"
+93301290,"STINGER","Stinger"
+93301300,"GIBSON","Gibson"
+93301310,"MAI TAI","Mai Tai"
+93301320,"TEQUILA SUNRISE","Tequila Sunrise"
+93301330,"GIN RICKEY","Gin Rickey"
+93301340,"GOLDEN CADILLAC","Golden Cadillac"
+93301360,"LONG ISLAND ICED TEA","Long Island iced tea"
+93301370,"FUZZY NAVEL COCKTAIL","Fuzzy Navel"
+93301400,"IRISH COFFEE (INCL COFFEE ROYALE)","Irish Coffee"
+93301450,"LIQUEUR W/ CREAM","Liqueur with cream"
+93301500,"FROZEN DAIQUIRI","Frozen daiquiri"
+93301510,"FROZEN MARGARITA","Frozen margarita"
+93301550,"EGGNOG, ALCOHOLIC","Eggnog, alcoholic"
+93301600,"GIN FIZZ","Gin fizz"
+93302000,"RUM, HOT BUTTERED","Rum, hot buttered"
+93302100,"ZOMBIE","Zombie"
+93401010,"WINE, TABLE, RED","Wine, table, red"
+93401020,"WINE, TABLE, WHITE","Wine, table, white"
+93401100,"WINE, RICE (INCLUDE SAKI)","Wine, rice"
+93401300,"WINE, COOKING (ASSUME COOKED)","Wine, cooking (assume cooked)"
+93402000,"WINE, DESSERT (INCLUDE MARSALA, PORT, MADEIRA)","Wine, dessert, sweet"
+93403000,"WINE, LIGHT","Wine, light"
+93404000,"WINE COOLER","Wine cooler"
+93404500,"SANGRIA","Sangria"
+93404600,"SANGRIA, PUERTO RICAN STYLE","Sangria, Puerto Rican style"
+93405000,"WINE SPRITZER","Wine spritzer"
+93406000,"GLUG (INCLUDE GLOGG, GLUHWEIN)","Glug"
+93501000,"BRANDY","Brandy"
+93502000,"WHISKEY","Whiskey"
+93503000,"GIN","Gin"
+93504000,"RUM","Rum"
+93504100,"RUM COOLER","Rum cooler"
+93505000,"VODKA","Vodka"
+94000100,"WATER, TAP","Water, tap"
+94100100,"WATER, BOTTLED, UNSWEETENED","Water, bottled, unsweetened"
+94100200,"WATER, BOTTLED, SWEETENED, WITH LOW OR NO CALORIE SWEETENER","Water, bottled, sweetened, with low or no calorie sweetener"
+94100300,"WATER, FRUIT FLAVORED, SWTND, W/ CORN SYRUP & LOWCAL SWTNR","Water, fruit flavored, sweetened, with high fructose corn syrup and low calorie sweetener"
+94210100,"PROPEL WATER","Propel Water"
+94210200,"GLACEAU WATER","Glaceau Water"
+94210300,"SOBE LIFEWATER","SoBe Lifewater"
+94220200,"GLACEAU WATER, LOW CALORIE","Glaceau Water, low calorie"
+94300100,"WATER, BABY, BOTTLED, UNSWEETENED","Water, baby, bottled, unsweetened"
+95101000,"BOOST, NUTRITIONAL DRINK, READY-TO-DRINK","Boost, nutritional drink, ready-to-drink"
+95101010,"BOOST PLUS, NUTRITIONAL DRINK, READY-TO-DRINK","Boost Plus, nutritional drink, ready-to-drink"
+95102000,"CARNATION INSTANT BREAKFAST, NUTRITIONAL DRINK, REGULAR, RTD","Carnation Instant Breakfast, nutritional drink, regular, ready-to-drink"
+95102010,"CARNATION INSTANT BREAKFAST, NUTRITIONAL DRINK, SUGAR FREE,","Carnation Instant Breakfast, nutritional drink, sugar free, ready-to-drink"
+95103000,"ENSURE, NUTRITIONAL SHAKE, READY-TO-DRINK","Ensure, nutritional shake, ready-to-drink"
+95103010,"ENSURE PLUS, NUTRITIONAL SHAKE, READY-TO-DRINK","Ensure Plus, nutritional shake, ready-to-drink"
+95104000,"GLUCERNA, NUTRITIONAL SHAKE, READY-TO-DRINK","Glucerna, nutritional shake, ready-to-drink"
+95105000,"KELLOGG'S SPECIAL K PROTEIN SHAKE","Kellogg's Special K Protein Shake"
+95106000,"MUSCLE MILK, READY-TO-DRINK","Muscle Milk, ready-to-drink"
+95106010,"MUSCLE MILK, LIGHT, READY-TO-DRINK","Muscle Milk, light, ready-to-drink"
+95110000,"SLIM FAST SHAKE, MEAL REPLACEMENT, REGULAR, READY-TO-DRINK","Slim Fast Shake, meal replacement, regular, ready-to-drink"
+95110010,"SLIM FAST SHAKE, MEAL REPLACEMENT, SUGAR FREE, RTD","Slim Fast Shake, meal replacement, sugar free, ready-to-drink"
+95110020,"SLIM FAST SHAKE, MEAL REPLACEMENT, HIGH PROTEIN, RTD","Slim Fast Shake, meal replacement, high protein, ready-to-drink"
+95120000,"NUTRITIONAL DRINK OR MEAL REPLACEMENT, READY-TO-DRINK, NFS","Nutritional drink or meal replacement, ready-to-drink, NFS"
+95120010,"NUTRITIONAL DRINK OR MEAL REPLACEMENT, HIGH PROTEIN, RTD","Nutritional drink or meal replacement, high protein, ready-to-drink, NFS"
+95120020,"NUTRITIONAL DRINK OR MEAL REPLACEMENT, HI PROT, LIGHT, RTD","Nutritional drink or meal replacement, high protein, light, ready-to-drink, NFS"
+95120050,"NUTRITIONAL DRINK OR MEAL REPLACEMENT, LIQUID, SOY-BASED","Nutritional drink or meal replacement, liquid, soy-based"
+95201000,"CARNATION INSTANT BREAKFAST, NUTRITIONAL DRINK MIX, REG,PDR","Carnation Instant Breakfast, nutritional drink mix, regular, powder"
+95201010,"CARNATION INSTANT BREAKFAST, NUTR DRINK MIX, SUGAR FREE,PDR","Carnation Instant Breakfast, nutritional drink mix, sugar free, powder"
+95201200,"EAS WHEY PROTEIN POWDER","EAS Whey Protein Powder"
+95201300,"EAS SOY PROTEIN POWDER","EAS Soy Protein Powder"
+95201500,"HERBALIFE, NUTRITIONAL SHAKE MIX, HIGH PROTEIN, POWDER","Herbalife, nutritional shake mix, high protein, powder"
+95201600,"ISOPURE PROTEIN POWDER","Isopure protein powder"
+95201700,"KELLOGG'S SPECIAL K20 PROTEIN WATER MIX","Kellogg's Special K20 Protein Water Mix"
+95202000,"MUSCLE MILK, REGULAR, POWDER","Muscle Milk, regular, powder"
+95202010,"MUSCLE MILK, LIGHT, POWDER","Muscle Milk, light, powder"
+95210000,"SLIM FAST SHAKE MIX, POWDER","Slim Fast Shake Mix, powder"
+95210010,"SLIM FAST SHAKE MIX, SUGAR FREE, POWDER","Slim Fast Shake Mix, sugar free, powder"
+95210020,"SLIM FAST SHAKE MIX, HIGH PROTEIN, POWDER","Slim Fast Shake Mix, high protein, powder"
+95220000,"NUTRITIONAL DRINK MIX OR MEAL REPLACEMENT, POWDER, NFS","Nutritional drink mix or meal replacement, powder, NFS"
+95220010,"NUTRITIONAL DRINK MIX OR MEAL REPLACEMENT, HIGH PRO, PDR,NFS","Nutritional drink mix or meal replacement, high protein, powder, NFS"
+95230000,"PROTEIN POWDER, WHEY BASED, NFS","Protein powder, whey based, NFS"
+95230010,"PROTEIN POWDER, SOY BASED, NFS","Protein powder, soy based, NFS"
+95230020,"PROTEIN POWDER, LIGHT, NFS","Protein powder, light, NFS"
+95230030,"PROTEIN POWDER, NFS","Protein powder, NFS"
+95310200,"FULL THROTTLE ENERGY DRINK","Full Throttle Energy Drink"
+95310400,"MONSTER ENERGY DRINK","Monster Energy Drink"
+95310500,"MOUNTAIN DEW AMP ENERGY DRINK","Mountain Dew AMP Energy Drink"
+95310550,"NO FEAR ENERGY DRINK","No Fear Energy Drink"
+95310555,"NO FEAR MOTHERLOAD ENERGY DRINK","No Fear Motherload Energy Drink"
+95310560,"NOS ENERGY DRINK","NOS Energy Drink"
+95310600,"RED BULL ENERGY DRINK","Red Bull Energy Drink"
+95310700,"ROCKSTAR ENERGY DRINK","Rockstar Energy Drink"
+95310750,"SOBE ENERGIZE ENERGY JUICE DRINK","SoBe Energize Energy Juice Drink"
+95310800,"VAULT ENERGY DRINK","Vault Energy Drink"
+95311000,"ENERGY DRINK","Energy Drink"
+95312400,"MONSTER ENERGY DRINK, LO CARB","Monster Energy Drink, Lo Carb"
+95312500,"MOUNTAIN DEW AMP ENERGY DRINK, SUGAR-FREE","Mountain Dew AMP Energy Drink, sugar-free"
+95312550,"NO FEAR ENERGY DRINK, SUGAR-FREE","No Fear Energy Drink, sugar-free"
+95312555,"NOS ENERGY DRINK, SUGAR-FREE","NOS Energy Drink, sugar-free"
+95312560,"CRANBERRY JUICE ENERGY DRINK, HI VIT C & B, W/LOW CAL SWTNR","Ocean Spray Cran-Energy Cranberry Energy Juice Drink"
+95312600,"RED BULL ENERGY DRINK, SUGAR-FREE","Red Bull Energy Drink, sugar-free"
+95312700,"ROCKSTAR ENERGY DRINK, SUGAR-FREE","Rockstar Energy Drink, sugar-free"
+95312800,"VAULT ZERO ENERGY DRINK","Vault Zero Energy Drink"
+95312900,"XS ENERGY DRINK","XS Energy Drink"
+95312905,"XS GOLD PLUS ENERGY DRINK","XS Gold Plus Energy Drink"
+95320200,"GATORADE THIRST QUENCHER SPORTS DRINK","Gatorade Thirst Quencher sports drink"
+95320500,"POWERADE SPORTS DRINK","Powerade sports drink"
+95321000,"FRUIT-FLAVORED THIRST QUENCHER BEVERAGE","Fruit-flavored thirst quencher beverage"
+95322200,"GATORADE G2 THIRST QUENCHER SPORTS DRINK, LOW CALORIE","Gatorade G2 Thirst Quencher sports drink, low calorie"
+95322500,"POWERADE ZERO SPORTS DRINK, LOW CALORIE","Powerade Zero sports drink, low calorie"
+95323000,"FRUIT-FLAV SPORTS DRINK OR THIRST QUENCHER BEVERAGE, LOW CAL","Fruit-flavored sports drink or thirst quencher beverage, low calorie"
+95330100,"FLUID REPLACEMENT, ELECTROLYTE SOLUTION","Fluid replacement, electrolyte solution"
+95330500,"FLUID REPLACEMENT, 5% GLUCOSE IN WATER","Fluid replacement, 5% glucose in water"
+95341000,"FUZE SLENDERIZE FORTIFIED LOW CALORIE FRUIT JUICE BEVERAGE","FUZE Slenderize fortified low calorie fruit juice beverage"
+95342000,"MONAVIE ACAI BLEND BEVERAGE","MonaVie acai blend beverage"
diff --git a/pandas/pandas/tests/io/sas/data/SSHSV1_A.csv b/pandas/pandas/tests/io/sas/data/SSHSV1_A.csv
new file mode 100644
index 0000000000000..d34efe4b895f7
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/SSHSV1_A.csv
@@ -0,0 +1,1427 @@
+"SEQN","SSXHE1"
+3,2
+8,1
+9,2
+22,1
+23,2
+30,1
+39,2
+49,1
+67,1
+74,2
+75,2
+76,1
+78,2
+109,1
+110,1
+128,2
+139,2
+140,1
+144,1
+154,2
+159,2
+180,2
+182,1
+185,2
+202,1
+203,1
+212,2
+214,1
+224,2
+226,2
+227,2
+231,2
+246,2
+251,2
+252,2
+278,2
+299,1
+306,2
+311,2
+336,1
+338,2
+343,1
+347,2
+348,2
+349,2
+356,2
+362,1
+378,1
+379,2
+384,1
+390,2
+394,1
+404,1
+418,2
+423,1
+433,2
+435,2
+445,2
+449,2
+450,2
+482,2
+485,2
+487,2
+490,2
+500,2
+501,1
+502,2
+518,1
+520,1
+525,2
+538,2
+556,2
+565,2
+589,2
+599,2
+601,2
+604,1
+609,1
+617,2
+619,2
+624,1
+625,2
+629,2
+635,2
+637,2
+649,2
+659,1
+664,2
+673,1
+679,2
+696,1
+700,1
+704,2
+707,2
+715,1
+721,1
+724,2
+726,2
+735,2
+745,2
+752,1
+759,1
+760,2
+766,2
+767,1
+771,2
+772,1
+777,1
+791,2
+795,1
+796,1
+814,2
+837,2
+845,2
+860,1
+872,1
+890,2
+900,1
+904,1
+912,1
+913,1
+916,2
+922,2
+927,1
+934,2
+938,2
+947,2
+951,2
+952,2
+967,2
+973,2
+985,2
+997,2
+1006,2
+1008,2
+1012,1
+1015,1
+1017,1
+1036,2
+1037,2
+1039,2
+1069,1
+1073,2
+1078,1
+1080,2
+1081,1
+1086,1
+1091,2
+1103,2
+1126,2
+1136,2
+1142,2
+1145,2
+1157,2
+1162,1
+1172,1
+1174,1
+1200,2
+1206,2
+1218,2
+1223,2
+1224,1
+1225,2
+1228,1
+1235,2
+1258,1
+1261,2
+1265,2
+1269,2
+1271,2
+1272,1
+1300,1
+1310,2
+1311,2
+1319,2
+1321,1
+1324,1
+1334,2
+1337,1
+1341,1
+1345,1
+1347,1
+1350,2
+1359,2
+1360,2
+1371,1
+1376,1
+1380,2
+1390,2
+1395,2
+1396,2
+1415,1
+1420,2
+1426,2
+1435,2
+1446,1
+1454,1
+1456,2
+1460,2
+1485,1
+1490,1
+1496,1
+1504,1
+1519,1
+1521,1
+1522,2
+1523,1
+1530,2
+1556,1
+1559,1
+1560,1
+1562,1
+1569,1
+1572,2
+1610,2
+1611,2
+1632,2
+1640,2
+1649,2
+1653,1
+1654,1
+1656,2
+1657,1
+1664,1
+1670,2
+1688,2
+1699,1
+1708,2
+1709,2
+1710,2
+1724,2
+1740,1
+1745,2
+1747,1
+1753,1
+1769,2
+1771,2
+1797,1
+1811,1
+1812,2
+1820,2
+1833,2
+1836,1
+1842,1
+1843,1
+1844,2
+1857,1
+1863,1
+1872,2
+1881,1
+1890,1
+1893,1
+1900,2
+1915,1
+1926,2
+1927,2
+1937,2
+1940,2
+1953,2
+1954,1
+1961,1
+1965,2
+2001,1
+2011,2
+2012,2
+2028,2
+2032,2
+2033,1
+2042,1
+2044,2
+2048,1
+2090,2
+2093,1
+2099,1
+2100,2
+2123,2
+2125,2
+2128,1
+2136,1
+2147,2
+2148,2
+2154,2
+2155,1
+2157,2
+2158,1
+2183,1
+2192,1
+2202,1
+2206,1
+2208,2
+2211,1
+2218,1
+2227,2
+2266,1
+2274,2
+2279,1
+2284,2
+2285,1
+2286,2
+2288,1
+2289,2
+2295,2
+2299,1
+2306,1
+2309,1
+2310,1
+2313,2
+2316,1
+2341,1
+2368,2
+2369,1
+2374,1
+2387,1
+2394,2
+2403,2
+2408,2
+2409,1
+2410,1
+2411,2
+2415,2
+2430,2
+2431,2
+2452,2
+2453,2
+2511,2
+2513,2
+2516,1
+2537,1
+2538,2
+2545,2
+2547,1
+2554,2
+2559,2
+2562,2
+2568,1
+2580,2
+2584,1
+2587,2
+2592,1
+2593,2
+2600,1
+2607,2
+2623,1
+2624,2
+2626,2
+2632,2
+2637,2
+2639,1
+2647,1
+2664,1
+2677,2
+2690,2
+2707,1
+2715,2
+2717,1
+2724,2
+2728,2
+2737,1
+2751,1
+2760,1
+2767,2
+2773,2
+2776,2
+2779,2
+2781,2
+2785,2
+2786,1
+2829,2
+2832,1
+2833,2
+2836,1
+2840,2
+2842,2
+2853,1
+2856,2
+2867,2
+2874,2
+2875,2
+2876,2
+2878,1
+2879,2
+2891,2
+2903,1
+2916,2
+2918,2
+2943,2
+2960,2
+2963,2
+2974,1
+2975,2
+2976,1
+2981,2
+2982,1
+2985,1
+2989,1
+3001,1
+3003,1
+3004,2
+3007,2
+3017,2
+3019,2
+3022,2
+3023,2
+3025,2
+3027,2
+3030,1
+3031,2
+3041,2
+3046,2
+3050,1
+3052,1
+3056,1
+3060,2
+3061,1
+3071,2
+3078,1
+3079,1
+3084,2
+3086,1
+3088,2
+3090,2
+3092,2
+3103,2
+3111,1
+3125,1
+3131,1
+3138,2
+3151,1
+3159,1
+3177,2
+3178,1
+3186,2
+3188,2
+3189,2
+3201,1
+3203,2
+3207,2
+3211,2
+3217,1
+3241,2
+3248,1
+3253,2
+3267,2
+3286,1
+3288,1
+3290,1
+3295,2
+3306,2
+3315,2
+3320,1
+3322,1
+3335,1
+3340,2
+3351,1
+3352,1
+3369,2
+3373,1
+3378,2
+3383,1
+3384,1
+3390,2
+3391,2
+3392,2
+3394,2
+3396,1
+3414,1
+3416,1
+3422,2
+3424,2
+3427,2
+3435,2
+3447,2
+3465,2
+3467,2
+3471,1
+3475,2
+3483,1
+3495,2
+3497,2
+3503,2
+3507,2
+3511,2
+3512,2
+3520,2
+3527,2
+3529,2
+3533,2
+3542,1
+3550,2
+3555,2
+3560,2
+3582,1
+3589,2
+3608,1
+3616,1
+3622,2
+3623,1
+3629,2
+3641,1
+3642,2
+3643,1
+3650,2
+3651,1
+3657,2
+3665,2
+3667,1
+3669,1
+3673,2
+3674,2
+3698,2
+3703,2
+3706,2
+3723,2
+3728,2
+3733,2
+3734,1
+3738,1
+3754,1
+3764,2
+3766,1
+3768,2
+3772,1
+3776,2
+3779,2
+3784,1
+3788,2
+3789,2
+3812,1
+3813,1
+3828,2
+3831,1
+3832,2
+3833,2
+3834,1
+3837,1
+3842,2
+3844,2
+3846,2
+3847,2
+3853,2
+3869,2
+3872,1
+3875,2
+3876,2
+3877,2
+3906,2
+3922,2
+3928,2
+3933,1
+3938,1
+3947,1
+3948,2
+3962,1
+3969,2
+3970,2
+3994,2
+4001,2
+4004,2
+4024,1
+4039,2
+4040,2
+4041,2
+4043,2
+4055,1
+4068,2
+4073,2
+4077,2
+4086,1
+4091,1
+4098,1
+4099,2
+4105,1
+4118,2
+4121,2
+4124,1
+4129,2
+4135,2
+4152,2
+4154,2
+4166,2
+4177,1
+4179,2
+4186,1
+4188,2
+4189,1
+4200,2
+4204,1
+4224,1
+4226,2
+4228,1
+4233,2
+4235,1
+4237,1
+4243,1
+4247,2
+4251,2
+4256,1
+4279,2
+4294,2
+4304,2
+4306,1
+4315,1
+4321,2
+4325,1
+4335,2
+4347,2
+4360,1
+4383,2
+4392,2
+4414,1
+4416,1
+4417,2
+4424,2
+4427,2
+4428,2
+4436,2
+4437,1
+4471,2
+4474,2
+4475,2
+4485,1
+4487,1
+4496,2
+4499,2
+4509,1
+4515,1
+4519,2
+4520,1
+4522,1
+4541,1
+4544,2
+4546,1
+4568,2
+4573,2
+4575,1
+4576,1
+4578,1
+4580,1
+4585,2
+4588,2
+4595,2
+4599,2
+4603,1
+4606,1
+4607,1
+4609,2
+4612,2
+4613,1
+4617,2
+4622,1
+4649,1
+4651,2
+4656,1
+4659,2
+4668,2
+4670,2
+4671,2
+4672,2
+4673,2
+4686,2
+4687,2
+4694,1
+4724,1
+4733,1
+4737,1
+4738,2
+4756,2
+4759,2
+4763,2
+4770,1
+4773,2
+4790,2
+4808,1
+4811,2
+4814,2
+4818,2
+4832,2
+4836,1
+4837,2
+4857,1
+4870,2
+4871,2
+4873,1
+4881,2
+4884,1
+4886,2
+4887,2
+4889,1
+4890,1
+4904,1
+4906,1
+4913,2
+4920,1
+4923,2
+4930,2
+4943,1
+4944,2
+4949,1
+4950,2
+4953,2
+4973,1
+4975,1
+4980,2
+4981,2
+4984,1
+5004,2
+5005,2
+5014,2
+5017,1
+5038,1
+5058,1
+5061,1
+5087,1
+5097,1
+5098,1
+5099,1
+5106,2
+5117,1
+5132,1
+5138,1
+5140,2
+5156,1
+5163,2
+5169,2
+5174,1
+5178,1
+5179,2
+5183,2
+5189,1
+5190,1
+5194,2
+5196,2
+5200,2
+5228,1
+5252,2
+5258,1
+5268,2
+5270,1
+5272,2
+5286,2
+5287,2
+5289,1
+5290,2
+5293,1
+5304,1
+5312,2
+5318,2
+5337,3
+5345,1
+5349,2
+5351,1
+5356,2
+5360,1
+5366,2
+5367,1
+5368,2
+5372,2
+5373,1
+5382,1
+5387,2
+5388,1
+5389,2
+5391,2
+5392,2
+5398,1
+5412,1
+5425,1
+5427,1
+5430,2
+5435,2
+5439,1
+5444,1
+5446,1
+5453,1
+5462,1
+5471,2
+5473,2
+5485,1
+5486,2
+5491,1
+5495,2
+5497,2
+5500,1
+5502,2
+5507,1
+5513,2
+5518,2
+5523,2
+5530,1
+5534,2
+5536,2
+5540,1
+5550,2
+5551,2
+5558,1
+5566,2
+5574,1
+5584,2
+5594,1
+5607,2
+5609,1
+5611,2
+5617,2
+5618,2
+5619,1
+5628,1
+5638,1
+5640,1
+5661,1
+5663,2
+5668,2
+5670,2
+5677,2
+5688,2
+5693,1
+5698,1
+5711,1
+5723,2
+5729,1
+5739,1
+5748,1
+5751,2
+5756,2
+5770,1
+5771,2
+5796,2
+5800,1
+5804,2
+5812,2
+5817,1
+5821,2
+5823,1
+5840,1
+5847,1
+5850,1
+5887,2
+5899,2
+5900,1
+5917,2
+5928,2
+5940,2
+5948,2
+5951,2
+5955,2
+5956,1
+5957,1
+5959,2
+5971,2
+5973,2
+5974,2
+5975,2
+5978,2
+5990,1
+5999,2
+6007,2
+6012,2
+6013,2
+6014,1
+6024,2
+6026,2
+6028,2
+6035,1
+6038,2
+6042,1
+6048,1
+6052,1
+6055,2
+6056,1
+6059,1
+6060,1
+6062,1
+6076,1
+6084,2
+6091,2
+6100,1
+6103,1
+6108,2
+6113,1
+6128,1
+6134,2
+6150,2
+6172,2
+6186,2
+6187,1
+6191,1
+6192,2
+6202,2
+6207,1
+6225,2
+6229,2
+6230,1
+6235,2
+6239,2
+6241,1
+6252,2
+6257,2
+6269,1
+6276,1
+6282,2
+6284,2
+6286,2
+6288,1
+6295,1
+6301,1
+6307,2
+6322,2
+6324,1
+6333,1
+6346,1
+6350,2
+6357,1
+6366,1
+6370,1
+6380,2
+6381,1
+6389,1
+6391,2
+6392,2
+6399,2
+6405,1
+6409,2
+6410,2
+6411,2
+6421,2
+6425,2
+6427,2
+6443,2
+6474,1
+6475,1
+6490,2
+6492,2
+6500,2
+6501,2
+6502,1
+6504,2
+6505,2
+6506,1
+6528,2
+6544,2
+6554,2
+6555,2
+6560,2
+6563,1
+6576,1
+6580,1
+6582,1
+6589,2
+6606,2
+6607,2
+6609,1
+6615,1
+6620,1
+6632,1
+6635,1
+6638,1
+6647,2
+6651,2
+6652,2
+6654,2
+6657,1
+6659,1
+6660,1
+6685,2
+6691,2
+6695,2
+6699,2
+6719,2
+6722,2
+6725,1
+6729,1
+6746,1
+6748,1
+6759,1
+6762,2
+6770,1
+6778,2
+6782,1
+6784,2
+6785,1
+6787,2
+6793,1
+6794,2
+6798,2
+6806,1
+6817,2
+6828,2
+6831,2
+6833,1
+6834,1
+6842,1
+6843,2
+6848,1
+6860,1
+6868,2
+6876,1
+6880,2
+6881,2
+6887,1
+6893,1
+6900,2
+6901,2
+6909,1
+6911,2
+6919,1
+6920,1
+6921,1
+6924,2
+6925,1
+6930,2
+6932,1
+6945,1
+6947,1
+6961,2
+6964,1
+6970,1
+6977,2
+6984,2
+6986,2
+6997,2
+7003,2
+7004,2
+7006,2
+7009,2
+7015,2
+7023,1
+7028,2
+7033,2
+7041,2
+7046,2
+7052,1
+7070,1
+7081,2
+7094,1
+7104,2
+7110,2
+7113,1
+7115,2
+7122,1
+7128,1
+7129,1
+7131,1
+7135,2
+7137,1
+7138,1
+7141,1
+7150,1
+7153,2
+7155,2
+7157,2
+7158,2
+7164,2
+7181,2
+7189,1
+7194,2
+7209,1
+7215,2
+7239,1
+7242,1
+7245,2
+7253,2
+7255,1
+7260,2
+7280,1
+7289,2
+7292,2
+7303,1
+7305,2
+7317,1
+7323,2
+7327,1
+7343,1
+7361,1
+7368,2
+7402,2
+7405,1
+7415,2
+7417,2
+7423,1
+7445,1
+7446,1
+7455,2
+7463,2
+7467,2
+7475,1
+7481,1
+7500,2
+7503,1
+7507,1
+7510,2
+7527,2
+7528,2
+7543,1
+7551,2
+7559,2
+7564,2
+7569,2
+7573,1
+7579,2
+7584,1
+7588,2
+7591,1
+7595,1
+7610,1
+7612,1
+7629,1
+7642,1
+7648,2
+7653,2
+7679,2
+7690,2
+7693,2
+7700,2
+7702,2
+7703,1
+7705,2
+7709,2
+7719,1
+7737,2
+7745,1
+7754,1
+7770,2
+7772,2
+7783,1
+7785,2
+7786,1
+7789,1
+7793,1
+7799,2
+7800,2
+7817,2
+7818,2
+7832,1
+7852,1
+7854,2
+7879,1
+7880,2
+7881,1
+7882,1
+7889,2
+7895,1
+7902,2
+7908,2
+7914,2
+7916,2
+7920,1
+7921,2
+7924,1
+7930,1
+7931,2
+7943,1
+7976,2
+7978,2
+8024,1
+8025,2
+8032,1
+8054,1
+8058,1
+8066,1
+8069,2
+8070,1
+8072,2
+8082,1
+8087,2
+8089,1
+8098,1
+8111,2
+8126,1
+8135,2
+8136,1
+8141,2
+8144,2
+8146,2
+8166,2
+8191,2
+8198,1
+8210,2
+8212,2
+8228,1
+8253,1
+8254,2
+8256,2
+8272,1
+8274,2
+8280,1
+8283,2
+8292,1
+8300,2
+8305,2
+8306,1
+8316,2
+8320,1
+8324,2
+8332,1
+8342,1
+8344,1
+8346,2
+8364,1
+8370,1
+8371,1
+8374,2
+8380,1
+8383,2
+8386,2
+8393,2
+8398,1
+8410,2
+8415,2
+8420,2
+8433,2
+8435,2
+8436,1
+8438,2
+8439,1
+8456,1
+8463,1
+8479,1
+8480,1
+8493,2
+8502,2
+8503,2
+8505,2
+8506,2
+8515,1
+8517,2
+8520,2
+8530,2
+8543,2
+8550,1
+8560,2
+8571,1
+8576,2
+8582,2
+8583,2
+8590,1
+8591,2
+8592,2
+8594,2
+8599,2
+8621,1
+8623,1
+8627,1
+8635,2
+8636,2
+8640,1
+8647,1
+8648,2
+8662,2
+8675,2
+8679,2
+8680,1
+8685,2
+8686,2
+8691,2
+8692,1
+8696,1
+8699,1
+8700,2
+8722,2
+8735,2
+8758,2
+8766,1
+8770,1
+8789,2
+8790,2
+8791,2
+8805,2
+8809,2
+8812,1
+8826,2
+8843,1
+8845,1
+8847,1
+8853,2
+8873,2
+8890,1
+8933,1
+8935,1
+8963,1
+8977,2
+8981,2
+8982,1
+8983,2
+8987,1
+8988,1
+8989,1
+9017,2
+9026,2
+9034,2
+9038,2
+9041,1
+9045,2
+9049,1
+9057,1
+9058,2
+9060,2
+9076,1
+9080,2
+9083,1
+9084,2
+9085,2
+9088,2
+9089,2
+9093,1
+9094,1
+9098,1
+9100,2
+9107,1
+9109,1
+9111,2
+9114,2
+9120,2
+9134,2
+9135,2
+9137,1
+9138,2
+9143,1
+9151,2
+9160,2
+9164,1
+9166,2
+9167,1
+9174,1
+9184,1
+9189,2
+9205,2
+9224,1
+9236,2
+9249,2
+9255,1
+9261,2
+9265,1
+9266,1
+9269,2
+9271,1
+9283,2
+9284,2
+9287,2
+9294,1
+9305,1
+9307,2
+9315,2
+9317,2
+9320,2
+9336,1
+9337,2
+9345,2
+9362,2
+9365,2
+9385,2
+9387,2
+9393,1
+9398,2
+9404,1
+9406,2
+9419,2
+9437,2
+9448,1
+9461,2
+9476,1
+9491,2
+9524,2
+9529,2
+9545,2
+9555,1
+9562,1
+9565,2
+9576,2
+9578,2
+9586,1
+9588,2
+9593,1
+9596,2
+9601,1
+9631,2
+9632,2
+9634,1
+9635,1
+9654,2
+9659,2
+9673,2
+9680,2
+9688,1
+9698,1
+9700,2
+9702,1
+9705,2
+9708,2
+9713,2
+9721,1
+9724,2
+9734,2
+9743,1
+9758,1
+9759,2
+9777,2
+9782,2
+9788,1
+9789,1
+9795,2
+9800,1
+9801,2
+9827,1
+9830,2
+9831,2
+9839,1
+9841,2
+9848,1
+9856,1
+9858,1
+9859,2
+9867,2
+9870,1
+9880,1
+9885,1
+9887,2
+9899,2
+9900,2
+9901,2
+9914,2
+9918,1
+9927,1
+9928,1
+9933,2
+9939,2
+9945,2
+9964,2
diff --git a/pandas/pandas/tests/io/sas/data/airline.csv b/pandas/pandas/tests/io/sas/data/airline.csv
new file mode 100644
index 0000000000000..a6336446003a9
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/airline.csv
@@ -0,0 +1,33 @@
+YEAR,Y,W,R,L,K
+1948.00000839,1.21399998665,0.243000000715,0.145400002599,1.41499996185,0.611999988556
+1949.0000084,1.35399997234,0.259999990463,0.218099996448,1.38399994373,0.559000015259
+1950.0,1.56900000572,0.277999997139,0.315699994564,1.38800001144,0.573000013828
+1951.0,1.94799995422,0.29699999094,0.393999993801,1.54999995232,0.56400001049
+1952.0,2.2650001049,0.310000002384,0.355899989605,1.80200004578,0.574000000954
+1953.0,2.73099994659,0.321999996901,0.359299987555,1.92599999905,0.711000025272
+1954.0,3.02500009537,0.335000008345,0.402500003576,1.96399998665,0.776000022888
+1955.0,3.56200003624,0.34999999404,0.396100014448,2.11599993706,0.827000021935
+1956.0,3.97900009155,0.361000001431,0.38220000267,2.43499994278,0.800000011921
+1957.00036436,4.42000007629,0.379000008106,0.30450001359,2.70700001717,0.921000003815
+1958.00096758,4.56300020218,0.391000002623,0.328399986029,2.70600008965,1.06700003147
+1959.00096116,5.38500022888,0.425999999046,0.38560000062,2.84599995613,1.08299994469
+1960.0,5.55399990082,0.441000014544,0.319299995899,3.08899998665,1.48099994659
+1961.0,5.46500015259,0.460000008345,0.307900011539,3.12199997902,1.73599994183
+1962.0,5.82499980927,0.485000014305,0.378300011158,3.18400001526,1.92599999905
+1963.0,6.87599992752,0.505999982357,0.418000012636,3.26300001144,2.04099988937
+1964.0,7.82299995422,0.537999987602,0.516300022602,3.41199994087,1.99699997902
+1965.0,9.11999988556,0.56400001049,0.587899982929,3.62299990654,2.25699996948
+1966.00082638,10.5120000839,0.586000025272,0.536899983883,4.07399988174,2.742000103
+1967.00001699,13.0200004578,0.621999979019,0.444299995899,4.71000003815,3.56399989128
+1968.00096202,15.2609996796,0.666000008583,0.305200010538,5.21700000763,4.76700019836
+1969.00023643,16.3129997253,0.731000006199,0.233199998736,5.5689997673,6.5110001564
+1970.00003926,16.0020008087,0.830999970436,0.188299998641,5.49499988556,7.62699985504
+1971.00096131,15.8760004044,0.90600001812,0.202299997211,5.33400011063,8.67300033569
+1972.0001291,16.6620006561,1.0,0.250600010157,5.34499979019,8.33100032806
+1973.00001162,17.013999939,1.05599999428,0.266799986362,5.66200017929,8.55700016022
+1974.00096128,19.3050003052,1.13100004196,0.266400009394,5.72900009155,9.50800037384
+1975.0,18.7210006714,1.24699997902,0.230100005865,5.72200012207,9.06200027466
+1976.0,19.25,1.375,0.345200002193,5.76200008392,8.26200008392
+1977.0,20.6469993591,1.54400002956,0.450800001621,5.87699985504,7.47399997711
+1978.0,22.7259998322,1.70299994946,0.587700009346,6.10799980164,7.10400009155
+1979.0,23.6189994812,1.77900004387,0.534600019455,6.85200023651,6.87400007248
diff --git a/pandas/pandas/tests/io/sas/data/airline.sas7bdat b/pandas/pandas/tests/io/sas/data/airline.sas7bdat
new file mode 100644
index 0000000000000..462299bc6a952
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/airline.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/airline.sas7bdat.gz b/pandas/pandas/tests/io/sas/data/airline.sas7bdat.gz
new file mode 100644
index 0000000000000..7b56e492295f4
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/airline.sas7bdat.gz differ
diff --git a/pandas/pandas/tests/io/sas/data/cars.sas7bdat b/pandas/pandas/tests/io/sas/data/cars.sas7bdat
new file mode 100644
index 0000000000000..ca5d3474c36ad
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/cars.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/corrupt.sas7bdat b/pandas/pandas/tests/io/sas/data/corrupt.sas7bdat
new file mode 100644
index 0000000000000..2941ffe3ecdf5
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/corrupt.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/dates_null.sas7bdat b/pandas/pandas/tests/io/sas/data/dates_null.sas7bdat
new file mode 100644
index 0000000000000..beadf1a34f42e
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/dates_null.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/datetime.csv b/pandas/pandas/tests/io/sas/data/datetime.csv
new file mode 100644
index 0000000000000..f0d82f7fc494e
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/datetime.csv
@@ -0,0 +1,5 @@
+Date1,Date2,DateTime,DateTimeHi,Taiw
+1677-09-22,1677-09-22,1677-09-21 00:12:44,1677-09-21 00:12:43.145225,1912-01-01
+1960-01-01,1960-01-01,1960-01-01 00:00:00,1960-01-01 00:00:00.000000,1960-01-01
+2016-02-29,2016-02-29,2016-02-29 23:59:59,2016-02-29 23:59:59.123456,2016-02-29
+2262-04-11,2262-04-11,2262-04-11 23:47:16,2262-04-11 23:47:16.854775,2262-04-11
diff --git a/pandas/pandas/tests/io/sas/data/datetime.sas7bdat b/pandas/pandas/tests/io/sas/data/datetime.sas7bdat
new file mode 100644
index 0000000000000..6469dbf29f8ee
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/datetime.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/load_log.sas7bdat b/pandas/pandas/tests/io/sas/data/load_log.sas7bdat
new file mode 100644
index 0000000000000..dc78925471baf
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/load_log.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/many_columns.csv b/pandas/pandas/tests/io/sas/data/many_columns.csv
new file mode 100644
index 0000000000000..307fc30f33b9f
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/many_columns.csv
@@ -0,0 +1,4 @@
+DATASRC,PDDOCID,age,agegt89,ASSESSA,ASSESS1,ASSESS3,ASSESS4,ASSESS5,ASSESS6,ASSESS7,week,BECK,conf1,conf2,conf3,demo3,demo4,demo5,demo6,demo7,demo11a,demo11b,demo11c,demo11d,derm1b,derm2,derm3,derm4,derm5a,derm5b,derm7,derm7a,derm7b,derm8,derm9,ECG3,ecgrtxt,ecgrhr,ecgrpr,ecgrqrs,ecgrqrsaxis,ecgrqt,ecgrqtc,ecgrrep,ecgrtime,mmse1,mmse2,mmse3,mmse4,mmse5,mmse6,mmse7,mmse8,mmse9,mmse10,mmse11,mmse12,mmse13,mmse14,mmse15,mmse16,mmse17,mmse18,mmse19,mmse20,mmse,mmsescor,mrf1,mrf2,mrf3,mrf4,mrf5,mrf6,mrf7,mrf8,mrf9,mrf10,mrf11,mrf12,mrf13,nvitl1s,nvitl1d,nvitl1r,nvitl2s,nvitl2d,nvitl2r,nvitl3s,nvitl3d,nvitl3r,nvitl4s,nvitl4d,nvitl4r,nvitl5,nvitl1,nvitl2,nvitl3,nvitl4,phys1,phys1a,phys14,phys15a,phys15b,phys15c,phys15d,phys16a,phys16b,phys16c,phys16d,phys17a,phys17b,phys17c,phys17d,phys18a,phys18b,phys18c,phys18d,phys19a,phys19b,phys20,phys22,phys24,phys26,phys28,PREG1,PREG2,updrsa,updrs1,updrs2,updrs3,updrs4,updrs5a,updrs6a,updrs7a,updrs8a,updrs9a,updrs10a,updrs11a,updrs12a,updrs13a,updrs14a,updrs15a,updrs16a,updrs17a,updrs18a,updrs19a,updrs20a1,updrs20b1,updrs20c1,updrs20d1,updrs20e1,updrs21a1,updrs21b1,updrs22a1,updrs22b1,updrs22c1,updrs22d1,updrs22e1,updrs23a1,updrs23b1,updrs24a1,updrs24b1,updrs25a1,updrs25b1,updrs26a1,updrs26b1,updrs26c1,updrs26d1,updrs27a,updrs28a,updrs29a,updrs30a,updrs31a,updrs32a,updrs33a,updrs34a,updrs35,updrs36,updrs37,updrs38,updrs39,updrs5b,updrs6b,updrs7b,updrs8b,updrs9b,updrs10b,updrs11b,updrs12b,updrs13b,updrs14b,updrs15b,updrs16b,updrs17b,updrs18b,updrs19b,updrs20a2,updrs20b2,updrs20c2,updrs20d2,updrs20e2,updrs21a2,updrs21b2,updrs22a2,updrs22b2,updrs22c2,updrs22d2,updrs22e2,updrs23a2,updrs23b2,updrs24a2,updrs24b2,updrs25a2,updrs25b2,updrs26a2,updrs26b2,updrs26c2,updrs26d2,updrs27b,updrs28b,updrs29b,updrs30b,updrs31b,updrs32b,updrs33b,updrs34b,updrs5c,updrs6c,updrs7c,updrs8c,updrs9c,updrs10c,updrs11c,updrs12c,updrs13c,updrs14c,updrs15c,updrs16c,updrs17c,updrs32c,updrs33c,updrs34c,updrsmental,updrsadl,updrsadlon,updrsadloff,updrsadlmin,updrstremor,updrstremortreat,updrstremormin,updrsrigid,updrsrigidtreat,updrsrigidmin,updrsmotor,updrsmotortreat,updrsmotormin,updrs,updrstrt,updrsmin,updrs4a,updrs41,updrs42,updrs43,updrs44,updrs45,updrs46,updrs47,updrs48,updrs49,updrs410,updrs411,vitl1s,vitl1d,vitl2,vitl3s,vitl3d,vitl4,vitl5,vitl6,assess,fbeck,conf,demo1,derm,ecg,ecgr,mrf,nvitl,fphys1,fpreg,fupdrs,fupdrs4,vitl,site,race,rImaged,rPD,rPDlt5,rAgeGt30,rHY,rMed,rMelanoma,rPreclude,rNeed,rEligible,gender,incsae,incsusp,incterm,increlated,inctermat,increason,incafter24,incendp,incres,disp2,disp3,disp4,disp6,inex1,inex2,inex3,inex4,inex5,inex6,inex7,inex8,inex9,inex10,inex11,inex12,inex13,inex14,inex15,inex16,inex17,inex18,inex19,inex20,inex21,inex22,inex23,inex24,inex25,inex26,inex27,inex28,treatment,treat,disp,inex,classify,enrollyr,demoyear,dob_yr,inexdays,demodays,onsetdays,diagdays,medstartdays,physdays,phys21dys,phys23dys,phys25dys,phys27dys,phys29dys,confdays,pregdays,nvitldays,nvitlscandays,vitldays,labdays,ecgdays,ecgtestdays,mrfdays,dermdays,dermexamdays,dermbiopdays,mmsedays,beckdays,updrdays,updr4days,assessdays,daystotherapy,dispdays,endpdys,termdys,SAEdys,resdys,lmeddys,wddays,VISIT_NO
+a030,ab304,43.0,0.0,0.0,0.0,,,,,,-2.0,0.0,1.0,1.0,,2.0,1.0,19.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,,,,,,,0.0,2.0,ABNORMAL,75.0,150.0,100.0,-3.0,410.0,460.0,2.0,1000.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,1.0,0.0,1.0,3.0,5.0,2.0,1.0,1.0,1.0,0.0,3.0,1.0,1.0,1.0,26.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,150.0,94.0,73.0,155.0,96.0,71.0,148.0,91.0,69.0,146.0,67.0,72.0,1.0,42840.0,46080.0,46980.0,30600.0,100.0,175.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,4.0,4.0,4.0,4.0,2.0,1.0,,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,0.0,0.0,0.5,0.0,0.0,0.0,1.0,1.0,2.0,2.0,1.0,1.5,0.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,1.0,1.0,1.0,2.5,95.0,95.0,7.0,,2.0,1.0,1.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.0,5.0,,,5.0,1.5,,1.5,7.5,,7.5,20.0,,20.0,25.0,,25.0,,,,,,,,,,,,,138.0,86.0,72.0,130.0,80.0,80.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,abc,1.0,1.0,1.0,0.0,1.0,34.0,5.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,1.0,,0.0,3.0,0.0,1.0,0.0,4.0,3.0,,1.0,1.0,1.0,1.0,1.0,1.0,,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,Placebo,1.0,1.0,1.0,1.0,2002.0,2002.0,1914.0,-28.0,-28.0,-404.0,-28.0,0.0,-28.0,,,,,-6.0,-28.0,-13.0,-13.0,-12.0,-28.0,-28.0,-28.0,-28.0,-28.0,-14.0,-14.0,,-28.0,-28.0,-28.0,,-28.0,,659.0,426.0,659.0,,,658.0,100.0,ab
+a030,ab304,43.0,0.0,0.0,0.0,,,,,,0.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1000.0,,,,,,,,,,,,,,,,,,,,,0.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,0.0,0.0,1.0,2.0,0.0,0.0,1.0,0.0,1.0,2.0,95.0,95.0,7.0,,2.0,1.0,2.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.0,3.0,,,3.0,0.0,,0.0,3.0,,3.0,13.0,,13.0,16.0,,16.0,,,,,,,,,,,,,140.0,86.0,76.0,132.0,80.0,84.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,abc,0.0,0.0,1.0,0.0,1.0,34.0,5.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,1.0,,0.0,3.0,0.0,1.0,0.0,4.0,3.0,,1.0,1.0,1.0,1.0,1.0,1.0,,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,Placebo,1.0,1.0,1.0,1.0,2002.0,,1914.0,-28.0,,,,0.0,,,,,,,,,,,0.0,0.0,,,,,,,,,0.0,,0.0,,659.0,426.0,659.0,,,658.0,100.0,ab
+a030,ab304,43.0,0.0,0.0,0.0,,,,,,4.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1000.0,,,,,,,,,,,,,,,,,,,,,0.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1.0,0.0,0.0,0.0,0.0,2.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.5,0.0,0.0,1.0,1.0,0.0,0.0,0.0,1.0,0.0,1.0,0.0,1.0,1.0,1.0,1.0,2.0,0.0,1.0,1.0,0.5,1.0,2.0,90.0,95.0,7.0,,2.0,2.0,2.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.0,5.0,,,5.0,0.5,,0.5,2.0,,2.0,16.0,,16.0,21.0,,21.0,0.0,,,,,,,,,,,,149.0,88.0,80.0,136.0,90.0,82.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,abc,0.0,0.0,1.0,1.0,1.0,34.0,5.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,1.0,,0.0,3.0,0.0,1.0,0.0,4.0,3.0,,1.0,1.0,1.0,1.0,1.0,1.0,,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,Placebo,1.0,1.0,1.0,1.0,2002.0,,1914.0,-28.0,,,,0.0,,,,,,,,,,,29.0,29.0,,,,,,,,,29.0,29.0,29.0,,659.0,426.0,659.0,,,658.0,100.0,ab
diff --git a/pandas/pandas/tests/io/sas/data/many_columns.sas7bdat b/pandas/pandas/tests/io/sas/data/many_columns.sas7bdat
new file mode 100644
index 0000000000000..582316fc59e18
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/many_columns.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/max_sas_date.sas7bdat b/pandas/pandas/tests/io/sas/data/max_sas_date.sas7bdat
new file mode 100644
index 0000000000000..b7838ebdcfeea
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/max_sas_date.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/paxraw_d_short.csv b/pandas/pandas/tests/io/sas/data/paxraw_d_short.csv
new file mode 100644
index 0000000000000..776799df5d8a2
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/paxraw_d_short.csv
@@ -0,0 +1,101 @@
+SEQN,PAXSTAT,PAXCAL,PAXDAY,PAXN,PAXHOUR,PAXMINUT,PAXINTEN,PAXSTEP
+31128,1,1,1,1,0,0,166,4
+31128,1,1,1,2,0,1,27,0
+31128,1,1,1,3,0,2,0,0
+31128,1,1,1,4,0,3,276,4
+31128,1,1,1,5,0,4,0,0
+31128,1,1,1,6,0,5,0,0
+31128,1,1,1,7,0,6,0,0
+31128,1,1,1,8,0,7,0,0
+31128,1,1,1,9,0,8,0,0
+31128,1,1,1,10,0,9,0,0
+31128,1,1,1,11,0,10,0,0
+31128,1,1,1,12,0,11,0,0
+31128,1,1,1,13,0,12,0,0
+31128,1,1,1,14,0,13,0,0
+31128,1,1,1,15,0,14,0,0
+31128,1,1,1,16,0,15,0,0
+31128,1,1,1,17,0,16,0,0
+31128,1,1,1,18,0,17,0,0
+31128,1,1,1,19,0,18,0,0
+31128,1,1,1,20,0,19,0,0
+31128,1,1,1,21,0,20,260,3
+31128,1,1,1,22,0,21,0,0
+31128,1,1,1,23,0,22,0,0
+31128,1,1,1,24,0,23,19,0
+31128,1,1,1,25,0,24,34,1
+31128,1,1,1,26,0,25,47,4
+31128,1,1,1,27,0,26,4,0
+31128,1,1,1,28,0,27,11,0
+31128,1,1,1,29,0,28,48,1
+31128,1,1,1,30,0,29,58,3
+31128,1,1,1,31,0,30,32,2
+31128,1,1,1,32,0,31,15,1
+31128,1,1,1,33,0,32,117,3
+31128,1,1,1,34,0,33,24,0
+31128,1,1,1,35,0,34,61,7
+31128,1,1,1,36,0,35,115,12
+31128,1,1,1,37,0,36,183,11
+31128,1,1,1,38,0,37,68,5
+31128,1,1,1,39,0,38,73,3
+31128,1,1,1,40,0,39,93,7
+31128,1,1,1,41,0,40,201,14
+31128,1,1,1,42,0,41,126,6
+31128,1,1,1,43,0,42,61,4
+31128,1,1,1,44,0,43,97,7
+31128,1,1,1,45,0,44,62,3
+31128,1,1,1,46,0,45,77,10
+31128,1,1,1,47,0,46,105,8
+31128,1,1,1,48,0,47,209,12
+31128,1,1,1,49,0,48,72,4
+31128,1,1,1,50,0,49,50,1
+31128,1,1,1,51,0,50,324,7
+31128,1,1,1,52,0,51,582,16
+31128,1,1,1,53,0,52,387,31
+31128,1,1,1,54,0,53,780,54
+31128,1,1,1,55,0,54,618,10
+31128,1,1,1,56,0,55,0,0
+31128,1,1,1,57,0,56,0,0
+31128,1,1,1,58,0,57,0,0
+31128,1,1,1,59,0,58,123,1
+31128,1,1,1,60,0,59,0,0
+31128,1,1,1,61,1,0,0,0
+31128,1,1,1,62,1,1,0,0
+31128,1,1,1,63,1,2,0,0
+31128,1,1,1,64,1,3,0,0
+31128,1,1,1,65,1,4,0,0
+31128,1,1,1,66,1,5,0,0
+31128,1,1,1,67,1,6,0,0
+31128,1,1,1,68,1,7,0,0
+31128,1,1,1,69,1,8,0,0
+31128,1,1,1,70,1,9,0,0
+31128,1,1,1,71,1,10,0,0
+31128,1,1,1,72,1,11,0,0
+31128,1,1,1,73,1,12,0,0
+31128,1,1,1,74,1,13,0,0
+31128,1,1,1,75,1,14,0,0
+31128,1,1,1,76,1,15,0,0
+31128,1,1,1,77,1,16,0,0
+31128,1,1,1,78,1,17,0,0
+31128,1,1,1,79,1,18,0,0
+31128,1,1,1,80,1,19,0,0
+31128,1,1,1,81,1,20,0,0
+31128,1,1,1,82,1,21,0,0
+31128,1,1,1,83,1,22,0,0
+31128,1,1,1,84,1,23,0,0
+31128,1,1,1,85,1,24,0,0
+31128,1,1,1,86,1,25,0,0
+31128,1,1,1,87,1,26,0,0
+31128,1,1,1,88,1,27,0,0
+31128,1,1,1,89,1,28,0,0
+31128,1,1,1,90,1,29,0,0
+31128,1,1,1,91,1,30,0,0
+31128,1,1,1,92,1,31,0,0
+31128,1,1,1,93,1,32,0,0
+31128,1,1,1,94,1,33,0,0
+31128,1,1,1,95,1,34,2,0
+31128,1,1,1,96,1,35,0,0
+31128,1,1,1,97,1,36,0,0
+31128,1,1,1,98,1,37,0,0
+31128,1,1,1,99,1,38,0,0
+31128,1,1,1,100,1,39,0,0
diff --git a/pandas/pandas/tests/io/sas/data/productsales.csv b/pandas/pandas/tests/io/sas/data/productsales.csv
new file mode 100644
index 0000000000000..1f6a4424e1a97
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/productsales.csv
@@ -0,0 +1,1441 @@
+ACTUAL,PREDICT,COUNTRY,REGION,DIVISION,PRODTYPE,PRODUCT,QUARTER,YEAR,MONTH
+925,850,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+999,297,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+608,846,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+642,533,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+656,646,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+948,486,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+612,717,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+114,564,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+685,230,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+657,494,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+608,903,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+353,266,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+107,190,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+354,139,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+101,217,CANADA,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+553,560,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+877,148,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+431,762,CANADA,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+511,457,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+157,532,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+520,629,CANADA,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+114,491,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+277,0,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+561,979,CANADA,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+220,585,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+444,267,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+178,487,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+756,764,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+329,312,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+910,531,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+530,536,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+101,773,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+515,143,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+730,126,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+993,862,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+954,754,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+267,410,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+347,701,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+991,204,CANADA,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+923,509,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+437,378,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+737,507,CANADA,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+104,49,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+840,876,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+704,66,CANADA,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+889,819,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+107,351,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+571,201,CANADA,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+688,209,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+544,51,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+954,135,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+445,47,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+829,379,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+464,758,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+968,475,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+842,343,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+721,507,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+966,269,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+332,699,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+328,824,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+355,497,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+506,44,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+585,522,CANADA,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+634,378,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+662,689,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+783,90,CANADA,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+786,720,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+710,343,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+950,457,CANADA,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+274,947,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+406,834,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+515,71,CANADA,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+35,282,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+995,538,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+670,679,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+406,601,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+825,577,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+467,908,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+709,819,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+522,687,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+688,157,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+956,111,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+129,31,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+687,790,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+877,795,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+845,379,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+425,114,CANADA,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+899,475,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+987,747,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+641,372,CANADA,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+448,415,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+341,955,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+137,356,CANADA,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+235,316,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+482,351,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+678,164,CANADA,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+240,386,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+605,113,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+274,68,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+422,885,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+763,575,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+561,743,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+339,816,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+877,203,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+192,581,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+604,815,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+55,333,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+87,40,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+942,672,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+912,23,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+768,948,CANADA,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+951,291,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+768,839,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+978,864,CANADA,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+20,337,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+298,95,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+193,535,CANADA,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+336,191,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+617,412,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+709,711,CANADA,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+5,425,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+164,215,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+422,948,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+424,544,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+854,764,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+168,446,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+8,957,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+748,967,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+682,11,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+300,110,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+672,263,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+894,215,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+944,965,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+403,423,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+596,753,CANADA,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+481,770,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+503,263,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+126,79,CANADA,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+721,441,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+271,858,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+721,667,CANADA,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+157,193,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+991,394,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+499,680,CANADA,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+284,414,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+705,770,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+737,679,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+745,7,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+633,713,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+983,851,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+591,944,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+42,130,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+771,485,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+465,23,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+296,193,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+890,7,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+312,919,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+777,768,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+364,854,CANADA,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+601,411,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+823,736,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+847,10,CANADA,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+490,311,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+387,348,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+688,458,CANADA,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+650,195,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+447,658,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+91,704,CANADA,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+197,807,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+51,861,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+570,873,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+423,933,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+524,355,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+416,794,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+789,645,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+551,700,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+400,831,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+361,800,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+189,830,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+554,828,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+585,12,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+281,501,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+629,914,CANADA,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+43,685,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+533,755,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+882,708,CANADA,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+790,595,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+600,32,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+148,49,CANADA,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+237,727,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+488,239,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+457,273,CANADA,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+401,986,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+181,544,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+995,182,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+120,197,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+119,435,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+319,974,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+333,524,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+923,688,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+634,750,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+493,155,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+461,860,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+304,102,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+641,425,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+992,224,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+202,408,CANADA,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+770,524,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+202,816,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+14,515,CANADA,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+134,793,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+977,460,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+174,732,CANADA,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+429,435,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+514,38,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+784,616,CANADA,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+973,225,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+511,402,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+30,697,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+895,567,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+557,231,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+282,372,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+909,15,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+276,866,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+234,452,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+479,663,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+782,982,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+755,813,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+689,523,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+496,871,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+24,511,CANADA,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+379,819,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+441,525,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+49,13,CANADA,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+243,694,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+295,782,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+395,839,CANADA,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+929,461,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+997,303,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+889,421,CANADA,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
+72,421,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+926,433,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+850,394,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+826,338,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+651,764,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+854,216,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+899,96,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+309,550,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+943,636,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+138,427,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+99,652,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+270,478,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+862,18,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+574,40,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+359,453,CANADA,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+958,987,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+791,26,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+284,101,CANADA,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+190,969,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+527,492,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+112,263,CANADA,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+271,593,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+643,923,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+554,146,CANADA,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+211,305,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+368,318,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+778,417,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+808,623,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+46,761,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+466,272,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+18,988,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+87,821,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+765,962,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+62,615,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+13,523,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+775,806,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+636,586,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+458,520,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+206,908,CANADA,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+310,30,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+813,247,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+22,647,CANADA,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+742,55,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+394,154,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+957,344,CANADA,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+205,95,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+198,665,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+638,145,CANADA,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+155,925,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+688,395,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+730,749,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+208,279,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+525,288,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+483,509,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+748,255,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+6,214,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+168,473,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+301,702,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+9,814,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+778,231,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+799,422,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+309,572,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+433,363,CANADA,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+969,919,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+181,355,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+787,992,CANADA,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+971,147,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+440,183,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+209,375,CANADA,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+537,77,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+364,308,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+377,660,CANADA,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+251,555,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+607,455,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+127,888,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+513,652,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+146,799,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+917,249,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+776,539,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+330,198,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+981,340,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+862,152,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+612,347,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+607,565,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+786,855,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+160,87,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+199,69,CANADA,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+972,807,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+870,565,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+494,798,CANADA,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+975,714,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+760,17,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+180,797,CANADA,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+256,422,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+422,621,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+859,661,CANADA,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+586,363,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+441,910,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+597,998,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+717,95,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+713,731,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+591,718,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+492,467,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+170,126,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+684,127,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+981,746,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+966,878,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+439,27,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+151,569,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+602,812,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+187,603,CANADA,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+415,506,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+61,185,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+839,692,CANADA,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+596,565,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+751,512,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+460,86,CANADA,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+922,399,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+153,672,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+928,801,CANADA,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+951,730,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+394,408,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+615,982,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+653,499,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+180,307,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+649,741,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+921,640,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+11,300,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+696,929,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+795,309,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+550,340,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+320,228,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+845,1000,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+245,21,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+142,583,CANADA,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+717,506,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+3,405,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+790,556,CANADA,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+646,72,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+230,103,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+938,262,CANADA,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+629,102,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+317,841,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+812,159,CANADA,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+141,570,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+64,375,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+207,298,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+435,32,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+96,760,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+252,338,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+956,149,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+633,343,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+190,151,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+227,44,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+24,583,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+420,230,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+910,907,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+709,783,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+810,117,CANADA,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+723,416,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+911,318,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+230,888,CANADA,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+448,60,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+945,596,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+508,576,CANADA,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+262,576,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+441,280,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+15,219,CANADA,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+795,133,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+301,273,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+304,86,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+49,400,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+576,364,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+669,63,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+325,929,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+272,344,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+80,768,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+46,668,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+223,407,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+774,536,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+784,657,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+92,215,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+67,966,CANADA,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+747,674,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+686,574,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+93,266,CANADA,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+192,680,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+51,362,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+498,412,CANADA,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+546,431,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+485,94,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+925,345,CANADA,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+292,445,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+540,632,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+21,855,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+100,36,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+49,250,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+353,427,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+911,367,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+823,245,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+278,893,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+576,490,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+655,88,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+763,964,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+88,62,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+746,506,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+927,680,CANADA,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+297,153,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+291,403,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+838,98,CANADA,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+112,376,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+509,477,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+472,50,CANADA,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+495,592,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+1000,813,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+241,740,CANADA,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+693,873,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+903,459,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+791,224,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+108,562,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+845,199,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+452,275,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+479,355,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+410,947,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+379,454,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+740,450,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+471,575,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+325,6,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+455,847,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+563,338,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+879,517,CANADA,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+312,630,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+587,381,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+628,864,CANADA,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+486,416,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+811,852,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+990,815,CANADA,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+35,23,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+764,527,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+619,693,CANADA,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
+996,977,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+554,549,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+540,951,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+140,390,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+554,204,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+724,78,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+693,613,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+866,745,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+833,56,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+164,887,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+753,651,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+60,691,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+688,767,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+883,709,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+109,417,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+950,326,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+438,599,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+286,818,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+342,13,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+383,185,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+80,140,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+322,717,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+749,852,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+606,125,GERMANY,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+641,325,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+494,648,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+428,365,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+936,120,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+597,347,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+728,638,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+933,732,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+663,465,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+394,262,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+334,947,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+114,694,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+89,482,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+874,600,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+674,94,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+347,323,GERMANY,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+105,49,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+286,70,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+669,844,GERMANY,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+786,773,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+104,68,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+770,110,GERMANY,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+263,42,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+900,171,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+630,644,GERMANY,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+597,408,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+185,45,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+175,522,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+576,166,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+957,885,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+993,713,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+500,838,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+410,267,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+592,967,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+64,529,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+208,656,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+273,665,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+906,419,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+429,776,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+961,971,GERMANY,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+338,248,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+472,486,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+903,674,GERMANY,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+299,603,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+948,492,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+931,512,GERMANY,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+570,391,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+97,313,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+674,758,GERMANY,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+468,304,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+430,846,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+893,912,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+519,810,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+267,122,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+908,102,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+176,161,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+673,450,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+798,215,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+291,765,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+583,557,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+442,739,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+951,811,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+430,780,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+559,645,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+726,365,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+944,597,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+497,126,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+388,655,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+81,604,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+111,280,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+288,115,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+845,205,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+745,672,GERMANY,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+352,339,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+234,70,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+167,528,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+606,220,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+670,691,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+764,197,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+659,239,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+996,50,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+424,135,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+899,972,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+392,475,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+555,868,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+860,451,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+114,565,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+943,116,GERMANY,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+365,385,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+249,375,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+192,357,GERMANY,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+328,230,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+311,829,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+576,971,GERMANY,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+915,280,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+522,853,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+625,953,GERMANY,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+873,874,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+498,578,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+808,768,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+742,178,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+744,916,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+30,917,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+747,633,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+672,107,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+564,523,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+785,924,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+825,481,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+243,240,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+959,819,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+123,602,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+714,538,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+252,632,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+715,952,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+670,480,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+81,700,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+653,726,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+795,526,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+182,410,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+725,307,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+101,73,GERMANY,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+143,232,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+15,993,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+742,652,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+339,761,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+39,428,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+465,4,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+889,101,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+856,869,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+358,271,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+452,633,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+387,481,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+824,302,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+185,245,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+151,941,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+419,721,GERMANY,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+643,893,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+63,898,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+202,94,GERMANY,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+332,962,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+723,71,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+148,108,GERMANY,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+840,71,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+601,767,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+962,323,GERMANY,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+166,982,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+531,614,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+963,839,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+994,388,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+978,296,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+72,429,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+33,901,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+428,350,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+413,581,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+737,583,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+85,92,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+916,647,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+785,771,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+302,26,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+1000,598,GERMANY,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+458,715,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+896,74,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+615,580,GERMANY,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+174,848,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+651,118,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+784,54,GERMANY,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+121,929,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+341,393,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+615,820,GERMANY,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+697,336,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+215,299,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+197,747,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+205,154,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+256,486,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+377,251,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+577,225,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+686,77,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+332,74,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+534,596,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+485,493,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+594,782,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+413,487,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+13,127,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+483,538,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+820,94,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+745,252,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+79,722,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+36,536,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+950,958,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+74,466,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+458,309,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+609,680,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+429,539,GERMANY,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+956,511,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+205,505,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+629,720,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+277,823,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+266,21,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+872,142,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+435,95,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+988,398,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+953,328,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+556,151,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+211,978,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+389,918,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+351,542,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+14,96,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+181,496,GERMANY,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+452,77,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+511,236,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+193,913,GERMANY,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+797,49,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+988,967,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+487,502,GERMANY,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+941,790,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+577,121,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+456,55,GERMANY,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
+982,739,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+593,683,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+702,610,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+528,248,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+873,530,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+301,889,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+769,245,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+724,473,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+466,938,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+774,150,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+111,772,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+954,201,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+780,945,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+210,177,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+93,378,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+332,83,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+186,803,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+782,398,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+41,215,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+222,194,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+992,287,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+477,410,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+948,50,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+817,204,GERMANY,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+597,239,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+649,637,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+3,938,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+731,788,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+181,399,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+468,576,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+891,187,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+226,703,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+28,455,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+609,244,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+224,868,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+230,353,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+216,101,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+282,924,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+501,144,GERMANY,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+320,0,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+720,910,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+464,259,GERMANY,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+363,107,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+49,63,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+223,270,GERMANY,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+452,554,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+210,154,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+444,205,GERMANY,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+222,441,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+678,183,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+25,459,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+57,810,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+981,268,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+740,916,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+408,742,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+966,522,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+107,299,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+488,677,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+759,709,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+504,310,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+99,160,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+503,698,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+724,540,GERMANY,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+309,901,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+625,34,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+294,536,GERMANY,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+890,780,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+501,716,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+34,532,GERMANY,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+203,871,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+140,199,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+845,845,GERMANY,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+774,591,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+645,378,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+986,942,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+296,686,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+936,720,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+341,546,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+32,845,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+277,667,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+548,627,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+727,142,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+812,655,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+168,556,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+150,459,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+136,89,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+695,726,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+363,38,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+853,60,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+621,369,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+764,381,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+669,465,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+772,981,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+228,758,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+261,31,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+821,237,GERMANY,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+100,285,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+465,94,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+350,561,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+991,143,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+910,95,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+206,341,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+263,388,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+374,272,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+875,890,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+810,734,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+398,364,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+565,619,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+417,517,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+291,781,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+251,327,GERMANY,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+449,48,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+774,809,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+386,73,GERMANY,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+22,936,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+940,400,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+132,736,GERMANY,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+103,211,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+152,271,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+952,855,GERMANY,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+872,923,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+748,854,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+749,769,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+876,271,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+860,383,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+900,29,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+705,185,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+913,351,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+315,560,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+466,840,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+233,517,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+906,949,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+148,633,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+661,636,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+847,138,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+768,481,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+866,408,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+475,130,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+112,813,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+136,661,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+763,311,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+388,872,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+996,643,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+486,174,GERMANY,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+494,528,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+771,124,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+49,126,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+322,440,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+878,881,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+827,292,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+852,873,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+716,357,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+81,247,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+916,18,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+673,395,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+242,620,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+914,946,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+902,72,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+707,691,GERMANY,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+223,95,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+619,878,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+254,757,GERMANY,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+688,898,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+477,172,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+280,419,GERMANY,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+546,849,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+630,807,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+455,599,GERMANY,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+505,59,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+823,790,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+891,574,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+840,96,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+436,376,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+168,352,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+177,741,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+727,12,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+278,157,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+443,10,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+905,544,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+881,817,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+507,754,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+363,425,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+603,492,GERMANY,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+473,485,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+128,369,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+105,560,GERMANY,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+325,651,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+711,326,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+983,180,GERMANY,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+241,935,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+71,403,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+395,345,GERMANY,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+168,278,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+512,376,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+291,104,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+776,543,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+271,798,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+946,333,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+195,833,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+165,132,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+238,629,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+409,337,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+720,300,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+309,470,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+812,875,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+441,237,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+500,272,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+517,860,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+924,415,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+572,140,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+768,367,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+692,195,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+28,245,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+202,285,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+76,98,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+421,932,GERMANY,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+636,898,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+52,330,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+184,603,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+739,280,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+841,507,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+65,202,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+623,513,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+517,132,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+636,21,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+845,657,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+232,195,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+26,323,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+680,299,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+364,811,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+572,739,GERMANY,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+145,889,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+644,189,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+87,698,GERMANY,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+620,646,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+535,562,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+661,753,GERMANY,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+884,425,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+689,693,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+646,941,GERMANY,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
+4,975,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+813,455,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+773,260,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+205,69,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+657,147,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+154,533,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+747,881,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+787,457,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+867,441,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+307,859,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+571,177,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+92,633,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+269,382,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+764,707,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+662,566,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+818,349,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+617,128,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+649,231,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+895,258,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+750,812,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+738,362,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+107,133,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+278,60,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+32,88,U.S.A.,EAST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+129,378,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+187,569,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+670,186,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+678,875,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+423,636,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+389,360,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+257,677,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+780,708,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+159,158,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+97,384,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+479,927,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+9,134,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+614,273,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+261,27,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+115,209,U.S.A.,EAST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+358,470,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+133,219,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+891,907,U.S.A.,EAST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+702,778,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+58,998,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+606,194,U.S.A.,EAST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+668,933,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+813,708,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+450,949,U.S.A.,EAST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+956,579,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+276,131,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+889,689,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+708,908,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+14,524,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+904,336,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+272,916,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+257,236,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+343,965,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+80,350,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+530,599,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+340,901,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+595,935,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+47,667,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+279,104,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+293,803,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+162,64,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+935,825,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+689,839,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+484,184,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+230,348,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+164,904,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+401,219,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+607,381,U.S.A.,EAST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+229,524,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+786,902,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+92,212,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+455,762,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+409,182,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+166,442,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+277,919,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+92,67,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+631,741,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+390,617,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+403,214,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+964,202,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+223,788,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+684,639,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+645,336,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+470,937,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+424,399,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+862,21,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+736,125,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+554,635,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+790,229,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+115,770,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+853,622,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+643,109,U.S.A.,EAST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+794,975,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+892,820,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+728,123,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+744,135,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+678,535,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+768,971,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+234,166,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+333,814,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+968,557,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+119,820,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+469,486,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+261,429,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+984,65,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+845,977,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+374,410,U.S.A.,EAST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+687,150,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+157,630,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+49,488,U.S.A.,EAST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+817,112,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+223,598,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+433,705,U.S.A.,EAST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+41,226,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+396,979,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+131,19,U.S.A.,EAST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+521,204,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+751,805,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+45,549,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+144,912,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+119,427,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+728,1,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+120,540,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+657,940,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+409,644,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+881,821,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+113,560,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+831,309,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+129,1000,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+76,945,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+260,931,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+882,504,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+157,950,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+443,278,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+111,225,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+497,6,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+321,124,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+194,206,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+684,320,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+634,270,U.S.A.,EAST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+622,278,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+689,447,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+120,170,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+374,87,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+926,384,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+687,574,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+600,585,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+779,947,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+223,984,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+628,189,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+326,364,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+836,49,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+361,851,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+444,643,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+501,143,U.S.A.,EAST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+743,763,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+861,987,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+203,264,U.S.A.,EAST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+762,439,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+705,750,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+153,37,U.S.A.,EAST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+436,95,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+428,79,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+804,832,U.S.A.,EAST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+805,649,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+860,838,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+104,439,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+434,207,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+912,804,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+571,875,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+267,473,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+415,845,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+261,91,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+746,630,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+30,185,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+662,317,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+916,88,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+415,607,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+514,35,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+756,680,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+461,78,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+460,117,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+305,440,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+198,652,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+234,249,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+638,658,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+88,563,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+751,737,U.S.A.,EAST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+816,789,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+437,988,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+715,220,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+780,946,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+245,986,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+201,129,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+815,433,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+865,492,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+634,306,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+901,154,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+789,206,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+882,81,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+953,882,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+862,848,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+628,664,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+765,389,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+741,182,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+61,505,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+470,861,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+869,263,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+650,400,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+750,556,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+602,497,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+54,181,U.S.A.,EAST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+384,619,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+161,332,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+977,669,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+615,487,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+783,994,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+977,331,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+375,739,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+298,665,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+104,921,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+713,862,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+556,662,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+323,517,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+391,352,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+593,166,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+906,859,U.S.A.,EAST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+130,571,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+613,976,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+58,466,U.S.A.,EAST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+314,79,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+67,864,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+654,623,U.S.A.,EAST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+312,170,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+349,662,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+415,763,U.S.A.,EAST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
+404,896,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-01-01
+22,973,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-02-01
+744,161,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1993,1993-03-01
+804,934,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-04-01
+101,697,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-05-01
+293,116,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1993,1993-06-01
+266,84,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-07-01
+372,604,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-08-01
+38,371,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1993,1993-09-01
+385,783,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-10-01
+262,335,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-11-01
+961,321,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1993,1993-12-01
+831,177,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-01-01
+579,371,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-02-01
+301,583,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,1,1994,1994-03-01
+693,364,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-04-01
+895,343,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-05-01
+320,854,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,2,1994,1994-06-01
+284,691,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-07-01
+362,387,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-08-01
+132,298,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,3,1994,1994-09-01
+42,635,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-10-01
+118,81,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-11-01
+42,375,U.S.A.,WEST,EDUCATION,FURNITURE,SOFA,4,1994,1994-12-01
+18,846,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-01-01
+512,933,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-02-01
+337,237,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1993,1993-03-01
+167,964,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-04-01
+749,382,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-05-01
+890,610,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1993,1993-06-01
+910,148,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-07-01
+403,837,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-08-01
+403,85,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1993,1993-09-01
+661,425,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-10-01
+485,633,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-11-01
+789,515,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1993,1993-12-01
+415,512,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-01-01
+418,156,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-02-01
+163,464,U.S.A.,WEST,EDUCATION,FURNITURE,BED,1,1994,1994-03-01
+298,813,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-04-01
+584,455,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-05-01
+797,366,U.S.A.,WEST,EDUCATION,FURNITURE,BED,2,1994,1994-06-01
+767,734,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-07-01
+984,451,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-08-01
+388,134,U.S.A.,WEST,EDUCATION,FURNITURE,BED,3,1994,1994-09-01
+924,547,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-10-01
+566,802,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-11-01
+390,61,U.S.A.,WEST,EDUCATION,FURNITURE,BED,4,1994,1994-12-01
+608,556,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-01-01
+840,202,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-02-01
+112,964,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1993,1993-03-01
+288,112,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-04-01
+408,445,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-05-01
+876,884,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1993,1993-06-01
+224,348,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-07-01
+133,564,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-08-01
+662,568,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1993,1993-09-01
+68,882,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-10-01
+626,542,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-11-01
+678,119,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1993,1993-12-01
+361,248,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-01-01
+464,868,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-02-01
+681,841,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,1,1994,1994-03-01
+377,484,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-04-01
+222,986,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-05-01
+972,39,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,2,1994,1994-06-01
+56,930,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-07-01
+695,252,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-08-01
+908,794,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,3,1994,1994-09-01
+328,658,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-10-01
+891,139,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-11-01
+265,331,U.S.A.,WEST,EDUCATION,OFFICE,TABLE,4,1994,1994-12-01
+251,261,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-01-01
+783,122,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-02-01
+425,296,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1993,1993-03-01
+859,391,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-04-01
+314,75,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-05-01
+153,731,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1993,1993-06-01
+955,883,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-07-01
+654,707,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-08-01
+693,97,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1993,1993-09-01
+757,390,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-10-01
+221,237,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-11-01
+942,496,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1993,1993-12-01
+31,814,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-01-01
+540,765,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-02-01
+352,308,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,1,1994,1994-03-01
+904,327,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-04-01
+436,266,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-05-01
+281,699,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,2,1994,1994-06-01
+801,599,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-07-01
+273,950,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-08-01
+716,117,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,3,1994,1994-09-01
+902,632,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-10-01
+341,35,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-11-01
+155,562,U.S.A.,WEST,EDUCATION,OFFICE,CHAIR,4,1994,1994-12-01
+796,144,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-01-01
+257,142,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-02-01
+611,273,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1993,1993-03-01
+6,915,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-04-01
+125,920,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-05-01
+745,294,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1993,1993-06-01
+437,681,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-07-01
+906,86,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-08-01
+844,764,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1993,1993-09-01
+413,269,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-10-01
+869,138,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-11-01
+403,834,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1993,1993-12-01
+137,112,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-01-01
+922,921,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-02-01
+202,859,U.S.A.,WEST,EDUCATION,OFFICE,DESK,1,1994,1994-03-01
+955,442,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-04-01
+781,593,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-05-01
+12,346,U.S.A.,WEST,EDUCATION,OFFICE,DESK,2,1994,1994-06-01
+931,312,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-07-01
+95,690,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-08-01
+795,344,U.S.A.,WEST,EDUCATION,OFFICE,DESK,3,1994,1994-09-01
+542,784,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-10-01
+935,639,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-11-01
+269,726,U.S.A.,WEST,EDUCATION,OFFICE,DESK,4,1994,1994-12-01
+197,596,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-01-01
+828,263,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-02-01
+461,194,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1993,1993-03-01
+35,895,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-04-01
+88,502,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-05-01
+832,342,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1993,1993-06-01
+900,421,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-07-01
+368,901,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-08-01
+201,474,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1993,1993-09-01
+758,571,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-10-01
+504,511,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-11-01
+864,379,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1993,1993-12-01
+574,68,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-01-01
+61,210,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-02-01
+565,478,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,1,1994,1994-03-01
+475,296,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-04-01
+44,664,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-05-01
+145,880,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,2,1994,1994-06-01
+813,607,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-07-01
+703,97,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-08-01
+757,908,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,3,1994,1994-09-01
+96,152,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-10-01
+860,622,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-11-01
+750,309,U.S.A.,WEST,CONSUMER,FURNITURE,SOFA,4,1994,1994-12-01
+585,912,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-01-01
+127,429,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-02-01
+669,580,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1993,1993-03-01
+708,179,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-04-01
+830,119,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-05-01
+550,369,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1993,1993-06-01
+762,882,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-07-01
+468,727,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-08-01
+151,823,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1993,1993-09-01
+103,783,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-10-01
+876,884,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-11-01
+881,891,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1993,1993-12-01
+116,909,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-01-01
+677,765,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-02-01
+477,180,U.S.A.,WEST,CONSUMER,FURNITURE,BED,1,1994,1994-03-01
+154,712,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-04-01
+331,175,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-05-01
+784,869,U.S.A.,WEST,CONSUMER,FURNITURE,BED,2,1994,1994-06-01
+563,820,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-07-01
+229,554,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-08-01
+451,126,U.S.A.,WEST,CONSUMER,FURNITURE,BED,3,1994,1994-09-01
+974,760,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-10-01
+484,446,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-11-01
+69,254,U.S.A.,WEST,CONSUMER,FURNITURE,BED,4,1994,1994-12-01
+755,516,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-01-01
+331,779,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-02-01
+482,987,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1993,1993-03-01
+632,318,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-04-01
+750,427,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-05-01
+618,86,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1993,1993-06-01
+935,553,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-07-01
+716,315,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-08-01
+205,328,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1993,1993-09-01
+215,521,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-10-01
+871,156,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-11-01
+552,841,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1993,1993-12-01
+619,623,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-01-01
+701,849,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-02-01
+104,438,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,1,1994,1994-03-01
+114,719,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-04-01
+854,906,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-05-01
+563,267,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,2,1994,1994-06-01
+73,542,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-07-01
+427,552,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-08-01
+348,428,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,3,1994,1994-09-01
+148,158,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-10-01
+895,379,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-11-01
+394,142,U.S.A.,WEST,CONSUMER,OFFICE,TABLE,4,1994,1994-12-01
+792,588,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-01-01
+175,506,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-02-01
+208,382,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1993,1993-03-01
+354,132,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-04-01
+163,652,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-05-01
+336,723,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1993,1993-06-01
+804,682,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-07-01
+863,382,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-08-01
+326,125,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1993,1993-09-01
+568,321,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-10-01
+691,922,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-11-01
+152,884,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1993,1993-12-01
+565,38,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-01-01
+38,194,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-02-01
+185,996,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,1,1994,1994-03-01
+318,532,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-04-01
+960,391,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-05-01
+122,104,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,2,1994,1994-06-01
+400,22,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-07-01
+301,650,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-08-01
+909,143,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,3,1994,1994-09-01
+433,999,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-10-01
+508,415,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-11-01
+648,350,U.S.A.,WEST,CONSUMER,OFFICE,CHAIR,4,1994,1994-12-01
+793,342,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-01-01
+129,215,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-02-01
+481,52,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1993,1993-03-01
+406,292,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-04-01
+512,862,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-05-01
+668,309,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1993,1993-06-01
+551,886,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-07-01
+124,172,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-08-01
+655,912,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1993,1993-09-01
+523,666,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-10-01
+739,656,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-11-01
+87,145,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1993,1993-12-01
+890,664,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-01-01
+665,639,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-02-01
+329,707,U.S.A.,WEST,CONSUMER,OFFICE,DESK,1,1994,1994-03-01
+417,891,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-04-01
+828,466,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-05-01
+298,451,U.S.A.,WEST,CONSUMER,OFFICE,DESK,2,1994,1994-06-01
+356,451,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-07-01
+909,874,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-08-01
+251,805,U.S.A.,WEST,CONSUMER,OFFICE,DESK,3,1994,1994-09-01
+526,426,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-10-01
+652,932,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-11-01
+573,581,U.S.A.,WEST,CONSUMER,OFFICE,DESK,4,1994,1994-12-01
diff --git a/pandas/pandas/tests/io/sas/data/productsales.sas7bdat b/pandas/pandas/tests/io/sas/data/productsales.sas7bdat
new file mode 100644
index 0000000000000..6f18c5a048115
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/productsales.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test1.sas7bdat b/pandas/pandas/tests/io/sas/data/test1.sas7bdat
new file mode 100644
index 0000000000000..951173ce4d9f9
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test1.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test10.sas7bdat b/pandas/pandas/tests/io/sas/data/test10.sas7bdat
new file mode 100644
index 0000000000000..a5fd43e6cb9ac
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test10.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test11.sas7bdat b/pandas/pandas/tests/io/sas/data/test11.sas7bdat
new file mode 100644
index 0000000000000..072aa683f66d9
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test11.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test12.sas7bdat b/pandas/pandas/tests/io/sas/data/test12.sas7bdat
new file mode 100644
index 0000000000000..e2a9db874948d
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test12.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test13.sas7bdat b/pandas/pandas/tests/io/sas/data/test13.sas7bdat
new file mode 100644
index 0000000000000..b1dc6f9f8eddc
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test13.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test14.sas7bdat b/pandas/pandas/tests/io/sas/data/test14.sas7bdat
new file mode 100644
index 0000000000000..5a958df51f0ce
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test14.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test15.sas7bdat b/pandas/pandas/tests/io/sas/data/test15.sas7bdat
new file mode 100644
index 0000000000000..c028d8041a3d3
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test15.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test16.sas7bdat b/pandas/pandas/tests/io/sas/data/test16.sas7bdat
new file mode 100644
index 0000000000000..867c3c51bbddd
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test16.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test2.sas7bdat b/pandas/pandas/tests/io/sas/data/test2.sas7bdat
new file mode 100644
index 0000000000000..ba0b8e8dcbb91
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test2.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test3.sas7bdat b/pandas/pandas/tests/io/sas/data/test3.sas7bdat
new file mode 100644
index 0000000000000..a061b1ddd0d45
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test3.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test4.sas7bdat b/pandas/pandas/tests/io/sas/data/test4.sas7bdat
new file mode 100644
index 0000000000000..addd6edf90830
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test4.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test5.sas7bdat b/pandas/pandas/tests/io/sas/data/test5.sas7bdat
new file mode 100644
index 0000000000000..ba741d5a635df
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test5.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test6.sas7bdat b/pandas/pandas/tests/io/sas/data/test6.sas7bdat
new file mode 100644
index 0000000000000..2d9b4b0466047
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test6.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test7.sas7bdat b/pandas/pandas/tests/io/sas/data/test7.sas7bdat
new file mode 100644
index 0000000000000..785b12cf175e3
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test7.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test8.sas7bdat b/pandas/pandas/tests/io/sas/data/test8.sas7bdat
new file mode 100644
index 0000000000000..67db5a143de07
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test8.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test9.sas7bdat b/pandas/pandas/tests/io/sas/data/test9.sas7bdat
new file mode 100644
index 0000000000000..d76a1f28033f4
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test9.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test_12659.csv b/pandas/pandas/tests/io/sas/data/test_12659.csv
new file mode 100644
index 0000000000000..cd83f7caa6aaf
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/test_12659.csv
@@ -0,0 +1,37 @@
+yearmonth,useGpCo,useGpVi,useSpec,useUrge,caseGpCo,caseGpVi,caseSpec,caseUrge,expendGpCo,expendGpVi,expendSpec,expendUrge
+201401,11,12,13,14,15,16,17,18,19,20,21,22
+201402,11,12,13,14,15,16,17,18,19,20,21,22
+201403,11,12,13,14,15,16,17,18,19,20,21,22
+201404,11,12,13,14,15,16,17,18,19,20,21,22
+201405,11,12,13,14,15,16,17,18,19,20,21,22
+201406,11,12,13,14,15,16,17,18,19,20,21,22
+201407,11,12,13,14,15,16,17,18,19,20,21,22
+201408,11,12,13,14,15,16,17,18,19,20,21,22
+201409,11,12,13,14,15,16,17,18,19,20,21,22
+201410,11,12,13,14,15,16,17,18,19,20,21,22
+201411,11,12,13,14,15,16,17,18,19,20,21,22
+201412,11,12,13,14,15,16,17,18,19,20,21,22
+201501,11,12,13,14,15,16,17,18,19,20,21,22
+201502,11,12,13,14,15,16,17,18,19,20,21,22
+201503,11,12,13,14,15,16,17,18,19,20,21,22
+201504,11,12,13,14,15,16,17,18,19,20,21,22
+201505,11,12,13,14,15,16,17,18,19,20,21,22
+201506,11,12,13,14,15,16,17,18,19,20,21,22
+201507,11,12,13,14,15,16,17,18,19,20,21,22
+201508,11,12,13,14,15,16,17,18,19,20,21,22
+201509,11,12,13,14,15,16,17,18,19,20,21,22
+201510,11,12,13,14,15,16,17,18,19,20,21,22
+201511,11,12,13,14,15,16,17,18,19,20,21,22
+201512,11,12,13,14,15,16,17,18,19,20,21,22
+201601,11,12,13,14,15,16,17,18,19,20,21,22
+201602,11,12,13,14,15,16,17,18,19,20,21,22
+201603,11,12,13,14,15,16,17,18,19,20,21,22
+201604,11,12,13,14,15,16,17,18,19,20,21,22
+201605,11,12,13,14,15,16,17,18,19,20,21,22
+201606,11,12,13,14,15,16,17,18,19,20,21,22
+201607,11,12,13,14,15,16,17,18,19,20,21,22
+201608,11,12,13,14,15,16,17,18,19,20,21,22
+201609,11,12,13,14,15,16,17,18,19,20,21,22
+201610,11,12,13,14,15,16,17,18,19,20,21,22
+201611,11,12,13,14,15,16,17,18,19,20,21,22
+201612,11,12,13,14,15,16,17,18,19,20,21,22
diff --git a/pandas/pandas/tests/io/sas/data/test_12659.sas7bdat b/pandas/pandas/tests/io/sas/data/test_12659.sas7bdat
new file mode 100644
index 0000000000000..bf91e931aa64a
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test_12659.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test_meta2_page.sas7bdat b/pandas/pandas/tests/io/sas/data/test_meta2_page.sas7bdat
new file mode 100644
index 0000000000000..9732cabc212cf
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/test_meta2_page.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/test_sas7bdat_1.csv b/pandas/pandas/tests/io/sas/data/test_sas7bdat_1.csv
new file mode 100644
index 0000000000000..3eb23e42448d1
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/test_sas7bdat_1.csv
@@ -0,0 +1,11 @@
+Column1,Column2,Column3,Column4,Column5,Column6,Column7,Column8,Column9,Column10,Column11,Column12,Column13,Column14,Column15,Column16,Column17,Column18,Column19,Column20,Column21,Column22,Column23,Column24,Column25,Column26,Column27,Column28,Column29,Column30,Column31,Column32,Column33,Column34,Column35,Column36,Column37,Column38,Column39,Column40,Column41,Column42,Column43,Column44,Column45,Column46,Column47,Column48,Column49,Column50,Column51,Column52,Column53,Column54,Column55,Column56,Column57,Column58,Column59,Column60,Column61,Column62,Column63,Column64,Column65,Column66,Column67,Column68,Column69,Column70,Column71,Column72,Column73,Column74,Column75,Column76,Column77,Column78,Column79,Column80,Column81,Column82,Column83,Column84,Column85,Column86,Column87,Column88,Column89,Column90,Column91,Column92,Column93,Column94,Column95,Column96,Column97,Column98,Column99,Column100
+0.636,pear,84,2170,0.103,apple,20,,0.621,apple,,9697,0.047,dog,7,2543,0.728,crocodile,55,2615,0.146,crocodile,10,2832,0.644,crocodile,6,9671,,crocodile,28,9126,0.433,crocodile,22,8117,0.318,dog,61,3363,0.938,pear,58,3700,0.844,dog,26,3989,0.132,crocodile,88,8240,0.325,,9,6102,0.032,apple,34,2987,0.651,crocodile,81,8778,,apple,91,9908,0.897,dog,26,3682,0.274,crocodile,75,1629,0.918,apple,9,7935,0.761,crocodile,,2398,0.914,apple,75,9204,0.946,pear,87,5587,0.940,apple,50,1611,0.480,apple,45,3230
+0.283,dog,49,6275,0.398,pear,50,339,0.561,apple,22,8596,0.661,pear,38,4928,0.709,crocodile,73,6011,0.239,crocodile,93,,0.093,crocodile,23,6198,0.757,dog,61,,0.593,pear,12,9571,,dog,6,892,0.883,pear,81,3363,0.166,pear,,1814,0.454,dog,52,1161,0.630,dog,43,159,0.398,apple,17,8194,0.905,dog,29,4752,0.382,pear,44,9302,0.247,apple,45,4645,0.321,apple,56,,0.415,,,5912,0.580,,60,1918,0.821,dog,68,6847,0.542,apple,83,4465,0.169,pear,53,5820,0.942,dog,76,4904
+0.452,pear,35,8627,0.117,pear,70,5704,0.209,apple,7,5079,0.961,pear,73,,0.414,dog,,9681,,pear,80,4183,0.131,crocodile,53,972,,apple,86,772,0.460,,89,5950,0.293,apple,25,288,0.314,dog,38,106,0.108,pear,49,979,0.481,dog,25,7904,0.270,dog,4,4891,,dog,32,9820,0.517,,64,9053,0.487,dog,78,7238,0.488,apple,54,4349,0.356,pear,73,9991,0.113,,63,5731,0.294,dog,58,3060,,pear,2,,0.004,dog,45,9872,0.024,dog,69,,0.336,pear,9,
+0.557,dog,29,5292,0.640,pear,34,2766,0.910,dog,26,2576,0.924,pear,85,3495,0.478,crocodile,43,1586,0.576,apple,79,4403,0.329,crocodile,,3572,0.702,dog,46,1913,0.147,,10,4292,0.368,crocodile,27,7037,0.137,pear,19,4545,0.999,apple,81,,0.095,pear,36,9195,0.494,pear,61,3393,,crocodile,27,8252,,dog,87,7968,0.845,apple,31,3801,0.514,dog,,6542,0.483,dog,58,4688,,pear,71,402,0.275,apple,51,3673,0.642,,82,4443,0.538,apple,10,114,0.155,,17,8170,0.493,pear,89,8566
+0.138,,55,1903,0.583,crocodile,34,4170,0.226,crocodile,11,3985,0.271,pear,,4624,,,43,2539,,dog,50,,0.469,crocodile,72,5999,0.476,apple,71,1184,0.333,crocodile,23,5075,0.120,pear,64,5242,0.561,apple,7,8206,0.103,,45,,0.972,,1,432,0.423,dog,4,1860,0.686,apple,43,9640,0.622,pear,15,91,0.795,apple,33,2149,0.248,dog,,,0.942,dog,85,987,0.492,pear,2,754,0.955,apple,48,1268,0.498,apple,83,6654,0.305,crocodile,81,423,0.013,,32,3217,0.046,,57,894
+0.948,dog,33,8962,0.691,pear,,2276,0.330,crocodile,27,1104,0.668,,73,2883,0.007,,18,3726,0.301,,85,3621,,apple,39,9646,0.305,pear,48,,0.103,dog,71,8216,0.813,dog,41,6387,0.114,apple,24,4428,0.122,dog,41,,0.558,pear,49,6455,0.119,pear,56,9390,0.482,apple,75,9168,0.437,dog,74,4101,0.557,dog,74,1631,0.159,dog,26,2218,0.274,crocodile,,7653,,dog,,9637,0.345,apple,82,440,0.604,apple,89,6848,0.177,pear,31,2597,0.847,dog,77,,0.700,crocodile,60,6088
+0.162,crocodile,17,8189,0.002,pear,30,5161,0.851,,,7230,0.886,dog,43,5277,0.086,dog,10,6891,0.510,dog,63,2275,0.641,dog,74,9202,0.452,pear,4,6580,0.008,,38,,0.491,apple,11,1469,0.672,dog,89,6182,0.477,apple,4,1877,0.241,dog,61,5083,,apple,78,2526,,,,7547,0.969,dog,22,1578,,dog,86,1211,0.221,apple,1,3252,0.146,,85,9278,0.676,dog,30,1218,0.012,apple,93,3159,0.681,crocodile,45,,0.825,crocodile,0,3749,0.831,,74,7021,0.534,pear,20,6122
+0.148,crocodile,37,1740,0.411,apple,23,8833,0.620,crocodile,5,439,0.372,apple,31,,0.881,dog,84,736,0.347,crocodile,46,224,0.080,crocodile,80,891,0.452,pear,82,3304,0.418,pear,85,2984,0.362,dog,65,,0.167,,65,5265,,apple,89,2101,0.864,apple,92,3636,0.382,crocodile,89,269,0.958,dog,23,1419,0.410,pear,86,8488,0.739,pear,,6576,0.613,dog,37,859,0.426,dog,15,8618,0.554,crocodile,51,4864,0.046,,,4421,0.995,dog,25,622,0.191,pear,84,536,0.405,apple,54,7081,0.575,crocodile,15,2570
+,pear,15,3679,0.102,pear,1,3227,0.157,pear,12,1627,0.955,pear,21,64,0.967,dog,14,2876,0.533,,74,4269,0.881,apple,58,5565,,dog,,6286,0.373,crocodile,46,6525,0.477,crocodile,18,1836,0.067,apple,66,3851,0.224,pear,,7233,0.443,,75,5577,0.392,crocodile,92,4353,0.549,pear,96,4884,0.581,crocodile,10,4058,,crocodile,22,8038,0.411,,94,9236,,pear,40,,0.770,crocodile,36,4591,0.854,dog,32,2902,0.724,apple,57,5499,0.566,pear,19,7577,0.853,,,,,crocodile,84,2709
+0.663,pear,,7735,0.086,apple,80,,,pear,16,9159,0.053,dog,52,3478,0.691,pear,49,9979,0.428,dog,46,5776,0.744,crocodile,3,9549,0.249,dog,90,1192,0.009,dog,68,8678,0.046,apple,62,1873,0.684,,31,7227,0.300,crocodile,83,9881,,,82,8272,0.621,crocodile,40,8171,0.002,crocodile,38,,0.580,pear,31,3094,,,5,5711,0.668,,30,217,0.672,dog,34,3184,0.105,pear,,1521,0.239,dog,28,6896,0.011,apple,,4509,0.691,,97,9247,0.167,pear,74,9824,0.945,crocodile,89,
diff --git a/pandas/pandas/tests/io/sas/data/test_sas7bdat_2.csv b/pandas/pandas/tests/io/sas/data/test_sas7bdat_2.csv
new file mode 100644
index 0000000000000..adc0587ae2797
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/data/test_sas7bdat_2.csv
@@ -0,0 +1,11 @@
+Column1,Column2,Column3,Column4,Column5,Column6,Column7,Column8,Column9,Column10,Column11,Column12,Column13,Column14,Column15,Column16,Column17,Column18,Column19,Column20,Column21,Column22,Column23,Column24,Column25,Column26,Column27,Column28,Column29,Column30,Column31,Column32,Column33,Column34,Column35,Column36,Column37,Column38,Column39,Column40,Column41,Column42,Column43,Column44,Column45,Column46,Column47,Column48,Column49,Column50,Column51,Column52,Column53,Column54,Column55,Column56,Column57,Column58,Column59,Column60,Column61,Column62,Column63,Column64,Column65,Column66,Column67,Column68,Column69,Column70,Column71,Column72,Column73,Column74,Column75,Column76,Column77,Column78,Column79,Column80,Column81,Column82,Column83,Column84,Column85,Column86,Column87,Column88,Column89,Column90,Column91,Column92,Column93,Column94,Column95,Column96,Column97,Column98,Column99,Column100
+0.636,高雄市,84,2170,0.103,부산,20,,0.621,부산,,9697,0.047,Иркутск,7,2543,0.728,鱷魚,55,2615,0.146,鱷魚,10,2832,0.644,鱷魚,6,9671,,鱷魚,28,9126,0.433,鱷魚,22,8117,0.318,Иркутск,61,3363,0.938,高雄市,58,3700,0.844,Иркутск,26,3989,0.132,鱷魚,88,8240,0.325,,9,6102,0.032,부산,34,2987,0.651,鱷魚,81,8778,,부산,91,9908,0.897,Иркутск,26,3682,0.274,鱷魚,75,1629,0.918,부산,9,7935,0.761,鱷魚,,2398,0.914,부산,75,9204,0.946,高雄市,87,5587,0.940,부산,50,1611,0.480,부산,45,3230
+0.283,Иркутск,49,6275,0.398,高雄市,50,339,0.561,부산,22,8596,0.661,高雄市,38,4928,0.709,鱷魚,73,6011,0.239,鱷魚,93,,0.093,鱷魚,23,6198,0.757,Иркутск,61,,0.593,高雄市,12,9571,,Иркутск,6,892,0.883,高雄市,81,3363,0.166,高雄市,,1814,0.454,Иркутск,52,1161,0.630,Иркутск,43,159,0.398,부산,17,8194,0.905,Иркутск,29,4752,0.382,高雄市,44,9302,0.247,부산,45,4645,0.321,부산,56,,0.415,,,5912,0.580,,60,1918,0.821,Иркутск,68,6847,0.542,부산,83,4465,0.169,高雄市,53,5820,0.942,Иркутск,76,4904
+0.452,高雄市,35,8627,0.117,高雄市,70,5704,0.209,부산,7,5079,0.961,高雄市,73,,0.414,Иркутск,,9681,,高雄市,80,4183,0.131,鱷魚,53,972,,부산,86,772,0.460,,89,5950,0.293,부산,25,288,0.314,Иркутск,38,106,0.108,高雄市,49,979,0.481,Иркутск,25,7904,0.270,Иркутск,4,4891,,Иркутск,32,9820,0.517,,64,9053,0.487,Иркутск,78,7238,0.488,부산,54,4349,0.356,高雄市,73,9991,0.113,,63,5731,0.294,Иркутск,58,3060,,高雄市,2,,0.004,Иркутск,45,9872,0.024,Иркутск,69,,0.336,高雄市,9,
+0.557,Иркутск,29,5292,0.640,高雄市,34,2766,0.910,Иркутск,26,2576,0.924,高雄市,85,3495,0.478,鱷魚,43,1586,0.576,부산,79,4403,0.329,鱷魚,,3572,0.702,Иркутск,46,1913,0.147,,10,4292,0.368,鱷魚,27,7037,0.137,高雄市,19,4545,0.999,부산,81,,0.095,高雄市,36,9195,0.494,高雄市,61,3393,,鱷魚,27,8252,,Иркутск,87,7968,0.845,부산,31,3801,0.514,Иркутск,,6542,0.483,Иркутск,58,4688,,高雄市,71,402,0.275,부산,51,3673,0.642,,82,4443,0.538,부산,10,114,0.155,,17,8170,0.493,高雄市,89,8566
+0.138,,55,1903,0.583,鱷魚,34,4170,0.226,鱷魚,11,3985,0.271,高雄市,,4624,,,43,2539,,Иркутск,50,,0.469,鱷魚,72,5999,0.476,부산,71,1184,0.333,鱷魚,23,5075,0.120,高雄市,64,5242,0.561,부산,7,8206,0.103,,45,,0.972,,1,432,0.423,Иркутск,4,1860,0.686,부산,43,9640,0.622,高雄市,15,91,0.795,부산,33,2149,0.248,Иркутск,,,0.942,Иркутск,85,987,0.492,高雄市,2,754,0.955,부산,48,1268,0.498,부산,83,6654,0.305,鱷魚,81,423,0.013,,32,3217,0.046,,57,894
+0.948,Иркутск,33,8962,0.691,高雄市,,2276,0.330,鱷魚,27,1104,0.668,,73,2883,0.007,,18,3726,0.301,,85,3621,,부산,39,9646,0.305,高雄市,48,,0.103,Иркутск,71,8216,0.813,Иркутск,41,6387,0.114,부산,24,4428,0.122,Иркутск,41,,0.558,高雄市,49,6455,0.119,高雄市,56,9390,0.482,부산,75,9168,0.437,Иркутск,74,4101,0.557,Иркутск,74,1631,0.159,Иркутск,26,2218,0.274,鱷魚,,7653,,Иркутск,,9637,0.345,부산,82,440,0.604,부산,89,6848,0.177,高雄市,31,2597,0.847,Иркутск,77,,0.700,鱷魚,60,6088
+0.162,鱷魚,17,8189,0.002,高雄市,30,5161,0.851,,,7230,0.886,Иркутск,43,5277,0.086,Иркутск,10,6891,0.510,Иркутск,63,2275,0.641,Иркутск,74,9202,0.452,高雄市,4,6580,0.008,,38,,0.491,부산,11,1469,0.672,Иркутск,89,6182,0.477,부산,4,1877,0.241,Иркутск,61,5083,,부산,78,2526,,,,7547,0.969,Иркутск,22,1578,,Иркутск,86,1211,0.221,부산,1,3252,0.146,,85,9278,0.676,Иркутск,30,1218,0.012,부산,93,3159,0.681,鱷魚,45,,0.825,鱷魚,0,3749,0.831,,74,7021,0.534,高雄市,20,6122
+0.148,鱷魚,37,1740,0.411,부산,23,8833,0.620,鱷魚,5,439,0.372,부산,31,,0.881,Иркутск,84,736,0.347,鱷魚,46,224,0.080,鱷魚,80,891,0.452,高雄市,82,3304,0.418,高雄市,85,2984,0.362,Иркутск,65,,0.167,,65,5265,,부산,89,2101,0.864,부산,92,3636,0.382,鱷魚,89,269,0.958,Иркутск,23,1419,0.410,高雄市,86,8488,0.739,高雄市,,6576,0.613,Иркутск,37,859,0.426,Иркутск,15,8618,0.554,鱷魚,51,4864,0.046,,,4421,0.995,Иркутск,25,622,0.191,高雄市,84,536,0.405,부산,54,7081,0.575,鱷魚,15,2570
+,高雄市,15,3679,0.102,高雄市,1,3227,0.157,高雄市,12,1627,0.955,高雄市,21,64,0.967,Иркутск,14,2876,0.533,,74,4269,0.881,부산,58,5565,,Иркутск,,6286,0.373,鱷魚,46,6525,0.477,鱷魚,18,1836,0.067,부산,66,3851,0.224,高雄市,,7233,0.443,,75,5577,0.392,鱷魚,92,4353,0.549,高雄市,96,4884,0.581,鱷魚,10,4058,,鱷魚,22,8038,0.411,,94,9236,,高雄市,40,,0.770,鱷魚,36,4591,0.854,Иркутск,32,2902,0.724,부산,57,5499,0.566,高雄市,19,7577,0.853,,,,,鱷魚,84,2709
+0.663,高雄市,,7735,0.086,부산,80,,,高雄市,16,9159,0.053,Иркутск,52,3478,0.691,高雄市,49,9979,0.428,Иркутск,46,5776,0.744,鱷魚,3,9549,0.249,Иркутск,90,1192,0.009,Иркутск,68,8678,0.046,부산,62,1873,0.684,,31,7227,0.300,鱷魚,83,9881,,,82,8272,0.621,鱷魚,40,8171,0.002,鱷魚,38,,0.580,高雄市,31,3094,,,5,5711,0.668,,30,217,0.672,Иркутск,34,3184,0.105,高雄市,,1521,0.239,Иркутск,28,6896,0.011,부산,,4509,0.691,,97,9247,0.167,高雄市,74,9824,0.945,鱷魚,89,
diff --git a/pandas/pandas/tests/io/sas/data/zero_rows.sas7bdat b/pandas/pandas/tests/io/sas/data/zero_rows.sas7bdat
new file mode 100644
index 0000000000000..a5ba95b639507
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/zero_rows.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/data/zero_variables.sas7bdat b/pandas/pandas/tests/io/sas/data/zero_variables.sas7bdat
new file mode 100644
index 0000000000000..85fec09447ec5
Binary files /dev/null and b/pandas/pandas/tests/io/sas/data/zero_variables.sas7bdat differ
diff --git a/pandas/pandas/tests/io/sas/test_byteswap.py b/pandas/pandas/tests/io/sas/test_byteswap.py
new file mode 100644
index 0000000000000..6d7f2f05d1b00
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/test_byteswap.py
@@ -0,0 +1,55 @@
+from hypothesis import (
+    assume,
+    example,
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas._libs.byteswap import (
+    read_double_with_byteswap,
+    read_float_with_byteswap,
+    read_uint16_with_byteswap,
+    read_uint32_with_byteswap,
+    read_uint64_with_byteswap,
+)
+
+import pandas._testing as tm
+
+
+@given(read_offset=st.integers(0, 11), number=st.integers(min_value=0))
+@example(number=2**16, read_offset=0)
+@example(number=2**32, read_offset=0)
+@example(number=2**64, read_offset=0)
+@pytest.mark.parametrize("int_type", [np.uint16, np.uint32, np.uint64])
+@pytest.mark.parametrize("should_byteswap", [True, False])
+def test_int_byteswap(read_offset, number, int_type, should_byteswap):
+    assume(number < 2 ** (8 * int_type(0).itemsize))
+    _test(number, int_type, read_offset, should_byteswap)
+
+
+@pytest.mark.filterwarnings("ignore:overflow encountered:RuntimeWarning")
+@given(read_offset=st.integers(0, 11), number=st.floats())
+@pytest.mark.parametrize("float_type", [np.float32, np.float64])
+@pytest.mark.parametrize("should_byteswap", [True, False])
+def test_float_byteswap(read_offset, number, float_type, should_byteswap):
+    _test(number, float_type, read_offset, should_byteswap)
+
+
+def _test(number, number_type, read_offset, should_byteswap):
+    number = number_type(number)
+    data = np.random.default_rng(2).integers(0, 256, size=20, dtype="uint8")
+    data[read_offset : read_offset + number.itemsize] = number[None].view("uint8")
+    swap_func = {
+        np.float32: read_float_with_byteswap,
+        np.float64: read_double_with_byteswap,
+        np.uint16: read_uint16_with_byteswap,
+        np.uint32: read_uint32_with_byteswap,
+        np.uint64: read_uint64_with_byteswap,
+    }[type(number)]
+    output_number = number_type(swap_func(bytes(data), read_offset, should_byteswap))
+    if should_byteswap:
+        tm.assert_equal(output_number, number.byteswap())
+    else:
+        tm.assert_equal(output_number, number)
diff --git a/pandas/pandas/tests/io/sas/test_sas.py b/pandas/pandas/tests/io/sas/test_sas.py
new file mode 100644
index 0000000000000..1e38baf4fc409
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/test_sas.py
@@ -0,0 +1,34 @@
+from io import StringIO
+
+import pytest
+
+from pandas import read_sas
+import pandas._testing as tm
+
+
+class TestSas:
+    def test_sas_buffer_format(self):
+        # see gh-14947
+        b = StringIO("")
+
+        msg = (
+            "If this is a buffer object rather than a string "
+            "name, you must specify a format string"
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_sas(b)
+
+    def test_sas_read_no_format_or_extension(self):
+        # see gh-24548
+        msg = "unable to infer format of SAS file.+"
+        with tm.ensure_clean("test_file_no_extension") as path:
+            with pytest.raises(ValueError, match=msg):
+                read_sas(path)
+
+
+def test_sas_archive(datapath):
+    fname_uncompressed = datapath("io", "sas", "data", "airline.sas7bdat")
+    df_uncompressed = read_sas(fname_uncompressed)
+    fname_compressed = datapath("io", "sas", "data", "airline.sas7bdat.gz")
+    df_compressed = read_sas(fname_compressed, format="sas7bdat")
+    tm.assert_frame_equal(df_uncompressed, df_compressed)
diff --git a/pandas/pandas/tests/io/sas/test_sas7bdat.py b/pandas/pandas/tests/io/sas/test_sas7bdat.py
new file mode 100644
index 0000000000000..a17cd27f8284e
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/test_sas7bdat.py
@@ -0,0 +1,416 @@
+import contextlib
+from datetime import datetime
+import io
+import os
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+from pandas.compat._constants import (
+    IS64,
+    WASM,
+)
+from pandas.errors import EmptyDataError
+
+import pandas as pd
+import pandas._testing as tm
+
+from pandas.io.sas.sas7bdat import SAS7BDATReader
+
+
+@pytest.fixture
+def dirpath(datapath):
+    return datapath("io", "sas", "data")
+
+
+@pytest.fixture(params=[(1, range(1, 16)), (2, [16])])
+def data_test_ix(request, dirpath):
+    i, test_ix = request.param
+    fname = os.path.join(dirpath, f"test_sas7bdat_{i}.csv")
+    df = pd.read_csv(fname)
+    epoch = datetime(1960, 1, 1)
+    t1 = pd.to_timedelta(df["Column4"], unit="D")
+    df["Column4"] = (epoch + t1).astype("M8[s]")
+    t2 = pd.to_timedelta(df["Column12"], unit="D")
+    df["Column12"] = (epoch + t2).astype("M8[s]")
+    for k in range(df.shape[1]):
+        col = df.iloc[:, k]
+        if col.dtype == np.int64:
+            df.isetitem(k, df.iloc[:, k].astype(np.float64))
+    return df, test_ix
+
+
+# https://github.com/cython/cython/issues/1720
+class TestSAS7BDAT:
+    @pytest.mark.slow
+    def test_from_file(self, dirpath, data_test_ix):
+        expected, test_ix = data_test_ix
+        for k in test_ix:
+            fname = os.path.join(dirpath, f"test{k}.sas7bdat")
+            df = pd.read_sas(fname, encoding="utf-8")
+            tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.slow
+    def test_from_buffer(self, dirpath, data_test_ix):
+        expected, test_ix = data_test_ix
+        for k in test_ix:
+            fname = os.path.join(dirpath, f"test{k}.sas7bdat")
+            with open(fname, "rb") as f:
+                byts = f.read()
+            buf = io.BytesIO(byts)
+            with pd.read_sas(
+                buf, format="sas7bdat", iterator=True, encoding="utf-8"
+            ) as rdr:
+                df = rdr.read()
+            tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.slow
+    def test_from_iterator(self, dirpath, data_test_ix):
+        expected, test_ix = data_test_ix
+        for k in test_ix:
+            fname = os.path.join(dirpath, f"test{k}.sas7bdat")
+            with pd.read_sas(fname, iterator=True, encoding="utf-8") as rdr:
+                df = rdr.read(2)
+                tm.assert_frame_equal(df, expected.iloc[0:2, :])
+                df = rdr.read(3)
+                tm.assert_frame_equal(df, expected.iloc[2:5, :])
+
+    @pytest.mark.slow
+    def test_path_pathlib(self, dirpath, data_test_ix):
+        expected, test_ix = data_test_ix
+        for k in test_ix:
+            fname = Path(os.path.join(dirpath, f"test{k}.sas7bdat"))
+            df = pd.read_sas(fname, encoding="utf-8")
+            tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("chunksize", (3, 5, 10, 11))
+    @pytest.mark.parametrize("k", range(1, 17))
+    def test_iterator_loop(self, dirpath, k, chunksize):
+        # github #13654
+        fname = os.path.join(dirpath, f"test{k}.sas7bdat")
+        with pd.read_sas(fname, chunksize=chunksize, encoding="utf-8") as rdr:
+            y = 0
+            for x in rdr:
+                y += x.shape[0]
+        assert y == rdr.row_count
+
+    def test_iterator_read_too_much(self, dirpath):
+        # github #14734
+        fname = os.path.join(dirpath, "test1.sas7bdat")
+        with pd.read_sas(
+            fname, format="sas7bdat", iterator=True, encoding="utf-8"
+        ) as rdr:
+            d1 = rdr.read(rdr.row_count + 20)
+
+        with pd.read_sas(fname, iterator=True, encoding="utf-8") as rdr:
+            d2 = rdr.read(rdr.row_count + 20)
+        tm.assert_frame_equal(d1, d2)
+
+
+def test_encoding_options(datapath):
+    fname = datapath("io", "sas", "data", "test1.sas7bdat")
+    df1 = pd.read_sas(fname)
+    df2 = pd.read_sas(fname, encoding="utf-8")
+    for col in df1.columns:
+        try:
+            df1[col] = df1[col].str.decode("utf-8")
+        except AttributeError:
+            pass
+    tm.assert_frame_equal(df1, df2)
+
+    with contextlib.closing(SAS7BDATReader(fname, convert_header_text=False)) as rdr:
+        df3 = rdr.read()
+    for x, y in zip(df1.columns, df3.columns):
+        assert x == y.decode()
+
+
+def test_encoding_infer(datapath):
+    fname = datapath("io", "sas", "data", "test1.sas7bdat")
+
+    with pd.read_sas(fname, encoding="infer", iterator=True) as df1_reader:
+        # check: is encoding inferred correctly from file
+        assert df1_reader.inferred_encoding == "cp1252"
+        df1 = df1_reader.read()
+
+    with pd.read_sas(fname, encoding="cp1252", iterator=True) as df2_reader:
+        df2 = df2_reader.read()
+
+    # check: reader reads correct information
+    tm.assert_frame_equal(df1, df2)
+
+
+def test_productsales(datapath):
+    fname = datapath("io", "sas", "data", "productsales.sas7bdat")
+    df = pd.read_sas(fname, encoding="utf-8")
+    fname = datapath("io", "sas", "data", "productsales.csv")
+    df0 = pd.read_csv(fname, parse_dates=["MONTH"])
+    vn = ["ACTUAL", "PREDICT", "QUARTER", "YEAR"]
+    df0[vn] = df0[vn].astype(np.float64)
+
+    df0["MONTH"] = df0["MONTH"].astype("M8[s]")
+    tm.assert_frame_equal(df, df0)
+
+
+def test_12659(datapath):
+    fname = datapath("io", "sas", "data", "test_12659.sas7bdat")
+    df = pd.read_sas(fname)
+    fname = datapath("io", "sas", "data", "test_12659.csv")
+    df0 = pd.read_csv(fname)
+    df0 = df0.astype(np.float64)
+    tm.assert_frame_equal(df, df0)
+
+
+def test_airline(datapath):
+    fname = datapath("io", "sas", "data", "airline.sas7bdat")
+    df = pd.read_sas(fname)
+    fname = datapath("io", "sas", "data", "airline.csv")
+    df0 = pd.read_csv(fname)
+    df0 = df0.astype(np.float64)
+    tm.assert_frame_equal(df, df0)
+
+
+@pytest.mark.skipif(WASM, reason="Pyodide/WASM has 32-bitness")
+def test_date_time(datapath):
+    # Support of different SAS date/datetime formats (PR #15871)
+    fname = datapath("io", "sas", "data", "datetime.sas7bdat")
+    df = pd.read_sas(fname)
+    fname = datapath("io", "sas", "data", "datetime.csv")
+    df0 = pd.read_csv(
+        fname, parse_dates=["Date1", "Date2", "DateTime", "DateTimeHi", "Taiw"]
+    )
+    # GH 19732: Timestamps imported from sas will incur floating point errors
+    # See GH#56014 for discussion of the correct "expected" results
+    #  We are really just testing that we are "close". This only seems to be
+    #  an issue near the implementation bounds.
+
+    df[df.columns[3]] = df.iloc[:, 3].dt.round("us")
+    df0["Date1"] = df0["Date1"].astype("M8[s]")
+    df0["Date2"] = df0["Date2"].astype("M8[s]")
+    df0["DateTime"] = df0["DateTime"].astype("M8[ms]")
+    df0["Taiw"] = df0["Taiw"].astype("M8[s]")
+
+    res = df0["DateTimeHi"].astype("M8[us]").dt.round("ms")
+    df0["DateTimeHi"] = res.astype("M8[ms]")
+
+    if not IS64:
+        # No good reason for this, just what we get on the CI
+        df0.loc[0, "DateTimeHi"] += np.timedelta64(1, "ms")
+        df0.loc[[2, 3], "DateTimeHi"] -= np.timedelta64(1, "ms")
+    tm.assert_frame_equal(df, df0)
+
+
+@pytest.mark.parametrize("column", ["WGT", "CYL"])
+def test_compact_numerical_values(datapath, column):
+    # Regression test for #21616
+    fname = datapath("io", "sas", "data", "cars.sas7bdat")
+    df = pd.read_sas(fname, encoding="latin-1")
+    # The two columns CYL and WGT in cars.sas7bdat have column
+    # width < 8 and only contain integral values.
+    # Test that pandas doesn't corrupt the numbers by adding
+    # decimals.
+    result = df[column]
+    expected = df[column].round()
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+
+def test_many_columns(datapath):
+    # Test for looking for column information in more places (PR #22628)
+    fname = datapath("io", "sas", "data", "many_columns.sas7bdat")
+
+    df = pd.read_sas(fname, encoding="latin-1")
+
+    fname = datapath("io", "sas", "data", "many_columns.csv")
+    df0 = pd.read_csv(fname, encoding="latin-1")
+    tm.assert_frame_equal(df, df0)
+
+
+def test_inconsistent_number_of_rows(datapath):
+    # Regression test for issue #16615. (PR #22628)
+    fname = datapath("io", "sas", "data", "load_log.sas7bdat")
+    df = pd.read_sas(fname, encoding="latin-1")
+    assert len(df) == 2097
+
+
+def test_zero_variables(datapath):
+    # Check if the SAS file has zero variables (PR #18184)
+    fname = datapath("io", "sas", "data", "zero_variables.sas7bdat")
+    with pytest.raises(EmptyDataError, match="No columns to parse from file"):
+        pd.read_sas(fname)
+
+
+@pytest.mark.parametrize("encoding", [None, "utf8"])
+def test_zero_rows(datapath, encoding):
+    # GH 18198
+    fname = datapath("io", "sas", "data", "zero_rows.sas7bdat")
+    result = pd.read_sas(fname, encoding=encoding)
+    str_value = b"a" if encoding is None else "a"
+    expected = pd.DataFrame([{"char_field": str_value, "num_field": 1.0}]).iloc[:0]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_corrupt_read(datapath):
+    # We don't really care about the exact failure, the important thing is
+    # that the resource should be cleaned up afterwards (BUG #35566)
+    fname = datapath("io", "sas", "data", "corrupt.sas7bdat")
+    msg = "'SAS7BDATReader' object has no attribute 'row_count'"
+    with pytest.raises(AttributeError, match=msg):
+        pd.read_sas(fname)
+
+
+@pytest.mark.xfail(WASM, reason="failing with currently set tolerances on WASM")
+def test_max_sas_date(datapath):
+    # GH 20927
+    # NB. max datetime in SAS dataset is 31DEC9999:23:59:59.999
+    #    but this is read as 29DEC9999:23:59:59.998993 by a buggy
+    #    sas7bdat module
+    # See also GH#56014 for discussion of the correct "expected" results.
+    fname = datapath("io", "sas", "data", "max_sas_date.sas7bdat")
+    df = pd.read_sas(fname, encoding="iso-8859-1")
+
+    expected = pd.DataFrame(
+        {
+            "text": ["max", "normal"],
+            "dt_as_float": [253717747199.999, 1880323199.999],
+            "dt_as_dt": np.array(
+                [
+                    datetime(9999, 12, 29, 23, 59, 59, 999000),
+                    datetime(2019, 8, 1, 23, 59, 59, 999000),
+                ],
+                dtype="M8[ms]",
+            ),
+            "date_as_float": [2936547.0, 21762.0],
+            "date_as_date": np.array(
+                [
+                    datetime(9999, 12, 29),
+                    datetime(2019, 8, 1),
+                ],
+                dtype="M8[s]",
+            ),
+        },
+        columns=["text", "dt_as_float", "dt_as_dt", "date_as_float", "date_as_date"],
+    )
+
+    if not IS64:
+        # No good reason for this, just what we get on the CI
+        expected.loc[:, "dt_as_dt"] -= np.timedelta64(1, "ms")
+
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.xfail(WASM, reason="failing with currently set tolerances on WASM")
+def test_max_sas_date_iterator(datapath):
+    # GH 20927
+    # when called as an iterator, only those chunks with a date > pd.Timestamp.max
+    # are returned as datetime.datetime, if this happens that whole chunk is returned
+    # as datetime.datetime
+    col_order = ["text", "dt_as_float", "dt_as_dt", "date_as_float", "date_as_date"]
+    fname = datapath("io", "sas", "data", "max_sas_date.sas7bdat")
+    results = []
+    for df in pd.read_sas(fname, encoding="iso-8859-1", chunksize=1):
+        # GH 19732: Timestamps imported from sas will incur floating point errors
+        df.reset_index(inplace=True, drop=True)
+        results.append(df)
+    expected = [
+        pd.DataFrame(
+            {
+                "text": ["max"],
+                "dt_as_float": [253717747199.999],
+                "dt_as_dt": np.array(
+                    [datetime(9999, 12, 29, 23, 59, 59, 999000)], dtype="M8[ms]"
+                ),
+                "date_as_float": [2936547.0],
+                "date_as_date": np.array([datetime(9999, 12, 29)], dtype="M8[s]"),
+            },
+            columns=col_order,
+        ),
+        pd.DataFrame(
+            {
+                "text": ["normal"],
+                "dt_as_float": [1880323199.999],
+                "dt_as_dt": np.array(["2019-08-01 23:59:59.999"], dtype="M8[ms]"),
+                "date_as_float": [21762.0],
+                "date_as_date": np.array(["2019-08-01"], dtype="M8[s]"),
+            },
+            columns=col_order,
+        ),
+    ]
+    if not IS64:
+        # No good reason for this, just what we get on the CI
+        expected[0].loc[0, "dt_as_dt"] -= np.timedelta64(1, "ms")
+        expected[1].loc[0, "dt_as_dt"] -= np.timedelta64(1, "ms")
+
+    tm.assert_frame_equal(results[0], expected[0])
+    tm.assert_frame_equal(results[1], expected[1])
+
+
+@pytest.mark.skipif(WASM, reason="Pyodide/WASM has 32-bitness")
+def test_null_date(datapath):
+    fname = datapath("io", "sas", "data", "dates_null.sas7bdat")
+    df = pd.read_sas(fname, encoding="utf-8")
+
+    expected = pd.DataFrame(
+        {
+            "datecol": np.array(
+                [
+                    datetime(9999, 12, 29),
+                    np.datetime64("NaT"),
+                ],
+                dtype="M8[s]",
+            ),
+            "datetimecol": np.array(
+                [
+                    datetime(9999, 12, 29, 23, 59, 59, 999000),
+                    np.datetime64("NaT"),
+                ],
+                dtype="M8[ms]",
+            ),
+        },
+    )
+    if not IS64:
+        # No good reason for this, just what we get on the CI
+        expected.loc[0, "datetimecol"] -= np.timedelta64(1, "ms")
+    tm.assert_frame_equal(df, expected)
+
+
+def test_meta2_page(datapath):
+    # GH 35545
+    fname = datapath("io", "sas", "data", "test_meta2_page.sas7bdat")
+    df = pd.read_sas(fname)
+    assert len(df) == 1000
+
+
+@pytest.mark.parametrize(
+    "test_file, override_offset, override_value, expected_msg",
+    [
+        ("test2.sas7bdat", 0x10000 + 55229, 0x80 | 0x0F, "Out of bounds"),
+        ("test2.sas7bdat", 0x10000 + 55229, 0x10, "unknown control byte"),
+        ("test3.sas7bdat", 118170, 184, "Out of bounds"),
+    ],
+)
+def test_rle_rdc_exceptions(
+    datapath, test_file, override_offset, override_value, expected_msg
+):
+    """Errors in RLE/RDC decompression should propagate."""
+    with open(datapath("io", "sas", "data", test_file), "rb") as fd:
+        data = bytearray(fd.read())
+    data[override_offset] = override_value
+    with pytest.raises(Exception, match=expected_msg):
+        pd.read_sas(io.BytesIO(data), format="sas7bdat")
+
+
+def test_0x40_control_byte(datapath):
+    # GH 31243
+    fname = datapath("io", "sas", "data", "0x40controlbyte.sas7bdat")
+    df = pd.read_sas(fname, encoding="ascii")
+    fname = datapath("io", "sas", "data", "0x40controlbyte.csv")
+    df0 = pd.read_csv(fname, dtype="str")
+    tm.assert_frame_equal(df, df0)
+
+
+def test_0x00_control_byte(datapath):
+    # GH 47099
+    fname = datapath("io", "sas", "data", "0x00controlbyte.sas7bdat.bz2")
+    df = next(pd.read_sas(fname, chunksize=11_000))
+    assert df.shape == (11_000, 20)
diff --git a/pandas/pandas/tests/io/sas/test_xport.py b/pandas/pandas/tests/io/sas/test_xport.py
new file mode 100644
index 0000000000000..30ca70d7e0009
--- /dev/null
+++ b/pandas/pandas/tests/io/sas/test_xport.py
@@ -0,0 +1,153 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+from pandas.io.sas.sasreader import read_sas
+
+# CSV versions of test xpt files were obtained using the R foreign library
+
+# Numbers in a SAS xport file are always float64, so need to convert
+# before making comparisons.
+
+
+def numeric_as_float(data):
+    for v in data.columns:
+        if data[v].dtype is np.dtype("int64"):
+            data[v] = data[v].astype(np.float64)
+
+
+class TestXport:
+    @pytest.mark.slow
+    def test1_basic(self, datapath):
+        # Tests with DEMO_G.xpt (all numeric file)
+
+        # Compare to this
+        file01 = datapath("io", "sas", "data", "DEMO_G.xpt")
+        data_csv = pd.read_csv(file01.replace(".xpt", ".csv"))
+        numeric_as_float(data_csv)
+
+        # Read full file
+        data = read_sas(file01, format="xport")
+        tm.assert_frame_equal(data, data_csv)
+        num_rows = data.shape[0]
+
+        # Test reading beyond end of file
+        with read_sas(file01, format="xport", iterator=True) as reader:
+            data = reader.read(num_rows + 100)
+        assert data.shape[0] == num_rows
+
+        # Test incremental read with `read` method.
+        with read_sas(file01, format="xport", iterator=True) as reader:
+            data = reader.read(10)
+        tm.assert_frame_equal(data, data_csv.iloc[0:10, :])
+
+        # Test incremental read with `get_chunk` method.
+        with read_sas(file01, format="xport", chunksize=10) as reader:
+            data = reader.get_chunk()
+        tm.assert_frame_equal(data, data_csv.iloc[0:10, :])
+
+        # Test read in loop
+        m = 0
+        with read_sas(file01, format="xport", chunksize=100) as reader:
+            for x in reader:
+                m += x.shape[0]
+        assert m == num_rows
+
+        # Read full file with `read_sas` method
+        data = read_sas(file01)
+        tm.assert_frame_equal(data, data_csv)
+
+    def test1_index(self, datapath):
+        # Tests with DEMO_G.xpt using index (all numeric file)
+
+        # Compare to this
+        file01 = datapath("io", "sas", "data", "DEMO_G.xpt")
+        data_csv = pd.read_csv(file01.replace(".xpt", ".csv"))
+        data_csv = data_csv.set_index("SEQN")
+        numeric_as_float(data_csv)
+
+        # Read full file
+        data = read_sas(file01, index="SEQN", format="xport")
+        tm.assert_frame_equal(data, data_csv, check_index_type=False)
+
+        # Test incremental read with `read` method.
+        with read_sas(file01, index="SEQN", format="xport", iterator=True) as reader:
+            data = reader.read(10)
+        tm.assert_frame_equal(data, data_csv.iloc[0:10, :], check_index_type=False)
+
+        # Test incremental read with `get_chunk` method.
+        with read_sas(file01, index="SEQN", format="xport", chunksize=10) as reader:
+            data = reader.get_chunk()
+        tm.assert_frame_equal(data, data_csv.iloc[0:10, :], check_index_type=False)
+
+    def test1_incremental(self, datapath):
+        # Test with DEMO_G.xpt, reading full file incrementally
+
+        file01 = datapath("io", "sas", "data", "DEMO_G.xpt")
+        data_csv = pd.read_csv(file01.replace(".xpt", ".csv"))
+        data_csv = data_csv.set_index("SEQN")
+        numeric_as_float(data_csv)
+
+        with read_sas(file01, index="SEQN", chunksize=1000) as reader:
+            all_data = list(reader)
+        data = pd.concat(all_data, axis=0)
+
+        tm.assert_frame_equal(data, data_csv, check_index_type=False)
+
+    def test2(self, datapath):
+        # Test with SSHSV1_A.xpt
+
+        file02 = datapath("io", "sas", "data", "SSHSV1_A.xpt")
+        # Compare to this
+        data_csv = pd.read_csv(file02.replace(".xpt", ".csv"))
+        numeric_as_float(data_csv)
+
+        data = read_sas(file02)
+        tm.assert_frame_equal(data, data_csv)
+
+    def test2_binary(self, datapath):
+        # Test with SSHSV1_A.xpt, read as a binary file
+
+        # Compare to this
+        file02 = datapath("io", "sas", "data", "SSHSV1_A.xpt")
+        data_csv = pd.read_csv(file02.replace(".xpt", ".csv"))
+        numeric_as_float(data_csv)
+
+        with open(file02, "rb") as fd:
+            # GH#35693 ensure that if we pass an open file, we
+            #  dont incorrectly close it in read_sas
+            data = read_sas(fd, format="xport")
+
+        tm.assert_frame_equal(data, data_csv)
+
+    def test_multiple_types(self, datapath):
+        # Test with DRXFCD_G.xpt (contains text and numeric variables)
+
+        # Compare to this
+        file03 = datapath("io", "sas", "data", "DRXFCD_G.xpt")
+        data_csv = pd.read_csv(file03.replace(".xpt", ".csv"))
+
+        data = read_sas(file03, encoding="utf-8")
+        tm.assert_frame_equal(data, data_csv)
+
+    def test_truncated_float_support(self, datapath):
+        # Test with paxraw_d_short.xpt, a shortened version of:
+        # http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PAXRAW_D.ZIP
+        # This file has truncated floats (5 bytes in this case).
+
+        # GH 11713
+        file04 = datapath("io", "sas", "data", "paxraw_d_short.xpt")
+        data_csv = pd.read_csv(file04.replace(".xpt", ".csv"))
+
+        data = read_sas(file04, format="xport")
+        tm.assert_frame_equal(data.astype("int64"), data_csv)
+
+    def test_cport_header_found_raises(self, datapath):
+        # Test with DEMO_PUF.cpt, the beginning of puf2019_1_fall.xpt
+        # from https://www.cms.gov/files/zip/puf2019.zip
+        # (despite the extension, it's a cpt file)
+        msg = "Header record indicates a CPORT file, which is not readable."
+        with pytest.raises(ValueError, match=msg):
+            read_sas(datapath("io", "sas", "data", "DEMO_PUF.cpt"), format="xport")
diff --git a/pandas/pandas/tests/io/test_clipboard.py b/pandas/pandas/tests/io/test_clipboard.py
new file mode 100644
index 0000000000000..b5e97314caf03
--- /dev/null
+++ b/pandas/pandas/tests/io/test_clipboard.py
@@ -0,0 +1,406 @@
+from textwrap import dedent
+
+import numpy as np
+import pytest
+
+from pandas.errors import (
+    PyperclipException,
+    PyperclipWindowsException,
+)
+
+import pandas as pd
+from pandas import (
+    NA,
+    DataFrame,
+    Series,
+    get_option,
+    read_clipboard,
+)
+import pandas._testing as tm
+
+from pandas.io.clipboard import (
+    CheckedCall,
+    _stringifyText,
+    init_qt_clipboard,
+)
+
+
+def build_kwargs(sep, excel):
+    kwargs = {}
+    if excel != "default":
+        kwargs["excel"] = excel
+    if sep != "default":
+        kwargs["sep"] = sep
+    return kwargs
+
+
+@pytest.fixture(
+    params=[
+        "delims",
+        "utf8",
+        "utf16",
+        "string",
+        "long",
+        "nonascii",
+        "colwidth",
+        "mixed",
+        "float",
+        "int",
+    ]
+)
+def df(request):
+    data_type = request.param
+
+    if data_type == "delims":
+        return DataFrame({"a": ['"a,\t"b|c', "d\tef`"], "b": ["hi'j", "k''lm"]})
+    elif data_type == "utf8":
+        return DataFrame({"a": ["µasd", "Ωœ∑`"], "b": ["øπ∆˚¬", "œ∑`®"]})
+    elif data_type == "utf16":
+        return DataFrame(
+            {"a": ["\U0001f44d\U0001f44d", "\U0001f44d\U0001f44d"], "b": ["abc", "def"]}
+        )
+    elif data_type == "string":
+        return DataFrame(
+            np.array([f"i-{i}" for i in range(15)]).reshape(5, 3), columns=list("abc")
+        )
+    elif data_type == "long":
+        max_rows = get_option("display.max_rows")
+        return DataFrame(
+            np.random.default_rng(2).integers(0, 10, size=(max_rows + 1, 3)),
+            columns=list("abc"),
+        )
+    elif data_type == "nonascii":
+        return DataFrame({"en": "in English".split(), "es": "en español".split()})
+    elif data_type == "colwidth":
+        _cw = get_option("display.max_colwidth") + 1
+        return DataFrame(
+            np.array(["x" * _cw for _ in range(15)]).reshape(5, 3), columns=list("abc")
+        )
+    elif data_type == "mixed":
+        return DataFrame(
+            {
+                "a": np.arange(1.0, 6.0) + 0.01,
+                "b": np.arange(1, 6).astype(np.int64),
+                "c": list("abcde"),
+            }
+        )
+    elif data_type == "float":
+        return DataFrame(np.random.default_rng(2).random((5, 3)), columns=list("abc"))
+    elif data_type == "int":
+        return DataFrame(
+            np.random.default_rng(2).integers(0, 10, (5, 3)), columns=list("abc")
+        )
+    else:
+        raise ValueError
+
+
+@pytest.fixture
+def mock_ctypes(monkeypatch):
+    """
+    Mocks WinError to help with testing the clipboard.
+    """
+
+    def _mock_win_error():
+        return "Window Error"
+
+    # Set raising to False because WinError won't exist on non-windows platforms
+    with monkeypatch.context() as m:
+        m.setattr("ctypes.WinError", _mock_win_error, raising=False)
+        yield
+
+
+@pytest.mark.usefixtures("mock_ctypes")
+def test_checked_call_with_bad_call(monkeypatch):
+    """
+    Give CheckCall a function that returns a falsey value and
+    mock get_errno so it returns false so an exception is raised.
+    """
+
+    def _return_false():
+        return False
+
+    monkeypatch.setattr("pandas.io.clipboard.get_errno", lambda: True)
+    msg = f"Error calling {_return_false.__name__} \\(Window Error\\)"
+
+    with pytest.raises(PyperclipWindowsException, match=msg):
+        CheckedCall(_return_false)()
+
+
+@pytest.mark.usefixtures("mock_ctypes")
+def test_checked_call_with_valid_call(monkeypatch):
+    """
+    Give CheckCall a function that returns a truthy value and
+    mock get_errno so it returns true so an exception is not raised.
+    The function should return the results from _return_true.
+    """
+
+    def _return_true():
+        return True
+
+    monkeypatch.setattr("pandas.io.clipboard.get_errno", lambda: False)
+
+    # Give CheckedCall a callable that returns a truthy value s
+    checked_call = CheckedCall(_return_true)
+    assert checked_call() is True
+
+
+@pytest.mark.parametrize(
+    "text",
+    [
+        "String_test",
+        True,
+        1,
+        1.0,
+        1j,
+    ],
+)
+def test_stringify_text(text):
+    valid_types = (str, int, float, bool)
+
+    if isinstance(text, valid_types):
+        result = _stringifyText(text)
+        assert result == str(text)
+    else:
+        msg = (
+            "only str, int, float, and bool values "
+            f"can be copied to the clipboard, not {type(text).__name__}"
+        )
+        with pytest.raises(PyperclipException, match=msg):
+            _stringifyText(text)
+
+
+@pytest.fixture
+def set_pyqt_clipboard(monkeypatch):
+    qt_cut, qt_paste = init_qt_clipboard()
+    with monkeypatch.context() as m:
+        m.setattr(pd.io.clipboard, "clipboard_set", qt_cut)
+        m.setattr(pd.io.clipboard, "clipboard_get", qt_paste)
+        yield
+
+
+@pytest.fixture
+def clipboard(qapp):
+    clip = qapp.clipboard()
+    yield clip
+    clip.clear()
+
+
+@pytest.mark.single_cpu
+@pytest.mark.clipboard
+@pytest.mark.usefixtures("set_pyqt_clipboard")
+@pytest.mark.usefixtures("clipboard")
+class TestClipboard:
+    # Test that default arguments copy as tab delimited
+    # Test that explicit delimiters are respected
+    @pytest.mark.parametrize("sep", [None, "\t", ",", "|"])
+    @pytest.mark.parametrize("encoding", [None, "UTF-8", "utf-8", "utf8"])
+    def test_round_trip_frame_sep(self, df, sep, encoding):
+        df.to_clipboard(excel=None, sep=sep, encoding=encoding)
+        result = read_clipboard(sep=sep or "\t", index_col=0, encoding=encoding)
+        tm.assert_frame_equal(df, result)
+
+    # Test white space separator
+    def test_round_trip_frame_string(self, df):
+        df.to_clipboard(excel=False, sep=None)
+        result = read_clipboard()
+        assert df.to_string() == result.to_string()
+        assert df.shape == result.shape
+
+    # Two character separator is not supported in to_clipboard
+    # Test that multi-character separators are not silently passed
+    def test_excel_sep_warning(self, df):
+        with tm.assert_produces_warning(
+            UserWarning,
+            match="to_clipboard in excel mode requires a single character separator.",
+            check_stacklevel=False,
+        ):
+            df.to_clipboard(excel=True, sep=r"\t")
+
+    # Separator is ignored when excel=False and should produce a warning
+    def test_copy_delim_warning(self, df):
+        with tm.assert_produces_warning(UserWarning, match="ignores the sep argument"):
+            df.to_clipboard(excel=False, sep="\t")
+
+    # Tests that the default behavior of to_clipboard is tab
+    # delimited and excel="True"
+    @pytest.mark.parametrize("sep", ["\t", None, "default"])
+    @pytest.mark.parametrize("excel", [True, None, "default"])
+    def test_clipboard_copy_tabs_default(self, sep, excel, df, clipboard):
+        kwargs = build_kwargs(sep, excel)
+        df.to_clipboard(**kwargs)
+        assert clipboard.text() == df.to_csv(sep="\t")
+
+    # Tests reading of white space separated tables
+    @pytest.mark.parametrize("sep", [None, "default"])
+    def test_clipboard_copy_strings(self, sep, df):
+        kwargs = build_kwargs(sep, False)
+        df.to_clipboard(**kwargs)
+        result = read_clipboard(sep=r"\s+")
+        assert result.to_string() == df.to_string()
+        assert df.shape == result.shape
+
+    def test_read_clipboard_infer_excel(self, clipboard):
+        # gh-19010: avoid warnings
+        clip_kwargs = {"engine": "python"}
+
+        text = dedent(
+            """
+            John James\tCharlie Mingus
+            1\t2
+            4\tHarry Carney
+            """.strip()
+        )
+        clipboard.setText(text)
+        df = read_clipboard(**clip_kwargs)
+
+        # excel data is parsed correctly
+        assert df.iloc[1, 1] == "Harry Carney"
+
+        # having diff tab counts doesn't trigger it
+        text = dedent(
+            """
+            a\t b
+            1  2
+            3  4
+            """.strip()
+        )
+        clipboard.setText(text)
+        res = read_clipboard(**clip_kwargs)
+
+        text = dedent(
+            """
+            a  b
+            1  2
+            3  4
+            """.strip()
+        )
+        clipboard.setText(text)
+        exp = read_clipboard(**clip_kwargs)
+
+        tm.assert_frame_equal(res, exp)
+
+    def test_infer_excel_with_nulls(self, clipboard):
+        # GH41108
+        text = "col1\tcol2\n1\tred\n\tblue\n2\tgreen"
+
+        clipboard.setText(text)
+        df = read_clipboard()
+        df_expected = DataFrame(
+            data={"col1": [1, None, 2], "col2": ["red", "blue", "green"]}
+        )
+
+        # excel data is parsed correctly
+        tm.assert_frame_equal(df, df_expected)
+
+    @pytest.mark.parametrize(
+        "multiindex",
+        [
+            (  # Can't use `dedent` here as it will remove the leading `\t`
+                "\n".join(
+                    [
+                        "\t\t\tcol1\tcol2",
+                        "A\t0\tTrue\t1\tred",
+                        "A\t1\tTrue\t\tblue",
+                        "B\t0\tFalse\t2\tgreen",
+                    ]
+                ),
+                [["A", "A", "B"], [0, 1, 0], [True, True, False]],
+            ),
+            (
+                "\n".join(
+                    ["\t\tcol1\tcol2", "A\t0\t1\tred", "A\t1\t\tblue", "B\t0\t2\tgreen"]
+                ),
+                [["A", "A", "B"], [0, 1, 0]],
+            ),
+        ],
+    )
+    def test_infer_excel_with_multiindex(self, clipboard, multiindex):
+        # GH41108
+
+        clipboard.setText(multiindex[0])
+        df = read_clipboard()
+        df_expected = DataFrame(
+            data={"col1": [1, None, 2], "col2": ["red", "blue", "green"]},
+            index=multiindex[1],
+        )
+
+        # excel data is parsed correctly
+        tm.assert_frame_equal(df, df_expected)
+
+    def test_invalid_encoding(self, df):
+        msg = "clipboard only supports utf-8 encoding"
+        # test case for testing invalid encoding
+        with pytest.raises(ValueError, match=msg):
+            df.to_clipboard(encoding="ascii")
+        with pytest.raises(NotImplementedError, match=msg):
+            read_clipboard(encoding="ascii")
+
+    @pytest.mark.parametrize("data", ["\U0001f44d...", "Ωœ∑`...", "abcd..."])
+    def test_raw_roundtrip(self, data):
+        # PR #25040 wide unicode wasn't copied correctly on PY3 on windows
+        df = DataFrame({"data": [data]})
+        df.to_clipboard()
+        result = read_clipboard()
+        tm.assert_frame_equal(df, result)
+
+    @pytest.mark.parametrize("engine", ["c", "python"])
+    def test_read_clipboard_dtype_backend(
+        self, clipboard, string_storage, dtype_backend, engine, using_infer_string
+    ):
+        # GH#50502
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            if engine == "c" and string_storage == "pyarrow":
+                # TODO avoid this exception?
+                string_dtype = pd.ArrowDtype(pa.large_string())
+            else:
+                string_dtype = pd.ArrowDtype(pa.string())
+        else:
+            string_dtype = pd.StringDtype(string_storage)
+
+        text = """a,b,c,d,e,f,g,h,i
+x,1,4.0,x,2,4.0,,True,False
+y,2,5.0,,,,,False,"""
+        clipboard.setText(text)
+
+        with pd.option_context("mode.string_storage", string_storage):
+            result = read_clipboard(sep=",", dtype_backend=dtype_backend, engine=engine)
+
+        expected = DataFrame(
+            {
+                "a": Series(["x", "y"], dtype=string_dtype),
+                "b": Series([1, 2], dtype="Int64"),
+                "c": Series([4.0, 5.0], dtype="Float64"),
+                "d": Series(["x", None], dtype=string_dtype),
+                "e": Series([2, NA], dtype="Int64"),
+                "f": Series([4.0, NA], dtype="Float64"),
+                "g": Series([NA, NA], dtype="Int64"),
+                "h": Series([True, False], dtype="boolean"),
+                "i": Series([False, NA], dtype="boolean"),
+            }
+        )
+        if dtype_backend == "pyarrow":
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                    for col in expected.columns
+                }
+            )
+            expected["g"] = ArrowExtensionArray(pa.array([None, None]))
+
+        if using_infer_string:
+            expected.columns = expected.columns.astype(
+                pd.StringDtype(string_storage, na_value=np.nan)
+            )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_invalid_dtype_backend(self):
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_clipboard(dtype_backend="numpy")
diff --git a/pandas/pandas/tests/io/test_common.py b/pandas/pandas/tests/io/test_common.py
new file mode 100644
index 0000000000000..e162815271ab3
--- /dev/null
+++ b/pandas/pandas/tests/io/test_common.py
@@ -0,0 +1,685 @@
+"""
+Tests for the pandas.io.common functionalities
+"""
+
+import codecs
+import errno
+from functools import partial
+from io import (
+    BytesIO,
+    StringIO,
+    UnsupportedOperation,
+)
+import mmap
+import os
+from pathlib import Path
+import pickle
+import tempfile
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    WASM,
+    is_platform_windows,
+)
+from pandas.compat.pyarrow import pa_version_under19p0
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+import pandas.io.common as icom
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+class CustomFSPath:
+    """For testing fspath on unknown objects"""
+
+    def __init__(self, path) -> None:
+        self.path = path
+
+    def __fspath__(self):
+        return self.path
+
+
+HERE = os.path.abspath(os.path.dirname(__file__))
+
+
+# https://github.com/cython/cython/issues/1720
+class TestCommonIOCapabilities:
+    data1 = """index,A,B,C,D
+foo,2,3,4,5
+bar,7,8,9,10
+baz,12,13,14,15
+qux,12,13,14,15
+foo2,12,13,14,15
+bar2,12,13,14,15
+"""
+
+    def test_expand_user(self):
+        filename = "~/sometest"
+        expanded_name = icom._expand_user(filename)
+
+        assert expanded_name != filename
+        assert os.path.isabs(expanded_name)
+        assert os.path.expanduser(filename) == expanded_name
+
+    def test_expand_user_normal_path(self):
+        filename = "/somefolder/sometest"
+        expanded_name = icom._expand_user(filename)
+
+        assert expanded_name == filename
+        assert os.path.expanduser(filename) == expanded_name
+
+    def test_stringify_path_pathlib(self):
+        rel_path = icom.stringify_path(Path("."))
+        assert rel_path == "."
+        redundant_path = icom.stringify_path(Path("foo//bar"))
+        assert redundant_path == os.path.join("foo", "bar")
+
+    def test_stringify_path_fspath(self):
+        p = CustomFSPath("foo/bar.csv")
+        result = icom.stringify_path(p)
+        assert result == "foo/bar.csv"
+
+    def test_stringify_file_and_path_like(self):
+        # GH 38125: do not stringify file objects that are also path-like
+        fsspec = pytest.importorskip("fsspec")
+        with tm.ensure_clean() as path:
+            with fsspec.open(f"file://{path}", mode="wb") as fsspec_obj:
+                assert fsspec_obj == icom.stringify_path(fsspec_obj)
+
+    @pytest.mark.parametrize("path_type", [str, CustomFSPath, Path])
+    def test_infer_compression_from_path(self, compression_format, path_type):
+        extension, expected = compression_format
+        path = path_type("foo/bar.csv" + extension)
+        compression = icom.infer_compression(path, compression="infer")
+        assert compression == expected
+
+    @pytest.mark.parametrize("path_type", [str, CustomFSPath, Path])
+    def test_get_handle_with_path(self, path_type):
+        with tempfile.TemporaryDirectory(dir=Path.home()) as tmp:
+            filename = path_type("~/" + Path(tmp).name + "/sometest")
+            with icom.get_handle(filename, "w") as handles:
+                assert Path(handles.handle.name).is_absolute()
+                assert os.path.expanduser(filename) == handles.handle.name
+
+    def test_get_handle_with_buffer(self):
+        with StringIO() as input_buffer:
+            with icom.get_handle(input_buffer, "r") as handles:
+                assert handles.handle == input_buffer
+            assert not input_buffer.closed
+        assert input_buffer.closed
+
+    # Test that BytesIOWrapper(get_handle) returns correct amount of bytes every time
+    def test_bytesiowrapper_returns_correct_bytes(self):
+        # Test latin1, ucs-2, and ucs-4 chars
+        data = """a,b,c
+1,2,3
+©,®,®
+Look,a snake,🐍"""
+        with icom.get_handle(StringIO(data), "rb", is_text=False) as handles:
+            result = b""
+            chunksize = 5
+            while True:
+                chunk = handles.handle.read(chunksize)
+                # Make sure each chunk is correct amount of bytes
+                assert len(chunk) <= chunksize
+                if len(chunk) < chunksize:
+                    # Can be less amount of bytes, but only at EOF
+                    # which happens when read returns empty
+                    assert len(handles.handle.read()) == 0
+                    result += chunk
+                    break
+                result += chunk
+            assert result == data.encode("utf-8")
+
+    # Test that pyarrow can handle a file opened with get_handle
+    def test_get_handle_pyarrow_compat(self):
+        pa_csv = pytest.importorskip("pyarrow.csv")
+
+        # Test latin1, ucs-2, and ucs-4 chars
+        data = """a,b,c
+1,2,3
+©,®,®
+Look,a snake,🐍"""
+        expected = pd.DataFrame(
+            {"a": ["1", "©", "Look"], "b": ["2", "®", "a snake"], "c": ["3", "®", "🐍"]}
+        )
+        s = StringIO(data)
+        with icom.get_handle(s, "rb", is_text=False) as handles:
+            df = pa_csv.read_csv(handles.handle).to_pandas()
+            if pa_version_under19p0:
+                expected = expected.astype("object")
+            tm.assert_frame_equal(df, expected)
+            assert not s.closed
+
+    def test_iterator(self):
+        with pd.read_csv(StringIO(self.data1), chunksize=1) as reader:
+            result = pd.concat(reader, ignore_index=True)
+        expected = pd.read_csv(StringIO(self.data1))
+        tm.assert_frame_equal(result, expected)
+
+        # GH12153
+        with pd.read_csv(StringIO(self.data1), chunksize=1) as it:
+            first = next(it)
+            tm.assert_frame_equal(first, expected.iloc[[0]])
+            tm.assert_frame_equal(pd.concat(it), expected.iloc[1:])
+
+    @pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+    @pytest.mark.parametrize(
+        "reader, module, error_class, fn_ext",
+        [
+            (pd.read_csv, "os", FileNotFoundError, "csv"),
+            (pd.read_fwf, "os", FileNotFoundError, "txt"),
+            (pd.read_excel, "xlrd", FileNotFoundError, "xlsx"),
+            (pd.read_feather, "pyarrow", OSError, "feather"),
+            (pd.read_hdf, "tables", FileNotFoundError, "h5"),
+            (pd.read_stata, "os", FileNotFoundError, "dta"),
+            (pd.read_sas, "os", FileNotFoundError, "sas7bdat"),
+            (pd.read_json, "os", FileNotFoundError, "json"),
+            (pd.read_pickle, "os", FileNotFoundError, "pickle"),
+        ],
+    )
+    def test_read_non_existent(self, reader, module, error_class, fn_ext):
+        pytest.importorskip(module)
+
+        path = os.path.join(HERE, "data", "does_not_exist." + fn_ext)
+        msg1 = rf"File (b')?.+does_not_exist\.{fn_ext}'? does not exist"
+        msg2 = rf"\[Errno 2\] No such file or directory: '.+does_not_exist\.{fn_ext}'"
+        msg3 = "Expected object or value"
+        msg4 = "path_or_buf needs to be a string file path or file-like"
+        msg5 = (
+            rf"\[Errno 2\] File .+does_not_exist\.{fn_ext} does not exist: "
+            rf"'.+does_not_exist\.{fn_ext}'"
+        )
+        msg6 = rf"\[Errno 2\] 没有那个文件或目录: '.+does_not_exist\.{fn_ext}'"
+        msg7 = (
+            rf"\[Errno 2\] File o directory non esistente: '.+does_not_exist\.{fn_ext}'"
+        )
+        msg8 = rf"Failed to open local file.+does_not_exist\.{fn_ext}"
+
+        with pytest.raises(
+            error_class,
+            match=rf"({msg1}|{msg2}|{msg3}|{msg4}|{msg5}|{msg6}|{msg7}|{msg8})",
+        ):
+            reader(path)
+
+    @pytest.mark.parametrize(
+        "method, module, error_class, fn_ext",
+        [
+            (pd.DataFrame.to_csv, "os", OSError, "csv"),
+            (pd.DataFrame.to_html, "os", OSError, "html"),
+            (pd.DataFrame.to_excel, "xlrd", OSError, "xlsx"),
+            (pd.DataFrame.to_feather, "pyarrow", OSError, "feather"),
+            (pd.DataFrame.to_parquet, "pyarrow", OSError, "parquet"),
+            (pd.DataFrame.to_stata, "os", OSError, "dta"),
+            (pd.DataFrame.to_json, "os", OSError, "json"),
+            (pd.DataFrame.to_pickle, "os", OSError, "pickle"),
+        ],
+    )
+    # NOTE: Missing parent directory for pd.DataFrame.to_hdf is handled by PyTables
+    def test_write_missing_parent_directory(self, method, module, error_class, fn_ext):
+        pytest.importorskip(module)
+
+        dummy_frame = pd.DataFrame({"a": [1, 2, 3], "b": [2, 3, 4], "c": [3, 4, 5]})
+
+        path = os.path.join(HERE, "data", "missing_folder", "does_not_exist." + fn_ext)
+
+        with pytest.raises(
+            error_class,
+            match=r"Cannot save file into a non-existent directory: .*missing_folder",
+        ):
+            method(dummy_frame, path)
+
+    @pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+    @pytest.mark.parametrize(
+        "reader, module, error_class, fn_ext",
+        [
+            (pd.read_csv, "os", FileNotFoundError, "csv"),
+            (pd.read_table, "os", FileNotFoundError, "csv"),
+            (pd.read_fwf, "os", FileNotFoundError, "txt"),
+            (pd.read_excel, "xlrd", FileNotFoundError, "xlsx"),
+            (pd.read_feather, "pyarrow", OSError, "feather"),
+            (pd.read_hdf, "tables", FileNotFoundError, "h5"),
+            (pd.read_stata, "os", FileNotFoundError, "dta"),
+            (pd.read_sas, "os", FileNotFoundError, "sas7bdat"),
+            (pd.read_json, "os", FileNotFoundError, "json"),
+            (pd.read_pickle, "os", FileNotFoundError, "pickle"),
+        ],
+    )
+    def test_read_expands_user_home_dir(
+        self, reader, module, error_class, fn_ext, monkeypatch
+    ):
+        pytest.importorskip(module)
+
+        path = os.path.join("~", "does_not_exist." + fn_ext)
+        monkeypatch.setattr(icom, "_expand_user", lambda x: os.path.join("foo", x))
+
+        msg1 = rf"File (b')?.+does_not_exist\.{fn_ext}'? does not exist"
+        msg2 = rf"\[Errno 2\] No such file or directory: '.+does_not_exist\.{fn_ext}'"
+        msg3 = "Unexpected character found when decoding 'false'"
+        msg4 = "path_or_buf needs to be a string file path or file-like"
+        msg5 = (
+            rf"\[Errno 2\] File .+does_not_exist\.{fn_ext} does not exist: "
+            rf"'.+does_not_exist\.{fn_ext}'"
+        )
+        msg6 = rf"\[Errno 2\] 没有那个文件或目录: '.+does_not_exist\.{fn_ext}'"
+        msg7 = (
+            rf"\[Errno 2\] File o directory non esistente: '.+does_not_exist\.{fn_ext}'"
+        )
+        msg8 = rf"Failed to open local file.+does_not_exist\.{fn_ext}"
+
+        with pytest.raises(
+            error_class,
+            match=rf"({msg1}|{msg2}|{msg3}|{msg4}|{msg5}|{msg6}|{msg7}|{msg8})",
+        ):
+            reader(path)
+
+    @pytest.mark.parametrize(
+        "reader, module, path",
+        [
+            (pd.read_csv, "os", ("io", "data", "csv", "iris.csv")),
+            (pd.read_table, "os", ("io", "data", "csv", "iris.csv")),
+            (
+                pd.read_fwf,
+                "os",
+                ("io", "data", "fixed_width", "fixed_width_format.txt"),
+            ),
+            (pd.read_excel, "xlrd", ("io", "data", "excel", "test1.xlsx")),
+            (
+                pd.read_feather,
+                "pyarrow",
+                ("io", "data", "feather", "feather-0_3_1.feather"),
+            ),
+            (
+                pd.read_hdf,
+                "tables",
+                ("io", "data", "legacy_hdf", "pytables_native2.h5"),
+            ),
+            (pd.read_stata, "os", ("io", "data", "stata", "stata10_115.dta")),
+            (pd.read_sas, "os", ("io", "sas", "data", "test1.sas7bdat")),
+            (pd.read_json, "os", ("io", "json", "data", "tsframe_v012.json")),
+            (
+                pd.read_pickle,
+                "os",
+                ("io", "data", "pickle", "categorical.0.25.0.pickle"),
+            ),
+        ],
+    )
+    def test_read_fspath_all(self, reader, module, path, datapath):
+        pytest.importorskip(module)
+        path = datapath(*path)
+
+        mypath = CustomFSPath(path)
+        result = reader(mypath)
+        expected = reader(path)
+
+        if path.endswith(".pickle"):
+            # categorical
+            tm.assert_categorical_equal(result, expected)
+        else:
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "writer_name, writer_kwargs, module",
+        [
+            ("to_csv", {}, "os"),
+            ("to_excel", {"engine": "openpyxl"}, "openpyxl"),
+            ("to_feather", {}, "pyarrow"),
+            ("to_html", {}, "os"),
+            ("to_json", {}, "os"),
+            ("to_latex", {}, "os"),
+            ("to_pickle", {}, "os"),
+            ("to_stata", {"time_stamp": pd.to_datetime("2019-01-01 00:00")}, "os"),
+        ],
+    )
+    def test_write_fspath_all(self, writer_name, writer_kwargs, module):
+        if writer_name in ["to_latex"]:  # uses Styler implementation
+            pytest.importorskip("jinja2")
+        p1 = tm.ensure_clean("string")
+        p2 = tm.ensure_clean("fspath")
+        df = pd.DataFrame({"A": [1, 2]})
+
+        with p1 as string, p2 as fspath:
+            pytest.importorskip(module)
+            mypath = CustomFSPath(fspath)
+            writer = getattr(df, writer_name)
+
+            writer(string, **writer_kwargs)
+            writer(mypath, **writer_kwargs)
+            with open(string, "rb") as f_str, open(fspath, "rb") as f_path:
+                if writer_name == "to_excel":
+                    # binary representation of excel contains time creation
+                    # data that causes flaky CI failures
+                    result = pd.read_excel(f_str, **writer_kwargs)
+                    expected = pd.read_excel(f_path, **writer_kwargs)
+                    tm.assert_frame_equal(result, expected)
+                else:
+                    result = f_str.read()
+                    expected = f_path.read()
+                    assert result == expected
+
+    def test_write_fspath_hdf5(self):
+        # Same test as write_fspath_all, except HDF5 files aren't
+        # necessarily byte-for-byte identical for a given dataframe, so we'll
+        # have to read and compare equality
+        pytest.importorskip("tables")
+
+        df = pd.DataFrame({"A": [1, 2]})
+        p1 = tm.ensure_clean("string")
+        p2 = tm.ensure_clean("fspath")
+
+        with p1 as string, p2 as fspath:
+            mypath = CustomFSPath(fspath)
+            df.to_hdf(mypath, key="bar")
+            df.to_hdf(string, key="bar")
+
+            result = pd.read_hdf(fspath, key="bar")
+            expected = pd.read_hdf(string, key="bar")
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.fixture
+def mmap_file(datapath):
+    return datapath("io", "data", "csv", "test_mmap.csv")
+
+
+class TestMMapWrapper:
+    @pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+    def test_constructor_bad_file(self, mmap_file):
+        non_file = StringIO("I am not a file")
+        non_file.fileno = lambda: -1
+
+        # the error raised is different on Windows
+        if is_platform_windows():
+            msg = "The parameter is incorrect"
+            err = OSError
+        else:
+            msg = "[Errno 22]"
+            err = mmap.error
+
+        with pytest.raises(err, match=msg):
+            icom._maybe_memory_map(non_file, True)
+
+        with open(mmap_file, encoding="utf-8") as target:
+            pass
+
+        msg = "I/O operation on closed file"
+        with pytest.raises(ValueError, match=msg):
+            icom._maybe_memory_map(target, True)
+
+    @pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+    def test_next(self, mmap_file):
+        with open(mmap_file, encoding="utf-8") as target:
+            lines = target.readlines()
+
+            with icom.get_handle(
+                target, "r", is_text=True, memory_map=True
+            ) as wrappers:
+                wrapper = wrappers.handle
+                assert isinstance(wrapper.buffer.buffer, mmap.mmap)
+
+                for line in lines:
+                    next_line = next(wrapper)
+                    assert next_line.strip() == line.strip()
+
+                with pytest.raises(StopIteration, match=r"^$"):
+                    next(wrapper)
+
+    def test_unknown_engine(self):
+        with tm.ensure_clean() as path:
+            df = pd.DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=pd.Index(list("ABCD")),
+                index=pd.Index([f"i-{i}" for i in range(30)]),
+            )
+            df.to_csv(path)
+            with pytest.raises(ValueError, match="Unknown engine"):
+                pd.read_csv(path, engine="pyt")
+
+    def test_binary_mode(self):
+        """
+        'encoding' shouldn't be passed to 'open' in binary mode.
+
+        GH 35058
+        """
+        with tm.ensure_clean() as path:
+            df = pd.DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=pd.Index(list("ABCD")),
+                index=pd.Index([f"i-{i}" for i in range(30)]),
+            )
+            df.to_csv(path, mode="w+b")
+            tm.assert_frame_equal(df, pd.read_csv(path, index_col=0))
+
+    @pytest.mark.parametrize("encoding", ["utf-16", "utf-32"])
+    @pytest.mark.parametrize("compression_", ["bz2", "xz"])
+    def test_warning_missing_utf_bom(self, encoding, compression_):
+        """
+        bz2 and xz do not write the byte order mark (BOM) for utf-16/32.
+
+        https://stackoverflow.com/questions/55171439
+
+        GH 35681
+        """
+        df = pd.DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        )
+        with tm.ensure_clean() as path:
+            with tm.assert_produces_warning(UnicodeWarning, match="byte order mark"):
+                df.to_csv(path, compression=compression_, encoding=encoding)
+
+            # reading should fail (otherwise we wouldn't need the warning)
+            msg = (
+                r"UTF-\d+ stream does not start with BOM|"
+                r"'utf-\d+' codec can't decode byte"
+            )
+            with pytest.raises(UnicodeError, match=msg):
+                pd.read_csv(path, compression=compression_, encoding=encoding)
+
+
+def test_is_fsspec_url():
+    assert icom.is_fsspec_url("gcs://pandas/somethingelse.com")
+    assert icom.is_fsspec_url("gs://pandas/somethingelse.com")
+    # the following is the only remote URL that is handled without fsspec
+    assert not icom.is_fsspec_url("http://pandas/somethingelse.com")
+    assert not icom.is_fsspec_url("random:pandas/somethingelse.com")
+    assert not icom.is_fsspec_url("/local/path")
+    assert not icom.is_fsspec_url("relative/local/path")
+    # fsspec URL in string should not be recognized
+    assert not icom.is_fsspec_url("this is not fsspec://url")
+    assert not icom.is_fsspec_url("{'url': 'gs://pandas/somethingelse.com'}")
+    # accept everything that conforms to RFC 3986 schema
+    assert icom.is_fsspec_url("RFC-3986+compliant.spec://something")
+
+
+@pytest.mark.parametrize("encoding", [None, "utf-8"])
+@pytest.mark.parametrize("format", ["csv", "json"])
+def test_codecs_encoding(encoding, format):
+    # GH39247
+    expected = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+    with tm.ensure_clean() as path:
+        with codecs.open(path, mode="w", encoding=encoding) as handle:
+            getattr(expected, f"to_{format}")(handle)
+        with codecs.open(path, mode="r", encoding=encoding) as handle:
+            if format == "csv":
+                df = pd.read_csv(handle, index_col=0)
+            else:
+                df = pd.read_json(handle)
+    tm.assert_frame_equal(expected, df)
+
+
+def test_codecs_get_writer_reader():
+    # GH39247
+    expected = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+    with tm.ensure_clean() as path:
+        with open(path, "wb") as handle:
+            with codecs.getwriter("utf-8")(handle) as encoded:
+                expected.to_csv(encoded)
+        with open(path, "rb") as handle:
+            with codecs.getreader("utf-8")(handle) as encoded:
+                df = pd.read_csv(encoded, index_col=0)
+    tm.assert_frame_equal(expected, df)
+
+
+@pytest.mark.parametrize(
+    "io_class,mode,msg",
+    [
+        (BytesIO, "t", "a bytes-like object is required, not 'str'"),
+        (StringIO, "b", "string argument expected, got 'bytes'"),
+    ],
+)
+def test_explicit_encoding(io_class, mode, msg):
+    # GH39247; this test makes sure that if a user provides mode="*t" or "*b",
+    # it is used. In the case of this test it leads to an error as intentionally the
+    # wrong mode is requested
+    expected = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+    with io_class() as buffer:
+        with pytest.raises(TypeError, match=msg):
+            expected.to_csv(buffer, mode=f"w{mode}")
+
+
+@pytest.mark.parametrize("encoding_errors", ["strict", "replace"])
+@pytest.mark.parametrize("format", ["csv", "json"])
+def test_encoding_errors(encoding_errors, format):
+    # GH39450
+    msg = "'utf-8' codec can't decode byte"
+    bad_encoding = b"\xe4"
+
+    if format == "csv":
+        content = b"," + bad_encoding + b"\n" + bad_encoding * 2 + b"," + bad_encoding
+        reader = partial(pd.read_csv, index_col=0)
+    else:
+        content = (
+            b'{"'
+            + bad_encoding * 2
+            + b'": {"'
+            + bad_encoding
+            + b'":"'
+            + bad_encoding
+            + b'"}}'
+        )
+        reader = partial(pd.read_json, orient="index")
+    with tm.ensure_clean() as path:
+        file = Path(path)
+        file.write_bytes(content)
+
+        if encoding_errors != "replace":
+            with pytest.raises(UnicodeDecodeError, match=msg):
+                reader(path, encoding_errors=encoding_errors)
+        else:
+            df = reader(path, encoding_errors=encoding_errors)
+            decoded = bad_encoding.decode(errors=encoding_errors)
+            expected = pd.DataFrame({decoded: [decoded]}, index=[decoded * 2])
+            tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("encoding_errors", [0, None])
+def test_encoding_errors_badtype(encoding_errors):
+    # GH 59075
+    content = StringIO("A,B\n1,2\n3,4\n")
+    reader = partial(pd.read_csv, encoding_errors=encoding_errors)
+    expected_error = "encoding_errors must be a string, got "
+    expected_error += f"{type(encoding_errors).__name__}"
+    with pytest.raises(ValueError, match=expected_error):
+        reader(content)
+
+
+def test_bad_encdoing_errors():
+    # GH 39777
+    with tm.ensure_clean() as path:
+        with pytest.raises(LookupError, match="unknown error handler name"):
+            icom.get_handle(path, "w", errors="bad")
+
+
+@pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+def test_errno_attribute():
+    # GH 13872
+    with pytest.raises(FileNotFoundError, match="\\[Errno 2\\]") as err:
+        pd.read_csv("doesnt_exist")
+        assert err.errno == errno.ENOENT
+
+
+def test_fail_mmap():
+    with pytest.raises(UnsupportedOperation, match="fileno"):
+        with BytesIO() as buffer:
+            icom.get_handle(buffer, "rb", memory_map=True)
+
+
+def test_close_on_error():
+    # GH 47136
+    class TestError:
+        def close(self):
+            raise OSError("test")
+
+    with pytest.raises(OSError, match="test"):
+        with BytesIO() as buffer:
+            with icom.get_handle(buffer, "rb") as handles:
+                handles.created_handles.append(TestError())
+
+
+@td.skip_if_no("fsspec", min_version="2023.1.0")
+@pytest.mark.parametrize("compression", [None, "infer"])
+def test_read_csv_chained_url_no_error(compression):
+    # GH 60100
+    tar_file_path = "pandas/tests/io/data/tar/test-csv.tar"
+    chained_file_url = f"tar://test.csv::file://{tar_file_path}"
+
+    result = pd.read_csv(chained_file_url, compression=compression, sep=";")
+    expected = pd.DataFrame({"1": {0: 3}, "2": {0: 4}})
+
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize(
+    "reader",
+    [
+        pd.read_csv,
+        pd.read_fwf,
+        pd.read_excel,
+        pd.read_feather,
+        pd.read_hdf,
+        pd.read_stata,
+        pd.read_sas,
+        pd.read_json,
+        pd.read_pickle,
+    ],
+)
+def test_pickle_reader(reader):
+    # GH 22265
+    with BytesIO() as buffer:
+        pickle.dump(reader, buffer)
+
+
+@td.skip_if_no("pyarrow")
+def test_pyarrow_read_csv_datetime_dtype():
+    # GH 59904
+    data = '"date"\n"20/12/2025"\n""\n"31/12/2020"'
+    result = pd.read_csv(
+        StringIO(data), parse_dates=["date"], dayfirst=True, dtype_backend="pyarrow"
+    )
+
+    expect_data = pd.to_datetime(["20/12/2025", pd.NaT, "31/12/2020"], dayfirst=True)
+    expect = pd.DataFrame({"date": expect_data})
+
+    tm.assert_frame_equal(expect, result)
diff --git a/pandas/pandas/tests/io/test_compression.py b/pandas/pandas/tests/io/test_compression.py
new file mode 100644
index 0000000000000..fd1e9b4fdf211
--- /dev/null
+++ b/pandas/pandas/tests/io/test_compression.py
@@ -0,0 +1,378 @@
+import gzip
+import io
+import os
+from pathlib import Path
+import subprocess
+import sys
+import tarfile
+import textwrap
+import time
+import zipfile
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_windows
+
+import pandas as pd
+import pandas._testing as tm
+
+import pandas.io.common as icom
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        pd.DataFrame(
+            100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            columns=["X", "Y", "Z"],
+        ),
+        pd.Series(100 * [0.123456, 0.234567, 0.567567], name="X"),
+    ],
+)
+@pytest.mark.parametrize("method", ["to_pickle", "to_json", "to_csv"])
+def test_compression_size(obj, method, compression_only):
+    if compression_only == "tar":
+        compression_only = {"method": "tar", "mode": "w:gz"}
+
+    with tm.ensure_clean() as path:
+        getattr(obj, method)(path, compression=compression_only)
+        compressed_size = os.path.getsize(path)
+        getattr(obj, method)(path, compression=None)
+        uncompressed_size = os.path.getsize(path)
+        assert uncompressed_size > compressed_size
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        pd.DataFrame(
+            100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            columns=["X", "Y", "Z"],
+        ),
+        pd.Series(100 * [0.123456, 0.234567, 0.567567], name="X"),
+    ],
+)
+@pytest.mark.parametrize("method", ["to_csv", "to_json"])
+def test_compression_size_fh(obj, method, compression_only):
+    with tm.ensure_clean() as path:
+        with icom.get_handle(
+            path,
+            "w:gz" if compression_only == "tar" else "w",
+            compression=compression_only,
+        ) as handles:
+            getattr(obj, method)(handles.handle)
+            assert not handles.handle.closed
+        compressed_size = os.path.getsize(path)
+    with tm.ensure_clean() as path:
+        with icom.get_handle(path, "w", compression=None) as handles:
+            getattr(obj, method)(handles.handle)
+            assert not handles.handle.closed
+        uncompressed_size = os.path.getsize(path)
+        assert uncompressed_size > compressed_size
+
+
+@pytest.mark.parametrize(
+    "write_method, write_kwargs, read_method",
+    [
+        ("to_csv", {"index": False}, pd.read_csv),
+        ("to_json", {}, pd.read_json),
+        ("to_pickle", {}, pd.read_pickle),
+    ],
+)
+def test_dataframe_compression_defaults_to_infer(
+    write_method, write_kwargs, read_method, compression_only, compression_to_extension
+):
+    # GH22004
+    input = pd.DataFrame([[1.0, 0, -4], [3.4, 5, 2]], columns=["X", "Y", "Z"])
+    extension = compression_to_extension[compression_only]
+    with tm.ensure_clean("compressed" + extension) as path:
+        getattr(input, write_method)(path, **write_kwargs)
+        output = read_method(path, compression=compression_only)
+    tm.assert_frame_equal(output, input)
+
+
+@pytest.mark.parametrize(
+    "write_method,write_kwargs,read_method,read_kwargs",
+    [
+        ("to_csv", {"index": False, "header": True}, pd.read_csv, {"squeeze": True}),
+        ("to_json", {}, pd.read_json, {"typ": "series"}),
+        ("to_pickle", {}, pd.read_pickle, {}),
+    ],
+)
+def test_series_compression_defaults_to_infer(
+    write_method,
+    write_kwargs,
+    read_method,
+    read_kwargs,
+    compression_only,
+    compression_to_extension,
+):
+    # GH22004
+    input = pd.Series([0, 5, -2, 10], name="X")
+    extension = compression_to_extension[compression_only]
+    with tm.ensure_clean("compressed" + extension) as path:
+        getattr(input, write_method)(path, **write_kwargs)
+        if "squeeze" in read_kwargs:
+            kwargs = read_kwargs.copy()
+            del kwargs["squeeze"]
+            output = read_method(path, compression=compression_only, **kwargs).squeeze(
+                "columns"
+            )
+        else:
+            output = read_method(path, compression=compression_only, **read_kwargs)
+    tm.assert_series_equal(output, input, check_names=False)
+
+
+def test_compression_warning(compression_only):
+    # Assert that passing a file object to to_csv while explicitly specifying a
+    # compression protocol triggers a RuntimeWarning, as per GH21227.
+    df = pd.DataFrame(
+        100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+        columns=["X", "Y", "Z"],
+    )
+    with tm.ensure_clean() as path:
+        with icom.get_handle(path, "w", compression=compression_only) as handles:
+            with tm.assert_produces_warning(RuntimeWarning, match="has no effect"):
+                df.to_csv(handles.handle, compression=compression_only)
+
+
+def test_compression_binary(compression_only):
+    """
+    Binary file handles support compression.
+
+    GH22555
+    """
+    df = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+
+    # with a file
+    with tm.ensure_clean() as path:
+        with open(path, mode="wb") as file:
+            df.to_csv(file, mode="wb", compression=compression_only)
+            file.seek(0)  # file shouldn't be closed
+        tm.assert_frame_equal(
+            df, pd.read_csv(path, index_col=0, compression=compression_only)
+        )
+
+    # with BytesIO
+    file = io.BytesIO()
+    df.to_csv(file, mode="wb", compression=compression_only)
+    file.seek(0)  # file shouldn't be closed
+    tm.assert_frame_equal(
+        df, pd.read_csv(file, index_col=0, compression=compression_only)
+    )
+
+
+def test_gzip_reproducibility_file_name():
+    """
+    Gzip should create reproducible archives with mtime.
+
+    Note: Archives created with different filenames will still be different!
+
+    GH 28103
+    """
+    df = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+    compression_options = {"method": "gzip", "mtime": 1}
+
+    # test for filename
+    with tm.ensure_clean() as path:
+        path = Path(path)
+        df.to_csv(path, compression=compression_options)
+        time.sleep(0.1)
+        output = path.read_bytes()
+        df.to_csv(path, compression=compression_options)
+        assert output == path.read_bytes()
+
+
+def test_gzip_reproducibility_file_object():
+    """
+    Gzip should create reproducible archives with mtime.
+
+    GH 28103
+    """
+    df = pd.DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=pd.Index(list("ABCD")),
+        index=pd.Index([f"i-{i}" for i in range(30)]),
+    )
+    compression_options = {"method": "gzip", "mtime": 1}
+
+    # test for file object
+    buffer = io.BytesIO()
+    df.to_csv(buffer, compression=compression_options, mode="wb")
+    output = buffer.getvalue()
+    time.sleep(0.1)
+    buffer = io.BytesIO()
+    df.to_csv(buffer, compression=compression_options, mode="wb")
+    assert output == buffer.getvalue()
+
+
+@pytest.mark.single_cpu
+def test_with_missing_lzma():
+    """Tests if import pandas works when lzma is not present."""
+    # https://github.com/pandas-dev/pandas/issues/27575
+    code = textwrap.dedent(
+        """\
+        import sys
+        sys.modules['lzma'] = None
+        import pandas
+        """
+    )
+    subprocess.check_output([sys.executable, "-c", code], stderr=subprocess.PIPE)
+
+
+@pytest.mark.single_cpu
+def test_with_missing_lzma_runtime():
+    """Tests if ModuleNotFoundError is hit when calling lzma without
+    having the module available.
+    """
+    code = textwrap.dedent(
+        """
+        import sys
+        import pytest
+        sys.modules['lzma'] = None
+        import pandas as pd
+        df = pd.DataFrame()
+        with pytest.raises(ModuleNotFoundError, match='import of lzma'):
+            df.to_csv('foo.csv', compression='xz')
+        """
+    )
+    subprocess.check_output([sys.executable, "-c", code], stderr=subprocess.PIPE)
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        pd.DataFrame(
+            100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            columns=["X", "Y", "Z"],
+        ),
+        pd.Series(100 * [0.123456, 0.234567, 0.567567], name="X"),
+    ],
+)
+@pytest.mark.parametrize("method", ["to_pickle", "to_json", "to_csv"])
+def test_gzip_compression_level(obj, method):
+    # GH33196
+    with tm.ensure_clean() as path:
+        getattr(obj, method)(path, compression="gzip")
+        compressed_size_default = os.path.getsize(path)
+        getattr(obj, method)(path, compression={"method": "gzip", "compresslevel": 1})
+        compressed_size_fast = os.path.getsize(path)
+        assert compressed_size_default < compressed_size_fast
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        pd.DataFrame(
+            100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            columns=["X", "Y", "Z"],
+        ),
+        pd.Series(100 * [0.123456, 0.234567, 0.567567], name="X"),
+    ],
+)
+@pytest.mark.parametrize("method", ["to_pickle", "to_json", "to_csv"])
+def test_xz_compression_level_read(obj, method):
+    with tm.ensure_clean() as path:
+        getattr(obj, method)(path, compression="xz")
+        compressed_size_default = os.path.getsize(path)
+        getattr(obj, method)(path, compression={"method": "xz", "preset": 1})
+        compressed_size_fast = os.path.getsize(path)
+        assert compressed_size_default < compressed_size_fast
+        if method == "to_csv":
+            pd.read_csv(path, compression="xz")
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        pd.DataFrame(
+            100 * [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+            columns=["X", "Y", "Z"],
+        ),
+        pd.Series(100 * [0.123456, 0.234567, 0.567567], name="X"),
+    ],
+)
+@pytest.mark.parametrize("method", ["to_pickle", "to_json", "to_csv"])
+def test_bzip_compression_level(obj, method):
+    """GH33196 bzip needs file size > 100k to show a size difference between
+    compression levels, so here we just check if the call works when
+    compression is passed as a dict.
+    """
+    with tm.ensure_clean() as path:
+        getattr(obj, method)(path, compression={"method": "bz2", "compresslevel": 1})
+
+
+@pytest.mark.parametrize(
+    "suffix,archive",
+    [
+        (".zip", zipfile.ZipFile),
+        (".tar", tarfile.TarFile),
+    ],
+)
+def test_empty_archive_zip(suffix, archive):
+    with tm.ensure_clean(filename=suffix) as path:
+        with archive(path, "w"):
+            pass
+        with pytest.raises(ValueError, match="Zero files found"):
+            pd.read_csv(path)
+
+
+def test_ambiguous_archive_zip():
+    with tm.ensure_clean(filename=".zip") as path:
+        with zipfile.ZipFile(path, "w") as file:
+            file.writestr("a.csv", "foo,bar")
+            file.writestr("b.csv", "foo,bar")
+        with pytest.raises(ValueError, match="Multiple files found in ZIP file"):
+            pd.read_csv(path)
+
+
+def test_ambiguous_archive_tar(tmp_path):
+    csvAPath = tmp_path / "a.csv"
+    with open(csvAPath, "w", encoding="utf-8") as a:
+        a.write("foo,bar\n")
+    csvBPath = tmp_path / "b.csv"
+    with open(csvBPath, "w", encoding="utf-8") as b:
+        b.write("foo,bar\n")
+
+    tarpath = tmp_path / "archive.tar"
+    with tarfile.TarFile(tarpath, "w") as tar:
+        tar.add(csvAPath, "a.csv")
+        tar.add(csvBPath, "b.csv")
+
+    with pytest.raises(ValueError, match="Multiple files found in TAR archive"):
+        pd.read_csv(tarpath)
+
+
+def test_tar_gz_to_different_filename():
+    with tm.ensure_clean(filename=".foo") as file:
+        pd.DataFrame(
+            [["1", "2"]],
+            columns=["foo", "bar"],
+        ).to_csv(file, compression={"method": "tar", "mode": "w:gz"}, index=False)
+        with gzip.open(file) as uncompressed:
+            with tarfile.TarFile(fileobj=uncompressed) as archive:
+                members = archive.getmembers()
+                assert len(members) == 1
+                content = archive.extractfile(members[0]).read().decode("utf8")
+
+                if is_platform_windows():
+                    expected = "foo,bar\r\n1,2\r\n"
+                else:
+                    expected = "foo,bar\n1,2\n"
+
+                assert content == expected
+
+
+def test_tar_no_error_on_close():
+    with io.BytesIO() as buffer:
+        with icom._BytesTarFile(fileobj=buffer, mode="w"):
+            pass
diff --git a/pandas/pandas/tests/io/test_feather.py b/pandas/pandas/tests/io/test_feather.py
new file mode 100644
index 0000000000000..e778193c147c1
--- /dev/null
+++ b/pandas/pandas/tests/io/test_feather.py
@@ -0,0 +1,296 @@
+"""test feather-format compat"""
+
+from datetime import datetime
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas.compat.pyarrow import (
+    pa_version_under18p0,
+    pa_version_under19p0,
+)
+
+import pandas as pd
+import pandas._testing as tm
+
+from pandas.io.feather_format import read_feather, to_feather  # isort:skip
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+pa = pytest.importorskip("pyarrow")
+
+
+@pytest.mark.single_cpu
+class TestFeather:
+    def check_error_on_write(self, df, exc, err_msg):
+        # check that we are raising the exception
+        # on writing
+
+        with pytest.raises(exc, match=err_msg):
+            with tm.ensure_clean() as path:
+                to_feather(df, path)
+
+    def check_external_error_on_write(self, df):
+        # check that we are raising the exception
+        # on writing
+
+        with tm.external_error_raised(Exception):
+            with tm.ensure_clean() as path:
+                to_feather(df, path)
+
+    def check_round_trip(self, df, expected=None, write_kwargs=None, **read_kwargs):
+        if write_kwargs is None:
+            write_kwargs = {}
+        if expected is None:
+            expected = df.copy()
+
+        with tm.ensure_clean() as path:
+            to_feather(df, path, **write_kwargs)
+
+            result = read_feather(path, **read_kwargs)
+
+            tm.assert_frame_equal(result, expected)
+
+    def test_error(self):
+        msg = "feather only support IO with DataFrames"
+        for obj in [
+            pd.Series([1, 2, 3]),
+            1,
+            "foo",
+            pd.Timestamp("20130101"),
+            np.array([1, 2, 3]),
+        ]:
+            self.check_error_on_write(obj, ValueError, msg)
+
+    def test_basic(self):
+        tz = zoneinfo.ZoneInfo("US/Eastern")
+        df = pd.DataFrame(
+            {
+                "string": list("abc"),
+                "int": list(range(1, 4)),
+                "uint": np.arange(3, 6).astype("u1"),
+                "float": np.arange(4.0, 7.0, dtype="float64"),
+                "float_with_null": [1.0, np.nan, 3],
+                "bool": [True, False, True],
+                "bool_with_null": [True, np.nan, False],
+                "cat": pd.Categorical(list("abc")),
+                "dt": pd.DatetimeIndex(
+                    list(pd.date_range("20130101", periods=3)), freq=None
+                ),
+                "dttz": pd.DatetimeIndex(
+                    list(pd.date_range("20130101", periods=3, tz=tz)),
+                    freq=None,
+                ),
+                "dt_with_null": [
+                    pd.Timestamp("20130101"),
+                    pd.NaT,
+                    pd.Timestamp("20130103"),
+                ],
+                "dtns": pd.DatetimeIndex(
+                    list(pd.date_range("20130101", periods=3, freq="ns")), freq=None
+                ),
+            }
+        )
+        df["periods"] = pd.period_range("2013", freq="M", periods=3)
+        df["timedeltas"] = pd.timedelta_range("1 day", periods=3)
+        df["intervals"] = pd.interval_range(0, 3, 3)
+
+        assert df.dttz.dtype.tz.key == "US/Eastern"
+
+        expected = df.copy()
+        expected.loc[1, "bool_with_null"] = None
+        self.check_round_trip(df, expected=expected)
+
+    def test_duplicate_columns(self):
+        # https://github.com/wesm/feather/issues/53
+        # not currently able to handle duplicate columns
+        df = pd.DataFrame(np.arange(12).reshape(4, 3), columns=list("aaa")).copy()
+        self.check_external_error_on_write(df)
+
+    def test_read_columns(self):
+        # GH 24025
+        df = pd.DataFrame(
+            {
+                "col1": list("abc"),
+                "col2": list(range(1, 4)),
+                "col3": list("xyz"),
+                "col4": list(range(4, 7)),
+            }
+        )
+        columns = ["col1", "col3"]
+        self.check_round_trip(df, expected=df[columns], columns=columns)
+
+    def test_read_columns_different_order(self):
+        # GH 33878
+        df = pd.DataFrame({"A": [1, 2], "B": ["x", "y"], "C": [True, False]})
+        expected = df[["B", "A"]]
+        self.check_round_trip(df, expected, columns=["B", "A"])
+
+    def test_unsupported_other(self):
+        # mixed python objects
+        df = pd.DataFrame({"a": ["a", 1, 2.0]})
+        self.check_external_error_on_write(df)
+
+    def test_rw_use_threads(self):
+        df = pd.DataFrame({"A": np.arange(100000)})
+        self.check_round_trip(df, use_threads=True)
+        self.check_round_trip(df, use_threads=False)
+
+    def test_path_pathlib(self):
+        df = pd.DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        ).reset_index()
+        result = tm.round_trip_pathlib(df.to_feather, read_feather)
+        tm.assert_frame_equal(df, result)
+
+    def test_passthrough_keywords(self):
+        df = pd.DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        ).reset_index()
+        self.check_round_trip(df, write_kwargs={"version": 1})
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_http_path(self, feather_file, httpserver):
+        # GH 29055
+        expected = read_feather(feather_file)
+        with open(feather_file, "rb") as f:
+            httpserver.serve_content(content=f.read())
+            res = read_feather(httpserver.url)
+        tm.assert_frame_equal(expected, res)
+
+    def test_read_feather_dtype_backend(
+        self, string_storage, dtype_backend, using_infer_string
+    ):
+        # GH#50765
+        df = pd.DataFrame(
+            {
+                "a": pd.Series([1, np.nan, 3], dtype="Int64"),
+                "b": pd.Series([1, 2, 3], dtype="Int64"),
+                "c": pd.Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": pd.Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": [True, False, None],
+                "f": [True, False, True],
+                "g": ["a", "b", "c"],
+                "h": ["a", "b", None],
+            }
+        )
+
+        with tm.ensure_clean() as path:
+            to_feather(df, path)
+            with pd.option_context("mode.string_storage", string_storage):
+                result = read_feather(path, dtype_backend=dtype_backend)
+
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            if using_infer_string:
+                string_dtype = pd.ArrowDtype(pa.large_string())
+            else:
+                string_dtype = pd.ArrowDtype(pa.string())
+        else:
+            string_dtype = pd.StringDtype(string_storage)
+
+        expected = pd.DataFrame(
+            {
+                "a": pd.Series([1, np.nan, 3], dtype="Int64"),
+                "b": pd.Series([1, 2, 3], dtype="Int64"),
+                "c": pd.Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": pd.Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": pd.Series([True, False, pd.NA], dtype="boolean"),
+                "f": pd.Series([True, False, True], dtype="boolean"),
+                "g": pd.Series(["a", "b", "c"], dtype=string_dtype),
+                "h": pd.Series(["a", "b", None], dtype=string_dtype),
+            }
+        )
+
+        if dtype_backend == "pyarrow":
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = pd.DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                    for col in expected.columns
+                }
+            )
+
+        if using_infer_string:
+            expected.columns = expected.columns.astype(
+                pd.StringDtype(string_storage, na_value=np.nan)
+            )
+        tm.assert_frame_equal(result, expected)
+
+    def test_int_columns_and_index(self):
+        df = pd.DataFrame({"a": [1, 2, 3]}, index=pd.Index([3, 4, 5], name="test"))
+        self.check_round_trip(df)
+
+    def test_invalid_dtype_backend(self):
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        df = pd.DataFrame({"int": list(range(1, 4))})
+        with tm.ensure_clean("tmp.feather") as path:
+            df.to_feather(path)
+            with pytest.raises(ValueError, match=msg):
+                read_feather(path, dtype_backend="numpy")
+
+    def test_string_inference(self, tmp_path, using_infer_string):
+        # GH#54431
+        path = tmp_path / "test_string_inference.p"
+        df = pd.DataFrame(data={"a": ["x", "y"]})
+        df.to_feather(path)
+        with pd.option_context("future.infer_string", True):
+            result = read_feather(path)
+        dtype = pd.StringDtype(na_value=np.nan)
+        expected = pd.DataFrame(
+            data={"a": ["x", "y"]}, dtype=pd.StringDtype(na_value=np.nan)
+        )
+        expected = pd.DataFrame(
+            data={"a": ["x", "y"]},
+            dtype=dtype,
+            columns=pd.Index(
+                ["a"],
+                dtype=object
+                if pa_version_under19p0 and not using_infer_string
+                else dtype,
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.skipif(pa_version_under18p0, reason="not supported before 18.0")
+    def test_string_inference_string_view_type(self, tmp_path):
+        # GH#54798
+        import pyarrow as pa
+        from pyarrow import feather
+
+        path = tmp_path / "string_view.parquet"
+        table = pa.table({"a": pa.array([None, "b", "c"], pa.string_view())})
+        feather.write_feather(table, path)
+
+        with pd.option_context("future.infer_string", True):
+            result = read_feather(path)
+
+            expected = pd.DataFrame(
+                data={"a": [None, "b", "c"]}, dtype=pd.StringDtype(na_value=np.nan)
+            )
+        tm.assert_frame_equal(result, expected)
+
+    def test_out_of_bounds_datetime_to_feather(self):
+        # GH#47832
+        df = pd.DataFrame(
+            {
+                "date": [
+                    datetime.fromisoformat("1654-01-01"),
+                    datetime.fromisoformat("1920-01-01"),
+                ],
+            }
+        )
+        self.check_round_trip(df)
diff --git a/pandas/pandas/tests/io/test_fsspec.py b/pandas/pandas/tests/io/test_fsspec.py
new file mode 100644
index 0000000000000..2e3e74a9d31ff
--- /dev/null
+++ b/pandas/pandas/tests/io/test_fsspec.py
@@ -0,0 +1,359 @@
+import io
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import HAS_PYARROW
+
+from pandas import (
+    DataFrame,
+    date_range,
+    read_csv,
+    read_excel,
+    read_feather,
+    read_json,
+    read_parquet,
+    read_pickle,
+    read_stata,
+    read_table,
+)
+import pandas._testing as tm
+from pandas.util import _test_decorators as td
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture
+def fsspectest():
+    pytest.importorskip("fsspec")
+    from fsspec import register_implementation
+    from fsspec.implementations.memory import MemoryFileSystem
+    from fsspec.registry import _registry as registry
+
+    class TestMemoryFS(MemoryFileSystem):
+        protocol = "testmem"
+        test = [None]
+
+        def __init__(self, **kwargs) -> None:
+            self.test[0] = kwargs.pop("test", None)
+            super().__init__(**kwargs)
+
+    register_implementation("testmem", TestMemoryFS, clobber=True)
+    yield TestMemoryFS()
+    registry.pop("testmem", None)
+    TestMemoryFS.test[0] = None
+    TestMemoryFS.store.clear()
+
+
+@pytest.fixture
+def df1():
+    return DataFrame(
+        {
+            "int": [1, 3],
+            "float": [2.0, np.nan],
+            "str": ["t", "s"],
+            "dt": date_range("2018-06-18", periods=2),
+        }
+    )
+
+
+@pytest.fixture
+def cleared_fs():
+    fsspec = pytest.importorskip("fsspec")
+
+    memfs = fsspec.filesystem("memory")
+    yield memfs
+    memfs.store.clear()
+
+
+def test_read_csv(cleared_fs, df1):
+    text = str(df1.to_csv(index=False)).encode()
+    with cleared_fs.open("test/test.csv", "wb") as w:
+        w.write(text)
+    df2 = read_csv("memory://test/test.csv", parse_dates=["dt"])
+
+    expected = df1.copy()
+    expected["dt"] = expected["dt"].astype("M8[s]")
+    tm.assert_frame_equal(df2, expected)
+
+
+def test_reasonable_error(monkeypatch, cleared_fs):
+    from fsspec.registry import known_implementations
+
+    with pytest.raises(ValueError, match="nosuchprotocol"):
+        read_csv("nosuchprotocol://test/test.csv")
+    err_msg = "test error message"
+    monkeypatch.setitem(
+        known_implementations,
+        "couldexist",
+        {"class": "unimportable.CouldExist", "err": err_msg},
+    )
+    with pytest.raises(ImportError, match=err_msg):
+        read_csv("couldexist://test/test.csv")
+
+
+def test_to_csv(cleared_fs, df1):
+    df1.to_csv("memory://test/test.csv", index=True)
+
+    df2 = read_csv("memory://test/test.csv", parse_dates=["dt"], index_col=0)
+
+    expected = df1.copy()
+    expected["dt"] = expected["dt"].astype("M8[s]")
+    tm.assert_frame_equal(df2, expected)
+
+
+def test_to_excel(cleared_fs, df1):
+    pytest.importorskip("openpyxl")
+    ext = "xlsx"
+    path = f"memory://test/test.{ext}"
+    df1.to_excel(path, index=True)
+
+    df2 = read_excel(path, parse_dates=["dt"], index_col=0)
+
+    expected = df1.copy()
+    expected["dt"] = expected["dt"].astype("M8[s]")
+    tm.assert_frame_equal(df2, expected)
+
+
+@pytest.mark.parametrize("binary_mode", [False, True])
+def test_to_csv_fsspec_object(cleared_fs, binary_mode, df1):
+    fsspec = pytest.importorskip("fsspec")
+
+    path = "memory://test/test.csv"
+    mode = "wb" if binary_mode else "w"
+    with fsspec.open(path, mode=mode).open() as fsspec_object:
+        df1.to_csv(fsspec_object, index=True)
+        assert not fsspec_object.closed
+
+    mode = mode.replace("w", "r")
+    with fsspec.open(path, mode=mode) as fsspec_object:
+        df2 = read_csv(
+            fsspec_object,
+            parse_dates=["dt"],
+            index_col=0,
+        )
+        assert not fsspec_object.closed
+
+    expected = df1.copy()
+    expected["dt"] = expected["dt"].astype("M8[s]")
+    tm.assert_frame_equal(df2, expected)
+
+
+def test_csv_options(fsspectest):
+    df = DataFrame({"a": [0]})
+    df.to_csv(
+        "testmem://test/test.csv", storage_options={"test": "csv_write"}, index=False
+    )
+    assert fsspectest.test[0] == "csv_write"
+    read_csv("testmem://test/test.csv", storage_options={"test": "csv_read"})
+    assert fsspectest.test[0] == "csv_read"
+
+
+def test_read_table_options(fsspectest):
+    # GH #39167
+    df = DataFrame({"a": [0]})
+    df.to_csv(
+        "testmem://test/test.csv", storage_options={"test": "csv_write"}, index=False
+    )
+    assert fsspectest.test[0] == "csv_write"
+    read_table("testmem://test/test.csv", storage_options={"test": "csv_read"})
+    assert fsspectest.test[0] == "csv_read"
+
+
+def test_excel_options(fsspectest):
+    pytest.importorskip("openpyxl")
+    extension = "xlsx"
+
+    df = DataFrame({"a": [0]})
+
+    path = f"testmem://test/test.{extension}"
+
+    df.to_excel(path, storage_options={"test": "write"}, index=False)
+    assert fsspectest.test[0] == "write"
+    read_excel(path, storage_options={"test": "read"})
+    assert fsspectest.test[0] == "read"
+
+
+@pytest.mark.xfail(
+    using_string_dtype() and HAS_PYARROW, reason="TODO(infer_string) fastparquet"
+)
+def test_to_parquet_new_file(cleared_fs, df1):
+    """Regression test for writing to a not-yet-existent GCS Parquet file."""
+    pytest.importorskip("fastparquet")
+
+    df1.to_parquet(
+        "memory://test/test.csv", index=True, engine="fastparquet", compression=None
+    )
+
+
+def test_arrowparquet_options(fsspectest):
+    """Regression test for writing to a not-yet-existent GCS Parquet file."""
+    pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [0]})
+    df.to_parquet(
+        "testmem://test/test.csv",
+        engine="pyarrow",
+        compression=None,
+        storage_options={"test": "parquet_write"},
+    )
+    assert fsspectest.test[0] == "parquet_write"
+    read_parquet(
+        "testmem://test/test.csv",
+        engine="pyarrow",
+        storage_options={"test": "parquet_read"},
+    )
+    assert fsspectest.test[0] == "parquet_read"
+
+
+def test_fastparquet_options(fsspectest):
+    """Regression test for writing to a not-yet-existent GCS Parquet file."""
+    pytest.importorskip("fastparquet")
+
+    df = DataFrame({"a": [0]})
+    df.to_parquet(
+        "testmem://test/test.csv",
+        engine="fastparquet",
+        compression=None,
+        storage_options={"test": "parquet_write"},
+    )
+    assert fsspectest.test[0] == "parquet_write"
+    read_parquet(
+        "testmem://test/test.csv",
+        engine="fastparquet",
+        storage_options={"test": "parquet_read"},
+    )
+    assert fsspectest.test[0] == "parquet_read"
+
+
+@pytest.mark.single_cpu
+def test_from_s3_csv(s3_public_bucket_with_data, tips_file, s3so):
+    pytest.importorskip("s3fs")
+    tm.assert_equal(
+        read_csv(
+            f"s3://{s3_public_bucket_with_data.name}/tips.csv", storage_options=s3so
+        ),
+        read_csv(tips_file),
+    )
+    # the following are decompressed by pandas, not fsspec
+    tm.assert_equal(
+        read_csv(
+            f"s3://{s3_public_bucket_with_data.name}/tips.csv.gz", storage_options=s3so
+        ),
+        read_csv(tips_file),
+    )
+    tm.assert_equal(
+        read_csv(
+            f"s3://{s3_public_bucket_with_data.name}/tips.csv.bz2", storage_options=s3so
+        ),
+        read_csv(tips_file),
+    )
+
+
+@pytest.mark.single_cpu
+@pytest.mark.parametrize("protocol", ["s3", "s3a", "s3n"])
+def test_s3_protocols(s3_public_bucket_with_data, tips_file, protocol, s3so):
+    pytest.importorskip("s3fs")
+    tm.assert_equal(
+        read_csv(
+            f"{protocol}://{s3_public_bucket_with_data.name}/tips.csv",
+            storage_options=s3so,
+        ),
+        read_csv(tips_file),
+    )
+
+
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string) fastparquet")
+@pytest.mark.single_cpu
+def test_s3_parquet(s3_public_bucket, s3so, df1):
+    pytest.importorskip("fastparquet")
+    pytest.importorskip("s3fs")
+
+    fn = f"s3://{s3_public_bucket.name}/test.parquet"
+    df1.to_parquet(
+        fn, index=False, engine="fastparquet", compression=None, storage_options=s3so
+    )
+    df2 = read_parquet(fn, engine="fastparquet", storage_options=s3so)
+    tm.assert_equal(df1, df2)
+
+
+@td.skip_if_installed("fsspec")
+def test_not_present_exception():
+    msg = "Missing optional dependency 'fsspec'|fsspec library is required"
+    with pytest.raises(ImportError, match=msg):
+        read_csv("memory://test/test.csv")
+
+
+def test_feather_options(fsspectest):
+    pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [0]})
+    df.to_feather("testmem://mockfile", storage_options={"test": "feather_write"})
+    assert fsspectest.test[0] == "feather_write"
+    out = read_feather("testmem://mockfile", storage_options={"test": "feather_read"})
+    assert fsspectest.test[0] == "feather_read"
+    tm.assert_frame_equal(df, out)
+
+
+def test_pickle_options(fsspectest):
+    df = DataFrame({"a": [0]})
+    df.to_pickle("testmem://mockfile", storage_options={"test": "pickle_write"})
+    assert fsspectest.test[0] == "pickle_write"
+    out = read_pickle("testmem://mockfile", storage_options={"test": "pickle_read"})
+    assert fsspectest.test[0] == "pickle_read"
+    tm.assert_frame_equal(df, out)
+
+
+def test_json_options(fsspectest, compression):
+    df = DataFrame({"a": [0]})
+    df.to_json(
+        "testmem://mockfile",
+        compression=compression,
+        storage_options={"test": "json_write"},
+    )
+    assert fsspectest.test[0] == "json_write"
+    out = read_json(
+        "testmem://mockfile",
+        compression=compression,
+        storage_options={"test": "json_read"},
+    )
+    assert fsspectest.test[0] == "json_read"
+    tm.assert_frame_equal(df, out)
+
+
+def test_stata_options(fsspectest):
+    df = DataFrame({"a": [0]})
+    df.to_stata(
+        "testmem://mockfile", storage_options={"test": "stata_write"}, write_index=False
+    )
+    assert fsspectest.test[0] == "stata_write"
+    out = read_stata("testmem://mockfile", storage_options={"test": "stata_read"})
+    assert fsspectest.test[0] == "stata_read"
+    tm.assert_frame_equal(df, out.astype("int64"))
+
+
+def test_markdown_options(fsspectest):
+    pytest.importorskip("tabulate")
+    df = DataFrame({"a": [0]})
+    df.to_markdown("testmem://mockfile", storage_options={"test": "md_write"})
+    assert fsspectest.test[0] == "md_write"
+    assert fsspectest.cat("testmem://mockfile")
+
+
+def test_non_fsspec_options():
+    pytest.importorskip("pyarrow")
+    with pytest.raises(ValueError, match="storage_options"):
+        read_csv("localfile", storage_options={"a": True})
+    with pytest.raises(ValueError, match="storage_options"):
+        # separate test for parquet, which has a different code path
+        read_parquet("localfile", storage_options={"a": True})
+    by = io.BytesIO()
+
+    with pytest.raises(ValueError, match="storage_options"):
+        read_csv(by, storage_options={"a": True})
+
+    df = DataFrame({"a": [0]})
+    with pytest.raises(ValueError, match="storage_options"):
+        df.to_parquet("nonfsspecpath", storage_options={"a": True})
diff --git a/pandas/pandas/tests/io/test_gcs.py b/pandas/pandas/tests/io/test_gcs.py
new file mode 100644
index 0000000000000..f68ef5fa2e0e5
--- /dev/null
+++ b/pandas/pandas/tests/io/test_gcs.py
@@ -0,0 +1,239 @@
+from io import BytesIO
+import os
+import pathlib
+import tarfile
+import zipfile
+
+import numpy as np
+import pytest
+
+from pandas.compat.pyarrow import pa_version_under17p0
+
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+    read_csv,
+    read_excel,
+    read_json,
+    read_parquet,
+)
+import pandas._testing as tm
+from pandas.util import _test_decorators as td
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture
+def gcs_buffer():
+    """Emulate GCS using a binary buffer."""
+    pytest.importorskip("gcsfs")
+    fsspec = pytest.importorskip("fsspec")
+
+    gcs_buffer = BytesIO()
+    gcs_buffer.close = lambda: True
+
+    class MockGCSFileSystem(fsspec.AbstractFileSystem):
+        @staticmethod
+        def open(*args, **kwargs):
+            gcs_buffer.seek(0)
+            return gcs_buffer
+
+        def ls(self, path, **kwargs):
+            # needed for pyarrow
+            return [{"name": path, "type": "file"}]
+
+    # Overwrites the default implementation from gcsfs to our mock class
+    fsspec.register_implementation("gs", MockGCSFileSystem, clobber=True)
+
+    return gcs_buffer
+
+
+# Patches pyarrow; other processes should not pick up change
+@pytest.mark.single_cpu
+@pytest.mark.parametrize("format", ["csv", "json", "parquet", "excel", "markdown"])
+def test_to_read_gcs(gcs_buffer, format, monkeypatch, capsys, request):
+    """
+    Test that many to/read functions support GCS.
+
+    GH 33987
+    """
+
+    df1 = DataFrame(
+        {
+            "int": [1, 3],
+            "float": [2.0, np.nan],
+            "str": ["t", "s"],
+            "dt": date_range("2018-06-18", periods=2),
+        }
+    )
+
+    path = f"gs://test/test.{format}"
+
+    if format == "csv":
+        df1.to_csv(path, index=True)
+        df2 = read_csv(path, parse_dates=["dt"], index_col=0)
+    elif format == "excel":
+        path = "gs://test/test.xlsx"
+        df1.to_excel(path)
+        df2 = read_excel(path, parse_dates=["dt"], index_col=0)
+    elif format == "json":
+        msg = (
+            "The default 'epoch' date format is deprecated and will be removed "
+            "in a future version, please use 'iso' date format instead."
+        )
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            df1.to_json(path)
+        df2 = read_json(path, convert_dates=["dt"])
+    elif format == "parquet":
+        pytest.importorskip("pyarrow")
+        pa_fs = pytest.importorskip("pyarrow.fs")
+
+        class MockFileSystem(pa_fs.FileSystem):
+            @staticmethod
+            def from_uri(path):
+                print("Using pyarrow filesystem")
+                to_local = pathlib.Path(path.replace("gs://", "")).absolute().as_uri()
+                return pa_fs.LocalFileSystem(to_local)
+
+        request.applymarker(
+            pytest.mark.xfail(
+                not pa_version_under17p0,
+                raises=TypeError,
+                reason="pyarrow 17 broke the mocked filesystem",
+            )
+        )
+        with monkeypatch.context() as m:
+            m.setattr(pa_fs, "FileSystem", MockFileSystem)
+            df1.to_parquet(path)
+            df2 = read_parquet(path)
+        captured = capsys.readouterr()
+        assert captured.out == "Using pyarrow filesystem\nUsing pyarrow filesystem\n"
+    elif format == "markdown":
+        pytest.importorskip("tabulate")
+        df1.to_markdown(path)
+        df2 = df1
+
+    expected = df1[:]
+    if format in ["csv", "excel"]:
+        expected["dt"] = expected["dt"].dt.as_unit("s")
+
+    tm.assert_frame_equal(df2, expected)
+
+
+def assert_equal_zip_safe(result: bytes, expected: bytes, compression: str):
+    """
+    For zip compression, only compare the CRC-32 checksum of the file contents
+    to avoid checking the time-dependent last-modified timestamp which
+    in some CI builds is off-by-one
+
+    See https://en.wikipedia.org/wiki/ZIP_(file_format)#File_headers
+    """
+    if compression == "zip":
+        # Only compare the CRC checksum of the file contents
+        with (
+            zipfile.ZipFile(BytesIO(result)) as exp,
+            zipfile.ZipFile(BytesIO(expected)) as res,
+        ):
+            for res_info, exp_info in zip(res.infolist(), exp.infolist()):
+                assert res_info.CRC == exp_info.CRC
+    elif compression == "tar":
+        with (
+            tarfile.open(fileobj=BytesIO(result)) as tar_exp,
+            tarfile.open(fileobj=BytesIO(expected)) as tar_res,
+        ):
+            for tar_res_info, tar_exp_info in zip(
+                tar_res.getmembers(), tar_exp.getmembers()
+            ):
+                actual_file = tar_res.extractfile(tar_res_info)
+                expected_file = tar_exp.extractfile(tar_exp_info)
+                assert (actual_file is None) == (expected_file is None)
+                if actual_file is not None and expected_file is not None:
+                    assert actual_file.read() == expected_file.read()
+    else:
+        assert result == expected
+
+
+@pytest.mark.parametrize("encoding", ["utf-8", "cp1251"])
+def test_to_csv_compression_encoding_gcs(
+    gcs_buffer, compression_only, encoding, compression_to_extension
+):
+    """
+    Compression and encoding should with GCS.
+
+    GH 35677 (to_csv, compression), GH 26124 (to_csv, encoding), and
+    GH 32392 (read_csv, encoding)
+    """
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD")),
+        index=Index([f"i-{i}" for i in range(30)]),
+    )
+
+    # reference of compressed and encoded file
+    compression = {"method": compression_only}
+    if compression_only == "gzip":
+        compression["mtime"] = 1  # be reproducible
+    buffer = BytesIO()
+    df.to_csv(buffer, compression=compression, encoding=encoding, mode="wb")
+
+    # write compressed file with explicit compression
+    path_gcs = "gs://test/test.csv"
+    df.to_csv(path_gcs, compression=compression, encoding=encoding)
+    res = gcs_buffer.getvalue()
+    expected = buffer.getvalue()
+    assert_equal_zip_safe(res, expected, compression_only)
+
+    read_df = read_csv(
+        path_gcs, index_col=0, compression=compression_only, encoding=encoding
+    )
+    tm.assert_frame_equal(df, read_df)
+
+    # write compressed file with implicit compression
+    file_ext = compression_to_extension[compression_only]
+    compression["method"] = "infer"
+    path_gcs += f".{file_ext}"
+    df.to_csv(path_gcs, compression=compression, encoding=encoding)
+
+    res = gcs_buffer.getvalue()
+    expected = buffer.getvalue()
+    assert_equal_zip_safe(res, expected, compression_only)
+
+    read_df = read_csv(path_gcs, index_col=0, compression="infer", encoding=encoding)
+    tm.assert_frame_equal(df, read_df)
+
+
+def test_to_parquet_gcs_new_file(monkeypatch, tmpdir):
+    """Regression test for writing to a not-yet-existent GCS Parquet file."""
+    pytest.importorskip("fastparquet")
+    pytest.importorskip("gcsfs")
+
+    from fsspec import AbstractFileSystem
+
+    df1 = DataFrame(
+        {
+            "int": [1, 3],
+            "float": [2.0, np.nan],
+            "str": ["t", "s"],
+            "dt": date_range("2018-06-18", periods=2),
+        }
+    )
+
+    class MockGCSFileSystem(AbstractFileSystem):
+        def open(self, path, mode="r", *args):
+            if "w" not in mode:
+                raise FileNotFoundError
+            return open(os.path.join(tmpdir, "test.parquet"), mode, encoding="utf-8")
+
+    monkeypatch.setattr("gcsfs.GCSFileSystem", MockGCSFileSystem)
+    df1.to_parquet(
+        "gs://test/test.csv", index=True, engine="fastparquet", compression=None
+    )
+
+
+@td.skip_if_installed("gcsfs")
+def test_gcs_not_present_exception():
+    with tm.external_error_raised(ImportError):
+        read_csv("gs://test/test.csv")
diff --git a/pandas/pandas/tests/io/test_html.py b/pandas/pandas/tests/io/test_html.py
new file mode 100644
index 0000000000000..bef28c4f027da
--- /dev/null
+++ b/pandas/pandas/tests/io/test_html.py
@@ -0,0 +1,1659 @@
+from collections.abc import Iterator
+from functools import partial
+from io import (
+    BytesIO,
+    StringIO,
+)
+import os
+from pathlib import Path
+import re
+import threading
+from urllib.error import URLError
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_windows
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    NA,
+    DataFrame,
+    MultiIndex,
+    Series,
+    Timestamp,
+    date_range,
+    read_csv,
+    read_html,
+    to_datetime,
+)
+import pandas._testing as tm
+
+from pandas.io.common import file_path_to_url
+
+
+@pytest.fixture(
+    params=[
+        "chinese_utf-16.html",
+        "chinese_utf-32.html",
+        "chinese_utf-8.html",
+        "letz_latin1.html",
+    ]
+)
+def html_encoding_file(request, datapath):
+    """Parametrized fixture for HTML encoding test filenames."""
+    return datapath("io", "data", "html_encoding", request.param)
+
+
+def assert_framelist_equal(list1, list2, *args, **kwargs):
+    assert len(list1) == len(list2), (
+        "lists are not of equal size "
+        f"len(list1) == {len(list1)}, "
+        f"len(list2) == {len(list2)}"
+    )
+    msg = "not all list elements are DataFrames"
+    both_frames = all(
+        map(
+            lambda x, y: isinstance(x, DataFrame) and isinstance(y, DataFrame),
+            list1,
+            list2,
+        )
+    )
+    assert both_frames, msg
+    for frame_i, frame_j in zip(list1, list2):
+        tm.assert_frame_equal(frame_i, frame_j, *args, **kwargs)
+        assert not frame_i.empty, "frames are both empty"
+
+
+def test_bs4_version_fails(monkeypatch, datapath):
+    bs4 = pytest.importorskip("bs4")
+    pytest.importorskip("html5lib")
+
+    monkeypatch.setattr(bs4, "__version__", "4.2")
+    with pytest.raises(ImportError, match="Pandas requires version"):
+        read_html(datapath("io", "data", "html", "spam.html"), flavor="bs4")
+
+
+def test_invalid_flavor():
+    url = "google.com"
+    flavor = "invalid flavor"
+    msg = r"\{" + flavor + r"\} is not a valid set of flavors"
+
+    with pytest.raises(ValueError, match=msg):
+        read_html(StringIO(url), match="google", flavor=flavor)
+
+
+def test_same_ordering(datapath):
+    pytest.importorskip("bs4")
+    pytest.importorskip("lxml")
+    pytest.importorskip("html5lib")
+
+    filename = datapath("io", "data", "html", "valid_markup.html")
+    dfs_lxml = read_html(filename, index_col=0, flavor=["lxml"])
+    dfs_bs4 = read_html(filename, index_col=0, flavor=["bs4"])
+    assert_framelist_equal(dfs_lxml, dfs_bs4)
+
+
+@pytest.fixture(
+    params=[
+        pytest.param("bs4", marks=[td.skip_if_no("bs4"), td.skip_if_no("html5lib")]),
+        pytest.param("lxml", marks=td.skip_if_no("lxml")),
+    ],
+)
+def flavor_read_html(request):
+    return partial(read_html, flavor=request.param)
+
+
+class TestReadHtml:
+    def test_literal_html_deprecation(self, flavor_read_html):
+        # GH 53785
+        msg = r"\[Errno 2\] No such file or director"
+
+        with pytest.raises(FileNotFoundError, match=msg):
+            flavor_read_html(
+                """<table>
+                <thead>
+                    <tr>
+                        <th>A</th>
+                        <th>B</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    <tr>
+                        <td>1</td>
+                        <td>2</td>
+                    </tr>
+                </tbody>
+                <tbody>
+                    <tr>
+                        <td>3</td>
+                        <td>4</td>
+                    </tr>
+                </tbody>
+            </table>"""
+            )
+
+    @pytest.fixture
+    def spam_data(self, datapath):
+        return datapath("io", "data", "html", "spam.html")
+
+    @pytest.fixture
+    def banklist_data(self, datapath):
+        return datapath("io", "data", "html", "banklist.html")
+
+    def test_to_html_compat(self, flavor_read_html):
+        df = (
+            DataFrame(
+                np.random.default_rng(2).random((4, 3)),
+                columns=pd.Index(list("abc")),
+            )
+            .map("{:.3f}".format)
+            .astype(float)
+        )
+        out = df.to_html()
+        res = flavor_read_html(
+            StringIO(out), attrs={"class": "dataframe"}, index_col=0
+        )[0]
+        tm.assert_frame_equal(res, df)
+
+    def test_dtype_backend(self, string_storage, dtype_backend, flavor_read_html):
+        # GH#50286
+        df = DataFrame(
+            {
+                "a": Series([1, np.nan, 3], dtype="Int64"),
+                "b": Series([1, 2, 3], dtype="Int64"),
+                "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": [True, False, None],
+                "f": [True, False, True],
+                "g": ["a", "b", "c"],
+                "h": ["a", "b", None],
+            }
+        )
+
+        out = df.to_html(index=False)
+        with pd.option_context("mode.string_storage", string_storage):
+            result = flavor_read_html(StringIO(out), dtype_backend=dtype_backend)[0]
+
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            string_dtype = pd.ArrowDtype(pa.string())
+        else:
+            string_dtype = pd.StringDtype(string_storage)
+
+        expected = DataFrame(
+            {
+                "a": Series([1, np.nan, 3], dtype="Int64"),
+                "b": Series([1, 2, 3], dtype="Int64"),
+                "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": Series([True, False, NA], dtype="boolean"),
+                "f": Series([True, False, True], dtype="boolean"),
+                "g": Series(["a", "b", "c"], dtype=string_dtype),
+                "h": Series(["a", "b", None], dtype=string_dtype),
+            }
+        )
+
+        if dtype_backend == "pyarrow":
+            import pyarrow as pa
+
+            from pandas.arrays import ArrowExtensionArray
+
+            expected = DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                    for col in expected.columns
+                }
+            )
+
+        # the storage of the str columns' Index is also affected by the
+        # string_storage setting -> ignore that for checking the result
+        tm.assert_frame_equal(result, expected, check_column_type=False)
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_banklist_url(self, httpserver, banklist_data, flavor_read_html):
+        with open(banklist_data, encoding="utf-8") as f:
+            httpserver.serve_content(content=f.read())
+            df1 = flavor_read_html(
+                # lxml cannot find attrs leave out for now
+                httpserver.url,
+                match="First Federal Bank of Florida",  # attrs={"class": "dataTable"}
+            )
+            # lxml cannot find attrs leave out for now
+            df2 = flavor_read_html(
+                httpserver.url,
+                match="Metcalf Bank",
+            )  # attrs={"class": "dataTable"})
+
+        assert_framelist_equal(df1, df2)
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_spam_url(self, httpserver, spam_data, flavor_read_html):
+        with open(spam_data, encoding="utf-8") as f:
+            httpserver.serve_content(content=f.read())
+            df1 = flavor_read_html(httpserver.url, match=".*Water.*")
+            df2 = flavor_read_html(httpserver.url, match="Unit")
+
+        assert_framelist_equal(df1, df2)
+
+    @pytest.mark.slow
+    def test_banklist(self, banklist_data, flavor_read_html):
+        df1 = flavor_read_html(
+            banklist_data, match=".*Florida.*", attrs={"id": "table"}
+        )
+        df2 = flavor_read_html(
+            banklist_data, match="Metcalf Bank", attrs={"id": "table"}
+        )
+
+        assert_framelist_equal(df1, df2)
+
+    def test_spam(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*")
+        df2 = flavor_read_html(spam_data, match="Unit")
+        assert_framelist_equal(df1, df2)
+
+        assert df1[0].iloc[0, 0] == "Proximates"
+        assert df1[0].columns[0] == "Nutrient"
+
+    def test_spam_no_match(self, spam_data, flavor_read_html):
+        dfs = flavor_read_html(spam_data)
+        for df in dfs:
+            assert isinstance(df, DataFrame)
+
+    def test_banklist_no_match(self, banklist_data, flavor_read_html):
+        dfs = flavor_read_html(banklist_data, attrs={"id": "table"})
+        for df in dfs:
+            assert isinstance(df, DataFrame)
+
+    def test_spam_header(self, spam_data, flavor_read_html):
+        df = flavor_read_html(spam_data, match=".*Water.*", header=2)[0]
+        assert df.columns[0] == "Proximates"
+        assert not df.empty
+
+    def test_skiprows_int(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=1)
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=1)
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_range(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=range(2))
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=range(2))
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_list(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=[1, 2])
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=[2, 1])
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_set(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows={1, 2})
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows={2, 1})
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_slice(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=1)
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=1)
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_slice_short(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=slice(2))
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=slice(2))
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_slice_long(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=slice(2, 5))
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=slice(4, 1, -1))
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_ndarray(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", skiprows=np.arange(2))
+        df2 = flavor_read_html(spam_data, match="Unit", skiprows=np.arange(2))
+
+        assert_framelist_equal(df1, df2)
+
+    def test_skiprows_invalid(self, spam_data, flavor_read_html):
+        with pytest.raises(TypeError, match=("is not a valid type for skipping rows")):
+            flavor_read_html(spam_data, match=".*Water.*", skiprows="asdf")
+
+    def test_index(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", index_col=0)
+        df2 = flavor_read_html(spam_data, match="Unit", index_col=0)
+        assert_framelist_equal(df1, df2)
+
+    def test_header_and_index_no_types(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", header=1, index_col=0)
+        df2 = flavor_read_html(spam_data, match="Unit", header=1, index_col=0)
+        assert_framelist_equal(df1, df2)
+
+    def test_header_and_index_with_types(self, spam_data, flavor_read_html):
+        df1 = flavor_read_html(spam_data, match=".*Water.*", header=1, index_col=0)
+        df2 = flavor_read_html(spam_data, match="Unit", header=1, index_col=0)
+        assert_framelist_equal(df1, df2)
+
+    def test_infer_types(self, spam_data, flavor_read_html):
+        # 10892 infer_types removed
+        df1 = flavor_read_html(spam_data, match=".*Water.*", index_col=0)
+        df2 = flavor_read_html(spam_data, match="Unit", index_col=0)
+        assert_framelist_equal(df1, df2)
+
+    def test_string_io(self, spam_data, flavor_read_html):
+        with open(spam_data, encoding="UTF-8") as f:
+            data1 = StringIO(f.read())
+
+        with open(spam_data, encoding="UTF-8") as f:
+            data2 = StringIO(f.read())
+
+        df1 = flavor_read_html(data1, match=".*Water.*")
+        df2 = flavor_read_html(data2, match="Unit")
+        assert_framelist_equal(df1, df2)
+
+    def test_string(self, spam_data, flavor_read_html):
+        with open(spam_data, encoding="UTF-8") as f:
+            data = f.read()
+
+        df1 = flavor_read_html(StringIO(data), match=".*Water.*")
+        df2 = flavor_read_html(StringIO(data), match="Unit")
+
+        assert_framelist_equal(df1, df2)
+
+    def test_file_like(self, spam_data, flavor_read_html):
+        with open(spam_data, encoding="UTF-8") as f:
+            df1 = flavor_read_html(f, match=".*Water.*")
+
+        with open(spam_data, encoding="UTF-8") as f:
+            df2 = flavor_read_html(f, match="Unit")
+
+        assert_framelist_equal(df1, df2)
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_bad_url_protocol(self, httpserver, flavor_read_html):
+        httpserver.serve_content("urlopen error unknown url type: git", code=404)
+        with pytest.raises(URLError, match="urlopen error unknown url type: git"):
+            flavor_read_html("git://github.com", match=".*Water.*")
+
+    @pytest.mark.slow
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_invalid_url(self, httpserver, flavor_read_html):
+        httpserver.serve_content("Name or service not known", code=404)
+        with pytest.raises((URLError, ValueError), match="HTTP Error 404: NOT FOUND"):
+            flavor_read_html(httpserver.url, match=".*Water.*")
+
+    @pytest.mark.slow
+    def test_file_url(self, banklist_data, flavor_read_html):
+        url = banklist_data
+        dfs = flavor_read_html(
+            file_path_to_url(os.path.abspath(url)), match="First", attrs={"id": "table"}
+        )
+        assert isinstance(dfs, list)
+        for df in dfs:
+            assert isinstance(df, DataFrame)
+
+    @pytest.mark.slow
+    def test_invalid_table_attrs(self, banklist_data, flavor_read_html):
+        url = banklist_data
+        with pytest.raises(ValueError, match="No tables found"):
+            flavor_read_html(
+                url, match="First Federal Bank of Florida", attrs={"id": "tasdfable"}
+            )
+
+    @pytest.mark.slow
+    def test_multiindex_header(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data, match="Metcalf", attrs={"id": "table"}, header=[0, 1]
+        )[0]
+        assert isinstance(df.columns, MultiIndex)
+
+    @pytest.mark.slow
+    def test_multiindex_index(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data, match="Metcalf", attrs={"id": "table"}, index_col=[0, 1]
+        )[0]
+        assert isinstance(df.index, MultiIndex)
+
+    @pytest.mark.slow
+    def test_multiindex_header_index(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data,
+            match="Metcalf",
+            attrs={"id": "table"},
+            header=[0, 1],
+            index_col=[0, 1],
+        )[0]
+        assert isinstance(df.columns, MultiIndex)
+        assert isinstance(df.index, MultiIndex)
+
+    @pytest.mark.slow
+    def test_multiindex_header_skiprows_tuples(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data,
+            match="Metcalf",
+            attrs={"id": "table"},
+            header=[0, 1],
+            skiprows=1,
+        )[0]
+        assert isinstance(df.columns, MultiIndex)
+
+    @pytest.mark.slow
+    def test_multiindex_header_skiprows(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data,
+            match="Metcalf",
+            attrs={"id": "table"},
+            header=[0, 1],
+            skiprows=1,
+        )[0]
+        assert isinstance(df.columns, MultiIndex)
+
+    @pytest.mark.slow
+    def test_multiindex_header_index_skiprows(self, banklist_data, flavor_read_html):
+        df = flavor_read_html(
+            banklist_data,
+            match="Metcalf",
+            attrs={"id": "table"},
+            header=[0, 1],
+            index_col=[0, 1],
+            skiprows=1,
+        )[0]
+        assert isinstance(df.index, MultiIndex)
+        assert isinstance(df.columns, MultiIndex)
+
+    @pytest.mark.slow
+    def test_regex_idempotency(self, banklist_data, flavor_read_html):
+        url = banklist_data
+        dfs = flavor_read_html(
+            file_path_to_url(os.path.abspath(url)),
+            match=re.compile(re.compile("Florida")),
+            attrs={"id": "table"},
+        )
+        assert isinstance(dfs, list)
+        for df in dfs:
+            assert isinstance(df, DataFrame)
+
+    def test_negative_skiprows(self, spam_data, flavor_read_html):
+        msg = r"\(you passed a negative value\)"
+        with pytest.raises(ValueError, match=msg):
+            flavor_read_html(spam_data, match="Water", skiprows=-1)
+
+    @pytest.fixture
+    def python_docs(self):
+        return """
+          <table class="contentstable" align="center"><tr>
+            <td width="50%">
+            <p class="biglink"><a class="biglink" href="whatsnew/2.7.html">What's new in Python 2.7?</a><br/>
+                <span class="linkdescr">or <a href="whatsnew/index.html">all "What's new" documents</a> since 2.0</span></p>
+            <p class="biglink"><a class="biglink" href="tutorial/index.html">Tutorial</a><br/>
+                <span class="linkdescr">start here</span></p>
+            <p class="biglink"><a class="biglink" href="library/index.html">Library Reference</a><br/>
+                <span class="linkdescr">keep this under your pillow</span></p>
+            <p class="biglink"><a class="biglink" href="reference/index.html">Language Reference</a><br/>
+                <span class="linkdescr">describes syntax and language elements</span></p>
+            <p class="biglink"><a class="biglink" href="using/index.html">Python Setup and Usage</a><br/>
+                <span class="linkdescr">how to use Python on different platforms</span></p>
+            <p class="biglink"><a class="biglink" href="howto/index.html">Python HOWTOs</a><br/>
+                <span class="linkdescr">in-depth documents on specific topics</span></p>
+            </td><td width="50%">
+            <p class="biglink"><a class="biglink" href="installing/index.html">Installing Python Modules</a><br/>
+                <span class="linkdescr">installing from the Python Package Index &amp; other sources</span></p>
+            <p class="biglink"><a class="biglink" href="distributing/index.html">Distributing Python Modules</a><br/>
+                <span class="linkdescr">publishing modules for installation by others</span></p>
+            <p class="biglink"><a class="biglink" href="extending/index.html">Extending and Embedding</a><br/>
+                <span class="linkdescr">tutorial for C/C++ programmers</span></p>
+            <p class="biglink"><a class="biglink" href="c-api/index.html">Python/C API</a><br/>
+                <span class="linkdescr">reference for C/C++ programmers</span></p>
+            <p class="biglink"><a class="biglink" href="faq/index.html">FAQs</a><br/>
+                <span class="linkdescr">frequently asked questions (with answers!)</span></p>
+            </td></tr>
+        </table>
+
+        <p><strong>Indices and tables:</strong></p>
+        <table class="contentstable" align="center"><tr>
+            <td width="50%">
+            <p class="biglink"><a class="biglink" href="py-modindex.html">Python Global Module Index</a><br/>
+                <span class="linkdescr">quick access to all modules</span></p>
+            <p class="biglink"><a class="biglink" href="genindex.html">General Index</a><br/>
+                <span class="linkdescr">all functions, classes, terms</span></p>
+            <p class="biglink"><a class="biglink" href="glossary.html">Glossary</a><br/>
+                <span class="linkdescr">the most important terms explained</span></p>
+            </td><td width="50%">
+            <p class="biglink"><a class="biglink" href="search.html">Search page</a><br/>
+                <span class="linkdescr">search this documentation</span></p>
+            <p class="biglink"><a class="biglink" href="contents.html">Complete Table of Contents</a><br/>
+                <span class="linkdescr">lists all sections and subsections</span></p>
+            </td></tr>
+        </table>
+        """  # noqa: E501
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_multiple_matches(self, python_docs, httpserver, flavor_read_html):
+        httpserver.serve_content(content=python_docs)
+        dfs = flavor_read_html(httpserver.url, match="Python")
+        assert len(dfs) > 1
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_python_docs_table(self, python_docs, httpserver, flavor_read_html):
+        httpserver.serve_content(content=python_docs)
+        dfs = flavor_read_html(httpserver.url, match="Python")
+        zz = [df.iloc[0, 0][0:4] for df in dfs]
+        assert sorted(zz) == ["Pyth", "What"]
+
+    def test_empty_tables(self, flavor_read_html):
+        """
+        Make sure that read_html ignores empty tables.
+        """
+        html = """
+            <table>
+                <thead>
+                    <tr>
+                        <th>A</th>
+                        <th>B</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    <tr>
+                        <td>1</td>
+                        <td>2</td>
+                    </tr>
+                </tbody>
+            </table>
+            <table>
+                <tbody>
+                </tbody>
+            </table>
+        """
+        result = flavor_read_html(StringIO(html))
+        assert len(result) == 1
+
+    def test_multiple_tbody(self, flavor_read_html):
+        # GH-20690
+        # Read all tbody tags within a single table.
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+            <thead>
+                <tr>
+                    <th>A</th>
+                    <th>B</th>
+                </tr>
+            </thead>
+            <tbody>
+                <tr>
+                    <td>1</td>
+                    <td>2</td>
+                </tr>
+            </tbody>
+            <tbody>
+                <tr>
+                    <td>3</td>
+                    <td>4</td>
+                </tr>
+            </tbody>
+        </table>"""
+            )
+        )[0]
+
+        expected = DataFrame(data=[[1, 2], [3, 4]], columns=["A", "B"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_header_and_one_column(self, flavor_read_html):
+        """
+        Don't fail with bs4 when there is a header and only one column
+        as described in issue #9178
+        """
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+                <thead>
+                    <tr>
+                        <th>Header</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    <tr>
+                        <td>first</td>
+                    </tr>
+                </tbody>
+            </table>"""
+            )
+        )[0]
+
+        expected = DataFrame(data={"Header": "first"}, index=[0])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_thead_without_tr(self, flavor_read_html):
+        """
+        Ensure parser adds <tr> within <thead> on malformed HTML.
+        """
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+            <thead>
+                <tr>
+                    <th>Country</th>
+                    <th>Municipality</th>
+                    <th>Year</th>
+                </tr>
+            </thead>
+            <tbody>
+                <tr>
+                    <td>Ukraine</td>
+                    <th>Odessa</th>
+                    <td>1944</td>
+                </tr>
+            </tbody>
+        </table>"""
+            )
+        )[0]
+
+        expected = DataFrame(
+            data=[["Ukraine", "Odessa", 1944]],
+            columns=["Country", "Municipality", "Year"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_tfoot_read(self, flavor_read_html):
+        """
+        Make sure that read_html reads tfoot, containing td or th.
+        Ignores empty tfoot
+        """
+        data_template = """<table>
+            <thead>
+                <tr>
+                    <th>A</th>
+                    <th>B</th>
+                </tr>
+            </thead>
+            <tbody>
+                <tr>
+                    <td>bodyA</td>
+                    <td>bodyB</td>
+                </tr>
+            </tbody>
+            <tfoot>
+                {footer}
+            </tfoot>
+        </table>"""
+
+        expected1 = DataFrame(data=[["bodyA", "bodyB"]], columns=["A", "B"])
+
+        expected2 = DataFrame(
+            data=[["bodyA", "bodyB"], ["footA", "footB"]], columns=["A", "B"]
+        )
+
+        data1 = data_template.format(footer="")
+        data2 = data_template.format(footer="<tr><td>footA</td><th>footB</th></tr>")
+
+        result1 = flavor_read_html(StringIO(data1))[0]
+        result2 = flavor_read_html(StringIO(data2))[0]
+
+        tm.assert_frame_equal(result1, expected1)
+        tm.assert_frame_equal(result2, expected2)
+
+    def test_parse_header_of_non_string_column(self, flavor_read_html):
+        # GH5048: if header is specified explicitly, an int column should be
+        # parsed as int while its header is parsed as str
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <td>S</td>
+                    <td>I</td>
+                </tr>
+                <tr>
+                    <td>text</td>
+                    <td>1944</td>
+                </tr>
+            </table>
+        """
+            ),
+            header=0,
+        )[0]
+
+        expected = DataFrame([["text", 1944]], columns=("S", "I"))
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.slow
+    def test_banklist_header(self, banklist_data, datapath, flavor_read_html):
+        from pandas.io.html import _remove_whitespace
+
+        def try_remove_ws(x):
+            try:
+                return _remove_whitespace(x)
+            except AttributeError:
+                return x
+
+        df = flavor_read_html(banklist_data, match="Metcalf", attrs={"id": "table"})[0]
+        ground_truth = read_csv(
+            datapath("io", "data", "csv", "banklist.csv"),
+            converters={"Updated Date": Timestamp, "Closing Date": Timestamp},
+        )
+        assert df.shape == ground_truth.shape
+        old = [
+            "First Vietnamese American Bank In Vietnamese",
+            "Westernbank Puerto Rico En Espanol",
+            "R-G Premier Bank of Puerto Rico En Espanol",
+            "Eurobank En Espanol",
+            "Sanderson State Bank En Espanol",
+            "Washington Mutual Bank (Including its subsidiary Washington "
+            "Mutual Bank FSB)",
+            "Silver State Bank En Espanol",
+            "AmTrade International Bank En Espanol",
+            "Hamilton Bank, NA En Espanol",
+            "The Citizens Savings Bank Pioneer Community Bank, Inc.",
+        ]
+        new = [
+            "First Vietnamese American Bank",
+            "Westernbank Puerto Rico",
+            "R-G Premier Bank of Puerto Rico",
+            "Eurobank",
+            "Sanderson State Bank",
+            "Washington Mutual Bank",
+            "Silver State Bank",
+            "AmTrade International Bank",
+            "Hamilton Bank, NA",
+            "The Citizens Savings Bank",
+        ]
+        dfnew = df.map(try_remove_ws).replace(old, new)
+        gtnew = ground_truth.map(try_remove_ws)
+        converted = dfnew
+        date_cols = ["Closing Date", "Updated Date"]
+        converted[date_cols] = converted[date_cols].apply(to_datetime)
+        tm.assert_frame_equal(converted, gtnew)
+
+    @pytest.mark.slow
+    def test_gold_canyon(self, banklist_data, flavor_read_html):
+        gc = "Gold Canyon"
+        with open(banklist_data, encoding="utf-8") as f:
+            raw_text = f.read()
+
+        assert gc in raw_text
+        df = flavor_read_html(
+            banklist_data, match="Gold Canyon", attrs={"id": "table"}
+        )[0]
+        assert gc in df.to_string()
+
+    def test_different_number_of_cols(self, flavor_read_html):
+        expected = flavor_read_html(
+            StringIO(
+                """<table>
+                        <thead>
+                            <tr style="text-align: right;">
+                            <th></th>
+                            <th>C_l0_g0</th>
+                            <th>C_l0_g1</th>
+                            <th>C_l0_g2</th>
+                            <th>C_l0_g3</th>
+                            <th>C_l0_g4</th>
+                            </tr>
+                        </thead>
+                        <tbody>
+                            <tr>
+                            <th>R_l0_g0</th>
+                            <td> 0.763</td>
+                            <td> 0.233</td>
+                            <td> nan</td>
+                            <td> nan</td>
+                            <td> nan</td>
+                            </tr>
+                            <tr>
+                            <th>R_l0_g1</th>
+                            <td> 0.244</td>
+                            <td> 0.285</td>
+                            <td> 0.392</td>
+                            <td> 0.137</td>
+                            <td> 0.222</td>
+                            </tr>
+                        </tbody>
+                    </table>"""
+            ),
+            index_col=0,
+        )[0]
+
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+                    <thead>
+                        <tr style="text-align: right;">
+                        <th></th>
+                        <th>C_l0_g0</th>
+                        <th>C_l0_g1</th>
+                        <th>C_l0_g2</th>
+                        <th>C_l0_g3</th>
+                        <th>C_l0_g4</th>
+                        </tr>
+                    </thead>
+                    <tbody>
+                        <tr>
+                        <th>R_l0_g0</th>
+                        <td> 0.763</td>
+                        <td> 0.233</td>
+                        </tr>
+                        <tr>
+                        <th>R_l0_g1</th>
+                        <td> 0.244</td>
+                        <td> 0.285</td>
+                        <td> 0.392</td>
+                        <td> 0.137</td>
+                        <td> 0.222</td>
+                        </tr>
+                    </tbody>
+                 </table>"""
+            ),
+            index_col=0,
+        )[0]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_colspan_rowspan_1(self, flavor_read_html):
+        # GH17054
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <th>A</th>
+                    <th colspan="1">B</th>
+                    <th rowspan="1">C</th>
+                </tr>
+                <tr>
+                    <td>a</td>
+                    <td>b</td>
+                    <td>c</td>
+                </tr>
+            </table>
+        """
+            )
+        )[0]
+
+        expected = DataFrame([["a", "b", "c"]], columns=["A", "B", "C"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_colspan_rowspan_copy_values(self, flavor_read_html):
+        # GH17054
+
+        # In ASCII, with lowercase letters being copies:
+        #
+        # X x Y Z W
+        # A B b z C
+
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <td colspan="2">X</td>
+                    <td>Y</td>
+                    <td rowspan="2">Z</td>
+                    <td>W</td>
+                </tr>
+                <tr>
+                    <td>A</td>
+                    <td colspan="2">B</td>
+                    <td>C</td>
+                </tr>
+            </table>
+        """
+            ),
+            header=0,
+        )[0]
+
+        expected = DataFrame(
+            data=[["A", "B", "B", "Z", "C"]], columns=["X", "X.1", "Y", "Z", "W"]
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_colspan_rowspan_both_not_1(self, flavor_read_html):
+        # GH17054
+
+        # In ASCII, with lowercase letters being copies:
+        #
+        # A B b b C
+        # a b b b D
+
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <td rowspan="2">A</td>
+                    <td rowspan="2" colspan="3">B</td>
+                    <td>C</td>
+                </tr>
+                <tr>
+                    <td>D</td>
+                </tr>
+            </table>
+        """
+            ),
+            header=0,
+        )[0]
+
+        expected = DataFrame(
+            data=[["A", "B", "B", "B", "D"]], columns=["A", "B", "B.1", "B.2", "C"]
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_rowspan_at_end_of_row(self, flavor_read_html):
+        # GH17054
+
+        # In ASCII, with lowercase letters being copies:
+        #
+        # A B
+        # C b
+
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <td>A</td>
+                    <td rowspan="2">B</td>
+                </tr>
+                <tr>
+                    <td>C</td>
+                </tr>
+            </table>
+        """
+            ),
+            header=0,
+        )[0]
+
+        expected = DataFrame(data=[["C", "B"]], columns=["A", "B"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_rowspan_only_rows(self, flavor_read_html):
+        # GH17054
+
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <td rowspan="3">A</td>
+                    <td rowspan="3">B</td>
+                </tr>
+            </table>
+        """
+            ),
+            header=0,
+        )[0]
+
+        expected = DataFrame(data=[["A", "B"], ["A", "B"]], columns=["A", "B"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_rowspan_in_header_overflowing_to_body(self, flavor_read_html):
+        # GH60210
+
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <th rowspan="2">A</th>
+                    <th>B</th>
+                </tr>
+                <tr>
+                    <td>1</td>
+                </tr>
+                <tr>
+                    <td>C</td>
+                    <td>2</td>
+                </tr>
+            </table>
+        """
+            )
+        )[0]
+
+        expected = DataFrame(data=[["A", 1], ["C", 2]], columns=["A", "B"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_header_inferred_from_rows_with_only_th(self, flavor_read_html):
+        # GH17054
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <th>A</th>
+                    <th>B</th>
+                </tr>
+                <tr>
+                    <th>a</th>
+                    <th>b</th>
+                </tr>
+                <tr>
+                    <td>1</td>
+                    <td>2</td>
+                </tr>
+            </table>
+        """
+            )
+        )[0]
+
+        columns = MultiIndex(levels=[["A", "B"], ["a", "b"]], codes=[[0, 1], [0, 1]])
+        expected = DataFrame(data=[[1, 2]], columns=columns)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_parse_dates_list(self, flavor_read_html):
+        df = DataFrame({"date": date_range("1/1/2001", periods=10)})
+
+        expected = df[:]
+        expected["date"] = expected["date"].dt.as_unit("s")
+
+        str_df = df.to_html()
+        res = flavor_read_html(StringIO(str_df), parse_dates=[1], index_col=0)
+        tm.assert_frame_equal(expected, res[0])
+        res = flavor_read_html(StringIO(str_df), parse_dates=["date"], index_col=0)
+        tm.assert_frame_equal(expected, res[0])
+
+    def test_wikipedia_states_table(self, datapath, flavor_read_html):
+        data = datapath("io", "data", "html", "wikipedia_states.html")
+        assert os.path.isfile(data), f"{data!r} is not a file"
+        assert os.path.getsize(data), f"{data!r} is an empty file"
+        result = flavor_read_html(data, match="Arizona", header=1)[0]
+        assert result.shape == (60, 12)
+        assert "Unnamed" in result.columns[-1]
+        assert result["sq mi"].dtype == np.dtype("float64")
+        assert np.allclose(result.loc[0, "sq mi"], 665384.04)
+
+    def test_wikipedia_states_multiindex(self, datapath, flavor_read_html):
+        data = datapath("io", "data", "html", "wikipedia_states.html")
+        result = flavor_read_html(data, match="Arizona", index_col=0)[0]
+        assert result.shape == (60, 11)
+        assert "Unnamed" in result.columns[-1][1]
+        assert result.columns.nlevels == 2
+        assert np.allclose(result.loc["Alaska", ("Total area[2]", "sq mi")], 665384.04)
+
+    def test_parser_error_on_empty_header_row(self, flavor_read_html):
+        result = flavor_read_html(
+            StringIO(
+                """
+                <table>
+                    <thead>
+                        <tr><th></th><th></tr>
+                        <tr><th>A</th><th>B</th></tr>
+                    </thead>
+                    <tbody>
+                        <tr><td>a</td><td>b</td></tr>
+                    </tbody>
+                </table>
+            """
+            ),
+            header=[0, 1],
+        )
+        expected = DataFrame(
+            [["a", "b"]],
+            columns=MultiIndex.from_tuples(
+                [("Unnamed: 0_level_0", "A"), ("Unnamed: 1_level_0", "B")]
+            ),
+        )
+        tm.assert_frame_equal(result[0], expected)
+
+    def test_decimal_rows(self, flavor_read_html):
+        # GH 12907
+        result = flavor_read_html(
+            StringIO(
+                """<html>
+            <body>
+             <table>
+                <thead>
+                    <tr>
+                        <th>Header</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    <tr>
+                        <td>1100#101</td>
+                    </tr>
+                </tbody>
+            </table>
+            </body>
+        </html>"""
+            ),
+            decimal="#",
+        )[0]
+
+        expected = DataFrame(data={"Header": 1100.101}, index=[0])
+
+        assert result["Header"].dtype == np.dtype("float64")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("arg", [True, False])
+    def test_bool_header_arg(self, spam_data, arg, flavor_read_html):
+        # GH 6114
+        msg = re.escape(
+            "Passing a bool to header is invalid. Use header=None for no header or "
+            "header=int or list-like of ints to specify the row(s) making up the "
+            "column names"
+        )
+        with pytest.raises(TypeError, match=msg):
+            flavor_read_html(spam_data, header=arg)
+
+    def test_converters(self, flavor_read_html):
+        # GH 13461
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+                 <thead>
+                   <tr>
+                     <th>a</th>
+                    </tr>
+                 </thead>
+                 <tbody>
+                   <tr>
+                     <td> 0.763</td>
+                   </tr>
+                   <tr>
+                     <td> 0.244</td>
+                   </tr>
+                 </tbody>
+               </table>"""
+            ),
+            converters={"a": str},
+        )[0]
+
+        expected = DataFrame({"a": ["0.763", "0.244"]})
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_na_values(self, flavor_read_html):
+        # GH 13461
+        result = flavor_read_html(
+            StringIO(
+                """<table>
+                 <thead>
+                   <tr>
+                     <th>a</th>
+                   </tr>
+                 </thead>
+                 <tbody>
+                   <tr>
+                     <td> 0.763</td>
+                   </tr>
+                   <tr>
+                     <td> 0.244</td>
+                   </tr>
+                 </tbody>
+               </table>"""
+            ),
+            na_values=[0.244],
+        )[0]
+
+        expected = DataFrame({"a": [0.763, np.nan]})
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_keep_default_na(self, flavor_read_html):
+        html_data = """<table>
+                        <thead>
+                            <tr>
+                            <th>a</th>
+                            </tr>
+                        </thead>
+                        <tbody>
+                            <tr>
+                            <td> N/A</td>
+                            </tr>
+                            <tr>
+                            <td> NA</td>
+                            </tr>
+                        </tbody>
+                    </table>"""
+
+        expected_df = DataFrame({"a": ["N/A", "NA"]})
+        html_df = flavor_read_html(StringIO(html_data), keep_default_na=False)[0]
+        tm.assert_frame_equal(expected_df, html_df)
+
+        expected_df = DataFrame({"a": [np.nan, np.nan]})
+        html_df = flavor_read_html(StringIO(html_data), keep_default_na=True)[0]
+        tm.assert_frame_equal(expected_df, html_df)
+
+    def test_preserve_empty_rows(self, flavor_read_html):
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <th>A</th>
+                    <th>B</th>
+                </tr>
+                <tr>
+                    <td>a</td>
+                    <td>b</td>
+                </tr>
+                <tr>
+                    <td></td>
+                    <td></td>
+                </tr>
+            </table>
+        """
+            )
+        )[0]
+
+        expected = DataFrame(data=[["a", "b"], [np.nan, np.nan]], columns=["A", "B"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_ignore_empty_rows_when_inferring_header(self, flavor_read_html):
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <thead>
+                    <tr><th></th><th></tr>
+                    <tr><th>A</th><th>B</th></tr>
+                    <tr><th>a</th><th>b</th></tr>
+                </thead>
+                <tbody>
+                    <tr><td>1</td><td>2</td></tr>
+                </tbody>
+            </table>
+        """
+            )
+        )[0]
+
+        columns = MultiIndex(levels=[["A", "B"], ["a", "b"]], codes=[[0, 1], [0, 1]])
+        expected = DataFrame(data=[[1, 2]], columns=columns)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_multiple_header_rows(self, flavor_read_html):
+        # Issue #13434
+        expected_df = DataFrame(
+            data=[("Hillary", 68, "D"), ("Bernie", 74, "D"), ("Donald", 69, "R")]
+        )
+        expected_df.columns = [
+            ["Unnamed: 0_level_0", "Age", "Party"],
+            ["Name", "Unnamed: 1_level_1", "Unnamed: 2_level_1"],
+        ]
+        html = expected_df.to_html(index=False)
+        html_df = flavor_read_html(StringIO(html))[0]
+        tm.assert_frame_equal(expected_df, html_df)
+
+    def test_works_on_valid_markup(self, datapath, flavor_read_html):
+        filename = datapath("io", "data", "html", "valid_markup.html")
+        dfs = flavor_read_html(filename, index_col=0)
+        assert isinstance(dfs, list)
+        assert isinstance(dfs[0], DataFrame)
+
+    @pytest.mark.slow
+    def test_fallback_success(self, datapath, flavor_read_html):
+        banklist_data = datapath("io", "data", "html", "banklist.html")
+
+        flavor_read_html(banklist_data, match=".*Water.*", flavor=["lxml", "html5lib"])
+
+    def test_to_html_timestamp(self):
+        rng = date_range("2000-01-01", periods=10)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)), index=rng)
+
+        result = df.to_html()
+        assert "2000-01-01" in result
+
+    def test_to_html_borderless(self):
+        df = DataFrame([{"A": 1, "B": 2}])
+        out_border_default = df.to_html()
+        out_border_true = df.to_html(border=True)
+        out_border_explicit_default = df.to_html(border=1)
+        out_border_nondefault = df.to_html(border=2)
+        out_border_zero = df.to_html(border=0)
+
+        out_border_false = df.to_html(border=False)
+
+        assert ' border="1"' in out_border_default
+        assert out_border_true == out_border_default
+        assert out_border_default == out_border_explicit_default
+        assert out_border_default != out_border_nondefault
+        assert ' border="2"' in out_border_nondefault
+        assert ' border="0"' not in out_border_zero
+        assert " border" not in out_border_false
+        assert out_border_zero == out_border_false
+
+    @pytest.mark.parametrize(
+        "displayed_only,exp0,exp1",
+        [
+            (True, ["foo"], None),
+            (False, ["foo  bar  baz  qux"], DataFrame(["foo"])),
+        ],
+    )
+    def test_displayed_only(self, displayed_only, exp0, exp1, flavor_read_html):
+        # GH 20027
+        data = """<html>
+          <body>
+            <table>
+              <tr>
+                <td>
+                  foo
+                  <span style="display:none;text-align:center">bar</span>
+                  <span style="display:none">baz</span>
+                  <span style="display: none">qux</span>
+                </td>
+              </tr>
+            </table>
+            <table style="display: none">
+              <tr>
+                <td>foo</td>
+              </tr>
+            </table>
+          </body>
+        </html>"""
+
+        exp0 = DataFrame(exp0)
+        dfs = flavor_read_html(StringIO(data), displayed_only=displayed_only)
+        tm.assert_frame_equal(dfs[0], exp0)
+
+        if exp1 is not None:
+            tm.assert_frame_equal(dfs[1], exp1)
+        else:
+            assert len(dfs) == 1  # Should not parse hidden table
+
+    @pytest.mark.parametrize("displayed_only", [True, False])
+    def test_displayed_only_with_many_elements(self, displayed_only, flavor_read_html):
+        html_table = """
+        <table>
+            <tr>
+                <th>A</th>
+                <th>B</th>
+            </tr>
+            <tr>
+                <td>1</td>
+                <td>2</td>
+            </tr>
+            <tr>
+                <td><span style="display:none"></span>4</td>
+                <td>5</td>
+            </tr>
+        </table>
+        """
+        result = flavor_read_html(StringIO(html_table), displayed_only=displayed_only)[
+            0
+        ]
+        expected = DataFrame({"A": [1, 4], "B": [2, 5]})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:You provided Unicode markup but also provided a value for "
+        "from_encoding.*:UserWarning"
+    )
+    def test_encode(self, html_encoding_file, flavor_read_html):
+        base_path = os.path.basename(html_encoding_file)
+        root = os.path.splitext(base_path)[0]
+        _, encoding = root.split("_")
+
+        try:
+            with open(html_encoding_file, "rb") as fobj:
+                from_string = flavor_read_html(
+                    BytesIO(fobj.read()), encoding=encoding, index_col=0
+                ).pop()
+
+            with open(html_encoding_file, "rb") as fobj:
+                from_file_like = flavor_read_html(
+                    BytesIO(fobj.read()), encoding=encoding, index_col=0
+                ).pop()
+
+            from_filename = flavor_read_html(
+                html_encoding_file, encoding=encoding, index_col=0
+            ).pop()
+            tm.assert_frame_equal(from_string, from_file_like)
+            tm.assert_frame_equal(from_string, from_filename)
+        except Exception:
+            # seems utf-16/32 fail on windows
+            if is_platform_windows():
+                if "16" in encoding or "32" in encoding:
+                    pytest.skip()
+            raise
+
+    def test_parse_failure_unseekable(self, flavor_read_html):
+        # Issue #17975
+
+        if flavor_read_html.keywords.get("flavor") == "lxml":
+            pytest.skip("Not applicable for lxml")
+
+        class UnseekableStringIO(StringIO):
+            def seekable(self):
+                return False
+
+        bad = UnseekableStringIO(
+            """
+            <table><tr><td>spam<foobr />eggs</td></tr></table>"""
+        )
+
+        assert flavor_read_html(bad)
+
+        with pytest.raises(ValueError, match="passed a non-rewindable file object"):
+            flavor_read_html(bad)
+
+    def test_parse_failure_rewinds(self, flavor_read_html):
+        # Issue #17975
+
+        class MockFile:
+            def __init__(self, data) -> None:
+                self.data = data
+                self.at_end = False
+
+            def read(self, size=None):
+                data = "" if self.at_end else self.data
+                self.at_end = True
+                return data
+
+            def seek(self, offset):
+                self.at_end = False
+
+            def seekable(self):
+                return True
+
+            def __next__(self): ...
+
+            def __iter__(self) -> Iterator:
+                # `is_file_like` depends on the presence of
+                # the __iter__ attribute.
+                return self
+
+        good = MockFile("<table><tr><td>spam<br />eggs</td></tr></table>")
+        bad = MockFile("<table><tr><td>spam<foobr />eggs</td></tr></table>")
+
+        assert flavor_read_html(good)
+        assert flavor_read_html(bad)
+
+    @pytest.mark.slow
+    @pytest.mark.single_cpu
+    def test_importcheck_thread_safety(self, datapath, flavor_read_html):
+        # see gh-16928
+
+        class ErrorThread(threading.Thread):
+            def run(self):
+                try:
+                    super().run()
+                except Exception as err:
+                    self.err = err
+                else:
+                    self.err = None
+
+        filename = datapath("io", "data", "html", "valid_markup.html")
+        helper_thread1 = ErrorThread(target=flavor_read_html, args=(filename,))
+        helper_thread2 = ErrorThread(target=flavor_read_html, args=(filename,))
+
+        helper_thread1.start()
+        helper_thread2.start()
+
+        while helper_thread1.is_alive() or helper_thread2.is_alive():
+            pass
+        assert None is helper_thread1.err is helper_thread2.err
+
+    def test_parse_path_object(self, datapath, flavor_read_html):
+        # GH 37705
+        file_path_string = datapath("io", "data", "html", "spam.html")
+        file_path = Path(file_path_string)
+        df1 = flavor_read_html(file_path_string)[0]
+        df2 = flavor_read_html(file_path)[0]
+        tm.assert_frame_equal(df1, df2)
+
+    def test_parse_br_as_space(self, flavor_read_html):
+        # GH 29528: pd.read_html() convert <br> to space
+        result = flavor_read_html(
+            StringIO(
+                """
+            <table>
+                <tr>
+                    <th>A</th>
+                </tr>
+                <tr>
+                    <td>word1<br>word2</td>
+                </tr>
+            </table>
+        """
+            )
+        )[0]
+
+        expected = DataFrame(data=[["word1 word2"]], columns=["A"])
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("arg", ["all", "body", "header", "footer"])
+    def test_extract_links(self, arg, flavor_read_html):
+        gh_13141_data = """
+          <table>
+            <tr>
+              <th>HTTP</th>
+              <th>FTP</th>
+              <th><a href="https://en.wiktionary.org/wiki/linkless">Linkless</a></th>
+            </tr>
+            <tr>
+              <td><a href="https://en.wikipedia.org/">Wikipedia</a></td>
+              <td>SURROUNDING <a href="ftp://ftp.us.debian.org/">Debian</a> TEXT</td>
+              <td>Linkless</td>
+            </tr>
+            <tfoot>
+              <tr>
+                <td><a href="https://en.wikipedia.org/wiki/Page_footer">Footer</a></td>
+                <td>
+                  Multiple <a href="1">links:</a> <a href="2">Only first captured.</a>
+                </td>
+              </tr>
+            </tfoot>
+          </table>
+          """
+
+        gh_13141_expected = {
+            "head_ignore": ["HTTP", "FTP", "Linkless"],
+            "head_extract": [
+                ("HTTP", None),
+                ("FTP", None),
+                ("Linkless", "https://en.wiktionary.org/wiki/linkless"),
+            ],
+            "body_ignore": ["Wikipedia", "SURROUNDING Debian TEXT", "Linkless"],
+            "body_extract": [
+                ("Wikipedia", "https://en.wikipedia.org/"),
+                ("SURROUNDING Debian TEXT", "ftp://ftp.us.debian.org/"),
+                ("Linkless", None),
+            ],
+            "footer_ignore": [
+                "Footer",
+                "Multiple links: Only first captured.",
+                None,
+            ],
+            "footer_extract": [
+                ("Footer", "https://en.wikipedia.org/wiki/Page_footer"),
+                ("Multiple links: Only first captured.", "1"),
+                None,
+            ],
+        }
+
+        data_exp = gh_13141_expected["body_ignore"]
+        foot_exp = gh_13141_expected["footer_ignore"]
+        head_exp = gh_13141_expected["head_ignore"]
+        if arg == "all":
+            data_exp = gh_13141_expected["body_extract"]
+            foot_exp = gh_13141_expected["footer_extract"]
+            head_exp = gh_13141_expected["head_extract"]
+        elif arg == "body":
+            data_exp = gh_13141_expected["body_extract"]
+        elif arg == "footer":
+            foot_exp = gh_13141_expected["footer_extract"]
+        elif arg == "header":
+            head_exp = gh_13141_expected["head_extract"]
+
+        result = flavor_read_html(StringIO(gh_13141_data), extract_links=arg)[0]
+        expected = DataFrame([data_exp, foot_exp], columns=head_exp)
+        expected = expected.fillna(np.nan)
+        tm.assert_frame_equal(result, expected)
+
+    def test_extract_links_bad(self, spam_data):
+        msg = (
+            "`extract_links` must be one of "
+            '{None, "header", "footer", "body", "all"}, got "incorrect"'
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_html(spam_data, extract_links="incorrect")
+
+    def test_extract_links_all_no_header(self, flavor_read_html):
+        # GH 48316
+        data = """
+        <table>
+          <tr>
+            <td>
+              <a href='https://google.com'>Google.com</a>
+            </td>
+          </tr>
+        </table>
+        """
+        result = flavor_read_html(StringIO(data), extract_links="all")[0]
+        expected = DataFrame([[("Google.com", "https://google.com")]])
+        tm.assert_frame_equal(result, expected)
+
+    def test_invalid_dtype_backend(self):
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        with pytest.raises(ValueError, match=msg):
+            read_html("test", dtype_backend="numpy")
+
+    def test_style_tag(self, flavor_read_html):
+        # GH 48316
+        data = """
+        <table>
+            <tr>
+                <th>
+                    <style>.style</style>
+                    A
+                    </th>
+                <th>B</th>
+            </tr>
+            <tr>
+                <td>A1</td>
+                <td>B1</td>
+            </tr>
+            <tr>
+                <td>A2</td>
+                <td>B2</td>
+            </tr>
+        </table>
+        """
+        result = flavor_read_html(StringIO(data))[0]
+        expected = DataFrame(data=[["A1", "B1"], ["A2", "B2"]], columns=["A", "B"])
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/test_http_headers.py b/pandas/pandas/tests/io/test_http_headers.py
new file mode 100644
index 0000000000000..3b9c8769ad9dc
--- /dev/null
+++ b/pandas/pandas/tests/io/test_http_headers.py
@@ -0,0 +1,174 @@
+"""
+Tests for the pandas custom headers in http(s) requests
+"""
+
+from functools import partial
+import gzip
+from io import BytesIO
+
+import pytest
+
+from pandas._config import using_string_dtype
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.network,
+    pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+    ),
+]
+
+
+def gzip_bytes(response_bytes):
+    with BytesIO() as bio:
+        with gzip.GzipFile(fileobj=bio, mode="w") as zipper:
+            zipper.write(response_bytes)
+        return bio.getvalue()
+
+
+def csv_responder(df):
+    return df.to_csv(index=False).encode("utf-8")
+
+
+def gz_csv_responder(df):
+    return gzip_bytes(csv_responder(df))
+
+
+def json_responder(df):
+    return df.to_json().encode("utf-8")
+
+
+def gz_json_responder(df):
+    return gzip_bytes(json_responder(df))
+
+
+def html_responder(df):
+    return df.to_html(index=False).encode("utf-8")
+
+
+def parquetpyarrow_reponder(df):
+    return df.to_parquet(index=False, engine="pyarrow")
+
+
+def parquetfastparquet_responder(df):
+    # the fastparquet engine doesn't like to write to a buffer
+    # it can do it via the open_with function being set appropriately
+    # however it automatically calls the close method and wipes the buffer
+    # so just overwrite that attribute on this instance to not do that
+
+    # protected by an importorskip in the respective test
+    import fsspec
+
+    df.to_parquet(
+        "memory://fastparquet_user_agent.parquet",
+        index=False,
+        engine="fastparquet",
+        compression=None,
+    )
+    with fsspec.open("memory://fastparquet_user_agent.parquet", "rb") as f:
+        return f.read()
+
+
+def pickle_respnder(df):
+    with BytesIO() as bio:
+        df.to_pickle(bio)
+        return bio.getvalue()
+
+
+def stata_responder(df):
+    with BytesIO() as bio:
+        df.to_stata(bio, write_index=False)
+        return bio.getvalue()
+
+
+@pytest.mark.parametrize(
+    "responder, read_method",
+    [
+        (csv_responder, pd.read_csv),
+        (json_responder, pd.read_json),
+        (
+            html_responder,
+            lambda *args, **kwargs: pd.read_html(*args, **kwargs)[0],
+        ),
+        pytest.param(
+            parquetpyarrow_reponder,
+            partial(pd.read_parquet, engine="pyarrow"),
+            marks=td.skip_if_no("pyarrow"),
+        ),
+        pytest.param(
+            parquetfastparquet_responder,
+            partial(pd.read_parquet, engine="fastparquet"),
+            marks=[
+                td.skip_if_no("fastparquet"),
+                td.skip_if_no("fsspec"),
+                pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string"),
+            ],
+        ),
+        (pickle_respnder, pd.read_pickle),
+        (stata_responder, pd.read_stata),
+        (gz_csv_responder, pd.read_csv),
+        (gz_json_responder, pd.read_json),
+    ],
+)
+@pytest.mark.parametrize(
+    "storage_options",
+    [
+        None,
+        {"User-Agent": "foo"},
+        {"User-Agent": "foo", "Auth": "bar"},
+    ],
+)
+def test_request_headers(responder, read_method, httpserver, storage_options):
+    expected = pd.DataFrame({"a": ["b"]})
+    default_headers = ["Accept-Encoding", "Host", "Connection", "User-Agent"]
+    if "gz" in responder.__name__:
+        extra = {"Content-Encoding": "gzip"}
+        if storage_options is None:
+            storage_options = extra
+        else:
+            storage_options |= extra
+    else:
+        extra = None
+    expected_headers = set(default_headers).union(
+        storage_options.keys() if storage_options else []
+    )
+    httpserver.serve_content(content=responder(expected), headers=extra)
+    result = read_method(httpserver.url, storage_options=storage_options)
+    tm.assert_frame_equal(result, expected)
+
+    request_headers = dict(httpserver.requests[0].headers)
+    for header in expected_headers:
+        exp = request_headers.pop(header)
+        if storage_options and header in storage_options:
+            assert exp == storage_options[header]
+    # No extra headers added
+    assert not request_headers
+
+
+@pytest.mark.parametrize(
+    "engine",
+    [
+        "pyarrow",
+        "fastparquet",
+    ],
+)
+def test_to_parquet_to_disk_with_storage_options(engine):
+    headers = {
+        "User-Agent": "custom",
+        "Auth": "other_custom",
+    }
+
+    pytest.importorskip(engine)
+
+    true_df = pd.DataFrame({"column_name": ["column_value"]})
+    msg = (
+        "storage_options passed with file object or non-fsspec file path|"
+        "storage_options passed with buffer, or non-supported URL"
+    )
+    with pytest.raises(ValueError, match=msg):
+        true_df.to_parquet("/tmp/junk.parquet", storage_options=headers, engine=engine)
diff --git a/pandas/pandas/tests/io/test_orc.py b/pandas/pandas/tests/io/test_orc.py
new file mode 100644
index 0000000000000..efb3dffecd856
--- /dev/null
+++ b/pandas/pandas/tests/io/test_orc.py
@@ -0,0 +1,442 @@
+"""test orc compat"""
+
+import datetime
+from decimal import Decimal
+from io import BytesIO
+import os
+import pathlib
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import read_orc
+import pandas._testing as tm
+from pandas.core.arrays import StringArray
+
+pytest.importorskip("pyarrow.orc")
+
+import pyarrow as pa
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+)
+
+
+@pytest.fixture
+def dirpath(datapath):
+    return datapath("io", "data", "orc")
+
+
+def test_orc_reader_empty(dirpath, using_infer_string):
+    columns = [
+        "boolean1",
+        "byte1",
+        "short1",
+        "int1",
+        "long1",
+        "float1",
+        "double1",
+        "bytes1",
+        "string1",
+    ]
+    dtypes = [
+        "bool",
+        "int8",
+        "int16",
+        "int32",
+        "int64",
+        "float32",
+        "float64",
+        "object",
+        "str" if using_infer_string else "object",
+    ]
+    expected = pd.DataFrame(index=pd.RangeIndex(0))
+    for colname, dtype in zip(columns, dtypes):
+        expected[colname] = pd.Series(dtype=dtype)
+    expected.columns = expected.columns.astype("str")
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.emptyFile.orc")
+    got = read_orc(inputfile, columns=columns)
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_reader_basic(dirpath):
+    data = {
+        "boolean1": np.array([False, True], dtype="bool"),
+        "byte1": np.array([1, 100], dtype="int8"),
+        "short1": np.array([1024, 2048], dtype="int16"),
+        "int1": np.array([65536, 65536], dtype="int32"),
+        "long1": np.array([9223372036854775807, 9223372036854775807], dtype="int64"),
+        "float1": np.array([1.0, 2.0], dtype="float32"),
+        "double1": np.array([-15.0, -5.0], dtype="float64"),
+        "bytes1": np.array([b"\x00\x01\x02\x03\x04", b""], dtype="object"),
+        "string1": np.array(["hi", "bye"], dtype="object"),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.test1.orc")
+    got = read_orc(inputfile, columns=data.keys())
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_reader_decimal(dirpath):
+    # Only testing the first 10 rows of data
+    data = {
+        "_col0": np.array(
+            [
+                Decimal("-1000.50000"),
+                Decimal("-999.60000"),
+                Decimal("-998.70000"),
+                Decimal("-997.80000"),
+                Decimal("-996.90000"),
+                Decimal("-995.10000"),
+                Decimal("-994.11000"),
+                Decimal("-993.12000"),
+                Decimal("-992.13000"),
+                Decimal("-991.14000"),
+            ],
+            dtype="object",
+        )
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.decimal.orc")
+    got = read_orc(inputfile).iloc[:10]
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_reader_date_low(dirpath):
+    data = {
+        "time": np.array(
+            [
+                "1900-05-05 12:34:56.100000",
+                "1900-05-05 12:34:56.100100",
+                "1900-05-05 12:34:56.100200",
+                "1900-05-05 12:34:56.100300",
+                "1900-05-05 12:34:56.100400",
+                "1900-05-05 12:34:56.100500",
+                "1900-05-05 12:34:56.100600",
+                "1900-05-05 12:34:56.100700",
+                "1900-05-05 12:34:56.100800",
+                "1900-05-05 12:34:56.100900",
+            ],
+            dtype="datetime64[ns]",
+        ),
+        "date": np.array(
+            [
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+                datetime.date(1900, 12, 25),
+            ],
+            dtype="object",
+        ),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.testDate1900.orc")
+    got = read_orc(inputfile).iloc[:10]
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_reader_date_high(dirpath):
+    data = {
+        "time": np.array(
+            [
+                "2038-05-05 12:34:56.100000",
+                "2038-05-05 12:34:56.100100",
+                "2038-05-05 12:34:56.100200",
+                "2038-05-05 12:34:56.100300",
+                "2038-05-05 12:34:56.100400",
+                "2038-05-05 12:34:56.100500",
+                "2038-05-05 12:34:56.100600",
+                "2038-05-05 12:34:56.100700",
+                "2038-05-05 12:34:56.100800",
+                "2038-05-05 12:34:56.100900",
+            ],
+            dtype="datetime64[ns]",
+        ),
+        "date": np.array(
+            [
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+                datetime.date(2038, 12, 25),
+            ],
+            dtype="object",
+        ),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.testDate2038.orc")
+    got = read_orc(inputfile).iloc[:10]
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_reader_snappy_compressed(dirpath):
+    data = {
+        "int1": np.array(
+            [
+                -1160101563,
+                1181413113,
+                2065821249,
+                -267157795,
+                172111193,
+                1752363137,
+                1406072123,
+                1911809390,
+                -1308542224,
+                -467100286,
+            ],
+            dtype="int32",
+        ),
+        "string1": np.array(
+            [
+                "f50dcb8",
+                "382fdaaa",
+                "90758c6",
+                "9e8caf3f",
+                "ee97332b",
+                "d634da1",
+                "2bea4396",
+                "d67d89e8",
+                "ad71007e",
+                "e8c82066",
+            ],
+            dtype="object",
+        ),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    inputfile = os.path.join(dirpath, "TestOrcFile.testSnappy.orc")
+    got = read_orc(inputfile).iloc[:10]
+
+    tm.assert_equal(expected, got)
+
+
+def test_orc_roundtrip_file(dirpath):
+    # GH44554
+    # PyArrow gained ORC write support with the current argument order
+    pytest.importorskip("pyarrow")
+
+    data = {
+        "boolean1": np.array([False, True], dtype="bool"),
+        "byte1": np.array([1, 100], dtype="int8"),
+        "short1": np.array([1024, 2048], dtype="int16"),
+        "int1": np.array([65536, 65536], dtype="int32"),
+        "long1": np.array([9223372036854775807, 9223372036854775807], dtype="int64"),
+        "float1": np.array([1.0, 2.0], dtype="float32"),
+        "double1": np.array([-15.0, -5.0], dtype="float64"),
+        "bytes1": np.array([b"\x00\x01\x02\x03\x04", b""], dtype="object"),
+        "string1": np.array(["hi", "bye"], dtype="object"),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    with tm.ensure_clean() as path:
+        expected.to_orc(path)
+        got = read_orc(path)
+
+        tm.assert_equal(expected, got)
+
+
+def test_orc_roundtrip_bytesio():
+    # GH44554
+    # PyArrow gained ORC write support with the current argument order
+    pytest.importorskip("pyarrow")
+
+    data = {
+        "boolean1": np.array([False, True], dtype="bool"),
+        "byte1": np.array([1, 100], dtype="int8"),
+        "short1": np.array([1024, 2048], dtype="int16"),
+        "int1": np.array([65536, 65536], dtype="int32"),
+        "long1": np.array([9223372036854775807, 9223372036854775807], dtype="int64"),
+        "float1": np.array([1.0, 2.0], dtype="float32"),
+        "double1": np.array([-15.0, -5.0], dtype="float64"),
+        "bytes1": np.array([b"\x00\x01\x02\x03\x04", b""], dtype="object"),
+        "string1": np.array(["hi", "bye"], dtype="object"),
+    }
+    expected = pd.DataFrame.from_dict(data)
+
+    bytes = expected.to_orc()
+    got = read_orc(BytesIO(bytes))
+
+    tm.assert_equal(expected, got)
+
+
+@pytest.mark.parametrize(
+    "orc_writer_dtypes_not_supported",
+    [
+        np.array([1, 20], dtype="uint64"),
+        pd.Series(["a", "b", "a"], dtype="category"),
+        [pd.Interval(left=0, right=2), pd.Interval(left=0, right=5)],
+        [pd.Period("2022-01-03", freq="D"), pd.Period("2022-01-04", freq="D")],
+    ],
+)
+def test_orc_writer_dtypes_not_supported(orc_writer_dtypes_not_supported):
+    # GH44554
+    # PyArrow gained ORC write support with the current argument order
+    pytest.importorskip("pyarrow")
+
+    df = pd.DataFrame({"unimpl": orc_writer_dtypes_not_supported})
+    msg = "The dtype of one or more columns is not supported yet."
+    with pytest.raises(NotImplementedError, match=msg):
+        df.to_orc()
+
+
+def test_orc_dtype_backend_pyarrow(using_infer_string):
+    pytest.importorskip("pyarrow")
+    df = pd.DataFrame(
+        {
+            "string": list("abc"),
+            "string_with_nan": ["a", np.nan, "c"],
+            "string_with_none": ["a", None, "c"],
+            "bytes": [b"foo", b"bar", None],
+            "int": list(range(1, 4)),
+            "float": np.arange(4.0, 7.0, dtype="float64"),
+            "float_with_nan": [2.0, np.nan, 3.0],
+            "bool": [True, False, True],
+            "bool_with_na": [True, False, None],
+            "datetime": pd.date_range("20130101", periods=3),
+            "datetime_with_nat": [
+                pd.Timestamp("20130101"),
+                pd.NaT,
+                pd.Timestamp("20130103"),
+            ],
+        }
+    )
+    # FIXME: without casting to ns we do not round-trip correctly
+    df["datetime_with_nat"] = df["datetime_with_nat"].astype("M8[ns]")
+
+    bytes_data = df.copy().to_orc()
+    result = read_orc(BytesIO(bytes_data), dtype_backend="pyarrow")
+
+    expected = pd.DataFrame(
+        {
+            col: pd.arrays.ArrowExtensionArray(pa.array(df[col], from_pandas=True))
+            for col in df.columns
+        }
+    )
+    if using_infer_string:
+        # ORC does not preserve distinction between string and large string
+        # -> the default large string comes back as string
+        string_dtype = pd.ArrowDtype(pa.string())
+        expected["string"] = expected["string"].astype(string_dtype)
+        expected["string_with_nan"] = expected["string_with_nan"].astype(string_dtype)
+        expected["string_with_none"] = expected["string_with_none"].astype(string_dtype)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_orc_dtype_backend_numpy_nullable():
+    # GH#50503
+    pytest.importorskip("pyarrow")
+    df = pd.DataFrame(
+        {
+            "string": list("abc"),
+            "string_with_nan": ["a", np.nan, "c"],
+            "string_with_none": ["a", None, "c"],
+            "int": list(range(1, 4)),
+            "int_with_nan": pd.Series([1, pd.NA, 3], dtype="Int64"),
+            "na_only": pd.Series([pd.NA, pd.NA, pd.NA], dtype="Int64"),
+            "float": np.arange(4.0, 7.0, dtype="float64"),
+            "float_with_nan": [2.0, np.nan, 3.0],
+            "bool": [True, False, True],
+            "bool_with_na": [True, False, None],
+        }
+    )
+
+    bytes_data = df.copy().to_orc()
+    result = read_orc(BytesIO(bytes_data), dtype_backend="numpy_nullable")
+
+    expected = pd.DataFrame(
+        {
+            "string": StringArray(np.array(["a", "b", "c"], dtype=np.object_)),
+            "string_with_nan": StringArray(
+                np.array(["a", pd.NA, "c"], dtype=np.object_)
+            ),
+            "string_with_none": StringArray(
+                np.array(["a", pd.NA, "c"], dtype=np.object_)
+            ),
+            "int": pd.Series([1, 2, 3], dtype="Int64"),
+            "int_with_nan": pd.Series([1, pd.NA, 3], dtype="Int64"),
+            "na_only": pd.Series([pd.NA, pd.NA, pd.NA], dtype="Int64"),
+            "float": pd.Series([4.0, 5.0, 6.0], dtype="Float64"),
+            "float_with_nan": pd.Series([2.0, pd.NA, 3.0], dtype="Float64"),
+            "bool": pd.Series([True, False, True], dtype="boolean"),
+            "bool_with_na": pd.Series([True, False, pd.NA], dtype="boolean"),
+        }
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_orc_uri_path():
+    expected = pd.DataFrame({"int": list(range(1, 4))})
+    with tm.ensure_clean("tmp.orc") as path:
+        expected.to_orc(path)
+        uri = pathlib.Path(path).as_uri()
+        result = read_orc(uri)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        pd.RangeIndex(start=2, stop=5, step=1),
+        pd.RangeIndex(start=0, stop=3, step=1, name="non-default"),
+        pd.Index([1, 2, 3]),
+    ],
+)
+def test_to_orc_non_default_index(index):
+    df = pd.DataFrame({"a": [1, 2, 3]}, index=index)
+    msg = (
+        "orc does not support serializing a non-default index|"
+        "orc does not serialize index meta-data"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.to_orc()
+
+
+def test_invalid_dtype_backend():
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    df = pd.DataFrame({"int": list(range(1, 4))})
+    with tm.ensure_clean("tmp.orc") as path:
+        df.to_orc(path)
+        with pytest.raises(ValueError, match=msg):
+            read_orc(path, dtype_backend="numpy")
+
+
+def test_string_inference(tmp_path):
+    # GH#54431
+    path = tmp_path / "test_string_inference.p"
+    df = pd.DataFrame(data={"a": ["x", "y"]})
+    df.to_orc(path)
+    with pd.option_context("future.infer_string", True):
+        result = read_orc(path)
+    expected = pd.DataFrame(
+        data={"a": ["x", "y"]},
+        dtype=pd.StringDtype(na_value=np.nan),
+        columns=pd.Index(["a"], dtype=pd.StringDtype(na_value=np.nan)),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/io/test_parquet.py b/pandas/pandas/tests/io/test_parquet.py
new file mode 100644
index 0000000000000..56a8e4c439164
--- /dev/null
+++ b/pandas/pandas/tests/io/test_parquet.py
@@ -0,0 +1,1492 @@
+"""test parquet compat"""
+
+import datetime
+from decimal import Decimal
+from io import BytesIO
+import os
+import pathlib
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import is_platform_windows
+from pandas.compat.pyarrow import (
+    pa_version_under11p0,
+    pa_version_under13p0,
+    pa_version_under15p0,
+    pa_version_under17p0,
+    pa_version_under19p0,
+)
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.util.version import Version
+
+from pandas.io.parquet import (
+    FastParquetImpl,
+    PyArrowImpl,
+    get_engine,
+    read_parquet,
+    to_parquet,
+)
+
+try:
+    import pyarrow
+
+    _HAVE_PYARROW = True
+except ImportError:
+    _HAVE_PYARROW = False
+
+try:
+    import fastparquet
+
+    _HAVE_FASTPARQUET = True
+except ImportError:
+    _HAVE_FASTPARQUET = False
+
+
+pytestmark = [
+    pytest.mark.filterwarnings("ignore:DataFrame._data is deprecated:FutureWarning"),
+    pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+    ),
+]
+
+
+# setup engines & skips
+@pytest.fixture(
+    params=[
+        pytest.param(
+            "fastparquet",
+            marks=[
+                pytest.mark.skipif(
+                    not _HAVE_FASTPARQUET,
+                    reason="fastparquet is not installed",
+                ),
+                pytest.mark.xfail(
+                    using_string_dtype(),
+                    reason="TODO(infer_string) fastparquet",
+                    strict=False,
+                ),
+            ],
+        ),
+        pytest.param(
+            "pyarrow",
+            marks=pytest.mark.skipif(
+                not _HAVE_PYARROW, reason="pyarrow is not installed"
+            ),
+        ),
+    ]
+)
+def engine(request):
+    return request.param
+
+
+@pytest.fixture
+def pa():
+    if not _HAVE_PYARROW:
+        pytest.skip("pyarrow is not installed")
+    return "pyarrow"
+
+
+@pytest.fixture
+def fp(request):
+    if not _HAVE_FASTPARQUET:
+        pytest.skip("fastparquet is not installed")
+    if using_string_dtype():
+        request.applymarker(
+            pytest.mark.xfail(reason="TODO(infer_string) fastparquet", strict=False)
+        )
+    return "fastparquet"
+
+
+@pytest.fixture
+def df_compat():
+    return pd.DataFrame({"A": [1, 2, 3], "B": "foo"}, columns=pd.Index(["A", "B"]))
+
+
+@pytest.fixture
+def df_cross_compat():
+    df = pd.DataFrame(
+        {
+            "a": list("abc"),
+            "b": list(range(1, 4)),
+            # 'c': np.arange(3, 6).astype('u1'),
+            "d": np.arange(4.0, 7.0, dtype="float64"),
+            "e": [True, False, True],
+            "f": pd.date_range("20130101", periods=3),
+            # 'g': pd.date_range('20130101', periods=3,
+            #                    tz='US/Eastern'),
+            # 'h': pd.date_range('20130101', periods=3, freq='ns')
+        }
+    )
+    return df
+
+
+@pytest.fixture
+def df_full():
+    return pd.DataFrame(
+        {
+            "string": list("abc"),
+            "string_with_nan": ["a", np.nan, "c"],
+            "string_with_none": ["a", None, "c"],
+            "bytes": [b"foo", b"bar", b"baz"],
+            "unicode": ["foo", "bar", "baz"],
+            "int": list(range(1, 4)),
+            "uint": np.arange(3, 6).astype("u1"),
+            "float": np.arange(4.0, 7.0, dtype="float64"),
+            "float_with_nan": [2.0, np.nan, 3.0],
+            "bool": [True, False, True],
+            "datetime": pd.date_range("20130101", periods=3),
+            "datetime_with_nat": [
+                pd.Timestamp("20130101"),
+                pd.NaT,
+                pd.Timestamp("20130103"),
+            ],
+        }
+    )
+
+
+@pytest.fixture(
+    params=[
+        datetime.datetime.now(datetime.timezone.utc),
+        datetime.datetime.now(datetime.timezone.min),
+        datetime.datetime.now(datetime.timezone.max),
+        datetime.datetime.strptime("2019-01-04T16:41:24+0200", "%Y-%m-%dT%H:%M:%S%z"),
+        datetime.datetime.strptime("2019-01-04T16:41:24+0215", "%Y-%m-%dT%H:%M:%S%z"),
+        datetime.datetime.strptime("2019-01-04T16:41:24-0200", "%Y-%m-%dT%H:%M:%S%z"),
+        datetime.datetime.strptime("2019-01-04T16:41:24-0215", "%Y-%m-%dT%H:%M:%S%z"),
+    ]
+)
+def timezone_aware_date_list(request):
+    return request.param
+
+
+def check_round_trip(
+    df,
+    engine=None,
+    path=None,
+    write_kwargs=None,
+    read_kwargs=None,
+    expected=None,
+    check_names=True,
+    check_like=False,
+    check_dtype=True,
+    repeat=2,
+):
+    """Verify parquet serializer and deserializer produce the same results.
+
+    Performs a pandas to disk and disk to pandas round trip,
+    then compares the 2 resulting DataFrames to verify equality.
+
+    Parameters
+    ----------
+    df: Dataframe
+    engine: str, optional
+        'pyarrow' or 'fastparquet'
+    path: str, optional
+    write_kwargs: dict of str:str, optional
+    read_kwargs: dict of str:str, optional
+    expected: DataFrame, optional
+        Expected deserialization result, otherwise will be equal to `df`
+    check_names: list of str, optional
+        Closed set of column names to be compared
+    check_like: bool, optional
+        If True, ignore the order of index & columns.
+    repeat: int, optional
+        How many times to repeat the test
+    """
+    write_kwargs = write_kwargs or {"compression": None}
+    read_kwargs = read_kwargs or {}
+
+    if expected is None:
+        expected = df
+
+    if engine:
+        write_kwargs["engine"] = engine
+        read_kwargs["engine"] = engine
+
+    def compare(repeat):
+        for _ in range(repeat):
+            df.to_parquet(path, **write_kwargs)
+            actual = read_parquet(path, **read_kwargs)
+
+            if "string_with_nan" in expected:
+                expected.loc[1, "string_with_nan"] = None
+            tm.assert_frame_equal(
+                expected,
+                actual,
+                check_names=check_names,
+                check_like=check_like,
+                check_dtype=check_dtype,
+            )
+
+    if path is None:
+        with tm.ensure_clean() as path:
+            compare(repeat)
+    else:
+        compare(repeat)
+
+
+def check_partition_names(path, expected):
+    """Check partitions of a parquet file are as expected.
+
+    Parameters
+    ----------
+    path: str
+        Path of the dataset.
+    expected: iterable of str
+        Expected partition names.
+    """
+    import pyarrow.dataset as ds
+
+    dataset = ds.dataset(path, partitioning="hive")
+    assert dataset.partitioning.schema.names == expected
+
+
+def test_invalid_engine(df_compat):
+    msg = "engine must be one of 'pyarrow', 'fastparquet'"
+    with pytest.raises(ValueError, match=msg):
+        check_round_trip(df_compat, "foo", "bar")
+
+
+def test_options_py(df_compat, pa, using_infer_string):
+    # use the set option
+    if using_infer_string and not pa_version_under19p0:
+        df_compat.columns = df_compat.columns.astype("str")
+
+    with pd.option_context("io.parquet.engine", "pyarrow"):
+        check_round_trip(df_compat)
+
+
+def test_options_fp(df_compat, fp):
+    # use the set option
+
+    with pd.option_context("io.parquet.engine", "fastparquet"):
+        check_round_trip(df_compat)
+
+
+def test_options_auto(df_compat, fp, pa):
+    # use the set option
+
+    with pd.option_context("io.parquet.engine", "auto"):
+        check_round_trip(df_compat)
+
+
+def test_options_get_engine(fp, pa):
+    assert isinstance(get_engine("pyarrow"), PyArrowImpl)
+    assert isinstance(get_engine("fastparquet"), FastParquetImpl)
+
+    with pd.option_context("io.parquet.engine", "pyarrow"):
+        assert isinstance(get_engine("auto"), PyArrowImpl)
+        assert isinstance(get_engine("pyarrow"), PyArrowImpl)
+        assert isinstance(get_engine("fastparquet"), FastParquetImpl)
+
+    with pd.option_context("io.parquet.engine", "fastparquet"):
+        assert isinstance(get_engine("auto"), FastParquetImpl)
+        assert isinstance(get_engine("pyarrow"), PyArrowImpl)
+        assert isinstance(get_engine("fastparquet"), FastParquetImpl)
+
+    with pd.option_context("io.parquet.engine", "auto"):
+        assert isinstance(get_engine("auto"), PyArrowImpl)
+        assert isinstance(get_engine("pyarrow"), PyArrowImpl)
+        assert isinstance(get_engine("fastparquet"), FastParquetImpl)
+
+
+def test_get_engine_auto_error_message():
+    # Expect different error messages from get_engine(engine="auto")
+    # if engines aren't installed vs. are installed but bad version
+    from pandas.compat._optional import VERSIONS
+
+    # Do we have engines installed, but a bad version of them?
+    pa_min_ver = VERSIONS.get("pyarrow")
+    fp_min_ver = VERSIONS.get("fastparquet")
+    have_pa_bad_version = (
+        False
+        if not _HAVE_PYARROW
+        else Version(pyarrow.__version__) < Version(pa_min_ver)
+    )
+    have_fp_bad_version = (
+        False
+        if not _HAVE_FASTPARQUET
+        else Version(fastparquet.__version__) < Version(fp_min_ver)
+    )
+    # Do we have usable engines installed?
+    have_usable_pa = _HAVE_PYARROW and not have_pa_bad_version
+    have_usable_fp = _HAVE_FASTPARQUET and not have_fp_bad_version
+
+    if not have_usable_pa and not have_usable_fp:
+        # No usable engines found.
+        if have_pa_bad_version:
+            match = f"Pandas requires version .{pa_min_ver}. or newer of .pyarrow."
+            with pytest.raises(ImportError, match=match):
+                get_engine("auto")
+        else:
+            match = "Missing optional dependency .pyarrow."
+            with pytest.raises(ImportError, match=match):
+                get_engine("auto")
+
+        if have_fp_bad_version:
+            match = f"Pandas requires version .{fp_min_ver}. or newer of .fastparquet."
+            with pytest.raises(ImportError, match=match):
+                get_engine("auto")
+        else:
+            match = "Missing optional dependency .fastparquet."
+            with pytest.raises(ImportError, match=match):
+                get_engine("auto")
+
+
+def test_cross_engine_pa_fp(df_cross_compat, pa, fp):
+    # cross-compat with differing reading/writing engines
+
+    df = df_cross_compat
+    with tm.ensure_clean() as path:
+        df.to_parquet(path, engine=pa, compression=None)
+
+        result = read_parquet(path, engine=fp)
+        tm.assert_frame_equal(result, df)
+
+        result = read_parquet(path, engine=fp, columns=["a", "d"])
+        tm.assert_frame_equal(result, df[["a", "d"]])
+
+
+def test_cross_engine_fp_pa(df_cross_compat, pa, fp):
+    # cross-compat with differing reading/writing engines
+    df = df_cross_compat
+    with tm.ensure_clean() as path:
+        df.to_parquet(path, engine=fp, compression=None)
+
+        result = read_parquet(path, engine=pa)
+        tm.assert_frame_equal(result, df)
+
+        result = read_parquet(path, engine=pa, columns=["a", "d"])
+        tm.assert_frame_equal(result, df[["a", "d"]])
+
+
+class Base:
+    def check_error_on_write(self, df, engine, exc, err_msg):
+        # check that we are raising the exception on writing
+        with tm.ensure_clean() as path:
+            with pytest.raises(exc, match=err_msg):
+                to_parquet(df, path, engine, compression=None)
+
+    def check_external_error_on_write(self, df, engine, exc):
+        # check that an external library is raising the exception on writing
+        with tm.ensure_clean() as path:
+            with tm.external_error_raised(exc):
+                to_parquet(df, path, engine, compression=None)
+
+
+class TestBasic(Base):
+    def test_error(self, engine):
+        for obj in [
+            pd.Series([1, 2, 3]),
+            1,
+            "foo",
+            pd.Timestamp("20130101"),
+            np.array([1, 2, 3]),
+        ]:
+            msg = "to_parquet only supports IO with DataFrames"
+            self.check_error_on_write(obj, engine, ValueError, msg)
+
+    def test_columns_dtypes(self, engine):
+        df = pd.DataFrame({"string": list("abc"), "int": list(range(1, 4))})
+
+        # unicode
+        df.columns = ["foo", "bar"]
+        check_round_trip(df, engine)
+
+    @pytest.mark.parametrize("compression", [None, "gzip", "snappy", "brotli"])
+    def test_compression(self, engine, compression):
+        df = pd.DataFrame({"A": [1, 2, 3]})
+        check_round_trip(df, engine, write_kwargs={"compression": compression})
+
+    def test_read_columns(self, engine):
+        # GH18154
+        df = pd.DataFrame({"string": list("abc"), "int": list(range(1, 4))})
+
+        expected = pd.DataFrame({"string": list("abc")})
+        check_round_trip(
+            df, engine, expected=expected, read_kwargs={"columns": ["string"]}
+        )
+
+    def test_read_filters(self, engine, tmp_path):
+        df = pd.DataFrame(
+            {
+                "int": list(range(4)),
+                "part": list("aabb"),
+            }
+        )
+
+        expected = pd.DataFrame({"int": [0, 1]})
+        check_round_trip(
+            df,
+            engine,
+            path=tmp_path,
+            expected=expected,
+            write_kwargs={"partition_cols": ["part"]},
+            read_kwargs={"filters": [("part", "==", "a")], "columns": ["int"]},
+            repeat=1,
+        )
+
+    def test_write_index(self):
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame({"A": [1, 2, 3]})
+        check_round_trip(df, "pyarrow")
+
+        indexes = [
+            [2, 3, 4],
+            pd.date_range("20130101", periods=3),
+            list("abc"),
+            [1, 3, 4],
+        ]
+        # non-default index
+        for index in indexes:
+            df.index = index
+            if isinstance(index, pd.DatetimeIndex):
+                df.index = df.index._with_freq(None)  # freq doesn't round-trip
+            check_round_trip(df, "pyarrow")
+
+        # index with meta-data
+        df.index = [0, 1, 2]
+        df.index.name = "foo"
+        check_round_trip(df, "pyarrow")
+
+    def test_write_multiindex(self, pa):
+        # Not supported in fastparquet as of 0.1.3 or older pyarrow version
+        engine = pa
+
+        df = pd.DataFrame({"A": [1, 2, 3]})
+        index = pd.MultiIndex.from_tuples([("a", 1), ("a", 2), ("b", 1)])
+        df.index = index
+        check_round_trip(df, engine)
+
+    def test_multiindex_with_columns(self, pa):
+        engine = pa
+        dates = pd.date_range("01-Jan-2018", "01-Dec-2018", freq="MS")
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((2 * len(dates), 3)),
+            columns=list("ABC"),
+        )
+        index1 = pd.MultiIndex.from_product(
+            [["Level1", "Level2"], dates], names=["level", "date"]
+        )
+        index2 = index1.copy(names=None)
+        for index in [index1, index2]:
+            df.index = index
+
+            check_round_trip(df, engine)
+            check_round_trip(
+                df, engine, read_kwargs={"columns": ["A", "B"]}, expected=df[["A", "B"]]
+            )
+
+    def test_write_ignoring_index(self, engine):
+        # ENH 20768
+        # Ensure index=False omits the index from the written Parquet file.
+        df = pd.DataFrame({"a": [1, 2, 3], "b": ["q", "r", "s"]})
+
+        write_kwargs = {"compression": None, "index": False}
+
+        # Because we're dropping the index, we expect the loaded dataframe to
+        # have the default integer index.
+        expected = df.reset_index(drop=True)
+
+        check_round_trip(df, engine, write_kwargs=write_kwargs, expected=expected)
+
+        # Ignore custom index
+        df = pd.DataFrame(
+            {"a": [1, 2, 3], "b": ["q", "r", "s"]}, index=["zyx", "wvu", "tsr"]
+        )
+
+        check_round_trip(df, engine, write_kwargs=write_kwargs, expected=expected)
+
+        # Ignore multi-indexes as well.
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        df = pd.DataFrame(
+            {"one": list(range(8)), "two": [-i for i in range(8)]}, index=arrays
+        )
+
+        expected = df.reset_index(drop=True)
+        check_round_trip(df, engine, write_kwargs=write_kwargs, expected=expected)
+
+    def test_write_column_multiindex(self, engine):
+        # Not able to write column multi-indexes with non-string column names.
+        mi_columns = pd.MultiIndex.from_tuples([("a", 1), ("a", 2), ("b", 1)])
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((4, 3)), columns=mi_columns
+        )
+
+        if engine == "fastparquet":
+            self.check_error_on_write(
+                df, engine, TypeError, "Column name must be a string"
+            )
+        elif engine == "pyarrow":
+            check_round_trip(df, engine)
+
+    def test_write_column_multiindex_nonstring(self, engine):
+        # GH #34777
+
+        # Not able to write column multi-indexes with non-string column names
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            [1, 2, 1, 2, 1, 2, 1, 2],
+        ]
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((8, 8)), columns=arrays
+        )
+        df.columns.names = ["Level1", "Level2"]
+        if engine == "fastparquet":
+            self.check_error_on_write(df, engine, ValueError, "Column name")
+        elif engine == "pyarrow":
+            check_round_trip(df, engine)
+
+    def test_write_column_multiindex_string(self, pa):
+        # GH #34777
+        # Not supported in fastparquet as of 0.1.3
+        engine = pa
+
+        # Write column multi-indexes with string column names
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((8, 8)), columns=arrays
+        )
+        df.columns.names = ["ColLevel1", "ColLevel2"]
+
+        check_round_trip(df, engine)
+
+    def test_write_column_index_string(self, pa):
+        # GH #34777
+        # Not supported in fastparquet as of 0.1.3
+        engine = pa
+
+        # Write column indexes with string column names
+        arrays = ["bar", "baz", "foo", "qux"]
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)), columns=arrays
+        )
+        df.columns.name = "StringCol"
+
+        check_round_trip(df, engine)
+
+    def test_write_column_index_nonstring(self, engine):
+        # GH #34777
+
+        # Write column indexes with string column names
+        arrays = [1, 2, 3, 4]
+        df = pd.DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)), columns=arrays
+        )
+        df.columns.name = "NonStringCol"
+        if engine == "fastparquet":
+            self.check_error_on_write(
+                df, engine, TypeError, "Column name must be a string"
+            )
+        else:
+            check_round_trip(df, engine)
+
+    def test_dtype_backend(self, engine, request):
+        pq = pytest.importorskip("pyarrow.parquet")
+
+        if engine == "fastparquet":
+            # We are manually disabling fastparquet's
+            # nullable dtype support pending discussion
+            mark = pytest.mark.xfail(
+                reason="Fastparquet nullable dtype support is disabled"
+            )
+            request.applymarker(mark)
+
+        table = pyarrow.table(
+            {
+                "a": pyarrow.array([1, 2, 3, None], "int64"),
+                "b": pyarrow.array([1, 2, 3, None], "uint8"),
+                "c": pyarrow.array(["a", "b", "c", None]),
+                "d": pyarrow.array([True, False, True, None]),
+                # Test that nullable dtypes used even in absence of nulls
+                "e": pyarrow.array([1, 2, 3, 4], "int64"),
+                # GH 45694
+                "f": pyarrow.array([1.0, 2.0, 3.0, None], "float32"),
+                "g": pyarrow.array([1.0, 2.0, 3.0, None], "float64"),
+            }
+        )
+        with tm.ensure_clean() as path:
+            # write manually with pyarrow to write integers
+            pq.write_table(table, path)
+            result1 = read_parquet(path, engine=engine)
+            result2 = read_parquet(path, engine=engine, dtype_backend="numpy_nullable")
+
+        assert result1["a"].dtype == np.dtype("float64")
+        expected = pd.DataFrame(
+            {
+                "a": pd.array([1, 2, 3, None], dtype="Int64"),
+                "b": pd.array([1, 2, 3, None], dtype="UInt8"),
+                "c": pd.array(["a", "b", "c", None], dtype="string"),
+                "d": pd.array([True, False, True, None], dtype="boolean"),
+                "e": pd.array([1, 2, 3, 4], dtype="Int64"),
+                "f": pd.array([1.0, 2.0, 3.0, None], dtype="Float32"),
+                "g": pd.array([1.0, 2.0, 3.0, None], dtype="Float64"),
+            }
+        )
+        if engine == "fastparquet":
+            # Fastparquet doesn't support string columns yet
+            # Only int and boolean
+            result2 = result2.drop("c", axis=1)
+            expected = expected.drop("c", axis=1)
+        tm.assert_frame_equal(result2, expected)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "Int64",
+            "UInt8",
+            "boolean",
+            "object",
+            "datetime64[ns, UTC]",
+            "float",
+            "period[D]",
+            "Float64",
+            "string",
+        ],
+    )
+    def test_read_empty_array(self, pa, dtype):
+        # GH #41241
+        df = pd.DataFrame(
+            {
+                "value": pd.array([], dtype=dtype),
+            }
+        )
+        pytest.importorskip("pyarrow", "11.0.0")
+        # GH 45694
+        expected = None
+        if dtype == "float":
+            expected = pd.DataFrame(
+                {
+                    "value": pd.array([], dtype="Float64"),
+                }
+            )
+        check_round_trip(
+            df, pa, read_kwargs={"dtype_backend": "numpy_nullable"}, expected=expected
+        )
+
+    @pytest.mark.network
+    @pytest.mark.single_cpu
+    def test_parquet_read_from_url(self, httpserver, datapath, df_compat, engine):
+        if engine != "auto":
+            pytest.importorskip(engine)
+        with open(datapath("io", "data", "parquet", "simple.parquet"), mode="rb") as f:
+            httpserver.serve_content(content=f.read())
+            df = read_parquet(httpserver.url, engine=engine)
+
+        expected = df_compat
+        if pa_version_under19p0:
+            expected.columns = expected.columns.astype(object)
+        tm.assert_frame_equal(df, expected)
+
+
+class TestParquetPyArrow(Base):
+    @pytest.mark.xfail(reason="datetime_with_nat unit doesn't round-trip")
+    def test_basic(self, pa, df_full):
+        df = df_full
+        pytest.importorskip("pyarrow", "11.0.0")
+
+        # additional supported types for pyarrow
+        dti = pd.date_range("20130101", periods=3, tz="Europe/Brussels")
+        dti = dti._with_freq(None)  # freq doesn't round-trip
+        df["datetime_tz"] = dti
+        df["bool_with_none"] = [True, None, True]
+
+        check_round_trip(df, pa)
+
+    def test_basic_subset_columns(self, pa, df_full):
+        # GH18628
+
+        df = df_full
+        # additional supported types for pyarrow
+        df["datetime_tz"] = pd.date_range("20130101", periods=3, tz="Europe/Brussels")
+
+        check_round_trip(
+            df,
+            pa,
+            expected=df[["string", "int"]],
+            read_kwargs={"columns": ["string", "int"]},
+        )
+
+    def test_to_bytes_without_path_or_buf_provided(self, pa, df_full):
+        # GH 37105
+        buf_bytes = df_full.to_parquet(engine=pa)
+        assert isinstance(buf_bytes, bytes)
+
+        buf_stream = BytesIO(buf_bytes)
+        res = read_parquet(buf_stream)
+
+        expected = df_full.copy()
+        expected.loc[1, "string_with_nan"] = None
+        if pa_version_under11p0:
+            expected["datetime_with_nat"] = expected["datetime_with_nat"].astype(
+                "M8[ns]"
+            )
+        else:
+            expected["datetime_with_nat"] = expected["datetime_with_nat"].astype(
+                "M8[ms]"
+            )
+        tm.assert_frame_equal(res, expected)
+
+    def test_duplicate_columns(self, pa):
+        # not currently able to handle duplicate columns
+        df = pd.DataFrame(np.arange(12).reshape(4, 3), columns=list("aaa")).copy()
+        self.check_error_on_write(df, pa, ValueError, "Duplicate column names found")
+
+    def test_timedelta(self, pa):
+        df = pd.DataFrame({"a": pd.timedelta_range("1 day", periods=3)})
+        check_round_trip(df, pa)
+
+    def test_unsupported(self, pa):
+        # mixed python objects
+        df = pd.DataFrame({"a": ["a", 1, 2.0]})
+        # pyarrow 0.11 raises ArrowTypeError
+        # older pyarrows raise ArrowInvalid
+        self.check_external_error_on_write(df, pa, pyarrow.ArrowException)
+
+    def test_unsupported_float16(self, pa):
+        # #44847, #44914
+        # Not able to write float 16 column using pyarrow.
+        data = np.arange(2, 10, dtype=np.float16)
+        df = pd.DataFrame(data=data, columns=["fp16"])
+        if pa_version_under15p0:
+            self.check_external_error_on_write(df, pa, pyarrow.ArrowException)
+        else:
+            check_round_trip(df, pa)
+
+    @pytest.mark.xfail(
+        is_platform_windows(),
+        reason=(
+            "PyArrow does not cleanup of partial files dumps when unsupported "
+            "dtypes are passed to_parquet function in windows"
+        ),
+    )
+    @pytest.mark.skipif(not pa_version_under15p0, reason="float16 works on 15")
+    @pytest.mark.parametrize("path_type", [str, pathlib.Path])
+    def test_unsupported_float16_cleanup(self, pa, path_type):
+        # #44847, #44914
+        # Not able to write float 16 column using pyarrow.
+        # Tests cleanup by pyarrow in case of an error
+        data = np.arange(2, 10, dtype=np.float16)
+        df = pd.DataFrame(data=data, columns=["fp16"])
+
+        with tm.ensure_clean() as path_str:
+            path = path_type(path_str)
+            with tm.external_error_raised(pyarrow.ArrowException):
+                df.to_parquet(path=path, engine=pa)
+            assert not os.path.isfile(path)
+
+    def test_categorical(self, pa):
+        # supported in >= 0.7.0
+        df = pd.DataFrame(
+            {
+                "a": pd.Categorical(list("abcdef")),
+                # test for null, out-of-order values, and unobserved category
+                "b": pd.Categorical(
+                    ["bar", "foo", "foo", "bar", None, "bar"],
+                    dtype=pd.CategoricalDtype(["foo", "bar", "baz"]),
+                ),
+                # test for ordered flag
+                "c": pd.Categorical(
+                    ["a", "b", "c", "a", "c", "b"],
+                    categories=["b", "c", "d"],
+                    ordered=True,
+                ),
+            }
+        )
+
+        check_round_trip(df, pa)
+
+    @pytest.mark.single_cpu
+    def test_s3_roundtrip_explicit_fs(self, df_compat, s3_public_bucket, pa, s3so):
+        s3fs = pytest.importorskip("s3fs")
+        s3 = s3fs.S3FileSystem(**s3so)
+        kw = {"filesystem": s3}
+        check_round_trip(
+            df_compat,
+            pa,
+            path=f"{s3_public_bucket.name}/pyarrow.parquet",
+            read_kwargs=kw,
+            write_kwargs=kw,
+        )
+
+    @pytest.mark.single_cpu
+    def test_s3_roundtrip(self, df_compat, s3_public_bucket, pa, s3so):
+        # GH #19134
+        s3so = {"storage_options": s3so}
+        check_round_trip(
+            df_compat,
+            pa,
+            path=f"s3://{s3_public_bucket.name}/pyarrow.parquet",
+            read_kwargs=s3so,
+            write_kwargs=s3so,
+        )
+
+    @pytest.mark.single_cpu
+    @pytest.mark.parametrize("partition_col", [["A"], []])
+    def test_s3_roundtrip_for_dir(
+        self, df_compat, s3_public_bucket, pa, partition_col, s3so
+    ):
+        pytest.importorskip("s3fs")
+        # GH #26388
+        expected_df = df_compat.copy()
+
+        # GH #35791
+        if partition_col:
+            expected_df = expected_df.astype(dict.fromkeys(partition_col, np.int32))
+            partition_col_type = "category"
+
+            expected_df[partition_col] = expected_df[partition_col].astype(
+                partition_col_type
+            )
+
+        check_round_trip(
+            df_compat,
+            pa,
+            expected=expected_df,
+            path=f"s3://{s3_public_bucket.name}/parquet_dir",
+            read_kwargs={"storage_options": s3so},
+            write_kwargs={
+                "partition_cols": partition_col,
+                "compression": None,
+                "storage_options": s3so,
+            },
+            check_like=True,
+            repeat=1,
+        )
+
+    def test_read_file_like_obj_support(self, df_compat, using_infer_string):
+        pytest.importorskip("pyarrow")
+        buffer = BytesIO()
+        df_compat.to_parquet(buffer)
+        df_from_buf = read_parquet(buffer)
+        if using_infer_string and not pa_version_under19p0:
+            df_compat.columns = df_compat.columns.astype("str")
+        tm.assert_frame_equal(df_compat, df_from_buf)
+
+    def test_expand_user(self, df_compat, monkeypatch):
+        pytest.importorskip("pyarrow")
+        monkeypatch.setenv("HOME", "TestingUser")
+        monkeypatch.setenv("USERPROFILE", "TestingUser")
+        with pytest.raises(OSError, match=r".*TestingUser.*"):
+            read_parquet("~/file.parquet")
+        with pytest.raises(OSError, match=r".*TestingUser.*"):
+            df_compat.to_parquet("~/file.parquet")
+
+    def test_partition_cols_supported(self, tmp_path, pa, df_full):
+        # GH #23283
+        partition_cols = ["bool", "int"]
+        df = df_full
+        df.to_parquet(tmp_path, partition_cols=partition_cols, compression=None)
+        check_partition_names(tmp_path, partition_cols)
+        assert read_parquet(tmp_path).shape == df.shape
+
+    def test_partition_cols_string(self, tmp_path, pa, df_full):
+        # GH #27117
+        partition_cols = "bool"
+        partition_cols_list = [partition_cols]
+        df = df_full
+        df.to_parquet(tmp_path, partition_cols=partition_cols, compression=None)
+        check_partition_names(tmp_path, partition_cols_list)
+        assert read_parquet(tmp_path).shape == df.shape
+
+    @pytest.mark.parametrize(
+        "path_type", [str, lambda x: x], ids=["string", "pathlib.Path"]
+    )
+    def test_partition_cols_pathlib(self, tmp_path, pa, df_compat, path_type):
+        # GH 35902
+
+        partition_cols = "B"
+        partition_cols_list = [partition_cols]
+        df = df_compat
+
+        path = path_type(tmp_path)
+        df.to_parquet(path, partition_cols=partition_cols_list)
+        assert read_parquet(path).shape == df.shape
+
+    def test_empty_dataframe(self, pa):
+        # GH #27339
+        df = pd.DataFrame(index=[], columns=[])
+        check_round_trip(df, pa)
+
+    def test_write_with_schema(self, pa):
+        import pyarrow
+
+        df = pd.DataFrame({"x": [0, 1]})
+        schema = pyarrow.schema([pyarrow.field("x", type=pyarrow.bool_())])
+        out_df = df.astype(bool)
+        check_round_trip(df, pa, write_kwargs={"schema": schema}, expected=out_df)
+
+    def test_additional_extension_arrays(self, pa, using_infer_string):
+        # test additional ExtensionArrays that are supported through the
+        # __arrow_array__ protocol
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame(
+            {
+                "a": pd.Series([1, 2, 3], dtype="Int64"),
+                "b": pd.Series([1, 2, 3], dtype="UInt32"),
+                "c": pd.Series(["a", None, "c"], dtype="string"),
+            }
+        )
+        if using_infer_string and pa_version_under19p0:
+            check_round_trip(df, pa, expected=df.astype({"c": "str"}))
+        else:
+            check_round_trip(df, pa)
+
+        df = pd.DataFrame({"a": pd.Series([1, 2, 3, None], dtype="Int64")})
+        check_round_trip(df, pa)
+
+    def test_pyarrow_backed_string_array(self, pa, string_storage, using_infer_string):
+        # test ArrowStringArray supported through the __arrow_array__ protocol
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame({"a": pd.Series(["a", None, "c"], dtype="string[pyarrow]")})
+        with pd.option_context("string_storage", string_storage):
+            if using_infer_string:
+                if pa_version_under19p0:
+                    expected = df.astype("str")
+                else:
+                    expected = df.astype(f"string[{string_storage}]")
+                expected.columns = expected.columns.astype("str")
+            else:
+                expected = df.astype(f"string[{string_storage}]")
+            check_round_trip(df, pa, expected=expected)
+
+    def test_additional_extension_types(self, pa):
+        # test additional ExtensionArrays that are supported through the
+        # __arrow_array__ protocol + by defining a custom ExtensionType
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame(
+            {
+                "c": pd.IntervalIndex.from_tuples([(0, 1), (1, 2), (3, 4)]),
+                "d": pd.period_range("2012-01-01", periods=3, freq="D"),
+                # GH-45881 issue with interval with datetime64[ns] subtype
+                "e": pd.IntervalIndex.from_breaks(
+                    pd.date_range("2012-01-01", periods=4, freq="D")
+                ),
+            }
+        )
+        check_round_trip(df, pa)
+
+    def test_timestamp_nanoseconds(self, pa):
+        # with version 2.6, pyarrow defaults to writing the nanoseconds, so
+        # this should work without error
+        # Note in previous pyarrows(<7.0.0), only the pseudo-version 2.0 was available
+        ver = "2.6"
+        df = pd.DataFrame({"a": pd.date_range("2017-01-01", freq="1ns", periods=10)})
+        check_round_trip(df, pa, write_kwargs={"version": ver})
+
+    def test_timezone_aware_index(self, pa, timezone_aware_date_list):
+        pytest.importorskip("pyarrow", "11.0.0")
+
+        idx = 5 * [timezone_aware_date_list]
+        df = pd.DataFrame(index=idx, data={"index_as_col": idx})
+
+        # see gh-36004
+        # compare time(zone) values only, skip their class:
+        # pyarrow always creates fixed offset timezones using pytz.FixedOffset()
+        # even if it was datetime.timezone() originally
+        #
+        # technically they are the same:
+        # they both implement datetime.tzinfo
+        # they both wrap datetime.timedelta()
+        # this use-case sets the resolution to 1 minute
+
+        expected = df[:]
+        if pa_version_under11p0:
+            expected.index = expected.index.as_unit("ns")
+        if timezone_aware_date_list.tzinfo != datetime.timezone.utc:
+            # pyarrow returns pytz.FixedOffset while pandas constructs datetime.timezone
+            # https://github.com/pandas-dev/pandas/issues/37286
+            try:
+                import pytz
+            except ImportError:
+                pass
+            else:
+                offset = df.index.tz.utcoffset(timezone_aware_date_list)
+                tz = pytz.FixedOffset(offset.total_seconds() / 60)
+                expected.index = expected.index.tz_convert(tz)
+                expected["index_as_col"] = expected["index_as_col"].dt.tz_convert(tz)
+        check_round_trip(df, pa, check_dtype=False, expected=expected)
+
+    def test_filter_row_groups(self, pa):
+        # https://github.com/pandas-dev/pandas/issues/26551
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame({"a": list(range(3))})
+        with tm.ensure_clean() as path:
+            df.to_parquet(path, engine=pa)
+            result = read_parquet(path, pa, filters=[("a", "==", 0)])
+        assert len(result) == 1
+
+    @pytest.mark.filterwarnings("ignore:make_block is deprecated:DeprecationWarning")
+    def test_read_dtype_backend_pyarrow_config(self, pa, df_full):
+        import pyarrow
+
+        df = df_full
+
+        # additional supported types for pyarrow
+        dti = pd.date_range("20130101", periods=3, tz="Europe/Brussels")
+        dti = dti._with_freq(None)  # freq doesn't round-trip
+        df["datetime_tz"] = dti
+        df["bool_with_none"] = [True, None, True]
+
+        pa_table = pyarrow.Table.from_pandas(df)
+        expected = pa_table.to_pandas(types_mapper=pd.ArrowDtype)
+        if pa_version_under13p0:
+            # pyarrow infers datetimes as us instead of ns
+            expected["datetime"] = expected["datetime"].astype("timestamp[us][pyarrow]")
+            expected["datetime_tz"] = expected["datetime_tz"].astype(
+                pd.ArrowDtype(pyarrow.timestamp(unit="us", tz="Europe/Brussels"))
+            )
+
+        expected["datetime_with_nat"] = expected["datetime_with_nat"].astype(
+            "timestamp[ms][pyarrow]"
+        )
+
+        check_round_trip(
+            df,
+            engine=pa,
+            read_kwargs={"dtype_backend": "pyarrow"},
+            expected=expected,
+        )
+
+    def test_read_dtype_backend_pyarrow_config_index(self, pa):
+        df = pd.DataFrame(
+            {"a": [1, 2]}, index=pd.Index([3, 4], name="test"), dtype="int64[pyarrow]"
+        )
+        expected = df.copy()
+        import pyarrow
+
+        if Version(pyarrow.__version__) > Version("11.0.0"):
+            expected.index = expected.index.astype("int64[pyarrow]")
+        check_round_trip(
+            df,
+            engine=pa,
+            read_kwargs={"dtype_backend": "pyarrow"},
+            expected=expected,
+        )
+
+    @pytest.mark.xfail(
+        pa_version_under17p0, reason="pa.pandas_compat passes 'datetime64' to .astype"
+    )
+    def test_columns_dtypes_not_invalid(self, pa):
+        df = pd.DataFrame({"string": list("abc"), "int": list(range(1, 4))})
+
+        # numeric
+        df.columns = [0, 1]
+        check_round_trip(df, pa)
+
+        # bytes
+        df.columns = [b"foo", b"bar"]
+        with pytest.raises(NotImplementedError, match="|S3"):
+            # Bytes fails on read_parquet
+            check_round_trip(df, pa)
+
+        # python object
+        df.columns = [
+            datetime.datetime(2011, 1, 1, 0, 0),
+            datetime.datetime(2011, 1, 1, 1, 1),
+        ]
+        check_round_trip(df, pa)
+
+    def test_empty_columns(self, pa):
+        # GH 52034
+        df = pd.DataFrame(index=pd.Index(["a", "b", "c"], name="custom name"))
+        check_round_trip(df, pa)
+
+    def test_df_attrs_persistence(self, tmp_path, pa):
+        path = tmp_path / "test_df_metadata.p"
+        df = pd.DataFrame(data={1: [1]})
+        df.attrs = {"test_attribute": 1}
+        df.to_parquet(path, engine=pa)
+        new_df = read_parquet(path, engine=pa)
+        assert new_df.attrs == df.attrs
+
+    def test_string_inference(self, tmp_path, pa, using_infer_string):
+        # GH#54431
+        path = tmp_path / "test_string_inference.p"
+        df = pd.DataFrame(data={"a": ["x", "y"]}, index=["a", "b"])
+        df.to_parquet(path, engine=pa)
+        with pd.option_context("future.infer_string", True):
+            result = read_parquet(path, engine=pa)
+        dtype = pd.StringDtype(na_value=np.nan)
+        expected = pd.DataFrame(
+            data={"a": ["x", "y"]},
+            dtype=dtype,
+            index=pd.Index(["a", "b"], dtype=dtype),
+            columns=pd.Index(
+                ["a"],
+                dtype=object
+                if pa_version_under19p0 and not using_infer_string
+                else dtype,
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.skipif(pa_version_under11p0, reason="not supported before 11.0")
+    def test_roundtrip_decimal(self, tmp_path, pa):
+        # GH#54768
+        import pyarrow as pa
+
+        path = tmp_path / "decimal.p"
+        df = pd.DataFrame({"a": [Decimal("123.00")]}, dtype="string[pyarrow]")
+        df.to_parquet(path, schema=pa.schema([("a", pa.decimal128(5))]))
+        result = read_parquet(path)
+        if pa_version_under19p0:
+            expected = pd.DataFrame({"a": ["123"]}, dtype="string[python]")
+        else:
+            expected = pd.DataFrame({"a": [Decimal("123.00")]}, dtype="object")
+        tm.assert_frame_equal(result, expected)
+
+    def test_infer_string_large_string_type(self, tmp_path, pa):
+        # GH#54798
+        import pyarrow as pa
+        import pyarrow.parquet as pq
+
+        path = tmp_path / "large_string.p"
+
+        table = pa.table({"a": pa.array([None, "b", "c"], pa.large_string())})
+        pq.write_table(table, path)
+
+        with pd.option_context("future.infer_string", True):
+            result = read_parquet(path)
+        expected = pd.DataFrame(
+            data={"a": [None, "b", "c"]},
+            dtype=pd.StringDtype(na_value=np.nan),
+            columns=pd.Index(["a"], dtype=pd.StringDtype(na_value=np.nan)),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    # NOTE: this test is not run by default, because it requires a lot of memory (>5GB)
+    # @pytest.mark.slow
+    # def test_string_column_above_2GB(self, tmp_path, pa):
+    #     # https://github.com/pandas-dev/pandas/issues/55606
+    #     # above 2GB of string data
+    #     v1 = b"x" * 100000000
+    #     v2 = b"x" * 147483646
+    #     df = pd.DataFrame({"strings": [v1] * 20 + [v2] + ["x"] * 20}, dtype="string")
+    #     df.to_parquet(tmp_path / "test.parquet")
+    #     result = read_parquet(tmp_path / "test.parquet")
+    #     assert result["strings"].dtype == "string"
+    # FIXME: don't leave commented-out
+
+    def test_non_nanosecond_timestamps(self, temp_file):
+        # GH#49236
+        pa = pytest.importorskip("pyarrow", "11.0.0")
+        pq = pytest.importorskip("pyarrow.parquet")
+
+        arr = pa.array([datetime.datetime(1600, 1, 1)], type=pa.timestamp("us"))
+        table = pa.table([arr], names=["timestamp"])
+        pq.write_table(table, temp_file)
+        result = read_parquet(temp_file)
+        expected = pd.DataFrame(
+            data={"timestamp": [datetime.datetime(1600, 1, 1)]},
+            dtype="datetime64[us]",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_maps_as_pydicts(self, pa):
+        pyarrow = pytest.importorskip("pyarrow", "13.0.0")
+
+        schema = pyarrow.schema(
+            [("foo", pyarrow.map_(pyarrow.string(), pyarrow.int64()))]
+        )
+        df = pd.DataFrame([{"foo": {"A": 1}}, {"foo": {"B": 2}}])
+        check_round_trip(
+            df,
+            pa,
+            write_kwargs={"schema": schema},
+            read_kwargs={"to_pandas_kwargs": {"maps_as_pydicts": "strict"}},
+        )
+
+
+class TestParquetFastParquet(Base):
+    def test_basic(self, fp, df_full, request):
+        pytz = pytest.importorskip("pytz")
+        import fastparquet
+
+        if Version(fastparquet.__version__) < Version("2024.11.0"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=("datetime_with_nat gets incorrect values"),
+                )
+            )
+
+        tz = pytz.timezone("US/Eastern")
+        df = df_full
+
+        dti = pd.date_range("20130101", periods=3, tz=tz)
+        dti = dti._with_freq(None)  # freq doesn't round-trip
+        df["datetime_tz"] = dti
+        df["timedelta"] = pd.timedelta_range("1 day", periods=3)
+        check_round_trip(df, fp)
+
+    def test_columns_dtypes_invalid(self, fp):
+        df = pd.DataFrame({"string": list("abc"), "int": list(range(1, 4))})
+
+        err = TypeError
+        msg = "Column name must be a string"
+
+        # numeric
+        df.columns = [0, 1]
+        self.check_error_on_write(df, fp, err, msg)
+
+        # bytes
+        df.columns = [b"foo", b"bar"]
+        self.check_error_on_write(df, fp, err, msg)
+
+        # python object
+        df.columns = [
+            datetime.datetime(2011, 1, 1, 0, 0),
+            datetime.datetime(2011, 1, 1, 1, 1),
+        ]
+        self.check_error_on_write(df, fp, err, msg)
+
+    def test_duplicate_columns(self, fp):
+        # not currently able to handle duplicate columns
+        df = pd.DataFrame(np.arange(12).reshape(4, 3), columns=list("aaa")).copy()
+        msg = "Cannot create parquet dataset with duplicate column names"
+        self.check_error_on_write(df, fp, ValueError, msg)
+
+    def test_bool_with_none(self, fp, request):
+        import fastparquet
+
+        if Version(fastparquet.__version__) < Version("2024.11.0") and Version(
+            np.__version__
+        ) >= Version("2.0.0"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=("fastparquet uses np.float_ in numpy2"),
+                )
+            )
+        df = pd.DataFrame({"a": [True, None, False]})
+        expected = pd.DataFrame({"a": [1.0, np.nan, 0.0]}, dtype="float16")
+        # Fastparquet bug in 0.7.1 makes it so that this dtype becomes
+        # float64
+        check_round_trip(df, fp, expected=expected, check_dtype=False)
+
+    def test_unsupported(self, fp):
+        # period
+        df = pd.DataFrame({"a": pd.period_range("2013", freq="M", periods=3)})
+        # error from fastparquet -> don't check exact error message
+        self.check_error_on_write(df, fp, ValueError, None)
+
+        # mixed
+        df = pd.DataFrame({"a": ["a", 1, 2.0]})
+        msg = "Can't infer object conversion type"
+        self.check_error_on_write(df, fp, ValueError, msg)
+
+    def test_categorical(self, fp):
+        df = pd.DataFrame({"a": pd.Categorical(list("abc"))})
+        check_round_trip(df, fp)
+
+    def test_filter_row_groups(self, fp):
+        d = {"a": list(range(3))}
+        df = pd.DataFrame(d)
+        with tm.ensure_clean() as path:
+            df.to_parquet(path, engine=fp, compression=None, row_group_offsets=1)
+            result = read_parquet(path, fp, filters=[("a", "==", 0)])
+        assert len(result) == 1
+
+    @pytest.mark.single_cpu
+    def test_s3_roundtrip(self, df_compat, s3_public_bucket, fp, s3so):
+        # GH #19134
+        check_round_trip(
+            df_compat,
+            fp,
+            path=f"s3://{s3_public_bucket.name}/fastparquet.parquet",
+            read_kwargs={"storage_options": s3so},
+            write_kwargs={"compression": None, "storage_options": s3so},
+        )
+
+    def test_partition_cols_supported(self, tmp_path, fp, df_full):
+        # GH #23283
+        partition_cols = ["bool", "int"]
+        df = df_full
+        df.to_parquet(
+            tmp_path,
+            engine="fastparquet",
+            partition_cols=partition_cols,
+            compression=None,
+        )
+        assert os.path.exists(tmp_path)
+        import fastparquet
+
+        actual_partition_cols = fastparquet.ParquetFile(str(tmp_path), False).cats
+        assert len(actual_partition_cols) == 2
+
+    def test_partition_cols_string(self, tmp_path, fp, df_full):
+        # GH #27117
+        partition_cols = "bool"
+        df = df_full
+        df.to_parquet(
+            tmp_path,
+            engine="fastparquet",
+            partition_cols=partition_cols,
+            compression=None,
+        )
+        assert os.path.exists(tmp_path)
+        import fastparquet
+
+        actual_partition_cols = fastparquet.ParquetFile(str(tmp_path), False).cats
+        assert len(actual_partition_cols) == 1
+
+    def test_partition_on_supported(self, tmp_path, fp, df_full):
+        # GH #23283
+        partition_cols = ["bool", "int"]
+        df = df_full
+        df.to_parquet(
+            tmp_path,
+            engine="fastparquet",
+            compression=None,
+            partition_on=partition_cols,
+        )
+        assert os.path.exists(tmp_path)
+        import fastparquet
+
+        actual_partition_cols = fastparquet.ParquetFile(str(tmp_path), False).cats
+        assert len(actual_partition_cols) == 2
+
+    def test_error_on_using_partition_cols_and_partition_on(
+        self, tmp_path, fp, df_full
+    ):
+        # GH #23283
+        partition_cols = ["bool", "int"]
+        df = df_full
+        msg = (
+            "Cannot use both partition_on and partition_cols. Use partition_cols for "
+            "partitioning data"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.to_parquet(
+                tmp_path,
+                engine="fastparquet",
+                compression=None,
+                partition_on=partition_cols,
+                partition_cols=partition_cols,
+            )
+
+    def test_empty_dataframe(self, fp):
+        # GH #27339
+        df = pd.DataFrame()
+        expected = df.copy()
+        check_round_trip(df, fp, expected=expected)
+
+    def test_timezone_aware_index(self, fp, timezone_aware_date_list, request):
+        import fastparquet
+
+        if Version(fastparquet.__version__) < Version("2024.11.0"):
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=(
+                        "fastparquet bug, see "
+                        "https://github.com/dask/fastparquet/issues/929"
+                    ),
+                )
+            )
+
+        idx = 5 * [timezone_aware_date_list]
+
+        df = pd.DataFrame(index=idx, data={"index_as_col": idx})
+
+        expected = df.copy()
+        expected.index.name = "index"
+        check_round_trip(df, fp, expected=expected)
+
+    def test_close_file_handle_on_read_error(self):
+        with tm.ensure_clean("test.parquet") as path:
+            pathlib.Path(path).write_bytes(b"breakit")
+            with tm.external_error_raised(Exception):  # Not important which exception
+                read_parquet(path, engine="fastparquet")
+            # The next line raises an error on Windows if the file is still open
+            pathlib.Path(path).unlink(missing_ok=False)
+
+    def test_bytes_file_name(self, engine):
+        # GH#48944
+        df = pd.DataFrame(data={"A": [0, 1], "B": [1, 0]})
+        with tm.ensure_clean("test.parquet") as path:
+            with open(path.encode(), "wb") as f:
+                df.to_parquet(f)
+
+            result = read_parquet(path, engine=engine)
+        tm.assert_frame_equal(result, df)
+
+    def test_filesystem_notimplemented(self):
+        pytest.importorskip("fastparquet")
+        df = pd.DataFrame(data={"A": [0, 1], "B": [1, 0]})
+        with tm.ensure_clean() as path:
+            with pytest.raises(
+                NotImplementedError, match="filesystem is not implemented"
+            ):
+                df.to_parquet(path, engine="fastparquet", filesystem="foo")
+
+        with tm.ensure_clean() as path:
+            pathlib.Path(path).write_bytes(b"foo")
+            with pytest.raises(
+                NotImplementedError, match="filesystem is not implemented"
+            ):
+                read_parquet(path, engine="fastparquet", filesystem="foo")
+
+    def test_invalid_filesystem(self):
+        pytest.importorskip("pyarrow")
+        df = pd.DataFrame(data={"A": [0, 1], "B": [1, 0]})
+        with tm.ensure_clean() as path:
+            with pytest.raises(
+                ValueError, match="filesystem must be a pyarrow or fsspec FileSystem"
+            ):
+                df.to_parquet(path, engine="pyarrow", filesystem="foo")
+
+        with tm.ensure_clean() as path:
+            pathlib.Path(path).write_bytes(b"foo")
+            with pytest.raises(
+                ValueError, match="filesystem must be a pyarrow or fsspec FileSystem"
+            ):
+                read_parquet(path, engine="pyarrow", filesystem="foo")
+
+    def test_unsupported_pa_filesystem_storage_options(self):
+        pa_fs = pytest.importorskip("pyarrow.fs")
+        df = pd.DataFrame(data={"A": [0, 1], "B": [1, 0]})
+        with tm.ensure_clean() as path:
+            with pytest.raises(
+                NotImplementedError,
+                match="storage_options not supported with a pyarrow FileSystem.",
+            ):
+                df.to_parquet(
+                    path,
+                    engine="pyarrow",
+                    filesystem=pa_fs.LocalFileSystem(),
+                    storage_options={"foo": "bar"},
+                )
+
+        with tm.ensure_clean() as path:
+            pathlib.Path(path).write_bytes(b"foo")
+            with pytest.raises(
+                NotImplementedError,
+                match="storage_options not supported with a pyarrow FileSystem.",
+            ):
+                read_parquet(
+                    path,
+                    engine="pyarrow",
+                    filesystem=pa_fs.LocalFileSystem(),
+                    storage_options={"foo": "bar"},
+                )
+
+    def test_invalid_dtype_backend(self, engine):
+        msg = (
+            "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+            "'pyarrow' are allowed."
+        )
+        df = pd.DataFrame({"int": list(range(1, 4))})
+        with tm.ensure_clean("tmp.parquet") as path:
+            df.to_parquet(path)
+            with pytest.raises(ValueError, match=msg):
+                read_parquet(path, dtype_backend="numpy")
diff --git a/pandas/pandas/tests/io/test_pickle.py b/pandas/pandas/tests/io/test_pickle.py
new file mode 100644
index 0000000000000..bab2c1561eb99
--- /dev/null
+++ b/pandas/pandas/tests/io/test_pickle.py
@@ -0,0 +1,524 @@
+"""
+manage legacy pickle tests
+
+How to add pickle tests:
+
+1. Install pandas version intended to output the pickle.
+
+2. Execute "generate_legacy_storage_files.py" to create the pickle.
+$ python generate_legacy_storage_files.py <output_dir> pickle
+
+3. Move the created pickle to "data/legacy_pickle/<version>" directory.
+"""
+
+from __future__ import annotations
+
+import bz2
+import datetime
+import functools
+from functools import partial
+import gzip
+import io
+import os
+from pathlib import Path
+import pickle
+import shutil
+import tarfile
+from typing import Any
+import uuid
+import zipfile
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_little_endian
+from pandas.compat._optional import import_optional_dependency
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.tests.io.generate_legacy_storage_files import create_pickle_data
+
+import pandas.io.common as icom
+from pandas.tseries.offsets import (
+    Day,
+    MonthEnd,
+)
+
+
+# ---------------------
+# comparison functions
+# ---------------------
+def compare_element(result, expected, typ):
+    if isinstance(expected, Index):
+        tm.assert_index_equal(expected, result)
+        return
+
+    if typ.startswith("sp_"):
+        tm.assert_equal(result, expected)
+    elif typ == "timestamp":
+        if expected is pd.NaT:
+            assert result is pd.NaT
+        else:
+            assert result == expected
+    else:
+        comparator = getattr(tm, f"assert_{typ}_equal", tm.assert_almost_equal)
+        comparator(result, expected)
+
+
+# ---------------------
+# tests
+# ---------------------
+
+
+def test_pickles(datapath):
+    pytest.importorskip("pytz")
+    if not is_platform_little_endian():
+        pytest.skip("known failure on non-little endian")
+
+    # For loop for compat with --strict-data-files
+    for legacy_pickle in Path(__file__).parent.glob("data/legacy_pickle/*/*.p*kl*"):
+        legacy_pickle = datapath(legacy_pickle)
+
+        data = pd.read_pickle(legacy_pickle)
+
+        for typ, dv in data.items():
+            for dt, result in dv.items():
+                expected = data[typ][dt]
+
+                if typ == "series" and dt == "ts":
+                    # GH 7748
+                    tm.assert_series_equal(result, expected)
+                    assert result.index.freq == expected.index.freq
+                    assert not result.index.freq.normalize
+                    tm.assert_series_equal(result > 0, expected > 0)
+
+                    # GH 9291
+                    freq = result.index.freq
+                    assert freq + Day(1) == Day(2)
+
+                    res = freq + pd.Timedelta(hours=1)
+                    assert isinstance(res, pd.Timedelta)
+                    assert res == pd.Timedelta(days=1, hours=1)
+
+                    res = freq + pd.Timedelta(nanoseconds=1)
+                    assert isinstance(res, pd.Timedelta)
+                    assert res == pd.Timedelta(days=1, nanoseconds=1)
+                elif typ == "index" and dt == "period":
+                    tm.assert_index_equal(result, expected)
+                    assert isinstance(result.freq, MonthEnd)
+                    assert result.freq == MonthEnd()
+                    assert result.freqstr == "M"
+                    tm.assert_index_equal(result.shift(2), expected.shift(2))
+                elif typ == "series" and dt in ("dt_tz", "cat"):
+                    tm.assert_series_equal(result, expected)
+                elif typ == "frame" and dt in (
+                    "dt_mixed_tzs",
+                    "cat_onecol",
+                    "cat_and_float",
+                ):
+                    tm.assert_frame_equal(result, expected)
+                else:
+                    compare_element(result, expected, typ)
+
+
+def python_pickler(obj, path):
+    with open(path, "wb") as fh:
+        pickle.dump(obj, fh, protocol=-1)
+
+
+def python_unpickler(path):
+    with open(path, "rb") as fh:
+        fh.seek(0)
+        return pickle.load(fh)
+
+
+def flatten(data: dict) -> list[tuple[str, Any]]:
+    """Flatten create_pickle_data"""
+    return [
+        (typ, example)
+        for typ, examples in data.items()
+        for example in examples.values()
+    ]
+
+
+@pytest.mark.parametrize(
+    "pickle_writer",
+    [
+        pytest.param(python_pickler, id="python"),
+        pytest.param(pd.to_pickle, id="pandas_proto_default"),
+        pytest.param(
+            functools.partial(pd.to_pickle, protocol=pickle.HIGHEST_PROTOCOL),
+            id="pandas_proto_highest",
+        ),
+        pytest.param(functools.partial(pd.to_pickle, protocol=4), id="pandas_proto_4"),
+        pytest.param(
+            functools.partial(pd.to_pickle, protocol=5),
+            id="pandas_proto_5",
+        ),
+    ],
+)
+@pytest.mark.parametrize("writer", [pd.to_pickle, python_pickler])
+@pytest.mark.parametrize("typ, expected", flatten(create_pickle_data()))
+def test_round_trip_current(typ, expected, pickle_writer, writer):
+    with tm.ensure_clean() as path:
+        # test writing with each pickler
+        pickle_writer(expected, path)
+
+        # test reading with each unpickler
+        result = pd.read_pickle(path)
+        compare_element(result, expected, typ)
+
+        result = python_unpickler(path)
+        compare_element(result, expected, typ)
+
+        # and the same for file objects (GH 35679)
+        with open(path, mode="wb") as handle:
+            writer(expected, path)
+            handle.seek(0)  # shouldn't close file handle
+        with open(path, mode="rb") as handle:
+            result = pd.read_pickle(handle)
+            handle.seek(0)  # shouldn't close file handle
+        compare_element(result, expected, typ)
+
+
+def test_pickle_path_pathlib():
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+    result = tm.round_trip_pathlib(df.to_pickle, pd.read_pickle)
+    tm.assert_frame_equal(df, result)
+
+
+# ---------------------
+# test pickle compression
+# ---------------------
+
+
+@pytest.fixture
+def get_random_path():
+    return f"__{uuid.uuid4()}__.pickle"
+
+
+class TestCompression:
+    _extension_to_compression = icom.extension_to_compression
+
+    def compress_file(self, src_path, dest_path, compression):
+        if compression is None:
+            shutil.copyfile(src_path, dest_path)
+            return
+
+        if compression == "gzip":
+            f = gzip.open(dest_path, "w")
+        elif compression == "bz2":
+            f = bz2.BZ2File(dest_path, "w")
+        elif compression == "zip":
+            with zipfile.ZipFile(dest_path, "w", compression=zipfile.ZIP_DEFLATED) as f:
+                f.write(src_path, os.path.basename(src_path))
+        elif compression == "tar":
+            with open(src_path, "rb") as fh:
+                with tarfile.open(dest_path, mode="w") as tar:
+                    tarinfo = tar.gettarinfo(src_path, os.path.basename(src_path))
+                    tar.addfile(tarinfo, fh)
+        elif compression == "xz":
+            import lzma
+
+            f = lzma.LZMAFile(dest_path, "w")
+        elif compression == "zstd":
+            f = import_optional_dependency("zstandard").open(dest_path, "wb")
+        else:
+            msg = f"Unrecognized compression type: {compression}"
+            raise ValueError(msg)
+
+        if compression not in ["zip", "tar"]:
+            with open(src_path, "rb") as fh:
+                with f:
+                    f.write(fh.read())
+
+    def test_write_explicit(self, compression, get_random_path):
+        base = get_random_path
+        path1 = base + ".compressed"
+        path2 = base + ".raw"
+
+        with tm.ensure_clean(path1) as p1, tm.ensure_clean(path2) as p2:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+
+            # write to compressed file
+            df.to_pickle(p1, compression=compression)
+
+            # decompress
+            with tm.decompress_file(p1, compression=compression) as f:
+                with open(p2, "wb") as fh:
+                    fh.write(f.read())
+
+            # read decompressed file
+            df2 = pd.read_pickle(p2, compression=None)
+
+            tm.assert_frame_equal(df, df2)
+
+    @pytest.mark.parametrize("compression", ["", "None", "bad", "7z"])
+    def test_write_explicit_bad(self, compression, get_random_path):
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        with tm.ensure_clean(get_random_path) as path:
+            with pytest.raises(ValueError, match="Unrecognized compression type"):
+                df.to_pickle(path, compression=compression)
+
+    def test_write_infer(self, compression_ext, get_random_path):
+        base = get_random_path
+        path1 = base + compression_ext
+        path2 = base + ".raw"
+        compression = self._extension_to_compression.get(compression_ext.lower())
+
+        with tm.ensure_clean(path1) as p1, tm.ensure_clean(path2) as p2:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+
+            # write to compressed file by inferred compression method
+            df.to_pickle(p1)
+
+            # decompress
+            with tm.decompress_file(p1, compression=compression) as f:
+                with open(p2, "wb") as fh:
+                    fh.write(f.read())
+
+            # read decompressed file
+            df2 = pd.read_pickle(p2, compression=None)
+
+            tm.assert_frame_equal(df, df2)
+
+    def test_read_explicit(self, compression, get_random_path):
+        base = get_random_path
+        path1 = base + ".raw"
+        path2 = base + ".compressed"
+
+        with tm.ensure_clean(path1) as p1, tm.ensure_clean(path2) as p2:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+
+            # write to uncompressed file
+            df.to_pickle(p1, compression=None)
+
+            # compress
+            self.compress_file(p1, p2, compression=compression)
+
+            # read compressed file
+            df2 = pd.read_pickle(p2, compression=compression)
+            tm.assert_frame_equal(df, df2)
+
+    def test_read_infer(self, compression_ext, get_random_path):
+        base = get_random_path
+        path1 = base + ".raw"
+        path2 = base + compression_ext
+        compression = self._extension_to_compression.get(compression_ext.lower())
+
+        with tm.ensure_clean(path1) as p1, tm.ensure_clean(path2) as p2:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+
+            # write to uncompressed file
+            df.to_pickle(p1, compression=None)
+
+            # compress
+            self.compress_file(p1, p2, compression=compression)
+
+            # read compressed file by inferred compression method
+            df2 = pd.read_pickle(p2)
+            tm.assert_frame_equal(df, df2)
+
+
+# ---------------------
+# test pickle compression
+# ---------------------
+
+
+class TestProtocol:
+    @pytest.mark.parametrize("protocol", [-1, 0, 1, 2])
+    def test_read(self, protocol, get_random_path):
+        with tm.ensure_clean(get_random_path) as path:
+            df = DataFrame(
+                1.1 * np.arange(120).reshape((30, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=Index([f"i-{i}" for i in range(30)], dtype=object),
+            )
+            df.to_pickle(path, protocol=protocol)
+            df2 = pd.read_pickle(path)
+            tm.assert_frame_equal(df, df2)
+
+
+def test_pickle_buffer_roundtrip():
+    with tm.ensure_clean() as path:
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        with open(path, "wb") as fh:
+            df.to_pickle(fh)
+        with open(path, "rb") as fh:
+            result = pd.read_pickle(fh)
+        tm.assert_frame_equal(df, result)
+
+
+def test_pickle_fsspec_roundtrip():
+    pytest.importorskip("fsspec")
+    with tm.ensure_clean():
+        mockurl = "memory://mockfile"
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )
+        df.to_pickle(mockurl)
+        result = pd.read_pickle(mockurl)
+        tm.assert_frame_equal(df, result)
+
+
+class MyTz(datetime.tzinfo):
+    def __init__(self) -> None:
+        pass
+
+
+def test_read_pickle_with_subclass():
+    # GH 12163
+    expected = Series(dtype=object), MyTz()
+    result = tm.round_trip_pickle(expected)
+
+    tm.assert_series_equal(result[0], expected[0])
+    assert isinstance(result[1], MyTz)
+
+
+def test_pickle_binary_object_compression(compression):
+    """
+    Read/write from binary file-objects w/wo compression.
+
+    GH 26237, GH 29054, and GH 29570
+    """
+    df = DataFrame(
+        1.1 * np.arange(120).reshape((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=Index([f"i-{i}" for i in range(30)], dtype=object),
+    )
+
+    # reference for compression
+    with tm.ensure_clean() as path:
+        df.to_pickle(path, compression=compression)
+        reference = Path(path).read_bytes()
+
+    # write
+    buffer = io.BytesIO()
+    df.to_pickle(buffer, compression=compression)
+    buffer.seek(0)
+
+    # gzip  and zip safe the filename: cannot compare the compressed content
+    assert buffer.getvalue() == reference or compression in ("gzip", "zip", "tar")
+
+    # read
+    read_df = pd.read_pickle(buffer, compression=compression)
+    buffer.seek(0)
+    tm.assert_frame_equal(df, read_df)
+
+
+def test_pickle_dataframe_with_multilevel_index(
+    multiindex_year_month_day_dataframe_random_data,
+    multiindex_dataframe_random_data,
+):
+    ymd = multiindex_year_month_day_dataframe_random_data
+    frame = multiindex_dataframe_random_data
+
+    def _test_roundtrip(frame):
+        unpickled = tm.round_trip_pickle(frame)
+        tm.assert_frame_equal(frame, unpickled)
+
+    _test_roundtrip(frame)
+    _test_roundtrip(frame.T)
+    _test_roundtrip(ymd)
+    _test_roundtrip(ymd.T)
+
+
+def test_pickle_timeseries_periodindex():
+    # GH#2891
+    prng = period_range("1/1/2011", "1/1/2012", freq="M")
+    ts = Series(np.random.default_rng(2).standard_normal(len(prng)), prng)
+    new_ts = tm.round_trip_pickle(ts)
+    assert new_ts.index.freqstr == "M"
+
+
+@pytest.mark.parametrize(
+    "name", [777, 777.0, "name", datetime.datetime(2001, 11, 11), (1, 2)]
+)
+def test_pickle_preserve_name(name):
+    unpickled = tm.round_trip_pickle(Series(np.arange(10, dtype=np.float64), name=name))
+    assert unpickled.name == name
+
+
+def test_pickle_datetimes(datetime_series):
+    unp_ts = tm.round_trip_pickle(datetime_series)
+    tm.assert_series_equal(unp_ts, datetime_series)
+
+
+def test_pickle_strings(string_series):
+    unp_series = tm.round_trip_pickle(string_series)
+    tm.assert_series_equal(unp_series, string_series)
+
+
+def test_pickle_preserves_block_ndim():
+    # GH#37631
+    ser = Series(list("abc")).astype("category").iloc[[0]]
+    res = tm.round_trip_pickle(ser)
+
+    assert res._mgr.blocks[0].ndim == 1
+    assert res._mgr.blocks[0].shape == (1,)
+
+    # GH#37631 OP issue was about indexing, underlying problem was pickle
+    tm.assert_series_equal(res[[True]], ser)
+
+
+@pytest.mark.parametrize("protocol", [pickle.DEFAULT_PROTOCOL, pickle.HIGHEST_PROTOCOL])
+def test_pickle_big_dataframe_compression(protocol, compression):
+    # GH#39002
+    df = DataFrame(range(100000))
+    result = tm.round_trip_pathlib(
+        partial(df.to_pickle, protocol=protocol, compression=compression),
+        partial(pd.read_pickle, compression=compression),
+    )
+    tm.assert_frame_equal(df, result)
+
+
+def test_pickle_frame_v124_unpickle_130(datapath):
+    # GH#42345 DataFrame created in 1.2.x, unpickle in 1.3.x
+    path = datapath(
+        Path(__file__).parent,
+        "data",
+        "legacy_pickle",
+        "1.2.4",
+        "empty_frame_v1_2_4-GH#42345.pkl",
+    )
+    with open(path, "rb") as fd:
+        df = pickle.load(fd)
+
+    expected = DataFrame(index=[], columns=[])
+    tm.assert_frame_equal(df, expected)
diff --git a/pandas/pandas/tests/io/test_s3.py b/pandas/pandas/tests/io/test_s3.py
new file mode 100644
index 0000000000000..79473895b662d
--- /dev/null
+++ b/pandas/pandas/tests/io/test_s3.py
@@ -0,0 +1,43 @@
+from io import BytesIO
+
+import pytest
+
+from pandas import read_csv
+
+
+def test_streaming_s3_objects():
+    # GH17135
+    # botocore gained iteration support in 1.10.47, can now be used in read_*
+    pytest.importorskip("botocore", minversion="1.10.47")
+    from botocore.response import StreamingBody
+
+    data = [b"foo,bar,baz\n1,2,3\n4,5,6\n", b"just,the,header\n"]
+    for el in data:
+        body = StreamingBody(BytesIO(el), content_length=len(el))
+        read_csv(body)
+
+
+@pytest.mark.single_cpu
+def test_read_without_creds_from_pub_bucket(s3_public_bucket_with_data, s3so):
+    # GH 34626
+    pytest.importorskip("s3fs")
+    result = read_csv(
+        f"s3://{s3_public_bucket_with_data.name}/tips.csv",
+        nrows=3,
+        storage_options=s3so,
+    )
+    assert len(result) == 3
+
+
+@pytest.mark.single_cpu
+def test_read_with_creds_from_pub_bucket(s3_public_bucket_with_data, s3so):
+    # Ensure we can read from a public bucket with credentials
+    # GH 34626
+    pytest.importorskip("s3fs")
+    df = read_csv(
+        f"s3://{s3_public_bucket_with_data.name}/tips.csv",
+        nrows=5,
+        header=None,
+        storage_options=s3so,
+    )
+    assert len(df) == 5
diff --git a/pandas/pandas/tests/io/test_spss.py b/pandas/pandas/tests/io/test_spss.py
new file mode 100644
index 0000000000000..950f74a686b8d
--- /dev/null
+++ b/pandas/pandas/tests/io/test_spss.py
@@ -0,0 +1,182 @@
+import datetime
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pyreadstat = pytest.importorskip("pyreadstat")
+
+
+# TODO(CoW) - detection of chained assignment in cython
+# https://github.com/pandas-dev/pandas/issues/51315
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+@pytest.mark.parametrize("path_klass", [lambda p: p, Path])
+def test_spss_labelled_num(path_klass, datapath):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = path_klass(datapath("io", "data", "spss", "labelled-num.sav"))
+
+    df = pd.read_spss(fname, convert_categoricals=True)
+    expected = pd.DataFrame({"VAR00002": "This is one"}, index=[0])
+    expected["VAR00002"] = pd.Categorical(expected["VAR00002"])
+    tm.assert_frame_equal(df, expected)
+
+    df = pd.read_spss(fname, convert_categoricals=False)
+    expected = pd.DataFrame({"VAR00002": 1.0}, index=[0])
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+def test_spss_labelled_num_na(datapath):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = datapath("io", "data", "spss", "labelled-num-na.sav")
+
+    df = pd.read_spss(fname, convert_categoricals=True)
+    expected = pd.DataFrame({"VAR00002": ["This is one", None]})
+    expected["VAR00002"] = pd.Categorical(expected["VAR00002"])
+    tm.assert_frame_equal(df, expected)
+
+    df = pd.read_spss(fname, convert_categoricals=False)
+    expected = pd.DataFrame({"VAR00002": [1.0, np.nan]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+def test_spss_labelled_str(datapath):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = datapath("io", "data", "spss", "labelled-str.sav")
+
+    df = pd.read_spss(fname, convert_categoricals=True)
+    expected = pd.DataFrame({"gender": ["Male", "Female"]})
+    expected["gender"] = pd.Categorical(expected["gender"])
+    tm.assert_frame_equal(df, expected)
+
+    df = pd.read_spss(fname, convert_categoricals=False)
+    expected = pd.DataFrame({"gender": ["M", "F"]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+def test_spss_kwargs(datapath):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = datapath("io", "data", "spss", "labelled-str.sav")
+
+    df = pd.read_spss(fname, convert_categoricals=True, row_limit=1)
+    expected = pd.DataFrame({"gender": ["Male"]}, dtype="category")
+    tm.assert_frame_equal(df, expected)
+
+    df = pd.read_spss(fname, convert_categoricals=False, row_offset=1)
+    expected = pd.DataFrame({"gender": ["F"]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+def test_spss_umlauts(datapath):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = datapath("io", "data", "spss", "umlauts.sav")
+
+    df = pd.read_spss(fname, convert_categoricals=True)
+    expected = pd.DataFrame(
+        {"var1": ["the ä umlaut", "the ü umlaut", "the ä umlaut", "the ö umlaut"]}
+    )
+    expected["var1"] = pd.Categorical(expected["var1"])
+    tm.assert_frame_equal(df, expected)
+
+    df = pd.read_spss(fname, convert_categoricals=False)
+    expected = pd.DataFrame({"var1": [1.0, 2.0, 1.0, 3.0]})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_spss_usecols(datapath):
+    # usecols must be list-like
+    fname = datapath("io", "data", "spss", "labelled-num.sav")
+
+    with pytest.raises(TypeError, match="usecols must be list-like."):
+        pd.read_spss(fname, usecols="VAR00002")
+
+
+def test_spss_umlauts_dtype_backend(datapath, dtype_backend):
+    # test file from the Haven project (https://haven.tidyverse.org/)
+    # Licence at LICENSES/HAVEN_LICENSE, LICENSES/HAVEN_MIT
+    fname = datapath("io", "data", "spss", "umlauts.sav")
+
+    df = pd.read_spss(fname, convert_categoricals=False, dtype_backend=dtype_backend)
+    expected = pd.DataFrame({"var1": [1.0, 2.0, 1.0, 3.0]}, dtype="Int64")
+
+    if dtype_backend == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+
+        from pandas.arrays import ArrowExtensionArray
+
+        expected = pd.DataFrame(
+            {
+                col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                for col in expected.columns
+            }
+        )
+
+    tm.assert_frame_equal(df, expected)
+
+
+def test_invalid_dtype_backend():
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        pd.read_spss("test", dtype_backend="numpy")
+
+
+@pytest.mark.filterwarnings("ignore::pandas.errors.ChainedAssignmentError")
+@pytest.mark.filterwarnings("ignore:ChainedAssignmentError:FutureWarning")
+def test_spss_metadata(datapath):
+    # GH 54264
+    fname = datapath("io", "data", "spss", "labelled-num.sav")
+
+    df = pd.read_spss(fname)
+    metadata = {
+        "column_names": ["VAR00002"],
+        "column_labels": [None],
+        "column_names_to_labels": {"VAR00002": None},
+        "file_encoding": "UTF-8",
+        "number_columns": 1,
+        "number_rows": 1,
+        "variable_value_labels": {"VAR00002": {1.0: "This is one"}},
+        "value_labels": {"labels0": {1.0: "This is one"}},
+        "variable_to_label": {"VAR00002": "labels0"},
+        "notes": [],
+        "original_variable_types": {"VAR00002": "F8.0"},
+        "readstat_variable_types": {"VAR00002": "double"},
+        "table_name": None,
+        "missing_ranges": {},
+        "missing_user_values": {},
+        "variable_storage_width": {"VAR00002": 8},
+        "variable_display_width": {"VAR00002": 8},
+        "variable_alignment": {"VAR00002": "unknown"},
+        "variable_measure": {"VAR00002": "unknown"},
+        "file_label": None,
+        "file_format": "sav/zsav",
+    }
+    if Version(pyreadstat.__version__) >= Version("1.2.4"):
+        metadata.update(
+            {
+                "creation_time": datetime.datetime(2015, 2, 6, 14, 33, 36),
+                "modification_time": datetime.datetime(2015, 2, 6, 14, 33, 36),
+            }
+        )
+    if Version(pyreadstat.__version__) >= Version("1.2.8"):
+        metadata["mr_sets"] = {}
+    tm.assert_dict_equal(df.attrs, metadata)
diff --git a/pandas/pandas/tests/io/test_sql.py b/pandas/pandas/tests/io/test_sql.py
new file mode 100644
index 0000000000000..13576c891ad2c
--- /dev/null
+++ b/pandas/pandas/tests/io/test_sql.py
@@ -0,0 +1,4414 @@
+from __future__ import annotations
+
+import contextlib
+from contextlib import closing
+import csv
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+from io import StringIO
+from pathlib import Path
+import sqlite3
+from typing import TYPE_CHECKING
+import uuid
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas._libs import lib
+from pandas.compat import pa_version_under14p1
+from pandas.compat._optional import import_optional_dependency
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    isna,
+    to_datetime,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+from pandas.io import sql
+from pandas.io.sql import (
+    SQLAlchemyEngine,
+    SQLDatabase,
+    SQLiteDatabase,
+    get_engine,
+    pandasSQL_builder,
+    read_sql_query,
+    read_sql_table,
+)
+
+if TYPE_CHECKING:
+    import sqlalchemy
+
+
+pytestmark = [
+    pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
+    ),
+    pytest.mark.single_cpu,
+]
+
+
+@pytest.fixture
+def sql_strings():
+    return {
+        "read_parameters": {
+            "sqlite": "SELECT * FROM iris WHERE Name=? AND SepalLength=?",
+            "mysql": "SELECT * FROM iris WHERE `Name`=%s AND `SepalLength`=%s",
+            "postgresql": 'SELECT * FROM iris WHERE "Name"=%s AND "SepalLength"=%s',
+        },
+        "read_named_parameters": {
+            "sqlite": """
+                SELECT * FROM iris WHERE Name=:name AND SepalLength=:length
+                """,
+            "mysql": """
+                SELECT * FROM iris WHERE
+                `Name`=%(name)s AND `SepalLength`=%(length)s
+                """,
+            "postgresql": """
+                SELECT * FROM iris WHERE
+                "Name"=%(name)s AND "SepalLength"=%(length)s
+                """,
+        },
+        "read_no_parameters_with_percent": {
+            "sqlite": "SELECT * FROM iris WHERE Name LIKE '%'",
+            "mysql": "SELECT * FROM iris WHERE `Name` LIKE '%'",
+            "postgresql": "SELECT * FROM iris WHERE \"Name\" LIKE '%'",
+        },
+    }
+
+
+def iris_table_metadata():
+    import sqlalchemy
+    from sqlalchemy import (
+        Column,
+        Double,
+        Float,
+        MetaData,
+        String,
+        Table,
+    )
+
+    dtype = Double if Version(sqlalchemy.__version__) >= Version("2.0.0") else Float
+    metadata = MetaData()
+    iris = Table(
+        "iris",
+        metadata,
+        Column("SepalLength", dtype),
+        Column("SepalWidth", dtype),
+        Column("PetalLength", dtype),
+        Column("PetalWidth", dtype),
+        Column("Name", String(200)),
+    )
+    return iris
+
+
+def create_and_load_iris_sqlite3(conn, iris_file: Path):
+    stmt = """CREATE TABLE iris (
+            "SepalLength" REAL,
+            "SepalWidth" REAL,
+            "PetalLength" REAL,
+            "PetalWidth" REAL,
+            "Name" TEXT
+        )"""
+
+    cur = conn.cursor()
+    cur.execute(stmt)
+    with iris_file.open(newline=None, encoding="utf-8") as csvfile:
+        reader = csv.reader(csvfile)
+        next(reader)
+        stmt = "INSERT INTO iris VALUES(?, ?, ?, ?, ?)"
+        # ADBC requires explicit types - no implicit str -> float conversion
+        records = []
+        records = [
+            (
+                float(row[0]),
+                float(row[1]),
+                float(row[2]),
+                float(row[3]),
+                row[4],
+            )
+            for row in reader
+        ]
+
+        cur.executemany(stmt, records)
+    cur.close()
+
+    conn.commit()
+
+
+def create_and_load_iris_postgresql(conn, iris_file: Path):
+    stmt = """CREATE TABLE iris (
+            "SepalLength" DOUBLE PRECISION,
+            "SepalWidth" DOUBLE PRECISION,
+            "PetalLength" DOUBLE PRECISION,
+            "PetalWidth" DOUBLE PRECISION,
+            "Name" TEXT
+        )"""
+    with conn.cursor() as cur:
+        cur.execute(stmt)
+        with iris_file.open(newline=None, encoding="utf-8") as csvfile:
+            reader = csv.reader(csvfile)
+            next(reader)
+            stmt = "INSERT INTO iris VALUES($1, $2, $3, $4, $5)"
+            # ADBC requires explicit types - no implicit str -> float conversion
+            records = [
+                (
+                    float(row[0]),
+                    float(row[1]),
+                    float(row[2]),
+                    float(row[3]),
+                    row[4],
+                )
+                for row in reader
+            ]
+
+            cur.executemany(stmt, records)
+
+    conn.commit()
+
+
+def create_and_load_iris(conn, iris_file: Path):
+    from sqlalchemy import insert
+
+    iris = iris_table_metadata()
+
+    with iris_file.open(newline=None, encoding="utf-8") as csvfile:
+        reader = csv.reader(csvfile)
+        header = next(reader)
+        params = [dict(zip(header, row)) for row in reader]
+        stmt = insert(iris).values(params)
+        with conn.begin() as con:
+            iris.drop(con, checkfirst=True)
+            iris.create(bind=con)
+            con.execute(stmt)
+
+
+def create_and_load_iris_view(conn):
+    stmt = "CREATE VIEW iris_view AS SELECT * FROM iris"
+    if isinstance(conn, sqlite3.Connection):
+        cur = conn.cursor()
+        cur.execute(stmt)
+    else:
+        adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+        if adbc and isinstance(conn, adbc.Connection):
+            with conn.cursor() as cur:
+                cur.execute(stmt)
+            conn.commit()
+        else:
+            from sqlalchemy import text
+
+            stmt = text(stmt)
+            with conn.begin() as con:
+                con.execute(stmt)
+
+
+def types_table_metadata(dialect: str):
+    from sqlalchemy import (
+        TEXT,
+        Boolean,
+        Column,
+        DateTime,
+        Float,
+        Integer,
+        MetaData,
+        Table,
+    )
+
+    date_type = TEXT if dialect == "sqlite" else DateTime
+    bool_type = Integer if dialect == "sqlite" else Boolean
+    metadata = MetaData()
+    types = Table(
+        "types",
+        metadata,
+        Column("TextCol", TEXT),
+        # error: Cannot infer type argument 1 of "Column"
+        Column("DateCol", date_type),  # type: ignore[misc]
+        Column("IntDateCol", Integer),
+        Column("IntDateOnlyCol", Integer),
+        Column("FloatCol", Float),
+        Column("IntCol", Integer),
+        # error: Cannot infer type argument 1 of "Column"
+        Column("BoolCol", bool_type),  # type: ignore[misc]
+        Column("IntColWithNull", Integer),
+        # error: Cannot infer type argument 1 of "Column"
+        Column("BoolColWithNull", bool_type),  # type: ignore[misc]
+    )
+    return types
+
+
+def create_and_load_types_sqlite3(conn, types_data: list[dict]):
+    stmt = """CREATE TABLE types (
+                    "TextCol" TEXT,
+                    "DateCol" TEXT,
+                    "IntDateCol" INTEGER,
+                    "IntDateOnlyCol" INTEGER,
+                    "FloatCol" REAL,
+                    "IntCol" INTEGER,
+                    "BoolCol" INTEGER,
+                    "IntColWithNull" INTEGER,
+                    "BoolColWithNull" INTEGER
+                )"""
+
+    ins_stmt = """
+                INSERT INTO types
+                VALUES(?, ?, ?, ?, ?, ?, ?, ?, ?)
+                """
+
+    if isinstance(conn, sqlite3.Connection):
+        cur = conn.cursor()
+        cur.execute(stmt)
+        cur.executemany(ins_stmt, types_data)
+    else:
+        with conn.cursor() as cur:
+            cur.execute(stmt)
+            cur.executemany(ins_stmt, types_data)
+
+        conn.commit()
+
+
+def create_and_load_types_postgresql(conn, types_data: list[dict]):
+    with conn.cursor() as cur:
+        stmt = """CREATE TABLE types (
+                        "TextCol" TEXT,
+                        "DateCol" TIMESTAMP,
+                        "IntDateCol" INTEGER,
+                        "IntDateOnlyCol" INTEGER,
+                        "FloatCol" DOUBLE PRECISION,
+                        "IntCol" INTEGER,
+                        "BoolCol" BOOLEAN,
+                        "IntColWithNull" INTEGER,
+                        "BoolColWithNull" BOOLEAN
+                    )"""
+        cur.execute(stmt)
+
+        stmt = """
+                INSERT INTO types
+                VALUES($1, $2::timestamp, $3, $4, $5, $6, $7, $8, $9)
+                """
+
+        cur.executemany(stmt, types_data)
+
+    conn.commit()
+
+
+def create_and_load_types(conn, types_data: list[dict], dialect: str):
+    from sqlalchemy import insert
+    from sqlalchemy.engine import Engine
+
+    types = types_table_metadata(dialect)
+
+    stmt = insert(types).values(types_data)
+    if isinstance(conn, Engine):
+        with conn.connect() as conn:
+            with conn.begin():
+                types.drop(conn, checkfirst=True)
+                types.create(bind=conn)
+                conn.execute(stmt)
+    else:
+        with conn.begin():
+            types.drop(conn, checkfirst=True)
+            types.create(bind=conn)
+            conn.execute(stmt)
+
+
+def create_and_load_postgres_datetz(conn):
+    from sqlalchemy import (
+        Column,
+        DateTime,
+        MetaData,
+        Table,
+        insert,
+    )
+    from sqlalchemy.engine import Engine
+
+    metadata = MetaData()
+    datetz = Table("datetz", metadata, Column("DateColWithTz", DateTime(timezone=True)))
+    datetz_data = [
+        {
+            "DateColWithTz": "2000-01-01 00:00:00-08:00",
+        },
+        {
+            "DateColWithTz": "2000-06-01 00:00:00-07:00",
+        },
+    ]
+    stmt = insert(datetz).values(datetz_data)
+    if isinstance(conn, Engine):
+        with conn.connect() as conn:
+            with conn.begin():
+                datetz.drop(conn, checkfirst=True)
+                datetz.create(bind=conn)
+                conn.execute(stmt)
+    else:
+        with conn.begin():
+            datetz.drop(conn, checkfirst=True)
+            datetz.create(bind=conn)
+            conn.execute(stmt)
+
+    # "2000-01-01 00:00:00-08:00" should convert to
+    # "2000-01-01 08:00:00"
+    # "2000-06-01 00:00:00-07:00" should convert to
+    # "2000-06-01 07:00:00"
+    # GH 6415
+    expected_data = [
+        Timestamp("2000-01-01 08:00:00", tz="UTC"),
+        Timestamp("2000-06-01 07:00:00", tz="UTC"),
+    ]
+    return Series(expected_data, name="DateColWithTz").astype("M8[us, UTC]")
+
+
+def check_iris_frame(frame: DataFrame):
+    pytype = frame.dtypes.iloc[0].type
+    row = frame.iloc[0]
+    assert issubclass(pytype, np.floating)
+    tm.assert_series_equal(
+        row, Series([5.1, 3.5, 1.4, 0.2, "Iris-setosa"], index=frame.columns, name=0)
+    )
+    assert frame.shape in ((150, 5), (8, 5))
+
+
+def count_rows(conn, table_name: str):
+    stmt = f"SELECT count(*) AS count_1 FROM {table_name}"
+    adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+    if isinstance(conn, sqlite3.Connection):
+        cur = conn.cursor()
+        return cur.execute(stmt).fetchone()[0]
+    elif adbc and isinstance(conn, adbc.Connection):
+        with conn.cursor() as cur:
+            cur.execute(stmt)
+            return cur.fetchone()[0]
+    else:
+        from sqlalchemy import create_engine
+        from sqlalchemy.engine import Engine
+
+        if isinstance(conn, str):
+            try:
+                engine = create_engine(conn)
+                with engine.connect() as conn:
+                    return conn.exec_driver_sql(stmt).scalar_one()
+            finally:
+                engine.dispose()
+        elif isinstance(conn, Engine):
+            with conn.connect() as conn:
+                return conn.exec_driver_sql(stmt).scalar_one()
+        else:
+            return conn.exec_driver_sql(stmt).scalar_one()
+
+
+@pytest.fixture
+def iris_path(datapath):
+    iris_path = datapath("io", "data", "csv", "iris.csv")
+    return Path(iris_path)
+
+
+@pytest.fixture
+def types_data():
+    return [
+        {
+            "TextCol": "first",
+            "DateCol": "2000-01-03 00:00:00",
+            "IntDateCol": 535852800,
+            "IntDateOnlyCol": 20101010,
+            "FloatCol": 10.10,
+            "IntCol": 1,
+            "BoolCol": False,
+            "IntColWithNull": 1,
+            "BoolColWithNull": False,
+        },
+        {
+            "TextCol": "first",
+            "DateCol": "2000-01-04 00:00:00",
+            "IntDateCol": 1356998400,
+            "IntDateOnlyCol": 20101212,
+            "FloatCol": 10.10,
+            "IntCol": 1,
+            "BoolCol": False,
+            "IntColWithNull": None,
+            "BoolColWithNull": None,
+        },
+    ]
+
+
+@pytest.fixture
+def types_data_frame(types_data):
+    dtypes = {
+        "TextCol": "str",
+        "DateCol": "str",
+        "IntDateCol": "int64",
+        "IntDateOnlyCol": "int64",
+        "FloatCol": "float",
+        "IntCol": "int64",
+        "BoolCol": "int64",
+        "IntColWithNull": "float",
+        "BoolColWithNull": "float",
+    }
+    df = DataFrame(types_data)
+    return df[dtypes.keys()].astype(dtypes)
+
+
+@pytest.fixture
+def test_frame1():
+    columns = ["index", "A", "B", "C", "D"]
+    data = [
+        (
+            "2000-01-03 00:00:00",
+            0.980268513777,
+            3.68573087906,
+            -0.364216805298,
+            -1.15973806169,
+        ),
+        (
+            "2000-01-04 00:00:00",
+            1.04791624281,
+            -0.0412318367011,
+            -0.16181208307,
+            0.212549316967,
+        ),
+        (
+            "2000-01-05 00:00:00",
+            0.498580885705,
+            0.731167677815,
+            -0.537677223318,
+            1.34627041952,
+        ),
+        (
+            "2000-01-06 00:00:00",
+            1.12020151869,
+            1.56762092543,
+            0.00364077397681,
+            0.67525259227,
+        ),
+    ]
+    return DataFrame(data, columns=columns)
+
+
+@pytest.fixture
+def test_frame3():
+    columns = ["index", "A", "B"]
+    data = [
+        ("2000-01-03 00:00:00", 2**31 - 1, -1.987670),
+        ("2000-01-04 00:00:00", -29, -0.0412318367011),
+        ("2000-01-05 00:00:00", 20000, 0.731167677815),
+        ("2000-01-06 00:00:00", -290867, 1.56762092543),
+    ]
+    return DataFrame(data, columns=columns)
+
+
+def get_all_views(conn):
+    if isinstance(conn, sqlite3.Connection):
+        c = conn.execute("SELECT name FROM sqlite_master WHERE type='view'")
+        return [view[0] for view in c.fetchall()]
+    else:
+        adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+        if adbc and isinstance(conn, adbc.Connection):
+            results = []
+            info = conn.adbc_get_objects().read_all().to_pylist()
+            for catalog in info:
+                catalog["catalog_name"]
+                for schema in catalog["catalog_db_schemas"]:
+                    schema["db_schema_name"]
+                    for table in schema["db_schema_tables"]:
+                        if table["table_type"] == "view":
+                            view_name = table["table_name"]
+                            results.append(view_name)
+
+            return results
+        else:
+            from sqlalchemy import inspect
+
+            return inspect(conn).get_view_names()
+
+
+def get_all_tables(conn):
+    if isinstance(conn, sqlite3.Connection):
+        c = conn.execute("SELECT name FROM sqlite_master WHERE type='table'")
+        return [table[0] for table in c.fetchall()]
+    else:
+        adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+
+        if adbc and isinstance(conn, adbc.Connection):
+            results = []
+            info = conn.adbc_get_objects().read_all().to_pylist()
+            for catalog in info:
+                for schema in catalog["catalog_db_schemas"]:
+                    for table in schema["db_schema_tables"]:
+                        if table["table_type"] == "table":
+                            table_name = table["table_name"]
+                            results.append(table_name)
+
+            return results
+        else:
+            from sqlalchemy import inspect
+
+            return inspect(conn).get_table_names()
+
+
+def drop_table(
+    table_name: str,
+    conn: sqlite3.Connection | sqlalchemy.engine.Engine | sqlalchemy.engine.Connection,
+):
+    if isinstance(conn, sqlite3.Connection):
+        conn.execute(f"DROP TABLE IF EXISTS {sql._get_valid_sqlite_name(table_name)}")
+        conn.commit()
+
+    else:
+        adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+        if adbc and isinstance(conn, adbc.Connection):
+            with conn.cursor() as cur:
+                cur.execute(f'DROP TABLE IF EXISTS "{table_name}"')
+        else:
+            with conn.begin() as con:
+                with sql.SQLDatabase(con) as db:
+                    db.drop_table(table_name)
+
+
+def drop_view(
+    view_name: str,
+    conn: sqlite3.Connection | sqlalchemy.engine.Engine | sqlalchemy.engine.Connection,
+):
+    import sqlalchemy
+
+    if isinstance(conn, sqlite3.Connection):
+        conn.execute(f"DROP VIEW IF EXISTS {sql._get_valid_sqlite_name(view_name)}")
+        conn.commit()
+    else:
+        adbc = import_optional_dependency("adbc_driver_manager.dbapi", errors="ignore")
+        if adbc and isinstance(conn, adbc.Connection):
+            with conn.cursor() as cur:
+                cur.execute(f'DROP VIEW IF EXISTS "{view_name}"')
+        else:
+            quoted_view = conn.engine.dialect.identifier_preparer.quote_identifier(
+                view_name
+            )
+            stmt = sqlalchemy.text(f"DROP VIEW IF EXISTS {quoted_view}")
+            with conn.begin() as con:
+                con.execute(stmt)  # type: ignore[union-attr]
+
+
+@pytest.fixture
+def mysql_pymysql_engine():
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+    pymysql = pytest.importorskip("pymysql")
+    engine = sqlalchemy.create_engine(
+        "mysql+pymysql://root@localhost:3306/pandas",
+        connect_args={"client_flag": pymysql.constants.CLIENT.MULTI_STATEMENTS},
+        poolclass=sqlalchemy.pool.NullPool,
+    )
+    yield engine
+    for view in get_all_views(engine):
+        drop_view(view, engine)
+    for tbl in get_all_tables(engine):
+        drop_table(tbl, engine)
+    engine.dispose()
+
+
+@pytest.fixture
+def mysql_pymysql_engine_iris(mysql_pymysql_engine, iris_path):
+    create_and_load_iris(mysql_pymysql_engine, iris_path)
+    create_and_load_iris_view(mysql_pymysql_engine)
+    return mysql_pymysql_engine
+
+
+@pytest.fixture
+def mysql_pymysql_engine_types(mysql_pymysql_engine, types_data):
+    create_and_load_types(mysql_pymysql_engine, types_data, "mysql")
+    return mysql_pymysql_engine
+
+
+@pytest.fixture
+def mysql_pymysql_conn(mysql_pymysql_engine):
+    with mysql_pymysql_engine.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def mysql_pymysql_conn_iris(mysql_pymysql_engine_iris):
+    with mysql_pymysql_engine_iris.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def mysql_pymysql_conn_types(mysql_pymysql_engine_types):
+    with mysql_pymysql_engine_types.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def postgresql_psycopg2_engine():
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+    pytest.importorskip("psycopg2")
+    engine = sqlalchemy.create_engine(
+        "postgresql+psycopg2://postgres:postgres@localhost:5432/pandas",
+        poolclass=sqlalchemy.pool.NullPool,
+    )
+    yield engine
+    for view in get_all_views(engine):
+        drop_view(view, engine)
+    for tbl in get_all_tables(engine):
+        drop_table(tbl, engine)
+    engine.dispose()
+
+
+@pytest.fixture
+def postgresql_psycopg2_engine_iris(postgresql_psycopg2_engine, iris_path):
+    create_and_load_iris(postgresql_psycopg2_engine, iris_path)
+    create_and_load_iris_view(postgresql_psycopg2_engine)
+    return postgresql_psycopg2_engine
+
+
+@pytest.fixture
+def postgresql_psycopg2_engine_types(postgresql_psycopg2_engine, types_data):
+    create_and_load_types(postgresql_psycopg2_engine, types_data, "postgres")
+    return postgresql_psycopg2_engine
+
+
+@pytest.fixture
+def postgresql_psycopg2_conn(postgresql_psycopg2_engine):
+    with postgresql_psycopg2_engine.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def postgresql_adbc_conn():
+    pytest.importorskip("pyarrow")
+    pytest.importorskip("adbc_driver_postgresql")
+    from adbc_driver_postgresql import dbapi
+
+    uri = "postgresql://postgres:postgres@localhost:5432/pandas"
+    with dbapi.connect(uri) as conn:
+        yield conn
+        for view in get_all_views(conn):
+            drop_view(view, conn)
+        for tbl in get_all_tables(conn):
+            drop_table(tbl, conn)
+        conn.commit()
+
+
+@pytest.fixture
+def postgresql_adbc_iris(postgresql_adbc_conn, iris_path):
+    import adbc_driver_manager as mgr
+
+    conn = postgresql_adbc_conn
+
+    try:
+        conn.adbc_get_table_schema("iris")
+    except mgr.ProgrammingError:
+        conn.rollback()
+        create_and_load_iris_postgresql(conn, iris_path)
+    try:
+        conn.adbc_get_table_schema("iris_view")
+    except mgr.ProgrammingError:  # note arrow-adbc issue 1022
+        conn.rollback()
+        create_and_load_iris_view(conn)
+    return conn
+
+
+@pytest.fixture
+def postgresql_adbc_types(postgresql_adbc_conn, types_data):
+    import adbc_driver_manager as mgr
+
+    conn = postgresql_adbc_conn
+
+    try:
+        conn.adbc_get_table_schema("types")
+    except mgr.ProgrammingError:
+        conn.rollback()
+        new_data = [tuple(entry.values()) for entry in types_data]
+
+        create_and_load_types_postgresql(conn, new_data)
+
+    return conn
+
+
+@pytest.fixture
+def postgresql_psycopg2_conn_iris(postgresql_psycopg2_engine_iris):
+    with postgresql_psycopg2_engine_iris.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def postgresql_psycopg2_conn_types(postgresql_psycopg2_engine_types):
+    with postgresql_psycopg2_engine_types.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def sqlite_str():
+    pytest.importorskip("sqlalchemy")
+    with tm.ensure_clean() as name:
+        yield f"sqlite:///{name}"
+
+
+@pytest.fixture
+def sqlite_engine(sqlite_str):
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+    engine = sqlalchemy.create_engine(sqlite_str, poolclass=sqlalchemy.pool.NullPool)
+    yield engine
+    for view in get_all_views(engine):
+        drop_view(view, engine)
+    for tbl in get_all_tables(engine):
+        drop_table(tbl, engine)
+    engine.dispose()
+
+
+@pytest.fixture
+def sqlite_conn(sqlite_engine):
+    with sqlite_engine.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def sqlite_str_iris(sqlite_str, iris_path):
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+    engine = sqlalchemy.create_engine(sqlite_str)
+    create_and_load_iris(engine, iris_path)
+    create_and_load_iris_view(engine)
+    engine.dispose()
+    return sqlite_str
+
+
+@pytest.fixture
+def sqlite_engine_iris(sqlite_engine, iris_path):
+    create_and_load_iris(sqlite_engine, iris_path)
+    create_and_load_iris_view(sqlite_engine)
+    return sqlite_engine
+
+
+@pytest.fixture
+def sqlite_conn_iris(sqlite_engine_iris):
+    with sqlite_engine_iris.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def sqlite_str_types(sqlite_str, types_data):
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+    engine = sqlalchemy.create_engine(sqlite_str)
+    create_and_load_types(engine, types_data, "sqlite")
+    engine.dispose()
+    return sqlite_str
+
+
+@pytest.fixture
+def sqlite_engine_types(sqlite_engine, types_data):
+    create_and_load_types(sqlite_engine, types_data, "sqlite")
+    return sqlite_engine
+
+
+@pytest.fixture
+def sqlite_conn_types(sqlite_engine_types):
+    with sqlite_engine_types.connect() as conn:
+        yield conn
+
+
+@pytest.fixture
+def sqlite_adbc_conn():
+    pytest.importorskip("pyarrow")
+    pytest.importorskip("adbc_driver_sqlite")
+    from adbc_driver_sqlite import dbapi
+
+    with tm.ensure_clean() as name:
+        uri = f"file:{name}"
+        with dbapi.connect(uri) as conn:
+            yield conn
+            for view in get_all_views(conn):
+                drop_view(view, conn)
+            for tbl in get_all_tables(conn):
+                drop_table(tbl, conn)
+            conn.commit()
+
+
+@pytest.fixture
+def sqlite_adbc_iris(sqlite_adbc_conn, iris_path):
+    import adbc_driver_manager as mgr
+
+    conn = sqlite_adbc_conn
+    try:
+        conn.adbc_get_table_schema("iris")
+    except mgr.ProgrammingError:
+        conn.rollback()
+        create_and_load_iris_sqlite3(conn, iris_path)
+    try:
+        conn.adbc_get_table_schema("iris_view")
+    except mgr.ProgrammingError:
+        conn.rollback()
+        create_and_load_iris_view(conn)
+    return conn
+
+
+@pytest.fixture
+def sqlite_adbc_types(sqlite_adbc_conn, types_data):
+    import adbc_driver_manager as mgr
+
+    conn = sqlite_adbc_conn
+    try:
+        conn.adbc_get_table_schema("types")
+    except mgr.ProgrammingError:
+        conn.rollback()
+        new_data = []
+        for entry in types_data:
+            entry["BoolCol"] = int(entry["BoolCol"])
+            if entry["BoolColWithNull"] is not None:
+                entry["BoolColWithNull"] = int(entry["BoolColWithNull"])
+            new_data.append(tuple(entry.values()))
+
+        create_and_load_types_sqlite3(conn, new_data)
+        conn.commit()
+
+    return conn
+
+
+@pytest.fixture
+def sqlite_buildin():
+    with contextlib.closing(sqlite3.connect(":memory:")) as closing_conn:
+        with closing_conn as conn:
+            yield conn
+
+
+@pytest.fixture
+def sqlite_buildin_iris(sqlite_buildin, iris_path):
+    create_and_load_iris_sqlite3(sqlite_buildin, iris_path)
+    create_and_load_iris_view(sqlite_buildin)
+    return sqlite_buildin
+
+
+@pytest.fixture
+def sqlite_buildin_types(sqlite_buildin, types_data):
+    types_data = [tuple(entry.values()) for entry in types_data]
+    create_and_load_types_sqlite3(sqlite_buildin, types_data)
+    return sqlite_buildin
+
+
+mysql_connectable = [
+    pytest.param("mysql_pymysql_engine", marks=pytest.mark.db),
+    pytest.param("mysql_pymysql_conn", marks=pytest.mark.db),
+]
+
+mysql_connectable_iris = [
+    pytest.param("mysql_pymysql_engine_iris", marks=pytest.mark.db),
+    pytest.param("mysql_pymysql_conn_iris", marks=pytest.mark.db),
+]
+
+mysql_connectable_types = [
+    pytest.param("mysql_pymysql_engine_types", marks=pytest.mark.db),
+    pytest.param("mysql_pymysql_conn_types", marks=pytest.mark.db),
+]
+
+postgresql_connectable = [
+    pytest.param("postgresql_psycopg2_engine", marks=pytest.mark.db),
+    pytest.param("postgresql_psycopg2_conn", marks=pytest.mark.db),
+]
+
+postgresql_connectable_iris = [
+    pytest.param("postgresql_psycopg2_engine_iris", marks=pytest.mark.db),
+    pytest.param("postgresql_psycopg2_conn_iris", marks=pytest.mark.db),
+]
+
+postgresql_connectable_types = [
+    pytest.param("postgresql_psycopg2_engine_types", marks=pytest.mark.db),
+    pytest.param("postgresql_psycopg2_conn_types", marks=pytest.mark.db),
+]
+
+sqlite_connectable = [
+    "sqlite_engine",
+    "sqlite_conn",
+    "sqlite_str",
+]
+
+sqlite_connectable_iris = [
+    "sqlite_engine_iris",
+    "sqlite_conn_iris",
+    "sqlite_str_iris",
+]
+
+sqlite_connectable_types = [
+    "sqlite_engine_types",
+    "sqlite_conn_types",
+    "sqlite_str_types",
+]
+
+sqlalchemy_connectable = mysql_connectable + postgresql_connectable + sqlite_connectable
+
+sqlalchemy_connectable_iris = (
+    mysql_connectable_iris + postgresql_connectable_iris + sqlite_connectable_iris
+)
+
+sqlalchemy_connectable_types = (
+    mysql_connectable_types + postgresql_connectable_types + sqlite_connectable_types
+)
+
+adbc_connectable = [
+    "sqlite_adbc_conn",
+    pytest.param("postgresql_adbc_conn", marks=pytest.mark.db),
+]
+
+adbc_connectable_iris = [
+    pytest.param("postgresql_adbc_iris", marks=pytest.mark.db),
+    "sqlite_adbc_iris",
+]
+
+adbc_connectable_types = [
+    pytest.param("postgresql_adbc_types", marks=pytest.mark.db),
+    "sqlite_adbc_types",
+]
+
+
+all_connectable = sqlalchemy_connectable + ["sqlite_buildin"] + adbc_connectable
+
+all_connectable_iris = (
+    sqlalchemy_connectable_iris + ["sqlite_buildin_iris"] + adbc_connectable_iris
+)
+
+all_connectable_types = (
+    sqlalchemy_connectable_types + ["sqlite_buildin_types"] + adbc_connectable_types
+)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_dataframe_to_sql(conn, test_frame1, request):
+    # GH 51086 if conn is sqlite_engine
+    conn = request.getfixturevalue(conn)
+    test_frame1.to_sql(name="test", con=conn, if_exists="append", index=False)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_dataframe_to_sql_empty(conn, test_frame1, request):
+    if conn == "postgresql_adbc_conn" and not using_string_dtype():
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="postgres ADBC driver < 1.2 cannot insert index with null type",
+            )
+        )
+
+    # GH 51086 if conn is sqlite_engine
+    conn = request.getfixturevalue(conn)
+    empty_df = test_frame1.iloc[:0]
+    empty_df.to_sql(name="test", con=conn, if_exists="append", index=False)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_dataframe_to_sql_arrow_dtypes(conn, request):
+    # GH 52046
+    pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {
+            "int": pd.array([1], dtype="int8[pyarrow]"),
+            "datetime": pd.array(
+                [datetime(2023, 1, 1)], dtype="timestamp[ns][pyarrow]"
+            ),
+            "date": pd.array([date(2023, 1, 1)], dtype="date32[day][pyarrow]"),
+            "timedelta": pd.array([timedelta(1)], dtype="duration[ns][pyarrow]"),
+            "string": pd.array(["a"], dtype="string[pyarrow]"),
+        }
+    )
+
+    if "adbc" in conn:
+        if conn == "sqlite_adbc_conn":
+            df = df.drop(columns=["timedelta"])
+        if pa_version_under14p1:
+            exp_warning = DeprecationWarning
+            msg = "is_sparse is deprecated"
+        else:
+            exp_warning = None
+            msg = ""
+    else:
+        exp_warning = UserWarning
+        msg = "the 'timedelta'"
+
+    conn = request.getfixturevalue(conn)
+    with tm.assert_produces_warning(exp_warning, match=msg, check_stacklevel=False):
+        df.to_sql(name="test_arrow", con=conn, if_exists="replace", index=False)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_dataframe_to_sql_arrow_dtypes_missing(conn, request, nulls_fixture):
+    # GH 52046
+    pytest.importorskip("pyarrow")
+    df = DataFrame(
+        {
+            "datetime": pd.array(
+                [datetime(2023, 1, 1), nulls_fixture], dtype="timestamp[ns][pyarrow]"
+            ),
+        }
+    )
+    conn = request.getfixturevalue(conn)
+    df.to_sql(name="test_arrow", con=conn, if_exists="replace", index=False)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize("method", [None, "multi"])
+def test_to_sql(conn, method, test_frame1, request):
+    if method == "multi" and "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'method' not implemented for ADBC drivers", strict=True
+            )
+        )
+
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.to_sql(test_frame1, "test_frame", method=method)
+        assert pandasSQL.has_table("test_frame")
+    assert count_rows(conn, "test_frame") == len(test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize(
+    "mode, num_row_coef", [("replace", 1), ("append", 2), ("delete_rows", 1)]
+)
+def test_to_sql_exist(conn, mode, num_row_coef, test_frame1, request):
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.to_sql(test_frame1, "test_frame", if_exists="fail")
+        pandasSQL.to_sql(test_frame1, "test_frame", if_exists=mode)
+        assert pandasSQL.has_table("test_frame")
+    assert count_rows(conn, "test_frame") == num_row_coef * len(test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_to_sql_exist_fail(conn, test_frame1, request):
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.to_sql(test_frame1, "test_frame", if_exists="fail")
+        assert pandasSQL.has_table("test_frame")
+
+        msg = "Table 'test_frame' already exists"
+        with pytest.raises(ValueError, match=msg):
+            pandasSQL.to_sql(test_frame1, "test_frame", if_exists="fail")
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_iris_query(conn, request):
+    conn = request.getfixturevalue(conn)
+    iris_frame = read_sql_query("SELECT * FROM iris", conn)
+    check_iris_frame(iris_frame)
+    iris_frame = pd.read_sql("SELECT * FROM iris", conn)
+    check_iris_frame(iris_frame)
+    iris_frame = pd.read_sql("SELECT * FROM iris where 0=1", conn)
+    assert iris_frame.shape == (0, 5)
+    assert "SepalWidth" in iris_frame.columns
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_iris_query_chunksize(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'chunksize' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn = request.getfixturevalue(conn)
+    iris_frame = concat(read_sql_query("SELECT * FROM iris", conn, chunksize=7))
+    check_iris_frame(iris_frame)
+    iris_frame = concat(pd.read_sql("SELECT * FROM iris", conn, chunksize=7))
+    check_iris_frame(iris_frame)
+    iris_frame = concat(pd.read_sql("SELECT * FROM iris where 0=1", conn, chunksize=7))
+    assert iris_frame.shape == (0, 5)
+    assert "SepalWidth" in iris_frame.columns
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_read_iris_query_expression_with_parameter(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'chunksize' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn = request.getfixturevalue(conn)
+    from sqlalchemy import (
+        MetaData,
+        Table,
+        create_engine,
+        select,
+    )
+
+    metadata = MetaData()
+    autoload_con = create_engine(conn) if isinstance(conn, str) else conn
+    iris = Table("iris", metadata, autoload_with=autoload_con)
+    iris_frame = read_sql_query(
+        select(iris), conn, params={"name": "Iris-setosa", "length": 5.1}
+    )
+    check_iris_frame(iris_frame)
+    if isinstance(conn, str):
+        autoload_con.dispose()
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_iris_query_string_with_parameter(conn, request, sql_strings):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'chunksize' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+
+    for db, query in sql_strings["read_parameters"].items():
+        if db in conn:
+            break
+    else:
+        raise KeyError(f"No part of {conn} found in sql_strings['read_parameters']")
+    conn = request.getfixturevalue(conn)
+    iris_frame = read_sql_query(query, conn, params=("Iris-setosa", 5.1))
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_read_iris_table(conn, request):
+    # GH 51015 if conn = sqlite_iris_str
+    conn = request.getfixturevalue(conn)
+    iris_frame = read_sql_table("iris", conn)
+    check_iris_frame(iris_frame)
+    iris_frame = pd.read_sql("iris", conn)
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_read_iris_table_chunksize(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="chunksize argument NotImplemented with ADBC")
+        )
+    conn = request.getfixturevalue(conn)
+    iris_frame = concat(read_sql_table("iris", conn, chunksize=7))
+    check_iris_frame(iris_frame)
+    iris_frame = concat(pd.read_sql("iris", conn, chunksize=7))
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_to_sql_callable(conn, test_frame1, request):
+    conn = request.getfixturevalue(conn)
+
+    check = []  # used to double check function below is really being used
+
+    def sample(pd_table, conn, keys, data_iter):
+        check.append(1)
+        data = [dict(zip(keys, row)) for row in data_iter]
+        conn.execute(pd_table.table.insert(), data)
+
+    with pandasSQL_builder(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.to_sql(test_frame1, "test_frame", method=sample)
+        assert pandasSQL.has_table("test_frame")
+    assert check == [1]
+    assert count_rows(conn, "test_frame") == len(test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable_types)
+def test_default_type_conversion(conn, request):
+    conn_name = conn
+    if conn_name == "sqlite_buildin_types":
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="sqlite_buildin connection does not implement read_sql_table"
+            )
+        )
+
+    conn = request.getfixturevalue(conn)
+    df = sql.read_sql_table("types", conn)
+
+    assert issubclass(df.FloatCol.dtype.type, np.floating)
+    assert issubclass(df.IntCol.dtype.type, np.integer)
+
+    # MySQL/sqlite has no real BOOL type
+    if "postgresql" in conn_name:
+        assert issubclass(df.BoolCol.dtype.type, np.bool_)
+    else:
+        assert issubclass(df.BoolCol.dtype.type, np.integer)
+
+    # Int column with NA values stays as float
+    assert issubclass(df.IntColWithNull.dtype.type, np.floating)
+
+    # Bool column with NA = int column with NA values => becomes float
+    if "postgresql" in conn_name:
+        assert issubclass(df.BoolColWithNull.dtype.type, object)
+    else:
+        assert issubclass(df.BoolColWithNull.dtype.type, np.floating)
+
+
+@pytest.mark.parametrize("conn", mysql_connectable)
+def test_read_procedure(conn, request):
+    conn = request.getfixturevalue(conn)
+
+    # GH 7324
+    # Although it is more an api test, it is added to the
+    # mysql tests as sqlite does not have stored procedures
+    from sqlalchemy import text
+    from sqlalchemy.engine import Engine
+
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3]})
+    df.to_sql(name="test_frame", con=conn, index=False)
+
+    proc = """DROP PROCEDURE IF EXISTS get_testdb;
+
+    CREATE PROCEDURE get_testdb ()
+
+    BEGIN
+        SELECT * FROM test_frame;
+    END"""
+    proc = text(proc)
+    if isinstance(conn, Engine):
+        with conn.connect() as engine_conn:
+            with engine_conn.begin():
+                engine_conn.execute(proc)
+    else:
+        with conn.begin():
+            conn.execute(proc)
+
+    res1 = sql.read_sql_query("CALL get_testdb();", conn)
+    tm.assert_frame_equal(df, res1)
+
+    # test delegation to read_sql_query
+    res2 = sql.read_sql("CALL get_testdb();", conn)
+    tm.assert_frame_equal(df, res2)
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+@pytest.mark.parametrize("expected_count", [2, "Success!"])
+def test_copy_from_callable_insertion_method(conn, expected_count, request):
+    # GH 8953
+    # Example in io.rst found under _io.sql.method
+    # not available in sqlite, mysql
+    def psql_insert_copy(table, conn, keys, data_iter):
+        # gets a DBAPI connection that can provide a cursor
+        dbapi_conn = conn.connection
+        with dbapi_conn.cursor() as cur:
+            s_buf = StringIO()
+            writer = csv.writer(s_buf)
+            writer.writerows(data_iter)
+            s_buf.seek(0)
+
+            columns = ", ".join([f'"{k}"' for k in keys])
+            if table.schema:
+                table_name = f"{table.schema}.{table.name}"
+            else:
+                table_name = table.name
+
+            sql_query = f"COPY {table_name} ({columns}) FROM STDIN WITH CSV"
+            cur.copy_expert(sql=sql_query, file=s_buf)
+        return expected_count
+
+    conn = request.getfixturevalue(conn)
+    expected = DataFrame({"col1": [1, 2], "col2": [0.1, 0.2], "col3": ["a", "n"]})
+    result_count = expected.to_sql(
+        name="test_frame", con=conn, index=False, method=psql_insert_copy
+    )
+    # GH 46891
+    if expected_count is None:
+        assert result_count is None
+    else:
+        assert result_count == expected_count
+    result = sql.read_sql_table("test_frame", conn)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+def test_insertion_method_on_conflict_do_nothing(conn, request):
+    # GH 15988: Example in to_sql docstring
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy.dialects.postgresql import insert
+    from sqlalchemy.engine import Engine
+    from sqlalchemy.sql import text
+
+    def insert_on_conflict(table, conn, keys, data_iter):
+        data = [dict(zip(keys, row)) for row in data_iter]
+        stmt = (
+            insert(table.table)
+            .values(data)
+            .on_conflict_do_nothing(index_elements=["a"])
+        )
+        result = conn.execute(stmt)
+        return result.rowcount
+
+    create_sql = text(
+        """
+    CREATE TABLE test_insert_conflict (
+        a  integer PRIMARY KEY,
+        b  numeric,
+        c  text
+    );
+    """
+    )
+    if isinstance(conn, Engine):
+        with conn.connect() as con:
+            with con.begin():
+                con.execute(create_sql)
+    else:
+        with conn.begin():
+            conn.execute(create_sql)
+
+    expected = DataFrame([[1, 2.1, "a"]], columns=list("abc"))
+    expected.to_sql(
+        name="test_insert_conflict", con=conn, if_exists="append", index=False
+    )
+
+    df_insert = DataFrame([[1, 3.2, "b"]], columns=list("abc"))
+    inserted = df_insert.to_sql(
+        name="test_insert_conflict",
+        con=conn,
+        index=False,
+        if_exists="append",
+        method=insert_on_conflict,
+    )
+    result = sql.read_sql_table("test_insert_conflict", conn)
+    tm.assert_frame_equal(result, expected)
+    assert inserted == 0
+
+    # Cleanup
+    with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.drop_table("test_insert_conflict")
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_to_sql_on_public_schema(conn, request):
+    if "sqlite" in conn or "mysql" in conn:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="test for public schema only specific to postgresql"
+            )
+        )
+
+    conn = request.getfixturevalue(conn)
+
+    test_data = DataFrame([[1, 2.1, "a"], [2, 3.1, "b"]], columns=list("abc"))
+    test_data.to_sql(
+        name="test_public_schema",
+        con=conn,
+        if_exists="append",
+        index=False,
+        schema="public",
+    )
+
+    df_out = sql.read_sql_table("test_public_schema", conn, schema="public")
+    tm.assert_frame_equal(test_data, df_out)
+
+
+@pytest.mark.parametrize("conn", mysql_connectable)
+def test_insertion_method_on_conflict_update(conn, request):
+    # GH 14553: Example in to_sql docstring
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy.dialects.mysql import insert
+    from sqlalchemy.engine import Engine
+    from sqlalchemy.sql import text
+
+    def insert_on_conflict(table, conn, keys, data_iter):
+        data = [dict(zip(keys, row)) for row in data_iter]
+        stmt = insert(table.table).values(data)
+        stmt = stmt.on_duplicate_key_update(b=stmt.inserted.b, c=stmt.inserted.c)
+        result = conn.execute(stmt)
+        return result.rowcount
+
+    create_sql = text(
+        """
+    CREATE TABLE test_insert_conflict (
+        a INT PRIMARY KEY,
+        b FLOAT,
+        c VARCHAR(10)
+    );
+    """
+    )
+    if isinstance(conn, Engine):
+        with conn.connect() as con:
+            with con.begin():
+                con.execute(create_sql)
+    else:
+        with conn.begin():
+            conn.execute(create_sql)
+
+    df = DataFrame([[1, 2.1, "a"]], columns=list("abc"))
+    df.to_sql(name="test_insert_conflict", con=conn, if_exists="append", index=False)
+
+    expected = DataFrame([[1, 3.2, "b"]], columns=list("abc"))
+    inserted = expected.to_sql(
+        name="test_insert_conflict",
+        con=conn,
+        index=False,
+        if_exists="append",
+        method=insert_on_conflict,
+    )
+    result = sql.read_sql_table("test_insert_conflict", conn)
+    tm.assert_frame_equal(result, expected)
+    assert inserted == 2
+
+    # Cleanup
+    with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.drop_table("test_insert_conflict")
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+def test_read_view_postgres(conn, request):
+    # GH 52969
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy.engine import Engine
+    from sqlalchemy.sql import text
+
+    table_name = f"group_{uuid.uuid4().hex}"
+    view_name = f"group_view_{uuid.uuid4().hex}"
+
+    sql_stmt = text(
+        f"""
+    CREATE TABLE {table_name} (
+        group_id INTEGER,
+        name TEXT
+    );
+    INSERT INTO {table_name} VALUES
+        (1, 'name');
+    CREATE VIEW {view_name}
+    AS
+    SELECT * FROM {table_name};
+    """
+    )
+    if isinstance(conn, Engine):
+        with conn.connect() as con:
+            with con.begin():
+                con.execute(sql_stmt)
+    else:
+        with conn.begin():
+            conn.execute(sql_stmt)
+    result = read_sql_table(view_name, conn)
+    expected = DataFrame({"group_id": [1], "name": "name"})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_read_view_sqlite(sqlite_buildin):
+    # GH 52969
+    create_table = """
+CREATE TABLE groups (
+   group_id INTEGER,
+   name TEXT
+);
+"""
+    insert_into = """
+INSERT INTO groups VALUES
+    (1, 'name');
+"""
+    create_view = """
+CREATE VIEW group_view
+AS
+SELECT * FROM groups;
+"""
+    sqlite_buildin.execute(create_table)
+    sqlite_buildin.execute(insert_into)
+    sqlite_buildin.execute(create_view)
+    result = pd.read_sql("SELECT * FROM group_view", sqlite_buildin)
+    expected = DataFrame({"group_id": [1], "name": "name"})
+    tm.assert_frame_equal(result, expected)
+
+
+def flavor(conn_name):
+    if "postgresql" in conn_name:
+        return "postgresql"
+    elif "sqlite" in conn_name:
+        return "sqlite"
+    elif "mysql" in conn_name:
+        return "mysql"
+
+    raise ValueError(f"unsupported connection: {conn_name}")
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_sql_iris_parameter(conn, request, sql_strings):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'params' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    query = sql_strings["read_parameters"][flavor(conn_name)]
+    params = ("Iris-setosa", 5.1)
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            iris_frame = pandasSQL.read_query(query, params=params)
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_sql_iris_named_parameter(conn, request, sql_strings):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'params' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    query = sql_strings["read_named_parameters"][flavor(conn_name)]
+    params = {"name": "Iris-setosa", "length": 5.1}
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            iris_frame = pandasSQL.read_query(query, params=params)
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_sql_iris_no_parameter_with_percent(conn, request, sql_strings):
+    if "mysql" in conn or ("postgresql" in conn and "adbc" not in conn):
+        request.applymarker(pytest.mark.xfail(reason="broken test"))
+
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+
+    query = sql_strings["read_no_parameters_with_percent"][flavor(conn_name)]
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            iris_frame = pandasSQL.read_query(query, params=None)
+    check_iris_frame(iris_frame)
+
+
+# -----------------------------------------------------------------------------
+# -- Testing the public API
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_api_read_sql_view(conn, request):
+    conn = request.getfixturevalue(conn)
+    iris_frame = sql.read_sql_query("SELECT * FROM iris_view", conn)
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_api_read_sql_with_chunksize_no_result(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="chunksize argument NotImplemented with ADBC")
+        )
+    conn = request.getfixturevalue(conn)
+    query = 'SELECT * FROM iris_view WHERE "SepalLength" < 0.0'
+    with_batch = sql.read_sql_query(query, conn, chunksize=5)
+    without_batch = sql.read_sql_query(query, conn)
+    tm.assert_frame_equal(concat(with_batch), without_batch)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql(conn, request, test_frame1):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame1", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame1")
+
+    sql.to_sql(test_frame1, "test_frame1", conn)
+    assert sql.has_table("test_frame1", conn)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_fail(conn, request, test_frame1):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame2", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame2")
+
+    sql.to_sql(test_frame1, "test_frame2", conn, if_exists="fail")
+    assert sql.has_table("test_frame2", conn)
+
+    msg = "Table 'test_frame2' already exists"
+    with pytest.raises(ValueError, match=msg):
+        sql.to_sql(test_frame1, "test_frame2", conn, if_exists="fail")
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_replace(conn, request, test_frame1):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame3", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame3")
+
+    sql.to_sql(test_frame1, "test_frame3", conn, if_exists="fail")
+    # Add to table again
+    sql.to_sql(test_frame1, "test_frame3", conn, if_exists="replace")
+    assert sql.has_table("test_frame3", conn)
+
+    num_entries = len(test_frame1)
+    num_rows = count_rows(conn, "test_frame3")
+
+    assert num_rows == num_entries
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_append(conn, request, test_frame1):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame4", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame4")
+
+    assert sql.to_sql(test_frame1, "test_frame4", conn, if_exists="fail") == 4
+
+    # Add to table again
+    assert sql.to_sql(test_frame1, "test_frame4", conn, if_exists="append") == 4
+    assert sql.has_table("test_frame4", conn)
+
+    num_entries = 2 * len(test_frame1)
+    num_rows = count_rows(conn, "test_frame4")
+
+    assert num_rows == num_entries
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_type_mapping(conn, request, test_frame3):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame5", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame5")
+
+    sql.to_sql(test_frame3, "test_frame5", conn, index=False)
+    result = sql.read_sql("SELECT * FROM test_frame5", conn)
+
+    tm.assert_frame_equal(test_frame3, result)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_series(conn, request):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_series", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_series")
+
+    s = Series(np.arange(5, dtype="int64"), name="series")
+    sql.to_sql(s, "test_series", conn, index=False)
+    s2 = sql.read_sql_query("SELECT * FROM test_series", conn)
+    tm.assert_frame_equal(s.to_frame(), s2)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_roundtrip(conn, request, test_frame1):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame_roundtrip", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame_roundtrip")
+
+    sql.to_sql(test_frame1, "test_frame_roundtrip", con=conn)
+    result = sql.read_sql_query("SELECT * FROM test_frame_roundtrip", con=conn)
+
+    # HACK!
+    if "adbc" in conn_name:
+        result = result.drop(columns="__index_level_0__")
+    else:
+        result = result.drop(columns="level_0")
+    tm.assert_frame_equal(result, test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_roundtrip_chunksize(conn, request, test_frame1):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="chunksize argument NotImplemented with ADBC")
+        )
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_frame_roundtrip", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_frame_roundtrip")
+
+    sql.to_sql(
+        test_frame1,
+        "test_frame_roundtrip",
+        con=conn,
+        index=False,
+        chunksize=2,
+    )
+    result = sql.read_sql_query("SELECT * FROM test_frame_roundtrip", con=conn)
+    tm.assert_frame_equal(result, test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_api_execute_sql(conn, request):
+    # drop_sql = "DROP TABLE IF EXISTS test"  # should already be done
+    conn = request.getfixturevalue(conn)
+    with sql.pandasSQL_builder(conn) as pandas_sql:
+        iris_results = pandas_sql.execute("SELECT * FROM iris")
+        row = iris_results.fetchone()
+        iris_results.close()
+    assert list(row) == [5.1, 3.5, 1.4, 0.2, "Iris-setosa"]
+
+
+@pytest.mark.parametrize("conn", all_connectable_types)
+def test_api_date_parsing(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    # Test date parsing in read_sql
+    # No Parsing
+    df = sql.read_sql_query("SELECT * FROM types", conn)
+    if not ("mysql" in conn_name or "postgres" in conn_name):
+        assert not issubclass(df.DateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_query("SELECT * FROM types", conn, parse_dates=["DateCol"])
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+    assert df.DateCol.tolist() == [
+        Timestamp(2000, 1, 3, 0, 0, 0),
+        Timestamp(2000, 1, 4, 0, 0, 0),
+    ]
+
+    df = sql.read_sql_query(
+        "SELECT * FROM types",
+        conn,
+        parse_dates={"DateCol": "%Y-%m-%d %H:%M:%S"},
+    )
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+    assert df.DateCol.tolist() == [
+        Timestamp(2000, 1, 3, 0, 0, 0),
+        Timestamp(2000, 1, 4, 0, 0, 0),
+    ]
+
+    df = sql.read_sql_query("SELECT * FROM types", conn, parse_dates=["IntDateCol"])
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+    assert df.IntDateCol.tolist() == [
+        Timestamp(1986, 12, 25, 0, 0, 0),
+        Timestamp(2013, 1, 1, 0, 0, 0),
+    ]
+
+    df = sql.read_sql_query(
+        "SELECT * FROM types", conn, parse_dates={"IntDateCol": "s"}
+    )
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+    assert df.IntDateCol.tolist() == [
+        Timestamp(1986, 12, 25, 0, 0, 0),
+        Timestamp(2013, 1, 1, 0, 0, 0),
+    ]
+
+    df = sql.read_sql_query(
+        "SELECT * FROM types",
+        conn,
+        parse_dates={"IntDateOnlyCol": "%Y%m%d"},
+    )
+    assert issubclass(df.IntDateOnlyCol.dtype.type, np.datetime64)
+    assert df.IntDateOnlyCol.tolist() == [
+        Timestamp("2010-10-10"),
+        Timestamp("2010-12-12"),
+    ]
+
+
+@pytest.mark.parametrize("conn", all_connectable_types)
+@pytest.mark.parametrize("error", ["raise", "coerce"])
+@pytest.mark.parametrize(
+    "read_sql, text, mode",
+    [
+        (sql.read_sql, "SELECT * FROM types", ("sqlalchemy", "fallback")),
+        (sql.read_sql, "types", ("sqlalchemy")),
+        (
+            sql.read_sql_query,
+            "SELECT * FROM types",
+            ("sqlalchemy", "fallback"),
+        ),
+        (sql.read_sql_table, "types", ("sqlalchemy")),
+    ],
+)
+def test_api_custom_dateparsing_error(
+    conn, request, read_sql, text, mode, error, types_data_frame
+):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if text == "types" and conn_name == "sqlite_buildin_types":
+        request.applymarker(
+            pytest.mark.xfail(reason="failing combination of arguments")
+        )
+
+    expected = types_data_frame.astype({"DateCol": "datetime64[s]"})
+
+    result = read_sql(
+        text,
+        con=conn,
+        parse_dates={
+            "DateCol": {"errors": error},
+        },
+    )
+    if "postgres" in conn_name:
+        # TODO: clean up types_data_frame fixture
+        result["BoolCol"] = result["BoolCol"].astype(int)
+        result["BoolColWithNull"] = result["BoolColWithNull"].astype(float)
+
+    if conn_name == "postgresql_adbc_types":
+        expected = expected.astype(
+            {
+                "IntDateCol": "int32",
+                "IntDateOnlyCol": "int32",
+                "IntCol": "int32",
+            }
+        )
+
+    if conn_name == "postgresql_adbc_types" and pa_version_under14p1:
+        expected["DateCol"] = expected["DateCol"].astype("datetime64[ns]")
+    elif "postgres" in conn_name or "mysql" in conn_name:
+        expected["DateCol"] = expected["DateCol"].astype("datetime64[us]")
+    else:
+        expected["DateCol"] = expected["DateCol"].astype("datetime64[s]")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable_types)
+def test_api_date_and_index(conn, request):
+    # Test case where same column appears in parse_date and index_col
+    conn = request.getfixturevalue(conn)
+    df = sql.read_sql_query(
+        "SELECT * FROM types",
+        conn,
+        index_col="DateCol",
+        parse_dates=["DateCol", "IntDateCol"],
+    )
+
+    assert issubclass(df.index.dtype.type, np.datetime64)
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_timedelta(conn, request):
+    # see #6921
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_timedelta", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_timedelta")
+
+    df = to_timedelta(Series(["00:00:01", "00:00:03"], name="foo")).to_frame()
+
+    if conn_name == "sqlite_adbc_conn":
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="sqlite ADBC driver doesn't implement timedelta",
+            )
+        )
+
+    if "adbc" in conn_name:
+        if pa_version_under14p1:
+            exp_warning = DeprecationWarning
+        else:
+            exp_warning = None
+    else:
+        exp_warning = UserWarning
+
+    with tm.assert_produces_warning(exp_warning, check_stacklevel=False):
+        result_count = df.to_sql(name="test_timedelta", con=conn)
+    assert result_count == 2
+    result = sql.read_sql_query("SELECT * FROM test_timedelta", conn)
+
+    if conn_name == "postgresql_adbc_conn":
+        # TODO: Postgres stores an INTERVAL, which ADBC reads as a Month-Day-Nano
+        # Interval; the default pandas type mapper maps this to a DateOffset
+        # but maybe we should try and restore the timedelta here?
+        expected = Series(
+            [
+                pd.DateOffset(months=0, days=0, microseconds=1000000, nanoseconds=0),
+                pd.DateOffset(months=0, days=0, microseconds=3000000, nanoseconds=0),
+            ],
+            name="foo",
+        )
+    else:
+        expected = df["foo"].astype("int64")
+    tm.assert_series_equal(result["foo"], expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_complex_raises(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = DataFrame({"a": [1 + 1j, 2j]})
+
+    if "adbc" in conn_name:
+        msg = "datatypes not supported"
+    else:
+        msg = "Complex datatypes not supported"
+    with pytest.raises(ValueError, match=msg):
+        assert df.to_sql("test_complex", con=conn) is None
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize(
+    "index_name,index_label,expected",
+    [
+        # no index name, defaults to 'index'
+        (None, None, "index"),
+        # specifying index_label
+        (None, "other_label", "other_label"),
+        # using the index name
+        ("index_name", None, "index_name"),
+        # has index name, but specifying index_label
+        ("index_name", "other_label", "other_label"),
+        # index name is integer
+        (0, None, "0"),
+        # index name is None but index label is integer
+        (None, 0, "0"),
+    ],
+)
+def test_api_to_sql_index_label(conn, request, index_name, index_label, expected):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="index_label argument NotImplemented with ADBC")
+        )
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_index_label", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_index_label")
+
+    temp_frame = DataFrame({"col1": range(4)})
+    temp_frame.index.name = index_name
+    query = "SELECT * FROM test_index_label"
+    sql.to_sql(temp_frame, "test_index_label", conn, index_label=index_label)
+    frame = sql.read_sql_query(query, conn)
+    assert frame.columns[0] == expected
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_to_sql_index_label_multiindex(conn, request):
+    conn_name = conn
+    if "mysql" in conn_name:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="MySQL can fail using TEXT without length as key", strict=False
+            )
+        )
+    elif "adbc" in conn_name:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="index_label argument NotImplemented with ADBC")
+        )
+
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_index_label", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_index_label")
+
+    expected_row_count = 4
+    temp_frame = DataFrame(
+        {"col1": range(4)},
+        index=MultiIndex.from_product([("A0", "A1"), ("B0", "B1")]),
+    )
+
+    # no index name, defaults to 'level_0' and 'level_1'
+    result = sql.to_sql(temp_frame, "test_index_label", conn)
+    assert result == expected_row_count
+    frame = sql.read_sql_query("SELECT * FROM test_index_label", conn)
+    assert frame.columns[0] == "level_0"
+    assert frame.columns[1] == "level_1"
+
+    # specifying index_label
+    result = sql.to_sql(
+        temp_frame,
+        "test_index_label",
+        conn,
+        if_exists="replace",
+        index_label=["A", "B"],
+    )
+    assert result == expected_row_count
+    frame = sql.read_sql_query("SELECT * FROM test_index_label", conn)
+    assert frame.columns[:2].tolist() == ["A", "B"]
+
+    # using the index name
+    temp_frame.index.names = ["A", "B"]
+    result = sql.to_sql(temp_frame, "test_index_label", conn, if_exists="replace")
+    assert result == expected_row_count
+    frame = sql.read_sql_query("SELECT * FROM test_index_label", conn)
+    assert frame.columns[:2].tolist() == ["A", "B"]
+
+    # has index name, but specifying index_label
+    result = sql.to_sql(
+        temp_frame,
+        "test_index_label",
+        conn,
+        if_exists="replace",
+        index_label=["C", "D"],
+    )
+    assert result == expected_row_count
+    frame = sql.read_sql_query("SELECT * FROM test_index_label", conn)
+    assert frame.columns[:2].tolist() == ["C", "D"]
+
+    msg = "Length of 'index_label' should match number of levels, which is 2"
+    with pytest.raises(ValueError, match=msg):
+        sql.to_sql(
+            temp_frame,
+            "test_index_label",
+            conn,
+            if_exists="replace",
+            index_label="C",
+        )
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_multiindex_roundtrip(conn, request):
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_multiindex_roundtrip", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_multiindex_roundtrip")
+
+    df = DataFrame.from_records(
+        [(1, 2.1, "line1"), (2, 1.5, "line2")],
+        columns=["A", "B", "C"],
+        index=["A", "B"],
+    )
+
+    df.to_sql(name="test_multiindex_roundtrip", con=conn)
+    result = sql.read_sql_query(
+        "SELECT * FROM test_multiindex_roundtrip", conn, index_col=["A", "B"]
+    )
+    tm.assert_frame_equal(df, result, check_index_type=True)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        None,
+        int,
+        float,
+        {"A": int, "B": float},
+    ],
+)
+def test_api_dtype_argument(conn, request, dtype):
+    # GH10285 Add dtype argument to read_sql_query
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_dtype_argument", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_dtype_argument")
+
+    df = DataFrame([[1.2, 3.4], [5.6, 7.8]], columns=["A", "B"])
+    assert df.to_sql(name="test_dtype_argument", con=conn) == 2
+
+    expected = df.astype(dtype)
+
+    if "postgres" in conn_name:
+        query = 'SELECT "A", "B" FROM test_dtype_argument'
+    else:
+        query = "SELECT A, B FROM test_dtype_argument"
+    result = sql.read_sql_query(query, con=conn, dtype=dtype)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_integer_col_names(conn, request):
+    conn = request.getfixturevalue(conn)
+    df = DataFrame([[1, 2], [3, 4]], columns=[0, 1])
+    sql.to_sql(df, "test_frame_integer_col_names", conn, if_exists="replace")
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_get_schema(conn, request, test_frame1):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'get_schema' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn = request.getfixturevalue(conn)
+    create_sql = sql.get_schema(test_frame1, "test", con=conn)
+    assert "CREATE" in create_sql
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_get_schema_with_schema(conn, request, test_frame1):
+    # GH28486
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'get_schema' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn = request.getfixturevalue(conn)
+    create_sql = sql.get_schema(test_frame1, "test", con=conn, schema="pypi")
+    assert "CREATE TABLE pypi." in create_sql
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_get_schema_dtypes(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'get_schema' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    float_frame = DataFrame({"a": [1.1, 1.2], "b": [2.1, 2.2]})
+
+    if conn_name == "sqlite_buildin":
+        dtype = "INTEGER"
+    else:
+        from sqlalchemy import Integer
+
+        dtype = Integer
+    create_sql = sql.get_schema(float_frame, "test", con=conn, dtype={"b": dtype})
+    assert "CREATE" in create_sql
+    assert "INTEGER" in create_sql
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_get_schema_keys(conn, request, test_frame1):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="'get_schema' not implemented for ADBC drivers",
+                strict=True,
+            )
+        )
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    frame = DataFrame({"Col1": [1.1, 1.2], "Col2": [2.1, 2.2]})
+    create_sql = sql.get_schema(frame, "test", con=conn, keys="Col1")
+
+    if "mysql" in conn_name:
+        constraint_sentence = "CONSTRAINT test_pk PRIMARY KEY (`Col1`)"
+    else:
+        constraint_sentence = 'CONSTRAINT test_pk PRIMARY KEY ("Col1")'
+    assert constraint_sentence in create_sql
+
+    # multiple columns as key (GH10385)
+    create_sql = sql.get_schema(test_frame1, "test", con=conn, keys=["A", "B"])
+    if "mysql" in conn_name:
+        constraint_sentence = "CONSTRAINT test_pk PRIMARY KEY (`A`, `B`)"
+    else:
+        constraint_sentence = 'CONSTRAINT test_pk PRIMARY KEY ("A", "B")'
+    assert constraint_sentence in create_sql
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_chunksize_read(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="chunksize argument NotImplemented with ADBC")
+        )
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_chunksize", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_chunksize")
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((22, 5)), columns=list("abcde")
+    )
+    df.to_sql(name="test_chunksize", con=conn, index=False)
+
+    # reading the query in one time
+    res1 = sql.read_sql_query("select * from test_chunksize", conn)
+
+    # reading the query in chunks with read_sql_query
+    res2 = DataFrame()
+    i = 0
+    sizes = [5, 5, 5, 5, 2]
+
+    for chunk in sql.read_sql_query("select * from test_chunksize", conn, chunksize=5):
+        res2 = concat([res2, chunk], ignore_index=True)
+        assert len(chunk) == sizes[i]
+        i += 1
+
+    tm.assert_frame_equal(res1, res2)
+
+    # reading the query in chunks with read_sql_query
+    if conn_name == "sqlite_buildin":
+        with pytest.raises(NotImplementedError, match=""):
+            sql.read_sql_table("test_chunksize", conn, chunksize=5)
+    else:
+        res3 = DataFrame()
+        i = 0
+        sizes = [5, 5, 5, 5, 2]
+
+        for chunk in sql.read_sql_table("test_chunksize", conn, chunksize=5):
+            res3 = concat([res3, chunk], ignore_index=True)
+            assert len(chunk) == sizes[i]
+            i += 1
+
+        tm.assert_frame_equal(res1, res3)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_categorical(conn, request):
+    if conn == "postgresql_adbc_conn":
+        adbc = import_optional_dependency("adbc_driver_postgresql", errors="ignore")
+        if adbc is not None and Version(adbc.__version__) < Version("0.9.0"):
+            request.node.add_marker(
+                pytest.mark.xfail(
+                    reason="categorical dtype not implemented for ADBC postgres driver",
+                    strict=True,
+                )
+            )
+    # GH8624
+    # test that categorical gets written correctly as dense column
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_categorical", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_categorical")
+
+    df = DataFrame(
+        {
+            "person_id": [1, 2, 3],
+            "person_name": ["John P. Doe", "Jane Dove", "John P. Doe"],
+        }
+    )
+    df2 = df.copy()
+    df2["person_name"] = df2["person_name"].astype("category")
+
+    df2.to_sql(name="test_categorical", con=conn, index=False)
+    res = sql.read_sql_query("SELECT * FROM test_categorical", conn)
+
+    tm.assert_frame_equal(res, df)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_unicode_column_name(conn, request):
+    # GH 11431
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_unicode", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_unicode")
+
+    df = DataFrame([[1, 2], [3, 4]], columns=["\xe9", "b"])
+    df.to_sql(name="test_unicode", con=conn, index=False)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_escaped_table_name(conn, request):
+    # GH 13206
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("d1187b08-4943-4c8d-a7f6", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("d1187b08-4943-4c8d-a7f6")
+
+    df = DataFrame({"A": [0, 1, 2], "B": [0.2, np.nan, 5.6]})
+    df.to_sql(name="d1187b08-4943-4c8d-a7f6", con=conn, index=False)
+
+    if "postgres" in conn_name:
+        query = 'SELECT * FROM "d1187b08-4943-4c8d-a7f6"'
+    else:
+        query = "SELECT * FROM `d1187b08-4943-4c8d-a7f6`"
+    res = sql.read_sql_query(query, conn)
+
+    tm.assert_frame_equal(res, df)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_api_read_sql_duplicate_columns(conn, request):
+    # GH#53117
+    if "adbc" in conn:
+        pa = pytest.importorskip("pyarrow")
+        if not (
+            Version(pa.__version__) >= Version("16.0")
+            and conn in ["sqlite_adbc_conn", "postgresql_adbc_conn"]
+        ):
+            request.node.add_marker(
+                pytest.mark.xfail(
+                    reason="pyarrow->pandas throws ValueError", strict=True
+                )
+            )
+    conn = request.getfixturevalue(conn)
+    if sql.has_table("test_table", conn):
+        with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+            pandasSQL.drop_table("test_table")
+
+    df = DataFrame({"a": [1, 2, 3], "b": [0.1, 0.2, 0.3], "c": 1})
+    df.to_sql(name="test_table", con=conn, index=False)
+
+    result = pd.read_sql("SELECT a, b, a +1 as a, c FROM test_table", conn)
+    expected = DataFrame(
+        [[1, 0.1, 2, 1], [2, 0.2, 3, 1], [3, 0.3, 4, 1]],
+        columns=["a", "b", "a", "c"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_read_table_columns(conn, request, test_frame1):
+    # test columns argument in read_table
+    conn_name = conn
+    if conn_name == "sqlite_buildin":
+        request.applymarker(pytest.mark.xfail(reason="Not Implemented"))
+
+    conn = request.getfixturevalue(conn)
+    sql.to_sql(test_frame1, "test_frame", conn)
+
+    cols = ["A", "B"]
+
+    result = sql.read_sql_table("test_frame", conn, columns=cols)
+    assert result.columns.tolist() == cols
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_read_table_index_col(conn, request, test_frame1):
+    # test columns argument in read_table
+    conn_name = conn
+    if conn_name == "sqlite_buildin":
+        request.applymarker(pytest.mark.xfail(reason="Not Implemented"))
+
+    conn = request.getfixturevalue(conn)
+    sql.to_sql(test_frame1, "test_frame", conn)
+
+    result = sql.read_sql_table("test_frame", conn, index_col="index")
+    assert result.index.names == ["index"]
+
+    result = sql.read_sql_table("test_frame", conn, index_col=["A", "B"])
+    assert result.index.names == ["A", "B"]
+
+    result = sql.read_sql_table(
+        "test_frame", conn, index_col=["A", "B"], columns=["C", "D"]
+    )
+    assert result.index.names == ["A", "B"]
+    assert result.columns.tolist() == ["C", "D"]
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_read_sql_delegate(conn, request):
+    if conn == "sqlite_buildin_iris":
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="sqlite_buildin connection does not implement read_sql_table"
+            )
+        )
+
+    conn = request.getfixturevalue(conn)
+    iris_frame1 = sql.read_sql_query("SELECT * FROM iris", conn)
+    iris_frame2 = sql.read_sql("SELECT * FROM iris", conn)
+    tm.assert_frame_equal(iris_frame1, iris_frame2)
+
+    iris_frame1 = sql.read_sql_table("iris", conn)
+    iris_frame2 = sql.read_sql("iris", conn)
+    tm.assert_frame_equal(iris_frame1, iris_frame2)
+
+
+def test_not_reflect_all_tables(sqlite_conn):
+    conn = sqlite_conn
+    from sqlalchemy import text
+    from sqlalchemy.engine import Engine
+
+    # create invalid table
+    query_list = [
+        text("CREATE TABLE invalid (x INTEGER, y UNKNOWN);"),
+        text("CREATE TABLE other_table (x INTEGER, y INTEGER);"),
+    ]
+
+    for query in query_list:
+        if isinstance(conn, Engine):
+            with conn.connect() as conn:
+                with conn.begin():
+                    conn.execute(query)
+        else:
+            with conn.begin():
+                conn.execute(query)
+
+    with tm.assert_produces_warning(None):
+        sql.read_sql_table("other_table", conn)
+        sql.read_sql_query("SELECT * FROM other_table", conn)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_warning_case_insensitive_table_name(conn, request, test_frame1):
+    conn_name = conn
+    if conn_name == "sqlite_buildin" or "adbc" in conn_name:
+        request.applymarker(pytest.mark.xfail(reason="Does not raise warning"))
+
+    conn = request.getfixturevalue(conn)
+    # see gh-7815
+    with tm.assert_produces_warning(
+        UserWarning,
+        match=(
+            r"The provided table name 'TABLE1' is not found exactly as such in "
+            r"the database after writing the table, possibly due to case "
+            r"sensitivity issues. Consider using lower case table names."
+        ),
+    ):
+        with sql.SQLDatabase(conn) as db:
+            db.check_case_sensitive("TABLE1", "")
+
+    # Test that the warning is certainly NOT triggered in a normal case.
+    with tm.assert_produces_warning(None):
+        test_frame1.to_sql(name="CaseSensitive", con=conn)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_sqlalchemy_type_mapping(conn, request):
+    conn = request.getfixturevalue(conn)
+    from sqlalchemy import TIMESTAMP
+
+    # Test Timestamp objects (no datetime64 because of timezone) (GH9085)
+    df = DataFrame(
+        {"time": to_datetime(["2014-12-12 01:54", "2014-12-11 02:54"], utc=True)}
+    )
+    with sql.SQLDatabase(conn) as db:
+        table = sql.SQLTable("test_type", db, frame=df)
+        # GH 9086: TIMESTAMP is the suggested type for datetimes with timezones
+        assert isinstance(table.table.c["time"].type, TIMESTAMP)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+@pytest.mark.parametrize(
+    "integer, expected",
+    [
+        ("int8", "SMALLINT"),
+        ("Int8", "SMALLINT"),
+        ("uint8", "SMALLINT"),
+        ("UInt8", "SMALLINT"),
+        ("int16", "SMALLINT"),
+        ("Int16", "SMALLINT"),
+        ("uint16", "INTEGER"),
+        ("UInt16", "INTEGER"),
+        ("int32", "INTEGER"),
+        ("Int32", "INTEGER"),
+        ("uint32", "BIGINT"),
+        ("UInt32", "BIGINT"),
+        ("int64", "BIGINT"),
+        ("Int64", "BIGINT"),
+        (int, "BIGINT" if np.dtype(int).name == "int64" else "INTEGER"),
+    ],
+)
+def test_sqlalchemy_integer_mapping(conn, request, integer, expected):
+    # GH35076 Map pandas integer to optimal SQLAlchemy integer type
+    conn = request.getfixturevalue(conn)
+    df = DataFrame([0, 1], columns=["a"], dtype=integer)
+    with sql.SQLDatabase(conn) as db:
+        table = sql.SQLTable("test_type", db, frame=df)
+
+        result = str(table.table.c.a.type)
+    assert result == expected
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+@pytest.mark.parametrize("integer", ["uint64", "UInt64"])
+def test_sqlalchemy_integer_overload_mapping(conn, request, integer):
+    conn = request.getfixturevalue(conn)
+    # GH35076 Map pandas integer to optimal SQLAlchemy integer type
+    df = DataFrame([0, 1], columns=["a"], dtype=integer)
+    with sql.SQLDatabase(conn) as db:
+        with pytest.raises(
+            ValueError, match="Unsigned 64 bit integer datatype is not supported"
+        ):
+            sql.SQLTable("test_type", db, frame=df)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_database_uri_string(conn, request, test_frame1):
+    pytest.importorskip("sqlalchemy")
+    conn = request.getfixturevalue(conn)
+    # Test read_sql and .to_sql method with a database URI (GH10654)
+    # db_uri = 'sqlite:///:memory:' # raises
+    # sqlalchemy.exc.OperationalError: (sqlite3.OperationalError) near
+    # "iris": syntax error [SQL: 'iris']
+    with tm.ensure_clean() as name:
+        db_uri = "sqlite:///" + name
+        table = "iris"
+        test_frame1.to_sql(name=table, con=db_uri, if_exists="replace", index=False)
+        test_frame2 = sql.read_sql(table, db_uri)
+        test_frame3 = sql.read_sql_table(table, db_uri)
+        query = "SELECT * FROM iris"
+        test_frame4 = sql.read_sql_query(query, db_uri)
+    tm.assert_frame_equal(test_frame1, test_frame2)
+    tm.assert_frame_equal(test_frame1, test_frame3)
+    tm.assert_frame_equal(test_frame1, test_frame4)
+
+
+@td.skip_if_installed("pg8000")
+@pytest.mark.parametrize("conn", all_connectable)
+def test_pg8000_sqlalchemy_passthrough_error(conn, request):
+    pytest.importorskip("sqlalchemy")
+    conn = request.getfixturevalue(conn)
+    # using driver that will not be installed on CI to trigger error
+    # in sqlalchemy.create_engine -> test passing of this error to user
+    db_uri = "postgresql+pg8000://user:pass@host/dbname"
+    with pytest.raises(ImportError, match="pg8000"):
+        sql.read_sql("select * from table", db_uri)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_query_by_text_obj(conn, request):
+    # WIP : GH10846
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    from sqlalchemy import text
+
+    if "postgres" in conn_name:
+        name_text = text('select * from iris where "Name"=:name')
+    else:
+        name_text = text("select * from iris where name=:name")
+    iris_df = sql.read_sql(name_text, conn, params={"name": "Iris-versicolor"})
+    all_names = set(iris_df["Name"])
+    assert all_names == {"Iris-versicolor"}
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_query_by_select_obj(conn, request):
+    conn = request.getfixturevalue(conn)
+    # WIP : GH10846
+    from sqlalchemy import (
+        bindparam,
+        select,
+    )
+
+    iris = iris_table_metadata()
+    name_select = select(iris).where(iris.c.Name == bindparam("name"))
+    iris_df = sql.read_sql(name_select, conn, params={"name": "Iris-setosa"})
+    all_names = set(iris_df["Name"])
+    assert all_names == {"Iris-setosa"}
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_column_with_percentage(conn, request):
+    # GH 37157
+    conn_name = conn
+    if conn_name == "sqlite_buildin":
+        request.applymarker(pytest.mark.xfail(reason="Not Implemented"))
+
+    conn = request.getfixturevalue(conn)
+    df = DataFrame({"A": [0, 1, 2], "%_variation": [3, 4, 5]})
+    df.to_sql(name="test_column_percentage", con=conn, index=False)
+
+    res = sql.read_sql_table("test_column_percentage", conn)
+
+    tm.assert_frame_equal(res, df)
+
+
+def test_sql_open_close(test_frame3):
+    # Test if the IO in the database still work if the connection closed
+    # between the writing and reading (as in many real situations).
+
+    with tm.ensure_clean() as name:
+        with closing(sqlite3.connect(name)) as conn:
+            assert sql.to_sql(test_frame3, "test_frame3_legacy", conn, index=False) == 4
+
+        with closing(sqlite3.connect(name)) as conn:
+            result = sql.read_sql_query("SELECT * FROM test_frame3_legacy;", conn)
+
+    tm.assert_frame_equal(test_frame3, result)
+
+
+@td.skip_if_installed("sqlalchemy")
+def test_con_string_import_error():
+    conn = "mysql://root@localhost/pandas"
+    msg = "Using URI string without sqlalchemy installed"
+    with pytest.raises(ImportError, match=msg):
+        sql.read_sql("SELECT * FROM iris", conn)
+
+
+@td.skip_if_installed("sqlalchemy")
+def test_con_unknown_dbapi2_class_does_not_error_without_sql_alchemy_installed():
+    class MockSqliteConnection:
+        def __init__(self, *args, **kwargs) -> None:
+            self.conn = sqlite3.Connection(*args, **kwargs)
+
+        def __getattr__(self, name):
+            return getattr(self.conn, name)
+
+        def close(self):
+            self.conn.close()
+
+    with contextlib.closing(MockSqliteConnection(":memory:")) as conn:
+        with tm.assert_produces_warning(UserWarning, match="only supports SQLAlchemy"):
+            sql.read_sql("SELECT 1", conn)
+
+
+def test_sqlite_read_sql_delegate(sqlite_buildin_iris):
+    conn = sqlite_buildin_iris
+    iris_frame1 = sql.read_sql_query("SELECT * FROM iris", conn)
+    iris_frame2 = sql.read_sql("SELECT * FROM iris", conn)
+    tm.assert_frame_equal(iris_frame1, iris_frame2)
+
+    msg = "Execution failed on sql 'iris': near \"iris\": syntax error"
+    with pytest.raises(sql.DatabaseError, match=msg):
+        sql.read_sql("iris", conn)
+
+
+def test_get_schema2(test_frame1):
+    # without providing a connection object (available for backwards comp)
+    create_sql = sql.get_schema(test_frame1, "test")
+    assert "CREATE" in create_sql
+
+
+def test_sqlite_type_mapping(sqlite_buildin):
+    # Test Timestamp objects (no datetime64 because of timezone) (GH9085)
+    conn = sqlite_buildin
+    df = DataFrame(
+        {"time": to_datetime(["2014-12-12 01:54", "2014-12-11 02:54"], utc=True)}
+    )
+    db = sql.SQLiteDatabase(conn)
+    table = sql.SQLiteTable("test_type", db, frame=df)
+    schema = table.sql_schema()
+    for col in schema.split("\n"):
+        if col.split()[0].strip('"') == "time":
+            assert col.split()[1] == "TIMESTAMP"
+
+
+# -----------------------------------------------------------------------------
+# -- Database flavor specific tests
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_create_table(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import inspect
+
+    temp_frame = DataFrame({"one": [1.0, 2.0, 3.0, 4.0], "two": [4.0, 3.0, 2.0, 1.0]})
+    with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+        assert pandasSQL.to_sql(temp_frame, "temp_frame") == 4
+
+    insp = inspect(conn)
+    assert insp.has_table("temp_frame")
+
+    # Cleanup
+    with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.drop_table("temp_frame")
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_drop_table(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import inspect
+
+    temp_frame = DataFrame({"one": [1.0, 2.0, 3.0, 4.0], "two": [4.0, 3.0, 2.0, 1.0]})
+    with sql.SQLDatabase(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            assert pandasSQL.to_sql(temp_frame, "temp_frame") == 4
+
+        insp = inspect(conn)
+        assert insp.has_table("temp_frame")
+
+        with pandasSQL.run_transaction():
+            pandasSQL.drop_table("temp_frame")
+        try:
+            insp.clear_cache()  # needed with SQLAlchemy 2.0, unavailable prior
+        except AttributeError:
+            pass
+        assert not insp.has_table("temp_frame")
+
+
+@pytest.mark.parametrize("conn_name", all_connectable)
+def test_delete_rows_success(conn_name, test_frame1, request):
+    table_name = "temp_delete_rows_frame"
+    conn = request.getfixturevalue(conn_name)
+
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            assert pandasSQL.to_sql(test_frame1, table_name) == test_frame1.shape[0]
+
+        with pandasSQL.run_transaction():
+            assert pandasSQL.delete_rows(table_name) is None
+
+        assert count_rows(conn, table_name) == 0
+        assert pandasSQL.has_table(table_name)
+
+
+@pytest.mark.parametrize("conn_name", all_connectable)
+def test_delete_rows_is_atomic(conn_name, request):
+    sqlalchemy = pytest.importorskip("sqlalchemy")
+
+    table_name = "temp_delete_rows_atomic_frame"
+    table_stmt = f"CREATE TABLE {table_name} (a INTEGER, b INTEGER UNIQUE NOT NULL)"
+
+    if conn_name != "sqlite_buildin" and "adbc" not in conn_name:
+        table_stmt = sqlalchemy.text(table_stmt)
+
+    # setting dtype is mandatory for adbc related tests
+    original_df = DataFrame({"a": [1, 2], "b": [3, 4]}, dtype="int32")
+    replacing_df = DataFrame({"a": [5, 6, 7], "b": [8, 8, 8]}, dtype="int32")
+
+    conn = request.getfixturevalue(conn_name)
+    pandasSQL = pandasSQL_builder(conn)
+
+    with pandasSQL.run_transaction() as cur:
+        cur.execute(table_stmt)
+
+    with pandasSQL.run_transaction():
+        pandasSQL.to_sql(original_df, table_name, if_exists="append", index=False)
+
+    # inserting duplicated values in a UNIQUE constraint column
+    with pytest.raises(pd.errors.DatabaseError):
+        with pandasSQL.run_transaction():
+            pandasSQL.to_sql(
+                replacing_df, table_name, if_exists="delete_rows", index=False
+            )
+
+    # failed "delete_rows" is rolled back preserving original data
+    with pandasSQL.run_transaction():
+        result_df = pandasSQL.read_query(f"SELECT * FROM {table_name}", dtype="int32")
+        tm.assert_frame_equal(result_df, original_df)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_roundtrip(conn, request, test_frame1):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    pandasSQL = pandasSQL_builder(conn)
+    with pandasSQL.run_transaction():
+        assert pandasSQL.to_sql(test_frame1, "test_frame_roundtrip") == 4
+        result = pandasSQL.read_query("SELECT * FROM test_frame_roundtrip")
+
+    if "adbc" in conn_name:
+        result = result.rename(columns={"__index_level_0__": "level_0"})
+    result.set_index("level_0", inplace=True)
+    # result.index.astype(int)
+
+    result.index.name = None
+
+    tm.assert_frame_equal(result, test_frame1)
+
+
+@pytest.mark.parametrize("conn", all_connectable_iris)
+def test_execute_sql(conn, request):
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            iris_results = pandasSQL.execute("SELECT * FROM iris")
+            row = iris_results.fetchone()
+            iris_results.close()
+    assert list(row) == [5.1, 3.5, 1.4, 0.2, "Iris-setosa"]
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_sqlalchemy_read_table(conn, request):
+    conn = request.getfixturevalue(conn)
+    iris_frame = sql.read_sql_table("iris", con=conn)
+    check_iris_frame(iris_frame)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_sqlalchemy_read_table_columns(conn, request):
+    conn = request.getfixturevalue(conn)
+    iris_frame = sql.read_sql_table(
+        "iris", con=conn, columns=["SepalLength", "SepalLength"]
+    )
+    tm.assert_index_equal(iris_frame.columns, Index(["SepalLength", "SepalLength__1"]))
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_iris)
+def test_read_table_absent_raises(conn, request):
+    conn = request.getfixturevalue(conn)
+    msg = "Table this_doesnt_exist not found"
+    with pytest.raises(ValueError, match=msg):
+        sql.read_sql_table("this_doesnt_exist", con=conn)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_types)
+def test_sqlalchemy_default_type_conversion(conn, request):
+    conn_name = conn
+    if conn_name == "sqlite_str":
+        pytest.skip("types tables not created in sqlite_str fixture")
+    elif "mysql" in conn_name or "sqlite" in conn_name:
+        request.applymarker(
+            pytest.mark.xfail(reason="boolean dtype not inferred properly")
+        )
+
+    conn = request.getfixturevalue(conn)
+    df = sql.read_sql_table("types", conn)
+
+    assert issubclass(df.FloatCol.dtype.type, np.floating)
+    assert issubclass(df.IntCol.dtype.type, np.integer)
+    assert issubclass(df.BoolCol.dtype.type, np.bool_)
+
+    # Int column with NA values stays as float
+    assert issubclass(df.IntColWithNull.dtype.type, np.floating)
+    # Bool column with NA values becomes object
+    assert issubclass(df.BoolColWithNull.dtype.type, object)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_bigint(conn, request):
+    # int64 should be converted to BigInteger, GH7433
+    conn = request.getfixturevalue(conn)
+    df = DataFrame(data={"i64": [2**62]})
+    assert df.to_sql(name="test_bigint", con=conn, index=False) == 1
+    result = sql.read_sql_table("test_bigint", conn)
+
+    tm.assert_frame_equal(df, result)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_types)
+def test_default_date_load(conn, request):
+    conn_name = conn
+    if conn_name == "sqlite_str":
+        pytest.skip("types tables not created in sqlite_str fixture")
+    elif "sqlite" in conn_name:
+        request.applymarker(
+            pytest.mark.xfail(reason="sqlite does not read date properly")
+        )
+
+    conn = request.getfixturevalue(conn)
+    df = sql.read_sql_table("types", conn)
+
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+@pytest.mark.parametrize("parse_dates", [None, ["DateColWithTz"]])
+def test_datetime_with_timezone_query(conn, request, parse_dates):
+    # edge case that converts postgresql datetime with time zone types
+    # to datetime64[ns,psycopg2.tz.FixedOffsetTimezone..], which is ok
+    # but should be more natural, so coerce to datetime64[ns] for now
+    conn = request.getfixturevalue(conn)
+    expected = create_and_load_postgres_datetz(conn)
+
+    # GH11216
+    df = read_sql_query("select * from datetz", conn, parse_dates=parse_dates)
+    col = df.DateColWithTz
+    tm.assert_series_equal(col, expected)
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+def test_datetime_with_timezone_query_chunksize(conn, request):
+    conn = request.getfixturevalue(conn)
+    expected = create_and_load_postgres_datetz(conn)
+
+    df = concat(
+        list(read_sql_query("select * from datetz", conn, chunksize=1)),
+        ignore_index=True,
+    )
+    col = df.DateColWithTz
+    tm.assert_series_equal(col, expected)
+
+
+@pytest.mark.parametrize("conn", postgresql_connectable)
+def test_datetime_with_timezone_table(conn, request):
+    conn = request.getfixturevalue(conn)
+    expected = create_and_load_postgres_datetz(conn)
+    result = sql.read_sql_table("datetz", conn)
+
+    exp_frame = expected.to_frame()
+    tm.assert_frame_equal(result, exp_frame)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_datetime_with_timezone_roundtrip(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    # GH 9086
+    # Write datetimetz data to a db and read it back
+    # For dbs that support timestamps with timezones, should get back UTC
+    # otherwise naive data should be returned
+    expected = DataFrame(
+        {"A": date_range("2013-01-01 09:00:00", periods=3, tz="US/Pacific", unit="us")}
+    )
+    assert expected.to_sql(name="test_datetime_tz", con=conn, index=False) == 3
+
+    if "postgresql" in conn_name:
+        # SQLAlchemy "timezones" (i.e. offsets) are coerced to UTC
+        expected["A"] = expected["A"].dt.tz_convert("UTC")
+    else:
+        # Otherwise, timestamps are returned as local, naive
+        expected["A"] = expected["A"].dt.tz_localize(None)
+
+    result = sql.read_sql_table("test_datetime_tz", conn)
+    tm.assert_frame_equal(result, expected)
+
+    result = sql.read_sql_query("SELECT * FROM test_datetime_tz", conn)
+    if "sqlite" in conn_name:
+        # read_sql_query does not return datetime type like read_sql_table
+        assert isinstance(result.loc[0, "A"], str)
+        result["A"] = to_datetime(result["A"]).dt.as_unit("us")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_out_of_bounds_datetime(conn, request):
+    # GH 26761
+    conn = request.getfixturevalue(conn)
+    data = DataFrame({"date": datetime(9999, 1, 1)}, index=[0])
+    assert data.to_sql(name="test_datetime_obb", con=conn, index=False) == 1
+    result = sql.read_sql_table("test_datetime_obb", conn)
+    expected = DataFrame(
+        np.array([datetime(9999, 1, 1)], dtype="M8[us]"), columns=["date"]
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_naive_datetimeindex_roundtrip(conn, request):
+    # GH 23510
+    # Ensure that a naive DatetimeIndex isn't converted to UTC
+    conn = request.getfixturevalue(conn)
+    dates = date_range("2018-01-01", periods=5, freq="6h", unit="us")._with_freq(None)
+    expected = DataFrame({"nums": range(5)}, index=dates)
+    assert expected.to_sql(name="foo_table", con=conn, index_label="info_date") == 5
+    result = sql.read_sql_table("foo_table", conn, index_col="info_date")
+    # result index with gain a name from a set_index operation; expected
+    tm.assert_frame_equal(result, expected, check_names=False)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable_types)
+def test_date_parsing(conn, request):
+    # No Parsing
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = sql.read_sql_table("types", conn)
+    expected_type = object if "sqlite" in conn_name else np.datetime64
+    assert issubclass(df.DateCol.dtype.type, expected_type)
+
+    df = sql.read_sql_table("types", conn, parse_dates=["DateCol"])
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_table("types", conn, parse_dates={"DateCol": "%Y-%m-%d %H:%M:%S"})
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_table(
+        "types",
+        conn,
+        parse_dates={"DateCol": {"format": "%Y-%m-%d %H:%M:%S"}},
+    )
+    assert issubclass(df.DateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_table("types", conn, parse_dates=["IntDateCol"])
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_table("types", conn, parse_dates={"IntDateCol": "s"})
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+
+    df = sql.read_sql_table("types", conn, parse_dates={"IntDateCol": {"unit": "s"}})
+    assert issubclass(df.IntDateCol.dtype.type, np.datetime64)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_datetime(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = DataFrame(
+        {"A": date_range("2013-01-01 09:00:00", periods=3), "B": np.arange(3.0)}
+    )
+    assert df.to_sql(name="test_datetime", con=conn) == 3
+
+    # with read_table -> type information from schema used
+    result = sql.read_sql_table("test_datetime", conn)
+    result = result.drop("index", axis=1)
+
+    expected = df[:]
+    expected["A"] = expected["A"].astype("M8[us]")
+    tm.assert_frame_equal(result, expected)
+
+    # with read_sql -> no type information -> sqlite has no native
+    result = sql.read_sql_query("SELECT * FROM test_datetime", conn)
+    result = result.drop("index", axis=1)
+    if "sqlite" in conn_name:
+        assert isinstance(result.loc[0, "A"], str)
+        result["A"] = to_datetime(result["A"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_datetime_NaT(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = DataFrame(
+        {"A": date_range("2013-01-01 09:00:00", periods=3), "B": np.arange(3.0)}
+    )
+    df.loc[1, "A"] = np.nan
+    assert df.to_sql(name="test_datetime", con=conn, index=False) == 3
+
+    # with read_table -> type information from schema used
+    result = sql.read_sql_table("test_datetime", conn)
+    expected = df[:]
+    expected["A"] = expected["A"].astype("M8[us]")
+    tm.assert_frame_equal(result, expected)
+
+    # with read_sql -> no type information -> sqlite has no native
+    result = sql.read_sql_query("SELECT * FROM test_datetime", conn)
+    if "sqlite" in conn_name:
+        assert isinstance(result.loc[0, "A"], str)
+        result["A"] = to_datetime(result["A"], errors="coerce")
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_datetime_date(conn, request):
+    # test support for datetime.date
+    conn = request.getfixturevalue(conn)
+    df = DataFrame([date(2014, 1, 1), date(2014, 1, 2)], columns=["a"])
+    assert df.to_sql(name="test_date", con=conn, index=False) == 2
+    res = read_sql_table("test_date", conn)
+    result = res["a"]
+    expected = to_datetime(df["a"])
+    # comes back as datetime64
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_datetime_time(conn, request, sqlite_buildin):
+    # test support for datetime.time
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = DataFrame([time(9, 0, 0), time(9, 1, 30)], columns=["a"])
+    assert df.to_sql(name="test_time", con=conn, index=False) == 2
+    res = read_sql_table("test_time", conn)
+    tm.assert_frame_equal(res, df)
+
+    # GH8341
+    # first, use the fallback to have the sqlite adapter put in place
+    sqlite_conn = sqlite_buildin
+    assert sql.to_sql(df, "test_time2", sqlite_conn, index=False) == 2
+    res = sql.read_sql_query("SELECT * FROM test_time2", sqlite_conn)
+    ref = df.map(lambda _: _.strftime("%H:%M:%S.%f"))
+    tm.assert_frame_equal(ref, res)  # check if adapter is in place
+    # then test if sqlalchemy is unaffected by the sqlite adapter
+    assert sql.to_sql(df, "test_time3", conn, index=False) == 2
+    if "sqlite" in conn_name:
+        res = sql.read_sql_query("SELECT * FROM test_time3", conn)
+        ref = df.map(lambda _: _.strftime("%H:%M:%S.%f"))
+        tm.assert_frame_equal(ref, res)
+    res = sql.read_sql_table("test_time3", conn)
+    tm.assert_frame_equal(df, res)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_mixed_dtype_insert(conn, request):
+    # see GH6509
+    conn = request.getfixturevalue(conn)
+    s1 = Series(2**25 + 1, dtype=np.int32)
+    s2 = Series(0.0, dtype=np.float32)
+    df = DataFrame({"s1": s1, "s2": s2})
+
+    # write and read again
+    assert df.to_sql(name="test_read_write", con=conn, index=False) == 1
+    df2 = sql.read_sql_table("test_read_write", conn)
+
+    tm.assert_frame_equal(df, df2, check_dtype=False, check_exact=True)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_nan_numeric(conn, request):
+    # NaNs in numeric float column
+    conn = request.getfixturevalue(conn)
+    df = DataFrame({"A": [0, 1, 2], "B": [0.2, np.nan, 5.6]})
+    assert df.to_sql(name="test_nan", con=conn, index=False) == 3
+
+    # with read_table
+    result = sql.read_sql_table("test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+    # with read_sql
+    result = sql.read_sql_query("SELECT * FROM test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_nan_fullcolumn(conn, request):
+    # full NaN column (numeric float column)
+    conn = request.getfixturevalue(conn)
+    df = DataFrame({"A": [0, 1, 2], "B": [np.nan, np.nan, np.nan]})
+    assert df.to_sql(name="test_nan", con=conn, index=False) == 3
+
+    # with read_table
+    result = sql.read_sql_table("test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+    # with read_sql -> not type info from table -> stays None
+    df["B"] = df["B"].astype("object")
+    df["B"] = None
+    result = sql.read_sql_query("SELECT * FROM test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_nan_string(conn, request):
+    # NaNs in string column
+    conn = request.getfixturevalue(conn)
+    df = DataFrame({"A": [0, 1, 2], "B": ["a", "b", np.nan]})
+    assert df.to_sql(name="test_nan", con=conn, index=False) == 3
+
+    # NaNs are coming back as None
+    df.loc[2, "B"] = None
+
+    # with read_table
+    result = sql.read_sql_table("test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+    # with read_sql
+    result = sql.read_sql_query("SELECT * FROM test_nan", conn)
+    tm.assert_frame_equal(result, df)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_to_sql_save_index(conn, request):
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(
+                reason="ADBC implementation does not create index", strict=True
+            )
+        )
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    df = DataFrame.from_records(
+        [(1, 2.1, "line1"), (2, 1.5, "line2")], columns=["A", "B", "C"], index=["A"]
+    )
+
+    tbl_name = "test_to_sql_saves_index"
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            assert pandasSQL.to_sql(df, tbl_name) == 2
+
+    if conn_name in {"sqlite_buildin", "sqlite_str"}:
+        ixs = sql.read_sql_query(
+            "SELECT * FROM sqlite_master WHERE type = 'index' "
+            f"AND tbl_name = '{tbl_name}'",
+            conn,
+        )
+        ix_cols = []
+        for ix_name in ixs.name:
+            ix_info = sql.read_sql_query(f"PRAGMA index_info({ix_name})", conn)
+            ix_cols.append(ix_info.name.tolist())
+    else:
+        from sqlalchemy import inspect
+
+        insp = inspect(conn)
+
+        ixs = insp.get_indexes(tbl_name)
+        ix_cols = [i["column_names"] for i in ixs]
+
+    assert ix_cols == [["A"]]
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_transactions(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+
+    stmt = "CREATE TABLE test_trans (A INT, B TEXT)"
+    if conn_name != "sqlite_buildin" and "adbc" not in conn_name:
+        from sqlalchemy import text
+
+        stmt = text(stmt)
+
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction() as trans:
+            trans.execute(stmt)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_transaction_rollback(conn, request):
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction() as trans:
+            stmt = "CREATE TABLE test_trans (A INT, B TEXT)"
+            if "adbc" in conn_name or isinstance(pandasSQL, SQLiteDatabase):
+                trans.execute(stmt)
+            else:
+                from sqlalchemy import text
+
+                stmt = text(stmt)
+                trans.execute(stmt)
+
+        class DummyException(Exception):
+            pass
+
+        # Make sure when transaction is rolled back, no rows get inserted
+        ins_sql = "INSERT INTO test_trans (A,B) VALUES (1, 'blah')"
+        if isinstance(pandasSQL, SQLDatabase):
+            from sqlalchemy import text
+
+            ins_sql = text(ins_sql)
+        try:
+            with pandasSQL.run_transaction() as trans:
+                trans.execute(ins_sql)
+                raise DummyException("error")
+        except DummyException:
+            # ignore raised exception
+            pass
+        with pandasSQL.run_transaction():
+            res = pandasSQL.read_query("SELECT * FROM test_trans")
+        assert len(res) == 0
+
+        # Make sure when transaction is committed, rows do get inserted
+        with pandasSQL.run_transaction() as trans:
+            trans.execute(ins_sql)
+            res2 = pandasSQL.read_query("SELECT * FROM test_trans")
+        assert len(res2) == 1
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_get_schema_create_table(conn, request, test_frame3):
+    # Use a dataframe without a bool column, since MySQL converts bool to
+    # TINYINT (which read_sql_table returns as an int and causes a dtype
+    # mismatch)
+    if conn == "sqlite_str":
+        request.applymarker(
+            pytest.mark.xfail(reason="test does not support sqlite_str fixture")
+        )
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import text
+    from sqlalchemy.engine import Engine
+
+    tbl = "test_get_schema_create_table"
+    create_sql = sql.get_schema(test_frame3, tbl, con=conn)
+    blank_test_df = test_frame3.iloc[:0]
+
+    create_sql = text(create_sql)
+    if isinstance(conn, Engine):
+        with conn.connect() as newcon:
+            with newcon.begin():
+                newcon.execute(create_sql)
+    else:
+        conn.execute(create_sql)
+    returned_df = sql.read_sql_table(tbl, conn)
+    tm.assert_frame_equal(returned_df, blank_test_df, check_index_type=False)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_dtype(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import (
+        TEXT,
+        String,
+    )
+    from sqlalchemy.schema import MetaData
+
+    cols = ["A", "B"]
+    data = [(0.8, True), (0.9, None)]
+    df = DataFrame(data, columns=cols)
+    assert df.to_sql(name="dtype_test", con=conn) == 2
+    assert df.to_sql(name="dtype_test2", con=conn, dtype={"B": TEXT}) == 2
+    meta = MetaData()
+    meta.reflect(bind=conn)
+    sqltype = meta.tables["dtype_test2"].columns["B"].type
+    assert isinstance(sqltype, TEXT)
+    msg = "The type of B is not a SQLAlchemy type"
+    with pytest.raises(ValueError, match=msg):
+        df.to_sql(name="error", con=conn, dtype={"B": str})
+
+    # GH9083
+    assert df.to_sql(name="dtype_test3", con=conn, dtype={"B": String(10)}) == 2
+    meta.reflect(bind=conn)
+    sqltype = meta.tables["dtype_test3"].columns["B"].type
+    assert isinstance(sqltype, String)
+    assert sqltype.length == 10
+
+    # single dtype
+    assert df.to_sql(name="single_dtype_test", con=conn, dtype=TEXT) == 2
+    meta.reflect(bind=conn)
+    sqltypea = meta.tables["single_dtype_test"].columns["A"].type
+    sqltypeb = meta.tables["single_dtype_test"].columns["B"].type
+    assert isinstance(sqltypea, TEXT)
+    assert isinstance(sqltypeb, TEXT)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_notna_dtype(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import (
+        Boolean,
+        DateTime,
+        Float,
+        Integer,
+    )
+    from sqlalchemy.schema import MetaData
+
+    cols = {
+        "Bool": Series([True, None]),
+        "Date": Series([datetime(2012, 5, 1), None]),
+        "Int": Series([1, None], dtype="object"),
+        "Float": Series([1.1, None]),
+    }
+    df = DataFrame(cols)
+
+    tbl = "notna_dtype_test"
+    assert df.to_sql(name=tbl, con=conn) == 2
+    _ = sql.read_sql_table(tbl, conn)
+    meta = MetaData()
+    meta.reflect(bind=conn)
+    my_type = Integer if "mysql" in conn_name else Boolean
+    col_dict = meta.tables[tbl].columns
+    assert isinstance(col_dict["Bool"].type, my_type)
+    assert isinstance(col_dict["Date"].type, DateTime)
+    assert isinstance(col_dict["Int"].type, Integer)
+    assert isinstance(col_dict["Float"].type, Float)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_double_precision(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("sqlite_str has no inspection system")
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import (
+        BigInteger,
+        Float,
+        Integer,
+    )
+    from sqlalchemy.schema import MetaData
+
+    V = 1.23456789101112131415
+
+    df = DataFrame(
+        {
+            "f32": Series([V], dtype="float32"),
+            "f64": Series([V], dtype="float64"),
+            "f64_as_f32": Series([V], dtype="float64"),
+            "i32": Series([5], dtype="int32"),
+            "i64": Series([5], dtype="int64"),
+        }
+    )
+
+    assert (
+        df.to_sql(
+            name="test_dtypes",
+            con=conn,
+            index=False,
+            if_exists="replace",
+            dtype={"f64_as_f32": Float(precision=23)},
+        )
+        == 1
+    )
+    res = sql.read_sql_table("test_dtypes", conn)
+
+    # check precision of float64
+    assert np.round(df["f64"].iloc[0], 14) == np.round(res["f64"].iloc[0], 14)
+
+    # check sql types
+    meta = MetaData()
+    meta.reflect(bind=conn)
+    col_dict = meta.tables["test_dtypes"].columns
+    assert str(col_dict["f32"].type) == str(col_dict["f64_as_f32"].type)
+    assert isinstance(col_dict["f32"].type, Float)
+    assert isinstance(col_dict["f64"].type, Float)
+    assert isinstance(col_dict["i32"].type, Integer)
+    assert isinstance(col_dict["i64"].type, BigInteger)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_connectable_issue_example(conn, request):
+    conn = request.getfixturevalue(conn)
+
+    # This tests the example raised in issue
+    # https://github.com/pandas-dev/pandas/issues/10104
+    from sqlalchemy.engine import Engine
+
+    def test_select(connection):
+        query = "SELECT test_foo_data FROM test_foo_data"
+        return sql.read_sql_query(query, con=connection)
+
+    def test_append(connection, data):
+        data.to_sql(name="test_foo_data", con=connection, if_exists="append")
+
+    def test_connectable(conn):
+        # https://github.com/sqlalchemy/sqlalchemy/commit/
+        # 00b5c10846e800304caa86549ab9da373b42fa5d#r48323973
+        foo_data = test_select(conn)
+        test_append(conn, foo_data)
+
+    def main(connectable):
+        if isinstance(connectable, Engine):
+            with connectable.connect() as conn:
+                with conn.begin():
+                    test_connectable(conn)
+        else:
+            test_connectable(connectable)
+
+    assert (
+        DataFrame({"test_foo_data": [0, 1, 2]}).to_sql(name="test_foo_data", con=conn)
+        == 3
+    )
+    main(conn)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+@pytest.mark.parametrize(
+    "input",
+    [{"foo": [np.inf]}, {"foo": [-np.inf]}, {"foo": [-np.inf], "infe0": ["bar"]}],
+)
+def test_to_sql_with_negative_npinf(conn, request, input):
+    # GH 34431
+
+    df = DataFrame(input)
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+
+    if "mysql" in conn_name:
+        # GH 36465
+        # The input {"foo": [-np.inf], "infe0": ["bar"]} does not raise any error
+        # for pymysql version >= 0.10
+        # TODO(GH#36465): remove this version check after GH 36465 is fixed
+        pymysql = pytest.importorskip("pymysql")
+
+        if Version(pymysql.__version__) < Version("1.0.3") and "infe0" in df.columns:
+            mark = pytest.mark.xfail(reason="GH 36465")
+            request.applymarker(mark)
+
+        msg = "Execution failed on sql"
+        with pytest.raises(pd.errors.DatabaseError, match=msg):
+            df.to_sql(name="foobar", con=conn, index=False)
+    else:
+        assert df.to_sql(name="foobar", con=conn, index=False) == 1
+        res = sql.read_sql_table("foobar", conn)
+        tm.assert_equal(df, res)
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_temporary_table(conn, request):
+    if conn == "sqlite_str":
+        pytest.skip("test does not work with str connection")
+
+    conn = request.getfixturevalue(conn)
+
+    from sqlalchemy import (
+        Column,
+        Integer,
+        Unicode,
+        select,
+    )
+    from sqlalchemy.orm import (
+        Session,
+        declarative_base,
+    )
+
+    test_data = "Hello, World!"
+    expected = DataFrame({"spam": [test_data]})
+    Base = declarative_base()
+
+    class Temporary(Base):
+        __tablename__ = "temp_test"
+        __table_args__ = {"prefixes": ["TEMPORARY"]}
+        id = Column(Integer, primary_key=True)
+        spam = Column(Unicode(30), nullable=False)
+
+    with Session(conn) as session:
+        with session.begin():
+            conn = session.connection()
+            Temporary.__table__.create(conn)
+            session.add(Temporary(spam=test_data))
+            session.flush()
+            df = sql.read_sql_query(sql=select(Temporary.spam), con=conn)
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_invalid_engine(conn, request, test_frame1):
+    if conn == "sqlite_buildin" or "adbc" in conn:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="SQLiteDatabase/ADBCDatabase does not raise for bad engine"
+            )
+        )
+
+    conn = request.getfixturevalue(conn)
+    msg = "engine must be one of 'auto', 'sqlalchemy'"
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pytest.raises(ValueError, match=msg):
+            pandasSQL.to_sql(test_frame1, "test_frame1", engine="bad_engine")
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_to_sql_with_sql_engine(conn, request, test_frame1):
+    """`to_sql` with the `engine` param"""
+    # mostly copied from this class's `_to_sql()` method
+    conn = request.getfixturevalue(conn)
+    with pandasSQL_builder(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            assert pandasSQL.to_sql(test_frame1, "test_frame1", engine="auto") == 4
+            assert pandasSQL.has_table("test_frame1")
+
+    num_entries = len(test_frame1)
+    num_rows = count_rows(conn, "test_frame1")
+    assert num_rows == num_entries
+
+
+@pytest.mark.parametrize("conn", sqlalchemy_connectable)
+def test_options_sqlalchemy(conn, request, test_frame1):
+    # use the set option
+    conn = request.getfixturevalue(conn)
+    with pd.option_context("io.sql.engine", "sqlalchemy"):
+        with pandasSQL_builder(conn) as pandasSQL:
+            with pandasSQL.run_transaction():
+                assert pandasSQL.to_sql(test_frame1, "test_frame1") == 4
+                assert pandasSQL.has_table("test_frame1")
+
+        num_entries = len(test_frame1)
+        num_rows = count_rows(conn, "test_frame1")
+        assert num_rows == num_entries
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_options_auto(conn, request, test_frame1):
+    # use the set option
+    conn = request.getfixturevalue(conn)
+    with pd.option_context("io.sql.engine", "auto"):
+        with pandasSQL_builder(conn) as pandasSQL:
+            with pandasSQL.run_transaction():
+                assert pandasSQL.to_sql(test_frame1, "test_frame1") == 4
+                assert pandasSQL.has_table("test_frame1")
+
+        num_entries = len(test_frame1)
+        num_rows = count_rows(conn, "test_frame1")
+        assert num_rows == num_entries
+
+
+def test_options_get_engine():
+    pytest.importorskip("sqlalchemy")
+    assert isinstance(get_engine("sqlalchemy"), SQLAlchemyEngine)
+
+    with pd.option_context("io.sql.engine", "sqlalchemy"):
+        assert isinstance(get_engine("auto"), SQLAlchemyEngine)
+        assert isinstance(get_engine("sqlalchemy"), SQLAlchemyEngine)
+
+    with pd.option_context("io.sql.engine", "auto"):
+        assert isinstance(get_engine("auto"), SQLAlchemyEngine)
+        assert isinstance(get_engine("sqlalchemy"), SQLAlchemyEngine)
+
+
+def test_get_engine_auto_error_message():
+    # Expect different error messages from get_engine(engine="auto")
+    # if engines aren't installed vs. are installed but bad version
+    pass
+    # TODO(GH#36893) fill this in when we add more engines
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize("func", ["read_sql", "read_sql_query"])
+def test_read_sql_dtype_backend(
+    conn,
+    request,
+    string_storage,
+    func,
+    dtype_backend,
+    dtype_backend_data,
+    dtype_backend_expected,
+):
+    # GH#50048
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    table = "test"
+    df = dtype_backend_data
+    df.to_sql(name=table, con=conn, index=False, if_exists="replace")
+
+    with pd.option_context("mode.string_storage", string_storage):
+        result = getattr(pd, func)(
+            f"Select * from {table}", conn, dtype_backend=dtype_backend
+        )
+        expected = dtype_backend_expected(string_storage, dtype_backend, conn_name)
+
+    tm.assert_frame_equal(result, expected)
+
+    if "adbc" in conn_name:
+        # adbc does not support chunksize argument
+        request.applymarker(
+            pytest.mark.xfail(reason="adbc does not support chunksize argument")
+        )
+
+    with pd.option_context("mode.string_storage", string_storage):
+        iterator = getattr(pd, func)(
+            f"Select * from {table}",
+            con=conn,
+            dtype_backend=dtype_backend,
+            chunksize=3,
+        )
+        expected = dtype_backend_expected(string_storage, dtype_backend, conn_name)
+        for result in iterator:
+            tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize("func", ["read_sql", "read_sql_table"])
+def test_read_sql_dtype_backend_table(
+    conn,
+    request,
+    string_storage,
+    func,
+    dtype_backend,
+    dtype_backend_data,
+    dtype_backend_expected,
+):
+    if "sqlite" in conn and "adbc" not in conn:
+        request.applymarker(
+            pytest.mark.xfail(
+                reason=(
+                    "SQLite actually returns proper boolean values via "
+                    "read_sql_table, but before pytest refactor was skipped"
+                )
+            )
+        )
+    # GH#50048
+    conn_name = conn
+    conn = request.getfixturevalue(conn)
+    table = "test"
+    df = dtype_backend_data
+    df.to_sql(name=table, con=conn, index=False, if_exists="replace")
+
+    with pd.option_context("mode.string_storage", string_storage):
+        result = getattr(pd, func)(table, conn, dtype_backend=dtype_backend)
+        expected = dtype_backend_expected(string_storage, dtype_backend, conn_name)
+    tm.assert_frame_equal(result, expected)
+
+    if "adbc" in conn_name:
+        # adbc does not support chunksize argument
+        return
+
+    with pd.option_context("mode.string_storage", string_storage):
+        iterator = getattr(pd, func)(
+            table,
+            conn,
+            dtype_backend=dtype_backend,
+            chunksize=3,
+        )
+        expected = dtype_backend_expected(string_storage, dtype_backend, conn_name)
+        for result in iterator:
+            tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize("func", ["read_sql", "read_sql_table", "read_sql_query"])
+def test_read_sql_invalid_dtype_backend_table(conn, request, func, dtype_backend_data):
+    conn = request.getfixturevalue(conn)
+    table = "test"
+    df = dtype_backend_data
+    df.to_sql(name=table, con=conn, index=False, if_exists="replace")
+
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        getattr(pd, func)(table, conn, dtype_backend="numpy")
+
+
+@pytest.fixture
+def dtype_backend_data() -> DataFrame:
+    return DataFrame(
+        {
+            "a": Series([1, np.nan, 3], dtype="Int64"),
+            "b": Series([1, 2, 3], dtype="Int64"),
+            "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+            "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+            "e": [True, False, None],
+            "f": [True, False, True],
+            "g": ["a", "b", "c"],
+            "h": ["a", "b", None],
+        }
+    )
+
+
+@pytest.fixture
+def dtype_backend_expected():
+    def func(string_storage, dtype_backend, conn_name) -> DataFrame:
+        string_dtype: pd.StringDtype | pd.ArrowDtype
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+            string_dtype = pd.ArrowDtype(pa.string())
+        else:
+            string_dtype = pd.StringDtype(string_storage)
+
+        df = DataFrame(
+            {
+                "a": Series([1, np.nan, 3], dtype="Int64"),
+                "b": Series([1, 2, 3], dtype="Int64"),
+                "c": Series([1.5, np.nan, 2.5], dtype="Float64"),
+                "d": Series([1.5, 2.0, 2.5], dtype="Float64"),
+                "e": Series([True, False, pd.NA], dtype="boolean"),
+                "f": Series([True, False, True], dtype="boolean"),
+                "g": Series(["a", "b", "c"], dtype=string_dtype),
+                "h": Series(["a", "b", None], dtype=string_dtype),
+            }
+        )
+        if dtype_backend == "pyarrow":
+            pa = pytest.importorskip("pyarrow")
+
+            from pandas.arrays import ArrowExtensionArray
+
+            df = DataFrame(
+                {
+                    col: ArrowExtensionArray(pa.array(df[col], from_pandas=True))
+                    for col in df.columns
+                }
+            )
+
+        if "mysql" in conn_name or "sqlite" in conn_name:
+            if dtype_backend == "numpy_nullable":
+                df = df.astype({"e": "Int64", "f": "Int64"})
+            else:
+                df = df.astype({"e": "int64[pyarrow]", "f": "int64[pyarrow]"})
+
+        return df
+
+    return func
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+def test_chunksize_empty_dtypes(conn, request):
+    # GH#50245
+    if "adbc" in conn:
+        request.node.add_marker(
+            pytest.mark.xfail(reason="chunksize argument NotImplemented with ADBC")
+        )
+    conn = request.getfixturevalue(conn)
+    dtypes = {"a": "int64", "b": "object"}
+    df = DataFrame(columns=["a", "b"]).astype(dtypes)
+    expected = df.copy()
+    df.to_sql(name="test", con=conn, index=False, if_exists="replace")
+
+    for result in read_sql_query(
+        "SELECT * FROM test",
+        conn,
+        dtype=dtypes,
+        chunksize=1,
+    ):
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("conn", all_connectable)
+@pytest.mark.parametrize("dtype_backend", [lib.no_default, "numpy_nullable"])
+@pytest.mark.parametrize("func", ["read_sql", "read_sql_query"])
+def test_read_sql_dtype(conn, request, func, dtype_backend):
+    # GH#50797
+    conn = request.getfixturevalue(conn)
+    table = "test"
+    df = DataFrame({"a": [1, 2, 3], "b": 5})
+    df.to_sql(name=table, con=conn, index=False, if_exists="replace")
+
+    result = getattr(pd, func)(
+        f"Select * from {table}",
+        conn,
+        dtype={"a": np.float64},
+        dtype_backend=dtype_backend,
+    )
+    expected = DataFrame(
+        {
+            "a": Series([1, 2, 3], dtype=np.float64),
+            "b": Series(
+                [5, 5, 5],
+                dtype="int64" if not dtype_backend == "numpy_nullable" else "Int64",
+            ),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_bigint_warning(sqlite_engine):
+    conn = sqlite_engine
+    # test no warning for BIGINT (to support int64) is raised (GH7433)
+    df = DataFrame({"a": [1, 2]}, dtype="int64")
+    assert df.to_sql(name="test_bigintwarning", con=conn, index=False) == 2
+
+    with tm.assert_produces_warning(None):
+        sql.read_sql_table("test_bigintwarning", conn)
+
+
+def test_valueerror_exception(sqlite_engine):
+    conn = sqlite_engine
+    df = DataFrame({"col1": [1, 2], "col2": [3, 4]})
+    with pytest.raises(ValueError, match="Empty table name specified"):
+        df.to_sql(name="", con=conn, if_exists="replace", index=False)
+
+
+def test_row_object_is_named_tuple(sqlite_engine):
+    conn = sqlite_engine
+    # GH 40682
+    # Test for the is_named_tuple() function
+    # Placed here due to its usage of sqlalchemy
+
+    from sqlalchemy import (
+        Column,
+        Integer,
+        String,
+    )
+    from sqlalchemy.orm import (
+        declarative_base,
+        sessionmaker,
+    )
+
+    BaseModel = declarative_base()
+
+    class Test(BaseModel):
+        __tablename__ = "test_frame"
+        id = Column(Integer, primary_key=True)
+        string_column = Column(String(50))
+
+    with conn.begin():
+        BaseModel.metadata.create_all(conn)
+    Session = sessionmaker(bind=conn)
+    with Session() as session:
+        df = DataFrame({"id": [0, 1], "string_column": ["hello", "world"]})
+        assert (
+            df.to_sql(name="test_frame", con=conn, index=False, if_exists="replace")
+            == 2
+        )
+        session.commit()
+        test_query = session.query(Test.id, Test.string_column)
+        df = DataFrame(test_query)
+
+    assert list(df.columns) == ["id", "string_column"]
+
+
+def test_read_sql_string_inference(sqlite_engine):
+    conn = sqlite_engine
+    # GH#54430
+    table = "test"
+    df = DataFrame({"a": ["x", "y"]})
+    df.to_sql(table, con=conn, index=False, if_exists="replace")
+
+    with pd.option_context("future.infer_string", True):
+        result = read_sql_table(table, conn)
+
+    dtype = pd.StringDtype(na_value=np.nan)
+    expected = DataFrame(
+        {"a": ["x", "y"]}, dtype=dtype, columns=Index(["a"], dtype=dtype)
+    )
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_roundtripping_datetimes(sqlite_engine):
+    conn = sqlite_engine
+    # GH#54877
+    df = DataFrame({"t": [datetime(2020, 12, 31, 12)]}, dtype="datetime64[ns]")
+    df.to_sql("test", conn, if_exists="replace", index=False)
+    result = pd.read_sql("select * from test", conn).iloc[0, 0]
+    assert result == "2020-12-31 12:00:00.000000"
+
+
+@pytest.fixture
+def sqlite_builtin_detect_types():
+    with contextlib.closing(
+        sqlite3.connect(":memory:", detect_types=sqlite3.PARSE_DECLTYPES)
+    ) as closing_conn:
+        with closing_conn as conn:
+            yield conn
+
+
+def test_roundtripping_datetimes_detect_types(sqlite_builtin_detect_types):
+    # https://github.com/pandas-dev/pandas/issues/55554
+    conn = sqlite_builtin_detect_types
+    df = DataFrame({"t": [datetime(2020, 12, 31, 12)]}, dtype="datetime64[ns]")
+    df.to_sql("test", conn, if_exists="replace", index=False)
+    result = pd.read_sql("select * from test", conn).iloc[0, 0]
+    assert result == Timestamp("2020-12-31 12:00:00.000000")
+
+
+@pytest.mark.db
+def test_psycopg2_schema_support(postgresql_psycopg2_engine):
+    conn = postgresql_psycopg2_engine
+
+    # only test this for postgresql (schema's not supported in
+    # mysql/sqlite)
+    df = DataFrame({"col1": [1, 2], "col2": [0.1, 0.2], "col3": ["a", "n"]})
+
+    # create a schema
+    with conn.connect() as con:
+        with con.begin():
+            con.exec_driver_sql("DROP SCHEMA IF EXISTS other CASCADE;")
+            con.exec_driver_sql("CREATE SCHEMA other;")
+
+    # write dataframe to different schema's
+    assert df.to_sql(name="test_schema_public", con=conn, index=False) == 2
+    assert (
+        df.to_sql(
+            name="test_schema_public_explicit",
+            con=conn,
+            index=False,
+            schema="public",
+        )
+        == 2
+    )
+    assert (
+        df.to_sql(name="test_schema_other", con=conn, index=False, schema="other") == 2
+    )
+
+    # read dataframes back in
+    res1 = sql.read_sql_table("test_schema_public", conn)
+    tm.assert_frame_equal(df, res1)
+    res2 = sql.read_sql_table("test_schema_public_explicit", conn)
+    tm.assert_frame_equal(df, res2)
+    res3 = sql.read_sql_table("test_schema_public_explicit", conn, schema="public")
+    tm.assert_frame_equal(df, res3)
+    res4 = sql.read_sql_table("test_schema_other", conn, schema="other")
+    tm.assert_frame_equal(df, res4)
+    msg = "Table test_schema_other not found"
+    with pytest.raises(ValueError, match=msg):
+        sql.read_sql_table("test_schema_other", conn, schema="public")
+
+    # different if_exists options
+
+    # create a schema
+    with conn.connect() as con:
+        with con.begin():
+            con.exec_driver_sql("DROP SCHEMA IF EXISTS other CASCADE;")
+            con.exec_driver_sql("CREATE SCHEMA other;")
+
+    # write dataframe with different if_exists options
+    assert (
+        df.to_sql(name="test_schema_other", con=conn, schema="other", index=False) == 2
+    )
+    df.to_sql(
+        name="test_schema_other",
+        con=conn,
+        schema="other",
+        index=False,
+        if_exists="replace",
+    )
+    assert (
+        df.to_sql(
+            name="test_schema_other",
+            con=conn,
+            schema="other",
+            index=False,
+            if_exists="append",
+        )
+        == 2
+    )
+    res = sql.read_sql_table("test_schema_other", conn, schema="other")
+    tm.assert_frame_equal(concat([df, df], ignore_index=True), res)
+
+
+@pytest.mark.db
+def test_self_join_date_columns(postgresql_psycopg2_engine):
+    # GH 44421
+    conn = postgresql_psycopg2_engine
+    from sqlalchemy.sql import text
+
+    create_table = text(
+        """
+    CREATE TABLE person
+    (
+        id serial constraint person_pkey primary key,
+        created_dt timestamp with time zone
+    );
+
+    INSERT INTO person
+        VALUES (1, '2021-01-01T00:00:00Z');
+    """
+    )
+    with conn.connect() as con:
+        with con.begin():
+            con.execute(create_table)
+
+    sql_query = (
+        'SELECT * FROM "person" AS p1 INNER JOIN "person" AS p2 ON p1.id = p2.id;'
+    )
+    result = pd.read_sql(sql_query, conn)
+    expected = DataFrame(
+        [[1, Timestamp("2021", tz="UTC")] * 2], columns=["id", "created_dt"] * 2
+    )
+    expected["created_dt"] = expected["created_dt"].astype("M8[us, UTC]")
+    tm.assert_frame_equal(result, expected)
+
+    # Cleanup
+    with sql.SQLDatabase(conn, need_transaction=True) as pandasSQL:
+        pandasSQL.drop_table("person")
+
+
+def test_create_and_drop_table(sqlite_engine):
+    conn = sqlite_engine
+    temp_frame = DataFrame({"one": [1.0, 2.0, 3.0, 4.0], "two": [4.0, 3.0, 2.0, 1.0]})
+    with sql.SQLDatabase(conn) as pandasSQL:
+        with pandasSQL.run_transaction():
+            assert pandasSQL.to_sql(temp_frame, "drop_test_frame") == 4
+
+        assert pandasSQL.has_table("drop_test_frame")
+
+        with pandasSQL.run_transaction():
+            pandasSQL.drop_table("drop_test_frame")
+
+        assert not pandasSQL.has_table("drop_test_frame")
+
+
+def test_sqlite_datetime_date(sqlite_buildin):
+    conn = sqlite_buildin
+    df = DataFrame([date(2014, 1, 1), date(2014, 1, 2)], columns=["a"])
+    assert df.to_sql(name="test_date", con=conn, index=False) == 2
+    res = read_sql_query("SELECT * FROM test_date", conn)
+    # comes back as strings
+    tm.assert_frame_equal(res, df.astype(str))
+
+
+@pytest.mark.parametrize("tz_aware", [False, True])
+def test_sqlite_datetime_time(tz_aware, sqlite_buildin):
+    conn = sqlite_buildin
+    # test support for datetime.time, GH #8341
+    if not tz_aware:
+        tz_times = [time(9, 0, 0), time(9, 1, 30)]
+    else:
+        tz_dt = date_range("2013-01-01 09:00:00", periods=2, tz="US/Pacific")
+        tz_times = Series(tz_dt.to_pydatetime()).map(lambda dt: dt.timetz())
+
+    df = DataFrame(tz_times, columns=["a"])
+
+    assert df.to_sql(name="test_time", con=conn, index=False) == 2
+    res = read_sql_query("SELECT * FROM test_time", conn)
+    # comes back as strings
+    expected = df.map(lambda _: _.strftime("%H:%M:%S.%f"))
+    tm.assert_frame_equal(res, expected)
+
+
+def get_sqlite_column_type(conn, table, column):
+    recs = conn.execute(f"PRAGMA table_info({table})")
+    for cid, name, ctype, not_null, default, pk in recs:
+        if name == column:
+            return ctype
+    raise ValueError(f"Table {table}, column {column} not found")
+
+
+def test_sqlite_test_dtype(sqlite_buildin):
+    conn = sqlite_buildin
+    cols = ["A", "B"]
+    data = [(0.8, True), (0.9, None)]
+    df = DataFrame(data, columns=cols)
+    assert df.to_sql(name="dtype_test", con=conn) == 2
+    assert df.to_sql(name="dtype_test2", con=conn, dtype={"B": "STRING"}) == 2
+
+    # sqlite stores Boolean values as INTEGER
+    assert get_sqlite_column_type(conn, "dtype_test", "B") == "INTEGER"
+
+    assert get_sqlite_column_type(conn, "dtype_test2", "B") == "STRING"
+    msg = r"B \(<class 'bool'>\) not a string"
+    with pytest.raises(ValueError, match=msg):
+        df.to_sql(name="error", con=conn, dtype={"B": bool})
+
+    # single dtype
+    assert df.to_sql(name="single_dtype_test", con=conn, dtype="STRING") == 2
+    assert get_sqlite_column_type(conn, "single_dtype_test", "A") == "STRING"
+    assert get_sqlite_column_type(conn, "single_dtype_test", "B") == "STRING"
+
+
+def test_sqlite_notna_dtype(sqlite_buildin):
+    conn = sqlite_buildin
+    cols = {
+        "Bool": Series([True, None]),
+        "Date": Series([datetime(2012, 5, 1), None]),
+        "Int": Series([1, None], dtype="object"),
+        "Float": Series([1.1, None]),
+    }
+    df = DataFrame(cols)
+
+    tbl = "notna_dtype_test"
+    assert df.to_sql(name=tbl, con=conn) == 2
+
+    assert get_sqlite_column_type(conn, tbl, "Bool") == "INTEGER"
+    assert get_sqlite_column_type(conn, tbl, "Date") == "TIMESTAMP"
+    assert get_sqlite_column_type(conn, tbl, "Int") == "INTEGER"
+    assert get_sqlite_column_type(conn, tbl, "Float") == "REAL"
+
+
+def test_sqlite_illegal_names(sqlite_buildin):
+    # For sqlite, these should work fine
+    conn = sqlite_buildin
+    df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+
+    msg = "Empty table or column name specified"
+    with pytest.raises(ValueError, match=msg):
+        df.to_sql(name="", con=conn)
+
+    for ndx, weird_name in enumerate(
+        [
+            "test_weird_name]",
+            "test_weird_name[",
+            "test_weird_name`",
+            'test_weird_name"',
+            "test_weird_name'",
+            "_b.test_weird_name_01-30",
+            '"_b.test_weird_name_01-30"',
+            "99beginswithnumber",
+            "12345",
+            "\xe9",
+        ]
+    ):
+        assert df.to_sql(name=weird_name, con=conn) == 2
+        sql.table_exists(weird_name, conn)
+
+        df2 = DataFrame([[1, 2], [3, 4]], columns=["a", weird_name])
+        c_tbl = f"test_weird_col_name{ndx:d}"
+        assert df2.to_sql(name=c_tbl, con=conn) == 2
+        sql.table_exists(c_tbl, conn)
+
+
+def format_query(sql, *args):
+    _formatters = {
+        datetime: "'{}'".format,
+        str: "'{}'".format,
+        np.str_: "'{}'".format,
+        bytes: "'{}'".format,
+        float: "{:.8f}".format,
+        int: "{:d}".format,
+        type(None): lambda x: "NULL",
+        np.float64: "{:.10f}".format,
+        bool: "'{!s}'".format,
+    }
+    processed_args = []
+    for arg in args:
+        if isinstance(arg, float) and isna(arg):
+            arg = None
+
+        formatter = _formatters[type(arg)]
+        processed_args.append(formatter(arg))
+
+    return sql % tuple(processed_args)
+
+
+def tquery(query, con=None):
+    """Replace removed sql.tquery function"""
+    with sql.pandasSQL_builder(con) as pandas_sql:
+        res = pandas_sql.execute(query).fetchall()
+    return None if res is None else list(res)
+
+
+def test_xsqlite_basic(sqlite_buildin):
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    assert sql.to_sql(frame, name="test_table", con=sqlite_buildin, index=False) == 10
+    result = sql.read_sql("select * from test_table", sqlite_buildin)
+
+    # HACK! Change this once indexes are handled properly.
+    result.index = frame.index
+
+    expected = frame
+    tm.assert_frame_equal(result, frame)
+
+    frame["txt"] = ["a"] * len(frame)
+    frame2 = frame.copy()
+    new_idx = Index(np.arange(len(frame2)), dtype=np.int64) + 10
+    frame2["Idx"] = new_idx.copy()
+    assert sql.to_sql(frame2, name="test_table2", con=sqlite_buildin, index=False) == 10
+    result = sql.read_sql("select * from test_table2", sqlite_buildin, index_col="Idx")
+    expected = frame.copy()
+    expected.index = new_idx
+    expected.index.name = "Idx"
+    tm.assert_frame_equal(expected, result)
+
+
+def test_xsqlite_write_row_by_row(sqlite_buildin):
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    frame.iloc[0, 0] = np.nan
+    create_sql = sql.get_schema(frame, "test")
+    cur = sqlite_buildin.cursor()
+    cur.execute(create_sql)
+
+    ins = "INSERT INTO test VALUES (%s, %s, %s, %s)"
+    for _, row in frame.iterrows():
+        fmt_sql = format_query(ins, *row)
+        tquery(fmt_sql, con=sqlite_buildin)
+
+    sqlite_buildin.commit()
+
+    result = sql.read_sql("select * from test", con=sqlite_buildin)
+    result.index = frame.index
+    tm.assert_frame_equal(result, frame, rtol=1e-3)
+
+
+def test_xsqlite_execute(sqlite_buildin):
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    create_sql = sql.get_schema(frame, "test")
+    cur = sqlite_buildin.cursor()
+    cur.execute(create_sql)
+    ins = "INSERT INTO test VALUES (?, ?, ?, ?)"
+
+    row = frame.iloc[0]
+    with sql.pandasSQL_builder(sqlite_buildin) as pandas_sql:
+        pandas_sql.execute(ins, tuple(row))
+    sqlite_buildin.commit()
+
+    result = sql.read_sql("select * from test", sqlite_buildin)
+    result.index = frame.index[:1]
+    tm.assert_frame_equal(result, frame[:1])
+
+
+def test_xsqlite_schema(sqlite_buildin):
+    frame = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    create_sql = sql.get_schema(frame, "test")
+    lines = create_sql.splitlines()
+    for line in lines:
+        tokens = line.split(" ")
+        if len(tokens) == 2 and tokens[0] == "A":
+            assert tokens[1] == "DATETIME"
+
+    create_sql = sql.get_schema(frame, "test", keys=["A", "B"])
+    lines = create_sql.splitlines()
+    assert 'PRIMARY KEY ("A", "B")' in create_sql
+    cur = sqlite_buildin.cursor()
+    cur.execute(create_sql)
+
+
+def test_xsqlite_execute_fail(sqlite_buildin):
+    create_sql = """
+    CREATE TABLE test
+    (
+    a TEXT,
+    b TEXT,
+    c REAL,
+    PRIMARY KEY (a, b)
+    );
+    """
+    cur = sqlite_buildin.cursor()
+    cur.execute(create_sql)
+
+    with sql.pandasSQL_builder(sqlite_buildin) as pandas_sql:
+        pandas_sql.execute("INSERT INTO test VALUES('foo', 'bar', 1.234)")
+        pandas_sql.execute("INSERT INTO test VALUES('foo', 'baz', 2.567)")
+
+        with pytest.raises(sql.DatabaseError, match="Execution failed on sql"):
+            pandas_sql.execute("INSERT INTO test VALUES('foo', 'bar', 7)")
+
+
+def test_xsqlite_execute_closed_connection():
+    create_sql = """
+    CREATE TABLE test
+    (
+    a TEXT,
+    b TEXT,
+    c REAL,
+    PRIMARY KEY (a, b)
+    );
+    """
+    with contextlib.closing(sqlite3.connect(":memory:")) as conn:
+        cur = conn.cursor()
+        cur.execute(create_sql)
+
+        with sql.pandasSQL_builder(conn) as pandas_sql:
+            pandas_sql.execute("INSERT INTO test VALUES('foo', 'bar', 1.234)")
+
+    msg = "Cannot operate on a closed database."
+    with pytest.raises(sqlite3.ProgrammingError, match=msg):
+        tquery("select * from test", con=conn)
+
+
+def test_xsqlite_keyword_as_column_names(sqlite_buildin):
+    df = DataFrame({"From": np.ones(5)})
+    assert sql.to_sql(df, con=sqlite_buildin, name="testkeywords", index=False) == 5
+
+
+def test_xsqlite_onecolumn_of_integer(sqlite_buildin):
+    # GH 3628
+    # a column_of_integers dataframe should transfer well to sql
+
+    mono_df = DataFrame([1, 2], columns=["c0"])
+    assert sql.to_sql(mono_df, con=sqlite_buildin, name="mono_df", index=False) == 2
+    # computing the sum via sql
+    con_x = sqlite_buildin
+    the_sum = sum(my_c0[0] for my_c0 in con_x.execute("select * from mono_df"))
+    # it should not fail, and gives 3 ( Issue #3628 )
+    assert the_sum == 3
+
+    result = sql.read_sql("select * from mono_df", con_x)
+    tm.assert_frame_equal(result, mono_df)
+
+
+def test_xsqlite_if_exists(sqlite_buildin):
+    df_if_exists_1 = DataFrame({"col1": [1, 2], "col2": ["A", "B"]})
+    df_if_exists_2 = DataFrame({"col1": [3, 4, 5], "col2": ["C", "D", "E"]})
+    table_name = "table_if_exists"
+    sql_select = f"SELECT * FROM {table_name}"
+
+    msg = "'notvalidvalue' is not valid for if_exists"
+    with pytest.raises(ValueError, match=msg):
+        sql.to_sql(
+            frame=df_if_exists_1,
+            con=sqlite_buildin,
+            name=table_name,
+            if_exists="notvalidvalue",
+        )
+    drop_table(table_name, sqlite_buildin)
+
+    # test if_exists='fail'
+    sql.to_sql(
+        frame=df_if_exists_1, con=sqlite_buildin, name=table_name, if_exists="fail"
+    )
+    msg = "Table 'table_if_exists' already exists"
+    with pytest.raises(ValueError, match=msg):
+        sql.to_sql(
+            frame=df_if_exists_1,
+            con=sqlite_buildin,
+            name=table_name,
+            if_exists="fail",
+        )
+    # test if_exists='replace'
+    sql.to_sql(
+        frame=df_if_exists_1,
+        con=sqlite_buildin,
+        name=table_name,
+        if_exists="replace",
+        index=False,
+    )
+    assert tquery(sql_select, con=sqlite_buildin) == [(1, "A"), (2, "B")]
+    assert (
+        sql.to_sql(
+            frame=df_if_exists_2,
+            con=sqlite_buildin,
+            name=table_name,
+            if_exists="replace",
+            index=False,
+        )
+        == 3
+    )
+    assert tquery(sql_select, con=sqlite_buildin) == [(3, "C"), (4, "D"), (5, "E")]
+    drop_table(table_name, sqlite_buildin)
+
+    # test if_exists='append'
+    assert (
+        sql.to_sql(
+            frame=df_if_exists_1,
+            con=sqlite_buildin,
+            name=table_name,
+            if_exists="fail",
+            index=False,
+        )
+        == 2
+    )
+    assert tquery(sql_select, con=sqlite_buildin) == [(1, "A"), (2, "B")]
+    assert (
+        sql.to_sql(
+            frame=df_if_exists_2,
+            con=sqlite_buildin,
+            name=table_name,
+            if_exists="append",
+            index=False,
+        )
+        == 3
+    )
+    assert tquery(sql_select, con=sqlite_buildin) == [
+        (1, "A"),
+        (2, "B"),
+        (3, "C"),
+        (4, "D"),
+        (5, "E"),
+    ]
+    drop_table(table_name, sqlite_buildin)
diff --git a/pandas/pandas/tests/io/test_stata.py b/pandas/pandas/tests/io/test_stata.py
new file mode 100644
index 0000000000000..9288b98d79fbe
--- /dev/null
+++ b/pandas/pandas/tests/io/test_stata.py
@@ -0,0 +1,2589 @@
+import bz2
+import datetime as dt
+from datetime import datetime
+import gzip
+import io
+import itertools
+import os
+import string
+import struct
+import tarfile
+import zipfile
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import CategoricalDtype
+import pandas._testing as tm
+from pandas.core.frame import (
+    DataFrame,
+    Series,
+)
+
+from pandas.io.parsers import read_csv
+from pandas.io.stata import (
+    CategoricalConversionWarning,
+    InvalidColumnName,
+    PossiblePrecisionLoss,
+    StataMissingValue,
+    StataReader,
+    StataWriter,
+    StataWriterUTF8,
+    ValueLabelTypeMismatch,
+    read_stata,
+)
+
+
+@pytest.fixture
+def mixed_frame():
+    return DataFrame(
+        {
+            "a": [1, 2, 3, 4],
+            "b": [1.0, 3.0, 27.0, 81.0],
+            "c": ["Atlanta", "Birmingham", "Cincinnati", "Detroit"],
+        }
+    )
+
+
+@pytest.fixture
+def parsed_114(datapath):
+    dta14_114 = datapath("io", "data", "stata", "stata5_114.dta")
+    parsed_114 = read_stata(dta14_114, convert_dates=True)
+    parsed_114.index.name = "index"
+    return parsed_114
+
+
+class TestStata:
+    def read_dta(self, file):
+        # Legacy default reader configuration
+        return read_stata(file, convert_dates=True)
+
+    def read_csv(self, file):
+        return read_csv(file, parse_dates=True)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_read_empty_dta(self, version, temp_file):
+        empty_ds = DataFrame(columns=["unit"])
+        # GH 7369, make sure can read a 0-obs dta file
+        path = temp_file
+        empty_ds.to_stata(path, write_index=False, version=version)
+        empty_ds2 = read_stata(path)
+        tm.assert_frame_equal(empty_ds, empty_ds2)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_read_empty_dta_with_dtypes(self, version, temp_file):
+        # GH 46240
+        # Fixing above bug revealed that types are not correctly preserved when
+        # writing empty DataFrames
+        empty_df_typed = DataFrame(
+            {
+                "i8": np.array([0], dtype=np.int8),
+                "i16": np.array([0], dtype=np.int16),
+                "i32": np.array([0], dtype=np.int32),
+                "i64": np.array([0], dtype=np.int64),
+                "u8": np.array([0], dtype=np.uint8),
+                "u16": np.array([0], dtype=np.uint16),
+                "u32": np.array([0], dtype=np.uint32),
+                "u64": np.array([0], dtype=np.uint64),
+                "f32": np.array([0], dtype=np.float32),
+                "f64": np.array([0], dtype=np.float64),
+            }
+        )
+        # GH 7369, make sure can read a 0-obs dta file
+        path = temp_file
+        empty_df_typed.to_stata(path, write_index=False, version=version)
+        empty_reread = read_stata(path)
+
+        expected = empty_df_typed
+        # No uint# support. Downcast since values in range for int#
+        expected["u8"] = expected["u8"].astype(np.int8)
+        expected["u16"] = expected["u16"].astype(np.int16)
+        expected["u32"] = expected["u32"].astype(np.int32)
+        # No int64 supported at all. Downcast since values in range for int32
+        expected["u64"] = expected["u64"].astype(np.int32)
+        expected["i64"] = expected["i64"].astype(np.int32)
+
+        tm.assert_frame_equal(expected, empty_reread)
+        tm.assert_series_equal(expected.dtypes, empty_reread.dtypes)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_read_index_col_none(self, version, temp_file):
+        df = DataFrame({"a": range(5), "b": ["b1", "b2", "b3", "b4", "b5"]})
+        # GH 7369, make sure can read a 0-obs dta file
+        path = temp_file
+        df.to_stata(path, write_index=False, version=version)
+        read_df = read_stata(path)
+
+        assert isinstance(read_df.index, pd.RangeIndex)
+        expected = df
+        expected["a"] = expected["a"].astype(np.int32)
+        tm.assert_frame_equal(read_df, expected, check_index_type=True)
+
+    @pytest.mark.parametrize(
+        "version", [102, 103, 104, 105, 108, 110, 111, 113, 114, 115, 117, 118, 119]
+    )
+    def test_read_dta1(self, version, datapath):
+        file = datapath("io", "data", "stata", f"stata1_{version}.dta")
+        parsed = self.read_dta(file)
+
+        # Pandas uses np.nan as missing value.
+        # Thus, all columns will be of type float, regardless of their name.
+        expected = DataFrame(
+            [(np.nan, np.nan, np.nan, np.nan, np.nan)],
+            columns=["float_miss", "double_miss", "byte_miss", "int_miss", "long_miss"],
+        )
+
+        # this is an oddity as really the nan should be float64, but
+        # the casting doesn't fail so need to match stata here
+        expected["float_miss"] = expected["float_miss"].astype(np.float32)
+
+        # Column names too long for older Stata formats
+        if version <= 108:
+            expected = expected.rename(
+                columns={
+                    "float_miss": "f_miss",
+                    "double_miss": "d_miss",
+                    "byte_miss": "b_miss",
+                    "int_miss": "i_miss",
+                    "long_miss": "l_miss",
+                }
+            )
+
+        tm.assert_frame_equal(parsed, expected)
+
+    def test_read_dta2(self, datapath):
+        expected = DataFrame.from_records(
+            [
+                (
+                    datetime(2006, 11, 19, 23, 13, 20),
+                    1479596223000,
+                    datetime(2010, 1, 20),
+                    datetime(2010, 1, 8),
+                    datetime(2010, 1, 1),
+                    datetime(1974, 7, 1),
+                    datetime(2010, 1, 1),
+                    datetime(2010, 1, 1),
+                ),
+                (
+                    datetime(1959, 12, 31, 20, 3, 20),
+                    -1479590,
+                    datetime(1953, 10, 2),
+                    datetime(1948, 6, 10),
+                    datetime(1955, 1, 1),
+                    datetime(1955, 7, 1),
+                    datetime(1955, 1, 1),
+                    datetime(2, 1, 1),
+                ),
+                (pd.NaT, pd.NaT, pd.NaT, pd.NaT, pd.NaT, pd.NaT, pd.NaT, pd.NaT),
+            ],
+            columns=[
+                "datetime_c",
+                "datetime_big_c",
+                "date",
+                "weekly_date",
+                "monthly_date",
+                "quarterly_date",
+                "half_yearly_date",
+                "yearly_date",
+            ],
+        )
+        # TODO(GH#55564): just pass M8[s] to the constructor
+        expected["datetime_c"] = expected["datetime_c"].astype("M8[ms]")
+        expected["date"] = expected["date"].astype("M8[s]")
+        expected["weekly_date"] = expected["weekly_date"].astype("M8[s]")
+        expected["monthly_date"] = expected["monthly_date"].astype("M8[s]")
+        expected["quarterly_date"] = expected["quarterly_date"].astype("M8[s]")
+        expected["half_yearly_date"] = expected["half_yearly_date"].astype("M8[s]")
+        expected["yearly_date"] = expected["yearly_date"].astype("M8[s]")
+
+        path1 = datapath("io", "data", "stata", "stata2_114.dta")
+        path2 = datapath("io", "data", "stata", "stata2_115.dta")
+        path3 = datapath("io", "data", "stata", "stata2_117.dta")
+
+        msg = "Leaving in Stata Internal Format"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            parsed_114 = self.read_dta(path1)
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            parsed_115 = self.read_dta(path2)
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            parsed_117 = self.read_dta(path3)
+            # FIXME: don't leave commented-out
+            # 113 is buggy due to limits of date format support in Stata
+            # parsed_113 = self.read_dta(
+            # datapath("io", "data", "stata", "stata2_113.dta")
+            # )
+
+        # FIXME: don't leave commented-out
+        # buggy test because of the NaT comparison on certain platforms
+        # Format 113 test fails since it does not support tc and tC formats
+        # tm.assert_frame_equal(parsed_113, expected)
+        tm.assert_frame_equal(parsed_114, expected)
+        tm.assert_frame_equal(parsed_115, expected)
+        tm.assert_frame_equal(parsed_117, expected)
+
+    @pytest.mark.parametrize(
+        "file", ["stata3_113", "stata3_114", "stata3_115", "stata3_117"]
+    )
+    def test_read_dta3(self, file, datapath):
+        file = datapath("io", "data", "stata", f"{file}.dta")
+        parsed = self.read_dta(file)
+
+        # match stata here
+        expected = self.read_csv(datapath("io", "data", "stata", "stata3.csv"))
+        expected = expected.astype(np.float32)
+        expected["year"] = expected["year"].astype(np.int16)
+        expected["quarter"] = expected["quarter"].astype(np.int8)
+
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.mark.parametrize("version", [110, 111, 113, 114, 115, 117])
+    def test_read_dta4(self, version, datapath):
+        file = datapath("io", "data", "stata", f"stata4_{version}.dta")
+        parsed = self.read_dta(file)
+
+        expected = DataFrame.from_records(
+            [
+                ["one", "ten", "one", "one", "one"],
+                ["two", "nine", "two", "two", "two"],
+                ["three", "eight", "three", "three", "three"],
+                ["four", "seven", 4, "four", "four"],
+                ["five", "six", 5, np.nan, "five"],
+                ["six", "five", 6, np.nan, "six"],
+                ["seven", "four", 7, np.nan, "seven"],
+                ["eight", "three", 8, np.nan, "eight"],
+                ["nine", "two", 9, np.nan, "nine"],
+                ["ten", "one", "ten", np.nan, "ten"],
+            ],
+            columns=[
+                "fully_labeled",
+                "fully_labeled2",
+                "incompletely_labeled",
+                "labeled_with_missings",
+                "float_labelled",
+            ],
+        )
+
+        # these are all categoricals
+        for col in expected:
+            orig = expected[col].copy()
+
+            categories = np.asarray(expected["fully_labeled"][orig.notna()])
+            if col == "incompletely_labeled":
+                categories = orig
+
+            cat = orig.astype("category")._values
+            cat = cat.set_categories(categories, ordered=True)
+            cat.categories.rename(None, inplace=True)
+
+            expected[col] = cat
+
+        # stata doesn't save .category metadata
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.mark.parametrize("version", [102, 103, 104, 105, 108])
+    def test_readold_dta4(self, version, datapath):
+        # This test is the same as test_read_dta4 above except that the columns
+        # had to be renamed to match the restrictions in older file format
+        file = datapath("io", "data", "stata", f"stata4_{version}.dta")
+        parsed = self.read_dta(file)
+
+        expected = DataFrame.from_records(
+            [
+                ["one", "ten", "one", "one", "one"],
+                ["two", "nine", "two", "two", "two"],
+                ["three", "eight", "three", "three", "three"],
+                ["four", "seven", 4, "four", "four"],
+                ["five", "six", 5, np.nan, "five"],
+                ["six", "five", 6, np.nan, "six"],
+                ["seven", "four", 7, np.nan, "seven"],
+                ["eight", "three", 8, np.nan, "eight"],
+                ["nine", "two", 9, np.nan, "nine"],
+                ["ten", "one", "ten", np.nan, "ten"],
+            ],
+            columns=[
+                "fulllab",
+                "fulllab2",
+                "incmplab",
+                "misslab",
+                "floatlab",
+            ],
+        )
+
+        # these are all categoricals
+        for col in expected:
+            orig = expected[col].copy()
+
+            categories = np.asarray(expected["fulllab"][orig.notna()])
+            if col == "incmplab":
+                categories = orig
+
+            cat = orig.astype("category")._values
+            cat = cat.set_categories(categories, ordered=True)
+            cat.categories.rename(None, inplace=True)
+
+            expected[col] = cat
+
+        # stata doesn't save .category metadata
+        tm.assert_frame_equal(parsed, expected)
+
+    # File containing strls
+    @pytest.mark.parametrize(
+        "file",
+        [
+            "stata12_117",
+            "stata12_be_117",
+            "stata12_118",
+            "stata12_be_118",
+            "stata12_119",
+            "stata12_be_119",
+        ],
+    )
+    def test_read_dta_strl(self, file, datapath):
+        parsed = self.read_dta(datapath("io", "data", "stata", f"{file}.dta"))
+        expected = DataFrame.from_records(
+            [
+                [1, "abc", "abcdefghi"],
+                [3, "cba", "qwertywertyqwerty"],
+                [93, "", "strl"],
+            ],
+            columns=["x", "y", "z"],
+        )
+
+        tm.assert_frame_equal(parsed, expected, check_dtype=False)
+
+    # 117 is not included in this list as it uses ASCII strings
+    @pytest.mark.parametrize(
+        "file",
+        [
+            "stata14_118",
+            "stata14_be_118",
+            "stata14_119",
+            "stata14_be_119",
+        ],
+    )
+    def test_read_dta118_119(self, file, datapath):
+        parsed_118 = self.read_dta(datapath("io", "data", "stata", f"{file}.dta"))
+        parsed_118["Bytes"] = parsed_118["Bytes"].astype("O")
+        expected = DataFrame.from_records(
+            [
+                ["Cat", "Bogota", "Bogotá", 1, 1.0, "option b Ünicode", 1.0],
+                ["Dog", "Boston", "Uzunköprü", np.nan, np.nan, np.nan, np.nan],
+                ["Plane", "Rome", "Tromsø", 0, 0.0, "option a", 0.0],
+                ["Potato", "Tokyo", "Elâzığ", -4, 4.0, 4, 4],  # noqa: RUF001
+                ["", "", "", 0, 0.3332999, "option a", 1 / 3.0],
+            ],
+            columns=[
+                "Things",
+                "Cities",
+                "Unicode_Cities_Strl",
+                "Ints",
+                "Floats",
+                "Bytes",
+                "Longs",
+            ],
+        )
+        expected["Floats"] = expected["Floats"].astype(np.float32)
+        for col in parsed_118.columns:
+            tm.assert_almost_equal(parsed_118[col], expected[col])
+
+        with StataReader(datapath("io", "data", "stata", f"{file}.dta")) as rdr:
+            vl = rdr.variable_labels()
+            vl_expected = {
+                "Unicode_Cities_Strl": "Here are some strls with Ünicode chars",
+                "Longs": "long data",
+                "Things": "Here are some things",
+                "Bytes": "byte data",
+                "Ints": "int data",
+                "Cities": "Here are some cities",
+                "Floats": "float data",
+            }
+            tm.assert_dict_equal(vl, vl_expected)
+
+            assert rdr.data_label == "This is a  Ünicode data label"
+
+    def test_read_write_dta5(self, temp_file):
+        original = DataFrame(
+            [(np.nan, np.nan, np.nan, np.nan, np.nan)],
+            columns=["float_miss", "double_miss", "byte_miss", "int_miss", "long_miss"],
+        )
+        original.index.name = "index"
+
+        path = temp_file
+        original.to_stata(path, convert_dates=None)
+        written_and_read_again = self.read_dta(path)
+
+        expected = original
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    def test_write_dta6(self, datapath, temp_file):
+        original = self.read_csv(datapath("io", "data", "stata", "stata3.csv"))
+        original.index.name = "index"
+        original.index = original.index.astype(np.int32)
+        original["year"] = original["year"].astype(np.int32)
+        original["quarter"] = original["quarter"].astype(np.int32)
+
+        path = temp_file
+        original.to_stata(path, convert_dates=None)
+        written_and_read_again = self.read_dta(path)
+        tm.assert_frame_equal(
+            written_and_read_again.set_index("index"),
+            original,
+            check_index_type=False,
+        )
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_read_write_dta10(self, version, temp_file, using_infer_string):
+        original = DataFrame(
+            data=[["string", "object", 1, 1.1, np.datetime64("2003-12-25")]],
+            columns=["string", "object", "integer", "floating", "datetime"],
+        )
+        original["object"] = Series(original["object"], dtype=object)
+        original.index.name = "index"
+        original.index = original.index.astype(np.int32)
+        original["integer"] = original["integer"].astype(np.int32)
+
+        path = temp_file
+        original.to_stata(path, convert_dates={"datetime": "tc"}, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        expected = original.copy()
+        # "tc" convert_dates means we store in ms
+        expected["datetime"] = expected["datetime"].astype("M8[ms]")
+        if using_infer_string:
+            expected["object"] = expected["object"].astype("str")
+
+        tm.assert_frame_equal(
+            written_and_read_again.set_index("index"),
+            expected,
+        )
+
+    def test_stata_doc_examples(self, temp_file):
+        path = temp_file
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)), columns=list("AB")
+        )
+        df.to_stata(path)
+
+    def test_write_preserves_original(self, temp_file):
+        # 9795
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)), columns=list("abcd")
+        )
+        df.loc[2, "a":"c"] = np.nan
+        df_copy = df.copy()
+        path = temp_file
+        df.to_stata(path, write_index=False)
+        tm.assert_frame_equal(df, df_copy)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_encoding(self, version, datapath, temp_file):
+        # GH 4626, proper encoding handling
+        raw = read_stata(datapath("io", "data", "stata", "stata1_encoding.dta"))
+        encoded = read_stata(datapath("io", "data", "stata", "stata1_encoding.dta"))
+        result = encoded.kreis1849[0]
+
+        expected = raw.kreis1849[0]
+        assert result == expected
+        assert isinstance(result, str)
+
+        path = temp_file
+        encoded.to_stata(path, write_index=False, version=version)
+        reread_encoded = read_stata(path)
+        tm.assert_frame_equal(encoded, reread_encoded)
+
+    def test_read_write_dta11(self, temp_file):
+        original = DataFrame(
+            [(1, 2, 3, 4)],
+            columns=[
+                "good",
+                "b\u00e4d",
+                "8number",
+                "astringwithmorethan32characters______",
+            ],
+        )
+        formatted = DataFrame(
+            [(1, 2, 3, 4)],
+            columns=["good", "b_d", "_8number", "astringwithmorethan32characters_"],
+        )
+        formatted.index.name = "index"
+        formatted = formatted.astype(np.int32)
+
+        path = temp_file
+        msg = "Not all pandas column names were valid Stata variable names"
+        with tm.assert_produces_warning(InvalidColumnName, match=msg):
+            original.to_stata(path, convert_dates=None)
+
+        written_and_read_again = self.read_dta(path)
+
+        expected = formatted
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_read_write_dta12(self, version, temp_file):
+        original = DataFrame(
+            [(1, 2, 3, 4, 5, 6)],
+            columns=[
+                "astringwithmorethan32characters_1",
+                "astringwithmorethan32characters_2",
+                "+",
+                "-",
+                "short",
+                "delete",
+            ],
+        )
+        formatted = DataFrame(
+            [(1, 2, 3, 4, 5, 6)],
+            columns=[
+                "astringwithmorethan32characters_",
+                "_0astringwithmorethan32character",
+                "_",
+                "_1_",
+                "_short",
+                "_delete",
+            ],
+        )
+        formatted.index.name = "index"
+        formatted = formatted.astype(np.int32)
+
+        path = temp_file
+        msg = "Not all pandas column names were valid Stata variable names"
+        with tm.assert_produces_warning(InvalidColumnName, match=msg):
+            original.to_stata(path, convert_dates=None, version=version)
+            # should get a warning for that format.
+
+        written_and_read_again = self.read_dta(path)
+
+        expected = formatted
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    def test_read_write_dta13(self, temp_file):
+        s1 = Series(2**9, dtype=np.int16)
+        s2 = Series(2**17, dtype=np.int32)
+        s3 = Series(2**33, dtype=np.int64)
+        original = DataFrame({"int16": s1, "int32": s2, "int64": s3})
+        original.index.name = "index"
+
+        formatted = original
+        formatted["int64"] = formatted["int64"].astype(np.float64)
+
+        path = temp_file
+        original.to_stata(path)
+        written_and_read_again = self.read_dta(path)
+
+        expected = formatted
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    @pytest.mark.parametrize(
+        "file", ["stata5_113", "stata5_114", "stata5_115", "stata5_117"]
+    )
+    def test_read_write_reread_dta14(
+        self, file, parsed_114, version, datapath, temp_file
+    ):
+        file = datapath("io", "data", "stata", f"{file}.dta")
+        parsed = self.read_dta(file)
+        parsed.index.name = "index"
+
+        tm.assert_frame_equal(parsed_114, parsed)
+
+        path = temp_file
+        parsed_114.to_stata(path, convert_dates={"date_td": "td"}, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        expected = parsed_114.copy()
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    @pytest.mark.parametrize(
+        "file", ["stata6_113", "stata6_114", "stata6_115", "stata6_117"]
+    )
+    def test_read_write_reread_dta15(self, file, datapath):
+        expected = self.read_csv(datapath("io", "data", "stata", "stata6.csv"))
+        expected["byte_"] = expected["byte_"].astype(np.int8)
+        expected["int_"] = expected["int_"].astype(np.int16)
+        expected["long_"] = expected["long_"].astype(np.int32)
+        expected["float_"] = expected["float_"].astype(np.float32)
+        expected["double_"] = expected["double_"].astype(np.float64)
+
+        # TODO(GH#55564): directly cast to M8[s]
+        arr = expected["date_td"].astype("Period[D]")._values.asfreq("s", how="S")
+        expected["date_td"] = arr.view("M8[s]")
+
+        file = datapath("io", "data", "stata", f"{file}.dta")
+        parsed = self.read_dta(file)
+
+        tm.assert_frame_equal(expected, parsed)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_timestamp_and_label(self, version, temp_file):
+        original = DataFrame([(1,)], columns=["variable"])
+        time_stamp = datetime(2000, 2, 29, 14, 21)
+        data_label = "This is a data file."
+        path = temp_file
+        original.to_stata(
+            path, time_stamp=time_stamp, data_label=data_label, version=version
+        )
+
+        with StataReader(path) as reader:
+            assert reader.time_stamp == "29 Feb 2000 14:21"
+            assert reader.data_label == data_label
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_invalid_timestamp(self, version, temp_file):
+        original = DataFrame([(1,)], columns=["variable"])
+        time_stamp = "01 Jan 2000, 00:00:00"
+        path = temp_file
+        msg = "time_stamp should be datetime type"
+        with pytest.raises(ValueError, match=msg):
+            original.to_stata(path, time_stamp=time_stamp, version=version)
+        assert not os.path.isfile(path)
+
+    def test_numeric_column_names(self, temp_file):
+        original = DataFrame(np.reshape(np.arange(25.0), (5, 5)))
+        original.index.name = "index"
+        path = temp_file
+        # should get a warning for that format.
+        msg = "Not all pandas column names were valid Stata variable names"
+        with tm.assert_produces_warning(InvalidColumnName, match=msg):
+            original.to_stata(path)
+
+        written_and_read_again = self.read_dta(path)
+
+        written_and_read_again = written_and_read_again.set_index("index")
+        columns = list(written_and_read_again.columns)
+        convert_col_name = lambda x: int(x[1])
+        written_and_read_again.columns = map(convert_col_name, columns)
+
+        expected = original
+        tm.assert_frame_equal(expected, written_and_read_again)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_nan_to_missing_value(self, version, temp_file):
+        s1 = Series(np.arange(4.0), dtype=np.float32)
+        s2 = Series(np.arange(4.0), dtype=np.float64)
+        s1[::2] = np.nan
+        s2[1::2] = np.nan
+        original = DataFrame({"s1": s1, "s2": s2})
+        original.index.name = "index"
+
+        path = temp_file
+        original.to_stata(path, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        written_and_read_again = written_and_read_again.set_index("index")
+        expected = original
+        tm.assert_frame_equal(written_and_read_again, expected)
+
+    def test_no_index(self, temp_file):
+        columns = ["x", "y"]
+        original = DataFrame(np.reshape(np.arange(10.0), (5, 2)), columns=columns)
+        original.index.name = "index_not_written"
+        path = temp_file
+        original.to_stata(path, write_index=False)
+        written_and_read_again = self.read_dta(path)
+        with pytest.raises(KeyError, match=original.index.name):
+            written_and_read_again["index_not_written"]
+
+    def test_string_no_dates(self, temp_file):
+        s1 = Series(["a", "A longer string"])
+        s2 = Series([1.0, 2.0], dtype=np.float64)
+        original = DataFrame({"s1": s1, "s2": s2})
+        original.index.name = "index"
+        path = temp_file
+        original.to_stata(path)
+        written_and_read_again = self.read_dta(path)
+
+        expected = original
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    def test_large_value_conversion(self, temp_file):
+        s0 = Series([1, 99], dtype=np.int8)
+        s1 = Series([1, 127], dtype=np.int8)
+        s2 = Series([1, 2**15 - 1], dtype=np.int16)
+        s3 = Series([1, 2**63 - 1], dtype=np.int64)
+        original = DataFrame({"s0": s0, "s1": s1, "s2": s2, "s3": s3})
+        original.index.name = "index"
+        path = temp_file
+        with tm.assert_produces_warning(PossiblePrecisionLoss, match="from int64 to"):
+            original.to_stata(path)
+
+        written_and_read_again = self.read_dta(path)
+
+        modified = original
+        modified["s1"] = Series(modified["s1"], dtype=np.int16)
+        modified["s2"] = Series(modified["s2"], dtype=np.int32)
+        modified["s3"] = Series(modified["s3"], dtype=np.float64)
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), modified)
+
+    def test_dates_invalid_column(self, temp_file):
+        original = DataFrame([datetime(2006, 11, 19, 23, 13, 20)])
+        original.index.name = "index"
+        path = temp_file
+        msg = "Not all pandas column names were valid Stata variable names"
+        with tm.assert_produces_warning(InvalidColumnName, match=msg):
+            original.to_stata(path, convert_dates={0: "tc"})
+
+        written_and_read_again = self.read_dta(path)
+
+        expected = original.copy()
+        expected.columns = ["_0"]
+        expected.index = original.index.astype(np.int32)
+        expected["_0"] = expected["_0"].astype("M8[ms]")
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    def test_105(self, datapath):
+        # Data obtained from:
+        # http://go.worldbank.org/ZXY29PVJ21
+        dpath = datapath("io", "data", "stata", "S4_EDUC1.dta")
+        df = read_stata(dpath)
+        df0 = [[1, 1, 3, -2], [2, 1, 2, -2], [4, 1, 1, -2]]
+        df0 = DataFrame(df0)
+        df0.columns = ["clustnum", "pri_schl", "psch_num", "psch_dis"]
+        df0["clustnum"] = df0["clustnum"].astype(np.int16)
+        df0["pri_schl"] = df0["pri_schl"].astype(np.int8)
+        df0["psch_num"] = df0["psch_num"].astype(np.int8)
+        df0["psch_dis"] = df0["psch_dis"].astype(np.float32)
+        tm.assert_frame_equal(df.head(3), df0)
+
+    def test_value_labels_old_format(self, datapath):
+        # GH 19417
+        #
+        # Test that value_labels() returns an empty dict if the file format
+        # predates supporting value labels.
+        dpath = datapath("io", "data", "stata", "S4_EDUC1.dta")
+        with StataReader(dpath) as reader:
+            assert reader.value_labels() == {}
+
+    def test_date_export_formats(self, temp_file):
+        columns = ["tc", "td", "tw", "tm", "tq", "th", "ty"]
+        conversions = {c: c for c in columns}
+        data = [datetime(2006, 11, 20, 23, 13, 20)] * len(columns)
+        original = DataFrame([data], columns=columns)
+        original.index.name = "index"
+        expected_values = [
+            datetime(2006, 11, 20, 23, 13, 20),  # Time
+            datetime(2006, 11, 20),  # Day
+            datetime(2006, 11, 19),  # Week
+            datetime(2006, 11, 1),  # Month
+            datetime(2006, 10, 1),  # Quarter year
+            datetime(2006, 7, 1),  # Half year
+            datetime(2006, 1, 1),
+        ]  # Year
+
+        expected = DataFrame(
+            [expected_values],
+            index=pd.Index([0], dtype=np.int32, name="index"),
+            columns=columns,
+            dtype="M8[s]",
+        )
+        expected["tc"] = expected["tc"].astype("M8[ms]")
+
+        path = temp_file
+        original.to_stata(path, convert_dates=conversions)
+        written_and_read_again = self.read_dta(path)
+
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    def test_write_missing_strings(self, temp_file):
+        original = DataFrame([["1"], [None]], columns=["foo"])
+
+        expected = DataFrame(
+            [["1"], [""]],
+            index=pd.RangeIndex(2, name="index"),
+            columns=["foo"],
+        )
+
+        path = temp_file
+        original.to_stata(path)
+        written_and_read_again = self.read_dta(path)
+
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    @pytest.mark.parametrize("byteorder", [">", "<"])
+    def test_bool_uint(self, byteorder, version, temp_file):
+        s0 = Series([0, 1, True], dtype=np.bool_)
+        s1 = Series([0, 1, 100], dtype=np.uint8)
+        s2 = Series([0, 1, 255], dtype=np.uint8)
+        s3 = Series([0, 1, 2**15 - 100], dtype=np.uint16)
+        s4 = Series([0, 1, 2**16 - 1], dtype=np.uint16)
+        s5 = Series([0, 1, 2**31 - 100], dtype=np.uint32)
+        s6 = Series([0, 1, 2**32 - 1], dtype=np.uint32)
+
+        original = DataFrame(
+            {"s0": s0, "s1": s1, "s2": s2, "s3": s3, "s4": s4, "s5": s5, "s6": s6}
+        )
+        original.index.name = "index"
+
+        path = temp_file
+        original.to_stata(path, byteorder=byteorder, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        written_and_read_again = written_and_read_again.set_index("index")
+
+        expected = original
+        expected_types = (
+            np.int8,
+            np.int8,
+            np.int16,
+            np.int16,
+            np.int32,
+            np.int32,
+            np.float64,
+        )
+        for c, t in zip(expected.columns, expected_types):
+            expected[c] = expected[c].astype(t)
+
+        tm.assert_frame_equal(written_and_read_again, expected)
+
+    def test_variable_labels(self, datapath):
+        with StataReader(datapath("io", "data", "stata", "stata7_115.dta")) as rdr:
+            sr_115 = rdr.variable_labels()
+        with StataReader(datapath("io", "data", "stata", "stata7_117.dta")) as rdr:
+            sr_117 = rdr.variable_labels()
+        keys = ("var1", "var2", "var3")
+        labels = ("label1", "label2", "label3")
+        for k, v in sr_115.items():
+            assert k in sr_117
+            assert v == sr_117[k]
+            assert k in keys
+            assert v in labels
+
+    def test_minimal_size_col(self, temp_file):
+        str_lens = (1, 100, 244)
+        s = {}
+        for str_len in str_lens:
+            s["s" + str(str_len)] = Series(
+                ["a" * str_len, "b" * str_len, "c" * str_len]
+            )
+        original = DataFrame(s)
+        path = temp_file
+        original.to_stata(path, write_index=False)
+
+        with StataReader(path) as sr:
+            sr._ensure_open()  # The `_*list` variables are initialized here
+            for variable, fmt, typ in zip(sr._varlist, sr._fmtlist, sr._typlist):
+                assert int(variable[1:]) == int(fmt[1:-1])
+                assert int(variable[1:]) == typ
+
+    def test_excessively_long_string(self, temp_file):
+        str_lens = (1, 244, 500)
+        s = {}
+        for str_len in str_lens:
+            s["s" + str(str_len)] = Series(
+                ["a" * str_len, "b" * str_len, "c" * str_len]
+            )
+        original = DataFrame(s)
+        msg = (
+            r"Fixed width strings in Stata \.dta files are limited to 244 "
+            r"\(or fewer\)\ncharacters\.  Column 's500' does not satisfy "
+            r"this restriction\. Use the\n'version=117' parameter to write "
+            r"the newer \(Stata 13 and later\) format\."
+        )
+        with pytest.raises(ValueError, match=msg):
+            path = temp_file
+            original.to_stata(path)
+
+    def test_missing_value_generator(self, temp_file):
+        types = ("b", "h", "l")
+        df = DataFrame([[0.0]], columns=["float_"])
+        path = temp_file
+        df.to_stata(path)
+        with StataReader(path) as rdr:
+            valid_range = rdr.VALID_RANGE
+        expected_values = ["." + chr(97 + i) for i in range(26)]
+        expected_values.insert(0, ".")
+        for t in types:
+            offset = valid_range[t][1]
+            for i in range(27):
+                val = StataMissingValue(offset + 1 + i)
+                assert val.string == expected_values[i]
+
+        # Test extremes for floats
+        val = StataMissingValue(struct.unpack("<f", b"\x00\x00\x00\x7f")[0])
+        assert val.string == "."
+        val = StataMissingValue(struct.unpack("<f", b"\x00\xd0\x00\x7f")[0])
+        assert val.string == ".z"
+
+        # Test extremes for floats
+        val = StataMissingValue(
+            struct.unpack("<d", b"\x00\x00\x00\x00\x00\x00\xe0\x7f")[0]
+        )
+        assert val.string == "."
+        val = StataMissingValue(
+            struct.unpack("<d", b"\x00\x00\x00\x00\x00\x1a\xe0\x7f")[0]
+        )
+        assert val.string == ".z"
+
+    @pytest.mark.parametrize("version", [113, 115, 117])
+    def test_missing_value_conversion(self, version, datapath):
+        columns = ["int8_", "int16_", "int32_", "float32_", "float64_"]
+        smv = StataMissingValue(101)
+        keys = sorted(smv.MISSING_VALUES.keys())
+        data = []
+        for i in range(27):
+            row = [StataMissingValue(keys[i + (j * 27)]) for j in range(5)]
+            data.append(row)
+        expected = DataFrame(data, columns=columns)
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata8_{version}.dta"),
+            convert_missing=True,
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.mark.parametrize("version", [104, 105, 108, 110, 111])
+    def test_missing_value_conversion_compat(self, version, datapath):
+        columns = ["int8_", "int16_", "int32_", "float32_", "float64_"]
+        smv = StataMissingValue(101)
+        keys = sorted(smv.MISSING_VALUES.keys())
+        data = []
+        row = [StataMissingValue(keys[j * 27]) for j in range(5)]
+        data.append(row)
+        expected = DataFrame(data, columns=columns)
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata8_{version}.dta"),
+            convert_missing=True,
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    # The byte type was not supported prior to the 104 format
+    @pytest.mark.parametrize("version", [102, 103])
+    def test_missing_value_conversion_compat_nobyte(self, version, datapath):
+        columns = ["int8_", "int16_", "int32_", "float32_", "float64_"]
+        smv = StataMissingValue(101)
+        keys = sorted(smv.MISSING_VALUES.keys())
+        data = []
+        row = [StataMissingValue(keys[j * 27]) for j in [1, 1, 2, 3, 4]]
+        data.append(row)
+        expected = DataFrame(data, columns=columns)
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata8_{version}.dta"),
+            convert_missing=True,
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    def test_big_dates(self, datapath, temp_file):
+        yr = [1960, 2000, 9999, 100, 2262, 1677]
+        mo = [1, 1, 12, 1, 4, 9]
+        dd = [1, 1, 31, 1, 22, 23]
+        hr = [0, 0, 23, 0, 0, 0]
+        mm = [0, 0, 59, 0, 0, 0]
+        ss = [0, 0, 59, 0, 0, 0]
+        expected = []
+        for year, month, day, hour, minute, second in zip(yr, mo, dd, hr, mm, ss):
+            row = []
+            for j in range(7):
+                if j == 0:
+                    row.append(datetime(year, month, day, hour, minute, second))
+                elif j == 6:
+                    row.append(datetime(year, 1, 1))
+                else:
+                    row.append(datetime(year, month, day))
+            expected.append(row)
+        expected.append([pd.NaT] * 7)
+        columns = [
+            "date_tc",
+            "date_td",
+            "date_tw",
+            "date_tm",
+            "date_tq",
+            "date_th",
+            "date_ty",
+        ]
+
+        # Fixes for weekly, quarterly,half,year
+        expected[2][2] = datetime(9999, 12, 24)
+        expected[2][3] = datetime(9999, 12, 1)
+        expected[2][4] = datetime(9999, 10, 1)
+        expected[2][5] = datetime(9999, 7, 1)
+        expected[4][2] = datetime(2262, 4, 16)
+        expected[4][3] = expected[4][4] = datetime(2262, 4, 1)
+        expected[4][5] = expected[4][6] = datetime(2262, 1, 1)
+        expected[5][2] = expected[5][3] = expected[5][4] = datetime(1677, 10, 1)
+        expected[5][5] = expected[5][6] = datetime(1678, 1, 1)
+
+        expected = DataFrame(expected, columns=columns, dtype=object)
+        expected["date_tc"] = expected["date_tc"].astype("M8[ms]")
+        expected["date_td"] = expected["date_td"].astype("M8[s]")
+        expected["date_tm"] = expected["date_tm"].astype("M8[s]")
+        expected["date_tw"] = expected["date_tw"].astype("M8[s]")
+        expected["date_tq"] = expected["date_tq"].astype("M8[s]")
+        expected["date_th"] = expected["date_th"].astype("M8[s]")
+        expected["date_ty"] = expected["date_ty"].astype("M8[s]")
+
+        parsed_115 = read_stata(datapath("io", "data", "stata", "stata9_115.dta"))
+        parsed_117 = read_stata(datapath("io", "data", "stata", "stata9_117.dta"))
+        tm.assert_frame_equal(expected, parsed_115)
+        tm.assert_frame_equal(expected, parsed_117)
+
+        date_conversion = {c: c[-2:] for c in columns}
+        # {c : c[-2:] for c in columns}
+        path = temp_file
+        expected.index.name = "index"
+        expected.to_stata(path, convert_dates=date_conversion)
+        written_and_read_again = self.read_dta(path)
+
+        tm.assert_frame_equal(
+            written_and_read_again.set_index("index"),
+            expected.set_index(expected.index.astype(np.int32)),
+        )
+
+    def test_dtype_conversion(self, datapath):
+        expected = self.read_csv(datapath("io", "data", "stata", "stata6.csv"))
+        expected["byte_"] = expected["byte_"].astype(np.int8)
+        expected["int_"] = expected["int_"].astype(np.int16)
+        expected["long_"] = expected["long_"].astype(np.int32)
+        expected["float_"] = expected["float_"].astype(np.float32)
+        expected["double_"] = expected["double_"].astype(np.float64)
+        expected["date_td"] = expected["date_td"].astype("M8[s]")
+
+        no_conversion = read_stata(
+            datapath("io", "data", "stata", "stata6_117.dta"), convert_dates=True
+        )
+        tm.assert_frame_equal(expected, no_conversion)
+
+        conversion = read_stata(
+            datapath("io", "data", "stata", "stata6_117.dta"),
+            convert_dates=True,
+            preserve_dtypes=False,
+        )
+
+        # read_csv types are the same
+        expected2 = self.read_csv(datapath("io", "data", "stata", "stata6.csv"))
+        expected2["date_td"] = expected["date_td"]
+
+        tm.assert_frame_equal(expected2, conversion)
+
+    def test_drop_column(self, datapath):
+        expected = self.read_csv(datapath("io", "data", "stata", "stata6.csv"))
+        expected["byte_"] = expected["byte_"].astype(np.int8)
+        expected["int_"] = expected["int_"].astype(np.int16)
+        expected["long_"] = expected["long_"].astype(np.int32)
+        expected["float_"] = expected["float_"].astype(np.float32)
+        expected["double_"] = expected["double_"].astype(np.float64)
+        expected["date_td"] = expected["date_td"].apply(
+            datetime.strptime, args=("%Y-%m-%d",)
+        )
+
+        columns = ["byte_", "int_", "long_"]
+        expected = expected[columns]
+        dropped = read_stata(
+            datapath("io", "data", "stata", "stata6_117.dta"),
+            convert_dates=True,
+            columns=columns,
+        )
+
+        tm.assert_frame_equal(expected, dropped)
+
+        # See PR 10757
+        columns = ["int_", "long_", "byte_"]
+        expected = expected[columns]
+        reordered = read_stata(
+            datapath("io", "data", "stata", "stata6_117.dta"),
+            convert_dates=True,
+            columns=columns,
+        )
+        tm.assert_frame_equal(expected, reordered)
+
+        msg = "columns contains duplicate entries"
+        with pytest.raises(ValueError, match=msg):
+            read_stata(
+                datapath("io", "data", "stata", "stata6_117.dta"),
+                convert_dates=True,
+                columns=["byte_", "byte_"],
+            )
+
+        msg = "The following columns were not found in the Stata data set: not_found"
+        with pytest.raises(ValueError, match=msg):
+            read_stata(
+                datapath("io", "data", "stata", "stata6_117.dta"),
+                convert_dates=True,
+                columns=["byte_", "int_", "long_", "not_found"],
+            )
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    @pytest.mark.filterwarnings(
+        "ignore:\\nStata value:pandas.io.stata.ValueLabelTypeMismatch"
+    )
+    def test_categorical_writing(self, version, temp_file):
+        original = DataFrame.from_records(
+            [
+                ["one", "ten", "one", "one", "one", 1],
+                ["two", "nine", "two", "two", "two", 2],
+                ["three", "eight", "three", "three", "three", 3],
+                ["four", "seven", 4, "four", "four", 4],
+                ["five", "six", 5, np.nan, "five", 5],
+                ["six", "five", 6, np.nan, "six", 6],
+                ["seven", "four", 7, np.nan, "seven", 7],
+                ["eight", "three", 8, np.nan, "eight", 8],
+                ["nine", "two", 9, np.nan, "nine", 9],
+                ["ten", "one", "ten", np.nan, "ten", 10],
+            ],
+            columns=[
+                "fully_labeled",
+                "fully_labeled2",
+                "incompletely_labeled",
+                "labeled_with_missings",
+                "float_labelled",
+                "unlabeled",
+            ],
+        )
+        path = temp_file
+        original.astype("category").to_stata(path, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        res = written_and_read_again.set_index("index")
+
+        expected = original
+        expected.index = expected.index.set_names("index")
+
+        expected["incompletely_labeled"] = expected["incompletely_labeled"].apply(str)
+        expected["unlabeled"] = expected["unlabeled"].apply(str)
+        for col in expected:
+            orig = expected[col]
+
+            cat = orig.astype("category")._values
+            cat = cat.as_ordered()
+            if col == "unlabeled":
+                cat = cat.set_categories(orig, ordered=True)
+
+            cat.categories.rename(None, inplace=True)
+
+            expected[col] = cat
+
+        tm.assert_frame_equal(res, expected)
+
+    def test_categorical_warnings_and_errors(self, temp_file):
+        # Warning for non-string labels
+        original = DataFrame.from_records(
+            [["a"], ["b"], ["c"], ["d"], [1]], columns=["Too_long"]
+        ).astype("category")
+
+        msg = "data file created has not lost information due to duplicate labels"
+        with tm.assert_produces_warning(ValueLabelTypeMismatch, match=msg):
+            original.to_stata(temp_file)
+            # should get a warning for mixed content
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_categorical_with_stata_missing_values(self, version, temp_file):
+        values = [["a" + str(i)] for i in range(120)]
+        values.append([np.nan])
+        original = DataFrame.from_records(values, columns=["many_labels"])
+        original = pd.concat(
+            [original[col].astype("category") for col in original], axis=1
+        )
+        original.index.name = "index"
+        path = temp_file
+        original.to_stata(path, version=version)
+        written_and_read_again = self.read_dta(path)
+
+        res = written_and_read_again.set_index("index")
+
+        expected = original
+        for col in expected:
+            cat = expected[col]._values
+            new_cats = cat.remove_unused_categories().categories
+            cat = cat.set_categories(new_cats, ordered=True)
+            expected[col] = cat
+        tm.assert_frame_equal(res, expected)
+
+    @pytest.mark.parametrize("file", ["stata10_115", "stata10_117"])
+    def test_categorical_order(self, file, datapath):
+        # Directly construct using expected codes
+        # Format is is_cat, col_name, labels (in order), underlying data
+        expected = [
+            (True, "ordered", ["a", "b", "c", "d", "e"], np.arange(5)),
+            (True, "reverse", ["a", "b", "c", "d", "e"], np.arange(5)[::-1]),
+            (True, "noorder", ["a", "b", "c", "d", "e"], np.array([2, 1, 4, 0, 3])),
+            (True, "floating", ["a", "b", "c", "d", "e"], np.arange(0, 5)),
+            (True, "float_missing", ["a", "d", "e"], np.array([0, 1, 2, -1, -1])),
+            (False, "nolabel", [1.0, 2.0, 3.0, 4.0, 5.0], np.arange(5)),
+            (True, "int32_mixed", ["d", 2, "e", "b", "a"], np.arange(5)),
+        ]
+        cols = []
+        for is_cat, col, labels, codes in expected:
+            if is_cat:
+                cols.append(
+                    (col, pd.Categorical.from_codes(codes, labels, ordered=True))
+                )
+            else:
+                cols.append((col, Series(labels, dtype=np.float32)))
+        expected = DataFrame.from_dict(dict(cols))
+
+        # Read with and with out categoricals, ensure order is identical
+        file = datapath("io", "data", "stata", f"{file}.dta")
+        parsed = read_stata(file)
+        tm.assert_frame_equal(expected, parsed)
+
+        # Check identity of codes
+        for col in expected:
+            if isinstance(expected[col].dtype, CategoricalDtype):
+                tm.assert_series_equal(expected[col].cat.codes, parsed[col].cat.codes)
+                tm.assert_index_equal(
+                    expected[col].cat.categories, parsed[col].cat.categories
+                )
+
+    @pytest.mark.parametrize("file", ["stata11_115", "stata11_117"])
+    def test_categorical_sorting(self, file, datapath):
+        parsed = read_stata(datapath("io", "data", "stata", f"{file}.dta"))
+
+        # Sort based on codes, not strings
+        parsed = parsed.sort_values("srh", na_position="first")
+
+        # Don't sort index
+        parsed.index = pd.RangeIndex(len(parsed))
+        codes = [-1, -1, 0, 1, 1, 1, 2, 2, 3, 4]
+        categories = ["Poor", "Fair", "Good", "Very good", "Excellent"]
+        cat = pd.Categorical.from_codes(
+            codes=codes, categories=categories, ordered=True
+        )
+        expected = Series(cat, name="srh")
+        tm.assert_series_equal(expected, parsed["srh"])
+
+    @pytest.mark.parametrize("file", ["stata10_115", "stata10_117"])
+    def test_categorical_ordering(self, file, datapath):
+        file = datapath("io", "data", "stata", f"{file}.dta")
+        parsed = read_stata(file)
+
+        parsed_unordered = read_stata(file, order_categoricals=False)
+        for col in parsed:
+            if not isinstance(parsed[col].dtype, CategoricalDtype):
+                continue
+            assert parsed[col].cat.ordered
+            assert not parsed_unordered[col].cat.ordered
+
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    @pytest.mark.parametrize(
+        "file",
+        [
+            "stata1_117",
+            "stata2_117",
+            "stata3_117",
+            "stata4_117",
+            "stata5_117",
+            "stata6_117",
+            "stata7_117",
+            "stata8_117",
+            "stata9_117",
+            "stata10_117",
+            "stata11_117",
+        ],
+    )
+    @pytest.mark.parametrize("chunksize", [1, 2])
+    @pytest.mark.parametrize("convert_categoricals", [False, True])
+    @pytest.mark.parametrize("convert_dates", [False, True])
+    def test_read_chunks_117(
+        self, file, chunksize, convert_categoricals, convert_dates, datapath
+    ):
+        fname = datapath("io", "data", "stata", f"{file}.dta")
+
+        parsed = read_stata(
+            fname,
+            convert_categoricals=convert_categoricals,
+            convert_dates=convert_dates,
+        )
+        with read_stata(
+            fname,
+            iterator=True,
+            convert_categoricals=convert_categoricals,
+            convert_dates=convert_dates,
+        ) as itr:
+            pos = 0
+            for j in range(5):
+                try:
+                    chunk = itr.read(chunksize)
+                except StopIteration:
+                    break
+                from_frame = parsed.iloc[pos : pos + chunksize, :].copy()
+                from_frame = self._convert_categorical(from_frame)
+                tm.assert_frame_equal(
+                    from_frame,
+                    chunk,
+                    check_dtype=False,
+                )
+                pos += chunksize
+
+    @staticmethod
+    def _convert_categorical(from_frame: DataFrame) -> DataFrame:
+        """
+        Emulate the categorical casting behavior we expect from roundtripping.
+        """
+        for col in from_frame:
+            ser = from_frame[col]
+            if isinstance(ser.dtype, CategoricalDtype):
+                cat = ser._values.remove_unused_categories()
+                if cat.categories.dtype == object:
+                    categories = pd.Index._with_infer(cat.categories._values)
+                    cat = cat.set_categories(categories)
+                elif cat.categories.dtype == "string" and len(cat.categories) == 0:
+                    # if the read categories are empty, it comes back as object dtype
+                    categories = cat.categories.astype(object)
+                    cat = cat.set_categories(categories)
+                from_frame[col] = cat
+        return from_frame
+
+    def test_iterator(self, datapath):
+        fname = datapath("io", "data", "stata", "stata3_117.dta")
+
+        parsed = read_stata(fname)
+
+        with read_stata(fname, iterator=True) as itr:
+            chunk = itr.read(5)
+            tm.assert_frame_equal(parsed.iloc[0:5, :], chunk)
+
+        with read_stata(fname, chunksize=5) as itr:
+            chunk = list(itr)
+            tm.assert_frame_equal(parsed.iloc[0:5, :], chunk[0])
+
+        with read_stata(fname, iterator=True) as itr:
+            chunk = itr.get_chunk(5)
+            tm.assert_frame_equal(parsed.iloc[0:5, :], chunk)
+
+        with read_stata(fname, chunksize=5) as itr:
+            chunk = itr.get_chunk()
+            tm.assert_frame_equal(parsed.iloc[0:5, :], chunk)
+
+        # GH12153
+        with read_stata(fname, chunksize=4) as itr:
+            from_chunks = pd.concat(itr)
+        tm.assert_frame_equal(parsed, from_chunks)
+
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    @pytest.mark.parametrize(
+        "file",
+        [
+            "stata2_115",
+            "stata3_115",
+            "stata4_115",
+            "stata5_115",
+            "stata6_115",
+            "stata7_115",
+            "stata8_115",
+            "stata9_115",
+            "stata10_115",
+            "stata11_115",
+        ],
+    )
+    @pytest.mark.parametrize("chunksize", [1, 2])
+    @pytest.mark.parametrize("convert_categoricals", [False, True])
+    @pytest.mark.parametrize("convert_dates", [False, True])
+    def test_read_chunks_115(
+        self, file, chunksize, convert_categoricals, convert_dates, datapath
+    ):
+        fname = datapath("io", "data", "stata", f"{file}.dta")
+
+        # Read the whole file
+        parsed = read_stata(
+            fname,
+            convert_categoricals=convert_categoricals,
+            convert_dates=convert_dates,
+        )
+
+        # Compare to what we get when reading by chunk
+        with read_stata(
+            fname,
+            iterator=True,
+            convert_dates=convert_dates,
+            convert_categoricals=convert_categoricals,
+        ) as itr:
+            pos = 0
+            for j in range(5):
+                try:
+                    chunk = itr.read(chunksize)
+                except StopIteration:
+                    break
+                from_frame = parsed.iloc[pos : pos + chunksize, :].copy()
+                from_frame = self._convert_categorical(from_frame)
+                tm.assert_frame_equal(
+                    from_frame,
+                    chunk,
+                    check_dtype=False,
+                )
+                pos += chunksize
+
+    def test_read_chunks_columns(self, datapath):
+        fname = datapath("io", "data", "stata", "stata3_117.dta")
+        columns = ["quarter", "cpi", "m1"]
+        chunksize = 2
+
+        parsed = read_stata(fname, columns=columns)
+        with read_stata(fname, iterator=True) as itr:
+            pos = 0
+            for j in range(5):
+                chunk = itr.read(chunksize, columns=columns)
+                if chunk is None:
+                    break
+                from_frame = parsed.iloc[pos : pos + chunksize, :]
+                tm.assert_frame_equal(from_frame, chunk, check_dtype=False)
+                pos += chunksize
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_write_variable_labels(self, version, mixed_frame, temp_file):
+        # GH 13631, add support for writing variable labels
+        mixed_frame.index.name = "index"
+        variable_labels = {"a": "City Rank", "b": "City Exponent", "c": "City"}
+        path = temp_file
+        mixed_frame.to_stata(path, variable_labels=variable_labels, version=version)
+        with StataReader(path) as sr:
+            read_labels = sr.variable_labels()
+        expected_labels = {
+            "index": "",
+            "a": "City Rank",
+            "b": "City Exponent",
+            "c": "City",
+        }
+        assert read_labels == expected_labels
+
+        variable_labels["index"] = "The Index"
+        path = temp_file
+        mixed_frame.to_stata(path, variable_labels=variable_labels, version=version)
+        with StataReader(path) as sr:
+            read_labels = sr.variable_labels()
+        assert read_labels == variable_labels
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_invalid_variable_labels(self, version, mixed_frame, temp_file):
+        mixed_frame.index.name = "index"
+        variable_labels = {"a": "very long" * 10, "b": "City Exponent", "c": "City"}
+        path = temp_file
+        msg = "Variable labels must be 80 characters or fewer"
+        with pytest.raises(ValueError, match=msg):
+            mixed_frame.to_stata(path, variable_labels=variable_labels, version=version)
+
+    @pytest.mark.parametrize("version", [114, 117])
+    def test_invalid_variable_label_encoding(self, version, mixed_frame, temp_file):
+        mixed_frame.index.name = "index"
+        variable_labels = {"a": "very long" * 10, "b": "City Exponent", "c": "City"}
+        variable_labels["a"] = "invalid character Œ"
+        path = temp_file
+        with pytest.raises(
+            ValueError, match="Variable labels must contain only characters"
+        ):
+            mixed_frame.to_stata(path, variable_labels=variable_labels, version=version)
+
+    def test_write_variable_label_errors(self, mixed_frame, temp_file):
+        values = ["\u03a1", "\u0391", "\u039d", "\u0394", "\u0391", "\u03a3"]
+
+        variable_labels_utf8 = {
+            "a": "City Rank",
+            "b": "City Exponent",
+            "c": "".join(values),
+        }
+
+        msg = (
+            "Variable labels must contain only characters that can be "
+            "encoded in Latin-1"
+        )
+        with pytest.raises(ValueError, match=msg):
+            path = temp_file
+            mixed_frame.to_stata(path, variable_labels=variable_labels_utf8)
+
+        variable_labels_long = {
+            "a": "City Rank",
+            "b": "City Exponent",
+            "c": "A very, very, very long variable label "
+            "that is too long for Stata which means "
+            "that it has more than 80 characters",
+        }
+
+        msg = "Variable labels must be 80 characters or fewer"
+        with pytest.raises(ValueError, match=msg):
+            path = temp_file
+            mixed_frame.to_stata(path, variable_labels=variable_labels_long)
+
+    def test_default_date_conversion(self, temp_file):
+        # GH 12259
+        dates = [
+            dt.datetime(1999, 12, 31, 12, 12, 12, 12000),
+            dt.datetime(2012, 12, 21, 12, 21, 12, 21000),
+            dt.datetime(1776, 7, 4, 7, 4, 7, 4000),
+        ]
+        original = DataFrame(
+            {
+                "nums": [1.0, 2.0, 3.0],
+                "strs": ["apple", "banana", "cherry"],
+                "dates": dates,
+            }
+        )
+
+        expected = original[:]
+        # "tc" for convert_dates below stores with "ms" resolution
+        expected["dates"] = expected["dates"].astype("M8[ms]")
+
+        path = temp_file
+        original.to_stata(path, write_index=False)
+        reread = read_stata(path, convert_dates=True)
+        tm.assert_frame_equal(expected, reread)
+
+        original.to_stata(path, write_index=False, convert_dates={"dates": "tc"})
+        direct = read_stata(path, convert_dates=True)
+        tm.assert_frame_equal(reread, direct)
+
+        dates_idx = original.columns.tolist().index("dates")
+        original.to_stata(path, write_index=False, convert_dates={dates_idx: "tc"})
+        direct = read_stata(path, convert_dates=True)
+        tm.assert_frame_equal(reread, direct)
+
+    def test_unsupported_type(self, temp_file):
+        original = DataFrame({"a": [1 + 2j, 2 + 4j]})
+
+        msg = "Data type complex128 not supported"
+        with pytest.raises(NotImplementedError, match=msg):
+            path = temp_file
+            original.to_stata(path)
+
+    def test_unsupported_datetype(self, temp_file):
+        dates = [
+            dt.datetime(1999, 12, 31, 12, 12, 12, 12000),
+            dt.datetime(2012, 12, 21, 12, 21, 12, 21000),
+            dt.datetime(1776, 7, 4, 7, 4, 7, 4000),
+        ]
+        original = DataFrame(
+            {
+                "nums": [1.0, 2.0, 3.0],
+                "strs": ["apple", "banana", "cherry"],
+                "dates": dates,
+            }
+        )
+
+        msg = "Format %tC not implemented"
+        with pytest.raises(NotImplementedError, match=msg):
+            path = temp_file
+            original.to_stata(path, convert_dates={"dates": "tC"})
+
+        dates = pd.date_range("1-1-1990", periods=3, tz="Asia/Hong_Kong")
+        original = DataFrame(
+            {
+                "nums": [1.0, 2.0, 3.0],
+                "strs": ["apple", "banana", "cherry"],
+                "dates": dates,
+            }
+        )
+        with pytest.raises(NotImplementedError, match="Data type datetime64"):
+            path = temp_file
+            original.to_stata(path)
+
+    def test_repeated_column_labels(self, datapath):
+        # GH 13923, 25772
+        msg = """
+Value labels for column ethnicsn are not unique. These cannot be converted to
+pandas categoricals.
+
+Either read the file with `convert_categoricals` set to False or use the
+low level interface in `StataReader` to separately read the values and the
+value_labels.
+
+The repeated labels are:\n-+\nwolof
+"""
+        with pytest.raises(ValueError, match=msg):
+            read_stata(
+                datapath("io", "data", "stata", "stata15.dta"),
+                convert_categoricals=True,
+            )
+
+    def test_stata_111(self, datapath):
+        # 111 is an old version but still used by current versions of
+        # SAS when exporting to Stata format. We do not know of any
+        # on-line documentation for this version.
+        df = read_stata(datapath("io", "data", "stata", "stata7_111.dta"))
+        original = DataFrame(
+            {
+                "y": [1, 1, 1, 1, 1, 0, 0, np.nan, 0, 0],
+                "x": [1, 2, 1, 3, np.nan, 4, 3, 5, 1, 6],
+                "w": [2, np.nan, 5, 2, 4, 4, 3, 1, 2, 3],
+                "z": ["a", "b", "c", "d", "e", "", "g", "h", "i", "j"],
+            }
+        )
+        original = original[["y", "x", "w", "z"]]
+        tm.assert_frame_equal(original, df)
+
+    def test_out_of_range_double(self, temp_file):
+        # GH 14618
+        df = DataFrame(
+            {
+                "ColumnOk": [0.0, np.finfo(np.double).eps, 4.49423283715579e307],
+                "ColumnTooBig": [0.0, np.finfo(np.double).eps, np.finfo(np.double).max],
+            }
+        )
+        msg = (
+            r"Column ColumnTooBig has a maximum value \(.+\) outside the range "
+            r"supported by Stata \(.+\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            path = temp_file
+            df.to_stata(path)
+
+    def test_out_of_range_float(self, temp_file):
+        original = DataFrame(
+            {
+                "ColumnOk": [
+                    0.0,
+                    np.finfo(np.float32).eps,
+                    np.finfo(np.float32).max / 10.0,
+                ],
+                "ColumnTooBig": [
+                    0.0,
+                    np.finfo(np.float32).eps,
+                    np.finfo(np.float32).max,
+                ],
+            }
+        )
+        original.index.name = "index"
+        for col in original:
+            original[col] = original[col].astype(np.float32)
+
+        path = temp_file
+        original.to_stata(path)
+        reread = read_stata(path)
+
+        original["ColumnTooBig"] = original["ColumnTooBig"].astype(np.float64)
+        expected = original
+        tm.assert_frame_equal(reread.set_index("index"), expected)
+
+    @pytest.mark.parametrize("infval", [np.inf, -np.inf])
+    def test_inf(self, infval, temp_file):
+        # GH 45350
+        df = DataFrame({"WithoutInf": [0.0, 1.0], "WithInf": [2.0, infval]})
+        msg = (
+            "Column WithInf contains infinity or -infinity"
+            "which is outside the range supported by Stata."
+        )
+        with pytest.raises(ValueError, match=msg):
+            path = temp_file
+            df.to_stata(path)
+
+    def test_path_pathlib(self):
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        )
+        df.index.name = "index"
+        reader = lambda x: read_stata(x).set_index("index")
+        result = tm.round_trip_pathlib(df.to_stata, reader)
+        tm.assert_frame_equal(df, result)
+
+    @pytest.mark.parametrize("write_index", [True, False])
+    def test_value_labels_iterator(self, write_index, temp_file):
+        # GH 16923
+        d = {"A": ["B", "E", "C", "A", "E"]}
+        df = DataFrame(data=d)
+        df["A"] = df["A"].astype("category")
+        path = temp_file
+        df.to_stata(path, write_index=write_index)
+
+        with read_stata(path, iterator=True) as dta_iter:
+            value_labels = dta_iter.value_labels()
+        assert value_labels == {"A": {0: "A", 1: "B", 2: "C", 3: "E"}}
+
+    def test_set_index(self, temp_file):
+        # GH 17328
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        )
+        df.index.name = "index"
+        path = temp_file
+        df.to_stata(path)
+        reread = read_stata(path, index_col="index")
+        tm.assert_frame_equal(df, reread)
+
+    @pytest.mark.parametrize(
+        "column", ["ms", "day", "week", "month", "qtr", "half", "yr"]
+    )
+    def test_date_parsing_ignores_format_details(self, column, datapath):
+        # GH 17797
+        #
+        # Test that display formats are ignored when determining if a numeric
+        # column is a date value.
+        #
+        # All date types are stored as numbers and format associated with the
+        # column denotes both the type of the date and the display format.
+        #
+        # STATA supports 9 date types which each have distinct units. We test 7
+        # of the 9 types, ignoring %tC and %tb. %tC is a variant of %tc that
+        # accounts for leap seconds and %tb relies on STATAs business calendar.
+        df = read_stata(datapath("io", "data", "stata", "stata13_dates.dta"))
+        unformatted = df.loc[0, column]
+        formatted = df.loc[0, column + "_fmt"]
+        assert unformatted == formatted
+
+    @pytest.mark.parametrize("byteorder", ["little", "big"])
+    def test_writer_117(self, byteorder, temp_file, using_infer_string):
+        original = DataFrame(
+            data=[
+                [
+                    "string",
+                    "object",
+                    1,
+                    1,
+                    1,
+                    1.1,
+                    1.1,
+                    np.datetime64("2003-12-25"),
+                    "a",
+                    "a" * 2045,
+                    "a" * 5000,
+                    "a",
+                ],
+                [
+                    "string-1",
+                    "object-1",
+                    1,
+                    1,
+                    1,
+                    1.1,
+                    1.1,
+                    np.datetime64("2003-12-26"),
+                    "b",
+                    "b" * 2045,
+                    "",
+                    "",
+                ],
+            ],
+            columns=[
+                "string",
+                "object",
+                "int8",
+                "int16",
+                "int32",
+                "float32",
+                "float64",
+                "datetime",
+                "s1",
+                "s2045",
+                "srtl",
+                "forced_strl",
+            ],
+        )
+        original["object"] = Series(original["object"], dtype=object)
+        original["int8"] = Series(original["int8"], dtype=np.int8)
+        original["int16"] = Series(original["int16"], dtype=np.int16)
+        original["int32"] = original["int32"].astype(np.int32)
+        original["float32"] = Series(original["float32"], dtype=np.float32)
+        original.index.name = "index"
+        copy = original.copy()
+        path = temp_file
+        original.to_stata(
+            path,
+            convert_dates={"datetime": "tc"},
+            byteorder=byteorder,
+            convert_strl=["forced_strl"],
+            version=117,
+        )
+        written_and_read_again = self.read_dta(path)
+
+        expected = original[:]
+        # "tc" for convert_dates means we store with "ms" resolution
+        expected["datetime"] = expected["datetime"].astype("M8[ms]")
+        if using_infer_string:
+            # object dtype (with only strings/None) comes back as string dtype
+            expected["object"] = expected["object"].astype("str")
+
+        tm.assert_frame_equal(
+            written_and_read_again.set_index("index"),
+            expected,
+        )
+        tm.assert_frame_equal(original, copy)
+
+    def test_convert_strl_name_swap(self, temp_file):
+        original = DataFrame(
+            [["a" * 3000, "A", "apple"], ["b" * 1000, "B", "banana"]],
+            columns=["long1" * 10, "long", 1],
+        )
+        original.index.name = "index"
+
+        msg = "Not all pandas column names were valid Stata variable names"
+        with tm.assert_produces_warning(InvalidColumnName, match=msg):
+            path = temp_file
+            original.to_stata(path, convert_strl=["long", 1], version=117)
+            reread = self.read_dta(path)
+            reread = reread.set_index("index")
+            reread.columns = original.columns
+            tm.assert_frame_equal(reread, original, check_index_type=False)
+
+    def test_invalid_date_conversion(self, temp_file):
+        # GH 12259
+        dates = [
+            dt.datetime(1999, 12, 31, 12, 12, 12, 12000),
+            dt.datetime(2012, 12, 21, 12, 21, 12, 21000),
+            dt.datetime(1776, 7, 4, 7, 4, 7, 4000),
+        ]
+        original = DataFrame(
+            {
+                "nums": [1.0, 2.0, 3.0],
+                "strs": ["apple", "banana", "cherry"],
+                "dates": dates,
+            }
+        )
+
+        path = temp_file
+        msg = "convert_dates key must be a column or an integer"
+        with pytest.raises(ValueError, match=msg):
+            original.to_stata(path, convert_dates={"wrong_name": "tc"})
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_nonfile_writing(self, version, temp_file):
+        # GH 21041
+        bio = io.BytesIO()
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        )
+        df.index.name = "index"
+        path = temp_file
+        df.to_stata(bio, version=version)
+        bio.seek(0)
+        with open(path, "wb") as dta:
+            dta.write(bio.read())
+        reread = read_stata(path, index_col="index")
+        tm.assert_frame_equal(df, reread)
+
+    def test_gzip_writing(self, temp_file):
+        # writing version 117 requires seek and cannot be used with gzip
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=pd.Index(list("ABCD")),
+            index=pd.Index([f"i-{i}" for i in range(30)]),
+        )
+        df.index.name = "index"
+        path = temp_file
+        with gzip.GzipFile(path, "wb") as gz:
+            df.to_stata(gz, version=114)
+        with gzip.GzipFile(path, "rb") as gz:
+            reread = read_stata(gz, index_col="index")
+        tm.assert_frame_equal(df, reread)
+
+    # 117 is not included in this list as it uses ASCII strings
+    @pytest.mark.parametrize(
+        "file",
+        [
+            "stata16_118",
+            "stata16_be_118",
+            "stata16_119",
+            "stata16_be_119",
+        ],
+    )
+    def test_unicode_dta_118_119(self, file, datapath):
+        unicode_df = self.read_dta(datapath("io", "data", "stata", f"{file}.dta"))
+
+        columns = ["utf8", "latin1", "ascii", "utf8_strl", "ascii_strl"]
+        values = [
+            ["ραηδας", "PÄNDÄS", "p", "ραηδας", "p"],
+            ["ƤĀńĐąŜ", "Ö", "a", "ƤĀńĐąŜ", "a"],
+            ["ᴘᴀᴎᴅᴀS", "Ü", "n", "ᴘᴀᴎᴅᴀS", "n"],
+            ["      ", "      ", "d", "      ", "d"],
+            [" ", "", "a", " ", "a"],
+            ["", "", "s", "", "s"],
+            ["", "", " ", "", " "],
+        ]
+        expected = DataFrame(values, columns=columns)
+
+        tm.assert_frame_equal(unicode_df, expected)
+
+    def test_mixed_string_strl(self, temp_file, using_infer_string):
+        # GH 23633
+        output = [{"mixed": "string" * 500, "number": 0}, {"mixed": None, "number": 1}]
+        output = DataFrame(output)
+        output.number = output.number.astype("int32")
+
+        path = temp_file
+        output.to_stata(path, write_index=False, version=117)
+        reread = read_stata(path)
+        expected = output.fillna("")
+        tm.assert_frame_equal(reread, expected)
+
+        # Check strl supports all None (null)
+        output["mixed"] = None
+        output.to_stata(path, write_index=False, convert_strl=["mixed"], version=117)
+        reread = read_stata(path)
+        expected = output.fillna("")
+        if using_infer_string:
+            expected["mixed"] = expected["mixed"].astype("str")
+        tm.assert_frame_equal(reread, expected)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_all_none_exception(self, version, temp_file):
+        output = [{"none": "none", "number": 0}, {"none": None, "number": 1}]
+        output = DataFrame(output)
+        output["none"] = None
+        with pytest.raises(ValueError, match="Column `none` cannot be exported"):
+            output.to_stata(temp_file, version=version)
+
+    @pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+    def test_invalid_file_not_written(self, version, temp_file):
+        content = "Here is one __�__ Another one __·__ Another one __½__"
+        df = DataFrame([content], columns=["invalid"])
+        msg1 = (
+            r"'latin-1' codec can't encode character '\\ufffd' "
+            r"in position 14: ordinal not in range\(256\)"
+        )
+        msg2 = (
+            "'ascii' codec can't decode byte 0xef in position 14: "
+            r"ordinal not in range\(128\)"
+        )
+        with pytest.raises(UnicodeEncodeError, match=f"{msg1}|{msg2}"):
+            df.to_stata(temp_file)
+
+    def test_strl_latin1(self, temp_file):
+        # GH 23573, correct GSO data to reflect correct size
+        output = DataFrame(
+            [["pandas"] * 2, ["þâÑÐŧ"] * 2], columns=["var_str", "var_strl"]
+        )
+
+        output.to_stata(temp_file, version=117, convert_strl=["var_strl"])
+        with open(temp_file, "rb") as reread:
+            content = reread.read()
+            expected = "þâÑÐŧ"
+            assert expected.encode("latin-1") in content
+            assert expected.encode("utf-8") in content
+            gsos = content.split(b"strls")[1][1:-2]
+            for gso in gsos.split(b"GSO")[1:]:
+                val = gso.split(b"\x00")[-2]
+                size = gso[gso.find(b"\x82") + 1]
+                assert len(val) == size - 1
+
+    def test_encoding_latin1_118(self, datapath):
+        # GH 25960
+        msg = """
+One or more strings in the dta file could not be decoded using utf-8, and
+so the fallback encoding of latin-1 is being used.  This can happen when a file
+has been incorrectly encoded by Stata or some other software. You should verify
+the string values returned are correct."""
+        # Move path outside of read_stata, or else assert_produces_warning
+        # will block pytests skip mechanism from triggering (failing the test)
+        # if the path is not present
+        path = datapath("io", "data", "stata", "stata1_encoding_118.dta")
+        with tm.assert_produces_warning(UnicodeWarning, filter_level="once") as w:
+            encoded = read_stata(path)
+            # with filter_level="always", produces 151 warnings which can be slow
+            assert len(w) == 1
+            assert w[0].message.args[0] == msg
+
+        expected = DataFrame([["Düsseldorf"]] * 151, columns=["kreis1849"])
+        tm.assert_frame_equal(encoded, expected)
+
+    @pytest.mark.slow
+    def test_stata_119(self, datapath):
+        # Gzipped since contains 32,999 variables and uncompressed is 20MiB
+        # Just validate that the reader reports correct number of variables
+        # to avoid high peak memory
+        with gzip.open(
+            datapath("io", "data", "stata", "stata1_119.dta.gz"), "rb"
+        ) as gz:
+            with StataReader(gz) as reader:
+                reader._ensure_open()
+                assert reader._nvar == 32999
+
+    @pytest.mark.parametrize("version", [118, 119, None])
+    @pytest.mark.parametrize("byteorder", ["little", "big"])
+    def test_utf8_writer(self, version, byteorder, temp_file):
+        cat = pd.Categorical(["a", "β", "ĉ"], ordered=True)
+        data = DataFrame(
+            [
+                [1.0, 1, "ᴬ", "ᴀ relatively long ŝtring"],
+                [2.0, 2, "ᴮ", ""],
+                [3.0, 3, "ᴰ", None],
+            ],
+            columns=["Å", "β", "ĉ", "strls"],
+        )
+        data["ᴐᴬᵀ"] = cat
+        variable_labels = {
+            "Å": "apple",
+            "β": "ᵈᵉᵊ",
+            "ĉ": "ᴎტჄႲႳႴႶႺ",
+            "strls": "Long Strings",
+            "ᴐᴬᵀ": "",
+        }
+        data_label = "ᴅaᵀa-label"
+        value_labels = {"β": {1: "label", 2: "æøå", 3: "ŋot valid latin-1"}}
+        data["β"] = data["β"].astype(np.int32)
+        writer = StataWriterUTF8(
+            temp_file,
+            data,
+            data_label=data_label,
+            convert_strl=["strls"],
+            variable_labels=variable_labels,
+            write_index=False,
+            byteorder=byteorder,
+            version=version,
+            value_labels=value_labels,
+        )
+        writer.write_file()
+        reread_encoded = read_stata(temp_file)
+        # Missing is intentionally converted to empty strl
+        data["strls"] = data["strls"].fillna("")
+        # Variable with value labels is reread as categorical
+        data["β"] = (
+            data["β"].replace(value_labels["β"]).astype("category").cat.as_ordered()
+        )
+        tm.assert_frame_equal(data, reread_encoded)
+        with StataReader(temp_file) as reader:
+            assert reader.data_label == data_label
+            assert reader.variable_labels() == variable_labels
+
+        data.to_stata(temp_file, version=version, write_index=False)
+        reread_to_stata = read_stata(temp_file)
+        tm.assert_frame_equal(data, reread_to_stata)
+
+    def test_writer_118_exceptions(self, temp_file):
+        df = DataFrame(np.zeros((1, 33000), dtype=np.int8))
+        with pytest.raises(ValueError, match="version must be either 118 or 119."):
+            StataWriterUTF8(temp_file, df, version=117)
+        with pytest.raises(ValueError, match="You must use version 119"):
+            StataWriterUTF8(temp_file, df, version=118)
+
+    @pytest.mark.parametrize(
+        "dtype_backend",
+        ["numpy_nullable", pytest.param("pyarrow", marks=td.skip_if_no("pyarrow"))],
+    )
+    def test_read_write_ea_dtypes(self, dtype_backend, temp_file, tmp_path):
+        df = DataFrame(
+            {
+                "a": [1, 2, None],
+                "b": ["a", "b", "c"],
+                "c": [True, False, None],
+                "d": [1.5, 2.5, 3.5],
+                "e": pd.date_range("2020-12-31", periods=3, freq="D"),
+            },
+            index=pd.Index([0, 1, 2], name="index"),
+        )
+        df = df.convert_dtypes(dtype_backend=dtype_backend)
+        stata_path = tmp_path / "test_stata.dta"
+        df.to_stata(stata_path, version=118)
+
+        df.to_stata(temp_file)
+        written_and_read_again = self.read_dta(temp_file)
+
+        expected = DataFrame(
+            {
+                "a": [1, 2, np.nan],
+                "b": ["a", "b", "c"],
+                "c": [1.0, 0, np.nan],
+                "d": [1.5, 2.5, 3.5],
+                # stata stores with ms unit, so unit does not round-trip exactly
+                "e": pd.date_range("2020-12-31", periods=3, freq="D", unit="ms"),
+            },
+            index=pd.RangeIndex(range(3), name="index"),
+        )
+
+        tm.assert_frame_equal(written_and_read_again.set_index("index"), expected)
+
+    @pytest.mark.parametrize("version", [113, 114, 115, 117, 118, 119])
+    def test_read_data_int_validranges(self, version, datapath):
+        expected = DataFrame(
+            {
+                "byte": np.array([-127, 100], dtype=np.int8),
+                "int": np.array([-32767, 32740], dtype=np.int16),
+                "long": np.array([-2147483647, 2147483620], dtype=np.int32),
+            }
+        )
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata_int_validranges_{version}.dta")
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    @pytest.mark.parametrize("version", [104, 105, 108, 110, 111])
+    def test_read_data_int_validranges_compat(self, version, datapath):
+        expected = DataFrame(
+            {
+                "byte": np.array([-128, 126], dtype=np.int8),
+                "int": np.array([-32768, 32766], dtype=np.int16),
+                "long": np.array([-2147483648, 2147483646], dtype=np.int32),
+            }
+        )
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata_int_validranges_{version}.dta")
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+    # The byte type was not supported prior to the 104 format
+    @pytest.mark.parametrize("version", [102, 103])
+    def test_read_data_int_validranges_compat_nobyte(self, version, datapath):
+        expected = DataFrame(
+            {
+                "byte": np.array([-128, 126], dtype=np.int16),
+                "int": np.array([-32768, 32766], dtype=np.int16),
+                "long": np.array([-2147483648, 2147483646], dtype=np.int32),
+            }
+        )
+
+        parsed = read_stata(
+            datapath("io", "data", "stata", f"stata_int_validranges_{version}.dta")
+        )
+        tm.assert_frame_equal(parsed, expected)
+
+
+@pytest.mark.parametrize("version", [105, 108, 110, 111, 113, 114])
+def test_backward_compat(version, datapath):
+    data_base = datapath("io", "data", "stata")
+    ref = os.path.join(data_base, "stata-compat-118.dta")
+    old = os.path.join(data_base, f"stata-compat-{version}.dta")
+    expected = read_stata(ref)
+    old_dta = read_stata(old)
+    tm.assert_frame_equal(old_dta, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize("version", [103, 104])
+def test_backward_compat_nodateconversion(version, datapath):
+    # The Stata data format prior to 105 did not support a date format
+    # so read the raw values for comparison
+    data_base = datapath("io", "data", "stata")
+    ref = os.path.join(data_base, "stata-compat-118.dta")
+    old = os.path.join(data_base, f"stata-compat-{version}.dta")
+    expected = read_stata(ref, convert_dates=False)
+    old_dta = read_stata(old, convert_dates=False)
+    tm.assert_frame_equal(old_dta, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize("version", [102])
+def test_backward_compat_nostring(version, datapath):
+    # The Stata data format prior to 105 did not support a date format
+    # so read the raw values for comparison
+    ref = datapath("io", "data", "stata", "stata-compat-118.dta")
+    old = datapath("io", "data", "stata", f"stata-compat-{version}.dta")
+    expected = read_stata(ref, convert_dates=False)
+    # The Stata data format prior to 103 did not support string data
+    expected = expected.drop(columns=["s10"])
+    old_dta = read_stata(old, convert_dates=False)
+    tm.assert_frame_equal(old_dta, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize("version", [105, 108, 110, 111, 113, 114, 118])
+def test_bigendian(version, datapath):
+    ref = datapath("io", "data", "stata", f"stata-compat-{version}.dta")
+    big = datapath("io", "data", "stata", f"stata-compat-be-{version}.dta")
+    expected = read_stata(ref)
+    big_dta = read_stata(big)
+    tm.assert_frame_equal(big_dta, expected)
+
+
+# Note: 102 format does not support big-endian byte order
+@pytest.mark.parametrize("version", [103, 104])
+def test_bigendian_nodateconversion(version, datapath):
+    # The Stata data format prior to 105 did not support a date format
+    # so read the raw values for comparison
+    ref = datapath("io", "data", "stata", f"stata-compat-{version}.dta")
+    big = datapath("io", "data", "stata", f"stata-compat-be-{version}.dta")
+    expected = read_stata(ref, convert_dates=False)
+    big_dta = read_stata(big, convert_dates=False)
+    tm.assert_frame_equal(big_dta, expected)
+
+
+def test_direct_read(datapath, monkeypatch):
+    file_path = datapath("io", "data", "stata", "stata-compat-118.dta")
+
+    # Test that opening a file path doesn't buffer the file.
+    with StataReader(file_path) as reader:
+        # Must not have been buffered to memory
+        assert not reader.read().empty
+        assert not isinstance(reader._path_or_buf, io.BytesIO)
+
+    # Test that we use a given fp exactly, if possible.
+    with open(file_path, "rb") as fp:
+        with StataReader(fp) as reader:
+            assert not reader.read().empty
+            assert reader._path_or_buf is fp
+
+    # Test that we use a given BytesIO exactly, if possible.
+    with open(file_path, "rb") as fp:
+        with io.BytesIO(fp.read()) as bio:
+            with StataReader(bio) as reader:
+                assert not reader.read().empty
+                assert reader._path_or_buf is bio
+
+
+@pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+@pytest.mark.parametrize("use_dict", [True, False])
+@pytest.mark.parametrize("infer", [True, False])
+def test_compression(
+    compression, version, use_dict, infer, compression_to_extension, tmp_path
+):
+    file_name = "dta_inferred_compression.dta"
+    if compression:
+        if use_dict:
+            file_ext = compression
+        else:
+            file_ext = compression_to_extension[compression]
+        file_name += f".{file_ext}"
+    compression_arg = compression
+    if infer:
+        compression_arg = "infer"
+    if use_dict:
+        compression_arg = {"method": compression}
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 2)), columns=list("AB")
+    )
+    df.index.name = "index"
+    path = tmp_path / file_name
+    path.touch()
+    df.to_stata(path, version=version, compression=compression_arg)
+    if compression == "gzip":
+        with gzip.open(path, "rb") as comp:
+            fp = io.BytesIO(comp.read())
+    elif compression == "zip":
+        with zipfile.ZipFile(path, "r") as comp:
+            fp = io.BytesIO(comp.read(comp.filelist[0]))
+    elif compression == "tar":
+        with tarfile.open(path) as tar:
+            fp = io.BytesIO(tar.extractfile(tar.getnames()[0]).read())
+    elif compression == "bz2":
+        with bz2.open(path, "rb") as comp:
+            fp = io.BytesIO(comp.read())
+    elif compression == "zstd":
+        zstd = pytest.importorskip("zstandard")
+        with zstd.open(path, "rb") as comp:
+            fp = io.BytesIO(comp.read())
+    elif compression == "xz":
+        lzma = pytest.importorskip("lzma")
+        with lzma.open(path, "rb") as comp:
+            fp = io.BytesIO(comp.read())
+    elif compression is None:
+        fp = path
+    reread = read_stata(fp, index_col="index")
+
+    expected = df
+    tm.assert_frame_equal(reread, expected)
+
+
+@pytest.mark.parametrize("method", ["zip", "infer"])
+@pytest.mark.parametrize("file_ext", [None, "dta", "zip"])
+def test_compression_dict(method, file_ext, tmp_path):
+    file_name = f"test.{file_ext}"
+    archive_name = "test.dta"
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 2)), columns=list("AB")
+    )
+    df.index.name = "index"
+    compression = {"method": method, "archive_name": archive_name}
+    path = tmp_path / file_name
+    path.touch()
+    df.to_stata(path, compression=compression)
+    if method == "zip" or file_ext == "zip":
+        with zipfile.ZipFile(path, "r") as zp:
+            assert len(zp.filelist) == 1
+            assert zp.filelist[0].filename == archive_name
+            fp = io.BytesIO(zp.read(zp.filelist[0]))
+    else:
+        fp = path
+    reread = read_stata(fp, index_col="index")
+
+    expected = df
+    tm.assert_frame_equal(reread, expected)
+
+
+@pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+def test_chunked_categorical(version, temp_file):
+    df = DataFrame({"cats": Series(["a", "b", "a", "b", "c"], dtype="category")})
+    df.index.name = "index"
+
+    expected = df.copy()
+
+    df.to_stata(temp_file, version=version)
+    with StataReader(temp_file, chunksize=2, order_categoricals=False) as reader:
+        for i, block in enumerate(reader):
+            block = block.set_index("index")
+            assert "cats" in block
+            tm.assert_series_equal(
+                block.cats,
+                expected.cats.iloc[2 * i : 2 * (i + 1)],
+                check_index_type=len(block) > 1,
+            )
+
+
+def test_chunked_categorical_partial(datapath):
+    dta_file = datapath("io", "data", "stata", "stata-dta-partially-labeled.dta")
+    values = ["a", "b", "a", "b", 3.0]
+    msg = "series with value labels are not fully labeled"
+    with StataReader(dta_file, chunksize=2) as reader:
+        with tm.assert_produces_warning(CategoricalConversionWarning, match=msg):
+            for i, block in enumerate(reader):
+                assert list(block.cats) == values[2 * i : 2 * (i + 1)]
+                if i < 2:
+                    idx = pd.Index(["a", "b"])
+                else:
+                    idx = pd.Index([3.0], dtype="float64")
+                tm.assert_index_equal(block.cats.cat.categories, idx)
+    with tm.assert_produces_warning(CategoricalConversionWarning, match=msg):
+        with StataReader(dta_file, chunksize=5) as reader:
+            large_chunk = reader.__next__()
+    direct = read_stata(dta_file)
+    tm.assert_frame_equal(direct, large_chunk)
+
+
+@pytest.mark.parametrize("chunksize", (-1, 0, "apple"))
+def test_iterator_errors(datapath, chunksize):
+    dta_file = datapath("io", "data", "stata", "stata-dta-partially-labeled.dta")
+    with pytest.raises(ValueError, match="chunksize must be a positive"):
+        with StataReader(dta_file, chunksize=chunksize):
+            pass
+
+
+def test_iterator_value_labels(temp_file):
+    # GH 31544
+    values = ["c_label", "b_label"] + ["a_label"] * 500
+    df = DataFrame({f"col{k}": pd.Categorical(values, ordered=True) for k in range(2)})
+    df.to_stata(temp_file, write_index=False)
+    expected = pd.Index(["a_label", "b_label", "c_label"])
+    with read_stata(temp_file, chunksize=100) as reader:
+        for j, chunk in enumerate(reader):
+            for i in range(2):
+                tm.assert_index_equal(chunk.dtypes.iloc[i].categories, expected)
+            tm.assert_frame_equal(chunk, df.iloc[j * 100 : (j + 1) * 100])
+
+
+def test_precision_loss(temp_file):
+    df = DataFrame(
+        [[sum(2**i for i in range(60)), sum(2**i for i in range(52))]],
+        columns=["big", "little"],
+    )
+    with tm.assert_produces_warning(
+        PossiblePrecisionLoss, match="Column converted from int64 to float64"
+    ):
+        df.to_stata(temp_file, write_index=False)
+    reread = read_stata(temp_file)
+    expected_dt = Series([np.float64, np.float64], index=["big", "little"])
+    tm.assert_series_equal(reread.dtypes, expected_dt)
+    assert reread.loc[0, "little"] == df.loc[0, "little"]
+    assert reread.loc[0, "big"] == float(df.loc[0, "big"])
+
+
+def test_compression_roundtrip(compression, temp_file):
+    df = DataFrame(
+        [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+        index=["A", "B"],
+        columns=["X", "Y", "Z"],
+    )
+    df.index.name = "index"
+
+    df.to_stata(temp_file, compression=compression)
+    reread = read_stata(temp_file, compression=compression, index_col="index")
+    tm.assert_frame_equal(df, reread)
+
+    # explicitly ensure file was compressed.
+    with tm.decompress_file(temp_file, compression) as fh:
+        contents = io.BytesIO(fh.read())
+    reread = read_stata(contents, index_col="index")
+    tm.assert_frame_equal(df, reread)
+
+
+@pytest.mark.parametrize("to_infer", [True, False])
+@pytest.mark.parametrize("read_infer", [True, False])
+def test_stata_compression(
+    compression_only, read_infer, to_infer, compression_to_extension, tmp_path
+):
+    compression = compression_only
+
+    ext = compression_to_extension[compression]
+    filename = f"test.{ext}"
+
+    df = DataFrame(
+        [[0.123456, 0.234567, 0.567567], [12.32112, 123123.2, 321321.2]],
+        index=["A", "B"],
+        columns=["X", "Y", "Z"],
+    )
+    df.index.name = "index"
+
+    to_compression = "infer" if to_infer else compression
+    read_compression = "infer" if read_infer else compression
+
+    path = tmp_path / filename
+    path.touch()
+    df.to_stata(path, compression=to_compression)
+    result = read_stata(path, compression=read_compression, index_col="index")
+    tm.assert_frame_equal(result, df)
+
+
+def test_non_categorical_value_labels(temp_file):
+    data = DataFrame(
+        {
+            "fully_labelled": [1, 2, 3, 3, 1],
+            "partially_labelled": [1.0, 2.0, np.nan, 9.0, np.nan],
+            "Y": [7, 7, 9, 8, 10],
+            "Z": pd.Categorical(["j", "k", "l", "k", "j"]),
+        }
+    )
+
+    path = temp_file
+    value_labels = {
+        "fully_labelled": {1: "one", 2: "two", 3: "three"},
+        "partially_labelled": {1.0: "one", 2.0: "two"},
+    }
+    expected = {**value_labels, "Z": {0: "j", 1: "k", 2: "l"}}
+
+    writer = StataWriter(path, data, value_labels=value_labels)
+    writer.write_file()
+
+    with StataReader(path) as reader:
+        reader_value_labels = reader.value_labels()
+        assert reader_value_labels == expected
+
+    msg = "Can't create value labels for notY, it wasn't found in the dataset."
+    value_labels = {"notY": {7: "label1", 8: "label2"}}
+    with pytest.raises(KeyError, match=msg):
+        StataWriter(path, data, value_labels=value_labels)
+
+    msg = (
+        "Can't create value labels for Z, value labels "
+        "can only be applied to numeric columns."
+    )
+    value_labels = {"Z": {1: "a", 2: "k", 3: "j", 4: "i"}}
+    with pytest.raises(ValueError, match=msg):
+        StataWriter(path, data, value_labels=value_labels)
+
+
+def test_non_categorical_value_label_name_conversion(temp_file):
+    # Check conversion of invalid variable names
+    data = DataFrame(
+        {
+            "invalid~!": [1, 1, 2, 3, 5, 8],  # Only alphanumeric and _
+            "6_invalid": [1, 1, 2, 3, 5, 8],  # Must start with letter or _
+            "invalid_name_longer_than_32_characters": [8, 8, 9, 9, 8, 8],  # Too long
+            "aggregate": [2, 5, 5, 6, 6, 9],  # Reserved words
+            (1, 2): [1, 2, 3, 4, 5, 6],  # Hashable non-string
+        }
+    )
+
+    value_labels = {
+        "invalid~!": {1: "label1", 2: "label2"},
+        "6_invalid": {1: "label1", 2: "label2"},
+        "invalid_name_longer_than_32_characters": {8: "eight", 9: "nine"},
+        "aggregate": {5: "five"},
+        (1, 2): {3: "three"},
+    }
+
+    expected = {
+        "invalid__": {1: "label1", 2: "label2"},
+        "_6_invalid": {1: "label1", 2: "label2"},
+        "invalid_name_longer_than_32_char": {8: "eight", 9: "nine"},
+        "_aggregate": {5: "five"},
+        "_1__2_": {3: "three"},
+    }
+
+    msg = "Not all pandas column names were valid Stata variable names"
+    with tm.assert_produces_warning(InvalidColumnName, match=msg):
+        data.to_stata(temp_file, value_labels=value_labels)
+
+    with StataReader(temp_file) as reader:
+        reader_value_labels = reader.value_labels()
+        assert reader_value_labels == expected
+
+
+def test_non_categorical_value_label_convert_categoricals_error(temp_file):
+    # Mapping more than one value to the same label is valid for Stata
+    # labels, but can't be read with convert_categoricals=True
+    value_labels = {
+        "repeated_labels": {10: "Ten", 20: "More than ten", 40: "More than ten"}
+    }
+
+    data = DataFrame(
+        {
+            "repeated_labels": [10, 10, 20, 20, 40, 40],
+        }
+    )
+
+    data.to_stata(temp_file, value_labels=value_labels)
+
+    with StataReader(temp_file, convert_categoricals=False) as reader:
+        reader_value_labels = reader.value_labels()
+    assert reader_value_labels == value_labels
+
+    col = "repeated_labels"
+    repeats = "-" * 80 + "\n" + "\n".join(["More than ten"])
+
+    msg = f"""
+Value labels for column {col} are not unique. These cannot be converted to
+pandas categoricals.
+
+Either read the file with `convert_categoricals` set to False or use the
+low level interface in `StataReader` to separately read the values and the
+value_labels.
+
+The repeated labels are:
+{repeats}
+"""
+    with pytest.raises(ValueError, match=msg):
+        read_stata(temp_file, convert_categoricals=True)
+
+
+@pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        pd.BooleanDtype,
+        pd.Int8Dtype,
+        pd.Int16Dtype,
+        pd.Int32Dtype,
+        pd.Int64Dtype,
+        pd.UInt8Dtype,
+        pd.UInt16Dtype,
+        pd.UInt32Dtype,
+        pd.UInt64Dtype,
+    ],
+)
+def test_nullable_support(dtype, version, temp_file):
+    df = DataFrame(
+        {
+            "a": Series([1.0, 2.0, 3.0]),
+            "b": Series([1, pd.NA, pd.NA], dtype=dtype.name),
+            "c": Series(["a", "b", None]),
+        }
+    )
+    dtype_name = df.b.dtype.numpy_dtype.name
+    # Only use supported names: no uint, bool or int64
+    dtype_name = dtype_name.replace("u", "")
+    if dtype_name == "int64":
+        dtype_name = "int32"
+    elif dtype_name == "bool":
+        dtype_name = "int8"
+    value = StataMissingValue.BASE_MISSING_VALUES[dtype_name]
+    smv = StataMissingValue(value)
+    expected_b = Series([1, smv, smv], dtype=object, name="b")
+    expected_c = Series(["a", "b", ""], name="c")
+    df.to_stata(temp_file, write_index=False, version=version)
+    reread = read_stata(temp_file, convert_missing=True)
+    tm.assert_series_equal(df.a, reread.a)
+    tm.assert_series_equal(reread.b, expected_b)
+    tm.assert_series_equal(reread.c, expected_c)
+
+
+def test_empty_frame(temp_file):
+    # GH 46240
+    # create an empty DataFrame with int64 and float64 dtypes
+    df = DataFrame(data={"a": range(3), "b": [1.0, 2.0, 3.0]}).head(0)
+    path = temp_file
+    df.to_stata(path, write_index=False, version=117)
+    # Read entire dataframe
+    df2 = read_stata(path)
+    assert "b" in df2
+    # Dtypes don't match since no support for int32
+    dtypes = Series({"a": np.dtype("int32"), "b": np.dtype("float64")})
+    tm.assert_series_equal(df2.dtypes, dtypes)
+    # read one column of empty .dta file
+    df3 = read_stata(path, columns=["a"])
+    assert "b" not in df3
+    tm.assert_series_equal(df3.dtypes, dtypes.loc[["a"]])
+
+
+@pytest.mark.parametrize("version", [114, 117, 118, 119, None])
+def test_many_strl(temp_file, version):
+    n = 65534
+    df = DataFrame(np.arange(n), columns=["col"])
+    lbls = ["".join(v) for v in itertools.product(*([string.ascii_letters] * 3))]
+    value_labels = {"col": {i: lbls[i] for i in range(n)}}
+    df.to_stata(temp_file, value_labels=value_labels, version=version)
diff --git a/pandas/pandas/tests/io/xml/__init__.py b/pandas/pandas/tests/io/xml/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/io/xml/conftest.py b/pandas/pandas/tests/io/xml/conftest.py
new file mode 100644
index 0000000000000..8900b0dc46754
--- /dev/null
+++ b/pandas/pandas/tests/io/xml/conftest.py
@@ -0,0 +1,105 @@
+from pathlib import Path
+
+import pytest
+
+
+@pytest.fixture
+def xml_data_path():
+    """
+    Returns a Path object to the XML example directory.
+
+    Examples
+    --------
+    >>> def test_read_xml(xml_data_path):
+    ...     pd.read_xml(xml_data_path / "file.xsl")
+    """
+    return Path(__file__).parent.parent / "data" / "xml"
+
+
+@pytest.fixture
+def xml_books(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `books.xml` example file.
+
+    Examples
+    --------
+    >>> def test_read_xml(xml_books):
+    ...     pd.read_xml(xml_books)
+    """
+    return datapath(xml_data_path / "books.xml")
+
+
+@pytest.fixture
+def xml_doc_ch_utf(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `doc_ch_utf.xml` example file.
+
+    Examples
+    --------
+    >>> def test_read_xml(xml_doc_ch_utf):
+    ...     pd.read_xml(xml_doc_ch_utf)
+    """
+    return datapath(xml_data_path / "doc_ch_utf.xml")
+
+
+@pytest.fixture
+def xml_baby_names(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `baby_names.xml` example file.
+
+    Examples
+    --------
+    >>> def test_read_xml(xml_baby_names):
+    ...     pd.read_xml(xml_baby_names)
+    """
+    return datapath(xml_data_path / "baby_names.xml")
+
+
+@pytest.fixture
+def kml_cta_rail_lines(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `cta_rail_lines.kml` example file.
+
+    Examples
+    --------
+    >>> def test_read_xml(kml_cta_rail_lines):
+    ...     pd.read_xml(
+    ...         kml_cta_rail_lines,
+    ...         xpath=".//k:Placemark",
+    ...         namespaces={"k": "http://www.opengis.net/kml/2.2"},
+    ...         stylesheet=xsl_flatten_doc,
+    ...     )
+    """
+    return datapath(xml_data_path / "cta_rail_lines.kml")
+
+
+@pytest.fixture
+def xsl_flatten_doc(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `flatten_doc.xsl` example file.
+
+    Examples
+    --------
+    >>> def test_read_xsl(xsl_flatten_doc, mode):
+    ...     with open(
+    ...         xsl_flatten_doc, mode, encoding="utf-8" if mode == "r" else None
+    ...     ) as f:
+    ...         xsl_obj = f.read()
+    """
+    return datapath(xml_data_path / "flatten_doc.xsl")
+
+
+@pytest.fixture
+def xsl_row_field_output(xml_data_path, datapath):
+    """
+    Returns the path (as `str`) to the `row_field_output.xsl` example file.
+
+    Examples
+    --------
+    >>> def test_read_xsl(xsl_row_field_output, mode):
+    ...     with open(
+    ...         xsl_row_field_output, mode, encoding="utf-8" if mode == "r" else None
+    ...     ) as f:
+    ...         xsl_obj = f.read()
+    """
+    return datapath(xml_data_path / "row_field_output.xsl")
diff --git a/pandas/pandas/tests/io/xml/test_to_xml.py b/pandas/pandas/tests/io/xml/test_to_xml.py
new file mode 100644
index 0000000000000..50fef2c5eb4eb
--- /dev/null
+++ b/pandas/pandas/tests/io/xml/test_to_xml.py
@@ -0,0 +1,1368 @@
+from __future__ import annotations
+
+from io import (
+    BytesIO,
+    StringIO,
+)
+import os
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+)
+import pandas._testing as tm
+
+from pandas.io.common import get_handle
+from pandas.io.xml import read_xml
+
+# CHECKLIST
+
+# [x] - ValueError: "Values for parser can only be lxml or etree."
+
+# etree
+# [x] - ImportError: "lxml not found, please install or use the etree parser."
+# [X] - TypeError: "...is not a valid type for attr_cols"
+# [X] - TypeError: "...is not a valid type for elem_cols"
+# [X] - LookupError: "unknown encoding"
+# [X] - KeyError: "...is not included in namespaces"
+# [X] - KeyError: "no valid column"
+# [X] - ValueError: "To use stylesheet, you need lxml installed..."
+# []  - OSError: (NEED PERMISSION ISSUE, DISK FULL, ETC.)
+# [X] - FileNotFoundError: "No such file or directory"
+# [X] - PermissionError: "Forbidden"
+
+# lxml
+# [X] - TypeError: "...is not a valid type for attr_cols"
+# [X] - TypeError: "...is not a valid type for elem_cols"
+# [X] - LookupError: "unknown encoding"
+# []  - OSError: (NEED PERMISSION ISSUE, DISK FULL, ETC.)
+# [X] - FileNotFoundError: "No such file or directory"
+# [X] - KeyError: "...is not included in namespaces"
+# [X] - KeyError: "no valid column"
+# [X] - ValueError: "stylesheet is not a url, file, or xml string."
+# []  - LookupError: (NEED WRONG ENCODING FOR FILE OUTPUT)
+# []  - URLError: (USUALLY DUE TO NETWORKING)
+# []  - HTTPError: (NEED AN ONLINE STYLESHEET)
+# [X] - OSError: "failed to load external entity"
+# [X] - XMLSyntaxError: "Opening and ending tag mismatch"
+# [X] - XSLTApplyError: "Cannot resolve URI"
+# [X] - XSLTParseError: "failed to compile"
+# [X] - PermissionError: "Forbidden"
+
+
+@pytest.fixture
+def geom_df():
+    return DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4, np.nan, 3],
+        }
+    )
+
+
+@pytest.fixture
+def planet_df():
+    return DataFrame(
+        {
+            "planet": [
+                "Mercury",
+                "Venus",
+                "Earth",
+                "Mars",
+                "Jupiter",
+                "Saturn",
+                "Uranus",
+                "Neptune",
+            ],
+            "type": [
+                "terrestrial",
+                "terrestrial",
+                "terrestrial",
+                "terrestrial",
+                "gas giant",
+                "gas giant",
+                "ice giant",
+                "ice giant",
+            ],
+            "location": [
+                "inner",
+                "inner",
+                "inner",
+                "inner",
+                "outer",
+                "outer",
+                "outer",
+                "outer",
+            ],
+            "mass": [
+                0.330114,
+                4.86747,
+                5.97237,
+                0.641712,
+                1898.187,
+                568.3174,
+                86.8127,
+                102.4126,
+            ],
+        }
+    )
+
+
+@pytest.fixture
+def from_file_expected():
+    return """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <index>0</index>
+    <category>cooking</category>
+    <title>Everyday Italian</title>
+    <author>Giada De Laurentiis</author>
+    <year>2005</year>
+    <price>30.0</price>
+  </row>
+  <row>
+    <index>1</index>
+    <category>children</category>
+    <title>Harry Potter</title>
+    <author>J K. Rowling</author>
+    <year>2005</year>
+    <price>29.99</price>
+  </row>
+  <row>
+    <index>2</index>
+    <category>web</category>
+    <title>Learning XML</title>
+    <author>Erik T. Ray</author>
+    <year>2003</year>
+    <price>39.95</price>
+  </row>
+</data>"""
+
+
+def equalize_decl(doc):
+    # etree and lxml differ on quotes and case in xml declaration
+    if doc is not None:
+        doc = doc.replace(
+            '<?xml version="1.0" encoding="utf-8"?',
+            "<?xml version='1.0' encoding='utf-8'?",
+        )
+    return doc
+
+
+@pytest.fixture(params=["rb", "r"])
+def mode(request):
+    return request.param
+
+
+@pytest.fixture(params=[pytest.param("lxml", marks=td.skip_if_no("lxml")), "etree"])
+def parser(request):
+    return request.param
+
+
+# FILE OUTPUT
+
+
+def test_file_output_str_read(xml_books, parser, from_file_expected):
+    df_file = read_xml(xml_books, parser=parser)
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(path, parser=parser)
+        with open(path, "rb") as f:
+            output = f.read().decode("utf-8").strip()
+
+        output = equalize_decl(output)
+
+        assert output == from_file_expected
+
+
+def test_file_output_bytes_read(xml_books, parser, from_file_expected):
+    df_file = read_xml(xml_books, parser=parser)
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(path, parser=parser)
+        with open(path, "rb") as f:
+            output = f.read().decode("utf-8").strip()
+
+        output = equalize_decl(output)
+
+        assert output == from_file_expected
+
+
+def test_str_output(xml_books, parser, from_file_expected):
+    df_file = read_xml(xml_books, parser=parser)
+
+    output = df_file.to_xml(parser=parser)
+    output = equalize_decl(output)
+
+    assert output == from_file_expected
+
+
+def test_wrong_file_path(parser, geom_df):
+    path = "/my/fake/path/output.xml"
+
+    with pytest.raises(
+        OSError,
+        match=(r"Cannot save file into a non-existent directory: .*path"),
+    ):
+        geom_df.to_xml(path, parser=parser)
+
+
+# INDEX
+
+
+def test_index_false(xml_books, parser):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <category>cooking</category>
+    <title>Everyday Italian</title>
+    <author>Giada De Laurentiis</author>
+    <year>2005</year>
+    <price>30.0</price>
+  </row>
+  <row>
+    <category>children</category>
+    <title>Harry Potter</title>
+    <author>J K. Rowling</author>
+    <year>2005</year>
+    <price>29.99</price>
+  </row>
+  <row>
+    <category>web</category>
+    <title>Learning XML</title>
+    <author>Erik T. Ray</author>
+    <year>2003</year>
+    <price>39.95</price>
+  </row>
+</data>"""
+
+    df_file = read_xml(xml_books, parser=parser)
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(path, index=False, parser=parser)
+        with open(path, "rb") as f:
+            output = f.read().decode("utf-8").strip()
+
+        output = equalize_decl(output)
+
+        assert output == expected
+
+
+def test_index_false_rename_row_root(xml_books, parser):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<books>
+  <book>
+    <category>cooking</category>
+    <title>Everyday Italian</title>
+    <author>Giada De Laurentiis</author>
+    <year>2005</year>
+    <price>30.0</price>
+  </book>
+  <book>
+    <category>children</category>
+    <title>Harry Potter</title>
+    <author>J K. Rowling</author>
+    <year>2005</year>
+    <price>29.99</price>
+  </book>
+  <book>
+    <category>web</category>
+    <title>Learning XML</title>
+    <author>Erik T. Ray</author>
+    <year>2003</year>
+    <price>39.95</price>
+  </book>
+</books>"""
+
+    df_file = read_xml(xml_books, parser=parser)
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(
+            path, index=False, root_name="books", row_name="book", parser=parser
+        )
+        with open(path, "rb") as f:
+            output = f.read().decode("utf-8").strip()
+
+        output = equalize_decl(output)
+
+        assert output == expected
+
+
+@pytest.mark.parametrize("typ", [int, str])
+def test_index_false_with_offset_input_index(parser, typ, geom_df):
+    """
+    Tests that the output does not contain the `<index>` field when the index of the
+    input Dataframe has an offset.
+
+    This is a regression test for issue #42458.
+    """
+
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+    offset_index = [typ(i) for i in range(10, 13)]
+    offset_geom_df = geom_df.copy()
+    offset_geom_df.index = Index(offset_index)
+    output = offset_geom_df.to_xml(index=False, parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# NA_REP
+
+na_expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+
+def test_na_elem_output(parser, geom_df):
+    output = geom_df.to_xml(parser=parser)
+    output = equalize_decl(output)
+
+    assert output == na_expected
+
+
+def test_na_empty_str_elem_option(parser, geom_df):
+    output = geom_df.to_xml(na_rep="", parser=parser)
+    output = equalize_decl(output)
+
+    assert output == na_expected
+
+
+def test_na_empty_elem_option(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides>0.0</sides>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(na_rep="0.0", parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# ATTR_COLS
+
+
+def test_attrs_cols_nan_output(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row index="0" shape="square" degrees="360" sides="4.0"/>
+  <row index="1" shape="circle" degrees="360"/>
+  <row index="2" shape="triangle" degrees="180" sides="3.0"/>
+</data>"""
+
+    output = geom_df.to_xml(attr_cols=["shape", "degrees", "sides"], parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_attrs_cols_prefix(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<doc:data xmlns:doc="http://example.xom">
+  <doc:row doc:index="0" doc:shape="square" \
+doc:degrees="360" doc:sides="4.0"/>
+  <doc:row doc:index="1" doc:shape="circle" \
+doc:degrees="360"/>
+  <doc:row doc:index="2" doc:shape="triangle" \
+doc:degrees="180" doc:sides="3.0"/>
+</doc:data>"""
+
+    output = geom_df.to_xml(
+        attr_cols=["index", "shape", "degrees", "sides"],
+        namespaces={"doc": "http://example.xom"},
+        prefix="doc",
+        parser=parser,
+    )
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_attrs_unknown_column(parser, geom_df):
+    with pytest.raises(KeyError, match=("no valid column")):
+        geom_df.to_xml(attr_cols=["shape", "degree", "sides"], parser=parser)
+
+
+def test_attrs_wrong_type(parser, geom_df):
+    with pytest.raises(TypeError, match=("is not a valid type for attr_cols")):
+        geom_df.to_xml(attr_cols='"shape", "degree", "sides"', parser=parser)
+
+
+# ELEM_COLS
+
+
+def test_elems_cols_nan_output(parser, geom_df):
+    elems_cols_expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+    <shape>square</shape>
+  </row>
+  <row>
+    <degrees>360</degrees>
+    <sides/>
+    <shape>circle</shape>
+  </row>
+  <row>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+    <shape>triangle</shape>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(
+        index=False, elem_cols=["degrees", "sides", "shape"], parser=parser
+    )
+    output = equalize_decl(output)
+
+    assert output == elems_cols_expected
+
+
+def test_elems_unknown_column(parser, geom_df):
+    with pytest.raises(KeyError, match=("no valid column")):
+        geom_df.to_xml(elem_cols=["shape", "degree", "sides"], parser=parser)
+
+
+def test_elems_wrong_type(parser, geom_df):
+    with pytest.raises(TypeError, match=("is not a valid type for elem_cols")):
+        geom_df.to_xml(elem_cols='"shape", "degree", "sides"', parser=parser)
+
+
+def test_elems_and_attrs_cols(parser, geom_df):
+    elems_cols_expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row shape="square">
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row shape="circle">
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row shape="triangle">
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(
+        index=False,
+        elem_cols=["degrees", "sides"],
+        attr_cols=["shape"],
+        parser=parser,
+    )
+    output = equalize_decl(output)
+
+    assert output == elems_cols_expected
+
+
+# HIERARCHICAL COLUMNS
+
+
+def test_hierarchical_columns(parser, planet_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <location>inner</location>
+    <type>terrestrial</type>
+    <count_mass>4</count_mass>
+    <sum_mass>11.81</sum_mass>
+    <mean_mass>2.95</mean_mass>
+  </row>
+  <row>
+    <location>outer</location>
+    <type>gas giant</type>
+    <count_mass>2</count_mass>
+    <sum_mass>2466.5</sum_mass>
+    <mean_mass>1233.25</mean_mass>
+  </row>
+  <row>
+    <location>outer</location>
+    <type>ice giant</type>
+    <count_mass>2</count_mass>
+    <sum_mass>189.23</sum_mass>
+    <mean_mass>94.61</mean_mass>
+  </row>
+  <row>
+    <location>All</location>
+    <type/>
+    <count_mass>8</count_mass>
+    <sum_mass>2667.54</sum_mass>
+    <mean_mass>333.44</mean_mass>
+  </row>
+</data>"""
+
+    pvt = planet_df.pivot_table(
+        index=["location", "type"],
+        values="mass",
+        aggfunc=["count", "sum", "mean"],
+        margins=True,
+    ).round(2)
+
+    output = pvt.to_xml(parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_hierarchical_attrs_columns(parser, planet_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row location="inner" type="terrestrial" count_mass="4" \
+sum_mass="11.81" mean_mass="2.95"/>
+  <row location="outer" type="gas giant" count_mass="2" \
+sum_mass="2466.5" mean_mass="1233.25"/>
+  <row location="outer" type="ice giant" count_mass="2" \
+sum_mass="189.23" mean_mass="94.61"/>
+  <row location="All" type="" count_mass="8" \
+sum_mass="2667.54" mean_mass="333.44"/>
+</data>"""
+
+    pvt = planet_df.pivot_table(
+        index=["location", "type"],
+        values="mass",
+        aggfunc=["count", "sum", "mean"],
+        margins=True,
+    ).round(2)
+
+    output = pvt.to_xml(attr_cols=list(pvt.reset_index().columns.values), parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# MULTIINDEX
+
+
+def test_multi_index(parser, planet_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <location>inner</location>
+    <type>terrestrial</type>
+    <count>4</count>
+    <sum>11.81</sum>
+    <mean>2.95</mean>
+  </row>
+  <row>
+    <location>outer</location>
+    <type>gas giant</type>
+    <count>2</count>
+    <sum>2466.5</sum>
+    <mean>1233.25</mean>
+  </row>
+  <row>
+    <location>outer</location>
+    <type>ice giant</type>
+    <count>2</count>
+    <sum>189.23</sum>
+    <mean>94.61</mean>
+  </row>
+</data>"""
+
+    agg = (
+        planet_df.groupby(["location", "type"])["mass"]
+        .agg(["count", "sum", "mean"])
+        .round(2)
+    )
+
+    output = agg.to_xml(parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_multi_index_attrs_cols(parser, planet_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row location="inner" type="terrestrial" count="4" \
+sum="11.81" mean="2.95"/>
+  <row location="outer" type="gas giant" count="2" \
+sum="2466.5" mean="1233.25"/>
+  <row location="outer" type="ice giant" count="2" \
+sum="189.23" mean="94.61"/>
+</data>"""
+
+    agg = (
+        planet_df.groupby(["location", "type"])["mass"]
+        .agg(["count", "sum", "mean"])
+        .round(2)
+    )
+    output = agg.to_xml(attr_cols=list(agg.reset_index().columns.values), parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# NAMESPACE
+
+
+def test_default_namespace(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data xmlns="http://example.com">
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(namespaces={"": "http://example.com"}, parser=parser)
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_unused_namespaces(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data xmlns:oth="http://other.org" xmlns:ex="http://example.com">
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(
+        namespaces={"oth": "http://other.org", "ex": "http://example.com"},
+        parser=parser,
+    )
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# PREFIX
+
+
+def test_namespace_prefix(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<doc:data xmlns:doc="http://example.com">
+  <doc:row>
+    <doc:index>0</doc:index>
+    <doc:shape>square</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides>4.0</doc:sides>
+  </doc:row>
+  <doc:row>
+    <doc:index>1</doc:index>
+    <doc:shape>circle</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides/>
+  </doc:row>
+  <doc:row>
+    <doc:index>2</doc:index>
+    <doc:shape>triangle</doc:shape>
+    <doc:degrees>180</doc:degrees>
+    <doc:sides>3.0</doc:sides>
+  </doc:row>
+</doc:data>"""
+
+    output = geom_df.to_xml(
+        namespaces={"doc": "http://example.com"}, prefix="doc", parser=parser
+    )
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+def test_missing_prefix_in_nmsp(parser, geom_df):
+    with pytest.raises(KeyError, match=("doc is not included in namespaces")):
+        geom_df.to_xml(
+            namespaces={"": "http://example.com"}, prefix="doc", parser=parser
+        )
+
+
+def test_namespace_prefix_and_default(parser, geom_df):
+    expected = """\
+<?xml version='1.0' encoding='utf-8'?>
+<doc:data xmlns:doc="http://other.org" xmlns="http://example.com">
+  <doc:row>
+    <doc:index>0</doc:index>
+    <doc:shape>square</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides>4.0</doc:sides>
+  </doc:row>
+  <doc:row>
+    <doc:index>1</doc:index>
+    <doc:shape>circle</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides/>
+  </doc:row>
+  <doc:row>
+    <doc:index>2</doc:index>
+    <doc:shape>triangle</doc:shape>
+    <doc:degrees>180</doc:degrees>
+    <doc:sides>3.0</doc:sides>
+  </doc:row>
+</doc:data>"""
+
+    output = geom_df.to_xml(
+        namespaces={"": "http://example.com", "doc": "http://other.org"},
+        prefix="doc",
+        parser=parser,
+    )
+    output = equalize_decl(output)
+
+    assert output == expected
+
+
+# ENCODING
+
+encoding_expected = """\
+<?xml version='1.0' encoding='ISO-8859-1'?>
+<data>
+  <row>
+    <index>0</index>
+    <rank>1</rank>
+    <malename>José</malename>
+    <femalename>Sofía</femalename>
+  </row>
+  <row>
+    <index>1</index>
+    <rank>2</rank>
+    <malename>Luis</malename>
+    <femalename>Valentina</femalename>
+  </row>
+  <row>
+    <index>2</index>
+    <rank>3</rank>
+    <malename>Carlos</malename>
+    <femalename>Isabella</femalename>
+  </row>
+  <row>
+    <index>3</index>
+    <rank>4</rank>
+    <malename>Juan</malename>
+    <femalename>Camila</femalename>
+  </row>
+  <row>
+    <index>4</index>
+    <rank>5</rank>
+    <malename>Jorge</malename>
+    <femalename>Valeria</femalename>
+  </row>
+</data>"""
+
+
+def test_encoding_option_str(xml_baby_names, parser):
+    df_file = read_xml(xml_baby_names, parser=parser, encoding="ISO-8859-1").head(5)
+
+    output = df_file.to_xml(encoding="ISO-8859-1", parser=parser)
+
+    if output is not None:
+        # etree and lxml differ on quotes and case in xml declaration
+        output = output.replace(
+            '<?xml version="1.0" encoding="ISO-8859-1"?',
+            "<?xml version='1.0' encoding='ISO-8859-1'?",
+        )
+
+    assert output == encoding_expected
+
+
+def test_correct_encoding_file(xml_baby_names):
+    pytest.importorskip("lxml")
+    df_file = read_xml(xml_baby_names, encoding="ISO-8859-1", parser="lxml")
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(path, index=False, encoding="ISO-8859-1", parser="lxml")
+
+
+@pytest.mark.parametrize("encoding", ["UTF-8", "UTF-16", "ISO-8859-1"])
+def test_wrong_encoding_option_lxml(xml_baby_names, parser, encoding):
+    pytest.importorskip("lxml")
+    df_file = read_xml(xml_baby_names, encoding="ISO-8859-1", parser="lxml")
+
+    with tm.ensure_clean("test.xml") as path:
+        df_file.to_xml(path, index=False, encoding=encoding, parser=parser)
+
+
+def test_misspelled_encoding(parser, geom_df):
+    with pytest.raises(LookupError, match=("unknown encoding")):
+        geom_df.to_xml(encoding="uft-8", parser=parser)
+
+
+# PRETTY PRINT
+
+
+def test_xml_declaration_pretty_print(geom_df):
+    pytest.importorskip("lxml")
+    expected = """\
+<data>
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+    output = geom_df.to_xml(xml_declaration=False)
+
+    assert output == expected
+
+
+def test_no_pretty_print_with_decl(parser, geom_df):
+    expected = (
+        "<?xml version='1.0' encoding='utf-8'?>\n"
+        "<data><row><index>0</index><shape>square</shape>"
+        "<degrees>360</degrees><sides>4.0</sides></row><row>"
+        "<index>1</index><shape>circle</shape><degrees>360"
+        "</degrees><sides/></row><row><index>2</index><shape>"
+        "triangle</shape><degrees>180</degrees><sides>3.0</sides>"
+        "</row></data>"
+    )
+
+    output = geom_df.to_xml(pretty_print=False, parser=parser)
+    output = equalize_decl(output)
+
+    # etree adds space for closed tags
+    if output is not None:
+        output = output.replace(" />", "/>")
+
+    assert output == expected
+
+
+def test_no_pretty_print_no_decl(parser, geom_df):
+    expected = (
+        "<data><row><index>0</index><shape>square</shape>"
+        "<degrees>360</degrees><sides>4.0</sides></row><row>"
+        "<index>1</index><shape>circle</shape><degrees>360"
+        "</degrees><sides/></row><row><index>2</index><shape>"
+        "triangle</shape><degrees>180</degrees><sides>3.0</sides>"
+        "</row></data>"
+    )
+
+    output = geom_df.to_xml(xml_declaration=False, pretty_print=False, parser=parser)
+
+    # etree adds space for closed tags
+    if output is not None:
+        output = output.replace(" />", "/>")
+
+    assert output == expected
+
+
+# PARSER
+
+
+@td.skip_if_installed("lxml")
+def test_default_parser_no_lxml(geom_df):
+    with pytest.raises(
+        ImportError, match=("lxml not found, please install or use the etree parser.")
+    ):
+        geom_df.to_xml()
+
+
+def test_unknown_parser(geom_df):
+    with pytest.raises(
+        ValueError, match=("Values for parser can only be lxml or etree.")
+    ):
+        geom_df.to_xml(parser="bs4")
+
+
+# STYLESHEET
+
+xsl_expected = """\
+<?xml version="1.0" encoding="utf-8"?>
+<data>
+  <row>
+    <field field="index">0</field>
+    <field field="shape">square</field>
+    <field field="degrees">360</field>
+    <field field="sides">4.0</field>
+  </row>
+  <row>
+    <field field="index">1</field>
+    <field field="shape">circle</field>
+    <field field="degrees">360</field>
+    <field field="sides"/>
+  </row>
+  <row>
+    <field field="index">2</field>
+    <field field="shape">triangle</field>
+    <field field="degrees">180</field>
+    <field field="sides">3.0</field>
+  </row>
+</data>"""
+
+
+def test_stylesheet_file_like(xsl_row_field_output, mode, geom_df):
+    pytest.importorskip("lxml")
+    with open(
+        xsl_row_field_output, mode, encoding="utf-8" if mode == "r" else None
+    ) as f:
+        assert geom_df.to_xml(stylesheet=f) == xsl_expected
+
+
+def test_stylesheet_io(xsl_row_field_output, mode, geom_df):
+    # note: By default the bodies of untyped functions are not checked,
+    # consider using --check-untyped-defs
+    pytest.importorskip("lxml")
+    xsl_obj: BytesIO | StringIO  # type: ignore[annotation-unchecked]
+
+    with open(
+        xsl_row_field_output, mode, encoding="utf-8" if mode == "r" else None
+    ) as f:
+        if mode == "rb":
+            xsl_obj = BytesIO(f.read())
+        else:
+            xsl_obj = StringIO(f.read())
+
+    output = geom_df.to_xml(stylesheet=xsl_obj)
+
+    assert output == xsl_expected
+
+
+def test_stylesheet_buffered_reader(xsl_row_field_output, mode, geom_df):
+    pytest.importorskip("lxml")
+    with open(
+        xsl_row_field_output, mode, encoding="utf-8" if mode == "r" else None
+    ) as f:
+        output = geom_df.to_xml(stylesheet=f)
+
+    assert output == xsl_expected
+
+
+def test_stylesheet_wrong_path(geom_df):
+    pytest.importorskip("lxml.etree")
+
+    xsl = os.path.join("does", "not", "exist", "row_field_output.xslt")
+
+    with pytest.raises(
+        FileNotFoundError, match=r"\[Errno 2\] No such file or director"
+    ):
+        geom_df.to_xml(stylesheet=xsl)
+
+
+@pytest.mark.parametrize("val", [StringIO(""), BytesIO(b"")])
+def test_empty_string_stylesheet(val, geom_df):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    msg = "|".join(
+        [
+            "Document is empty",
+            "Start tag expected, '<' not found",
+            # Seen on Mac with lxml 4.9.1
+            r"None \(line 0\)",
+        ]
+    )
+
+    with pytest.raises(lxml_etree.XMLSyntaxError, match=msg):
+        geom_df.to_xml(stylesheet=val)
+
+
+def test_incorrect_xsl_syntax(geom_df):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="xml" encoding="utf-8" indent="yes" >
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="@*|node()">
+        <xsl:copy>
+            <xsl:apply-templates select="@*|node()"/>
+        </xsl:copy>
+    </xsl:template>
+
+    <xsl:template match="row/*">
+        <field>
+            <xsl:attribute name="field">
+                <xsl:value-of select="name()"/>
+            </xsl:attribute>
+            <xsl:value-of select="text()"/>
+        </field>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    with pytest.raises(
+        lxml_etree.XMLSyntaxError, match="Opening and ending tag mismatch"
+    ):
+        geom_df.to_xml(stylesheet=StringIO(xsl))
+
+
+def test_incorrect_xsl_eval(geom_df):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="xml" encoding="utf-8" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="@*|node(*)">
+        <xsl:copy>
+            <xsl:apply-templates select="@*|node()"/>
+        </xsl:copy>
+    </xsl:template>
+
+    <xsl:template match="row/*">
+        <field>
+            <xsl:attribute name="field">
+                <xsl:value-of select="name()"/>
+            </xsl:attribute>
+            <xsl:value-of select="text()"/>
+        </field>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    with pytest.raises(lxml_etree.XSLTParseError, match="failed to compile"):
+        geom_df.to_xml(stylesheet=StringIO(xsl))
+
+
+def test_incorrect_xsl_apply(geom_df):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="xml" encoding="utf-8" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="@*|node()">
+        <xsl:copy>
+            <xsl:copy-of select="document('non_existent.xml')/*"/>
+        </xsl:copy>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    with pytest.raises(lxml_etree.XSLTApplyError, match="Cannot resolve URI"):
+        with tm.ensure_clean("test.xml") as path:
+            geom_df.to_xml(path, stylesheet=StringIO(xsl))
+
+
+def test_stylesheet_with_etree(geom_df):
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="xml" encoding="utf-8" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="@*|node(*)">
+        <xsl:copy>
+            <xsl:apply-templates select="@*|node()"/>
+        </xsl:copy>
+    </xsl:template>"""
+
+    with pytest.raises(ValueError, match="To use stylesheet, you need lxml installed"):
+        geom_df.to_xml(parser="etree", stylesheet=StringIO(xsl))
+
+
+def test_style_to_csv(geom_df):
+    pytest.importorskip("lxml")
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="text" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:param name="delim">,</xsl:param>
+    <xsl:template match="/data">
+        <xsl:text>,shape,degrees,sides&#xa;</xsl:text>
+        <xsl:apply-templates select="row"/>
+    </xsl:template>
+
+    <xsl:template match="row">
+        <xsl:value-of select="concat(index, $delim, shape, $delim,
+                                     degrees, $delim, sides)"/>
+         <xsl:text>&#xa;</xsl:text>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    out_csv = geom_df.to_csv(lineterminator="\n")
+
+    if out_csv is not None:
+        out_csv = out_csv.strip()
+    out_xml = geom_df.to_xml(stylesheet=StringIO(xsl))
+
+    assert out_csv == out_xml
+
+
+def test_style_to_string(geom_df):
+    pytest.importorskip("lxml")
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="text" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:param name="delim"><xsl:text>               </xsl:text></xsl:param>
+    <xsl:template match="/data">
+        <xsl:text>      shape  degrees  sides&#xa;</xsl:text>
+        <xsl:apply-templates select="row"/>
+    </xsl:template>
+
+    <xsl:template match="row">
+        <xsl:value-of select="concat(index, ' ',
+                                     substring($delim, 1, string-length('triangle')
+                                               - string-length(shape) + 1),
+                                     shape,
+                                     substring($delim, 1, string-length(name(degrees))
+                                               - string-length(degrees) + 2),
+                                     degrees,
+                                     substring($delim, 1, string-length(name(sides))
+                                               - string-length(sides) + 2),
+                                     sides)"/>
+         <xsl:text>&#xa;</xsl:text>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    out_str = geom_df.to_string()
+    out_xml = geom_df.to_xml(na_rep="NaN", stylesheet=StringIO(xsl))
+
+    assert out_xml == out_str
+
+
+def test_style_to_json(geom_df):
+    pytest.importorskip("lxml")
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="text" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:param name="quot">"</xsl:param>
+
+    <xsl:template match="/data">
+        <xsl:text>{"shape":{</xsl:text>
+        <xsl:apply-templates select="descendant::row/shape"/>
+        <xsl:text>},"degrees":{</xsl:text>
+        <xsl:apply-templates select="descendant::row/degrees"/>
+        <xsl:text>},"sides":{</xsl:text>
+        <xsl:apply-templates select="descendant::row/sides"/>
+        <xsl:text>}}</xsl:text>
+    </xsl:template>
+
+    <xsl:template match="shape|degrees|sides">
+        <xsl:variable name="val">
+            <xsl:if test = ".=''">
+                <xsl:value-of select="'null'"/>
+            </xsl:if>
+            <xsl:if test = "number(text()) = text()">
+                <xsl:value-of select="text()"/>
+            </xsl:if>
+            <xsl:if test = "number(text()) != text()">
+                <xsl:value-of select="concat($quot, text(), $quot)"/>
+            </xsl:if>
+        </xsl:variable>
+        <xsl:value-of select="concat($quot, preceding-sibling::index,
+                                     $quot,':', $val)"/>
+        <xsl:if test="preceding-sibling::index != //row[last()]/index">
+            <xsl:text>,</xsl:text>
+        </xsl:if>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    out_json = geom_df.to_json()
+    out_xml = geom_df.to_xml(stylesheet=StringIO(xsl))
+
+    assert out_json == out_xml
+
+
+# COMPRESSION
+
+
+geom_xml = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <index>0</index>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4.0</sides>
+  </row>
+  <row>
+    <index>1</index>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <index>2</index>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+
+def test_compression_output(parser, compression_only, geom_df):
+    with tm.ensure_clean() as path:
+        geom_df.to_xml(path, parser=parser, compression=compression_only)
+
+        with get_handle(
+            path,
+            "r",
+            compression=compression_only,
+        ) as handle_obj:
+            output = handle_obj.handle.read()
+
+    output = equalize_decl(output)
+
+    assert geom_xml == output.strip()
+
+
+def test_filename_and_suffix_comp(
+    parser, compression_only, geom_df, compression_to_extension
+):
+    compfile = "xml." + compression_to_extension[compression_only]
+    with tm.ensure_clean(filename=compfile) as path:
+        geom_df.to_xml(path, parser=parser, compression=compression_only)
+
+        with get_handle(
+            path,
+            "r",
+            compression=compression_only,
+        ) as handle_obj:
+            output = handle_obj.handle.read()
+
+    output = equalize_decl(output)
+
+    assert geom_xml == output.strip()
+
+
+def test_ea_dtypes(any_numeric_ea_dtype, parser):
+    # GH#43903
+    expected = """<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <index>0</index>
+    <a/>
+  </row>
+</data>"""
+    df = DataFrame({"a": [NA]}).astype(any_numeric_ea_dtype)
+    result = df.to_xml(parser=parser)
+    assert equalize_decl(result).strip() == expected
+
+
+def test_unsuported_compression(parser, geom_df):
+    with pytest.raises(ValueError, match="Unrecognized compression type"):
+        with tm.ensure_clean() as path:
+            geom_df.to_xml(path, parser=parser, compression="7z")
+
+
+# STORAGE OPTIONS
+
+
+@pytest.mark.single_cpu
+def test_s3_permission_output(parser, s3_public_bucket, geom_df):
+    s3fs = pytest.importorskip("s3fs")
+    pytest.importorskip("lxml")
+
+    with tm.external_error_raised((PermissionError, FileNotFoundError)):
+        fs = s3fs.S3FileSystem(anon=True)
+        fs.ls(s3_public_bucket.name)
+
+        geom_df.to_xml(
+            f"s3://{s3_public_bucket.name}/geom.xml", compression="zip", parser=parser
+        )
diff --git a/pandas/pandas/tests/io/xml/test_xml.py b/pandas/pandas/tests/io/xml/test_xml.py
new file mode 100644
index 0000000000000..d897d251909fe
--- /dev/null
+++ b/pandas/pandas/tests/io/xml/test_xml.py
@@ -0,0 +1,2063 @@
+from __future__ import annotations
+
+from io import (
+    BytesIO,
+    StringIO,
+)
+from lzma import LZMAError
+import os
+from tarfile import ReadError
+from urllib.error import HTTPError
+from xml.etree.ElementTree import ParseError
+from zipfile import BadZipFile
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+from pandas.compat._optional import import_optional_dependency
+from pandas.errors import (
+    EmptyDataError,
+    ParserError,
+)
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    NA,
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+from pandas.io.common import get_handle
+from pandas.io.xml import read_xml
+
+# CHECK LIST
+
+# [x] - ValueError: "Values for parser can only be lxml or etree."
+
+# etree
+# [X] - ImportError: "lxml not found, please install or use the etree parser."
+# [X] - TypeError: "expected str, bytes or os.PathLike object, not NoneType"
+# [X] - ValueError: "Either element or attributes can be parsed not both."
+# [X] - ValueError: "xpath does not return any nodes..."
+# [X] - SyntaxError: "You have used an incorrect or unsupported XPath"
+# [X] - ValueError: "names does not match length of child elements in xpath."
+# [X] - TypeError: "...is not a valid type for names"
+# [X] - ValueError: "To use stylesheet, you need lxml installed..."
+# []  - URLError: (GENERAL ERROR WITH HTTPError AS SUBCLASS)
+# [X] - HTTPError: "HTTP Error 404: Not Found"
+# []  - OSError: (GENERAL ERROR WITH FileNotFoundError AS SUBCLASS)
+# [X] - FileNotFoundError: "No such file or directory"
+# []  - ParseError    (FAILSAFE CATCH ALL FOR VERY COMPLEX XML)
+# [X] - UnicodeDecodeError: "'utf-8' codec can't decode byte 0xe9..."
+# [X] - UnicodeError: "UTF-16 stream does not start with BOM"
+# [X] - BadZipFile: "File is not a zip file"
+# [X] - OSError: "Invalid data stream"
+# [X] - LZMAError: "Input format not supported by decoder"
+# [X] - ValueError: "Unrecognized compression type"
+# [X] - PermissionError: "Forbidden"
+
+# lxml
+# [X] - ValueError: "Either element or attributes can be parsed not both."
+# [X] - AttributeError: "__enter__"
+# [X] - XSLTApplyError: "Cannot resolve URI"
+# [X] - XSLTParseError: "document is not a stylesheet"
+# [X] - ValueError: "xpath does not return any nodes."
+# [X] - XPathEvalError: "Invalid expression"
+# []  - XPathSyntaxError: (OLD VERSION IN lxml FOR XPATH ERRORS)
+# [X] - TypeError: "empty namespace prefix is not supported in XPath"
+# [X] - ValueError: "names does not match length of child elements in xpath."
+# [X] - TypeError: "...is not a valid type for names"
+# [X] - LookupError: "unknown encoding"
+# []  - URLError: (USUALLY DUE TO NETWORKING)
+# [X  - HTTPError: "HTTP Error 404: Not Found"
+# [X] - OSError: "failed to load external entity"
+# [X] - XMLSyntaxError: "Start tag expected, '<' not found"
+# []  - ParserError: (FAILSAFE CATCH ALL FOR VERY COMPLEX XML
+# [X] - ValueError: "Values for parser can only be lxml or etree."
+# [X] - UnicodeDecodeError: "'utf-8' codec can't decode byte 0xe9..."
+# [X] - UnicodeError: "UTF-16 stream does not start with BOM"
+# [X] - BadZipFile: "File is not a zip file"
+# [X] - OSError: "Invalid data stream"
+# [X] - LZMAError: "Input format not supported by decoder"
+# [X] - ValueError: "Unrecognized compression type"
+# [X] - PermissionError: "Forbidden"
+
+geom_df = DataFrame(
+    {
+        "shape": ["square", "circle", "triangle"],
+        "degrees": [360, 360, 180],
+        "sides": [4, np.nan, 3],
+    }
+)
+
+xml_default_nmsp = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data xmlns="http://example.com">
+  <row>
+    <shape>square</shape>
+    <degrees>360</degrees>
+    <sides>4</sides>
+  </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>180</degrees>
+    <sides>3</sides>
+  </row>
+</data>"""
+
+xml_prefix_nmsp = """\
+<?xml version='1.0' encoding='utf-8'?>
+<doc:data xmlns:doc="http://example.com">
+  <doc:row>
+    <doc:shape>square</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides>4.0</doc:sides>
+  </doc:row>
+  <doc:row>
+    <doc:shape>circle</doc:shape>
+    <doc:degrees>360</doc:degrees>
+    <doc:sides/>
+  </doc:row>
+  <doc:row>
+    <doc:shape>triangle</doc:shape>
+    <doc:degrees>180</doc:degrees>
+    <doc:sides>3.0</doc:sides>
+  </doc:row>
+</doc:data>"""
+
+
+df_kml = DataFrame(
+    {
+        "id": {
+            0: "ID_00001",
+            1: "ID_00002",
+            2: "ID_00003",
+            3: "ID_00004",
+            4: "ID_00005",
+        },
+        "name": {
+            0: "Blue Line (Forest Park)",
+            1: "Red, Purple Line",
+            2: "Red, Purple Line",
+            3: "Red, Purple Line",
+            4: "Red, Purple Line",
+        },
+        "styleUrl": {
+            0: "#LineStyle01",
+            1: "#LineStyle01",
+            2: "#LineStyle01",
+            3: "#LineStyle01",
+            4: "#LineStyle01",
+        },
+        "extrude": {0: 0, 1: 0, 2: 0, 3: 0, 4: 0},
+        "altitudeMode": {
+            0: "clampedToGround",
+            1: "clampedToGround",
+            2: "clampedToGround",
+            3: "clampedToGround",
+            4: "clampedToGround",
+        },
+        "coordinates": {
+            0: (
+                "-87.77678526964958,41.8708863930319,0 "
+                "-87.77826234150609,41.87097820122218,0 "
+                "-87.78251583439344,41.87130129991005,0 "
+                "-87.78418294588424,41.87145055520308,0 "
+                "-87.7872369165933,41.8717239119163,0 "
+                "-87.79160214925886,41.87210797280065,0"
+            ),
+            1: (
+                "-87.65758750947528,41.96427269188822,0 "
+                "-87.65802133507393,41.96581929055245,0 "
+                "-87.65819033925305,41.96621846093642,0 "
+                "-87.6583189819129,41.96650362897086,0 "
+                "-87.65835858701473,41.96669002089185,0 "
+                "-87.65838428411853,41.96688150295095,0 "
+                "-87.65842208882658,41.96745896091846,0 "
+                "-87.65846556843937,41.9683761425439,0 "
+                "-87.65849296214573,41.96913893870342,0"
+            ),
+            2: (
+                "-87.65492939166126,41.95377494531437,0 "
+                "-87.65557043199591,41.95376544118533,0 "
+                "-87.65606302030132,41.95376391658746,0 "
+                "-87.65623502146268,41.95377379126367,0 "
+                "-87.65634748981634,41.95380103566435,0 "
+                "-87.65646537904269,41.95387703994676,0 "
+                "-87.65656532461145,41.95396622645799,0 "
+                "-87.65664760856414,41.95404201996044,0 "
+                "-87.65671750555913,41.95416647054043,0 "
+                "-87.65673983607117,41.95429949810849,0 "
+                "-87.65673866475777,41.95441024240925,0 "
+                "-87.6567690255541,41.95490657227902,0 "
+                "-87.65683672482363,41.95692259283837,0 "
+                "-87.6568900886376,41.95861070983142,0 "
+                "-87.65699865558875,41.96181418669004,0 "
+                "-87.65756347177603,41.96397045777844,0 "
+                "-87.65758750947528,41.96427269188822,0"
+            ),
+            3: (
+                "-87.65362593118043,41.94742799535678,0 "
+                "-87.65363554415794,41.94819886386848,0 "
+                "-87.6536456393239,41.95059994675451,0 "
+                "-87.65365831235026,41.95108288489359,0 "
+                "-87.6536604873874,41.9519954657554,0 "
+                "-87.65362592053201,41.95245597302328,0 "
+                "-87.65367158496069,41.95311153649393,0 "
+                "-87.65368468595476,41.9533202828916,0 "
+                "-87.65369271253692,41.95343095587119,0 "
+                "-87.65373335834569,41.95351536301472,0 "
+                "-87.65378605844126,41.95358212680591,0 "
+                "-87.65385067928185,41.95364452823767,0 "
+                "-87.6539390793817,41.95370263886964,0 "
+                "-87.6540786298351,41.95373403675265,0 "
+                "-87.65430648647626,41.9537535411832,0 "
+                "-87.65492939166126,41.95377494531437,0"
+            ),
+            4: (
+                "-87.65345391792157,41.94217681262115,0 "
+                "-87.65342448305786,41.94237224420864,0 "
+                "-87.65339745703922,41.94268217746244,0 "
+                "-87.65337753982941,41.94288140770284,0 "
+                "-87.65336256753105,41.94317369618263,0 "
+                "-87.65338799707138,41.94357253961736,0 "
+                "-87.65340240886648,41.94389158188269,0 "
+                "-87.65341837392448,41.94406444407721,0 "
+                "-87.65342275247338,41.94421065714904,0 "
+                "-87.65347469646018,41.94434829382345,0 "
+                "-87.65351486483024,41.94447699917548,0 "
+                "-87.65353483605053,41.9453896864472,0 "
+                "-87.65361975532807,41.94689193720703,0 "
+                "-87.65362593118043,41.94742799535678,0"
+            ),
+        },
+    },
+    index=range(5),
+)
+
+
+def test_literal_xml_raises():
+    # GH 53809
+    pytest.importorskip("lxml")
+    msg = "|".join([r".*No such file or directory", r".*Invalid argument"])
+
+    with pytest.raises((FileNotFoundError, OSError), match=msg):
+        read_xml(xml_default_nmsp)
+
+
+@pytest.fixture(params=["rb", "r"])
+def mode(request):
+    return request.param
+
+
+@pytest.fixture(params=[pytest.param("lxml", marks=td.skip_if_no("lxml")), "etree"])
+def parser(request):
+    return request.param
+
+
+def read_xml_iterparse(data, **kwargs):
+    with tm.ensure_clean() as path:
+        with open(path, "w", encoding="utf-8") as f:
+            f.write(data)
+        return read_xml(path, **kwargs)
+
+
+def read_xml_iterparse_comp(comp_path, compression_only, **kwargs):
+    with get_handle(comp_path, "r", compression=compression_only) as handles:
+        with tm.ensure_clean() as path:
+            with open(path, "w", encoding="utf-8") as f:
+                f.write(handles.handle.read())
+            return read_xml(path, **kwargs)
+
+
+# FILE / URL
+
+
+def test_parser_consistency_file(xml_books):
+    pytest.importorskip("lxml")
+    df_file_lxml = read_xml(xml_books, parser="lxml")
+    df_file_etree = read_xml(xml_books, parser="etree")
+
+    df_iter_lxml = read_xml(
+        xml_books,
+        parser="lxml",
+        iterparse={"book": ["category", "title", "year", "author", "price"]},
+    )
+    df_iter_etree = read_xml(
+        xml_books,
+        parser="etree",
+        iterparse={"book": ["category", "title", "year", "author", "price"]},
+    )
+
+    tm.assert_frame_equal(df_file_lxml, df_file_etree)
+    tm.assert_frame_equal(df_file_lxml, df_iter_lxml)
+    tm.assert_frame_equal(df_iter_lxml, df_iter_etree)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_parser_consistency_url(parser, httpserver):
+    httpserver.serve_content(content=xml_default_nmsp)
+
+    df_xpath = read_xml(StringIO(xml_default_nmsp), parser=parser)
+    df_iter = read_xml(
+        BytesIO(xml_default_nmsp.encode()),
+        parser=parser,
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    tm.assert_frame_equal(df_xpath, df_iter)
+
+
+def test_file_like(xml_books, parser, mode):
+    with open(xml_books, mode, encoding="utf-8" if mode == "r" else None) as f:
+        df_file = read_xml(f, parser=parser)
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_file, df_expected)
+
+
+def test_file_io(xml_books, parser, mode):
+    with open(xml_books, mode, encoding="utf-8" if mode == "r" else None) as f:
+        xml_obj = f.read()
+
+    df_io = read_xml(
+        (BytesIO(xml_obj) if isinstance(xml_obj, bytes) else StringIO(xml_obj)),
+        parser=parser,
+    )
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_io, df_expected)
+
+
+def test_file_buffered_reader_string(xml_books, parser, mode):
+    with open(xml_books, mode, encoding="utf-8" if mode == "r" else None) as f:
+        xml_obj = f.read()
+
+    if mode == "rb":
+        xml_obj = StringIO(xml_obj.decode())
+    elif mode == "r":
+        xml_obj = StringIO(xml_obj)
+
+    df_str = read_xml(xml_obj, parser=parser)
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_str, df_expected)
+
+
+def test_file_buffered_reader_no_xml_declaration(xml_books, parser, mode):
+    with open(xml_books, mode, encoding="utf-8" if mode == "r" else None) as f:
+        next(f)
+        xml_obj = f.read()
+
+    if mode == "rb":
+        xml_obj = StringIO(xml_obj.decode())
+    elif mode == "r":
+        xml_obj = StringIO(xml_obj)
+
+    df_str = read_xml(xml_obj, parser=parser)
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_str, df_expected)
+
+
+def test_string_charset(parser):
+    txt = "<中文標籤><row><c1>1</c1><c2>2</c2></row></中文標籤>"
+
+    df_str = read_xml(StringIO(txt), parser=parser)
+
+    df_expected = DataFrame({"c1": 1, "c2": 2}, index=range(1))
+
+    tm.assert_frame_equal(df_str, df_expected)
+
+
+def test_file_charset(xml_doc_ch_utf, parser):
+    df_file = read_xml(xml_doc_ch_utf, parser=parser)
+
+    df_expected = DataFrame(
+        {
+            "問": [
+                "問  若箇是邪而言破邪 何者是正而道(Sorry, this is Big5 only)申正",
+                "問 既破有得申無得 亦應但破性執申假名以不",
+                "問 既破性申假 亦應但破有申無 若有無兩洗 亦應性假雙破耶",
+            ],
+            "答": [
+                "".join(
+                    [
+                        "答  邪既無量 正亦多途  大略為言不出二種 謂",
+                        "有得與無得 有得是邪須破 無得是正須申\n\t\t故",
+                    ]
+                ),
+                None,
+                "答  不例  有無皆是性 所以須雙破 既分性假異 故有破不破",
+            ],
+            "a": [
+                None,
+                "答 性執是有得 假名是無得  今破有得申無得 即是破性執申假名也",
+                None,
+            ],
+        }
+    )
+
+    tm.assert_frame_equal(df_file, df_expected)
+
+
+def test_file_handle_close(xml_books, parser):
+    with open(xml_books, "rb") as f:
+        read_xml(BytesIO(f.read()), parser=parser)
+
+        assert not f.closed
+
+
+@pytest.mark.parametrize("val", ["", b""])
+def test_empty_string_lxml(val):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    msg = "|".join(
+        [
+            "Document is empty",
+            # Seen on Mac with lxml 4.91
+            r"None \(line 0\)",
+        ]
+    )
+    if isinstance(val, str):
+        data = StringIO(val)
+    else:
+        data = BytesIO(val)
+    with pytest.raises(lxml_etree.XMLSyntaxError, match=msg):
+        read_xml(data, parser="lxml")
+
+
+@pytest.mark.parametrize("val", ["", b""])
+def test_empty_string_etree(val):
+    if isinstance(val, str):
+        data = StringIO(val)
+    else:
+        data = BytesIO(val)
+    with pytest.raises(ParseError, match="no element found"):
+        read_xml(data, parser="etree")
+
+
+@pytest.mark.skipif(WASM, reason="limited file system access on WASM")
+def test_wrong_file_path(parser):
+    filename = os.path.join("does", "not", "exist", "books.xml")
+
+    with pytest.raises(
+        FileNotFoundError, match=r"\[Errno 2\] No such file or directory"
+    ):
+        read_xml(filename, parser=parser)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_url(httpserver, xml_file):
+    pytest.importorskip("lxml")
+    with open(xml_file, encoding="utf-8") as f:
+        httpserver.serve_content(content=f.read())
+        df_url = read_xml(httpserver.url, xpath=".//book[count(*)=4]")
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_url, df_expected)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_wrong_url(parser, httpserver):
+    httpserver.serve_content("NOT FOUND", code=404)
+    with pytest.raises(HTTPError, match=("HTTP Error 404: NOT FOUND")):
+        read_xml(httpserver.url, xpath=".//book[count(*)=4]", parser=parser)
+
+
+# CONTENT
+
+
+def test_whitespace(parser):
+    xml = """
+      <data>
+        <row sides=" 4 ">
+          <shape>
+              square
+          </shape>
+          <degrees>&#009;360&#009;</degrees>
+        </row>
+        <row sides=" 0 ">
+          <shape>
+              circle
+          </shape>
+          <degrees>&#009;360&#009;</degrees>
+        </row>
+        <row sides=" 3 ">
+          <shape>
+              triangle
+          </shape>
+          <degrees>&#009;180&#009;</degrees>
+        </row>
+      </data>"""
+
+    df_xpath = read_xml(StringIO(xml), parser=parser, dtype="string")
+
+    df_iter = read_xml_iterparse(
+        xml,
+        parser=parser,
+        iterparse={"row": ["sides", "shape", "degrees"]},
+        dtype="string",
+    )
+
+    df_expected = DataFrame(
+        {
+            "sides": [" 4 ", " 0 ", " 3 "],
+            "shape": [
+                "\n              square\n          ",
+                "\n              circle\n          ",
+                "\n              triangle\n          ",
+            ],
+            "degrees": ["\t360\t", "\t360\t", "\t180\t"],
+        },
+        dtype="string",
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+# XPATH
+
+
+def test_empty_xpath_lxml(xml_books):
+    pytest.importorskip("lxml")
+    with pytest.raises(ValueError, match=("xpath does not return any nodes")):
+        read_xml(xml_books, xpath=".//python", parser="lxml")
+
+
+def test_bad_xpath_etree(xml_books):
+    with pytest.raises(
+        SyntaxError, match=("You have used an incorrect or unsupported XPath")
+    ):
+        read_xml(xml_books, xpath=".//[book]", parser="etree")
+
+
+def test_bad_xpath_lxml(xml_books):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    with pytest.raises(lxml_etree.XPathEvalError, match=("Invalid expression")):
+        read_xml(xml_books, xpath=".//[book]", parser="lxml")
+
+
+# NAMESPACE
+
+
+def test_default_namespace(parser):
+    df_nmsp = read_xml(
+        StringIO(xml_default_nmsp),
+        xpath=".//ns:row",
+        namespaces={"ns": "http://example.com"},
+        parser=parser,
+    )
+
+    df_iter = read_xml_iterparse(
+        xml_default_nmsp,
+        parser=parser,
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    tm.assert_frame_equal(df_nmsp, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_prefix_namespace(parser):
+    df_nmsp = read_xml(
+        StringIO(xml_prefix_nmsp),
+        xpath=".//doc:row",
+        namespaces={"doc": "http://example.com"},
+        parser=parser,
+    )
+    df_iter = read_xml_iterparse(
+        xml_prefix_nmsp, parser=parser, iterparse={"row": ["shape", "degrees", "sides"]}
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    tm.assert_frame_equal(df_nmsp, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_consistency_default_namespace():
+    pytest.importorskip("lxml")
+    df_lxml = read_xml(
+        StringIO(xml_default_nmsp),
+        xpath=".//ns:row",
+        namespaces={"ns": "http://example.com"},
+        parser="lxml",
+    )
+
+    df_etree = read_xml(
+        StringIO(xml_default_nmsp),
+        xpath=".//doc:row",
+        namespaces={"doc": "http://example.com"},
+        parser="etree",
+    )
+
+    tm.assert_frame_equal(df_lxml, df_etree)
+
+
+def test_consistency_prefix_namespace():
+    pytest.importorskip("lxml")
+    df_lxml = read_xml(
+        StringIO(xml_prefix_nmsp),
+        xpath=".//doc:row",
+        namespaces={"doc": "http://example.com"},
+        parser="lxml",
+    )
+
+    df_etree = read_xml(
+        StringIO(xml_prefix_nmsp),
+        xpath=".//doc:row",
+        namespaces={"doc": "http://example.com"},
+        parser="etree",
+    )
+
+    tm.assert_frame_equal(df_lxml, df_etree)
+
+
+# PREFIX
+
+
+def test_missing_prefix_with_default_namespace(xml_books, parser):
+    with pytest.raises(ValueError, match=("xpath does not return any nodes")):
+        read_xml(xml_books, xpath=".//Placemark", parser=parser)
+
+
+def test_missing_prefix_definition_etree(kml_cta_rail_lines):
+    with pytest.raises(SyntaxError, match=("you used an undeclared namespace prefix")):
+        read_xml(kml_cta_rail_lines, xpath=".//kml:Placemark", parser="etree")
+
+
+def test_missing_prefix_definition_lxml(kml_cta_rail_lines):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    with pytest.raises(lxml_etree.XPathEvalError, match=("Undefined namespace prefix")):
+        read_xml(kml_cta_rail_lines, xpath=".//kml:Placemark", parser="lxml")
+
+
+@pytest.mark.parametrize("key", ["", None])
+def test_none_namespace_prefix(key):
+    pytest.importorskip("lxml")
+    with pytest.raises(
+        TypeError, match=("empty namespace prefix is not supported in XPath")
+    ):
+        read_xml(
+            StringIO(xml_default_nmsp),
+            xpath=".//kml:Placemark",
+            namespaces={key: "http://www.opengis.net/kml/2.2"},
+            parser="lxml",
+        )
+
+
+# ELEMS AND ATTRS
+
+
+def test_file_elems_and_attrs(xml_books, parser):
+    df_file = read_xml(xml_books, parser=parser)
+    df_iter = read_xml(
+        xml_books,
+        parser=parser,
+        iterparse={"book": ["category", "title", "author", "year", "price"]},
+    )
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_file, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_file_only_attrs(xml_books, parser):
+    df_file = read_xml(xml_books, attrs_only=True, parser=parser)
+    df_iter = read_xml(xml_books, parser=parser, iterparse={"book": ["category"]})
+    df_expected = DataFrame({"category": ["cooking", "children", "web"]})
+
+    tm.assert_frame_equal(df_file, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_file_only_elems(xml_books, parser):
+    df_file = read_xml(xml_books, elems_only=True, parser=parser)
+    df_iter = read_xml(
+        xml_books,
+        parser=parser,
+        iterparse={"book": ["title", "author", "year", "price"]},
+    )
+    df_expected = DataFrame(
+        {
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_file, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_elem_and_attrs_only(kml_cta_rail_lines, parser):
+    with pytest.raises(
+        ValueError,
+        match=("Either element or attributes can be parsed not both"),
+    ):
+        read_xml(kml_cta_rail_lines, elems_only=True, attrs_only=True, parser=parser)
+
+
+def test_empty_attrs_only(parser):
+    xml = """
+      <data>
+        <row>
+          <shape sides="4">square</shape>
+          <degrees>360</degrees>
+        </row>
+        <row>
+          <shape sides="0">circle</shape>
+          <degrees>360</degrees>
+        </row>
+        <row>
+          <shape sides="3">triangle</shape>
+          <degrees>180</degrees>
+        </row>
+      </data>"""
+
+    with pytest.raises(
+        ValueError,
+        match=("xpath does not return any nodes or attributes"),
+    ):
+        read_xml(StringIO(xml), xpath="./row", attrs_only=True, parser=parser)
+
+
+def test_empty_elems_only(parser):
+    xml = """
+      <data>
+        <row sides="4" shape="square" degrees="360"/>
+        <row sides="0" shape="circle" degrees="360"/>
+        <row sides="3" shape="triangle" degrees="180"/>
+      </data>"""
+
+    with pytest.raises(
+        ValueError,
+        match=("xpath does not return any nodes or attributes"),
+    ):
+        read_xml(StringIO(xml), xpath="./row", elems_only=True, parser=parser)
+
+
+def test_attribute_centric_xml():
+    pytest.importorskip("lxml")
+    xml = """\
+<?xml version="1.0" encoding="UTF-8"?>
+<TrainSchedule>
+      <Stations>
+         <station Name="Manhattan" coords="31,460,195,498"/>
+         <station Name="Laraway Road" coords="63,409,194,455"/>
+         <station Name="179th St (Orland Park)" coords="0,364,110,395"/>
+         <station Name="153rd St (Orland Park)" coords="7,333,113,362"/>
+         <station Name="143rd St (Orland Park)" coords="17,297,115,330"/>
+         <station Name="Palos Park" coords="128,281,239,303"/>
+         <station Name="Palos Heights" coords="148,257,283,279"/>
+         <station Name="Worth" coords="170,230,248,255"/>
+         <station Name="Chicago Ridge" coords="70,187,208,214"/>
+         <station Name="Oak Lawn" coords="166,159,266,185"/>
+         <station Name="Ashburn" coords="197,133,336,157"/>
+         <station Name="Wrightwood" coords="219,106,340,133"/>
+         <station Name="Chicago Union Sta" coords="220,0,360,43"/>
+      </Stations>
+</TrainSchedule>"""
+
+    df_lxml = read_xml(StringIO(xml), xpath=".//station")
+    df_etree = read_xml(StringIO(xml), xpath=".//station", parser="etree")
+
+    df_iter_lx = read_xml_iterparse(xml, iterparse={"station": ["Name", "coords"]})
+    df_iter_et = read_xml_iterparse(
+        xml, parser="etree", iterparse={"station": ["Name", "coords"]}
+    )
+
+    tm.assert_frame_equal(df_lxml, df_etree)
+    tm.assert_frame_equal(df_iter_lx, df_iter_et)
+
+
+# NAMES
+
+
+def test_names_option_output(xml_books, parser):
+    df_file = read_xml(
+        xml_books, names=["Col1", "Col2", "Col3", "Col4", "Col5"], parser=parser
+    )
+    df_iter = read_xml(
+        xml_books,
+        parser=parser,
+        names=["Col1", "Col2", "Col3", "Col4", "Col5"],
+        iterparse={"book": ["category", "title", "author", "year", "price"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "Col1": ["cooking", "children", "web"],
+            "Col2": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "Col3": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "Col4": [2005, 2005, 2003],
+            "Col5": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_file, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_repeat_names(parser):
+    xml = """\
+<shapes>
+  <shape type="2D">
+    <name>circle</name>
+    <type>curved</type>
+  </shape>
+  <shape type="3D">
+    <name>sphere</name>
+    <type>curved</type>
+  </shape>
+</shapes>"""
+    df_xpath = read_xml(
+        StringIO(xml),
+        xpath=".//shape",
+        parser=parser,
+        names=["type_dim", "shape", "type_edge"],
+    )
+
+    df_iter = read_xml_iterparse(
+        xml,
+        parser=parser,
+        iterparse={"shape": ["type", "name", "type"]},
+        names=["type_dim", "shape", "type_edge"],
+    )
+
+    df_expected = DataFrame(
+        {
+            "type_dim": ["2D", "3D"],
+            "shape": ["circle", "sphere"],
+            "type_edge": ["curved", "curved"],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_repeat_values_new_names(parser):
+    xml = """\
+<shapes>
+  <shape>
+    <name>rectangle</name>
+    <family>rectangle</family>
+  </shape>
+  <shape>
+    <name>square</name>
+    <family>rectangle</family>
+  </shape>
+  <shape>
+    <name>ellipse</name>
+    <family>ellipse</family>
+  </shape>
+  <shape>
+    <name>circle</name>
+    <family>ellipse</family>
+  </shape>
+</shapes>"""
+    df_xpath = read_xml(
+        StringIO(xml), xpath=".//shape", parser=parser, names=["name", "group"]
+    )
+
+    df_iter = read_xml_iterparse(
+        xml,
+        parser=parser,
+        iterparse={"shape": ["name", "family"]},
+        names=["name", "group"],
+    )
+
+    df_expected = DataFrame(
+        {
+            "name": ["rectangle", "square", "ellipse", "circle"],
+            "group": ["rectangle", "rectangle", "ellipse", "ellipse"],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_repeat_elements(parser):
+    xml = """\
+<shapes>
+  <shape>
+    <value item="name">circle</value>
+    <value item="family">ellipse</value>
+    <value item="degrees">360</value>
+    <value item="sides">0</value>
+  </shape>
+  <shape>
+    <value item="name">triangle</value>
+    <value item="family">polygon</value>
+    <value item="degrees">180</value>
+    <value item="sides">3</value>
+  </shape>
+  <shape>
+    <value item="name">square</value>
+    <value item="family">polygon</value>
+    <value item="degrees">360</value>
+    <value item="sides">4</value>
+  </shape>
+</shapes>"""
+    df_xpath = read_xml(
+        StringIO(xml),
+        xpath=".//shape",
+        parser=parser,
+        names=["name", "family", "degrees", "sides"],
+    )
+
+    df_iter = read_xml_iterparse(
+        xml,
+        parser=parser,
+        iterparse={"shape": ["value", "value", "value", "value"]},
+        names=["name", "family", "degrees", "sides"],
+    )
+
+    df_expected = DataFrame(
+        {
+            "name": ["circle", "triangle", "square"],
+            "family": ["ellipse", "polygon", "polygon"],
+            "degrees": [360, 180, 360],
+            "sides": [0, 3, 4],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_names_option_wrong_length(xml_books, parser):
+    with pytest.raises(ValueError, match=("names does not match length")):
+        read_xml(xml_books, names=["Col1", "Col2", "Col3"], parser=parser)
+
+
+def test_names_option_wrong_type(xml_books, parser):
+    with pytest.raises(TypeError, match=("is not a valid type for names")):
+        read_xml(xml_books, names="Col1, Col2, Col3", parser=parser)
+
+
+# ENCODING
+
+
+def test_wrong_encoding(xml_baby_names, parser):
+    with pytest.raises(UnicodeDecodeError, match=("'utf-8' codec can't decode")):
+        read_xml(xml_baby_names, parser=parser)
+
+
+def test_utf16_encoding(xml_baby_names, parser):
+    with pytest.raises(
+        UnicodeError,
+        match=(
+            "UTF-16 stream does not start with BOM|"
+            "'utf-16(-le)?' codec can't decode byte"
+        ),
+    ):
+        read_xml(xml_baby_names, encoding="UTF-16", parser=parser)
+
+
+def test_unknown_encoding(xml_baby_names, parser):
+    with pytest.raises(LookupError, match=("unknown encoding: UFT-8")):
+        read_xml(xml_baby_names, encoding="UFT-8", parser=parser)
+
+
+def test_ascii_encoding(xml_baby_names, parser):
+    with pytest.raises(UnicodeDecodeError, match=("'ascii' codec can't decode byte")):
+        read_xml(xml_baby_names, encoding="ascii", parser=parser)
+
+
+def test_parser_consistency_with_encoding(xml_baby_names):
+    pytest.importorskip("lxml")
+    df_xpath_lxml = read_xml(xml_baby_names, parser="lxml", encoding="ISO-8859-1")
+    df_xpath_etree = read_xml(xml_baby_names, parser="etree", encoding="iso-8859-1")
+
+    df_iter_lxml = read_xml(
+        xml_baby_names,
+        parser="lxml",
+        encoding="ISO-8859-1",
+        iterparse={"row": ["rank", "malename", "femalename"]},
+    )
+    df_iter_etree = read_xml(
+        xml_baby_names,
+        parser="etree",
+        encoding="ISO-8859-1",
+        iterparse={"row": ["rank", "malename", "femalename"]},
+    )
+
+    tm.assert_frame_equal(df_xpath_lxml, df_xpath_etree)
+    tm.assert_frame_equal(df_xpath_etree, df_iter_etree)
+    tm.assert_frame_equal(df_iter_lxml, df_iter_etree)
+
+
+def test_wrong_encoding_for_lxml():
+    pytest.importorskip("lxml")
+    # GH#45133
+    data = """<data>
+  <row>
+    <a>c</a>
+  </row>
+</data>
+"""
+    with pytest.raises(TypeError, match="encoding None"):
+        read_xml(StringIO(data), parser="lxml", encoding=None)
+
+
+def test_none_encoding_etree():
+    # GH#45133
+    data = """<data>
+  <row>
+    <a>c</a>
+  </row>
+</data>
+"""
+    result = read_xml(StringIO(data), parser="etree", encoding=None)
+    expected = DataFrame({"a": ["c"]})
+    tm.assert_frame_equal(result, expected)
+
+
+# PARSER
+
+
+@td.skip_if_installed("lxml")
+def test_default_parser_no_lxml(xml_books):
+    with pytest.raises(
+        ImportError, match=("lxml not found, please install or use the etree parser.")
+    ):
+        read_xml(xml_books)
+
+
+def test_wrong_parser(xml_books):
+    with pytest.raises(
+        ValueError, match=("Values for parser can only be lxml or etree.")
+    ):
+        read_xml(xml_books, parser="bs4")
+
+
+# STYLESHEET
+
+
+def test_stylesheet_file(kml_cta_rail_lines, xsl_flatten_doc):
+    pytest.importorskip("lxml")
+    df_style = read_xml(
+        kml_cta_rail_lines,
+        xpath=".//k:Placemark",
+        namespaces={"k": "http://www.opengis.net/kml/2.2"},
+        stylesheet=xsl_flatten_doc,
+    )
+
+    df_iter = read_xml(
+        kml_cta_rail_lines,
+        iterparse={
+            "Placemark": [
+                "id",
+                "name",
+                "styleUrl",
+                "extrude",
+                "altitudeMode",
+                "coordinates",
+            ]
+        },
+    )
+
+    tm.assert_frame_equal(df_kml, df_style)
+    tm.assert_frame_equal(df_kml, df_iter)
+
+
+def test_stylesheet_file_like(kml_cta_rail_lines, xsl_flatten_doc, mode):
+    pytest.importorskip("lxml")
+    with open(xsl_flatten_doc, mode, encoding="utf-8" if mode == "r" else None) as f:
+        df_style = read_xml(
+            kml_cta_rail_lines,
+            xpath=".//k:Placemark",
+            namespaces={"k": "http://www.opengis.net/kml/2.2"},
+            stylesheet=f,
+        )
+
+    tm.assert_frame_equal(df_kml, df_style)
+
+
+def test_stylesheet_io(kml_cta_rail_lines, xsl_flatten_doc, mode):
+    # note: By default the bodies of untyped functions are not checked,
+    # consider using --check-untyped-defs
+    pytest.importorskip("lxml")
+    xsl_obj: BytesIO | StringIO  # type: ignore[annotation-unchecked]
+
+    with open(xsl_flatten_doc, mode, encoding="utf-8" if mode == "r" else None) as f:
+        if mode == "rb":
+            xsl_obj = BytesIO(f.read())
+        else:
+            xsl_obj = StringIO(f.read())
+
+    df_style = read_xml(
+        kml_cta_rail_lines,
+        xpath=".//k:Placemark",
+        namespaces={"k": "http://www.opengis.net/kml/2.2"},
+        stylesheet=xsl_obj,
+    )
+
+    tm.assert_frame_equal(df_kml, df_style)
+
+
+def test_stylesheet_buffered_reader(kml_cta_rail_lines, xsl_flatten_doc, mode):
+    pytest.importorskip("lxml")
+    with open(xsl_flatten_doc, mode, encoding="utf-8" if mode == "r" else None) as f:
+        df_style = read_xml(
+            kml_cta_rail_lines,
+            xpath=".//k:Placemark",
+            namespaces={"k": "http://www.opengis.net/kml/2.2"},
+            stylesheet=f,
+        )
+
+    tm.assert_frame_equal(df_kml, df_style)
+
+
+def test_style_charset():
+    pytest.importorskip("lxml")
+    xml = "<中文標籤><row><c1>1</c1><c2>2</c2></row></中文標籤>"
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+ <xsl:output omit-xml-declaration="yes" indent="yes"/>
+ <xsl:strip-space elements="*"/>
+
+ <xsl:template match="node()|@*">
+     <xsl:copy>
+       <xsl:apply-templates select="node()|@*"/>
+     </xsl:copy>
+ </xsl:template>
+
+ <xsl:template match="中文標籤">
+     <根>
+       <xsl:apply-templates />
+     </根>
+ </xsl:template>
+
+</xsl:stylesheet>"""
+
+    df_orig = read_xml(StringIO(xml))
+    df_style = read_xml(StringIO(xml), stylesheet=StringIO(xsl))
+
+    tm.assert_frame_equal(df_orig, df_style)
+
+
+def test_not_stylesheet(kml_cta_rail_lines, xml_books):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    with pytest.raises(
+        lxml_etree.XSLTParseError, match=("document is not a stylesheet")
+    ):
+        read_xml(kml_cta_rail_lines, stylesheet=xml_books)
+
+
+def test_incorrect_xsl_syntax(kml_cta_rail_lines):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                              xmlns:k="http://www.opengis.net/kml/2.2"/>
+    <xsl:output method="xml" omit-xml-declaration="yes"
+                cdata-section-elements="k:description" indent="yes"/>
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="node()|@*">
+     <xsl:copy>
+       <xsl:apply-templates select="node()|@*"/>
+     </xsl:copy>
+    </xsl:template>
+
+    <xsl:template match="k:MultiGeometry|k:LineString">
+        <xsl:apply-templates select='*'/>
+    </xsl:template>
+
+    <xsl:template match="k:description|k:Snippet|k:Style"/>
+</xsl:stylesheet>"""
+
+    with pytest.raises(
+        lxml_etree.XMLSyntaxError, match="Extra content at the end of the document"
+    ):
+        read_xml(kml_cta_rail_lines, stylesheet=StringIO(xsl))
+
+
+def test_incorrect_xsl_eval(kml_cta_rail_lines):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                              xmlns:k="http://www.opengis.net/kml/2.2">
+    <xsl:output method="xml" omit-xml-declaration="yes"
+                cdata-section-elements="k:description" indent="yes"/>
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="node(*)|@*">
+     <xsl:copy>
+       <xsl:apply-templates select="node()|@*"/>
+     </xsl:copy>
+    </xsl:template>
+
+    <xsl:template match="k:MultiGeometry|k:LineString">
+        <xsl:apply-templates select='*'/>
+    </xsl:template>
+
+    <xsl:template match="k:description|k:Snippet|k:Style"/>
+</xsl:stylesheet>"""
+
+    with pytest.raises(lxml_etree.XSLTParseError, match="failed to compile"):
+        read_xml(kml_cta_rail_lines, stylesheet=StringIO(xsl))
+
+
+def test_incorrect_xsl_apply(kml_cta_rail_lines):
+    lxml_etree = pytest.importorskip("lxml.etree")
+
+    xsl = """\
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+    <xsl:output method="xml" encoding="utf-8" indent="yes" />
+    <xsl:strip-space elements="*"/>
+
+    <xsl:template match="@*|node()">
+        <xsl:copy>
+            <xsl:copy-of select="document('non_existent.xml')/*"/>
+        </xsl:copy>
+    </xsl:template>
+</xsl:stylesheet>"""
+
+    with pytest.raises(lxml_etree.XSLTApplyError, match="Cannot resolve URI"):
+        read_xml(kml_cta_rail_lines, stylesheet=StringIO(xsl))
+
+
+def test_wrong_stylesheet(kml_cta_rail_lines, xml_data_path):
+    pytest.importorskip("lxml.etree")
+
+    xsl = xml_data_path / "flatten_doesnt_exist.xsl"
+
+    with pytest.raises(
+        FileNotFoundError, match=r"\[Errno 2\] No such file or directory"
+    ):
+        read_xml(kml_cta_rail_lines, stylesheet=xsl)
+
+
+def test_stylesheet_file_close(kml_cta_rail_lines, xsl_flatten_doc, mode):
+    # note: By default the bodies of untyped functions are not checked,
+    # consider using --check-untyped-defs
+    pytest.importorskip("lxml")
+    xsl_obj: BytesIO | StringIO  # type: ignore[annotation-unchecked]
+
+    with open(xsl_flatten_doc, mode, encoding="utf-8" if mode == "r" else None) as f:
+        if mode == "rb":
+            xsl_obj = BytesIO(f.read())
+        else:
+            xsl_obj = StringIO(f.read())
+
+        read_xml(kml_cta_rail_lines, stylesheet=xsl_obj)
+
+        assert not f.closed
+
+
+def test_stylesheet_with_etree(kml_cta_rail_lines, xsl_flatten_doc):
+    pytest.importorskip("lxml")
+    with pytest.raises(
+        ValueError, match=("To use stylesheet, you need lxml installed")
+    ):
+        read_xml(kml_cta_rail_lines, parser="etree", stylesheet=xsl_flatten_doc)
+
+
+@pytest.mark.parametrize("val", [StringIO(""), BytesIO(b"")])
+def test_empty_stylesheet(val, kml_cta_rail_lines):
+    lxml_etree = pytest.importorskip("lxml.etree")
+    with pytest.raises(lxml_etree.XMLSyntaxError):
+        read_xml(kml_cta_rail_lines, stylesheet=val)
+
+
+# ITERPARSE
+def test_file_like_iterparse(xml_books, parser, mode):
+    with open(xml_books, mode, encoding="utf-8" if mode == "r" else None) as f:
+        if mode == "r" and parser == "lxml":
+            with pytest.raises(
+                TypeError, match=("reading file objects must return bytes objects")
+            ):
+                read_xml(
+                    f,
+                    parser=parser,
+                    iterparse={
+                        "book": ["category", "title", "year", "author", "price"]
+                    },
+                )
+            return None
+        else:
+            df_filelike = read_xml(
+                f,
+                parser=parser,
+                iterparse={"book": ["category", "title", "year", "author", "price"]},
+            )
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_filelike, df_expected)
+
+
+def test_file_io_iterparse(xml_books, parser, mode):
+    funcIO = StringIO if mode == "r" else BytesIO
+    with open(
+        xml_books,
+        mode,
+        encoding="utf-8" if mode == "r" else None,
+    ) as f:
+        with funcIO(f.read()) as b:
+            if mode == "r" and parser == "lxml":
+                with pytest.raises(
+                    TypeError, match=("reading file objects must return bytes objects")
+                ):
+                    read_xml(
+                        b,
+                        parser=parser,
+                        iterparse={
+                            "book": ["category", "title", "year", "author", "price"]
+                        },
+                    )
+                return None
+            else:
+                df_fileio = read_xml(
+                    b,
+                    parser=parser,
+                    iterparse={
+                        "book": ["category", "title", "year", "author", "price"]
+                    },
+                )
+
+    df_expected = DataFrame(
+        {
+            "category": ["cooking", "children", "web"],
+            "title": ["Everyday Italian", "Harry Potter", "Learning XML"],
+            "author": ["Giada De Laurentiis", "J K. Rowling", "Erik T. Ray"],
+            "year": [2005, 2005, 2003],
+            "price": [30.00, 29.99, 39.95],
+        }
+    )
+
+    tm.assert_frame_equal(df_fileio, df_expected)
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_url_path_error(parser, httpserver, xml_file):
+    with open(xml_file, encoding="utf-8") as f:
+        httpserver.serve_content(content=f.read())
+        with pytest.raises(
+            ParserError, match=("iterparse is designed for large XML files")
+        ):
+            read_xml(
+                httpserver.url,
+                parser=parser,
+                iterparse={"row": ["shape", "degrees", "sides", "date"]},
+            )
+
+
+def test_compression_error(parser, compression_only):
+    with tm.ensure_clean(filename="geom_xml.zip") as path:
+        geom_df.to_xml(path, parser=parser, compression=compression_only)
+
+        with pytest.raises(
+            ParserError, match=("iterparse is designed for large XML files")
+        ):
+            read_xml(
+                path,
+                parser=parser,
+                iterparse={"row": ["shape", "degrees", "sides", "date"]},
+                compression=compression_only,
+            )
+
+
+def test_wrong_dict_type(xml_books, parser):
+    with pytest.raises(TypeError, match="list is not a valid type for iterparse"):
+        read_xml(
+            xml_books,
+            parser=parser,
+            iterparse=["category", "title", "year", "author", "price"],
+        )
+
+
+def test_wrong_dict_value(xml_books, parser):
+    with pytest.raises(
+        TypeError, match="<class 'str'> is not a valid type for value in iterparse"
+    ):
+        read_xml(xml_books, parser=parser, iterparse={"book": "category"})
+
+
+def test_bad_xml(parser):
+    bad_xml = """\
+<?xml version='1.0' encoding='utf-8'?>
+  <row>
+    <shape>square</shape>
+    <degrees>00360</degrees>
+    <sides>4.0</sides>
+    <date>2020-01-01</date>
+   </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>00360</degrees>
+    <sides/>
+    <date>2021-01-01</date>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>00180</degrees>
+    <sides>3.0</sides>
+    <date>2022-01-01</date>
+  </row>
+"""
+    with tm.ensure_clean(filename="bad.xml") as path:
+        with open(path, "w", encoding="utf-8") as f:
+            f.write(bad_xml)
+
+        with pytest.raises(
+            SyntaxError,
+            match=(
+                "Extra content at the end of the document|junk after document element"
+            ),
+        ):
+            read_xml(
+                path,
+                parser=parser,
+                parse_dates=["date"],
+                iterparse={"row": ["shape", "degrees", "sides", "date"]},
+            )
+
+
+def test_comment(parser):
+    xml = """\
+<!-- comment before root -->
+<shapes>
+  <!-- comment within root -->
+  <shape>
+    <name>circle</name>
+    <type>2D</type>
+  </shape>
+  <shape>
+    <name>sphere</name>
+    <type>3D</type>
+    <!-- comment within child -->
+  </shape>
+  <!-- comment within root -->
+</shapes>
+<!-- comment after root -->"""
+
+    df_xpath = read_xml(StringIO(xml), xpath=".//shape", parser=parser)
+
+    df_iter = read_xml_iterparse(
+        xml, parser=parser, iterparse={"shape": ["name", "type"]}
+    )
+
+    df_expected = DataFrame(
+        {
+            "name": ["circle", "sphere"],
+            "type": ["2D", "3D"],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_dtd(parser):
+    xml = """\
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE non-profits [
+    <!ELEMENT shapes (shape*) >
+    <!ELEMENT shape ( name, type )>
+    <!ELEMENT name (#PCDATA)>
+]>
+<shapes>
+  <shape>
+    <name>circle</name>
+    <type>2D</type>
+  </shape>
+  <shape>
+    <name>sphere</name>
+    <type>3D</type>
+  </shape>
+</shapes>"""
+
+    df_xpath = read_xml(StringIO(xml), xpath=".//shape", parser=parser)
+
+    df_iter = read_xml_iterparse(
+        xml, parser=parser, iterparse={"shape": ["name", "type"]}
+    )
+
+    df_expected = DataFrame(
+        {
+            "name": ["circle", "sphere"],
+            "type": ["2D", "3D"],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_processing_instruction(parser):
+    xml = """\
+<?xml version="1.0" encoding="UTF-8"?>
+<?xml-stylesheet type="text/xsl" href="style.xsl"?>
+<?display table-view?>
+<?sort alpha-ascending?>
+<?textinfo whitespace is allowed ?>
+<?elementnames <shape>, <name>, <type> ?>
+<shapes>
+  <shape>
+    <name>circle</name>
+    <type>2D</type>
+  </shape>
+  <shape>
+    <name>sphere</name>
+    <type>3D</type>
+  </shape>
+</shapes>"""
+
+    df_xpath = read_xml(StringIO(xml), xpath=".//shape", parser=parser)
+
+    df_iter = read_xml_iterparse(
+        xml, parser=parser, iterparse={"shape": ["name", "type"]}
+    )
+
+    df_expected = DataFrame(
+        {
+            "name": ["circle", "sphere"],
+            "type": ["2D", "3D"],
+        }
+    )
+
+    tm.assert_frame_equal(df_xpath, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_no_result(xml_books, parser):
+    with pytest.raises(
+        ParserError, match="No result from selected items in iterparse."
+    ):
+        read_xml(
+            xml_books,
+            parser=parser,
+            iterparse={"node": ["attr1", "elem1", "elem2", "elem3"]},
+        )
+
+
+def test_empty_data(xml_books, parser):
+    with pytest.raises(EmptyDataError, match="No columns to parse from file"):
+        read_xml(
+            xml_books,
+            parser=parser,
+            iterparse={"book": ["attr1", "elem1", "elem2", "elem3"]},
+        )
+
+
+def test_online_stylesheet():
+    pytest.importorskip("lxml")
+    xml = """\
+<?xml version="1.0" encoding="UTF-8"?>
+<catalog>
+  <cd>
+    <title>Empire Burlesque</title>
+    <artist>Bob Dylan</artist>
+    <country>USA</country>
+    <company>Columbia</company>
+    <price>10.90</price>
+    <year>1985</year>
+  </cd>
+  <cd>
+    <title>Hide your heart</title>
+    <artist>Bonnie Tyler</artist>
+    <country>UK</country>
+    <company>CBS Records</company>
+    <price>9.90</price>
+    <year>1988</year>
+  </cd>
+  <cd>
+    <title>Greatest Hits</title>
+    <artist>Dolly Parton</artist>
+    <country>USA</country>
+    <company>RCA</company>
+    <price>9.90</price>
+    <year>1982</year>
+  </cd>
+  <cd>
+    <title>Still got the blues</title>
+    <artist>Gary Moore</artist>
+    <country>UK</country>
+    <company>Virgin records</company>
+    <price>10.20</price>
+    <year>1990</year>
+  </cd>
+  <cd>
+    <title>Eros</title>
+    <artist>Eros Ramazzotti</artist>
+    <country>EU</country>
+    <company>BMG</company>
+    <price>9.90</price>
+    <year>1997</year>
+  </cd>
+  <cd>
+    <title>One night only</title>
+    <artist>Bee Gees</artist>
+    <country>UK</country>
+    <company>Polydor</company>
+    <price>10.90</price>
+    <year>1998</year>
+  </cd>
+  <cd>
+    <title>Sylvias Mother</title>
+    <artist>Dr.Hook</artist>
+    <country>UK</country>
+    <company>CBS</company>
+    <price>8.10</price>
+    <year>1973</year>
+  </cd>
+  <cd>
+    <title>Maggie May</title>
+    <artist>Rod Stewart</artist>
+    <country>UK</country>
+    <company>Pickwick</company>
+    <price>8.50</price>
+    <year>1990</year>
+  </cd>
+  <cd>
+    <title>Romanza</title>
+    <artist>Andrea Bocelli</artist>
+    <country>EU</country>
+    <company>Polydor</company>
+    <price>10.80</price>
+    <year>1996</year>
+  </cd>
+  <cd>
+    <title>When a man loves a woman</title>
+    <artist>Percy Sledge</artist>
+    <country>USA</country>
+    <company>Atlantic</company>
+    <price>8.70</price>
+    <year>1987</year>
+  </cd>
+  <cd>
+    <title>Black angel</title>
+    <artist>Savage Rose</artist>
+    <country>EU</country>
+    <company>Mega</company>
+    <price>10.90</price>
+    <year>1995</year>
+  </cd>
+  <cd>
+    <title>1999 Grammy Nominees</title>
+    <artist>Many</artist>
+    <country>USA</country>
+    <company>Grammy</company>
+    <price>10.20</price>
+    <year>1999</year>
+  </cd>
+  <cd>
+    <title>For the good times</title>
+    <artist>Kenny Rogers</artist>
+    <country>UK</country>
+    <company>Mucik Master</company>
+    <price>8.70</price>
+    <year>1995</year>
+  </cd>
+  <cd>
+    <title>Big Willie style</title>
+    <artist>Will Smith</artist>
+    <country>USA</country>
+    <company>Columbia</company>
+    <price>9.90</price>
+    <year>1997</year>
+  </cd>
+  <cd>
+    <title>Tupelo Honey</title>
+    <artist>Van Morrison</artist>
+    <country>UK</country>
+    <company>Polydor</company>
+    <price>8.20</price>
+    <year>1971</year>
+  </cd>
+  <cd>
+    <title>Soulsville</title>
+    <artist>Jorn Hoel</artist>
+    <country>Norway</country>
+    <company>WEA</company>
+    <price>7.90</price>
+    <year>1996</year>
+  </cd>
+  <cd>
+    <title>The very best of</title>
+    <artist>Cat Stevens</artist>
+    <country>UK</country>
+    <company>Island</company>
+    <price>8.90</price>
+    <year>1990</year>
+  </cd>
+  <cd>
+    <title>Stop</title>
+    <artist>Sam Brown</artist>
+    <country>UK</country>
+    <company>A and M</company>
+    <price>8.90</price>
+    <year>1988</year>
+  </cd>
+  <cd>
+    <title>Bridge of Spies</title>
+    <artist>T`Pau</artist>
+    <country>UK</country>
+    <company>Siren</company>
+    <price>7.90</price>
+    <year>1987</year>
+  </cd>
+  <cd>
+    <title>Private Dancer</title>
+    <artist>Tina Turner</artist>
+    <country>UK</country>
+    <company>Capitol</company>
+    <price>8.90</price>
+    <year>1983</year>
+  </cd>
+  <cd>
+    <title>Midt om natten</title>
+    <artist>Kim Larsen</artist>
+    <country>EU</country>
+    <company>Medley</company>
+    <price>7.80</price>
+    <year>1983</year>
+  </cd>
+  <cd>
+    <title>Pavarotti Gala Concert</title>
+    <artist>Luciano Pavarotti</artist>
+    <country>UK</country>
+    <company>DECCA</company>
+    <price>9.90</price>
+    <year>1991</year>
+  </cd>
+  <cd>
+    <title>The dock of the bay</title>
+    <artist>Otis Redding</artist>
+    <country>USA</country>
+    <COMPANY>Stax Records</COMPANY>
+    <PRICE>7.90</PRICE>
+    <YEAR>1968</YEAR>
+  </cd>
+  <cd>
+    <title>Picture book</title>
+    <artist>Simply Red</artist>
+    <country>EU</country>
+    <company>Elektra</company>
+    <price>7.20</price>
+    <year>1985</year>
+  </cd>
+  <cd>
+    <title>Red</title>
+    <artist>The Communards</artist>
+    <country>UK</country>
+    <company>London</company>
+    <price>7.80</price>
+    <year>1987</year>
+  </cd>
+  <cd>
+    <title>Unchain my heart</title>
+    <artist>Joe Cocker</artist>
+    <country>USA</country>
+    <company>EMI</company>
+    <price>8.20</price>
+    <year>1987</year>
+  </cd>
+</catalog>
+"""
+    xsl = """\
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+<xsl:template match="/">
+<html>
+<body>
+  <h2>My CD Collection</h2>
+  <table border="1">
+    <tr bgcolor="#9acd32">
+      <th style="text-align:left">Title</th>
+      <th style="text-align:left">Artist</th>
+    </tr>
+    <xsl:for-each select="catalog/cd">
+    <tr>
+      <td><xsl:value-of select="title"/></td>
+      <td><xsl:value-of select="artist"/></td>
+    </tr>
+    </xsl:for-each>
+  </table>
+</body>
+</html>
+</xsl:template>
+</xsl:stylesheet>
+"""
+
+    df_xsl = read_xml(
+        StringIO(xml),
+        xpath=".//tr[td and position() <= 6]",
+        names=["title", "artist"],
+        stylesheet=StringIO(xsl),
+    )
+
+    df_expected = DataFrame(
+        {
+            "title": {
+                0: "Empire Burlesque",
+                1: "Hide your heart",
+                2: "Greatest Hits",
+                3: "Still got the blues",
+                4: "Eros",
+            },
+            "artist": {
+                0: "Bob Dylan",
+                1: "Bonnie Tyler",
+                2: "Dolly Parton",
+                3: "Gary Moore",
+                4: "Eros Ramazzotti",
+            },
+        }
+    )
+
+    tm.assert_frame_equal(df_expected, df_xsl)
+
+
+# COMPRESSION
+
+
+def test_compression_read(parser, compression_only):
+    with tm.ensure_clean() as comp_path:
+        geom_df.to_xml(
+            comp_path, index=False, parser=parser, compression=compression_only
+        )
+
+        df_xpath = read_xml(comp_path, parser=parser, compression=compression_only)
+
+        df_iter = read_xml_iterparse_comp(
+            comp_path,
+            compression_only,
+            parser=parser,
+            iterparse={"row": ["shape", "degrees", "sides"]},
+            compression=compression_only,
+        )
+
+    tm.assert_frame_equal(df_xpath, geom_df)
+    tm.assert_frame_equal(df_iter, geom_df)
+
+
+def test_wrong_compression(parser, compression, compression_only):
+    actual_compression = compression
+    attempted_compression = compression_only
+
+    if actual_compression == attempted_compression:
+        pytest.skip(f"{actual_compression} == {attempted_compression}")
+
+    errors = {
+        "bz2": (OSError, "Invalid data stream"),
+        "gzip": (OSError, "Not a gzipped file"),
+        "zip": (BadZipFile, "File is not a zip file"),
+        "tar": (ReadError, "file could not be opened successfully"),
+    }
+    zstd = import_optional_dependency("zstandard", errors="ignore")
+    if zstd is not None:
+        errors["zstd"] = (zstd.ZstdError, "Unknown frame descriptor")
+    lzma = import_optional_dependency("lzma", errors="ignore")
+    if lzma is not None:
+        errors["xz"] = (LZMAError, "Input format not supported by decoder")
+    error_cls, error_str = errors[attempted_compression]
+
+    with tm.ensure_clean() as path:
+        geom_df.to_xml(path, parser=parser, compression=actual_compression)
+
+        with pytest.raises(error_cls, match=error_str):
+            read_xml(path, parser=parser, compression=attempted_compression)
+
+
+def test_unsuported_compression(parser):
+    with pytest.raises(ValueError, match="Unrecognized compression type"):
+        with tm.ensure_clean() as path:
+            read_xml(path, parser=parser, compression="7z")
+
+
+# STORAGE OPTIONS
+
+
+@pytest.mark.network
+@pytest.mark.single_cpu
+def test_s3_parser_consistency(s3_public_bucket_with_data, s3so):
+    pytest.importorskip("s3fs")
+    pytest.importorskip("lxml")
+    s3 = f"s3://{s3_public_bucket_with_data.name}/books.xml"
+
+    df_lxml = read_xml(s3, parser="lxml", storage_options=s3so)
+
+    df_etree = read_xml(s3, parser="etree", storage_options=s3so)
+
+    tm.assert_frame_equal(df_lxml, df_etree)
+
+
+def test_read_xml_nullable_dtypes(
+    parser, string_storage, dtype_backend, using_infer_string
+):
+    # GH#50500
+    data = """<?xml version='1.0' encoding='utf-8'?>
+<data xmlns="http://example.com">
+<row>
+  <a>x</a>
+  <b>1</b>
+  <c>4.0</c>
+  <d>x</d>
+  <e>2</e>
+  <f>4.0</f>
+  <g></g>
+  <h>True</h>
+  <i>False</i>
+</row>
+<row>
+  <a>y</a>
+  <b>2</b>
+  <c>5.0</c>
+  <d></d>
+  <e></e>
+  <f></f>
+  <g></g>
+  <h>False</h>
+  <i></i>
+</row>
+</data>"""
+
+    with pd.option_context("mode.string_storage", string_storage):
+        result = read_xml(StringIO(data), parser=parser, dtype_backend=dtype_backend)
+
+    if dtype_backend == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+        string_dtype = pd.ArrowDtype(pa.string())
+    else:
+        string_dtype = pd.StringDtype(string_storage)
+
+    expected = DataFrame(
+        {
+            "a": Series(["x", "y"], dtype=string_dtype),
+            "b": Series([1, 2], dtype="Int64"),
+            "c": Series([4.0, 5.0], dtype="Float64"),
+            "d": Series(["x", None], dtype=string_dtype),
+            "e": Series([2, NA], dtype="Int64"),
+            "f": Series([4.0, NA], dtype="Float64"),
+            "g": Series([NA, NA], dtype="Int64"),
+            "h": Series([True, False], dtype="boolean"),
+            "i": Series([False, NA], dtype="boolean"),
+        }
+    )
+
+    if dtype_backend == "pyarrow":
+        pa = pytest.importorskip("pyarrow")
+        from pandas.arrays import ArrowExtensionArray
+
+        expected = DataFrame(
+            {
+                col: ArrowExtensionArray(pa.array(expected[col], from_pandas=True))
+                for col in expected.columns
+            }
+        )
+        expected["g"] = ArrowExtensionArray(pa.array([None, None]))
+
+    # the storage of the str columns' Index is also affected by the
+    # string_storage setting -> ignore that for checking the result
+    tm.assert_frame_equal(result, expected, check_column_type=False)
+
+
+def test_invalid_dtype_backend():
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        read_xml("test", dtype_backend="numpy")
diff --git a/pandas/pandas/tests/io/xml/test_xml_dtypes.py b/pandas/pandas/tests/io/xml/test_xml_dtypes.py
new file mode 100644
index 0000000000000..96ef50f9d7149
--- /dev/null
+++ b/pandas/pandas/tests/io/xml/test_xml_dtypes.py
@@ -0,0 +1,431 @@
+from __future__ import annotations
+
+from io import StringIO
+
+import pytest
+
+from pandas.errors import ParserWarning
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    to_datetime,
+)
+import pandas._testing as tm
+
+from pandas.io.xml import read_xml
+
+
+@pytest.fixture(params=[pytest.param("lxml", marks=td.skip_if_no("lxml")), "etree"])
+def parser(request):
+    return request.param
+
+
+@pytest.fixture(
+    params=[None, {"book": ["category", "title", "author", "year", "price"]}]
+)
+def iterparse(request):
+    return request.param
+
+
+def read_xml_iterparse(data, **kwargs):
+    with tm.ensure_clean() as path:
+        with open(path, "w", encoding="utf-8") as f:
+            f.write(data)
+        return read_xml(path, **kwargs)
+
+
+xml_types = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <shape>square</shape>
+    <degrees>00360</degrees>
+    <sides>4.0</sides>
+   </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>00360</degrees>
+    <sides/>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>00180</degrees>
+    <sides>3.0</sides>
+  </row>
+</data>"""
+
+xml_dates = """<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <shape>square</shape>
+    <degrees>00360</degrees>
+    <sides>4.0</sides>
+    <date>2020-01-01</date>
+   </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>00360</degrees>
+    <sides/>
+    <date>2021-01-01</date>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>00180</degrees>
+    <sides>3.0</sides>
+    <date>2022-01-01</date>
+  </row>
+</data>"""
+
+
+# DTYPE
+
+
+def test_dtype_single_str(parser):
+    df_result = read_xml(StringIO(xml_types), dtype={"degrees": "str"}, parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_types,
+        parser=parser,
+        dtype={"degrees": "str"},
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": ["00360", "00360", "00180"],
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_dtypes_all_str(parser):
+    df_result = read_xml(StringIO(xml_dates), dtype="string", parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        dtype="string",
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": ["00360", "00360", "00180"],
+            "sides": ["4.0", None, "3.0"],
+            "date": ["2020-01-01", "2021-01-01", "2022-01-01"],
+        },
+        dtype="string",
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_dtypes_with_names(parser):
+    df_result = read_xml(
+        StringIO(xml_dates),
+        names=["Col1", "Col2", "Col3", "Col4"],
+        dtype={"Col2": "string", "Col3": "Int64", "Col4": "datetime64[ns]"},
+        parser=parser,
+    )
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        names=["Col1", "Col2", "Col3", "Col4"],
+        dtype={"Col2": "string", "Col3": "Int64", "Col4": "datetime64[ns]"},
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "Col1": ["square", "circle", "triangle"],
+            "Col2": Series(["00360", "00360", "00180"]).astype("string"),
+            "Col3": Series([4.0, float("nan"), 3.0]).astype("Int64"),
+            "Col4": DatetimeIndex(
+                ["2020-01-01", "2021-01-01", "2022-01-01"], dtype="M8[ns]"
+            ),
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_dtype_nullable_int(parser):
+    df_result = read_xml(StringIO(xml_types), dtype={"sides": "Int64"}, parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_types,
+        parser=parser,
+        dtype={"sides": "Int64"},
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": Series([4.0, float("nan"), 3.0]).astype("Int64"),
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_dtype_float(parser):
+    df_result = read_xml(StringIO(xml_types), dtype={"degrees": "float"}, parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_types,
+        parser=parser,
+        dtype={"degrees": "float"},
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": Series([360, 360, 180]).astype("float"),
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_wrong_dtype(xml_books, parser, iterparse):
+    with pytest.raises(
+        ValueError, match=('Unable to parse string "Everyday Italian" at position 0')
+    ):
+        read_xml(
+            xml_books, dtype={"title": "Int64"}, parser=parser, iterparse=iterparse
+        )
+
+
+def test_both_dtype_converters(parser):
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": ["00360", "00360", "00180"],
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    with tm.assert_produces_warning(ParserWarning, match="Both a converter and dtype"):
+        df_result = read_xml(
+            StringIO(xml_types),
+            dtype={"degrees": "str"},
+            converters={"degrees": str},
+            parser=parser,
+        )
+        df_iter = read_xml_iterparse(
+            xml_types,
+            dtype={"degrees": "str"},
+            converters={"degrees": str},
+            parser=parser,
+            iterparse={"row": ["shape", "degrees", "sides"]},
+        )
+
+        tm.assert_frame_equal(df_result, df_expected)
+        tm.assert_frame_equal(df_iter, df_expected)
+
+
+# CONVERTERS
+
+
+def test_converters_str(parser):
+    df_result = read_xml(
+        StringIO(xml_types), converters={"degrees": str}, parser=parser
+    )
+    df_iter = read_xml_iterparse(
+        xml_types,
+        parser=parser,
+        converters={"degrees": str},
+        iterparse={"row": ["shape", "degrees", "sides"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": ["00360", "00360", "00180"],
+            "sides": [4.0, float("nan"), 3.0],
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_converters_date(parser):
+    convert_to_datetime = lambda x: to_datetime(x)
+    df_result = read_xml(
+        StringIO(xml_dates), converters={"date": convert_to_datetime}, parser=parser
+    )
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        converters={"date": convert_to_datetime},
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+            "date": to_datetime(["2020-01-01", "2021-01-01", "2022-01-01"]),
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_wrong_converters_type(xml_books, parser, iterparse):
+    with pytest.raises(TypeError, match=("Type converters must be a dict or subclass")):
+        read_xml(
+            xml_books, converters={"year", str}, parser=parser, iterparse=iterparse
+        )
+
+
+def test_callable_func_converters(xml_books, parser, iterparse):
+    with pytest.raises(TypeError, match=("'float' object is not callable")):
+        read_xml(
+            xml_books, converters={"year": float()}, parser=parser, iterparse=iterparse
+        )
+
+
+def test_callable_str_converters(xml_books, parser, iterparse):
+    with pytest.raises(TypeError, match=("'str' object is not callable")):
+        read_xml(
+            xml_books, converters={"year": "float"}, parser=parser, iterparse=iterparse
+        )
+
+
+# PARSE DATES
+
+
+def test_parse_dates_column_name(parser):
+    df_result = read_xml(StringIO(xml_dates), parse_dates=["date"], parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        parse_dates=["date"],
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+            "date": to_datetime(["2020-01-01", "2021-01-01", "2022-01-01"]),
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_parse_dates_column_index(parser):
+    df_result = read_xml(StringIO(xml_dates), parse_dates=[3], parser=parser)
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        parse_dates=[3],
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+            "date": to_datetime(["2020-01-01", "2021-01-01", "2022-01-01"]),
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_parse_dates_true(parser):
+    df_result = read_xml(StringIO(xml_dates), parse_dates=True, parser=parser)
+
+    df_iter = read_xml_iterparse(
+        xml_dates,
+        parser=parser,
+        parse_dates=True,
+        iterparse={"row": ["shape", "degrees", "sides", "date"]},
+    )
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+            "date": ["2020-01-01", "2021-01-01", "2022-01-01"],
+        }
+    )
+
+    tm.assert_frame_equal(df_result, df_expected)
+    tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_day_first_parse_dates(parser):
+    xml = """\
+<?xml version='1.0' encoding='utf-8'?>
+<data>
+  <row>
+    <shape>square</shape>
+    <degrees>00360</degrees>
+    <sides>4.0</sides>
+    <date>31/12/2020</date>
+   </row>
+  <row>
+    <shape>circle</shape>
+    <degrees>00360</degrees>
+    <sides/>
+    <date>31/12/2021</date>
+  </row>
+  <row>
+    <shape>triangle</shape>
+    <degrees>00180</degrees>
+    <sides>3.0</sides>
+    <date>31/12/2022</date>
+  </row>
+</data>"""
+
+    df_expected = DataFrame(
+        {
+            "shape": ["square", "circle", "triangle"],
+            "degrees": [360, 360, 180],
+            "sides": [4.0, float("nan"), 3.0],
+            "date": to_datetime(["2020-12-31", "2021-12-31", "2022-12-31"]),
+        }
+    )
+
+    with tm.assert_produces_warning(
+        UserWarning, match="Parsing dates in %d/%m/%Y format"
+    ):
+        df_result = read_xml(StringIO(xml), parse_dates=["date"], parser=parser)
+        df_iter = read_xml_iterparse(
+            xml,
+            parse_dates=["date"],
+            parser=parser,
+            iterparse={"row": ["shape", "degrees", "sides", "date"]},
+        )
+
+        tm.assert_frame_equal(df_result, df_expected)
+        tm.assert_frame_equal(df_iter, df_expected)
+
+
+def test_wrong_parse_dates_type(xml_books, parser, iterparse):
+    with pytest.raises(TypeError, match="Only booleans and lists are accepted"):
+        read_xml(xml_books, parse_dates={"date"}, parser=parser, iterparse=iterparse)
diff --git a/pandas/pandas/tests/libs/__init__.py b/pandas/pandas/tests/libs/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/libs/test_hashtable.py b/pandas/pandas/tests/libs/test_hashtable.py
new file mode 100644
index 0000000000000..50b561aefcf49
--- /dev/null
+++ b/pandas/pandas/tests/libs/test_hashtable.py
@@ -0,0 +1,781 @@
+from collections import namedtuple
+from collections.abc import Generator
+from contextlib import contextmanager
+import re
+import struct
+import tracemalloc
+
+import numpy as np
+import pytest
+
+from pandas._libs import hashtable as ht
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.algorithms import isin
+
+
+@contextmanager
+def activated_tracemalloc() -> Generator[None, None, None]:
+    tracemalloc.start()
+    try:
+        yield
+    finally:
+        tracemalloc.stop()
+
+
+def get_allocated_khash_memory():
+    snapshot = tracemalloc.take_snapshot()
+    snapshot = snapshot.filter_traces(
+        (tracemalloc.DomainFilter(True, ht.get_hashtable_trace_domain()),)
+    )
+    return sum(x.size for x in snapshot.traces)
+
+
+@pytest.mark.parametrize(
+    "table_type, dtype",
+    [
+        (ht.PyObjectHashTable, np.object_),
+        (ht.Complex128HashTable, np.complex128),
+        (ht.Int64HashTable, np.int64),
+        (ht.UInt64HashTable, np.uint64),
+        (ht.Float64HashTable, np.float64),
+        (ht.Complex64HashTable, np.complex64),
+        (ht.Int32HashTable, np.int32),
+        (ht.UInt32HashTable, np.uint32),
+        (ht.Float32HashTable, np.float32),
+        (ht.Int16HashTable, np.int16),
+        (ht.UInt16HashTable, np.uint16),
+        (ht.Int8HashTable, np.int8),
+        (ht.UInt8HashTable, np.uint8),
+        (ht.IntpHashTable, np.intp),
+    ],
+)
+class TestHashTable:
+    def test_get_set_contains_len(self, table_type, dtype):
+        index = 5
+        table = table_type(55)
+        assert len(table) == 0
+        assert index not in table
+
+        table.set_item(index, 42)
+        assert len(table) == 1
+        assert index in table
+        assert table.get_item(index) == 42
+
+        table.set_item(index + 1, 41)
+        assert index in table
+        assert index + 1 in table
+        assert len(table) == 2
+        assert table.get_item(index) == 42
+        assert table.get_item(index + 1) == 41
+
+        table.set_item(index, 21)
+        assert index in table
+        assert index + 1 in table
+        assert len(table) == 2
+        assert table.get_item(index) == 21
+        assert table.get_item(index + 1) == 41
+        assert index + 2 not in table
+
+        table.set_item(index + 1, 21)
+        assert index in table
+        assert index + 1 in table
+        assert len(table) == 2
+        assert table.get_item(index) == 21
+        assert table.get_item(index + 1) == 21
+
+        with pytest.raises(KeyError, match=str(index + 2)):
+            table.get_item(index + 2)
+
+    def test_get_set_contains_len_mask(self, table_type, dtype):
+        if table_type == ht.PyObjectHashTable:
+            pytest.skip("Mask not supported for object")
+        index = 5
+        table = table_type(55, uses_mask=True)
+        assert len(table) == 0
+        assert index not in table
+
+        table.set_item(index, 42)
+        assert len(table) == 1
+        assert index in table
+        assert table.get_item(index) == 42
+        with pytest.raises(KeyError, match="NA"):
+            table.get_na()
+
+        table.set_item(index + 1, 41)
+        table.set_na(41)
+        assert pd.NA in table
+        assert index in table
+        assert index + 1 in table
+        assert len(table) == 3
+        assert table.get_item(index) == 42
+        assert table.get_item(index + 1) == 41
+        assert table.get_na() == 41
+
+        table.set_na(21)
+        assert index in table
+        assert index + 1 in table
+        assert len(table) == 3
+        assert table.get_item(index + 1) == 41
+        assert table.get_na() == 21
+        assert index + 2 not in table
+
+        with pytest.raises(KeyError, match=str(index + 2)):
+            table.get_item(index + 2)
+
+    def test_map_keys_to_values(self, table_type, dtype, writable):
+        # only Int64HashTable has this method
+        if table_type == ht.Int64HashTable:
+            N = 77
+            table = table_type()
+            keys = np.arange(N).astype(dtype)
+            vals = np.arange(N).astype(np.int64) + N
+            keys.flags.writeable = writable
+            vals.flags.writeable = writable
+            table.map_keys_to_values(keys, vals)
+            for i in range(N):
+                assert table.get_item(keys[i]) == i + N
+
+    def test_map_locations(self, table_type, dtype, writable):
+        N = 8
+        table = table_type()
+        keys = (np.arange(N) + N).astype(dtype)
+        keys.flags.writeable = writable
+        table.map_locations(keys)
+        for i in range(N):
+            assert table.get_item(keys[i]) == i
+
+    def test_map_locations_mask(self, table_type, dtype, writable):
+        if table_type == ht.PyObjectHashTable:
+            pytest.skip("Mask not supported for object")
+        N = 3
+        table = table_type(uses_mask=True)
+        keys = (np.arange(N) + N).astype(dtype)
+        keys.flags.writeable = writable
+        table.map_locations(keys, np.array([False, False, True]))
+        for i in range(N - 1):
+            assert table.get_item(keys[i]) == i
+
+        with pytest.raises(KeyError, match=re.escape(str(keys[N - 1]))):
+            table.get_item(keys[N - 1])
+
+        assert table.get_na() == 2
+
+    def test_lookup(self, table_type, dtype, writable):
+        N = 3
+        table = table_type()
+        keys = (np.arange(N) + N).astype(dtype)
+        keys.flags.writeable = writable
+        table.map_locations(keys)
+        result = table.lookup(keys)
+        expected = np.arange(N)
+        tm.assert_numpy_array_equal(result.astype(np.int64), expected.astype(np.int64))
+
+    def test_lookup_wrong(self, table_type, dtype):
+        if dtype in (np.int8, np.uint8):
+            N = 100
+        else:
+            N = 512
+        table = table_type()
+        keys = (np.arange(N) + N).astype(dtype)
+        table.map_locations(keys)
+        wrong_keys = np.arange(N).astype(dtype)
+        result = table.lookup(wrong_keys)
+        assert np.all(result == -1)
+
+    def test_lookup_mask(self, table_type, dtype, writable):
+        if table_type == ht.PyObjectHashTable:
+            pytest.skip("Mask not supported for object")
+        N = 3
+        table = table_type(uses_mask=True)
+        keys = (np.arange(N) + N).astype(dtype)
+        mask = np.array([False, True, False])
+        keys.flags.writeable = writable
+        table.map_locations(keys, mask)
+        result = table.lookup(keys, mask)
+        expected = np.arange(N)
+        tm.assert_numpy_array_equal(result.astype(np.int64), expected.astype(np.int64))
+
+        result = table.lookup(np.array([1 + N]).astype(dtype), np.array([False]))
+        tm.assert_numpy_array_equal(
+            result.astype(np.int64), np.array([-1], dtype=np.int64)
+        )
+
+    def test_unique(self, table_type, dtype, writable):
+        if dtype in (np.int8, np.uint8):
+            N = 88
+        else:
+            N = 1000
+        table = table_type()
+        expected = (np.arange(N) + N).astype(dtype)
+        keys = np.repeat(expected, 5)
+        keys.flags.writeable = writable
+        unique = table.unique(keys)
+        tm.assert_numpy_array_equal(unique, expected)
+
+    def test_tracemalloc_works(self, table_type, dtype):
+        if dtype in (np.int8, np.uint8):
+            N = 256
+        else:
+            N = 30000
+        keys = np.arange(N).astype(dtype)
+        with activated_tracemalloc():
+            table = table_type()
+            table.map_locations(keys)
+            used = get_allocated_khash_memory()
+            my_size = table.sizeof()
+            assert used == my_size
+            del table
+            assert get_allocated_khash_memory() == 0
+
+    def test_tracemalloc_for_empty(self, table_type, dtype):
+        with activated_tracemalloc():
+            table = table_type()
+            used = get_allocated_khash_memory()
+            my_size = table.sizeof()
+            assert used == my_size
+            del table
+            assert get_allocated_khash_memory() == 0
+
+    def test_get_state(self, table_type, dtype):
+        table = table_type(1000)
+        state = table.get_state()
+        assert state["size"] == 0
+        assert state["n_occupied"] == 0
+        assert "n_buckets" in state
+        assert "upper_bound" in state
+
+    @pytest.mark.parametrize("N", range(1, 110))
+    def test_no_reallocation(self, table_type, dtype, N):
+        keys = np.arange(N).astype(dtype)
+        preallocated_table = table_type(N)
+        n_buckets_start = preallocated_table.get_state()["n_buckets"]
+        preallocated_table.map_locations(keys)
+        n_buckets_end = preallocated_table.get_state()["n_buckets"]
+        # original number of buckets was enough:
+        assert n_buckets_start == n_buckets_end
+        # check with clean table (not too much preallocated)
+        clean_table = table_type()
+        clean_table.map_locations(keys)
+        assert n_buckets_start == clean_table.get_state()["n_buckets"]
+
+
+class TestHashTableUnsorted:
+    # TODO: moved from test_algos; may be redundancies with other tests
+    def test_string_hashtable_set_item_signature(self):
+        # GH#30419 fix typing in StringHashTable.set_item to prevent segfault
+        tbl = ht.StringHashTable()
+
+        tbl.set_item("key", 1)
+        assert tbl.get_item("key") == 1
+
+        with pytest.raises(TypeError, match="'key' has incorrect type"):
+            # key arg typed as string, not object
+            tbl.set_item(4, 6)
+        with pytest.raises(TypeError, match="'val' has incorrect type"):
+            tbl.get_item(4)
+
+    def test_lookup_nan(self, writable):
+        # GH#21688 ensure we can deal with readonly memory views
+        xs = np.array([2.718, 3.14, np.nan, -7, 5, 2, 3])
+        xs.setflags(write=writable)
+        m = ht.Float64HashTable()
+        m.map_locations(xs)
+        tm.assert_numpy_array_equal(m.lookup(xs), np.arange(len(xs), dtype=np.intp))
+
+    def test_add_signed_zeros(self):
+        # GH#21866 inconsistent hash-function for float64
+        # default hash-function would lead to different hash-buckets
+        # for 0.0 and -0.0 if there are more than 2^30 hash-buckets
+        # but this would mean 16GB
+        N = 4  # 12 * 10**8 would trigger the error, if you have enough memory
+        m = ht.Float64HashTable(N)
+        m.set_item(0.0, 0)
+        m.set_item(-0.0, 0)
+        assert len(m) == 1  # 0.0 and -0.0 are equivalent
+
+    def test_add_different_nans(self):
+        # GH#21866 inconsistent hash-function for float64
+        # create different nans from bit-patterns:
+        NAN1 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000000))[0]
+        NAN2 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000001))[0]
+        assert NAN1 != NAN1
+        assert NAN2 != NAN2
+        # default hash function would lead to different hash-buckets
+        # for NAN1 and NAN2 even if there are only 4 buckets:
+        m = ht.Float64HashTable()
+        m.set_item(NAN1, 0)
+        m.set_item(NAN2, 0)
+        assert len(m) == 1  # NAN1 and NAN2 are equivalent
+
+    def test_lookup_overflow(self, writable):
+        xs = np.array([1, 2, 2**63], dtype=np.uint64)
+        # GH 21688 ensure we can deal with readonly memory views
+        xs.setflags(write=writable)
+        m = ht.UInt64HashTable()
+        m.map_locations(xs)
+        tm.assert_numpy_array_equal(m.lookup(xs), np.arange(len(xs), dtype=np.intp))
+
+    @pytest.mark.parametrize("nvals", [0, 10])  # resizing to 0 is special case
+    @pytest.mark.parametrize(
+        "htable, uniques, dtype, safely_resizes",
+        [
+            (ht.PyObjectHashTable, ht.ObjectVector, "object", False),
+            (ht.StringHashTable, ht.ObjectVector, "object", True),
+            (ht.Float64HashTable, ht.Float64Vector, "float64", False),
+            (ht.Int64HashTable, ht.Int64Vector, "int64", False),
+            (ht.Int32HashTable, ht.Int32Vector, "int32", False),
+            (ht.UInt64HashTable, ht.UInt64Vector, "uint64", False),
+        ],
+    )
+    def test_vector_resize(
+        self, writable, htable, uniques, dtype, safely_resizes, nvals
+    ):
+        # Test for memory errors after internal vector
+        # reallocations (GH 7157)
+        # Changed from using np.random.default_rng(2).rand to range
+        # which could cause flaky CI failures when safely_resizes=False
+        vals = np.array(range(1000), dtype=dtype)
+
+        # GH 21688 ensures we can deal with read-only memory views
+        vals.setflags(write=writable)
+
+        # initialise instances; cannot initialise in parametrization,
+        # as otherwise external views would be held on the array (which is
+        # one of the things this test is checking)
+        htable = htable()
+        uniques = uniques()
+
+        # get_labels may append to uniques
+        htable.get_labels(vals[:nvals], uniques, 0, -1)
+        # to_array() sets an external_view_exists flag on uniques.
+        tmp = uniques.to_array()
+        oldshape = tmp.shape
+
+        # subsequent get_labels() calls can no longer append to it
+        # (except for StringHashTables + ObjectVector)
+        if safely_resizes:
+            htable.get_labels(vals, uniques, 0, -1)
+        else:
+            with pytest.raises(ValueError, match="external reference.*"):
+                htable.get_labels(vals, uniques, 0, -1)
+
+        uniques.to_array()  # should not raise here
+        assert tmp.shape == oldshape
+
+    @pytest.mark.parametrize(
+        "hashtable",
+        [
+            ht.PyObjectHashTable,
+            ht.StringHashTable,
+            ht.Float64HashTable,
+            ht.Int64HashTable,
+            ht.Int32HashTable,
+            ht.UInt64HashTable,
+        ],
+    )
+    def test_hashtable_large_sizehint(self, hashtable):
+        # GH#22729 smoketest for not raising when passing a large size_hint
+        size_hint = np.iinfo(np.uint32).max + 1
+        hashtable(size_hint=size_hint)
+
+
+class TestPyObjectHashTableWithNans:
+    def test_nan_float(self):
+        nan1 = float("nan")
+        nan2 = float("nan")
+        assert nan1 is not nan2
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+
+    def test_nan_complex_both(self):
+        nan1 = complex(float("nan"), float("nan"))
+        nan2 = complex(float("nan"), float("nan"))
+        assert nan1 is not nan2
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+
+    def test_nan_complex_real(self):
+        nan1 = complex(float("nan"), 1)
+        nan2 = complex(float("nan"), 1)
+        other = complex(float("nan"), 2)
+        assert nan1 is not nan2
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+        with pytest.raises(KeyError, match=re.escape(repr(other))):
+            table.get_item(other)
+
+    def test_nan_complex_imag(self):
+        nan1 = complex(1, float("nan"))
+        nan2 = complex(1, float("nan"))
+        other = complex(2, float("nan"))
+        assert nan1 is not nan2
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+        with pytest.raises(KeyError, match=re.escape(repr(other))):
+            table.get_item(other)
+
+    def test_nan_in_tuple(self):
+        nan1 = (float("nan"),)
+        nan2 = (float("nan"),)
+        assert nan1[0] is not nan2[0]
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+
+    def test_nan_in_nested_tuple(self):
+        nan1 = (1, (2, (float("nan"),)))
+        nan2 = (1, (2, (float("nan"),)))
+        other = (1, 2)
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+        with pytest.raises(KeyError, match=re.escape(repr(other))):
+            table.get_item(other)
+
+    def test_nan_in_namedtuple(self):
+        T = namedtuple("T", ["x"])
+        nan1 = T(float("nan"))
+        nan2 = T(float("nan"))
+        assert nan1.x is not nan2.x
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+
+    def test_nan_in_nested_namedtuple(self):
+        T = namedtuple("T", ["x", "y"])
+        nan1 = T(1, (2, (float("nan"),)))
+        nan2 = T(1, (2, (float("nan"),)))
+        other = T(1, 2)
+        table = ht.PyObjectHashTable()
+        table.set_item(nan1, 42)
+        assert table.get_item(nan2) == 42
+        with pytest.raises(KeyError, match=re.escape(repr(other))):
+            table.get_item(other)
+
+
+def test_hash_equal_tuple_with_nans():
+    a = (float("nan"), (float("nan"), float("nan")))
+    b = (float("nan"), (float("nan"), float("nan")))
+    assert ht.object_hash(a) == ht.object_hash(b)
+    assert ht.objects_are_equal(a, b)
+
+
+def test_hash_equal_namedtuple_with_nans():
+    T = namedtuple("T", ["x", "y"])
+    a = T(float("nan"), (float("nan"), float("nan")))
+    b = T(float("nan"), (float("nan"), float("nan")))
+    assert ht.object_hash(a) == ht.object_hash(b)
+    assert ht.objects_are_equal(a, b)
+
+
+def test_hash_equal_namedtuple_and_tuple():
+    T = namedtuple("T", ["x", "y"])
+    a = T(1, (2, 3))
+    b = (1, (2, 3))
+    assert ht.object_hash(a) == ht.object_hash(b)
+    assert ht.objects_are_equal(a, b)
+
+
+def test_get_labels_groupby_for_Int64(writable):
+    table = ht.Int64HashTable()
+    vals = np.array([1, 2, -1, 2, 1, -1], dtype=np.int64)
+    vals.flags.writeable = writable
+    arr, unique = table.get_labels_groupby(vals)
+    expected_arr = np.array([0, 1, -1, 1, 0, -1], dtype=np.intp)
+    expected_unique = np.array([1, 2], dtype=np.int64)
+    tm.assert_numpy_array_equal(arr, expected_arr)
+    tm.assert_numpy_array_equal(unique, expected_unique)
+
+
+def test_tracemalloc_works_for_StringHashTable():
+    N = 1000
+    keys = np.arange(N).astype(np.str_).astype(np.object_)
+    with activated_tracemalloc():
+        table = ht.StringHashTable()
+        table.map_locations(keys)
+        used = get_allocated_khash_memory()
+        my_size = table.sizeof()
+        assert used == my_size
+        del table
+        assert get_allocated_khash_memory() == 0
+
+
+def test_tracemalloc_for_empty_StringHashTable():
+    with activated_tracemalloc():
+        table = ht.StringHashTable()
+        used = get_allocated_khash_memory()
+        my_size = table.sizeof()
+        assert used == my_size
+        del table
+        assert get_allocated_khash_memory() == 0
+
+
+@pytest.mark.parametrize("N", range(1, 110))
+def test_no_reallocation_StringHashTable(N):
+    keys = np.arange(N).astype(np.str_).astype(np.object_)
+    preallocated_table = ht.StringHashTable(N)
+    n_buckets_start = preallocated_table.get_state()["n_buckets"]
+    preallocated_table.map_locations(keys)
+    n_buckets_end = preallocated_table.get_state()["n_buckets"]
+    # original number of buckets was enough:
+    assert n_buckets_start == n_buckets_end
+    # check with clean table (not too much preallocated)
+    clean_table = ht.StringHashTable()
+    clean_table.map_locations(keys)
+    assert n_buckets_start == clean_table.get_state()["n_buckets"]
+
+
+@pytest.mark.parametrize(
+    "table_type, dtype",
+    [
+        (ht.Float64HashTable, np.float64),
+        (ht.Float32HashTable, np.float32),
+        (ht.Complex128HashTable, np.complex128),
+        (ht.Complex64HashTable, np.complex64),
+    ],
+)
+class TestHashTableWithNans:
+    def test_get_set_contains_len(self, table_type, dtype):
+        index = float("nan")
+        table = table_type()
+        assert index not in table
+
+        table.set_item(index, 42)
+        assert len(table) == 1
+        assert index in table
+        assert table.get_item(index) == 42
+
+        table.set_item(index, 41)
+        assert len(table) == 1
+        assert index in table
+        assert table.get_item(index) == 41
+
+    def test_map_locations(self, table_type, dtype):
+        N = 10
+        table = table_type()
+        keys = np.full(N, np.nan, dtype=dtype)
+        table.map_locations(keys)
+        assert len(table) == 1
+        assert table.get_item(np.nan) == N - 1
+
+    def test_unique(self, table_type, dtype):
+        N = 1020
+        table = table_type()
+        keys = np.full(N, np.nan, dtype=dtype)
+        unique = table.unique(keys)
+        assert np.all(np.isnan(unique)) and len(unique) == 1
+
+
+def test_unique_for_nan_objects_floats():
+    table = ht.PyObjectHashTable()
+    keys = np.array([float("nan") for i in range(50)], dtype=np.object_)
+    unique = table.unique(keys)
+    assert len(unique) == 1
+
+
+def test_unique_for_nan_objects_complex():
+    table = ht.PyObjectHashTable()
+    keys = np.array([complex(float("nan"), 1.0) for i in range(50)], dtype=np.object_)
+    unique = table.unique(keys)
+    assert len(unique) == 1
+
+
+def test_unique_for_nan_objects_tuple():
+    table = ht.PyObjectHashTable()
+    keys = np.array(
+        [1] + [(1.0, (float("nan"), 1.0)) for i in range(50)], dtype=np.object_
+    )
+    unique = table.unique(keys)
+    assert len(unique) == 2
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        np.object_,
+        np.complex128,
+        np.int64,
+        np.uint64,
+        np.float64,
+        np.complex64,
+        np.int32,
+        np.uint32,
+        np.float32,
+        np.int16,
+        np.uint16,
+        np.int8,
+        np.uint8,
+        np.intp,
+    ],
+)
+class TestHelpFunctions:
+    def test_value_count(self, dtype, writable):
+        N = 43
+        expected = (np.arange(N) + N).astype(dtype)
+        values = np.repeat(expected, 5)
+        values.flags.writeable = writable
+        keys, counts, _ = ht.value_count(values, False)
+        tm.assert_numpy_array_equal(np.sort(keys), expected)
+        assert np.all(counts == 5)
+
+    def test_value_count_mask(self, dtype):
+        if dtype == np.object_:
+            pytest.skip("mask not implemented for object dtype")
+        values = np.array([1] * 5, dtype=dtype)
+        mask = np.zeros((5,), dtype=np.bool_)
+        mask[1] = True
+        mask[4] = True
+        keys, counts, na_counter = ht.value_count(values, False, mask=mask)
+        assert len(keys) == 2
+        assert na_counter == 2
+
+    def test_value_count_stable(self, dtype, writable):
+        # GH12679
+        values = np.array([2, 1, 5, 22, 3, -1, 8]).astype(dtype)
+        values.flags.writeable = writable
+        keys, counts, _ = ht.value_count(values, False)
+        tm.assert_numpy_array_equal(keys, values)
+        assert np.all(counts == 1)
+
+    def test_duplicated_first(self, dtype, writable):
+        N = 100
+        values = np.repeat(np.arange(N).astype(dtype), 5)
+        values.flags.writeable = writable
+        result = ht.duplicated(values)
+        expected = np.ones_like(values, dtype=np.bool_)
+        expected[::5] = False
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_ismember_yes(self, dtype, writable):
+        N = 127
+        arr = np.arange(N).astype(dtype)
+        values = np.arange(N).astype(dtype)
+        arr.flags.writeable = writable
+        values.flags.writeable = writable
+        result = ht.ismember(arr, values)
+        expected = np.ones_like(values, dtype=np.bool_)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_ismember_no(self, dtype):
+        N = 17
+        arr = np.arange(N).astype(dtype)
+        values = (np.arange(N) + N).astype(dtype)
+        result = ht.ismember(arr, values)
+        expected = np.zeros_like(values, dtype=np.bool_)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_mode(self, dtype, writable):
+        if dtype in (np.int8, np.uint8):
+            N = 53
+        else:
+            N = 11111
+        values = np.repeat(np.arange(N).astype(dtype), 5)
+        values[0] = 42
+        values.flags.writeable = writable
+        result = ht.mode(values, False)[0]
+        assert result == 42
+
+    def test_mode_stable(self, dtype, writable):
+        values = np.array([2, 1, 5, 22, 3, -1, 8]).astype(dtype)
+        values.flags.writeable = writable
+        keys = ht.mode(values, False)[0]
+        tm.assert_numpy_array_equal(keys, values)
+
+
+def test_modes_with_nans():
+    # GH42688, nans aren't mangled
+    nulls = [pd.NA, np.nan, pd.NaT, None]
+    values = np.array([True] + nulls * 2, dtype=np.object_)
+    modes = ht.mode(values, False)[0]
+    assert modes.size == len(nulls)
+
+
+def test_unique_label_indices_intp(writable):
+    keys = np.array([1, 2, 2, 2, 1, 3], dtype=np.intp)
+    keys.flags.writeable = writable
+    result = ht.unique_label_indices(keys)
+    expected = np.array([0, 1, 5], dtype=np.intp)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_unique_label_indices():
+    a = np.random.default_rng(2).integers(1, 1 << 10, 1 << 15).astype(np.intp)
+
+    left = ht.unique_label_indices(a)
+    right = np.unique(a, return_index=True)[1]
+
+    tm.assert_numpy_array_equal(left, right, check_dtype=False)
+
+    a[np.random.default_rng(2).choice(len(a), 10)] = -1
+    left = ht.unique_label_indices(a)
+    right = np.unique(a, return_index=True)[1][1:]
+    tm.assert_numpy_array_equal(left, right, check_dtype=False)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        np.float64,
+        np.float32,
+        np.complex128,
+        np.complex64,
+    ],
+)
+class TestHelpFunctionsWithNans:
+    def test_value_count(self, dtype):
+        values = np.array([np.nan, np.nan, np.nan], dtype=dtype)
+        keys, counts, _ = ht.value_count(values, True)
+        assert len(keys) == 0
+        keys, counts, _ = ht.value_count(values, False)
+        assert len(keys) == 1 and np.all(np.isnan(keys))
+        assert counts[0] == 3
+
+    def test_duplicated_first(self, dtype):
+        values = np.array([np.nan, np.nan, np.nan], dtype=dtype)
+        result = ht.duplicated(values)
+        expected = np.array([False, True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_ismember_yes(self, dtype):
+        arr = np.array([np.nan, np.nan, np.nan], dtype=dtype)
+        values = np.array([np.nan, np.nan], dtype=dtype)
+        result = ht.ismember(arr, values)
+        expected = np.array([True, True, True], dtype=np.bool_)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_ismember_no(self, dtype):
+        arr = np.array([np.nan, np.nan, np.nan], dtype=dtype)
+        values = np.array([1], dtype=dtype)
+        result = ht.ismember(arr, values)
+        expected = np.array([False, False, False], dtype=np.bool_)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_mode(self, dtype):
+        values = np.array([42, np.nan, np.nan, np.nan], dtype=dtype)
+        assert ht.mode(values, True)[0] == 42
+        assert np.isnan(ht.mode(values, False)[0])
+
+
+def test_ismember_tuple_with_nans():
+    # GH-41836
+    values = np.empty(2, dtype=object)
+    values[:] = [("a", float("nan")), ("b", 1)]
+    comps = [("a", float("nan"))]
+
+    result = isin(values, comps)
+    expected = np.array([True, False], dtype=np.bool_)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_float_complex_int_are_equal_as_objects():
+    values = ["a", 5, 5.0, 5.0 + 0j]
+    comps = list(range(129))
+    result = isin(np.array(values, dtype=object), np.asarray(comps))
+    expected = np.array([False, True, True, True], dtype=np.bool_)
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/libs/test_join.py b/pandas/pandas/tests/libs/test_join.py
new file mode 100644
index 0000000000000..bf8b4fabc54cb
--- /dev/null
+++ b/pandas/pandas/tests/libs/test_join.py
@@ -0,0 +1,388 @@
+import numpy as np
+import pytest
+
+from pandas._libs import join as libjoin
+from pandas._libs.join import (
+    inner_join,
+    left_outer_join,
+)
+
+import pandas._testing as tm
+
+
+class TestIndexer:
+    @pytest.mark.parametrize(
+        "dtype", ["int32", "int64", "float32", "float64", "object"]
+    )
+    def test_outer_join_indexer(self, dtype):
+        indexer = libjoin.outer_join_indexer
+
+        left = np.arange(3, dtype=dtype)
+        right = np.arange(2, 5, dtype=dtype)
+        empty = np.array([], dtype=dtype)
+
+        result, lindexer, rindexer = indexer(left, right)
+        assert isinstance(result, np.ndarray)
+        assert isinstance(lindexer, np.ndarray)
+        assert isinstance(rindexer, np.ndarray)
+        tm.assert_numpy_array_equal(result, np.arange(5, dtype=dtype))
+        exp = np.array([0, 1, 2, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(lindexer, exp)
+        exp = np.array([-1, -1, 0, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(rindexer, exp)
+
+        result, lindexer, rindexer = indexer(empty, right)
+        tm.assert_numpy_array_equal(result, right)
+        exp = np.array([-1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(lindexer, exp)
+        exp = np.array([0, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(rindexer, exp)
+
+        result, lindexer, rindexer = indexer(left, empty)
+        tm.assert_numpy_array_equal(result, left)
+        exp = np.array([0, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(lindexer, exp)
+        exp = np.array([-1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(rindexer, exp)
+
+    def test_cython_left_outer_join(self):
+        left = np.array([0, 1, 2, 1, 2, 0, 0, 1, 2, 3, 3], dtype=np.intp)
+        right = np.array([1, 1, 0, 4, 2, 2, 1], dtype=np.intp)
+        max_group = 5
+
+        ls, rs = left_outer_join(left, right, max_group)
+
+        exp_ls = left.argsort(kind="mergesort")
+        exp_rs = right.argsort(kind="mergesort")
+
+        exp_li = np.array([0, 1, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 7, 7, 8, 8, 9, 10])
+        exp_ri = np.array(
+            [0, 0, 0, 1, 2, 3, 1, 2, 3, 1, 2, 3, 4, 5, 4, 5, 4, 5, -1, -1]
+        )
+
+        exp_ls = exp_ls.take(exp_li)
+        exp_ls[exp_li == -1] = -1
+
+        exp_rs = exp_rs.take(exp_ri)
+        exp_rs[exp_ri == -1] = -1
+
+        tm.assert_numpy_array_equal(ls, exp_ls, check_dtype=False)
+        tm.assert_numpy_array_equal(rs, exp_rs, check_dtype=False)
+
+    def test_cython_right_outer_join(self):
+        left = np.array([0, 1, 2, 1, 2, 0, 0, 1, 2, 3, 3], dtype=np.intp)
+        right = np.array([1, 1, 0, 4, 2, 2, 1], dtype=np.intp)
+        max_group = 5
+
+        rs, ls = left_outer_join(right, left, max_group)
+
+        exp_ls = left.argsort(kind="mergesort")
+        exp_rs = right.argsort(kind="mergesort")
+
+        #            0        1        1        1
+        exp_li = np.array(
+            [
+                0,
+                1,
+                2,
+                3,
+                4,
+                5,
+                3,
+                4,
+                5,
+                3,
+                4,
+                5,
+                #            2        2        4
+                6,
+                7,
+                8,
+                6,
+                7,
+                8,
+                -1,
+            ]
+        )
+        exp_ri = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6])
+
+        exp_ls = exp_ls.take(exp_li)
+        exp_ls[exp_li == -1] = -1
+
+        exp_rs = exp_rs.take(exp_ri)
+        exp_rs[exp_ri == -1] = -1
+
+        tm.assert_numpy_array_equal(ls, exp_ls)
+        tm.assert_numpy_array_equal(rs, exp_rs)
+
+    def test_cython_inner_join(self):
+        left = np.array([0, 1, 2, 1, 2, 0, 0, 1, 2, 3, 3], dtype=np.intp)
+        right = np.array([1, 1, 0, 4, 2, 2, 1, 4], dtype=np.intp)
+        max_group = 5
+
+        ls, rs = inner_join(left, right, max_group)
+
+        exp_ls = left.argsort(kind="mergesort")
+        exp_rs = right.argsort(kind="mergesort")
+
+        exp_li = np.array([0, 1, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 7, 7, 8, 8])
+        exp_ri = np.array([0, 0, 0, 1, 2, 3, 1, 2, 3, 1, 2, 3, 4, 5, 4, 5, 4, 5])
+
+        exp_ls = exp_ls.take(exp_li)
+        exp_ls[exp_li == -1] = -1
+
+        exp_rs = exp_rs.take(exp_ri)
+        exp_rs[exp_ri == -1] = -1
+
+        tm.assert_numpy_array_equal(ls, exp_ls)
+        tm.assert_numpy_array_equal(rs, exp_rs)
+
+
+def test_left_join_indexer_unique(writable):
+    a = np.array([1, 2, 3, 4, 5], dtype=np.int64)
+    b = np.array([2, 2, 3, 4, 4], dtype=np.int64)
+    # GH#37312, GH#37264
+    a.setflags(write=writable)
+    b.setflags(write=writable)
+
+    result = libjoin.left_join_indexer_unique(b, a)
+    expected = np.array([1, 1, 2, 3, 3], dtype=np.intp)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_left_outer_join_bug():
+    left = np.array(
+        [
+            0,
+            1,
+            0,
+            1,
+            1,
+            2,
+            3,
+            1,
+            0,
+            2,
+            1,
+            2,
+            0,
+            1,
+            1,
+            2,
+            3,
+            2,
+            3,
+            2,
+            1,
+            1,
+            3,
+            0,
+            3,
+            2,
+            3,
+            0,
+            0,
+            2,
+            3,
+            2,
+            0,
+            3,
+            1,
+            3,
+            0,
+            1,
+            3,
+            0,
+            0,
+            1,
+            0,
+            3,
+            1,
+            0,
+            1,
+            0,
+            1,
+            1,
+            0,
+            2,
+            2,
+            2,
+            2,
+            2,
+            0,
+            3,
+            1,
+            2,
+            0,
+            0,
+            3,
+            1,
+            3,
+            2,
+            2,
+            0,
+            1,
+            3,
+            0,
+            2,
+            3,
+            2,
+            3,
+            3,
+            2,
+            3,
+            3,
+            1,
+            3,
+            2,
+            0,
+            0,
+            3,
+            1,
+            1,
+            1,
+            0,
+            2,
+            3,
+            3,
+            1,
+            2,
+            0,
+            3,
+            1,
+            2,
+            0,
+            2,
+        ],
+        dtype=np.intp,
+    )
+
+    right = np.array([3, 1], dtype=np.intp)
+    max_groups = 4
+
+    lidx, ridx = libjoin.left_outer_join(left, right, max_groups, sort=False)
+
+    exp_lidx = np.arange(len(left), dtype=np.intp)
+    exp_ridx = -np.ones(len(left), dtype=np.intp)
+
+    exp_ridx[left == 1] = 1
+    exp_ridx[left == 3] = 0
+
+    tm.assert_numpy_array_equal(lidx, exp_lidx)
+    tm.assert_numpy_array_equal(ridx, exp_ridx)
+
+
+def test_inner_join_indexer():
+    a = np.array([1, 2, 3, 4, 5], dtype=np.int64)
+    b = np.array([0, 3, 5, 7, 9], dtype=np.int64)
+
+    index, ares, bres = libjoin.inner_join_indexer(a, b)
+
+    index_exp = np.array([3, 5], dtype=np.int64)
+    tm.assert_almost_equal(index, index_exp)
+
+    aexp = np.array([2, 4], dtype=np.intp)
+    bexp = np.array([1, 2], dtype=np.intp)
+    tm.assert_almost_equal(ares, aexp)
+    tm.assert_almost_equal(bres, bexp)
+
+    a = np.array([5], dtype=np.int64)
+    b = np.array([5], dtype=np.int64)
+
+    index, ares, bres = libjoin.inner_join_indexer(a, b)
+    tm.assert_numpy_array_equal(index, np.array([5], dtype=np.int64))
+    tm.assert_numpy_array_equal(ares, np.array([0], dtype=np.intp))
+    tm.assert_numpy_array_equal(bres, np.array([0], dtype=np.intp))
+
+
+def test_outer_join_indexer():
+    a = np.array([1, 2, 3, 4, 5], dtype=np.int64)
+    b = np.array([0, 3, 5, 7, 9], dtype=np.int64)
+
+    index, ares, bres = libjoin.outer_join_indexer(a, b)
+
+    index_exp = np.array([0, 1, 2, 3, 4, 5, 7, 9], dtype=np.int64)
+    tm.assert_almost_equal(index, index_exp)
+
+    aexp = np.array([-1, 0, 1, 2, 3, 4, -1, -1], dtype=np.intp)
+    bexp = np.array([0, -1, -1, 1, -1, 2, 3, 4], dtype=np.intp)
+    tm.assert_almost_equal(ares, aexp)
+    tm.assert_almost_equal(bres, bexp)
+
+    a = np.array([5], dtype=np.int64)
+    b = np.array([5], dtype=np.int64)
+
+    index, ares, bres = libjoin.outer_join_indexer(a, b)
+    tm.assert_numpy_array_equal(index, np.array([5], dtype=np.int64))
+    tm.assert_numpy_array_equal(ares, np.array([0], dtype=np.intp))
+    tm.assert_numpy_array_equal(bres, np.array([0], dtype=np.intp))
+
+
+def test_left_join_indexer():
+    a = np.array([1, 2, 3, 4, 5], dtype=np.int64)
+    b = np.array([0, 3, 5, 7, 9], dtype=np.int64)
+
+    index, ares, bres = libjoin.left_join_indexer(a, b)
+
+    tm.assert_almost_equal(index, a)
+
+    aexp = np.array([0, 1, 2, 3, 4], dtype=np.intp)
+    bexp = np.array([-1, -1, 1, -1, 2], dtype=np.intp)
+    tm.assert_almost_equal(ares, aexp)
+    tm.assert_almost_equal(bres, bexp)
+
+    a = np.array([5], dtype=np.int64)
+    b = np.array([5], dtype=np.int64)
+
+    index, ares, bres = libjoin.left_join_indexer(a, b)
+    tm.assert_numpy_array_equal(index, np.array([5], dtype=np.int64))
+    tm.assert_numpy_array_equal(ares, np.array([0], dtype=np.intp))
+    tm.assert_numpy_array_equal(bres, np.array([0], dtype=np.intp))
+
+
+def test_left_join_indexer2():
+    idx = np.array([1, 1, 2, 5], dtype=np.int64)
+    idx2 = np.array([1, 2, 5, 7, 9], dtype=np.int64)
+
+    res, lidx, ridx = libjoin.left_join_indexer(idx2, idx)
+
+    exp_res = np.array([1, 1, 2, 5, 7, 9], dtype=np.int64)
+    tm.assert_almost_equal(res, exp_res)
+
+    exp_lidx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+    tm.assert_almost_equal(lidx, exp_lidx)
+
+    exp_ridx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+    tm.assert_almost_equal(ridx, exp_ridx)
+
+
+def test_outer_join_indexer2():
+    idx = np.array([1, 1, 2, 5], dtype=np.int64)
+    idx2 = np.array([1, 2, 5, 7, 9], dtype=np.int64)
+
+    res, lidx, ridx = libjoin.outer_join_indexer(idx2, idx)
+
+    exp_res = np.array([1, 1, 2, 5, 7, 9], dtype=np.int64)
+    tm.assert_almost_equal(res, exp_res)
+
+    exp_lidx = np.array([0, 0, 1, 2, 3, 4], dtype=np.intp)
+    tm.assert_almost_equal(lidx, exp_lidx)
+
+    exp_ridx = np.array([0, 1, 2, 3, -1, -1], dtype=np.intp)
+    tm.assert_almost_equal(ridx, exp_ridx)
+
+
+def test_inner_join_indexer2():
+    idx = np.array([1, 1, 2, 5], dtype=np.int64)
+    idx2 = np.array([1, 2, 5, 7, 9], dtype=np.int64)
+
+    res, lidx, ridx = libjoin.inner_join_indexer(idx2, idx)
+
+    exp_res = np.array([1, 1, 2, 5], dtype=np.int64)
+    tm.assert_almost_equal(res, exp_res)
+
+    exp_lidx = np.array([0, 0, 1, 2], dtype=np.intp)
+    tm.assert_almost_equal(lidx, exp_lidx)
+
+    exp_ridx = np.array([0, 1, 2, 3], dtype=np.intp)
+    tm.assert_almost_equal(ridx, exp_ridx)
diff --git a/pandas/pandas/tests/libs/test_lib.py b/pandas/pandas/tests/libs/test_lib.py
new file mode 100644
index 0000000000000..17dae1879f3b8
--- /dev/null
+++ b/pandas/pandas/tests/libs/test_lib.py
@@ -0,0 +1,299 @@
+import pickle
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    Timedelta,
+    lib,
+    writers as libwriters,
+)
+from pandas.compat import IS64
+
+from pandas import Index
+import pandas._testing as tm
+
+
+class TestMisc:
+    def test_max_len_string_array(self):
+        arr = a = np.array(["foo", "b", np.nan], dtype="object")
+        assert libwriters.max_len_string_array(arr) == 3
+
+        # unicode
+        arr = a.astype("U").astype(object)
+        assert libwriters.max_len_string_array(arr) == 3
+
+        # bytes for python3
+        arr = a.astype("S").astype(object)
+        assert libwriters.max_len_string_array(arr) == 3
+
+        # raises
+        msg = "No matching signature found"
+        with pytest.raises(TypeError, match=msg):
+            libwriters.max_len_string_array(arr.astype("U"))
+
+    def test_fast_unique_multiple_list_gen_sort(self):
+        keys = [["p", "a"], ["n", "d"], ["a", "s"]]
+
+        gen = (key for key in keys)
+        expected = np.array(["a", "d", "n", "p", "s"])
+        out = lib.fast_unique_multiple_list_gen(gen, sort=True)
+        tm.assert_numpy_array_equal(np.array(out), expected)
+
+        gen = (key for key in keys)
+        expected = np.array(["p", "a", "n", "d", "s"])
+        out = lib.fast_unique_multiple_list_gen(gen, sort=False)
+        tm.assert_numpy_array_equal(np.array(out), expected)
+
+    def test_fast_multiget_timedelta_resos(self):
+        # This will become relevant for test_constructor_dict_timedelta64_index
+        #  once Timedelta constructor preserves reso when passed a
+        #  np.timedelta64 object
+        td = Timedelta(days=1)
+
+        mapping1 = {td: 1}
+        mapping2 = {td.as_unit("s"): 1}
+
+        oindex = Index([td * n for n in range(3)])._values.astype(object)
+
+        expected = lib.fast_multiget(mapping1, oindex)
+        result = lib.fast_multiget(mapping2, oindex)
+        tm.assert_numpy_array_equal(result, expected)
+
+        # case that can't be cast to td64ns
+        td = Timedelta(np.timedelta64(146000, "D"))
+        assert hash(td) == hash(td.as_unit("ms"))
+        assert hash(td) == hash(td.as_unit("us"))
+        mapping1 = {td: 1}
+        mapping2 = {td.as_unit("ms"): 1}
+
+        oindex = Index([td * n for n in range(3)])._values.astype(object)
+
+        expected = lib.fast_multiget(mapping1, oindex)
+        result = lib.fast_multiget(mapping2, oindex)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestIndexing:
+    def test_maybe_indices_to_slice_left_edge(self):
+        target = np.arange(100)
+
+        # slice
+        indices = np.array([], dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize("end", [1, 2, 5, 20, 99])
+    @pytest.mark.parametrize("step", [1, 2, 4])
+    def test_maybe_indices_to_slice_left_edge_not_slice_end_steps(self, end, step):
+        target = np.arange(100)
+        indices = np.arange(0, end, step, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+        # reverse
+        indices = indices[::-1]
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize(
+        "case", [[2, 1, 2, 0], [2, 2, 1, 0], [0, 1, 2, 1], [-2, 0, 2], [2, 0, -2]]
+    )
+    def test_maybe_indices_to_slice_left_edge_not_slice(self, case):
+        # not slice
+        target = np.arange(100)
+        indices = np.array(case, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize("start", [0, 2, 5, 20, 97, 98])
+    @pytest.mark.parametrize("step", [1, 2, 4])
+    def test_maybe_indices_to_slice_right_edge(self, start, step):
+        target = np.arange(100)
+
+        # slice
+        indices = np.arange(start, 99, step, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+        # reverse
+        indices = indices[::-1]
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    def test_maybe_indices_to_slice_right_edge_not_slice(self):
+        # not slice
+        target = np.arange(100)
+        indices = np.array([97, 98, 99, 100], dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+
+        msg = "index 100 is out of bounds for axis (0|1) with size 100"
+
+        with pytest.raises(IndexError, match=msg):
+            target[indices]
+        with pytest.raises(IndexError, match=msg):
+            target[maybe_slice]
+
+        indices = np.array([100, 99, 98, 97], dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+
+        with pytest.raises(IndexError, match=msg):
+            target[indices]
+        with pytest.raises(IndexError, match=msg):
+            target[maybe_slice]
+
+    @pytest.mark.parametrize(
+        "case", [[99, 97, 99, 96], [99, 99, 98, 97], [98, 98, 97, 96]]
+    )
+    def test_maybe_indices_to_slice_right_edge_cases(self, case):
+        target = np.arange(100)
+        indices = np.array(case, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize("step", [1, 2, 4, 5, 8, 9])
+    def test_maybe_indices_to_slice_both_edges(self, step):
+        target = np.arange(10)
+
+        # slice
+        indices = np.arange(0, 9, step, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+        # reverse
+        indices = indices[::-1]
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize("case", [[4, 2, 0, -2], [2, 2, 1, 0], [0, 1, 2, 1]])
+    def test_maybe_indices_to_slice_both_edges_not_slice(self, case):
+        # not slice
+        target = np.arange(10)
+        indices = np.array(case, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize("start, end", [(2, 10), (5, 25), (65, 97)])
+    @pytest.mark.parametrize("step", [1, 2, 4, 20])
+    def test_maybe_indices_to_slice_middle(self, start, end, step):
+        target = np.arange(100)
+
+        # slice
+        indices = np.arange(start, end, step, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+        # reverse
+        indices = indices[::-1]
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    @pytest.mark.parametrize(
+        "case", [[14, 12, 10, 12], [12, 12, 11, 10], [10, 11, 12, 11]]
+    )
+    def test_maybe_indices_to_slice_middle_not_slice(self, case):
+        # not slice
+        target = np.arange(100)
+        indices = np.array(case, dtype=np.intp)
+        maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
+
+        assert not isinstance(maybe_slice, slice)
+        tm.assert_numpy_array_equal(maybe_slice, indices)
+        tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
+
+    def test_maybe_booleans_to_slice(self):
+        arr = np.array([0, 0, 1, 1, 1, 0, 1], dtype=np.uint8)
+        result = lib.maybe_booleans_to_slice(arr)
+        assert result.dtype == np.bool_
+
+        result = lib.maybe_booleans_to_slice(arr[:0])
+        assert result == slice(0, 0)
+
+    def test_get_reverse_indexer(self):
+        indexer = np.array([-1, -1, 1, 2, 0, -1, 3, 4], dtype=np.intp)
+        result = lib.get_reverse_indexer(indexer, 5)
+        expected = np.array([4, 2, 3, 6, 7], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["int64", "int32"])
+    def test_is_range_indexer(self, dtype):
+        # GH#50592
+        left = np.arange(0, 100, dtype=dtype)
+        assert lib.is_range_indexer(left, 100)
+
+    @pytest.mark.skipif(
+        not IS64,
+        reason="2**31 is too big for Py_ssize_t on 32-bit. "
+        "It doesn't matter though since you cannot create an array that long on 32-bit",
+    )
+    @pytest.mark.parametrize("dtype", ["int64", "int32"])
+    def test_is_range_indexer_big_n(self, dtype):
+        # GH53616
+        left = np.arange(0, 100, dtype=dtype)
+
+        assert not lib.is_range_indexer(left, 2**31)
+
+    @pytest.mark.parametrize("dtype", ["int64", "int32"])
+    def test_is_range_indexer_not_equal(self, dtype):
+        # GH#50592
+        left = np.array([1, 2], dtype=dtype)
+        assert not lib.is_range_indexer(left, 2)
+
+    @pytest.mark.parametrize("dtype", ["int64", "int32"])
+    def test_is_range_indexer_not_equal_shape(self, dtype):
+        # GH#50592
+        left = np.array([0, 1, 2], dtype=dtype)
+        assert not lib.is_range_indexer(left, 2)
+
+
+def test_cache_readonly_preserve_docstrings():
+    # GH18197
+    assert Index.hasnans.__doc__ is not None
+
+
+def test_no_default_pickle():
+    # GH#40397
+    obj = tm.round_trip_pickle(lib.no_default)
+    assert obj is lib.no_default
+
+
+def test_ensure_string_array_copy():
+    # ensure the original array is not modified in case of copy=False with
+    # pickle-roundtripped object dtype array
+    # https://github.com/pandas-dev/pandas/issues/54654
+    arr = np.array(["a", None], dtype=object)
+    arr = pickle.loads(pickle.dumps(arr))
+    result = lib.ensure_string_array(arr, copy=False)
+    assert not np.shares_memory(arr, result)
+    assert arr[1] is None
+    assert result[1] is np.nan
diff --git a/pandas/pandas/tests/libs/test_libalgos.py b/pandas/pandas/tests/libs/test_libalgos.py
new file mode 100644
index 0000000000000..42d09c72aab2b
--- /dev/null
+++ b/pandas/pandas/tests/libs/test_libalgos.py
@@ -0,0 +1,162 @@
+from datetime import datetime
+from itertools import permutations
+
+import numpy as np
+
+from pandas._libs import algos as libalgos
+
+import pandas._testing as tm
+
+
+def test_ensure_platform_int():
+    arr = np.arange(100, dtype=np.intp)
+
+    result = libalgos.ensure_platform_int(arr)
+    assert result is arr
+
+
+def test_is_lexsorted():
+    failure = [
+        np.array(
+            ([3] * 32) + ([2] * 32) + ([1] * 32) + ([0] * 32),
+            dtype="int64",
+        ),
+        np.array(
+            list(range(31))[::-1] * 4,
+            dtype="int64",
+        ),
+    ]
+
+    assert not libalgos.is_lexsorted(failure)
+
+
+def test_groupsort_indexer():
+    a = np.random.default_rng(2).integers(0, 1000, 100).astype(np.intp)
+    b = np.random.default_rng(2).integers(0, 1000, 100).astype(np.intp)
+
+    result = libalgos.groupsort_indexer(a, 1000)[0]
+
+    # need to use a stable sort
+    # np.argsort returns int, groupsort_indexer
+    # always returns intp
+    expected = np.argsort(a, kind="mergesort")
+    expected = expected.astype(np.intp)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+    # compare with lexsort
+    # np.lexsort returns int, groupsort_indexer
+    # always returns intp
+    key = a * 1000 + b
+    result = libalgos.groupsort_indexer(key, 1000000)[0]
+    expected = np.lexsort((b, a))
+    expected = expected.astype(np.intp)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+class TestPadBackfill:
+    def test_backfill(self):
+        old = np.array([1, 5, 10], dtype=np.int64)
+        new = np.array(list(range(12)), dtype=np.int64)
+
+        filler = libalgos.backfill["int64_t"](old, new)
+
+        expect_filler = np.array([0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(filler, expect_filler)
+
+        # corner case
+        old = np.array([1, 4], dtype=np.int64)
+        new = np.array(list(range(5, 10)), dtype=np.int64)
+        filler = libalgos.backfill["int64_t"](old, new)
+
+        expect_filler = np.array([-1, -1, -1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(filler, expect_filler)
+
+    def test_pad(self):
+        old = np.array([1, 5, 10], dtype=np.int64)
+        new = np.array(list(range(12)), dtype=np.int64)
+
+        filler = libalgos.pad["int64_t"](old, new)
+
+        expect_filler = np.array([-1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(filler, expect_filler)
+
+        # corner case
+        old = np.array([5, 10], dtype=np.int64)
+        new = np.arange(5, dtype=np.int64)
+        filler = libalgos.pad["int64_t"](old, new)
+        expect_filler = np.array([-1, -1, -1, -1, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(filler, expect_filler)
+
+    def test_pad_backfill_object_segfault(self):
+        old = np.array([], dtype="O")
+        new = np.array([datetime(2010, 12, 31)], dtype="O")
+
+        result = libalgos.pad["object"](old, new)
+        expected = np.array([-1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = libalgos.pad["object"](new, old)
+        expected = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = libalgos.backfill["object"](old, new)
+        expected = np.array([-1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = libalgos.backfill["object"](new, old)
+        expected = np.array([], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestInfinity:
+    def test_infinity_sort(self):
+        # GH#13445
+        # numpy's argsort can be unhappy if something is less than
+        # itself.  Instead, let's give our infinities a self-consistent
+        # ordering, but outside the float extended real line.
+
+        Inf = libalgos.Infinity()
+        NegInf = libalgos.NegInfinity()
+
+        ref_nums = [NegInf, float("-inf"), -1e100, 0, 1e100, float("inf"), Inf]
+
+        assert all(Inf >= x for x in ref_nums)
+        assert all(Inf > x or x is Inf for x in ref_nums)
+        assert Inf >= Inf and Inf == Inf
+        assert not Inf < Inf and not Inf > Inf
+        assert libalgos.Infinity() == libalgos.Infinity()
+        assert not libalgos.Infinity() != libalgos.Infinity()
+
+        assert all(NegInf <= x for x in ref_nums)
+        assert all(NegInf < x or x is NegInf for x in ref_nums)
+        assert NegInf <= NegInf and NegInf == NegInf
+        assert not NegInf < NegInf and not NegInf > NegInf
+        assert libalgos.NegInfinity() == libalgos.NegInfinity()
+        assert not libalgos.NegInfinity() != libalgos.NegInfinity()
+
+        for perm in permutations(ref_nums):
+            assert sorted(perm) == ref_nums
+
+        # smoke tests
+        np.array([libalgos.Infinity()] * 32).argsort()
+        np.array([libalgos.NegInfinity()] * 32).argsort()
+
+    def test_infinity_against_nan(self):
+        Inf = libalgos.Infinity()
+        NegInf = libalgos.NegInfinity()
+
+        assert not Inf > np.nan
+        assert not Inf >= np.nan
+        assert not Inf < np.nan
+        assert not Inf <= np.nan
+        assert not Inf == np.nan
+        assert Inf != np.nan
+
+        assert not NegInf > np.nan
+        assert not NegInf >= np.nan
+        assert not NegInf < np.nan
+        assert not NegInf <= np.nan
+        assert not NegInf == np.nan
+        assert NegInf != np.nan
diff --git a/pandas/pandas/tests/plotting/__init__.py b/pandas/pandas/tests/plotting/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/plotting/common.py b/pandas/pandas/tests/plotting/common.py
new file mode 100644
index 0000000000000..d8c49d6d47f28
--- /dev/null
+++ b/pandas/pandas/tests/plotting/common.py
@@ -0,0 +1,579 @@
+"""
+Module consolidating common testing functions for checking plotting.
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas.core.dtypes.api import is_list_like
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from matplotlib.axes import Axes
+
+
+def _check_legend_labels(axes, labels=None, visible=True):
+    """
+    Check each axes has expected legend labels
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    labels : list-like
+        expected legend labels
+    visible : bool
+        expected legend visibility. labels are checked only when visible is
+        True
+    """
+    if visible and (labels is None):
+        raise ValueError("labels must be specified when visible is True")
+    axes = _flatten_visible(axes)
+    for ax in axes:
+        if visible:
+            assert ax.get_legend() is not None
+            _check_text_labels(ax.get_legend().get_texts(), labels)
+        else:
+            assert ax.get_legend() is None
+
+
+def _check_legend_marker(ax, expected_markers=None, visible=True):
+    """
+    Check ax has expected legend markers
+
+    Parameters
+    ----------
+    ax : matplotlib Axes object
+    expected_markers : list-like
+        expected legend markers
+    visible : bool
+        expected legend visibility. labels are checked only when visible is
+        True
+    """
+    if visible and (expected_markers is None):
+        raise ValueError("Markers must be specified when visible is True")
+    if visible:
+        handles, _ = ax.get_legend_handles_labels()
+        markers = [handle.get_marker() for handle in handles]
+        assert markers == expected_markers
+    else:
+        assert ax.get_legend() is None
+
+
+def _check_data(xp, rs):
+    """
+    Check each axes has identical lines
+
+    Parameters
+    ----------
+    xp : matplotlib Axes object
+    rs : matplotlib Axes object
+    """
+    xp_lines = xp.get_lines()
+    rs_lines = rs.get_lines()
+
+    assert len(xp_lines) == len(rs_lines)
+    for xpl, rsl in zip(xp_lines, rs_lines):
+        xpdata = xpl.get_xydata()
+        rsdata = rsl.get_xydata()
+        tm.assert_almost_equal(xpdata, rsdata)
+
+
+def _check_visible(collections, visible=True):
+    """
+    Check each artist is visible or not
+
+    Parameters
+    ----------
+    collections : matplotlib Artist or its list-like
+        target Artist or its list or collection
+    visible : bool
+        expected visibility
+    """
+    from matplotlib.collections import Collection
+
+    if not isinstance(collections, Collection) and not is_list_like(collections):
+        collections = [collections]
+
+    for patch in collections:
+        assert patch.get_visible() == visible
+
+
+def _check_patches_all_filled(axes: Axes | Sequence[Axes], filled: bool = True) -> None:
+    """
+    Check for each artist whether it is filled or not
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    filled : bool
+        expected filling
+    """
+
+    axes = _flatten_visible(axes)
+    for ax in axes:
+        for patch in ax.patches:
+            assert patch.fill == filled
+
+
+def _get_colors_mapped(series, colors):
+    unique = series.unique()
+    # unique and colors length can be differed
+    # depending on slice value
+    mapped = dict(zip(unique, colors))
+    return [mapped[v] for v in series.values]
+
+
+def _check_colors(collections, linecolors=None, facecolors=None, mapping=None):
+    """
+    Check each artist has expected line colors and face colors
+
+    Parameters
+    ----------
+    collections : list-like
+        list or collection of target artist
+    linecolors : list-like which has the same length as collections
+        list of expected line colors
+    facecolors : list-like which has the same length as collections
+        list of expected face colors
+    mapping : Series
+        Series used for color grouping key
+        used for andrew_curves, parallel_coordinates, radviz test
+    """
+    from matplotlib import colors
+    from matplotlib.collections import (
+        Collection,
+        LineCollection,
+        PolyCollection,
+    )
+    from matplotlib.lines import Line2D
+
+    conv = colors.ColorConverter
+    if linecolors is not None:
+        if mapping is not None:
+            linecolors = _get_colors_mapped(mapping, linecolors)
+            linecolors = linecolors[: len(collections)]
+
+        assert len(collections) == len(linecolors)
+        for patch, color in zip(collections, linecolors):
+            if isinstance(patch, Line2D):
+                result = patch.get_color()
+                # Line2D may contains string color expression
+                result = conv.to_rgba(result)
+            elif isinstance(patch, (PolyCollection, LineCollection)):
+                result = tuple(patch.get_edgecolor()[0])
+            else:
+                result = patch.get_edgecolor()
+
+            expected = conv.to_rgba(color)
+            assert result == expected
+
+    if facecolors is not None:
+        if mapping is not None:
+            facecolors = _get_colors_mapped(mapping, facecolors)
+            facecolors = facecolors[: len(collections)]
+
+        assert len(collections) == len(facecolors)
+        for patch, color in zip(collections, facecolors):
+            if isinstance(patch, Collection):
+                # returned as list of np.array
+                result = patch.get_facecolor()[0]
+            else:
+                result = patch.get_facecolor()
+
+            if isinstance(result, np.ndarray):
+                result = tuple(result)
+
+            expected = conv.to_rgba(color)
+            assert result == expected
+
+
+def _check_text_labels(texts, expected):
+    """
+    Check each text has expected labels
+
+    Parameters
+    ----------
+    texts : matplotlib Text object, or its list-like
+        target text, or its list
+    expected : str or list-like which has the same length as texts
+        expected text label, or its list
+    """
+    if not is_list_like(texts):
+        assert texts.get_text() == expected
+    else:
+        labels = [t.get_text() for t in texts]
+        assert len(labels) == len(expected)
+        for label, e in zip(labels, expected):
+            assert label == e
+
+
+def _check_ticks_props(axes, xlabelsize=None, xrot=None, ylabelsize=None, yrot=None):
+    """
+    Check each axes has expected tick properties
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    xlabelsize : number
+        expected xticks font size
+    xrot : number
+        expected xticks rotation
+    ylabelsize : number
+        expected yticks font size
+    yrot : number
+        expected yticks rotation
+    """
+    from matplotlib.ticker import NullFormatter
+
+    axes = _flatten_visible(axes)
+    for ax in axes:
+        if xlabelsize is not None or xrot is not None:
+            if isinstance(ax.xaxis.get_minor_formatter(), NullFormatter):
+                # If minor ticks has NullFormatter, rot / fontsize are not
+                # retained
+                labels = ax.get_xticklabels()
+            else:
+                labels = ax.get_xticklabels() + ax.get_xticklabels(minor=True)
+
+            for label in labels:
+                if xlabelsize is not None:
+                    tm.assert_almost_equal(label.get_fontsize(), xlabelsize)
+                if xrot is not None:
+                    tm.assert_almost_equal(label.get_rotation(), xrot)
+
+        if ylabelsize is not None or yrot is not None:
+            if isinstance(ax.yaxis.get_minor_formatter(), NullFormatter):
+                labels = ax.get_yticklabels()
+            else:
+                labels = ax.get_yticklabels() + ax.get_yticklabels(minor=True)
+
+            for label in labels:
+                if ylabelsize is not None:
+                    tm.assert_almost_equal(label.get_fontsize(), ylabelsize)
+                if yrot is not None:
+                    tm.assert_almost_equal(label.get_rotation(), yrot)
+
+
+def _check_ax_scales(axes, xaxis="linear", yaxis="linear"):
+    """
+    Check each axes has expected scales
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    xaxis : {'linear', 'log'}
+        expected xaxis scale
+    yaxis : {'linear', 'log'}
+        expected yaxis scale
+    """
+    axes = _flatten_visible(axes)
+    for ax in axes:
+        assert ax.xaxis.get_scale() == xaxis
+        assert ax.yaxis.get_scale() == yaxis
+
+
+def _check_axes_shape(axes, axes_num=None, layout=None, figsize=None):
+    """
+    Check expected number of axes is drawn in expected layout
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    axes_num : number
+        expected number of axes. Unnecessary axes should be set to
+        invisible.
+    layout : tuple
+        expected layout, (expected number of rows , columns)
+    figsize : tuple
+        expected figsize. default is matplotlib default
+    """
+    from pandas.plotting._matplotlib.tools import flatten_axes
+
+    if figsize is None:
+        figsize = (6.4, 4.8)
+    visible_axes = _flatten_visible(axes)
+
+    if axes_num is not None:
+        assert len(visible_axes) == axes_num
+        for ax in visible_axes:
+            # check something drawn on visible axes
+            assert len(ax.get_children()) > 0
+
+    if layout is not None:
+        x_set = set()
+        y_set = set()
+        for ax in flatten_axes(axes):
+            # check axes coordinates to estimate layout
+            points = ax.get_position().get_points()
+            x_set.add(points[0][0])
+            y_set.add(points[0][1])
+        result = (len(y_set), len(x_set))
+        assert result == layout
+
+    tm.assert_numpy_array_equal(
+        visible_axes[0].figure.get_size_inches(),
+        np.array(figsize, dtype=np.float64),
+    )
+
+
+def _flatten_visible(axes: Axes | Sequence[Axes]) -> Sequence[Axes]:
+    """
+    Flatten axes, and filter only visible
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+
+    """
+    from pandas.plotting._matplotlib.tools import flatten_axes
+
+    axes_ndarray = flatten_axes(axes)
+    axes = [ax for ax in axes_ndarray if ax.get_visible()]
+    return axes
+
+
+def _check_has_errorbars(axes, xerr=0, yerr=0):
+    """
+    Check axes has expected number of errorbars
+
+    Parameters
+    ----------
+    axes : matplotlib Axes object, or its list-like
+    xerr : number
+        expected number of x errorbar
+    yerr : number
+        expected number of y errorbar
+    """
+    axes = _flatten_visible(axes)
+    for ax in axes:
+        containers = ax.containers
+        xerr_count = 0
+        yerr_count = 0
+        for c in containers:
+            has_xerr = getattr(c, "has_xerr", False)
+            has_yerr = getattr(c, "has_yerr", False)
+            if has_xerr:
+                xerr_count += 1
+            if has_yerr:
+                yerr_count += 1
+        assert xerr == xerr_count
+        assert yerr == yerr_count
+
+
+def _check_box_return_type(
+    returned, return_type, expected_keys=None, check_ax_title=True
+):
+    """
+    Check box returned type is correct
+
+    Parameters
+    ----------
+    returned : object to be tested, returned from boxplot
+    return_type : str
+        return_type passed to boxplot
+    expected_keys : list-like, optional
+        group labels in subplot case. If not passed,
+        the function checks assuming boxplot uses single ax
+    check_ax_title : bool
+        Whether to check the ax.title is the same as expected_key
+        Intended to be checked by calling from ``boxplot``.
+        Normal ``plot`` doesn't attach ``ax.title``, it must be disabled.
+    """
+    from matplotlib.axes import Axes
+
+    types = {"dict": dict, "axes": Axes, "both": tuple}
+    if expected_keys is None:
+        # should be fixed when the returning default is changed
+        if return_type is None:
+            return_type = "dict"
+
+        assert isinstance(returned, types[return_type])
+        if return_type == "both":
+            assert isinstance(returned.ax, Axes)
+            assert isinstance(returned.lines, dict)
+    else:
+        # should be fixed when the returning default is changed
+        if return_type is None:
+            for r in _flatten_visible(returned):
+                assert isinstance(r, Axes)
+            return
+
+        assert isinstance(returned, Series)
+
+        assert sorted(returned.keys()) == sorted(expected_keys)
+        for key, value in returned.items():
+            assert isinstance(value, types[return_type])
+            # check returned dict has correct mapping
+            if return_type == "axes":
+                if check_ax_title:
+                    assert value.get_title() == key
+            elif return_type == "both":
+                if check_ax_title:
+                    assert value.ax.get_title() == key
+                assert isinstance(value.ax, Axes)
+                assert isinstance(value.lines, dict)
+            elif return_type == "dict":
+                line = value["medians"][0]
+                axes = line.axes
+                if check_ax_title:
+                    assert axes.get_title() == key
+            else:
+                raise AssertionError
+
+
+def _check_grid_settings(obj, kinds, kws=None):
+    # Make sure plot defaults to rcParams['axes.grid'] setting, GH 9792
+
+    import matplotlib as mpl
+
+    def is_grid_on():
+        xticks = mpl.pyplot.gca().xaxis.get_major_ticks()
+        yticks = mpl.pyplot.gca().yaxis.get_major_ticks()
+        xoff = all(not g.gridline.get_visible() for g in xticks)
+        yoff = all(not g.gridline.get_visible() for g in yticks)
+
+        return not (xoff and yoff)
+
+    if kws is None:
+        kws = {}
+    spndx = 1
+    for kind in kinds:
+        mpl.pyplot.subplot(1, 4 * len(kinds), spndx)
+        spndx += 1
+        mpl.rc("axes", grid=False)
+        obj.plot(kind=kind, **kws)
+        assert not is_grid_on()
+        mpl.pyplot.clf()
+
+        mpl.pyplot.subplot(1, 4 * len(kinds), spndx)
+        spndx += 1
+        mpl.rc("axes", grid=True)
+        obj.plot(kind=kind, grid=False, **kws)
+        assert not is_grid_on()
+        mpl.pyplot.clf()
+
+        if kind not in ["pie", "hexbin", "scatter"]:
+            mpl.pyplot.subplot(1, 4 * len(kinds), spndx)
+            spndx += 1
+            mpl.rc("axes", grid=True)
+            obj.plot(kind=kind, **kws)
+            assert is_grid_on()
+            mpl.pyplot.clf()
+
+            mpl.pyplot.subplot(1, 4 * len(kinds), spndx)
+            spndx += 1
+            mpl.rc("axes", grid=False)
+            obj.plot(kind=kind, grid=True, **kws)
+            assert is_grid_on()
+            mpl.pyplot.clf()
+
+
+def _unpack_cycler(rcParams, field="color"):
+    """
+    Auxiliary function for correctly unpacking cycler after MPL >= 1.5
+    """
+    return [v[field] for v in rcParams["axes.prop_cycle"]]
+
+
+def get_x_axis(ax):
+    return ax._shared_axes["x"]
+
+
+def get_y_axis(ax):
+    return ax._shared_axes["y"]
+
+
+def assert_is_valid_plot_return_object(objs) -> None:
+    from matplotlib.artist import Artist
+    from matplotlib.axes import Axes
+
+    if isinstance(objs, (Series, np.ndarray)):
+        if isinstance(objs, Series):
+            objs = objs._values
+        for el in objs.reshape(-1):
+            msg = (
+                "one of 'objs' is not a matplotlib Axes instance, "
+                f"type encountered {type(el).__name__!r}"
+            )
+            assert isinstance(el, (Axes, dict)), msg
+    else:
+        msg = (
+            "objs is neither an ndarray of Artist instances nor a single "
+            "ArtistArtist instance, tuple, or dict, 'objs' is a "
+            f"{type(objs).__name__!r}"
+        )
+        assert isinstance(objs, (Artist, tuple, dict)), msg
+
+
+def _check_plot_works(f, default_axes=False, **kwargs):
+    """
+    Create plot and ensure that plot return object is valid.
+
+    Parameters
+    ----------
+    f : func
+        Plotting function.
+    default_axes : bool, optional
+        If False (default):
+            - If `ax` not in `kwargs`, then create subplot(211) and plot there
+            - Create new subplot(212) and plot there as well
+            - Mind special corner case for bootstrap_plot (see `_gen_two_subplots`)
+        If True:
+            - Simply run plotting function with kwargs provided
+            - All required axes instances will be created automatically
+            - It is recommended to use it when the plotting function
+            creates multiple axes itself. It helps avoid warnings like
+            'UserWarning: To output multiple subplots,
+            the figure containing the passed axes is being cleared'
+    **kwargs
+        Keyword arguments passed to the plotting function.
+
+    Returns
+    -------
+    Plot object returned by the last plotting.
+    """
+    import matplotlib.pyplot as plt
+
+    if default_axes:
+        gen_plots = _gen_default_plot
+    else:
+        gen_plots = _gen_two_subplots
+
+    ret = None
+    fig = kwargs.get("figure", plt.gcf())
+    fig.clf()
+
+    for ret in gen_plots(f, fig, **kwargs):
+        assert_is_valid_plot_return_object(ret)
+
+    return ret
+
+
+def _gen_default_plot(f, fig, **kwargs):
+    """
+    Create plot in a default way.
+    """
+    yield f(**kwargs)
+
+
+def _gen_two_subplots(f, fig, **kwargs):
+    """
+    Create plot on two subplots forcefully created.
+    """
+    if "ax" not in kwargs:
+        fig.add_subplot(211)
+    yield f(**kwargs)
+
+    if f is pd.plotting.bootstrap_plot:
+        assert "ax" not in kwargs
+    else:
+        kwargs["ax"] = fig.add_subplot(212)
+    yield f(**kwargs)
diff --git a/pandas/pandas/tests/plotting/conftest.py b/pandas/pandas/tests/plotting/conftest.py
new file mode 100644
index 0000000000000..eb5a1f1f6382e
--- /dev/null
+++ b/pandas/pandas/tests/plotting/conftest.py
@@ -0,0 +1,39 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    to_datetime,
+)
+
+
+@pytest.fixture(autouse=True)
+def autouse_mpl_cleanup(mpl_cleanup):
+    pass
+
+
+@pytest.fixture
+def hist_df():
+    n = 50
+    rng = np.random.default_rng(10)
+    gender = rng.choice(["Male", "Female"], size=n)
+    classroom = rng.choice(["A", "B", "C"], size=n)
+
+    hist_df = DataFrame(
+        {
+            "gender": gender,
+            "classroom": classroom,
+            "height": rng.normal(66, 4, size=n),
+            "weight": rng.normal(161, 32, size=n),
+            "category": rng.integers(4, size=n),
+            "datetime": to_datetime(
+                rng.integers(
+                    812419200000000000,
+                    819331200000000000,
+                    size=n,
+                    dtype=np.int64,
+                )
+            ),
+        }
+    )
+    return hist_df
diff --git a/pandas/pandas/tests/plotting/frame/__init__.py b/pandas/pandas/tests/plotting/frame/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/plotting/frame/test_frame.py b/pandas/pandas/tests/plotting/frame/test_frame.py
new file mode 100644
index 0000000000000..d18f098267599
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_frame.py
@@ -0,0 +1,2633 @@
+"""Test cases for DataFrame.plot"""
+
+from datetime import (
+    date,
+    datetime,
+)
+import gc
+import itertools
+import re
+import string
+import weakref
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.api import is_list_like
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    Series,
+    bdate_range,
+    date_range,
+    option_context,
+    plotting,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_ax_scales,
+    _check_axes_shape,
+    _check_box_return_type,
+    _check_colors,
+    _check_data,
+    _check_grid_settings,
+    _check_has_errorbars,
+    _check_legend_labels,
+    _check_plot_works,
+    _check_text_labels,
+    _check_ticks_props,
+    _check_visible,
+    get_y_axis,
+)
+from pandas.util.version import Version
+
+from pandas.io.formats.printing import pprint_thing
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+
+class TestDataFramePlots:
+    @pytest.mark.slow
+    def test_plot(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        _check_plot_works(df.plot, grid=False)
+
+    @pytest.mark.slow
+    def test_plot_subplots(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        # _check_plot_works adds an ax so use default_axes=True to avoid warning
+        axes = _check_plot_works(df.plot, default_axes=True, subplots=True)
+        _check_axes_shape(axes, axes_num=4, layout=(4, 1))
+
+    @pytest.mark.slow
+    def test_plot_subplots_negative_layout(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        axes = _check_plot_works(
+            df.plot,
+            default_axes=True,
+            subplots=True,
+            layout=(-1, 2),
+        )
+        _check_axes_shape(axes, axes_num=4, layout=(2, 2))
+
+    @pytest.mark.slow
+    def test_plot_subplots_use_index(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        axes = _check_plot_works(
+            df.plot,
+            default_axes=True,
+            subplots=True,
+            use_index=False,
+        )
+        _check_ticks_props(axes, xrot=0)
+        _check_axes_shape(axes, axes_num=4, layout=(4, 1))
+
+    @pytest.mark.xfail(reason="Api changed in 3.6.0")
+    @pytest.mark.slow
+    def test_plot_invalid_arg(self):
+        df = DataFrame({"x": [1, 2], "y": [3, 4]})
+        msg = "'Line2D' object has no property 'blarg'"
+        with pytest.raises(AttributeError, match=msg):
+            df.plot.line(blarg=True)
+
+    @pytest.mark.slow
+    def test_plot_tick_props(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+
+        ax = _check_plot_works(df.plot, use_index=True)
+        _check_ticks_props(ax, xrot=0)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"yticks": [1, 5, 10]},
+            {"xticks": [1, 5, 10]},
+            {"ylim": (-100, 100), "xlim": (-100, 100)},
+            {"default_axes": True, "subplots": True, "title": "blah"},
+        ],
+    )
+    def test_plot_other_args(self, kwargs):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        _check_plot_works(df.plot, **kwargs)
+
+    @pytest.mark.slow
+    def test_plot_visible_ax(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        # We have to redo it here because _check_plot_works does two plots,
+        # once without an ax kwarg and once with an ax kwarg and the new sharex
+        # behaviour does not remove the visibility of the latter axis (as ax is
+        # present).  see: https://github.com/pandas-dev/pandas/issues/9737
+
+        axes = df.plot(subplots=True, title="blah")
+        _check_axes_shape(axes, axes_num=3, layout=(3, 1))
+        for ax in axes[:2]:
+            _check_visible(ax.xaxis)  # xaxis must be visible for grid
+            _check_visible(ax.get_xticklabels(), visible=False)
+            _check_visible(ax.get_xticklabels(minor=True), visible=False)
+            _check_visible([ax.xaxis.get_label()], visible=False)
+        for ax in [axes[2]]:
+            _check_visible(ax.xaxis)
+            _check_visible(ax.get_xticklabels())
+            _check_visible([ax.xaxis.get_label()])
+            _check_ticks_props(ax, xrot=0)
+
+    @pytest.mark.slow
+    def test_plot_title(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        _check_plot_works(df.plot, title="blah")
+
+    @pytest.mark.slow
+    def test_plot_multiindex(self):
+        tuples = zip(string.ascii_letters[:10], range(10))
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=MultiIndex.from_tuples(tuples),
+        )
+        ax = _check_plot_works(df.plot, use_index=True)
+        _check_ticks_props(ax, xrot=0)
+
+    @pytest.mark.slow
+    def test_plot_multiindex_unicode(self):
+        # unicode
+        index = MultiIndex.from_tuples(
+            [
+                ("\u03b1", 0),
+                ("\u03b1", 1),
+                ("\u03b2", 2),
+                ("\u03b2", 3),
+                ("\u03b3", 4),
+                ("\u03b3", 5),
+                ("\u03b4", 6),
+                ("\u03b4", 7),
+            ],
+            names=["i0", "i1"],
+        )
+        columns = MultiIndex.from_tuples(
+            [("bar", "\u0394"), ("bar", "\u0395")], names=["c0", "c1"]
+        )
+        df = DataFrame(
+            np.random.default_rng(2).integers(0, 10, (8, 2)),
+            columns=columns,
+            index=index,
+        )
+        _check_plot_works(df.plot, title="\u03a3")
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("layout", [None, (-1, 1)])
+    def test_plot_single_column_bar(self, layout):
+        # GH 6951
+        # Test with single column
+        df = DataFrame({"x": np.random.default_rng(2).random(10)})
+        axes = _check_plot_works(df.plot.bar, subplots=True, layout=layout)
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+
+    @pytest.mark.slow
+    def test_plot_passed_ax(self):
+        # When ax is supplied and required number of axes is 1,
+        # passed ax should be used:
+        df = DataFrame({"x": np.random.default_rng(2).random(10)})
+        _, ax = mpl.pyplot.subplots()
+        axes = df.plot.bar(subplots=True, ax=ax)
+        assert len(axes) == 1
+        result = ax.axes
+        assert result is axes[0]
+
+    @pytest.mark.parametrize(
+        "cols, x, y",
+        [
+            [list("ABCDE"), "A", "B"],
+            [["A", "B"], "A", "B"],
+            [["C", "A"], "C", "A"],
+            [["A", "C"], "A", "C"],
+            [["B", "C"], "B", "C"],
+            [["A", "D"], "A", "D"],
+            [["A", "E"], "A", "E"],
+        ],
+    )
+    def test_nullable_int_plot(self, cols, x, y):
+        # GH 32073
+        dates = ["2008", "2009", None, "2011", "2012"]
+        df = DataFrame(
+            {
+                "A": [1, 2, 3, 4, 5],
+                "B": [1, 2, 3, 4, 5],
+                "C": np.array([7, 5, np.nan, 3, 2], dtype=object),
+                "D": pd.to_datetime(dates, format="%Y").view("i8"),
+                "E": pd.to_datetime(dates, format="%Y", utc=True).view("i8"),
+            }
+        )
+
+        _check_plot_works(df[cols].plot, x=x, y=y)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("plot", ["line", "bar", "hist", "pie"])
+    def test_integer_array_plot_series(self, plot):
+        # GH 25587
+        arr = pd.array([1, 2, 3, 4], dtype="UInt32")
+
+        s = Series(arr)
+        _check_plot_works(getattr(s.plot, plot))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "plot, kwargs",
+        [
+            ["line", {}],
+            ["bar", {}],
+            ["hist", {}],
+            ["pie", {"y": "y"}],
+            ["scatter", {"x": "x", "y": "y"}],
+            ["hexbin", {"x": "x", "y": "y"}],
+        ],
+    )
+    def test_integer_array_plot_df(self, plot, kwargs):
+        # GH 25587
+        arr = pd.array([1, 2, 3, 4], dtype="UInt32")
+        df = DataFrame({"x": arr, "y": arr})
+        _check_plot_works(getattr(df.plot, plot), **kwargs)
+
+    def test_nonnumeric_exclude(self):
+        df = DataFrame({"A": ["x", "y", "z"], "B": [1, 2, 3]})
+        ax = df.plot()
+        assert len(ax.get_lines()) == 1  # B was plotted
+
+    def test_implicit_label(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=["a", "b", "c"]
+        )
+        ax = df.plot(x="a", y="b")
+        _check_text_labels(ax.xaxis.get_label(), "a")
+
+    def test_donot_overwrite_index_name(self):
+        # GH 8494
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((2, 2)), columns=["a", "b"]
+        )
+        df.index.name = "NAME"
+        df.plot(y="b", label="LABEL")
+        assert df.index.name == "NAME"
+
+    def test_plot_xy(self):
+        # columns.inferred_type == 'string'
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        _check_data(df.plot(x=0, y=1), df.set_index("A")["B"].plot())
+        _check_data(df.plot(x=0), df.set_index("A").plot())
+        _check_data(df.plot(y=0), df.B.plot())
+        _check_data(df.plot(x="A", y="B"), df.set_index("A").B.plot())
+        _check_data(df.plot(x="A"), df.set_index("A").plot())
+        _check_data(df.plot(y="B"), df.B.plot())
+
+    def test_plot_xy_int_cols(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        # columns.inferred_type == 'integer'
+        df.columns = np.arange(1, len(df.columns) + 1)
+        _check_data(df.plot(x=1, y=2), df.set_index(1)[2].plot())
+        _check_data(df.plot(x=1), df.set_index(1).plot())
+        _check_data(df.plot(y=1), df[1].plot())
+
+    def test_plot_xy_figsize_and_title(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=5, freq="B"),
+        )
+        # figsize and title
+        ax = df.plot(x=1, y=2, title="Test", figsize=(16, 8))
+        _check_text_labels(ax.title, "Test")
+        _check_axes_shape(ax, axes_num=1, layout=(1, 1), figsize=(16.0, 8.0))
+
+        # columns.inferred_type == 'mixed'
+        # TODO add MultiIndex test
+
+    @pytest.mark.parametrize(
+        "input_log, expected_log", [(True, "log"), ("sym", "symlog")]
+    )
+    def test_logscales(self, input_log, expected_log):
+        df = DataFrame({"a": np.arange(100)}, index=np.arange(100))
+
+        ax = df.plot(logy=input_log)
+        _check_ax_scales(ax, yaxis=expected_log)
+        assert ax.get_yscale() == expected_log
+
+        ax = df.plot(logx=input_log)
+        _check_ax_scales(ax, xaxis=expected_log)
+        assert ax.get_xscale() == expected_log
+
+        ax = df.plot(loglog=input_log)
+        _check_ax_scales(ax, xaxis=expected_log, yaxis=expected_log)
+        assert ax.get_xscale() == expected_log
+        assert ax.get_yscale() == expected_log
+
+    @pytest.mark.parametrize("input_param", ["logx", "logy", "loglog"])
+    def test_invalid_logscale(self, input_param):
+        # GH: 24867
+        df = DataFrame({"a": np.arange(100)}, index=np.arange(100))
+
+        msg = f"keyword '{input_param}' should be bool, None, or 'sym', not 'sm'"
+        with pytest.raises(ValueError, match=msg):
+            df.plot(**{input_param: "sm"})
+
+        msg = f"PiePlot ignores the '{input_param}' keyword"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            df.plot.pie(subplots=True, **{input_param: True})
+
+    def test_xcompat(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        ax = df.plot(x_compat=True)
+        lines = ax.get_lines()
+        assert not isinstance(lines[0].get_xdata(), PeriodIndex)
+        _check_ticks_props(ax, xrot=30)
+
+    def test_xcompat_plot_params(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        plotting.plot_params["xaxis.compat"] = True
+        ax = df.plot()
+        lines = ax.get_lines()
+        assert not isinstance(lines[0].get_xdata(), PeriodIndex)
+        _check_ticks_props(ax, xrot=30)
+
+    def test_xcompat_plot_params_x_compat(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        plotting.plot_params["x_compat"] = False
+
+        ax = df.plot()
+        lines = ax.get_lines()
+        assert not isinstance(lines[0].get_xdata(), PeriodIndex)
+        msg = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert isinstance(PeriodIndex(lines[0].get_xdata()), PeriodIndex)
+
+    def test_xcompat_plot_params_context_manager(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        # useful if you're plotting a bunch together
+        with plotting.plot_params.use("x_compat", True):
+            ax = df.plot()
+            lines = ax.get_lines()
+            assert not isinstance(lines[0].get_xdata(), PeriodIndex)
+            _check_ticks_props(ax, xrot=30)
+
+    def test_xcompat_plot_period(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        ax = df.plot()
+        lines = ax.get_lines()
+        assert not isinstance(lines[0].get_xdata(), PeriodIndex)
+        msg = r"PeriodDtype\[B\] is deprecated "
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert isinstance(PeriodIndex(lines[0].get_xdata()), PeriodIndex)
+        _check_ticks_props(ax, xrot=0)
+
+    def test_period_compat(self):
+        # GH 9012
+        # period-array conversions
+        df = DataFrame(
+            np.random.default_rng(2).random((21, 2)),
+            index=bdate_range(datetime(2000, 1, 1), datetime(2000, 1, 31)),
+            columns=["a", "b"],
+        )
+
+        df.plot()
+        mpl.pyplot.axhline(y=0)
+
+    @pytest.mark.parametrize("index_dtype", [np.int64, np.float64])
+    def test_unsorted_index(self, index_dtype):
+        df = DataFrame(
+            {"y": np.arange(100)},
+            index=Index(np.arange(99, -1, -1), dtype=index_dtype),
+            dtype=np.int64,
+        )
+        ax = df.plot()
+        lines = ax.get_lines()[0]
+        rs = lines.get_xydata()
+        rs = Series(rs[:, 1], rs[:, 0], dtype=np.int64, name="y")
+        tm.assert_series_equal(rs, df.y, check_index_type=False)
+
+    @pytest.mark.parametrize(
+        "df",
+        [
+            DataFrame({"y": [0.0, 1.0, 2.0, 3.0]}, index=[1.0, 0.0, 3.0, 2.0]),
+            DataFrame(
+                {"y": [0.0, 1.0, np.nan, 3.0, 4.0, 5.0, 6.0]},
+                index=[1.0, 0.0, 3.0, 2.0, np.nan, 3.0, 2.0],
+            ),
+        ],
+    )
+    def test_unsorted_index_lims(self, df):
+        ax = df.plot()
+        xmin, xmax = ax.get_xlim()
+        lines = ax.get_lines()
+        assert xmin <= np.nanmin(lines[0].get_data()[0])
+        assert xmax >= np.nanmax(lines[0].get_data()[0])
+
+    def test_unsorted_index_lims_x_y(self):
+        df = DataFrame({"y": [0.0, 1.0, 2.0, 3.0], "z": [91.0, 90.0, 93.0, 92.0]})
+        ax = df.plot(x="z", y="y")
+        xmin, xmax = ax.get_xlim()
+        lines = ax.get_lines()
+        assert xmin <= np.nanmin(lines[0].get_data()[0])
+        assert xmax >= np.nanmax(lines[0].get_data()[0])
+
+    def test_negative_log(self):
+        df = -DataFrame(
+            np.random.default_rng(2).random((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+        msg = "Log-y scales are not supported in area plot"
+        with pytest.raises(ValueError, match=msg):
+            df.plot.area(logy=True)
+        with pytest.raises(ValueError, match=msg):
+            df.plot.area(loglog=True)
+
+    def _compare_stacked_y_cood(self, normal_lines, stacked_lines):
+        base = np.zeros(len(normal_lines[0].get_data()[1]))
+        for nl, sl in zip(normal_lines, stacked_lines):
+            base += nl.get_data()[1]  # get y coordinates
+            sy = sl.get_data()[1]
+            tm.assert_numpy_array_equal(base, sy)
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    @pytest.mark.parametrize("mult", [1, -1])
+    def test_line_area_stacked(self, kind, mult):
+        df = mult * DataFrame(
+            np.random.default_rng(2).random((6, 4)), columns=["w", "x", "y", "z"]
+        )
+
+        ax1 = _check_plot_works(df.plot, kind=kind, stacked=False)
+        ax2 = _check_plot_works(df.plot, kind=kind, stacked=True)
+        self._compare_stacked_y_cood(ax1.lines, ax2.lines)
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_line_area_stacked_sep_df(self, kind):
+        # each column has either positive or negative value
+        sep_df = DataFrame(
+            {
+                "w": np.random.default_rng(2).random(6),
+                "x": np.random.default_rng(2).random(6),
+                "y": -np.random.default_rng(2).random(6),
+                "z": -np.random.default_rng(2).random(6),
+            }
+        )
+        ax1 = _check_plot_works(sep_df.plot, kind=kind, stacked=False)
+        ax2 = _check_plot_works(sep_df.plot, kind=kind, stacked=True)
+        self._compare_stacked_y_cood(ax1.lines[:2], ax2.lines[:2])
+        self._compare_stacked_y_cood(ax1.lines[2:], ax2.lines[2:])
+
+    def test_line_area_stacked_mixed(self):
+        mixed_df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["w", "x", "y", "z"],
+        )
+        _check_plot_works(mixed_df.plot, stacked=False)
+
+        msg = (
+            "When stacked is True, each column must be either all positive or "
+            "all negative. Column 'w' contains both positive and negative "
+            "values"
+        )
+        with pytest.raises(ValueError, match=msg):
+            mixed_df.plot(stacked=True)
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_line_area_stacked_positive_idx(self, kind):
+        df = DataFrame(
+            np.random.default_rng(2).random((6, 4)), columns=["w", "x", "y", "z"]
+        )
+        # Use an index with strictly positive values, preventing
+        #  matplotlib from warning about ignoring xlim
+        df2 = df.set_index(df.index + 1)
+        _check_plot_works(df2.plot, kind=kind, logx=True, stacked=True)
+
+    @pytest.mark.parametrize(
+        "idx", [range(4), date_range("2023-01-1", freq="D", periods=4)]
+    )
+    def test_line_area_nan_df(self, idx):
+        values1 = [1, 2, np.nan, 3]
+        values2 = [3, np.nan, 2, 1]
+        df = DataFrame({"a": values1, "b": values2}, index=idx)
+
+        ax = _check_plot_works(df.plot)
+        masked1 = ax.lines[0].get_ydata()
+        masked2 = ax.lines[1].get_ydata()
+        # remove nan for comparison purpose
+
+        exp = np.array([1, 2, 3], dtype=np.float64)
+        tm.assert_numpy_array_equal(np.delete(masked1.data, 2), exp)
+
+        exp = np.array([3, 2, 1], dtype=np.float64)
+        tm.assert_numpy_array_equal(np.delete(masked2.data, 1), exp)
+        tm.assert_numpy_array_equal(masked1.mask, np.array([False, False, True, False]))
+        tm.assert_numpy_array_equal(masked2.mask, np.array([False, True, False, False]))
+
+    @pytest.mark.parametrize(
+        "idx", [range(4), date_range("2023-01-1", freq="D", periods=4)]
+    )
+    def test_line_area_nan_df_stacked(self, idx):
+        values1 = [1, 2, np.nan, 3]
+        values2 = [3, np.nan, 2, 1]
+        df = DataFrame({"a": values1, "b": values2}, index=idx)
+
+        expected1 = np.array([1, 2, 0, 3], dtype=np.float64)
+        expected2 = np.array([3, 0, 2, 1], dtype=np.float64)
+
+        ax = _check_plot_works(df.plot, stacked=True)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected1)
+        tm.assert_numpy_array_equal(ax.lines[1].get_ydata(), expected1 + expected2)
+
+    @pytest.mark.parametrize(
+        "idx", [range(4), date_range("2023-01-1", freq="D", periods=4)]
+    )
+    @pytest.mark.parametrize("kwargs", [{}, {"stacked": False}])
+    def test_line_area_nan_df_stacked_area(self, idx, kwargs):
+        values1 = [1, 2, np.nan, 3]
+        values2 = [3, np.nan, 2, 1]
+        df = DataFrame({"a": values1, "b": values2}, index=idx)
+
+        expected1 = np.array([1, 2, 0, 3], dtype=np.float64)
+        expected2 = np.array([3, 0, 2, 1], dtype=np.float64)
+
+        ax = _check_plot_works(df.plot.area, **kwargs)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected1)
+        if kwargs:
+            tm.assert_numpy_array_equal(ax.lines[1].get_ydata(), expected2)
+        else:
+            tm.assert_numpy_array_equal(ax.lines[1].get_ydata(), expected1 + expected2)
+
+        ax = _check_plot_works(df.plot.area, stacked=False)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected1)
+        tm.assert_numpy_array_equal(ax.lines[1].get_ydata(), expected2)
+
+    @pytest.mark.parametrize("kwargs", [{}, {"secondary_y": True}])
+    def test_line_lim(self, kwargs):
+        df = DataFrame(np.random.default_rng(2).random((6, 3)), columns=["x", "y", "z"])
+        ax = df.plot(**kwargs)
+        xmin, xmax = ax.get_xlim()
+        lines = ax.get_lines()
+        assert xmin <= lines[0].get_data()[0][0]
+        assert xmax >= lines[0].get_data()[0][-1]
+
+    def test_line_lim_subplots(self):
+        df = DataFrame(np.random.default_rng(2).random((6, 3)), columns=["x", "y", "z"])
+        axes = df.plot(secondary_y=True, subplots=True)
+        _check_axes_shape(axes, axes_num=3, layout=(3, 1))
+        for ax in axes:
+            assert hasattr(ax, "left_ax")
+            assert not hasattr(ax, "right_ax")
+            xmin, xmax = ax.get_xlim()
+            lines = ax.get_lines()
+            assert xmin <= lines[0].get_data()[0][0]
+            assert xmax >= lines[0].get_data()[0][-1]
+
+    @pytest.mark.xfail(
+        strict=False,
+        reason="2020-12-01 this has been failing periodically on the "
+        "ymin==0 assertion for a week or so.",
+    )
+    @pytest.mark.parametrize("stacked", [True, False])
+    def test_area_lim(self, stacked):
+        df = DataFrame(
+            np.random.default_rng(2).random((6, 4)), columns=["x", "y", "z", "four"]
+        )
+
+        neg_df = -df
+
+        ax = _check_plot_works(df.plot.area, stacked=stacked)
+        xmin, xmax = ax.get_xlim()
+        ymin, ymax = ax.get_ylim()
+        lines = ax.get_lines()
+        assert xmin <= lines[0].get_data()[0][0]
+        assert xmax >= lines[0].get_data()[0][-1]
+        assert ymin == 0
+
+        ax = _check_plot_works(neg_df.plot.area, stacked=stacked)
+        ymin, ymax = ax.get_ylim()
+        assert ymax == 0
+
+    def test_area_sharey_dont_overwrite(self):
+        # GH37942
+        df = DataFrame(np.random.default_rng(2).random((4, 2)), columns=["x", "y"])
+        fig, (ax1, ax2) = mpl.pyplot.subplots(1, 2, sharey=True)
+
+        df.plot(ax=ax1, kind="area")
+        df.plot(ax=ax2, kind="area")
+
+        assert get_y_axis(ax1).joined(ax1, ax2)
+        assert get_y_axis(ax2).joined(ax1, ax2)
+
+    @pytest.mark.parametrize("stacked", [True, False])
+    def test_bar_linewidth(self, stacked):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        ax = df.plot.bar(stacked=stacked, linewidth=2)
+        for r in ax.patches:
+            assert r.get_linewidth() == 2
+
+    def test_bar_linewidth_subplots(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # subplots
+        axes = df.plot.bar(linewidth=2, subplots=True)
+        _check_axes_shape(axes, axes_num=5, layout=(5, 1))
+        for ax in axes:
+            for r in ax.patches:
+                assert r.get_linewidth() == 2
+
+    @pytest.mark.parametrize(
+        "meth, dim", [("bar", "get_width"), ("barh", "get_height")]
+    )
+    @pytest.mark.parametrize("stacked", [True, False])
+    def test_bar_barwidth(self, meth, dim, stacked):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        width = 0.9
+
+        ax = getattr(df.plot, meth)(stacked=stacked, width=width)
+        for r in ax.patches:
+            if not stacked:
+                assert getattr(r, dim)() == width / len(df.columns)
+            else:
+                assert getattr(r, dim)() == width
+
+    @pytest.mark.parametrize(
+        "meth, dim", [("bar", "get_width"), ("barh", "get_height")]
+    )
+    def test_barh_barwidth_subplots(self, meth, dim):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        width = 0.9
+
+        axes = getattr(df.plot, meth)(width=width, subplots=True)
+        for ax in axes:
+            for r in ax.patches:
+                assert getattr(r, dim)() == width
+
+    def test_bar_bottom_left_bottom(self):
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+        ax = df.plot.bar(stacked=False, bottom=1)
+        result = [p.get_y() for p in ax.patches]
+        assert result == [1] * 25
+
+        ax = df.plot.bar(stacked=True, bottom=[-1, -2, -3, -4, -5])
+        result = [p.get_y() for p in ax.patches[:5]]
+        assert result == [-1, -2, -3, -4, -5]
+
+    def test_bar_bottom_left_left(self):
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+        ax = df.plot.barh(stacked=False, left=np.array([1, 1, 1, 1, 1]))
+        result = [p.get_x() for p in ax.patches]
+        assert result == [1] * 25
+
+        ax = df.plot.barh(stacked=True, left=[1, 2, 3, 4, 5])
+        result = [p.get_x() for p in ax.patches[:5]]
+        assert result == [1, 2, 3, 4, 5]
+
+    def test_bar_bottom_left_subplots(self):
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+        axes = df.plot.bar(subplots=True, bottom=-1)
+        for ax in axes:
+            result = [p.get_y() for p in ax.patches]
+            assert result == [-1] * 5
+
+        axes = df.plot.barh(subplots=True, left=np.array([1, 1, 1, 1, 1]))
+        for ax in axes:
+            result = [p.get_x() for p in ax.patches]
+            assert result == [1] * 5
+
+    def test_bar_nan(self):
+        df = DataFrame({"A": [10, np.nan, 20], "B": [5, 10, 20], "C": [1, 2, 3]})
+        ax = df.plot.bar()
+        expected = [10, 0, 20, 5, 10, 20, 1, 2, 3]
+        result = [p.get_height() for p in ax.patches]
+        assert result == expected
+
+    def test_bar_nan_stacked(self):
+        df = DataFrame({"A": [10, np.nan, 20], "B": [5, 10, 20], "C": [1, 2, 3]})
+        ax = df.plot.bar(stacked=True)
+        expected = [10, 0, 20, 5, 10, 20, 1, 2, 3]
+        result = [p.get_height() for p in ax.patches]
+        assert result == expected
+
+        result = [p.get_y() for p in ax.patches]
+        expected = [0.0, 0.0, 0.0, 10.0, 0.0, 20.0, 15.0, 10.0, 40.0]
+        assert result == expected
+
+    def test_bar_stacked_label_position_with_zero_height(self):
+        # GH 59429
+        df = DataFrame({"A": [3, 0, 1], "B": [0, 2, 4], "C": [5, 0, 2]})
+        ax = df.plot.bar(stacked=True)
+        ax.bar_label(ax.containers[-1])
+        expected = [8.0, 2.0, 7.0]
+        result = [text.xy[1] for text in ax.texts]
+        tm.assert_almost_equal(result, expected)
+        plt.close("all")
+
+    @pytest.mark.parametrize("idx", [Index, pd.CategoricalIndex])
+    def test_bar_categorical(self, idx):
+        # GH 13019
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 5)),
+            index=idx(list("ABCDEF")),
+            columns=idx(list("abcde")),
+        )
+
+        ax = df.plot.bar()
+        ticks = ax.xaxis.get_ticklocs()
+        tm.assert_numpy_array_equal(ticks, np.array([0, 1, 2, 3, 4, 5]))
+        assert ax.get_xlim() == (-0.5, 5.5)
+        # check left-edge of bars
+        assert ax.patches[0].get_x() == -0.25
+        assert ax.patches[-1].get_x() == 5.15
+
+        ax = df.plot.bar(stacked=True)
+        tm.assert_numpy_array_equal(ticks, np.array([0, 1, 2, 3, 4, 5]))
+        assert ax.get_xlim() == (-0.5, 5.5)
+        assert ax.patches[0].get_x() == -0.25
+        assert ax.patches[-1].get_x() == 4.75
+
+    @pytest.mark.parametrize("x, y", [("x", "y"), (1, 2)])
+    def test_plot_scatter(self, x, y):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+
+        _check_plot_works(df.plot.scatter, x=x, y=y)
+
+    def test_plot_scatter_error(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+        msg = re.escape("scatter() missing 1 required positional argument: 'y'")
+        with pytest.raises(TypeError, match=msg):
+            df.plot.scatter(x="x")
+        msg = re.escape("scatter() missing 1 required positional argument: 'x'")
+        with pytest.raises(TypeError, match=msg):
+            df.plot.scatter(y="y")
+
+    def test_plot_scatter_shape(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+        # GH 6951
+        axes = df.plot(x="x", y="y", kind="scatter", subplots=True)
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+
+    def test_raise_error_on_datetime_time_data(self):
+        # GH 8113, datetime.time type is not supported by matplotlib in scatter
+        df = DataFrame(np.random.default_rng(2).standard_normal(10), columns=["a"])
+        df["dtime"] = date_range(start="2014-01-01", freq="h", periods=10).time
+        msg = "must be a string or a (real )?number, not 'datetime.time'"
+
+        with pytest.raises(TypeError, match=msg):
+            df.plot(kind="scatter", x="dtime", y="a")
+
+    @pytest.mark.parametrize("x, y", [("dates", "vals"), (0, 1)])
+    def test_scatterplot_datetime_data(self, x, y):
+        # GH 30391
+        dates = date_range(start=date(2019, 1, 1), periods=12, freq="W")
+        vals = np.random.default_rng(2).normal(0, 1, len(dates))
+        df = DataFrame({"dates": dates, "vals": vals})
+
+        _check_plot_works(df.plot.scatter, x=x, y=y)
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    @pytest.mark.parametrize("x, y", [("a", "b"), (0, 1)])
+    @pytest.mark.parametrize("b_col", [[2, 3, 4], ["a", "b", "c"]])
+    def test_scatterplot_object_data(self, b_col, x, y, infer_string):
+        # GH 18755
+        with option_context("future.infer_string", infer_string):
+            df = DataFrame({"a": ["A", "B", "C"], "b": b_col})
+
+            _check_plot_works(df.plot.scatter, x=x, y=y)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    @pytest.mark.parametrize(
+        "categories",
+        (["setosa", "versicolor", "virginica"], ["versicolor", "virginica", "setosa"]),
+    )
+    def test_scatterplot_color_by_categorical(self, ordered, categories):
+        df = DataFrame(
+            [[5.1, 3.5], [4.9, 3.0], [7.0, 3.2], [6.4, 3.2], [5.9, 3.0]],
+            columns=["length", "width"],
+        )
+        df["species"] = pd.Categorical(
+            ["setosa", "setosa", "virginica", "virginica", "versicolor"],
+            ordered=ordered,
+            categories=categories,
+        )
+        ax = df.plot.scatter(x=0, y=1, c="species")
+        (colorbar_collection,) = ax.collections
+        colorbar = colorbar_collection.colorbar
+
+        expected_ticks = np.array([0.5, 1.5, 2.5])
+        result_ticks = colorbar.get_ticks()
+        tm.assert_numpy_array_equal(result_ticks, expected_ticks)
+
+        expected_boundaries = np.array([0.0, 1.0, 2.0, 3.0])
+        result_boundaries = colorbar._boundaries
+        tm.assert_numpy_array_equal(result_boundaries, expected_boundaries)
+
+        expected_yticklabels = categories
+        result_yticklabels = [i.get_text() for i in colorbar.ax.get_ymajorticklabels()]
+        assert all(i == j for i, j in zip(result_yticklabels, expected_yticklabels))
+
+    @pytest.mark.parametrize("x, y", [("x", "y"), ("y", "x"), ("y", "y")])
+    def test_plot_scatter_with_categorical_data(self, x, y):
+        # after fixing GH 18755, should be able to plot categorical data
+        df = DataFrame({"x": [1, 2, 3, 4], "y": pd.Categorical(["a", "b", "a", "c"])})
+
+        _check_plot_works(df.plot.scatter, x=x, y=y)
+
+    @pytest.mark.parametrize("x, y, c", [("x", "y", "z"), (0, 1, 2)])
+    def test_plot_scatter_with_c(self, x, y, c):
+        df = DataFrame(
+            np.random.default_rng(2).integers(low=0, high=100, size=(6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+
+        ax = df.plot.scatter(x=x, y=y, c=c)
+        # default to Greys
+        assert ax.collections[0].cmap.name == "Greys"
+
+        assert ax.collections[0].colorbar.ax.get_ylabel() == "z"
+
+    def test_plot_scatter_with_c_props(self):
+        df = DataFrame(
+            np.random.default_rng(2).integers(low=0, high=100, size=(6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["x", "y", "z", "four"],
+        )
+        cm = "cubehelix"
+        ax = df.plot.scatter(x="x", y="y", c="z", colormap=cm)
+        assert ax.collections[0].cmap.name == cm
+
+        # verify turning off colorbar works
+        ax = df.plot.scatter(x="x", y="y", c="z", colorbar=False)
+        assert ax.collections[0].colorbar is None
+
+        # verify that we can still plot a solid color
+        ax = df.plot.scatter(x=0, y=1, c="red")
+        assert ax.collections[0].colorbar is None
+        _check_colors(ax.collections, facecolors=["r"])
+
+    def test_plot_scatter_with_c_array(self):
+        # Ensure that we can pass an np.array straight through to matplotlib,
+        # this functionality was accidentally removed previously.
+        # See https://github.com/pandas-dev/pandas/issues/8852 for bug report
+        #
+        # Exercise colormap path and non-colormap path as they are independent
+        #
+        df = DataFrame({"A": [1, 2], "B": [3, 4]})
+        red_rgba = [1.0, 0.0, 0.0, 1.0]
+        green_rgba = [0.0, 1.0, 0.0, 1.0]
+        rgba_array = np.array([red_rgba, green_rgba])
+        ax = df.plot.scatter(x="A", y="B", c=rgba_array)
+        # expect the face colors of the points in the non-colormap path to be
+        # identical to the values we supplied, normally we'd be on shaky ground
+        # comparing floats for equality but here we expect them to be
+        # identical.
+        tm.assert_numpy_array_equal(ax.collections[0].get_facecolor(), rgba_array)
+        # we don't test the colors of the faces in this next plot because they
+        # are dependent on the spring colormap, which may change its colors
+        # later.
+        float_array = np.array([0.0, 1.0])
+        df.plot.scatter(x="A", y="B", c=float_array, cmap="spring")
+
+    def test_plot_scatter_with_s(self):
+        # this refers to GH 32904
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)) * 100, columns=["a", "b", "c"]
+        )
+
+        ax = df.plot.scatter(x="a", y="b", s="c")
+        tm.assert_numpy_array_equal(df["c"].values, right=ax.collections[0].get_sizes())
+
+    def test_plot_scatter_with_norm(self):
+        # added while fixing GH 45809
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)) * 100, columns=["a", "b", "c"]
+        )
+        norm = mpl.colors.LogNorm()
+        ax = df.plot.scatter(x="a", y="b", c="c", norm=norm)
+        assert ax.collections[0].norm is norm
+
+    def test_plot_scatter_without_norm(self):
+        # added while fixing GH 45809
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)) * 100, columns=["a", "b", "c"]
+        )
+        ax = df.plot.scatter(x="a", y="b", c="c")
+        plot_norm = ax.collections[0].norm
+        color_min_max = (df.c.min(), df.c.max())
+        default_norm = mpl.colors.Normalize(*color_min_max)
+        for value in df.c:
+            assert plot_norm(value) == default_norm(value)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {},
+            {"legend": False},
+            {"default_axes": True, "subplots": True},
+            {"stacked": True},
+        ],
+    )
+    def test_plot_bar(self, kwargs):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+
+        _check_plot_works(df.plot.bar, **kwargs)
+
+    @pytest.mark.slow
+    def test_plot_bar_int_col(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 15)),
+            index=list(string.ascii_letters[:10]),
+            columns=range(15),
+        )
+        _check_plot_works(df.plot.bar)
+
+    @pytest.mark.slow
+    def test_plot_bar_ticks(self):
+        df = DataFrame({"a": [0, 1], "b": [1, 0]})
+        ax = _check_plot_works(df.plot.bar)
+        _check_ticks_props(ax, xrot=90)
+
+        ax = df.plot.bar(rot=35, fontsize=10)
+        _check_ticks_props(ax, xrot=35, xlabelsize=10, ylabelsize=10)
+
+    @pytest.mark.slow
+    def test_plot_barh_ticks(self):
+        df = DataFrame({"a": [0, 1], "b": [1, 0]})
+        ax = _check_plot_works(df.plot.barh)
+        _check_ticks_props(ax, yrot=0)
+
+        ax = df.plot.barh(rot=55, fontsize=11)
+        _check_ticks_props(ax, yrot=55, ylabelsize=11, xlabelsize=11)
+
+    def test_boxplot(self, hist_df):
+        df = hist_df
+        numeric_cols = df._get_numeric_data().columns
+        labels = [pprint_thing(c) for c in numeric_cols]
+
+        ax = _check_plot_works(df.plot.box)
+        _check_text_labels(ax.get_xticklabels(), labels)
+        tm.assert_numpy_array_equal(
+            ax.xaxis.get_ticklocs(), np.arange(1, len(numeric_cols) + 1)
+        )
+        assert len(ax.lines) == 7 * len(numeric_cols)
+
+    def test_boxplot_series(self, hist_df):
+        df = hist_df
+        series = df["height"]
+        axes = series.plot.box(rot=40)
+        _check_ticks_props(axes, xrot=40, yrot=0)
+
+        _check_plot_works(series.plot.box)
+
+    def test_boxplot_series_positions(self, hist_df):
+        df = hist_df
+        positions = np.array([1, 6, 7])
+        ax = df.plot.box(positions=positions)
+        numeric_cols = df._get_numeric_data().columns
+        labels = [pprint_thing(c) for c in numeric_cols]
+        _check_text_labels(ax.get_xticklabels(), labels)
+        tm.assert_numpy_array_equal(ax.xaxis.get_ticklocs(), positions)
+        assert len(ax.lines) == 7 * len(numeric_cols)
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    @pytest.mark.xfail(
+        Version(mpl.__version__) >= Version("3.10"),
+        reason="Fails starting with matplotlib 3.10",
+    )
+    def test_boxplot_vertical(self, hist_df):
+        df = hist_df
+        numeric_cols = df._get_numeric_data().columns
+        labels = [pprint_thing(c) for c in numeric_cols]
+
+        # if horizontal, yticklabels are rotated
+        kwargs = (
+            {"vert": False}
+            if Version(mpl.__version__) < Version("3.10")
+            else {"orientation": "horizontal"}
+        )
+        ax = df.plot.box(rot=50, fontsize=8, **kwargs)
+        _check_ticks_props(ax, xrot=0, yrot=50, ylabelsize=8)
+        _check_text_labels(ax.get_yticklabels(), labels)
+        assert len(ax.lines) == 7 * len(numeric_cols)
+
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    @pytest.mark.xfail(
+        Version(mpl.__version__) >= Version("3.10"),
+        reason="Fails starting with matplotlib version 3.10",
+    )
+    def test_boxplot_vertical_subplots(self, hist_df):
+        df = hist_df
+        numeric_cols = df._get_numeric_data().columns
+        labels = [pprint_thing(c) for c in numeric_cols]
+        kwargs = (
+            {"vert": False}
+            if Version(mpl.__version__) < Version("3.10")
+            else {"orientation": "horizontal"}
+        )
+        axes = _check_plot_works(
+            df.plot.box, default_axes=True, subplots=True, logx=True, **kwargs
+        )
+        _check_axes_shape(axes, axes_num=3, layout=(1, 3))
+        _check_ax_scales(axes, xaxis="log")
+        for ax, label in zip(axes, labels):
+            _check_text_labels(ax.get_yticklabels(), [label])
+            assert len(ax.lines) == 7
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    @pytest.mark.xfail(
+        Version(mpl.__version__) >= Version("3.10"),
+        reason="Fails starting with matplotlib 3.10",
+    )
+    def test_boxplot_vertical_positions(self, hist_df):
+        df = hist_df
+        numeric_cols = df._get_numeric_data().columns
+        labels = [pprint_thing(c) for c in numeric_cols]
+        positions = np.array([3, 2, 8])
+        kwargs = (
+            {"vert": False}
+            if Version(mpl.__version__) < Version("3.10")
+            else {"orientation": "horizontal"}
+        )
+        ax = df.plot.box(positions=positions, **kwargs)
+        _check_text_labels(ax.get_yticklabels(), labels)
+        tm.assert_numpy_array_equal(ax.yaxis.get_ticklocs(), positions)
+        assert len(ax.lines) == 7 * len(numeric_cols)
+
+    def test_boxplot_return_type_invalid(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+        msg = "return_type must be {None, 'axes', 'dict', 'both'}"
+        with pytest.raises(ValueError, match=msg):
+            df.plot.box(return_type="not_a_type")
+
+    @pytest.mark.parametrize("return_type", ["dict", "axes", "both"])
+    def test_boxplot_return_type_invalid_type(self, return_type):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+        result = df.plot.box(return_type=return_type)
+        _check_box_return_type(result, return_type)
+
+    def test_kde_df(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        ax = _check_plot_works(df.plot, kind="kde")
+        expected = [pprint_thing(c) for c in df.columns]
+        _check_legend_labels(ax, labels=expected)
+        _check_ticks_props(ax, xrot=0)
+
+    def test_kde_df_rot(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        ax = df.plot(kind="kde", rot=20, fontsize=5)
+        _check_ticks_props(ax, xrot=20, xlabelsize=5, ylabelsize=5)
+
+    def test_kde_df_subplots(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        axes = _check_plot_works(
+            df.plot,
+            default_axes=True,
+            kind="kde",
+            subplots=True,
+        )
+        _check_axes_shape(axes, axes_num=4, layout=(4, 1))
+
+    def test_kde_df_logy(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        axes = df.plot(kind="kde", logy=True, subplots=True)
+        _check_ax_scales(axes, yaxis="log")
+
+    def test_kde_missing_vals(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).uniform(size=(100, 4)))
+        df.loc[0, 0] = np.nan
+        _check_plot_works(df.plot, kind="kde")
+
+    def test_hist_df(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((100, 4)))
+
+        ax = _check_plot_works(df.plot.hist)
+        expected = [pprint_thing(c) for c in df.columns]
+        _check_legend_labels(ax, labels=expected)
+
+        axes = _check_plot_works(
+            df.plot.hist,
+            default_axes=True,
+            subplots=True,
+            logy=True,
+        )
+        _check_axes_shape(axes, axes_num=4, layout=(4, 1))
+        _check_ax_scales(axes, yaxis="log")
+
+    def test_hist_df_series(self):
+        series = Series(np.random.default_rng(2).random(10))
+        axes = series.plot.hist(rot=40)
+        _check_ticks_props(axes, xrot=40, yrot=0)
+
+    def test_hist_df_series_cumulative_density(self):
+        series = Series(np.random.default_rng(2).random(10))
+        ax = series.plot.hist(cumulative=True, bins=4, density=True)
+        # height of last bin (index 5) must be 1.0
+        rects = [x for x in ax.get_children() if isinstance(x, mpl.patches.Rectangle)]
+        tm.assert_almost_equal(rects[-1].get_height(), 1.0)
+
+    def test_hist_df_series_cumulative(self):
+        series = Series(np.random.default_rng(2).random(10))
+        ax = series.plot.hist(cumulative=True, bins=4)
+        rects = [x for x in ax.get_children() if isinstance(x, mpl.patches.Rectangle)]
+
+        tm.assert_almost_equal(rects[-2].get_height(), 10.0)
+
+    def test_hist_df_orientation(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        # if horizontal, yticklabels are rotated
+        axes = df.plot.hist(rot=50, fontsize=8, orientation="horizontal")
+        _check_ticks_props(axes, xrot=0, yrot=50, ylabelsize=8)
+
+    @pytest.mark.parametrize("weight_shape", [(100,), (100, 2)])
+    def test_hist_weights(self, weight_shape):
+        # GH 33173
+        weights = 0.1 * np.ones(shape=weight_shape)
+        df = DataFrame(
+            dict(zip(["A", "B"], np.random.default_rng(2).standard_normal((2, 100))))
+        )
+
+        ax1 = _check_plot_works(df.plot, kind="hist", weights=weights)
+        ax2 = _check_plot_works(df.plot, kind="hist")
+
+        patch_height_with_weights = [patch.get_height() for patch in ax1.patches]
+
+        # original heights with no weights, and we manually multiply with example
+        # weights, so after multiplication, they should be almost same
+        expected_patch_height = [0.1 * patch.get_height() for patch in ax2.patches]
+
+        tm.assert_almost_equal(patch_height_with_weights, expected_patch_height)
+
+    def _check_box_coord(
+        self,
+        patches,
+        expected_y=None,
+        expected_h=None,
+        expected_x=None,
+        expected_w=None,
+    ):
+        result_y = np.array([p.get_y() for p in patches])
+        result_height = np.array([p.get_height() for p in patches])
+        result_x = np.array([p.get_x() for p in patches])
+        result_width = np.array([p.get_width() for p in patches])
+        # dtype is depending on above values, no need to check
+
+        if expected_y is not None:
+            tm.assert_numpy_array_equal(result_y, expected_y, check_dtype=False)
+        if expected_h is not None:
+            tm.assert_numpy_array_equal(result_height, expected_h, check_dtype=False)
+        if expected_x is not None:
+            tm.assert_numpy_array_equal(result_x, expected_x, check_dtype=False)
+        if expected_w is not None:
+            tm.assert_numpy_array_equal(result_width, expected_w, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            {
+                "A": np.repeat(np.array([1, 2, 3, 4, 5]), np.array([10, 9, 8, 7, 6])),
+                "B": np.repeat(np.array([1, 2, 3, 4, 5]), np.array([8, 8, 8, 8, 8])),
+                "C": np.repeat(np.array([1, 2, 3, 4, 5]), np.array([6, 7, 8, 9, 10])),
+            },
+            {
+                "A": np.repeat(
+                    np.array([np.nan, 1, 2, 3, 4, 5]), np.array([3, 10, 9, 8, 7, 6])
+                ),
+                "B": np.repeat(
+                    np.array([1, np.nan, 2, 3, 4, 5]), np.array([8, 3, 8, 8, 8, 8])
+                ),
+                "C": np.repeat(
+                    np.array([1, 2, 3, np.nan, 4, 5]), np.array([6, 7, 8, 3, 9, 10])
+                ),
+            },
+        ],
+    )
+    def test_hist_df_coord(self, data):
+        df = DataFrame(data)
+
+        ax = df.plot.hist(bins=5)
+        self._check_box_coord(
+            ax.patches[:5],
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            ax.patches[5:10],
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            ax.patches[10:],
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([6, 7, 8, 9, 10]),
+        )
+
+        ax = df.plot.hist(bins=5, stacked=True)
+        self._check_box_coord(
+            ax.patches[:5],
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            ax.patches[5:10],
+            expected_y=np.array([10, 9, 8, 7, 6]),
+            expected_h=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            ax.patches[10:],
+            expected_y=np.array([18, 17, 16, 15, 14]),
+            expected_h=np.array([6, 7, 8, 9, 10]),
+        )
+
+        axes = df.plot.hist(bins=5, stacked=True, subplots=True)
+        self._check_box_coord(
+            axes[0].patches,
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            axes[1].patches,
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            axes[2].patches,
+            expected_y=np.array([0, 0, 0, 0, 0]),
+            expected_h=np.array([6, 7, 8, 9, 10]),
+        )
+
+        # horizontal
+        ax = df.plot.hist(bins=5, orientation="horizontal")
+        self._check_box_coord(
+            ax.patches[:5],
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            ax.patches[5:10],
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            ax.patches[10:],
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([6, 7, 8, 9, 10]),
+        )
+
+        ax = df.plot.hist(bins=5, stacked=True, orientation="horizontal")
+        self._check_box_coord(
+            ax.patches[:5],
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            ax.patches[5:10],
+            expected_x=np.array([10, 9, 8, 7, 6]),
+            expected_w=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            ax.patches[10:],
+            expected_x=np.array([18, 17, 16, 15, 14]),
+            expected_w=np.array([6, 7, 8, 9, 10]),
+        )
+
+        axes = df.plot.hist(
+            bins=5, stacked=True, subplots=True, orientation="horizontal"
+        )
+        self._check_box_coord(
+            axes[0].patches,
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([10, 9, 8, 7, 6]),
+        )
+        self._check_box_coord(
+            axes[1].patches,
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([8, 8, 8, 8, 8]),
+        )
+        self._check_box_coord(
+            axes[2].patches,
+            expected_x=np.array([0, 0, 0, 0, 0]),
+            expected_w=np.array([6, 7, 8, 9, 10]),
+        )
+
+    def test_plot_int_columns(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((100, 4))).cumsum()
+        _check_plot_works(df.plot, legend=True)
+
+    @pytest.mark.parametrize(
+        "markers",
+        [
+            {0: "^", 1: "+", 2: "o"},
+            {0: "^", 1: "+"},
+            ["^", "+", "o"],
+            ["^", "+"],
+        ],
+    )
+    def test_style_by_column(self, markers):
+        fig = plt.gcf()
+        fig.clf()
+        fig.add_subplot(111)
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 3)))
+        ax = df.plot(style=markers)
+        for idx, line in enumerate(ax.get_lines()[: len(markers)]):
+            assert line.get_marker() == markers[idx]
+
+    def test_line_label_none(self):
+        s = Series([1, 2])
+        ax = s.plot()
+        assert ax.get_legend() is None
+
+        ax = s.plot(legend=True)
+        assert ax.get_legend().get_texts()[0].get_text() == ""
+
+    @pytest.mark.parametrize(
+        "props, expected",
+        [
+            ("boxprops", "boxes"),
+            ("whiskerprops", "whiskers"),
+            ("capprops", "caps"),
+            ("medianprops", "medians"),
+        ],
+    )
+    def test_specified_props_kwd_plot_box(self, props, expected):
+        # GH 30346
+        df = DataFrame({k: np.random.default_rng(2).random(100) for k in "ABC"})
+        kwd = {props: {"color": "C1"}}
+        result = df.plot.box(return_type="dict", **kwd)
+
+        assert result[expected][0].get_color() == "C1"
+
+    def test_unordered_ts(self):
+        # GH#2609, GH#55906
+        index = [date(2012, 10, 1), date(2012, 9, 1), date(2012, 8, 1)]
+        values = [3.0, 2.0, 1.0]
+        df = DataFrame(
+            np.array(values),
+            index=index,
+            columns=["test"],
+        )
+        ax = df.plot()
+        xticks = ax.lines[0].get_xdata()
+        tm.assert_numpy_array_equal(xticks, np.array(index, dtype=object))
+        ydata = ax.lines[0].get_ydata()
+        tm.assert_numpy_array_equal(ydata, np.array(values))
+
+        # even though we don't sort the data before passing it to matplotlib,
+        # the ticks are sorted
+        xticks = ax.xaxis.get_ticklabels()
+        xlocs = [x.get_position()[0] for x in xticks]
+        assert Index(xlocs).is_monotonic_increasing
+        xlabels = [x.get_text() for x in xticks]
+        assert pd.to_datetime(xlabels, format="%Y-%m-%d").is_monotonic_increasing
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._common_kinds)
+    def test_kind_both_ways(self, kind):
+        pytest.importorskip("scipy")
+        df = DataFrame({"x": [1, 2, 3]})
+        df.plot(kind=kind)
+        getattr(df.plot, kind)()
+
+    @pytest.mark.parametrize("kind", ["scatter", "hexbin"])
+    def test_kind_both_ways_x_y(self, kind):
+        pytest.importorskip("scipy")
+        df = DataFrame({"x": [1, 2, 3]})
+        df.plot("x", "x", kind=kind)
+        getattr(df.plot, kind)("x", "x")
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._common_kinds)
+    def test_all_invalid_plot_data(self, kind):
+        df = DataFrame(list("abcd"))
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            df.plot(kind=kind)
+
+    @pytest.mark.parametrize(
+        "kind", list(plotting.PlotAccessor._common_kinds) + ["area"]
+    )
+    def test_partially_invalid_plot_data_numeric(self, kind):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            dtype=object,
+        )
+        df[np.random.default_rng(2).random(df.shape[0]) > 0.5] = "a"
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            df.plot(kind=kind)
+
+    def test_invalid_kind(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        msg = "invalid_plot_kind is not a valid plot kind"
+        with pytest.raises(ValueError, match=msg):
+            df.plot(kind="invalid_plot_kind")
+
+    @pytest.mark.parametrize(
+        "x,y,lbl",
+        [
+            (["B", "C"], "A", "a"),
+            (["A"], ["B", "C"], ["b", "c"]),
+        ],
+    )
+    def test_invalid_xy_args(self, x, y, lbl):
+        # GH 18671, 19699 allows y to be list-like but not x
+        df = DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+        with pytest.raises(ValueError, match="x must be a label or position"):
+            df.plot(x=x, y=y, label=lbl)
+
+    def test_bad_label(self):
+        df = DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+        msg = "label should be list-like and same length as y"
+        with pytest.raises(ValueError, match=msg):
+            df.plot(x="A", y=["B", "C"], label="bad_label")
+
+    @pytest.mark.parametrize("x,y", [("A", "B"), (["A"], "B")])
+    def test_invalid_xy_args_dup_cols(self, x, y):
+        # GH 18671, 19699 allows y to be list-like but not x
+        df = DataFrame([[1, 3, 5], [2, 4, 6]], columns=list("AAB"))
+        with pytest.raises(ValueError, match="x must be a label or position"):
+            df.plot(x=x, y=y)
+
+    @pytest.mark.parametrize(
+        "x,y,lbl,colors",
+        [
+            ("A", ["B"], ["b"], ["red"]),
+            ("A", ["B", "C"], ["b", "c"], ["red", "blue"]),
+            (0, [1, 2], ["bokeh", "cython"], ["green", "yellow"]),
+        ],
+    )
+    def test_y_listlike(self, x, y, lbl, colors):
+        # GH 19699: tests list-like y and verifies lbls & colors
+        df = DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+        _check_plot_works(df.plot, x="A", y=y, label=lbl)
+
+        ax = df.plot(x=x, y=y, label=lbl, color=colors)
+        assert len(ax.lines) == len(y)
+        _check_colors(ax.get_lines(), linecolors=colors)
+
+    @pytest.mark.parametrize("x,y,colnames", [(0, 1, ["A", "B"]), (1, 0, [0, 1])])
+    def test_xy_args_integer(self, x, y, colnames):
+        # GH 20056: tests integer args for xy and checks col names
+        df = DataFrame({"A": [1, 2], "B": [3, 4]})
+        df.columns = colnames
+        _check_plot_works(df.plot, x=x, y=y)
+
+    def test_hexbin_basic(self):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+
+        ax = df.plot.hexbin(x="A", y="B", gridsize=10)
+        # TODO: need better way to test. This just does existence.
+        assert len(ax.collections) == 1
+
+    def test_hexbin_basic_subplots(self):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+        # GH 6951
+        axes = df.plot.hexbin(x="A", y="B", subplots=True)
+        # hexbin should have 2 axes in the figure, 1 for plotting and another
+        # is colorbar
+        assert len(axes[0].figure.axes) == 2
+        # return value is single axes
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+
+    @pytest.mark.parametrize("reduce_C", [None, np.std])
+    def test_hexbin_with_c(self, reduce_C):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+
+        ax = df.plot.hexbin(x="A", y="B", C="C", reduce_C_function=reduce_C)
+        assert len(ax.collections) == 1
+
+    @pytest.mark.parametrize(
+        "kwargs, expected",
+        [
+            ({}, "BuGn"),  # default cmap
+            ({"colormap": "cubehelix"}, "cubehelix"),
+            ({"cmap": "YlGn"}, "YlGn"),
+        ],
+    )
+    def test_hexbin_cmap(self, kwargs, expected):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+        ax = df.plot.hexbin(x="A", y="B", **kwargs)
+        assert ax.collections[0].cmap.name == expected
+
+    def test_pie_df_err(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((5, 3)),
+            columns=["X", "Y", "Z"],
+            index=["a", "b", "c", "d", "e"],
+        )
+        msg = "pie requires either y column or 'subplots=True'"
+        with pytest.raises(ValueError, match=msg):
+            df.plot.pie()
+
+    @pytest.mark.parametrize("y", ["Y", 2])
+    def test_pie_df(self, y):
+        df = DataFrame(
+            np.random.default_rng(2).random((5, 3)),
+            columns=["X", "Y", "Z"],
+            index=["a", "b", "c", "d", "e"],
+        )
+        ax = _check_plot_works(df.plot.pie, y=y)
+        _check_text_labels(ax.texts, df.index)
+
+    def test_pie_df_subplots(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((5, 3)),
+            columns=["X", "Y", "Z"],
+            index=["a", "b", "c", "d", "e"],
+        )
+        axes = _check_plot_works(
+            df.plot.pie,
+            default_axes=True,
+            subplots=True,
+        )
+        assert len(axes) == len(df.columns)
+        for ax in axes:
+            _check_text_labels(ax.texts, df.index)
+        for ax, ylabel in zip(axes, df.columns):
+            assert ax.get_ylabel() == ""
+
+    def test_pie_df_labels_colors(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((5, 3)),
+            columns=["X", "Y", "Z"],
+            index=["a", "b", "c", "d", "e"],
+        )
+        labels = ["A", "B", "C", "D", "E"]
+        color_args = ["r", "g", "b", "c", "m"]
+        axes = _check_plot_works(
+            df.plot.pie,
+            default_axes=True,
+            subplots=True,
+            labels=labels,
+            colors=color_args,
+        )
+        assert len(axes) == len(df.columns)
+
+        for ax in axes:
+            _check_text_labels(ax.texts, labels)
+            _check_colors(ax.patches, facecolors=color_args)
+
+    def test_pie_df_nan(self):
+        df = DataFrame(np.random.default_rng(2).random((4, 4)))
+        for i in range(4):
+            df.iloc[i, i] = np.nan
+        _, axes = mpl.pyplot.subplots(ncols=4)
+
+        # GH 37668
+        kwargs = {"normalize": True}
+
+        with tm.assert_produces_warning(None):
+            df.plot.pie(subplots=True, ax=axes, legend=True, **kwargs)
+
+        base_expected = ["0", "1", "2", "3"]
+        for i, ax in enumerate(axes):
+            expected = list(base_expected)  # force copy
+            expected[i] = ""
+            result = [x.get_text() for x in ax.texts]
+            assert result == expected
+
+            # legend labels
+            # NaN's not included in legend with subplots
+            # see https://github.com/pandas-dev/pandas/issues/8390
+            result_labels = [x.get_text() for x in ax.get_legend().get_texts()]
+            expected_labels = base_expected[:i] + base_expected[i + 1 :]
+            assert result_labels == expected_labels
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"logy": True},
+            {"logx": True, "logy": True},
+            {"loglog": True},
+        ],
+    )
+    def test_errorbar_plot(self, kwargs):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        d_err = {"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}
+        df_err = DataFrame(d_err)
+
+        # check line plots
+        ax = _check_plot_works(df.plot, yerr=df_err, **kwargs)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    @pytest.mark.slow
+    def test_errorbar_plot_bar(self):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        d_err = {"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}
+        df_err = DataFrame(d_err)
+        ax = _check_plot_works(
+            (df + 1).plot, yerr=df_err, xerr=df_err, kind="bar", log=True
+        )
+        _check_has_errorbars(ax, xerr=2, yerr=2)
+
+    @pytest.mark.slow
+    def test_errorbar_plot_yerr_array(self):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        # yerr is raw error values
+        ax = _check_plot_works(df["y"].plot, yerr=np.ones(12) * 0.4)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+        ax = _check_plot_works(df.plot, yerr=np.ones((2, 12)) * 0.4)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("yerr", ["yerr", "誤差"])
+    def test_errorbar_plot_column_name(self, yerr):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        df[yerr] = np.ones(12) * 0.2
+
+        ax = _check_plot_works(df.plot, yerr=yerr)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+        ax = _check_plot_works(df.plot, y="y", x="x", yerr=yerr)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+    @pytest.mark.slow
+    def test_errorbar_plot_external_valueerror(self):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        with tm.external_error_raised(ValueError):
+            df.plot(yerr=np.random.default_rng(2).standard_normal(11))
+
+    @pytest.mark.slow
+    def test_errorbar_plot_external_typeerror(self):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        df_err = DataFrame({"x": ["zzz"] * 12, "y": ["zzz"] * 12})
+        with tm.external_error_raised(TypeError):
+            df.plot(yerr=df_err)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["line", "bar", "barh"])
+    @pytest.mark.parametrize(
+        "y_err",
+        [
+            Series(np.ones(12) * 0.2, name="x"),
+            DataFrame({"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}),
+        ],
+    )
+    def test_errorbar_plot_different_yerr(self, kind, y_err):
+        df = DataFrame({"x": np.arange(12), "y": np.arange(12, 0, -1)})
+
+        ax = _check_plot_works(df.plot, yerr=y_err, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["line", "bar", "barh"])
+    @pytest.mark.parametrize(
+        "y_err, x_err",
+        [
+            (
+                DataFrame({"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}),
+                DataFrame({"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}),
+            ),
+            (Series(np.ones(12) * 0.2, name="x"), Series(np.ones(12) * 0.2, name="x")),
+            (0.2, 0.2),
+        ],
+    )
+    def test_errorbar_plot_different_yerr_xerr(self, kind, y_err, x_err):
+        df = DataFrame({"x": np.arange(12), "y": np.arange(12, 0, -1)})
+        ax = _check_plot_works(df.plot, yerr=y_err, xerr=x_err, kind=kind)
+        _check_has_errorbars(ax, xerr=2, yerr=2)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["line", "bar", "barh"])
+    def test_errorbar_plot_different_yerr_xerr_subplots(self, kind):
+        df = DataFrame({"x": np.arange(12), "y": np.arange(12, 0, -1)})
+        df_err = DataFrame({"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4})
+        axes = _check_plot_works(
+            df.plot,
+            default_axes=True,
+            yerr=df_err,
+            xerr=df_err,
+            subplots=True,
+            kind=kind,
+        )
+        _check_has_errorbars(axes, xerr=1, yerr=1)
+
+    @pytest.mark.xfail(reason="Iterator is consumed", raises=ValueError)
+    def test_errorbar_plot_iterator(self):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+
+        # yerr is iterator
+        ax = _check_plot_works(df.plot, yerr=itertools.repeat(0.1, len(df)))
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    def test_errorbar_with_integer_column_names(self):
+        # test with integer column names
+        df = DataFrame(np.abs(np.random.default_rng(2).standard_normal((10, 2))))
+        df_err = DataFrame(np.abs(np.random.default_rng(2).standard_normal((10, 2))))
+        ax = _check_plot_works(df.plot, yerr=df_err)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+        ax = _check_plot_works(df.plot, y=0, yerr=1)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["line", "bar"])
+    def test_errorbar_with_partial_columns_kind(self, kind):
+        df = DataFrame(np.abs(np.random.default_rng(2).standard_normal((10, 3))))
+        df_err = DataFrame(
+            np.abs(np.random.default_rng(2).standard_normal((10, 2))), columns=[0, 2]
+        )
+        ax = _check_plot_works(df.plot, yerr=df_err, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    @pytest.mark.slow
+    def test_errorbar_with_partial_columns_dti(self):
+        df = DataFrame(np.abs(np.random.default_rng(2).standard_normal((10, 3))))
+        df_err = DataFrame(
+            np.abs(np.random.default_rng(2).standard_normal((10, 2))), columns=[0, 2]
+        )
+        ix = date_range("1/1/2000", periods=10, freq="ME")
+        df.set_index(ix, inplace=True)
+        df_err.set_index(ix, inplace=True)
+        ax = _check_plot_works(df.plot, yerr=df_err, kind="line")
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("err_box", [lambda x: x, DataFrame])
+    def test_errorbar_with_partial_columns_box(self, err_box):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        df = DataFrame(d)
+        err = err_box({"x": np.ones(12) * 0.2, "z": np.ones(12) * 0.4})
+        ax = _check_plot_works(df.plot, yerr=err)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+    @pytest.mark.parametrize("kind", ["line", "bar", "barh"])
+    def test_errorbar_timeseries(self, kind):
+        d = {"x": np.arange(12), "y": np.arange(12, 0, -1)}
+        d_err = {"x": np.ones(12) * 0.2, "y": np.ones(12) * 0.4}
+
+        # check time-series plots
+        ix = date_range("1/1/2000", "1/1/2001", freq="ME")
+        tdf = DataFrame(d, index=ix)
+        tdf_err = DataFrame(d_err, index=ix)
+
+        ax = _check_plot_works(tdf.plot, yerr=tdf_err, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+        ax = _check_plot_works(tdf.plot, yerr=d_err, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+        ax = _check_plot_works(tdf.plot, y="y", yerr=tdf_err["x"], kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+        ax = _check_plot_works(tdf.plot, y="y", yerr="x", kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+        ax = _check_plot_works(tdf.plot, yerr=tdf_err, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=2)
+
+        axes = _check_plot_works(
+            tdf.plot,
+            default_axes=True,
+            kind=kind,
+            yerr=tdf_err,
+            subplots=True,
+        )
+        _check_has_errorbars(axes, xerr=0, yerr=1)
+
+    def test_errorbar_asymmetrical(self):
+        err = np.random.default_rng(2).random((3, 2, 5))
+
+        # each column is [0, 1, 2, 3, 4], [3, 4, 5, 6, 7]...
+        df = DataFrame(np.arange(15).reshape(3, 5)).T
+
+        ax = df.plot(yerr=err, xerr=err / 2)
+
+        yerr_0_0 = ax.collections[1].get_paths()[0].vertices[:, 1]
+        expected_0_0 = err[0, :, 0] * np.array([-1, 1])
+        tm.assert_almost_equal(yerr_0_0, expected_0_0)
+
+        msg = re.escape(
+            "Asymmetrical error bars should be provided with the shape (3, 2, 5)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.plot(yerr=err.T)
+
+    def test_table(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        _check_plot_works(df.plot, table=True)
+        _check_plot_works(df.plot, table=df)
+
+        # GH 35945 UserWarning
+        with tm.assert_produces_warning(None):
+            ax = df.plot()
+            assert len(ax.tables) == 0
+            plotting.table(ax, df.T)
+            assert len(ax.tables) == 1
+
+    def test_errorbar_scatter(self):
+        df = DataFrame(
+            np.abs(np.random.default_rng(2).standard_normal((5, 2))),
+            index=range(5),
+            columns=["x", "y"],
+        )
+        df_err = DataFrame(
+            np.abs(np.random.default_rng(2).standard_normal((5, 2))) / 5,
+            index=range(5),
+            columns=["x", "y"],
+        )
+
+        ax = _check_plot_works(df.plot.scatter, x="x", y="y")
+        _check_has_errorbars(ax, xerr=0, yerr=0)
+        ax = _check_plot_works(df.plot.scatter, x="x", y="y", xerr=df_err)
+        _check_has_errorbars(ax, xerr=1, yerr=0)
+
+        ax = _check_plot_works(df.plot.scatter, x="x", y="y", yerr=df_err)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+        ax = _check_plot_works(df.plot.scatter, x="x", y="y", xerr=df_err, yerr=df_err)
+        _check_has_errorbars(ax, xerr=1, yerr=1)
+
+    def test_errorbar_scatter_color(self):
+        def _check_errorbar_color(containers, expected, has_err="has_xerr"):
+            lines = []
+            errs = next(c.lines for c in ax.containers if getattr(c, has_err, False))
+            for el in errs:
+                if is_list_like(el):
+                    lines.extend(el)
+                else:
+                    lines.append(el)
+            err_lines = [x for x in lines if x in ax.collections]
+            _check_colors(err_lines, linecolors=np.array([expected] * len(err_lines)))
+
+        # GH 8081
+        df = DataFrame(
+            np.abs(np.random.default_rng(2).standard_normal((10, 5))),
+            columns=["a", "b", "c", "d", "e"],
+        )
+        ax = df.plot.scatter(x="a", y="b", xerr="d", yerr="e", c="red")
+        _check_has_errorbars(ax, xerr=1, yerr=1)
+        _check_errorbar_color(ax.containers, "red", has_err="has_xerr")
+        _check_errorbar_color(ax.containers, "red", has_err="has_yerr")
+
+        ax = df.plot.scatter(x="a", y="b", yerr="e", color="green")
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+        _check_errorbar_color(ax.containers, "green", has_err="has_yerr")
+
+    def test_scatter_unknown_colormap(self):
+        # GH#48726
+        df = DataFrame({"a": [1, 2, 3], "b": 4})
+        with pytest.raises((ValueError, KeyError), match="'unknown' is not a"):
+            df.plot(x="a", y="b", colormap="unknown", kind="scatter")
+
+    def test_sharex_and_ax(self):
+        # https://github.com/pandas-dev/pandas/issues/9737 using gridspec,
+        # the axis in fig.get_axis() are sorted differently than pandas
+        # expected them, so make sure that only the right ones are removed
+        gs, axes = _generate_4_axes_via_gridspec()
+
+        df = DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, 6],
+                "b": [1, 2, 3, 4, 5, 6],
+                "c": [1, 2, 3, 4, 5, 6],
+                "d": [1, 2, 3, 4, 5, 6],
+            }
+        )
+
+        def _check(axes):
+            for ax in axes:
+                assert len(ax.lines) == 1
+                _check_visible(ax.get_yticklabels(), visible=True)
+            for ax in [axes[0], axes[2]]:
+                _check_visible(ax.get_xticklabels(), visible=False)
+                _check_visible(ax.get_xticklabels(minor=True), visible=False)
+            for ax in [axes[1], axes[3]]:
+                _check_visible(ax.get_xticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(minor=True), visible=True)
+
+        for ax in axes:
+            df.plot(x="a", y="b", title="title", ax=ax, sharex=True)
+        gs.tight_layout(plt.gcf())
+        _check(axes)
+        plt.close("all")
+
+        gs, axes = _generate_4_axes_via_gridspec()
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=axes, sharex=True)
+        _check(axes)
+
+    def test_sharex_false_and_ax(self):
+        # https://github.com/pandas-dev/pandas/issues/9737 using gridspec,
+        # the axis in fig.get_axis() are sorted differently than pandas
+        # expected them, so make sure that only the right ones are removed
+        df = DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, 6],
+                "b": [1, 2, 3, 4, 5, 6],
+                "c": [1, 2, 3, 4, 5, 6],
+                "d": [1, 2, 3, 4, 5, 6],
+            }
+        )
+        gs, axes = _generate_4_axes_via_gridspec()
+        # without sharex, no labels should be touched!
+        for ax in axes:
+            df.plot(x="a", y="b", title="title", ax=ax)
+
+        gs.tight_layout(plt.gcf())
+        for ax in axes:
+            assert len(ax.lines) == 1
+            _check_visible(ax.get_yticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+
+    def test_sharey_and_ax(self):
+        # https://github.com/pandas-dev/pandas/issues/9737 using gridspec,
+        # the axis in fig.get_axis() are sorted differently than pandas
+        # expected them, so make sure that only the right ones are removed
+        gs, axes = _generate_4_axes_via_gridspec()
+
+        df = DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, 6],
+                "b": [1, 2, 3, 4, 5, 6],
+                "c": [1, 2, 3, 4, 5, 6],
+                "d": [1, 2, 3, 4, 5, 6],
+            }
+        )
+
+        def _check(axes):
+            for ax in axes:
+                assert len(ax.lines) == 1
+                _check_visible(ax.get_xticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(minor=True), visible=True)
+            for ax in [axes[0], axes[1]]:
+                _check_visible(ax.get_yticklabels(), visible=True)
+            for ax in [axes[2], axes[3]]:
+                _check_visible(ax.get_yticklabels(), visible=False)
+
+        for ax in axes:
+            df.plot(x="a", y="b", title="title", ax=ax, sharey=True)
+        gs.tight_layout(plt.gcf())
+        _check(axes)
+        plt.close("all")
+
+        gs, axes = _generate_4_axes_via_gridspec()
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=axes, sharey=True)
+
+        gs.tight_layout(plt.gcf())
+        _check(axes)
+
+    def test_sharey_and_ax_tight(self):
+        # https://github.com/pandas-dev/pandas/issues/9737 using gridspec,
+        df = DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, 6],
+                "b": [1, 2, 3, 4, 5, 6],
+                "c": [1, 2, 3, 4, 5, 6],
+                "d": [1, 2, 3, 4, 5, 6],
+            }
+        )
+        gs, axes = _generate_4_axes_via_gridspec()
+        # without sharex, no labels should be touched!
+        for ax in axes:
+            df.plot(x="a", y="b", title="title", ax=ax)
+
+        gs.tight_layout(plt.gcf())
+        for ax in axes:
+            assert len(ax.lines) == 1
+            _check_visible(ax.get_yticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._all_kinds)
+    def test_memory_leak(self, kind):
+        """Check that every plot type gets properly collected."""
+        pytest.importorskip("scipy")
+        args = {}
+        if kind in ["hexbin", "scatter", "pie"]:
+            df = DataFrame(
+                {
+                    "A": np.random.default_rng(2).uniform(size=20),
+                    "B": np.random.default_rng(2).uniform(size=20),
+                    "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+                }
+            )
+            args = {"x": "A", "y": "B"}
+        elif kind == "area":
+            df = DataFrame(
+                np.random.default_rng(2).standard_normal((10, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=date_range("2000-01-01", periods=10, freq="B"),
+            ).abs()
+        else:
+            df = DataFrame(
+                np.random.default_rng(2).standard_normal((10, 4)),
+                columns=Index(list("ABCD"), dtype=object),
+                index=date_range("2000-01-01", periods=10, freq="B"),
+            )
+
+        # Use a weakref so we can see if the object gets collected without
+        # also preventing it from being collected
+        ref = weakref.ref(df.plot(kind=kind, **args))
+
+        # have matplotlib delete all the figures
+        plt.close("all")
+        # force a garbage collection
+        gc.collect()
+        assert ref() is None
+
+    def test_df_gridspec_patterns_vert_horiz(self):
+        # GH 10819
+        ts = Series(
+            np.random.default_rng(2).standard_normal(10),
+            index=date_range("1/1/2000", periods=10),
+        )
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=ts.index,
+            columns=list("AB"),
+        )
+
+        def _get_vertical_grid():
+            gs = mpl.gridspec.GridSpec(3, 1)
+            fig = plt.figure()
+            ax1 = fig.add_subplot(gs[:2, :])
+            ax2 = fig.add_subplot(gs[2, :])
+            return ax1, ax2
+
+        def _get_horizontal_grid():
+            gs = mpl.gridspec.GridSpec(1, 3)
+            fig = plt.figure()
+            ax1 = fig.add_subplot(gs[:, :2])
+            ax2 = fig.add_subplot(gs[:, 2])
+            return ax1, ax2
+
+        for ax1, ax2 in [_get_vertical_grid(), _get_horizontal_grid()]:
+            ax1 = ts.plot(ax=ax1)
+            assert len(ax1.lines) == 1
+            ax2 = df.plot(ax=ax2)
+            assert len(ax2.lines) == 2
+            for ax in [ax1, ax2]:
+                _check_visible(ax.get_yticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(minor=True), visible=True)
+            plt.close("all")
+
+        # subplots=True
+        for ax1, ax2 in [_get_vertical_grid(), _get_horizontal_grid()]:
+            axes = df.plot(subplots=True, ax=[ax1, ax2])
+            assert len(ax1.lines) == 1
+            assert len(ax2.lines) == 1
+            for ax in axes:
+                _check_visible(ax.get_yticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(), visible=True)
+                _check_visible(ax.get_xticklabels(minor=True), visible=True)
+            plt.close("all")
+
+        # vertical / subplots / sharex=True / sharey=True
+        ax1, ax2 = _get_vertical_grid()
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=[ax1, ax2], sharex=True, sharey=True)
+        assert len(axes[0].lines) == 1
+        assert len(axes[1].lines) == 1
+        for ax in [ax1, ax2]:
+            # yaxis are visible because there is only one column
+            _check_visible(ax.get_yticklabels(), visible=True)
+        # xaxis of axes0 (top) are hidden
+        _check_visible(axes[0].get_xticklabels(), visible=False)
+        _check_visible(axes[0].get_xticklabels(minor=True), visible=False)
+        _check_visible(axes[1].get_xticklabels(), visible=True)
+        _check_visible(axes[1].get_xticklabels(minor=True), visible=True)
+        plt.close("all")
+
+        # horizontal / subplots / sharex=True / sharey=True
+        ax1, ax2 = _get_horizontal_grid()
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=[ax1, ax2], sharex=True, sharey=True)
+        assert len(axes[0].lines) == 1
+        assert len(axes[1].lines) == 1
+        _check_visible(axes[0].get_yticklabels(), visible=True)
+        # yaxis of axes1 (right) are hidden
+        _check_visible(axes[1].get_yticklabels(), visible=False)
+        for ax in [ax1, ax2]:
+            # xaxis are visible because there is only one column
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+        plt.close("all")
+
+    def test_df_gridspec_patterns_boxed(self):
+        # GH 10819
+        ts = Series(
+            np.random.default_rng(2).standard_normal(10),
+            index=date_range("1/1/2000", periods=10),
+        )
+
+        # boxed
+        def _get_boxed_grid():
+            gs = mpl.gridspec.GridSpec(3, 3)
+            fig = plt.figure()
+            ax1 = fig.add_subplot(gs[:2, :2])
+            ax2 = fig.add_subplot(gs[:2, 2])
+            ax3 = fig.add_subplot(gs[2, :2])
+            ax4 = fig.add_subplot(gs[2, 2])
+            return ax1, ax2, ax3, ax4
+
+        axes = _get_boxed_grid()
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            index=ts.index,
+            columns=list("ABCD"),
+        )
+        axes = df.plot(subplots=True, ax=axes)
+        for ax in axes:
+            assert len(ax.lines) == 1
+            # axis are visible because these are not shared
+            _check_visible(ax.get_yticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+        plt.close("all")
+
+        # subplots / sharex=True / sharey=True
+        axes = _get_boxed_grid()
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=axes, sharex=True, sharey=True)
+        for ax in axes:
+            assert len(ax.lines) == 1
+        for ax in [axes[0], axes[2]]:  # left column
+            _check_visible(ax.get_yticklabels(), visible=True)
+        for ax in [axes[1], axes[3]]:  # right column
+            _check_visible(ax.get_yticklabels(), visible=False)
+        for ax in [axes[0], axes[1]]:  # top row
+            _check_visible(ax.get_xticklabels(), visible=False)
+            _check_visible(ax.get_xticklabels(minor=True), visible=False)
+        for ax in [axes[2], axes[3]]:  # bottom row
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+        plt.close("all")
+
+    def test_df_grid_settings(self):
+        # Make sure plot defaults to rcParams['axes.grid'] setting, GH 9792
+        _check_grid_settings(
+            DataFrame({"a": [1, 2, 3], "b": [2, 3, 4]}),
+            plotting.PlotAccessor._dataframe_kinds,
+            kws={"x": "a", "y": "b"},
+        )
+
+    def test_plain_axes(self):
+        # supplied ax itself is a SubplotAxes, but figure contains also
+        # a plain Axes object (GH11556)
+        fig, ax = mpl.pyplot.subplots()
+        fig.add_axes([0.2, 0.2, 0.2, 0.2])
+        Series(np.random.default_rng(2).random(10)).plot(ax=ax)
+
+    def test_plain_axes_df(self):
+        # supplied ax itself is a plain Axes, but because the cmap keyword
+        # a new ax is created for the colorbar -> also multiples axes (GH11520)
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(8),
+                "b": np.random.default_rng(2).standard_normal(8),
+            }
+        )
+        fig = mpl.pyplot.figure()
+        ax = fig.add_axes((0, 0, 1, 1))
+        df.plot(kind="scatter", ax=ax, x="a", y="b", c="a", cmap="hsv")
+
+    def test_plain_axes_make_axes_locatable(self):
+        # other examples
+        fig, ax = mpl.pyplot.subplots()
+        from mpl_toolkits.axes_grid1 import make_axes_locatable
+
+        divider = make_axes_locatable(ax)
+        cax = divider.append_axes("right", size="5%", pad=0.05)
+        Series(np.random.default_rng(2).random(10)).plot(ax=ax)
+        Series(np.random.default_rng(2).random(10)).plot(ax=cax)
+
+    def test_plain_axes_make_inset_axes(self):
+        fig, ax = mpl.pyplot.subplots()
+        from mpl_toolkits.axes_grid1.inset_locator import inset_axes
+
+        iax = inset_axes(ax, width="30%", height=1.0, loc=3)
+        Series(np.random.default_rng(2).random(10)).plot(ax=ax)
+        Series(np.random.default_rng(2).random(10)).plot(ax=iax)
+
+    @pytest.mark.parametrize("method", ["line", "barh", "bar"])
+    def test_secondary_axis_font_size(self, method):
+        # GH: 12565
+        df = (
+            DataFrame(
+                np.random.default_rng(2).standard_normal((15, 2)), columns=list("AB")
+            )
+            .assign(C=lambda df: df.B.cumsum())
+            .assign(D=lambda df: df.C * 1.1)
+        )
+
+        fontsize = 20
+        sy = ["C", "D"]
+
+        kwargs = {"secondary_y": sy, "fontsize": fontsize, "mark_right": True}
+        ax = getattr(df.plot, method)(**kwargs)
+        _check_ticks_props(axes=ax.right_ax, ylabelsize=fontsize)
+
+    def test_x_string_values_ticks(self):
+        # Test if string plot index have a fixed xtick position
+        # GH: 7612, GH: 22334
+        df = DataFrame(
+            {
+                "sales": [3, 2, 3],
+                "visits": [20, 42, 28],
+                "day": ["Monday", "Tuesday", "Wednesday"],
+            }
+        )
+        ax = df.plot.area(x="day")
+        ax.set_xlim(-1, 3)
+        xticklabels = [t.get_text() for t in ax.get_xticklabels()]
+        labels_position = dict(zip(xticklabels, ax.get_xticks()))
+        # Testing if the label stayed at the right position
+        assert labels_position["Monday"] == 0.0
+        assert labels_position["Tuesday"] == 1.0
+        assert labels_position["Wednesday"] == 2.0
+
+    def test_x_multiindex_values_ticks(self):
+        # Test if multiindex plot index have a fixed xtick position
+        # GH: 15912
+        index = MultiIndex.from_product([[2012, 2013], [1, 2]])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)),
+            columns=["A", "B"],
+            index=index,
+        )
+        ax = df.plot()
+        ax.set_xlim(-1, 4)
+        xticklabels = [t.get_text() for t in ax.get_xticklabels()]
+        labels_position = dict(zip(xticklabels, ax.get_xticks()))
+        # Testing if the label stayed at the right position
+        assert labels_position["(2012, 1)"] == 0.0
+        assert labels_position["(2012, 2)"] == 1.0
+        assert labels_position["(2013, 1)"] == 2.0
+        assert labels_position["(2013, 2)"] == 3.0
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_xlim_plot_line(self, kind):
+        # test if xlim is set correctly in plot.line and plot.area
+        # GH 27686
+        df = DataFrame([2, 4], index=[1, 2])
+        ax = df.plot(kind=kind)
+        xlims = ax.get_xlim()
+        assert xlims[0] < 1
+        assert xlims[1] > 2
+
+    def test_xlim_plot_line_correctly_in_mixed_plot_type(self):
+        # test if xlim is set correctly when ax contains multiple different kinds
+        # of plots, GH 27686
+        fig, ax = mpl.pyplot.subplots()
+
+        indexes = ["k1", "k2", "k3", "k4"]
+        df = DataFrame(
+            {
+                "s1": [1000, 2000, 1500, 2000],
+                "s2": [900, 1400, 2000, 3000],
+                "s3": [1500, 1500, 1600, 1200],
+                "secondary_y": [1, 3, 4, 3],
+            },
+            index=indexes,
+        )
+        df[["s1", "s2", "s3"]].plot.bar(ax=ax, stacked=False)
+        df[["secondary_y"]].plot(ax=ax, secondary_y=True)
+
+        xlims = ax.get_xlim()
+        assert xlims[0] < 0
+        assert xlims[1] > 3
+
+        # make sure axis labels are plotted correctly as well
+        xticklabels = [t.get_text() for t in ax.get_xticklabels()]
+        assert xticklabels == indexes
+
+    def test_plot_no_rows(self):
+        # GH 27758
+        df = DataFrame(columns=["foo"], dtype=int)
+        assert df.empty
+        ax = df.plot()
+        assert len(ax.get_lines()) == 1
+        line = ax.get_lines()[0]
+        assert len(line.get_xdata()) == 0
+        assert len(line.get_ydata()) == 0
+
+    def test_plot_no_numeric_data(self):
+        df = DataFrame(["a", "b", "c"])
+        with pytest.raises(TypeError, match="no numeric data to plot"):
+            df.plot()
+
+    @pytest.mark.parametrize(
+        "kind", ("line", "bar", "barh", "hist", "kde", "density", "area", "pie")
+    )
+    def test_group_subplot(self, kind):
+        pytest.importorskip("scipy")
+        d = {
+            "a": np.arange(10),
+            "b": np.arange(10) + 1,
+            "c": np.arange(10) + 1,
+            "d": np.arange(10),
+            "e": np.arange(10),
+        }
+        df = DataFrame(d)
+
+        axes = df.plot(subplots=[("b", "e"), ("c", "d")], kind=kind)
+        assert len(axes) == 3  # 2 groups + single column a
+
+        expected_labels = (["b", "e"], ["c", "d"], ["a"])
+        for ax, labels in zip(axes, expected_labels):
+            if kind != "pie":
+                _check_legend_labels(ax, labels=labels)
+            if kind == "line":
+                assert len(ax.lines) == len(labels)
+
+    def test_group_subplot_series_notimplemented(self):
+        ser = Series(range(1))
+        msg = "An iterable subplots for a Series"
+        with pytest.raises(NotImplementedError, match=msg):
+            ser.plot(subplots=[("a",)])
+
+    def test_group_subplot_multiindex_notimplemented(self):
+        df = DataFrame(np.eye(2), columns=MultiIndex.from_tuples([(0, 1), (1, 2)]))
+        msg = "An iterable subplots for a DataFrame with a MultiIndex"
+        with pytest.raises(NotImplementedError, match=msg):
+            df.plot(subplots=[(0, 1)])
+
+    def test_group_subplot_nonunique_cols_notimplemented(self):
+        df = DataFrame(np.eye(2), columns=["a", "a"])
+        msg = "An iterable subplots for a DataFrame with non-unique"
+        with pytest.raises(NotImplementedError, match=msg):
+            df.plot(subplots=[("a",)])
+
+    @pytest.mark.parametrize(
+        "subplots, expected_msg",
+        [
+            (123, "subplots should be a bool or an iterable"),
+            ("a", "each entry should be a list/tuple"),  # iterable of non-iterable
+            ((1,), "each entry should be a list/tuple"),  # iterable of non-iterable
+            (("a",), "each entry should be a list/tuple"),  # iterable of strings
+        ],
+    )
+    def test_group_subplot_bad_input(self, subplots, expected_msg):
+        # Make sure error is raised when subplots is not a properly
+        # formatted iterable. Only iterables of iterables are permitted, and
+        # entries should not be strings.
+        d = {"a": np.arange(10), "b": np.arange(10)}
+        df = DataFrame(d)
+
+        with pytest.raises(ValueError, match=expected_msg):
+            df.plot(subplots=subplots)
+
+    def test_group_subplot_invalid_column_name(self):
+        d = {"a": np.arange(10), "b": np.arange(10)}
+        df = DataFrame(d)
+
+        if Version(np.__version__) < Version("2.0.0"):
+            with pytest.raises(ValueError, match=r"Column label\(s\) \['bad_name'\]"):
+                df.plot(subplots=[("a", "bad_name")])
+        else:
+            with pytest.raises(
+                ValueError, match=r"Column label\(s\) \[np\.str\_\('bad_name'\)\]"
+            ):
+                df.plot(subplots=[("a", "bad_name")])
+
+    def test_group_subplot_duplicated_column(self):
+        d = {"a": np.arange(10), "b": np.arange(10), "c": np.arange(10)}
+        df = DataFrame(d)
+
+        with pytest.raises(ValueError, match="should be in only one subplot"):
+            df.plot(subplots=[("a", "b"), ("a", "c")])
+
+    @pytest.mark.parametrize("kind", ("box", "scatter", "hexbin"))
+    def test_group_subplot_invalid_kind(self, kind):
+        d = {"a": np.arange(10), "b": np.arange(10)}
+        df = DataFrame(d)
+        with pytest.raises(
+            ValueError, match="When subplots is an iterable, kind must be one of"
+        ):
+            df.plot(subplots=[("a", "b")], kind=kind)
+
+    @pytest.mark.parametrize(
+        "index_name, old_label, new_label",
+        [
+            (None, "", "new"),
+            ("old", "old", "new"),
+            (None, "", ""),
+            (None, "", 1),
+            (None, "", [1, 2]),
+        ],
+    )
+    @pytest.mark.parametrize("kind", ["line", "area", "bar"])
+    def test_xlabel_ylabel_dataframe_single_plot(
+        self, kind, index_name, old_label, new_label
+    ):
+        # GH 9093
+        df = DataFrame([[1, 2], [2, 5]], columns=["Type A", "Type B"])
+        df.index.name = index_name
+
+        # default is the ylabel is not shown and xlabel is index name
+        ax = df.plot(kind=kind)
+        assert ax.get_xlabel() == old_label
+        assert ax.get_ylabel() == ""
+
+        # old xlabel will be overridden and assigned ylabel will be used as ylabel
+        ax = df.plot(kind=kind, ylabel=new_label, xlabel=new_label)
+        assert ax.get_ylabel() == str(new_label)
+        assert ax.get_xlabel() == str(new_label)
+
+    @pytest.mark.parametrize(
+        "xlabel, ylabel",
+        [
+            (None, None),
+            ("X Label", None),
+            (None, "Y Label"),
+            ("X Label", "Y Label"),
+        ],
+    )
+    @pytest.mark.parametrize("kind", ["scatter", "hexbin"])
+    def test_xlabel_ylabel_dataframe_plane_plot(self, kind, xlabel, ylabel):
+        # GH 37001
+        xcol = "Type A"
+        ycol = "Type B"
+        df = DataFrame([[1, 2], [2, 5]], columns=[xcol, ycol])
+
+        # default is the labels are column names
+        ax = df.plot(kind=kind, x=xcol, y=ycol, xlabel=xlabel, ylabel=ylabel)
+        assert ax.get_xlabel() == (xcol if xlabel is None else xlabel)
+        assert ax.get_ylabel() == (ycol if ylabel is None else ylabel)
+
+    @pytest.mark.parametrize("secondary_y", (False, True))
+    def test_secondary_y(self, secondary_y):
+        ax_df = DataFrame([0]).plot(
+            secondary_y=secondary_y, ylabel="Y", ylim=(0, 100), yticks=[99]
+        )
+        for ax in ax_df.figure.axes:
+            if ax.yaxis.get_visible():
+                assert ax.get_ylabel() == "Y"
+                assert ax.get_ylim() == (0, 100)
+                assert ax.get_yticks()[0] == 99
+
+    @pytest.mark.slow
+    def test_plot_no_warning(self):
+        # GH 55138
+        # TODO(3.0): this can be removed once Period[B] deprecation is enforced
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        with tm.assert_produces_warning(False):
+            _ = df.plot()
+            _ = df.T.plot()
+
+    @pytest.mark.parametrize("freq", ["h", "7h", "60min", "120min", "3M"])
+    def test_plot_period_index_makes_no_right_shift(self, freq):
+        # GH#57587
+        idx = pd.period_range("01/01/2000", freq=freq, periods=4)
+        df = DataFrame(
+            np.array([0, 1, 0, 1]),
+            index=idx,
+            columns=["A"],
+        )
+        expected = idx.values
+
+        ax = df.plot()
+        result = ax.get_lines()[0].get_xdata()
+        assert all(str(result[i]) == str(expected[i]) for i in range(4))
+
+    def test_plot_display_xlabel_and_xticks(self):
+        # GH#44050
+        df = DataFrame(np.random.default_rng(2).random((10, 2)), columns=["a", "b"])
+        ax = df.plot.hexbin(x="a", y="b")
+
+        _check_visible([ax.xaxis.get_label()], visible=True)
+        _check_visible(ax.get_xticklabels(), visible=True)
+
+
+def _generate_4_axes_via_gridspec():
+    gs = mpl.gridspec.GridSpec(2, 2)
+    ax_tl = plt.subplot(gs[0, 0])
+    ax_ll = plt.subplot(gs[1, 0])
+    ax_tr = plt.subplot(gs[0, 1])
+    ax_lr = plt.subplot(gs[1, 1])
+
+    return gs, [ax_tl, ax_ll, ax_tr, ax_lr]
diff --git a/pandas/pandas/tests/plotting/frame/test_frame_color.py b/pandas/pandas/tests/plotting/frame/test_frame_color.py
new file mode 100644
index 0000000000000..74ee45664e01a
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_frame_color.py
@@ -0,0 +1,738 @@
+"""Test cases for DataFrame.plot"""
+
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_colors,
+    _check_plot_works,
+    _unpack_cycler,
+)
+from pandas.util.version import Version
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+cm = pytest.importorskip("matplotlib.cm")
+
+
+def _check_colors_box(bp, box_c, whiskers_c, medians_c, caps_c="k", fliers_c=None):
+    if fliers_c is None:
+        fliers_c = "k"
+    _check_colors(bp["boxes"], linecolors=[box_c] * len(bp["boxes"]))
+    _check_colors(bp["whiskers"], linecolors=[whiskers_c] * len(bp["whiskers"]))
+    _check_colors(bp["medians"], linecolors=[medians_c] * len(bp["medians"]))
+    _check_colors(bp["fliers"], linecolors=[fliers_c] * len(bp["fliers"]))
+    _check_colors(bp["caps"], linecolors=[caps_c] * len(bp["caps"]))
+
+
+class TestDataFrameColor:
+    @pytest.mark.parametrize("color", list(range(10)))
+    def test_mpl2_color_cycle_str(self, color):
+        # GH 15516
+        color = f"C{color}"
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=["a", "b", "c"]
+        )
+        _check_plot_works(df.plot, color=color)
+
+    def test_color_single_series_list(self):
+        # GH 3486
+        df = DataFrame({"A": [1, 2, 3]})
+        _check_plot_works(df.plot, color=["red"])
+
+    @pytest.mark.parametrize("color", [(1, 0, 0), (1, 0, 0, 0.5)])
+    def test_rgb_tuple_color(self, color):
+        # GH 16695
+        df = DataFrame({"x": [1, 2], "y": [3, 4]})
+        _check_plot_works(df.plot, x="x", y="y", color=color)
+
+    def test_color_empty_string(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        with pytest.raises(ValueError, match="Invalid color argument:"):
+            df.plot(color="")
+
+    def test_color_and_style_arguments(self):
+        df = DataFrame({"x": [1, 2], "y": [3, 4]})
+        # passing both 'color' and 'style' arguments should be allowed
+        # if there is no color symbol in the style strings:
+        ax = df.plot(color=["red", "black"], style=["-", "--"])
+        # check that the linestyles are correctly set:
+        linestyle = [line.get_linestyle() for line in ax.lines]
+        assert linestyle == ["-", "--"]
+        # check that the colors are correctly set:
+        color = [line.get_color() for line in ax.lines]
+        assert color == ["red", "black"]
+        # passing both 'color' and 'style' arguments should not be allowed
+        # if there is a color symbol in the style strings:
+        msg = (
+            "Cannot pass 'style' string with a color symbol and 'color' keyword "
+            "argument. Please use one or the other or pass 'style' without a color "
+            "symbol"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.plot(color=["red", "black"], style=["k-", "r--"])
+
+    @pytest.mark.parametrize(
+        "color, expected",
+        [
+            ("green", ["green"] * 4),
+            (["yellow", "red", "green", "blue"], ["yellow", "red", "green", "blue"]),
+        ],
+    )
+    def test_color_and_marker(self, color, expected):
+        # GH 21003
+        df = DataFrame(np.random.default_rng(2).random((7, 4)))
+        ax = df.plot(color=color, style="d--")
+        # check colors
+        result = [i.get_color() for i in ax.lines]
+        assert result == expected
+        # check markers and linestyles
+        assert all(i.get_linestyle() == "--" for i in ax.lines)
+        assert all(i.get_marker() == "d" for i in ax.lines)
+
+    def test_color_and_style(self):
+        color = {"g": "black", "h": "brown"}
+        style = {"g": "-", "h": "--"}
+        expected_color = ["black", "brown"]
+        expected_style = ["-", "--"]
+        df = DataFrame({"g": [1, 2], "h": [2, 3]}, index=[1, 2])
+        ax = df.plot.line(color=color, style=style)
+        color = [i.get_color() for i in ax.lines]
+        style = [i.get_linestyle() for i in ax.lines]
+        assert color == expected_color
+        assert style == expected_style
+
+    def test_bar_colors(self):
+        default_colors = _unpack_cycler(plt.rcParams)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.bar()
+        _check_colors(ax.patches[::5], facecolors=default_colors[:5])
+
+    def test_bar_colors_custom(self):
+        custom_colors = "rgcby"
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.bar(color=custom_colors)
+        _check_colors(ax.patches[::5], facecolors=custom_colors)
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_bar_colors_cmap(self, colormap):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        ax = df.plot.bar(colormap=colormap)
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, 5)]
+        _check_colors(ax.patches[::5], facecolors=rgba_colors)
+
+    def test_bar_colors_single_col(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.loc[:, [0]].plot.bar(color="DodgerBlue")
+        _check_colors([ax.patches[0]], facecolors=["DodgerBlue"])
+
+    def test_bar_colors_green(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot(kind="bar", color="green")
+        _check_colors(ax.patches[::5], facecolors=["green"] * 5)
+
+    def test_bar_user_colors(self):
+        df = DataFrame(
+            {"A": range(4), "B": range(1, 5), "color": ["red", "blue", "blue", "red"]}
+        )
+        # This should *only* work when `y` is specified, else
+        # we use one color per column
+        ax = df.plot.bar(y="A", color=df["color"])
+        result = [p.get_facecolor() for p in ax.patches]
+        expected = [
+            (1.0, 0.0, 0.0, 1.0),
+            (0.0, 0.0, 1.0, 1.0),
+            (0.0, 0.0, 1.0, 1.0),
+            (1.0, 0.0, 0.0, 1.0),
+        ]
+        assert result == expected
+
+    def test_if_scatterplot_colorbar_affects_xaxis_visibility(self):
+        # addressing issue #10611, to ensure colobar does not
+        # interfere with x-axis label and ticklabels with
+        # ipython inline backend.
+        random_array = np.random.default_rng(2).random((10, 3))
+        df = DataFrame(random_array, columns=["A label", "B label", "C label"])
+
+        ax1 = df.plot.scatter(x="A label", y="B label")
+        ax2 = df.plot.scatter(x="A label", y="B label", c="C label")
+
+        vis1 = [vis.get_visible() for vis in ax1.xaxis.get_minorticklabels()]
+        vis2 = [vis.get_visible() for vis in ax2.xaxis.get_minorticklabels()]
+        assert vis1 == vis2
+
+        vis1 = [vis.get_visible() for vis in ax1.xaxis.get_majorticklabels()]
+        vis2 = [vis.get_visible() for vis in ax2.xaxis.get_majorticklabels()]
+        assert vis1 == vis2
+
+        assert (
+            ax1.xaxis.get_label().get_visible() == ax2.xaxis.get_label().get_visible()
+        )
+
+    def test_if_hexbin_xaxis_label_is_visible(self):
+        # addressing issue #10678, to ensure colobar does not
+        # interfere with x-axis label and ticklabels with
+        # ipython inline backend.
+        random_array = np.random.default_rng(2).random((10, 3))
+        df = DataFrame(random_array, columns=["A label", "B label", "C label"])
+
+        ax = df.plot.hexbin("A label", "B label", gridsize=12)
+        assert all(vis.get_visible() for vis in ax.xaxis.get_minorticklabels())
+        assert all(vis.get_visible() for vis in ax.xaxis.get_majorticklabels())
+        assert ax.xaxis.get_label().get_visible()
+
+    def test_if_scatterplot_colorbars_are_next_to_parent_axes(self):
+        random_array = np.random.default_rng(2).random((10, 3))
+        df = DataFrame(random_array, columns=["A label", "B label", "C label"])
+
+        fig, axes = plt.subplots(1, 2)
+        df.plot.scatter("A label", "B label", c="C label", ax=axes[0])
+        df.plot.scatter("A label", "B label", c="C label", ax=axes[1])
+        plt.tight_layout()
+
+        points = np.array([ax.get_position().get_points() for ax in fig.axes])
+        axes_x_coords = points[:, :, 0]
+        parent_distance = axes_x_coords[1, :] - axes_x_coords[0, :]
+        colorbar_distance = axes_x_coords[3, :] - axes_x_coords[2, :]
+        assert np.isclose(parent_distance, colorbar_distance, atol=1e-7).all()
+
+    @pytest.mark.parametrize("cmap", [None, "Greys"])
+    def test_scatter_with_c_column_name_with_colors(self, cmap):
+        # https://github.com/pandas-dev/pandas/issues/34316
+
+        df = DataFrame(
+            [[5.1, 3.5], [4.9, 3.0], [7.0, 3.2], [6.4, 3.2], [5.9, 3.0]],
+            columns=["length", "width"],
+        )
+        df["species"] = ["r", "r", "g", "g", "b"]
+        if cmap is not None:
+            with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+                ax = df.plot.scatter(x=0, y=1, cmap=cmap, c="species")
+        else:
+            ax = df.plot.scatter(x=0, y=1, c="species", cmap=cmap)
+
+        assert len(np.unique(ax.collections[0].get_facecolor(), axis=0)) == 3  # r/g/b
+        assert (
+            np.unique(ax.collections[0].get_facecolor(), axis=0)
+            == np.array(
+                [[0.0, 0.0, 1.0, 1.0], [0.0, 0.5, 0.0, 1.0], [1.0, 0.0, 0.0, 1.0]]
+            )  # r/g/b
+        ).all()
+        assert ax.collections[0].colorbar is None
+
+    def test_scatter_with_c_column_name_without_colors(self):
+        # Given
+        colors = ["NY", "MD", "MA", "CA"]
+        color_count = 4  # 4 unique colors
+
+        # When
+        df = DataFrame(
+            {
+                "dataX": range(100),
+                "dataY": range(100),
+                "color": (colors[i % len(colors)] for i in range(100)),
+            }
+        )
+
+        # Then
+        ax = df.plot.scatter("dataX", "dataY", c="color")
+        assert len(np.unique(ax.collections[0].get_facecolor(), axis=0)) == color_count
+
+        # Given
+        colors = ["r", "g", "not-a-color"]
+        color_count = 3
+        # Also, since not all are mpl-colors, points matching 'r' or 'g'
+        # are not necessarily red or green
+
+        # When
+        df = DataFrame(
+            {
+                "dataX": range(100),
+                "dataY": range(100),
+                "color": (colors[i % len(colors)] for i in range(100)),
+            }
+        )
+
+        # Then
+        ax = df.plot.scatter("dataX", "dataY", c="color")
+        assert len(np.unique(ax.collections[0].get_facecolor(), axis=0)) == color_count
+
+    def test_scatter_colors(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3], "c": [1, 2, 3]})
+        with pytest.raises(TypeError, match="Specify exactly one of `c` and `color`"):
+            df.plot.scatter(x="a", y="b", c="c", color="green")
+
+    def test_scatter_colors_not_raising_warnings(self):
+        # GH-53908. Do not raise UserWarning: No data for colormapping
+        # provided via 'c'. Parameters 'cmap' will be ignored
+        df = DataFrame({"x": [1, 2, 3], "y": [1, 2, 3]})
+        with tm.assert_produces_warning(None):
+            ax = df.plot.scatter(x="x", y="y", c="b")
+            assert (
+                len(np.unique(ax.collections[0].get_facecolor(), axis=0)) == 1
+            )  # blue
+            assert (
+                np.unique(ax.collections[0].get_facecolor(), axis=0)
+                == np.array([[0.0, 0.0, 1.0, 1.0]])
+            ).all()  # blue
+
+    def test_scatter_colors_default(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3], "c": [1, 2, 3]})
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+
+        ax = df.plot.scatter(x="a", y="b", c="c")
+        tm.assert_numpy_array_equal(
+            ax.collections[0].get_facecolor()[0],
+            np.array(mpl.colors.ColorConverter.to_rgba(default_colors[0])),
+        )
+
+    def test_scatter_colors_white(self):
+        df = DataFrame({"a": [1, 2, 3], "b": [1, 2, 3], "c": [1, 2, 3]})
+        ax = df.plot.scatter(x="a", y="b", color="white")
+        tm.assert_numpy_array_equal(
+            ax.collections[0].get_facecolor()[0],
+            np.array([1, 1, 1, 1], dtype=np.float64),
+        )
+
+    def test_scatter_colorbar_different_cmap(self):
+        # GH 33389
+        df = DataFrame({"x": [1, 2, 3], "y": [1, 3, 2], "c": [1, 2, 3]})
+        df["x2"] = df["x"] + 1
+
+        _, ax = plt.subplots()
+        df.plot("x", "y", c="c", kind="scatter", cmap="cividis", ax=ax)
+        df.plot("x2", "y", c="c", kind="scatter", cmap="magma", ax=ax)
+
+        assert ax.collections[0].cmap.name == "cividis"
+        assert ax.collections[1].cmap.name == "magma"
+
+    def test_line_colors(self):
+        custom_colors = "rgcby"
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        ax = df.plot(color=custom_colors)
+        _check_colors(ax.get_lines(), linecolors=custom_colors)
+
+        plt.close("all")
+
+        ax2 = df.plot(color=custom_colors)
+        lines2 = ax2.get_lines()
+
+        for l1, l2 in zip(ax.get_lines(), lines2):
+            assert l1.get_color() == l2.get_color()
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_line_colors_cmap(self, colormap):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot(colormap=colormap)
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        _check_colors(ax.get_lines(), linecolors=rgba_colors)
+
+    def test_line_colors_single_col(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # make color a list if plotting one column frame
+        # handles cases like df.plot(color='DodgerBlue')
+        ax = df.loc[:, [0]].plot(color="DodgerBlue")
+        _check_colors(ax.lines, linecolors=["DodgerBlue"])
+
+    def test_line_colors_single_color(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot(color="red")
+        _check_colors(ax.get_lines(), linecolors=["red"] * 5)
+
+    def test_line_colors_hex(self):
+        # GH 10299
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        custom_colors = ["#FF0000", "#0000FF", "#FFFF00", "#000000", "#FFFFFF"]
+        ax = df.plot(color=custom_colors)
+        _check_colors(ax.get_lines(), linecolors=custom_colors)
+
+    def test_dont_modify_colors(self):
+        colors = ["r", "g", "b"]
+        DataFrame(np.random.default_rng(2).random((10, 2))).plot(color=colors)
+        assert len(colors) == 3
+
+    def test_line_colors_and_styles_subplots(self):
+        # GH 9894
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        axes = df.plot(subplots=True)
+        for ax, c in zip(axes, list(default_colors)):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    @pytest.mark.parametrize("color", ["k", "green"])
+    def test_line_colors_and_styles_subplots_single_color_str(self, color):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        axes = df.plot(subplots=True, color=color)
+        for ax in axes:
+            _check_colors(ax.get_lines(), linecolors=[color])
+
+    @pytest.mark.parametrize("color", ["rgcby", list("rgcby")])
+    def test_line_colors_and_styles_subplots_custom_colors(self, color):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        axes = df.plot(color=color, subplots=True)
+        for ax, c in zip(axes, list(color)):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    def test_line_colors_and_styles_subplots_colormap_hex(self):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # GH 10299
+        custom_colors = ["#FF0000", "#0000FF", "#FFFF00", "#000000", "#FFFFFF"]
+        axes = df.plot(color=custom_colors, subplots=True)
+        for ax, c in zip(axes, list(custom_colors)):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    @pytest.mark.parametrize("cmap", ["jet", cm.jet])
+    def test_line_colors_and_styles_subplots_colormap_subplot(self, cmap):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        axes = df.plot(colormap=cmap, subplots=True)
+        for ax, c in zip(axes, rgba_colors):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    def test_line_colors_and_styles_subplots_single_col(self):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # make color a list if plotting one column frame
+        # handles cases like df.plot(color='DodgerBlue')
+        axes = df.loc[:, [0]].plot(color="DodgerBlue", subplots=True)
+        _check_colors(axes[0].lines, linecolors=["DodgerBlue"])
+
+    def test_line_colors_and_styles_subplots_single_char(self):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # single character style
+        axes = df.plot(style="r", subplots=True)
+        for ax in axes:
+            _check_colors(ax.get_lines(), linecolors=["r"])
+
+    def test_line_colors_and_styles_subplots_list_styles(self):
+        # GH 9894
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # list of styles
+        styles = list("rgcby")
+        axes = df.plot(style=styles, subplots=True)
+        for ax, c in zip(axes, styles):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    def test_area_colors(self):
+        custom_colors = "rgcby"
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+
+        ax = df.plot.area(color=custom_colors)
+        _check_colors(ax.get_lines(), linecolors=custom_colors)
+        poly = [
+            o
+            for o in ax.get_children()
+            if isinstance(o, mpl.collections.PolyCollection)
+        ]
+        _check_colors(poly, facecolors=custom_colors)
+
+        handles, _ = ax.get_legend_handles_labels()
+        _check_colors(handles, facecolors=custom_colors)
+
+        for h in handles:
+            assert h.get_alpha() is None
+
+    def test_area_colors_poly(self):
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+        ax = df.plot.area(colormap="jet")
+        jet_colors = [mpl.cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        _check_colors(ax.get_lines(), linecolors=jet_colors)
+        poly = [
+            o
+            for o in ax.get_children()
+            if isinstance(o, mpl.collections.PolyCollection)
+        ]
+        _check_colors(poly, facecolors=jet_colors)
+
+        handles, _ = ax.get_legend_handles_labels()
+        _check_colors(handles, facecolors=jet_colors)
+        for h in handles:
+            assert h.get_alpha() is None
+
+    def test_area_colors_stacked_false(self):
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+        jet_colors = [mpl.cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        # When stacked=False, alpha is set to 0.5
+        ax = df.plot.area(colormap=mpl.cm.jet, stacked=False)
+        _check_colors(ax.get_lines(), linecolors=jet_colors)
+        poly = [
+            o
+            for o in ax.get_children()
+            if isinstance(o, mpl.collections.PolyCollection)
+        ]
+        jet_with_alpha = [(c[0], c[1], c[2], 0.5) for c in jet_colors]
+        _check_colors(poly, facecolors=jet_with_alpha)
+
+        handles, _ = ax.get_legend_handles_labels()
+        linecolors = jet_with_alpha
+        _check_colors(handles[: len(jet_colors)], linecolors=linecolors)
+        for h in handles:
+            assert h.get_alpha() == 0.5
+
+    def test_hist_colors(self):
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.hist()
+        _check_colors(ax.patches[::10], facecolors=default_colors[:5])
+
+    def test_hist_colors_single_custom(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        custom_colors = "rgcby"
+        ax = df.plot.hist(color=custom_colors)
+        _check_colors(ax.patches[::10], facecolors=custom_colors)
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_hist_colors_cmap(self, colormap):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.hist(colormap=colormap)
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, 5)]
+        _check_colors(ax.patches[::10], facecolors=rgba_colors)
+
+    def test_hist_colors_single_col(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.loc[:, [0]].plot.hist(color="DodgerBlue")
+        _check_colors([ax.patches[0]], facecolors=["DodgerBlue"])
+
+    def test_hist_colors_single_color(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot(kind="hist", color="green")
+        _check_colors(ax.patches[::10], facecolors=["green"] * 5)
+
+    def test_kde_colors(self):
+        pytest.importorskip("scipy")
+        custom_colors = "rgcby"
+        df = DataFrame(np.random.default_rng(2).random((5, 5)))
+
+        ax = df.plot.kde(color=custom_colors)
+        _check_colors(ax.get_lines(), linecolors=custom_colors)
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_kde_colors_cmap(self, colormap):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.kde(colormap=colormap)
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        _check_colors(ax.get_lines(), linecolors=rgba_colors)
+
+    def test_kde_colors_and_styles_subplots(self):
+        pytest.importorskip("scipy")
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+
+        axes = df.plot(kind="kde", subplots=True)
+        for ax, c in zip(axes, list(default_colors)):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    @pytest.mark.parametrize("colormap", ["k", "red"])
+    def test_kde_colors_and_styles_subplots_single_col_str(self, colormap):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        axes = df.plot(kind="kde", color=colormap, subplots=True)
+        for ax in axes:
+            _check_colors(ax.get_lines(), linecolors=[colormap])
+
+    def test_kde_colors_and_styles_subplots_custom_color(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        custom_colors = "rgcby"
+        axes = df.plot(kind="kde", color=custom_colors, subplots=True)
+        for ax, c in zip(axes, list(custom_colors)):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_kde_colors_and_styles_subplots_cmap(self, colormap):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        rgba_colors = [cm.jet(n) for n in np.linspace(0, 1, len(df))]
+        axes = df.plot(kind="kde", colormap=colormap, subplots=True)
+        for ax, c in zip(axes, rgba_colors):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    def test_kde_colors_and_styles_subplots_single_col(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # make color a list if plotting one column frame
+        # handles cases like df.plot(color='DodgerBlue')
+        axes = df.loc[:, [0]].plot(kind="kde", color="DodgerBlue", subplots=True)
+        _check_colors(axes[0].lines, linecolors=["DodgerBlue"])
+
+    def test_kde_colors_and_styles_subplots_single_char(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # list of styles
+        # single character style
+        axes = df.plot(kind="kde", style="r", subplots=True)
+        for ax in axes:
+            _check_colors(ax.get_lines(), linecolors=["r"])
+
+    def test_kde_colors_and_styles_subplots_list(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # list of styles
+        styles = list("rgcby")
+        axes = df.plot(kind="kde", style=styles, subplots=True)
+        for ax, c in zip(axes, styles):
+            _check_colors(ax.get_lines(), linecolors=[c])
+
+    def test_boxplot_colors(self):
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        bp = df.plot.box(return_type="dict")
+        _check_colors_box(
+            bp,
+            default_colors[0],
+            default_colors[0],
+            default_colors[2],
+            default_colors[0],
+        )
+
+    def test_boxplot_colors_dict_colors(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        dict_colors = {
+            "boxes": "#572923",
+            "whiskers": "#982042",
+            "medians": "#804823",
+            "caps": "#123456",
+        }
+        bp = df.plot.box(color=dict_colors, sym="r+", return_type="dict")
+        _check_colors_box(
+            bp,
+            dict_colors["boxes"],
+            dict_colors["whiskers"],
+            dict_colors["medians"],
+            dict_colors["caps"],
+            "r",
+        )
+
+    def test_boxplot_colors_default_color(self):
+        default_colors = _unpack_cycler(mpl.pyplot.rcParams)
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # partial colors
+        dict_colors = {"whiskers": "c", "medians": "m"}
+        bp = df.plot.box(color=dict_colors, return_type="dict")
+        _check_colors_box(bp, default_colors[0], "c", "m", default_colors[0])
+
+    @pytest.mark.parametrize("colormap", ["jet", cm.jet])
+    def test_boxplot_colors_cmap(self, colormap):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        bp = df.plot.box(colormap=colormap, return_type="dict")
+        jet_colors = [cm.jet(n) for n in np.linspace(0, 1, 3)]
+        _check_colors_box(
+            bp, jet_colors[0], jet_colors[0], jet_colors[2], jet_colors[0]
+        )
+
+    def test_boxplot_colors_single(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # string color is applied to all artists except fliers
+        bp = df.plot.box(color="DodgerBlue", return_type="dict")
+        _check_colors_box(bp, "DodgerBlue", "DodgerBlue", "DodgerBlue", "DodgerBlue")
+
+    def test_boxplot_colors_tuple(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # tuple is also applied to all artists except fliers
+        bp = df.plot.box(color=(0, 1, 0), sym="#123456", return_type="dict")
+        _check_colors_box(bp, (0, 1, 0), (0, 1, 0), (0, 1, 0), (0, 1, 0), "#123456")
+
+    def test_boxplot_colors_invalid(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        msg = re.escape(
+            "color dict contains invalid key 'xxxx'. The key must be either "
+            "['boxes', 'whiskers', 'medians', 'caps']"
+        )
+        with pytest.raises(ValueError, match=msg):
+            # Color contains invalid key results in ValueError
+            df.plot.box(color={"boxes": "red", "xxxx": "blue"})
+
+    def test_default_color_cycle(self):
+        import cycler
+
+        colors = list("rgbk")
+        plt.rcParams["axes.prop_cycle"] = cycler.cycler("color", colors)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 3)))
+        ax = df.plot()
+
+        expected = _unpack_cycler(plt.rcParams)[:3]
+        _check_colors(ax.get_lines(), linecolors=expected)
+
+    def test_no_color_bar(self):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+        ax = df.plot.hexbin(x="A", y="B", colorbar=None)
+        assert ax.collections[0].colorbar is None
+
+    def test_mixing_cmap_and_colormap_raises(self):
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).uniform(size=20),
+                "B": np.random.default_rng(2).uniform(size=20),
+                "C": np.arange(20) + np.random.default_rng(2).uniform(size=20),
+            }
+        )
+        msg = "Only specify one of `cmap` and `colormap`"
+        with pytest.raises(TypeError, match=msg):
+            df.plot.hexbin(x="A", y="B", cmap="YlGn", colormap="BuGn")
+
+    def test_passed_bar_colors(self):
+        color_tuples = [(0.9, 0, 0, 1), (0, 0.9, 0, 1), (0, 0, 0.9, 1)]
+        colormap = mpl.colors.ListedColormap(color_tuples)
+        barplot = DataFrame([[1, 2, 3]]).plot(kind="bar", cmap=colormap)
+        assert color_tuples == [c.get_facecolor() for c in barplot.patches]
+
+    def test_rcParams_bar_colors(self):
+        color_tuples = [(0.9, 0, 0, 1), (0, 0.9, 0, 1), (0, 0, 0.9, 1)]
+        with mpl.rc_context(rc={"axes.prop_cycle": mpl.cycler("color", color_tuples)}):
+            barplot = DataFrame([[1, 2, 3]]).plot(kind="bar")
+        assert color_tuples == [c.get_facecolor() for c in barplot.patches]
+
+    def test_colors_of_columns_with_same_name(self):
+        # ISSUE 11136 -> https://github.com/pandas-dev/pandas/issues/11136
+        # Creating a DataFrame with duplicate column labels and testing colors of them.
+        df = DataFrame({"b": [0, 1, 0], "a": [1, 2, 3]})
+        df1 = DataFrame({"a": [2, 4, 6]})
+        df_concat = pd.concat([df, df1], axis=1)
+        result = df_concat.plot()
+        legend = result.get_legend()
+        if Version(mpl.__version__) < Version("3.7"):
+            handles = legend.legendHandles
+        else:
+            handles = legend.legend_handles
+        for legend, line in zip(handles, result.lines):
+            assert legend.get_color() == line.get_color()
+
+    def test_invalid_colormap(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 2)), columns=["A", "B"]
+        )
+        msg = "(is not a valid value)|(is not a known colormap)"
+        with pytest.raises((ValueError, KeyError), match=msg):
+            df.plot(colormap="invalid_colormap")
+
+    def test_dataframe_none_color(self):
+        # GH51953
+        df = DataFrame([[1, 2, 3]])
+        ax = df.plot(color=None)
+        expected = _unpack_cycler(mpl.pyplot.rcParams)[:3]
+        _check_colors(ax.get_lines(), linecolors=expected)
diff --git a/pandas/pandas/tests/plotting/frame/test_frame_groupby.py b/pandas/pandas/tests/plotting/frame/test_frame_groupby.py
new file mode 100644
index 0000000000000..b7e147bbabde5
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_frame_groupby.py
@@ -0,0 +1,72 @@
+"""Test cases for DataFrame.plot"""
+
+import pytest
+
+from pandas import DataFrame
+from pandas.tests.plotting.common import _check_visible
+
+pytest.importorskip("matplotlib")
+
+
+class TestDataFramePlotsGroupby:
+    def _assert_ytickslabels_visibility(self, axes, expected):
+        for ax, exp in zip(axes, expected):
+            _check_visible(ax.get_yticklabels(), visible=exp)
+
+    def _assert_xtickslabels_visibility(self, axes, expected):
+        for ax, exp in zip(axes, expected):
+            _check_visible(ax.get_xticklabels(), visible=exp)
+
+    @pytest.mark.parametrize(
+        "kwargs, expected",
+        [
+            # behavior without keyword
+            ({}, [True, False, True, False]),
+            # set sharey=True should be identical
+            ({"sharey": True}, [True, False, True, False]),
+            # sharey=False, all yticklabels should be visible
+            ({"sharey": False}, [True, True, True, True]),
+        ],
+    )
+    def test_groupby_boxplot_sharey(self, kwargs, expected):
+        # https://github.com/pandas-dev/pandas/issues/20968
+        # sharey can now be switched check whether the right
+        # pair of axes is turned on or off
+        df = DataFrame(
+            {
+                "a": [-1.43, -0.15, -3.70, -1.43, -0.14],
+                "b": [0.56, 0.84, 0.29, 0.56, 0.85],
+                "c": [0, 1, 2, 3, 1],
+            },
+            index=[0, 1, 2, 3, 4],
+        )
+        axes = df.groupby("c").boxplot(**kwargs)
+        self._assert_ytickslabels_visibility(axes, expected)
+
+    @pytest.mark.parametrize(
+        "kwargs, expected",
+        [
+            # behavior without keyword
+            ({}, [True, True, True, True]),
+            # set sharex=False should be identical
+            ({"sharex": False}, [True, True, True, True]),
+            # sharex=True, xticklabels should be visible
+            # only for bottom plots
+            ({"sharex": True}, [False, False, True, True]),
+        ],
+    )
+    def test_groupby_boxplot_sharex(self, kwargs, expected):
+        # https://github.com/pandas-dev/pandas/issues/20968
+        # sharex can now be switched check whether the right
+        # pair of axes is turned on or off
+
+        df = DataFrame(
+            {
+                "a": [-1.43, -0.15, -3.70, -1.43, -0.14],
+                "b": [0.56, 0.84, 0.29, 0.56, 0.85],
+                "c": [0, 1, 2, 3, 1],
+            },
+            index=[0, 1, 2, 3, 4],
+        )
+        axes = df.groupby("c").boxplot(**kwargs)
+        self._assert_xtickslabels_visibility(axes, expected)
diff --git a/pandas/pandas/tests/plotting/frame/test_frame_legend.py b/pandas/pandas/tests/plotting/frame/test_frame_legend.py
new file mode 100644
index 0000000000000..a9723fe4ef871
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_frame_legend.py
@@ -0,0 +1,269 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    date_range,
+)
+from pandas.tests.plotting.common import (
+    _check_legend_labels,
+    _check_legend_marker,
+    _check_text_labels,
+)
+from pandas.util.version import Version
+
+mpl = pytest.importorskip("matplotlib")
+
+
+class TestFrameLegend:
+    @pytest.mark.xfail(
+        reason=(
+            "Open bug in matplotlib "
+            "https://github.com/matplotlib/matplotlib/issues/11357"
+        )
+    )
+    def test_mixed_yerr(self):
+        # https://github.com/pandas-dev/pandas/issues/39522
+        df = DataFrame([{"x": 1, "a": 1, "b": 1}, {"x": 2, "a": 2, "b": 3}])
+
+        ax = df.plot("x", "a", c="orange", yerr=0.1, label="orange")
+        df.plot("x", "b", c="blue", yerr=None, ax=ax, label="blue")
+
+        legend = ax.get_legend()
+        if Version(mpl.__version__) < Version("3.7"):
+            result_handles = legend.legendHandles
+        else:
+            result_handles = legend.legend_handles
+
+        assert isinstance(result_handles[0], mpl.collections.LineCollection)
+        assert isinstance(result_handles[1], mpl.lines.Line2D)
+
+    def test_legend_false(self):
+        # https://github.com/pandas-dev/pandas/issues/40044
+        df = DataFrame({"a": [1, 1], "b": [2, 3]})
+        df2 = DataFrame({"d": [2.5, 2.5]})
+
+        ax = df.plot(legend=True, color={"a": "blue", "b": "green"}, secondary_y="b")
+        df2.plot(legend=True, color={"d": "red"}, ax=ax)
+        legend = ax.get_legend()
+        if Version(mpl.__version__) < Version("3.7"):
+            handles = legend.legendHandles
+        else:
+            handles = legend.legend_handles
+        result = [handle.get_color() for handle in handles]
+        expected = ["blue", "green", "red"]
+        assert result == expected
+
+    @pytest.mark.parametrize("kind", ["line", "bar", "barh", "kde", "area", "hist"])
+    def test_df_legend_labels(self, kind):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).random((3, 3)), columns=["a", "b", "c"])
+        df2 = DataFrame(
+            np.random.default_rng(2).random((3, 3)), columns=["d", "e", "f"]
+        )
+        df3 = DataFrame(
+            np.random.default_rng(2).random((3, 3)), columns=["g", "h", "i"]
+        )
+        df4 = DataFrame(
+            np.random.default_rng(2).random((3, 3)), columns=["j", "k", "l"]
+        )
+
+        ax = df.plot(kind=kind, legend=True)
+        _check_legend_labels(ax, labels=df.columns)
+
+        ax = df2.plot(kind=kind, legend=False, ax=ax)
+        _check_legend_labels(ax, labels=df.columns)
+
+        ax = df3.plot(kind=kind, legend=True, ax=ax)
+        _check_legend_labels(ax, labels=df.columns.union(df3.columns))
+
+        ax = df4.plot(kind=kind, legend="reverse", ax=ax)
+        expected = list(df.columns.union(df3.columns)) + list(reversed(df4.columns))
+        _check_legend_labels(ax, labels=expected)
+
+    def test_df_legend_labels_secondary_y(self):
+        pytest.importorskip("scipy")
+        df = DataFrame(np.random.default_rng(2).random((3, 3)), columns=["a", "b", "c"])
+        df2 = DataFrame(
+            np.random.default_rng(2).random((3, 3)), columns=["d", "e", "f"]
+        )
+        df3 = DataFrame(
+            np.random.default_rng(2).random((3, 3)), columns=["g", "h", "i"]
+        )
+        # Secondary Y
+        ax = df.plot(legend=True, secondary_y="b")
+        _check_legend_labels(ax, labels=["a", "b (right)", "c"])
+        ax = df2.plot(legend=False, ax=ax)
+        _check_legend_labels(ax, labels=["a", "b (right)", "c"])
+        ax = df3.plot(kind="bar", legend=True, secondary_y="h", ax=ax)
+        _check_legend_labels(ax, labels=["a", "b (right)", "c", "g", "h (right)", "i"])
+
+    def test_df_legend_labels_time_series(self):
+        # Time Series
+        pytest.importorskip("scipy")
+        ind = date_range("1/1/2014", periods=3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["a", "b", "c"],
+            index=ind,
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["d", "e", "f"],
+            index=ind,
+        )
+        df3 = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["g", "h", "i"],
+            index=ind,
+        )
+        ax = df.plot(legend=True, secondary_y="b")
+        _check_legend_labels(ax, labels=["a", "b (right)", "c"])
+        ax = df2.plot(legend=False, ax=ax)
+        _check_legend_labels(ax, labels=["a", "b (right)", "c"])
+        ax = df3.plot(legend=True, ax=ax)
+        _check_legend_labels(ax, labels=["a", "b (right)", "c", "g", "h", "i"])
+
+    def test_df_legend_labels_time_series_scatter(self):
+        # Time Series
+        pytest.importorskip("scipy")
+        ind = date_range("1/1/2014", periods=3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["a", "b", "c"],
+            index=ind,
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["d", "e", "f"],
+            index=ind,
+        )
+        df3 = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["g", "h", "i"],
+            index=ind,
+        )
+        # scatter
+        ax = df.plot.scatter(x="a", y="b", label="data1")
+        _check_legend_labels(ax, labels=["data1"])
+        ax = df2.plot.scatter(x="d", y="e", legend=False, label="data2", ax=ax)
+        _check_legend_labels(ax, labels=["data1"])
+        ax = df3.plot.scatter(x="g", y="h", label="data3", ax=ax)
+        _check_legend_labels(ax, labels=["data1", "data3"])
+
+    def test_df_legend_labels_time_series_no_mutate(self):
+        pytest.importorskip("scipy")
+        ind = date_range("1/1/2014", periods=3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 3)),
+            columns=["a", "b", "c"],
+            index=ind,
+        )
+        # ensure label args pass through and
+        # index name does not mutate
+        # column names don't mutate
+        df5 = df.set_index("a")
+        ax = df5.plot(y="b")
+        _check_legend_labels(ax, labels=["b"])
+        ax = df5.plot(y="b", label="LABEL_b")
+        _check_legend_labels(ax, labels=["LABEL_b"])
+        _check_text_labels(ax.xaxis.get_label(), "a")
+        ax = df5.plot(y="c", label="LABEL_c", ax=ax)
+        _check_legend_labels(ax, labels=["LABEL_b", "LABEL_c"])
+        assert df5.columns.tolist() == ["b", "c"]
+
+    def test_missing_marker_multi_plots_on_same_ax(self):
+        # GH 18222
+        df = DataFrame(data=[[1, 1, 1, 1], [2, 2, 4, 8]], columns=["x", "r", "g", "b"])
+        _, ax = mpl.pyplot.subplots(nrows=1, ncols=3)
+        # Left plot
+        df.plot(x="x", y="r", linewidth=0, marker="o", color="r", ax=ax[0])
+        df.plot(x="x", y="g", linewidth=1, marker="x", color="g", ax=ax[0])
+        df.plot(x="x", y="b", linewidth=1, marker="o", color="b", ax=ax[0])
+        _check_legend_labels(ax[0], labels=["r", "g", "b"])
+        _check_legend_marker(ax[0], expected_markers=["o", "x", "o"])
+        # Center plot
+        df.plot(x="x", y="b", linewidth=1, marker="o", color="b", ax=ax[1])
+        df.plot(x="x", y="r", linewidth=0, marker="o", color="r", ax=ax[1])
+        df.plot(x="x", y="g", linewidth=1, marker="x", color="g", ax=ax[1])
+        _check_legend_labels(ax[1], labels=["b", "r", "g"])
+        _check_legend_marker(ax[1], expected_markers=["o", "o", "x"])
+        # Right plot
+        df.plot(x="x", y="g", linewidth=1, marker="x", color="g", ax=ax[2])
+        df.plot(x="x", y="b", linewidth=1, marker="o", color="b", ax=ax[2])
+        df.plot(x="x", y="r", linewidth=0, marker="o", color="r", ax=ax[2])
+        _check_legend_labels(ax[2], labels=["g", "b", "r"])
+        _check_legend_marker(ax[2], expected_markers=["x", "o", "o"])
+
+    def test_legend_name(self):
+        multi = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)),
+            columns=[np.array(["a", "a", "b", "b"]), np.array(["x", "y", "x", "y"])],
+        )
+        multi.columns.names = ["group", "individual"]
+
+        ax = multi.plot()
+        leg_title = ax.legend_.get_title()
+        _check_text_labels(leg_title, "group,individual")
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot(legend=True, ax=ax)
+        leg_title = ax.legend_.get_title()
+        _check_text_labels(leg_title, "group,individual")
+
+        df.columns.name = "new"
+        ax = df.plot(legend=False, ax=ax)
+        leg_title = ax.legend_.get_title()
+        _check_text_labels(leg_title, "group,individual")
+
+        ax = df.plot(legend=True, ax=ax)
+        leg_title = ax.legend_.get_title()
+        _check_text_labels(leg_title, "new")
+
+    @pytest.mark.parametrize(
+        "kind",
+        [
+            "line",
+            "bar",
+            "barh",
+            pytest.param("kde", marks=td.skip_if_no("scipy")),
+            "area",
+            "hist",
+        ],
+    )
+    def test_no_legend(self, kind):
+        df = DataFrame(np.random.default_rng(2).random((3, 3)), columns=["a", "b", "c"])
+        ax = df.plot(kind=kind, legend=False)
+        _check_legend_labels(ax, visible=False)
+
+    def test_missing_markers_legend(self):
+        # 14958
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 3)), columns=["A", "B", "C"]
+        )
+        ax = df.plot(y=["A"], marker="x", linestyle="solid")
+        df.plot(y=["B"], marker="o", linestyle="dotted", ax=ax)
+        df.plot(y=["C"], marker="<", linestyle="dotted", ax=ax)
+
+        _check_legend_labels(ax, labels=["A", "B", "C"])
+        _check_legend_marker(ax, expected_markers=["x", "o", "<"])
+
+    def test_missing_markers_legend_using_style(self):
+        # 14563
+        df = DataFrame(
+            {
+                "A": [1, 2, 3, 4, 5, 6],
+                "B": [2, 4, 1, 3, 2, 4],
+                "C": [3, 3, 2, 6, 4, 2],
+                "X": [1, 2, 3, 4, 5, 6],
+            }
+        )
+
+        _, ax = mpl.pyplot.subplots()
+        for kind in "ABC":
+            df.plot("X", kind, label=kind, ax=ax, style=".")
+
+        _check_legend_labels(ax, labels=["A", "B", "C"])
+        _check_legend_marker(ax, expected_markers=[".", ".", "."])
diff --git a/pandas/pandas/tests/plotting/frame/test_frame_subplots.py b/pandas/pandas/tests/plotting/frame/test_frame_subplots.py
new file mode 100644
index 0000000000000..b44725a01fe23
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_frame_subplots.py
@@ -0,0 +1,752 @@
+"""Test cases for DataFrame.plot"""
+
+import string
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_linux
+from pandas.compat.numpy import np_version_gte1p24
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_axes_shape,
+    _check_box_return_type,
+    _check_legend_labels,
+    _check_ticks_props,
+    _check_visible,
+    _flatten_visible,
+)
+
+from pandas.io.formats.printing import pprint_thing
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+
+class TestDataFramePlotsSubplots:
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["bar", "barh", "line", "area"])
+    def test_subplots(self, kind):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+
+        axes = df.plot(kind=kind, subplots=True, sharex=True, legend=True)
+        _check_axes_shape(axes, axes_num=3, layout=(3, 1))
+        assert axes.shape == (3,)
+
+        for ax, column in zip(axes, df.columns):
+            _check_legend_labels(ax, labels=[pprint_thing(column)])
+
+        for ax in axes[:-2]:
+            _check_visible(ax.xaxis)  # xaxis must be visible for grid
+            _check_visible(ax.get_xticklabels(), visible=False)
+            if kind != "bar":
+                # change https://github.com/pandas-dev/pandas/issues/26714
+                _check_visible(ax.get_xticklabels(minor=True), visible=False)
+            _check_visible(ax.xaxis.get_label(), visible=False)
+            _check_visible(ax.get_yticklabels())
+
+        _check_visible(axes[-1].xaxis)
+        _check_visible(axes[-1].get_xticklabels())
+        _check_visible(axes[-1].get_xticklabels(minor=True))
+        _check_visible(axes[-1].xaxis.get_label())
+        _check_visible(axes[-1].get_yticklabels())
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["bar", "barh", "line", "area"])
+    def test_subplots_no_share_x(self, kind):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        axes = df.plot(kind=kind, subplots=True, sharex=False)
+        for ax in axes:
+            _check_visible(ax.xaxis)
+            _check_visible(ax.get_xticklabels())
+            _check_visible(ax.get_xticklabels(minor=True))
+            _check_visible(ax.xaxis.get_label())
+            _check_visible(ax.get_yticklabels())
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["bar", "barh", "line", "area"])
+    def test_subplots_no_legend(self, kind):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        axes = df.plot(kind=kind, subplots=True, legend=False)
+        for ax in axes:
+            assert ax.get_legend() is None
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_subplots_timeseries(self, kind):
+        idx = date_range(start="2014-07-01", freq="ME", periods=10)
+        df = DataFrame(np.random.default_rng(2).random((10, 3)), index=idx)
+
+        axes = df.plot(kind=kind, subplots=True, sharex=True)
+        _check_axes_shape(axes, axes_num=3, layout=(3, 1))
+
+        for ax in axes[:-2]:
+            # GH 7801
+            _check_visible(ax.xaxis)  # xaxis must be visible for grid
+            _check_visible(ax.get_xticklabels(), visible=False)
+            _check_visible(ax.get_xticklabels(minor=True), visible=False)
+            _check_visible(ax.xaxis.get_label(), visible=False)
+            _check_visible(ax.get_yticklabels())
+
+        _check_visible(axes[-1].xaxis)
+        _check_visible(axes[-1].get_xticklabels())
+        _check_visible(axes[-1].get_xticklabels(minor=True))
+        _check_visible(axes[-1].xaxis.get_label())
+        _check_visible(axes[-1].get_yticklabels())
+        _check_ticks_props(axes, xrot=0)
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_subplots_timeseries_rot(self, kind):
+        idx = date_range(start="2014-07-01", freq="ME", periods=10)
+        df = DataFrame(np.random.default_rng(2).random((10, 3)), index=idx)
+        axes = df.plot(kind=kind, subplots=True, sharex=False, rot=45, fontsize=7)
+        for ax in axes:
+            _check_visible(ax.xaxis)
+            _check_visible(ax.get_xticklabels())
+            _check_visible(ax.get_xticklabels(minor=True))
+            _check_visible(ax.xaxis.get_label())
+            _check_visible(ax.get_yticklabels())
+            _check_ticks_props(ax, xlabelsize=7, xrot=45, ylabelsize=7)
+
+    @pytest.mark.parametrize(
+        "col", ["numeric", "timedelta", "datetime_no_tz", "datetime_all_tz"]
+    )
+    def test_subplots_timeseries_y_axis(self, col):
+        # GH16953
+        data = {
+            "numeric": np.array([1, 2, 5]),
+            "timedelta": [
+                pd.Timedelta(-10, unit="s"),
+                pd.Timedelta(10, unit="m"),
+                pd.Timedelta(10, unit="h"),
+            ],
+            "datetime_no_tz": [
+                pd.to_datetime("2017-08-01 00:00:00"),
+                pd.to_datetime("2017-08-01 02:00:00"),
+                pd.to_datetime("2017-08-02 00:00:00"),
+            ],
+            "datetime_all_tz": [
+                pd.to_datetime("2017-08-01 00:00:00", utc=True),
+                pd.to_datetime("2017-08-01 02:00:00", utc=True),
+                pd.to_datetime("2017-08-02 00:00:00", utc=True),
+            ],
+            "text": ["This", "should", "fail"],
+        }
+        testdata = DataFrame(data)
+
+        ax = testdata.plot(y=col)
+        result = ax.get_lines()[0].get_data()[1]
+        expected = testdata[col].values
+        assert (result == expected).all()
+
+    def test_subplots_timeseries_y_text_error(self):
+        # GH16953
+        data = {
+            "numeric": np.array([1, 2, 5]),
+            "text": ["This", "should", "fail"],
+        }
+        testdata = DataFrame(data)
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            testdata.plot(y="text")
+
+    @pytest.mark.xfail(reason="not support for period, categorical, datetime_mixed_tz")
+    def test_subplots_timeseries_y_axis_not_supported(self):
+        """
+        This test will fail for:
+            period:
+                since period isn't yet implemented in ``select_dtypes``
+                and because it will need a custom value converter +
+                tick formatter (as was done for x-axis plots)
+
+            categorical:
+                 because it will need a custom value converter +
+                 tick formatter (also doesn't work for x-axis, as of now)
+
+            datetime_mixed_tz:
+                because of the way how pandas handles ``Series`` of
+                ``datetime`` objects with different timezone,
+                generally converting ``datetime`` objects in a tz-aware
+                form could help with this problem
+        """
+        data = {
+            "numeric": np.array([1, 2, 5]),
+            "period": [
+                pd.Period("2017-08-01 00:00:00", freq="h"),
+                pd.Period("2017-08-01 02:00", freq="h"),
+                pd.Period("2017-08-02 00:00:00", freq="h"),
+            ],
+            "categorical": pd.Categorical(
+                ["c", "b", "a"], categories=["a", "b", "c"], ordered=False
+            ),
+            "datetime_mixed_tz": [
+                pd.to_datetime("2017-08-01 00:00:00", utc=True),
+                pd.to_datetime("2017-08-01 02:00:00"),
+                pd.to_datetime("2017-08-02 00:00:00"),
+            ],
+        }
+        testdata = DataFrame(data)
+        ax_period = testdata.plot(x="numeric", y="period")
+        assert (
+            ax_period.get_lines()[0].get_data()[1] == testdata["period"].values
+        ).all()
+        ax_categorical = testdata.plot(x="numeric", y="categorical")
+        assert (
+            ax_categorical.get_lines()[0].get_data()[1]
+            == testdata["categorical"].values
+        ).all()
+        ax_datetime_mixed_tz = testdata.plot(x="numeric", y="datetime_mixed_tz")
+        assert (
+            ax_datetime_mixed_tz.get_lines()[0].get_data()[1]
+            == testdata["datetime_mixed_tz"].values
+        ).all()
+
+    @pytest.mark.parametrize(
+        "layout, exp_layout",
+        [
+            [(2, 2), (2, 2)],
+            [(-1, 2), (2, 2)],
+            [(2, -1), (2, 2)],
+            [(1, 4), (1, 4)],
+            [(-1, 4), (1, 4)],
+            [(4, -1), (4, 1)],
+        ],
+    )
+    def test_subplots_layout_multi_column(self, layout, exp_layout):
+        # GH 6667
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+
+        axes = df.plot(subplots=True, layout=layout)
+        _check_axes_shape(axes, axes_num=3, layout=exp_layout)
+        assert axes.shape == exp_layout
+
+    def test_subplots_layout_multi_column_error(self):
+        # GH 6667
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        msg = "Layout of 1x1 must be larger than required size 3"
+
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, layout=(1, 1))
+
+        msg = "At least one dimension of layout must be positive"
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, layout=(-1, -1))
+
+    @pytest.mark.parametrize(
+        "kwargs, expected_axes_num, expected_layout, expected_shape",
+        [
+            ({}, 1, (1, 1), (1,)),
+            ({"layout": (3, 3)}, 1, (3, 3), (3, 3)),
+        ],
+    )
+    def test_subplots_layout_single_column(
+        self, kwargs, expected_axes_num, expected_layout, expected_shape
+    ):
+        # GH 6667
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 1)),
+            index=list(string.ascii_letters[:10]),
+        )
+        axes = df.plot(subplots=True, **kwargs)
+        _check_axes_shape(
+            axes,
+            axes_num=expected_axes_num,
+            layout=expected_layout,
+        )
+        assert axes.shape == expected_shape
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("idx", [range(5), date_range("1/1/2000", periods=5)])
+    def test_subplots_warnings(self, idx):
+        # GH 9464
+        with tm.assert_produces_warning(None):
+            df = DataFrame(np.random.default_rng(2).standard_normal((5, 4)), index=idx)
+            df.plot(subplots=True, layout=(3, 2))
+
+    def test_subplots_multiple_axes(self):
+        # GH 5353, 6970, GH 7069
+        fig, axes = mpl.pyplot.subplots(2, 3)
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+
+        returned = df.plot(subplots=True, ax=axes[0], sharex=False, sharey=False)
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        assert returned.shape == (3,)
+        assert returned[0].figure is fig
+        # draw on second row
+        returned = df.plot(subplots=True, ax=axes[1], sharex=False, sharey=False)
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        assert returned.shape == (3,)
+        assert returned[0].figure is fig
+        _check_axes_shape(axes, axes_num=6, layout=(2, 3))
+
+    def test_subplots_multiple_axes_error(self):
+        # GH 5353, 6970, GH 7069
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=list(string.ascii_letters[:10]),
+        )
+        msg = "The number of passed axes must be 3, the same as the output plot"
+        _, axes = mpl.pyplot.subplots(2, 3)
+
+        with pytest.raises(ValueError, match=msg):
+            # pass different number of axes from required
+            df.plot(subplots=True, ax=axes)
+
+    @pytest.mark.parametrize(
+        "layout, exp_layout",
+        [
+            [(2, 1), (2, 2)],
+            [(2, -1), (2, 2)],
+            [(-1, 2), (2, 2)],
+        ],
+    )
+    def test_subplots_multiple_axes_2_dim(self, layout, exp_layout):
+        # GH 5353, 6970, GH 7069
+        # pass 2-dim axes and invalid layout
+        # invalid layout should not affect to input and return value
+        # (show warning is tested in
+        # TestDataFrameGroupByPlots.test_grouped_box_multiple_axes
+        _, axes = mpl.pyplot.subplots(2, 2)
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 4)),
+            index=list(string.ascii_letters[:10]),
+        )
+        with tm.assert_produces_warning(UserWarning, match="layout keyword is ignored"):
+            returned = df.plot(
+                subplots=True, ax=axes, layout=layout, sharex=False, sharey=False
+            )
+            _check_axes_shape(returned, axes_num=4, layout=exp_layout)
+            assert returned.shape == (4,)
+
+    def test_subplots_multiple_axes_single_col(self):
+        # GH 5353, 6970, GH 7069
+        # single column
+        _, axes = mpl.pyplot.subplots(1, 1)
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 1)),
+            index=list(string.ascii_letters[:10]),
+        )
+
+        axes = df.plot(subplots=True, ax=[axes], sharex=False, sharey=False)
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+        assert axes.shape == (1,)
+
+    def test_subplots_ts_share_axes(self):
+        # GH 3964
+        _, axes = mpl.pyplot.subplots(3, 3, sharex=True, sharey=True)
+        mpl.pyplot.subplots_adjust(left=0.05, right=0.95, hspace=0.3, wspace=0.3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 9)),
+            index=date_range(start="2014-07-01", freq="ME", periods=10),
+        )
+        for i, ax in enumerate(axes.ravel()):
+            df[i].plot(ax=ax, fontsize=5)
+
+        # Rows other than bottom should not be visible
+        for ax in axes[0:-1].ravel():
+            _check_visible(ax.get_xticklabels(), visible=False)
+
+        # Bottom row should be visible
+        for ax in axes[-1].ravel():
+            _check_visible(ax.get_xticklabels(), visible=True)
+
+        # First column should be visible
+        for ax in axes[[0, 1, 2], [0]].ravel():
+            _check_visible(ax.get_yticklabels(), visible=True)
+
+        # Other columns should not be visible
+        for ax in axes[[0, 1, 2], [1]].ravel():
+            _check_visible(ax.get_yticklabels(), visible=False)
+        for ax in axes[[0, 1, 2], [2]].ravel():
+            _check_visible(ax.get_yticklabels(), visible=False)
+
+    def test_subplots_sharex_axes_existing_axes(self):
+        # GH 9158
+        d = {"A": [1.0, 2.0, 3.0, 4.0], "B": [4.0, 3.0, 2.0, 1.0], "C": [5, 1, 3, 4]}
+        df = DataFrame(d, index=date_range("2014 10 11", "2014 10 14"))
+
+        axes = df[["A", "B"]].plot(subplots=True)
+        df["C"].plot(ax=axes[0], secondary_y=True)
+
+        _check_visible(axes[0].get_xticklabels(), visible=False)
+        _check_visible(axes[1].get_xticklabels(), visible=True)
+        for ax in axes.ravel():
+            _check_visible(ax.get_yticklabels(), visible=True)
+
+    def test_subplots_dup_columns(self):
+        # GH 10962
+        df = DataFrame(np.random.default_rng(2).random((5, 5)), columns=list("aaaaa"))
+        axes = df.plot(subplots=True)
+        for ax in axes:
+            _check_legend_labels(ax, labels=["a"])
+            assert len(ax.lines) == 1
+
+    def test_subplots_dup_columns_secondary_y(self):
+        # GH 10962
+        df = DataFrame(np.random.default_rng(2).random((5, 5)), columns=list("aaaaa"))
+        axes = df.plot(subplots=True, secondary_y="a")
+        for ax in axes:
+            # (right) is only attached when subplots=False
+            _check_legend_labels(ax, labels=["a"])
+            assert len(ax.lines) == 1
+
+    def test_subplots_dup_columns_secondary_y_no_subplot(self):
+        # GH 10962
+        df = DataFrame(np.random.default_rng(2).random((5, 5)), columns=list("aaaaa"))
+        ax = df.plot(secondary_y="a")
+        _check_legend_labels(ax, labels=["a (right)"] * 5)
+        assert len(ax.lines) == 0
+        assert len(ax.right_ax.lines) == 5
+
+    @pytest.mark.xfail(
+        np_version_gte1p24 and is_platform_linux(),
+        reason="Weird rounding problems",
+        strict=False,
+    )
+    def test_bar_log_no_subplots(self):
+        # GH3254, GH3298 matplotlib/matplotlib#1882, #1892
+        # regressions in 1.2.1
+        expected = np.array([0.1, 1.0, 10.0, 100])
+
+        # no subplots
+        df = DataFrame({"A": [3] * 5, "B": list(range(1, 6))}, index=range(5))
+        ax = df.plot.bar(grid=True, log=True)
+        tm.assert_numpy_array_equal(ax.yaxis.get_ticklocs(), expected)
+
+    @pytest.mark.xfail(
+        np_version_gte1p24 and is_platform_linux(),
+        reason="Weird rounding problems",
+        strict=False,
+    )
+    def test_bar_log_subplots(self):
+        expected = np.array([0.1, 1.0, 10.0, 100.0, 1000.0, 1e4])
+
+        ax = DataFrame([Series([200, 300]), Series([300, 500])]).plot.bar(
+            log=True, subplots=True
+        )
+
+        tm.assert_numpy_array_equal(ax[0].yaxis.get_ticklocs(), expected)
+        tm.assert_numpy_array_equal(ax[1].yaxis.get_ticklocs(), expected)
+
+    def test_boxplot_subplots_return_type_default(self, hist_df):
+        df = hist_df
+
+        # normal style: return_type=None
+        result = df.plot.box(subplots=True)
+        assert isinstance(result, Series)
+        _check_box_return_type(
+            result, None, expected_keys=["height", "weight", "category"]
+        )
+
+    @pytest.mark.parametrize("rt", ["dict", "axes", "both"])
+    def test_boxplot_subplots_return_type(self, hist_df, rt):
+        df = hist_df
+        returned = df.plot.box(return_type=rt, subplots=True)
+        _check_box_return_type(
+            returned,
+            rt,
+            expected_keys=["height", "weight", "category"],
+            check_ax_title=False,
+        )
+
+    def test_df_subplots_patterns_minorticks(self):
+        # GH 10657
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=date_range("1/1/2000", periods=10),
+            columns=list("AB"),
+        )
+
+        # shared subplots
+        _, axes = plt.subplots(2, 1, sharex=True)
+        axes = df.plot(subplots=True, ax=axes)
+        for ax in axes:
+            assert len(ax.lines) == 1
+            _check_visible(ax.get_yticklabels(), visible=True)
+        # xaxis of 1st ax must be hidden
+        _check_visible(axes[0].get_xticklabels(), visible=False)
+        _check_visible(axes[0].get_xticklabels(minor=True), visible=False)
+        _check_visible(axes[1].get_xticklabels(), visible=True)
+        _check_visible(axes[1].get_xticklabels(minor=True), visible=True)
+
+    def test_df_subplots_patterns_minorticks_1st_ax_hidden(self):
+        # GH 10657
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=date_range("1/1/2000", periods=10),
+            columns=list("AB"),
+        )
+        _, axes = plt.subplots(2, 1)
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            axes = df.plot(subplots=True, ax=axes, sharex=True)
+        for ax in axes:
+            assert len(ax.lines) == 1
+            _check_visible(ax.get_yticklabels(), visible=True)
+        # xaxis of 1st ax must be hidden
+        _check_visible(axes[0].get_xticklabels(), visible=False)
+        _check_visible(axes[0].get_xticklabels(minor=True), visible=False)
+        _check_visible(axes[1].get_xticklabels(), visible=True)
+        _check_visible(axes[1].get_xticklabels(minor=True), visible=True)
+
+    def test_df_subplots_patterns_minorticks_not_shared(self):
+        # GH 10657
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=date_range("1/1/2000", periods=10),
+            columns=list("AB"),
+        )
+        # not shared
+        _, axes = plt.subplots(2, 1)
+        axes = df.plot(subplots=True, ax=axes)
+        for ax in axes:
+            assert len(ax.lines) == 1
+            _check_visible(ax.get_yticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(), visible=True)
+            _check_visible(ax.get_xticklabels(minor=True), visible=True)
+
+    def test_subplots_sharex_false(self):
+        # test when sharex is set to False, two plots should have different
+        # labels, GH 25160
+        df = DataFrame(np.random.default_rng(2).random((10, 2)))
+        df.iloc[5:, 1] = np.nan
+        df.iloc[:5, 0] = np.nan
+
+        _, axs = mpl.pyplot.subplots(2, 1)
+        df.plot.line(ax=axs, subplots=True, sharex=False)
+
+        expected_ax1 = np.arange(4.5, 10, 0.5)
+        expected_ax2 = np.arange(-0.5, 5, 0.5)
+
+        tm.assert_numpy_array_equal(axs[0].get_xticks(), expected_ax1)
+        tm.assert_numpy_array_equal(axs[1].get_xticks(), expected_ax2)
+
+    def test_subplots_constrained_layout(self, temp_file):
+        # GH 25261
+        idx = date_range(start="now", periods=10)
+        df = DataFrame(np.random.default_rng(2).random((10, 3)), index=idx)
+        kwargs = {}
+        if hasattr(mpl.pyplot.Figure, "get_constrained_layout"):
+            kwargs["constrained_layout"] = True
+        _, axes = mpl.pyplot.subplots(2, **kwargs)
+        with tm.assert_produces_warning(None):
+            df.plot(ax=axes[0])
+            with temp_file.open(mode="wb") as path:
+                mpl.pyplot.savefig(path)
+
+    @pytest.mark.parametrize(
+        "index_name, old_label, new_label",
+        [
+            (None, "", "new"),
+            ("old", "old", "new"),
+            (None, "", ""),
+            (None, "", 1),
+            (None, "", [1, 2]),
+        ],
+    )
+    @pytest.mark.parametrize("kind", ["line", "area", "bar"])
+    def test_xlabel_ylabel_dataframe_subplots(
+        self, kind, index_name, old_label, new_label
+    ):
+        # GH 9093
+        df = DataFrame([[1, 2], [2, 5]], columns=["Type A", "Type B"])
+        df.index.name = index_name
+
+        # default is the ylabel is not shown and xlabel is index name
+        axes = df.plot(kind=kind, subplots=True)
+        assert all(ax.get_ylabel() == "" for ax in axes)
+        assert all(ax.get_xlabel() == old_label for ax in axes)
+
+        # old xlabel will be overridden and assigned ylabel will be used as ylabel
+        axes = df.plot(kind=kind, ylabel=new_label, xlabel=new_label, subplots=True)
+        assert all(ax.get_ylabel() == str(new_label) for ax in axes)
+        assert all(ax.get_xlabel() == str(new_label) for ax in axes)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            # stacked center
+            {"kind": "bar", "stacked": True},
+            {"kind": "bar", "stacked": True, "width": 0.9},
+            {"kind": "barh", "stacked": True},
+            {"kind": "barh", "stacked": True, "width": 0.9},
+            # center
+            {"kind": "bar", "stacked": False},
+            {"kind": "bar", "stacked": False, "width": 0.9},
+            {"kind": "barh", "stacked": False},
+            {"kind": "barh", "stacked": False, "width": 0.9},
+            # subplots center
+            {"kind": "bar", "subplots": True},
+            {"kind": "bar", "subplots": True, "width": 0.9},
+            {"kind": "barh", "subplots": True},
+            {"kind": "barh", "subplots": True, "width": 0.9},
+            # align edge
+            {"kind": "bar", "stacked": True, "align": "edge"},
+            {"kind": "bar", "stacked": True, "width": 0.9, "align": "edge"},
+            {"kind": "barh", "stacked": True, "align": "edge"},
+            {"kind": "barh", "stacked": True, "width": 0.9, "align": "edge"},
+            {"kind": "bar", "stacked": False, "align": "edge"},
+            {"kind": "bar", "stacked": False, "width": 0.9, "align": "edge"},
+            {"kind": "barh", "stacked": False, "align": "edge"},
+            {"kind": "barh", "stacked": False, "width": 0.9, "align": "edge"},
+            {"kind": "bar", "subplots": True, "align": "edge"},
+            {"kind": "bar", "subplots": True, "width": 0.9, "align": "edge"},
+            {"kind": "barh", "subplots": True, "align": "edge"},
+            {"kind": "barh", "subplots": True, "width": 0.9, "align": "edge"},
+        ],
+    )
+    def test_bar_align_multiple_columns(self, kwargs):
+        # GH2157
+        df = DataFrame({"A": [3] * 5, "B": list(range(5))}, index=range(5))
+        self._check_bar_alignment(df, **kwargs)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"kind": "bar", "stacked": False},
+            {"kind": "bar", "stacked": True},
+            {"kind": "barh", "stacked": False},
+            {"kind": "barh", "stacked": True},
+            {"kind": "bar", "subplots": True},
+            {"kind": "barh", "subplots": True},
+        ],
+    )
+    def test_bar_align_single_column(self, kwargs):
+        df = DataFrame(np.random.default_rng(2).standard_normal(5))
+        self._check_bar_alignment(df, **kwargs)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"kind": "bar", "stacked": False},
+            {"kind": "bar", "stacked": True},
+            {"kind": "barh", "stacked": False},
+            {"kind": "barh", "stacked": True},
+            {"kind": "bar", "subplots": True},
+            {"kind": "barh", "subplots": True},
+        ],
+    )
+    def test_bar_barwidth_position(self, kwargs):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        self._check_bar_alignment(df, width=0.9, position=0.2, **kwargs)
+
+    @pytest.mark.parametrize("w", [1, 1.0])
+    def test_bar_barwidth_position_int(self, w):
+        # GH 12979
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        ax = df.plot.bar(stacked=True, width=w)
+        ticks = ax.xaxis.get_ticklocs()
+        tm.assert_numpy_array_equal(ticks, np.array([0, 1, 2, 3, 4]))
+        assert ax.get_xlim() == (-0.75, 4.75)
+        # check left-edge of bars
+        assert ax.patches[0].get_x() == -0.5
+        assert ax.patches[-1].get_x() == 3.5
+
+    @pytest.mark.parametrize(
+        "kind, kwargs",
+        [
+            ["bar", {"stacked": True}],
+            ["barh", {"stacked": False}],
+            ["barh", {"stacked": True}],
+            ["bar", {"subplots": True}],
+            ["barh", {"subplots": True}],
+        ],
+    )
+    def test_bar_barwidth_position_int_width_1(self, kind, kwargs):
+        # GH 12979
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        self._check_bar_alignment(df, kind=kind, width=1, **kwargs)
+
+    def _check_bar_alignment(
+        self,
+        df,
+        kind="bar",
+        stacked=False,
+        subplots=False,
+        align="center",
+        width=0.5,
+        position=0.5,
+    ):
+        axes = df.plot(
+            kind=kind,
+            stacked=stacked,
+            subplots=subplots,
+            align=align,
+            width=width,
+            position=position,
+            grid=True,
+        )
+
+        axes = _flatten_visible(axes)
+
+        for ax in axes:
+            if kind == "bar":
+                axis = ax.xaxis
+                ax_min, ax_max = ax.get_xlim()
+                min_edge = min(p.get_x() for p in ax.patches)
+                max_edge = max(p.get_x() + p.get_width() for p in ax.patches)
+            elif kind == "barh":
+                axis = ax.yaxis
+                ax_min, ax_max = ax.get_ylim()
+                min_edge = min(p.get_y() for p in ax.patches)
+                max_edge = max(p.get_y() + p.get_height() for p in ax.patches)
+            else:
+                raise ValueError
+
+            # GH 7498
+            # compare margins between lim and bar edges
+            tm.assert_almost_equal(ax_min, min_edge - 0.25)
+            tm.assert_almost_equal(ax_max, max_edge + 0.25)
+
+            p = ax.patches[0]
+            if kind == "bar" and (stacked is True or subplots is True):
+                edge = p.get_x()
+                center = edge + p.get_width() * position
+            elif kind == "bar" and stacked is False:
+                center = p.get_x() + p.get_width() * len(df.columns) * position
+                edge = p.get_x()
+            elif kind == "barh" and (stacked is True or subplots is True):
+                center = p.get_y() + p.get_height() * position
+                edge = p.get_y()
+            elif kind == "barh" and stacked is False:
+                center = p.get_y() + p.get_height() * len(df.columns) * position
+                edge = p.get_y()
+            else:
+                raise ValueError
+
+            # Check the ticks locates on integer
+            assert (axis.get_ticklocs() == np.arange(len(df))).all()
+
+            if align == "center":
+                # Check whether the bar locates on center
+                tm.assert_almost_equal(axis.get_ticklocs()[0], center)
+            elif align == "edge":
+                # Check whether the bar's edge starts from the tick
+                tm.assert_almost_equal(axis.get_ticklocs()[0], edge)
+            else:
+                raise ValueError
+
+        return axes
diff --git a/pandas/pandas/tests/plotting/frame/test_hist_box_by.py b/pandas/pandas/tests/plotting/frame/test_hist_box_by.py
new file mode 100644
index 0000000000000..a9250fa8347cc
--- /dev/null
+++ b/pandas/pandas/tests/plotting/frame/test_hist_box_by.py
@@ -0,0 +1,342 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_axes_shape,
+    _check_plot_works,
+    get_x_axis,
+    get_y_axis,
+)
+
+pytest.importorskip("matplotlib")
+
+
+@pytest.fixture
+def hist_df():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((30, 2)), columns=["A", "B"]
+    )
+    df["C"] = np.random.default_rng(2).choice(["a", "b", "c"], 30)
+    df["D"] = np.random.default_rng(2).choice(["a", "b", "c"], 30)
+    return df
+
+
+class TestHistWithBy:
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "by, column, titles, legends",
+        [
+            ("C", "A", ["a", "b", "c"], [["A"]] * 3),
+            ("C", ["A", "B"], ["a", "b", "c"], [["A", "B"]] * 3),
+            ("C", None, ["a", "b", "c"], [["A", "B"]] * 3),
+            (
+                ["C", "D"],
+                "A",
+                [
+                    "(a, a)",
+                    "(b, b)",
+                    "(c, c)",
+                ],
+                [["A"]] * 3,
+            ),
+            (
+                ["C", "D"],
+                ["A", "B"],
+                [
+                    "(a, a)",
+                    "(b, b)",
+                    "(c, c)",
+                ],
+                [["A", "B"]] * 3,
+            ),
+            (
+                ["C", "D"],
+                None,
+                [
+                    "(a, a)",
+                    "(b, b)",
+                    "(c, c)",
+                ],
+                [["A", "B"]] * 3,
+            ),
+        ],
+    )
+    def test_hist_plot_by_argument(self, by, column, titles, legends, hist_df):
+        # GH 15079
+        axes = _check_plot_works(
+            hist_df.plot.hist, column=column, by=by, default_axes=True
+        )
+        result_titles = [ax.get_title() for ax in axes]
+        result_legends = [
+            [legend.get_text() for legend in ax.get_legend().texts] for ax in axes
+        ]
+
+        assert result_legends == legends
+        assert result_titles == titles
+
+    @pytest.mark.parametrize(
+        "by, column, titles, legends",
+        [
+            (0, "A", ["a", "b", "c"], [["A"]] * 3),
+            (0, None, ["a", "b", "c"], [["A", "B"]] * 3),
+            (
+                [0, "D"],
+                "A",
+                [
+                    "(a, a)",
+                    "(b, b)",
+                    "(c, c)",
+                ],
+                [["A"]] * 3,
+            ),
+        ],
+    )
+    def test_hist_plot_by_0(self, by, column, titles, legends, hist_df):
+        # GH 15079
+        df = hist_df.copy()
+        df = df.rename(columns={"C": 0})
+
+        axes = _check_plot_works(df.plot.hist, default_axes=True, column=column, by=by)
+        result_titles = [ax.get_title() for ax in axes]
+        result_legends = [
+            [legend.get_text() for legend in ax.get_legend().texts] for ax in axes
+        ]
+
+        assert result_legends == legends
+        assert result_titles == titles
+
+    @pytest.mark.parametrize(
+        "by, column",
+        [
+            ([], ["A"]),
+            ([], ["A", "B"]),
+            ((), None),
+            ((), ["A", "B"]),
+        ],
+    )
+    def test_hist_plot_empty_list_string_tuple_by(self, by, column, hist_df):
+        # GH 15079
+        msg = "No group keys passed"
+        with pytest.raises(ValueError, match=msg):
+            _check_plot_works(
+                hist_df.plot.hist, default_axes=True, column=column, by=by
+            )
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "by, column, layout, axes_num",
+        [
+            (["C"], "A", (2, 2), 3),
+            ("C", "A", (2, 2), 3),
+            (["C"], ["A"], (1, 3), 3),
+            ("C", None, (3, 1), 3),
+            ("C", ["A", "B"], (3, 1), 3),
+            (["C", "D"], "A", (9, 1), 3),
+            (["C", "D"], "A", (3, 3), 3),
+            (["C", "D"], ["A"], (5, 2), 3),
+            (["C", "D"], ["A", "B"], (9, 1), 3),
+            (["C", "D"], None, (9, 1), 3),
+            (["C", "D"], ["A", "B"], (5, 2), 3),
+        ],
+    )
+    def test_hist_plot_layout_with_by(self, by, column, layout, axes_num, hist_df):
+        # GH 15079
+        # _check_plot_works adds an ax so catch warning. see GH #13188
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(
+                hist_df.plot.hist, column=column, by=by, layout=layout
+            )
+        _check_axes_shape(axes, axes_num=axes_num, layout=layout)
+
+    @pytest.mark.parametrize(
+        "msg, by, layout",
+        [
+            ("larger than required size", ["C", "D"], (1, 1)),
+            (re.escape("Layout must be a tuple of (rows, columns)"), "C", (1,)),
+            ("At least one dimension of layout must be positive", "C", (-1, -1)),
+        ],
+    )
+    def test_hist_plot_invalid_layout_with_by_raises(self, msg, by, layout, hist_df):
+        # GH 15079, test if error is raised when invalid layout is given
+
+        with pytest.raises(ValueError, match=msg):
+            hist_df.plot.hist(column=["A", "B"], by=by, layout=layout)
+
+    @pytest.mark.slow
+    def test_axis_share_x_with_by(self, hist_df):
+        # GH 15079
+        ax1, ax2, ax3 = hist_df.plot.hist(column="A", by="C", sharex=True)
+
+        # share x
+        assert get_x_axis(ax1).joined(ax1, ax2)
+        assert get_x_axis(ax2).joined(ax1, ax2)
+        assert get_x_axis(ax3).joined(ax1, ax3)
+        assert get_x_axis(ax3).joined(ax2, ax3)
+
+        # don't share y
+        assert not get_y_axis(ax1).joined(ax1, ax2)
+        assert not get_y_axis(ax2).joined(ax1, ax2)
+        assert not get_y_axis(ax3).joined(ax1, ax3)
+        assert not get_y_axis(ax3).joined(ax2, ax3)
+
+    @pytest.mark.slow
+    def test_axis_share_y_with_by(self, hist_df):
+        # GH 15079
+        ax1, ax2, ax3 = hist_df.plot.hist(column="A", by="C", sharey=True)
+
+        # share y
+        assert get_y_axis(ax1).joined(ax1, ax2)
+        assert get_y_axis(ax2).joined(ax1, ax2)
+        assert get_y_axis(ax3).joined(ax1, ax3)
+        assert get_y_axis(ax3).joined(ax2, ax3)
+
+        # don't share x
+        assert not get_x_axis(ax1).joined(ax1, ax2)
+        assert not get_x_axis(ax2).joined(ax1, ax2)
+        assert not get_x_axis(ax3).joined(ax1, ax3)
+        assert not get_x_axis(ax3).joined(ax2, ax3)
+
+    @pytest.mark.parametrize("figsize", [(12, 8), (20, 10)])
+    def test_figure_shape_hist_with_by(self, figsize, hist_df):
+        # GH 15079
+        axes = hist_df.plot.hist(column="A", by="C", figsize=figsize)
+        _check_axes_shape(axes, axes_num=3, figsize=figsize)
+
+
+class TestBoxWithBy:
+    @pytest.mark.parametrize(
+        "by, column, titles, xticklabels",
+        [
+            ("C", "A", ["A"], [["a", "b", "c"]]),
+            (
+                ["C", "D"],
+                "A",
+                ["A"],
+                [
+                    [
+                        "(a, a)",
+                        "(b, b)",
+                        "(c, c)",
+                    ]
+                ],
+            ),
+            ("C", ["A", "B"], ["A", "B"], [["a", "b", "c"]] * 2),
+            (
+                ["C", "D"],
+                ["A", "B"],
+                ["A", "B"],
+                [
+                    [
+                        "(a, a)",
+                        "(b, b)",
+                        "(c, c)",
+                    ]
+                ]
+                * 2,
+            ),
+            (["C"], None, ["A", "B"], [["a", "b", "c"]] * 2),
+        ],
+    )
+    def test_box_plot_by_argument(self, by, column, titles, xticklabels, hist_df):
+        # GH 15079
+        axes = _check_plot_works(
+            hist_df.plot.box, default_axes=True, column=column, by=by
+        )
+        result_titles = [ax.get_title() for ax in axes]
+        result_xticklabels = [
+            [label.get_text() for label in ax.get_xticklabels()] for ax in axes
+        ]
+
+        assert result_xticklabels == xticklabels
+        assert result_titles == titles
+
+    @pytest.mark.parametrize(
+        "by, column, titles, xticklabels",
+        [
+            (0, "A", ["A"], [["a", "b", "c"]]),
+            (
+                [0, "D"],
+                "A",
+                ["A"],
+                [
+                    [
+                        "(a, a)",
+                        "(b, b)",
+                        "(c, c)",
+                    ]
+                ],
+            ),
+            (0, None, ["A", "B"], [["a", "b", "c"]] * 2),
+        ],
+    )
+    def test_box_plot_by_0(self, by, column, titles, xticklabels, hist_df):
+        # GH 15079
+        df = hist_df.copy()
+        df = df.rename(columns={"C": 0})
+
+        axes = _check_plot_works(df.plot.box, default_axes=True, column=column, by=by)
+        result_titles = [ax.get_title() for ax in axes]
+        result_xticklabels = [
+            [label.get_text() for label in ax.get_xticklabels()] for ax in axes
+        ]
+
+        assert result_xticklabels == xticklabels
+        assert result_titles == titles
+
+    @pytest.mark.parametrize(
+        "by, column",
+        [
+            ([], ["A"]),
+            ((), "A"),
+            ([], None),
+            ((), ["A", "B"]),
+        ],
+    )
+    def test_box_plot_with_none_empty_list_by(self, by, column, hist_df):
+        # GH 15079
+        msg = "No group keys passed"
+        with pytest.raises(ValueError, match=msg):
+            _check_plot_works(hist_df.plot.box, default_axes=True, column=column, by=by)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "by, column, layout, axes_num",
+        [
+            (["C"], "A", (1, 1), 1),
+            ("C", "A", (1, 1), 1),
+            ("C", None, (2, 1), 2),
+            ("C", ["A", "B"], (1, 2), 2),
+            (["C", "D"], "A", (1, 1), 1),
+            (["C", "D"], None, (1, 2), 2),
+        ],
+    )
+    def test_box_plot_layout_with_by(self, by, column, layout, axes_num, hist_df):
+        # GH 15079
+        axes = _check_plot_works(
+            hist_df.plot.box, default_axes=True, column=column, by=by, layout=layout
+        )
+        _check_axes_shape(axes, axes_num=axes_num, layout=layout)
+
+    @pytest.mark.parametrize(
+        "msg, by, layout",
+        [
+            ("larger than required size", ["C", "D"], (1, 1)),
+            (re.escape("Layout must be a tuple of (rows, columns)"), "C", (1,)),
+            ("At least one dimension of layout must be positive", "C", (-1, -1)),
+        ],
+    )
+    def test_box_plot_invalid_layout_with_by_raises(self, msg, by, layout, hist_df):
+        # GH 15079, test if error is raised when invalid layout is given
+
+        with pytest.raises(ValueError, match=msg):
+            hist_df.plot.box(column=["A", "B"], by=by, layout=layout)
+
+    @pytest.mark.parametrize("figsize", [(12, 8), (20, 10)])
+    def test_figure_shape_hist_with_by(self, figsize, hist_df):
+        # GH 15079
+        axes = hist_df.plot.box(column="A", by="C", figsize=figsize)
+        _check_axes_shape(axes, axes_num=1, figsize=figsize)
diff --git a/pandas/pandas/tests/plotting/test_backend.py b/pandas/pandas/tests/plotting/test_backend.py
new file mode 100644
index 0000000000000..8a8643415ae12
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_backend.py
@@ -0,0 +1,98 @@
+import sys
+import types
+
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas
+
+
+@pytest.fixture
+def dummy_backend():
+    db = types.ModuleType("pandas_dummy_backend")
+    setattr(db, "plot", lambda *args, **kwargs: "used_dummy")
+    return db
+
+
+@pytest.fixture
+def restore_backend():
+    """Restore the plotting backend to matplotlib"""
+    with pandas.option_context("plotting.backend", "matplotlib"):
+        yield
+
+
+def test_backend_is_not_module():
+    msg = "Could not find plotting backend 'not_an_existing_module'."
+    with pytest.raises(ValueError, match=msg):
+        pandas.set_option("plotting.backend", "not_an_existing_module")
+
+    assert pandas.options.plotting.backend == "matplotlib"
+
+
+def test_backend_is_correct(monkeypatch, restore_backend, dummy_backend):
+    monkeypatch.setitem(sys.modules, "pandas_dummy_backend", dummy_backend)
+
+    pandas.set_option("plotting.backend", "pandas_dummy_backend")
+    assert pandas.get_option("plotting.backend") == "pandas_dummy_backend"
+    assert (
+        pandas.plotting._core._get_plot_backend("pandas_dummy_backend") is dummy_backend
+    )
+
+
+def test_backend_can_be_set_in_plot_call(monkeypatch, restore_backend, dummy_backend):
+    monkeypatch.setitem(sys.modules, "pandas_dummy_backend", dummy_backend)
+    df = pandas.DataFrame([1, 2, 3])
+
+    assert pandas.get_option("plotting.backend") == "matplotlib"
+    assert df.plot(backend="pandas_dummy_backend") == "used_dummy"
+
+
+def test_register_entrypoint(restore_backend, tmp_path, monkeypatch, dummy_backend):
+    monkeypatch.syspath_prepend(tmp_path)
+    monkeypatch.setitem(sys.modules, "pandas_dummy_backend", dummy_backend)
+
+    dist_info = tmp_path / "my_backend-0.0.0.dist-info"
+    dist_info.mkdir()
+    # entry_point name should not match module name - otherwise pandas will
+    # fall back to backend lookup by module name
+    (dist_info / "entry_points.txt").write_bytes(
+        b"[pandas_plotting_backends]\nmy_ep_backend = pandas_dummy_backend\n"
+    )
+
+    assert pandas.plotting._core._get_plot_backend("my_ep_backend") is dummy_backend
+
+    with pandas.option_context("plotting.backend", "my_ep_backend"):
+        assert pandas.plotting._core._get_plot_backend() is dummy_backend
+
+
+def test_setting_backend_without_plot_raises(monkeypatch):
+    # GH-28163
+    module = types.ModuleType("pandas_plot_backend")
+    monkeypatch.setitem(sys.modules, "pandas_plot_backend", module)
+
+    assert pandas.options.plotting.backend == "matplotlib"
+    with pytest.raises(
+        ValueError, match="Could not find plotting backend 'pandas_plot_backend'."
+    ):
+        pandas.set_option("plotting.backend", "pandas_plot_backend")
+
+    assert pandas.options.plotting.backend == "matplotlib"
+
+
+@td.skip_if_installed("matplotlib")
+def test_no_matplotlib_ok():
+    msg = (
+        'matplotlib is required for plotting when the default backend "matplotlib" is '
+        "selected."
+    )
+    with pytest.raises(ImportError, match=msg):
+        pandas.plotting._core._get_plot_backend("matplotlib")
+
+
+def test_extra_kinds_ok(monkeypatch, restore_backend, dummy_backend):
+    # https://github.com/pandas-dev/pandas/pull/28647
+    monkeypatch.setitem(sys.modules, "pandas_dummy_backend", dummy_backend)
+    pandas.set_option("plotting.backend", "pandas_dummy_backend")
+    df = pandas.DataFrame({"A": [1, 2, 3]})
+    df.plot(kind="not a real kind")
diff --git a/pandas/pandas/tests/plotting/test_boxplot_method.py b/pandas/pandas/tests/plotting/test_boxplot_method.py
new file mode 100644
index 0000000000000..2267b6197cd80
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_boxplot_method.py
@@ -0,0 +1,774 @@
+"""Test cases for .boxplot method"""
+
+from __future__ import annotations
+
+import itertools
+import string
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    date_range,
+    plotting,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_axes_shape,
+    _check_box_return_type,
+    _check_plot_works,
+    _check_ticks_props,
+    _check_visible,
+)
+from pandas.util.version import Version
+
+from pandas.io.formats.printing import pprint_thing
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+
+def _check_ax_limits(col, ax):
+    y_min, y_max = ax.get_ylim()
+    assert y_min <= col.min()
+    assert y_max >= col.max()
+
+
+if Version(mpl.__version__) < Version("3.10"):
+    verts: list[dict[str, bool | str]] = [{"vert": False}, {"vert": True}]
+else:
+    verts = [{"orientation": "horizontal"}, {"orientation": "vertical"}]
+
+
+@pytest.fixture(params=verts)
+def vert(request):
+    return request.param
+
+
+class TestDataFramePlots:
+    def test_stacked_boxplot_set_axis(self):
+        # GH2980
+        n = 30
+        df = DataFrame(
+            {
+                "Clinical": np.random.default_rng(2).choice([0, 1, 2, 3], n),
+                "Confirmed": np.random.default_rng(2).choice([0, 1, 2, 3], n),
+                "Discarded": np.random.default_rng(2).choice([0, 1, 2, 3], n),
+            },
+            index=np.arange(0, n),
+        )
+        ax = df.plot(kind="bar", stacked=True)
+        assert [int(x.get_text()) for x in ax.get_xticklabels()] == df.index.to_list()
+        ax.set_xticks(np.arange(0, n, 10))
+        plt.draw()  # Update changes
+        assert [int(x.get_text()) for x in ax.get_xticklabels()] == list(
+            np.arange(0, n, 10)
+        )
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs, warn",
+        [
+            [{"return_type": "dict"}, None],
+            [{"column": ["one", "two"]}, None],
+            [{"column": ["one", "two"], "by": "indic"}, UserWarning],
+            [{"column": ["one"], "by": ["indic", "indic2"]}, None],
+            [{"by": "indic"}, UserWarning],
+            [{"by": ["indic", "indic2"]}, UserWarning],
+            [{"notch": 1}, None],
+            [{"by": "indic", "notch": 1}, UserWarning],
+        ],
+    )
+    def test_boxplot_legacy1(self, kwargs, warn):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+        df["indic"] = ["foo", "bar"] * 3
+        df["indic2"] = ["foo", "bar", "foo"] * 2
+
+        # _check_plot_works can add an ax so catch warning. see GH #13188
+        with tm.assert_produces_warning(warn, check_stacklevel=False):
+            _check_plot_works(df.boxplot, **kwargs)
+
+    def test_boxplot_legacy1_series(self):
+        ser = Series(np.random.default_rng(2).standard_normal(6))
+        _check_plot_works(plotting._core.boxplot, data=ser, return_type="dict")
+
+    def test_boxplot_legacy2(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+        df["Y"] = Series(["A"] * 10)
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(df.boxplot, by="X")
+
+    def test_boxplot_legacy2_with_ax(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+        df["Y"] = Series(["A"] * 10)
+        # When ax is supplied and required number of axes is 1,
+        # passed ax should be used:
+        _, ax = mpl.pyplot.subplots()
+        axes = df.boxplot("Col1", by="X", ax=ax)
+        ax_axes = ax.axes
+        assert ax_axes is axes
+
+    def test_boxplot_legacy2_with_ax_return_type(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+        df["Y"] = Series(["A"] * 10)
+        fig, ax = mpl.pyplot.subplots()
+        axes = df.groupby("Y").boxplot(ax=ax, return_type="axes")
+        ax_axes = ax.axes
+        assert ax_axes is axes["A"]
+
+    def test_boxplot_legacy2_with_multi_col(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+        df["Y"] = Series(["A"] * 10)
+        # Multiple columns with an ax argument should use same figure
+        fig, ax = mpl.pyplot.subplots()
+        msg = "the figure containing the passed axes is being cleared"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            axes = df.boxplot(
+                column=["Col1", "Col2"], by="X", ax=ax, return_type="axes"
+            )
+        assert axes["Col1"].get_figure() is fig
+
+    def test_boxplot_legacy2_by_none(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(["A", "A", "A", "A", "A", "B", "B", "B", "B", "B"])
+        df["Y"] = Series(["A"] * 10)
+        # When by is None, check that all relevant lines are present in the
+        # dict
+        _, ax = mpl.pyplot.subplots()
+        d = df.boxplot(ax=ax, return_type="dict")
+        lines = list(itertools.chain.from_iterable(d.values()))
+        assert len(ax.get_lines()) == len(lines)
+
+    def test_boxplot_return_type_none(self, hist_df):
+        # GH 12216; return_type=None & by=None -> axes
+        result = hist_df.boxplot()
+        assert isinstance(result, mpl.pyplot.Axes)
+
+    def test_boxplot_return_type_legacy(self):
+        # API change in https://github.com/pandas-dev/pandas/pull/7096
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+        msg = "return_type must be {'axes', 'dict', 'both'}"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(return_type="NOT_A_TYPE")
+
+        result = df.boxplot()
+        _check_box_return_type(result, "axes")
+
+    @pytest.mark.parametrize("return_type", ["dict", "axes", "both"])
+    def test_boxplot_return_type_legacy_return_type(self, return_type):
+        # API change in https://github.com/pandas-dev/pandas/pull/7096
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 4)),
+            index=list(string.ascii_letters[:6]),
+            columns=["one", "two", "three", "four"],
+        )
+        with tm.assert_produces_warning(False):
+            result = df.boxplot(return_type=return_type)
+        _check_box_return_type(result, return_type)
+
+    def test_boxplot_axis_limits(self, hist_df):
+        df = hist_df.copy()
+        df["age"] = np.random.default_rng(2).integers(1, 20, df.shape[0])
+        # One full row
+        height_ax, weight_ax = df.boxplot(["height", "weight"], by="category")
+        _check_ax_limits(df["height"], height_ax)
+        _check_ax_limits(df["weight"], weight_ax)
+        assert weight_ax._sharey == height_ax
+
+    def test_boxplot_axis_limits_two_rows(self, hist_df):
+        df = hist_df.copy()
+        df["age"] = np.random.default_rng(2).integers(1, 20, df.shape[0])
+        # Two rows, one partial
+        p = df.boxplot(["height", "weight", "age"], by="category")
+        height_ax, weight_ax, age_ax = p[0, 0], p[0, 1], p[1, 0]
+        dummy_ax = p[1, 1]
+
+        _check_ax_limits(df["height"], height_ax)
+        _check_ax_limits(df["weight"], weight_ax)
+        _check_ax_limits(df["age"], age_ax)
+        assert weight_ax._sharey == height_ax
+        assert age_ax._sharey == height_ax
+        assert dummy_ax._sharey is None
+
+    def test_boxplot_empty_column(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((20, 4)))
+        df.loc[:, 0] = np.nan
+        _check_plot_works(df.boxplot, return_type="axes")
+
+    def test_figsize(self):
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 5)), columns=["A", "B", "C", "D", "E"]
+        )
+        result = df.boxplot(return_type="axes", figsize=(12, 8))
+        assert result.figure.bbox_inches.width == 12
+        assert result.figure.bbox_inches.height == 8
+
+    def test_fontsize(self):
+        df = DataFrame({"a": [1, 2, 3, 4, 5, 6]})
+        _check_ticks_props(df.boxplot("a", fontsize=16), xlabelsize=16, ylabelsize=16)
+
+    def test_boxplot_numeric_data(self):
+        # GH 22799
+        df = DataFrame(
+            {
+                "a": date_range("2012-01-01", periods=10),
+                "b": np.random.default_rng(2).standard_normal(10),
+                "c": np.random.default_rng(2).standard_normal(10) + 2,
+                "d": date_range("2012-01-01", periods=10).astype(str),
+                "e": date_range("2012-01-01", periods=10, tz="UTC"),
+                "f": timedelta_range("1 days", periods=10),
+            }
+        )
+        ax = df.plot(kind="box")
+        assert [x.get_text() for x in ax.get_xticklabels()] == ["b", "c"]
+
+    @pytest.mark.parametrize(
+        "colors_kwd, expected",
+        [
+            (
+                {"boxes": "r", "whiskers": "b", "medians": "g", "caps": "c"},
+                {"boxes": "r", "whiskers": "b", "medians": "g", "caps": "c"},
+            ),
+            ({"boxes": "r"}, {"boxes": "r"}),
+            ("r", {"boxes": "r", "whiskers": "r", "medians": "r", "caps": "r"}),
+        ],
+    )
+    def test_color_kwd(self, colors_kwd, expected):
+        # GH: 26214
+        df = DataFrame(np.random.default_rng(2).random((10, 2)))
+        result = df.boxplot(color=colors_kwd, return_type="dict")
+        for k, v in expected.items():
+            assert result[k][0].get_color() == v
+
+    @pytest.mark.parametrize(
+        "scheme,expected",
+        [
+            (
+                "dark_background",
+                {
+                    "boxes": "#8dd3c7",
+                    "whiskers": "#8dd3c7",
+                    "medians": "#bfbbd9",
+                    "caps": "#8dd3c7",
+                },
+            ),
+            (
+                "default",
+                {
+                    "boxes": "#1f77b4",
+                    "whiskers": "#1f77b4",
+                    "medians": "#2ca02c",
+                    "caps": "#1f77b4",
+                },
+            ),
+        ],
+    )
+    def test_colors_in_theme(self, scheme, expected):
+        # GH: 40769
+        df = DataFrame(np.random.default_rng(2).random((10, 2)))
+        plt.style.use(scheme)
+        result = df.plot.box(return_type="dict")
+        for k, v in expected.items():
+            assert result[k][0].get_color() == v
+
+    @pytest.mark.parametrize(
+        "dict_colors, msg",
+        [({"boxes": "r", "invalid_key": "r"}, "invalid key 'invalid_key'")],
+    )
+    def test_color_kwd_errors(self, dict_colors, msg):
+        # GH: 26214
+        df = DataFrame(np.random.default_rng(2).random((10, 2)))
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(color=dict_colors, return_type="dict")
+
+    @pytest.mark.parametrize(
+        "props, expected",
+        [
+            ("boxprops", "boxes"),
+            ("whiskerprops", "whiskers"),
+            ("capprops", "caps"),
+            ("medianprops", "medians"),
+        ],
+    )
+    def test_specified_props_kwd(self, props, expected):
+        # GH 30346
+        df = DataFrame({k: np.random.default_rng(2).random(10) for k in "ABC"})
+        kwd = {props: {"color": "C1"}}
+        result = df.boxplot(return_type="dict", **kwd)
+
+        assert result[expected][0].get_color() == "C1"
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    def test_plot_xlabel_ylabel(self, vert):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+                "group": np.random.default_rng(2).choice(["group1", "group2"], 10),
+            }
+        )
+        xlabel, ylabel = "x", "y"
+        ax = df.plot(kind="box", xlabel=xlabel, ylabel=ylabel, **vert)
+        assert ax.get_xlabel() == xlabel
+        assert ax.get_ylabel() == ylabel
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    def test_plot_box(self, vert):
+        # GH 54941
+        rng = np.random.default_rng(2)
+        df1 = DataFrame(rng.integers(0, 100, size=(10, 4)), columns=list("ABCD"))
+        df2 = DataFrame(rng.integers(0, 100, size=(10, 4)), columns=list("ABCD"))
+
+        xlabel, ylabel = "x", "y"
+        _, axs = plt.subplots(ncols=2, figsize=(10, 7), sharey=True)
+        df1.plot.box(ax=axs[0], xlabel=xlabel, ylabel=ylabel, **vert)
+        df2.plot.box(ax=axs[1], xlabel=xlabel, ylabel=ylabel, **vert)
+        for ax in axs:
+            assert ax.get_xlabel() == xlabel
+            assert ax.get_ylabel() == ylabel
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    def test_boxplot_xlabel_ylabel(self, vert):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+                "group": np.random.default_rng(2).choice(["group1", "group2"], 10),
+            }
+        )
+        xlabel, ylabel = "x", "y"
+        ax = df.boxplot(xlabel=xlabel, ylabel=ylabel, **vert)
+        assert ax.get_xlabel() == xlabel
+        assert ax.get_ylabel() == ylabel
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    def test_boxplot_group_xlabel_ylabel(self, vert):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+                "group": np.random.default_rng(2).choice(["group1", "group2"], 10),
+            }
+        )
+        xlabel, ylabel = "x", "y"
+        ax = df.boxplot(by="group", xlabel=xlabel, ylabel=ylabel, **vert)
+        for subplot in ax:
+            assert subplot.get_xlabel() == xlabel
+            assert subplot.get_ylabel() == ylabel
+
+    @pytest.mark.filterwarnings("ignore:set_ticklabels:UserWarning")
+    def test_boxplot_group_no_xlabel_ylabel(self, vert, request):
+        if Version(mpl.__version__) >= Version("3.10") and vert == {
+            "orientation": "horizontal"
+        }:
+            request.applymarker(
+                pytest.mark.xfail(reason=f"{vert} fails starting with matplotlib 3.10")
+            )
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+                "group": np.random.default_rng(2).choice(["group1", "group2"], 10),
+            }
+        )
+        ax = df.boxplot(by="group", **vert)
+        for subplot in ax:
+            target_label = (
+                subplot.get_xlabel()
+                if vert == {"vert": True} or vert == {"orientation": "vertical"}
+                else subplot.get_ylabel()
+            )
+            assert target_label == pprint_thing(["group"])
+
+
+class TestDataFrameGroupByPlots:
+    def test_boxplot_legacy1(self, hist_df):
+        grouped = hist_df.groupby(by="gender")
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(grouped.boxplot, return_type="axes")
+        _check_axes_shape(list(axes.values), axes_num=2, layout=(1, 2))
+
+    def test_boxplot_legacy1_return_type(self, hist_df):
+        grouped = hist_df.groupby(by="gender")
+        axes = _check_plot_works(grouped.boxplot, subplots=False, return_type="axes")
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+
+    @pytest.mark.slow
+    def test_boxplot_legacy2(self):
+        tuples = zip(string.ascii_letters[:10], range(10))
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=MultiIndex.from_tuples(tuples),
+        )
+        grouped = df.groupby(level=1)
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(grouped.boxplot, return_type="axes")
+        _check_axes_shape(list(axes.values), axes_num=10, layout=(4, 3))
+
+    @pytest.mark.slow
+    def test_boxplot_legacy2_return_type(self):
+        tuples = zip(string.ascii_letters[:10], range(10))
+        df = DataFrame(
+            np.random.default_rng(2).random((10, 3)),
+            index=MultiIndex.from_tuples(tuples),
+        )
+        grouped = df.groupby(level=1)
+        axes = _check_plot_works(grouped.boxplot, subplots=False, return_type="axes")
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+
+    def test_grouped_plot_fignums(self):
+        n = 10
+        weight = Series(np.random.default_rng(2).normal(166, 20, size=n))
+        height = Series(np.random.default_rng(2).normal(60, 10, size=n))
+        gender = np.random.default_rng(2).choice(["male", "female"], size=n)
+        df = DataFrame({"height": height, "weight": weight, "gender": gender})
+        gb = df.groupby("gender")
+
+        res = gb.plot()
+        assert len(mpl.pyplot.get_fignums()) == 2
+        assert len(res) == 2
+        plt.close("all")
+
+        res = gb.boxplot(return_type="axes")
+        assert len(mpl.pyplot.get_fignums()) == 1
+        assert len(res) == 2
+
+    def test_grouped_plot_fignums_excluded_col(self):
+        n = 10
+        weight = Series(np.random.default_rng(2).normal(166, 20, size=n))
+        height = Series(np.random.default_rng(2).normal(60, 10, size=n))
+        gender = np.random.default_rng(2).choice(["male", "female"], size=n)
+        df = DataFrame({"height": height, "weight": weight, "gender": gender})
+        # now works with GH 5610 as gender is excluded
+        df.groupby("gender").hist()
+
+    @pytest.mark.slow
+    def test_grouped_box_return_type(self, hist_df):
+        df = hist_df
+
+        # old style: return_type=None
+        result = df.boxplot(by="gender")
+        assert isinstance(result, np.ndarray)
+        _check_box_return_type(
+            result, None, expected_keys=["height", "weight", "category"]
+        )
+
+    @pytest.mark.slow
+    def test_grouped_box_return_type_groupby(self, hist_df):
+        df = hist_df
+        # now for groupby
+        result = df.groupby("gender").boxplot(return_type="dict")
+        _check_box_return_type(result, "dict", expected_keys=["Male", "Female"])
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("return_type", ["dict", "axes", "both"])
+    def test_grouped_box_return_type_arg(self, hist_df, return_type):
+        df = hist_df
+
+        returned = df.groupby("classroom").boxplot(return_type=return_type)
+        _check_box_return_type(returned, return_type, expected_keys=["A", "B", "C"])
+
+        returned = df.boxplot(by="classroom", return_type=return_type)
+        _check_box_return_type(
+            returned, return_type, expected_keys=["height", "weight", "category"]
+        )
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("return_type", ["dict", "axes", "both"])
+    def test_grouped_box_return_type_arg_duplcate_cats(self, return_type):
+        columns2 = "X B C D A".split()
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((6, 5)), columns=columns2
+        )
+        categories2 = "A B".split()
+        df2["category"] = categories2 * 3
+
+        returned = df2.groupby("category").boxplot(return_type=return_type)
+        _check_box_return_type(returned, return_type, expected_keys=categories2)
+
+        returned = df2.boxplot(by="category", return_type=return_type)
+        _check_box_return_type(returned, return_type, expected_keys=columns2)
+
+    @pytest.mark.slow
+    def test_grouped_box_layout_too_small(self, hist_df):
+        df = hist_df
+
+        msg = "Layout of 1x1 must be larger than required size 2"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(column=["weight", "height"], by=df.gender, layout=(1, 1))
+
+    @pytest.mark.slow
+    def test_grouped_box_layout_needs_by(self, hist_df):
+        df = hist_df
+        msg = "The 'layout' keyword is not supported when 'by' is None"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(
+                column=["height", "weight", "category"],
+                layout=(2, 1),
+                return_type="dict",
+            )
+
+    @pytest.mark.slow
+    def test_grouped_box_layout_positive_layout(self, hist_df):
+        df = hist_df
+        msg = "At least one dimension of layout must be positive"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(column=["weight", "height"], by=df.gender, layout=(-1, -1))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "gb_key, axes_num, rows",
+        [["gender", 2, 1], ["category", 4, 2], ["classroom", 3, 2]],
+    )
+    def test_grouped_box_layout_positive_layout_axes(
+        self, hist_df, gb_key, axes_num, rows
+    ):
+        df = hist_df
+        # _check_plot_works adds an ax so catch warning. see GH #13188 GH 6769
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(
+                df.groupby(gb_key).boxplot, column="height", return_type="dict"
+            )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=axes_num, layout=(rows, 2))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "col, visible", [["height", False], ["weight", True], ["category", True]]
+    )
+    def test_grouped_box_layout_visible(self, hist_df, col, visible):
+        df = hist_df
+        # GH 5897
+        axes = df.boxplot(
+            column=["height", "weight", "category"], by="gender", return_type="axes"
+        )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=3, layout=(2, 2))
+        ax = axes[col]
+        _check_visible(ax.get_xticklabels(), visible=visible)
+        _check_visible([ax.xaxis.get_label()], visible=visible)
+
+    @pytest.mark.slow
+    def test_grouped_box_layout_shape(self, hist_df):
+        df = hist_df
+        df.groupby("classroom").boxplot(
+            column=["height", "weight", "category"], return_type="dict"
+        )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=3, layout=(2, 2))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("cols", [2, -1])
+    def test_grouped_box_layout_works(self, hist_df, cols):
+        df = hist_df
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(
+                df.groupby("category").boxplot,
+                column="height",
+                layout=(3, cols),
+                return_type="dict",
+            )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=4, layout=(3, 2))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("rows, res", [[4, 4], [-1, 3]])
+    def test_grouped_box_layout_axes_shape_rows(self, hist_df, rows, res):
+        df = hist_df
+        df.boxplot(
+            column=["height", "weight", "category"], by="gender", layout=(rows, 1)
+        )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=3, layout=(res, 1))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("cols, res", [[4, 4], [-1, 3]])
+    def test_grouped_box_layout_axes_shape_cols_groupby(self, hist_df, cols, res):
+        df = hist_df
+        df.groupby("classroom").boxplot(
+            column=["height", "weight", "category"],
+            layout=(1, cols),
+            return_type="dict",
+        )
+        _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=3, layout=(1, res))
+
+    @pytest.mark.slow
+    def test_grouped_box_multiple_axes(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+
+        # check warning to ignore sharex / sharey
+        # this check should be done in the first function which
+        # passes multiple axes to plot, hist or boxplot
+        # location should be changed if other test is added
+        # which has earlier alphabetical order
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            _, axes = mpl.pyplot.subplots(2, 2)
+            df.groupby("category").boxplot(column="height", return_type="axes", ax=axes)
+            _check_axes_shape(mpl.pyplot.gcf().axes, axes_num=4, layout=(2, 2))
+
+    @pytest.mark.slow
+    def test_grouped_box_multiple_axes_on_fig(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+        fig, axes = mpl.pyplot.subplots(2, 3)
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            returned = df.boxplot(
+                column=["height", "weight", "category"],
+                by="gender",
+                return_type="axes",
+                ax=axes[0],
+            )
+        returned = np.array(list(returned.values))
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        tm.assert_numpy_array_equal(returned, axes[0])
+        assert returned[0].figure is fig
+
+        # draw on second row
+        with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+            returned = df.groupby("classroom").boxplot(
+                column=["height", "weight", "category"], return_type="axes", ax=axes[1]
+            )
+        returned = np.array(list(returned.values))
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        tm.assert_numpy_array_equal(returned, axes[1])
+        assert returned[0].figure is fig
+
+    @pytest.mark.slow
+    def test_grouped_box_multiple_axes_ax_error(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+        msg = "The number of passed axes must be 3, the same as the output plot"
+        _, axes = mpl.pyplot.subplots(2, 3)
+        with pytest.raises(ValueError, match=msg):
+            # pass different number of axes from required
+            with tm.assert_produces_warning(UserWarning, match="sharex and sharey"):
+                axes = df.groupby("classroom").boxplot(ax=axes)
+
+    def test_fontsize(self):
+        df = DataFrame({"a": [1, 2, 3, 4, 5, 6], "b": [0, 0, 0, 1, 1, 1]})
+        _check_ticks_props(
+            df.boxplot("a", by="b", fontsize=16), xlabelsize=16, ylabelsize=16
+        )
+
+    @pytest.mark.parametrize(
+        "col, expected_xticklabel",
+        [
+            ("v", ["(a, v)", "(b, v)", "(c, v)", "(d, v)", "(e, v)"]),
+            (["v"], ["(a, v)", "(b, v)", "(c, v)", "(d, v)", "(e, v)"]),
+            ("v1", ["(a, v1)", "(b, v1)", "(c, v1)", "(d, v1)", "(e, v1)"]),
+            (
+                ["v", "v1"],
+                [
+                    "(a, v)",
+                    "(a, v1)",
+                    "(b, v)",
+                    "(b, v1)",
+                    "(c, v)",
+                    "(c, v1)",
+                    "(d, v)",
+                    "(d, v1)",
+                    "(e, v)",
+                    "(e, v1)",
+                ],
+            ),
+            (
+                None,
+                [
+                    "(a, v)",
+                    "(a, v1)",
+                    "(b, v)",
+                    "(b, v1)",
+                    "(c, v)",
+                    "(c, v1)",
+                    "(d, v)",
+                    "(d, v1)",
+                    "(e, v)",
+                    "(e, v1)",
+                ],
+            ),
+        ],
+    )
+    def test_groupby_boxplot_subplots_false(self, col, expected_xticklabel):
+        # GH 16748
+        df = DataFrame(
+            {
+                "cat": np.random.default_rng(2).choice(list("abcde"), 100),
+                "v": np.random.default_rng(2).random(100),
+                "v1": np.random.default_rng(2).random(100),
+            }
+        )
+        grouped = df.groupby("cat")
+
+        axes = _check_plot_works(
+            grouped.boxplot, subplots=False, column=col, return_type="axes"
+        )
+
+        result_xticklabel = [x.get_text() for x in axes.get_xticklabels()]
+        assert expected_xticklabel == result_xticklabel
+
+    def test_groupby_boxplot_object(self, hist_df):
+        # GH 43480
+        df = hist_df.astype("object")
+        grouped = df.groupby("gender")
+        msg = "boxplot method requires numerical columns, nothing to plot"
+        with pytest.raises(ValueError, match=msg):
+            _check_plot_works(grouped.boxplot, subplots=False)
+
+    def test_boxplot_multiindex_column(self):
+        # GH 16748
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = list(zip(*arrays))
+        index = MultiIndex.from_tuples(tuples, names=["first", "second"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 8)),
+            index=["A", "B", "C"],
+            columns=index,
+        )
+
+        col = [("bar", "one"), ("bar", "two")]
+        axes = _check_plot_works(df.boxplot, column=col, return_type="axes")
+
+        expected_xticklabel = ["(bar, one)", "(bar, two)"]
+        result_xticklabel = [x.get_text() for x in axes.get_xticklabels()]
+        assert expected_xticklabel == result_xticklabel
+
+    @pytest.mark.parametrize("group", ["X", ["X", "Y"]])
+    def test_boxplot_multi_groupby_groups(self, group):
+        # GH 14701
+        rows = 20
+        df = DataFrame(
+            np.random.default_rng(12).normal(size=(rows, 2)), columns=["Col1", "Col2"]
+        )
+        df["X"] = Series(np.repeat(["A", "B"], int(rows / 2)))
+        df["Y"] = Series(np.tile(["C", "D"], int(rows / 2)))
+        grouped = df.groupby(group)
+        _check_plot_works(df.boxplot, by=group, default_axes=True)
+        _check_plot_works(df.plot.box, by=group, default_axes=True)
+        _check_plot_works(grouped.boxplot, default_axes=True)
diff --git a/pandas/pandas/tests/plotting/test_common.py b/pandas/pandas/tests/plotting/test_common.py
new file mode 100644
index 0000000000000..20daf59356248
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_common.py
@@ -0,0 +1,60 @@
+import pytest
+
+from pandas import DataFrame
+from pandas.tests.plotting.common import (
+    _check_plot_works,
+    _check_ticks_props,
+    _gen_two_subplots,
+)
+
+plt = pytest.importorskip("matplotlib.pyplot")
+
+
+class TestCommon:
+    def test__check_ticks_props(self):
+        # GH 34768
+        df = DataFrame({"b": [0, 1, 0], "a": [1, 2, 3]})
+        ax = _check_plot_works(df.plot, rot=30)
+        ax.yaxis.set_tick_params(rotation=30)
+        msg = "expected 0.00000 but got "
+        with pytest.raises(AssertionError, match=msg):
+            _check_ticks_props(ax, xrot=0)
+        with pytest.raises(AssertionError, match=msg):
+            _check_ticks_props(ax, xlabelsize=0)
+        with pytest.raises(AssertionError, match=msg):
+            _check_ticks_props(ax, yrot=0)
+        with pytest.raises(AssertionError, match=msg):
+            _check_ticks_props(ax, ylabelsize=0)
+
+    def test__gen_two_subplots_with_ax(self):
+        fig = plt.gcf()
+        gen = _gen_two_subplots(f=lambda **kwargs: None, fig=fig, ax="test")
+        # On the first yield, no subplot should be added since ax was passed
+        next(gen)
+        assert fig.get_axes() == []
+        # On the second, the one axis should match fig.subplot(2, 1, 2)
+        next(gen)
+        axes = fig.get_axes()
+        assert len(axes) == 1
+        subplot_geometry = list(axes[0].get_subplotspec().get_geometry()[:-1])
+        subplot_geometry[-1] += 1
+        assert subplot_geometry == [2, 1, 2]
+
+    def test_colorbar_layout(self):
+        fig = plt.figure()
+
+        axes = fig.subplot_mosaic(
+            """
+            AB
+            CC
+            """
+        )
+
+        x = [1, 2, 3]
+        y = [1, 2, 3]
+
+        cs0 = axes["A"].scatter(x, y)
+        axes["B"].scatter(x, y)
+
+        fig.colorbar(cs0, ax=[axes["A"], axes["B"]], location="right")
+        DataFrame(x).plot(ax=axes["C"])
diff --git a/pandas/pandas/tests/plotting/test_converter.py b/pandas/pandas/tests/plotting/test_converter.py
new file mode 100644
index 0000000000000..cfdfa7f723599
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_converter.py
@@ -0,0 +1,391 @@
+from datetime import (
+    date,
+    datetime,
+)
+import subprocess
+import sys
+
+import numpy as np
+import pytest
+
+import pandas._config.config as cf
+
+from pandas._libs.tslibs import to_offset
+
+from pandas import (
+    Index,
+    Period,
+    PeriodIndex,
+    Series,
+    Timestamp,
+    arrays,
+    date_range,
+)
+import pandas._testing as tm
+
+from pandas.plotting import (
+    deregister_matplotlib_converters,
+    register_matplotlib_converters,
+)
+from pandas.tseries.offsets import (
+    Day,
+    Micro,
+    Milli,
+    Second,
+)
+
+plt = pytest.importorskip("matplotlib.pyplot")
+dates = pytest.importorskip("matplotlib.dates")
+units = pytest.importorskip("matplotlib.units")
+
+from pandas.plotting._matplotlib import converter
+
+
+@pytest.mark.single_cpu
+def test_registry_mpl_resets():
+    # Check that Matplotlib converters are properly reset (see issue #27481)
+    code = (
+        "import matplotlib.units as units; "
+        "import matplotlib.dates as mdates; "
+        "n_conv = len(units.registry); "
+        "import pandas as pd; "
+        "pd.plotting.register_matplotlib_converters(); "
+        "pd.plotting.deregister_matplotlib_converters(); "
+        "assert len(units.registry) == n_conv"
+    )
+    call = [sys.executable, "-c", code]
+    subprocess.check_output(call)
+
+
+def test_timtetonum_accepts_unicode():
+    assert converter.time2num("00:01") == converter.time2num("00:01")
+
+
+class TestRegistration:
+    @pytest.mark.single_cpu
+    def test_dont_register_by_default(self):
+        # Run in subprocess to ensure a clean state
+        code = (
+            "import matplotlib.units; "
+            "import pandas as pd; "
+            "units = dict(matplotlib.units.registry); "
+            "assert pd.Timestamp not in units"
+        )
+        call = [sys.executable, "-c", code]
+        assert subprocess.check_call(call) == 0
+
+    def test_registering_no_warning(self):
+        s = Series(range(12), index=date_range("2017", periods=12))
+        _, ax = plt.subplots()
+
+        # Set to the "warn" state, in case this isn't the first test run
+        register_matplotlib_converters()
+        ax.plot(s.index, s.values)
+
+    def test_pandas_plots_register(self):
+        s = Series(range(12), index=date_range("2017", periods=12))
+        # Set to the "warn" state, in case this isn't the first test run
+        with tm.assert_produces_warning(None) as w:
+            s.plot()
+
+        assert len(w) == 0
+
+    def test_matplotlib_formatters(self):
+        # Can't make any assertion about the start state.
+        # We we check that toggling converters off removes it, and toggling it
+        # on restores it.
+
+        with cf.option_context("plotting.matplotlib.register_converters", True):
+            with cf.option_context("plotting.matplotlib.register_converters", False):
+                assert Timestamp not in units.registry
+            assert Timestamp in units.registry
+
+    def test_option_no_warning(self):
+        s = Series(range(12), index=date_range("2017", periods=12))
+        _, ax = plt.subplots()
+
+        # Test without registering first, no warning
+        with cf.option_context("plotting.matplotlib.register_converters", False):
+            ax.plot(s.index, s.values)
+
+        # Now test with registering
+        register_matplotlib_converters()
+        with cf.option_context("plotting.matplotlib.register_converters", False):
+            ax.plot(s.index, s.values)
+
+    def test_registry_resets(self):
+        # make a copy, to reset to
+        original = dict(units.registry)
+
+        try:
+            # get to a known state
+            units.registry.clear()
+            date_converter = dates.DateConverter()
+            units.registry[datetime] = date_converter
+            units.registry[date] = date_converter
+
+            register_matplotlib_converters()
+            assert units.registry[date] is not date_converter
+            deregister_matplotlib_converters()
+            assert units.registry[date] is date_converter
+
+        finally:
+            # restore original stater
+            units.registry.clear()
+            for k, v in original.items():
+                units.registry[k] = v
+
+
+class TestDateTimeConverter:
+    @pytest.fixture
+    def dtc(self):
+        return converter.DatetimeConverter()
+
+    def test_convert_accepts_unicode(self, dtc):
+        r1 = dtc.convert("2000-01-01 12:22", None, None)
+        r2 = dtc.convert("2000-01-01 12:22", None, None)
+        assert r1 == r2, "DatetimeConverter.convert should accept unicode"
+
+    def test_conversion(self, dtc):
+        rs = dtc.convert(["2012-1-1"], None, None)[0]
+        xp = dates.date2num(datetime(2012, 1, 1))
+        assert rs == xp
+
+        rs = dtc.convert("2012-1-1", None, None)
+        assert rs == xp
+
+        rs = dtc.convert(date(2012, 1, 1), None, None)
+        assert rs == xp
+
+        rs = dtc.convert("2012-1-1", None, None)
+        assert rs == xp
+
+        rs = dtc.convert(Timestamp("2012-1-1"), None, None)
+        assert rs == xp
+
+        # also testing datetime64 dtype (GH8614)
+        rs = dtc.convert("2012-01-01", None, None)
+        assert rs == xp
+
+        rs = dtc.convert("2012-01-01 00:00:00+0000", None, None)
+        assert rs == xp
+
+        rs = dtc.convert(
+            np.array(["2012-01-01 00:00:00+0000", "2012-01-02 00:00:00+0000"]),
+            None,
+            None,
+        )
+        assert rs[0] == xp
+
+        # we have a tz-aware date (constructed to that when we turn to utc it
+        # is the same as our sample)
+        ts = Timestamp("2012-01-01").tz_localize("UTC").tz_convert("US/Eastern")
+        rs = dtc.convert(ts, None, None)
+        assert rs == xp
+
+        rs = dtc.convert(ts.to_pydatetime(), None, None)
+        assert rs == xp
+
+        rs = dtc.convert(Index([ts - Day(1), ts]), None, None)
+        assert rs[1] == xp
+
+        rs = dtc.convert(Index([ts - Day(1), ts]).to_pydatetime(), None, None)
+        assert rs[1] == xp
+
+    def test_conversion_float(self, dtc):
+        rtol = 0.5 * 10**-9
+
+        rs = dtc.convert(Timestamp("2012-1-1 01:02:03", tz="UTC"), None, None)
+        xp = dates.date2num(Timestamp("2012-1-1 01:02:03", tz="UTC"))
+        tm.assert_almost_equal(rs, xp, rtol=rtol)
+
+        rs = dtc.convert(
+            Timestamp("2012-1-1 09:02:03", tz="Asia/Hong_Kong"), None, None
+        )
+        tm.assert_almost_equal(rs, xp, rtol=rtol)
+
+        rs = dtc.convert(datetime(2012, 1, 1, 1, 2, 3), None, None)
+        tm.assert_almost_equal(rs, xp, rtol=rtol)
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            [date(1677, 1, 1), date(1677, 1, 2)],
+            [datetime(1677, 1, 1, 12), datetime(1677, 1, 2, 12)],
+        ],
+    )
+    def test_conversion_outofbounds_datetime(self, dtc, values):
+        # 2579
+        rs = dtc.convert(values, None, None)
+        xp = dates.date2num(values)
+        tm.assert_numpy_array_equal(rs, xp)
+        rs = dtc.convert(values[0], None, None)
+        xp = dates.date2num(values[0])
+        assert rs == xp
+
+    @pytest.mark.parametrize(
+        "time,format_expected",
+        [
+            (0, "00:00"),  # time2num(datetime.time.min)
+            (86399.999999, "23:59:59.999999"),  # time2num(datetime.time.max)
+            (90000, "01:00"),
+            (3723, "01:02:03"),
+            (39723.2, "11:02:03.200"),
+        ],
+    )
+    def test_time_formatter(self, time, format_expected):
+        # issue 18478
+        result = converter.TimeFormatter(None)(time)
+        assert result == format_expected
+
+    @pytest.mark.parametrize("freq", ("B", "ms", "s"))
+    def test_dateindex_conversion(self, freq, dtc):
+        rtol = 10**-9
+        dateindex = date_range("2020-01-01", periods=10, freq=freq)
+        rs = dtc.convert(dateindex, None, None)
+        xp = dates.date2num(dateindex._mpl_repr())
+        tm.assert_almost_equal(rs, xp, rtol=rtol)
+
+    @pytest.mark.parametrize("offset", [Second(), Milli(), Micro(50)])
+    def test_resolution(self, offset, dtc):
+        # Matplotlib's time representation using floats cannot distinguish
+        # intervals smaller than ~10 microsecond in the common range of years.
+        ts1 = Timestamp("2012-1-1")
+        ts2 = ts1 + offset
+        val1 = dtc.convert(ts1, None, None)
+        val2 = dtc.convert(ts2, None, None)
+        if not val1 < val2:
+            raise AssertionError(f"{val1} is not less than {val2}.")
+
+    def test_convert_nested(self, dtc):
+        inner = [Timestamp("2017-01-01"), Timestamp("2017-01-02")]
+        data = [inner, inner]
+        result = dtc.convert(data, None, None)
+        expected = [dtc.convert(x, None, None) for x in data]
+        assert (np.array(result) == expected).all()
+
+
+class TestPeriodConverter:
+    @pytest.fixture
+    def pc(self):
+        return converter.PeriodConverter()
+
+    @pytest.fixture
+    def axis(self):
+        class Axis:
+            pass
+
+        axis = Axis()
+        axis.freq = "D"
+        return axis
+
+    def test_convert_accepts_unicode(self, pc, axis):
+        r1 = pc.convert("2012-1-1", None, axis)
+        r2 = pc.convert("2012-1-1", None, axis)
+        assert r1 == r2
+
+    def test_conversion(self, pc, axis):
+        rs = pc.convert(["2012-1-1"], None, axis)[0]
+        xp = Period("2012-1-1").ordinal
+        assert rs == xp
+
+        rs = pc.convert("2012-1-1", None, axis)
+        assert rs == xp
+
+        rs = pc.convert([date(2012, 1, 1)], None, axis)[0]
+        assert rs == xp
+
+        rs = pc.convert(date(2012, 1, 1), None, axis)
+        assert rs == xp
+
+        rs = pc.convert([Timestamp("2012-1-1")], None, axis)[0]
+        assert rs == xp
+
+        rs = pc.convert(Timestamp("2012-1-1"), None, axis)
+        assert rs == xp
+
+        rs = pc.convert("2012-01-01", None, axis)
+        assert rs == xp
+
+        rs = pc.convert("2012-01-01 00:00:00+0000", None, axis)
+        assert rs == xp
+
+        rs = pc.convert(
+            np.array(
+                ["2012-01-01 00:00:00", "2012-01-02 00:00:00"],
+                dtype="datetime64[ns]",
+            ),
+            None,
+            axis,
+        )
+        assert rs[0] == xp
+
+    def test_integer_passthrough(self, pc, axis):
+        # GH9012
+        rs = pc.convert([0, 1], None, axis)
+        xp = [0, 1]
+        assert rs == xp
+
+    def test_convert_nested(self, pc, axis):
+        data = ["2012-1-1", "2012-1-2"]
+        r1 = pc.convert([data, data], None, axis)
+        r2 = [pc.convert(data, None, axis) for _ in range(2)]
+        assert r1 == r2
+
+
+class TestTimeDeltaConverter:
+    """Test timedelta converter"""
+
+    @pytest.mark.parametrize(
+        "x, decimal, format_expected",
+        [
+            (0.0, 0, "00:00:00"),
+            (3972320000000, 1, "01:06:12.3"),
+            (713233432000000, 2, "8 days 06:07:13.43"),
+            (32423432000000, 4, "09:00:23.4320"),
+        ],
+    )
+    def test_format_timedelta_ticks(self, x, decimal, format_expected):
+        tdc = converter.TimeSeries_TimedeltaFormatter
+        result = tdc.format_timedelta_ticks(x, pos=None, n_decimals=decimal)
+        assert result == format_expected
+
+    @pytest.mark.parametrize("view_interval", [(1, 2), (2, 1)])
+    def test_call_w_different_view_intervals(self, view_interval, monkeypatch):
+        # previously broke on reversed xlmits; see GH37454
+        class mock_axis:
+            def get_view_interval(self):
+                return view_interval
+
+        tdc = converter.TimeSeries_TimedeltaFormatter()
+        monkeypatch.setattr(tdc, "axis", mock_axis())
+        tdc(0.0, 0)
+
+
+@pytest.mark.parametrize("year_span", [11.25, 30, 80, 150, 400, 800, 1500, 2500, 3500])
+# The range is limited to 11.25 at the bottom by if statements in
+# the _quarterly_finder() function
+def test_quarterly_finder(year_span):
+    vmin = -1000
+    vmax = vmin + year_span * 4
+    span = vmax - vmin + 1
+    if span < 45:
+        pytest.skip("the quarterly finder is only invoked if the span is >= 45")
+    nyears = span / 4
+    (min_anndef, maj_anndef) = converter._get_default_annual_spacing(nyears)
+    result = converter._quarterly_finder(vmin, vmax, to_offset("QE"))
+    quarters = PeriodIndex(
+        arrays.PeriodArray(np.array([x[0] for x in result]), dtype="period[Q]")
+    )
+    majors = np.array([x[1] for x in result])
+    minors = np.array([x[2] for x in result])
+    major_quarters = quarters[majors]
+    minor_quarters = quarters[minors]
+    check_major_years = major_quarters.year % maj_anndef == 0
+    check_minor_years = minor_quarters.year % min_anndef == 0
+    check_major_quarters = major_quarters.quarter == 1
+    check_minor_quarters = minor_quarters.quarter == 1
+    assert np.all(check_major_years)
+    assert np.all(check_minor_years)
+    assert np.all(check_major_quarters)
+    assert np.all(check_minor_quarters)
diff --git a/pandas/pandas/tests/plotting/test_datetimelike.py b/pandas/pandas/tests/plotting/test_datetimelike.py
new file mode 100644
index 0000000000000..1275f3d6f7d6d
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_datetimelike.py
@@ -0,0 +1,1719 @@
+"""Test cases for time series specific (freq conversion, etc)"""
+
+from datetime import (
+    date,
+    datetime,
+    time,
+    timedelta,
+)
+import pickle
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    BaseOffset,
+    to_offset,
+)
+
+from pandas.core.dtypes.dtypes import PeriodDtype
+
+from pandas import (
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    concat,
+    isna,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import (
+    DatetimeIndex,
+    bdate_range,
+    date_range,
+)
+from pandas.core.indexes.period import (
+    Period,
+    PeriodIndex,
+    period_range,
+)
+from pandas.core.indexes.timedeltas import timedelta_range
+from pandas.tests.plotting.common import _check_ticks_props
+
+from pandas.tseries.offsets import WeekOfMonth
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+import pandas.plotting._matplotlib.converter as conv
+
+
+class TestTSPlot:
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    def test_ts_plot_with_tz(self, tz_aware_fixture):
+        # GH2877, GH17173, GH31205, GH31580
+        tz = tz_aware_fixture
+        index = date_range("1/1/2011", periods=2, freq="h", tz=tz)
+        ts = Series([188.5, 328.25], index=index)
+        _check_plot_works(ts.plot)
+        ax = ts.plot()
+        xdata = next(iter(ax.get_lines())).get_xdata()
+        # Check first and last points' labels are correct
+        assert (xdata[0].hour, xdata[0].minute) == (0, 0)
+        assert (xdata[-1].hour, xdata[-1].minute) == (1, 0)
+
+    def test_fontsize_set_correctly(self):
+        # For issue #8765
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 9)), index=range(10)
+        )
+        _, ax = mpl.pyplot.subplots()
+        df.plot(fontsize=2, ax=ax)
+        for label in ax.get_xticklabels() + ax.get_yticklabels():
+            assert label.get_fontsize() == 2
+
+    def test_frame_inferred(self):
+        # inferred freq
+        idx = date_range("1/1/1987", freq="MS", periods=10)
+        idx = DatetimeIndex(idx.values, freq=None)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)), index=idx
+        )
+        _check_plot_works(df.plot)
+
+        # axes freq
+        idx = idx[0:4].union(idx[6:])
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)), index=idx
+        )
+        _check_plot_works(df2.plot)
+
+    def test_frame_inferred_n_gt_1(self):
+        # N > 1
+        idx = date_range("2008-1-1 00:15:00", freq="15min", periods=10)
+        idx = DatetimeIndex(idx.values, freq=None)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)), index=idx
+        )
+        _check_plot_works(df.plot)
+
+    def test_is_error_nozeroindex(self):
+        # GH11858
+        i = np.array([1, 2, 3])
+        a = DataFrame(i, index=i)
+        _check_plot_works(a.plot, xerr=a)
+        _check_plot_works(a.plot, yerr=a)
+
+    def test_nonnumeric_exclude(self):
+        idx = date_range("1/1/1987", freq="YE", periods=3)
+        df = DataFrame({"A": ["x", "y", "z"], "B": [1, 2, 3]}, idx)
+
+        fig, ax = mpl.pyplot.subplots()
+        df.plot(ax=ax)  # it works
+        assert len(ax.get_lines()) == 1  # B was plotted
+
+    def test_nonnumeric_exclude_error(self):
+        idx = date_range("1/1/1987", freq="YE", periods=3)
+        df = DataFrame({"A": ["x", "y", "z"], "B": [1, 2, 3]}, idx)
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            df["A"].plot()
+
+    @pytest.mark.parametrize("freq", ["s", "min", "h", "D", "W", "M", "Q", "Y"])
+    def test_tsplot_period(self, freq):
+        idx = period_range("12/31/1999", freq=freq, periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(ser.plot, ax=ax)
+
+    @pytest.mark.parametrize(
+        "freq", ["s", "min", "h", "D", "W", "ME", "QE-DEC", "YE", "1B30Min"]
+    )
+    def test_tsplot_datetime(self, freq):
+        idx = date_range("12/31/1999", freq=freq, periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(ser.plot, ax=ax)
+
+    def test_tsplot(self):
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        _, ax = mpl.pyplot.subplots()
+        ts.plot(style="k", ax=ax)
+        color = (0.0, 0.0, 0.0, 1)
+        assert color == ax.get_lines()[0].get_color()
+
+    @pytest.mark.parametrize("index", [None, date_range("2020-01-01", periods=10)])
+    def test_both_style_and_color(self, index):
+        ts = Series(np.arange(10, dtype=np.float64), index=index)
+        msg = (
+            "Cannot pass 'style' string with a color symbol and 'color' "
+            "keyword argument. Please use one or the other or pass 'style' "
+            "without a color symbol"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ts.plot(style="b-", color="#000099")
+
+    @pytest.mark.parametrize("freq", ["ms", "us"])
+    def test_high_freq(self, freq):
+        _, ax = mpl.pyplot.subplots()
+        rng = date_range("1/1/2012", periods=10, freq=freq)
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _check_plot_works(ser.plot, ax=ax)
+
+    def test_get_datevalue(self):
+        assert conv.get_datevalue(None, "D") is None
+        assert conv.get_datevalue(1987, "Y") == 1987
+        assert (
+            conv.get_datevalue(Period(1987, "Y"), "M") == Period("1987-12", "M").ordinal
+        )
+        assert conv.get_datevalue("1/1/1987", "D") == Period("1987-1-1", "D").ordinal
+
+    @pytest.mark.parametrize(
+        "freq, expected_string",
+        [["YE-DEC", "t = 2014  y = 1.000000"], ["D", "t = 2014-01-01  y = 1.000000"]],
+    )
+    def test_ts_plot_format_coord(self, freq, expected_string):
+        ser = Series(1, index=date_range("2014-01-01", periods=3, freq=freq))
+        _, ax = mpl.pyplot.subplots()
+        ser.plot(ax=ax)
+        first_line = ax.get_lines()[0]
+        first_x = first_line.get_xdata()[0].ordinal
+        first_y = first_line.get_ydata()[0]
+        assert expected_string == ax.format_coord(first_x, first_y)
+
+    @pytest.mark.parametrize("freq", ["s", "min", "h", "D", "W", "M", "Q", "Y"])
+    def test_line_plot_period_series(self, freq):
+        idx = period_range("12/31/1999", freq=freq, periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        _check_plot_works(ser.plot, ser.index.freq)
+
+    @pytest.mark.parametrize(
+        "frqncy", ["1s", "3s", "5min", "7h", "4D", "8W", "11M", "3Y"]
+    )
+    def test_line_plot_period_mlt_series(self, frqncy):
+        # test period index line plot for series with multiples (`mlt`) of the
+        # frequency (`frqncy`) rule code. tests resolution of issue #14763
+        idx = period_range("12/31/1999", freq=frqncy, periods=10)
+        s = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        _check_plot_works(s.plot, s.index.freq.rule_code)
+
+    @pytest.mark.parametrize(
+        "freq", ["s", "min", "h", "D", "W", "ME", "QE-DEC", "YE", "1B30Min"]
+    )
+    def test_line_plot_datetime_series(self, freq):
+        idx = date_range("12/31/1999", freq=freq, periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        _check_plot_works(ser.plot, ser.index.freq.rule_code)
+
+    @pytest.mark.parametrize("freq", ["s", "min", "h", "D", "W", "ME", "QE", "YE"])
+    def test_line_plot_period_frame(self, freq):
+        idx = date_range("12/31/1999", freq=freq, periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)),
+            index=idx,
+            columns=["A", "B", "C"],
+        )
+        _check_plot_works(df.plot, df.index.freq)
+
+    @pytest.mark.parametrize(
+        "frqncy", ["1s", "3s", "5min", "7h", "4D", "8W", "11M", "3Y"]
+    )
+    def test_line_plot_period_mlt_frame(self, frqncy):
+        # test period index line plot for DataFrames with multiples (`mlt`)
+        # of the frequency (`frqncy`) rule code. tests resolution of issue
+        # #14763
+        idx = period_range("12/31/1999", freq=frqncy, periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)),
+            index=idx,
+            columns=["A", "B", "C"],
+        )
+        freq = df.index.freq.rule_code
+        _check_plot_works(df.plot, freq)
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    @pytest.mark.parametrize(
+        "freq", ["s", "min", "h", "D", "W", "ME", "QE-DEC", "YE", "1B30Min"]
+    )
+    def test_line_plot_datetime_frame(self, freq):
+        idx = date_range("12/31/1999", freq=freq, periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)),
+            index=idx,
+            columns=["A", "B", "C"],
+        )
+        freq = PeriodDtype(df.index.freq)._freqstr
+        freq = df.index.to_period(freq).freq
+        _check_plot_works(df.plot, freq)
+
+    @pytest.mark.parametrize(
+        "freq", ["s", "min", "h", "D", "W", "ME", "QE-DEC", "YE", "1B30Min"]
+    )
+    def test_line_plot_inferred_freq(self, freq):
+        idx = date_range("12/31/1999", freq=freq, periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        ser = Series(ser.values, Index(np.asarray(ser.index)))
+        _check_plot_works(ser.plot, ser.index.inferred_freq)
+
+        ser = ser.iloc[[0, 3, 5, 6]]
+        _check_plot_works(ser.plot)
+
+    def test_fake_inferred_business(self):
+        _, ax = mpl.pyplot.subplots()
+        rng = date_range("2001-1-1", "2001-1-10")
+        ts = Series(range(len(rng)), index=rng)
+        ts = concat([ts[:3], ts[5:]])
+        ts.plot(ax=ax)
+        assert not hasattr(ax, "freq")
+
+    def test_plot_offset_freq(self):
+        ser = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        _check_plot_works(ser.plot)
+
+    def test_plot_offset_freq_business(self):
+        dr = date_range("2023-01-01", freq="BQS", periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(len(dr)), index=dr)
+        _check_plot_works(ser.plot)
+
+    def test_plot_multiple_inferred_freq(self):
+        dr = Index([datetime(2000, 1, 1), datetime(2000, 1, 6), datetime(2000, 1, 11)])
+        ser = Series(np.random.default_rng(2).standard_normal(len(dr)), index=dr)
+        _check_plot_works(ser.plot)
+
+    def test_irreg_hf(self):
+        idx = date_range("2012-6-22 21:59:51", freq="s", periods=10)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 2)), index=idx
+        )
+
+        irreg = df.iloc[[0, 1, 3, 4]]
+        _, ax = mpl.pyplot.subplots()
+        irreg.plot(ax=ax)
+        diffs = Series(ax.get_lines()[0].get_xydata()[:, 0]).diff()
+
+        sec = 1.0 / 24 / 60 / 60
+        assert (np.fabs(diffs[1:] - [sec, sec * 2, sec]) < 1e-8).all()
+
+    def test_irreg_hf_object(self):
+        idx = date_range("2012-6-22 21:59:51", freq="s", periods=10)
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 2)), index=idx
+        )
+        _, ax = mpl.pyplot.subplots()
+        df2.index = df2.index.astype(object)
+        df2.plot(ax=ax)
+        diffs = Series(ax.get_lines()[0].get_xydata()[:, 0]).diff()
+        sec = 1.0 / 24 / 60 / 60
+        assert (np.fabs(diffs[1:] - sec) < 1e-8).all()
+
+    def test_irregular_datetime64_repr_bug(self):
+        ser = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        ser = ser.iloc[[0, 1, 2, 7]]
+
+        _, ax = mpl.pyplot.subplots()
+
+        ret = ser.plot(ax=ax)
+        assert ret is not None
+
+        for rs, xp in zip(ax.get_lines()[0].get_xdata(), ser.index):
+            assert rs == xp
+
+    def test_business_freq(self):
+        bts = Series(range(5), period_range("2020-01-01", periods=5))
+        msg = r"PeriodDtype\[B\] is deprecated"
+        dt = bts.index[0].to_timestamp()
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            bts.index = period_range(start=dt, periods=len(bts), freq="B")
+        _, ax = mpl.pyplot.subplots()
+        bts.plot(ax=ax)
+        assert ax.get_lines()[0].get_xydata()[0, 0] == bts.index[0].ordinal
+        idx = ax.get_lines()[0].get_xdata()
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert PeriodIndex(data=idx).freqstr == "B"
+
+    def test_business_freq_convert(self):
+        bts = Series(
+            np.arange(50, dtype=np.float64),
+            index=date_range("2020-01-01", periods=50, freq="B"),
+        ).asfreq("BME")
+        ts = bts.to_period("M")
+        _, ax = mpl.pyplot.subplots()
+        bts.plot(ax=ax)
+        assert ax.get_lines()[0].get_xydata()[0, 0] == ts.index[0].ordinal
+        idx = ax.get_lines()[0].get_xdata()
+        assert PeriodIndex(data=idx).freqstr == "M"
+
+    def test_freq_with_no_period_alias(self):
+        # GH34487
+        freq = WeekOfMonth()
+        bts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        ).asfreq(freq)
+        _, ax = mpl.pyplot.subplots()
+        bts.plot(ax=ax)
+
+        idx = ax.get_lines()[0].get_xdata()
+        msg = "freq not specified and cannot be inferred"
+        with pytest.raises(ValueError, match=msg):
+            PeriodIndex(data=idx)
+
+    def test_nonzero_base(self):
+        # GH2571
+        idx = date_range("2012-12-20", periods=24, freq="h") + timedelta(minutes=30)
+        df = DataFrame(np.arange(24), index=idx)
+        _, ax = mpl.pyplot.subplots()
+        df.plot(ax=ax)
+        rs = ax.get_lines()[0].get_xdata()
+        assert not Index(rs).is_normalized
+
+    def test_dataframe(self):
+        bts = DataFrame(
+            {
+                "a": Series(
+                    np.arange(10, dtype=np.float64),
+                    index=date_range("2020-01-01", periods=10),
+                )
+            }
+        )
+        _, ax = mpl.pyplot.subplots()
+        bts.plot(ax=ax)
+        idx = ax.get_lines()[0].get_xdata()
+        tm.assert_index_equal(bts.index.to_period(), PeriodIndex(idx))
+
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    @pytest.mark.parametrize(
+        "obj",
+        [
+            Series(
+                np.arange(10, dtype=np.float64),
+                index=date_range("2020-01-01", periods=10),
+            ),
+            DataFrame(
+                {
+                    "a": Series(
+                        np.arange(10, dtype=np.float64),
+                        index=date_range("2020-01-01", periods=10),
+                    ),
+                    "b": Series(
+                        np.arange(10, dtype=np.float64),
+                        index=date_range("2020-01-01", periods=10),
+                    )
+                    + 1,
+                }
+            ),
+        ],
+    )
+    def test_axis_limits(self, obj):
+        _, ax = mpl.pyplot.subplots()
+        obj.plot(ax=ax)
+        xlim = ax.get_xlim()
+        ax.set_xlim(xlim[0] - 5, xlim[1] + 10)
+        result = ax.get_xlim()
+        assert result[0] == xlim[0] - 5
+        assert result[1] == xlim[1] + 10
+
+        # string
+        expected = (Period("1/1/2000", ax.freq), Period("4/1/2000", ax.freq))
+        ax.set_xlim("1/1/2000", "4/1/2000")
+        result = ax.get_xlim()
+        assert int(result[0]) == expected[0].ordinal
+        assert int(result[1]) == expected[1].ordinal
+
+        # datetime
+        expected = (Period("1/1/2000", ax.freq), Period("4/1/2000", ax.freq))
+        ax.set_xlim(datetime(2000, 1, 1), datetime(2000, 4, 1))
+        result = ax.get_xlim()
+        assert int(result[0]) == expected[0].ordinal
+        assert int(result[1]) == expected[1].ordinal
+
+    def test_get_finder(self):
+        assert conv.get_finder(to_offset("B")) == conv._daily_finder
+        assert conv.get_finder(to_offset("D")) == conv._daily_finder
+        assert conv.get_finder(to_offset("ME")) == conv._monthly_finder
+        assert conv.get_finder(to_offset("QE")) == conv._quarterly_finder
+        assert conv.get_finder(to_offset("YE")) == conv._annual_finder
+        assert conv.get_finder(to_offset("W")) == conv._daily_finder
+
+    def test_finder_daily(self):
+        day_lst = [10, 40, 252, 400, 950, 2750, 10000]
+
+        msg = "Period with BDay freq is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            xpl1 = xpl2 = [Period("1999-1-1", freq="B").ordinal] * len(day_lst)
+        rs1 = []
+        rs2 = []
+        for n in day_lst:
+            rng = bdate_range("1999-1-1", periods=n)
+            ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+            _, ax = mpl.pyplot.subplots()
+            ser.plot(ax=ax)
+            xaxis = ax.get_xaxis()
+            rs1.append(xaxis.get_majorticklocs()[0])
+
+            vmin, vmax = ax.get_xlim()
+            ax.set_xlim(vmin + 0.9, vmax)
+            rs2.append(xaxis.get_majorticklocs()[0])
+            mpl.pyplot.close(ax.get_figure())
+
+        assert rs1 == xpl1
+        assert rs2 == xpl2
+
+    def test_finder_quarterly(self):
+        yrs = [3.5, 11]
+
+        xpl1 = xpl2 = [Period("1988Q1").ordinal] * len(yrs)
+        rs1 = []
+        rs2 = []
+        for n in yrs:
+            rng = period_range("1987Q2", periods=int(n * 4), freq="Q")
+            ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+            _, ax = mpl.pyplot.subplots()
+            ser.plot(ax=ax)
+            xaxis = ax.get_xaxis()
+            rs1.append(xaxis.get_majorticklocs()[0])
+
+            (vmin, vmax) = ax.get_xlim()
+            ax.set_xlim(vmin + 0.9, vmax)
+            rs2.append(xaxis.get_majorticklocs()[0])
+            mpl.pyplot.close(ax.get_figure())
+
+        assert rs1 == xpl1
+        assert rs2 == xpl2
+
+    def test_finder_monthly(self):
+        yrs = [1.15, 2.5, 4, 11]
+
+        xpl1 = xpl2 = [Period("Jan 1988").ordinal] * len(yrs)
+        rs1 = []
+        rs2 = []
+        for n in yrs:
+            rng = period_range("1987Q2", periods=int(n * 12), freq="M")
+            ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+            _, ax = mpl.pyplot.subplots()
+            ser.plot(ax=ax)
+            xaxis = ax.get_xaxis()
+            rs1.append(xaxis.get_majorticklocs()[0])
+
+            vmin, vmax = ax.get_xlim()
+            ax.set_xlim(vmin + 0.9, vmax)
+            rs2.append(xaxis.get_majorticklocs()[0])
+            mpl.pyplot.close(ax.get_figure())
+
+        assert rs1 == xpl1
+        assert rs2 == xpl2
+
+    def test_finder_monthly_long(self):
+        rng = period_range("1988Q1", periods=24 * 12, freq="M")
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ser.plot(ax=ax)
+        xaxis = ax.get_xaxis()
+        rs = xaxis.get_majorticklocs()[0]
+        xp = Period("1989Q1", "M").ordinal
+        assert rs == xp
+
+    def test_finder_annual(self):
+        xp = [1987, 1988, 1990, 1990, 1995, 2020, 2070, 2170]
+        xp = [Period(x, freq="Y").ordinal for x in xp]
+        rs = []
+        for nyears in [5, 10, 19, 49, 99, 199, 599, 1001]:
+            rng = period_range("1987", periods=nyears, freq="Y")
+            ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+            _, ax = mpl.pyplot.subplots()
+            ser.plot(ax=ax)
+            xaxis = ax.get_xaxis()
+            rs.append(xaxis.get_majorticklocs()[0])
+            mpl.pyplot.close(ax.get_figure())
+
+        assert rs == xp
+
+    @pytest.mark.slow
+    def test_finder_minutely(self):
+        nminutes = 1 * 24 * 60
+        rng = date_range("1/1/1999", freq="Min", periods=nminutes)
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ser.plot(ax=ax)
+        xaxis = ax.get_xaxis()
+        rs = xaxis.get_majorticklocs()[0]
+        xp = Period("1/1/1999", freq="Min").ordinal
+
+        assert rs == xp
+
+    def test_finder_hourly(self):
+        nhours = 23
+        rng = date_range("1/1/1999", freq="h", periods=nhours)
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ser.plot(ax=ax)
+        xaxis = ax.get_xaxis()
+        rs = xaxis.get_majorticklocs()[0]
+        xp = Period("1/1/1999", freq="h").ordinal
+
+        assert rs == xp
+
+    def test_gaps(self):
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        ts.iloc[5:7] = np.nan
+        _, ax = mpl.pyplot.subplots()
+        ts.plot(ax=ax)
+        lines = ax.get_lines()
+        assert len(lines) == 1
+        line = lines[0]
+        data = line.get_xydata()
+
+        data = np.ma.MaskedArray(data, mask=isna(data), fill_value=np.nan)
+
+        assert isinstance(data, np.ma.core.MaskedArray)
+        mask = data.mask
+        assert mask[5:7, 1].all()
+
+    def test_gaps_irregular(self):
+        # irregular
+        ts = Series(
+            np.arange(30, dtype=np.float64), index=date_range("2020-01-01", periods=30)
+        )
+        ts = ts.iloc[[0, 1, 2, 5, 7, 9, 12, 15, 20]]
+        ts.iloc[2:5] = np.nan
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot(ax=ax)
+        lines = ax.get_lines()
+        assert len(lines) == 1
+        line = lines[0]
+        data = line.get_xydata()
+
+        data = np.ma.MaskedArray(data, mask=isna(data), fill_value=np.nan)
+
+        assert isinstance(data, np.ma.core.MaskedArray)
+        mask = data.mask
+        assert mask[2:5, 1].all()
+
+    def test_gaps_non_ts(self):
+        # non-ts
+        idx = [0, 1, 2, 5, 7, 9, 12, 15, 20]
+        ser = Series(np.random.default_rng(2).standard_normal(len(idx)), idx)
+        ser.iloc[2:5] = np.nan
+        _, ax = mpl.pyplot.subplots()
+        ser.plot(ax=ax)
+        lines = ax.get_lines()
+        assert len(lines) == 1
+        line = lines[0]
+        data = line.get_xydata()
+        data = np.ma.MaskedArray(data, mask=isna(data), fill_value=np.nan)
+
+        assert isinstance(data, np.ma.core.MaskedArray)
+        mask = data.mask
+        assert mask[2:5, 1].all()
+
+    def test_gap_upsample(self):
+        low = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        low.iloc[5:7] = np.nan
+        _, ax = mpl.pyplot.subplots()
+        low.plot(ax=ax)
+
+        idxh = date_range(low.index[0], low.index[-1], freq="12h")
+        s = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        s.plot(secondary_y=True)
+        lines = ax.get_lines()
+        assert len(lines) == 1
+        assert len(ax.right_ax.get_lines()) == 1
+
+        line = lines[0]
+        data = line.get_xydata()
+        data = np.ma.MaskedArray(data, mask=isna(data), fill_value=np.nan)
+
+        assert isinstance(data, np.ma.core.MaskedArray)
+        mask = data.mask
+        assert mask[5:7, 1].all()
+
+    def test_secondary_y(self):
+        ser = Series(np.random.default_rng(2).standard_normal(10))
+        fig, _ = mpl.pyplot.subplots()
+        ax = ser.plot(secondary_y=True)
+        assert hasattr(ax, "left_ax")
+        assert not hasattr(ax, "right_ax")
+        axes = fig.get_axes()
+        line = ax.get_lines()[0]
+        xp = Series(line.get_ydata(), line.get_xdata())
+        tm.assert_series_equal(ser, xp)
+        assert ax.get_yaxis().get_ticks_position() == "right"
+        assert not axes[0].get_yaxis().get_visible()
+
+    def test_secondary_y_yaxis(self):
+        Series(np.random.default_rng(2).standard_normal(10))
+        ser2 = Series(np.random.default_rng(2).standard_normal(10))
+        _, ax2 = mpl.pyplot.subplots()
+        ser2.plot(ax=ax2)
+        assert ax2.get_yaxis().get_ticks_position() == "left"
+
+    def test_secondary_both(self):
+        ser = Series(np.random.default_rng(2).standard_normal(10))
+        ser2 = Series(np.random.default_rng(2).standard_normal(10))
+        ax = ser2.plot()
+        ax2 = ser.plot(secondary_y=True)
+        assert ax.get_yaxis().get_visible()
+        assert not hasattr(ax, "left_ax")
+        assert hasattr(ax, "right_ax")
+        assert hasattr(ax2, "left_ax")
+        assert not hasattr(ax2, "right_ax")
+
+    def test_secondary_y_ts(self):
+        idx = date_range("1/1/2000", periods=10)
+        ser = Series(np.random.default_rng(2).standard_normal(10), idx)
+        fig, _ = mpl.pyplot.subplots()
+        ax = ser.plot(secondary_y=True)
+        assert hasattr(ax, "left_ax")
+        assert not hasattr(ax, "right_ax")
+        axes = fig.get_axes()
+        line = ax.get_lines()[0]
+        xp = Series(line.get_ydata(), line.get_xdata()).to_timestamp()
+        tm.assert_series_equal(ser, xp)
+        assert ax.get_yaxis().get_ticks_position() == "right"
+        assert not axes[0].get_yaxis().get_visible()
+
+    def test_secondary_y_ts_yaxis(self):
+        idx = date_range("1/1/2000", periods=10)
+        ser2 = Series(np.random.default_rng(2).standard_normal(10), idx)
+        _, ax2 = mpl.pyplot.subplots()
+        ser2.plot(ax=ax2)
+        assert ax2.get_yaxis().get_ticks_position() == "left"
+
+    def test_secondary_y_ts_visible(self):
+        idx = date_range("1/1/2000", periods=10)
+        ser2 = Series(np.random.default_rng(2).standard_normal(10), idx)
+        ax = ser2.plot()
+        assert ax.get_yaxis().get_visible()
+
+    def test_secondary_kde(self):
+        pytest.importorskip("scipy")
+        ser = Series(np.random.default_rng(2).standard_normal(10))
+        fig, ax = mpl.pyplot.subplots()
+        ax = ser.plot(secondary_y=True, kind="density", ax=ax)
+        assert hasattr(ax, "left_ax")
+        assert not hasattr(ax, "right_ax")
+        axes = fig.get_axes()
+        assert axes[1].get_yaxis().get_ticks_position() == "right"
+
+    def test_secondary_bar(self):
+        ser = Series(np.random.default_rng(2).standard_normal(10))
+        fig, ax = mpl.pyplot.subplots()
+        ser.plot(secondary_y=True, kind="bar", ax=ax)
+        axes = fig.get_axes()
+        assert axes[1].get_yaxis().get_ticks_position() == "right"
+
+    def test_secondary_frame(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=["a", "b", "c"]
+        )
+        axes = df.plot(secondary_y=["a", "c"], subplots=True)
+        assert axes[0].get_yaxis().get_ticks_position() == "right"
+        assert axes[1].get_yaxis().get_ticks_position() == "left"
+        assert axes[2].get_yaxis().get_ticks_position() == "right"
+
+    def test_secondary_bar_frame(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 3)), columns=["a", "b", "c"]
+        )
+        axes = df.plot(kind="bar", secondary_y=["a", "c"], subplots=True)
+        assert axes[0].get_yaxis().get_ticks_position() == "right"
+        assert axes[1].get_yaxis().get_ticks_position() == "left"
+        assert axes[2].get_yaxis().get_ticks_position() == "right"
+
+    def test_mixed_freq_regular_first(self):
+        # TODO
+        s1 = Series(
+            np.arange(20, dtype=np.float64),
+            index=date_range("2020-01-01", periods=20, freq="B"),
+        )
+        s2 = s1.iloc[[0, 5, 10, 11, 12, 13, 14, 15]]
+
+        # it works!
+        _, ax = mpl.pyplot.subplots()
+        s1.plot(ax=ax)
+
+        ax2 = s2.plot(style="g", ax=ax)
+        lines = ax2.get_lines()
+        msg = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            idx1 = PeriodIndex(lines[0].get_xdata())
+            idx2 = PeriodIndex(lines[1].get_xdata())
+
+            tm.assert_index_equal(idx1, s1.index.to_period("B"))
+            tm.assert_index_equal(idx2, s2.index.to_period("B"))
+
+            left, right = ax2.get_xlim()
+            pidx = s1.index.to_period()
+        assert left <= pidx[0].ordinal
+        assert right >= pidx[-1].ordinal
+
+    def test_mixed_freq_irregular_first(self):
+        s1 = Series(
+            np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+        )
+        s2 = s1.iloc[[0, 5, 10, 11, 12, 13, 14, 15]]
+        _, ax = mpl.pyplot.subplots()
+        s2.plot(style="g", ax=ax)
+        s1.plot(ax=ax)
+        assert not hasattr(ax, "freq")
+        lines = ax.get_lines()
+        x1 = lines[0].get_xdata()
+        tm.assert_numpy_array_equal(x1, s2.index.astype(object).values)
+        x2 = lines[1].get_xdata()
+        tm.assert_numpy_array_equal(x2, s1.index.astype(object).values)
+
+    def test_mixed_freq_regular_first_df(self):
+        # GH 9852
+        s1 = Series(
+            np.arange(20, dtype=np.float64),
+            index=date_range("2020-01-01", periods=20, freq="B"),
+        ).to_frame()
+        s2 = s1.iloc[[0, 5, 10, 11, 12, 13, 14, 15], :]
+        _, ax = mpl.pyplot.subplots()
+        s1.plot(ax=ax)
+        ax2 = s2.plot(style="g", ax=ax)
+        lines = ax2.get_lines()
+        msg = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            idx1 = PeriodIndex(lines[0].get_xdata())
+            idx2 = PeriodIndex(lines[1].get_xdata())
+            assert idx1.equals(s1.index.to_period("B"))
+            assert idx2.equals(s2.index.to_period("B"))
+            left, right = ax2.get_xlim()
+            pidx = s1.index.to_period()
+        assert left <= pidx[0].ordinal
+        assert right >= pidx[-1].ordinal
+
+    def test_mixed_freq_irregular_first_df(self):
+        # GH 9852
+        s1 = Series(
+            np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+        ).to_frame()
+        s2 = s1.iloc[[0, 5, 10, 11, 12, 13, 14, 15], :]
+        _, ax = mpl.pyplot.subplots()
+        s2.plot(style="g", ax=ax)
+        s1.plot(ax=ax)
+        assert not hasattr(ax, "freq")
+        lines = ax.get_lines()
+        x1 = lines[0].get_xdata()
+        tm.assert_numpy_array_equal(x1, s2.index.astype(object).values)
+        x2 = lines[1].get_xdata()
+        tm.assert_numpy_array_equal(x2, s1.index.astype(object).values)
+
+    def test_mixed_freq_hf_first(self):
+        idxh = date_range("1/1/1999", periods=365, freq="D")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        high.plot(ax=ax)
+        low.plot(ax=ax)
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == "D"
+
+    def test_mixed_freq_alignment(self):
+        ts_ind = date_range("2012-01-01 13:00", "2012-01-02", freq="h")
+        ts_data = np.random.default_rng(2).standard_normal(12)
+
+        ts = Series(ts_data, index=ts_ind)
+        ts2 = ts.asfreq("min").interpolate()
+
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot(ax=ax)
+        ts2.plot(style="r", ax=ax)
+
+        assert ax.lines[0].get_xdata()[0] == ax.lines[1].get_xdata()[0]
+
+    def test_mixed_freq_lf_first(self):
+        idxh = date_range("1/1/1999", periods=365, freq="D")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        low.plot(legend=True, ax=ax)
+        high.plot(legend=True, ax=ax)
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == "D"
+        leg = ax.get_legend()
+        assert len(leg.texts) == 2
+        mpl.pyplot.close(ax.get_figure())
+
+    def test_mixed_freq_lf_first_hourly(self):
+        idxh = date_range("1/1/1999", periods=240, freq="min")
+        idxl = date_range("1/1/1999", periods=4, freq="h")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        low.plot(ax=ax)
+        high.plot(ax=ax)
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == "min"
+
+    @pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+    def test_mixed_freq_irreg_period(self):
+        ts = Series(
+            np.arange(30, dtype=np.float64), index=date_range("2020-01-01", periods=30)
+        )
+        irreg = ts.iloc[[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 16, 17, 18, 29]]
+        msg = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            rng = period_range("1/3/2000", periods=30, freq="B")
+        ps = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _, ax = mpl.pyplot.subplots()
+        irreg.plot(ax=ax)
+        ps.plot(ax=ax)
+
+    def test_mixed_freq_shared_ax(self):
+        # GH13341, using sharex=True
+        idx1 = date_range("2015-01-01", periods=3, freq="ME")
+        idx2 = idx1[:1].union(idx1[2:])
+        s1 = Series(range(len(idx1)), idx1)
+        s2 = Series(range(len(idx2)), idx2)
+
+        _, (ax1, ax2) = mpl.pyplot.subplots(nrows=2, sharex=True)
+        s1.plot(ax=ax1)
+        s2.plot(ax=ax2)
+
+        assert ax1.freq == "M"
+        assert ax2.freq == "M"
+        assert ax1.lines[0].get_xydata()[0, 0] == ax2.lines[0].get_xydata()[0, 0]
+
+    def test_mixed_freq_shared_ax_twin_x(self):
+        # GH13341, using sharex=True
+        idx1 = date_range("2015-01-01", periods=3, freq="ME")
+        idx2 = idx1[:1].union(idx1[2:])
+        s1 = Series(range(len(idx1)), idx1)
+        s2 = Series(range(len(idx2)), idx2)
+        # using twinx
+        _, ax1 = mpl.pyplot.subplots()
+        ax2 = ax1.twinx()
+        s1.plot(ax=ax1)
+        s2.plot(ax=ax2)
+
+        assert ax1.lines[0].get_xydata()[0, 0] == ax2.lines[0].get_xydata()[0, 0]
+
+    @pytest.mark.xfail(reason="TODO (GH14330, GH14322)")
+    def test_mixed_freq_shared_ax_twin_x_irregular_first(self):
+        # GH13341, using sharex=True
+        idx1 = date_range("2015-01-01", periods=3, freq="ME")
+        idx2 = idx1[:1].union(idx1[2:])
+        s1 = Series(range(len(idx1)), idx1)
+        s2 = Series(range(len(idx2)), idx2)
+        _, ax1 = mpl.pyplot.subplots()
+        ax2 = ax1.twinx()
+        s2.plot(ax=ax1)
+        s1.plot(ax=ax2)
+        assert ax1.lines[0].get_xydata()[0, 0] == ax2.lines[0].get_xydata()[0, 0]
+
+    def test_nat_handling(self):
+        _, ax = mpl.pyplot.subplots()
+
+        dti = DatetimeIndex(["2015-01-01", NaT, "2015-01-03"])
+        s = Series(range(len(dti)), dti)
+        s.plot(ax=ax)
+        xdata = ax.get_lines()[0].get_xdata()
+        # plot x data is bounded by index values
+        assert s.index.min() <= Series(xdata).min()
+        assert Series(xdata).max() <= s.index.max()
+
+    def test_to_weekly_resampling_disallow_how_kwd(self):
+        idxh = date_range("1/1/1999", periods=52, freq="W")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        high.plot(ax=ax)
+
+        msg = (
+            "'how' is not a valid keyword for plotting functions. If plotting "
+            "multiple objects on shared axes, resample manually first."
+        )
+        with pytest.raises(ValueError, match=msg):
+            low.plot(ax=ax, how="foo")
+
+    def test_to_weekly_resampling(self):
+        idxh = date_range("1/1/1999", periods=52, freq="W")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        high.plot(ax=ax)
+        low.plot(ax=ax)
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == idxh.freq
+
+    def test_from_weekly_resampling(self):
+        idxh = date_range("1/1/1999", periods=52, freq="W")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        low.plot(ax=ax)
+        high.plot(ax=ax)
+
+        expected_h = idxh.to_period().asi8.astype(np.float64)
+        expected_l = np.array(
+            [1514, 1519, 1523, 1527, 1531, 1536, 1540, 1544, 1549, 1553, 1558, 1562],
+            dtype=np.float64,
+        )
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == idxh.freq
+            xdata = line.get_xdata(orig=False)
+            if len(xdata) == 12:  # idxl lines
+                tm.assert_numpy_array_equal(xdata, expected_l)
+            else:
+                tm.assert_numpy_array_equal(xdata, expected_h)
+
+    @pytest.mark.parametrize("kind1, kind2", [("line", "area"), ("area", "line")])
+    def test_from_resampling_area_line_mixed(self, kind1, kind2):
+        idxh = date_range("1/1/1999", periods=52, freq="W")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = DataFrame(
+            np.random.default_rng(2).random((len(idxh), 3)),
+            index=idxh,
+            columns=[0, 1, 2],
+        )
+        low = DataFrame(
+            np.random.default_rng(2).random((len(idxl), 3)),
+            index=idxl,
+            columns=[0, 1, 2],
+        )
+
+        _, ax = mpl.pyplot.subplots()
+        low.plot(kind=kind1, stacked=True, ax=ax)
+        high.plot(kind=kind2, stacked=True, ax=ax)
+
+        # check low dataframe result
+        expected_x = np.array(
+            [
+                1514,
+                1519,
+                1523,
+                1527,
+                1531,
+                1536,
+                1540,
+                1544,
+                1549,
+                1553,
+                1558,
+                1562,
+            ],
+            dtype=np.float64,
+        )
+        expected_y = np.zeros(len(expected_x), dtype=np.float64)
+        for i in range(3):
+            line = ax.lines[i]
+            assert PeriodIndex(line.get_xdata()).freq == idxh.freq
+            tm.assert_numpy_array_equal(line.get_xdata(orig=False), expected_x)
+            # check stacked values are correct
+            expected_y += low[i].values
+            tm.assert_numpy_array_equal(line.get_ydata(orig=False), expected_y)
+
+        # check high dataframe result
+        expected_x = idxh.to_period().asi8.astype(np.float64)
+        expected_y = np.zeros(len(expected_x), dtype=np.float64)
+        for i in range(3):
+            line = ax.lines[3 + i]
+            assert PeriodIndex(data=line.get_xdata()).freq == idxh.freq
+            tm.assert_numpy_array_equal(line.get_xdata(orig=False), expected_x)
+            expected_y += high[i].values
+            tm.assert_numpy_array_equal(line.get_ydata(orig=False), expected_y)
+
+    @pytest.mark.parametrize("kind1, kind2", [("line", "area"), ("area", "line")])
+    def test_from_resampling_area_line_mixed_high_to_low(self, kind1, kind2):
+        idxh = date_range("1/1/1999", periods=52, freq="W")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = DataFrame(
+            np.random.default_rng(2).random((len(idxh), 3)),
+            index=idxh,
+            columns=[0, 1, 2],
+        )
+        low = DataFrame(
+            np.random.default_rng(2).random((len(idxl), 3)),
+            index=idxl,
+            columns=[0, 1, 2],
+        )
+        _, ax = mpl.pyplot.subplots()
+        high.plot(kind=kind1, stacked=True, ax=ax)
+        low.plot(kind=kind2, stacked=True, ax=ax)
+
+        # check high dataframe result
+        expected_x = idxh.to_period().asi8.astype(np.float64)
+        expected_y = np.zeros(len(expected_x), dtype=np.float64)
+        for i in range(3):
+            line = ax.lines[i]
+            assert PeriodIndex(data=line.get_xdata()).freq == idxh.freq
+            tm.assert_numpy_array_equal(line.get_xdata(orig=False), expected_x)
+            expected_y += high[i].values
+            tm.assert_numpy_array_equal(line.get_ydata(orig=False), expected_y)
+
+        # check low dataframe result
+        expected_x = np.array(
+            [
+                1514,
+                1519,
+                1523,
+                1527,
+                1531,
+                1536,
+                1540,
+                1544,
+                1549,
+                1553,
+                1558,
+                1562,
+            ],
+            dtype=np.float64,
+        )
+        expected_y = np.zeros(len(expected_x), dtype=np.float64)
+        for i in range(3):
+            lines = ax.lines[3 + i]
+            assert PeriodIndex(data=lines.get_xdata()).freq == idxh.freq
+            tm.assert_numpy_array_equal(lines.get_xdata(orig=False), expected_x)
+            expected_y += low[i].values
+            tm.assert_numpy_array_equal(lines.get_ydata(orig=False), expected_y)
+
+    def test_mixed_freq_second_millisecond(self):
+        # GH 7772, GH 7760
+        idxh = date_range("2014-07-01 09:00", freq="s", periods=5)
+        idxl = date_range("2014-07-01 09:00", freq="100ms", periods=50)
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        # high to low
+        _, ax = mpl.pyplot.subplots()
+        high.plot(ax=ax)
+        low.plot(ax=ax)
+        assert len(ax.get_lines()) == 2
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == "ms"
+
+    def test_mixed_freq_second_millisecond_low_to_high(self):
+        # GH 7772, GH 7760
+        idxh = date_range("2014-07-01 09:00", freq="s", periods=5)
+        idxl = date_range("2014-07-01 09:00", freq="100ms", periods=50)
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        # low to high
+        _, ax = mpl.pyplot.subplots()
+        low.plot(ax=ax)
+        high.plot(ax=ax)
+        assert len(ax.get_lines()) == 2
+        for line in ax.get_lines():
+            assert PeriodIndex(data=line.get_xdata()).freq == "ms"
+
+    def test_irreg_dtypes(self):
+        # date
+        idx = [date(2000, 1, 1), date(2000, 1, 5), date(2000, 1, 20)]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((len(idx), 3)),
+            Index(idx, dtype=object),
+        )
+        _check_plot_works(df.plot)
+
+    def test_irreg_dtypes_dt64(self):
+        # np.datetime64
+        idx = date_range("1/1/2000", periods=10)
+        idx = idx[[0, 2, 5, 9]].astype(object)
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(idx), 3)), idx)
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(df.plot, ax=ax)
+
+    def test_time(self):
+        t = datetime(1, 1, 1, 3, 30, 0)
+        deltas = np.random.default_rng(2).integers(1, 20, 3).cumsum()
+        ts = np.array([(t + timedelta(minutes=int(x))).time() for x in deltas])
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(len(ts)),
+                "b": np.random.default_rng(2).standard_normal(len(ts)),
+            },
+            index=ts,
+        )
+        _, ax = mpl.pyplot.subplots()
+        df.plot(ax=ax)
+
+        # verify tick labels
+        ticks = ax.get_xticks()
+        labels = ax.get_xticklabels()
+        for _tick, _label in zip(ticks, labels):
+            m, s = divmod(int(_tick), 60)
+            h, m = divmod(m, 60)
+            rs = _label.get_text()
+            if len(rs) > 0:
+                if s != 0:
+                    xp = time(h, m, s).strftime("%H:%M:%S")
+                else:
+                    xp = time(h, m, s).strftime("%H:%M")
+                assert xp == rs
+
+    def test_time_change_xlim(self):
+        t = datetime(1, 1, 1, 3, 30, 0)
+        deltas = np.random.default_rng(2).integers(1, 20, 3).cumsum()
+        ts = np.array([(t + timedelta(minutes=int(x))).time() for x in deltas])
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(len(ts)),
+                "b": np.random.default_rng(2).standard_normal(len(ts)),
+            },
+            index=ts,
+        )
+        _, ax = mpl.pyplot.subplots()
+        df.plot(ax=ax)
+
+        # verify tick labels
+        ticks = ax.get_xticks()
+        labels = ax.get_xticklabels()
+        for _tick, _label in zip(ticks, labels):
+            m, s = divmod(int(_tick), 60)
+            h, m = divmod(m, 60)
+            rs = _label.get_text()
+            if len(rs) > 0:
+                if s != 0:
+                    xp = time(h, m, s).strftime("%H:%M:%S")
+                else:
+                    xp = time(h, m, s).strftime("%H:%M")
+                assert xp == rs
+
+        # change xlim
+        ax.set_xlim("1:30", "5:00")
+
+        # check tick labels again
+        ticks = ax.get_xticks()
+        labels = ax.get_xticklabels()
+        for _tick, _label in zip(ticks, labels):
+            m, s = divmod(int(_tick), 60)
+            h, m = divmod(m, 60)
+            rs = _label.get_text()
+            if len(rs) > 0:
+                if s != 0:
+                    xp = time(h, m, s).strftime("%H:%M:%S")
+                else:
+                    xp = time(h, m, s).strftime("%H:%M")
+                assert xp == rs
+
+    def test_time_musec(self):
+        t = datetime(1, 1, 1, 3, 30, 0)
+        deltas = np.random.default_rng(2).integers(1, 20, 3).cumsum()
+        ts = np.array([(t + timedelta(microseconds=int(x))).time() for x in deltas])
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(len(ts)),
+                "b": np.random.default_rng(2).standard_normal(len(ts)),
+            },
+            index=ts,
+        )
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(ax=ax)
+
+        # verify tick labels
+        ticks = ax.get_xticks()
+        labels = ax.get_xticklabels()
+        for _tick, _label in zip(ticks, labels):
+            m, s = divmod(int(_tick), 60)
+
+            us = round((_tick - int(_tick)) * 1e6)
+
+            h, m = divmod(m, 60)
+            rs = _label.get_text()
+            if len(rs) > 0:
+                if (us % 1000) != 0:
+                    xp = time(h, m, s, us).strftime("%H:%M:%S.%f")
+                elif (us // 1000) != 0:
+                    xp = time(h, m, s, us).strftime("%H:%M:%S.%f")[:-3]
+                elif s != 0:
+                    xp = time(h, m, s, us).strftime("%H:%M:%S")
+                else:
+                    xp = time(h, m, s, us).strftime("%H:%M")
+                assert xp == rs
+
+    def test_secondary_upsample(self):
+        idxh = date_range("1/1/1999", periods=365, freq="D")
+        idxl = date_range("1/1/1999", periods=12, freq="ME")
+        high = Series(np.random.default_rng(2).standard_normal(len(idxh)), idxh)
+        low = Series(np.random.default_rng(2).standard_normal(len(idxl)), idxl)
+        _, ax = mpl.pyplot.subplots()
+        low.plot(ax=ax)
+        ax = high.plot(secondary_y=True, ax=ax)
+        for line in ax.get_lines():
+            assert PeriodIndex(line.get_xdata()).freq == "D"
+        assert hasattr(ax, "left_ax")
+        assert not hasattr(ax, "right_ax")
+        for line in ax.left_ax.get_lines():
+            assert PeriodIndex(line.get_xdata()).freq == "D"
+
+    def test_secondary_legend(self):
+        fig = mpl.pyplot.figure()
+        ax = fig.add_subplot(211)
+
+        # ts
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df.plot(secondary_y=["A", "B"], ax=ax)
+        leg = ax.get_legend()
+        assert len(leg.get_lines()) == 4
+        assert leg.get_texts()[0].get_text() == "A (right)"
+        assert leg.get_texts()[1].get_text() == "B (right)"
+        assert leg.get_texts()[2].get_text() == "C"
+        assert leg.get_texts()[3].get_text() == "D"
+        assert ax.right_ax.get_legend() is None
+        colors = set()
+        for line in leg.get_lines():
+            colors.add(line.get_color())
+
+        # TODO: color cycle problems
+        assert len(colors) == 4
+
+    def test_secondary_legend_right(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        fig = mpl.pyplot.figure()
+        ax = fig.add_subplot(211)
+        df.plot(secondary_y=["A", "C"], mark_right=False, ax=ax)
+        leg = ax.get_legend()
+        assert len(leg.get_lines()) == 4
+        assert leg.get_texts()[0].get_text() == "A"
+        assert leg.get_texts()[1].get_text() == "B"
+        assert leg.get_texts()[2].get_text() == "C"
+        assert leg.get_texts()[3].get_text() == "D"
+
+    def test_secondary_legend_bar(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        fig, ax = mpl.pyplot.subplots()
+        df.plot(kind="bar", secondary_y=["A"], ax=ax)
+        leg = ax.get_legend()
+        assert leg.get_texts()[0].get_text() == "A (right)"
+        assert leg.get_texts()[1].get_text() == "B"
+
+    def test_secondary_legend_bar_right(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        fig, ax = mpl.pyplot.subplots()
+        df.plot(kind="bar", secondary_y=["A"], mark_right=False, ax=ax)
+        leg = ax.get_legend()
+        assert leg.get_texts()[0].get_text() == "A"
+        assert leg.get_texts()[1].get_text() == "B"
+
+    def test_secondary_legend_multi_col(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        fig = mpl.pyplot.figure()
+        ax = fig.add_subplot(211)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        ax = df.plot(secondary_y=["C", "D"], ax=ax)
+        leg = ax.get_legend()
+        assert len(leg.get_lines()) == 4
+        assert ax.right_ax.get_legend() is None
+        colors = set()
+        for line in leg.get_lines():
+            colors.add(line.get_color())
+
+        # TODO: color cycle problems
+        assert len(colors) == 4
+
+    def test_secondary_legend_nonts(self):
+        # non-ts
+        df = DataFrame(
+            1.1 * np.arange(40).reshape((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(10)], dtype=object),
+        )
+        fig = mpl.pyplot.figure()
+        ax = fig.add_subplot(211)
+        ax = df.plot(secondary_y=["A", "B"], ax=ax)
+        leg = ax.get_legend()
+        assert len(leg.get_lines()) == 4
+        assert ax.right_ax.get_legend() is None
+        colors = set()
+        for line in leg.get_lines():
+            colors.add(line.get_color())
+
+        # TODO: color cycle problems
+        assert len(colors) == 4
+
+    def test_secondary_legend_nonts_multi_col(self):
+        # non-ts
+        df = DataFrame(
+            1.1 * np.arange(40).reshape((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(10)], dtype=object),
+        )
+        fig = mpl.pyplot.figure()
+        ax = fig.add_subplot(211)
+        ax = df.plot(secondary_y=["C", "D"], ax=ax)
+        leg = ax.get_legend()
+        assert len(leg.get_lines()) == 4
+        assert ax.right_ax.get_legend() is None
+        colors = set()
+        for line in leg.get_lines():
+            colors.add(line.get_color())
+
+        # TODO: color cycle problems
+        assert len(colors) == 4
+
+    @pytest.mark.xfail(reason="Api changed in 3.6.0")
+    def test_format_date_axis(self):
+        rng = date_range("1/1/2012", periods=12, freq="ME")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(rng), 3)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(ax=ax)
+        xaxis = ax.get_xaxis()
+        for line in xaxis.get_ticklabels():
+            if len(line.get_text()) > 0:
+                assert line.get_rotation() == 30
+
+    def test_ax_plot(self):
+        x = date_range(start="2012-01-02", periods=10, freq="D")
+        y = list(range(len(x)))
+        _, ax = mpl.pyplot.subplots()
+        lines = ax.plot(x, y, label="Y")
+        tm.assert_index_equal(DatetimeIndex(lines[0].get_xdata()), x)
+
+    def test_mpl_nopandas(self):
+        dates = [date(2008, 12, 31), date(2009, 1, 31)]
+        values1 = np.arange(10.0, 11.0, 0.5)
+        values2 = np.arange(11.0, 12.0, 0.5)
+
+        _, ax = mpl.pyplot.subplots()
+        (
+            line1,
+            line2,
+        ) = ax.plot(
+            [x.toordinal() for x in dates],
+            values1,
+            "-",
+            [x.toordinal() for x in dates],
+            values2,
+            "-",
+            linewidth=4,
+        )
+
+        exp = np.array([x.toordinal() for x in dates], dtype=np.float64)
+        tm.assert_numpy_array_equal(line1.get_xydata()[:, 0], exp)
+        tm.assert_numpy_array_equal(line2.get_xydata()[:, 0], exp)
+
+    def test_irregular_ts_shared_ax_xlim(self):
+        # GH 2960
+        ts = Series(
+            np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+        )
+        ts_irregular = ts.iloc[[1, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 17, 18]]
+
+        # plot the left section of the irregular series, then the right section
+        _, ax = mpl.pyplot.subplots()
+        ts_irregular[:5].plot(ax=ax)
+        ts_irregular[5:].plot(ax=ax)
+
+        # check that axis limits are correct
+        left, right = ax.get_xlim()
+        assert left <= conv.DatetimeConverter.convert(ts_irregular.index.min(), "", ax)
+        assert right >= conv.DatetimeConverter.convert(ts_irregular.index.max(), "", ax)
+
+    def test_secondary_y_non_ts_xlim(self):
+        # GH 3490 - non-timeseries with secondary y
+        index_1 = [1, 2, 3, 4]
+        index_2 = [5, 6, 7, 8]
+        s1 = Series(1, index=index_1)
+        s2 = Series(2, index=index_2)
+
+        _, ax = mpl.pyplot.subplots()
+        s1.plot(ax=ax)
+        left_before, right_before = ax.get_xlim()
+        s2.plot(secondary_y=True, ax=ax)
+        left_after, right_after = ax.get_xlim()
+
+        assert left_before >= left_after
+        assert right_before < right_after
+
+    def test_secondary_y_regular_ts_xlim(self):
+        # GH 3490 - regular-timeseries with secondary y
+        index_1 = date_range(start="2000-01-01", periods=4, freq="D")
+        index_2 = date_range(start="2000-01-05", periods=4, freq="D")
+        s1 = Series(1, index=index_1)
+        s2 = Series(2, index=index_2)
+
+        _, ax = mpl.pyplot.subplots()
+        s1.plot(ax=ax)
+        left_before, right_before = ax.get_xlim()
+        s2.plot(secondary_y=True, ax=ax)
+        left_after, right_after = ax.get_xlim()
+
+        assert left_before >= left_after
+        assert right_before < right_after
+
+    def test_secondary_y_mixed_freq_ts_xlim(self):
+        # GH 3490 - mixed frequency timeseries with secondary y
+        rng = date_range("2000-01-01", periods=10, freq="min")
+        ts = Series(1, index=rng)
+
+        _, ax = mpl.pyplot.subplots()
+        ts.plot(ax=ax)
+        left_before, right_before = ax.get_xlim()
+        ts.resample("D").mean().plot(secondary_y=True, ax=ax)
+        left_after, right_after = ax.get_xlim()
+
+        # a downsample should not have changed either limit
+        assert left_before == left_after
+        assert right_before == right_after
+
+    def test_secondary_y_irregular_ts_xlim(self):
+        # GH 3490 - irregular-timeseries with secondary y
+        ts = Series(
+            np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+        )
+        ts_irregular = ts.iloc[[1, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 17, 18]]
+
+        _, ax = mpl.pyplot.subplots()
+        ts_irregular[:5].plot(ax=ax)
+        # plot higher-x values on secondary axis
+        ts_irregular[5:].plot(secondary_y=True, ax=ax)
+        # ensure secondary limits aren't overwritten by plot on primary
+        ts_irregular[:5].plot(ax=ax)
+
+        left, right = ax.get_xlim()
+        assert left <= conv.DatetimeConverter.convert(ts_irregular.index.min(), "", ax)
+        assert right >= conv.DatetimeConverter.convert(ts_irregular.index.max(), "", ax)
+
+    def test_plot_outofbounds_datetime(self):
+        # 2579 - checking this does not raise
+        values = [date(1677, 1, 1), date(1677, 1, 2)]
+        _, ax = mpl.pyplot.subplots()
+        ax.plot(values)
+
+        values = [datetime(1677, 1, 1, 12), datetime(1677, 1, 2, 12)]
+        ax.plot(values)
+
+    def test_format_timedelta_ticks_narrow(self):
+        expected_labels = [f"00:00:00.0000000{i:0>2d}" for i in np.arange(10)]
+
+        rng = timedelta_range("0", periods=10, freq="ns")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(rng), 3)), rng)
+        _, ax = mpl.pyplot.subplots()
+        df.plot(fontsize=2, ax=ax)
+        mpl.pyplot.draw()
+        labels = ax.get_xticklabels()
+
+        result_labels = [x.get_text() for x in labels]
+        assert len(result_labels) == len(expected_labels)
+        assert result_labels == expected_labels
+
+    def test_format_timedelta_ticks_wide(self):
+        expected_labels = [
+            "00:00:00",
+            "1 days 03:46:40",
+            "2 days 07:33:20",
+            "3 days 11:20:00",
+            "4 days 15:06:40",
+            "5 days 18:53:20",
+            "6 days 22:40:00",
+            "8 days 02:26:40",
+            "9 days 06:13:20",
+        ]
+
+        rng = timedelta_range("0", periods=10, freq="1 D")
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(rng), 3)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(fontsize=2, ax=ax)
+        mpl.pyplot.draw()
+        labels = ax.get_xticklabels()
+
+        result_labels = [x.get_text() for x in labels]
+        assert len(result_labels) == len(expected_labels)
+        assert result_labels == expected_labels
+
+    def test_timedelta_plot(self):
+        # test issue #8711
+        s = Series(range(5), timedelta_range("1day", periods=5))
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(s.plot, ax=ax)
+
+    def test_timedelta_long_period(self):
+        # test long period
+        index = timedelta_range("1 day 2 hr 30 min 10 s", periods=10, freq="1 D")
+        s = Series(np.random.default_rng(2).standard_normal(len(index)), index)
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(s.plot, ax=ax)
+
+    def test_timedelta_short_period(self):
+        # test short period
+        index = timedelta_range("1 day 2 hr 30 min 10 s", periods=10, freq="1 ns")
+        s = Series(np.random.default_rng(2).standard_normal(len(index)), index)
+        _, ax = mpl.pyplot.subplots()
+        _check_plot_works(s.plot, ax=ax)
+
+    def test_hist(self):
+        # https://github.com/matplotlib/matplotlib/issues/8459
+        rng = date_range("1/1/2011", periods=10, freq="h")
+        x = rng
+        w1 = np.arange(0, 1, 0.1)
+        w2 = np.arange(0, 1, 0.1)[::-1]
+        _, ax = mpl.pyplot.subplots()
+        ax.hist([x, x], weights=[w1, w2])
+
+    def test_overlapping_datetime(self):
+        # GB 6608
+        s1 = Series(
+            [1, 2, 3],
+            index=[
+                datetime(1995, 12, 31),
+                datetime(2000, 12, 31),
+                datetime(2005, 12, 31),
+            ],
+        )
+        s2 = Series(
+            [1, 2, 3],
+            index=[
+                datetime(1997, 12, 31),
+                datetime(2003, 12, 31),
+                datetime(2008, 12, 31),
+            ],
+        )
+
+        # plot first series, then add the second series to those axes,
+        # then try adding the first series again
+        _, ax = mpl.pyplot.subplots()
+        s1.plot(ax=ax)
+        s2.plot(ax=ax)
+        s1.plot(ax=ax)
+
+    @pytest.mark.xfail(reason="GH9053 matplotlib does not use ax.xaxis.converter")
+    def test_add_matplotlib_datetime64(self):
+        # GH9053 - ensure that a plot with PeriodConverter still understands
+        # datetime64 data. This still fails because matplotlib overrides the
+        # ax.xaxis.converter with a DatetimeConverter
+        s = Series(
+            np.random.default_rng(2).standard_normal(10),
+            index=date_range("1970-01-02", periods=10),
+        )
+        ax = s.plot()
+        with tm.assert_produces_warning(DeprecationWarning):
+            # multi-dimensional indexing
+            ax.plot(s.index, s.values, color="g")
+        l1, l2 = ax.lines
+        tm.assert_numpy_array_equal(l1.get_xydata(), l2.get_xydata())
+
+    def test_matplotlib_scatter_datetime64(self):
+        # https://github.com/matplotlib/matplotlib/issues/11391
+        df = DataFrame(np.random.default_rng(2).random((10, 2)), columns=["x", "y"])
+        df["time"] = date_range("2018-01-01", periods=10, freq="D")
+        _, ax = mpl.pyplot.subplots()
+        ax.scatter(x="time", y="y", data=df)
+        mpl.pyplot.draw()
+        label = ax.get_xticklabels()[0]
+        expected = "2018-01-01"
+        assert label.get_text() == expected
+
+    def test_check_xticks_rot(self):
+        # https://github.com/pandas-dev/pandas/issues/29460
+        # regular time series
+        x = to_datetime(["2020-05-01", "2020-05-02", "2020-05-03"])
+        df = DataFrame({"x": x, "y": [1, 2, 3]})
+        axes = df.plot(x="x", y="y")
+        _check_ticks_props(axes, xrot=0)
+
+    def test_check_xticks_rot_irregular(self):
+        # irregular time series
+        x = to_datetime(["2020-05-01", "2020-05-02", "2020-05-04"])
+        df = DataFrame({"x": x, "y": [1, 2, 3]})
+        axes = df.plot(x="x", y="y")
+        _check_ticks_props(axes, xrot=30)
+
+    def test_check_xticks_rot_use_idx(self):
+        # irregular time series
+        x = to_datetime(["2020-05-01", "2020-05-02", "2020-05-04"])
+        df = DataFrame({"x": x, "y": [1, 2, 3]})
+        # use timeseries index or not
+        axes = df.set_index("x").plot(y="y", use_index=True)
+        _check_ticks_props(axes, xrot=30)
+        axes = df.set_index("x").plot(y="y", use_index=False)
+        _check_ticks_props(axes, xrot=0)
+
+    def test_check_xticks_rot_sharex(self):
+        # irregular time series
+        x = to_datetime(["2020-05-01", "2020-05-02", "2020-05-04"])
+        df = DataFrame({"x": x, "y": [1, 2, 3]})
+        # separate subplots
+        axes = df.plot(x="x", y="y", subplots=True, sharex=True)
+        _check_ticks_props(axes, xrot=30)
+        axes = df.plot(x="x", y="y", subplots=True, sharex=False)
+        _check_ticks_props(axes, xrot=0)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            date_range("2020-01-01", periods=5),
+            date_range("2020-01-01", periods=5, tz="UTC"),
+            timedelta_range("1 day", periods=5, freq="D"),
+            period_range("2020-01-01", periods=5, freq="D"),
+            Index([date(2000, 1, i) for i in [1, 3, 6, 20, 22]], dtype=object),
+            range(5),
+        ],
+    )
+    def test_pickle_fig(self, temp_file, frame_or_series, idx):
+        # GH18439, GH#24088, statsmodels#4772
+        df = frame_or_series(range(5), index=idx)
+        fig, ax = plt.subplots(1, 1)
+        df.plot(ax=ax)
+        with temp_file.open(mode="wb") as path:
+            pickle.dump(fig, path)
+
+
+def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
+    fig = plt.gcf()
+
+    fig.clf()
+    ax = fig.add_subplot(211)
+    orig_ax = kwargs.pop("ax", plt.gca())
+    orig_axfreq = getattr(orig_ax, "freq", None)
+
+    ret = f(*args, **kwargs)
+    assert ret is not None  # do something more intelligent
+
+    ax = kwargs.pop("ax", plt.gca())
+    if series is not None:
+        dfreq = series.index.freq
+        if isinstance(dfreq, BaseOffset):
+            dfreq = dfreq.rule_code
+        if orig_axfreq is None:
+            assert ax.freq == dfreq
+
+    if freq is not None and orig_axfreq is None:
+        assert to_offset(ax.freq, is_period=True) == freq
+
+    ax = fig.add_subplot(212)
+    kwargs["ax"] = ax
+    ret = f(*args, **kwargs)
+    assert ret is not None  # TODO: do something more intelligent
diff --git a/pandas/pandas/tests/plotting/test_groupby.py b/pandas/pandas/tests/plotting/test_groupby.py
new file mode 100644
index 0000000000000..0cb125d822fd1
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_groupby.py
@@ -0,0 +1,154 @@
+"""Test cases for GroupBy.plot"""
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+from pandas.tests.plotting.common import (
+    _check_axes_shape,
+    _check_legend_labels,
+)
+
+pytest.importorskip("matplotlib")
+
+
+class TestDataFrameGroupByPlots:
+    def test_series_groupby_plotting_nominally_works(self):
+        n = 10
+        weight = Series(np.random.default_rng(2).normal(166, 20, size=n))
+        gender = np.random.default_rng(2).choice(["male", "female"], size=n)
+
+        weight.groupby(gender).plot()
+
+    def test_series_groupby_plotting_nominally_works_hist(self):
+        n = 10
+        height = Series(np.random.default_rng(2).normal(60, 10, size=n))
+        gender = np.random.default_rng(2).choice(["male", "female"], size=n)
+        height.groupby(gender).hist()
+
+    def test_series_groupby_plotting_nominally_works_alpha(self):
+        n = 10
+        height = Series(np.random.default_rng(2).normal(60, 10, size=n))
+        gender = np.random.default_rng(2).choice(["male", "female"], size=n)
+        # Regression test for GH8733
+        height.groupby(gender).plot(alpha=0.5)
+
+    def test_plotting_with_float_index_works(self):
+        # GH 7025
+        df = DataFrame(
+            {
+                "def": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+                "val": np.random.default_rng(2).standard_normal(9),
+            },
+            index=[1.0, 2.0, 3.0, 1.0, 2.0, 3.0, 1.0, 2.0, 3.0],
+        )
+
+        df.groupby("def")["val"].plot()
+
+    def test_plotting_with_float_index_works_apply(self):
+        # GH 7025
+        df = DataFrame(
+            {
+                "def": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+                "val": np.random.default_rng(2).standard_normal(9),
+            },
+            index=[1.0, 2.0, 3.0, 1.0, 2.0, 3.0, 1.0, 2.0, 3.0],
+        )
+        df.groupby("def")["val"].apply(lambda x: x.plot())
+
+    def test_hist_single_row(self):
+        # GH10214
+        bins = np.arange(80, 100 + 2, 1)
+        df = DataFrame({"Name": ["AAA", "BBB"], "ByCol": [1, 2], "Mark": [85, 89]})
+        df["Mark"].hist(by=df["ByCol"], bins=bins)
+
+    def test_hist_single_row_single_bycol(self):
+        # GH10214
+        bins = np.arange(80, 100 + 2, 1)
+        df = DataFrame({"Name": ["AAA"], "ByCol": [1], "Mark": [85]})
+        df["Mark"].hist(by=df["ByCol"], bins=bins)
+
+    def test_plot_submethod_works(self):
+        df = DataFrame({"x": [1, 2, 3, 4, 5], "y": [1, 2, 3, 2, 1], "z": list("ababa")})
+        df.groupby("z").plot.scatter("x", "y")
+
+    def test_plot_submethod_works_line(self):
+        df = DataFrame({"x": [1, 2, 3, 4, 5], "y": [1, 2, 3, 2, 1], "z": list("ababa")})
+        df.groupby("z")["x"].plot.line()
+
+    def test_plot_kwargs(self):
+        df = DataFrame({"x": [1, 2, 3, 4, 5], "y": [1, 2, 3, 2, 1], "z": list("ababa")})
+
+        res = df.groupby("z").plot(kind="scatter", x="x", y="y")
+        # check that a scatter plot is effectively plotted: the axes should
+        # contain a PathCollection from the scatter plot (GH11805)
+        assert len(res["a"].collections) == 1
+
+    def test_plot_kwargs_scatter(self):
+        df = DataFrame({"x": [1, 2, 3, 4, 5], "y": [1, 2, 3, 2, 1], "z": list("ababa")})
+        res = df.groupby("z").plot.scatter(x="x", y="y")
+        assert len(res["a"].collections) == 1
+
+    @pytest.mark.parametrize("column, expected_axes_num", [(None, 2), ("b", 1)])
+    def test_groupby_hist_frame_with_legend(self, column, expected_axes_num):
+        # GH 6279 - DataFrameGroupBy histogram can have a legend
+        expected_layout = (1, expected_axes_num)
+        expected_labels = column or [["a"], ["b"]]
+
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+        g = df.groupby("c")
+
+        for axes in g.hist(legend=True, column=column):
+            _check_axes_shape(axes, axes_num=expected_axes_num, layout=expected_layout)
+            for ax, expected_label in zip(axes[0], expected_labels):
+                _check_legend_labels(ax, expected_label)
+
+    @pytest.mark.parametrize("column", [None, "b"])
+    def test_groupby_hist_frame_with_legend_raises(self, column):
+        # GH 6279 - DataFrameGroupBy histogram with legend and label raises
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+        g = df.groupby("c")
+
+        with pytest.raises(ValueError, match="Cannot use both legend and label"):
+            g.hist(legend=True, column=column, label="d")
+
+    def test_groupby_hist_series_with_legend(self):
+        # GH 6279 - SeriesGroupBy histogram can have a legend
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+        g = df.groupby("c")
+
+        for ax in g["a"].hist(legend=True):
+            _check_axes_shape(ax, axes_num=1, layout=(1, 1))
+            _check_legend_labels(ax, ["1", "2"])
+
+    def test_groupby_hist_series_with_legend_raises(self):
+        # GH 6279 - SeriesGroupBy histogram with legend and label raises
+        index = Index(15 * ["1"] + 15 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+        g = df.groupby("c")
+
+        with pytest.raises(ValueError, match="Cannot use both legend and label"):
+            g.hist(legend=True, label="d")
diff --git a/pandas/pandas/tests/plotting/test_hist_method.py b/pandas/pandas/tests/plotting/test_hist_method.py
new file mode 100644
index 0000000000000..410b658065d8d
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_hist_method.py
@@ -0,0 +1,957 @@
+"""Test cases for .hist method"""
+
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_ax_scales,
+    _check_axes_shape,
+    _check_colors,
+    _check_legend_labels,
+    _check_patches_all_filled,
+    _check_plot_works,
+    _check_text_labels,
+    _check_ticks_props,
+    get_x_axis,
+    get_y_axis,
+)
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+from pandas.plotting._matplotlib.hist import _grouped_hist
+
+
+@pytest.fixture
+def ts():
+    return Series(
+        np.arange(30, dtype=np.float64),
+        index=date_range("2020-01-01", periods=30, freq="B"),
+        name="ts",
+    )
+
+
+class TestSeriesPlots:
+    @pytest.mark.parametrize("kwargs", [{}, {"grid": False}, {"figsize": (8, 10)}])
+    def test_hist_legacy_kwargs(self, ts, kwargs):
+        _check_plot_works(ts.hist, **kwargs)
+
+    @pytest.mark.parametrize("kwargs", [{}, {"bins": 5}])
+    def test_hist_legacy_kwargs_warning(self, ts, kwargs):
+        # _check_plot_works adds an ax so catch warning. see GH #13188
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(ts.hist, by=ts.index.month, **kwargs)
+
+    def test_hist_legacy_ax(self, ts):
+        fig, ax = mpl.pyplot.subplots(1, 1)
+        _check_plot_works(ts.hist, ax=ax, default_axes=True)
+
+    def test_hist_legacy_ax_and_fig(self, ts):
+        fig, ax = mpl.pyplot.subplots(1, 1)
+        _check_plot_works(ts.hist, ax=ax, figure=fig, default_axes=True)
+
+    def test_hist_legacy_fig(self, ts):
+        fig, _ = mpl.pyplot.subplots(1, 1)
+        _check_plot_works(ts.hist, figure=fig, default_axes=True)
+
+    def test_hist_legacy_multi_ax(self, ts):
+        fig, (ax1, ax2) = mpl.pyplot.subplots(1, 2)
+        _check_plot_works(ts.hist, figure=fig, ax=ax1, default_axes=True)
+        _check_plot_works(ts.hist, figure=fig, ax=ax2, default_axes=True)
+
+    def test_hist_legacy_by_fig_error(self, ts):
+        fig, _ = mpl.pyplot.subplots(1, 1)
+        msg = (
+            "Cannot pass 'figure' when using the 'by' argument, since a new 'Figure' "
+            "instance will be created"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ts.hist(by=ts.index, figure=fig)
+
+    def test_hist_bins_legacy(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        ax = df.hist(bins=2)[0][0]
+        assert len(ax.patches) == 2
+
+    def test_hist_layout(self, hist_df):
+        df = hist_df
+        msg = "The 'layout' keyword is not supported when 'by' is None"
+        with pytest.raises(ValueError, match=msg):
+            df.height.hist(layout=(1, 1))
+
+        with pytest.raises(ValueError, match=msg):
+            df.height.hist(layout=[1, 1])
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "by, layout, axes_num, res_layout",
+        [
+            ["gender", (2, 1), 2, (2, 1)],
+            ["gender", (3, -1), 2, (3, 1)],
+            ["category", (4, 1), 4, (4, 1)],
+            ["category", (2, -1), 4, (2, 2)],
+            ["category", (3, -1), 4, (3, 2)],
+            ["category", (-1, 4), 4, (1, 4)],
+            ["classroom", (2, 2), 3, (2, 2)],
+        ],
+    )
+    def test_hist_layout_with_by(self, hist_df, by, layout, axes_num, res_layout):
+        df = hist_df
+
+        # _check_plot_works adds an `ax` kwarg to the method call
+        # so we get a warning about an axis being cleared, even
+        # though we don't explicitly pass one, see GH #13188
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(df.height.hist, by=getattr(df, by), layout=layout)
+        _check_axes_shape(axes, axes_num=axes_num, layout=res_layout)
+
+    def test_hist_layout_with_by_shape(self, hist_df):
+        df = hist_df
+
+        axes = df.height.hist(by=df.category, layout=(4, 2), figsize=(12, 7))
+        _check_axes_shape(axes, axes_num=4, layout=(4, 2), figsize=(12, 7))
+
+    def test_hist_no_overlap(self):
+        x = Series(np.random.default_rng(2).standard_normal(2))
+        y = Series(np.random.default_rng(2).standard_normal(2))
+        plt.subplot(121)
+        x.hist()
+        plt.subplot(122)
+        y.hist()
+        fig = plt.gcf()
+        axes = fig.axes
+        assert len(axes) == 2
+
+    def test_hist_by_no_extra_plots(self, hist_df):
+        df = hist_df
+        df.height.hist(by=df.gender)
+        assert len(mpl.pyplot.get_fignums()) == 1
+
+    def test_plot_fails_when_ax_differs_from_figure(self, ts):
+        fig1 = plt.figure(1)
+        fig2 = plt.figure(2)
+        ax1 = fig1.add_subplot(111)
+        msg = "passed axis not bound to passed figure"
+        with pytest.raises(AssertionError, match=msg):
+            ts.hist(ax=ax1, figure=fig2)
+
+    @pytest.mark.parametrize(
+        "histtype, expected",
+        [
+            ("bar", True),
+            ("barstacked", True),
+            ("step", False),
+            ("stepfilled", True),
+        ],
+    )
+    def test_histtype_argument(self, histtype, expected):
+        # GH23992 Verify functioning of histtype argument
+        ser = Series(np.random.default_rng(2).integers(1, 10))
+        ax = ser.hist(histtype=histtype)
+        _check_patches_all_filled(ax, filled=expected)
+
+    @pytest.mark.parametrize(
+        "by, expected_axes_num, expected_layout", [(None, 1, (1, 1)), ("b", 2, (1, 2))]
+    )
+    def test_hist_with_legend(self, by, expected_axes_num, expected_layout):
+        # GH 6279 - Series histogram can have a legend
+        index = 5 * ["1"] + 5 * ["2"]
+        s = Series(np.random.default_rng(2).standard_normal(10), index=index, name="a")
+        s.index.name = "b"
+
+        # Use default_axes=True when plotting method generate subplots itself
+        axes = _check_plot_works(s.hist, default_axes=True, legend=True, by=by)
+        _check_axes_shape(axes, axes_num=expected_axes_num, layout=expected_layout)
+        _check_legend_labels(axes, "a")
+
+    @pytest.mark.parametrize("by", [None, "b"])
+    def test_hist_with_legend_raises(self, by):
+        # GH 6279 - Series histogram with legend and label raises
+        index = 5 * ["1"] + 5 * ["2"]
+        s = Series(np.random.default_rng(2).standard_normal(10), index=index, name="a")
+        s.index.name = "b"
+
+        with pytest.raises(ValueError, match="Cannot use both legend and label"):
+            s.hist(legend=True, by=by, label="c")
+
+    def test_hist_kwargs(self, ts):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.hist(bins=5, ax=ax)
+        assert len(ax.patches) == 5
+        _check_text_labels(ax.yaxis.get_label(), "Frequency")
+
+    def test_hist_kwargs_horizontal(self, ts):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.hist(bins=5, ax=ax)
+        ax = ts.plot.hist(orientation="horizontal", ax=ax)
+        _check_text_labels(ax.xaxis.get_label(), "Frequency")
+
+    def test_hist_kwargs_align(self, ts):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.hist(bins=5, ax=ax)
+        ax = ts.plot.hist(align="left", stacked=True, ax=ax)
+
+    @pytest.mark.xfail(reason="Api changed in 3.6.0")
+    def test_hist_kde(self, ts):
+        pytest.importorskip("scipy")
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.hist(logy=True, ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        xlabels = ax.get_xticklabels()
+        # ticks are values, thus ticklabels are blank
+        _check_text_labels(xlabels, [""] * len(xlabels))
+        ylabels = ax.get_yticklabels()
+        _check_text_labels(ylabels, [""] * len(ylabels))
+
+    def test_hist_kde_plot_works(self, ts):
+        pytest.importorskip("scipy")
+        _check_plot_works(ts.plot.kde)
+
+    def test_hist_kde_density_works(self, ts):
+        pytest.importorskip("scipy")
+        _check_plot_works(ts.plot.density)
+
+    @pytest.mark.xfail(reason="Api changed in 3.6.0")
+    def test_hist_kde_logy(self, ts):
+        pytest.importorskip("scipy")
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.kde(logy=True, ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        xlabels = ax.get_xticklabels()
+        _check_text_labels(xlabels, [""] * len(xlabels))
+        ylabels = ax.get_yticklabels()
+        _check_text_labels(ylabels, [""] * len(ylabels))
+
+    def test_hist_kde_color_bins(self, ts):
+        pytest.importorskip("scipy")
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.hist(logy=True, bins=10, color="b", ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        assert len(ax.patches) == 10
+        _check_colors(ax.patches, facecolors=["b"] * 10)
+
+    def test_hist_kde_color(self, ts):
+        pytest.importorskip("scipy")
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.kde(logy=True, color="r", ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        lines = ax.get_lines()
+        assert len(lines) == 1
+        _check_colors(lines, ["r"])
+
+
+class TestDataFramePlots:
+    @pytest.mark.slow
+    def test_hist_df_legacy(self, hist_df):
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(hist_df.hist)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_layout(self):
+        # make sure layout is handled
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        df[2] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(df.hist, grid=False)
+        _check_axes_shape(axes, axes_num=3, layout=(2, 2))
+        assert not axes[1, 1].get_visible()
+
+        _check_plot_works(df[[2]].hist)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_layout2(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 1)))
+        _check_plot_works(df.hist)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_layout3(self):
+        # make sure layout is handled
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+        df[5] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(df.hist, layout=(4, 2))
+        _check_axes_shape(axes, axes_num=6, layout=(4, 2))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs", [{"sharex": True, "sharey": True}, {"figsize": (8, 10)}, {"bins": 5}]
+    )
+    def test_hist_df_legacy_layout_kwargs(self, kwargs):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 5)))
+        df[5] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        # make sure sharex, sharey is handled
+        # handle figsize arg
+        # check bins argument
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            _check_plot_works(df.hist, **kwargs)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_layout_labelsize_rot(self, frame_or_series):
+        # make sure xlabelsize and xrot are handled
+        obj = frame_or_series(range(10))
+        xf, yf = 20, 18
+        xrot, yrot = 30, 40
+        axes = obj.hist(xlabelsize=xf, xrot=xrot, ylabelsize=yf, yrot=yrot)
+        _check_ticks_props(axes, xlabelsize=xf, xrot=xrot, ylabelsize=yf, yrot=yrot)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_rectangles(self):
+        ser = Series(range(10))
+        ax = ser.hist(cumulative=True, bins=4, density=True)
+        # height of last bin (index 5) must be 1.0
+        rects = [x for x in ax.get_children() if isinstance(x, mpl.patches.Rectangle)]
+        tm.assert_almost_equal(rects[-1].get_height(), 1.0)
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_scale(self):
+        ser = Series(range(10))
+        ax = ser.hist(log=True)
+        # scale of y must be 'log'
+        _check_ax_scales(ax, yaxis="log")
+
+    @pytest.mark.slow
+    def test_hist_df_legacy_external_error(self):
+        ser = Series(range(10))
+        # propagate attr exception from matplotlib.Axes.hist
+        with tm.external_error_raised(AttributeError):
+            ser.hist(foo="bar")
+
+    def test_hist_non_numerical_or_datetime_raises(self):
+        # gh-10444, GH32590
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).random(10),
+                "b": np.random.default_rng(2).integers(0, 10, 10),
+                "c": to_datetime(
+                    np.random.default_rng(2).integers(
+                        1582800000000000000, 1583500000000000000, 10, dtype=np.int64
+                    )
+                ),
+                "d": to_datetime(
+                    np.random.default_rng(2).integers(
+                        1582800000000000000, 1583500000000000000, 10, dtype=np.int64
+                    ),
+                    utc=True,
+                ),
+            }
+        )
+        df_o = df.astype(object)
+
+        msg = "hist method requires numerical or datetime columns, nothing to plot."
+        with pytest.raises(ValueError, match=msg):
+            df_o.hist()
+
+    @pytest.mark.parametrize(
+        "layout_test",
+        (
+            {"layout": None, "expected_size": (2, 2)},  # default is 2x2
+            {"layout": (2, 2), "expected_size": (2, 2)},
+            {"layout": (4, 1), "expected_size": (4, 1)},
+            {"layout": (1, 4), "expected_size": (1, 4)},
+            {"layout": (3, 3), "expected_size": (3, 3)},
+            {"layout": (-1, 4), "expected_size": (1, 4)},
+            {"layout": (4, -1), "expected_size": (4, 1)},
+            {"layout": (-1, 2), "expected_size": (2, 2)},
+            {"layout": (2, -1), "expected_size": (2, 2)},
+        ),
+    )
+    def test_hist_layout(self, layout_test):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        df[2] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        axes = df.hist(layout=layout_test["layout"])
+        expected = layout_test["expected_size"]
+        _check_axes_shape(axes, axes_num=3, layout=expected)
+
+    def test_hist_layout_error(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        df[2] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        # layout too small for all 4 plots
+        msg = "Layout of 1x1 must be larger than required size 3"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(layout=(1, 1))
+
+        # invalid format for layout
+        msg = re.escape("Layout must be a tuple of (rows, columns)")
+        with pytest.raises(ValueError, match=msg):
+            df.hist(layout=(1,))
+        msg = "At least one dimension of layout must be positive"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(layout=(-1, -1))
+
+    # GH 9351
+    def test_tight_layout(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        df[2] = to_datetime(
+            np.random.default_rng(2).integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        # Use default_axes=True when plotting method generate subplots itself
+        _check_plot_works(df.hist, default_axes=True)
+        mpl.pyplot.tight_layout()
+
+    def test_hist_subplot_xrot(self):
+        # GH 30288
+        df = DataFrame(
+            {
+                "length": [1.5, 0.5, 1.2, 0.9, 3],
+                "animal": ["pig", "rabbit", "pig", "pig", "rabbit"],
+            }
+        )
+        # Use default_axes=True when plotting method generate subplots itself
+        axes = _check_plot_works(
+            df.hist,
+            default_axes=True,
+            column="length",
+            by="animal",
+            bins=5,
+            xrot=0,
+        )
+        _check_ticks_props(axes, xrot=0)
+
+    @pytest.mark.parametrize(
+        "column, expected",
+        [
+            (None, ["width", "length", "height"]),
+            (["length", "width", "height"], ["length", "width", "height"]),
+        ],
+    )
+    def test_hist_column_order_unchanged(self, column, expected):
+        # GH29235
+
+        df = DataFrame(
+            {
+                "width": [0.7, 0.2, 0.15, 0.2, 1.1],
+                "length": [1.5, 0.5, 1.2, 0.9, 3],
+                "height": [3, 0.5, 3.4, 2, 1],
+            },
+            index=["pig", "rabbit", "duck", "chicken", "horse"],
+        )
+
+        # Use default_axes=True when plotting method generate subplots itself
+        axes = _check_plot_works(
+            df.hist,
+            default_axes=True,
+            column=column,
+            layout=(1, 3),
+        )
+        result = [axes[0, i].get_title() for i in range(3)]
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "histtype, expected",
+        [
+            ("bar", True),
+            ("barstacked", True),
+            ("step", False),
+            ("stepfilled", True),
+        ],
+    )
+    def test_histtype_argument(self, histtype, expected):
+        # GH23992 Verify functioning of histtype argument
+        df = DataFrame(
+            np.random.default_rng(2).integers(1, 10, size=(10, 2)), columns=["a", "b"]
+        )
+        ax = df.hist(histtype=histtype)
+        _check_patches_all_filled(ax, filled=expected)
+
+    @pytest.mark.parametrize("by", [None, "c"])
+    @pytest.mark.parametrize("column", [None, "b"])
+    def test_hist_with_legend(self, by, column):
+        # GH 6279 - DataFrame histogram can have a legend
+        expected_axes_num = 1 if by is None and column is not None else 2
+        expected_layout = (1, expected_axes_num)
+        expected_labels = column or ["a", "b"]
+        if by is not None:
+            expected_labels = [expected_labels] * 2
+
+        index = Index(5 * ["1"] + 5 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+
+        # Use default_axes=True when plotting method generate subplots itself
+        axes = _check_plot_works(
+            df.hist,
+            default_axes=True,
+            legend=True,
+            by=by,
+            column=column,
+        )
+
+        _check_axes_shape(axes, axes_num=expected_axes_num, layout=expected_layout)
+        if by is None and column is None:
+            axes = axes[0]
+        for expected_label, ax in zip(expected_labels, axes):
+            _check_legend_labels(ax, expected_label)
+
+    @pytest.mark.parametrize("by", [None, "c"])
+    @pytest.mark.parametrize("column", [None, "b"])
+    def test_hist_with_legend_raises(self, by, column):
+        # GH 6279 - DataFrame histogram with legend and label raises
+        index = Index(5 * ["1"] + 5 * ["2"], name="c")
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            index=index,
+            columns=["a", "b"],
+        )
+
+        with pytest.raises(ValueError, match="Cannot use both legend and label"):
+            df.hist(legend=True, by=by, column=column, label="d")
+
+    def test_hist_df_kwargs(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 2)))
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot.hist(bins=5, ax=ax)
+        assert len(ax.patches) == 10
+
+    def test_hist_df_with_nonnumerics(self):
+        # GH 9853
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=["A", "B", "C", "D"],
+        )
+        df["E"] = ["x", "y"] * 5
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot.hist(bins=5, ax=ax)
+        assert len(ax.patches) == 20
+
+    def test_hist_df_with_nonnumerics_no_bins(self):
+        # GH 9853
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=["A", "B", "C", "D"],
+        )
+        df["E"] = ["x", "y"] * 5
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot.hist(ax=ax)  # bins=10
+        assert len(ax.patches) == 40
+
+    def test_hist_secondary_legend(self):
+        # GH 9610
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), columns=list("abcd")
+        )
+
+        # primary -> secondary
+        _, ax = mpl.pyplot.subplots()
+        ax = df["a"].plot.hist(legend=True, ax=ax)
+        df["b"].plot.hist(ax=ax, legend=True, secondary_y=True)
+        # both legends are drawn on left ax
+        # left and right axis must be visible
+        _check_legend_labels(ax, labels=["a", "b (right)"])
+        assert ax.get_yaxis().get_visible()
+        assert ax.right_ax.get_yaxis().get_visible()
+
+    def test_hist_secondary_secondary(self):
+        # GH 9610
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), columns=list("abcd")
+        )
+        # secondary -> secondary
+        _, ax = mpl.pyplot.subplots()
+        ax = df["a"].plot.hist(legend=True, secondary_y=True, ax=ax)
+        df["b"].plot.hist(ax=ax, legend=True, secondary_y=True)
+        # both legends are draw on left ax
+        # left axis must be invisible, right axis must be visible
+        _check_legend_labels(ax.left_ax, labels=["a (right)", "b (right)"])
+        assert not ax.left_ax.get_yaxis().get_visible()
+        assert ax.get_yaxis().get_visible()
+
+    def test_hist_secondary_primary(self):
+        # GH 9610
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), columns=list("abcd")
+        )
+        # secondary -> primary
+        _, ax = mpl.pyplot.subplots()
+        ax = df["a"].plot.hist(legend=True, secondary_y=True, ax=ax)
+        # right axes is returned
+        df["b"].plot.hist(ax=ax, legend=True)
+        # both legends are draw on left ax
+        # left and right axis must be visible
+        _check_legend_labels(ax.left_ax, labels=["a (right)", "b"])
+        assert ax.left_ax.get_yaxis().get_visible()
+        assert ax.get_yaxis().get_visible()
+
+    def test_hist_with_nans_and_weights(self):
+        # GH 48884
+        df = DataFrame(
+            [[np.nan, 0.2, 0.3], [0.4, np.nan, np.nan], [0.7, 0.8, 0.9]],
+            columns=list("abc"),
+        )
+        weights = np.array([0.25, 0.3, 0.45])
+        no_nan_df = DataFrame([[0.4, 0.2, 0.3], [0.7, 0.8, 0.9]], columns=list("abc"))
+        no_nan_weights = np.array([[0.3, 0.25, 0.25], [0.45, 0.45, 0.45]])
+
+        _, ax0 = mpl.pyplot.subplots()
+        df.plot.hist(ax=ax0, weights=weights)
+        rects = [x for x in ax0.get_children() if isinstance(x, mpl.patches.Rectangle)]
+        heights = [rect.get_height() for rect in rects]
+        _, ax1 = mpl.pyplot.subplots()
+        no_nan_df.plot.hist(ax=ax1, weights=no_nan_weights)
+        no_nan_rects = [
+            x for x in ax1.get_children() if isinstance(x, mpl.patches.Rectangle)
+        ]
+        no_nan_heights = [rect.get_height() for rect in no_nan_rects]
+        assert all(h0 == h1 for h0, h1 in zip(heights, no_nan_heights))
+
+        idxerror_weights = np.array([[0.3, 0.25], [0.45, 0.45]])
+
+        msg = "weights must have the same shape as data, or be a single column"
+        _, ax2 = mpl.pyplot.subplots()
+        with pytest.raises(ValueError, match=msg):
+            no_nan_df.plot.hist(ax=ax2, weights=idxerror_weights)
+
+
+class TestDataFrameGroupByPlots:
+    def test_grouped_hist_legacy(self):
+        rs = np.random.default_rng(10)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+
+        axes = _grouped_hist(df.A, by=df.C)
+        _check_axes_shape(axes, axes_num=4, layout=(2, 2))
+
+    def test_grouped_hist_legacy_axes_shape_no_col(self):
+        rs = np.random.default_rng(10)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+        axes = df.hist(by=df.C)
+        _check_axes_shape(axes, axes_num=4, layout=(2, 2))
+
+    def test_grouped_hist_legacy_single_key(self):
+        rs = np.random.default_rng(2)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+        # group by a key with single value
+        axes = df.hist(by="D", rot=30)
+        _check_axes_shape(axes, axes_num=1, layout=(1, 1))
+        _check_ticks_props(axes, xrot=30)
+
+    def test_grouped_hist_legacy_grouped_hist_kwargs(self):
+        rs = np.random.default_rng(2)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        # make sure kwargs to hist are handled
+        xf, yf = 20, 18
+        xrot, yrot = 30, 40
+
+        axes = _grouped_hist(
+            df.A,
+            by=df.C,
+            cumulative=True,
+            bins=4,
+            xlabelsize=xf,
+            xrot=xrot,
+            ylabelsize=yf,
+            yrot=yrot,
+            density=True,
+        )
+        # height of last bin (index 5) must be 1.0
+        for ax in axes.ravel():
+            rects = [
+                x for x in ax.get_children() if isinstance(x, mpl.patches.Rectangle)
+            ]
+            height = rects[-1].get_height()
+            tm.assert_almost_equal(height, 1.0)
+        _check_ticks_props(axes, xlabelsize=xf, xrot=xrot, ylabelsize=yf, yrot=yrot)
+
+    def test_grouped_hist_legacy_grouped_hist(self):
+        rs = np.random.default_rng(2)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+        axes = _grouped_hist(df.A, by=df.C, log=True)
+        # scale of y must be 'log'
+        _check_ax_scales(axes, yaxis="log")
+
+    def test_grouped_hist_legacy_external_err(self):
+        rs = np.random.default_rng(2)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+        # propagate attr exception from matplotlib.Axes.hist
+        with tm.external_error_raised(AttributeError):
+            _grouped_hist(df.A, by=df.C, foo="bar")
+
+    def test_grouped_hist_legacy_figsize_err(self):
+        rs = np.random.default_rng(2)
+        df = DataFrame(rs.standard_normal((10, 1)), columns=["A"])
+        df["B"] = to_datetime(
+            rs.integers(
+                812419200000000000,
+                819331200000000000,
+                size=10,
+                dtype=np.int64,
+            )
+        )
+        df["C"] = rs.integers(0, 4, 10)
+        df["D"] = ["X"] * 10
+        msg = "Specify figure size by tuple instead"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(by="C", figsize="default")
+
+    def test_grouped_hist_legacy2(self):
+        n = 10
+        weight = Series(np.random.default_rng(2).normal(166, 20, size=n))
+        height = Series(np.random.default_rng(2).normal(60, 10, size=n))
+        gender_int = np.random.default_rng(2).choice([0, 1], size=n)
+        df_int = DataFrame({"height": height, "weight": weight, "gender": gender_int})
+        gb = df_int.groupby("gender")
+        axes = gb.hist()
+        assert len(axes) == 2
+        assert len(mpl.pyplot.get_fignums()) == 2
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "msg, plot_col, by_col, layout",
+        [
+            [
+                "Layout of 1x1 must be larger than required size 2",
+                "weight",
+                "gender",
+                (1, 1),
+            ],
+            [
+                "Layout of 1x3 must be larger than required size 4",
+                "height",
+                "category",
+                (1, 3),
+            ],
+            [
+                "At least one dimension of layout must be positive",
+                "height",
+                "category",
+                (-1, -1),
+            ],
+        ],
+    )
+    def test_grouped_hist_layout_error(self, hist_df, msg, plot_col, by_col, layout):
+        df = hist_df
+        with pytest.raises(ValueError, match=msg):
+            df.hist(column=plot_col, by=getattr(df, by_col), layout=layout)
+
+    @pytest.mark.slow
+    def test_grouped_hist_layout_warning(self, hist_df):
+        df = hist_df
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(
+                df.hist, column="height", by=df.gender, layout=(2, 1)
+            )
+        _check_axes_shape(axes, axes_num=2, layout=(2, 1))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "layout, check_layout, figsize",
+        [[(4, 1), (4, 1), None], [(-1, 1), (4, 1), None], [(4, 2), (4, 2), (12, 8)]],
+    )
+    def test_grouped_hist_layout_figsize(self, hist_df, layout, check_layout, figsize):
+        df = hist_df
+        axes = df.hist(column="height", by=df.category, layout=layout, figsize=figsize)
+        _check_axes_shape(axes, axes_num=4, layout=check_layout, figsize=figsize)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kwargs", [{}, {"column": "height", "layout": (2, 2)}])
+    def test_grouped_hist_layout_by_warning(self, hist_df, kwargs):
+        df = hist_df
+        # GH 6769
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(df.hist, by="classroom", **kwargs)
+        _check_axes_shape(axes, axes_num=3, layout=(2, 2))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "kwargs, axes_num, layout",
+        [
+            [{"by": "gender", "layout": (3, 5)}, 2, (3, 5)],
+            [{"column": ["height", "weight", "category"]}, 3, (2, 2)],
+        ],
+    )
+    def test_grouped_hist_layout_axes(self, hist_df, kwargs, axes_num, layout):
+        df = hist_df
+        axes = df.hist(**kwargs)
+        _check_axes_shape(axes, axes_num=axes_num, layout=layout)
+
+    def test_grouped_hist_multiple_axes(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+
+        fig, axes = mpl.pyplot.subplots(2, 3)
+        returned = df.hist(column=["height", "weight", "category"], ax=axes[0])
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        tm.assert_numpy_array_equal(returned, axes[0])
+        assert returned[0].figure is fig
+
+    def test_grouped_hist_multiple_axes_no_cols(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+
+        fig, axes = mpl.pyplot.subplots(2, 3)
+        returned = df.hist(by="classroom", ax=axes[1])
+        _check_axes_shape(returned, axes_num=3, layout=(1, 3))
+        tm.assert_numpy_array_equal(returned, axes[1])
+        assert returned[0].figure is fig
+
+    def test_grouped_hist_multiple_axes_error(self, hist_df):
+        # GH 6970, GH 7069
+        df = hist_df
+        fig, axes = mpl.pyplot.subplots(2, 3)
+        # pass different number of axes from required
+        msg = "The number of passed axes must be 1, the same as the output plot"
+        with pytest.raises(ValueError, match=msg):
+            axes = df.hist(column="height", ax=axes)
+
+    def test_axis_share_x(self, hist_df):
+        df = hist_df
+        # GH4089
+        ax1, ax2 = df.hist(column="height", by=df.gender, sharex=True)
+
+        # share x
+        assert get_x_axis(ax1).joined(ax1, ax2)
+        assert get_x_axis(ax2).joined(ax1, ax2)
+
+        # don't share y
+        assert not get_y_axis(ax1).joined(ax1, ax2)
+        assert not get_y_axis(ax2).joined(ax1, ax2)
+
+    def test_axis_share_y(self, hist_df):
+        df = hist_df
+        ax1, ax2 = df.hist(column="height", by=df.gender, sharey=True)
+
+        # share y
+        assert get_y_axis(ax1).joined(ax1, ax2)
+        assert get_y_axis(ax2).joined(ax1, ax2)
+
+        # don't share x
+        assert not get_x_axis(ax1).joined(ax1, ax2)
+        assert not get_x_axis(ax2).joined(ax1, ax2)
+
+    def test_axis_share_xy(self, hist_df):
+        df = hist_df
+        ax1, ax2 = df.hist(column="height", by=df.gender, sharex=True, sharey=True)
+
+        # share both x and y
+        assert get_x_axis(ax1).joined(ax1, ax2)
+        assert get_x_axis(ax2).joined(ax1, ax2)
+
+        assert get_y_axis(ax1).joined(ax1, ax2)
+        assert get_y_axis(ax2).joined(ax1, ax2)
+
+    @pytest.mark.parametrize(
+        "histtype, expected",
+        [
+            ("bar", True),
+            ("barstacked", True),
+            ("step", False),
+            ("stepfilled", True),
+        ],
+    )
+    def test_histtype_argument(self, histtype, expected):
+        # GH23992 Verify functioning of histtype argument
+        df = DataFrame(
+            np.random.default_rng(2).integers(1, 10, size=(10, 2)), columns=["a", "b"]
+        )
+        ax = df.hist(by="a", histtype=histtype)
+        _check_patches_all_filled(ax, filled=expected)
diff --git a/pandas/pandas/tests/plotting/test_misc.py b/pandas/pandas/tests/plotting/test_misc.py
new file mode 100644
index 0000000000000..43e1255404784
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_misc.py
@@ -0,0 +1,683 @@
+"""Test cases for misc plot functions"""
+
+import os
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+    interval_range,
+    period_range,
+    plotting,
+    read_csv,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_colors,
+    _check_legend_labels,
+    _check_plot_works,
+    _check_text_labels,
+    _check_ticks_props,
+)
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+cm = pytest.importorskip("matplotlib.cm")
+
+from pandas.plotting._matplotlib.style import get_standard_colors
+
+
+@pytest.fixture
+def iris(datapath) -> DataFrame:
+    """
+    The iris dataset as a DataFrame.
+    """
+    return read_csv(datapath("io", "data", "csv", "iris.csv"))
+
+
+@td.skip_if_installed("matplotlib")
+def test_import_error_message():
+    # GH-19810
+    df = DataFrame({"A": [1, 2]})
+
+    with pytest.raises(ImportError, match="matplotlib is required for plotting"):
+        df.plot()
+
+
+def test_get_accessor_args():
+    func = plotting._core.PlotAccessor._get_call_args
+
+    msg = "Called plot accessor for type list, expected Series or DataFrame"
+    with pytest.raises(TypeError, match=msg):
+        func(backend_name="", data=[], args=[], kwargs={})
+
+    msg = "should not be called with positional arguments"
+    with pytest.raises(TypeError, match=msg):
+        func(backend_name="", data=Series(dtype=object), args=["line", None], kwargs={})
+
+    x, y, kind, kwargs = func(
+        backend_name="",
+        data=DataFrame(),
+        args=["x"],
+        kwargs={"y": "y", "kind": "bar", "grid": False},
+    )
+    assert x == "x"
+    assert y == "y"
+    assert kind == "bar"
+    assert kwargs == {"grid": False}
+
+    x, y, kind, kwargs = func(
+        backend_name="pandas.plotting._matplotlib",
+        data=Series(dtype=object),
+        args=[],
+        kwargs={},
+    )
+    assert x is None
+    assert y is None
+    assert kind == "line"
+    assert len(kwargs) == 24
+
+
+@pytest.mark.parametrize("kind", plotting.PlotAccessor._all_kinds)
+@pytest.mark.parametrize(
+    "data", [DataFrame(np.arange(15).reshape(5, 3)), Series(range(5))]
+)
+@pytest.mark.parametrize(
+    "index",
+    [
+        Index(range(5)),
+        date_range("2020-01-01", periods=5),
+        period_range("2020-01-01", periods=5),
+    ],
+)
+def test_savefig(kind, data, index):
+    fig, ax = plt.subplots()
+    data.index = index
+    kwargs = {}
+    if kind in ["hexbin", "scatter", "pie"]:
+        if isinstance(data, Series):
+            pytest.skip(f"{kind} not supported with Series")
+        kwargs = {"x": 0, "y": 1}
+    data.plot(kind=kind, ax=ax, **kwargs)
+    fig.savefig(os.devnull)
+
+
+class TestSeriesPlots:
+    def test_autocorrelation_plot(self):
+        ser = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        # Ensure no UserWarning when making plot
+        with tm.assert_produces_warning(None):
+            _check_plot_works(plotting.autocorrelation_plot, series=ser)
+            _check_plot_works(plotting.autocorrelation_plot, series=ser.values)
+
+            ax = plotting.autocorrelation_plot(ser, label="Test")
+        _check_legend_labels(ax, labels=["Test"])
+
+    @pytest.mark.parametrize("kwargs", [{}, {"lag": 5}])
+    def test_lag_plot(self, kwargs):
+        ser = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        _check_plot_works(plotting.lag_plot, series=ser, **kwargs)
+
+    def test_bootstrap_plot(self):
+        ser = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        _check_plot_works(plotting.bootstrap_plot, series=ser, size=10)
+
+
+class TestDataFramePlots:
+    @pytest.mark.parametrize("pass_axis", [False, True])
+    def test_scatter_matrix_axis(self, pass_axis):
+        pytest.importorskip("scipy")
+        scatter_matrix = plotting.scatter_matrix
+
+        ax = None
+        if pass_axis:
+            _, ax = mpl.pyplot.subplots(3, 3)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 3)))
+
+        # we are plotting multiples on a sub-plot
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(
+                scatter_matrix,
+                frame=df,
+                range_padding=0.1,
+                ax=ax,
+            )
+        axes0_labels = axes[0][0].yaxis.get_majorticklabels()
+        # GH 5662
+        expected = ["-2", "-1", "0"]
+        _check_text_labels(axes0_labels, expected)
+        _check_ticks_props(axes, xlabelsize=8, xrot=90, ylabelsize=8, yrot=0)
+
+    @pytest.mark.parametrize("pass_axis", [False, True])
+    def test_scatter_matrix_axis_smaller(self, pass_axis):
+        pytest.importorskip("scipy")
+        scatter_matrix = plotting.scatter_matrix
+
+        ax = None
+        if pass_axis:
+            _, ax = mpl.pyplot.subplots(3, 3)
+
+        df = DataFrame(np.random.default_rng(11).standard_normal((10, 3)))
+        df[0] = (df[0] - 2) / 3
+
+        # we are plotting multiples on a sub-plot
+        with tm.assert_produces_warning(UserWarning, check_stacklevel=False):
+            axes = _check_plot_works(
+                scatter_matrix,
+                frame=df,
+                range_padding=0.1,
+                ax=ax,
+            )
+        axes0_labels = axes[0][0].yaxis.get_majorticklabels()
+        expected = ["-1.25", "-1.0", "-0.75", "-0.5"]
+        _check_text_labels(axes0_labels, expected)
+        _check_ticks_props(axes, xlabelsize=8, xrot=90, ylabelsize=8, yrot=0)
+
+    @pytest.mark.slow
+    def test_andrews_curves_no_warning(self, iris):
+        # Ensure no UserWarning when making plot
+        with tm.assert_produces_warning(None):
+            _check_plot_works(plotting.andrews_curves, frame=iris, class_column="Name")
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "linecolors",
+        [
+            ("#556270", "#4ECDC4", "#C7F464"),
+            ["dodgerblue", "aquamarine", "seagreen"],
+        ],
+    )
+    @pytest.mark.parametrize(
+        "df",
+        [
+            "iris",
+            DataFrame(
+                {
+                    "A": np.random.default_rng(2).standard_normal(10),
+                    "B": np.random.default_rng(2).standard_normal(10),
+                    "C": np.random.default_rng(2).standard_normal(10),
+                    "Name": ["A"] * 10,
+                }
+            ),
+        ],
+    )
+    def test_andrews_curves_linecolors(self, request, df, linecolors):
+        if isinstance(df, str):
+            df = request.getfixturevalue(df)
+        ax = _check_plot_works(
+            plotting.andrews_curves, frame=df, class_column="Name", color=linecolors
+        )
+        _check_colors(
+            ax.get_lines()[:10], linecolors=linecolors, mapping=df["Name"][:10]
+        )
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "df",
+        [
+            "iris",
+            DataFrame(
+                {
+                    "A": np.random.default_rng(2).standard_normal(10),
+                    "B": np.random.default_rng(2).standard_normal(10),
+                    "C": np.random.default_rng(2).standard_normal(10),
+                    "Name": ["A"] * 10,
+                }
+            ),
+        ],
+    )
+    def test_andrews_curves_cmap(self, request, df):
+        if isinstance(df, str):
+            df = request.getfixturevalue(df)
+        cmaps = [cm.jet(n) for n in np.linspace(0, 1, df["Name"].nunique())]
+        ax = _check_plot_works(
+            plotting.andrews_curves, frame=df, class_column="Name", color=cmaps
+        )
+        _check_colors(ax.get_lines()[:10], linecolors=cmaps, mapping=df["Name"][:10])
+
+    @pytest.mark.slow
+    def test_andrews_curves_handle(self):
+        colors = ["b", "g", "r"]
+        df = DataFrame({"A": [1, 2, 3], "B": [1, 2, 3], "C": [1, 2, 3], "Name": colors})
+        ax = plotting.andrews_curves(df, "Name", color=colors)
+        handles, _ = ax.get_legend_handles_labels()
+        _check_colors(handles, linecolors=colors)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "color",
+        [("#556270", "#4ECDC4", "#C7F464"), ["dodgerblue", "aquamarine", "seagreen"]],
+    )
+    def test_parallel_coordinates_colors(self, iris, color):
+        df = iris
+
+        ax = _check_plot_works(
+            plotting.parallel_coordinates, frame=df, class_column="Name", color=color
+        )
+        _check_colors(ax.get_lines()[:10], linecolors=color, mapping=df["Name"][:10])
+
+    @pytest.mark.slow
+    def test_parallel_coordinates_cmap(self, iris):
+        df = iris
+
+        ax = _check_plot_works(
+            plotting.parallel_coordinates,
+            frame=df,
+            class_column="Name",
+            colormap=cm.jet,
+        )
+        cmaps = [mpl.cm.jet(n) for n in np.linspace(0, 1, df["Name"].nunique())]
+        _check_colors(ax.get_lines()[:10], linecolors=cmaps, mapping=df["Name"][:10])
+
+    @pytest.mark.slow
+    def test_parallel_coordinates_line_diff(self, iris):
+        df = iris
+
+        ax = _check_plot_works(
+            plotting.parallel_coordinates, frame=df, class_column="Name"
+        )
+        nlines = len(ax.get_lines())
+        nxticks = len(ax.xaxis.get_ticklabels())
+
+        ax = _check_plot_works(
+            plotting.parallel_coordinates, frame=df, class_column="Name", axvlines=False
+        )
+        assert len(ax.get_lines()) == (nlines - nxticks)
+
+    @pytest.mark.slow
+    def test_parallel_coordinates_handles(self, iris):
+        df = iris
+        colors = ["b", "g", "r"]
+        df = DataFrame({"A": [1, 2, 3], "B": [1, 2, 3], "C": [1, 2, 3], "Name": colors})
+        ax = plotting.parallel_coordinates(df, "Name", color=colors)
+        handles, _ = ax.get_legend_handles_labels()
+        _check_colors(handles, linecolors=colors)
+
+    # not sure if this is indicative of a problem
+    @pytest.mark.filterwarnings("ignore:Attempting to set:UserWarning")
+    def test_parallel_coordinates_with_sorted_labels(self):
+        # GH 15908
+        df = DataFrame(
+            {
+                "feat": list(range(30)),
+                "class": [2 for _ in range(10)]
+                + [3 for _ in range(10)]
+                + [1 for _ in range(10)],
+            }
+        )
+        ax = plotting.parallel_coordinates(df, "class", sort_labels=True)
+        polylines, labels = ax.get_legend_handles_labels()
+        color_label_tuples = zip(
+            [polyline.get_color() for polyline in polylines], labels
+        )
+        ordered_color_label_tuples = sorted(color_label_tuples, key=lambda x: x[1])
+        prev_next_tupels = zip(
+            list(ordered_color_label_tuples[0:-1]), list(ordered_color_label_tuples[1:])
+        )
+        for prev, nxt in prev_next_tupels:
+            # labels and colors are ordered strictly increasing
+            assert prev[1] < nxt[1] and prev[0] < nxt[0]
+
+    def test_radviz_no_warning(self, iris):
+        # Ensure no UserWarning when making plot
+        with tm.assert_produces_warning(None):
+            _check_plot_works(plotting.radviz, frame=iris, class_column="Name")
+
+    @pytest.mark.parametrize(
+        "color",
+        [("#556270", "#4ECDC4", "#C7F464"), ["dodgerblue", "aquamarine", "seagreen"]],
+    )
+    def test_radviz_color(self, iris, color):
+        df = iris
+        ax = _check_plot_works(
+            plotting.radviz, frame=df, class_column="Name", color=color
+        )
+        # skip Circle drawn as ticks
+        patches = [p for p in ax.patches[:20] if p.get_label() != ""]
+        _check_colors(patches[:10], facecolors=color, mapping=df["Name"][:10])
+
+    def test_radviz_color_cmap(self, iris):
+        df = iris
+        ax = _check_plot_works(
+            plotting.radviz, frame=df, class_column="Name", colormap=cm.jet
+        )
+        cmaps = [mpl.cm.jet(n) for n in np.linspace(0, 1, df["Name"].nunique())]
+        patches = [p for p in ax.patches[:20] if p.get_label() != ""]
+        _check_colors(patches, facecolors=cmaps, mapping=df["Name"][:10])
+
+    def test_radviz_colors_handles(self):
+        colors = [[0.0, 0.0, 1.0, 1.0], [0.0, 0.5, 1.0, 1.0], [1.0, 0.0, 0.0, 1.0]]
+        df = DataFrame(
+            {"A": [1, 2, 3], "B": [2, 1, 3], "C": [3, 2, 1], "Name": ["b", "g", "r"]}
+        )
+        ax = plotting.radviz(df, "Name", color=colors)
+        handles, _ = ax.get_legend_handles_labels()
+        _check_colors(handles, facecolors=colors)
+
+    def test_subplot_titles(self, iris):
+        df = iris.drop("Name", axis=1).head()
+        # Use the column names as the subplot titles
+        title = list(df.columns)
+
+        # Case len(title) == len(df)
+        plot = df.plot(subplots=True, title=title)
+        assert [p.get_title() for p in plot] == title
+
+    def test_subplot_titles_too_much(self, iris):
+        df = iris.drop("Name", axis=1).head()
+        # Use the column names as the subplot titles
+        title = list(df.columns)
+        # Case len(title) > len(df)
+        msg = (
+            "The length of `title` must equal the number of columns if "
+            "using `title` of type `list` and `subplots=True`"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, title=title + ["kittens > puppies"])
+
+    def test_subplot_titles_too_little(self, iris):
+        df = iris.drop("Name", axis=1).head()
+        # Use the column names as the subplot titles
+        title = list(df.columns)
+        msg = (
+            "The length of `title` must equal the number of columns if "
+            "using `title` of type `list` and `subplots=True`"
+        )
+        # Case len(title) < len(df)
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, title=title[:2])
+
+    def test_subplot_titles_subplots_false(self, iris):
+        df = iris.drop("Name", axis=1).head()
+        # Use the column names as the subplot titles
+        title = list(df.columns)
+        # Case subplots=False and title is of type list
+        msg = (
+            "Using `title` of type `list` is not supported unless "
+            "`subplots=True` is passed"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=False, title=title)
+
+    def test_subplot_titles_numeric_square_layout(self, iris):
+        df = iris.drop("Name", axis=1).head()
+        # Use the column names as the subplot titles
+        title = list(df.columns)
+        # Case df with 3 numeric columns but layout of (2,2)
+        plot = df.drop("SepalWidth", axis=1).plot(
+            subplots=True, layout=(2, 2), title=title[:-1]
+        )
+        title_list = [ax.get_title() for sublist in plot for ax in sublist]
+        assert title_list == title[:3] + [""]
+
+    def test_get_standard_colors_random_seed(self):
+        # GH17525
+        df = DataFrame(np.zeros((10, 10)))
+
+        # Make sure that the random seed isn't reset by get_standard_colors
+        plotting.parallel_coordinates(df, 0)
+        rand1 = np.random.default_rng(None).random()
+        plotting.parallel_coordinates(df, 0)
+        rand2 = np.random.default_rng(None).random()
+        assert rand1 != rand2
+
+    def test_get_standard_colors_consistency(self):
+        # GH17525
+        # Make sure it produces the same colors every time it's called
+        color1 = get_standard_colors(1, color_type="random")
+        color2 = get_standard_colors(1, color_type="random")
+        assert color1 == color2
+
+    def test_get_standard_colors_default_num_colors(self):
+        # Make sure the default color_types returns the specified amount
+        color1 = get_standard_colors(1, color_type="default")
+        color2 = get_standard_colors(9, color_type="default")
+        color3 = get_standard_colors(20, color_type="default")
+        assert len(color1) == 1
+        assert len(color2) == 9
+        assert len(color3) == 20
+
+    def test_plot_single_color(self):
+        # Example from #20585. All 3 bars should have the same color
+        df = DataFrame(
+            {
+                "account-start": ["2017-02-03", "2017-03-03", "2017-01-01"],
+                "client": ["Alice Anders", "Bob Baker", "Charlie Chaplin"],
+                "balance": [-1432.32, 10.43, 30000.00],
+                "db-id": [1234, 2424, 251],
+                "proxy-id": [525, 1525, 2542],
+                "rank": [52, 525, 32],
+            }
+        )
+        ax = df.client.value_counts().plot.bar()
+        colors = [rect.get_facecolor() for rect in ax.get_children()[0:3]]
+        assert all(color == colors[0] for color in colors)
+
+    def test_get_standard_colors_no_appending(self):
+        # GH20726
+
+        # Make sure not to add more colors so that matplotlib can cycle
+        # correctly.
+        color_before = mpl.cm.gnuplot(range(5))
+        color_after = get_standard_colors(1, color=color_before)
+        assert len(color_after) == len(color_before)
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((48, 4)), columns=list("ABCD")
+        )
+
+        color_list = mpl.cm.gnuplot(np.linspace(0, 1, 16))
+        p = df.A.plot.bar(figsize=(16, 7), color=color_list)
+        assert p.patches[1].get_facecolor() == p.patches[17].get_facecolor()
+
+    @pytest.mark.parametrize("kind", ["bar", "line"])
+    def test_dictionary_color(self, kind):
+        # issue-8193
+        # Test plot color dictionary format
+        data_files = ["a", "b"]
+
+        expected = [(0.5, 0.24, 0.6), (0.3, 0.7, 0.7)]
+
+        df1 = DataFrame(np.random.default_rng(2).random((2, 2)), columns=data_files)
+        dic_color = {"b": (0.3, 0.7, 0.7), "a": (0.5, 0.24, 0.6)}
+
+        ax = df1.plot(kind=kind, color=dic_color)
+        if kind == "bar":
+            colors = [rect.get_facecolor()[0:-1] for rect in ax.get_children()[0:3:2]]
+        else:
+            colors = [rect.get_color() for rect in ax.get_lines()[0:2]]
+        assert all(color == expected[index] for index, color in enumerate(colors))
+
+    def test_bar_plot(self):
+        # GH38947
+        # Test bar plot with string and int index
+        expected = [mpl.text.Text(0, 0, "0"), mpl.text.Text(1, 0, "Total")]
+
+        df = DataFrame(
+            {
+                "a": [1, 2],
+            },
+            index=Index([0, "Total"]),
+        )
+        plot_bar = df.plot.bar()
+        assert all(
+            (a.get_text() == b.get_text())
+            for a, b in zip(plot_bar.get_xticklabels(), expected)
+        )
+
+    def test_barh_plot_labels_mixed_integer_string(self):
+        # GH39126
+        # Test barh plot with string and integer at the same column
+        df = DataFrame([{"word": 1, "value": 0}, {"word": "knowledge", "value": 2}])
+        plot_barh = df.plot.barh(x="word", legend=None)
+        expected_yticklabels = [
+            mpl.text.Text(0, 0, "1"),
+            mpl.text.Text(0, 1, "knowledge"),
+        ]
+        assert all(
+            actual.get_text() == expected.get_text()
+            for actual, expected in zip(
+                plot_barh.get_yticklabels(), expected_yticklabels
+            )
+        )
+
+    def test_has_externally_shared_axis_x_axis(self):
+        # GH33819
+        # Test _has_externally_shared_axis() works for x-axis
+        func = plotting._matplotlib.tools._has_externally_shared_axis
+
+        fig = mpl.pyplot.figure()
+        plots = fig.subplots(2, 4)
+
+        # Create *externally* shared axes for first and third columns
+        plots[0][0] = fig.add_subplot(231, sharex=plots[1][0])
+        plots[0][2] = fig.add_subplot(233, sharex=plots[1][2])
+
+        # Create *internally* shared axes for second and third columns
+        plots[0][1].twinx()
+        plots[0][2].twinx()
+
+        # First  column is only externally shared
+        # Second column is only internally shared
+        # Third  column is both
+        # Fourth column is neither
+        assert func(plots[0][0], "x")
+        assert not func(plots[0][1], "x")
+        assert func(plots[0][2], "x")
+        assert not func(plots[0][3], "x")
+
+    def test_has_externally_shared_axis_y_axis(self):
+        # GH33819
+        # Test _has_externally_shared_axis() works for y-axis
+        func = plotting._matplotlib.tools._has_externally_shared_axis
+
+        fig = mpl.pyplot.figure()
+        plots = fig.subplots(4, 2)
+
+        # Create *externally* shared axes for first and third rows
+        plots[0][0] = fig.add_subplot(321, sharey=plots[0][1])
+        plots[2][0] = fig.add_subplot(325, sharey=plots[2][1])
+
+        # Create *internally* shared axes for second and third rows
+        plots[1][0].twiny()
+        plots[2][0].twiny()
+
+        # First  row is only externally shared
+        # Second row is only internally shared
+        # Third  row is both
+        # Fourth row is neither
+        assert func(plots[0][0], "y")
+        assert not func(plots[1][0], "y")
+        assert func(plots[2][0], "y")
+        assert not func(plots[3][0], "y")
+
+    def test_has_externally_shared_axis_invalid_compare_axis(self):
+        # GH33819
+        # Test _has_externally_shared_axis() raises an exception when
+        # passed an invalid value as compare_axis parameter
+        func = plotting._matplotlib.tools._has_externally_shared_axis
+
+        fig = mpl.pyplot.figure()
+        plots = fig.subplots(4, 2)
+
+        # Create arbitrary axes
+        plots[0][0] = fig.add_subplot(321, sharey=plots[0][1])
+
+        # Check that an invalid compare_axis value triggers the expected exception
+        msg = "needs 'x' or 'y' as a second parameter"
+        with pytest.raises(ValueError, match=msg):
+            func(plots[0][0], "z")
+
+    def test_externally_shared_axes(self):
+        # Example from GH33819
+        # Create data
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).standard_normal(10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+
+        # Create figure
+        fig = mpl.pyplot.figure()
+        plots = fig.subplots(2, 3)
+
+        # Create *externally* shared axes
+        plots[0][0] = fig.add_subplot(231, sharex=plots[1][0])
+        # note: no plots[0][1] that's the twin only case
+        plots[0][2] = fig.add_subplot(233, sharex=plots[1][2])
+
+        # Create *internally* shared axes
+        # note: no plots[0][0] that's the external only case
+        twin_ax1 = plots[0][1].twinx()
+        twin_ax2 = plots[0][2].twinx()
+
+        # Plot data to primary axes
+        df["a"].plot(ax=plots[0][0], title="External share only").set_xlabel(
+            "this label should never be visible"
+        )
+        df["a"].plot(ax=plots[1][0])
+
+        df["a"].plot(ax=plots[0][1], title="Internal share (twin) only").set_xlabel(
+            "this label should always be visible"
+        )
+        df["a"].plot(ax=plots[1][1])
+
+        df["a"].plot(ax=plots[0][2], title="Both").set_xlabel(
+            "this label should never be visible"
+        )
+        df["a"].plot(ax=plots[1][2])
+
+        # Plot data to twinned axes
+        df["b"].plot(ax=twin_ax1, color="green")
+        df["b"].plot(ax=twin_ax2, color="yellow")
+
+        assert not plots[0][0].xaxis.get_label().get_visible()
+        assert plots[0][1].xaxis.get_label().get_visible()
+        assert not plots[0][2].xaxis.get_label().get_visible()
+
+    def test_plot_bar_axis_units_timestamp_conversion(self):
+        # GH 38736
+        # Ensure string x-axis from the second plot will not be converted to datetime
+        # due to axis data from first plot
+        df = DataFrame(
+            [1.0],
+            index=[Timestamp("2022-02-22 22:22:22")],
+        )
+        _check_plot_works(df.plot)
+        s = Series({"A": 1.0})
+        _check_plot_works(s.plot.bar)
+
+    def test_bar_plt_xaxis_intervalrange(self):
+        # GH 38969
+        # Ensure IntervalIndex x-axis produces a bar plot as expected
+        expected = [mpl.text.Text(0, 0, "([0, 1],)"), mpl.text.Text(1, 0, "([1, 2],)")]
+        s = Series(
+            [1, 2],
+            index=[interval_range(0, 2, closed="both")],
+        )
+        _check_plot_works(s.plot.bar)
+        assert all(
+            (a.get_text() == b.get_text())
+            for a, b in zip(s.plot.bar().get_xticklabels(), expected)
+        )
diff --git a/pandas/pandas/tests/plotting/test_series.py b/pandas/pandas/tests/plotting/test_series.py
new file mode 100644
index 0000000000000..c3b0219971446
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_series.py
@@ -0,0 +1,973 @@
+"""Test cases for Series.plot"""
+
+from datetime import datetime
+from itertools import chain
+
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_linux
+from pandas.compat.numpy import np_version_gte1p24
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+    period_range,
+    plotting,
+)
+import pandas._testing as tm
+from pandas.tests.plotting.common import (
+    _check_ax_scales,
+    _check_axes_shape,
+    _check_colors,
+    _check_grid_settings,
+    _check_has_errorbars,
+    _check_legend_labels,
+    _check_plot_works,
+    _check_text_labels,
+    _check_ticks_props,
+    _unpack_cycler,
+    get_y_axis,
+)
+
+from pandas.tseries.offsets import CustomBusinessDay
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+
+from pandas.plotting._matplotlib.converter import DatetimeConverter
+from pandas.plotting._matplotlib.style import get_standard_colors
+
+
+@pytest.fixture
+def ts():
+    return Series(
+        np.arange(10, dtype=np.float64),
+        index=date_range("2020-01-01", periods=10),
+        name="ts",
+    )
+
+
+@pytest.fixture
+def series():
+    return Series(
+        range(10), dtype=np.float64, name="series", index=[f"i_{i}" for i in range(10)]
+    )
+
+
+class TestSeriesPlots:
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kwargs", [{"label": "foo"}, {"use_index": False}])
+    def test_plot(self, ts, kwargs):
+        _check_plot_works(ts.plot, **kwargs)
+
+    @pytest.mark.slow
+    def test_plot_tick_props(self, ts):
+        axes = _check_plot_works(ts.plot, rot=0)
+        _check_ticks_props(axes, xrot=0)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "scale, exp_scale",
+        [
+            [{"logy": True}, {"yaxis": "log"}],
+            [{"logx": True}, {"xaxis": "log"}],
+            [{"loglog": True}, {"xaxis": "log", "yaxis": "log"}],
+        ],
+    )
+    def test_plot_scales(self, ts, scale, exp_scale):
+        ax = _check_plot_works(ts.plot, style=".", **scale)
+        _check_ax_scales(ax, **exp_scale)
+
+    @pytest.mark.slow
+    def test_plot_ts_bar(self, ts):
+        _check_plot_works(ts[:10].plot.bar)
+
+    @pytest.mark.slow
+    def test_plot_ts_area_stacked(self, ts):
+        _check_plot_works(ts.plot.area, stacked=False)
+
+    def test_plot_iseries(self):
+        ser = Series(range(5), period_range("2020-01-01", periods=5))
+        _check_plot_works(ser.plot)
+
+    @pytest.mark.parametrize(
+        "kind",
+        [
+            "line",
+            "bar",
+            "barh",
+            pytest.param("kde", marks=td.skip_if_no("scipy")),
+            "hist",
+            "box",
+        ],
+    )
+    def test_plot_series_kinds(self, series, kind):
+        _check_plot_works(series[:5].plot, kind=kind)
+
+    def test_plot_series_barh(self, series):
+        _check_plot_works(series[:10].plot.barh)
+
+    def test_plot_series_bar_ax(self):
+        ax = _check_plot_works(
+            Series(np.random.default_rng(2).standard_normal(10)).plot.bar, color="black"
+        )
+        _check_colors([ax.patches[0]], facecolors=["black"])
+
+    @pytest.mark.parametrize("kwargs", [{}, {"layout": (-1, 1)}, {"layout": (1, -1)}])
+    def test_plot_6951(self, ts, kwargs):
+        # GH 6951
+        ax = _check_plot_works(ts.plot, subplots=True, **kwargs)
+        _check_axes_shape(ax, axes_num=1, layout=(1, 1))
+
+    def test_plot_figsize_and_title(self, series):
+        # figsize and title
+        _, ax = mpl.pyplot.subplots()
+        ax = series.plot(title="Test", figsize=(16, 8), ax=ax)
+        _check_text_labels(ax.title, "Test")
+        _check_axes_shape(ax, axes_num=1, layout=(1, 1), figsize=(16, 8))
+
+    def test_dont_modify_rcParams(self):
+        # GH 8242
+        key = "axes.prop_cycle"
+        colors = mpl.pyplot.rcParams[key]
+        _, ax = mpl.pyplot.subplots()
+        Series([1, 2, 3]).plot(ax=ax)
+        assert colors == mpl.pyplot.rcParams[key]
+
+    @pytest.mark.parametrize("kwargs", [{}, {"secondary_y": True}])
+    def test_ts_line_lim(self, ts, kwargs):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot(ax=ax, **kwargs)
+        xmin, xmax = ax.get_xlim()
+        lines = ax.get_lines()
+        assert xmin <= lines[0].get_data(orig=False)[0][0]
+        assert xmax >= lines[0].get_data(orig=False)[0][-1]
+
+    def test_ts_area_lim(self, ts):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.area(stacked=False, ax=ax)
+        xmin, xmax = ax.get_xlim()
+        line = ax.get_lines()[0].get_data(orig=False)[0]
+        assert xmin <= line[0]
+        assert xmax >= line[-1]
+        _check_ticks_props(ax, xrot=0)
+
+    def test_ts_area_lim_xcompat(self, ts):
+        # GH 7471
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.area(stacked=False, x_compat=True, ax=ax)
+        xmin, xmax = ax.get_xlim()
+        line = ax.get_lines()[0].get_data(orig=False)[0]
+        assert xmin <= line[0]
+        assert xmax >= line[-1]
+        _check_ticks_props(ax, xrot=30)
+
+    def test_ts_tz_area_lim_xcompat(self, ts):
+        tz_ts = ts.copy()
+        tz_ts.index = tz_ts.tz_localize("GMT").tz_convert("CET")
+        _, ax = mpl.pyplot.subplots()
+        ax = tz_ts.plot.area(stacked=False, x_compat=True, ax=ax)
+        xmin, xmax = ax.get_xlim()
+        line = ax.get_lines()[0].get_data(orig=False)[0]
+        assert xmin <= line[0]
+        assert xmax >= line[-1]
+        _check_ticks_props(ax, xrot=0)
+
+    def test_ts_tz_area_lim_xcompat_secondary_y(self, ts):
+        tz_ts = ts.copy()
+        tz_ts.index = tz_ts.tz_localize("GMT").tz_convert("CET")
+        _, ax = mpl.pyplot.subplots()
+        ax = tz_ts.plot.area(stacked=False, secondary_y=True, ax=ax)
+        xmin, xmax = ax.get_xlim()
+        line = ax.get_lines()[0].get_data(orig=False)[0]
+        assert xmin <= line[0]
+        assert xmax >= line[-1]
+        _check_ticks_props(ax, xrot=0)
+
+    def test_area_sharey_dont_overwrite(self, ts):
+        # GH37942
+        fig, (ax1, ax2) = mpl.pyplot.subplots(1, 2, sharey=True)
+
+        abs(ts).plot(ax=ax1, kind="area")
+        abs(ts).plot(ax=ax2, kind="area")
+
+        assert get_y_axis(ax1).joined(ax1, ax2)
+        assert get_y_axis(ax2).joined(ax1, ax2)
+
+    def test_label(self):
+        s = Series([1, 2])
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(label="LABEL", legend=True, ax=ax)
+        _check_legend_labels(ax, labels=["LABEL"])
+
+    def test_label_none(self):
+        s = Series([1, 2])
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(legend=True, ax=ax)
+        _check_legend_labels(ax, labels=[""])
+
+    def test_label_ser_name(self):
+        s = Series([1, 2], name="NAME")
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(legend=True, ax=ax)
+        _check_legend_labels(ax, labels=["NAME"])
+
+    def test_label_ser_name_override(self):
+        s = Series([1, 2], name="NAME")
+        # override the default
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(legend=True, label="LABEL", ax=ax)
+        _check_legend_labels(ax, labels=["LABEL"])
+
+    def test_label_ser_name_override_dont_draw(self):
+        s = Series([1, 2], name="NAME")
+        # Add lebel info, but don't draw
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(legend=False, label="LABEL", ax=ax)
+        assert ax.get_legend() is None  # Hasn't been drawn
+        ax.legend()  # draw it
+        _check_legend_labels(ax, labels=["LABEL"])
+
+    def test_boolean(self):
+        # GH 23719
+        s = Series([False, False, True])
+        _check_plot_works(s.plot, include_bool=True)
+
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            _check_plot_works(s.plot)
+
+    @pytest.mark.parametrize("index", [None, date_range("2020-01-01", periods=4)])
+    def test_line_area_nan_series(self, index):
+        values = [1, 2, np.nan, 3]
+        d = Series(values, index=index)
+        ax = _check_plot_works(d.plot)
+        masked = ax.lines[0].get_ydata()
+        # remove nan for comparison purpose
+        exp = np.array([1, 2, 3], dtype=np.float64)
+        tm.assert_numpy_array_equal(np.delete(masked.data, 2), exp)
+        tm.assert_numpy_array_equal(masked.mask, np.array([False, False, True, False]))
+
+        expected = np.array([1, 2, 0, 3], dtype=np.float64)
+        ax = _check_plot_works(d.plot, stacked=True)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected)
+        ax = _check_plot_works(d.plot.area)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected)
+        ax = _check_plot_works(d.plot.area, stacked=False)
+        tm.assert_numpy_array_equal(ax.lines[0].get_ydata(), expected)
+
+    def test_line_use_index_false(self):
+        s = Series([1, 2, 3], index=["a", "b", "c"])
+        s.index.name = "The Index"
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(use_index=False, ax=ax)
+        label = ax.get_xlabel()
+        assert label == ""
+
+    def test_line_use_index_false_diff_var(self):
+        s = Series([1, 2, 3], index=["a", "b", "c"])
+        s.index.name = "The Index"
+        _, ax = mpl.pyplot.subplots()
+        ax2 = s.plot.bar(use_index=False, ax=ax)
+        label2 = ax2.get_xlabel()
+        assert label2 == ""
+
+    @pytest.mark.xfail(
+        np_version_gte1p24 and is_platform_linux(),
+        reason="Weird rounding problems",
+        strict=False,
+    )
+    @pytest.mark.parametrize("axis, meth", [("yaxis", "bar"), ("xaxis", "barh")])
+    def test_bar_log(self, axis, meth):
+        expected = np.array([1e-1, 1e0, 1e1, 1e2, 1e3, 1e4])
+
+        _, ax = mpl.pyplot.subplots()
+        ax = getattr(Series([200, 500]).plot, meth)(log=True, ax=ax)
+        tm.assert_numpy_array_equal(getattr(ax, axis).get_ticklocs(), expected)
+
+    @pytest.mark.xfail(
+        np_version_gte1p24 and is_platform_linux(),
+        reason="Weird rounding problems",
+        strict=False,
+    )
+    @pytest.mark.parametrize(
+        "axis, kind, res_meth",
+        [["yaxis", "bar", "get_ylim"], ["xaxis", "barh", "get_xlim"]],
+    )
+    def test_bar_log_kind_bar(self, axis, kind, res_meth):
+        # GH 9905
+        expected = np.array([1e-5, 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1])
+
+        _, ax = mpl.pyplot.subplots()
+        ax = Series([0.1, 0.01, 0.001]).plot(log=True, kind=kind, ax=ax)
+        ymin = 0.0007943282347242822
+        ymax = 0.12589254117941673
+        res = getattr(ax, res_meth)()
+        tm.assert_almost_equal(res[0], ymin)
+        tm.assert_almost_equal(res[1], ymax)
+        tm.assert_numpy_array_equal(getattr(ax, axis).get_ticklocs(), expected)
+
+    def test_bar_ignore_index(self):
+        df = Series([1, 2, 3, 4], index=["a", "b", "c", "d"])
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot.bar(use_index=False, ax=ax)
+        _check_text_labels(ax.get_xticklabels(), ["0", "1", "2", "3"])
+
+    def test_bar_user_colors(self):
+        s = Series([1, 2, 3, 4])
+        ax = s.plot.bar(color=["red", "blue", "blue", "red"])
+        result = [p.get_facecolor() for p in ax.patches]
+        expected = [
+            (1.0, 0.0, 0.0, 1.0),
+            (0.0, 0.0, 1.0, 1.0),
+            (0.0, 0.0, 1.0, 1.0),
+            (1.0, 0.0, 0.0, 1.0),
+        ]
+        assert result == expected
+
+    def test_rotation_default(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        # Default rot 0
+        _, ax = mpl.pyplot.subplots()
+        axes = df.plot(ax=ax)
+        _check_ticks_props(axes, xrot=0)
+
+    def test_rotation_30(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+        _, ax = mpl.pyplot.subplots()
+        axes = df.plot(rot=30, ax=ax)
+        _check_ticks_props(axes, xrot=30)
+
+    def test_irregular_datetime(self):
+        rng = date_range("1/1/2000", "1/15/2000")
+        rng = rng[[0, 1, 2, 3, 5, 9, 10, 11, 12]]
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+        _, ax = mpl.pyplot.subplots()
+        ax = ser.plot(ax=ax)
+        xp = DatetimeConverter.convert(datetime(1999, 1, 1), "", ax)
+        ax.set_xlim("1/1/1999", "1/1/2001")
+        assert xp == ax.get_xlim()[0]
+        _check_ticks_props(ax, xrot=30)
+
+    def test_unsorted_index_xlim(self):
+        ser = Series(
+            [0.0, 1.0, np.nan, 3.0, 4.0, 5.0, 6.0],
+            index=[1.0, 0.0, 3.0, 2.0, np.nan, 3.0, 2.0],
+        )
+        _, ax = mpl.pyplot.subplots()
+        ax = ser.plot(ax=ax)
+        xmin, xmax = ax.get_xlim()
+        lines = ax.get_lines()
+        assert xmin <= np.nanmin(lines[0].get_data(orig=False)[0])
+        assert xmax >= np.nanmax(lines[0].get_data(orig=False)[0])
+
+    def test_pie_series(self):
+        # if sum of values is less than 1.0, pie handle them as rate and draw
+        # semicircle.
+        series = Series(
+            np.random.default_rng(2).integers(1, 5),
+            index=["a", "b", "c", "d", "e"],
+            name="YLABEL",
+        )
+        ax = _check_plot_works(series.plot.pie)
+        _check_text_labels(ax.texts, series.index)
+        assert ax.get_ylabel() == ""
+
+    def test_pie_arrow_type(self):
+        # GH 59192
+        pytest.importorskip("pyarrow")
+        ser = Series([1, 2, 3, 4], dtype="int32[pyarrow]")
+        _check_plot_works(ser.plot.pie)
+
+    def test_pie_series_no_label(self):
+        series = Series(
+            np.random.default_rng(2).integers(1, 5),
+            index=["a", "b", "c", "d", "e"],
+            name="YLABEL",
+        )
+        ax = _check_plot_works(series.plot.pie, labels=None)
+        _check_text_labels(ax.texts, [""] * 5)
+
+    def test_pie_series_less_colors_than_elements(self):
+        series = Series(
+            np.random.default_rng(2).integers(1, 5),
+            index=["a", "b", "c", "d", "e"],
+            name="YLABEL",
+        )
+        color_args = ["r", "g", "b"]
+        ax = _check_plot_works(series.plot.pie, colors=color_args)
+
+        color_expected = ["r", "g", "b", "r", "g"]
+        _check_colors(ax.patches, facecolors=color_expected)
+
+    def test_pie_series_labels_and_colors(self):
+        series = Series(
+            np.random.default_rng(2).integers(1, 5),
+            index=["a", "b", "c", "d", "e"],
+            name="YLABEL",
+        )
+        # with labels and colors
+        labels = ["A", "B", "C", "D", "E"]
+        color_args = ["r", "g", "b", "c", "m"]
+        ax = _check_plot_works(series.plot.pie, labels=labels, colors=color_args)
+        _check_text_labels(ax.texts, labels)
+        _check_colors(ax.patches, facecolors=color_args)
+
+    def test_pie_series_autopct_and_fontsize(self):
+        series = Series(
+            np.random.default_rng(2).integers(1, 5),
+            index=["a", "b", "c", "d", "e"],
+            name="YLABEL",
+        )
+        color_args = ["r", "g", "b", "c", "m"]
+        ax = _check_plot_works(
+            series.plot.pie, colors=color_args, autopct="%.2f", fontsize=7
+        )
+        pcts = [f"{s * 100:.2f}" for s in series.values / series.sum()]
+        expected_texts = list(chain.from_iterable(zip(series.index, pcts)))
+        _check_text_labels(ax.texts, expected_texts)
+        for t in ax.texts:
+            assert t.get_fontsize() == 7
+
+    def test_pie_series_negative_raises(self):
+        # includes negative value
+        series = Series([1, 2, 0, 4, -1], index=["a", "b", "c", "d", "e"])
+        with pytest.raises(ValueError, match="pie plot doesn't allow negative values"):
+            series.plot.pie()
+
+    def test_pie_series_nan(self):
+        # includes nan
+        series = Series([1, 2, np.nan, 4], index=["a", "b", "c", "d"], name="YLABEL")
+        ax = _check_plot_works(series.plot.pie)
+        _check_text_labels(ax.texts, ["a", "b", "", "d"])
+
+    def test_pie_nan(self):
+        s = Series([1, np.nan, 1, 1])
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot.pie(legend=True, ax=ax)
+        expected = ["0", "", "2", "3"]
+        result = [x.get_text() for x in ax.texts]
+        assert result == expected
+
+    def test_df_series_secondary_legend(self):
+        # GH 9779
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=list("abc")
+        )
+        s = Series(np.random.default_rng(2).standard_normal(10), name="x")
+
+        # primary -> secondary (without passing ax)
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(ax=ax)
+        s.plot(legend=True, secondary_y=True, ax=ax)
+        # both legends are drawn on left ax
+        # left and right axis must be visible
+        _check_legend_labels(ax, labels=["a", "b", "c", "x (right)"])
+        assert ax.get_yaxis().get_visible()
+        assert ax.right_ax.get_yaxis().get_visible()
+
+    def test_df_series_secondary_legend_both(self):
+        # GH 9779
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=list("abc")
+        )
+        s = Series(np.random.default_rng(2).standard_normal(10), name="x")
+        # secondary -> secondary (without passing ax)
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(secondary_y=True, ax=ax)
+        s.plot(legend=True, secondary_y=True, ax=ax)
+        # both legends are drawn on left ax
+        # left axis must be invisible and right axis must be visible
+        expected = ["a (right)", "b (right)", "c (right)", "x (right)"]
+        _check_legend_labels(ax.left_ax, labels=expected)
+        assert not ax.left_ax.get_yaxis().get_visible()
+        assert ax.get_yaxis().get_visible()
+
+    def test_df_series_secondary_legend_both_with_axis_2(self):
+        # GH 9779
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)), columns=list("abc")
+        )
+        s = Series(np.random.default_rng(2).standard_normal(10), name="x")
+        # secondary -> secondary (with passing ax)
+        _, ax = mpl.pyplot.subplots()
+        ax = df.plot(secondary_y=True, mark_right=False, ax=ax)
+        s.plot(ax=ax, legend=True, secondary_y=True)
+        # both legends are drawn on left ax
+        # left axis must be invisible and right axis must be visible
+        expected = ["a", "b", "c", "x (right)"]
+        _check_legend_labels(ax.left_ax, expected)
+        assert not ax.left_ax.get_yaxis().get_visible()
+        assert ax.get_yaxis().get_visible()
+
+    @pytest.mark.parametrize(
+        "input_logy, expected_scale", [(True, "log"), ("sym", "symlog")]
+    )
+    @pytest.mark.parametrize("secondary_kwarg", [{}, {"secondary_y": True}])
+    def test_secondary_logy(self, input_logy, expected_scale, secondary_kwarg):
+        # GH 25545, GH 24980
+        s1 = Series(np.random.default_rng(2).standard_normal(10))
+        ax1 = s1.plot(logy=input_logy, **secondary_kwarg)
+        assert ax1.get_yscale() == expected_scale
+
+    def test_plot_fails_with_dupe_color_and_style(self):
+        x = Series(np.random.default_rng(2).standard_normal(2))
+        _, ax = mpl.pyplot.subplots()
+        msg = (
+            "Cannot pass 'style' string with a color symbol and 'color' keyword "
+            "argument. Please use one or the other or pass 'style' without a color "
+            "symbol"
+        )
+        with pytest.raises(ValueError, match=msg):
+            x.plot(style="k--", color="k", ax=ax)
+
+    @pytest.mark.parametrize(
+        "bw_method, ind",
+        [
+            ["scott", 20],
+            [None, 20],
+            [None, np.int_(20)],
+            [0.5, np.linspace(-100, 100, 20)],
+        ],
+    )
+    def test_kde_kwargs(self, ts, bw_method, ind):
+        pytest.importorskip("scipy")
+        _check_plot_works(ts.plot.kde, bw_method=bw_method, ind=ind)
+
+    @pytest.mark.parametrize(
+        "bw_method, ind, weights",
+        [
+            ["scott", 20, None],
+            [None, 20, None],
+            [None, np.int_(20), None],
+            [0.5, np.linspace(-100, 100, 20), None],
+            ["scott", 40, np.linspace(0.0, 2.0, 50)],
+        ],
+    )
+    def test_kde_kwargs_weights(self, bw_method, ind, weights):
+        # GH59337
+        pytest.importorskip("scipy")
+        s = Series(np.random.default_rng(2).uniform(size=50))
+        _check_plot_works(s.plot.kde, bw_method=bw_method, ind=ind, weights=weights)
+
+    def test_density_kwargs(self, ts):
+        pytest.importorskip("scipy")
+        sample_points = np.linspace(-100, 100, 20)
+        _check_plot_works(ts.plot.density, bw_method=0.5, ind=sample_points)
+
+    def test_kde_kwargs_check_axes(self, ts):
+        pytest.importorskip("scipy")
+        _, ax = mpl.pyplot.subplots()
+        sample_points = np.linspace(-100, 100, 20)
+        ax = ts.plot.kde(logy=True, bw_method=0.5, ind=sample_points, ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        _check_text_labels(ax.yaxis.get_label(), "Density")
+
+    def test_kde_missing_vals(self):
+        pytest.importorskip("scipy")
+        s = Series(np.random.default_rng(2).uniform(size=50))
+        s[0] = np.nan
+        axes = _check_plot_works(s.plot.kde)
+
+        # gh-14821: check if the values have any missing values
+        assert any(~np.isnan(axes.lines[0].get_xdata()))
+
+    @pytest.mark.xfail(reason="Api changed in 3.6.0")
+    def test_boxplot_series(self, ts):
+        _, ax = mpl.pyplot.subplots()
+        ax = ts.plot.box(logy=True, ax=ax)
+        _check_ax_scales(ax, yaxis="log")
+        xlabels = ax.get_xticklabels()
+        _check_text_labels(xlabels, [ts.name])
+        ylabels = ax.get_yticklabels()
+        _check_text_labels(ylabels, [""] * len(ylabels))
+
+    @pytest.mark.parametrize(
+        "kind",
+        plotting.PlotAccessor._common_kinds + plotting.PlotAccessor._series_kinds,
+    )
+    def test_kind_kwarg(self, kind):
+        pytest.importorskip("scipy")
+        s = Series(range(3))
+        _, ax = mpl.pyplot.subplots()
+        s.plot(kind=kind, ax=ax)
+        mpl.pyplot.close()
+
+    @pytest.mark.parametrize(
+        "kind",
+        plotting.PlotAccessor._common_kinds + plotting.PlotAccessor._series_kinds,
+    )
+    def test_kind_attr(self, kind):
+        pytest.importorskip("scipy")
+        s = Series(range(3))
+        _, ax = mpl.pyplot.subplots()
+        getattr(s.plot, kind)()
+        mpl.pyplot.close()
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._common_kinds)
+    def test_invalid_plot_data(self, kind):
+        s = Series(list("abcd"))
+        _, ax = mpl.pyplot.subplots()
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            s.plot(kind=kind, ax=ax)
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._common_kinds)
+    def test_valid_object_plot(self, kind):
+        pytest.importorskip("scipy")
+        s = Series(range(10), dtype=object)
+        _check_plot_works(s.plot, kind=kind)
+
+    @pytest.mark.parametrize("kind", plotting.PlotAccessor._common_kinds)
+    def test_partially_invalid_plot_data(self, kind):
+        s = Series(["a", "b", 1.0, 2])
+        _, ax = mpl.pyplot.subplots()
+        msg = "no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            s.plot(kind=kind, ax=ax)
+
+    def test_invalid_kind(self):
+        s = Series([1, 2])
+        with pytest.raises(ValueError, match="invalid_kind is not a valid plot kind"):
+            s.plot(kind="invalid_kind")
+
+    def test_dup_datetime_index_plot(self):
+        dr1 = date_range("1/1/2009", periods=4)
+        dr2 = date_range("1/2/2009", periods=4)
+        index = dr1.append(dr2)
+        values = np.random.default_rng(2).standard_normal(index.size)
+        s = Series(values, index=index)
+        _check_plot_works(s.plot)
+
+    def test_errorbar_asymmetrical(self):
+        # GH9536
+        s = Series(np.arange(10), name="x")
+        err = np.random.default_rng(2).random((2, 10))
+
+        ax = s.plot(yerr=err, xerr=err)
+
+        result = np.vstack([i.vertices[:, 1] for i in ax.collections[1].get_paths()])
+        expected = (err.T * np.array([-1, 1])) + s.to_numpy().reshape(-1, 1)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_errorbar_asymmetrical_error(self):
+        # GH9536
+        s = Series(np.arange(10), name="x")
+        msg = (
+            "Asymmetrical error bars should be provided "
+            f"with the shape \\(2, {len(s)}\\)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            s.plot(yerr=np.random.default_rng(2).random((2, 11)))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize("kind", ["line", "bar"])
+    @pytest.mark.parametrize(
+        "yerr",
+        [
+            Series(np.abs(np.random.default_rng(2).standard_normal(10))),
+            np.abs(np.random.default_rng(2).standard_normal(10)),
+            list(np.abs(np.random.default_rng(2).standard_normal(10))),
+            DataFrame(
+                np.abs(np.random.default_rng(2).standard_normal((10, 2))),
+                columns=["x", "y"],
+            ),
+        ],
+    )
+    def test_errorbar_plot(self, kind, yerr):
+        s = Series(np.arange(10), name="x")
+        ax = _check_plot_works(s.plot, yerr=yerr, kind=kind)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+    @pytest.mark.slow
+    def test_errorbar_plot_yerr_0(self):
+        s = Series(np.arange(10), name="x")
+        s_err = np.abs(np.random.default_rng(2).standard_normal(10))
+        ax = _check_plot_works(s.plot, xerr=s_err)
+        _check_has_errorbars(ax, xerr=1, yerr=0)
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "yerr",
+        [
+            Series(np.abs(np.random.default_rng(2).standard_normal(12))),
+            DataFrame(
+                np.abs(np.random.default_rng(2).standard_normal((12, 2))),
+                columns=["x", "y"],
+            ),
+        ],
+    )
+    def test_errorbar_plot_ts(self, yerr):
+        # test time series plotting
+        ix = date_range("1/1/2000", "1/1/2001", freq="ME")
+        ts = Series(np.arange(12), index=ix, name="x")
+        yerr.index = ix
+
+        ax = _check_plot_works(ts.plot, yerr=yerr)
+        _check_has_errorbars(ax, xerr=0, yerr=1)
+
+    @pytest.mark.slow
+    def test_errorbar_plot_invalid_yerr_shape(self):
+        s = Series(np.arange(10), name="x")
+        # check incorrect lengths and types
+        with tm.external_error_raised(ValueError):
+            s.plot(yerr=np.arange(11))
+
+    @pytest.mark.slow
+    def test_errorbar_plot_invalid_yerr(self):
+        s = Series(np.arange(10), name="x")
+        s_err = ["zzz"] * 10
+        with tm.external_error_raised(TypeError):
+            s.plot(yerr=s_err)
+
+    @pytest.mark.slow
+    def test_table_true(self, series):
+        _check_plot_works(series.plot, table=True)
+
+    @pytest.mark.slow
+    def test_table_self(self, series):
+        _check_plot_works(series.plot, table=series)
+
+    @pytest.mark.slow
+    def test_series_grid_settings(self):
+        # Make sure plot defaults to rcParams['axes.grid'] setting, GH 9792
+        pytest.importorskip("scipy")
+        _check_grid_settings(
+            Series([1, 2, 3]),
+            plotting.PlotAccessor._series_kinds + plotting.PlotAccessor._common_kinds,
+        )
+
+    @pytest.mark.parametrize("c", ["r", "red", "green", "#FF0000"])
+    def test_standard_colors(self, c):
+        result = get_standard_colors(1, color=c)
+        assert result == [c]
+
+        result = get_standard_colors(1, color=[c])
+        assert result == [c]
+
+        result = get_standard_colors(3, color=c)
+        assert result == [c] * 3
+
+        result = get_standard_colors(3, color=[c])
+        assert result == [c] * 3
+
+    def test_standard_colors_all(self):
+        # multiple colors like mediumaquamarine
+        for c in mpl.colors.cnames:
+            result = get_standard_colors(num_colors=1, color=c)
+            assert result == [c]
+
+            result = get_standard_colors(num_colors=1, color=[c])
+            assert result == [c]
+
+            result = get_standard_colors(num_colors=3, color=c)
+            assert result == [c] * 3
+
+            result = get_standard_colors(num_colors=3, color=[c])
+            assert result == [c] * 3
+
+        # single letter colors like k
+        for c in mpl.colors.ColorConverter.colors:
+            result = get_standard_colors(num_colors=1, color=c)
+            assert result == [c]
+
+            result = get_standard_colors(num_colors=1, color=[c])
+            assert result == [c]
+
+            result = get_standard_colors(num_colors=3, color=c)
+            assert result == [c] * 3
+
+            result = get_standard_colors(num_colors=3, color=[c])
+            assert result == [c] * 3
+
+    def test_series_plot_color_kwargs(self):
+        # GH1890
+        _, ax = mpl.pyplot.subplots()
+        ax = Series(np.arange(12) + 1).plot(color="green", ax=ax)
+        _check_colors(ax.get_lines(), linecolors=["green"])
+
+    def test_time_series_plot_color_kwargs(self):
+        # #1890
+        _, ax = mpl.pyplot.subplots()
+        ax = Series(np.arange(12) + 1, index=date_range("1/1/2000", periods=12)).plot(
+            color="green", ax=ax
+        )
+        _check_colors(ax.get_lines(), linecolors=["green"])
+
+    def test_time_series_plot_color_with_empty_kwargs(self):
+        def_colors = _unpack_cycler(mpl.rcParams)
+        index = date_range("1/1/2000", periods=12)
+        s = Series(np.arange(1, 13), index=index)
+
+        ncolors = 3
+
+        _, ax = mpl.pyplot.subplots()
+        for i in range(ncolors):
+            ax = s.plot(ax=ax)
+        _check_colors(ax.get_lines(), linecolors=def_colors[:ncolors])
+
+    def test_xticklabels(self):
+        # GH11529
+        s = Series(np.arange(10), index=[f"P{i:02d}" for i in range(10)])
+        _, ax = mpl.pyplot.subplots()
+        ax = s.plot(xticks=[0, 3, 5, 9], ax=ax)
+        exp = [f"P{i:02d}" for i in [0, 3, 5, 9]]
+        _check_text_labels(ax.get_xticklabels(), exp)
+
+    def test_xtick_barPlot(self):
+        # GH28172
+        s = Series(range(10), index=[f"P{i:02d}" for i in range(10)])
+        ax = s.plot.bar(xticks=range(0, 11, 2))
+        exp = np.array(list(range(0, 11, 2)))
+        tm.assert_numpy_array_equal(exp, ax.get_xticks())
+
+    def test_custom_business_day_freq(self):
+        # GH7222
+        s = Series(
+            range(100, 121),
+            index=pd.bdate_range(
+                start="2014-05-01",
+                end="2014-06-01",
+                freq=CustomBusinessDay(holidays=["2014-05-26"]),
+            ),
+        )
+
+        _check_plot_works(s.plot)
+
+    @pytest.mark.xfail(
+        reason="GH#24426, see also "
+        "github.com/pandas-dev/pandas/commit/"
+        "ef1bd69fa42bbed5d09dd17f08c44fc8bfc2b685#r61470674"
+    )
+    def test_plot_accessor_updates_on_inplace(self):
+        ser = Series([1, 2, 3, 4])
+        _, ax = mpl.pyplot.subplots()
+        ax = ser.plot(ax=ax)
+        before = ax.xaxis.get_ticklocs()
+
+        ser.drop([0, 1], inplace=True)
+        _, ax = mpl.pyplot.subplots()
+        after = ax.xaxis.get_ticklocs()
+        tm.assert_numpy_array_equal(before, after)
+
+    @pytest.mark.parametrize("kind", ["line", "area"])
+    def test_plot_xlim_for_series(self, kind):
+        # test if xlim is also correctly plotted in Series for line and area
+        # GH 27686
+        s = Series([2, 3])
+        _, ax = mpl.pyplot.subplots()
+        s.plot(kind=kind, ax=ax)
+        xlims = ax.get_xlim()
+
+        assert xlims[0] < 0
+        assert xlims[1] > 1
+
+    def test_plot_no_rows(self):
+        # GH 27758
+        df = Series(dtype=int)
+        assert df.empty
+        ax = df.plot()
+        assert len(ax.get_lines()) == 1
+        line = ax.get_lines()[0]
+        assert len(line.get_xdata()) == 0
+        assert len(line.get_ydata()) == 0
+
+    def test_plot_no_numeric_data(self):
+        df = Series(["a", "b", "c"])
+        with pytest.raises(TypeError, match="no numeric data to plot"):
+            df.plot()
+
+    @pytest.mark.parametrize(
+        "data, index",
+        [
+            ([1, 2, 3, 4], [3, 2, 1, 0]),
+            ([10, 50, 20, 30], [1910, 1920, 1980, 1950]),
+        ],
+    )
+    def test_plot_order(self, data, index):
+        # GH38865 Verify plot order of a Series
+        ser = Series(data=data, index=index)
+        ax = ser.plot(kind="bar")
+
+        expected = ser.tolist()
+        result = [
+            patch.get_bbox().ymax
+            for patch in sorted(ax.patches, key=lambda patch: patch.get_bbox().xmax)
+        ]
+        assert expected == result
+
+    def test_style_single_ok(self):
+        s = Series([1, 2])
+        ax = s.plot(style="s", color="C3")
+        assert ax.lines[0].get_color() == "C3"
+
+    @pytest.mark.parametrize(
+        "index_name, old_label, new_label",
+        [(None, "", "new"), ("old", "old", "new"), (None, "", "")],
+    )
+    @pytest.mark.parametrize("kind", ["line", "area", "bar", "barh", "hist"])
+    def test_xlabel_ylabel_series(self, kind, index_name, old_label, new_label):
+        # GH 9093
+        ser = Series([1, 2, 3, 4])
+        ser.index.name = index_name
+
+        # default is the ylabel is not shown and xlabel is index name (reverse for barh)
+        ax = ser.plot(kind=kind)
+        if kind == "barh":
+            assert ax.get_xlabel() == ""
+            assert ax.get_ylabel() == old_label
+        elif kind == "hist":
+            assert ax.get_xlabel() == ""
+            assert ax.get_ylabel() == "Frequency"
+        else:
+            assert ax.get_ylabel() == ""
+            assert ax.get_xlabel() == old_label
+
+        # old xlabel will be overridden and assigned ylabel will be used as ylabel
+        ax = ser.plot(kind=kind, ylabel=new_label, xlabel=new_label)
+        assert ax.get_ylabel() == new_label
+        assert ax.get_xlabel() == new_label
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            pd.timedelta_range(start=0, periods=2, freq="D"),
+            [pd.Timedelta(days=1), pd.Timedelta(days=2)],
+        ],
+    )
+    def test_timedelta_index(self, index):
+        # GH37454
+        xlims = (3, 1)
+        ax = Series([1, 2], index=index).plot(xlim=(xlims))
+        assert ax.get_xlim() == (3, 1)
+
+    def test_series_none_color(self):
+        # GH51953
+        series = Series([1, 2, 3])
+        ax = series.plot(color=None)
+        expected = _unpack_cycler(mpl.pyplot.rcParams)[:1]
+        _check_colors(ax.get_lines(), linecolors=expected)
+
+    @pytest.mark.slow
+    def test_plot_no_warning(self, ts):
+        # GH 55138
+        # TODO(3.0): this can be removed once Period[B] deprecation is enforced
+        with tm.assert_produces_warning(False):
+            _ = ts.plot()
+
+    def test_secondary_y_subplot_axis_labels(self):
+        # GH#14102
+        s1 = Series([5, 7, 6, 8, 7], index=[1, 2, 3, 4, 5])
+        s2 = Series([6, 4, 5, 3, 4], index=[1, 2, 3, 4, 5])
+
+        ax = plt.subplot(2, 1, 1)
+        s1.plot(ax=ax)
+        s2.plot(ax=ax, secondary_y=True)
+        ax2 = plt.subplot(2, 1, 2)
+        s1.plot(ax=ax2)
+        assert len(ax.xaxis.get_minor_ticks()) == 0
+        assert len(ax.get_xticklabels()) > 0
diff --git a/pandas/pandas/tests/plotting/test_style.py b/pandas/pandas/tests/plotting/test_style.py
new file mode 100644
index 0000000000000..f9c89e0a7893f
--- /dev/null
+++ b/pandas/pandas/tests/plotting/test_style.py
@@ -0,0 +1,149 @@
+import pytest
+
+from pandas import Series
+
+mpl = pytest.importorskip("matplotlib")
+plt = pytest.importorskip("matplotlib.pyplot")
+from pandas.plotting._matplotlib.style import get_standard_colors
+
+
+class TestGetStandardColors:
+    @pytest.mark.parametrize(
+        "num_colors, expected",
+        [
+            (3, ["red", "green", "blue"]),
+            (5, ["red", "green", "blue", "red", "green"]),
+            (7, ["red", "green", "blue", "red", "green", "blue", "red"]),
+            (2, ["red", "green"]),
+            (1, ["red"]),
+        ],
+    )
+    def test_default_colors_named_from_prop_cycle(self, num_colors, expected):
+        mpl_params = {
+            "axes.prop_cycle": plt.cycler(color=["red", "green", "blue"]),
+        }
+        with mpl.rc_context(rc=mpl_params):
+            result = get_standard_colors(num_colors=num_colors)
+            assert result == expected
+
+    @pytest.mark.parametrize(
+        "num_colors, expected",
+        [
+            (1, ["b"]),
+            (3, ["b", "g", "r"]),
+            (4, ["b", "g", "r", "y"]),
+            (5, ["b", "g", "r", "y", "b"]),
+            (7, ["b", "g", "r", "y", "b", "g", "r"]),
+        ],
+    )
+    def test_default_colors_named_from_prop_cycle_string(self, num_colors, expected):
+        mpl_params = {
+            "axes.prop_cycle": plt.cycler(color="bgry"),
+        }
+        with mpl.rc_context(rc=mpl_params):
+            result = get_standard_colors(num_colors=num_colors)
+            assert result == expected
+
+    @pytest.mark.parametrize(
+        "num_colors, expected_name",
+        [
+            (1, ["C0"]),
+            (3, ["C0", "C1", "C2"]),
+            (
+                12,
+                [
+                    "C0",
+                    "C1",
+                    "C2",
+                    "C3",
+                    "C4",
+                    "C5",
+                    "C6",
+                    "C7",
+                    "C8",
+                    "C9",
+                    "C0",
+                    "C1",
+                ],
+            ),
+        ],
+    )
+    def test_default_colors_named_undefined_prop_cycle(self, num_colors, expected_name):
+        with mpl.rc_context(rc={}):
+            expected = [mpl.colors.to_hex(x) for x in expected_name]
+            result = get_standard_colors(num_colors=num_colors)
+            assert result == expected
+
+    @pytest.mark.parametrize(
+        "num_colors, expected",
+        [
+            (1, ["red", "green", (0.1, 0.2, 0.3)]),
+            (2, ["red", "green", (0.1, 0.2, 0.3)]),
+            (3, ["red", "green", (0.1, 0.2, 0.3)]),
+            (4, ["red", "green", (0.1, 0.2, 0.3), "red"]),
+        ],
+    )
+    def test_user_input_color_sequence(self, num_colors, expected):
+        color = ["red", "green", (0.1, 0.2, 0.3)]
+        result = get_standard_colors(color=color, num_colors=num_colors)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "num_colors, expected",
+        [
+            (1, ["r", "g", "b", "k"]),
+            (2, ["r", "g", "b", "k"]),
+            (3, ["r", "g", "b", "k"]),
+            (4, ["r", "g", "b", "k"]),
+            (5, ["r", "g", "b", "k", "r"]),
+            (6, ["r", "g", "b", "k", "r", "g"]),
+        ],
+    )
+    def test_user_input_color_string(self, num_colors, expected):
+        color = "rgbk"
+        result = get_standard_colors(color=color, num_colors=num_colors)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "num_colors, expected",
+        [
+            (1, [(0.1, 0.2, 0.3)]),
+            (2, [(0.1, 0.2, 0.3), (0.1, 0.2, 0.3)]),
+            (3, [(0.1, 0.2, 0.3), (0.1, 0.2, 0.3), (0.1, 0.2, 0.3)]),
+        ],
+    )
+    def test_user_input_color_floats(self, num_colors, expected):
+        color = (0.1, 0.2, 0.3)
+        result = get_standard_colors(color=color, num_colors=num_colors)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "color, num_colors, expected",
+        [
+            ("Crimson", 1, ["Crimson"]),
+            ("DodgerBlue", 2, ["DodgerBlue", "DodgerBlue"]),
+            ("firebrick", 3, ["firebrick", "firebrick", "firebrick"]),
+        ],
+    )
+    def test_user_input_named_color_string(self, color, num_colors, expected):
+        result = get_standard_colors(color=color, num_colors=num_colors)
+        assert result == expected
+
+    @pytest.mark.parametrize("color", ["", [], (), Series([], dtype="object")])
+    def test_empty_color_raises(self, color):
+        with pytest.raises(ValueError, match="Invalid color argument"):
+            get_standard_colors(color=color, num_colors=1)
+
+    @pytest.mark.parametrize(
+        "color",
+        [
+            "bad_color",
+            ("red", "green", "bad_color"),
+            (0.1,),
+            (0.1, 0.2),
+            (0.1, 0.2, 0.3, 0.4, 0.5),  # must be either 3 or 4 floats
+        ],
+    )
+    def test_bad_color_raises(self, color):
+        with pytest.raises(ValueError, match="Invalid color"):
+            get_standard_colors(color=color, num_colors=5)
diff --git a/pandas/pandas/tests/reductions/__init__.py b/pandas/pandas/tests/reductions/__init__.py
new file mode 100644
index 0000000000000..e3851753b6742
--- /dev/null
+++ b/pandas/pandas/tests/reductions/__init__.py
@@ -0,0 +1,4 @@
+"""
+Tests for reductions where we want to test for matching behavior across
+Array, Index, Series, and DataFrame methods.
+"""
diff --git a/pandas/pandas/tests/reductions/test_reductions.py b/pandas/pandas/tests/reductions/test_reductions.py
new file mode 100644
index 0000000000000..a7bb80727206e
--- /dev/null
+++ b/pandas/pandas/tests/reductions/test_reductions.py
@@ -0,0 +1,1670 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+from decimal import Decimal
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    NaT,
+    Period,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    date_range,
+    isna,
+    period_range,
+    timedelta_range,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.core import nanops
+from pandas.core.arrays.string_arrow import ArrowStringArrayNumpySemantics
+
+
+def get_objs():
+    indexes = [
+        Index([True, False] * 5, name="a"),
+        Index(np.arange(10), dtype=np.int64, name="a"),
+        Index(np.arange(10), dtype=np.float64, name="a"),
+        DatetimeIndex(date_range("2020-01-01", periods=10), name="a"),
+        DatetimeIndex(date_range("2020-01-01", periods=10), name="a").tz_localize(
+            tz="US/Eastern"
+        ),
+        PeriodIndex(period_range("2020-01-01", periods=10, freq="D"), name="a"),
+        Index([str(i) for i in range(10)], name="a"),
+    ]
+
+    arr = np.random.default_rng(2).standard_normal(10)
+    series = [Series(arr, index=idx, name="a") for idx in indexes]
+
+    objs = indexes + series
+    return objs
+
+
+class TestReductions:
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    @pytest.mark.parametrize("opname", ["max", "min"])
+    @pytest.mark.parametrize("obj", get_objs())
+    def test_ops(self, opname, obj):
+        result = getattr(obj, opname)()
+        if not isinstance(obj, PeriodIndex):
+            if isinstance(obj.values, ArrowStringArrayNumpySemantics):
+                # max not on the interface
+                expected = getattr(np.array(obj.values), opname)()
+            else:
+                expected = getattr(obj.values, opname)()
+        else:
+            expected = Period(ordinal=getattr(obj.asi8, opname)(), freq=obj.freq)
+
+        if getattr(obj, "tz", None) is not None:
+            # We need to de-localize before comparing to the numpy-produced result
+            expected = expected.astype("M8[ns]").astype("int64")
+            assert result._value == expected
+        else:
+            assert result == expected
+
+    @pytest.mark.parametrize("opname", ["max", "min"])
+    @pytest.mark.parametrize(
+        "dtype, val",
+        [
+            ("object", 2.0),
+            ("float64", 2.0),
+            ("datetime64[ns]", datetime(2011, 11, 1)),
+            ("Int64", 2),
+            ("boolean", True),
+        ],
+    )
+    def test_nanminmax(self, opname, dtype, val, index_or_series):
+        # GH#7261
+        klass = index_or_series
+
+        def check_missing(res):
+            if dtype == "datetime64[ns]":
+                return res is NaT
+            elif dtype in ["Int64", "boolean"]:
+                return res is pd.NA
+            else:
+                return isna(res)
+
+        obj = klass([None], dtype=dtype)
+        assert check_missing(getattr(obj, opname)())
+        assert check_missing(getattr(obj, opname)(skipna=False))
+
+        obj = klass([], dtype=dtype)
+        assert check_missing(getattr(obj, opname)())
+        assert check_missing(getattr(obj, opname)(skipna=False))
+
+        if dtype == "object":
+            # generic test with object only works for empty / all NaN
+            return
+
+        obj = klass([None, val], dtype=dtype)
+        assert getattr(obj, opname)() == val
+        assert check_missing(getattr(obj, opname)(skipna=False))
+
+        obj = klass([None, val, None], dtype=dtype)
+        assert getattr(obj, opname)() == val
+        assert check_missing(getattr(obj, opname)(skipna=False))
+
+    @pytest.mark.parametrize("opname", ["max", "min"])
+    def test_nanargminmax(self, opname, index_or_series):
+        # GH#7261
+        klass = index_or_series
+        arg_op = "arg" + opname if klass is Index else "idx" + opname
+
+        obj = klass([NaT, datetime(2011, 11, 1)])
+        assert getattr(obj, arg_op)() == 1
+
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            getattr(obj, arg_op)(skipna=False)
+
+        obj = klass([NaT, datetime(2011, 11, 1), NaT])
+        # check DatetimeIndex non-monotonic path
+        assert getattr(obj, arg_op)() == 1
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            getattr(obj, arg_op)(skipna=False)
+
+    @pytest.mark.parametrize("opname", ["max", "min"])
+    @pytest.mark.parametrize("dtype", ["M8[ns]", "datetime64[ns, UTC]"])
+    def test_nanops_empty_object(self, opname, index_or_series, dtype):
+        klass = index_or_series
+        arg_op = "arg" + opname if klass is Index else "idx" + opname
+
+        obj = klass([], dtype=dtype)
+
+        assert getattr(obj, opname)() is NaT
+        assert getattr(obj, opname)(skipna=False) is NaT
+
+        with pytest.raises(ValueError, match="empty sequence"):
+            getattr(obj, arg_op)()
+        with pytest.raises(ValueError, match="empty sequence"):
+            getattr(obj, arg_op)(skipna=False)
+
+    def test_argminmax(self):
+        obj = Index(np.arange(5, dtype="int64"))
+        assert obj.argmin() == 0
+        assert obj.argmax() == 4
+
+        obj = Index([np.nan, 1, np.nan, 2])
+        assert obj.argmin() == 1
+        assert obj.argmax() == 3
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmax(skipna=False)
+
+        obj = Index([np.nan])
+        with pytest.raises(ValueError, match="Encountered all NA values"):
+            obj.argmin()
+        with pytest.raises(ValueError, match="Encountered all NA values"):
+            obj.argmax()
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmax(skipna=False)
+
+        obj = Index([NaT, datetime(2011, 11, 1), datetime(2011, 11, 2), NaT])
+        assert obj.argmin() == 1
+        assert obj.argmax() == 2
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmax(skipna=False)
+
+        obj = Index([NaT])
+        with pytest.raises(ValueError, match="Encountered all NA values"):
+            obj.argmin()
+        with pytest.raises(ValueError, match="Encountered all NA values"):
+            obj.argmax()
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            obj.argmax(skipna=False)
+
+    @pytest.mark.parametrize("op, expected_col", [["max", "a"], ["min", "b"]])
+    def test_same_tz_min_max_axis_1(self, op, expected_col):
+        # GH 10390
+        df = DataFrame(
+            date_range("2016-01-01 00:00:00", periods=3, tz="UTC"), columns=["a"]
+        )
+        df["b"] = df.a.subtract(Timedelta(seconds=3600))
+        result = getattr(df, op)(axis=1)
+        expected = df[expected_col].rename(None)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("func", ["maximum", "minimum"])
+    def test_numpy_reduction_with_tz_aware_dtype(self, tz_aware_fixture, func):
+        # GH 15552
+        tz = tz_aware_fixture
+        arg = pd.to_datetime(["2019"]).tz_localize(tz)
+        expected = Series(arg)
+        result = getattr(np, func)(expected, expected)
+        tm.assert_series_equal(result, expected)
+
+    def test_nan_int_timedelta_sum(self):
+        # GH 27185
+        df = DataFrame(
+            {
+                "A": Series([1, 2, NaT], dtype="timedelta64[ns]"),
+                "B": Series([1, 2, np.nan], dtype="Int64"),
+            }
+        )
+        expected = Series({"A": Timedelta(3), "B": 3})
+        result = df.sum()
+        tm.assert_series_equal(result, expected)
+
+
+class TestIndexReductions:
+    # Note: the name TestIndexReductions indicates these tests
+    #  were moved from a Index-specific test file, _not_ that these tests are
+    #  intended long-term to be Index-specific
+
+    @pytest.mark.parametrize(
+        "start,stop,step",
+        [
+            (0, 400, 3),
+            (500, 0, -6),
+            (-(10**6), 10**6, 4),
+            (10**6, -(10**6), -4),
+            (0, 10, 20),
+        ],
+    )
+    def test_max_min_range(self, start, stop, step):
+        # GH#17607
+        idx = RangeIndex(start, stop, step)
+        expected = idx._values.max()
+        result = idx.max()
+        assert result == expected
+
+        # skipna should be irrelevant since RangeIndex should never have NAs
+        result2 = idx.max(skipna=False)
+        assert result2 == expected
+
+        expected = idx._values.min()
+        result = idx.min()
+        assert result == expected
+
+        # skipna should be irrelevant since RangeIndex should never have NAs
+        result2 = idx.min(skipna=False)
+        assert result2 == expected
+
+        # empty
+        idx = RangeIndex(start, stop, -step)
+        assert isna(idx.max())
+        assert isna(idx.min())
+
+    def test_minmax_timedelta64(self):
+        # monotonic
+        idx1 = TimedeltaIndex(["1 days", "2 days", "3 days"])
+        assert idx1.is_monotonic_increasing
+
+        # non-monotonic
+        idx2 = TimedeltaIndex(["1 days", np.nan, "3 days", "NaT"])
+        assert not idx2.is_monotonic_increasing
+
+        for idx in [idx1, idx2]:
+            assert idx.min() == Timedelta("1 days")
+            assert idx.max() == Timedelta("3 days")
+            assert idx.argmin() == 0
+            assert idx.argmax() == 2
+
+    @pytest.mark.parametrize("op", ["min", "max"])
+    def test_minmax_timedelta_empty_or_na(self, op):
+        # Return NaT
+        obj = TimedeltaIndex([])
+        assert getattr(obj, op)() is NaT
+
+        obj = TimedeltaIndex([NaT])
+        assert getattr(obj, op)() is NaT
+
+        obj = TimedeltaIndex([NaT, NaT, NaT])
+        assert getattr(obj, op)() is NaT
+
+    def test_numpy_minmax_timedelta64(self):
+        td = timedelta_range("16815 days", "16820 days", freq="D")
+
+        assert np.min(td) == Timedelta("16815 days")
+        assert np.max(td) == Timedelta("16820 days")
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.min(td, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.max(td, out=0)
+
+        assert np.argmin(td) == 0
+        assert np.argmax(td) == 5
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmin(td, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmax(td, out=0)
+
+    def test_timedelta_ops(self):
+        # GH#4984
+        # make sure ops return Timedelta
+        s = Series(
+            [Timestamp("20130101") + timedelta(seconds=i * i) for i in range(10)]
+        )
+        td = s.diff()
+
+        result = td.mean()
+        expected = to_timedelta(timedelta(seconds=9))
+        assert result == expected
+
+        result = td.to_frame().mean()
+        assert result[0] == expected
+
+        result = td.quantile(0.1)
+        expected = Timedelta(np.timedelta64(2600, "ms"))
+        assert result == expected
+
+        result = td.median()
+        expected = to_timedelta("00:00:09")
+        assert result == expected
+
+        result = td.to_frame().median()
+        assert result[0] == expected
+
+        # GH#6462
+        # consistency in returned values for sum
+        result = td.sum()
+        expected = to_timedelta("00:01:21")
+        assert result == expected
+
+        result = td.to_frame().sum()
+        assert result[0] == expected
+
+        # std
+        result = td.std()
+        expected = to_timedelta(Series(td.dropna().values).std())
+        assert result == expected
+
+        result = td.to_frame().std()
+        assert result[0] == expected
+
+        # GH#10040
+        # make sure NaT is properly handled by median()
+        s = Series([Timestamp("2015-02-03"), Timestamp("2015-02-07")])
+        assert s.diff().median() == timedelta(days=4)
+
+        s = Series(
+            [Timestamp("2015-02-03"), Timestamp("2015-02-07"), Timestamp("2015-02-15")]
+        )
+        assert s.diff().median() == timedelta(days=6)
+
+    @pytest.mark.parametrize("opname", ["skew", "kurt", "sem", "prod", "var"])
+    def test_invalid_td64_reductions(self, opname):
+        s = Series(
+            [Timestamp("20130101") + timedelta(seconds=i * i) for i in range(10)]
+        )
+        td = s.diff()
+
+        msg = "|".join(
+            [
+                f"reduction operation '{opname}' not allowed for this dtype",
+                rf"cannot perform {opname} with type timedelta64\[ns\]",
+                f"does not support operation '{opname}'",
+            ]
+        )
+
+        with pytest.raises(TypeError, match=msg):
+            getattr(td, opname)()
+
+        with pytest.raises(TypeError, match=msg):
+            getattr(td.to_frame(), opname)(numeric_only=False)
+
+    def test_minmax_tz(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        # monotonic
+        idx1 = DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"], tz=tz)
+        assert idx1.is_monotonic_increasing
+
+        # non-monotonic
+        idx2 = DatetimeIndex(
+            ["2011-01-01", NaT, "2011-01-03", "2011-01-02", NaT], tz=tz
+        )
+        assert not idx2.is_monotonic_increasing
+
+        for idx in [idx1, idx2]:
+            assert idx.min() == Timestamp("2011-01-01", tz=tz)
+            assert idx.max() == Timestamp("2011-01-03", tz=tz)
+            assert idx.argmin() == 0
+            assert idx.argmax() == 2
+
+    @pytest.mark.parametrize("op", ["min", "max"])
+    def test_minmax_nat_datetime64(self, op):
+        # Return NaT
+        obj = DatetimeIndex([])
+        assert isna(getattr(obj, op)())
+
+        obj = DatetimeIndex([NaT])
+        assert isna(getattr(obj, op)())
+
+        obj = DatetimeIndex([NaT, NaT, NaT])
+        assert isna(getattr(obj, op)())
+
+    def test_numpy_minmax_integer(self):
+        # GH#26125
+        idx = Index([1, 2, 3])
+
+        expected = idx.values.max()
+        result = np.max(idx)
+        assert result == expected
+
+        expected = idx.values.min()
+        result = np.min(idx)
+        assert result == expected
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.min(idx, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.max(idx, out=0)
+
+        expected = idx.values.argmax()
+        result = np.argmax(idx)
+        assert result == expected
+
+        expected = idx.values.argmin()
+        result = np.argmin(idx)
+        assert result == expected
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmin(idx, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmax(idx, out=0)
+
+    def test_numpy_minmax_range(self):
+        # GH#26125
+        idx = RangeIndex(0, 10, 3)
+
+        result = np.max(idx)
+        assert result == 9
+
+        result = np.min(idx)
+        assert result == 0
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.min(idx, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.max(idx, out=0)
+
+        # No need to test again argmax/argmin compat since the implementation
+        # is the same as basic integer index
+
+    def test_numpy_minmax_datetime64(self):
+        dr = date_range(start="2016-01-15", end="2016-01-20")
+
+        assert np.min(dr) == Timestamp("2016-01-15 00:00:00")
+        assert np.max(dr) == Timestamp("2016-01-20 00:00:00")
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.min(dr, out=0)
+
+        with pytest.raises(ValueError, match=errmsg):
+            np.max(dr, out=0)
+
+        assert np.argmin(dr) == 0
+        assert np.argmax(dr) == 5
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmin(dr, out=0)
+
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmax(dr, out=0)
+
+    def test_minmax_period(self):
+        # monotonic
+        idx1 = PeriodIndex([NaT, "2011-01-01", "2011-01-02", "2011-01-03"], freq="D")
+        assert not idx1.is_monotonic_increasing
+        assert idx1[1:].is_monotonic_increasing
+
+        # non-monotonic
+        idx2 = PeriodIndex(
+            ["2011-01-01", NaT, "2011-01-03", "2011-01-02", NaT], freq="D"
+        )
+        assert not idx2.is_monotonic_increasing
+
+        for idx in [idx1, idx2]:
+            assert idx.min() == Period("2011-01-01", freq="D")
+            assert idx.max() == Period("2011-01-03", freq="D")
+        assert idx1.argmin() == 1
+        assert idx2.argmin() == 0
+        assert idx1.argmax() == 3
+        assert idx2.argmax() == 2
+
+    @pytest.mark.parametrize("op", ["min", "max"])
+    @pytest.mark.parametrize("data", [[], [NaT], [NaT, NaT, NaT]])
+    def test_minmax_period_empty_nat(self, op, data):
+        # Return NaT
+        obj = PeriodIndex(data, freq="M")
+        result = getattr(obj, op)()
+        assert result is NaT
+
+    def test_numpy_minmax_period(self):
+        pr = period_range(start="2016-01-15", end="2016-01-20")
+
+        assert np.min(pr) == Period("2016-01-15", freq="D")
+        assert np.max(pr) == Period("2016-01-20", freq="D")
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.min(pr, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.max(pr, out=0)
+
+        assert np.argmin(pr) == 0
+        assert np.argmax(pr) == 5
+
+        errmsg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmin(pr, out=0)
+        with pytest.raises(ValueError, match=errmsg):
+            np.argmax(pr, out=0)
+
+    def test_min_max_categorical(self):
+        ci = pd.CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
+        msg = (
+            r"Categorical is not ordered for operation min\n"
+            r"you can use .as_ordered\(\) to change the Categorical to an ordered one\n"
+        )
+        with pytest.raises(TypeError, match=msg):
+            ci.min()
+        msg = (
+            r"Categorical is not ordered for operation max\n"
+            r"you can use .as_ordered\(\) to change the Categorical to an ordered one\n"
+        )
+        with pytest.raises(TypeError, match=msg):
+            ci.max()
+
+        ci = pd.CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=True)
+        assert ci.min() == "c"
+        assert ci.max() == "b"
+
+
+class TestSeriesReductions:
+    # Note: the name TestSeriesReductions indicates these tests
+    #  were moved from a series-specific test file, _not_ that these tests are
+    #  intended long-term to be series-specific
+
+    def test_sum_inf(self):
+        s = Series(np.random.default_rng(2).standard_normal(10))
+        s2 = s.copy()
+
+        s[5:8] = np.inf
+        s2[5:8] = np.nan
+
+        assert np.isinf(s.sum())
+
+        arr = np.random.default_rng(2).standard_normal((100, 100)).astype("f4")
+        arr[:, 2] = np.inf
+
+        res = nanops.nansum(arr, axis=1)
+        assert np.isinf(res).all()
+
+    @pytest.mark.parametrize(
+        "dtype", ["float64", "Float32", "Int64", "boolean", "object"]
+    )
+    @pytest.mark.parametrize("use_bottleneck", [True, False])
+    @pytest.mark.parametrize("method, unit", [("sum", 0.0), ("prod", 1.0)])
+    def test_empty(self, method, unit, use_bottleneck, dtype):
+        with pd.option_context("use_bottleneck", use_bottleneck):
+            # GH#9422 / GH#18921
+            # Entirely empty
+            s = Series([], dtype=dtype)
+            # NA by default
+            result = getattr(s, method)()
+            assert result == unit
+
+            # Explicit
+            result = getattr(s, method)(min_count=0)
+            assert result == unit
+
+            result = getattr(s, method)(min_count=1)
+            assert isna(result)
+
+            # Skipna, default
+            result = getattr(s, method)(skipna=True)
+            result == unit
+
+            # Skipna, explicit
+            result = getattr(s, method)(skipna=True, min_count=0)
+            assert result == unit
+
+            result = getattr(s, method)(skipna=True, min_count=1)
+            assert isna(result)
+
+            result = getattr(s, method)(skipna=False, min_count=0)
+            assert result == unit
+
+            result = getattr(s, method)(skipna=False, min_count=1)
+            assert isna(result)
+
+            # All-NA
+            s = Series([np.nan], dtype=dtype)
+            # NA by default
+            result = getattr(s, method)()
+            assert result == unit
+
+            # Explicit
+            result = getattr(s, method)(min_count=0)
+            assert result == unit
+
+            result = getattr(s, method)(min_count=1)
+            assert isna(result)
+
+            # Skipna, default
+            result = getattr(s, method)(skipna=True)
+            result == unit
+
+            # skipna, explicit
+            result = getattr(s, method)(skipna=True, min_count=0)
+            assert result == unit
+
+            result = getattr(s, method)(skipna=True, min_count=1)
+            assert isna(result)
+
+            # Mix of valid, empty
+            s = Series([np.nan, 1], dtype=dtype)
+            # Default
+            result = getattr(s, method)()
+            assert result == 1.0
+
+            # Explicit
+            result = getattr(s, method)(min_count=0)
+            assert result == 1.0
+
+            result = getattr(s, method)(min_count=1)
+            assert result == 1.0
+
+            # Skipna
+            result = getattr(s, method)(skipna=True)
+            assert result == 1.0
+
+            result = getattr(s, method)(skipna=True, min_count=0)
+            assert result == 1.0
+
+            # GH#844 (changed in GH#9422)
+            df = DataFrame(np.empty((10, 0)), dtype=dtype)
+            assert (getattr(df, method)(axis=1) == unit).all()
+
+            s = Series([1], dtype=dtype)
+            result = getattr(s, method)(min_count=2)
+            assert isna(result)
+
+            result = getattr(s, method)(skipna=False, min_count=2)
+            assert isna(result)
+
+            s = Series([np.nan], dtype=dtype)
+            result = getattr(s, method)(min_count=2)
+            assert isna(result)
+
+            s = Series([np.nan, 1], dtype=dtype)
+            result = getattr(s, method)(min_count=2)
+            assert isna(result)
+
+    @pytest.mark.parametrize("method", ["mean", "var"])
+    @pytest.mark.parametrize("dtype", ["Float64", "Int64", "boolean"])
+    def test_ops_consistency_on_empty_nullable(self, method, dtype):
+        # GH#34814
+        # consistency for nullable dtypes on empty or ALL-NA mean
+
+        # empty series
+        eser = Series([], dtype=dtype)
+        result = getattr(eser, method)()
+        assert result is pd.NA
+
+        # ALL-NA series
+        nser = Series([np.nan], dtype=dtype)
+        result = getattr(nser, method)()
+        assert result is pd.NA
+
+    @pytest.mark.parametrize("method", ["mean", "median", "std", "var"])
+    def test_ops_consistency_on_empty(self, method):
+        # GH#7869
+        # consistency on empty
+
+        # float
+        result = getattr(Series(dtype=float), method)()
+        assert isna(result)
+
+        # timedelta64[ns]
+        tdser = Series([], dtype="m8[ns]")
+        if method == "var":
+            msg = "|".join(
+                [
+                    "operation 'var' not allowed",
+                    r"cannot perform var with type timedelta64\[ns\]",
+                    "does not support operation 'var'",
+                ]
+            )
+            with pytest.raises(TypeError, match=msg):
+                getattr(tdser, method)()
+        else:
+            result = getattr(tdser, method)()
+            assert result is NaT
+
+    def test_nansum_buglet(self):
+        ser = Series([1.0, np.nan], index=[0, 1])
+        result = np.nansum(ser)
+        tm.assert_almost_equal(result, 1)
+
+    @pytest.mark.parametrize("use_bottleneck", [True, False])
+    @pytest.mark.parametrize("dtype", ["int32", "int64"])
+    def test_sum_overflow_int(self, use_bottleneck, dtype):
+        with pd.option_context("use_bottleneck", use_bottleneck):
+            # GH#6915
+            # overflowing on the smaller int dtypes
+            v = np.arange(5000000, dtype=dtype)
+            s = Series(v)
+
+            result = s.sum(skipna=False)
+            assert int(result) == v.sum(dtype="int64")
+            result = s.min(skipna=False)
+            assert int(result) == 0
+            result = s.max(skipna=False)
+            assert int(result) == v[-1]
+
+    @pytest.mark.parametrize("use_bottleneck", [True, False])
+    @pytest.mark.parametrize("dtype", ["float32", "float64"])
+    def test_sum_overflow_float(self, use_bottleneck, dtype):
+        with pd.option_context("use_bottleneck", use_bottleneck):
+            v = np.arange(5000000, dtype=dtype)
+            s = Series(v)
+
+            result = s.sum(skipna=False)
+            assert result == v.sum(dtype=dtype)
+            result = s.min(skipna=False)
+            assert np.allclose(float(result), 0.0)
+            result = s.max(skipna=False)
+            assert np.allclose(float(result), v[-1])
+
+    def test_mean_masked_overflow(self):
+        # GH#48378
+        val = 100_000_000_000_000_000
+        n_elements = 100
+        na = np.array([val] * n_elements)
+        ser = Series([val] * n_elements, dtype="Int64")
+
+        result_numpy = np.mean(na)
+        result_masked = ser.mean()
+        assert result_masked - result_numpy == 0
+        assert result_masked == 1e17
+
+    @pytest.mark.parametrize("ddof, exp", [(1, 2.5), (0, 2.0)])
+    def test_var_masked_array(self, ddof, exp):
+        # GH#48379
+        ser = Series([1, 2, 3, 4, 5], dtype="Int64")
+        ser_numpy_dtype = Series([1, 2, 3, 4, 5], dtype="int64")
+        result = ser.var(ddof=ddof)
+        result_numpy_dtype = ser_numpy_dtype.var(ddof=ddof)
+        assert result == result_numpy_dtype
+        assert result == exp
+
+    @pytest.mark.parametrize("dtype", ("m8[ns]", "M8[ns]", "M8[ns, UTC]"))
+    def test_empty_timeseries_reductions_return_nat(self, dtype, skipna):
+        # covers GH#11245
+        assert Series([], dtype=dtype).min(skipna=skipna) is NaT
+        assert Series([], dtype=dtype).max(skipna=skipna) is NaT
+
+    def test_numpy_argmin(self):
+        # See GH#16830
+        data = np.arange(1, 11)
+
+        s = Series(data, index=data)
+        result = np.argmin(s)
+
+        expected = np.argmin(data)
+        assert result == expected
+
+        result = s.argmin()
+
+        assert result == expected
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argmin(s, out=data)
+
+    def test_numpy_argmax(self):
+        # See GH#16830
+        data = np.arange(1, 11)
+
+        ser = Series(data, index=data)
+        result = np.argmax(ser)
+        expected = np.argmax(data)
+        assert result == expected
+
+        result = ser.argmax()
+
+        assert result == expected
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.argmax(ser, out=data)
+
+    def test_idxmin_dt64index(self, unit):
+        # GH#43587 should have NaT instead of NaN
+        dti = DatetimeIndex(["NaT", "2015-02-08", "NaT"]).as_unit(unit)
+        ser = Series([1.0, 2.0, np.nan], index=dti)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            ser.idxmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            ser.idxmax(skipna=False)
+
+        df = ser.to_frame()
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            df.idxmin(skipna=False)
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            df.idxmax(skipna=False)
+
+    def test_idxmin(self):
+        # test idxmin
+        # _check_stat_op approach can not be used here because of isna check.
+        string_series = Series(range(20), dtype=np.float64, name="series")
+
+        # add some NaNs
+        string_series[5:15] = np.nan
+
+        # skipna or no
+        assert string_series[string_series.idxmin()] == string_series.min()
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            string_series.idxmin(skipna=False)
+
+        # no NaNs
+        nona = string_series.dropna()
+        assert nona[nona.idxmin()] == nona.min()
+        assert nona.index.values.tolist().index(nona.idxmin()) == nona.values.argmin()
+
+        # all NaNs
+        allna = string_series * np.nan
+        msg = "Encountered all NA values"
+        with pytest.raises(ValueError, match=msg):
+            allna.idxmin()
+
+        # datetime64[ns]
+        s = Series(date_range("20130102", periods=6))
+        result = s.idxmin()
+        assert result == 0
+
+        s[0] = np.nan
+        result = s.idxmin()
+        assert result == 1
+
+    def test_idxmax(self):
+        # test idxmax
+        # _check_stat_op approach can not be used here because of isna check.
+        string_series = Series(range(20), dtype=np.float64, name="series")
+
+        # add some NaNs
+        string_series[5:15] = np.nan
+
+        # skipna or no
+        assert string_series[string_series.idxmax()] == string_series.max()
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            assert isna(string_series.idxmax(skipna=False))
+
+        # no NaNs
+        nona = string_series.dropna()
+        assert nona[nona.idxmax()] == nona.max()
+        assert nona.index.values.tolist().index(nona.idxmax()) == nona.values.argmax()
+
+        # all NaNs
+        allna = string_series * np.nan
+        msg = "Encountered all NA values"
+        with pytest.raises(ValueError, match=msg):
+            allna.idxmax()
+
+        s = Series(date_range("20130102", periods=6))
+        result = s.idxmax()
+        assert result == 5
+
+        s[5] = np.nan
+        result = s.idxmax()
+        assert result == 4
+
+        # Index with float64 dtype
+        # GH#5914
+        s = Series([1, 2, 3], [1.1, 2.1, 3.1])
+        result = s.idxmax()
+        assert result == 3.1
+        result = s.idxmin()
+        assert result == 1.1
+
+        s = Series(s.index, s.index)
+        result = s.idxmax()
+        assert result == 3.1
+        result = s.idxmin()
+        assert result == 1.1
+
+    def test_all_any(self):
+        ts = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        bool_series = ts > 0
+        assert not bool_series.all()
+        assert bool_series.any()
+
+        # Alternative types, with implicit 'object' dtype.
+        s = Series(["abc", True])
+        assert s.any()
+
+    def test_numpy_all_any(self, index_or_series):
+        # GH#40180
+        idx = index_or_series([0, 1, 2])
+        assert not np.all(idx)
+        assert np.any(idx)
+        idx = Index([1, 2, 3])
+        assert np.all(idx)
+
+    def test_all_any_skipna(self):
+        # Check skipna, with implicit 'object' dtype.
+        s1 = Series([np.nan, True])
+        s2 = Series([np.nan, False])
+        assert s1.all(skipna=False)  # nan && True => True
+        assert s1.all(skipna=True)
+        assert s2.any(skipna=False)
+        assert not s2.any(skipna=True)
+
+    def test_all_any_bool_only(self):
+        s = Series([False, False, True, True, False, True], index=[0, 0, 1, 1, 2, 2])
+
+        # GH#47500 - test bool_only works
+        assert s.any(bool_only=True)
+        assert not s.all(bool_only=True)
+
+    def test_any_all_object_dtype(self, all_boolean_reductions, skipna):
+        # GH#12863
+        ser = Series(["a", "b", "c", "d", "e"], dtype=object)
+        result = getattr(ser, all_boolean_reductions)(skipna=skipna)
+        expected = True
+
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "data", [[False, None], [None, False], [False, np.nan], [np.nan, False]]
+    )
+    def test_any_all_object_dtype_missing(self, data, all_boolean_reductions):
+        # GH#27709
+        ser = Series(data)
+        result = getattr(ser, all_boolean_reductions)(skipna=False)
+
+        # None is treated is False, but np.nan is treated as True
+        expected = all_boolean_reductions == "any" and None not in data
+        assert result == expected
+
+    @pytest.mark.parametrize("dtype", ["boolean", "Int64", "UInt64", "Float64"])
+    @pytest.mark.parametrize(
+        # expected_data indexed as [[skipna=False/any, skipna=False/all],
+        #                           [skipna=True/any, skipna=True/all]]
+        "data,expected_data",
+        [
+            ([0, 0, 0], [[False, False], [False, False]]),
+            ([1, 1, 1], [[True, True], [True, True]]),
+            ([pd.NA, pd.NA, pd.NA], [[pd.NA, pd.NA], [False, True]]),
+            ([0, pd.NA, 0], [[pd.NA, False], [False, False]]),
+            ([1, pd.NA, 1], [[True, pd.NA], [True, True]]),
+            ([1, pd.NA, 0], [[True, False], [True, False]]),
+        ],
+    )
+    def test_any_all_nullable_kleene_logic(
+        self, all_boolean_reductions, skipna, data, dtype, expected_data
+    ):
+        # GH-37506, GH-41967
+        ser = Series(data, dtype=dtype)
+        expected = expected_data[skipna][all_boolean_reductions == "all"]
+
+        result = getattr(ser, all_boolean_reductions)(skipna=skipna)
+        assert (result is pd.NA and expected is pd.NA) or result == expected
+
+    def test_any_axis1_bool_only(self):
+        # GH#32432
+        df = DataFrame({"A": [True, False], "B": [1, 2]})
+        result = df.any(axis=1, bool_only=True)
+        expected = Series([True, False])
+        tm.assert_series_equal(result, expected)
+
+    def test_any_all_datetimelike(self):
+        # GH#38723 these may not be the desired long-term behavior (GH#34479)
+        #  but in the interim should be internally consistent
+        dta = date_range("1995-01-02", periods=3)._data
+        ser = Series(dta)
+        df = DataFrame(ser)
+
+        # GH#34479
+        msg = "datetime64 type does not support operation '(any|all)'"
+        with pytest.raises(TypeError, match=msg):
+            dta.all()
+        with pytest.raises(TypeError, match=msg):
+            dta.any()
+
+        with pytest.raises(TypeError, match=msg):
+            ser.all()
+        with pytest.raises(TypeError, match=msg):
+            ser.any()
+
+        with pytest.raises(TypeError, match=msg):
+            df.any().all()
+        with pytest.raises(TypeError, match=msg):
+            df.all().all()
+
+        dta = dta.tz_localize("UTC")
+        ser = Series(dta)
+        df = DataFrame(ser)
+        # GH#34479
+        with pytest.raises(TypeError, match=msg):
+            dta.all()
+        with pytest.raises(TypeError, match=msg):
+            dta.any()
+
+        with pytest.raises(TypeError, match=msg):
+            ser.all()
+        with pytest.raises(TypeError, match=msg):
+            ser.any()
+
+        with pytest.raises(TypeError, match=msg):
+            df.any().all()
+        with pytest.raises(TypeError, match=msg):
+            df.all().all()
+
+        tda = dta - dta[0]
+        ser = Series(tda)
+        df = DataFrame(ser)
+
+        assert tda.any()
+        assert not tda.all()
+
+        assert ser.any()
+        assert not ser.all()
+
+        assert df.any().all()
+        assert not df.all().any()
+
+    def test_any_all_string_dtype(self, any_string_dtype):
+        # GH#54591
+        if (
+            isinstance(any_string_dtype, pd.StringDtype)
+            and any_string_dtype.na_value is pd.NA
+        ):
+            # the nullable string dtype currently still raise an error
+            # https://github.com/pandas-dev/pandas/issues/51939
+            ser = Series(["a", "b"], dtype=any_string_dtype)
+            with pytest.raises(TypeError):
+                ser.any()
+            with pytest.raises(TypeError):
+                ser.all()
+            return
+
+        ser = Series(["", "a"], dtype=any_string_dtype)
+        assert ser.any()
+        assert not ser.all()
+        assert ser.any(skipna=False)
+        assert not ser.all(skipna=False)
+
+        ser = Series([np.nan, "a"], dtype=any_string_dtype)
+        assert ser.any()
+        assert ser.all()
+        assert ser.any(skipna=False)
+        assert ser.all(skipna=False)  # NaN is considered truthy
+
+        ser = Series([np.nan, ""], dtype=any_string_dtype)
+        assert not ser.any()
+        assert not ser.all()
+        assert ser.any(skipna=False)  # NaN is considered truthy
+        assert not ser.all(skipna=False)
+
+        ser = Series(["a", "b"], dtype=any_string_dtype)
+        assert ser.any()
+        assert ser.all()
+        assert ser.any(skipna=False)
+        assert ser.all(skipna=False)
+
+        ser = Series([], dtype=any_string_dtype)
+        assert not ser.any()
+        assert ser.all()
+        assert not ser.any(skipna=False)
+        assert ser.all(skipna=False)
+
+        ser = Series([""], dtype=any_string_dtype)
+        assert not ser.any()
+        assert not ser.all()
+        assert not ser.any(skipna=False)
+        assert not ser.all(skipna=False)
+
+        ser = Series([np.nan], dtype=any_string_dtype)
+        assert not ser.any()
+        assert ser.all()
+        assert ser.any(skipna=False)  # NaN is considered truthy
+        assert ser.all(skipna=False)  # NaN is considered truthy
+
+    def test_timedelta64_analytics(self):
+        # index min/max
+        dti = date_range("2012-1-1", periods=3, freq="D")
+        td = Series(dti) - Timestamp("20120101")
+
+        result = td.idxmin()
+        assert result == 0
+
+        result = td.idxmax()
+        assert result == 2
+
+        # GH#2982
+        # with NaT
+        td[0] = np.nan
+
+        result = td.idxmin()
+        assert result == 1
+
+        result = td.idxmax()
+        assert result == 2
+
+        # abs
+        s1 = Series(date_range("20120101", periods=3))
+        s2 = Series(date_range("20120102", periods=3))
+        expected = Series(s2 - s1)
+
+        result = np.abs(s1 - s2)
+        tm.assert_series_equal(result, expected)
+
+        result = (s1 - s2).abs()
+        tm.assert_series_equal(result, expected)
+
+        # max/min
+        result = td.max()
+        expected = Timedelta("2 days")
+        assert result == expected
+
+        result = td.min()
+        expected = Timedelta("1 days")
+        assert result == expected
+
+    def test_assert_idxminmax_empty_raises(self):
+        """
+        Cases where ``Series.argmax`` and related should raise an exception
+        """
+        test_input = Series([], dtype="float64")
+        msg = "attempt to get argmin of an empty sequence"
+        with pytest.raises(ValueError, match=msg):
+            test_input.idxmin()
+        with pytest.raises(ValueError, match=msg):
+            test_input.idxmin(skipna=False)
+        msg = "attempt to get argmax of an empty sequence"
+        with pytest.raises(ValueError, match=msg):
+            test_input.idxmax()
+        with pytest.raises(ValueError, match=msg):
+            test_input.idxmax(skipna=False)
+
+    def test_idxminmax_object_dtype(self, using_infer_string):
+        # pre-2.1 object-dtype was disallowed for argmin/max
+        ser = Series(["foo", "bar", "baz"])
+        assert ser.idxmax() == 0
+        assert ser.idxmax(skipna=False) == 0
+        assert ser.idxmin() == 1
+        assert ser.idxmin(skipna=False) == 1
+
+        ser2 = Series([(1,), (2,)])
+        assert ser2.idxmax() == 1
+        assert ser2.idxmax(skipna=False) == 1
+        assert ser2.idxmin() == 0
+        assert ser2.idxmin(skipna=False) == 0
+
+        if not using_infer_string:
+            # attempting to compare np.nan with string raises
+            ser3 = Series(["foo", "foo", "bar", "bar", None, np.nan, "baz"])
+            msg = "'>' not supported between instances of 'float' and 'str'"
+            with pytest.raises(TypeError, match=msg):
+                ser3.idxmax()
+            with pytest.raises(TypeError, match=msg):
+                ser3.idxmax(skipna=False)
+            msg = "'<' not supported between instances of 'float' and 'str'"
+            with pytest.raises(TypeError, match=msg):
+                ser3.idxmin()
+            with pytest.raises(TypeError, match=msg):
+                ser3.idxmin(skipna=False)
+
+    def test_idxminmax_object_frame(self):
+        # GH#4279
+        df = DataFrame([["zimm", 2.5], ["biff", 1.0], ["bid", 12.0]])
+        res = df.idxmax()
+        exp = Series([0, 2])
+        tm.assert_series_equal(res, exp)
+
+    def test_idxminmax_object_tuples(self):
+        # GH#43697
+        ser = Series([(1, 3), (2, 2), (3, 1)])
+        assert ser.idxmax() == 2
+        assert ser.idxmin() == 0
+        assert ser.idxmax(skipna=False) == 2
+        assert ser.idxmin(skipna=False) == 0
+
+    def test_idxminmax_object_decimals(self):
+        # GH#40685
+        df = DataFrame(
+            {
+                "idx": [0, 1],
+                "x": [Decimal("8.68"), Decimal("42.23")],
+                "y": [Decimal("7.11"), Decimal("79.61")],
+            }
+        )
+        res = df.idxmax()
+        exp = Series({"idx": 1, "x": 1, "y": 1})
+        tm.assert_series_equal(res, exp)
+
+        res2 = df.idxmin()
+        exp2 = exp - 1
+        tm.assert_series_equal(res2, exp2)
+
+    def test_argminmax_object_ints(self):
+        # GH#18021
+        ser = Series([0, 1], dtype="object")
+        assert ser.argmax() == 1
+        assert ser.argmin() == 0
+        assert ser.argmax(skipna=False) == 1
+        assert ser.argmin(skipna=False) == 0
+
+    def test_idxminmax_with_inf(self):
+        # For numeric data with NA and Inf (GH #13595)
+        s = Series([0, -np.inf, np.inf, np.nan])
+
+        assert s.idxmin() == 1
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            s.idxmin(skipna=False)
+
+        assert s.idxmax() == 2
+        with pytest.raises(ValueError, match="Encountered an NA value"):
+            s.idxmax(skipna=False)
+
+    def test_sum_uint64(self):
+        # GH 53401
+        s = Series([10000000000000000000], dtype="uint64")
+        result = s.sum()
+        expected = np.uint64(10000000000000000000)
+        tm.assert_almost_equal(result, expected)
+
+    def test_signedness_preserved_after_sum(self):
+        # GH 37491
+        ser = Series([1, 2, 3, 4])
+
+        assert ser.astype("uint8").sum().dtype == "uint64"
+
+
+class TestDatetime64SeriesReductions:
+    # Note: the name TestDatetime64SeriesReductions indicates these tests
+    #  were moved from a series-specific test file, _not_ that these tests are
+    #  intended long-term to be series-specific
+
+    @pytest.mark.parametrize(
+        "nat_ser",
+        [
+            Series([NaT, NaT]),
+            Series([NaT, Timedelta("nat")]),
+            Series([Timedelta("nat"), Timedelta("nat")]),
+        ],
+    )
+    def test_minmax_nat_series(self, nat_ser):
+        # GH#23282
+        assert nat_ser.min() is NaT
+        assert nat_ser.max() is NaT
+        assert nat_ser.min(skipna=False) is NaT
+        assert nat_ser.max(skipna=False) is NaT
+
+    @pytest.mark.parametrize(
+        "nat_df",
+        [
+            [NaT, NaT],
+            [NaT, Timedelta("nat")],
+            [Timedelta("nat"), Timedelta("nat")],
+        ],
+    )
+    def test_minmax_nat_dataframe(self, nat_df):
+        # GH#23282
+        nat_df = DataFrame(nat_df)
+        assert nat_df.min()[0] is NaT
+        assert nat_df.max()[0] is NaT
+        assert nat_df.min(skipna=False)[0] is NaT
+        assert nat_df.max(skipna=False)[0] is NaT
+
+    def test_min_max(self):
+        rng = date_range("1/1/2000", "12/31/2000")
+        rng2 = rng.take(np.random.default_rng(2).permutation(len(rng)))
+
+        the_min = rng2.min()
+        the_max = rng2.max()
+        assert isinstance(the_min, Timestamp)
+        assert isinstance(the_max, Timestamp)
+        assert the_min == rng[0]
+        assert the_max == rng[-1]
+
+        assert rng.min() == rng[0]
+        assert rng.max() == rng[-1]
+
+    def test_min_max_series(self):
+        rng = date_range("1/1/2000", periods=10, freq="4h")
+        lvls = ["A", "A", "A", "B", "B", "B", "C", "C", "C", "C"]
+        df = DataFrame(
+            {
+                "TS": rng,
+                "V": np.random.default_rng(2).standard_normal(len(rng)),
+                "L": lvls,
+            }
+        )
+
+        result = df.TS.max()
+        exp = Timestamp(df.TS.iat[-1])
+        assert isinstance(result, Timestamp)
+        assert result == exp
+
+        result = df.TS.min()
+        exp = Timestamp(df.TS.iat[0])
+        assert isinstance(result, Timestamp)
+        assert result == exp
+
+
+class TestCategoricalSeriesReductions:
+    # Note: the name TestCategoricalSeriesReductions indicates these tests
+    #  were moved from a series-specific test file, _not_ that these tests are
+    #  intended long-term to be series-specific
+
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_unordered_raises(self, function):
+        # unordered cats have no min/max
+        cat = Series(Categorical(["a", "b", "c", "d"], ordered=False))
+        msg = f"Categorical is not ordered for operation {function}"
+        with pytest.raises(TypeError, match=msg):
+            getattr(cat, function)()
+
+    @pytest.mark.parametrize(
+        "values, categories",
+        [
+            (list("abc"), list("abc")),
+            (list("abc"), list("cba")),
+            (list("abc") + [np.nan], list("cba")),
+            ([1, 2, 3], [3, 2, 1]),
+            ([1, 2, 3, np.nan], [3, 2, 1]),
+        ],
+    )
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_ordered(self, values, categories, function):
+        # GH 25303
+        cat = Series(Categorical(values, categories=categories, ordered=True))
+        result = getattr(cat, function)(skipna=True)
+        expected = categories[0] if function == "min" else categories[2]
+        assert result == expected
+
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_ordered_with_nan_only(self, function, skipna):
+        # https://github.com/pandas-dev/pandas/issues/33450
+        cat = Series(Categorical([np.nan], categories=[1, 2], ordered=True))
+        result = getattr(cat, function)(skipna=skipna)
+        assert result is np.nan
+
+    @pytest.mark.parametrize("function", ["min", "max"])
+    def test_min_max_skipna(self, function, skipna):
+        cat = Series(
+            Categorical(["a", "b", np.nan, "a"], categories=["b", "a"], ordered=True)
+        )
+        result = getattr(cat, function)(skipna=skipna)
+
+        if skipna is True:
+            expected = "b" if function == "min" else "a"
+            assert result == expected
+        else:
+            assert result is np.nan
+
+
+class TestSeriesMode:
+    # Note: the name TestSeriesMode indicates these tests
+    #  were moved from a series-specific test file, _not_ that these tests are
+    #  intended long-term to be series-specific
+
+    def test_mode_empty(self, dropna):
+        s = Series([], dtype=np.float64)
+        result = s.mode(dropna)
+        tm.assert_series_equal(result, s)
+
+    @pytest.mark.parametrize(
+        "dropna, data, expected",
+        [
+            (True, [1, 1, 1, 2], [1]),
+            (True, [1, 1, 1, 2, 3, 3, 3], [1, 3]),
+            (False, [1, 1, 1, 2], [1]),
+            (False, [1, 1, 1, 2, 3, 3, 3], [1, 3]),
+        ],
+    )
+    def test_mode_numerical(self, dropna, data, expected, any_real_numpy_dtype):
+        s = Series(data, dtype=any_real_numpy_dtype)
+        result = s.mode(dropna)
+        expected = Series(expected, dtype=any_real_numpy_dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dropna, expected", [(True, [1.0]), (False, [1, np.nan])])
+    def test_mode_numerical_nan(self, dropna, expected):
+        s = Series([1, 1, 2, np.nan, np.nan])
+        result = s.mode(dropna)
+        expected = Series(expected)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [
+            (True, ["b"], ["bar"]),
+            (False, ["b"], [np.nan]),
+        ],
+    )
+    def test_mode_object(self, dropna, expected1, expected2):
+        # Test string and object types.
+        data = ["a"] * 2 + ["b"] * 3
+
+        s = Series(data, dtype="c")
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype="c")
+        tm.assert_series_equal(result, expected1)
+
+        data = ["foo", "bar", "bar", np.nan, np.nan, np.nan]
+
+        s = Series(data, dtype=object)
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype=object)
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [
+            (True, ["b"], ["bar"]),
+            (False, ["b"], [np.nan]),
+        ],
+    )
+    def test_mode_string(self, dropna, expected1, expected2, any_string_dtype):
+        # Test string and object types.
+        data = ["a"] * 2 + ["b"] * 3
+
+        s = Series(data, dtype=any_string_dtype)
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype=any_string_dtype)
+        tm.assert_series_equal(result, expected1)
+
+        data = ["foo", "bar", "bar", np.nan, np.nan, np.nan]
+
+        s = Series(data, dtype=any_string_dtype)
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype=any_string_dtype)
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [(True, ["foo"], ["foo"]), (False, ["foo"], [np.nan])],
+    )
+    def test_mode_mixeddtype(self, dropna, expected1, expected2):
+        s = Series([1, "foo", "foo"])
+        result = s.mode(dropna)
+        expected = Series(expected1, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        s = Series([1, "foo", "foo", np.nan, np.nan, np.nan])
+        result = s.mode(dropna)
+        expected = Series(expected2, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [
+            (
+                True,
+                ["1900-05-03", "2011-01-03", "2013-01-02"],
+                ["2011-01-03", "2013-01-02"],
+            ),
+            (False, [np.nan], [np.nan, "2011-01-03", "2013-01-02"]),
+        ],
+    )
+    def test_mode_datetime(self, dropna, expected1, expected2):
+        s = Series(
+            ["2011-01-03", "2013-01-02", "1900-05-03", "nan", "nan"], dtype="M8[ns]"
+        )
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype="M8[ns]")
+        tm.assert_series_equal(result, expected1)
+
+        s = Series(
+            [
+                "2011-01-03",
+                "2013-01-02",
+                "1900-05-03",
+                "2011-01-03",
+                "2013-01-02",
+                "nan",
+                "nan",
+            ],
+            dtype="M8[ns]",
+        )
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype="M8[ns]")
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [
+            (True, ["-1 days", "0 days", "1 days"], ["2 min", "1 day"]),
+            (False, [np.nan], [np.nan, "2 min", "1 day"]),
+        ],
+    )
+    def test_mode_timedelta(self, dropna, expected1, expected2):
+        # gh-5986: Test timedelta types.
+
+        s = Series(
+            ["1 days", "-1 days", "0 days", "nan", "nan"], dtype="timedelta64[ns]"
+        )
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype="timedelta64[ns]")
+        tm.assert_series_equal(result, expected1)
+
+        s = Series(
+            [
+                "1 day",
+                "1 day",
+                "-1 day",
+                "-1 day 2 min",
+                "2 min",
+                "2 min",
+                "nan",
+                "nan",
+            ],
+            dtype="timedelta64[ns]",
+        )
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype="timedelta64[ns]")
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2, expected3",
+        [
+            (
+                True,
+                Categorical([1, 2], categories=[1, 2]),
+                Categorical(["a"], categories=[1, "a"]),
+                Categorical([3, 1], categories=[3, 2, 1], ordered=True),
+            ),
+            (
+                False,
+                Categorical([np.nan], categories=[1, 2]),
+                Categorical([np.nan, "a"], categories=[1, "a"]),
+                Categorical([np.nan, 3, 1], categories=[3, 2, 1], ordered=True),
+            ),
+        ],
+    )
+    def test_mode_category(self, dropna, expected1, expected2, expected3):
+        s = Series(Categorical([1, 2, np.nan, np.nan]))
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype="category")
+        tm.assert_series_equal(result, expected1)
+
+        s = Series(Categorical([1, "a", "a", np.nan, np.nan]))
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype="category")
+        tm.assert_series_equal(result, expected2)
+
+        s = Series(
+            Categorical(
+                [1, 1, 2, 3, 3, np.nan, np.nan], categories=[3, 2, 1], ordered=True
+            )
+        )
+        result = s.mode(dropna)
+        expected3 = Series(expected3, dtype="category")
+        tm.assert_series_equal(result, expected3)
+
+    @pytest.mark.parametrize(
+        "dropna, expected1, expected2",
+        [(True, [2**63], [1, 2**63]), (False, [2**63], [1, 2**63])],
+    )
+    def test_mode_intoverflow(self, dropna, expected1, expected2):
+        # Test for uint64 overflow.
+        s = Series([1, 2**63, 2**63], dtype=np.uint64)
+        result = s.mode(dropna)
+        expected1 = Series(expected1, dtype=np.uint64)
+        tm.assert_series_equal(result, expected1)
+
+        s = Series([1, 2**63], dtype=np.uint64)
+        result = s.mode(dropna)
+        expected2 = Series(expected2, dtype=np.uint64)
+        tm.assert_series_equal(result, expected2)
+
+    def test_mode_sort_with_na(self):
+        s = Series([1, "foo", "foo", np.nan, np.nan])
+        expected = Series(["foo", np.nan], dtype=object)
+        result = s.mode(dropna=False)
+        tm.assert_series_equal(result, expected)
+
+    def test_mode_boolean_with_na(self):
+        # GH#42107
+        ser = Series([True, False, True, pd.NA], dtype="boolean")
+        result = ser.mode()
+        expected = Series([True], dtype="boolean")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "array,expected,dtype",
+        [
+            (
+                [0, 1j, 1, 1, 1 + 1j, 1 + 2j],
+                [1],
+                np.complex128,
+            ),
+            (
+                [0, 1j, 1, 1, 1 + 1j, 1 + 2j],
+                [1],
+                np.complex64,
+            ),
+            (
+                [1 + 1j, 2j, 1 + 1j],
+                [1 + 1j],
+                np.complex128,
+            ),
+        ],
+    )
+    def test_single_mode_value_complex(self, array, expected, dtype):
+        result = Series(array, dtype=dtype).mode()
+        expected = Series(expected, dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "array,expected,dtype",
+        [
+            (
+                # no modes
+                [0, 1j, 1, 1 + 1j, 1 + 2j],
+                [0j, 1j, 1 + 0j, 1 + 1j, 1 + 2j],
+                np.complex128,
+            ),
+            (
+                [1 + 1j, 2j, 1 + 1j, 2j, 3],
+                [2j, 1 + 1j],
+                np.complex64,
+            ),
+        ],
+    )
+    def test_multimode_complex(self, array, expected, dtype):
+        # GH 17927
+        # mode tries to sort multimodal series.
+        # Complex numbers are sorted by their magnitude
+        result = Series(array, dtype=dtype).mode()
+        expected = Series(expected, dtype=dtype)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/reductions/test_stat_reductions.py b/pandas/pandas/tests/reductions/test_stat_reductions.py
new file mode 100644
index 0000000000000..4af1ca1d4800a
--- /dev/null
+++ b/pandas/pandas/tests/reductions/test_stat_reductions.py
@@ -0,0 +1,274 @@
+"""
+Tests for statistical reductions of 2nd moment or higher: var, skew, kurt, ...
+"""
+
+import inspect
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeLikeStatReductions:
+    def test_dt64_mean(self, tz_naive_fixture, index_or_series_or_array):
+        tz = tz_naive_fixture
+
+        dti = date_range("2001-01-01", periods=11, tz=tz)
+        # shuffle so that we are not just working with monotone-increasing
+        dti = dti.take([4, 1, 3, 10, 9, 7, 8, 5, 0, 2, 6])
+        dtarr = dti._data
+
+        obj = index_or_series_or_array(dtarr)
+        assert obj.mean() == pd.Timestamp("2001-01-06", tz=tz)
+        assert obj.mean(skipna=False) == pd.Timestamp("2001-01-06", tz=tz)
+
+        # dtarr[-2] will be the first date 2001-01-1
+        dtarr[-2] = pd.NaT
+
+        obj = index_or_series_or_array(dtarr)
+        assert obj.mean() == pd.Timestamp("2001-01-06 07:12:00", tz=tz)
+        assert obj.mean(skipna=False) is pd.NaT
+
+    @pytest.mark.parametrize("freq", ["s", "h", "D", "W", "B"])
+    def test_period_mean(self, index_or_series_or_array, freq):
+        # GH#24757
+        dti = date_range("2001-01-01", periods=11)
+        # shuffle so that we are not just working with monotone-increasing
+        dti = dti.take([4, 1, 3, 10, 9, 7, 8, 5, 0, 2, 6])
+
+        warn = FutureWarning if freq == "B" else None
+        msg = r"PeriodDtype\[B\] is deprecated"
+        with tm.assert_produces_warning(warn, match=msg):
+            parr = dti._data.to_period(freq)
+        obj = index_or_series_or_array(parr)
+        with pytest.raises(TypeError, match="ambiguous"):
+            obj.mean()
+        with pytest.raises(TypeError, match="ambiguous"):
+            obj.mean(skipna=True)
+
+        # parr[-2] will be the first date 2001-01-1
+        parr[-2] = pd.NaT
+
+        with pytest.raises(TypeError, match="ambiguous"):
+            obj.mean()
+        with pytest.raises(TypeError, match="ambiguous"):
+            obj.mean(skipna=True)
+
+    def test_td64_mean(self, index_or_series_or_array):
+        m8values = np.array([0, 3, -2, -7, 1, 2, -1, 3, 5, -2, 4], "m8[D]")
+        tdi = pd.TimedeltaIndex(m8values).as_unit("ns")
+
+        tdarr = tdi._data
+        obj = index_or_series_or_array(tdarr, copy=False)
+
+        result = obj.mean()
+        expected = np.array(tdarr).mean()
+        assert result == expected
+
+        tdarr[0] = pd.NaT
+        assert obj.mean(skipna=False) is pd.NaT
+
+        result2 = obj.mean(skipna=True)
+        assert result2 == tdi[1:].mean()
+
+        # exact equality fails by 1 nanosecond
+        assert result2.round("us") == (result * 11.0 / 10).round("us")
+
+
+class TestSeriesStatReductions:
+    # Note: the name TestSeriesStatReductions indicates these tests
+    #  were moved from a series-specific test file, _not_ that these tests are
+    #  intended long-term to be series-specific
+
+    def _check_stat_op(
+        self, name, alternate, string_series_, check_objects=False, check_allna=False
+    ):
+        with pd.option_context("use_bottleneck", False):
+            f = getattr(Series, name)
+
+            # add some NaNs
+            string_series_[5:15] = np.nan
+
+            # mean, idxmax, idxmin, min, and max are valid for dates
+            if name not in ["max", "min", "mean", "median", "std"]:
+                ds = Series(date_range("1/1/2001", periods=10))
+                msg = f"does not support operation '{name}'"
+                with pytest.raises(TypeError, match=msg):
+                    f(ds)
+
+            # skipna or no
+            assert pd.notna(f(string_series_))
+            assert pd.isna(f(string_series_, skipna=False))
+
+            # check the result is correct
+            nona = string_series_.dropna()
+            tm.assert_almost_equal(f(nona), alternate(nona.values))
+            tm.assert_almost_equal(f(string_series_), alternate(nona.values))
+
+            allna = string_series_ * np.nan
+
+            if check_allna:
+                assert np.isnan(f(allna))
+
+            # dtype=object with None, it works!
+            s = Series([1, 2, 3, None, 5])
+            f(s)
+
+            # GH#2888
+            items = [0]
+            items.extend(range(2**40, 2**40 + 1000))
+            s = Series(items, dtype="int64")
+            tm.assert_almost_equal(float(f(s)), float(alternate(s.values)))
+
+            # check date range
+            if check_objects:
+                s = Series(pd.bdate_range("1/1/2000", periods=10))
+                res = f(s)
+                exp = alternate(s)
+                assert res == exp
+
+            # check on string data
+            if name not in ["sum", "min", "max"]:
+                with pytest.raises(TypeError, match=None):
+                    f(Series(list("abc")))
+
+            # Invalid axis.
+            msg = "No axis named 1 for object type Series"
+            with pytest.raises(ValueError, match=msg):
+                f(string_series_, axis=1)
+
+            if "numeric_only" in inspect.getfullargspec(f).args:
+                # only the index is string; dtype is float
+                f(string_series_, numeric_only=True)
+
+    def test_sum(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("sum", np.sum, string_series, check_allna=False)
+
+    def test_mean(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("mean", np.mean, string_series)
+
+    def test_median(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("median", np.median, string_series)
+
+        # test with integers, test failure
+        int_ts = Series(np.ones(10, dtype=int), index=range(10))
+        tm.assert_almost_equal(np.median(int_ts), int_ts.median())
+
+    def test_prod(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("prod", np.prod, string_series)
+
+    def test_min(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("min", np.min, string_series, check_objects=True)
+
+    def test_max(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        self._check_stat_op("max", np.max, string_series, check_objects=True)
+
+    def test_var_std(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        datetime_series = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+
+        alt = lambda x: np.std(x, ddof=1)
+        self._check_stat_op("std", alt, string_series)
+
+        alt = lambda x: np.var(x, ddof=1)
+        self._check_stat_op("var", alt, string_series)
+
+        result = datetime_series.std(ddof=4)
+        expected = np.std(datetime_series.values, ddof=4)
+        tm.assert_almost_equal(result, expected)
+
+        result = datetime_series.var(ddof=4)
+        expected = np.var(datetime_series.values, ddof=4)
+        tm.assert_almost_equal(result, expected)
+
+        # 1 - element series with ddof=1
+        s = datetime_series.iloc[[0]]
+        result = s.var(ddof=1)
+        assert pd.isna(result)
+
+        result = s.std(ddof=1)
+        assert pd.isna(result)
+
+    def test_sem(self):
+        string_series = Series(range(20), dtype=np.float64, name="series")
+        datetime_series = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+
+        alt = lambda x: np.std(x, ddof=1) / np.sqrt(len(x))
+        self._check_stat_op("sem", alt, string_series)
+
+        result = datetime_series.sem(ddof=4)
+        expected = np.std(datetime_series.values, ddof=4) / np.sqrt(
+            len(datetime_series.values)
+        )
+        tm.assert_almost_equal(result, expected)
+
+        # 1 - element series with ddof=1
+        s = datetime_series.iloc[[0]]
+        result = s.sem(ddof=1)
+        assert pd.isna(result)
+
+    def test_skew(self):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        string_series = Series(range(20), dtype=np.float64, name="series")
+
+        alt = lambda x: sp_stats.skew(x, bias=False)
+        self._check_stat_op("skew", alt, string_series)
+
+        # test corner cases, skew() returns NaN unless there's at least 3
+        # values
+        min_N = 3
+        for i in range(1, min_N + 1):
+            s = Series(np.ones(i))
+            df = DataFrame(np.ones((i, i)))
+            if i < min_N:
+                assert np.isnan(s.skew())
+                assert np.isnan(df.skew()).all()
+            else:
+                assert 0 == s.skew()
+                assert isinstance(s.skew(), np.float64)  # GH53482
+                assert (df.skew() == 0).all()
+
+    def test_kurt(self):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        string_series = Series(range(20), dtype=np.float64, name="series")
+
+        alt = lambda x: sp_stats.kurtosis(x, bias=False)
+        self._check_stat_op("kurt", alt, string_series)
+
+    def test_kurt_corner(self):
+        # test corner cases, kurt() returns NaN unless there's at least 4
+        # values
+        min_N = 4
+        for i in range(1, min_N + 1):
+            s = Series(np.ones(i))
+            df = DataFrame(np.ones((i, i)))
+            if i < min_N:
+                assert np.isnan(s.kurt())
+                assert np.isnan(df.kurt()).all()
+            else:
+                assert 0 == s.kurt()
+                assert isinstance(s.kurt(), np.float64)  # GH53482
+                assert (df.kurt() == 0).all()
diff --git a/pandas/pandas/tests/resample/__init__.py b/pandas/pandas/tests/resample/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/resample/conftest.py b/pandas/pandas/tests/resample/conftest.py
new file mode 100644
index 0000000000000..6c45ece5d8fb9
--- /dev/null
+++ b/pandas/pandas/tests/resample/conftest.py
@@ -0,0 +1,33 @@
+import pytest
+
+# The various methods we support
+downsample_methods = [
+    "min",
+    "max",
+    "first",
+    "last",
+    "sum",
+    "mean",
+    "sem",
+    "median",
+    "prod",
+    "var",
+    "std",
+    "ohlc",
+    "quantile",
+]
+upsample_methods = ["count", "size"]
+series_methods = ["nunique"]
+resample_methods = downsample_methods + upsample_methods + series_methods
+
+
+@pytest.fixture(params=downsample_methods)
+def downsample_method(request):
+    """Fixture for parametrization of Grouper downsample methods."""
+    return request.param
+
+
+@pytest.fixture(params=resample_methods)
+def resample_method(request):
+    """Fixture for parametrization of Grouper resample methods."""
+    return request.param
diff --git a/pandas/pandas/tests/resample/test_base.py b/pandas/pandas/tests/resample/test_base.py
new file mode 100644
index 0000000000000..b2d9f6c0e3eb0
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_base.py
@@ -0,0 +1,564 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_extension_array_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    NaT,
+    PeriodIndex,
+    Series,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+from pandas.core.groupby.groupby import DataError
+from pandas.core.groupby.grouper import Grouper
+from pandas.core.indexes.datetimes import date_range
+from pandas.core.indexes.period import period_range
+from pandas.core.indexes.timedeltas import timedelta_range
+from pandas.core.resample import _asfreq_compat
+
+
+@pytest.fixture(
+    params=[
+        "linear",
+        "time",
+        "index",
+        "values",
+        "nearest",
+        "zero",
+        "slinear",
+        "quadratic",
+        "cubic",
+        "barycentric",
+        "krogh",
+        "from_derivatives",
+        "piecewise_polynomial",
+        "pchip",
+        "akima",
+    ],
+)
+def all_1d_no_arg_interpolation_methods(request):
+    return request.param
+
+
+@pytest.mark.parametrize("freq", ["2D", "1h"])
+@pytest.mark.parametrize(
+    "index",
+    [
+        timedelta_range("1 day", "10 day", freq="D"),
+        date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+    ],
+)
+def test_asfreq(frame_or_series, index, freq):
+    obj = frame_or_series(range(len(index)), index=index)
+    idx_range = date_range if isinstance(index, DatetimeIndex) else timedelta_range
+
+    result = obj.resample(freq).asfreq()
+    new_index = idx_range(obj.index[0], obj.index[-1], freq=freq)
+    expected = obj.reindex(new_index)
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        timedelta_range("1 day", "10 day", freq="D"),
+        date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+    ],
+)
+def test_asfreq_fill_value(index):
+    # test for fill value during resampling, issue 3715
+
+    ser = Series(range(len(index)), index=index, name="a")
+    idx_range = date_range if isinstance(index, DatetimeIndex) else timedelta_range
+
+    result = ser.resample("1h").asfreq()
+    new_index = idx_range(ser.index[0], ser.index[-1], freq="1h")
+    expected = ser.reindex(new_index)
+    tm.assert_series_equal(result, expected)
+
+    # Explicit cast to float to avoid implicit cast when setting None
+    frame = ser.astype("float").to_frame("value")
+    frame.iloc[1] = None
+    result = frame.resample("1h").asfreq(fill_value=4.0)
+    new_index = idx_range(frame.index[0], frame.index[-1], freq="1h")
+    expected = frame.reindex(new_index, fill_value=4.0)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        timedelta_range("1 day", "10 day", freq="D"),
+        date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+        period_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+    ],
+)
+def test_resample_interpolate(index):
+    # GH#12925
+    df = DataFrame(range(len(index)), index=index)
+    warn = None
+    if isinstance(df.index, PeriodIndex):
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        result = df.resample("1min").asfreq().interpolate()
+        expected = df.resample("1min").interpolate()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_interpolate_regular_sampling_off_grid(
+    all_1d_no_arg_interpolation_methods,
+):
+    pytest.importorskip("scipy")
+    # GH#21351
+    index = date_range("2000-01-01 00:01:00", periods=5, freq="2h")
+    ser = Series(np.arange(5.0), index)
+
+    method = all_1d_no_arg_interpolation_methods
+    # Resample to 1 hour sampling and interpolate with the given method
+    ser_resampled = ser.resample("1h").interpolate(method)
+
+    # Check that none of the resampled values are NaN, except the first one
+    # which lies 1 minute before the first actual data point
+    assert np.isnan(ser_resampled.iloc[0])
+    assert not ser_resampled.iloc[1:].isna().any()
+
+    if method not in ["nearest", "zero"]:
+        # Check that the resampled values are close to the expected values
+        # except for methods with known inaccuracies
+        assert np.all(
+            np.isclose(ser_resampled.values[1:], np.arange(0.5, 4.5, 0.5), rtol=1.0e-1)
+        )
+
+
+def test_resample_interpolate_irregular_sampling(all_1d_no_arg_interpolation_methods):
+    pytest.importorskip("scipy")
+    # GH#21351
+    ser = Series(
+        np.linspace(0.0, 1.0, 5),
+        index=DatetimeIndex(
+            [
+                "2000-01-01 00:00:03",
+                "2000-01-01 00:00:22",
+                "2000-01-01 00:00:24",
+                "2000-01-01 00:00:31",
+                "2000-01-01 00:00:39",
+            ]
+        ),
+    )
+
+    # Resample to 5 second sampling and interpolate with the given method
+    ser_resampled = ser.resample("5s").interpolate(all_1d_no_arg_interpolation_methods)
+
+    # Check that none of the resampled values are NaN, except the first one
+    # which lies 3 seconds before the first actual data point
+    assert np.isnan(ser_resampled.iloc[0])
+    assert not ser_resampled.iloc[1:].isna().any()
+
+
+def test_raises_on_non_datetimelike_index():
+    # this is a non datetimelike index
+    xp = DataFrame()
+    msg = (
+        "Only valid with DatetimeIndex, TimedeltaIndex or PeriodIndex, "
+        "but got an instance of 'RangeIndex'"
+    )
+    with pytest.raises(TypeError, match=msg):
+        xp.resample("YE")
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        PeriodIndex([], freq="D", name="a"),
+        DatetimeIndex([], name="a"),
+        TimedeltaIndex([], name="a"),
+    ],
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+def test_resample_empty_series(freq, index, resample_method):
+    # GH12771 & GH12868
+
+    ser = Series(index=index, dtype=float)
+    if freq == "ME" and isinstance(ser.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.resample(freq)
+        return
+    elif freq == "ME" and isinstance(ser.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    warn = None
+    if isinstance(ser.index, PeriodIndex):
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = ser.resample(freq)
+    result = getattr(rs, resample_method)()
+
+    if resample_method == "ohlc":
+        expected = DataFrame(
+            [], index=ser.index[:0], columns=["open", "high", "low", "close"]
+        )
+        expected.index = _asfreq_compat(ser.index, freq)
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+    else:
+        expected = ser.copy()
+        expected.index = _asfreq_compat(ser.index, freq)
+        tm.assert_series_equal(result, expected, check_dtype=False)
+
+    tm.assert_index_equal(result.index, expected.index)
+    assert result.index.freq == expected.index.freq
+
+
+@pytest.mark.parametrize(
+    "freq",
+    [
+        pytest.param("ME", marks=pytest.mark.xfail(reason="Don't know why this fails")),
+        "D",
+        "h",
+    ],
+)
+def test_resample_nat_index_series(freq, resample_method):
+    # GH39227
+
+    ser = Series(range(5), index=PeriodIndex([NaT] * 5, freq=freq))
+
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        rs = ser.resample(freq)
+    result = getattr(rs, resample_method)()
+
+    if resample_method == "ohlc":
+        expected = DataFrame(
+            [], index=ser.index[:0], columns=["open", "high", "low", "close"]
+        )
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+    else:
+        expected = ser[:0].copy()
+        tm.assert_series_equal(result, expected, check_dtype=False)
+    tm.assert_index_equal(result.index, expected.index)
+    assert result.index.freq == expected.index.freq
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        PeriodIndex([], freq="D", name="a"),
+        DatetimeIndex([], name="a"),
+        TimedeltaIndex([], name="a"),
+    ],
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+@pytest.mark.parametrize("resample_method", ["count", "size"])
+def test_resample_count_empty_series(freq, index, resample_method):
+    # GH28427
+    ser = Series(index=index)
+    if freq == "ME" and isinstance(ser.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.resample(freq)
+        return
+    elif freq == "ME" and isinstance(ser.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    warn = None
+    if isinstance(ser.index, PeriodIndex):
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = ser.resample(freq)
+
+    result = getattr(rs, resample_method)()
+
+    index = _asfreq_compat(ser.index, freq)
+
+    expected = Series([], dtype="int64", index=index, name=ser.name)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index", [DatetimeIndex([]), TimedeltaIndex([]), PeriodIndex([], freq="D")]
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+def test_resample_empty_dataframe(index, freq, resample_method):
+    # GH13212
+    df = DataFrame(index=index)
+    # count retains dimensions too
+    if freq == "ME" and isinstance(df.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.resample(freq, group_keys=False)
+        return
+    elif freq == "ME" and isinstance(df.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    warn = None
+    if isinstance(df.index, PeriodIndex):
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = df.resample(freq, group_keys=False)
+    result = getattr(rs, resample_method)()
+    if resample_method == "ohlc":
+        # TODO: no tests with len(df.columns) > 0
+        mi = MultiIndex.from_product([df.columns, ["open", "high", "low", "close"]])
+        expected = DataFrame([], index=df.index[:0], columns=mi, dtype=np.float64)
+        expected.index = _asfreq_compat(df.index, freq)
+
+    elif resample_method != "size":
+        expected = df.copy()
+    else:
+        # GH14962
+        expected = Series([], dtype=np.int64)
+
+    expected.index = _asfreq_compat(df.index, freq)
+
+    tm.assert_index_equal(result.index, expected.index)
+    assert result.index.freq == expected.index.freq
+    tm.assert_almost_equal(result, expected)
+
+    # test size for GH13212 (currently stays as df)
+
+
+@pytest.mark.parametrize(
+    "index", [DatetimeIndex([]), TimedeltaIndex([]), PeriodIndex([], freq="D")]
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+def test_resample_count_empty_dataframe(freq, index):
+    # GH28427
+    empty_frame_dti = DataFrame(index=index, columns=Index(["a"], dtype=object))
+
+    if freq == "ME" and isinstance(empty_frame_dti.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            empty_frame_dti.resample(freq)
+        return
+    elif freq == "ME" and isinstance(empty_frame_dti.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    warn = None
+    if isinstance(empty_frame_dti.index, PeriodIndex):
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = empty_frame_dti.resample(freq)
+    result = rs.count()
+
+    index = _asfreq_compat(empty_frame_dti.index, freq)
+
+    expected = DataFrame(dtype="int64", index=index, columns=Index(["a"], dtype=object))
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index", [DatetimeIndex([]), TimedeltaIndex([]), PeriodIndex([], freq="D")]
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+def test_resample_size_empty_dataframe(freq, index):
+    # GH28427
+
+    empty_frame_dti = DataFrame(index=index, columns=Index(["a"], dtype=object))
+
+    if freq == "ME" and isinstance(empty_frame_dti.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            empty_frame_dti.resample(freq)
+        return
+    elif freq == "ME" and isinstance(empty_frame_dti.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    msg = "Resampling with a PeriodIndex"
+    warn = None
+    if isinstance(empty_frame_dti.index, PeriodIndex):
+        warn = FutureWarning
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = empty_frame_dti.resample(freq)
+    result = rs.size()
+
+    index = _asfreq_compat(empty_frame_dti.index, freq)
+
+    expected = Series([], dtype="int64", index=index)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        PeriodIndex([], freq="M", name="a"),
+        DatetimeIndex([], name="a"),
+        TimedeltaIndex([], name="a"),
+    ],
+)
+@pytest.mark.parametrize("dtype", [float, int, object, "datetime64[ns]"])
+@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
+def test_resample_empty_dtypes(index, dtype, resample_method):
+    # Empty series were sometimes causing a segfault (for the functions
+    # with Cython bounds-checking disabled) or an IndexError.  We just run
+    # them to ensure they no longer do.  (GH #10228)
+    warn = None
+    if isinstance(index, PeriodIndex):
+        # GH#53511
+        index = PeriodIndex([], freq="B", name=index.name)
+        warn = FutureWarning
+    msg = "Resampling with a PeriodIndex is deprecated"
+
+    empty_series_dti = Series([], index, dtype)
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = empty_series_dti.resample("D", group_keys=False)
+    try:
+        getattr(rs, resample_method)()
+    except DataError:
+        # Ignore these since some combinations are invalid
+        # (ex: doing mean with dtype of np.object_)
+        pass
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        PeriodIndex([], freq="D", name="a"),
+        DatetimeIndex([], name="a"),
+        TimedeltaIndex([], name="a"),
+    ],
+)
+@pytest.mark.parametrize("freq", ["ME", "D", "h"])
+def test_apply_to_empty_series(index, freq):
+    # GH 14313
+    ser = Series(index=index)
+
+    if freq == "ME" and isinstance(ser.index, TimedeltaIndex):
+        msg = (
+            "Resampling on a TimedeltaIndex requires fixed-duration `freq`, "
+            "e.g. '24h' or '3D', not <MonthEnd>"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.resample(freq)
+        return
+    elif freq == "ME" and isinstance(ser.index, PeriodIndex):
+        # index is PeriodIndex, so convert to corresponding Period freq
+        freq = "M"
+
+    msg = "Resampling with a PeriodIndex"
+    warn = None
+    if isinstance(ser.index, PeriodIndex):
+        warn = FutureWarning
+
+    with tm.assert_produces_warning(warn, match=msg):
+        rs = ser.resample(freq, group_keys=False)
+
+    result = rs.apply(lambda x: 1)
+    with tm.assert_produces_warning(warn, match=msg):
+        expected = ser.resample(freq).apply("sum")
+
+    tm.assert_series_equal(result, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        timedelta_range("1 day", "10 day", freq="D"),
+        date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+        period_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+    ],
+)
+def test_resampler_is_iterable(index):
+    # GH 15314
+    series = Series(range(len(index)), index=index)
+    freq = "h"
+    tg = Grouper(freq=freq, convention="start")
+    msg = "Resampling with a PeriodIndex"
+    warn = None
+    if isinstance(series.index, PeriodIndex):
+        warn = FutureWarning
+
+    with tm.assert_produces_warning(warn, match=msg):
+        grouped = series.groupby(tg)
+
+    with tm.assert_produces_warning(warn, match=msg):
+        resampled = series.resample(freq)
+    for (rk, rv), (gk, gv) in zip(resampled, grouped):
+        assert rk == gk
+        tm.assert_series_equal(rv, gv)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        timedelta_range("1 day", "10 day", freq="D"),
+        date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+        period_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D"),
+    ],
+)
+def test_resample_quantile(index):
+    # GH 15023
+    ser = Series(range(len(index)), index=index)
+    q = 0.75
+    freq = "h"
+
+    msg = "Resampling with a PeriodIndex"
+    warn = None
+    if isinstance(ser.index, PeriodIndex):
+        warn = FutureWarning
+    with tm.assert_produces_warning(warn, match=msg):
+        result = ser.resample(freq).quantile(q)
+        expected = ser.resample(freq).agg(lambda x: x.quantile(q)).rename(ser.name)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["first", "last"])
+def test_first_last_skipna(any_real_nullable_dtype, skipna, how):
+    # GH#57019
+    if is_extension_array_dtype(any_real_nullable_dtype):
+        na_value = Series(dtype=any_real_nullable_dtype).dtype.na_value
+    else:
+        na_value = np.nan
+    df = DataFrame(
+        {
+            "a": [2, 1, 1, 2],
+            "b": [na_value, 3.0, na_value, 4.0],
+            "c": [na_value, 3.0, na_value, 4.0],
+        },
+        index=date_range("2020-01-01", periods=4, freq="D"),
+        dtype=any_real_nullable_dtype,
+    )
+    rs = df.resample("ME")
+    method = getattr(rs, how)
+    result = method(skipna=skipna)
+
+    ts = pd.to_datetime("2020-01-31").as_unit("ns")
+    gb = df.groupby(df.shape[0] * [ts])
+    expected = getattr(gb, how)(skipna=skipna)
+    expected.index.freq = "ME"
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/resample/test_datetime_index.py b/pandas/pandas/tests/resample/test_datetime_index.py
new file mode 100644
index 0000000000000..3a7fd548ca961
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_datetime_index.py
@@ -0,0 +1,2164 @@
+from datetime import datetime
+from functools import partial
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._typing import DatetimeNaTType
+from pandas.compat import is_platform_windows
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timedelta,
+    Timestamp,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+from pandas.core.groupby.grouper import Grouper
+from pandas.core.indexes.datetimes import date_range
+from pandas.core.indexes.period import (
+    Period,
+    period_range,
+)
+from pandas.core.resample import (
+    DatetimeIndex,
+    _get_timestamp_range_edges,
+)
+
+from pandas.tseries import offsets
+from pandas.tseries.offsets import Minute
+
+
+@pytest.fixture
+def simple_date_range_series():
+    """
+    Series with date range index and random data for test purposes.
+    """
+
+    def _simple_date_range_series(start, end, freq="D"):
+        rng = date_range(start, end, freq=freq)
+        return Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    return _simple_date_range_series
+
+
+def test_custom_grouper(unit):
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="Min")
+    dti = index.as_unit(unit)
+    s = Series(np.array([1] * len(dti)), index=dti, dtype="int64")
+
+    b = Grouper(freq=Minute(5))
+    g = s.groupby(b)
+
+    # check all cython functions work
+    g.ohlc()  # doesn't use _cython_agg_general
+    funcs = ["sum", "mean", "prod", "min", "max", "var"]
+    for f in funcs:
+        g._cython_agg_general(f, alt=None, numeric_only=True)
+
+    b = Grouper(freq=Minute(5), closed="right", label="right")
+    g = s.groupby(b)
+    # check all cython functions work
+    g.ohlc()  # doesn't use _cython_agg_general
+    funcs = ["sum", "mean", "prod", "min", "max", "var"]
+    for f in funcs:
+        g._cython_agg_general(f, alt=None, numeric_only=True)
+
+    assert g.ngroups == 2593
+    assert notna(g.mean()).all()
+
+    # construct expected val
+    arr = [1] + [5] * 2592
+    idx = dti[0:-1:5]
+    idx = idx.append(dti[-1:])
+    idx = DatetimeIndex(idx, freq="5min").as_unit(unit)
+    expect = Series(arr, index=idx)
+
+    # GH2763 - return input dtype if we can
+    result = g.agg("sum")
+    tm.assert_series_equal(result, expect)
+
+
+def test_custom_grouper_df(unit):
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+    b = Grouper(freq=Minute(5), closed="right", label="right")
+    dti = index.as_unit(unit)
+    df = DataFrame(
+        np.random.default_rng(2).random((len(dti), 10)), index=dti, dtype="float64"
+    )
+    r = df.groupby(b).agg("sum")
+
+    assert len(r.columns) == 10
+    assert len(r.index) == 2593
+
+
+@pytest.mark.parametrize(
+    "closed, expected",
+    [
+        (
+            "right",
+            lambda s: Series(
+                [s.iloc[0], s[1:6].mean(), s[6:11].mean(), s[11:].mean()],
+                index=date_range("1/1/2000", periods=4, freq="5min", name="index"),
+            ),
+        ),
+        (
+            "left",
+            lambda s: Series(
+                [s[:5].mean(), s[5:10].mean(), s[10:].mean()],
+                index=date_range(
+                    "1/1/2000 00:05", periods=3, freq="5min", name="index"
+                ),
+            ),
+        ),
+    ],
+)
+def test_resample_basic(closed, expected, unit):
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    s = Series(range(len(index)), index=index)
+    s.index.name = "index"
+    s.index = s.index.as_unit(unit)
+    expected = expected(s)
+    expected.index = expected.index.as_unit(unit)
+    result = s.resample("5min", closed=closed, label="right").mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_integerarray(unit):
+    # GH 25580, resample on IntegerArray
+    ts = Series(
+        range(9),
+        index=date_range("1/1/2000", periods=9, freq="min").as_unit(unit),
+        dtype="Int64",
+    )
+    result = ts.resample("3min").sum()
+    expected = Series(
+        [3, 12, 21],
+        index=date_range("1/1/2000", periods=3, freq="3min").as_unit(unit),
+        dtype="Int64",
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ts.resample("3min").mean()
+    expected = Series(
+        [1, 4, 7],
+        index=date_range("1/1/2000", periods=3, freq="3min").as_unit(unit),
+        dtype="Float64",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_basic_grouper(unit):
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    s = Series(range(len(index)), index=index)
+    s.index.name = "index"
+    s.index = s.index.as_unit(unit)
+    result = s.resample("5Min").last()
+    grouper = Grouper(freq=Minute(5), closed="left", label="left")
+    expected = s.groupby(grouper).agg(lambda x: x.iloc[-1])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:The 'convention' keyword in Series.resample:FutureWarning"
+)
+@pytest.mark.parametrize(
+    "keyword,value",
+    [("label", "righttt"), ("closed", "righttt"), ("convention", "starttt")],
+)
+def test_resample_string_kwargs(keyword, value, unit):
+    # see gh-19303
+    # Check that wrong keyword argument strings raise an error
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    series = Series(range(len(index)), index=index)
+    series.index.name = "index"
+    series.index = series.index.as_unit(unit)
+    msg = f"Unsupported value {value} for `{keyword}`"
+    with pytest.raises(ValueError, match=msg):
+        series.resample("5min", **({keyword: value}))
+
+
+def test_resample_how(downsample_method, unit):
+    if downsample_method == "ohlc":
+        pytest.skip("covered by test_resample_how_ohlc")
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    s = Series(range(len(index)), index=index)
+    s.index.name = "index"
+    s.index = s.index.as_unit(unit)
+    grouplist = np.ones_like(s)
+    grouplist[0] = 0
+    grouplist[1:6] = 1
+    grouplist[6:11] = 2
+    grouplist[11:] = 3
+    expected = s.groupby(grouplist).agg(downsample_method)
+    expected.index = date_range(
+        "1/1/2000", periods=4, freq="5min", name="index"
+    ).as_unit(unit)
+
+    result = getattr(
+        s.resample("5min", closed="right", label="right"), downsample_method
+    )()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_how_ohlc(unit):
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    s = Series(range(len(index)), index=index)
+    s.index.name = "index"
+    s.index = s.index.as_unit(unit)
+    grouplist = np.ones_like(s)
+    grouplist[0] = 0
+    grouplist[1:6] = 1
+    grouplist[6:11] = 2
+    grouplist[11:] = 3
+
+    def _ohlc(group):
+        if isna(group).all():
+            return np.repeat(np.nan, 4)
+        return [group.iloc[0], group.max(), group.min(), group.iloc[-1]]
+
+    expected = DataFrame(
+        s.groupby(grouplist).agg(_ohlc).values.tolist(),
+        index=date_range("1/1/2000", periods=4, freq="5min", name="index").as_unit(
+            unit
+        ),
+        columns=["open", "high", "low", "close"],
+    )
+
+    result = s.resample("5min", closed="right", label="right").ohlc()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_how_callables(unit):
+    # GH#7929
+    data = np.arange(5, dtype=np.int64)
+    msg = "'d' is deprecated and will be removed in a future version."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        ind = date_range(start="2014-01-01", periods=len(data), freq="d").as_unit(unit)
+    df = DataFrame({"A": data, "B": data}, index=ind)
+
+    def fn(x, a=1):
+        return str(type(x))
+
+    class FnClass:
+        def __call__(self, x):
+            return str(type(x))
+
+    df_standard = df.resample("ME").apply(fn)
+    df_lambda = df.resample("ME").apply(lambda x: str(type(x)))
+    df_partial = df.resample("ME").apply(partial(fn))
+    df_partial2 = df.resample("ME").apply(partial(fn, a=2))
+    df_class = df.resample("ME").apply(FnClass())
+
+    tm.assert_frame_equal(df_standard, df_lambda)
+    tm.assert_frame_equal(df_standard, df_partial)
+    tm.assert_frame_equal(df_standard, df_partial2)
+    tm.assert_frame_equal(df_standard, df_class)
+
+
+def test_resample_rounding(unit):
+    # GH 8371
+    # odd results when rounding is needed
+
+    ts = [
+        "2014-11-08 00:00:01",
+        "2014-11-08 00:00:02",
+        "2014-11-08 00:00:02",
+        "2014-11-08 00:00:03",
+        "2014-11-08 00:00:07",
+        "2014-11-08 00:00:07",
+        "2014-11-08 00:00:08",
+        "2014-11-08 00:00:08",
+        "2014-11-08 00:00:08",
+        "2014-11-08 00:00:09",
+        "2014-11-08 00:00:10",
+        "2014-11-08 00:00:11",
+        "2014-11-08 00:00:11",
+        "2014-11-08 00:00:13",
+        "2014-11-08 00:00:14",
+        "2014-11-08 00:00:15",
+        "2014-11-08 00:00:17",
+        "2014-11-08 00:00:20",
+        "2014-11-08 00:00:21",
+    ]
+    df = DataFrame({"value": [1] * 19}, index=pd.to_datetime(ts))
+    df.index = df.index.as_unit(unit)
+
+    result = df.resample("6s").sum()
+    expected = DataFrame(
+        {"value": [4, 9, 4, 2]},
+        index=date_range("2014-11-08", freq="6s", periods=4).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("7s").sum()
+    expected = DataFrame(
+        {"value": [4, 10, 4, 1]},
+        index=date_range("2014-11-08", freq="7s", periods=4).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("11s").sum()
+    expected = DataFrame(
+        {"value": [11, 8]},
+        index=date_range("2014-11-08", freq="11s", periods=2).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("13s").sum()
+    expected = DataFrame(
+        {"value": [13, 6]},
+        index=date_range("2014-11-08", freq="13s", periods=2).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("17s").sum()
+    expected = DataFrame(
+        {"value": [16, 3]},
+        index=date_range("2014-11-08", freq="17s", periods=2).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_basic_from_daily(unit):
+    # from daily
+    dti = date_range(
+        start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="D", name="index"
+    ).as_unit(unit)
+
+    s = Series(np.random.default_rng(2).random(len(dti)), dti)
+
+    # to weekly
+    msg = "'w-sun' is deprecated and will be removed in a future version."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = s.resample("w-sun").last()
+
+    assert len(result) == 3
+    assert (result.index.dayofweek == [6, 6, 6]).all()
+    assert result.iloc[0] == s["1/2/2005"]
+    assert result.iloc[1] == s["1/9/2005"]
+    assert result.iloc[2] == s.iloc[-1]
+
+    result = s.resample("W-MON").last()
+    assert len(result) == 2
+    assert (result.index.dayofweek == [0, 0]).all()
+    assert result.iloc[0] == s["1/3/2005"]
+    assert result.iloc[1] == s["1/10/2005"]
+
+    result = s.resample("W-TUE").last()
+    assert len(result) == 2
+    assert (result.index.dayofweek == [1, 1]).all()
+    assert result.iloc[0] == s["1/4/2005"]
+    assert result.iloc[1] == s["1/10/2005"]
+
+    result = s.resample("W-WED").last()
+    assert len(result) == 2
+    assert (result.index.dayofweek == [2, 2]).all()
+    assert result.iloc[0] == s["1/5/2005"]
+    assert result.iloc[1] == s["1/10/2005"]
+
+    result = s.resample("W-THU").last()
+    assert len(result) == 2
+    assert (result.index.dayofweek == [3, 3]).all()
+    assert result.iloc[0] == s["1/6/2005"]
+    assert result.iloc[1] == s["1/10/2005"]
+
+    result = s.resample("W-FRI").last()
+    assert len(result) == 2
+    assert (result.index.dayofweek == [4, 4]).all()
+    assert result.iloc[0] == s["1/7/2005"]
+    assert result.iloc[1] == s["1/10/2005"]
+
+    # to biz day
+    result = s.resample("B").last()
+    assert len(result) == 7
+    assert (result.index.dayofweek == [4, 0, 1, 2, 3, 4, 0]).all()
+
+    assert result.iloc[0] == s["1/2/2005"]
+    assert result.iloc[1] == s["1/3/2005"]
+    assert result.iloc[5] == s["1/9/2005"]
+    assert result.index.name == "index"
+
+
+def test_resample_upsampling_picked_but_not_correct(unit):
+    # Test for issue #3020
+    dates = date_range("01-Jan-2014", "05-Jan-2014", freq="D").as_unit(unit)
+    series = Series(1, index=dates)
+
+    result = series.resample("D").mean()
+    assert result.index[0] == dates[0]
+
+    # GH 5955
+    # incorrect deciding to upsample when the axis frequency matches the
+    # resample frequency
+
+    s = Series(
+        np.arange(1.0, 6), index=[datetime(1975, 1, i, 12, 0) for i in range(1, 6)]
+    )
+    s.index = s.index.as_unit(unit)
+    expected = Series(
+        np.arange(1.0, 6),
+        index=date_range("19750101", periods=5, freq="D").as_unit(unit),
+    )
+
+    result = s.resample("D").count()
+    tm.assert_series_equal(result, Series(1, index=expected.index))
+
+    result1 = s.resample("D").sum()
+    result2 = s.resample("D").mean()
+    tm.assert_series_equal(result1, expected)
+    tm.assert_series_equal(result2, expected)
+
+
+@pytest.mark.parametrize("f", ["sum", "mean", "prod", "min", "max", "var"])
+def test_resample_frame_basic_cy_funcs(f, unit):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((50, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=50, freq="B"),
+    )
+    df.index = df.index.as_unit(unit)
+
+    b = Grouper(freq="ME")
+    g = df.groupby(b)
+
+    # check all cython functions work
+    g._cython_agg_general(f, alt=None, numeric_only=True)
+
+
+@pytest.mark.parametrize("freq", ["YE", "ME"])
+def test_resample_frame_basic_M_A(freq, unit):
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((50, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=50, freq="B"),
+    )
+    df.index = df.index.as_unit(unit)
+    result = df.resample(freq).mean()
+    tm.assert_series_equal(result["A"], df["A"].resample(freq).mean())
+
+
+def test_resample_upsample(unit):
+    # from daily
+    dti = date_range(
+        start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="D", name="index"
+    ).as_unit(unit)
+
+    s = Series(np.random.default_rng(2).random(len(dti)), dti)
+
+    # to minutely, by padding
+    result = s.resample("Min").ffill()
+    assert len(result) == 12961
+    assert result.iloc[0] == s.iloc[0]
+    assert result.iloc[-1] == s.iloc[-1]
+
+    assert result.index.name == "index"
+
+
+def test_resample_how_method(unit):
+    # GH9915
+    s = Series(
+        [11, 22],
+        index=[
+            Timestamp("2015-03-31 21:48:52.672000"),
+            Timestamp("2015-03-31 21:49:52.739000"),
+        ],
+    )
+    s.index = s.index.as_unit(unit)
+    expected = Series(
+        [11, np.nan, np.nan, np.nan, np.nan, np.nan, 22],
+        index=DatetimeIndex(
+            [
+                Timestamp("2015-03-31 21:48:50"),
+                Timestamp("2015-03-31 21:49:00"),
+                Timestamp("2015-03-31 21:49:10"),
+                Timestamp("2015-03-31 21:49:20"),
+                Timestamp("2015-03-31 21:49:30"),
+                Timestamp("2015-03-31 21:49:40"),
+                Timestamp("2015-03-31 21:49:50"),
+            ],
+            freq="10s",
+        ),
+    )
+    expected.index = expected.index.as_unit(unit)
+    tm.assert_series_equal(s.resample("10s").mean(), expected)
+
+
+def test_resample_extra_index_point(unit):
+    # GH#9756
+    index = date_range(start="20150101", end="20150331", freq="BME").as_unit(unit)
+    expected = DataFrame({"A": Series([21, 41, 63], index=index)})
+
+    index = date_range(start="20150101", end="20150331", freq="B").as_unit(unit)
+    df = DataFrame({"A": Series(range(len(index)), index=index)}, dtype="int64")
+    result = df.resample("BME").last()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_upsample_with_limit(unit):
+    rng = date_range("1/1/2000", periods=3, freq="5min").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+
+    result = ts.resample("min").ffill(limit=2)
+    expected = ts.reindex(result.index, method="ffill", limit=2)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("freq", ["1D", "10h", "5Min", "10s"])
+@pytest.mark.parametrize("rule", ["YE", "3ME", "15D", "30h", "15Min", "30s"])
+def test_nearest_upsample_with_limit(tz_aware_fixture, freq, rule, unit):
+    # GH 33939
+    rng = date_range("1/1/2000", periods=3, freq=freq, tz=tz_aware_fixture).as_unit(
+        unit
+    )
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+
+    result = ts.resample(rule).nearest(limit=2)
+    expected = ts.reindex(result.index, method="nearest", limit=2)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_ohlc(unit):
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="Min")
+    s = Series(range(len(index)), index=index)
+    s.index.name = "index"
+    s.index = s.index.as_unit(unit)
+
+    grouper = Grouper(freq=Minute(5))
+    expect = s.groupby(grouper).agg(lambda x: x.iloc[-1])
+    result = s.resample("5Min").ohlc()
+
+    assert len(result) == len(expect)
+    assert len(result.columns) == 4
+
+    xs = result.iloc[-2]
+    assert xs["open"] == s.iloc[-6]
+    assert xs["high"] == s[-6:-1].max()
+    assert xs["low"] == s[-6:-1].min()
+    assert xs["close"] == s.iloc[-2]
+
+    xs = result.iloc[0]
+    assert xs["open"] == s.iloc[0]
+    assert xs["high"] == s[:5].max()
+    assert xs["low"] == s[:5].min()
+    assert xs["close"] == s.iloc[4]
+
+
+def test_resample_ohlc_result(unit):
+    # GH 12332
+    index = date_range("1-1-2000", "2-15-2000", freq="h").as_unit(unit)
+    index = index.union(date_range("4-15-2000", "5-15-2000", freq="h").as_unit(unit))
+    s = Series(range(len(index)), index=index)
+
+    a = s.loc[:"4-15-2000"].resample("30min").ohlc()
+    assert isinstance(a, DataFrame)
+
+    b = s.loc[:"4-14-2000"].resample("30min").ohlc()
+    assert isinstance(b, DataFrame)
+
+
+def test_resample_ohlc_result_odd_period(unit):
+    # GH12348
+    # raising on odd period
+    rng = date_range("2013-12-30", "2014-01-07").as_unit(unit)
+    index = rng.drop(
+        [
+            Timestamp("2014-01-01"),
+            Timestamp("2013-12-31"),
+            Timestamp("2014-01-04"),
+            Timestamp("2014-01-05"),
+        ]
+    )
+    df = DataFrame(data=np.arange(len(index)), index=index)
+    result = df.resample("B").mean()
+    expected = df.reindex(index=date_range(rng[0], rng[-1], freq="B").as_unit(unit))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_ohlc_dataframe(unit):
+    df = (
+        DataFrame(
+            {
+                "PRICE": {
+                    Timestamp("2011-01-06 10:59:05", tz=None): 24990,
+                    Timestamp("2011-01-06 12:43:33", tz=None): 25499,
+                    Timestamp("2011-01-06 12:54:09", tz=None): 25499,
+                },
+                "VOLUME": {
+                    Timestamp("2011-01-06 10:59:05", tz=None): 1500000000,
+                    Timestamp("2011-01-06 12:43:33", tz=None): 5000000000,
+                    Timestamp("2011-01-06 12:54:09", tz=None): 100000000,
+                },
+            }
+        )
+    ).reindex(["VOLUME", "PRICE"], axis=1)
+    df.index = df.index.as_unit(unit)
+    df.columns.name = "Cols"
+    res = df.resample("h").ohlc()
+    exp = pd.concat(
+        [df["VOLUME"].resample("h").ohlc(), df["PRICE"].resample("h").ohlc()],
+        axis=1,
+        keys=df.columns,
+    )
+    assert exp.columns.names[0] == "Cols"
+    tm.assert_frame_equal(exp, res)
+
+    df.columns = [["a", "b"], ["c", "d"]]
+    res = df.resample("h").ohlc()
+    exp.columns = pd.MultiIndex.from_tuples(
+        [
+            ("a", "c", "open"),
+            ("a", "c", "high"),
+            ("a", "c", "low"),
+            ("a", "c", "close"),
+            ("b", "d", "open"),
+            ("b", "d", "high"),
+            ("b", "d", "low"),
+            ("b", "d", "close"),
+        ]
+    )
+    tm.assert_frame_equal(exp, res)
+
+    # dupe columns fail atm
+    # df.columns = ['PRICE', 'PRICE']
+
+
+def test_resample_reresample(unit):
+    dti = date_range(
+        start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="D"
+    ).as_unit(unit)
+    s = Series(np.random.default_rng(2).random(len(dti)), dti)
+    bs = s.resample("B", closed="right", label="right").mean()
+    result = bs.resample("8h").mean()
+    assert len(result) == 25
+    assert isinstance(result.index.freq, offsets.DateOffset)
+    assert result.index.freq == offsets.Hour(8)
+
+
+@pytest.mark.parametrize(
+    "freq, expected_kwargs",
+    [
+        ["YE-DEC", {"start": "1990", "end": "2000", "freq": "Y-DEC"}],
+        ["YE-JUN", {"start": "1990", "end": "2000", "freq": "Y-JUN"}],
+        ["ME", {"start": "1990-01", "end": "2000-01", "freq": "M"}],
+    ],
+)
+def test_resample_timestamp_to_period(
+    simple_date_range_series, freq, expected_kwargs, unit
+):
+    ts = simple_date_range_series("1/1/1990", "1/1/2000")
+    ts.index = ts.index.as_unit(unit)
+
+    result = ts.resample(freq).mean().to_period()
+    expected = ts.resample(freq).mean()
+    expected.index = period_range(**expected_kwargs)
+    tm.assert_series_equal(result, expected)
+
+
+def test_ohlc_5min(unit):
+    def _ohlc(group):
+        if isna(group).all():
+            return np.repeat(np.nan, 4)
+        return [group.iloc[0], group.max(), group.min(), group.iloc[-1]]
+
+    rng = date_range("1/1/2000 00:00:00", "1/1/2000 5:59:50", freq="10s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    resampled = ts.resample("5min", closed="right", label="right").ohlc()
+
+    assert (resampled.loc["1/1/2000 00:00"] == ts.iloc[0]).all()
+
+    exp = _ohlc(ts[1:31])
+    assert (resampled.loc["1/1/2000 00:05"] == exp).all()
+
+    exp = _ohlc(ts["1/1/2000 5:55:01":])
+    assert (resampled.loc["1/1/2000 6:00:00"] == exp).all()
+
+
+def test_downsample_non_unique(unit):
+    rng = date_range("1/1/2000", "2/29/2000").as_unit(unit)
+    rng2 = rng.repeat(5).values
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng2)), index=rng2)
+
+    result = ts.resample("ME").mean()
+
+    expected = ts.groupby(lambda x: x.month).mean()
+    assert len(result) == 2
+    tm.assert_almost_equal(result.iloc[0], expected[1])
+    tm.assert_almost_equal(result.iloc[1], expected[2])
+
+
+def test_asfreq_non_unique(unit):
+    # GH #1077
+    rng = date_range("1/1/2000", "2/29/2000").as_unit(unit)
+    rng2 = rng.repeat(2).values
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng2)), index=rng2)
+
+    msg = "cannot reindex on an axis with duplicate labels"
+    with pytest.raises(ValueError, match=msg):
+        ts.asfreq("B")
+
+
+@pytest.mark.parametrize("freq", ["min", "5min", "15min", "30min", "4h", "12h"])
+def test_resample_anchored_ticks(freq, unit):
+    # If a fixed delta (5 minute, 4 hour) evenly divides a day, we should
+    # "anchor" the origin at midnight so we get regular intervals rather
+    # than starting from the first timestamp which might start in the
+    # middle of a desired interval
+
+    rng = date_range("1/1/2000 04:00:00", periods=86400, freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    ts[:2] = np.nan  # so results are the same
+    result = ts[2:].resample(freq, closed="left", label="left").mean()
+    expected = ts.resample(freq, closed="left", label="left").mean()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("end", [1, 2])
+def test_resample_single_group(end, unit):
+    mysum = lambda x: x.sum()
+
+    rng = date_range("2000-1-1", f"2000-{end}-10", freq="D").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    tm.assert_series_equal(ts.resample("ME").sum(), ts.resample("ME").apply(mysum))
+
+
+def test_resample_single_group_std(unit):
+    # GH 3849
+    s = Series(
+        [30.1, 31.6],
+        index=[Timestamp("20070915 15:30:00"), Timestamp("20070915 15:40:00")],
+    )
+    s.index = s.index.as_unit(unit)
+    expected = Series(
+        [0.75], index=DatetimeIndex([Timestamp("20070915")], freq="D").as_unit(unit)
+    )
+    result = s.resample("D").apply(lambda x: np.std(x))
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_offset(unit):
+    # GH 31809
+
+    rng = date_range("1/1/2000 00:00:00", "1/1/2000 02:00", freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    resampled = ts.resample("5min", offset="2min").mean()
+    exp_rng = date_range("12/31/1999 23:57:00", "1/1/2000 01:57", freq="5min").as_unit(
+        unit
+    )
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"origin": "1999-12-31 23:57:00"},
+        {"origin": Timestamp("1970-01-01 00:02:00")},
+        {"origin": "epoch", "offset": "2m"},
+        # origin of '1999-31-12 12:02:00' should be equivalent for this case
+        {"origin": "1999-12-31 12:02:00"},
+        {"offset": "-3m"},
+    ],
+)
+def test_resample_origin(kwargs, unit):
+    # GH 31809
+    rng = date_range("2000-01-01 00:00:00", "2000-01-01 02:00", freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    exp_rng = date_range(
+        "1999-12-31 23:57:00", "2000-01-01 01:57", freq="5min"
+    ).as_unit(unit)
+
+    resampled = ts.resample("5min", **kwargs).mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+
+@pytest.mark.parametrize(
+    "origin", ["invalid_value", "epch", "startday", "startt", "2000-30-30", object()]
+)
+def test_resample_bad_origin(origin, unit):
+    rng = date_range("2000-01-01 00:00:00", "2000-01-01 02:00", freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    msg = (
+        "'origin' should be equal to 'epoch', 'start', 'start_day', "
+        "'end', 'end_day' or should be a Timestamp convertible type. Got "
+        f"'{origin}' instead."
+    )
+    with pytest.raises(ValueError, match=msg):
+        ts.resample("5min", origin=origin)
+
+
+@pytest.mark.parametrize("offset", ["invalid_value", "12dayys", "2000-30-30", object()])
+def test_resample_bad_offset(offset, unit):
+    rng = date_range("2000-01-01 00:00:00", "2000-01-01 02:00", freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    msg = f"'offset' should be a Timedelta convertible type. Got '{offset}' instead."
+    with pytest.raises(ValueError, match=msg):
+        ts.resample("5min", offset=offset)
+
+
+def test_resample_origin_prime_freq(unit):
+    # GH 31809
+    start, end = "2000-10-01 23:30:00", "2000-10-02 00:30:00"
+    rng = date_range(start, end, freq="7min").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    exp_rng = date_range(
+        "2000-10-01 23:14:00", "2000-10-02 00:22:00", freq="17min"
+    ).as_unit(unit)
+    resampled = ts.resample("17min").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+    resampled = ts.resample("17min", origin="start_day").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    exp_rng = date_range(
+        "2000-10-01 23:30:00", "2000-10-02 00:21:00", freq="17min"
+    ).as_unit(unit)
+    resampled = ts.resample("17min", origin="start").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+    resampled = ts.resample("17min", offset="23h30min").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+    resampled = ts.resample("17min", origin="start_day", offset="23h30min").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    exp_rng = date_range(
+        "2000-10-01 23:18:00", "2000-10-02 00:26:00", freq="17min"
+    ).as_unit(unit)
+    resampled = ts.resample("17min", origin="epoch").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    exp_rng = date_range(
+        "2000-10-01 23:24:00", "2000-10-02 00:15:00", freq="17min"
+    ).as_unit(unit)
+    resampled = ts.resample("17min", origin="2000-01-01").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+
+def test_resample_origin_with_tz(unit):
+    # GH 31809
+    msg = "The origin must have the same timezone as the index."
+
+    tz = "Europe/Paris"
+    rng = date_range(
+        "2000-01-01 00:00:00", "2000-01-01 02:00", freq="s", tz=tz
+    ).as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    exp_rng = date_range(
+        "1999-12-31 23:57:00", "2000-01-01 01:57", freq="5min", tz=tz
+    ).as_unit(unit)
+    resampled = ts.resample("5min", origin="1999-12-31 23:57:00+00:00").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    # origin of '1999-31-12 12:02:00+03:00' should be equivalent for this case
+    resampled = ts.resample("5min", origin="1999-12-31 12:02:00+03:00").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    resampled = ts.resample("5min", origin="epoch", offset="2m").mean()
+    tm.assert_index_equal(resampled.index, exp_rng)
+
+    with pytest.raises(ValueError, match=msg):
+        ts.resample("5min", origin="12/31/1999 23:57:00").mean()
+
+    # if the series is not tz aware, origin should not be tz aware
+    rng = date_range("2000-01-01 00:00:00", "2000-01-01 02:00", freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    with pytest.raises(ValueError, match=msg):
+        ts.resample("5min", origin="12/31/1999 23:57:00+03:00").mean()
+
+
+def test_resample_origin_epoch_with_tz_day_vs_24h(unit):
+    # GH 34474
+    start, end = "2000-10-01 23:30:00+0500", "2000-12-02 00:30:00+0500"
+    rng = date_range(start, end, freq="7min").as_unit(unit)
+    random_values = np.random.default_rng(2).standard_normal(len(rng))
+    ts_1 = Series(random_values, index=rng)
+
+    result_1 = ts_1.resample("D", origin="epoch").mean()
+    result_2 = ts_1.resample("24h", origin="epoch").mean()
+    tm.assert_series_equal(result_1, result_2)
+
+    # check that we have the same behavior with epoch even if we are not timezone aware
+    ts_no_tz = ts_1.tz_localize(None)
+    result_3 = ts_no_tz.resample("D", origin="epoch").mean()
+    result_4 = ts_no_tz.resample("24h", origin="epoch").mean()
+    tm.assert_series_equal(result_1, result_3.tz_localize(rng.tz), check_freq=False)
+    tm.assert_series_equal(result_1, result_4.tz_localize(rng.tz), check_freq=False)
+
+    # check that we have the similar results with two different timezones (+2H and +5H)
+    start, end = "2000-10-01 23:30:00+0200", "2000-12-02 00:30:00+0200"
+    rng = date_range(start, end, freq="7min").as_unit(unit)
+    ts_2 = Series(random_values, index=rng)
+    result_5 = ts_2.resample("D", origin="epoch").mean()
+    result_6 = ts_2.resample("24h", origin="epoch").mean()
+    tm.assert_series_equal(result_1.tz_localize(None), result_5.tz_localize(None))
+    tm.assert_series_equal(result_1.tz_localize(None), result_6.tz_localize(None))
+
+
+def test_resample_origin_with_day_freq_on_dst(unit):
+    # GH 31809
+    tz = "America/Chicago"
+
+    def _create_series(values, timestamps, freq="D"):
+        return Series(
+            values,
+            index=DatetimeIndex(
+                [Timestamp(t, tz=tz) for t in timestamps], freq=freq, ambiguous=True
+            ).as_unit(unit),
+        )
+
+    # test classical behavior of origin in a DST context
+    start = Timestamp("2013-11-02", tz=tz)
+    end = Timestamp("2013-11-03 23:59", tz=tz)
+    rng = date_range(start, end, freq="1h").as_unit(unit)
+    ts = Series(np.ones(len(rng)), index=rng)
+
+    expected = _create_series([24.0, 25.0], ["2013-11-02", "2013-11-03"])
+    for origin in ["epoch", "start", "start_day", start, None]:
+        result = ts.resample("D", origin=origin).sum()
+        tm.assert_series_equal(result, expected)
+
+    # test complex behavior of origin/offset in a DST context
+    start = Timestamp("2013-11-03", tz=tz)
+    end = Timestamp("2013-11-03 23:59", tz=tz)
+    rng = date_range(start, end, freq="1h").as_unit(unit)
+    ts = Series(np.ones(len(rng)), index=rng)
+
+    expected_ts = ["2013-11-02 22:00-05:00", "2013-11-03 22:00-06:00"]
+    expected = _create_series([23.0, 2.0], expected_ts)
+    result = ts.resample("D", origin="start", offset="-2h").sum()
+    tm.assert_series_equal(result, expected)
+
+    expected_ts = ["2013-11-02 22:00-05:00", "2013-11-03 21:00-06:00"]
+    expected = _create_series([22.0, 3.0], expected_ts, freq="24h")
+    result = ts.resample("24h", origin="start", offset="-2h").sum()
+    tm.assert_series_equal(result, expected)
+
+    expected_ts = ["2013-11-02 02:00-05:00", "2013-11-03 02:00-06:00"]
+    expected = _create_series([3.0, 22.0], expected_ts)
+    result = ts.resample("D", origin="start", offset="2h").sum()
+    tm.assert_series_equal(result, expected)
+
+    expected_ts = ["2013-11-02 23:00-05:00", "2013-11-03 23:00-06:00"]
+    expected = _create_series([24.0, 1.0], expected_ts)
+    result = ts.resample("D", origin="start", offset="-1h").sum()
+    tm.assert_series_equal(result, expected)
+
+    expected_ts = ["2013-11-02 01:00-05:00", "2013-11-03 01:00:00-0500"]
+    expected = _create_series([1.0, 24.0], expected_ts)
+    result = ts.resample("D", origin="start", offset="1h").sum()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_dst_midnight_last_nonexistent():
+    # GH 58380
+    ts = Series(
+        1,
+        date_range("2024-04-19", "2024-04-20", tz="Africa/Cairo", freq="15min"),
+    )
+
+    expected = Series([len(ts)], index=DatetimeIndex([ts.index[0]], freq="7D"))
+
+    result = ts.resample("7D").sum()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_daily_anchored(unit):
+    rng = date_range("1/1/2000 0:00:00", periods=10000, freq="min").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    ts[:2] = np.nan  # so results are the same
+
+    result = ts[2:].resample("D", closed="left", label="left").mean()
+    expected = ts.resample("D", closed="left", label="left").mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_to_period_monthly_buglet(unit):
+    # GH #1259
+
+    rng = date_range("1/1/2000", "12/31/2000").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    result = ts.resample("ME").mean().to_period()
+    exp_index = period_range("Jan-2000", "Dec-2000", freq="M")
+    tm.assert_index_equal(result.index, exp_index)
+
+
+def test_period_with_agg():
+    # aggregate a period resampler with a lambda
+    s2 = Series(
+        np.random.default_rng(2).integers(0, 5, 50),
+        index=period_range("2012-01-01", freq="h", periods=50),
+        dtype="float64",
+    )
+
+    expected = s2.to_timestamp().resample("D").mean().to_period()
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        rs = s2.resample("D")
+    result = rs.agg(lambda x: x.mean())
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_segfault(unit):
+    # GH 8573
+    # segfaulting in older versions
+    all_wins_and_wagers = [
+        (1, datetime(2013, 10, 1, 16, 20), 1, 0),
+        (2, datetime(2013, 10, 1, 16, 10), 1, 0),
+        (2, datetime(2013, 10, 1, 18, 15), 1, 0),
+        (2, datetime(2013, 10, 1, 16, 10, 31), 1, 0),
+    ]
+
+    df = DataFrame.from_records(
+        all_wins_and_wagers, columns=("ID", "timestamp", "A", "B")
+    ).set_index("timestamp")
+    df.index = df.index.as_unit(unit)
+    result = df.groupby("ID").resample("5min").sum()
+    expected = df.groupby("ID").apply(lambda x: x.resample("5min").sum())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_dtype_preservation(unit):
+    # GH 12202
+    # validation tests for dtype preservation
+
+    df = DataFrame(
+        {
+            "date": date_range(start="2016-01-01", periods=4, freq="W").as_unit(unit),
+            "group": [1, 1, 2, 2],
+            "val": Series([5, 6, 7, 8], dtype="int32"),
+        }
+    ).set_index("date")
+
+    result = df.resample("1D").ffill()
+    assert result.val.dtype == np.int32
+
+    result = df.groupby("group").resample("1D").ffill()
+    assert result.val.dtype == np.int32
+
+
+def test_resample_dtype_coercion(unit):
+    pytest.importorskip("scipy.interpolate")
+
+    # GH 16361
+    df = {"a": [1, 3, 1, 4]}
+    df = DataFrame(df, index=date_range("2017-01-01", "2017-01-04").as_unit(unit))
+
+    expected = df.astype("float64").resample("h").mean()["a"].interpolate("cubic")
+
+    result = df.resample("h")["a"].mean().interpolate("cubic")
+    tm.assert_series_equal(result, expected)
+
+    result = df.resample("h").mean()["a"].interpolate("cubic")
+    tm.assert_series_equal(result, expected)
+
+
+def test_weekly_resample_buglet(unit):
+    # #1327
+    rng = date_range("1/1/2000", freq="B", periods=20).as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    resampled = ts.resample("W").mean()
+    expected = ts.resample("W-SUN").mean()
+    tm.assert_series_equal(resampled, expected)
+
+
+def test_monthly_resample_error(unit):
+    # #1451
+    dates = date_range("4/16/2012 20:00", periods=5000, freq="h").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(dates)), index=dates)
+    # it works!
+    ts.resample("ME")
+
+
+def test_nanosecond_resample_error():
+    # GH 12307 - Values falls after last bin when
+    # Resampling using pd.tseries.offsets.Nano as period
+    start = 1443707890427
+    exp_start = 1443707890400
+    indx = date_range(start=pd.to_datetime(start), periods=10, freq="100ns")
+    ts = Series(range(len(indx)), index=indx)
+    r = ts.resample(pd.tseries.offsets.Nano(100))
+    result = r.agg("mean")
+
+    exp_indx = date_range(start=pd.to_datetime(exp_start), periods=10, freq="100ns")
+    exp = Series(range(len(exp_indx)), index=exp_indx, dtype=float)
+
+    tm.assert_series_equal(result, exp)
+
+
+def test_resample_anchored_intraday(unit):
+    # #1471, #1458
+
+    rng = date_range("1/1/2012", "4/1/2012", freq="100min").as_unit(unit)
+    df = DataFrame(rng.month, index=rng)
+
+    result = df.resample("ME").mean()
+    expected = df.resample("ME").mean().to_period()
+    expected = expected.to_timestamp(how="end")
+    expected.index += Timedelta(1, "ns") - Timedelta(1, "D")
+    expected.index = expected.index.as_unit(unit)._with_freq("infer")
+    assert expected.index.freq == "ME"
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("ME", closed="left").mean()
+    exp = df.shift(1, freq="D").resample("ME").mean().to_period()
+    exp = exp.to_timestamp(how="end")
+
+    exp.index = exp.index + Timedelta(1, "ns") - Timedelta(1, "D")
+    exp.index = exp.index.as_unit(unit)._with_freq("infer")
+    assert exp.index.freq == "ME"
+    tm.assert_frame_equal(result, exp)
+
+
+def test_resample_anchored_intraday2(unit):
+    rng = date_range("1/1/2012", "4/1/2012", freq="100min").as_unit(unit)
+    df = DataFrame(rng.month, index=rng)
+
+    result = df.resample("QE").mean()
+    expected = df.resample("QE").mean().to_period()
+    expected = expected.to_timestamp(how="end")
+    expected.index += Timedelta(1, "ns") - Timedelta(1, "D")
+    expected.index._data.freq = "QE"
+    expected.index._freq = lib.no_default
+    expected.index = expected.index.as_unit(unit)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.resample("QE", closed="left").mean()
+    expected = df.shift(1, freq="D").resample("QE").mean()
+    expected = expected.to_period()
+    expected = expected.to_timestamp(how="end")
+    expected.index += Timedelta(1, "ns") - Timedelta(1, "D")
+    expected.index._data.freq = "QE"
+    expected.index._freq = lib.no_default
+    expected.index = expected.index.as_unit(unit)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_anchored_intraday3(simple_date_range_series, unit):
+    ts = simple_date_range_series("2012-04-29 23:00", "2012-04-30 5:00", freq="h")
+    ts.index = ts.index.as_unit(unit)
+    resampled = ts.resample("ME").mean()
+    assert len(resampled) == 1
+
+
+@pytest.mark.parametrize("freq", ["MS", "BMS", "QS-MAR", "YS-DEC", "YS-JUN"])
+def test_resample_anchored_monthstart(simple_date_range_series, freq, unit):
+    ts = simple_date_range_series("1/1/2000", "12/31/2002")
+    ts.index = ts.index.as_unit(unit)
+    ts.resample(freq).mean()
+
+
+@pytest.mark.parametrize("label, sec", [[None, 2.0], ["right", "4.2"]])
+def test_resample_anchored_multiday(label, sec):
+    # When resampling a range spanning multiple days, ensure that the
+    # start date gets used to determine the offset.  Fixes issue where
+    # a one day period is not a multiple of the frequency.
+    #
+    # See: https://github.com/pandas-dev/pandas/issues/8683
+
+    index1 = date_range("2014-10-14 23:06:23.206", periods=3, freq="400ms")
+    index2 = date_range("2014-10-15 23:00:00", periods=2, freq="2200ms")
+    index = index1.union(index2)
+
+    s = Series(np.random.default_rng(2).standard_normal(5), index=index)
+
+    # Ensure left closing works
+    result = s.resample("2200ms", label=label).mean()
+    assert result.index[-1] == Timestamp(f"2014-10-15 23:00:{sec}00")
+
+
+def test_corner_cases(unit):
+    # miscellaneous test coverage
+
+    rng = date_range("1/1/2000", periods=12, freq="min").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    result = ts.resample("5min", closed="right", label="left").mean()
+    ex_index = date_range("1999-12-31 23:55", periods=4, freq="5min").as_unit(unit)
+    tm.assert_index_equal(result.index, ex_index)
+
+
+def test_corner_cases_date(simple_date_range_series, unit):
+    # resample to periods
+    ts = simple_date_range_series("2000-04-28", "2000-04-30 11:00", freq="h")
+    ts.index = ts.index.as_unit(unit)
+    result = ts.resample("ME").mean().to_period()
+    assert len(result) == 1
+    assert result.index[0] == Period("2000-04", freq="M")
+
+
+def test_anchored_lowercase_buglet(unit):
+    dates = date_range("4/16/2012 20:00", periods=50000, freq="s").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(dates)), index=dates)
+    # it works!
+    msg = "'d' is deprecated and will be removed in a future version."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        ts.resample("d").mean()
+
+
+def test_upsample_apply_functions(unit):
+    # #1596
+    rng = date_range("2012-06-12", periods=4, freq="h").as_unit(unit)
+
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    result = ts.resample("20min").aggregate(["mean", "sum"])
+    assert isinstance(result, DataFrame)
+
+
+def test_resample_not_monotonic(unit):
+    rng = date_range("2012-06-12", periods=200, freq="h").as_unit(unit)
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    ts = ts.take(np.random.default_rng(2).permutation(len(ts)))
+
+    result = ts.resample("D").sum()
+    exp = ts.sort_index().resample("D").sum()
+    tm.assert_series_equal(result, exp)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "int64",
+        "int32",
+        "float64",
+        pytest.param(
+            "float32",
+            marks=pytest.mark.xfail(
+                reason="Empty groups cause x.mean() to return float64"
+            ),
+        ),
+    ],
+)
+def test_resample_median_bug_1688(dtype, unit):
+    # GH#55958
+    dti = DatetimeIndex(
+        [datetime(2012, 1, 1, 0, 0, 0), datetime(2012, 1, 1, 0, 5, 0)]
+    ).as_unit(unit)
+    df = DataFrame(
+        [1, 2],
+        index=dti,
+        dtype=dtype,
+    )
+
+    result = df.resample("min").apply(lambda x: x.mean())
+    exp = df.asfreq("min")
+    tm.assert_frame_equal(result, exp)
+
+    result = df.resample("min").median()
+    exp = df.asfreq("min")
+    tm.assert_frame_equal(result, exp)
+
+
+def test_how_lambda_functions(simple_date_range_series, unit):
+    ts = simple_date_range_series("1/1/2000", "4/1/2000")
+    ts.index = ts.index.as_unit(unit)
+
+    result = ts.resample("ME").apply(lambda x: x.mean())
+    exp = ts.resample("ME").mean()
+    tm.assert_series_equal(result, exp)
+
+    foo_exp = ts.resample("ME").mean()
+    foo_exp.name = "foo"
+    bar_exp = ts.resample("ME").std()
+    bar_exp.name = "bar"
+
+    result = ts.resample("ME").apply([lambda x: x.mean(), lambda x: x.std(ddof=1)])
+    result.columns = ["foo", "bar"]
+    tm.assert_series_equal(result["foo"], foo_exp)
+    tm.assert_series_equal(result["bar"], bar_exp)
+
+    # this is a MI Series, so comparing the names of the results
+    # doesn't make sense
+    result = ts.resample("ME").aggregate(
+        {"foo": lambda x: x.mean(), "bar": lambda x: x.std(ddof=1)}
+    )
+    tm.assert_series_equal(result["foo"], foo_exp, check_names=False)
+    tm.assert_series_equal(result["bar"], bar_exp, check_names=False)
+
+
+def test_resample_unequal_times(unit):
+    # #1772
+    start = datetime(1999, 3, 1, 5)
+    # end hour is less than start
+    end = datetime(2012, 7, 31, 4)
+    bad_ind = date_range(start, end, freq="30min").as_unit(unit)
+    df = DataFrame({"close": 1}, index=bad_ind)
+
+    # it works!
+    df.resample("YS").sum()
+
+
+def test_resample_consistency(unit):
+    # GH 6418
+    # resample with bfill / limit / reindex consistency
+
+    i30 = date_range("2002-02-02", periods=4, freq="30min").as_unit(unit)
+    s = Series(np.arange(4.0), index=i30)
+    s.iloc[2] = np.nan
+
+    # Upsample by factor 3 with reindex() and resample() methods:
+    i10 = date_range(i30[0], i30[-1], freq="10min").as_unit(unit)
+
+    s10 = s.reindex(index=i10, method="bfill")
+    s10_2 = s.reindex(index=i10, method="bfill", limit=2)
+    with tm.assert_produces_warning(FutureWarning):
+        rl = s.reindex_like(s10, method="bfill", limit=2)
+    r10_2 = s.resample("10Min").bfill(limit=2)
+    r10 = s.resample("10Min").bfill()
+
+    # s10_2, r10, r10_2, rl should all be equal
+    tm.assert_series_equal(s10_2, r10)
+    tm.assert_series_equal(s10_2, r10_2)
+    tm.assert_series_equal(s10_2, rl)
+
+
+dates1: list[DatetimeNaTType] = [
+    datetime(2014, 10, 1),
+    datetime(2014, 9, 3),
+    datetime(2014, 11, 5),
+    datetime(2014, 9, 5),
+    datetime(2014, 10, 8),
+    datetime(2014, 7, 15),
+]
+
+dates2: list[DatetimeNaTType] = (
+    dates1[:2] + [pd.NaT] + dates1[2:4] + [pd.NaT] + dates1[4:]
+)
+dates3 = [pd.NaT] + dates1 + [pd.NaT]
+
+
+@pytest.mark.parametrize("dates", [dates1, dates2, dates3])
+def test_resample_timegrouper(dates, unit):
+    # GH 7227
+    dates = DatetimeIndex(dates).as_unit(unit)
+    df = DataFrame({"A": dates, "B": np.arange(len(dates))})
+    result = df.set_index("A").resample("ME").count()
+    exp_idx = DatetimeIndex(
+        ["2014-07-31", "2014-08-31", "2014-09-30", "2014-10-31", "2014-11-30"],
+        freq="ME",
+        name="A",
+    ).as_unit(unit)
+    expected = DataFrame({"B": [1, 0, 2, 2, 1]}, index=exp_idx)
+    if df["A"].isna().any():
+        expected.index = expected.index._with_freq(None)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(Grouper(freq="ME", key="A")).count()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dates", [dates1, dates2, dates3])
+def test_resample_timegrouper2(dates, unit):
+    dates = DatetimeIndex(dates).as_unit(unit)
+
+    df = DataFrame({"A": dates, "B": np.arange(len(dates)), "C": np.arange(len(dates))})
+    result = df.set_index("A").resample("ME").count()
+
+    exp_idx = DatetimeIndex(
+        ["2014-07-31", "2014-08-31", "2014-09-30", "2014-10-31", "2014-11-30"],
+        freq="ME",
+        name="A",
+    ).as_unit(unit)
+    expected = DataFrame(
+        {"B": [1, 0, 2, 2, 1], "C": [1, 0, 2, 2, 1]},
+        index=exp_idx,
+        columns=["B", "C"],
+    )
+    if df["A"].isna().any():
+        expected.index = expected.index._with_freq(None)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby(Grouper(freq="ME", key="A")).count()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_nunique(unit):
+    # GH 12352
+    df = DataFrame(
+        {
+            "ID": {
+                Timestamp("2015-06-05 00:00:00"): "0010100903",
+                Timestamp("2015-06-08 00:00:00"): "0010150847",
+            },
+            "DATE": {
+                Timestamp("2015-06-05 00:00:00"): "2015-06-05",
+                Timestamp("2015-06-08 00:00:00"): "2015-06-08",
+            },
+        }
+    )
+    df.index = df.index.as_unit(unit)
+    r = df.resample("D")
+    g = df.groupby(Grouper(freq="D"))
+    expected = df.groupby(Grouper(freq="D")).ID.apply(lambda x: x.nunique())
+    assert expected.name == "ID"
+
+    for t in [r, g]:
+        result = t.ID.nunique()
+        tm.assert_series_equal(result, expected)
+
+    result = df.ID.resample("D").nunique()
+    tm.assert_series_equal(result, expected)
+
+    result = df.ID.groupby(Grouper(freq="D")).nunique()
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_nunique_preserves_column_level_names(unit):
+    # see gh-23222
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((5, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=5, freq="D"),
+    ).abs()
+    df.index = df.index.as_unit(unit)
+    df.columns = pd.MultiIndex.from_arrays(
+        [df.columns.tolist()] * 2, names=["lev0", "lev1"]
+    )
+    result = df.resample("1h").nunique()
+    tm.assert_index_equal(df.columns, result.columns)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x: x.nunique(),
+        lambda x: x.agg(Series.nunique),
+        lambda x: x.agg("nunique"),
+    ],
+    ids=["nunique", "series_nunique", "nunique_str"],
+)
+def test_resample_nunique_with_date_gap(func, unit):
+    # GH 13453
+    # Since all elements are unique, these should all be the same
+    index = date_range("1-1-2000", "2-15-2000", freq="h").as_unit(unit)
+    index2 = date_range("4-15-2000", "5-15-2000", freq="h").as_unit(unit)
+    index3 = index.append(index2)
+    s = Series(range(len(index3)), index=index3, dtype="int64")
+    r = s.resample("ME")
+    result = r.count()
+    expected = func(r)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_group_info(unit):
+    # GH10914
+
+    # use a fixed seed to always have the same uniques
+    n = 100
+    k = 10
+    prng = np.random.default_rng(2)
+
+    dr = date_range(start="2015-08-27", periods=n // 10, freq="min").as_unit(unit)
+    ts = Series(prng.integers(0, n // k, n).astype("int64"), index=prng.choice(dr, n))
+
+    left = ts.resample("30min").nunique()
+    ix = date_range(start=ts.index.min(), end=ts.index.max(), freq="30min").as_unit(
+        unit
+    )
+
+    vals = ts.values
+    bins = np.searchsorted(ix.values, ts.index, side="right")
+
+    sorter = np.lexsort((vals, bins))
+    vals, bins = vals[sorter], bins[sorter]
+
+    mask = np.r_[True, vals[1:] != vals[:-1]]
+    mask |= np.r_[True, bins[1:] != bins[:-1]]
+
+    arr = np.bincount(bins[mask] - 1, minlength=len(ix)).astype("int64", copy=False)
+    right = Series(arr, index=ix)
+
+    tm.assert_series_equal(left, right)
+
+
+def test_resample_size(unit):
+    n = 10000
+    dr = date_range("2015-09-19", periods=n, freq="min").as_unit(unit)
+    ts = Series(
+        np.random.default_rng(2).standard_normal(n),
+        index=np.random.default_rng(2).choice(dr, n),
+    )
+
+    left = ts.resample("7min").size()
+    ix = date_range(start=left.index.min(), end=ts.index.max(), freq="7min").as_unit(
+        unit
+    )
+
+    bins = np.searchsorted(ix.values, ts.index.values, side="right")
+    val = np.bincount(bins, minlength=len(ix) + 1)[1:].astype("int64", copy=False)
+
+    right = Series(val, index=ix)
+    tm.assert_series_equal(left, right)
+
+
+def test_resample_across_dst():
+    # The test resamples a DatetimeIndex with values before and after a
+    # DST change
+    # Issue: 14682
+
+    # The DatetimeIndex we will start with
+    # (note that DST happens at 03:00+02:00 -> 02:00+01:00)
+    # 2016-10-30 02:23:00+02:00, 2016-10-30 02:23:00+01:00
+    df1 = DataFrame([1477786980, 1477790580], columns=["ts"])
+    dti1 = DatetimeIndex(
+        pd.to_datetime(df1.ts, unit="s")
+        .dt.tz_localize("UTC")
+        .dt.tz_convert("Europe/Madrid")
+    )
+
+    # The expected DatetimeIndex after resampling.
+    # 2016-10-30 02:00:00+02:00, 2016-10-30 02:00:00+01:00
+    df2 = DataFrame([1477785600, 1477789200], columns=["ts"])
+    dti2 = DatetimeIndex(
+        pd.to_datetime(df2.ts, unit="s")
+        .dt.tz_localize("UTC")
+        .dt.tz_convert("Europe/Madrid"),
+        freq="h",
+    )
+    df = DataFrame([5, 5], index=dti1)
+
+    result = df.resample(rule="h").sum()
+    expected = DataFrame([5, 5], index=dti2)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_with_dst_time_change(unit):
+    # GH 24972
+    index = (
+        DatetimeIndex([1478064900001000000, 1480037118776792000], tz="UTC")
+        .tz_convert("America/Chicago")
+        .as_unit(unit)
+    )
+
+    df = DataFrame([1, 2], index=index)
+    result = df.groupby(Grouper(freq="1D")).last()
+    expected_index_values = date_range(
+        "2016-11-02", "2016-11-24", freq="D", tz="America/Chicago"
+    ).as_unit(unit)
+
+    index = DatetimeIndex(expected_index_values)
+    expected = DataFrame([1.0] + ([np.nan] * 21) + [2.0], index=index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_dst_anchor(unit):
+    # 5172
+    dti = DatetimeIndex([datetime(2012, 11, 4, 23)], tz="US/Eastern").as_unit(unit)
+    df = DataFrame([5], index=dti)
+
+    dti = DatetimeIndex(df.index.normalize(), freq="D").as_unit(unit)
+    expected = DataFrame([5], index=dti)
+    tm.assert_frame_equal(df.resample(rule="D").sum(), expected)
+    df.resample(rule="MS").sum()
+    tm.assert_frame_equal(
+        df.resample(rule="MS").sum(),
+        DataFrame(
+            [5],
+            index=DatetimeIndex(
+                [datetime(2012, 11, 1)], tz="US/Eastern", freq="MS"
+            ).as_unit(unit),
+        ),
+    )
+
+
+def test_resample_dst_anchor2(unit):
+    dti = date_range(
+        "2013-09-30", "2013-11-02", freq="30Min", tz="Europe/Paris"
+    ).as_unit(unit)
+    values = range(dti.size)
+    df = DataFrame({"a": values, "b": values, "c": values}, index=dti, dtype="int64")
+    how = {"a": "min", "b": "max", "c": "count"}
+
+    rs = df.resample("W-MON")
+    result = rs.agg(how)[["a", "b", "c"]]
+    expected = DataFrame(
+        {
+            "a": [0, 48, 384, 720, 1056, 1394],
+            "b": [47, 383, 719, 1055, 1393, 1586],
+            "c": [48, 336, 336, 336, 338, 193],
+        },
+        index=date_range(
+            "9/30/2013", "11/4/2013", freq="W-MON", tz="Europe/Paris"
+        ).as_unit(unit),
+    )
+    tm.assert_frame_equal(
+        result,
+        expected,
+        "W-MON Frequency",
+    )
+
+    rs2 = df.resample("2W-MON")
+    result2 = rs2.agg(how)[["a", "b", "c"]]
+    expected2 = DataFrame(
+        {
+            "a": [0, 48, 720, 1394],
+            "b": [47, 719, 1393, 1586],
+            "c": [48, 672, 674, 193],
+        },
+        index=date_range(
+            "9/30/2013", "11/11/2013", freq="2W-MON", tz="Europe/Paris"
+        ).as_unit(unit),
+    )
+    tm.assert_frame_equal(
+        result2,
+        expected2,
+        "2W-MON Frequency",
+    )
+
+    rs3 = df.resample("MS")
+    result3 = rs3.agg(how)[["a", "b", "c"]]
+    expected3 = DataFrame(
+        {"a": [0, 48, 1538], "b": [47, 1537, 1586], "c": [48, 1490, 49]},
+        index=date_range("9/1/2013", "11/1/2013", freq="MS", tz="Europe/Paris").as_unit(
+            unit
+        ),
+    )
+    tm.assert_frame_equal(
+        result3,
+        expected3,
+        "MS Frequency",
+    )
+
+    rs4 = df.resample("2MS")
+    result4 = rs4.agg(how)[["a", "b", "c"]]
+    expected4 = DataFrame(
+        {"a": [0, 1538], "b": [1537, 1586], "c": [1538, 49]},
+        index=date_range(
+            "9/1/2013", "11/1/2013", freq="2MS", tz="Europe/Paris"
+        ).as_unit(unit),
+    )
+    tm.assert_frame_equal(
+        result4,
+        expected4,
+        "2MS Frequency",
+    )
+
+    df_daily = df["10/26/2013":"10/29/2013"]
+    rs_d = df_daily.resample("D")
+    result_d = rs_d.agg({"a": "min", "b": "max", "c": "count"})[["a", "b", "c"]]
+    expected_d = DataFrame(
+        {
+            "a": [1248, 1296, 1346, 1394],
+            "b": [1295, 1345, 1393, 1441],
+            "c": [48, 50, 48, 48],
+        },
+        index=date_range(
+            "10/26/2013", "10/29/2013", freq="D", tz="Europe/Paris"
+        ).as_unit(unit),
+    )
+    tm.assert_frame_equal(
+        result_d,
+        expected_d,
+        "D Frequency",
+    )
+
+
+def test_downsample_across_dst(unit):
+    # GH 8531
+    tz = zoneinfo.ZoneInfo("Europe/Berlin")
+    dt = datetime(2014, 10, 26)
+    dates = date_range(dt.astimezone(tz), periods=4, freq="2h").as_unit(unit)
+    result = Series(5, index=dates).resample("h").mean()
+    expected = Series(
+        [5.0, np.nan] * 3 + [5.0],
+        index=date_range(dt.astimezone(tz), periods=7, freq="h").as_unit(unit),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_downsample_across_dst_weekly(unit):
+    # GH 9119, GH 21459
+    df = DataFrame(
+        index=DatetimeIndex(
+            ["2017-03-25", "2017-03-26", "2017-03-27", "2017-03-28", "2017-03-29"],
+            tz="Europe/Amsterdam",
+        ).as_unit(unit),
+        data=[11, 12, 13, 14, 15],
+    )
+    result = df.resample("1W").sum()
+    expected = DataFrame(
+        [23, 42],
+        index=DatetimeIndex(
+            ["2017-03-26", "2017-04-02"], tz="Europe/Amsterdam", freq="W"
+        ).as_unit(unit),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_downsample_across_dst_weekly_2(unit):
+    # GH 9119, GH 21459
+    idx = date_range("2013-04-01", "2013-05-01", tz="Europe/London", freq="h").as_unit(
+        unit
+    )
+    s = Series(index=idx, dtype=np.float64)
+    result = s.resample("W").mean()
+    expected = Series(
+        index=date_range("2013-04-07", freq="W", periods=5, tz="Europe/London").as_unit(
+            unit
+        ),
+        dtype=np.float64,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_downsample_dst_at_midnight(unit):
+    # GH 25758
+    start = datetime(2018, 11, 3, 12)
+    end = datetime(2018, 11, 5, 12)
+    index = date_range(start, end, freq="1h").as_unit(unit)
+    index = index.tz_localize("UTC").tz_convert("America/Havana")
+    data = list(range(len(index)))
+    dataframe = DataFrame(data, index=index)
+    result = dataframe.groupby(Grouper(freq="1D")).mean()
+
+    dti = date_range("2018-11-03", periods=3).tz_localize(
+        "America/Havana", ambiguous=True
+    )
+    dti = DatetimeIndex(dti, freq="D").as_unit(unit)
+    expected = DataFrame([7.5, 28.0, 44.5], index=dti)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_with_nat(unit):
+    # GH 13020
+    index = DatetimeIndex(
+        [
+            pd.NaT,
+            "1970-01-01 00:00:00",
+            pd.NaT,
+            "1970-01-01 00:00:01",
+            "1970-01-01 00:00:02",
+        ]
+    ).as_unit(unit)
+    frame = DataFrame([2, 3, 5, 7, 11], index=index)
+
+    index_1s = DatetimeIndex(
+        ["1970-01-01 00:00:00", "1970-01-01 00:00:01", "1970-01-01 00:00:02"]
+    ).as_unit(unit)
+    frame_1s = DataFrame([3.0, 7.0, 11.0], index=index_1s)
+    tm.assert_frame_equal(frame.resample("1s").mean(), frame_1s)
+
+    index_2s = DatetimeIndex(["1970-01-01 00:00:00", "1970-01-01 00:00:02"]).as_unit(
+        unit
+    )
+    frame_2s = DataFrame([5.0, 11.0], index=index_2s)
+    tm.assert_frame_equal(frame.resample("2s").mean(), frame_2s)
+
+    index_3s = DatetimeIndex(["1970-01-01 00:00:00"]).as_unit(unit)
+    frame_3s = DataFrame([7.0], index=index_3s)
+    tm.assert_frame_equal(frame.resample("3s").mean(), frame_3s)
+
+    tm.assert_frame_equal(frame.resample("60s").mean(), frame_3s)
+
+
+def test_resample_datetime_values(unit):
+    # GH 13119
+    # check that datetime dtype is preserved when NaT values are
+    # introduced by the resampling
+
+    dates = [datetime(2016, 1, 15), datetime(2016, 1, 19)]
+    df = DataFrame({"timestamp": dates}, index=dates)
+    df.index = df.index.as_unit(unit)
+
+    exp = Series(
+        [datetime(2016, 1, 15), pd.NaT, datetime(2016, 1, 19)],
+        index=date_range("2016-01-15", periods=3, freq="2D").as_unit(unit),
+        name="timestamp",
+    )
+
+    res = df.resample("2D").first()["timestamp"]
+    tm.assert_series_equal(res, exp)
+    res = df["timestamp"].resample("2D").first()
+    tm.assert_series_equal(res, exp)
+
+
+def test_resample_apply_with_additional_args(unit):
+    # GH 14615
+    index = date_range("1/1/2000 00:00:00", "1/1/2000 00:13:00", freq="Min")
+    series = Series(range(len(index)), index=index)
+    series.index.name = "index"
+
+    def f(data, add_arg):
+        return np.mean(data) * add_arg
+
+    series.index = series.index.as_unit(unit)
+
+    multiplier = 10
+    result = series.resample("D").apply(f, multiplier)
+    expected = series.resample("D").mean().multiply(multiplier)
+    tm.assert_series_equal(result, expected)
+
+    # Testing as kwarg
+    result = series.resample("D").apply(f, add_arg=multiplier)
+    expected = series.resample("D").mean().multiply(multiplier)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_apply_with_additional_args2():
+    # Testing dataframe
+    def f(data, add_arg):
+        return np.mean(data) * add_arg
+
+    multiplier = 10
+
+    df = DataFrame({"A": 1, "B": 2}, index=date_range("2017", periods=10))
+    result = df.groupby("A").resample("D").agg(f, multiplier).astype(float)
+    expected = df.groupby("A").resample("D").mean().multiply(multiplier)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("k", [1, 2, 3])
+@pytest.mark.parametrize(
+    "n1, freq1, n2, freq2",
+    [
+        (30, "s", 0.5, "Min"),
+        (60, "s", 1, "Min"),
+        (3600, "s", 1, "h"),
+        (60, "Min", 1, "h"),
+        (21600, "s", 0.25, "D"),
+        (86400, "s", 1, "D"),
+        (43200, "s", 0.5, "D"),
+        (1440, "Min", 1, "D"),
+        (12, "h", 0.5, "D"),
+        (24, "h", 1, "D"),
+    ],
+)
+def test_resample_equivalent_offsets(n1, freq1, n2, freq2, k, unit):
+    # GH 24127
+    n1_ = n1 * k
+    n2_ = n2 * k
+    dti = date_range("1991-09-05", "1991-09-12", freq=freq1).as_unit(unit)
+    ser = Series(range(len(dti)), index=dti)
+
+    result1 = ser.resample(str(n1_) + freq1).mean()
+    result2 = ser.resample(str(n2_) + freq2).mean()
+    tm.assert_series_equal(result1, result2)
+
+
+@pytest.mark.parametrize(
+    "first,last,freq,exp_first,exp_last",
+    [
+        ("19910905", "19920406", "D", "19910905", "19920407"),
+        ("19910905 00:00", "19920406 06:00", "D", "19910905", "19920407"),
+        ("19910905 06:00", "19920406 06:00", "h", "19910905 06:00", "19920406 07:00"),
+        ("19910906", "19920406", "ME", "19910831", "19920430"),
+        ("19910831", "19920430", "ME", "19910831", "19920531"),
+        ("1991-08", "1992-04", "ME", "19910831", "19920531"),
+    ],
+)
+def test_get_timestamp_range_edges(first, last, freq, exp_first, exp_last, unit):
+    first = Period(first)
+    first = first.to_timestamp(first.freq).as_unit(unit)
+    last = Period(last)
+    last = last.to_timestamp(last.freq).as_unit(unit)
+
+    exp_first = Timestamp(exp_first)
+    exp_last = Timestamp(exp_last)
+
+    freq = pd.tseries.frequencies.to_offset(freq)
+    result = _get_timestamp_range_edges(first, last, freq, unit="ns")
+    expected = (exp_first, exp_last)
+    assert result == expected
+
+
+@pytest.mark.parametrize("duplicates", [True, False])
+def test_resample_apply_product(duplicates, unit):
+    # GH 5586
+    index = date_range(start="2012-01-31", freq="ME", periods=12).as_unit(unit)
+
+    ts = Series(range(12), index=index)
+    df = DataFrame({"A": ts, "B": ts + 2})
+    if duplicates:
+        df.columns = ["A", "A"]
+
+    result = df.resample("QE").apply(np.prod)
+    expected = DataFrame(
+        np.array([[0, 24], [60, 210], [336, 720], [990, 1716]], dtype=np.int64),
+        index=DatetimeIndex(
+            ["2012-03-31", "2012-06-30", "2012-09-30", "2012-12-31"], freq="QE-DEC"
+        ).as_unit(unit),
+        columns=df.columns,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "first,last,freq_in,freq_out,exp_last",
+    [
+        (
+            "2020-03-28",
+            "2020-03-31",
+            "D",
+            "24h",
+            "2020-03-30 01:00",
+        ),  # includes transition into DST
+        (
+            "2020-03-28",
+            "2020-10-27",
+            "D",
+            "24h",
+            "2020-10-27 00:00",
+        ),  # includes transition into and out of DST
+        (
+            "2020-10-25",
+            "2020-10-27",
+            "D",
+            "24h",
+            "2020-10-26 23:00",
+        ),  # includes transition out of DST
+        (
+            "2020-03-28",
+            "2020-03-31",
+            "24h",
+            "D",
+            "2020-03-30 00:00",
+        ),  # same as above, but from 24H to D
+        ("2020-03-28", "2020-10-27", "24h", "D", "2020-10-27 00:00"),
+        ("2020-10-25", "2020-10-27", "24h", "D", "2020-10-26 00:00"),
+    ],
+)
+def test_resample_calendar_day_with_dst(
+    first: str, last: str, freq_in: str, freq_out: str, exp_last: str, unit
+):
+    # GH 35219
+    ts = Series(
+        1.0, date_range(first, last, freq=freq_in, tz="Europe/Amsterdam").as_unit(unit)
+    )
+    result = ts.resample(freq_out).ffill()
+    expected = Series(
+        1.0,
+        date_range(first, exp_last, freq=freq_out, tz="Europe/Amsterdam").as_unit(unit),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max", "first", "last"])
+def test_resample_aggregate_functions_min_count(func, unit):
+    # GH#37768
+    index = date_range(start="2020", freq="ME", periods=3).as_unit(unit)
+    ser = Series([1, np.nan, np.nan], index)
+    result = getattr(ser.resample("QE"), func)(min_count=2)
+    expected = Series(
+        [np.nan],
+        index=DatetimeIndex(["2020-03-31"], freq="QE-DEC").as_unit(unit),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_unsigned_int(any_unsigned_int_numpy_dtype, unit):
+    # gh-43329
+    df = DataFrame(
+        index=date_range(start="2000-01-01", end="2000-01-03 23", freq="12h").as_unit(
+            unit
+        ),
+        columns=["x"],
+        data=[0, 1, 0] * 2,
+        dtype=any_unsigned_int_numpy_dtype,
+    )
+    df = df.loc[(df.index < "2000-01-02") | (df.index > "2000-01-03"), :]
+
+    result = df.resample("D").max()
+
+    expected = DataFrame(
+        [1, np.nan, 0],
+        columns=["x"],
+        index=date_range(start="2000-01-01", end="2000-01-03 23", freq="D").as_unit(
+            unit
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_long_rule_non_nano():
+    # https://github.com/pandas-dev/pandas/issues/51024
+    idx = date_range("0300-01-01", "2000-01-01", unit="s", freq="100YE")
+    ser = Series([1, 4, 2, 8, 5, 7, 1, 4, 2, 8, 5, 7, 1, 4, 2, 8, 5], index=idx)
+    result = ser.resample("200YE").mean()
+    expected_idx = DatetimeIndex(
+        np.array(
+            [
+                "0300-12-31",
+                "0500-12-31",
+                "0700-12-31",
+                "0900-12-31",
+                "1100-12-31",
+                "1300-12-31",
+                "1500-12-31",
+                "1700-12-31",
+                "1900-12-31",
+            ]
+        ).astype("datetime64[s]"),
+        freq="200YE-DEC",
+    )
+    expected = Series([1.0, 3.0, 6.5, 4.0, 3.0, 6.5, 4.0, 3.0, 6.5], index=expected_idx)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_empty_series_with_tz():
+    # GH#53664
+    df = DataFrame({"ts": [], "values": []}).astype(
+        {"ts": "datetime64[ns, Atlantic/Faroe]"}
+    )
+    result = df.resample("2MS", on="ts", closed="left", label="left", origin="start")[
+        "values"
+    ].sum()
+
+    expected_idx = DatetimeIndex(
+        [], freq="2MS", name="ts", dtype="datetime64[ns, Atlantic/Faroe]"
+    )
+    expected = Series([], index=expected_idx, name="values", dtype="float64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("freq", ["2M", "2m", "2Q", "2Q-SEP", "2q-sep", "1Y", "2Y-MAR"])
+def test_resample_M_Q_Y_raises(freq):
+    msg = f"Invalid frequency: {freq}"
+
+    s = Series(range(10), index=date_range("20130101", freq="D", periods=10))
+    with pytest.raises(ValueError, match=msg):
+        s.resample(freq).mean()
+
+
+@pytest.mark.parametrize("freq", ["2BM", "1bm", "1BQ", "2BQ-MAR", "2bq=-mar"])
+def test_resample_BM_BQ_raises(freq):
+    msg = f"Invalid frequency: {freq}"
+
+    s = Series(range(10), index=date_range("20130101", freq="D", periods=10))
+    with pytest.raises(ValueError, match=msg):
+        s.resample(freq).mean()
+
+
+@pytest.mark.parametrize(
+    "freq,freq_depr,data",
+    [
+        ("1W-SUN", "1w-sun", ["2013-01-06"]),
+        ("1D", "1d", ["2013-01-01"]),
+        ("1B", "1b", ["2013-01-01"]),
+        ("1C", "1c", ["2013-01-01"]),
+    ],
+)
+def test_resample_depr_lowercase_frequency(freq, freq_depr, data):
+    msg = f"'{freq_depr[1:]}' is deprecated and will be removed in a future version."
+
+    s = Series(range(5), index=date_range("20130101", freq="h", periods=5))
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = s.resample(freq_depr).mean()
+
+    exp_dti = DatetimeIndex(data=data, dtype="datetime64[ns]", freq=freq)
+    expected = Series(2.0, index=exp_dti)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_ms_closed_right(unit):
+    # https://github.com/pandas-dev/pandas/issues/55271
+    dti = date_range(start="2020-01-31", freq="1min", periods=6000, unit=unit)
+    df = DataFrame({"ts": dti}, index=dti)
+    grouped = df.resample("MS", closed="right")
+    result = grouped.last()
+    exp_dti = DatetimeIndex(
+        [datetime(2020, 1, 1), datetime(2020, 2, 1)], freq="MS"
+    ).as_unit(unit)
+    expected = DataFrame(
+        {"ts": [datetime(2020, 2, 1), datetime(2020, 2, 4, 3, 59)]},
+        index=exp_dti,
+    ).astype(f"M8[{unit}]")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("freq", ["B", "C"])
+def test_resample_c_b_closed_right(freq: str, unit):
+    # https://github.com/pandas-dev/pandas/issues/55281
+    dti = date_range(start="2020-01-31", freq="1min", periods=6000, unit=unit)
+    df = DataFrame({"ts": dti}, index=dti)
+    grouped = df.resample(freq, closed="right")
+    result = grouped.last()
+
+    exp_dti = DatetimeIndex(
+        [
+            datetime(2020, 1, 30),
+            datetime(2020, 1, 31),
+            datetime(2020, 2, 3),
+            datetime(2020, 2, 4),
+        ],
+        freq=freq,
+    ).as_unit(unit)
+    expected = DataFrame(
+        {
+            "ts": [
+                datetime(2020, 1, 31),
+                datetime(2020, 2, 3),
+                datetime(2020, 2, 4),
+                datetime(2020, 2, 4, 3, 59),
+            ]
+        },
+        index=exp_dti,
+    ).astype(f"M8[{unit}]")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_b_55282(unit):
+    # https://github.com/pandas-dev/pandas/issues/55282
+    dti = date_range("2023-09-26", periods=6, freq="12h", unit=unit)
+    ser = Series([1, 2, 3, 4, 5, 6], index=dti)
+    result = ser.resample("B", closed="right", label="right").mean()
+
+    exp_dti = DatetimeIndex(
+        [
+            datetime(2023, 9, 26),
+            datetime(2023, 9, 27),
+            datetime(2023, 9, 28),
+            datetime(2023, 9, 29),
+        ],
+        freq="B",
+    ).as_unit(unit)
+    expected = Series(
+        [1.0, 2.5, 4.5, 6.0],
+        index=exp_dti,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+@pytest.mark.parametrize(
+    "tz",
+    [
+        None,
+        pytest.param(
+            "UTC",
+            marks=pytest.mark.xfail(
+                condition=is_platform_windows(),
+                reason="TODO: Set ARROW_TIMEZONE_DATABASE env var in CI",
+            ),
+        ),
+    ],
+)
+def test_arrow_timestamp_resample(tz):
+    # GH 56371
+    idx = Series(date_range("2020-01-01", periods=5), dtype="timestamp[ns][pyarrow]")
+    if tz is not None:
+        idx = idx.dt.tz_localize(tz)
+    expected = Series(np.arange(5, dtype=np.float64), index=idx)
+    result = expected.resample("1D").mean()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("freq", ["1A", "2A-MAR"])
+def test_resample_A_raises(freq):
+    msg = f"Invalid frequency: {freq[1:]}"
+
+    s = Series(range(10), index=date_range("20130101", freq="D", periods=10))
+    with pytest.raises(ValueError, match=msg):
+        s.resample(freq).mean()
diff --git a/pandas/pandas/tests/resample/test_period_index.py b/pandas/pandas/tests/resample/test_period_index.py
new file mode 100644
index 0000000000000..e17529dfab00c
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_period_index.py
@@ -0,0 +1,1070 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+import re
+import warnings
+import zoneinfo
+
+import dateutil
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.ccalendar import (
+    DAYS,
+    MONTHS,
+)
+from pandas._libs.tslibs.period import IncompatibleFrequency
+from pandas.errors import InvalidIndexError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import date_range
+from pandas.core.indexes.period import (
+    Period,
+    PeriodIndex,
+    period_range,
+)
+from pandas.core.resample import _get_period_range_edges
+
+from pandas.tseries import offsets
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Resampling with a PeriodIndex is deprecated:FutureWarning"
+)
+
+
+@pytest.fixture
+def simple_period_range_series():
+    """
+    Series with period range index and random data for test purposes.
+    """
+
+    def _simple_period_range_series(start, end, freq="D"):
+        with warnings.catch_warnings():
+            # suppress Period[B] deprecation warning
+            msg = "|".join(["Period with BDay freq", r"PeriodDtype\[B\] is deprecated"])
+            warnings.filterwarnings(
+                "ignore",
+                msg,
+                category=FutureWarning,
+            )
+            rng = period_range(start, end, freq=freq)
+        return Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    return _simple_period_range_series
+
+
+class TestPeriodIndex:
+    @pytest.mark.parametrize("freq", ["2D", "1h", "2h"])
+    def test_asfreq(self, frame_or_series, freq):
+        # GH 12884, 15944
+
+        obj = frame_or_series(range(5), index=period_range("2020-01-01", periods=5))
+
+        expected = obj.to_timestamp().resample(freq).asfreq()
+        result = obj.to_timestamp().resample(freq).asfreq()
+        tm.assert_almost_equal(result, expected)
+
+        start = obj.index[0].to_timestamp(how="start")
+        end = (obj.index[-1] + obj.index.freq).to_timestamp(how="start")
+        new_index = date_range(start=start, end=end, freq=freq, inclusive="left")
+        expected = obj.to_timestamp().reindex(new_index).to_period(freq)
+
+        result = obj.resample(freq).asfreq()
+        tm.assert_almost_equal(result, expected)
+
+        result = obj.resample(freq).asfreq().to_timestamp().to_period()
+        tm.assert_almost_equal(result, expected)
+
+    def test_asfreq_fill_value(self):
+        # test for fill value during resampling, issue 3715
+
+        index = period_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+        s = Series(range(len(index)), index=index)
+        new_index = date_range(
+            s.index[0].to_timestamp(how="start"),
+            (s.index[-1]).to_timestamp(how="start"),
+            freq="1h",
+        )
+        expected = s.to_timestamp().reindex(new_index, fill_value=4.0)
+        result = s.to_timestamp().resample("1h").asfreq(fill_value=4.0)
+        tm.assert_series_equal(result, expected)
+
+        frame = s.to_frame("value")
+        new_index = date_range(
+            frame.index[0].to_timestamp(how="start"),
+            (frame.index[-1]).to_timestamp(how="start"),
+            freq="1h",
+        )
+        expected = frame.to_timestamp().reindex(new_index, fill_value=3.0)
+        result = frame.to_timestamp().resample("1h").asfreq(fill_value=3.0)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("freq", ["h", "12h", "2D", "W"])
+    @pytest.mark.parametrize("kwargs", [{"on": "date"}, {"level": "d"}])
+    def test_selection(self, freq, kwargs):
+        # This is a bug, these should be implemented
+        # GH 14008
+        index = period_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+        rng = np.arange(len(index), dtype=np.int64)
+        df = DataFrame(
+            {"date": index, "a": rng},
+            index=pd.MultiIndex.from_arrays([rng, index], names=["v", "d"]),
+        )
+        msg = (
+            "Resampling from level= or on= selection with a PeriodIndex is "
+            r"not currently supported, use \.set_index\(\.\.\.\) to "
+            "explicitly set index"
+        )
+        with pytest.raises(NotImplementedError, match=msg):
+            df.resample(freq, **kwargs)
+
+    @pytest.mark.parametrize("month", MONTHS)
+    @pytest.mark.parametrize("meth", ["ffill", "bfill"])
+    @pytest.mark.parametrize("conv", ["start", "end"])
+    @pytest.mark.parametrize(
+        ("offset", "period"), [("D", "D"), ("B", "B"), ("ME", "M"), ("QE", "Q")]
+    )
+    def test_annual_upsample_cases(
+        self, offset, period, conv, meth, month, simple_period_range_series
+    ):
+        ts = simple_period_range_series("1/1/1990", "12/31/1991", freq=f"Y-{month}")
+        warn = FutureWarning if period == "B" else None
+        msg = r"PeriodDtype\[B\] is deprecated"
+        if warn is None:
+            msg = "Resampling with a PeriodIndex is deprecated"
+            warn = FutureWarning
+        with tm.assert_produces_warning(warn, match=msg):
+            result = getattr(ts.resample(period, convention=conv), meth)()
+            expected = result.to_timestamp(period, how=conv)
+            expected = expected.asfreq(offset, meth).to_period()
+        tm.assert_series_equal(result, expected)
+
+    def test_basic_downsample(self, simple_period_range_series):
+        ts = simple_period_range_series("1/1/1990", "6/30/1995", freq="M")
+        result = ts.resample("Y-DEC").mean()
+
+        expected = ts.groupby(ts.index.year).mean()
+        expected.index = period_range("1/1/1990", "6/30/1995", freq="Y-DEC")
+        tm.assert_series_equal(result, expected)
+
+        # this is ok
+        tm.assert_series_equal(ts.resample("Y-DEC").mean(), result)
+        tm.assert_series_equal(ts.resample("Y").mean(), result)
+
+    @pytest.mark.parametrize(
+        "rule,expected_error_msg",
+        [
+            ("Y-DEC", "<YearEnd: month=12>"),
+            ("Q-MAR", "<QuarterEnd: startingMonth=3>"),
+            ("M", "<MonthEnd>"),
+            ("W-THU", "<Week: weekday=3>"),
+        ],
+    )
+    def test_not_subperiod(self, simple_period_range_series, rule, expected_error_msg):
+        # These are incompatible period rules for resampling
+        ts = simple_period_range_series("1/1/1990", "6/30/1995", freq="W-WED")
+        msg = (
+            "Frequency <Week: weekday=2> cannot be resampled to "
+            f"{expected_error_msg}, as they are not sub or super periods"
+        )
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            ts.resample(rule).mean()
+
+    @pytest.mark.parametrize("freq", ["D", "2D"])
+    def test_basic_upsample(self, freq, simple_period_range_series):
+        ts = simple_period_range_series("1/1/1990", "6/30/1995", freq="M")
+        result = ts.resample("Y-DEC").mean()
+
+        msg = "The 'convention' keyword in Series.resample is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            resampled = result.resample(freq, convention="end").ffill()
+        expected = result.to_timestamp(freq, how="end")
+        expected = expected.asfreq(freq, "ffill").to_period(freq)
+        tm.assert_series_equal(resampled, expected)
+
+    def test_upsample_with_limit(self):
+        rng = period_range("1/1/2000", periods=5, freq="Y")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), rng)
+
+        msg = "The 'convention' keyword in Series.resample is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = ts.resample("M", convention="end").ffill(limit=2)
+        expected = ts.asfreq("M").reindex(result.index, method="ffill", limit=2)
+        tm.assert_series_equal(result, expected)
+
+    def test_annual_upsample(self, simple_period_range_series):
+        ts = simple_period_range_series("1/1/1990", "12/31/1995", freq="Y-DEC")
+        df = DataFrame({"a": ts})
+        rdf = df.resample("D").ffill()
+        exp = df["a"].resample("D").ffill()
+        tm.assert_series_equal(rdf["a"], exp)
+
+    def test_annual_upsample2(self):
+        rng = period_range("2000", "2003", freq="Y-DEC")
+        ts = Series([1, 2, 3, 4], index=rng)
+
+        result = ts.resample("M").ffill()
+        ex_index = period_range("2000-01", "2003-12", freq="M")
+
+        expected = ts.asfreq("M", how="start").reindex(ex_index, method="ffill")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("month", MONTHS)
+    @pytest.mark.parametrize("convention", ["start", "end"])
+    @pytest.mark.parametrize(
+        ("offset", "period"), [("D", "D"), ("B", "B"), ("ME", "M")]
+    )
+    def test_quarterly_upsample(
+        self, month, offset, period, convention, simple_period_range_series
+    ):
+        freq = f"Q-{month}"
+        ts = simple_period_range_series("1/1/1990", "12/31/1995", freq=freq)
+        warn = FutureWarning if period == "B" else None
+        msg = r"PeriodDtype\[B\] is deprecated"
+        if warn is None:
+            msg = "Resampling with a PeriodIndex is deprecated"
+            warn = FutureWarning
+        with tm.assert_produces_warning(warn, match=msg):
+            result = ts.resample(period, convention=convention).ffill()
+            expected = result.to_timestamp(period, how=convention)
+            expected = expected.asfreq(offset, "ffill").to_period()
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("target", ["D", "B"])
+    @pytest.mark.parametrize("convention", ["start", "end"])
+    def test_monthly_upsample(self, target, convention, simple_period_range_series):
+        ts = simple_period_range_series("1/1/1990", "12/31/1995", freq="M")
+
+        warn = None if target == "D" else FutureWarning
+        msg = r"PeriodDtype\[B\] is deprecated"
+        if warn is None:
+            msg = "Resampling with a PeriodIndex is deprecated"
+            warn = FutureWarning
+        with tm.assert_produces_warning(warn, match=msg):
+            result = ts.resample(target, convention=convention).ffill()
+            expected = result.to_timestamp(target, how=convention)
+            expected = expected.asfreq(target, "ffill").to_period()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_basic(self):
+        # GH3609
+        s = Series(
+            range(100),
+            index=date_range("20130101", freq="s", periods=100, name="idx"),
+            dtype="float",
+        )
+        s[10:30] = np.nan
+        index = PeriodIndex(
+            [Period("2013-01-01 00:00", "min"), Period("2013-01-01 00:01", "min")],
+            name="idx",
+        )
+        expected = Series([34.5, 79.5], index=index)
+        result = s.to_period().resample("min").mean()
+        tm.assert_series_equal(result, expected)
+        result2 = s.resample("min").mean().to_period()
+        tm.assert_series_equal(result2, expected)
+
+    @pytest.mark.parametrize(
+        "freq,expected_vals", [("M", [31, 29, 31, 9]), ("2M", [31 + 29, 31 + 9])]
+    )
+    def test_resample_count(self, freq, expected_vals):
+        # GH12774
+        series = Series(1, index=period_range(start="2000", periods=100))
+        result = series.resample(freq).count()
+        expected_index = period_range(
+            start="2000", freq=freq, periods=len(expected_vals)
+        )
+        expected = Series(expected_vals, index=expected_index)
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_same_freq(self, resample_method):
+        # GH12770
+        series = Series(range(3), index=period_range(start="2000", periods=3, freq="M"))
+        expected = series
+
+        result = getattr(series.resample("M"), resample_method)()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_incompat_freq(self):
+        msg = (
+            "Frequency <MonthEnd> cannot be resampled to <Week: weekday=6>, "
+            "as they are not sub or super periods"
+        )
+        pi = period_range(start="2000", periods=3, freq="M")
+        ser = Series(range(3), index=pi)
+        rs = ser.resample("W")
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            # TODO: should this raise at the resample call instead of at the mean call?
+            rs.mean()
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            zoneinfo.ZoneInfo("America/Los_Angeles"),
+            dateutil.tz.gettz("America/Los_Angeles"),
+        ],
+    )
+    def test_with_local_timezone(self, tz):
+        # see gh-5430
+        local_timezone = tz
+
+        start = datetime(
+            year=2013, month=11, day=1, hour=0, minute=0, tzinfo=timezone.utc
+        )
+        # 1 day later
+        end = datetime(
+            year=2013, month=11, day=2, hour=0, minute=0, tzinfo=timezone.utc
+        )
+
+        index = date_range(start, end, freq="h", name="idx")
+
+        series = Series(1, index=index)
+        series = series.tz_convert(local_timezone)
+        msg = "Converting to PeriodArray/Index representation will drop timezone"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = series.resample("D").mean().to_period()
+
+        # Create the expected series
+        # Index is moved back a day with the timezone conversion from UTC to
+        # Pacific
+        expected_index = (
+            period_range(start=start, end=end, freq="D", name="idx") - offsets.Day()
+        )
+        expected = Series(1.0, index=expected_index)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            zoneinfo.ZoneInfo("America/Los_Angeles"),
+            dateutil.tz.gettz("America/Los_Angeles"),
+        ],
+    )
+    def test_resample_with_tz(self, tz, unit):
+        # GH 13238
+        dti = date_range("2017-01-01", periods=48, freq="h", tz=tz, unit=unit)
+        ser = Series(2, index=dti)
+        result = ser.resample("D").mean()
+        exp_dti = pd.DatetimeIndex(
+            ["2017-01-01", "2017-01-02"], tz=tz, freq="D"
+        ).as_unit(unit)
+        expected = Series(
+            2.0,
+            index=exp_dti,
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_nonexistent_time_bin_edge(self):
+        # GH 19375
+        index = date_range("2017-03-12", "2017-03-12 1:45:00", freq="15min")
+        s = Series(np.zeros(len(index)), index=index)
+        expected = s.tz_localize("US/Pacific")
+        expected.index = pd.DatetimeIndex(expected.index, freq="900s")
+        result = expected.resample("900s").mean()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_nonexistent_time_bin_edge2(self):
+        # GH 23742
+        index = date_range(start="2017-10-10", end="2017-10-20", freq="1h")
+        index = index.tz_localize("UTC").tz_convert("America/Sao_Paulo")
+        df = DataFrame(data=list(range(len(index))), index=index)
+        result = df.groupby(pd.Grouper(freq="1D")).count()
+        expected = date_range(
+            start="2017-10-09",
+            end="2017-10-20",
+            freq="D",
+            tz="America/Sao_Paulo",
+            nonexistent="shift_forward",
+            inclusive="left",
+        )
+        tm.assert_index_equal(result.index, expected)
+
+    def test_resample_ambiguous_time_bin_edge(self):
+        # GH 10117
+        idx = date_range(
+            "2014-10-25 22:00:00",
+            "2014-10-26 00:30:00",
+            freq="30min",
+            tz="Europe/London",
+        )
+        expected = Series(np.zeros(len(idx)), index=idx)
+        result = expected.resample("30min").mean()
+        tm.assert_series_equal(result, expected)
+
+    def test_fill_method_and_how_upsample(self):
+        # GH2073
+        s = Series(
+            np.arange(9, dtype="int64"),
+            index=date_range("2010-01-01", periods=9, freq="QE"),
+        )
+        last = s.resample("ME").ffill()
+        both = s.resample("ME").ffill().resample("ME").last().astype("int64")
+        tm.assert_series_equal(last, both)
+
+    @pytest.mark.parametrize("day", DAYS)
+    @pytest.mark.parametrize("target", ["D", "B"])
+    @pytest.mark.parametrize("convention", ["start", "end"])
+    def test_weekly_upsample(self, day, target, convention, simple_period_range_series):
+        freq = f"W-{day}"
+        ts = simple_period_range_series("1/1/1990", "12/31/1995", freq=freq)
+
+        warn = None if target == "D" else FutureWarning
+        msg = r"PeriodDtype\[B\] is deprecated"
+        if warn is None:
+            msg = "Resampling with a PeriodIndex is deprecated"
+            warn = FutureWarning
+        with tm.assert_produces_warning(warn, match=msg):
+            result = ts.resample(target, convention=convention).ffill()
+            expected = result.to_timestamp(target, how=convention)
+            expected = expected.asfreq(target, "ffill").to_period()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_to_timestamps(self, simple_period_range_series):
+        ts = simple_period_range_series("1/1/1990", "12/31/1995", freq="M")
+
+        result = ts.resample("Y-DEC").mean().to_timestamp()
+        expected = ts.resample("Y-DEC").mean().to_timestamp(how="start")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("month", MONTHS)
+    def test_resample_to_quarterly(self, simple_period_range_series, month):
+        ts = simple_period_range_series("1990", "1992", freq=f"Y-{month}")
+        quar_ts = ts.resample(f"Q-{month}").ffill()
+
+        stamps = ts.to_timestamp("D", how="start")
+        qdates = period_range(
+            ts.index[0].asfreq("D", "start"),
+            ts.index[-1].asfreq("D", "end"),
+            freq=f"Q-{month}",
+        )
+
+        expected = stamps.reindex(qdates.to_timestamp("D", "s"), method="ffill")
+        expected.index = qdates
+
+        tm.assert_series_equal(quar_ts, expected)
+
+    @pytest.mark.parametrize("how", ["start", "end"])
+    def test_resample_to_quarterly_start_end(self, simple_period_range_series, how):
+        # conforms, but different month
+        ts = simple_period_range_series("1990", "1992", freq="Y-JUN")
+        msg = "The 'convention' keyword in Series.resample is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = ts.resample("Q-MAR", convention=how).ffill()
+        expected = ts.asfreq("Q-MAR", how=how)
+        expected = expected.reindex(result.index, method="ffill")
+
+        # FIXME: don't leave commented-out
+        # .to_timestamp('D')
+        # expected = expected.resample('Q-MAR').ffill()
+
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_fill_missing(self):
+        rng = PeriodIndex([2000, 2005, 2007, 2009], freq="Y")
+
+        s = Series(np.random.default_rng(2).standard_normal(4), index=rng)
+
+        stamps = s.to_timestamp()
+        filled = s.resample("Y").ffill()
+        expected = stamps.resample("YE").ffill().to_period("Y")
+        tm.assert_series_equal(filled, expected)
+
+    def test_cant_fill_missing_dups(self):
+        rng = PeriodIndex([2000, 2005, 2005, 2007, 2007], freq="Y")
+        s = Series(np.random.default_rng(2).standard_normal(5), index=rng)
+        msg = "Reindexing only valid with uniquely valued Index objects"
+        with pytest.raises(InvalidIndexError, match=msg):
+            s.resample("Y").ffill()
+
+    def test_resample_5minute(self):
+        rng = period_range("1/1/2000", "1/5/2000", freq="min")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+        expected = ts.to_timestamp().resample("5min").mean()
+        result = ts.resample("5min").mean().to_timestamp()
+        tm.assert_series_equal(result, expected)
+
+        expected = expected.to_period("5min")
+        result = ts.resample("5min").mean()
+        tm.assert_series_equal(result, expected)
+        result = ts.resample("5min").mean().to_timestamp().to_period()
+        tm.assert_series_equal(result, expected)
+
+    def test_upsample_daily_business_daily(self, simple_period_range_series):
+        ts = simple_period_range_series("1/1/2000", "2/1/2000", freq="B")
+
+        result = ts.resample("D").asfreq()
+        expected = ts.asfreq("D").reindex(period_range("1/3/2000", "2/1/2000"))
+        tm.assert_series_equal(result, expected)
+
+        ts = simple_period_range_series("1/1/2000", "2/1/2000")
+        msg = "The 'convention' keyword in Series.resample is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = ts.resample("h", convention="s").asfreq()
+        exp_rng = period_range("1/1/2000", "2/1/2000 23:00", freq="h")
+        expected = ts.asfreq("h", how="s").reindex(exp_rng)
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_irregular_sparse(self):
+        dr = date_range(start="1/1/2012", freq="5min", periods=1000)
+        s = Series(np.array(100), index=dr)
+        # subset the data.
+        subset = s[:"2012-01-04 06:55"]
+
+        result = subset.resample("10min").apply(len)
+        expected = s.resample("10min").apply(len).loc[result.index]
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_weekly_all_na(self):
+        rng = date_range("1/1/2000", periods=10, freq="W-WED")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        result = ts.resample("W-THU").asfreq()
+
+        assert result.isna().all()
+
+        result = ts.resample("W-THU").asfreq().ffill()[:-1]
+        expected = ts.asfreq("W-THU").ffill()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_tz_localized(self, unit):
+        dr = date_range(start="2012-4-13", end="2012-5-1", unit=unit)
+        ts = Series(range(len(dr)), index=dr)
+
+        ts_utc = ts.tz_localize("UTC")
+        ts_local = ts_utc.tz_convert("America/Los_Angeles")
+
+        result = ts_local.resample("W").mean()
+
+        ts_local_naive = ts_local.copy()
+        ts_local_naive.index = ts_local_naive.index.tz_localize(None)
+
+        exp = ts_local_naive.resample("W").mean().tz_localize("America/Los_Angeles")
+        exp.index = pd.DatetimeIndex(exp.index, freq="W")
+
+        tm.assert_series_equal(result, exp)
+
+        # it works
+        result = ts_local.resample("D").mean()
+
+    def test_resample_tz_localized2(self):
+        # #2245
+        idx = date_range(
+            "2001-09-20 15:59", "2001-09-20 16:00", freq="min", tz="Australia/Sydney"
+        )
+        s = Series([1, 2], index=idx)
+
+        result = s.resample("D", closed="right", label="right").mean()
+        ex_index = date_range("2001-09-21", periods=1, freq="D", tz="Australia/Sydney")
+        expected = Series([1.5], index=ex_index)
+
+        tm.assert_series_equal(result, expected)
+
+        # for good measure
+        msg = "Converting to PeriodArray/Index representation will drop timezone "
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = s.resample("D").mean().to_period()
+        ex_index = period_range("2001-09-20", periods=1, freq="D")
+        expected = Series([1.5], index=ex_index)
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_tz_localized3(self):
+        # GH 6397
+        # comparing an offset that doesn't propagate tz's
+        rng = date_range("1/1/2011", periods=20000, freq="h")
+        rng = rng.tz_localize("EST")
+        ts = DataFrame(index=rng)
+        ts["first"] = np.random.default_rng(2).standard_normal(len(rng))
+        ts["second"] = np.cumsum(np.random.default_rng(2).standard_normal(len(rng)))
+        expected = DataFrame(
+            {
+                "first": ts.resample("YE").sum()["first"],
+                "second": ts.resample("YE").mean()["second"],
+            },
+            columns=["first", "second"],
+        )
+        result = (
+            ts.resample("YE")
+            .agg({"first": "sum", "second": "mean"})
+            .reindex(columns=["first", "second"])
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_closed_left_corner(self):
+        # #1465
+        s = Series(
+            np.random.default_rng(2).standard_normal(21),
+            index=date_range(start="1/1/2012 9:30", freq="1min", periods=21),
+        )
+        s.iloc[0] = np.nan
+
+        result = s.resample("10min", closed="left", label="right").mean()
+        exp = s[1:].resample("10min", closed="left", label="right").mean()
+        tm.assert_series_equal(result, exp)
+
+        result = s.resample("10min", closed="left", label="left").mean()
+        exp = s[1:].resample("10min", closed="left", label="left").mean()
+
+        ex_index = date_range(start="1/1/2012 9:30", freq="10min", periods=3)
+
+        tm.assert_index_equal(result.index, ex_index)
+        tm.assert_series_equal(result, exp)
+
+    def test_quarterly_resampling(self):
+        rng = period_range("2000Q1", periods=10, freq="Q-DEC")
+        ts = Series(np.arange(10), index=rng)
+
+        result = ts.resample("Y").mean()
+        exp = ts.to_timestamp().resample("YE").mean().to_period()
+        tm.assert_series_equal(result, exp)
+
+    def test_resample_weekly_bug_1726(self):
+        # 8/6/12 is a Monday
+        ind = date_range(start="8/6/2012", end="8/26/2012", freq="D")
+        n = len(ind)
+        data = [[x] * 5 for x in range(n)]
+        df = DataFrame(data, columns=["open", "high", "low", "close", "vol"], index=ind)
+
+        # it works!
+        df.resample("W-MON", closed="left", label="left").first()
+
+    def test_resample_with_dst_time_change(self):
+        # GH 15549
+        index = (
+            pd.DatetimeIndex([1457537600000000000, 1458059600000000000])
+            .tz_localize("UTC")
+            .tz_convert("America/Chicago")
+        )
+        df = DataFrame([1, 2], index=index)
+        result = df.resample("12h", closed="right", label="right").last().ffill()
+
+        expected_index_values = [
+            "2016-03-09 12:00:00-06:00",
+            "2016-03-10 00:00:00-06:00",
+            "2016-03-10 12:00:00-06:00",
+            "2016-03-11 00:00:00-06:00",
+            "2016-03-11 12:00:00-06:00",
+            "2016-03-12 00:00:00-06:00",
+            "2016-03-12 12:00:00-06:00",
+            "2016-03-13 00:00:00-06:00",
+            "2016-03-13 13:00:00-05:00",
+            "2016-03-14 01:00:00-05:00",
+            "2016-03-14 13:00:00-05:00",
+            "2016-03-15 01:00:00-05:00",
+            "2016-03-15 13:00:00-05:00",
+        ]
+        index = (
+            pd.to_datetime(expected_index_values, utc=True)
+            .tz_convert("America/Chicago")
+            .as_unit(index.unit)
+        )
+        index = pd.DatetimeIndex(index, freq="12h")
+        expected = DataFrame(
+            [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0],
+            index=index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_resample_bms_2752(self):
+        # GH2753
+        timeseries = Series(
+            index=pd.bdate_range("20000101", "20000201"), dtype=np.float64
+        )
+        res1 = timeseries.resample("BMS").mean()
+        res2 = timeseries.resample("BMS").mean().resample("B").mean()
+        assert res1.index[0] == Timestamp("20000103")
+        assert res1.index[0] == res2.index[0]
+
+    @pytest.mark.xfail(reason="Commented out for more than 3 years. Should this work?")
+    def test_monthly_convention_span(self):
+        rng = period_range("2000-01", periods=3, freq="ME")
+        ts = Series(np.arange(3), index=rng)
+
+        # hacky way to get same thing
+        exp_index = period_range("2000-01-01", "2000-03-31", freq="D")
+        expected = ts.asfreq("D", how="end").reindex(exp_index)
+        expected = expected.fillna(method="bfill")
+
+        result = ts.resample("D").mean()
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "from_freq, to_freq", [("D", "ME"), ("QE", "YE"), ("ME", "QE"), ("D", "W")]
+    )
+    def test_default_right_closed_label(self, from_freq, to_freq):
+        idx = date_range(start="8/15/2012", periods=100, freq=from_freq)
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(idx), 2)), idx)
+
+        resampled = df.resample(to_freq).mean()
+        tm.assert_frame_equal(
+            resampled, df.resample(to_freq, closed="right", label="right").mean()
+        )
+
+    @pytest.mark.parametrize(
+        "from_freq, to_freq",
+        [("D", "MS"), ("QE", "YS"), ("ME", "QS"), ("h", "D"), ("min", "h")],
+    )
+    def test_default_left_closed_label(self, from_freq, to_freq):
+        idx = date_range(start="8/15/2012", periods=100, freq=from_freq)
+        df = DataFrame(np.random.default_rng(2).standard_normal((len(idx), 2)), idx)
+
+        resampled = df.resample(to_freq).mean()
+        tm.assert_frame_equal(
+            resampled, df.resample(to_freq, closed="left", label="left").mean()
+        )
+
+    def test_all_values_single_bin(self):
+        # GH#2070
+        index = period_range(start="2012-01-01", end="2012-12-31", freq="M")
+        ser = Series(np.random.default_rng(2).standard_normal(len(index)), index=index)
+
+        result = ser.resample("Y").mean()
+        tm.assert_almost_equal(result.iloc[0], ser.mean())
+
+    def test_evenly_divisible_with_no_extra_bins(self):
+        # GH#4076
+        # when the frequency is evenly divisible, sometimes extra bins
+
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((9, 3)),
+            index=date_range("2000-1-1", periods=9),
+        )
+        result = df.resample("5D").mean()
+        expected = pd.concat([df.iloc[0:5].mean(), df.iloc[5:].mean()], axis=1).T
+        expected.index = pd.DatetimeIndex(
+            [Timestamp("2000-1-1"), Timestamp("2000-1-6")], dtype="M8[ns]", freq="5D"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_evenly_divisible_with_no_extra_bins2(self):
+        index = date_range(start="2001-5-4", periods=28)
+        df = DataFrame(
+            [
+                {
+                    "REST_KEY": 1,
+                    "DLY_TRN_QT": 80,
+                    "DLY_SLS_AMT": 90,
+                    "COOP_DLY_TRN_QT": 30,
+                    "COOP_DLY_SLS_AMT": 20,
+                }
+            ]
+            * 28
+            + [
+                {
+                    "REST_KEY": 2,
+                    "DLY_TRN_QT": 70,
+                    "DLY_SLS_AMT": 10,
+                    "COOP_DLY_TRN_QT": 50,
+                    "COOP_DLY_SLS_AMT": 20,
+                }
+            ]
+            * 28,
+            index=index.append(index),
+        ).sort_index()
+
+        index = date_range("2001-5-4", periods=4, freq="7D")
+        expected = DataFrame(
+            [
+                {
+                    "REST_KEY": 14,
+                    "DLY_TRN_QT": 14,
+                    "DLY_SLS_AMT": 14,
+                    "COOP_DLY_TRN_QT": 14,
+                    "COOP_DLY_SLS_AMT": 14,
+                }
+            ]
+            * 4,
+            index=index,
+        )
+        result = df.resample("7D").count()
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            [
+                {
+                    "REST_KEY": 21,
+                    "DLY_TRN_QT": 1050,
+                    "DLY_SLS_AMT": 700,
+                    "COOP_DLY_TRN_QT": 560,
+                    "COOP_DLY_SLS_AMT": 280,
+                }
+            ]
+            * 4,
+            index=index,
+        )
+        result = df.resample("7D").sum()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("freq, period_mult", [("h", 24), ("12h", 2)])
+    def test_upsampling_ohlc(self, freq, period_mult):
+        # GH 13083
+        pi = period_range(start="2000", freq="D", periods=10)
+        s = Series(range(len(pi)), index=pi)
+        expected = s.to_timestamp().resample(freq).ohlc().to_period(freq)
+
+        # timestamp-based resampling doesn't include all sub-periods
+        # of the last original period, so extend accordingly:
+        new_index = period_range(start="2000", freq=freq, periods=period_mult * len(pi))
+        expected = expected.reindex(new_index)
+        result = s.resample(freq).ohlc()
+        tm.assert_frame_equal(result, expected)
+
+        result = s.resample(freq).ohlc().to_timestamp().to_period()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "periods, values",
+        [
+            (
+                [
+                    pd.NaT,
+                    "1970-01-01 00:00:00",
+                    pd.NaT,
+                    "1970-01-01 00:00:02",
+                    "1970-01-01 00:00:03",
+                ],
+                [2, 3, 5, 7, 11],
+            ),
+            (
+                [
+                    pd.NaT,
+                    pd.NaT,
+                    "1970-01-01 00:00:00",
+                    pd.NaT,
+                    pd.NaT,
+                    pd.NaT,
+                    "1970-01-01 00:00:02",
+                    "1970-01-01 00:00:03",
+                    pd.NaT,
+                    pd.NaT,
+                ],
+                [1, 2, 3, 5, 6, 8, 7, 11, 12, 13],
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "freq, expected_values",
+        [
+            ("1s", [3, np.nan, 7, 11]),
+            ("2s", [3, (7 + 11) / 2]),
+            ("3s", [(3 + 7) / 2, 11]),
+        ],
+    )
+    def test_resample_with_nat(self, periods, values, freq, expected_values):
+        # GH 13224
+        index = PeriodIndex(periods, freq="s")
+        frame = DataFrame(values, index=index)
+
+        expected_index = period_range(
+            "1970-01-01 00:00:00", periods=len(expected_values), freq=freq
+        )
+        expected = DataFrame(expected_values, index=expected_index)
+        msg = "Resampling with a PeriodIndex is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            rs = frame.resample(freq)
+        result = rs.mean()
+        tm.assert_frame_equal(result, expected)
+
+    def test_resample_with_only_nat(self):
+        # GH 13224
+        pi = PeriodIndex([pd.NaT] * 3, freq="s")
+        frame = DataFrame([2, 3, 5], index=pi, columns=["a"])
+        expected_index = PeriodIndex(data=[], freq=pi.freq)
+        expected = DataFrame(index=expected_index, columns=["a"], dtype="float64")
+        result = frame.resample("1s").mean()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "start,end,start_freq,end_freq,offset",
+        [
+            ("19910905", "19910909 03:00", "h", "24h", "10h"),
+            ("19910905", "19910909 12:00", "h", "24h", "10h"),
+            ("19910905", "19910909 23:00", "h", "24h", "10h"),
+            ("19910905 10:00", "19910909", "h", "24h", "10h"),
+            ("19910905 10:00", "19910909 10:00", "h", "24h", "10h"),
+            ("19910905", "19910909 10:00", "h", "24h", "10h"),
+            ("19910905 12:00", "19910909", "h", "24h", "10h"),
+            ("19910905 12:00", "19910909 03:00", "h", "24h", "10h"),
+            ("19910905 12:00", "19910909 12:00", "h", "24h", "10h"),
+            ("19910905 12:00", "19910909 12:00", "h", "24h", "34h"),
+            ("19910905 12:00", "19910909 12:00", "h", "17h", "10h"),
+            ("19910905 12:00", "19910909 12:00", "h", "17h", "3h"),
+            ("19910905", "19910913 06:00", "2h", "24h", "10h"),
+            ("19910905", "19910905 01:39", "Min", "5Min", "3Min"),
+            ("19910905", "19910905 03:18", "2Min", "5Min", "3Min"),
+        ],
+    )
+    def test_resample_with_offset(self, start, end, start_freq, end_freq, offset):
+        # GH 23882 & 31809
+        pi = period_range(start, end, freq=start_freq)
+        ser = Series(np.arange(len(pi)), index=pi)
+        msg = "Resampling with a PeriodIndex is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            rs = ser.resample(end_freq, offset=offset)
+        result = rs.mean()
+        result = result.to_timestamp(end_freq)
+
+        expected = ser.to_timestamp().resample(end_freq, offset=offset).mean()
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_with_offset_month(self):
+        # GH 23882 & 31809
+        pi = period_range("19910905 12:00", "19910909 1:00", freq="h")
+        ser = Series(np.arange(len(pi)), index=pi)
+        msg = "Resampling with a PeriodIndex is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            rs = ser.resample("M", offset="3h")
+        result = rs.mean()
+        result = result.to_timestamp("M")
+        expected = ser.to_timestamp().resample("ME", offset="3h").mean()
+        # TODO: is non-tick the relevant characteristic? (GH 33815)
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "first,last,freq,freq_to_offset,exp_first,exp_last",
+        [
+            ("19910905", "19920406", "D", "D", "19910905", "19920406"),
+            ("19910905 00:00", "19920406 06:00", "D", "D", "19910905", "19920406"),
+            (
+                "19910905 06:00",
+                "19920406 06:00",
+                "h",
+                "h",
+                "19910905 06:00",
+                "19920406 06:00",
+            ),
+            ("19910906", "19920406", "M", "ME", "1991-09", "1992-04"),
+            ("19910831", "19920430", "M", "ME", "1991-08", "1992-04"),
+            ("1991-08", "1992-04", "M", "ME", "1991-08", "1992-04"),
+        ],
+    )
+    def test_get_period_range_edges(
+        self, first, last, freq, freq_to_offset, exp_first, exp_last
+    ):
+        first = Period(first)
+        last = Period(last)
+
+        exp_first = Period(exp_first, freq=freq)
+        exp_last = Period(exp_last, freq=freq)
+
+        freq = pd.tseries.frequencies.to_offset(freq_to_offset)
+        result = _get_period_range_edges(first, last, freq)
+        expected = (exp_first, exp_last)
+        assert result == expected
+
+    def test_sum_min_count(self):
+        # GH 19974
+        index = date_range(start="2018", freq="ME", periods=6)
+        data = np.ones(6)
+        data[3:6] = np.nan
+        s = Series(data, index).to_period()
+        msg = "Resampling with a PeriodIndex is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            rs = s.resample("Q")
+        result = rs.sum(min_count=1)
+        expected = Series(
+            [3.0, np.nan], index=PeriodIndex(["2018Q1", "2018Q2"], freq="Q-DEC")
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_resample_t_l_deprecated(self):
+        # GH#52536
+        msg_t = "Invalid frequency: T"
+        msg_l = "Invalid frequency: L"
+
+        with pytest.raises(ValueError, match=msg_l):
+            period_range(
+                "2020-01-01 00:00:00 00:00", "2020-01-01 00:00:00 00:01", freq="L"
+            )
+        rng_l = period_range(
+            "2020-01-01 00:00:00 00:00", "2020-01-01 00:00:00 00:01", freq="ms"
+        )
+        ser = Series(np.arange(len(rng_l)), index=rng_l)
+
+        with pytest.raises(ValueError, match=msg_t):
+            ser.resample("T").mean()
+
+    @pytest.mark.parametrize(
+        "freq, freq_depr, freq_depr_res",
+        [
+            ("2Q", "2q", "2y"),
+            ("2M", "2m", "2q"),
+        ],
+    )
+    def test_resample_lowercase_frequency_raises(self, freq, freq_depr, freq_depr_res):
+        msg = f"Invalid frequency: {freq_depr}"
+        with pytest.raises(ValueError, match=msg):
+            period_range("2020-01-01", "2020-08-01", freq=freq_depr)
+
+        msg = f"Invalid frequency: {freq_depr_res}"
+        rng = period_range("2020-01-01", "2020-08-01", freq=freq)
+        ser = Series(np.arange(len(rng)), index=rng)
+        with pytest.raises(ValueError, match=msg):
+            ser.resample(freq_depr_res).mean()
+
+    @pytest.mark.parametrize(
+        "offset",
+        [
+            offsets.MonthBegin(),
+            offsets.BYearBegin(2),
+            offsets.BusinessHour(2),
+        ],
+    )
+    def test_asfreq_invalid_period_offset(self, offset, frame_or_series):
+        # GH#55785
+        msg = re.escape(f"{offset} is not supported as period frequency")
+
+        obj = frame_or_series(range(5), index=period_range("2020-01-01", periods=5))
+        with pytest.raises(ValueError, match=msg):
+            obj.asfreq(freq=offset)
+
+
+@pytest.mark.parametrize(
+    "freq",
+    [
+        ("2ME"),
+        ("2QE"),
+        ("2QE-FEB"),
+        ("2YE"),
+        ("2YE-MAR"),
+        ("2me"),
+        ("2qe"),
+        ("2ye-mar"),
+    ],
+)
+def test_resample_frequency_ME_QE_YE_raises(frame_or_series, freq):
+    # GH#9586
+    msg = f"{freq[1:]} is not supported as period frequency"
+
+    obj = frame_or_series(range(5), index=period_range("2020-01-01", periods=5))
+    msg = f"Invalid frequency: {freq}"
+    with pytest.raises(ValueError, match=msg):
+        obj.resample(freq)
+
+
+def test_corner_cases_period(simple_period_range_series):
+    # miscellaneous test coverage
+    len0pts = simple_period_range_series("2007-01", "2010-05", freq="M")[:0]
+    # it works
+    msg = "Resampling with a PeriodIndex is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = len0pts.resample("Y-DEC").mean()
+    assert len(result) == 0
+
+
+@pytest.mark.parametrize("freq", ["2BME", "2CBME", "2SME", "2BQE-FEB", "2BYE-MAR"])
+def test_resample_frequency_invalid_freq(frame_or_series, freq):
+    # GH#9586
+    msg = f"Invalid frequency: {freq}"
+
+    obj = frame_or_series(range(5), index=period_range("2020-01-01", periods=5))
+    with pytest.raises(ValueError, match=msg):
+        obj.resample(freq)
diff --git a/pandas/pandas/tests/resample/test_resample_api.py b/pandas/pandas/tests/resample/test_resample_api.py
new file mode 100644
index 0000000000000..da1774cf22587
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_resample_api.py
@@ -0,0 +1,989 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    NamedAgg,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import date_range
+
+
+@pytest.fixture
+def dti():
+    return date_range(start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="Min")
+
+
+@pytest.fixture
+def _test_series(dti):
+    return Series(np.random.default_rng(2).random(len(dti)), dti)
+
+
+@pytest.fixture
+def test_frame(dti, _test_series):
+    return DataFrame({"A": _test_series, "B": _test_series, "C": np.arange(len(dti))})
+
+
+def test_str(_test_series):
+    r = _test_series.resample("h")
+    assert (
+        "DatetimeIndexResampler [freq=<Hour>, closed=left, "
+        "label=left, convention=start, origin=start_day]" in str(r)
+    )
+
+    r = _test_series.resample("h", origin="2000-01-01")
+    assert (
+        "DatetimeIndexResampler [freq=<Hour>, closed=left, "
+        "label=left, convention=start, origin=2000-01-01 00:00:00]" in str(r)
+    )
+
+
+def test_api(_test_series):
+    r = _test_series.resample("h")
+    result = r.mean()
+    assert isinstance(result, Series)
+    assert len(result) == 217
+
+    r = _test_series.to_frame().resample("h")
+    result = r.mean()
+    assert isinstance(result, DataFrame)
+    assert len(result) == 217
+
+
+def test_groupby_resample_api():
+    # GH 12448
+    # .groupby(...).resample(...) hitting warnings
+    # when appropriate
+    df = DataFrame(
+        {
+            "date": date_range(start="2016-01-01", periods=4, freq="W"),
+            "group": [1, 1, 2, 2],
+            "val": [5, 6, 7, 8],
+        }
+    ).set_index("date")
+
+    # replication step
+    i = (
+        date_range("2016-01-03", periods=8).tolist()
+        + date_range("2016-01-17", periods=8).tolist()
+    )
+    index = pd.MultiIndex.from_arrays([[1] * 8 + [2] * 8, i], names=["group", "date"])
+    expected = DataFrame({"val": [5] * 7 + [6] + [7] * 7 + [8]}, index=index)
+    result = df.groupby("group").apply(lambda x: x.resample("1D").ffill())[["val"]]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_on_api():
+    # GH 15021
+    # .groupby(...).resample(on=...) results in an unexpected
+    # keyword warning.
+    df = DataFrame(
+        {
+            "key": ["A", "B"] * 5,
+            "dates": date_range("2016-01-01", periods=10),
+            "values": np.random.default_rng(2).standard_normal(10),
+        }
+    )
+
+    expected = df.set_index("dates").groupby("key").resample("D").mean()
+    result = df.groupby("key").resample("D", on="dates").mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_group_keys():
+    df = DataFrame({"A": 1, "B": 2}, index=date_range("2000", periods=10))
+    expected = df.copy()
+
+    # group_keys=False
+    g = df.resample("5D", group_keys=False)
+    result = g.apply(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+
+    # group_keys defaults to False
+    g = df.resample("5D")
+    result = g.apply(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+
+    # group_keys=True
+    expected.index = pd.MultiIndex.from_arrays(
+        [
+            pd.to_datetime(["2000-01-01", "2000-01-06"]).as_unit("ns").repeat(5),
+            expected.index,
+        ]
+    )
+    g = df.resample("5D", group_keys=True)
+    result = g.apply(lambda x: x)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_pipe(test_frame, _test_series):
+    # GH17905
+
+    # series
+    r = _test_series.resample("h")
+    expected = r.max() - r.mean()
+    result = r.pipe(lambda x: x.max() - x.mean())
+    tm.assert_series_equal(result, expected)
+
+    # dataframe
+    r = test_frame.resample("h")
+    expected = r.max() - r.mean()
+    result = r.pipe(lambda x: x.max() - x.mean())
+    tm.assert_frame_equal(result, expected)
+
+
+def test_getitem(test_frame):
+    r = test_frame.resample("h")
+    tm.assert_index_equal(r._selected_obj.columns, test_frame.columns)
+
+    r = test_frame.resample("h")["B"]
+    assert r._selected_obj.name == test_frame.columns[1]
+
+    # technically this is allowed
+    r = test_frame.resample("h")["A", "B"]
+    tm.assert_index_equal(r._selected_obj.columns, test_frame.columns[[0, 1]])
+
+    r = test_frame.resample("h")["A", "B"]
+    tm.assert_index_equal(r._selected_obj.columns, test_frame.columns[[0, 1]])
+
+
+@pytest.mark.parametrize("key", [["D"], ["A", "D"]])
+def test_select_bad_cols(key, test_frame):
+    g = test_frame.resample("h")
+    # 'A' should not be referenced as a bad column...
+    # will have to rethink regex if you change message!
+    msg = r"^\"Columns not found: 'D'\"$"
+    with pytest.raises(KeyError, match=msg):
+        g[key]
+
+
+def test_attribute_access(test_frame):
+    r = test_frame.resample("h")
+    tm.assert_series_equal(r.A.sum(), r["A"].sum())
+
+
+@pytest.mark.parametrize("attr", ["groups", "ngroups", "indices"])
+def test_api_compat_before_use(attr):
+    # make sure that we are setting the binner
+    # on these attributes
+    rng = date_range("1/1/2012", periods=100, freq="s")
+    ts = Series(np.arange(len(rng)), index=rng)
+    rs = ts.resample("30s")
+
+    # before use
+    getattr(rs, attr)
+
+    # after grouper is initialized is ok
+    rs.mean()
+    getattr(rs, attr)
+
+
+def tests_raises_on_nuisance(test_frame, using_infer_string):
+    df = test_frame
+    df["D"] = "foo"
+    r = df.resample("h")
+    result = r[["A", "B"]].mean()
+    expected = pd.concat([r.A.mean(), r.B.mean()], axis=1)
+    tm.assert_frame_equal(result, expected)
+
+    expected = r[["A", "B", "C"]].mean()
+    msg = re.escape("agg function failed [how->mean,dtype->")
+    if using_infer_string:
+        msg = "dtype 'str' does not support operation 'mean'"
+    with pytest.raises(TypeError, match=msg):
+        r.mean()
+    result = r.mean(numeric_only=True)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_downsample_but_actually_upsampling():
+    # this is reindex / asfreq
+    rng = date_range("1/1/2012", periods=100, freq="s")
+    ts = Series(np.arange(len(rng), dtype="int64"), index=rng)
+    result = ts.resample("20s").asfreq()
+    expected = Series(
+        [0, 20, 40, 60, 80],
+        index=date_range("2012-01-01 00:00:00", freq="20s", periods=5),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_combined_up_downsampling_of_irregular():
+    # since we are really doing an operation like this
+    # ts2.resample('2s').mean().ffill()
+    # preserve these semantics
+
+    rng = date_range("1/1/2012", periods=100, freq="s")
+    ts = Series(np.arange(len(rng)), index=rng)
+    ts2 = ts.iloc[[0, 1, 2, 3, 5, 7, 11, 15, 16, 25, 30]]
+
+    result = ts2.resample("2s").mean().ffill()
+    expected = Series(
+        [
+            0.5,
+            2.5,
+            5.0,
+            7.0,
+            7.0,
+            11.0,
+            11.0,
+            15.0,
+            16.0,
+            16.0,
+            16.0,
+            16.0,
+            25.0,
+            25.0,
+            25.0,
+            30.0,
+        ],
+        index=pd.DatetimeIndex(
+            [
+                "2012-01-01 00:00:00",
+                "2012-01-01 00:00:02",
+                "2012-01-01 00:00:04",
+                "2012-01-01 00:00:06",
+                "2012-01-01 00:00:08",
+                "2012-01-01 00:00:10",
+                "2012-01-01 00:00:12",
+                "2012-01-01 00:00:14",
+                "2012-01-01 00:00:16",
+                "2012-01-01 00:00:18",
+                "2012-01-01 00:00:20",
+                "2012-01-01 00:00:22",
+                "2012-01-01 00:00:24",
+                "2012-01-01 00:00:26",
+                "2012-01-01 00:00:28",
+                "2012-01-01 00:00:30",
+            ],
+            dtype="datetime64[ns]",
+            freq="2s",
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_transform_series(_test_series):
+    r = _test_series.resample("20min")
+    expected = _test_series.groupby(pd.Grouper(freq="20min")).transform("mean")
+    result = r.transform("mean")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("on", [None, "date"])
+def test_transform_frame(on):
+    # GH#47079
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+    index.name = "date"
+    df = DataFrame(
+        np.random.default_rng(2).random((10, 2)), columns=list("AB"), index=index
+    )
+    expected = df.groupby(pd.Grouper(freq="20min")).transform("mean")
+    if on == "date":
+        # Move date to being a column; result will then have a RangeIndex
+        expected = expected.reset_index(drop=True)
+        df = df.reset_index()
+
+    r = df.resample("20min", on=on)
+    result = r.transform("mean")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        lambda x: x.resample("20min", group_keys=False),
+        lambda x: x.groupby(pd.Grouper(freq="20min"), group_keys=False),
+    ],
+    ids=["resample", "groupby"],
+)
+def test_apply_without_aggregation(func, _test_series):
+    # both resample and groupby should work w/o aggregation
+    t = func(_test_series)
+    result = t.apply(lambda x: x)
+    tm.assert_series_equal(result, _test_series)
+
+
+def test_apply_without_aggregation2(_test_series):
+    grouped = _test_series.to_frame(name="foo").resample("20min", group_keys=False)
+    result = grouped["foo"].apply(lambda x: x)
+    tm.assert_series_equal(result, _test_series.rename("foo"))
+
+
+def test_agg_consistency():
+    # make sure that we are consistent across
+    # similar aggregations with and w/o selection list
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((1000, 3)),
+        index=date_range("1/1/2012", freq="s", periods=1000),
+        columns=["A", "B", "C"],
+    )
+
+    r = df.resample("3min")
+
+    msg = r"Label\(s\) \['r1', 'r2'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        r.agg({"r1": "mean", "r2": "sum"})
+
+
+def test_agg_consistency_int_str_column_mix():
+    # GH#39025
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((1000, 2)),
+        index=date_range("1/1/2012", freq="s", periods=1000),
+        columns=[1, "a"],
+    )
+
+    r = df.resample("3min")
+
+    msg = r"Label\(s\) \[2, 'b'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        r.agg({2: "mean", "b": "sum"})
+
+
+# TODO(GH#14008): once GH 14008 is fixed, move these tests into
+# `Base` test class
+
+
+@pytest.fixture
+def index():
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+    index.name = "date"
+    return index
+
+
+@pytest.fixture
+def df(index):
+    frame = DataFrame(
+        np.random.default_rng(2).random((10, 2)), columns=list("AB"), index=index
+    )
+    return frame
+
+
+@pytest.fixture
+def df_col(df):
+    return df.reset_index()
+
+
+@pytest.fixture
+def df_mult(df_col, index):
+    df_mult = df_col.copy()
+    df_mult.index = pd.MultiIndex.from_arrays(
+        [range(10), index], names=["index", "date"]
+    )
+    return df_mult
+
+
+@pytest.fixture
+def a_mean(df):
+    return df.resample("2D")["A"].mean()
+
+
+@pytest.fixture
+def a_std(df):
+    return df.resample("2D")["A"].std()
+
+
+@pytest.fixture
+def a_sum(df):
+    return df.resample("2D")["A"].sum()
+
+
+@pytest.fixture
+def b_mean(df):
+    return df.resample("2D")["B"].mean()
+
+
+@pytest.fixture
+def b_std(df):
+    return df.resample("2D")["B"].std()
+
+
+@pytest.fixture
+def b_sum(df):
+    return df.resample("2D")["B"].sum()
+
+
+@pytest.fixture
+def df_resample(df):
+    return df.resample("2D")
+
+
+@pytest.fixture
+def df_col_resample(df_col):
+    return df_col.resample("2D", on="date")
+
+
+@pytest.fixture
+def df_mult_resample(df_mult):
+    return df_mult.resample("2D", level="date")
+
+
+@pytest.fixture
+def df_grouper_resample(df):
+    return df.groupby(pd.Grouper(freq="2D"))
+
+
+@pytest.fixture(
+    params=["df_resample", "df_col_resample", "df_mult_resample", "df_grouper_resample"]
+)
+def cases(request):
+    return request.getfixturevalue(request.param)
+
+
+def test_agg_mixed_column_aggregation(cases, a_mean, a_std, b_mean, b_std, request):
+    expected = pd.concat([a_mean, a_std, b_mean, b_std], axis=1)
+    expected.columns = pd.MultiIndex.from_product([["A", "B"], ["mean", "<lambda_0>"]])
+    # "date" is an index and a column, so get included in the agg
+    if "df_mult" in request.node.callspec.id:
+        date_mean = cases["date"].mean()
+        date_std = cases["date"].std()
+        expected = pd.concat([date_mean, date_std, expected], axis=1)
+        expected.columns = pd.MultiIndex.from_product(
+            [["date", "A", "B"], ["mean", "<lambda_0>"]]
+        )
+    result = cases.aggregate([np.mean, lambda x: np.std(x, ddof=1)])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg",
+    [
+        {"func": {"A": np.mean, "B": lambda x: np.std(x, ddof=1)}},
+        {"A": ("A", np.mean), "B": ("B", lambda x: np.std(x, ddof=1))},
+        {"A": NamedAgg("A", np.mean), "B": NamedAgg("B", lambda x: np.std(x, ddof=1))},
+    ],
+)
+def test_agg_both_mean_std_named_result(cases, a_mean, b_std, agg):
+    expected = pd.concat([a_mean, b_std], axis=1)
+    result = cases.aggregate(**agg)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+def test_agg_both_mean_std_dict_of_list(cases, a_mean, a_std):
+    expected = pd.concat([a_mean, a_std], axis=1)
+    expected.columns = pd.MultiIndex.from_tuples([("A", "mean"), ("A", "std")])
+    result = cases.aggregate({"A": ["mean", "std"]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg", [{"func": ["mean", "sum"]}, {"mean": "mean", "sum": "sum"}]
+)
+def test_agg_both_mean_sum(cases, a_mean, a_sum, agg):
+    expected = pd.concat([a_mean, a_sum], axis=1)
+    expected.columns = ["mean", "sum"]
+    result = cases["A"].aggregate(**agg)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "agg",
+    [
+        {"A": {"mean": "mean", "sum": "sum"}},
+        {
+            "A": {"mean": "mean", "sum": "sum"},
+            "B": {"mean2": "mean", "sum2": "sum"},
+        },
+    ],
+)
+def test_agg_dict_of_dict_specificationerror(cases, agg):
+    msg = "nested renamer is not supported"
+    with pytest.raises(pd.errors.SpecificationError, match=msg):
+        cases.aggregate(agg)
+
+
+def test_agg_dict_of_lists(cases, a_mean, a_std, b_mean, b_std):
+    expected = pd.concat([a_mean, a_std, b_mean, b_std], axis=1)
+    expected.columns = pd.MultiIndex.from_tuples(
+        [("A", "mean"), ("A", "std"), ("B", "mean"), ("B", "std")]
+    )
+    result = cases.aggregate({"A": ["mean", "std"], "B": ["mean", "std"]})
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize(
+    "agg",
+    [
+        {"func": {"A": np.sum, "B": lambda x: np.std(x, ddof=1)}},
+        {"A": ("A", np.sum), "B": ("B", lambda x: np.std(x, ddof=1))},
+        {"A": NamedAgg("A", np.sum), "B": NamedAgg("B", lambda x: np.std(x, ddof=1))},
+    ],
+)
+def test_agg_with_lambda(cases, agg):
+    # passed lambda
+    rcustom = cases["B"].apply(lambda x: np.std(x, ddof=1))
+    expected = pd.concat([cases["A"].sum(), rcustom], axis=1)
+    result = cases.agg(**agg)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize(
+    "agg",
+    [
+        {"func": {"result1": np.sum, "result2": np.mean}},
+        {"A": ("result1", np.sum), "B": ("result2", np.mean)},
+        {"A": NamedAgg("result1", np.sum), "B": NamedAgg("result2", np.mean)},
+    ],
+)
+def test_agg_no_column(cases, agg):
+    msg = r"Label\(s\) \['result1', 'result2'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        cases[["A", "B"]].agg(**agg)
+
+
+@pytest.mark.parametrize(
+    "cols, agg",
+    [
+        [None, {"A": ["sum", "std"], "B": ["mean", "std"]}],
+        [
+            [
+                "A",
+                "B",
+            ],
+            {"A": ["sum", "std"], "B": ["mean", "std"]},
+        ],
+    ],
+)
+def test_agg_specificationerror_nested(cases, cols, agg, a_sum, a_std, b_mean, b_std):
+    # agg with different hows
+    # equivalent of using a selection list / or not
+    expected = pd.concat([a_sum, a_std, b_mean, b_std], axis=1)
+    expected.columns = pd.MultiIndex.from_tuples(
+        [("A", "sum"), ("A", "std"), ("B", "mean"), ("B", "std")]
+    )
+    if cols is not None:
+        obj = cases[cols]
+    else:
+        obj = cases
+
+    result = obj.agg(agg)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize(
+    "agg", [{"A": ["sum", "std"]}, {"A": ["sum", "std"], "B": ["mean", "std"]}]
+)
+def test_agg_specificationerror_series(cases, agg):
+    msg = "nested renamer is not supported"
+
+    # series like aggs
+    with pytest.raises(pd.errors.SpecificationError, match=msg):
+        cases["A"].agg(agg)
+
+
+def test_agg_specificationerror_invalid_names(cases):
+    # errors
+    # invalid names in the agg specification
+    msg = r"Label\(s\) \['B'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        cases[["A"]].agg({"A": ["sum", "std"], "B": ["mean", "std"]})
+
+
+def test_agg_nested_dicts():
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+    index.name = "date"
+    df = DataFrame(
+        np.random.default_rng(2).random((10, 2)), columns=list("AB"), index=index
+    )
+    df_col = df.reset_index()
+    df_mult = df_col.copy()
+    df_mult.index = pd.MultiIndex.from_arrays(
+        [range(10), df.index], names=["index", "date"]
+    )
+    r = df.resample("2D")
+    cases = [
+        r,
+        df_col.resample("2D", on="date"),
+        df_mult.resample("2D", level="date"),
+        df.groupby(pd.Grouper(freq="2D")),
+    ]
+
+    msg = "nested renamer is not supported"
+    for t in cases:
+        with pytest.raises(pd.errors.SpecificationError, match=msg):
+            t.aggregate({"r1": {"A": ["mean", "sum"]}, "r2": {"B": ["mean", "sum"]}})
+
+    for t in cases:
+        with pytest.raises(pd.errors.SpecificationError, match=msg):
+            t[["A", "B"]].agg(
+                {"A": {"ra": ["mean", "std"]}, "B": {"rb": ["mean", "std"]}}
+            )
+
+        with pytest.raises(pd.errors.SpecificationError, match=msg):
+            t.agg({"A": {"ra": ["mean", "std"]}, "B": {"rb": ["mean", "std"]}})
+
+
+def test_try_aggregate_non_existing_column():
+    # GH 16766
+    data = [
+        {"dt": datetime(2017, 6, 1, 0), "x": 1.0, "y": 2.0},
+        {"dt": datetime(2017, 6, 1, 1), "x": 2.0, "y": 2.0},
+        {"dt": datetime(2017, 6, 1, 2), "x": 3.0, "y": 1.5},
+    ]
+    df = DataFrame(data).set_index("dt")
+
+    # Error as we don't have 'z' column
+    msg = r"Label\(s\) \['z'\] do not exist"
+    with pytest.raises(KeyError, match=msg):
+        df.resample("30min").agg({"x": ["mean"], "y": ["median"], "z": ["sum"]})
+
+
+def test_agg_list_like_func_with_args():
+    # 50624
+    df = DataFrame(
+        {"x": [1, 2, 3]}, index=date_range("2020-01-01", periods=3, freq="D")
+    )
+
+    def foo1(x, a=1, c=0):
+        return x + a + c
+
+    def foo2(x, b=2, c=0):
+        return x + b + c
+
+    msg = r"foo1\(\) got an unexpected keyword argument 'b'"
+    with pytest.raises(TypeError, match=msg):
+        df.resample("D").agg([foo1, foo2], 3, b=3, c=4)
+
+    result = df.resample("D").agg([foo1, foo2], 3, c=4)
+    expected = DataFrame(
+        [[8, 8], [9, 9], [10, 10]],
+        index=date_range("2020-01-01", periods=3, freq="D"),
+        columns=pd.MultiIndex.from_tuples([("x", "foo1"), ("x", "foo2")]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_selection_api_validation():
+    # GH 13500
+    index = date_range(datetime(2005, 1, 1), datetime(2005, 1, 10), freq="D")
+
+    rng = np.arange(len(index), dtype=np.int64)
+    df = DataFrame(
+        {"date": index, "a": rng},
+        index=pd.MultiIndex.from_arrays([rng, index], names=["v", "d"]),
+    )
+    df_exp = DataFrame({"a": rng}, index=index)
+
+    # non DatetimeIndex
+    msg = (
+        "Only valid with DatetimeIndex, TimedeltaIndex or PeriodIndex, "
+        "but got an instance of 'Index'"
+    )
+    with pytest.raises(TypeError, match=msg):
+        df.resample("2D", level="v")
+
+    msg = "The Grouper cannot specify both a key and a level!"
+    with pytest.raises(ValueError, match=msg):
+        df.resample("2D", on="date", level="d")
+
+    msg = "unhashable type: 'list'"
+    with pytest.raises(TypeError, match=msg):
+        df.resample("2D", on=["a", "date"])
+
+    msg = r"\"Level \['a', 'date'\] not found\""
+    with pytest.raises(KeyError, match=msg):
+        df.resample("2D", level=["a", "date"])
+
+    # upsampling not allowed
+    msg = (
+        "Upsampling from level= or on= selection is not supported, use "
+        r"\.set_index\(\.\.\.\) to explicitly set index to datetime-like"
+    )
+    with pytest.raises(ValueError, match=msg):
+        df.resample("2D", level="d").asfreq()
+    with pytest.raises(ValueError, match=msg):
+        df.resample("2D", on="date").asfreq()
+
+    exp = df_exp.resample("2D").sum()
+    exp.index.name = "date"
+    result = df.resample("2D", on="date").sum()
+    tm.assert_frame_equal(exp, result)
+
+    exp.index.name = "d"
+    with pytest.raises(
+        TypeError, match="datetime64 type does not support operation 'sum'"
+    ):
+        df.resample("2D", level="d").sum()
+    result = df.resample("2D", level="d").sum(numeric_only=True)
+    tm.assert_frame_equal(exp, result)
+
+
+@pytest.mark.parametrize(
+    "col_name", ["t2", "t2x", "t2q", "T_2M", "t2p", "t2m", "t2m1", "T2M"]
+)
+def test_agg_with_datetime_index_list_agg_func(col_name):
+    # GH 22660
+    # The parametrized column names would get converted to dates by our
+    # date parser. Some would result in OutOfBoundsError (ValueError) while
+    # others would result in OverflowError when passed into Timestamp.
+    # We catch these errors and move on to the correct branch.
+    df = DataFrame(
+        list(range(200)),
+        index=date_range(
+            start="2017-01-01", freq="15min", periods=200, tz="Europe/Berlin"
+        ),
+        columns=[col_name],
+    )
+    result = df.resample("1D").aggregate(["mean"])
+    expected = DataFrame(
+        [47.5, 143.5, 195.5],
+        index=date_range(start="2017-01-01", freq="D", periods=3, tz="Europe/Berlin"),
+        columns=pd.MultiIndex(levels=[[col_name], ["mean"]], codes=[[0], [0]]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_agg_readonly():
+    # GH#31710 cython needs to allow readonly data
+    index = date_range("2020-01-01", "2020-01-02", freq="1h")
+    arr = np.zeros_like(index)
+    arr.setflags(write=False)
+
+    ser = Series(arr, index=index)
+    rs = ser.resample("1D")
+
+    expected = Series([pd.Timestamp(0), pd.Timestamp(0)], index=index[::24])
+
+    result = rs.agg("last")
+    tm.assert_series_equal(result, expected)
+
+    result = rs.agg("first")
+    tm.assert_series_equal(result, expected)
+
+    result = rs.agg("max")
+    tm.assert_series_equal(result, expected)
+
+    result = rs.agg("min")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start,end,freq,data,resample_freq,origin,closed,exp_data,exp_end,exp_periods",
+    [
+        (
+            "2000-10-01 23:30:00",
+            "2000-10-02 00:26:00",
+            "7min",
+            [0, 3, 6, 9, 12, 15, 18, 21, 24],
+            "17min",
+            "end",
+            None,
+            [0, 18, 27, 63],
+            "20001002 00:26:00",
+            4,
+        ),
+        (
+            "20200101 8:26:35",
+            "20200101 9:31:58",
+            "77s",
+            [1] * 51,
+            "7min",
+            "end",
+            "right",
+            [1, 6, 5, 6, 5, 6, 5, 6, 5, 6],
+            "2020-01-01 09:30:45",
+            10,
+        ),
+        (
+            "2000-10-01 23:30:00",
+            "2000-10-02 00:26:00",
+            "7min",
+            [0, 3, 6, 9, 12, 15, 18, 21, 24],
+            "17min",
+            "end",
+            "left",
+            [0, 18, 27, 39, 24],
+            "20001002 00:43:00",
+            5,
+        ),
+        (
+            "2000-10-01 23:30:00",
+            "2000-10-02 00:26:00",
+            "7min",
+            [0, 3, 6, 9, 12, 15, 18, 21, 24],
+            "17min",
+            "end_day",
+            None,
+            [3, 15, 45, 45],
+            "2000-10-02 00:29:00",
+            4,
+        ),
+    ],
+)
+def test_end_and_end_day_origin(
+    start,
+    end,
+    freq,
+    data,
+    resample_freq,
+    origin,
+    closed,
+    exp_data,
+    exp_end,
+    exp_periods,
+):
+    rng = date_range(start, end, freq=freq)
+    ts = Series(data, index=rng)
+
+    res = ts.resample(resample_freq, origin=origin, closed=closed).sum()
+    expected = Series(
+        exp_data,
+        index=date_range(end=exp_end, freq=resample_freq, periods=exp_periods),
+    )
+
+    tm.assert_series_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    # expected_data is a string when op raises a ValueError
+    "method, numeric_only, expected_data",
+    [
+        ("sum", True, {"num": [25]}),
+        ("sum", False, {"cat": ["cat_1cat_2"], "num": [25]}),
+        ("sum", lib.no_default, {"cat": ["cat_1cat_2"], "num": [25]}),
+        ("prod", True, {"num": [100]}),
+        ("prod", False, "can't multiply sequence"),
+        ("prod", lib.no_default, "can't multiply sequence"),
+        ("min", True, {"num": [5]}),
+        ("min", False, {"cat": ["cat_1"], "num": [5]}),
+        ("min", lib.no_default, {"cat": ["cat_1"], "num": [5]}),
+        ("max", True, {"num": [20]}),
+        ("max", False, {"cat": ["cat_2"], "num": [20]}),
+        ("max", lib.no_default, {"cat": ["cat_2"], "num": [20]}),
+        ("first", True, {"num": [5]}),
+        ("first", False, {"cat": ["cat_1"], "num": [5]}),
+        ("first", lib.no_default, {"cat": ["cat_1"], "num": [5]}),
+        ("last", True, {"num": [20]}),
+        ("last", False, {"cat": ["cat_2"], "num": [20]}),
+        ("last", lib.no_default, {"cat": ["cat_2"], "num": [20]}),
+        ("mean", True, {"num": [12.5]}),
+        ("mean", False, "Could not convert"),
+        ("mean", lib.no_default, "Could not convert"),
+        ("median", True, {"num": [12.5]}),
+        ("median", False, r"Cannot convert \['cat_1' 'cat_2'\] to numeric"),
+        ("median", lib.no_default, r"Cannot convert \['cat_1' 'cat_2'\] to numeric"),
+        ("std", True, {"num": [10.606601717798213]}),
+        ("std", False, "could not convert string to float"),
+        ("std", lib.no_default, "could not convert string to float"),
+        ("var", True, {"num": [112.5]}),
+        ("var", False, "could not convert string to float"),
+        ("var", lib.no_default, "could not convert string to float"),
+        ("sem", True, {"num": [7.5]}),
+        ("sem", False, "could not convert string to float"),
+        ("sem", lib.no_default, "could not convert string to float"),
+    ],
+)
+def test_frame_downsample_method(
+    method, numeric_only, expected_data, using_infer_string
+):
+    # GH#46442 test if `numeric_only` behave as expected for DataFrameGroupBy
+
+    index = date_range("2018-01-01", periods=2, freq="D")
+    expected_index = date_range("2018-12-31", periods=1, freq="YE")
+    df = DataFrame({"cat": ["cat_1", "cat_2"], "num": [5, 20]}, index=index)
+    resampled = df.resample("YE")
+    if numeric_only is lib.no_default:
+        kwargs = {}
+    else:
+        kwargs = {"numeric_only": numeric_only}
+
+    func = getattr(resampled, method)
+    if isinstance(expected_data, str):
+        if method in ("var", "mean", "median", "prod"):
+            klass = TypeError
+            msg = re.escape(f"agg function failed [how->{method},dtype->")
+            if using_infer_string:
+                msg = f"dtype 'str' does not support operation '{method}'"
+        elif method in ["sum", "std", "sem"] and using_infer_string:
+            klass = TypeError
+            msg = f"dtype 'str' does not support operation '{method}'"
+        else:
+            klass = ValueError
+            msg = expected_data
+        with pytest.raises(klass, match=msg):
+            _ = func(**kwargs)
+    else:
+        result = func(**kwargs)
+        expected = DataFrame(expected_data, index=expected_index)
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, numeric_only, expected_data",
+    [
+        ("sum", True, ()),
+        ("sum", False, ["cat_1cat_2"]),
+        ("sum", lib.no_default, ["cat_1cat_2"]),
+        ("prod", True, ()),
+        ("prod", False, ()),
+        ("prod", lib.no_default, ()),
+        ("min", True, ()),
+        ("min", False, ["cat_1"]),
+        ("min", lib.no_default, ["cat_1"]),
+        ("max", True, ()),
+        ("max", False, ["cat_2"]),
+        ("max", lib.no_default, ["cat_2"]),
+        ("first", True, ()),
+        ("first", False, ["cat_1"]),
+        ("first", lib.no_default, ["cat_1"]),
+        ("last", True, ()),
+        ("last", False, ["cat_2"]),
+        ("last", lib.no_default, ["cat_2"]),
+    ],
+)
+def test_series_downsample_method(
+    method, numeric_only, expected_data, using_infer_string
+):
+    # GH#46442 test if `numeric_only` behave as expected for SeriesGroupBy
+
+    index = date_range("2018-01-01", periods=2, freq="D")
+    expected_index = date_range("2018-12-31", periods=1, freq="YE")
+    df = Series(["cat_1", "cat_2"], index=index)
+    resampled = df.resample("YE")
+    kwargs = {} if numeric_only is lib.no_default else {"numeric_only": numeric_only}
+
+    func = getattr(resampled, method)
+    if numeric_only and numeric_only is not lib.no_default:
+        msg = rf"Cannot use numeric_only=True with SeriesGroupBy\.{method}"
+        with pytest.raises(TypeError, match=msg):
+            func(**kwargs)
+    elif method == "prod":
+        msg = re.escape("agg function failed [how->prod,dtype->")
+        if using_infer_string:
+            msg = "dtype 'str' does not support operation 'prod'"
+        with pytest.raises(TypeError, match=msg):
+            func(**kwargs)
+
+    else:
+        result = func(**kwargs)
+        expected = Series(expected_data, index=expected_index)
+        tm.assert_series_equal(result, expected)
+
+
+def test_resample_empty():
+    # GH#52484
+    df = DataFrame(
+        index=pd.to_datetime(
+            ["2018-01-01 00:00:00", "2018-01-01 12:00:00", "2018-01-02 00:00:00"]
+        )
+    )
+    expected = DataFrame(
+        index=pd.to_datetime(
+            [
+                "2018-01-01 00:00:00",
+                "2018-01-01 08:00:00",
+                "2018-01-01 16:00:00",
+                "2018-01-02 00:00:00",
+            ]
+        )
+    )
+    result = df.resample("8h").mean()
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/resample/test_resampler_grouper.py b/pandas/pandas/tests/resample/test_resampler_grouper.py
new file mode 100644
index 0000000000000..e7850f96b3b0f
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_resampler_grouper.py
@@ -0,0 +1,654 @@
+from textwrap import dedent
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import is_platform_windows
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    TimedeltaIndex,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import date_range
+
+
+@pytest.fixture
+def test_frame():
+    return DataFrame(
+        {"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)},
+        index=date_range("1/1/2000", freq="s", periods=40),
+    )
+
+
+def test_tab_complete_ipython6_warning(ip):
+    from IPython.core.completer import provisionalcompleter
+
+    code = dedent(
+        """\
+    import numpy as np
+    from pandas import Series, date_range
+    data = np.arange(10, dtype=np.float64)
+    index = date_range("2020-01-01", periods=len(data))
+    s = Series(data, index=index)
+    rs = s.resample("D")
+    """
+    )
+    ip.run_cell(code)
+
+    # GH 31324 newer jedi version raises Deprecation warning;
+    #  appears resolved 2021-02-02
+    with tm.assert_produces_warning(None, raise_on_extra_warnings=False):
+        with provisionalcompleter("ignore"):
+            list(ip.Completer.completions("rs.", 1))
+
+
+def test_deferred_with_groupby():
+    # GH 12486
+    # support deferred resample ops with groupby
+    data = [
+        ["2010-01-01", "A", 2],
+        ["2010-01-02", "A", 3],
+        ["2010-01-05", "A", 8],
+        ["2010-01-10", "A", 7],
+        ["2010-01-13", "A", 3],
+        ["2010-01-01", "B", 5],
+        ["2010-01-03", "B", 2],
+        ["2010-01-04", "B", 1],
+        ["2010-01-11", "B", 7],
+        ["2010-01-14", "B", 3],
+    ]
+
+    df = DataFrame(data, columns=["date", "id", "score"])
+    df.date = pd.to_datetime(df.date)
+
+    def f_0(x):
+        return x.set_index("date").resample("D").asfreq()
+
+    expected = df.groupby("id").apply(f_0)
+    result = df.set_index("date").groupby("id").resample("D").asfreq()
+    tm.assert_frame_equal(result, expected)
+
+    df = DataFrame(
+        {
+            "date": date_range(start="2016-01-01", periods=4, freq="W"),
+            "group": [1, 1, 2, 2],
+            "val": [5, 6, 7, 8],
+        }
+    ).set_index("date")
+
+    def f_1(x):
+        return x.resample("1D").ffill()
+
+    expected = df.groupby("group").apply(f_1)
+    result = df.groupby("group").resample("1D").ffill()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_getitem(test_frame):
+    g = test_frame.groupby("A")
+
+    expected = g.B.apply(lambda x: x.resample("2s").mean())
+
+    result = g.resample("2s").B.mean()
+    tm.assert_series_equal(result, expected)
+
+    result = g.B.resample("2s").mean()
+    tm.assert_series_equal(result, expected)
+
+    result = g.resample("2s").mean().B
+    tm.assert_series_equal(result, expected)
+
+
+def test_getitem_multiple():
+    # GH 13174
+    # multiple calls after selection causing an issue with aliasing
+    data = [{"id": 1, "buyer": "A"}, {"id": 2, "buyer": "B"}]
+    df = DataFrame(data, index=date_range("2016-01-01", periods=2))
+    r = df.groupby("id").resample("1D")
+    result = r["buyer"].count()
+
+    exp_mi = pd.MultiIndex.from_arrays([[1, 2], df.index], names=("id", None))
+    expected = Series(
+        [1, 1],
+        index=exp_mi,
+        name="buyer",
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = r["buyer"].count()
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_resample_on_api_with_getitem():
+    # GH 17813
+    df = DataFrame(
+        {"id": list("aabbb"), "date": date_range("1-1-2016", periods=5), "data": 1}
+    )
+    exp = df.set_index("date").groupby("id").resample("2D")["data"].sum()
+    result = df.groupby("id").resample("2D", on="date")["data"].sum()
+    tm.assert_series_equal(result, exp)
+
+
+def test_groupby_with_origin():
+    # GH 31809
+
+    freq = "1399min"  # prime number that is smaller than 24h
+    start, end = "1/1/2000 00:00:00", "1/31/2000 00:00"
+    middle = "1/15/2000 00:00:00"
+
+    rng = date_range(start, end, freq="1231min")  # prime number
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+    ts2 = ts[middle:end]
+
+    # proves that grouper without a fixed origin does not work
+    # when dealing with unusual frequencies
+    simple_grouper = pd.Grouper(freq=freq)
+    count_ts = ts.groupby(simple_grouper).agg("count")
+    count_ts = count_ts[middle:end]
+    count_ts2 = ts2.groupby(simple_grouper).agg("count")
+    with pytest.raises(AssertionError, match="Index are different"):
+        tm.assert_index_equal(count_ts.index, count_ts2.index)
+
+    # test origin on 1970-01-01 00:00:00
+    origin = Timestamp(0)
+    adjusted_grouper = pd.Grouper(freq=freq, origin=origin)
+    adjusted_count_ts = ts.groupby(adjusted_grouper).agg("count")
+    adjusted_count_ts = adjusted_count_ts[middle:end]
+    adjusted_count_ts2 = ts2.groupby(adjusted_grouper).agg("count")
+    tm.assert_series_equal(adjusted_count_ts, adjusted_count_ts2)
+
+    # test origin on 2049-10-18 20:00:00
+    origin_future = Timestamp(0) + pd.Timedelta("1399min") * 30_000
+    adjusted_grouper2 = pd.Grouper(freq=freq, origin=origin_future)
+    adjusted2_count_ts = ts.groupby(adjusted_grouper2).agg("count")
+    adjusted2_count_ts = adjusted2_count_ts[middle:end]
+    adjusted2_count_ts2 = ts2.groupby(adjusted_grouper2).agg("count")
+    tm.assert_series_equal(adjusted2_count_ts, adjusted2_count_ts2)
+
+    # both grouper use an adjusted timestamp that is a multiple of 1399 min
+    # they should be equals even if the adjusted_timestamp is in the future
+    tm.assert_series_equal(adjusted_count_ts, adjusted2_count_ts2)
+
+
+def test_nearest():
+    # GH 17496
+    # Resample nearest
+    index = date_range("1/1/2000", periods=3, freq="min")
+    result = Series(range(3), index=index).resample("20s").nearest()
+
+    expected = Series(
+        [0, 0, 1, 1, 1, 2, 2],
+        index=pd.DatetimeIndex(
+            [
+                "2000-01-01 00:00:00",
+                "2000-01-01 00:00:20",
+                "2000-01-01 00:00:40",
+                "2000-01-01 00:01:00",
+                "2000-01-01 00:01:20",
+                "2000-01-01 00:01:40",
+                "2000-01-01 00:02:00",
+            ],
+            dtype="datetime64[ns]",
+            freq="20s",
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        "first",
+        "last",
+        "median",
+        "sem",
+        "sum",
+        "mean",
+        "min",
+        "max",
+        "size",
+        "count",
+        "nearest",
+        "bfill",
+        "ffill",
+        "asfreq",
+        "ohlc",
+    ],
+)
+def test_methods(f, test_frame):
+    g = test_frame.groupby("A")
+    r = g.resample("2s")
+
+    result = getattr(r, f)()
+    expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
+    tm.assert_equal(result, expected)
+
+
+def test_methods_nunique(test_frame):
+    # series only
+    g = test_frame.groupby("A")
+    r = g.resample("2s")
+    result = r.B.nunique()
+    expected = g.B.apply(lambda x: x.resample("2s").nunique())
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("f", ["std", "var"])
+def test_methods_std_var(f, test_frame):
+    g = test_frame.groupby("A")
+    r = g.resample("2s")
+    result = getattr(r, f)(ddof=1)
+    expected = g.apply(lambda x: getattr(x.resample("2s"), f)(ddof=1))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply(test_frame):
+    g = test_frame.groupby("A")
+    r = g.resample("2s")
+
+    # reduction
+    expected = g.resample("2s").sum()
+
+    def f_0(x):
+        return x.resample("2s").sum()
+
+    result = r.apply(f_0)
+    tm.assert_frame_equal(result, expected)
+
+    def f_1(x):
+        return x.resample("2s").apply(lambda y: y.sum())
+
+    result = g.apply(f_1)
+    # y.sum() results in int64 instead of int32 on 32-bit architectures
+    expected = expected.astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_with_mutated_index():
+    # GH 15169
+    index = date_range("1-1-2015", "12-31-15", freq="D")
+    df = DataFrame(
+        data={"col1": np.random.default_rng(2).random(len(index))}, index=index
+    )
+
+    def f(x):
+        s = Series([1, 2], index=["a", "b"])
+        return s
+
+    expected = df.groupby(pd.Grouper(freq="ME")).apply(f)
+
+    result = df.resample("ME").apply(f)
+    tm.assert_frame_equal(result, expected)
+
+    # A case for series
+    expected = df["col1"].groupby(pd.Grouper(freq="ME"), group_keys=False).apply(f)
+    result = df["col1"].resample("ME").apply(f)
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_columns_multilevel():
+    # GH 16231
+    cols = pd.MultiIndex.from_tuples([("A", "a", "", "one"), ("B", "b", "i", "two")])
+    ind = date_range(start="2017-01-01", freq="15Min", periods=8)
+    df = DataFrame(np.array([0] * 16).reshape(8, 2), index=ind, columns=cols)
+    agg_dict = {col: (np.sum if col[3] == "one" else np.mean) for col in df.columns}
+    result = df.resample("h").apply(lambda x: agg_dict[x.name](x))
+    expected = DataFrame(
+        2 * [[0, 0.0]],
+        index=date_range(start="2017-01-01", freq="1h", periods=2),
+        columns=pd.MultiIndex.from_tuples(
+            [("A", "a", "", "one"), ("B", "b", "i", "two")]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_non_naive_index():
+    def weighted_quantile(series, weights, q):
+        series = series.sort_values()
+        cumsum = weights.reindex(series.index).fillna(0).cumsum()
+        cutoff = cumsum.iloc[-1] * q
+        return series[cumsum >= cutoff].iloc[0]
+
+    times = date_range("2017-6-23 18:00", periods=8, freq="15min", tz="UTC")
+    data = Series([1.0, 1, 1, 1, 1, 2, 2, 0], index=times)
+    weights = Series([160.0, 91, 65, 43, 24, 10, 1, 0], index=times)
+
+    result = data.resample("D").apply(weighted_quantile, weights=weights, q=0.5)
+    ind = date_range(
+        "2017-06-23 00:00:00+00:00", "2017-06-23 00:00:00+00:00", freq="D", tz="UTC"
+    )
+    expected = Series([1.0], index=ind)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_groupby_with_label(unit):
+    # GH 13235
+    index = date_range("2000-01-01", freq="2D", periods=5, unit=unit)
+    df = DataFrame(index=index, data={"col0": [0, 0, 1, 1, 2], "col1": [1, 1, 1, 1, 1]})
+    result = df.groupby("col0").resample("1W", label="left").sum()
+
+    mi = [
+        np.array([0, 0, 1, 2], dtype=np.int64),
+        np.array(
+            ["1999-12-26", "2000-01-02", "2000-01-02", "2000-01-02"],
+            dtype=f"M8[{unit}]",
+        ),
+    ]
+    mindex = pd.MultiIndex.from_arrays(mi, names=["col0", None])
+    expected = DataFrame(data={"col1": [1, 1, 2, 1]}, index=mindex)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_consistency_with_window(test_frame):
+    # consistent return values with window
+    df = test_frame
+    expected = Index([1, 2, 3], name="A")
+    result = df.groupby("A").resample("2s").mean()
+    assert result.index.nlevels == 2
+    tm.assert_index_equal(result.index.levels[0], expected)
+
+    result = df.groupby("A").rolling(20).mean()
+    assert result.index.nlevels == 2
+    tm.assert_index_equal(result.index.levels[0], expected)
+
+
+def test_median_duplicate_columns():
+    # GH 14233
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((20, 3)),
+        columns=list("aaa"),
+        index=date_range("2012-01-01", periods=20, freq="s"),
+    )
+    result = df.resample("5s").median()
+    df.columns = ["a", "b", "c"]
+    expected = df.resample("5s").median()
+    expected.columns = result.columns
+    tm.assert_frame_equal(result, expected)
+
+
+def test_apply_to_one_column_of_df():
+    # GH: 36951
+    df = DataFrame(
+        {"col": range(10), "col1": range(10, 20)},
+        index=date_range("2012-01-01", periods=10, freq="20min"),
+    )
+
+    # access "col" via getattr -> make sure we handle AttributeError
+    result = df.resample("h").apply(lambda group: group.col.sum())
+    expected = Series(
+        [3, 12, 21, 9], index=date_range("2012-01-01", periods=4, freq="h")
+    )
+    tm.assert_series_equal(result, expected)
+
+    # access "col" via _getitem__ -> make sure we handle KeyErrpr
+    result = df.resample("h").apply(lambda group: group["col"].sum())
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_groupby_agg():
+    # GH: 33548
+    df = DataFrame(
+        {
+            "cat": [
+                "cat_1",
+                "cat_1",
+                "cat_2",
+                "cat_1",
+                "cat_2",
+                "cat_1",
+                "cat_2",
+                "cat_1",
+            ],
+            "num": [5, 20, 22, 3, 4, 30, 10, 50],
+            "date": [
+                "2019-2-1",
+                "2018-02-03",
+                "2020-3-11",
+                "2019-2-2",
+                "2019-2-2",
+                "2018-12-4",
+                "2020-3-11",
+                "2020-12-12",
+            ],
+        }
+    )
+    df["date"] = pd.to_datetime(df["date"])
+
+    resampled = df.groupby("cat").resample("YE", on="date")
+    expected = resampled[["num"]].sum()
+    result = resampled.agg({"num": "sum"})
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_groupby_agg_listlike():
+    # GH 42905
+    ts = Timestamp("2021-02-28 00:00:00")
+    df = DataFrame({"class": ["beta"], "value": [69]}, index=Index([ts], name="date"))
+    resampled = df.groupby("class").resample("ME")["value"]
+    result = resampled.agg(["sum", "size"])
+    expected = DataFrame(
+        [[69, 1]],
+        index=pd.MultiIndex.from_tuples([("beta", ts)], names=["class", "date"]),
+        columns=["sum", "size"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", [["a"], ["a", "b"]])
+def test_empty(keys):
+    # GH 26411
+    df = DataFrame([], columns=["a", "b"], index=TimedeltaIndex([]))
+    result = df.groupby(keys).resample(rule=pd.to_timedelta("00:00:01")).mean()
+    expected_columns = ["b"] if keys == ["a"] else []
+    expected = (
+        DataFrame(columns=["a", "b"])
+        .set_index(keys, drop=False)
+        .set_index(TimedeltaIndex([]), append=True)[expected_columns]
+    )
+    if len(keys) == 1:
+        expected.index.name = keys[0]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
+@pytest.mark.parametrize("consolidate", [True, False])
+def test_resample_groupby_agg_object_dtype_all_nan(consolidate):
+    # https://github.com/pandas-dev/pandas/issues/39329
+
+    dates = date_range("2020-01-01", periods=15, freq="D")
+    df1 = DataFrame({"key": "A", "date": dates, "col1": range(15), "col_object": "val"})
+    df2 = DataFrame({"key": "B", "date": dates, "col1": range(15)})
+    df = pd.concat([df1, df2], ignore_index=True)
+    if consolidate:
+        df = df._consolidate()
+
+    result = df.groupby(["key"]).resample("W", on="date").min()
+    idx = pd.MultiIndex.from_arrays(
+        [
+            ["A"] * 3 + ["B"] * 3,
+            pd.to_datetime(["2020-01-05", "2020-01-12", "2020-01-19"] * 2).as_unit(
+                "ns"
+            ),
+        ],
+        names=["key", "date"],
+    )
+    expected = DataFrame(
+        {
+            "col1": [0, 5, 12] * 2,
+            "col_object": ["val"] * 3 + [np.nan] * 3,
+        },
+        index=idx,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_with_list_of_keys():
+    # GH 47362
+    df = DataFrame(
+        data={
+            "date": date_range(start="2016-01-01", periods=8),
+            "group": [0, 0, 0, 0, 1, 1, 1, 1],
+            "val": [1, 7, 5, 2, 3, 10, 5, 1],
+        }
+    )
+    result = df.groupby("group").resample("2D", on="date")[["val"]].mean()
+
+    mi_exp = pd.MultiIndex.from_arrays(
+        [[0, 0, 1, 1], df["date"]._values[::2]], names=["group", "date"]
+    )
+    expected = DataFrame(
+        data={
+            "val": [4.0, 3.5, 6.5, 3.0],
+        },
+        index=mi_exp,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", [["a"], ["a", "b"]])
+def test_resample_no_index(keys):
+    # GH 47705
+    df = DataFrame([], columns=["a", "b", "date"])
+    df["date"] = pd.to_datetime(df["date"])
+    df = df.set_index("date")
+    result = df.groupby(keys).resample(rule=pd.to_timedelta("00:00:01")).mean()
+    expected_columns = ["b"] if keys == ["a"] else []
+    expected = DataFrame(columns=["a", "b", "date"]).set_index(keys, drop=False)
+    expected["date"] = pd.to_datetime(expected["date"])
+    expected = expected.set_index("date", append=True, drop=True)[expected_columns]
+    if len(keys) == 1:
+        expected.index.name = keys[0]
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_no_columns():
+    # GH#52484
+    df = DataFrame(
+        index=Index(
+            pd.to_datetime(
+                ["2018-01-01 00:00:00", "2018-01-01 12:00:00", "2018-01-02 00:00:00"]
+            ),
+            name="date",
+        )
+    )
+    result = df.groupby([0, 0, 1]).resample(rule=pd.to_timedelta("06:00:00")).mean()
+    index = pd.to_datetime(
+        [
+            "2018-01-01 00:00:00",
+            "2018-01-01 06:00:00",
+            "2018-01-01 12:00:00",
+            "2018-01-02 00:00:00",
+        ]
+    )
+    expected = DataFrame(
+        index=pd.MultiIndex(
+            levels=[np.array([0, 1], dtype=np.intp), index],
+            codes=[[0, 0, 0, 1], [0, 1, 2, 3]],
+            names=[None, "date"],
+        )
+    )
+
+    # GH#52710 - Index comes out as 32-bit on 64-bit Windows
+    tm.assert_frame_equal(result, expected, check_index_type=not is_platform_windows())
+
+
+def test_groupby_resample_size_all_index_same():
+    # GH 46826
+    df = DataFrame(
+        {"A": [1] * 3 + [2] * 3 + [1] * 3 + [2] * 3, "B": np.arange(12)},
+        index=date_range("31/12/2000 18:00", freq="h", periods=12),
+    )
+    result = df.groupby("A").resample("D").size()
+
+    mi_exp = pd.MultiIndex.from_arrays(
+        [
+            [1, 1, 2, 2],
+            pd.DatetimeIndex(["2000-12-31", "2001-01-01"] * 2, dtype="M8[ns]"),
+        ],
+        names=["A", None],
+    )
+    expected = Series(
+        3,
+        index=mi_exp,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_groupby_resample_on_index_with_list_of_keys():
+    # GH 50840
+    df = DataFrame(
+        data={
+            "group": [0, 0, 0, 0, 1, 1, 1, 1],
+            "val": [3, 1, 4, 1, 5, 9, 2, 6],
+        },
+        index=date_range(start="2016-01-01", periods=8, name="date"),
+    )
+    result = df.groupby("group").resample("2D")[["val"]].mean()
+
+    mi_exp = pd.MultiIndex.from_arrays(
+        [[0, 0, 1, 1], df.index[::2]], names=["group", "date"]
+    )
+    expected = DataFrame(
+        data={
+            "val": [2.0, 2.5, 7.0, 4.0],
+        },
+        index=mi_exp,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_on_index_with_list_of_keys_multi_columns():
+    # GH 50876
+    df = DataFrame(
+        data={
+            "group": [0, 0, 0, 0, 1, 1, 1, 1],
+            "first_val": [3, 1, 4, 1, 5, 9, 2, 6],
+            "second_val": [2, 7, 1, 8, 2, 8, 1, 8],
+            "third_val": [1, 4, 1, 4, 2, 1, 3, 5],
+        },
+        index=date_range(start="2016-01-01", periods=8, name="date"),
+    )
+    result = df.groupby("group").resample("2D")[["first_val", "second_val"]].mean()
+
+    mi_exp = pd.MultiIndex.from_arrays(
+        [[0, 0, 1, 1], df.index[::2]], names=["group", "date"]
+    )
+    expected = DataFrame(
+        data={
+            "first_val": [2.0, 2.5, 7.0, 4.0],
+            "second_val": [4.5, 4.5, 5.0, 4.5],
+        },
+        index=mi_exp,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_on_index_with_list_of_keys_missing_column():
+    # GH 50876
+    df = DataFrame(
+        data={
+            "group": [0, 0, 0, 0, 1, 1, 1, 1],
+            "val": [3, 1, 4, 1, 5, 9, 2, 6],
+        },
+        index=Series(
+            date_range(start="2016-01-01", periods=8),
+            name="date",
+        ),
+    )
+    gb = df.groupby("group")
+    rs = gb.resample("2D")
+    with pytest.raises(KeyError, match="Columns not found"):
+        rs[["val_not_in_dataframe"]]
diff --git a/pandas/pandas/tests/resample/test_time_grouper.py b/pandas/pandas/tests/resample/test_time_grouper.py
new file mode 100644
index 0000000000000..3cc95922e7f2f
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_time_grouper.py
@@ -0,0 +1,442 @@
+from datetime import datetime
+from operator import methodcaller
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas.core.groupby.grouper import Grouper
+from pandas.core.indexes.datetimes import date_range
+
+
+@pytest.fixture
+def test_series():
+    return Series(
+        np.random.default_rng(2).standard_normal(1000),
+        index=date_range("1/1/2000", periods=1000),
+    )
+
+
+def test_apply(test_series):
+    grouper = Grouper(freq="YE", label="right", closed="right")
+
+    grouped = test_series.groupby(grouper)
+
+    def f(x):
+        return x.sort_values()[-3:]
+
+    applied = grouped.apply(f)
+    expected = test_series.groupby(lambda x: x.year).apply(f)
+
+    applied.index = applied.index.droplevel(0)
+    expected.index = expected.index.droplevel(0)
+    tm.assert_series_equal(applied, expected)
+
+
+def test_count(test_series):
+    test_series[::3] = np.nan
+
+    expected = test_series.groupby(lambda x: x.year).count()
+
+    grouper = Grouper(freq="YE", label="right", closed="right")
+    result = test_series.groupby(grouper).count()
+    expected.index = result.index
+    tm.assert_series_equal(result, expected)
+
+    result = test_series.resample("YE").count()
+    expected.index = result.index
+    tm.assert_series_equal(result, expected)
+
+
+def test_numpy_reduction(test_series):
+    result = test_series.resample("YE", closed="right").prod()
+    expected = test_series.groupby(lambda x: x.year).agg(np.prod)
+    expected.index = result.index
+    tm.assert_series_equal(result, expected)
+
+
+def test_apply_iteration():
+    # #2300
+    N = 1000
+    ind = date_range(start="2000-01-01", freq="D", periods=N)
+    df = DataFrame({"open": 1, "close": 2}, index=ind)
+    tg = Grouper(freq="ME")
+
+    grouper, _ = tg._get_grouper(df)
+
+    # Errors
+    grouped = df.groupby(grouper, group_keys=False)
+
+    def f(df):
+        return df["close"] / df["open"]
+
+    # it works!
+    result = grouped.apply(f)
+    tm.assert_index_equal(result.index, df.index)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        Index([1, 2]),
+        Index(["a", "b"]),
+        Index([1.1, 2.2]),
+        pd.MultiIndex.from_arrays([[1, 2], ["a", "b"]]),
+    ],
+)
+def test_fails_on_no_datetime_index(index):
+    name = type(index).__name__
+    df = DataFrame({"a": range(len(index))}, index=index)
+
+    msg = (
+        "Only valid with DatetimeIndex, TimedeltaIndex "
+        f"or PeriodIndex, but got an instance of '{name}'"
+    )
+    with pytest.raises(TypeError, match=msg):
+        df.groupby(Grouper(freq="D"))
+
+
+def test_aaa_group_order():
+    # GH 12840
+    # check TimeGrouper perform stable sorts
+    n = 20
+    data = np.random.default_rng(2).standard_normal((n, 4))
+    df = DataFrame(data, columns=["A", "B", "C", "D"])
+    df["key"] = [
+        datetime(2013, 1, 1),
+        datetime(2013, 1, 2),
+        datetime(2013, 1, 3),
+        datetime(2013, 1, 4),
+        datetime(2013, 1, 5),
+    ] * 4
+    grouped = df.groupby(Grouper(key="key", freq="D"))
+
+    tm.assert_frame_equal(grouped.get_group(datetime(2013, 1, 1)), df[::5])
+    tm.assert_frame_equal(grouped.get_group(datetime(2013, 1, 2)), df[1::5])
+    tm.assert_frame_equal(grouped.get_group(datetime(2013, 1, 3)), df[2::5])
+    tm.assert_frame_equal(grouped.get_group(datetime(2013, 1, 4)), df[3::5])
+    tm.assert_frame_equal(grouped.get_group(datetime(2013, 1, 5)), df[4::5])
+
+
+def test_aggregate_normal(resample_method):
+    """Check TimeGrouper's aggregation is identical as normal groupby."""
+
+    data = np.random.default_rng(2).standard_normal((20, 4))
+    normal_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    normal_df["key"] = [1, 2, 3, 4, 5] * 4
+
+    dt_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    dt_df["key"] = Index(
+        [
+            datetime(2013, 1, 1),
+            datetime(2013, 1, 2),
+            datetime(2013, 1, 3),
+            datetime(2013, 1, 4),
+            datetime(2013, 1, 5),
+        ]
+        * 4,
+        dtype="M8[ns]",
+    )
+
+    normal_grouped = normal_df.groupby("key")
+    dt_grouped = dt_df.groupby(Grouper(key="key", freq="D"))
+
+    expected = getattr(normal_grouped, resample_method)()
+    dt_result = getattr(dt_grouped, resample_method)()
+    expected.index = date_range(start="2013-01-01", freq="D", periods=5, name="key")
+    tm.assert_equal(expected, dt_result)
+
+
+@pytest.mark.xfail(reason="if TimeGrouper is used included, 'nth' doesn't work yet")
+def test_aggregate_nth():
+    """Check TimeGrouper's aggregation is identical as normal groupby."""
+
+    data = np.random.default_rng(2).standard_normal((20, 4))
+    normal_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    normal_df["key"] = [1, 2, 3, 4, 5] * 4
+
+    dt_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    dt_df["key"] = [
+        datetime(2013, 1, 1),
+        datetime(2013, 1, 2),
+        datetime(2013, 1, 3),
+        datetime(2013, 1, 4),
+        datetime(2013, 1, 5),
+    ] * 4
+
+    normal_grouped = normal_df.groupby("key")
+    dt_grouped = dt_df.groupby(Grouper(key="key", freq="D"))
+
+    expected = normal_grouped.nth(3)
+    expected.index = date_range(start="2013-01-01", freq="D", periods=5, name="key")
+    dt_result = dt_grouped.nth(3)
+    tm.assert_frame_equal(expected, dt_result)
+
+
+@pytest.mark.parametrize(
+    "method, method_args, unit",
+    [
+        ("sum", {}, 0),
+        ("sum", {"min_count": 0}, 0),
+        ("sum", {"min_count": 1}, np.nan),
+        ("prod", {}, 1),
+        ("prod", {"min_count": 0}, 1),
+        ("prod", {"min_count": 1}, np.nan),
+    ],
+)
+def test_resample_entirely_nat_window(method, method_args, unit):
+    ser = Series([0] * 2 + [np.nan] * 2, index=date_range("2017", periods=4))
+    result = methodcaller(method, **method_args)(ser.resample("2D"))
+
+    exp_dti = pd.DatetimeIndex(["2017-01-01", "2017-01-03"], dtype="M8[ns]", freq="2D")
+    expected = Series([0.0, unit], index=exp_dti)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func, fill_value",
+    [("min", np.nan), ("max", np.nan), ("sum", 0), ("prod", 1), ("count", 0)],
+)
+def test_aggregate_with_nat(func, fill_value):
+    # check TimeGrouper's aggregation is identical as normal groupby
+    # if NaT is included, 'var', 'std', 'mean', 'first','last'
+    # and 'nth' doesn't work yet
+
+    n = 20
+    data = np.random.default_rng(2).standard_normal((n, 4)).astype("int64")
+    normal_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    normal_df["key"] = [1, 2, np.nan, 4, 5] * 4
+
+    dt_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    dt_df["key"] = Index(
+        [
+            datetime(2013, 1, 1),
+            datetime(2013, 1, 2),
+            pd.NaT,
+            datetime(2013, 1, 4),
+            datetime(2013, 1, 5),
+        ]
+        * 4,
+        dtype="M8[ns]",
+    )
+
+    normal_grouped = normal_df.groupby("key")
+    dt_grouped = dt_df.groupby(Grouper(key="key", freq="D"))
+
+    normal_result = getattr(normal_grouped, func)()
+    dt_result = getattr(dt_grouped, func)()
+
+    pad = DataFrame([[fill_value] * 4], index=[3], columns=["A", "B", "C", "D"])
+    expected = pd.concat([normal_result, pad])
+    expected = expected.sort_index()
+    dti = date_range(
+        start="2013-01-01",
+        freq="D",
+        periods=5,
+        name="key",
+        unit=dt_df["key"]._values.unit,
+    )
+    expected.index = dti._with_freq(None)  # TODO: is this desired?
+    tm.assert_frame_equal(expected, dt_result)
+    assert dt_result.index.name == "key"
+
+
+def test_aggregate_with_nat_size():
+    # GH 9925
+    n = 20
+    data = np.random.default_rng(2).standard_normal((n, 4)).astype("int64")
+    normal_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    normal_df["key"] = [1, 2, np.nan, 4, 5] * 4
+
+    dt_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    dt_df["key"] = Index(
+        [
+            datetime(2013, 1, 1),
+            datetime(2013, 1, 2),
+            pd.NaT,
+            datetime(2013, 1, 4),
+            datetime(2013, 1, 5),
+        ]
+        * 4,
+        dtype="M8[ns]",
+    )
+
+    normal_grouped = normal_df.groupby("key")
+    dt_grouped = dt_df.groupby(Grouper(key="key", freq="D"))
+
+    normal_result = normal_grouped.size()
+    dt_result = dt_grouped.size()
+
+    pad = Series([0], index=[3])
+    expected = pd.concat([normal_result, pad])
+    expected = expected.sort_index()
+    expected.index = date_range(
+        start="2013-01-01",
+        freq="D",
+        periods=5,
+        name="key",
+        unit=dt_df["key"]._values.unit,
+    )._with_freq(None)
+    tm.assert_series_equal(expected, dt_result)
+    assert dt_result.index.name == "key"
+
+
+def test_repr():
+    # GH18203
+    result = repr(Grouper(key="A", freq="h"))
+    expected = (
+        "TimeGrouper(key='A', freq=<Hour>, sort=True, dropna=True, "
+        "closed='left', label='left', how='mean', "
+        "convention='e', origin='start_day')"
+    )
+    assert result == expected
+
+    result = repr(Grouper(key="A", freq="h", origin="2000-01-01"))
+    expected = (
+        "TimeGrouper(key='A', freq=<Hour>, sort=True, dropna=True, "
+        "closed='left', label='left', how='mean', "
+        "convention='e', origin=Timestamp('2000-01-01 00:00:00'))"
+    )
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "method, method_args, expected_values",
+    [
+        ("sum", {}, [1, 0, 1]),
+        ("sum", {"min_count": 0}, [1, 0, 1]),
+        ("sum", {"min_count": 1}, [1, np.nan, 1]),
+        ("sum", {"min_count": 2}, [np.nan, np.nan, np.nan]),
+        ("prod", {}, [1, 1, 1]),
+        ("prod", {"min_count": 0}, [1, 1, 1]),
+        ("prod", {"min_count": 1}, [1, np.nan, 1]),
+        ("prod", {"min_count": 2}, [np.nan, np.nan, np.nan]),
+    ],
+)
+def test_upsample_sum(method, method_args, expected_values):
+    ser = Series(1, index=date_range("2017", periods=2, freq="h"))
+    resampled = ser.resample("30min")
+    index = pd.DatetimeIndex(
+        ["2017-01-01T00:00:00", "2017-01-01T00:30:00", "2017-01-01T01:00:00"],
+        dtype="M8[ns]",
+        freq="30min",
+    )
+    result = methodcaller(method, **method_args)(resampled)
+    expected = Series(expected_values, index=index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.fixture
+def groupy_test_df():
+    return DataFrame(
+        {"price": [10, 11, 9], "volume": [50, 60, 50]},
+        index=date_range("01/01/2018", periods=3, freq="W"),
+    )
+
+
+def test_groupby_resample_interpolate_raises(groupy_test_df):
+    # GH 35325
+
+    # Make a copy of the test data frame that has index.name=None
+    groupy_test_df_without_index_name = groupy_test_df.copy()
+    groupy_test_df_without_index_name.index.name = None
+
+    dfs = [groupy_test_df, groupy_test_df_without_index_name]
+
+    for df in dfs:
+        with pytest.raises(
+            NotImplementedError,
+            match="Direct interpolation of MultiIndex data frames is not supported",
+        ):
+            df.groupby("volume").resample("1D").interpolate(method="linear")
+
+
+def test_groupby_resample_interpolate_with_apply_syntax(groupy_test_df):
+    # GH 35325
+
+    # Make a copy of the test data frame that has index.name=None
+    groupy_test_df_without_index_name = groupy_test_df.copy()
+    groupy_test_df_without_index_name.index.name = None
+
+    dfs = [groupy_test_df, groupy_test_df_without_index_name]
+
+    for df in dfs:
+        result = df.groupby("volume").apply(
+            lambda x: x.resample("1D").interpolate(method="linear"),
+        )
+
+        volume = [50] * 15 + [60]
+        week_starting = list(date_range("2018-01-07", "2018-01-21")) + [
+            Timestamp("2018-01-14")
+        ]
+        expected_ind = pd.MultiIndex.from_arrays(
+            [volume, week_starting],
+            names=["volume", df.index.name],
+        )
+
+        expected = DataFrame(
+            data={
+                "price": [
+                    10.0,
+                    9.928571428571429,
+                    9.857142857142858,
+                    9.785714285714286,
+                    9.714285714285714,
+                    9.642857142857142,
+                    9.571428571428571,
+                    9.5,
+                    9.428571428571429,
+                    9.357142857142858,
+                    9.285714285714286,
+                    9.214285714285714,
+                    9.142857142857142,
+                    9.071428571428571,
+                    9.0,
+                    11.0,
+                ]
+            },
+            index=expected_ind,
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_interpolate_with_apply_syntax_off_grid(groupy_test_df):
+    """Similar test as test_groupby_resample_interpolate_with_apply_syntax but
+    with resampling that results in missing anchor points when interpolating.
+    See GH#21351."""
+    # GH#21351
+    result = groupy_test_df.groupby("volume").apply(
+        lambda x: x.resample("265h").interpolate(method="linear")
+    )
+
+    volume = [50, 50, 60]
+    week_starting = pd.DatetimeIndex(
+        [
+            Timestamp("2018-01-07"),
+            Timestamp("2018-01-18 01:00:00"),
+            Timestamp("2018-01-14"),
+        ]
+    ).as_unit("ns")
+    expected_ind = pd.MultiIndex.from_arrays(
+        [volume, week_starting],
+        names=["volume", "week_starting"],
+    )
+
+    expected = DataFrame(
+        data={
+            "price": [
+                10.0,
+                9.21131,
+                11.0,
+            ]
+        },
+        index=expected_ind,
+    )
+    tm.assert_frame_equal(result, expected, check_names=False)
diff --git a/pandas/pandas/tests/resample/test_timedelta.py b/pandas/pandas/tests/resample/test_timedelta.py
new file mode 100644
index 0000000000000..309810b656ed3
--- /dev/null
+++ b/pandas/pandas/tests/resample/test_timedelta.py
@@ -0,0 +1,219 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.indexes.timedeltas import timedelta_range
+
+
+def test_asfreq_bug():
+    df = DataFrame(data=[1, 3], index=[timedelta(), timedelta(minutes=3)])
+    result = df.resample("1min").asfreq()
+    expected = DataFrame(
+        data=[1, np.nan, np.nan, 3],
+        index=timedelta_range("0 day", periods=4, freq="1min"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_with_nat():
+    # GH 13223
+    index = pd.to_timedelta(["0s", pd.NaT, "2s"])
+    result = DataFrame({"value": [2, 3, 5]}, index).resample("1s").mean()
+    expected = DataFrame(
+        {"value": [2.5, np.nan, 5.0]},
+        index=timedelta_range("0 day", periods=3, freq="1s"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_as_freq_with_subperiod():
+    # GH 13022
+    index = timedelta_range("00:00:00", "00:10:00", freq="5min")
+    df = DataFrame(data={"value": [1, 5, 10]}, index=index)
+    result = df.resample("2min").asfreq()
+    expected_data = {"value": [1, np.nan, np.nan, np.nan, np.nan, 10]}
+    expected = DataFrame(
+        data=expected_data, index=timedelta_range("00:00:00", "00:10:00", freq="2min")
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_with_timedeltas():
+    expected = DataFrame({"A": np.arange(1480)})
+    expected = expected.groupby(expected.index // 30).sum()
+    expected.index = timedelta_range("0 days", freq="30min", periods=50)
+
+    df = DataFrame(
+        {"A": np.arange(1480)}, index=pd.to_timedelta(np.arange(1480), unit="min")
+    )
+    result = df.resample("30min").sum()
+
+    tm.assert_frame_equal(result, expected)
+
+    s = df["A"]
+    result = s.resample("30min").sum()
+    tm.assert_series_equal(result, expected["A"])
+
+
+def test_resample_single_period_timedelta():
+    s = Series(list(range(5)), index=timedelta_range("1 day", freq="s", periods=5))
+    result = s.resample("2s").sum()
+    expected = Series([1, 5, 4], index=timedelta_range("1 day", freq="2s", periods=3))
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_timedelta_idempotency():
+    # GH 12072
+    index = timedelta_range("0", periods=9, freq="10ms")
+    series = Series(range(9), index=index)
+    result = series.resample("10ms").mean()
+    expected = series.astype(float)
+    tm.assert_series_equal(result, expected)
+
+
+def test_resample_offset_with_timedeltaindex():
+    # GH 10530 & 31809
+    rng = timedelta_range(start="0s", periods=25, freq="s")
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    with_base = ts.resample("2s", offset="5s").mean()
+    without_base = ts.resample("2s").mean()
+
+    exp_without_base = timedelta_range(start="0s", end="25s", freq="2s")
+    exp_with_base = timedelta_range(start="5s", end="29s", freq="2s")
+
+    tm.assert_index_equal(without_base.index, exp_without_base)
+    tm.assert_index_equal(with_base.index, exp_with_base)
+
+
+def test_resample_categorical_data_with_timedeltaindex():
+    # GH #12169
+    df = DataFrame({"Group_obj": "A"}, index=pd.to_timedelta(list(range(20)), unit="s"))
+    df["Group"] = df["Group_obj"].astype("category")
+    result = df.resample("10s").agg(lambda x: (x.value_counts().index[0]))
+    exp_tdi = pd.TimedeltaIndex(np.array([0, 10], dtype="m8[s]"), freq="10s").as_unit(
+        "ns"
+    )
+    expected = DataFrame(
+        {"Group_obj": ["A", "A"], "Group": ["A", "A"]},
+        index=exp_tdi,
+    )
+    expected = expected.reindex(["Group_obj", "Group"], axis=1)
+    expected["Group"] = expected["Group_obj"].astype("category")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_timedelta_values():
+    # GH 13119
+    # check that timedelta dtype is preserved when NaT values are
+    # introduced by the resampling
+
+    times = timedelta_range("1 day", "6 day", freq="4D")
+    df = DataFrame({"time": times}, index=times)
+
+    times2 = timedelta_range("1 day", "6 day", freq="2D")
+    exp = Series(times2, index=times2, name="time")
+    exp.iloc[1] = pd.NaT
+
+    res = df.resample("2D").first()["time"]
+    tm.assert_series_equal(res, exp)
+    res = df["time"].resample("2D").first()
+    tm.assert_series_equal(res, exp)
+
+
+@pytest.mark.parametrize(
+    "start, end, freq, resample_freq",
+    [
+        ("8h", "21h59min50s", "10s", "3h"),  # GH 30353 example
+        ("3h", "22h", "1h", "5h"),
+        ("527D", "5006D", "3D", "10D"),
+        ("1D", "10D", "1D", "2D"),  # GH 13022 example
+        # tests that worked before GH 33498:
+        ("8h", "21h59min50s", "10s", "2h"),
+        ("0h", "21h59min50s", "10s", "3h"),
+        ("10D", "85D", "D", "2D"),
+    ],
+)
+def test_resample_timedelta_edge_case(start, end, freq, resample_freq):
+    # GH 33498
+    # check that the timedelta bins does not contains an extra bin
+    idx = timedelta_range(start=start, end=end, freq=freq)
+    s = Series(np.arange(len(idx)), index=idx)
+    result = s.resample(resample_freq).min()
+    expected_index = timedelta_range(freq=resample_freq, start=start, end=end)
+    tm.assert_index_equal(result.index, expected_index)
+    assert result.index.freq == expected_index.freq
+    assert not np.isnan(result.iloc[-1])
+
+
+@pytest.mark.parametrize("duplicates", [True, False])
+def test_resample_with_timedelta_yields_no_empty_groups(duplicates):
+    # GH 10603
+    df = DataFrame(
+        np.random.default_rng(2).normal(size=(10000, 4)),
+        index=timedelta_range(start="0s", periods=10000, freq="3906250ns"),
+    )
+    if duplicates:
+        # case with non-unique columns
+        df.columns = ["A", "B", "A", "C"]
+
+    result = df.loc["1s":, :].resample("3s").apply(lambda x: len(x))
+
+    expected = DataFrame(
+        [[768] * 4] * 12 + [[528] * 4],
+        index=timedelta_range(start="1s", periods=13, freq="3s"),
+    )
+    expected.columns = df.columns
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_quantile_timedelta(unit):
+    # GH: 29485
+    dtype = np.dtype(f"m8[{unit}]")
+    df = DataFrame(
+        {"value": pd.to_timedelta(np.arange(4), unit="s").astype(dtype)},
+        index=pd.date_range("20200101", periods=4, tz="UTC"),
+    )
+    result = df.resample("2D").quantile(0.99)
+    expected = DataFrame(
+        {
+            "value": [
+                pd.Timedelta("0 days 00:00:00.990000"),
+                pd.Timedelta("0 days 00:00:02.990000"),
+            ]
+        },
+        index=pd.date_range("20200101", periods=2, tz="UTC", freq="2D"),
+    ).astype(dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_resample_closed_right():
+    # GH#45414
+    idx = pd.Index([pd.Timedelta(seconds=120 + i * 30) for i in range(10)])
+    ser = Series(range(10), index=idx)
+    result = ser.resample("min", closed="right", label="right").sum()
+    expected = Series(
+        [0, 3, 7, 11, 15, 9],
+        index=pd.TimedeltaIndex(
+            [pd.Timedelta(seconds=120 + i * 60) for i in range(6)], freq="min"
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_arrow_duration_resample():
+    # GH 56371
+    idx = pd.Index(timedelta_range("1 day", periods=5), dtype="duration[ns][pyarrow]")
+    expected = Series(np.arange(5, dtype=np.float64), index=idx)
+    result = expected.resample("1D").mean()
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/__init__.py b/pandas/pandas/tests/reshape/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/reshape/concat/__init__.py b/pandas/pandas/tests/reshape/concat/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/reshape/concat/test_append.py b/pandas/pandas/tests/reshape/concat/test_append.py
new file mode 100644
index 0000000000000..6e18ccfc70e06
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_append.py
@@ -0,0 +1,376 @@
+import datetime as dt
+from itertools import combinations
+
+import dateutil
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    concat,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestAppend:
+    def test_append(self, sort, float_frame):
+        mixed_frame = float_frame.copy()
+        mixed_frame["foo"] = "bar"
+
+        begin_index = float_frame.index[:5]
+        end_index = float_frame.index[5:]
+
+        begin_frame = float_frame.reindex(begin_index)
+        end_frame = float_frame.reindex(end_index)
+
+        appended = begin_frame._append(end_frame)
+        tm.assert_almost_equal(appended["A"], float_frame["A"])
+
+        del end_frame["A"]
+        partial_appended = begin_frame._append(end_frame, sort=sort)
+        assert "A" in partial_appended
+
+        partial_appended = end_frame._append(begin_frame, sort=sort)
+        assert "A" in partial_appended
+
+        # mixed type handling
+        appended = mixed_frame[:5]._append(mixed_frame[5:])
+        tm.assert_frame_equal(appended, mixed_frame)
+
+        # what to test here
+        mixed_appended = mixed_frame[:5]._append(float_frame[5:], sort=sort)
+        mixed_appended2 = float_frame[:5]._append(mixed_frame[5:], sort=sort)
+
+        # all equal except 'foo' column
+        tm.assert_frame_equal(
+            mixed_appended.reindex(columns=["A", "B", "C", "D"]),
+            mixed_appended2.reindex(columns=["A", "B", "C", "D"]),
+        )
+
+    def test_append_empty(self, float_frame):
+        empty = DataFrame()
+
+        appended = float_frame._append(empty)
+        tm.assert_frame_equal(float_frame, appended)
+        assert appended is not float_frame
+
+        appended = empty._append(float_frame)
+        tm.assert_frame_equal(float_frame, appended)
+        assert appended is not float_frame
+
+    def test_append_overlap_raises(self, float_frame):
+        msg = "Indexes have overlapping values"
+        with pytest.raises(ValueError, match=msg):
+            float_frame._append(float_frame, verify_integrity=True)
+
+    def test_append_new_columns(self):
+        # see gh-6129: new columns
+        df = DataFrame({"a": {"x": 1, "y": 2}, "b": {"x": 3, "y": 4}})
+        row = Series([5, 6, 7], index=["a", "b", "c"], name="z")
+        expected = DataFrame(
+            {
+                "a": {"x": 1, "y": 2, "z": 5},
+                "b": {"x": 3, "y": 4, "z": 6},
+                "c": {"z": 7},
+            }
+        )
+        result = df._append(row)
+        tm.assert_frame_equal(result, expected)
+
+    def test_append_length0_frame(self, sort):
+        df = DataFrame(columns=["A", "B", "C"])
+        df3 = DataFrame(index=[0, 1], columns=["A", "B"])
+        df5 = df._append(df3, sort=sort)
+
+        expected = DataFrame(index=[0, 1], columns=["A", "B", "C"])
+        tm.assert_frame_equal(df5, expected)
+
+    def test_append_records(self):
+        arr1 = np.zeros((2,), dtype=("i4,f4,S10"))
+        arr1[:] = [(1, 2.0, "Hello"), (2, 3.0, "World")]
+
+        arr2 = np.zeros((3,), dtype=("i4,f4,S10"))
+        arr2[:] = [(3, 4.0, "foo"), (5, 6.0, "bar"), (7.0, 8.0, "baz")]
+
+        df1 = DataFrame(arr1)
+        df2 = DataFrame(arr2)
+
+        result = df1._append(df2, ignore_index=True)
+        expected = DataFrame(np.concatenate((arr1, arr2)))
+        tm.assert_frame_equal(result, expected)
+
+    # rewrite sort fixture, since we also want to test default of None
+    def test_append_sorts(self, sort):
+        df1 = DataFrame({"a": [1, 2], "b": [1, 2]}, columns=["b", "a"])
+        df2 = DataFrame({"a": [1, 2], "c": [3, 4]}, index=[2, 3])
+
+        result = df1._append(df2, sort=sort)
+
+        # for None / True
+        expected = DataFrame(
+            {"b": [1, 2, None, None], "a": [1, 2, 1, 2], "c": [None, None, 3, 4]},
+            columns=["a", "b", "c"],
+        )
+        if sort is False:
+            expected = expected[["b", "a", "c"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_append_different_columns(self, sort):
+        df = DataFrame(
+            {
+                "bools": np.random.default_rng(2).standard_normal(10) > 0,
+                "ints": np.random.default_rng(2).integers(0, 10, 10),
+                "floats": np.random.default_rng(2).standard_normal(10),
+                "strings": ["foo", "bar"] * 5,
+            }
+        )
+
+        a = df[:5].loc[:, ["bools", "ints", "floats"]]
+        b = df[5:].loc[:, ["strings", "ints", "floats"]]
+
+        appended = a._append(b, sort=sort)
+        assert isna(appended["strings"][0:4]).all()
+        assert isna(appended["bools"][5:]).all()
+
+    def test_append_many(self, sort, float_frame):
+        chunks = [
+            float_frame[:5],
+            float_frame[5:10],
+            float_frame[10:15],
+            float_frame[15:],
+        ]
+
+        result = chunks[0]._append(chunks[1:])
+        tm.assert_frame_equal(result, float_frame)
+
+        chunks[-1] = chunks[-1].copy()
+        chunks[-1]["foo"] = "bar"
+        result = chunks[0]._append(chunks[1:], sort=sort)
+        tm.assert_frame_equal(result.loc[:, float_frame.columns], float_frame)
+        assert (result["foo"][15:] == "bar").all()
+        assert result["foo"][:15].isna().all()
+
+    def test_append_preserve_index_name(self):
+        # #980
+        df1 = DataFrame(columns=["A", "B", "C"])
+        df1 = df1.set_index(["A"])
+        df2 = DataFrame(data=[[1, 4, 7], [2, 5, 8], [3, 6, 9]], columns=["A", "B", "C"])
+        df2 = df2.set_index(["A"])
+
+        result = df1._append(df2)
+        assert result.index.name == "A"
+
+    indexes_can_append = [
+        pd.RangeIndex(3),
+        Index([4, 5, 6]),
+        Index([4.5, 5.5, 6.5]),
+        Index(list("abc")),
+        pd.CategoricalIndex("A B C".split()),
+        pd.CategoricalIndex("D E F".split(), ordered=True),
+        pd.IntervalIndex.from_breaks([7, 8, 9, 10]),
+        pd.DatetimeIndex(
+            [
+                dt.datetime(2013, 1, 3, 0, 0),
+                dt.datetime(2013, 1, 3, 6, 10),
+                dt.datetime(2013, 1, 3, 7, 12),
+            ]
+        ),
+        pd.MultiIndex.from_arrays(["A B C".split(), "D E F".split()]),
+    ]
+
+    @pytest.mark.parametrize(
+        "index", indexes_can_append, ids=lambda x: type(x).__name__
+    )
+    def test_append_same_columns_type(self, index):
+        # GH18359
+
+        # df wider than ser
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=index)
+        ser_index = index[:2]
+        ser = Series([7, 8], index=ser_index, name=2)
+        result = df._append(ser)
+        expected = DataFrame(
+            [[1, 2, 3.0], [4, 5, 6], [7, 8, np.nan]], index=[0, 1, 2], columns=index
+        )
+        # integer dtype is preserved for columns present in ser.index
+        assert expected.dtypes.iloc[0].kind == "i"
+        assert expected.dtypes.iloc[1].kind == "i"
+
+        tm.assert_frame_equal(result, expected)
+
+        # ser wider than df
+        ser_index = index
+        index = index[:2]
+        df = DataFrame([[1, 2], [4, 5]], columns=index)
+        ser = Series([7, 8, 9], index=ser_index, name=2)
+        result = df._append(ser)
+        expected = DataFrame(
+            [[1, 2, np.nan], [4, 5, np.nan], [7, 8, 9]],
+            index=[0, 1, 2],
+            columns=ser_index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "df_columns, series_index",
+        combinations(indexes_can_append, r=2),
+        ids=lambda x: type(x).__name__,
+    )
+    def test_append_different_columns_types(self, df_columns, series_index):
+        # GH18359
+        # See also test 'test_append_different_columns_types_raises' below
+        # for errors raised when appending
+
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=df_columns)
+        ser = Series([7, 8, 9], index=series_index, name=2)
+
+        result = df._append(ser)
+        idx_diff = ser.index.difference(df_columns)
+        combined_columns = Index(df_columns.tolist()).append(idx_diff)
+        expected = DataFrame(
+            [
+                [1.0, 2.0, 3.0, np.nan, np.nan, np.nan],
+                [4, 5, 6, np.nan, np.nan, np.nan],
+                [np.nan, np.nan, np.nan, 7, 8, 9],
+            ],
+            index=[0, 1, 2],
+            columns=combined_columns,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_append_dtype_coerce(self, sort):
+        # GH 4993
+        # appending with datetime will incorrectly convert datetime64
+
+        df1 = DataFrame(
+            index=[1, 2],
+            data=[dt.datetime(2013, 1, 1, 0, 0), dt.datetime(2013, 1, 2, 0, 0)],
+            columns=["start_time"],
+        )
+        df2 = DataFrame(
+            index=[4, 5],
+            data=[
+                [dt.datetime(2013, 1, 3, 0, 0), dt.datetime(2013, 1, 3, 6, 10)],
+                [dt.datetime(2013, 1, 4, 0, 0), dt.datetime(2013, 1, 4, 7, 10)],
+            ],
+            columns=["start_time", "end_time"],
+        )
+
+        expected = concat(
+            [
+                Series(
+                    [
+                        pd.NaT,
+                        pd.NaT,
+                        dt.datetime(2013, 1, 3, 6, 10),
+                        dt.datetime(2013, 1, 4, 7, 10),
+                    ],
+                    name="end_time",
+                ),
+                Series(
+                    [
+                        dt.datetime(2013, 1, 1, 0, 0),
+                        dt.datetime(2013, 1, 2, 0, 0),
+                        dt.datetime(2013, 1, 3, 0, 0),
+                        dt.datetime(2013, 1, 4, 0, 0),
+                    ],
+                    name="start_time",
+                ),
+            ],
+            axis=1,
+            sort=sort,
+        )
+        result = df1._append(df2, ignore_index=True, sort=sort)
+        if sort:
+            expected = expected[["end_time", "start_time"]]
+        else:
+            expected = expected[["start_time", "end_time"]]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_append_missing_column_proper_upcast(self, sort):
+        df1 = DataFrame({"A": np.array([1, 2, 3, 4], dtype="i8")})
+        df2 = DataFrame({"B": np.array([True, False, True, False], dtype=bool)})
+
+        appended = df1._append(df2, ignore_index=True, sort=sort)
+        assert appended["A"].dtype == "f8"
+        assert appended["B"].dtype == "O"
+
+    def test_append_empty_frame_to_series_with_dateutil_tz(self):
+        # GH 23682
+        date = Timestamp("2018-10-24 07:30:00", tz=dateutil.tz.tzutc())
+        ser = Series({"a": 1.0, "b": 2.0, "date": date})
+        df = DataFrame(columns=["c", "d"])
+        result_a = df._append(ser, ignore_index=True)
+        expected = DataFrame(
+            [[np.nan, np.nan, 1.0, 2.0, date]], columns=["c", "d", "a", "b", "date"]
+        )
+        # These columns get cast to object after append
+        expected["c"] = expected["c"].astype(object)
+        expected["d"] = expected["d"].astype(object)
+        tm.assert_frame_equal(result_a, expected)
+
+        expected = DataFrame(
+            [[np.nan, np.nan, 1.0, 2.0, date]] * 2, columns=["c", "d", "a", "b", "date"]
+        )
+        expected["c"] = expected["c"].astype(object)
+        expected["d"] = expected["d"].astype(object)
+        result_b = result_a._append(ser, ignore_index=True)
+        tm.assert_frame_equal(result_b, expected)
+
+        result = df._append([ser, ser], ignore_index=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_append_empty_tz_frame_with_datetime64ns(self):
+        # https://github.com/pandas-dev/pandas/issues/35460
+        df = DataFrame(columns=["a"]).astype("datetime64[ns, UTC]")
+
+        # pd.NaT gets inferred as tz-naive, so append result is tz-naive
+        result = df._append({"a": pd.NaT}, ignore_index=True)
+        expected = DataFrame({"a": [pd.NaT]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
+
+        # also test with typed value to append
+        df = DataFrame(columns=["a"]).astype("datetime64[ns, UTC]")
+        other = Series({"a": pd.NaT}, dtype="datetime64[ns]")
+        result = df._append(other, ignore_index=True)
+        tm.assert_frame_equal(result, expected)
+
+        # mismatched tz
+        other = Series({"a": pd.NaT}, dtype="datetime64[ns, US/Pacific]")
+        result = df._append(other, ignore_index=True)
+        expected = DataFrame({"a": [pd.NaT]}).astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype_str", ["datetime64[ns, UTC]", "datetime64[ns]", "Int64", "int64"]
+    )
+    @pytest.mark.parametrize("val", [1, "NaT"])
+    def test_append_empty_frame_with_timedelta64ns_nat(self, dtype_str, val):
+        # https://github.com/pandas-dev/pandas/issues/35460
+        df = DataFrame(columns=["a"]).astype(dtype_str)
+
+        other = DataFrame({"a": [np.timedelta64(val, "ns")]})
+        result = df._append(other, ignore_index=True)
+
+        expected = other.astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype_str", ["datetime64[ns, UTC]", "datetime64[ns]", "Int64", "int64"]
+    )
+    @pytest.mark.parametrize("val", [1, "NaT"])
+    def test_append_frame_with_timedelta64ns_nat(self, dtype_str, val):
+        # https://github.com/pandas-dev/pandas/issues/35460
+        df = DataFrame({"a": pd.array([1], dtype=dtype_str)})
+
+        other = DataFrame({"a": [np.timedelta64(val, "ns")]})
+        result = df._append(other, ignore_index=True)
+
+        expected = DataFrame({"a": [df.iloc[0, 0], other.iloc[0, 0]]}, dtype=object)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_append_common.py b/pandas/pandas/tests/reshape/concat/test_append_common.py
new file mode 100644
index 0000000000000..d0ff950e7985f
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_append_common.py
@@ -0,0 +1,750 @@
+import zoneinfo
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(
+    params=list(
+        {
+            "bool": [True, False, True],
+            "int64": [1, 2, 3],
+            "float64": [1.1, np.nan, 3.3],
+            "category": Categorical(["X", "Y", "Z"]),
+            "object": ["a", "b", "c"],
+            "datetime64[s]": [
+                pd.Timestamp("2011-01-01"),
+                pd.Timestamp("2011-01-02"),
+                pd.Timestamp("2011-01-03"),
+            ],
+            "datetime64[s, US/Eastern]": [
+                pd.Timestamp("2011-01-01", tz="US/Eastern"),
+                pd.Timestamp("2011-01-02", tz="US/Eastern"),
+                pd.Timestamp("2011-01-03", tz="US/Eastern"),
+            ],
+            "timedelta64[ns]": [
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+                pd.Timedelta("3 days"),
+            ],
+            "period[M]": [
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02", freq="M"),
+                pd.Period("2011-03", freq="M"),
+            ],
+        }.items()
+    )
+)
+def item(request):
+    key, data = request.param
+    return key, data
+
+
+@pytest.fixture
+def item2(item):
+    return item
+
+
+class TestConcatAppendCommon:
+    """
+    Test common dtype coercion rules between concat and append.
+    """
+
+    def test_dtypes(self, item, index_or_series, using_infer_string):
+        # to confirm test case covers intended dtypes
+        typ, vals = item
+        obj = index_or_series(vals)
+        if typ == "object" and using_infer_string:
+            typ = "string"
+        if isinstance(obj, Index):
+            assert obj.dtype == typ
+        elif isinstance(obj, Series):
+            if typ.startswith("period"):
+                assert obj.dtype == "Period[M]"
+            else:
+                assert obj.dtype == typ
+
+    def test_concatlike_same_dtypes(self, item):
+        # GH 13660
+        typ1, vals1 = item
+
+        vals2 = vals1
+        vals3 = vals1
+
+        if typ1 == "category":
+            exp_data = Categorical(list(vals1) + list(vals2))
+            exp_data3 = Categorical(list(vals1) + list(vals2) + list(vals3))
+        else:
+            exp_data = vals1 + vals2
+            exp_data3 = vals1 + vals2 + vals3
+
+        # ----- Index ----- #
+
+        # index.append
+        res = Index(vals1).append(Index(vals2))
+        exp = Index(exp_data)
+        tm.assert_index_equal(res, exp)
+
+        # 3 elements
+        res = Index(vals1).append([Index(vals2), Index(vals3)])
+        exp = Index(exp_data3)
+        tm.assert_index_equal(res, exp)
+
+        # index.append name mismatch
+        i1 = Index(vals1, name="x")
+        i2 = Index(vals2, name="y")
+        res = i1.append(i2)
+        exp = Index(exp_data)
+        tm.assert_index_equal(res, exp)
+
+        # index.append name match
+        i1 = Index(vals1, name="x")
+        i2 = Index(vals2, name="x")
+        res = i1.append(i2)
+        exp = Index(exp_data, name="x")
+        tm.assert_index_equal(res, exp)
+
+        # cannot append non-index
+        with pytest.raises(TypeError, match="all inputs must be Index"):
+            Index(vals1).append(vals2)
+
+        with pytest.raises(TypeError, match="all inputs must be Index"):
+            Index(vals1).append([Index(vals2), vals3])
+
+        # ----- Series ----- #
+
+        # series.append
+        res = Series(vals1)._append(Series(vals2), ignore_index=True)
+        exp = Series(exp_data)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # concat
+        res = pd.concat([Series(vals1), Series(vals2)], ignore_index=True)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # 3 elements
+        res = Series(vals1)._append([Series(vals2), Series(vals3)], ignore_index=True)
+        exp = Series(exp_data3)
+        tm.assert_series_equal(res, exp)
+
+        res = pd.concat(
+            [Series(vals1), Series(vals2), Series(vals3)],
+            ignore_index=True,
+        )
+        tm.assert_series_equal(res, exp)
+
+        # name mismatch
+        s1 = Series(vals1, name="x")
+        s2 = Series(vals2, name="y")
+        res = s1._append(s2, ignore_index=True)
+        exp = Series(exp_data)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        res = pd.concat([s1, s2], ignore_index=True)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # name match
+        s1 = Series(vals1, name="x")
+        s2 = Series(vals2, name="x")
+        res = s1._append(s2, ignore_index=True)
+        exp = Series(exp_data, name="x")
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        res = pd.concat([s1, s2], ignore_index=True)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # cannot append non-index
+        msg = (
+            r"cannot concatenate object of type '.+'; "
+            "only Series and DataFrame objs are valid"
+        )
+        with pytest.raises(TypeError, match=msg):
+            Series(vals1)._append(vals2)
+
+        with pytest.raises(TypeError, match=msg):
+            Series(vals1)._append([Series(vals2), vals3])
+
+        with pytest.raises(TypeError, match=msg):
+            pd.concat([Series(vals1), vals2])
+
+        with pytest.raises(TypeError, match=msg):
+            pd.concat([Series(vals1), Series(vals2), vals3])
+
+    def test_concatlike_dtypes_coercion(self, item, item2, request):
+        # GH 13660
+        typ1, vals1 = item
+        typ2, vals2 = item2
+
+        vals3 = vals2
+
+        # basically infer
+        exp_index_dtype = None
+        exp_series_dtype = None
+
+        if typ1 == typ2:
+            pytest.skip("same dtype is tested in test_concatlike_same_dtypes")
+        elif typ1 == "category" or typ2 == "category":
+            pytest.skip("categorical type tested elsewhere")
+
+        # specify expected dtype
+        if typ1 == "bool" and typ2 in ("int64", "float64"):
+            # series coerces to numeric based on numpy rule
+            # index doesn't because bool is object dtype
+            exp_series_dtype = typ2
+            mark = pytest.mark.xfail(reason="GH#39187 casting to object")
+            request.applymarker(mark)
+        elif typ2 == "bool" and typ1 in ("int64", "float64"):
+            exp_series_dtype = typ1
+            mark = pytest.mark.xfail(reason="GH#39187 casting to object")
+            request.applymarker(mark)
+        elif typ1 in {"datetime64[ns, US/Eastern]", "timedelta64[ns]"} or typ2 in {
+            "datetime64[ns, US/Eastern]",
+            "timedelta64[ns]",
+        }:
+            exp_index_dtype = object
+            exp_series_dtype = object
+
+        exp_data = vals1 + vals2
+        exp_data3 = vals1 + vals2 + vals3
+
+        # ----- Index ----- #
+
+        # index.append
+        # GH#39817
+        res = Index(vals1).append(Index(vals2))
+        exp = Index(exp_data, dtype=exp_index_dtype)
+        tm.assert_index_equal(res, exp)
+
+        # 3 elements
+        res = Index(vals1).append([Index(vals2), Index(vals3)])
+        exp = Index(exp_data3, dtype=exp_index_dtype)
+        tm.assert_index_equal(res, exp)
+
+        # ----- Series ----- #
+
+        # series._append
+        # GH#39817
+        res = Series(vals1)._append(Series(vals2), ignore_index=True)
+        exp = Series(exp_data, dtype=exp_series_dtype)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # concat
+        # GH#39817
+        res = pd.concat([Series(vals1), Series(vals2)], ignore_index=True)
+        tm.assert_series_equal(res, exp, check_index_type=True)
+
+        # 3 elements
+        # GH#39817
+        res = Series(vals1)._append([Series(vals2), Series(vals3)], ignore_index=True)
+        exp = Series(exp_data3, dtype=exp_series_dtype)
+        tm.assert_series_equal(res, exp)
+
+        # GH#39817
+        res = pd.concat(
+            [Series(vals1), Series(vals2), Series(vals3)],
+            ignore_index=True,
+        )
+        tm.assert_series_equal(res, exp)
+
+    def test_concatlike_common_coerce_to_pandas_object(self):
+        # GH 13626
+        # result must be Timestamp/Timedelta, not datetime.datetime/timedelta
+        dti = pd.DatetimeIndex(["2011-01-01", "2011-01-02"])
+        tdi = pd.TimedeltaIndex(["1 days", "2 days"])
+
+        exp = Index(
+            [
+                pd.Timestamp("2011-01-01"),
+                pd.Timestamp("2011-01-02"),
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+            ]
+        )
+
+        res = dti.append(tdi)
+        tm.assert_index_equal(res, exp)
+        assert isinstance(res[0], pd.Timestamp)
+        assert isinstance(res[-1], pd.Timedelta)
+
+        dts = Series(dti)
+        tds = Series(tdi)
+        res = dts._append(tds)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+        assert isinstance(res.iloc[0], pd.Timestamp)
+        assert isinstance(res.iloc[-1], pd.Timedelta)
+
+        res = pd.concat([dts, tds])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+        assert isinstance(res.iloc[0], pd.Timestamp)
+        assert isinstance(res.iloc[-1], pd.Timedelta)
+
+    def test_concatlike_datetimetz(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH 7795
+        dti1 = pd.DatetimeIndex(["2011-01-01", "2011-01-02"], tz=tz)
+        dti2 = pd.DatetimeIndex(["2012-01-01", "2012-01-02"], tz=tz)
+
+        exp = pd.DatetimeIndex(
+            ["2011-01-01", "2011-01-02", "2012-01-01", "2012-01-02"], tz=tz
+        )
+
+        res = dti1.append(dti2)
+        tm.assert_index_equal(res, exp)
+
+        dts1 = Series(dti1)
+        dts2 = Series(dti2)
+        res = dts1._append(dts2)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([dts1, dts2])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+    @pytest.mark.parametrize("tz", ["UTC", "US/Eastern", "Asia/Tokyo", "EST5EDT"])
+    def test_concatlike_datetimetz_short(self, tz):
+        # GH#7795
+        ix1 = pd.date_range(start="2014-07-15", end="2014-07-17", freq="D", tz=tz)
+        ix2 = pd.DatetimeIndex(["2014-07-11", "2014-07-21"], tz=tz)
+        df1 = DataFrame(0, index=ix1, columns=["A", "B"])
+        df2 = DataFrame(0, index=ix2, columns=["A", "B"])
+
+        exp_idx = pd.DatetimeIndex(
+            ["2014-07-15", "2014-07-16", "2014-07-17", "2014-07-11", "2014-07-21"],
+            tz=tz,
+        ).as_unit("ns")
+        exp = DataFrame(0, index=exp_idx, columns=["A", "B"])
+
+        tm.assert_frame_equal(df1._append(df2), exp)
+        tm.assert_frame_equal(pd.concat([df1, df2]), exp)
+
+    def test_concatlike_datetimetz_to_object(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH 13660
+
+        # different tz coerces to object
+        dti1 = pd.DatetimeIndex(["2011-01-01", "2011-01-02"], tz=tz)
+        dti2 = pd.DatetimeIndex(["2012-01-01", "2012-01-02"])
+
+        exp = Index(
+            [
+                pd.Timestamp("2011-01-01", tz=tz),
+                pd.Timestamp("2011-01-02", tz=tz),
+                pd.Timestamp("2012-01-01"),
+                pd.Timestamp("2012-01-02"),
+            ],
+            dtype=object,
+        )
+
+        res = dti1.append(dti2)
+        tm.assert_index_equal(res, exp)
+
+        dts1 = Series(dti1)
+        dts2 = Series(dti2)
+        res = dts1._append(dts2)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([dts1, dts2])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        # different tz
+        tz_diff = zoneinfo.ZoneInfo("US/Hawaii")
+        dti3 = pd.DatetimeIndex(["2012-01-01", "2012-01-02"], tz=tz_diff)
+
+        exp = Index(
+            [
+                pd.Timestamp("2011-01-01", tz=tz),
+                pd.Timestamp("2011-01-02", tz=tz),
+                pd.Timestamp("2012-01-01", tz=tz_diff),
+                pd.Timestamp("2012-01-02", tz=tz_diff),
+            ],
+            dtype=object,
+        )
+
+        res = dti1.append(dti3)
+        tm.assert_index_equal(res, exp)
+
+        dts1 = Series(dti1)
+        dts3 = Series(dti3)
+        res = dts1._append(dts3)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([dts1, dts3])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+    def test_concatlike_common_period(self):
+        # GH 13660
+        pi1 = pd.PeriodIndex(["2011-01", "2011-02"], freq="M")
+        pi2 = pd.PeriodIndex(["2012-01", "2012-02"], freq="M")
+
+        exp = pd.PeriodIndex(["2011-01", "2011-02", "2012-01", "2012-02"], freq="M")
+
+        res = pi1.append(pi2)
+        tm.assert_index_equal(res, exp)
+
+        ps1 = Series(pi1)
+        ps2 = Series(pi2)
+        res = ps1._append(ps2)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([ps1, ps2])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+    def test_concatlike_common_period_diff_freq_to_object(self):
+        # GH 13221
+        pi1 = pd.PeriodIndex(["2011-01", "2011-02"], freq="M")
+        pi2 = pd.PeriodIndex(["2012-01-01", "2012-02-01"], freq="D")
+
+        exp = Index(
+            [
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02", freq="M"),
+                pd.Period("2012-01-01", freq="D"),
+                pd.Period("2012-02-01", freq="D"),
+            ],
+            dtype=object,
+        )
+
+        res = pi1.append(pi2)
+        tm.assert_index_equal(res, exp)
+
+        ps1 = Series(pi1)
+        ps2 = Series(pi2)
+        res = ps1._append(ps2)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([ps1, ps2])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+    def test_concatlike_common_period_mixed_dt_to_object(self):
+        # GH 13221
+        # different datetimelike
+        pi1 = pd.PeriodIndex(["2011-01", "2011-02"], freq="M")
+        tdi = pd.TimedeltaIndex(["1 days", "2 days"])
+        exp = Index(
+            [
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02", freq="M"),
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+            ],
+            dtype=object,
+        )
+
+        res = pi1.append(tdi)
+        tm.assert_index_equal(res, exp)
+
+        ps1 = Series(pi1)
+        tds = Series(tdi)
+        res = ps1._append(tds)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([ps1, tds])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        # inverse
+        exp = Index(
+            [
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02", freq="M"),
+            ],
+            dtype=object,
+        )
+
+        res = tdi.append(pi1)
+        tm.assert_index_equal(res, exp)
+
+        ps1 = Series(pi1)
+        tds = Series(tdi)
+        res = tds._append(ps1)
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+        res = pd.concat([tds, ps1])
+        tm.assert_series_equal(res, Series(exp, index=[0, 1, 0, 1]))
+
+    def test_concat_categorical(self):
+        # GH 13524
+
+        # same categories -> category
+        s1 = Series([1, 2, np.nan], dtype="category")
+        s2 = Series([2, 1, 2], dtype="category")
+
+        exp = Series([1, 2, np.nan, 2, 1, 2], dtype="category")
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        # partially different categories => not-category
+        s1 = Series([3, 2], dtype="category")
+        s2 = Series([2, 1], dtype="category")
+
+        exp = Series([3, 2, 2, 1])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        # completely different categories (same dtype) => not-category
+        s1 = Series([10, 11, np.nan], dtype="category")
+        s2 = Series([np.nan, 1, 3, 2], dtype="category")
+
+        exp = Series([10, 11, np.nan, np.nan, 1, 3, 2], dtype=np.float64)
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+    def test_union_categorical_same_categories_different_order(self):
+        # https://github.com/pandas-dev/pandas/issues/19096
+        a = Series(Categorical(["a", "b", "c"], categories=["a", "b", "c"]))
+        b = Series(Categorical(["a", "b", "c"], categories=["b", "a", "c"]))
+        result = pd.concat([a, b], ignore_index=True)
+        expected = Series(
+            Categorical(["a", "b", "c", "a", "b", "c"], categories=["a", "b", "c"])
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_categorical_coercion(self):
+        # GH 13524
+
+        # category + not-category => not-category
+        s1 = Series([1, 2, np.nan], dtype="category")
+        s2 = Series([2, 1, 2])
+
+        exp = Series([1, 2, np.nan, 2, 1, 2], dtype=np.float64)
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        # result shouldn't be affected by 1st elem dtype
+        exp = Series([2, 1, 2, 1, 2, np.nan], dtype=np.float64)
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        # all values are not in category => not-category
+        s1 = Series([3, 2], dtype="category")
+        s2 = Series([2, 1])
+
+        exp = Series([3, 2, 2, 1])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        exp = Series([2, 1, 3, 2])
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        # completely different categories => not-category
+        s1 = Series([10, 11, np.nan], dtype="category")
+        s2 = Series([1, 3, 2])
+
+        exp = Series([10, 11, np.nan, 1, 3, 2], dtype=np.float64)
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        exp = Series([1, 3, 2, 10, 11, np.nan], dtype=np.float64)
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        # different dtype => not-category
+        s1 = Series([10, 11, np.nan], dtype="category")
+        s2 = Series(["a", "b", "c"])
+
+        exp = Series([10, 11, np.nan, "a", "b", "c"])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        exp = Series(["a", "b", "c", 10, 11, np.nan])
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        # if normal series only contains NaN-likes => not-category
+        s1 = Series([10, 11], dtype="category")
+        s2 = Series([np.nan, np.nan, np.nan])
+
+        exp = Series([10, 11, np.nan, np.nan, np.nan])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        exp = Series([np.nan, np.nan, np.nan, 10, 11])
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+    def test_concat_categorical_3elem_coercion(self):
+        # GH 13524
+
+        # mixed dtypes => not-category
+        s1 = Series([1, 2, np.nan], dtype="category")
+        s2 = Series([2, 1, 2], dtype="category")
+        s3 = Series([1, 2, 1, 2, np.nan])
+
+        exp = Series([1, 2, np.nan, 2, 1, 2, 1, 2, 1, 2, np.nan], dtype="float")
+        tm.assert_series_equal(pd.concat([s1, s2, s3], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append([s2, s3], ignore_index=True), exp)
+
+        exp = Series([1, 2, 1, 2, np.nan, 1, 2, np.nan, 2, 1, 2], dtype="float")
+        tm.assert_series_equal(pd.concat([s3, s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s3._append([s1, s2], ignore_index=True), exp)
+
+        # values are all in either category => not-category
+        s1 = Series([4, 5, 6], dtype="category")
+        s2 = Series([1, 2, 3], dtype="category")
+        s3 = Series([1, 3, 4])
+
+        exp = Series([4, 5, 6, 1, 2, 3, 1, 3, 4])
+        tm.assert_series_equal(pd.concat([s1, s2, s3], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append([s2, s3], ignore_index=True), exp)
+
+        exp = Series([1, 3, 4, 4, 5, 6, 1, 2, 3])
+        tm.assert_series_equal(pd.concat([s3, s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s3._append([s1, s2], ignore_index=True), exp)
+
+        # values are all in either category => not-category
+        s1 = Series([4, 5, 6], dtype="category")
+        s2 = Series([1, 2, 3], dtype="category")
+        s3 = Series([10, 11, 12])
+
+        exp = Series([4, 5, 6, 1, 2, 3, 10, 11, 12])
+        tm.assert_series_equal(pd.concat([s1, s2, s3], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append([s2, s3], ignore_index=True), exp)
+
+        exp = Series([10, 11, 12, 4, 5, 6, 1, 2, 3])
+        tm.assert_series_equal(pd.concat([s3, s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s3._append([s1, s2], ignore_index=True), exp)
+
+    def test_concat_categorical_multi_coercion(self):
+        # GH 13524
+
+        s1 = Series([1, 3], dtype="category")
+        s2 = Series([3, 4], dtype="category")
+        s3 = Series([2, 3])
+        s4 = Series([2, 2], dtype="category")
+        s5 = Series([1, np.nan])
+        s6 = Series([1, 3, 2], dtype="category")
+
+        # mixed dtype, values are all in categories => not-category
+        exp = Series([1, 3, 3, 4, 2, 3, 2, 2, 1, np.nan, 1, 3, 2])
+        res = pd.concat([s1, s2, s3, s4, s5, s6], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+        res = s1._append([s2, s3, s4, s5, s6], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+
+        exp = Series([1, 3, 2, 1, np.nan, 2, 2, 2, 3, 3, 4, 1, 3])
+        res = pd.concat([s6, s5, s4, s3, s2, s1], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+        res = s6._append([s5, s4, s3, s2, s1], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+
+    def test_concat_categorical_ordered(self):
+        # GH 13524
+
+        s1 = Series(Categorical([1, 2, np.nan], ordered=True))
+        s2 = Series(Categorical([2, 1, 2], ordered=True))
+
+        exp = Series(Categorical([1, 2, np.nan, 2, 1, 2], ordered=True))
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        exp = Series(Categorical([1, 2, np.nan, 2, 1, 2, 1, 2, np.nan], ordered=True))
+        tm.assert_series_equal(pd.concat([s1, s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append([s2, s1], ignore_index=True), exp)
+
+    def test_concat_categorical_coercion_nan(self):
+        # GH 13524
+
+        # some edge cases
+        # category + not-category => not category
+        s1 = Series(np.array([np.nan, np.nan], dtype=np.float64), dtype="category")
+        s2 = Series([np.nan, 1])
+
+        exp = Series([np.nan, np.nan, np.nan, 1])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        s1 = Series([1, np.nan], dtype="category")
+        s2 = Series([np.nan, np.nan])
+
+        exp = Series([1, np.nan, np.nan, np.nan], dtype="float")
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        # mixed dtype, all nan-likes => not-category
+        s1 = Series([np.nan, np.nan], dtype="category")
+        s2 = Series([np.nan, np.nan])
+
+        exp = Series([np.nan, np.nan, np.nan, np.nan])
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        # all category nan-likes => category
+        s1 = Series([np.nan, np.nan], dtype="category")
+        s2 = Series([np.nan, np.nan], dtype="category")
+
+        exp = Series([np.nan, np.nan, np.nan, np.nan], dtype="category")
+
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+    def test_concat_categorical_empty(self):
+        # GH 13524
+
+        s1 = Series([], dtype="category")
+        s2 = Series([1, 2], dtype="category")
+        exp = s2.astype(object)
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+        s1 = Series([], dtype="category")
+        s2 = Series([], dtype="category")
+
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), s2)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), s2)
+
+        s1 = Series([], dtype="category")
+        s2 = Series([], dtype="object")
+
+        # different dtype => not-category
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), s2)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), s2)
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), s2)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), s2)
+
+        s1 = Series([], dtype="category")
+        s2 = Series([np.nan, np.nan])
+
+        exp = Series([np.nan, np.nan], dtype=object)
+        tm.assert_series_equal(pd.concat([s1, s2], ignore_index=True), exp)
+        tm.assert_series_equal(s1._append(s2, ignore_index=True), exp)
+
+        tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
+        tm.assert_series_equal(s2._append(s1, ignore_index=True), exp)
+
+    def test_categorical_concat_append(self):
+        cat = Categorical(["a", "b"], categories=["a", "b"])
+        vals = [1, 2]
+        df = DataFrame({"cats": cat, "vals": vals})
+        cat2 = Categorical(["a", "b", "a", "b"], categories=["a", "b"])
+        vals2 = [1, 2, 1, 2]
+        exp = DataFrame({"cats": cat2, "vals": vals2}, index=Index([0, 1, 0, 1]))
+
+        tm.assert_frame_equal(pd.concat([df, df]), exp)
+        tm.assert_frame_equal(df._append(df), exp)
+
+        # GH 13524 can concat different categories
+        cat3 = Categorical(["a", "b"], categories=["a", "b", "c"])
+        vals3 = [1, 2]
+        df_different_categories = DataFrame({"cats": cat3, "vals": vals3})
+
+        res = pd.concat([df, df_different_categories], ignore_index=True)
+        exp = DataFrame({"cats": list("abab"), "vals": [1, 2, 1, 2]})
+        tm.assert_frame_equal(res, exp)
+
+        res = df._append(df_different_categories, ignore_index=True)
+        tm.assert_frame_equal(res, exp)
diff --git a/pandas/pandas/tests/reshape/concat/test_categorical.py b/pandas/pandas/tests/reshape/concat/test_categorical.py
new file mode 100644
index 0000000000000..8e6a14e6bfb8f
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_categorical.py
@@ -0,0 +1,271 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestCategoricalConcat:
+    def test_categorical_concat(self, sort):
+        # See GH 10177
+        df1 = DataFrame(
+            np.arange(18, dtype="int64").reshape(6, 3), columns=["a", "b", "c"]
+        )
+
+        df2 = DataFrame(np.arange(14, dtype="int64").reshape(7, 2), columns=["a", "c"])
+
+        cat_values = ["one", "one", "two", "one", "two", "two", "one"]
+        df2["h"] = Series(Categorical(cat_values))
+
+        res = pd.concat((df1, df2), axis=0, ignore_index=True, sort=sort)
+        exp = DataFrame(
+            {
+                "a": [0, 3, 6, 9, 12, 15, 0, 2, 4, 6, 8, 10, 12],
+                "b": [
+                    1,
+                    4,
+                    7,
+                    10,
+                    13,
+                    16,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                ],
+                "c": [2, 5, 8, 11, 14, 17, 1, 3, 5, 7, 9, 11, 13],
+                "h": [None] * 6 + cat_values,
+            }
+        )
+        exp["h"] = exp["h"].astype(df2["h"].dtype)
+        tm.assert_frame_equal(res, exp)
+
+    def test_categorical_concat_dtypes(self, using_infer_string):
+        # GH8143
+        index = ["cat", "obj", "num"]
+        cat = Categorical(["a", "b", "c"])
+        obj = Series(["a", "b", "c"])
+        num = Series([1, 2, 3])
+        df = pd.concat([Series(cat), obj, num], axis=1, keys=index)
+
+        result = df.dtypes == (object if not using_infer_string else "str")
+        expected = Series([False, True, False], index=index)
+        tm.assert_series_equal(result, expected)
+
+        result = df.dtypes == "int64"
+        expected = Series([False, False, True], index=index)
+        tm.assert_series_equal(result, expected)
+
+        result = df.dtypes == "category"
+        expected = Series([True, False, False], index=index)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_categoricalindex(self):
+        # GH 16111, categories that aren't lexsorted
+        categories = [9, 0, 1, 2, 3]
+
+        a = Series(1, index=pd.CategoricalIndex([9, 0], categories=categories))
+        b = Series(2, index=pd.CategoricalIndex([0, 1], categories=categories))
+        c = Series(3, index=pd.CategoricalIndex([1, 2], categories=categories))
+
+        result = pd.concat([a, b, c], axis=1)
+
+        exp_idx = pd.CategoricalIndex([9, 0, 1, 2], categories=categories)
+        exp = DataFrame(
+            {
+                0: [1, 1, np.nan, np.nan],
+                1: [np.nan, 2, 2, np.nan],
+                2: [np.nan, np.nan, 3, 3],
+            },
+            columns=[0, 1, 2],
+            index=exp_idx,
+        )
+        tm.assert_frame_equal(result, exp)
+
+    def test_categorical_concat_preserve(self):
+        # GH 8641  series concat not preserving category dtype
+        # GH 13524 can concat different categories
+        s = Series(list("abc"), dtype="category")
+        s2 = Series(list("abd"), dtype="category")
+
+        exp = Series(list("abcabd"))
+        res = pd.concat([s, s2], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+
+        exp = Series(list("abcabc"), dtype="category")
+        res = pd.concat([s, s], ignore_index=True)
+        tm.assert_series_equal(res, exp)
+
+        exp = Series(list("abcabc"), index=[0, 1, 2, 0, 1, 2], dtype="category")
+        res = pd.concat([s, s])
+        tm.assert_series_equal(res, exp)
+
+        a = Series(np.arange(6, dtype="int64"))
+        b = Series(list("aabbca"))
+
+        df2 = DataFrame({"A": a, "B": b.astype(CategoricalDtype(list("cab")))})
+        res = pd.concat([df2, df2])
+        exp = DataFrame(
+            {
+                "A": pd.concat([a, a]),
+                "B": pd.concat([b, b]).astype(CategoricalDtype(list("cab"))),
+            }
+        )
+        tm.assert_frame_equal(res, exp)
+
+    def test_categorical_index_preserver(self):
+        a = Series(np.arange(6, dtype="int64"))
+        b = Series(list("aabbca"))
+
+        df2 = DataFrame(
+            {"A": a, "B": b.astype(CategoricalDtype(list("cab")))}
+        ).set_index("B")
+        result = pd.concat([df2, df2])
+        expected = DataFrame(
+            {
+                "A": pd.concat([a, a]),
+                "B": pd.concat([b, b]).astype(CategoricalDtype(list("cab"))),
+            }
+        ).set_index("B")
+        tm.assert_frame_equal(result, expected)
+
+        # wrong categories -> uses concat_compat, which casts to object
+        df3 = DataFrame(
+            {"A": a, "B": Categorical(b, categories=list("abe"))}
+        ).set_index("B")
+        result = pd.concat([df2, df3])
+        expected = pd.concat(
+            [
+                df2.set_axis(df2.index.astype(object), axis=0),
+                df3.set_axis(df3.index.astype(object), axis=0),
+            ]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_categorical_tz(self):
+        # GH-23816
+        a = Series(pd.date_range("2017-01-01", periods=2, tz="US/Pacific"))
+        b = Series(["a", "b"], dtype="category")
+        result = pd.concat([a, b], ignore_index=True)
+        expected = Series(
+            [
+                pd.Timestamp("2017-01-01", tz="US/Pacific"),
+                pd.Timestamp("2017-01-02", tz="US/Pacific"),
+                "a",
+                "b",
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_categorical_datetime(self):
+        # GH-39443
+        df1 = DataFrame(
+            {"x": Series(datetime(2021, 1, 1), index=[0], dtype="category")}
+        )
+        df2 = DataFrame(
+            {"x": Series(datetime(2021, 1, 2), index=[1], dtype="category")}
+        )
+
+        result = pd.concat([df1, df2])
+        expected = DataFrame(
+            {"x": Series([datetime(2021, 1, 1), datetime(2021, 1, 2)])}
+        )
+
+        tm.assert_equal(result, expected)
+
+    def test_concat_categorical_unchanged(self):
+        # GH-12007
+        # test fix for when concat on categorical and float
+        # coerces dtype categorical -> float
+        df = DataFrame(Series(["a", "b", "c"], dtype="category", name="A"))
+        ser = Series([0, 1, 2], index=[0, 1, 3], name="B")
+        result = pd.concat([df, ser], axis=1)
+        expected = DataFrame(
+            {
+                "A": Series(["a", "b", "c", np.nan], dtype="category"),
+                "B": Series([0, 1, np.nan, 2], dtype="float"),
+            }
+        )
+        tm.assert_equal(result, expected)
+
+    def test_categorical_concat_gh7864(self):
+        # GH 7864
+        # make sure ordering is preserved
+        df = DataFrame({"id": [1, 2, 3, 4, 5, 6], "raw_grade": list("abbaae")})
+        df["grade"] = Categorical(df["raw_grade"])
+        df["grade"].cat.set_categories(["e", "a", "b"])
+
+        df1 = df[0:3]
+        df2 = df[3:]
+
+        tm.assert_index_equal(df["grade"].cat.categories, df1["grade"].cat.categories)
+        tm.assert_index_equal(df["grade"].cat.categories, df2["grade"].cat.categories)
+
+        dfx = pd.concat([df1, df2])
+        tm.assert_index_equal(df["grade"].cat.categories, dfx["grade"].cat.categories)
+
+        dfa = df1._append(df2)
+        tm.assert_index_equal(df["grade"].cat.categories, dfa["grade"].cat.categories)
+
+    def test_categorical_index_upcast(self):
+        # GH 17629
+        # test upcasting to object when concatenating on categorical indexes
+        # with non-identical categories
+
+        a = DataFrame({"foo": [1, 2]}, index=Categorical(["foo", "bar"]))
+        b = DataFrame({"foo": [4, 3]}, index=Categorical(["baz", "bar"]))
+
+        res = pd.concat([a, b])
+        exp = DataFrame({"foo": [1, 2, 4, 3]}, index=["foo", "bar", "baz", "bar"])
+
+        tm.assert_equal(res, exp)
+
+        a = Series([1, 2], index=Categorical(["foo", "bar"]))
+        b = Series([4, 3], index=Categorical(["baz", "bar"]))
+
+        res = pd.concat([a, b])
+        exp = Series([1, 2, 4, 3], index=["foo", "bar", "baz", "bar"])
+
+        tm.assert_equal(res, exp)
+
+    def test_categorical_missing_from_one_frame(self):
+        # GH 25412
+        df1 = DataFrame({"f1": [1, 2, 3]})
+        df2 = DataFrame({"f1": [2, 3, 1], "f2": Series([4, 4, 4]).astype("category")})
+        result = pd.concat([df1, df2], sort=True)
+        dtype = CategoricalDtype([4])
+        expected = DataFrame(
+            {
+                "f1": [1, 2, 3, 2, 3, 1],
+                "f2": Categorical.from_codes([-1, -1, -1, 0, 0, 0], dtype=dtype),
+            },
+            index=[0, 1, 2, 0, 1, 2],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_categorical_same_categories_different_order(self):
+        # https://github.com/pandas-dev/pandas/issues/24845
+
+        c1 = pd.CategoricalIndex(["a", "a"], categories=["a", "b"], ordered=False)
+        c2 = pd.CategoricalIndex(["b", "b"], categories=["b", "a"], ordered=False)
+        c3 = pd.CategoricalIndex(
+            ["a", "a", "b", "b"], categories=["a", "b"], ordered=False
+        )
+
+        df1 = DataFrame({"A": [1, 2]}, index=c1)
+        df2 = DataFrame({"A": [3, 4]}, index=c2)
+
+        result = pd.concat((df1, df2))
+        expected = DataFrame({"A": [1, 2, 3, 4]}, index=c3)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_concat.py b/pandas/pandas/tests/reshape/concat/test_concat.py
new file mode 100644
index 0000000000000..d3edee17366f7
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_concat.py
@@ -0,0 +1,945 @@
+from collections import (
+    abc,
+    deque,
+)
+from collections.abc import Iterator
+from datetime import datetime
+from decimal import Decimal
+import itertools
+
+import numpy as np
+import pytest
+
+from pandas.errors import InvalidIndexError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import SparseArray
+from pandas.tests.extension.decimal import to_decimal
+
+
+class TestConcatenate:
+    def test_append_concat(self):
+        # GH#1815
+        d1 = date_range("12/31/1990", "12/31/1999", freq="YE-DEC")
+        d2 = date_range("12/31/2000", "12/31/2009", freq="YE-DEC")
+
+        s1 = Series(np.random.default_rng(2).standard_normal(10), d1)
+        s2 = Series(np.random.default_rng(2).standard_normal(10), d2)
+
+        s1 = s1.to_period()
+        s2 = s2.to_period()
+
+        # drops index
+        result = concat([s1, s2])
+        assert isinstance(result.index, PeriodIndex)
+        assert result.index[0] == s1.index[0]
+
+    def test_concat_copy(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 3)))
+        df2 = DataFrame(np.random.default_rng(2).integers(0, 10, size=4).reshape(4, 1))
+        df3 = DataFrame({5: "foo"}, index=range(4))
+
+        result = concat([df, df2, df3], axis=1)
+
+        for block in result._mgr.blocks:
+            arr = block.values
+            if arr.dtype.kind == "f":
+                assert arr.base is df._mgr.blocks[0].values.base
+            elif arr.dtype.kind in ["i", "u"]:
+                assert arr.base is df2._mgr.blocks[0].values.base
+            elif arr.dtype == object:
+                assert arr.base is not None
+            elif arr.dtype == "string":
+                tm.shares_memory(arr, df3._mgr.blocks[0].values)
+
+        # Float block was consolidated.
+        df4 = DataFrame(np.random.default_rng(2).standard_normal((4, 1)))
+        result = concat([df, df2, df3, df4], axis=1)
+        for blocks in result._mgr.blocks:
+            arr = blocks.values
+            if arr.dtype.kind == "f":
+                # this is a view on some array in either df or df4
+                assert any(
+                    np.shares_memory(arr, block.values)
+                    for block in itertools.chain(df._mgr.blocks, df4._mgr.blocks)
+                )
+            elif arr.dtype.kind in ["i", "u"]:
+                assert arr.base is df2._mgr.blocks[0].values.base
+            elif arr.dtype == object:
+                # this is a view on df3
+                assert any(
+                    np.shares_memory(arr, block.values) for block in df3._mgr.blocks
+                )
+
+    def test_concat_with_group_keys(self):
+        # axis=0
+        df = DataFrame(np.random.default_rng(2).standard_normal((3, 4)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((4, 4)))
+
+        result = concat([df, df2], keys=[0, 1])
+        exp_index = MultiIndex.from_arrays(
+            [[0, 0, 0, 1, 1, 1, 1], [0, 1, 2, 0, 1, 2, 3]]
+        )
+        expected = DataFrame(np.r_[df.values, df2.values], index=exp_index)
+        tm.assert_frame_equal(result, expected)
+
+        result = concat([df, df], keys=[0, 1])
+        exp_index2 = MultiIndex.from_arrays([[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]])
+        expected = DataFrame(np.r_[df.values, df.values], index=exp_index2)
+        tm.assert_frame_equal(result, expected)
+
+        # axis=1
+        df = DataFrame(np.random.default_rng(2).standard_normal((4, 3)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((4, 4)))
+
+        result = concat([df, df2], keys=[0, 1], axis=1)
+        expected = DataFrame(np.c_[df.values, df2.values], columns=exp_index)
+        tm.assert_frame_equal(result, expected)
+
+        result = concat([df, df], keys=[0, 1], axis=1)
+        expected = DataFrame(np.c_[df.values, df.values], columns=exp_index2)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_keys_specific_levels(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        pieces = [df.iloc[:, [0, 1]], df.iloc[:, [2]], df.iloc[:, [3]]]
+        level = ["three", "two", "one", "zero"]
+        result = concat(
+            pieces,
+            axis=1,
+            keys=["one", "two", "three"],
+            levels=[level],
+            names=["group_key"],
+        )
+
+        tm.assert_index_equal(result.columns.levels[0], Index(level, name="group_key"))
+        tm.assert_index_equal(result.columns.levels[1], Index([0, 1, 2, 3]))
+
+        assert result.columns.names == ["group_key", None]
+
+    @pytest.mark.parametrize("mapping", ["mapping", "dict"])
+    def test_concat_mapping(self, mapping, non_dict_mapping_subclass):
+        constructor = dict if mapping == "dict" else non_dict_mapping_subclass
+        frames = constructor(
+            {
+                "foo": DataFrame(np.random.default_rng(2).standard_normal((4, 3))),
+                "bar": DataFrame(np.random.default_rng(2).standard_normal((4, 3))),
+                "baz": DataFrame(np.random.default_rng(2).standard_normal((4, 3))),
+                "qux": DataFrame(np.random.default_rng(2).standard_normal((4, 3))),
+            }
+        )
+
+        sorted_keys = list(frames.keys())
+
+        result = concat(frames)
+        expected = concat([frames[k] for k in sorted_keys], keys=sorted_keys)
+        tm.assert_frame_equal(result, expected)
+
+        result = concat(frames, axis=1)
+        expected = concat([frames[k] for k in sorted_keys], keys=sorted_keys, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        keys = ["baz", "foo", "bar"]
+        result = concat(frames, keys=keys)
+        expected = concat([frames[k] for k in keys], keys=keys)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_keys_and_levels(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((1, 3)))
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((1, 4)))
+
+        levels = [["foo", "baz"], ["one", "two"]]
+        names = ["first", "second"]
+        result = concat(
+            [df, df2, df, df2],
+            keys=[("foo", "one"), ("foo", "two"), ("baz", "one"), ("baz", "two")],
+            levels=levels,
+            names=names,
+        )
+        expected = concat([df, df2, df, df2])
+        exp_index = MultiIndex(
+            levels=levels + [[0]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1], [0, 0, 0, 0]],
+            names=names + [None],
+        )
+        expected.index = exp_index
+
+        tm.assert_frame_equal(result, expected)
+
+        # no names
+        result = concat(
+            [df, df2, df, df2],
+            keys=[("foo", "one"), ("foo", "two"), ("baz", "one"), ("baz", "two")],
+            levels=levels,
+        )
+        assert result.index.names == (None,) * 3
+
+        # no levels
+        result = concat(
+            [df, df2, df, df2],
+            keys=[("foo", "one"), ("foo", "two"), ("baz", "one"), ("baz", "two")],
+            names=["first", "second"],
+        )
+        assert result.index.names == ("first", "second", None)
+        tm.assert_index_equal(
+            result.index.levels[0], Index(["baz", "foo"], name="first")
+        )
+
+    def test_concat_keys_levels_no_overlap(self):
+        # GH #1406
+        df = DataFrame(np.random.default_rng(2).standard_normal((1, 3)), index=["a"])
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((1, 4)), index=["b"])
+
+        msg = "Values not found in passed level"
+        with pytest.raises(ValueError, match=msg):
+            concat([df, df], keys=["one", "two"], levels=[["foo", "bar", "baz"]])
+
+        msg = "Key one not in level"
+        with pytest.raises(ValueError, match=msg):
+            concat([df, df2], keys=["one", "two"], levels=[["foo", "bar", "baz"]])
+
+    def test_crossed_dtypes_weird_corner(self):
+        columns = ["A", "B", "C", "D"]
+        df1 = DataFrame(
+            {
+                "A": np.array([1, 2, 3, 4], dtype="f8"),
+                "B": np.array([1, 2, 3, 4], dtype="i8"),
+                "C": np.array([1, 2, 3, 4], dtype="f8"),
+                "D": np.array([1, 2, 3, 4], dtype="i8"),
+            },
+            columns=columns,
+        )
+
+        df2 = DataFrame(
+            {
+                "A": np.array([1, 2, 3, 4], dtype="i8"),
+                "B": np.array([1, 2, 3, 4], dtype="f8"),
+                "C": np.array([1, 2, 3, 4], dtype="i8"),
+                "D": np.array([1, 2, 3, 4], dtype="f8"),
+            },
+            columns=columns,
+        )
+
+        appended = concat([df1, df2], ignore_index=True)
+        expected = DataFrame(
+            np.concatenate([df1.values, df2.values], axis=0), columns=columns
+        )
+        tm.assert_frame_equal(appended, expected)
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((1, 3)), index=["a"])
+        df2 = DataFrame(np.random.default_rng(2).standard_normal((1, 4)), index=["b"])
+        result = concat([df, df2], keys=["one", "two"], names=["first", "second"])
+        assert result.index.names == ("first", "second")
+
+    def test_with_mixed_tuples(self, sort):
+        # 10697
+        # columns have mixed tuples, so handle properly
+        df1 = DataFrame({"A": "foo", ("B", 1): "bar"}, index=range(2))
+        df2 = DataFrame({"B": "foo", ("B", 1): "bar"}, index=range(2))
+
+        # it works
+        concat([df1, df2], sort=sort)
+
+    def test_concat_mixed_objs_columns(self):
+        # Test column-wise concat for mixed series/frames (axis=1)
+        # G2385
+
+        index = date_range("01-Jan-2013", periods=10, freq="h")
+        arr = np.arange(10, dtype="int64")
+        s1 = Series(arr, index=index)
+        s2 = Series(arr, index=index)
+        df = DataFrame(arr.reshape(-1, 1), index=index)
+
+        expected = DataFrame(
+            np.repeat(arr, 2).reshape(-1, 2), index=index, columns=[0, 0]
+        )
+        result = concat([df, df], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            np.repeat(arr, 2).reshape(-1, 2), index=index, columns=[0, 1]
+        )
+        result = concat([s1, s2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            np.repeat(arr, 3).reshape(-1, 3), index=index, columns=[0, 1, 2]
+        )
+        result = concat([s1, s2, s1], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(
+            np.repeat(arr, 5).reshape(-1, 5), index=index, columns=[0, 0, 1, 2, 3]
+        )
+        result = concat([s1, df, s2, s2, s1], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # with names
+        s1.name = "foo"
+        expected = DataFrame(
+            np.repeat(arr, 3).reshape(-1, 3), index=index, columns=["foo", 0, 0]
+        )
+        result = concat([s1, df, s2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        s2.name = "bar"
+        expected = DataFrame(
+            np.repeat(arr, 3).reshape(-1, 3), index=index, columns=["foo", 0, "bar"]
+        )
+        result = concat([s1, df, s2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # ignore index
+        expected = DataFrame(
+            np.repeat(arr, 3).reshape(-1, 3), index=index, columns=[0, 1, 2]
+        )
+        result = concat([s1, df, s2], axis=1, ignore_index=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_mixed_objs_index(self):
+        # Test row-wise concat for mixed series/frames with a common name
+        # GH2385, GH15047
+
+        index = date_range("01-Jan-2013", periods=10, freq="h")
+        arr = np.arange(10, dtype="int64")
+        s1 = Series(arr, index=index)
+        s2 = Series(arr, index=index)
+        df = DataFrame(arr.reshape(-1, 1), index=index)
+
+        expected = DataFrame(
+            np.tile(arr, 3).reshape(-1, 1), index=index.tolist() * 3, columns=[0]
+        )
+        result = concat([s1, df, s2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_mixed_objs_index_names(self):
+        # Test row-wise concat for mixed series/frames with distinct names
+        # GH2385, GH15047
+
+        index = date_range("01-Jan-2013", periods=10, freq="h")
+        arr = np.arange(10, dtype="int64")
+        s1 = Series(arr, index=index, name="foo")
+        s2 = Series(arr, index=index, name="bar")
+        df = DataFrame(arr.reshape(-1, 1), index=index)
+
+        expected = DataFrame(
+            np.kron(np.where(np.identity(3) == 1, 1, np.nan), arr).T,
+            index=index.tolist() * 3,
+            columns=["foo", 0, "bar"],
+        )
+        result = concat([s1, df, s2])
+        tm.assert_frame_equal(result, expected)
+
+        # Rename all series to 0 when ignore_index=True
+        expected = DataFrame(np.tile(arr, 3).reshape(-1, 1), columns=[0])
+        result = concat([s1, df, s2], ignore_index=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_dtype_coercion(self):
+        # 12411
+        df = DataFrame({"date": [pd.Timestamp("20130101").tz_localize("UTC"), pd.NaT]})
+
+        result = concat([df.iloc[[0]], df.iloc[[1]]])
+        tm.assert_series_equal(result.dtypes, df.dtypes)
+
+        # 12045
+        df = DataFrame({"date": [datetime(2012, 1, 1), datetime(1012, 1, 2)]})
+        result = concat([df.iloc[[0]], df.iloc[[1]]])
+        tm.assert_series_equal(result.dtypes, df.dtypes)
+
+        # 11594
+        df = DataFrame({"text": ["some words"] + [None] * 9})
+        result = concat([df.iloc[[0]], df.iloc[[1]]])
+        tm.assert_series_equal(result.dtypes, df.dtypes)
+
+    def test_concat_single_with_key(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+
+        result = concat([df], keys=["foo"])
+        expected = concat([df, df], keys=["foo", "bar"])
+        tm.assert_frame_equal(result, expected[:10])
+
+    def test_concat_no_items_raises(self):
+        with pytest.raises(ValueError, match="No objects to concatenate"):
+            concat([])
+
+    def test_concat_exclude_none(self):
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+
+        pieces = [df[:5], None, None, df[5:]]
+        result = concat(pieces)
+        tm.assert_frame_equal(result, df)
+        with pytest.raises(ValueError, match="All objects passed were None"):
+            concat([None, None])
+
+    def test_concat_keys_with_none(self):
+        # #1649
+        df0 = DataFrame([[10, 20, 30], [10, 20, 30], [10, 20, 30]])
+
+        result = concat({"a": None, "b": df0, "c": df0[:2], "d": df0[:1], "e": df0})
+        expected = concat({"b": df0, "c": df0[:2], "d": df0[:1], "e": df0})
+        tm.assert_frame_equal(result, expected)
+
+        result = concat(
+            [None, df0, df0[:2], df0[:1], df0], keys=["a", "b", "c", "d", "e"]
+        )
+        expected = concat([df0, df0[:2], df0[:1], df0], keys=["b", "c", "d", "e"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [range, RangeIndex])
+    @pytest.mark.parametrize("include_none", [True, False])
+    def test_concat_preserves_rangeindex(self, klass, include_none):
+        df = DataFrame([1, 2])
+        df2 = DataFrame([3, 4])
+        data = [df, None, df2, None] if include_none else [df, df2]
+        keys_length = 4 if include_none else 2
+        result = concat(data, keys=klass(keys_length))
+        expected = DataFrame(
+            [1, 2, 3, 4],
+            index=MultiIndex(
+                levels=(
+                    RangeIndex(start=0, stop=keys_length, step=keys_length / 2),
+                    RangeIndex(start=0, stop=2, step=1),
+                ),
+                codes=(
+                    np.array([0, 0, 1, 1], dtype=np.int8),
+                    np.array([0, 1, 0, 1], dtype=np.int8),
+                ),
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_bug_1719(self):
+        ts1 = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+        ts2 = ts1[::2]
+
+        # to join with union
+        # these two are of different length!
+        left = concat([ts1, ts2], join="outer", axis=1)
+        right = concat([ts2, ts1], join="outer", axis=1)
+
+        assert len(left) == len(right)
+
+    def test_concat_bug_2972(self):
+        ts0 = Series(np.zeros(5))
+        ts1 = Series(np.ones(5))
+        ts0.name = ts1.name = "same name"
+        result = concat([ts0, ts1], axis=1)
+
+        expected = DataFrame({0: ts0, 1: ts1})
+        expected.columns = ["same name", "same name"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_bug_3602(self):
+        # GH 3602, duplicate columns
+        df1 = DataFrame(
+            {
+                "firmNo": [0, 0, 0, 0],
+                "prc": [6, 6, 6, 6],
+                "stringvar": ["rrr", "rrr", "rrr", "rrr"],
+            }
+        )
+        df2 = DataFrame(
+            {"C": [9, 10, 11, 12], "misc": [1, 2, 3, 4], "prc": [6, 6, 6, 6]}
+        )
+        expected = DataFrame(
+            [
+                [0, 6, "rrr", 9, 1, 6],
+                [0, 6, "rrr", 10, 2, 6],
+                [0, 6, "rrr", 11, 3, 6],
+                [0, 6, "rrr", 12, 4, 6],
+            ]
+        )
+        expected.columns = ["firmNo", "prc", "stringvar", "C", "misc", "prc"]
+
+        result = concat([df1, df2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_iterables(self):
+        # GH8645 check concat works with tuples, list, generators, and weird
+        # stuff like deque and custom iterables
+        df1 = DataFrame([1, 2, 3])
+        df2 = DataFrame([4, 5, 6])
+        expected = DataFrame([1, 2, 3, 4, 5, 6])
+        tm.assert_frame_equal(concat((df1, df2), ignore_index=True), expected)
+        tm.assert_frame_equal(concat([df1, df2], ignore_index=True), expected)
+        tm.assert_frame_equal(
+            concat((df for df in (df1, df2)), ignore_index=True), expected
+        )
+        tm.assert_frame_equal(concat(deque((df1, df2)), ignore_index=True), expected)
+
+        class CustomIterator1:
+            def __len__(self) -> int:
+                return 2
+
+            def __getitem__(self, index):
+                try:
+                    return {0: df1, 1: df2}[index]
+                except KeyError as err:
+                    raise IndexError from err
+
+        tm.assert_frame_equal(concat(CustomIterator1(), ignore_index=True), expected)
+
+        class CustomIterator2(abc.Iterable):
+            def __iter__(self) -> Iterator:
+                yield df1
+                yield df2
+
+        tm.assert_frame_equal(concat(CustomIterator2(), ignore_index=True), expected)
+
+    def test_concat_order(self):
+        # GH 17344, GH#47331
+        dfs = [DataFrame(index=range(3), columns=["a", 1, None])]
+        dfs += [DataFrame(index=range(3), columns=[None, 1, "a"]) for _ in range(100)]
+
+        result = concat(dfs, sort=True).columns
+        expected = Index([1, "a", None])
+        tm.assert_index_equal(result, expected)
+
+    def test_concat_different_extension_dtypes_upcasts(self):
+        a = Series(pd.array([1, 2], dtype="Int64"))
+        b = Series(to_decimal([1, 2]))
+
+        result = concat([a, b], ignore_index=True)
+        expected = Series([1, 2, Decimal(1), Decimal(2)], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_ordered_dict(self):
+        # GH 21510
+        expected = concat(
+            [Series(range(3)), Series(range(4))], keys=["First", "Another"]
+        )
+        result = concat({"First": Series(range(3)), "Another": Series(range(4))})
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_duplicate_indices_raise(self):
+        # GH 45888: test raise for concat DataFrames with duplicate indices
+        # https://github.com/pandas-dev/pandas/issues/36263
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal(5),
+            index=[0, 1, 2, 3, 3],
+            columns=["a"],
+        )
+        df2 = DataFrame(
+            np.random.default_rng(2).standard_normal(5),
+            index=[0, 1, 2, 2, 4],
+            columns=["b"],
+        )
+        msg = "Reindexing only valid with uniquely valued Index objects"
+        with pytest.raises(InvalidIndexError, match=msg):
+            concat([df1, df2], axis=1)
+
+
+def test_concat_no_unnecessary_upcast(float_numpy_dtype, frame_or_series):
+    # GH 13247
+    dims = frame_or_series(dtype=object).ndim
+    dt = float_numpy_dtype
+
+    dfs = [
+        frame_or_series(np.array([1], dtype=dt, ndmin=dims)),
+        frame_or_series(np.array([np.nan], dtype=dt, ndmin=dims)),
+        frame_or_series(np.array([5], dtype=dt, ndmin=dims)),
+    ]
+    x = concat(dfs)
+    assert x.values.dtype == dt
+
+
+def test_concat_will_upcast(frame_or_series, any_signed_int_numpy_dtype):
+    dt = any_signed_int_numpy_dtype
+    dims = frame_or_series().ndim
+    dfs = [
+        frame_or_series(np.array([1], dtype=dt, ndmin=dims)),
+        frame_or_series(np.array([np.nan], ndmin=dims)),
+        frame_or_series(np.array([5], dtype=dt, ndmin=dims)),
+    ]
+    x = concat(dfs)
+    assert x.values.dtype == "float64"
+
+
+def test_concat_empty_and_non_empty_frame_regression():
+    # GH 18178 regression test
+    df1 = DataFrame({"foo": [1]})
+    df2 = DataFrame({"foo": []})
+    expected = DataFrame({"foo": [1.0]})
+    result = concat([df1, df2])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_sparse():
+    # GH 23557
+    a = Series(SparseArray([0, 1, 2]))
+    expected = DataFrame(data=[[0, 0], [1, 1], [2, 2]]).astype(
+        pd.SparseDtype(np.int64, 0)
+    )
+    result = concat([a, a], axis=1)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_dense_sparse():
+    # GH 30668
+    dtype = pd.SparseDtype(np.float64, None)
+    a = Series(pd.arrays.SparseArray([1, None]), dtype=dtype)
+    b = Series([1], dtype=float)
+    expected = Series(data=[1, None, 1], index=[0, 1, 0]).astype(dtype)
+    result = concat([a, b], axis=0)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("keys", [["e", "f", "f"], ["f", "e", "f"]])
+def test_duplicate_keys(keys):
+    # GH 33654
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    s1 = Series([7, 8, 9], name="c")
+    s2 = Series([10, 11, 12], name="d")
+    result = concat([df, s1, s2], axis=1, keys=keys)
+    expected_values = [[1, 4, 7, 10], [2, 5, 8, 11], [3, 6, 9, 12]]
+    expected_columns = MultiIndex.from_tuples(
+        [(keys[0], "a"), (keys[0], "b"), (keys[1], "c"), (keys[2], "d")]
+    )
+    expected = DataFrame(expected_values, columns=expected_columns)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_duplicate_keys_same_frame():
+    # GH 43595
+    keys = ["e", "e"]
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    result = concat([df, df], axis=1, keys=keys)
+    expected_values = [[1, 4, 1, 4], [2, 5, 2, 5], [3, 6, 3, 6]]
+    expected_columns = MultiIndex.from_tuples(
+        [(keys[0], "a"), (keys[0], "b"), (keys[1], "a"), (keys[1], "b")]
+    )
+    expected = DataFrame(expected_values, columns=expected_columns)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+)
+@pytest.mark.parametrize(
+    "obj",
+    [
+        tm.SubclassedDataFrame({"A": np.arange(0, 10)}),
+        tm.SubclassedSeries(np.arange(0, 10), name="A"),
+    ],
+)
+def test_concat_preserves_subclass(obj):
+    # GH28330 -- preserve subclass
+
+    result = concat([obj, obj])
+    assert isinstance(result, type(obj))
+
+
+def test_concat_frame_axis0_extension_dtypes():
+    # preserve extension dtype (through common_dtype mechanism)
+    df1 = DataFrame({"a": pd.array([1, 2, 3], dtype="Int64")})
+    df2 = DataFrame({"a": np.array([4, 5, 6])})
+
+    result = concat([df1, df2], ignore_index=True)
+    expected = DataFrame({"a": [1, 2, 3, 4, 5, 6]}, dtype="Int64")
+    tm.assert_frame_equal(result, expected)
+
+    result = concat([df2, df1], ignore_index=True)
+    expected = DataFrame({"a": [4, 5, 6, 1, 2, 3]}, dtype="Int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_preserves_extension_int64_dtype():
+    # GH 24768
+    df_a = DataFrame({"a": [-1]}, dtype="Int64")
+    df_b = DataFrame({"b": [1]}, dtype="Int64")
+    result = concat([df_a, df_b], ignore_index=True)
+    expected = DataFrame({"a": [-1, None], "b": [None, 1]}, dtype="Int64")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype1,dtype2,expected_dtype",
+    [
+        ("bool", "bool", "bool"),
+        ("boolean", "bool", "boolean"),
+        ("bool", "boolean", "boolean"),
+        ("boolean", "boolean", "boolean"),
+    ],
+)
+def test_concat_bool_types(dtype1, dtype2, expected_dtype):
+    # GH 42800
+    ser1 = Series([True, False], dtype=dtype1)
+    ser2 = Series([False, True], dtype=dtype2)
+    result = concat([ser1, ser2], ignore_index=True)
+    expected = Series([True, False, False, True], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("keys", "integrity"),
+    [
+        (["red"] * 3, True),
+        (["red"] * 3, False),
+        (["red", "blue", "red"], False),
+        (["red", "blue", "red"], True),
+    ],
+)
+def test_concat_repeated_keys(keys, integrity):
+    # GH: 20816
+    series_list = [Series({"a": 1}), Series({"b": 2}), Series({"c": 3})]
+    result = concat(series_list, keys=keys, verify_integrity=integrity)
+    tuples = list(zip(keys, ["a", "b", "c"]))
+    expected = Series([1, 2, 3], index=MultiIndex.from_tuples(tuples))
+    tm.assert_series_equal(result, expected)
+
+
+def test_concat_null_object_with_dti():
+    # GH#40841
+    dti = pd.DatetimeIndex(
+        ["2021-04-08 21:21:14+00:00"], dtype="datetime64[ns, UTC]", name="Time (UTC)"
+    )
+    right = DataFrame(data={"C": [0.5274]}, index=dti)
+
+    idx = Index([None], dtype="object", name="Maybe Time (UTC)")
+    left = DataFrame(data={"A": [None], "B": [np.nan]}, index=idx)
+
+    result = concat([left, right], axis="columns")
+
+    exp_index = Index([None, dti[0]], dtype=object)
+    expected = DataFrame(
+        {
+            "A": np.array([None, np.nan], dtype=object),
+            "B": [np.nan, np.nan],
+            "C": [np.nan, 0.5274],
+        },
+        index=exp_index,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_multiindex_with_empty_rangeindex():
+    # GH#41234
+    mi = MultiIndex.from_tuples([("B", 1), ("C", 1)])
+    df1 = DataFrame([[1, 2]], columns=mi)
+    df2 = DataFrame(index=[1], columns=RangeIndex(0))
+
+    result = concat([df1, df2])
+    expected = DataFrame([[1, 2], [np.nan, np.nan]], columns=mi)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        Series(data=[1, 2]),
+        DataFrame(
+            data={
+                "col1": [1, 2],
+            }
+        ),
+        DataFrame(dtype=float),
+        Series(dtype=float),
+    ],
+)
+def test_concat_drop_attrs(data):
+    # GH#41828
+    df1 = data.copy()
+    df1.attrs = {1: 1}
+    df2 = data.copy()
+    df2.attrs = {1: 2}
+    df = concat([df1, df2])
+    assert len(df.attrs) == 0
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        Series(data=[1, 2]),
+        DataFrame(
+            data={
+                "col1": [1, 2],
+            }
+        ),
+        DataFrame(dtype=float),
+        Series(dtype=float),
+    ],
+)
+def test_concat_retain_attrs(data):
+    # GH#41828
+    df1 = data.copy()
+    df1.attrs = {1: 1}
+    df2 = data.copy()
+    df2.attrs = {1: 1}
+    df = concat([df1, df2])
+    assert df.attrs[1] == 1
+
+
+@pytest.mark.parametrize("df_dtype", ["float64", "int64", "datetime64[ns]"])
+@pytest.mark.parametrize("empty_dtype", [None, "float64", "object"])
+def test_concat_ignore_empty_object_float(empty_dtype, df_dtype):
+    # https://github.com/pandas-dev/pandas/issues/45637
+    df = DataFrame({"foo": [1, 2], "bar": [1, 2]}, dtype=df_dtype)
+    empty = DataFrame(columns=["foo", "bar"], dtype=empty_dtype)
+
+    needs_update = False
+    if df_dtype == "datetime64[ns]" or (
+        df_dtype == "float64" and empty_dtype != "float64"
+    ):
+        needs_update = True
+
+    result = concat([empty, df])
+    expected = df
+    if df_dtype == "int64":
+        # TODO what exact behaviour do we want for integer eventually?
+        if empty_dtype == "float64":
+            expected = df.astype("float64")
+        else:
+            expected = df.astype("object")
+
+    if needs_update:
+        # GH#40893 changed the expected here to retain dependence on empty
+        expected = expected.astype(object)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("df_dtype", ["float64", "int64", "datetime64[ns]"])
+@pytest.mark.parametrize("empty_dtype", [None, "float64", "object"])
+def test_concat_ignore_all_na_object_float(empty_dtype, df_dtype):
+    df = DataFrame({"foo": [1, 2], "bar": [1, 2]}, dtype=df_dtype)
+    empty = DataFrame({"foo": [np.nan], "bar": [np.nan]}, dtype=empty_dtype)
+
+    if df_dtype == "int64":
+        # TODO what exact behaviour do we want for integer eventually?
+        if empty_dtype == "object":
+            df_dtype = "object"
+        else:
+            df_dtype = "float64"
+
+    needs_update = False
+    if empty_dtype != df_dtype and empty_dtype is not None:
+        needs_update = True
+    elif df_dtype == "datetime64[ns]":
+        needs_update = True
+
+    result = concat([empty, df], ignore_index=True)
+
+    expected = DataFrame({"foo": [np.nan, 1, 2], "bar": [np.nan, 1, 2]}, dtype=df_dtype)
+    if needs_update:
+        # GH#40893 changed the expected here to retain dependence on empty
+        expected = expected.astype(object)
+        expected.iloc[0] = np.nan
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_ignore_empty_from_reindex():
+    # https://github.com/pandas-dev/pandas/pull/43507#issuecomment-920375856
+    df1 = DataFrame({"a": [1], "b": [pd.Timestamp("2012-01-01")]})
+    df2 = DataFrame({"a": [2]})
+
+    aligned = df2.reindex(columns=df1.columns)
+
+    result = concat([df1, aligned], ignore_index=True)
+
+    expected = DataFrame(
+        {
+            "a": [1, 2],
+            "b": pd.array([pd.Timestamp("2012-01-01"), np.nan], dtype=object),
+        },
+        dtype=object,
+    )
+    expected["a"] = expected["a"].astype("int64")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_mismatched_keys_length():
+    # GH#43485
+    ser = Series(range(5))
+    sers = [ser + n for n in range(4)]
+    keys = ["A", "B", "C"]
+
+    msg = r"The length of the keys"
+    with pytest.raises(ValueError, match=msg):
+        concat(sers, keys=keys, axis=1)
+    with pytest.raises(ValueError, match=msg):
+        concat(sers, keys=keys, axis=0)
+    with pytest.raises(ValueError, match=msg):
+        concat((x for x in sers), keys=(y for y in keys), axis=1)
+    with pytest.raises(ValueError, match=msg):
+        concat((x for x in sers), keys=(y for y in keys), axis=0)
+
+
+def test_concat_multiindex_with_category():
+    df1 = DataFrame(
+        {
+            "c1": Series(list("abc"), dtype="category"),
+            "c2": Series(list("eee"), dtype="category"),
+            "i2": Series([1, 2, 3]),
+        }
+    )
+    df1 = df1.set_index(["c1", "c2"])
+    df2 = DataFrame(
+        {
+            "c1": Series(list("abc"), dtype="category"),
+            "c2": Series(list("eee"), dtype="category"),
+            "i2": Series([4, 5, 6]),
+        }
+    )
+    df2 = df2.set_index(["c1", "c2"])
+    result = concat([df1, df2])
+    expected = DataFrame(
+        {
+            "c1": Series(list("abcabc"), dtype="category"),
+            "c2": Series(list("eeeeee"), dtype="category"),
+            "i2": Series([1, 2, 3, 4, 5, 6]),
+        }
+    )
+    expected = expected.set_index(["c1", "c2"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_ea_upcast():
+    # GH#54848
+    df1 = DataFrame(["a"], dtype="string")
+    df2 = DataFrame([1], dtype="Int64")
+    result = concat([df1, df2])
+    expected = DataFrame(["a", 1], index=[0, 0])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_none_with_timezone_timestamp():
+    # GH#52093
+    df1 = DataFrame([{"A": None}])
+    df2 = DataFrame([{"A": pd.Timestamp("1990-12-20 00:00:00+00:00")}])
+    result = concat([df1, df2], ignore_index=True)
+    expected = DataFrame(
+        {"A": [None, pd.Timestamp("1990-12-20 00:00:00+00:00")]}, dtype=object
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_with_series_and_frame_returns_rangeindex_columns():
+    ser = Series([0])
+    df = DataFrame([1, 2])
+    result = concat([ser, df])
+    expected = DataFrame([0, 1, 2], index=[0, 0, 1])
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+
+def test_concat_with_moot_ignore_index_and_keys():
+    df1 = DataFrame([[0]])
+    df2 = DataFrame([[42]])
+
+    ignore_index = True
+    keys = ["df1", "df2"]
+    msg = f"Cannot set {ignore_index=} and specify keys. Either should be used."
+    with pytest.raises(ValueError, match=msg):
+        concat([df1, df2], keys=keys, ignore_index=ignore_index)
diff --git a/pandas/pandas/tests/reshape/concat/test_dataframe.py b/pandas/pandas/tests/reshape/concat/test_dataframe.py
new file mode 100644
index 0000000000000..2ad46ac009928
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_dataframe.py
@@ -0,0 +1,216 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+
+class TestDataFrameConcat:
+    def test_concat_multiple_frames_dtypes(self):
+        # GH#2759
+        df1 = DataFrame(data=np.ones((10, 2)), columns=["foo", "bar"], dtype=np.float64)
+        df2 = DataFrame(data=np.ones((10, 2)), dtype=np.float32)
+        results = concat((df1, df2), axis=1).dtypes
+        expected = Series(
+            [np.dtype("float64")] * 2 + [np.dtype("float32")] * 2,
+            index=["foo", "bar", 0, 1],
+        )
+        tm.assert_series_equal(results, expected)
+
+    def test_concat_tuple_keys(self):
+        # GH#14438
+        df1 = DataFrame(np.ones((2, 2)), columns=list("AB"))
+        df2 = DataFrame(np.ones((3, 2)) * 2, columns=list("AB"))
+        results = concat((df1, df2), keys=[("bee", "bah"), ("bee", "boo")])
+        expected = DataFrame(
+            {
+                "A": {
+                    ("bee", "bah", 0): 1.0,
+                    ("bee", "bah", 1): 1.0,
+                    ("bee", "boo", 0): 2.0,
+                    ("bee", "boo", 1): 2.0,
+                    ("bee", "boo", 2): 2.0,
+                },
+                "B": {
+                    ("bee", "bah", 0): 1.0,
+                    ("bee", "bah", 1): 1.0,
+                    ("bee", "boo", 0): 2.0,
+                    ("bee", "boo", 1): 2.0,
+                    ("bee", "boo", 2): 2.0,
+                },
+            }
+        )
+        tm.assert_frame_equal(results, expected)
+
+    def test_concat_named_keys(self):
+        # GH#14252
+        df = DataFrame({"foo": [1, 2], "bar": [0.1, 0.2]})
+        index = Index(["a", "b"], name="baz")
+        concatted_named_from_keys = concat([df, df], keys=index)
+        expected_named = DataFrame(
+            {"foo": [1, 2, 1, 2], "bar": [0.1, 0.2, 0.1, 0.2]},
+            index=pd.MultiIndex.from_product((["a", "b"], [0, 1]), names=["baz", None]),
+        )
+        tm.assert_frame_equal(concatted_named_from_keys, expected_named)
+
+        index_no_name = Index(["a", "b"], name=None)
+        concatted_named_from_names = concat([df, df], keys=index_no_name, names=["baz"])
+        tm.assert_frame_equal(concatted_named_from_names, expected_named)
+
+        concatted_unnamed = concat([df, df], keys=index_no_name)
+        expected_unnamed = DataFrame(
+            {"foo": [1, 2, 1, 2], "bar": [0.1, 0.2, 0.1, 0.2]},
+            index=pd.MultiIndex.from_product((["a", "b"], [0, 1]), names=[None, None]),
+        )
+        tm.assert_frame_equal(concatted_unnamed, expected_unnamed)
+
+    def test_concat_axis_parameter(self):
+        # GH#14369
+        df1 = DataFrame({"A": [0.1, 0.2]}, index=range(2))
+        df2 = DataFrame({"A": [0.3, 0.4]}, index=range(2))
+
+        # Index/row/0 DataFrame
+        expected_index = DataFrame({"A": [0.1, 0.2, 0.3, 0.4]}, index=[0, 1, 0, 1])
+
+        concatted_index = concat([df1, df2], axis="index")
+        tm.assert_frame_equal(concatted_index, expected_index)
+
+        concatted_row = concat([df1, df2], axis="rows")
+        tm.assert_frame_equal(concatted_row, expected_index)
+
+        concatted_0 = concat([df1, df2], axis=0)
+        tm.assert_frame_equal(concatted_0, expected_index)
+
+        # Columns/1 DataFrame
+        expected_columns = DataFrame(
+            [[0.1, 0.3], [0.2, 0.4]], index=[0, 1], columns=["A", "A"]
+        )
+
+        concatted_columns = concat([df1, df2], axis="columns")
+        tm.assert_frame_equal(concatted_columns, expected_columns)
+
+        concatted_1 = concat([df1, df2], axis=1)
+        tm.assert_frame_equal(concatted_1, expected_columns)
+
+        series1 = Series([0.1, 0.2])
+        series2 = Series([0.3, 0.4])
+
+        # Index/row/0 Series
+        expected_index_series = Series([0.1, 0.2, 0.3, 0.4], index=[0, 1, 0, 1])
+
+        concatted_index_series = concat([series1, series2], axis="index")
+        tm.assert_series_equal(concatted_index_series, expected_index_series)
+
+        concatted_row_series = concat([series1, series2], axis="rows")
+        tm.assert_series_equal(concatted_row_series, expected_index_series)
+
+        concatted_0_series = concat([series1, series2], axis=0)
+        tm.assert_series_equal(concatted_0_series, expected_index_series)
+
+        # Columns/1 Series
+        expected_columns_series = DataFrame(
+            [[0.1, 0.3], [0.2, 0.4]], index=[0, 1], columns=[0, 1]
+        )
+
+        concatted_columns_series = concat([series1, series2], axis="columns")
+        tm.assert_frame_equal(concatted_columns_series, expected_columns_series)
+
+        concatted_1_series = concat([series1, series2], axis=1)
+        tm.assert_frame_equal(concatted_1_series, expected_columns_series)
+
+        # Testing ValueError
+        with pytest.raises(ValueError, match="No axis named"):
+            concat([series1, series2], axis="something")
+
+    def test_concat_numerical_names(self):
+        # GH#15262, GH#12223
+        df = DataFrame(
+            {"col": range(9)},
+            dtype="int32",
+            index=(
+                pd.MultiIndex.from_product(
+                    [["A0", "A1", "A2"], ["B0", "B1", "B2"]], names=[1, 2]
+                )
+            ),
+        )
+        result = concat((df.iloc[:2, :], df.iloc[-2:, :]))
+        expected = DataFrame(
+            {"col": [0, 1, 7, 8]},
+            dtype="int32",
+            index=pd.MultiIndex.from_tuples(
+                [("A0", "B0"), ("A0", "B1"), ("A2", "B1"), ("A2", "B2")], names=[1, 2]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_astype_dup_col(self):
+        # GH#23049
+        df = DataFrame([{"a": "b"}])
+        df = concat([df, df], axis=1)
+
+        result = df.astype("category")
+        expected = DataFrame(
+            np.array(["b", "b"]).reshape(1, 2), columns=["a", "a"]
+        ).astype("category")
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_dataframe_keys_bug(self, sort):
+        t1 = DataFrame(
+            {"value": Series([1, 2, 3], index=Index(["a", "b", "c"], name="id"))}
+        )
+        t2 = DataFrame({"value": Series([7, 8], index=Index(["a", "b"], name="id"))})
+
+        # it works
+        result = concat([t1, t2], axis=1, keys=["t1", "t2"], sort=sort)
+        assert list(result.columns) == [("t1", "value"), ("t2", "value")]
+
+    def test_concat_bool_with_int(self):
+        # GH#42092 we may want to change this to return object, but that
+        #  would need a deprecation
+        df1 = DataFrame(Series([True, False, True, True], dtype="bool"))
+        df2 = DataFrame(Series([1, 0, 1], dtype="int64"))
+
+        result = concat([df1, df2])
+        expected = concat([df1.astype("int64"), df2])
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_duplicates_in_index_with_keys(self):
+        # GH#42651
+        index = [1, 1, 3]
+        data = [1, 2, 3]
+
+        df = DataFrame(data=data, index=index)
+        result = concat([df], keys=["A"], names=["ID", "date"])
+        mi = pd.MultiIndex.from_product([["A"], index], names=["ID", "date"])
+        expected = DataFrame(data=data, index=mi)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_index_equal(result.index.levels[1], Index([1, 3], name="date"))
+
+    def test_outer_sort_columns(self):
+        # GH#47127
+        df1 = DataFrame({"A": [0], "B": [1], 0: 1})
+        df2 = DataFrame({"A": [100]})
+        result = concat([df1, df2], ignore_index=True, join="outer", sort=True)
+        expected = DataFrame({0: [1.0, np.nan], "A": [0, 100], "B": [1.0, np.nan]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_inner_sort_columns(self):
+        # GH#47127
+        df1 = DataFrame({"A": [0], "B": [1], 0: 1})
+        df2 = DataFrame({"A": [100], 0: 2})
+        result = concat([df1, df2], ignore_index=True, join="inner", sort=True)
+        expected = DataFrame({0: [1, 2], "A": [0, 100]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_sort_columns_one_df(self):
+        # GH#47127
+        df1 = DataFrame({"A": [100], 0: 2})
+        result = concat([df1], ignore_index=True, join="inner", sort=True)
+        expected = DataFrame({0: [2], "A": [100]})
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_datetimes.py b/pandas/pandas/tests/reshape/concat/test_datetimes.py
new file mode 100644
index 0000000000000..0cf3192ea3a74
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_datetimes.py
@@ -0,0 +1,589 @@
+import datetime as dt
+from datetime import datetime
+
+import dateutil
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+
+class TestDatetimeConcat:
+    def test_concat_datetime64_block(self):
+        rng = date_range("1/1/2000", periods=10)
+
+        df = DataFrame({"time": rng})
+
+        result = concat([df, df])
+        assert (result.iloc[:10]["time"] == rng).all()
+        assert (result.iloc[10:]["time"] == rng).all()
+
+    def test_concat_datetime_datetime64_frame(self):
+        # GH#2624
+        rows = []
+        rows.append([datetime(2010, 1, 1), 1])
+        rows.append([datetime(2010, 1, 2), "hi"])
+
+        df2_obj = DataFrame.from_records(rows, columns=["date", "test"])
+
+        ind = date_range(start="2000/1/1", freq="D", periods=10)
+        df1 = DataFrame({"date": ind, "test": range(10)})
+
+        # it works!
+        concat([df1, df2_obj])
+
+    def test_concat_datetime_timezone(self):
+        # GH 18523
+        idx1 = date_range("2011-01-01", periods=3, freq="h", tz="Europe/Paris")
+        idx2 = date_range(start=idx1[0], end=idx1[-1], freq="h")
+        df1 = DataFrame({"a": [1, 2, 3]}, index=idx1)
+        df2 = DataFrame({"b": [1, 2, 3]}, index=idx2)
+        result = concat([df1, df2], axis=1)
+
+        exp_idx = DatetimeIndex(
+            [
+                "2011-01-01 00:00:00+01:00",
+                "2011-01-01 01:00:00+01:00",
+                "2011-01-01 02:00:00+01:00",
+            ],
+            dtype="M8[ns, Europe/Paris]",
+            freq="h",
+        )
+        expected = DataFrame(
+            [[1, 1], [2, 2], [3, 3]], index=exp_idx, columns=["a", "b"]
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        idx3 = date_range("2011-01-01", periods=3, freq="h", tz="Asia/Tokyo")
+        df3 = DataFrame({"b": [1, 2, 3]}, index=idx3)
+        result = concat([df1, df3], axis=1)
+
+        exp_idx = DatetimeIndex(
+            [
+                "2010-12-31 15:00:00+00:00",
+                "2010-12-31 16:00:00+00:00",
+                "2010-12-31 17:00:00+00:00",
+                "2010-12-31 23:00:00+00:00",
+                "2011-01-01 00:00:00+00:00",
+                "2011-01-01 01:00:00+00:00",
+            ]
+        ).as_unit("ns")
+
+        expected = DataFrame(
+            [
+                [np.nan, 1],
+                [np.nan, 2],
+                [np.nan, 3],
+                [1, np.nan],
+                [2, np.nan],
+                [3, np.nan],
+            ],
+            index=exp_idx,
+            columns=["a", "b"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        # GH 13783: Concat after resample
+        result = concat([df1.resample("h").mean(), df2.resample("h").mean()], sort=True)
+        expected = DataFrame(
+            {"a": [1, 2, 3] + [np.nan] * 3, "b": [np.nan] * 3 + [1, 2, 3]},
+            index=idx1.append(idx1),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_datetimeindex_freq(self):
+        # GH 3232
+        # Monotonic index result
+        dr = date_range("01-Jan-2013", periods=100, freq="50ms", tz="UTC")
+        data = list(range(100))
+        expected = DataFrame(data, index=dr)
+        result = concat([expected[:50], expected[50:]])
+        tm.assert_frame_equal(result, expected)
+
+        # Non-monotonic index result
+        result = concat([expected[50:], expected[:50]])
+        expected = DataFrame(data[50:] + data[:50], index=dr[50:].append(dr[:50]))
+        expected.index._data.freq = None
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_multiindex_datetime_object_index(self):
+        # https://github.com/pandas-dev/pandas/issues/11058
+        idx = Index(
+            [dt.date(2013, 1, 1), dt.date(2014, 1, 1), dt.date(2015, 1, 1)],
+            dtype="object",
+        )
+
+        s = Series(
+            ["a", "b"],
+            index=MultiIndex.from_arrays(
+                [
+                    [1, 2],
+                    idx[:-1],
+                ],
+                names=["first", "second"],
+            ),
+        )
+        s2 = Series(
+            ["a", "b"],
+            index=MultiIndex.from_arrays(
+                [[1, 2], idx[::2]],
+                names=["first", "second"],
+            ),
+        )
+        mi = MultiIndex.from_arrays(
+            [[1, 2, 2], idx],
+            names=["first", "second"],
+        )
+        assert mi.levels[1].dtype == object
+
+        expected = DataFrame(
+            [["a", "a"], ["b", np.nan], [np.nan, "b"]],
+            index=mi,
+        )
+        result = concat([s, s2], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_NaT_series(self):
+        # GH 11693
+        # test for merging NaT series with datetime series.
+        x = Series(
+            date_range("20151124 08:00", "20151124 09:00", freq="1h", tz="US/Eastern")
+        )
+        y = Series(pd.NaT, index=[0, 1], dtype="datetime64[ns, US/Eastern]")
+        expected = Series([x[0], x[1], pd.NaT, pd.NaT])
+
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+        # all NaT with tz
+        expected = Series(pd.NaT, index=range(4), dtype="datetime64[ns, US/Eastern]")
+        result = concat([y, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_NaT_series2(self):
+        # without tz
+        x = Series(date_range("20151124 08:00", "20151124 09:00", freq="1h"))
+        y = Series(date_range("20151124 10:00", "20151124 11:00", freq="1h"))
+        y[:] = pd.NaT
+        expected = Series([x[0], x[1], pd.NaT, pd.NaT])
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+        # all NaT without tz
+        x[:] = pd.NaT
+        expected = Series(pd.NaT, index=range(4), dtype="datetime64[ns]")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_concat_NaT_dataframes(self, tz):
+        # GH 12396
+
+        dti = DatetimeIndex([pd.NaT, pd.NaT], tz=tz)
+        first = DataFrame({0: dti})
+        second = DataFrame(
+            [[Timestamp("2015/01/01", tz=tz)], [Timestamp("2016/01/01", tz=tz)]],
+            index=[2, 3],
+        )
+        expected = DataFrame(
+            [
+                pd.NaT,
+                pd.NaT,
+                Timestamp("2015/01/01", tz=tz),
+                Timestamp("2016/01/01", tz=tz),
+            ]
+        )
+
+        result = concat([first, second], axis=0)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz1", [None, "UTC"])
+    @pytest.mark.parametrize("tz2", [None, "UTC"])
+    @pytest.mark.parametrize("item", [pd.NaT, Timestamp("20150101").as_unit("ns")])
+    def test_concat_NaT_dataframes_all_NaT_axis_0(self, tz1, tz2, item):
+        # GH 12396
+
+        # tz-naive
+        first = DataFrame([[pd.NaT], [pd.NaT]]).apply(lambda x: x.dt.tz_localize(tz1))
+        second = DataFrame([item]).apply(lambda x: x.dt.tz_localize(tz2))
+
+        result = concat([first, second], axis=0)
+        expected = DataFrame(Series([pd.NaT, pd.NaT, item], index=[0, 1, 0]))
+        expected = expected.apply(lambda x: x.dt.tz_localize(tz2))
+        if tz1 != tz2:
+            expected = expected.astype(object)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz1", [None, "UTC"])
+    @pytest.mark.parametrize("tz2", [None, "UTC"])
+    def test_concat_NaT_dataframes_all_NaT_axis_1(self, tz1, tz2):
+        # GH 12396
+
+        first = DataFrame(Series([pd.NaT, pd.NaT]).dt.tz_localize(tz1))
+        second = DataFrame(Series([pd.NaT]).dt.tz_localize(tz2), columns=[1])
+        expected = DataFrame(
+            {
+                0: Series([pd.NaT, pd.NaT]).dt.tz_localize(tz1),
+                1: Series([pd.NaT, pd.NaT]).dt.tz_localize(tz2),
+            }
+        )
+        result = concat([first, second], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("tz1", [None, "UTC"])
+    @pytest.mark.parametrize("tz2", [None, "UTC"])
+    def test_concat_NaT_series_dataframe_all_NaT(self, tz1, tz2):
+        # GH 12396
+
+        # tz-naive
+        first = Series([pd.NaT, pd.NaT]).dt.tz_localize(tz1)
+        second = DataFrame(
+            [
+                [Timestamp("2015/01/01", tz=tz2)],
+                [Timestamp("2016/01/01", tz=tz2)],
+            ],
+            index=[2, 3],
+        )
+
+        expected = DataFrame(
+            [
+                pd.NaT,
+                pd.NaT,
+                Timestamp("2015/01/01", tz=tz2),
+                Timestamp("2016/01/01", tz=tz2),
+            ]
+        )
+        if tz1 != tz2:
+            expected = expected.astype(object)
+
+        result = concat([first, second])
+        tm.assert_frame_equal(result, expected)
+
+
+class TestTimezoneConcat:
+    def test_concat_tz_series(self):
+        # gh-11755: tz and no tz
+        x = Series(date_range("20151124 08:00", "20151124 09:00", freq="1h", tz="UTC"))
+        y = Series(date_range("2012-01-01", "2012-01-02"))
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_tz_series2(self):
+        # gh-11887: concat tz and object
+        x = Series(date_range("20151124 08:00", "20151124 09:00", freq="1h", tz="UTC"))
+        y = Series(["a", "b"])
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_tz_series3(self, unit, unit2):
+        # see gh-12217 and gh-12306
+        # Concatenating two UTC times
+        first = DataFrame([[datetime(2016, 1, 1)]], dtype=f"M8[{unit}]")
+        first[0] = first[0].dt.tz_localize("UTC")
+
+        second = DataFrame([[datetime(2016, 1, 2)]], dtype=f"M8[{unit2}]")
+        second[0] = second[0].dt.tz_localize("UTC")
+
+        result = concat([first, second])
+        exp_unit = tm.get_finest_unit(unit, unit2)
+        assert result[0].dtype == f"datetime64[{exp_unit}, UTC]"
+
+    def test_concat_tz_series4(self, unit, unit2):
+        # Concatenating two London times
+        first = DataFrame([[datetime(2016, 1, 1)]], dtype=f"M8[{unit}]")
+        first[0] = first[0].dt.tz_localize("Europe/London")
+
+        second = DataFrame([[datetime(2016, 1, 2)]], dtype=f"M8[{unit2}]")
+        second[0] = second[0].dt.tz_localize("Europe/London")
+
+        result = concat([first, second])
+        exp_unit = tm.get_finest_unit(unit, unit2)
+        assert result[0].dtype == f"datetime64[{exp_unit}, Europe/London]"
+
+    def test_concat_tz_series5(self, unit, unit2):
+        # Concatenating 2+1 London times
+        first = DataFrame(
+            [[datetime(2016, 1, 1)], [datetime(2016, 1, 2)]], dtype=f"M8[{unit}]"
+        )
+        first[0] = first[0].dt.tz_localize("Europe/London")
+
+        second = DataFrame([[datetime(2016, 1, 3)]], dtype=f"M8[{unit2}]")
+        second[0] = second[0].dt.tz_localize("Europe/London")
+
+        result = concat([first, second])
+        exp_unit = tm.get_finest_unit(unit, unit2)
+        assert result[0].dtype == f"datetime64[{exp_unit}, Europe/London]"
+
+    def test_concat_tz_series6(self, unit, unit2):
+        # Concatenating 1+2 London times
+        first = DataFrame([[datetime(2016, 1, 1)]], dtype=f"M8[{unit}]")
+        first[0] = first[0].dt.tz_localize("Europe/London")
+
+        second = DataFrame(
+            [[datetime(2016, 1, 2)], [datetime(2016, 1, 3)]], dtype=f"M8[{unit2}]"
+        )
+        second[0] = second[0].dt.tz_localize("Europe/London")
+
+        result = concat([first, second])
+        exp_unit = tm.get_finest_unit(unit, unit2)
+        assert result[0].dtype == f"datetime64[{exp_unit}, Europe/London]"
+
+    def test_concat_tz_series_tzlocal(self):
+        # see gh-13583
+        x = [
+            Timestamp("2011-01-01", tz=dateutil.tz.tzlocal()),
+            Timestamp("2011-02-01", tz=dateutil.tz.tzlocal()),
+        ]
+        y = [
+            Timestamp("2012-01-01", tz=dateutil.tz.tzlocal()),
+            Timestamp("2012-02-01", tz=dateutil.tz.tzlocal()),
+        ]
+
+        result = concat([Series(x), Series(y)], ignore_index=True)
+        tm.assert_series_equal(result, Series(x + y))
+        assert result.dtype == "datetime64[s, tzlocal()]"
+
+    def test_concat_tz_series_with_datetimelike(self):
+        # see gh-12620: tz and timedelta
+        x = [
+            Timestamp("2011-01-01", tz="US/Eastern"),
+            Timestamp("2011-02-01", tz="US/Eastern"),
+        ]
+        y = [pd.Timedelta("1 day"), pd.Timedelta("2 day")]
+        result = concat([Series(x), Series(y)], ignore_index=True)
+        tm.assert_series_equal(result, Series(x + y, dtype="object"))
+
+        # tz and period
+        y = [pd.Period("2011-03", freq="M"), pd.Period("2011-04", freq="M")]
+        result = concat([Series(x), Series(y)], ignore_index=True)
+        tm.assert_series_equal(result, Series(x + y, dtype="object"))
+
+    def test_concat_tz_frame(self):
+        df2 = DataFrame(
+            {
+                "A": Timestamp("20130102", tz="US/Eastern"),
+                "B": Timestamp("20130603", tz="CET"),
+            },
+            index=range(5),
+        )
+
+        # concat
+        df3 = concat([df2.A.to_frame(), df2.B.to_frame()], axis=1)
+        tm.assert_frame_equal(df2, df3)
+
+    def test_concat_multiple_tzs(self):
+        # GH#12467
+        # combining datetime tz-aware and naive DataFrames
+        ts1 = Timestamp("2015-01-01", tz=None)
+        ts2 = Timestamp("2015-01-01", tz="UTC")
+        ts3 = Timestamp("2015-01-01", tz="EST")
+
+        df1 = DataFrame({"time": [ts1]})
+        df2 = DataFrame({"time": [ts2]})
+        df3 = DataFrame({"time": [ts3]})
+
+        results = concat([df1, df2]).reset_index(drop=True)
+        expected = DataFrame({"time": [ts1, ts2]}, dtype=object)
+        tm.assert_frame_equal(results, expected)
+
+        results = concat([df1, df3]).reset_index(drop=True)
+        expected = DataFrame({"time": [ts1, ts3]}, dtype=object)
+        tm.assert_frame_equal(results, expected)
+
+        results = concat([df2, df3]).reset_index(drop=True)
+        expected = DataFrame({"time": [ts2, ts3]})
+        tm.assert_frame_equal(results, expected)
+
+    def test_concat_multiindex_with_tz(self):
+        # GH 6606
+        df = DataFrame(
+            {
+                "dt": DatetimeIndex(
+                    [
+                        datetime(2014, 1, 1),
+                        datetime(2014, 1, 2),
+                        datetime(2014, 1, 3),
+                    ],
+                    dtype="M8[ns, US/Pacific]",
+                ),
+                "b": ["A", "B", "C"],
+                "c": [1, 2, 3],
+                "d": [4, 5, 6],
+            }
+        )
+        df = df.set_index(["dt", "b"])
+
+        exp_idx1 = DatetimeIndex(
+            ["2014-01-01", "2014-01-02", "2014-01-03"] * 2,
+            dtype="M8[ns, US/Pacific]",
+            name="dt",
+        )
+        exp_idx2 = Index(["A", "B", "C"] * 2, name="b")
+        exp_idx = MultiIndex.from_arrays([exp_idx1, exp_idx2])
+        expected = DataFrame(
+            {"c": [1, 2, 3] * 2, "d": [4, 5, 6] * 2}, index=exp_idx, columns=["c", "d"]
+        )
+
+        result = concat([df, df])
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_tz_not_aligned(self):
+        # GH#22796
+        ts = pd.to_datetime([1, 2]).tz_localize("UTC")
+        a = DataFrame({"A": ts})
+        b = DataFrame({"A": ts, "B": ts})
+        result = concat([a, b], sort=True, ignore_index=True)
+        expected = DataFrame(
+            {"A": list(ts) + list(ts), "B": [pd.NaT, pd.NaT] + list(ts)}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "t1",
+        [
+            "2015-01-01",
+            pytest.param(
+                pd.NaT,
+                marks=pytest.mark.xfail(
+                    reason="GH23037 incorrect dtype when concatenating"
+                ),
+            ),
+        ],
+    )
+    def test_concat_tz_NaT(self, t1):
+        # GH#22796
+        # Concatenating tz-aware multicolumn DataFrames
+        ts1 = Timestamp(t1, tz="UTC")
+        ts2 = Timestamp("2015-01-01", tz="UTC")
+        ts3 = Timestamp("2015-01-01", tz="UTC")
+
+        df1 = DataFrame([[ts1, ts2]])
+        df2 = DataFrame([[ts3]])
+
+        result = concat([df1, df2])
+        expected = DataFrame([[ts1, ts2], [ts3, pd.NaT]], index=[0, 0])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_tz_with_empty(self):
+        # GH 9188
+        result = concat(
+            [DataFrame(date_range("2000", periods=1, tz="UTC")), DataFrame()]
+        )
+        expected = DataFrame(date_range("2000", periods=1, tz="UTC"))
+        tm.assert_frame_equal(result, expected)
+
+
+class TestPeriodConcat:
+    def test_concat_period_series(self):
+        x = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="D"))
+        y = Series(pd.PeriodIndex(["2015-10-01", "2016-01-01"], freq="D"))
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="Period[D]")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_period_multiple_freq_series(self):
+        x = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="D"))
+        y = Series(pd.PeriodIndex(["2015-10-01", "2016-01-01"], freq="M"))
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "object"
+
+    def test_concat_period_other_series(self):
+        x = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="D"))
+        y = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="M"))
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "object"
+
+    def test_concat_period_other_series2(self):
+        # non-period
+        x = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="D"))
+        y = Series(DatetimeIndex(["2015-11-01", "2015-12-01"]))
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "object"
+
+    def test_concat_period_other_series3(self):
+        x = Series(pd.PeriodIndex(["2015-11-01", "2015-12-01"], freq="D"))
+        y = Series(["A", "B"])
+        expected = Series([x[0], x[1], y[0], y[1]], dtype="object")
+        result = concat([x, y], ignore_index=True)
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "object"
+
+
+def test_concat_timedelta64_block():
+    rng = to_timedelta(np.arange(10), unit="s")
+
+    df = DataFrame({"time": rng})
+
+    result = concat([df, df])
+    tm.assert_frame_equal(result.iloc[:10], df, check_index_type=False)
+    tm.assert_frame_equal(result.iloc[10:], df, check_index_type=False)
+
+
+def test_concat_multiindex_datetime_nat():
+    # GH#44900
+    left = DataFrame({"a": 1}, index=MultiIndex.from_tuples([(1, pd.NaT)]))
+    right = DataFrame(
+        {"b": 2}, index=MultiIndex.from_tuples([(1, pd.NaT), (2, pd.NaT)])
+    )
+    result = concat([left, right], axis="columns")
+    expected = DataFrame(
+        {"a": [1.0, np.nan], "b": 2}, MultiIndex.from_tuples([(1, pd.NaT), (2, pd.NaT)])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_concat_float_datetime64():
+    # GH#32934
+    df_time = DataFrame({"A": pd.array(["2000"], dtype="datetime64[ns]")})
+    df_float = DataFrame({"A": pd.array([1.0], dtype="float64")})
+
+    expected = DataFrame(
+        {
+            "A": [
+                pd.array(["2000"], dtype="datetime64[ns]")[0],
+                pd.array([1.0], dtype="float64")[0],
+            ]
+        },
+        index=[0, 0],
+    )
+    result = concat([df_time, df_float])
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame({"A": pd.array([], dtype="object")})
+    result = concat([df_time.iloc[:0], df_float.iloc[:0]])
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame({"A": pd.array([1.0], dtype="object")})
+    result = concat([df_time.iloc[:0], df_float])
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame({"A": pd.array(["2000"], dtype="datetime64[ns]")}).astype(
+        object
+    )
+
+    result = concat([df_time, df_float.iloc[:0]])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_empty.py b/pandas/pandas/tests/reshape/concat/test_empty.py
new file mode 100644
index 0000000000000..4869cfbf3a556
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_empty.py
@@ -0,0 +1,293 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    RangeIndex,
+    Series,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestEmptyConcat:
+    def test_handle_empty_objects(self, sort, using_infer_string):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)), columns=list("abcd")
+        )
+
+        dfcopy = df[:5].copy()
+        dfcopy["foo"] = "bar"
+        empty = df[5:5]
+
+        frames = [dfcopy, empty, empty, df[5:]]
+        concatted = concat(frames, axis=0, sort=sort)
+
+        expected = df.reindex(columns=["a", "b", "c", "d", "foo"])
+        expected["foo"] = expected["foo"].astype(
+            object if not using_infer_string else "str"
+        )
+        expected.loc[0:4, "foo"] = "bar"
+
+        tm.assert_frame_equal(concatted, expected)
+
+        # empty as first element with time series
+        # GH3259
+        df = DataFrame(
+            {"A": range(10000)}, index=date_range("20130101", periods=10000, freq="s")
+        )
+        empty = DataFrame()
+        result = concat([df, empty], axis=1)
+        tm.assert_frame_equal(result, df)
+        result = concat([empty, df], axis=1)
+        tm.assert_frame_equal(result, df)
+
+        result = concat([df, empty])
+        tm.assert_frame_equal(result, df)
+        result = concat([empty, df])
+        tm.assert_frame_equal(result, df)
+
+    def test_concat_empty_series(self):
+        # GH 11082
+        s1 = Series([1, 2, 3], name="x")
+        s2 = Series(name="y", dtype="float64")
+        res = concat([s1, s2], axis=1)
+        exp = DataFrame(
+            {"x": [1, 2, 3], "y": [np.nan, np.nan, np.nan]},
+            index=RangeIndex(3),
+        )
+        tm.assert_frame_equal(res, exp)
+
+        s1 = Series([1, 2, 3], name="x")
+        s2 = Series(name="y", dtype="float64")
+        res = concat([s1, s2], axis=0)
+        # name will be reset
+        exp = Series([1, 2, 3], dtype="float64")
+        tm.assert_series_equal(res, exp)
+
+        # empty Series with no name
+        s1 = Series([1, 2, 3], name="x")
+        s2 = Series(name=None, dtype="float64")
+        res = concat([s1, s2], axis=1)
+        exp = DataFrame(
+            {"x": [1, 2, 3], 0: [np.nan, np.nan, np.nan]},
+            columns=["x", 0],
+            index=RangeIndex(3),
+        )
+        tm.assert_frame_equal(res, exp)
+
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    @pytest.mark.parametrize("values", [[], [1, 2, 3]])
+    def test_concat_empty_series_timelike(self, tz, values):
+        # GH 18447
+
+        first = Series([], dtype="M8[ns]").dt.tz_localize(tz)
+        dtype = None if values else np.float64
+        second = Series(values, dtype=dtype)
+
+        expected = DataFrame(
+            {
+                0: Series([pd.NaT] * len(values), dtype="M8[ns]").dt.tz_localize(tz),
+                1: values,
+            }
+        )
+        result = concat([first, second], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "left,right,expected",
+        [
+            # booleans
+            (np.bool_, np.int32, np.object_),  # changed from int32 in 2.0 GH#39817
+            (np.bool_, np.float32, np.object_),
+            # datetime-like
+            ("m8[ns]", np.bool_, np.object_),
+            ("m8[ns]", np.int64, np.object_),
+            ("M8[ns]", np.bool_, np.object_),
+            ("M8[ns]", np.int64, np.object_),
+            # categorical
+            ("category", "category", "category"),
+            ("category", "object", "object"),
+        ],
+    )
+    def test_concat_empty_series_dtypes(self, left, right, expected):
+        # GH#39817, GH#45101
+        result = concat([Series(dtype=left), Series(dtype=right)])
+        assert result.dtype == expected
+
+    @pytest.mark.parametrize(
+        "dtype", ["float64", "int8", "uint8", "bool", "m8[ns]", "M8[ns]"]
+    )
+    def test_concat_empty_series_dtypes_match_roundtrips(self, dtype):
+        dtype = np.dtype(dtype)
+
+        result = concat([Series(dtype=dtype)])
+        assert result.dtype == dtype
+
+        result = concat([Series(dtype=dtype), Series(dtype=dtype)])
+        assert result.dtype == dtype
+
+    @pytest.mark.parametrize("dtype", ["float64", "int8", "uint8", "m8[ns]", "M8[ns]"])
+    @pytest.mark.parametrize(
+        "dtype2",
+        ["float64", "int8", "uint8", "m8[ns]", "M8[ns]"],
+    )
+    def test_concat_empty_series_dtypes_roundtrips(self, dtype, dtype2):
+        # round-tripping with self & like self
+        if dtype == dtype2:
+            pytest.skip("same dtype is not applicable for test")
+
+        def int_result_type(dtype, dtype2):
+            typs = {dtype.kind, dtype2.kind}
+            if not len(typs - {"i", "u", "b"}) and (
+                dtype.kind == "i" or dtype2.kind == "i"
+            ):
+                return "i"
+            elif not len(typs - {"u", "b"}) and (
+                dtype.kind == "u" or dtype2.kind == "u"
+            ):
+                return "u"
+            return None
+
+        def float_result_type(dtype, dtype2):
+            typs = {dtype.kind, dtype2.kind}
+            if not len(typs - {"f", "i", "u"}) and (
+                dtype.kind == "f" or dtype2.kind == "f"
+            ):
+                return "f"
+            return None
+
+        def get_result_type(dtype, dtype2):
+            result = float_result_type(dtype, dtype2)
+            if result is not None:
+                return result
+            result = int_result_type(dtype, dtype2)
+            if result is not None:
+                return result
+            return "O"
+
+        dtype = np.dtype(dtype)
+        dtype2 = np.dtype(dtype2)
+        expected = get_result_type(dtype, dtype2)
+        result = concat([Series(dtype=dtype), Series(dtype=dtype2)]).dtype
+        assert result.kind == expected
+
+    def test_concat_empty_series_dtypes_triple(self):
+        assert (
+            concat(
+                [Series(dtype="M8[ns]"), Series(dtype=np.bool_), Series(dtype=np.int64)]
+            ).dtype
+            == np.object_
+        )
+
+    def test_concat_empty_series_dtype_category_with_array(self):
+        # GH#18515
+        assert (
+            concat(
+                [Series(np.array([]), dtype="category"), Series(dtype="float64")]
+            ).dtype
+            == "float64"
+        )
+
+    def test_concat_empty_series_dtypes_sparse(self):
+        result = concat(
+            [
+                Series(dtype="float64").astype("Sparse"),
+                Series(dtype="float64").astype("Sparse"),
+            ]
+        )
+        assert result.dtype == "Sparse[float64]"
+
+        result = concat(
+            [Series(dtype="float64").astype("Sparse"), Series(dtype="float64")]
+        )
+        expected = pd.SparseDtype(np.float64)
+        assert result.dtype == expected
+
+        result = concat(
+            [Series(dtype="float64").astype("Sparse"), Series(dtype="object")]
+        )
+        expected = pd.SparseDtype("object")
+        assert result.dtype == expected
+
+    def test_concat_empty_df_object_dtype(self):
+        # GH 9149
+        df_1 = DataFrame({"Row": [0, 1, 1], "EmptyCol": np.nan, "NumberCol": [1, 2, 3]})
+        df_2 = DataFrame(columns=df_1.columns)
+        result = concat([df_1, df_2], axis=0)
+        expected = df_1.astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_empty_dataframe_dtypes(self):
+        df = DataFrame(columns=list("abc"))
+        df["a"] = df["a"].astype(np.bool_)
+        df["b"] = df["b"].astype(np.int32)
+        df["c"] = df["c"].astype(np.float64)
+
+        result = concat([df, df])
+        assert result["a"].dtype == np.bool_
+        assert result["b"].dtype == np.int32
+        assert result["c"].dtype == np.float64
+
+        result = concat([df, df.astype(np.float64)])
+        assert result["a"].dtype == np.object_
+        assert result["b"].dtype == np.float64
+        assert result["c"].dtype == np.float64
+
+    def test_concat_inner_join_empty(self):
+        # GH 15328
+        df_empty = DataFrame()
+        df_a = DataFrame({"a": [1, 2]}, index=[0, 1], dtype="int64")
+        df_expected = DataFrame({"a": []}, index=RangeIndex(0), dtype="int64")
+
+        result = concat([df_a, df_empty], axis=1, join="inner")
+        tm.assert_frame_equal(result, df_expected)
+
+        result = concat([df_a, df_empty], axis=1, join="outer")
+        tm.assert_frame_equal(result, df_a)
+
+    def test_empty_dtype_coerce(self):
+        # xref to #12411
+        # xref to #12045
+        # xref to #11594
+        # see below
+
+        # 10571
+        df1 = DataFrame(data=[[1, None], [2, None]], columns=["a", "b"])
+        df2 = DataFrame(data=[[3, None], [4, None]], columns=["a", "b"])
+        result = concat([df1, df2])
+        expected = df1.dtypes
+        tm.assert_series_equal(result.dtypes, expected)
+
+    def test_concat_empty_dataframe(self):
+        # 39037
+        df1 = DataFrame(columns=["a", "b"])
+        df2 = DataFrame(columns=["b", "c"])
+        result = concat([df1, df2, df1])
+        expected = DataFrame(columns=["a", "b", "c"])
+        tm.assert_frame_equal(result, expected)
+
+        df3 = DataFrame(columns=["a", "b"])
+        df4 = DataFrame(columns=["b"])
+        result = concat([df3, df4])
+        expected = DataFrame(columns=["a", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_empty_dataframe_different_dtypes(self, using_infer_string):
+        # 39037
+        df1 = DataFrame({"a": [1, 2, 3], "b": ["a", "b", "c"]})
+        df2 = DataFrame({"a": [1, 2, 3]})
+
+        result = concat([df1[:0], df2[:0]])
+        assert result["a"].dtype == np.int64
+        assert result["b"].dtype == np.object_ if not using_infer_string else "str"
+
+    def test_concat_to_empty_ea(self):
+        """48510 `concat` to an empty EA should maintain type EA dtype."""
+        df_empty = DataFrame({"a": pd.array([], dtype=pd.Int64Dtype())})
+        df_new = DataFrame({"a": pd.array([1, 2, 3], dtype=pd.Int64Dtype())})
+        expected = df_new.copy()
+        result = concat([df_empty, df_new])
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_index.py b/pandas/pandas/tests/reshape/concat/test_index.py
new file mode 100644
index 0000000000000..17f1a9d4ecbf1
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_index.py
@@ -0,0 +1,467 @@
+from copy import deepcopy
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+
+class TestIndexConcat:
+    def test_concat_ignore_index(self, sort):
+        frame1 = DataFrame(
+            {"test1": ["a", "b", "c"], "test2": [1, 2, 3], "test3": [4.5, 3.2, 1.2]}
+        )
+        frame2 = DataFrame({"test3": [5.2, 2.2, 4.3]})
+        frame1.index = Index(["x", "y", "z"])
+        frame2.index = Index(["x", "y", "q"])
+
+        v1 = concat([frame1, frame2], axis=1, ignore_index=True, sort=sort)
+
+        nan = np.nan
+        expected = DataFrame(
+            [
+                [nan, nan, nan, 4.3],
+                ["a", 1, 4.5, 5.2],
+                ["b", 2, 3.2, 2.2],
+                ["c", 3, 1.2, nan],
+            ],
+            index=Index(["q", "x", "y", "z"]),
+        )
+        if not sort:
+            expected = expected.loc[["x", "y", "z", "q"]]
+
+        tm.assert_frame_equal(v1, expected)
+
+    @pytest.mark.parametrize(
+        "name_in1,name_in2,name_in3,name_out",
+        [
+            ("idx", "idx", "idx", "idx"),
+            ("idx", "idx", None, None),
+            ("idx", None, None, None),
+            ("idx1", "idx2", None, None),
+            ("idx1", "idx1", "idx2", None),
+            ("idx1", "idx2", "idx3", None),
+            (None, None, None, None),
+        ],
+    )
+    def test_concat_same_index_names(self, name_in1, name_in2, name_in3, name_out):
+        # GH13475
+        indices = [
+            Index(["a", "b", "c"], name=name_in1),
+            Index(["b", "c", "d"], name=name_in2),
+            Index(["c", "d", "e"], name=name_in3),
+        ]
+        frames = [
+            DataFrame({c: [0, 1, 2]}, index=i) for i, c in zip(indices, ["x", "y", "z"])
+        ]
+        result = concat(frames, axis=1)
+
+        exp_ind = Index(["a", "b", "c", "d", "e"], name=name_out)
+        expected = DataFrame(
+            {
+                "x": [0, 1, 2, np.nan, np.nan],
+                "y": [np.nan, 0, 1, 2, np.nan],
+                "z": [np.nan, np.nan, 0, 1, 2],
+            },
+            index=exp_ind,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_rename_index(self):
+        a = DataFrame(
+            np.random.default_rng(2).random((3, 3)),
+            columns=list("ABC"),
+            index=Index(list("abc"), name="index_a"),
+        )
+        b = DataFrame(
+            np.random.default_rng(2).random((3, 3)),
+            columns=list("ABC"),
+            index=Index(list("abc"), name="index_b"),
+        )
+
+        result = concat([a, b], keys=["key0", "key1"], names=["lvl0", "lvl1"])
+
+        exp = concat([a, b], keys=["key0", "key1"], names=["lvl0"])
+        names = list(exp.index.names)
+        names[1] = "lvl1"
+        exp.index.set_names(names, inplace=True)
+
+        tm.assert_frame_equal(result, exp)
+        assert result.index.names == exp.index.names
+
+    def test_concat_copy_index_series(self, axis):
+        # GH 29879
+        ser = Series([1, 2])
+        comb = concat([ser, ser], axis=axis)
+        if axis in [0, "index"]:
+            assert comb.index is not ser.index
+        else:
+            assert comb.index is ser.index
+
+    def test_concat_copy_index_frame(self, axis):
+        # GH 29879
+        df = DataFrame([[1, 2], [3, 4]], columns=["a", "b"])
+        comb = concat([df, df], axis=axis)
+        if axis in [0, "index"]:
+            assert not comb.index.is_(df.index)
+            assert comb.columns.is_(df.columns)
+        elif axis in [1, "columns"]:
+            assert comb.index.is_(df.index)
+            assert not comb.columns.is_(df.columns)
+
+    def test_default_index(self):
+        # is_series and ignore_index
+        s1 = Series([1, 2, 3], name="x")
+        s2 = Series([4, 5, 6], name="y")
+        res = concat([s1, s2], axis=1, ignore_index=True)
+        assert isinstance(res.columns, pd.RangeIndex)
+        exp = DataFrame([[1, 4], [2, 5], [3, 6]])
+        # use check_index_type=True to check the result have
+        # RangeIndex (default index)
+        tm.assert_frame_equal(res, exp, check_index_type=True, check_column_type=True)
+
+        # is_series and all inputs have no names
+        s1 = Series([1, 2, 3])
+        s2 = Series([4, 5, 6])
+        res = concat([s1, s2], axis=1, ignore_index=False)
+        assert isinstance(res.columns, pd.RangeIndex)
+        exp = DataFrame([[1, 4], [2, 5], [3, 6]])
+        exp.columns = pd.RangeIndex(2)
+        tm.assert_frame_equal(res, exp, check_index_type=True, check_column_type=True)
+
+        # is_dataframe and ignore_index
+        df1 = DataFrame({"A": [1, 2], "B": [5, 6]})
+        df2 = DataFrame({"A": [3, 4], "B": [7, 8]})
+
+        res = concat([df1, df2], axis=0, ignore_index=True)
+        exp = DataFrame([[1, 5], [2, 6], [3, 7], [4, 8]], columns=["A", "B"])
+        tm.assert_frame_equal(res, exp, check_index_type=True, check_column_type=True)
+
+        res = concat([df1, df2], axis=1, ignore_index=True)
+        exp = DataFrame([[1, 5, 3, 7], [2, 6, 4, 8]])
+        tm.assert_frame_equal(res, exp, check_index_type=True, check_column_type=True)
+
+    def test_dups_index(self):
+        # GH 4771
+
+        # single dtypes
+        df = DataFrame(
+            np.random.default_rng(2).integers(0, 10, size=40).reshape(10, 4),
+            columns=["A", "A", "C", "C"],
+        )
+
+        result = concat([df, df], axis=1)
+        tm.assert_frame_equal(result.iloc[:, :4], df)
+        tm.assert_frame_equal(result.iloc[:, 4:], df)
+
+        result = concat([df, df], axis=0)
+        tm.assert_frame_equal(result.iloc[:10], df)
+        tm.assert_frame_equal(result.iloc[10:], df)
+
+        # multi dtypes
+        df = concat(
+            [
+                DataFrame(
+                    np.random.default_rng(2).standard_normal((10, 4)),
+                    columns=["A", "A", "B", "B"],
+                ),
+                DataFrame(
+                    np.random.default_rng(2).integers(0, 10, size=20).reshape(10, 2),
+                    columns=["A", "C"],
+                ),
+            ],
+            axis=1,
+        )
+
+        result = concat([df, df], axis=1)
+        tm.assert_frame_equal(result.iloc[:, :6], df)
+        tm.assert_frame_equal(result.iloc[:, 6:], df)
+
+        result = concat([df, df], axis=0)
+        tm.assert_frame_equal(result.iloc[:10], df)
+        tm.assert_frame_equal(result.iloc[10:], df)
+
+        # append
+        result = df.iloc[0:8, :]._append(df.iloc[8:])
+        tm.assert_frame_equal(result, df)
+
+        result = df.iloc[0:8, :]._append(df.iloc[8:9])._append(df.iloc[9:10])
+        tm.assert_frame_equal(result, df)
+
+        expected = concat([df, df], axis=0)
+        result = df._append(df)
+        tm.assert_frame_equal(result, expected)
+
+
+class TestMultiIndexConcat:
+    def test_concat_multiindex_with_keys(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+        index = frame.index
+        result = concat([frame, frame], keys=[0, 1], names=["iteration"])
+
+        assert result.index.names == ("iteration",) + index.names
+        tm.assert_frame_equal(result.loc[0], frame)
+        tm.assert_frame_equal(result.loc[1], frame)
+        assert result.index.nlevels == 3
+
+    def test_concat_multiindex_with_none_in_index_names(self):
+        # GH 15787
+        index = MultiIndex.from_product([[1], range(5)], names=["level1", None])
+        df = DataFrame({"col": range(5)}, index=index, dtype=np.int32)
+
+        result = concat([df, df], keys=[1, 2], names=["level2"])
+        index = MultiIndex.from_product(
+            [[1, 2], [1], range(5)], names=["level2", "level1", None]
+        )
+        expected = DataFrame({"col": list(range(5)) * 2}, index=index, dtype=np.int32)
+        tm.assert_frame_equal(result, expected)
+
+        result = concat([df, df[:2]], keys=[1, 2], names=["level2"])
+        level2 = [1] * 5 + [2] * 2
+        level1 = [1] * 7
+        no_name = list(range(5)) + list(range(2))
+        tuples = list(zip(level2, level1, no_name))
+        index = MultiIndex.from_tuples(tuples, names=["level2", "level1", None])
+        expected = DataFrame({"col": no_name}, index=index, dtype=np.int32)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_multiindex_rangeindex(self):
+        # GH13542
+        # when multi-index levels are RangeIndex objects
+        # there is a bug in concat with objects of len 1
+
+        df = DataFrame(np.random.default_rng(2).standard_normal((9, 2)))
+        df.index = MultiIndex(
+            levels=[pd.RangeIndex(3), pd.RangeIndex(3)],
+            codes=[np.repeat(np.arange(3), 3), np.tile(np.arange(3), 3)],
+        )
+
+        res = concat([df.iloc[[2, 3, 4], :], df.iloc[[5], :]])
+        exp = df.iloc[[2, 3, 4, 5], :]
+        tm.assert_frame_equal(res, exp)
+
+    def test_concat_multiindex_dfs_with_deepcopy(self):
+        # GH 9967
+        example_multiindex1 = MultiIndex.from_product([["a"], ["b"]])
+        example_dataframe1 = DataFrame([0], index=example_multiindex1)
+
+        example_multiindex2 = MultiIndex.from_product([["a"], ["c"]])
+        example_dataframe2 = DataFrame([1], index=example_multiindex2)
+
+        example_dict = {"s1": example_dataframe1, "s2": example_dataframe2}
+        expected_index = MultiIndex(
+            levels=[["s1", "s2"], ["a"], ["b", "c"]],
+            codes=[[0, 1], [0, 0], [0, 1]],
+            names=["testname", None, None],
+        )
+        expected = DataFrame([[0], [1]], index=expected_index)
+        result_copy = concat(deepcopy(example_dict), names=["testname"])
+        tm.assert_frame_equal(result_copy, expected)
+        result_no_copy = concat(example_dict, names=["testname"])
+        tm.assert_frame_equal(result_no_copy, expected)
+
+    @pytest.mark.parametrize(
+        "mi1_list",
+        [
+            [["a"], range(2)],
+            [["b"], np.arange(2.0, 4.0)],
+            [["c"], ["A", "B"]],
+            [["d"], pd.date_range(start="2017", end="2018", periods=2)],
+        ],
+    )
+    @pytest.mark.parametrize(
+        "mi2_list",
+        [
+            [["a"], range(2)],
+            [["b"], np.arange(2.0, 4.0)],
+            [["c"], ["A", "B"]],
+            [["d"], pd.date_range(start="2017", end="2018", periods=2)],
+        ],
+    )
+    def test_concat_with_various_multiindex_dtypes(
+        self, mi1_list: list, mi2_list: list
+    ):
+        # GitHub #23478
+        mi1 = MultiIndex.from_product(mi1_list)
+        mi2 = MultiIndex.from_product(mi2_list)
+
+        df1 = DataFrame(np.zeros((1, len(mi1))), columns=mi1)
+        df2 = DataFrame(np.zeros((1, len(mi2))), columns=mi2)
+
+        if mi1_list[0] == mi2_list[0]:
+            expected_mi = MultiIndex(
+                levels=[mi1_list[0], list(mi1_list[1])],
+                codes=[[0, 0, 0, 0], [0, 1, 0, 1]],
+            )
+        else:
+            expected_mi = MultiIndex(
+                levels=[
+                    mi1_list[0] + mi2_list[0],
+                    list(mi1_list[1]) + list(mi2_list[1]),
+                ],
+                codes=[[0, 0, 1, 1], [0, 1, 2, 3]],
+            )
+
+        expected_df = DataFrame(np.zeros((1, len(expected_mi))), columns=expected_mi)
+
+        with tm.assert_produces_warning(None):
+            result_df = concat((df1, df2), axis=1)
+
+        tm.assert_frame_equal(expected_df, result_df)
+
+    def test_concat_multiindex_(self):
+        # GitHub #44786
+        df = DataFrame({"col": ["a", "b", "c"]}, index=["1", "2", "2"])
+        df = concat([df], keys=["X"])
+
+        iterables = [["X"], ["1", "2", "2"]]
+        result_index = df.index
+        expected_index = MultiIndex.from_product(iterables)
+
+        tm.assert_index_equal(result_index, expected_index)
+
+        result_df = df
+        expected_df = DataFrame(
+            {"col": ["a", "b", "c"]}, index=MultiIndex.from_product(iterables)
+        )
+        tm.assert_frame_equal(result_df, expected_df)
+
+    def test_concat_with_key_not_unique(self, performance_warning):
+        # GitHub #46519
+        df1 = DataFrame({"name": [1]})
+        df2 = DataFrame({"name": [2]})
+        df3 = DataFrame({"name": [3]})
+        df_a = concat([df1, df2, df3], keys=["x", "y", "x"])
+        # the warning is caused by indexing unsorted multi-index
+        with tm.assert_produces_warning(
+            performance_warning, match="indexing past lexsort depth"
+        ):
+            out_a = df_a.loc[("x", 0), :]
+        df_b = DataFrame(
+            {"name": [1, 2, 3]},
+            index=MultiIndex(
+                levels=[["x", "y"], range(1)], codes=[[0, 1, 0], [0, 0, 0]]
+            ),
+        )
+        with tm.assert_produces_warning(
+            performance_warning, match="indexing past lexsort depth"
+        ):
+            out_b = df_b.loc[("x", 0)]
+
+        tm.assert_frame_equal(out_a, out_b)
+
+        df1 = DataFrame({"name": ["a", "a", "b"]})
+        df2 = DataFrame({"name": ["a", "b"]})
+        df3 = DataFrame({"name": ["c", "d"]})
+        df_a = concat([df1, df2, df3], keys=["x", "y", "x"])
+        with tm.assert_produces_warning(
+            performance_warning, match="indexing past lexsort depth"
+        ):
+            out_a = df_a.loc[("x", 0), :]
+
+        df_b = DataFrame(
+            {
+                "a": ["x", "x", "x", "y", "y", "x", "x"],
+                "b": [0, 1, 2, 0, 1, 0, 1],
+                "name": list("aababcd"),
+            }
+        ).set_index(["a", "b"])
+        df_b.index.names = [None, None]
+        with tm.assert_produces_warning(
+            performance_warning, match="indexing past lexsort depth"
+        ):
+            out_b = df_b.loc[("x", 0), :]
+
+        tm.assert_frame_equal(out_a, out_b)
+
+    def test_concat_with_duplicated_levels(self):
+        # keyword levels should be unique
+        df1 = DataFrame({"A": [1]}, index=["x"])
+        df2 = DataFrame({"A": [1]}, index=["y"])
+        msg = r"Level values not unique: \['x', 'y', 'y'\]"
+        with pytest.raises(ValueError, match=msg):
+            concat([df1, df2], keys=["x", "y"], levels=[["x", "y", "y"]])
+
+    @pytest.mark.parametrize("levels", [[["x", "y"]], [["x", "y", "y"]]])
+    def test_concat_with_levels_with_none_keys(self, levels):
+        df1 = DataFrame({"A": [1]}, index=["x"])
+        df2 = DataFrame({"A": [1]}, index=["y"])
+        msg = "levels supported only when keys is not None"
+        with pytest.raises(ValueError, match=msg):
+            concat([df1, df2], levels=levels)
+
+    def test_concat_range_index_result(self):
+        # GH#47501
+        df1 = DataFrame({"a": [1, 2]})
+        df2 = DataFrame({"b": [1, 2]})
+
+        result = concat([df1, df2], sort=True, axis=1)
+        expected = DataFrame({"a": [1, 2], "b": [1, 2]})
+        tm.assert_frame_equal(result, expected)
+        expected_index = pd.RangeIndex(0, 2)
+        tm.assert_index_equal(result.index, expected_index, exact=True)
+
+    def test_concat_index_keep_dtype(self):
+        # GH#47329
+        df1 = DataFrame([[0, 1, 1]], columns=Index([1, 2, 3], dtype="object"))
+        df2 = DataFrame([[0, 1]], columns=Index([1, 2], dtype="object"))
+        result = concat([df1, df2], ignore_index=True, join="outer", sort=True)
+        expected = DataFrame(
+            [[0, 1, 1.0], [0, 1, np.nan]], columns=Index([1, 2, 3], dtype="object")
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_index_keep_dtype_ea_numeric(self, any_numeric_ea_dtype):
+        # GH#47329
+        df1 = DataFrame(
+            [[0, 1, 1]], columns=Index([1, 2, 3], dtype=any_numeric_ea_dtype)
+        )
+        df2 = DataFrame([[0, 1]], columns=Index([1, 2], dtype=any_numeric_ea_dtype))
+        result = concat([df1, df2], ignore_index=True, join="outer", sort=True)
+        expected = DataFrame(
+            [[0, 1, 1.0], [0, 1, np.nan]],
+            columns=Index([1, 2, 3], dtype=any_numeric_ea_dtype),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["Int8", "Int16", "Int32"])
+    def test_concat_index_find_common(self, dtype):
+        # GH#47329
+        df1 = DataFrame([[0, 1, 1]], columns=Index([1, 2, 3], dtype=dtype))
+        df2 = DataFrame([[0, 1]], columns=Index([1, 2], dtype="Int32"))
+        result = concat([df1, df2], ignore_index=True, join="outer", sort=True)
+        expected = DataFrame(
+            [[0, 1, 1.0], [0, 1, np.nan]], columns=Index([1, 2, 3], dtype="Int32")
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_axis_1_sort_false_rangeindex(self, using_infer_string):
+        # GH 46675
+        s1 = Series(["a", "b", "c"])
+        s2 = Series(["a", "b"])
+        s3 = Series(["a", "b", "c", "d"])
+        s4 = Series([], dtype=object if not using_infer_string else "str")
+        result = concat(
+            [s1, s2, s3, s4], sort=False, join="outer", ignore_index=False, axis=1
+        )
+        expected = DataFrame(
+            [
+                ["a"] * 3 + [np.nan],
+                ["b"] * 3 + [np.nan],
+                ["c", np.nan] * 2,
+                [np.nan] * 2 + ["d"] + [np.nan],
+            ],
+            dtype=object if not using_infer_string else "str",
+        )
+        tm.assert_frame_equal(
+            result, expected, check_index_type=True, check_column_type=True
+        )
diff --git a/pandas/pandas/tests/reshape/concat/test_invalid.py b/pandas/pandas/tests/reshape/concat/test_invalid.py
new file mode 100644
index 0000000000000..b1c44fc0debc3
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_invalid.py
@@ -0,0 +1,54 @@
+from io import StringIO
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    concat,
+    read_csv,
+)
+import pandas._testing as tm
+
+
+class TestInvalidConcat:
+    @pytest.mark.parametrize("obj", [1, {}, [1, 2], (1, 2)])
+    def test_concat_invalid(self, obj):
+        # trying to concat a ndframe with a non-ndframe
+        df1 = DataFrame(range(2))
+        msg = (
+            f"cannot concatenate object of type '{type(obj)}'; "
+            "only Series and DataFrame objs are valid"
+        )
+        with pytest.raises(TypeError, match=msg):
+            concat([df1, obj])
+
+    def test_concat_invalid_first_argument(self):
+        df1 = DataFrame(range(2))
+        msg = (
+            "first argument must be an iterable of pandas "
+            'objects, you passed an object of type "DataFrame"'
+        )
+        with pytest.raises(TypeError, match=msg):
+            concat(df1)
+
+    def test_concat_generator_obj(self):
+        # generator ok though
+        concat(DataFrame(np.random.default_rng(2).random((5, 5))) for _ in range(3))
+
+    def test_concat_textreader_obj(self):
+        # text reader ok
+        # GH6583
+        data = """index,A,B,C,D
+                  foo,2,3,4,5
+                  bar,7,8,9,10
+                  baz,12,13,14,15
+                  qux,12,13,14,15
+                  foo2,12,13,14,15
+                  bar2,12,13,14,15
+               """
+
+        with read_csv(StringIO(data), chunksize=1) as reader:
+            result = concat(reader, ignore_index=True)
+        expected = read_csv(StringIO(data))
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/concat/test_series.py b/pandas/pandas/tests/reshape/concat/test_series.py
new file mode 100644
index 0000000000000..3523340bb2858
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_series.py
@@ -0,0 +1,172 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesConcat:
+    def test_concat_series(self):
+        ts = Series(
+            np.arange(20, dtype=np.float64),
+            index=date_range("2020-01-01", periods=20),
+            name="foo",
+        )
+        ts.name = "foo"
+
+        pieces = [ts[:5], ts[5:15], ts[15:]]
+
+        result = concat(pieces)
+        tm.assert_series_equal(result, ts)
+        assert result.name == ts.name
+
+        result = concat(pieces, keys=[0, 1, 2])
+        expected = ts.copy()
+        exp_codes = [np.repeat([0, 1, 2], [len(x) for x in pieces]), np.arange(len(ts))]
+        exp_index = MultiIndex(
+            levels=[[0, 1, 2], DatetimeIndex(ts.index.to_numpy(dtype="M8[ns]"))],
+            codes=exp_codes,
+        )
+        expected.index = exp_index
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_empty_and_non_empty_series_regression(self):
+        # GH 18187 regression test
+        s1 = Series([1])
+        s2 = Series([], dtype=object)
+
+        expected = s1.astype(object)
+        result = concat([s1, s2])
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_series_axis1(self):
+        ts = Series(
+            np.arange(10, dtype=np.float64), index=date_range("2020-01-01", periods=10)
+        )
+
+        pieces = [ts[:-2], ts[2:], ts[2:-2]]
+
+        result = concat(pieces, axis=1)
+        expected = DataFrame(pieces).T
+        tm.assert_frame_equal(result, expected)
+
+        result = concat(pieces, keys=["A", "B", "C"], axis=1)
+        expected = DataFrame(pieces, index=["A", "B", "C"]).T
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_series_axis1_preserves_series_names(self):
+        # preserve series names, #2489
+        s = Series(np.random.default_rng(2).standard_normal(5), name="A")
+        s2 = Series(np.random.default_rng(2).standard_normal(5), name="B")
+
+        result = concat([s, s2], axis=1)
+        expected = DataFrame({"A": s, "B": s2})
+        tm.assert_frame_equal(result, expected)
+
+        s2.name = None
+        result = concat([s, s2], axis=1)
+        tm.assert_index_equal(result.columns, Index(["A", 0], dtype="object"))
+
+    def test_concat_series_axis1_with_reindex(self, sort):
+        # must reindex, #2603
+        s = Series(
+            np.random.default_rng(2).standard_normal(3), index=["c", "a", "b"], name="A"
+        )
+        s2 = Series(
+            np.random.default_rng(2).standard_normal(4),
+            index=["d", "a", "b", "c"],
+            name="B",
+        )
+        result = concat([s, s2], axis=1, sort=sort)
+        expected = DataFrame({"A": s, "B": s2}, index=["c", "a", "b", "d"])
+        if sort:
+            expected = expected.sort_index()
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_series_axis1_names_applied(self):
+        # ensure names argument is not ignored on axis=1, #23490
+        s = Series([1, 2, 3])
+        s2 = Series([4, 5, 6])
+        result = concat([s, s2], axis=1, keys=["a", "b"], names=["A"])
+        expected = DataFrame(
+            [[1, 4], [2, 5], [3, 6]], columns=Index(["a", "b"], name="A")
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = concat([s, s2], axis=1, keys=[("a", 1), ("b", 2)], names=["A", "B"])
+        expected = DataFrame(
+            [[1, 4], [2, 5], [3, 6]],
+            columns=MultiIndex.from_tuples([("a", 1), ("b", 2)], names=["A", "B"]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_series_axis1_same_names_ignore_index(self):
+        dates = date_range("01-Jan-2013", "01-Jan-2014", freq="MS")[0:-1]
+        s1 = Series(
+            np.random.default_rng(2).standard_normal(len(dates)),
+            index=dates,
+            name="value",
+        )
+        s2 = Series(
+            np.random.default_rng(2).standard_normal(len(dates)),
+            index=dates,
+            name="value",
+        )
+
+        result = concat([s1, s2], axis=1, ignore_index=True)
+        expected = Index(range(2))
+
+        tm.assert_index_equal(result.columns, expected, exact=True)
+
+    @pytest.mark.parametrize("s1name", [np.int64(190), 190])
+    def test_concat_series_name_npscalar_tuple(self, s1name):
+        # GH21015
+        s2name = (43, 0)
+        s1 = Series({"a": 1, "b": 2}, name=s1name)
+        s2 = Series({"c": 5, "d": 6}, name=s2name)
+        result = concat([s1, s2])
+        expected = Series({"a": 1, "b": 2, "c": 5, "d": 6})
+        tm.assert_series_equal(result, expected)
+
+    def test_concat_series_partial_columns_names(self):
+        # GH10698
+        named_series = Series([1, 2], name="foo")
+        unnamed_series1 = Series([1, 2])
+        unnamed_series2 = Series([4, 5])
+
+        result = concat([named_series, unnamed_series1, unnamed_series2], axis=1)
+        expected = DataFrame(
+            {"foo": [1, 2], 0: [1, 2], 1: [4, 5]}, columns=["foo", 0, 1]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = concat(
+            [named_series, unnamed_series1, unnamed_series2],
+            axis=1,
+            keys=["red", "blue", "yellow"],
+        )
+        expected = DataFrame(
+            {"red": [1, 2], "blue": [1, 2], "yellow": [4, 5]},
+            columns=["red", "blue", "yellow"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = concat(
+            [named_series, unnamed_series1, unnamed_series2], axis=1, ignore_index=True
+        )
+        expected = DataFrame({0: [1, 2], 1: [1, 2], 2: [4, 5]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_series_length_one_reversed(self, frame_or_series):
+        # GH39401
+        obj = frame_or_series([100])
+        result = concat([obj.iloc[::-1]])
+        tm.assert_equal(result, obj)
diff --git a/pandas/pandas/tests/reshape/concat/test_sort.py b/pandas/pandas/tests/reshape/concat/test_sort.py
new file mode 100644
index 0000000000000..2724f81958893
--- /dev/null
+++ b/pandas/pandas/tests/reshape/concat/test_sort.py
@@ -0,0 +1,118 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestConcatSort:
+    def test_concat_sorts_columns(self, sort):
+        # GH-4588
+        df1 = DataFrame({"a": [1, 2], "b": [1, 2]}, columns=["b", "a"])
+        df2 = DataFrame({"a": [3, 4], "c": [5, 6]})
+
+        # for sort=True/None
+        expected = DataFrame(
+            {"a": [1, 2, 3, 4], "b": [1, 2, None, None], "c": [None, None, 5, 6]},
+            columns=["a", "b", "c"],
+        )
+
+        if sort is False:
+            expected = expected[["b", "a", "c"]]
+
+        # default
+        with tm.assert_produces_warning(None):
+            result = pd.concat([df1, df2], ignore_index=True, sort=sort)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_sorts_index(self, sort):
+        df1 = DataFrame({"a": [1, 2, 3]}, index=["c", "a", "b"])
+        df2 = DataFrame({"b": [1, 2]}, index=["a", "b"])
+
+        # For True/None
+        expected = DataFrame(
+            {"a": [2, 3, 1], "b": [1, 2, None]},
+            index=["a", "b", "c"],
+            columns=["a", "b"],
+        )
+        if sort is False:
+            expected = expected.loc[["c", "a", "b"]]
+
+        # Warn and sort by default
+        with tm.assert_produces_warning(None):
+            result = pd.concat([df1, df2], axis=1, sort=sort)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_inner_sort(self, sort):
+        # https://github.com/pandas-dev/pandas/pull/20613
+        df1 = DataFrame(
+            {"a": [1, 2], "b": [1, 2], "c": [1, 2]}, columns=["b", "a", "c"]
+        )
+        df2 = DataFrame({"a": [1, 2], "b": [3, 4]}, index=[3, 4])
+
+        with tm.assert_produces_warning(None):
+            # unset sort should *not* warn for inner join
+            # since that never sorted
+            result = pd.concat([df1, df2], sort=sort, join="inner", ignore_index=True)
+
+        expected = DataFrame({"b": [1, 2, 3, 4], "a": [1, 2, 1, 2]}, columns=["b", "a"])
+        if sort is True:
+            expected = expected[["a", "b"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_aligned_sort(self):
+        # GH-4588
+        df = DataFrame({"c": [1, 2], "b": [3, 4], "a": [5, 6]}, columns=["c", "b", "a"])
+        result = pd.concat([df, df], sort=True, ignore_index=True)
+        expected = DataFrame(
+            {"a": [5, 6, 5, 6], "b": [3, 4, 3, 4], "c": [1, 2, 1, 2]},
+            columns=["a", "b", "c"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pd.concat(
+            [df, df[["c", "b"]]], join="inner", sort=True, ignore_index=True
+        )
+        expected = expected[["b", "c"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_aligned_sort_does_not_raise(self):
+        # GH-4588
+        # We catch TypeErrors from sorting internally and do not re-raise.
+        df = DataFrame({1: [1, 2], "a": [3, 4]}, columns=[1, "a"])
+        expected = DataFrame({1: [1, 2, 1, 2], "a": [3, 4, 3, 4]}, columns=[1, "a"])
+        result = pd.concat([df, df], ignore_index=True, sort=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_frame_with_sort_false(self):
+        # GH 43375
+        result = pd.concat(
+            [DataFrame({i: i}, index=[i]) for i in range(2, 0, -1)], sort=False
+        )
+        expected = DataFrame([[2, np.nan], [np.nan, 1]], index=[2, 1], columns=[2, 1])
+
+        tm.assert_frame_equal(result, expected)
+
+        # GH 37937
+        df1 = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]}, index=[1, 2, 3])
+        df2 = DataFrame({"c": [7, 8, 9], "d": [10, 11, 12]}, index=[3, 1, 6])
+        result = pd.concat([df2, df1], axis=1, sort=False)
+        expected = DataFrame(
+            [
+                [7.0, 10.0, 3.0, 6.0],
+                [8.0, 11.0, 1.0, 4.0],
+                [9.0, 12.0, np.nan, np.nan],
+                [np.nan, np.nan, 2.0, 5.0],
+            ],
+            index=[3, 1, 6, 2],
+            columns=["c", "d", "a", "b"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_concat_sort_none_raises(self):
+        # GH#41518
+        df = DataFrame({1: [1, 2], "a": [3, 4]})
+        msg = "The 'sort' keyword only accepts boolean values; None was passed."
+        with pytest.raises(ValueError, match=msg):
+            pd.concat([df, df], sort=None)
diff --git a/pandas/pandas/tests/reshape/data/cut_data.csv b/pandas/pandas/tests/reshape/data/cut_data.csv
new file mode 100644
index 0000000000000..c198ec77e45da
--- /dev/null
+++ b/pandas/pandas/tests/reshape/data/cut_data.csv
@@ -0,0 +1 @@
+1.001 0.994 0.9951 0.9956 0.9956 0.9951 0.9949 1.001 0.994 0.9938 0.9908 0.9947 0.992 0.9912 1.0002 0.9914 0.9928 0.9892 0.9917 0.9955 0.9892 0.9912 0.993 0.9937 0.9951 0.9955 0.993 0.9961 0.9914 0.9906 0.9974 0.9934 0.992 0.9939 0.9962 0.9905 0.9934 0.9906 0.9999 0.9999 0.9937 0.9937 0.9954 0.9934 0.9934 0.9931 0.994 0.9939 0.9954 0.995 0.9917 0.9914 0.991 0.9911 0.993 0.9908 0.9962 0.9972 0.9931 0.9926 0.9951 0.9972 0.991 0.9931 0.9927 0.9934 0.9903 0.992 0.9926 0.9962 0.9956 0.9958 0.9964 0.9941 0.9926 0.9962 0.9898 0.9912 0.9961 0.9949 0.9929 0.9985 0.9946 0.9966 0.9974 0.9975 0.9974 0.9972 0.9974 0.9975 0.9974 0.9957 0.99 0.9899 0.9916 0.9969 0.9979 0.9913 0.9956 0.9979 0.9975 0.9962 0.997 1 0.9975 0.9974 0.9962 0.999 0.999 0.9927 0.9959 1 0.9982 0.9968 0.9968 0.994 0.9914 0.9911 0.9982 0.9982 0.9934 0.9984 0.9952 0.9952 0.9928 0.9912 0.994 0.9958 0.9924 0.9924 0.994 0.9958 0.9979 0.9982 0.9961 0.9979 0.992 0.9975 0.9917 0.9923 0.9927 0.9975 0.992 0.9947 0.9921 0.9905 0.9918 0.9951 0.9917 0.994 0.9934 0.9968 0.994 0.9919 0.9966 0.9979 0.9979 0.9898 0.9894 0.9894 0.9898 0.998 0.9932 0.9979 0.997 0.9972 0.9974 0.9896 0.9968 0.9958 0.9906 0.9917 0.9902 0.9918 0.999 0.9927 0.991 0.9972 0.9931 0.995 0.9951 0.9936 1.001 0.9979 0.997 0.9972 0.9954 0.9924 0.9906 0.9962 0.9962 1.001 0.9928 0.9942 0.9942 0.9942 0.9942 0.9961 0.998 0.9961 0.9984 0.998 0.9973 0.9949 0.9924 0.9972 0.9958 0.9968 0.9938 0.993 0.994 0.9918 0.9958 0.9944 0.9912 0.9961 0.9939 0.9961 0.9989 0.9938 0.9939 0.9971 0.9912 0.9936 0.9929 0.9998 0.9938 0.9969 0.9938 0.9998 0.9972 0.9976 0.9976 0.9979 0.9979 0.9979 0.9979 0.9972 0.9918 0.9982 0.9985 0.9944 0.9903 0.9934 0.9975 0.9923 0.99 0.9905 0.9905 0.996 0.9964 0.998 0.9975 0.9913 0.9932 0.9935 0.9927 0.9927 0.9912 0.9904 0.9939 0.9996 0.9944 0.9977 0.9912 0.9996 0.9965 0.9944 0.9945 0.9944 0.9965 0.9944 0.9972 0.9949 0.9966 0.9954 0.9954 0.9915 0.9919 0.9916 0.99 0.9909 0.9938 0.9982 0.9988 0.9961 0.9978 0.9979 0.9979 0.9979 0.9979 0.9945 1 0.9957 0.9968 0.9934 0.9976 0.9932 0.997 0.9923 0.9914 0.992 0.9914 0.9914 0.9949 0.9949 0.995 0.995 0.9927 0.9928 0.9917 0.9918 0.9954 0.9941 0.9941 0.9934 0.9927 0.9938 0.9933 0.9934 0.9927 0.9938 0.9927 0.9946 0.993 0.9946 0.9976 0.9944 0.9978 0.992 0.9912 0.9927 0.9906 0.9954 0.9923 0.9906 0.991 0.9972 0.9945 0.9934 0.9964 0.9948 0.9962 0.9931 0.993 0.9942 0.9906 0.9995 0.998 0.997 0.9914 0.992 0.9924 0.992 0.9937 0.9978 0.9978 0.9927 0.994 0.9935 0.9968 0.9941 0.9942 0.9978 0.9923 0.9912 0.9923 0.9927 0.9931 0.9941 0.9927 0.9931 0.9934 0.9936 0.9893 0.9893 0.9919 0.9924 0.9927 0.9919 0.9924 0.9975 0.9969 0.9936 0.991 0.9893 0.9906 0.9941 0.995 0.9983 0.9983 0.9916 0.9957 0.99 0.9976 0.992 0.9917 0.9917 0.9993 0.9908 0.9917 0.9976 0.9934 1 0.9918 0.992 0.9896 0.9932 0.992 0.9917 0.9999 0.998 0.9918 0.9918 0.9999 0.998 0.9927 0.9959 0.9927 0.9929 0.9898 0.9954 0.9954 0.9954 0.9954 0.9954 0.9954 0.9974 0.9936 0.9978 0.9974 0.9927 0.9934 0.9938 0.9922 0.992 0.9935 0.9906 0.9934 0.9934 0.9913 0.9938 0.9898 0.9975 0.9975 0.9937 0.9914 0.9982 0.9982 0.9929 0.9971 0.9921 0.9931 0.9924 0.9929 0.9982 0.9892 0.9956 0.9924 0.9971 0.9956 0.9982 0.9973 0.9932 0.9976 0.9962 0.9956 0.9932 0.9976 0.9992 0.9983 0.9937 0.99 0.9944 0.9938 0.9965 0.9893 0.9927 0.994 0.9928 0.9964 0.9917 0.9972 0.9964 0.9954 0.993 0.9928 0.9916 0.9936 0.9962 0.9899 0.9898 0.996 0.9907 0.994 0.9913 0.9976 0.9904 0.992 0.9976 0.999 0.9975 0.9937 0.9937 0.998 0.998 0.9944 0.9938 0.9907 0.9938 0.9921 0.9908 0.9931 0.9915 0.9952 0.9926 0.9934 0.992 0.9918 0.9942 0.9942 0.9942 0.9901 0.9898 0.9902 0.9934 0.9906 0.9898 0.9896 0.9922 0.9947 0.9945 0.9976 0.9976 0.9976 0.9987 0.9987 0.9976 0.992 0.9955 0.9953 0.9976 0.992 0.9952 0.9983 0.9933 0.9958 0.9922 0.9928 0.9976 0.9976 0.9916 0.9901 0.9976 0.9901 0.9916 0.9982 0.993 0.9969 0.991 0.9953 0.9924 0.9969 0.9928 0.9945 0.9967 0.9944 0.9928 0.9929 0.9948 0.9976 0.9912 0.9987 0.99 0.991 0.9933 0.9933 0.9899 0.9912 0.9912 0.9976 0.994 0.9947 0.9954 0.993 0.9954 0.9963 0.992 0.9926 0.995 0.9983 0.992 0.9968 0.9905 0.9904 0.9926 0.9968 0.9928 0.9949 0.9909 0.9937 0.9914 0.9905 0.9904 0.9924 0.9924 0.9965 0.9965 0.9993 0.9965 0.9908 0.992 0.9978 0.9978 0.9978 0.9978 0.9912 0.9928 0.9928 0.993 0.9993 0.9965 0.9937 0.9913 0.9934 0.9952 0.9983 0.9957 0.9957 0.9916 0.9999 0.9999 0.9936 0.9972 0.9933 0.9934 0.9931 0.9976 0.9937 0.9937 0.991 0.9979 0.9971 0.9969 0.9968 0.9961 0.993 0.9973 0.9944 0.9986 0.9986 0.9986 0.9986 0.9972 0.9917 0.992 0.9932 0.9936 0.9915 0.9922 0.9934 0.9952 0.9972 0.9934 0.9958 0.9944 0.9908 0.9958 0.9925 0.9966 0.9972 0.9912 0.995 0.9928 0.9968 0.9955 0.9981 0.991 0.991 0.991 0.992 0.9931 0.997 0.9948 0.9923 0.9976 0.9938 0.9984 0.9972 0.9922 0.9935 0.9944 0.9942 0.9944 0.9997 0.9977 0.9912 0.9982 0.9982 0.9983 0.998 0.9894 0.9927 0.9917 0.9904 0.993 0.9941 0.9943 0.99855 0.99345 0.998 0.9916 0.9916 0.99475 0.99325 0.9933 0.9969 1.0002 0.9933 0.9937 0.99685 0.99455 0.9917 0.99035 0.9914 0.99225 0.99155 0.9954 0.99455 0.9924 0.99695 0.99655 0.9934 0.998 0.9971 0.9948 0.998 0.9971 0.99215 0.9948 0.9915 0.99115 0.9932 0.9977 0.99535 0.99165 0.9953 0.9928 0.9958 0.9928 0.9928 0.9964 0.9987 0.9953 0.9932 0.9907 0.99755 0.99935 0.9932 0.9932 0.9958 0.99585 1.00055 0.9985 0.99505 0.992 0.9988 0.99175 0.9962 0.9962 0.9942 0.9927 0.9927 0.99985 0.997 0.9918 0.99215 0.99865 0.9992 1.0006 0.99135 0.99715 0.9992 1.0006 0.99865 0.99815 0.99815 0.99815 0.9949 0.99815 0.99815 0.99225 0.99445 0.99225 0.99335 0.99625 0.9971 0.9983 0.99445 0.99085 0.9977 0.9953 0.99775 0.99795 0.99505 0.9977 0.9975 0.99745 0.9976 0.99775 0.9953 0.9932 0.99405 1 0.99785 0.9939 0.9939 0.99675 0.9939 0.99675 0.98965 0.9971 0.99445 0.9945 0.9939 0.9958 0.9956 0.99055 0.9959 0.9925 0.9963 0.9935 0.99105 0.99045 0.9963 0.99155 0.99085 0.99085 0.99085 0.9924 0.9924 0.99975 0.99975 0.99315 0.9917 0.9917 0.99845 0.9921 0.99975 0.9909 0.99315 0.99855 0.9934 0.9978 0.9934 0.9949 0.99855 0.9986 0.99725 0.9946 0.99255 0.9996 0.9939 0.99 0.9937 0.9886 0.9934 1 0.9994 0.9926 0.9956 0.9978 0.9915 0.9939 0.9932 0.993 0.9898 0.9921 0.9932 0.9919 0.993 0.9953 0.9928 0.9928 0.9976 0.9906 0.9918 0.99185 0.9918 0.99185 0.994 0.9908 0.9928 0.9896 0.9908 0.9918 0.9952 0.9923 0.9915 0.9952 0.9947 0.9983 0.9975 0.995 0.9944 0.994 0.9944 0.9908 0.99795 0.9985 0.99425 0.99425 0.9943 0.9924 0.9946 0.9924 0.995 0.9919 0.99 0.9923 0.9956 0.9978 0.9978 0.9967 0.9934 0.9936 0.9932 0.9934 0.998 0.9978 0.9929 0.9974 0.99685 0.99495 0.99745 0.99505 0.992 0.9978 0.9956 0.9982 0.99485 0.9971 0.99265 0.9904 0.9965 0.9946 0.99965 0.9935 0.996 0.9942 0.9936 0.9965 0.9928 0.9928 0.9965 0.9936 0.9938 0.9926 0.9926 0.9983 0.9983 0.992 0.9983 0.9923 0.9972 0.9928 0.9928 0.9994 0.991 0.9906 0.9894 0.9898 0.9994 0.991 0.9925 0.9956 0.9946 0.9966 0.9951 0.9927 0.9927 0.9951 0.9894 0.9907 0.9925 0.9928 0.9941 0.9941 0.9925 0.9935 0.9932 0.9944 0.9972 0.994 0.9956 0.9927 0.9924 0.9966 0.9997 0.9936 0.9936 0.9952 0.9952 0.9928 0.9911 0.993 0.9911 0.9932 0.993 0.993 0.9932 0.9932 0.9943 0.9968 0.9994 0.9926 0.9968 0.9932 0.9916 0.9946 0.9925 0.9925 0.9935 0.9962 0.9928 0.993 0.993 0.9956 0.9941 0.9972 0.9948 0.9955 0.9972 0.9972 0.9983 0.9942 0.9936 0.9956 0.9953 0.9918 0.995 0.992 0.9952 1.001 0.9924 0.9932 0.9937 0.9918 0.9934 0.991 0.9962 0.9932 0.9908 0.9962 0.9918 0.9941 0.9931 0.9981 0.9931 0.9944 0.992 0.9966 0.9956 0.9956 0.9949 1.0002 0.9942 0.9923 0.9917 0.9931 0.992 1.0002 0.9953 0.9951 0.9974 0.9904 0.9974 0.9944 1.0004 0.9952 0.9956 0.995 0.995 0.9995 0.9942 0.9977 0.992 0.992 0.9995 0.9934 1.0006 0.9982 0.9928 0.9945 0.9963 0.9906 0.9956 0.9942 0.9962 0.9894 0.995 0.9908 0.9914 0.9938 0.9977 0.9922 0.992 0.9903 0.9893 0.9952 0.9903 0.9912 0.9983 0.9937 0.9932 0.9928 0.9922 0.9976 0.9922 0.9974 0.998 0.9931 0.9911 0.9944 0.9937 0.9974 0.989 0.992 0.9928 0.9918 0.9936 0.9944 0.9988 0.994 0.9953 0.9986 0.9914 0.9934 0.996 0.9937 0.9921 0.998 0.996 0.9933 0.9933 0.9959 0.9936 0.9953 0.9938 0.9952 0.9959 0.9959 0.9937 0.992 0.9967 0.9944 0.9998 0.9998 0.9942 0.9998 0.9945 0.9998 0.9946 0.9942 0.9928 0.9946 0.9927 0.9938 0.9918 0.9945 0.9966 0.9954 0.9913 0.9931 0.9986 0.9965 0.9984 0.9952 0.9956 0.9949 0.9954 0.996 0.9931 0.992 0.9912 0.9978 0.9938 0.9914 0.9932 0.9944 0.9913 0.9948 0.998 0.9998 0.9964 0.9992 0.9948 0.9998 0.998 0.9939 0.992 0.9922 0.9955 0.9917 0.9917 0.9954 0.9986 0.9955 0.9917 0.9907 0.9922 0.9958 0.993 0.9917 0.9926 0.9959 0.9906 0.9993 0.993 0.9906 0.992 0.992 0.994 0.9959 0.9908 0.9902 0.9908 0.9943 0.9921 0.9911 0.9986 0.992 0.992 0.9943 0.9937 0.993 0.9902 0.9928 0.9896 0.998 0.9954 0.9938 0.9918 0.9896 0.9944 0.9999 0.9953 0.992 0.9925 0.9981 0.9952 0.9927 0.9927 0.9911 0.9936 0.9959 0.9946 0.9948 0.9955 0.9951 0.9952 0.9946 0.9946 0.9944 0.9938 0.9963 0.991 1.0003 0.9966 0.9993 1.0003 0.9938 0.9965 0.9938 0.9993 0.9938 1.0003 0.9966 0.9942 0.9928 0.991 0.9911 0.9977 0.9927 0.9911 0.991 0.9912 0.9907 0.9902 0.992 0.994 0.9966 0.993 0.993 0.993 0.9966 0.9942 0.9925 0.9925 0.9928 0.995 0.9939 0.9958 0.9952 1 0.9948 0.99 0.9958 0.9948 0.9949 0.997 0.9927 0.9938 0.9949 0.9953 0.997 0.9932 0.9927 0.9932 0.9955 0.9914 0.991 0.992 0.9924 0.9927 0.9911 0.9958 0.9928 0.9902 0.994 0.994 0.9972 1.0004 0.991 0.9918 0.995 0.9941 0.9956 0.9956 0.9959 0.9922 0.9931 0.9959 0.9984 0.9908 0.991 0.9928 0.9936 0.9941 0.9924 0.9917 0.9906 0.995 0.9956 0.9955 0.9907 1 0.9953 0.9911 0.9922 0.9951 0.9948 0.9906 0.994 0.9907 0.9927 0.9914 0.9958 1 0.9984 0.9941 0.9944 0.998 0.998 0.9902 0.9911 0.9929 0.993 0.9918 0.992 0.9932 0.992 0.994 0.9923 0.993 0.9956 0.9907 0.99 0.9918 0.9926 0.995 0.99 0.99 0.9946 0.9907 0.9898 0.9918 0.9986 0.9986 0.9928 0.9986 0.9979 0.994 0.9937 0.9938 0.9942 0.9944 0.993 0.9986 0.9932 0.9934 0.9928 0.9925 0.9944 0.9909 0.9932 0.9934 1.0001 0.992 0.9916 0.998 0.9919 0.9925 0.9977 0.9944 0.991 0.99 0.9917 0.9923 0.9928 0.9923 0.9928 0.9902 0.9893 0.9917 0.9982 1.0005 0.9923 0.9951 0.9956 0.998 0.9928 0.9938 0.9914 0.9955 0.9924 0.9911 0.9917 0.9917 0.9932 0.9955 0.9929 0.9955 0.9958 1.0012 0.9968 0.9911 0.9924 0.991 0.9946 0.9928 0.9946 0.9917 0.9918 0.9926 0.9931 0.9932 0.9903 0.9928 0.9929 0.9958 0.9955 0.9911 0.9938 0.9942 0.9945 0.9962 0.992 0.9927 0.9948 0.9945 0.9942 0.9952 0.9942 0.9958 0.9918 0.9932 1.0004 0.9972 0.9998 0.9918 0.9918 0.9964 0.9936 0.9931 0.9938 0.9934 0.99 0.9914 0.9904 0.994 0.9938 0.9933 0.9909 0.9942 0.9945 0.9954 0.996 0.9991 0.993 0.9942 0.9934 0.9939 0.9937 0.994 0.9926 0.9951 0.9952 0.9935 0.9938 0.9939 0.9933 0.9927 0.998 0.9997 0.9981 0.992 0.9954 0.992 0.9997 0.9981 0.9943 0.9941 0.9936 0.9996 0.9932 0.9926 0.9936 0.992 0.9936 0.9996 0.993 0.9924 0.9928 0.9926 0.9952 0.9945 0.9945 0.9903 0.9932 0.9953 0.9936 0.9912 0.9962 0.9965 0.9932 0.9967 0.9953 0.9963 0.992 0.991 0.9958 0.99 0.991 0.9958 0.9938 0.9996 0.9946 0.9974 0.9945 0.9946 0.9974 0.9957 0.9931 0.9947 0.9953 0.9931 0.9946 0.9978 0.9989 1.0004 0.9938 0.9934 0.9978 0.9956 0.9982 0.9948 0.9956 0.9982 0.9926 0.991 0.9945 0.9916 0.9953 0.9938 0.9956 0.9906 0.9956 0.9932 0.9914 0.9938 0.996 0.9906 0.98815 0.9942 0.9903 0.9906 0.9935 1.0024 0.9968 0.9906 0.9941 0.9919 0.9928 0.9958 0.9932 0.9957 0.9937 0.9982 0.9928 0.9919 0.9956 0.9957 0.9954 0.993 0.9954 0.9987 0.9956 0.9928 0.9951 0.993 0.9928 0.9926 0.9938 1.0001 0.9933 0.9952 0.9934 0.9988 0.993 0.9952 0.9948 0.9998 0.9971 0.9998 0.9962 0.9948 0.99 0.9942 0.9965 0.9912 0.9978 0.9928 1.0103 0.9956 0.9936 0.9929 0.9966 0.9964 0.996 0.9959 0.9954 0.9914 1.0103 1.0004 0.9911 0.9938 0.9927 0.9922 0.9924 0.9963 0.9936 0.9951 0.9951 0.9955 0.9961 0.9936 0.992 0.9944 0.9944 1.0008 0.9962 0.9986 0.9986 1 0.9986 0.9982 1 0.9949 0.9915 0.9951 0.9986 0.9927 0.9955 0.9952 0.9928 0.9982 0.9914 0.9927 0.9918 0.9944 0.9969 0.9955 0.9954 0.9955 0.9921 0.9934 0.9998 0.9946 0.9984 0.9924 0.9939 0.995 0.9957 0.9953 0.9912 0.9939 0.9921 0.9954 0.9933 0.9941 0.995 0.9977 0.9912 0.9945 0.9952 0.9924 0.9986 0.9953 0.9939 0.9929 0.9988 0.9906 0.9914 0.9978 0.9928 0.9948 0.9978 0.9946 0.9908 0.9954 0.9906 0.99705 0.9982 0.9932 0.9977 0.994 0.9982 0.9929 0.9924 0.9966 0.9921 0.9967 0.9934 0.9914 0.99705 0.9961 0.9967 0.9926 0.99605 0.99435 0.9948 0.9916 0.997 0.9961 0.9967 0.9961 0.9955 0.9922 0.9918 0.9955 0.9941 0.9955 0.9955 0.9924 0.9973 0.999 0.9941 0.9922 0.9922 0.9953 0.9945 0.9945 0.9957 0.9932 0.9945 0.9913 0.9909 0.9939 0.991 0.9954 0.9943 0.993 1.0002 0.9946 0.9953 0.9918 0.9936 0.9984 0.9956 0.9966 0.9942 0.9984 0.9956 0.9966 0.9974 0.9944 1.0008 0.9974 1.0008 0.9928 0.9944 0.9908 0.9917 0.9911 0.9912 0.9953 0.9932 0.9896 0.9889 0.9912 0.9926 0.9911 0.9964 0.9974 0.9944 0.9974 0.9964 0.9963 0.9948 0.9948 0.9953 0.9948 0.9953 0.9949 0.9988 0.9954 0.992 0.9984 0.9954 0.9926 0.992 0.9976 0.9972 0.991 0.998 0.9966 0.998 1.0007 0.992 0.9925 0.991 0.9934 0.9955 0.9944 0.9981 0.9968 0.9946 0.9946 0.9981 0.9946 0.997 0.9924 0.9958 0.994 0.9958 0.9984 0.9948 0.9932 0.9952 0.9924 0.9945 0.9976 0.9976 0.9938 0.9997 0.994 0.9921 0.9986 0.9987 0.9991 0.9987 0.9991 0.9991 0.9948 0.9987 0.993 0.9988 1 0.9932 0.9991 0.9989 1 1 0.9952 0.9969 0.9966 0.9966 0.9976 0.99 0.9988 0.9942 0.9984 0.9932 0.9969 0.9966 0.9933 0.9916 0.9914 0.9966 0.9958 0.9926 0.9939 0.9953 0.9906 0.9914 0.9958 0.9926 0.9991 0.9994 0.9976 0.9966 0.9953 0.9923 0.993 0.9931 0.9932 0.9926 0.9938 0.9966 0.9974 0.9924 0.9948 0.9964 0.9924 0.9966 0.9974 0.9938 0.9928 0.9959 1.0001 0.9959 1.0001 0.9968 0.9932 0.9954 0.9992 0.9932 0.9939 0.9952 0.9996 0.9966 0.9925 0.996 0.9996 0.9973 0.9937 0.9966 1.0017 0.993 0.993 0.9959 0.9958 1.0017 0.9958 0.9979 0.9941 0.997 0.9934 0.9927 0.9944 0.9927 0.9963 1.0011 1.0011 0.9959 0.9973 0.9966 0.9932 0.9984 0.999 0.999 0.999 0.999 0.999 1.0006 0.9937 0.9954 0.997 0.9912 0.9939 0.999 0.9957 0.9926 0.9994 1.0004 0.9994 1.0004 1.0004 1.0002 0.9922 0.9922 0.9934 0.9926 0.9941 0.9994 1.0004 0.9924 0.9948 0.9935 0.9918 0.9948 0.9924 0.9979 0.993 0.994 0.991 0.993 0.9922 0.9979 0.9937 0.9928 0.9965 0.9928 0.9991 0.9948 0.9925 0.9958 0.9962 0.9965 0.9951 0.9944 0.9916 0.9987 0.9928 0.9926 0.9934 0.9944 0.9949 0.9926 0.997 0.9949 0.9948 0.992 0.9964 0.9926 0.9982 0.9955 0.9955 0.9958 0.9997 1.0001 1.0001 0.9918 0.9918 0.9931 1.0001 0.9926 0.9966 0.9932 0.9969 0.9925 0.9914 0.996 0.9952 0.9934 0.9939 0.9939 0.9906 0.9901 0.9948 0.995 0.9953 0.9953 0.9952 0.996 0.9948 0.9951 0.9931 0.9962 0.9948 0.9959 0.9962 0.9958 0.9948 0.9948 0.994 0.9942 0.9942 0.9948 0.9964 0.9958 0.9932 0.9986 0.9986 0.9988 0.9953 0.9983 1 0.9951 0.9983 0.9906 0.9981 0.9936 0.9951 0.9953 1.0005 0.9972 1 0.9969 1.0001 1.0001 1.0001 0.9934 0.9969 1.0001 0.9902 0.993 0.9914 0.9941 0.9967 0.9918 0.998 0.9967 0.9918 0.9957 0.9986 0.9958 0.9948 0.9918 0.9923 0.9998 0.9998 0.9914 0.9939 0.9966 0.995 0.9966 0.994 0.9972 0.9998 0.9998 0.9982 0.9924 0.9972 0.997 0.9954 0.9962 0.9972 0.9921 0.9905 0.9998 0.993 0.9941 0.9994 0.9962 0.992 0.9922 0.994 0.9897 0.9954 0.99 0.9948 0.9922 0.998 0.9944 0.9944 0.9986 0.9986 0.9986 0.9986 0.9986 0.996 0.9999 0.9986 0.9986 0.996 0.9951 0.9999 0.993 0.9982 0.992 0.9963 0.995 0.9956 0.997 0.9936 0.9935 0.9963 0.9967 0.9912 0.9981 0.9966 0.9967 0.9963 0.9935 0.9902 0.99 0.996 0.9966 0.9962 0.994 0.996 0.994 0.9944 0.9974 0.996 0.9922 0.9917 0.9918 0.9936 0.9938 0.9918 0.9939 0.9917 0.9981 0.9941 0.9928 0.9952 0.9898 0.9914 0.9981 0.9957 0.998 0.9957 0.9986 0.9983 0.9982 0.997 0.9947 0.997 0.9947 0.99416 0.99516 0.99496 0.9974 0.99579 0.9983 0.99471 0.9974 0.99644 0.99579 0.99699 0.99758 0.9977 0.99397 0.9983 0.99471 0.99243 0.9962 1.00182 0.99384 0.99582 0.9962 0.9924 0.99466 0.99212 0.99449 0.99748 0.99449 0.99748 0.99475 0.99189 0.99827 0.99752 0.99827 0.99479 0.99752 0.99642 1.00047 0.99382 0.99784 0.99486 0.99537 0.99382 0.99838 0.99566 0.99268 0.99566 0.99468 0.9933 0.99307 0.99907 0.99907 0.99907 0.99907 0.99471 0.99471 0.99907 0.99148 0.99383 0.99365 0.99272 0.99148 0.99235 0.99508 0.9946 0.99674 0.99018 0.99235 0.99084 0.99856 0.99591 0.9975 0.9944 0.99173 0.99378 0.99805 0.99534 0.99232 0.99805 0.99078 0.99534 0.99061 0.99182 0.9966 0.9912 0.99779 0.99814 0.99096 0.99379 0.99426 0.99228 0.99335 0.99595 0.99297 0.99687 0.99297 0.99687 0.99445 0.9986 0.99154 0.9981 0.98993 1.00241 0.99716 0.99437 0.9972 0.99756 0.99509 0.99572 0.99756 0.99175 0.99254 0.99509 0.99676 0.9979 0.99194 0.99077 0.99782 0.99942 0.99708 0.99353 0.99256 0.99199 0.9918 0.99354 0.99244 0.99831 0.99396 0.99724 0.99524 0.9927 0.99802 0.99512 0.99438 0.99679 0.99652 0.99698 0.99474 0.99511 0.99582 0.99125 0.99256 0.9911 0.99168 0.9911 0.99556 1.00098 0.99516 0.99516 0.99518 0.99347 0.9929 0.99347 0.99841 0.99362 0.99361 0.9914 0.99114 0.9925 0.99453 0.9938 0.9938 0.99806 0.9961 1.00016 0.9916 0.99116 0.99319 0.99517 0.99514 0.99566 0.99166 0.99587 0.99558 0.99117 0.99399 0.99741 0.99405 0.99622 1.00051 0.99803 0.99405 0.99773 0.99397 0.99622 0.99713 0.99274 1.00118 0.99176 0.9969 0.99771 0.99411 0.99771 0.99411 0.99194 0.99558 0.99194 0.99558 0.99577 0.99564 0.99578 0.99888 1.00014 0.99441 0.99594 0.99437 0.99594 0.9979 0.99434 0.99203 0.998 0.99316 0.998 0.99314 0.99316 0.99612 0.99295 0.99394 0.99642 0.99642 0.99248 0.99268 0.99954 0.99692 0.99592 0.99592 0.99692 0.99822 0.99822 0.99402 0.99404 0.99787 0.99347 0.99838 0.99839 0.99375 0.99155 0.9936 0.99434 0.9922 0.99571 0.99658 0.99076 0.99496 0.9937 0.99076 0.99542 0.99825 0.99289 0.99432 0.99523 0.99542 0.9959 0.99543 0.99662 0.99088 0.99088 0.99922 0.9966 0.99466 0.99922 0.99836 0.99836 0.99238 0.99645 1 1 0.99376 1 0.99513 0.99556 0.99556 0.99543 0.99886 0.99526 0.99166 0.99691 0.99732 0.99573 0.99656 0.99112 0.99214 0.99165 0.99004 0.99463 0.99683 0.99004 0.99596 0.99898 0.99114 0.99508 0.99306 0.99898 0.99508 0.99114 0.99342 0.99345 0.99772 0.99239 0.99502 0.99502 0.99479 0.99207 0.99497 0.99828 0.99542 0.99542 0.99228 0.99706 0.99497 0.99669 0.99828 0.99269 0.99196 0.99662 0.99475 0.99544 0.99944 0.99475 0.99544 0.9966 0.99066 0.9907 0.99066 0.998 0.9907 0.99066 0.99307 0.99106 0.99696 0.99106 0.99307 0.99167 0.99902 0.98992 0.99182 0.99556 0.99582 0.99182 0.98972 0.99352 0.9946 0.99273 0.99628 0.99582 0.99553 0.98914 0.99354 0.99976 0.99808 0.99808 0.99808 0.99808 0.99808 0.99808 0.9919 0.99808 0.99499 0.99655 0.99615 0.99296 0.99482 0.99079 0.99366 0.99434 0.98958 0.99434 0.99938 0.99059 0.99835 0.98958 0.99159 0.99159 0.98931 0.9938 0.99558 0.99563 0.98931 0.99691 0.9959 0.99159 0.99628 0.99076 0.99678 0.99678 0.99678 0.99089 0.99537 1.0002 0.99628 0.99089 0.99678 0.99076 0.99332 0.99316 0.99272 0.99636 0.99202 0.99148 0.99064 0.99884 0.99773 1.00013 0.98974 0.99773 1.00013 0.99112 0.99136 0.99132 0.99642 0.99488 0.99527 0.99578 0.99352 0.99199 0.99198 0.99756 0.99578 0.99561 0.99347 0.98936 0.99786 0.99705 0.9942 0.9948 0.99116 0.99688 0.98974 0.99542 0.99154 0.99118 0.99044 0.9914 0.9979 0.98892 0.99114 0.99188 0.99583 0.98892 0.98892 0.99704 0.9911 0.99334 0.99334 0.99094 0.99014 0.99304 0.99652 0.98944 0.99772 0.99367 0.99304 0.99183 0.99126 0.98944 0.99577 0.99772 0.99652 0.99428 0.99388 0.99208 0.99256 0.99388 0.9925 0.99904 0.99216 0.99208 0.99428 0.99165 0.99924 0.99924 0.99924 0.9956 0.99562 0.9972 0.99924 0.9958 0.99976 0.99976 0.99296 0.9957 0.9958 0.99579 0.99541 0.99976 0.99518 0.99168 0.99276 0.99085 0.99873 0.99172 0.99312 0.99276 0.9972 0.99278 0.99092 0.9962 0.99053 0.99858 0.9984 0.99335 0.99053 0.9949 0.9962 0.99092 0.99532 0.99727 0.99026 0.99668 0.99727 0.9952 0.99144 0.99144 0.99015 0.9914 0.99693 0.99035 0.99693 0.99035 0.99006 0.99126 0.98994 0.98985 0.9971 0.99882 0.99477 0.99478 0.99576 0.99578 0.99354 0.99244 0.99084 0.99612 0.99356 0.98952 0.99612 0.99084 0.99244 0.99955 0.99374 0.9892 0.99144 0.99352 0.99352 0.9935 0.99237 0.99144 0.99022 0.99032 1.03898 0.99587 0.99587 0.99587 0.99976 0.99354 0.99976 0.99552 0.99552 0.99587 0.99604 0.99584 0.98894 0.9963 0.993 0.98894 0.9963 0.99068 0.98964 0.99604 0.99584 0.9923 0.99437 0.993 0.99238 0.99801 0.99802 0.99566 0.99067 0.99066 0.9929 0.9934 0.99067 0.98912 0.99066 0.99228 0.98912 0.9958 0.99052 0.99312 0.9968 0.99502 0.99084 0.99573 0.99256 0.9959 0.99084 0.99084 0.99644 0.99526 0.9954 0.99095 0.99188 0.9909 0.99256 0.9959 0.99581 0.99132 0.98936 0.99136 0.99142 0.99232 0.99232 0.993 0.99311 0.99132 0.98993 0.99208 0.99776 0.99839 0.99574 0.99093 0.99156 0.99278 0.9924 0.98984 0.99035 0.9924 0.99165 0.9923 0.99278 0.99008 0.98964 0.99156 0.9909 0.98984 0.9889 0.99178 0.99076 0.9889 0.99046 0.98999 0.98946 0.98976 0.99046 0.99672 0.99482 0.98945 0.98883 0.99362 0.99075 0.99436 0.98988 0.99158 0.99265 0.99195 0.99168 0.9918 0.99313 0.9895 0.9932 0.99848 0.9909 0.99014 0.9952 0.99652 0.99848 0.99104 0.99772 0.9922 0.99076 0.99622 0.9902 0.99114 0.9938 0.99594 0.9902 0.99035 0.99032 0.99558 0.99622 0.99076 0.99413 0.99043 0.99043 0.98982 0.98934 0.9902 0.99449 0.99629 0.9948 0.98984 0.99326 0.99834 0.99555 0.98975 0.99216 0.99216 0.99834 0.9901 0.98975 0.99573 0.99326 0.99215 0.98993 0.99218 0.99555 0.99564 0.99564 0.99397 0.99576 0.99601 0.99564 0.99397 0.98713 0.99308 0.99308 0.99582 0.99494 0.9929 0.99471 0.9929 0.9929 0.99037 0.99304 0.99026 0.98986 0.99471 0.98951 0.99634 0.99368 0.99792 0.99026 0.99362 0.98919 0.99835 0.99835 0.99038 0.99104 0.99038 0.99286 0.99296 0.99835 0.9954 0.9914 0.99286 0.99604 0.99604 0.99119 0.99007 0.99507 0.99596 0.99011 0.99184 0.99469 0.99469 0.99406 0.99305 0.99096 0.98956 0.9921 0.99496 0.99406 0.99406 0.9888 0.98942 0.99082 0.98802 17.3 1.4 1.3 1.6 5.25 2.4 14.6 11.8 1.5 1.8 7.7 2 1.8 1.4 16.7 8.1 8 4.7 8.1 2.1 16.7 6.4 1.5 7.6 1.5 12.4 1.3 1.7 8.1 7.1 7.6 2.3 6.5 1.4 12.7 1.6 1.1 1.2 6.5 4.6 0.6 10.6 4.6 4.8 2.7 12.6 0.6 9.2 6.6 7 8.45 11.1 18.15 18.15 4.1 4.1 4.6 18.15 4.9 8.3 1.4 11.5 1.8 1.6 2.4 4.9 1.8 4.3 4.4 1.4 1.6 1.3 5.2 5.6 5.3 4.9 2.4 1.6 2.1 1.4 7.1 1.6 10.7 11.1 10.7 1.6 1.6 1.5 1.5 1.6 1.6 8 7.7 2.7 15.1 15.1 8.9 6 12.3 13.1 6.7 12.3 2.3 11.1 1.5 6.7 6 15.2 10.2 13.1 10.7 17.1 17.1 17.1 1.9 10.7 17.1 1.2 1.2 3.1 1.5 10.7 4.9 12.6 10.7 4.9 12.15 12 1.7 2.6 1.4 1.9 16.9 16.9 2.1 7 7.1 5.9 7.1 8.7 13.2 15.3 15.3 13.2 2.7 10.65 10 6.8 15.6 13.2 5.1 3 15.3 2.1 1.9 8.6 8.75 3.6 4.7 1.3 1.8 9.7 4 2.4 4.7 18.8 1.8 1.8 12.8 12.8 12.8 12.8 12.8 7.8 16.75 12.8 12.8 7.8 5.4 16.75 1.3 10.1 3.8 10.9 6.6 9.8 11.7 1.2 1.4 9.6 12.2 2.6 10.7 4.9 12.2 9.6 1.4 1.1 1 8.2 11.3 7.3 2.3 8.2 2.1 2 10 15.75 3.9 2 1.5 1.6 1.4 1.5 1.4 2 13.8 1.3 3.8 6.9 2.2 1.6 13.8 10.8 12.8 10.8 15.3 12.1 12 11.6 9.2 11.6 9.2 2.8 1.6 6.1 8.5 7.8 14.9 6.2 8.5 8.2 7.8 10.6 11.2 11.6 7.1 14.9 6.2 1.7 7.7 17.3 1.4 7.7 7.7 3.4 1.6 1.4 1.4 10.4 1.4 10.4 4.1 2.8 15.7 10.9 15.7 6.5 10.9 5.9 17.3 1.4 13.5 8.5 6.2 1.4 14.95 7.7 1.3 7.7 1.3 1.3 1.3 15.6 15.6 15.6 15.6 4.9 5 15.6 6.5 1.4 2.7 1.2 6.5 6.4 6.9 7.2 10.6 3.5 6.4 2.3 12.05 7 11.8 1.4 5 2.2 14.6 1.6 1.3 14.6 2.8 1.6 3.3 6.3 8.1 1.6 10.6 11.8 1.7 8.1 1.4 1.3 1.8 7.2 1.1 11.95 1.1 11.95 2.2 12.7 1.4 10.6 1.9 17.8 10.2 4.8 9.8 8.4 7.2 4.8 8.4 4.5 1.4 7.2 11 11.1 2.6 2 10.1 13.3 11.4 1.3 1.4 1.4 7 2 1.2 12.9 5 10.1 3.75 1.7 12.6 1.3 1.6 7.6 8.1 14.9 6 6 7.2 3 1.2 2 4.9 2 8.9 16.45 2 1.9 5.1 4.4 5.8 4.4 12.9 1.3 1.3 1.2 2.7 1.7 8.2 1.5 1.5 12.9 3.9 17.75 4.9 1.6 1.4 2 2 8.2 2.1 1.8 8.5 4.45 5.8 13 2.7 7.3 19.1 8.8 2.7 7.4 2.3 6.85 11.4 0.9 19.35 7.9 11.75 7.7 3 7.7 3 1.5 7.5 1.5 7.5 8.3 7.05 8.4 13.9 17.5 5.6 9.4 4.8 9.4 9.7 6.3 1.6 14.6 2.5 14.6 2.6 2.5 8.2 1.5 2.3 10 10 1.6 1.6 16 10.4 7.4 7.4 10.4 16.05 16.05 2.6 2.5 10.8 1.2 12.1 11.95 1.7 0.8 1.4 1.3 6.3 10.3 15.55 1.5 1.5 1.4 1.5 7.9 13 1 4.85 7.1 7.9 7.5 7.6 10.3 1.7 1.7 19.95 7.7 5.3 19.95 12.7 12.7 1.5 11.3 18.1 18.1 7 18.1 6.4 1.4 1.4 3.1 14.1 7.7 5.2 11.6 10.4 7.5 11.2 0.8 1.4 4.7 3.1 4 11.3 3.1 8.1 14.8 1.4 8.1 3.5 14.8 8.1 1.4 1.5 1.5 12.8 1.6 7.1 7.1 11.2 1.7 6.7 17.3 8.6 8.6 1.5 12.1 6.7 10.7 17.3 1.8 1.4 7.5 4.8 7.1 16.9 4.8 7.1 11.3 1.1 1.2 1.1 12.9 1.2 1.1 1.2 2.3 10 2.3 1.2 1.4 14.9 1.8 1.8 7 8.6 1.8 1.1 1.3 4.9 1.9 10.4 10 8.6 1.7 1.7 18.95 12.8 12.8 12.8 12.8 12.8 12.8 0.7 12.8 1.4 13.3 8.5 1.5 11.7 5 1.2 2.1 1.4 2.1 16 1.1 15.3 1.4 2.8 2.8 0.9 2.5 8.1 8.2 0.9 11.1 7.8 2.8 10.1 3.2 14.2 14.2 14.2 2.9 6 20.4 10.1 2.9 14.2 3.2 0.95 1.7 1.7 9 1.3 1.4 2.4 16 11.4 14.35 2.1 11.4 14.35 1.1 1.1 1.2 15.8 5.2 5.2 9.6 5.2 1.2 0.8 14.45 9.6 6.9 3.4 2.3 11 5.95 5.1 5.4 1.2 12.6 1 6.6 1.5 1 1.1 6.6 8.2 2 1.4 2 7.5 2 2 13.3 2.85 5.6 5.6 1 3.2 1 7.1 2.4 11.2 9.5 1 1.8 2.6 2.4 8 11.2 7.1 3.3 10.3 1.2 1.6 10.3 9.65 16.4 1.5 1.2 3.3 5 16.3 16.3 16.3 6.5 6.4 10.2 16.3 7.4 13.7 13.7 1.3 7.4 7.4 7.45 7.2 13.7 10.4 1.1 6.5 4.6 13.9 5.2 1.7 6.5 16.4 3.6 1.5 12.4 1.7 6.2 6.2 2.6 1.7 9.3 12.4 1.5 9.1 12 4.8 12.3 12 2.7 3.6 3.6 4.3 1.8 11.8 1.8 11.8 1.8 1.4 6.6 1.55 0.7 6.4 11.8 4.3 5.1 5.8 5.9 1.3 1.4 1.2 7.4 10.8 1.8 7.4 1.2 1.4 14.4 1.7 3.6 3.6 10.05 10.05 10.5 1.9 3.6 1.65 1.9 65.8 6.85 7.4 7.4 20.2 11 20.2 6.2 6.2 6.85 8 8.2 2.2 10.1 7.2 2.2 10.1 1.6 1.3 8 8.2 5.3 14 7.2 1.6 11.8 9.6 6.1 2.7 3.6 1.7 1.6 2.7 1 0.9 1.6 1 10.6 2 1.2 6.2 9.2 5 6.3 3.3 8 1.2 1.2 16.2 11.6 7.2 1.1 3.4 1.4 3.3 8 9.3 2.3 0.9 3.5 1.7 1.3 1.3 5.6 7.4 2.3 1 1.5 10 14.9 9.3 1 1 5.9 5 1.25 3.9 5 0.8 1 5.9 1.6 1.3 1 1.1 1.25 1.4 1.2 5 1.4 1.7 1.8 1.6 1.5 1.7 13.9 5.9 2.1 1.1 6.7 2.7 6.7 3.95 7.75 10.6 1.6 2.5 0.7 11.1 5.15 4.7 9.7 1.7 1.4 2 7.5 9.7 0.8 13.1 1.1 2.2 8.9 1.1 0.9 1.7 6.9 1.1 1 1 7.6 8.9 2.2 1.2 1 1 3.1 1.95 2.2 8.75 11.9 2.7 5.45 6.3 14.4 7.8 1.6 9.1 9.1 14.4 1.3 1.6 11.3 6.3 0.7 1.25 0.7 7.8 10.3 10.3 7.8 8.7 8.3 10.3 7.8 1.2 8.3 8.3 6.2 5 1.8 1.6 1.8 1.8 2.9 6 0.9 1.1 1.6 5.45 14.05 8 13.1 4.9 1.3 2.2 14.9 14.9 0.95 1.4 0.95 1.7 5.6 14.9 7.1 1.2 9.6 11.4 11.4 7.9 5 11.1 8 3.8 10.55 10.2 10.2 9.8 6.3 1.1 4.5 6.3 10.9 9.8 9.8 0.8 0.8 1.2 1.3 9.8 10.2 10.9 6.3 6.3 1.2 0.9 1.1 4.5 3.7 18.1 1.35 5.5 3.1 12.85 19.8 8.25 12.85 3.8 6.9 8.25 11.7 4.6 4 19.8 12.85 1.2 8.9 11.7 6.2 14.8 14.8 10.8 1.6 8.3 8.4 2.5 3.5 17.2 2.1 12.2 11.8 16.8 17.2 1.1 14.7 5.5 6.1 1.2 1.3 8.7 1.7 8.7 10.2 4.5 5.9 1.7 1.4 5.4 7.9 1.1 7 7 7.6 7 12.3 15.3 12.3 1.2 2.3 6.1 7.6 10.2 4.1 2.9 8.5 1.5 3.1 7.9 3.5 4.9 1.1 7 1.2 4.5 2.6 9.9 4.5 9.5 1.5 3.2 2.6 11.2 3.2 2.3 4.9 4.9 1.4 1.5 6.7 2.1 4.3 10.9 7 2.3 2.5 2.6 3.2 2.5 14.7 4.5 2.2 1.9 1.6 17.3 4.2 4.2 2.5 1.9 1.4 0.8 8 1.6 1.7 5.5 17.3 8.6 6.9 2.1 2.2 1.5 2.5 17.6 4.2 2.9 4.8 11.9 0.9 1.3 6.4 4.3 11.9 8.1 1.3 0.9 17.2 17.2 17.2 8.7 17.2 8.7 7.5 17.2 4.6 3.7 2.2 7.4 15.1 7.4 4.8 7.9 1 15.1 7.4 4.8 4.6 1.4 6.2 6.1 5.1 6.3 0.9 2.3 6.6 7.5 8.6 11.9 2.3 7.1 4.3 1.1 1 7.9 1 1 1 7.3 1.7 1.3 6.4 1.8 1.5 3.8 7.9 1 1.2 5.3 9.1 6.5 9.1 6.3 5.1 6.5 2.4 9.1 7.5 5 6.75 1.2 1.6 16.05 5 12.4 0.95 4.6 1.7 1 1.3 5 2.5 2.6 2.1 12.75 1.1 12.4 3.7 2.65 2.5 8.2 7.3 1.1 6.6 7 14.5 11.8 3 3.7 6 4.6 2.5 3.3 1 1.1 1.4 3.3 8.55 2.5 6.7 3.8 4.5 4.6 4.2 11.3 5.5 4.2 2.2 14.5 14.5 14.5 14.5 14.5 14.5 1.5 18.75 3.6 1.4 5.1 10.5 2 2.6 9.2 1.8 5.7 2.4 1.9 1.4 0.9 4.6 1.4 9.2 1.4 1.8 2.3 2.3 4.4 6.4 2.9 2.8 2.9 4.4 8.2 1 2.9 7 1.8 1.5 7 8.2 7.6 2.3 8.7 1 2.9 6.7 5 1.9 2 1.9 8.5 12.6 5.2 2.1 1.1 1.3 1.1 9.2 1.2 1.1 8.3 1.8 1.4 15.7 4.35 1.8 1.6 2 5 1.8 1.3 1 1.4 8.1 8.6 3.7 5.7 2.35 13.65 13.65 13.65 15.2 4.6 1.2 4.6 6.65 13.55 13.65 9.8 10.3 6.7 15.2 9.9 7.2 1.1 8.3 11.25 12.8 9.65 12.6 12.2 8.3 11.25 1.3 9.9 7.2 1.1 1.1 4.8 1.1 1.4 1.7 10.6 1.4 1.1 5.55 2.1 1.7 9 1.7 1.8 4.7 11.3 3.6 6.9 3.6 4.9 6.95 1.9 4.7 11.3 1.8 11.3 8.2 8.3 9.55 8.4 7.8 7.8 10.2 5.5 7.8 7.4 3.3 5 3.3 5 1.3 1.2 7.4 7.8 9.9 0.7 4.6 5.6 9.5 14.8 4.6 2.1 11.6 1.2 11.6 2.1 20.15 4.7 4.3 14.5 4.9 14.55 14.55 10.05 4.9 14.5 14.55 15.25 3.15 1.3 5.2 1.1 7.1 8.8 18.5 8.8 1.4 1.2 5 1.6 18.75 6 9.4 9.7 4.75 6 5.35 5.35 6.8 6.9 1.4 0.9 1.2 1.3 2.6 12 9.85 3.85 2 1.6 7.8 1.9 2 10.3 1.1 12 3.85 9.85 2 4 1.1 10.4 6.1 1.8 10.4 4.7 4 1.1 6.4 8.15 6.1 4.8 1.2 1.1 1.4 7.4 1.8 1 15.5 15.5 8.4 2.4 3.95 19.95 2 3 15.5 8.4 14.3 4.2 1.4 3 4.9 2.4 14.3 10.7 11 1.4 1.2 12.9 10.8 1.3 2 1.8 1.2 7.5 9.7 3.8 7.2 9.7 6.3 6.3 0.8 8.6 6.3 3.1 7.2 7.1 6.4 14.7 7.2 7.1 1.9 1.2 4.8 1.2 3.4 4.3 8.5 1.8 1.8 19.5 8.5 19.9 8.3 1.8 1.1 16.65 16.65 16.65 0.9 6.1 10.2 0.9 16.65 3.85 4.4 4.5 3.2 4.5 4.4 9.7 4.2 4.2 1.1 9.7 4.2 5.6 4.2 1.6 1.6 1.1 14.6 2.6 1.2 7.25 6.55 7 1.5 1.4 7.25 1 4.2 17.5 17.5 17.5 1.5 1.3 3.9 4.2 7.6 1 1.1 11.8 1.4 9.7 12.9 1.6 7.2 7.1 1.9 8.8 7.2 1.4 14.3 14.3 8.8 1.4 1.8 14.3 7.2 1.2 11.8 0.9 12.6 26.05 4.7 12.6 1.2 26.05 6.1 11.8 0.9 5.6 5.3 5.7 8 8 17.6 8 8.8 1.5 1.4 4.8 2.4 3.7 4.9 5.7 5.7 4.9 2 5.1 4.5 3.2 6.65 1.6 4 17.75 1.4 17.75 7.2 5.7 8.5 11.4 5.4 2.7 4.3 1.2 1.8 1.3 5.7 2.7 11.7 4.3 11 1.6 11.6 6.2 1.8 1.2 1 2.4 1.2 8.2 18.8 9.6 12.9 9.2 1.2 12.9 8 12.9 1.6 12 2.5 9.2 4.4 8.8 9.6 8 18.8 1.3 1.2 12.9 1.2 1.6 1.5 18.15 13.1 13.1 13.1 13.1 1 1.6 11.8 1.4 1 13.1 10.6 10.4 1.1 7.4 1.2 3.4 18.15 8 2.5 2 2 6.9 1.2 9.4 2.9 6.9 5.4 1.3 20.8 10.3 1.3 1.6 13.1 1.8 8 1.6 1.4 14.7 14.7 14.7 14.7 14.7 14.7 14.7 1.8 10.6 12.5 6.8 14.7 2.9 1.4 1.4 2.1 7.4 2.9 1.4 1.4 7.4 5 2.5 6.1 2.7 2.1 12.9 12.9 12.9 13.7 12.9 2.4 9.8 13.7 1.3 12.1 6.1 7.7 6.1 1.4 7.7 12.1 6.8 9.2 8.3 17.4 2.7 12.8 8.2 8.1 8.2 8.3 8 11.8 12 1.7 17.4 13.9 10.7 2 2.2 1.3 1.1 2 6.4 1.3 1.1 10.7 6.4 6.3 6.4 15.1 2 2 2.2 12.1 8.8 8.8 5.1 6.8 6.8 3.7 12.2 5.7 8.1 2.5 4 6.8 1 5.1 5.8 10.6 3.5 3.5 16.4 4.8 3.3 1.2 1.2 4.8 3.3 2.5 8.7 1.6 4 2.5 16.2 9 16.2 1.4 7 9 3.1 1.5 4.6 4.8 4.6 1.5 2.7 6.3 7.2 7.2 12.4 6.6 6.6 4 4.8 1.3 7.2 11.1 12.4 9.8 6.6 13.3 11.7 8 1.6 16.55 1.5 10.2 6.6 17.8 17.8 1.5 7.4 17.8 2 7.4 2 17.8 12.1 8.2 1.5 8.7 3.5 6.4 2.1 7.7 12.3 1.3 8.7 3.5 1.1 2.8 3.5 1.9 3.8 3.8 2.4 4.8 4.8 6.2 1.3 3.8 1.5 4.8 1.9 6.2 7.9 1.6 1.4 2.6 14.8 2.4 0.9 0.9 1.2 9.9 3.9 15.6 15.6 1.5 1.6 7.8 5.6 1.3 16.7 7.95 6.7 1.1 6.3 8.9 1 1.5 6.6 6.2 6.3 2.1 2.2 5.4 8.9 1 17.9 2.6 1.3 17.9 2.6 2.3 4.3 7.1 7.1 11.9 11.7 5.8 3.8 12.4 6.5 7.1 7.6 7.9 2.8 10.6 2.8 1.5 7.6 7.9 1.7 7.6 7.5 1.7 1.7 12.1 4.5 1.7 8 7.6 8.6 8.6 14.6 1.6 8.6 14.6 1.1 3.7 8.9 8.9 4.7 8.9 3.1 5.8 5.8 5.8 1 15.8 1.5 5.2 1.5 2.5 1 15.8 5.9 3.1 3.1 5.8 11.5 18 4.8 8.5 1.6 18 4.8 5.9 1.1 8.5 13.1 4.1 2.9 13.1 1.1 1.5 7.75 1.15 1 17.8 5.7 17.8 7.4 1.4 1.4 1 4.4 1.6 7.9 15.5 15.5 15.5 15.5 17.55 13.5 13.5 1.3 15.5 11.6 7.9 15.5 17.55 11.6 13.15 1.9 13.5 1.3 6.1 6.1 1.9 1.9 1.6 11.3 8.4 8.3 8.4 12.2 8 1.3 12.7 1.3 10.5 12.5 9.6 1.5 1.5 7.8 10.8 12.5 8.6 1.2 14.5 3.7 1.1 1.1 3.8 4.6 10.2 7.9 2.4 10.7 4.9 10.7 1.1 7.9 5.6 2.4 14.2 9.5 9.5 4.1 4.7 1.4 0.9 20.3 3.5 2.7 1.2 1.2 2 1.1 1.5 1.2 18.1 18.1 3.6 3.5 12.1 17.45 12.1 3 1.6 5.7 5.6 6.8 15.6 6 1.8 8.6 8.6 11.5 7.8 2.4 5 8.6 1.5 5.4 11.9 11.9 9 10 11.9 11.9 15.5 5.4 15 1.4 9.4 3.7 15 1.4 6.5 1.4 6.3 13.7 13.7 13.7 13.7 13.7 13.7 1.5 1.6 1.4 3.5 1 1.4 1.5 13.7 1.6 5.2 1.4 11.9 2.4 3.2 1.7 4.2 15.4 13 5.6 9.7 2.5 4 15.4 1.2 2 1.2 5.1 1.4 1.2 6.5 1.3 6.5 2.7 1.3 7.4 12.9 1.3 1.2 2.6 2.3 1.3 10.5 2.6 14.4 1.2 3.1 1.7 6 11.8 6.2 1.4 12.1 12.1 12.1 3.9 4.6 12.1 1.2 8.1 3.9 1.1 6.5 10.1 10.7 3.2 12.4 5.2 5 2.5 9.2 6.9 2 15 15 1.2 15 1.8 10.8 3.9 4.2 2 13.5 13.3 2.2 1.4 1.6 2.2 14.8 1.8 14.8 1.3 9.9 5.1 5.1 1.5 1.5 11.1 5.25 2.3 7.9 8 1.4 5.25 2.3 2.3 3.5 13.7 9.9 15.4 16 16 16 16 2.4 5.5 2.3 16.8 16 17.8 17.8 6.8 6.8 6.8 6.8 1.6 4.7 11.8 17.8 15.7 5.8 15.7 9 15.7 5.8 8.8 10.2 6.6 6.5 8.9 11.1 4.2 1.6 7.4 11.5 1.6 2 4.8 9.8 1.9 4.2 1.6 7.3 5.4 10.4 1.9 7.3 5.4 7.7 11.5 1.2 2.2 1 8.2 8.3 8.2 9.3 8.1 8.2 8.3 13.9 13.9 13.9 13.9 13.9 13.9 13.9 2 13.9 15.7 1.2 1.5 1.2 3.2 1.2 2.6 13.2 10.4 5.7 2.5 1.6 1.4 7.4 2.5 5.6 3.6 7.5 5.8 1.6 1.5 2.9 11.2 9.65 10.1 3.2 11.2 11.45 9.65 4.5 2.7 3.5 1.7 2.1 4.8 5 2.6 6.6 5 7.3 5 1.7 2.6 8.2 8.2 5 1.2 7.1 9.5 15.8 15.5 15.8 17.05 12.7 12.3 11.8 11.8 11.8 12.3 11.8 13.6 5.2 6.2 7.9 7.9 3.3 2.8 7.9 3.3 6.3 4.9 10.4 4.9 10.4 16 6.3 2.2 17.3 17.3 17.3 17.3 2.2 2.2 17.3 6.6 6.5 12.3 5 2.8 13.6 2.8 5.4 10.9 1.7 9.15 4.5 9.15 1.4 5.9 16.4 1.2 16.4 5.9 7.8 7.8 2.8 2.9 2.5 12.8 12.2 7.7 2.8 2.9 17.3 19.3 19.3 19.3 2.7 6.4 17.3 2.4 2.8 1.7 15.4 15.4 4.1 6.6 1.2 2.1 1 1.1 1.4 1.6 9.8 1.9 1.3 7.9 7.9 4.5 22.6 7.9 3.5 1.2 4.5 2 7.8 0.9 2.9 2.9 3.5 4.2 9.7 10.5 1.1 16.1 1.1 8.1 6.2 7.7 2.4 16.3 2.3 8.4 8.5 6 1.1 1.75 2.6 1.3 2.1 1.1 1.1 2.8 9 2.8 2.2 5.1 3.5 12.7 7.5 2 3.5 14.3 9.8 12.7 12.7 5.1 3.5 12.7 12.9 12.9 1.3 10.5 1.5 12.7 12.9 1.2 6.2 8.8 3.9 1.3 9.1 9.1 3.9 1.8 2.1 1.4 14.7 9.1 1.9 1.8 9.6 3.9 1.3 11.8 1.9 12 7.9 9.3 4.6 2.2 10.2 10.6 1.4 9.1 11.1 9.1 4.4 2.8 1.1 1.3 1.2 3.3 9.7 2.3 1.1 11.4 1.2 14.7 13.8 1.3 6.3 7.9 2 11.8 1.2 10 5.2 1.2 7.2 9.9 5.3 13.55 2.2 9.9 4.3 13 13.55 1 1.1 6.9 13.4 4.6 9.9 3 5.8 12.9 3.2 0.8 2.5 2.4 7.2 7.3 6.3 4.25 1.2 2 4.25 4.7 4.5 1.4 4.1 5.3 4.2 6.65 8.2 2.6 2.6 2 12.2 2.3 8.2 5 10.7 10.8 1.7 1.3 1.7 12.7 1.3 1.2 1.3 5.7 3.4 1.1 1 1 1.65 6.8 6.8 4.9 1.4 2.5 10.8 10.8 10.8 10.8 2.8 1.3 2 1.1 8.2 6 6.1 8.2 8.8 6.1 6 1.2 11.4 1.3 1.3 6.2 3.2 4.5 9.9 6.2 11.4 1.3 1.3 0.9 0.7 1 1 10.4 1.3 12.5 12.5 12.5 12.5 19.25 1.1 12.5 19.25 9 1.2 9 1.3 12.8 12.8 7.6 7.6 1.4 8.3 9 1.85 12.55 1.4 1.8 4 12.55 9 3 1.85 7.9 2.6 1.2 7.1 7.9 1.3 10.7 7.7 8.4 10.7 12.7 1.8 7.7 10.5 1.6 1.85 10.5 10.5 1 1.2 1.7 1.6 9 1.9 1.2 1.5 3.9 3.6 1.2 5 2.9 10.4 11.4 18.35 18.4 1.2 7.1 1.3 1.5 10.2 2.2 3.5 3.5 3.9 7.4 7.4 11 1.5 3.9 5.4 1.5 5 1.2 13 13 13 13 8.6 1.7 1.2 1.2 1.2 2 19.4 0.8 6.3 6.4 12.1 12.1 12.9 2.4 4.3 4.2 12.9 1.7 2.2 12.1 3.4 7.4 7.3 1.1 1.1 1.4 14.5 8 1.1 1.1 2.2 5.8 0.9 6.4 10.9 7.3 8.3 1.3 3.3 1 1.1 1 5.1 3.2 12.6 3.7 1.7 5.1 1 1.3 1.5 4.6 10.3 6.1 6.1 1.2 10.3 9.9 1.6 1.1 1.5 1.2 1.5 1.1 11.5 7.8 7.4 1.45 8.9 1.1 1 2.5 1.1 2.4 2.3 5.1 2.5 8.9 2.5 8.9 1.6 1.4 3.9 13.7 13.7 9.2 7.8 7.6 7.7 3 1.3 4 1.1 2 1.9 1.4 4.5 10.1 6.6 1.9 12.4 1.6 2.5 1.2 2.5 0.8 0.9 8.1 8.1 11.75 1.3 1.9 8.3 8.1 5.7 1.9 1.2 11.75 2.2 0.9 1.3 1.6 8 1.2 1.1 0.8
\ No newline at end of file
diff --git a/pandas/pandas/tests/reshape/merge/__init__.py b/pandas/pandas/tests/reshape/merge/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/reshape/merge/test_join.py b/pandas/pandas/tests/reshape/merge/test_join.py
new file mode 100644
index 0000000000000..65bfea0b9beea
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_join.py
@@ -0,0 +1,1128 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    bdate_range,
+    concat,
+    merge,
+    option_context,
+)
+import pandas._testing as tm
+
+
+def get_test_data(ngroups=8, n=50):
+    unique_groups = list(range(ngroups))
+    arr = np.asarray(np.tile(unique_groups, n // ngroups))
+
+    if len(arr) < n:
+        arr = np.asarray(list(arr) + unique_groups[: n - len(arr)])
+
+    np.random.default_rng(2).shuffle(arr)
+    return arr
+
+
+class TestJoin:
+    # aggregate multiple columns
+    @pytest.fixture
+    def df(self):
+        df = DataFrame(
+            {
+                "key1": get_test_data(),
+                "key2": get_test_data(),
+                "data1": np.random.default_rng(2).standard_normal(50),
+                "data2": np.random.default_rng(2).standard_normal(50),
+            }
+        )
+
+        # exclude a couple keys for fun
+        df = df[df["key2"] > 1]
+        return df
+
+    @pytest.fixture
+    def df2(self):
+        return DataFrame(
+            {
+                "key1": get_test_data(n=10),
+                "key2": get_test_data(ngroups=4, n=10),
+                "value": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+
+    @pytest.fixture
+    def target_source(self):
+        data = {
+            "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+            "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+            "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+            "D": bdate_range("1/1/2009", periods=5),
+        }
+        target = DataFrame(data, index=Index(["a", "b", "c", "d", "e"], dtype=object))
+
+        # Join on string value
+
+        source = DataFrame(
+            {"MergedA": data["A"], "MergedD": data["D"]}, index=data["C"]
+        )
+        return target, source
+
+    def test_left_outer_join(self, df, df2):
+        joined_key2 = merge(df, df2, on="key2")
+        _check_join(df, df2, joined_key2, ["key2"], how="left")
+
+        joined_both = merge(df, df2)
+        _check_join(df, df2, joined_both, ["key1", "key2"], how="left")
+
+    def test_right_outer_join(self, df, df2):
+        joined_key2 = merge(df, df2, on="key2", how="right")
+        _check_join(df, df2, joined_key2, ["key2"], how="right")
+
+        joined_both = merge(df, df2, how="right")
+        _check_join(df, df2, joined_both, ["key1", "key2"], how="right")
+
+    def test_full_outer_join(self, df, df2):
+        joined_key2 = merge(df, df2, on="key2", how="outer")
+        _check_join(df, df2, joined_key2, ["key2"], how="outer")
+
+        joined_both = merge(df, df2, how="outer")
+        _check_join(df, df2, joined_both, ["key1", "key2"], how="outer")
+
+    def test_inner_join(self, df, df2):
+        joined_key2 = merge(df, df2, on="key2", how="inner")
+        _check_join(df, df2, joined_key2, ["key2"], how="inner")
+
+        joined_both = merge(df, df2, how="inner")
+        _check_join(df, df2, joined_both, ["key1", "key2"], how="inner")
+
+    def test_handle_overlap(self, df, df2):
+        joined = merge(df, df2, on="key2", suffixes=(".foo", ".bar"))
+
+        assert "key1.foo" in joined
+        assert "key1.bar" in joined
+
+    def test_handle_overlap_arbitrary_key(self, df, df2):
+        joined = merge(
+            df,
+            df2,
+            left_on="key2",
+            right_on="key1",
+            suffixes=(".foo", ".bar"),
+        )
+        assert "key1.foo" in joined
+        assert "key2.bar" in joined
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    def test_join_on(self, target_source, infer_string):
+        target, source = target_source
+
+        merged = target.join(source, on="C")
+        tm.assert_series_equal(merged["MergedA"], target["A"], check_names=False)
+        tm.assert_series_equal(merged["MergedD"], target["D"], check_names=False)
+
+        # join with duplicates (fix regression from DataFrame/Matrix merge)
+        df = DataFrame({"key": ["a", "a", "b", "b", "c"]})
+        df2 = DataFrame({"value": [0, 1, 2]}, index=["a", "b", "c"])
+        joined = df.join(df2, on="key")
+        expected = DataFrame(
+            {"key": ["a", "a", "b", "b", "c"], "value": [0, 0, 1, 1, 2]}
+        )
+        tm.assert_frame_equal(joined, expected)
+
+        # Test when some are missing
+        df_a = DataFrame([[1], [2], [3]], index=["a", "b", "c"], columns=["one"])
+        df_b = DataFrame([["foo"], ["bar"]], index=[1, 2], columns=["two"])
+        df_c = DataFrame([[1], [2]], index=[1, 2], columns=["three"])
+        joined = df_a.join(df_b, on="one")
+        joined = joined.join(df_c, on="one")
+        assert np.isnan(joined["two"]["c"])
+        assert np.isnan(joined["three"]["c"])
+
+        # merge column not p resent
+        with pytest.raises(KeyError, match="^'E'$"):
+            target.join(source, on="E")
+
+        # overlap
+        msg = (
+            "You are trying to merge on float64 and object|str columns for key "
+            "'A'. If you wish to proceed you should use pd.concat"
+        )
+        with pytest.raises(ValueError, match=msg):
+            target.join(source, on="A")
+
+    def test_join_on_fails_with_different_right_index(self):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=3),
+                "b": np.random.default_rng(2).standard_normal(3),
+            }
+        )
+        df2 = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            },
+            index=MultiIndex.from_product([range(5), ["A", "B"]]),
+        )
+        msg = r'len\(left_on\) must equal the number of levels in the index of "right"'
+        with pytest.raises(ValueError, match=msg):
+            merge(df, df2, left_on="a", right_index=True)
+
+    def test_join_on_fails_with_different_left_index(self):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=3),
+                "b": np.random.default_rng(2).standard_normal(3),
+            },
+            index=MultiIndex.from_arrays([range(3), list("abc")]),
+        )
+        df2 = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+        msg = r'len\(right_on\) must equal the number of levels in the index of "left"'
+        with pytest.raises(ValueError, match=msg):
+            merge(df, df2, right_on="b", left_index=True)
+
+    def test_join_on_fails_with_different_column_counts(self):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=3),
+                "b": np.random.default_rng(2).standard_normal(3),
+            }
+        )
+        df2 = DataFrame(
+            {
+                "a": np.random.default_rng(2).choice(["m", "f"], size=10),
+                "b": np.random.default_rng(2).standard_normal(10),
+            },
+            index=MultiIndex.from_product([range(5), ["A", "B"]]),
+        )
+        msg = r"len\(right_on\) must equal len\(left_on\)"
+        with pytest.raises(ValueError, match=msg):
+            merge(df, df2, right_on="a", left_on=["a", "b"])
+
+    @pytest.mark.parametrize("wrong_type", [2, "str", None, np.array([0, 1])])
+    def test_join_on_fails_with_wrong_object_type(self, wrong_type):
+        # GH12081 - original issue
+
+        # GH21220 - merging of Series and DataFrame is now allowed
+        # Edited test to remove the Series object from test parameters
+
+        df = DataFrame({"a": [1, 1]})
+        msg = (
+            "Can only merge Series or DataFrame objects, "
+            f"a {type(wrong_type)} was passed"
+        )
+        with pytest.raises(TypeError, match=msg):
+            merge(wrong_type, df, left_on="a", right_on="a")
+        with pytest.raises(TypeError, match=msg):
+            merge(df, wrong_type, left_on="a", right_on="a")
+
+    def test_join_on_pass_vector(self, target_source):
+        target, source = target_source
+        expected = target.join(source, on="C")
+        expected = expected.rename(columns={"C": "key_0"})
+        expected = expected[["key_0", "A", "B", "D", "MergedA", "MergedD"]]
+
+        join_col = target.pop("C")
+        result = target.join(source, on=join_col)
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_with_len0(self, target_source):
+        # nothing to merge
+        target, source = target_source
+        merged = target.join(source.reindex([]), on="C")
+        for col in source:
+            assert col in merged
+            assert merged[col].isna().all()
+
+        merged2 = target.join(source.reindex([]), on="C", how="inner")
+        tm.assert_index_equal(merged2.columns, merged.columns)
+        assert len(merged2) == 0
+
+    def test_join_on_inner(self):
+        df = DataFrame({"key": ["a", "a", "d", "b", "b", "c"]})
+        df2 = DataFrame({"value": [0, 1]}, index=["a", "b"])
+
+        joined = df.join(df2, on="key", how="inner")
+
+        expected = df.join(df2, on="key")
+        expected = expected[expected["value"].notna()]
+        tm.assert_series_equal(joined["key"], expected["key"])
+        tm.assert_series_equal(joined["value"], expected["value"], check_dtype=False)
+        tm.assert_index_equal(joined.index, expected.index)
+
+    def test_join_on_singlekey_list(self):
+        df = DataFrame({"key": ["a", "a", "b", "b", "c"]})
+        df2 = DataFrame({"value": [0, 1, 2]}, index=["a", "b", "c"])
+
+        # corner cases
+        joined = df.join(df2, on=["key"])
+        expected = df.join(df2, on="key")
+
+        tm.assert_frame_equal(joined, expected)
+
+    def test_join_on_series(self, target_source):
+        target, source = target_source
+        result = target.join(source["MergedA"], on="C")
+        expected = target.join(source[["MergedA"]], on="C")
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_on_series_buglet(self):
+        # GH #638
+        df = DataFrame({"a": [1, 1]})
+        ds = Series([2], index=[1], name="b")
+        result = df.join(ds, on="a")
+        expected = DataFrame({"a": [1, 1], "b": [2, 2]}, index=df.index)
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_index_mixed(self, join_type):
+        # no overlapping blocks
+        df1 = DataFrame(index=np.arange(10))
+        df1["bool"] = True
+        df1["string"] = "foo"
+
+        df2 = DataFrame(index=np.arange(5, 15))
+        df2["int"] = 1
+        df2["float"] = 1.0
+
+        joined = df1.join(df2, how=join_type)
+        expected = _join_by_hand(df1, df2, how=join_type)
+        tm.assert_frame_equal(joined, expected)
+
+        joined = df2.join(df1, how=join_type)
+        expected = _join_by_hand(df2, df1, how=join_type)
+        tm.assert_frame_equal(joined, expected)
+
+    def test_join_index_mixed_overlap(self):
+        df1 = DataFrame(
+            {"A": 1.0, "B": 2, "C": "foo", "D": True},
+            index=np.arange(10),
+            columns=["A", "B", "C", "D"],
+        )
+        assert df1["B"].dtype == np.int64
+        assert df1["D"].dtype == np.bool_
+
+        df2 = DataFrame(
+            {"A": 1.0, "B": 2, "C": "foo", "D": True},
+            index=np.arange(0, 10, 2),
+            columns=["A", "B", "C", "D"],
+        )
+
+        # overlap
+        joined = df1.join(df2, lsuffix="_one", rsuffix="_two")
+        expected_columns = [
+            "A_one",
+            "B_one",
+            "C_one",
+            "D_one",
+            "A_two",
+            "B_two",
+            "C_two",
+            "D_two",
+        ]
+        df1.columns = expected_columns[:4]
+        df2.columns = expected_columns[4:]
+        expected = _join_by_hand(df1, df2)
+        tm.assert_frame_equal(joined, expected)
+
+    def test_join_empty_bug(self):
+        # generated an exception in 0.4.3
+        x = DataFrame()
+        x.join(DataFrame([3], index=[0], columns=["A"]), how="outer")
+
+    def test_join_unconsolidated(self):
+        # GH #331
+        a = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 2)), columns=["a", "b"]
+        )
+        c = Series(np.random.default_rng(2).standard_normal(30))
+        a["c"] = c
+        d = DataFrame(np.random.default_rng(2).standard_normal((30, 1)), columns=["q"])
+
+        # it works!
+        a.join(d)
+        d.join(a)
+
+    def test_join_multiindex(self):
+        index1 = MultiIndex.from_arrays(
+            [["a", "a", "a", "b", "b", "b"], [1, 2, 3, 1, 2, 3]],
+            names=["first", "second"],
+        )
+
+        index2 = MultiIndex.from_arrays(
+            [["b", "b", "b", "c", "c", "c"], [1, 2, 3, 1, 2, 3]],
+            names=["first", "second"],
+        )
+
+        df1 = DataFrame(
+            data=np.random.default_rng(2).standard_normal(6),
+            index=index1,
+            columns=["var X"],
+        )
+        df2 = DataFrame(
+            data=np.random.default_rng(2).standard_normal(6),
+            index=index2,
+            columns=["var Y"],
+        )
+
+        df1 = df1.sort_index(level=0)
+        df2 = df2.sort_index(level=0)
+
+        joined = df1.join(df2, how="outer")
+        ex_index = Index(index1.values).union(Index(index2.values))
+        expected = df1.reindex(ex_index).join(df2.reindex(ex_index))
+        expected.index.names = index1.names
+        tm.assert_frame_equal(joined, expected)
+        assert joined.index.names == index1.names
+
+        df1 = df1.sort_index(level=1)
+        df2 = df2.sort_index(level=1)
+
+        joined = df1.join(df2, how="outer").sort_index(level=0)
+        ex_index = Index(index1.values).union(Index(index2.values))
+        expected = df1.reindex(ex_index).join(df2.reindex(ex_index))
+        expected.index.names = index1.names
+
+        tm.assert_frame_equal(joined, expected)
+        assert joined.index.names == index1.names
+
+    def test_join_inner_multiindex(self, lexsorted_two_level_string_multiindex):
+        key1 = ["bar", "bar", "bar", "foo", "foo", "baz", "baz", "qux", "qux", "snap"]
+        key2 = [
+            "two",
+            "one",
+            "three",
+            "one",
+            "two",
+            "one",
+            "two",
+            "two",
+            "three",
+            "one",
+        ]
+
+        data = np.random.default_rng(2).standard_normal(len(key1))
+        data = DataFrame({"key1": key1, "key2": key2, "data": data})
+
+        index = lexsorted_two_level_string_multiindex
+        to_join = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 3)),
+            index=index,
+            columns=["j_one", "j_two", "j_three"],
+        )
+
+        joined = data.join(to_join, on=["key1", "key2"], how="inner")
+        expected = merge(
+            data,
+            to_join.reset_index(),
+            left_on=["key1", "key2"],
+            right_on=["first", "second"],
+            how="inner",
+            sort=False,
+        )
+
+        expected2 = merge(
+            to_join,
+            data,
+            right_on=["key1", "key2"],
+            left_index=True,
+            how="inner",
+            sort=False,
+        )
+        tm.assert_frame_equal(joined, expected2.reindex_like(joined))
+
+        expected2 = merge(
+            to_join,
+            data,
+            right_on=["key1", "key2"],
+            left_index=True,
+            how="inner",
+            sort=False,
+        )
+
+        expected = expected.drop(["first", "second"], axis=1)
+        expected.index = joined.index
+
+        assert joined.index.is_monotonic_increasing
+        tm.assert_frame_equal(joined, expected)
+
+        # _assert_same_contents(expected, expected2.loc[:, expected.columns])
+
+    def test_join_hierarchical_mixed_raises(self):
+        # GH 2024
+        # GH 40993: For raising, enforced in 2.0
+        df = DataFrame([(1, 2, 3), (4, 5, 6)], columns=["a", "b", "c"])
+        new_df = df.groupby(["a"]).agg({"b": ["mean", "sum"]})
+        other_df = DataFrame([(1, 2, 3), (7, 10, 6)], columns=["a", "b", "d"])
+        other_df.set_index("a", inplace=True)
+        # GH 9455, 12219
+        with pytest.raises(
+            pd.errors.MergeError, match="Not allowed to merge between different levels"
+        ):
+            merge(new_df, other_df, left_index=True, right_index=True)
+
+    def test_join_float64_float32(self):
+        a = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 2)),
+            columns=["a", "b"],
+            dtype=np.float64,
+        )
+        b = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 1)),
+            columns=["c"],
+            dtype=np.float32,
+        )
+        joined = a.join(b)
+        assert joined.dtypes["a"] == "float64"
+        assert joined.dtypes["b"] == "float64"
+        assert joined.dtypes["c"] == "float32"
+
+        a = np.random.default_rng(2).integers(0, 5, 100).astype("int64")
+        b = np.random.default_rng(2).random(100).astype("float64")
+        c = np.random.default_rng(2).random(100).astype("float32")
+        df = DataFrame({"a": a, "b": b, "c": c})
+        xpdf = DataFrame({"a": a, "b": b, "c": c})
+        s = DataFrame(
+            np.random.default_rng(2).random(5).astype("float32"), columns=["md"]
+        )
+        rs = df.merge(s, left_on="a", right_index=True)
+        assert rs.dtypes["a"] == "int64"
+        assert rs.dtypes["b"] == "float64"
+        assert rs.dtypes["c"] == "float32"
+        assert rs.dtypes["md"] == "float32"
+
+        xp = xpdf.merge(s, left_on="a", right_index=True)
+        tm.assert_frame_equal(rs, xp)
+
+    def test_join_many_non_unique_index(self):
+        df1 = DataFrame({"a": [1, 1], "b": [1, 1], "c": [10, 20]})
+        df2 = DataFrame({"a": [1, 1], "b": [1, 2], "d": [100, 200]})
+        df3 = DataFrame({"a": [1, 1], "b": [1, 2], "e": [1000, 2000]})
+        idf1 = df1.set_index(["a", "b"])
+        idf2 = df2.set_index(["a", "b"])
+        idf3 = df3.set_index(["a", "b"])
+
+        result = idf1.join([idf2, idf3], how="outer")
+
+        df_partially_merged = merge(df1, df2, on=["a", "b"], how="outer")
+        expected = merge(df_partially_merged, df3, on=["a", "b"], how="outer")
+
+        result = result.reset_index()
+        expected = expected[result.columns]
+        expected["a"] = expected.a.astype("int64")
+        expected["b"] = expected.b.astype("int64")
+        tm.assert_frame_equal(result, expected)
+
+        df1 = DataFrame({"a": [1, 1, 1], "b": [1, 1, 1], "c": [10, 20, 30]})
+        df2 = DataFrame({"a": [1, 1, 1], "b": [1, 1, 2], "d": [100, 200, 300]})
+        df3 = DataFrame({"a": [1, 1, 1], "b": [1, 1, 2], "e": [1000, 2000, 3000]})
+        idf1 = df1.set_index(["a", "b"])
+        idf2 = df2.set_index(["a", "b"])
+        idf3 = df3.set_index(["a", "b"])
+        result = idf1.join([idf2, idf3], how="inner")
+
+        df_partially_merged = merge(df1, df2, on=["a", "b"], how="inner")
+        expected = merge(df_partially_merged, df3, on=["a", "b"], how="inner")
+
+        result = result.reset_index()
+
+        tm.assert_frame_equal(result, expected.loc[:, result.columns])
+
+        # GH 11519
+        df = DataFrame(
+            {
+                "A": ["foo", "bar", "foo", "bar", "foo", "bar", "foo", "foo"],
+                "B": ["one", "one", "two", "three", "two", "two", "one", "three"],
+                "C": np.random.default_rng(2).standard_normal(8),
+                "D": np.random.default_rng(2).standard_normal(8),
+            }
+        )
+        s = Series(
+            np.repeat(np.arange(8), 2), index=np.repeat(np.arange(8), 2), name="TEST"
+        )
+        inner = df.join(s, how="inner")
+        outer = df.join(s, how="outer")
+        left = df.join(s, how="left")
+        right = df.join(s, how="right")
+        tm.assert_frame_equal(inner, outer)
+        tm.assert_frame_equal(inner, left)
+        tm.assert_frame_equal(inner, right)
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    def test_join_sort(self, infer_string):
+        with option_context("future.infer_string", infer_string):
+            left = DataFrame(
+                {"key": ["foo", "bar", "baz", "foo"], "value": [1, 2, 3, 4]}
+            )
+            right = DataFrame({"value2": ["a", "b", "c"]}, index=["bar", "baz", "foo"])
+
+            joined = left.join(right, on="key", sort=True)
+            expected = DataFrame(
+                {
+                    "key": ["bar", "baz", "foo", "foo"],
+                    "value": [2, 3, 1, 4],
+                    "value2": ["a", "b", "c", "c"],
+                },
+                index=[1, 2, 0, 3],
+            )
+            tm.assert_frame_equal(joined, expected)
+
+            # smoke test
+            joined = left.join(right, on="key", sort=False)
+            tm.assert_index_equal(joined.index, Index(range(4)), exact=True)
+
+    def test_join_mixed_non_unique_index(self):
+        # GH 12814, unorderable types in py3 with a non-unique index
+        df1 = DataFrame({"a": [1, 2, 3, 4]}, index=[1, 2, 3, "a"])
+        df2 = DataFrame({"b": [5, 6, 7, 8]}, index=[1, 3, 3, 4])
+        result = df1.join(df2)
+        expected = DataFrame(
+            {"a": [1, 2, 3, 3, 4], "b": [5, np.nan, 6, 7, np.nan]},
+            index=[1, 2, 3, 3, "a"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df3 = DataFrame({"a": [1, 2, 3, 4]}, index=[1, 2, 2, "a"])
+        df4 = DataFrame({"b": [5, 6, 7, 8]}, index=[1, 2, 3, 4])
+        result = df3.join(df4)
+        expected = DataFrame(
+            {"a": [1, 2, 3, 4], "b": [5, 6, 6, np.nan]}, index=[1, 2, 2, "a"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_non_unique_period_index(self):
+        # GH #16871
+        index = pd.period_range("2016-01-01", periods=16, freq="M")
+        df = DataFrame(list(range(len(index))), index=index, columns=["pnum"])
+        df2 = concat([df, df])
+        result = df.join(df2, how="inner", rsuffix="_df2")
+        expected = DataFrame(
+            np.tile(np.arange(16, dtype=np.int64).repeat(2).reshape(-1, 1), 2),
+            columns=["pnum", "pnum_df2"],
+            index=df2.sort_index().index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_mixed_type_join_with_suffix(self, using_infer_string):
+        # GH #916
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((20, 6)),
+            columns=["a", "b", "c", "d", "e", "f"],
+        )
+        df.insert(0, "id", 0)
+        df.insert(5, "dt", "foo")
+
+        grouped = df.groupby("id")
+        msg = re.escape("agg function failed [how->mean,dtype->")
+        if using_infer_string:
+            msg = "dtype 'str' does not support operation 'mean'"
+        with pytest.raises(TypeError, match=msg):
+            grouped.mean()
+        mn = grouped.mean(numeric_only=True)
+        cn = grouped.count()
+
+        # it works!
+        mn.join(cn, rsuffix="_right")
+
+    def test_join_many(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 6)), columns=list("abcdef")
+        )
+        df_list = [df[["a", "b"]], df[["c", "d"]], df[["e", "f"]]]
+
+        joined = df_list[0].join(df_list[1:])
+        tm.assert_frame_equal(joined, df)
+
+        df_list = [df[["a", "b"]][:-2], df[["c", "d"]][2:], df[["e", "f"]][1:9]]
+
+        def _check_diff_index(df_list, result, exp_index):
+            reindexed = [x.reindex(exp_index) for x in df_list]
+            expected = reindexed[0].join(reindexed[1:])
+            tm.assert_frame_equal(result, expected)
+
+        # different join types
+        joined = df_list[0].join(df_list[1:], how="outer")
+        _check_diff_index(df_list, joined, df.index)
+
+        joined = df_list[0].join(df_list[1:])
+        _check_diff_index(df_list, joined, df_list[0].index)
+
+        joined = df_list[0].join(df_list[1:], how="inner")
+        _check_diff_index(df_list, joined, df.index[2:8])
+
+        msg = "Joining multiple DataFrames only supported for joining on index"
+        with pytest.raises(ValueError, match=msg):
+            df_list[0].join(df_list[1:], on="a")
+
+    def test_join_many_mixed(self):
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((8, 4)),
+            columns=["A", "B", "C", "D"],
+        )
+        df["key"] = ["foo", "bar"] * 4
+        df1 = df.loc[:, ["A", "B"]]
+        df2 = df.loc[:, ["C", "D"]]
+        df3 = df.loc[:, ["key"]]
+
+        result = df1.join([df2, df3])
+        tm.assert_frame_equal(result, df)
+
+    def test_join_dups(self):
+        # joining dups
+        df = concat(
+            [
+                DataFrame(
+                    np.random.default_rng(2).standard_normal((10, 4)),
+                    columns=["A", "A", "B", "B"],
+                ),
+                DataFrame(
+                    np.random.default_rng(2).integers(0, 10, size=20).reshape(10, 2),
+                    columns=["A", "C"],
+                ),
+            ],
+            axis=1,
+        )
+
+        expected = concat([df, df], axis=1)
+        result = df.join(df, rsuffix="_2")
+        result.columns = expected.columns
+        tm.assert_frame_equal(result, expected)
+
+        # GH 4975, invalid join on dups
+        w = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)), columns=["x", "y"]
+        )
+        x = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)), columns=["x", "y"]
+        )
+        y = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)), columns=["x", "y"]
+        )
+        z = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 2)), columns=["x", "y"]
+        )
+
+        dta = x.merge(y, left_index=True, right_index=True).merge(
+            z, left_index=True, right_index=True, how="outer"
+        )
+        # GH 40991: As of 2.0 causes duplicate columns
+        with pytest.raises(
+            pd.errors.MergeError,
+            match="Passing 'suffixes' which cause duplicate columns",
+        ):
+            dta.merge(w, left_index=True, right_index=True)
+
+    def test_join_multi_to_multi(self, join_type):
+        # GH 20475
+        leftindex = MultiIndex.from_product(
+            [list("abc"), list("xy"), [1, 2]], names=["abc", "xy", "num"]
+        )
+        left = DataFrame({"v1": range(12)}, index=leftindex)
+
+        rightindex = MultiIndex.from_product(
+            [list("abc"), list("xy")], names=["abc", "xy"]
+        )
+        right = DataFrame({"v2": [100 * i for i in range(1, 7)]}, index=rightindex)
+
+        result = left.join(right, on=["abc", "xy"], how=join_type)
+        expected = (
+            left.reset_index()
+            .merge(right.reset_index(), on=["abc", "xy"], how=join_type)
+            .set_index(["abc", "xy", "num"])
+        )
+        tm.assert_frame_equal(expected, result)
+
+        msg = r'len\(left_on\) must equal the number of levels in the index of "right"'
+        with pytest.raises(ValueError, match=msg):
+            left.join(right, on="xy", how=join_type)
+
+        with pytest.raises(ValueError, match=msg):
+            right.join(left, on=["abc", "xy"], how=join_type)
+
+    def test_join_on_tz_aware_datetimeindex(self):
+        # GH 23931, 26335
+        df1 = DataFrame(
+            {
+                "date": pd.date_range(
+                    start="2018-01-01", periods=5, tz="America/Chicago"
+                ),
+                "vals": list("abcde"),
+            }
+        )
+
+        df2 = DataFrame(
+            {
+                "date": pd.date_range(
+                    start="2018-01-03", periods=5, tz="America/Chicago"
+                ),
+                "vals_2": list("tuvwx"),
+            }
+        )
+        result = df1.join(df2.set_index("date"), on="date")
+        expected = df1.copy()
+        expected["vals_2"] = Series([np.nan] * 2 + list("tuv"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_datetime_string(self):
+        # GH 5647
+        dfa = DataFrame(
+            [
+                ["2012-08-02", "L", 10],
+                ["2012-08-02", "J", 15],
+                ["2013-04-06", "L", 20],
+                ["2013-04-06", "J", 25],
+            ],
+            columns=["x", "y", "a"],
+        )
+        dfa["x"] = pd.to_datetime(dfa["x"]).astype("M8[ns]")
+        dfb = DataFrame(
+            [["2012-08-02", "J", 1], ["2013-04-06", "L", 2]],
+            columns=["x", "y", "z"],
+            index=[2, 4],
+        )
+        dfb["x"] = pd.to_datetime(dfb["x"]).astype("M8[ns]")
+        result = dfb.join(dfa.set_index(["x", "y"]), on=["x", "y"])
+        expected = DataFrame(
+            [
+                [Timestamp("2012-08-02 00:00:00"), "J", 1, 15],
+                [Timestamp("2013-04-06 00:00:00"), "L", 2, 20],
+            ],
+            index=[2, 4],
+            columns=["x", "y", "z", "a"],
+        )
+        expected["x"] = expected["x"].astype("M8[ns]")
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_with_categorical_index(self):
+        # GH47812
+        ix = ["a", "b"]
+        id1 = pd.CategoricalIndex(ix, categories=ix)
+        id2 = pd.CategoricalIndex(reversed(ix), categories=reversed(ix))
+
+        df1 = DataFrame({"c1": ix}, index=id1)
+        df2 = DataFrame({"c2": reversed(ix)}, index=id2)
+        result = df1.join(df2)
+        expected = DataFrame(
+            {"c1": ["a", "b"], "c2": ["a", "b"]},
+            index=pd.CategoricalIndex(["a", "b"], categories=["a", "b"]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def _check_join(left, right, result, join_col, how="left", lsuffix="_x", rsuffix="_y"):
+    # some smoke tests
+    for c in join_col:
+        assert result[c].notna().all()
+
+    left_grouped = left.groupby(join_col)
+    right_grouped = right.groupby(join_col)
+
+    for group_key, group in result.groupby(join_col):
+        l_joined = _restrict_to_columns(group, left.columns, lsuffix)
+        r_joined = _restrict_to_columns(group, right.columns, rsuffix)
+
+        try:
+            lgroup = left_grouped.get_group(group_key)
+        except KeyError as err:
+            if how in ("left", "inner"):
+                raise AssertionError(
+                    f"key {group_key} should not have been in the join"
+                ) from err
+
+            _assert_all_na(l_joined, left.columns, join_col)
+        else:
+            _assert_same_contents(l_joined, lgroup)
+
+        try:
+            rgroup = right_grouped.get_group(group_key)
+        except KeyError as err:
+            if how in ("right", "inner"):
+                raise AssertionError(
+                    f"key {group_key} should not have been in the join"
+                ) from err
+
+            _assert_all_na(r_joined, right.columns, join_col)
+        else:
+            _assert_same_contents(r_joined, rgroup)
+
+
+def _restrict_to_columns(group, columns, suffix):
+    found = [
+        c for c in group.columns if c in columns or c.replace(suffix, "") in columns
+    ]
+
+    # filter
+    group = group.loc[:, found]
+
+    # get rid of suffixes, if any
+    group = group.rename(columns=lambda x: x.replace(suffix, ""))
+
+    # put in the right order...
+    group = group.loc[:, columns]
+
+    return group
+
+
+def _assert_same_contents(join_chunk, source):
+    NA_SENTINEL = -1234567  # drop_duplicates not so NA-friendly...
+
+    jvalues = join_chunk.fillna(NA_SENTINEL).drop_duplicates().values
+    svalues = source.fillna(NA_SENTINEL).drop_duplicates().values
+
+    rows = {tuple(row) for row in jvalues}
+    assert len(rows) == len(source)
+    assert all(tuple(row) in rows for row in svalues)
+
+
+def _assert_all_na(join_chunk, source_columns, join_col):
+    for c in source_columns:
+        if c in join_col:
+            continue
+        assert join_chunk[c].isna().all()
+
+
+def _join_by_hand(a, b, how="left"):
+    join_index = a.index.join(b.index, how=how)
+
+    a_re = a.reindex(join_index)
+    b_re = b.reindex(join_index)
+
+    result_columns = a.columns.append(b.columns)
+
+    for col, s in b_re.items():
+        a_re[col] = s
+    return a_re.reindex(columns=result_columns)
+
+
+def test_join_inner_multiindex_deterministic_order():
+    # GH: 36910
+    left = DataFrame(
+        data={"e": 5},
+        index=MultiIndex.from_tuples([(1, 2, 4)], names=("a", "b", "d")),
+    )
+    right = DataFrame(
+        data={"f": 6}, index=MultiIndex.from_tuples([(2, 3)], names=("b", "c"))
+    )
+    result = left.join(right, how="inner")
+    expected = DataFrame(
+        {"e": [5], "f": [6]},
+        index=MultiIndex.from_tuples([(1, 2, 4, 3)], names=("a", "b", "d", "c")),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("input_col", "output_cols"), [("b", ["a", "b"]), ("a", ["a_x", "a_y"])]
+)
+def test_join_cross(input_col, output_cols):
+    # GH#5401
+    left = DataFrame({"a": [1, 3]})
+    right = DataFrame({input_col: [3, 4]})
+    result = left.join(right, how="cross", lsuffix="_x", rsuffix="_y")
+    expected = DataFrame({output_cols[0]: [1, 1, 3, 3], output_cols[1]: [3, 4, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_join_multiindex_one_level(join_type):
+    # GH#36909
+    left = DataFrame(
+        data={"c": 3}, index=MultiIndex.from_tuples([(1, 2)], names=("a", "b"))
+    )
+    right = DataFrame(data={"d": 4}, index=MultiIndex.from_tuples([(2,)], names=("b",)))
+    result = left.join(right, how=join_type)
+    if join_type == "right":
+        expected = DataFrame(
+            {"c": [3], "d": [4]},
+            index=MultiIndex.from_tuples([(2, 1)], names=["b", "a"]),
+        )
+    else:
+        expected = DataFrame(
+            {"c": [3], "d": [4]},
+            index=MultiIndex.from_tuples([(1, 2)], names=["a", "b"]),
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "categories, values",
+    [
+        (["Y", "X"], ["Y", "X", "X"]),
+        ([2, 1], [2, 1, 1]),
+        ([2.5, 1.5], [2.5, 1.5, 1.5]),
+        (
+            [Timestamp("2020-12-31"), Timestamp("2019-12-31")],
+            [Timestamp("2020-12-31"), Timestamp("2019-12-31"), Timestamp("2019-12-31")],
+        ),
+    ],
+)
+def test_join_multiindex_not_alphabetical_categorical(categories, values):
+    # GH#38502
+    left = DataFrame(
+        {
+            "first": ["A", "A"],
+            "second": Categorical(categories, categories=categories),
+            "value": [1, 2],
+        }
+    ).set_index(["first", "second"])
+    right = DataFrame(
+        {
+            "first": ["A", "A", "B"],
+            "second": Categorical(values, categories=categories),
+            "value": [3, 4, 5],
+        }
+    ).set_index(["first", "second"])
+    result = left.join(right, lsuffix="_left", rsuffix="_right")
+
+    expected = DataFrame(
+        {
+            "first": ["A", "A"],
+            "second": Categorical(categories, categories=categories),
+            "value_left": [1, 2],
+            "value_right": [3, 4],
+        }
+    ).set_index(["first", "second"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "left_empty, how, exp",
+    [
+        (False, "left", "left"),
+        (False, "right", "empty"),
+        (False, "inner", "empty"),
+        (False, "outer", "left"),
+        (False, "cross", "empty"),
+        (True, "left", "empty"),
+        (True, "right", "right"),
+        (True, "inner", "empty"),
+        (True, "outer", "right"),
+        (True, "cross", "empty"),
+    ],
+)
+def test_join_empty(left_empty, how, exp):
+    left = DataFrame({"A": [2, 1], "B": [3, 4]}, dtype="int64").set_index("A")
+    right = DataFrame({"A": [1], "C": [5]}, dtype="int64").set_index("A")
+
+    if left_empty:
+        left = left.head(0)
+    else:
+        right = right.head(0)
+
+    result = left.join(right, how=how)
+
+    if exp == "left":
+        expected = DataFrame({"A": [2, 1], "B": [3, 4], "C": [np.nan, np.nan]})
+        expected = expected.set_index("A")
+    elif exp == "right":
+        expected = DataFrame({"B": [np.nan], "A": [1], "C": [5]})
+        expected = expected.set_index("A")
+    elif exp == "empty":
+        expected = DataFrame(columns=["B", "C"], dtype="int64")
+        if how != "cross":
+            expected = expected.rename_axis("A")
+    if how == "outer":
+        expected = expected.sort_index()
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_join_empty_uncomparable_columns():
+    # GH 57048
+    df1 = DataFrame()
+    df2 = DataFrame(columns=["test"])
+    df3 = DataFrame(columns=["foo", ("bar", "baz")])
+
+    result = df1 + df2
+    expected = DataFrame(columns=["test"])
+    tm.assert_frame_equal(result, expected)
+
+    result = df2 + df3
+    expected = DataFrame(columns=[("bar", "baz"), "foo", "test"])
+    tm.assert_frame_equal(result, expected)
+
+    result = df1 + df3
+    expected = DataFrame(columns=[("bar", "baz"), "foo"])
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "how, values",
+    [
+        ("inner", [0, 1, 2]),
+        ("outer", [0, 1, 2]),
+        ("left", [0, 1, 2]),
+        ("right", [0, 2, 1]),
+    ],
+)
+def test_join_multiindex_categorical_output_index_dtype(how, values):
+    # GH#50906
+    df1 = DataFrame(
+        {
+            "a": Categorical([0, 1, 2]),
+            "b": Categorical([0, 1, 2]),
+            "c": [0, 1, 2],
+        }
+    ).set_index(["a", "b"])
+
+    df2 = DataFrame(
+        {
+            "a": Categorical([0, 2, 1]),
+            "b": Categorical([0, 2, 1]),
+            "d": [0, 2, 1],
+        }
+    ).set_index(["a", "b"])
+
+    expected = DataFrame(
+        {
+            "a": Categorical(values),
+            "b": Categorical(values),
+            "c": values,
+            "d": values,
+        }
+    ).set_index(["a", "b"])
+
+    result = df1.join(df2, how=how)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_join_multiindex_with_none_as_label():
+    # GH 58721
+    df1 = DataFrame(
+        {"A": [1]},
+        index=MultiIndex.from_tuples([(3, 3)], names=["X", None]),
+    )
+    df2 = DataFrame(
+        {"B": [2]},
+        index=MultiIndex.from_tuples([(3, 3)], names=[None, "X"]),
+    )
+
+    result12 = df1.join(df2)
+    expected12 = DataFrame(
+        {"A": [1], "B": [2]},
+        index=MultiIndex.from_tuples([(3, 3)], names=["X", None]),
+    )
+    tm.assert_frame_equal(result12, expected12)
+
+    result21 = df2.join(df1)
+    expected21 = DataFrame(
+        {"B": [2], "A": [1]},
+        index=MultiIndex.from_tuples([(3, 3)], names=[None, "X"]),
+    )
+    tm.assert_frame_equal(result21, expected21)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge.py b/pandas/pandas/tests/reshape/merge/test_merge.py
new file mode 100644
index 0000000000000..f0f67aebd85ec
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge.py
@@ -0,0 +1,3062 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+import re
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import (
+    is_object_dtype,
+    is_string_dtype,
+)
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+from pandas.core.reshape.concat import concat
+from pandas.core.reshape.merge import (
+    MergeError,
+    merge,
+)
+
+
+def get_test_data(ngroups=8, n=50):
+    unique_groups = list(range(ngroups))
+    arr = np.asarray(np.tile(unique_groups, n // ngroups))
+
+    if len(arr) < n:
+        arr = np.asarray(list(arr) + unique_groups[: n - len(arr)])
+
+    np.random.default_rng(2).shuffle(arr)
+    return arr
+
+
+@pytest.fixture
+def dfs_for_indicator():
+    df1 = DataFrame({"col1": [0, 1], "col_conflict": [1, 2], "col_left": ["a", "b"]})
+    df2 = DataFrame(
+        {
+            "col1": [1, 2, 3, 4, 5],
+            "col_conflict": [1, 2, 3, 4, 5],
+            "col_right": [2, 2, 2, 2, 2],
+        }
+    )
+    return df1, df2
+
+
+class TestMerge:
+    @pytest.fixture
+    def df(self):
+        df = DataFrame(
+            {
+                "key1": get_test_data(),
+                "key2": get_test_data(),
+                "data1": np.random.default_rng(2).standard_normal(50),
+                "data2": np.random.default_rng(2).standard_normal(50),
+            }
+        )
+
+        # exclude a couple keys for fun
+        df = df[df["key2"] > 1]
+        return df
+
+    @pytest.fixture
+    def df2(self):
+        return DataFrame(
+            {
+                "key1": get_test_data(n=10),
+                "key2": get_test_data(ngroups=4, n=10),
+                "value": np.random.default_rng(2).standard_normal(10),
+            }
+        )
+
+    @pytest.fixture
+    def left(self):
+        return DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "e", "a"],
+                "v1": np.random.default_rng(2).standard_normal(7),
+            }
+        )
+
+    def test_merge_inner_join_empty(self):
+        # GH 15328
+        df_empty = DataFrame()
+        df_a = DataFrame({"a": [1, 2]}, index=[0, 1], dtype="int64")
+        result = merge(df_empty, df_a, left_index=True, right_index=True)
+        expected = DataFrame({"a": []}, dtype="int64")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_common(self, df, df2):
+        joined = merge(df, df2)
+        exp = merge(df, df2, on=["key1", "key2"])
+        tm.assert_frame_equal(joined, exp)
+
+    def test_merge_non_string_columns(self):
+        # https://github.com/pandas-dev/pandas/issues/17962
+        # Checks that method runs for non string column names
+        left = DataFrame(
+            {0: [1, 0, 1, 0], 1: [0, 1, 0, 0], 2: [0, 0, 2, 0], 3: [1, 0, 0, 3]}
+        )
+
+        right = left.astype(float)
+        expected = left
+        result = merge(left, right)
+        tm.assert_frame_equal(expected, result)
+
+    def test_merge_index_as_on_arg(self, df, df2):
+        # GH14355
+
+        left = df.set_index("key1")
+        right = df2.set_index("key1")
+        result = merge(left, right, on="key1")
+        expected = merge(df, df2, on="key1").set_index("key1")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_index_singlekey_right_vs_left(self):
+        left = DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "e", "a"],
+                "v1": np.random.default_rng(2).standard_normal(7),
+            }
+        )
+        right = DataFrame(
+            {"v2": np.random.default_rng(2).standard_normal(4)},
+            index=["d", "b", "c", "a"],
+        )
+
+        merged1 = merge(
+            left, right, left_on="key", right_index=True, how="left", sort=False
+        )
+        merged2 = merge(
+            right, left, right_on="key", left_index=True, how="right", sort=False
+        )
+        tm.assert_frame_equal(merged1, merged2.loc[:, merged1.columns])
+
+        merged1 = merge(
+            left, right, left_on="key", right_index=True, how="left", sort=True
+        )
+        merged2 = merge(
+            right, left, right_on="key", left_index=True, how="right", sort=True
+        )
+        tm.assert_frame_equal(merged1, merged2.loc[:, merged1.columns])
+
+    def test_merge_index_singlekey_inner(self):
+        left = DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "e", "a"],
+                "v1": np.random.default_rng(2).standard_normal(7),
+            }
+        )
+        right = DataFrame(
+            {"v2": np.random.default_rng(2).standard_normal(4)},
+            index=["d", "b", "c", "a"],
+        )
+
+        # inner join
+        result = merge(left, right, left_on="key", right_index=True, how="inner")
+        expected = left.join(right, on="key").loc[result.index]
+        tm.assert_frame_equal(result, expected)
+
+        result = merge(right, left, right_on="key", left_index=True, how="inner")
+        expected = left.join(right, on="key").loc[result.index]
+        tm.assert_frame_equal(result, expected.loc[:, result.columns])
+
+    def test_merge_misspecified(self, df, df2, left):
+        right = DataFrame(
+            {"v2": np.random.default_rng(2).standard_normal(4)},
+            index=["d", "b", "c", "a"],
+        )
+        msg = "Must pass right_on or right_index=True"
+        with pytest.raises(pd.errors.MergeError, match=msg):
+            merge(left, right, left_index=True)
+        msg = "Must pass left_on or left_index=True"
+        with pytest.raises(pd.errors.MergeError, match=msg):
+            merge(left, right, right_index=True)
+
+        msg = (
+            'Can only pass argument "on" OR "left_on" and "right_on", not '
+            "a combination of both"
+        )
+        with pytest.raises(pd.errors.MergeError, match=msg):
+            merge(left, left, left_on="key", on="key")
+
+        msg = r"len\(right_on\) must equal len\(left_on\)"
+        with pytest.raises(ValueError, match=msg):
+            merge(df, df2, left_on=["key1"], right_on=["key1", "key2"])
+
+    def test_index_and_on_parameters_confusion(self, df, df2):
+        msg = "right_index parameter must be of type bool, not <class 'list'>"
+        with pytest.raises(ValueError, match=msg):
+            merge(
+                df,
+                df2,
+                how="left",
+                left_index=False,
+                right_index=["key1", "key2"],
+            )
+        msg = "left_index parameter must be of type bool, not <class 'list'>"
+        with pytest.raises(ValueError, match=msg):
+            merge(
+                df,
+                df2,
+                how="left",
+                left_index=["key1", "key2"],
+                right_index=False,
+            )
+        with pytest.raises(ValueError, match=msg):
+            merge(
+                df,
+                df2,
+                how="left",
+                left_index=["key1", "key2"],
+                right_index=["key1", "key2"],
+            )
+
+    def test_merge_overlap(self, left):
+        merged = merge(left, left, on="key")
+        exp_len = (left["key"].value_counts() ** 2).sum()
+        assert len(merged) == exp_len
+        assert "v1_x" in merged
+        assert "v1_y" in merged
+
+    def test_merge_different_column_key_names(self):
+        left = DataFrame({"lkey": ["foo", "bar", "baz", "foo"], "value": [1, 2, 3, 4]})
+        right = DataFrame({"rkey": ["foo", "bar", "qux", "foo"], "value": [5, 6, 7, 8]})
+
+        merged = left.merge(
+            right, left_on="lkey", right_on="rkey", how="outer", sort=True
+        )
+
+        exp = Series(["bar", "baz", "foo", "foo", "foo", "foo", np.nan], name="lkey")
+        tm.assert_series_equal(merged["lkey"], exp)
+
+        exp = Series(["bar", np.nan, "foo", "foo", "foo", "foo", "qux"], name="rkey")
+        tm.assert_series_equal(merged["rkey"], exp)
+
+        exp = Series([2, 3, 1, 1, 4, 4, np.nan], name="value_x")
+        tm.assert_series_equal(merged["value_x"], exp)
+
+        exp = Series([6, np.nan, 5, 8, 5, 8, 7], name="value_y")
+        tm.assert_series_equal(merged["value_y"], exp)
+
+    def test_merge_copy(self):
+        left = DataFrame({"a": 0, "b": 1}, index=range(10))
+        right = DataFrame({"c": "foo", "d": "bar"}, index=range(10))
+
+        merged = merge(left, right, left_index=True, right_index=True)
+
+        merged["a"] = 6
+        assert (left["a"] == 0).all()
+
+        merged["d"] = "peekaboo"
+        assert (right["d"] == "bar").all()
+
+    def test_merge_nocopy(self, using_infer_string):
+        left = DataFrame({"a": 0, "b": 1}, index=range(10))
+        right = DataFrame({"c": "foo", "d": "bar"}, index=range(10))
+
+        merged = merge(left, right, left_index=True, right_index=True)
+
+        assert np.shares_memory(merged["a"]._values, left["a"]._values)
+        if not using_infer_string:
+            assert np.shares_memory(merged["d"]._values, right["d"]._values)
+
+    def test_intelligently_handle_join_key(self):
+        # #733, be a bit more 1337 about not returning unconsolidated DataFrame
+
+        left = DataFrame(
+            {"key": [1, 1, 2, 2, 3], "value": list(range(5))}, columns=["value", "key"]
+        )
+        right = DataFrame({"key": [1, 1, 2, 3, 4, 5], "rvalue": list(range(6))})
+
+        joined = merge(left, right, on="key", how="outer")
+        expected = DataFrame(
+            {
+                "key": [1, 1, 1, 1, 2, 2, 3, 4, 5],
+                "value": np.array([0, 0, 1, 1, 2, 3, 4, np.nan, np.nan]),
+                "rvalue": [0, 1, 0, 1, 2, 2, 3, 4, 5],
+            },
+            columns=["value", "key", "rvalue"],
+        )
+        tm.assert_frame_equal(joined, expected)
+
+    def test_merge_join_key_dtype_cast(self):
+        # #8596
+
+        df1 = DataFrame({"key": [1], "v1": [10]})
+        df2 = DataFrame({"key": [2], "v1": [20]})
+        df = merge(df1, df2, how="outer")
+        assert df["key"].dtype == "int64"
+
+        df1 = DataFrame({"key": [True], "v1": [1]})
+        df2 = DataFrame({"key": [False], "v1": [0]})
+        df = merge(df1, df2, how="outer")
+
+        # GH13169
+        # GH#40073
+        assert df["key"].dtype == "bool"
+
+        df1 = DataFrame({"val": [1]})
+        df2 = DataFrame({"val": [2]})
+        lkey = np.array([1])
+        rkey = np.array([2])
+        df = merge(df1, df2, left_on=lkey, right_on=rkey, how="outer")
+        assert df["key_0"].dtype == np.dtype(int)
+
+    def test_handle_join_key_pass_array(self):
+        left = DataFrame(
+            {"key": [1, 1, 2, 2, 3], "value": np.arange(5)},
+            columns=["value", "key"],
+            dtype="int64",
+        )
+        right = DataFrame({"rvalue": np.arange(6)}, dtype="int64")
+        key = np.array([1, 1, 2, 3, 4, 5], dtype="int64")
+
+        merged = merge(left, right, left_on="key", right_on=key, how="outer")
+        merged2 = merge(right, left, left_on=key, right_on="key", how="outer")
+
+        tm.assert_series_equal(merged["key"], merged2["key"])
+        assert merged["key"].notna().all()
+        assert merged2["key"].notna().all()
+
+        left = DataFrame({"value": np.arange(5)}, columns=["value"])
+        right = DataFrame({"rvalue": np.arange(6)})
+        lkey = np.array([1, 1, 2, 2, 3])
+        rkey = np.array([1, 1, 2, 3, 4, 5])
+
+        merged = merge(left, right, left_on=lkey, right_on=rkey, how="outer")
+        expected = Series([1, 1, 1, 1, 2, 2, 3, 4, 5], dtype=int, name="key_0")
+        tm.assert_series_equal(merged["key_0"], expected)
+
+        left = DataFrame({"value": np.arange(3)})
+        right = DataFrame({"rvalue": np.arange(6)})
+
+        key = np.array([0, 1, 1, 2, 2, 3], dtype=np.int64)
+        merged = merge(left, right, left_index=True, right_on=key, how="outer")
+        tm.assert_series_equal(merged["key_0"], Series(key, name="key_0"))
+
+    def test_no_overlap_more_informative_error(self):
+        dt = datetime.now()
+        df1 = DataFrame({"x": ["a"]}, index=[dt])
+
+        df2 = DataFrame({"y": ["b", "c"]}, index=[dt, dt])
+
+        msg = (
+            "No common columns to perform merge on. "
+            f"Merge options: left_on={None}, right_on={None}, "
+            f"left_index={False}, right_index={False}"
+        )
+
+        with pytest.raises(MergeError, match=msg):
+            merge(df1, df2)
+
+    def test_merge_non_unique_indexes(self):
+        dt = datetime(2012, 5, 1)
+        dt2 = datetime(2012, 5, 2)
+        dt3 = datetime(2012, 5, 3)
+        dt4 = datetime(2012, 5, 4)
+
+        df1 = DataFrame({"x": ["a"]}, index=[dt])
+        df2 = DataFrame({"y": ["b", "c"]}, index=[dt, dt])
+        _check_merge(df1, df2)
+
+        # Not monotonic
+        df1 = DataFrame({"x": ["a", "b", "q"]}, index=[dt2, dt, dt4])
+        df2 = DataFrame(
+            {"y": ["c", "d", "e", "f", "g", "h"]}, index=[dt3, dt3, dt2, dt2, dt, dt]
+        )
+        _check_merge(df1, df2)
+
+        df1 = DataFrame({"x": ["a", "b"]}, index=[dt, dt])
+        df2 = DataFrame({"y": ["c", "d"]}, index=[dt, dt])
+        _check_merge(df1, df2)
+
+    def test_merge_non_unique_index_many_to_many(self):
+        dt = datetime(2012, 5, 1)
+        dt2 = datetime(2012, 5, 2)
+        dt3 = datetime(2012, 5, 3)
+        df1 = DataFrame({"x": ["a", "b", "c", "d"]}, index=[dt2, dt2, dt, dt])
+        df2 = DataFrame(
+            {"y": ["e", "f", "g", " h", "i"]}, index=[dt2, dt2, dt3, dt, dt]
+        )
+        _check_merge(df1, df2)
+
+    def test_left_merge_empty_dataframe(self):
+        left = DataFrame({"key": [1], "value": [2]})
+        right = DataFrame({"key": []})
+
+        result = merge(left, right, on="key", how="left")
+        tm.assert_frame_equal(result, left)
+
+        result = merge(right, left, on="key", how="right")
+        tm.assert_frame_equal(result, left)
+
+    def test_merge_empty_dataframe(self, index, join_type):
+        # GH52777
+        left = DataFrame([], index=index[:0])
+        right = left.copy()
+
+        result = left.join(right, how=join_type)
+        tm.assert_frame_equal(result, left)
+
+    @pytest.mark.parametrize(
+        "kwarg",
+        [
+            {"left_index": True, "right_index": True},
+            {"left_index": True, "right_on": "x"},
+            {"left_on": "a", "right_index": True},
+            {"left_on": "a", "right_on": "x"},
+        ],
+    )
+    def test_merge_left_empty_right_empty(self, join_type, kwarg):
+        # GH 10824
+        left = DataFrame(columns=["a", "b", "c"])
+        right = DataFrame(columns=["x", "y", "z"])
+
+        exp_in = DataFrame(columns=["a", "b", "c", "x", "y", "z"], dtype=object)
+
+        result = merge(left, right, how=join_type, **kwarg)
+        tm.assert_frame_equal(result, exp_in)
+
+    def test_merge_left_empty_right_notempty(self):
+        # GH 10824
+        left = DataFrame(columns=["a", "b", "c"])
+        right = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=["x", "y", "z"])
+
+        exp_out = DataFrame(
+            {
+                "a": np.array([np.nan] * 3, dtype=object),
+                "b": np.array([np.nan] * 3, dtype=object),
+                "c": np.array([np.nan] * 3, dtype=object),
+                "x": [1, 4, 7],
+                "y": [2, 5, 8],
+                "z": [3, 6, 9],
+            },
+            columns=["a", "b", "c", "x", "y", "z"],
+        )
+        exp_in = exp_out[0:0]  # make empty DataFrame keeping dtype
+
+        def check1(exp, kwarg):
+            result = merge(left, right, how="inner", **kwarg)
+            tm.assert_frame_equal(result, exp)
+            result = merge(left, right, how="left", **kwarg)
+            tm.assert_frame_equal(result, exp)
+
+        def check2(exp, kwarg):
+            result = merge(left, right, how="right", **kwarg)
+            tm.assert_frame_equal(result, exp)
+            result = merge(left, right, how="outer", **kwarg)
+            tm.assert_frame_equal(result, exp)
+
+        for kwarg in [
+            {"left_index": True, "right_index": True},
+            {"left_index": True, "right_on": "x"},
+        ]:
+            check1(exp_in, kwarg)
+            check2(exp_out, kwarg)
+
+        kwarg = {"left_on": "a", "right_index": True}
+        check1(exp_in, kwarg)
+        exp_out["a"] = [0, 1, 2]
+        check2(exp_out, kwarg)
+
+        kwarg = {"left_on": "a", "right_on": "x"}
+        check1(exp_in, kwarg)
+        exp_out["a"] = np.array([np.nan] * 3, dtype=object)
+        check2(exp_out, kwarg)
+
+    def test_merge_left_notempty_right_empty(self):
+        # GH 10824
+        left = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], columns=["a", "b", "c"])
+        right = DataFrame(columns=["x", "y", "z"])
+
+        exp_out = DataFrame(
+            {
+                "a": [1, 4, 7],
+                "b": [2, 5, 8],
+                "c": [3, 6, 9],
+                "x": np.array([np.nan] * 3, dtype=object),
+                "y": np.array([np.nan] * 3, dtype=object),
+                "z": np.array([np.nan] * 3, dtype=object),
+            },
+            columns=["a", "b", "c", "x", "y", "z"],
+        )
+        exp_in = exp_out[0:0]  # make empty DataFrame keeping dtype
+        # result will have object dtype
+        exp_in.index = exp_in.index.astype(object)
+
+        def check1(exp, kwarg):
+            result = merge(left, right, how="inner", **kwarg)
+            tm.assert_frame_equal(result, exp)
+            result = merge(left, right, how="right", **kwarg)
+            tm.assert_frame_equal(result, exp)
+
+        def check2(exp, kwarg):
+            result = merge(left, right, how="left", **kwarg)
+            tm.assert_frame_equal(result, exp)
+            result = merge(left, right, how="outer", **kwarg)
+            tm.assert_frame_equal(result, exp)
+
+            # TODO: should the next loop be un-indented? doing so breaks this test
+            for kwarg in [
+                {"left_index": True, "right_index": True},
+                {"left_index": True, "right_on": "x"},
+                {"left_on": "a", "right_index": True},
+                {"left_on": "a", "right_on": "x"},
+            ]:
+                check1(exp_in, kwarg)
+                check2(exp_out, kwarg)
+
+    @pytest.mark.parametrize(
+        "series_of_dtype",
+        [
+            Series([1], dtype="int64"),
+            Series([1], dtype="Int64"),
+            Series([1.23]),
+            Series(["foo"]),
+            Series([True]),
+            Series([pd.Timestamp("2018-01-01")]),
+            Series([pd.Timestamp("2018-01-01", tz="US/Eastern")]),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "series_of_dtype2",
+        [
+            Series([1], dtype="int64"),
+            Series([1], dtype="Int64"),
+            Series([1.23]),
+            Series(["foo"]),
+            Series([True]),
+            Series([pd.Timestamp("2018-01-01")]),
+            Series([pd.Timestamp("2018-01-01", tz="US/Eastern")]),
+        ],
+    )
+    def test_merge_empty_frame(self, series_of_dtype, series_of_dtype2):
+        # GH 25183
+        df = DataFrame(
+            {"key": series_of_dtype, "value": series_of_dtype2},
+            columns=["key", "value"],
+        )
+        df_empty = df[:0]
+        expected = DataFrame(
+            {
+                "key": Series(dtype=df.dtypes["key"]),
+                "value_x": Series(dtype=df.dtypes["value"]),
+                "value_y": Series(dtype=df.dtypes["value"]),
+            },
+            columns=["key", "value_x", "value_y"],
+        )
+        actual = df_empty.merge(df, on="key")
+        tm.assert_frame_equal(actual, expected)
+
+    @pytest.mark.parametrize(
+        "series_of_dtype",
+        [
+            Series([1], dtype="int64"),
+            Series([1], dtype="Int64"),
+            Series([1.23]),
+            Series(["foo"]),
+            Series([True]),
+            Series([pd.Timestamp("2018-01-01")]),
+            Series([pd.Timestamp("2018-01-01", tz="US/Eastern")]),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "series_of_dtype_all_na",
+        [
+            Series([np.nan], dtype="Int64"),
+            Series([np.nan], dtype="float"),
+            Series([np.nan], dtype="object"),
+            Series([pd.NaT]),
+        ],
+    )
+    def test_merge_all_na_column(self, series_of_dtype, series_of_dtype_all_na):
+        # GH 25183
+        df_left = DataFrame(
+            {"key": series_of_dtype, "value": series_of_dtype_all_na},
+            columns=["key", "value"],
+        )
+        df_right = DataFrame(
+            {"key": series_of_dtype, "value": series_of_dtype_all_na},
+            columns=["key", "value"],
+        )
+        expected = DataFrame(
+            {
+                "key": series_of_dtype,
+                "value_x": series_of_dtype_all_na,
+                "value_y": series_of_dtype_all_na,
+            },
+            columns=["key", "value_x", "value_y"],
+        )
+        actual = df_left.merge(df_right, on="key")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_merge_nosort(self):
+        # GH#2098
+
+        d = {
+            "var1": np.random.default_rng(2).integers(0, 10, size=10),
+            "var2": np.random.default_rng(2).integers(0, 10, size=10),
+            "var3": [
+                datetime(2012, 1, 12),
+                datetime(2011, 2, 4),
+                datetime(2010, 2, 3),
+                datetime(2012, 1, 12),
+                datetime(2011, 2, 4),
+                datetime(2012, 4, 3),
+                datetime(2012, 3, 4),
+                datetime(2008, 5, 1),
+                datetime(2010, 2, 3),
+                datetime(2012, 2, 3),
+            ],
+        }
+        df = DataFrame.from_dict(d)
+        var3 = df.var3.unique()
+        var3 = np.sort(var3)
+        new = DataFrame.from_dict(
+            {"var3": var3, "var8": np.random.default_rng(2).random(7)}
+        )
+
+        result = df.merge(new, on="var3", sort=False)
+        exp = merge(df, new, on="var3", sort=False)
+        tm.assert_frame_equal(result, exp)
+
+        assert (df.var3.unique() == result.var3.unique()).all()
+
+    @pytest.mark.parametrize(
+        ("sort", "values"), [(False, [1, 1, 0, 1, 1]), (True, [0, 1, 1, 1, 1])]
+    )
+    @pytest.mark.parametrize("how", ["left", "right"])
+    def test_merge_same_order_left_right(self, sort, values, how):
+        # GH#35382
+        df = DataFrame({"a": [1, 0, 1]})
+
+        result = df.merge(df, on="a", how=how, sort=sort)
+        expected = DataFrame(values, columns=["a"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_nan_right(self):
+        df1 = DataFrame({"i1": [0, 1], "i2": [0, 1]})
+        df2 = DataFrame({"i1": [0], "i3": [0]})
+        result = df1.join(df2, on="i1", rsuffix="_")
+        expected = (
+            DataFrame(
+                {
+                    "i1": {0: 0.0, 1: 1},
+                    "i2": {0: 0, 1: 1},
+                    "i1_": {0: 0, 1: np.nan},
+                    "i3": {0: 0.0, 1: np.nan},
+                    None: {0: 0, 1: 0},
+                },
+                columns=Index(["i1", "i2", "i1_", "i3", None], dtype=object),
+            )
+            .set_index(None)
+            .reset_index()[["i1", "i2", "i1_", "i3"]]
+        )
+        result.columns = result.columns.astype("object")
+        tm.assert_frame_equal(result, expected, check_dtype=False)
+
+    def test_merge_nan_right2(self):
+        df1 = DataFrame({"i1": [0, 1], "i2": [0.5, 1.5]})
+        df2 = DataFrame({"i1": [0], "i3": [0.7]})
+        result = df1.join(df2, rsuffix="_", on="i1")
+        expected = DataFrame(
+            {
+                "i1": {0: 0, 1: 1},
+                "i1_": {0: 0.0, 1: np.nan},
+                "i2": {0: 0.5, 1: 1.5},
+                "i3": {0: 0.69999999999999996, 1: np.nan},
+            }
+        )[["i1", "i2", "i1_", "i3"]]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+    )
+    def test_merge_type(self, df, df2):
+        class NotADataFrame(DataFrame):
+            @property
+            def _constructor(self):
+                return NotADataFrame
+
+        nad = NotADataFrame(df)
+        result = nad.merge(df2, on="key1")
+
+        assert isinstance(result, NotADataFrame)
+
+    def test_join_append_timedeltas(self):
+        # timedelta64 issues with join/merge
+        # GH 5695
+
+        d = DataFrame.from_dict(
+            {"d": [datetime(2013, 11, 5, 5, 56)], "t": [timedelta(0, 22500)]}
+        )
+        df = DataFrame(columns=list("dt"))
+        df = concat([df, d], ignore_index=True)
+        result = concat([df, d], ignore_index=True)
+        expected = DataFrame(
+            {
+                "d": [datetime(2013, 11, 5, 5, 56), datetime(2013, 11, 5, 5, 56)],
+                "t": [timedelta(0, 22500), timedelta(0, 22500)],
+            },
+            dtype=object,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_append_timedeltas2(self):
+        # timedelta64 issues with join/merge
+        # GH 5695
+        td = np.timedelta64(300000000)
+        lhs = DataFrame(Series([td, td], index=["A", "B"]))
+        rhs = DataFrame(Series([td], index=["A"]))
+
+        result = lhs.join(rhs, rsuffix="r", how="left")
+        expected = DataFrame(
+            {
+                "0": Series([td, td], index=list("AB")),
+                "0r": Series([td, pd.NaT], index=list("AB")),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("unit", ["D", "h", "m", "s", "ms", "us", "ns"])
+    def test_other_datetime_unit(self, unit):
+        # GH 13389
+        df1 = DataFrame({"entity_id": [101, 102]})
+        ser = Series([None, None], index=[101, 102], name="days")
+
+        dtype = f"datetime64[{unit}]"
+
+        if unit in ["D", "h", "m"]:
+            # not supported so we cast to the nearest supported unit, seconds
+            exp_dtype = "datetime64[s]"
+        else:
+            exp_dtype = dtype
+        df2 = ser.astype(exp_dtype).to_frame("days")
+        assert df2["days"].dtype == exp_dtype
+
+        result = df1.merge(df2, left_on="entity_id", right_index=True)
+
+        days = np.array(["nat", "nat"], dtype=exp_dtype)
+        days = pd.core.arrays.DatetimeArray._simple_new(days, dtype=days.dtype)
+        exp = DataFrame(
+            {
+                "entity_id": [101, 102],
+                "days": days,
+            },
+            columns=["entity_id", "days"],
+        )
+        assert exp["days"].dtype == exp_dtype
+        tm.assert_frame_equal(result, exp)
+
+    @pytest.mark.parametrize("unit", ["D", "h", "m", "s", "ms", "us", "ns"])
+    def test_other_timedelta_unit(self, unit):
+        # GH 13389
+        df1 = DataFrame({"entity_id": [101, 102]})
+        ser = Series([None, None], index=[101, 102], name="days")
+
+        dtype = f"m8[{unit}]"
+        if unit in ["D", "h", "m"]:
+            # We cannot astype, instead do nearest supported unit, i.e. "s"
+            msg = "Supported resolutions are 's', 'ms', 'us', 'ns'"
+            with pytest.raises(ValueError, match=msg):
+                ser.astype(dtype)
+
+            df2 = ser.astype("m8[s]").to_frame("days")
+        else:
+            df2 = ser.astype(dtype).to_frame("days")
+            assert df2["days"].dtype == dtype
+
+        result = df1.merge(df2, left_on="entity_id", right_index=True)
+
+        exp = DataFrame(
+            {"entity_id": [101, 102], "days": np.array(["nat", "nat"], dtype=dtype)},
+            columns=["entity_id", "days"],
+        )
+        tm.assert_frame_equal(result, exp)
+
+    def test_overlapping_columns_error_message(self):
+        df = DataFrame({"key": [1, 2, 3], "v1": [4, 5, 6], "v2": [7, 8, 9]})
+        df2 = DataFrame({"key": [1, 2, 3], "v1": [4, 5, 6], "v2": [7, 8, 9]})
+
+        df.columns = ["key", "foo", "foo"]
+        df2.columns = ["key", "bar", "bar"]
+        expected = DataFrame(
+            {
+                "key": [1, 2, 3],
+                "v1": [4, 5, 6],
+                "v2": [7, 8, 9],
+                "v3": [4, 5, 6],
+                "v4": [7, 8, 9],
+            }
+        )
+        expected.columns = ["key", "foo", "foo", "bar", "bar"]
+        tm.assert_frame_equal(merge(df, df2), expected)
+
+        # #2649, #10639
+        df2.columns = ["key1", "foo", "foo"]
+        msg = r"Data columns not unique: Index\(\['foo'\], dtype='object|str'\)"
+        with pytest.raises(MergeError, match=msg):
+            merge(df, df2)
+
+    def test_merge_on_datetime64tz(self):
+        # GH11405
+        left = DataFrame(
+            {
+                "key": pd.date_range("20151010", periods=2, tz="US/Eastern"),
+                "value": [1, 2],
+            }
+        )
+        right = DataFrame(
+            {
+                "key": pd.date_range("20151011", periods=3, tz="US/Eastern"),
+                "value": [1, 2, 3],
+            }
+        )
+
+        expected = DataFrame(
+            {
+                "key": pd.date_range("20151010", periods=4, tz="US/Eastern"),
+                "value_x": [1, 2, np.nan, np.nan],
+                "value_y": [np.nan, 1, 2, 3],
+            }
+        )
+        result = merge(left, right, on="key", how="outer")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_datetime64tz_values(self):
+        left = DataFrame(
+            {
+                "key": [1, 2],
+                "value": pd.date_range("20151010", periods=2, tz="US/Eastern"),
+            }
+        )
+        right = DataFrame(
+            {
+                "key": [2, 3],
+                "value": pd.date_range("20151011", periods=2, tz="US/Eastern"),
+            }
+        )
+        expected = DataFrame(
+            {
+                "key": [1, 2, 3],
+                "value_x": list(pd.date_range("20151010", periods=2, tz="US/Eastern"))
+                + [pd.NaT],
+                "value_y": [pd.NaT]
+                + list(pd.date_range("20151011", periods=2, tz="US/Eastern")),
+            }
+        )
+        result = merge(left, right, on="key", how="outer")
+        tm.assert_frame_equal(result, expected)
+        assert result["value_x"].dtype == "datetime64[ns, US/Eastern]"
+        assert result["value_y"].dtype == "datetime64[ns, US/Eastern]"
+
+    def test_merge_on_datetime64tz_empty(self):
+        # https://github.com/pandas-dev/pandas/issues/25014
+        dtz = pd.DatetimeTZDtype(tz="UTC")
+        right = DataFrame(
+            {
+                "date": DatetimeIndex(["2018"], dtype=dtz),
+                "value": [4.0],
+                "date2": DatetimeIndex(["2019"], dtype=dtz),
+            },
+            columns=["date", "value", "date2"],
+        )
+        left = right[:0]
+        result = left.merge(right, on="date")
+        expected = DataFrame(
+            {
+                "date": Series(dtype=dtz),
+                "value_x": Series(dtype=float),
+                "date2_x": Series(dtype=dtz),
+                "value_y": Series(dtype=float),
+                "date2_y": Series(dtype=dtz),
+            },
+            columns=["date", "value_x", "date2_x", "value_y", "date2_y"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_datetime64tz_with_dst_transition(self):
+        # GH 18885
+        df1 = DataFrame(
+            pd.date_range("2017-10-29 01:00", periods=4, freq="h", tz="Europe/Madrid"),
+            columns=["date"],
+        )
+        df1["value"] = 1
+        df2 = DataFrame(
+            {
+                "date": pd.to_datetime(
+                    [
+                        "2017-10-29 03:00:00",
+                        "2017-10-29 04:00:00",
+                        "2017-10-29 05:00:00",
+                    ]
+                ),
+                "value": 2,
+            }
+        )
+        df2["date"] = df2["date"].dt.tz_localize("UTC").dt.tz_convert("Europe/Madrid")
+        result = merge(df1, df2, how="outer", on="date")
+        expected = DataFrame(
+            {
+                "date": pd.date_range(
+                    "2017-10-29 01:00", periods=7, freq="h", tz="Europe/Madrid"
+                ),
+                "value_x": [1] * 4 + [np.nan] * 3,
+                "value_y": [np.nan] * 4 + [2] * 3,
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_non_unique_period_index(self):
+        # GH #16871
+        index = pd.period_range("2016-01-01", periods=16, freq="M")
+        df = DataFrame(list(range(len(index))), index=index, columns=["pnum"])
+        df2 = concat([df, df])
+        result = df.merge(df2, left_index=True, right_index=True, how="inner")
+        expected = DataFrame(
+            np.tile(np.arange(16, dtype=np.int64).repeat(2).reshape(-1, 1), 2),
+            columns=["pnum_x", "pnum_y"],
+            index=df2.sort_index().index,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_on_periods(self):
+        left = DataFrame(
+            {"key": pd.period_range("20151010", periods=2, freq="D"), "value": [1, 2]}
+        )
+        right = DataFrame(
+            {
+                "key": pd.period_range("20151011", periods=3, freq="D"),
+                "value": [1, 2, 3],
+            }
+        )
+
+        expected = DataFrame(
+            {
+                "key": pd.period_range("20151010", periods=4, freq="D"),
+                "value_x": [1, 2, np.nan, np.nan],
+                "value_y": [np.nan, 1, 2, 3],
+            }
+        )
+        result = merge(left, right, on="key", how="outer")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_period_values(self):
+        left = DataFrame(
+            {"key": [1, 2], "value": pd.period_range("20151010", periods=2, freq="D")}
+        )
+        right = DataFrame(
+            {"key": [2, 3], "value": pd.period_range("20151011", periods=2, freq="D")}
+        )
+
+        exp_x = pd.period_range("20151010", periods=2, freq="D")
+        exp_y = pd.period_range("20151011", periods=2, freq="D")
+        expected = DataFrame(
+            {
+                "key": [1, 2, 3],
+                "value_x": list(exp_x) + [pd.NaT],
+                "value_y": [pd.NaT] + list(exp_y),
+            }
+        )
+        result = merge(left, right, on="key", how="outer")
+        tm.assert_frame_equal(result, expected)
+        assert result["value_x"].dtype == "Period[D]"
+        assert result["value_y"].dtype == "Period[D]"
+
+    def test_indicator(self, dfs_for_indicator):
+        # PR #10054. xref #7412 and closes #8790.
+        df1, df2 = dfs_for_indicator
+        df1_copy = df1.copy()
+
+        df2_copy = df2.copy()
+
+        df_result = DataFrame(
+            {
+                "col1": [0, 1, 2, 3, 4, 5],
+                "col_conflict_x": [1, 2, np.nan, np.nan, np.nan, np.nan],
+                "col_left": ["a", "b", np.nan, np.nan, np.nan, np.nan],
+                "col_conflict_y": [np.nan, 1, 2, 3, 4, 5],
+                "col_right": [np.nan, 2, 2, 2, 2, 2],
+            }
+        )
+        df_result["_merge"] = Categorical(
+            [
+                "left_only",
+                "both",
+                "right_only",
+                "right_only",
+                "right_only",
+                "right_only",
+            ],
+            categories=["left_only", "right_only", "both"],
+        )
+
+        df_result = df_result[
+            [
+                "col1",
+                "col_conflict_x",
+                "col_left",
+                "col_conflict_y",
+                "col_right",
+                "_merge",
+            ]
+        ]
+
+        test = merge(df1, df2, on="col1", how="outer", indicator=True)
+        tm.assert_frame_equal(test, df_result)
+        test = df1.merge(df2, on="col1", how="outer", indicator=True)
+        tm.assert_frame_equal(test, df_result)
+
+        # No side effects
+        tm.assert_frame_equal(df1, df1_copy)
+        tm.assert_frame_equal(df2, df2_copy)
+
+        # Check with custom name
+        df_result_custom_name = df_result
+        df_result_custom_name = df_result_custom_name.rename(
+            columns={"_merge": "custom_name"}
+        )
+
+        test_custom_name = merge(
+            df1, df2, on="col1", how="outer", indicator="custom_name"
+        )
+        tm.assert_frame_equal(test_custom_name, df_result_custom_name)
+        test_custom_name = df1.merge(
+            df2, on="col1", how="outer", indicator="custom_name"
+        )
+        tm.assert_frame_equal(test_custom_name, df_result_custom_name)
+
+    def test_merge_indicator_arg_validation(self, dfs_for_indicator):
+        # Check only accepts strings and booleans
+        df1, df2 = dfs_for_indicator
+
+        msg = "indicator option can only accept boolean or string arguments"
+        with pytest.raises(ValueError, match=msg):
+            merge(df1, df2, on="col1", how="outer", indicator=5)
+        with pytest.raises(ValueError, match=msg):
+            df1.merge(df2, on="col1", how="outer", indicator=5)
+
+    def test_merge_indicator_result_integrity(self, dfs_for_indicator):
+        # Check result integrity
+        df1, df2 = dfs_for_indicator
+
+        test2 = merge(df1, df2, on="col1", how="left", indicator=True)
+        assert (test2._merge != "right_only").all()
+        test2 = df1.merge(df2, on="col1", how="left", indicator=True)
+        assert (test2._merge != "right_only").all()
+
+        test3 = merge(df1, df2, on="col1", how="right", indicator=True)
+        assert (test3._merge != "left_only").all()
+        test3 = df1.merge(df2, on="col1", how="right", indicator=True)
+        assert (test3._merge != "left_only").all()
+
+        test4 = merge(df1, df2, on="col1", how="inner", indicator=True)
+        assert (test4._merge == "both").all()
+        test4 = df1.merge(df2, on="col1", how="inner", indicator=True)
+        assert (test4._merge == "both").all()
+
+    def test_merge_indicator_invalid(self, dfs_for_indicator):
+        # Check if working name in df
+        df1, _ = dfs_for_indicator
+
+        for i in ["_right_indicator", "_left_indicator", "_merge"]:
+            df_badcolumn = DataFrame({"col1": [1, 2], i: [2, 2]})
+
+            msg = (
+                "Cannot use `indicator=True` option when data contains a "
+                f"column named {i}|"
+                "Cannot use name of an existing column for indicator column"
+            )
+            with pytest.raises(ValueError, match=msg):
+                merge(df1, df_badcolumn, on="col1", how="outer", indicator=True)
+            with pytest.raises(ValueError, match=msg):
+                df1.merge(df_badcolumn, on="col1", how="outer", indicator=True)
+
+        # Check for name conflict with custom name
+        df_badcolumn = DataFrame({"col1": [1, 2], "custom_column_name": [2, 2]})
+
+        msg = "Cannot use name of an existing column for indicator column"
+        with pytest.raises(ValueError, match=msg):
+            merge(
+                df1,
+                df_badcolumn,
+                on="col1",
+                how="outer",
+                indicator="custom_column_name",
+            )
+        with pytest.raises(ValueError, match=msg):
+            df1.merge(
+                df_badcolumn, on="col1", how="outer", indicator="custom_column_name"
+            )
+
+    def test_merge_indicator_multiple_columns(self):
+        # Merge on multiple columns
+        df3 = DataFrame({"col1": [0, 1], "col2": ["a", "b"]})
+
+        df4 = DataFrame({"col1": [1, 1, 3], "col2": ["b", "x", "y"]})
+
+        hand_coded_result = DataFrame(
+            {"col1": [0, 1, 1, 3], "col2": ["a", "b", "x", "y"]}
+        )
+        hand_coded_result["_merge"] = Categorical(
+            ["left_only", "both", "right_only", "right_only"],
+            categories=["left_only", "right_only", "both"],
+        )
+
+        test5 = merge(df3, df4, on=["col1", "col2"], how="outer", indicator=True)
+        tm.assert_frame_equal(test5, hand_coded_result)
+        test5 = df3.merge(df4, on=["col1", "col2"], how="outer", indicator=True)
+        tm.assert_frame_equal(test5, hand_coded_result)
+
+    def test_validation(self):
+        left = DataFrame(
+            {"a": ["a", "b", "c", "d"], "b": ["cat", "dog", "weasel", "horse"]},
+            index=range(4),
+        )
+
+        right = DataFrame(
+            {
+                "a": ["a", "b", "c", "d", "e"],
+                "c": ["meow", "bark", "um... weasel noise?", "nay", "chirp"],
+            },
+            index=range(5),
+        )
+
+        # Make sure no side effects.
+        left_copy = left.copy()
+        right_copy = right.copy()
+
+        result = merge(left, right, left_index=True, right_index=True, validate="1:1")
+        tm.assert_frame_equal(left, left_copy)
+        tm.assert_frame_equal(right, right_copy)
+
+        # make sure merge still correct
+        expected = DataFrame(
+            {
+                "a_x": ["a", "b", "c", "d"],
+                "b": ["cat", "dog", "weasel", "horse"],
+                "a_y": ["a", "b", "c", "d"],
+                "c": ["meow", "bark", "um... weasel noise?", "nay"],
+            },
+            index=range(4),
+            columns=["a_x", "b", "a_y", "c"],
+        )
+
+        result = merge(
+            left, right, left_index=True, right_index=True, validate="one_to_one"
+        )
+        tm.assert_frame_equal(result, expected)
+
+        expected_2 = DataFrame(
+            {
+                "a": ["a", "b", "c", "d"],
+                "b": ["cat", "dog", "weasel", "horse"],
+                "c": ["meow", "bark", "um... weasel noise?", "nay"],
+            },
+            index=range(4),
+        )
+
+        result = merge(left, right, on="a", validate="1:1")
+        tm.assert_frame_equal(left, left_copy)
+        tm.assert_frame_equal(right, right_copy)
+        tm.assert_frame_equal(result, expected_2)
+
+        result = merge(left, right, on="a", validate="one_to_one")
+        tm.assert_frame_equal(result, expected_2)
+
+        # One index, one column
+        expected_3 = DataFrame(
+            {
+                "b": ["cat", "dog", "weasel", "horse"],
+                "a": ["a", "b", "c", "d"],
+                "c": ["meow", "bark", "um... weasel noise?", "nay"],
+            },
+            columns=["b", "a", "c"],
+            index=range(4),
+        )
+
+        left_index_reset = left.set_index("a")
+        result = merge(
+            left_index_reset,
+            right,
+            left_index=True,
+            right_on="a",
+            validate="one_to_one",
+        )
+        tm.assert_frame_equal(result, expected_3)
+
+        # Dups on right
+        right_w_dups = concat([right, DataFrame({"a": ["e"], "c": ["moo"]}, index=[4])])
+        merge(
+            left,
+            right_w_dups,
+            left_index=True,
+            right_index=True,
+            validate="one_to_many",
+        )
+
+        msg = "Merge keys are not unique in right dataset; not a one-to-one merge"
+        with pytest.raises(MergeError, match=msg):
+            merge(
+                left,
+                right_w_dups,
+                left_index=True,
+                right_index=True,
+                validate="one_to_one",
+            )
+
+        with pytest.raises(MergeError, match=msg):
+            merge(left, right_w_dups, on="a", validate="one_to_one")
+
+        # Dups on left
+        left_w_dups = concat(
+            [left, DataFrame({"a": ["a"], "c": ["cow"]}, index=[3])], sort=True
+        )
+        merge(
+            left_w_dups,
+            right,
+            left_index=True,
+            right_index=True,
+            validate="many_to_one",
+        )
+
+        msg = "Merge keys are not unique in left dataset; not a one-to-one merge"
+        with pytest.raises(MergeError, match=msg):
+            merge(
+                left_w_dups,
+                right,
+                left_index=True,
+                right_index=True,
+                validate="one_to_one",
+            )
+
+        with pytest.raises(MergeError, match=msg):
+            merge(left_w_dups, right, on="a", validate="one_to_one")
+
+        # Dups on both
+        merge(left_w_dups, right_w_dups, on="a", validate="many_to_many")
+
+        msg = "Merge keys are not unique in right dataset; not a many-to-one merge"
+        with pytest.raises(MergeError, match=msg):
+            merge(
+                left_w_dups,
+                right_w_dups,
+                left_index=True,
+                right_index=True,
+                validate="many_to_one",
+            )
+
+        msg = "Merge keys are not unique in left dataset; not a one-to-many merge"
+        with pytest.raises(MergeError, match=msg):
+            merge(left_w_dups, right_w_dups, on="a", validate="one_to_many")
+
+        # Check invalid arguments
+        msg = (
+            '"jibberish" is not a valid argument. '
+            "Valid arguments are:\n"
+            '- "1:1"\n'
+            '- "1:m"\n'
+            '- "m:1"\n'
+            '- "m:m"\n'
+            '- "one_to_one"\n'
+            '- "one_to_many"\n'
+            '- "many_to_one"\n'
+            '- "many_to_many"'
+        )
+        with pytest.raises(ValueError, match=msg):
+            merge(left, right, on="a", validate="jibberish")
+
+        # Two column merge, dups in both, but jointly no dups.
+        left = DataFrame(
+            {
+                "a": ["a", "a", "b", "b"],
+                "b": [0, 1, 0, 1],
+                "c": ["cat", "dog", "weasel", "horse"],
+            },
+            index=range(4),
+        )
+
+        right = DataFrame(
+            {
+                "a": ["a", "a", "b"],
+                "b": [0, 1, 0],
+                "d": ["meow", "bark", "um... weasel noise?"],
+            },
+            index=range(3),
+        )
+
+        expected_multi = DataFrame(
+            {
+                "a": ["a", "a", "b"],
+                "b": [0, 1, 0],
+                "c": ["cat", "dog", "weasel"],
+                "d": ["meow", "bark", "um... weasel noise?"],
+            },
+            index=range(3),
+        )
+
+        msg = (
+            "Merge keys are not unique in either left or right dataset; "
+            "not a one-to-one merge"
+        )
+        with pytest.raises(MergeError, match=msg):
+            merge(left, right, on="a", validate="1:1")
+
+        result = merge(left, right, on=["a", "b"], validate="1:1")
+        tm.assert_frame_equal(result, expected_multi)
+
+    def test_merge_two_empty_df_no_division_error(self):
+        # GH17776, PR #17846
+        a = DataFrame({"a": [], "b": [], "c": []})
+        with np.errstate(divide="raise"):
+            merge(a, a, on=("a", "b"))
+
+    @pytest.mark.parametrize("how", ["right", "outer"])
+    @pytest.mark.parametrize(
+        "index,expected_index",
+        [
+            (
+                CategoricalIndex([1, 2, 4]),
+                CategoricalIndex([1, 2, 4, None, None, None]),
+            ),
+            (
+                DatetimeIndex(
+                    ["2001-01-01", "2002-02-02", "2003-03-03"], dtype="M8[ns]"
+                ),
+                DatetimeIndex(
+                    ["2001-01-01", "2002-02-02", "2003-03-03", pd.NaT, pd.NaT, pd.NaT],
+                    dtype="M8[ns]",
+                ),
+            ),
+            *[
+                (
+                    Index([1, 2, 3], dtype=dtyp),
+                    Index([1, 2, 3, None, None, None], dtype=np.float64),
+                )
+                for dtyp in tm.ALL_REAL_NUMPY_DTYPES
+            ],
+            (
+                IntervalIndex.from_tuples([(1, 2), (2, 3), (3, 4)]),
+                IntervalIndex.from_tuples(
+                    [(1, 2), (2, 3), (3, 4), np.nan, np.nan, np.nan]
+                ),
+            ),
+            (
+                PeriodIndex(["2001-01-01", "2001-01-02", "2001-01-03"], freq="D"),
+                PeriodIndex(
+                    ["2001-01-01", "2001-01-02", "2001-01-03", pd.NaT, pd.NaT, pd.NaT],
+                    freq="D",
+                ),
+            ),
+            (
+                TimedeltaIndex(["1D", "2D", "3D"]),
+                TimedeltaIndex(["1D", "2D", "3D", pd.NaT, pd.NaT, pd.NaT]),
+            ),
+        ],
+    )
+    def test_merge_on_index_with_more_values(self, how, index, expected_index):
+        # GH 24212
+        # pd.merge gets [0, 1, 2, -1, -1, -1] as left_indexer, ensure that
+        # -1 is interpreted as a missing value instead of the last element
+        df1 = DataFrame({"a": [0, 1, 2], "key": [0, 1, 2]}, index=index)
+        df2 = DataFrame({"b": [0, 1, 2, 3, 4, 5]})
+        result = df1.merge(df2, left_on="key", right_index=True, how=how)
+        expected = DataFrame(
+            [
+                [0, 0, 0],
+                [1, 1, 1],
+                [2, 2, 2],
+                [np.nan, 3, 3],
+                [np.nan, 4, 4],
+                [np.nan, 5, 5],
+            ],
+            columns=["a", "key", "b"],
+        )
+        expected.set_index(expected_index, inplace=True)
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_right_index_right(self):
+        # Note: the expected output here is probably incorrect.
+        # See https://github.com/pandas-dev/pandas/issues/17257 for more.
+        # We include this as a regression test for GH-24897.
+        left = DataFrame({"a": [1, 2, 3], "key": [0, 1, 1]})
+        right = DataFrame({"b": [1, 2, 3]})
+
+        expected = DataFrame(
+            {"a": [1, 2, 3, None], "key": [0, 1, 1, 2], "b": [1, 2, 2, 3]},
+            columns=["a", "key", "b"],
+            index=[0, 1, 2, np.nan],
+        )
+        result = left.merge(right, left_on="key", right_index=True, how="right")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("how", ["left", "right"])
+    def test_merge_preserves_row_order(self, how):
+        # GH 27453
+        left_df = DataFrame({"animal": ["dog", "pig"], "max_speed": [40, 11]})
+        right_df = DataFrame({"animal": ["quetzal", "pig"], "max_speed": [80, 11]})
+        result = left_df.merge(right_df, on=["animal", "max_speed"], how=how)
+        if how == "right":
+            expected = DataFrame({"animal": ["quetzal", "pig"], "max_speed": [80, 11]})
+        else:
+            expected = DataFrame({"animal": ["dog", "pig"], "max_speed": [40, 11]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_take_missing_values_from_index_of_other_dtype(self):
+        # GH 24212
+        left = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "key": Categorical(["a", "a", "b"], categories=list("abc")),
+            }
+        )
+        right = DataFrame({"b": [1, 2, 3]}, index=CategoricalIndex(["a", "b", "c"]))
+        result = left.merge(right, left_on="key", right_index=True, how="right")
+        expected = DataFrame(
+            {
+                "a": [1, 2, 3, None],
+                "key": Categorical(["a", "a", "b", "c"]),
+                "b": [1, 1, 2, 3],
+            },
+            index=[0, 1, 2, np.nan],
+        )
+        expected = expected.reindex(columns=["a", "key", "b"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_readonly(self):
+        # https://github.com/pandas-dev/pandas/issues/27943
+        data1 = DataFrame(
+            np.arange(20).reshape((4, 5)) + 1, columns=["a", "b", "c", "d", "e"]
+        )
+        data2 = DataFrame(
+            np.arange(20).reshape((5, 4)) + 1, columns=["a", "b", "x", "y"]
+        )
+
+        # make each underlying block array / column array read-only
+        for block in data1._mgr.blocks:
+            block.values.flags.writeable = False
+
+        data1.merge(data2)  # no error
+
+    def test_merge_how_validation(self):
+        # https://github.com/pandas-dev/pandas/issues/59422
+        data1 = DataFrame(
+            np.arange(20).reshape((4, 5)) + 1, columns=["a", "b", "c", "d", "e"]
+        )
+        data2 = DataFrame(
+            np.arange(20).reshape((5, 4)) + 1, columns=["a", "b", "x", "y"]
+        )
+        msg = (
+            "'full' is not a valid Merge type: left, right, inner, outer, "
+            "left_anti, right_anti, cross, asof"
+        )
+        with pytest.raises(ValueError, match=re.escape(msg)):
+            data1.merge(data2, how="full")
+
+
+def _check_merge(x, y):
+    for how in ["inner", "left", "outer"]:
+        for sort in [True, False]:
+            result = x.join(y, how=how, sort=sort)
+
+            expected = merge(x.reset_index(), y.reset_index(), how=how, sort=sort)
+            expected = expected.set_index("index")
+
+            # TODO check_names on merge?
+            tm.assert_frame_equal(result, expected, check_names=False)
+
+
+class TestMergeDtypes:
+    @pytest.mark.parametrize("dtype", [object, "category"])
+    def test_different(self, dtype):
+        left = DataFrame(
+            {
+                "A": ["foo", "bar"],
+                "B": Series(["foo", "bar"]).astype("category"),
+                "C": [1, 2],
+                "D": [1.0, 2.0],
+                "E": Series([1, 2], dtype="uint64"),
+                "F": Series([1, 2], dtype="int32"),
+            }
+        )
+        right_vals = Series(["foo", "bar"], dtype=dtype)
+        right = DataFrame({"A": right_vals})
+
+        # GH 9780
+        # We allow merging on object and categorical cols and cast
+        # categorical cols to object
+        result = merge(left, right, on="A")
+        assert is_object_dtype(result.A.dtype) or is_string_dtype(result.A.dtype)
+
+    @pytest.mark.parametrize("d2", [np.int64, np.float64, np.float32, np.float16])
+    def test_join_multi_dtypes(self, any_int_numpy_dtype, d2):
+        dtype1 = np.dtype(any_int_numpy_dtype)
+        dtype2 = np.dtype(d2)
+
+        left = DataFrame(
+            {
+                "k1": np.array([0, 1, 2] * 8, dtype=dtype1),
+                "k2": ["foo", "bar"] * 12,
+                "v": np.array(np.arange(24), dtype=np.int64),
+            }
+        )
+
+        index = MultiIndex.from_tuples([(2, "bar"), (1, "foo")])
+        right = DataFrame({"v2": np.array([5, 7], dtype=dtype2)}, index=index)
+
+        result = left.join(right, on=["k1", "k2"])
+
+        expected = left.copy()
+
+        if dtype2.kind == "i":
+            dtype2 = np.dtype("float64")
+        expected["v2"] = np.array(np.nan, dtype=dtype2)
+        expected.loc[(expected.k1 == 2) & (expected.k2 == "bar"), "v2"] = 5
+        expected.loc[(expected.k1 == 1) & (expected.k2 == "foo"), "v2"] = 7
+
+        tm.assert_frame_equal(result, expected)
+
+        result = left.join(right, on=["k1", "k2"], sort=True)
+        expected.sort_values(["k1", "k2"], kind="mergesort", inplace=True)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "int_vals, float_vals, exp_vals",
+        [
+            ([1, 2, 3], [1.0, 2.0, 3.0], {"X": [1, 2, 3], "Y": [1.0, 2.0, 3.0]}),
+            ([1, 2, 3], [1.0, 3.0], {"X": [1, 3], "Y": [1.0, 3.0]}),
+            ([1, 2], [1.0, 2.0, 3.0], {"X": [1, 2], "Y": [1.0, 2.0]}),
+        ],
+    )
+    def test_merge_on_ints_floats(self, int_vals, float_vals, exp_vals):
+        # GH 16572
+        # Check that float column is not cast to object if
+        # merging on float and int columns
+        A = DataFrame({"X": int_vals})
+        B = DataFrame({"Y": float_vals})
+        expected = DataFrame(exp_vals)
+
+        result = A.merge(B, left_on="X", right_on="Y")
+        tm.assert_frame_equal(result, expected)
+
+        result = B.merge(A, left_on="Y", right_on="X")
+        tm.assert_frame_equal(result, expected[["Y", "X"]])
+
+    def test_merge_key_dtype_cast(self):
+        # GH 17044
+        df1 = DataFrame({"key": [1.0, 2.0], "v1": [10, 20]}, columns=["key", "v1"])
+        df2 = DataFrame({"key": [2], "v2": [200]}, columns=["key", "v2"])
+        result = df1.merge(df2, on="key", how="left")
+        expected = DataFrame(
+            {"key": [1.0, 2.0], "v1": [10, 20], "v2": [np.nan, 200.0]},
+            columns=["key", "v1", "v2"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_on_ints_floats_warning(self):
+        # GH 16572
+        # merge will produce a warning when merging on int and
+        # float columns where the float values are not exactly
+        # equal to their int representation
+        A = DataFrame({"X": [1, 2, 3]})
+        B = DataFrame({"Y": [1.1, 2.5, 3.0]})
+        expected = DataFrame({"X": [3], "Y": [3.0]})
+
+        msg = "the float values are not equal to their int representation"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = A.merge(B, left_on="X", right_on="Y")
+            tm.assert_frame_equal(result, expected)
+
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            result = B.merge(A, left_on="Y", right_on="X")
+            tm.assert_frame_equal(result, expected[["Y", "X"]])
+
+        # test no warning if float has NaNs
+        B = DataFrame({"Y": [np.nan, np.nan, 3.0]})
+
+        with tm.assert_produces_warning(None):
+            result = B.merge(A, left_on="Y", right_on="X")
+            tm.assert_frame_equal(result, expected[["Y", "X"]])
+
+    def test_merge_incompat_infer_boolean_object(self):
+        # GH21119: bool + object bool merge OK
+        df1 = DataFrame({"key": Series([True, False], dtype=object)})
+        df2 = DataFrame({"key": [True, False]})
+
+        expected = DataFrame({"key": [True, False]}, dtype=object)
+        result = merge(df1, df2, on="key")
+        tm.assert_frame_equal(result, expected)
+        result = merge(df2, df1, on="key")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_incompat_infer_boolean_object_with_missing(self):
+        # GH21119: bool + object bool merge OK
+        # with missing value
+        df1 = DataFrame({"key": Series([True, False, np.nan], dtype=object)})
+        df2 = DataFrame({"key": [True, False]})
+
+        expected = DataFrame({"key": [True, False]}, dtype=object)
+        result = merge(df1, df2, on="key")
+        tm.assert_frame_equal(result, expected)
+        result = merge(df2, df1, on="key")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "df1_vals, df2_vals",
+        [
+            # merge on category coerces to object
+            ([0, 1, 2], Series(["a", "b", "a"]).astype("category")),
+            ([0.0, 1.0, 2.0], Series(["a", "b", "a"]).astype("category")),
+            # no not infer
+            ([0, 1], Series([False, True], dtype=object)),
+            ([0, 1], Series([False, True], dtype=bool)),
+        ],
+    )
+    def test_merge_incompat_dtypes_are_ok(self, df1_vals, df2_vals):
+        # these are explicitly allowed incompat merges, that pass thru
+        # the result type is dependent on if the values on the rhs are
+        # inferred, otherwise these will be coerced to object
+
+        df1 = DataFrame({"A": df1_vals})
+        df2 = DataFrame({"A": df2_vals})
+
+        result = merge(df1, df2, on=["A"])
+        assert is_object_dtype(result.A.dtype)
+        result = merge(df2, df1, on=["A"])
+        assert is_object_dtype(result.A.dtype) or is_string_dtype(result.A.dtype)
+
+    @pytest.mark.parametrize(
+        "df1_vals, df2_vals",
+        [
+            # do not infer to numeric
+            (Series([1, 2], dtype="uint64"), ["a", "b", "c"]),
+            (Series([1, 2], dtype="int32"), ["a", "b", "c"]),
+            ([0, 1, 2], ["0", "1", "2"]),
+            ([0.0, 1.0, 2.0], ["0", "1", "2"]),
+            (
+                pd.date_range("1/1/2011", periods=2, freq="D"),
+                ["2011-01-01", "2011-01-02"],
+            ),
+            (pd.date_range("1/1/2011", periods=2, freq="D"), [0, 1]),
+            (pd.date_range("1/1/2011", periods=2, freq="D"), [0.0, 1.0]),
+            (
+                pd.date_range("20130101", periods=3),
+                pd.date_range("20130101", periods=3, tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_merge_incompat_dtypes_error(self, df1_vals, df2_vals):
+        # GH 9780, GH 15800
+        # Raise a ValueError when a user tries to merge on
+        # dtypes that are incompatible (e.g., obj and int/float)
+
+        df1 = DataFrame({"A": df1_vals})
+        df2 = DataFrame({"A": df2_vals})
+
+        msg = (
+            f"You are trying to merge on {df1['A'].dtype} and {df2['A'].dtype} "
+            "columns for key 'A'. If you wish to proceed you should use pd.concat"
+        )
+        msg = re.escape(msg)
+        with pytest.raises(ValueError, match=msg):
+            merge(df1, df2, on=["A"])
+
+        # Check that error still raised when swapping order of dataframes
+        msg = (
+            f"You are trying to merge on {df2['A'].dtype} and {df1['A'].dtype} "
+            "columns for key 'A'. If you wish to proceed you should use pd.concat"
+        )
+        msg = re.escape(msg)
+        with pytest.raises(ValueError, match=msg):
+            merge(df2, df1, on=["A"])
+
+        # Check that error still raised when merging on multiple columns
+        # The error message should mention the first incompatible column
+        if len(df1_vals) == len(df2_vals):
+            # Column A in df1 and df2 is of compatible (the same) dtype
+            # Columns B and C in df1 and df2 are of incompatible dtypes
+            df3 = DataFrame({"A": df2_vals, "B": df1_vals, "C": df1_vals})
+            df4 = DataFrame({"A": df2_vals, "B": df2_vals, "C": df2_vals})
+
+            # Check that error raised correctly when merging all columns A, B, and C
+            # The error message should mention key 'B'
+            msg = (
+                f"You are trying to merge on {df3['B'].dtype} and {df4['B'].dtype} "
+                "columns for key 'B'. If you wish to proceed you should use pd.concat"
+            )
+            msg = re.escape(msg)
+            with pytest.raises(ValueError, match=msg):
+                merge(df3, df4)
+
+            # Check that error raised correctly when merging columns A and C
+            # The error message should mention key 'C'
+            msg = (
+                f"You are trying to merge on {df3['C'].dtype} and {df4['C'].dtype} "
+                "columns for key 'C'. If you wish to proceed you should use pd.concat"
+            )
+            msg = re.escape(msg)
+            with pytest.raises(ValueError, match=msg):
+                merge(df3, df4, on=["A", "C"])
+
+    @pytest.mark.parametrize(
+        "expected_data, how",
+        [
+            ([1, 2], "outer"),
+            ([], "inner"),
+            ([2], "right"),
+            ([1], "left"),
+        ],
+    )
+    def test_merge_EA_dtype(self, any_numeric_ea_dtype, how, expected_data):
+        # GH#40073
+        d1 = DataFrame([(1,)], columns=["id"], dtype=any_numeric_ea_dtype)
+        d2 = DataFrame([(2,)], columns=["id"], dtype=any_numeric_ea_dtype)
+        result = merge(d1, d2, how=how)
+        exp_index = RangeIndex(len(expected_data))
+        expected = DataFrame(
+            expected_data, index=exp_index, columns=["id"], dtype=any_numeric_ea_dtype
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "expected_data, how",
+        [
+            (["a", "b"], "outer"),
+            ([], "inner"),
+            (["b"], "right"),
+            (["a"], "left"),
+        ],
+    )
+    def test_merge_string_dtype(self, how, expected_data, any_string_dtype):
+        # GH#40073
+        d1 = DataFrame([("a",)], columns=["id"], dtype=any_string_dtype)
+        d2 = DataFrame([("b",)], columns=["id"], dtype=any_string_dtype)
+        result = merge(d1, d2, how=how)
+        exp_idx = RangeIndex(len(expected_data))
+        expected = DataFrame(
+            expected_data, index=exp_idx, columns=["id"], dtype=any_string_dtype
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "how, expected_data",
+        [
+            ("inner", [[True, 1, 4], [False, 5, 3]]),
+            ("outer", [[False, 5, 3], [True, 1, 4]]),
+            ("left", [[True, 1, 4], [False, 5, 3]]),
+            ("right", [[False, 5, 3], [True, 1, 4]]),
+        ],
+    )
+    def test_merge_bool_dtype(self, how, expected_data):
+        # GH#40073
+        df1 = DataFrame({"A": [True, False], "B": [1, 5]})
+        df2 = DataFrame({"A": [False, True], "C": [3, 4]})
+        result = merge(df1, df2, how=how)
+        expected = DataFrame(expected_data, columns=["A", "B", "C"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_ea_with_string(self, join_type, string_dtype):
+        # GH 43734 Avoid the use of `assign` with multi-index
+        df1 = DataFrame(
+            data={
+                ("lvl0", "lvl1-a"): ["1", "2", "3", "4", None],
+                ("lvl0", "lvl1-b"): ["4", "5", "6", "7", "8"],
+            },
+            dtype=pd.StringDtype(),
+        )
+        df1_copy = df1.copy()
+        df2 = DataFrame(
+            data={
+                ("lvl0", "lvl1-a"): ["1", "2", "3", pd.NA, "5"],
+                ("lvl0", "lvl1-c"): ["7", "8", "9", pd.NA, "11"],
+            },
+            dtype=string_dtype,
+        )
+        df2_copy = df2.copy()
+        merged = merge(left=df1, right=df2, on=[("lvl0", "lvl1-a")], how=join_type)
+
+        # No change in df1 and df2
+        tm.assert_frame_equal(df1, df1_copy)
+        tm.assert_frame_equal(df2, df2_copy)
+
+        # Check the expected types for the merged data frame
+        expected = Series(
+            [np.dtype("O"), pd.StringDtype(), np.dtype("O")],
+            index=MultiIndex.from_tuples(
+                [("lvl0", "lvl1-a"), ("lvl0", "lvl1-b"), ("lvl0", "lvl1-c")]
+            ),
+        )
+        tm.assert_series_equal(merged.dtypes, expected)
+
+    @pytest.mark.parametrize(
+        "left_empty, how, exp",
+        [
+            (False, "left", "left"),
+            (False, "right", "empty"),
+            (False, "inner", "empty"),
+            (False, "outer", "left"),
+            (False, "cross", "empty_cross"),
+            (True, "left", "empty"),
+            (True, "right", "right"),
+            (True, "inner", "empty"),
+            (True, "outer", "right"),
+            (True, "cross", "empty_cross"),
+        ],
+    )
+    def test_merge_empty(self, left_empty, how, exp):
+        left = DataFrame({"A": [2, 1], "B": [3, 4]})
+        right = DataFrame({"A": [1], "C": [5]}, dtype="int64")
+
+        if left_empty:
+            left = left.head(0)
+        else:
+            right = right.head(0)
+
+        result = left.merge(right, how=how)
+
+        if exp == "left":
+            expected = DataFrame({"A": [2, 1], "B": [3, 4], "C": [np.nan, np.nan]})
+        elif exp == "right":
+            expected = DataFrame({"A": [1], "B": [np.nan], "C": [5]})
+        elif exp == "empty":
+            expected = DataFrame(columns=["A", "B", "C"], dtype="int64")
+        elif exp == "empty_cross":
+            expected = DataFrame(columns=["A_x", "B", "A_y", "C"], dtype="int64")
+
+        if how == "outer":
+            expected = expected.sort_values("A", ignore_index=True)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_with_uintc_columns(self):
+        df1 = DataFrame({"a": ["foo", "bar"], "b": np.array([1, 2], dtype=np.uintc)})
+        df2 = DataFrame({"a": ["foo", "baz"], "b": np.array([3, 4], dtype=np.uintc)})
+        result = df1.merge(df2, how="outer")
+        expected = DataFrame(
+            {
+                "a": ["bar", "baz", "foo", "foo"],
+                "b": np.array([2, 4, 1, 3], dtype=np.uintc),
+            }
+        )
+        tm.assert_frame_equal(result.reset_index(drop=True), expected)
+
+    def test_merge_with_intc_columns(self):
+        df1 = DataFrame({"a": ["foo", "bar"], "b": np.array([1, 2], dtype=np.intc)})
+        df2 = DataFrame({"a": ["foo", "baz"], "b": np.array([3, 4], dtype=np.intc)})
+        result = df1.merge(df2, how="outer")
+        expected = DataFrame(
+            {
+                "a": ["bar", "baz", "foo", "foo"],
+                "b": np.array([2, 4, 1, 3], dtype=np.intc),
+            }
+        )
+        tm.assert_frame_equal(result.reset_index(drop=True), expected)
+
+    def test_merge_intc_non_monotonic(self):
+        df = DataFrame({"join_key": Series([0, 2, 1], dtype=np.intc)})
+        df_details = DataFrame(
+            {"join_key": Series([0, 1, 2], dtype=np.intc), "value": ["a", "b", "c"]}
+        )
+        merged = df.merge(df_details, on="join_key", how="left")
+        expected = DataFrame(
+            {"join_key": np.array([0, 2, 1], dtype=np.intc), "value": ["a", "c", "b"]}
+        )
+        tm.assert_frame_equal(merged.reset_index(drop=True), expected)
+
+
+@pytest.fixture
+def left():
+    return DataFrame(
+        {
+            "X": Series(
+                np.random.default_rng(2).choice(["foo", "bar"], size=(10,))
+            ).astype(CategoricalDtype(["foo", "bar"])),
+            "Y": np.random.default_rng(2).choice(["one", "two", "three"], size=(10,)),
+        }
+    )
+
+
+@pytest.fixture
+def right():
+    return DataFrame(
+        {
+            "X": Series(["foo", "bar"]).astype(CategoricalDtype(["foo", "bar"])),
+            "Z": [1, 2],
+        }
+    )
+
+
+class TestMergeCategorical:
+    def test_identical(self, left, using_infer_string):
+        # merging on the same, should preserve dtypes
+        merged = merge(left, left, on="X")
+        result = merged.dtypes.sort_index()
+        dtype = np.dtype("O") if not using_infer_string else "str"
+        expected = Series(
+            [CategoricalDtype(categories=["foo", "bar"]), dtype, dtype],
+            index=["X", "Y_x", "Y_y"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_basic(self, left, right, using_infer_string):
+        # we have matching Categorical dtypes in X
+        # so should preserve the merged column
+        merged = merge(left, right, on="X")
+        result = merged.dtypes.sort_index()
+        dtype = np.dtype("O") if not using_infer_string else "str"
+        expected = Series(
+            [
+                CategoricalDtype(categories=["foo", "bar"]),
+                dtype,
+                np.dtype("int64"),
+            ],
+            index=["X", "Y", "Z"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_merge_categorical(self):
+        # GH 9426
+
+        right = DataFrame(
+            {
+                "c": {0: "a", 1: "b", 2: "c", 3: "d", 4: "e"},
+                "d": {0: "null", 1: "null", 2: "null", 3: "null", 4: "null"},
+            }
+        )
+        left = DataFrame(
+            {
+                "a": {0: "f", 1: "f", 2: "f", 3: "f", 4: "f"},
+                "b": {0: "g", 1: "g", 2: "g", 3: "g", 4: "g"},
+            }
+        )
+        df = merge(left, right, how="left", left_on="b", right_on="c")
+
+        # object-object
+        expected = df.copy()
+
+        # object-cat
+        # note that we propagate the category
+        # because we don't have any matching rows
+        cright = right.copy()
+        cright["d"] = cright["d"].astype("category")
+        result = merge(left, cright, how="left", left_on="b", right_on="c")
+        expected["d"] = expected["d"].astype(CategoricalDtype(["null"]))
+        tm.assert_frame_equal(result, expected)
+
+        # cat-object
+        cleft = left.copy()
+        cleft["b"] = cleft["b"].astype("category")
+        result = merge(cleft, cright, how="left", left_on="b", right_on="c")
+        tm.assert_frame_equal(result, expected)
+
+        # cat-cat
+        cright = right.copy()
+        cright["d"] = cright["d"].astype("category")
+        cleft = left.copy()
+        cleft["b"] = cleft["b"].astype("category")
+        result = merge(cleft, cright, how="left", left_on="b", right_on="c")
+        tm.assert_frame_equal(result, expected)
+
+    def tests_merge_categorical_unordered_equal(self):
+        # GH-19551
+        df1 = DataFrame(
+            {
+                "Foo": Categorical(["A", "B", "C"], categories=["A", "B", "C"]),
+                "Left": ["A0", "B0", "C0"],
+            }
+        )
+
+        df2 = DataFrame(
+            {
+                "Foo": Categorical(["C", "B", "A"], categories=["C", "B", "A"]),
+                "Right": ["C1", "B1", "A1"],
+            }
+        )
+        result = merge(df1, df2, on=["Foo"])
+        expected = DataFrame(
+            {
+                "Foo": Categorical(["A", "B", "C"]),
+                "Left": ["A0", "B0", "C0"],
+                "Right": ["A1", "B1", "C1"],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_multiindex_merge_with_unordered_categoricalindex(self, ordered):
+        # GH 36973
+        pcat = CategoricalDtype(categories=["P2", "P1"], ordered=ordered)
+        df1 = DataFrame(
+            {
+                "id": ["C", "C", "D"],
+                "p": Categorical(["P2", "P1", "P2"], dtype=pcat),
+                "a": [0, 1, 2],
+            }
+        ).set_index(["id", "p"])
+        df2 = DataFrame(
+            {
+                "id": ["A", "C", "C"],
+                "p": Categorical(["P2", "P2", "P1"], dtype=pcat),
+                "d1": [10, 11, 12],
+            }
+        ).set_index(["id", "p"])
+        result = merge(df1, df2, how="left", left_index=True, right_index=True)
+        expected = DataFrame(
+            {
+                "id": ["C", "C", "D"],
+                "p": Categorical(["P2", "P1", "P2"], dtype=pcat),
+                "a": [0, 1, 2],
+                "d1": [11.0, 12.0, np.nan],
+            }
+        ).set_index(["id", "p"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_other_columns(self, left, right, using_infer_string):
+        # non-merge columns should preserve if possible
+        right = right.assign(Z=right.Z.astype("category"))
+
+        merged = merge(left, right, on="X")
+        result = merged.dtypes.sort_index()
+        dtype = np.dtype("O") if not using_infer_string else "str"
+        expected = Series(
+            [
+                CategoricalDtype(categories=["foo", "bar"]),
+                dtype,
+                CategoricalDtype(categories=[1, 2]),
+            ],
+            index=["X", "Y", "Z"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        # categories are preserved
+        assert left.X.values._categories_match_up_to_permutation(merged.X.values)
+        assert right.Z.values._categories_match_up_to_permutation(merged.Z.values)
+
+    @pytest.mark.parametrize(
+        "change",
+        [
+            lambda x: x,
+            lambda x: x.astype(CategoricalDtype(["foo", "bar", "bah"])),
+            lambda x: x.astype(CategoricalDtype(ordered=True)),
+        ],
+    )
+    def test_dtype_on_merged_different(
+        self, change, join_type, left, right, using_infer_string
+    ):
+        # our merging columns, X now has 2 different dtypes
+        # so we must be object as a result
+
+        X = change(right.X.astype("object"))
+        right = right.assign(X=X)
+        assert isinstance(left.X.values.dtype, CategoricalDtype)
+        # assert not left.X.values._categories_match_up_to_permutation(right.X.values)
+
+        merged = merge(left, right, on="X", how=join_type)
+
+        result = merged.dtypes.sort_index()
+        dtype = np.dtype("O") if not using_infer_string else "str"
+        expected = Series([dtype, dtype, np.dtype("int64")], index=["X", "Y", "Z"])
+        tm.assert_series_equal(result, expected)
+
+    def test_self_join_multiple_categories(self):
+        # GH 16767
+        # non-duplicates should work with multiple categories
+        m = 5
+        df = DataFrame(
+            {
+                "a": ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j"] * m,
+                "b": ["t", "w", "x", "y", "z"] * 2 * m,
+                "c": [
+                    letter
+                    for each in ["m", "n", "u", "p", "o"]
+                    for letter in [each] * 2 * m
+                ],
+                "d": [
+                    letter
+                    for each in [
+                        "aa",
+                        "bb",
+                        "cc",
+                        "dd",
+                        "ee",
+                        "ff",
+                        "gg",
+                        "hh",
+                        "ii",
+                        "jj",
+                    ]
+                    for letter in [each] * m
+                ],
+            }
+        )
+
+        # change them all to categorical variables
+        df = df.apply(lambda x: x.astype("category"))
+
+        # self-join should equal ourselves
+        result = merge(df, df, on=list(df.columns))
+
+        tm.assert_frame_equal(result, df)
+
+    def test_dtype_on_categorical_dates(self):
+        # GH 16900
+        # dates should not be coerced to ints
+
+        df = DataFrame(
+            [[date(2001, 1, 1), 1.1], [date(2001, 1, 2), 1.3]], columns=["date", "num2"]
+        )
+        df["date"] = df["date"].astype("category")
+
+        df2 = DataFrame(
+            [[date(2001, 1, 1), 1.3], [date(2001, 1, 3), 1.4]], columns=["date", "num4"]
+        )
+        df2["date"] = df2["date"].astype("category")
+
+        expected_outer = DataFrame(
+            [
+                [pd.Timestamp("2001-01-01").date(), 1.1, 1.3],
+                [pd.Timestamp("2001-01-02").date(), 1.3, np.nan],
+                [pd.Timestamp("2001-01-03").date(), np.nan, 1.4],
+            ],
+            columns=["date", "num2", "num4"],
+        )
+        result_outer = merge(df, df2, how="outer", on=["date"])
+        tm.assert_frame_equal(result_outer, expected_outer)
+
+        expected_inner = DataFrame(
+            [[pd.Timestamp("2001-01-01").date(), 1.1, 1.3]],
+            columns=["date", "num2", "num4"],
+        )
+        result_inner = merge(df, df2, how="inner", on=["date"])
+        tm.assert_frame_equal(result_inner, expected_inner)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    @pytest.mark.parametrize(
+        "category_column,categories,expected_categories",
+        [
+            ([False, True, True, False], [True, False], [True, False]),
+            ([2, 1, 1, 2], [1, 2], [1, 2]),
+            (["False", "True", "True", "False"], ["True", "False"], ["True", "False"]),
+        ],
+    )
+    def test_merging_with_bool_or_int_cateorical_column(
+        self, category_column, categories, expected_categories, ordered
+    ):
+        # GH 17187
+        # merging with a boolean/int categorical column
+        df1 = DataFrame({"id": [1, 2, 3, 4], "cat": category_column})
+        df1["cat"] = df1["cat"].astype(CategoricalDtype(categories, ordered=ordered))
+        df2 = DataFrame({"id": [2, 4], "num": [1, 9]})
+        result = df1.merge(df2)
+        expected = DataFrame({"id": [2, 4], "cat": expected_categories, "num": [1, 9]})
+        expected["cat"] = expected["cat"].astype(
+            CategoricalDtype(categories, ordered=ordered)
+        )
+        tm.assert_frame_equal(expected, result)
+
+    def test_merge_on_int_array(self):
+        # GH 23020
+        df = DataFrame({"A": Series([1, 2, np.nan], dtype="Int64"), "B": 1})
+        result = merge(df, df, on="A")
+        expected = DataFrame(
+            {"A": Series([1, 2, np.nan], dtype="Int64"), "B_x": 1, "B_y": 1}
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+class TestMergeOnIndexes:
+    @pytest.mark.parametrize(
+        "how, sort, expected",
+        [
+            ("inner", False, DataFrame({"a": [20, 10], "b": [200, 100]}, index=[2, 1])),
+            ("inner", True, DataFrame({"a": [10, 20], "b": [100, 200]}, index=[1, 2])),
+            (
+                "left",
+                False,
+                DataFrame({"a": [20, 10, 0], "b": [200, 100, np.nan]}, index=[2, 1, 0]),
+            ),
+            (
+                "left",
+                True,
+                DataFrame({"a": [0, 10, 20], "b": [np.nan, 100, 200]}, index=[0, 1, 2]),
+            ),
+            (
+                "right",
+                False,
+                DataFrame(
+                    {"a": [np.nan, 10, 20], "b": [300, 100, 200]}, index=[3, 1, 2]
+                ),
+            ),
+            (
+                "right",
+                True,
+                DataFrame(
+                    {"a": [10, 20, np.nan], "b": [100, 200, 300]}, index=[1, 2, 3]
+                ),
+            ),
+            (
+                "outer",
+                False,
+                DataFrame(
+                    {"a": [0, 10, 20, np.nan], "b": [np.nan, 100, 200, 300]},
+                    index=[0, 1, 2, 3],
+                ),
+            ),
+            (
+                "outer",
+                True,
+                DataFrame(
+                    {"a": [0, 10, 20, np.nan], "b": [np.nan, 100, 200, 300]},
+                    index=[0, 1, 2, 3],
+                ),
+            ),
+        ],
+    )
+    def test_merge_on_indexes(self, how, sort, expected):
+        left_df = DataFrame({"a": [20, 10, 0]}, index=[2, 1, 0])
+        right_df = DataFrame({"b": [300, 100, 200]}, index=[3, 1, 2])
+        result = merge(
+            left_df, right_df, left_index=True, right_index=True, how=how, sort=sort
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        Index([1, 2, 4], dtype=dtyp, name="index_col")
+        for dtyp in tm.ALL_REAL_NUMPY_DTYPES
+    ]
+    + [
+        CategoricalIndex(["A", "B", "C"], categories=["A", "B", "C"], name="index_col"),
+        RangeIndex(start=0, stop=3, name="index_col"),
+        DatetimeIndex(["2018-01-01", "2018-01-02", "2018-01-03"], name="index_col"),
+    ],
+    ids=lambda x: f"{type(x).__name__}[{x.dtype}]",
+)
+def test_merge_index_types(index):
+    # gh-20777
+    # assert key access is consistent across index types
+    left = DataFrame({"left_data": [1, 2, 3]}, index=index)
+    right = DataFrame({"right_data": [1.0, 2.0, 3.0]}, index=index)
+
+    result = left.merge(right, on=["index_col"])
+
+    expected = DataFrame(
+        {"left_data": [1, 2, 3], "right_data": [1.0, 2.0, 3.0]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "on,left_on,right_on,left_index,right_index,nm",
+    [
+        (["outer", "inner"], None, None, False, False, "B"),
+        (None, None, None, True, True, "B"),
+        (None, ["outer", "inner"], None, False, True, "B"),
+        (None, None, ["outer", "inner"], True, False, "B"),
+        (["outer", "inner"], None, None, False, False, None),
+        (None, None, None, True, True, None),
+        (None, ["outer", "inner"], None, False, True, None),
+        (None, None, ["outer", "inner"], True, False, None),
+    ],
+)
+def test_merge_series(on, left_on, right_on, left_index, right_index, nm):
+    # GH 21220
+    a = DataFrame(
+        {"A": [1, 2, 3, 4]},
+        index=MultiIndex.from_product([["a", "b"], [0, 1]], names=["outer", "inner"]),
+    )
+    b = Series(
+        [1, 2, 3, 4],
+        index=MultiIndex.from_product([["a", "b"], [1, 2]], names=["outer", "inner"]),
+        name=nm,
+    )
+    expected = DataFrame(
+        {"A": [2, 4], "B": [1, 3]},
+        index=MultiIndex.from_product([["a", "b"], [1]], names=["outer", "inner"]),
+    )
+    if nm is not None:
+        result = merge(
+            a,
+            b,
+            on=on,
+            left_on=left_on,
+            right_on=right_on,
+            left_index=left_index,
+            right_index=right_index,
+        )
+        tm.assert_frame_equal(result, expected)
+    else:
+        msg = "Cannot merge a Series without a name"
+        with pytest.raises(ValueError, match=msg):
+            result = merge(
+                a,
+                b,
+                on=on,
+                left_on=left_on,
+                right_on=right_on,
+                left_index=left_index,
+                right_index=right_index,
+            )
+
+
+def test_merge_series_multilevel():
+    # GH#47946
+    # GH 40993: For raising, enforced in 2.0
+    a = DataFrame(
+        {"A": [1, 2, 3, 4]},
+        index=MultiIndex.from_product([["a", "b"], [0, 1]], names=["outer", "inner"]),
+    )
+    b = Series(
+        [1, 2, 3, 4],
+        index=MultiIndex.from_product([["a", "b"], [1, 2]], names=["outer", "inner"]),
+        name=("B", "C"),
+    )
+    with pytest.raises(
+        MergeError, match="Not allowed to merge between different levels"
+    ):
+        merge(a, b, on=["outer", "inner"])
+
+
+@pytest.mark.parametrize(
+    "col1, col2, kwargs, expected_cols",
+    [
+        (0, 0, {"suffixes": ("", "_dup")}, ["0", "0_dup"]),
+        (0, 0, {"suffixes": (None, "_dup")}, [0, "0_dup"]),
+        (0, 0, {"suffixes": ("_x", "_y")}, ["0_x", "0_y"]),
+        (0, 0, {"suffixes": ["_x", "_y"]}, ["0_x", "0_y"]),
+        ("a", 0, {"suffixes": (None, "_y")}, ["a", 0]),
+        (0.0, 0.0, {"suffixes": ("_x", None)}, ["0.0_x", 0.0]),
+        ("b", "b", {"suffixes": (None, "_y")}, ["b", "b_y"]),
+        ("a", "a", {"suffixes": ("_x", None)}, ["a_x", "a"]),
+        ("a", "b", {"suffixes": ("_x", None)}, ["a", "b"]),
+        ("a", "a", {"suffixes": (None, "_x")}, ["a", "a_x"]),
+        (0, 0, {"suffixes": ("_a", None)}, ["0_a", 0]),
+        ("a", "a", {}, ["a_x", "a_y"]),
+        (0, 0, {}, ["0_x", "0_y"]),
+    ],
+)
+def test_merge_suffix(col1, col2, kwargs, expected_cols):
+    # issue: 24782
+    a = DataFrame({col1: [1, 2, 3]})
+    b = DataFrame({col2: [4, 5, 6]})
+
+    expected = DataFrame([[1, 4], [2, 5], [3, 6]], columns=expected_cols)
+
+    result = a.merge(b, left_index=True, right_index=True, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(a, b, left_index=True, right_index=True, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "how,expected",
+    [
+        (
+            "right",
+            {"A": [100, 200, 300], "B1": [60, 70, np.nan], "B2": [600, 700, 800]},
+        ),
+        (
+            "outer",
+            {
+                "A": [1, 100, 200, 300],
+                "B1": [80, 60, 70, np.nan],
+                "B2": [np.nan, 600, 700, 800],
+            },
+        ),
+    ],
+)
+def test_merge_duplicate_suffix(how, expected):
+    left_df = DataFrame({"A": [100, 200, 1], "B": [60, 70, 80]})
+    right_df = DataFrame({"A": [100, 200, 300], "B": [600, 700, 800]})
+    result = merge(left_df, right_df, on="A", how=how, suffixes=("_x", "_x"))
+    expected = DataFrame(expected)
+    expected.columns = ["A", "B_x", "B_x"]
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "col1, col2, suffixes",
+    [("a", "a", (None, None)), ("a", "a", ("", None)), (0, 0, (None, ""))],
+)
+def test_merge_suffix_error(col1, col2, suffixes):
+    # issue: 24782
+    a = DataFrame({col1: [1, 2, 3]})
+    b = DataFrame({col2: [3, 4, 5]})
+
+    # TODO: might reconsider current raise behaviour, see issue 24782
+    msg = "columns overlap but no suffix specified"
+    with pytest.raises(ValueError, match=msg):
+        merge(a, b, left_index=True, right_index=True, suffixes=suffixes)
+
+
+@pytest.mark.parametrize("suffixes", [{"left", "right"}, {"left": 0, "right": 0}])
+def test_merge_suffix_raises(suffixes):
+    a = DataFrame({"a": [1, 2, 3]})
+    b = DataFrame({"b": [3, 4, 5]})
+
+    with pytest.raises(TypeError, match="Passing 'suffixes' as a"):
+        merge(a, b, left_index=True, right_index=True, suffixes=suffixes)
+
+
+@pytest.mark.parametrize(
+    "col1, col2, suffixes, msg",
+    [
+        ("a", "a", ("a", "b", "c"), r"too many values to unpack \(expected 2\)"),
+        ("a", "a", tuple("a"), r"not enough values to unpack \(expected 2, got 1\)"),
+    ],
+)
+def test_merge_suffix_length_error(col1, col2, suffixes, msg):
+    a = DataFrame({col1: [1, 2, 3]})
+    b = DataFrame({col2: [3, 4, 5]})
+
+    with pytest.raises(ValueError, match=msg):
+        merge(a, b, left_index=True, right_index=True, suffixes=suffixes)
+
+
+@pytest.mark.parametrize("cat_dtype", ["one", "two"])
+@pytest.mark.parametrize("reverse", [True, False])
+def test_merge_equal_cat_dtypes(cat_dtype, reverse):
+    # see gh-22501
+    cat_dtypes = {
+        "one": CategoricalDtype(categories=["a", "b", "c"], ordered=False),
+        "two": CategoricalDtype(categories=["a", "b", "c"], ordered=False),
+    }
+
+    df1 = DataFrame(
+        {"foo": Series(["a", "b", "c"]).astype(cat_dtypes["one"]), "left": [1, 2, 3]}
+    ).set_index("foo")
+
+    data_foo = ["a", "b", "c"]
+    data_right = [1, 2, 3]
+
+    if reverse:
+        data_foo.reverse()
+        data_right.reverse()
+
+    df2 = DataFrame(
+        {"foo": Series(data_foo).astype(cat_dtypes[cat_dtype]), "right": data_right}
+    ).set_index("foo")
+
+    result = df1.merge(df2, left_index=True, right_index=True)
+
+    expected = DataFrame(
+        {
+            "left": [1, 2, 3],
+            "right": [1, 2, 3],
+            "foo": Series(["a", "b", "c"]).astype(cat_dtypes["one"]),
+        }
+    ).set_index("foo")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_equal_cat_dtypes2():
+    # see gh-22501
+    cat_dtype = CategoricalDtype(categories=["a", "b", "c"], ordered=False)
+
+    # Test Data
+    df1 = DataFrame(
+        {"foo": Series(["a", "b"]).astype(cat_dtype), "left": [1, 2]}
+    ).set_index("foo")
+
+    df2 = DataFrame(
+        {"foo": Series(["a", "b", "c"]).astype(cat_dtype), "right": [3, 2, 1]}
+    ).set_index("foo")
+
+    result = df1.merge(df2, left_index=True, right_index=True)
+
+    expected = DataFrame(
+        {"left": [1, 2], "right": [3, 2], "foo": Series(["a", "b"]).astype(cat_dtype)}
+    ).set_index("foo")
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_on_cat_and_ext_array():
+    # GH 28668
+    right = DataFrame(
+        {"a": Series([pd.Interval(0, 1), pd.Interval(1, 2)], dtype="interval")}
+    )
+    left = right.copy()
+    left["a"] = left["a"].astype("category")
+
+    result = merge(left, right, how="inner", on="a")
+    expected = right.copy()
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_multiindex_columns():
+    # Issue #28518
+    # Verify that merging two dataframes give the expected labels
+    # The original cause of this issue come from a bug lexsort_depth and is tested in
+    # test_lexsort_depth
+
+    letters = ["a", "b", "c", "d"]
+    numbers = ["1", "2", "3"]
+    index = MultiIndex.from_product((letters, numbers), names=["outer", "inner"])
+
+    frame_x = DataFrame(columns=index)
+    frame_x["id"] = ""
+    frame_y = DataFrame(columns=index)
+    frame_y["id"] = ""
+
+    l_suf = "_x"
+    r_suf = "_y"
+    result = frame_x.merge(frame_y, on="id", suffixes=((l_suf, r_suf)))
+
+    # Constructing the expected results
+    tuples = [(letter + l_suf, num) for letter in letters for num in numbers]
+    tuples += [("id", "")]
+    tuples += [(letter + r_suf, num) for letter in letters for num in numbers]
+
+    expected_index = MultiIndex.from_tuples(tuples, names=["outer", "inner"])
+    expected = DataFrame(columns=expected_index)
+
+    tm.assert_frame_equal(result, expected, check_dtype=False)
+
+
+def test_merge_datetime_upcast_dtype():
+    # https://github.com/pandas-dev/pandas/issues/31208
+    df1 = DataFrame({"x": ["a", "b", "c"], "y": ["1", "2", "4"]})
+    df2 = DataFrame(
+        {"y": ["1", "2", "3"], "z": pd.to_datetime(["2000", "2001", "2002"])}
+    )
+    result = merge(df1, df2, how="left", on="y")
+    expected = DataFrame(
+        {
+            "x": ["a", "b", "c"],
+            "y": ["1", "2", "4"],
+            "z": pd.to_datetime(["2000", "2001", "NaT"]),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("n_categories", [5, 128])
+def test_categorical_non_unique_monotonic(n_categories):
+    # GH 28189
+    # With n_categories as 5, we test the int8 case is hit in libjoin,
+    # with n_categories as 128 we test the int16 case.
+    left_index = CategoricalIndex([0] + list(range(n_categories)))
+    df1 = DataFrame(range(n_categories + 1), columns=["value"], index=left_index)
+    df2 = DataFrame(
+        [[6]],
+        columns=["value"],
+        index=CategoricalIndex([0], categories=list(range(n_categories))),
+    )
+
+    result = merge(df1, df2, how="left", left_index=True, right_index=True)
+    expected = DataFrame(
+        [[i, 6.0] if i < 2 else [i, np.nan] for i in range(n_categories + 1)],
+        columns=["value_x", "value_y"],
+        index=left_index,
+    )
+    tm.assert_frame_equal(expected, result)
+
+
+def test_merge_join_categorical_multiindex():
+    # From issue 16627
+    a = {
+        "Cat1": Categorical(["a", "b", "a", "c", "a", "b"], ["a", "b", "c"]),
+        "Int1": [0, 1, 0, 1, 0, 0],
+    }
+    a = DataFrame(a)
+
+    b = {
+        "Cat": Categorical(["a", "b", "c", "a", "b", "c"], ["a", "b", "c"]),
+        "Int": [0, 0, 0, 1, 1, 1],
+        "Factor": [1.1, 1.2, 1.3, 1.4, 1.5, 1.6],
+    }
+    b = DataFrame(b).set_index(["Cat", "Int"])["Factor"]
+
+    expected = merge(
+        a,
+        b.reset_index(),
+        left_on=["Cat1", "Int1"],
+        right_on=["Cat", "Int"],
+        how="left",
+    )
+    expected = expected.drop(["Cat", "Int"], axis=1)
+    result = a.join(b, on=["Cat1", "Int1"])
+    tm.assert_frame_equal(expected, result)
+
+    # Same test, but with ordered categorical
+    a = {
+        "Cat1": Categorical(
+            ["a", "b", "a", "c", "a", "b"], ["b", "a", "c"], ordered=True
+        ),
+        "Int1": [0, 1, 0, 1, 0, 0],
+    }
+    a = DataFrame(a)
+
+    b = {
+        "Cat": Categorical(
+            ["a", "b", "c", "a", "b", "c"], ["b", "a", "c"], ordered=True
+        ),
+        "Int": [0, 0, 0, 1, 1, 1],
+        "Factor": [1.1, 1.2, 1.3, 1.4, 1.5, 1.6],
+    }
+    b = DataFrame(b).set_index(["Cat", "Int"])["Factor"]
+
+    expected = merge(
+        a,
+        b.reset_index(),
+        left_on=["Cat1", "Int1"],
+        right_on=["Cat", "Int"],
+        how="left",
+    )
+    expected = expected.drop(["Cat", "Int"], axis=1)
+    result = a.join(b, on=["Cat1", "Int1"])
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize("func", ["merge", "merge_asof"])
+@pytest.mark.parametrize(
+    ("kwargs", "err_msg"),
+    [
+        ({"left_on": "a", "left_index": True}, ["left_on", "left_index"]),
+        ({"right_on": "a", "right_index": True}, ["right_on", "right_index"]),
+    ],
+)
+def test_merge_join_cols_error_reporting_duplicates(func, kwargs, err_msg):
+    # GH: 16228
+    left = DataFrame({"a": [1, 2], "b": [3, 4]})
+    right = DataFrame({"a": [1, 1], "c": [5, 6]})
+    msg = rf'Can only pass argument "{err_msg[0]}" OR "{err_msg[1]}" not both\.'
+    with pytest.raises(MergeError, match=msg):
+        getattr(pd, func)(left, right, **kwargs)
+
+
+@pytest.mark.parametrize("func", ["merge", "merge_asof"])
+@pytest.mark.parametrize(
+    ("kwargs", "err_msg"),
+    [
+        ({"left_on": "a"}, ["right_on", "right_index"]),
+        ({"right_on": "a"}, ["left_on", "left_index"]),
+    ],
+)
+def test_merge_join_cols_error_reporting_missing(func, kwargs, err_msg):
+    # GH: 16228
+    left = DataFrame({"a": [1, 2], "b": [3, 4]})
+    right = DataFrame({"a": [1, 1], "c": [5, 6]})
+    msg = rf'Must pass "{err_msg[0]}" OR "{err_msg[1]}"\.'
+    with pytest.raises(MergeError, match=msg):
+        getattr(pd, func)(left, right, **kwargs)
+
+
+@pytest.mark.parametrize("func", ["merge", "merge_asof"])
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"right_index": True},
+        {"left_index": True},
+    ],
+)
+def test_merge_join_cols_error_reporting_on_and_index(func, kwargs):
+    # GH: 16228
+    left = DataFrame({"a": [1, 2], "b": [3, 4]})
+    right = DataFrame({"a": [1, 1], "c": [5, 6]})
+    msg = (
+        r'Can only pass argument "on" OR "left_index" '
+        r'and "right_index", not a combination of both\.'
+    )
+    with pytest.raises(MergeError, match=msg):
+        getattr(pd, func)(left, right, on="a", **kwargs)
+
+
+def test_merge_right_left_index():
+    # GH#38616
+    left = DataFrame({"x": [1, 1], "z": ["foo", "foo"]})
+    right = DataFrame({"x": [1, 1], "z": ["foo", "foo"]})
+    result = merge(left, right, how="right", left_index=True, right_on="x")
+    expected = DataFrame(
+        {
+            "x": [1, 1],
+            "x_x": [1, 1],
+            "z_x": ["foo", "foo"],
+            "x_y": [1, 1],
+            "z_y": ["foo", "foo"],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_result_empty_index_and_on():
+    # GH#33814
+    df1 = DataFrame({"a": [1], "b": [2]}).set_index(["a", "b"])
+    df2 = DataFrame({"b": [1]}).set_index(["b"])
+    expected = DataFrame({"a": [], "b": []}, dtype=np.int64).set_index(["a", "b"])
+    result = merge(df1, df2, left_on=["b"], right_index=True)
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(df2, df1, left_index=True, right_on=["b"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_suffixes_produce_dup_columns_raises():
+    # GH#22818; Enforced in 2.0
+    left = DataFrame({"a": [1, 2, 3], "b": 1, "b_x": 2})
+    right = DataFrame({"a": [1, 2, 3], "b": 2})
+
+    with pytest.raises(MergeError, match="Passing 'suffixes' which cause duplicate"):
+        merge(left, right, on="a")
+
+    with pytest.raises(MergeError, match="Passing 'suffixes' which cause duplicate"):
+        merge(right, left, on="a", suffixes=("_y", "_x"))
+
+
+def test_merge_duplicate_columns_with_suffix_no_warning():
+    # GH#22818
+    # Do not raise warning when duplicates are caused by duplicates in origin
+    left = DataFrame([[1, 1, 1], [2, 2, 2]], columns=["a", "b", "b"])
+    right = DataFrame({"a": [1, 3], "b": 2})
+    result = merge(left, right, on="a")
+    expected = DataFrame([[1, 1, 1, 2]], columns=["a", "b_x", "b_x", "b_y"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_duplicate_columns_with_suffix_causing_another_duplicate_raises():
+    # GH#22818, Enforced in 2.0
+    # This should raise warning because suffixes cause another collision
+    left = DataFrame([[1, 1, 1, 1], [2, 2, 2, 2]], columns=["a", "b", "b", "b_x"])
+    right = DataFrame({"a": [1, 3], "b": 2})
+    with pytest.raises(MergeError, match="Passing 'suffixes' which cause duplicate"):
+        merge(left, right, on="a")
+
+
+def test_merge_string_float_column_result():
+    # GH 13353
+    df1 = DataFrame([[1, 2], [3, 4]], columns=Index(["a", 114.0]))
+    df2 = DataFrame([[9, 10], [11, 12]], columns=["x", "y"])
+    result = merge(df2, df1, how="inner", left_index=True, right_index=True)
+    expected = DataFrame(
+        [[9, 10, 1, 2], [11, 12, 3, 4]], columns=Index(["x", "y", "a", 114.0])
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_mergeerror_on_left_index_mismatched_dtypes():
+    # GH 22449
+    df_1 = DataFrame(data=["X"], columns=["C"], index=[22])
+    df_2 = DataFrame(data=["X"], columns=["C"], index=[999])
+    with pytest.raises(MergeError, match="Can only pass argument"):
+        merge(df_1, df_2, on=["C"], left_index=True)
+
+
+def test_merge_on_left_categoricalindex():
+    # GH#48464 don't raise when left_on is a CategoricalIndex
+    ci = CategoricalIndex(range(3))
+
+    right = DataFrame({"A": ci, "B": range(3)})
+    left = DataFrame({"C": range(3, 6)})
+
+    res = merge(left, right, left_on=ci, right_on="A")
+    expected = merge(left, right, left_on=ci._data, right_on="A")
+    tm.assert_frame_equal(res, expected)
+
+
+@pytest.mark.parametrize("dtype", [None, "Int64"])
+def test_merge_outer_with_NaN(dtype):
+    # GH#43550
+    left = DataFrame({"key": [1, 2], "col1": [1, 2]}, dtype=dtype)
+    right = DataFrame({"key": [np.nan, np.nan], "col2": [3, 4]}, dtype=dtype)
+    result = merge(left, right, on="key", how="outer")
+    expected = DataFrame(
+        {
+            "key": [1, 2, np.nan, np.nan],
+            "col1": [1, 2, np.nan, np.nan],
+            "col2": [np.nan, np.nan, 3, 4],
+        },
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # switch left and right
+    result = merge(right, left, on="key", how="outer")
+    expected = DataFrame(
+        {
+            "key": [1, 2, np.nan, np.nan],
+            "col2": [np.nan, np.nan, 3, 4],
+            "col1": [1, 2, np.nan, np.nan],
+        },
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_different_index_names():
+    # GH#45094
+    left = DataFrame({"a": [1]}, index=Index([1], name="c"))
+    right = DataFrame({"a": [1]}, index=Index([1], name="d"))
+    result = merge(left, right, left_on="c", right_on="d")
+    expected = DataFrame({"a_x": [1], "a_y": 1})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_ea(any_numeric_ea_dtype, join_type):
+    # GH#44240
+    left = DataFrame({"a": [1, 2, 3], "b": 1}, dtype=any_numeric_ea_dtype)
+    right = DataFrame({"a": [1, 2, 3], "c": 2}, dtype=any_numeric_ea_dtype)
+    result = left.merge(right, how=join_type)
+    expected = DataFrame({"a": [1, 2, 3], "b": 1, "c": 2}, dtype=any_numeric_ea_dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_ea_and_non_ea(any_numeric_ea_dtype, join_type):
+    # GH#44240
+    left = DataFrame({"a": [1, 2, 3], "b": 1}, dtype=any_numeric_ea_dtype)
+    right = DataFrame({"a": [1, 2, 3], "c": 2}, dtype=any_numeric_ea_dtype.lower())
+    result = left.merge(right, how=join_type)
+    expected = DataFrame(
+        {
+            "a": Series([1, 2, 3], dtype=any_numeric_ea_dtype),
+            "b": Series([1, 1, 1], dtype=any_numeric_ea_dtype),
+            "c": Series([2, 2, 2], dtype=any_numeric_ea_dtype.lower()),
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["int64", "int64[pyarrow]"])
+def test_merge_arrow_and_numpy_dtypes(dtype):
+    # GH#52406
+    pytest.importorskip("pyarrow")
+    df = DataFrame({"a": [1, 2]}, dtype=dtype)
+    df2 = DataFrame({"a": [1, 2]}, dtype="int64[pyarrow]")
+    result = df.merge(df2)
+    expected = df.copy()
+    tm.assert_frame_equal(result, expected)
+
+    result = df2.merge(df)
+    expected = df2.copy()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("tz", [None, "America/Chicago"])
+def test_merge_datetime_different_resolution(tz, join_type):
+    # https://github.com/pandas-dev/pandas/issues/53200
+    vals = [
+        pd.Timestamp(2023, 5, 12, tz=tz),
+        pd.Timestamp(2023, 5, 13, tz=tz),
+        pd.Timestamp(2023, 5, 14, tz=tz),
+    ]
+    df1 = DataFrame({"t": vals[:2], "a": [1.0, 2.0]})
+    df1["t"] = df1["t"].dt.as_unit("ns")
+    df2 = DataFrame({"t": vals[1:], "b": [1.0, 2.0]})
+    df2["t"] = df2["t"].dt.as_unit("s")
+
+    expected = DataFrame({"t": vals, "a": [1.0, 2.0, np.nan], "b": [np.nan, 1.0, 2.0]})
+    expected["t"] = expected["t"].dt.as_unit("ns")
+    if join_type == "inner":
+        expected = expected.iloc[[1]].reset_index(drop=True)
+    elif join_type == "left":
+        expected = expected.iloc[[0, 1]]
+    elif join_type == "right":
+        expected = expected.iloc[[1, 2]].reset_index(drop=True)
+
+    result = df1.merge(df2, on="t", how=join_type)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_multiindex_single_level():
+    # GH52331
+    df = DataFrame({"col": ["A", "B"]})
+    df2 = DataFrame(
+        data={"b": [100]},
+        index=MultiIndex.from_tuples([("A",), ("C",)], names=["col"]),
+    )
+    expected = DataFrame({"col": ["A", "B"], "b": [100, np.nan]})
+
+    result = df.merge(df2, left_on=["col"], right_index=True, how="left")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("on_index", [True, False])
+@pytest.mark.parametrize("left_unique", [True, False])
+@pytest.mark.parametrize("left_monotonic", [True, False])
+@pytest.mark.parametrize("right_unique", [True, False])
+@pytest.mark.parametrize("right_monotonic", [True, False])
+def test_merge_combinations(
+    join_type,
+    sort,
+    on_index,
+    left_unique,
+    left_monotonic,
+    right_unique,
+    right_monotonic,
+):
+    how = join_type
+    # GH 54611
+    left = [2, 3]
+    if left_unique:
+        left.append(4 if left_monotonic else 1)
+    else:
+        left.append(3 if left_monotonic else 2)
+
+    right = [2, 3]
+    if right_unique:
+        right.append(4 if right_monotonic else 1)
+    else:
+        right.append(3 if right_monotonic else 2)
+
+    left = DataFrame({"key": left})
+    right = DataFrame({"key": right})
+
+    if on_index:
+        left = left.set_index("key")
+        right = right.set_index("key")
+        on_kwargs = {"left_index": True, "right_index": True}
+    else:
+        on_kwargs = {"on": "key"}
+
+    result = merge(left, right, how=how, sort=sort, **on_kwargs)
+
+    if on_index:
+        left = left.reset_index()
+        right = right.reset_index()
+
+    if how in ["left", "right", "inner"]:
+        if how in ["left", "inner"]:
+            expected, other, other_unique = left, right, right_unique
+        else:
+            expected, other, other_unique = right, left, left_unique
+        if how == "inner":
+            keep_values = set(left["key"].values).intersection(right["key"].values)
+            keep_mask = expected["key"].isin(keep_values)
+            expected = expected[keep_mask]
+        if sort:
+            expected = expected.sort_values("key")
+        if not other_unique:
+            other_value_counts = other["key"].value_counts()
+            repeats = other_value_counts.reindex(expected["key"].values, fill_value=1)
+            repeats = repeats.astype(np.intp)
+            expected = expected["key"].repeat(repeats.values)
+            expected = expected.to_frame()
+    elif how == "outer":
+        left_counts = left["key"].value_counts()
+        right_counts = right["key"].value_counts()
+        expected_counts = left_counts.mul(right_counts, fill_value=1)
+        expected_counts = expected_counts.astype(np.intp)
+        expected = expected_counts.index.values.repeat(expected_counts.values)
+        expected = DataFrame({"key": expected})
+        expected = expected.sort_values("key")
+
+    if on_index:
+        expected = expected.set_index("key")
+    else:
+        expected = expected.reset_index(drop=True)
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_ea_int_and_float_numpy():
+    # GH#46178
+    df1 = DataFrame([1.0, np.nan], dtype=pd.Int64Dtype())
+    df2 = DataFrame([1.5])
+    expected = DataFrame(columns=[0], dtype="Int64")
+
+    with tm.assert_produces_warning(UserWarning, match="You are merging"):
+        result = df1.merge(df2)
+    tm.assert_frame_equal(result, expected)
+
+    with tm.assert_produces_warning(UserWarning, match="You are merging"):
+        result = df2.merge(df1)
+    tm.assert_frame_equal(result, expected.astype("float64"))
+
+    df2 = DataFrame([1.0])
+    expected = DataFrame([1], columns=[0], dtype="Int64")
+    result = df1.merge(df2)
+    tm.assert_frame_equal(result, expected)
+
+    result = df2.merge(df1)
+    tm.assert_frame_equal(result, expected.astype("float64"))
+
+
+def test_merge_arrow_string_index(any_string_dtype):
+    # GH#54894
+    pytest.importorskip("pyarrow")
+    left = DataFrame({"a": ["a", "b"]}, dtype=any_string_dtype)
+    right = DataFrame({"b": 1}, index=Index(["a", "c"], dtype=any_string_dtype))
+    result = left.merge(right, left_on="a", right_index=True, how="left")
+    expected = DataFrame(
+        {"a": Series(["a", "b"], dtype=any_string_dtype), "b": [1, np.nan]}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("left_empty", [True, False])
+@pytest.mark.parametrize("right_empty", [True, False])
+def test_merge_empty_frames_column_order(left_empty, right_empty):
+    # GH 51929
+    df1 = DataFrame(1, index=[0], columns=["A", "B"])
+    df2 = DataFrame(1, index=[0], columns=["A", "C", "D"])
+
+    if left_empty:
+        df1 = df1.iloc[:0]
+    if right_empty:
+        df2 = df2.iloc[:0]
+
+    result = merge(df1, df2, on=["A"], how="outer")
+    expected = DataFrame(1, index=range(1), columns=["A", "B", "C", "D"])
+    if left_empty and right_empty:
+        expected = expected.iloc[:0]
+    elif left_empty:
+        expected["B"] = np.nan
+    elif right_empty:
+        expected[["C", "D"]] = np.nan
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("how", ["left", "right", "inner", "outer"])
+def test_merge_datetime_and_timedelta(how):
+    left = DataFrame({"key": Series([1, None], dtype="datetime64[ns]")})
+    right = DataFrame({"key": Series([1], dtype="timedelta64[ns]")})
+
+    msg = (
+        f"You are trying to merge on {left['key'].dtype} and {right['key'].dtype} "
+        "columns for key 'key'. If you wish to proceed you should use pd.concat"
+    )
+    with pytest.raises(ValueError, match=re.escape(msg)):
+        left.merge(right, on="key", how=how)
+
+    msg = (
+        f"You are trying to merge on {right['key'].dtype} and {left['key'].dtype} "
+        "columns for key 'key'. If you wish to proceed you should use pd.concat"
+    )
+    with pytest.raises(ValueError, match=re.escape(msg)):
+        right.merge(left, on="key", how=how)
+
+
+def test_merge_on_all_nan_column():
+    # GH#59421
+    left = DataFrame({"x": [1, 2, 3], "y": [np.nan, np.nan, np.nan], "z": [4, 5, 6]})
+    right = DataFrame({"x": [1, 2, 3], "y": [np.nan, np.nan, np.nan], "zz": [4, 5, 6]})
+    result = left.merge(right, on=["x", "y"], how="outer")
+    # Should not trigger array bounds eerror with bounds checking or asan enabled.
+    expected = DataFrame(
+        {"x": [1, 2, 3], "y": [np.nan, np.nan, np.nan], "z": [4, 5, 6], "zz": [4, 5, 6]}
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge_antijoin.py b/pandas/pandas/tests/reshape/merge/test_merge_antijoin.py
new file mode 100644
index 0000000000000..006622c6e5e94
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge_antijoin.py
@@ -0,0 +1,280 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+)
+import pandas._testing as tm
+from pandas.core.reshape.merge import merge
+
+
+def test_merge_antijoin():
+    # GH#42916
+    left = DataFrame({"A": [1, 2, 3]}, index=["a", "b", "c"])
+    right = DataFrame({"B": [1, 2, 4]}, index=["a", "b", "d"])
+
+    result = merge(left, right, how="left_anti", left_index=True, right_index=True)
+    expected = DataFrame({"A": [3], "B": [np.nan]}, index=["c"])
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_index=True, right_index=True)
+    expected = DataFrame({"A": [np.nan], "B": [4]}, index=["d"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_on_different_columns():
+    left = DataFrame({"A": [1.0, 2.0, 3.0], "B": ["a", "b", "c"]}).astype({"B": object})
+    right = DataFrame({"C": [1.0, 2.0, 4.0], "D": ["a", "d", "b"]}).astype(
+        {"D": object}
+    )
+
+    result = merge(left, right, how="left_anti", left_on="B", right_on="D")
+    expected = DataFrame(
+        {
+            "A": [3.0],
+            "B": ["c"],
+            "C": [np.nan],
+            "D": [np.nan],
+        },
+        index=[2],
+    ).astype({"B": object, "D": object})
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="B", right_on="D")
+    expected = DataFrame(
+        {
+            "A": [np.nan],
+            "B": [np.nan],
+            "C": [2.0],
+            "D": ["d"],
+        },
+        index=[1],
+    ).astype({"B": object, "D": object})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_nonunique_keys():
+    left = DataFrame({"A": [1.0, 2.0, 3.0], "B": ["a", "b", "b"]}).astype({"B": object})
+    right = DataFrame({"C": [1.0, 2.0, 4.0], "D": ["b", "d", "d"]}).astype(
+        {"D": object}
+    )
+
+    result = merge(left, right, how="left_anti", left_on="B", right_on="D")
+    expected = DataFrame(
+        {
+            "A": [1.0],
+            "B": ["a"],
+            "C": [np.nan],
+            "D": [np.nan],
+        },
+        index=[0],
+    ).astype({"B": object, "D": object})
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="B", right_on="D")
+    expected = DataFrame(
+        {
+            "A": [np.nan, np.nan],
+            "B": [np.nan, np.nan],
+            "C": [2.0, 4.0],
+            "D": ["d", "d"],
+        },
+        index=[2, 3],
+    ).astype({"B": object, "D": object})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_same_df():
+    left = DataFrame({"A": [1, 2, 3]}, index=["a", "b", "c"], dtype=np.int64)
+    result = merge(left, left, how="left_anti", left_index=True, right_index=True)
+    expected = DataFrame([], columns=["A_x", "A_y"], dtype=np.int64)
+    tm.assert_frame_equal(result, expected, check_index_type=False)
+
+
+def test_merge_antijoin_nans():
+    left = DataFrame({"A": [1.0, 2.0, np.nan], "C": ["a", "b", "c"]}).astype(
+        {"C": object}
+    )
+    right = DataFrame({"A": [3.0, 2.0, np.nan], "D": ["d", "e", "f"]}).astype(
+        {"D": object}
+    )
+    result = merge(left, right, how="left_anti", on="A")
+    expected = DataFrame({"A": [1.0], "C": ["a"], "D": [np.nan]}).astype(
+        {"C": object, "D": object}
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_on_datetime64tz():
+    # GH11405
+    left = DataFrame(
+        {
+            "key": pd.date_range("20151010", periods=2, tz="US/Eastern"),
+            "value": [1.0, 2.0],
+        }
+    )
+    right = DataFrame(
+        {
+            "key": pd.date_range("20151011", periods=3, tz="US/Eastern"),
+            "value": [1.0, 2.0, 3.0],
+        }
+    )
+
+    expected = DataFrame(
+        {
+            "key": pd.date_range("20151010", periods=1, tz="US/Eastern"),
+            "value_x": [1.0],
+            "value_y": [np.nan],
+        },
+        index=[0],
+    )
+    result = merge(left, right, on="key", how="left_anti")
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame(
+        {
+            "key": pd.date_range("20151012", periods=2, tz="US/Eastern"),
+            "value_x": [np.nan, np.nan],
+            "value_y": [2.0, 3.0],
+        },
+        index=[1, 2],
+    )
+    result = merge(left, right, on="key", how="right_anti")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_multiindex():
+    left = DataFrame(
+        {
+            "A": [1, 2, 3],
+            "B": [4, 5, 6],
+        },
+        index=MultiIndex.from_tuples(
+            [("a", "x"), ("b", "y"), ("c", "z")], names=["first", "second"]
+        ),
+    )
+    right = DataFrame(
+        {
+            "C": [7, 8, 9],
+            "D": [10, 11, 12],
+        },
+        index=MultiIndex.from_tuples(
+            [("a", "x"), ("b", "y"), ("c", "w")], names=["first", "second"]
+        ),
+    )
+
+    result = merge(left, right, how="left_anti", left_index=True, right_index=True)
+    expected = DataFrame(
+        {
+            "A": [3],
+            "B": [6],
+            "C": [np.nan],
+            "D": [np.nan],
+        },
+        index=MultiIndex.from_tuples([("c", "z")], names=["first", "second"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_index=True, right_index=True)
+    expected = DataFrame(
+        {
+            "A": [np.nan],
+            "B": [np.nan],
+            "C": [9],
+            "D": [12],
+        },
+        index=MultiIndex.from_tuples([("c", "w")], names=["first", "second"]),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "Int64",
+        pytest.param("int64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        pytest.param("timestamp[s][pyarrow]", marks=td.skip_if_no("pyarrow")),
+        pytest.param("string[pyarrow]", marks=td.skip_if_no("pyarrow")),
+    ],
+)
+def test_merge_antijoin_extension_dtype(dtype):
+    left = DataFrame(
+        {
+            "join_col": [1, 3, 5],
+            "left_val": [1, 2, 3],
+        }
+    )
+    right = DataFrame(
+        {
+            "join_col": [2, 3, 4],
+            "right_val": [1, 2, 3],
+        }
+    )
+    left = left.astype({"join_col": dtype})
+    right = right.astype({"join_col": dtype})
+    result = merge(left, right, how="left_anti", on="join_col")
+    expected = DataFrame(
+        {
+            "join_col": [1, 5],
+            "left_val": [1, 3],
+            "right_val": [np.nan, np.nan],
+        },
+        index=[0, 2],
+    )
+    expected = expected.astype({"join_col": dtype})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_empty_dataframe():
+    left = DataFrame({"A": [], "B": []})
+    right = DataFrame({"C": [], "D": []})
+
+    result = merge(left, right, how="left_anti", left_on="A", right_on="C")
+    expected = DataFrame({"A": [], "B": [], "C": [], "D": []})
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="A", right_on="C")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_no_common_elements():
+    left = DataFrame({"A": [1, 2, 3]})
+    right = DataFrame({"B": [4, 5, 6]})
+
+    result = merge(left, right, how="left_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [1, 2, 3], "B": [np.nan, np.nan, np.nan]})
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [np.nan, np.nan, np.nan], "B": [4, 5, 6]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_with_null_values():
+    left = DataFrame({"A": [1.0, 2.0, None, 4.0]})
+    right = DataFrame({"B": [2.0, None, 5.0]})
+
+    result = merge(left, right, how="left_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [1.0, 4.0], "B": [np.nan, np.nan]}, index=[0, 3])
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [np.nan], "B": [5.0]}, index=[2])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_antijoin_with_mixed_dtypes():
+    left = DataFrame({"A": [1, "2", 3.0]})
+    right = DataFrame({"B": ["2", 3.0, 4]})
+
+    result = merge(left, right, how="left_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [1], "B": [np.nan]}, dtype=object)
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(left, right, how="right_anti", left_on="A", right_on="B")
+    expected = DataFrame({"A": [np.nan], "B": [4]}, dtype=object, index=[2])
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge_asof.py b/pandas/pandas/tests/reshape/merge/test_merge_asof.py
new file mode 100644
index 0000000000000..f7b0876c5a605
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge_asof.py
@@ -0,0 +1,3635 @@
+import datetime
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Index,
+    Timedelta,
+    merge_asof,
+    option_context,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.reshape.merge import MergeError
+
+
+class TestAsOfMerge:
+    def prep_data(self, df, dedupe=False):
+        if dedupe:
+            df = df.drop_duplicates(["time", "ticker"], keep="last").reset_index(
+                drop=True
+            )
+        df.time = to_datetime(df.time)
+        return df
+
+    @pytest.fixture
+    def trades(self):
+        df = pd.DataFrame(
+            [
+                ["20160525 13:30:00.023", "MSFT", "51.9500", "75", "NASDAQ"],
+                ["20160525 13:30:00.038", "MSFT", "51.9500", "155", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.7700", "100", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9200", "100", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "200", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "300", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "600", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "44", "NASDAQ"],
+                ["20160525 13:30:00.074", "AAPL", "98.6700", "478343", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6700", "478343", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6600", "6", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "30", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "75", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "20", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "35", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "10", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.5500", "6", "ARCA"],
+                ["20160525 13:30:00.075", "AAPL", "98.5500", "6", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "1000", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "200", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "300", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "400", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "600", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "200", "ARCA"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "783", "NASDAQ"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "100", "NASDAQ"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "100", "NASDAQ"],
+            ],
+            columns="time,ticker,price,quantity,marketCenter".split(","),
+        )
+        df["price"] = df["price"].astype("float64")
+        df["quantity"] = df["quantity"].astype("int64")
+        return self.prep_data(df)
+
+    @pytest.fixture
+    def quotes(self):
+        df = pd.DataFrame(
+            [
+                ["20160525 13:30:00.023", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.023", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.041", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.072", "GOOG", "720.50", "720.88"],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.92", "51.95"],
+            ],
+            columns="time,ticker,bid,ask".split(","),
+        )
+        df["bid"] = df["bid"].astype("float64")
+        df["ask"] = df["ask"].astype("float64")
+        return self.prep_data(df, dedupe=True)
+
+    @pytest.fixture
+    def asof(self):
+        df = pd.DataFrame(
+            [
+                [
+                    "20160525 13:30:00.023",
+                    "MSFT",
+                    "51.95",
+                    "75",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.038",
+                    "MSFT",
+                    "51.95",
+                    "155",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.77",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.92",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "200",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "300",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "600",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "44",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.074",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.66",
+                    "6",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "30",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "75",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "20",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "35",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "10",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "1000",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "300",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "400",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "600",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "783",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+            ],
+            columns="time,ticker,price,quantity,marketCenter,bid,ask".split(","),
+        )
+        df["price"] = df["price"].astype("float64")
+        df["quantity"] = df["quantity"].astype("int64")
+        df["bid"] = df["bid"].astype("float64")
+        df["ask"] = df["ask"].astype("float64")
+        return self.prep_data(df)
+
+    @pytest.fixture
+    def tolerance(self):
+        df = pd.DataFrame(
+            [
+                [
+                    "20160525 13:30:00.023",
+                    "MSFT",
+                    "51.95",
+                    "75",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.038",
+                    "MSFT",
+                    "51.95",
+                    "155",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.77",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.92",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "200",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "300",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "600",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "44",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.074",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.66",
+                    "6",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "30",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "75",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "20",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "35",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "10",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "1000",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "300",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "400",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "600",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "783",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+            ],
+            columns="time,ticker,price,quantity,marketCenter,bid,ask".split(","),
+        )
+        df["price"] = df["price"].astype("float64")
+        df["quantity"] = df["quantity"].astype("int64")
+        df["bid"] = df["bid"].astype("float64")
+        df["ask"] = df["ask"].astype("float64")
+        return self.prep_data(df)
+
+    def test_examples1(self):
+        """doc-string examples"""
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 6, 7], "right_val": [1, 2, 3, 6, 7]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [1, 3, 7]}
+        )
+
+        result = merge_asof(left, right, on="a")
+        tm.assert_frame_equal(result, expected)
+
+    def test_examples2(self, unit):
+        """doc-string examples"""
+        if unit == "s":
+            pytest.skip(
+                "This test is invalid for unit='s' because that would "
+                "round the trades['time']]"
+            )
+        trades = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.038",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.048",
+                    ]
+                ).astype(f"M8[{unit}]"),
+                "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+            },
+            columns=["time", "ticker", "price", "quantity"],
+        )
+
+        quotes = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.030",
+                        "20160525 13:30:00.041",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.049",
+                        "20160525 13:30:00.072",
+                        "20160525 13:30:00.075",
+                    ]
+                ).astype(f"M8[{unit}]"),
+                "ticker": [
+                    "GOOG",
+                    "MSFT",
+                    "MSFT",
+                    "MSFT",
+                    "GOOG",
+                    "AAPL",
+                    "GOOG",
+                    "MSFT",
+                ],
+                "bid": [720.50, 51.95, 51.97, 51.99, 720.50, 97.99, 720.50, 52.01],
+                "ask": [720.93, 51.96, 51.98, 52.00, 720.93, 98.01, 720.88, 52.03],
+            },
+            columns=["time", "ticker", "bid", "ask"],
+        )
+
+        merge_asof(trades, quotes, on="time", by="ticker")
+
+        merge_asof(trades, quotes, on="time", by="ticker", tolerance=Timedelta("2ms"))
+
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.038",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.048",
+                    ]
+                ).astype(f"M8[{unit}]"),
+                "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+                "bid": [np.nan, 51.97, np.nan, np.nan, np.nan],
+                "ask": [np.nan, 51.98, np.nan, np.nan, np.nan],
+            },
+            columns=["time", "ticker", "price", "quantity", "bid", "ask"],
+        )
+
+        result = merge_asof(
+            trades,
+            quotes,
+            on="time",
+            by="ticker",
+            tolerance=Timedelta("10ms"),
+            allow_exact_matches=False,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_examples3(self):
+        """doc-string examples"""
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 6, 7], "right_val": [1, 2, 3, 6, 7]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [1, 6, np.nan]}
+        )
+
+        result = merge_asof(left, right, on="a", direction="forward")
+        tm.assert_frame_equal(result, expected)
+
+    def test_examples4(self):
+        """doc-string examples"""
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 6, 7], "right_val": [1, 2, 3, 6, 7]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [1, 6, 7]}
+        )
+
+        result = merge_asof(left, right, on="a", direction="nearest")
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic(self, trades, asof, quotes):
+        expected = asof
+
+        result = merge_asof(trades, quotes, on="time", by="ticker")
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_categorical(self, trades, asof, quotes):
+        expected = asof
+        trades.ticker = trades.ticker.astype("category")
+        quotes.ticker = quotes.ticker.astype("category")
+        expected.ticker = expected.ticker.astype("category")
+
+        result = merge_asof(trades, quotes, on="time", by="ticker")
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_left_index(self, trades, asof, quotes):
+        # GH14253
+        expected = asof
+        trades = trades.set_index("time")
+
+        result = merge_asof(
+            trades, quotes, left_index=True, right_on="time", by="ticker"
+        )
+        # left-only index uses right"s index, oddly
+        expected.index = result.index
+        # time column appears after left"s columns
+        expected = expected[result.columns]
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_right_index(self, trades, asof, quotes):
+        expected = asof
+        quotes = quotes.set_index("time")
+
+        result = merge_asof(
+            trades, quotes, left_on="time", right_index=True, by="ticker"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_left_index_right_index(self, trades, asof, quotes):
+        expected = asof.set_index("time")
+        trades = trades.set_index("time")
+        quotes = quotes.set_index("time")
+
+        result = merge_asof(
+            trades, quotes, left_index=True, right_index=True, by="ticker"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_multi_index_left(self, trades, quotes):
+        # MultiIndex is prohibited
+        trades = trades.set_index(["time", "price"])
+        quotes = quotes.set_index("time")
+        with pytest.raises(MergeError, match="left can only have one index"):
+            merge_asof(trades, quotes, left_index=True, right_index=True)
+
+    def test_multi_index_right(self, trades, quotes):
+        # MultiIndex is prohibited
+        trades = trades.set_index("time")
+        quotes = quotes.set_index(["time", "bid"])
+        with pytest.raises(MergeError, match="right can only have one index"):
+            merge_asof(trades, quotes, left_index=True, right_index=True)
+
+    def test_on_and_index_left_on(self, trades, quotes):
+        # "on" parameter and index together is prohibited
+        trades = trades.set_index("time")
+        quotes = quotes.set_index("time")
+        msg = 'Can only pass argument "left_on" OR "left_index" not both.'
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades, quotes, left_on="price", left_index=True, right_index=True
+            )
+
+    def test_on_and_index_right_on(self, trades, quotes):
+        trades = trades.set_index("time")
+        quotes = quotes.set_index("time")
+        msg = 'Can only pass argument "right_on" OR "right_index" not both.'
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades, quotes, right_on="bid", left_index=True, right_index=True
+            )
+
+    def test_basic_left_by_right_by(self, trades, asof, quotes):
+        # GH14253
+        expected = asof
+
+        result = merge_asof(
+            trades, quotes, on="time", left_by="ticker", right_by="ticker"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_missing_right_by(self, trades, asof, quotes):
+        expected = asof
+
+        q = quotes[quotes.ticker != "MSFT"]
+        result = merge_asof(trades, q, on="time", by="ticker")
+        expected.loc[expected.ticker == "MSFT", ["bid", "ask"]] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    def test_multiby(self):
+        # GH13936
+        trades = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.046",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.050",
+                    ]
+                ),
+                "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+                "exch": ["ARCA", "NSDQ", "NSDQ", "BATS", "NSDQ"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+            },
+            columns=["time", "ticker", "exch", "price", "quantity"],
+        )
+
+        quotes = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.030",
+                        "20160525 13:30:00.041",
+                        "20160525 13:30:00.045",
+                        "20160525 13:30:00.049",
+                    ]
+                ),
+                "ticker": ["GOOG", "MSFT", "MSFT", "MSFT", "GOOG", "AAPL"],
+                "exch": ["BATS", "NSDQ", "ARCA", "ARCA", "NSDQ", "ARCA"],
+                "bid": [720.51, 51.95, 51.97, 51.99, 720.50, 97.99],
+                "ask": [720.92, 51.96, 51.98, 52.00, 720.93, 98.01],
+            },
+            columns=["time", "ticker", "exch", "bid", "ask"],
+        )
+
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.046",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.050",
+                    ]
+                ),
+                "ticker": ["MSFT", "MSFT", "GOOG", "GOOG", "AAPL"],
+                "exch": ["ARCA", "NSDQ", "NSDQ", "BATS", "NSDQ"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+                "bid": [np.nan, 51.95, 720.50, 720.51, np.nan],
+                "ask": [np.nan, 51.96, 720.93, 720.92, np.nan],
+            },
+            columns=["time", "ticker", "exch", "price", "quantity", "bid", "ask"],
+        )
+
+        result = merge_asof(trades, quotes, on="time", by=["ticker", "exch"])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["object", "string"])
+    def test_multiby_heterogeneous_types(self, dtype):
+        # GH13936
+        trades = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.046",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.050",
+                    ]
+                ),
+                "ticker": [0, 0, 1, 1, 2],
+                "exch": ["ARCA", "NSDQ", "NSDQ", "BATS", "NSDQ"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+            },
+            columns=["time", "ticker", "exch", "price", "quantity"],
+        )
+        trades = trades.astype({"ticker": dtype, "exch": dtype})
+
+        quotes = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.030",
+                        "20160525 13:30:00.041",
+                        "20160525 13:30:00.045",
+                        "20160525 13:30:00.049",
+                    ]
+                ),
+                "ticker": [1, 0, 0, 0, 1, 2],
+                "exch": ["BATS", "NSDQ", "ARCA", "ARCA", "NSDQ", "ARCA"],
+                "bid": [720.51, 51.95, 51.97, 51.99, 720.50, 97.99],
+                "ask": [720.92, 51.96, 51.98, 52.00, 720.93, 98.01],
+            },
+            columns=["time", "ticker", "exch", "bid", "ask"],
+        )
+        quotes = quotes.astype({"ticker": dtype, "exch": dtype})
+
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.023",
+                        "20160525 13:30:00.046",
+                        "20160525 13:30:00.048",
+                        "20160525 13:30:00.050",
+                    ]
+                ),
+                "ticker": [0, 0, 1, 1, 2],
+                "exch": ["ARCA", "NSDQ", "NSDQ", "BATS", "NSDQ"],
+                "price": [51.95, 51.95, 720.77, 720.92, 98.00],
+                "quantity": [75, 155, 100, 100, 100],
+                "bid": [np.nan, 51.95, 720.50, 720.51, np.nan],
+                "ask": [np.nan, 51.96, 720.93, 720.92, np.nan],
+            },
+            columns=["time", "ticker", "exch", "price", "quantity", "bid", "ask"],
+        )
+        expected = expected.astype({"ticker": dtype, "exch": dtype})
+
+        result = merge_asof(trades, quotes, on="time", by=["ticker", "exch"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_mismatched_index_dtype(self):
+        # similar to test_multiby_indexed, but we change the dtype on left.index
+        left = pd.DataFrame(
+            [
+                [to_datetime("20160602"), 1, "a"],
+                [to_datetime("20160602"), 2, "a"],
+                [to_datetime("20160603"), 1, "b"],
+                [to_datetime("20160603"), 2, "b"],
+            ],
+            columns=["time", "k1", "k2"],
+        ).set_index("time")
+        # different dtype for the index
+        left.index = left.index - pd.Timestamp(0)
+
+        right = pd.DataFrame(
+            [
+                [to_datetime("20160502"), 1, "a", 1.0],
+                [to_datetime("20160502"), 2, "a", 2.0],
+                [to_datetime("20160503"), 1, "b", 3.0],
+                [to_datetime("20160503"), 2, "b", 4.0],
+            ],
+            columns=["time", "k1", "k2", "value"],
+        ).set_index("time")
+
+        msg = "incompatible merge keys"
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(left, right, left_index=True, right_index=True, by=["k1", "k2"])
+
+    def test_multiby_indexed(self):
+        # GH15676
+        left = pd.DataFrame(
+            [
+                [to_datetime("20160602"), 1, "a"],
+                [to_datetime("20160602"), 2, "a"],
+                [to_datetime("20160603"), 1, "b"],
+                [to_datetime("20160603"), 2, "b"],
+            ],
+            columns=["time", "k1", "k2"],
+        ).set_index("time")
+
+        right = pd.DataFrame(
+            [
+                [to_datetime("20160502"), 1, "a", 1.0],
+                [to_datetime("20160502"), 2, "a", 2.0],
+                [to_datetime("20160503"), 1, "b", 3.0],
+                [to_datetime("20160503"), 2, "b", 4.0],
+            ],
+            columns=["time", "k1", "k2", "value"],
+        ).set_index("time")
+
+        expected = pd.DataFrame(
+            [
+                [to_datetime("20160602"), 1, "a", 1.0],
+                [to_datetime("20160602"), 2, "a", 2.0],
+                [to_datetime("20160603"), 1, "b", 3.0],
+                [to_datetime("20160603"), 2, "b", 4.0],
+            ],
+            columns=["time", "k1", "k2", "value"],
+        ).set_index("time")
+
+        result = merge_asof(
+            left, right, left_index=True, right_index=True, by=["k1", "k2"]
+        )
+
+        tm.assert_frame_equal(expected, result)
+
+        with pytest.raises(
+            MergeError, match="left_by and right_by must be the same length"
+        ):
+            merge_asof(
+                left,
+                right,
+                left_index=True,
+                right_index=True,
+                left_by=["k1", "k2"],
+                right_by=["k1"],
+            )
+
+    def test_basic2(self, datapath):
+        expected = pd.DataFrame(
+            [
+                [
+                    "20160525 13:30:00.023",
+                    "MSFT",
+                    "51.95",
+                    "75",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.038",
+                    "MSFT",
+                    "51.95",
+                    "155",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.77",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.92",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "200",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "300",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "600",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "44",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.074",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.66",
+                    "6",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "30",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "75",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "20",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "35",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "10",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.55",
+                    "6",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "1000",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "300",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "400",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "600",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "783",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.084",
+                    "AAPL",
+                    "98.64",
+                    "40",
+                    "NASDAQ",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.084",
+                    "AAPL",
+                    "98.55",
+                    "149",
+                    "EDGX",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.086",
+                    "AAPL",
+                    "98.56",
+                    "500",
+                    "ARCA",
+                    "98.55",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "647",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "300",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "50",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "50",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "70",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "70",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "1",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "62",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "10",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.104",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.105",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.105",
+                    "AAPL",
+                    "98.63",
+                    "700",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.106",
+                    "AAPL",
+                    "98.63",
+                    "61",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.107",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.107",
+                    "AAPL",
+                    "98.63",
+                    "53",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.108",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.108",
+                    "AAPL",
+                    "98.63",
+                    "839",
+                    "ARCA",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.115",
+                    "AAPL",
+                    "98.63",
+                    "5",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.118",
+                    "AAPL",
+                    "98.63",
+                    "295",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.118",
+                    "AAPL",
+                    "98.63",
+                    "5",
+                    "EDGX",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.128",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.128",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "NASDAQ",
+                    "98.62",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.128",
+                    "MSFT",
+                    "51.92",
+                    "100",
+                    "ARCA",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "100",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "10",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "59",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "31",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "69",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "12",
+                    "NASDAQ",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "12",
+                    "EDGX",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "100",
+                    "ARCA",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.129",
+                    "AAPL",
+                    "98.62",
+                    "100",
+                    "ARCA",
+                    "98.61",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.130",
+                    "MSFT",
+                    "51.95",
+                    "317",
+                    "ARCA",
+                    "51.93",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.130",
+                    "MSFT",
+                    "51.95",
+                    "283",
+                    "ARCA",
+                    "51.93",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.135",
+                    "MSFT",
+                    "51.93",
+                    "100",
+                    "EDGX",
+                    "51.92",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.135",
+                    "AAPL",
+                    "98.62",
+                    "100",
+                    "ARCA",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "12",
+                    "NASDAQ",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "88",
+                    "NASDAQ",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "162",
+                    "NASDAQ",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.61",
+                    "100",
+                    "BATS",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "61",
+                    "ARCA",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "25",
+                    "ARCA",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.144",
+                    "AAPL",
+                    "98.62",
+                    "14",
+                    "ARCA",
+                    "98.61",
+                    "98.62",
+                ],
+                [
+                    "20160525 13:30:00.145",
+                    "AAPL",
+                    "98.62",
+                    "12",
+                    "ARCA",
+                    "98.6",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.145",
+                    "AAPL",
+                    "98.62",
+                    "100",
+                    "ARCA",
+                    "98.6",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.145",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "NASDAQ",
+                    "98.6",
+                    "98.63",
+                ],
+                [
+                    "20160525 13:30:00.145",
+                    "AAPL",
+                    "98.63",
+                    "100",
+                    "NASDAQ",
+                    "98.6",
+                    "98.63",
+                ],
+            ],
+            columns="time,ticker,price,quantity,marketCenter,bid,ask".split(","),
+        )
+        expected["price"] = expected["price"].astype("float64")
+        expected["quantity"] = expected["quantity"].astype("int64")
+        expected["bid"] = expected["bid"].astype("float64")
+        expected["ask"] = expected["ask"].astype("float64")
+        expected = self.prep_data(expected)
+
+        trades = pd.DataFrame(
+            [
+                ["20160525 13:30:00.023", "MSFT", "51.9500", "75", "NASDAQ"],
+                ["20160525 13:30:00.038", "MSFT", "51.9500", "155", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.7700", "100", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9200", "100", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "200", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "300", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "600", "NASDAQ"],
+                ["20160525 13:30:00.048", "GOOG", "720.9300", "44", "NASDAQ"],
+                ["20160525 13:30:00.074", "AAPL", "98.6700", "478343", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6700", "478343", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6600", "6", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "30", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "75", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "20", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "35", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.6500", "10", "NASDAQ"],
+                ["20160525 13:30:00.075", "AAPL", "98.5500", "6", "ARCA"],
+                ["20160525 13:30:00.075", "AAPL", "98.5500", "6", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "1000", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "200", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "300", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "400", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "600", "ARCA"],
+                ["20160525 13:30:00.076", "AAPL", "98.5600", "200", "ARCA"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "783", "NASDAQ"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "100", "NASDAQ"],
+                ["20160525 13:30:00.078", "MSFT", "51.9500", "100", "NASDAQ"],
+                ["20160525 13:30:00.084", "AAPL", "98.6400", "40", "NASDAQ"],
+                ["20160525 13:30:00.084", "AAPL", "98.5500", "149", "EDGX"],
+                ["20160525 13:30:00.086", "AAPL", "98.5600", "500", "ARCA"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "647", "EDGX"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "300", "EDGX"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "50", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "50", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "70", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "70", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "1", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "62", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "10", "NASDAQ"],
+                ["20160525 13:30:00.104", "AAPL", "98.6300", "100", "ARCA"],
+                ["20160525 13:30:00.105", "AAPL", "98.6300", "100", "ARCA"],
+                ["20160525 13:30:00.105", "AAPL", "98.6300", "700", "ARCA"],
+                ["20160525 13:30:00.106", "AAPL", "98.6300", "61", "EDGX"],
+                ["20160525 13:30:00.107", "AAPL", "98.6300", "100", "ARCA"],
+                ["20160525 13:30:00.107", "AAPL", "98.6300", "53", "ARCA"],
+                ["20160525 13:30:00.108", "AAPL", "98.6300", "100", "ARCA"],
+                ["20160525 13:30:00.108", "AAPL", "98.6300", "839", "ARCA"],
+                ["20160525 13:30:00.115", "AAPL", "98.6300", "5", "EDGX"],
+                ["20160525 13:30:00.118", "AAPL", "98.6300", "295", "EDGX"],
+                ["20160525 13:30:00.118", "AAPL", "98.6300", "5", "EDGX"],
+                ["20160525 13:30:00.128", "AAPL", "98.6300", "100", "NASDAQ"],
+                ["20160525 13:30:00.128", "AAPL", "98.6300", "100", "NASDAQ"],
+                ["20160525 13:30:00.128", "MSFT", "51.9200", "100", "ARCA"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "100", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "10", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "59", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "31", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "69", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "12", "NASDAQ"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "12", "EDGX"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "100", "ARCA"],
+                ["20160525 13:30:00.129", "AAPL", "98.6200", "100", "ARCA"],
+                ["20160525 13:30:00.130", "MSFT", "51.9500", "317", "ARCA"],
+                ["20160525 13:30:00.130", "MSFT", "51.9500", "283", "ARCA"],
+                ["20160525 13:30:00.135", "MSFT", "51.9300", "100", "EDGX"],
+                ["20160525 13:30:00.135", "AAPL", "98.6200", "100", "ARCA"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "12", "NASDAQ"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "88", "NASDAQ"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "162", "NASDAQ"],
+                ["20160525 13:30:00.144", "AAPL", "98.6100", "100", "BATS"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "61", "ARCA"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "25", "ARCA"],
+                ["20160525 13:30:00.144", "AAPL", "98.6200", "14", "ARCA"],
+                ["20160525 13:30:00.145", "AAPL", "98.6200", "12", "ARCA"],
+                ["20160525 13:30:00.145", "AAPL", "98.6200", "100", "ARCA"],
+                ["20160525 13:30:00.145", "AAPL", "98.6300", "100", "NASDAQ"],
+                ["20160525 13:30:00.145", "AAPL", "98.6300", "100", "NASDAQ"],
+            ],
+            columns="time,ticker,price,quantity,marketCenter".split(","),
+        )
+        trades["price"] = trades["price"].astype("float64")
+        trades["quantity"] = trades["quantity"].astype("int64")
+        trades = self.prep_data(trades)
+
+        quotes = pd.DataFrame(
+            [
+                ["20160525 13:30:00.023", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.023", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.041", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.048", "GOOG", "720.50", "720.93"],
+                ["20160525 13:30:00.072", "GOOG", "720.50", "720.88"],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.076", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.95", "51.95"],
+                ["20160525 13:30:00.078", "MSFT", "51.92", "51.95"],
+                ["20160525 13:30:00.079", "MSFT", "51.92", "51.95"],
+                ["20160525 13:30:00.080", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.084", "AAPL", "98.55", "98.56"],
+                ["20160525 13:30:00.086", "AAPL", "98.55", "98.63"],
+                ["20160525 13:30:00.088", "AAPL", "98.65", "98.63"],
+                ["20160525 13:30:00.089", "AAPL", "98.63", "98.63"],
+                ["20160525 13:30:00.104", "AAPL", "98.63", "98.63"],
+                ["20160525 13:30:00.104", "AAPL", "98.63", "98.63"],
+                ["20160525 13:30:00.104", "AAPL", "98.63", "98.63"],
+                ["20160525 13:30:00.104", "AAPL", "98.63", "98.63"],
+                ["20160525 13:30:00.104", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.105", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.107", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.115", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.115", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.118", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.128", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.128", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.129", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.129", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.129", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.129", "AAPL", "98.62", "98.63"],
+                ["20160525 13:30:00.129", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.130", "MSFT", "51.93", "51.95"],
+                ["20160525 13:30:00.130", "MSFT", "51.93", "51.95"],
+                ["20160525 13:30:00.130", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.131", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.131", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.135", "MSFT", "51.92", "51.95"],
+                ["20160525 13:30:00.135", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.136", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.136", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.144", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.144", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.145", "AAPL", "98.61", "98.62"],
+                ["20160525 13:30:00.145", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.145", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.145", "AAPL", "98.60", "98.63"],
+                ["20160525 13:30:00.145", "AAPL", "98.61", "98.63"],
+                ["20160525 13:30:00.145", "AAPL", "98.60", "98.63"],
+            ],
+            columns="time,ticker,bid,ask".split(","),
+        )
+        quotes["bid"] = quotes["bid"].astype("float64")
+        quotes["ask"] = quotes["ask"].astype("float64")
+        quotes = self.prep_data(quotes, dedupe=True)
+
+        result = merge_asof(trades, quotes, on="time", by="ticker")
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_no_by(self, trades, asof, quotes):
+        f = (
+            lambda x: x[x.ticker == "MSFT"]
+            .drop("ticker", axis=1)
+            .reset_index(drop=True)
+        )
+
+        # just use a single ticker
+        expected = f(asof)
+        trades = f(trades)
+        quotes = f(quotes)
+
+        result = merge_asof(trades, quotes, on="time")
+        tm.assert_frame_equal(result, expected)
+
+    def test_valid_join_keys(self, trades, quotes):
+        msg = r"incompatible merge keys \[1\] .* must be the same type"
+
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(trades, quotes, left_on="time", right_on="bid", by="ticker")
+
+        with pytest.raises(MergeError, match="can only asof on a key for left"):
+            merge_asof(trades, quotes, on=["time", "ticker"], by="ticker")
+
+        with pytest.raises(MergeError, match="can only asof on a key for left"):
+            merge_asof(trades, quotes, by="ticker")
+
+    def test_with_duplicates(self, datapath, trades, quotes, asof):
+        q = (
+            pd.concat([quotes, quotes])
+            .sort_values(["time", "ticker"])
+            .reset_index(drop=True)
+        )
+        result = merge_asof(trades, q, on="time", by="ticker")
+        expected = self.prep_data(asof)
+        tm.assert_frame_equal(result, expected)
+
+    def test_with_duplicates_no_on(self):
+        df1 = pd.DataFrame({"key": [1, 1, 3], "left_val": [1, 2, 3]})
+        df2 = pd.DataFrame({"key": [1, 2, 2], "right_val": [1, 2, 3]})
+        result = merge_asof(df1, df2, on="key")
+        expected = pd.DataFrame(
+            {"key": [1, 1, 3], "left_val": [1, 2, 3], "right_val": [1, 1, 3]}
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_valid_allow_exact_matches(self, trades, quotes):
+        msg = "allow_exact_matches must be boolean, passed foo"
+
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades, quotes, on="time", by="ticker", allow_exact_matches="foo"
+            )
+
+    def test_valid_tolerance(self, trades, quotes):
+        # dti
+        merge_asof(trades, quotes, on="time", by="ticker", tolerance=Timedelta("1s"))
+
+        # integer
+        merge_asof(
+            trades.reset_index(),
+            quotes.reset_index(),
+            on="index",
+            by="ticker",
+            tolerance=1,
+        )
+
+        msg = r"incompatible tolerance .*, must be compat with type .*"
+
+        # incompat
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(trades, quotes, on="time", by="ticker", tolerance=1)
+
+        # invalid
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades.reset_index(),
+                quotes.reset_index(),
+                on="index",
+                by="ticker",
+                tolerance=1.0,
+            )
+
+        msg = "tolerance must be positive"
+
+        # invalid negative
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades, quotes, on="time", by="ticker", tolerance=-Timedelta("1s")
+            )
+
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(
+                trades.reset_index(),
+                quotes.reset_index(),
+                on="index",
+                by="ticker",
+                tolerance=-1,
+            )
+
+    def test_non_sorted(self, trades, quotes):
+        trades = trades.sort_values("time", ascending=False)
+        quotes = quotes.sort_values("time", ascending=False)
+
+        # we require that we are already sorted on time & quotes
+        assert not trades.time.is_monotonic_increasing
+        assert not quotes.time.is_monotonic_increasing
+        with pytest.raises(ValueError, match="left keys must be sorted"):
+            merge_asof(trades, quotes, on="time", by="ticker")
+
+        trades = trades.sort_values("time")
+        assert trades.time.is_monotonic_increasing
+        assert not quotes.time.is_monotonic_increasing
+        with pytest.raises(ValueError, match="right keys must be sorted"):
+            merge_asof(trades, quotes, on="time", by="ticker")
+
+        quotes = quotes.sort_values("time")
+        assert trades.time.is_monotonic_increasing
+        assert quotes.time.is_monotonic_increasing
+
+        # ok, though has dupes
+        merge_asof(trades, quotes, on="time", by="ticker")
+
+    @pytest.mark.parametrize(
+        "tolerance_ts",
+        [Timedelta("1day"), datetime.timedelta(days=1)],
+        ids=["Timedelta", "datetime.timedelta"],
+    )
+    def test_tolerance(self, tolerance_ts, trades, quotes, tolerance):
+        result = merge_asof(
+            trades, quotes, on="time", by="ticker", tolerance=tolerance_ts
+        )
+        expected = tolerance
+        tm.assert_frame_equal(result, expected)
+
+    def test_tolerance_forward(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 7, 11], "right_val": [1, 2, 3, 7, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [1, np.nan, 11]}
+        )
+
+        result = merge_asof(left, right, on="a", direction="forward", tolerance=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_tolerance_nearest(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 7, 11], "right_val": [1, 2, 3, 7, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [1, np.nan, 11]}
+        )
+
+        result = merge_asof(left, right, on="a", direction="nearest", tolerance=1)
+        tm.assert_frame_equal(result, expected)
+
+    def test_tolerance_tz(self, unit):
+        # GH 14844
+        left = pd.DataFrame(
+            {
+                "date": pd.date_range(
+                    start=to_datetime("2016-01-02"),
+                    freq="D",
+                    periods=5,
+                    tz=datetime.timezone.utc,
+                    unit=unit,
+                ),
+                "value1": np.arange(5),
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "date": pd.date_range(
+                    start=to_datetime("2016-01-01"),
+                    freq="D",
+                    periods=5,
+                    tz=datetime.timezone.utc,
+                    unit=unit,
+                ),
+                "value2": list("ABCDE"),
+            }
+        )
+        result = merge_asof(left, right, on="date", tolerance=Timedelta("1 day"))
+
+        expected = pd.DataFrame(
+            {
+                "date": pd.date_range(
+                    start=to_datetime("2016-01-02"),
+                    freq="D",
+                    periods=5,
+                    tz=datetime.timezone.utc,
+                    unit=unit,
+                ),
+                "value1": np.arange(5),
+                "value2": list("BCDEE"),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_tolerance_float(self):
+        # GH22981
+        left = pd.DataFrame({"a": [1.1, 3.5, 10.9], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame(
+            {"a": [1.0, 2.5, 3.3, 7.5, 11.5], "right_val": [1.0, 2.5, 3.3, 7.5, 11.5]}
+        )
+
+        expected = pd.DataFrame(
+            {
+                "a": [1.1, 3.5, 10.9],
+                "left_val": ["a", "b", "c"],
+                "right_val": [1, 3.3, np.nan],
+            }
+        )
+
+        result = merge_asof(left, right, on="a", direction="nearest", tolerance=0.5)
+        tm.assert_frame_equal(result, expected)
+
+    def test_index_tolerance(self, trades, quotes, tolerance):
+        # GH 15135
+        expected = tolerance.set_index("time")
+        trades = trades.set_index("time")
+        quotes = quotes.set_index("time")
+
+        result = merge_asof(
+            trades,
+            quotes,
+            left_index=True,
+            right_index=True,
+            by="ticker",
+            tolerance=Timedelta("1day"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches(self, trades, quotes):
+        result = merge_asof(
+            trades, quotes, on="time", by="ticker", allow_exact_matches=False
+        )
+        df = pd.DataFrame(
+            [
+                [
+                    "20160525 13:30:00.023",
+                    "MSFT",
+                    "51.95",
+                    "75",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.038",
+                    "MSFT",
+                    "51.95",
+                    "155",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.77",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.92",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "200",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "300",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "600",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "44",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.074",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.66",
+                    "6",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "30",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "75",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "20",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "35",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "10",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "6", "ARCA", np.nan, np.nan],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "6", "ARCA", np.nan, np.nan],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "1000",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "300",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "400",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "600",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "783",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+            ],
+            columns="time,ticker,price,quantity,marketCenter,bid,ask".split(","),
+        )
+        df["price"] = df["price"].astype("float64")
+        df["quantity"] = df["quantity"].astype("int64")
+        df["bid"] = df["bid"].astype("float64")
+        df["ask"] = df["ask"].astype("float64")
+        expected = self.prep_data(df)
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_forward(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 7, 11], "right_val": [1, 2, 3, 7, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [2, 7, 11]}
+        )
+
+        result = merge_asof(
+            left, right, on="a", direction="forward", allow_exact_matches=False
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_nearest(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 2, 3, 7, 11], "right_val": [1, 2, 3, 7, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [2, 3, 11]}
+        )
+
+        result = merge_asof(
+            left, right, on="a", direction="nearest", allow_exact_matches=False
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_and_tolerance(self, trades, quotes):
+        result = merge_asof(
+            trades,
+            quotes,
+            on="time",
+            by="ticker",
+            tolerance=Timedelta("100ms"),
+            allow_exact_matches=False,
+        )
+        df = pd.DataFrame(
+            [
+                [
+                    "20160525 13:30:00.023",
+                    "MSFT",
+                    "51.95",
+                    "75",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.038",
+                    "MSFT",
+                    "51.95",
+                    "155",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.77",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.92",
+                    "100",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "200",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "300",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "600",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.048",
+                    "GOOG",
+                    "720.93",
+                    "44",
+                    "NASDAQ",
+                    "720.5",
+                    "720.93",
+                ],
+                [
+                    "20160525 13:30:00.074",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.67",
+                    "478343",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.66",
+                    "6",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "30",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "75",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "20",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "35",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                [
+                    "20160525 13:30:00.075",
+                    "AAPL",
+                    "98.65",
+                    "10",
+                    "NASDAQ",
+                    np.nan,
+                    np.nan,
+                ],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "6", "ARCA", np.nan, np.nan],
+                ["20160525 13:30:00.075", "AAPL", "98.55", "6", "ARCA", np.nan, np.nan],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "1000",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "300",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "400",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "600",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.076",
+                    "AAPL",
+                    "98.56",
+                    "200",
+                    "ARCA",
+                    "98.55",
+                    "98.56",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "783",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+                [
+                    "20160525 13:30:00.078",
+                    "MSFT",
+                    "51.95",
+                    "100",
+                    "NASDAQ",
+                    "51.95",
+                    "51.95",
+                ],
+            ],
+            columns="time,ticker,price,quantity,marketCenter,bid,ask".split(","),
+        )
+        df["price"] = df["price"].astype("float64")
+        df["quantity"] = df["quantity"].astype("int64")
+        df["bid"] = df["bid"].astype("float64")
+        df["ask"] = df["ask"].astype("float64")
+        expected = self.prep_data(df)
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_and_tolerance2(self):
+        # GH 13695
+        df1 = pd.DataFrame(
+            {"time": to_datetime(["2016-07-15 13:30:00.030"]), "username": ["bob"]}
+        )
+        df2 = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    ["2016-07-15 13:30:00.000", "2016-07-15 13:30:00.030"]
+                ),
+                "version": [1, 2],
+            }
+        )
+
+        result = merge_asof(df1, df2, on="time")
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(["2016-07-15 13:30:00.030"]),
+                "username": ["bob"],
+                "version": [2],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = merge_asof(df1, df2, on="time", allow_exact_matches=False)
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(["2016-07-15 13:30:00.030"]),
+                "username": ["bob"],
+                "version": [1],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = merge_asof(
+            df1,
+            df2,
+            on="time",
+            allow_exact_matches=False,
+            tolerance=Timedelta("10ms"),
+        )
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(["2016-07-15 13:30:00.030"]),
+                "username": ["bob"],
+                "version": [np.nan],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_and_tolerance3(self):
+        # GH 13709
+        df1 = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    ["2016-07-15 13:30:00.030", "2016-07-15 13:30:00.030"]
+                ),
+                "username": ["bob", "charlie"],
+            }
+        )
+        df2 = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    ["2016-07-15 13:30:00.000", "2016-07-15 13:30:00.030"]
+                ),
+                "version": [1, 2],
+            }
+        )
+
+        result = merge_asof(
+            df1,
+            df2,
+            on="time",
+            allow_exact_matches=False,
+            tolerance=Timedelta("10ms"),
+        )
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    ["2016-07-15 13:30:00.030", "2016-07-15 13:30:00.030"]
+                ),
+                "username": ["bob", "charlie"],
+                "version": [np.nan, np.nan],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_and_tolerance_forward(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 3, 4, 6, 11], "right_val": [1, 3, 4, 6, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [np.nan, 6, 11]}
+        )
+
+        result = merge_asof(
+            left,
+            right,
+            on="a",
+            direction="forward",
+            allow_exact_matches=False,
+            tolerance=1,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_allow_exact_matches_and_tolerance_nearest(self):
+        # GH14887
+
+        left = pd.DataFrame({"a": [1, 5, 10], "left_val": ["a", "b", "c"]})
+        right = pd.DataFrame({"a": [1, 3, 4, 6, 11], "right_val": [1, 3, 4, 7, 11]})
+
+        expected = pd.DataFrame(
+            {"a": [1, 5, 10], "left_val": ["a", "b", "c"], "right_val": [np.nan, 4, 11]}
+        )
+
+        result = merge_asof(
+            left,
+            right,
+            on="a",
+            direction="nearest",
+            allow_exact_matches=False,
+            tolerance=1,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_forward_by(self):
+        # GH14887
+
+        left = pd.DataFrame(
+            {
+                "a": [1, 5, 10, 12, 15],
+                "b": ["X", "X", "Y", "Z", "Y"],
+                "left_val": ["a", "b", "c", "d", "e"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "a": [1, 6, 11, 15, 16],
+                "b": ["X", "Z", "Y", "Z", "Y"],
+                "right_val": [1, 6, 11, 15, 16],
+            }
+        )
+
+        expected = pd.DataFrame(
+            {
+                "a": [1, 5, 10, 12, 15],
+                "b": ["X", "X", "Y", "Z", "Y"],
+                "left_val": ["a", "b", "c", "d", "e"],
+                "right_val": [1, np.nan, 11, 15, 16],
+            }
+        )
+
+        result = merge_asof(left, right, on="a", by="b", direction="forward")
+        tm.assert_frame_equal(result, expected)
+
+    def test_nearest_by(self):
+        # GH14887
+
+        left = pd.DataFrame(
+            {
+                "a": [1, 5, 10, 12, 15],
+                "b": ["X", "X", "Z", "Z", "Y"],
+                "left_val": ["a", "b", "c", "d", "e"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "a": [1, 6, 11, 15, 16],
+                "b": ["X", "Z", "Z", "Z", "Y"],
+                "right_val": [1, 6, 11, 15, 16],
+            }
+        )
+
+        expected = pd.DataFrame(
+            {
+                "a": [1, 5, 10, 12, 15],
+                "b": ["X", "X", "Z", "Z", "Y"],
+                "left_val": ["a", "b", "c", "d", "e"],
+                "right_val": [1, 1, 11, 11, 16],
+            }
+        )
+
+        result = merge_asof(left, right, on="a", by="b", direction="nearest")
+        tm.assert_frame_equal(result, expected)
+
+    def test_by_int(self):
+        # we specialize by type, so test that this is correct
+        df1 = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.020",
+                        "20160525 13:30:00.030",
+                        "20160525 13:30:00.040",
+                        "20160525 13:30:00.050",
+                        "20160525 13:30:00.060",
+                    ]
+                ),
+                "key": [1, 2, 1, 3, 2],
+                "value1": [1.1, 1.2, 1.3, 1.4, 1.5],
+            },
+            columns=["time", "key", "value1"],
+        )
+
+        df2 = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.015",
+                        "20160525 13:30:00.020",
+                        "20160525 13:30:00.025",
+                        "20160525 13:30:00.035",
+                        "20160525 13:30:00.040",
+                        "20160525 13:30:00.055",
+                        "20160525 13:30:00.060",
+                        "20160525 13:30:00.065",
+                    ]
+                ),
+                "key": [2, 1, 1, 3, 2, 1, 2, 3],
+                "value2": [2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8],
+            },
+            columns=["time", "key", "value2"],
+        )
+
+        result = merge_asof(df1, df2, on="time", by="key")
+
+        expected = pd.DataFrame(
+            {
+                "time": to_datetime(
+                    [
+                        "20160525 13:30:00.020",
+                        "20160525 13:30:00.030",
+                        "20160525 13:30:00.040",
+                        "20160525 13:30:00.050",
+                        "20160525 13:30:00.060",
+                    ]
+                ),
+                "key": [1, 2, 1, 3, 2],
+                "value1": [1.1, 1.2, 1.3, 1.4, 1.5],
+                "value2": [2.2, 2.1, 2.3, 2.4, 2.7],
+            },
+            columns=["time", "key", "value1", "value2"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_on_float(self):
+        # mimics how to determine the minimum-price variation
+        df1 = pd.DataFrame(
+            {
+                "price": [5.01, 0.0023, 25.13, 340.05, 30.78, 1040.90, 0.0078],
+                "symbol": list("ABCDEFG"),
+            },
+            columns=["symbol", "price"],
+        )
+
+        df2 = pd.DataFrame(
+            {"price": [0.0, 1.0, 100.0], "mpv": [0.0001, 0.01, 0.05]},
+            columns=["price", "mpv"],
+        )
+
+        df1 = df1.sort_values("price").reset_index(drop=True)
+
+        result = merge_asof(df1, df2, on="price")
+
+        expected = pd.DataFrame(
+            {
+                "symbol": list("BGACEDF"),
+                "price": [0.0023, 0.0078, 5.01, 25.13, 30.78, 340.05, 1040.90],
+                "mpv": [0.0001, 0.0001, 0.01, 0.01, 0.01, 0.05, 0.05],
+            },
+            columns=["symbol", "price", "mpv"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_on_specialized_type(self, any_real_numpy_dtype):
+        # see gh-13936
+        dtype = np.dtype(any_real_numpy_dtype).type
+
+        df1 = pd.DataFrame(
+            {"value": [5, 2, 25, 100, 78, 120, 79], "symbol": list("ABCDEFG")},
+            columns=["symbol", "value"],
+        )
+        df1.value = dtype(df1.value)
+
+        df2 = pd.DataFrame(
+            {"value": [0, 80, 120, 125], "result": list("xyzw")},
+            columns=["value", "result"],
+        )
+        df2.value = dtype(df2.value)
+
+        df1 = df1.sort_values("value").reset_index(drop=True)
+        result = merge_asof(df1, df2, on="value")
+
+        expected = pd.DataFrame(
+            {
+                "symbol": list("BACEGDF"),
+                "value": [2, 5, 25, 78, 79, 100, 120],
+                "result": list("xxxxxyz"),
+            },
+            columns=["symbol", "value", "result"],
+        )
+        expected.value = dtype(expected.value)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_on_specialized_type_by_int(self, any_real_numpy_dtype):
+        # see gh-13936
+        dtype = np.dtype(any_real_numpy_dtype).type
+
+        df1 = pd.DataFrame(
+            {
+                "value": [5, 2, 25, 100, 78, 120, 79],
+                "key": [1, 2, 3, 2, 3, 1, 2],
+                "symbol": list("ABCDEFG"),
+            },
+            columns=["symbol", "key", "value"],
+        )
+        df1.value = dtype(df1.value)
+
+        df2 = pd.DataFrame(
+            {"value": [0, 80, 120, 125], "key": [1, 2, 2, 3], "result": list("xyzw")},
+            columns=["value", "key", "result"],
+        )
+        df2.value = dtype(df2.value)
+
+        df1 = df1.sort_values("value").reset_index(drop=True)
+        result = merge_asof(df1, df2, on="value", by="key")
+
+        expected = pd.DataFrame(
+            {
+                "symbol": list("BACEGDF"),
+                "key": [2, 1, 3, 3, 2, 2, 1],
+                "value": [2, 5, 25, 78, 79, 100, 120],
+                "result": [np.nan, "x", np.nan, np.nan, np.nan, "y", "x"],
+            },
+            columns=["symbol", "key", "value", "result"],
+        )
+        expected.value = dtype(expected.value)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_on_float_by_int(self):
+        # type specialize both "by" and "on" parameters
+        df1 = pd.DataFrame(
+            {
+                "symbol": list("AAABBBCCC"),
+                "exch": [1, 2, 3, 1, 2, 3, 1, 2, 3],
+                "price": [
+                    3.26,
+                    3.2599,
+                    3.2598,
+                    12.58,
+                    12.59,
+                    12.5,
+                    378.15,
+                    378.2,
+                    378.25,
+                ],
+            },
+            columns=["symbol", "exch", "price"],
+        )
+
+        df2 = pd.DataFrame(
+            {
+                "exch": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+                "price": [0.0, 1.0, 100.0, 0.0, 5.0, 100.0, 0.0, 5.0, 1000.0],
+                "mpv": [0.0001, 0.01, 0.05, 0.0001, 0.01, 0.1, 0.0001, 0.25, 1.0],
+            },
+            columns=["exch", "price", "mpv"],
+        )
+
+        df1 = df1.sort_values("price").reset_index(drop=True)
+        df2 = df2.sort_values("price").reset_index(drop=True)
+
+        result = merge_asof(df1, df2, on="price", by="exch")
+
+        expected = pd.DataFrame(
+            {
+                "symbol": list("AAABBBCCC"),
+                "exch": [3, 2, 1, 3, 1, 2, 1, 2, 3],
+                "price": [
+                    3.2598,
+                    3.2599,
+                    3.26,
+                    12.5,
+                    12.58,
+                    12.59,
+                    378.15,
+                    378.2,
+                    378.25,
+                ],
+                "mpv": [0.0001, 0.0001, 0.01, 0.25, 0.01, 0.01, 0.05, 0.1, 0.25],
+            },
+            columns=["symbol", "exch", "price", "mpv"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_datatype_error_raises(self):
+        msg = r"Incompatible merge dtype, .*, both sides must have numeric dtype"
+
+        left = pd.DataFrame({"left_val": [1, 5, 10], "a": ["a", "b", "c"]})
+        right = pd.DataFrame({"right_val": [1, 2, 3, 6, 7], "a": [1, 2, 3, 6, 7]})
+
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(left, right, on="a")
+
+    def test_merge_datatype_categorical_error_raises(self):
+        msg = (
+            r"incompatible merge keys \[0\] .* both sides category, "
+            "but not equal ones"
+        )
+
+        left = pd.DataFrame(
+            {"left_val": [1, 5, 10], "a": pd.Categorical(["a", "b", "c"])}
+        )
+        right = pd.DataFrame(
+            {
+                "right_val": [1, 2, 3, 6, 7],
+                "a": pd.Categorical(["a", "X", "c", "X", "b"]),
+            }
+        )
+
+        with pytest.raises(MergeError, match=msg):
+            merge_asof(left, right, on="a")
+
+    def test_merge_groupby_multiple_column_with_categorical_column(self):
+        # GH 16454
+        df = pd.DataFrame({"x": [0], "y": [0], "z": pd.Categorical([0])})
+        result = merge_asof(df, df, on="x", by=["y", "z"])
+        expected = pd.DataFrame({"x": [0], "y": [0], "z": pd.Categorical([0])})
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "func", [lambda x: x, to_datetime], ids=["numeric", "datetime"]
+    )
+    @pytest.mark.parametrize("side", ["left", "right"])
+    def test_merge_on_nans(self, func, side):
+        # GH 23189
+        msg = f"Merge keys contain null values on {side} side"
+        nulls = func([1.0, 5.0, np.nan])
+        non_nulls = func([1.0, 5.0, 10.0])
+        df_null = pd.DataFrame({"a": nulls, "left_val": ["a", "b", "c"]})
+        df = pd.DataFrame({"a": non_nulls, "right_val": [1, 6, 11]})
+
+        with pytest.raises(ValueError, match=msg):
+            if side == "left":
+                merge_asof(df_null, df, on="a")
+            else:
+                merge_asof(df, df_null, on="a")
+
+    def test_by_nullable(self, any_numeric_ea_dtype, using_infer_string):
+        # Note: this test passes if instead of using pd.array we use
+        #  np.array([np.nan, 1]).  Other than that, I (@jbrockmendel)
+        #  have NO IDEA what the expected behavior is.
+        # TODO(GH#32306): may be relevant to the expected behavior here.
+
+        arr = pd.array([pd.NA, 0, 1], dtype=any_numeric_ea_dtype)
+        if arr.dtype.kind in ["i", "u"]:
+            max_val = np.iinfo(arr.dtype.numpy_dtype).max
+        else:
+            max_val = np.finfo(arr.dtype.numpy_dtype).max
+        # set value s.t. (at least for integer dtypes) arr._values_for_argsort
+        #  is not an injection
+        arr[2] = max_val
+
+        left = pd.DataFrame(
+            {
+                "by_col1": arr,
+                "by_col2": ["HELLO", "To", "You"],
+                "on_col": [2, 4, 6],
+                "value": ["a", "c", "e"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "by_col1": arr,
+                "by_col2": ["WORLD", "Wide", "Web"],
+                "on_col": [1, 2, 6],
+                "value": ["b", "d", "f"],
+            }
+        )
+
+        result = merge_asof(left, right, by=["by_col1", "by_col2"], on="on_col")
+        expected = pd.DataFrame(
+            {
+                "by_col1": arr,
+                "by_col2": ["HELLO", "To", "You"],
+                "on_col": [2, 4, 6],
+                "value_x": ["a", "c", "e"],
+            }
+        )
+        expected["value_y"] = np.array([np.nan, np.nan, np.nan], dtype=object)
+        if using_infer_string:
+            expected["value_y"] = expected["value_y"].astype("str")
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_by_col_tz_aware(self):
+        # GH 21184
+        left = pd.DataFrame(
+            {
+                "by_col": pd.DatetimeIndex(["2018-01-01"]).tz_localize("UTC"),
+                "on_col": [2],
+                "values": ["a"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "by_col": pd.DatetimeIndex(["2018-01-01"]).tz_localize("UTC"),
+                "on_col": [1],
+                "values": ["b"],
+            }
+        )
+        result = merge_asof(left, right, by="by_col", on="on_col")
+        expected = pd.DataFrame(
+            [[pd.Timestamp("2018-01-01", tz="UTC"), 2, "a", "b"]],
+            columns=["by_col", "on_col", "values_x", "values_y"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_by_mixed_tz_aware(self, using_infer_string):
+        # GH 26649
+        left = pd.DataFrame(
+            {
+                "by_col1": pd.DatetimeIndex(["2018-01-01"]).tz_localize("UTC"),
+                "by_col2": ["HELLO"],
+                "on_col": [2],
+                "value": ["a"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "by_col1": pd.DatetimeIndex(["2018-01-01"]).tz_localize("UTC"),
+                "by_col2": ["WORLD"],
+                "on_col": [1],
+                "value": ["b"],
+            }
+        )
+        result = merge_asof(left, right, by=["by_col1", "by_col2"], on="on_col")
+        expected = pd.DataFrame(
+            [[pd.Timestamp("2018-01-01", tz="UTC"), "HELLO", 2, "a"]],
+            columns=["by_col1", "by_col2", "on_col", "value_x"],
+        )
+        expected["value_y"] = np.array([np.nan], dtype=object)
+        if using_infer_string:
+            expected["value_y"] = expected["value_y"].astype("str")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["float64", "int16", "m8[ns]", "M8[us]"])
+    def test_by_dtype(self, dtype):
+        # GH 55453, GH 22794
+        left = pd.DataFrame(
+            {
+                "by_col": np.array([1], dtype=dtype),
+                "on_col": [2],
+                "value": ["a"],
+            }
+        )
+        right = pd.DataFrame(
+            {
+                "by_col": np.array([1], dtype=dtype),
+                "on_col": [1],
+                "value": ["b"],
+            }
+        )
+        result = merge_asof(left, right, by="by_col", on="on_col")
+        expected = pd.DataFrame(
+            {
+                "by_col": np.array([1], dtype=dtype),
+                "on_col": [2],
+                "value_x": ["a"],
+                "value_y": ["b"],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_timedelta_tolerance_nearest(self, unit):
+        # GH 27642
+        if unit == "s":
+            pytest.skip(
+                "This test is invalid with unit='s' because that would "
+                "round left['time']"
+            )
+
+        left = pd.DataFrame(
+            list(zip([0, 5, 10, 15, 20, 25], [0, 1, 2, 3, 4, 5])),
+            columns=["time", "left"],
+        )
+
+        left["time"] = pd.to_timedelta(left["time"], "ms").astype(f"m8[{unit}]")
+
+        right = pd.DataFrame(
+            list(zip([0, 3, 9, 12, 15, 18], [0, 1, 2, 3, 4, 5])),
+            columns=["time", "right"],
+        )
+
+        right["time"] = pd.to_timedelta(right["time"], "ms").astype(f"m8[{unit}]")
+
+        expected = pd.DataFrame(
+            list(
+                zip(
+                    [0, 5, 10, 15, 20, 25],
+                    [0, 1, 2, 3, 4, 5],
+                    [0, np.nan, 2, 4, np.nan, np.nan],
+                )
+            ),
+            columns=["time", "left", "right"],
+        )
+
+        expected["time"] = pd.to_timedelta(expected["time"], "ms").astype(f"m8[{unit}]")
+
+        result = merge_asof(
+            left, right, on="time", tolerance=Timedelta("1ms"), direction="nearest"
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_int_type_tolerance(self, any_int_dtype):
+        # GH #28870
+
+        left = pd.DataFrame({"a": [0, 10, 20], "left_val": [1, 2, 3]})
+        right = pd.DataFrame({"a": [5, 15, 25], "right_val": [1, 2, 3]})
+        left["a"] = left["a"].astype(any_int_dtype)
+        right["a"] = right["a"].astype(any_int_dtype)
+
+        expected = pd.DataFrame(
+            {"a": [0, 10, 20], "left_val": [1, 2, 3], "right_val": [np.nan, 1.0, 2.0]}
+        )
+        expected["a"] = expected["a"].astype(any_int_dtype)
+
+        result = merge_asof(left, right, on="a", tolerance=10)
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_index_column_tz(self):
+        # GH 29864
+        index = pd.date_range("2019-10-01", freq="30min", periods=5, tz="UTC")
+        left = pd.DataFrame([0.9, 0.8, 0.7, 0.6], columns=["xyz"], index=index[1:])
+        right = pd.DataFrame({"from_date": index, "abc": [2.46] * 4 + [2.19]})
+        result = merge_asof(
+            left=left, right=right, left_index=True, right_on=["from_date"]
+        )
+        expected = pd.DataFrame(
+            {
+                "xyz": [0.9, 0.8, 0.7, 0.6],
+                "from_date": index[1:],
+                "abc": [2.46] * 3 + [2.19],
+            },
+            index=pd.date_range(
+                "2019-10-01 00:30:00", freq="30min", periods=4, tz="UTC"
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = merge_asof(
+            left=right, right=left, right_index=True, left_on=["from_date"]
+        )
+        expected = pd.DataFrame(
+            {
+                "from_date": index,
+                "abc": [2.46] * 4 + [2.19],
+                "xyz": [np.nan, 0.9, 0.8, 0.7, 0.6],
+            },
+            index=Index([0, 1, 2, 3, 4]),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_left_index_right_index_tolerance(self, unit):
+        # https://github.com/pandas-dev/pandas/issues/35558
+        if unit == "s":
+            pytest.skip(
+                "This test is invalid with unit='s' because that would round dr1"
+            )
+
+        dr1 = pd.date_range(
+            start="1/1/2020", end="1/20/2020", freq="2D", unit=unit
+        ) + Timedelta(seconds=0.4).as_unit(unit)
+        dr2 = pd.date_range(start="1/1/2020", end="2/1/2020", unit=unit)
+
+        df1 = pd.DataFrame({"val1": "foo"}, index=pd.DatetimeIndex(dr1))
+        df2 = pd.DataFrame({"val2": "bar"}, index=pd.DatetimeIndex(dr2))
+
+        expected = pd.DataFrame(
+            {"val1": "foo", "val2": "bar"}, index=pd.DatetimeIndex(dr1)
+        )
+        result = merge_asof(
+            df1,
+            df2,
+            left_index=True,
+            right_index=True,
+            tolerance=Timedelta(seconds=0.5),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+@pytest.mark.parametrize(
+    "kwargs", [{"on": "x"}, {"left_index": True, "right_index": True}]
+)
+@pytest.mark.parametrize(
+    "data",
+    [["2019-06-01 00:09:12", "2019-06-01 00:10:29"], [1.0, "2019-06-01 00:10:29"]],
+)
+def test_merge_asof_non_numerical_dtype(kwargs, data, infer_string):
+    # GH#29130
+    with option_context("future.infer_string", infer_string):
+        left = pd.DataFrame({"x": data}, index=data)
+        right = pd.DataFrame({"x": data}, index=data)
+        with pytest.raises(
+            MergeError,
+            match=r"Incompatible merge dtype, .*, both sides must have numeric dtype",
+        ):
+            merge_asof(left, right, **kwargs)
+
+
+def test_merge_asof_non_numerical_dtype_object():
+    # GH#29130
+    left = pd.DataFrame({"a": ["12", "13", "15"], "left_val1": ["a", "b", "c"]})
+    right = pd.DataFrame({"a": ["a", "b", "c"], "left_val": ["d", "e", "f"]})
+    with pytest.raises(
+        MergeError,
+        match=r"Incompatible merge dtype, .*, both sides must have numeric dtype",
+    ):
+        merge_asof(
+            left,
+            right,
+            left_on="left_val1",
+            right_on="a",
+            left_by="a",
+            right_by="left_val",
+        )
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"right_index": True, "left_index": True},
+        {"left_on": "left_time", "right_index": True},
+        {"left_index": True, "right_on": "right"},
+    ],
+)
+def test_merge_asof_index_behavior(kwargs):
+    # GH 33463
+    index = Index([1, 5, 10], name="test")
+    left = pd.DataFrame({"left": ["a", "b", "c"], "left_time": [1, 4, 10]}, index=index)
+    right = pd.DataFrame({"right": [1, 2, 3, 6, 7]}, index=[1, 2, 3, 6, 7])
+    result = merge_asof(left, right, **kwargs)
+
+    expected = pd.DataFrame(
+        {"left": ["a", "b", "c"], "left_time": [1, 4, 10], "right": [1, 3, 7]},
+        index=index,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_numeric_column_in_index():
+    # GH#34488
+    left = pd.DataFrame({"b": [10, 11, 12]}, index=Index([1, 2, 3], name="a"))
+    right = pd.DataFrame({"c": [20, 21, 22]}, index=Index([0, 2, 3], name="a"))
+
+    result = merge_asof(left, right, left_on="a", right_on="a")
+    expected = pd.DataFrame({"a": [1, 2, 3], "b": [10, 11, 12], "c": [20, 21, 22]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_numeric_column_in_multiindex():
+    # GH#34488
+    left = pd.DataFrame(
+        {"b": [10, 11, 12]},
+        index=pd.MultiIndex.from_arrays([[1, 2, 3], ["a", "b", "c"]], names=["a", "z"]),
+    )
+    right = pd.DataFrame(
+        {"c": [20, 21, 22]},
+        index=pd.MultiIndex.from_arrays([[1, 2, 3], ["x", "y", "z"]], names=["a", "y"]),
+    )
+
+    result = merge_asof(left, right, left_on="a", right_on="a")
+    expected = pd.DataFrame({"a": [1, 2, 3], "b": [10, 11, 12], "c": [20, 21, 22]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_numeri_column_in_index_object_dtype():
+    # GH#34488
+    left = pd.DataFrame({"b": [10, 11, 12]}, index=Index(["1", "2", "3"], name="a"))
+    right = pd.DataFrame({"c": [20, 21, 22]}, index=Index(["m", "n", "o"], name="a"))
+
+    with pytest.raises(
+        MergeError,
+        match=r"Incompatible merge dtype, .*, both sides must have numeric dtype",
+    ):
+        merge_asof(left, right, left_on="a", right_on="a")
+
+    left = left.reset_index().set_index(["a", "b"])
+    right = right.reset_index().set_index(["a", "c"])
+
+    with pytest.raises(
+        MergeError,
+        match=r"Incompatible merge dtype, .*, both sides must have numeric dtype",
+    ):
+        merge_asof(left, right, left_on="a", right_on="a")
+
+
+def test_merge_asof_array_as_on(unit):
+    # GH#42844
+    dti = pd.DatetimeIndex(
+        ["2021/01/01 00:37", "2021/01/01 01:40"], dtype=f"M8[{unit}]"
+    )
+    right = pd.DataFrame(
+        {
+            "a": [2, 6],
+            "ts": dti,
+        }
+    )
+    ts_merge = pd.date_range(
+        start=pd.Timestamp("2021/01/01 00:00"), periods=3, freq="1h", unit=unit
+    )
+    left = pd.DataFrame({"b": [4, 8, 7]})
+    result = merge_asof(
+        left,
+        right,
+        left_on=ts_merge,
+        right_on="ts",
+        allow_exact_matches=False,
+        direction="backward",
+    )
+    expected = pd.DataFrame({"b": [4, 8, 7], "a": [np.nan, 2, 6], "ts": ts_merge})
+    tm.assert_frame_equal(result, expected)
+
+    result = merge_asof(
+        right,
+        left,
+        left_on="ts",
+        right_on=ts_merge,
+        allow_exact_matches=False,
+        direction="backward",
+    )
+    expected = pd.DataFrame(
+        {
+            "a": [2, 6],
+            "ts": dti,
+            "b": [4, 8],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_raise_for_duplicate_columns():
+    # GH#50102
+    left = pd.DataFrame([[1, 2, "a"]], columns=["a", "a", "left_val"])
+    right = pd.DataFrame([[1, 1, 1]], columns=["a", "a", "right_val"])
+
+    with pytest.raises(ValueError, match="column label 'a'"):
+        merge_asof(left, right, on="a")
+
+    with pytest.raises(ValueError, match="column label 'a'"):
+        merge_asof(left, right, left_on="a", right_on="right_val")
+
+    with pytest.raises(ValueError, match="column label 'a'"):
+        merge_asof(left, right, left_on="left_val", right_on="a")
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "Int64",
+        pytest.param("int64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        pytest.param("timestamp[s][pyarrow]", marks=td.skip_if_no("pyarrow")),
+    ],
+)
+def test_merge_asof_extension_dtype(dtype):
+    # GH 52904
+    left = pd.DataFrame(
+        {
+            "join_col": [1, 3, 5],
+            "left_val": [1, 2, 3],
+        }
+    )
+    right = pd.DataFrame(
+        {
+            "join_col": [2, 3, 4],
+            "right_val": [1, 2, 3],
+        }
+    )
+    left = left.astype({"join_col": dtype})
+    right = right.astype({"join_col": dtype})
+    result = merge_asof(left, right, on="join_col")
+    expected = pd.DataFrame(
+        {
+            "join_col": [1, 3, 5],
+            "left_val": [1, 2, 3],
+            "right_val": [np.nan, 2.0, 3.0],
+        }
+    )
+    expected = expected.astype({"join_col": dtype})
+    tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_merge_asof_pyarrow_td_tolerance():
+    # GH 56486
+    ser = pd.Series(
+        [datetime.datetime(2023, 1, 1)], dtype="timestamp[us, UTC][pyarrow]"
+    )
+    df = pd.DataFrame(
+        {
+            "timestamp": ser,
+            "value": [1],
+        }
+    )
+    result = merge_asof(df, df, on="timestamp", tolerance=Timedelta("1s"))
+    expected = pd.DataFrame(
+        {
+            "timestamp": ser,
+            "value_x": [1],
+            "value_y": [1],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_read_only_ndarray():
+    # GH 53513
+    left = pd.Series([2], index=[2], name="left")
+    right = pd.Series([1], index=[1], name="right")
+    # set to read-only
+    left.index.values.flags.writeable = False
+    right.index.values.flags.writeable = False
+    result = merge_asof(left, right, left_index=True, right_index=True)
+    expected = pd.DataFrame({"left": [2], "right": [1]}, index=[2])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_asof_multiby_with_categorical():
+    # GH 43541
+    left = pd.DataFrame(
+        {
+            "c1": pd.Categorical(["a", "a", "b", "b"], categories=["a", "b"]),
+            "c2": ["x"] * 4,
+            "t": [1] * 4,
+            "v": range(4),
+        }
+    )
+    right = pd.DataFrame(
+        {
+            "c1": pd.Categorical(["b", "b"], categories=["b", "a"]),
+            "c2": ["x"] * 2,
+            "t": [1, 2],
+            "v": range(2),
+        }
+    )
+    result = merge_asof(
+        left,
+        right,
+        by=["c1", "c2"],
+        on="t",
+        direction="forward",
+        suffixes=["_left", "_right"],
+    )
+    expected = pd.DataFrame(
+        {
+            "c1": pd.Categorical(["a", "a", "b", "b"], categories=["a", "b"]),
+            "c2": ["x"] * 4,
+            "t": [1] * 4,
+            "v_left": range(4),
+            "v_right": [np.nan, np.nan, 0.0, 0.0],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge_cross.py b/pandas/pandas/tests/reshape/merge/test_merge_cross.py
new file mode 100644
index 0000000000000..6ab80cf0e0823
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge_cross.py
@@ -0,0 +1,109 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.reshape.merge import (
+    MergeError,
+    merge,
+)
+
+
+@pytest.mark.parametrize(
+    ("input_col", "output_cols"), [("b", ["a", "b"]), ("a", ["a_x", "a_y"])]
+)
+def test_merge_cross(input_col, output_cols):
+    # GH#5401
+    left = DataFrame({"a": [1, 3]})
+    right = DataFrame({input_col: [3, 4]})
+    left_copy = left.copy()
+    right_copy = right.copy()
+    result = merge(left, right, how="cross")
+    expected = DataFrame({output_cols[0]: [1, 1, 3, 3], output_cols[1]: [3, 4, 3, 4]})
+    tm.assert_frame_equal(result, expected)
+    tm.assert_frame_equal(left, left_copy)
+    tm.assert_frame_equal(right, right_copy)
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"left_index": True},
+        {"right_index": True},
+        {"on": "a"},
+        {"left_on": "a"},
+        {"right_on": "b"},
+    ],
+)
+def test_merge_cross_error_reporting(kwargs):
+    # GH#5401
+    left = DataFrame({"a": [1, 3]})
+    right = DataFrame({"b": [3, 4]})
+    msg = (
+        "Can not pass on, right_on, left_on or set right_index=True or left_index=True"
+    )
+    with pytest.raises(MergeError, match=msg):
+        merge(left, right, how="cross", **kwargs)
+
+
+def test_merge_cross_mixed_dtypes():
+    # GH#5401
+    left = DataFrame(["a", "b", "c"], columns=["A"])
+    right = DataFrame(range(2), columns=["B"])
+    result = merge(left, right, how="cross")
+    expected = DataFrame({"A": ["a", "a", "b", "b", "c", "c"], "B": [0, 1, 0, 1, 0, 1]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_cross_more_than_one_column():
+    # GH#5401
+    left = DataFrame({"A": list("ab"), "B": [2, 1]})
+    right = DataFrame({"C": range(2), "D": range(4, 6)})
+    result = merge(left, right, how="cross")
+    expected = DataFrame(
+        {
+            "A": ["a", "a", "b", "b"],
+            "B": [2, 2, 1, 1],
+            "C": [0, 1, 0, 1],
+            "D": [4, 5, 4, 5],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_merge_cross_null_values(nulls_fixture):
+    # GH#5401
+    left = DataFrame({"a": [1, nulls_fixture]})
+    right = DataFrame({"b": ["a", "b"], "c": [1.0, 2.0]})
+    result = merge(left, right, how="cross")
+    expected = DataFrame(
+        {
+            "a": [1, 1, nulls_fixture, nulls_fixture],
+            "b": ["a", "b", "a", "b"],
+            "c": [1.0, 2.0, 1.0, 2.0],
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_join_cross_error_reporting():
+    # GH#5401
+    left = DataFrame({"a": [1, 3]})
+    right = DataFrame({"a": [3, 4]})
+    msg = (
+        "Can not pass on, right_on, left_on or set right_index=True or left_index=True"
+    )
+    with pytest.raises(MergeError, match=msg):
+        left.join(right, how="cross", on="a")
+
+
+def test_merge_cross_series():
+    # GH#54055
+    ls = Series([1, 2, 3, 4], index=[1, 2, 3, 4], name="left")
+    rs = Series([3, 4, 5, 6], index=[3, 4, 5, 6], name="right")
+    res = merge(ls, rs, how="cross")
+
+    expected = merge(ls.to_frame(), rs.to_frame(), how="cross")
+    tm.assert_frame_equal(res, expected)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge_index_as_string.py b/pandas/pandas/tests/reshape/merge/test_merge_index_as_string.py
new file mode 100644
index 0000000000000..b4271d4face4f
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge_index_as_string.py
@@ -0,0 +1,186 @@
+import numpy as np
+import pytest
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df1():
+    return DataFrame(
+        {
+            "outer": [1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 4],
+            "inner": [1, 2, 3, 1, 2, 3, 4, 1, 2, 1, 2],
+            "v1": np.linspace(0, 1, 11),
+        }
+    )
+
+
+@pytest.fixture
+def df2():
+    return DataFrame(
+        {
+            "outer": [1, 1, 1, 1, 1, 1, 2, 2, 3, 3, 3, 3],
+            "inner": [1, 2, 2, 3, 3, 4, 2, 3, 1, 1, 2, 3],
+            "v2": np.linspace(10, 11, 12),
+        }
+    )
+
+
+@pytest.fixture(params=[[], ["outer"], ["outer", "inner"]])
+def left_df(request, df1):
+    """Construct left test DataFrame with specified levels
+    (any of 'outer', 'inner', and 'v1')
+    """
+    levels = request.param
+    if levels:
+        df1 = df1.set_index(levels)
+
+    return df1
+
+
+@pytest.fixture(params=[[], ["outer"], ["outer", "inner"]])
+def right_df(request, df2):
+    """Construct right test DataFrame with specified levels
+    (any of 'outer', 'inner', and 'v2')
+    """
+    levels = request.param
+
+    if levels:
+        df2 = df2.set_index(levels)
+
+    return df2
+
+
+def compute_expected(df_left, df_right, on=None, left_on=None, right_on=None, how=None):
+    """
+    Compute the expected merge result for the test case.
+
+    This method computes the expected result of merging two DataFrames on
+    a combination of their columns and index levels. It does so by
+    explicitly dropping/resetting their named index levels, performing a
+    merge on their columns, and then finally restoring the appropriate
+    index in the result.
+
+    Parameters
+    ----------
+    df_left : DataFrame
+        The left DataFrame (may have zero or more named index levels)
+    df_right : DataFrame
+        The right DataFrame (may have zero or more named index levels)
+    on : list of str
+        The on parameter to the merge operation
+    left_on : list of str
+        The left_on parameter to the merge operation
+    right_on : list of str
+        The right_on parameter to the merge operation
+    how : str
+        The how parameter to the merge operation
+
+    Returns
+    -------
+    DataFrame
+        The expected merge result
+    """
+    # Handle on param if specified
+    if on is not None:
+        left_on, right_on = on, on
+
+    # Compute input named index levels
+    left_levels = [n for n in df_left.index.names if n is not None]
+    right_levels = [n for n in df_right.index.names if n is not None]
+
+    # Compute output named index levels
+    output_levels = [i for i in left_on if i in right_levels and i in left_levels]
+
+    # Drop index levels that aren't involved in the merge
+    drop_left = [n for n in left_levels if n not in left_on]
+    if drop_left:
+        df_left = df_left.reset_index(drop_left, drop=True)
+
+    drop_right = [n for n in right_levels if n not in right_on]
+    if drop_right:
+        df_right = df_right.reset_index(drop_right, drop=True)
+
+    # Convert remaining index levels to columns
+    reset_left = [n for n in left_levels if n in left_on]
+    if reset_left:
+        df_left = df_left.reset_index(level=reset_left)
+
+    reset_right = [n for n in right_levels if n in right_on]
+    if reset_right:
+        df_right = df_right.reset_index(level=reset_right)
+
+    # Perform merge
+    expected = df_left.merge(df_right, left_on=left_on, right_on=right_on, how=how)
+
+    # Restore index levels
+    if output_levels:
+        expected = expected.set_index(output_levels)
+
+    return expected
+
+
+@pytest.mark.parametrize(
+    "on,how",
+    [
+        (["outer"], "inner"),
+        (["inner"], "left"),
+        (["outer", "inner"], "right"),
+        (["inner", "outer"], "outer"),
+    ],
+)
+def test_merge_indexes_and_columns_on(left_df, right_df, on, how):
+    # Construct expected result
+    expected = compute_expected(left_df, right_df, on=on, how=how)
+
+    # Perform merge
+    result = left_df.merge(right_df, on=on, how=how)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize(
+    "left_on,right_on,how",
+    [
+        (["outer"], ["outer"], "inner"),
+        (["inner"], ["inner"], "right"),
+        (["outer", "inner"], ["outer", "inner"], "left"),
+        (["inner", "outer"], ["inner", "outer"], "outer"),
+    ],
+)
+def test_merge_indexes_and_columns_lefton_righton(
+    left_df, right_df, left_on, right_on, how
+):
+    # Construct expected result
+    expected = compute_expected(
+        left_df, right_df, left_on=left_on, right_on=right_on, how=how
+    )
+
+    # Perform merge
+    result = left_df.merge(right_df, left_on=left_on, right_on=right_on, how=how)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+@pytest.mark.parametrize("left_index", ["inner", ["inner", "outer"]])
+def test_join_indexes_and_columns_on(df1, df2, left_index, join_type):
+    # Construct left_df
+    left_df = df1.set_index(left_index)
+
+    # Construct right_df
+    right_df = df2.set_index(["outer", "inner"])
+
+    # Result
+    expected = (
+        left_df.reset_index()
+        .join(
+            right_df, on=["outer", "inner"], how=join_type, lsuffix="_x", rsuffix="_y"
+        )
+        .set_index(left_index)
+    )
+
+    # Perform join
+    result = left_df.join(
+        right_df, on=["outer", "inner"], how=join_type, lsuffix="_x", rsuffix="_y"
+    )
+
+    tm.assert_frame_equal(result, expected, check_like=True)
diff --git a/pandas/pandas/tests/reshape/merge/test_merge_ordered.py b/pandas/pandas/tests/reshape/merge/test_merge_ordered.py
new file mode 100644
index 0000000000000..7deecf6bbf4a4
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_merge_ordered.py
@@ -0,0 +1,241 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    merge_ordered,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def left():
+    return DataFrame({"key": ["a", "c", "e"], "lvalue": [1, 2.0, 3]})
+
+
+@pytest.fixture
+def right():
+    return DataFrame({"key": ["b", "c", "d", "f"], "rvalue": [1, 2, 3.0, 4]})
+
+
+class TestMergeOrdered:
+    def test_basic(self, left, right):
+        result = merge_ordered(left, right, on="key")
+        expected = DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "f"],
+                "lvalue": [1, np.nan, 2, np.nan, 3, np.nan],
+                "rvalue": [np.nan, 1, 2, 3, np.nan, 4],
+            }
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_ffill(self, left, right):
+        result = merge_ordered(left, right, on="key", fill_method="ffill")
+        expected = DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "f"],
+                "lvalue": [1.0, 1, 2, 2, 3, 3.0],
+                "rvalue": [np.nan, 1, 2, 3, 3, 4],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_multigroup(self, left, right):
+        left = pd.concat([left, left], ignore_index=True)
+
+        left["group"] = ["a"] * 3 + ["b"] * 3
+
+        result = merge_ordered(
+            left, right, on="key", left_by="group", fill_method="ffill"
+        )
+        expected = DataFrame(
+            {
+                "key": ["a", "b", "c", "d", "e", "f"] * 2,
+                "lvalue": [1.0, 1, 2, 2, 3, 3.0] * 2,
+                "rvalue": [np.nan, 1, 2, 3, 3, 4] * 2,
+            }
+        )
+        expected["group"] = ["a"] * 6 + ["b"] * 6
+
+        tm.assert_frame_equal(result, expected.loc[:, result.columns])
+
+        result2 = merge_ordered(
+            right, left, on="key", right_by="group", fill_method="ffill"
+        )
+        tm.assert_frame_equal(result, result2.loc[:, result.columns])
+
+        result = merge_ordered(left, right, on="key", left_by="group")
+        assert result["group"].notna().all()
+
+    @pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+    )
+    def test_merge_type(self, left, right):
+        class NotADataFrame(DataFrame):
+            @property
+            def _constructor(self):
+                return NotADataFrame
+
+        nad = NotADataFrame(left)
+        result = nad.merge(right, on="key")
+
+        assert isinstance(result, NotADataFrame)
+
+    @pytest.mark.parametrize(
+        "df_seq, pattern",
+        [
+            ((), "[Nn]o objects"),
+            ([], "[Nn]o objects"),
+            ({}, "[Nn]o objects"),
+            ([None], "objects.*None"),
+            ([None, None], "objects.*None"),
+        ],
+    )
+    def test_empty_sequence_concat(self, df_seq, pattern):
+        # GH 9157
+        with pytest.raises(ValueError, match=pattern):
+            pd.concat(df_seq)
+
+    @pytest.mark.parametrize(
+        "arg", [[DataFrame()], [None, DataFrame()], [DataFrame(), None]]
+    )
+    def test_empty_sequence_concat_ok(self, arg):
+        pd.concat(arg)
+
+    def test_doc_example(self):
+        left = DataFrame(
+            {
+                "group": list("aaabbb"),
+                "key": ["a", "c", "e", "a", "c", "e"],
+                "lvalue": [1, 2, 3] * 2,
+            }
+        )
+
+        right = DataFrame({"key": ["b", "c", "d"], "rvalue": [1, 2, 3]})
+
+        result = merge_ordered(left, right, fill_method="ffill", left_by="group")
+
+        expected = DataFrame(
+            {
+                "group": list("aaaaabbbbb"),
+                "key": ["a", "b", "c", "d", "e"] * 2,
+                "lvalue": [1, 1, 2, 2, 3] * 2,
+                "rvalue": [np.nan, 1, 2, 3, 3] * 2,
+            }
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "left, right, on, left_by, right_by, expected",
+        [
+            (
+                {"G": ["g", "g"], "H": ["h", "h"], "T": [1, 3]},
+                {"T": [2], "E": [1]},
+                ["T"],
+                ["G", "H"],
+                None,
+                {
+                    "G": ["g"] * 3,
+                    "H": ["h"] * 3,
+                    "T": [1, 2, 3],
+                    "E": [np.nan, 1.0, np.nan],
+                },
+            ),
+            (
+                {"G": ["g", "g"], "H": ["h", "h"], "T": [1, 3]},
+                {"T": [2], "E": [1]},
+                "T",
+                ["G", "H"],
+                None,
+                {
+                    "G": ["g"] * 3,
+                    "H": ["h"] * 3,
+                    "T": [1, 2, 3],
+                    "E": [np.nan, 1.0, np.nan],
+                },
+            ),
+            (
+                {"T": [2], "E": [1]},
+                {"G": ["g", "g"], "H": ["h", "h"], "T": [1, 3]},
+                ["T"],
+                None,
+                ["G", "H"],
+                {
+                    "T": [1, 2, 3],
+                    "E": [np.nan, 1.0, np.nan],
+                    "G": ["g"] * 3,
+                    "H": ["h"] * 3,
+                },
+            ),
+        ],
+    )
+    def test_list_type_by(self, left, right, on, left_by, right_by, expected):
+        # GH 35269
+        left = DataFrame(left)
+        right = DataFrame(right)
+        result = merge_ordered(
+            left=left,
+            right=right,
+            on=on,
+            left_by=left_by,
+            right_by=right_by,
+        )
+        expected = DataFrame(expected)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_left_by_length_equals_to_right_shape0(self):
+        # GH 38166
+        left = DataFrame([["g", "h", 1], ["g", "h", 3]], columns=list("GHE"))
+        right = DataFrame([[2, 1]], columns=list("ET"))
+        result = merge_ordered(left, right, on="E", left_by=["G", "H"])
+        expected = DataFrame(
+            {"G": ["g"] * 3, "H": ["h"] * 3, "E": [1, 2, 3], "T": [np.nan, 1.0, np.nan]}
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_elements_not_in_by_but_in_df(self):
+        # GH 38167
+        left = DataFrame([["g", "h", 1], ["g", "h", 3]], columns=list("GHE"))
+        right = DataFrame([[2, 1]], columns=list("ET"))
+        msg = r"\{'h'\} not found in left columns"
+        with pytest.raises(KeyError, match=msg):
+            merge_ordered(left, right, on="E", left_by=["G", "h"])
+
+    @pytest.mark.parametrize("invalid_method", ["linear", "carrot"])
+    def test_ffill_validate_fill_method(self, left, right, invalid_method):
+        # GH 55884
+        with pytest.raises(
+            ValueError, match=re.escape("fill_method must be 'ffill' or None")
+        ):
+            merge_ordered(left, right, on="key", fill_method=invalid_method)
+
+    def test_ffill_left_merge(self):
+        # GH 57010
+        df1 = DataFrame(
+            {
+                "key": ["a", "c", "e", "a", "c", "e"],
+                "lvalue": [1, 2, 3, 1, 2, 3],
+                "group": ["a", "a", "a", "b", "b", "b"],
+            }
+        )
+        df2 = DataFrame({"key": ["b", "c", "d"], "rvalue": [1, 2, 3]})
+        result = merge_ordered(
+            df1, df2, fill_method="ffill", left_by="group", how="left"
+        )
+        expected = DataFrame(
+            {
+                "key": ["a", "c", "e", "a", "c", "e"],
+                "lvalue": [1, 2, 3, 1, 2, 3],
+                "group": ["a", "a", "a", "b", "b", "b"],
+                "rvalue": [np.nan, 2.0, 2.0, np.nan, 2.0, 2.0],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/merge/test_multi.py b/pandas/pandas/tests/reshape/merge/test_multi.py
new file mode 100644
index 0000000000000..7ae2fffa04205
--- /dev/null
+++ b/pandas/pandas/tests/reshape/merge/test_multi.py
@@ -0,0 +1,930 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    Timestamp,
+    option_context,
+)
+import pandas._testing as tm
+from pandas.core.reshape.concat import concat
+from pandas.core.reshape.merge import merge
+
+
+@pytest.fixture
+def left():
+    """left dataframe (not multi-indexed) for multi-index join tests"""
+    # a little relevant example with NAs
+    key1 = ["bar", "bar", "bar", "foo", "foo", "baz", "baz", "qux", "qux", "snap"]
+    key2 = ["two", "one", "three", "one", "two", "one", "two", "two", "three", "one"]
+
+    data = np.random.default_rng(2).standard_normal(len(key1))
+    return DataFrame({"key1": key1, "key2": key2, "data": data})
+
+
+@pytest.fixture
+def right(multiindex_dataframe_random_data):
+    """right dataframe (multi-indexed) for multi-index join tests"""
+    df = multiindex_dataframe_random_data
+    df.index.names = ["key1", "key2"]
+
+    df.columns = ["j_one", "j_two", "j_three"]
+    return df
+
+
+@pytest.fixture
+def left_multi():
+    return DataFrame(
+        {
+            "Origin": ["A", "A", "B", "B", "C"],
+            "Destination": ["A", "B", "A", "C", "A"],
+            "Period": ["AM", "AM", "IP", "AM", "OP"],
+            "TripPurp": ["hbw", "nhb", "hbo", "nhb", "hbw"],
+            "Trips": [1987, 3647, 2470, 4296, 4444],
+        },
+        columns=["Origin", "Destination", "Period", "TripPurp", "Trips"],
+    ).set_index(["Origin", "Destination", "Period", "TripPurp"])
+
+
+@pytest.fixture
+def right_multi():
+    return DataFrame(
+        {
+            "Origin": ["A", "A", "B", "B", "C", "C", "E"],
+            "Destination": ["A", "B", "A", "B", "A", "B", "F"],
+            "Period": ["AM", "AM", "IP", "AM", "OP", "IP", "AM"],
+            "LinkType": ["a", "b", "c", "b", "a", "b", "a"],
+            "Distance": [100, 80, 90, 80, 75, 35, 55],
+        },
+        columns=["Origin", "Destination", "Period", "LinkType", "Distance"],
+    ).set_index(["Origin", "Destination", "Period", "LinkType"])
+
+
+@pytest.fixture
+def on_cols_multi():
+    return ["Origin", "Destination", "Period"]
+
+
+class TestMergeMulti:
+    def test_merge_on_multikey(self, left, right, join_type):
+        on_cols = ["key1", "key2"]
+        result = left.join(right, on=on_cols, how=join_type).reset_index(drop=True)
+
+        expected = merge(left, right.reset_index(), on=on_cols, how=join_type)
+
+        tm.assert_frame_equal(result, expected)
+
+        result = left.join(right, on=on_cols, how=join_type, sort=True).reset_index(
+            drop=True
+        )
+
+        expected = merge(
+            left, right.reset_index(), on=on_cols, how=join_type, sort=True
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+    )
+    def test_left_join_multi_index(self, sort, infer_string):
+        with option_context("future.infer_string", infer_string):
+            icols = ["1st", "2nd", "3rd"]
+
+            def bind_cols(df):
+                iord = lambda a: 0 if a != a else ord(a)
+                f = lambda ts: ts.map(iord) - ord("a")
+                return f(df["1st"]) + f(df["3rd"]) * 1e2 + df["2nd"].fillna(0) * 10
+
+            def run_asserts(left, right, sort):
+                res = left.join(right, on=icols, how="left", sort=sort)
+
+                assert len(left) < len(res) + 1
+                assert not res["4th"].isna().any()
+                assert not res["5th"].isna().any()
+
+                tm.assert_series_equal(res["4th"], -res["5th"], check_names=False)
+                result = bind_cols(res.iloc[:, :-2])
+                tm.assert_series_equal(res["4th"], result, check_names=False)
+                assert result.name is None
+
+                if sort:
+                    tm.assert_frame_equal(res, res.sort_values(icols, kind="mergesort"))
+
+                out = merge(left, right.reset_index(), on=icols, sort=sort, how="left")
+
+                res.index = RangeIndex(len(res))
+                tm.assert_frame_equal(out, res)
+
+            lc = list(map(chr, np.arange(ord("a"), ord("z") + 1)))
+            left = DataFrame(
+                np.random.default_rng(2).choice(lc, (50, 2)), columns=["1st", "3rd"]
+            )
+            # Explicit cast to float to avoid implicit cast when setting nan
+            left.insert(
+                1,
+                "2nd",
+                np.random.default_rng(2).integers(0, 10, len(left)).astype("float"),
+            )
+            right = left.sample(frac=1, random_state=np.random.default_rng(2))
+
+            left["4th"] = bind_cols(left)
+            right["5th"] = -bind_cols(right)
+            right.set_index(icols, inplace=True)
+
+            run_asserts(left, right, sort)
+
+            # inject some nulls
+            left.loc[1::4, "1st"] = np.nan
+            left.loc[2::5, "2nd"] = np.nan
+            left.loc[3::6, "3rd"] = np.nan
+            left["4th"] = bind_cols(left)
+
+            i = np.random.default_rng(2).permutation(len(left))
+            right = left.iloc[i, :-1]
+            right["5th"] = -bind_cols(right)
+            right.set_index(icols, inplace=True)
+
+            run_asserts(left, right, sort)
+
+    def test_merge_right_vs_left(self, left, right, sort):
+        # compare left vs right merge with multikey
+        on_cols = ["key1", "key2"]
+        merged_left_right = left.merge(
+            right, left_on=on_cols, right_index=True, how="left", sort=sort
+        )
+
+        merge_right_left = right.merge(
+            left, right_on=on_cols, left_index=True, how="right", sort=sort
+        )
+
+        # Reorder columns
+        merge_right_left = merge_right_left[merged_left_right.columns]
+
+        tm.assert_frame_equal(merged_left_right, merge_right_left)
+
+    def test_merge_multiple_cols_with_mixed_cols_index(self):
+        # GH29522
+        s = Series(
+            range(6),
+            MultiIndex.from_product([["A", "B"], [1, 2, 3]], names=["lev1", "lev2"]),
+            name="Amount",
+        )
+        df = DataFrame({"lev1": list("AAABBB"), "lev2": [1, 2, 3, 1, 2, 3], "col": 0})
+        result = merge(df, s.reset_index(), on=["lev1", "lev2"])
+        expected = DataFrame(
+            {
+                "lev1": list("AAABBB"),
+                "lev2": [1, 2, 3, 1, 2, 3],
+                "col": [0] * 6,
+                "Amount": range(6),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_compress_group_combinations(self):
+        # ~ 40000000 possible unique groups
+        key1 = [str(i) for i in range(10000)]
+        key1 = np.tile(key1, 2)
+        key2 = key1[::-1]
+
+        df = DataFrame(
+            {
+                "key1": key1,
+                "key2": key2,
+                "value1": np.random.default_rng(2).standard_normal(20000),
+            }
+        )
+
+        df2 = DataFrame(
+            {
+                "key1": key1[::2],
+                "key2": key2[::2],
+                "value2": np.random.default_rng(2).standard_normal(10000),
+            }
+        )
+
+        # just to hit the label compression code path
+        merge(df, df2, how="outer")
+
+    def test_left_join_index_preserve_order(self):
+        on_cols = ["k1", "k2"]
+        left = DataFrame(
+            {
+                "k1": [0, 1, 2] * 8,
+                "k2": ["foo", "bar"] * 12,
+                "v": np.array(np.arange(24), dtype=np.int64),
+            }
+        )
+
+        index = MultiIndex.from_tuples([(2, "bar"), (1, "foo")])
+        right = DataFrame({"v2": [5, 7]}, index=index)
+
+        result = left.join(right, on=on_cols)
+
+        expected = left.copy()
+        expected["v2"] = np.nan
+        expected.loc[(expected.k1 == 2) & (expected.k2 == "bar"), "v2"] = 5
+        expected.loc[(expected.k1 == 1) & (expected.k2 == "foo"), "v2"] = 7
+
+        tm.assert_frame_equal(result, expected)
+
+        result.sort_values(on_cols, kind="mergesort", inplace=True)
+        expected = left.join(right, on=on_cols, sort=True)
+
+        tm.assert_frame_equal(result, expected)
+
+        # test join with multi dtypes blocks
+        left = DataFrame(
+            {
+                "k1": [0, 1, 2] * 8,
+                "k2": ["foo", "bar"] * 12,
+                "k3": np.array([0, 1, 2] * 8, dtype=np.float32),
+                "v": np.array(np.arange(24), dtype=np.int32),
+            }
+        )
+
+        index = MultiIndex.from_tuples([(2, "bar"), (1, "foo")])
+        right = DataFrame({"v2": [5, 7]}, index=index)
+
+        result = left.join(right, on=on_cols)
+
+        expected = left.copy()
+        expected["v2"] = np.nan
+        expected.loc[(expected.k1 == 2) & (expected.k2 == "bar"), "v2"] = 5
+        expected.loc[(expected.k1 == 1) & (expected.k2 == "foo"), "v2"] = 7
+
+        tm.assert_frame_equal(result, expected)
+
+        result = result.sort_values(on_cols, kind="mergesort")
+        expected = left.join(right, on=on_cols, sort=True)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_left_join_index_multi_match_multiindex(self):
+        left = DataFrame(
+            [
+                ["X", "Y", "C", "a"],
+                ["W", "Y", "C", "e"],
+                ["V", "Q", "A", "h"],
+                ["V", "R", "D", "i"],
+                ["X", "Y", "D", "b"],
+                ["X", "Y", "A", "c"],
+                ["W", "Q", "B", "f"],
+                ["W", "R", "C", "g"],
+                ["V", "Y", "C", "j"],
+                ["X", "Y", "B", "d"],
+            ],
+            columns=["cola", "colb", "colc", "tag"],
+            index=[3, 2, 0, 1, 7, 6, 4, 5, 9, 8],
+        )
+
+        right = DataFrame(
+            [
+                ["W", "R", "C", 0],
+                ["W", "Q", "B", 3],
+                ["W", "Q", "B", 8],
+                ["X", "Y", "A", 1],
+                ["X", "Y", "A", 4],
+                ["X", "Y", "B", 5],
+                ["X", "Y", "C", 6],
+                ["X", "Y", "C", 9],
+                ["X", "Q", "C", -6],
+                ["X", "R", "C", -9],
+                ["V", "Y", "C", 7],
+                ["V", "R", "D", 2],
+                ["V", "R", "D", -1],
+                ["V", "Q", "A", -3],
+            ],
+            columns=["col1", "col2", "col3", "val"],
+        ).set_index(["col1", "col2", "col3"])
+
+        result = left.join(right, on=["cola", "colb", "colc"], how="left")
+
+        expected = DataFrame(
+            [
+                ["X", "Y", "C", "a", 6],
+                ["X", "Y", "C", "a", 9],
+                ["W", "Y", "C", "e", np.nan],
+                ["V", "Q", "A", "h", -3],
+                ["V", "R", "D", "i", 2],
+                ["V", "R", "D", "i", -1],
+                ["X", "Y", "D", "b", np.nan],
+                ["X", "Y", "A", "c", 1],
+                ["X", "Y", "A", "c", 4],
+                ["W", "Q", "B", "f", 3],
+                ["W", "Q", "B", "f", 8],
+                ["W", "R", "C", "g", 0],
+                ["V", "Y", "C", "j", 7],
+                ["X", "Y", "B", "d", 5],
+            ],
+            columns=["cola", "colb", "colc", "tag", "val"],
+            index=[3, 3, 2, 0, 1, 1, 7, 6, 6, 4, 4, 5, 9, 8],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        result = left.join(right, on=["cola", "colb", "colc"], how="left", sort=True)
+
+        expected = expected.sort_values(["cola", "colb", "colc"], kind="mergesort")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_left_join_index_multi_match(self):
+        left = DataFrame(
+            [["c", 0], ["b", 1], ["a", 2], ["b", 3]],
+            columns=["tag", "val"],
+            index=[2, 0, 1, 3],
+        )
+
+        right = DataFrame(
+            [
+                ["a", "v"],
+                ["c", "w"],
+                ["c", "x"],
+                ["d", "y"],
+                ["a", "z"],
+                ["c", "r"],
+                ["e", "q"],
+                ["c", "s"],
+            ],
+            columns=["tag", "char"],
+        ).set_index("tag")
+
+        result = left.join(right, on="tag", how="left")
+
+        expected = DataFrame(
+            [
+                ["c", 0, "w"],
+                ["c", 0, "x"],
+                ["c", 0, "r"],
+                ["c", 0, "s"],
+                ["b", 1, np.nan],
+                ["a", 2, "v"],
+                ["a", 2, "z"],
+                ["b", 3, np.nan],
+            ],
+            columns=["tag", "val", "char"],
+            index=[2, 2, 2, 2, 0, 1, 1, 3],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        result = left.join(right, on="tag", how="left", sort=True)
+        expected2 = expected.sort_values("tag", kind="mergesort")
+
+        tm.assert_frame_equal(result, expected2)
+
+        # GH7331 - maintain left frame order in left merge
+        result = merge(left, right.reset_index(), how="left", on="tag")
+        expected.index = RangeIndex(len(expected))
+        tm.assert_frame_equal(result, expected)
+
+    def test_left_merge_na_buglet(self):
+        left = DataFrame(
+            {
+                "id": list("abcde"),
+                "v1": np.random.default_rng(2).standard_normal(5),
+                "v2": np.random.default_rng(2).standard_normal(5),
+                "dummy": list("abcde"),
+                "v3": np.random.default_rng(2).standard_normal(5),
+            },
+            columns=["id", "v1", "v2", "dummy", "v3"],
+        )
+        right = DataFrame(
+            {
+                "id": ["a", "b", np.nan, np.nan, np.nan],
+                "sv3": [1.234, 5.678, np.nan, np.nan, np.nan],
+            }
+        )
+
+        result = merge(left, right, on="id", how="left")
+
+        rdf = right.drop(["id"], axis=1)
+        expected = left.join(rdf)
+        tm.assert_frame_equal(result, expected)
+
+    def test_merge_na_keys(self):
+        data = [
+            [1950, "A", 1.5],
+            [1950, "B", 1.5],
+            [1955, "B", 1.5],
+            [1960, "B", np.nan],
+            [1970, "B", 4.0],
+            [1950, "C", 4.0],
+            [1960, "C", np.nan],
+            [1965, "C", 3.0],
+            [1970, "C", 4.0],
+        ]
+
+        frame = DataFrame(data, columns=["year", "panel", "data"])
+
+        other_data = [
+            [1960, "A", np.nan],
+            [1970, "A", np.nan],
+            [1955, "A", np.nan],
+            [1965, "A", np.nan],
+            [1965, "B", np.nan],
+            [1955, "C", np.nan],
+        ]
+        other = DataFrame(other_data, columns=["year", "panel", "data"])
+
+        result = frame.merge(other, how="outer")
+
+        expected = frame.fillna(-999).merge(other.fillna(-999), how="outer")
+        expected = expected.replace(-999, np.nan)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [None, np.asarray, Series, Index])
+    def test_merge_datetime_index(self, klass):
+        # see gh-19038
+        df = DataFrame(
+            [1, 2, 3], ["2016-01-01", "2017-01-01", "2018-01-01"], columns=["a"]
+        )
+        df.index = pd.to_datetime(df.index)
+        on_vector = df.index.year
+
+        if klass is not None:
+            on_vector = klass(on_vector)
+
+        exp_years = np.array([2016, 2017, 2018], dtype=np.int32)
+        expected = DataFrame({"a": [1, 2, 3], "key_1": exp_years})
+
+        result = df.merge(df, on=["a", on_vector], how="inner")
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame({"key_0": exp_years, "a_x": [1, 2, 3], "a_y": [1, 2, 3]})
+
+        result = df.merge(df, on=[df.index.year], how="inner")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("merge_type", ["left", "right"])
+    def test_merge_datetime_multi_index_empty_df(self, merge_type):
+        # see gh-36895
+
+        left = DataFrame(
+            data={
+                "data": [1.5, 1.5],
+            },
+            index=MultiIndex.from_tuples(
+                [[Timestamp("1950-01-01"), "A"], [Timestamp("1950-01-02"), "B"]],
+                names=["date", "panel"],
+            ),
+        )
+
+        right = DataFrame(
+            index=MultiIndex.from_tuples([], names=["date", "panel"]), columns=["state"]
+        )
+
+        expected_index = MultiIndex.from_tuples(
+            [[Timestamp("1950-01-01"), "A"], [Timestamp("1950-01-02"), "B"]],
+            names=["date", "panel"],
+        )
+
+        if merge_type == "left":
+            expected = DataFrame(
+                data={
+                    "data": [1.5, 1.5],
+                    "state": np.array([np.nan, np.nan], dtype=object),
+                },
+                index=expected_index,
+            )
+            results_merge = left.merge(right, how="left", on=["date", "panel"])
+            results_join = left.join(right, how="left")
+        else:
+            expected = DataFrame(
+                data={
+                    "state": np.array([np.nan, np.nan], dtype=object),
+                    "data": [1.5, 1.5],
+                },
+                index=expected_index,
+            )
+            results_merge = right.merge(left, how="right", on=["date", "panel"])
+            results_join = right.join(left, how="right")
+
+        tm.assert_frame_equal(results_merge, expected)
+        tm.assert_frame_equal(results_join, expected)
+
+    @pytest.fixture
+    def household(self):
+        household = DataFrame(
+            {
+                "household_id": [1, 2, 3],
+                "male": [0, 1, 0],
+                "wealth": [196087.3, 316478.7, 294750],
+            },
+            columns=["household_id", "male", "wealth"],
+        ).set_index("household_id")
+        return household
+
+    @pytest.fixture
+    def portfolio(self):
+        portfolio = DataFrame(
+            {
+                "household_id": [1, 2, 2, 3, 3, 3, 4],
+                "asset_id": [
+                    "nl0000301109",
+                    "nl0000289783",
+                    "gb00b03mlx29",
+                    "gb00b03mlx29",
+                    "lu0197800237",
+                    "nl0000289965",
+                    np.nan,
+                ],
+                "name": [
+                    "ABN Amro",
+                    "Robeco",
+                    "Royal Dutch Shell",
+                    "Royal Dutch Shell",
+                    "AAB Eastern Europe Equity Fund",
+                    "Postbank BioTech Fonds",
+                    np.nan,
+                ],
+                "share": [1.0, 0.4, 0.6, 0.15, 0.6, 0.25, 1.0],
+            },
+            columns=["household_id", "asset_id", "name", "share"],
+        ).set_index(["household_id", "asset_id"])
+        return portfolio
+
+    @pytest.fixture
+    def expected(self):
+        expected = (
+            DataFrame(
+                {
+                    "male": [0, 1, 1, 0, 0, 0],
+                    "wealth": [
+                        196087.3,
+                        316478.7,
+                        316478.7,
+                        294750.0,
+                        294750.0,
+                        294750.0,
+                    ],
+                    "name": [
+                        "ABN Amro",
+                        "Robeco",
+                        "Royal Dutch Shell",
+                        "Royal Dutch Shell",
+                        "AAB Eastern Europe Equity Fund",
+                        "Postbank BioTech Fonds",
+                    ],
+                    "share": [1.00, 0.40, 0.60, 0.15, 0.60, 0.25],
+                    "household_id": [1, 2, 2, 3, 3, 3],
+                    "asset_id": [
+                        "nl0000301109",
+                        "nl0000289783",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "lu0197800237",
+                        "nl0000289965",
+                    ],
+                }
+            )
+            .set_index(["household_id", "asset_id"])
+            .reindex(columns=["male", "wealth", "name", "share"])
+        )
+        return expected
+
+    def test_join_multi_levels(self, portfolio, household, expected):
+        portfolio = portfolio.copy()
+        household = household.copy()
+
+        # GH 3662
+        # merge multi-levels
+        result = household.join(portfolio, how="inner")
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_multi_levels_merge_equivalence(self, portfolio, household, expected):
+        portfolio = portfolio.copy()
+        household = household.copy()
+
+        # equivalency
+        result = merge(
+            household.reset_index(),
+            portfolio.reset_index(),
+            on=["household_id"],
+            how="inner",
+        ).set_index(["household_id", "asset_id"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_multi_levels_outer(self, portfolio, household, expected):
+        portfolio = portfolio.copy()
+        household = household.copy()
+
+        result = household.join(portfolio, how="outer")
+        expected = concat(
+            [
+                expected,
+                (
+                    DataFrame(
+                        {"share": [1.00]},
+                        index=MultiIndex.from_tuples(
+                            [(4, np.nan)], names=["household_id", "asset_id"]
+                        ),
+                    )
+                ),
+            ],
+            axis=0,
+            sort=True,
+        ).reindex(columns=expected.columns)
+        tm.assert_frame_equal(result, expected, check_index_type=False)
+
+    def test_join_multi_levels_invalid(self, portfolio, household):
+        portfolio = portfolio.copy()
+        household = household.copy()
+
+        # invalid cases
+        household.index.name = "foo"
+
+        with pytest.raises(
+            ValueError, match="cannot join with no overlapping index names"
+        ):
+            household.join(portfolio, how="inner")
+
+        portfolio2 = portfolio.copy()
+        portfolio2.index.set_names(["household_id", "foo"])
+
+        with pytest.raises(ValueError, match="columns overlap but no suffix specified"):
+            portfolio2.join(portfolio, how="inner")
+
+    def test_join_multi_levels2(self):
+        # some more advanced merges
+        # GH6360
+        household = DataFrame(
+            {
+                "household_id": [1, 2, 2, 3, 3, 3, 4],
+                "asset_id": [
+                    "nl0000301109",
+                    "nl0000301109",
+                    "gb00b03mlx29",
+                    "gb00b03mlx29",
+                    "lu0197800237",
+                    "nl0000289965",
+                    np.nan,
+                ],
+                "share": [1.0, 0.4, 0.6, 0.15, 0.6, 0.25, 1.0],
+            },
+            columns=["household_id", "asset_id", "share"],
+        ).set_index(["household_id", "asset_id"])
+
+        log_return = DataFrame(
+            {
+                "asset_id": [
+                    "gb00b03mlx29",
+                    "gb00b03mlx29",
+                    "gb00b03mlx29",
+                    "lu0197800237",
+                    "lu0197800237",
+                ],
+                "t": [233, 234, 235, 180, 181],
+                "log_return": [
+                    0.09604978,
+                    -0.06524096,
+                    0.03532373,
+                    0.03025441,
+                    0.036997,
+                ],
+            }
+        ).set_index(["asset_id", "t"])
+
+        expected = (
+            DataFrame(
+                {
+                    "household_id": [2, 2, 2, 3, 3, 3, 3, 3],
+                    "asset_id": [
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "lu0197800237",
+                        "lu0197800237",
+                    ],
+                    "t": [233, 234, 235, 233, 234, 235, 180, 181],
+                    "share": [0.6, 0.6, 0.6, 0.15, 0.15, 0.15, 0.6, 0.6],
+                    "log_return": [
+                        0.09604978,
+                        -0.06524096,
+                        0.03532373,
+                        0.09604978,
+                        -0.06524096,
+                        0.03532373,
+                        0.03025441,
+                        0.036997,
+                    ],
+                }
+            )
+            .set_index(["household_id", "asset_id", "t"])
+            .reindex(columns=["share", "log_return"])
+        )
+
+        # this is the equivalency
+        result = merge(
+            household.reset_index(),
+            log_return.reset_index(),
+            on=["asset_id"],
+            how="inner",
+        ).set_index(["household_id", "asset_id", "t"])
+        tm.assert_frame_equal(result, expected)
+
+        expected = (
+            DataFrame(
+                {
+                    "household_id": [2, 2, 2, 3, 3, 3, 3, 3, 3, 1, 2, 4],
+                    "asset_id": [
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "gb00b03mlx29",
+                        "lu0197800237",
+                        "lu0197800237",
+                        "nl0000289965",
+                        "nl0000301109",
+                        "nl0000301109",
+                        None,
+                    ],
+                    "t": [
+                        233,
+                        234,
+                        235,
+                        233,
+                        234,
+                        235,
+                        180,
+                        181,
+                        None,
+                        None,
+                        None,
+                        None,
+                    ],
+                    "share": [
+                        0.6,
+                        0.6,
+                        0.6,
+                        0.15,
+                        0.15,
+                        0.15,
+                        0.6,
+                        0.6,
+                        0.25,
+                        1.0,
+                        0.4,
+                        1.0,
+                    ],
+                    "log_return": [
+                        0.09604978,
+                        -0.06524096,
+                        0.03532373,
+                        0.09604978,
+                        -0.06524096,
+                        0.03532373,
+                        0.03025441,
+                        0.036997,
+                        None,
+                        None,
+                        None,
+                        None,
+                    ],
+                }
+            )
+            .set_index(["household_id", "asset_id", "t"])
+            .reindex(columns=["share", "log_return"])
+        )
+
+        result = merge(
+            household.reset_index(),
+            log_return.reset_index(),
+            on=["asset_id"],
+            how="outer",
+        ).set_index(["household_id", "asset_id", "t"])
+
+        tm.assert_frame_equal(result, expected)
+
+
+class TestJoinMultiMulti:
+    def test_join_multi_multi(self, left_multi, right_multi, join_type, on_cols_multi):
+        left_names = left_multi.index.names
+        right_names = right_multi.index.names
+        if join_type == "right":
+            level_order = right_names + left_names.difference(right_names)
+        else:
+            level_order = left_names + right_names.difference(left_names)
+        # Multi-index join tests
+        expected = (
+            merge(
+                left_multi.reset_index(),
+                right_multi.reset_index(),
+                how=join_type,
+                on=on_cols_multi,
+            )
+            .set_index(level_order)
+            .sort_index()
+        )
+
+        result = left_multi.join(right_multi, how=join_type).sort_index()
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_multi_empty_frames(
+        self, left_multi, right_multi, join_type, on_cols_multi
+    ):
+        left_multi = left_multi.drop(columns=left_multi.columns)
+        right_multi = right_multi.drop(columns=right_multi.columns)
+
+        left_names = left_multi.index.names
+        right_names = right_multi.index.names
+        if join_type == "right":
+            level_order = right_names + left_names.difference(right_names)
+        else:
+            level_order = left_names + right_names.difference(left_names)
+
+        expected = (
+            merge(
+                left_multi.reset_index(),
+                right_multi.reset_index(),
+                how=join_type,
+                on=on_cols_multi,
+            )
+            .set_index(level_order)
+            .sort_index()
+        )
+
+        result = left_multi.join(right_multi, how=join_type).sort_index()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [None, np.asarray, Series, Index])
+    def test_merge_datetime_index(self, box):
+        # see gh-19038
+        df = DataFrame(
+            [1, 2, 3], ["2016-01-01", "2017-01-01", "2018-01-01"], columns=["a"]
+        )
+        df.index = pd.to_datetime(df.index)
+        on_vector = df.index.year
+
+        if box is not None:
+            on_vector = box(on_vector)
+
+        exp_years = np.array([2016, 2017, 2018], dtype=np.int32)
+        expected = DataFrame({"a": [1, 2, 3], "key_1": exp_years})
+
+        result = df.merge(df, on=["a", on_vector], how="inner")
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame({"key_0": exp_years, "a_x": [1, 2, 3], "a_y": [1, 2, 3]})
+
+        result = df.merge(df, on=[df.index.year], how="inner")
+        tm.assert_frame_equal(result, expected)
+
+    def test_single_common_level(self):
+        index_left = MultiIndex.from_tuples(
+            [("K0", "X0"), ("K0", "X1"), ("K1", "X2")], names=["key", "X"]
+        )
+
+        left = DataFrame(
+            {"A": ["A0", "A1", "A2"], "B": ["B0", "B1", "B2"]}, index=index_left
+        )
+
+        index_right = MultiIndex.from_tuples(
+            [("K0", "Y0"), ("K1", "Y1"), ("K2", "Y2"), ("K2", "Y3")], names=["key", "Y"]
+        )
+
+        right = DataFrame(
+            {"C": ["C0", "C1", "C2", "C3"], "D": ["D0", "D1", "D2", "D3"]},
+            index=index_right,
+        )
+
+        result = left.join(right)
+        expected = merge(
+            left.reset_index(), right.reset_index(), on=["key"], how="inner"
+        ).set_index(["key", "X", "Y"])
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_join_multi_wrong_order(self):
+        # GH 25760
+        # GH 28956
+
+        midx1 = MultiIndex.from_product([[1, 2], [3, 4]], names=["a", "b"])
+        midx3 = MultiIndex.from_tuples([(4, 1), (3, 2), (3, 1)], names=["b", "a"])
+
+        left = DataFrame(index=midx1, data={"x": [10, 20, 30, 40]})
+        right = DataFrame(index=midx3, data={"y": ["foo", "bar", "fing"]})
+
+        result = left.join(right)
+
+        expected = DataFrame(
+            index=midx1,
+            data={"x": [10, 20, 30, 40], "y": ["fing", "foo", "bar", np.nan]},
+        )
+
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_crosstab.py b/pandas/pandas/tests/reshape/test_crosstab.py
new file mode 100644
index 0000000000000..070c756e8c928
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_crosstab.py
@@ -0,0 +1,879 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    crosstab,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df():
+    df = DataFrame(
+        {
+            "A": [
+                "foo",
+                "foo",
+                "foo",
+                "foo",
+                "bar",
+                "bar",
+                "bar",
+                "bar",
+                "foo",
+                "foo",
+                "foo",
+            ],
+            "B": [
+                "one",
+                "one",
+                "one",
+                "two",
+                "one",
+                "one",
+                "one",
+                "two",
+                "two",
+                "two",
+                "one",
+            ],
+            "C": [
+                "dull",
+                "dull",
+                "shiny",
+                "dull",
+                "dull",
+                "shiny",
+                "shiny",
+                "dull",
+                "shiny",
+                "shiny",
+                "shiny",
+            ],
+            "D": np.random.default_rng(2).standard_normal(11),
+            "E": np.random.default_rng(2).standard_normal(11),
+            "F": np.random.default_rng(2).standard_normal(11),
+        }
+    )
+
+    return pd.concat([df, df], ignore_index=True)
+
+
+class TestCrosstab:
+    def test_crosstab_single(self, df):
+        result = crosstab(df["A"], df["C"])
+        expected = df.groupby(["A", "C"]).size().unstack()
+        tm.assert_frame_equal(result, expected.fillna(0).astype(np.int64))
+
+    def test_crosstab_multiple(self, df):
+        result = crosstab(df["A"], [df["B"], df["C"]])
+        expected = df.groupby(["A", "B", "C"]).size()
+        expected = expected.unstack("B").unstack("C").fillna(0).astype(np.int64)
+        tm.assert_frame_equal(result, expected)
+
+        result = crosstab([df["B"], df["C"]], df["A"])
+        expected = df.groupby(["B", "C", "A"]).size()
+        expected = expected.unstack("A").fillna(0).astype(np.int64)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [np.array, list, tuple])
+    def test_crosstab_ndarray(self, box):
+        # GH 44076
+        a = box(np.random.default_rng(2).integers(0, 5, size=100))
+        b = box(np.random.default_rng(2).integers(0, 3, size=100))
+        c = box(np.random.default_rng(2).integers(0, 10, size=100))
+
+        df = DataFrame({"a": a, "b": b, "c": c})
+
+        result = crosstab(a, [b, c], rownames=["a"], colnames=("b", "c"))
+        expected = crosstab(df["a"], [df["b"], df["c"]])
+        tm.assert_frame_equal(result, expected)
+
+        result = crosstab([b, c], a, colnames=["a"], rownames=("b", "c"))
+        expected = crosstab([df["b"], df["c"]], df["a"])
+        tm.assert_frame_equal(result, expected)
+
+        # assign arbitrary names
+        result = crosstab(a, c)
+        expected = crosstab(df["a"], df["c"])
+        expected.index.names = ["row_0"]
+        expected.columns.names = ["col_0"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_non_aligned(self):
+        # GH 17005
+        a = Series([0, 1, 1], index=["a", "b", "c"])
+        b = Series([3, 4, 3, 4, 3], index=["a", "b", "c", "d", "f"])
+        c = np.array([3, 4, 3], dtype=np.int64)
+
+        expected = DataFrame(
+            [[1, 0], [1, 1]],
+            index=Index([0, 1], name="row_0"),
+            columns=Index([3, 4], name="col_0"),
+        )
+
+        result = crosstab(a, b)
+        tm.assert_frame_equal(result, expected)
+
+        result = crosstab(a, c)
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_margins(self):
+        a = np.random.default_rng(2).integers(0, 7, size=100)
+        b = np.random.default_rng(2).integers(0, 3, size=100)
+        c = np.random.default_rng(2).integers(0, 5, size=100)
+
+        df = DataFrame({"a": a, "b": b, "c": c})
+
+        result = crosstab(a, [b, c], rownames=["a"], colnames=("b", "c"), margins=True)
+
+        assert result.index.names == ("a",)
+        assert result.columns.names == ["b", "c"]
+
+        all_cols = result["All", ""]
+        exp_cols = df.groupby(["a"]).size().astype("i8")
+        # to keep index.name
+        exp_margin = Series([len(df)], index=Index(["All"], name="a"))
+        exp_cols = pd.concat([exp_cols, exp_margin])
+        exp_cols.name = ("All", "")
+
+        tm.assert_series_equal(all_cols, exp_cols)
+
+        all_rows = result.loc["All"]
+        exp_rows = df.groupby(["b", "c"]).size().astype("i8")
+        exp_rows = pd.concat([exp_rows, Series([len(df)], index=[("All", "")])])
+        exp_rows.name = "All"
+
+        exp_rows = exp_rows.reindex(all_rows.index)
+        exp_rows = exp_rows.fillna(0).astype(np.int64)
+        tm.assert_series_equal(all_rows, exp_rows)
+
+    def test_crosstab_margins_set_margin_name(self):
+        # GH 15972
+        a = np.random.default_rng(2).integers(0, 7, size=100)
+        b = np.random.default_rng(2).integers(0, 3, size=100)
+        c = np.random.default_rng(2).integers(0, 5, size=100)
+
+        df = DataFrame({"a": a, "b": b, "c": c})
+
+        result = crosstab(
+            a,
+            [b, c],
+            rownames=["a"],
+            colnames=("b", "c"),
+            margins=True,
+            margins_name="TOTAL",
+        )
+
+        assert result.index.names == ("a",)
+        assert result.columns.names == ["b", "c"]
+
+        all_cols = result["TOTAL", ""]
+        exp_cols = df.groupby(["a"]).size().astype("i8")
+        # to keep index.name
+        exp_margin = Series([len(df)], index=Index(["TOTAL"], name="a"))
+        exp_cols = pd.concat([exp_cols, exp_margin])
+        exp_cols.name = ("TOTAL", "")
+
+        tm.assert_series_equal(all_cols, exp_cols)
+
+        all_rows = result.loc["TOTAL"]
+        exp_rows = df.groupby(["b", "c"]).size().astype("i8")
+        exp_rows = pd.concat([exp_rows, Series([len(df)], index=[("TOTAL", "")])])
+        exp_rows.name = "TOTAL"
+
+        exp_rows = exp_rows.reindex(all_rows.index)
+        exp_rows = exp_rows.fillna(0).astype(np.int64)
+        tm.assert_series_equal(all_rows, exp_rows)
+
+        msg = "margins_name argument must be a string"
+        for margins_name in [666, None, ["a", "b"]]:
+            with pytest.raises(ValueError, match=msg):
+                crosstab(
+                    a,
+                    [b, c],
+                    rownames=["a"],
+                    colnames=("b", "c"),
+                    margins=True,
+                    margins_name=margins_name,
+                )
+
+    def test_crosstab_pass_values(self):
+        a = np.random.default_rng(2).integers(0, 7, size=100)
+        b = np.random.default_rng(2).integers(0, 3, size=100)
+        c = np.random.default_rng(2).integers(0, 5, size=100)
+        values = np.random.default_rng(2).standard_normal(100)
+
+        table = crosstab(
+            [a, b], c, values, aggfunc="sum", rownames=["foo", "bar"], colnames=["baz"]
+        )
+
+        df = DataFrame({"foo": a, "bar": b, "baz": c, "values": values})
+
+        expected = df.pivot_table(
+            "values", index=["foo", "bar"], columns="baz", aggfunc="sum"
+        )
+        tm.assert_frame_equal(table, expected)
+
+    def test_crosstab_dropna(self):
+        # GH 3820
+        a = np.array(["foo", "foo", "foo", "bar", "bar", "foo", "foo"], dtype=object)
+        b = np.array(["one", "one", "two", "one", "two", "two", "two"], dtype=object)
+        c = np.array(
+            ["dull", "dull", "dull", "dull", "dull", "shiny", "shiny"], dtype=object
+        )
+        res = crosstab(a, [b, c], rownames=["a"], colnames=["b", "c"], dropna=False)
+        m = MultiIndex.from_tuples(
+            [("one", "dull"), ("one", "shiny"), ("two", "dull"), ("two", "shiny")],
+            names=["b", "c"],
+        )
+        tm.assert_index_equal(res.columns, m)
+
+    def test_crosstab_no_overlap(self):
+        # GS 10291
+
+        s1 = Series([1, 2, 3], index=[1, 2, 3])
+        s2 = Series([4, 5, 6], index=[4, 5, 6])
+
+        actual = crosstab(s1, s2)
+        expected = DataFrame(
+            index=Index([], dtype="int64", name="row_0"),
+            columns=Index([], dtype="int64", name="col_0"),
+        )
+
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna(self):
+        # GH 12577
+        # pivot_table counts null into margin ('All')
+        # when margins=true and dropna=true
+
+        df = DataFrame({"a": [1, 2, 2, 2, 2, np.nan], "b": [3, 3, 4, 4, 4, 4]})
+        actual = crosstab(df.a, df.b, margins=True, dropna=True)
+        expected = DataFrame([[1, 0, 1], [1, 3, 4], [2, 3, 5]])
+        expected.index = Index([1.0, 2.0, "All"], name="a")
+        expected.columns = Index([3, 4, "All"], name="b")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna2(self):
+        df = DataFrame(
+            {"a": [1, np.nan, np.nan, np.nan, 2, np.nan], "b": [3, np.nan, 4, 4, 4, 4]}
+        )
+        actual = crosstab(df.a, df.b, margins=True, dropna=True)
+        expected = DataFrame([[1, 0, 1], [0, 1, 1], [1, 1, 2]])
+        expected.index = Index([1.0, 2.0, "All"], name="a")
+        expected.columns = Index([3.0, 4.0, "All"], name="b")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna3(self):
+        df = DataFrame(
+            {"a": [1, np.nan, np.nan, np.nan, np.nan, 2], "b": [3, 3, 4, 4, 4, 4]}
+        )
+        actual = crosstab(df.a, df.b, margins=True, dropna=True)
+        expected = DataFrame([[1, 0, 1], [0, 1, 1], [1, 1, 2]])
+        expected.index = Index([1.0, 2.0, "All"], name="a")
+        expected.columns = Index([3, 4, "All"], name="b")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna4(self):
+        # GH 12642
+        # _add_margins raises KeyError: Level None not found
+        # when margins=True and dropna=False
+        # GH: 10772: Keep np.nan in result with dropna=False
+        df = DataFrame({"a": [1, 2, 2, 2, 2, np.nan], "b": [3, 3, 4, 4, 4, 4]})
+        actual = crosstab(df.a, df.b, margins=True, dropna=False)
+        expected = DataFrame([[1, 0, 1.0], [1, 3, 4.0], [0, 1, np.nan], [2, 4, 6.0]])
+        expected.index = Index([1.0, 2.0, np.nan, "All"], name="a")
+        expected.columns = Index([3, 4, "All"], name="b")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna5(self):
+        # GH: 10772: Keep np.nan in result with dropna=False
+        df = DataFrame(
+            {"a": [1, np.nan, np.nan, np.nan, 2, np.nan], "b": [3, np.nan, 4, 4, 4, 4]}
+        )
+        actual = crosstab(df.a, df.b, margins=True, dropna=False)
+        expected = DataFrame(
+            [[1, 0, 0, 1.0], [0, 1, 0, 1.0], [0, 3, 1, np.nan], [1, 4, 0, 6.0]]
+        )
+        expected.index = Index([1.0, 2.0, np.nan, "All"], name="a")
+        expected.columns = Index([3.0, 4.0, np.nan, "All"], name="b")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_margin_dropna6(self):
+        # GH: 10772: Keep np.nan in result with dropna=False
+        a = np.array(["foo", "foo", "foo", "bar", "bar", "foo", "foo"], dtype=object)
+        b = np.array(["one", "one", "two", "one", "two", np.nan, "two"], dtype=object)
+        c = np.array(
+            ["dull", "dull", "dull", "dull", "dull", "shiny", "shiny"], dtype=object
+        )
+
+        actual = crosstab(
+            a, [b, c], rownames=["a"], colnames=["b", "c"], margins=True, dropna=False
+        )
+        m = MultiIndex.from_arrays(
+            [
+                ["one", "one", "two", "two", np.nan, np.nan, "All"],
+                ["dull", "shiny", "dull", "shiny", "dull", "shiny", ""],
+            ],
+            names=["b", "c"],
+        )
+        expected = DataFrame(
+            [[1, 0, 1, 0, 0, 0, 2], [2, 0, 1, 1, 0, 1, 5], [3, 0, 2, 1, 0, 0, 7]],
+            columns=m,
+        )
+        expected.index = Index(["bar", "foo", "All"], name="a")
+        tm.assert_frame_equal(actual, expected)
+
+        actual = crosstab(
+            [a, b], c, rownames=["a", "b"], colnames=["c"], margins=True, dropna=False
+        )
+        m = MultiIndex.from_arrays(
+            [
+                ["bar", "bar", "bar", "foo", "foo", "foo", "All"],
+                ["one", "two", np.nan, "one", "two", np.nan, ""],
+            ],
+            names=["a", "b"],
+        )
+        expected = DataFrame(
+            [
+                [1, 0, 1.0],
+                [1, 0, 1.0],
+                [0, 0, np.nan],
+                [2, 0, 2.0],
+                [1, 1, 2.0],
+                [0, 1, np.nan],
+                [5, 2, 7.0],
+            ],
+            index=m,
+        )
+        expected.columns = Index(["dull", "shiny", "All"], name="c")
+        tm.assert_frame_equal(actual, expected)
+
+        actual = crosstab(
+            [a, b], c, rownames=["a", "b"], colnames=["c"], margins=True, dropna=True
+        )
+        m = MultiIndex.from_arrays(
+            [["bar", "bar", "foo", "foo", "All"], ["one", "two", "one", "two", ""]],
+            names=["a", "b"],
+        )
+        expected = DataFrame(
+            [[1, 0, 1], [1, 0, 1], [2, 0, 2], [1, 1, 2], [5, 1, 6]], index=m
+        )
+        expected.columns = Index(["dull", "shiny", "All"], name="c")
+        tm.assert_frame_equal(actual, expected)
+
+    def test_crosstab_normalize(self):
+        # Issue 12578
+        df = DataFrame(
+            {"a": [1, 2, 2, 2, 2], "b": [3, 3, 4, 4, 4], "c": [1, 1, np.nan, 1, 1]}
+        )
+
+        rindex = Index([1, 2], name="a")
+        cindex = Index([3, 4], name="b")
+        full_normal = DataFrame([[0.2, 0], [0.2, 0.6]], index=rindex, columns=cindex)
+        row_normal = DataFrame([[1.0, 0], [0.25, 0.75]], index=rindex, columns=cindex)
+        col_normal = DataFrame([[0.5, 0], [0.5, 1.0]], index=rindex, columns=cindex)
+
+        # Check all normalize args
+        tm.assert_frame_equal(crosstab(df.a, df.b, normalize="all"), full_normal)
+        tm.assert_frame_equal(crosstab(df.a, df.b, normalize=True), full_normal)
+        tm.assert_frame_equal(crosstab(df.a, df.b, normalize="index"), row_normal)
+        tm.assert_frame_equal(crosstab(df.a, df.b, normalize="columns"), col_normal)
+        tm.assert_frame_equal(
+            crosstab(df.a, df.b, normalize=1),
+            crosstab(df.a, df.b, normalize="columns"),
+        )
+        tm.assert_frame_equal(
+            crosstab(df.a, df.b, normalize=0), crosstab(df.a, df.b, normalize="index")
+        )
+
+        row_normal_margins = DataFrame(
+            [[1.0, 0], [0.25, 0.75], [0.4, 0.6]],
+            index=Index([1, 2, "All"], name="a", dtype="object"),
+            columns=Index([3, 4], name="b", dtype="object"),
+        )
+        col_normal_margins = DataFrame(
+            [[0.5, 0, 0.2], [0.5, 1.0, 0.8]],
+            index=Index([1, 2], name="a", dtype="object"),
+            columns=Index([3, 4, "All"], name="b", dtype="object"),
+        )
+
+        all_normal_margins = DataFrame(
+            [[0.2, 0, 0.2], [0.2, 0.6, 0.8], [0.4, 0.6, 1]],
+            index=Index([1, 2, "All"], name="a", dtype="object"),
+            columns=Index([3, 4, "All"], name="b", dtype="object"),
+        )
+        tm.assert_frame_equal(
+            crosstab(df.a, df.b, normalize="index", margins=True), row_normal_margins
+        )
+        tm.assert_frame_equal(
+            crosstab(df.a, df.b, normalize="columns", margins=True), col_normal_margins
+        )
+        tm.assert_frame_equal(
+            crosstab(df.a, df.b, normalize=True, margins=True), all_normal_margins
+        )
+
+    def test_crosstab_normalize_arrays(self):
+        # GH#12578
+        df = DataFrame(
+            {"a": [1, 2, 2, 2, 2], "b": [3, 3, 4, 4, 4], "c": [1, 1, np.nan, 1, 1]}
+        )
+
+        # Test arrays
+        crosstab(
+            [np.array([1, 1, 2, 2]), np.array([1, 2, 1, 2])], np.array([1, 2, 1, 2])
+        )
+
+        # Test with aggfunc
+        norm_counts = DataFrame(
+            [[0.25, 0, 0.25], [0.25, 0.5, 0.75], [0.5, 0.5, 1]],
+            index=Index([1, 2, "All"], name="a", dtype="object"),
+            columns=Index([3, 4, "All"], name="b"),
+        )
+        test_case = crosstab(
+            df.a, df.b, df.c, aggfunc="count", normalize="all", margins=True
+        )
+        tm.assert_frame_equal(test_case, norm_counts)
+
+        df = DataFrame(
+            {"a": [1, 2, 2, 2, 2], "b": [3, 3, 4, 4, 4], "c": [0, 4, np.nan, 3, 3]}
+        )
+
+        norm_sum = DataFrame(
+            [[0, 0, 0.0], [0.4, 0.6, 1], [0.4, 0.6, 1]],
+            index=Index([1, 2, "All"], name="a", dtype="object"),
+            columns=Index([3, 4, "All"], name="b", dtype="object"),
+        )
+        test_case = crosstab(
+            df.a, df.b, df.c, aggfunc=np.sum, normalize="all", margins=True
+        )
+        tm.assert_frame_equal(test_case, norm_sum)
+
+    def test_crosstab_with_empties(self):
+        # Check handling of empties
+        df = DataFrame(
+            {
+                "a": [1, 2, 2, 2, 2],
+                "b": [3, 3, 4, 4, 4],
+                "c": [np.nan, np.nan, np.nan, np.nan, np.nan],
+            }
+        )
+
+        empty = DataFrame(
+            [[0.0, 0.0], [0.0, 0.0]],
+            index=Index([1, 2], name="a", dtype="int64"),
+            columns=Index([3, 4], name="b"),
+        )
+
+        for i in [True, "index", "columns"]:
+            calculated = crosstab(df.a, df.b, values=df.c, aggfunc="count", normalize=i)
+            tm.assert_frame_equal(empty, calculated)
+
+        nans = DataFrame(
+            [[0.0, np.nan], [0.0, 0.0]],
+            index=Index([1, 2], name="a", dtype="int64"),
+            columns=Index([3, 4], name="b"),
+        )
+
+        calculated = crosstab(df.a, df.b, values=df.c, aggfunc="count", normalize=False)
+        tm.assert_frame_equal(nans, calculated)
+
+    def test_crosstab_errors(self):
+        # Issue 12578
+
+        df = DataFrame(
+            {"a": [1, 2, 2, 2, 2], "b": [3, 3, 4, 4, 4], "c": [1, 1, np.nan, 1, 1]}
+        )
+
+        error = "values cannot be used without an aggfunc."
+        with pytest.raises(ValueError, match=error):
+            crosstab(df.a, df.b, values=df.c)
+
+        error = "aggfunc cannot be used without values"
+        with pytest.raises(ValueError, match=error):
+            crosstab(df.a, df.b, aggfunc=np.mean)
+
+        error = "Not a valid normalize argument"
+        with pytest.raises(ValueError, match=error):
+            crosstab(df.a, df.b, normalize="42")
+
+        with pytest.raises(ValueError, match=error):
+            crosstab(df.a, df.b, normalize=42)
+
+        error = "Not a valid margins argument"
+        with pytest.raises(ValueError, match=error):
+            crosstab(df.a, df.b, normalize="all", margins=42)
+
+    def test_crosstab_with_categorial_columns(self):
+        # GH 8860
+        df = DataFrame(
+            {
+                "MAKE": ["Honda", "Acura", "Tesla", "Honda", "Honda", "Acura"],
+                "MODEL": ["Sedan", "Sedan", "Electric", "Pickup", "Sedan", "Sedan"],
+            }
+        )
+        categories = ["Sedan", "Electric", "Pickup"]
+        df["MODEL"] = df["MODEL"].astype("category").cat.set_categories(categories)
+        result = crosstab(df["MAKE"], df["MODEL"])
+
+        expected_index = Index(["Acura", "Honda", "Tesla"], name="MAKE")
+        expected_columns = CategoricalIndex(
+            categories, categories=categories, ordered=False, name="MODEL"
+        )
+        expected_data = [[2, 0, 0], [2, 0, 1], [0, 1, 0]]
+        expected = DataFrame(
+            expected_data, index=expected_index, columns=expected_columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_with_numpy_size(self):
+        # GH 4003
+        df = DataFrame(
+            {
+                "A": ["one", "one", "two", "three"] * 6,
+                "B": ["A", "B", "C"] * 8,
+                "C": ["foo", "foo", "foo", "bar", "bar", "bar"] * 4,
+                "D": np.random.default_rng(2).standard_normal(24),
+                "E": np.random.default_rng(2).standard_normal(24),
+            }
+        )
+        result = crosstab(
+            index=[df["A"], df["B"]],
+            columns=[df["C"]],
+            margins=True,
+            aggfunc=np.size,
+            values=df["D"],
+        )
+        expected_index = MultiIndex(
+            levels=[["All", "one", "three", "two"], ["", "A", "B", "C"]],
+            codes=[[1, 1, 1, 2, 2, 2, 3, 3, 3, 0], [1, 2, 3, 1, 2, 3, 1, 2, 3, 0]],
+            names=["A", "B"],
+        )
+        expected_column = Index(["bar", "foo", "All"], name="C")
+        expected_data = np.array(
+            [
+                [2.0, 2.0, 4.0],
+                [2.0, 2.0, 4.0],
+                [2.0, 2.0, 4.0],
+                [2.0, np.nan, 2.0],
+                [np.nan, 2.0, 2.0],
+                [2.0, np.nan, 2.0],
+                [np.nan, 2.0, 2.0],
+                [2.0, np.nan, 2.0],
+                [np.nan, 2.0, 2.0],
+                [12.0, 12.0, 24.0],
+            ]
+        )
+        expected = DataFrame(
+            expected_data, index=expected_index, columns=expected_column
+        )
+        # aggfunc is np.size, resulting in integers
+        expected["All"] = expected["All"].astype("int64")
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_duplicate_names(self):
+        # GH 13279 / 22529
+
+        s1 = Series(range(3), name="foo")
+        s2_foo = Series(range(1, 4), name="foo")
+        s2_bar = Series(range(1, 4), name="bar")
+        s3 = Series(range(3), name="waldo")
+
+        # check result computed with duplicate labels against
+        # result computed with unique labels, then relabelled
+        mapper = {"bar": "foo"}
+
+        # duplicate row, column labels
+        result = crosstab(s1, s2_foo)
+        expected = crosstab(s1, s2_bar).rename_axis(columns=mapper, axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # duplicate row, unique column labels
+        result = crosstab([s1, s2_foo], s3)
+        expected = crosstab([s1, s2_bar], s3).rename_axis(index=mapper, axis=0)
+        tm.assert_frame_equal(result, expected)
+
+        # unique row, duplicate column labels
+        result = crosstab(s3, [s1, s2_foo])
+        expected = crosstab(s3, [s1, s2_bar]).rename_axis(columns=mapper, axis=1)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("names", [["a", ("b", "c")], [("a", "b"), "c"]])
+    def test_crosstab_tuple_name(self, names):
+        s1 = Series(range(3), name=names[0])
+        s2 = Series(range(1, 4), name=names[1])
+
+        mi = MultiIndex.from_arrays([range(3), range(1, 4)], names=names)
+        expected = Series(1, index=mi).unstack(1, fill_value=0)
+
+        result = crosstab(s1, s2)
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_both_tuple_names(self):
+        # GH 18321
+        s1 = Series(range(3), name=("a", "b"))
+        s2 = Series(range(3), name=("c", "d"))
+
+        expected = DataFrame(
+            np.eye(3, dtype="int64"),
+            index=Index(range(3), name=("a", "b")),
+            columns=Index(range(3), name=("c", "d")),
+        )
+        result = crosstab(s1, s2)
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_unsorted_order(self):
+        df = DataFrame({"b": [3, 1, 2], "a": [5, 4, 6]}, index=["C", "A", "B"])
+        result = crosstab(df.index, [df.b, df.a])
+        e_idx = Index(["A", "B", "C"], name="row_0")
+        e_columns = MultiIndex.from_tuples([(1, 4), (2, 6), (3, 5)], names=["b", "a"])
+        expected = DataFrame(
+            [[1, 0, 0], [0, 1, 0], [0, 0, 1]], index=e_idx, columns=e_columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_crosstab_normalize_multiple_columns(self):
+        # GH 15150
+        df = DataFrame(
+            {
+                "A": ["one", "one", "two", "three"] * 6,
+                "B": ["A", "B", "C"] * 8,
+                "C": ["foo", "foo", "foo", "bar", "bar", "bar"] * 4,
+                "D": [0] * 24,
+                "E": [0] * 24,
+            }
+        )
+
+        result = crosstab(
+            [df.A, df.B],
+            df.C,
+            values=df.D,
+            aggfunc=np.sum,
+            normalize=True,
+            margins=True,
+        )
+        expected = DataFrame(
+            np.array([0] * 29 + [1], dtype=float).reshape(10, 3),
+            columns=Index(["bar", "foo", "All"], name="C"),
+            index=MultiIndex.from_tuples(
+                [
+                    ("one", "A"),
+                    ("one", "B"),
+                    ("one", "C"),
+                    ("three", "A"),
+                    ("three", "B"),
+                    ("three", "C"),
+                    ("two", "A"),
+                    ("two", "B"),
+                    ("two", "C"),
+                    ("All", ""),
+                ],
+                names=["A", "B"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_margin_normalize(self):
+        # GH 27500
+        df = DataFrame(
+            {
+                "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+                "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+                "C": [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+                "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+                "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+            }
+        )
+        # normalize on index
+        result = crosstab(
+            [df.A, df.B], df.C, margins=True, margins_name="Sub-Total", normalize=0
+        )
+        expected = DataFrame(
+            [[0.5, 0.5], [0.5, 0.5], [0.666667, 0.333333], [0, 1], [0.444444, 0.555556]]
+        )
+        expected.index = MultiIndex(
+            levels=[["Sub-Total", "bar", "foo"], ["", "one", "two"]],
+            codes=[[1, 1, 2, 2, 0], [1, 2, 1, 2, 0]],
+            names=["A", "B"],
+        )
+        expected.columns = Index(["large", "small"], name="C")
+        tm.assert_frame_equal(result, expected)
+
+        # normalize on columns
+        result = crosstab(
+            [df.A, df.B], df.C, margins=True, margins_name="Sub-Total", normalize=1
+        )
+        expected = DataFrame(
+            [
+                [0.25, 0.2, 0.222222],
+                [0.25, 0.2, 0.222222],
+                [0.5, 0.2, 0.333333],
+                [0, 0.4, 0.222222],
+            ]
+        )
+        expected.columns = Index(["large", "small", "Sub-Total"], name="C")
+        expected.index = MultiIndex(
+            levels=[["bar", "foo"], ["one", "two"]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+            names=["A", "B"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        # normalize on both index and column
+        result = crosstab(
+            [df.A, df.B], df.C, margins=True, margins_name="Sub-Total", normalize=True
+        )
+        expected = DataFrame(
+            [
+                [0.111111, 0.111111, 0.222222],
+                [0.111111, 0.111111, 0.222222],
+                [0.222222, 0.111111, 0.333333],
+                [0.000000, 0.222222, 0.222222],
+                [0.444444, 0.555555, 1],
+            ]
+        )
+        expected.columns = Index(["large", "small", "Sub-Total"], name="C")
+        expected.index = MultiIndex(
+            levels=[["Sub-Total", "bar", "foo"], ["", "one", "two"]],
+            codes=[[1, 1, 2, 2, 0], [1, 2, 1, 2, 0]],
+            names=["A", "B"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_margin_normalize_multiple_columns(self):
+        # GH 35144
+        # use multiple columns with margins and normalization
+        df = DataFrame(
+            {
+                "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+                "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+                "C": [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+                "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+                "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+            }
+        )
+        result = crosstab(
+            index=df.C,
+            columns=[df.A, df.B],
+            margins=True,
+            margins_name="margin",
+            normalize=True,
+        )
+        expected = DataFrame(
+            [
+                [0.111111, 0.111111, 0.222222, 0.000000, 0.444444],
+                [0.111111, 0.111111, 0.111111, 0.222222, 0.555556],
+                [0.222222, 0.222222, 0.333333, 0.222222, 1.0],
+            ],
+            index=["large", "small", "margin"],
+        )
+        expected.columns = MultiIndex(
+            levels=[["bar", "foo", "margin"], ["", "one", "two"]],
+            codes=[[0, 0, 1, 1, 2], [1, 2, 1, 2, 0]],
+            names=["A", "B"],
+        )
+        expected.index.name = "C"
+        tm.assert_frame_equal(result, expected)
+
+    def test_margin_support_Float(self):
+        # GH 50313
+        # use Float64 formats and function aggfunc with margins
+        df = DataFrame(
+            {"A": [1, 2, 2, 1], "B": [3, 3, 4, 5], "C": [-1.0, 10.0, 1.0, 10.0]},
+            dtype="Float64",
+        )
+        result = crosstab(
+            df["A"],
+            df["B"],
+            values=df["C"],
+            aggfunc="sum",
+            margins=True,
+        )
+        expected = DataFrame(
+            [
+                [-1.0, pd.NA, 10.0, 9.0],
+                [10.0, 1.0, pd.NA, 11.0],
+                [9.0, 1.0, 10.0, 20.0],
+            ],
+            index=Index([1.0, 2.0, "All"], dtype="object", name="A"),
+            columns=Index([3.0, 4.0, 5.0, "All"], dtype="object", name="B"),
+            dtype="Float64",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_margin_with_ordered_categorical_column(self):
+        # GH 25278
+        df = DataFrame(
+            {
+                "First": ["B", "B", "C", "A", "B", "C"],
+                "Second": ["C", "B", "B", "B", "C", "A"],
+            }
+        )
+        df["First"] = df["First"].astype(CategoricalDtype(ordered=True))
+        customized_categories_order = ["C", "A", "B"]
+        df["First"] = df["First"].cat.reorder_categories(customized_categories_order)
+        result = crosstab(df["First"], df["Second"], margins=True)
+
+        expected_index = Index(["C", "A", "B", "All"], name="First")
+        expected_columns = Index(["A", "B", "C", "All"], name="Second")
+        expected_data = [[1, 1, 0, 2], [0, 1, 0, 1], [0, 1, 2, 3], [1, 3, 2, 6]]
+        expected = DataFrame(
+            expected_data, index=expected_index, columns=expected_columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("a_dtype", ["category", "int64"])
+@pytest.mark.parametrize("b_dtype", ["category", "int64"])
+def test_categoricals(a_dtype, b_dtype):
+    # https://github.com/pandas-dev/pandas/issues/37465
+    g = np.random.default_rng(2)
+    a = Series(g.integers(0, 3, size=100)).astype(a_dtype)
+    b = Series(g.integers(0, 2, size=100)).astype(b_dtype)
+    result = crosstab(a, b, margins=True, dropna=False)
+    columns = Index([0, 1, "All"], dtype="object", name="col_0")
+    index = Index([0, 1, 2, "All"], dtype="object", name="row_0")
+    values = [[10, 18, 28], [23, 16, 39], [17, 16, 33], [50, 50, 100]]
+    expected = DataFrame(values, index, columns)
+    tm.assert_frame_equal(result, expected)
+
+    # Verify when categorical does not have all values present
+    a.loc[a == 1] = 2
+    a_is_cat = isinstance(a.dtype, CategoricalDtype)
+    assert not a_is_cat or a.value_counts().loc[1] == 0
+    result = crosstab(a, b, margins=True, dropna=False)
+    values = [[10, 18, 28], [0, 0, 0], [40, 32, 72], [50, 50, 100]]
+    expected = DataFrame(values, index, columns)
+    if not a_is_cat:
+        expected = expected.loc[[0, 2, "All"]]
+        expected["All"] = expected["All"].astype("int64")
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_cut.py b/pandas/pandas/tests/reshape/test_cut.py
new file mode 100644
index 0000000000000..63332fe4658e5
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_cut.py
@@ -0,0 +1,812 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    Series,
+    TimedeltaIndex,
+    Timestamp,
+    cut,
+    date_range,
+    interval_range,
+    isna,
+    qcut,
+    timedelta_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+import pandas.core.reshape.tile as tmod
+
+
+def test_simple():
+    data = np.ones(5, dtype="int64")
+    result = cut(data, 4, labels=False)
+
+    expected = np.array([1, 1, 1, 1, 1])
+    tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize("func", [list, np.array])
+def test_bins(func):
+    data = func([0.2, 1.4, 2.5, 6.2, 9.7, 2.1])
+    result, bins = cut(data, 3, retbins=True)
+
+    intervals = IntervalIndex.from_breaks(bins.round(3))
+    intervals = intervals.take([0, 0, 0, 1, 2, 0])
+    expected = Categorical(intervals, ordered=True)
+
+    tm.assert_categorical_equal(result, expected)
+    tm.assert_almost_equal(bins, np.array([0.1905, 3.36666667, 6.53333333, 9.7]))
+
+
+def test_right():
+    data = np.array([0.2, 1.4, 2.5, 6.2, 9.7, 2.1, 2.575])
+    result, bins = cut(data, 4, right=True, retbins=True)
+
+    intervals = IntervalIndex.from_breaks(bins.round(3))
+    expected = Categorical(intervals, ordered=True)
+    expected = expected.take([0, 0, 0, 2, 3, 0, 0])
+
+    tm.assert_categorical_equal(result, expected)
+    tm.assert_almost_equal(bins, np.array([0.1905, 2.575, 4.95, 7.325, 9.7]))
+
+
+def test_no_right():
+    data = np.array([0.2, 1.4, 2.5, 6.2, 9.7, 2.1, 2.575])
+    result, bins = cut(data, 4, right=False, retbins=True)
+
+    intervals = IntervalIndex.from_breaks(bins.round(3), closed="left")
+    intervals = intervals.take([0, 0, 0, 2, 3, 0, 1])
+    expected = Categorical(intervals, ordered=True)
+
+    tm.assert_categorical_equal(result, expected)
+    tm.assert_almost_equal(bins, np.array([0.2, 2.575, 4.95, 7.325, 9.7095]))
+
+
+def test_bins_from_interval_index():
+    c = cut(range(5), 3)
+    expected = c
+    result = cut(range(5), bins=expected.categories)
+    tm.assert_categorical_equal(result, expected)
+
+    expected = Categorical.from_codes(
+        np.append(c.codes, -1), categories=c.categories, ordered=True
+    )
+    result = cut(range(6), bins=expected.categories)
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_bins_from_interval_index_doc_example():
+    # Make sure we preserve the bins.
+    ages = np.array([10, 15, 13, 12, 23, 25, 28, 59, 60])
+    c = cut(ages, bins=[0, 18, 35, 70])
+    expected = IntervalIndex.from_tuples([(0, 18), (18, 35), (35, 70)])
+    tm.assert_index_equal(c.categories, expected)
+
+    result = cut([25, 20, 50], bins=c.categories)
+    tm.assert_index_equal(result.categories, expected)
+    tm.assert_numpy_array_equal(result.codes, np.array([1, 1, 2], dtype="int8"))
+
+
+def test_bins_not_overlapping_from_interval_index():
+    # see gh-23980
+    msg = "Overlapping IntervalIndex is not accepted"
+    ii = IntervalIndex.from_tuples([(0, 10), (2, 12), (4, 14)])
+
+    with pytest.raises(ValueError, match=msg):
+        cut([5, 6], bins=ii)
+
+
+def test_bins_not_monotonic():
+    msg = "bins must increase monotonically"
+    data = [0.2, 1.4, 2.5, 6.2, 9.7, 2.1]
+
+    with pytest.raises(ValueError, match=msg):
+        cut(data, [0.1, 1.5, 1, 10])
+
+
+@pytest.mark.parametrize(
+    "x, bins, expected",
+    [
+        (
+            date_range("2017-12-31", periods=3),
+            [Timestamp.min, Timestamp("2018-01-01"), Timestamp.max],
+            IntervalIndex.from_tuples(
+                [
+                    (Timestamp.min, Timestamp("2018-01-01")),
+                    (Timestamp("2018-01-01"), Timestamp.max),
+                ]
+            ),
+        ),
+        (
+            [-1, 0, 1],
+            np.array(
+                [np.iinfo(np.int64).min, 0, np.iinfo(np.int64).max], dtype="int64"
+            ),
+            IntervalIndex.from_tuples(
+                [(np.iinfo(np.int64).min, 0), (0, np.iinfo(np.int64).max)]
+            ),
+        ),
+        (
+            [
+                np.timedelta64(-1, "ns"),
+                np.timedelta64(0, "ns"),
+                np.timedelta64(1, "ns"),
+            ],
+            np.array(
+                [
+                    np.timedelta64(-np.iinfo(np.int64).max, "ns"),
+                    np.timedelta64(0, "ns"),
+                    np.timedelta64(np.iinfo(np.int64).max, "ns"),
+                ]
+            ),
+            IntervalIndex.from_tuples(
+                [
+                    (
+                        np.timedelta64(-np.iinfo(np.int64).max, "ns"),
+                        np.timedelta64(0, "ns"),
+                    ),
+                    (
+                        np.timedelta64(0, "ns"),
+                        np.timedelta64(np.iinfo(np.int64).max, "ns"),
+                    ),
+                ]
+            ),
+        ),
+    ],
+)
+def test_bins_monotonic_not_overflowing(x, bins, expected):
+    # GH 26045
+    result = cut(x, bins)
+    tm.assert_index_equal(result.categories, expected)
+
+
+def test_wrong_num_labels():
+    msg = "Bin labels must be one fewer than the number of bin edges"
+    data = [0.2, 1.4, 2.5, 6.2, 9.7, 2.1]
+
+    with pytest.raises(ValueError, match=msg):
+        cut(data, [0, 1, 10], labels=["foo", "bar", "baz"])
+
+
+@pytest.mark.parametrize(
+    "x,bins,msg",
+    [
+        ([], 2, "Cannot cut empty array"),
+        ([1, 2, 3], 0.5, "`bins` should be a positive integer"),
+    ],
+)
+def test_cut_corner(x, bins, msg):
+    with pytest.raises(ValueError, match=msg):
+        cut(x, bins)
+
+
+@pytest.mark.parametrize("arg", [2, np.eye(2), DataFrame(np.eye(2))])
+@pytest.mark.parametrize("cut_func", [cut, qcut])
+def test_cut_not_1d_arg(arg, cut_func):
+    msg = "Input array must be 1 dimensional"
+    with pytest.raises(ValueError, match=msg):
+        cut_func(arg, 2)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        [0, 1, 2, 3, 4, np.inf],
+        [-np.inf, 0, 1, 2, 3, 4],
+        [-np.inf, 0, 1, 2, 3, 4, np.inf],
+    ],
+)
+def test_int_bins_with_inf(data):
+    # GH 24314
+    msg = "cannot specify integer `bins` when input data contains infinity"
+    with pytest.raises(ValueError, match=msg):
+        cut(data, bins=3)
+
+
+def test_cut_out_of_range_more():
+    # see gh-1511
+    name = "x"
+
+    ser = Series([0, -1, 0, 1, -3], name=name)
+    ind = cut(ser, [0, 1], labels=False)
+
+    exp = Series([np.nan, np.nan, np.nan, 0, np.nan], name=name)
+    tm.assert_series_equal(ind, exp)
+
+
+@pytest.mark.parametrize(
+    "right,breaks,closed",
+    [
+        (True, [-1e-3, 0.25, 0.5, 0.75, 1], "right"),
+        (False, [0, 0.25, 0.5, 0.75, 1 + 1e-3], "left"),
+    ],
+)
+def test_labels(right, breaks, closed):
+    arr = np.tile(np.arange(0, 1.01, 0.1), 4)
+
+    result, bins = cut(arr, 4, retbins=True, right=right)
+    ex_levels = IntervalIndex.from_breaks(breaks, closed=closed)
+    tm.assert_index_equal(result.categories, ex_levels)
+
+
+def test_cut_pass_series_name_to_factor():
+    name = "foo"
+    ser = Series(np.random.default_rng(2).standard_normal(100), name=name)
+
+    factor = cut(ser, 4)
+    assert factor.name == name
+
+
+def test_label_precision():
+    arr = np.arange(0, 0.73, 0.01)
+    result = cut(arr, 4, precision=2)
+
+    ex_levels = IntervalIndex.from_breaks([-0.00072, 0.18, 0.36, 0.54, 0.72])
+    tm.assert_index_equal(result.categories, ex_levels)
+
+
+@pytest.mark.parametrize("labels", [None, False])
+def test_na_handling(labels):
+    arr = np.arange(0, 0.75, 0.01)
+    arr[::3] = np.nan
+
+    result = cut(arr, 4, labels=labels)
+    result = np.asarray(result)
+
+    expected = np.where(isna(arr), np.nan, result)
+    tm.assert_almost_equal(result, expected)
+
+
+def test_inf_handling():
+    data = np.arange(6)
+    data_ser = Series(data, dtype="int64")
+
+    bins = [-np.inf, 2, 4, np.inf]
+    result = cut(data, bins)
+    result_ser = cut(data_ser, bins)
+
+    ex_uniques = IntervalIndex.from_breaks(bins)
+    tm.assert_index_equal(result.categories, ex_uniques)
+
+    assert result[5] == Interval(4, np.inf)
+    assert result[0] == Interval(-np.inf, 2)
+    assert result_ser[5] == Interval(4, np.inf)
+    assert result_ser[0] == Interval(-np.inf, 2)
+
+
+def test_cut_out_of_bounds():
+    arr = np.random.default_rng(2).standard_normal(100)
+    result = cut(arr, [-1, 0, 1])
+
+    mask = isna(result)
+    ex_mask = (arr < -1) | (arr > 1)
+    tm.assert_numpy_array_equal(mask, ex_mask)
+
+
+@pytest.mark.parametrize(
+    "get_labels,get_expected",
+    [
+        (
+            lambda labels: labels,
+            lambda labels: Categorical(
+                ["Medium"] + 4 * ["Small"] + ["Medium", "Large"],
+                categories=labels,
+                ordered=True,
+            ),
+        ),
+        (
+            lambda labels: Categorical.from_codes([0, 1, 2], labels),
+            lambda labels: Categorical.from_codes([1] + 4 * [0] + [1, 2], labels),
+        ),
+    ],
+)
+def test_cut_pass_labels(get_labels, get_expected):
+    bins = [0, 25, 50, 100]
+    arr = [50, 5, 10, 15, 20, 30, 70]
+    labels = ["Small", "Medium", "Large"]
+
+    result = cut(arr, bins, labels=get_labels(labels))
+    tm.assert_categorical_equal(result, get_expected(labels))
+
+
+def test_cut_pass_labels_compat():
+    # see gh-16459
+    arr = [50, 5, 10, 15, 20, 30, 70]
+    labels = ["Good", "Medium", "Bad"]
+
+    result = cut(arr, 3, labels=labels)
+    exp = cut(arr, 3, labels=Categorical(labels, categories=labels, ordered=True))
+    tm.assert_categorical_equal(result, exp)
+
+
+@pytest.mark.parametrize("x", [np.arange(11.0), np.arange(11.0) / 1e10])
+def test_round_frac_just_works(x):
+    # It works.
+    cut(x, 2)
+
+
+@pytest.mark.parametrize(
+    "val,precision,expected",
+    [
+        (-117.9998, 3, -118),
+        (117.9998, 3, 118),
+        (117.9998, 2, 118),
+        (0.000123456, 2, 0.00012),
+    ],
+)
+def test_round_frac(val, precision, expected):
+    # see gh-1979
+    result = tmod._round_frac(val, precision=precision)
+    assert result == expected
+
+
+def test_cut_return_intervals():
+    ser = Series([0, 1, 2, 3, 4, 5, 6, 7, 8])
+    result = cut(ser, 3)
+
+    exp_bins = np.linspace(0, 8, num=4).round(3)
+    exp_bins[0] -= 0.008
+
+    expected = Series(
+        IntervalIndex.from_breaks(exp_bins, closed="right").take(
+            [0, 0, 0, 1, 1, 1, 2, 2, 2]
+        )
+    ).astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_ret_bins():
+    # see gh-8589
+    ser = Series(np.arange(4))
+    result, bins = cut(ser, 2, retbins=True)
+
+    expected = Series(
+        IntervalIndex.from_breaks([-0.003, 1.5, 3], closed="right").repeat(2)
+    ).astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,msg",
+    [
+        ({"duplicates": "drop"}, None),
+        ({}, "Bin edges must be unique"),
+        ({"duplicates": "raise"}, "Bin edges must be unique"),
+        ({"duplicates": "foo"}, "invalid value for 'duplicates' parameter"),
+    ],
+)
+def test_cut_duplicates_bin(kwargs, msg):
+    # see gh-20947
+    bins = [0, 2, 4, 6, 10, 10]
+    values = Series(np.array([1, 3, 5, 7, 9]), index=["a", "b", "c", "d", "e"])
+
+    if msg is not None:
+        with pytest.raises(ValueError, match=msg):
+            cut(values, bins, **kwargs)
+    else:
+        result = cut(values, bins, **kwargs)
+        expected = cut(values, pd.unique(np.asarray(bins)))
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("data", [9.0, -9.0, 0.0])
+@pytest.mark.parametrize("length", [1, 2])
+def test_single_bin(data, length):
+    # see gh-14652, gh-15428
+    ser = Series([data] * length)
+    result = cut(ser, 1, labels=False)
+
+    expected = Series([0] * length, dtype=np.intp)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "array_1_writeable,array_2_writeable", [(True, True), (True, False), (False, False)]
+)
+def test_cut_read_only(array_1_writeable, array_2_writeable):
+    # issue 18773
+    array_1 = np.arange(0, 100, 10)
+    array_1.flags.writeable = array_1_writeable
+
+    array_2 = np.arange(0, 100, 10)
+    array_2.flags.writeable = array_2_writeable
+
+    hundred_elements = np.arange(100)
+    tm.assert_categorical_equal(
+        cut(hundred_elements, array_1), cut(hundred_elements, array_2)
+    )
+
+
+@pytest.mark.parametrize(
+    "conv",
+    [
+        lambda v: Timestamp(v),
+        lambda v: to_datetime(v),
+        lambda v: np.datetime64(v),
+        lambda v: Timestamp(v).to_pydatetime(),
+    ],
+)
+def test_datetime_bin(conv):
+    data = [np.datetime64("2012-12-13"), np.datetime64("2012-12-15")]
+    bin_data = ["2012-12-12", "2012-12-14", "2012-12-16"]
+
+    expected = Series(
+        IntervalIndex(
+            [
+                Interval(Timestamp(bin_data[0]), Timestamp(bin_data[1])),
+                Interval(Timestamp(bin_data[1]), Timestamp(bin_data[2])),
+            ]
+        )
+    )
+
+    bins = [conv(v) for v in bin_data]
+    result = Series(cut(data, bins=bins))
+
+    if type(bins[0]) is datetime:
+        # The bins have microsecond dtype -> so does result
+        expected = expected.astype("interval[datetime64[us]]")
+
+    expected = expected.astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("box", [Series, Index, np.array, list])
+def test_datetime_cut(unit, box):
+    # see gh-14714
+    #
+    # Testing time data when it comes in various collection types.
+    data = to_datetime(["2013-01-01", "2013-01-02", "2013-01-03"]).astype(f"M8[{unit}]")
+    data = box(data)
+    result, _ = cut(data, 3, retbins=True)
+
+    if unit == "s":
+        # See https://github.com/pandas-dev/pandas/pull/56101#discussion_r1405325425
+        # for why we round to 8 seconds instead of 7
+        left = DatetimeIndex(
+            ["2012-12-31 23:57:08", "2013-01-01 16:00:00", "2013-01-02 08:00:00"],
+            dtype=f"M8[{unit}]",
+        )
+    else:
+        left = DatetimeIndex(
+            [
+                "2012-12-31 23:57:07.200000",
+                "2013-01-01 16:00:00",
+                "2013-01-02 08:00:00",
+            ],
+            dtype=f"M8[{unit}]",
+        )
+    right = DatetimeIndex(
+        ["2013-01-01 16:00:00", "2013-01-02 08:00:00", "2013-01-03 00:00:00"],
+        dtype=f"M8[{unit}]",
+    )
+
+    exp_intervals = IntervalIndex.from_arrays(left, right)
+    expected = Series(exp_intervals).astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(Series(result), expected)
+
+
+@pytest.mark.parametrize("box", [list, np.array, Index, Series])
+def test_datetime_tz_cut_mismatched_tzawareness(box):
+    # GH#54964
+    bins = box(
+        [
+            Timestamp("2013-01-01 04:57:07.200000"),
+            Timestamp("2013-01-01 21:00:00"),
+            Timestamp("2013-01-02 13:00:00"),
+            Timestamp("2013-01-03 05:00:00"),
+        ]
+    )
+    ser = Series(date_range("20130101", periods=3, tz="US/Eastern"))
+
+    msg = "Cannot use timezone-naive bins with timezone-aware values"
+    with pytest.raises(ValueError, match=msg):
+        cut(ser, bins)
+
+
+@pytest.mark.parametrize(
+    "bins",
+    [
+        3,
+        [
+            Timestamp("2013-01-01 04:57:07.200000", tz="UTC").tz_convert("US/Eastern"),
+            Timestamp("2013-01-01 21:00:00", tz="UTC").tz_convert("US/Eastern"),
+            Timestamp("2013-01-02 13:00:00", tz="UTC").tz_convert("US/Eastern"),
+            Timestamp("2013-01-03 05:00:00", tz="UTC").tz_convert("US/Eastern"),
+        ],
+    ],
+)
+@pytest.mark.parametrize("box", [list, np.array, Index, Series])
+def test_datetime_tz_cut(bins, box):
+    # see gh-19872
+    tz = "US/Eastern"
+    ser = Series(date_range("20130101", periods=3, tz=tz))
+
+    if not isinstance(bins, int):
+        bins = box(bins)
+
+    result = cut(ser, bins)
+    ii = IntervalIndex(
+        [
+            Interval(
+                Timestamp("2012-12-31 23:57:07.200000", tz=tz),
+                Timestamp("2013-01-01 16:00:00", tz=tz),
+            ),
+            Interval(
+                Timestamp("2013-01-01 16:00:00", tz=tz),
+                Timestamp("2013-01-02 08:00:00", tz=tz),
+            ),
+            Interval(
+                Timestamp("2013-01-02 08:00:00", tz=tz),
+                Timestamp("2013-01-03 00:00:00", tz=tz),
+            ),
+        ]
+    )
+    if isinstance(bins, int):
+        # the dtype is inferred from ser, which has nanosecond unit
+        ii = ii.astype("interval[datetime64[ns, US/Eastern]]")
+    expected = Series(ii).astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(result, expected)
+
+
+def test_datetime_nan_error():
+    msg = "bins must be of datetime64 dtype"
+
+    with pytest.raises(ValueError, match=msg):
+        cut(date_range("20130101", periods=3), bins=[0, 2, 4])
+
+
+def test_datetime_nan_mask():
+    result = cut(
+        date_range("20130102", periods=5), bins=date_range("20130101", periods=2)
+    )
+
+    mask = result.categories.isna()
+    tm.assert_numpy_array_equal(mask, np.array([False]))
+
+    mask = result.isna()
+    tm.assert_numpy_array_equal(mask, np.array([False, True, True, True, True]))
+
+
+@pytest.mark.parametrize("tz", [None, "UTC", "US/Pacific"])
+def test_datetime_cut_roundtrip(tz, unit):
+    # see gh-19891
+    ser = Series(date_range("20180101", periods=3, tz=tz, unit=unit))
+    result, result_bins = cut(ser, 2, retbins=True)
+
+    expected = cut(ser, result_bins)
+    tm.assert_series_equal(result, expected)
+
+    if unit == "s":
+        # TODO: constructing DatetimeIndex with dtype="M8[s]" without truncating
+        #  the first entry here raises in array_to_datetime. Should truncate
+        #  instead of raising?
+        # See https://github.com/pandas-dev/pandas/pull/56101#discussion_r1405325425
+        # for why we round to 8 seconds instead of 7
+        expected_bins = DatetimeIndex(
+            ["2017-12-31 23:57:08", "2018-01-02 00:00:00", "2018-01-03 00:00:00"],
+            dtype=f"M8[{unit}]",
+        )
+    else:
+        expected_bins = DatetimeIndex(
+            [
+                "2017-12-31 23:57:07.200000",
+                "2018-01-02 00:00:00",
+                "2018-01-03 00:00:00",
+            ],
+            dtype=f"M8[{unit}]",
+        )
+    expected_bins = expected_bins.tz_localize(tz)
+    tm.assert_index_equal(result_bins, expected_bins)
+
+
+def test_timedelta_cut_roundtrip():
+    # see gh-19891
+    ser = Series(timedelta_range("1day", periods=3))
+    result, result_bins = cut(ser, 2, retbins=True)
+
+    expected = cut(ser, result_bins)
+    tm.assert_series_equal(result, expected)
+
+    expected_bins = TimedeltaIndex(
+        ["0 days 23:57:07.200000", "2 days 00:00:00", "3 days 00:00:00"]
+    )
+    tm.assert_index_equal(result_bins, expected_bins)
+
+
+@pytest.mark.parametrize("bins", [6, 7])
+@pytest.mark.parametrize(
+    "box, compare",
+    [
+        (Series, tm.assert_series_equal),
+        (np.array, tm.assert_categorical_equal),
+        (list, tm.assert_equal),
+    ],
+)
+def test_cut_bool_coercion_to_int(bins, box, compare):
+    # issue 20303
+    data_expected = box([0, 1, 1, 0, 1] * 10)
+    data_result = box([False, True, True, False, True] * 10)
+    expected = cut(data_expected, bins, duplicates="drop")
+    result = cut(data_result, bins, duplicates="drop")
+    compare(result, expected)
+
+
+@pytest.mark.parametrize("labels", ["foo", 1, True])
+def test_cut_incorrect_labels(labels):
+    # GH 13318
+    values = range(5)
+    msg = "Bin labels must either be False, None or passed in as a list-like argument"
+    with pytest.raises(ValueError, match=msg):
+        cut(values, 4, labels=labels)
+
+
+@pytest.mark.parametrize("bins", [3, [0, 5, 15]])
+@pytest.mark.parametrize("right", [True, False])
+@pytest.mark.parametrize("include_lowest", [True, False])
+def test_cut_nullable_integer(bins, right, include_lowest):
+    a = np.random.default_rng(2).integers(0, 10, size=50).astype(float)
+    a[::2] = np.nan
+    result = cut(
+        pd.array(a, dtype="Int64"), bins, right=right, include_lowest=include_lowest
+    )
+    expected = cut(a, bins, right=right, include_lowest=include_lowest)
+    tm.assert_categorical_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, bins, labels, expected_codes, expected_labels",
+    [
+        ([15, 17, 19], [14, 16, 18, 20], ["A", "B", "A"], [0, 1, 0], ["A", "B"]),
+        ([1, 3, 5], [0, 2, 4, 6, 8], [2, 0, 1, 2], [2, 0, 1], [0, 1, 2]),
+    ],
+)
+def test_cut_non_unique_labels(data, bins, labels, expected_codes, expected_labels):
+    # GH 33141
+    result = cut(data, bins=bins, labels=labels, ordered=False)
+    expected = Categorical.from_codes(
+        expected_codes, categories=expected_labels, ordered=False
+    )
+    tm.assert_categorical_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, bins, labels, expected_codes, expected_labels",
+    [
+        ([15, 17, 19], [14, 16, 18, 20], ["C", "B", "A"], [0, 1, 2], ["C", "B", "A"]),
+        ([1, 3, 5], [0, 2, 4, 6, 8], [3, 0, 1, 2], [0, 1, 2], [3, 0, 1, 2]),
+    ],
+)
+def test_cut_unordered_labels(data, bins, labels, expected_codes, expected_labels):
+    # GH 33141
+    result = cut(data, bins=bins, labels=labels, ordered=False)
+    expected = Categorical.from_codes(
+        expected_codes, categories=expected_labels, ordered=False
+    )
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_cut_unordered_with_missing_labels_raises_error():
+    # GH 33141
+    msg = "'labels' must be provided if 'ordered = False'"
+    with pytest.raises(ValueError, match=msg):
+        cut([0.5, 3], bins=[0, 1, 2], ordered=False)
+
+
+def test_cut_unordered_with_series_labels():
+    # https://github.com/pandas-dev/pandas/issues/36603
+    ser = Series([1, 2, 3, 4, 5])
+    bins = Series([0, 2, 4, 6])
+    labels = Series(["a", "b", "c"])
+    result = cut(ser, bins=bins, labels=labels, ordered=False)
+    expected = Series(["a", "a", "b", "b", "c"], dtype="category")
+    tm.assert_series_equal(result, expected)
+
+
+def test_cut_no_warnings():
+    df = DataFrame({"value": np.random.default_rng(2).integers(0, 100, 20)})
+    labels = [f"{i} - {i + 9}" for i in range(0, 100, 10)]
+    with tm.assert_produces_warning(False):
+        df["group"] = cut(df.value, range(0, 105, 10), right=False, labels=labels)
+
+
+def test_cut_with_duplicated_index_lowest_included():
+    # GH 42185
+    expected = Series(
+        [Interval(-0.001, 2, closed="right")] * 3
+        + [Interval(2, 4, closed="right"), Interval(-0.001, 2, closed="right")],
+        index=[0, 1, 2, 3, 0],
+        dtype="category",
+    ).cat.as_ordered()
+
+    ser = Series([0, 1, 2, 3, 0], index=[0, 1, 2, 3, 0])
+    result = cut(ser, bins=[0, 2, 4], include_lowest=True)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.filterwarnings("ignore:invalid value encountered in cast:RuntimeWarning")
+def test_cut_with_nonexact_categorical_indices():
+    # GH 42424
+
+    ser = Series(range(100))
+    ser1 = cut(ser, 10).value_counts().head(5)
+    ser2 = cut(ser, 10).value_counts().tail(5)
+    result = DataFrame({"1": ser1, "2": ser2})
+
+    index = pd.CategoricalIndex(
+        [
+            Interval(-0.099, 9.9, closed="right"),
+            Interval(9.9, 19.8, closed="right"),
+            Interval(19.8, 29.7, closed="right"),
+            Interval(29.7, 39.6, closed="right"),
+            Interval(39.6, 49.5, closed="right"),
+            Interval(49.5, 59.4, closed="right"),
+            Interval(59.4, 69.3, closed="right"),
+            Interval(69.3, 79.2, closed="right"),
+            Interval(79.2, 89.1, closed="right"),
+            Interval(89.1, 99, closed="right"),
+        ],
+        ordered=True,
+    )
+
+    expected = DataFrame(
+        {"1": [10] * 5 + [np.nan] * 5, "2": [np.nan] * 5 + [10] * 5}, index=index
+    )
+
+    tm.assert_frame_equal(expected, result)
+
+
+def test_cut_with_timestamp_tuple_labels():
+    # GH 40661
+    labels = [(Timestamp(10),), (Timestamp(20),), (Timestamp(30),)]
+    result = cut([2, 4, 6], bins=[1, 3, 5, 7], labels=labels)
+
+    expected = Categorical.from_codes([0, 1, 2], labels, ordered=True)
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_cut_bins_datetime_intervalindex():
+    # https://github.com/pandas-dev/pandas/issues/46218
+    bins = interval_range(Timestamp("2022-02-25"), Timestamp("2022-02-27"), freq="1D")
+    # passing Series instead of list is important to trigger bug
+    result = cut(Series([Timestamp("2022-02-26")]).astype("M8[ns]"), bins=bins)
+    expected = Categorical.from_codes([0], bins, ordered=True)
+    tm.assert_categorical_equal(result.array, expected)
+
+
+def test_cut_with_nullable_int64():
+    # GH 30787
+    series = Series([0, 1, 2, 3, 4, pd.NA, 6, 7], dtype="Int64")
+    bins = [0, 2, 4, 6, 8]
+    intervals = IntervalIndex.from_breaks(bins)
+
+    expected = Series(
+        Categorical.from_codes([-1, 0, 0, 1, 1, -1, 2, 3], intervals, ordered=True)
+    )
+
+    result = cut(series, bins=bins)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_cut_datetime_array_no_attributeerror():
+    # GH 55431
+    ser = Series(to_datetime(["2023-10-06 12:00:00+0000", "2023-10-07 12:00:00+0000"]))
+
+    result = cut(ser.array, bins=2)
+
+    categories = result.categories
+    expected = Categorical.from_codes([0, 1], categories=categories, ordered=True)
+
+    tm.assert_categorical_equal(
+        result, expected, check_dtype=True, check_category_order=True
+    )
diff --git a/pandas/pandas/tests/reshape/test_from_dummies.py b/pandas/pandas/tests/reshape/test_from_dummies.py
new file mode 100644
index 0000000000000..c7b7992a78232
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_from_dummies.py
@@ -0,0 +1,451 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    from_dummies,
+    get_dummies,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def dummies_basic():
+    return DataFrame(
+        {
+            "col1_a": [1, 0, 1],
+            "col1_b": [0, 1, 0],
+            "col2_a": [0, 1, 0],
+            "col2_b": [1, 0, 0],
+            "col2_c": [0, 0, 1],
+        },
+    )
+
+
+@pytest.fixture
+def dummies_with_unassigned():
+    return DataFrame(
+        {
+            "col1_a": [1, 0, 0],
+            "col1_b": [0, 1, 0],
+            "col2_a": [0, 1, 0],
+            "col2_b": [0, 0, 0],
+            "col2_c": [0, 0, 1],
+        },
+    )
+
+
+def test_error_wrong_data_type():
+    dummies = [0, 1, 0]
+    with pytest.raises(
+        TypeError,
+        match=r"Expected 'data' to be a 'DataFrame'; Received 'data' of type: list",
+    ):
+        from_dummies(dummies)
+
+
+def test_error_no_prefix_contains_unassigned():
+    dummies = DataFrame({"a": [1, 0, 0], "b": [0, 1, 0]})
+    with pytest.raises(
+        ValueError,
+        match=(
+            r"Dummy DataFrame contains unassigned value\(s\); "
+            r"First instance in row: 2"
+        ),
+    ):
+        from_dummies(dummies)
+
+
+def test_error_no_prefix_wrong_default_category_type():
+    dummies = DataFrame({"a": [1, 0, 1], "b": [0, 1, 1]})
+    with pytest.raises(
+        TypeError,
+        match=(
+            r"Expected 'default_category' to be of type 'None', 'Hashable', or 'dict'; "
+            r"Received 'default_category' of type: list"
+        ),
+    ):
+        from_dummies(dummies, default_category=["c", "d"])
+
+
+def test_error_no_prefix_multi_assignment():
+    dummies = DataFrame({"a": [1, 0, 1], "b": [0, 1, 1]})
+    with pytest.raises(
+        ValueError,
+        match=(
+            r"Dummy DataFrame contains multi-assignment\(s\); "
+            r"First instance in row: 2"
+        ),
+    ):
+        from_dummies(dummies)
+
+
+def test_error_no_prefix_contains_nan():
+    dummies = DataFrame({"a": [1, 0, 0], "b": [0, 1, np.nan]})
+    with pytest.raises(
+        ValueError, match=r"Dummy DataFrame contains NA value in column: 'b'"
+    ):
+        from_dummies(dummies)
+
+
+def test_error_contains_non_dummies():
+    dummies = DataFrame(
+        {"a": [1, 6, 3, 1], "b": [0, 1, 0, 2], "c": ["c1", "c2", "c3", "c4"]}
+    )
+    with pytest.raises(
+        TypeError,
+        match=r"Passed DataFrame contains non-dummy data",
+    ):
+        from_dummies(dummies)
+
+
+def test_error_with_prefix_multiple_seperators():
+    dummies = DataFrame(
+        {
+            "col1_a": [1, 0, 1],
+            "col1_b": [0, 1, 0],
+            "col2-a": [0, 1, 0],
+            "col2-b": [1, 0, 1],
+        },
+    )
+    with pytest.raises(
+        ValueError,
+        match=(r"Separator not specified for column: col2-a"),
+    ):
+        from_dummies(dummies, sep="_")
+
+
+def test_error_with_prefix_sep_wrong_type(dummies_basic):
+    with pytest.raises(
+        TypeError,
+        match=(
+            r"Expected 'sep' to be of type 'str' or 'None'; "
+            r"Received 'sep' of type: list"
+        ),
+    ):
+        from_dummies(dummies_basic, sep=["_"])
+
+
+def test_error_with_prefix_contains_unassigned(dummies_with_unassigned):
+    with pytest.raises(
+        ValueError,
+        match=(
+            r"Dummy DataFrame contains unassigned value\(s\); "
+            r"First instance in row: 2"
+        ),
+    ):
+        from_dummies(dummies_with_unassigned, sep="_")
+
+
+def test_error_with_prefix_default_category_wrong_type(dummies_with_unassigned):
+    with pytest.raises(
+        TypeError,
+        match=(
+            r"Expected 'default_category' to be of type 'None', 'Hashable', or 'dict'; "
+            r"Received 'default_category' of type: list"
+        ),
+    ):
+        from_dummies(dummies_with_unassigned, sep="_", default_category=["x", "y"])
+
+
+def test_error_with_prefix_default_category_dict_not_complete(
+    dummies_with_unassigned,
+):
+    with pytest.raises(
+        ValueError,
+        match=(
+            r"Length of 'default_category' \(1\) did not match "
+            r"the length of the columns being encoded \(2\)"
+        ),
+    ):
+        from_dummies(dummies_with_unassigned, sep="_", default_category={"col1": "x"})
+
+
+def test_error_with_prefix_contains_nan(dummies_basic):
+    # Set float64 dtype to avoid upcast when setting np.nan
+    dummies_basic["col2_c"] = dummies_basic["col2_c"].astype("float64")
+    dummies_basic.loc[2, "col2_c"] = np.nan
+    with pytest.raises(
+        ValueError, match=r"Dummy DataFrame contains NA value in column: 'col2_c'"
+    ):
+        from_dummies(dummies_basic, sep="_")
+
+
+def test_error_with_prefix_contains_non_dummies(dummies_basic):
+    # Set object dtype to avoid upcast when setting "str"
+    dummies_basic["col2_c"] = dummies_basic["col2_c"].astype(object)
+    dummies_basic.loc[2, "col2_c"] = "str"
+    with pytest.raises(TypeError, match=r"Passed DataFrame contains non-dummy data"):
+        from_dummies(dummies_basic, sep="_")
+
+
+def test_error_with_prefix_double_assignment():
+    dummies = DataFrame(
+        {
+            "col1_a": [1, 0, 1],
+            "col1_b": [1, 1, 0],
+            "col2_a": [0, 1, 0],
+            "col2_b": [1, 0, 0],
+            "col2_c": [0, 0, 1],
+        },
+    )
+    with pytest.raises(
+        ValueError,
+        match=(
+            r"Dummy DataFrame contains multi-assignment\(s\); "
+            r"First instance in row: 0"
+        ),
+    ):
+        from_dummies(dummies, sep="_")
+
+
+def test_roundtrip_series_to_dataframe():
+    categories = Series(["a", "b", "c", "a"])
+    dummies = get_dummies(categories)
+    result = from_dummies(dummies)
+    expected = DataFrame({"": ["a", "b", "c", "a"]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_roundtrip_single_column_dataframe():
+    categories = DataFrame({"": ["a", "b", "c", "a"]})
+    dummies = get_dummies(categories)
+    result = from_dummies(dummies, sep="_")
+    expected = categories
+    tm.assert_frame_equal(result, expected)
+
+
+def test_roundtrip_with_prefixes():
+    categories = DataFrame({"col1": ["a", "b", "a"], "col2": ["b", "a", "c"]})
+    dummies = get_dummies(categories)
+    result = from_dummies(dummies, sep="_")
+    expected = categories
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_string_cats_basic():
+    dummies = DataFrame({"a": [1, 0, 0, 1], "b": [0, 1, 0, 0], "c": [0, 0, 1, 0]})
+    expected = DataFrame({"": ["a", "b", "c", "a"]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_string_cats_basic_bool_values():
+    dummies = DataFrame(
+        {
+            "a": [True, False, False, True],
+            "b": [False, True, False, False],
+            "c": [False, False, True, False],
+        }
+    )
+    expected = DataFrame({"": ["a", "b", "c", "a"]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_string_cats_basic_mixed_bool_values():
+    dummies = DataFrame(
+        {"a": [1, 0, 0, 1], "b": [False, True, False, False], "c": [0, 0, 1, 0]}
+    )
+    expected = DataFrame({"": ["a", "b", "c", "a"]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_int_cats_basic():
+    dummies = DataFrame(
+        {1: [1, 0, 0, 0], 25: [0, 1, 0, 0], 2: [0, 0, 1, 0], 5: [0, 0, 0, 1]}
+    )
+    expected = DataFrame({"": [1, 25, 2, 5]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_float_cats_basic():
+    dummies = DataFrame(
+        {1.0: [1, 0, 0, 0], 25.0: [0, 1, 0, 0], 2.5: [0, 0, 1, 0], 5.84: [0, 0, 0, 1]}
+    )
+    expected = DataFrame({"": [1.0, 25.0, 2.5, 5.84]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_mixed_cats_basic():
+    dummies = DataFrame(
+        {
+            1.23: [1, 0, 0, 0, 0],
+            "c": [0, 1, 0, 0, 0],
+            2: [0, 0, 1, 0, 0],
+            False: [0, 0, 0, 1, 0],
+            None: [0, 0, 0, 0, 1],
+        }
+    )
+    expected = DataFrame({"": [1.23, "c", 2, False, None]}, dtype="object")
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_no_prefix_string_cats_contains_get_dummies_NaN_column():
+    dummies = DataFrame({"a": [1, 0, 0], "b": [0, 1, 0], "NaN": [0, 0, 1]})
+    expected = DataFrame({"": ["a", "b", "NaN"]})
+    result = from_dummies(dummies)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "default_category, expected",
+    [
+        pytest.param(
+            "c",
+            {"": ["a", "b", "c"]},
+            id="default_category is a str",
+        ),
+        pytest.param(
+            1,
+            {"": ["a", "b", 1]},
+            id="default_category is a int",
+        ),
+        pytest.param(
+            1.25,
+            {"": ["a", "b", 1.25]},
+            id="default_category is a float",
+        ),
+        pytest.param(
+            0,
+            {"": ["a", "b", 0]},
+            id="default_category is a 0",
+        ),
+        pytest.param(
+            False,
+            {"": ["a", "b", False]},
+            id="default_category is a bool",
+        ),
+        pytest.param(
+            (1, 2),
+            {"": ["a", "b", (1, 2)]},
+            id="default_category is a tuple",
+        ),
+    ],
+)
+def test_no_prefix_string_cats_default_category(
+    default_category, expected, using_infer_string
+):
+    dummies = DataFrame({"a": [1, 0, 0], "b": [0, 1, 0]})
+    result = from_dummies(dummies, default_category=default_category)
+    expected = DataFrame(expected)
+    if using_infer_string:
+        expected[""] = expected[""].astype("str")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_with_prefix_basic(dummies_basic):
+    expected = DataFrame({"col1": ["a", "b", "a"], "col2": ["b", "a", "c"]})
+    result = from_dummies(dummies_basic, sep="_")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_with_prefix_contains_get_dummies_NaN_column():
+    dummies = DataFrame(
+        {
+            "col1_a": [1, 0, 0],
+            "col1_b": [0, 1, 0],
+            "col1_NaN": [0, 0, 1],
+            "col2_a": [0, 1, 0],
+            "col2_b": [0, 0, 0],
+            "col2_c": [0, 0, 1],
+            "col2_NaN": [1, 0, 0],
+        },
+    )
+    expected = DataFrame({"col1": ["a", "b", "NaN"], "col2": ["NaN", "a", "c"]})
+    result = from_dummies(dummies, sep="_")
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "default_category, expected",
+    [
+        pytest.param(
+            "x",
+            {"col1": ["a", "b", "x"], "col2": ["x", "a", "c"]},
+            id="default_category is a str",
+        ),
+        pytest.param(
+            0,
+            {"col1": ["a", "b", 0], "col2": [0, "a", "c"]},
+            id="default_category is a 0",
+        ),
+        pytest.param(
+            False,
+            {"col1": ["a", "b", False], "col2": [False, "a", "c"]},
+            id="default_category is a False",
+        ),
+        pytest.param(
+            {"col2": 1, "col1": 2.5},
+            {"col1": ["a", "b", 2.5], "col2": [1, "a", "c"]},
+            id="default_category is a dict with int and float values",
+        ),
+        pytest.param(
+            {"col2": None, "col1": False},
+            {"col1": ["a", "b", False], "col2": [None, "a", "c"]},
+            id="default_category is a dict with bool and None values",
+        ),
+        pytest.param(
+            {"col2": (1, 2), "col1": [1.25, False]},
+            {"col1": ["a", "b", [1.25, False]], "col2": [(1, 2), "a", "c"]},
+            id="default_category is a dict with list and tuple values",
+        ),
+    ],
+)
+def test_with_prefix_default_category(
+    dummies_with_unassigned, default_category, expected, using_infer_string
+):
+    result = from_dummies(
+        dummies_with_unassigned, sep="_", default_category=default_category
+    )
+    expected = DataFrame(expected)
+    if using_infer_string:
+        expected = expected.astype("str")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ea_categories():
+    # GH 54300
+    df = DataFrame({"a": [1, 0, 0, 1], "b": [0, 1, 0, 0], "c": [0, 0, 1, 0]})
+    df.columns = df.columns.astype("string[python]")
+    result = from_dummies(df)
+    expected = DataFrame({"": Series(list("abca"), dtype="string[python]")})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ea_categories_with_sep():
+    # GH 54300
+    df = DataFrame(
+        {
+            "col1_a": [1, 0, 1],
+            "col1_b": [0, 1, 0],
+            "col2_a": [0, 1, 0],
+            "col2_b": [1, 0, 0],
+            "col2_c": [0, 0, 1],
+        }
+    )
+    df.columns = df.columns.astype("string[python]")
+    result = from_dummies(df, sep="_")
+    expected = DataFrame(
+        {
+            "col1": Series(list("aba"), dtype="string[python]"),
+            "col2": Series(list("bac"), dtype="string[python]"),
+        }
+    )
+    expected.columns = expected.columns.astype("string[python]")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_maintain_original_index():
+    # GH 54300
+    df = DataFrame(
+        {"a": [1, 0, 0, 1], "b": [0, 1, 0, 0], "c": [0, 0, 1, 0]}, index=list("abcd")
+    )
+    result = from_dummies(df)
+    expected = DataFrame({"": list("abca")}, index=list("abcd"))
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_get_dummies.py b/pandas/pandas/tests/reshape/test_get_dummies.py
new file mode 100644
index 0000000000000..9ce2c925a368b
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_get_dummies.py
@@ -0,0 +1,741 @@
+import re
+import unicodedata
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import is_integer_dtype
+
+import pandas as pd
+from pandas import (
+    ArrowDtype,
+    Categorical,
+    CategoricalDtype,
+    CategoricalIndex,
+    DataFrame,
+    Index,
+    RangeIndex,
+    Series,
+    SparseDtype,
+    get_dummies,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+try:
+    import pyarrow as pa
+except ImportError:
+    pa = None
+
+
+class TestGetDummies:
+    @pytest.fixture
+    def df(self):
+        return DataFrame({"A": ["a", "b", "a"], "B": ["b", "b", "c"], "C": [1, 2, 3]})
+
+    @pytest.fixture(params=["uint8", "i8", np.float64, bool, None])
+    def dtype(self, request):
+        return np.dtype(request.param)
+
+    @pytest.fixture(params=["dense", "sparse"])
+    def sparse(self, request):
+        # params are strings to simplify reading test results,
+        # e.g. TestGetDummies::test_basic[uint8-sparse] instead of [uint8-True]
+        return request.param == "sparse"
+
+    def effective_dtype(self, dtype):
+        if dtype is None:
+            return np.uint8
+        return dtype
+
+    def test_get_dummies_raises_on_dtype_object(self, df):
+        msg = "dtype=object is not a valid dtype for get_dummies"
+        with pytest.raises(ValueError, match=msg):
+            get_dummies(df, dtype="object")
+
+    def test_get_dummies_basic(self, sparse, dtype):
+        s_list = list("abc")
+        s_series = Series(s_list)
+        s_series_index = Series(s_list, list("ABC"))
+
+        expected = DataFrame(
+            {"a": [1, 0, 0], "b": [0, 1, 0], "c": [0, 0, 1]},
+            dtype=self.effective_dtype(dtype),
+        )
+        if sparse:
+            if dtype.kind == "b":
+                expected = expected.apply(SparseArray, fill_value=False)
+            else:
+                expected = expected.apply(SparseArray, fill_value=0.0)
+        result = get_dummies(s_list, sparse=sparse, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(s_series, sparse=sparse, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        expected.index = list("ABC")
+        result = get_dummies(s_series_index, sparse=sparse, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_basic_types(self, sparse, dtype, using_infer_string):
+        # GH 10531
+        s_list = list("abc")
+        s_series = Series(s_list)
+        s_df = DataFrame(
+            {"a": [0, 1, 0, 1, 2], "b": ["A", "A", "B", "C", "C"], "c": [2, 3, 3, 3, 2]}
+        )
+
+        expected = DataFrame(
+            {"a": [1, 0, 0], "b": [0, 1, 0], "c": [0, 0, 1]},
+            dtype=self.effective_dtype(dtype),
+            columns=list("abc"),
+        )
+        if sparse:
+            if is_integer_dtype(dtype):
+                fill_value = 0
+            elif dtype == bool:
+                fill_value = False
+            else:
+                fill_value = 0.0
+
+            expected = expected.apply(SparseArray, fill_value=fill_value)
+        result = get_dummies(s_list, sparse=sparse, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(s_series, sparse=sparse, dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(s_df, columns=s_df.columns, sparse=sparse, dtype=dtype)
+        if sparse:
+            dtype_name = f"Sparse[{self.effective_dtype(dtype).name}, {fill_value}]"
+        else:
+            dtype_name = self.effective_dtype(dtype).name
+
+        expected = Series({dtype_name: 8}, name="count")
+        result = result.dtypes.value_counts()
+        result.index = [str(i) for i in result.index]
+        tm.assert_series_equal(result, expected)
+
+        result = get_dummies(s_df, columns=["a"], sparse=sparse, dtype=dtype)
+
+        key = "str" if using_infer_string else "object"
+        expected_counts = {"int64": 1, key: 1}
+        expected_counts[dtype_name] = 3 + expected_counts.get(dtype_name, 0)
+
+        expected = Series(expected_counts, name="count").sort_index()
+        result = result.dtypes.value_counts()
+        result.index = [str(i) for i in result.index]
+        result = result.sort_index()
+        tm.assert_series_equal(result, expected)
+
+    def test_get_dummies_just_na(self, sparse):
+        just_na_list = [np.nan]
+        just_na_series = Series(just_na_list)
+        just_na_series_index = Series(just_na_list, index=["A"])
+
+        res_list = get_dummies(just_na_list, sparse=sparse)
+        res_series = get_dummies(just_na_series, sparse=sparse)
+        res_series_index = get_dummies(just_na_series_index, sparse=sparse)
+
+        assert res_list.empty
+        assert res_series.empty
+        assert res_series_index.empty
+
+        assert res_list.index.tolist() == [0]
+        assert res_series.index.tolist() == [0]
+        assert res_series_index.index.tolist() == ["A"]
+
+    def test_get_dummies_include_na(self, sparse, dtype):
+        s = ["a", "b", np.nan]
+        res = get_dummies(s, sparse=sparse, dtype=dtype)
+        exp = DataFrame(
+            {"a": [1, 0, 0], "b": [0, 1, 0]}, dtype=self.effective_dtype(dtype)
+        )
+        if sparse:
+            if dtype.kind == "b":
+                exp = exp.apply(SparseArray, fill_value=False)
+            else:
+                exp = exp.apply(SparseArray, fill_value=0.0)
+        tm.assert_frame_equal(res, exp)
+
+        # Sparse dataframes do not allow nan labelled columns, see #GH8822
+        res_na = get_dummies(s, dummy_na=True, sparse=sparse, dtype=dtype)
+        exp_na = DataFrame(
+            {np.nan: [0, 0, 1], "a": [1, 0, 0], "b": [0, 1, 0]},
+            dtype=self.effective_dtype(dtype),
+        )
+        exp_na = exp_na.reindex(["a", "b", np.nan], axis=1)
+        # hack (NaN handling in assert_index_equal)
+        exp_na.columns = res_na.columns
+        if sparse:
+            if dtype.kind == "b":
+                exp_na = exp_na.apply(SparseArray, fill_value=False)
+            else:
+                exp_na = exp_na.apply(SparseArray, fill_value=0.0)
+        tm.assert_frame_equal(res_na, exp_na)
+
+        res_just_na = get_dummies([np.nan], dummy_na=True, sparse=sparse, dtype=dtype)
+        exp_just_na = DataFrame(
+            Series(1, index=[0]), columns=[np.nan], dtype=self.effective_dtype(dtype)
+        )
+        tm.assert_numpy_array_equal(res_just_na.values, exp_just_na.values)
+
+    def test_get_dummies_unicode(self, sparse):
+        # See GH 6885 - get_dummies chokes on unicode values
+        e = "e"
+        eacute = unicodedata.lookup("LATIN SMALL LETTER E WITH ACUTE")
+        s = [e, eacute, eacute]
+        res = get_dummies(s, prefix="letter", sparse=sparse)
+        exp = DataFrame(
+            {"letter_e": [True, False, False], f"letter_{eacute}": [False, True, True]}
+        )
+        if sparse:
+            exp = exp.apply(SparseArray, fill_value=False)
+        tm.assert_frame_equal(res, exp)
+
+    def test_dataframe_dummies_all_obj(self, df, sparse):
+        df = df[["A", "B"]]
+        result = get_dummies(df, sparse=sparse)
+        expected = DataFrame(
+            {"A_a": [1, 0, 1], "A_b": [0, 1, 0], "B_b": [1, 1, 0], "B_c": [0, 0, 1]},
+            dtype=bool,
+        )
+        if sparse:
+            expected = DataFrame(
+                {
+                    "A_a": SparseArray([1, 0, 1], dtype="bool"),
+                    "A_b": SparseArray([0, 1, 0], dtype="bool"),
+                    "B_b": SparseArray([1, 1, 0], dtype="bool"),
+                    "B_c": SparseArray([0, 0, 1], dtype="bool"),
+                }
+            )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_string_dtype(self, df, any_string_dtype):
+        # GH44965
+        df = df[["A", "B"]]
+        df = df.astype({"A": "str", "B": any_string_dtype})
+        result = get_dummies(df)
+        expected = DataFrame(
+            {
+                "A_a": [1, 0, 1],
+                "A_b": [0, 1, 0],
+                "B_b": [1, 1, 0],
+                "B_c": [0, 0, 1],
+            },
+            dtype=bool,
+        )
+        if any_string_dtype == "string" and any_string_dtype.na_value is pd.NA:
+            expected[["B_b", "B_c"]] = expected[["B_b", "B_c"]].astype("boolean")
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_mix_default(self, df, sparse, dtype):
+        result = get_dummies(df, sparse=sparse, dtype=dtype)
+        if sparse:
+            arr = SparseArray
+            if dtype.kind == "b":
+                typ = SparseDtype(dtype, False)
+            else:
+                typ = SparseDtype(dtype, 0)
+        else:
+            arr = np.array
+            typ = dtype
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3],
+                "A_a": arr([1, 0, 1], dtype=typ),
+                "A_b": arr([0, 1, 0], dtype=typ),
+                "B_b": arr([1, 1, 0], dtype=typ),
+                "B_c": arr([0, 0, 1], dtype=typ),
+            }
+        )
+        expected = expected[["C", "A_a", "A_b", "B_b", "B_c"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_prefix_list(self, df, sparse):
+        prefixes = ["from_A", "from_B"]
+        result = get_dummies(df, prefix=prefixes, sparse=sparse)
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3],
+                "from_A_a": [True, False, True],
+                "from_A_b": [False, True, False],
+                "from_B_b": [True, True, False],
+                "from_B_c": [False, False, True],
+            },
+        )
+        expected[["C"]] = df[["C"]]
+        cols = ["from_A_a", "from_A_b", "from_B_b", "from_B_c"]
+        expected = expected[["C"] + cols]
+
+        typ = SparseArray if sparse else Series
+        expected[cols] = expected[cols].apply(lambda x: typ(x))
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_prefix_str(self, df, sparse):
+        # not that you should do this...
+        result = get_dummies(df, prefix="bad", sparse=sparse)
+        bad_columns = ["bad_a", "bad_b", "bad_b", "bad_c"]
+        expected = DataFrame(
+            [
+                [1, True, False, True, False],
+                [2, False, True, True, False],
+                [3, True, False, False, True],
+            ],
+            columns=["C"] + bad_columns,
+        )
+        expected = expected.astype({"C": np.int64})
+        if sparse:
+            # work around astyping & assigning with duplicate columns
+            # https://github.com/pandas-dev/pandas/issues/14427
+            expected = pd.concat(
+                [
+                    Series([1, 2, 3], name="C"),
+                    Series([True, False, True], name="bad_a", dtype="Sparse[bool]"),
+                    Series([False, True, False], name="bad_b", dtype="Sparse[bool]"),
+                    Series([True, True, False], name="bad_b", dtype="Sparse[bool]"),
+                    Series([False, False, True], name="bad_c", dtype="Sparse[bool]"),
+                ],
+                axis=1,
+            )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_subset(self, df, sparse):
+        result = get_dummies(df, prefix=["from_A"], columns=["A"], sparse=sparse)
+        expected = DataFrame(
+            {
+                "B": ["b", "b", "c"],
+                "C": [1, 2, 3],
+                "from_A_a": [1, 0, 1],
+                "from_A_b": [0, 1, 0],
+            },
+        )
+        cols = expected.columns
+        expected[cols[1:]] = expected[cols[1:]].astype(bool)
+        expected[["C"]] = df[["C"]]
+        if sparse:
+            cols = ["from_A_a", "from_A_b"]
+            expected[cols] = expected[cols].astype(SparseDtype("bool", False))
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_prefix_sep(self, df, sparse):
+        result = get_dummies(df, prefix_sep="..", sparse=sparse)
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3],
+                "A..a": [True, False, True],
+                "A..b": [False, True, False],
+                "B..b": [True, True, False],
+                "B..c": [False, False, True],
+            },
+        )
+        expected[["C"]] = df[["C"]]
+        expected = expected[["C", "A..a", "A..b", "B..b", "B..c"]]
+        if sparse:
+            cols = ["A..a", "A..b", "B..b", "B..c"]
+            expected[cols] = expected[cols].astype(SparseDtype("bool", False))
+
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(df, prefix_sep=["..", "__"], sparse=sparse)
+        expected = expected.rename(columns={"B..b": "B__b", "B..c": "B__c"})
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(df, prefix_sep={"A": "..", "B": "__"}, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_prefix_bad_length(self, df, sparse):
+        msg = re.escape(
+            "Length of 'prefix' (1) did not match the length of the columns being "
+            "encoded (2)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            get_dummies(df, prefix=["too few"], sparse=sparse)
+
+    def test_dataframe_dummies_prefix_sep_bad_length(self, df, sparse):
+        msg = re.escape(
+            "Length of 'prefix_sep' (1) did not match the length of the columns being "
+            "encoded (2)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            get_dummies(df, prefix_sep=["bad"], sparse=sparse)
+
+    def test_dataframe_dummies_prefix_dict(self, sparse):
+        prefixes = {"A": "from_A", "B": "from_B"}
+        df = DataFrame({"C": [1, 2, 3], "A": ["a", "b", "a"], "B": ["b", "b", "c"]})
+        result = get_dummies(df, prefix=prefixes, sparse=sparse)
+
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3],
+                "from_A_a": [1, 0, 1],
+                "from_A_b": [0, 1, 0],
+                "from_B_b": [1, 1, 0],
+                "from_B_c": [0, 0, 1],
+            }
+        )
+
+        columns = ["from_A_a", "from_A_b", "from_B_b", "from_B_c"]
+        expected[columns] = expected[columns].astype(bool)
+        if sparse:
+            expected[columns] = expected[columns].astype(SparseDtype("bool", False))
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_with_na(self, df, sparse, dtype):
+        df.loc[3, :] = [np.nan, np.nan, np.nan]
+        result = get_dummies(df, dummy_na=True, sparse=sparse, dtype=dtype).sort_index(
+            axis=1
+        )
+
+        if sparse:
+            arr = SparseArray
+            if dtype.kind == "b":
+                typ = SparseDtype(dtype, False)
+            else:
+                typ = SparseDtype(dtype, 0)
+        else:
+            arr = np.array
+            typ = dtype
+
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3, np.nan],
+                "A_a": arr([1, 0, 1, 0], dtype=typ),
+                "A_b": arr([0, 1, 0, 0], dtype=typ),
+                "A_nan": arr([0, 0, 0, 1], dtype=typ),
+                "B_b": arr([1, 1, 0, 0], dtype=typ),
+                "B_c": arr([0, 0, 1, 0], dtype=typ),
+                "B_nan": arr([0, 0, 0, 1], dtype=typ),
+            }
+        ).sort_index(axis=1)
+
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(df, dummy_na=False, sparse=sparse, dtype=dtype)
+        expected = expected[["C", "A_a", "A_b", "B_b", "B_c"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_with_categorical(self, df, sparse, dtype):
+        df["cat"] = Categorical(["x", "y", "y"])
+        result = get_dummies(df, sparse=sparse, dtype=dtype).sort_index(axis=1)
+        if sparse:
+            arr = SparseArray
+            if dtype.kind == "b":
+                typ = SparseDtype(dtype, False)
+            else:
+                typ = SparseDtype(dtype, 0)
+        else:
+            arr = np.array
+            typ = dtype
+
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3],
+                "A_a": arr([1, 0, 1], dtype=typ),
+                "A_b": arr([0, 1, 0], dtype=typ),
+                "B_b": arr([1, 1, 0], dtype=typ),
+                "B_c": arr([0, 0, 1], dtype=typ),
+                "cat_x": arr([1, 0, 0], dtype=typ),
+                "cat_y": arr([0, 1, 1], dtype=typ),
+            }
+        ).sort_index(axis=1)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "get_dummies_kwargs,expected",
+        [
+            (
+                {"data": DataFrame({"ä": ["a"]})},
+                "ä_a",
+            ),
+            (
+                {"data": DataFrame({"x": ["ä"]})},
+                "x_ä",
+            ),
+            (
+                {"data": DataFrame({"x": ["a"]}), "prefix": "ä"},
+                "ä_a",
+            ),
+            (
+                {"data": DataFrame({"x": ["a"]}), "prefix_sep": "ä"},
+                "xäa",
+            ),
+        ],
+    )
+    def test_dataframe_dummies_unicode(self, get_dummies_kwargs, expected):
+        # GH22084 get_dummies incorrectly encodes unicode characters
+        # in dataframe column names
+        result = get_dummies(**get_dummies_kwargs)
+        expected = DataFrame({expected: [True]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_basic_drop_first(self, sparse):
+        # GH12402 Add a new parameter `drop_first` to avoid collinearity
+        # Basic case
+        s_list = list("abc")
+        s_series = Series(s_list)
+        s_series_index = Series(s_list, list("ABC"))
+
+        expected = DataFrame({"b": [0, 1, 0], "c": [0, 0, 1]}, dtype=bool)
+
+        result = get_dummies(s_list, drop_first=True, sparse=sparse)
+        if sparse:
+            expected = expected.apply(SparseArray, fill_value=False)
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(s_series, drop_first=True, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+        expected.index = list("ABC")
+        result = get_dummies(s_series_index, drop_first=True, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_basic_drop_first_one_level(self, sparse):
+        # Test the case that categorical variable only has one level.
+        s_list = list("aaa")
+        s_series = Series(s_list)
+        s_series_index = Series(s_list, list("ABC"))
+
+        expected = DataFrame(index=RangeIndex(3))
+
+        result = get_dummies(s_list, drop_first=True, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(s_series, drop_first=True, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame(index=list("ABC"))
+        result = get_dummies(s_series_index, drop_first=True, sparse=sparse)
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_basic_drop_first_NA(self, sparse):
+        # Test NA handling together with drop_first
+        s_NA = ["a", "b", np.nan]
+        res = get_dummies(s_NA, drop_first=True, sparse=sparse)
+        exp = DataFrame({"b": [0, 1, 0]}, dtype=bool)
+        if sparse:
+            exp = exp.apply(SparseArray, fill_value=False)
+
+        tm.assert_frame_equal(res, exp)
+
+        res_na = get_dummies(s_NA, dummy_na=True, drop_first=True, sparse=sparse)
+        exp_na = DataFrame({"b": [0, 1, 0], np.nan: [0, 0, 1]}, dtype=bool).reindex(
+            ["b", np.nan], axis=1
+        )
+        if sparse:
+            exp_na = exp_na.apply(SparseArray, fill_value=False)
+        tm.assert_frame_equal(res_na, exp_na)
+
+        res_just_na = get_dummies(
+            [np.nan], dummy_na=True, drop_first=True, sparse=sparse
+        )
+        exp_just_na = DataFrame(index=RangeIndex(1))
+        tm.assert_frame_equal(res_just_na, exp_just_na)
+
+    def test_dataframe_dummies_drop_first(self, df, sparse):
+        df = df[["A", "B"]]
+        result = get_dummies(df, drop_first=True, sparse=sparse)
+        expected = DataFrame({"A_b": [0, 1, 0], "B_c": [0, 0, 1]}, dtype=bool)
+        if sparse:
+            expected = expected.apply(SparseArray, fill_value=False)
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_drop_first_with_categorical(self, df, sparse, dtype):
+        df["cat"] = Categorical(["x", "y", "y"])
+        result = get_dummies(df, drop_first=True, sparse=sparse)
+        expected = DataFrame(
+            {"C": [1, 2, 3], "A_b": [0, 1, 0], "B_c": [0, 0, 1], "cat_y": [0, 1, 1]}
+        )
+        cols = ["A_b", "B_c", "cat_y"]
+        expected[cols] = expected[cols].astype(bool)
+        expected = expected[["C", "A_b", "B_c", "cat_y"]]
+        if sparse:
+            for col in cols:
+                expected[col] = SparseArray(expected[col])
+        tm.assert_frame_equal(result, expected)
+
+    def test_dataframe_dummies_drop_first_with_na(self, df, sparse):
+        df.loc[3, :] = [np.nan, np.nan, np.nan]
+        result = get_dummies(
+            df, dummy_na=True, drop_first=True, sparse=sparse
+        ).sort_index(axis=1)
+        expected = DataFrame(
+            {
+                "C": [1, 2, 3, np.nan],
+                "A_b": [0, 1, 0, 0],
+                "A_nan": [0, 0, 0, 1],
+                "B_c": [0, 0, 1, 0],
+                "B_nan": [0, 0, 0, 1],
+            }
+        )
+        cols = ["A_b", "A_nan", "B_c", "B_nan"]
+        expected[cols] = expected[cols].astype(bool)
+        expected = expected.sort_index(axis=1)
+        if sparse:
+            for col in cols:
+                expected[col] = SparseArray(expected[col])
+
+        tm.assert_frame_equal(result, expected)
+
+        result = get_dummies(df, dummy_na=False, drop_first=True, sparse=sparse)
+        expected = expected[["C", "A_b", "B_c"]]
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_int_int(self):
+        data = Series([1, 2, 1])
+        result = get_dummies(data)
+        expected = DataFrame([[1, 0], [0, 1], [1, 0]], columns=[1, 2], dtype=bool)
+        tm.assert_frame_equal(result, expected)
+
+        data = Series(Categorical(["a", "b", "a"]))
+        result = get_dummies(data)
+        expected = DataFrame(
+            [[1, 0], [0, 1], [1, 0]], columns=Categorical(["a", "b"]), dtype=bool
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_int_df(self, dtype):
+        data = DataFrame(
+            {
+                "A": [1, 2, 1],
+                "B": Categorical(["a", "b", "a"]),
+                "C": [1, 2, 1],
+                "D": [1.0, 2.0, 1.0],
+            }
+        )
+        columns = ["C", "D", "A_1", "A_2", "B_a", "B_b"]
+        expected = DataFrame(
+            [[1, 1.0, 1, 0, 1, 0], [2, 2.0, 0, 1, 0, 1], [1, 1.0, 1, 0, 1, 0]],
+            columns=columns,
+        )
+        expected[columns[2:]] = expected[columns[2:]].astype(dtype)
+        result = get_dummies(data, columns=["A", "B"], dtype=dtype)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("ordered", [True, False])
+    def test_dataframe_dummies_preserve_categorical_dtype(self, dtype, ordered):
+        # GH13854
+        cat = Categorical(list("xy"), categories=list("xyz"), ordered=ordered)
+        result = get_dummies(cat, dtype=dtype)
+
+        data = np.array([[1, 0, 0], [0, 1, 0]], dtype=self.effective_dtype(dtype))
+        cols = CategoricalIndex(
+            cat.categories, categories=cat.categories, ordered=ordered
+        )
+        expected = DataFrame(data, columns=cols, dtype=self.effective_dtype(dtype))
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("sparse", [True, False])
+    def test_get_dummies_dont_sparsify_all_columns(self, sparse):
+        # GH18914
+        df = DataFrame.from_dict({"GDP": [1, 2], "Nation": ["AB", "CD"]})
+        df = get_dummies(df, columns=["Nation"], sparse=sparse)
+        df2 = df.reindex(columns=["GDP"])
+
+        tm.assert_frame_equal(df[["GDP"]], df2)
+
+    def test_get_dummies_duplicate_columns(self, df):
+        # GH20839
+        df.columns = ["A", "A", "A"]
+        result = get_dummies(df).sort_index(axis=1)
+
+        expected = DataFrame(
+            [
+                [1, True, False, True, False],
+                [2, False, True, True, False],
+                [3, True, False, False, True],
+            ],
+            columns=["A", "A_a", "A_b", "A_b", "A_c"],
+        ).sort_index(axis=1)
+
+        expected = expected.astype({"A": np.int64})
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_all_sparse(self):
+        df = DataFrame({"A": [1, 2]})
+        result = get_dummies(df, columns=["A"], sparse=True)
+        dtype = SparseDtype("bool", False)
+        expected = DataFrame(
+            {
+                "A_1": SparseArray([1, 0], dtype=dtype),
+                "A_2": SparseArray([0, 1], dtype=dtype),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("values", ["baz"])
+    def test_get_dummies_with_string_values(self, values):
+        # issue #28383
+        df = DataFrame(
+            {
+                "bar": [1, 2, 3, 4, 5, 6],
+                "foo": ["one", "one", "one", "two", "two", "two"],
+                "baz": ["A", "B", "C", "A", "B", "C"],
+                "zoo": ["x", "y", "z", "q", "w", "t"],
+            }
+        )
+
+        msg = "Input must be a list-like for parameter `columns`"
+
+        with pytest.raises(TypeError, match=msg):
+            get_dummies(df, columns=values)
+
+    def test_get_dummies_ea_dtype_series(self, any_numeric_ea_and_arrow_dtype):
+        # GH#32430
+        ser = Series(list("abca"))
+        result = get_dummies(ser, dtype=any_numeric_ea_and_arrow_dtype)
+        expected = DataFrame(
+            {"a": [1, 0, 0, 1], "b": [0, 1, 0, 0], "c": [0, 0, 1, 0]},
+            dtype=any_numeric_ea_and_arrow_dtype,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_get_dummies_ea_dtype_dataframe(self, any_numeric_ea_and_arrow_dtype):
+        # GH#32430
+        df = DataFrame({"x": list("abca")})
+        result = get_dummies(df, dtype=any_numeric_ea_and_arrow_dtype)
+        expected = DataFrame(
+            {"x_a": [1, 0, 0, 1], "x_b": [0, 1, 0, 0], "x_c": [0, 0, 1, 0]},
+            dtype=any_numeric_ea_and_arrow_dtype,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype_type", ["string", "category"])
+    def test_get_dummies_ea_dtype(self, dtype_type, string_dtype_no_object):
+        # GH#56273
+        dtype = string_dtype_no_object
+        exp_dtype = "boolean" if dtype.na_value is pd.NA else "bool"
+        if dtype_type == "category":
+            dtype = CategoricalDtype(Index(["a"], dtype))
+        df = DataFrame({"name": Series(["a"], dtype=dtype), "x": 1})
+        result = get_dummies(df)
+        expected = DataFrame({"x": 1, "name_a": Series([True], dtype=exp_dtype)})
+        tm.assert_frame_equal(result, expected)
+
+    @td.skip_if_no("pyarrow")
+    def test_get_dummies_arrow_dtype(self):
+        # GH#56273
+        df = DataFrame({"name": Series(["a"], dtype=ArrowDtype(pa.string())), "x": 1})
+        result = get_dummies(df)
+        expected = DataFrame({"x": 1, "name_a": Series([True], dtype="bool[pyarrow]")})
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {
+                "name": Series(
+                    ["a"],
+                    dtype=CategoricalDtype(Index(["a"], dtype=ArrowDtype(pa.string()))),
+                ),
+                "x": 1,
+            }
+        )
+        result = get_dummies(df)
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_melt.py b/pandas/pandas/tests/reshape/test_melt.py
new file mode 100644
index 0000000000000..95aa5291cb45a
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_melt.py
@@ -0,0 +1,1272 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+    lreshape,
+    melt,
+    wide_to_long,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df():
+    res = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    res["id1"] = (res["A"] > 0).astype(np.int64)
+    res["id2"] = (res["B"] > 0).astype(np.int64)
+    return res
+
+
+@pytest.fixture
+def df1():
+    res = DataFrame(
+        [
+            [1.067683, -1.110463, 0.20867],
+            [-1.321405, 0.368915, -1.055342],
+            [-0.807333, 0.08298, -0.873361],
+        ]
+    )
+    res.columns = [list("ABC"), list("abc")]
+    res.columns.names = ["CAP", "low"]
+    return res
+
+
+@pytest.fixture
+def var_name():
+    return "var"
+
+
+@pytest.fixture
+def value_name():
+    return "val"
+
+
+class TestMelt:
+    def test_top_level_method(self, df):
+        result = melt(df)
+        assert result.columns.tolist() == ["variable", "value"]
+
+    def test_method_signatures(self, df, df1, var_name, value_name):
+        tm.assert_frame_equal(df.melt(), melt(df))
+
+        tm.assert_frame_equal(
+            df.melt(id_vars=["id1", "id2"], value_vars=["A", "B"]),
+            melt(df, id_vars=["id1", "id2"], value_vars=["A", "B"]),
+        )
+
+        tm.assert_frame_equal(
+            df.melt(var_name=var_name, value_name=value_name),
+            melt(df, var_name=var_name, value_name=value_name),
+        )
+
+        tm.assert_frame_equal(df1.melt(col_level=0), melt(df1, col_level=0))
+
+    def test_default_col_names(self, df):
+        result = df.melt()
+        assert result.columns.tolist() == ["variable", "value"]
+
+        result1 = df.melt(id_vars=["id1"])
+        assert result1.columns.tolist() == ["id1", "variable", "value"]
+
+        result2 = df.melt(id_vars=["id1", "id2"])
+        assert result2.columns.tolist() == ["id1", "id2", "variable", "value"]
+
+    def test_value_vars(self, df):
+        result3 = df.melt(id_vars=["id1", "id2"], value_vars="A")
+        assert len(result3) == 10
+
+        result4 = df.melt(id_vars=["id1", "id2"], value_vars=["A", "B"])
+        expected4 = DataFrame(
+            {
+                "id1": df["id1"].tolist() * 2,
+                "id2": df["id2"].tolist() * 2,
+                "variable": ["A"] * 10 + ["B"] * 10,
+                "value": (df["A"].tolist() + df["B"].tolist()),
+            },
+            columns=["id1", "id2", "variable", "value"],
+        )
+        tm.assert_frame_equal(result4, expected4)
+
+    @pytest.mark.parametrize("type_", (tuple, list, np.array))
+    def test_value_vars_types(self, type_, df):
+        # GH 15348
+        expected = DataFrame(
+            {
+                "id1": df["id1"].tolist() * 2,
+                "id2": df["id2"].tolist() * 2,
+                "variable": ["A"] * 10 + ["B"] * 10,
+                "value": (df["A"].tolist() + df["B"].tolist()),
+            },
+            columns=["id1", "id2", "variable", "value"],
+        )
+        result = df.melt(id_vars=["id1", "id2"], value_vars=type_(("A", "B")))
+        tm.assert_frame_equal(result, expected)
+
+    def test_vars_work_with_multiindex(self, df1):
+        expected = DataFrame(
+            {
+                ("A", "a"): df1[("A", "a")],
+                "CAP": ["B"] * len(df1),
+                "low": ["b"] * len(df1),
+                "value": df1[("B", "b")],
+            },
+            columns=[("A", "a"), "CAP", "low", "value"],
+        )
+
+        result = df1.melt(id_vars=[("A", "a")], value_vars=[("B", "b")])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "id_vars, value_vars, col_level, expected",
+        [
+            (
+                ["A"],
+                ["B"],
+                0,
+                {
+                    "A": {0: 1.067683, 1: -1.321405, 2: -0.807333},
+                    "CAP": {0: "B", 1: "B", 2: "B"},
+                    "value": {0: -1.110463, 1: 0.368915, 2: 0.08298},
+                },
+            ),
+            (
+                ["a"],
+                ["b"],
+                1,
+                {
+                    "a": {0: 1.067683, 1: -1.321405, 2: -0.807333},
+                    "low": {0: "b", 1: "b", 2: "b"},
+                    "value": {0: -1.110463, 1: 0.368915, 2: 0.08298},
+                },
+            ),
+        ],
+    )
+    def test_single_vars_work_with_multiindex(
+        self, id_vars, value_vars, col_level, expected, df1
+    ):
+        result = df1.melt(id_vars, value_vars, col_level=col_level)
+        expected = DataFrame(expected)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "id_vars, value_vars",
+        [
+            [("A", "a"), [("B", "b")]],
+            [[("A", "a")], ("B", "b")],
+            [("A", "a"), ("B", "b")],
+        ],
+    )
+    def test_tuple_vars_fail_with_multiindex(self, id_vars, value_vars, df1):
+        # melt should fail with an informative error message if
+        # the columns have a MultiIndex and a tuple is passed
+        # for id_vars or value_vars.
+        msg = r"(id|value)_vars must be a list of tuples when columns are a MultiIndex"
+        with pytest.raises(ValueError, match=msg):
+            df1.melt(id_vars=id_vars, value_vars=value_vars)
+
+    def test_custom_var_name(self, df, var_name):
+        result5 = df.melt(var_name=var_name)
+        assert result5.columns.tolist() == ["var", "value"]
+
+        result6 = df.melt(id_vars=["id1"], var_name=var_name)
+        assert result6.columns.tolist() == ["id1", "var", "value"]
+
+        result7 = df.melt(id_vars=["id1", "id2"], var_name=var_name)
+        assert result7.columns.tolist() == ["id1", "id2", "var", "value"]
+
+        result8 = df.melt(id_vars=["id1", "id2"], value_vars="A", var_name=var_name)
+        assert result8.columns.tolist() == ["id1", "id2", "var", "value"]
+
+        result9 = df.melt(
+            id_vars=["id1", "id2"], value_vars=["A", "B"], var_name=var_name
+        )
+        expected9 = DataFrame(
+            {
+                "id1": df["id1"].tolist() * 2,
+                "id2": df["id2"].tolist() * 2,
+                var_name: ["A"] * 10 + ["B"] * 10,
+                "value": (df["A"].tolist() + df["B"].tolist()),
+            },
+            columns=["id1", "id2", var_name, "value"],
+        )
+        tm.assert_frame_equal(result9, expected9)
+
+    def test_custom_value_name(self, df, value_name):
+        result10 = df.melt(value_name=value_name)
+        assert result10.columns.tolist() == ["variable", "val"]
+
+        result11 = df.melt(id_vars=["id1"], value_name=value_name)
+        assert result11.columns.tolist() == ["id1", "variable", "val"]
+
+        result12 = df.melt(id_vars=["id1", "id2"], value_name=value_name)
+        assert result12.columns.tolist() == ["id1", "id2", "variable", "val"]
+
+        result13 = df.melt(
+            id_vars=["id1", "id2"], value_vars="A", value_name=value_name
+        )
+        assert result13.columns.tolist() == ["id1", "id2", "variable", "val"]
+
+        result14 = df.melt(
+            id_vars=["id1", "id2"], value_vars=["A", "B"], value_name=value_name
+        )
+        expected14 = DataFrame(
+            {
+                "id1": df["id1"].tolist() * 2,
+                "id2": df["id2"].tolist() * 2,
+                "variable": ["A"] * 10 + ["B"] * 10,
+                value_name: (df["A"].tolist() + df["B"].tolist()),
+            },
+            columns=["id1", "id2", "variable", value_name],
+        )
+        tm.assert_frame_equal(result14, expected14)
+
+    def test_custom_var_and_value_name(self, df, value_name, var_name):
+        result15 = df.melt(var_name=var_name, value_name=value_name)
+        assert result15.columns.tolist() == ["var", "val"]
+
+        result16 = df.melt(id_vars=["id1"], var_name=var_name, value_name=value_name)
+        assert result16.columns.tolist() == ["id1", "var", "val"]
+
+        result17 = df.melt(
+            id_vars=["id1", "id2"], var_name=var_name, value_name=value_name
+        )
+        assert result17.columns.tolist() == ["id1", "id2", "var", "val"]
+
+        result18 = df.melt(
+            id_vars=["id1", "id2"],
+            value_vars="A",
+            var_name=var_name,
+            value_name=value_name,
+        )
+        assert result18.columns.tolist() == ["id1", "id2", "var", "val"]
+
+        result19 = df.melt(
+            id_vars=["id1", "id2"],
+            value_vars=["A", "B"],
+            var_name=var_name,
+            value_name=value_name,
+        )
+        expected19 = DataFrame(
+            {
+                "id1": df["id1"].tolist() * 2,
+                "id2": df["id2"].tolist() * 2,
+                var_name: ["A"] * 10 + ["B"] * 10,
+                value_name: (df["A"].tolist() + df["B"].tolist()),
+            },
+            columns=["id1", "id2", var_name, value_name],
+        )
+        tm.assert_frame_equal(result19, expected19)
+
+        df20 = df.copy()
+        df20.columns.name = "foo"
+        result20 = df20.melt()
+        assert result20.columns.tolist() == ["foo", "value"]
+
+    @pytest.mark.parametrize("col_level", [0, "CAP"])
+    def test_col_level(self, col_level, df1):
+        res = df1.melt(col_level=col_level)
+        assert res.columns.tolist() == ["CAP", "value"]
+
+    def test_multiindex(self, df1):
+        res = df1.melt()
+        assert res.columns.tolist() == ["CAP", "low", "value"]
+
+    @pytest.mark.parametrize(
+        "col",
+        [
+            date_range("2010", periods=5, tz="US/Pacific"),
+            pd.Categorical(["a", "b", "c", "a", "d"]),
+            [0, 1, 0, 0, 0],
+        ],
+    )
+    def test_pandas_dtypes(self, col):
+        # GH 15785
+        col = pd.Series(col)
+        df = DataFrame(
+            {"klass": range(5), "col": col, "attr1": [1, 0, 0, 0, 0], "attr2": col}
+        )
+        expected_value = pd.concat([pd.Series([1, 0, 0, 0, 0]), col], ignore_index=True)
+        result = melt(
+            df, id_vars=["klass", "col"], var_name="attribute", value_name="value"
+        )
+        expected = DataFrame(
+            {
+                0: list(range(5)) * 2,
+                1: pd.concat([col] * 2, ignore_index=True),
+                2: ["attr1"] * 5 + ["attr2"] * 5,
+                3: expected_value,
+            }
+        )
+        expected.columns = ["klass", "col", "attribute", "value"]
+        tm.assert_frame_equal(result, expected)
+
+    def test_preserve_category(self):
+        # GH 15853
+        data = DataFrame({"A": [1, 2], "B": pd.Categorical(["X", "Y"])})
+        result = melt(data, ["B"], ["A"])
+        expected = DataFrame(
+            {"B": pd.Categorical(["X", "Y"]), "variable": ["A", "A"], "value": [1, 2]}
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_missing_columns_raises(self):
+        # GH-23575
+        # This test is to ensure that pandas raises an error if melting is
+        # attempted with column names absent from the dataframe
+
+        # Generate data
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((5, 4)), columns=list("abcd")
+        )
+
+        # Try to melt with missing `value_vars` column name
+        msg = "The following id_vars or value_vars are not present in the DataFrame:"
+        with pytest.raises(KeyError, match=msg):
+            df.melt(["a", "b"], ["C", "d"])
+
+        # Try to melt with missing `id_vars` column name
+        with pytest.raises(KeyError, match=msg):
+            df.melt(["A", "b"], ["c", "d"])
+
+        # Multiple missing
+        with pytest.raises(
+            KeyError,
+            match=msg,
+        ):
+            df.melt(["a", "b", "not_here", "or_there"], ["c", "d"])
+
+        # Multiindex melt fails if column is missing from multilevel melt
+        df.columns = [list("ABCD"), list("abcd")]
+        with pytest.raises(KeyError, match=msg):
+            df.melt([("E", "a")], [("B", "b")])
+        # Multiindex fails if column is missing from single level melt
+        with pytest.raises(KeyError, match=msg):
+            df.melt(["A"], ["F"], col_level=0)
+
+    def test_melt_mixed_int_str_id_vars(self):
+        # GH 29718
+        df = DataFrame({0: ["foo"], "a": ["bar"], "b": [1], "d": [2]})
+        result = melt(df, id_vars=[0, "a"], value_vars=["b", "d"])
+        expected = DataFrame(
+            {0: ["foo"] * 2, "a": ["bar"] * 2, "variable": list("bd"), "value": [1, 2]}
+        )
+        # the df's columns are mixed type and thus object -> preserves object dtype
+        expected["variable"] = expected["variable"].astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_mixed_int_str_value_vars(self):
+        # GH 29718
+        df = DataFrame({0: ["foo"], "a": ["bar"]})
+        result = melt(df, value_vars=[0, "a"])
+        expected = DataFrame({"variable": [0, "a"], "value": ["foo", "bar"]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_ignore_index(self):
+        # GH 17440
+        df = DataFrame({"foo": [0], "bar": [1]}, index=["first"])
+        result = melt(df, ignore_index=False)
+        expected = DataFrame(
+            {"variable": ["foo", "bar"], "value": [0, 1]}, index=["first", "first"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_ignore_multiindex(self):
+        # GH 17440
+        index = pd.MultiIndex.from_tuples(
+            [("first", "second"), ("first", "third")], names=["baz", "foobar"]
+        )
+        df = DataFrame({"foo": [0, 1], "bar": [2, 3]}, index=index)
+        result = melt(df, ignore_index=False)
+
+        expected_index = pd.MultiIndex.from_tuples(
+            [("first", "second"), ("first", "third")] * 2, names=["baz", "foobar"]
+        )
+        expected = DataFrame(
+            {"variable": ["foo"] * 2 + ["bar"] * 2, "value": [0, 1, 2, 3]},
+            index=expected_index,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_ignore_index_name_and_type(self):
+        # GH 17440
+        index = Index(["foo", "bar"], dtype="category", name="baz")
+        df = DataFrame({"x": [0, 1], "y": [2, 3]}, index=index)
+        result = melt(df, ignore_index=False)
+
+        expected_index = Index(["foo", "bar"] * 2, dtype="category", name="baz")
+        expected = DataFrame(
+            {"variable": ["x", "x", "y", "y"], "value": [0, 1, 2, 3]},
+            index=expected_index,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_with_duplicate_columns(self):
+        # GH#41951
+        df = DataFrame([["id", 2, 3]], columns=["a", "b", "b"])
+        result = df.melt(id_vars=["a"], value_vars=["b"])
+        expected = DataFrame(
+            [["id", "b", 2], ["id", "b", 3]], columns=["a", "variable", "value"]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["Int8", "Int64"])
+    def test_melt_ea_dtype(self, dtype):
+        # GH#41570
+        df = DataFrame(
+            {
+                "a": pd.Series([1, 2], dtype="Int8"),
+                "b": pd.Series([3, 4], dtype=dtype),
+            }
+        )
+        result = df.melt()
+        expected = DataFrame(
+            {
+                "variable": ["a", "a", "b", "b"],
+                "value": pd.Series([1, 2, 3, 4], dtype=dtype),
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_ea_columns(self):
+        # GH 54297
+        df = DataFrame(
+            {
+                "A": {0: "a", 1: "b", 2: "c"},
+                "B": {0: 1, 1: 3, 2: 5},
+                "C": {0: 2, 1: 4, 2: 6},
+            }
+        )
+        df.columns = df.columns.astype("string[python]")
+        result = df.melt(id_vars=["A"], value_vars=["B"])
+        expected = DataFrame(
+            {
+                "A": list("abc"),
+                "variable": pd.Series(["B"] * 3, dtype="string[python]"),
+                "value": [1, 3, 5],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_preserves_datetime(self):
+        df = DataFrame(
+            data=[
+                {
+                    "type": "A0",
+                    "start_date": pd.Timestamp("2023/03/01", tz="Asia/Tokyo"),
+                    "end_date": pd.Timestamp("2023/03/10", tz="Asia/Tokyo"),
+                },
+                {
+                    "type": "A1",
+                    "start_date": pd.Timestamp("2023/03/01", tz="Asia/Tokyo"),
+                    "end_date": pd.Timestamp("2023/03/11", tz="Asia/Tokyo"),
+                },
+            ],
+            index=["aaaa", "bbbb"],
+        )
+        result = df.melt(
+            id_vars=["type"],
+            value_vars=["start_date", "end_date"],
+            var_name="start/end",
+            value_name="date",
+        )
+        expected = DataFrame(
+            {
+                "type": {0: "A0", 1: "A1", 2: "A0", 3: "A1"},
+                "start/end": {
+                    0: "start_date",
+                    1: "start_date",
+                    2: "end_date",
+                    3: "end_date",
+                },
+                "date": {
+                    0: pd.Timestamp("2023-03-01 00:00:00+0900", tz="Asia/Tokyo"),
+                    1: pd.Timestamp("2023-03-01 00:00:00+0900", tz="Asia/Tokyo"),
+                    2: pd.Timestamp("2023-03-10 00:00:00+0900", tz="Asia/Tokyo"),
+                    3: pd.Timestamp("2023-03-11 00:00:00+0900", tz="Asia/Tokyo"),
+                },
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_allows_non_scalar_id_vars(self):
+        df = DataFrame(
+            data={"a": [1, 2, 3], "b": [4, 5, 6]},
+            index=["11", "22", "33"],
+        )
+        result = df.melt(
+            id_vars="a",
+            var_name=0,
+            value_name=1,
+        )
+        expected = DataFrame({"a": [1, 2, 3], 0: ["b"] * 3, 1: [4, 5, 6]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_allows_non_string_var_name(self):
+        df = DataFrame(
+            data={"a": [1, 2, 3], "b": [4, 5, 6]},
+            index=["11", "22", "33"],
+        )
+        result = df.melt(
+            id_vars=["a"],
+            var_name=0,
+            value_name=1,
+        )
+        expected = DataFrame({"a": [1, 2, 3], 0: ["b"] * 3, 1: [4, 5, 6]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_melt_non_scalar_var_name_raises(self):
+        df = DataFrame(
+            data={"a": [1, 2, 3], "b": [4, 5, 6]},
+            index=["11", "22", "33"],
+        )
+        with pytest.raises(ValueError, match=r".* must be a scalar."):
+            df.melt(id_vars=["a"], var_name=[1, 2])
+
+    def test_melt_multiindex_columns_var_name(self):
+        # GH 58033
+        df = DataFrame({("A", "a"): [1], ("A", "b"): [2]})
+
+        expected = DataFrame(
+            [("A", "a", 1), ("A", "b", 2)], columns=["first", "second", "value"]
+        )
+
+        tm.assert_frame_equal(df.melt(var_name=["first", "second"]), expected)
+        tm.assert_frame_equal(df.melt(var_name=["first"]), expected[["first", "value"]])
+
+    def test_melt_multiindex_columns_var_name_too_many(self):
+        # GH 58033
+        df = DataFrame({("A", "a"): [1], ("A", "b"): [2]})
+
+        with pytest.raises(
+            ValueError, match="but the dataframe columns only have 2 levels"
+        ):
+            df.melt(var_name=["first", "second", "third"])
+
+
+class TestLreshape:
+    def test_pairs(self):
+        data = {
+            "birthdt": [
+                "08jan2009",
+                "20dec2008",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+            ],
+            "birthwt": [1766, 3301, 1454, 3139, 4133],
+            "id": [101, 102, 103, 104, 105],
+            "sex": ["Male", "Female", "Female", "Female", "Female"],
+            "visitdt1": [
+                "11jan2009",
+                "22dec2008",
+                "04jan2009",
+                "29dec2008",
+                "20jan2009",
+            ],
+            "visitdt2": ["21jan2009", np.nan, "22jan2009", "31dec2008", "03feb2009"],
+            "visitdt3": ["05feb2009", np.nan, np.nan, "02jan2009", "15feb2009"],
+            "wt1": [1823, 3338, 1549, 3298, 4306],
+            "wt2": [2011.0, np.nan, 1892.0, 3338.0, 4575.0],
+            "wt3": [2293.0, np.nan, np.nan, 3377.0, 4805.0],
+        }
+
+        df = DataFrame(data)
+
+        spec = {
+            "visitdt": [f"visitdt{i:d}" for i in range(1, 4)],
+            "wt": [f"wt{i:d}" for i in range(1, 4)],
+        }
+        result = lreshape(df, spec)
+
+        exp_data = {
+            "birthdt": [
+                "08jan2009",
+                "20dec2008",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+                "08jan2009",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+                "08jan2009",
+                "21dec2008",
+                "11jan2009",
+            ],
+            "birthwt": [
+                1766,
+                3301,
+                1454,
+                3139,
+                4133,
+                1766,
+                1454,
+                3139,
+                4133,
+                1766,
+                3139,
+                4133,
+            ],
+            "id": [101, 102, 103, 104, 105, 101, 103, 104, 105, 101, 104, 105],
+            "sex": [
+                "Male",
+                "Female",
+                "Female",
+                "Female",
+                "Female",
+                "Male",
+                "Female",
+                "Female",
+                "Female",
+                "Male",
+                "Female",
+                "Female",
+            ],
+            "visitdt": [
+                "11jan2009",
+                "22dec2008",
+                "04jan2009",
+                "29dec2008",
+                "20jan2009",
+                "21jan2009",
+                "22jan2009",
+                "31dec2008",
+                "03feb2009",
+                "05feb2009",
+                "02jan2009",
+                "15feb2009",
+            ],
+            "wt": [
+                1823.0,
+                3338.0,
+                1549.0,
+                3298.0,
+                4306.0,
+                2011.0,
+                1892.0,
+                3338.0,
+                4575.0,
+                2293.0,
+                3377.0,
+                4805.0,
+            ],
+        }
+        exp = DataFrame(exp_data, columns=result.columns)
+        tm.assert_frame_equal(result, exp)
+
+        result = lreshape(df, spec, dropna=False)
+        exp_data = {
+            "birthdt": [
+                "08jan2009",
+                "20dec2008",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+                "08jan2009",
+                "20dec2008",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+                "08jan2009",
+                "20dec2008",
+                "30dec2008",
+                "21dec2008",
+                "11jan2009",
+            ],
+            "birthwt": [
+                1766,
+                3301,
+                1454,
+                3139,
+                4133,
+                1766,
+                3301,
+                1454,
+                3139,
+                4133,
+                1766,
+                3301,
+                1454,
+                3139,
+                4133,
+            ],
+            "id": [
+                101,
+                102,
+                103,
+                104,
+                105,
+                101,
+                102,
+                103,
+                104,
+                105,
+                101,
+                102,
+                103,
+                104,
+                105,
+            ],
+            "sex": [
+                "Male",
+                "Female",
+                "Female",
+                "Female",
+                "Female",
+                "Male",
+                "Female",
+                "Female",
+                "Female",
+                "Female",
+                "Male",
+                "Female",
+                "Female",
+                "Female",
+                "Female",
+            ],
+            "visitdt": [
+                "11jan2009",
+                "22dec2008",
+                "04jan2009",
+                "29dec2008",
+                "20jan2009",
+                "21jan2009",
+                np.nan,
+                "22jan2009",
+                "31dec2008",
+                "03feb2009",
+                "05feb2009",
+                np.nan,
+                np.nan,
+                "02jan2009",
+                "15feb2009",
+            ],
+            "wt": [
+                1823.0,
+                3338.0,
+                1549.0,
+                3298.0,
+                4306.0,
+                2011.0,
+                np.nan,
+                1892.0,
+                3338.0,
+                4575.0,
+                2293.0,
+                np.nan,
+                np.nan,
+                3377.0,
+                4805.0,
+            ],
+        }
+        exp = DataFrame(exp_data, columns=result.columns)
+        tm.assert_frame_equal(result, exp)
+
+        spec = {
+            "visitdt": [f"visitdt{i:d}" for i in range(1, 3)],
+            "wt": [f"wt{i:d}" for i in range(1, 4)],
+        }
+        msg = "All column lists must be same length"
+        with pytest.raises(ValueError, match=msg):
+            lreshape(df, spec)
+
+
+class TestWideToLong:
+    def test_simple(self):
+        x = np.random.default_rng(2).standard_normal(3)
+        df = DataFrame(
+            {
+                "A1970": {0: "a", 1: "b", 2: "c"},
+                "A1980": {0: "d", 1: "e", 2: "f"},
+                "B1970": {0: 2.5, 1: 1.2, 2: 0.7},
+                "B1980": {0: 3.2, 1: 1.3, 2: 0.1},
+                "X": dict(zip(range(3), x)),
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": x.tolist() + x.tolist(),
+            "A": ["a", "b", "c", "d", "e", "f"],
+            "B": [2.5, 1.2, 0.7, 3.2, 1.3, 0.1],
+            "year": [1970, 1970, 1970, 1980, 1980, 1980],
+            "id": [0, 1, 2, 0, 1, 2],
+        }
+        expected = DataFrame(exp_data)
+        expected = expected.set_index(["id", "year"])[["X", "A", "B"]]
+        result = wide_to_long(df, ["A", "B"], i="id", j="year")
+        tm.assert_frame_equal(result, expected)
+
+    def test_stubs(self):
+        # GH9204 wide_to_long call should not modify 'stubs' list
+        df = DataFrame([[0, 1, 2, 3, 8], [4, 5, 6, 7, 9]])
+        df.columns = ["id", "inc1", "inc2", "edu1", "edu2"]
+        stubs = ["inc", "edu"]
+
+        wide_to_long(df, stubs, i="id", j="age")
+
+        assert stubs == ["inc", "edu"]
+
+    def test_separating_character(self):
+        # GH14779
+
+        x = np.random.default_rng(2).standard_normal(3)
+        df = DataFrame(
+            {
+                "A.1970": {0: "a", 1: "b", 2: "c"},
+                "A.1980": {0: "d", 1: "e", 2: "f"},
+                "B.1970": {0: 2.5, 1: 1.2, 2: 0.7},
+                "B.1980": {0: 3.2, 1: 1.3, 2: 0.1},
+                "X": dict(zip(range(3), x)),
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": x.tolist() + x.tolist(),
+            "A": ["a", "b", "c", "d", "e", "f"],
+            "B": [2.5, 1.2, 0.7, 3.2, 1.3, 0.1],
+            "year": [1970, 1970, 1970, 1980, 1980, 1980],
+            "id": [0, 1, 2, 0, 1, 2],
+        }
+        expected = DataFrame(exp_data)
+        expected = expected.set_index(["id", "year"])[["X", "A", "B"]]
+        result = wide_to_long(df, ["A", "B"], i="id", j="year", sep=".")
+        tm.assert_frame_equal(result, expected)
+
+    def test_escapable_characters(self):
+        x = np.random.default_rng(2).standard_normal(3)
+        df = DataFrame(
+            {
+                "A(quarterly)1970": {0: "a", 1: "b", 2: "c"},
+                "A(quarterly)1980": {0: "d", 1: "e", 2: "f"},
+                "B(quarterly)1970": {0: 2.5, 1: 1.2, 2: 0.7},
+                "B(quarterly)1980": {0: 3.2, 1: 1.3, 2: 0.1},
+                "X": dict(zip(range(3), x)),
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": x.tolist() + x.tolist(),
+            "A(quarterly)": ["a", "b", "c", "d", "e", "f"],
+            "B(quarterly)": [2.5, 1.2, 0.7, 3.2, 1.3, 0.1],
+            "year": [1970, 1970, 1970, 1980, 1980, 1980],
+            "id": [0, 1, 2, 0, 1, 2],
+        }
+        expected = DataFrame(exp_data)
+        expected = expected.set_index(["id", "year"])[
+            ["X", "A(quarterly)", "B(quarterly)"]
+        ]
+        result = wide_to_long(df, ["A(quarterly)", "B(quarterly)"], i="id", j="year")
+        tm.assert_frame_equal(result, expected)
+
+    def test_unbalanced(self):
+        # test that we can have a varying amount of time variables
+        df = DataFrame(
+            {
+                "A2010": [1.0, 2.0],
+                "A2011": [3.0, 4.0],
+                "B2010": [5.0, 6.0],
+                "X": ["X1", "X2"],
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": ["X1", "X2", "X1", "X2"],
+            "A": [1.0, 2.0, 3.0, 4.0],
+            "B": [5.0, 6.0, np.nan, np.nan],
+            "id": [0, 1, 0, 1],
+            "year": [2010, 2010, 2011, 2011],
+        }
+        expected = DataFrame(exp_data)
+        expected = expected.set_index(["id", "year"])[["X", "A", "B"]]
+        result = wide_to_long(df, ["A", "B"], i="id", j="year")
+        tm.assert_frame_equal(result, expected)
+
+    def test_character_overlap(self):
+        # Test we handle overlapping characters in both id_vars and value_vars
+        df = DataFrame(
+            {
+                "A11": ["a11", "a22", "a33"],
+                "A12": ["a21", "a22", "a23"],
+                "B11": ["b11", "b12", "b13"],
+                "B12": ["b21", "b22", "b23"],
+                "BB11": [1, 2, 3],
+                "BB12": [4, 5, 6],
+                "BBBX": [91, 92, 93],
+                "BBBZ": [91, 92, 93],
+            }
+        )
+        df["id"] = df.index
+        expected = DataFrame(
+            {
+                "BBBX": [91, 92, 93, 91, 92, 93],
+                "BBBZ": [91, 92, 93, 91, 92, 93],
+                "A": ["a11", "a22", "a33", "a21", "a22", "a23"],
+                "B": ["b11", "b12", "b13", "b21", "b22", "b23"],
+                "BB": [1, 2, 3, 4, 5, 6],
+                "id": [0, 1, 2, 0, 1, 2],
+                "year": [11, 11, 11, 12, 12, 12],
+            }
+        )
+        expected = expected.set_index(["id", "year"])[["BBBX", "BBBZ", "A", "B", "BB"]]
+        result = wide_to_long(df, ["A", "B", "BB"], i="id", j="year")
+        tm.assert_frame_equal(result.sort_index(axis=1), expected.sort_index(axis=1))
+
+    def test_invalid_separator(self):
+        # if an invalid separator is supplied a empty data frame is returned
+        sep = "nope!"
+        df = DataFrame(
+            {
+                "A2010": [1.0, 2.0],
+                "A2011": [3.0, 4.0],
+                "B2010": [5.0, 6.0],
+                "X": ["X1", "X2"],
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": "",
+            "A2010": [],
+            "A2011": [],
+            "B2010": [],
+            "id": [],
+            "year": [],
+            "A": [],
+            "B": [],
+        }
+        expected = DataFrame(exp_data).astype({"year": np.int64})
+        expected = expected.set_index(["id", "year"])[
+            ["X", "A2010", "A2011", "B2010", "A", "B"]
+        ]
+        expected.index = expected.index.set_levels([0, 1], level=0)
+        result = wide_to_long(df, ["A", "B"], i="id", j="year", sep=sep)
+        tm.assert_frame_equal(result.sort_index(axis=1), expected.sort_index(axis=1))
+
+    def test_num_string_disambiguation(self):
+        # Test that we can disambiguate number value_vars from
+        # string value_vars
+        df = DataFrame(
+            {
+                "A11": ["a11", "a22", "a33"],
+                "A12": ["a21", "a22", "a23"],
+                "B11": ["b11", "b12", "b13"],
+                "B12": ["b21", "b22", "b23"],
+                "BB11": [1, 2, 3],
+                "BB12": [4, 5, 6],
+                "Arating": [91, 92, 93],
+                "Arating_old": [91, 92, 93],
+            }
+        )
+        df["id"] = df.index
+        expected = DataFrame(
+            {
+                "Arating": [91, 92, 93, 91, 92, 93],
+                "Arating_old": [91, 92, 93, 91, 92, 93],
+                "A": ["a11", "a22", "a33", "a21", "a22", "a23"],
+                "B": ["b11", "b12", "b13", "b21", "b22", "b23"],
+                "BB": [1, 2, 3, 4, 5, 6],
+                "id": [0, 1, 2, 0, 1, 2],
+                "year": [11, 11, 11, 12, 12, 12],
+            }
+        )
+        expected = expected.set_index(["id", "year"])[
+            ["Arating", "Arating_old", "A", "B", "BB"]
+        ]
+        result = wide_to_long(df, ["A", "B", "BB"], i="id", j="year")
+        tm.assert_frame_equal(result.sort_index(axis=1), expected.sort_index(axis=1))
+
+    def test_invalid_suffixtype(self):
+        # If all stubs names end with a string, but a numeric suffix is
+        # assumed,  an empty data frame is returned
+        df = DataFrame(
+            {
+                "Aone": [1.0, 2.0],
+                "Atwo": [3.0, 4.0],
+                "Bone": [5.0, 6.0],
+                "X": ["X1", "X2"],
+            }
+        )
+        df["id"] = df.index
+        exp_data = {
+            "X": "",
+            "Aone": [],
+            "Atwo": [],
+            "Bone": [],
+            "id": [],
+            "year": [],
+            "A": [],
+            "B": [],
+        }
+        expected = DataFrame(exp_data).astype({"year": np.int64})
+
+        expected = expected.set_index(["id", "year"])
+        expected.index = expected.index.set_levels([0, 1], level=0)
+        result = wide_to_long(df, ["A", "B"], i="id", j="year")
+        tm.assert_frame_equal(result.sort_index(axis=1), expected.sort_index(axis=1))
+
+    def test_multiple_id_columns(self):
+        # Taken from http://www.ats.ucla.edu/stat/stata/modules/reshapel.htm
+        df = DataFrame(
+            {
+                "famid": [1, 1, 1, 2, 2, 2, 3, 3, 3],
+                "birth": [1, 2, 3, 1, 2, 3, 1, 2, 3],
+                "ht1": [2.8, 2.9, 2.2, 2, 1.8, 1.9, 2.2, 2.3, 2.1],
+                "ht2": [3.4, 3.8, 2.9, 3.2, 2.8, 2.4, 3.3, 3.4, 2.9],
+            }
+        )
+        expected = DataFrame(
+            {
+                "ht": [
+                    2.8,
+                    3.4,
+                    2.9,
+                    3.8,
+                    2.2,
+                    2.9,
+                    2.0,
+                    3.2,
+                    1.8,
+                    2.8,
+                    1.9,
+                    2.4,
+                    2.2,
+                    3.3,
+                    2.3,
+                    3.4,
+                    2.1,
+                    2.9,
+                ],
+                "famid": [1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3],
+                "birth": [1, 1, 2, 2, 3, 3, 1, 1, 2, 2, 3, 3, 1, 1, 2, 2, 3, 3],
+                "age": [1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2],
+            }
+        )
+        expected = expected.set_index(["famid", "birth", "age"])[["ht"]]
+        result = wide_to_long(df, "ht", i=["famid", "birth"], j="age")
+        tm.assert_frame_equal(result, expected)
+
+    def test_non_unique_idvars(self):
+        # GH16382
+        # Raise an error message if non unique id vars (i) are passed
+        df = DataFrame(
+            {"A_A1": [1, 2, 3, 4, 5], "B_B1": [1, 2, 3, 4, 5], "x": [1, 1, 1, 1, 1]}
+        )
+        msg = "the id variables need to uniquely identify each row"
+        with pytest.raises(ValueError, match=msg):
+            wide_to_long(df, ["A_A", "B_B"], i="x", j="colname")
+
+    def test_cast_j_int(self):
+        df = DataFrame(
+            {
+                "actor_1": ["CCH Pounder", "Johnny Depp", "Christoph Waltz"],
+                "actor_2": ["Joel David Moore", "Orlando Bloom", "Rory Kinnear"],
+                "actor_fb_likes_1": [1000.0, 40000.0, 11000.0],
+                "actor_fb_likes_2": [936.0, 5000.0, 393.0],
+                "title": ["Avatar", "Pirates of the Caribbean", "Spectre"],
+            }
+        )
+
+        expected = DataFrame(
+            {
+                "actor": [
+                    "CCH Pounder",
+                    "Johnny Depp",
+                    "Christoph Waltz",
+                    "Joel David Moore",
+                    "Orlando Bloom",
+                    "Rory Kinnear",
+                ],
+                "actor_fb_likes": [1000.0, 40000.0, 11000.0, 936.0, 5000.0, 393.0],
+                "num": [1, 1, 1, 2, 2, 2],
+                "title": [
+                    "Avatar",
+                    "Pirates of the Caribbean",
+                    "Spectre",
+                    "Avatar",
+                    "Pirates of the Caribbean",
+                    "Spectre",
+                ],
+            }
+        ).set_index(["title", "num"])
+        result = wide_to_long(
+            df, ["actor", "actor_fb_likes"], i="title", j="num", sep="_"
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_identical_stubnames(self):
+        df = DataFrame(
+            {
+                "A2010": [1.0, 2.0],
+                "A2011": [3.0, 4.0],
+                "B2010": [5.0, 6.0],
+                "A": ["X1", "X2"],
+            }
+        )
+        msg = "stubname can't be identical to a column name"
+        with pytest.raises(ValueError, match=msg):
+            wide_to_long(df, ["A", "B"], i="A", j="colname")
+
+    def test_nonnumeric_suffix(self):
+        df = DataFrame(
+            {
+                "treatment_placebo": [1.0, 2.0],
+                "treatment_test": [3.0, 4.0],
+                "result_placebo": [5.0, 6.0],
+                "A": ["X1", "X2"],
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": ["X1", "X2", "X1", "X2"],
+                "colname": ["placebo", "placebo", "test", "test"],
+                "result": [5.0, 6.0, np.nan, np.nan],
+                "treatment": [1.0, 2.0, 3.0, 4.0],
+            }
+        )
+        expected = expected.set_index(["A", "colname"])
+        result = wide_to_long(
+            df, ["result", "treatment"], i="A", j="colname", suffix="[a-z]+", sep="_"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_mixed_type_suffix(self):
+        df = DataFrame(
+            {
+                "A": ["X1", "X2"],
+                "result_1": [0, 9],
+                "result_foo": [5.0, 6.0],
+                "treatment_1": [1.0, 2.0],
+                "treatment_foo": [3.0, 4.0],
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": ["X1", "X2", "X1", "X2"],
+                "colname": ["1", "1", "foo", "foo"],
+                "result": [0.0, 9.0, 5.0, 6.0],
+                "treatment": [1.0, 2.0, 3.0, 4.0],
+            }
+        ).set_index(["A", "colname"])
+        result = wide_to_long(
+            df, ["result", "treatment"], i="A", j="colname", suffix=".+", sep="_"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_float_suffix(self):
+        df = DataFrame(
+            {
+                "treatment_1.1": [1.0, 2.0],
+                "treatment_2.1": [3.0, 4.0],
+                "result_1.2": [5.0, 6.0],
+                "result_1": [0, 9],
+                "A": ["X1", "X2"],
+            }
+        )
+        expected = DataFrame(
+            {
+                "A": ["X1", "X2", "X1", "X2", "X1", "X2", "X1", "X2"],
+                "colname": [1.2, 1.2, 1.0, 1.0, 1.1, 1.1, 2.1, 2.1],
+                "result": [5.0, 6.0, 0.0, 9.0, np.nan, np.nan, np.nan, np.nan],
+                "treatment": [np.nan, np.nan, np.nan, np.nan, 1.0, 2.0, 3.0, 4.0],
+            }
+        )
+        expected = expected.set_index(["A", "colname"])
+        result = wide_to_long(
+            df, ["result", "treatment"], i="A", j="colname", suffix="[0-9.]+", sep="_"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_col_substring_of_stubname(self):
+        # GH22468
+        # Don't raise ValueError when a column name is a substring
+        # of a stubname that's been passed as a string
+        wide_data = {
+            "node_id": {0: 0, 1: 1, 2: 2, 3: 3, 4: 4},
+            "A": {0: 0.80, 1: 0.0, 2: 0.25, 3: 1.0, 4: 0.81},
+            "PA0": {0: 0.74, 1: 0.56, 2: 0.56, 3: 0.98, 4: 0.6},
+            "PA1": {0: 0.77, 1: 0.64, 2: 0.52, 3: 0.98, 4: 0.67},
+            "PA3": {0: 0.34, 1: 0.70, 2: 0.52, 3: 0.98, 4: 0.67},
+        }
+        wide_df = DataFrame.from_dict(wide_data)
+        expected = wide_to_long(wide_df, stubnames=["PA"], i=["node_id", "A"], j="time")
+        result = wide_to_long(wide_df, stubnames="PA", i=["node_id", "A"], j="time")
+        tm.assert_frame_equal(result, expected)
+
+    def test_raise_of_column_name_value(self):
+        # GH34731, enforced in 2.0
+        # raise a ValueError if the resultant value column name matches
+        # a name in the dataframe already (default name is "value")
+        df = DataFrame({"col": list("ABC"), "value": range(10, 16, 2)})
+
+        with pytest.raises(
+            ValueError, match=re.escape("value_name (value) cannot match")
+        ):
+            df.melt(id_vars="value", value_name="value")
+
+    def test_missing_stubname(self, any_string_dtype):
+        # GH46044
+        df = DataFrame({"id": ["1", "2"], "a-1": [100, 200], "a-2": [300, 400]})
+        df = df.astype({"id": any_string_dtype})
+        result = wide_to_long(
+            df,
+            stubnames=["a", "b"],
+            i="id",
+            j="num",
+            sep="-",
+        )
+        index = Index(
+            [("1", 1), ("2", 1), ("1", 2), ("2", 2)],
+            name=("id", "num"),
+        )
+        expected = DataFrame(
+            {"a": [100, 200, 300, 400], "b": [np.nan] * 4},
+            index=index,
+        )
+        new_level = expected.index.levels[0].astype(any_string_dtype)
+        if any_string_dtype == "object":
+            new_level = expected.index.levels[0].astype("str")
+        expected.index = expected.index.set_levels(new_level, level=0)
+        tm.assert_frame_equal(result, expected)
+
+
+def test_wide_to_long_string_columns(string_storage):
+    # GH 57066
+    string_dtype = pd.StringDtype(string_storage, na_value=np.nan)
+    df = DataFrame(
+        {
+            "ID": {0: 1},
+            "R_test1": {0: 1},
+            "R_test2": {0: 1},
+            "R_test3": {0: 2},
+            "D": {0: 1},
+        }
+    )
+    df.columns = df.columns.astype(string_dtype)
+    result = wide_to_long(
+        df, stubnames="R", i="ID", j="UNPIVOTED", sep="_", suffix=".*"
+    )
+    expected = DataFrame(
+        [[1, 1], [1, 1], [1, 2]],
+        columns=Index(["D", "R"]),
+        index=pd.MultiIndex.from_arrays(
+            [
+                [1, 1, 1],
+                Index(["test1", "test2", "test3"], dtype=string_dtype),
+            ],
+            names=["ID", "UNPIVOTED"],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_pivot.py b/pandas/pandas/tests/reshape/test_pivot.py
new file mode 100644
index 0000000000000..374d236c8ff39
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_pivot.py
@@ -0,0 +1,2829 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+from itertools import product
+import re
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat.numpy import np_version_gte1p25
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Grouper,
+    Index,
+    MultiIndex,
+    Series,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+from pandas.core.reshape import reshape as reshape_lib
+from pandas.core.reshape.pivot import pivot_table
+
+
+class TestPivotTable:
+    @pytest.fixture
+    def data(self):
+        return DataFrame(
+            {
+                "A": [
+                    "foo",
+                    "foo",
+                    "foo",
+                    "foo",
+                    "bar",
+                    "bar",
+                    "bar",
+                    "bar",
+                    "foo",
+                    "foo",
+                    "foo",
+                ],
+                "B": [
+                    "one",
+                    "one",
+                    "one",
+                    "two",
+                    "one",
+                    "one",
+                    "one",
+                    "two",
+                    "two",
+                    "two",
+                    "one",
+                ],
+                "C": [
+                    "dull",
+                    "dull",
+                    "shiny",
+                    "dull",
+                    "dull",
+                    "shiny",
+                    "shiny",
+                    "dull",
+                    "shiny",
+                    "shiny",
+                    "shiny",
+                ],
+                "D": np.random.default_rng(2).standard_normal(11),
+                "E": np.random.default_rng(2).standard_normal(11),
+                "F": np.random.default_rng(2).standard_normal(11),
+            }
+        )
+
+    def test_pivot_table(self, observed, data):
+        index = ["A", "B"]
+        columns = "C"
+        table = pivot_table(
+            data, values="D", index=index, columns=columns, observed=observed
+        )
+
+        table2 = data.pivot_table(
+            values="D", index=index, columns=columns, observed=observed
+        )
+        tm.assert_frame_equal(table, table2)
+
+        # this works
+        pivot_table(data, values="D", index=index, observed=observed)
+
+        if len(index) > 1:
+            assert table.index.names == tuple(index)
+        else:
+            assert table.index.name == index[0]
+
+        if len(columns) > 1:
+            assert table.columns.names == columns
+        else:
+            assert table.columns.name == columns[0]
+
+        expected = data.groupby(index + [columns])["D"].agg("mean").unstack()
+        tm.assert_frame_equal(table, expected)
+
+    def test_pivot_table_categorical_observed_equal(self, observed):
+        # issue #24923
+        df = DataFrame(
+            {"col1": list("abcde"), "col2": list("fghij"), "col3": [1, 2, 3, 4, 5]}
+        )
+
+        expected = df.pivot_table(
+            index="col1", values="col3", columns="col2", aggfunc="sum", fill_value=0
+        )
+
+        expected.index = expected.index.astype("category")
+        expected.columns = expected.columns.astype("category")
+
+        df.col1 = df.col1.astype("category")
+        df.col2 = df.col2.astype("category")
+
+        result = df.pivot_table(
+            index="col1",
+            values="col3",
+            columns="col2",
+            aggfunc="sum",
+            fill_value=0,
+            observed=observed,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_nocols(self):
+        df = DataFrame(
+            {"rows": ["a", "b", "c"], "cols": ["x", "y", "z"], "values": [1, 2, 3]}
+        )
+        rs = df.pivot_table(columns="cols", aggfunc="sum")
+        xp = df.pivot_table(index="cols", aggfunc="sum").T
+        tm.assert_frame_equal(rs, xp)
+
+        rs = df.pivot_table(columns="cols", aggfunc={"values": "mean"})
+        xp = df.pivot_table(index="cols", aggfunc={"values": "mean"}).T
+        tm.assert_frame_equal(rs, xp)
+
+    def test_pivot_table_dropna(self):
+        df = DataFrame(
+            {
+                "amount": {0: 60000, 1: 100000, 2: 50000, 3: 30000},
+                "customer": {0: "A", 1: "A", 2: "B", 3: "C"},
+                "month": {0: 201307, 1: 201309, 2: 201308, 3: 201310},
+                "product": {0: "a", 1: "b", 2: "c", 3: "d"},
+                "quantity": {0: 2000000, 1: 500000, 2: 1000000, 3: 1000000},
+            }
+        )
+        pv_col = df.pivot_table(
+            "quantity", "month", ["customer", "product"], dropna=False
+        )
+        pv_ind = df.pivot_table(
+            "quantity", ["customer", "product"], "month", dropna=False
+        )
+
+        m = MultiIndex.from_tuples(
+            [
+                ("A", "a"),
+                ("A", "b"),
+                ("A", "c"),
+                ("A", "d"),
+                ("B", "a"),
+                ("B", "b"),
+                ("B", "c"),
+                ("B", "d"),
+                ("C", "a"),
+                ("C", "b"),
+                ("C", "c"),
+                ("C", "d"),
+            ],
+            names=["customer", "product"],
+        )
+        tm.assert_index_equal(pv_col.columns, m)
+        tm.assert_index_equal(pv_ind.index, m)
+
+    def test_pivot_table_categorical(self):
+        cat1 = Categorical(
+            ["a", "a", "b", "b"], categories=["a", "b", "z"], ordered=True
+        )
+        cat2 = Categorical(
+            ["c", "d", "c", "d"], categories=["c", "d", "y"], ordered=True
+        )
+        df = DataFrame({"A": cat1, "B": cat2, "values": [1, 2, 3, 4]})
+        result = pivot_table(
+            df, values="values", index=["A", "B"], dropna=True, observed=False
+        )
+
+        exp_index = MultiIndex.from_arrays([cat1, cat2], names=["A", "B"])
+        expected = DataFrame({"values": [1.0, 2.0, 3.0, 4.0]}, index=exp_index)
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_dropna_categoricals(self, dropna):
+        # GH 15193
+        categories = ["a", "b", "c", "d"]
+
+        df = DataFrame(
+            {
+                "A": ["a", "a", "a", "b", "b", "b", "c", "c", "c"],
+                "B": [1, 2, 3, 1, 2, 3, 1, 2, 3],
+                "C": range(9),
+            }
+        )
+
+        df["A"] = df["A"].astype(CategoricalDtype(categories, ordered=False))
+        result = df.pivot_table(
+            index="B", columns="A", values="C", dropna=dropna, observed=False
+        )
+        expected_columns = Series(["a", "b", "c"], name="A")
+        expected_columns = expected_columns.astype(
+            CategoricalDtype(categories, ordered=False)
+        )
+        expected_index = Series([1, 2, 3], name="B")
+        expected = DataFrame(
+            [[0.0, 3.0, 6.0], [1.0, 4.0, 7.0], [2.0, 5.0, 8.0]],
+            index=expected_index,
+            columns=expected_columns,
+        )
+        if not dropna:
+            # add back the non observed to compare
+            expected = expected.reindex(columns=Categorical(categories)).astype("float")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_with_non_observable_dropna(self, dropna):
+        # gh-21133
+        df = DataFrame(
+            {
+                "A": Categorical(
+                    [np.nan, "low", "high", "low", "high"],
+                    categories=["low", "high"],
+                    ordered=True,
+                ),
+                "B": [0.0, 1.0, 2.0, 3.0, 4.0],
+            }
+        )
+
+        result = df.pivot_table(index="A", values="B", dropna=dropna, observed=False)
+        if dropna:
+            values = [2.0, 3.0]
+            codes = [0, 1]
+        else:
+            # GH: 10772
+            values = [2.0, 3.0, 0.0]
+            codes = [0, 1, -1]
+        expected = DataFrame(
+            {"B": values},
+            index=Index(
+                Categorical.from_codes(codes, categories=["low", "high"], ordered=True),
+                name="A",
+            ),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_with_non_observable_dropna_multi_cat(self, dropna):
+        # gh-21378
+        df = DataFrame(
+            {
+                "A": Categorical(
+                    ["left", "low", "high", "low", "high"],
+                    categories=["low", "high", "left"],
+                    ordered=True,
+                ),
+                "B": range(5),
+            }
+        )
+
+        result = df.pivot_table(index="A", values="B", dropna=dropna, observed=False)
+        expected = DataFrame(
+            {"B": [2.0, 3.0, 0.0]},
+            index=Index(
+                Categorical.from_codes(
+                    [0, 1, 2], categories=["low", "high", "left"], ordered=True
+                ),
+                name="A",
+            ),
+        )
+        if not dropna:
+            expected["B"] = expected["B"].astype(float)
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "left_right", [([0] * 4, [1] * 4), (range(3), range(1, 4))]
+    )
+    def test_pivot_with_interval_index(self, left_right, dropna, closed):
+        # GH 25814
+        left, right = left_right
+        interval_values = Categorical(pd.IntervalIndex.from_arrays(left, right, closed))
+        df = DataFrame({"A": interval_values, "B": 1})
+
+        result = df.pivot_table(index="A", values="B", dropna=dropna, observed=False)
+        expected = DataFrame(
+            {"B": 1.0}, index=Index(interval_values.unique(), name="A")
+        )
+        if not dropna:
+            expected = expected.astype(float)
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_with_interval_index_margins(self):
+        # GH 25815
+        ordered_cat = pd.IntervalIndex.from_arrays([0, 0, 1, 1], [1, 1, 2, 2])
+        df = DataFrame(
+            {
+                "A": np.arange(4, 0, -1, dtype=np.intp),
+                "B": ["a", "b", "a", "b"],
+                "C": Categorical(ordered_cat, ordered=True).sort_values(
+                    ascending=False
+                ),
+            }
+        )
+
+        pivot_tab = pivot_table(
+            df,
+            index="C",
+            columns="B",
+            values="A",
+            aggfunc="sum",
+            margins=True,
+            observed=False,
+        )
+
+        result = pivot_tab["All"]
+        expected = Series(
+            [3, 7, 10],
+            index=Index([pd.Interval(0, 1), pd.Interval(1, 2), "All"], name="C"),
+            name="All",
+            dtype=np.intp,
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_pass_array(self, data):
+        result = data.pivot_table("D", index=data.A, columns=data.C)
+        expected = data.pivot_table("D", index="A", columns="C")
+        tm.assert_frame_equal(result, expected)
+
+    def test_pass_function(self, data):
+        result = data.pivot_table("D", index=lambda x: x // 5, columns=data.C)
+        expected = data.pivot_table("D", index=data.index // 5, columns="C")
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_multiple(self, data):
+        index = ["A", "B"]
+        columns = "C"
+        table = pivot_table(data, index=index, columns=columns)
+        expected = data.groupby(index + [columns]).agg("mean").unstack()
+        tm.assert_frame_equal(table, expected)
+
+    def test_pivot_dtypes(self):
+        # can convert dtypes
+        f = DataFrame(
+            {
+                "a": ["cat", "bat", "cat", "bat"],
+                "v": [1, 2, 3, 4],
+                "i": ["a", "b", "a", "b"],
+            }
+        )
+        assert f.dtypes["v"] == "int64"
+
+        z = pivot_table(
+            f, values="v", index=["a"], columns=["i"], fill_value=0, aggfunc="sum"
+        )
+        result = z.dtypes
+        expected = Series([np.dtype("int64")] * 2, index=Index(list("ab"), name="i"))
+        tm.assert_series_equal(result, expected)
+
+        # cannot convert dtypes
+        f = DataFrame(
+            {
+                "a": ["cat", "bat", "cat", "bat"],
+                "v": [1.5, 2.5, 3.5, 4.5],
+                "i": ["a", "b", "a", "b"],
+            }
+        )
+        assert f.dtypes["v"] == "float64"
+
+        z = pivot_table(
+            f, values="v", index=["a"], columns=["i"], fill_value=0, aggfunc="mean"
+        )
+        result = z.dtypes
+        expected = Series([np.dtype("float64")] * 2, index=Index(list("ab"), name="i"))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "columns,values",
+        [
+            ("bool1", ["float1", "float2"]),
+            ("bool1", ["float1", "float2", "bool1"]),
+            ("bool2", ["float1", "float2", "bool1"]),
+        ],
+    )
+    def test_pivot_preserve_dtypes(self, columns, values):
+        # GH 7142 regression test
+        v = np.arange(5, dtype=np.float64)
+        df = DataFrame(
+            {"float1": v, "float2": v + 2.0, "bool1": v <= 2, "bool2": v <= 3}
+        )
+
+        df_res = df.reset_index().pivot_table(
+            index="index", columns=columns, values=values
+        )
+
+        result = dict(df_res.dtypes)
+        expected = {col: np.dtype("float64") for col in df_res}
+        assert result == expected
+
+    def test_pivot_no_values(self):
+        # GH 14380
+        idx = pd.DatetimeIndex(
+            ["2011-01-01", "2011-02-01", "2011-01-02", "2011-01-01", "2011-01-02"]
+        )
+        df = DataFrame({"A": [1, 2, 3, 4, 5]}, index=idx)
+        res = df.pivot_table(index=df.index.month, columns=df.index.day)
+
+        exp_columns = MultiIndex.from_tuples([("A", 1), ("A", 2)])
+        exp_columns = exp_columns.set_levels(
+            exp_columns.levels[1].astype(np.int32), level=1
+        )
+        exp = DataFrame(
+            [[2.5, 4.0], [2.0, np.nan]],
+            index=Index([1, 2], dtype=np.int32),
+            columns=exp_columns,
+        )
+        tm.assert_frame_equal(res, exp)
+
+        df = DataFrame(
+            {
+                "A": [1, 2, 3, 4, 5],
+                "dt": date_range("2011-01-01", freq="D", periods=5),
+            },
+            index=idx,
+        )
+        res = df.pivot_table(index=df.index.month, columns=Grouper(key="dt", freq="ME"))
+        exp_columns = MultiIndex.from_arrays(
+            [["A"], pd.DatetimeIndex(["2011-01-31"], dtype="M8[ns]")],
+            names=[None, "dt"],
+        )
+        exp = DataFrame(
+            [3.25, 2.0], index=Index([1, 2], dtype=np.int32), columns=exp_columns
+        )
+        tm.assert_frame_equal(res, exp)
+
+        res = df.pivot_table(
+            index=Grouper(freq="YE"), columns=Grouper(key="dt", freq="ME")
+        )
+        exp = DataFrame(
+            [3.0],
+            index=pd.DatetimeIndex(["2011-12-31"], freq="YE"),
+            columns=exp_columns,
+        )
+        tm.assert_frame_equal(res, exp)
+
+    def test_pivot_multi_values(self, data):
+        result = pivot_table(
+            data, values=["D", "E"], index="A", columns=["B", "C"], fill_value=0
+        )
+        expected = pivot_table(
+            data.drop(["F"], axis=1), index="A", columns=["B", "C"], fill_value=0
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_multi_functions(self, data):
+        f = lambda func: pivot_table(
+            data, values=["D", "E"], index=["A", "B"], columns="C", aggfunc=func
+        )
+        result = f(["mean", "std"])
+        means = f("mean")
+        stds = f("std")
+        expected = concat([means, stds], keys=["mean", "std"], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+        # margins not supported??
+        f = lambda func: pivot_table(
+            data,
+            values=["D", "E"],
+            index=["A", "B"],
+            columns="C",
+            aggfunc=func,
+            margins=True,
+        )
+        result = f(["mean", "std"])
+        means = f("mean")
+        stds = f("std")
+        expected = concat([means, stds], keys=["mean", "std"], axis=1)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_index_with_nan(self, method):
+        # GH 3588
+        nan = np.nan
+        df = DataFrame(
+            {
+                "a": ["R1", "R2", nan, "R4"],
+                "b": ["C1", "C2", "C3", "C4"],
+                "c": [10, 15, 17, 20],
+            }
+        )
+        if method:
+            result = df.pivot(index="a", columns="b", values="c")
+        else:
+            result = pd.pivot(df, index="a", columns="b", values="c")
+        expected = DataFrame(
+            [
+                [nan, nan, 17, nan],
+                [10, nan, nan, nan],
+                [nan, 15, nan, nan],
+                [nan, nan, nan, 20],
+            ],
+            index=Index([nan, "R1", "R2", "R4"], name="a"),
+            columns=Index(["C1", "C2", "C3", "C4"], name="b"),
+        )
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(df.pivot(index="b", columns="a", values="c"), expected.T)
+
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_index_with_nan_dates(self, method):
+        # GH9491
+        df = DataFrame(
+            {
+                "a": date_range("2014-02-01", periods=6, freq="D"),
+                "c": 100 + np.arange(6),
+            }
+        )
+        df["b"] = df["a"] - pd.Timestamp("2014-02-02")
+        df.loc[1, "a"] = df.loc[3, "a"] = np.nan
+        df.loc[1, "b"] = df.loc[4, "b"] = np.nan
+
+        if method:
+            pv = df.pivot(index="a", columns="b", values="c")
+        else:
+            pv = pd.pivot(df, index="a", columns="b", values="c")
+        assert pv.notna().values.sum() == len(df)
+
+        for _, row in df.iterrows():
+            assert pv.loc[row["a"], row["b"]] == row["c"]
+
+        if method:
+            result = df.pivot(index="b", columns="a", values="c")
+        else:
+            result = pd.pivot(df, index="b", columns="a", values="c")
+        tm.assert_frame_equal(result, pv.T)
+
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_with_tz(self, method, unit):
+        # GH 5878
+        df = DataFrame(
+            {
+                "dt1": pd.DatetimeIndex(
+                    [
+                        datetime(2013, 1, 1, 9, 0),
+                        datetime(2013, 1, 2, 9, 0),
+                        datetime(2013, 1, 1, 9, 0),
+                        datetime(2013, 1, 2, 9, 0),
+                    ],
+                    dtype=f"M8[{unit}, US/Pacific]",
+                ),
+                "dt2": pd.DatetimeIndex(
+                    [
+                        datetime(2014, 1, 1, 9, 0),
+                        datetime(2014, 1, 1, 9, 0),
+                        datetime(2014, 1, 2, 9, 0),
+                        datetime(2014, 1, 2, 9, 0),
+                    ],
+                    dtype=f"M8[{unit}, Asia/Tokyo]",
+                ),
+                "data1": np.arange(4, dtype="int64"),
+                "data2": np.arange(4, dtype="int64"),
+            }
+        )
+
+        exp_col1 = Index(["data1", "data1", "data2", "data2"])
+        exp_col2 = pd.DatetimeIndex(
+            ["2014/01/01 09:00", "2014/01/02 09:00"] * 2,
+            name="dt2",
+            dtype=f"M8[{unit}, Asia/Tokyo]",
+        )
+        exp_col = MultiIndex.from_arrays([exp_col1, exp_col2])
+        exp_idx = pd.DatetimeIndex(
+            ["2013/01/01 09:00", "2013/01/02 09:00"],
+            name="dt1",
+            dtype=f"M8[{unit}, US/Pacific]",
+        )
+        expected = DataFrame(
+            [[0, 2, 0, 2], [1, 3, 1, 3]],
+            index=exp_idx,
+            columns=exp_col,
+        )
+
+        if method:
+            pv = df.pivot(index="dt1", columns="dt2")
+        else:
+            pv = pd.pivot(df, index="dt1", columns="dt2")
+        tm.assert_frame_equal(pv, expected)
+
+        expected = DataFrame(
+            [[0, 2], [1, 3]],
+            index=exp_idx,
+            columns=exp_col2[:2],
+        )
+
+        if method:
+            pv = df.pivot(index="dt1", columns="dt2", values="data1")
+        else:
+            pv = pd.pivot(df, index="dt1", columns="dt2", values="data1")
+        tm.assert_frame_equal(pv, expected)
+
+    def test_pivot_tz_in_values(self):
+        # GH 14948
+        df = DataFrame(
+            [
+                {
+                    "uid": "aa",
+                    "ts": pd.Timestamp("2016-08-12 13:00:00-0700", tz="US/Pacific"),
+                },
+                {
+                    "uid": "aa",
+                    "ts": pd.Timestamp("2016-08-12 08:00:00-0700", tz="US/Pacific"),
+                },
+                {
+                    "uid": "aa",
+                    "ts": pd.Timestamp("2016-08-12 14:00:00-0700", tz="US/Pacific"),
+                },
+                {
+                    "uid": "aa",
+                    "ts": pd.Timestamp("2016-08-25 11:00:00-0700", tz="US/Pacific"),
+                },
+                {
+                    "uid": "aa",
+                    "ts": pd.Timestamp("2016-08-25 13:00:00-0700", tz="US/Pacific"),
+                },
+            ]
+        )
+
+        df = df.set_index("ts").reset_index()
+        mins = df.ts.map(lambda x: x.replace(hour=0, minute=0, second=0, microsecond=0))
+
+        result = pivot_table(
+            df.set_index("ts").reset_index(),
+            values="ts",
+            index=["uid"],
+            columns=[mins],
+            aggfunc="min",
+        )
+        expected = DataFrame(
+            [
+                [
+                    pd.Timestamp("2016-08-12 08:00:00-0700", tz="US/Pacific"),
+                    pd.Timestamp("2016-08-25 11:00:00-0700", tz="US/Pacific"),
+                ]
+            ],
+            index=Index(["aa"], name="uid"),
+            columns=pd.DatetimeIndex(
+                [
+                    pd.Timestamp("2016-08-12 00:00:00", tz="US/Pacific"),
+                    pd.Timestamp("2016-08-25 00:00:00", tz="US/Pacific"),
+                ],
+                name="ts",
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_periods(self, method):
+        df = DataFrame(
+            {
+                "p1": [
+                    pd.Period("2013-01-01", "D"),
+                    pd.Period("2013-01-02", "D"),
+                    pd.Period("2013-01-01", "D"),
+                    pd.Period("2013-01-02", "D"),
+                ],
+                "p2": [
+                    pd.Period("2013-01", "M"),
+                    pd.Period("2013-01", "M"),
+                    pd.Period("2013-02", "M"),
+                    pd.Period("2013-02", "M"),
+                ],
+                "data1": np.arange(4, dtype="int64"),
+                "data2": np.arange(4, dtype="int64"),
+            }
+        )
+
+        exp_col1 = Index(["data1", "data1", "data2", "data2"])
+        exp_col2 = pd.PeriodIndex(["2013-01", "2013-02"] * 2, name="p2", freq="M")
+        exp_col = MultiIndex.from_arrays([exp_col1, exp_col2])
+        expected = DataFrame(
+            [[0, 2, 0, 2], [1, 3, 1, 3]],
+            index=pd.PeriodIndex(["2013-01-01", "2013-01-02"], name="p1", freq="D"),
+            columns=exp_col,
+        )
+        if method:
+            pv = df.pivot(index="p1", columns="p2")
+        else:
+            pv = pd.pivot(df, index="p1", columns="p2")
+        tm.assert_frame_equal(pv, expected)
+
+        expected = DataFrame(
+            [[0, 2], [1, 3]],
+            index=pd.PeriodIndex(["2013-01-01", "2013-01-02"], name="p1", freq="D"),
+            columns=pd.PeriodIndex(["2013-01", "2013-02"], name="p2", freq="M"),
+        )
+        if method:
+            pv = df.pivot(index="p1", columns="p2", values="data1")
+        else:
+            pv = pd.pivot(df, index="p1", columns="p2", values="data1")
+        tm.assert_frame_equal(pv, expected)
+
+    def test_pivot_periods_with_margins(self):
+        # GH 28323
+        df = DataFrame(
+            {
+                "a": [1, 1, 2, 2],
+                "b": [
+                    pd.Period("2019Q1"),
+                    pd.Period("2019Q2"),
+                    pd.Period("2019Q1"),
+                    pd.Period("2019Q2"),
+                ],
+                "x": 1.0,
+            }
+        )
+
+        expected = DataFrame(
+            data=1.0,
+            index=Index([1, 2, "All"], name="a"),
+            columns=Index([pd.Period("2019Q1"), pd.Period("2019Q2"), "All"], name="b"),
+        )
+
+        result = df.pivot_table(index="a", columns="b", values="x", margins=True)
+        tm.assert_frame_equal(expected, result)
+
+    @pytest.mark.parametrize("box", [list, np.array, Series, Index])
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_with_list_like_values(self, box, method):
+        # issue #17160
+        values = box(["baz", "zoo"])
+        df = DataFrame(
+            {
+                "foo": ["one", "one", "one", "two", "two", "two"],
+                "bar": ["A", "B", "C", "A", "B", "C"],
+                "baz": [1, 2, 3, 4, 5, 6],
+                "zoo": ["x", "y", "z", "q", "w", "t"],
+            }
+        )
+
+        if method:
+            result = df.pivot(index="foo", columns="bar", values=values)
+        else:
+            result = pd.pivot(df, index="foo", columns="bar", values=values)
+
+        data = [[1, 2, 3, "x", "y", "z"], [4, 5, 6, "q", "w", "t"]]
+        index = Index(data=["one", "two"], name="foo")
+        columns = MultiIndex(
+            levels=[["baz", "zoo"], ["A", "B", "C"]],
+            codes=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]],
+            names=[None, "bar"],
+        )
+        expected = DataFrame(data=data, index=index, columns=columns)
+        expected["baz"] = expected["baz"].astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            ["bar", "baz"],
+            np.array(["bar", "baz"]),
+            Series(["bar", "baz"]),
+            Index(["bar", "baz"]),
+        ],
+    )
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_with_list_like_values_nans(self, values, method):
+        # issue #17160
+        df = DataFrame(
+            {
+                "foo": ["one", "one", "one", "two", "two", "two"],
+                "bar": ["A", "B", "C", "A", "B", "C"],
+                "baz": [1, 2, 3, 4, 5, 6],
+                "zoo": ["x", "y", "z", "q", "w", "t"],
+            }
+        )
+
+        if method:
+            result = df.pivot(index="zoo", columns="foo", values=values)
+        else:
+            result = pd.pivot(df, index="zoo", columns="foo", values=values)
+
+        data = [
+            [np.nan, "A", np.nan, 4],
+            [np.nan, "C", np.nan, 6],
+            [np.nan, "B", np.nan, 5],
+            ["A", np.nan, 1, np.nan],
+            ["B", np.nan, 2, np.nan],
+            ["C", np.nan, 3, np.nan],
+        ]
+        index = Index(data=["q", "t", "w", "x", "y", "z"], name="zoo")
+        columns = MultiIndex(
+            levels=[["bar", "baz"], ["one", "two"]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+            names=[None, "foo"],
+        )
+        expected = DataFrame(data=data, index=index, columns=columns)
+        expected["baz"] = expected["baz"].astype(object)
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_columns_none_raise_error(self):
+        # GH 30924
+        df = DataFrame({"col1": ["a", "b", "c"], "col2": [1, 2, 3], "col3": [1, 2, 3]})
+        msg = r"pivot\(\) missing 1 required keyword-only argument: 'columns'"
+        with pytest.raises(TypeError, match=msg):
+            df.pivot(index="col1", values="col3")
+
+    @pytest.mark.xfail(
+        reason="MultiIndexed unstack with tuple names fails with KeyError GH#19966"
+    )
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_with_multiindex(self, method):
+        # issue #17160
+        index = Index(data=[0, 1, 2, 3, 4, 5])
+        data = [
+            ["one", "A", 1, "x"],
+            ["one", "B", 2, "y"],
+            ["one", "C", 3, "z"],
+            ["two", "A", 4, "q"],
+            ["two", "B", 5, "w"],
+            ["two", "C", 6, "t"],
+        ]
+        columns = MultiIndex(
+            levels=[["bar", "baz"], ["first", "second"]],
+            codes=[[0, 0, 1, 1], [0, 1, 0, 1]],
+        )
+        df = DataFrame(data=data, index=index, columns=columns, dtype="object")
+        if method:
+            result = df.pivot(
+                index=("bar", "first"),
+                columns=("bar", "second"),
+                values=("baz", "first"),
+            )
+        else:
+            result = pd.pivot(
+                df,
+                index=("bar", "first"),
+                columns=("bar", "second"),
+                values=("baz", "first"),
+            )
+
+        data = {
+            "A": Series([1, 4], index=["one", "two"]),
+            "B": Series([2, 5], index=["one", "two"]),
+            "C": Series([3, 6], index=["one", "two"]),
+        }
+        expected = DataFrame(data)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("method", [True, False])
+    def test_pivot_with_tuple_of_values(self, method):
+        # issue #17160
+        df = DataFrame(
+            {
+                "foo": ["one", "one", "one", "two", "two", "two"],
+                "bar": ["A", "B", "C", "A", "B", "C"],
+                "baz": [1, 2, 3, 4, 5, 6],
+                "zoo": ["x", "y", "z", "q", "w", "t"],
+            }
+        )
+        with pytest.raises(KeyError, match=r"^\('bar', 'baz'\)$"):
+            # tuple is seen as a single column name
+            if method:
+                df.pivot(index="zoo", columns="foo", values=("bar", "baz"))
+            else:
+                pd.pivot(df, index="zoo", columns="foo", values=("bar", "baz"))
+
+    def _check_output(
+        self,
+        result,
+        values_col,
+        data,
+        index=None,
+        columns=None,
+        margins_col="All",
+    ):
+        if index is None:
+            index = ["A", "B"]
+        if columns is None:
+            columns = ["C"]
+        col_margins = result.loc[result.index[:-1], margins_col]
+        expected_col_margins = data.groupby(index)[values_col].mean()
+        tm.assert_series_equal(col_margins, expected_col_margins, check_names=False)
+        assert col_margins.name == margins_col
+
+        result = result.sort_index()
+        index_margins = result.loc[(margins_col, "")].iloc[:-1]
+
+        expected_ix_margins = data.groupby(columns)[values_col].mean()
+        tm.assert_series_equal(index_margins, expected_ix_margins, check_names=False)
+        assert index_margins.name == (margins_col, "")
+
+        grand_total_margins = result.loc[(margins_col, ""), margins_col]
+        expected_total_margins = data[values_col].mean()
+        assert grand_total_margins == expected_total_margins
+
+    def test_margins(self, data):
+        # column specified
+        result = data.pivot_table(
+            values="D", index=["A", "B"], columns="C", margins=True, aggfunc="mean"
+        )
+        self._check_output(result, "D", data)
+
+        # Set a different margins_name (not 'All')
+        result = data.pivot_table(
+            values="D",
+            index=["A", "B"],
+            columns="C",
+            margins=True,
+            aggfunc="mean",
+            margins_name="Totals",
+        )
+        self._check_output(result, "D", data, margins_col="Totals")
+
+        # no column specified
+        table = data.pivot_table(
+            index=["A", "B"], columns="C", margins=True, aggfunc="mean"
+        )
+        for value_col in table.columns.levels[0]:
+            self._check_output(table[value_col], value_col, data)
+
+    def test_no_col(self, data, using_infer_string):
+        # no col
+
+        # to help with a buglet
+        data.columns = [k * 2 for k in data.columns]
+        msg = re.escape("agg function failed [how->mean,dtype->")
+        if using_infer_string:
+            msg = "dtype 'str' does not support operation 'mean'"
+        with pytest.raises(TypeError, match=msg):
+            data.pivot_table(index=["AA", "BB"], margins=True, aggfunc="mean")
+        table = data.drop(columns="CC").pivot_table(
+            index=["AA", "BB"], margins=True, aggfunc="mean"
+        )
+        for value_col in table.columns:
+            totals = table.loc[("All", ""), value_col]
+            assert totals == data[value_col].mean()
+
+        with pytest.raises(TypeError, match=msg):
+            data.pivot_table(index=["AA", "BB"], margins=True, aggfunc="mean")
+        table = data.drop(columns="CC").pivot_table(
+            index=["AA", "BB"], margins=True, aggfunc="mean"
+        )
+        for item in ["DD", "EE", "FF"]:
+            totals = table.loc[("All", ""), item]
+            assert totals == data[item].mean()
+
+    @pytest.mark.parametrize(
+        "columns, aggfunc, values, expected_columns",
+        [
+            (
+                "A",
+                "mean",
+                [[5.5, 5.5, 2.2, 2.2], [8.0, 8.0, 4.4, 4.4]],
+                Index(["bar", "All", "foo", "All"], name="A"),
+            ),
+            (
+                ["A", "B"],
+                "sum",
+                [
+                    [9, 13, 22, 5, 6, 11],
+                    [14, 18, 32, 11, 11, 22],
+                ],
+                MultiIndex.from_tuples(
+                    [
+                        ("bar", "one"),
+                        ("bar", "two"),
+                        ("bar", "All"),
+                        ("foo", "one"),
+                        ("foo", "two"),
+                        ("foo", "All"),
+                    ],
+                    names=["A", "B"],
+                ),
+            ),
+        ],
+    )
+    def test_margin_with_only_columns_defined(
+        self, columns, aggfunc, values, expected_columns, using_infer_string
+    ):
+        # GH 31016
+        df = DataFrame(
+            {
+                "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+                "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+                "C": [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+                "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+                "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+            }
+        )
+        if aggfunc != "sum":
+            msg = re.escape("agg function failed [how->mean,dtype->")
+            if using_infer_string:
+                msg = "dtype 'str' does not support operation 'mean'"
+            with pytest.raises(TypeError, match=msg):
+                df.pivot_table(columns=columns, margins=True, aggfunc=aggfunc)
+        if "B" not in columns:
+            df = df.drop(columns="B")
+        result = df.drop(columns="C").pivot_table(
+            columns=columns, margins=True, aggfunc=aggfunc
+        )
+        expected = DataFrame(values, index=Index(["D", "E"]), columns=expected_columns)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_margins_dtype(self, data):
+        # GH 17013
+
+        df = data.copy()
+        df[["D", "E", "F"]] = np.arange(len(df) * 3).reshape(len(df), 3).astype("i8")
+
+        mi_val = list(product(["bar", "foo"], ["one", "two"])) + [("All", "")]
+        mi = MultiIndex.from_tuples(mi_val, names=("A", "B"))
+        expected = DataFrame(
+            {"dull": [12, 21, 3, 9, 45], "shiny": [33, 0, 36, 51, 120]}, index=mi
+        ).rename_axis("C", axis=1)
+        expected["All"] = expected["dull"] + expected["shiny"]
+
+        result = df.pivot_table(
+            values="D",
+            index=["A", "B"],
+            columns="C",
+            margins=True,
+            aggfunc="sum",
+            fill_value=0,
+        )
+
+        tm.assert_frame_equal(expected, result)
+
+    def test_margins_dtype_len(self, data):
+        mi_val = list(product(["bar", "foo"], ["one", "two"])) + [("All", "")]
+        mi = MultiIndex.from_tuples(mi_val, names=("A", "B"))
+        expected = DataFrame(
+            {"dull": [1, 1, 2, 1, 5], "shiny": [2, 0, 2, 2, 6]}, index=mi
+        ).rename_axis("C", axis=1)
+        expected["All"] = expected["dull"] + expected["shiny"]
+
+        result = data.pivot_table(
+            values="D",
+            index=["A", "B"],
+            columns="C",
+            margins=True,
+            aggfunc=len,
+            fill_value=0,
+        )
+
+        tm.assert_frame_equal(expected, result)
+
+    @pytest.mark.parametrize("cols", [(1, 2), ("a", "b"), (1, "b"), ("a", 1)])
+    def test_pivot_table_multiindex_only(self, cols):
+        # GH 17038
+        df2 = DataFrame({cols[0]: [1, 2, 3], cols[1]: [1, 2, 3], "v": [4, 5, 6]})
+
+        result = df2.pivot_table(values="v", columns=cols)
+        expected = DataFrame(
+            [[4.0, 5.0, 6.0]],
+            columns=MultiIndex.from_tuples([(1, 1), (2, 2), (3, 3)], names=cols),
+            index=Index(["v"], dtype="str" if cols == ("a", "b") else "object"),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_retains_tz(self):
+        dti = date_range("2016-01-01", periods=3, tz="Europe/Amsterdam")
+        df = DataFrame(
+            {
+                "A": np.random.default_rng(2).standard_normal(3),
+                "B": np.random.default_rng(2).standard_normal(3),
+                "C": dti,
+            }
+        )
+        result = df.pivot_table(index=["B", "C"], dropna=False)
+
+        # check tz retention
+        assert result.index.levels[1].equals(dti)
+
+    def test_pivot_integer_columns(self):
+        # caused by upstream bug in unstack
+
+        d = date.min
+        data = list(
+            product(
+                ["foo", "bar"],
+                ["A", "B", "C"],
+                ["x1", "x2"],
+                [d + timedelta(i) for i in range(20)],
+                [1.0],
+            )
+        )
+        df = DataFrame(data)
+        table = df.pivot_table(values=4, index=[0, 1, 3], columns=[2])
+
+        df2 = df.rename(columns=str)
+        table2 = df2.pivot_table(values="4", index=["0", "1", "3"], columns=["2"])
+
+        tm.assert_frame_equal(table, table2, check_names=False)
+
+    def test_pivot_no_level_overlap(self):
+        # GH #1181
+
+        data = DataFrame(
+            {
+                "a": ["a", "a", "a", "a", "b", "b", "b", "b"] * 2,
+                "b": [0, 0, 0, 0, 1, 1, 1, 1] * 2,
+                "c": (["foo"] * 4 + ["bar"] * 4) * 2,
+                "value": np.random.default_rng(2).standard_normal(16),
+            }
+        )
+
+        table = data.pivot_table("value", index="a", columns=["b", "c"])
+
+        grouped = data.groupby(["a", "b", "c"])["value"].mean()
+        expected = grouped.unstack("b").unstack("c").dropna(axis=1, how="all")
+        tm.assert_frame_equal(table, expected)
+
+    def test_pivot_columns_lexsorted(self):
+        n = 10000
+
+        dtype = np.dtype(
+            [
+                ("Index", object),
+                ("Symbol", object),
+                ("Year", int),
+                ("Month", int),
+                ("Day", int),
+                ("Quantity", int),
+                ("Price", float),
+            ]
+        )
+
+        products = np.array(
+            [
+                ("SP500", "ADBE"),
+                ("SP500", "NVDA"),
+                ("SP500", "ORCL"),
+                ("NDQ100", "AAPL"),
+                ("NDQ100", "MSFT"),
+                ("NDQ100", "GOOG"),
+                ("FTSE", "DGE.L"),
+                ("FTSE", "TSCO.L"),
+                ("FTSE", "GSK.L"),
+            ],
+            dtype=[("Index", object), ("Symbol", object)],
+        )
+        items = np.empty(n, dtype=dtype)
+        iproduct = np.random.default_rng(2).integers(0, len(products), n)
+        items["Index"] = products["Index"][iproduct]
+        items["Symbol"] = products["Symbol"][iproduct]
+        dr = date_range(date(2000, 1, 1), date(2010, 12, 31))
+        dates = dr[np.random.default_rng(2).integers(0, len(dr), n)]
+        items["Year"] = dates.year
+        items["Month"] = dates.month
+        items["Day"] = dates.day
+        items["Price"] = np.random.default_rng(2).lognormal(4.0, 2.0, n)
+
+        df = DataFrame(items)
+
+        pivoted = df.pivot_table(
+            "Price",
+            index=["Month", "Day"],
+            columns=["Index", "Symbol", "Year"],
+            aggfunc="mean",
+        )
+
+        assert pivoted.columns.is_monotonic_increasing
+
+    def test_pivot_complex_aggfunc(self, data):
+        f = {"D": ["std"], "E": ["sum"]}
+        expected = data.groupby(["A", "B"]).agg(f).unstack("B")
+        result = data.pivot_table(index="A", columns="B", aggfunc=f)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_margins_no_values_no_cols(self, data):
+        # Regression test on pivot table: no values or cols passed.
+        result = data[["A", "B"]].pivot_table(
+            index=["A", "B"], aggfunc=len, margins=True
+        )
+        result_list = result.tolist()
+        assert sum(result_list[:-1]) == result_list[-1]
+
+    def test_margins_no_values_two_rows(self, data):
+        # Regression test on pivot table: no values passed but rows are a
+        # multi-index
+        result = data[["A", "B", "C"]].pivot_table(
+            index=["A", "B"], columns="C", aggfunc=len, margins=True
+        )
+        assert result.All.tolist() == [3.0, 1.0, 4.0, 3.0, 11.0]
+
+    def test_margins_no_values_one_row_one_col(self, data):
+        # Regression test on pivot table: no values passed but row and col
+        # defined
+        result = data[["A", "B"]].pivot_table(
+            index="A", columns="B", aggfunc=len, margins=True
+        )
+        assert result.All.tolist() == [4.0, 7.0, 11.0]
+
+    def test_margins_no_values_two_row_two_cols(self, data):
+        # Regression test on pivot table: no values passed but rows and cols
+        # are multi-indexed
+        data["D"] = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k"]
+        result = data[["A", "B", "C", "D"]].pivot_table(
+            index=["A", "B"], columns=["C", "D"], aggfunc=len, margins=True
+        )
+        assert result.All.tolist() == [3.0, 1.0, 4.0, 3.0, 11.0]
+
+    @pytest.mark.parametrize("margin_name", ["foo", "one", 666, None, ["a", "b"]])
+    def test_pivot_table_with_margins_set_margin_name(self, margin_name, data):
+        # see gh-3335
+        msg = (
+            f'Conflicting name "{margin_name}" in margins|'
+            "margins_name argument must be a string"
+        )
+        with pytest.raises(ValueError, match=msg):
+            # multi-index index
+            pivot_table(
+                data,
+                values="D",
+                index=["A", "B"],
+                columns=["C"],
+                margins=True,
+                margins_name=margin_name,
+            )
+        with pytest.raises(ValueError, match=msg):
+            # multi-index column
+            pivot_table(
+                data,
+                values="D",
+                index=["C"],
+                columns=["A", "B"],
+                margins=True,
+                margins_name=margin_name,
+            )
+        with pytest.raises(ValueError, match=msg):
+            # non-multi-index index/column
+            pivot_table(
+                data,
+                values="D",
+                index=["A"],
+                columns=["B"],
+                margins=True,
+                margins_name=margin_name,
+            )
+
+    def test_pivot_timegrouper(self):
+        df = DataFrame(
+            {
+                "Branch": "A A A A A A A B".split(),
+                "Buyer": "Carl Mark Carl Carl Joe Joe Joe Carl".split(),
+                "Quantity": [1, 3, 5, 1, 8, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 1, 1),
+                    datetime(2013, 1, 1),
+                    datetime(2013, 10, 1),
+                    datetime(2013, 10, 2),
+                    datetime(2013, 10, 1),
+                    datetime(2013, 10, 2),
+                    datetime(2013, 12, 2),
+                    datetime(2013, 12, 2),
+                ],
+            }
+        ).set_index("Date")
+
+        expected = DataFrame(
+            np.array([10, 18, 3], dtype="int64").reshape(1, 3),
+            index=pd.DatetimeIndex([datetime(2013, 12, 31)], freq="YE"),
+            columns="Carl Joe Mark".split(),
+        )
+        expected.index.name = "Date"
+        expected.columns.name = "Buyer"
+
+        result = pivot_table(
+            df,
+            index=Grouper(freq="YE"),
+            columns="Buyer",
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index="Buyer",
+            columns=Grouper(freq="YE"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+        expected = DataFrame(
+            np.array([1, np.nan, 3, 9, 18, np.nan]).reshape(2, 3),
+            index=pd.DatetimeIndex(
+                [datetime(2013, 1, 1), datetime(2013, 7, 1)], freq="6MS"
+            ),
+            columns="Carl Joe Mark".split(),
+        )
+        expected.index.name = "Date"
+        expected.columns.name = "Buyer"
+
+        result = pivot_table(
+            df,
+            index=Grouper(freq="6MS"),
+            columns="Buyer",
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index="Buyer",
+            columns=Grouper(freq="6MS"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+        # passing the name
+        df = df.reset_index()
+        result = pivot_table(
+            df,
+            index=Grouper(freq="6MS", key="Date"),
+            columns="Buyer",
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index="Buyer",
+            columns=Grouper(freq="6MS", key="Date"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+        msg = "'The grouper name foo is not found'"
+        with pytest.raises(KeyError, match=msg):
+            pivot_table(
+                df,
+                index=Grouper(freq="6MS", key="foo"),
+                columns="Buyer",
+                values="Quantity",
+                aggfunc="sum",
+            )
+        with pytest.raises(KeyError, match=msg):
+            pivot_table(
+                df,
+                index="Buyer",
+                columns=Grouper(freq="6MS", key="foo"),
+                values="Quantity",
+                aggfunc="sum",
+            )
+
+        # passing the level
+        df = df.set_index("Date")
+        result = pivot_table(
+            df,
+            index=Grouper(freq="6MS", level="Date"),
+            columns="Buyer",
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index="Buyer",
+            columns=Grouper(freq="6MS", level="Date"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+        msg = "The level foo is not valid"
+        with pytest.raises(ValueError, match=msg):
+            pivot_table(
+                df,
+                index=Grouper(freq="6MS", level="foo"),
+                columns="Buyer",
+                values="Quantity",
+                aggfunc="sum",
+            )
+        with pytest.raises(ValueError, match=msg):
+            pivot_table(
+                df,
+                index="Buyer",
+                columns=Grouper(freq="6MS", level="foo"),
+                values="Quantity",
+                aggfunc="sum",
+            )
+
+    def test_pivot_timegrouper_double(self):
+        # double grouper
+        df = DataFrame(
+            {
+                "Branch": "A A A A A A A B".split(),
+                "Buyer": "Carl Mark Carl Carl Joe Joe Joe Carl".split(),
+                "Quantity": [1, 3, 5, 1, 8, 1, 9, 3],
+                "Date": [
+                    datetime(2013, 11, 1, 13, 0),
+                    datetime(2013, 9, 1, 13, 5),
+                    datetime(2013, 10, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 11, 1, 20, 0),
+                    datetime(2013, 10, 2, 10, 0),
+                    datetime(2013, 10, 2, 12, 0),
+                    datetime(2013, 12, 5, 14, 0),
+                ],
+                "PayDay": [
+                    datetime(2013, 10, 4, 0, 0),
+                    datetime(2013, 10, 15, 13, 5),
+                    datetime(2013, 9, 5, 20, 0),
+                    datetime(2013, 11, 2, 10, 0),
+                    datetime(2013, 10, 7, 20, 0),
+                    datetime(2013, 9, 5, 10, 0),
+                    datetime(2013, 12, 30, 12, 0),
+                    datetime(2013, 11, 20, 14, 0),
+                ],
+            }
+        )
+
+        result = pivot_table(
+            df,
+            index=Grouper(freq="ME", key="Date"),
+            columns=Grouper(freq="ME", key="PayDay"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        expected = DataFrame(
+            np.array(
+                [
+                    np.nan,
+                    3,
+                    np.nan,
+                    np.nan,
+                    6,
+                    np.nan,
+                    1,
+                    9,
+                    np.nan,
+                    9,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    np.nan,
+                    3,
+                    np.nan,
+                ]
+            ).reshape(4, 4),
+            index=pd.DatetimeIndex(
+                [
+                    datetime(2013, 9, 30),
+                    datetime(2013, 10, 31),
+                    datetime(2013, 11, 30),
+                    datetime(2013, 12, 31),
+                ],
+                freq="ME",
+            ),
+            columns=pd.DatetimeIndex(
+                [
+                    datetime(2013, 9, 30),
+                    datetime(2013, 10, 31),
+                    datetime(2013, 11, 30),
+                    datetime(2013, 12, 31),
+                ],
+                freq="ME",
+            ),
+        )
+        expected.index.name = "Date"
+        expected.columns.name = "PayDay"
+
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index=Grouper(freq="ME", key="PayDay"),
+            columns=Grouper(freq="ME", key="Date"),
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+        tuples = [
+            (datetime(2013, 9, 30), datetime(2013, 10, 31)),
+            (datetime(2013, 10, 31), datetime(2013, 9, 30)),
+            (datetime(2013, 10, 31), datetime(2013, 11, 30)),
+            (datetime(2013, 10, 31), datetime(2013, 12, 31)),
+            (datetime(2013, 11, 30), datetime(2013, 10, 31)),
+            (datetime(2013, 12, 31), datetime(2013, 11, 30)),
+        ]
+        idx = MultiIndex.from_tuples(tuples, names=["Date", "PayDay"])
+        expected = DataFrame(
+            np.array(
+                [3, np.nan, 6, np.nan, 1, np.nan, 9, np.nan, 9, np.nan, np.nan, 3]
+            ).reshape(6, 2),
+            index=idx,
+            columns=["A", "B"],
+        )
+        expected.columns.name = "Branch"
+
+        result = pivot_table(
+            df,
+            index=[Grouper(freq="ME", key="Date"), Grouper(freq="ME", key="PayDay")],
+            columns=["Branch"],
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index=["Branch"],
+            columns=[Grouper(freq="ME", key="Date"), Grouper(freq="ME", key="PayDay")],
+            values="Quantity",
+            aggfunc="sum",
+        )
+        tm.assert_frame_equal(result, expected.T)
+
+    def test_pivot_datetime_tz(self):
+        dates1 = pd.DatetimeIndex(
+            [
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+            ],
+            dtype="M8[ns, US/Pacific]",
+            name="dt1",
+        )
+        dates2 = pd.DatetimeIndex(
+            [
+                "2013-01-01 15:00:00",
+                "2013-01-01 15:00:00",
+                "2013-01-01 15:00:00",
+                "2013-02-01 15:00:00",
+                "2013-02-01 15:00:00",
+                "2013-02-01 15:00:00",
+            ],
+            dtype="M8[ns, Asia/Tokyo]",
+        )
+        df = DataFrame(
+            {
+                "label": ["a", "a", "a", "b", "b", "b"],
+                "dt1": dates1,
+                "dt2": dates2,
+                "value1": np.arange(6, dtype="int64"),
+                "value2": [1, 2] * 3,
+            }
+        )
+
+        exp_idx = dates1[:3]
+        exp_col1 = Index(["value1", "value1"])
+        exp_col2 = Index(["a", "b"], name="label")
+        exp_col = MultiIndex.from_arrays([exp_col1, exp_col2])
+        expected = DataFrame(
+            [[0.0, 3.0], [1.0, 4.0], [2.0, 5.0]], index=exp_idx, columns=exp_col
+        )
+        result = pivot_table(df, index=["dt1"], columns=["label"], values=["value1"])
+        tm.assert_frame_equal(result, expected)
+
+        exp_col1 = Index(["sum", "sum", "sum", "sum", "mean", "mean", "mean", "mean"])
+        exp_col2 = Index(["value1", "value1", "value2", "value2"] * 2)
+        exp_col3 = pd.DatetimeIndex(
+            ["2013-01-01 15:00:00", "2013-02-01 15:00:00"] * 4,
+            dtype="M8[ns, Asia/Tokyo]",
+            name="dt2",
+        )
+        exp_col = MultiIndex.from_arrays([exp_col1, exp_col2, exp_col3])
+        expected1 = DataFrame(
+            np.array(
+                [
+                    [
+                        0,
+                        3,
+                        1,
+                        2,
+                    ],
+                    [1, 4, 2, 1],
+                    [2, 5, 1, 2],
+                ],
+                dtype="int64",
+            ),
+            index=exp_idx,
+            columns=exp_col[:4],
+        )
+        expected2 = DataFrame(
+            np.array(
+                [
+                    [0.0, 3.0, 1.0, 2.0],
+                    [1.0, 4.0, 2.0, 1.0],
+                    [2.0, 5.0, 1.0, 2.0],
+                ],
+            ),
+            index=exp_idx,
+            columns=exp_col[4:],
+        )
+        expected = concat([expected1, expected2], axis=1)
+
+        result = pivot_table(
+            df,
+            index=["dt1"],
+            columns=["dt2"],
+            values=["value1", "value2"],
+            aggfunc=["sum", "mean"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_dtaccessor(self):
+        # GH 8103
+        dates1 = pd.DatetimeIndex(
+            [
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+                "2011-07-19 07:00:00",
+                "2011-07-19 08:00:00",
+                "2011-07-19 09:00:00",
+            ]
+        )
+        dates2 = pd.DatetimeIndex(
+            [
+                "2013-01-01 15:00:00",
+                "2013-01-01 15:00:00",
+                "2013-01-01 15:00:00",
+                "2013-02-01 15:00:00",
+                "2013-02-01 15:00:00",
+                "2013-02-01 15:00:00",
+            ]
+        )
+        df = DataFrame(
+            {
+                "label": ["a", "a", "a", "b", "b", "b"],
+                "dt1": dates1,
+                "dt2": dates2,
+                "value1": np.arange(6, dtype="int64"),
+                "value2": [1, 2] * 3,
+            }
+        )
+
+        result = pivot_table(
+            df, index="label", columns=df["dt1"].dt.hour, values="value1"
+        )
+
+        exp_idx = Index(["a", "b"], name="label")
+        expected = DataFrame(
+            {7: [0.0, 3.0], 8: [1.0, 4.0], 9: [2.0, 5.0]},
+            index=exp_idx,
+            columns=Index([7, 8, 9], dtype=np.int32, name="dt1"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df, index=df["dt2"].dt.month, columns=df["dt1"].dt.hour, values="value1"
+        )
+
+        expected = DataFrame(
+            {7: [0.0, 3.0], 8: [1.0, 4.0], 9: [2.0, 5.0]},
+            index=Index([1, 2], dtype=np.int32, name="dt2"),
+            columns=Index([7, 8, 9], dtype=np.int32, name="dt1"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index=df["dt2"].dt.year.values,
+            columns=[df["dt1"].dt.hour, df["dt2"].dt.month],
+            values="value1",
+        )
+
+        exp_col = MultiIndex.from_arrays(
+            [
+                np.array([7, 7, 8, 8, 9, 9], dtype=np.int32),
+                np.array([1, 2] * 3, dtype=np.int32),
+            ],
+            names=["dt1", "dt2"],
+        )
+        expected = DataFrame(
+            np.array([[0.0, 3.0, 1.0, 4.0, 2.0, 5.0]]),
+            index=Index([2013], dtype=np.int32),
+            columns=exp_col,
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(
+            df,
+            index=np.array(["X", "X", "X", "X", "Y", "Y"]),
+            columns=[df["dt1"].dt.hour, df["dt2"].dt.month],
+            values="value1",
+        )
+        expected = DataFrame(
+            np.array(
+                [[0, 3, 1, np.nan, 2, np.nan], [np.nan, np.nan, np.nan, 4, np.nan, 5]]
+            ),
+            index=["X", "Y"],
+            columns=exp_col,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_daily(self):
+        rng = date_range("1/1/2000", "12/31/2004", freq="D")
+        ts = Series(np.arange(len(rng)), index=rng)
+
+        result = pivot_table(
+            DataFrame(ts), index=ts.index.year, columns=ts.index.dayofyear
+        )
+        result.columns = result.columns.droplevel(0)
+
+        doy = np.asarray(ts.index.dayofyear)
+
+        expected = {}
+        for y in ts.index.year.unique().values:
+            mask = ts.index.year == y
+            expected[y] = Series(ts.values[mask], index=doy[mask])
+        expected = DataFrame(expected, dtype=float).T
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(result, expected)
+
+    def test_monthly(self):
+        rng = date_range("1/1/2000", "12/31/2004", freq="ME")
+        ts = Series(np.arange(len(rng)), index=rng)
+
+        result = pivot_table(DataFrame(ts), index=ts.index.year, columns=ts.index.month)
+        result.columns = result.columns.droplevel(0)
+
+        month = np.asarray(ts.index.month)
+        expected = {}
+        for y in ts.index.year.unique().values:
+            mask = ts.index.year == y
+            expected[y] = Series(ts.values[mask], index=month[mask])
+        expected = DataFrame(expected, dtype=float).T
+        expected.index = expected.index.astype(np.int32)
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_with_iterator_values(self, data):
+        # GH 12017
+        aggs = {"D": "sum", "E": "mean"}
+
+        pivot_values_list = pivot_table(
+            data, index=["A"], values=list(aggs.keys()), aggfunc=aggs
+        )
+
+        pivot_values_keys = pivot_table(
+            data, index=["A"], values=aggs.keys(), aggfunc=aggs
+        )
+        tm.assert_frame_equal(pivot_values_keys, pivot_values_list)
+
+        agg_values_gen = (value for value in aggs)
+        pivot_values_gen = pivot_table(
+            data, index=["A"], values=agg_values_gen, aggfunc=aggs
+        )
+        tm.assert_frame_equal(pivot_values_gen, pivot_values_list)
+
+    def test_pivot_table_margins_name_with_aggfunc_list(self):
+        # GH 13354
+        margins_name = "Weekly"
+        costs = DataFrame(
+            {
+                "item": ["bacon", "cheese", "bacon", "cheese"],
+                "cost": [2.5, 4.5, 3.2, 3.3],
+                "day": ["ME", "ME", "T", "T"],
+            }
+        )
+        table = costs.pivot_table(
+            index="item",
+            columns="day",
+            margins=True,
+            margins_name=margins_name,
+            aggfunc=["mean", "max"],
+        )
+        ix = Index(["bacon", "cheese", margins_name], name="item")
+        tups = [
+            ("mean", "cost", "ME"),
+            ("mean", "cost", "T"),
+            ("mean", "cost", margins_name),
+            ("max", "cost", "ME"),
+            ("max", "cost", "T"),
+            ("max", "cost", margins_name),
+        ]
+        cols = MultiIndex.from_tuples(tups, names=[None, None, "day"])
+        expected = DataFrame(table.values, index=ix, columns=cols)
+        tm.assert_frame_equal(table, expected)
+
+    def test_categorical_margins(self, observed):
+        # GH 10989
+        df = DataFrame(
+            {"x": np.arange(8), "y": np.arange(8) // 4, "z": np.arange(8) % 2}
+        )
+
+        expected = DataFrame([[1.0, 2.0, 1.5], [5, 6, 5.5], [3, 4, 3.5]])
+        expected.index = Index([0, 1, "All"], name="y")
+        expected.columns = Index([0, 1, "All"], name="z")
+
+        table = df.pivot_table("x", "y", "z", dropna=observed, margins=True)
+        tm.assert_frame_equal(table, expected)
+
+    def test_categorical_margins_category(self, observed):
+        df = DataFrame(
+            {"x": np.arange(8), "y": np.arange(8) // 4, "z": np.arange(8) % 2}
+        )
+
+        expected = DataFrame([[1.0, 2.0, 1.5], [5, 6, 5.5], [3, 4, 3.5]])
+        expected.index = Index([0, 1, "All"], name="y")
+        expected.columns = Index([0, 1, "All"], name="z")
+
+        df.y = df.y.astype("category")
+        df.z = df.z.astype("category")
+        table = df.pivot_table(
+            "x", "y", "z", dropna=observed, margins=True, observed=False
+        )
+        tm.assert_frame_equal(table, expected)
+
+    def test_margins_casted_to_float(self):
+        # GH 24893
+        df = DataFrame(
+            {
+                "A": [2, 4, 6, 8],
+                "B": [1, 4, 5, 8],
+                "C": [1, 3, 4, 6],
+                "D": ["X", "X", "Y", "Y"],
+            }
+        )
+
+        result = pivot_table(df, index="D", margins=True)
+        expected = DataFrame(
+            {"A": [3.0, 7.0, 5], "B": [2.5, 6.5, 4.5], "C": [2.0, 5.0, 3.5]},
+            index=Index(["X", "Y", "All"], name="D"),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_with_categorical(self, observed, ordered):
+        # gh-21370
+        idx = [np.nan, "low", "high", "low", np.nan]
+        col = [np.nan, "A", "B", np.nan, "A"]
+        df = DataFrame(
+            {
+                "In": Categorical(idx, categories=["low", "high"], ordered=ordered),
+                "Col": Categorical(col, categories=["A", "B"], ordered=ordered),
+                "Val": range(1, 6),
+            }
+        )
+        # case with index/columns/value
+        result = df.pivot_table(
+            index="In", columns="Col", values="Val", observed=observed
+        )
+
+        expected_cols = pd.CategoricalIndex(["A", "B"], ordered=ordered, name="Col")
+
+        expected = DataFrame(data=[[2.0, np.nan], [np.nan, 3.0]], columns=expected_cols)
+        expected.index = Index(
+            Categorical(["low", "high"], categories=["low", "high"], ordered=ordered),
+            name="In",
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        # case with columns/value
+        result = df.pivot_table(columns="Col", values="Val", observed=observed)
+
+        expected = DataFrame(
+            data=[[3.5, 3.0]], columns=expected_cols, index=Index(["Val"])
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_categorical_aggfunc(self, observed):
+        # GH 9534
+        df = DataFrame(
+            {"C1": ["A", "B", "C", "C"], "C2": ["a", "a", "b", "b"], "V": [1, 2, 3, 4]}
+        )
+        df["C1"] = df["C1"].astype("category")
+        result = df.pivot_table(
+            "V",
+            index="C1",
+            columns="C2",
+            dropna=observed,
+            aggfunc="count",
+            observed=False,
+        )
+
+        expected_index = pd.CategoricalIndex(
+            ["A", "B", "C"], categories=["A", "B", "C"], ordered=False, name="C1"
+        )
+        expected_columns = Index(["a", "b"], name="C2")
+        expected_data = np.array([[1, 0], [1, 0], [0, 2]], dtype=np.int64)
+        expected = DataFrame(
+            expected_data, index=expected_index, columns=expected_columns
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_categorical_pivot_index_ordering(self, observed):
+        # GH 8731
+        df = DataFrame(
+            {
+                "Sales": [100, 120, 220],
+                "Month": ["January", "January", "January"],
+                "Year": [2013, 2014, 2013],
+            }
+        )
+        months = [
+            "January",
+            "February",
+            "March",
+            "April",
+            "May",
+            "June",
+            "July",
+            "August",
+            "September",
+            "October",
+            "November",
+            "December",
+        ]
+        df["Month"] = df["Month"].astype("category").cat.set_categories(months)
+        result = df.pivot_table(
+            values="Sales",
+            index="Month",
+            columns="Year",
+            observed=observed,
+            aggfunc="sum",
+        )
+        expected_columns = Index([2013, 2014], name="Year", dtype="int64")
+        expected_index = pd.CategoricalIndex(
+            months, categories=months, ordered=False, name="Month"
+        )
+        expected_data = [[320, 120]] + [[0, 0]] * 11
+        expected = DataFrame(
+            expected_data, index=expected_index, columns=expected_columns
+        )
+        if observed:
+            expected = expected.loc[["January"]]
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_not_series(self):
+        # GH 4386
+        # pivot_table always returns a DataFrame
+        # when values is not list like and columns is None
+        # and aggfunc is not instance of list
+        df = DataFrame({"col1": [3, 4, 5], "col2": ["C", "D", "E"], "col3": [1, 3, 9]})
+
+        result = df.pivot_table("col1", index=["col3", "col2"], aggfunc="sum")
+        m = MultiIndex.from_arrays([[1, 3, 9], ["C", "D", "E"]], names=["col3", "col2"])
+        expected = DataFrame([3, 4, 5], index=m, columns=["col1"])
+
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot_table("col1", index="col3", columns="col2", aggfunc="sum")
+        expected = DataFrame(
+            [[3, np.nan, np.nan], [np.nan, 4, np.nan], [np.nan, np.nan, 5]],
+            index=Index([1, 3, 9], name="col3"),
+            columns=Index(["C", "D", "E"], name="col2"),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot_table("col1", index="col3", aggfunc=["sum"])
+        m = MultiIndex.from_arrays([["sum"], ["col1"]])
+        expected = DataFrame([3, 4, 5], index=Index([1, 3, 9], name="col3"), columns=m)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_margins_name_unicode(self):
+        # issue #13292
+        greek = "\u0394\u03bf\u03ba\u03b9\u03bc\u03ae"
+        frame = DataFrame({"foo": [1, 2, 3]}, columns=Index(["foo"], dtype=object))
+        table = pivot_table(
+            frame, index=["foo"], aggfunc=len, margins=True, margins_name=greek
+        )
+        index = Index([1, 2, 3, greek], dtype="object", name="foo")
+        expected = DataFrame(index=index, columns=[])
+        tm.assert_frame_equal(table, expected)
+
+    def test_pivot_string_as_func(self):
+        # GH #18713
+        # for correctness purposes
+        data = DataFrame(
+            {
+                "A": [
+                    "foo",
+                    "foo",
+                    "foo",
+                    "foo",
+                    "bar",
+                    "bar",
+                    "bar",
+                    "bar",
+                    "foo",
+                    "foo",
+                    "foo",
+                ],
+                "B": [
+                    "one",
+                    "one",
+                    "one",
+                    "two",
+                    "one",
+                    "one",
+                    "one",
+                    "two",
+                    "two",
+                    "two",
+                    "one",
+                ],
+                "C": range(11),
+            }
+        )
+
+        result = pivot_table(data, index="A", columns="B", aggfunc="sum")
+        mi = MultiIndex(
+            levels=[["C"], ["one", "two"]], codes=[[0, 0], [0, 1]], names=[None, "B"]
+        )
+        expected = DataFrame(
+            {("C", "one"): {"bar": 15, "foo": 13}, ("C", "two"): {"bar": 7, "foo": 20}},
+            columns=mi,
+        ).rename_axis("A")
+        tm.assert_frame_equal(result, expected)
+
+        result = pivot_table(data, index="A", columns="B", aggfunc=["sum", "mean"])
+        mi = MultiIndex(
+            levels=[["sum", "mean"], ["C"], ["one", "two"]],
+            codes=[[0, 0, 1, 1], [0, 0, 0, 0], [0, 1, 0, 1]],
+            names=[None, None, "B"],
+        )
+        expected = DataFrame(
+            {
+                ("mean", "C", "one"): {"bar": 5.0, "foo": 3.25},
+                ("mean", "C", "two"): {"bar": 7.0, "foo": 6.666666666666667},
+                ("sum", "C", "one"): {"bar": 15, "foo": 13},
+                ("sum", "C", "two"): {"bar": 7, "foo": 20},
+            },
+            columns=mi,
+        ).rename_axis("A")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("kwargs", [{"a": 2}, {"a": 2, "b": 3}, {"b": 3, "a": 2}])
+    def test_pivot_table_kwargs(self, kwargs):
+        # GH#57884
+        def f(x, a, b=3):
+            return x.sum() * a + b
+
+        def g(x):
+            return f(x, **kwargs)
+
+        df = DataFrame(
+            {
+                "A": ["good", "bad", "good", "bad", "good"],
+                "B": ["one", "two", "one", "three", "two"],
+                "X": [2, 5, 4, 20, 10],
+            }
+        )
+        result = pivot_table(
+            df, index="A", columns="B", values="X", aggfunc=f, **kwargs
+        )
+        expected = pivot_table(df, index="A", columns="B", values="X", aggfunc=g)
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "kwargs", [{}, {"b": 10}, {"a": 3}, {"a": 3, "b": 10}, {"b": 10, "a": 3}]
+    )
+    def test_pivot_table_kwargs_margin(self, data, kwargs):
+        # GH#57884
+        def f(x, a=5, b=7):
+            return (x.sum() + b) * a
+
+        def g(x):
+            return f(x, **kwargs)
+
+        result = data.pivot_table(
+            values="D",
+            index=["A", "B"],
+            columns="C",
+            aggfunc=f,
+            margins=True,
+            fill_value=0,
+            **kwargs,
+        )
+
+        expected = data.pivot_table(
+            values="D",
+            index=["A", "B"],
+            columns="C",
+            aggfunc=g,
+            margins=True,
+            fill_value=0,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "f, f_numpy",
+        [
+            ("sum", np.sum),
+            ("mean", np.mean),
+            ("min", np.min),
+            (["sum", "mean"], [np.sum, np.mean]),
+            (["sum", "min"], [np.sum, np.min]),
+            (["max", "mean"], [np.max, np.mean]),
+        ],
+    )
+    def test_pivot_string_func_vs_func(self, f, f_numpy, data):
+        # GH #18713
+        # for consistency purposes
+        data = data.drop(columns="C")
+        result = pivot_table(data, index="A", columns="B", aggfunc=f)
+        expected = pivot_table(data, index="A", columns="B", aggfunc=f_numpy)
+
+        if not np_version_gte1p25 and isinstance(f_numpy, list):
+            # Prior to 1.25, np.min/np.max would come through as amin and amax
+            mapper = {"amin": "min", "amax": "max", "sum": "sum", "mean": "mean"}
+            expected.columns = expected.columns.map(
+                lambda x: (mapper[x[0]], x[1], x[2])
+            )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.slow
+    def test_pivot_number_of_levels_larger_than_int32(
+        self, performance_warning, monkeypatch
+    ):
+        # GH 20601
+        # GH 26314: Change ValueError to PerformanceWarning
+        class MockUnstacker(reshape_lib._Unstacker):
+            def __init__(self, *args, **kwargs) -> None:
+                # __init__ will raise the warning
+                super().__init__(*args, **kwargs)
+                raise Exception("Don't compute final result.")
+
+        with monkeypatch.context() as m:
+            m.setattr(reshape_lib, "_Unstacker", MockUnstacker)
+            df = DataFrame(
+                {"ind1": np.arange(2**16), "ind2": np.arange(2**16), "count": 0}
+            )
+
+            msg = "The following operation may generate"
+            with tm.assert_produces_warning(performance_warning, match=msg):
+                with pytest.raises(Exception, match="Don't compute final result."):
+                    df.pivot_table(
+                        index="ind1", columns="ind2", values="count", aggfunc="count"
+                    )
+
+    def test_pivot_table_aggfunc_dropna(self, dropna):
+        # GH 22159
+        df = DataFrame(
+            {
+                "fruit": ["apple", "peach", "apple"],
+                "size": [1, 1, 2],
+                "taste": [7, 6, 6],
+            }
+        )
+
+        def ret_one(x):
+            return 1
+
+        def ret_sum(x):
+            return sum(x)
+
+        def ret_none(x):
+            return np.nan
+
+        result = pivot_table(
+            df, columns="fruit", aggfunc=[ret_sum, ret_none, ret_one], dropna=dropna
+        )
+
+        data = [[3, 1, np.nan, np.nan, 1, 1], [13, 6, np.nan, np.nan, 1, 1]]
+        col = MultiIndex.from_product(
+            [["ret_sum", "ret_none", "ret_one"], ["apple", "peach"]],
+            names=[None, "fruit"],
+        )
+        expected = DataFrame(data, index=["size", "taste"], columns=col)
+
+        if dropna:
+            expected = expected.dropna(axis="columns")
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_aggfunc_scalar_dropna(self, dropna):
+        # GH 22159
+        df = DataFrame(
+            {"A": ["one", "two", "one"], "x": [3, np.nan, 2], "y": [1, np.nan, np.nan]}
+        )
+
+        result = pivot_table(df, columns="A", aggfunc="mean", dropna=dropna)
+
+        data = [[2.5, np.nan], [1, np.nan]]
+        col = Index(["one", "two"], name="A")
+        expected = DataFrame(data, index=["x", "y"], columns=col)
+
+        if dropna:
+            expected = expected.dropna(axis="columns")
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("margins", [True, False])
+    def test_pivot_table_empty_aggfunc(self, margins):
+        # GH 9186 & GH 13483 & GH 49240
+        df = DataFrame(
+            {
+                "A": [2, 2, 3, 3, 2],
+                "id": [5, 6, 7, 8, 9],
+                "C": ["p", "q", "q", "p", "q"],
+                "D": [None, None, None, None, None],
+            }
+        )
+        result = df.pivot_table(
+            index="A", columns="D", values="id", aggfunc=np.size, margins=margins
+        )
+        exp_cols = Index([], name="D")
+        expected = DataFrame(index=Index([], dtype="int64", name="A"), columns=exp_cols)
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_no_column_raises(self):
+        # GH 10326
+        def agg(arr):
+            return np.mean(arr)
+
+        df = DataFrame({"X": [0, 0, 1, 1], "Y": [0, 1, 0, 1], "Z": [10, 20, 30, 40]})
+        with pytest.raises(KeyError, match="notpresent"):
+            df.pivot_table("notpresent", "X", "Y", aggfunc=agg)
+
+    def test_pivot_table_multiindex_columns_doctest_case(self):
+        # The relevant characteristic is that the call
+        #  to maybe_downcast_to_dtype(agged[v], data[v].dtype) in
+        #  __internal_pivot_table has `agged[v]` a DataFrame instead of Series,
+        #  In this case this is because agged.columns is a MultiIndex and 'v'
+        #  is only indexing on its first level.
+        df = DataFrame(
+            {
+                "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+                "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+                "C": [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+                "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+                "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+            }
+        )
+
+        table = pivot_table(
+            df,
+            values=["D", "E"],
+            index=["A", "C"],
+            aggfunc={"D": "mean", "E": ["min", "max", "mean"]},
+        )
+        cols = MultiIndex.from_tuples(
+            [("D", "mean"), ("E", "max"), ("E", "mean"), ("E", "min")]
+        )
+        index = MultiIndex.from_tuples(
+            [("bar", "large"), ("bar", "small"), ("foo", "large"), ("foo", "small")],
+            names=["A", "C"],
+        )
+        vals = np.array(
+            [
+                [5.5, 9.0, 7.5, 6.0],
+                [5.5, 9.0, 8.5, 8.0],
+                [2.0, 5.0, 4.5, 4.0],
+                [2.33333333, 6.0, 4.33333333, 2.0],
+            ]
+        )
+        expected = DataFrame(vals, columns=cols, index=index)
+        expected[("E", "min")] = expected[("E", "min")].astype(np.int64)
+        expected[("E", "max")] = expected[("E", "max")].astype(np.int64)
+        tm.assert_frame_equal(table, expected)
+
+    def test_pivot_table_sort_false(self):
+        # GH#39143
+        df = DataFrame(
+            {
+                "a": ["d1", "d4", "d3"],
+                "col": ["a", "b", "c"],
+                "num": [23, 21, 34],
+                "year": ["2018", "2018", "2019"],
+            }
+        )
+        result = df.pivot_table(
+            index=["a", "col"], columns="year", values="num", aggfunc="sum", sort=False
+        )
+        expected = DataFrame(
+            [[23, np.nan], [21, np.nan], [np.nan, 34]],
+            columns=Index(["2018", "2019"], name="year"),
+            index=MultiIndex.from_arrays(
+                [["d1", "d4", "d3"], ["a", "b", "c"]], names=["a", "col"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_nullable_margins(self):
+        # GH#48681
+        df = DataFrame(
+            {"a": "A", "b": [1, 2], "sales": Series([10, 11], dtype="Int64")}
+        )
+
+        result = df.pivot_table(index="b", columns="a", margins=True, aggfunc="sum")
+        expected = DataFrame(
+            [[10, 10], [11, 11], [21, 21]],
+            index=Index([1, 2, "All"], name="b"),
+            columns=MultiIndex.from_tuples(
+                [("sales", "A"), ("sales", "All")], names=[None, "a"]
+            ),
+            dtype="Int64",
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_sort_false_with_multiple_values(self):
+        df = DataFrame(
+            {
+                "firstname": ["John", "Michael"],
+                "lastname": ["Foo", "Bar"],
+                "height": [173, 182],
+                "age": [47, 33],
+            }
+        )
+        result = df.pivot_table(
+            index=["lastname", "firstname"], values=["height", "age"], sort=False
+        )
+        expected = DataFrame(
+            [[173.0, 47.0], [182.0, 33.0]],
+            columns=["height", "age"],
+            index=MultiIndex.from_tuples(
+                [("Foo", "John"), ("Bar", "Michael")],
+                names=["lastname", "firstname"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_with_margins_and_numeric_columns(self):
+        # GH 26568
+        df = DataFrame([["a", "x", 1], ["a", "y", 2], ["b", "y", 3], ["b", "z", 4]])
+        df.columns = [10, 20, 30]
+
+        result = df.pivot_table(
+            index=10, columns=20, values=30, aggfunc="sum", fill_value=0, margins=True
+        )
+
+        expected = DataFrame([[1, 2, 0, 3], [0, 3, 4, 7], [1, 5, 4, 10]])
+        expected.columns = ["x", "y", "z", "All"]
+        expected.index = ["a", "b", "All"]
+        expected.columns.name = 20
+        expected.index.name = 10
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype,expected_dtype", [("Int64", "Float64"), ("int64", "float64")]
+    )
+    def test_pivot_ea_dtype_dropna(self, dropna, dtype, expected_dtype):
+        # GH#47477
+        # GH#47971
+        df = DataFrame({"x": "a", "y": "b", "age": Series([20, 40], dtype=dtype)})
+        result = df.pivot_table(
+            index="x", columns="y", values="age", aggfunc="mean", dropna=dropna
+        )
+        expected = DataFrame(
+            [[30]],
+            index=Index(["a"], name="x"),
+            columns=Index(["b"], name="y"),
+            dtype=expected_dtype,
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_datetime_warning(self):
+        # GH#48683
+        df = DataFrame(
+            {
+                "a": "A",
+                "b": [1, 2],
+                "date": pd.Timestamp("2019-12-31"),
+                "sales": [10.0, 11],
+            }
+        )
+        with tm.assert_produces_warning(None):
+            result = df.pivot_table(
+                index=["b", "date"], columns="a", margins=True, aggfunc="sum"
+            )
+        expected = DataFrame(
+            [[10.0, 10.0], [11.0, 11.0], [21.0, 21.0]],
+            index=MultiIndex.from_arrays(
+                [
+                    Index([1, 2, "All"], name="b"),
+                    Index(
+                        [pd.Timestamp("2019-12-31"), pd.Timestamp("2019-12-31"), ""],
+                        dtype=object,
+                        name="date",
+                    ),
+                ]
+            ),
+            columns=MultiIndex.from_tuples(
+                [("sales", "A"), ("sales", "All")], names=[None, "a"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_with_mixed_nested_tuples(self):
+        # GH 50342
+        df = DataFrame(
+            {
+                "A": ["foo", "foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar"],
+                "B": ["one", "one", "one", "two", "two", "one", "one", "two", "two"],
+                "C": [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+                "D": [1, 2, 2, 3, 3, 4, 5, 6, 7],
+                "E": [2, 4, 5, 5, 6, 6, 8, 9, 9],
+                ("col5",): [
+                    "foo",
+                    "foo",
+                    "foo",
+                    "foo",
+                    "foo",
+                    "bar",
+                    "bar",
+                    "bar",
+                    "bar",
+                ],
+                ("col6", 6): [
+                    "one",
+                    "one",
+                    "one",
+                    "two",
+                    "two",
+                    "one",
+                    "one",
+                    "two",
+                    "two",
+                ],
+                (7, "seven"): [
+                    "small",
+                    "large",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                    "small",
+                    "small",
+                    "large",
+                ],
+            }
+        )
+        result = pivot_table(
+            df, values="D", index=["A", "B"], columns=[(7, "seven")], aggfunc="sum"
+        )
+        expected = DataFrame(
+            [[4.0, 5.0], [7.0, 6.0], [4.0, 1.0], [np.nan, 6.0]],
+            columns=Index(["large", "small"], name=(7, "seven")),
+            index=MultiIndex.from_arrays(
+                [["bar", "bar", "foo", "foo"], ["one", "two"] * 2], names=["A", "B"]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_table_aggfunc_nunique_with_different_values(self):
+        test = DataFrame(
+            {
+                "a": range(10),
+                "b": range(10),
+                "c": range(10),
+                "d": range(10),
+            }
+        )
+
+        columnval = MultiIndex.from_arrays(
+            [
+                ["nunique" for i in range(10)],
+                ["c" for i in range(10)],
+                range(10),
+            ],
+            names=(None, None, "b"),
+        )
+        nparr = np.full((10, 10), np.nan)
+        np.fill_diagonal(nparr, 1.0)
+
+        expected = DataFrame(nparr, index=Index(range(10), name="a"), columns=columnval)
+        result = test.pivot_table(
+            index=[
+                "a",
+            ],
+            columns=[
+                "b",
+            ],
+            values=[
+                "c",
+            ],
+            aggfunc=["nunique"],
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+
+class TestPivot:
+    def test_pivot(self):
+        data = {
+            "index": ["A", "B", "C", "C", "B", "A"],
+            "columns": ["One", "One", "One", "Two", "Two", "Two"],
+            "values": [1.0, 2.0, 3.0, 3.0, 2.0, 1.0],
+        }
+
+        frame = DataFrame(data)
+        pivoted = frame.pivot(index="index", columns="columns", values="values")
+
+        expected = DataFrame(
+            {
+                "One": {"A": 1.0, "B": 2.0, "C": 3.0},
+                "Two": {"A": 1.0, "B": 2.0, "C": 3.0},
+            }
+        )
+
+        expected.index.name, expected.columns.name = "index", "columns"
+        tm.assert_frame_equal(pivoted, expected)
+
+        # name tracking
+        assert pivoted.index.name == "index"
+        assert pivoted.columns.name == "columns"
+
+        # don't specify values
+        pivoted = frame.pivot(index="index", columns="columns")
+        assert pivoted.index.name == "index"
+        assert pivoted.columns.names == (None, "columns")
+
+    def test_pivot_duplicates(self):
+        data = DataFrame(
+            {
+                "a": ["bar", "bar", "foo", "foo", "foo"],
+                "b": ["one", "two", "one", "one", "two"],
+                "c": [1.0, 2.0, 3.0, 3.0, 4.0],
+            }
+        )
+        with pytest.raises(ValueError, match="duplicate entries"):
+            data.pivot(index="a", columns="b", values="c")
+
+    def test_pivot_empty(self):
+        df = DataFrame(columns=["a", "b", "c"])
+        result = df.pivot(index="a", columns="b", values="c")
+        expected = DataFrame(index=[], columns=[])
+        tm.assert_frame_equal(result, expected, check_names=False)
+
+    def test_pivot_integer_bug(self, any_string_dtype):
+        df = DataFrame(
+            data=[("A", "1", "A1"), ("B", "2", "B2")], dtype=any_string_dtype
+        )
+
+        result = df.pivot(index=1, columns=0, values=2)
+        expected_columns = Index(["A", "B"], name=0, dtype=any_string_dtype)
+        if any_string_dtype == "object":
+            expected_columns = expected_columns.astype("str")
+        tm.assert_index_equal(result.columns, expected_columns)
+
+    def test_pivot_index_none(self):
+        # GH#3962
+        data = {
+            "index": ["A", "B", "C", "C", "B", "A"],
+            "columns": ["One", "One", "One", "Two", "Two", "Two"],
+            "values": [1.0, 2.0, 3.0, 3.0, 2.0, 1.0],
+        }
+
+        frame = DataFrame(data).set_index("index")
+        result = frame.pivot(columns="columns", values="values")
+        expected = DataFrame(
+            {
+                "One": {"A": 1.0, "B": 2.0, "C": 3.0},
+                "Two": {"A": 1.0, "B": 2.0, "C": 3.0},
+            }
+        )
+
+        expected.index.name, expected.columns.name = "index", "columns"
+        tm.assert_frame_equal(result, expected)
+
+        # omit values
+        result = frame.pivot(columns="columns")
+
+        expected.columns = MultiIndex.from_tuples(
+            [("values", "One"), ("values", "Two")], names=[None, "columns"]
+        )
+        expected.index.name = "index"
+        tm.assert_frame_equal(result, expected, check_names=False)
+        assert result.index.name == "index"
+        assert result.columns.names == (None, "columns")
+        expected.columns = expected.columns.droplevel(0)
+        result = frame.pivot(columns="columns", values="values")
+
+        expected.columns.name = "columns"
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_index_list_values_none_immutable_args(self):
+        # GH37635
+        df = DataFrame(
+            {
+                "lev1": [1, 1, 1, 2, 2, 2],
+                "lev2": [1, 1, 2, 1, 1, 2],
+                "lev3": [1, 2, 1, 2, 1, 2],
+                "lev4": [1, 2, 3, 4, 5, 6],
+                "values": [0, 1, 2, 3, 4, 5],
+            }
+        )
+        index = ["lev1", "lev2"]
+        columns = ["lev3"]
+        result = df.pivot(index=index, columns=columns)
+
+        expected = DataFrame(
+            np.array(
+                [
+                    [1.0, 2.0, 0.0, 1.0],
+                    [3.0, np.nan, 2.0, np.nan],
+                    [5.0, 4.0, 4.0, 3.0],
+                    [np.nan, 6.0, np.nan, 5.0],
+                ]
+            ),
+            index=MultiIndex.from_arrays(
+                [(1, 1, 2, 2), (1, 2, 1, 2)], names=["lev1", "lev2"]
+            ),
+            columns=MultiIndex.from_arrays(
+                [("lev4", "lev4", "values", "values"), (1, 2, 1, 2)],
+                names=[None, "lev3"],
+            ),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+        assert index == ["lev1", "lev2"]
+        assert columns == ["lev3"]
+
+    def test_pivot_columns_not_given(self):
+        # GH#48293
+        df = DataFrame({"a": [1], "b": 1})
+        with pytest.raises(TypeError, match="missing 1 required keyword-only argument"):
+            df.pivot()
+
+    # this still fails because columns=None gets passed down to unstack as level=None
+    # while at that point None was converted to NaN
+    @pytest.mark.xfail(
+        using_string_dtype(), reason="TODO(infer_string) None is cast to NaN"
+    )
+    def test_pivot_columns_is_none(self):
+        # GH#48293
+        df = DataFrame({None: [1], "b": 2, "c": 3})
+        result = df.pivot(columns=None)
+        expected = DataFrame({("b", 1): [2], ("c", 1): 3})
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot(columns=None, index="b")
+        expected = DataFrame({("c", 1): 3}, index=Index([2], name="b"))
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot(columns=None, index="b", values="c")
+        expected = DataFrame({1: 3}, index=Index([2], name="b"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_index_is_none(self, using_infer_string):
+        # GH#48293
+        df = DataFrame({None: [1], "b": 2, "c": 3})
+
+        result = df.pivot(columns="b", index=None)
+        expected = DataFrame({("c", 2): 3}, index=[1])
+        expected.columns.names = [None, "b"]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot(columns="b", index=None, values="c")
+        expected = DataFrame(3, index=[1], columns=Index([2], name="b"))
+        if using_infer_string:
+            expected.index.name = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_values_is_none(self):
+        # GH#48293
+        df = DataFrame({None: [1], "b": 2, "c": 3})
+
+        result = df.pivot(columns="b", index="c", values=None)
+        expected = DataFrame(
+            1, index=Index([3], name="c"), columns=Index([2], name="b")
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.pivot(columns="b", values=None)
+        expected = DataFrame(1, index=[0], columns=Index([2], name="b"))
+        tm.assert_frame_equal(result, expected)
+
+    def test_pivot_not_changing_index_name(self):
+        # GH#52692
+        df = DataFrame({"one": ["a"], "two": 0, "three": 1})
+        expected = df.copy(deep=True)
+        df.pivot(index="one", columns="two", values="three")
+        tm.assert_frame_equal(df, expected)
+
+    def test_pivot_table_empty_dataframe_correct_index(self):
+        # GH 21932
+        df = DataFrame([], columns=["a", "b", "value"])
+        pivot = df.pivot_table(index="a", columns="b", values="value", aggfunc="count")
+
+        expected = Index([], dtype="object", name="b")
+        tm.assert_index_equal(pivot.columns, expected)
+
+    def test_pivot_table_handles_explicit_datetime_types(self):
+        # GH#43574
+        df = DataFrame(
+            [
+                {"a": "x", "date_str": "2023-01-01", "amount": 1},
+                {"a": "y", "date_str": "2023-01-02", "amount": 2},
+                {"a": "z", "date_str": "2023-01-03", "amount": 3},
+            ]
+        )
+        df["date"] = pd.to_datetime(df["date_str"])
+
+        with tm.assert_produces_warning(False):
+            pivot = df.pivot_table(
+                index=["a", "date"], values=["amount"], aggfunc="sum", margins=True
+            )
+
+        expected = MultiIndex.from_tuples(
+            [
+                ("x", datetime.strptime("2023-01-01 00:00:00", "%Y-%m-%d %H:%M:%S")),
+                ("y", datetime.strptime("2023-01-02 00:00:00", "%Y-%m-%d %H:%M:%S")),
+                ("z", datetime.strptime("2023-01-03 00:00:00", "%Y-%m-%d %H:%M:%S")),
+                ("All", ""),
+            ],
+            names=["a", "date"],
+        )
+        tm.assert_index_equal(pivot.index, expected)
+
+    def test_pivot_table_with_margins_and_numeric_column_names(self):
+        # GH#26568
+        df = DataFrame([["a", "x", 1], ["a", "y", 2], ["b", "y", 3], ["b", "z", 4]])
+
+        result = df.pivot_table(
+            index=0, columns=1, values=2, aggfunc="sum", fill_value=0, margins=True
+        )
+
+        expected = DataFrame(
+            [[1, 2, 0, 3], [0, 3, 4, 7], [1, 5, 4, 10]],
+            columns=Index(["x", "y", "z", "All"], name=1),
+            index=Index(["a", "b", "All"], name=0),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("m", [1, 10])
+    def test_unstack_copy(self, m):
+        # GH#56633
+        levels = np.arange(m)
+        index = MultiIndex.from_product([levels] * 2)
+        values = np.arange(m * m * 100).reshape(m * m, 100)
+        df = DataFrame(values, index, np.arange(100))
+        df_orig = df.copy()
+        result = df.unstack(sort=False)
+        result.iloc[0, 0] = -1
+        tm.assert_frame_equal(df, df_orig)
+
+    def test_pivot_empty_with_datetime(self):
+        # GH#59126
+        df = DataFrame(
+            {
+                "timestamp": Series([], dtype=pd.DatetimeTZDtype(tz="UTC")),
+                "category": Series([], dtype=str),
+                "value": Series([], dtype=str),
+            }
+        )
+        df_pivoted = df.pivot_table(
+            index="category", columns="value", values="timestamp"
+        )
+        assert df_pivoted.empty
+
+    def test_pivot_margins_with_none_index(self):
+        # GH#58722
+        df = DataFrame(
+            {
+                "x": [1, 1, 2],
+                "y": [3, 3, 4],
+                "z": [5, 5, 6],
+                "w": [7, 8, 9],
+            }
+        )
+        result = df.pivot_table(
+            index=None,
+            columns=["y", "z"],
+            values="w",
+            margins=True,
+            aggfunc="count",
+        )
+        expected = DataFrame(
+            [[2, 2, 1, 1]],
+            index=["w"],
+            columns=MultiIndex(
+                levels=[[3, 4], [5, 6, "All"]],
+                codes=[[0, 0, 1, 1], [0, 2, 1, 2]],
+                names=["y", "z"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_pivot_multilevel.py b/pandas/pandas/tests/reshape/test_pivot_multilevel.py
new file mode 100644
index 0000000000000..2c9d54c3db72c
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_pivot_multilevel.py
@@ -0,0 +1,252 @@
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "input_index, input_columns, input_values, "
+    "expected_values, expected_columns, expected_index",
+    [
+        (
+            ["lev4"],
+            "lev3",
+            "values",
+            [
+                [0.0, np.nan],
+                [np.nan, 1.0],
+                [2.0, np.nan],
+                [np.nan, 3.0],
+                [4.0, np.nan],
+                [np.nan, 5.0],
+                [6.0, np.nan],
+                [np.nan, 7.0],
+            ],
+            Index([1, 2], name="lev3"),
+            Index([1, 2, 3, 4, 5, 6, 7, 8], name="lev4"),
+        ),
+        (
+            ["lev4"],
+            "lev3",
+            lib.no_default,
+            [
+                [1.0, np.nan, 1.0, np.nan, 0.0, np.nan],
+                [np.nan, 1.0, np.nan, 1.0, np.nan, 1.0],
+                [1.0, np.nan, 2.0, np.nan, 2.0, np.nan],
+                [np.nan, 1.0, np.nan, 2.0, np.nan, 3.0],
+                [2.0, np.nan, 1.0, np.nan, 4.0, np.nan],
+                [np.nan, 2.0, np.nan, 1.0, np.nan, 5.0],
+                [2.0, np.nan, 2.0, np.nan, 6.0, np.nan],
+                [np.nan, 2.0, np.nan, 2.0, np.nan, 7.0],
+            ],
+            MultiIndex.from_tuples(
+                [
+                    ("lev1", 1),
+                    ("lev1", 2),
+                    ("lev2", 1),
+                    ("lev2", 2),
+                    ("values", 1),
+                    ("values", 2),
+                ],
+                names=[None, "lev3"],
+            ),
+            Index([1, 2, 3, 4, 5, 6, 7, 8], name="lev4"),
+        ),
+        (
+            ["lev1", "lev2"],
+            "lev3",
+            "values",
+            [[0, 1], [2, 3], [4, 5], [6, 7]],
+            Index([1, 2], name="lev3"),
+            MultiIndex.from_tuples(
+                [(1, 1), (1, 2), (2, 1), (2, 2)], names=["lev1", "lev2"]
+            ),
+        ),
+        (
+            ["lev1", "lev2"],
+            "lev3",
+            lib.no_default,
+            [[1, 2, 0, 1], [3, 4, 2, 3], [5, 6, 4, 5], [7, 8, 6, 7]],
+            MultiIndex.from_tuples(
+                [("lev4", 1), ("lev4", 2), ("values", 1), ("values", 2)],
+                names=[None, "lev3"],
+            ),
+            MultiIndex.from_tuples(
+                [(1, 1), (1, 2), (2, 1), (2, 2)], names=["lev1", "lev2"]
+            ),
+        ),
+    ],
+)
+def test_pivot_list_like_index(
+    input_index,
+    input_columns,
+    input_values,
+    expected_values,
+    expected_columns,
+    expected_index,
+):
+    # GH 21425, test when index is given a list
+    df = pd.DataFrame(
+        {
+            "lev1": [1, 1, 1, 1, 2, 2, 2, 2],
+            "lev2": [1, 1, 2, 2, 1, 1, 2, 2],
+            "lev3": [1, 2, 1, 2, 1, 2, 1, 2],
+            "lev4": [1, 2, 3, 4, 5, 6, 7, 8],
+            "values": [0, 1, 2, 3, 4, 5, 6, 7],
+        }
+    )
+
+    result = df.pivot(index=input_index, columns=input_columns, values=input_values)
+    expected = pd.DataFrame(
+        expected_values, columns=expected_columns, index=expected_index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "input_index, input_columns, input_values, "
+    "expected_values, expected_columns, expected_index",
+    [
+        (
+            "lev4",
+            ["lev3"],
+            "values",
+            [
+                [0.0, np.nan],
+                [np.nan, 1.0],
+                [2.0, np.nan],
+                [np.nan, 3.0],
+                [4.0, np.nan],
+                [np.nan, 5.0],
+                [6.0, np.nan],
+                [np.nan, 7.0],
+            ],
+            Index([1, 2], name="lev3"),
+            Index([1, 2, 3, 4, 5, 6, 7, 8], name="lev4"),
+        ),
+        (
+            ["lev1", "lev2"],
+            ["lev3"],
+            "values",
+            [[0, 1], [2, 3], [4, 5], [6, 7]],
+            Index([1, 2], name="lev3"),
+            MultiIndex.from_tuples(
+                [(1, 1), (1, 2), (2, 1), (2, 2)], names=["lev1", "lev2"]
+            ),
+        ),
+        (
+            ["lev1"],
+            ["lev2", "lev3"],
+            "values",
+            [[0, 1, 2, 3], [4, 5, 6, 7]],
+            MultiIndex.from_tuples(
+                [(1, 1), (1, 2), (2, 1), (2, 2)], names=["lev2", "lev3"]
+            ),
+            Index([1, 2], name="lev1"),
+        ),
+        (
+            ["lev1", "lev2"],
+            ["lev3", "lev4"],
+            "values",
+            [
+                [0.0, 1.0, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+                [np.nan, np.nan, 2.0, 3.0, np.nan, np.nan, np.nan, np.nan],
+                [np.nan, np.nan, np.nan, np.nan, 4.0, 5.0, np.nan, np.nan],
+                [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, 6.0, 7.0],
+            ],
+            MultiIndex.from_tuples(
+                [(1, 1), (2, 2), (1, 3), (2, 4), (1, 5), (2, 6), (1, 7), (2, 8)],
+                names=["lev3", "lev4"],
+            ),
+            MultiIndex.from_tuples(
+                [(1, 1), (1, 2), (2, 1), (2, 2)], names=["lev1", "lev2"]
+            ),
+        ),
+    ],
+)
+def test_pivot_list_like_columns(
+    input_index,
+    input_columns,
+    input_values,
+    expected_values,
+    expected_columns,
+    expected_index,
+):
+    # GH 21425, test when columns is given a list
+    df = pd.DataFrame(
+        {
+            "lev1": [1, 1, 1, 1, 2, 2, 2, 2],
+            "lev2": [1, 1, 2, 2, 1, 1, 2, 2],
+            "lev3": [1, 2, 1, 2, 1, 2, 1, 2],
+            "lev4": [1, 2, 3, 4, 5, 6, 7, 8],
+            "values": [0, 1, 2, 3, 4, 5, 6, 7],
+        }
+    )
+
+    result = df.pivot(index=input_index, columns=input_columns, values=input_values)
+    expected = pd.DataFrame(
+        expected_values, columns=expected_columns, index=expected_index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_pivot_multiindexed_rows_and_cols():
+    # GH 36360
+
+    df = pd.DataFrame(
+        data=np.arange(12).reshape(4, 3),
+        columns=MultiIndex.from_tuples(
+            [(0, 0), (0, 1), (0, 2)], names=["col_L0", "col_L1"]
+        ),
+        index=MultiIndex.from_tuples(
+            [(0, 0, 0), (0, 0, 1), (1, 1, 1), (1, 0, 0)],
+            names=["idx_L0", "idx_L1", "idx_L2"],
+        ),
+    )
+
+    res = df.pivot_table(
+        index=["idx_L0"],
+        columns=["idx_L1"],
+        values=[(0, 1)],
+        aggfunc=lambda col: col.values.sum(),
+    )
+
+    expected = pd.DataFrame(
+        data=[[5, np.nan], [10, 7.0]],
+        columns=MultiIndex.from_tuples(
+            [(0, 1, 0), (0, 1, 1)], names=["col_L0", "col_L1", "idx_L1"]
+        ),
+        index=Index([0, 1], dtype="int64", name="idx_L0"),
+    )
+    expected = expected.astype("float64")
+
+    tm.assert_frame_equal(res, expected)
+
+
+def test_pivot_df_multiindex_index_none():
+    # GH 23955
+    df = pd.DataFrame(
+        [
+            ["A", "A1", "label1", 1],
+            ["A", "A2", "label2", 2],
+            ["B", "A1", "label1", 3],
+            ["B", "A2", "label2", 4],
+        ],
+        columns=["index_1", "index_2", "label", "value"],
+    )
+    df = df.set_index(["index_1", "index_2"])
+
+    result = df.pivot(columns="label", values="value")
+    expected = pd.DataFrame(
+        [[1.0, np.nan], [np.nan, 2.0], [3.0, np.nan], [np.nan, 4.0]],
+        index=df.index,
+        columns=Index(["label1", "label2"], name="label"),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/reshape/test_qcut.py b/pandas/pandas/tests/reshape/test_qcut.py
new file mode 100644
index 0000000000000..b2e9f26e1c407
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_qcut.py
@@ -0,0 +1,321 @@
+import os
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DatetimeIndex,
+    Interval,
+    IntervalIndex,
+    NaT,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    cut,
+    date_range,
+    isna,
+    qcut,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+
+from pandas.tseries.offsets import Day
+
+
+def test_qcut():
+    arr = np.random.default_rng(2).standard_normal(1000)
+
+    # We store the bins as Index that have been
+    # rounded to comparisons are a bit tricky.
+    labels, _ = qcut(arr, 4, retbins=True)
+    ex_bins = np.quantile(arr, [0, 0.25, 0.5, 0.75, 1.0])
+
+    result = labels.categories.left.values
+    assert np.allclose(result, ex_bins[:-1], atol=1e-2)
+
+    result = labels.categories.right.values
+    assert np.allclose(result, ex_bins[1:], atol=1e-2)
+
+    ex_levels = cut(arr, ex_bins, include_lowest=True)
+    tm.assert_categorical_equal(labels, ex_levels)
+
+
+def test_qcut_bounds():
+    arr = np.random.default_rng(2).standard_normal(1000)
+
+    factor = qcut(arr, 10, labels=False)
+    assert len(np.unique(factor)) == 10
+
+
+def test_qcut_specify_quantiles():
+    arr = np.random.default_rng(2).standard_normal(100)
+    factor = qcut(arr, [0, 0.25, 0.5, 0.75, 1.0])
+
+    expected = qcut(arr, 4)
+    tm.assert_categorical_equal(factor, expected)
+
+
+def test_qcut_all_bins_same():
+    with pytest.raises(ValueError, match="edges.*unique"):
+        qcut([0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 3)
+
+
+def test_qcut_include_lowest():
+    values = np.arange(10)
+    ii = qcut(values, 4)
+
+    ex_levels = IntervalIndex(
+        [
+            Interval(-0.001, 2.25),
+            Interval(2.25, 4.5),
+            Interval(4.5, 6.75),
+            Interval(6.75, 9),
+        ]
+    )
+    tm.assert_index_equal(ii.categories, ex_levels)
+
+
+def test_qcut_nas():
+    arr = np.random.default_rng(2).standard_normal(100)
+    arr[:20] = np.nan
+
+    result = qcut(arr, 4)
+    assert isna(result[:20]).all()
+
+
+def test_qcut_index():
+    result = qcut([0, 2], 2)
+    intervals = [Interval(-0.001, 1), Interval(1, 2)]
+
+    expected = Categorical(intervals, ordered=True)
+    tm.assert_categorical_equal(result, expected)
+
+
+def test_qcut_binning_issues(datapath):
+    # see gh-1978, gh-1979
+    cut_file = datapath(os.path.join("reshape", "data", "cut_data.csv"))
+    arr = np.loadtxt(cut_file)
+    result = qcut(arr, 20)
+
+    starts = []
+    ends = []
+
+    for lev in np.unique(result):
+        s = lev.left
+        e = lev.right
+        assert s != e
+
+        starts.append(float(s))
+        ends.append(float(e))
+
+    for (sp, sn), (ep, en) in zip(
+        zip(starts[:-1], starts[1:]), zip(ends[:-1], ends[1:])
+    ):
+        assert sp < sn
+        assert ep < en
+        assert ep <= sn
+
+
+def test_qcut_return_intervals():
+    ser = Series([0, 1, 2, 3, 4, 5, 6, 7, 8])
+    res = qcut(ser, [0, 0.333, 0.666, 1])
+
+    exp_levels = np.array(
+        [Interval(-0.001, 2.664), Interval(2.664, 5.328), Interval(5.328, 8)]
+    )
+    exp = Series(exp_levels.take([0, 0, 0, 1, 1, 1, 2, 2, 2])).astype(
+        CategoricalDtype(ordered=True)
+    )
+    tm.assert_series_equal(res, exp)
+
+
+@pytest.mark.parametrize("labels", ["foo", 1, True])
+def test_qcut_incorrect_labels(labels):
+    # GH 13318
+    values = range(5)
+    msg = "Bin labels must either be False, None or passed in as a list-like argument"
+    with pytest.raises(ValueError, match=msg):
+        qcut(values, 4, labels=labels)
+
+
+@pytest.mark.parametrize("labels", [["a", "b", "c"], list(range(3))])
+def test_qcut_wrong_length_labels(labels):
+    # GH 13318
+    values = range(10)
+    msg = "Bin labels must be one fewer than the number of bin edges"
+    with pytest.raises(ValueError, match=msg):
+        qcut(values, 4, labels=labels)
+
+
+@pytest.mark.parametrize(
+    "labels, expected",
+    [
+        (["a", "b", "c"], ["a", "b", "c"]),
+        (list(range(3)), [0, 1, 2]),
+    ],
+)
+def test_qcut_list_like_labels(labels, expected):
+    # GH 13318
+    values = range(3)
+    result = qcut(values, 3, labels=labels)
+    expected = Categorical(expected, ordered=True)
+    tm.assert_categorical_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "kwargs,msg",
+    [
+        ({"duplicates": "drop"}, None),
+        ({}, "Bin edges must be unique"),
+        ({"duplicates": "raise"}, "Bin edges must be unique"),
+        ({"duplicates": "foo"}, "invalid value for 'duplicates' parameter"),
+    ],
+)
+def test_qcut_duplicates_bin(kwargs, msg):
+    # see gh-7751
+    values = [0, 0, 0, 0, 1, 2, 3]
+
+    if msg is not None:
+        with pytest.raises(ValueError, match=msg):
+            qcut(values, 3, **kwargs)
+    else:
+        result = qcut(values, 3, **kwargs)
+        expected = IntervalIndex([Interval(-0.001, 1), Interval(1, 3)])
+        tm.assert_index_equal(result.categories, expected)
+
+
+@pytest.mark.parametrize(
+    "data,start,end", [(9.0, 8.999, 9.0), (0.0, -0.001, 0.0), (-9.0, -9.001, -9.0)]
+)
+@pytest.mark.parametrize("length", [1, 2])
+@pytest.mark.parametrize("labels", [None, False])
+def test_single_quantile(data, start, end, length, labels):
+    # see gh-15431
+    ser = Series([data] * length)
+    result = qcut(ser, 1, labels=labels)
+
+    if labels is None:
+        intervals = IntervalIndex([Interval(start, end)] * length, closed="right")
+        expected = Series(intervals).astype(CategoricalDtype(ordered=True))
+    else:
+        expected = Series([0] * length, dtype=np.intp)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser",
+    [
+        DatetimeIndex(["20180101", NaT, "20180103"]),
+        TimedeltaIndex(["0 days", NaT, "2 days"]),
+    ],
+    ids=lambda x: str(x.dtype),
+)
+def test_qcut_nat(ser, unit):
+    # see gh-19768
+    ser = Series(ser)
+    ser = ser.dt.as_unit(unit)
+    td = Timedelta(1, unit=unit).as_unit(unit)
+
+    left = Series([ser[0] - td, np.nan, ser[2] - Day()], dtype=ser.dtype)
+    right = Series([ser[2] - Day(), np.nan, ser[2]], dtype=ser.dtype)
+    intervals = IntervalIndex.from_arrays(left, right)
+    expected = Series(Categorical(intervals, ordered=True))
+
+    result = qcut(ser, 2)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("bins", [3, np.linspace(0, 1, 4)])
+def test_datetime_tz_qcut(bins):
+    # see gh-19872
+    tz = "US/Eastern"
+    ser = Series(date_range("20130101", periods=3, tz=tz))
+
+    result = qcut(ser, bins)
+    expected = Series(
+        IntervalIndex(
+            [
+                Interval(
+                    Timestamp("2012-12-31 23:59:59.999999999", tz=tz),
+                    Timestamp("2013-01-01 16:00:00", tz=tz),
+                ),
+                Interval(
+                    Timestamp("2013-01-01 16:00:00", tz=tz),
+                    Timestamp("2013-01-02 08:00:00", tz=tz),
+                ),
+                Interval(
+                    Timestamp("2013-01-02 08:00:00", tz=tz),
+                    Timestamp("2013-01-03 00:00:00", tz=tz),
+                ),
+            ]
+        )
+    ).astype(CategoricalDtype(ordered=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "arg,expected_bins",
+    [
+        [
+            timedelta_range("1day", periods=3),
+            TimedeltaIndex(["1 days", "2 days", "3 days"]),
+        ],
+        [
+            date_range("20180101", periods=3),
+            DatetimeIndex(["2018-01-01", "2018-01-02", "2018-01-03"]),
+        ],
+    ],
+)
+def test_date_like_qcut_bins(arg, expected_bins, unit):
+    # see gh-19891
+    arg = arg.as_unit(unit)
+    expected_bins = expected_bins.as_unit(unit)
+    ser = Series(arg)
+    result, result_bins = qcut(ser, 2, retbins=True)
+    tm.assert_index_equal(result_bins, expected_bins)
+
+
+@pytest.mark.parametrize("bins", [6, 7])
+@pytest.mark.parametrize(
+    "box, compare",
+    [
+        (Series, tm.assert_series_equal),
+        (np.array, tm.assert_categorical_equal),
+        (list, tm.assert_equal),
+    ],
+)
+def test_qcut_bool_coercion_to_int(bins, box, compare):
+    # issue 20303
+    data_expected = box([0, 1, 1, 0, 1] * 10)
+    data_result = box([False, True, True, False, True] * 10)
+    expected = qcut(data_expected, bins, duplicates="drop")
+    result = qcut(data_result, bins, duplicates="drop")
+    compare(result, expected)
+
+
+@pytest.mark.parametrize("q", [2, 5, 10])
+def test_qcut_nullable_integer(q, any_numeric_ea_dtype):
+    arr = pd.array(np.arange(100), dtype=any_numeric_ea_dtype)
+    arr[::2] = pd.NA
+
+    result = qcut(arr, q)
+    expected = qcut(arr.astype(float), q)
+
+    tm.assert_categorical_equal(result, expected)
+
+
+@pytest.mark.parametrize("scale", [1.0, 1 / 3, 17.0])
+@pytest.mark.parametrize("q", [3, 7, 9])
+@pytest.mark.parametrize("precision", [1, 3, 16])
+def test_qcut_contains(scale, q, precision):
+    # GH-59355
+    arr = (scale * np.arange(q + 1)).round(precision)
+    result = qcut(arr, q, precision=precision)
+
+    for value, bucket in zip(arr, result):
+        assert value in bucket
diff --git a/pandas/pandas/tests/reshape/test_union_categoricals.py b/pandas/pandas/tests/reshape/test_union_categoricals.py
new file mode 100644
index 0000000000000..081feae6fc43f
--- /dev/null
+++ b/pandas/pandas/tests/reshape/test_union_categoricals.py
@@ -0,0 +1,369 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.concat import union_categoricals
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestUnionCategoricals:
+    @pytest.mark.parametrize(
+        "a, b, combined",
+        [
+            (list("abc"), list("abd"), list("abcabd")),
+            ([0, 1, 2], [2, 3, 4], [0, 1, 2, 2, 3, 4]),
+            ([0, 1.2, 2], [2, 3.4, 4], [0, 1.2, 2, 2, 3.4, 4]),
+            (
+                ["b", "b", np.nan, "a"],
+                ["a", np.nan, "c"],
+                ["b", "b", np.nan, "a", "a", np.nan, "c"],
+            ),
+            (
+                pd.date_range("2014-01-01", "2014-01-05"),
+                pd.date_range("2014-01-06", "2014-01-07"),
+                pd.date_range("2014-01-01", "2014-01-07"),
+            ),
+            (
+                pd.date_range("2014-01-01", "2014-01-05", tz="US/Central"),
+                pd.date_range("2014-01-06", "2014-01-07", tz="US/Central"),
+                pd.date_range("2014-01-01", "2014-01-07", tz="US/Central"),
+            ),
+            (
+                pd.period_range("2014-01-01", "2014-01-05"),
+                pd.period_range("2014-01-06", "2014-01-07"),
+                pd.period_range("2014-01-01", "2014-01-07"),
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("box", [Categorical, CategoricalIndex, Series])
+    def test_union_categorical(self, a, b, combined, box):
+        # GH 13361
+        result = union_categoricals([box(Categorical(a)), box(Categorical(b))])
+        expected = Categorical(combined)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categorical_ordered_appearance(self):
+        # new categories ordered by appearance
+        s = Categorical(["x", "y", "z"])
+        s2 = Categorical(["a", "b", "c"])
+        result = union_categoricals([s, s2])
+        expected = Categorical(
+            ["x", "y", "z", "a", "b", "c"], categories=["x", "y", "z", "a", "b", "c"]
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categorical_ordered_true(self):
+        s = Categorical([0, 1.2, 2], ordered=True)
+        s2 = Categorical([0, 1.2, 2], ordered=True)
+        result = union_categoricals([s, s2])
+        expected = Categorical([0, 1.2, 2, 0, 1.2, 2], ordered=True)
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categorical_match_types(self):
+        # must exactly match types
+        s = Categorical([0, 1.2, 2])
+        s2 = Categorical([2, 3, 4])
+        msg = "dtype of categories must be the same"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([s, s2])
+
+    def test_union_categorical_empty(self):
+        msg = "No Categoricals to union"
+        with pytest.raises(ValueError, match=msg):
+            union_categoricals([])
+
+    def test_union_categoricals_nan(self):
+        # GH 13759
+        res = union_categoricals(
+            [Categorical([1, 2, np.nan]), Categorical([3, 2, np.nan])]
+        )
+        exp = Categorical([1, 2, np.nan, 3, 2, np.nan])
+        tm.assert_categorical_equal(res, exp)
+
+        res = union_categoricals(
+            [Categorical(["A", "B"]), Categorical(["B", "B", np.nan])]
+        )
+        exp = Categorical(["A", "B", "B", "B", np.nan])
+        tm.assert_categorical_equal(res, exp)
+
+        val1 = [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-03-01"), pd.NaT]
+        val2 = [pd.NaT, pd.Timestamp("2011-01-01"), pd.Timestamp("2011-02-01")]
+
+        res = union_categoricals([Categorical(val1), Categorical(val2)])
+        exp = Categorical(
+            val1 + val2,
+            categories=[
+                pd.Timestamp("2011-01-01"),
+                pd.Timestamp("2011-03-01"),
+                pd.Timestamp("2011-02-01"),
+            ],
+        )
+        tm.assert_categorical_equal(res, exp)
+
+        # all NaN
+        res = union_categoricals(
+            [
+                Categorical(np.array([np.nan, np.nan], dtype=object)),
+                Categorical(["X"], categories=pd.Index(["X"], dtype=object)),
+            ]
+        )
+        exp = Categorical([np.nan, np.nan, "X"])
+        tm.assert_categorical_equal(res, exp)
+
+        res = union_categoricals(
+            [Categorical([np.nan, np.nan]), Categorical([np.nan, np.nan])]
+        )
+        exp = Categorical([np.nan, np.nan, np.nan, np.nan])
+        tm.assert_categorical_equal(res, exp)
+
+    @pytest.mark.parametrize("val", [[], ["1"]])
+    def test_union_categoricals_empty(self, val, request, using_infer_string):
+        # GH 13759
+        if using_infer_string and val == ["1"]:
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason="TDOD(infer_string) object and strings dont match"
+                )
+            )
+        res = union_categoricals([Categorical([]), Categorical(val)])
+        exp = Categorical(val)
+        tm.assert_categorical_equal(res, exp)
+
+    def test_union_categorical_same_category(self):
+        # check fastpath
+        c1 = Categorical([1, 2, 3, 4], categories=[1, 2, 3, 4])
+        c2 = Categorical([3, 2, 1, np.nan], categories=[1, 2, 3, 4])
+        res = union_categoricals([c1, c2])
+        exp = Categorical([1, 2, 3, 4, 3, 2, 1, np.nan], categories=[1, 2, 3, 4])
+        tm.assert_categorical_equal(res, exp)
+
+    def test_union_categorical_same_category_str(self):
+        c1 = Categorical(["z", "z", "z"], categories=["x", "y", "z"])
+        c2 = Categorical(["x", "x", "x"], categories=["x", "y", "z"])
+        res = union_categoricals([c1, c2])
+        exp = Categorical(["z", "z", "z", "x", "x", "x"], categories=["x", "y", "z"])
+        tm.assert_categorical_equal(res, exp)
+
+    def test_union_categorical_same_categories_different_order(self):
+        # https://github.com/pandas-dev/pandas/issues/19096
+        c1 = Categorical(["a", "b", "c"], categories=["a", "b", "c"])
+        c2 = Categorical(["a", "b", "c"], categories=["b", "a", "c"])
+        result = union_categoricals([c1, c2])
+        expected = Categorical(
+            ["a", "b", "c", "a", "b", "c"], categories=["a", "b", "c"]
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_ordered(self):
+        c1 = Categorical([1, 2, 3], ordered=True)
+        c2 = Categorical([1, 2, 3], ordered=False)
+
+        msg = "Categorical.ordered must be the same"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2])
+
+        res = union_categoricals([c1, c1])
+        exp = Categorical([1, 2, 3, 1, 2, 3], ordered=True)
+        tm.assert_categorical_equal(res, exp)
+
+        c1 = Categorical([1, 2, 3, np.nan], ordered=True)
+        c2 = Categorical([3, 2], categories=[1, 2, 3], ordered=True)
+
+        res = union_categoricals([c1, c2])
+        exp = Categorical([1, 2, 3, np.nan, 3, 2], ordered=True)
+        tm.assert_categorical_equal(res, exp)
+
+        c1 = Categorical([1, 2, 3], ordered=True)
+        c2 = Categorical([1, 2, 3], categories=[3, 2, 1], ordered=True)
+
+        msg = "to union ordered Categoricals, all categories must be the same"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2])
+
+    def test_union_categoricals_ignore_order(self):
+        # GH 15219
+        c1 = Categorical([1, 2, 3], ordered=True)
+        c2 = Categorical([1, 2, 3], ordered=False)
+
+        res = union_categoricals([c1, c2], ignore_order=True)
+        exp = Categorical([1, 2, 3, 1, 2, 3])
+        tm.assert_categorical_equal(res, exp)
+
+        msg = "Categorical.ordered must be the same"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2], ignore_order=False)
+
+        res = union_categoricals([c1, c1], ignore_order=True)
+        exp = Categorical([1, 2, 3, 1, 2, 3])
+        tm.assert_categorical_equal(res, exp)
+
+        res = union_categoricals([c1, c1], ignore_order=False)
+        exp = Categorical([1, 2, 3, 1, 2, 3], categories=[1, 2, 3], ordered=True)
+        tm.assert_categorical_equal(res, exp)
+
+        c1 = Categorical([1, 2, 3, np.nan], ordered=True)
+        c2 = Categorical([3, 2], categories=[1, 2, 3], ordered=True)
+
+        res = union_categoricals([c1, c2], ignore_order=True)
+        exp = Categorical([1, 2, 3, np.nan, 3, 2])
+        tm.assert_categorical_equal(res, exp)
+
+        c1 = Categorical([1, 2, 3], ordered=True)
+        c2 = Categorical([1, 2, 3], categories=[3, 2, 1], ordered=True)
+
+        res = union_categoricals([c1, c2], ignore_order=True)
+        exp = Categorical([1, 2, 3, 1, 2, 3])
+        tm.assert_categorical_equal(res, exp)
+
+        res = union_categoricals([c2, c1], ignore_order=True, sort_categories=True)
+        exp = Categorical([1, 2, 3, 1, 2, 3], categories=[1, 2, 3])
+        tm.assert_categorical_equal(res, exp)
+
+        c1 = Categorical([1, 2, 3], ordered=True)
+        c2 = Categorical([4, 5, 6], ordered=True)
+        result = union_categoricals([c1, c2], ignore_order=True)
+        expected = Categorical([1, 2, 3, 4, 5, 6])
+        tm.assert_categorical_equal(result, expected)
+
+        msg = "to union ordered Categoricals, all categories must be the same"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2], ignore_order=False)
+
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2])
+
+    def test_union_categoricals_sort(self):
+        # GH 13846
+        c1 = Categorical(["x", "y", "z"])
+        c2 = Categorical(["a", "b", "c"])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical(
+            ["x", "y", "z", "a", "b", "c"], categories=["a", "b", "c", "x", "y", "z"]
+        )
+        tm.assert_categorical_equal(result, expected)
+
+        # fastpath
+        c1 = Categorical(["a", "b"], categories=["b", "a", "c"])
+        c2 = Categorical(["b", "c"], categories=["b", "a", "c"])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical(["a", "b", "b", "c"], categories=["a", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Categorical(["a", "b"], categories=["c", "a", "b"])
+        c2 = Categorical(["b", "c"], categories=["c", "a", "b"])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical(["a", "b", "b", "c"], categories=["a", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+        # fastpath - skip resort
+        c1 = Categorical(["a", "b"], categories=["a", "b", "c"])
+        c2 = Categorical(["b", "c"], categories=["a", "b", "c"])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical(["a", "b", "b", "c"], categories=["a", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Categorical(["x", np.nan])
+        c2 = Categorical([np.nan, "b"])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical(["x", np.nan, np.nan, "b"], categories=["b", "x"])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Categorical([np.nan])
+        c2 = Categorical([np.nan])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical([np.nan, np.nan])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Categorical([])
+        c2 = Categorical([])
+        result = union_categoricals([c1, c2], sort_categories=True)
+        expected = Categorical([])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Categorical(["b", "a"], categories=["b", "a", "c"], ordered=True)
+        c2 = Categorical(["a", "c"], categories=["b", "a", "c"], ordered=True)
+        msg = "Cannot use sort_categories=True with ordered Categoricals"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, c2], sort_categories=True)
+
+    def test_union_categoricals_sort_false(self):
+        # GH 13846
+        c1 = Categorical(["x", "y", "z"])
+        c2 = Categorical(["a", "b", "c"])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical(
+            ["x", "y", "z", "a", "b", "c"], categories=["x", "y", "z", "a", "b", "c"]
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_fastpath(self):
+        # fastpath
+        c1 = Categorical(["a", "b"], categories=["b", "a", "c"])
+        c2 = Categorical(["b", "c"], categories=["b", "a", "c"])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical(["a", "b", "b", "c"], categories=["b", "a", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_skipresort(self):
+        # fastpath - skip resort
+        c1 = Categorical(["a", "b"], categories=["a", "b", "c"])
+        c2 = Categorical(["b", "c"], categories=["a", "b", "c"])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical(["a", "b", "b", "c"], categories=["a", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_one_nan(self):
+        c1 = Categorical(["x", np.nan])
+        c2 = Categorical([np.nan, "b"])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical(["x", np.nan, np.nan, "b"], categories=["x", "b"])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_only_nan(self):
+        c1 = Categorical([np.nan])
+        c2 = Categorical([np.nan])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical([np.nan, np.nan])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_empty(self):
+        c1 = Categorical([])
+        c2 = Categorical([])
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical([])
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categoricals_sort_false_ordered_true(self):
+        c1 = Categorical(["b", "a"], categories=["b", "a", "c"], ordered=True)
+        c2 = Categorical(["a", "c"], categories=["b", "a", "c"], ordered=True)
+        result = union_categoricals([c1, c2], sort_categories=False)
+        expected = Categorical(
+            ["b", "a", "a", "c"], categories=["b", "a", "c"], ordered=True
+        )
+        tm.assert_categorical_equal(result, expected)
+
+    def test_union_categorical_unwrap(self):
+        # GH 14173
+        c1 = Categorical(["a", "b"])
+        c2 = Series(["b", "c"], dtype="category")
+        result = union_categoricals([c1, c2])
+        expected = Categorical(["a", "b", "b", "c"])
+        tm.assert_categorical_equal(result, expected)
+
+        c2 = CategoricalIndex(c2)
+        result = union_categoricals([c1, c2])
+        tm.assert_categorical_equal(result, expected)
+
+        c1 = Series(c1)
+        result = union_categoricals([c1, c2])
+        tm.assert_categorical_equal(result, expected)
+
+        msg = "all components to combine must be Categorical"
+        with pytest.raises(TypeError, match=msg):
+            union_categoricals([c1, ["a", "b", "c"]])
diff --git a/pandas/pandas/tests/scalar/__init__.py b/pandas/pandas/tests/scalar/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/interval/__init__.py b/pandas/pandas/tests/scalar/interval/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/interval/test_arithmetic.py b/pandas/pandas/tests/scalar/interval/test_arithmetic.py
new file mode 100644
index 0000000000000..603763227cb88
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_arithmetic.py
@@ -0,0 +1,192 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Interval,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestIntervalArithmetic:
+    def test_interval_add(self, closed):
+        interval = Interval(0, 1, closed=closed)
+        expected = Interval(1, 2, closed=closed)
+
+        result = interval + 1
+        assert result == expected
+
+        result = 1 + interval
+        assert result == expected
+
+        result = interval
+        result += 1
+        assert result == expected
+
+        msg = r"unsupported operand type\(s\) for \+"
+        with pytest.raises(TypeError, match=msg):
+            interval + interval
+
+        with pytest.raises(TypeError, match=msg):
+            interval + "foo"
+
+    def test_interval_sub(self, closed):
+        interval = Interval(0, 1, closed=closed)
+        expected = Interval(-1, 0, closed=closed)
+
+        result = interval - 1
+        assert result == expected
+
+        result = interval
+        result -= 1
+        assert result == expected
+
+        msg = r"unsupported operand type\(s\) for -"
+        with pytest.raises(TypeError, match=msg):
+            interval - interval
+
+        with pytest.raises(TypeError, match=msg):
+            interval - "foo"
+
+    def test_interval_mult(self, closed):
+        interval = Interval(0, 1, closed=closed)
+        expected = Interval(0, 2, closed=closed)
+
+        result = interval * 2
+        assert result == expected
+
+        result = 2 * interval
+        assert result == expected
+
+        result = interval
+        result *= 2
+        assert result == expected
+
+        msg = r"unsupported operand type\(s\) for \*"
+        with pytest.raises(TypeError, match=msg):
+            interval * interval
+
+        msg = r"can\'t multiply sequence by non-int"
+        with pytest.raises(TypeError, match=msg):
+            interval * "foo"
+
+    def test_interval_div(self, closed):
+        interval = Interval(0, 1, closed=closed)
+        expected = Interval(0, 0.5, closed=closed)
+
+        result = interval / 2.0
+        assert result == expected
+
+        result = interval
+        result /= 2.0
+        assert result == expected
+
+        msg = r"unsupported operand type\(s\) for /"
+        with pytest.raises(TypeError, match=msg):
+            interval / interval
+
+        with pytest.raises(TypeError, match=msg):
+            interval / "foo"
+
+    def test_interval_floordiv(self, closed):
+        interval = Interval(1, 2, closed=closed)
+        expected = Interval(0, 1, closed=closed)
+
+        result = interval // 2
+        assert result == expected
+
+        result = interval
+        result //= 2
+        assert result == expected
+
+        msg = r"unsupported operand type\(s\) for //"
+        with pytest.raises(TypeError, match=msg):
+            interval // interval
+
+        with pytest.raises(TypeError, match=msg):
+            interval // "foo"
+
+    @pytest.mark.parametrize("method", ["__add__", "__sub__"])
+    @pytest.mark.parametrize(
+        "interval",
+        [
+            Interval(
+                Timestamp("2017-01-01 00:00:00"), Timestamp("2018-01-01 00:00:00")
+            ),
+            Interval(Timedelta(days=7), Timedelta(days=14)),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "delta", [Timedelta(days=7), timedelta(7), np.timedelta64(7, "D")]
+    )
+    def test_time_interval_add_subtract_timedelta(self, interval, delta, method):
+        # https://github.com/pandas-dev/pandas/issues/32023
+        result = getattr(interval, method)(delta)
+        left = getattr(interval.left, method)(delta)
+        right = getattr(interval.right, method)(delta)
+        expected = Interval(left, right)
+
+        assert result == expected
+
+    @pytest.mark.parametrize("interval", [Interval(1, 2), Interval(1.0, 2.0)])
+    @pytest.mark.parametrize(
+        "delta", [Timedelta(days=7), timedelta(7), np.timedelta64(7, "D")]
+    )
+    def test_numeric_interval_add_timedelta_raises(self, interval, delta):
+        # https://github.com/pandas-dev/pandas/issues/32023
+        msg = "|".join(
+            [
+                "unsupported operand",
+                "cannot use operands",
+                "Only numeric, Timestamp and Timedelta endpoints are allowed",
+            ]
+        )
+        with pytest.raises((TypeError, ValueError), match=msg):
+            interval + delta
+
+        with pytest.raises((TypeError, ValueError), match=msg):
+            delta + interval
+
+    @pytest.mark.parametrize("klass", [timedelta, np.timedelta64, Timedelta])
+    def test_timedelta_add_timestamp_interval(self, klass):
+        delta = klass(0)
+        expected = Interval(Timestamp("2020-01-01"), Timestamp("2020-02-01"))
+
+        result = delta + expected
+        assert result == expected
+
+        result = expected + delta
+        assert result == expected
+
+
+class TestIntervalComparisons:
+    def test_interval_equal(self):
+        assert Interval(0, 1) == Interval(0, 1, closed="right")
+        assert Interval(0, 1) != Interval(0, 1, closed="left")
+        assert Interval(0, 1) != 0
+
+    def test_interval_comparison(self):
+        msg = (
+            "'<' not supported between instances of "
+            "'pandas._libs.interval.Interval' and 'int'"
+        )
+        with pytest.raises(TypeError, match=msg):
+            Interval(0, 1) < 2
+
+        assert Interval(0, 1) < Interval(1, 2)
+        assert Interval(0, 1) < Interval(0, 2)
+        assert Interval(0, 1) < Interval(0.5, 1.5)
+        assert Interval(0, 1) <= Interval(0, 1)
+        assert Interval(0, 1) > Interval(-1, 2)
+        assert Interval(0, 1) >= Interval(0, 1)
+
+    def test_equality_comparison_broadcasts_over_array(self):
+        # https://github.com/pandas-dev/pandas/issues/35931
+        interval = Interval(0, 1)
+        arr = np.array([interval, interval])
+        result = interval == arr
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/scalar/interval/test_constructors.py b/pandas/pandas/tests/scalar/interval/test_constructors.py
new file mode 100644
index 0000000000000..a4bc00b923434
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_constructors.py
@@ -0,0 +1,51 @@
+import pytest
+
+from pandas import (
+    Interval,
+    Period,
+    Timestamp,
+)
+
+
+class TestIntervalConstructors:
+    @pytest.mark.parametrize(
+        "left, right",
+        [
+            ("a", "z"),
+            (("a", "b"), ("c", "d")),
+            (list("AB"), list("ab")),
+            (Interval(0, 1), Interval(1, 2)),
+            (Period("2018Q1", freq="Q"), Period("2018Q1", freq="Q")),
+        ],
+    )
+    def test_construct_errors(self, left, right):
+        # GH#23013
+        msg = "Only numeric, Timestamp and Timedelta endpoints are allowed"
+        with pytest.raises(ValueError, match=msg):
+            Interval(left, right)
+
+    def test_constructor_errors(self):
+        msg = "invalid option for 'closed': foo"
+        with pytest.raises(ValueError, match=msg):
+            Interval(0, 1, closed="foo")
+
+        msg = "left side of interval must be <= right side"
+        with pytest.raises(ValueError, match=msg):
+            Interval(1, 0)
+
+    @pytest.mark.parametrize(
+        "tz_left, tz_right", [(None, "UTC"), ("UTC", None), ("UTC", "US/Eastern")]
+    )
+    def test_constructor_errors_tz(self, tz_left, tz_right):
+        # GH#18538
+        left = Timestamp("2017-01-01", tz=tz_left)
+        right = Timestamp("2017-01-02", tz=tz_right)
+
+        if tz_left is None or tz_right is None:
+            error = TypeError
+            msg = "Cannot compare tz-naive and tz-aware timestamps"
+        else:
+            error = ValueError
+            msg = "left and right must have the same time zone"
+        with pytest.raises(error, match=msg):
+            Interval(left, right)
diff --git a/pandas/pandas/tests/scalar/interval/test_contains.py b/pandas/pandas/tests/scalar/interval/test_contains.py
new file mode 100644
index 0000000000000..8dfca117a658b
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_contains.py
@@ -0,0 +1,73 @@
+import pytest
+
+from pandas import (
+    Interval,
+    Timedelta,
+    Timestamp,
+)
+
+
+class TestContains:
+    def test_contains(self):
+        interval = Interval(0, 1)
+        assert 0.5 in interval
+        assert 1 in interval
+        assert 0 not in interval
+
+        interval_both = Interval(0, 1, "both")
+        assert 0 in interval_both
+        assert 1 in interval_both
+
+        interval_neither = Interval(0, 1, closed="neither")
+        assert 0 not in interval_neither
+        assert 0.5 in interval_neither
+        assert 1 not in interval_neither
+
+    def test_contains_interval(self, inclusive_endpoints_fixture):
+        interval1 = Interval(0, 1, "both")
+        interval2 = Interval(0, 1, inclusive_endpoints_fixture)
+        assert interval1 in interval1
+        assert interval2 in interval2
+        assert interval2 in interval1
+        assert interval1 not in interval2 or inclusive_endpoints_fixture == "both"
+
+    def test_contains_infinite_length(self):
+        interval1 = Interval(0, 1, "both")
+        interval2 = Interval(float("-inf"), float("inf"), "neither")
+        assert interval1 in interval2
+        assert interval2 not in interval1
+
+    def test_contains_zero_length(self):
+        interval1 = Interval(0, 1, "both")
+        interval2 = Interval(-1, -1, "both")
+        interval3 = Interval(0.5, 0.5, "both")
+        assert interval2 not in interval1
+        assert interval3 in interval1
+        assert interval2 not in interval3 and interval3 not in interval2
+        assert interval1 not in interval2 and interval1 not in interval3
+
+    @pytest.mark.parametrize(
+        "type1",
+        [
+            (0, 1),
+            (Timestamp(2000, 1, 1, 0), Timestamp(2000, 1, 1, 1)),
+            (Timedelta("0h"), Timedelta("1h")),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "type2",
+        [
+            (0, 1),
+            (Timestamp(2000, 1, 1, 0), Timestamp(2000, 1, 1, 1)),
+            (Timedelta("0h"), Timedelta("1h")),
+        ],
+    )
+    def test_contains_mixed_types(self, type1, type2):
+        interval1 = Interval(*type1)
+        interval2 = Interval(*type2)
+        if type1 == type2:
+            assert interval1 in interval2
+        else:
+            msg = "^'<=' not supported between instances of"
+            with pytest.raises(TypeError, match=msg):
+                interval1 in interval2
diff --git a/pandas/pandas/tests/scalar/interval/test_formats.py b/pandas/pandas/tests/scalar/interval/test_formats.py
new file mode 100644
index 0000000000000..6bf7aa91df3ce
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_formats.py
@@ -0,0 +1,11 @@
+from pandas import Interval
+
+
+def test_interval_repr():
+    interval = Interval(0, 1)
+    assert repr(interval) == "Interval(0, 1, closed='right')"
+    assert str(interval) == "(0, 1]"
+
+    interval_left = Interval(0, 1, closed="left")
+    assert repr(interval_left) == "Interval(0, 1, closed='left')"
+    assert str(interval_left) == "[0, 1)"
diff --git a/pandas/pandas/tests/scalar/interval/test_interval.py b/pandas/pandas/tests/scalar/interval/test_interval.py
new file mode 100644
index 0000000000000..91b31e82f9c52
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_interval.py
@@ -0,0 +1,87 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Interval,
+    Timedelta,
+    Timestamp,
+)
+
+
+@pytest.fixture
+def interval():
+    return Interval(0, 1)
+
+
+class TestInterval:
+    def test_properties(self, interval):
+        assert interval.closed == "right"
+        assert interval.left == 0
+        assert interval.right == 1
+        assert interval.mid == 0.5
+
+    def test_hash(self, interval):
+        # should not raise
+        hash(interval)
+
+    @pytest.mark.parametrize(
+        "left, right, expected",
+        [
+            (0, 5, 5),
+            (-2, 5.5, 7.5),
+            (10, 10, 0),
+            (10, np.inf, np.inf),
+            (-np.inf, -5, np.inf),
+            (-np.inf, np.inf, np.inf),
+            (Timedelta("0 days"), Timedelta("5 days"), Timedelta("5 days")),
+            (Timedelta("10 days"), Timedelta("10 days"), Timedelta("0 days")),
+            (Timedelta("1h10min"), Timedelta("5h5min"), Timedelta("3h55min")),
+            (Timedelta("5s"), Timedelta("1h"), Timedelta("59min55s")),
+        ],
+    )
+    def test_length(self, left, right, expected):
+        # GH 18789
+        iv = Interval(left, right)
+        result = iv.length
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "left, right, expected",
+        [
+            ("2017-01-01", "2017-01-06", "5 days"),
+            ("2017-01-01", "2017-01-01 12:00:00", "12 hours"),
+            ("2017-01-01 12:00", "2017-01-01 12:00:00", "0 days"),
+            ("2017-01-01 12:01", "2017-01-05 17:31:00", "4 days 5 hours 30 min"),
+        ],
+    )
+    @pytest.mark.parametrize("tz", (None, "UTC", "CET", "US/Eastern"))
+    def test_length_timestamp(self, tz, left, right, expected):
+        # GH 18789
+        iv = Interval(Timestamp(left, tz=tz), Timestamp(right, tz=tz))
+        result = iv.length
+        expected = Timedelta(expected)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "left, right",
+        [
+            (0, 1),
+            (Timedelta("0 days"), Timedelta("1 day")),
+            (Timestamp("2018-01-01"), Timestamp("2018-01-02")),
+            (
+                Timestamp("2018-01-01", tz="US/Eastern"),
+                Timestamp("2018-01-02", tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_is_empty(self, left, right, closed):
+        # GH27219
+        # non-empty always return False
+        iv = Interval(left, right, closed)
+        assert iv.is_empty is False
+
+        # same endpoint is empty except when closed='both' (contains one point)
+        iv = Interval(left, left, closed)
+        result = iv.is_empty
+        expected = closed != "both"
+        assert result is expected
diff --git a/pandas/pandas/tests/scalar/interval/test_overlaps.py b/pandas/pandas/tests/scalar/interval/test_overlaps.py
new file mode 100644
index 0000000000000..7fcf59d7bb4af
--- /dev/null
+++ b/pandas/pandas/tests/scalar/interval/test_overlaps.py
@@ -0,0 +1,67 @@
+import pytest
+
+from pandas import (
+    Interval,
+    Timedelta,
+    Timestamp,
+)
+
+
+@pytest.fixture(
+    params=[
+        (Timedelta("0 days"), Timedelta("1 day")),
+        (Timestamp("2018-01-01"), Timedelta("1 day")),
+        (0, 1),
+    ],
+    ids=lambda x: type(x[0]).__name__,
+)
+def start_shift(request):
+    """
+    Fixture for generating intervals of types from a start value and a shift
+    value that can be added to start to generate an endpoint
+    """
+    return request.param
+
+
+class TestOverlaps:
+    def test_overlaps_self(self, start_shift, closed):
+        start, shift = start_shift
+        interval = Interval(start, start + shift, closed)
+        assert interval.overlaps(interval)
+
+    def test_overlaps_nested(self, start_shift, closed, other_closed):
+        start, shift = start_shift
+        interval1 = Interval(start, start + 3 * shift, other_closed)
+        interval2 = Interval(start + shift, start + 2 * shift, closed)
+
+        # nested intervals should always overlap
+        assert interval1.overlaps(interval2)
+
+    def test_overlaps_disjoint(self, start_shift, closed, other_closed):
+        start, shift = start_shift
+        interval1 = Interval(start, start + shift, other_closed)
+        interval2 = Interval(start + 2 * shift, start + 3 * shift, closed)
+
+        # disjoint intervals should never overlap
+        assert not interval1.overlaps(interval2)
+
+    def test_overlaps_endpoint(self, start_shift, closed, other_closed):
+        start, shift = start_shift
+        interval1 = Interval(start, start + shift, other_closed)
+        interval2 = Interval(start + shift, start + 2 * shift, closed)
+
+        # overlap if shared endpoint is closed for both (overlap at a point)
+        result = interval1.overlaps(interval2)
+        expected = interval1.closed_right and interval2.closed_left
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "other",
+        [10, True, "foo", Timedelta("1 day"), Timestamp("2018-01-01")],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_overlaps_invalid_type(self, other):
+        interval = Interval(0, 1)
+        msg = f"`other` must be an Interval, got {type(other).__name__}"
+        with pytest.raises(TypeError, match=msg):
+            interval.overlaps(other)
diff --git a/pandas/pandas/tests/scalar/period/__init__.py b/pandas/pandas/tests/scalar/period/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/period/test_arithmetic.py b/pandas/pandas/tests/scalar/period/test_arithmetic.py
new file mode 100644
index 0000000000000..97e486f9af060
--- /dev/null
+++ b/pandas/pandas/tests/scalar/period/test_arithmetic.py
@@ -0,0 +1,487 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.period import IncompatibleFrequency
+
+from pandas import (
+    NaT,
+    Period,
+    Timedelta,
+    Timestamp,
+    offsets,
+)
+
+
+class TestPeriodArithmetic:
+    def test_add_overflow_raises(self):
+        # GH#55503
+        per = Timestamp.max.to_period("ns")
+
+        msg = "|".join(
+            [
+                "Python int too large to convert to C long",
+                # windows, 32bit linux builds
+                "int too big to convert",
+            ]
+        )
+        with pytest.raises(OverflowError, match=msg):
+            per + 1
+
+        msg = "value too large"
+        with pytest.raises(OverflowError, match=msg):
+            per + Timedelta(1)
+        with pytest.raises(OverflowError, match=msg):
+            per + offsets.Nano(1)
+
+    def test_period_add_integer(self):
+        per1 = Period(freq="D", year=2008, month=1, day=1)
+        per2 = Period(freq="D", year=2008, month=1, day=2)
+        assert per1 + 1 == per2
+        assert 1 + per1 == per2
+
+    def test_period_add_invalid(self):
+        # GH#4731
+        per1 = Period(freq="D", year=2008, month=1, day=1)
+        per2 = Period(freq="D", year=2008, month=1, day=2)
+
+        msg = "|".join(
+            [
+                r"unsupported operand type\(s\)",
+                "can only concatenate str",
+                "must be str, not Period",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            per1 + "str"
+        with pytest.raises(TypeError, match=msg):
+            "str" + per1
+        with pytest.raises(TypeError, match=msg):
+            per1 + per2
+
+    def test_period_sub_period_annual(self):
+        left, right = Period("2011", freq="Y"), Period("2007", freq="Y")
+        result = left - right
+        assert result == 4 * right.freq
+
+        msg = r"Input has different freq=M from Period\(freq=Y-DEC\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            left - Period("2007-01", freq="M")
+
+    def test_period_sub_period(self):
+        per1 = Period("2011-01-01", freq="D")
+        per2 = Period("2011-01-15", freq="D")
+
+        off = per1.freq
+        assert per1 - per2 == -14 * off
+        assert per2 - per1 == 14 * off
+
+        msg = r"Input has different freq=M from Period\(freq=D\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            per1 - Period("2011-02", freq="M")
+
+    @pytest.mark.parametrize("n", [1, 2, 3, 4])
+    def test_sub_n_gt_1_ticks(self, tick_classes, n):
+        # GH#23878
+        p1 = Period("19910905", freq=tick_classes(n))
+        p2 = Period("19920406", freq=tick_classes(n))
+
+        expected = Period(str(p2), freq=p2.freq.base) - Period(
+            str(p1), freq=p1.freq.base
+        )
+
+        assert (p2 - p1) == expected
+
+    @pytest.mark.parametrize("normalize", [True, False])
+    @pytest.mark.parametrize("n", [1, 2, 3, 4])
+    @pytest.mark.parametrize(
+        "offset, kwd_name",
+        [
+            (offsets.YearEnd, "month"),
+            (offsets.QuarterEnd, "startingMonth"),
+            (offsets.MonthEnd, None),
+            (offsets.Week, "weekday"),
+        ],
+    )
+    def test_sub_n_gt_1_offsets(self, offset, kwd_name, n, normalize):
+        # GH#23878
+        kwds = {kwd_name: 3} if kwd_name is not None else {}
+        p1_d = "19910905"
+        p2_d = "19920406"
+        p1 = Period(p1_d, freq=offset(n, normalize, **kwds))
+        p2 = Period(p2_d, freq=offset(n, normalize, **kwds))
+
+        expected = Period(p2_d, freq=p2.freq.base) - Period(p1_d, freq=p1.freq.base)
+
+        assert (p2 - p1) == expected
+
+    def test_period_add_offset(self):
+        # freq is DateOffset
+        for freq in ["Y", "2Y", "3Y"]:
+            per = Period("2011", freq=freq)
+            exp = Period("2013", freq=freq)
+            assert per + offsets.YearEnd(2) == exp
+            assert offsets.YearEnd(2) + per == exp
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(365, "D"),
+                timedelta(365),
+            ]:
+                msg = "Input has different freq|Input cannot be converted to Period"
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per + off
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    off + per
+
+        for freq in ["M", "2M", "3M"]:
+            per = Period("2011-03", freq=freq)
+            exp = Period("2011-05", freq=freq)
+            assert per + offsets.MonthEnd(2) == exp
+            assert offsets.MonthEnd(2) + per == exp
+
+            exp = Period("2012-03", freq=freq)
+            assert per + offsets.MonthEnd(12) == exp
+            assert offsets.MonthEnd(12) + per == exp
+
+            msg = "|".join(
+                [
+                    "Input has different freq",
+                    "Input cannot be converted to Period",
+                ]
+            )
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(365, "D"),
+                timedelta(365),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per + off
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    off + per
+
+        # freq is Tick
+        for freq in ["D", "2D", "3D"]:
+            per = Period("2011-04-01", freq=freq)
+
+            exp = Period("2011-04-06", freq=freq)
+            assert per + offsets.Day(5) == exp
+            assert offsets.Day(5) + per == exp
+
+            exp = Period("2011-04-02", freq=freq)
+            assert per + offsets.Hour(24) == exp
+            assert offsets.Hour(24) + per == exp
+
+            exp = Period("2011-04-03", freq=freq)
+            assert per + np.timedelta64(2, "D") == exp
+            assert np.timedelta64(2, "D") + per == exp
+
+            exp = Period("2011-04-02", freq=freq)
+            assert per + np.timedelta64(3600 * 24, "s") == exp
+            assert np.timedelta64(3600 * 24, "s") + per == exp
+
+            exp = Period("2011-03-30", freq=freq)
+            assert per + timedelta(-2) == exp
+            assert timedelta(-2) + per == exp
+
+            exp = Period("2011-04-03", freq=freq)
+            assert per + timedelta(hours=48) == exp
+            assert timedelta(hours=48) + per == exp
+
+            msg = "|".join(
+                [
+                    "Input has different freq",
+                    "Input cannot be converted to Period",
+                ]
+            )
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(4, "h"),
+                timedelta(hours=23),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per + off
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    off + per
+
+        for freq in ["h", "2h", "3h"]:
+            per = Period("2011-04-01 09:00", freq=freq)
+
+            exp = Period("2011-04-03 09:00", freq=freq)
+            assert per + offsets.Day(2) == exp
+            assert offsets.Day(2) + per == exp
+
+            exp = Period("2011-04-01 12:00", freq=freq)
+            assert per + offsets.Hour(3) == exp
+            assert offsets.Hour(3) + per == exp
+
+            msg = "cannot use operands with types"
+            exp = Period("2011-04-01 12:00", freq=freq)
+            assert per + np.timedelta64(3, "h") == exp
+            assert np.timedelta64(3, "h") + per == exp
+
+            exp = Period("2011-04-01 10:00", freq=freq)
+            assert per + np.timedelta64(3600, "s") == exp
+            assert np.timedelta64(3600, "s") + per == exp
+
+            exp = Period("2011-04-01 11:00", freq=freq)
+            assert per + timedelta(minutes=120) == exp
+            assert timedelta(minutes=120) + per == exp
+
+            exp = Period("2011-04-05 12:00", freq=freq)
+            assert per + timedelta(days=4, minutes=180) == exp
+            assert timedelta(days=4, minutes=180) + per == exp
+
+            msg = "|".join(
+                [
+                    "Input has different freq",
+                    "Input cannot be converted to Period",
+                ]
+            )
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(3200, "s"),
+                timedelta(hours=23, minutes=30),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per + off
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    off + per
+
+    def test_period_sub_offset(self):
+        # freq is DateOffset
+        msg = "|".join(
+            [
+                "Input has different freq",
+                "Input cannot be converted to Period",
+            ]
+        )
+
+        for freq in ["Y", "2Y", "3Y"]:
+            per = Period("2011", freq=freq)
+            assert per - offsets.YearEnd(2) == Period("2009", freq=freq)
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(365, "D"),
+                timedelta(365),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per - off
+
+        for freq in ["M", "2M", "3M"]:
+            per = Period("2011-03", freq=freq)
+            assert per - offsets.MonthEnd(2) == Period("2011-01", freq=freq)
+            assert per - offsets.MonthEnd(12) == Period("2010-03", freq=freq)
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(365, "D"),
+                timedelta(365),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per - off
+
+        # freq is Tick
+        for freq in ["D", "2D", "3D"]:
+            per = Period("2011-04-01", freq=freq)
+            assert per - offsets.Day(5) == Period("2011-03-27", freq=freq)
+            assert per - offsets.Hour(24) == Period("2011-03-31", freq=freq)
+            assert per - np.timedelta64(2, "D") == Period("2011-03-30", freq=freq)
+            assert per - np.timedelta64(3600 * 24, "s") == Period(
+                "2011-03-31", freq=freq
+            )
+            assert per - timedelta(-2) == Period("2011-04-03", freq=freq)
+            assert per - timedelta(hours=48) == Period("2011-03-30", freq=freq)
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(4, "h"),
+                timedelta(hours=23),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per - off
+
+        for freq in ["h", "2h", "3h"]:
+            per = Period("2011-04-01 09:00", freq=freq)
+            assert per - offsets.Day(2) == Period("2011-03-30 09:00", freq=freq)
+            assert per - offsets.Hour(3) == Period("2011-04-01 06:00", freq=freq)
+            assert per - np.timedelta64(3, "h") == Period("2011-04-01 06:00", freq=freq)
+            assert per - np.timedelta64(3600, "s") == Period(
+                "2011-04-01 08:00", freq=freq
+            )
+            assert per - timedelta(minutes=120) == Period("2011-04-01 07:00", freq=freq)
+            assert per - timedelta(days=4, minutes=180) == Period(
+                "2011-03-28 06:00", freq=freq
+            )
+
+            for off in [
+                offsets.YearBegin(2),
+                offsets.MonthBegin(1),
+                offsets.Minute(),
+                np.timedelta64(3200, "s"),
+                timedelta(hours=23, minutes=30),
+            ]:
+                with pytest.raises(IncompatibleFrequency, match=msg):
+                    per - off
+
+    @pytest.mark.parametrize("freq", ["M", "2M", "3M"])
+    def test_period_addsub_nat(self, freq):
+        # GH#13071
+        per = Period("2011-01", freq=freq)
+
+        # For subtraction, NaT is treated as another Period object
+        assert NaT - per is NaT
+        assert per - NaT is NaT
+
+        # For addition, NaT is treated as offset-like
+        assert NaT + per is NaT
+        assert per + NaT is NaT
+
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "m"])
+    def test_period_add_sub_td64_nat(self, unit):
+        # GH#47196
+        per = Period("2022-06-01", "D")
+        nat = np.timedelta64("NaT", unit)
+
+        assert per + nat is NaT
+        assert nat + per is NaT
+        assert per - nat is NaT
+
+        with pytest.raises(TypeError, match="unsupported operand"):
+            nat - per
+
+    def test_period_ops_offset(self):
+        per = Period("2011-04-01", freq="D")
+        result = per + offsets.Day()
+        exp = Period("2011-04-02", freq="D")
+        assert result == exp
+
+        result = per - offsets.Day(2)
+        exp = Period("2011-03-30", freq="D")
+        assert result == exp
+
+        msg = r"Input cannot be converted to Period\(freq=D\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            per + offsets.Hour(2)
+
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            per - offsets.Hour(2)
+
+    def test_period_add_timestamp_raises(self):
+        # GH#17983
+        ts = Timestamp("2017")
+        per = Period("2017", freq="M")
+
+        msg = r"unsupported operand type\(s\) for \+: 'Timestamp' and 'Period'"
+        with pytest.raises(TypeError, match=msg):
+            ts + per
+
+        msg = r"unsupported operand type\(s\) for \+: 'Period' and 'Timestamp'"
+        with pytest.raises(TypeError, match=msg):
+            per + ts
+
+
+class TestPeriodComparisons:
+    def test_period_comparison_same_freq(self):
+        jan = Period("2000-01", "M")
+        feb = Period("2000-02", "M")
+
+        assert not jan == feb
+        assert jan != feb
+        assert jan < feb
+        assert jan <= feb
+        assert not jan > feb
+        assert not jan >= feb
+
+    def test_period_comparison_same_period_different_object(self):
+        # Separate Period objects for the same period
+        left = Period("2000-01", "M")
+        right = Period("2000-01", "M")
+
+        assert left == right
+        assert left >= right
+        assert left <= right
+        assert not left < right
+        assert not left > right
+
+    def test_period_comparison_mismatched_freq(self):
+        jan = Period("2000-01", "M")
+        day = Period("2012-01-01", "D")
+
+        assert not jan == day
+        assert jan != day
+        msg = r"Input has different freq=D from Period\(freq=M\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            jan < day
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            jan <= day
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            jan > day
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            jan >= day
+
+    def test_period_comparison_invalid_type(self):
+        jan = Period("2000-01", "M")
+
+        assert not jan == 1
+        assert jan != 1
+
+        int_or_per = "'(Period|int)'"
+        msg = f"not supported between instances of {int_or_per} and {int_or_per}"
+        for left, right in [(jan, 1), (1, jan)]:
+            with pytest.raises(TypeError, match=msg):
+                left > right
+            with pytest.raises(TypeError, match=msg):
+                left >= right
+            with pytest.raises(TypeError, match=msg):
+                left < right
+            with pytest.raises(TypeError, match=msg):
+                left <= right
+
+    def test_period_comparison_nat(self):
+        per = Period("2011-01-01", freq="D")
+
+        ts = Timestamp("2011-01-01")
+        # confirm Period('NaT') work identical with Timestamp('NaT')
+        for left, right in [
+            (NaT, per),
+            (per, NaT),
+            (NaT, ts),
+            (ts, NaT),
+        ]:
+            assert not left < right
+            assert not left > right
+            assert not left == right
+            assert left != right
+            assert not left <= right
+            assert not left >= right
+
+    @pytest.mark.parametrize(
+        "scalar, expected",
+        ((0, False), (Period("2000-01", "M"), True)),
+    )
+    def test_period_comparison_numpy_zerodim_arr(self, scalar, expected):
+        zerodim_arr = np.array(scalar)
+        per = Period("2000-01", "M")
+
+        assert (per == zerodim_arr) is expected
+        assert (zerodim_arr == per) is expected
diff --git a/pandas/pandas/tests/scalar/period/test_asfreq.py b/pandas/pandas/tests/scalar/period/test_asfreq.py
new file mode 100644
index 0000000000000..0ae5389a3e9b5
--- /dev/null
+++ b/pandas/pandas/tests/scalar/period/test_asfreq.py
@@ -0,0 +1,825 @@
+import pytest
+
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+from pandas.errors import OutOfBoundsDatetime
+
+from pandas import (
+    Period,
+    Timestamp,
+    offsets,
+)
+import pandas._testing as tm
+
+bday_msg = "Period with BDay freq is deprecated"
+
+
+class TestFreqConversion:
+    """Test frequency conversion of date objects"""
+
+    @pytest.mark.filterwarnings("ignore:Period with BDay:FutureWarning")
+    @pytest.mark.parametrize("freq", ["Y", "Q", "M", "W", "B", "D"])
+    def test_asfreq_near_zero(self, freq):
+        # GH#19643, GH#19650
+        per = Period("0001-01-01", freq=freq)
+        tup1 = (per.year, per.hour, per.day)
+
+        prev = per - 1
+        assert prev.ordinal == per.ordinal - 1
+        tup2 = (prev.year, prev.month, prev.day)
+        assert tup2 < tup1
+
+    def test_asfreq_near_zero_weekly(self):
+        # GH#19834
+        per1 = Period("0001-01-01", "D") + 6
+        per2 = Period("0001-01-01", "D") - 6
+        week1 = per1.asfreq("W")
+        week2 = per2.asfreq("W")
+        assert week1 != week2
+        assert week1.asfreq("D", "E") >= per1
+        assert week2.asfreq("D", "S") <= per2
+
+    def test_to_timestamp_out_of_bounds(self):
+        # GH#19643, used to incorrectly give Timestamp in 1754
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            per = Period("0001-01-01", freq="B")
+        msg = "Out of bounds nanosecond timestamp"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+                per.to_timestamp()
+
+    def test_asfreq_corner(self):
+        val = Period(freq="Y", year=2007)
+        result1 = val.asfreq("5min")
+        result2 = val.asfreq("min")
+        expected = Period("2007-12-31 23:59", freq="min")
+        assert result1.ordinal == expected.ordinal
+        assert result1.freqstr == "5min"
+        assert result2.ordinal == expected.ordinal
+        assert result2.freqstr == "min"
+
+    def test_conv_annual(self):
+        # frequency conversion tests: from Annual Frequency
+        msg = INVALID_FREQ_ERR_MSG
+
+        ival_A = Period(freq="Y", year=2007)
+
+        ival_AJAN = Period(freq="Y-JAN", year=2007)
+        ival_AJUN = Period(freq="Y-JUN", year=2007)
+        ival_ANOV = Period(freq="Y-NOV", year=2007)
+
+        ival_A_to_Q_start = Period(freq="Q", year=2007, quarter=1)
+        ival_A_to_Q_end = Period(freq="Q", year=2007, quarter=4)
+        ival_A_to_M_start = Period(freq="M", year=2007, month=1)
+        ival_A_to_M_end = Period(freq="M", year=2007, month=12)
+        ival_A_to_W_start = Period(freq="W", year=2007, month=1, day=1)
+        ival_A_to_W_end = Period(freq="W", year=2007, month=12, day=31)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_A_to_B_start = Period(freq="B", year=2007, month=1, day=1)
+            ival_A_to_B_end = Period(freq="B", year=2007, month=12, day=31)
+        ival_A_to_D_start = Period(freq="D", year=2007, month=1, day=1)
+        ival_A_to_D_end = Period(freq="D", year=2007, month=12, day=31)
+        ival_A_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_A_to_H_end = Period(freq="h", year=2007, month=12, day=31, hour=23)
+        ival_A_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_A_to_T_end = Period(
+            freq="Min", year=2007, month=12, day=31, hour=23, minute=59
+        )
+        ival_A_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_A_to_S_end = Period(
+            freq="s", year=2007, month=12, day=31, hour=23, minute=59, second=59
+        )
+
+        ival_AJAN_to_D_end = Period(freq="D", year=2007, month=1, day=31)
+        ival_AJAN_to_D_start = Period(freq="D", year=2006, month=2, day=1)
+        ival_AJUN_to_D_end = Period(freq="D", year=2007, month=6, day=30)
+        ival_AJUN_to_D_start = Period(freq="D", year=2006, month=7, day=1)
+        ival_ANOV_to_D_end = Period(freq="D", year=2007, month=11, day=30)
+        ival_ANOV_to_D_start = Period(freq="D", year=2006, month=12, day=1)
+
+        assert ival_A.asfreq("Q", "s") == ival_A_to_Q_start
+        assert ival_A.asfreq("Q", "e") == ival_A_to_Q_end
+        assert ival_A.asfreq("M", "s") == ival_A_to_M_start
+        assert ival_A.asfreq("M", "E") == ival_A_to_M_end
+        assert ival_A.asfreq("W", "s") == ival_A_to_W_start
+        assert ival_A.asfreq("W", "E") == ival_A_to_W_end
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_A.asfreq("B", "s") == ival_A_to_B_start
+            assert ival_A.asfreq("B", "E") == ival_A_to_B_end
+        assert ival_A.asfreq("D", "s") == ival_A_to_D_start
+        assert ival_A.asfreq("D", "E") == ival_A_to_D_end
+        with pytest.raises(ValueError, match=msg):
+            assert ival_A.asfreq("H", "s") == ival_A_to_H_start
+            assert ival_A.asfreq("H", "E") == ival_A_to_H_end
+        assert ival_A.asfreq("min", "s") == ival_A_to_T_start
+        assert ival_A.asfreq("min", "E") == ival_A_to_T_end
+        with pytest.raises(ValueError, match=msg):
+            assert ival_A.asfreq("T", "s") == ival_A_to_T_start
+            assert ival_A.asfreq("T", "E") == ival_A_to_T_end
+            assert ival_A.asfreq("S", "S") == ival_A_to_S_start
+            assert ival_A.asfreq("S", "E") == ival_A_to_S_end
+
+        assert ival_AJAN.asfreq("D", "s") == ival_AJAN_to_D_start
+        assert ival_AJAN.asfreq("D", "E") == ival_AJAN_to_D_end
+
+        assert ival_AJUN.asfreq("D", "s") == ival_AJUN_to_D_start
+        assert ival_AJUN.asfreq("D", "E") == ival_AJUN_to_D_end
+
+        assert ival_ANOV.asfreq("D", "s") == ival_ANOV_to_D_start
+        assert ival_ANOV.asfreq("D", "E") == ival_ANOV_to_D_end
+
+        assert ival_A.asfreq("Y") == ival_A
+
+    def test_conv_quarterly(self):
+        # frequency conversion tests: from Quarterly Frequency
+
+        ival_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_Q_end_of_year = Period(freq="Q", year=2007, quarter=4)
+
+        ival_QEJAN = Period(freq="Q-JAN", year=2007, quarter=1)
+        ival_QEJUN = Period(freq="Q-JUN", year=2007, quarter=1)
+
+        ival_Q_to_A = Period(freq="Y", year=2007)
+        ival_Q_to_M_start = Period(freq="M", year=2007, month=1)
+        ival_Q_to_M_end = Period(freq="M", year=2007, month=3)
+        ival_Q_to_W_start = Period(freq="W", year=2007, month=1, day=1)
+        ival_Q_to_W_end = Period(freq="W", year=2007, month=3, day=31)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_Q_to_B_start = Period(freq="B", year=2007, month=1, day=1)
+            ival_Q_to_B_end = Period(freq="B", year=2007, month=3, day=30)
+        ival_Q_to_D_start = Period(freq="D", year=2007, month=1, day=1)
+        ival_Q_to_D_end = Period(freq="D", year=2007, month=3, day=31)
+        ival_Q_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_Q_to_H_end = Period(freq="h", year=2007, month=3, day=31, hour=23)
+        ival_Q_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_Q_to_T_end = Period(
+            freq="Min", year=2007, month=3, day=31, hour=23, minute=59
+        )
+        ival_Q_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_Q_to_S_end = Period(
+            freq="s", year=2007, month=3, day=31, hour=23, minute=59, second=59
+        )
+
+        ival_QEJAN_to_D_start = Period(freq="D", year=2006, month=2, day=1)
+        ival_QEJAN_to_D_end = Period(freq="D", year=2006, month=4, day=30)
+
+        ival_QEJUN_to_D_start = Period(freq="D", year=2006, month=7, day=1)
+        ival_QEJUN_to_D_end = Period(freq="D", year=2006, month=9, day=30)
+
+        assert ival_Q.asfreq("Y") == ival_Q_to_A
+        assert ival_Q_end_of_year.asfreq("Y") == ival_Q_to_A
+
+        assert ival_Q.asfreq("M", "s") == ival_Q_to_M_start
+        assert ival_Q.asfreq("M", "E") == ival_Q_to_M_end
+        assert ival_Q.asfreq("W", "s") == ival_Q_to_W_start
+        assert ival_Q.asfreq("W", "E") == ival_Q_to_W_end
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_Q.asfreq("B", "s") == ival_Q_to_B_start
+            assert ival_Q.asfreq("B", "E") == ival_Q_to_B_end
+        assert ival_Q.asfreq("D", "s") == ival_Q_to_D_start
+        assert ival_Q.asfreq("D", "E") == ival_Q_to_D_end
+        assert ival_Q.asfreq("h", "s") == ival_Q_to_H_start
+        assert ival_Q.asfreq("h", "E") == ival_Q_to_H_end
+        assert ival_Q.asfreq("Min", "s") == ival_Q_to_T_start
+        assert ival_Q.asfreq("Min", "E") == ival_Q_to_T_end
+        assert ival_Q.asfreq("s", "s") == ival_Q_to_S_start
+        assert ival_Q.asfreq("s", "E") == ival_Q_to_S_end
+
+        assert ival_QEJAN.asfreq("D", "s") == ival_QEJAN_to_D_start
+        assert ival_QEJAN.asfreq("D", "E") == ival_QEJAN_to_D_end
+        assert ival_QEJUN.asfreq("D", "s") == ival_QEJUN_to_D_start
+        assert ival_QEJUN.asfreq("D", "E") == ival_QEJUN_to_D_end
+
+        assert ival_Q.asfreq("Q") == ival_Q
+
+    def test_conv_monthly(self):
+        # frequency conversion tests: from Monthly Frequency
+
+        ival_M = Period(freq="M", year=2007, month=1)
+        ival_M_end_of_year = Period(freq="M", year=2007, month=12)
+        ival_M_end_of_quarter = Period(freq="M", year=2007, month=3)
+        ival_M_to_A = Period(freq="Y", year=2007)
+        ival_M_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_M_to_W_start = Period(freq="W", year=2007, month=1, day=1)
+        ival_M_to_W_end = Period(freq="W", year=2007, month=1, day=31)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_M_to_B_start = Period(freq="B", year=2007, month=1, day=1)
+            ival_M_to_B_end = Period(freq="B", year=2007, month=1, day=31)
+        ival_M_to_D_start = Period(freq="D", year=2007, month=1, day=1)
+        ival_M_to_D_end = Period(freq="D", year=2007, month=1, day=31)
+        ival_M_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_M_to_H_end = Period(freq="h", year=2007, month=1, day=31, hour=23)
+        ival_M_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_M_to_T_end = Period(
+            freq="Min", year=2007, month=1, day=31, hour=23, minute=59
+        )
+        ival_M_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_M_to_S_end = Period(
+            freq="s", year=2007, month=1, day=31, hour=23, minute=59, second=59
+        )
+
+        assert ival_M.asfreq("Y") == ival_M_to_A
+        assert ival_M_end_of_year.asfreq("Y") == ival_M_to_A
+        assert ival_M.asfreq("Q") == ival_M_to_Q
+        assert ival_M_end_of_quarter.asfreq("Q") == ival_M_to_Q
+
+        assert ival_M.asfreq("W", "s") == ival_M_to_W_start
+        assert ival_M.asfreq("W", "E") == ival_M_to_W_end
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_M.asfreq("B", "s") == ival_M_to_B_start
+            assert ival_M.asfreq("B", "E") == ival_M_to_B_end
+        assert ival_M.asfreq("D", "s") == ival_M_to_D_start
+        assert ival_M.asfreq("D", "E") == ival_M_to_D_end
+        assert ival_M.asfreq("h", "s") == ival_M_to_H_start
+        assert ival_M.asfreq("h", "E") == ival_M_to_H_end
+        assert ival_M.asfreq("Min", "s") == ival_M_to_T_start
+        assert ival_M.asfreq("Min", "E") == ival_M_to_T_end
+        assert ival_M.asfreq("s", "s") == ival_M_to_S_start
+        assert ival_M.asfreq("s", "E") == ival_M_to_S_end
+
+        assert ival_M.asfreq("M") == ival_M
+
+    def test_conv_weekly(self):
+        # frequency conversion tests: from Weekly Frequency
+        ival_W = Period(freq="W", year=2007, month=1, day=1)
+
+        ival_WSUN = Period(freq="W", year=2007, month=1, day=7)
+        ival_WSAT = Period(freq="W-SAT", year=2007, month=1, day=6)
+        ival_WFRI = Period(freq="W-FRI", year=2007, month=1, day=5)
+        ival_WTHU = Period(freq="W-THU", year=2007, month=1, day=4)
+        ival_WWED = Period(freq="W-WED", year=2007, month=1, day=3)
+        ival_WTUE = Period(freq="W-TUE", year=2007, month=1, day=2)
+        ival_WMON = Period(freq="W-MON", year=2007, month=1, day=1)
+
+        ival_WSUN_to_D_start = Period(freq="D", year=2007, month=1, day=1)
+        ival_WSUN_to_D_end = Period(freq="D", year=2007, month=1, day=7)
+        ival_WSAT_to_D_start = Period(freq="D", year=2006, month=12, day=31)
+        ival_WSAT_to_D_end = Period(freq="D", year=2007, month=1, day=6)
+        ival_WFRI_to_D_start = Period(freq="D", year=2006, month=12, day=30)
+        ival_WFRI_to_D_end = Period(freq="D", year=2007, month=1, day=5)
+        ival_WTHU_to_D_start = Period(freq="D", year=2006, month=12, day=29)
+        ival_WTHU_to_D_end = Period(freq="D", year=2007, month=1, day=4)
+        ival_WWED_to_D_start = Period(freq="D", year=2006, month=12, day=28)
+        ival_WWED_to_D_end = Period(freq="D", year=2007, month=1, day=3)
+        ival_WTUE_to_D_start = Period(freq="D", year=2006, month=12, day=27)
+        ival_WTUE_to_D_end = Period(freq="D", year=2007, month=1, day=2)
+        ival_WMON_to_D_start = Period(freq="D", year=2006, month=12, day=26)
+        ival_WMON_to_D_end = Period(freq="D", year=2007, month=1, day=1)
+
+        ival_W_end_of_year = Period(freq="W", year=2007, month=12, day=31)
+        ival_W_end_of_quarter = Period(freq="W", year=2007, month=3, day=31)
+        ival_W_end_of_month = Period(freq="W", year=2007, month=1, day=31)
+        ival_W_to_A = Period(freq="Y", year=2007)
+        ival_W_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_W_to_M = Period(freq="M", year=2007, month=1)
+
+        if Period(freq="D", year=2007, month=12, day=31).weekday == 6:
+            ival_W_to_A_end_of_year = Period(freq="Y", year=2007)
+        else:
+            ival_W_to_A_end_of_year = Period(freq="Y", year=2008)
+
+        if Period(freq="D", year=2007, month=3, day=31).weekday == 6:
+            ival_W_to_Q_end_of_quarter = Period(freq="Q", year=2007, quarter=1)
+        else:
+            ival_W_to_Q_end_of_quarter = Period(freq="Q", year=2007, quarter=2)
+
+        if Period(freq="D", year=2007, month=1, day=31).weekday == 6:
+            ival_W_to_M_end_of_month = Period(freq="M", year=2007, month=1)
+        else:
+            ival_W_to_M_end_of_month = Period(freq="M", year=2007, month=2)
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_W_to_B_start = Period(freq="B", year=2007, month=1, day=1)
+            ival_W_to_B_end = Period(freq="B", year=2007, month=1, day=5)
+        ival_W_to_D_start = Period(freq="D", year=2007, month=1, day=1)
+        ival_W_to_D_end = Period(freq="D", year=2007, month=1, day=7)
+        ival_W_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_W_to_H_end = Period(freq="h", year=2007, month=1, day=7, hour=23)
+        ival_W_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_W_to_T_end = Period(
+            freq="Min", year=2007, month=1, day=7, hour=23, minute=59
+        )
+        ival_W_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_W_to_S_end = Period(
+            freq="s", year=2007, month=1, day=7, hour=23, minute=59, second=59
+        )
+
+        assert ival_W.asfreq("Y") == ival_W_to_A
+        assert ival_W_end_of_year.asfreq("Y") == ival_W_to_A_end_of_year
+
+        assert ival_W.asfreq("Q") == ival_W_to_Q
+        assert ival_W_end_of_quarter.asfreq("Q") == ival_W_to_Q_end_of_quarter
+
+        assert ival_W.asfreq("M") == ival_W_to_M
+        assert ival_W_end_of_month.asfreq("M") == ival_W_to_M_end_of_month
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_W.asfreq("B", "s") == ival_W_to_B_start
+            assert ival_W.asfreq("B", "E") == ival_W_to_B_end
+
+        assert ival_W.asfreq("D", "s") == ival_W_to_D_start
+        assert ival_W.asfreq("D", "E") == ival_W_to_D_end
+
+        assert ival_WSUN.asfreq("D", "s") == ival_WSUN_to_D_start
+        assert ival_WSUN.asfreq("D", "E") == ival_WSUN_to_D_end
+        assert ival_WSAT.asfreq("D", "s") == ival_WSAT_to_D_start
+        assert ival_WSAT.asfreq("D", "E") == ival_WSAT_to_D_end
+        assert ival_WFRI.asfreq("D", "s") == ival_WFRI_to_D_start
+        assert ival_WFRI.asfreq("D", "E") == ival_WFRI_to_D_end
+        assert ival_WTHU.asfreq("D", "s") == ival_WTHU_to_D_start
+        assert ival_WTHU.asfreq("D", "E") == ival_WTHU_to_D_end
+        assert ival_WWED.asfreq("D", "s") == ival_WWED_to_D_start
+        assert ival_WWED.asfreq("D", "E") == ival_WWED_to_D_end
+        assert ival_WTUE.asfreq("D", "s") == ival_WTUE_to_D_start
+        assert ival_WTUE.asfreq("D", "E") == ival_WTUE_to_D_end
+        assert ival_WMON.asfreq("D", "s") == ival_WMON_to_D_start
+        assert ival_WMON.asfreq("D", "E") == ival_WMON_to_D_end
+
+        assert ival_W.asfreq("h", "s") == ival_W_to_H_start
+        assert ival_W.asfreq("h", "E") == ival_W_to_H_end
+        assert ival_W.asfreq("Min", "s") == ival_W_to_T_start
+        assert ival_W.asfreq("Min", "E") == ival_W_to_T_end
+        assert ival_W.asfreq("s", "s") == ival_W_to_S_start
+        assert ival_W.asfreq("s", "E") == ival_W_to_S_end
+
+        assert ival_W.asfreq("W") == ival_W
+
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            ival_W.asfreq("WK")
+
+    def test_conv_weekly_legacy(self):
+        # frequency conversion tests: from Weekly Frequency
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK", year=2007, month=1, day=1)
+
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-SAT", year=2007, month=1, day=6)
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-FRI", year=2007, month=1, day=5)
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-THU", year=2007, month=1, day=4)
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-WED", year=2007, month=1, day=3)
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-TUE", year=2007, month=1, day=2)
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK-MON", year=2007, month=1, day=1)
+
+    def test_conv_business(self):
+        # frequency conversion tests: from Business Frequency"
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_B = Period(freq="B", year=2007, month=1, day=1)
+            ival_B_end_of_year = Period(freq="B", year=2007, month=12, day=31)
+            ival_B_end_of_quarter = Period(freq="B", year=2007, month=3, day=30)
+            ival_B_end_of_month = Period(freq="B", year=2007, month=1, day=31)
+            ival_B_end_of_week = Period(freq="B", year=2007, month=1, day=5)
+
+        ival_B_to_A = Period(freq="Y", year=2007)
+        ival_B_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_B_to_M = Period(freq="M", year=2007, month=1)
+        ival_B_to_W = Period(freq="W", year=2007, month=1, day=7)
+        ival_B_to_D = Period(freq="D", year=2007, month=1, day=1)
+        ival_B_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_B_to_H_end = Period(freq="h", year=2007, month=1, day=1, hour=23)
+        ival_B_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_B_to_T_end = Period(
+            freq="Min", year=2007, month=1, day=1, hour=23, minute=59
+        )
+        ival_B_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_B_to_S_end = Period(
+            freq="s", year=2007, month=1, day=1, hour=23, minute=59, second=59
+        )
+
+        assert ival_B.asfreq("Y") == ival_B_to_A
+        assert ival_B_end_of_year.asfreq("Y") == ival_B_to_A
+        assert ival_B.asfreq("Q") == ival_B_to_Q
+        assert ival_B_end_of_quarter.asfreq("Q") == ival_B_to_Q
+        assert ival_B.asfreq("M") == ival_B_to_M
+        assert ival_B_end_of_month.asfreq("M") == ival_B_to_M
+        assert ival_B.asfreq("W") == ival_B_to_W
+        assert ival_B_end_of_week.asfreq("W") == ival_B_to_W
+
+        assert ival_B.asfreq("D") == ival_B_to_D
+
+        assert ival_B.asfreq("h", "s") == ival_B_to_H_start
+        assert ival_B.asfreq("h", "E") == ival_B_to_H_end
+        assert ival_B.asfreq("Min", "s") == ival_B_to_T_start
+        assert ival_B.asfreq("Min", "E") == ival_B_to_T_end
+        assert ival_B.asfreq("s", "s") == ival_B_to_S_start
+        assert ival_B.asfreq("s", "E") == ival_B_to_S_end
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_B.asfreq("B") == ival_B
+
+    def test_conv_daily(self):
+        # frequency conversion tests: from Business Frequency"
+
+        ival_D = Period(freq="D", year=2007, month=1, day=1)
+        ival_D_end_of_year = Period(freq="D", year=2007, month=12, day=31)
+        ival_D_end_of_quarter = Period(freq="D", year=2007, month=3, day=31)
+        ival_D_end_of_month = Period(freq="D", year=2007, month=1, day=31)
+        ival_D_end_of_week = Period(freq="D", year=2007, month=1, day=7)
+
+        ival_D_friday = Period(freq="D", year=2007, month=1, day=5)
+        ival_D_saturday = Period(freq="D", year=2007, month=1, day=6)
+        ival_D_sunday = Period(freq="D", year=2007, month=1, day=7)
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_B_friday = Period(freq="B", year=2007, month=1, day=5)
+            ival_B_monday = Period(freq="B", year=2007, month=1, day=8)
+
+        ival_D_to_A = Period(freq="Y", year=2007)
+
+        ival_Deoq_to_AJAN = Period(freq="Y-JAN", year=2008)
+        ival_Deoq_to_AJUN = Period(freq="Y-JUN", year=2007)
+        ival_Deoq_to_ADEC = Period(freq="Y-DEC", year=2007)
+
+        ival_D_to_QEJAN = Period(freq="Q-JAN", year=2007, quarter=4)
+        ival_D_to_QEJUN = Period(freq="Q-JUN", year=2007, quarter=3)
+        ival_D_to_QEDEC = Period(freq="Q-DEC", year=2007, quarter=1)
+
+        ival_D_to_M = Period(freq="M", year=2007, month=1)
+        ival_D_to_W = Period(freq="W", year=2007, month=1, day=7)
+
+        ival_D_to_H_start = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_D_to_H_end = Period(freq="h", year=2007, month=1, day=1, hour=23)
+        ival_D_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_D_to_T_end = Period(
+            freq="Min", year=2007, month=1, day=1, hour=23, minute=59
+        )
+        ival_D_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_D_to_S_end = Period(
+            freq="s", year=2007, month=1, day=1, hour=23, minute=59, second=59
+        )
+
+        assert ival_D.asfreq("Y") == ival_D_to_A
+
+        assert ival_D_end_of_quarter.asfreq("Y-JAN") == ival_Deoq_to_AJAN
+        assert ival_D_end_of_quarter.asfreq("Y-JUN") == ival_Deoq_to_AJUN
+        assert ival_D_end_of_quarter.asfreq("Y-DEC") == ival_Deoq_to_ADEC
+
+        assert ival_D_end_of_year.asfreq("Y") == ival_D_to_A
+        assert ival_D_end_of_quarter.asfreq("Q") == ival_D_to_QEDEC
+        assert ival_D.asfreq("Q-JAN") == ival_D_to_QEJAN
+        assert ival_D.asfreq("Q-JUN") == ival_D_to_QEJUN
+        assert ival_D.asfreq("Q-DEC") == ival_D_to_QEDEC
+        assert ival_D.asfreq("M") == ival_D_to_M
+        assert ival_D_end_of_month.asfreq("M") == ival_D_to_M
+        assert ival_D.asfreq("W") == ival_D_to_W
+        assert ival_D_end_of_week.asfreq("W") == ival_D_to_W
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_D_friday.asfreq("B") == ival_B_friday
+            assert ival_D_saturday.asfreq("B", "s") == ival_B_friday
+            assert ival_D_saturday.asfreq("B", "E") == ival_B_monday
+            assert ival_D_sunday.asfreq("B", "s") == ival_B_friday
+            assert ival_D_sunday.asfreq("B", "E") == ival_B_monday
+
+        assert ival_D.asfreq("h", "s") == ival_D_to_H_start
+        assert ival_D.asfreq("h", "E") == ival_D_to_H_end
+        assert ival_D.asfreq("Min", "s") == ival_D_to_T_start
+        assert ival_D.asfreq("Min", "E") == ival_D_to_T_end
+        assert ival_D.asfreq("s", "s") == ival_D_to_S_start
+        assert ival_D.asfreq("s", "E") == ival_D_to_S_end
+
+        assert ival_D.asfreq("D") == ival_D
+
+    def test_conv_hourly(self):
+        # frequency conversion tests: from Hourly Frequency"
+
+        ival_H = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_H_end_of_year = Period(freq="h", year=2007, month=12, day=31, hour=23)
+        ival_H_end_of_quarter = Period(freq="h", year=2007, month=3, day=31, hour=23)
+        ival_H_end_of_month = Period(freq="h", year=2007, month=1, day=31, hour=23)
+        ival_H_end_of_week = Period(freq="h", year=2007, month=1, day=7, hour=23)
+        ival_H_end_of_day = Period(freq="h", year=2007, month=1, day=1, hour=23)
+        ival_H_end_of_bus = Period(freq="h", year=2007, month=1, day=1, hour=23)
+
+        ival_H_to_A = Period(freq="Y", year=2007)
+        ival_H_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_H_to_M = Period(freq="M", year=2007, month=1)
+        ival_H_to_W = Period(freq="W", year=2007, month=1, day=7)
+        ival_H_to_D = Period(freq="D", year=2007, month=1, day=1)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_H_to_B = Period(freq="B", year=2007, month=1, day=1)
+
+        ival_H_to_T_start = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0
+        )
+        ival_H_to_T_end = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=59
+        )
+        ival_H_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_H_to_S_end = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=59, second=59
+        )
+
+        assert ival_H.asfreq("Y") == ival_H_to_A
+        assert ival_H_end_of_year.asfreq("Y") == ival_H_to_A
+        assert ival_H.asfreq("Q") == ival_H_to_Q
+        assert ival_H_end_of_quarter.asfreq("Q") == ival_H_to_Q
+        assert ival_H.asfreq("M") == ival_H_to_M
+        assert ival_H_end_of_month.asfreq("M") == ival_H_to_M
+        assert ival_H.asfreq("W") == ival_H_to_W
+        assert ival_H_end_of_week.asfreq("W") == ival_H_to_W
+        assert ival_H.asfreq("D") == ival_H_to_D
+        assert ival_H_end_of_day.asfreq("D") == ival_H_to_D
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_H.asfreq("B") == ival_H_to_B
+            assert ival_H_end_of_bus.asfreq("B") == ival_H_to_B
+
+        assert ival_H.asfreq("Min", "s") == ival_H_to_T_start
+        assert ival_H.asfreq("Min", "E") == ival_H_to_T_end
+        assert ival_H.asfreq("s", "s") == ival_H_to_S_start
+        assert ival_H.asfreq("s", "E") == ival_H_to_S_end
+
+        assert ival_H.asfreq("h") == ival_H
+
+    def test_conv_minutely(self):
+        # frequency conversion tests: from Minutely Frequency"
+
+        ival_T = Period(freq="Min", year=2007, month=1, day=1, hour=0, minute=0)
+        ival_T_end_of_year = Period(
+            freq="Min", year=2007, month=12, day=31, hour=23, minute=59
+        )
+        ival_T_end_of_quarter = Period(
+            freq="Min", year=2007, month=3, day=31, hour=23, minute=59
+        )
+        ival_T_end_of_month = Period(
+            freq="Min", year=2007, month=1, day=31, hour=23, minute=59
+        )
+        ival_T_end_of_week = Period(
+            freq="Min", year=2007, month=1, day=7, hour=23, minute=59
+        )
+        ival_T_end_of_day = Period(
+            freq="Min", year=2007, month=1, day=1, hour=23, minute=59
+        )
+        ival_T_end_of_bus = Period(
+            freq="Min", year=2007, month=1, day=1, hour=23, minute=59
+        )
+        ival_T_end_of_hour = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=59
+        )
+
+        ival_T_to_A = Period(freq="Y", year=2007)
+        ival_T_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_T_to_M = Period(freq="M", year=2007, month=1)
+        ival_T_to_W = Period(freq="W", year=2007, month=1, day=7)
+        ival_T_to_D = Period(freq="D", year=2007, month=1, day=1)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_T_to_B = Period(freq="B", year=2007, month=1, day=1)
+        ival_T_to_H = Period(freq="h", year=2007, month=1, day=1, hour=0)
+
+        ival_T_to_S_start = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        ival_T_to_S_end = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=59
+        )
+
+        assert ival_T.asfreq("Y") == ival_T_to_A
+        assert ival_T_end_of_year.asfreq("Y") == ival_T_to_A
+        assert ival_T.asfreq("Q") == ival_T_to_Q
+        assert ival_T_end_of_quarter.asfreq("Q") == ival_T_to_Q
+        assert ival_T.asfreq("M") == ival_T_to_M
+        assert ival_T_end_of_month.asfreq("M") == ival_T_to_M
+        assert ival_T.asfreq("W") == ival_T_to_W
+        assert ival_T_end_of_week.asfreq("W") == ival_T_to_W
+        assert ival_T.asfreq("D") == ival_T_to_D
+        assert ival_T_end_of_day.asfreq("D") == ival_T_to_D
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_T.asfreq("B") == ival_T_to_B
+            assert ival_T_end_of_bus.asfreq("B") == ival_T_to_B
+        assert ival_T.asfreq("h") == ival_T_to_H
+        assert ival_T_end_of_hour.asfreq("h") == ival_T_to_H
+
+        assert ival_T.asfreq("s", "s") == ival_T_to_S_start
+        assert ival_T.asfreq("s", "E") == ival_T_to_S_end
+
+        assert ival_T.asfreq("Min") == ival_T
+
+    def test_conv_secondly(self):
+        # frequency conversion tests: from Secondly Frequency"
+
+        ival_S = Period(freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=0)
+        ival_S_end_of_year = Period(
+            freq="s", year=2007, month=12, day=31, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_quarter = Period(
+            freq="s", year=2007, month=3, day=31, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_month = Period(
+            freq="s", year=2007, month=1, day=31, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_week = Period(
+            freq="s", year=2007, month=1, day=7, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_day = Period(
+            freq="s", year=2007, month=1, day=1, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_bus = Period(
+            freq="s", year=2007, month=1, day=1, hour=23, minute=59, second=59
+        )
+        ival_S_end_of_hour = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=59, second=59
+        )
+        ival_S_end_of_minute = Period(
+            freq="s", year=2007, month=1, day=1, hour=0, minute=0, second=59
+        )
+
+        ival_S_to_A = Period(freq="Y", year=2007)
+        ival_S_to_Q = Period(freq="Q", year=2007, quarter=1)
+        ival_S_to_M = Period(freq="M", year=2007, month=1)
+        ival_S_to_W = Period(freq="W", year=2007, month=1, day=7)
+        ival_S_to_D = Period(freq="D", year=2007, month=1, day=1)
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            ival_S_to_B = Period(freq="B", year=2007, month=1, day=1)
+        ival_S_to_H = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        ival_S_to_T = Period(freq="Min", year=2007, month=1, day=1, hour=0, minute=0)
+
+        assert ival_S.asfreq("Y") == ival_S_to_A
+        assert ival_S_end_of_year.asfreq("Y") == ival_S_to_A
+        assert ival_S.asfreq("Q") == ival_S_to_Q
+        assert ival_S_end_of_quarter.asfreq("Q") == ival_S_to_Q
+        assert ival_S.asfreq("M") == ival_S_to_M
+        assert ival_S_end_of_month.asfreq("M") == ival_S_to_M
+        assert ival_S.asfreq("W") == ival_S_to_W
+        assert ival_S_end_of_week.asfreq("W") == ival_S_to_W
+        assert ival_S.asfreq("D") == ival_S_to_D
+        assert ival_S_end_of_day.asfreq("D") == ival_S_to_D
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert ival_S.asfreq("B") == ival_S_to_B
+            assert ival_S_end_of_bus.asfreq("B") == ival_S_to_B
+        assert ival_S.asfreq("h") == ival_S_to_H
+        assert ival_S_end_of_hour.asfreq("h") == ival_S_to_H
+        assert ival_S.asfreq("Min") == ival_S_to_T
+        assert ival_S_end_of_minute.asfreq("Min") == ival_S_to_T
+
+        assert ival_S.asfreq("s") == ival_S
+
+    def test_conv_microsecond(self):
+        # GH#31475 Avoid floating point errors dropping the start_time to
+        #  before the beginning of the Period
+        per = Period("2020-01-30 15:57:27.576166", freq="us")
+        assert per.ordinal == 1580399847576166
+
+        start = per.start_time
+        expected = Timestamp("2020-01-30 15:57:27.576166")
+        assert start == expected
+        assert start._value == per.ordinal * 1000
+
+        per2 = Period("2300-01-01", "us")
+        msg = "2300-01-01"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            per2.start_time
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            per2.end_time
+
+    def test_asfreq_mult(self):
+        # normal freq to mult freq
+        p = Period(freq="Y", year=2007)
+        # ordinal will not change
+        for freq in ["3Y", offsets.YearEnd(3)]:
+            result = p.asfreq(freq)
+            expected = Period("2007", freq="3Y")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+        # ordinal will not change
+        for freq in ["3Y", offsets.YearEnd(3)]:
+            result = p.asfreq(freq, how="S")
+            expected = Period("2007", freq="3Y")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+
+        # mult freq to normal freq
+        p = Period(freq="3Y", year=2007)
+        # ordinal will change because how=E is the default
+        for freq in ["Y", offsets.YearEnd()]:
+            result = p.asfreq(freq)
+            expected = Period("2009", freq="Y")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+        # ordinal will not change
+        for freq in ["Y", offsets.YearEnd()]:
+            result = p.asfreq(freq, how="s")
+            expected = Period("2007", freq="Y")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+
+        p = Period(freq="Y", year=2007)
+        for freq in ["2M", offsets.MonthEnd(2)]:
+            result = p.asfreq(freq)
+            expected = Period("2007-12", freq="2M")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+        for freq in ["2M", offsets.MonthEnd(2)]:
+            result = p.asfreq(freq, how="s")
+            expected = Period("2007-01", freq="2M")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+
+        p = Period(freq="3Y", year=2007)
+        for freq in ["2M", offsets.MonthEnd(2)]:
+            result = p.asfreq(freq)
+            expected = Period("2009-12", freq="2M")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+        for freq in ["2M", offsets.MonthEnd(2)]:
+            result = p.asfreq(freq, how="s")
+            expected = Period("2007-01", freq="2M")
+
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+
+    def test_asfreq_combined(self):
+        # normal freq to combined freq
+        p = Period("2007", freq="h")
+
+        # ordinal will not change
+        expected = Period("2007", freq="25h")
+        for freq, how in zip(["1D1h", "1h1D"], ["E", "S"]):
+            result = p.asfreq(freq, how=how)
+            assert result == expected
+            assert result.ordinal == expected.ordinal
+            assert result.freq == expected.freq
+
+        # combined freq to normal freq
+        p1 = Period(freq="1D1h", year=2007)
+        p2 = Period(freq="1h1D", year=2007)
+
+        # ordinal will change because how=E is the default
+        result1 = p1.asfreq("h")
+        result2 = p2.asfreq("h")
+        expected = Period("2007-01-02", freq="h")
+        assert result1 == expected
+        assert result1.ordinal == expected.ordinal
+        assert result1.freq == expected.freq
+        assert result2 == expected
+        assert result2.ordinal == expected.ordinal
+        assert result2.freq == expected.freq
+
+        # ordinal will not change
+        result1 = p1.asfreq("h", how="S")
+        result2 = p2.asfreq("h", how="S")
+        expected = Period("2007-01-01", freq="h")
+        assert result1 == expected
+        assert result1.ordinal == expected.ordinal
+        assert result1.freq == expected.freq
+        assert result2 == expected
+        assert result2.ordinal == expected.ordinal
+        assert result2.freq == expected.freq
+
+    def test_asfreq_MS(self):
+        initial = Period("2013")
+
+        assert initial.asfreq(freq="M", how="S") == Period("2013-01", "M")
+
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            initial.asfreq(freq="MS", how="S")
+
+        with pytest.raises(ValueError, match=msg):
+            Period("2013-01", "MS")
diff --git a/pandas/pandas/tests/scalar/period/test_period.py b/pandas/pandas/tests/scalar/period/test_period.py
new file mode 100644
index 0000000000000..baaedaa853565
--- /dev/null
+++ b/pandas/pandas/tests/scalar/period/test_period.py
@@ -0,0 +1,1170 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+from pandas._libs.tslibs.ccalendar import (
+    DAYS,
+    MONTHS,
+)
+from pandas._libs.tslibs.np_datetime import OutOfBoundsDatetime
+from pandas._libs.tslibs.parsing import DateParseError
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+
+from pandas import (
+    NaT,
+    Period,
+    Timedelta,
+    Timestamp,
+    offsets,
+)
+import pandas._testing as tm
+
+bday_msg = "Period with BDay freq is deprecated"
+
+
+class TestPeriodDisallowedFreqs:
+    @pytest.mark.parametrize(
+        "freq, freq_msg",
+        [
+            (offsets.BYearBegin(), "BYearBegin"),
+            (offsets.YearBegin(2), "YearBegin"),
+            (offsets.QuarterBegin(startingMonth=12), "QuarterBegin"),
+            (offsets.BusinessMonthEnd(2), "BusinessMonthEnd"),
+        ],
+    )
+    def test_offsets_not_supported(self, freq, freq_msg):
+        # GH#55785
+        msg = re.escape(f"{freq} is not supported as period frequency")
+        with pytest.raises(ValueError, match=msg):
+            Period(year=2014, freq=freq)
+
+    def test_custom_business_day_freq_raises(self):
+        # GH#52534
+        msg = "C is not supported as period frequency"
+        with pytest.raises(ValueError, match=msg):
+            Period("2023-04-10", freq="C")
+        msg = f"{offsets.CustomBusinessDay().base} is not supported as period frequency"
+        with pytest.raises(ValueError, match=msg):
+            Period("2023-04-10", freq=offsets.CustomBusinessDay())
+
+    def test_invalid_frequency_error_message(self):
+        msg = "WOM-1MON is not supported as period frequency"
+        with pytest.raises(ValueError, match=msg):
+            Period("2012-01-02", freq="WOM-1MON")
+
+    def test_invalid_frequency_period_error_message(self):
+        msg = "Invalid frequency: ME"
+        with pytest.raises(ValueError, match=msg):
+            Period("2012-01-02", freq="ME")
+
+
+class TestPeriodConstruction:
+    def test_from_td64nat_raises(self):
+        # GH#44507
+        td = NaT.to_numpy("m8[ns]")
+
+        msg = "Value must be Period, string, integer, or datetime"
+        with pytest.raises(ValueError, match=msg):
+            Period(td)
+
+        with pytest.raises(ValueError, match=msg):
+            Period(td, freq="D")
+
+    def test_construction(self):
+        i1 = Period("1/1/2005", freq="M")
+        i2 = Period("Jan 2005")
+
+        assert i1 == i2
+
+        # GH#54105 - Period can be confusingly instantiated with lowercase freq
+        # TODO: raise in the future an error when passing lowercase freq
+        i1 = Period("2005", freq="Y")
+        i2 = Period("2005")
+
+        assert i1 == i2
+
+        i4 = Period("2005", freq="M")
+        assert i1 != i4
+
+        i1 = Period.now(freq="Q")
+        i2 = Period(datetime.now(), freq="Q")
+
+        assert i1 == i2
+
+        # Pass in freq as a keyword argument sometimes as a test for
+        # https://github.com/pandas-dev/pandas/issues/53369
+        i1 = Period.now(freq="D")
+        i2 = Period(datetime.now(), freq="D")
+        i3 = Period.now(offsets.Day())
+
+        assert i1 == i2
+        assert i1 == i3
+
+        i1 = Period("1982", freq="min")
+        msg = "'MIN' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            i2 = Period("1982", freq="MIN")
+        assert i1 == i2
+
+        i1 = Period(year=2005, month=3, day=1, freq="D")
+        i2 = Period("3/1/2005", freq="D")
+        assert i1 == i2
+
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            i3 = Period(year=2005, month=3, day=1, freq="d")
+        assert i1 == i3
+
+        i1 = Period("2007-01-01 09:00:00.001")
+        expected = Period(datetime(2007, 1, 1, 9, 0, 0, 1000), freq="ms")
+        assert i1 == expected
+
+        expected = Period("2007-01-01 09:00:00.001", freq="ms")
+        assert i1 == expected
+
+        i1 = Period("2007-01-01 09:00:00.00101")
+        expected = Period(datetime(2007, 1, 1, 9, 0, 0, 1010), freq="us")
+        assert i1 == expected
+
+        expected = Period("2007-01-01 09:00:00.00101", freq="us")
+        assert i1 == expected
+
+        msg = "Must supply freq for ordinal value"
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=200701)
+
+        msg = "Invalid frequency: X"
+        with pytest.raises(ValueError, match=msg):
+            Period("2007-1-1", freq="X")
+
+    def test_tuple_freq_disallowed(self):
+        # GH#34703 tuple freq disallowed
+        with pytest.raises(TypeError, match="pass as a string instead"):
+            Period("1982", freq=("Min", 1))
+
+        with pytest.raises(TypeError, match="pass as a string instead"):
+            Period("2006-12-31", ("w", 1))
+
+    def test_construction_from_timestamp_nanos(self):
+        # GH#46811 don't drop nanos from Timestamp
+        ts = Timestamp("2022-04-20 09:23:24.123456789")
+        per = Period(ts, freq="ns")
+
+        # should losslessly round-trip, not lose the 789
+        rt = per.to_timestamp()
+        assert rt == ts
+
+        # same thing but from a datetime64 object
+        dt64 = ts.asm8
+        per2 = Period(dt64, freq="ns")
+        rt2 = per2.to_timestamp()
+        assert rt2.asm8 == dt64
+
+    def test_construction_bday(self):
+        # Biz day construction, roll forward if non-weekday
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            i1 = Period("3/10/12", freq="B")
+            i2 = Period("3/10/12", freq="D")
+            assert i1 == i2.asfreq("B")
+            i2 = Period("3/11/12", freq="D")
+            assert i1 == i2.asfreq("B")
+            i2 = Period("3/12/12", freq="D")
+            assert i1 == i2.asfreq("B")
+
+            i3 = Period("3/10/12", freq="b")
+            assert i1 == i3
+
+            i1 = Period(year=2012, month=3, day=10, freq="B")
+            i2 = Period("3/12/12", freq="B")
+            assert i1 == i2
+
+    def test_construction_quarter(self):
+        i1 = Period(year=2005, quarter=1, freq="Q")
+        i2 = Period("1/1/2005", freq="Q")
+        assert i1 == i2
+
+        i1 = Period(year=2005, quarter=3, freq="Q")
+        i2 = Period("9/1/2005", freq="Q")
+        assert i1 == i2
+
+        i1 = Period("2005Q1")
+        i2 = Period(year=2005, quarter=1, freq="Q")
+        i3 = Period("2005q1")
+        assert i1 == i2
+        assert i1 == i3
+
+        i1 = Period("05Q1")
+        assert i1 == i2
+        lower = Period("05q1")
+        assert i1 == lower
+
+        i1 = Period("1Q2005")
+        assert i1 == i2
+        lower = Period("1q2005")
+        assert i1 == lower
+
+        i1 = Period("1Q05")
+        assert i1 == i2
+        lower = Period("1q05")
+        assert i1 == lower
+
+        i1 = Period("4Q1984")
+        assert i1.year == 1984
+        lower = Period("4q1984")
+        assert i1 == lower
+
+    def test_construction_month(self):
+        expected = Period("2007-01", freq="M")
+        i1 = Period("200701", freq="M")
+        assert i1 == expected
+
+        i1 = Period("200701", freq="M")
+        assert i1 == expected
+
+        i1 = Period(200701, freq="M")
+        assert i1 == expected
+
+        i1 = Period(ordinal=200701, freq="M")
+        assert i1.year == 18695
+
+        i1 = Period(datetime(2007, 1, 1), freq="M")
+        i2 = Period("200701", freq="M")
+        assert i1 == i2
+
+        i1 = Period(date(2007, 1, 1), freq="M")
+        i2 = Period(datetime(2007, 1, 1), freq="M")
+        i3 = Period(np.datetime64("2007-01-01"), freq="M")
+        i4 = Period("2007-01-01 00:00:00", freq="M")
+        i5 = Period("2007-01-01 00:00:00.000", freq="M")
+        assert i1 == i2
+        assert i1 == i3
+        assert i1 == i4
+        assert i1 == i5
+
+    def test_period_constructor_offsets(self):
+        assert Period("1/1/2005", freq=offsets.MonthEnd()) == Period(
+            "1/1/2005", freq="M"
+        )
+        assert Period("2005", freq=offsets.YearEnd()) == Period("2005", freq="Y")
+        assert Period("2005", freq=offsets.MonthEnd()) == Period("2005", freq="M")
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert Period("3/10/12", freq=offsets.BusinessDay()) == Period(
+                "3/10/12", freq="B"
+            )
+        assert Period("3/10/12", freq=offsets.Day()) == Period("3/10/12", freq="D")
+
+        assert Period(
+            year=2005, quarter=1, freq=offsets.QuarterEnd(startingMonth=12)
+        ) == Period(year=2005, quarter=1, freq="Q")
+        assert Period(
+            year=2005, quarter=2, freq=offsets.QuarterEnd(startingMonth=12)
+        ) == Period(year=2005, quarter=2, freq="Q")
+
+        assert Period(year=2005, month=3, day=1, freq=offsets.Day()) == Period(
+            year=2005, month=3, day=1, freq="D"
+        )
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert Period(year=2012, month=3, day=10, freq=offsets.BDay()) == Period(
+                year=2012, month=3, day=10, freq="B"
+            )
+
+        expected = Period("2005-03-01", freq="3D")
+        assert Period(year=2005, month=3, day=1, freq=offsets.Day(3)) == expected
+        assert Period(year=2005, month=3, day=1, freq="3D") == expected
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert Period(year=2012, month=3, day=10, freq=offsets.BDay(3)) == Period(
+                year=2012, month=3, day=10, freq="3B"
+            )
+
+        assert Period(200701, freq=offsets.MonthEnd()) == Period(200701, freq="M")
+
+        i1 = Period(ordinal=200701, freq=offsets.MonthEnd())
+        i2 = Period(ordinal=200701, freq="M")
+        assert i1 == i2
+        assert i1.year == 18695
+        assert i2.year == 18695
+
+        i1 = Period(datetime(2007, 1, 1), freq="M")
+        i2 = Period("200701", freq="M")
+        assert i1 == i2
+
+        i1 = Period(date(2007, 1, 1), freq="M")
+        i2 = Period(datetime(2007, 1, 1), freq="M")
+        i3 = Period(np.datetime64("2007-01-01"), freq="M")
+        i4 = Period("2007-01-01 00:00:00", freq="M")
+        i5 = Period("2007-01-01 00:00:00.000", freq="M")
+        assert i1 == i2
+        assert i1 == i3
+        assert i1 == i4
+        assert i1 == i5
+
+        i1 = Period("2007-01-01 09:00:00.001")
+        expected = Period(datetime(2007, 1, 1, 9, 0, 0, 1000), freq="ms")
+        assert i1 == expected
+
+        expected = Period("2007-01-01 09:00:00.001", freq="ms")
+        assert i1 == expected
+
+        i1 = Period("2007-01-01 09:00:00.00101")
+        expected = Period(datetime(2007, 1, 1, 9, 0, 0, 1010), freq="us")
+        assert i1 == expected
+
+        expected = Period("2007-01-01 09:00:00.00101", freq="us")
+        assert i1 == expected
+
+    def test_invalid_arguments(self):
+        msg = "Must supply freq for datetime value"
+        with pytest.raises(ValueError, match=msg):
+            Period(datetime.now())
+        with pytest.raises(ValueError, match=msg):
+            Period(datetime.now().date())
+
+        msg = "Value must be Period, string, integer, or datetime"
+        with pytest.raises(ValueError, match=msg):
+            Period(1.6, freq="D")
+        msg = "Ordinal must be an integer"
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=1.6, freq="D")
+        msg = "Only value or ordinal but not both should be given but not both"
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=2, value=1, freq="D")
+
+        msg = "If value is None, freq cannot be None"
+        with pytest.raises(ValueError, match=msg):
+            Period(month=1)
+
+        msg = '^Given date string "-2000" not likely a datetime$'
+        with pytest.raises(ValueError, match=msg):
+            Period("-2000", "Y")
+        msg = "day is out of range for month"
+        with pytest.raises(DateParseError, match=msg):
+            Period("0", "Y")
+        msg = "Unknown datetime string format, unable to parse"
+        with pytest.raises(DateParseError, match=msg):
+            Period("1/1/-2000", "Y")
+
+    def test_constructor_corner(self):
+        expected = Period("2007-01", freq="2M")
+        assert Period(year=2007, month=1, freq="2M") == expected
+
+        assert Period(None) is NaT
+
+        p = Period("2007-01-01", freq="D")
+
+        result = Period(p, freq="Y")
+        exp = Period("2007", freq="Y")
+        assert result == exp
+
+    def test_constructor_infer_freq(self):
+        p = Period("2007-01-01")
+        assert p.freq == "D"
+
+        p = Period("2007-01-01 07")
+        assert p.freq == "h"
+
+        p = Period("2007-01-01 07:10")
+        assert p.freq == "min"
+
+        p = Period("2007-01-01 07:10:15")
+        assert p.freq == "s"
+
+        p = Period("2007-01-01 07:10:15.123")
+        assert p.freq == "ms"
+
+        # We see that there are 6 digits after the decimal, so get microsecond
+        #  even though they are all zeros.
+        p = Period("2007-01-01 07:10:15.123000")
+        assert p.freq == "us"
+
+        p = Period("2007-01-01 07:10:15.123400")
+        assert p.freq == "us"
+
+    def test_multiples(self):
+        result1 = Period("1989", freq="2Y")
+        result2 = Period("1989", freq="Y")
+        assert result1.ordinal == result2.ordinal
+        assert result1.freqstr == "2Y-DEC"
+        assert result2.freqstr == "Y-DEC"
+        assert result1.freq == offsets.YearEnd(2)
+        assert result2.freq == offsets.YearEnd()
+
+        assert (result1 + 1).ordinal == result1.ordinal + 2
+        assert (1 + result1).ordinal == result1.ordinal + 2
+        assert (result1 - 1).ordinal == result2.ordinal - 2
+        assert (-1 + result1).ordinal == result2.ordinal - 2
+
+    @pytest.mark.parametrize("month", MONTHS)
+    def test_period_cons_quarterly(self, month):
+        # bugs in scikits.timeseries
+        freq = f"Q-{month}"
+        exp = Period("1989Q3", freq=freq)
+        assert "1989Q3" in str(exp)
+        stamp = exp.to_timestamp("D", how="end")
+        p = Period(stamp, freq=freq)
+        assert p == exp
+
+        stamp = exp.to_timestamp("3D", how="end")
+        p = Period(stamp, freq=freq)
+        assert p == exp
+
+    @pytest.mark.parametrize("month", MONTHS)
+    def test_period_cons_annual(self, month):
+        # bugs in scikits.timeseries
+        freq = f"Y-{month}"
+        exp = Period("1989", freq=freq)
+        stamp = exp.to_timestamp("D", how="end") + timedelta(days=30)
+        p = Period(stamp, freq=freq)
+
+        assert p == exp + 1
+        assert isinstance(p, Period)
+
+    @pytest.mark.parametrize("day", DAYS)
+    @pytest.mark.parametrize("num", range(10, 17))
+    def test_period_cons_weekly(self, num, day):
+        daystr = f"2011-02-{num}"
+        freq = f"W-{day}"
+
+        result = Period(daystr, freq=freq)
+        expected = Period(daystr, freq="D").asfreq(freq)
+        assert result == expected
+        assert isinstance(result, Period)
+
+    def test_parse_week_str_roundstrip(self):
+        # GH#50803
+        per = Period("2017-01-23/2017-01-29")
+        assert per.freq.freqstr == "W-SUN"
+
+        per = Period("2017-01-24/2017-01-30")
+        assert per.freq.freqstr == "W-MON"
+
+        msg = "Could not parse as weekly-freq Period"
+        with pytest.raises(ValueError, match=msg):
+            # not 6 days apart
+            Period("2016-01-23/2017-01-29")
+
+    def test_period_from_ordinal(self):
+        p = Period("2011-01", freq="M")
+        res = Period._from_ordinal(p.ordinal, freq=p.freq)
+        assert p == res
+        assert isinstance(res, Period)
+
+    @pytest.mark.parametrize("freq", ["Y", "M", "D", "h"])
+    def test_construct_from_nat_string_and_freq(self, freq):
+        per = Period("NaT", freq=freq)
+        assert per is NaT
+
+        per = Period("NaT", freq="2" + freq)
+        assert per is NaT
+
+        per = Period("NaT", freq="3" + freq)
+        assert per is NaT
+
+    def test_period_cons_nat(self):
+        p = Period("nat", freq="W-SUN")
+        assert p is NaT
+
+        p = Period(iNaT, freq="D")
+        assert p is NaT
+
+        p = Period(iNaT, freq="3D")
+        assert p is NaT
+
+        p = Period(iNaT, freq="1D1h")
+        assert p is NaT
+
+        p = Period("NaT")
+        assert p is NaT
+
+        p = Period(iNaT)
+        assert p is NaT
+
+    def test_period_cons_mult(self):
+        p1 = Period("2011-01", freq="3M")
+        p2 = Period("2011-01", freq="M")
+        assert p1.ordinal == p2.ordinal
+
+        assert p1.freq == offsets.MonthEnd(3)
+        assert p1.freqstr == "3M"
+
+        assert p2.freq == offsets.MonthEnd()
+        assert p2.freqstr == "M"
+
+        result = p1 + 1
+        assert result.ordinal == (p2 + 3).ordinal
+
+        assert result.freq == p1.freq
+        assert result.freqstr == "3M"
+
+        result = p1 - 1
+        assert result.ordinal == (p2 - 3).ordinal
+        assert result.freq == p1.freq
+        assert result.freqstr == "3M"
+
+        msg = "Frequency must be positive, because it represents span: -3M"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="-3M")
+
+        msg = "Frequency must be positive, because it represents span: 0M"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="0M")
+
+    def test_period_cons_combined(self):
+        p = [
+            (
+                Period("2011-01", freq="1D1h"),
+                Period("2011-01", freq="1h1D"),
+                Period("2011-01", freq="h"),
+            ),
+            (
+                Period(ordinal=1, freq="1D1h"),
+                Period(ordinal=1, freq="1h1D"),
+                Period(ordinal=1, freq="h"),
+            ),
+        ]
+
+        for p1, p2, p3 in p:
+            assert p1.ordinal == p3.ordinal
+            assert p2.ordinal == p3.ordinal
+
+            assert p1.freq == offsets.Hour(25)
+            assert p1.freqstr == "25h"
+
+            assert p2.freq == offsets.Hour(25)
+            assert p2.freqstr == "25h"
+
+            assert p3.freq == offsets.Hour()
+            assert p3.freqstr == "h"
+
+            result = p1 + 1
+            assert result.ordinal == (p3 + 25).ordinal
+            assert result.freq == p1.freq
+            assert result.freqstr == "25h"
+
+            result = p2 + 1
+            assert result.ordinal == (p3 + 25).ordinal
+            assert result.freq == p2.freq
+            assert result.freqstr == "25h"
+
+            result = p1 - 1
+            assert result.ordinal == (p3 - 25).ordinal
+            assert result.freq == p1.freq
+            assert result.freqstr == "25h"
+
+            result = p2 - 1
+            assert result.ordinal == (p3 - 25).ordinal
+            assert result.freq == p2.freq
+            assert result.freqstr == "25h"
+
+        msg = "Frequency must be positive, because it represents span: -25h"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="-1D1h")
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="-1h1D")
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=1, freq="-1D1h")
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=1, freq="-1h1D")
+
+        msg = "Frequency must be positive, because it represents span: 0D"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="0D0h")
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=1, freq="0D0h")
+
+        # You can only combine together day and intraday offsets
+        msg = "Invalid frequency: 1W1D"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="1W1D")
+        msg = "Invalid frequency: 1D1W"
+        with pytest.raises(ValueError, match=msg):
+            Period("2011-01", freq="1D1W")
+
+    @pytest.mark.parametrize("day", ["1970/01/01 ", "2020-12-31 ", "1981/09/13 "])
+    @pytest.mark.parametrize("hour", ["00:00:00", "00:00:01", "23:59:59", "12:00:59"])
+    @pytest.mark.parametrize(
+        "sec_float, expected",
+        [
+            (".000000001", 1),
+            (".000000999", 999),
+            (".123456789", 789),
+            (".999999999", 999),
+            (".999999000", 0),
+            # Test femtoseconds, attoseconds, picoseconds are dropped like Timestamp
+            (".999999001123", 1),
+            (".999999001123456", 1),
+            (".999999001123456789", 1),
+        ],
+    )
+    def test_period_constructor_nanosecond(self, day, hour, sec_float, expected):
+        # GH 34621
+
+        assert Period(day + hour + sec_float).start_time.nanosecond == expected
+
+    @pytest.mark.parametrize("hour", range(24))
+    def test_period_large_ordinal(self, hour):
+        # Issue #36430
+        # Integer overflow for Period over the maximum timestamp
+        p = Period(ordinal=2562048 + hour, freq="1h")
+        assert p.hour == hour
+
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    @pytest.mark.parametrize(
+        "freq,freq_depr",
+        [("2W", "2w"), ("2W-FRI", "2w-fri"), ("2D", "2d"), ("2B", "2b")],
+    )
+    def test_period_deprecated_lowercase_freq(self, freq, freq_depr):
+        # GH#58998
+        msg = (
+            f"'{freq_depr[1:]}' is deprecated and will be removed in a future version."
+        )
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = Period("2016-03-01 09:00", freq=freq_depr)
+
+        expected = Period("2016-03-01 09:00", freq=freq)
+        assert result == expected
+
+
+class TestPeriodMethods:
+    def test_round_trip(self):
+        p = Period("2000Q1")
+        new_p = tm.round_trip_pickle(p)
+        assert new_p == p
+
+    def test_hash(self):
+        assert hash(Period("2011-01", freq="M")) == hash(Period("2011-01", freq="M"))
+
+        assert hash(Period("2011-01-01", freq="D")) != hash(Period("2011-01", freq="M"))
+
+        assert hash(Period("2011-01", freq="3M")) != hash(Period("2011-01", freq="2M"))
+
+        assert hash(Period("2011-01", freq="M")) != hash(Period("2011-02", freq="M"))
+
+    # --------------------------------------------------------------
+    # to_timestamp
+
+    def test_to_timestamp_mult(self):
+        p = Period("2011-01", freq="M")
+        assert p.to_timestamp(how="S") == Timestamp("2011-01-01")
+        expected = Timestamp("2011-02-01") - Timedelta(1, "ns")
+        assert p.to_timestamp(how="E") == expected
+
+        p = Period("2011-01", freq="3M")
+        assert p.to_timestamp(how="S") == Timestamp("2011-01-01")
+        expected = Timestamp("2011-04-01") - Timedelta(1, "ns")
+        assert p.to_timestamp(how="E") == expected
+
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_to_timestamp(self):
+        p = Period("1982", freq="Y")
+        start_ts = p.to_timestamp(how="S")
+        aliases = ["s", "StarT", "BEGIn"]
+        for a in aliases:
+            assert start_ts == p.to_timestamp("D", how=a)
+            # freq with mult should not affect to the result
+            assert start_ts == p.to_timestamp("3D", how=a)
+
+        end_ts = p.to_timestamp(how="E")
+        aliases = ["e", "end", "FINIsH"]
+        for a in aliases:
+            assert end_ts == p.to_timestamp("D", how=a)
+            assert end_ts == p.to_timestamp("3D", how=a)
+
+        from_lst = ["Y", "Q", "M", "W", "B", "D", "h", "Min", "s"]
+
+        def _ex(p):
+            if p.freq == "B":
+                return p.start_time + Timedelta(days=1, nanoseconds=-1)
+            return Timestamp((p + p.freq).start_time._value - 1)
+
+        for fcode in from_lst:
+            p = Period("1982", freq=fcode)
+            result = p.to_timestamp().to_period(fcode)
+            assert result == p
+
+            assert p.start_time == p.to_timestamp(how="S")
+
+            assert p.end_time == _ex(p)
+
+        # Frequency other than daily
+
+        p = Period("1985", freq="Y")
+
+        result = p.to_timestamp("h", how="end")
+        expected = Timestamp(1986, 1, 1) - Timedelta(1, "ns")
+        assert result == expected
+        result = p.to_timestamp("3h", how="end")
+        assert result == expected
+
+        result = p.to_timestamp("min", how="end")
+        expected = Timestamp(1986, 1, 1) - Timedelta(1, "ns")
+        assert result == expected
+        result = p.to_timestamp("2min", how="end")
+        assert result == expected
+
+        result = p.to_timestamp(how="end")
+        expected = Timestamp(1986, 1, 1) - Timedelta(1, "ns")
+        assert result == expected
+
+        expected = datetime(1985, 1, 1)
+        result = p.to_timestamp("h", how="start")
+        assert result == expected
+        result = p.to_timestamp("min", how="start")
+        assert result == expected
+        result = p.to_timestamp("s", how="start")
+        assert result == expected
+        result = p.to_timestamp("3h", how="start")
+        assert result == expected
+        result = p.to_timestamp("5s", how="start")
+        assert result == expected
+
+    def test_to_timestamp_business_end(self):
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            per = Period("1990-01-05", "B")  # Friday
+            result = per.to_timestamp("B", how="E")
+
+        expected = Timestamp("1990-01-06") - Timedelta(nanoseconds=1)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "ts, expected",
+        [
+            ("1970-01-01 00:00:00", 0),
+            ("1970-01-01 00:00:00.000001", 1),
+            ("1970-01-01 00:00:00.00001", 10),
+            ("1970-01-01 00:00:00.499", 499000),
+            ("1999-12-31 23:59:59.999", 999000),
+            ("1999-12-31 23:59:59.999999", 999999),
+            ("2050-12-31 23:59:59.5", 500000),
+            ("2050-12-31 23:59:59.500001", 500001),
+            ("2050-12-31 23:59:59.123456", 123456),
+        ],
+    )
+    @pytest.mark.parametrize("freq", [None, "us", "ns"])
+    def test_to_timestamp_microsecond(self, ts, expected, freq):
+        # GH 24444
+        result = Period(ts).to_timestamp(freq=freq).microsecond
+        assert result == expected
+
+    # --------------------------------------------------------------
+    # Rendering: __repr__, strftime, etc
+
+    @pytest.mark.parametrize(
+        "str_ts,freq,str_res,str_freq",
+        (
+            ("Jan-2000", None, "2000-01", "M"),
+            ("2000-12-15", None, "2000-12-15", "D"),
+            (
+                "2000-12-15 13:45:26.123456789",
+                "ns",
+                "2000-12-15 13:45:26.123456789",
+                "ns",
+            ),
+            ("2000-12-15 13:45:26.123456789", "us", "2000-12-15 13:45:26.123456", "us"),
+            ("2000-12-15 13:45:26.123456", None, "2000-12-15 13:45:26.123456", "us"),
+            ("2000-12-15 13:45:26.123456789", "ms", "2000-12-15 13:45:26.123", "ms"),
+            ("2000-12-15 13:45:26.123", None, "2000-12-15 13:45:26.123", "ms"),
+            ("2000-12-15 13:45:26", "s", "2000-12-15 13:45:26", "s"),
+            ("2000-12-15 13:45:26", "min", "2000-12-15 13:45", "min"),
+            ("2000-12-15 13:45:26", "h", "2000-12-15 13:00", "h"),
+            ("2000-12-15", "Y", "2000", "Y-DEC"),
+            ("2000-12-15", "Q", "2000Q4", "Q-DEC"),
+            ("2000-12-15", "M", "2000-12", "M"),
+            ("2000-12-15", "W", "2000-12-11/2000-12-17", "W-SUN"),
+            ("2000-12-15", "D", "2000-12-15", "D"),
+            ("2000-12-15", "B", "2000-12-15", "B"),
+        ),
+    )
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_repr(self, str_ts, freq, str_res, str_freq):
+        p = Period(str_ts, freq=freq)
+        assert str(p) == str_res
+        assert repr(p) == f"Period('{str_res}', '{str_freq}')"
+
+    def test_repr_nat(self):
+        p = Period("nat", freq="M")
+        assert repr(NaT) in repr(p)
+
+    def test_strftime(self):
+        # GH#3363
+        p = Period("2000-1-1 12:34:12", freq="s")
+        res = p.strftime("%Y-%m-%d %H:%M:%S")
+        assert res == "2000-01-01 12:34:12"
+        assert isinstance(res, str)
+
+
+class TestPeriodProperties:
+    """Test properties such as year, month, weekday, etc...."""
+
+    @pytest.mark.parametrize("freq", ["Y", "M", "D", "h"])
+    def test_is_leap_year(self, freq):
+        # GH 13727
+        p = Period("2000-01-01 00:00:00", freq=freq)
+        assert p.is_leap_year
+        assert isinstance(p.is_leap_year, bool)
+
+        p = Period("1999-01-01 00:00:00", freq=freq)
+        assert not p.is_leap_year
+
+        p = Period("2004-01-01 00:00:00", freq=freq)
+        assert p.is_leap_year
+
+        p = Period("2100-01-01 00:00:00", freq=freq)
+        assert not p.is_leap_year
+
+    def test_quarterly_negative_ordinals(self):
+        p = Period(ordinal=-1, freq="Q-DEC")
+        assert p.year == 1969
+        assert p.quarter == 4
+        assert isinstance(p, Period)
+
+        p = Period(ordinal=-2, freq="Q-DEC")
+        assert p.year == 1969
+        assert p.quarter == 3
+        assert isinstance(p, Period)
+
+        p = Period(ordinal=-2, freq="M")
+        assert p.year == 1969
+        assert p.month == 11
+        assert isinstance(p, Period)
+
+    def test_freq_str(self):
+        i1 = Period("1982", freq="Min")
+        assert i1.freq == offsets.Minute()
+        assert i1.freqstr == "min"
+
+    @pytest.mark.filterwarnings(
+        "ignore:Period with BDay freq is deprecated:FutureWarning"
+    )
+    def test_period_deprecated_freq(self):
+        cases = {
+            "M": ["MTH", "MONTH", "MONTHLY", "Mth", "month", "monthly"],
+            "B": ["BUS", "BUSINESS", "BUSINESSLY", "WEEKDAY", "bus"],
+            "D": ["DAY", "DLY", "DAILY", "Day", "Dly", "Daily"],
+            "h": ["HR", "HOUR", "HRLY", "HOURLY", "hr", "Hour", "HRly"],
+            "min": ["minute", "MINUTE", "MINUTELY", "minutely"],
+            "s": ["sec", "SEC", "SECOND", "SECONDLY", "second"],
+            "ms": ["MILLISECOND", "MILLISECONDLY", "millisecond"],
+            "us": ["MICROSECOND", "MICROSECONDLY", "microsecond"],
+            "ns": ["NANOSECOND", "NANOSECONDLY", "nanosecond"],
+        }
+
+        msg = INVALID_FREQ_ERR_MSG
+        for exp, freqs in cases.items():
+            for freq in freqs:
+                with pytest.raises(ValueError, match=msg):
+                    Period("2016-03-01 09:00", freq=freq)
+                with pytest.raises(ValueError, match=msg):
+                    Period(ordinal=1, freq=freq)
+
+            # check supported freq-aliases still works
+            p1 = Period("2016-03-01 09:00", freq=exp)
+            p2 = Period(ordinal=1, freq=exp)
+            assert isinstance(p1, Period)
+            assert isinstance(p2, Period)
+
+    @staticmethod
+    def _period_constructor(bound, offset):
+        return Period(
+            year=bound.year,
+            month=bound.month,
+            day=bound.day,
+            hour=bound.hour,
+            minute=bound.minute,
+            second=bound.second + offset,
+            freq="us",
+        )
+
+    @pytest.mark.parametrize("bound, offset", [(Timestamp.min, -1), (Timestamp.max, 1)])
+    @pytest.mark.parametrize("period_property", ["start_time", "end_time"])
+    def test_outer_bounds_start_and_end_time(self, bound, offset, period_property):
+        # GH #13346
+        period = TestPeriodProperties._period_constructor(bound, offset)
+        with pytest.raises(OutOfBoundsDatetime, match="Out of bounds nanosecond"):
+            getattr(period, period_property)
+
+    @pytest.mark.parametrize("bound, offset", [(Timestamp.min, -1), (Timestamp.max, 1)])
+    @pytest.mark.parametrize("period_property", ["start_time", "end_time"])
+    def test_inner_bounds_start_and_end_time(self, bound, offset, period_property):
+        # GH #13346
+        period = TestPeriodProperties._period_constructor(bound, -offset)
+        expected = period.to_timestamp().round(freq="s")
+        assert getattr(period, period_property).round(freq="s") == expected
+        expected = (bound - offset * Timedelta(1, unit="s")).floor("s")
+        assert getattr(period, period_property).floor("s") == expected
+
+    def test_start_time(self):
+        freq_lst = ["Y", "Q", "M", "D", "h", "min", "s"]
+        xp = datetime(2012, 1, 1)
+        for f in freq_lst:
+            p = Period("2012", freq=f)
+            assert p.start_time == xp
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert Period("2012", freq="B").start_time == datetime(2012, 1, 2)
+        assert Period("2012", freq="W").start_time == datetime(2011, 12, 26)
+
+    def test_end_time(self):
+        p = Period("2012", freq="Y")
+
+        def _ex(*args):
+            return Timestamp(Timestamp(datetime(*args)).as_unit("ns")._value - 1)
+
+        xp = _ex(2013, 1, 1)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="Q")
+        xp = _ex(2012, 4, 1)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="M")
+        xp = _ex(2012, 2, 1)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="D")
+        xp = _ex(2012, 1, 2)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="h")
+        xp = _ex(2012, 1, 1, 1)
+        assert xp == p.end_time
+
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            p = Period("2012", freq="B")
+            xp = _ex(2012, 1, 3)
+            assert xp == p.end_time
+
+        p = Period("2012", freq="W")
+        xp = _ex(2012, 1, 2)
+        assert xp == p.end_time
+
+        # Test for GH 11738
+        p = Period("2012", freq="15D")
+        xp = _ex(2012, 1, 16)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="1D1h")
+        xp = _ex(2012, 1, 2, 1)
+        assert xp == p.end_time
+
+        p = Period("2012", freq="1h1D")
+        xp = _ex(2012, 1, 2, 1)
+        assert xp == p.end_time
+
+    def test_end_time_business_friday(self):
+        # GH#34449
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            per = Period("1990-01-05", "B")
+            result = per.end_time
+
+        expected = Timestamp("1990-01-06") - Timedelta(nanoseconds=1)
+        assert result == expected
+
+    def test_anchor_week_end_time(self):
+        def _ex(*args):
+            return Timestamp(Timestamp(datetime(*args)).as_unit("ns")._value - 1)
+
+        p = Period("2013-1-1", "W-SAT")
+        xp = _ex(2013, 1, 6)
+        assert p.end_time == xp
+
+    def test_properties_annually(self):
+        # Test properties on Periods with annually frequency.
+        a_date = Period(freq="Y", year=2007)
+        assert a_date.year == 2007
+
+    def test_properties_quarterly(self):
+        # Test properties on Periods with daily frequency.
+        qedec_date = Period(freq="Q-DEC", year=2007, quarter=1)
+        qejan_date = Period(freq="Q-JAN", year=2007, quarter=1)
+        qejun_date = Period(freq="Q-JUN", year=2007, quarter=1)
+        for x in range(3):
+            for qd in (qedec_date, qejan_date, qejun_date):
+                assert (qd + x).qyear == 2007
+                assert (qd + x).quarter == x + 1
+
+    def test_properties_monthly(self):
+        # Test properties on Periods with daily frequency.
+        m_date = Period(freq="M", year=2007, month=1)
+        for x in range(11):
+            m_ival_x = m_date + x
+            assert m_ival_x.year == 2007
+            if 1 <= x + 1 <= 3:
+                assert m_ival_x.quarter == 1
+            elif 4 <= x + 1 <= 6:
+                assert m_ival_x.quarter == 2
+            elif 7 <= x + 1 <= 9:
+                assert m_ival_x.quarter == 3
+            elif 10 <= x + 1 <= 12:
+                assert m_ival_x.quarter == 4
+            assert m_ival_x.month == x + 1
+
+    def test_properties_weekly(self):
+        # Test properties on Periods with daily frequency.
+        w_date = Period(freq="W", year=2007, month=1, day=7)
+        assert w_date.year == 2007
+        assert w_date.quarter == 1
+        assert w_date.month == 1
+        assert w_date.week == 1
+        assert (w_date - 1).week == 52
+        assert w_date.days_in_month == 31
+        assert Period(freq="W", year=2012, month=2, day=1).days_in_month == 29
+
+    def test_properties_weekly_legacy(self):
+        # Test properties on Periods with daily frequency.
+        w_date = Period(freq="W", year=2007, month=1, day=7)
+        assert w_date.year == 2007
+        assert w_date.quarter == 1
+        assert w_date.month == 1
+        assert w_date.week == 1
+        assert (w_date - 1).week == 52
+        assert w_date.days_in_month == 31
+
+        exp = Period(freq="W", year=2012, month=2, day=1)
+        assert exp.days_in_month == 29
+
+        msg = INVALID_FREQ_ERR_MSG
+        with pytest.raises(ValueError, match=msg):
+            Period(freq="WK", year=2007, month=1, day=7)
+
+    def test_properties_daily(self):
+        # Test properties on Periods with daily frequency.
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            b_date = Period(freq="B", year=2007, month=1, day=1)
+        assert b_date.year == 2007
+        assert b_date.quarter == 1
+        assert b_date.month == 1
+        assert b_date.day == 1
+        assert b_date.weekday == 0
+        assert b_date.dayofyear == 1
+        assert b_date.days_in_month == 31
+        with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+            assert Period(freq="B", year=2012, month=2, day=1).days_in_month == 29
+
+        d_date = Period(freq="D", year=2007, month=1, day=1)
+
+        assert d_date.year == 2007
+        assert d_date.quarter == 1
+        assert d_date.month == 1
+        assert d_date.day == 1
+        assert d_date.weekday == 0
+        assert d_date.dayofyear == 1
+        assert d_date.days_in_month == 31
+        assert Period(freq="D", year=2012, month=2, day=1).days_in_month == 29
+
+    def test_properties_hourly(self):
+        # Test properties on Periods with hourly frequency.
+        h_date1 = Period(freq="h", year=2007, month=1, day=1, hour=0)
+        h_date2 = Period(freq="2h", year=2007, month=1, day=1, hour=0)
+
+        for h_date in [h_date1, h_date2]:
+            assert h_date.year == 2007
+            assert h_date.quarter == 1
+            assert h_date.month == 1
+            assert h_date.day == 1
+            assert h_date.weekday == 0
+            assert h_date.dayofyear == 1
+            assert h_date.hour == 0
+            assert h_date.days_in_month == 31
+            assert (
+                Period(freq="h", year=2012, month=2, day=1, hour=0).days_in_month == 29
+            )
+
+    def test_properties_minutely(self):
+        # Test properties on Periods with minutely frequency.
+        t_date = Period(freq="Min", year=2007, month=1, day=1, hour=0, minute=0)
+        assert t_date.quarter == 1
+        assert t_date.month == 1
+        assert t_date.day == 1
+        assert t_date.weekday == 0
+        assert t_date.dayofyear == 1
+        assert t_date.hour == 0
+        assert t_date.minute == 0
+        assert t_date.days_in_month == 31
+        assert (
+            Period(freq="D", year=2012, month=2, day=1, hour=0, minute=0).days_in_month
+            == 29
+        )
+
+    def test_properties_secondly(self):
+        # Test properties on Periods with secondly frequency.
+        s_date = Period(
+            freq="Min", year=2007, month=1, day=1, hour=0, minute=0, second=0
+        )
+        assert s_date.year == 2007
+        assert s_date.quarter == 1
+        assert s_date.month == 1
+        assert s_date.day == 1
+        assert s_date.weekday == 0
+        assert s_date.dayofyear == 1
+        assert s_date.hour == 0
+        assert s_date.minute == 0
+        assert s_date.second == 0
+        assert s_date.days_in_month == 31
+        assert (
+            Period(
+                freq="Min", year=2012, month=2, day=1, hour=0, minute=0, second=0
+            ).days_in_month
+            == 29
+        )
+
+
+class TestPeriodComparisons:
+    def test_sort_periods(self):
+        jan = Period("2000-01", "M")
+        feb = Period("2000-02", "M")
+        mar = Period("2000-03", "M")
+        periods = [mar, jan, feb]
+        correctPeriods = [jan, feb, mar]
+        assert sorted(periods) == correctPeriods
+
+
+def test_period_immutable():
+    # see gh-17116
+    msg = "not writable"
+
+    per = Period("2014Q1")
+    with pytest.raises(AttributeError, match=msg):
+        per.ordinal = 14
+
+    freq = per.freq
+    with pytest.raises(AttributeError, match=msg):
+        per.freq = 2 * freq
+
+
+def test_small_year_parsing():
+    per1 = Period("0001-01-07", "D")
+    assert per1.year == 1
+    assert per1.day == 7
+
+
+def test_negone_ordinals():
+    freqs = ["Y", "M", "Q", "D", "h", "min", "s"]
+
+    period = Period(ordinal=-1, freq="D")
+    for freq in freqs:
+        repr(period.asfreq(freq))
+
+    for freq in freqs:
+        period = Period(ordinal=-1, freq=freq)
+        repr(period)
+        assert period.year == 1969
+
+    with tm.assert_produces_warning(FutureWarning, match=bday_msg):
+        period = Period(ordinal=-1, freq="B")
+    repr(period)
+    period = Period(ordinal=-1, freq="W")
+    repr(period)
diff --git a/pandas/pandas/tests/scalar/test_na_scalar.py b/pandas/pandas/tests/scalar/test_na_scalar.py
new file mode 100644
index 0000000000000..287b7557f50f9
--- /dev/null
+++ b/pandas/pandas/tests/scalar/test_na_scalar.py
@@ -0,0 +1,316 @@
+from datetime import (
+    date,
+    time,
+    timedelta,
+)
+import pickle
+
+import numpy as np
+import pytest
+
+from pandas._libs.missing import NA
+
+from pandas.core.dtypes.common import is_scalar
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_singleton():
+    assert NA is NA
+    new_NA = type(NA)()
+    assert new_NA is NA
+
+
+def test_repr():
+    assert repr(NA) == "<NA>"
+    assert str(NA) == "<NA>"
+
+
+def test_format():
+    # GH-34740
+    assert format(NA) == "<NA>"
+    assert format(NA, ">10") == "      <NA>"
+    assert format(NA, "xxx") == "<NA>"  # NA is flexible, accept any format spec
+
+    assert f"{NA}" == "<NA>"
+    assert f"{NA:>10}" == "      <NA>"
+    assert f"{NA:xxx}" == "<NA>"
+
+
+def test_truthiness():
+    msg = "boolean value of NA is ambiguous"
+
+    with pytest.raises(TypeError, match=msg):
+        bool(NA)
+
+    with pytest.raises(TypeError, match=msg):
+        not NA
+
+
+def test_hashable():
+    assert hash(NA) == hash(NA)
+    d = {NA: "test"}
+    assert d[NA] == "test"
+
+
+@pytest.mark.parametrize(
+    "other", [NA, 1, 1.0, "a", b"a", np.int64(1), np.nan], ids=repr
+)
+def test_arithmetic_ops(all_arithmetic_functions, other):
+    op = all_arithmetic_functions
+
+    if op.__name__ in ("pow", "rpow", "rmod") and isinstance(other, (str, bytes)):
+        pytest.skip(reason=f"{op.__name__} with NA and {other} not defined.")
+    if op.__name__ in ("divmod", "rdivmod"):
+        assert op(NA, other) is (NA, NA)
+    else:
+        if op.__name__ == "rpow":
+            # avoid special case
+            other += 1
+        assert op(NA, other) is NA
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        NA,
+        1,
+        1.0,
+        "a",
+        b"a",
+        np.int64(1),
+        np.nan,
+        np.bool_(True),
+        time(0),
+        date(1, 2, 3),
+        timedelta(1),
+        pd.NaT,
+    ],
+)
+def test_comparison_ops(comparison_op, other):
+    assert comparison_op(NA, other) is NA
+    assert comparison_op(other, NA) is NA
+
+
+@pytest.mark.parametrize(
+    "value",
+    [
+        0,
+        0.0,
+        -0,
+        -0.0,
+        False,
+        np.bool_(False),
+        np.int_(0),
+        np.float64(0),
+        np.int_(-0),
+        np.float64(-0),
+    ],
+)
+@pytest.mark.parametrize("asarray", [True, False])
+def test_pow_special(value, asarray):
+    if asarray:
+        value = np.array([value])
+    result = NA**value
+
+    if asarray:
+        result = result[0]
+    else:
+        # this assertion isn't possible for ndarray.
+        assert isinstance(result, type(value))
+    assert result == 1
+
+
+@pytest.mark.parametrize(
+    "value", [1, 1.0, True, np.bool_(True), np.int_(1), np.float64(1)]
+)
+@pytest.mark.parametrize("asarray", [True, False])
+def test_rpow_special(value, asarray):
+    if asarray:
+        value = np.array([value])
+    result = value**NA
+
+    if asarray:
+        result = result[0]
+    elif not isinstance(value, (np.float64, np.bool_, np.int_)):
+        # this assertion isn't possible with asarray=True
+        assert isinstance(result, type(value))
+
+    assert result == value
+
+
+@pytest.mark.parametrize("value", [-1, -1.0, np.int_(-1), np.float64(-1)])
+@pytest.mark.parametrize("asarray", [True, False])
+def test_rpow_minus_one(value, asarray):
+    if asarray:
+        value = np.array([value])
+    result = value**NA
+
+    if asarray:
+        result = result[0]
+
+    assert pd.isna(result)
+
+
+def test_unary_ops():
+    assert +NA is NA
+    assert -NA is NA
+    assert abs(NA) is NA
+    assert ~NA is NA
+
+
+def test_logical_and():
+    assert NA & True is NA
+    assert True & NA is NA
+    assert NA & False is False
+    assert False & NA is False
+    assert NA & NA is NA
+
+    msg = "unsupported operand type"
+    with pytest.raises(TypeError, match=msg):
+        NA & 5
+
+
+def test_logical_or():
+    assert NA | True is True
+    assert True | NA is True
+    assert NA | False is NA
+    assert False | NA is NA
+    assert NA | NA is NA
+
+    msg = "unsupported operand type"
+    with pytest.raises(TypeError, match=msg):
+        NA | 5
+
+
+def test_logical_xor():
+    assert NA ^ True is NA
+    assert True ^ NA is NA
+    assert NA ^ False is NA
+    assert False ^ NA is NA
+    assert NA ^ NA is NA
+
+    msg = "unsupported operand type"
+    with pytest.raises(TypeError, match=msg):
+        NA ^ 5
+
+
+def test_logical_not():
+    assert ~NA is NA
+
+
+@pytest.mark.parametrize("shape", [(3,), (3, 3), (1, 2, 3)])
+def test_arithmetic_ndarray(shape, all_arithmetic_functions):
+    op = all_arithmetic_functions
+    a = np.zeros(shape)
+    if op.__name__ == "pow":
+        a += 5
+    result = op(NA, a)
+    expected = np.full(a.shape, NA, dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_is_scalar():
+    assert is_scalar(NA) is True
+
+
+def test_isna():
+    assert pd.isna(NA) is True
+    assert pd.notna(NA) is False
+
+
+def test_series_isna():
+    s = pd.Series([1, NA], dtype=object)
+    expected = pd.Series([False, True])
+    tm.assert_series_equal(s.isna(), expected)
+
+
+def test_ufunc():
+    assert np.log(NA) is NA
+    assert np.add(NA, 1) is NA
+    result = np.divmod(NA, 1)
+    assert result[0] is NA and result[1] is NA
+
+    result = np.frexp(NA)
+    assert result[0] is NA and result[1] is NA
+
+
+def test_ufunc_raises():
+    msg = "ufunc method 'at'"
+    with pytest.raises(ValueError, match=msg):
+        np.log.at(NA, 0)
+
+
+def test_binary_input_not_dunder():
+    a = np.array([1, 2, 3])
+    expected = np.array([NA, NA, NA], dtype=object)
+    result = np.logaddexp(a, NA)
+    tm.assert_numpy_array_equal(result, expected)
+
+    result = np.logaddexp(NA, a)
+    tm.assert_numpy_array_equal(result, expected)
+
+    # all NA, multiple inputs
+    assert np.logaddexp(NA, NA) is NA
+
+    result = np.modf(NA, NA)
+    assert len(result) == 2
+    assert all(x is NA for x in result)
+
+
+def test_divmod_ufunc():
+    # binary in, binary out.
+    a = np.array([1, 2, 3])
+    expected = np.array([NA, NA, NA], dtype=object)
+
+    result = np.divmod(a, NA)
+    assert isinstance(result, tuple)
+    for arr in result:
+        tm.assert_numpy_array_equal(arr, expected)
+        tm.assert_numpy_array_equal(arr, expected)
+
+    result = np.divmod(NA, a)
+    for arr in result:
+        tm.assert_numpy_array_equal(arr, expected)
+        tm.assert_numpy_array_equal(arr, expected)
+
+
+def test_integer_hash_collision_dict():
+    # GH 30013
+    result = {NA: "foo", hash(NA): "bar"}
+
+    assert result[NA] == "foo"
+    assert result[hash(NA)] == "bar"
+
+
+def test_integer_hash_collision_set():
+    # GH 30013
+    result = {NA, hash(NA)}
+
+    assert len(result) == 2
+    assert NA in result
+    assert hash(NA) in result
+
+
+def test_pickle_roundtrip():
+    # https://github.com/pandas-dev/pandas/issues/31847
+    result = pickle.loads(pickle.dumps(NA))
+    assert result is NA
+
+
+def test_pickle_roundtrip_pandas():
+    result = tm.round_trip_pickle(NA)
+    assert result is NA
+
+
+@pytest.mark.parametrize(
+    "values, dtype", [([1, 2, NA], "Int64"), (["A", "B", NA], "string")]
+)
+@pytest.mark.parametrize("as_frame", [True, False])
+def test_pickle_roundtrip_containers(as_frame, values, dtype):
+    s = pd.Series(pd.array(values, dtype=dtype))
+    if as_frame:
+        s = s.to_frame(name="A")
+    result = tm.round_trip_pickle(s)
+    tm.assert_equal(result, s)
diff --git a/pandas/pandas/tests/scalar/test_nat.py b/pandas/pandas/tests/scalar/test_nat.py
new file mode 100644
index 0000000000000..b20df43dd49a6
--- /dev/null
+++ b/pandas/pandas/tests/scalar/test_nat.py
@@ -0,0 +1,710 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+import operator
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+from pandas.compat.numpy import np_version_gte1p24p3
+
+from pandas import (
+    DatetimeIndex,
+    DatetimeTZDtype,
+    Index,
+    NaT,
+    Period,
+    Series,
+    Timedelta,
+    TimedeltaIndex,
+    Timestamp,
+    isna,
+    offsets,
+)
+import pandas._testing as tm
+from pandas.core import roperator
+from pandas.core.arrays import (
+    DatetimeArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+
+
+class TestNaTFormatting:
+    def test_repr(self):
+        assert repr(NaT) == "NaT"
+
+    def test_str(self):
+        assert str(NaT) == "NaT"
+
+    def test_isoformat(self):
+        assert NaT.isoformat() == "NaT"
+
+
+@pytest.mark.parametrize(
+    "nat,idx",
+    [
+        (Timestamp("NaT"), DatetimeArray),
+        (Timedelta("NaT"), TimedeltaArray),
+        (Period("NaT", freq="M"), PeriodArray),
+    ],
+)
+def test_nat_fields(nat, idx):
+    for field in idx._field_ops:
+        # weekday is a property of DTI, but a method
+        # on NaT/Timestamp for compat with datetime
+        if field == "weekday":
+            continue
+
+        result = getattr(NaT, field)
+        assert np.isnan(result)
+
+        result = getattr(nat, field)
+        assert np.isnan(result)
+
+    for field in idx._bool_ops:
+        result = getattr(NaT, field)
+        assert result is False
+
+        result = getattr(nat, field)
+        assert result is False
+
+
+def test_nat_vector_field_access():
+    idx = DatetimeIndex(["1/1/2000", None, None, "1/4/2000"])
+
+    for field in DatetimeArray._field_ops:
+        # weekday is a property of DTI, but a method
+        # on NaT/Timestamp for compat with datetime
+        if field == "weekday":
+            continue
+
+        result = getattr(idx, field)
+        expected = Index([getattr(x, field) for x in idx])
+        tm.assert_index_equal(result, expected)
+
+    ser = Series(idx)
+
+    for field in DatetimeArray._field_ops:
+        # weekday is a property of DTI, but a method
+        # on NaT/Timestamp for compat with datetime
+        if field == "weekday":
+            continue
+
+        result = getattr(ser.dt, field)
+        expected = [getattr(x, field) for x in idx]
+        tm.assert_series_equal(result, Series(expected))
+
+    for field in DatetimeArray._bool_ops:
+        result = getattr(ser.dt, field)
+        expected = [getattr(x, field) for x in idx]
+        tm.assert_series_equal(result, Series(expected))
+
+
+@pytest.mark.parametrize("klass", [Timestamp, Timedelta, Period])
+@pytest.mark.parametrize(
+    "value", [None, np.nan, iNaT, float("nan"), NaT, "NaT", "nat", "", "NAT"]
+)
+def test_identity(klass, value):
+    assert klass(value) is NaT
+
+
+@pytest.mark.parametrize("klass", [Timestamp, Timedelta])
+@pytest.mark.parametrize("method", ["round", "floor", "ceil"])
+@pytest.mark.parametrize("freq", ["s", "5s", "min", "5min", "h", "5h"])
+def test_round_nat(klass, method, freq):
+    # see gh-14940
+    ts = klass("nat")
+
+    round_method = getattr(ts, method)
+    assert round_method(freq) is ts
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        "astimezone",
+        "combine",
+        "ctime",
+        "dst",
+        "fromordinal",
+        "fromtimestamp",
+        "fromisocalendar",
+        "isocalendar",
+        "strftime",
+        "strptime",
+        "time",
+        "timestamp",
+        "timetuple",
+        "timetz",
+        "toordinal",
+        "tzname",
+        "utcfromtimestamp",
+        "utcnow",
+        "utcoffset",
+        "utctimetuple",
+    ],
+)
+def test_nat_methods_raise(method):
+    # see gh-9513, gh-17329
+    msg = f"NaTType does not support {method}"
+
+    with pytest.raises(ValueError, match=msg):
+        getattr(NaT, method)()
+
+
+@pytest.mark.parametrize("method", ["weekday", "isoweekday"])
+def test_nat_methods_nan(method):
+    # see gh-9513, gh-17329
+    assert np.isnan(getattr(NaT, method)())
+
+
+@pytest.mark.parametrize(
+    "method", ["date", "now", "replace", "today", "tz_convert", "tz_localize"]
+)
+def test_nat_methods_nat(method):
+    # see gh-8254, gh-9513, gh-17329
+    assert getattr(NaT, method)() is NaT
+
+
+@pytest.mark.parametrize(
+    "get_nat", [lambda x: NaT, lambda x: Timedelta(x), lambda x: Timestamp(x)]
+)
+def test_nat_iso_format(get_nat):
+    # see gh-12300
+    assert get_nat("NaT").isoformat() == "NaT"
+    assert get_nat("NaT").isoformat(timespec="nanoseconds") == "NaT"
+
+
+@pytest.mark.parametrize(
+    "klass,expected",
+    [
+        (Timestamp, ["normalize", "to_julian_date", "to_period", "unit"]),
+        (
+            Timedelta,
+            [
+                "components",
+                "resolution_string",
+                "to_pytimedelta",
+                "to_timedelta64",
+                "unit",
+                "view",
+            ],
+        ),
+    ],
+)
+def test_missing_public_nat_methods(klass, expected):
+    # see gh-17327
+    #
+    # NaT should have *most* of the Timestamp and Timedelta methods.
+    # Here, we check which public methods NaT does not have. We
+    # ignore any missing private methods.
+    nat_names = dir(NaT)
+    klass_names = dir(klass)
+
+    missing = [x for x in klass_names if x not in nat_names and not x.startswith("_")]
+    missing.sort()
+
+    assert missing == expected
+
+
+def _get_overlap_public_nat_methods(klass, as_tuple=False):
+    """
+    Get overlapping public methods between NaT and another class.
+
+    Parameters
+    ----------
+    klass : type
+        The class to compare with NaT
+    as_tuple : bool, default False
+        Whether to return a list of tuples of the form (klass, method).
+
+    Returns
+    -------
+    overlap : list
+    """
+    nat_names = dir(NaT)
+    klass_names = dir(klass)
+
+    overlap = [
+        x
+        for x in nat_names
+        if x in klass_names and not x.startswith("_") and callable(getattr(klass, x))
+    ]
+
+    # Timestamp takes precedence over Timedelta in terms of overlap.
+    if klass is Timedelta:
+        ts_names = dir(Timestamp)
+        overlap = [x for x in overlap if x not in ts_names]
+
+    if as_tuple:
+        overlap = [(klass, method) for method in overlap]
+
+    overlap.sort()
+    return overlap
+
+
+@pytest.mark.parametrize(
+    "klass,expected",
+    [
+        (
+            Timestamp,
+            [
+                "as_unit",
+                "astimezone",
+                "ceil",
+                "combine",
+                "ctime",
+                "date",
+                "day_name",
+                "dst",
+                "floor",
+                "fromisocalendar",
+                "fromisoformat",
+                "fromordinal",
+                "fromtimestamp",
+                "isocalendar",
+                "isoformat",
+                "isoweekday",
+                "month_name",
+                "now",
+                "replace",
+                "round",
+                "strftime",
+                "strptime",
+                "time",
+                "timestamp",
+                "timetuple",
+                "timetz",
+                "to_datetime64",
+                "to_numpy",
+                "to_pydatetime",
+                "today",
+                "toordinal",
+                "tz_convert",
+                "tz_localize",
+                "tzname",
+                "utcfromtimestamp",
+                "utcnow",
+                "utcoffset",
+                "utctimetuple",
+                "weekday",
+            ],
+        ),
+        (Timedelta, ["total_seconds"]),
+    ],
+)
+def test_overlap_public_nat_methods(klass, expected):
+    # see gh-17327
+    #
+    # NaT should have *most* of the Timestamp and Timedelta methods.
+    # In case when Timestamp, Timedelta, and NaT are overlap, the overlap
+    # is considered to be with Timestamp and NaT, not Timedelta.
+    assert _get_overlap_public_nat_methods(klass) == expected
+
+
+@pytest.mark.parametrize(
+    "compare",
+    (
+        _get_overlap_public_nat_methods(Timestamp, True)
+        + _get_overlap_public_nat_methods(Timedelta, True)
+    ),
+    ids=lambda x: f"{x[0].__name__}.{x[1]}",
+)
+def test_nat_doc_strings(compare):
+    # see gh-17327
+    #
+    # The docstrings for overlapping methods should match.
+    klass, method = compare
+    klass_doc = getattr(klass, method).__doc__
+
+    if klass == Timestamp and method == "isoformat":
+        pytest.skip(
+            "Ignore differences with Timestamp.isoformat() as they're intentional"
+        )
+
+    if method == "to_numpy":
+        # GH#44460 can return either dt64 or td64 depending on dtype,
+        #  different docstring is intentional
+        pytest.skip(f"different docstring for {method} is intentional")
+
+    nat_doc = getattr(NaT, method).__doc__
+    assert klass_doc == nat_doc
+
+
+_ops = {
+    "left_plus_right": lambda a, b: a + b,
+    "right_plus_left": lambda a, b: b + a,
+    "left_minus_right": lambda a, b: a - b,
+    "right_minus_left": lambda a, b: b - a,
+    "left_times_right": lambda a, b: a * b,
+    "right_times_left": lambda a, b: b * a,
+    "left_div_right": lambda a, b: a / b,
+    "right_div_left": lambda a, b: b / a,
+}
+
+
+@pytest.mark.parametrize("op_name", list(_ops.keys()))
+@pytest.mark.parametrize(
+    "value,val_type",
+    [
+        (2, "scalar"),
+        (1.5, "floating"),
+        (np.nan, "floating"),
+        ("foo", "str"),
+        (timedelta(3600), "timedelta"),
+        (Timedelta("5s"), "timedelta"),
+        (datetime(2014, 1, 1), "timestamp"),
+        (Timestamp("2014-01-01"), "timestamp"),
+        (Timestamp("2014-01-01", tz="UTC"), "timestamp"),
+        (Timestamp("2014-01-01", tz="US/Eastern"), "timestamp"),
+        (datetime(2014, 1, 1).astimezone(zoneinfo.ZoneInfo("Asia/Tokyo")), "timestamp"),
+    ],
+)
+def test_nat_arithmetic_scalar(op_name, value, val_type):
+    # see gh-6873
+    invalid_ops = {
+        "scalar": {"right_div_left"},
+        "floating": {
+            "right_div_left",
+            "left_minus_right",
+            "right_minus_left",
+            "left_plus_right",
+            "right_plus_left",
+        },
+        "str": set(_ops.keys()),
+        "timedelta": {"left_times_right", "right_times_left"},
+        "timestamp": {
+            "left_times_right",
+            "right_times_left",
+            "left_div_right",
+            "right_div_left",
+        },
+    }
+
+    op = _ops[op_name]
+
+    if op_name in invalid_ops.get(val_type, set()):
+        if (
+            val_type == "timedelta"
+            and "times" in op_name
+            and isinstance(value, Timedelta)
+        ):
+            typs = "(Timedelta|NaTType)"
+            msg = rf"unsupported operand type\(s\) for \*: '{typs}' and '{typs}'"
+        elif val_type == "str":
+            # un-specific check here because the message comes from str
+            #  and varies by method
+            msg = "|".join(
+                [
+                    "can only concatenate str",
+                    "unsupported operand type",
+                    "can't multiply sequence",
+                    "Can't convert 'NaTType'",
+                    "must be str, not NaTType",
+                ]
+            )
+        else:
+            msg = "unsupported operand type"
+
+        with pytest.raises(TypeError, match=msg):
+            op(NaT, value)
+    else:
+        if val_type == "timedelta" and "div" in op_name:
+            expected = np.nan
+        else:
+            expected = NaT
+
+        assert op(NaT, value) is expected
+
+
+@pytest.mark.parametrize(
+    "val,expected", [(np.nan, NaT), (NaT, np.nan), (np.timedelta64("NaT"), np.nan)]
+)
+def test_nat_rfloordiv_timedelta(val, expected):
+    # see gh-#18846
+    #
+    # See also test_timedelta.TestTimedeltaArithmetic.test_floordiv
+    td = Timedelta(hours=3, minutes=4)
+    assert td // val is expected
+
+
+@pytest.mark.parametrize(
+    "op_name",
+    ["left_plus_right", "right_plus_left", "left_minus_right", "right_minus_left"],
+)
+@pytest.mark.parametrize(
+    "value",
+    [
+        DatetimeIndex(["2011-01-01", "2011-01-02"], dtype="M8[ns]", name="x"),
+        DatetimeIndex(
+            ["2011-01-01", "2011-01-02"], dtype="M8[ns, US/Eastern]", name="x"
+        ),
+        DatetimeArray._from_sequence(["2011-01-01", "2011-01-02"], dtype="M8[ns]"),
+        DatetimeArray._from_sequence(
+            ["2011-01-01", "2011-01-02"], dtype=DatetimeTZDtype(tz="US/Pacific")
+        ),
+        TimedeltaIndex(["1 day", "2 day"], name="x"),
+    ],
+)
+def test_nat_arithmetic_index(op_name, value):
+    # see gh-11718
+    exp_name = "x"
+    exp_data = [NaT] * 2
+
+    if value.dtype.kind == "M" and "plus" in op_name:
+        expected = DatetimeIndex(exp_data, tz=value.tz, name=exp_name)
+    else:
+        expected = TimedeltaIndex(exp_data, name=exp_name)
+    expected = expected.as_unit(value.unit)
+
+    if not isinstance(value, Index):
+        expected = expected.array
+
+    op = _ops[op_name]
+    result = op(NaT, value)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "op_name",
+    ["left_plus_right", "right_plus_left", "left_minus_right", "right_minus_left"],
+)
+@pytest.mark.parametrize("box", [TimedeltaIndex, Series, TimedeltaArray._from_sequence])
+def test_nat_arithmetic_td64_vector(op_name, box):
+    # see gh-19124
+    vec = box(["1 day", "2 day"], dtype="timedelta64[ns]")
+    box_nat = box([NaT, NaT], dtype="timedelta64[ns]")
+    tm.assert_equal(_ops[op_name](vec, NaT), box_nat)
+
+
+@pytest.mark.parametrize(
+    "dtype,op,out_dtype",
+    [
+        ("datetime64[ns]", operator.add, "datetime64[ns]"),
+        ("datetime64[ns]", roperator.radd, "datetime64[ns]"),
+        ("datetime64[ns]", operator.sub, "timedelta64[ns]"),
+        ("datetime64[ns]", roperator.rsub, "timedelta64[ns]"),
+        ("timedelta64[ns]", operator.add, "datetime64[ns]"),
+        ("timedelta64[ns]", roperator.radd, "datetime64[ns]"),
+        ("timedelta64[ns]", operator.sub, "datetime64[ns]"),
+        ("timedelta64[ns]", roperator.rsub, "timedelta64[ns]"),
+    ],
+)
+def test_nat_arithmetic_ndarray(dtype, op, out_dtype):
+    other = np.arange(10).astype(dtype)
+    result = op(NaT, other)
+
+    expected = np.empty(other.shape, dtype=out_dtype)
+    expected.fill("NaT")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_nat_pinned_docstrings():
+    # see gh-17327
+    assert NaT.ctime.__doc__ == Timestamp.ctime.__doc__
+
+
+def test_to_numpy_alias():
+    # GH 24653: alias .to_numpy() for scalars
+    expected = NaT.to_datetime64()
+    result = NaT.to_numpy()
+
+    assert isna(expected) and isna(result)
+
+    # GH#44460
+    result = NaT.to_numpy("M8[s]")
+    assert isinstance(result, np.datetime64)
+    assert result.dtype == "M8[s]"
+
+    result = NaT.to_numpy("m8[ns]")
+    assert isinstance(result, np.timedelta64)
+    assert result.dtype == "m8[ns]"
+
+    result = NaT.to_numpy("m8[s]")
+    assert isinstance(result, np.timedelta64)
+    assert result.dtype == "m8[s]"
+
+    with pytest.raises(ValueError, match="NaT.to_numpy dtype must be a "):
+        NaT.to_numpy(np.int64)
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        Timedelta(0),
+        Timedelta(0).to_pytimedelta(),
+        pytest.param(
+            Timedelta(0).to_timedelta64(),
+            marks=pytest.mark.xfail(
+                not np_version_gte1p24p3,
+                reason="td64 doesn't return NotImplemented, see numpy#17017",
+                # When this xfail is fixed, test_nat_comparisons_numpy
+                #  can be removed.
+            ),
+        ),
+        Timestamp(0),
+        Timestamp(0).to_pydatetime(),
+        pytest.param(
+            Timestamp(0).to_datetime64(),
+            marks=pytest.mark.xfail(
+                not np_version_gte1p24p3,
+                reason="dt64 doesn't return NotImplemented, see numpy#17017",
+            ),
+        ),
+        Timestamp(0).tz_localize("UTC"),
+        NaT,
+    ],
+)
+def test_nat_comparisons(compare_operators_no_eq_ne, other):
+    # GH 26039
+    opname = compare_operators_no_eq_ne
+
+    assert getattr(NaT, opname)(other) is False
+
+    op = getattr(operator, opname.strip("_"))
+    assert op(NaT, other) is False
+    assert op(other, NaT) is False
+
+
+@pytest.mark.parametrize("other", [np.timedelta64(0, "ns"), np.datetime64("now", "ns")])
+def test_nat_comparisons_numpy(other):
+    # Once numpy#17017 is fixed and the xfailed cases in test_nat_comparisons
+    #  pass, this test can be removed
+    assert not NaT == other
+    assert NaT != other
+    assert not NaT < other
+    assert not NaT > other
+    assert not NaT <= other
+    assert not NaT >= other
+
+
+@pytest.mark.parametrize("other_and_type", [("foo", "str"), (2, "int"), (2.0, "float")])
+@pytest.mark.parametrize(
+    "symbol_and_op",
+    [("<=", operator.le), ("<", operator.lt), (">=", operator.ge), (">", operator.gt)],
+)
+def test_nat_comparisons_invalid(other_and_type, symbol_and_op):
+    # GH#35585
+    other, other_type = other_and_type
+    symbol, op = symbol_and_op
+
+    assert not NaT == other
+    assert not other == NaT
+
+    assert NaT != other
+    assert other != NaT
+
+    msg = f"'{symbol}' not supported between instances of 'NaTType' and '{other_type}'"
+    with pytest.raises(TypeError, match=msg):
+        op(NaT, other)
+
+    msg = f"'{symbol}' not supported between instances of '{other_type}' and 'NaTType'"
+    with pytest.raises(TypeError, match=msg):
+        op(other, NaT)
+
+
+@pytest.mark.parametrize(
+    "other",
+    [
+        np.array(["foo"] * 2, dtype=object),
+        np.array([2, 3], dtype="int64"),
+        np.array([2.0, 3.5], dtype="float64"),
+    ],
+    ids=["str", "int", "float"],
+)
+def test_nat_comparisons_invalid_ndarray(other):
+    # GH#40722
+    expected = np.array([False, False])
+    result = NaT == other
+    tm.assert_numpy_array_equal(result, expected)
+    result = other == NaT
+    tm.assert_numpy_array_equal(result, expected)
+
+    expected = np.array([True, True])
+    result = NaT != other
+    tm.assert_numpy_array_equal(result, expected)
+    result = other != NaT
+    tm.assert_numpy_array_equal(result, expected)
+
+    for symbol, op in [
+        ("<=", operator.le),
+        ("<", operator.lt),
+        (">=", operator.ge),
+        (">", operator.gt),
+    ]:
+        msg = f"'{symbol}' not supported between"
+
+        with pytest.raises(TypeError, match=msg):
+            op(NaT, other)
+
+        if other.dtype == np.dtype("object"):
+            # uses the reverse operator, so symbol changes
+            msg = None
+        with pytest.raises(TypeError, match=msg):
+            op(other, NaT)
+
+
+def test_compare_date(fixed_now_ts):
+    # GH#39151 comparing NaT with date object is deprecated
+    # See also: tests.scalar.timestamps.test_comparisons::test_compare_date
+
+    dt = fixed_now_ts.to_pydatetime().date()
+
+    msg = "Cannot compare NaT with datetime.date object"
+    for left, right in [(NaT, dt), (dt, NaT)]:
+        assert not left == right
+        assert left != right
+
+        with pytest.raises(TypeError, match=msg):
+            left < right
+        with pytest.raises(TypeError, match=msg):
+            left <= right
+        with pytest.raises(TypeError, match=msg):
+            left > right
+        with pytest.raises(TypeError, match=msg):
+            left >= right
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        offsets.YearEnd(2),
+        offsets.YearBegin(2),
+        offsets.MonthBegin(1),
+        offsets.MonthEnd(2),
+        offsets.MonthEnd(12),
+        offsets.Day(2),
+        offsets.Day(5),
+        offsets.Hour(24),
+        offsets.Hour(3),
+        offsets.Minute(),
+        np.timedelta64(3, "h"),
+        np.timedelta64(4, "h"),
+        np.timedelta64(3200, "s"),
+        np.timedelta64(3600, "s"),
+        np.timedelta64(3600 * 24, "s"),
+        np.timedelta64(2, "D"),
+        np.timedelta64(365, "D"),
+        timedelta(-2),
+        timedelta(365),
+        timedelta(minutes=120),
+        timedelta(days=4, minutes=180),
+        timedelta(hours=23),
+        timedelta(hours=23, minutes=30),
+        timedelta(hours=48),
+    ],
+)
+def test_nat_addsub_tdlike_scalar(obj):
+    assert NaT + obj is NaT
+    assert obj + NaT is NaT
+    assert NaT - obj is NaT
+
+
+def test_pickle():
+    # GH#4606
+    p = tm.round_trip_pickle(NaT)
+    assert p is NaT
diff --git a/pandas/pandas/tests/scalar/timedelta/__init__.py b/pandas/pandas/tests/scalar/timedelta/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/timedelta/methods/__init__.py b/pandas/pandas/tests/scalar/timedelta/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/timedelta/methods/test_as_unit.py b/pandas/pandas/tests/scalar/timedelta/methods/test_as_unit.py
new file mode 100644
index 0000000000000..8660141e5a537
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/methods/test_as_unit.py
@@ -0,0 +1,80 @@
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.errors import OutOfBoundsTimedelta
+
+from pandas import Timedelta
+
+
+class TestAsUnit:
+    def test_as_unit(self):
+        td = Timedelta(days=1)
+
+        assert td.as_unit("ns") is td
+
+        res = td.as_unit("us")
+        assert res._value == td._value // 1000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == td._value
+        assert rt._creso == td._creso
+
+        res = td.as_unit("ms")
+        assert res._value == td._value // 1_000_000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == td._value
+        assert rt._creso == td._creso
+
+        res = td.as_unit("s")
+        assert res._value == td._value // 1_000_000_000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == td._value
+        assert rt._creso == td._creso
+
+    def test_as_unit_overflows(self):
+        # microsecond that would be just out of bounds for nano
+        us = 9223372800000000
+        td = Timedelta._from_value_and_reso(us, NpyDatetimeUnit.NPY_FR_us.value)
+
+        msg = "Cannot cast 106752 days 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            td.as_unit("ns")
+
+        res = td.as_unit("ms")
+        assert res._value == us // 1000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+    def test_as_unit_rounding(self):
+        td = Timedelta(microseconds=1500)
+        res = td.as_unit("ms")
+
+        expected = Timedelta(milliseconds=1)
+        assert res == expected
+
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+        assert res._value == 1
+
+        with pytest.raises(ValueError, match="Cannot losslessly convert units"):
+            td.as_unit("ms", round_ok=False)
+
+    def test_as_unit_non_nano(self):
+        # case where we are going neither to nor from nano
+        td = Timedelta(days=1).as_unit("ms")
+        assert td.days == 1
+        assert td._value == 86_400_000
+        assert td.components.days == 1
+        assert td._d == 1
+        assert td.total_seconds() == 86400
+
+        res = td.as_unit("us")
+        assert res._value == 86_400_000_000
+        assert res.components.days == 1
+        assert res.components.hours == 0
+        assert res._d == 1
+        assert res._h == 0
+        assert res.total_seconds() == 86400
diff --git a/pandas/pandas/tests/scalar/timedelta/methods/test_round.py b/pandas/pandas/tests/scalar/timedelta/methods/test_round.py
new file mode 100644
index 0000000000000..96cb1c07d2b76
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/methods/test_round.py
@@ -0,0 +1,188 @@
+from hypothesis import (
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._libs.tslibs import iNaT
+from pandas.errors import OutOfBoundsTimedelta
+
+from pandas import Timedelta
+
+
+class TestTimedeltaRound:
+    @pytest.mark.parametrize(
+        "freq,s1,s2",
+        [
+            # This first case has s1, s2 being the same as t1,t2 below
+            (
+                "ns",
+                "1 days 02:34:56.789123456",
+                "-1 days 02:34:56.789123456",
+            ),
+            (
+                "us",
+                "1 days 02:34:56.789123000",
+                "-1 days 02:34:56.789123000",
+            ),
+            (
+                "ms",
+                "1 days 02:34:56.789000000",
+                "-1 days 02:34:56.789000000",
+            ),
+            ("s", "1 days 02:34:57", "-1 days 02:34:57"),
+            ("2s", "1 days 02:34:56", "-1 days 02:34:56"),
+            ("5s", "1 days 02:34:55", "-1 days 02:34:55"),
+            ("min", "1 days 02:35:00", "-1 days 02:35:00"),
+            ("12min", "1 days 02:36:00", "-1 days 02:36:00"),
+            ("h", "1 days 03:00:00", "-1 days 03:00:00"),
+            ("D", "1 days", "-1 days"),
+        ],
+    )
+    def test_round(self, freq, s1, s2):
+        s1 = Timedelta(s1)
+        s2 = Timedelta(s2)
+        t1 = Timedelta("1 days 02:34:56.789123456")
+        t2 = Timedelta("-1 days 02:34:56.789123456")
+
+        r1 = t1.round(freq)
+        assert r1 == s1
+        r2 = t2.round(freq)
+        assert r2 == s2
+
+    def test_round_invalid(self):
+        t1 = Timedelta("1 days 02:34:56.789123456")
+
+        for freq, msg in [
+            ("YE", "<YearEnd: month=12> is a non-fixed frequency"),
+            ("ME", "<MonthEnd> is a non-fixed frequency"),
+            ("foobar", "Invalid frequency: foobar"),
+        ]:
+            with pytest.raises(ValueError, match=msg):
+                t1.round(freq)
+
+    @pytest.mark.skip_ubsan
+    def test_round_implementation_bounds(self):
+        # See also: analogous test for Timestamp
+        # GH#38964
+        result = Timedelta.min.ceil("s")
+        expected = Timedelta.min + Timedelta(seconds=1) - Timedelta(145224193)
+        assert result == expected
+
+        result = Timedelta.max.floor("s")
+        expected = Timedelta.max - Timedelta(854775807)
+        assert result == expected
+
+        msg = (
+            r"Cannot round -106752 days \+00:12:43.145224193 to freq=s without overflow"
+        )
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta.min.floor("s")
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta.min.round("s")
+
+        msg = "Cannot round 106751 days 23:47:16.854775807 to freq=s without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta.max.ceil("s")
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta.max.round("s")
+
+    @pytest.mark.skip_ubsan
+    @given(val=st.integers(min_value=iNaT + 1, max_value=lib.i8max))
+    @pytest.mark.parametrize(
+        "method", [Timedelta.round, Timedelta.floor, Timedelta.ceil]
+    )
+    def test_round_sanity(self, val, method):
+        cls = Timedelta
+        err_cls = OutOfBoundsTimedelta
+
+        val = np.int64(val)
+        td = cls(val)
+
+        def checker(ts, nanos, unit):
+            # First check that we do raise in cases where we should
+            if nanos == 1:
+                pass
+            else:
+                div, mod = divmod(ts._value, nanos)
+                diff = int(nanos - mod)
+                lb = ts._value - mod
+                assert lb <= ts._value  # i.e. no overflows with python ints
+                ub = ts._value + diff
+                assert ub > ts._value  # i.e. no overflows with python ints
+
+                msg = "without overflow"
+                if mod == 0:
+                    # We should never be raising in this
+                    pass
+                elif method is cls.ceil:
+                    if ub > cls.max._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif method is cls.floor:
+                    if lb < cls.min._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif mod >= diff:
+                    if ub > cls.max._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif lb < cls.min._value:
+                    with pytest.raises(err_cls, match=msg):
+                        method(ts, unit)
+                    return
+
+            res = method(ts, unit)
+
+            td = res - ts
+            diff = abs(td._value)
+            assert diff < nanos
+            assert res._value % nanos == 0
+
+            if method is cls.round:
+                assert diff <= nanos / 2
+            elif method is cls.floor:
+                assert res <= ts
+            elif method is cls.ceil:
+                assert res >= ts
+
+        nanos = 1
+        checker(td, nanos, "ns")
+
+        nanos = 1000
+        checker(td, nanos, "us")
+
+        nanos = 1_000_000
+        checker(td, nanos, "ms")
+
+        nanos = 1_000_000_000
+        checker(td, nanos, "s")
+
+        nanos = 60 * 1_000_000_000
+        checker(td, nanos, "min")
+
+        nanos = 60 * 60 * 1_000_000_000
+        checker(td, nanos, "h")
+
+        nanos = 24 * 60 * 60 * 1_000_000_000
+        checker(td, nanos, "D")
+
+    def test_round_non_nano(self, unit):
+        td = Timedelta("1 days 02:34:57").as_unit(unit)
+
+        res = td.round("min")
+        assert res == Timedelta("1 days 02:35:00")
+        assert res._creso == td._creso
+
+        res = td.floor("min")
+        assert res == Timedelta("1 days 02:34:00")
+        assert res._creso == td._creso
+
+        res = td.ceil("min")
+        assert res == Timedelta("1 days 02:35:00")
+        assert res._creso == td._creso
diff --git a/pandas/pandas/tests/scalar/timedelta/test_arithmetic.py b/pandas/pandas/tests/scalar/timedelta/test_arithmetic.py
new file mode 100644
index 0000000000000..2183a5851ea9c
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/test_arithmetic.py
@@ -0,0 +1,1185 @@
+"""
+Tests for scalar Timedelta arithmetic ops
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+import operator
+
+import numpy as np
+import pytest
+
+from pandas.errors import OutOfBoundsTimedelta
+
+import pandas as pd
+from pandas import (
+    NaT,
+    Timedelta,
+    Timestamp,
+    offsets,
+)
+import pandas._testing as tm
+from pandas.core import ops
+
+
+class TestTimedeltaAdditionSubtraction:
+    """
+    Tests for Timedelta methods:
+
+        __add__, __radd__,
+        __sub__, __rsub__
+    """
+
+    @pytest.mark.parametrize(
+        "ten_seconds",
+        [
+            Timedelta(10, unit="s"),
+            timedelta(seconds=10),
+            np.timedelta64(10, "s"),
+            np.timedelta64(10000000000, "ns"),
+            offsets.Second(10),
+        ],
+    )
+    def test_td_add_sub_ten_seconds(self, ten_seconds):
+        # GH#6808
+        base = Timestamp("20130101 09:01:12.123456")
+        expected_add = Timestamp("20130101 09:01:22.123456")
+        expected_sub = Timestamp("20130101 09:01:02.123456")
+
+        result = base + ten_seconds
+        assert result == expected_add
+
+        result = base - ten_seconds
+        assert result == expected_sub
+
+    @pytest.mark.parametrize(
+        "one_day_ten_secs",
+        [
+            Timedelta("1 day, 00:00:10"),
+            Timedelta("1 days, 00:00:10"),
+            timedelta(days=1, seconds=10),
+            np.timedelta64(1, "D") + np.timedelta64(10, "s"),
+            offsets.Day() + offsets.Second(10),
+        ],
+    )
+    def test_td_add_sub_one_day_ten_seconds(self, one_day_ten_secs):
+        # GH#6808
+        base = Timestamp("20130102 09:01:12.123456")
+        expected_add = Timestamp("20130103 09:01:22.123456")
+        expected_sub = Timestamp("20130101 09:01:02.123456")
+
+        result = base + one_day_ten_secs
+        assert result == expected_add
+
+        result = base - one_day_ten_secs
+        assert result == expected_sub
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_datetimelike_scalar(self, op):
+        # GH#19738
+        td = Timedelta(10, unit="D")
+
+        result = op(td, datetime(2016, 1, 1))
+        if op is operator.add:
+            # datetime + Timedelta does _not_ call Timedelta.__radd__,
+            # so we get a datetime back instead of a Timestamp
+            assert isinstance(result, Timestamp)
+        assert result == Timestamp(2016, 1, 11)
+
+        result = op(td, Timestamp("2018-01-12 18:09"))
+        assert isinstance(result, Timestamp)
+        assert result == Timestamp("2018-01-22 18:09")
+
+        result = op(td, np.datetime64("2018-01-12"))
+        assert isinstance(result, Timestamp)
+        assert result == Timestamp("2018-01-22")
+
+        result = op(td, NaT)
+        assert result is NaT
+
+    def test_td_add_timestamp_overflow(self):
+        ts = Timestamp("1700-01-01").as_unit("ns")
+        msg = "Cannot cast 259987 from D to 'ns' without overflow."
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            ts + Timedelta(13 * 19999, unit="D")
+
+        msg = "Cannot cast 259987 days 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            ts + timedelta(days=13 * 19999)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_td(self, op):
+        td = Timedelta(10, unit="D")
+
+        result = op(td, Timedelta(days=10))
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=20)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_pytimedelta(self, op):
+        td = Timedelta(10, unit="D")
+        result = op(td, timedelta(days=9))
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=19)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_timedelta64(self, op):
+        td = Timedelta(10, unit="D")
+        result = op(td, np.timedelta64(-4, "D"))
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=6)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_offset(self, op):
+        td = Timedelta(10, unit="D")
+
+        result = op(td, offsets.Hour(6))
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=10, hours=6)
+
+    def test_td_sub_td(self):
+        td = Timedelta(10, unit="D")
+        expected = Timedelta(0, unit="ns")
+        result = td - td
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+    def test_td_sub_pytimedelta(self):
+        td = Timedelta(10, unit="D")
+        expected = Timedelta(0, unit="ns")
+
+        result = td - td.to_pytimedelta()
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = td.to_pytimedelta() - td
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+    def test_td_sub_timedelta64(self):
+        td = Timedelta(10, unit="D")
+        expected = Timedelta(0, unit="ns")
+
+        result = td - td.to_timedelta64()
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = td.to_timedelta64() - td
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+    def test_td_sub_nat(self):
+        # In this context pd.NaT is treated as timedelta-like
+        td = Timedelta(10, unit="D")
+        result = td - NaT
+        assert result is NaT
+
+    def test_td_sub_td64_nat(self):
+        td = Timedelta(10, unit="D")
+        td_nat = np.timedelta64("NaT")
+
+        result = td - td_nat
+        assert result is NaT
+
+        result = td_nat - td
+        assert result is NaT
+
+    def test_td_sub_offset(self):
+        td = Timedelta(10, unit="D")
+        result = td - offsets.Hour(1)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(239, unit="h")
+
+    def test_td_add_sub_numeric_raises(self):
+        td = Timedelta(10, unit="D")
+        msg = "unsupported operand type"
+        for other in [2, 2.0, np.int64(2), np.float64(2)]:
+            with pytest.raises(TypeError, match=msg):
+                td + other
+            with pytest.raises(TypeError, match=msg):
+                other + td
+            with pytest.raises(TypeError, match=msg):
+                td - other
+            with pytest.raises(TypeError, match=msg):
+                other - td
+
+    def test_td_add_sub_int_ndarray(self):
+        td = Timedelta("1 day")
+        other = np.array([1])
+
+        msg = r"unsupported operand type\(s\) for \+: 'Timedelta' and 'int'"
+        with pytest.raises(TypeError, match=msg):
+            td + np.array([1])
+
+        msg = "|".join(
+            [
+                (
+                    r"unsupported operand type\(s\) for \+: 'numpy.ndarray' "
+                    "and 'Timedelta'"
+                ),
+                # This message goes on to say "Please do not rely on this error;
+                #  it may not be given on all Python implementations"
+                "Concatenation operation is not implemented for NumPy arrays",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            other + td
+        msg = r"unsupported operand type\(s\) for -: 'Timedelta' and 'int'"
+        with pytest.raises(TypeError, match=msg):
+            td - other
+        msg = r"unsupported operand type\(s\) for -: 'numpy.ndarray' and 'Timedelta'"
+        with pytest.raises(TypeError, match=msg):
+            other - td
+
+    def test_td_rsub_nat(self):
+        td = Timedelta(10, unit="D")
+        result = NaT - td
+        assert result is NaT
+
+        result = np.datetime64("NaT") - td
+        assert result is NaT
+
+    def test_td_rsub_offset(self):
+        result = offsets.Hour(1) - Timedelta(10, unit="D")
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(-239, unit="h")
+
+    def test_td_sub_timedeltalike_object_dtype_array(self):
+        # GH#21980
+        arr = np.array([Timestamp("20130101 9:01"), Timestamp("20121230 9:02")])
+        exp = np.array([Timestamp("20121231 9:01"), Timestamp("20121229 9:02")])
+        res = arr - Timedelta("1D")
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_td_sub_mixed_most_timedeltalike_object_dtype_array(self):
+        # GH#21980
+        now = Timestamp("2021-11-09 09:54:00")
+        arr = np.array([now, Timedelta("1D"), np.timedelta64(2, "h")])
+        exp = np.array(
+            [
+                now - Timedelta("1D"),
+                Timedelta("0D"),
+                np.timedelta64(2, "h") - Timedelta("1D"),
+            ]
+        )
+        res = arr - Timedelta("1D")
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_td_rsub_mixed_most_timedeltalike_object_dtype_array(self):
+        # GH#21980
+        now = Timestamp("2021-11-09 09:54:00")
+        arr = np.array([now, Timedelta("1D"), np.timedelta64(2, "h")])
+        msg = r"unsupported operand type\(s\) for \-: 'Timedelta' and 'Timestamp'"
+        with pytest.raises(TypeError, match=msg):
+            Timedelta("1D") - arr
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_timedeltalike_object_dtype_array(self, op):
+        # GH#21980
+        arr = np.array([Timestamp("20130101 9:01"), Timestamp("20121230 9:02")])
+        exp = np.array([Timestamp("20130102 9:01"), Timestamp("20121231 9:02")])
+        res = op(arr, Timedelta("1D"))
+        tm.assert_numpy_array_equal(res, exp)
+
+    @pytest.mark.parametrize("op", [operator.add, ops.radd])
+    def test_td_add_mixed_timedeltalike_object_dtype_array(self, op):
+        # GH#21980
+        now = Timestamp("2021-11-09 09:54:00")
+        arr = np.array([now, Timedelta("1D")])
+        exp = np.array([now + Timedelta("1D"), Timedelta("2D")])
+        res = op(arr, Timedelta("1D"))
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_td_add_sub_td64_ndarray(self):
+        td = Timedelta("1 day")
+
+        other = np.array([td.to_timedelta64()])
+        expected = np.array([Timedelta("2 Days").to_timedelta64()])
+
+        result = td + other
+        tm.assert_numpy_array_equal(result, expected)
+        result = other + td
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = td - other
+        tm.assert_numpy_array_equal(result, expected * 0)
+        result = other - td
+        tm.assert_numpy_array_equal(result, expected * 0)
+
+    def test_td_add_sub_dt64_ndarray(self):
+        td = Timedelta("1 day")
+        other = np.array(["2000-01-01"], dtype="M8[ns]")
+
+        expected = np.array(["2000-01-02"], dtype="M8[ns]")
+        tm.assert_numpy_array_equal(td + other, expected)
+        tm.assert_numpy_array_equal(other + td, expected)
+
+        expected = np.array(["1999-12-31"], dtype="M8[ns]")
+        tm.assert_numpy_array_equal(-td + other, expected)
+        tm.assert_numpy_array_equal(other - td, expected)
+
+    def test_td_add_sub_ndarray_0d(self):
+        td = Timedelta("1 day")
+        other = np.array(td.asm8)
+
+        result = td + other
+        assert isinstance(result, Timedelta)
+        assert result == 2 * td
+
+        result = other + td
+        assert isinstance(result, Timedelta)
+        assert result == 2 * td
+
+        result = other - td
+        assert isinstance(result, Timedelta)
+        assert result == 0 * td
+
+        result = td - other
+        assert isinstance(result, Timedelta)
+        assert result == 0 * td
+
+
+class TestTimedeltaMultiplicationDivision:
+    """
+    Tests for Timedelta methods:
+
+        __mul__, __rmul__,
+        __div__, __rdiv__,
+        __truediv__, __rtruediv__,
+        __floordiv__, __rfloordiv__,
+        __mod__, __rmod__,
+        __divmod__, __rdivmod__
+    """
+
+    # ---------------------------------------------------------------
+    # Timedelta.__mul__, __rmul__
+
+    @pytest.mark.parametrize(
+        "td_nat", [NaT, np.timedelta64("NaT", "ns"), np.timedelta64("NaT")]
+    )
+    @pytest.mark.parametrize("op", [operator.mul, ops.rmul])
+    def test_td_mul_nat(self, op, td_nat):
+        # GH#19819
+        td = Timedelta(10, unit="D")
+        typs = "|".join(["numpy.timedelta64", "NaTType", "Timedelta"])
+        msg = "|".join(
+            [
+                rf"unsupported operand type\(s\) for \*: '{typs}' and '{typs}'",
+                r"ufunc '?multiply'? cannot use operands with types",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            op(td, td_nat)
+
+    @pytest.mark.parametrize("nan", [np.nan, np.float64("NaN"), float("nan")])
+    @pytest.mark.parametrize("op", [operator.mul, ops.rmul])
+    def test_td_mul_nan(self, op, nan):
+        # np.float64('NaN') has a 'dtype' attr, avoid treating as array
+        td = Timedelta(10, unit="D")
+        result = op(td, nan)
+        assert result is NaT
+
+    @pytest.mark.parametrize("op", [operator.mul, ops.rmul])
+    def test_td_mul_scalar(self, op):
+        # GH#19738
+        td = Timedelta(minutes=3)
+
+        result = op(td, 2)
+        assert result == Timedelta(minutes=6)
+
+        result = op(td, 1.5)
+        assert result == Timedelta(minutes=4, seconds=30)
+
+        assert op(td, np.nan) is NaT
+
+        assert op(-1, td)._value == -1 * td._value
+        assert op(-1.0, td)._value == -1.0 * td._value
+
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            # timedelta * datetime is gibberish
+            op(td, Timestamp(2016, 1, 2))
+
+        with pytest.raises(TypeError, match=msg):
+            # invalid multiply with another timedelta
+            op(td, td)
+
+    def test_td_mul_numeric_ndarray(self):
+        td = Timedelta("1 day")
+        other = np.array([2])
+        expected = np.array([Timedelta("2 Days").to_timedelta64()])
+
+        result = td * other
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = other * td
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_td_mul_numeric_ndarray_0d(self):
+        td = Timedelta("1 day")
+        other = np.array(2, dtype=np.int64)
+        assert other.ndim == 0
+        expected = Timedelta("2 days")
+
+        res = td * other
+        assert type(res) is Timedelta
+        assert res == expected
+
+        res = other * td
+        assert type(res) is Timedelta
+        assert res == expected
+
+    def test_td_mul_td64_ndarray_invalid(self):
+        td = Timedelta("1 day")
+        other = np.array([Timedelta("2 Days").to_timedelta64()])
+
+        msg = (
+            "ufunc '?multiply'? cannot use operands with types "
+            rf"dtype\('{tm.ENDIAN}m8\[ns\]'\) and dtype\('{tm.ENDIAN}m8\[ns\]'\)"
+        )
+        with pytest.raises(TypeError, match=msg):
+            td * other
+        with pytest.raises(TypeError, match=msg):
+            other * td
+
+    # ---------------------------------------------------------------
+    # Timedelta.__div__, __truediv__
+
+    def test_td_div_timedeltalike_scalar(self):
+        # GH#19738
+        td = Timedelta(10, unit="D")
+
+        result = td / offsets.Hour(1)
+        assert result == 240
+
+        assert td / td == 1
+        assert td / np.timedelta64(60, "h") == 4
+
+        assert np.isnan(td / NaT)
+
+    def test_td_div_td64_non_nano(self):
+        # truediv
+        td = Timedelta("1 days 2 hours 3 ns")
+        result = td / np.timedelta64(1, "D")
+        assert result == td._value / (86400 * 10**9)
+        result = td / np.timedelta64(1, "s")
+        assert result == td._value / 10**9
+        result = td / np.timedelta64(1, "ns")
+        assert result == td._value
+
+        # floordiv
+        td = Timedelta("1 days 2 hours 3 ns")
+        result = td // np.timedelta64(1, "D")
+        assert result == 1
+        result = td // np.timedelta64(1, "s")
+        assert result == 93600
+        result = td // np.timedelta64(1, "ns")
+        assert result == td._value
+
+    def test_td_div_numeric_scalar(self):
+        # GH#19738
+        td = Timedelta(10, unit="D")
+
+        result = td / 2
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=5)
+
+        result = td / 5
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(days=2)
+
+    @pytest.mark.parametrize(
+        "nan",
+        [
+            np.nan,
+            np.float64("NaN"),
+            float("nan"),
+        ],
+    )
+    def test_td_div_nan(self, nan):
+        # np.float64('NaN') has a 'dtype' attr, avoid treating as array
+        td = Timedelta(10, unit="D")
+        result = td / nan
+        assert result is NaT
+
+        result = td // nan
+        assert result is NaT
+
+    def test_td_div_td64_ndarray(self):
+        td = Timedelta("1 day")
+
+        other = np.array([Timedelta("2 Days").to_timedelta64()])
+        expected = np.array([0.5])
+
+        result = td / other
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = other / td
+        tm.assert_numpy_array_equal(result, expected * 4)
+
+    def test_td_div_ndarray_0d(self):
+        td = Timedelta("1 day")
+
+        other = np.array(1)
+        res = td / other
+        assert isinstance(res, Timedelta)
+        assert res == td
+
+    # ---------------------------------------------------------------
+    # Timedelta.__rdiv__
+
+    def test_td_rdiv_timedeltalike_scalar(self):
+        # GH#19738
+        td = Timedelta(10, unit="D")
+        result = offsets.Hour(1) / td
+        assert result == 1 / 240.0
+
+        assert np.timedelta64(60, "h") / td == 0.25
+
+    def test_td_rdiv_na_scalar(self):
+        # GH#31869 None gets cast to NaT
+        td = Timedelta(10, unit="D")
+
+        result = NaT / td
+        assert np.isnan(result)
+
+        result = None / td
+        assert np.isnan(result)
+
+        result = np.timedelta64("NaT") / td
+        assert np.isnan(result)
+
+        msg = r"unsupported operand type\(s\) for /: 'numpy.datetime64' and 'Timedelta'"
+        with pytest.raises(TypeError, match=msg):
+            np.datetime64("NaT") / td
+
+        msg = r"unsupported operand type\(s\) for /: 'float' and 'Timedelta'"
+        with pytest.raises(TypeError, match=msg):
+            np.nan / td
+
+    def test_td_rdiv_ndarray(self):
+        td = Timedelta(10, unit="D")
+
+        arr = np.array([td], dtype=object)
+        result = arr / td
+        expected = np.array([1], dtype=np.float64)
+        tm.assert_numpy_array_equal(result, expected)
+
+        arr = np.array([None])
+        result = arr / td
+        expected = np.array([np.nan])
+        tm.assert_numpy_array_equal(result, expected)
+
+        arr = np.array([np.nan], dtype=object)
+        msg = r"unsupported operand type\(s\) for /: 'float' and 'Timedelta'"
+        with pytest.raises(TypeError, match=msg):
+            arr / td
+
+        arr = np.array([np.nan], dtype=np.float64)
+        msg = "cannot use operands with types dtype"
+        with pytest.raises(TypeError, match=msg):
+            arr / td
+
+    def test_td_rdiv_ndarray_0d(self):
+        td = Timedelta(10, unit="D")
+
+        arr = np.array(td.asm8)
+
+        assert arr / td == 1
+
+    # ---------------------------------------------------------------
+    # Timedelta.__floordiv__
+
+    def test_td_floordiv_timedeltalike_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+        scalar = Timedelta(hours=3, minutes=3)
+
+        assert td // scalar == 1
+        assert -td // scalar.to_pytimedelta() == -2
+        assert (2 * td) // scalar.to_timedelta64() == 2
+
+    def test_td_floordiv_null_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+
+        assert td // np.nan is NaT
+        assert np.isnan(td // NaT)
+        assert np.isnan(td // np.timedelta64("NaT"))
+
+    def test_td_floordiv_offsets(self):
+        # GH#19738
+        td = Timedelta(hours=3, minutes=4)
+        assert td // offsets.Hour(1) == 3
+        assert td // offsets.Minute(2) == 92
+
+    def test_td_floordiv_invalid_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+
+        msg = "|".join(
+            [
+                r"Invalid dtype datetime64\[D\] for __floordiv__",
+                "'dtype' is an invalid keyword argument for this function",
+                "this function got an unexpected keyword argument 'dtype'",
+                r"ufunc '?floor_divide'? cannot use operands with types",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            td // np.datetime64("2016-01-01", dtype="datetime64[us]")
+
+    def test_td_floordiv_numeric_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+
+        expected = Timedelta(hours=1, minutes=32)
+        assert td // 2 == expected
+        assert td // 2.0 == expected
+        assert td // np.float64(2.0) == expected
+        assert td // np.int32(2.0) == expected
+        assert td // np.uint8(2.0) == expected
+
+    def test_td_floordiv_timedeltalike_array(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+        scalar = Timedelta(hours=3, minutes=3)
+
+        # Array-like others
+        assert td // np.array(scalar.to_timedelta64()) == 1
+
+        res = (3 * td) // np.array([scalar.to_timedelta64()])
+        expected = np.array([3], dtype=np.int64)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = (10 * td) // np.array([scalar.to_timedelta64(), np.timedelta64("NaT")])
+        expected = np.array([10, np.nan])
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_td_floordiv_numeric_series(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=4)
+        ser = pd.Series([1], dtype=np.int64)
+        res = td // ser
+        assert res.dtype.kind == "m"
+
+    # ---------------------------------------------------------------
+    # Timedelta.__rfloordiv__
+
+    def test_td_rfloordiv_timedeltalike_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+        scalar = Timedelta(hours=3, minutes=4)
+
+        # scalar others
+        # x // Timedelta is defined only for timedelta-like x. int-like,
+        # float-like, and date-like, in particular, should all either
+        # a) raise TypeError directly or
+        # b) return NotImplemented, following which the reversed
+        #    operation will raise TypeError.
+        assert td.__rfloordiv__(scalar) == 1
+        assert (-td).__rfloordiv__(scalar.to_pytimedelta()) == -2
+        assert (2 * td).__rfloordiv__(scalar.to_timedelta64()) == 0
+
+    def test_td_rfloordiv_null_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+
+        assert np.isnan(td.__rfloordiv__(NaT))
+        assert np.isnan(td.__rfloordiv__(np.timedelta64("NaT")))
+
+    def test_td_rfloordiv_offsets(self):
+        # GH#19738
+        assert offsets.Hour(1) // Timedelta(minutes=25) == 2
+
+    def test_td_rfloordiv_invalid_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+
+        dt64 = np.datetime64("2016-01-01", "us")
+
+        assert td.__rfloordiv__(dt64) is NotImplemented
+
+        msg = (
+            r"unsupported operand type\(s\) for //: 'numpy.datetime64' and 'Timedelta'"
+        )
+        with pytest.raises(TypeError, match=msg):
+            dt64 // td
+
+    def test_td_rfloordiv_numeric_scalar(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+
+        assert td.__rfloordiv__(np.nan) is NotImplemented
+        assert td.__rfloordiv__(3.5) is NotImplemented
+        assert td.__rfloordiv__(2) is NotImplemented
+        assert td.__rfloordiv__(np.float64(2.0)) is NotImplemented
+        assert td.__rfloordiv__(np.uint8(9)) is NotImplemented
+        assert td.__rfloordiv__(np.int32(2.0)) is NotImplemented
+
+        msg = r"unsupported operand type\(s\) for //: '.*' and 'Timedelta"
+        with pytest.raises(TypeError, match=msg):
+            np.float64(2.0) // td
+        with pytest.raises(TypeError, match=msg):
+            np.uint8(9) // td
+        with pytest.raises(TypeError, match=msg):
+            # deprecated GH#19761, enforced GH#29797
+            np.int32(2.0) // td
+
+    def test_td_rfloordiv_timedeltalike_array(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+        scalar = Timedelta(hours=3, minutes=4)
+
+        # Array-like others
+        assert td.__rfloordiv__(np.array(scalar.to_timedelta64())) == 1
+
+        res = td.__rfloordiv__(np.array([(3 * scalar).to_timedelta64()]))
+        expected = np.array([3], dtype=np.int64)
+        tm.assert_numpy_array_equal(res, expected)
+
+        arr = np.array([(10 * scalar).to_timedelta64(), np.timedelta64("NaT")])
+        res = td.__rfloordiv__(arr)
+        expected = np.array([10, np.nan])
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_td_rfloordiv_intarray(self):
+        # deprecated GH#19761, enforced GH#29797
+        ints = np.array([1349654400, 1349740800, 1349827200, 1349913600]) * 10**9
+
+        msg = "Invalid dtype"
+        with pytest.raises(TypeError, match=msg):
+            ints // Timedelta(1, unit="s")
+
+    def test_td_rfloordiv_numeric_series(self):
+        # GH#18846
+        td = Timedelta(hours=3, minutes=3)
+        ser = pd.Series([1], dtype=np.int64)
+        res = td.__rfloordiv__(ser)
+        assert res is NotImplemented
+
+        msg = "Invalid dtype"
+        with pytest.raises(TypeError, match=msg):
+            # Deprecated GH#19761, enforced GH#29797
+            ser // td
+
+    # ----------------------------------------------------------------
+    # Timedelta.__mod__, __rmod__
+
+    def test_mod_timedeltalike(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+
+        # Timedelta-like others
+        result = td % Timedelta(hours=6)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(hours=1)
+
+        result = td % timedelta(minutes=60)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(0)
+
+        result = td % NaT
+        assert result is NaT
+
+    def test_mod_timedelta64_nat(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+
+        result = td % np.timedelta64("NaT", "ns")
+        assert result is NaT
+
+    def test_mod_timedelta64(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+
+        result = td % np.timedelta64(2, "h")
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(hours=1)
+
+    def test_mod_offset(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+
+        result = td % offsets.Hour(5)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(hours=2)
+
+    def test_mod_numeric(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+
+        # Numeric Others
+        result = td % 2
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(0)
+
+        result = td % 1e12
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(minutes=3, seconds=20)
+
+        result = td % int(1e12)
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(minutes=3, seconds=20)
+
+    def test_mod_invalid(self):
+        # GH#19365
+        td = Timedelta(hours=37)
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            td % Timestamp("2018-01-22")
+
+        with pytest.raises(TypeError, match=msg):
+            td % []
+
+    def test_rmod_pytimedelta(self):
+        # GH#19365
+        td = Timedelta(minutes=3)
+
+        result = timedelta(minutes=4) % td
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(minutes=1)
+
+    def test_rmod_timedelta64(self):
+        # GH#19365
+        td = Timedelta(minutes=3)
+        result = np.timedelta64(5, "m") % td
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta(minutes=2)
+
+    def test_rmod_invalid(self):
+        # GH#19365
+        td = Timedelta(minutes=3)
+
+        msg = "unsupported operand"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp("2018-01-22") % td
+
+        with pytest.raises(TypeError, match=msg):
+            15 % td
+
+        with pytest.raises(TypeError, match=msg):
+            16.0 % td
+
+        msg = "Invalid dtype int"
+        with pytest.raises(TypeError, match=msg):
+            np.array([22, 24]) % td
+
+    # ----------------------------------------------------------------
+    # Timedelta.__divmod__, __rdivmod__
+
+    def test_divmod_numeric(self):
+        # GH#19365
+        td = Timedelta(days=2, hours=6)
+
+        result = divmod(td, 53 * 3600 * 1e9)
+        assert result[0] == Timedelta(1, unit="ns")
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(hours=1)
+
+        assert result
+        result = divmod(td, np.nan)
+        assert result[0] is NaT
+        assert result[1] is NaT
+
+    def test_divmod(self):
+        # GH#19365
+        td = Timedelta(days=2, hours=6)
+
+        result = divmod(td, timedelta(days=1))
+        assert result[0] == 2
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(hours=6)
+
+        result = divmod(td, 54)
+        assert result[0] == Timedelta(hours=1)
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(0)
+
+        result = divmod(td, NaT)
+        assert np.isnan(result[0])
+        assert result[1] is NaT
+
+    def test_divmod_offset(self):
+        # GH#19365
+        td = Timedelta(days=2, hours=6)
+
+        result = divmod(td, offsets.Hour(-4))
+        assert result[0] == -14
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(hours=-2)
+
+    def test_divmod_invalid(self):
+        # GH#19365
+        td = Timedelta(days=2, hours=6)
+
+        msg = r"unsupported operand type\(s\) for //: 'Timedelta' and 'Timestamp'"
+        with pytest.raises(TypeError, match=msg):
+            divmod(td, Timestamp("2018-01-22"))
+
+    def test_rdivmod_pytimedelta(self):
+        # GH#19365
+        result = divmod(timedelta(days=2, hours=6), Timedelta(days=1))
+        assert result[0] == 2
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(hours=6)
+
+    def test_rdivmod_offset(self):
+        result = divmod(offsets.Hour(54), Timedelta(hours=-4))
+        assert result[0] == -14
+        assert isinstance(result[1], Timedelta)
+        assert result[1] == Timedelta(hours=-2)
+
+    def test_rdivmod_invalid(self):
+        # GH#19365
+        td = Timedelta(minutes=3)
+        msg = "unsupported operand type"
+
+        with pytest.raises(TypeError, match=msg):
+            divmod(Timestamp("2018-01-22"), td)
+
+        with pytest.raises(TypeError, match=msg):
+            divmod(15, td)
+
+        with pytest.raises(TypeError, match=msg):
+            divmod(16.0, td)
+
+        msg = "Invalid dtype int"
+        with pytest.raises(TypeError, match=msg):
+            divmod(np.array([22, 24]), td)
+
+    # ----------------------------------------------------------------
+
+    @pytest.mark.parametrize(
+        "op", [operator.mul, ops.rmul, operator.truediv, ops.rdiv, ops.rsub]
+    )
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            [Timestamp("20130101 9:01"), Timestamp("20121230 9:02")],
+            [Timestamp("2021-11-09 09:54:00"), Timedelta("1D")],
+        ],
+    )
+    def test_td_op_timedelta_timedeltalike_array(self, op, arr):
+        arr = np.array(arr)
+        msg = "unsupported operand type|cannot use operands with types"
+        with pytest.raises(TypeError, match=msg):
+            op(arr, Timedelta("1D"))
+
+
+class TestTimedeltaComparison:
+    @pytest.mark.skip_ubsan
+    def test_compare_pytimedelta_bounds(self):
+        # GH#49021 don't overflow on comparison with very large pytimedeltas
+
+        for unit in ["ns", "us"]:
+            tdmax = Timedelta.max.as_unit(unit).max
+            tdmin = Timedelta.min.as_unit(unit).min
+
+            assert tdmax < timedelta.max
+            assert tdmax <= timedelta.max
+            assert not tdmax > timedelta.max
+            assert not tdmax >= timedelta.max
+            assert tdmax != timedelta.max
+            assert not tdmax == timedelta.max
+
+            assert tdmin > timedelta.min
+            assert tdmin >= timedelta.min
+            assert not tdmin < timedelta.min
+            assert not tdmin <= timedelta.min
+            assert tdmin != timedelta.min
+            assert not tdmin == timedelta.min
+
+        # But the "ms" and "s"-reso bounds extend pass pytimedelta
+        for unit in ["ms", "s"]:
+            tdmax = Timedelta.max.as_unit(unit).max
+            tdmin = Timedelta.min.as_unit(unit).min
+
+            assert tdmax > timedelta.max
+            assert tdmax >= timedelta.max
+            assert not tdmax < timedelta.max
+            assert not tdmax <= timedelta.max
+            assert tdmax != timedelta.max
+            assert not tdmax == timedelta.max
+
+            assert tdmin < timedelta.min
+            assert tdmin <= timedelta.min
+            assert not tdmin > timedelta.min
+            assert not tdmin >= timedelta.min
+            assert tdmin != timedelta.min
+            assert not tdmin == timedelta.min
+
+    def test_compare_pytimedelta_bounds2(self):
+        # a pytimedelta outside the microsecond bounds
+        pytd = timedelta(days=999999999, seconds=86399)
+        # NB: np.timedelta64(td, "s"") incorrectly overflows
+        td64 = np.timedelta64(pytd.days, "D") + np.timedelta64(pytd.seconds, "s")
+        td = Timedelta(td64)
+        assert td.days == pytd.days
+        assert td.seconds == pytd.seconds
+
+        assert td == pytd
+        assert not td != pytd
+        assert not td < pytd
+        assert not td > pytd
+        assert td <= pytd
+        assert td >= pytd
+
+        td2 = td - Timedelta(seconds=1).as_unit("s")
+        assert td2 != pytd
+        assert not td2 == pytd
+        assert td2 < pytd
+        assert td2 <= pytd
+        assert not td2 > pytd
+        assert not td2 >= pytd
+
+    def test_compare_tick(self, tick_classes):
+        cls = tick_classes
+
+        off = cls(4)
+        td = off._as_pd_timedelta
+        assert isinstance(td, Timedelta)
+
+        assert td == off
+        assert not td != off
+        assert td <= off
+        assert td >= off
+        assert not td < off
+        assert not td > off
+
+        assert not td == 2 * off
+        assert td != 2 * off
+        assert td <= 2 * off
+        assert td < 2 * off
+        assert not td >= 2 * off
+        assert not td > 2 * off
+
+    def test_comparison_object_array(self):
+        # analogous to GH#15183
+        td = Timedelta("2 days")
+        other = Timedelta("3 hours")
+
+        arr = np.array([other, td], dtype=object)
+        res = arr == td
+        expected = np.array([False, True], dtype=bool)
+        assert (res == expected).all()
+
+        # 2D case
+        arr = np.array([[other, td], [td, other]], dtype=object)
+        res = arr != td
+        expected = np.array([[True, False], [False, True]], dtype=bool)
+        assert res.shape == expected.shape
+        assert (res == expected).all()
+
+    def test_compare_timedelta_ndarray(self):
+        # GH#11835
+        periods = [Timedelta("0 days 01:00:00"), Timedelta("0 days 01:00:00")]
+        arr = np.array(periods)
+        result = arr[0] > arr
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_compare_td64_ndarray(self):
+        # GG#33441
+        arr = np.arange(5).astype("timedelta64[ns]")
+        td = Timedelta(arr[1])
+
+        expected = np.array([False, True, False, False, False], dtype=bool)
+
+        result = td == arr
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = arr == td
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = td != arr
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        result = arr != td
+        tm.assert_numpy_array_equal(result, ~expected)
+
+    def test_compare_custom_object(self):
+        """
+        Make sure non supported operations on Timedelta returns NonImplemented
+        and yields to other operand (GH#20829).
+        """
+
+        class CustomClass:
+            def __init__(self, cmp_result=None) -> None:
+                self.cmp_result = cmp_result
+
+            def generic_result(self):
+                if self.cmp_result is None:
+                    return NotImplemented
+                else:
+                    return self.cmp_result
+
+            def __eq__(self, other):
+                return self.generic_result()
+
+            def __gt__(self, other):
+                return self.generic_result()
+
+        t = Timedelta("1s")
+
+        assert t != "string"
+        assert t != 1
+        assert t != CustomClass()
+        assert t != CustomClass(cmp_result=False)
+
+        assert t < CustomClass(cmp_result=True)
+        assert not t < CustomClass(cmp_result=False)
+
+        assert t == CustomClass(cmp_result=True)
+
+    @pytest.mark.parametrize("val", ["string", 1])
+    def test_compare_unknown_type(self, val):
+        # GH#20829
+        t = Timedelta("1s")
+        msg = "not supported between instances of 'Timedelta' and '(int|str)'"
+        with pytest.raises(TypeError, match=msg):
+            t >= val
+        with pytest.raises(TypeError, match=msg):
+            t > val
+        with pytest.raises(TypeError, match=msg):
+            t <= val
+        with pytest.raises(TypeError, match=msg):
+            t < val
+
+
+def test_ops_notimplemented():
+    class Other:
+        pass
+
+    other = Other()
+
+    td = Timedelta("1 day")
+    assert td.__add__(other) is NotImplemented
+    assert td.__sub__(other) is NotImplemented
+    assert td.__truediv__(other) is NotImplemented
+    assert td.__mul__(other) is NotImplemented
+    assert td.__floordiv__(other) is NotImplemented
+
+
+def test_ops_error_str():
+    # GH#13624
+    td = Timedelta("1 day")
+
+    for left, right in [(td, "a"), ("a", td)]:
+        msg = "|".join(
+            [
+                "unsupported operand type",
+                r'can only concatenate str \(not "Timedelta"\) to str',
+                "must be str, not Timedelta",
+            ]
+        )
+        with pytest.raises(TypeError, match=msg):
+            left + right
+
+        msg = "not supported between instances of"
+        with pytest.raises(TypeError, match=msg):
+            left > right
+
+        assert not left == right
+        assert left != right
diff --git a/pandas/pandas/tests/scalar/timedelta/test_constructors.py b/pandas/pandas/tests/scalar/timedelta/test_constructors.py
new file mode 100644
index 0000000000000..45caeb1733590
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/test_constructors.py
@@ -0,0 +1,693 @@
+from datetime import timedelta
+from itertools import product
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import OutOfBoundsTimedelta
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+
+from pandas import (
+    Index,
+    NaT,
+    Timedelta,
+    TimedeltaIndex,
+    offsets,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+
+class TestTimedeltaConstructorUnitKeyword:
+    @pytest.mark.parametrize("unit", ["Y", "y", "M"])
+    def test_unit_m_y_raises(self, unit):
+        msg = "Units 'M', 'Y', and 'y' are no longer supported"
+
+        with pytest.raises(ValueError, match=msg):
+            Timedelta(10, unit)
+
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(10, unit)
+
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta([1, 2], unit)
+
+    @pytest.mark.parametrize(
+        "unit,unit_depr",
+        [
+            ("W", "w"),
+            ("D", "d"),
+            ("min", "MIN"),
+            ("s", "S"),
+            ("h", "H"),
+            ("ms", "MS"),
+            ("us", "US"),
+        ],
+    )
+    def test_unit_deprecated(self, unit, unit_depr):
+        # GH#52536
+        msg = f"'{unit_depr}' is deprecated and will be removed in a future version."
+
+        expected = Timedelta(1, unit=unit)
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = Timedelta(1, unit=unit_depr)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "unit, np_unit",
+        [("W", "W")]
+        + [(value, "D") for value in ["D", "days", "day", "Days", "Day"]]
+        + [
+            (value, "m")
+            for value in [
+                "m",
+                "minute",
+                "min",
+                "minutes",
+                "Minute",
+                "Min",
+                "Minutes",
+            ]
+        ]
+        + [
+            (value, "s")
+            for value in [
+                "s",
+                "seconds",
+                "sec",
+                "second",
+                "Seconds",
+                "Sec",
+                "Second",
+            ]
+        ]
+        + [
+            (value, "ms")
+            for value in [
+                "ms",
+                "milliseconds",
+                "millisecond",
+                "milli",
+                "millis",
+                "Milliseconds",
+                "Millisecond",
+                "Milli",
+                "Millis",
+            ]
+        ]
+        + [
+            (value, "us")
+            for value in [
+                "us",
+                "microseconds",
+                "microsecond",
+                "micro",
+                "micros",
+                "Microseconds",
+                "Microsecond",
+                "Micro",
+                "Micros",
+            ]
+        ]
+        + [
+            (value, "ns")
+            for value in [
+                "ns",
+                "nanoseconds",
+                "nanosecond",
+                "nano",
+                "nanos",
+                "Nanoseconds",
+                "Nanosecond",
+                "Nano",
+                "Nanos",
+            ]
+        ],
+    )
+    @pytest.mark.parametrize("wrapper", [np.array, list, Index])
+    def test_unit_parser(self, unit, np_unit, wrapper):
+        # validate all units, GH 6855, GH 21762
+        # array-likes
+        expected = TimedeltaIndex(
+            [np.timedelta64(i, np_unit) for i in np.arange(5).tolist()],
+            dtype="m8[ns]",
+        )
+        # TODO(2.0): the desired output dtype may have non-nano resolution
+
+        result = to_timedelta(wrapper(range(5)), unit=unit)
+        tm.assert_index_equal(result, expected)
+
+        str_repr = [f"{x}{unit}" for x in np.arange(5)]
+        result = to_timedelta(wrapper(str_repr))
+        tm.assert_index_equal(result, expected)
+        result = to_timedelta(wrapper(str_repr))
+        tm.assert_index_equal(result, expected)
+
+        # scalar
+        expected = Timedelta(np.timedelta64(2, np_unit).astype("timedelta64[ns]"))
+        result = to_timedelta(2, unit=unit)
+        assert result == expected
+        result = Timedelta(2, unit=unit)
+        assert result == expected
+
+        result = to_timedelta(f"2{unit}")
+        assert result == expected
+        result = Timedelta(f"2{unit}")
+        assert result == expected
+
+    @pytest.mark.parametrize("unit", ["T", "t", "L", "l", "U", "u", "N", "n"])
+    def test_unit_T_L_N_U_raises(self, unit):
+        msg = f"invalid unit abbreviation: {unit}"
+        with pytest.raises(ValueError, match=msg):
+            Timedelta(1, unit=unit)
+
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(10, unit)
+
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta([1, 2], unit)
+
+
+def test_construct_from_kwargs_overflow():
+    # GH#55503
+    msg = "seconds=86400000000000000000, milliseconds=0, microseconds=0, nanoseconds=0"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        Timedelta(days=10**6)
+    msg = "seconds=60000000000000000000, milliseconds=0, microseconds=0, nanoseconds=0"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        Timedelta(minutes=10**9)
+
+
+def test_construct_with_weeks_unit_overflow():
+    # GH#47268 don't silently wrap around
+    with pytest.raises(OutOfBoundsTimedelta, match="without overflow"):
+        Timedelta(1000000000000000000, unit="W")
+
+    with pytest.raises(OutOfBoundsTimedelta, match="without overflow"):
+        Timedelta(1000000000000000000.0, unit="W")
+
+
+def test_construct_from_td64_with_unit():
+    # ignore the unit, as it may cause silently overflows leading to incorrect
+    #  results, and in non-overflow cases is irrelevant GH#46827
+    obj = np.timedelta64(123456789000000000, "h")
+
+    with pytest.raises(OutOfBoundsTimedelta, match="123456789000000000 hours"):
+        Timedelta(obj, unit="ps")
+
+    with pytest.raises(OutOfBoundsTimedelta, match="123456789000000000 hours"):
+        Timedelta(obj, unit="ns")
+
+    with pytest.raises(OutOfBoundsTimedelta, match="123456789000000000 hours"):
+        Timedelta(obj)
+
+
+def test_from_td64_retain_resolution():
+    # case where we retain millisecond resolution
+    obj = np.timedelta64(12345, "ms")
+
+    td = Timedelta(obj)
+    assert td._value == obj.view("i8")
+    assert td._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+    # Case where we cast to nearest-supported reso
+    obj2 = np.timedelta64(1234, "D")
+    td2 = Timedelta(obj2)
+    assert td2._creso == NpyDatetimeUnit.NPY_FR_s.value
+    assert td2 == obj2
+    assert td2.days == 1234
+
+    # Case that _would_ overflow if we didn't support non-nano
+    obj3 = np.timedelta64(1000000000000000000, "us")
+    td3 = Timedelta(obj3)
+    assert td3.total_seconds() == 1000000000000
+    assert td3._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+
+def test_from_pytimedelta_us_reso():
+    # pytimedelta has microsecond resolution, so Timedelta(pytd) inherits that
+    td = timedelta(days=4, minutes=3)
+    result = Timedelta(td)
+    assert result.to_pytimedelta() == td
+    assert result._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+
+def test_from_tick_reso():
+    tick = offsets.Nano()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    tick = offsets.Micro()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+    tick = offsets.Milli()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+    tick = offsets.Second()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+    # everything above Second gets cast to the closest supported reso: second
+    tick = offsets.Minute()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+    tick = offsets.Hour()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+    tick = offsets.Day()
+    assert Timedelta(tick)._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+
+def test_construction():
+    expected = np.timedelta64(10, "D").astype("m8[ns]").view("i8")
+    assert Timedelta(10, unit="D")._value == expected
+    assert Timedelta(10.0, unit="D")._value == expected
+    assert Timedelta("10 days")._value == expected
+    assert Timedelta(days=10)._value == expected
+    assert Timedelta(days=10.0)._value == expected
+
+    expected += np.timedelta64(10, "s").astype("m8[ns]").view("i8")
+    assert Timedelta("10 days 00:00:10")._value == expected
+    assert Timedelta(days=10, seconds=10)._value == expected
+    assert Timedelta(days=10, milliseconds=10 * 1000)._value == expected
+    assert Timedelta(days=10, microseconds=10 * 1000 * 1000)._value == expected
+
+    # rounding cases
+    assert Timedelta(82739999850000)._value == 82739999850000
+    assert "0 days 22:58:59.999850" in str(Timedelta(82739999850000))
+    assert Timedelta(123072001000000)._value == 123072001000000
+    assert "1 days 10:11:12.001" in str(Timedelta(123072001000000))
+
+    # string conversion with/without leading zero
+    # GH#9570
+    assert Timedelta("0:00:00") == timedelta(hours=0)
+    assert Timedelta("00:00:00") == timedelta(hours=0)
+    assert Timedelta("-1:00:00") == -timedelta(hours=1)
+    assert Timedelta("-01:00:00") == -timedelta(hours=1)
+
+    # more strings & abbrevs
+    # GH#8190
+    assert Timedelta("1 h") == timedelta(hours=1)
+    assert Timedelta("1 hour") == timedelta(hours=1)
+    assert Timedelta("1 hr") == timedelta(hours=1)
+    assert Timedelta("1 hours") == timedelta(hours=1)
+    assert Timedelta("-1 hours") == -timedelta(hours=1)
+    assert Timedelta("1 m") == timedelta(minutes=1)
+    assert Timedelta("1.5 m") == timedelta(seconds=90)
+    assert Timedelta("1 minute") == timedelta(minutes=1)
+    assert Timedelta("1 minutes") == timedelta(minutes=1)
+    assert Timedelta("1 s") == timedelta(seconds=1)
+    assert Timedelta("1 second") == timedelta(seconds=1)
+    assert Timedelta("1 seconds") == timedelta(seconds=1)
+    assert Timedelta("1 ms") == timedelta(milliseconds=1)
+    assert Timedelta("1 milli") == timedelta(milliseconds=1)
+    assert Timedelta("1 millisecond") == timedelta(milliseconds=1)
+    assert Timedelta("1 us") == timedelta(microseconds=1)
+    assert Timedelta("1 µs") == timedelta(microseconds=1)
+    assert Timedelta("1 micros") == timedelta(microseconds=1)
+    assert Timedelta("1 microsecond") == timedelta(microseconds=1)
+    assert Timedelta("1.5 microsecond") == Timedelta("00:00:00.000001500")
+    assert Timedelta("1 ns") == Timedelta("00:00:00.000000001")
+    assert Timedelta("1 nano") == Timedelta("00:00:00.000000001")
+    assert Timedelta("1 nanosecond") == Timedelta("00:00:00.000000001")
+
+    # combos
+    assert Timedelta("10 days 1 hour") == timedelta(days=10, hours=1)
+    assert Timedelta("10 days 1 h") == timedelta(days=10, hours=1)
+    assert Timedelta("10 days 1 h 1m 1s") == timedelta(
+        days=10, hours=1, minutes=1, seconds=1
+    )
+    assert Timedelta("-10 days 1 h 1m 1s") == -timedelta(
+        days=10, hours=1, minutes=1, seconds=1
+    )
+    assert Timedelta("-10 days 1 h 1m 1s") == -timedelta(
+        days=10, hours=1, minutes=1, seconds=1
+    )
+    assert Timedelta("-10 days 1 h 1m 1s 3us") == -timedelta(
+        days=10, hours=1, minutes=1, seconds=1, microseconds=3
+    )
+    assert Timedelta("-10 days 1 h 1.5m 1s 3us") == -timedelta(
+        days=10, hours=1, minutes=1, seconds=31, microseconds=3
+    )
+
+    # Currently invalid as it has a - on the hh:mm:dd part
+    # (only allowed on the days)
+    msg = "only leading negative signs are allowed"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta("-10 days -1 h 1.5m 1s 3us")
+
+    # only leading neg signs are allowed
+    with pytest.raises(ValueError, match=msg):
+        Timedelta("10 days -1 h 1.5m 1s 3us")
+
+    # no units specified
+    msg = "no units specified"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta("3.1415")
+
+    # invalid construction
+    msg = "cannot construct a Timedelta"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta()
+
+    msg = "unit abbreviation w/o a number"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta("foo")
+
+    msg = (
+        "cannot construct a Timedelta from the passed arguments, allowed keywords are "
+    )
+    with pytest.raises(ValueError, match=msg):
+        Timedelta(day=10)
+
+    # floats
+    expected = np.timedelta64(10, "s").astype("m8[ns]").view("i8") + np.timedelta64(
+        500, "ms"
+    ).astype("m8[ns]").view("i8")
+    assert Timedelta(10.5, unit="s")._value == expected
+
+    # offset
+    assert to_timedelta(offsets.Hour(2)) == Timedelta(hours=2)
+    assert Timedelta(offsets.Hour(2)) == Timedelta(hours=2)
+    assert Timedelta(offsets.Second(2)) == Timedelta(seconds=2)
+
+    # GH#11995: unicode
+    expected = Timedelta("1h")
+    result = Timedelta("1h")
+    assert result == expected
+    assert to_timedelta(offsets.Hour(2)) == Timedelta("0 days, 02:00:00")
+
+    msg = "unit abbreviation w/o a number"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta("foo bar")
+
+
+@pytest.mark.parametrize(
+    "item",
+    list(
+        {
+            "days": "D",
+            "seconds": "s",
+            "microseconds": "us",
+            "milliseconds": "ms",
+            "minutes": "m",
+            "hours": "h",
+            "weeks": "W",
+        }.items()
+    ),
+)
+@pytest.mark.parametrize(
+    "npdtype", [np.int64, np.int32, np.int16, np.float64, np.float32, np.float16]
+)
+def test_td_construction_with_np_dtypes(npdtype, item):
+    # GH#8757: test construction with np dtypes
+    pykwarg, npkwarg = item
+    expected = np.timedelta64(1, npkwarg).astype("m8[ns]").view("i8")
+    assert Timedelta(**{pykwarg: npdtype(1)})._value == expected
+
+
+@pytest.mark.parametrize(
+    "val",
+    [
+        "1s",
+        "-1s",
+        "1us",
+        "-1us",
+        "1 day",
+        "-1 day",
+        "-23:59:59.999999",
+        "-1 days +23:59:59.999999",
+        "-1ns",
+        "1ns",
+        "-23:59:59.999999999",
+    ],
+)
+def test_td_from_repr_roundtrip(val):
+    # round-trip both for string and value
+    td = Timedelta(val)
+    assert Timedelta(td._value) == td
+
+    assert Timedelta(str(td)) == td
+    assert Timedelta(td._repr_base(format="all")) == td
+    assert Timedelta(td._repr_base()) == td
+
+
+def test_overflow_on_construction():
+    # GH#3374
+    value = Timedelta("1day")._value * 20169940
+    msg = "Cannot cast 1742682816000000000000 from ns to 'ns' without overflow"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        Timedelta(value)
+
+    # xref GH#17637
+    msg = "Cannot cast 139993 from D to 'ns' without overflow"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        Timedelta(7 * 19999, unit="D")
+
+    # used to overflow before non-ns support
+    td = Timedelta(timedelta(days=13 * 19999))
+    assert td._creso == NpyDatetimeUnit.NPY_FR_us.value
+    assert td.days == 13 * 19999
+
+
+@pytest.mark.parametrize(
+    "val, unit",
+    [
+        (15251, "W"),  # 1
+        (106752, "D"),  # change from previous:
+        (2562048, "h"),  # 0 hours
+        (153722868, "m"),  # 13 minutes
+        (9223372037, "s"),  # 44 seconds
+    ],
+)
+def test_construction_out_of_bounds_td64ns(val, unit):
+    # TODO: parametrize over units just above/below the implementation bounds
+    #  once GH#38964 is resolved
+
+    # Timedelta.max is just under 106752 days
+    td64 = np.timedelta64(val, unit)
+    assert td64.astype("m8[ns]").view("i8") < 0  # i.e. naive astype will be wrong
+
+    td = Timedelta(td64)
+    if unit != "M":
+        # with unit="M" the conversion to "s" is poorly defined
+        #  (and numpy issues DeprecationWarning)
+        assert td.asm8 == td64
+    assert td.asm8.dtype == "m8[s]"
+    msg = r"Cannot cast 1067\d\d days .* to unit='ns' without overflow"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        td.as_unit("ns")
+
+    # But just back in bounds and we are OK
+    assert Timedelta(td64 - 1) == td64 - 1
+
+    td64 *= -1
+    assert td64.astype("m8[ns]").view("i8") > 0  # i.e. naive astype will be wrong
+
+    td2 = Timedelta(td64)
+    msg = r"Cannot cast -1067\d\d days .* to unit='ns' without overflow"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        td2.as_unit("ns")
+
+    # But just back in bounds and we are OK
+    assert Timedelta(td64 + 1) == td64 + 1
+
+
+@pytest.mark.parametrize(
+    "val, unit",
+    [
+        (15251 * 10**9, "W"),
+        (106752 * 10**9, "D"),
+        (2562048 * 10**9, "h"),
+        (153722868 * 10**9, "m"),
+    ],
+)
+def test_construction_out_of_bounds_td64s(val, unit):
+    td64 = np.timedelta64(val, unit)
+    with pytest.raises(OutOfBoundsTimedelta, match=str(td64)):
+        Timedelta(td64)
+
+    # But just back in bounds and we are OK
+    assert Timedelta(td64 - 10**9) == td64 - 10**9
+
+
+@pytest.mark.parametrize(
+    "fmt,exp",
+    [
+        (
+            "P6DT0H50M3.010010012S",
+            Timedelta(
+                days=6,
+                minutes=50,
+                seconds=3,
+                milliseconds=10,
+                microseconds=10,
+                nanoseconds=12,
+            ),
+        ),
+        (
+            "P-6DT0H50M3.010010012S",
+            Timedelta(
+                days=-6,
+                minutes=50,
+                seconds=3,
+                milliseconds=10,
+                microseconds=10,
+                nanoseconds=12,
+            ),
+        ),
+        ("P4DT12H30M5S", Timedelta(days=4, hours=12, minutes=30, seconds=5)),
+        ("P0DT0H0M0.000000123S", Timedelta(nanoseconds=123)),
+        ("P0DT0H0M0.00001S", Timedelta(microseconds=10)),
+        ("P0DT0H0M0.001S", Timedelta(milliseconds=1)),
+        ("P0DT0H1M0S", Timedelta(minutes=1)),
+        ("P1DT25H61M61S", Timedelta(days=1, hours=25, minutes=61, seconds=61)),
+        ("PT1S", Timedelta(seconds=1)),
+        ("PT0S", Timedelta(seconds=0)),
+        ("P1WT0S", Timedelta(days=7, seconds=0)),
+        ("P1D", Timedelta(days=1)),
+        ("P1DT1H", Timedelta(days=1, hours=1)),
+        ("P1W", Timedelta(days=7)),
+        ("PT300S", Timedelta(seconds=300)),
+        ("P1DT0H0M00000000000S", Timedelta(days=1)),
+        ("PT-6H3M", Timedelta(hours=-6, minutes=3)),
+        ("-PT6H3M", Timedelta(hours=-6, minutes=-3)),
+        ("-PT-6H+3M", Timedelta(hours=6, minutes=-3)),
+    ],
+)
+def test_iso_constructor(fmt, exp):
+    assert Timedelta(fmt) == exp
+
+
+@pytest.mark.parametrize(
+    "fmt",
+    [
+        "PPPPPPPPPPPP",
+        "PDTHMS",
+        "P0DT999H999M999S",
+        "P1DT0H0M0.0000000000000S",
+        "P1DT0H0M0.S",
+        "P",
+        "-P",
+    ],
+)
+def test_iso_constructor_raises(fmt):
+    msg = f"Invalid ISO 8601 Duration format - {fmt}"
+    with pytest.raises(ValueError, match=msg):
+        Timedelta(fmt)
+
+
+@pytest.mark.parametrize(
+    "constructed_td, conversion",
+    [
+        (Timedelta(nanoseconds=100), "100ns"),
+        (
+            Timedelta(
+                days=1,
+                hours=1,
+                minutes=1,
+                weeks=1,
+                seconds=1,
+                milliseconds=1,
+                microseconds=1,
+                nanoseconds=1,
+            ),
+            694861001001001,
+        ),
+        (Timedelta(microseconds=1) + Timedelta(nanoseconds=1), "1us1ns"),
+        (Timedelta(microseconds=1) - Timedelta(nanoseconds=1), "999ns"),
+        (Timedelta(microseconds=1) + 5 * Timedelta(nanoseconds=-2), "990ns"),
+    ],
+)
+def test_td_constructor_on_nanoseconds(constructed_td, conversion):
+    # GH#9273
+    assert constructed_td == Timedelta(conversion)
+
+
+def test_td_constructor_value_error():
+    msg = "Invalid type <class 'str'>. Must be int or float."
+    with pytest.raises(TypeError, match=msg):
+        Timedelta(nanoseconds="abc")
+
+
+def test_timedelta_constructor_identity():
+    # Test for #30543
+    expected = Timedelta(np.timedelta64(1, "s"))
+    result = Timedelta(expected)
+    assert result is expected
+
+
+def test_timedelta_pass_td_and_kwargs_raises():
+    # don't silently ignore the kwargs GH#48898
+    td = Timedelta(days=1)
+    msg = (
+        "Cannot pass both a Timedelta input and timedelta keyword arguments, "
+        r"got \['days'\]"
+    )
+    with pytest.raises(ValueError, match=msg):
+        Timedelta(td, days=2)
+
+
+@pytest.mark.parametrize(
+    "constructor, value, unit",
+    [
+        (Timedelta, "10s", "ms"),
+        (to_timedelta, "10s", "ms"),
+        (to_timedelta, ["1", 2, 3], "s"),
+    ],
+)
+def test_string_with_unit(constructor, value, unit):
+    with pytest.raises(ValueError, match="unit must not be specified"):
+        constructor(value, unit=unit)
+
+
+@pytest.mark.parametrize(
+    "value",
+    [
+        "".join(elements)
+        for repetition in (1, 2)
+        for elements in product("+-, ", repeat=repetition)
+    ],
+)
+def test_string_without_numbers(value):
+    # GH39710 Timedelta input string with only symbols and no digits raises an error
+    msg = (
+        "symbols w/o a number"
+        if value != "--"
+        else "only leading negative signs are allowed"
+    )
+    with pytest.raises(ValueError, match=msg):
+        Timedelta(value)
+
+
+def test_timedelta_new_npnat():
+    # GH#48898
+    nat = np.timedelta64("NaT", "h")
+    assert Timedelta(nat) is NaT
+
+
+def test_subclass_respected():
+    # GH#49579
+    class MyCustomTimedelta(Timedelta):
+        pass
+
+    td = MyCustomTimedelta("1 minute")
+    assert isinstance(td, MyCustomTimedelta)
+
+
+def test_non_nano_value():
+    # https://github.com/pandas-dev/pandas/issues/49076
+    result = Timedelta(10, unit="D").as_unit("s").value
+    # `.value` shows nanoseconds, even though unit is 's'
+    assert result == 864000000000000
+
+    # out-of-nanoseconds-bounds `.value` raises informative message
+    msg = (
+        r"Cannot convert Timedelta to nanoseconds without overflow. "
+        r"Use `.asm8.view\('i8'\)` to cast represent Timedelta in its "
+        r"own unit \(here, s\).$"
+    )
+    td = Timedelta(1_000, "D").as_unit("s") * 1_000
+    with pytest.raises(OverflowError, match=msg):
+        td.value
+    # check that the suggested workaround actually works
+    result = td.asm8.view("i8")
+    assert result == 86400000000
diff --git a/pandas/pandas/tests/scalar/timedelta/test_formats.py b/pandas/pandas/tests/scalar/timedelta/test_formats.py
new file mode 100644
index 0000000000000..1aafeec2ceed5
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/test_formats.py
@@ -0,0 +1,109 @@
+import pytest
+
+from pandas import Timedelta
+
+
+@pytest.mark.parametrize(
+    "td, expected_repr",
+    [
+        (Timedelta(10, unit="D"), "Timedelta('10 days 00:00:00')"),
+        (Timedelta(10, unit="s"), "Timedelta('0 days 00:00:10')"),
+        (Timedelta(10, unit="ms"), "Timedelta('0 days 00:00:00.010000')"),
+        (Timedelta(-10, unit="ms"), "Timedelta('-1 days +23:59:59.990000')"),
+    ],
+)
+def test_repr(td, expected_repr):
+    assert repr(td) == expected_repr
+
+
+@pytest.mark.parametrize(
+    "td, expected_iso",
+    [
+        (
+            Timedelta(
+                days=6,
+                minutes=50,
+                seconds=3,
+                milliseconds=10,
+                microseconds=10,
+                nanoseconds=12,
+            ),
+            "P6DT0H50M3.010010012S",
+        ),
+        (Timedelta(days=4, hours=12, minutes=30, seconds=5), "P4DT12H30M5S"),
+        (Timedelta(nanoseconds=123), "P0DT0H0M0.000000123S"),
+        # trim nano
+        (Timedelta(microseconds=10), "P0DT0H0M0.00001S"),
+        # trim micro
+        (Timedelta(milliseconds=1), "P0DT0H0M0.001S"),
+        # don't strip every 0
+        (Timedelta(minutes=1), "P0DT0H1M0S"),
+    ],
+)
+def test_isoformat(td, expected_iso):
+    assert td.isoformat() == expected_iso
+
+
+class TestReprBase:
+    def test_none(self):
+        delta_1d = Timedelta(1, unit="D")
+        delta_0d = Timedelta(0, unit="D")
+        delta_1s = Timedelta(1, unit="s")
+        delta_500ms = Timedelta(500, unit="ms")
+
+        drepr = lambda x: x._repr_base()
+        assert drepr(delta_1d) == "1 days"
+        assert drepr(-delta_1d) == "-1 days"
+        assert drepr(delta_0d) == "0 days"
+        assert drepr(delta_1s) == "0 days 00:00:01"
+        assert drepr(delta_500ms) == "0 days 00:00:00.500000"
+        assert drepr(delta_1d + delta_1s) == "1 days 00:00:01"
+        assert drepr(-delta_1d + delta_1s) == "-1 days +00:00:01"
+        assert drepr(delta_1d + delta_500ms) == "1 days 00:00:00.500000"
+        assert drepr(-delta_1d + delta_500ms) == "-1 days +00:00:00.500000"
+
+    def test_sub_day(self):
+        delta_1d = Timedelta(1, unit="D")
+        delta_0d = Timedelta(0, unit="D")
+        delta_1s = Timedelta(1, unit="s")
+        delta_500ms = Timedelta(500, unit="ms")
+
+        drepr = lambda x: x._repr_base(format="sub_day")
+        assert drepr(delta_1d) == "1 days"
+        assert drepr(-delta_1d) == "-1 days"
+        assert drepr(delta_0d) == "00:00:00"
+        assert drepr(delta_1s) == "00:00:01"
+        assert drepr(delta_500ms) == "00:00:00.500000"
+        assert drepr(delta_1d + delta_1s) == "1 days 00:00:01"
+        assert drepr(-delta_1d + delta_1s) == "-1 days +00:00:01"
+        assert drepr(delta_1d + delta_500ms) == "1 days 00:00:00.500000"
+        assert drepr(-delta_1d + delta_500ms) == "-1 days +00:00:00.500000"
+
+    def test_long(self):
+        delta_1d = Timedelta(1, unit="D")
+        delta_0d = Timedelta(0, unit="D")
+        delta_1s = Timedelta(1, unit="s")
+        delta_500ms = Timedelta(500, unit="ms")
+
+        drepr = lambda x: x._repr_base(format="long")
+        assert drepr(delta_1d) == "1 days 00:00:00"
+        assert drepr(-delta_1d) == "-1 days +00:00:00"
+        assert drepr(delta_0d) == "0 days 00:00:00"
+        assert drepr(delta_1s) == "0 days 00:00:01"
+        assert drepr(delta_500ms) == "0 days 00:00:00.500000"
+        assert drepr(delta_1d + delta_1s) == "1 days 00:00:01"
+        assert drepr(-delta_1d + delta_1s) == "-1 days +00:00:01"
+        assert drepr(delta_1d + delta_500ms) == "1 days 00:00:00.500000"
+        assert drepr(-delta_1d + delta_500ms) == "-1 days +00:00:00.500000"
+
+    def test_all(self):
+        delta_1d = Timedelta(1, unit="D")
+        delta_0d = Timedelta(0, unit="D")
+        delta_1ns = Timedelta(1, unit="ns")
+
+        drepr = lambda x: x._repr_base(format="all")
+        assert drepr(delta_1d) == "1 days 00:00:00.000000000"
+        assert drepr(-delta_1d) == "-1 days +00:00:00.000000000"
+        assert drepr(delta_0d) == "0 days 00:00:00.000000000"
+        assert drepr(delta_1ns) == "0 days 00:00:00.000000001"
+        assert drepr(-delta_1d + delta_1ns) == "-1 days +00:00:00.000000001"
diff --git a/pandas/pandas/tests/scalar/timedelta/test_timedelta.py b/pandas/pandas/tests/scalar/timedelta/test_timedelta.py
new file mode 100644
index 0000000000000..8be2ec846a6d9
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timedelta/test_timedelta.py
@@ -0,0 +1,726 @@
+"""test the scalar Timedelta"""
+
+from datetime import timedelta
+import sys
+
+from hypothesis import (
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._libs.missing import NA
+from pandas._libs.tslibs import (
+    NaT,
+    iNaT,
+)
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.errors import OutOfBoundsTimedelta
+
+from pandas import (
+    Timedelta,
+    to_timedelta,
+)
+import pandas._testing as tm
+
+
+class TestNonNano:
+    @pytest.fixture(params=["s", "ms", "us"])
+    def unit_str(self, request):
+        return request.param
+
+    @pytest.fixture
+    def unit(self, unit_str):
+        # 7, 8, 9 correspond to second, millisecond, and microsecond, respectively
+        attr = f"NPY_FR_{unit_str}"
+        return getattr(NpyDatetimeUnit, attr).value
+
+    @pytest.fixture
+    def val(self, unit):
+        # microsecond that would be just out of bounds for nano
+        us = 9223372800000000
+        if unit == NpyDatetimeUnit.NPY_FR_us.value:
+            value = us
+        elif unit == NpyDatetimeUnit.NPY_FR_ms.value:
+            value = us // 1000
+        else:
+            value = us // 1_000_000
+        return value
+
+    @pytest.fixture
+    def td(self, unit, val):
+        return Timedelta._from_value_and_reso(val, unit)
+
+    def test_from_value_and_reso(self, unit, val):
+        # Just checking that the fixture is giving us what we asked for
+        td = Timedelta._from_value_and_reso(val, unit)
+        assert td._value == val
+        assert td._creso == unit
+        assert td.days == 106752
+
+    def test_unary_non_nano(self, td, unit):
+        assert abs(td)._creso == unit
+        assert (-td)._creso == unit
+        assert (+td)._creso == unit
+
+    def test_sub_preserves_reso(self, td, unit):
+        res = td - td
+        expected = Timedelta._from_value_and_reso(0, unit)
+        assert res == expected
+        assert res._creso == unit
+
+    def test_mul_preserves_reso(self, td, unit):
+        # The td fixture should always be far from the implementation
+        #  bound, so doubling does not risk overflow.
+        res = td * 2
+        assert res._value == td._value * 2
+        assert res._creso == unit
+
+    def test_cmp_cross_reso(self, td):
+        # numpy gets this wrong because of silent overflow
+        other = Timedelta(days=106751, unit="ns")
+        assert other < td
+        assert td > other
+        assert not other == td
+        assert td != other
+
+    def test_to_pytimedelta(self, td):
+        res = td.to_pytimedelta()
+        expected = timedelta(days=106752)
+        assert type(res) is timedelta
+        assert res == expected
+
+    def test_to_timedelta64(self, td, unit):
+        for res in [td.to_timedelta64(), td.to_numpy(), td.asm8]:
+            assert isinstance(res, np.timedelta64)
+            assert res.view("i8") == td._value
+            if unit == NpyDatetimeUnit.NPY_FR_s.value:
+                assert res.dtype == "m8[s]"
+            elif unit == NpyDatetimeUnit.NPY_FR_ms.value:
+                assert res.dtype == "m8[ms]"
+            elif unit == NpyDatetimeUnit.NPY_FR_us.value:
+                assert res.dtype == "m8[us]"
+
+    def test_truediv_timedeltalike(self, td):
+        assert td / td == 1
+        assert (2.5 * td) / td == 2.5
+
+        other = Timedelta(td._value)
+        msg = "Cannot cast 106752 days 00:00:00 to unit='ns' without overflow."
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            td / other
+
+        # Timedelta(other.to_pytimedelta()) has microsecond resolution,
+        #  so the division doesn't require casting all the way to nanos,
+        #  so succeeds
+        res = other.to_pytimedelta() / td
+        expected = other.to_pytimedelta() / td.to_pytimedelta()
+        assert res == expected
+
+        # if there's no overflow, we cast to the higher reso
+        left = Timedelta._from_value_and_reso(50, NpyDatetimeUnit.NPY_FR_us.value)
+        right = Timedelta._from_value_and_reso(50, NpyDatetimeUnit.NPY_FR_ms.value)
+        result = left / right
+        assert result == 0.001
+
+        result = right / left
+        assert result == 1000
+
+    def test_truediv_numeric(self, td):
+        assert td / np.nan is NaT
+
+        res = td / 2
+        assert res._value == td._value / 2
+        assert res._creso == td._creso
+
+        res = td / 2.0
+        assert res._value == td._value / 2
+        assert res._creso == td._creso
+
+    def test_truediv_na_type_not_supported(self, td):
+        msg_td_floordiv_na = (
+            r"unsupported operand type\(s\) for /: 'Timedelta' and 'NAType'"
+        )
+        with pytest.raises(TypeError, match=msg_td_floordiv_na):
+            td / NA
+
+        msg_na_floordiv_td = (
+            r"unsupported operand type\(s\) for /: 'NAType' and 'Timedelta'"
+        )
+        with pytest.raises(TypeError, match=msg_na_floordiv_td):
+            NA / td
+
+    def test_floordiv_timedeltalike(self, td):
+        assert td // td == 1
+        assert (2.5 * td) // td == 2
+
+        other = Timedelta(td._value)
+        msg = "Cannot cast 106752 days 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            td // other
+
+        # Timedelta(other.to_pytimedelta()) has microsecond resolution,
+        #  so the floordiv doesn't require casting all the way to nanos,
+        #  so succeeds
+        res = other.to_pytimedelta() // td
+        assert res == 0
+
+        # if there's no overflow, we cast to the higher reso
+        left = Timedelta._from_value_and_reso(50050, NpyDatetimeUnit.NPY_FR_us.value)
+        right = Timedelta._from_value_and_reso(50, NpyDatetimeUnit.NPY_FR_ms.value)
+        result = left // right
+        assert result == 1
+        result = right // left
+        assert result == 0
+
+    def test_floordiv_numeric(self, td):
+        assert td // np.nan is NaT
+
+        res = td // 2
+        assert res._value == td._value // 2
+        assert res._creso == td._creso
+
+        res = td // 2.0
+        assert res._value == td._value // 2
+        assert res._creso == td._creso
+
+        assert td // np.array(np.nan) is NaT
+
+        res = td // np.array(2)
+        assert res._value == td._value // 2
+        assert res._creso == td._creso
+
+        res = td // np.array(2.0)
+        assert res._value == td._value // 2
+        assert res._creso == td._creso
+
+    def test_floordiv_na_type_not_supported(self, td):
+        msg_td_floordiv_na = (
+            r"unsupported operand type\(s\) for //: 'Timedelta' and 'NAType'"
+        )
+        with pytest.raises(TypeError, match=msg_td_floordiv_na):
+            td // NA
+
+        msg_na_floordiv_td = (
+            r"unsupported operand type\(s\) for //: 'NAType' and 'Timedelta'"
+        )
+        with pytest.raises(TypeError, match=msg_na_floordiv_td):
+            NA // td
+
+    def test_addsub_mismatched_reso(self, td):
+        # need to cast to since td is out of bounds for ns, so
+        #  so we would raise OverflowError without casting
+        other = Timedelta(days=1).as_unit("us")
+
+        # td is out of bounds for ns
+        result = td + other
+        assert result._creso == other._creso
+        assert result.days == td.days + 1
+
+        result = other + td
+        assert result._creso == other._creso
+        assert result.days == td.days + 1
+
+        result = td - other
+        assert result._creso == other._creso
+        assert result.days == td.days - 1
+
+        result = other - td
+        assert result._creso == other._creso
+        assert result.days == 1 - td.days
+
+        other2 = Timedelta(500)
+        msg = "Cannot cast 106752 days 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            td + other2
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            other2 + td
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            td - other2
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            other2 - td
+
+    def test_min(self, td):
+        assert td.min <= td
+        assert td.min._creso == td._creso
+        assert td.min._value == NaT._value + 1
+
+    def test_max(self, td):
+        assert td.max >= td
+        assert td.max._creso == td._creso
+        assert td.max._value == np.iinfo(np.int64).max
+
+    def test_resolution(self, td):
+        expected = Timedelta._from_value_and_reso(1, td._creso)
+        result = td.resolution
+        assert result == expected
+        assert result._creso == expected._creso
+
+    def test_hash(self) -> None:
+        # GH#54037
+        second_resolution_max = Timedelta(0).as_unit("s").max
+
+        assert hash(second_resolution_max)
+
+
+def test_timedelta_class_min_max_resolution():
+    # when accessed on the class (as opposed to an instance), we default
+    #  to nanoseconds
+    assert Timedelta.min == Timedelta(NaT._value + 1)
+    assert Timedelta.min._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    assert Timedelta.max == Timedelta(np.iinfo(np.int64).max)
+    assert Timedelta.max._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    assert Timedelta.resolution == Timedelta(1)
+    assert Timedelta.resolution._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+
+class TestTimedeltaUnaryOps:
+    def test_invert(self):
+        td = Timedelta(10, unit="D")
+
+        msg = "bad operand type for unary ~"
+        with pytest.raises(TypeError, match=msg):
+            ~td
+
+        # check this matches pytimedelta and timedelta64
+        with pytest.raises(TypeError, match=msg):
+            ~(td.to_pytimedelta())
+
+        umsg = "ufunc 'invert' not supported for the input types"
+        with pytest.raises(TypeError, match=umsg):
+            ~(td.to_timedelta64())
+
+    def test_unary_ops(self):
+        td = Timedelta(10, unit="D")
+
+        # __neg__, __pos__
+        assert -td == Timedelta(-10, unit="D")
+        assert -td == Timedelta("-10D")
+        assert +td == Timedelta(10, unit="D")
+
+        # __abs__, __abs__(__neg__)
+        assert abs(td) == td
+        assert abs(-td) == td
+        assert abs(-td) == Timedelta("10D")
+
+
+class TestTimedeltas:
+    @pytest.mark.parametrize(
+        "unit, value, expected",
+        [
+            ("us", 9.999, 9999),
+            ("ms", 9.999999, 9999999),
+            ("s", 9.999999999, 9999999999),
+        ],
+    )
+    def test_rounding_on_int_unit_construction(self, unit, value, expected):
+        # GH 12690
+        result = Timedelta(value, unit=unit)
+        assert result._value == expected
+        result = Timedelta(str(value) + unit)
+        assert result._value == expected
+
+    def test_total_seconds_scalar(self):
+        # see gh-10939
+        rng = Timedelta("1 days, 10:11:12.100123456")
+        expt = 1 * 86400 + 10 * 3600 + 11 * 60 + 12 + 100123456.0 / 1e9
+        tm.assert_almost_equal(rng.total_seconds(), expt)
+
+        rng = Timedelta(np.nan)
+        assert np.isnan(rng.total_seconds())
+
+    def test_conversion(self):
+        for td in [Timedelta(10, unit="D"), Timedelta("1 days, 10:11:12.012345")]:
+            pydt = td.to_pytimedelta()
+            assert td == Timedelta(pydt)
+            assert td == pydt
+            assert isinstance(pydt, timedelta) and not isinstance(pydt, Timedelta)
+
+            assert td == np.timedelta64(td._value, "ns")
+            td64 = td.to_timedelta64()
+
+            assert td64 == np.timedelta64(td._value, "ns")
+            assert td == td64
+
+            assert isinstance(td64, np.timedelta64)
+
+        # this is NOT equal and cannot be roundtripped (because of the nanos)
+        td = Timedelta("1 days, 10:11:12.012345678")
+        assert td != td.to_pytimedelta()
+
+    def test_fields(self):
+        def check(value):
+            # that we are int
+            assert isinstance(value, int)
+
+        # compat to datetime.timedelta
+        rng = to_timedelta("1 days, 10:11:12")
+        assert rng.days == 1
+        assert rng.seconds == 10 * 3600 + 11 * 60 + 12
+        assert rng.microseconds == 0
+        assert rng.nanoseconds == 0
+
+        msg = "'Timedelta' object has no attribute '{}'"
+        with pytest.raises(AttributeError, match=msg.format("hours")):
+            rng.hours
+        with pytest.raises(AttributeError, match=msg.format("minutes")):
+            rng.minutes
+        with pytest.raises(AttributeError, match=msg.format("milliseconds")):
+            rng.milliseconds
+
+        # GH 10050
+        check(rng.days)
+        check(rng.seconds)
+        check(rng.microseconds)
+        check(rng.nanoseconds)
+
+        td = Timedelta("-1 days, 10:11:12")
+        assert abs(td) == Timedelta("13:48:48")
+        assert str(td) == "-1 days +10:11:12"
+        assert -td == Timedelta("0 days 13:48:48")
+        assert -Timedelta("-1 days, 10:11:12")._value == 49728000000000
+        assert Timedelta("-1 days, 10:11:12")._value == -49728000000000
+
+        rng = to_timedelta("-1 days, 10:11:12.100123456")
+        assert rng.days == -1
+        assert rng.seconds == 10 * 3600 + 11 * 60 + 12
+        assert rng.microseconds == 100 * 1000 + 123
+        assert rng.nanoseconds == 456
+        msg = "'Timedelta' object has no attribute '{}'"
+        with pytest.raises(AttributeError, match=msg.format("hours")):
+            rng.hours
+        with pytest.raises(AttributeError, match=msg.format("minutes")):
+            rng.minutes
+        with pytest.raises(AttributeError, match=msg.format("milliseconds")):
+            rng.milliseconds
+
+        # components
+        tup = to_timedelta(-1, "us").components
+        assert tup.days == -1
+        assert tup.hours == 23
+        assert tup.minutes == 59
+        assert tup.seconds == 59
+        assert tup.milliseconds == 999
+        assert tup.microseconds == 999
+        assert tup.nanoseconds == 0
+
+        # GH 10050
+        check(tup.days)
+        check(tup.hours)
+        check(tup.minutes)
+        check(tup.seconds)
+        check(tup.milliseconds)
+        check(tup.microseconds)
+        check(tup.nanoseconds)
+
+        tup = Timedelta("-1 days 1 us").components
+        assert tup.days == -2
+        assert tup.hours == 23
+        assert tup.minutes == 59
+        assert tup.seconds == 59
+        assert tup.milliseconds == 999
+        assert tup.microseconds == 999
+        assert tup.nanoseconds == 0
+
+    # TODO: this is a test of to_timedelta string parsing
+    def test_iso_conversion(self):
+        # GH #21877
+        expected = Timedelta(1, unit="s")
+        assert to_timedelta("P0DT0H0M1S") == expected
+
+    # TODO: this is a test of to_timedelta returning NaT
+    def test_nat_converters(self):
+        result = to_timedelta("nat").to_numpy()
+        assert result.dtype.kind == "M"
+        assert result.astype("int64") == iNaT
+
+        result = to_timedelta("nan").to_numpy()
+        assert result.dtype.kind == "M"
+        assert result.astype("int64") == iNaT
+
+    def test_numeric_conversions(self):
+        assert Timedelta(0) == np.timedelta64(0, "ns")
+        assert Timedelta(10) == np.timedelta64(10, "ns")
+        assert Timedelta(10, unit="ns") == np.timedelta64(10, "ns")
+
+        assert Timedelta(10, unit="us") == np.timedelta64(10, "us")
+        assert Timedelta(10, unit="ms") == np.timedelta64(10, "ms")
+        assert Timedelta(10, unit="s") == np.timedelta64(10, "s")
+        assert Timedelta(10, unit="D") == np.timedelta64(10, "D")
+
+    def test_timedelta_conversions(self):
+        assert Timedelta(timedelta(seconds=1)) == np.timedelta64(1, "s").astype(
+            "m8[ns]"
+        )
+        assert Timedelta(timedelta(microseconds=1)) == np.timedelta64(1, "us").astype(
+            "m8[ns]"
+        )
+        assert Timedelta(timedelta(days=1)) == np.timedelta64(1, "D").astype("m8[ns]")
+
+    def test_to_numpy_alias(self):
+        # GH 24653: alias .to_numpy() for scalars
+        td = Timedelta("10m7s")
+        assert td.to_timedelta64() == td.to_numpy()
+
+        # GH#44460
+        msg = "dtype and copy arguments are ignored"
+        with pytest.raises(ValueError, match=msg):
+            td.to_numpy("m8[s]")
+        with pytest.raises(ValueError, match=msg):
+            td.to_numpy(copy=True)
+
+    def test_identity(self):
+        td = Timedelta(10, unit="D")
+        assert isinstance(td, Timedelta)
+        assert isinstance(td, timedelta)
+
+    def test_short_format_converters(self):
+        def conv(v):
+            return v.astype("m8[ns]")
+
+        assert Timedelta("10") == np.timedelta64(10, "ns")
+        assert Timedelta("10ns") == np.timedelta64(10, "ns")
+        assert Timedelta("100") == np.timedelta64(100, "ns")
+        assert Timedelta("100ns") == np.timedelta64(100, "ns")
+
+        assert Timedelta("1000") == np.timedelta64(1000, "ns")
+        assert Timedelta("1000ns") == np.timedelta64(1000, "ns")
+
+        msg = "'NS' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert Timedelta("1000NS") == np.timedelta64(1000, "ns")
+
+        assert Timedelta("10us") == np.timedelta64(10000, "ns")
+        assert Timedelta("100us") == np.timedelta64(100000, "ns")
+        assert Timedelta("1000us") == np.timedelta64(1000000, "ns")
+        assert Timedelta("1000Us") == np.timedelta64(1000000, "ns")
+        assert Timedelta("1000uS") == np.timedelta64(1000000, "ns")
+
+        assert Timedelta("1ms") == np.timedelta64(1000000, "ns")
+        assert Timedelta("10ms") == np.timedelta64(10000000, "ns")
+        assert Timedelta("100ms") == np.timedelta64(100000000, "ns")
+        assert Timedelta("1000ms") == np.timedelta64(1000000000, "ns")
+
+        assert Timedelta("-1s") == -np.timedelta64(1000000000, "ns")
+        assert Timedelta("1s") == np.timedelta64(1000000000, "ns")
+        assert Timedelta("10s") == np.timedelta64(10000000000, "ns")
+        assert Timedelta("100s") == np.timedelta64(100000000000, "ns")
+        assert Timedelta("1000s") == np.timedelta64(1000000000000, "ns")
+
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert Timedelta("1d") == conv(np.timedelta64(1, "D"))
+        assert Timedelta("-1D") == -conv(np.timedelta64(1, "D"))
+        assert Timedelta("1D") == conv(np.timedelta64(1, "D"))
+        assert Timedelta("10D") == conv(np.timedelta64(10, "D"))
+        assert Timedelta("100D") == conv(np.timedelta64(100, "D"))
+        assert Timedelta("1000D") == conv(np.timedelta64(1000, "D"))
+        assert Timedelta("10000D") == conv(np.timedelta64(10000, "D"))
+
+        # space
+        assert Timedelta(" 10000D ") == conv(np.timedelta64(10000, "D"))
+        assert Timedelta(" - 10000D ") == -conv(np.timedelta64(10000, "D"))
+
+        # invalid
+        msg = "invalid unit abbreviation"
+        with pytest.raises(ValueError, match=msg):
+            Timedelta("1foo")
+        msg = "unit abbreviation w/o a number"
+        with pytest.raises(ValueError, match=msg):
+            Timedelta("foo")
+
+    def test_full_format_converters(self):
+        def conv(v):
+            return v.astype("m8[ns]")
+
+        d1 = np.timedelta64(1, "D")
+
+        assert Timedelta("1days") == conv(d1)
+        assert Timedelta("1days,") == conv(d1)
+        assert Timedelta("- 1days,") == -conv(d1)
+
+        assert Timedelta("00:00:01") == conv(np.timedelta64(1, "s"))
+        assert Timedelta("06:00:01") == conv(np.timedelta64(6 * 3600 + 1, "s"))
+        assert Timedelta("06:00:01.0") == conv(np.timedelta64(6 * 3600 + 1, "s"))
+        assert Timedelta("06:00:01.01") == conv(
+            np.timedelta64(1000 * (6 * 3600 + 1) + 10, "ms")
+        )
+
+        assert Timedelta("- 1days, 00:00:01") == conv(-d1 + np.timedelta64(1, "s"))
+        assert Timedelta("1days, 06:00:01") == conv(
+            d1 + np.timedelta64(6 * 3600 + 1, "s")
+        )
+        assert Timedelta("1days, 06:00:01.01") == conv(
+            d1 + np.timedelta64(1000 * (6 * 3600 + 1) + 10, "ms")
+        )
+
+        # invalid
+        msg = "have leftover units"
+        with pytest.raises(ValueError, match=msg):
+            Timedelta("- 1days, 00")
+
+    def test_pickle(self):
+        v = Timedelta("1 days 10:11:12.0123456")
+        v_p = tm.round_trip_pickle(v)
+        assert v == v_p
+
+    def test_timedelta_hash_equality(self):
+        # GH 11129
+        v = Timedelta(1, "D")
+        td = timedelta(days=1)
+        assert hash(v) == hash(td)
+
+        d = {td: 2}
+        assert d[v] == 2
+
+        tds = [Timedelta(seconds=1) + Timedelta(days=n) for n in range(20)]
+        assert all(hash(td) == hash(td.to_pytimedelta()) for td in tds)
+
+        # python timedeltas drop ns resolution
+        ns_td = Timedelta(1, "ns")
+        assert hash(ns_td) != hash(ns_td.to_pytimedelta())
+
+    @pytest.mark.skip_ubsan
+    @pytest.mark.xfail(
+        reason="pd.Timedelta violates the Python hash invariant (GH#44504).",
+    )
+    @given(
+        st.integers(
+            min_value=(-sys.maxsize - 1) // 500,
+            max_value=sys.maxsize // 500,
+        )
+    )
+    def test_hash_equality_invariance(self, half_microseconds: int) -> None:
+        # GH#44504
+
+        nanoseconds = half_microseconds * 500
+
+        pandas_timedelta = Timedelta(nanoseconds)
+        numpy_timedelta = np.timedelta64(nanoseconds)
+
+        # See: https://docs.python.org/3/glossary.html#term-hashable
+        # Hashable objects which compare equal must have the same hash value.
+        assert pandas_timedelta != numpy_timedelta or hash(pandas_timedelta) == hash(
+            numpy_timedelta
+        )
+
+    def test_implementation_limits(self):
+        min_td = Timedelta(Timedelta.min)
+        max_td = Timedelta(Timedelta.max)
+
+        # GH 12727
+        # timedelta limits correspond to int64 boundaries
+        assert min_td._value == iNaT + 1
+        assert max_td._value == lib.i8max
+
+        # Beyond lower limit, a NAT before the Overflow
+        assert (min_td - Timedelta(1, "ns")) is NaT
+
+        msg = "int too (large|big) to convert"
+        with pytest.raises(OverflowError, match=msg):
+            min_td - Timedelta(2, "ns")
+
+        with pytest.raises(OverflowError, match=msg):
+            max_td + Timedelta(1, "ns")
+
+        # Same tests using the internal nanosecond values
+        td = Timedelta(min_td._value - 1, "ns")
+        assert td is NaT
+
+        msg = "Cannot cast -9223372036854775809 from ns to 'ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta(min_td._value - 2, "ns")
+
+        msg = "Cannot cast 9223372036854775808 from ns to 'ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            Timedelta(max_td._value + 1, "ns")
+
+    def test_total_seconds_precision(self):
+        # GH 19458
+        assert Timedelta("30s").total_seconds() == 30.0
+        assert Timedelta("0").total_seconds() == 0.0
+        assert Timedelta("-2s").total_seconds() == -2.0
+        assert Timedelta("5.324s").total_seconds() == 5.324
+        assert (Timedelta("30s").total_seconds() - 30.0) < 1e-20
+        assert (30.0 - Timedelta("30s").total_seconds()) < 1e-20
+
+    def test_resolution_string(self):
+        assert Timedelta(days=1).resolution_string == "D"
+        assert Timedelta(days=1, hours=6).resolution_string == "h"
+        assert Timedelta(days=1, minutes=6).resolution_string == "min"
+        assert Timedelta(days=1, seconds=6).resolution_string == "s"
+        assert Timedelta(days=1, milliseconds=6).resolution_string == "ms"
+        assert Timedelta(days=1, microseconds=6).resolution_string == "us"
+        assert Timedelta(days=1, nanoseconds=6).resolution_string == "ns"
+
+    def test_resolution_deprecated(self):
+        # GH#21344
+        td = Timedelta(days=4, hours=3)
+        result = td.resolution
+        assert result == Timedelta(nanoseconds=1)
+
+        # Check that the attribute is available on the class, mirroring
+        #  the stdlib timedelta behavior
+        result = Timedelta.resolution
+        assert result == Timedelta(nanoseconds=1)
+
+    @pytest.mark.parametrize(
+        "unit,unit_depr",
+        [
+            ("W", "w"),
+            ("D", "d"),
+            ("min", "MIN"),
+            ("s", "S"),
+            ("h", "H"),
+            ("ms", "MS"),
+            ("us", "US"),
+        ],
+    )
+    def test_unit_deprecated(self, unit, unit_depr):
+        # GH#59051
+        msg = f"'{unit_depr}' is deprecated and will be removed in a future version."
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = Timedelta(1, unit_depr)
+        assert result == Timedelta(1, unit)
+
+
+@pytest.mark.parametrize(
+    "value, expected",
+    [
+        (Timedelta("10s"), True),
+        (Timedelta("-10s"), True),
+        (Timedelta(10, unit="ns"), True),
+        (Timedelta(0, unit="ns"), False),
+        (Timedelta(-10, unit="ns"), True),
+        (Timedelta(None), True),
+        (NaT, True),
+    ],
+)
+def test_truthiness(value, expected):
+    # https://github.com/pandas-dev/pandas/issues/21484
+    assert bool(value) is expected
+
+
+def test_timedelta_attribute_precision():
+    # GH 31354
+    td = Timedelta(1552211999999999872, unit="ns")
+    result = td.days * 86400
+    result += td.seconds
+    result *= 1000000
+    result += td.microseconds
+    result *= 1000
+    result += td.nanoseconds
+    expected = td._value
+    assert result == expected
+
+
+def test_to_pytimedelta_large_values():
+    td = Timedelta(1152921504609987375, unit="ns")
+    result = td.to_pytimedelta()
+    expected = timedelta(days=13343, seconds=86304, microseconds=609987)
+    assert result == expected
diff --git a/pandas/pandas/tests/scalar/timestamp/__init__.py b/pandas/pandas/tests/scalar/timestamp/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/__init__.py b/pandas/pandas/tests/scalar/timestamp/methods/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_as_unit.py b/pandas/pandas/tests/scalar/timestamp/methods/test_as_unit.py
new file mode 100644
index 0000000000000..5572c5936fb12
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_as_unit.py
@@ -0,0 +1,86 @@
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.errors import OutOfBoundsDatetime
+
+from pandas import Timestamp
+
+
+class TestTimestampAsUnit:
+    def test_as_unit(self):
+        ts = Timestamp("1970-01-01").as_unit("ns")
+        assert ts.unit == "ns"
+
+        assert ts.as_unit("ns") is ts
+
+        res = ts.as_unit("us")
+        assert res._value == ts._value // 1000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == ts._value
+        assert rt._creso == ts._creso
+
+        res = ts.as_unit("ms")
+        assert res._value == ts._value // 1_000_000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == ts._value
+        assert rt._creso == ts._creso
+
+        res = ts.as_unit("s")
+        assert res._value == ts._value // 1_000_000_000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+        rt = res.as_unit("ns")
+        assert rt._value == ts._value
+        assert rt._creso == ts._creso
+
+    def test_as_unit_overflows(self):
+        # microsecond that would be just out of bounds for nano
+        us = 9223372800000000
+        ts = Timestamp._from_value_and_reso(us, NpyDatetimeUnit.NPY_FR_us.value, None)
+
+        msg = "Cannot cast 2262-04-12 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            ts.as_unit("ns")
+
+        res = ts.as_unit("ms")
+        assert res._value == us // 1000
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+
+    def test_as_unit_rounding(self):
+        ts = Timestamp(1_500_000)  # i.e. 1500 microseconds
+        res = ts.as_unit("ms")
+
+        expected = Timestamp(1_000_000)  # i.e. 1 millisecond
+        assert res == expected
+
+        assert res._creso == NpyDatetimeUnit.NPY_FR_ms.value
+        assert res._value == 1
+
+        with pytest.raises(ValueError, match="Cannot losslessly convert units"):
+            ts.as_unit("ms", round_ok=False)
+
+    def test_as_unit_non_nano(self):
+        # case where we are going neither to nor from nano
+        ts = Timestamp("1970-01-02").as_unit("ms")
+        assert ts.year == 1970
+        assert ts.month == 1
+        assert ts.day == 2
+        assert ts.hour == ts.minute == ts.second == ts.microsecond == ts.nanosecond == 0
+
+        res = ts.as_unit("s")
+        assert res._value == 24 * 3600
+        assert res.year == 1970
+        assert res.month == 1
+        assert res.day == 2
+        assert (
+            res.hour
+            == res.minute
+            == res.second
+            == res.microsecond
+            == res.nanosecond
+            == 0
+        )
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_normalize.py b/pandas/pandas/tests/scalar/timestamp/methods/test_normalize.py
new file mode 100644
index 0000000000000..60f249c602bd6
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_normalize.py
@@ -0,0 +1,21 @@
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+
+
+class TestTimestampNormalize:
+    @pytest.mark.parametrize("arg", ["2013-11-30", "2013-11-30 12:00:00"])
+    def test_normalize(self, tz_naive_fixture, arg, unit):
+        tz = tz_naive_fixture
+        ts = Timestamp(arg, tz=tz).as_unit(unit)
+        result = ts.normalize()
+        expected = Timestamp("2013-11-30", tz=tz)
+        assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+    def test_normalize_pre_epoch_dates(self):
+        # GH: 36294
+        result = Timestamp("1969-01-01 09:00:00").normalize()
+        expected = Timestamp("1969-01-01 00:00:00")
+        assert result == expected
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_replace.py b/pandas/pandas/tests/scalar/timestamp/methods/test_replace.py
new file mode 100644
index 0000000000000..f15ea0e485cae
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_replace.py
@@ -0,0 +1,197 @@
+from datetime import datetime
+import zoneinfo
+
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    Timestamp,
+    conversion,
+)
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.compat import WASM
+import pandas.util._test_decorators as td
+
+import pandas._testing as tm
+
+
+class TestTimestampReplace:
+    def test_replace_out_of_pydatetime_bounds(self):
+        # GH#50348
+        ts = Timestamp("2016-01-01").as_unit("ns")
+
+        msg = "Out of bounds timestamp: 99999-01-01 00:00:00 with frequency 'ns'"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            ts.replace(year=99_999)
+
+        ts = ts.as_unit("ms")
+        result = ts.replace(year=99_999)
+        assert result.year == 99_999
+        assert result._value == Timestamp(np.datetime64("99999-01-01", "ms"))._value
+
+    def test_replace_non_nano(self):
+        ts = Timestamp._from_value_and_reso(
+            91514880000000000, NpyDatetimeUnit.NPY_FR_us.value, None
+        )
+        assert ts.to_pydatetime() == datetime(4869, 12, 28)
+
+        result = ts.replace(year=4900)
+        assert result._creso == ts._creso
+        assert result.to_pydatetime() == datetime(4900, 12, 28)
+
+    def test_replace_naive(self):
+        # GH#14621, GH#7825
+        ts = Timestamp("2016-01-01 09:00:00")
+        result = ts.replace(hour=0)
+        expected = Timestamp("2016-01-01 00:00:00")
+        assert result == expected
+
+    def test_replace_aware(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH#14621, GH#7825
+        # replacing datetime components with and w/o presence of a timezone
+        ts = Timestamp("2016-01-01 09:00:00", tz=tz)
+        result = ts.replace(hour=0)
+        expected = Timestamp("2016-01-01 00:00:00", tz=tz)
+        assert result == expected
+
+    def test_replace_preserves_nanos(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH#14621, GH#7825
+        ts = Timestamp("2016-01-01 09:00:00.000000123", tz=tz)
+        result = ts.replace(hour=0)
+        expected = Timestamp("2016-01-01 00:00:00.000000123", tz=tz)
+        assert result == expected
+
+    def test_replace_multiple(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH#14621, GH#7825
+        # replacing datetime components with and w/o presence of a timezone
+        # test all
+        ts = Timestamp("2016-01-01 09:00:00.000000123", tz=tz)
+        result = ts.replace(
+            year=2015,
+            month=2,
+            day=2,
+            hour=0,
+            minute=5,
+            second=5,
+            microsecond=5,
+            nanosecond=5,
+        )
+        expected = Timestamp("2015-02-02 00:05:05.000005005", tz=tz)
+        assert result == expected
+
+    def test_replace_invalid_kwarg(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH#14621, GH#7825
+        ts = Timestamp("2016-01-01 09:00:00.000000123", tz=tz)
+        msg = r"replace\(\) got an unexpected keyword argument"
+        with pytest.raises(TypeError, match=msg):
+            ts.replace(foo=5)
+
+    def test_replace_integer_args(self, tz_aware_fixture):
+        tz = tz_aware_fixture
+        # GH#14621, GH#7825
+        ts = Timestamp("2016-01-01 09:00:00.000000123", tz=tz)
+        msg = "value must be an integer, received <class 'float'> for hour"
+        with pytest.raises(ValueError, match=msg):
+            ts.replace(hour=0.1)
+
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    def test_replace_tzinfo_equiv_tz_localize_none(self):
+        # GH#14621, GH#7825
+        # assert conversion to naive is the same as replacing tzinfo with None
+        ts = Timestamp("2013-11-03 01:59:59.999999-0400", tz="US/Eastern")
+        assert ts.tz_localize(None) == ts.replace(tzinfo=None)
+
+    @td.skip_if_windows
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    def test_replace_tzinfo(self):
+        # GH#15683
+        dt = datetime(2016, 3, 27, 1, fold=1)
+        tzinfo = dt.astimezone(zoneinfo.ZoneInfo("Europe/Berlin")).tzinfo
+
+        result_dt = dt.replace(tzinfo=tzinfo)
+        result_pd = Timestamp(dt).replace(tzinfo=tzinfo)
+
+        # datetime.timestamp() converts in the local timezone
+        with tm.set_timezone("UTC"):
+            assert result_dt.timestamp() == result_pd.timestamp()
+
+        assert result_dt == result_pd
+        assert result_dt == result_pd.to_pydatetime()
+
+        result_dt = dt.replace(tzinfo=tzinfo).replace(tzinfo=None)
+        result_pd = Timestamp(dt).replace(tzinfo=tzinfo).replace(tzinfo=None)
+
+        # datetime.timestamp() converts in the local timezone
+        with tm.set_timezone("UTC"):
+            assert result_dt.timestamp() == result_pd.timestamp()
+
+        assert result_dt == result_pd
+        assert result_dt == result_pd.to_pydatetime()
+
+    @pytest.mark.parametrize(
+        "tz, normalize",
+        [
+            ("pytz/US/Eastern", lambda x: x.tzinfo.normalize(x)),
+            (gettz("US/Eastern"), lambda x: x),
+        ],
+    )
+    def test_replace_across_dst(self, tz, normalize):
+        # GH#18319 check that 1) timezone is correctly normalized and
+        # 2) that hour is not incorrectly changed by this normalization
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        ts_naive = Timestamp("2017-12-03 16:03:30")
+        ts_aware = conversion.localize_pydatetime(ts_naive, tz)
+
+        # Preliminary sanity-check
+        assert ts_aware == normalize(ts_aware)
+
+        # Replace across DST boundary
+        ts2 = ts_aware.replace(month=6)
+
+        # Check that `replace` preserves hour literal
+        assert (ts2.hour, ts2.minute) == (ts_aware.hour, ts_aware.minute)
+
+        # Check that post-replace object is appropriately normalized
+        ts2b = normalize(ts2)
+        assert ts2 == ts2b
+
+    def test_replace_dst_border(self, unit):
+        # Gh 7825
+        t = Timestamp("2013-11-3", tz="America/Chicago").as_unit(unit)
+        result = t.replace(hour=3)
+        expected = Timestamp("2013-11-3 03:00:00", tz="America/Chicago")
+        assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+    @pytest.mark.parametrize("fold", [0, 1])
+    @pytest.mark.parametrize("tz", ["dateutil/Europe/London", "Europe/London"])
+    def test_replace_dst_fold(self, fold, tz, unit):
+        # GH 25017
+        d = datetime(2019, 10, 27, 2, 30)
+        ts = Timestamp(d, tz=tz).as_unit(unit)
+        result = ts.replace(hour=1, fold=fold)
+        expected = Timestamp(datetime(2019, 10, 27, 1, 30)).tz_localize(
+            tz, ambiguous=not fold
+        )
+        assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+    @pytest.mark.parametrize("fold", [0, 1])
+    def test_replace_preserves_fold(self, fold):
+        # GH#37610. Check that replace preserves Timestamp fold property
+        tz = gettz("Europe/Moscow")
+
+        ts = Timestamp(
+            year=2009, month=10, day=25, hour=2, minute=30, fold=fold, tzinfo=tz
+        )
+        ts_replaced = ts.replace(second=1)
+
+        assert ts_replaced.fold == fold
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_round.py b/pandas/pandas/tests/scalar/timestamp/methods/test_round.py
new file mode 100644
index 0000000000000..6b27e5e6c5554
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_round.py
@@ -0,0 +1,371 @@
+from hypothesis import (
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._libs.tslibs import (
+    NaT,
+    OutOfBoundsDatetime,
+    Timedelta,
+    Timestamp,
+    iNaT,
+    to_offset,
+)
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+
+import pandas._testing as tm
+
+
+class TestTimestampRound:
+    def test_round_division_by_zero_raises(self):
+        ts = Timestamp("2016-01-01")
+
+        msg = "Division by zero in rounding"
+        with pytest.raises(ValueError, match=msg):
+            ts.round("0ns")
+
+    @pytest.mark.parametrize(
+        "timestamp, freq, expected",
+        [
+            ("20130101 09:10:11", "D", "20130101"),
+            ("20130101 19:10:11", "D", "20130102"),
+            ("20130201 12:00:00", "D", "20130202"),
+            ("20130104 12:00:00", "D", "20130105"),
+            ("2000-01-05 05:09:15.13", "D", "2000-01-05 00:00:00"),
+            ("2000-01-05 05:09:15.13", "h", "2000-01-05 05:00:00"),
+            ("2000-01-05 05:09:15.13", "s", "2000-01-05 05:09:15"),
+        ],
+    )
+    def test_round_frequencies(self, timestamp, freq, expected):
+        dt = Timestamp(timestamp)
+        result = dt.round(freq)
+        expected = Timestamp(expected)
+        assert result == expected
+
+    def test_round_tzaware(self):
+        dt = Timestamp("20130101 09:10:11", tz="US/Eastern")
+        result = dt.round("D")
+        expected = Timestamp("20130101", tz="US/Eastern")
+        assert result == expected
+
+        dt = Timestamp("20130101 09:10:11", tz="US/Eastern")
+        result = dt.round("s")
+        assert result == dt
+
+    def test_round_30min(self):
+        # round
+        dt = Timestamp("20130104 12:32:00")
+        result = dt.round("30Min")
+        expected = Timestamp("20130104 12:30:00")
+        assert result == expected
+
+    def test_round_subsecond(self):
+        # GH#14440 & GH#15578
+        result = Timestamp("2016-10-17 12:00:00.0015").round("ms")
+        expected = Timestamp("2016-10-17 12:00:00.002000")
+        assert result == expected
+
+        result = Timestamp("2016-10-17 12:00:00.00149").round("ms")
+        expected = Timestamp("2016-10-17 12:00:00.001000")
+        assert result == expected
+
+        ts = Timestamp("2016-10-17 12:00:00.0015")
+        for freq in ["us", "ns"]:
+            assert ts == ts.round(freq)
+
+        result = Timestamp("2016-10-17 12:00:00.001501031").round("10ns")
+        expected = Timestamp("2016-10-17 12:00:00.001501030")
+        assert result == expected
+
+    def test_round_nonstandard_freq(self):
+        with tm.assert_produces_warning(False):
+            Timestamp("2016-10-17 12:00:00.001501031").round("1010ns")
+
+    def test_round_invalid_arg(self):
+        stamp = Timestamp("2000-01-05 05:09:15.13")
+        with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+            stamp.round("foo")
+
+    @pytest.mark.parametrize(
+        "test_input, rounder, freq, expected",
+        [
+            ("2117-01-01 00:00:45", "floor", "15s", "2117-01-01 00:00:45"),
+            ("2117-01-01 00:00:45", "ceil", "15s", "2117-01-01 00:00:45"),
+            (
+                "2117-01-01 00:00:45.000000012",
+                "floor",
+                "10ns",
+                "2117-01-01 00:00:45.000000010",
+            ),
+            (
+                "1823-01-01 00:00:01.000000012",
+                "ceil",
+                "10ns",
+                "1823-01-01 00:00:01.000000020",
+            ),
+            ("1823-01-01 00:00:01", "floor", "1s", "1823-01-01 00:00:01"),
+            ("1823-01-01 00:00:01", "ceil", "1s", "1823-01-01 00:00:01"),
+            ("NaT", "floor", "1s", "NaT"),
+            ("NaT", "ceil", "1s", "NaT"),
+        ],
+    )
+    def test_ceil_floor_edge(self, test_input, rounder, freq, expected):
+        dt = Timestamp(test_input)
+        func = getattr(dt, rounder)
+        result = func(freq)
+
+        if dt is NaT:
+            assert result is NaT
+        else:
+            expected = Timestamp(expected)
+            assert result == expected
+
+    @pytest.mark.parametrize(
+        "test_input, freq, expected",
+        [
+            ("2018-01-01 00:02:06", "2s", "2018-01-01 00:02:06"),
+            ("2018-01-01 00:02:00", "2min", "2018-01-01 00:02:00"),
+            ("2018-01-01 00:04:00", "4min", "2018-01-01 00:04:00"),
+            ("2018-01-01 00:15:00", "15min", "2018-01-01 00:15:00"),
+            ("2018-01-01 00:20:00", "20min", "2018-01-01 00:20:00"),
+            ("2018-01-01 03:00:00", "3h", "2018-01-01 03:00:00"),
+        ],
+    )
+    @pytest.mark.parametrize("rounder", ["ceil", "floor", "round"])
+    def test_round_minute_freq(self, test_input, freq, expected, rounder):
+        # Ensure timestamps that shouldn't round dont!
+        # GH#21262
+
+        dt = Timestamp(test_input)
+        expected = Timestamp(expected)
+        func = getattr(dt, rounder)
+        result = func(freq)
+        assert result == expected
+
+    def test_ceil(self, unit):
+        dt = Timestamp("20130101 09:10:11").as_unit(unit)
+        result = dt.ceil("D")
+        expected = Timestamp("20130102")
+        assert result == expected
+        assert result._creso == dt._creso
+
+    def test_floor(self, unit):
+        dt = Timestamp("20130101 09:10:11").as_unit(unit)
+        result = dt.floor("D")
+        expected = Timestamp("20130101")
+        assert result == expected
+        assert result._creso == dt._creso
+
+    @pytest.mark.parametrize("method", ["ceil", "round", "floor"])
+    def test_round_dst_border_ambiguous(self, method, unit):
+        # GH 18946 round near "fall back" DST
+        ts = Timestamp("2017-10-29 00:00:00", tz="UTC").tz_convert("Europe/Madrid")
+        ts = ts.as_unit(unit)
+        result = getattr(ts, method)("h", ambiguous=True)
+        assert result == ts
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+        result = getattr(ts, method)("h", ambiguous=False)
+        expected = Timestamp("2017-10-29 01:00:00", tz="UTC").tz_convert(
+            "Europe/Madrid"
+        )
+        assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+        result = getattr(ts, method)("h", ambiguous="NaT")
+        assert result is NaT
+
+        msg = "Cannot infer dst time"
+        with pytest.raises(ValueError, match=msg):
+            getattr(ts, method)("h", ambiguous="raise")
+
+    @pytest.mark.parametrize(
+        "method, ts_str, freq",
+        [
+            ["ceil", "2018-03-11 01:59:00-0600", "5min"],
+            ["round", "2018-03-11 01:59:00-0600", "5min"],
+            ["floor", "2018-03-11 03:01:00-0500", "2h"],
+        ],
+    )
+    def test_round_dst_border_nonexistent(self, method, ts_str, freq, unit):
+        # GH 23324 round near "spring forward" DST
+        ts = Timestamp(ts_str, tz="America/Chicago").as_unit(unit)
+        result = getattr(ts, method)(freq, nonexistent="shift_forward")
+        expected = Timestamp("2018-03-11 03:00:00", tz="America/Chicago")
+        assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+        result = getattr(ts, method)(freq, nonexistent="NaT")
+        assert result is NaT
+
+        msg = "2018-03-11 02:00:00"
+        with pytest.raises(ValueError, match=msg):
+            getattr(ts, method)(freq, nonexistent="raise")
+
+    @pytest.mark.parametrize(
+        "timestamp",
+        [
+            "2018-01-01 0:0:0.124999360",
+            "2018-01-01 0:0:0.125000367",
+            "2018-01-01 0:0:0.125500",
+            "2018-01-01 0:0:0.126500",
+            "2018-01-01 12:00:00",
+            "2019-01-01 12:00:00",
+        ],
+    )
+    @pytest.mark.parametrize(
+        "freq",
+        [
+            "2ns",
+            "3ns",
+            "4ns",
+            "5ns",
+            "6ns",
+            "7ns",
+            "250ns",
+            "500ns",
+            "750ns",
+            "1us",
+            "19us",
+            "250us",
+            "500us",
+            "750us",
+            "1s",
+            "2s",
+            "3s",
+            "1D",
+        ],
+    )
+    def test_round_int64(self, timestamp, freq):
+        # check that all rounding modes are accurate to int64 precision
+        # see GH#22591
+        dt = Timestamp(timestamp).as_unit("ns")
+        unit = to_offset(freq).nanos
+
+        # test floor
+        result = dt.floor(freq)
+        assert result._value % unit == 0, f"floor not a {freq} multiple"
+        assert 0 <= dt._value - result._value < unit, "floor error"
+
+        # test ceil
+        result = dt.ceil(freq)
+        assert result._value % unit == 0, f"ceil not a {freq} multiple"
+        assert 0 <= result._value - dt._value < unit, "ceil error"
+
+        # test round
+        result = dt.round(freq)
+        assert result._value % unit == 0, f"round not a {freq} multiple"
+        assert abs(result._value - dt._value) <= unit // 2, "round error"
+        if unit % 2 == 0 and abs(result._value - dt._value) == unit // 2:
+            # round half to even
+            assert result._value // unit % 2 == 0, "round half to even error"
+
+    def test_round_implementation_bounds(self):
+        # See also: analogous test for Timedelta
+        result = Timestamp.min.ceil("s")
+        expected = Timestamp(1677, 9, 21, 0, 12, 44)
+        assert result == expected
+
+        result = Timestamp.max.floor("s")
+        expected = Timestamp.max - Timedelta(854775807)
+        assert result == expected
+
+        msg = "Cannot round 1677-09-21 00:12:43.145224193 to freq=<Second>"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.min.floor("s")
+
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.min.round("s")
+
+        msg = "Cannot round 2262-04-11 23:47:16.854775807 to freq=<Second>"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.max.ceil("s")
+
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.max.round("s")
+
+    @given(val=st.integers(iNaT + 1, lib.i8max))
+    @pytest.mark.parametrize(
+        "method", [Timestamp.round, Timestamp.floor, Timestamp.ceil]
+    )
+    def test_round_sanity(self, val, method):
+        cls = Timestamp
+        err_cls = OutOfBoundsDatetime
+
+        val = np.int64(val)
+        ts = cls(val)
+
+        def checker(ts, nanos, unit):
+            # First check that we do raise in cases where we should
+            if nanos == 1:
+                pass
+            else:
+                div, mod = divmod(ts._value, nanos)
+                diff = int(nanos - mod)
+                lb = ts._value - mod
+                assert lb <= ts._value  # i.e. no overflows with python ints
+                ub = ts._value + diff
+                assert ub > ts._value  # i.e. no overflows with python ints
+
+                msg = "without overflow"
+                if mod == 0:
+                    # We should never be raising in this
+                    pass
+                elif method is cls.ceil:
+                    if ub > cls.max._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif method is cls.floor:
+                    if lb < cls.min._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif mod >= diff:
+                    if ub > cls.max._value:
+                        with pytest.raises(err_cls, match=msg):
+                            method(ts, unit)
+                        return
+                elif lb < cls.min._value:
+                    with pytest.raises(err_cls, match=msg):
+                        method(ts, unit)
+                    return
+
+            res = method(ts, unit)
+
+            td = res - ts
+            diff = abs(td._value)
+            assert diff < nanos
+            assert res._value % nanos == 0
+
+            if method is cls.round:
+                assert diff <= nanos / 2
+            elif method is cls.floor:
+                assert res <= ts
+            elif method is cls.ceil:
+                assert res >= ts
+
+        nanos = 1
+        checker(ts, nanos, "ns")
+
+        nanos = 1000
+        checker(ts, nanos, "us")
+
+        nanos = 1_000_000
+        checker(ts, nanos, "ms")
+
+        nanos = 1_000_000_000
+        checker(ts, nanos, "s")
+
+        nanos = 60 * 1_000_000_000
+        checker(ts, nanos, "min")
+
+        nanos = 60 * 60 * 1_000_000_000
+        checker(ts, nanos, "h")
+
+        nanos = 24 * 60 * 60 * 1_000_000_000
+        checker(ts, nanos, "D")
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_timestamp_method.py b/pandas/pandas/tests/scalar/timestamp/methods/test_timestamp_method.py
new file mode 100644
index 0000000000000..beacaaf04e6b2
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_timestamp_method.py
@@ -0,0 +1,34 @@
+# NB: This is for the Timestamp.timestamp *method* specifically, not
+# the Timestamp class in general.
+from datetime import timezone
+
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas.compat import WASM
+import pandas.util._test_decorators as td
+
+import pandas._testing as tm
+
+
+class TestTimestampMethod:
+    @td.skip_if_windows
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    def test_timestamp(self, fixed_now_ts):
+        # GH#17329
+        # tz-naive --> treat it as if it were UTC for purposes of timestamp()
+        ts = fixed_now_ts
+        uts = ts.replace(tzinfo=timezone.utc)
+        assert ts.timestamp() == uts.timestamp()
+
+        tsc = Timestamp("2014-10-11 11:00:01.12345678", tz="US/Central")
+        utsc = tsc.tz_convert("UTC")
+
+        # utsc is a different representation of the same time
+        assert tsc.timestamp() == utsc.timestamp()
+
+        # datetime.timestamp() converts in the local timezone
+        with tm.set_timezone("UTC"):
+            # should agree with datetime.timestamp method
+            dt = ts.to_pydatetime()
+            assert dt.timestamp() == ts.timestamp()
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_to_julian_date.py b/pandas/pandas/tests/scalar/timestamp/methods/test_to_julian_date.py
new file mode 100644
index 0000000000000..7769614b601a4
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_to_julian_date.py
@@ -0,0 +1,28 @@
+from pandas import Timestamp
+
+
+class TestTimestampToJulianDate:
+    def test_compare_1700(self):
+        ts = Timestamp("1700-06-23")
+        res = ts.to_julian_date()
+        assert res == 2_342_145.5
+
+    def test_compare_2000(self):
+        ts = Timestamp("2000-04-12")
+        res = ts.to_julian_date()
+        assert res == 2_451_646.5
+
+    def test_compare_2100(self):
+        ts = Timestamp("2100-08-12")
+        res = ts.to_julian_date()
+        assert res == 2_488_292.5
+
+    def test_compare_hour01(self):
+        ts = Timestamp("2000-08-12T01:00:00")
+        res = ts.to_julian_date()
+        assert res == 2_451_768.5416666666666666
+
+    def test_compare_hour13(self):
+        ts = Timestamp("2000-08-12T13:00:00")
+        res = ts.to_julian_date()
+        assert res == 2_451_769.0416666666666666
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_to_pydatetime.py b/pandas/pandas/tests/scalar/timestamp/methods/test_to_pydatetime.py
new file mode 100644
index 0000000000000..07e57b51a7f1e
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_to_pydatetime.py
@@ -0,0 +1,83 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import pytest
+
+from pandas._libs.tslibs.timezones import dateutil_gettz as gettz
+import pandas.util._test_decorators as td
+
+from pandas import Timestamp
+import pandas._testing as tm
+
+
+class TestTimestampToPyDatetime:
+    def test_to_pydatetime_fold(self):
+        # GH#45087
+        tzstr = "dateutil/usr/share/zoneinfo/America/Chicago"
+        ts = Timestamp(year=2013, month=11, day=3, hour=1, minute=0, fold=1, tz=tzstr)
+        dt = ts.to_pydatetime()
+        assert dt.fold == 1
+
+    def test_to_pydatetime_nonzero_nano(self):
+        ts = Timestamp("2011-01-01 9:00:00.123456789")
+
+        # Warn the user of data loss (nanoseconds).
+        msg = "Discarding nonzero nanoseconds in conversion"
+        with tm.assert_produces_warning(UserWarning, match=msg):
+            expected = datetime(2011, 1, 1, 9, 0, 0, 123456)
+            result = ts.to_pydatetime()
+            assert result == expected
+
+    def test_timestamp_to_datetime(self):
+        stamp = Timestamp("20090415", tz="US/Eastern")
+        dtval = stamp.to_pydatetime()
+        assert stamp == dtval
+        assert stamp.tzinfo == dtval.tzinfo
+
+    def test_timestamp_to_pydatetime_dateutil(self):
+        stamp = Timestamp("20090415", tz="dateutil/US/Eastern")
+        dtval = stamp.to_pydatetime()
+        assert stamp == dtval
+        assert stamp.tzinfo == dtval.tzinfo
+
+    def test_timestamp_to_pydatetime_explicit_pytz(self):
+        pytz = pytest.importorskip("pytz")
+        stamp = Timestamp("20090415", tz=pytz.timezone("US/Eastern"))
+        dtval = stamp.to_pydatetime()
+        assert stamp == dtval
+        assert stamp.tzinfo == dtval.tzinfo
+
+    @td.skip_if_windows
+    def test_timestamp_to_pydatetime_explicit_dateutil(self):
+        stamp = Timestamp("20090415", tz=gettz("US/Eastern"))
+        dtval = stamp.to_pydatetime()
+        assert stamp == dtval
+        assert stamp.tzinfo == dtval.tzinfo
+
+    def test_to_pydatetime_bijective(self):
+        # Ensure that converting to datetime and back only loses precision
+        # by going from nanoseconds to microseconds.
+        exp_warning = None if Timestamp.max.nanosecond == 0 else UserWarning
+        with tm.assert_produces_warning(exp_warning):
+            pydt_max = Timestamp.max.to_pydatetime()
+
+        assert (
+            Timestamp(pydt_max).as_unit("ns")._value / 1000
+            == Timestamp.max._value / 1000
+        )
+
+        exp_warning = None if Timestamp.min.nanosecond == 0 else UserWarning
+        with tm.assert_produces_warning(exp_warning):
+            pydt_min = Timestamp.min.to_pydatetime()
+
+        # The next assertion can be enabled once GH#39221 is merged
+        #  assert pydt_min < Timestamp.min  # this is bc nanos are dropped
+        tdus = timedelta(microseconds=1)
+        assert pydt_min + tdus > Timestamp.min
+
+        assert (
+            Timestamp(pydt_min + tdus).as_unit("ns")._value / 1000
+            == Timestamp.min._value / 1000
+        )
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_tz_convert.py b/pandas/pandas/tests/scalar/timestamp/methods/test_tz_convert.py
new file mode 100644
index 0000000000000..a7bb3b90805ab
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_tz_convert.py
@@ -0,0 +1,51 @@
+import dateutil
+import pytest
+
+from pandas._libs.tslibs import timezones
+import pandas.util._test_decorators as td
+
+from pandas import Timestamp
+
+
+class TestTimestampTZConvert:
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_astimezone(self, tzstr):
+        # astimezone is an alias for tz_convert, so keep it with
+        # the tz_convert tests
+        utcdate = Timestamp("3/11/2012 22:00", tz="UTC")
+        expected = utcdate.tz_convert(tzstr)
+        result = utcdate.astimezone(tzstr)
+        assert expected == result
+        assert isinstance(result, Timestamp)
+
+    @pytest.mark.parametrize(
+        "stamp",
+        [
+            "2014-02-01 09:00",
+            "2014-07-08 09:00",
+            "2014-11-01 17:00",
+            "2014-11-05 00:00",
+        ],
+    )
+    def test_tz_convert_roundtrip(self, stamp, tz_aware_fixture):
+        tz = tz_aware_fixture
+
+        ts = Timestamp(stamp, tz="UTC")
+        converted = ts.tz_convert(tz)
+
+        reset = converted.tz_convert(None)
+        assert reset == Timestamp(stamp)
+        assert reset.tzinfo is None
+        assert reset == converted.tz_convert("UTC").tz_localize(None)
+
+    @td.skip_if_windows
+    def test_tz_convert_utc_with_system_utc(self):
+        # from system utc to real utc
+        ts = Timestamp("2001-01-05 11:56", tz=timezones.maybe_get_tz("dateutil/UTC"))
+        # check that the time hasn't changed.
+        assert ts == ts.tz_convert(dateutil.tz.tzutc())
+
+        # from system utc to real utc
+        ts = Timestamp("2001-01-05 11:56", tz=timezones.maybe_get_tz("dateutil/UTC"))
+        # check that the time hasn't changed.
+        assert ts == ts.tz_convert(dateutil.tz.tzutc())
diff --git a/pandas/pandas/tests/scalar/timestamp/methods/test_tz_localize.py b/pandas/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
new file mode 100644
index 0000000000000..cb7ac5fa6f1da
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
@@ -0,0 +1,336 @@
+from datetime import timedelta
+import re
+import zoneinfo
+
+from dateutil.tz import gettz
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.errors import OutOfBoundsDatetime
+
+from pandas import (
+    NaT,
+    Timestamp,
+)
+
+
+class TestTimestampTZLocalize:
+    @pytest.mark.skip_ubsan
+    def test_tz_localize_pushes_out_of_bounds(self):
+        # GH#12677
+        # tz_localize that pushes away from the boundary is OK
+        pytz = pytest.importorskip("pytz")
+        msg = (
+            f"Converting {Timestamp.min.strftime('%Y-%m-%d %H:%M:%S')} "
+            f"underflows past {Timestamp.min}"
+        )
+        pac = Timestamp.min.tz_localize(pytz.timezone("US/Pacific"))
+        assert pac._value > Timestamp.min._value
+        pac.tz_convert("Asia/Tokyo")  # tz_convert doesn't change value
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.min.tz_localize(pytz.timezone("Asia/Tokyo"))
+
+        # tz_localize that pushes away from the boundary is OK
+        msg = (
+            f"Converting {Timestamp.max.strftime('%Y-%m-%d %H:%M:%S')} "
+            f"overflows past {Timestamp.max}"
+        )
+        tokyo = Timestamp.max.tz_localize(pytz.timezone("Asia/Tokyo"))
+        assert tokyo._value < Timestamp.max._value
+        tokyo.tz_convert("US/Pacific")  # tz_convert doesn't change value
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp.max.tz_localize(pytz.timezone("US/Pacific"))
+
+    @pytest.mark.parametrize(
+        "tz",
+        [zoneinfo.ZoneInfo("US/Central"), "dateutil/US/Central", "pytz/US/Central"],
+    )
+    def test_tz_localize_ambiguous_bool(self, unit, tz):
+        # make sure that we are correctly accepting bool values as ambiguous
+        # GH#14402
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        ts = Timestamp("2015-11-01 01:00:03").as_unit(unit)
+        expected0 = Timestamp("2015-11-01 01:00:03-0500", tz=tz)
+        expected1 = Timestamp("2015-11-01 01:00:03-0600", tz=tz)
+
+        msg = "Cannot infer dst time from 2015-11-01 01:00:03"
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize(tz)
+
+        result = ts.tz_localize(tz, ambiguous=True)
+        assert result == expected0
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+        result = ts.tz_localize(tz, ambiguous=False)
+        assert result == expected1
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+    def test_tz_localize_ambiguous(self):
+        ts = Timestamp("2014-11-02 01:00")
+        ts_dst = ts.tz_localize("US/Eastern", ambiguous=True)
+        ts_no_dst = ts.tz_localize("US/Eastern", ambiguous=False)
+
+        assert ts_no_dst._value - ts_dst._value == 3600
+        msg = re.escape(
+            "'ambiguous' parameter must be one of: "
+            "True, False, 'NaT', 'raise' (default)"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize("US/Eastern", ambiguous="infer")
+
+        # GH#8025
+        msg = "Cannot localize tz-aware Timestamp, use tz_convert for conversions"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp("2011-01-01", tz="US/Eastern").tz_localize("Asia/Tokyo")
+
+        msg = "Cannot convert tz-naive Timestamp, use tz_localize to localize"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp("2011-01-01").tz_convert("Asia/Tokyo")
+
+    @pytest.mark.parametrize(
+        "stamp, tz",
+        [
+            ("2015-03-08 02:00", "US/Eastern"),
+            ("2015-03-08 02:30", "US/Pacific"),
+            ("2015-03-29 02:00", "Europe/Paris"),
+            ("2015-03-29 02:30", "Europe/Belgrade"),
+        ],
+    )
+    def test_tz_localize_nonexistent(self, stamp, tz):
+        # GH#13057
+        ts = Timestamp(stamp)
+        with pytest.raises(ValueError, match=stamp):
+            ts.tz_localize(tz)
+        # GH 22644
+        with pytest.raises(ValueError, match=stamp):
+            ts.tz_localize(tz, nonexistent="raise")
+        assert ts.tz_localize(tz, nonexistent="NaT") is NaT
+
+    @pytest.mark.parametrize(
+        "stamp, tz, forward_expected, backward_expected",
+        [
+            (
+                "2015-03-29 02:00:00",
+                "Europe/Warsaw",
+                "2015-03-29 03:00:00",
+                "2015-03-29 01:59:59",
+            ),  # utc+1 -> utc+2
+            (
+                "2023-03-12 02:00:00",
+                "America/Los_Angeles",
+                "2023-03-12 03:00:00",
+                "2023-03-12 01:59:59",
+            ),  # utc-8 -> utc-7
+            (
+                "2023-03-26 01:00:00",
+                "Europe/London",
+                "2023-03-26 02:00:00",
+                "2023-03-26 00:59:59",
+            ),  # utc+0 -> utc+1
+            (
+                "2023-03-26 00:00:00",
+                "Atlantic/Azores",
+                "2023-03-26 01:00:00",
+                "2023-03-25 23:59:59",
+            ),  # utc-1 -> utc+0
+        ],
+    )
+    def test_tz_localize_nonexistent_shift(
+        self, stamp, tz, forward_expected, backward_expected
+    ):
+        ts = Timestamp(stamp)
+        forward_ts = ts.tz_localize(tz, nonexistent="shift_forward")
+        assert forward_ts == Timestamp(forward_expected, tz=tz)
+        backward_ts = ts.tz_localize(tz, nonexistent="shift_backward")
+        assert backward_ts == Timestamp(backward_expected, tz=tz)
+
+    def test_tz_localize_ambiguous_raise(self):
+        # GH#13057
+        ts = Timestamp("2015-11-1 01:00")
+        msg = "Cannot infer dst time from 2015-11-01 01:00:00,"
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize("US/Pacific", ambiguous="raise")
+
+    def test_tz_localize_nonexistent_invalid_arg(self, warsaw):
+        # GH 22644
+        tz = warsaw
+        ts = Timestamp("2015-03-29 02:00:00")
+        msg = (
+            "The nonexistent argument must be one of 'raise', 'NaT', "
+            "'shift_forward', 'shift_backward' or a timedelta object"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize(tz, nonexistent="foo")
+
+    @pytest.mark.parametrize(
+        "stamp",
+        [
+            "2014-02-01 09:00",
+            "2014-07-08 09:00",
+            "2014-11-01 17:00",
+            "2014-11-05 00:00",
+        ],
+    )
+    def test_tz_localize_roundtrip(self, stamp, tz_aware_fixture):
+        tz = tz_aware_fixture
+        ts = Timestamp(stamp)
+        localized = ts.tz_localize(tz)
+        assert localized == Timestamp(stamp, tz=tz)
+
+        msg = "Cannot localize tz-aware Timestamp"
+        with pytest.raises(TypeError, match=msg):
+            localized.tz_localize(tz)
+
+        reset = localized.tz_localize(None)
+        assert reset == ts
+        assert reset.tzinfo is None
+
+    def test_tz_localize_ambiguous_compat(self):
+        # validate that pytz and dateutil are compat for dst
+        # when the transition happens
+        pytz = pytest.importorskip("pytz")
+        naive = Timestamp("2013-10-27 01:00:00")
+
+        pytz_zone = pytz.timezone("Europe/London")
+        dateutil_zone = "dateutil/Europe/London"
+        result_pytz = naive.tz_localize(pytz_zone, ambiguous=False)
+        result_dateutil = naive.tz_localize(dateutil_zone, ambiguous=False)
+        assert result_pytz._value == result_dateutil._value
+        assert result_pytz._value == 1382835600
+
+        # fixed ambiguous behavior
+        # see gh-14621, GH#45087
+        assert result_pytz.to_pydatetime().tzname() == "GMT"
+        assert result_dateutil.to_pydatetime().tzname() == "GMT"
+        assert str(result_pytz) == str(result_dateutil)
+
+        # 1 hour difference
+        result_pytz = naive.tz_localize(pytz_zone, ambiguous=True)
+        result_dateutil = naive.tz_localize(dateutil_zone, ambiguous=True)
+        assert result_pytz._value == result_dateutil._value
+        assert result_pytz._value == 1382832000
+
+        # see gh-14621
+        assert str(result_pytz) == str(result_dateutil)
+        assert (
+            result_pytz.to_pydatetime().tzname()
+            == result_dateutil.to_pydatetime().tzname()
+        )
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "pytz/US/Eastern",
+            gettz("US/Eastern"),
+            zoneinfo.ZoneInfo("US/Eastern"),
+            "dateutil/US/Eastern",
+        ],
+    )
+    def test_timestamp_tz_localize(self, tz):
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        stamp = Timestamp("3/11/2012 04:00")
+
+        result = stamp.tz_localize(tz)
+        expected = Timestamp("3/11/2012 04:00", tz=tz)
+        assert result.hour == expected.hour
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "start_ts, tz, end_ts, shift",
+        [
+            ["2015-03-29 02:20:00", "Europe/Warsaw", "2015-03-29 03:00:00", "forward"],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 01:59:59.999999999",
+                "backward",
+            ],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 03:20:00",
+                timedelta(hours=1),
+            ],
+            [
+                "2015-03-29 02:20:00",
+                "Europe/Warsaw",
+                "2015-03-29 01:20:00",
+                timedelta(hours=-1),
+            ],
+            ["2018-03-11 02:33:00", "US/Pacific", "2018-03-11 03:00:00", "forward"],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 01:59:59.999999999",
+                "backward",
+            ],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 03:33:00",
+                timedelta(hours=1),
+            ],
+            [
+                "2018-03-11 02:33:00",
+                "US/Pacific",
+                "2018-03-11 01:33:00",
+                timedelta(hours=-1),
+            ],
+        ],
+    )
+    @pytest.mark.parametrize("tz_type", ["", "dateutil/"])
+    def test_timestamp_tz_localize_nonexistent_shift(
+        self, start_ts, tz, end_ts, shift, tz_type, unit
+    ):
+        # GH 8917, 24466
+        tz = tz_type + tz
+        if isinstance(shift, str):
+            shift = "shift_" + shift
+        ts = Timestamp(start_ts).as_unit(unit)
+        result = ts.tz_localize(tz, nonexistent=shift)
+        expected = Timestamp(end_ts).tz_localize(tz)
+
+        if unit == "us":
+            assert result == expected.replace(nanosecond=0)
+        elif unit == "ms":
+            micros = expected.microsecond - expected.microsecond % 1000
+            assert result == expected.replace(microsecond=micros, nanosecond=0)
+        elif unit == "s":
+            assert result == expected.replace(microsecond=0, nanosecond=0)
+        else:
+            assert result == expected
+        assert result._creso == getattr(NpyDatetimeUnit, f"NPY_FR_{unit}").value
+
+    @pytest.mark.parametrize("offset", [-1, 1])
+    def test_timestamp_tz_localize_nonexistent_shift_invalid(self, offset, warsaw):
+        # GH 8917, 24466
+        tz = warsaw
+        ts = Timestamp("2015-03-29 02:20:00")
+        msg = "The provided timedelta will relocalize on a nonexistent time"
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize(tz, nonexistent=timedelta(seconds=offset))
+
+    def test_timestamp_tz_localize_nonexistent_NaT(self, warsaw, unit):
+        # GH 8917
+        tz = warsaw
+        ts = Timestamp("2015-03-29 02:20:00").as_unit(unit)
+        result = ts.tz_localize(tz, nonexistent="NaT")
+        assert result is NaT
+
+    def test_timestamp_tz_localize_nonexistent_raise(self, warsaw, unit):
+        # GH 8917
+        tz = warsaw
+        ts = Timestamp("2015-03-29 02:20:00").as_unit(unit)
+        msg = "2015-03-29 02:20:00"
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize(tz, nonexistent="raise")
+        msg = (
+            "The nonexistent argument must be one of 'raise', 'NaT', "
+            "'shift_forward', 'shift_backward' or a timedelta object"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ts.tz_localize(tz, nonexistent="foo")
diff --git a/pandas/pandas/tests/scalar/timestamp/test_arithmetic.py b/pandas/pandas/tests/scalar/timestamp/test_arithmetic.py
new file mode 100644
index 0000000000000..d65d425620c84
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_arithmetic.py
@@ -0,0 +1,346 @@
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+
+from dateutil.tz import gettz
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+    Timedelta,
+    Timestamp,
+    offsets,
+    to_offset,
+)
+
+import pandas._testing as tm
+
+
+class TestTimestampArithmetic:
+    def test_overflow_offset(self):
+        # no overflow expected
+
+        stamp = Timestamp("2000/1/1")
+        offset_no_overflow = to_offset("D") * 100
+
+        expected = Timestamp("2000/04/10")
+        assert stamp + offset_no_overflow == expected
+
+        assert offset_no_overflow + stamp == expected
+
+        expected = Timestamp("1999/09/23")
+        assert stamp - offset_no_overflow == expected
+
+    def test_overflow_offset_raises(self):
+        # xref https://github.com/statsmodels/statsmodels/issues/3374
+        # ends up multiplying really large numbers which overflow
+
+        stamp = Timestamp("2017-01-13 00:00:00").as_unit("ns")
+        offset_overflow = 20169940 * offsets.Day(1)
+        lmsg2 = r"Cannot cast -?20169940 days \+?00:00:00 to unit='ns' without overflow"
+
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg2):
+            stamp + offset_overflow
+
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg2):
+            offset_overflow + stamp
+
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg2):
+            stamp - offset_overflow
+
+        # xref https://github.com/pandas-dev/pandas/issues/14080
+        # used to crash, so check for proper overflow exception
+
+        stamp = Timestamp("2000/1/1").as_unit("ns")
+        offset_overflow = to_offset("D") * 100**5
+
+        lmsg3 = (
+            r"Cannot cast -?10000000000 days \+?00:00:00 to unit='ns' without overflow"
+        )
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg3):
+            stamp + offset_overflow
+
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg3):
+            offset_overflow + stamp
+
+        with pytest.raises(OutOfBoundsTimedelta, match=lmsg3):
+            stamp - offset_overflow
+
+    def test_overflow_timestamp_raises(self):
+        # https://github.com/pandas-dev/pandas/issues/31774
+        msg = "Result is too large"
+        a = Timestamp("2101-01-01 00:00:00").as_unit("ns")
+        b = Timestamp("1688-01-01 00:00:00").as_unit("ns")
+
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            a - b
+
+        # but we're OK for timestamp and datetime.datetime
+        assert (a - b.to_pydatetime()) == (a.to_pydatetime() - b)
+
+    def test_delta_preserve_nanos(self):
+        val = Timestamp(1337299200000000123)
+        result = val + timedelta(1)
+        assert result.nanosecond == val.nanosecond
+
+    def test_rsub_dtscalars(self, tz_naive_fixture):
+        # In particular, check that datetime64 - Timestamp works GH#28286
+        td = Timedelta(1235345642000)
+        ts = Timestamp("2021-01-01", tz=tz_naive_fixture)
+        other = ts + td
+
+        assert other - ts == td
+        assert other.to_pydatetime() - ts == td
+        if tz_naive_fixture is None:
+            assert other.to_datetime64() - ts == td
+        else:
+            msg = "Cannot subtract tz-naive and tz-aware datetime-like objects"
+            with pytest.raises(TypeError, match=msg):
+                other.to_datetime64() - ts
+
+    def test_timestamp_sub_datetime(self):
+        dt = datetime(2013, 10, 12)
+        ts = Timestamp(datetime(2013, 10, 13))
+        assert (ts - dt).days == 1
+        assert (dt - ts).days == -1
+
+    def test_subtract_tzaware_datetime(self):
+        t1 = Timestamp("2020-10-22T22:00:00+00:00")
+        t2 = datetime(2020, 10, 22, 22, tzinfo=timezone.utc)
+
+        result = t1 - t2
+
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta("0 days")
+
+    def test_subtract_timestamp_from_different_timezone(self):
+        t1 = Timestamp("20130101").tz_localize("US/Eastern")
+        t2 = Timestamp("20130101").tz_localize("CET")
+
+        result = t1 - t2
+
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta("0 days 06:00:00")
+
+    def test_subtracting_involving_datetime_with_different_tz(self):
+        t1 = datetime(2013, 1, 1, tzinfo=timezone(timedelta(hours=-5)))
+        t2 = Timestamp("20130101").tz_localize("CET")
+
+        result = t1 - t2
+
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta("0 days 06:00:00")
+
+        result = t2 - t1
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta("-1 days +18:00:00")
+
+    def test_subtracting_different_timezones(self, tz_aware_fixture):
+        t_raw = Timestamp("20130101")
+        t_UTC = t_raw.tz_localize("UTC")
+        t_diff = t_UTC.tz_convert(tz_aware_fixture) + Timedelta("0 days 05:00:00")
+
+        result = t_diff - t_UTC
+
+        assert isinstance(result, Timedelta)
+        assert result == Timedelta("0 days 05:00:00")
+
+    def test_addition_subtraction_types(self):
+        # Assert on the types resulting from Timestamp +/- various date/time
+        # objects
+        dt = datetime(2014, 3, 4)
+        td = timedelta(seconds=1)
+        ts = Timestamp(dt)
+
+        msg = "Addition/subtraction of integers"
+        with pytest.raises(TypeError, match=msg):
+            # GH#22535 add/sub with integers is deprecated
+            ts + 1
+        with pytest.raises(TypeError, match=msg):
+            ts - 1
+
+        # Timestamp + datetime not supported, though subtraction is supported
+        # and yields timedelta more tests in tseries/base/tests/test_base.py
+        assert type(ts - dt) == Timedelta
+        assert type(ts + td) == Timestamp
+        assert type(ts - td) == Timestamp
+
+        # Timestamp +/- datetime64 not supported, so not tested (could possibly
+        # assert error raised?)
+        td64 = np.timedelta64(1, "D")
+        assert type(ts + td64) == Timestamp
+        assert type(ts - td64) == Timestamp
+
+    @pytest.mark.parametrize(
+        "td", [Timedelta(hours=3), np.timedelta64(3, "h"), timedelta(hours=3)]
+    )
+    def test_radd_tdscalar(self, td, fixed_now_ts):
+        # GH#24775 timedelta64+Timestamp should not raise
+        ts = fixed_now_ts
+        assert td + ts == ts + td
+
+    @pytest.mark.parametrize(
+        "other,expected_difference",
+        [
+            (np.timedelta64(-123, "ns"), -123),
+            (np.timedelta64(1234567898, "ns"), 1234567898),
+            (np.timedelta64(-123, "us"), -123000),
+            (np.timedelta64(-123, "ms"), -123000000),
+        ],
+    )
+    def test_timestamp_add_timedelta64_unit(self, other, expected_difference):
+        now = datetime.now(timezone.utc)
+        ts = Timestamp(now).as_unit("ns")
+        result = ts + other
+        valdiff = result._value - ts._value
+        assert valdiff == expected_difference
+
+        ts2 = Timestamp(now)
+        assert ts2 + other == result
+
+    @pytest.mark.parametrize(
+        "ts",
+        [
+            Timestamp("1776-07-04"),
+            Timestamp("1776-07-04", tz="UTC"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "other",
+        [
+            1,
+            np.int64(1),
+            np.array([1, 2], dtype=np.int32),
+            np.array([3, 4], dtype=np.uint64),
+        ],
+    )
+    def test_add_int_with_freq(self, ts, other):
+        msg = "Addition/subtraction of integers and integer-arrays"
+        with pytest.raises(TypeError, match=msg):
+            ts + other
+        with pytest.raises(TypeError, match=msg):
+            other + ts
+
+        with pytest.raises(TypeError, match=msg):
+            ts - other
+
+        msg = "unsupported operand type"
+        with pytest.raises(TypeError, match=msg):
+            other - ts
+
+    @pytest.mark.parametrize("shape", [(6,), (2, 3)])
+    def test_addsub_m8ndarray(self, shape):
+        # GH#33296
+        ts = Timestamp("2020-04-04 15:45").as_unit("ns")
+        other = np.arange(6).astype("m8[h]").reshape(shape)
+
+        result = ts + other
+
+        ex_stamps = [ts + Timedelta(hours=n) for n in range(6)]
+        expected = np.array([x.asm8 for x in ex_stamps], dtype="M8[ns]").reshape(shape)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = other + ts
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = ts - other
+        ex_stamps = [ts - Timedelta(hours=n) for n in range(6)]
+        expected = np.array([x.asm8 for x in ex_stamps], dtype="M8[ns]").reshape(shape)
+        tm.assert_numpy_array_equal(result, expected)
+
+        msg = r"unsupported operand type\(s\) for -: 'numpy.ndarray' and 'Timestamp'"
+        with pytest.raises(TypeError, match=msg):
+            other - ts
+
+    @pytest.mark.parametrize("shape", [(6,), (2, 3)])
+    def test_addsub_m8ndarray_tzaware(self, shape):
+        # GH#33296
+        ts = Timestamp("2020-04-04 15:45", tz="US/Pacific")
+
+        other = np.arange(6).astype("m8[h]").reshape(shape)
+
+        result = ts + other
+
+        ex_stamps = [ts + Timedelta(hours=n) for n in range(6)]
+        expected = np.array(ex_stamps).reshape(shape)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = other + ts
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = ts - other
+        ex_stamps = [ts - Timedelta(hours=n) for n in range(6)]
+        expected = np.array(ex_stamps).reshape(shape)
+        tm.assert_numpy_array_equal(result, expected)
+
+        msg = r"unsupported operand type\(s\) for -: 'numpy.ndarray' and 'Timestamp'"
+        with pytest.raises(TypeError, match=msg):
+            other - ts
+
+    def test_subtract_different_utc_objects(self, utc_fixture, utc_fixture2):
+        # GH 32619
+        dt = datetime(2021, 1, 1)
+        ts1 = Timestamp(dt, tz=utc_fixture)
+        ts2 = Timestamp(dt, tz=utc_fixture2)
+        result = ts1 - ts2
+        expected = Timedelta(0)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "pytz/US/Eastern",
+            gettz("US/Eastern"),
+            "US/Eastern",
+            "dateutil/US/Eastern",
+        ],
+    )
+    def test_timestamp_add_timedelta_push_over_dst_boundary(self, tz):
+        # GH#1389
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        # 4 hours before DST transition
+        stamp = Timestamp("3/10/2012 22:00", tz=tz)
+
+        result = stamp + timedelta(hours=6)
+
+        # spring forward, + "7" hours
+        expected = Timestamp("3/11/2012 05:00", tz=tz)
+
+        assert result == expected
+
+    def test_timestamp_dst_transition(self):
+        # GH 60084
+        dt_str = "2023-11-05 01:00-08:00"
+        tz_str = "America/Los_Angeles"
+
+        ts1 = Timestamp(dt_str, tz=tz_str)
+        ts2 = ts1 + Timedelta(hours=0)
+
+        assert ts1 == ts2
+        assert hash(ts1) == hash(ts2)
+
+
+class SubDatetime(datetime):
+    pass
+
+
+@pytest.mark.parametrize(
+    "lh,rh",
+    [
+        (SubDatetime(2000, 1, 1), Timedelta(hours=1)),
+        (Timedelta(hours=1), SubDatetime(2000, 1, 1)),
+    ],
+)
+def test_dt_subclass_add_timedelta(lh, rh):
+    # GH#25851
+    # ensure that subclassed datetime works for
+    # Timedelta operations
+    result = lh + rh
+    expected = SubDatetime(2000, 1, 1, 1)
+    assert result == expected
diff --git a/pandas/pandas/tests/scalar/timestamp/test_comparisons.py b/pandas/pandas/tests/scalar/timestamp/test_comparisons.py
new file mode 100644
index 0000000000000..b2590c43e1ece
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_comparisons.py
@@ -0,0 +1,313 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+import operator
+
+import numpy as np
+import pytest
+
+from pandas import Timestamp
+import pandas._testing as tm
+
+
+class TestTimestampComparison:
+    def test_compare_non_nano_dt64(self):
+        # don't raise when converting dt64 to Timestamp in __richcmp__
+        dt = np.datetime64("1066-10-14")
+        ts = Timestamp(dt)
+
+        assert dt == ts
+
+    def test_comparison_dt64_ndarray(self):
+        ts = Timestamp("2021-01-01")
+        ts2 = Timestamp("2019-04-05")
+        arr = np.array([[ts.asm8, ts2.asm8]], dtype="M8[ns]")
+
+        result = ts == arr
+        expected = np.array([[True, False]], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = arr == ts
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = ts != arr
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        result = arr != ts
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        result = ts2 < arr
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = arr < ts2
+        tm.assert_numpy_array_equal(result, np.array([[False, False]], dtype=bool))
+
+        result = ts2 <= arr
+        tm.assert_numpy_array_equal(result, np.array([[True, True]], dtype=bool))
+
+        result = arr <= ts2
+        tm.assert_numpy_array_equal(result, ~expected)
+
+        result = ts >= arr
+        tm.assert_numpy_array_equal(result, np.array([[True, True]], dtype=bool))
+
+        result = arr >= ts
+        tm.assert_numpy_array_equal(result, np.array([[True, False]], dtype=bool))
+
+    @pytest.mark.parametrize("reverse", [True, False])
+    def test_comparison_dt64_ndarray_tzaware(self, reverse, comparison_op):
+        ts = Timestamp("2021-01-01 00:00:00.00000", tz="UTC")
+        arr = np.array([ts.asm8, ts.asm8], dtype="M8[ns]")
+
+        left, right = ts, arr
+        if reverse:
+            left, right = arr, ts
+
+        if comparison_op is operator.eq:
+            expected = np.array([False, False], dtype=bool)
+            result = comparison_op(left, right)
+            tm.assert_numpy_array_equal(result, expected)
+        elif comparison_op is operator.ne:
+            expected = np.array([True, True], dtype=bool)
+            result = comparison_op(left, right)
+            tm.assert_numpy_array_equal(result, expected)
+        else:
+            msg = "Cannot compare tz-naive and tz-aware timestamps"
+            with pytest.raises(TypeError, match=msg):
+                comparison_op(left, right)
+
+    def test_comparison_object_array(self):
+        # GH#15183
+        ts = Timestamp("2011-01-03 00:00:00-0500", tz="US/Eastern")
+        other = Timestamp("2011-01-01 00:00:00-0500", tz="US/Eastern")
+        naive = Timestamp("2011-01-01 00:00:00")
+
+        arr = np.array([other, ts], dtype=object)
+        res = arr == ts
+        expected = np.array([False, True], dtype=bool)
+        assert (res == expected).all()
+
+        # 2D case
+        arr = np.array([[other, ts], [ts, other]], dtype=object)
+        res = arr != ts
+        expected = np.array([[True, False], [False, True]], dtype=bool)
+        assert res.shape == expected.shape
+        assert (res == expected).all()
+
+        # tzaware mismatch
+        arr = np.array([naive], dtype=object)
+        msg = "Cannot compare tz-naive and tz-aware timestamps"
+        with pytest.raises(TypeError, match=msg):
+            arr < ts
+
+    def test_comparison(self):
+        # 5-18-2012 00:00:00.000
+        stamp = 1337299200000000000
+
+        val = Timestamp(stamp)
+
+        assert val == val
+        assert not val != val
+        assert not val < val
+        assert val <= val
+        assert not val > val
+        assert val >= val
+
+        other = datetime(2012, 5, 18)
+        assert val == other
+        assert not val != other
+        assert not val < other
+        assert val <= other
+        assert not val > other
+        assert val >= other
+
+        other = Timestamp(stamp + 100)
+
+        assert val != other
+        assert val != other
+        assert val < other
+        assert val <= other
+        assert other > val
+        assert other >= val
+
+    def test_compare_invalid(self):
+        # GH#8058
+        val = Timestamp("20130101 12:01:02")
+        assert not val == "foo"
+        assert not val == 10.0
+        assert not val == 1
+        assert not val == []
+        assert not val == {"foo": 1}
+        assert not val == np.float64(1)
+        assert not val == np.int64(1)
+
+        assert val != "foo"
+        assert val != 10.0
+        assert val != 1
+        assert val != []
+        assert val != {"foo": 1}
+        assert val != np.float64(1)
+        assert val != np.int64(1)
+
+    @pytest.mark.parametrize("tz", [None, "US/Pacific"])
+    def test_compare_date(self, tz):
+        # GH#36131 comparing Timestamp with date object is deprecated
+        ts = Timestamp("2021-01-01 00:00:00.00000", tz=tz)
+        dt = ts.to_pydatetime().date()
+        # in 2.0 we disallow comparing pydate objects with Timestamps,
+        #  following the stdlib datetime behavior.
+
+        msg = "Cannot compare Timestamp with datetime.date"
+        for left, right in [(ts, dt), (dt, ts)]:
+            assert not left == right
+            assert left != right
+
+            with pytest.raises(TypeError, match=msg):
+                left < right
+            with pytest.raises(TypeError, match=msg):
+                left <= right
+            with pytest.raises(TypeError, match=msg):
+                left > right
+            with pytest.raises(TypeError, match=msg):
+                left >= right
+
+    def test_cant_compare_tz_naive_w_aware(self, utc_fixture):
+        # see GH#1404
+        a = Timestamp("3/12/2012")
+        b = Timestamp("3/12/2012", tz=utc_fixture)
+
+        msg = "Cannot compare tz-naive and tz-aware timestamps"
+        assert not a == b
+        assert a != b
+        with pytest.raises(TypeError, match=msg):
+            a < b
+        with pytest.raises(TypeError, match=msg):
+            a <= b
+        with pytest.raises(TypeError, match=msg):
+            a > b
+        with pytest.raises(TypeError, match=msg):
+            a >= b
+
+        assert not b == a
+        assert b != a
+        with pytest.raises(TypeError, match=msg):
+            b < a
+        with pytest.raises(TypeError, match=msg):
+            b <= a
+        with pytest.raises(TypeError, match=msg):
+            b > a
+        with pytest.raises(TypeError, match=msg):
+            b >= a
+
+        assert not a == b.to_pydatetime()
+        assert not a.to_pydatetime() == b
+
+    def test_timestamp_compare_scalars(self):
+        # case where ndim == 0
+        lhs = np.datetime64(datetime(2013, 12, 6))
+        rhs = Timestamp("now")
+        nat = Timestamp("nat")
+
+        ops = {"gt": "lt", "lt": "gt", "ge": "le", "le": "ge", "eq": "eq", "ne": "ne"}
+
+        for left, right in ops.items():
+            left_f = getattr(operator, left)
+            right_f = getattr(operator, right)
+            expected = left_f(lhs, rhs)
+
+            result = right_f(rhs, lhs)
+            assert result == expected
+
+            expected = left_f(rhs, nat)
+            result = right_f(nat, rhs)
+            assert result == expected
+
+    def test_timestamp_compare_with_early_datetime(self):
+        # e.g. datetime.min
+        stamp = Timestamp("2012-01-01")
+
+        assert not stamp == datetime.min
+        assert not stamp == datetime(1600, 1, 1)
+        assert not stamp == datetime(2700, 1, 1)
+        assert stamp != datetime.min
+        assert stamp != datetime(1600, 1, 1)
+        assert stamp != datetime(2700, 1, 1)
+        assert stamp > datetime(1600, 1, 1)
+        assert stamp >= datetime(1600, 1, 1)
+        assert stamp < datetime(2700, 1, 1)
+        assert stamp <= datetime(2700, 1, 1)
+
+        other = Timestamp.min.to_pydatetime(warn=False)
+        assert other - timedelta(microseconds=1) < Timestamp.min
+
+    def test_timestamp_compare_oob_dt64(self):
+        us = np.timedelta64(1, "us")
+        other = np.datetime64(Timestamp.min).astype("M8[us]")
+
+        # This may change if the implementation bound is dropped to match
+        #  DatetimeArray/DatetimeIndex GH#24124
+        assert Timestamp.min > other
+        # Note: numpy gets the reversed comparison wrong
+
+        other = np.datetime64(Timestamp.max).astype("M8[us]")
+        assert Timestamp.max > other  # not actually OOB
+        assert other < Timestamp.max
+
+        assert Timestamp.max < other + us
+        # Note: numpy gets the reversed comparison wrong
+
+        # GH-42794
+        other = datetime(9999, 9, 9)
+        assert Timestamp.min < other
+        assert other > Timestamp.min
+        assert Timestamp.max < other
+        assert other > Timestamp.max
+
+        other = datetime(1, 1, 1)
+        assert Timestamp.max > other
+        assert other < Timestamp.max
+        assert Timestamp.min > other
+        assert other < Timestamp.min
+
+    def test_compare_zerodim_array(self, fixed_now_ts):
+        # GH#26916
+        ts = fixed_now_ts
+        dt64 = np.datetime64("2016-01-01", "ns")
+        arr = np.array(dt64)
+        assert arr.ndim == 0
+
+        result = arr < ts
+        assert result is np.bool_(True)
+        result = arr > ts
+        assert result is np.bool_(False)
+
+
+def test_rich_comparison_with_unsupported_type():
+    # Comparisons with unsupported objects should return NotImplemented
+    # (it previously raised TypeError, see #24011)
+
+    class Inf:
+        def __lt__(self, o):
+            return False
+
+        def __le__(self, o):
+            return isinstance(o, Inf)
+
+        def __gt__(self, o):
+            return not isinstance(o, Inf)
+
+        def __ge__(self, o):
+            return True
+
+        def __eq__(self, other) -> bool:
+            return isinstance(other, Inf)
+
+    inf = Inf()
+    timestamp = Timestamp("2018-11-30")
+
+    for left, right in [(inf, timestamp), (timestamp, inf)]:
+        assert left > right or left < right
+        assert left >= right or left <= right
+        assert not left == right
+        assert left != right
diff --git a/pandas/pandas/tests/scalar/timestamp/test_constructors.py b/pandas/pandas/tests/scalar/timestamp/test_constructors.py
new file mode 100644
index 0000000000000..2c97c4a32e0aa
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_constructors.py
@@ -0,0 +1,1086 @@
+import calendar
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+    timezone,
+)
+import zoneinfo
+
+import dateutil.tz
+from dateutil.tz import (
+    gettz,
+    tzoffset,
+    tzutc,
+)
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas.errors import OutOfBoundsDatetime
+
+from pandas import (
+    NA,
+    NaT,
+    Period,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestTimestampConstructorUnitKeyword:
+    @pytest.mark.parametrize("typ", [int, float])
+    def test_constructor_int_float_with_YM_unit(self, typ):
+        # GH#47266 avoid the conversions in cast_from_unit
+        val = typ(150)
+
+        ts = Timestamp(val, unit="Y")
+        expected = Timestamp("2120-01-01")
+        assert ts == expected
+
+        ts = Timestamp(val, unit="M")
+        expected = Timestamp("1982-07-01")
+        assert ts == expected
+
+    @pytest.mark.parametrize("typ", [int, float])
+    def test_construct_from_int_float_with_unit_out_of_bound_raises(self, typ):
+        # GH#50870  make sure we get a OutOfBoundsDatetime instead of OverflowError
+        val = typ(150000000000000)
+
+        msg = f"cannot convert input {val} with the unit 'D'"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp(val, unit="D")
+
+    def test_constructor_float_not_round_with_YM_unit_raises(self):
+        # GH#47267 avoid the conversions in cast_from-unit
+
+        msg = "Conversion of non-round float with unit=[MY] is ambiguous"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(150.5, unit="Y")
+
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(150.5, unit="M")
+
+    @pytest.mark.parametrize(
+        "value, check_kwargs",
+        [
+            [946688461000000000, {}],
+            [946688461000000000 / 1000, {"unit": "us"}],
+            [946688461000000000 / 1_000_000, {"unit": "ms"}],
+            [946688461000000000 / 1_000_000_000, {"unit": "s"}],
+            [10957, {"unit": "D", "h": 0}],
+            [
+                (946688461000000000 + 500000) / 1000000000,
+                {"unit": "s", "us": 499, "ns": 964},
+            ],
+            [
+                (946688461000000000 + 500000000) / 1000000000,
+                {"unit": "s", "us": 500000},
+            ],
+            [(946688461000000000 + 500000) / 1000000, {"unit": "ms", "us": 500}],
+            [(946688461000000000 + 500000) / 1000, {"unit": "us", "us": 500}],
+            [(946688461000000000 + 500000000) / 1000000, {"unit": "ms", "us": 500000}],
+            [946688461000000000 / 1000.0 + 5, {"unit": "us", "us": 5}],
+            [946688461000000000 / 1000.0 + 5000, {"unit": "us", "us": 5000}],
+            [946688461000000000 / 1000000.0 + 0.5, {"unit": "ms", "us": 500}],
+            [946688461000000000 / 1000000.0 + 0.005, {"unit": "ms", "us": 5, "ns": 5}],
+            [946688461000000000 / 1000000000.0 + 0.5, {"unit": "s", "us": 500000}],
+            [10957 + 0.5, {"unit": "D", "h": 12}],
+        ],
+    )
+    def test_construct_with_unit(self, value, check_kwargs):
+        def check(value, unit=None, h=1, s=1, us=0, ns=0):
+            stamp = Timestamp(value, unit=unit)
+            assert stamp.year == 2000
+            assert stamp.month == 1
+            assert stamp.day == 1
+            assert stamp.hour == h
+            if unit != "D":
+                assert stamp.minute == 1
+                assert stamp.second == s
+                assert stamp.microsecond == us
+            else:
+                assert stamp.minute == 0
+                assert stamp.second == 0
+                assert stamp.microsecond == 0
+            assert stamp.nanosecond == ns
+
+        check(value, **check_kwargs)
+
+
+class TestTimestampConstructorFoldKeyword:
+    def test_timestamp_constructor_invalid_fold_raise(self):
+        # Test for GH#25057
+        # Valid fold values are only [None, 0, 1]
+        msg = "Valid values for the fold argument are None, 0, or 1."
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(123, fold=2)
+
+    def test_timestamp_constructor_pytz_fold_raise(self):
+        # Test for GH#25057
+        # pytz doesn't support fold. Check that we raise
+        # if fold is passed with pytz
+        pytz = pytest.importorskip("pytz")
+        msg = "pytz timezones do not support fold. Please use dateutil timezones."
+        tz = pytz.timezone("Europe/London")
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(datetime(2019, 10, 27, 0, 30, 0, 0), tz=tz, fold=0)
+
+    @pytest.mark.parametrize("fold", [0, 1])
+    @pytest.mark.parametrize(
+        "ts_input",
+        [
+            1572136200000000000,
+            1572136200000000000.0,
+            np.datetime64(1572136200000000000, "ns"),
+            "2019-10-27 01:30:00+01:00",
+            datetime(2019, 10, 27, 0, 30, 0, 0, tzinfo=timezone.utc),
+        ],
+    )
+    def test_timestamp_constructor_fold_conflict(self, ts_input, fold):
+        # Test for GH#25057
+        # Check that we raise on fold conflict
+        msg = (
+            "Cannot pass fold with possibly unambiguous input: int, float, "
+            "numpy.datetime64, str, or timezone-aware datetime-like. "
+            "Pass naive datetime-like or build Timestamp from components."
+        )
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(ts_input=ts_input, fold=fold)
+
+    @pytest.mark.parametrize("tz", ["dateutil/Europe/London", None])
+    @pytest.mark.parametrize("fold", [0, 1])
+    def test_timestamp_constructor_retain_fold(self, tz, fold):
+        # Test for GH#25057
+        # Check that we retain fold
+        ts = Timestamp(year=2019, month=10, day=27, hour=1, minute=30, tz=tz, fold=fold)
+        result = ts.fold
+        expected = fold
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "dateutil/Europe/London",
+            zoneinfo.ZoneInfo("Europe/London"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "ts_input,fold_out",
+        [
+            (1572136200000000000, 0),
+            (1572139800000000000, 1),
+            ("2019-10-27 01:30:00+01:00", 0),
+            ("2019-10-27 01:30:00+00:00", 1),
+            (datetime(2019, 10, 27, 1, 30, 0, 0, fold=0), 0),
+            (datetime(2019, 10, 27, 1, 30, 0, 0, fold=1), 1),
+        ],
+    )
+    def test_timestamp_constructor_infer_fold_from_value(self, tz, ts_input, fold_out):
+        # Test for GH#25057
+        # Check that we infer fold correctly based on timestamps since utc
+        # or strings
+        ts = Timestamp(ts_input, tz=tz)
+        result = ts.fold
+        expected = fold_out
+        assert result == expected
+
+    @pytest.mark.parametrize("tz", ["dateutil/Europe/London"])
+    @pytest.mark.parametrize(
+        "fold,value_out",
+        [
+            (0, 1572136200000000),
+            (1, 1572139800000000),
+        ],
+    )
+    def test_timestamp_constructor_adjust_value_for_fold(self, tz, fold, value_out):
+        # Test for GH#25057
+        # Check that we adjust value for fold correctly
+        # based on timestamps since utc
+        ts_input = datetime(2019, 10, 27, 1, 30)
+        ts = Timestamp(ts_input, tz=tz, fold=fold)
+        result = ts._value
+        expected = value_out
+        assert result == expected
+
+
+class TestTimestampConstructorPositionalAndKeywordSupport:
+    def test_constructor_positional(self):
+        # see GH#10758
+        msg = "'NoneType' object cannot be interpreted as an integer"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp(2000, 1)
+
+        msg = "month must be in 1..12"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(2000, 0, 1)
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(2000, 13, 1)
+
+        msg = "day is out of range for month"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(2000, 1, 0)
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(2000, 1, 32)
+
+        # see gh-11630
+        assert repr(Timestamp(2015, 11, 12)) == repr(Timestamp("20151112"))
+        assert repr(Timestamp(2015, 11, 12, 1, 2, 3, 999999)) == repr(
+            Timestamp("2015-11-12 01:02:03.999999")
+        )
+
+    def test_constructor_keyword(self):
+        # GH#10758
+        msg = "function missing required argument 'day'|Required argument 'day'"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp(year=2000, month=1)
+
+        msg = "month must be in 1..12"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(year=2000, month=0, day=1)
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(year=2000, month=13, day=1)
+
+        msg = "day is out of range for month"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(year=2000, month=1, day=0)
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(year=2000, month=1, day=32)
+
+        assert repr(Timestamp(year=2015, month=11, day=12)) == repr(
+            Timestamp("20151112")
+        )
+
+        assert repr(
+            Timestamp(
+                year=2015,
+                month=11,
+                day=12,
+                hour=1,
+                minute=2,
+                second=3,
+                microsecond=999999,
+            )
+        ) == repr(Timestamp("2015-11-12 01:02:03.999999"))
+
+    @pytest.mark.parametrize(
+        "arg",
+        [
+            "year",
+            "month",
+            "day",
+            "hour",
+            "minute",
+            "second",
+            "microsecond",
+            "nanosecond",
+        ],
+    )
+    def test_invalid_date_kwarg_with_string_input(self, arg):
+        kwarg = {arg: 1}
+        msg = "Cannot pass a date attribute keyword argument"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2010-10-10 12:59:59.999999999", **kwarg)
+
+    @pytest.mark.parametrize("kwargs", [{}, {"year": 2020}, {"year": 2020, "month": 1}])
+    def test_constructor_missing_keyword(self, kwargs):
+        # GH#31200
+
+        # The exact error message of datetime() depends on its version
+        msg1 = r"function missing required argument '(year|month|day)' \(pos [123]\)"
+        msg2 = r"Required argument '(year|month|day)' \(pos [123]\) not found"
+        msg = "|".join([msg1, msg2])
+
+        with pytest.raises(TypeError, match=msg):
+            Timestamp(**kwargs)
+
+    def test_constructor_positional_with_tzinfo(self):
+        # GH#31929
+        ts = Timestamp(2020, 12, 31, tzinfo=timezone.utc)
+        expected = Timestamp("2020-12-31", tzinfo=timezone.utc)
+        assert ts == expected
+
+    @pytest.mark.parametrize("kwd", ["nanosecond", "microsecond", "second", "minute"])
+    def test_constructor_positional_keyword_mixed_with_tzinfo(self, kwd, request):
+        # TODO: if we passed microsecond with a keyword we would mess up
+        #  xref GH#45307
+        if kwd != "nanosecond":
+            # nanosecond is keyword-only as of 2.0, others are not
+            mark = pytest.mark.xfail(reason="GH#45307")
+            request.applymarker(mark)
+
+        kwargs = {kwd: 4}
+        ts = Timestamp(2020, 12, 31, tzinfo=timezone.utc, **kwargs)
+
+        td_kwargs = {kwd + "s": 4}
+        td = Timedelta(**td_kwargs)
+        expected = Timestamp("2020-12-31", tz=timezone.utc) + td
+        assert ts == expected
+
+
+class TestTimestampClassMethodConstructors:
+    # Timestamp constructors other than __new__
+
+    def test_utcnow_deprecated(self):
+        # GH#56680
+        msg = "Timestamp.utcnow is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            Timestamp.utcnow()
+
+    def test_utcfromtimestamp_deprecated(self):
+        # GH#56680
+        msg = "Timestamp.utcfromtimestamp is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            Timestamp.utcfromtimestamp(43)
+
+    def test_constructor_strptime(self):
+        # GH#25016
+        # Test support for Timestamp.strptime
+        fmt = "%Y%m%d-%H%M%S-%f%z"
+        ts = "20190129-235348-000001+0000"
+        msg = r"Timestamp.strptime\(\) is not implemented"
+        with pytest.raises(NotImplementedError, match=msg):
+            Timestamp.strptime(ts, fmt)
+
+    def test_constructor_fromisocalendar(self):
+        # GH#30395
+        expected_timestamp = Timestamp("2000-01-03 00:00:00")
+        expected_stdlib = datetime.fromisocalendar(2000, 1, 1)
+        result = Timestamp.fromisocalendar(2000, 1, 1)
+        assert result == expected_timestamp
+        assert result == expected_stdlib
+        assert isinstance(result, Timestamp)
+
+    def test_constructor_fromordinal(self):
+        base = datetime(2000, 1, 1)
+
+        ts = Timestamp.fromordinal(base.toordinal())
+        assert base == ts
+        assert base.toordinal() == ts.toordinal()
+
+        ts = Timestamp.fromordinal(base.toordinal(), tz="US/Eastern")
+        assert Timestamp("2000-01-01", tz="US/Eastern") == ts
+        assert base.toordinal() == ts.toordinal()
+
+        # GH#3042
+        dt = datetime(2011, 4, 16, 0, 0)
+        ts = Timestamp.fromordinal(dt.toordinal())
+        assert ts.to_pydatetime() == dt
+
+        # with a tzinfo
+        stamp = Timestamp("2011-4-16", tz="US/Eastern")
+        dt_tz = stamp.to_pydatetime()
+        ts = Timestamp.fromordinal(dt_tz.toordinal(), tz="US/Eastern")
+        assert ts.to_pydatetime() == dt_tz
+
+    def test_now(self):
+        # GH#9000
+        ts_from_string = Timestamp("now")
+        ts_from_method = Timestamp.now()
+        ts_datetime = datetime.now()
+
+        ts_from_string_tz = Timestamp("now", tz="US/Eastern")
+        ts_from_method_tz = Timestamp.now(tz="US/Eastern")
+
+        # Check that the delta between the times is less than 1s (arbitrarily
+        # small)
+        delta = Timedelta(seconds=1)
+        assert abs(ts_from_method - ts_from_string) < delta
+        assert abs(ts_datetime - ts_from_method) < delta
+        assert abs(ts_from_method_tz - ts_from_string_tz) < delta
+        assert (
+            abs(
+                ts_from_string_tz.tz_localize(None)
+                - ts_from_method_tz.tz_localize(None)
+            )
+            < delta
+        )
+
+    def test_today(self):
+        ts_from_string = Timestamp("today")
+        ts_from_method = Timestamp.today()
+        ts_datetime = datetime.today()
+
+        ts_from_string_tz = Timestamp("today", tz="US/Eastern")
+        ts_from_method_tz = Timestamp.today(tz="US/Eastern")
+
+        # Check that the delta between the times is less than 1s (arbitrarily
+        # small)
+        delta = Timedelta(seconds=1)
+        assert abs(ts_from_method - ts_from_string) < delta
+        assert abs(ts_datetime - ts_from_method) < delta
+        assert abs(ts_from_method_tz - ts_from_string_tz) < delta
+        assert (
+            abs(
+                ts_from_string_tz.tz_localize(None)
+                - ts_from_method_tz.tz_localize(None)
+            )
+            < delta
+        )
+
+
+class TestTimestampResolutionInference:
+    def test_construct_from_time_unit(self):
+        # GH#54097 only passing a time component, no date
+        ts = Timestamp("01:01:01.111")
+        assert ts.unit == "ms"
+
+    def test_constructor_str_infer_reso(self):
+        # non-iso8601 path
+
+        # _parse_delimited_date path
+        ts = Timestamp("01/30/2023")
+        assert ts.unit == "s"
+
+        # _parse_dateabbr_string path
+        ts = Timestamp("2015Q1")
+        assert ts.unit == "s"
+
+        # dateutil_parse path
+        ts = Timestamp("2016-01-01 1:30:01 PM")
+        assert ts.unit == "s"
+
+        ts = Timestamp("2016 June 3 15:25:01.345")
+        assert ts.unit == "ms"
+
+        ts = Timestamp("300-01-01")
+        assert ts.unit == "s"
+
+        ts = Timestamp("300 June 1:30:01.300")
+        assert ts.unit == "ms"
+
+        # dateutil path -> don't drop trailing zeros
+        ts = Timestamp("01-01-2013T00:00:00.000000000+0000")
+        assert ts.unit == "ns"
+
+        ts = Timestamp("2016/01/02 03:04:05.001000 UTC")
+        assert ts.unit == "us"
+
+        # higher-than-nanosecond -> we drop the trailing bits
+        ts = Timestamp("01-01-2013T00:00:00.000000002100+0000")
+        assert ts == Timestamp("01-01-2013T00:00:00.000000002+0000")
+        assert ts.unit == "ns"
+
+        # GH#56208 minute reso through the ISO8601 path with tz offset
+        ts = Timestamp("2020-01-01 00:00+00:00")
+        assert ts.unit == "s"
+
+        ts = Timestamp("2020-01-01 00+00:00")
+        assert ts.unit == "s"
+
+    @pytest.mark.parametrize("method", ["now", "today"])
+    def test_now_today_unit(self, method):
+        # GH#55879
+        ts_from_method = getattr(Timestamp, method)()
+        ts_from_string = Timestamp(method)
+        assert ts_from_method.unit == ts_from_string.unit == "us"
+
+
+class TestTimestampConstructors:
+    def test_weekday_but_no_day_raises(self):
+        # GH#52659
+        msg = "Parsing datetimes with weekday but no day information is not supported"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2023 Sept Thu")
+
+    def test_construct_from_string_invalid_raises(self):
+        # dateutil (weirdly) parses "200622-12-31" as
+        #  datetime(2022, 6, 20, 12, 0, tzinfo=tzoffset(None, -111600)
+        #  which besides being mis-parsed, is a tzoffset that will cause
+        #  str(ts) to raise ValueError.  Ensure we raise in the constructor
+        #  instead.
+        # see test_to_datetime_malformed_raise for analogous to_datetime test
+        with pytest.raises(ValueError, match="gives an invalid tzoffset"):
+            Timestamp("200622-12-31")
+
+    def test_constructor_from_iso8601_str_with_offset_reso(self):
+        # GH#49737
+        ts = Timestamp("2016-01-01 04:05:06-01:00")
+        assert ts.unit == "s"
+
+        ts = Timestamp("2016-01-01 04:05:06.000-01:00")
+        assert ts.unit == "ms"
+
+        ts = Timestamp("2016-01-01 04:05:06.000000-01:00")
+        assert ts.unit == "us"
+
+        ts = Timestamp("2016-01-01 04:05:06.000000001-01:00")
+        assert ts.unit == "ns"
+
+    def test_constructor_from_date_second_reso(self):
+        # GH#49034 constructing from a pydate object gets lowest supported
+        #  reso, i.e. seconds
+        obj = date(2012, 9, 1)
+        ts = Timestamp(obj)
+        assert ts.unit == "s"
+
+    def test_constructor_datetime64_with_tz(self):
+        # GH#42288, GH#24559
+        dt = np.datetime64("1970-01-01 05:00:00")
+        tzstr = "UTC+05:00"
+
+        # pre-2.0 this interpreted dt as a UTC time. in 2.0 this is treated
+        #  as a wall-time, consistent with DatetimeIndex behavior
+        ts = Timestamp(dt, tz=tzstr)
+
+        alt = Timestamp(dt).tz_localize(tzstr)
+        assert ts == alt
+        assert ts.hour == 5
+
+    def test_constructor(self):
+        base_str = "2014-07-01 09:00"
+        base_dt = datetime(2014, 7, 1, 9)
+        base_expected = 1_404_205_200_000_000_000
+
+        # confirm base representation is correct
+        assert calendar.timegm(base_dt.timetuple()) * 1_000_000_000 == base_expected
+
+        tests = [
+            (base_str, base_dt, base_expected),
+            (
+                "2014-07-01 10:00",
+                datetime(2014, 7, 1, 10),
+                base_expected + 3600 * 1_000_000_000,
+            ),
+            (
+                "2014-07-01 09:00:00.000008000",
+                datetime(2014, 7, 1, 9, 0, 0, 8),
+                base_expected + 8000,
+            ),
+            (
+                "2014-07-01 09:00:00.000000005",
+                Timestamp("2014-07-01 09:00:00.000000005"),
+                base_expected + 5,
+            ),
+        ]
+
+        timezones = [
+            (None, 0),
+            ("UTC", 0),
+            (timezone.utc, 0),
+            ("Asia/Tokyo", 9),
+            ("US/Eastern", -4),
+            ("dateutil/US/Pacific", -7),
+            (timezone(timedelta(hours=-3)), -3),
+            (dateutil.tz.tzoffset(None, 18000), 5),
+        ]
+
+        for date_str, date_obj, expected in tests:
+            for result in [Timestamp(date_str), Timestamp(date_obj)]:
+                result = result.as_unit("ns")  # test originally written before non-nano
+                # only with timestring
+                assert result.as_unit("ns")._value == expected
+
+                # re-creation shouldn't affect to internal value
+                result = Timestamp(result)
+                assert result.as_unit("ns")._value == expected
+
+            # with timezone
+            for tz, offset in timezones:
+                for result in [Timestamp(date_str, tz=tz), Timestamp(date_obj, tz=tz)]:
+                    result = result.as_unit(
+                        "ns"
+                    )  # test originally written before non-nano
+                    expected_tz = expected - offset * 3600 * 1_000_000_000
+                    assert result.as_unit("ns")._value == expected_tz
+
+                    # should preserve tz
+                    result = Timestamp(result)
+                    assert result.as_unit("ns")._value == expected_tz
+
+                    # should convert to UTC
+                    if tz is not None:
+                        result = Timestamp(result).tz_convert("UTC")
+                    else:
+                        result = Timestamp(result, tz="UTC")
+                    expected_utc = expected - offset * 3600 * 1_000_000_000
+                    assert result.as_unit("ns")._value == expected_utc
+
+    def test_constructor_with_stringoffset(self):
+        # GH 7833
+        base_str = "2014-07-01 11:00:00+02:00"
+        base_dt = datetime(2014, 7, 1, 9)
+        base_expected = 1_404_205_200_000_000_000
+
+        # confirm base representation is correct
+        assert calendar.timegm(base_dt.timetuple()) * 1_000_000_000 == base_expected
+
+        tests = [
+            (base_str, base_expected),
+            ("2014-07-01 12:00:00+02:00", base_expected + 3600 * 1_000_000_000),
+            ("2014-07-01 11:00:00.000008000+02:00", base_expected + 8000),
+            ("2014-07-01 11:00:00.000000005+02:00", base_expected + 5),
+        ]
+
+        timezones = [
+            ("UTC", 0),
+            (timezone.utc, 0),
+            ("Asia/Tokyo", 9),
+            ("US/Eastern", -4),
+            ("dateutil/US/Pacific", -7),
+            (timezone(timedelta(hours=-3)), -3),
+            (dateutil.tz.tzoffset(None, 18000), 5),
+        ]
+
+        for date_str, expected in tests:
+            for result in [Timestamp(date_str)]:
+                # only with timestring
+                assert result.as_unit("ns")._value == expected
+
+                # re-creation shouldn't affect to internal value
+                result = Timestamp(result)
+                assert result.as_unit("ns")._value == expected
+
+            # with timezone
+            for tz, offset in timezones:
+                result = Timestamp(date_str, tz=tz)
+                expected_tz = expected
+                assert result.as_unit("ns")._value == expected_tz
+
+                # should preserve tz
+                result = Timestamp(result)
+                assert result.as_unit("ns")._value == expected_tz
+
+                # should convert to UTC
+                result = Timestamp(result).tz_convert("UTC")
+                expected_utc = expected
+                assert result.as_unit("ns")._value == expected_utc
+
+        # This should be 2013-11-01 05:00 in UTC
+        # converted to Chicago tz
+        result = Timestamp("2013-11-01 00:00:00-0500", tz="America/Chicago")
+        assert result._value == Timestamp("2013-11-01 05:00")._value
+        expected = "Timestamp('2013-11-01 00:00:00-0500', tz='America/Chicago')"
+        assert repr(result) == expected
+        assert result == eval(repr(result))
+
+        # This should be 2013-11-01 05:00 in UTC
+        # converted to Tokyo tz (+09:00)
+        result = Timestamp("2013-11-01 00:00:00-0500", tz="Asia/Tokyo")
+        assert result._value == Timestamp("2013-11-01 05:00")._value
+        expected = "Timestamp('2013-11-01 14:00:00+0900', tz='Asia/Tokyo')"
+        assert repr(result) == expected
+        assert result == eval(repr(result))
+
+        # GH11708
+        # This should be 2015-11-18 10:00 in UTC
+        # converted to Asia/Katmandu
+        result = Timestamp("2015-11-18 15:45:00+05:45", tz="Asia/Katmandu")
+        assert result._value == Timestamp("2015-11-18 10:00")._value
+        expected = "Timestamp('2015-11-18 15:45:00+0545', tz='Asia/Katmandu')"
+        assert repr(result) == expected
+        assert result == eval(repr(result))
+
+        # This should be 2015-11-18 10:00 in UTC
+        # converted to Asia/Kolkata
+        result = Timestamp("2015-11-18 15:30:00+05:30", tz="Asia/Kolkata")
+        assert result._value == Timestamp("2015-11-18 10:00")._value
+        expected = "Timestamp('2015-11-18 15:30:00+0530', tz='Asia/Kolkata')"
+        assert repr(result) == expected
+        assert result == eval(repr(result))
+
+    def test_constructor_invalid(self):
+        msg = "Cannot convert input"
+        with pytest.raises(TypeError, match=msg):
+            Timestamp(slice(2))
+        msg = "Cannot convert Period"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(Period("1000-01-01"))
+
+    def test_constructor_invalid_tz(self):
+        # GH#17690
+        msg = (
+            "Argument 'tzinfo' has incorrect type "
+            r"\(expected datetime.tzinfo, got str\)"
+        )
+        with pytest.raises(TypeError, match=msg):
+            Timestamp("2017-10-22", tzinfo="US/Eastern")
+
+        msg = "at most one of"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2017-10-22", tzinfo=timezone.utc, tz="UTC")
+
+        msg = "Cannot pass a date attribute keyword argument when passing a date string"
+        with pytest.raises(ValueError, match=msg):
+            # GH#5168
+            # case where user tries to pass tz as an arg, not kwarg, gets
+            # interpreted as `year`
+            Timestamp("2012-01-01", "US/Pacific")
+
+    def test_constructor_tz_or_tzinfo(self):
+        # GH#17943, GH#17690, GH#5168
+        stamps = [
+            Timestamp(year=2017, month=10, day=22, tz="UTC"),
+            Timestamp(year=2017, month=10, day=22, tzinfo=timezone.utc),
+            Timestamp(year=2017, month=10, day=22, tz=timezone.utc),
+            Timestamp(datetime(2017, 10, 22), tzinfo=timezone.utc),
+            Timestamp(datetime(2017, 10, 22), tz="UTC"),
+            Timestamp(datetime(2017, 10, 22), tz=timezone.utc),
+        ]
+        assert all(ts == stamps[0] for ts in stamps)
+
+    @pytest.mark.parametrize(
+        "result",
+        [
+            Timestamp(datetime(2000, 1, 2, 3, 4, 5, 6), nanosecond=1),
+            Timestamp(
+                year=2000,
+                month=1,
+                day=2,
+                hour=3,
+                minute=4,
+                second=5,
+                microsecond=6,
+                nanosecond=1,
+            ),
+            Timestamp(
+                year=2000,
+                month=1,
+                day=2,
+                hour=3,
+                minute=4,
+                second=5,
+                microsecond=6,
+                nanosecond=1,
+                tz="UTC",
+            ),
+            Timestamp(2000, 1, 2, 3, 4, 5, 6, None, nanosecond=1),
+            Timestamp(2000, 1, 2, 3, 4, 5, 6, tz=timezone.utc, nanosecond=1),
+        ],
+    )
+    def test_constructor_nanosecond(self, result):
+        # GH 18898
+        # As of 2.0 (GH 49416), nanosecond should not be accepted positionally
+        expected = Timestamp(datetime(2000, 1, 2, 3, 4, 5, 6), tz=result.tz)
+        expected = expected + Timedelta(nanoseconds=1)
+        assert result == expected
+
+    @pytest.mark.parametrize("z", ["Z0", "Z00"])
+    def test_constructor_invalid_Z0_isostring(self, z):
+        # GH 8910
+        msg = f"Unknown datetime string format, unable to parse: 2014-11-02 01:00{z}"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(f"2014-11-02 01:00{z}")
+
+    def test_out_of_bounds_integer_value(self):
+        # GH#26651 check that we raise OutOfBoundsDatetime, not OverflowError
+        msg = str(Timestamp.max._value * 2)
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp(Timestamp.max._value * 2)
+        msg = str(Timestamp.min._value * 2)
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp(Timestamp.min._value * 2)
+
+    def test_out_of_bounds_value(self):
+        one_us = np.timedelta64(1).astype("timedelta64[us]")
+
+        # By definition we can't go out of bounds in [ns], so we
+        # convert the datetime64s to [us] so we can go out of bounds
+        min_ts_us = np.datetime64(Timestamp.min).astype("M8[us]") + one_us
+        max_ts_us = np.datetime64(Timestamp.max).astype("M8[us]")
+
+        # No error for the min/max datetimes
+        Timestamp(min_ts_us)
+        Timestamp(max_ts_us)
+
+        # We used to raise on these before supporting non-nano
+        us_val = NpyDatetimeUnit.NPY_FR_us.value
+        assert Timestamp(min_ts_us - one_us)._creso == us_val
+        assert Timestamp(max_ts_us + one_us)._creso == us_val
+
+        # https://github.com/numpy/numpy/issues/22346 for why
+        #  we can't use the same construction as above with minute resolution
+
+        # too_low, too_high are the _just_ outside the range of M8[s]
+        too_low = np.datetime64("-292277022657-01-27T08:29", "m")
+        too_high = np.datetime64("292277026596-12-04T15:31", "m")
+
+        msg = "Out of bounds"
+        # One us less than the minimum is an error
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(too_low)
+
+        # One us more than the maximum is an error
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(too_high)
+
+    def test_out_of_bounds_string(self):
+        msg = "Cannot cast .* to unit='ns' without overflow"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("1676-01-01").as_unit("ns")
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2263-01-01").as_unit("ns")
+
+        ts = Timestamp("2263-01-01")
+        assert ts.unit == "s"
+
+        ts = Timestamp("1676-01-01")
+        assert ts.unit == "s"
+
+    def test_barely_out_of_bounds(self):
+        # GH#19529
+        # GH#19382 close enough to bounds that dropping nanos would result
+        # in an in-bounds datetime
+        msg = "Out of bounds nanosecond timestamp: 2262-04-11 23:47:16"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp("2262-04-11 23:47:16.854775808")
+
+    @pytest.mark.skip_ubsan
+    def test_bounds_with_different_units(self):
+        out_of_bounds_dates = ("1677-09-21", "2262-04-12")
+
+        time_units = ("D", "h", "m", "s", "ms", "us")
+
+        for date_string in out_of_bounds_dates:
+            for unit in time_units:
+                dt64 = np.datetime64(date_string, unit)
+                ts = Timestamp(dt64)
+                if unit in ["s", "ms", "us"]:
+                    # We can preserve the input unit
+                    assert ts._value == dt64.view("i8")
+                else:
+                    # we chose the closest unit that we _do_ support
+                    assert ts._creso == NpyDatetimeUnit.NPY_FR_s.value
+
+        # With more extreme cases, we can't even fit inside second resolution
+        info = np.iinfo(np.int64)
+        msg = "Out of bounds second timestamp:"
+        for value in [info.min + 1, info.max]:
+            for unit in ["D", "h", "m"]:
+                dt64 = np.datetime64(value, unit)
+                with pytest.raises(OutOfBoundsDatetime, match=msg):
+                    Timestamp(dt64)
+
+        in_bounds_dates = ("1677-09-23", "2262-04-11")
+
+        for date_string in in_bounds_dates:
+            for unit in time_units:
+                dt64 = np.datetime64(date_string, unit)
+                Timestamp(dt64)
+
+    @pytest.mark.parametrize("arg", ["001-01-01", "0001-01-01"])
+    def test_out_of_bounds_string_consistency(self, arg):
+        # GH 15829
+        msg = "Cannot cast 0001-01-01 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp(arg).as_unit("ns")
+
+        ts = Timestamp(arg)
+        assert ts.unit == "s"
+        assert ts.year == ts.month == ts.day == 1
+
+    def test_min_valid(self):
+        # Ensure that Timestamp.min is a valid Timestamp
+        Timestamp(Timestamp.min)
+
+    def test_max_valid(self):
+        # Ensure that Timestamp.max is a valid Timestamp
+        Timestamp(Timestamp.max)
+
+    @pytest.mark.parametrize("offset", ["+0300", "+0200"])
+    def test_construct_timestamp_near_dst(self, offset):
+        # GH 20854
+        expected = Timestamp(f"2016-10-30 03:00:00{offset}", tz="Europe/Helsinki")
+        result = Timestamp(expected).tz_convert("Europe/Helsinki")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "arg", ["2013/01/01 00:00:00+09:00", "2013-01-01 00:00:00+09:00"]
+    )
+    def test_construct_with_different_string_format(self, arg):
+        # GH 12064
+        result = Timestamp(arg)
+        expected = Timestamp(datetime(2013, 1, 1), tz=timezone(timedelta(hours=9)))
+        assert result == expected
+
+    @pytest.mark.parametrize("box", [datetime, Timestamp])
+    def test_raise_tz_and_tzinfo_in_datetime_input(self, box):
+        # GH 23579
+        kwargs = {"year": 2018, "month": 1, "day": 1, "tzinfo": timezone.utc}
+        msg = "Cannot pass a datetime or Timestamp"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(box(**kwargs), tz="US/Pacific")
+        msg = "Cannot pass a datetime or Timestamp"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(box(**kwargs), tzinfo=zoneinfo.ZoneInfo("US/Pacific"))
+
+    def test_dont_convert_dateutil_utc_to_default_utc(self):
+        result = Timestamp(datetime(2018, 1, 1), tz=tzutc())
+        expected = Timestamp(datetime(2018, 1, 1)).tz_localize(tzutc())
+        assert result == expected
+
+    def test_constructor_subclassed_datetime(self):
+        # GH 25851
+        # ensure that subclassed datetime works for
+        # Timestamp creation
+        class SubDatetime(datetime):
+            pass
+
+        data = SubDatetime(2000, 1, 1)
+        result = Timestamp(data)
+        expected = Timestamp(2000, 1, 1)
+        assert result == expected
+
+    def test_timestamp_constructor_tz_utc(self):
+        utc_stamp = Timestamp("3/11/2012 05:00", tz="utc")
+        assert utc_stamp.tzinfo is timezone.utc
+        assert utc_stamp.hour == 5
+
+        utc_stamp = Timestamp("3/11/2012 05:00").tz_localize("utc")
+        assert utc_stamp.hour == 5
+
+    def test_timestamp_to_datetime_tzoffset(self):
+        tzinfo = tzoffset(None, 7200)
+        expected = Timestamp("3/11/2012 04:00", tz=tzinfo)
+        result = Timestamp(expected.to_pydatetime())
+        assert expected == result
+
+    def test_timestamp_constructor_near_dst_boundary(self):
+        # GH#11481 & GH#15777
+        # Naive string timestamps were being localized incorrectly
+        # with tz_convert_from_utc_single instead of tz_localize_to_utc
+
+        for tz in ["Europe/Brussels", "Europe/Prague"]:
+            result = Timestamp("2015-10-25 01:00", tz=tz)
+            expected = Timestamp("2015-10-25 01:00").tz_localize(tz)
+            assert result == expected
+
+            msg = "Cannot infer dst time from 2015-10-25 02:00:00"
+            with pytest.raises(ValueError, match=msg):
+                Timestamp("2015-10-25 02:00", tz=tz)
+
+        result = Timestamp("2017-03-26 01:00", tz="Europe/Paris")
+        expected = Timestamp("2017-03-26 01:00").tz_localize("Europe/Paris")
+        assert result == expected
+
+        msg = "2017-03-26 02:00"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2017-03-26 02:00", tz="Europe/Paris")
+
+        # GH#11708
+        naive = Timestamp("2015-11-18 10:00:00")
+        result = naive.tz_localize("UTC").tz_convert("Asia/Kolkata")
+        expected = Timestamp("2015-11-18 15:30:00+0530", tz="Asia/Kolkata")
+        assert result == expected
+
+        # GH#15823
+        result = Timestamp("2017-03-26 00:00", tz="Europe/Paris")
+        expected = Timestamp("2017-03-26 00:00:00+0100", tz="Europe/Paris")
+        assert result == expected
+
+        result = Timestamp("2017-03-26 01:00", tz="Europe/Paris")
+        expected = Timestamp("2017-03-26 01:00:00+0100", tz="Europe/Paris")
+        assert result == expected
+
+        msg = "2017-03-26 02:00"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2017-03-26 02:00", tz="Europe/Paris")
+
+        result = Timestamp("2017-03-26 02:00:00+0100", tz="Europe/Paris")
+        naive = Timestamp(result.as_unit("ns")._value)
+        expected = naive.tz_localize("UTC").tz_convert("Europe/Paris")
+        assert result == expected
+
+        result = Timestamp("2017-03-26 03:00", tz="Europe/Paris")
+        expected = Timestamp("2017-03-26 03:00:00+0200", tz="Europe/Paris")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "tz",
+        [
+            "pytz/US/Eastern",
+            gettz("US/Eastern"),
+            "US/Eastern",
+            "dateutil/US/Eastern",
+        ],
+    )
+    def test_timestamp_constructed_by_date_and_tz(self, tz):
+        # GH#2993, Timestamp cannot be constructed by datetime.date
+        # and tz correctly
+        if isinstance(tz, str) and tz.startswith("pytz/"):
+            pytz = pytest.importorskip("pytz")
+            tz = pytz.timezone(tz.removeprefix("pytz/"))
+        result = Timestamp(date(2012, 3, 11), tz=tz)
+
+        expected = Timestamp("3/11/2012", tz=tz)
+        assert result.hour == expected.hour
+        assert result == expected
+
+    def test_explicit_tz_none(self):
+        # GH#48688
+        msg = "Passed data is timezone-aware, incompatible with 'tz=None'"
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(datetime(2022, 1, 1, tzinfo=timezone.utc), tz=None)
+
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2022-01-01 00:00:00", tzinfo=timezone.utc, tz=None)
+
+        with pytest.raises(ValueError, match=msg):
+            Timestamp("2022-01-01 00:00:00-0400", tz=None)
+
+
+def test_constructor_ambiguous_dst():
+    # GH 24329
+    # Make sure that calling Timestamp constructor
+    # on Timestamp created from ambiguous time
+    # doesn't change Timestamp.value
+    ts = Timestamp(1382835600000000000, tz="dateutil/Europe/London")
+    expected = ts._value
+    result = Timestamp(ts)._value
+    assert result == expected
+
+
+@pytest.mark.parametrize("epoch", [1552211999999999872, 1552211999999999999])
+def test_constructor_before_dst_switch(epoch):
+    # GH 31043
+    # Make sure that calling Timestamp constructor
+    # on time just before DST switch doesn't lead to
+    # nonexistent time or value change
+    ts = Timestamp(epoch, tz="dateutil/America/Los_Angeles")
+    result = ts.tz.dst(ts)
+    expected = timedelta(seconds=0)
+    assert Timestamp(ts)._value == epoch
+    assert result == expected
+
+
+def test_timestamp_constructor_identity():
+    # Test for #30543
+    expected = Timestamp("2017-01-01T12")
+    result = Timestamp(expected)
+    assert result is expected
+
+
+@pytest.mark.parametrize("nano", [-1, 1000])
+def test_timestamp_nano_range(nano):
+    # GH 48255
+    with pytest.raises(ValueError, match="nanosecond must be in 0..999"):
+        Timestamp(year=2022, month=1, day=1, nanosecond=nano)
+
+
+def test_non_nano_value():
+    # https://github.com/pandas-dev/pandas/issues/49076
+    result = Timestamp("1800-01-01", unit="s").value
+    # `.value` shows nanoseconds, even though unit is 's'
+    assert result == -5364662400000000000
+
+    # out-of-nanoseconds-bounds `.value` raises informative message
+    msg = (
+        r"Cannot convert Timestamp to nanoseconds without overflow. "
+        r"Use `.asm8.view\('i8'\)` to cast represent Timestamp in its "
+        r"own unit \(here, s\).$"
+    )
+    ts = Timestamp("0300-01-01")
+    with pytest.raises(OverflowError, match=msg):
+        ts.value
+    # check that the suggested workaround actually works
+    result = ts.asm8.view("i8")
+    assert result == -52700112000
+
+
+@pytest.mark.parametrize("na_value", [None, np.nan, np.datetime64("NaT"), NaT, NA])
+def test_timestamp_constructor_na_value(na_value):
+    # GH45481
+    result = Timestamp(na_value)
+    expected = NaT
+    assert result is expected
diff --git a/pandas/pandas/tests/scalar/timestamp/test_formats.py b/pandas/pandas/tests/scalar/timestamp/test_formats.py
new file mode 100644
index 0000000000000..7b20f0a17556d
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_formats.py
@@ -0,0 +1,206 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+import pprint
+
+import dateutil.tz
+import pytest
+
+from pandas.compat import WASM
+
+from pandas import Timestamp
+
+ts_no_ns = Timestamp(
+    year=2019,
+    month=5,
+    day=18,
+    hour=15,
+    minute=17,
+    second=8,
+    microsecond=132263,
+)
+ts_no_ns_year1 = Timestamp(
+    year=1,
+    month=5,
+    day=18,
+    hour=15,
+    minute=17,
+    second=8,
+    microsecond=132263,
+)
+ts_ns = Timestamp(
+    year=2019,
+    month=5,
+    day=18,
+    hour=15,
+    minute=17,
+    second=8,
+    microsecond=132263,
+    nanosecond=123,
+)
+ts_ns_tz = Timestamp(
+    year=2019,
+    month=5,
+    day=18,
+    hour=15,
+    minute=17,
+    second=8,
+    microsecond=132263,
+    nanosecond=123,
+    tz="UTC",
+)
+ts_no_us = Timestamp(
+    year=2019,
+    month=5,
+    day=18,
+    hour=15,
+    minute=17,
+    second=8,
+    microsecond=0,
+    nanosecond=123,
+)
+
+
+@pytest.mark.parametrize(
+    "ts, timespec, expected_iso",
+    [
+        (ts_no_ns, "auto", "2019-05-18T15:17:08.132263"),
+        (ts_no_ns, "seconds", "2019-05-18T15:17:08"),
+        (ts_no_ns, "nanoseconds", "2019-05-18T15:17:08.132263000"),
+        (ts_no_ns_year1, "seconds", "0001-05-18T15:17:08"),
+        (ts_no_ns_year1, "nanoseconds", "0001-05-18T15:17:08.132263000"),
+        (ts_ns, "auto", "2019-05-18T15:17:08.132263123"),
+        (ts_ns, "hours", "2019-05-18T15"),
+        (ts_ns, "minutes", "2019-05-18T15:17"),
+        (ts_ns, "seconds", "2019-05-18T15:17:08"),
+        (ts_ns, "milliseconds", "2019-05-18T15:17:08.132"),
+        (ts_ns, "microseconds", "2019-05-18T15:17:08.132263"),
+        (ts_ns, "nanoseconds", "2019-05-18T15:17:08.132263123"),
+        (ts_ns_tz, "auto", "2019-05-18T15:17:08.132263123+00:00"),
+        (ts_ns_tz, "hours", "2019-05-18T15+00:00"),
+        (ts_ns_tz, "minutes", "2019-05-18T15:17+00:00"),
+        (ts_ns_tz, "seconds", "2019-05-18T15:17:08+00:00"),
+        (ts_ns_tz, "milliseconds", "2019-05-18T15:17:08.132+00:00"),
+        (ts_ns_tz, "microseconds", "2019-05-18T15:17:08.132263+00:00"),
+        (ts_ns_tz, "nanoseconds", "2019-05-18T15:17:08.132263123+00:00"),
+        (ts_no_us, "auto", "2019-05-18T15:17:08.000000123"),
+    ],
+)
+def test_isoformat(ts, timespec, expected_iso):
+    assert ts.isoformat(timespec=timespec) == expected_iso
+
+
+class TestTimestampRendering:
+    @pytest.mark.parametrize(
+        "tz", ["UTC", "Asia/Tokyo", "US/Eastern", "dateutil/America/Los_Angeles"]
+    )
+    @pytest.mark.parametrize("freq", ["D", "M", "S", "N"])
+    @pytest.mark.parametrize(
+        "date", ["2014-03-07", "2014-01-01 09:00", "2014-01-01 00:00:00.000000001"]
+    )
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    def test_repr(self, date, freq, tz):
+        # avoid to match with timezone name
+        freq_repr = f"'{freq}'"
+        if tz.startswith("dateutil"):
+            tz_repr = tz.replace("dateutil", "")
+        else:
+            tz_repr = tz
+
+        date_only = Timestamp(date)
+        assert date in repr(date_only)
+        assert tz_repr not in repr(date_only)
+        assert freq_repr not in repr(date_only)
+        assert date_only == eval(repr(date_only))
+
+        date_tz = Timestamp(date, tz=tz)
+        assert date in repr(date_tz)
+        assert tz_repr in repr(date_tz)
+        assert freq_repr not in repr(date_tz)
+        assert date_tz == eval(repr(date_tz))
+
+    def test_repr_utcoffset(self):
+        # This can cause the tz field to be populated, but it's redundant to
+        # include this information in the date-string.
+        date_with_utc_offset = Timestamp("2014-03-13 00:00:00-0400")
+        assert "2014-03-13 00:00:00-0400" in repr(date_with_utc_offset)
+        assert "tzoffset" not in repr(date_with_utc_offset)
+        assert "UTC-04:00" in repr(date_with_utc_offset)
+        expr = repr(date_with_utc_offset)
+        assert date_with_utc_offset == eval(expr)
+
+    def test_timestamp_repr_pre1900(self):
+        # pre-1900
+        stamp = Timestamp("1850-01-01", tz="US/Eastern")
+        repr(stamp)
+
+        iso8601 = "1850-01-01 01:23:45.012345"
+        stamp = Timestamp(iso8601, tz="US/Eastern")
+        result = repr(stamp)
+        assert iso8601 in result
+
+    def test_pprint(self):
+        # GH#12622
+        nested_obj = {"foo": 1, "bar": [{"w": {"a": Timestamp("2011-01-01")}}] * 10}
+        result = pprint.pformat(nested_obj, width=50)
+        expected = r"""{'bar': [{'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}},
+         {'w': {'a': Timestamp('2011-01-01 00:00:00')}}],
+ 'foo': 1}"""
+        assert result == expected
+
+    def test_to_timestamp_repr_is_code(self):
+        zs = [
+            Timestamp("99-04-17 00:00:00", tz="UTC"),
+            Timestamp("2001-04-17 00:00:00", tz="UTC"),
+            Timestamp("2001-04-17 00:00:00", tz="America/Los_Angeles"),
+            Timestamp("2001-04-17 00:00:00", tz=None),
+        ]
+        for z in zs:
+            assert eval(repr(z)) == z
+
+    def test_repr_matches_pydatetime_no_tz(self):
+        dt_date = datetime(2013, 1, 2)
+        assert str(dt_date) == str(Timestamp(dt_date))
+
+        dt_datetime = datetime(2013, 1, 2, 12, 1, 3)
+        assert str(dt_datetime) == str(Timestamp(dt_datetime))
+
+        dt_datetime_us = datetime(2013, 1, 2, 12, 1, 3, 45)
+        assert str(dt_datetime_us) == str(Timestamp(dt_datetime_us))
+
+        ts_nanos_only = Timestamp(200)
+        assert str(ts_nanos_only) == "1970-01-01 00:00:00.000000200"
+
+        ts_nanos_micros = Timestamp(1200)
+        assert str(ts_nanos_micros) == "1970-01-01 00:00:00.000001200"
+
+    def test_repr_matches_pydatetime_tz_stdlib(self):
+        dt_date = datetime(2013, 1, 2, tzinfo=timezone.utc)
+        assert str(dt_date) == str(Timestamp(dt_date))
+
+        dt_datetime = datetime(2013, 1, 2, 12, 1, 3, tzinfo=timezone.utc)
+        assert str(dt_datetime) == str(Timestamp(dt_datetime))
+
+        dt_datetime_us = datetime(2013, 1, 2, 12, 1, 3, 45, tzinfo=timezone.utc)
+        assert str(dt_datetime_us) == str(Timestamp(dt_datetime_us))
+
+    def test_repr_matches_pydatetime_tz_dateutil(self):
+        utc = dateutil.tz.tzutc()
+
+        dt_date = datetime(2013, 1, 2, tzinfo=utc)
+        assert str(dt_date) == str(Timestamp(dt_date))
+
+        dt_datetime = datetime(2013, 1, 2, 12, 1, 3, tzinfo=utc)
+        assert str(dt_datetime) == str(Timestamp(dt_datetime))
+
+        dt_datetime_us = datetime(2013, 1, 2, 12, 1, 3, 45, tzinfo=utc)
+        assert str(dt_datetime_us) == str(Timestamp(dt_datetime_us))
diff --git a/pandas/pandas/tests/scalar/timestamp/test_timestamp.py b/pandas/pandas/tests/scalar/timestamp/test_timestamp.py
new file mode 100644
index 0000000000000..38d0ddfbc13bd
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_timestamp.py
@@ -0,0 +1,931 @@
+"""test the scalar Timestamp"""
+
+import calendar
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+import locale
+import time
+import unicodedata
+import zoneinfo
+
+from dateutil.tz import (
+    tzlocal,
+    tzutc,
+)
+from hypothesis import (
+    given,
+    strategies as st,
+)
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas._libs.tslibs.timezones import (
+    dateutil_gettz as gettz,
+    get_timezone,
+    maybe_get_tz,
+    tz_compare,
+)
+from pandas.compat import IS64
+
+from pandas import (
+    NaT,
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+from pandas.tseries.frequencies import to_offset
+
+
+class TestTimestampProperties:
+    def test_properties_business(self):
+        freq = to_offset("B")
+
+        ts = Timestamp("2017-10-01")
+        assert ts.dayofweek == 6
+        assert ts.day_of_week == 6
+        assert ts.is_month_start  # not a weekday
+        assert not freq.is_month_start(ts)
+        assert freq.is_month_start(ts + Timedelta(days=1))
+        assert not freq.is_quarter_start(ts)
+        assert freq.is_quarter_start(ts + Timedelta(days=1))
+
+        ts = Timestamp("2017-09-30")
+        assert ts.dayofweek == 5
+        assert ts.day_of_week == 5
+        assert ts.is_month_end
+        assert not freq.is_month_end(ts)
+        assert freq.is_month_end(ts - Timedelta(days=1))
+        assert ts.is_quarter_end
+        assert not freq.is_quarter_end(ts)
+        assert freq.is_quarter_end(ts - Timedelta(days=1))
+
+    @pytest.mark.parametrize(
+        "attr, expected",
+        [
+            ["year", 2014],
+            ["month", 12],
+            ["day", 31],
+            ["hour", 23],
+            ["minute", 59],
+            ["second", 0],
+            ["microsecond", 0],
+            ["nanosecond", 0],
+            ["dayofweek", 2],
+            ["day_of_week", 2],
+            ["quarter", 4],
+            ["dayofyear", 365],
+            ["day_of_year", 365],
+            ["week", 1],
+            ["daysinmonth", 31],
+        ],
+    )
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_fields(self, attr, expected, tz):
+        # GH 10050
+        # GH 13303
+        ts = Timestamp("2014-12-31 23:59:00", tz=tz)
+        result = getattr(ts, attr)
+        # that we are int like
+        assert isinstance(result, int)
+        assert result == expected
+
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_millisecond_raises(self, tz):
+        ts = Timestamp("2014-12-31 23:59:00", tz=tz)
+        msg = "'Timestamp' object has no attribute 'millisecond'"
+        with pytest.raises(AttributeError, match=msg):
+            ts.millisecond
+
+    @pytest.mark.parametrize(
+        "start", ["is_month_start", "is_quarter_start", "is_year_start"]
+    )
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_is_start(self, start, tz):
+        ts = Timestamp("2014-01-01 00:00:00", tz=tz)
+        assert getattr(ts, start)
+
+    @pytest.mark.parametrize("end", ["is_month_end", "is_year_end", "is_quarter_end"])
+    @pytest.mark.parametrize("tz", [None, "US/Eastern"])
+    def test_is_end(self, end, tz):
+        ts = Timestamp("2014-12-31 23:59:59", tz=tz)
+        assert getattr(ts, end)
+
+    # GH 12806
+    @pytest.mark.parametrize("tz", [None, "EST"])
+    # error: Unsupported operand types for + ("List[None]" and "List[str]")
+    @pytest.mark.parametrize(
+        "time_locale",
+        [None] + tm.get_locales(),  # type: ignore[operator]
+    )
+    def test_names(self, tz, time_locale):
+        # GH 17354
+        # Test .day_name(), .month_name
+        data = Timestamp("2017-08-28 23:00:00", tz=tz)
+        if time_locale is None:
+            expected_day = "Monday"
+            expected_month = "August"
+        else:
+            with tm.set_locale(time_locale, locale.LC_TIME):
+                expected_day = calendar.day_name[0].capitalize()
+                expected_month = calendar.month_name[8].capitalize()
+
+        result_day = data.day_name(time_locale)
+        result_month = data.month_name(time_locale)
+
+        # Work around https://github.com/pandas-dev/pandas/issues/22342
+        # different normalizations
+        expected_day = unicodedata.normalize("NFD", expected_day)
+        expected_month = unicodedata.normalize("NFD", expected_month)
+
+        result_day = unicodedata.normalize("NFD", result_day)
+        result_month = unicodedata.normalize("NFD", result_month)
+
+        assert result_day == expected_day
+        assert result_month == expected_month
+
+        # Test NaT
+        nan_ts = Timestamp(NaT)
+        assert np.isnan(nan_ts.day_name(time_locale))
+        assert np.isnan(nan_ts.month_name(time_locale))
+
+    def test_is_leap_year(self, tz_naive_fixture):
+        tz = tz_naive_fixture
+        if not IS64 and tz == tzlocal():
+            # https://github.com/dateutil/dateutil/issues/197
+            pytest.skip(
+                "tzlocal() on a 32 bit platform causes internal overflow errors"
+            )
+        # GH 13727
+        dt = Timestamp("2000-01-01 00:00:00", tz=tz)
+        assert dt.is_leap_year
+        assert isinstance(dt.is_leap_year, bool)
+
+        dt = Timestamp("1999-01-01 00:00:00", tz=tz)
+        assert not dt.is_leap_year
+
+        dt = Timestamp("2004-01-01 00:00:00", tz=tz)
+        assert dt.is_leap_year
+
+        dt = Timestamp("2100-01-01 00:00:00", tz=tz)
+        assert not dt.is_leap_year
+
+    def test_woy_boundary(self):
+        # make sure weeks at year boundaries are correct
+        d = datetime(2013, 12, 31)
+        result = Timestamp(d).week
+        expected = 1  # ISO standard
+        assert result == expected
+
+        d = datetime(2008, 12, 28)
+        result = Timestamp(d).week
+        expected = 52  # ISO standard
+        assert result == expected
+
+        d = datetime(2009, 12, 31)
+        result = Timestamp(d).week
+        expected = 53  # ISO standard
+        assert result == expected
+
+        d = datetime(2010, 1, 1)
+        result = Timestamp(d).week
+        expected = 53  # ISO standard
+        assert result == expected
+
+        d = datetime(2010, 1, 3)
+        result = Timestamp(d).week
+        expected = 53  # ISO standard
+        assert result == expected
+
+        result = np.array(
+            [
+                Timestamp(datetime(*args)).week
+                for args in [(2000, 1, 1), (2000, 1, 2), (2005, 1, 1), (2005, 1, 2)]
+            ]
+        )
+        assert (result == [52, 52, 53, 53]).all()
+
+    def test_resolution(self):
+        # GH#21336, GH#21365
+        dt = Timestamp("2100-01-01 00:00:00.000000000")
+        assert dt.resolution == Timedelta(nanoseconds=1)
+
+        # Check that the attribute is available on the class, mirroring
+        #  the stdlib datetime behavior
+        assert Timestamp.resolution == Timedelta(nanoseconds=1)
+
+        assert dt.as_unit("us").resolution == Timedelta(microseconds=1)
+        assert dt.as_unit("ms").resolution == Timedelta(milliseconds=1)
+        assert dt.as_unit("s").resolution == Timedelta(seconds=1)
+
+    @pytest.mark.parametrize(
+        "date_string, expected",
+        [
+            ("0000-2-29", 1),
+            ("0000-3-1", 2),
+            ("1582-10-14", 3),
+            ("-0040-1-1", 4),
+            ("2023-06-18", 6),
+        ],
+    )
+    def test_dow_historic(self, date_string, expected):
+        # GH 53738
+        ts = Timestamp(date_string)
+        dow = ts.weekday()
+        assert dow == expected
+
+    @given(
+        ts=st.datetimes(),
+        sign=st.sampled_from(["-", ""]),
+    )
+    def test_dow_parametric(self, ts, sign):
+        # GH 53738
+        ts = (
+            f"{sign}{str(ts.year).zfill(4)}"
+            f"-{str(ts.month).zfill(2)}"
+            f"-{str(ts.day).zfill(2)}"
+        )
+        result = Timestamp(ts).weekday()
+        expected = (
+            (np.datetime64(ts) - np.datetime64("1970-01-01")).astype("int64") - 4
+        ) % 7
+        assert result == expected
+
+
+class TestTimestamp:
+    @pytest.mark.parametrize("tz", [None, zoneinfo.ZoneInfo("US/Pacific")])
+    def test_disallow_setting_tz(self, tz):
+        # GH#3746
+        ts = Timestamp("2010")
+        msg = "Cannot directly set timezone"
+        with pytest.raises(AttributeError, match=msg):
+            ts.tz = tz
+
+    def test_default_to_stdlib_utc(self):
+        msg = "Timestamp.utcnow is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            assert Timestamp.utcnow().tz is timezone.utc
+        assert Timestamp.now("UTC").tz is timezone.utc
+        assert Timestamp("2016-01-01", tz="UTC").tz is timezone.utc
+
+    def test_tz(self):
+        tstr = "2014-02-01 09:00"
+        ts = Timestamp(tstr)
+        local = ts.tz_localize("Asia/Tokyo")
+        assert local.hour == 9
+        assert local == Timestamp(tstr, tz="Asia/Tokyo")
+        conv = local.tz_convert("US/Eastern")
+        assert conv == Timestamp("2014-01-31 19:00", tz="US/Eastern")
+        assert conv.hour == 19
+
+        # preserves nanosecond
+        ts = Timestamp(tstr) + offsets.Nano(5)
+        local = ts.tz_localize("Asia/Tokyo")
+        assert local.hour == 9
+        assert local.nanosecond == 5
+        conv = local.tz_convert("US/Eastern")
+        assert conv.nanosecond == 5
+        assert conv.hour == 19
+
+    def test_utc_z_designator(self):
+        assert get_timezone(Timestamp("2014-11-02 01:00Z").tzinfo) is timezone.utc
+
+    def test_asm8(self):
+        ns = [Timestamp.min._value, Timestamp.max._value, 1000]
+
+        for n in ns:
+            assert (
+                Timestamp(n).asm8.view("i8") == np.datetime64(n, "ns").view("i8") == n
+            )
+
+        assert Timestamp("nat").asm8.view("i8") == np.datetime64("nat", "ns").view("i8")
+
+    def test_class_ops(self):
+        def compare(x, y):
+            assert int((Timestamp(x)._value - Timestamp(y)._value) / 1e9) == 0
+
+        compare(Timestamp.now(), datetime.now())
+        compare(Timestamp.now("UTC"), datetime.now(timezone.utc))
+        compare(Timestamp.now("UTC"), datetime.now(tzutc()))
+        msg = "Timestamp.utcnow is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            compare(Timestamp.utcnow(), datetime.now(timezone.utc))
+        compare(Timestamp.today(), datetime.today())
+        current_time = calendar.timegm(datetime.now().utctimetuple())
+
+        msg = "Timestamp.utcfromtimestamp is deprecated"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            ts_utc = Timestamp.utcfromtimestamp(current_time)
+        assert ts_utc.timestamp() == current_time
+        compare(
+            Timestamp.fromtimestamp(current_time), datetime.fromtimestamp(current_time)
+        )
+        compare(
+            # Support tz kwarg in Timestamp.fromtimestamp
+            Timestamp.fromtimestamp(current_time, "UTC"),
+            datetime.fromtimestamp(current_time, timezone.utc),
+        )
+        compare(
+            # Support tz kwarg in Timestamp.fromtimestamp
+            Timestamp.fromtimestamp(current_time, tz="UTC"),
+            datetime.fromtimestamp(current_time, timezone.utc),
+        )
+
+        date_component = datetime.now(timezone.utc)
+        time_component = (date_component + timedelta(minutes=10)).time()
+        compare(
+            Timestamp.combine(date_component, time_component),
+            datetime.combine(date_component, time_component),
+        )
+
+    def test_basics_nanos(self):
+        val = np.int64(946_684_800_000_000_000).view("M8[ns]")
+        stamp = Timestamp(val.view("i8") + 500)
+        assert stamp.year == 2000
+        assert stamp.month == 1
+        assert stamp.microsecond == 0
+        assert stamp.nanosecond == 500
+
+        # GH 14415
+        val = np.iinfo(np.int64).min + 80_000_000_000_000
+        stamp = Timestamp(val)
+        assert stamp.year == 1677
+        assert stamp.month == 9
+        assert stamp.day == 21
+        assert stamp.microsecond == 145224
+        assert stamp.nanosecond == 192
+
+    def test_roundtrip(self):
+        # test value to string and back conversions
+        # further test accessors
+        base = Timestamp("20140101 00:00:00").as_unit("ns")
+
+        result = Timestamp(base._value + Timedelta("5ms")._value)
+        assert result == Timestamp(f"{base}.005000")
+        assert result.microsecond == 5000
+
+        result = Timestamp(base._value + Timedelta("5us")._value)
+        assert result == Timestamp(f"{base}.000005")
+        assert result.microsecond == 5
+
+        result = Timestamp(base._value + Timedelta("5ns")._value)
+        assert result == Timestamp(f"{base}.000000005")
+        assert result.nanosecond == 5
+        assert result.microsecond == 0
+
+        result = Timestamp(base._value + Timedelta("6ms 5us")._value)
+        assert result == Timestamp(f"{base}.006005")
+        assert result.microsecond == 5 + 6 * 1000
+
+        result = Timestamp(base._value + Timedelta("200ms 5us")._value)
+        assert result == Timestamp(f"{base}.200005")
+        assert result.microsecond == 5 + 200 * 1000
+
+    def test_hash_equivalent(self):
+        d = {datetime(2011, 1, 1): 5}
+        stamp = Timestamp(datetime(2011, 1, 1))
+        assert d[stamp] == 5
+
+    @pytest.mark.parametrize(
+        "timezone, year, month, day, hour",
+        [["America/Chicago", 2013, 11, 3, 1], ["America/Santiago", 2021, 4, 3, 23]],
+    )
+    def test_hash_timestamp_with_fold(self, timezone, year, month, day, hour):
+        # see gh-33931
+        test_timezone = gettz(timezone)
+        transition_1 = Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=0,
+            tzinfo=test_timezone,
+        )
+        transition_2 = Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=1,
+            tzinfo=test_timezone,
+        )
+        assert hash(transition_1) == hash(transition_2)
+
+
+class TestTimestampNsOperations:
+    def test_nanosecond_string_parsing(self):
+        ts = Timestamp("2013-05-01 07:15:45.123456789")
+        # GH 7878
+        expected_repr = "2013-05-01 07:15:45.123456789"
+        expected_value = 1_367_392_545_123_456_789
+        assert ts._value == expected_value
+        assert expected_repr in repr(ts)
+
+        ts = Timestamp("2013-05-01 07:15:45.123456789+09:00", tz="Asia/Tokyo")
+        assert ts._value == expected_value - 9 * 3600 * 1_000_000_000
+        assert expected_repr in repr(ts)
+
+        ts = Timestamp("2013-05-01 07:15:45.123456789", tz="UTC")
+        assert ts._value == expected_value
+        assert expected_repr in repr(ts)
+
+        ts = Timestamp("2013-05-01 07:15:45.123456789", tz="US/Eastern")
+        assert ts._value == expected_value + 4 * 3600 * 1_000_000_000
+        assert expected_repr in repr(ts)
+
+        # GH 10041
+        ts = Timestamp("20130501T071545.123456789")
+        assert ts._value == expected_value
+        assert expected_repr in repr(ts)
+
+    def test_nanosecond_timestamp(self):
+        # GH 7610
+        expected = 1_293_840_000_000_000_005
+        t = Timestamp("2011-01-01") + offsets.Nano(5)
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000005')"
+        assert t._value == expected
+        assert t.nanosecond == 5
+
+        t = Timestamp(t)
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000005')"
+        assert t._value == expected
+        assert t.nanosecond == 5
+
+        t = Timestamp("2011-01-01 00:00:00.000000005")
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000005')"
+        assert t._value == expected
+        assert t.nanosecond == 5
+
+        expected = 1_293_840_000_000_000_010
+        t = t + offsets.Nano(5)
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000010')"
+        assert t._value == expected
+        assert t.nanosecond == 10
+
+        t = Timestamp(t)
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000010')"
+        assert t._value == expected
+        assert t.nanosecond == 10
+
+        t = Timestamp("2011-01-01 00:00:00.000000010")
+        assert repr(t) == "Timestamp('2011-01-01 00:00:00.000000010')"
+        assert t._value == expected
+        assert t.nanosecond == 10
+
+
+class TestTimestampConversion:
+    def test_conversion(self):
+        # GH#9255
+        ts = Timestamp("2000-01-01").as_unit("ns")
+
+        result = ts.to_pydatetime()
+        expected = datetime(2000, 1, 1)
+        assert result == expected
+        assert type(result) == type(expected)
+
+        result = ts.to_datetime64()
+        expected = np.datetime64(ts._value, "ns")
+        assert result == expected
+        assert type(result) == type(expected)
+        assert result.dtype == expected.dtype
+
+    def test_to_period_tz_warning(self):
+        # GH#21333 make sure a warning is issued when timezone
+        # info is lost
+        ts = Timestamp("2009-04-15 16:17:18", tz="US/Eastern")
+        with tm.assert_produces_warning(UserWarning, match="drop timezone information"):
+            ts.to_period("D")
+
+    def test_to_numpy_alias(self):
+        # GH 24653: alias .to_numpy() for scalars
+        ts = Timestamp(datetime.now())
+        assert ts.to_datetime64() == ts.to_numpy()
+
+        # GH#44460
+        msg = "dtype and copy arguments are ignored"
+        with pytest.raises(ValueError, match=msg):
+            ts.to_numpy("M8[s]")
+        with pytest.raises(ValueError, match=msg):
+            ts.to_numpy(copy=True)
+
+
+class TestNonNano:
+    @pytest.fixture(params=["s", "ms", "us"])
+    def reso(self, request):
+        return request.param
+
+    @pytest.fixture
+    def dt64(self, reso):
+        # cases that are in-bounds for nanosecond, so we can compare against
+        #  the existing implementation.
+        return np.datetime64("2016-01-01", reso)
+
+    @pytest.fixture
+    def ts(self, dt64):
+        return Timestamp._from_dt64(dt64)
+
+    @pytest.fixture
+    def ts_tz(self, ts, tz_aware_fixture):
+        tz = maybe_get_tz(tz_aware_fixture)
+        return Timestamp._from_value_and_reso(ts._value, ts._creso, tz)
+
+    def test_non_nano_construction(self, dt64, ts, reso):
+        assert ts._value == dt64.view("i8")
+
+        if reso == "s":
+            assert ts._creso == NpyDatetimeUnit.NPY_FR_s.value
+        elif reso == "ms":
+            assert ts._creso == NpyDatetimeUnit.NPY_FR_ms.value
+        elif reso == "us":
+            assert ts._creso == NpyDatetimeUnit.NPY_FR_us.value
+
+    def test_non_nano_fields(self, dt64, ts):
+        alt = Timestamp(dt64)
+
+        assert ts.year == alt.year
+        assert ts.month == alt.month
+        assert ts.day == alt.day
+        assert ts.hour == ts.minute == ts.second == ts.microsecond == 0
+        assert ts.nanosecond == 0
+
+        assert ts.to_julian_date() == alt.to_julian_date()
+        assert ts.weekday() == alt.weekday()
+        assert ts.isoweekday() == alt.isoweekday()
+
+    def test_start_end_fields(self, ts):
+        assert ts.is_year_start
+        assert ts.is_quarter_start
+        assert ts.is_month_start
+        assert not ts.is_year_end
+        assert not ts.is_month_end
+        assert not ts.is_month_end
+
+        # 2016-01-01 is a Friday, so is year/quarter/month start with this freq
+        assert ts.is_year_start
+        assert ts.is_quarter_start
+        assert ts.is_month_start
+        assert not ts.is_year_end
+        assert not ts.is_month_end
+        assert not ts.is_month_end
+
+    def test_day_name(self, dt64, ts):
+        alt = Timestamp(dt64)
+        assert ts.day_name() == alt.day_name()
+
+    def test_month_name(self, dt64, ts):
+        alt = Timestamp(dt64)
+        assert ts.month_name() == alt.month_name()
+
+    def test_tz_convert(self, ts):
+        ts = Timestamp._from_value_and_reso(ts._value, ts._creso, timezone.utc)
+
+        tz = zoneinfo.ZoneInfo("US/Pacific")
+        result = ts.tz_convert(tz)
+
+        assert isinstance(result, Timestamp)
+        assert result._creso == ts._creso
+        assert tz_compare(result.tz, tz)
+
+    def test_repr(self, dt64, ts):
+        alt = Timestamp(dt64)
+
+        assert str(ts) == str(alt)
+        assert repr(ts) == repr(alt)
+
+    def test_comparison(self, dt64, ts):
+        alt = Timestamp(dt64)
+
+        assert ts == dt64
+        assert dt64 == ts
+        assert ts == alt
+        assert alt == ts
+
+        assert not ts != dt64
+        assert not dt64 != ts
+        assert not ts != alt
+        assert not alt != ts
+
+        assert not ts < dt64
+        assert not dt64 < ts
+        assert not ts < alt
+        assert not alt < ts
+
+        assert not ts > dt64
+        assert not dt64 > ts
+        assert not ts > alt
+        assert not alt > ts
+
+        assert ts >= dt64
+        assert dt64 >= ts
+        assert ts >= alt
+        assert alt >= ts
+
+        assert ts <= dt64
+        assert dt64 <= ts
+        assert ts <= alt
+        assert alt <= ts
+
+    def test_cmp_cross_reso(self):
+        # numpy gets this wrong because of silent overflow
+        dt64 = np.datetime64(9223372800, "s")  # won't fit in M8[ns]
+        ts = Timestamp._from_dt64(dt64)
+
+        # subtracting 3600*24 gives a datetime64 that _can_ fit inside the
+        #  nanosecond implementation bounds.
+        other = Timestamp(dt64 - 3600 * 24).as_unit("ns")
+        assert other < ts
+        assert other.asm8 > ts.asm8  # <- numpy gets this wrong
+        assert ts > other
+        assert ts.asm8 < other.asm8  # <- numpy gets this wrong
+        assert not other == ts
+        assert ts != other
+
+    @pytest.mark.xfail(reason="Dispatches to np.datetime64 which is wrong")
+    def test_cmp_cross_reso_reversed_dt64(self):
+        dt64 = np.datetime64(106752, "D")  # won't fit in M8[ns]
+        ts = Timestamp._from_dt64(dt64)
+        other = Timestamp(dt64 - 1)
+
+        assert other.asm8 < ts
+
+    def test_pickle(self, ts, tz_aware_fixture):
+        tz = tz_aware_fixture
+        tz = maybe_get_tz(tz)
+        ts = Timestamp._from_value_and_reso(ts._value, ts._creso, tz)
+        rt = tm.round_trip_pickle(ts)
+        assert rt._creso == ts._creso
+        assert rt == ts
+
+    def test_normalize(self, dt64, ts):
+        alt = Timestamp(dt64)
+        result = ts.normalize()
+        assert result._creso == ts._creso
+        assert result == alt.normalize()
+
+    def test_asm8(self, dt64, ts):
+        rt = ts.asm8
+        assert rt == dt64
+        assert rt.dtype == dt64.dtype
+
+    def test_to_numpy(self, dt64, ts):
+        res = ts.to_numpy()
+        assert res == dt64
+        assert res.dtype == dt64.dtype
+
+    def test_to_datetime64(self, dt64, ts):
+        res = ts.to_datetime64()
+        assert res == dt64
+        assert res.dtype == dt64.dtype
+
+    def test_timestamp(self, dt64, ts):
+        alt = Timestamp(dt64)
+        assert ts.timestamp() == alt.timestamp()
+
+    def test_to_period(self, dt64, ts):
+        alt = Timestamp(dt64)
+        assert ts.to_period("D") == alt.to_period("D")
+
+    @pytest.mark.parametrize(
+        "td", [timedelta(days=4), Timedelta(days=4), np.timedelta64(4, "D")]
+    )
+    def test_addsub_timedeltalike_non_nano(self, dt64, ts, td):
+        exp_reso = max(ts._creso, Timedelta(td)._creso)
+
+        result = ts - td
+        expected = Timestamp(dt64) - td
+        assert isinstance(result, Timestamp)
+        assert result._creso == exp_reso
+        assert result == expected
+
+        result = ts + td
+        expected = Timestamp(dt64) + td
+        assert isinstance(result, Timestamp)
+        assert result._creso == exp_reso
+        assert result == expected
+
+        result = td + ts
+        expected = td + Timestamp(dt64)
+        assert isinstance(result, Timestamp)
+        assert result._creso == exp_reso
+        assert result == expected
+
+    def test_addsub_offset(self, ts_tz):
+        # specifically non-Tick offset
+        off = offsets.YearEnd(1)
+        result = ts_tz + off
+
+        assert isinstance(result, Timestamp)
+        assert result._creso == ts_tz._creso
+        if ts_tz.month == 12 and ts_tz.day == 31:
+            assert result.year == ts_tz.year + 1
+        else:
+            assert result.year == ts_tz.year
+        assert result.day == 31
+        assert result.month == 12
+        assert tz_compare(result.tz, ts_tz.tz)
+
+        result = ts_tz - off
+
+        assert isinstance(result, Timestamp)
+        assert result._creso == ts_tz._creso
+        assert result.year == ts_tz.year - 1
+        assert result.day == 31
+        assert result.month == 12
+        assert tz_compare(result.tz, ts_tz.tz)
+
+    def test_sub_datetimelike_mismatched_reso(self, ts_tz):
+        # case with non-lossy rounding
+        ts = ts_tz
+
+        # choose a unit for `other` that doesn't match ts_tz's;
+        #  this construction ensures we get cases with other._creso < ts._creso
+        #  and cases with other._creso > ts._creso
+        unit = {
+            NpyDatetimeUnit.NPY_FR_us.value: "ms",
+            NpyDatetimeUnit.NPY_FR_ms.value: "s",
+            NpyDatetimeUnit.NPY_FR_s.value: "us",
+        }[ts._creso]
+        other = ts.as_unit(unit)
+        assert other._creso != ts._creso
+
+        result = ts - other
+        assert isinstance(result, Timedelta)
+        assert result._value == 0
+        assert result._creso == max(ts._creso, other._creso)
+
+        result = other - ts
+        assert isinstance(result, Timedelta)
+        assert result._value == 0
+        assert result._creso == max(ts._creso, other._creso)
+
+        if ts._creso < other._creso:
+            # Case where rounding is lossy
+            other2 = other + Timedelta._from_value_and_reso(1, other._creso)
+            exp = ts.as_unit(other.unit) - other2
+
+            res = ts - other2
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+
+            res = other2 - ts
+            assert res == -exp
+            assert res._creso == max(ts._creso, other._creso)
+        else:
+            ts2 = ts + Timedelta._from_value_and_reso(1, ts._creso)
+            exp = ts2 - other.as_unit(ts2.unit)
+
+            res = ts2 - other
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+            res = other - ts2
+            assert res == -exp
+            assert res._creso == max(ts._creso, other._creso)
+
+    def test_sub_timedeltalike_mismatched_reso(self, ts_tz):
+        # case with non-lossy rounding
+        ts = ts_tz
+
+        # choose a unit for `other` that doesn't match ts_tz's;
+        #  this construction ensures we get cases with other._creso < ts._creso
+        #  and cases with other._creso > ts._creso
+        unit = {
+            NpyDatetimeUnit.NPY_FR_us.value: "ms",
+            NpyDatetimeUnit.NPY_FR_ms.value: "s",
+            NpyDatetimeUnit.NPY_FR_s.value: "us",
+        }[ts._creso]
+        other = Timedelta(0).as_unit(unit)
+        assert other._creso != ts._creso
+
+        result = ts + other
+        assert isinstance(result, Timestamp)
+        assert result == ts
+        assert result._creso == max(ts._creso, other._creso)
+
+        result = other + ts
+        assert isinstance(result, Timestamp)
+        assert result == ts
+        assert result._creso == max(ts._creso, other._creso)
+
+        if ts._creso < other._creso:
+            # Case where rounding is lossy
+            other2 = other + Timedelta._from_value_and_reso(1, other._creso)
+            exp = ts.as_unit(other.unit) + other2
+            res = ts + other2
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+            res = other2 + ts
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+        else:
+            ts2 = ts + Timedelta._from_value_and_reso(1, ts._creso)
+            exp = ts2 + other.as_unit(ts2.unit)
+
+            res = ts2 + other
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+            res = other + ts2
+            assert res == exp
+            assert res._creso == max(ts._creso, other._creso)
+
+    def test_addition_doesnt_downcast_reso(self):
+        # https://github.com/pandas-dev/pandas/pull/48748#pullrequestreview-1122635413
+        ts = Timestamp(year=2022, month=1, day=1, microsecond=999999).as_unit("us")
+        td = Timedelta(microseconds=1).as_unit("us")
+        res = ts + td
+        assert res._creso == ts._creso
+
+    def test_sub_timedelta64_mismatched_reso(self, ts_tz):
+        ts = ts_tz
+
+        res = ts + np.timedelta64(1, "ns")
+        exp = ts.as_unit("ns") + np.timedelta64(1, "ns")
+        assert exp == res
+        assert exp._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    def test_min(self, ts):
+        assert ts.min <= ts
+        assert ts.min._creso == ts._creso
+        assert ts.min._value == NaT._value + 1
+
+    def test_max(self, ts):
+        assert ts.max >= ts
+        assert ts.max._creso == ts._creso
+        assert ts.max._value == np.iinfo(np.int64).max
+
+    def test_resolution(self, ts):
+        expected = Timedelta._from_value_and_reso(1, ts._creso)
+        result = ts.resolution
+        assert result == expected
+        assert result._creso == expected._creso
+
+    def test_out_of_ns_bounds(self):
+        # https://github.com/pandas-dev/pandas/issues/51060
+        result = Timestamp(-52700112000, unit="s")
+        assert result == Timestamp("0300-01-01")
+        assert result.to_numpy() == np.datetime64("0300-01-01T00:00:00", "s")
+
+
+def test_timestamp_class_min_max_resolution():
+    # when accessed on the class (as opposed to an instance), we default
+    #  to nanoseconds
+    assert Timestamp.min == Timestamp(NaT._value + 1)
+    assert Timestamp.min._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    assert Timestamp.max == Timestamp(np.iinfo(np.int64).max)
+    assert Timestamp.max._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+    assert Timestamp.resolution == Timedelta(1)
+    assert Timestamp.resolution._creso == NpyDatetimeUnit.NPY_FR_ns.value
+
+
+def test_delimited_date():
+    # https://github.com/pandas-dev/pandas/issues/50231
+    with tm.assert_produces_warning(None):
+        result = Timestamp("13-01-2000")
+    expected = Timestamp(2000, 1, 13)
+    assert result == expected
+
+
+def test_utctimetuple():
+    # GH 32174
+    ts = Timestamp("2000-01-01", tz="UTC")
+    result = ts.utctimetuple()
+    expected = time.struct_time((2000, 1, 1, 0, 0, 0, 5, 1, 0))
+    assert result == expected
+
+
+def test_negative_dates():
+    # https://github.com/pandas-dev/pandas/issues/50787
+    ts = Timestamp("-2000-01-01")
+    msg = (
+        " not yet supported on Timestamps which are outside the range of "
+        "Python's standard library. For now, please call the components you need "
+        r"\(such as `.year` and `.month`\) and construct your string from there.$"
+    )
+    func = "^strftime"
+    with pytest.raises(NotImplementedError, match=func + msg):
+        ts.strftime("%Y")
+
+    msg = (
+        " not yet supported on Timestamps which "
+        "are outside the range of Python's standard library. "
+    )
+    func = "^date"
+    with pytest.raises(NotImplementedError, match=func + msg):
+        ts.date()
+    func = "^isocalendar"
+    with pytest.raises(NotImplementedError, match=func + msg):
+        ts.isocalendar()
+    func = "^timetuple"
+    with pytest.raises(NotImplementedError, match=func + msg):
+        ts.timetuple()
+    func = "^toordinal"
+    with pytest.raises(NotImplementedError, match=func + msg):
+        ts.toordinal()
diff --git a/pandas/pandas/tests/scalar/timestamp/test_timezones.py b/pandas/pandas/tests/scalar/timestamp/test_timezones.py
new file mode 100644
index 0000000000000..8f2ee3ef45075
--- /dev/null
+++ b/pandas/pandas/tests/scalar/timestamp/test_timezones.py
@@ -0,0 +1,25 @@
+"""
+Tests for Timestamp timezone-related methods
+"""
+
+from datetime import datetime
+
+from pandas._libs.tslibs import timezones
+
+from pandas import Timestamp
+
+
+class TestTimestampTZOperations:
+    # ------------------------------------------------------------------
+
+    def test_timestamp_timetz_equivalent_with_datetime_tz(self, tz_naive_fixture):
+        # GH21358
+        tz = timezones.maybe_get_tz(tz_naive_fixture)
+
+        stamp = Timestamp("2018-06-04 10:20:30", tz=tz)
+        _datetime = datetime(2018, 6, 4, hour=10, minute=20, second=30, tzinfo=tz)
+
+        result = stamp.timetz()
+        expected = _datetime.timetz()
+
+        assert result == expected
diff --git a/pandas/pandas/tests/series/__init__.py b/pandas/pandas/tests/series/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/series/accessors/__init__.py b/pandas/pandas/tests/series/accessors/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/series/accessors/test_cat_accessor.py b/pandas/pandas/tests/series/accessors/test_cat_accessor.py
new file mode 100644
index 0000000000000..ce8ea27ea1fa2
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_cat_accessor.py
@@ -0,0 +1,258 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays.categorical import CategoricalAccessor
+from pandas.core.indexes.accessors import Properties
+
+
+class TestCatAccessor:
+    @pytest.mark.parametrize(
+        "method",
+        [
+            lambda x: x.cat.set_categories([1, 2, 3]),
+            lambda x: x.cat.reorder_categories([2, 3, 1], ordered=True),
+            lambda x: x.cat.rename_categories([1, 2, 3]),
+            lambda x: x.cat.remove_unused_categories(),
+            lambda x: x.cat.remove_categories([2]),
+            lambda x: x.cat.add_categories([4]),
+            lambda x: x.cat.as_ordered(),
+            lambda x: x.cat.as_unordered(),
+        ],
+    )
+    def test_getname_categorical_accessor(self, method):
+        # GH#17509
+        ser = Series([1, 2, 3], name="A").astype("category")
+        expected = "A"
+        result = method(ser).name
+        assert result == expected
+
+    def test_cat_accessor(self):
+        ser = Series(Categorical(["a", "b", np.nan, "a"]))
+        tm.assert_index_equal(ser.cat.categories, Index(["a", "b"]))
+        assert not ser.cat.ordered, False
+
+        exp = Categorical(["a", "b", np.nan, "a"], categories=["b", "a"])
+
+        res = ser.cat.set_categories(["b", "a"])
+        tm.assert_categorical_equal(res.values, exp)
+
+        ser[:] = "a"
+        ser = ser.cat.remove_unused_categories()
+        tm.assert_index_equal(ser.cat.categories, Index(["a"]))
+
+    def test_cat_accessor_api(self):
+        # GH#9322
+
+        assert Series.cat is CategoricalAccessor
+        ser = Series(list("aabbcde")).astype("category")
+        assert isinstance(ser.cat, CategoricalAccessor)
+
+        invalid = Series([1])
+        with pytest.raises(AttributeError, match="only use .cat accessor"):
+            invalid.cat
+        assert not hasattr(invalid, "cat")
+
+    def test_cat_accessor_no_new_attributes(self):
+        # https://github.com/pandas-dev/pandas/issues/10673
+        cat = Series(list("aabbcde")).astype("category")
+        with pytest.raises(AttributeError, match="You cannot add any new attribute"):
+            cat.cat.xlabel = "a"
+
+    def test_categorical_delegations(self):
+        # invalid accessor
+        msg = r"Can only use \.cat accessor with a 'category' dtype"
+        with pytest.raises(AttributeError, match=msg):
+            Series([1, 2, 3]).cat
+        with pytest.raises(AttributeError, match=msg):
+            Series([1, 2, 3]).cat()
+        with pytest.raises(AttributeError, match=msg):
+            Series(["a", "b", "c"]).cat
+        with pytest.raises(AttributeError, match=msg):
+            Series(np.arange(5.0)).cat
+        with pytest.raises(AttributeError, match=msg):
+            Series([Timestamp("20130101")]).cat
+
+        # Series should delegate calls to '.categories', '.codes', '.ordered'
+        # and the methods '.set_categories()' 'drop_unused_categories()' to the
+        # categorical
+        ser = Series(Categorical(["a", "b", "c", "a"], ordered=True))
+        exp_categories = Index(["a", "b", "c"])
+        tm.assert_index_equal(ser.cat.categories, exp_categories)
+        ser = ser.cat.rename_categories([1, 2, 3])
+        exp_categories = Index([1, 2, 3])
+        tm.assert_index_equal(ser.cat.categories, exp_categories)
+
+        exp_codes = Series([0, 1, 2, 0], dtype="int8")
+        tm.assert_series_equal(ser.cat.codes, exp_codes)
+
+        assert ser.cat.ordered
+        ser = ser.cat.as_unordered()
+        assert not ser.cat.ordered
+
+        ser = ser.cat.as_ordered()
+        assert ser.cat.ordered
+
+        # reorder
+        ser = Series(Categorical(["a", "b", "c", "a"], ordered=True))
+        exp_categories = Index(["c", "b", "a"])
+        exp_values = np.array(["a", "b", "c", "a"], dtype=np.object_)
+        ser = ser.cat.set_categories(["c", "b", "a"])
+        tm.assert_index_equal(ser.cat.categories, exp_categories)
+        tm.assert_numpy_array_equal(ser.values.__array__(), exp_values)
+        tm.assert_numpy_array_equal(ser.__array__(), exp_values)
+
+        # remove unused categories
+        ser = Series(Categorical(["a", "b", "b", "a"], categories=["a", "b", "c"]))
+        exp_categories = Index(["a", "b"])
+        exp_values = np.array(["a", "b", "b", "a"], dtype=np.object_)
+        ser = ser.cat.remove_unused_categories()
+        tm.assert_index_equal(ser.cat.categories, exp_categories)
+        tm.assert_numpy_array_equal(ser.values.__array__(), exp_values)
+        tm.assert_numpy_array_equal(ser.__array__(), exp_values)
+
+        # This method is likely to be confused, so test that it raises an error
+        # on wrong inputs:
+        msg = "'Series' object has no attribute 'set_categories'"
+        with pytest.raises(AttributeError, match=msg):
+            ser.set_categories([4, 3, 2, 1])
+
+        # right: ser.cat.set_categories([4,3,2,1])
+
+        # GH#18862 (let Series.cat.rename_categories take callables)
+        ser = Series(Categorical(["a", "b", "c", "a"], ordered=True))
+        result = ser.cat.rename_categories(lambda x: x.upper())
+        expected = Series(
+            Categorical(["A", "B", "C", "A"], categories=["A", "B", "C"], ordered=True)
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "idx",
+        [
+            date_range("1/1/2015", periods=5),
+            date_range("1/1/2015", periods=5, tz="MET"),
+            period_range("1/1/2015", freq="D", periods=5),
+            timedelta_range("1 days", "10 days"),
+        ],
+    )
+    def test_dt_accessor_api_for_categorical(self, idx):
+        # https://github.com/pandas-dev/pandas/issues/10661
+
+        ser = Series(idx)
+        cat = ser.astype("category")
+
+        # only testing field (like .day)
+        # and bool (is_month_start)
+        attr_names = type(ser._values)._datetimelike_ops
+
+        assert isinstance(cat.dt, Properties)
+
+        special_func_defs = [
+            ("strftime", ("%Y-%m-%d",), {}),
+            ("round", ("D",), {}),
+            ("floor", ("D",), {}),
+            ("ceil", ("D",), {}),
+            ("asfreq", ("D",), {}),
+            ("as_unit", ("s"), {}),
+        ]
+        if idx.dtype == "M8[ns]":
+            # exclude dt64tz since that is already localized and would raise
+            tup = ("tz_localize", ("UTC",), {})
+            special_func_defs.append(tup)
+        elif idx.dtype.kind == "M":
+            # exclude dt64 since that is not localized so would raise
+            tup = ("tz_convert", ("EST",), {})
+            special_func_defs.append(tup)
+
+        _special_func_names = [f[0] for f in special_func_defs]
+
+        _ignore_names = ["components", "tz_localize", "tz_convert"]
+
+        func_names = [
+            fname
+            for fname in dir(ser.dt)
+            if not (
+                fname.startswith("_")
+                or fname in attr_names
+                or fname in _special_func_names
+                or fname in _ignore_names
+            )
+        ]
+
+        func_defs = [(fname, (), {}) for fname in func_names]
+        func_defs.extend(
+            f_def for f_def in special_func_defs if f_def[0] in dir(ser.dt)
+        )
+
+        for func, args, kwargs in func_defs:
+            warn_cls = []
+            if func == "to_period" and getattr(idx, "tz", None) is not None:
+                # dropping TZ
+                warn_cls.append(UserWarning)
+            elif func == "to_pytimedelta":
+                # GH 57463
+                warn_cls.append(FutureWarning)
+            if warn_cls:
+                warn_cls = tuple(warn_cls)
+            else:
+                warn_cls = None
+            with tm.assert_produces_warning(warn_cls):
+                res = getattr(cat.dt, func)(*args, **kwargs)
+                exp = getattr(ser.dt, func)(*args, **kwargs)
+
+            tm.assert_equal(res, exp)
+
+        for attr in attr_names:
+            res = getattr(cat.dt, attr)
+            exp = getattr(ser.dt, attr)
+
+            tm.assert_equal(res, exp)
+
+    def test_dt_accessor_api_for_categorical_invalid(self):
+        invalid = Series([1, 2, 3]).astype("category")
+        msg = "Can only use .dt accessor with datetimelike"
+
+        with pytest.raises(AttributeError, match=msg):
+            invalid.dt
+        assert not hasattr(invalid, "str")
+
+    def test_set_categories_setitem(self):
+        # GH#43334
+
+        df = DataFrame({"Survived": [1, 0, 1], "Sex": [0, 1, 1]}, dtype="category")
+
+        df["Survived"] = df["Survived"].cat.rename_categories(["No", "Yes"])
+        df["Sex"] = df["Sex"].cat.rename_categories(["female", "male"])
+
+        # values should not be coerced to NaN
+        assert list(df["Sex"]) == ["female", "male", "male"]
+        assert list(df["Survived"]) == ["Yes", "No", "Yes"]
+
+        df["Sex"] = Categorical(df["Sex"], categories=["female", "male"], ordered=False)
+        df["Survived"] = Categorical(
+            df["Survived"], categories=["No", "Yes"], ordered=False
+        )
+
+        # values should not be coerced to NaN
+        assert list(df["Sex"]) == ["female", "male", "male"]
+        assert list(df["Survived"]) == ["Yes", "No", "Yes"]
+
+    def test_categorical_of_booleans_is_boolean(self):
+        # https://github.com/pandas-dev/pandas/issues/46313
+        df = DataFrame(
+            {"int_cat": [1, 2, 3], "bool_cat": [True, False, False]}, dtype="category"
+        )
+        value = df["bool_cat"].cat.categories.dtype
+        expected = np.dtype(np.bool_)
+        assert value is expected
diff --git a/pandas/pandas/tests/series/accessors/test_dt_accessor.py b/pandas/pandas/tests/series/accessors/test_dt_accessor.py
new file mode 100644
index 0000000000000..2c441a6ed91c1
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_dt_accessor.py
@@ -0,0 +1,830 @@
+import calendar
+from datetime import (
+    date,
+    datetime,
+    time,
+)
+import locale
+import unicodedata
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.timezones import maybe_get_tz
+
+from pandas.core.dtypes.common import (
+    is_integer_dtype,
+    is_list_like,
+)
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Period,
+    PeriodIndex,
+    Series,
+    StringDtype,
+    TimedeltaIndex,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    DatetimeArray,
+    PeriodArray,
+    TimedeltaArray,
+)
+
+ok_for_period = PeriodArray._datetimelike_ops
+ok_for_period_methods = ["strftime", "to_timestamp", "asfreq"]
+ok_for_dt = DatetimeArray._datetimelike_ops
+ok_for_dt_methods = [
+    "to_period",
+    "to_pydatetime",
+    "tz_localize",
+    "tz_convert",
+    "normalize",
+    "strftime",
+    "round",
+    "floor",
+    "ceil",
+    "day_name",
+    "month_name",
+    "isocalendar",
+    "as_unit",
+]
+ok_for_td = TimedeltaArray._datetimelike_ops
+ok_for_td_methods = [
+    "components",
+    "to_pytimedelta",
+    "total_seconds",
+    "round",
+    "floor",
+    "ceil",
+    "as_unit",
+]
+
+
+def get_dir(ser):
+    # check limited display api
+    results = [r for r in ser.dt.__dir__() if not r.startswith("_")]
+    return sorted(set(results))
+
+
+class TestSeriesDatetimeValues:
+    def _compare(self, ser, name):
+        # GH 7207, 11128
+        # test .dt namespace accessor
+
+        def get_expected(ser, prop):
+            result = getattr(Index(ser._values), prop)
+            if isinstance(result, np.ndarray):
+                if is_integer_dtype(result):
+                    result = result.astype("int64")
+            elif not is_list_like(result) or isinstance(result, DataFrame):
+                return result
+            return Series(result, index=ser.index, name=ser.name)
+
+        left = getattr(ser.dt, name)
+        right = get_expected(ser, name)
+        if not (is_list_like(left) and is_list_like(right)):
+            assert left == right
+        elif isinstance(left, DataFrame):
+            tm.assert_frame_equal(left, right)
+        else:
+            tm.assert_series_equal(left, right)
+
+    @pytest.mark.parametrize("freq", ["D", "s", "ms"])
+    def test_dt_namespace_accessor_datetime64(self, freq):
+        # GH#7207, GH#11128
+        # test .dt namespace accessor
+
+        # datetimeindex
+        dti = date_range("20130101", periods=5, freq=freq)
+        ser = Series(dti, name="xxx")
+
+        for prop in ok_for_dt:
+            # we test freq below
+            if prop != "freq":
+                self._compare(ser, prop)
+
+        for prop in ok_for_dt_methods:
+            getattr(ser.dt, prop)
+
+        result = ser.dt.to_pydatetime()
+        assert isinstance(result, Series)
+        assert result.dtype == object
+
+        result = ser.dt.tz_localize("US/Eastern")
+        exp_values = DatetimeIndex(ser.values).tz_localize("US/Eastern")
+        expected = Series(exp_values, index=ser.index, name="xxx")
+        tm.assert_series_equal(result, expected)
+
+        tz_result = result.dt.tz
+        assert str(tz_result) == "US/Eastern"
+        freq_result = ser.dt.freq
+        assert freq_result == DatetimeIndex(ser.values, freq="infer").freq
+
+        # let's localize, then convert
+        result = ser.dt.tz_localize("UTC").dt.tz_convert("US/Eastern")
+        exp_values = (
+            DatetimeIndex(ser.values).tz_localize("UTC").tz_convert("US/Eastern")
+        )
+        expected = Series(exp_values, index=ser.index, name="xxx")
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_namespace_accessor_datetime64tz(self):
+        # GH#7207, GH#11128
+        # test .dt namespace accessor
+
+        # datetimeindex with tz
+        dti = date_range("20130101", periods=5, tz="US/Eastern")
+        ser = Series(dti, name="xxx")
+        for prop in ok_for_dt:
+            # we test freq below
+            if prop != "freq":
+                self._compare(ser, prop)
+
+        for prop in ok_for_dt_methods:
+            getattr(ser.dt, prop)
+
+        result = ser.dt.to_pydatetime()
+        assert isinstance(result, Series)
+        assert result.dtype == object
+
+        result = ser.dt.tz_convert("CET")
+        expected = Series(ser._values.tz_convert("CET"), index=ser.index, name="xxx")
+        tm.assert_series_equal(result, expected)
+
+        tz_result = result.dt.tz
+        assert str(tz_result) == "CET"
+        freq_result = ser.dt.freq
+        assert freq_result == DatetimeIndex(ser.values, freq="infer").freq
+
+    def test_dt_namespace_accessor_timedelta(self):
+        # GH#7207, GH#11128
+        # test .dt namespace accessor
+
+        # timedelta index
+        cases = [
+            Series(
+                timedelta_range("1 day", periods=5), index=list("abcde"), name="xxx"
+            ),
+            Series(timedelta_range("1 day 01:23:45", periods=5, freq="s"), name="xxx"),
+            Series(
+                timedelta_range("2 days 01:23:45.012345", periods=5, freq="ms"),
+                name="xxx",
+            ),
+        ]
+        for ser in cases:
+            for prop in ok_for_td:
+                # we test freq below
+                if prop != "freq":
+                    self._compare(ser, prop)
+
+            for prop in ok_for_td_methods:
+                getattr(ser.dt, prop)
+
+            result = ser.dt.components
+            assert isinstance(result, DataFrame)
+            tm.assert_index_equal(result.index, ser.index)
+
+            msg = "The behavior of TimedeltaProperties.to_pytimedelta is deprecated"
+            with tm.assert_produces_warning(FutureWarning, match=msg):
+                result = ser.dt.to_pytimedelta()
+            assert isinstance(result, np.ndarray)
+            assert result.dtype == object
+
+            result = ser.dt.total_seconds()
+            assert isinstance(result, Series)
+            assert result.dtype == "float64"
+
+            freq_result = ser.dt.freq
+            assert freq_result == TimedeltaIndex(ser.values, freq="infer").freq
+
+    def test_dt_namespace_accessor_period(self):
+        # GH#7207, GH#11128
+        # test .dt namespace accessor
+
+        # periodindex
+        pi = period_range("20130101", periods=5, freq="D")
+        ser = Series(pi, name="xxx")
+
+        for prop in ok_for_period:
+            # we test freq below
+            if prop != "freq":
+                self._compare(ser, prop)
+
+        for prop in ok_for_period_methods:
+            getattr(ser.dt, prop)
+
+        freq_result = ser.dt.freq
+        assert freq_result == PeriodIndex(ser.values).freq
+
+    def test_dt_namespace_accessor_index_and_values(self):
+        # both
+        index = date_range("20130101", periods=3, freq="D")
+        dti = date_range("20140204", periods=3, freq="s")
+        ser = Series(dti, index=index, name="xxx")
+        exp = Series(
+            np.array([2014, 2014, 2014], dtype="int32"), index=index, name="xxx"
+        )
+        tm.assert_series_equal(ser.dt.year, exp)
+
+        exp = Series(np.array([2, 2, 2], dtype="int32"), index=index, name="xxx")
+        tm.assert_series_equal(ser.dt.month, exp)
+
+        exp = Series(np.array([0, 1, 2], dtype="int32"), index=index, name="xxx")
+        tm.assert_series_equal(ser.dt.second, exp)
+
+        exp = Series([ser.iloc[0]] * 3, index=index, name="xxx")
+        tm.assert_series_equal(ser.dt.normalize(), exp)
+
+    def test_dt_accessor_limited_display_api(self):
+        # tznaive
+        ser = Series(date_range("20130101", periods=5, freq="D"), name="xxx")
+        results = get_dir(ser)
+        tm.assert_almost_equal(results, sorted(set(ok_for_dt + ok_for_dt_methods)))
+
+        # tzaware
+        ser = Series(date_range("2015-01-01", "2016-01-01", freq="min"), name="xxx")
+        ser = ser.dt.tz_localize("UTC").dt.tz_convert("America/Chicago")
+        results = get_dir(ser)
+        tm.assert_almost_equal(results, sorted(set(ok_for_dt + ok_for_dt_methods)))
+
+        # Period
+        idx = period_range("20130101", periods=5, freq="D", name="xxx")
+        ser = Series(idx)
+        results = get_dir(ser)
+        tm.assert_almost_equal(
+            results, sorted(set(ok_for_period + ok_for_period_methods))
+        )
+
+    def test_dt_accessor_ambiguous_freq_conversions(self):
+        # GH#11295
+        # ambiguous time error on the conversions
+        ser = Series(date_range("2015-01-01", "2016-01-01", freq="min"), name="xxx")
+        ser = ser.dt.tz_localize("UTC").dt.tz_convert("America/Chicago")
+
+        exp_values = date_range(
+            "2015-01-01", "2016-01-01", freq="min", tz="UTC"
+        ).tz_convert("America/Chicago")
+        # freq not preserved by tz_localize above
+        exp_values = exp_values._with_freq(None)
+        expected = Series(exp_values, name="xxx")
+        tm.assert_series_equal(ser, expected)
+
+    def test_dt_accessor_not_writeable(self):
+        # no setting allowed
+        ser = Series(date_range("20130101", periods=5, freq="D"), name="xxx")
+        with pytest.raises(ValueError, match="modifications"):
+            ser.dt.hour = 5
+
+        # trying to set a copy
+        with tm.raises_chained_assignment_error():
+            ser.dt.hour[0] = 5
+
+    @pytest.mark.parametrize(
+        "method, dates",
+        [
+            ["round", ["2012-01-02", "2012-01-02", "2012-01-01"]],
+            ["floor", ["2012-01-01", "2012-01-01", "2012-01-01"]],
+            ["ceil", ["2012-01-02", "2012-01-02", "2012-01-02"]],
+        ],
+    )
+    def test_dt_round(self, method, dates):
+        # round
+        ser = Series(
+            pd.to_datetime(
+                ["2012-01-01 13:00:00", "2012-01-01 12:01:00", "2012-01-01 08:00:00"]
+            ),
+            name="xxx",
+        )
+        result = getattr(ser.dt, method)("D")
+        expected = Series(pd.to_datetime(dates), name="xxx")
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_round_tz(self):
+        ser = Series(
+            pd.to_datetime(
+                ["2012-01-01 13:00:00", "2012-01-01 12:01:00", "2012-01-01 08:00:00"]
+            ),
+            name="xxx",
+        )
+        result = ser.dt.tz_localize("UTC").dt.tz_convert("US/Eastern").dt.round("D")
+
+        exp_values = pd.to_datetime(
+            ["2012-01-01", "2012-01-01", "2012-01-01"]
+        ).tz_localize("US/Eastern")
+        expected = Series(exp_values, name="xxx")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["ceil", "round", "floor"])
+    def test_dt_round_tz_ambiguous(self, method):
+        # GH 18946 round near "fall back" DST
+        df1 = DataFrame(
+            [
+                pd.to_datetime("2017-10-29 02:00:00+02:00", utc=True),
+                pd.to_datetime("2017-10-29 02:00:00+01:00", utc=True),
+                pd.to_datetime("2017-10-29 03:00:00+01:00", utc=True),
+            ],
+            columns=["date"],
+        )
+        df1["date"] = df1["date"].dt.tz_convert("Europe/Madrid")
+        # infer
+        result = getattr(df1.date.dt, method)("h", ambiguous="infer")
+        expected = df1["date"]
+        tm.assert_series_equal(result, expected)
+
+        # bool-array
+        result = getattr(df1.date.dt, method)("h", ambiguous=[True, False, False])
+        tm.assert_series_equal(result, expected)
+
+        # NaT
+        result = getattr(df1.date.dt, method)("h", ambiguous="NaT")
+        expected = df1["date"].copy()
+        expected.iloc[0:2] = pd.NaT
+        tm.assert_series_equal(result, expected)
+
+        # raise
+        with tm.external_error_raised(ValueError):
+            getattr(df1.date.dt, method)("h", ambiguous="raise")
+
+    @pytest.mark.parametrize(
+        "method, ts_str, freq",
+        [
+            ["ceil", "2018-03-11 01:59:00-0600", "5min"],
+            ["round", "2018-03-11 01:59:00-0600", "5min"],
+            ["floor", "2018-03-11 03:01:00-0500", "2h"],
+        ],
+    )
+    def test_dt_round_tz_nonexistent(self, method, ts_str, freq):
+        # GH 23324 round near "spring forward" DST
+        ser = Series([pd.Timestamp(ts_str, tz="America/Chicago")])
+        result = getattr(ser.dt, method)(freq, nonexistent="shift_forward")
+        expected = Series([pd.Timestamp("2018-03-11 03:00:00", tz="America/Chicago")])
+        tm.assert_series_equal(result, expected)
+
+        result = getattr(ser.dt, method)(freq, nonexistent="NaT")
+        expected = Series([pd.NaT]).dt.tz_localize(result.dt.tz)
+        tm.assert_series_equal(result, expected)
+
+        with pytest.raises(ValueError, match="2018-03-11 02:00:00"):
+            getattr(ser.dt, method)(freq, nonexistent="raise")
+
+    @pytest.mark.parametrize("freq", ["ns", "us", "1000us"])
+    def test_dt_round_nonnano_higher_resolution_no_op(self, freq):
+        # GH 52761
+        ser = Series(
+            ["2020-05-31 08:00:00", "2000-12-31 04:00:05", "1800-03-14 07:30:20"],
+            dtype="datetime64[ms]",
+        )
+        expected = ser.copy()
+        result = ser.dt.round(freq)
+        tm.assert_series_equal(result, expected)
+
+        assert not np.shares_memory(ser.array._ndarray, result.array._ndarray)
+
+    def test_dt_namespace_accessor_categorical(self):
+        # GH 19468
+        dti = DatetimeIndex(["20171111", "20181212"]).repeat(2)
+        ser = Series(pd.Categorical(dti), name="foo")
+        result = ser.dt.year
+        expected = Series([2017, 2017, 2018, 2018], dtype="int32", name="foo")
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_tz_localize_categorical(self, tz_aware_fixture):
+        # GH 27952
+        tz = tz_aware_fixture
+        datetimes = Series(
+            ["2019-01-01", "2019-01-01", "2019-01-02"], dtype="datetime64[ns]"
+        )
+        categorical = datetimes.astype("category")
+        result = categorical.dt.tz_localize(tz)
+        expected = datetimes.dt.tz_localize(tz)
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_tz_convert_categorical(self, tz_aware_fixture):
+        # GH 27952
+        tz = tz_aware_fixture
+        datetimes = Series(
+            ["2019-01-01", "2019-01-01", "2019-01-02"], dtype="datetime64[ns, MET]"
+        )
+        categorical = datetimes.astype("category")
+        result = categorical.dt.tz_convert(tz)
+        expected = datetimes.dt.tz_convert(tz)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("accessor", ["year", "month", "day"])
+    def test_dt_other_accessors_categorical(self, accessor):
+        # GH 27952
+        datetimes = Series(
+            ["2018-01-01", "2018-01-01", "2019-01-02"], dtype="datetime64[ns]"
+        )
+        categorical = datetimes.astype("category")
+        result = getattr(categorical.dt, accessor)
+        expected = getattr(datetimes.dt, accessor)
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_accessor_no_new_attributes(self):
+        # https://github.com/pandas-dev/pandas/issues/10673
+        ser = Series(date_range("20130101", periods=5, freq="D"))
+        with pytest.raises(AttributeError, match="You cannot add any new attribute"):
+            ser.dt.xlabel = "a"
+
+    # error: Unsupported operand types for + ("List[None]" and "List[str]")
+    @pytest.mark.parametrize(
+        "time_locale",
+        [None] + tm.get_locales(),  # type: ignore[operator]
+    )
+    def test_dt_accessor_datetime_name_accessors(self, time_locale):
+        # Test Monday -> Sunday and January -> December, in that sequence
+        if time_locale is None:
+            # If the time_locale is None, day-name and month_name should
+            # return the english attributes
+            expected_days = [
+                "Monday",
+                "Tuesday",
+                "Wednesday",
+                "Thursday",
+                "Friday",
+                "Saturday",
+                "Sunday",
+            ]
+            expected_months = [
+                "January",
+                "February",
+                "March",
+                "April",
+                "May",
+                "June",
+                "July",
+                "August",
+                "September",
+                "October",
+                "November",
+                "December",
+            ]
+        else:
+            with tm.set_locale(time_locale, locale.LC_TIME):
+                expected_days = calendar.day_name[:]
+                expected_months = calendar.month_name[1:]
+
+        ser = Series(date_range(freq="D", start=datetime(1998, 1, 1), periods=365))
+        english_days = [
+            "Monday",
+            "Tuesday",
+            "Wednesday",
+            "Thursday",
+            "Friday",
+            "Saturday",
+            "Sunday",
+        ]
+        for day, name, eng_name in zip(range(4, 11), expected_days, english_days):
+            name = name.capitalize()
+            assert ser.dt.day_name(locale=time_locale)[day] == name
+            assert ser.dt.day_name(locale=None)[day] == eng_name
+        ser = pd.concat([ser, Series([pd.NaT])])
+        assert np.isnan(ser.dt.day_name(locale=time_locale).iloc[-1])
+
+        ser = Series(date_range(freq="ME", start="2012", end="2013"))
+        result = ser.dt.month_name(locale=time_locale)
+        expected = Series([month.capitalize() for month in expected_months])
+
+        # work around https://github.com/pandas-dev/pandas/issues/22342
+        result = result.str.normalize("NFD")
+        expected = expected.str.normalize("NFD")
+
+        tm.assert_series_equal(result, expected)
+
+        for s_date, expected in zip(ser, expected_months):
+            result = s_date.month_name(locale=time_locale)
+            expected = expected.capitalize()
+
+            result = unicodedata.normalize("NFD", result)
+            expected = unicodedata.normalize("NFD", expected)
+
+            assert result == expected
+
+        ser = pd.concat([ser, Series([pd.NaT])])
+        assert np.isnan(ser.dt.month_name(locale=time_locale).iloc[-1])
+
+    def test_strftime(self):
+        # GH 10086
+        ser = Series(date_range("20130101", periods=5))
+        result = ser.dt.strftime("%Y/%m/%d")
+        expected = Series(
+            ["2013/01/01", "2013/01/02", "2013/01/03", "2013/01/04", "2013/01/05"]
+        )
+        tm.assert_series_equal(result, expected)
+
+        ser = Series(date_range("2015-02-03 11:22:33.4567", periods=5))
+        result = ser.dt.strftime("%Y/%m/%d %H-%M-%S")
+        expected = Series(
+            [
+                "2015/02/03 11-22-33",
+                "2015/02/04 11-22-33",
+                "2015/02/05 11-22-33",
+                "2015/02/06 11-22-33",
+                "2015/02/07 11-22-33",
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+        ser = Series(period_range("20130101", periods=5))
+        result = ser.dt.strftime("%Y/%m/%d")
+        expected = Series(
+            ["2013/01/01", "2013/01/02", "2013/01/03", "2013/01/04", "2013/01/05"]
+        )
+        tm.assert_series_equal(result, expected)
+
+        ser = Series(period_range("2015-02-03 11:22:33.4567", periods=5, freq="s"))
+        result = ser.dt.strftime("%Y/%m/%d %H-%M-%S")
+        expected = Series(
+            [
+                "2015/02/03 11-22-33",
+                "2015/02/03 11-22-34",
+                "2015/02/03 11-22-35",
+                "2015/02/03 11-22-36",
+                "2015/02/03 11-22-37",
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_strftime_dt64_days(self):
+        ser = Series(date_range("20130101", periods=5))
+        ser.iloc[0] = pd.NaT
+        result = ser.dt.strftime("%Y/%m/%d")
+        expected = Series(
+            [np.nan, "2013/01/02", "2013/01/03", "2013/01/04", "2013/01/05"]
+        )
+        tm.assert_series_equal(result, expected)
+
+        datetime_index = date_range("20150301", periods=5)
+        result = datetime_index.strftime("%Y/%m/%d")
+
+        expected = Index(
+            ["2015/03/01", "2015/03/02", "2015/03/03", "2015/03/04", "2015/03/05"],
+        )
+        # dtype may be S10 or U10 depending on python version
+        tm.assert_index_equal(result, expected)
+
+    def test_strftime_period_days(self, using_infer_string):
+        period_index = period_range("20150301", periods=5)
+        result = period_index.strftime("%Y/%m/%d")
+        expected = Index(
+            ["2015/03/01", "2015/03/02", "2015/03/03", "2015/03/04", "2015/03/05"],
+            dtype="=U10",
+        )
+        if using_infer_string:
+            expected = expected.astype(StringDtype(na_value=np.nan))
+        tm.assert_index_equal(result, expected)
+
+    def test_strftime_dt64_microsecond_resolution(self):
+        ser = Series([datetime(2013, 1, 1, 2, 32, 59), datetime(2013, 1, 2, 14, 32, 1)])
+        result = ser.dt.strftime("%Y-%m-%d %H:%M:%S")
+        expected = Series(["2013-01-01 02:32:59", "2013-01-02 14:32:01"])
+        tm.assert_series_equal(result, expected)
+
+    def test_strftime_period_hours(self):
+        ser = Series(period_range("20130101", periods=4, freq="h"))
+        result = ser.dt.strftime("%Y/%m/%d %H:%M:%S")
+        expected = Series(
+            [
+                "2013/01/01 00:00:00",
+                "2013/01/01 01:00:00",
+                "2013/01/01 02:00:00",
+                "2013/01/01 03:00:00",
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_strftime_period_minutes(self):
+        ser = Series(period_range("20130101", periods=4, freq="ms"))
+        result = ser.dt.strftime("%Y/%m/%d %H:%M:%S.%l")
+        expected = Series(
+            [
+                "2013/01/01 00:00:00.000",
+                "2013/01/01 00:00:00.001",
+                "2013/01/01 00:00:00.002",
+                "2013/01/01 00:00:00.003",
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            DatetimeIndex(["2019-01-01", pd.NaT]),
+            PeriodIndex(["2019-01-01", pd.NaT], dtype="period[D]"),
+        ],
+    )
+    def test_strftime_nat(self, data):
+        # GH 29578
+        ser = Series(data)
+        result = ser.dt.strftime("%Y-%m-%d")
+        expected = Series(["2019-01-01", np.nan])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data", [DatetimeIndex([pd.NaT]), PeriodIndex([pd.NaT], dtype="period[D]")]
+    )
+    def test_strftime_all_nat(self, data):
+        # https://github.com/pandas-dev/pandas/issues/45858
+        ser = Series(data)
+        with tm.assert_produces_warning(None):
+            result = ser.dt.strftime("%Y-%m-%d")
+        expected = Series([np.nan], dtype="str")
+        tm.assert_series_equal(result, expected)
+
+    def test_valid_dt_with_missing_values(self):
+        # GH 8689
+        ser = Series(date_range("20130101", periods=5, freq="D"))
+        ser.iloc[2] = pd.NaT
+
+        for attr in ["microsecond", "nanosecond", "second", "minute", "hour", "day"]:
+            expected = getattr(ser.dt, attr).copy()
+            expected.iloc[2] = np.nan
+            result = getattr(ser.dt, attr)
+            tm.assert_series_equal(result, expected)
+
+        result = ser.dt.date
+        expected = Series(
+            [
+                date(2013, 1, 1),
+                date(2013, 1, 2),
+                pd.NaT,
+                date(2013, 1, 4),
+                date(2013, 1, 5),
+            ],
+            dtype="object",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = ser.dt.time
+        expected = Series([time(0), time(0), pd.NaT, time(0), time(0)], dtype="object")
+        tm.assert_series_equal(result, expected)
+
+    def test_dt_accessor_api(self):
+        # GH 9322
+        from pandas.core.indexes.accessors import (
+            CombinedDatetimelikeProperties,
+            DatetimeProperties,
+        )
+
+        assert Series.dt is CombinedDatetimelikeProperties
+
+        ser = Series(date_range("2000-01-01", periods=3))
+        assert isinstance(ser.dt, DatetimeProperties)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            np.arange(5),
+            list("abcde"),
+            np.random.default_rng(2).standard_normal(5),
+        ],
+    )
+    def test_dt_accessor_invalid(self, data):
+        # GH#9322 check that series with incorrect dtypes don't have attr
+        ser = Series(data)
+        with pytest.raises(AttributeError, match="only use .dt accessor"):
+            ser.dt
+        assert not hasattr(ser, "dt")
+
+    def test_dt_accessor_updates_on_inplace(self):
+        ser = Series(date_range("2018-01-01", periods=10))
+        ser[2] = None
+        return_value = ser.fillna(pd.Timestamp("2018-01-01"), inplace=True)
+        assert return_value is None
+        result = ser.dt.date
+        assert result[0] == result[2]
+
+    def test_date_tz(self):
+        # GH11757
+        rng = DatetimeIndex(
+            ["2014-04-04 23:56", "2014-07-18 21:24", "2015-11-22 22:14"],
+            tz="US/Eastern",
+        )
+        ser = Series(rng)
+        expected = Series([date(2014, 4, 4), date(2014, 7, 18), date(2015, 11, 22)])
+        tm.assert_series_equal(ser.dt.date, expected)
+        tm.assert_series_equal(ser.apply(lambda x: x.date()), expected)
+
+    def test_dt_timetz_accessor(self, tz_naive_fixture):
+        # GH21358
+        tz = maybe_get_tz(tz_naive_fixture)
+
+        dtindex = DatetimeIndex(
+            ["2014-04-04 23:56", "2014-07-18 21:24", "2015-11-22 22:14"], tz=tz
+        )
+        ser = Series(dtindex)
+        expected = Series(
+            [time(23, 56, tzinfo=tz), time(21, 24, tzinfo=tz), time(22, 14, tzinfo=tz)]
+        )
+        result = ser.dt.timetz
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "input_series, expected_output",
+        [
+            [["2020-01-01"], [[2020, 1, 3]]],
+            [[pd.NaT], [[np.nan, np.nan, np.nan]]],
+            [["2019-12-31", "2019-12-29"], [[2020, 1, 2], [2019, 52, 7]]],
+            [["2010-01-01", pd.NaT], [[2009, 53, 5], [np.nan, np.nan, np.nan]]],
+            # see GH#36032
+            [["2016-01-08", "2016-01-04"], [[2016, 1, 5], [2016, 1, 1]]],
+            [["2016-01-07", "2016-01-01"], [[2016, 1, 4], [2015, 53, 5]]],
+        ],
+    )
+    def test_isocalendar(self, input_series, expected_output):
+        result = pd.to_datetime(Series(input_series)).dt.isocalendar()
+        expected_frame = DataFrame(
+            expected_output, columns=["year", "week", "day"], dtype="UInt32"
+        )
+        tm.assert_frame_equal(result, expected_frame)
+
+    def test_hour_index(self):
+        dt_series = Series(
+            date_range(start="2021-01-01", periods=5, freq="h"),
+            index=[2, 6, 7, 8, 11],
+            dtype="category",
+        )
+        result = dt_series.dt.hour
+        expected = Series(
+            [0, 1, 2, 3, 4],
+            dtype="int32",
+            index=[2, 6, 7, 8, 11],
+        )
+        tm.assert_series_equal(result, expected)
+
+
+class TestSeriesPeriodValuesDtAccessor:
+    @pytest.mark.parametrize(
+        "input_vals",
+        [
+            [Period("2016-01", freq="M"), Period("2016-02", freq="M")],
+            [Period("2016-01-01", freq="D"), Period("2016-01-02", freq="D")],
+            [
+                Period("2016-01-01 00:00:00", freq="h"),
+                Period("2016-01-01 01:00:00", freq="h"),
+            ],
+            [
+                Period("2016-01-01 00:00:00", freq="M"),
+                Period("2016-01-01 00:01:00", freq="M"),
+            ],
+            [
+                Period("2016-01-01 00:00:00", freq="s"),
+                Period("2016-01-01 00:00:01", freq="s"),
+            ],
+        ],
+    )
+    def test_end_time_timevalues(self, input_vals):
+        # GH#17157
+        # Check that the time part of the Period is adjusted by end_time
+        # when using the dt accessor on a Series
+        dtype = pd.PeriodDtype(input_vals[0].freq)
+        input_vals = PeriodArray._from_sequence(np.asarray(input_vals), dtype=dtype)
+
+        ser = Series(input_vals)
+        result = ser.dt.end_time
+        expected = ser.apply(lambda x: x.end_time)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("input_vals", [("2001"), ("NaT")])
+    def test_to_period(self, input_vals):
+        # GH#21205
+        expected = Series([input_vals], dtype="Period[D]")
+        result = Series([input_vals], dtype="datetime64[ns]").dt.to_period("D")
+        tm.assert_series_equal(result, expected)
+
+
+def test_normalize_pre_epoch_dates():
+    # GH: 36294
+    ser = pd.to_datetime(Series(["1969-01-01 09:00:00", "2016-01-01 09:00:00"]))
+    result = ser.dt.normalize()
+    expected = pd.to_datetime(Series(["1969-01-01", "2016-01-01"]))
+    tm.assert_series_equal(result, expected)
+
+
+def test_day_attribute_non_nano_beyond_int32():
+    # GH 52386
+    data = np.array(
+        [
+            136457654736252,
+            134736784364431,
+            245345345545332,
+            223432411,
+            2343241,
+            3634548734,
+            23234,
+        ],
+        dtype="timedelta64[s]",
+    )
+    ser = Series(data)
+    result = ser.dt.days
+    expected = Series([1579371003, 1559453522, 2839645203, 2586, 27, 42066, 0])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/accessors/test_list_accessor.py b/pandas/pandas/tests/series/accessors/test_list_accessor.py
new file mode 100644
index 0000000000000..bec8ca13a2f5f
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_list_accessor.py
@@ -0,0 +1,160 @@
+import re
+
+import pytest
+
+from pandas import (
+    ArrowDtype,
+    Series,
+)
+import pandas._testing as tm
+
+pa = pytest.importorskip("pyarrow")
+
+from pandas.compat import pa_version_under11p0
+
+
+@pytest.mark.parametrize(
+    "list_dtype",
+    (
+        pa.list_(pa.int64()),
+        pa.list_(pa.int64(), list_size=3),
+        pa.large_list(pa.int64()),
+    ),
+)
+def test_list_getitem(list_dtype):
+    ser = Series(
+        [[1, 2, 3], [4, None, 5], None],
+        dtype=ArrowDtype(list_dtype),
+        name="a",
+    )
+    actual = ser.list[1]
+    expected = Series([2, None, None], dtype="int64[pyarrow]", name="a")
+    tm.assert_series_equal(actual, expected)
+
+
+def test_list_getitem_index():
+    # GH 58425
+    ser = Series(
+        [[1, 2, 3], [4, None, 5], None],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+        index=[1, 3, 7],
+        name="a",
+    )
+    actual = ser.list[1]
+    expected = Series(
+        [2, None, None],
+        dtype="int64[pyarrow]",
+        index=[1, 3, 7],
+        name="a",
+    )
+    tm.assert_series_equal(actual, expected)
+
+
+def test_list_getitem_slice():
+    ser = Series(
+        [[1, 2, 3], [4, None, 5], None],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+        index=[1, 3, 7],
+        name="a",
+    )
+    if pa_version_under11p0:
+        with pytest.raises(
+            NotImplementedError, match="List slice not supported by pyarrow "
+        ):
+            ser.list[1:None:None]
+    else:
+        actual = ser.list[1:None:None]
+        expected = Series(
+            [[2, 3], [None, 5], None],
+            dtype=ArrowDtype(pa.list_(pa.int64())),
+            index=[1, 3, 7],
+            name="a",
+        )
+        tm.assert_series_equal(actual, expected)
+
+
+def test_list_len():
+    ser = Series(
+        [[1, 2, 3], [4, None], None],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+        name="a",
+    )
+    actual = ser.list.len()
+    expected = Series([3, 2, None], dtype=ArrowDtype(pa.int32()), name="a")
+    tm.assert_series_equal(actual, expected)
+
+
+def test_list_flatten():
+    ser = Series(
+        [[1, 2, 3], None, [4, None], [], [7, 8]],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+        name="a",
+    )
+    actual = ser.list.flatten()
+    expected = Series(
+        [1, 2, 3, 4, None, 7, 8],
+        dtype=ArrowDtype(pa.int64()),
+        index=[0, 0, 0, 2, 2, 4, 4],
+        name="a",
+    )
+    tm.assert_series_equal(actual, expected)
+
+
+def test_list_getitem_slice_invalid():
+    ser = Series(
+        [[1, 2, 3], [4, None, 5], None],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+    )
+    if pa_version_under11p0:
+        with pytest.raises(
+            NotImplementedError, match="List slice not supported by pyarrow "
+        ):
+            ser.list[1:None:0]
+    else:
+        with pytest.raises(pa.lib.ArrowInvalid, match=re.escape("`step` must be >= 1")):
+            ser.list[1:None:0]
+
+
+def test_list_accessor_non_list_dtype():
+    ser = Series(
+        [1, 2, 4],
+        dtype=ArrowDtype(pa.int64()),
+    )
+    with pytest.raises(
+        AttributeError,
+        match=re.escape(
+            "Can only use the '.list' accessor with 'list[pyarrow]' dtype, "
+            "not int64[pyarrow]."
+        ),
+    ):
+        ser.list[1:None:0]
+
+
+@pytest.mark.parametrize(
+    "list_dtype",
+    (
+        pa.list_(pa.int64()),
+        pa.list_(pa.int64(), list_size=3),
+        pa.large_list(pa.int64()),
+    ),
+)
+def test_list_getitem_invalid_index(list_dtype):
+    ser = Series(
+        [[1, 2, 3], [4, None, 5], None],
+        dtype=ArrowDtype(list_dtype),
+    )
+    with pytest.raises(pa.lib.ArrowInvalid, match="Index -1 is out of bounds"):
+        ser.list[-1]
+    with pytest.raises(pa.lib.ArrowInvalid, match="Index 5 is out of bounds"):
+        ser.list[5]
+    with pytest.raises(ValueError, match="key must be an int or slice, got str"):
+        ser.list["abc"]
+
+
+def test_list_accessor_not_iterable():
+    ser = Series(
+        [[1, 2, 3], [4, None], None],
+        dtype=ArrowDtype(pa.list_(pa.int64())),
+    )
+    with pytest.raises(TypeError, match="'ListAccessor' object is not iterable"):
+        iter(ser.list)
diff --git a/pandas/pandas/tests/series/accessors/test_sparse_accessor.py b/pandas/pandas/tests/series/accessors/test_sparse_accessor.py
new file mode 100644
index 0000000000000..118095b5dcdbc
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_sparse_accessor.py
@@ -0,0 +1,9 @@
+from pandas import Series
+
+
+class TestSparseAccessor:
+    def test_sparse_accessor_updates_on_inplace(self):
+        ser = Series([1, 1, 2, 3], dtype="Sparse[int]")
+        return_value = ser.drop([0, 1], inplace=True)
+        assert return_value is None
+        assert ser.sparse.density == 1.0
diff --git a/pandas/pandas/tests/series/accessors/test_str_accessor.py b/pandas/pandas/tests/series/accessors/test_str_accessor.py
new file mode 100644
index 0000000000000..ff530459b78fb
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_str_accessor.py
@@ -0,0 +1,26 @@
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestStrAccessor:
+    def test_str_attribute(self):
+        # GH#9068
+        methods = ["strip", "rstrip", "lstrip"]
+        ser = Series([" jack", "jill ", " jesse ", "frank"])
+        for method in methods:
+            expected = Series([getattr(str, method)(x) for x in ser.values])
+            tm.assert_series_equal(getattr(Series.str, method)(ser.str), expected)
+
+        # str accessor only valid with string values
+        ser = Series(range(5))
+        msg = "Can only use .str accessor with string values, not integer"
+        with pytest.raises(AttributeError, match=msg):
+            ser.str.repeat(2)
+
+    def test_str_accessor_updates_on_inplace(self):
+        ser = Series(list("abc"))
+        return_value = ser.drop([0], inplace=True)
+        assert return_value is None
+        assert len(ser.str.lower()) == 2
diff --git a/pandas/pandas/tests/series/accessors/test_struct_accessor.py b/pandas/pandas/tests/series/accessors/test_struct_accessor.py
new file mode 100644
index 0000000000000..80aea75fda406
--- /dev/null
+++ b/pandas/pandas/tests/series/accessors/test_struct_accessor.py
@@ -0,0 +1,196 @@
+import re
+
+import pytest
+
+from pandas.compat.pyarrow import (
+    pa_version_under11p0,
+    pa_version_under13p0,
+)
+
+from pandas import (
+    ArrowDtype,
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+pa = pytest.importorskip("pyarrow")
+pc = pytest.importorskip("pyarrow.compute")
+
+
+def test_struct_accessor_dtypes():
+    ser = Series(
+        [],
+        dtype=ArrowDtype(
+            pa.struct(
+                [
+                    ("int_col", pa.int64()),
+                    ("string_col", pa.string()),
+                    (
+                        "struct_col",
+                        pa.struct(
+                            [
+                                ("int_col", pa.int64()),
+                                ("float_col", pa.float64()),
+                            ]
+                        ),
+                    ),
+                ]
+            )
+        ),
+    )
+    actual = ser.struct.dtypes
+    expected = Series(
+        [
+            ArrowDtype(pa.int64()),
+            ArrowDtype(pa.string()),
+            ArrowDtype(
+                pa.struct(
+                    [
+                        ("int_col", pa.int64()),
+                        ("float_col", pa.float64()),
+                    ]
+                )
+            ),
+        ],
+        index=Index(["int_col", "string_col", "struct_col"]),
+    )
+    tm.assert_series_equal(actual, expected)
+
+
+@pytest.mark.skipif(pa_version_under13p0, reason="pyarrow>=13.0.0 required")
+def test_struct_accessor_field():
+    index = Index([-100, 42, 123])
+    ser = Series(
+        [
+            {"rice": 1.0, "maize": -1, "wheat": "a"},
+            {"rice": 2.0, "maize": 0, "wheat": "b"},
+            {"rice": 3.0, "maize": 1, "wheat": "c"},
+        ],
+        dtype=ArrowDtype(
+            pa.struct(
+                [
+                    ("rice", pa.float64()),
+                    ("maize", pa.int64()),
+                    ("wheat", pa.string()),
+                ]
+            )
+        ),
+        index=index,
+    )
+    by_name = ser.struct.field("maize")
+    by_name_expected = Series(
+        [-1, 0, 1],
+        dtype=ArrowDtype(pa.int64()),
+        index=index,
+        name="maize",
+    )
+    tm.assert_series_equal(by_name, by_name_expected)
+
+    by_index = ser.struct.field(2)
+    by_index_expected = Series(
+        ["a", "b", "c"],
+        dtype=ArrowDtype(pa.string()),
+        index=index,
+        name="wheat",
+    )
+    tm.assert_series_equal(by_index, by_index_expected)
+
+
+def test_struct_accessor_field_with_invalid_name_or_index():
+    ser = Series([], dtype=ArrowDtype(pa.struct([("field", pa.int64())])))
+
+    with pytest.raises(ValueError, match="name_or_index must be an int, str,"):
+        ser.struct.field(1.1)
+
+
+@pytest.mark.skipif(pa_version_under11p0, reason="pyarrow>=11.0.0 required")
+def test_struct_accessor_explode():
+    index = Index([-100, 42, 123])
+    ser = Series(
+        [
+            {"painted": 1, "snapping": {"sea": "green"}},
+            {"painted": 2, "snapping": {"sea": "leatherback"}},
+            {"painted": 3, "snapping": {"sea": "hawksbill"}},
+        ],
+        dtype=ArrowDtype(
+            pa.struct(
+                [
+                    ("painted", pa.int64()),
+                    ("snapping", pa.struct([("sea", pa.string())])),
+                ]
+            )
+        ),
+        index=index,
+    )
+    actual = ser.struct.explode()
+    expected = DataFrame(
+        {
+            "painted": Series([1, 2, 3], index=index, dtype=ArrowDtype(pa.int64())),
+            "snapping": Series(
+                [{"sea": "green"}, {"sea": "leatherback"}, {"sea": "hawksbill"}],
+                index=index,
+                dtype=ArrowDtype(pa.struct([("sea", pa.string())])),
+            ),
+        },
+    )
+    tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "invalid",
+    [
+        pytest.param(Series([1, 2, 3], dtype="int64"), id="int64"),
+        pytest.param(
+            Series(["a", "b", "c"], dtype="string[pyarrow]"), id="string-pyarrow"
+        ),
+    ],
+)
+def test_struct_accessor_api_for_invalid(invalid):
+    with pytest.raises(
+        AttributeError,
+        match=re.escape(
+            "Can only use the '.struct' accessor with 'struct[pyarrow]' dtype, "
+            f"not {invalid.dtype}."
+        ),
+    ):
+        invalid.struct
+
+
+@pytest.mark.parametrize(
+    ["indices", "name"],
+    [
+        (0, "int_col"),
+        ([1, 2], "str_col"),
+        (pc.field("int_col"), "int_col"),
+        ("int_col", "int_col"),
+        (b"string_col", b"string_col"),
+        ([b"string_col"], "string_col"),
+    ],
+)
+@pytest.mark.skipif(pa_version_under13p0, reason="pyarrow>=13.0.0 required")
+def test_struct_accessor_field_expanded(indices, name):
+    arrow_type = pa.struct(
+        [
+            ("int_col", pa.int64()),
+            (
+                "struct_col",
+                pa.struct(
+                    [
+                        ("int_col", pa.int64()),
+                        ("float_col", pa.float64()),
+                        ("str_col", pa.string()),
+                    ]
+                ),
+            ),
+            (b"string_col", pa.string()),
+        ]
+    )
+
+    data = pa.array([], type=arrow_type)
+    ser = Series(data, dtype=ArrowDtype(arrow_type))
+    expected = pc.struct_field(data, indices)
+    result = ser.struct.field(indices)
+    tm.assert_equal(result.array._pa_array.combine_chunks(), expected)
+    assert result.name == name
diff --git a/pandas/pandas/tests/series/indexing/__init__.py b/pandas/pandas/tests/series/indexing/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/series/indexing/test_datetime.py b/pandas/pandas/tests/series/indexing/test_datetime.py
new file mode 100644
index 0000000000000..97cafc33611ed
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_datetime.py
@@ -0,0 +1,493 @@
+"""
+Also test support for datetime64[ns] in Series / DataFrame
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+import re
+
+from dateutil.tz import (
+    gettz,
+    tzutc,
+)
+import numpy as np
+import pytest
+
+from pandas._libs import index as libindex
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+def test_fancy_getitem():
+    dti = date_range(
+        freq="WOM-1FRI", start=datetime(2005, 1, 1), end=datetime(2010, 1, 1)
+    )
+
+    s = Series(np.arange(len(dti)), index=dti)
+
+    assert s["1/2/2009"] == 48
+    assert s["2009-1-2"] == 48
+    assert s[datetime(2009, 1, 2)] == 48
+    assert s[Timestamp(datetime(2009, 1, 2))] == 48
+    with pytest.raises(KeyError, match=r"^'2009-1-3'$"):
+        s["2009-1-3"]
+    tm.assert_series_equal(
+        s["3/6/2009":"2009-06-05"], s[datetime(2009, 3, 6) : datetime(2009, 6, 5)]
+    )
+
+
+def test_fancy_setitem():
+    dti = date_range(
+        freq="WOM-1FRI", start=datetime(2005, 1, 1), end=datetime(2010, 1, 1)
+    )
+
+    s = Series(np.arange(len(dti)), index=dti)
+
+    s["1/2/2009"] = -2
+    assert s.iloc[48] == -2
+    s["1/2/2009":"2009-06-05"] = -3
+    assert (s[48:54] == -3).all()
+
+
+@pytest.mark.parametrize("tz_source", ["pytz", "dateutil"])
+def test_getitem_setitem_datetime_tz(tz_source):
+    if tz_source == "pytz":
+        pytz = pytest.importorskip(tz_source)
+        tzget = pytz.timezone
+    else:
+        # handle special case for utc in dateutil
+        tzget = lambda x: tzutc() if x == "UTC" else gettz(x)
+
+    N = 50
+    # testing with timezone, GH #2785
+    rng = date_range("1/1/1990", periods=N, freq="h", tz=tzget("US/Eastern"))
+    ts = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+
+    # also test Timestamp tz handling, GH #2789
+    result = ts.copy()
+    result["1990-01-01 09:00:00+00:00"] = 0
+    result["1990-01-01 09:00:00+00:00"] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+    result = ts.copy()
+    result["1990-01-01 03:00:00-06:00"] = 0
+    result["1990-01-01 03:00:00-06:00"] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+    # repeat with datetimes
+    result = ts.copy()
+    result[datetime(1990, 1, 1, 9, tzinfo=tzget("UTC"))] = 0
+    result[datetime(1990, 1, 1, 9, tzinfo=tzget("UTC"))] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+    result = ts.copy()
+    dt = Timestamp(1990, 1, 1, 3).tz_localize(tzget("US/Central"))
+    dt = dt.to_pydatetime()
+    result[dt] = 0
+    result[dt] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+
+def test_getitem_setitem_datetimeindex():
+    N = 50
+    # testing with timezone, GH #2785
+    rng = date_range("1/1/1990", periods=N, freq="h", tz="US/Eastern")
+    ts = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+
+    result = ts["1990-01-01 04:00:00"]
+    expected = ts.iloc[4]
+    assert result == expected
+
+    result = ts.copy()
+    result["1990-01-01 04:00:00"] = 0
+    result["1990-01-01 04:00:00"] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+    result = ts["1990-01-01 04:00:00":"1990-01-01 07:00:00"]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    result = ts.copy()
+    result["1990-01-01 04:00:00":"1990-01-01 07:00:00"] = 0
+    result["1990-01-01 04:00:00":"1990-01-01 07:00:00"] = ts[4:8]
+    tm.assert_series_equal(result, ts)
+
+    lb = "1990-01-01 04:00:00"
+    rb = "1990-01-01 07:00:00"
+    # GH#18435 strings get a pass from tzawareness compat
+    result = ts[(ts.index >= lb) & (ts.index <= rb)]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    lb = "1990-01-01 04:00:00-0500"
+    rb = "1990-01-01 07:00:00-0500"
+    result = ts[(ts.index >= lb) & (ts.index <= rb)]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    # But we do not give datetimes a pass on tzawareness compat
+    msg = "Cannot compare tz-naive and tz-aware datetime-like objects"
+    naive = datetime(1990, 1, 1, 4)
+    for key in [naive, Timestamp(naive), np.datetime64(naive, "ns")]:
+        with pytest.raises(KeyError, match=re.escape(repr(key))):
+            # GH#36148 as of 2.0 we require tzawareness-compat
+            ts[key]
+
+    result = ts.copy()
+    # GH#36148 as of 2.0 we do not ignore tzawareness mismatch in indexing,
+    #  so setting it as a new key casts to object rather than matching
+    #  rng[4]
+    result[naive] = ts.iloc[4]
+    assert result.index.dtype == object
+    tm.assert_index_equal(result.index[:-1], rng.astype(object))
+    assert result.index[-1] == naive
+
+    msg = "Cannot compare tz-naive and tz-aware datetime-like objects"
+    with pytest.raises(TypeError, match=msg):
+        # GH#36148 require tzawareness compat as of 2.0
+        ts[naive : datetime(1990, 1, 1, 7)]
+
+    result = ts.copy()
+    with pytest.raises(TypeError, match=msg):
+        # GH#36148 require tzawareness compat as of 2.0
+        result[naive : datetime(1990, 1, 1, 7)] = 0
+    with pytest.raises(TypeError, match=msg):
+        # GH#36148 require tzawareness compat as of 2.0
+        result[naive : datetime(1990, 1, 1, 7)] = 99
+    # the __setitems__ here failed, so result should still match ts
+    tm.assert_series_equal(result, ts)
+
+    lb = naive
+    rb = datetime(1990, 1, 1, 7)
+    msg = r"Invalid comparison between dtype=datetime64\[ns, US/Eastern\] and datetime"
+    with pytest.raises(TypeError, match=msg):
+        # tznaive vs tzaware comparison is invalid
+        # see GH#18376, GH#18162
+        ts[(ts.index >= lb) & (ts.index <= rb)]
+
+    lb = Timestamp(naive).tz_localize(rng.tzinfo)
+    rb = Timestamp(datetime(1990, 1, 1, 7)).tz_localize(rng.tzinfo)
+    result = ts[(ts.index >= lb) & (ts.index <= rb)]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    result = ts[ts.index[4]]
+    expected = ts.iloc[4]
+    assert result == expected
+
+    result = ts[ts.index[4:8]]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    result = ts.copy()
+    result[ts.index[4:8]] = 0
+    result.iloc[4:8] = ts.iloc[4:8]
+    tm.assert_series_equal(result, ts)
+
+    # also test partial date slicing
+    result = ts["1990-01-02"]
+    expected = ts[24:48]
+    tm.assert_series_equal(result, expected)
+
+    result = ts.copy()
+    result["1990-01-02"] = 0
+    result["1990-01-02"] = ts[24:48]
+    tm.assert_series_equal(result, ts)
+
+
+def test_getitem_setitem_periodindex():
+    N = 50
+    rng = period_range("1/1/1990", periods=N, freq="h")
+    ts = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+
+    result = ts["1990-01-01 04"]
+    expected = ts.iloc[4]
+    assert result == expected
+
+    result = ts.copy()
+    result["1990-01-01 04"] = 0
+    result["1990-01-01 04"] = ts.iloc[4]
+    tm.assert_series_equal(result, ts)
+
+    result = ts["1990-01-01 04":"1990-01-01 07"]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    result = ts.copy()
+    result["1990-01-01 04":"1990-01-01 07"] = 0
+    result["1990-01-01 04":"1990-01-01 07"] = ts[4:8]
+    tm.assert_series_equal(result, ts)
+
+    lb = "1990-01-01 04"
+    rb = "1990-01-01 07"
+    result = ts[(ts.index >= lb) & (ts.index <= rb)]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    # GH 2782
+    result = ts[ts.index[4]]
+    expected = ts.iloc[4]
+    assert result == expected
+
+    result = ts[ts.index[4:8]]
+    expected = ts[4:8]
+    tm.assert_series_equal(result, expected)
+
+    result = ts.copy()
+    result[ts.index[4:8]] = 0
+    result.iloc[4:8] = ts.iloc[4:8]
+    tm.assert_series_equal(result, ts)
+
+
+def test_datetime_indexing():
+    index = date_range("1/1/2000", "1/7/2000")
+    index = index.repeat(3)
+
+    s = Series(len(index), index=index)
+    stamp = Timestamp("1/8/2000")
+
+    with pytest.raises(KeyError, match=re.escape(repr(stamp))):
+        s[stamp]
+    s[stamp] = 0
+    assert s[stamp] == 0
+
+    # not monotonic
+    s = Series(len(index), index=index)
+    s = s[::-1]
+
+    with pytest.raises(KeyError, match=re.escape(repr(stamp))):
+        s[stamp]
+    s[stamp] = 0
+    assert s[stamp] == 0
+
+
+# test duplicates in time series
+
+
+def test_indexing_with_duplicate_datetimeindex(
+    rand_series_with_duplicate_datetimeindex,
+):
+    ts = rand_series_with_duplicate_datetimeindex
+
+    uniques = ts.index.unique()
+    for date in uniques:
+        result = ts[date]
+
+        mask = ts.index == date
+        total = (ts.index == date).sum()
+        expected = ts[mask]
+        if total > 1:
+            tm.assert_series_equal(result, expected)
+        else:
+            tm.assert_almost_equal(result, expected.iloc[0])
+
+        cp = ts.copy()
+        cp[date] = 0
+        expected = Series(np.where(mask, 0, ts), index=ts.index)
+        tm.assert_series_equal(cp, expected)
+
+    key = datetime(2000, 1, 6)
+    with pytest.raises(KeyError, match=re.escape(repr(key))):
+        ts[key]
+
+    # new index
+    ts[datetime(2000, 1, 6)] = 0
+    assert ts[datetime(2000, 1, 6)] == 0
+
+
+def test_loc_getitem_over_size_cutoff(monkeypatch):
+    # #1821
+
+    monkeypatch.setattr(libindex, "_SIZE_CUTOFF", 1000)
+
+    # create large list of non periodic datetime
+    dates = []
+    sec = timedelta(seconds=1)
+    half_sec = timedelta(microseconds=500000)
+    d = datetime(2011, 12, 5, 20, 30)
+    n = 1100
+    for i in range(n):
+        dates.append(d)
+        dates.append(d + sec)
+        dates.append(d + sec + half_sec)
+        dates.append(d + sec + sec + half_sec)
+        d += 3 * sec
+
+    # duplicate some values in the list
+    duplicate_positions = np.random.default_rng(2).integers(0, len(dates) - 1, 20)
+    for p in duplicate_positions:
+        dates[p + 1] = dates[p]
+
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((len(dates), 4)),
+        index=dates,
+        columns=list("ABCD"),
+    )
+
+    pos = n * 3
+    timestamp = df.index[pos]
+    assert timestamp in df.index
+
+    # it works!
+    df.loc[timestamp]
+    assert len(df.loc[[timestamp]]) > 0
+
+
+def test_indexing_over_size_cutoff_period_index(monkeypatch):
+    # GH 27136
+
+    monkeypatch.setattr(libindex, "_SIZE_CUTOFF", 1000)
+
+    n = 1100
+    idx = period_range("1/1/2000", freq="min", periods=n)
+    assert idx._engine.over_size_threshold
+
+    s = Series(np.random.default_rng(2).standard_normal(len(idx)), index=idx)
+
+    pos = n - 1
+    timestamp = idx[pos]
+    assert timestamp in s.index
+
+    # it works!
+    s[timestamp]
+    assert len(s.loc[[timestamp]]) > 0
+
+
+def test_indexing_unordered():
+    # GH 2437
+    rng = date_range(start="2011-01-01", end="2011-01-15")
+    ts = Series(np.random.default_rng(2).random(len(rng)), index=rng)
+    ts2 = pd.concat([ts[0:4], ts[-4:], ts[4:-4]])
+
+    for t in ts.index:
+        expected = ts[t]
+        result = ts2[t]
+        assert expected == result
+
+    # GH 3448 (ranges)
+    def compare(slobj):
+        result = ts2[slobj].copy()
+        result = result.sort_index()
+        expected = ts[slobj]
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+    for key in [
+        slice("2011-01-01", "2011-01-15"),
+        slice("2010-12-30", "2011-01-15"),
+        slice("2011-01-01", "2011-01-16"),
+        # partial ranges
+        slice("2011-01-01", "2011-01-6"),
+        slice("2011-01-06", "2011-01-8"),
+        slice("2011-01-06", "2011-01-12"),
+    ]:
+        with pytest.raises(
+            KeyError, match="Value based partial slicing on non-monotonic"
+        ):
+            compare(key)
+
+    # single values
+    result = ts2["2011"].sort_index()
+    expected = ts["2011"]
+    expected.index = expected.index._with_freq(None)
+    tm.assert_series_equal(result, expected)
+
+
+def test_indexing_unordered2():
+    # diff freq
+    rng = date_range(datetime(2005, 1, 1), periods=20, freq="ME")
+    ts = Series(np.arange(len(rng)), index=rng)
+    ts = ts.take(np.random.default_rng(2).permutation(20))
+
+    result = ts["2005"]
+    for t in result.index:
+        assert t.year == 2005
+
+
+def test_indexing():
+    idx = date_range("2001-1-1", periods=20, freq="ME")
+    ts = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+
+    # getting
+
+    # GH 3070, make sure semantics work on Series/Frame
+    result = ts["2001"]
+    tm.assert_series_equal(result, ts.iloc[:12])
+
+    df = DataFrame({"A": ts})
+
+    # GH#36179 pre-2.0 df["2001"] operated as slicing on rows. in 2.0 it behaves
+    #  like any other key, so raises
+    with pytest.raises(KeyError, match="2001"):
+        df["2001"]
+
+    # setting
+    ts = Series(np.random.default_rng(2).random(len(idx)), index=idx)
+    expected = ts.copy()
+    expected.iloc[:12] = 1
+    ts["2001"] = 1
+    tm.assert_series_equal(ts, expected)
+
+    expected = df.copy()
+    expected.iloc[:12, 0] = 1
+    df.loc["2001", "A"] = 1
+    tm.assert_frame_equal(df, expected)
+
+
+def test_getitem_str_month_with_datetimeindex():
+    # GH3546 (not including times on the last day)
+    idx = date_range(start="2013-05-31 00:00", end="2013-05-31 23:00", freq="h")
+    ts = Series(range(len(idx)), index=idx)
+    expected = ts["2013-05"]
+    tm.assert_series_equal(expected, ts)
+
+    idx = date_range(start="2013-05-31 00:00", end="2013-05-31 23:59", freq="s")
+    ts = Series(range(len(idx)), index=idx)
+    expected = ts["2013-05"]
+    tm.assert_series_equal(expected, ts)
+
+
+def test_getitem_str_year_with_datetimeindex():
+    idx = [
+        Timestamp("2013-05-31 00:00"),
+        Timestamp(datetime(2013, 5, 31, 23, 59, 59, 999999)),
+    ]
+    ts = Series(range(len(idx)), index=idx)
+    expected = ts["2013"]
+    tm.assert_series_equal(expected, ts)
+
+
+def test_getitem_str_second_with_datetimeindex():
+    # GH14826, indexing with a seconds resolution string / datetime object
+    df = DataFrame(
+        np.random.default_rng(2).random((5, 5)),
+        columns=["open", "high", "low", "close", "volume"],
+        index=date_range("2012-01-02 18:01:00", periods=5, tz="US/Central", freq="s"),
+    )
+
+    # this is a single date, so will raise
+    with pytest.raises(KeyError, match=r"^'2012-01-02 18:01:02'$"):
+        df["2012-01-02 18:01:02"]
+
+    msg = r"Timestamp\('2012-01-02 18:01:02-0600', tz='US/Central'\)"
+    with pytest.raises(KeyError, match=msg):
+        df[df.index[2]]
+
+
+def test_compare_datetime_with_all_none():
+    # GH#54870
+    ser = Series(["2020-01-01", "2020-01-02"], dtype="datetime64[ns]")
+    ser2 = Series([None, None])
+    result = ser > ser2
+    expected = Series([False, False])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/indexing/test_delitem.py b/pandas/pandas/tests/series/indexing/test_delitem.py
new file mode 100644
index 0000000000000..7440ef2692c47
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_delitem.py
@@ -0,0 +1,69 @@
+import pytest
+
+from pandas import (
+    Index,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesDelItem:
+    def test_delitem(self):
+        # GH#5542
+        # should delete the item inplace
+        s = Series(range(5))
+        del s[0]
+
+        expected = Series(range(1, 5), index=range(1, 5))
+        tm.assert_series_equal(s, expected)
+
+        del s[1]
+        expected = Series(range(2, 5), index=range(2, 5))
+        tm.assert_series_equal(s, expected)
+
+        # only 1 left, del, add, del
+        s = Series(1)
+        del s[0]
+        tm.assert_series_equal(s, Series(dtype="int64", index=Index([], dtype="int64")))
+        s[0] = 1
+        tm.assert_series_equal(s, Series(1))
+        del s[0]
+        tm.assert_series_equal(s, Series(dtype="int64", index=Index([], dtype="int64")))
+
+    def test_delitem_object_index(self):
+        # Index(dtype=object)
+        s = Series(1, index=Index(["a"], dtype="str"))
+        del s["a"]
+        tm.assert_series_equal(s, Series(dtype="int64", index=Index([], dtype="str")))
+        s["a"] = 1
+        tm.assert_series_equal(s, Series(1, index=Index(["a"], dtype="str")))
+        del s["a"]
+        tm.assert_series_equal(s, Series(dtype="int64", index=Index([], dtype="str")))
+
+    def test_delitem_missing_key(self):
+        # empty
+        s = Series(dtype=object)
+
+        with pytest.raises(KeyError, match=r"^0$"):
+            del s[0]
+
+    def test_delitem_extension_dtype(self):
+        # GH#40386
+        # DatetimeTZDtype
+        dti = date_range("2016-01-01", periods=3, tz="US/Pacific")
+        ser = Series(dti)
+
+        expected = ser[[0, 2]]
+        del ser[1]
+        assert ser.dtype == dti.dtype
+        tm.assert_series_equal(ser, expected)
+
+        # PeriodDtype
+        pi = dti.tz_localize(None).to_period("D")
+        ser = Series(pi)
+
+        expected = ser[:2]
+        del ser[2]
+        assert ser.dtype == pi.dtype
+        tm.assert_series_equal(ser, expected)
diff --git a/pandas/pandas/tests/series/indexing/test_get.py b/pandas/pandas/tests/series/indexing/test_get.py
new file mode 100644
index 0000000000000..5ff92ca89efba
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_get.py
@@ -0,0 +1,228 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DatetimeIndex,
+    Index,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_get():
+    # GH 6383
+    s = Series(
+        np.array(
+            [
+                43,
+                48,
+                60,
+                48,
+                50,
+                51,
+                50,
+                45,
+                57,
+                48,
+                56,
+                45,
+                51,
+                39,
+                55,
+                43,
+                54,
+                52,
+                51,
+                54,
+            ]
+        )
+    )
+
+    result = s.get(25, 0)
+    expected = 0
+    assert result == expected
+
+    s = Series(
+        np.array(
+            [
+                43,
+                48,
+                60,
+                48,
+                50,
+                51,
+                50,
+                45,
+                57,
+                48,
+                56,
+                45,
+                51,
+                39,
+                55,
+                43,
+                54,
+                52,
+                51,
+                54,
+            ]
+        ),
+        index=Index(
+            [
+                25.0,
+                36.0,
+                49.0,
+                64.0,
+                81.0,
+                100.0,
+                121.0,
+                144.0,
+                169.0,
+                196.0,
+                1225.0,
+                1296.0,
+                1369.0,
+                1444.0,
+                1521.0,
+                1600.0,
+                1681.0,
+                1764.0,
+                1849.0,
+                1936.0,
+            ],
+            dtype=np.float64,
+        ),
+    )
+
+    result = s.get(25, 0)
+    expected = 43
+    assert result == expected
+
+    # GH 7407
+    # with a boolean accessor
+    df = pd.DataFrame({"i": [0] * 3, "b": [False] * 3})
+    vc = df.i.value_counts()
+    result = vc.get(99, default="Missing")
+    assert result == "Missing"
+
+    vc = df.b.value_counts()
+    result = vc.get(False, default="Missing")
+    assert result == 3
+
+    result = vc.get(True, default="Missing")
+    assert result == "Missing"
+
+
+def test_get_nan(float_numpy_dtype):
+    # GH 8569
+    s = Index(range(10), dtype=float_numpy_dtype).to_series()
+    assert s.get(np.nan) is None
+    assert s.get(np.nan, default="Missing") == "Missing"
+
+
+def test_get_nan_multiple(float_numpy_dtype):
+    # GH 8569
+    # ensure that fixing "test_get_nan" above hasn't broken get
+    # with multiple elements
+    s = Index(range(10), dtype=float_numpy_dtype).to_series()
+
+    idx = [2, 30]
+    assert s.get(idx) is None
+
+    idx = [2, np.nan]
+    assert s.get(idx) is None
+
+    # GH 17295 - all missing keys
+    idx = [20, 30]
+    assert s.get(idx) is None
+
+    idx = [np.nan, np.nan]
+    assert s.get(idx) is None
+
+
+def test_get_with_default():
+    # GH#7725
+    d0 = ["a", "b", "c", "d"]
+    d1 = np.arange(4, dtype="int64")
+
+    for data, index in ((d0, d1), (d1, d0)):
+        s = Series(data, index=index)
+        for i, d in zip(index, data):
+            assert s.get(i) == d
+            assert s.get(i, d) == d
+            assert s.get(i, "z") == d
+
+            assert s.get("e", "z") == "z"
+            assert s.get("e", "e") == "e"
+
+            assert s.get(10, "z") == "z"
+            assert s.get(10, 10) == 10
+
+
+@pytest.mark.parametrize(
+    "arr",
+    [
+        np.random.default_rng(2).standard_normal(10),
+        DatetimeIndex(date_range("2020-01-01", periods=10), name="a").tz_localize(
+            tz="US/Eastern"
+        ),
+    ],
+)
+def test_get_with_ea(arr):
+    # GH#21260
+    ser = Series(arr, index=[2 * i for i in range(len(arr))])
+    assert ser.get(4) == ser.iloc[2]
+
+    result = ser.get([4, 6])
+    expected = ser.iloc[[2, 3]]
+    tm.assert_series_equal(result, expected)
+
+    result = ser.get(slice(2))
+    expected = ser.iloc[[0, 1]]
+    tm.assert_series_equal(result, expected)
+
+    assert ser.get(-1) is None
+    assert ser.get(ser.index.max() + 1) is None
+
+    ser = Series(arr[:6], index=list("abcdef"))
+    assert ser.get("c") == ser.iloc[2]
+
+    result = ser.get(slice("b", "d"))
+    expected = ser.iloc[[1, 2, 3]]
+    tm.assert_series_equal(result, expected)
+
+    result = ser.get("Z")
+    assert result is None
+
+    # As of 3.0, ints are treated as labels
+    assert ser.get(4) is None
+    assert ser.get(-1) is None
+    assert ser.get(len(ser)) is None
+
+    # GH#21257
+    ser = Series(arr)
+    ser2 = ser[::2]
+    assert ser2.get(1) is None
+
+
+def test_getitem_get(string_series, object_series):
+    for obj in [string_series, object_series]:
+        idx = obj.index[5]
+
+        assert obj[idx] == obj.get(idx)
+        assert obj[idx] == obj.iloc[5]
+
+    # As of 3.0, ints are treated as labels
+    assert string_series.get(-1) is None
+    assert string_series.iloc[5] == string_series.get(string_series.index[5])
+
+
+def test_get_none():
+    # GH#5652
+    s1 = Series(dtype=object)
+    s2 = Series(dtype=object, index=list("abc"))
+    for s in [s1, s2]:
+        result = s.get(None)
+        assert result is None
diff --git a/pandas/pandas/tests/series/indexing/test_getitem.py b/pandas/pandas/tests/series/indexing/test_getitem.py
new file mode 100644
index 0000000000000..8ba5e8452711d
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_getitem.py
@@ -0,0 +1,718 @@
+"""
+Series.__getitem__ test classes are organized by the type of key passed.
+"""
+
+from datetime import (
+    date,
+    datetime,
+    time,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    conversion,
+    timezones,
+)
+from pandas.compat.numpy import np_version_gt2
+
+from pandas.core.dtypes.common import is_scalar
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.indexing import IndexingError
+
+from pandas.tseries.offsets import BDay
+
+
+class TestSeriesGetitemScalars:
+    def test_getitem_object_index_float_string(self):
+        # GH#17286
+        ser = Series([1] * 4, index=Index(["a", "b", "c", 1.0]))
+        assert ser["a"] == 1
+        assert ser[1.0] == 1
+
+    def test_getitem_float_keys_tuple_values(self):
+        # see GH#13509
+
+        # unique Index
+        ser = Series([(1, 1), (2, 2), (3, 3)], index=[0.0, 0.1, 0.2], name="foo")
+        result = ser[0.0]
+        assert result == (1, 1)
+
+        # non-unique Index
+        expected = Series([(1, 1), (2, 2)], index=[0.0, 0.0], name="foo")
+        ser = Series([(1, 1), (2, 2), (3, 3)], index=[0.0, 0.0, 0.2], name="foo")
+
+        result = ser[0.0]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_unrecognized_scalar(self):
+        # GH#32684 a scalar key that is not recognized by lib.is_scalar
+
+        # a series that might be produced via `frame.dtypes`
+        ser = Series([1, 2], index=[np.dtype("O"), np.dtype("i8")])
+
+        key = ser.index[1]
+
+        result = ser[key]
+        assert result == 2
+
+    def test_getitem_negative_out_of_bounds(self):
+        ser = Series(["a"] * 10, index=["a"] * 10)
+
+        with pytest.raises(KeyError, match="^-11$"):
+            ser[-11]
+
+    def test_getitem_out_of_bounds_indexerror(self, datetime_series):
+        # don't segfault, GH#495
+        N = len(datetime_series)
+        with pytest.raises(KeyError, match=str(N)):
+            datetime_series[N]
+
+    def test_getitem_out_of_bounds_empty_rangeindex_keyerror(self):
+        # GH#917
+        # With a RangeIndex, an int key gives a KeyError
+        ser = Series([], dtype=object)
+        with pytest.raises(KeyError, match="-1"):
+            ser[-1]
+
+    def test_getitem_keyerror_with_integer_index(self, any_int_numpy_dtype):
+        dtype = any_int_numpy_dtype
+        ser = Series(
+            np.random.default_rng(2).standard_normal(6),
+            index=Index([0, 0, 1, 1, 2, 2], dtype=dtype),
+        )
+
+        with pytest.raises(KeyError, match=r"^5$"):
+            ser[5]
+
+        with pytest.raises(KeyError, match=r"^'c'$"):
+            ser["c"]
+
+        # not monotonic
+        ser = Series(
+            np.random.default_rng(2).standard_normal(6), index=[2, 2, 0, 0, 1, 1]
+        )
+
+        with pytest.raises(KeyError, match=r"^5$"):
+            ser[5]
+
+        with pytest.raises(KeyError, match=r"^'c'$"):
+            ser["c"]
+
+    def test_getitem_int64(self, datetime_series):
+        if np_version_gt2:
+            msg = r"^np.int64\(5\)$"
+        else:
+            msg = "^5$"
+        idx = np.int64(5)
+        with pytest.raises(KeyError, match=msg):
+            datetime_series[idx]
+
+    def test_getitem_full_range(self):
+        # github.com/pandas-dev/pandas/commit/4f433773141d2eb384325714a2776bcc5b2e20f7
+        ser = Series(range(5), index=list(range(5)))
+        result = ser[list(range(5))]
+        tm.assert_series_equal(result, ser)
+
+    # ------------------------------------------------------------------
+    # Series with DatetimeIndex
+
+    @pytest.mark.parametrize("tzstr", ["Europe/Berlin", "dateutil/Europe/Berlin"])
+    def test_getitem_pydatetime_tz(self, tzstr):
+        tz = timezones.maybe_get_tz(tzstr)
+
+        index = date_range(
+            start="2012-12-24 16:00", end="2012-12-24 18:00", freq="h", tz=tzstr
+        )
+        ts = Series(index=index, data=index.hour)
+        time_pandas = Timestamp("2012-12-24 17:00", tz=tzstr)
+
+        dt = datetime(2012, 12, 24, 17, 0)
+        time_datetime = conversion.localize_pydatetime(dt, tz)
+        assert ts[time_pandas] == ts[time_datetime]
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_string_index_alias_tz_aware(self, tz):
+        rng = date_range("1/1/2000", periods=10, tz=tz)
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        result = ser["1/3/2000"]
+        tm.assert_almost_equal(result, ser.iloc[2])
+
+    def test_getitem_time_object(self):
+        rng = date_range("1/1/2000", "1/5/2000", freq="5min")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        mask = (rng.hour == 9) & (rng.minute == 30)
+        result = ts[time(9, 30)]
+        expected = ts[mask]
+        result.index = result.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+    # ------------------------------------------------------------------
+    # Series with CategoricalIndex
+
+    def test_getitem_scalar_categorical_index(self):
+        cats = Categorical([Timestamp("12-31-1999"), Timestamp("12-31-2000")])
+
+        ser = Series([1, 2], index=cats)
+
+        expected = ser.iloc[0]
+        result = ser[cats[0]]
+        assert result == expected
+
+    def test_getitem_numeric_categorical_listlike_matches_scalar(self):
+        # GH#15470
+        ser = Series(["a", "b", "c"], index=pd.CategoricalIndex([2, 1, 0]))
+
+        # 0 is treated as a label
+        assert ser[0] == "c"
+
+        # the listlike analogue should also be treated as labels
+        res = ser[[0]]
+        expected = ser.iloc[-1:]
+        tm.assert_series_equal(res, expected)
+
+        res2 = ser[[0, 1, 2]]
+        tm.assert_series_equal(res2, ser.iloc[::-1])
+
+    def test_getitem_integer_categorical_not_positional(self):
+        # GH#14865
+        ser = Series(["a", "b", "c"], index=Index([1, 2, 3], dtype="category"))
+        assert ser.get(3) == "c"
+        assert ser[3] == "c"
+
+    def test_getitem_str_with_timedeltaindex(self):
+        rng = timedelta_range("1 day 10:11:12", freq="h", periods=500)
+        ser = Series(np.arange(len(rng)), index=rng)
+
+        key = "6 days, 23:11:12"
+        indexer = rng.get_loc(key)
+        assert indexer == 133
+
+        result = ser[key]
+        assert result == ser.iloc[133]
+
+        msg = r"^Timedelta\('50 days 00:00:00'\)$"
+        with pytest.raises(KeyError, match=msg):
+            rng.get_loc("50 days")
+        with pytest.raises(KeyError, match=msg):
+            ser["50 days"]
+
+    def test_getitem_bool_index_positional(self):
+        # GH#48653
+        ser = Series({True: 1, False: 0})
+        with pytest.raises(KeyError, match="^0$"):
+            ser[0]
+
+
+class TestSeriesGetitemSlices:
+    def test_getitem_partial_str_slice_with_datetimeindex(self):
+        # GH#34860
+        arr = date_range("1/1/2008", "1/1/2009")
+        ser = arr.to_series()
+        result = ser["2008"]
+
+        rng = date_range(start="2008-01-01", end="2008-12-31")
+        expected = Series(rng, index=rng)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_slice_strings_with_datetimeindex(self):
+        idx = DatetimeIndex(
+            ["1/1/2000", "1/2/2000", "1/2/2000", "1/3/2000", "1/4/2000"]
+        )
+
+        ts = Series(np.random.default_rng(2).standard_normal(len(idx)), index=idx)
+
+        result = ts["1/2/2000":]
+        expected = ts[1:]
+        tm.assert_series_equal(result, expected)
+
+        result = ts["1/2/2000":"1/3/2000"]
+        expected = ts[1:4]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_partial_str_slice_with_timedeltaindex(self):
+        rng = timedelta_range("1 day 10:11:12", freq="h", periods=500)
+        ser = Series(np.arange(len(rng)), index=rng)
+
+        result = ser["5 day":"6 day"]
+        expected = ser.iloc[86:134]
+        tm.assert_series_equal(result, expected)
+
+        result = ser["5 day":]
+        expected = ser.iloc[86:]
+        tm.assert_series_equal(result, expected)
+
+        result = ser[:"6 day"]
+        expected = ser.iloc[:134]
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_partial_str_slice_high_reso_with_timedeltaindex(self):
+        # higher reso
+        rng = timedelta_range("1 day 10:11:12", freq="us", periods=2000)
+        ser = Series(np.arange(len(rng)), index=rng)
+
+        result = ser["1 day 10:11:12":]
+        expected = ser.iloc[0:]
+        tm.assert_series_equal(result, expected)
+
+        result = ser["1 day 10:11:12.001":]
+        expected = ser.iloc[1000:]
+        tm.assert_series_equal(result, expected)
+
+        result = ser["1 days, 10:11:12.001001"]
+        assert result == ser.iloc[1001]
+
+    def test_getitem_slice_2d(self, datetime_series):
+        # GH#30588 multi-dimensional indexing deprecated
+        with pytest.raises(ValueError, match="Multi-dimensional indexing"):
+            datetime_series[:, np.newaxis]
+
+    def test_getitem_median_slice_bug(self):
+        index = date_range("20090415", "20090519", freq="2B")
+        ser = Series(np.random.default_rng(2).standard_normal(13), index=index)
+
+        indexer = [slice(6, 7, None)]
+        msg = "Indexing with a single-item list"
+        with pytest.raises(ValueError, match=msg):
+            # GH#31299
+            ser[indexer]
+        # but we're OK with a single-element tuple
+        result = ser[(indexer[0],)]
+        expected = ser[indexer[0]]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "slc, positions",
+        [
+            [slice(date(2018, 1, 1), None), [0, 1, 2]],
+            [slice(date(2019, 1, 2), None), [2]],
+            [slice(date(2020, 1, 1), None), []],
+            [slice(None, date(2020, 1, 1)), [0, 1, 2]],
+            [slice(None, date(2019, 1, 1)), [0]],
+        ],
+    )
+    def test_getitem_slice_date(self, slc, positions):
+        # https://github.com/pandas-dev/pandas/issues/31501
+        ser = Series(
+            [0, 1, 2],
+            DatetimeIndex(["2019-01-01", "2019-01-01T06:00:00", "2019-01-02"]),
+        )
+        result = ser[slc]
+        expected = ser.take(positions)
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_slice_float_raises(self, datetime_series):
+        msg = (
+            "cannot do slice indexing on DatetimeIndex with these indexers "
+            r"\[{key}\] of type float"
+        )
+        with pytest.raises(TypeError, match=msg.format(key=r"4\.0")):
+            datetime_series[4.0:10.0]
+
+        with pytest.raises(TypeError, match=msg.format(key=r"4\.5")):
+            datetime_series[4.5:10.0]
+
+    def test_getitem_slice_bug(self):
+        ser = Series(range(10), index=list(range(10)))
+        result = ser[-12:]
+        tm.assert_series_equal(result, ser)
+
+        result = ser[-7:]
+        tm.assert_series_equal(result, ser[3:])
+
+        result = ser[:-12]
+        tm.assert_series_equal(result, ser[:0])
+
+    def test_getitem_slice_integers(self):
+        ser = Series(
+            np.random.default_rng(2).standard_normal(8),
+            index=[2, 4, 6, 8, 10, 12, 14, 16],
+        )
+
+        result = ser[:4]
+        expected = Series(ser.values[:4], index=[2, 4, 6, 8])
+        tm.assert_series_equal(result, expected)
+
+
+class TestSeriesGetitemListLike:
+    @pytest.mark.parametrize("box", [list, np.array, Index, Series])
+    def test_getitem_no_matches(self, box):
+        # GH#33462 we expect the same behavior for list/ndarray/Index/Series
+        ser = Series(["A", "B"])
+
+        key = Series(["C"], dtype=object)
+        key = box(key)
+
+        msg = r"None of \[Index\(\['C'\], dtype='object|str'\)\] are in the \[index\]"
+        with pytest.raises(KeyError, match=msg):
+            ser[key]
+
+    def test_getitem_intlist_intindex_periodvalues(self):
+        ser = Series(period_range("2000-01-01", periods=10, freq="D"))
+
+        result = ser[[2, 4]]
+        exp = Series(
+            [pd.Period("2000-01-03", freq="D"), pd.Period("2000-01-05", freq="D")],
+            index=[2, 4],
+            dtype="Period[D]",
+        )
+        tm.assert_series_equal(result, exp)
+        assert result.dtype == "Period[D]"
+
+    @pytest.mark.parametrize("box", [list, np.array, Index])
+    def test_getitem_intlist_intervalindex_non_int(self, box):
+        # GH#33404 fall back to positional since ints are unambiguous;
+        #  changed in 3.0 to never fallback
+        dti = date_range("2000-01-03", periods=3)._with_freq(None)
+        ii = pd.IntervalIndex.from_breaks(dti)
+        ser = Series(range(len(ii)), index=ii)
+
+        key = box([0])
+        msg = r"None of \[Index\(\[0\], dtype='int(32|64)'\)\] are in the \[index\]"
+        with pytest.raises(KeyError, match=msg):
+            ser[key]
+
+    @pytest.mark.parametrize("box", [list, np.array, Index])
+    @pytest.mark.parametrize("dtype", [np.int64, np.float64, np.uint64])
+    def test_getitem_intlist_multiindex_numeric_level(self, dtype, box):
+        # GH#33404 do _not_ fall back to positional since ints are ambiguous
+        idx = Index(range(4)).astype(dtype)
+        dti = date_range("2000-01-03", periods=3)
+        mi = pd.MultiIndex.from_product([idx, dti])
+        ser = Series(range(len(mi))[::-1], index=mi)
+
+        key = box([5])
+        with pytest.raises(KeyError, match="5"):
+            ser[key]
+
+    def test_getitem_uint_array_key(self, any_unsigned_int_numpy_dtype):
+        # GH #37218
+        ser = Series([1, 2, 3])
+        key = np.array([4], dtype=any_unsigned_int_numpy_dtype)
+
+        with pytest.raises(KeyError, match="4"):
+            ser[key]
+        with pytest.raises(KeyError, match="4"):
+            ser.loc[key]
+
+
+class TestGetitemBooleanMask:
+    def test_getitem_boolean(self, string_series):
+        ser = string_series
+        mask = ser > ser.median()
+
+        # passing list is OK
+        result = ser[list(mask)]
+        expected = ser[mask]
+        tm.assert_series_equal(result, expected)
+        tm.assert_index_equal(result.index, ser.index[mask])
+
+    def test_getitem_boolean_empty(self):
+        ser = Series([], dtype=np.int64)
+        ser.index.name = "index_name"
+        ser = ser[ser.isna()]
+        assert ser.index.name == "index_name"
+        assert ser.dtype == np.int64
+
+        # GH#5877
+        # indexing with empty series
+        ser = Series(["A", "B"], dtype=object)
+        expected = Series(dtype=object, index=Index([], dtype="int64"))
+        result = ser[Series([], dtype=object)]
+        tm.assert_series_equal(result, expected)
+
+        # invalid because of the boolean indexer
+        # that's empty or not-aligned
+        msg = (
+            r"Unalignable boolean Series provided as indexer \(index of "
+            r"the boolean Series and of the indexed object do not match"
+        )
+        with pytest.raises(IndexingError, match=msg):
+            ser[Series([], dtype=bool)]
+
+        with pytest.raises(IndexingError, match=msg):
+            ser[Series([True], dtype=bool)]
+
+    def test_getitem_boolean_object(self, string_series):
+        # using column from DataFrame
+
+        ser = string_series
+        mask = ser > ser.median()
+        omask = mask.astype(object)
+
+        # getitem
+        result = ser[omask]
+        expected = ser[mask]
+        tm.assert_series_equal(result, expected)
+
+        # setitem
+        s2 = ser.copy()
+        cop = ser.copy()
+        cop[omask] = 5
+        s2[mask] = 5
+        tm.assert_series_equal(cop, s2)
+
+        # nans raise exception
+        omask[5:10] = np.nan
+        msg = "Cannot mask with non-boolean array containing NA / NaN values"
+        with pytest.raises(ValueError, match=msg):
+            ser[omask]
+        with pytest.raises(ValueError, match=msg):
+            ser[omask] = 5
+
+    def test_getitem_boolean_dt64_copies(self):
+        # GH#36210
+        dti = date_range("2016-01-01", periods=4, tz="US/Pacific")
+        key = np.array([True, True, False, False])
+
+        ser = Series(dti._data)
+
+        res = ser[key]
+        assert res._values._ndarray.base is None
+
+        # compare with numeric case for reference
+        ser2 = Series(range(4))
+        res2 = ser2[key]
+        assert res2._values.base is None
+
+    def test_getitem_boolean_corner(self, datetime_series):
+        ts = datetime_series
+        mask_shifted = ts.shift(1, freq=BDay()) > ts.median()
+
+        msg = (
+            r"Unalignable boolean Series provided as indexer \(index of "
+            r"the boolean Series and of the indexed object do not match"
+        )
+        with pytest.raises(IndexingError, match=msg):
+            ts[mask_shifted]
+
+        with pytest.raises(IndexingError, match=msg):
+            ts.loc[mask_shifted]
+
+    def test_getitem_boolean_different_order(self, string_series):
+        ordered = string_series.sort_values()
+
+        sel = string_series[ordered > 0]
+        exp = string_series[string_series > 0]
+        tm.assert_series_equal(sel, exp)
+
+    def test_getitem_boolean_contiguous_preserve_freq(self):
+        rng = date_range("1/1/2000", "3/1/2000", freq="B")
+
+        mask = np.zeros(len(rng), dtype=bool)
+        mask[10:20] = True
+
+        masked = rng[mask]
+        expected = rng[10:20]
+        assert expected.freq == rng.freq
+        tm.assert_index_equal(masked, expected)
+
+        mask[22] = True
+        masked = rng[mask]
+        assert masked.freq is None
+
+
+class TestGetitemCallable:
+    def test_getitem_callable(self):
+        # GH#12533
+        ser = Series(4, index=list("ABCD"))
+        result = ser[lambda x: "A"]
+        assert result == ser.loc["A"]
+
+        result = ser[lambda x: ["A", "B"]]
+        expected = ser.loc[["A", "B"]]
+        tm.assert_series_equal(result, expected)
+
+        result = ser[lambda x: [True, False, True, True]]
+        expected = ser.iloc[[0, 2, 3]]
+        tm.assert_series_equal(result, expected)
+
+
+def test_getitem_generator(string_series):
+    gen = (x > 0 for x in string_series)
+    result = string_series[gen]
+    result2 = string_series[iter(string_series > 0)]
+    expected = string_series[string_series > 0]
+    tm.assert_series_equal(result, expected)
+    tm.assert_series_equal(result2, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        [0, 1],
+        date_range("2012-01-01", periods=2),
+        date_range("2012-01-01", periods=2, tz="CET"),
+    ],
+)
+def test_getitem_ndim_deprecated(data):
+    series = Series(data)
+    with pytest.raises(ValueError, match="Multi-dimensional indexing"):
+        series[:, None]
+
+
+def test_getitem_multilevel_scalar_slice_not_implemented(
+    multiindex_year_month_day_dataframe_random_data,
+):
+    # not implementing this for now
+    df = multiindex_year_month_day_dataframe_random_data
+    ser = df["A"]
+
+    msg = r"\(2000, slice\(3, 4, None\)\)"
+    with pytest.raises(TypeError, match=msg):
+        ser[2000, 3:4]
+
+
+def test_getitem_dataframe_raises():
+    rng = list(range(10))
+    ser = Series(10, index=rng)
+    df = DataFrame(rng, index=rng)
+    msg = (
+        "Indexing a Series with DataFrame is not supported, "
+        "use the appropriate DataFrame column"
+    )
+    with pytest.raises(TypeError, match=msg):
+        ser[df > 5]
+
+
+def test_getitem_assignment_series_alignment():
+    # https://github.com/pandas-dev/pandas/issues/37427
+    # with getitem, when assigning with a Series, it is not first aligned
+    ser = Series(range(10))
+    idx = np.array([2, 4, 9])
+    ser[idx] = Series([10, 11, 12])
+    expected = Series([0, 1, 10, 3, 11, 5, 6, 7, 8, 12])
+    tm.assert_series_equal(ser, expected)
+
+
+def test_getitem_duplicate_index_mistyped_key_raises_keyerror():
+    # GH#29189 float_index.get_loc(None) should raise KeyError, not TypeError
+    ser = Series([2, 5, 6, 8], index=[2.0, 4.0, 4.0, 5.0])
+    with pytest.raises(KeyError, match="None"):
+        ser[None]
+
+    with pytest.raises(KeyError, match="None"):
+        ser.index.get_loc(None)
+
+    with pytest.raises(KeyError, match="None"):
+        ser.index._engine.get_loc(None)
+
+
+def test_getitem_1tuple_slice_without_multiindex():
+    ser = Series(range(5))
+    key = (slice(3),)
+
+    result = ser[key]
+    expected = ser[key[0]]
+    tm.assert_series_equal(result, expected)
+
+
+def test_getitem_preserve_name(datetime_series):
+    result = datetime_series[datetime_series > 0]
+    assert result.name == datetime_series.name
+
+    result = datetime_series[5:10]
+    assert result.name == datetime_series.name
+
+
+def test_getitem_with_integer_labels():
+    # integer indexes, be careful
+    ser = Series(
+        np.random.default_rng(2).standard_normal(10), index=list(range(0, 20, 2))
+    )
+    inds = [0, 2, 5, 7, 8]
+    arr_inds = np.array([0, 2, 5, 7, 8])
+    with pytest.raises(KeyError, match="not in index"):
+        ser[inds]
+
+    with pytest.raises(KeyError, match="not in index"):
+        ser[arr_inds]
+
+
+def test_getitem_missing(datetime_series):
+    # missing
+    d = datetime_series.index[0] - BDay()
+    msg = r"Timestamp\('1999-12-31 00:00:00'\)"
+    with pytest.raises(KeyError, match=msg):
+        datetime_series[d]
+
+
+def test_getitem_fancy(string_series, object_series):
+    msg = r"None of \[Index\(\[1, 2, 3\], dtype='int(32|64)'\)\] are in the \[index\]"
+    with pytest.raises(KeyError, match=msg):
+        string_series[[1, 2, 3]]
+    with pytest.raises(KeyError, match=msg):
+        object_series[[1, 2, 3]]
+
+
+def test_getitem_box_float64(datetime_series):
+    with pytest.raises(KeyError, match="^5$"):
+        datetime_series[5]
+
+
+def test_getitem_unordered_dup():
+    obj = Series(range(5), index=["c", "a", "a", "b", "b"])
+    assert is_scalar(obj["c"])
+    assert obj["c"] == 0
+
+
+def test_getitem_dups():
+    ser = Series(range(5), index=["A", "A", "B", "C", "C"], dtype=np.int64)
+    expected = Series([3, 4], index=["C", "C"], dtype=np.int64)
+    result = ser["C"]
+    tm.assert_series_equal(result, expected)
+
+
+def test_getitem_categorical_str():
+    # GH#31765
+    ser = Series(range(5), index=Categorical(["a", "b", "c", "a", "b"]))
+    result = ser["a"]
+    expected = ser.iloc[[0, 3]]
+    tm.assert_series_equal(result, expected)
+
+
+def test_slice_can_reorder_not_uniquely_indexed():
+    ser = Series(1, index=["a", "a", "b", "b", "c"])
+    ser[::-1]  # it works!
+
+
+@pytest.mark.parametrize("index_vals", ["aabcd", "aadcb"])
+def test_duplicated_index_getitem_positional_indexer(index_vals):
+    # GH 11747; changed in 3.0 integers are treated as always-labels
+    s = Series(range(5), index=list(index_vals))
+
+    with pytest.raises(KeyError, match="^3$"):
+        s[3]
+
+
+class TestGetitemDeprecatedIndexers:
+    @pytest.mark.parametrize("key", [{1}, {1: 1}])
+    def test_getitem_dict_and_set_deprecated(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2, 3])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser[key]
+
+    @pytest.mark.parametrize("key", [{1}, {1: 1}])
+    def test_setitem_dict_and_set_disallowed(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2, 3])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser[key] = 1
diff --git a/pandas/pandas/tests/series/indexing/test_indexing.py b/pandas/pandas/tests/series/indexing/test_indexing.py
new file mode 100644
index 0000000000000..02efd850f9640
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_indexing.py
@@ -0,0 +1,503 @@
+"""test get/set & misc"""
+
+from datetime import timedelta
+import re
+
+import numpy as np
+import pytest
+
+from pandas.errors import IndexingError
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+    IndexSlice,
+    MultiIndex,
+    NaT,
+    Series,
+    Timedelta,
+    Timestamp,
+    concat,
+    date_range,
+    isna,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+def test_basic_indexing():
+    s = Series(
+        np.random.default_rng(2).standard_normal(5), index=["a", "b", "a", "a", "b"]
+    )
+
+    with pytest.raises(KeyError, match="^5$"):
+        s[5]
+
+    with pytest.raises(KeyError, match=r"^'c'$"):
+        s["c"]
+
+    s = s.sort_index()
+
+    with pytest.raises(KeyError, match="^5$"):
+        s[5]
+
+
+def test_getitem_numeric_should_not_fallback_to_positional(any_numeric_dtype):
+    # GH51053
+    dtype = any_numeric_dtype
+    idx = Index([1, 0, 1], dtype=dtype)
+    ser = Series(range(3), index=idx)
+    result = ser[1]
+    expected = Series([0, 2], index=Index([1, 1], dtype=dtype))
+    tm.assert_series_equal(result, expected, check_exact=True)
+
+
+def test_setitem_numeric_should_not_fallback_to_positional(any_numeric_dtype):
+    # GH51053
+    dtype = any_numeric_dtype
+    idx = Index([1, 0, 1], dtype=dtype)
+    ser = Series(range(3), index=idx)
+    ser[1] = 10
+    expected = Series([10, 1, 10], index=idx)
+    tm.assert_series_equal(ser, expected, check_exact=True)
+
+
+def test_basic_getitem_with_labels(datetime_series):
+    indices = datetime_series.index[[5, 10, 15]]
+
+    result = datetime_series[indices]
+    expected = datetime_series.reindex(indices)
+    tm.assert_series_equal(result, expected)
+
+    result = datetime_series[indices[0] : indices[2]]
+    expected = datetime_series.loc[indices[0] : indices[2]]
+    tm.assert_series_equal(result, expected)
+
+
+def test_basic_getitem_dt64tz_values():
+    # GH12089
+    # with tz for values
+    ser = Series(
+        date_range("2011-01-01", periods=3, tz="US/Eastern"), index=["a", "b", "c"]
+    )
+    expected = Timestamp("2011-01-01", tz="US/Eastern")
+    result = ser.loc["a"]
+    assert result == expected
+    result = ser.iloc[0]
+    assert result == expected
+    result = ser["a"]
+    assert result == expected
+
+
+def test_getitem_setitem_ellipsis():
+    s = Series(np.random.default_rng(2).standard_normal(10))
+
+    result = s[...]
+    tm.assert_series_equal(result, s)
+
+
+@pytest.mark.parametrize(
+    "result_1, duplicate_item, expected_1",
+    [
+        [
+            {1: 12, 2: [1, 2, 2, 3]},
+            {1: 313},
+            Series({1: 12}, dtype=object),
+        ],
+        [
+            {1: [1, 2, 3], 2: [1, 2, 2, 3]},
+            {1: [1, 2, 3]},
+            Series({1: [1, 2, 3]}),
+        ],
+    ],
+)
+def test_getitem_with_duplicates_indices(result_1, duplicate_item, expected_1):
+    # GH 17610
+    result_1 = Series(result_1)
+    duplicate_item = Series(duplicate_item)
+    result = result_1._append(duplicate_item)
+    expected = expected_1._append(duplicate_item)
+    tm.assert_series_equal(result[1], expected)
+    assert result[2] == result_1[2]
+
+
+def test_getitem_setitem_integers():
+    # caused bug without test
+    s = Series([1, 2, 3], ["a", "b", "c"])
+
+    assert s.iloc[0] == s["a"]
+    s.iloc[0] = 5
+    tm.assert_almost_equal(s["a"], 5)
+
+
+def test_series_box_timestamp():
+    rng = date_range("20090415", "20090519", freq="B")
+    ser = Series(rng)
+    assert isinstance(ser[0], Timestamp)
+    assert isinstance(ser.at[1], Timestamp)
+    assert isinstance(ser.iat[2], Timestamp)
+    assert isinstance(ser.loc[3], Timestamp)
+    assert isinstance(ser.iloc[4], Timestamp)
+
+    ser = Series(rng, index=rng)
+    assert isinstance(ser[rng[0]], Timestamp)
+    assert isinstance(ser.at[rng[1]], Timestamp)
+    assert isinstance(ser.iat[2], Timestamp)
+    assert isinstance(ser.loc[rng[3]], Timestamp)
+    assert isinstance(ser.iloc[4], Timestamp)
+
+
+def test_series_box_timedelta():
+    rng = timedelta_range("1 day 1 s", periods=5, freq="h")
+    ser = Series(rng)
+    assert isinstance(ser[0], Timedelta)
+    assert isinstance(ser.at[1], Timedelta)
+    assert isinstance(ser.iat[2], Timedelta)
+    assert isinstance(ser.loc[3], Timedelta)
+    assert isinstance(ser.iloc[4], Timedelta)
+
+
+def test_getitem_ambiguous_keyerror(indexer_sl):
+    ser = Series(range(10), index=list(range(0, 20, 2)))
+    with pytest.raises(KeyError, match=r"^1$"):
+        indexer_sl(ser)[1]
+
+
+def test_getitem_dups_with_missing(indexer_sl):
+    # breaks reindex, so need to use .loc internally
+    # GH 4246
+    ser = Series([1, 2, 3, 4], ["foo", "bar", "foo", "bah"])
+    with pytest.raises(KeyError, match=re.escape("['bam'] not in index")):
+        indexer_sl(ser)[["foo", "bar", "bah", "bam"]]
+
+
+def test_setitem_ambiguous_keyerror(indexer_sl):
+    s = Series(range(10), index=list(range(0, 20, 2)))
+
+    # equivalent of an append
+    s2 = s.copy()
+    indexer_sl(s2)[1] = 5
+    expected = concat([s, Series([5], index=[1])])
+    tm.assert_series_equal(s2, expected)
+
+
+def test_setitem(datetime_series):
+    datetime_series[datetime_series.index[5]] = np.nan
+    datetime_series.iloc[[1, 2, 17]] = np.nan
+    datetime_series.iloc[6] = np.nan
+    assert np.isnan(datetime_series.iloc[6])
+    assert np.isnan(datetime_series.iloc[2])
+    datetime_series[np.isnan(datetime_series)] = 5
+    assert not np.isnan(datetime_series.iloc[2])
+
+
+def test_setslice(datetime_series):
+    sl = datetime_series[5:20]
+    assert len(sl) == len(sl.index)
+    assert sl.index.is_unique is True
+
+
+def test_basic_getitem_setitem_corner(datetime_series):
+    # invalid tuples, e.g. td.ts[:, None] vs. td.ts[:, 2]
+    msg = "key of type tuple not found and not a MultiIndex"
+    with pytest.raises(KeyError, match=msg):
+        datetime_series[:, 2]
+    with pytest.raises(KeyError, match=msg):
+        datetime_series[:, 2] = 2
+
+    # weird lists. [slice(0, 5)] raises but not two slices
+    msg = "Indexing with a single-item list"
+    with pytest.raises(ValueError, match=msg):
+        # GH#31299
+        datetime_series[[slice(None, 5)]]
+
+    # but we're OK with a single-element tuple
+    result = datetime_series[(slice(None, 5),)]
+    expected = datetime_series[:5]
+    tm.assert_series_equal(result, expected)
+
+    # OK
+    msg = r"unhashable type(: 'slice')?"
+    with pytest.raises(TypeError, match=msg):
+        datetime_series[[5, [None, None]]]
+    with pytest.raises(TypeError, match=msg):
+        datetime_series[[5, [None, None]]] = 2
+
+
+def test_slice(string_series, object_series):
+    original = string_series.copy()
+    numSlice = string_series[10:20]
+    numSliceEnd = string_series[-10:]
+    objSlice = object_series[10:20]
+
+    assert string_series.index[9] not in numSlice.index
+    assert object_series.index[9] not in objSlice.index
+
+    assert len(numSlice) == len(numSlice.index)
+    assert string_series[numSlice.index[0]] == numSlice[numSlice.index[0]]
+
+    assert numSlice.index[1] == string_series.index[11]
+    tm.assert_numpy_array_equal(np.array(numSliceEnd), np.array(string_series)[-10:])
+
+    # Test return view.
+    sl = string_series[10:20]
+    sl[:] = 0
+
+    # Doesn't modify parent (CoW)
+    tm.assert_series_equal(string_series, original)
+
+
+def test_timedelta_assignment():
+    # GH 8209
+    s = Series([], dtype=object)
+    s.loc["B"] = timedelta(1)
+    expected = Series(Timedelta("1 days"), dtype="timedelta64[ns]", index=["B"])
+    tm.assert_series_equal(s, expected)
+
+    s = s.reindex(s.index.insert(0, "A"))
+    expected = Series(
+        [np.nan, Timedelta("1 days")], dtype="timedelta64[ns]", index=["A", "B"]
+    )
+    tm.assert_series_equal(s, expected)
+
+    s.loc["A"] = timedelta(1)
+    expected = Series(Timedelta("1 days"), dtype="timedelta64[ns]", index=["A", "B"])
+    tm.assert_series_equal(s, expected)
+
+
+def test_underlying_data_conversion():
+    # GH 4080
+    df = DataFrame({c: [1, 2, 3] for c in ["a", "b", "c"]})
+    return_value = df.set_index(["a", "b", "c"], inplace=True)
+    assert return_value is None
+    s = Series([1], index=[(2, 2, 2)])
+    df["val"] = 0
+    df_original = df.copy()
+    df
+
+    with tm.raises_chained_assignment_error():
+        df["val"].update(s)
+    expected = df_original
+    tm.assert_frame_equal(df, expected)
+
+
+def test_preserve_refs(datetime_series):
+    seq = datetime_series.iloc[[5, 10, 15]]
+    seq.iloc[1] = np.nan
+    assert not np.isnan(datetime_series.iloc[10])
+
+
+def test_multilevel_preserve_name(lexsorted_two_level_string_multiindex, indexer_sl):
+    index = lexsorted_two_level_string_multiindex
+    ser = Series(
+        np.random.default_rng(2).standard_normal(len(index)), index=index, name="sth"
+    )
+
+    result = indexer_sl(ser)["foo"]
+    assert result.name == ser.name
+
+
+# miscellaneous methods
+
+
+@pytest.mark.parametrize(
+    "index",
+    [
+        date_range("2014-01-01", periods=20, freq="MS"),
+        period_range("2014-01", periods=20, freq="M"),
+        timedelta_range("0", periods=20, freq="h"),
+    ],
+)
+def test_slice_with_negative_step(index):
+    keystr1 = str(index[9])
+    keystr2 = str(index[13])
+
+    ser = Series(np.arange(20), index)
+    SLC = IndexSlice
+
+    for key in [keystr1, index[9]]:
+        tm.assert_indexing_slices_equivalent(ser, SLC[key::-1], SLC[9::-1])
+        tm.assert_indexing_slices_equivalent(ser, SLC[:key:-1], SLC[:8:-1])
+
+        for key2 in [keystr2, index[13]]:
+            tm.assert_indexing_slices_equivalent(ser, SLC[key2:key:-1], SLC[13:8:-1])
+            tm.assert_indexing_slices_equivalent(ser, SLC[key:key2:-1], SLC[0:0:-1])
+
+
+def test_tuple_index():
+    # GH 35534 - Selecting values when a Series has an Index of tuples
+    s = Series([1, 2], index=[("a",), ("b",)])
+    assert s[("a",)] == 1
+    assert s[("b",)] == 2
+    s[("b",)] = 3
+    assert s[("b",)] == 3
+
+
+def test_frozenset_index():
+    # GH35747 - Selecting values when a Series has an Index of frozenset
+    idx0, idx1 = frozenset("a"), frozenset("b")
+    s = Series([1, 2], index=[idx0, idx1])
+    assert s[idx0] == 1
+    assert s[idx1] == 2
+    s[idx1] = 3
+    assert s[idx1] == 3
+
+
+def test_loc_setitem_all_false_indexer():
+    # GH#45778
+    ser = Series([1, 2], index=["a", "b"])
+    expected = ser.copy()
+    rhs = Series([6, 7], index=["a", "b"])
+    ser.loc[ser > 100] = rhs
+    tm.assert_series_equal(ser, expected)
+
+
+def test_loc_boolean_indexer_non_matching_index():
+    # GH#46551
+    ser = Series([1])
+    result = ser.loc[Series([NA, False], dtype="boolean")]
+    expected = Series([], dtype="int64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_loc_boolean_indexer_miss_matching_index():
+    # GH#46551
+    ser = Series([1])
+    indexer = Series([NA, False], dtype="boolean", index=[1, 2])
+    with pytest.raises(IndexingError, match="Unalignable"):
+        ser.loc[indexer]
+
+
+def test_loc_setitem_nested_data_enlargement():
+    # GH#48614
+    df = DataFrame({"a": [1]})
+    ser = Series({"label": df})
+    ser.loc["new_label"] = df
+    expected = Series({"label": df, "new_label": df})
+    tm.assert_series_equal(ser, expected)
+
+
+def test_loc_ea_numeric_index_oob_slice_end():
+    # GH#50161
+    ser = Series(1, index=Index([0, 1, 2], dtype="Int64"))
+    result = ser.loc[2:3]
+    expected = Series(1, index=Index([2], dtype="Int64"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_getitem_bool_int_key():
+    # GH#48653
+    ser = Series({True: 1, False: 0})
+    with pytest.raises(KeyError, match="0"):
+        ser.loc[0]
+
+
+@pytest.mark.parametrize("val", [{}, {"b": "x"}])
+@pytest.mark.parametrize("indexer", [[], [False, False], slice(0, -1), np.array([])])
+def test_setitem_empty_indexer(indexer, val):
+    # GH#45981
+    df = DataFrame({"a": [1, 2], **val})
+    expected = df.copy()
+    df.loc[indexer] = 1.5
+    tm.assert_frame_equal(df, expected)
+
+
+class TestDeprecatedIndexers:
+    @pytest.mark.parametrize("key", [{1}, {1: 1}])
+    def test_getitem_dict_and_set_deprecated(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser.loc[key]
+
+    @pytest.mark.parametrize("key", [{1}, {1: 1}, ({1}, 2), ({1: 1}, 2)])
+    def test_getitem_dict_and_set_deprecated_multiindex(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2], index=MultiIndex.from_tuples([(1, 2), (3, 4)]))
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser.loc[key]
+
+    @pytest.mark.parametrize("key", [{1}, {1: 1}])
+    def test_setitem_dict_and_set_disallowed(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2])
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser.loc[key] = 1
+
+    @pytest.mark.parametrize("key", [{1}, {1: 1}, ({1}, 2), ({1: 1}, 2)])
+    def test_setitem_dict_and_set_disallowed_multiindex(self, key):
+        # GH#42825 enforced in 2.0
+        ser = Series([1, 2], index=MultiIndex.from_tuples([(1, 2), (3, 4)]))
+        with pytest.raises(TypeError, match="as an indexer is not supported"):
+            ser.loc[key] = 1
+
+
+class TestSetitemValidation:
+    # This is adapted from pandas/tests/arrays/masked/test_indexing.py
+    def _check_setitem_invalid(self, ser, invalid, indexer):
+        orig_ser = ser.copy()
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[indexer] = invalid
+            ser = orig_ser.copy()
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser.iloc[indexer] = invalid
+            ser = orig_ser.copy()
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser.loc[indexer] = invalid
+            ser = orig_ser.copy()
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[:] = invalid
+
+    def _check_setitem_valid(self, ser, value, indexer):
+        orig_ser = ser.copy()
+
+        ser[indexer] = value
+        ser = orig_ser.copy()
+
+        ser.iloc[indexer] = value
+        ser = orig_ser.copy()
+
+        ser.loc[indexer] = value
+        ser = orig_ser.copy()
+
+        ser[:] = value
+
+    _invalid_scalars = [
+        1 + 2j,
+        "True",
+        "1",
+        "1.0",
+        NaT,
+        np.datetime64("NaT"),
+        np.timedelta64("NaT"),
+    ]
+    _indexers = [0, [0], slice(0, 1), [True, False, False], slice(None, None, None)]
+
+    @pytest.mark.parametrize(
+        "invalid", _invalid_scalars + [1, 1.0, np.int64(1), np.float64(1)]
+    )
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_bool(self, invalid, indexer):
+        ser = Series([True, False, False], dtype="bool")
+        self._check_setitem_invalid(ser, invalid, indexer)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True, 1.5, np.float64(1.5)])
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_int(self, invalid, any_int_numpy_dtype, indexer):
+        ser = Series([1, 2, 3], dtype=any_int_numpy_dtype)
+        if isna(invalid) and invalid is not NaT and not np.isnat(invalid):
+            self._check_setitem_valid(ser, invalid, indexer)
+        else:
+            self._check_setitem_invalid(ser, invalid, indexer)
+
+    @pytest.mark.parametrize("invalid", _invalid_scalars + [True])
+    @pytest.mark.parametrize("indexer", _indexers)
+    def test_setitem_validation_scalar_float(self, invalid, float_numpy_dtype, indexer):
+        ser = Series([1, 2, None], dtype=float_numpy_dtype)
+        self._check_setitem_invalid(ser, invalid, indexer)
diff --git a/pandas/pandas/tests/series/indexing/test_mask.py b/pandas/pandas/tests/series/indexing/test_mask.py
new file mode 100644
index 0000000000000..3c21cd0d5ca64
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_mask.py
@@ -0,0 +1,69 @@
+import numpy as np
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+def test_mask():
+    # compare with tested results in test_where
+    s = Series(np.random.default_rng(2).standard_normal(5))
+    cond = s > 0
+
+    rs = s.where(~cond, np.nan)
+    tm.assert_series_equal(rs, s.mask(cond))
+
+    rs = s.where(~cond)
+    rs2 = s.mask(cond)
+    tm.assert_series_equal(rs, rs2)
+
+    rs = s.where(~cond, -s)
+    rs2 = s.mask(cond, -s)
+    tm.assert_series_equal(rs, rs2)
+
+    cond = Series([True, False, False, True, False], index=s.index)
+    s2 = -(s.abs())
+    rs = s2.where(~cond[:3])
+    rs2 = s2.mask(cond[:3])
+    tm.assert_series_equal(rs, rs2)
+
+    rs = s2.where(~cond[:3], -s2)
+    rs2 = s2.mask(cond[:3], -s2)
+    tm.assert_series_equal(rs, rs2)
+
+    msg = "Array conditional must be same shape as self"
+    with pytest.raises(ValueError, match=msg):
+        s.mask(1)
+    with pytest.raises(ValueError, match=msg):
+        s.mask(cond[:3].values, -s)
+
+
+def test_mask_casts():
+    # dtype changes
+    ser = Series([1, 2, 3, 4])
+    result = ser.mask(ser > 2, np.nan)
+    expected = Series([1, 2, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_mask_casts2():
+    # see gh-21891
+    ser = Series([1, 2])
+    res = ser.mask([True, False])
+
+    exp = Series([np.nan, 2])
+    tm.assert_series_equal(res, exp)
+
+
+def test_mask_inplace():
+    s = Series(np.random.default_rng(2).standard_normal(5))
+    cond = s > 0
+
+    rs = s.copy()
+    rs.mask(cond, inplace=True)
+    tm.assert_series_equal(rs.dropna(), s[~cond])
+    tm.assert_series_equal(rs, s.mask(cond))
+
+    rs = s.copy()
+    rs.mask(cond, -s, inplace=True)
+    tm.assert_series_equal(rs, s.mask(cond, -s))
diff --git a/pandas/pandas/tests/series/indexing/test_set_value.py b/pandas/pandas/tests/series/indexing/test_set_value.py
new file mode 100644
index 0000000000000..3bf89da53d923
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_set_value.py
@@ -0,0 +1,49 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+def test_series_set_value(using_infer_string):
+    # GH#1561, GH#51363 as of 3.0 we do not do inference in Index.insert
+
+    dates = [datetime(2001, 1, 1), datetime(2001, 1, 2)]
+    if using_infer_string:
+        index = DatetimeIndex(dates)
+    else:
+        index = Index(dates, dtype=object)
+
+    s = Series(dtype=object)
+    s._set_value(dates[0], 1.0)
+    s._set_value(dates[1], np.nan)
+
+    expected = Series([1.0, np.nan], index=index)
+
+    tm.assert_series_equal(s, expected)
+
+
+def test_set_value_dt64(datetime_series):
+    idx = datetime_series.index[10]
+    res = datetime_series._set_value(idx, 0)
+    assert res is None
+    assert datetime_series[idx] == 0
+
+
+def test_set_value_str_index(string_series):
+    # equiv
+    ser = string_series.copy()
+    res = ser._set_value("foobar", 0)
+    assert res is None
+    assert ser.index[-1] == "foobar"
+    assert ser["foobar"] == 0
+
+    ser2 = string_series.copy()
+    ser2.loc["foobar"] = 0
+    assert ser2.index[-1] == "foobar"
+    assert ser2["foobar"] == 0
diff --git a/pandas/pandas/tests/series/indexing/test_setitem.py b/pandas/pandas/tests/series/indexing/test_setitem.py
new file mode 100644
index 0000000000000..964f0b90b3c41
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_setitem.py
@@ -0,0 +1,1840 @@
+import contextlib
+from datetime import (
+    date,
+    datetime,
+)
+from decimal import Decimal
+import os
+
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gte1p24
+from pandas.errors import IndexingError
+
+from pandas.core.dtypes.common import is_list_like
+
+from pandas import (
+    NA,
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    Period,
+    Series,
+    StringDtype,
+    Timedelta,
+    Timestamp,
+    array,
+    concat,
+    date_range,
+    interval_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries.offsets import BDay
+
+
+class TestSetitemDT64Values:
+    def test_setitem_none_nan(self):
+        series = Series(date_range("1/1/2000", periods=10))
+        series[3] = None
+        assert series[3] is NaT
+
+        series[3:5] = None
+        assert series[4] is NaT
+
+        series[5] = np.nan
+        assert series[5] is NaT
+
+        series[5:7] = np.nan
+        assert series[6] is NaT
+
+    def test_setitem_multiindex_empty_slice(self):
+        # https://github.com/pandas-dev/pandas/issues/35878
+        idx = MultiIndex.from_tuples([("a", 1), ("b", 2)])
+        result = Series([1, 2], index=idx)
+        expected = result.copy()
+        result.loc[[]] = 0
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_with_string_index(self):
+        # GH#23451
+        # Set object dtype to avoid upcast when setting date.today()
+        ser = Series([1, 2, 3], index=["Date", "b", "other"], dtype=object)
+        ser["Date"] = date.today()
+        assert ser.Date == date.today()
+        assert ser["Date"] == date.today()
+
+    def test_setitem_tuple_with_datetimetz_values(self):
+        # GH#20441
+        arr = date_range("2017", periods=4, tz="US/Eastern")
+        index = [(0, 1), (0, 2), (0, 3), (0, 4)]
+        result = Series(arr, index=index)
+        expected = result.copy()
+        result[(0, 1)] = np.nan
+        expected.iloc[0] = np.nan
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "UTC", "Asia/Tokyo"])
+    def test_setitem_with_tz(self, tz, indexer_sli):
+        orig = Series(date_range("2016-01-01", freq="h", periods=3, tz=tz))
+        assert orig.dtype == f"datetime64[ns, {tz}]"
+
+        exp = Series(
+            [
+                Timestamp("2016-01-01 00:00", tz=tz),
+                Timestamp("2011-01-01 00:00", tz=tz),
+                Timestamp("2016-01-01 02:00", tz=tz),
+            ],
+            dtype=orig.dtype,
+        )
+
+        # scalar
+        ser = orig.copy()
+        indexer_sli(ser)[1] = Timestamp("2011-01-01", tz=tz)
+        tm.assert_series_equal(ser, exp)
+
+        # vector
+        vals = Series(
+            [Timestamp("2011-01-01", tz=tz), Timestamp("2012-01-01", tz=tz)],
+            index=[1, 2],
+            dtype=orig.dtype,
+        )
+        assert vals.dtype == f"datetime64[ns, {tz}]"
+
+        exp = Series(
+            [
+                Timestamp("2016-01-01 00:00", tz=tz),
+                Timestamp("2011-01-01 00:00", tz=tz),
+                Timestamp("2012-01-01 00:00", tz=tz),
+            ],
+            dtype=orig.dtype,
+        )
+
+        ser = orig.copy()
+        indexer_sli(ser)[[1, 2]] = vals
+        tm.assert_series_equal(ser, exp)
+
+    def test_setitem_with_tz_dst(self, indexer_sli):
+        # GH#14146 trouble setting values near DST boundary
+        tz = "US/Eastern"
+        orig = Series(date_range("2016-11-06", freq="h", periods=3, tz=tz))
+        assert orig.dtype == f"datetime64[ns, {tz}]"
+
+        exp = Series(
+            [
+                Timestamp("2016-11-06 00:00-04:00", tz=tz),
+                Timestamp("2011-01-01 00:00-05:00", tz=tz),
+                Timestamp("2016-11-06 01:00-05:00", tz=tz),
+            ],
+            dtype=orig.dtype,
+        )
+
+        # scalar
+        ser = orig.copy()
+        indexer_sli(ser)[1] = Timestamp("2011-01-01", tz=tz)
+        tm.assert_series_equal(ser, exp)
+
+        # vector
+        vals = Series(
+            [Timestamp("2011-01-01", tz=tz), Timestamp("2012-01-01", tz=tz)],
+            index=[1, 2],
+            dtype=orig.dtype,
+        )
+        assert vals.dtype == f"datetime64[ns, {tz}]"
+
+        exp = Series(
+            [
+                Timestamp("2016-11-06 00:00", tz=tz),
+                Timestamp("2011-01-01 00:00", tz=tz),
+                Timestamp("2012-01-01 00:00", tz=tz),
+            ],
+            dtype=orig.dtype,
+        )
+
+        ser = orig.copy()
+        indexer_sli(ser)[[1, 2]] = vals
+        tm.assert_series_equal(ser, exp)
+
+    def test_object_series_setitem_dt64array_exact_match(self):
+        # make sure the dt64 isn't cast by numpy to integers
+        # https://github.com/numpy/numpy/issues/12550
+
+        ser = Series({"X": np.nan}, dtype=object)
+
+        indexer = [True]
+
+        # "exact_match" -> size of array being set matches size of ser
+        value = np.array([4], dtype="M8[ns]")
+
+        ser.iloc[indexer] = value
+
+        expected = Series([value[0]], index=["X"], dtype=object)
+        assert all(isinstance(x, np.datetime64) for x in expected.values)
+
+        tm.assert_series_equal(ser, expected)
+
+
+class TestSetitemScalarIndexer:
+    def test_setitem_negative_out_of_bounds(self):
+        # As of 3.0, int keys are treated as labels, so this becomes
+        #  setitem-with-expansion
+        ser = Series(["a"] * 10, index=["a"] * 10)
+        ser[-11] = "foo"
+        exp = Series(["a"] * 10 + ["foo"], index=["a"] * 10 + [-11])
+        tm.assert_series_equal(ser, exp)
+
+    @pytest.mark.parametrize("indexer", [tm.loc, tm.at])
+    @pytest.mark.parametrize("ser_index", [0, 1])
+    def test_setitem_series_object_dtype(self, indexer, ser_index):
+        # GH#38303
+        ser = Series([0, 0], dtype="object")
+        idxr = indexer(ser)
+        idxr[0] = Series([42], index=[ser_index])
+        expected = Series([Series([42], index=[ser_index]), 0], dtype="object")
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize("index, exp_value", [(0, 42), (1, np.nan)])
+    def test_setitem_series(self, index, exp_value):
+        # GH#38303
+        ser = Series([0, 0])
+        ser.loc[0] = Series([42], index=[index])
+        expected = Series([exp_value, 0])
+        tm.assert_series_equal(ser, expected)
+
+
+class TestSetitemSlices:
+    def test_setitem_slice_float_raises(self, datetime_series):
+        msg = (
+            "cannot do slice indexing on DatetimeIndex with these indexers "
+            r"\[{key}\] of type float"
+        )
+        with pytest.raises(TypeError, match=msg.format(key=r"4\.0")):
+            datetime_series[4.0:10.0] = 0
+
+        with pytest.raises(TypeError, match=msg.format(key=r"4\.5")):
+            datetime_series[4.5:10.0] = 0
+
+    def test_setitem_slice(self):
+        ser = Series(range(10), index=list(range(10)))
+        ser[-12:] = 0
+        assert (ser == 0).all()
+
+        ser[:-12] = 5
+        assert (ser == 0).all()
+
+    def test_setitem_slice_integers(self):
+        ser = Series(
+            np.random.default_rng(2).standard_normal(8),
+            index=[2, 4, 6, 8, 10, 12, 14, 16],
+        )
+
+        ser[:4] = 0
+        assert (ser[:4] == 0).all()
+        assert not (ser[4:] == 0).any()
+
+    def test_setitem_slicestep(self):
+        # caught this bug when writing tests
+        series = Series(
+            np.arange(20, dtype=np.float64), index=np.arange(20, dtype=np.int64)
+        )
+
+        series[::2] = 0
+        assert (series[::2] == 0).all()
+
+    def test_setitem_multiindex_slice(self, indexer_sli):
+        # GH 8856
+        mi = MultiIndex.from_product(([0, 1], list("abcde")))
+        result = Series(np.arange(10, dtype=np.int64), mi)
+        indexer_sli(result)[::4] = 100
+        expected = Series([100, 1, 2, 3, 100, 5, 6, 7, 100, 9], mi)
+        tm.assert_series_equal(result, expected)
+
+
+class TestSetitemBooleanMask:
+    def test_setitem_mask_cast(self):
+        # GH#2746
+        # need to upcast
+        ser = Series([1, 2], index=[1, 2], dtype="int64")
+        ser[[True, False]] = Series([0], index=[1], dtype="int64")
+        expected = Series([0, 2], index=[1, 2], dtype="int64")
+
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_mask_align_and_promote(self):
+        # GH#8387: test that changing types does not break alignment
+        ts = Series(
+            np.random.default_rng(2).standard_normal(100), index=np.arange(100, 0, -1)
+        ).round(5)
+        mask = ts > 0
+        left = ts.copy()
+        right = ts[mask].copy().map(str)
+        with pytest.raises(TypeError, match="Invalid value"):
+            left[mask] = right
+
+    def test_setitem_mask_promote_strs(self):
+        ser = Series([0, 1, 2, 0])
+        mask = ser > 0
+        ser2 = ser[mask].map(str)
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[mask] = ser2
+
+    def test_setitem_mask_promote(self):
+        ser = Series([0, "foo", "bar", 0])
+        mask = Series([False, True, True, False])
+        ser2 = ser[mask]
+        ser[mask] = ser2
+
+        expected = Series([0, "foo", "bar", 0])
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_boolean(self, string_series):
+        mask = string_series > string_series.median()
+
+        # similar indexed series
+        result = string_series.copy()
+        result[mask] = string_series * 2
+        expected = string_series * 2
+        tm.assert_series_equal(result[mask], expected[mask])
+
+        # needs alignment
+        result = string_series.copy()
+        result[mask] = (string_series * 2)[0:5]
+        expected = (string_series * 2)[0:5].reindex_like(string_series)
+        expected[-mask] = string_series[mask]
+        tm.assert_series_equal(result[mask], expected[mask])
+
+    def test_setitem_boolean_corner(self, datetime_series):
+        ts = datetime_series
+        mask_shifted = ts.shift(1, freq=BDay()) > ts.median()
+
+        msg = (
+            r"Unalignable boolean Series provided as indexer \(index of "
+            r"the boolean Series and of the indexed object do not match"
+        )
+        with pytest.raises(IndexingError, match=msg):
+            ts[mask_shifted] = 1
+
+        with pytest.raises(IndexingError, match=msg):
+            ts.loc[mask_shifted] = 1
+
+    def test_setitem_boolean_different_order(self, string_series):
+        ordered = string_series.sort_values()
+
+        copy = string_series.copy()
+        copy[ordered > 0] = 0
+
+        expected = string_series.copy()
+        expected[expected > 0] = 0
+
+        tm.assert_series_equal(copy, expected)
+
+    @pytest.mark.parametrize("func", [list, np.array, Series])
+    def test_setitem_boolean_python_list(self, func):
+        # GH19406
+        ser = Series([None, "b", None])
+        mask = func([True, False, True])
+        ser[mask] = ["a", "c"]
+        expected = Series(["a", "b", "c"])
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_boolean_nullable_int_types(self, any_numeric_ea_dtype):
+        # GH: 26468
+        ser = Series([5, 6, 7, 8], dtype=any_numeric_ea_dtype)
+        ser[ser > 6] = Series(range(4), dtype=any_numeric_ea_dtype)
+        expected = Series([5, 6, 2, 3], dtype=any_numeric_ea_dtype)
+        tm.assert_series_equal(ser, expected)
+
+        ser = Series([5, 6, 7, 8], dtype=any_numeric_ea_dtype)
+        ser.loc[ser > 6] = Series(range(4), dtype=any_numeric_ea_dtype)
+        tm.assert_series_equal(ser, expected)
+
+        ser = Series([5, 6, 7, 8], dtype=any_numeric_ea_dtype)
+        loc_ser = Series(range(4), dtype=any_numeric_ea_dtype)
+        ser.loc[ser > 6] = loc_ser.loc[loc_ser > 1]
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_with_bool_mask_and_values_matching_n_trues_in_length(self):
+        # GH#30567
+        ser = Series([None] * 10)
+        mask = [False] * 3 + [True] * 5 + [False] * 2
+        ser[mask] = range(5)
+        result = ser
+        expected = Series([None] * 3 + list(range(5)) + [None] * 2, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    def test_setitem_nan_with_bool(self):
+        # GH 13034
+        result = Series([True, False, True])
+        with pytest.raises(TypeError, match="Invalid value"):
+            result[0] = np.nan
+
+    def test_setitem_mask_smallint_upcast(self):
+        orig = Series([1, 2, 3], dtype="int8")
+        alt = np.array([999, 1000, 1001], dtype=np.int64)
+
+        mask = np.array([True, False, True])
+
+        ser = orig.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[mask] = Series(alt)
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser.mask(mask, alt, inplace=True)
+
+        res = ser.where(~mask, Series(alt))
+        expected = Series([999, 2, 1001])
+        tm.assert_series_equal(res, expected)
+
+    def test_setitem_mask_smallint_no_upcast(self):
+        # like test_setitem_mask_smallint_upcast, but while we can't hold 'alt',
+        #  we *can* hold alt[mask] without casting
+        orig = Series([1, 2, 3], dtype="uint8")
+        alt = Series([245, 1000, 246], dtype=np.int64)
+
+        mask = np.array([True, False, True])
+
+        ser = orig.copy()
+        ser[mask] = alt
+        expected = Series([245, 2, 246], dtype="uint8")
+        tm.assert_series_equal(ser, expected)
+
+        ser2 = orig.copy()
+        ser2.mask(mask, alt, inplace=True)
+        tm.assert_series_equal(ser2, expected)
+
+        # TODO: ser.where(~mask, alt) unnecessarily upcasts to int64
+        ser3 = orig.copy()
+        res = ser3.where(~mask, alt)
+        tm.assert_series_equal(res, expected, check_dtype=False)
+
+
+class TestSetitemViewCopySemantics:
+    def test_setitem_invalidates_datetime_index_freq(self):
+        # GH#24096 altering a datetime64tz Series inplace invalidates the
+        #  `freq` attribute on the underlying DatetimeIndex
+
+        dti = date_range("20130101", periods=3, tz="US/Eastern")
+        ts = dti[1]
+        ser = Series(dti)
+        assert ser._values is not dti
+        assert ser._values._ndarray.base is dti._data._ndarray.base
+        assert dti.freq == "D"
+        ser.iloc[1] = NaT
+        assert ser._values.freq is None
+
+        # check that the DatetimeIndex was not altered in place
+        assert ser._values is not dti
+        assert ser._values._ndarray.base is not dti._data._ndarray.base
+        assert dti[1] == ts
+        assert dti.freq == "D"
+
+    def test_dt64tz_setitem_does_not_mutate_dti(self):
+        # GH#21907, GH#24096
+        dti = date_range("2016-01-01", periods=10, tz="US/Pacific")
+        ts = dti[0]
+        ser = Series(dti)
+        assert ser._values is not dti
+        assert ser._values._ndarray.base is dti._data._ndarray.base
+        assert ser._mgr.blocks[0].values._ndarray.base is dti._data._ndarray.base
+
+        assert ser._mgr.blocks[0].values is not dti
+
+        ser[::3] = NaT
+        assert ser[0] is NaT
+        assert dti[0] == ts
+
+
+class TestSetitemCallable:
+    def test_setitem_callable_key(self):
+        # GH#12533
+        ser = Series([1, 2, 3, 4], index=list("ABCD"))
+        ser[lambda x: "A"] = -1
+
+        expected = Series([-1, 2, 3, 4], index=list("ABCD"))
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_callable_other(self):
+        # GH#13299
+        inc = lambda x: x + 1
+
+        # set object dtype to avoid upcast when setting inc
+        ser = Series([1, 2, -1, 4], dtype=object)
+        ser[ser < 0] = inc
+
+        expected = Series([1, 2, inc, 4])
+        tm.assert_series_equal(ser, expected)
+
+
+class TestSetitemWithExpansion:
+    def test_setitem_empty_series(self, using_infer_string):
+        # GH#10193
+        key = Timestamp("2012-01-01")
+        series = Series(dtype=object)
+        series[key] = 47
+        expected = Series(
+            47, index=[key] if using_infer_string else Index([key], dtype=object)
+        )
+        tm.assert_series_equal(series, expected)
+
+    def test_setitem_empty_series_datetimeindex_preserves_freq(self):
+        # GH#33573 our index should retain its freq
+        dti = DatetimeIndex([], freq="D", dtype="M8[ns]")
+        series = Series([], index=dti, dtype=object)
+        key = Timestamp("2012-01-01")
+        series[key] = 47
+        expected = Series(47, DatetimeIndex([key], freq="D").as_unit("ns"))
+        tm.assert_series_equal(series, expected)
+        assert series.index.freq == expected.index.freq
+
+    def test_setitem_empty_series_timestamp_preserves_dtype(self):
+        # GH 21881
+        timestamp = Timestamp(1412526600000000000)
+        series = Series([timestamp], index=["timestamp"], dtype=object)
+        expected = series["timestamp"]
+
+        series = Series([], dtype=object)
+        series["anything"] = 300.0
+        series["timestamp"] = timestamp
+        result = series["timestamp"]
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "td",
+        [
+            Timedelta("9 days"),
+            Timedelta("9 days").to_timedelta64(),
+            Timedelta("9 days").to_pytimedelta(),
+        ],
+    )
+    def test_append_timedelta_does_not_cast(self, td, using_infer_string, request):
+        # GH#22717 inserting a Timedelta should _not_ cast to int64
+        if using_infer_string and not isinstance(td, Timedelta):
+            # TODO: GH#56010
+            request.applymarker(pytest.mark.xfail(reason="inferred as string"))
+
+        expected = Series(["x", td], index=[0, "td"], dtype=object)
+
+        ser = Series(["x"])
+        ser["td"] = td
+        tm.assert_series_equal(ser, expected)
+        assert isinstance(ser["td"], Timedelta)
+
+        ser = Series(["x"])
+        ser.loc["td"] = Timedelta("9 days")
+        tm.assert_series_equal(ser, expected)
+        assert isinstance(ser["td"], Timedelta)
+
+    def test_setitem_with_expansion_type_promotion(self):
+        # GH#12599
+        ser = Series(dtype=object)
+        ser["a"] = Timestamp("2016-01-01")
+        ser["b"] = 3.0
+        ser["c"] = "foo"
+        expected = Series([Timestamp("2016-01-01"), 3.0, "foo"], index=["a", "b", "c"])
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_not_contained(self, string_series):
+        # set item that's not contained
+        ser = string_series.copy()
+        assert "foobar" not in ser.index
+        ser["foobar"] = 1
+
+        app = Series([1], index=["foobar"], name="series")
+        expected = concat([string_series, app])
+        tm.assert_series_equal(ser, expected)
+
+    def test_setitem_keep_precision(self, any_numeric_ea_dtype):
+        # GH#32346
+        ser = Series([1, 2], dtype=any_numeric_ea_dtype)
+        ser[2] = 10
+        expected = Series([1, 2, 10], dtype=any_numeric_ea_dtype)
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize(
+        "na, target_na, dtype, target_dtype, indexer, raises",
+        [
+            (NA, NA, "Int64", "Int64", 1, False),
+            (NA, NA, "Int64", "Int64", 2, False),
+            (NA, np.nan, "int64", "float64", 1, False),
+            (NA, np.nan, "int64", "float64", 2, False),
+            (NaT, NaT, "int64", "object", 1, True),
+            (NaT, NaT, "int64", "object", 2, False),
+            (np.nan, NA, "Int64", "Int64", 1, False),
+            (np.nan, NA, "Int64", "Int64", 2, False),
+            (np.nan, NA, "Float64", "Float64", 1, False),
+            (np.nan, NA, "Float64", "Float64", 2, False),
+            (np.nan, np.nan, "int64", "float64", 1, False),
+            (np.nan, np.nan, "int64", "float64", 2, False),
+        ],
+    )
+    def test_setitem_enlarge_with_na(
+        self, na, target_na, dtype, target_dtype, indexer, raises
+    ):
+        # GH#32346
+        ser = Series([1, 2], dtype=dtype)
+        if raises:
+            with pytest.raises(TypeError, match="Invalid value"):
+                ser[indexer] = na
+        else:
+            ser[indexer] = na
+            expected_values = [1, target_na] if indexer == 1 else [1, 2, target_na]
+            expected = Series(expected_values, dtype=target_dtype)
+            tm.assert_series_equal(ser, expected)
+
+    def test_setitem_enlargement_object_none(self, nulls_fixture, using_infer_string):
+        # GH#48665
+        ser = Series(["a", "b"])
+        ser[3] = nulls_fixture
+        dtype = (
+            "str"
+            if using_infer_string and not isinstance(nulls_fixture, Decimal)
+            else object
+        )
+        expected = Series(["a", "b", nulls_fixture], index=[0, 1, 3], dtype=dtype)
+        tm.assert_series_equal(ser, expected)
+        if using_infer_string:
+            ser[3] is np.nan
+        else:
+            assert ser[3] is nulls_fixture
+
+
+def test_setitem_scalar_into_readonly_backing_data():
+    # GH#14359: test that you cannot mutate a read only buffer
+
+    array = np.zeros(5)
+    array.flags.writeable = False  # make the array immutable
+    series = Series(array, copy=False)
+
+    for n in series.index:
+        msg = "assignment destination is read-only"
+        with pytest.raises(ValueError, match=msg):
+            series[n] = 1
+
+        assert array[n] == 0
+
+
+def test_setitem_slice_into_readonly_backing_data():
+    # GH#14359: test that you cannot mutate a read only buffer
+
+    array = np.zeros(5)
+    array.flags.writeable = False  # make the array immutable
+    series = Series(array, copy=False)
+
+    msg = "assignment destination is read-only"
+    with pytest.raises(ValueError, match=msg):
+        series[1:3] = 1
+
+    assert not array.any()
+
+
+def test_setitem_categorical_assigning_ops():
+    orig = Series(Categorical(["b", "b"], categories=["a", "b"]))
+    ser = orig.copy()
+    ser[:] = "a"
+    exp = Series(Categorical(["a", "a"], categories=["a", "b"]))
+    tm.assert_series_equal(ser, exp)
+
+    ser = orig.copy()
+    ser[1] = "a"
+    exp = Series(Categorical(["b", "a"], categories=["a", "b"]))
+    tm.assert_series_equal(ser, exp)
+
+    ser = orig.copy()
+    ser[ser.index > 0] = "a"
+    exp = Series(Categorical(["b", "a"], categories=["a", "b"]))
+    tm.assert_series_equal(ser, exp)
+
+    ser = orig.copy()
+    ser[[False, True]] = "a"
+    exp = Series(Categorical(["b", "a"], categories=["a", "b"]))
+    tm.assert_series_equal(ser, exp)
+
+    ser = orig.copy()
+    ser.index = ["x", "y"]
+    ser["y"] = "a"
+    exp = Series(Categorical(["b", "a"], categories=["a", "b"]), index=["x", "y"])
+    tm.assert_series_equal(ser, exp)
+
+
+def test_setitem_nan_into_categorical():
+    # ensure that one can set something to np.nan
+    ser = Series(Categorical([1, 2, 3]))
+    exp = Series(Categorical([1, np.nan, 3], categories=[1, 2, 3]))
+    ser[1] = np.nan
+    tm.assert_series_equal(ser, exp)
+
+
+class TestSetitemCasting:
+    @pytest.mark.parametrize("unique", [True, False])
+    @pytest.mark.parametrize("val", [3, 3.0, "3"], ids=type)
+    def test_setitem_non_bool_into_bool(self, val, indexer_sli, unique):
+        # dont cast these 3-like values to bool
+        ser = Series([True, False])
+        if not unique:
+            ser.index = [1, 1]
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            indexer_sli(ser)[1] = val
+
+    def test_setitem_boolean_array_into_npbool(self):
+        # GH#45462
+        ser = Series([True, False, True])
+        values = ser._values
+        arr = array([True, False, None])
+
+        ser[:2] = arr[:2]  # no NAs -> can set inplace
+        assert ser._values is values
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[1:] = arr[1:]  # has an NA -> cast to boolean dtype
+
+
+class SetitemCastingEquivalents:
+    """
+    Check each of several methods that _should_ be equivalent to `obj[key] = val`
+
+    We assume that
+        - obj.index is the default Index(range(len(obj)))
+        - the setitem does not expand the obj
+    """
+
+    @pytest.fixture
+    def is_inplace(self, obj, expected):
+        """
+        Whether we expect the setting to be in-place or not.
+        """
+        return expected.dtype == obj.dtype
+
+    def check_indexer(self, obj, key, expected, val, indexer, is_inplace):
+        orig = obj
+        obj = obj.copy()
+        arr = obj._values
+
+        indexer(obj)[key] = val
+        tm.assert_series_equal(obj, expected)
+
+        self._check_inplace(is_inplace, orig, arr, obj)
+
+    def _check_inplace(self, is_inplace, orig, arr, obj):
+        if is_inplace is None:
+            # We are not (yet) checking whether setting is inplace or not
+            pass
+        elif is_inplace:
+            if arr.dtype.kind in ["m", "M"]:
+                # We may not have the same DTA/TDA, but will have the same
+                #  underlying data
+                assert arr._ndarray is obj._values._ndarray
+            else:
+                assert obj._values is arr
+        else:
+            # otherwise original array should be unchanged
+            tm.assert_equal(arr, orig._values)
+
+    def test_int_key(self, obj, key, expected, raises, val, indexer_sli, is_inplace):
+        if not isinstance(key, int):
+            pytest.skip("Not relevant for int key")
+        if raises:
+            ctx = pytest.raises(TypeError, match="Invalid value")
+        else:
+            ctx = contextlib.nullcontext()
+
+        with ctx:
+            self.check_indexer(obj, key, expected, val, indexer_sli, is_inplace)
+
+        if indexer_sli is tm.loc:
+            with ctx:
+                self.check_indexer(obj, key, expected, val, tm.at, is_inplace)
+        elif indexer_sli is tm.iloc:
+            with ctx:
+                self.check_indexer(obj, key, expected, val, tm.iat, is_inplace)
+
+        rng = range(key, key + 1)
+        with ctx:
+            self.check_indexer(obj, rng, expected, val, indexer_sli, is_inplace)
+
+        if indexer_sli is not tm.loc:
+            # Note: no .loc because that handles slice edges differently
+            slc = slice(key, key + 1)
+            with ctx:
+                self.check_indexer(obj, slc, expected, val, indexer_sli, is_inplace)
+
+        ilkey = [key]
+        with ctx:
+            self.check_indexer(obj, ilkey, expected, val, indexer_sli, is_inplace)
+
+        indkey = np.array(ilkey)
+        with ctx:
+            self.check_indexer(obj, indkey, expected, val, indexer_sli, is_inplace)
+
+        genkey = (x for x in [key])
+        with ctx:
+            self.check_indexer(obj, genkey, expected, val, indexer_sli, is_inplace)
+
+    def test_slice_key(self, obj, key, expected, raises, val, indexer_sli, is_inplace):
+        if not isinstance(key, slice):
+            pytest.skip("Not relevant for slice key")
+        if raises:
+            ctx = pytest.raises(TypeError, match="Invalid value")
+        else:
+            ctx = contextlib.nullcontext()
+
+        if indexer_sli is not tm.loc:
+            # Note: no .loc because that handles slice edges differently
+            with ctx:
+                self.check_indexer(obj, key, expected, val, indexer_sli, is_inplace)
+
+        ilkey = list(range(len(obj)))[key]
+        with ctx:
+            self.check_indexer(obj, ilkey, expected, val, indexer_sli, is_inplace)
+
+        indkey = np.array(ilkey)
+        with ctx:
+            self.check_indexer(obj, indkey, expected, val, indexer_sli, is_inplace)
+
+        genkey = (x for x in indkey)
+        with ctx:
+            self.check_indexer(obj, genkey, expected, val, indexer_sli, is_inplace)
+
+    def test_mask_key(self, obj, key, expected, raises, val, indexer_sli):
+        # setitem with boolean mask
+        mask = np.zeros(obj.shape, dtype=bool)
+        mask[key] = True
+
+        obj = obj.copy()
+
+        if is_list_like(val) and len(val) < mask.sum():
+            msg = "boolean index did not match indexed array along dimension"
+            with pytest.raises(IndexError, match=msg):
+                indexer_sli(obj)[mask] = val
+            return
+
+        if raises:
+            with pytest.raises(TypeError, match="Invalid value"):
+                indexer_sli(obj)[mask] = val
+        else:
+            indexer_sli(obj)[mask] = val
+
+    def test_series_where(self, obj, key, expected, raises, val, is_inplace):
+        mask = np.zeros(obj.shape, dtype=bool)
+        mask[key] = True
+
+        if is_list_like(val) and len(val) < len(obj):
+            # Series.where is not valid here
+            msg = "operands could not be broadcast together with shapes"
+            with pytest.raises(ValueError, match=msg):
+                obj.where(~mask, val)
+            return
+
+        orig = obj
+        obj = obj.copy()
+        arr = obj._values
+
+        res = obj.where(~mask, val)
+
+        if val is NA and res.dtype == object:
+            expected = expected.fillna(NA)
+        elif val is None and res.dtype == object:
+            assert expected.dtype == object
+            expected = expected.copy()
+            expected[expected.isna()] = None
+        tm.assert_series_equal(res, expected)
+
+        self._check_inplace(is_inplace, orig, arr, obj)
+
+    def test_index_where(self, obj, key, expected, raises, val):
+        mask = np.zeros(obj.shape, dtype=bool)
+        mask[key] = True
+
+        res = Index(obj).where(~mask, val)
+        expected_idx = Index(expected, dtype=expected.dtype)
+        tm.assert_index_equal(res, expected_idx)
+
+    def test_index_putmask(self, obj, key, expected, raises, val):
+        mask = np.zeros(obj.shape, dtype=bool)
+        mask[key] = True
+
+        res = Index(obj).putmask(mask, val)
+        tm.assert_index_equal(res, Index(expected, dtype=expected.dtype))
+
+
+@pytest.mark.parametrize(
+    "obj,expected,key,raises",
+    [
+        pytest.param(
+            # GH#45568 setting a valid NA value into IntervalDtype[int] should
+            #  cast to IntervalDtype[float]
+            Series(interval_range(1, 5)),
+            Series(
+                [Interval(1, 2), np.nan, Interval(3, 4), Interval(4, 5)],
+                dtype="interval[float64]",
+            ),
+            1,
+            True,
+            id="interval_int_na_value",
+        ),
+        pytest.param(
+            # these induce dtype changes
+            Series([2, 3, 4, 5, 6, 7, 8, 9, 10]),
+            Series([np.nan, 3, np.nan, 5, np.nan, 7, np.nan, 9, np.nan]),
+            slice(None, None, 2),
+            False,
+            id="int_series_slice_key_step",
+        ),
+        pytest.param(
+            Series([True, True, False, False]),
+            Series([np.nan, True, np.nan, False], dtype=object),
+            slice(None, None, 2),
+            True,
+            id="bool_series_slice_key_step",
+        ),
+        pytest.param(
+            # these induce dtype changes
+            Series(np.arange(10)),
+            Series([np.nan, np.nan, np.nan, np.nan, np.nan, 5, 6, 7, 8, 9]),
+            slice(None, 5),
+            False,
+            id="int_series_slice_key",
+        ),
+        pytest.param(
+            # changes dtype GH#4463
+            Series([1, 2, 3]),
+            Series([np.nan, 2, 3]),
+            0,
+            False,
+            id="int_series_int_key",
+        ),
+        pytest.param(
+            # changes dtype GH#4463
+            Series([False]),
+            Series([np.nan], dtype=object),
+            # TODO: maybe go to float64 since we are changing the _whole_ Series?
+            0,
+            True,
+            id="bool_series_int_key_change_all",
+        ),
+        pytest.param(
+            # changes dtype GH#4463
+            Series([False, True]),
+            Series([np.nan, True], dtype=object),
+            0,
+            True,
+            id="bool_series_int_key",
+        ),
+    ],
+)
+class TestSetitemCastingEquivalents(SetitemCastingEquivalents):
+    @pytest.fixture(params=[np.nan, np.float64("NaN"), None, NA])
+    def val(self, request):
+        """
+        NA values that should generally be valid_na for *all* dtypes.
+
+        Include both python float NaN and np.float64; only np.float64 has a
+        `dtype` attribute.
+        """
+        return request.param
+
+
+class TestSetitemTimedelta64IntoNumeric(SetitemCastingEquivalents):
+    # timedelta64 should not be treated as integers when setting into
+    #  numeric Series
+
+    @pytest.fixture
+    def val(self):
+        td = np.timedelta64(4, "ns")
+        return td
+        # TODO: could also try np.full((1,), td)
+
+    @pytest.fixture(params=[complex, int, float])
+    def dtype(self, request):
+        return request.param
+
+    @pytest.fixture
+    def obj(self, dtype):
+        arr = np.arange(5).astype(dtype)
+        ser = Series(arr)
+        return ser
+
+    @pytest.fixture
+    def expected(self, dtype):
+        arr = np.arange(5).astype(dtype)
+        ser = Series(arr)
+        ser = ser.astype(object)
+        ser.iloc[0] = np.timedelta64(4, "ns")
+        return ser
+
+    @pytest.fixture
+    def key(self):
+        return 0
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+class TestSetitemDT64IntoInt(SetitemCastingEquivalents):
+    # GH#39619 dont cast dt64 to int when doing this setitem
+
+    @pytest.fixture(params=["M8[ns]", "m8[ns]"])
+    def dtype(self, request):
+        return request.param
+
+    @pytest.fixture
+    def scalar(self, dtype):
+        val = np.datetime64("2021-01-18 13:25:00", "ns")
+        if dtype == "m8[ns]":
+            val = val - val
+        return val
+
+    @pytest.fixture
+    def expected(self, scalar):
+        expected = Series([scalar, scalar, 3], dtype=object)
+        assert isinstance(expected[0], type(scalar))
+        return expected
+
+    @pytest.fixture
+    def obj(self):
+        return Series([1, 2, 3])
+
+    @pytest.fixture
+    def key(self):
+        return slice(None, -1)
+
+    @pytest.fixture(params=[None, list, np.array])
+    def val(self, scalar, request):
+        box = request.param
+        if box is None:
+            return scalar
+        return box([scalar, scalar])
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+class TestSetitemNAPeriodDtype(SetitemCastingEquivalents):
+    # Setting compatible NA values into Series with PeriodDtype
+
+    @pytest.fixture
+    def expected(self, key):
+        exp = Series(period_range("2000-01-01", periods=10, freq="D"))
+        exp._values.view("i8")[key] = NaT._value
+        assert exp[key] is NaT or all(x is NaT for x in exp[key])
+        return exp
+
+    @pytest.fixture
+    def obj(self):
+        return Series(period_range("2000-01-01", periods=10, freq="D"))
+
+    @pytest.fixture(params=[3, slice(3, 5)])
+    def key(self, request):
+        return request.param
+
+    @pytest.fixture(params=[None, np.nan])
+    def val(self, request):
+        return request.param
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+class TestSetitemNADatetimeLikeDtype(SetitemCastingEquivalents):
+    # some nat-like values should be cast to datetime64/timedelta64 when
+    #  inserting into a datetime64/timedelta64 series.  Others should coerce
+    #  to object and retain their dtypes.
+    # GH#18586 for td64 and boolean mask case
+
+    @pytest.fixture(
+        params=["m8[ns]", "M8[ns]", "datetime64[ns, UTC]", "datetime64[ns, US/Central]"]
+    )
+    def dtype(self, request):
+        return request.param
+
+    @pytest.fixture
+    def obj(self, dtype):
+        i8vals = date_range("2016-01-01", periods=3).asi8
+        idx = Index(i8vals, dtype=dtype)
+        assert idx.dtype == dtype
+        return Series(idx)
+
+    @pytest.fixture(
+        params=[
+            None,
+            np.nan,
+            NaT,
+            np.timedelta64("NaT", "ns"),
+            np.datetime64("NaT", "ns"),
+        ]
+    )
+    def val(self, request):
+        return request.param
+
+    @pytest.fixture
+    def is_inplace(self, val, obj):
+        # td64   -> cast to object iff val is datetime64("NaT")
+        # dt64   -> cast to object iff val is timedelta64("NaT")
+        # dt64tz -> cast to object with anything _but_ NaT
+        return val is NaT or val is None or val is np.nan or obj.dtype == val.dtype
+
+    @pytest.fixture
+    def expected(self, obj, val, is_inplace):
+        dtype = obj.dtype if is_inplace else object
+        expected = Series([val] + list(obj[1:]), dtype=dtype)
+        return expected
+
+    @pytest.fixture
+    def key(self):
+        return 0
+
+    @pytest.fixture
+    def raises(self, is_inplace):
+        return False if is_inplace else True
+
+
+class TestSetitemMismatchedTZCastsToObject(SetitemCastingEquivalents):
+    # GH#24024
+    @pytest.fixture
+    def obj(self):
+        return Series(date_range("2000", periods=2, tz="US/Central"))
+
+    @pytest.fixture
+    def val(self):
+        return Timestamp("2000", tz="US/Eastern")
+
+    @pytest.fixture
+    def key(self):
+        return 0
+
+    @pytest.fixture
+    def expected(self, obj, val):
+        # pre-2.0 this would cast to object, in 2.0 we cast the val to
+        #  the target tz
+        expected = Series(
+            [
+                val.tz_convert("US/Central"),
+                Timestamp("2000-01-02 00:00:00-06:00", tz="US/Central"),
+            ],
+            dtype=obj.dtype,
+        )
+        return expected
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "obj,expected",
+    [
+        # For numeric series, we should coerce to NaN.
+        (Series([1, 2, 3]), Series([np.nan, 2, 3])),
+        (Series([1.0, 2.0, 3.0]), Series([np.nan, 2.0, 3.0])),
+        # For datetime series, we should coerce to NaT.
+        (
+            Series([datetime(2000, 1, 1), datetime(2000, 1, 2), datetime(2000, 1, 3)]),
+            Series([NaT, datetime(2000, 1, 2), datetime(2000, 1, 3)]),
+        ),
+        # For objects, we should preserve the None value.
+        (Series(["foo", "bar", "baz"]), Series([None, "bar", "baz"])),
+    ],
+)
+class TestSeriesNoneCoercion(SetitemCastingEquivalents):
+    @pytest.fixture
+    def key(self):
+        return 0
+
+    @pytest.fixture
+    def val(self):
+        return None
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+class TestSetitemFloatIntervalWithIntIntervalValues(SetitemCastingEquivalents):
+    # GH#44201 Cast to shared IntervalDtype rather than object
+
+    def test_setitem_example(self):
+        # Just a case here to make obvious what this test class is aimed at
+        idx = IntervalIndex.from_breaks(range(4))
+        obj = Series(idx)
+        val = Interval(0.5, 1.5)
+
+        with pytest.raises(TypeError, match="Invalid value"):
+            obj[0] = val
+
+    @pytest.fixture
+    def obj(self):
+        """
+        Fixture to create a Series [(0, 1], (1, 2], (2, 3]]
+        """
+        idx = IntervalIndex.from_breaks(range(4))
+        return Series(idx)
+
+    @pytest.fixture
+    def val(self):
+        """
+        Fixture to get an interval (0.5, 1.5]
+        """
+        return Interval(0.5, 1.5)
+
+    @pytest.fixture
+    def key(self):
+        """
+        Fixture to get a key 0
+        """
+        return 0
+
+    @pytest.fixture
+    def expected(self, obj, val):
+        """
+        Fixture to get a Series [(0.5, 1.5], (1.0, 2.0], (2.0, 3.0]]
+        """
+        data = [val] + list(obj[1:])
+        idx = IntervalIndex(data, dtype="Interval[float64]")
+        return Series(idx)
+
+    @pytest.fixture
+    def raises(self):
+        """
+        Fixture to enable raising pytest exceptions
+        """
+        return True
+
+
+class TestSetitemRangeIntoIntegerSeries(SetitemCastingEquivalents):
+    # GH#44261 Setting a range with sufficiently-small integers into
+    #  small-itemsize integer dtypes should not need to upcast
+
+    @pytest.fixture
+    def obj(self, any_int_numpy_dtype):
+        dtype = np.dtype(any_int_numpy_dtype)
+        ser = Series(range(5), dtype=dtype)
+        return ser
+
+    @pytest.fixture
+    def val(self):
+        return range(2, 4)
+
+    @pytest.fixture
+    def key(self):
+        return slice(0, 2)
+
+    @pytest.fixture
+    def expected(self, any_int_numpy_dtype):
+        dtype = np.dtype(any_int_numpy_dtype)
+        exp = Series([2, 3, 2, 3, 4], dtype=dtype)
+        return exp
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "val, raises",
+    [
+        (np.array([2.0, 3.0]), False),
+        (np.array([2.5, 3.5]), True),
+        (
+            np.array([2**65, 2**65 + 1], dtype=np.float64),
+            True,
+        ),  # all ints, but can't cast
+    ],
+)
+class TestSetitemFloatNDarrayIntoIntegerSeries(SetitemCastingEquivalents):
+    @pytest.fixture
+    def obj(self):
+        return Series(range(5), dtype=np.int64)
+
+    @pytest.fixture
+    def key(self):
+        return slice(0, 2)
+
+    @pytest.fixture
+    def expected(self, val):
+        if val[0] == 2:
+            # NB: this condition is based on currently-hardcoded "val" cases
+            dtype = np.int64
+        else:
+            dtype = np.float64
+        res_values = np.array(range(5), dtype=dtype)
+        res_values[:2] = val
+        return Series(res_values)
+
+
+@pytest.mark.parametrize("val", [512, np.int16(512)])
+class TestSetitemIntoIntegerSeriesNeedsUpcast(SetitemCastingEquivalents):
+    @pytest.fixture
+    def obj(self):
+        return Series([1, 2, 3], dtype=np.int8)
+
+    @pytest.fixture
+    def key(self):
+        return 1
+
+    @pytest.fixture
+    def expected(self):
+        return Series([1, 512, 3], dtype=np.int16)
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+@pytest.mark.parametrize("val", [2**33 + 1.0, 2**33 + 1.1, 2**62])
+class TestSmallIntegerSetitemUpcast(SetitemCastingEquivalents):
+    # https://github.com/pandas-dev/pandas/issues/39584#issuecomment-941212124
+    @pytest.fixture
+    def obj(self):
+        return Series([1, 2, 3], dtype="i4")
+
+    @pytest.fixture
+    def key(self):
+        return 0
+
+    @pytest.fixture
+    def expected(self, val):
+        if val % 1 != 0:
+            dtype = "f8"
+        else:
+            dtype = "i8"
+        return Series([val, 2, 3], dtype=dtype)
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+class CoercionTest(SetitemCastingEquivalents):
+    # Tests ported from tests.indexing.test_coercion
+
+    @pytest.fixture
+    def key(self):
+        return 1
+
+    @pytest.fixture
+    def expected(self, obj, key, val, exp_dtype):
+        vals = list(obj)
+        vals[key] = val
+        return Series(vals, dtype=exp_dtype)
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [(np.int32(1), np.int8, None), (np.int16(2**9), np.int16, True)],
+)
+class TestCoercionInt8(CoercionTest):
+    # previously test_setitem_series_int8 in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series([1, 2, 3, 4], dtype=np.int8)
+
+
+@pytest.mark.parametrize("val", [1, 1.1, 1 + 1j, True])
+@pytest.mark.parametrize("exp_dtype", [object])
+class TestCoercionObject(CoercionTest):
+    # previously test_setitem_series_object in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series(["a", "b", "c", "d"], dtype=object)
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, object, True),
+        ("e", StringDtype(na_value=np.nan), False),
+    ],
+)
+class TestCoercionString(CoercionTest):
+    @pytest.fixture
+    def obj(self):
+        return Series(["a", "b", "c", "d"], dtype=StringDtype(na_value=np.nan))
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, np.complex128, False),
+        (1.1, np.complex128, False),
+        (1 + 1j, np.complex128, False),
+        (True, object, True),
+    ],
+)
+class TestCoercionComplex(CoercionTest):
+    # previously test_setitem_series_complex128 in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series([1 + 1j, 2 + 2j, 3 + 3j, 4 + 4j])
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, object, True),
+        ("3", object, True),
+        (3, object, True),
+        (1.1, object, True),
+        (1 + 1j, object, True),
+        (True, bool, False),
+    ],
+)
+class TestCoercionBool(CoercionTest):
+    # previously test_setitem_series_bool in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series([True, False, True, False], dtype=bool)
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, np.int64, False),
+        (1.1, np.float64, True),
+        (1 + 1j, np.complex128, True),
+        (True, object, True),
+    ],
+)
+class TestCoercionInt64(CoercionTest):
+    # previously test_setitem_series_int64 in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series([1, 2, 3, 4])
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, np.float64, False),
+        (1.1, np.float64, False),
+        (1 + 1j, np.complex128, True),
+        (True, object, True),
+    ],
+)
+class TestCoercionFloat64(CoercionTest):
+    # previously test_setitem_series_float64 in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series([1.1, 2.2, 3.3, 4.4])
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (1, np.float32, False),
+        pytest.param(
+            1.1,
+            np.float32,
+            False,
+            marks=pytest.mark.xfail(
+                (
+                    not np_version_gte1p24
+                    or (
+                        np_version_gte1p24
+                        and os.environ.get("NPY_PROMOTION_STATE", "weak") != "weak"
+                    )
+                ),
+                reason="np.float32(1.1) ends up as 1.100000023841858, so "
+                "np_can_hold_element raises and we cast to float64",
+            ),
+        ),
+        (1 + 1j, np.complex128, True),
+        (True, object, True),
+        (np.uint8(2), np.float32, False),
+        (np.uint32(2), np.float32, False),
+        # float32 cannot hold np.iinfo(np.uint32).max exactly
+        # (closest it can hold is 4294967300.0 which off by 5.0), so
+        # we cast to float64
+        (np.uint32(np.iinfo(np.uint32).max), np.float64, True),
+        (np.uint64(2), np.float32, False),
+        (np.int64(2), np.float32, False),
+    ],
+)
+class TestCoercionFloat32(CoercionTest):
+    @pytest.fixture
+    def obj(self):
+        return Series([1.1, 2.2, 3.3, 4.4], dtype=np.float32)
+
+    def test_slice_key(self, obj, key, expected, raises, val, indexer_sli, is_inplace):
+        super().test_slice_key(obj, key, expected, raises, val, indexer_sli, is_inplace)
+
+        if isinstance(val, float):
+            # the xfail would xpass bc test_slice_key short-circuits
+            raise AssertionError("xfail not relevant for this test.")
+
+
+@pytest.mark.parametrize(
+    "exp_dtype",
+    [
+        "M8[ms]",
+        "M8[ms, UTC]",
+        "m8[ms]",
+    ],
+)
+class TestCoercionDatetime64HigherReso(CoercionTest):
+    @pytest.fixture
+    def obj(self, exp_dtype):
+        idx = date_range("2011-01-01", freq="D", periods=4, unit="s")
+        if exp_dtype == "m8[ms]":
+            idx = idx - Timestamp("1970-01-01")
+            assert idx.dtype == "m8[s]"
+        elif exp_dtype == "M8[ms, UTC]":
+            idx = idx.tz_localize("UTC")
+        return Series(idx)
+
+    @pytest.fixture
+    def val(self, exp_dtype):
+        ts = Timestamp("2011-01-02 03:04:05.678").as_unit("ms")
+        if exp_dtype == "m8[ms]":
+            return ts - Timestamp("1970-01-01")
+        elif exp_dtype == "M8[ms, UTC]":
+            return ts.tz_localize("UTC")
+        return ts
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (Timestamp("2012-01-01"), "datetime64[ns]", False),
+        (1, object, True),
+        ("x", object, True),
+    ],
+)
+class TestCoercionDatetime64(CoercionTest):
+    # previously test_setitem_series_datetime64 in tests.indexing.test_coercion
+
+    @pytest.fixture
+    def obj(self):
+        return Series(date_range("2011-01-01", freq="D", periods=4))
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (Timestamp("2012-01-01", tz="US/Eastern"), "datetime64[ns, US/Eastern]", False),
+        # pre-2.0, a mis-matched tz would end up casting to object
+        (Timestamp("2012-01-01", tz="US/Pacific"), "datetime64[ns, US/Eastern]", False),
+        (Timestamp("2012-01-01"), object, True),
+        (1, object, True),
+    ],
+)
+class TestCoercionDatetime64TZ(CoercionTest):
+    # previously test_setitem_series_datetime64tz in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        tz = "US/Eastern"
+        return Series(date_range("2011-01-01", freq="D", periods=4, tz=tz))
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "val,exp_dtype,raises",
+    [
+        (Timedelta("12 day"), "timedelta64[ns]", False),
+        (1, object, True),
+        ("x", object, True),
+    ],
+)
+class TestCoercionTimedelta64(CoercionTest):
+    # previously test_setitem_series_timedelta64 in tests.indexing.test_coercion
+    @pytest.fixture
+    def obj(self):
+        return Series(timedelta_range("1 day", periods=4))
+
+    @pytest.fixture
+    def raises(self):
+        return False
+
+
+@pytest.mark.parametrize(
+    "val", ["foo", Period("2016", freq="Y"), Interval(1, 2, closed="both")]
+)
+@pytest.mark.parametrize("exp_dtype", [object])
+class TestPeriodIntervalCoercion(CoercionTest):
+    # GH#45768
+    @pytest.fixture(
+        params=[
+            period_range("2016-01-01", periods=3, freq="D"),
+            interval_range(1, 5),
+        ]
+    )
+    def obj(self, request):
+        return Series(request.param)
+
+    @pytest.fixture
+    def raises(self):
+        return True
+
+
+def test_20643():
+    # closed by GH#45121
+    orig = Series([0, 1, 2], index=["a", "b", "c"])
+
+    ser = orig.copy()
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.at["b"] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.loc["b"] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser["b"] = 2.7
+
+    ser = orig.copy()
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.iat[1] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.iloc[1] = 2.7
+
+    orig_df = orig.to_frame("A")
+
+    df = orig_df.copy()
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.at["b", "A"] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.loc["b", "A"] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.iloc[1, 0] = 2.7
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        df.iat[1, 0] = 2.7
+
+
+def test_20643_comment():
+    # https://github.com/pandas-dev/pandas/issues/20643#issuecomment-431244590
+    # fixed sometime prior to GH#45121
+    orig = Series([0, 1, 2], index=["a", "b", "c"])
+    expected = Series([np.nan, 1, 2], index=["a", "b", "c"])
+
+    ser = orig.copy()
+    ser.iat[0] = None
+    tm.assert_series_equal(ser, expected)
+
+    ser = orig.copy()
+    ser.iloc[0] = None
+    tm.assert_series_equal(ser, expected)
+
+
+def test_15413():
+    # fixed by GH#45121
+    ser = Series([1, 2, 3])
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser[ser == 2] += 0.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser[1] += 0.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.loc[1] += 0.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.iloc[1] += 0.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.iat[1] += 0.5
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser.at[1] += 0.5
+
+
+def test_32878_int_itemsize():
+    # Fixed by GH#45121
+    arr = np.arange(5).astype("i4")
+    ser = Series(arr)
+    val = np.int64(np.iinfo(np.int64).max)
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser[0] = val
+
+
+def test_32878_complex_itemsize():
+    arr = np.arange(5).astype("c8")
+    ser = Series(arr)
+    val = np.finfo(np.float64).max
+    val = val.astype("c16")
+
+    # GH#32878 used to coerce val to inf+0.000000e+00j
+    with pytest.raises(TypeError, match="Invalid value"):
+        ser[0] = val
+
+
+def test_37692(indexer_al):
+    # GH#37692
+    ser = Series([1, 2, 3], index=["a", "b", "c"])
+    with pytest.raises(TypeError, match="Invalid value"):
+        indexer_al(ser)["b"] = "test"
+
+
+def test_setitem_bool_int_float_consistency(indexer_sli):
+    # GH#21513
+    # bool-with-int and bool-with-float both upcast to object
+    #  int-with-float and float-with-int are both non-casting so long
+    #  as the setitem can be done losslessly
+    for dtype in [np.float64, np.int64]:
+        ser = Series(0, index=range(3), dtype=dtype)
+        with pytest.raises(TypeError, match="Invalid value"):
+            indexer_sli(ser)[0] = True
+
+        ser = Series(0, index=range(3), dtype=bool)
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser[0] = dtype(1)
+
+    # 1.0 can be held losslessly, so no casting
+    ser = Series(0, index=range(3), dtype=np.int64)
+    indexer_sli(ser)[0] = np.float64(1.0)
+    assert ser.dtype == np.int64
+
+    # 1 can be held losslessly, so no casting
+    ser = Series(0, index=range(3), dtype=np.float64)
+    indexer_sli(ser)[0] = np.int64(1)
+
+
+def test_setitem_positional_with_casting():
+    # GH#45070 case where in __setitem__ we get a KeyError, then when
+    #  we fallback we *also* get a ValueError if we try to set inplace.
+    # As of 3.0 we always treat int keys as labels, so this becomes
+    #  setitem-with-expansion
+    ser = Series([1, 2, 3], index=["a", "b", "c"])
+
+    ser[0] = "X"
+    expected = Series([1, 2, 3, "X"], index=["a", "b", "c", 0], dtype=object)
+    tm.assert_series_equal(ser, expected)
+
+
+def test_setitem_positional_float_into_int_coerces():
+    # Case where we hit a KeyError and then trying to set in-place incorrectly
+    #  casts a float to an int;
+    # As of 3.0 we always treat int keys as labels, so this becomes
+    #  setitem-with-expansion
+    ser = Series([1, 2, 3], index=["a", "b", "c"])
+
+    ser[0] = 1.5
+    expected = Series([1, 2, 3, 1.5], index=["a", "b", "c", 0])
+    tm.assert_series_equal(ser, expected)
+
+
+def test_setitem_int_not_positional():
+    # GH#42215 deprecated falling back to positional on __setitem__ with an
+    #  int not contained in the index; enforced in 2.0
+    ser = Series([1, 2, 3, 4], index=[1.1, 2.1, 3.0, 4.1])
+    assert not ser.index._should_fallback_to_positional
+    # assert not ser.index.astype(object)._should_fallback_to_positional
+
+    # 3.0 is in our index, so post-enforcement behavior is unchanged
+    ser[3] = 10
+    expected = Series([1, 2, 10, 4], index=ser.index)
+    tm.assert_series_equal(ser, expected)
+
+    # pre-enforcement `ser[5] = 5` raised IndexError
+    ser[5] = 5
+    expected = Series([1, 2, 10, 4, 5], index=[1.1, 2.1, 3.0, 4.1, 5.0])
+    tm.assert_series_equal(ser, expected)
+
+    ii = IntervalIndex.from_breaks(range(10))[::2]
+    ser2 = Series(range(len(ii)), index=ii)
+    exp_index = ii.astype(object).append(Index([4]))
+    expected2 = Series([0, 1, 2, 3, 4, 9], index=exp_index)
+    # pre-enforcement `ser2[4] = 9` interpreted 4 as positional
+    ser2[4] = 9
+    tm.assert_series_equal(ser2, expected2)
+
+    mi = MultiIndex.from_product([ser.index, ["A", "B"]])
+    ser3 = Series(range(len(mi)), index=mi)
+    expected3 = ser3.copy()
+    expected3.loc[4] = 99
+    # pre-enforcement `ser3[4] = 99` interpreted 4 as positional
+    ser3[4] = 99
+    tm.assert_series_equal(ser3, expected3)
+
+
+def test_setitem_with_bool_indexer():
+    # GH#42530
+
+    df = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    result = df.pop("b").copy()
+    result[[True, False, False]] = 9
+    expected = Series(data=[9, 5, 6], name="b")
+    tm.assert_series_equal(result, expected)
+
+    df.loc[[True, False, False], "a"] = 10
+    expected = DataFrame({"a": [10, 2, 3]})
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize("size", range(2, 6))
+@pytest.mark.parametrize(
+    "mask", [[True, False, False, False, False], [True, False], [False]]
+)
+@pytest.mark.parametrize(
+    "item", [2.0, np.nan, np.finfo(float).max, np.finfo(float).min]
+)
+@pytest.mark.parametrize("box", [np.array, list, tuple])
+def test_setitem_bool_indexer_dont_broadcast_length1_values(size, mask, item, box):
+    # GH#44265
+    # see also tests.series.indexing.test_where.test_broadcast
+
+    selection = np.resize(mask, size)
+
+    data = np.arange(size, dtype=float)
+
+    ser = Series(data)
+
+    if selection.sum() != 1:
+        msg = (
+            "cannot set using a list-like indexer with a different "
+            "length than the value"
+        )
+        with pytest.raises(ValueError, match=msg):
+            # GH#44265
+            ser[selection] = box([item])
+    else:
+        # In this corner case setting is equivalent to setting with the unboxed
+        #  item
+        ser[selection] = box([item])
+
+        expected = Series(np.arange(size, dtype=float))
+        expected[selection] = item
+        tm.assert_series_equal(ser, expected)
+
+
+def test_setitem_empty_mask_dont_upcast_dt64():
+    dti = date_range("2016-01-01", periods=3)
+    ser = Series(dti)
+    orig = ser.copy()
+    mask = np.zeros(3, dtype=bool)
+
+    ser[mask] = "foo"
+    assert ser.dtype == dti.dtype  # no-op -> dont upcast
+    tm.assert_series_equal(ser, orig)
+
+    ser.mask(mask, "foo", inplace=True)
+    assert ser.dtype == dti.dtype  # no-op -> dont upcast
+    tm.assert_series_equal(ser, orig)
diff --git a/pandas/pandas/tests/series/indexing/test_take.py b/pandas/pandas/tests/series/indexing/test_take.py
new file mode 100644
index 0000000000000..43fbae8908966
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_take.py
@@ -0,0 +1,50 @@
+import pytest
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+
+def test_take_validate_axis():
+    # GH#51022
+    ser = Series([-1, 5, 6, 2, 4])
+
+    msg = "No axis named foo for object type Series"
+    with pytest.raises(ValueError, match=msg):
+        ser.take([1, 2], axis="foo")
+
+
+def test_take():
+    ser = Series([-1, 5, 6, 2, 4])
+
+    actual = ser.take([1, 3, 4])
+    expected = Series([5, 2, 4], index=[1, 3, 4])
+    tm.assert_series_equal(actual, expected)
+
+    actual = ser.take([-1, 3, 4])
+    expected = Series([4, 2, 4], index=[4, 3, 4])
+    tm.assert_series_equal(actual, expected)
+
+    msg = "indices are out-of-bounds"
+    with pytest.raises(IndexError, match=msg):
+        ser.take([1, 10])
+    with pytest.raises(IndexError, match=msg):
+        ser.take([2, 5])
+
+
+def test_take_categorical():
+    # https://github.com/pandas-dev/pandas/issues/20664
+    ser = Series(pd.Categorical(["a", "b", "c"]))
+    result = ser.take([-2, -2, 0])
+    expected = Series(
+        pd.Categorical(["b", "b", "a"], categories=["a", "b", "c"]), index=[1, 1, 0]
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_take_slice_raises():
+    ser = Series([-1, 5, 6, 2, 4])
+
+    msg = "Series.take requires a sequence of integers, not slice"
+    with pytest.raises(TypeError, match=msg):
+        ser.take(slice(0, 3, 1))
diff --git a/pandas/pandas/tests/series/indexing/test_where.py b/pandas/pandas/tests/series/indexing/test_where.py
new file mode 100644
index 0000000000000..663ee8ad0ee38
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_where.py
@@ -0,0 +1,445 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_integer
+
+import pandas as pd
+from pandas import (
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+
+
+def test_where_unsafe_int(any_signed_int_numpy_dtype):
+    s = Series(np.arange(10), dtype=any_signed_int_numpy_dtype)
+    mask = s < 5
+
+    s[mask] = range(2, 7)
+    expected = Series(
+        list(range(2, 7)) + list(range(5, 10)),
+        dtype=any_signed_int_numpy_dtype,
+    )
+
+    tm.assert_series_equal(s, expected)
+
+
+def test_where_unsafe_float(float_numpy_dtype):
+    s = Series(np.arange(10), dtype=float_numpy_dtype)
+    mask = s < 5
+
+    s[mask] = range(2, 7)
+    data = list(range(2, 7)) + list(range(5, 10))
+    expected = Series(data, dtype=float_numpy_dtype)
+
+    tm.assert_series_equal(s, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype,expected_dtype",
+    [
+        (np.int8, np.float64),
+        (np.int16, np.float64),
+        (np.int32, np.float64),
+        (np.int64, np.float64),
+        (np.float32, np.float32),
+        (np.float64, np.float64),
+    ],
+)
+def test_where_unsafe_upcast(dtype, expected_dtype):
+    # see gh-9743
+    s = Series(np.arange(10), dtype=dtype)
+    values = [2.5, 3.5, 4.5, 5.5, 6.5]
+    mask = s < 5
+    if np.dtype(dtype).kind == np.dtype(expected_dtype).kind == "f":
+        s[mask] = values
+        expected = Series(values + list(range(5, 10)), dtype=expected_dtype)
+        tm.assert_series_equal(s, expected)
+    else:
+        with pytest.raises(TypeError, match="Invalid value"):
+            s[mask] = values
+
+
+def test_where_unsafe():
+    # see gh-9731
+    s = Series(np.arange(10), dtype="int64")
+    values = [2.5, 3.5, 4.5, 5.5]
+
+    mask = s > 5
+    expected = Series(list(range(6)) + values, dtype="float64")
+
+    with pytest.raises(TypeError, match="Invalid value"):
+        s[mask] = values
+    s = s.astype("float64")
+    s[mask] = values
+
+    # see gh-3235
+    s = Series(np.arange(10), dtype="int64")
+    mask = s < 5
+    s[mask] = range(2, 7)
+    expected = Series(list(range(2, 7)) + list(range(5, 10)), dtype="int64")
+    tm.assert_series_equal(s, expected)
+    assert s.dtype == expected.dtype
+
+    s = Series(np.arange(10), dtype="int64")
+    mask = s > 5
+    s[mask] = [0] * 4
+    expected = Series([0, 1, 2, 3, 4, 5] + [0] * 4, dtype="int64")
+    tm.assert_series_equal(s, expected)
+
+    s = Series(np.arange(10))
+    mask = s > 5
+
+    msg = "cannot set using a list-like indexer with a different length than the value"
+    with pytest.raises(ValueError, match=msg):
+        s[mask] = [5, 4, 3, 2, 1]
+
+    with pytest.raises(ValueError, match=msg):
+        s[mask] = [0] * 5
+
+    # dtype changes
+    s = Series([1, 2, 3, 4])
+    result = s.where(s > 2, np.nan)
+    expected = Series([np.nan, np.nan, 3, 4])
+    tm.assert_series_equal(result, expected)
+
+    # GH 4667
+    # setting with None changes dtype
+    s = Series(range(10)).astype(float)
+    s[8] = None
+    result = s[8]
+    assert isna(result)
+
+    s = Series(range(10)).astype(float)
+    s[s > 8] = None
+    result = s[isna(s)]
+    expected = Series(np.nan, index=[9])
+    tm.assert_series_equal(result, expected)
+
+
+def test_where():
+    s = Series(np.random.default_rng(2).standard_normal(5))
+    cond = s > 0
+
+    rs = s.where(cond).dropna()
+    rs2 = s[cond]
+    tm.assert_series_equal(rs, rs2)
+
+    rs = s.where(cond, -s)
+    tm.assert_series_equal(rs, s.abs())
+
+    rs = s.where(cond)
+    assert s.shape == rs.shape
+    assert rs is not s
+
+    # test alignment
+    cond = Series([True, False, False, True, False], index=s.index)
+    s2 = -(s.abs())
+
+    expected = s2[cond].reindex(s2.index[:3]).reindex(s2.index)
+    rs = s2.where(cond[:3])
+    tm.assert_series_equal(rs, expected)
+
+    expected = s2.abs()
+    expected.iloc[0] = s2[0]
+    rs = s2.where(cond[:3], -s2)
+    tm.assert_series_equal(rs, expected)
+
+
+def test_where_error():
+    s = Series(np.random.default_rng(2).standard_normal(5))
+    cond = s > 0
+
+    msg = "Array conditional must be same shape as self"
+    with pytest.raises(ValueError, match=msg):
+        s.where(1)
+    with pytest.raises(ValueError, match=msg):
+        s.where(cond[:3].values, -s)
+
+    # GH 2745
+    s = Series([1, 2])
+    s[[True, False]] = [0, 1]
+    expected = Series([0, 2])
+    tm.assert_series_equal(s, expected)
+
+    # failures
+    msg = "cannot set using a list-like indexer with a different length than the value"
+    with pytest.raises(ValueError, match=msg):
+        s[[True, False]] = [0, 2, 3]
+
+    with pytest.raises(ValueError, match=msg):
+        s[[True, False]] = []
+
+
+@pytest.mark.parametrize("klass", [list, tuple, np.array, Series])
+def test_where_array_like(klass):
+    # see gh-15414
+    s = Series([1, 2, 3])
+    cond = [False, True, True]
+    expected = Series([np.nan, 2, 3])
+
+    result = s.where(klass(cond))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "cond",
+    [
+        [1, 0, 1],
+        Series([2, 5, 7]),
+        ["True", "False", "True"],
+        [Timestamp("2017-01-01"), pd.NaT, Timestamp("2017-01-02")],
+    ],
+)
+def test_where_invalid_input(cond):
+    # see gh-15414: only boolean arrays accepted
+    s = Series([1, 2, 3])
+    msg = "Boolean array expected for the condition"
+
+    with pytest.raises(TypeError, match=msg):
+        s.where(cond)
+
+    msg = "Array conditional must be same shape as self"
+    with pytest.raises(ValueError, match=msg):
+        s.where([True])
+
+
+def test_where_ndframe_align():
+    msg = "Array conditional must be same shape as self"
+    s = Series([1, 2, 3])
+
+    cond = [True]
+    with pytest.raises(ValueError, match=msg):
+        s.where(cond)
+
+    expected = Series([1, np.nan, np.nan])
+
+    out = s.where(Series(cond))
+    tm.assert_series_equal(out, expected)
+
+    cond = np.array([False, True, False, True])
+    with pytest.raises(ValueError, match=msg):
+        s.where(cond)
+
+    expected = Series([np.nan, 2, np.nan])
+
+    out = s.where(Series(cond))
+    tm.assert_series_equal(out, expected)
+
+
+def test_where_setitem_invalid():
+    # GH 2702
+    # make sure correct exceptions are raised on invalid list assignment
+
+    msg = (
+        lambda x: f"cannot set using a {x} indexer with a "
+        "different length than the value"
+    )
+    # slice
+    s = Series(list("abc"), dtype=object)
+
+    with pytest.raises(ValueError, match=msg("slice")):
+        s[0:3] = list(range(27))
+
+    s[0:3] = list(range(3))
+    expected = Series([0, 1, 2])
+    tm.assert_series_equal(s.astype(np.int64), expected)
+
+    # slice with step
+    s = Series(list("abcdef"), dtype=object)
+
+    with pytest.raises(ValueError, match=msg("slice")):
+        s[0:4:2] = list(range(27))
+
+    s = Series(list("abcdef"), dtype=object)
+    s[0:4:2] = list(range(2))
+    expected = Series([0, "b", 1, "d", "e", "f"])
+    tm.assert_series_equal(s, expected)
+
+    # neg slices
+    s = Series(list("abcdef"), dtype=object)
+
+    with pytest.raises(ValueError, match=msg("slice")):
+        s[:-1] = list(range(27))
+
+    s[-3:-1] = list(range(2))
+    expected = Series(["a", "b", "c", 0, 1, "f"])
+    tm.assert_series_equal(s, expected)
+
+    # list
+    s = Series(list("abc"), dtype=object)
+
+    with pytest.raises(ValueError, match=msg("list-like")):
+        s[[0, 1, 2]] = list(range(27))
+
+    s = Series(list("abc"), dtype=object)
+
+    with pytest.raises(ValueError, match=msg("list-like")):
+        s[[0, 1, 2]] = list(range(2))
+
+    # scalar
+    s = Series(list("abc"), dtype=object)
+    s[0] = list(range(10))
+    expected = Series([list(range(10)), "b", "c"])
+    tm.assert_series_equal(s, expected)
+
+
+@pytest.mark.parametrize("size", range(2, 6))
+@pytest.mark.parametrize(
+    "mask", [[True, False, False, False, False], [True, False], [False]]
+)
+@pytest.mark.parametrize(
+    "item", [2.0, np.nan, np.finfo(float).max, np.finfo(float).min]
+)
+@pytest.mark.parametrize("box", [np.array, list, tuple])
+def test_broadcast(size, mask, item, box):
+    # GH#8801, GH#4195
+    selection = np.resize(mask, size)
+
+    data = np.arange(size, dtype=float)
+
+    # Construct the expected series by taking the source
+    # data or item based on the selection
+    expected = Series(
+        [item if use_item else data[i] for i, use_item in enumerate(selection)]
+    )
+
+    s = Series(data)
+
+    s[selection] = item
+    tm.assert_series_equal(s, expected)
+
+    s = Series(data)
+    result = s.where(~selection, box([item]))
+    tm.assert_series_equal(result, expected)
+
+    s = Series(data)
+    result = s.mask(selection, box([item]))
+    tm.assert_series_equal(result, expected)
+
+
+def test_where_inplace():
+    s = Series(np.random.default_rng(2).standard_normal(5))
+    cond = s > 0
+
+    rs = s.copy()
+
+    rs.where(cond, inplace=True)
+    tm.assert_series_equal(rs.dropna(), s[cond])
+    tm.assert_series_equal(rs, s.where(cond))
+
+    rs = s.copy()
+    rs.where(cond, -s, inplace=True)
+    tm.assert_series_equal(rs, s.where(cond, -s))
+
+
+def test_where_dups():
+    # GH 4550
+    # where crashes with dups in index
+    s1 = Series(list(range(3)))
+    s2 = Series(list(range(3)))
+    comb = pd.concat([s1, s2])
+    result = comb.where(comb < 2)
+    expected = Series([0, 1, np.nan, 0, 1, np.nan], index=[0, 1, 2, 0, 1, 2])
+    tm.assert_series_equal(result, expected)
+
+    # GH 4548
+    # inplace updating not working with dups
+    comb[comb < 1] = 5
+    expected = Series([5, 1, 2, 5, 1, 2], index=[0, 1, 2, 0, 1, 2])
+    tm.assert_series_equal(comb, expected)
+
+    comb[comb < 2] += 10
+    expected = Series([5, 11, 2, 5, 11, 2], index=[0, 1, 2, 0, 1, 2])
+    tm.assert_series_equal(comb, expected)
+
+
+def test_where_numeric_with_string():
+    # GH 9280
+    s = Series([1, 2, 3])
+    w = s.where(s > 1, "X")
+
+    assert not is_integer(w[0])
+    assert is_integer(w[1])
+    assert is_integer(w[2])
+    assert isinstance(w[0], str)
+    assert w.dtype == "object"
+
+    w = s.where(s > 1, ["X", "Y", "Z"])
+    assert not is_integer(w[0])
+    assert is_integer(w[1])
+    assert is_integer(w[2])
+    assert isinstance(w[0], str)
+    assert w.dtype == "object"
+
+    w = s.where(s > 1, np.array(["X", "Y", "Z"]))
+    assert not is_integer(w[0])
+    assert is_integer(w[1])
+    assert is_integer(w[2])
+    assert isinstance(w[0], str)
+    assert w.dtype == "object"
+
+
+def test_where_datetimetz():
+    # GH 15701
+    timestamps = ["2016-12-31 12:00:04+00:00", "2016-12-31 12:00:04.010000+00:00"]
+    ser = Series([Timestamp(t) for t in timestamps], dtype="datetime64[ns, UTC]")
+    rs = ser.where(Series([False, True]))
+    expected = Series([pd.NaT, ser[1]], dtype="datetime64[ns, UTC]")
+    tm.assert_series_equal(rs, expected)
+
+
+def test_where_sparse():
+    # GH#17198 make sure we dont get an AttributeError for sp_index
+    ser = Series(pd.arrays.SparseArray([1, 2]))
+    result = ser.where(ser >= 2, 0)
+    expected = Series(pd.arrays.SparseArray([0, 2]))
+    tm.assert_series_equal(result, expected)
+
+
+def test_where_empty_series_and_empty_cond_having_non_bool_dtypes():
+    # https://github.com/pandas-dev/pandas/issues/34592
+    ser = Series([], dtype=float)
+    result = ser.where([])
+    tm.assert_series_equal(result, ser)
+
+
+def test_where_categorical(frame_or_series):
+    # https://github.com/pandas-dev/pandas/issues/18888
+    exp = frame_or_series(
+        pd.Categorical(["A", "A", "B", "B", np.nan], categories=["A", "B", "C"]),
+        dtype="category",
+    )
+    df = frame_or_series(["A", "A", "B", "B", "C"], dtype="category")
+    res = df.where(df != "C")
+    tm.assert_equal(exp, res)
+
+
+def test_where_datetimelike_categorical(tz_naive_fixture):
+    # GH#37682
+    tz = tz_naive_fixture
+
+    dr = date_range("2001-01-01", periods=3, tz=tz)._with_freq(None)
+    lvals = pd.DatetimeIndex([dr[0], dr[1], pd.NaT])
+    rvals = pd.Categorical([dr[0], pd.NaT, dr[2]])
+
+    mask = np.array([True, True, False])
+
+    # DatetimeIndex.where
+    res = lvals.where(mask, rvals)
+    tm.assert_index_equal(res, dr)
+
+    # DatetimeArray.where
+    res = lvals._data._where(mask, rvals)
+    tm.assert_datetime_array_equal(res, dr._data)
+
+    # Series.where
+    res = Series(lvals).where(mask, rvals)
+    tm.assert_series_equal(res, Series(dr))
+
+    # DataFrame.where
+    res = pd.DataFrame(lvals).where(mask[:, None], pd.DataFrame(rvals))
+
+    tm.assert_frame_equal(res, pd.DataFrame(dr))
diff --git a/pandas/pandas/tests/series/indexing/test_xs.py b/pandas/pandas/tests/series/indexing/test_xs.py
new file mode 100644
index 0000000000000..a67f3ec708f24
--- /dev/null
+++ b/pandas/pandas/tests/series/indexing/test_xs.py
@@ -0,0 +1,82 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_xs_datetimelike_wrapping():
+    # GH#31630 a case where we shouldn't wrap datetime64 in Timestamp
+    arr = date_range("2016-01-01", periods=3)._data._ndarray
+
+    ser = Series(arr, dtype=object)
+    for i in range(len(ser)):
+        ser.iloc[i] = arr[i]
+    assert ser.dtype == object
+    assert isinstance(ser[0], np.datetime64)
+
+    result = ser.xs(0)
+    assert isinstance(result, np.datetime64)
+
+
+class TestXSWithMultiIndex:
+    def test_xs_level_series(self, multiindex_dataframe_random_data):
+        df = multiindex_dataframe_random_data
+        ser = df["A"]
+        expected = ser[:, "two"]
+        result = df.xs("two", level=1)["A"]
+        tm.assert_series_equal(result, expected)
+
+    def test_series_getitem_multiindex_xs_by_label(self):
+        # GH#5684
+        idx = MultiIndex.from_tuples(
+            [("a", "one"), ("a", "two"), ("b", "one"), ("b", "two")]
+        )
+        ser = Series([1, 2, 3, 4], index=idx)
+        return_value = ser.index.set_names(["L1", "L2"], inplace=True)
+        assert return_value is None
+        expected = Series([1, 3], index=["a", "b"])
+        return_value = expected.index.set_names(["L1"], inplace=True)
+        assert return_value is None
+
+        result = ser.xs("one", level="L2")
+        tm.assert_series_equal(result, expected)
+
+    def test_series_getitem_multiindex_xs(self):
+        # GH#6258
+        dt = list(date_range("20130903", periods=3))
+        idx = MultiIndex.from_product([list("AB"), dt])
+        ser = Series([1, 3, 4, 1, 3, 4], index=idx)
+        expected = Series([1, 1], index=list("AB"))
+
+        result = ser.xs("20130903", level=1)
+        tm.assert_series_equal(result, expected)
+
+    def test_series_xs_droplevel_false(self):
+        # GH: 19056
+        mi = MultiIndex.from_tuples(
+            [("a", "x"), ("a", "y"), ("b", "x")], names=["level1", "level2"]
+        )
+        ser = Series([1, 1, 1], index=mi)
+        result = ser.xs("a", axis=0, drop_level=False)
+        expected = Series(
+            [1, 1],
+            index=MultiIndex.from_tuples(
+                [("a", "x"), ("a", "y")], names=["level1", "level2"]
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_xs_key_as_list(self):
+        # GH#41760
+        mi = MultiIndex.from_tuples([("a", "x")], names=["level1", "level2"])
+        ser = Series([1], index=mi)
+        with pytest.raises(TypeError, match="list keys are not supported"):
+            ser.xs(["a", "x"], axis=0, drop_level=False)
+
+        with pytest.raises(TypeError, match="list keys are not supported"):
+            ser.xs(["a"], axis=0, drop_level=False)
diff --git a/pandas/pandas/tests/series/methods/__init__.py b/pandas/pandas/tests/series/methods/__init__.py
new file mode 100644
index 0000000000000..bcb0d30f405e2
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/__init__.py
@@ -0,0 +1,7 @@
+"""
+Test files dedicated to individual (stand-alone) Series methods
+
+Ideally these files/tests should correspond 1-to-1 with tests.frame.methods
+
+These may also present opportunities for sharing/de-duplicating test code.
+"""
diff --git a/pandas/pandas/tests/series/methods/test_add_prefix_suffix.py b/pandas/pandas/tests/series/methods/test_add_prefix_suffix.py
new file mode 100644
index 0000000000000..289a56b98b7e1
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_add_prefix_suffix.py
@@ -0,0 +1,41 @@
+import pytest
+
+from pandas import Index
+import pandas._testing as tm
+
+
+def test_add_prefix_suffix(string_series):
+    with_prefix = string_series.add_prefix("foo#")
+    expected = Index([f"foo#{c}" for c in string_series.index])
+    tm.assert_index_equal(with_prefix.index, expected)
+
+    with_suffix = string_series.add_suffix("#foo")
+    expected = Index([f"{c}#foo" for c in string_series.index])
+    tm.assert_index_equal(with_suffix.index, expected)
+
+    with_pct_prefix = string_series.add_prefix("%")
+    expected = Index([f"%{c}" for c in string_series.index])
+    tm.assert_index_equal(with_pct_prefix.index, expected)
+
+    with_pct_suffix = string_series.add_suffix("%")
+    expected = Index([f"{c}%" for c in string_series.index])
+    tm.assert_index_equal(with_pct_suffix.index, expected)
+
+
+def test_add_prefix_suffix_axis(string_series):
+    # GH 47819
+    with_prefix = string_series.add_prefix("foo#", axis=0)
+    expected = Index([f"foo#{c}" for c in string_series.index])
+    tm.assert_index_equal(with_prefix.index, expected)
+
+    with_pct_suffix = string_series.add_suffix("#foo", axis=0)
+    expected = Index([f"{c}#foo" for c in string_series.index])
+    tm.assert_index_equal(with_pct_suffix.index, expected)
+
+
+def test_add_prefix_suffix_invalid_axis(string_series):
+    with pytest.raises(ValueError, match="No axis named 1 for object type Series"):
+        string_series.add_prefix("foo#", axis=1)
+
+    with pytest.raises(ValueError, match="No axis named 1 for object type Series"):
+        string_series.add_suffix("foo#", axis=1)
diff --git a/pandas/pandas/tests/series/methods/test_align.py b/pandas/pandas/tests/series/methods/test_align.py
new file mode 100644
index 0000000000000..f7e4dbe11f3d3
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_align.py
@@ -0,0 +1,199 @@
+from datetime import timezone
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "first_slice,second_slice",
+    [
+        [[2, None], [None, -5]],
+        [[None, 0], [None, -5]],
+        [[None, -5], [None, 0]],
+        [[None, 0], [None, 0]],
+    ],
+)
+@pytest.mark.parametrize("fill", [None, -1])
+def test_align(datetime_series, first_slice, second_slice, join_type, fill):
+    a = datetime_series[slice(*first_slice)]
+    b = datetime_series[slice(*second_slice)]
+
+    aa, ab = a.align(b, join=join_type, fill_value=fill)
+
+    join_index = a.index.join(b.index, how=join_type)
+    if fill is not None:
+        diff_a = aa.index.difference(join_index)
+        diff_b = ab.index.difference(join_index)
+        if len(diff_a) > 0:
+            assert (aa.reindex(diff_a) == fill).all()
+        if len(diff_b) > 0:
+            assert (ab.reindex(diff_b) == fill).all()
+
+    ea = a.reindex(join_index)
+    eb = b.reindex(join_index)
+
+    if fill is not None:
+        ea = ea.fillna(fill)
+        eb = eb.fillna(fill)
+
+    tm.assert_series_equal(aa, ea)
+    tm.assert_series_equal(ab, eb)
+    assert aa.name == "ts"
+    assert ea.name == "ts"
+    assert ab.name == "ts"
+    assert eb.name == "ts"
+
+
+def test_align_nocopy(datetime_series):
+    b = datetime_series[:5].copy()
+
+    # do copy
+    a = datetime_series.copy()
+    ra, _ = a.align(b, join="left")
+    ra[:5] = 5
+    assert not (a[:5] == 5).any()
+
+    # do not copy
+    a = datetime_series.copy()
+    ra, _ = a.align(b, join="left")
+    ra[:5] = 5
+    assert not (a[:5] == 5).any()
+
+    # do copy
+    a = datetime_series.copy()
+    b = datetime_series[:5].copy()
+    _, rb = a.align(b, join="right")
+    rb[:3] = 5
+    assert not (b[:3] == 5).any()
+
+    # do not copy
+    a = datetime_series.copy()
+    b = datetime_series[:5].copy()
+    _, rb = a.align(b, join="right")
+    rb[:2] = 5
+    assert not (b[:2] == 5).any()
+
+
+def test_align_same_index(datetime_series):
+    a, b = datetime_series.align(datetime_series)
+    assert a.index.is_(datetime_series.index)
+    assert b.index.is_(datetime_series.index)
+
+    a, b = datetime_series.align(datetime_series)
+    assert a.index is not datetime_series.index
+    assert b.index is not datetime_series.index
+    assert a.index.is_(datetime_series.index)
+    assert b.index.is_(datetime_series.index)
+
+
+def test_align_multiindex():
+    # GH 10665
+
+    midx = pd.MultiIndex.from_product(
+        [range(2), range(3), range(2)], names=("a", "b", "c")
+    )
+    idx = pd.Index(range(2), name="b")
+    s1 = Series(np.arange(12, dtype="int64"), index=midx)
+    s2 = Series(np.arange(2, dtype="int64"), index=idx)
+
+    # these must be the same results (but flipped)
+    res1l, res1r = s1.align(s2, join="left")
+    res2l, res2r = s2.align(s1, join="right")
+
+    expl = s1
+    tm.assert_series_equal(expl, res1l)
+    tm.assert_series_equal(expl, res2r)
+    expr = Series([0, 0, 1, 1, np.nan, np.nan] * 2, index=midx)
+    tm.assert_series_equal(expr, res1r)
+    tm.assert_series_equal(expr, res2l)
+
+    res1l, res1r = s1.align(s2, join="right")
+    res2l, res2r = s2.align(s1, join="left")
+
+    exp_idx = pd.MultiIndex.from_product(
+        [range(2), range(2), range(2)], names=("a", "b", "c")
+    )
+    expl = Series([0, 1, 2, 3, 6, 7, 8, 9], index=exp_idx)
+    tm.assert_series_equal(expl, res1l)
+    tm.assert_series_equal(expl, res2r)
+    expr = Series([0, 0, 1, 1] * 2, index=exp_idx)
+    tm.assert_series_equal(expr, res1r)
+    tm.assert_series_equal(expr, res2l)
+
+
+def test_align_dt64tzindex_mismatched_tzs():
+    idx1 = date_range("2001", periods=5, freq="h", tz="US/Eastern")
+    ser = Series(np.random.default_rng(2).standard_normal(len(idx1)), index=idx1)
+    ser_central = ser.tz_convert("US/Central")
+    # different timezones convert to UTC
+
+    new1, new2 = ser.align(ser_central)
+    assert new1.index.tz is timezone.utc
+    assert new2.index.tz is timezone.utc
+
+
+def test_align_periodindex(join_type):
+    rng = period_range("1/1/2000", "1/1/2010", freq="Y")
+    ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+    # TODO: assert something?
+    ts.align(ts[::2], join=join_type)
+
+
+def test_align_stringindex(any_string_dtype):
+    left = Series(range(3), index=pd.Index(["a", "b", "d"], dtype=any_string_dtype))
+    right = Series(range(3), index=pd.Index(["a", "b", "c"], dtype=any_string_dtype))
+    result_left, result_right = left.align(right)
+
+    expected_idx = pd.Index(["a", "b", "c", "d"], dtype=any_string_dtype)
+    expected_left = Series([0, 1, np.nan, 2], index=expected_idx)
+    expected_right = Series([0, 1, 2, np.nan], index=expected_idx)
+
+    tm.assert_series_equal(result_left, expected_left)
+    tm.assert_series_equal(result_right, expected_right)
+
+
+def test_align_left_fewer_levels():
+    # GH#45224
+    left = Series([2], index=pd.MultiIndex.from_tuples([(1, 3)], names=["a", "c"]))
+    right = Series(
+        [1], index=pd.MultiIndex.from_tuples([(1, 2, 3)], names=["a", "b", "c"])
+    )
+    result_left, result_right = left.align(right)
+
+    expected_right = Series(
+        [1], index=pd.MultiIndex.from_tuples([(1, 3, 2)], names=["a", "c", "b"])
+    )
+    expected_left = Series(
+        [2], index=pd.MultiIndex.from_tuples([(1, 3, 2)], names=["a", "c", "b"])
+    )
+    tm.assert_series_equal(result_left, expected_left)
+    tm.assert_series_equal(result_right, expected_right)
+
+
+def test_align_left_different_named_levels():
+    # GH#45224
+    left = Series(
+        [2], index=pd.MultiIndex.from_tuples([(1, 4, 3)], names=["a", "d", "c"])
+    )
+    right = Series(
+        [1], index=pd.MultiIndex.from_tuples([(1, 2, 3)], names=["a", "b", "c"])
+    )
+    result_left, result_right = left.align(right)
+
+    expected_left = Series(
+        [2], index=pd.MultiIndex.from_tuples([(1, 4, 3, 2)], names=["a", "d", "c", "b"])
+    )
+    expected_right = Series(
+        [1], index=pd.MultiIndex.from_tuples([(1, 4, 3, 2)], names=["a", "d", "c", "b"])
+    )
+    tm.assert_series_equal(result_left, expected_left)
+    tm.assert_series_equal(result_right, expected_right)
diff --git a/pandas/pandas/tests/series/methods/test_argsort.py b/pandas/pandas/tests/series/methods/test_argsort.py
new file mode 100644
index 0000000000000..019efe8683347
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_argsort.py
@@ -0,0 +1,76 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    Timestamp,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestSeriesArgsort:
+    def test_argsort_axis(self):
+        # GH#54257
+        ser = Series(range(3))
+
+        msg = "No axis named 2 for object type Series"
+        with pytest.raises(ValueError, match=msg):
+            ser.argsort(axis=2)
+
+    def test_argsort_numpy(self, datetime_series):
+        ser = datetime_series
+        res = np.argsort(ser).values
+        expected = np.argsort(np.array(ser))
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_argsort_numpy_missing(self):
+        data = [0.1, np.nan, 0.2, np.nan, 0.3]
+        ser = Series(data)
+        result = np.argsort(ser)
+        expected = np.argsort(np.array(data))
+
+        tm.assert_numpy_array_equal(result.values, expected)
+
+    def test_argsort(self, datetime_series):
+        argsorted = datetime_series.argsort()
+        assert issubclass(argsorted.dtype.type, np.integer)
+
+    def test_argsort_dt64(self, unit):
+        # GH#2967 (introduced bug in 0.11-dev I think)
+        ser = Series(
+            [Timestamp(f"201301{i:02d}") for i in range(1, 6)], dtype=f"M8[{unit}]"
+        )
+        assert ser.dtype == f"datetime64[{unit}]"
+        shifted = ser.shift(-1)
+        assert shifted.dtype == f"datetime64[{unit}]"
+        assert isna(shifted[4])
+
+        result = ser.argsort()
+        expected = Series(range(5), dtype=np.intp)
+        tm.assert_series_equal(result, expected)
+
+        result = shifted.argsort()
+        expected = Series(list(range(4)) + [4], dtype=np.intp)
+        tm.assert_series_equal(result, expected)
+
+    def test_argsort_stable(self):
+        ser = Series(np.random.default_rng(2).integers(0, 100, size=10000))
+        mindexer = ser.argsort(kind="mergesort")
+        qindexer = ser.argsort()
+
+        mexpected = np.argsort(ser.values, kind="mergesort")
+        qexpected = np.argsort(ser.values, kind="quicksort")
+
+        tm.assert_series_equal(mindexer.astype(np.intp), Series(mexpected))
+        tm.assert_series_equal(qindexer.astype(np.intp), Series(qexpected))
+        msg = (
+            r"ndarray Expected type <class 'numpy\.ndarray'>, "
+            r"found <class 'pandas\.Series'> instead"
+        )
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_numpy_array_equal(qindexer, mindexer)
+
+    def test_argsort_preserve_name(self, datetime_series):
+        result = datetime_series.argsort()
+        assert result.name == datetime_series.name
diff --git a/pandas/pandas/tests/series/methods/test_asof.py b/pandas/pandas/tests/series/methods/test_asof.py
new file mode 100644
index 0000000000000..2acc2921e5efc
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_asof.py
@@ -0,0 +1,205 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import IncompatibleFrequency
+
+from pandas import (
+    DatetimeIndex,
+    PeriodIndex,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+    notna,
+    offsets,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesAsof:
+    def test_asof_nanosecond_index_access(self):
+        ts = Timestamp("20130101").as_unit("ns")._value
+        dti = DatetimeIndex([ts + 50 + i for i in range(100)])
+        ser = Series(np.random.default_rng(2).standard_normal(100), index=dti)
+
+        first_value = ser.asof(ser.index[0])
+
+        # GH#46903 previously incorrectly was "day"
+        assert dti.resolution == "nanosecond"
+
+        # this used to not work bc parsing was done by dateutil that didn't
+        #  handle nanoseconds
+        assert first_value == ser["2013-01-01 00:00:00.000000050"]
+
+        expected_ts = np.datetime64("2013-01-01 00:00:00.000000050", "ns")
+        assert first_value == ser[Timestamp(expected_ts)]
+
+    def test_basic(self):
+        # array or list or dates
+        N = 50
+        rng = date_range("1/1/1990", periods=N, freq="53s")
+        ts = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+        ts.iloc[15:30] = np.nan
+        dates = date_range("1/1/1990", periods=N * 3, freq="25s")
+
+        result = ts.asof(dates)
+        assert notna(result).all()
+        lb = ts.index[14]
+        ub = ts.index[30]
+
+        result = ts.asof(list(dates))
+        assert notna(result).all()
+        lb = ts.index[14]
+        ub = ts.index[30]
+
+        mask = (result.index >= lb) & (result.index < ub)
+        rs = result[mask]
+        assert (rs == ts[lb]).all()
+
+        val = result[result.index[result.index >= ub][0]]
+        assert ts[ub] == val
+
+    def test_scalar(self):
+        N = 30
+        rng = date_range("1/1/1990", periods=N, freq="53s")
+        # Explicit cast to float avoid implicit cast when setting nan
+        ts = Series(np.arange(N), index=rng, dtype="float")
+        ts.iloc[5:10] = np.nan
+        ts.iloc[15:20] = np.nan
+
+        val1 = ts.asof(ts.index[7])
+        val2 = ts.asof(ts.index[19])
+
+        assert val1 == ts.iloc[4]
+        assert val2 == ts.iloc[14]
+
+        # accepts strings
+        val1 = ts.asof(str(ts.index[7]))
+        assert val1 == ts.iloc[4]
+
+        # in there
+        result = ts.asof(ts.index[3])
+        assert result == ts.iloc[3]
+
+        # no as of value
+        d = ts.index[0] - offsets.BDay()
+        assert np.isnan(ts.asof(d))
+
+    def test_with_nan(self):
+        # basic asof test
+        rng = date_range("1/1/2000", "1/2/2000", freq="4h")
+        s = Series(np.arange(len(rng)), index=rng)
+        r = s.resample("2h").mean()
+
+        result = r.asof(r.index)
+        expected = Series(
+            [0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6.0],
+            index=date_range("1/1/2000", "1/2/2000", freq="2h"),
+        )
+        tm.assert_series_equal(result, expected)
+
+        r.iloc[3:5] = np.nan
+        result = r.asof(r.index)
+        expected = Series(
+            [0, 0, 1, 1, 1, 1, 3, 3, 4, 4, 5, 5, 6.0],
+            index=date_range("1/1/2000", "1/2/2000", freq="2h"),
+        )
+        tm.assert_series_equal(result, expected)
+
+        r.iloc[-3:] = np.nan
+        result = r.asof(r.index)
+        expected = Series(
+            [0, 0, 1, 1, 1, 1, 3, 3, 4, 4, 4, 4, 4.0],
+            index=date_range("1/1/2000", "1/2/2000", freq="2h"),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_periodindex(self):
+        # array or list or dates
+        N = 50
+        rng = period_range("1/1/1990", periods=N, freq="h")
+        ts = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+        ts.iloc[15:30] = np.nan
+        dates = date_range("1/1/1990", periods=N * 3, freq="37min")
+
+        result = ts.asof(dates)
+        assert notna(result).all()
+        lb = ts.index[14]
+        ub = ts.index[30]
+
+        result = ts.asof(list(dates))
+        assert notna(result).all()
+        lb = ts.index[14]
+        ub = ts.index[30]
+
+        pix = PeriodIndex(result.index.values, freq="h")
+        mask = (pix >= lb) & (pix < ub)
+        rs = result[mask]
+        assert (rs == ts[lb]).all()
+
+        ts.iloc[5:10] = np.nan
+        ts.iloc[15:20] = np.nan
+
+        val1 = ts.asof(ts.index[7])
+        val2 = ts.asof(ts.index[19])
+
+        assert val1 == ts.iloc[4]
+        assert val2 == ts.iloc[14]
+
+        # accepts strings
+        val1 = ts.asof(str(ts.index[7]))
+        assert val1 == ts.iloc[4]
+
+        # in there
+        assert ts.asof(ts.index[3]) == ts.iloc[3]
+
+        # no as of value
+        d = ts.index[0].to_timestamp() - offsets.BDay()
+        assert isna(ts.asof(d))
+
+        # Mismatched freq
+        msg = "Input has different freq"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            ts.asof(rng.asfreq("D"))
+
+    def test_errors(self):
+        s = Series(
+            [1, 2, 3],
+            index=[Timestamp("20130101"), Timestamp("20130103"), Timestamp("20130102")],
+        )
+
+        # non-monotonic
+        assert not s.index.is_monotonic_increasing
+        with pytest.raises(ValueError, match="requires a sorted index"):
+            s.asof(s.index[0])
+
+        # subset with Series
+        N = 10
+        rng = date_range("1/1/1990", periods=N, freq="53s")
+        s = Series(np.random.default_rng(2).standard_normal(N), index=rng)
+        with pytest.raises(ValueError, match="not valid for Series"):
+            s.asof(s.index[0], subset="foo")
+
+    def test_all_nans(self):
+        # GH 15713
+        # series is all nans
+
+        # testing non-default indexes
+        N = 50
+        rng = date_range("1/1/1990", periods=N, freq="53s")
+
+        dates = date_range("1/1/1990", periods=N * 3, freq="25s")
+        result = Series(np.nan, index=rng).asof(dates)
+        expected = Series(np.nan, index=dates)
+        tm.assert_series_equal(result, expected)
+
+        # testing scalar input
+        date = date_range("1/1/1990", periods=N * 3, freq="25s")[0]
+        result = Series(np.nan, index=rng).asof(date)
+        assert isna(result)
+
+        # test name is propagated
+        result = Series(np.nan, index=[1, 2, 3, 4], name="test").asof([4, 5])
+        expected = Series(np.nan, index=[4, 5], name="test")
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_astype.py b/pandas/pandas/tests/series/methods/test_astype.py
new file mode 100644
index 0000000000000..4a7e204ee4161
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_astype.py
@@ -0,0 +1,683 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+from importlib import reload
+import string
+import sys
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import iNaT
+import pandas.util._test_decorators as td
+
+from pandas import (
+    NA,
+    Categorical,
+    CategoricalDtype,
+    DatetimeTZDtype,
+    Index,
+    Interval,
+    NaT,
+    Series,
+    Timedelta,
+    Timestamp,
+    cut,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+def rand_str(nchars: int) -> str:
+    """
+    Generate one random byte string.
+    """
+    RANDS_CHARS = np.array(
+        list(string.ascii_letters + string.digits), dtype=(np.str_, 1)
+    )
+    return "".join(np.random.default_rng(2).choice(RANDS_CHARS, nchars))
+
+
+class TestAstypeAPI:
+    def test_astype_unitless_dt64_raises(self):
+        # GH#47844
+        ser = Series(["1970-01-01", "1970-01-01", "1970-01-01"], dtype="datetime64[ns]")
+        df = ser.to_frame()
+
+        msg = "Casting to unit-less dtype 'datetime64' is not supported"
+        with pytest.raises(TypeError, match=msg):
+            ser.astype(np.datetime64)
+        with pytest.raises(TypeError, match=msg):
+            df.astype(np.datetime64)
+        with pytest.raises(TypeError, match=msg):
+            ser.astype("datetime64")
+        with pytest.raises(TypeError, match=msg):
+            df.astype("datetime64")
+
+    def test_arg_for_errors_in_astype(self):
+        # see GH#14878
+        ser = Series([1, 2, 3])
+
+        msg = (
+            r"Expected value of kwarg 'errors' to be one of \['raise', "
+            r"'ignore'\]\. Supplied value is 'False'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.astype(np.float64, errors=False)
+
+        ser.astype(np.int8, errors="raise")
+
+    @pytest.mark.parametrize("dtype_class", [dict, Series])
+    def test_astype_dict_like(self, dtype_class):
+        # see GH#7271
+        ser = Series(range(0, 10, 2), name="abc")
+
+        dt1 = dtype_class({"abc": str})
+        result = ser.astype(dt1)
+        expected = Series(["0", "2", "4", "6", "8"], name="abc", dtype="str")
+        tm.assert_series_equal(result, expected)
+
+        dt2 = dtype_class({"abc": "float64"})
+        result = ser.astype(dt2)
+        expected = Series([0.0, 2.0, 4.0, 6.0, 8.0], dtype="float64", name="abc")
+        tm.assert_series_equal(result, expected)
+
+        dt3 = dtype_class({"abc": str, "def": str})
+        msg = (
+            "Only the Series name can be used for the key in Series dtype "
+            r"mappings\."
+        )
+        with pytest.raises(KeyError, match=msg):
+            ser.astype(dt3)
+
+        dt4 = dtype_class({0: str})
+        with pytest.raises(KeyError, match=msg):
+            ser.astype(dt4)
+
+        # GH#16717
+        # if dtypes provided is empty, it should error
+        if dtype_class is Series:
+            dt5 = dtype_class({}, dtype=object)
+        else:
+            dt5 = dtype_class({})
+
+        with pytest.raises(KeyError, match=msg):
+            ser.astype(dt5)
+
+
+class TestAstype:
+    @pytest.mark.parametrize("tz", [None, "UTC", "US/Pacific"])
+    def test_astype_object_to_dt64_non_nano(self, tz):
+        # GH#55756, GH#54620
+        ts = Timestamp("2999-01-01")
+        dtype = "M8[us]"
+        if tz is not None:
+            dtype = f"M8[us, {tz}]"
+        vals = [ts, "2999-01-02 03:04:05.678910", 2500]
+        ser = Series(vals, dtype=object)
+        result = ser.astype(dtype)
+
+        # The 2500 is interpreted as microseconds, consistent with what
+        #  we would get if we created DatetimeIndexes from vals[:2] and vals[2:]
+        #  and concated the results.
+        pointwise = [
+            vals[0].tz_localize(tz),
+            Timestamp(vals[1], tz=tz),
+            to_datetime(vals[2], unit="us", utc=True).tz_convert(tz),
+        ]
+        exp_vals = [x.as_unit("us").asm8 for x in pointwise]
+        exp_arr = np.array(exp_vals, dtype="M8[us]")
+        expected = Series(exp_arr, dtype="M8[us]")
+        if tz is not None:
+            expected = expected.dt.tz_localize("UTC").dt.tz_convert(tz)
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_mixed_object_to_dt64tz(self):
+        # pre-2.0 this raised ValueError bc of tz mismatch
+        # xref GH#32581
+        ts = Timestamp("2016-01-04 05:06:07", tz="US/Pacific")
+        ts2 = ts.tz_convert("Asia/Tokyo")
+
+        ser = Series([ts, ts2], dtype=object)
+        res = ser.astype("datetime64[ns, Europe/Brussels]")
+        expected = Series(
+            [ts.tz_convert("Europe/Brussels"), ts2.tz_convert("Europe/Brussels")],
+            dtype="datetime64[ns, Europe/Brussels]",
+        )
+        tm.assert_series_equal(res, expected)
+
+    @pytest.mark.parametrize("dtype", np.typecodes["All"])
+    def test_astype_empty_constructor_equality(self, dtype):
+        # see GH#15524
+
+        if dtype not in (
+            "S",
+            "V",  # poor support (if any) currently
+            "M",
+            "m",  # Generic timestamps raise a ValueError. Already tested.
+        ):
+            init_empty = Series([], dtype=dtype)
+            as_type_empty = Series([]).astype(dtype)
+            tm.assert_series_equal(init_empty, as_type_empty)
+
+    @pytest.mark.parametrize("dtype", [str, np.str_])
+    @pytest.mark.parametrize(
+        "data",
+        [
+            [string.digits * 10, rand_str(63), rand_str(64), rand_str(1000)],
+            [string.digits * 10, rand_str(63), rand_str(64), np.nan, 1.0],
+        ],
+    )
+    def test_astype_str_map(self, dtype, data, using_infer_string):
+        # see GH#4405
+        series = Series(data)
+        using_string_dtype = using_infer_string and dtype is str
+        result = series.astype(dtype)
+        if using_string_dtype:
+            expected = series.map(lambda val: str(val) if val is not np.nan else np.nan)
+        else:
+            expected = series.map(str)
+            if using_infer_string:
+                expected = expected.astype(object)
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_float_to_period(self):
+        result = Series([np.nan]).astype("period[D]")
+        expected = Series([NaT], dtype="period[D]")
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_no_pandas_dtype(self):
+        # https://github.com/pandas-dev/pandas/pull/24866
+        ser = Series([1, 2], dtype="int64")
+        # Don't have NumpyEADtype in the public API, so we use `.array.dtype`,
+        # which is a NumpyEADtype.
+        result = ser.astype(ser.array.dtype)
+        tm.assert_series_equal(result, ser)
+
+    @pytest.mark.parametrize("dtype", [np.datetime64, np.timedelta64])
+    def test_astype_generic_timestamp_no_frequency(self, dtype, request):
+        # see GH#15524, GH#15987
+        data = [1]
+        ser = Series(data)
+
+        if np.dtype(dtype).name not in ["timedelta64", "datetime64"]:
+            mark = pytest.mark.xfail(reason="GH#33890 Is assigned ns unit")
+            request.applymarker(mark)
+
+        msg = (
+            rf"The '{dtype.__name__}' dtype has no unit\. "
+            rf"Please pass in '{dtype.__name__}\[ns\]' instead."
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.astype(dtype)
+
+    def test_astype_dt64_to_str(self):
+        # GH#10442 : testing astype(str) is correct for Series/DatetimeIndex
+        dti = date_range("2012-01-01", periods=3)
+        result = Series(dti).astype(str)
+        expected = Series(["2012-01-01", "2012-01-02", "2012-01-03"], dtype="str")
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_dt64tz_to_str(self):
+        # GH#10442 : testing astype(str) is correct for Series/DatetimeIndex
+        dti_tz = date_range("2012-01-01", periods=3, tz="US/Eastern")
+        result = Series(dti_tz).astype(str)
+        expected = Series(
+            [
+                "2012-01-01 00:00:00-05:00",
+                "2012-01-02 00:00:00-05:00",
+                "2012-01-03 00:00:00-05:00",
+            ],
+            dtype="str",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_datetime(self, unit):
+        ser = Series(iNaT, dtype=f"M8[{unit}]", index=range(5))
+
+        ser = ser.astype("O")
+        assert ser.dtype == np.object_
+
+        ser = Series([datetime(2001, 1, 2, 0, 0)])
+
+        ser = ser.astype("O")
+        assert ser.dtype == np.object_
+
+        ser = Series(
+            [datetime(2001, 1, 2, 0, 0) for i in range(3)], dtype=f"M8[{unit}]"
+        )
+
+        ser[1] = np.nan
+        assert ser.dtype == f"M8[{unit}]"
+
+        ser = ser.astype("O")
+        assert ser.dtype == np.object_
+
+    def test_astype_datetime64tz(self):
+        ser = Series(date_range("20130101", periods=3, tz="US/Eastern"))
+
+        # astype
+        result = ser.astype(object)
+        expected = Series(ser.astype(object), dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        result = Series(ser.values).dt.tz_localize("UTC").dt.tz_convert(ser.dt.tz)
+        tm.assert_series_equal(result, ser)
+
+        # astype - object, preserves on construction
+        result = Series(ser.astype(object))
+        expected = ser.astype(object)
+        tm.assert_series_equal(result, expected)
+
+        # astype - datetime64[ns, tz]
+        msg = "Cannot use .astype to convert from timezone-naive"
+        with pytest.raises(TypeError, match=msg):
+            # dt64->dt64tz astype deprecated
+            Series(ser.values).astype("datetime64[ns, US/Eastern]")
+
+        with pytest.raises(TypeError, match=msg):
+            # dt64->dt64tz astype deprecated
+            Series(ser.values).astype(ser.dtype)
+
+        result = ser.astype("datetime64[ns, CET]")
+        expected = Series(date_range("20130101 06:00:00", periods=3, tz="CET"))
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_str_cast_dt64(self):
+        # see GH#9757
+        ts = Series([Timestamp("2010-01-04 00:00:00")])
+        res = ts.astype(str)
+
+        expected = Series(["2010-01-04"], dtype="str")
+        tm.assert_series_equal(res, expected)
+
+        ts = Series([Timestamp("2010-01-04 00:00:00", tz="US/Eastern")])
+        res = ts.astype(str)
+
+        expected = Series(["2010-01-04 00:00:00-05:00"], dtype="str")
+        tm.assert_series_equal(res, expected)
+
+    def test_astype_str_cast_td64(self):
+        # see GH#9757
+
+        td = Series([Timedelta(1, unit="D")])
+        ser = td.astype(str)
+
+        expected = Series(["1 days"], dtype="str")
+        tm.assert_series_equal(ser, expected)
+
+    def test_dt64_series_astype_object(self):
+        dt64ser = Series(date_range("20130101", periods=3))
+        result = dt64ser.astype(object)
+        assert isinstance(result.iloc[0], datetime)
+        assert result.dtype == np.object_
+
+    def test_td64_series_astype_object(self):
+        tdser = Series(["59 Days", "59 Days", "NaT"], dtype="timedelta64[ns]")
+        result = tdser.astype(object)
+        assert isinstance(result.iloc[0], timedelta)
+        assert result.dtype == np.object_
+
+    @pytest.mark.parametrize(
+        "data, dtype",
+        [
+            (["x", "y", "z"], "string[python]"),
+            pytest.param(
+                ["x", "y", "z"],
+                "string[pyarrow]",
+                marks=td.skip_if_no("pyarrow"),
+            ),
+            (["x", "y", "z"], "category"),
+            (3 * [Timestamp("2020-01-01", tz="UTC")], None),
+            (3 * [Interval(0, 1)], None),
+        ],
+    )
+    @pytest.mark.parametrize("errors", ["raise", "ignore"])
+    def test_astype_ignores_errors_for_extension_dtypes(self, data, dtype, errors):
+        # https://github.com/pandas-dev/pandas/issues/35471
+        ser = Series(data, dtype=dtype)
+        if errors == "ignore":
+            expected = ser
+            result = ser.astype(float, errors="ignore")
+            tm.assert_series_equal(result, expected)
+        else:
+            msg = "(Cannot cast)|(could not convert)"
+            with pytest.raises((ValueError, TypeError), match=msg):
+                ser.astype(float, errors=errors)
+
+    def test_astype_from_float_to_str(self, any_float_dtype):
+        # https://github.com/pandas-dev/pandas/issues/36451
+        ser = Series([0.1], dtype=any_float_dtype)
+        result = ser.astype(str)
+        expected = Series(["0.1"], dtype="str")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "value, string_value",
+        [
+            (None, "None"),
+            (np.nan, "nan"),
+            (NA, "<NA>"),
+        ],
+    )
+    def test_astype_to_str_preserves_na(self, value, string_value, using_infer_string):
+        # https://github.com/pandas-dev/pandas/issues/36904
+        ser = Series(["a", "b", value], dtype=object)
+        result = ser.astype(str)
+        expected = Series(
+            ["a", "b", None if using_infer_string else string_value], dtype="str"
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["float32", "float64", "int64", "int32"])
+    def test_astype(self, dtype):
+        ser = Series(np.random.default_rng(2).standard_normal(5), name="foo")
+        as_typed = ser.astype(dtype)
+
+        assert as_typed.dtype == dtype
+        assert as_typed.name == ser.name
+
+    @pytest.mark.parametrize("value", [np.nan, np.inf])
+    def test_astype_cast_nan_inf_int(self, any_int_numpy_dtype, value):
+        # gh-14265: check NaN and inf raise error when converting to int
+        msg = "Cannot convert non-finite values \\(NA or inf\\) to integer"
+        ser = Series([value])
+
+        with pytest.raises(ValueError, match=msg):
+            ser.astype(any_int_numpy_dtype)
+
+    def test_astype_cast_object_int_fail(self, any_int_numpy_dtype):
+        arr = Series(["car", "house", "tree", "1"])
+        msg = r"invalid literal for int\(\) with base 10: 'car'"
+        with pytest.raises(ValueError, match=msg):
+            arr.astype(any_int_numpy_dtype)
+
+    def test_astype_float_to_uint_negatives_raise(
+        self, float_numpy_dtype, any_unsigned_int_numpy_dtype
+    ):
+        # GH#45151 We don't cast negative numbers to nonsense values
+        # TODO: same for EA float/uint dtypes, signed integers?
+        arr = np.arange(5).astype(float_numpy_dtype) - 3  # includes negatives
+        ser = Series(arr)
+
+        msg = "Cannot losslessly cast from .* to .*"
+        with pytest.raises(ValueError, match=msg):
+            ser.astype(any_unsigned_int_numpy_dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            ser.to_frame().astype(any_unsigned_int_numpy_dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            # We currently catch and re-raise in Index.astype
+            Index(ser).astype(any_unsigned_int_numpy_dtype)
+
+        with pytest.raises(ValueError, match=msg):
+            ser.array.astype(any_unsigned_int_numpy_dtype)
+
+    def test_astype_cast_object_int(self):
+        arr = Series(["1", "2", "3", "4"], dtype=object)
+        result = arr.astype(int)
+
+        tm.assert_series_equal(result, Series(np.arange(1, 5)))
+
+    def test_astype_unicode(self, using_infer_string):
+        # see GH#7758: A bit of magic is required to set
+        # default encoding to utf-8
+        digits = string.digits
+        test_series = [
+            Series([digits * 10, rand_str(63), rand_str(64), rand_str(1000)]),
+            Series(["データーサイエンス、お前はもう死んでいる"]),
+        ]
+
+        former_encoding = None
+
+        if sys.getdefaultencoding() == "utf-8":
+            # GH#45326 as of 2.0 Series.astype matches Index.astype by handling
+            #  bytes with obj.decode() instead of str(obj)
+            item = "野菜食べないとやばい"
+            ser = Series([item.encode()])
+            result = ser.astype(np.str_)
+            expected = Series([item], dtype=object)
+            tm.assert_series_equal(result, expected)
+
+        for ser in test_series:
+            res = ser.astype(np.str_)
+            expec = ser.map(str)
+            if using_infer_string:
+                expec = expec.astype(object)
+            tm.assert_series_equal(res, expec)
+
+        # Restore the former encoding
+        if former_encoding is not None and former_encoding != "utf-8":
+            reload(sys)
+            sys.setdefaultencoding(former_encoding)
+
+    def test_astype_bytes(self):
+        # GH#39474
+        result = Series(["foo", "bar", "baz"]).astype(bytes)
+        assert result.dtypes == np.dtype("S3")
+
+    def test_astype_nan_to_bool(self):
+        # GH#43018
+        ser = Series(np.nan, dtype="object")
+        result = ser.astype("bool")
+        expected = Series(True, dtype="bool")
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_ea_to_datetimetzdtype(self, any_numeric_ea_dtype):
+        # GH37553
+        ser = Series([4, 0, 9], dtype=any_numeric_ea_dtype)
+        result = ser.astype(DatetimeTZDtype(tz="US/Pacific"))
+
+        expected = Series(
+            {
+                0: Timestamp("1969-12-31 16:00:00.000000004-08:00", tz="US/Pacific"),
+                1: Timestamp("1969-12-31 16:00:00.000000000-08:00", tz="US/Pacific"),
+                2: Timestamp("1969-12-31 16:00:00.000000009-08:00", tz="US/Pacific"),
+            }
+        )
+
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_retain_attrs(self, any_numpy_dtype):
+        # GH#44414
+        ser = Series([0, 1, 2, 3])
+        ser.attrs["Location"] = "Michigan"
+
+        result = ser.astype(any_numpy_dtype).attrs
+        expected = ser.attrs
+
+        tm.assert_dict_equal(expected, result)
+
+
+class TestAstypeString:
+    @pytest.mark.parametrize(
+        "data, dtype",
+        [
+            ([True, NA], "boolean"),
+            (["A", NA], "category"),
+            (["2020-10-10", "2020-10-10"], "datetime64[ns]"),
+            (["2020-10-10", "2020-10-10", NaT], "datetime64[ns]"),
+            (
+                ["2012-01-01 00:00:00-05:00", NaT],
+                "datetime64[ns, US/Eastern]",
+            ),
+            ([1, None], "UInt16"),
+            (["1/1/2021", "2/1/2021"], "period[M]"),
+            (["1/1/2021", "2/1/2021", NaT], "period[M]"),
+            (["1 Day", "59 Days", NaT], "timedelta64[ns]"),
+            # currently no way to parse IntervalArray from a list of strings
+        ],
+    )
+    def test_astype_string_to_extension_dtype_roundtrip(
+        self, data, dtype, request, nullable_string_dtype
+    ):
+        if dtype == "boolean":
+            mark = pytest.mark.xfail(
+                reason="TODO StringArray.astype() with missing values #GH40566"
+            )
+            request.applymarker(mark)
+        # GH-40351
+        ser = Series(data, dtype=dtype)
+
+        # Note: just passing .astype(dtype) fails for dtype="category"
+        #  with bc ser.dtype.categories will be object dtype whereas
+        #  result.dtype.categories will have string dtype
+        result = ser.astype(nullable_string_dtype).astype(ser.dtype)
+        tm.assert_series_equal(result, ser)
+
+
+class TestAstypeCategorical:
+    def test_astype_categorical_to_other(self):
+        cat = Categorical([f"{i} - {i + 499}" for i in range(0, 10000, 500)])
+        ser = Series(np.random.default_rng(2).integers(0, 10000, 100)).sort_values()
+        ser = cut(ser, range(0, 10500, 500), right=False, labels=cat)
+
+        expected = ser
+        tm.assert_series_equal(ser.astype("category"), expected)
+        tm.assert_series_equal(ser.astype(CategoricalDtype()), expected)
+        msg = r"Cannot cast object|str dtype to float64"
+        with pytest.raises(ValueError, match=msg):
+            ser.astype("float64")
+
+        cat = Series(Categorical(["a", "b", "b", "a", "a", "c", "c", "c"]))
+        exp = Series(["a", "b", "b", "a", "a", "c", "c", "c"], dtype="str")
+        tm.assert_series_equal(cat.astype("str"), exp)
+        s2 = Series(Categorical(["1", "2", "3", "4"]))
+        exp2 = Series([1, 2, 3, 4]).astype("int")
+        tm.assert_series_equal(s2.astype("int"), exp2)
+
+        # object don't sort correctly, so just compare that we have the same
+        # values
+        def cmp(a, b):
+            tm.assert_almost_equal(np.sort(np.unique(a)), np.sort(np.unique(b)))
+
+        expected = Series(np.array(ser.values), name="value_group")
+        cmp(ser.astype("object"), expected)
+        cmp(ser.astype(np.object_), expected)
+
+        # array conversion
+        tm.assert_almost_equal(np.array(ser), np.array(ser.values))
+
+        tm.assert_series_equal(ser.astype("category"), ser)
+        tm.assert_series_equal(ser.astype(CategoricalDtype()), ser)
+
+        roundtrip_expected = ser.cat.set_categories(
+            ser.cat.categories.sort_values()
+        ).cat.remove_unused_categories()
+        result = ser.astype("object").astype("category")
+        tm.assert_series_equal(result, roundtrip_expected)
+        result = ser.astype("object").astype(CategoricalDtype())
+        tm.assert_series_equal(result, roundtrip_expected)
+
+    def test_astype_categorical_invalid_conversions(self):
+        # invalid conversion (these are NOT a dtype)
+        cat = Categorical([f"{i} - {i + 499}" for i in range(0, 10000, 500)])
+        ser = Series(np.random.default_rng(2).integers(0, 10000, 100)).sort_values()
+        ser = cut(ser, range(0, 10500, 500), right=False, labels=cat)
+
+        msg = (
+            "dtype '<class 'pandas.core.arrays.categorical.Categorical'>' "
+            "not understood"
+        )
+        with pytest.raises(TypeError, match=msg):
+            ser.astype(Categorical)
+        with pytest.raises(TypeError, match=msg):
+            ser.astype("object").astype(Categorical)
+
+    def test_astype_categoricaldtype(self):
+        ser = Series(["a", "b", "a"])
+        result = ser.astype(CategoricalDtype(["a", "b"], ordered=True))
+        expected = Series(Categorical(["a", "b", "a"], ordered=True))
+        tm.assert_series_equal(result, expected)
+
+        result = ser.astype(CategoricalDtype(["a", "b"], ordered=False))
+        expected = Series(Categorical(["a", "b", "a"], ordered=False))
+        tm.assert_series_equal(result, expected)
+
+        result = ser.astype(CategoricalDtype(["a", "b", "c"], ordered=False))
+        expected = Series(
+            Categorical(["a", "b", "a"], categories=["a", "b", "c"], ordered=False)
+        )
+        tm.assert_series_equal(result, expected)
+        tm.assert_index_equal(result.cat.categories, Index(["a", "b", "c"]))
+
+    @pytest.mark.parametrize("name", [None, "foo"])
+    @pytest.mark.parametrize("dtype_ordered", [True, False])
+    @pytest.mark.parametrize("series_ordered", [True, False])
+    def test_astype_categorical_to_categorical(
+        self, name, dtype_ordered, series_ordered
+    ):
+        # GH#10696, GH#18593
+        s_data = list("abcaacbab")
+        s_dtype = CategoricalDtype(list("bac"), ordered=series_ordered)
+        ser = Series(s_data, dtype=s_dtype, name=name)
+
+        # unspecified categories
+        dtype = CategoricalDtype(ordered=dtype_ordered)
+        result = ser.astype(dtype)
+        exp_dtype = CategoricalDtype(s_dtype.categories, dtype_ordered)
+        expected = Series(s_data, name=name, dtype=exp_dtype)
+        tm.assert_series_equal(result, expected)
+
+        # different categories
+        dtype = CategoricalDtype(list("adc"), dtype_ordered)
+        result = ser.astype(dtype)
+        expected = Series(s_data, name=name, dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        if dtype_ordered is False:
+            # not specifying ordered, so only test once
+            expected = ser
+            result = ser.astype("category")
+            tm.assert_series_equal(result, expected)
+
+    def test_astype_bool_missing_to_categorical(self):
+        # GH-19182
+        ser = Series([True, False, np.nan])
+        assert ser.dtypes == np.object_
+
+        result = ser.astype(CategoricalDtype(categories=[True, False]))
+        expected = Series(Categorical([True, False, np.nan], categories=[True, False]))
+        tm.assert_series_equal(result, expected)
+
+    def test_astype_categories_raises(self):
+        # deprecated GH#17636, removed in GH#27141
+        ser = Series(["a", "b", "a"])
+        with pytest.raises(TypeError, match="got an unexpected"):
+            ser.astype("category", categories=["a", "b"], ordered=True)
+
+    @pytest.mark.parametrize("items", [["a", "b", "c", "a"], [1, 2, 3, 1]])
+    def test_astype_from_categorical(self, items):
+        ser = Series(items)
+        exp = Series(Categorical(items))
+        res = ser.astype("category")
+        tm.assert_series_equal(res, exp)
+
+    def test_astype_from_categorical_with_keywords(self):
+        # with keywords
+        lst = ["a", "b", "c", "a"]
+        ser = Series(lst)
+        exp = Series(Categorical(lst, ordered=True))
+        res = ser.astype(CategoricalDtype(None, ordered=True))
+        tm.assert_series_equal(res, exp)
+
+        exp = Series(Categorical(lst, categories=list("abcdef"), ordered=True))
+        res = ser.astype(CategoricalDtype(list("abcdef"), ordered=True))
+        tm.assert_series_equal(res, exp)
+
+    def test_astype_timedelta64_with_np_nan(self):
+        # GH45798
+        result = Series([Timedelta(1), np.nan], dtype="timedelta64[ns]")
+        expected = Series([Timedelta(1), NaT], dtype="timedelta64[ns]")
+        tm.assert_series_equal(result, expected)
+
+    @td.skip_if_no("pyarrow")
+    def test_astype_int_na_string(self):
+        # GH#57418
+        ser = Series([12, NA], dtype="Int64[pyarrow]")
+        result = ser.astype("string[pyarrow]")
+        expected = Series(["12", NA], dtype="string[pyarrow]")
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_autocorr.py b/pandas/pandas/tests/series/methods/test_autocorr.py
new file mode 100644
index 0000000000000..c1d768cf02f37
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_autocorr.py
@@ -0,0 +1,30 @@
+import numpy as np
+
+
+class TestAutoCorr:
+    def test_autocorr(self, datetime_series):
+        # Just run the function
+        corr1 = datetime_series.autocorr()
+
+        # Now run it with the lag parameter
+        corr2 = datetime_series.autocorr(lag=1)
+
+        # corr() with lag needs Series of at least length 2
+        if len(datetime_series) <= 2:
+            assert np.isnan(corr1)
+            assert np.isnan(corr2)
+        else:
+            assert corr1 == corr2
+
+        # Choose a random lag between 1 and length of Series - 2
+        # and compare the result with the Series corr() function
+        n = 1 + np.random.default_rng(2).integers(max(1, len(datetime_series) - 2))
+        corr1 = datetime_series.corr(datetime_series.shift(n))
+        corr2 = datetime_series.autocorr(lag=n)
+
+        # corr() with lag needs Series of at least length 2
+        if len(datetime_series) <= 2:
+            assert np.isnan(corr1)
+            assert np.isnan(corr2)
+        else:
+            assert corr1 == corr2
diff --git a/pandas/pandas/tests/series/methods/test_between.py b/pandas/pandas/tests/series/methods/test_between.py
new file mode 100644
index 0000000000000..f035767e2ce0e
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_between.py
@@ -0,0 +1,74 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    bdate_range,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestBetween:
+    def test_between(self):
+        series = Series(date_range("1/1/2000", periods=10))
+        left, right = series[[2, 7]]
+
+        result = series.between(left, right)
+        expected = (series >= left) & (series <= right)
+        tm.assert_series_equal(result, expected)
+
+    def test_between_datetime_object_dtype(self):
+        ser = Series(bdate_range("1/1/2000", periods=20), dtype=object)
+        ser[::2] = np.nan
+
+        result = ser[ser.between(ser[3], ser[17])]
+        expected = ser[3:18].dropna()
+        tm.assert_series_equal(result, expected)
+
+        result = ser[ser.between(ser[3], ser[17], inclusive="neither")]
+        expected = ser[5:16].dropna()
+        tm.assert_series_equal(result, expected)
+
+    def test_between_period_values(self):
+        ser = Series(period_range("2000-01-01", periods=10, freq="D"))
+        left, right = ser[[2, 7]]
+        result = ser.between(left, right)
+        expected = (ser >= left) & (ser <= right)
+        tm.assert_series_equal(result, expected)
+
+    def test_between_inclusive_string(self):
+        # GH 40628
+        series = Series(date_range("1/1/2000", periods=10))
+        left, right = series[[2, 7]]
+
+        result = series.between(left, right, inclusive="both")
+        expected = (series >= left) & (series <= right)
+        tm.assert_series_equal(result, expected)
+
+        result = series.between(left, right, inclusive="left")
+        expected = (series >= left) & (series < right)
+        tm.assert_series_equal(result, expected)
+
+        result = series.between(left, right, inclusive="right")
+        expected = (series > left) & (series <= right)
+        tm.assert_series_equal(result, expected)
+
+        result = series.between(left, right, inclusive="neither")
+        expected = (series > left) & (series < right)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("inclusive", ["yes", True, False])
+    def test_between_error_args(self, inclusive):
+        # GH 40628
+        series = Series(date_range("1/1/2000", periods=10))
+        left, right = series[[2, 7]]
+
+        value_error_msg = (
+            "Inclusive has to be either string of 'both','left', 'right', or 'neither'."
+        )
+
+        series = Series(date_range("1/1/2000", periods=10))
+        with pytest.raises(ValueError, match=value_error_msg):
+            series.between(left, right, inclusive=inclusive)
diff --git a/pandas/pandas/tests/series/methods/test_case_when.py b/pandas/pandas/tests/series/methods/test_case_when.py
new file mode 100644
index 0000000000000..7cb60a11644a3
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_case_when.py
@@ -0,0 +1,148 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    array as pd_array,
+    date_range,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture
+def df():
+    """
+    base dataframe for testing
+    """
+    return DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+
+
+def test_case_when_caselist_is_not_a_list(df):
+    """
+    Raise ValueError if caselist is not a list.
+    """
+    msg = "The caselist argument should be a list; "
+    msg += "instead got.+"
+    with pytest.raises(TypeError, match=msg):  # GH39154
+        df["a"].case_when(caselist=())
+
+
+def test_case_when_no_caselist(df):
+    """
+    Raise ValueError if no caselist is provided.
+    """
+    msg = "provide at least one boolean condition, "
+    msg += "with a corresponding replacement."
+    with pytest.raises(ValueError, match=msg):  # GH39154
+        df["a"].case_when([])
+
+
+def test_case_when_odd_caselist(df):
+    """
+    Raise ValueError if no of caselist is odd.
+    """
+    msg = "Argument 0 must have length 2; "
+    msg += "a condition and replacement; instead got length 3."
+
+    with pytest.raises(ValueError, match=msg):
+        df["a"].case_when([(df["a"].eq(1), 1, df.a.gt(1))])
+
+
+def test_case_when_raise_error_from_mask(df):
+    """
+    Raise Error from within Series.mask
+    """
+    msg = "Failed to apply condition0 and replacement0."
+    with pytest.raises(ValueError, match=msg):
+        df["a"].case_when([(df["a"].eq(1), [1, 2])])
+
+
+def test_case_when_single_condition(df):
+    """
+    Test output on a single condition.
+    """
+    result = Series([np.nan, np.nan, np.nan]).case_when([(df.a.eq(1), 1)])
+    expected = Series([1, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_multiple_conditions(df):
+    """
+    Test output when booleans are derived from a computation
+    """
+    result = Series([np.nan, np.nan, np.nan]).case_when(
+        [(df.a.eq(1), 1), (Series([False, True, False]), 2)]
+    )
+    expected = Series([1, 2, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_multiple_conditions_replacement_list(df):
+    """
+    Test output when replacement is a list
+    """
+    result = Series([np.nan, np.nan, np.nan]).case_when(
+        [([True, False, False], 1), (df["a"].gt(1) & df["b"].eq(5), [1, 2, 3])]
+    )
+    expected = Series([1, 2, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_multiple_conditions_replacement_extension_dtype(df):
+    """
+    Test output when replacement has an extension dtype
+    """
+    result = Series([np.nan, np.nan, np.nan]).case_when(
+        [
+            ([True, False, False], 1),
+            (df["a"].gt(1) & df["b"].eq(5), pd_array([1, 2, 3], dtype="Int64")),
+        ],
+    )
+    expected = Series([1, 2, np.nan], dtype="Float64")
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_multiple_conditions_replacement_series(df):
+    """
+    Test output when replacement is a Series
+    """
+    result = Series([np.nan, np.nan, np.nan]).case_when(
+        [
+            (np.array([True, False, False]), 1),
+            (df["a"].gt(1) & df["b"].eq(5), Series([1, 2, 3])),
+        ],
+    )
+    expected = Series([1, 2, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_non_range_index():
+    """
+    Test output if index is not RangeIndex
+    """
+    rng = np.random.default_rng(seed=123)
+    dates = date_range("1/1/2000", periods=8)
+    df = DataFrame(
+        rng.standard_normal(size=(8, 4)), index=dates, columns=["A", "B", "C", "D"]
+    )
+    result = Series(5, index=df.index, name="A").case_when([(df.A.gt(0), df.B)])
+    expected = df.A.mask(df.A.gt(0), df.B).where(df.A.gt(0), 5)
+    tm.assert_series_equal(result, expected)
+
+
+def test_case_when_callable():
+    """
+    Test output on a callable
+    """
+    # https://numpy.org/doc/stable/reference/generated/numpy.piecewise.html
+    x = np.linspace(-2.5, 2.5, 6)
+    ser = Series(x)
+    result = ser.case_when(
+        caselist=[
+            (lambda df: df < 0, lambda df: -df),
+            (lambda df: df >= 0, lambda df: df),
+        ]
+    )
+    expected = np.piecewise(x, [x < 0, x >= 0], [lambda x: -x, lambda x: x])
+    tm.assert_series_equal(result, Series(expected))
diff --git a/pandas/pandas/tests/series/methods/test_clip.py b/pandas/pandas/tests/series/methods/test_clip.py
new file mode 100644
index 0000000000000..8ed422fc118dc
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_clip.py
@@ -0,0 +1,162 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas.errors import OutOfBoundsDatetime
+
+import pandas as pd
+from pandas import (
+    Series,
+    Timestamp,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+
+class TestSeriesClip:
+    def test_clip(self, datetime_series):
+        val = datetime_series.median()
+
+        assert datetime_series.clip(lower=val).min() == val
+        assert datetime_series.clip(upper=val).max() == val
+
+        result = datetime_series.clip(-0.5, 0.5)
+        expected = np.clip(datetime_series, -0.5, 0.5)
+        tm.assert_series_equal(result, expected)
+        assert isinstance(expected, Series)
+
+    def test_clip_types_and_nulls(self):
+        sers = [
+            Series([np.nan, 1.0, 2.0, 3.0]),
+            Series([None, "a", "b", "c"]),
+            Series(pd.to_datetime([np.nan, 1, 2, 3], unit="D")),
+        ]
+
+        for s in sers:
+            thresh = s[2]
+            lower = s.clip(lower=thresh)
+            upper = s.clip(upper=thresh)
+            assert lower[notna(lower)].min() == thresh
+            assert upper[notna(upper)].max() == thresh
+            assert list(isna(s)) == list(isna(lower))
+            assert list(isna(s)) == list(isna(upper))
+
+    def test_series_clipping_with_na_values(self, any_numeric_ea_dtype, nulls_fixture):
+        # Ensure that clipping method can handle NA values with out failing
+        # GH#40581
+
+        if nulls_fixture is pd.NaT:
+            # constructor will raise, see
+            #  test_constructor_mismatched_null_nullable_dtype
+            pytest.skip("See test_constructor_mismatched_null_nullable_dtype")
+
+        ser = Series([nulls_fixture, 1.0, 3.0], dtype=any_numeric_ea_dtype)
+        s_clipped_upper = ser.clip(upper=2.0)
+        s_clipped_lower = ser.clip(lower=2.0)
+
+        expected_upper = Series([nulls_fixture, 1.0, 2.0], dtype=any_numeric_ea_dtype)
+        expected_lower = Series([nulls_fixture, 2.0, 3.0], dtype=any_numeric_ea_dtype)
+
+        tm.assert_series_equal(s_clipped_upper, expected_upper)
+        tm.assert_series_equal(s_clipped_lower, expected_lower)
+
+    def test_clip_with_na_args(self):
+        """Should process np.nan argument as None"""
+        # GH#17276
+        s = Series([1, 2, 3])
+
+        tm.assert_series_equal(s.clip(np.nan), Series([1, 2, 3]))
+        tm.assert_series_equal(s.clip(upper=np.nan, lower=np.nan), Series([1, 2, 3]))
+
+        # GH#19992
+
+        res = s.clip(lower=[0, 4, np.nan])
+        tm.assert_series_equal(res, Series([1, 4, 3.0]))
+        res = s.clip(upper=[1, np.nan, 1])
+        tm.assert_series_equal(res, Series([1, 2, 1.0]))
+
+        # GH#40420
+        s = Series([1, 2, 3])
+        result = s.clip(0, [np.nan, np.nan, np.nan])
+        tm.assert_series_equal(s, result)
+
+    def test_clip_against_series(self):
+        # GH#6966
+
+        s = Series([1.0, 1.0, 4.0])
+
+        lower = Series([1.0, 2.0, 3.0])
+        upper = Series([1.5, 2.5, 3.5])
+
+        tm.assert_series_equal(s.clip(lower, upper), Series([1.0, 2.0, 3.5]))
+        tm.assert_series_equal(s.clip(1.5, upper), Series([1.5, 1.5, 3.5]))
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize("upper", [[1, 2, 3], np.asarray([1, 2, 3])])
+    def test_clip_against_list_like(self, inplace, upper):
+        # GH#15390
+        original = Series([5, 6, 7])
+        result = original.clip(upper=upper, inplace=inplace)
+        expected = Series([1, 2, 3])
+
+        if inplace:
+            result = original
+        tm.assert_series_equal(result, expected, check_exact=True)
+
+    def test_clip_with_datetimes(self):
+        # GH#11838
+        # naive and tz-aware datetimes
+
+        t = Timestamp("2015-12-01 09:30:30")
+        s = Series([Timestamp("2015-12-01 09:30:00"), Timestamp("2015-12-01 09:31:00")])
+        result = s.clip(upper=t)
+        expected = Series(
+            [Timestamp("2015-12-01 09:30:00"), Timestamp("2015-12-01 09:30:30")]
+        )
+        tm.assert_series_equal(result, expected)
+
+        t = Timestamp("2015-12-01 09:30:30", tz="US/Eastern")
+        s = Series(
+            [
+                Timestamp("2015-12-01 09:30:00", tz="US/Eastern"),
+                Timestamp("2015-12-01 09:31:00", tz="US/Eastern"),
+            ]
+        )
+        result = s.clip(upper=t)
+        expected = Series(
+            [
+                Timestamp("2015-12-01 09:30:00", tz="US/Eastern"),
+                Timestamp("2015-12-01 09:30:30", tz="US/Eastern"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_clip_with_timestamps_and_oob_datetimes_object(self):
+        # GH-42794
+        ser = Series([datetime(1, 1, 1), datetime(9999, 9, 9)], dtype=object)
+
+        result = ser.clip(lower=Timestamp.min, upper=Timestamp.max)
+        expected = Series([Timestamp.min, Timestamp.max], dtype=object)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_clip_with_timestamps_and_oob_datetimes_non_nano(self):
+        # GH#56410
+        dtype = "M8[us]"
+        ser = Series([datetime(1, 1, 1), datetime(9999, 9, 9)], dtype=dtype)
+
+        msg = (
+            r"Incompatible \(high-resolution\) value for dtype='datetime64\[us\]'. "
+            "Explicitly cast before operating"
+        )
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            ser.clip(lower=Timestamp.min, upper=Timestamp.max)
+
+        lower = Timestamp.min.as_unit("us")
+        upper = Timestamp.max.as_unit("us")
+        result = ser.clip(lower=lower, upper=upper)
+        expected = Series([lower, upper], dtype=dtype)
+
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_combine.py b/pandas/pandas/tests/series/methods/test_combine.py
new file mode 100644
index 0000000000000..75d47e3daa103
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_combine.py
@@ -0,0 +1,17 @@
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestCombine:
+    def test_combine_scalar(self):
+        # GH#21248
+        # Note - combine() with another Series is tested elsewhere because
+        # it is used when testing operators
+        ser = Series([i * 10 for i in range(5)])
+        result = ser.combine(3, lambda x, y: x + y)
+        expected = Series([i * 10 + 3 for i in range(5)])
+        tm.assert_series_equal(result, expected)
+
+        result = ser.combine(22, lambda x, y: min(x, y))
+        expected = Series([min(i * 10, 22) for i in range(5)])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_combine_first.py b/pandas/pandas/tests/series/methods/test_combine_first.py
new file mode 100644
index 0000000000000..51d6704e1905b
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_combine_first.py
@@ -0,0 +1,146 @@
+from datetime import datetime
+
+import numpy as np
+
+import pandas as pd
+from pandas import (
+    Period,
+    Series,
+    date_range,
+    period_range,
+    to_datetime,
+)
+import pandas._testing as tm
+
+
+class TestCombineFirst:
+    def test_combine_first_period_datetime(self):
+        # GH#3367
+        didx = date_range(start="1950-01-31", end="1950-07-31", freq="ME")
+        pidx = period_range(start=Period("1950-1"), end=Period("1950-7"), freq="M")
+        # check to be consistent with DatetimeIndex
+        for idx in [didx, pidx]:
+            a = Series([1, np.nan, np.nan, 4, 5, np.nan, 7], index=idx)
+            b = Series([9, 9, 9, 9, 9, 9, 9], index=idx)
+
+            result = a.combine_first(b)
+            expected = Series([1, 9, 9, 4, 5, 9, 7], index=idx, dtype=np.float64)
+            tm.assert_series_equal(result, expected)
+
+    def test_combine_first_name(self, datetime_series):
+        result = datetime_series.combine_first(datetime_series[:5])
+        assert result.name == datetime_series.name
+
+    def test_combine_first(self, using_infer_string):
+        values = np.arange(20, dtype=np.float64)
+        series = Series(values, index=np.arange(20, dtype=np.int64))
+
+        series_copy = series * 2
+        series_copy[::2] = np.nan
+
+        # nothing used from the input
+        combined = series.combine_first(series_copy)
+
+        tm.assert_series_equal(combined, series)
+
+        # Holes filled from input
+        combined = series_copy.combine_first(series)
+        assert np.isfinite(combined).all()
+
+        tm.assert_series_equal(combined[::2], series[::2])
+        tm.assert_series_equal(combined[1::2], series_copy[1::2])
+
+        # mixed types
+        index = pd.Index([str(i) for i in range(20)])
+        floats = Series(np.random.default_rng(2).standard_normal(20), index=index)
+        strings = Series([str(i) for i in range(10)], index=index[::2], dtype=object)
+
+        combined = strings.combine_first(floats)
+
+        tm.assert_series_equal(strings, combined.loc[index[::2]])
+        tm.assert_series_equal(floats[1::2].astype(object), combined.loc[index[1::2]])
+
+        # corner case
+        ser = Series([1.0, 2, 3], index=[0, 1, 2])
+        empty = Series([], index=[], dtype=object)
+        result = ser.combine_first(empty)
+        if not using_infer_string:
+            ser.index = ser.index.astype("O")
+        tm.assert_series_equal(result, ser.astype(object))
+
+    def test_combine_first_dt64(self, unit):
+        s0 = to_datetime(Series(["2010", np.nan])).dt.as_unit(unit)
+        s1 = to_datetime(Series([np.nan, "2011"])).dt.as_unit(unit)
+        rs = s0.combine_first(s1)
+        xp = to_datetime(Series(["2010", "2011"])).dt.as_unit(unit)
+        tm.assert_series_equal(rs, xp)
+
+        s0 = to_datetime(Series(["2010", np.nan])).dt.as_unit(unit)
+        s1 = Series([np.nan, "2011"])
+        rs = s0.combine_first(s1)
+
+        xp = Series([datetime(2010, 1, 1), "2011"], dtype=f"datetime64[{unit}]")
+
+        tm.assert_series_equal(rs, xp)
+
+    def test_combine_first_dt_tz_values(self, tz_naive_fixture):
+        ser1 = Series(
+            pd.DatetimeIndex(["20150101", "20150102", "20150103"], tz=tz_naive_fixture),
+            name="ser1",
+        )
+        ser2 = Series(
+            pd.DatetimeIndex(["20160514", "20160515", "20160516"], tz=tz_naive_fixture),
+            index=[2, 3, 4],
+            name="ser2",
+        )
+        result = ser1.combine_first(ser2)
+        exp_vals = pd.DatetimeIndex(
+            ["20150101", "20150102", "20150103", "20160515", "20160516"],
+            tz=tz_naive_fixture,
+        )
+        exp = Series(exp_vals, name="ser1")
+        tm.assert_series_equal(exp, result)
+
+    def test_combine_first_timezone_series_with_empty_series(self):
+        # GH 41800
+        time_index = date_range(
+            datetime(2021, 1, 1, 1),
+            datetime(2021, 1, 1, 10),
+            freq="h",
+            tz="Europe/Rome",
+        )
+        s1 = Series(range(10), index=time_index)
+        s2 = Series(index=time_index)
+        result = s1.combine_first(s2)
+        tm.assert_series_equal(result, s1.astype(np.float64))
+
+    def test_combine_first_preserves_dtype(self):
+        # GH51764
+        s1 = Series([1666880195890293744, 1666880195890293837])
+        s2 = Series([1, 2, 3])
+        result = s1.combine_first(s2)
+        expected = Series([1666880195890293744, 1666880195890293837, 3])
+        tm.assert_series_equal(result, expected)
+
+    def test_combine_mixed_timezone(self):
+        # GH 26283
+        uniform_tz = Series({pd.Timestamp("2019-05-01", tz="UTC"): 1.0})
+        multi_tz = Series(
+            {
+                pd.Timestamp("2019-05-01 01:00:00+0100", tz="Europe/London"): 2.0,
+                pd.Timestamp("2019-05-02", tz="UTC"): 3.0,
+            }
+        )
+
+        result = uniform_tz.combine_first(multi_tz)
+        expected = Series(
+            [1.0, 3.0],
+            index=pd.Index(
+                [
+                    pd.Timestamp("2019-05-01 00:00:00+00:00", tz="UTC"),
+                    pd.Timestamp("2019-05-02 00:00:00+00:00", tz="UTC"),
+                ],
+                dtype="object",
+            ),
+        )
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_compare.py b/pandas/pandas/tests/series/methods/test_compare.py
new file mode 100644
index 0000000000000..2a57d5139b62c
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_compare.py
@@ -0,0 +1,140 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_compare_axis(axis):
+    # GH#30429
+    s1 = pd.Series(["a", "b", "c"])
+    s2 = pd.Series(["x", "b", "z"])
+
+    result = s1.compare(s2, align_axis=axis)
+
+    if axis in (1, "columns"):
+        indices = pd.Index([0, 2])
+        columns = pd.Index(["self", "other"])
+        expected = pd.DataFrame(
+            [["a", "x"], ["c", "z"]], index=indices, columns=columns
+        )
+        tm.assert_frame_equal(result, expected)
+    else:
+        indices = pd.MultiIndex.from_product([[0, 2], ["self", "other"]])
+        expected = pd.Series(["a", "x", "c", "z"], index=indices)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keep_shape, keep_equal",
+    [
+        (True, False),
+        (False, True),
+        (True, True),
+        # False, False case is already covered in test_compare_axis
+    ],
+)
+def test_compare_various_formats(keep_shape, keep_equal):
+    s1 = pd.Series(["a", "b", "c"])
+    s2 = pd.Series(["x", "b", "z"])
+
+    result = s1.compare(s2, keep_shape=keep_shape, keep_equal=keep_equal)
+
+    if keep_shape:
+        indices = pd.Index([0, 1, 2])
+        columns = pd.Index(["self", "other"])
+        if keep_equal:
+            expected = pd.DataFrame(
+                [["a", "x"], ["b", "b"], ["c", "z"]], index=indices, columns=columns
+            )
+        else:
+            expected = pd.DataFrame(
+                [["a", "x"], [np.nan, np.nan], ["c", "z"]],
+                index=indices,
+                columns=columns,
+            )
+    else:
+        indices = pd.Index([0, 2])
+        columns = pd.Index(["self", "other"])
+        expected = pd.DataFrame(
+            [["a", "x"], ["c", "z"]], index=indices, columns=columns
+        )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compare_with_equal_nulls():
+    # We want to make sure two NaNs are considered the same
+    # and dropped where applicable
+    s1 = pd.Series(["a", "b", np.nan])
+    s2 = pd.Series(["x", "b", np.nan])
+
+    result = s1.compare(s2)
+    expected = pd.DataFrame([["a", "x"]], columns=["self", "other"])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_compare_with_non_equal_nulls():
+    # We want to make sure the relevant NaNs do not get dropped
+    s1 = pd.Series(["a", "b", "c"])
+    s2 = pd.Series(["x", "b", np.nan])
+
+    result = s1.compare(s2, align_axis=0)
+
+    indices = pd.MultiIndex.from_product([[0, 2], ["self", "other"]])
+    expected = pd.Series(["a", "x", "c", np.nan], index=indices)
+    tm.assert_series_equal(result, expected)
+
+
+def test_compare_multi_index():
+    index = pd.MultiIndex.from_arrays([[0, 0, 1], [0, 1, 2]])
+    s1 = pd.Series(["a", "b", "c"], index=index)
+    s2 = pd.Series(["x", "b", "z"], index=index)
+
+    result = s1.compare(s2, align_axis=0)
+
+    indices = pd.MultiIndex.from_arrays(
+        [[0, 0, 1, 1], [0, 0, 2, 2], ["self", "other", "self", "other"]]
+    )
+    expected = pd.Series(["a", "x", "c", "z"], index=indices)
+    tm.assert_series_equal(result, expected)
+
+
+def test_compare_different_indices():
+    msg = "Can only compare identically-labeled Series objects"
+    ser1 = pd.Series([1, 2, 3], index=["a", "b", "c"])
+    ser2 = pd.Series([1, 2, 3], index=["a", "b", "d"])
+    with pytest.raises(ValueError, match=msg):
+        ser1.compare(ser2)
+
+
+def test_compare_different_lengths():
+    msg = "Can only compare identically-labeled Series objects"
+    ser1 = pd.Series([1, 2, 3])
+    ser2 = pd.Series([1, 2, 3, 4])
+    with pytest.raises(ValueError, match=msg):
+        ser1.compare(ser2)
+
+
+def test_compare_datetime64_and_string():
+    # Issue https://github.com/pandas-dev/pandas/issues/45506
+    # Catch OverflowError when comparing datetime64 and string
+    data = [
+        {"a": "2015-07-01", "b": "08335394550"},
+        {"a": "2015-07-02", "b": "+49 (0) 0345 300033"},
+        {"a": "2015-07-03", "b": "+49(0)2598 04457"},
+        {"a": "2015-07-04", "b": "0741470003"},
+        {"a": "2015-07-05", "b": "04181 83668"},
+    ]
+    dtypes = {"a": "datetime64[ns]", "b": "string"}
+    df = pd.DataFrame(data=data).astype(dtypes)
+
+    result_eq1 = df["a"].eq(df["b"])
+    result_eq2 = df["a"] == df["b"]
+    result_neq = df["a"] != df["b"]
+
+    expected_eq = pd.Series([False] * 5)  # For .eq and ==
+    expected_neq = pd.Series([True] * 5)  # For !=
+
+    tm.assert_series_equal(result_eq1, expected_eq)
+    tm.assert_series_equal(result_eq2, expected_eq)
+    tm.assert_series_equal(result_neq, expected_neq)
diff --git a/pandas/pandas/tests/series/methods/test_convert_dtypes.py b/pandas/pandas/tests/series/methods/test_convert_dtypes.py
new file mode 100644
index 0000000000000..d373386108ff6
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_convert_dtypes.py
@@ -0,0 +1,310 @@
+from itertools import product
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+import pandas as pd
+import pandas._testing as tm
+
+
+class TestSeriesConvertDtypes:
+    @pytest.mark.parametrize(
+        "data, maindtype, expected_default, expected_other",
+        [
+            (
+                # data
+                [1, 2, 3],
+                # original dtype
+                np.dtype("int32"),
+                # default expected dtype
+                "Int32",
+                # exceptions on expected dtype
+                {("convert_integer", False): np.dtype("int32")},
+            ),
+            (
+                [1, 2, 3],
+                np.dtype("int64"),
+                "Int64",
+                {("convert_integer", False): np.dtype("int64")},
+            ),
+            (
+                ["x", "y", "z"],
+                np.dtype("O"),
+                pd.StringDtype(),
+                {("convert_string", False): np.dtype("O")},
+            ),
+            (
+                [True, False, np.nan],
+                np.dtype("O"),
+                pd.BooleanDtype(),
+                {("convert_boolean", False): np.dtype("O")},
+            ),
+            (
+                ["h", "i", np.nan],
+                np.dtype("O"),
+                pd.StringDtype(),
+                {("convert_string", False): np.dtype("O")},
+            ),
+            (  # GH32117
+                ["h", "i", 1],
+                np.dtype("O"),
+                np.dtype("O"),
+                {},
+            ),
+            (
+                [10, np.nan, 20],
+                np.dtype("float"),
+                "Int64",
+                {
+                    ("convert_integer", False, "convert_floating", True): "Float64",
+                    ("convert_integer", False, "convert_floating", False): np.dtype(
+                        "float"
+                    ),
+                },
+            ),
+            (
+                [np.nan, 100.5, 200],
+                np.dtype("float"),
+                "Float64",
+                {("convert_floating", False): np.dtype("float")},
+            ),
+            (
+                [3, 4, 5],
+                "Int8",
+                "Int8",
+                {},
+            ),
+            (
+                [[1, 2], [3, 4], [5]],
+                None,
+                np.dtype("O"),
+                {},
+            ),
+            (
+                [4, 5, 6],
+                np.dtype("uint32"),
+                "UInt32",
+                {("convert_integer", False): np.dtype("uint32")},
+            ),
+            (
+                [-10, 12, 13],
+                np.dtype("i1"),
+                "Int8",
+                {("convert_integer", False): np.dtype("i1")},
+            ),
+            (
+                [1.2, 1.3],
+                np.dtype("float32"),
+                "Float32",
+                {("convert_floating", False): np.dtype("float32")},
+            ),
+            (
+                [1, 2.0],
+                object,
+                "Int64",
+                {
+                    ("convert_integer", False): "Float64",
+                    ("convert_integer", False, "convert_floating", False): np.dtype(
+                        "float"
+                    ),
+                    ("infer_objects", False): np.dtype("object"),
+                },
+            ),
+            (
+                [1, 2.5],
+                object,
+                "Float64",
+                {
+                    ("convert_floating", False): np.dtype("float"),
+                    ("infer_objects", False): np.dtype("object"),
+                },
+            ),
+            (["a", "b"], pd.CategoricalDtype(), pd.CategoricalDtype(), {}),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("s"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                {},
+            ),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("ms"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                {},
+            ),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("us"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                {},
+            ),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("ns"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                pd.DatetimeTZDtype(tz="UTC"),
+                {},
+            ),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("ns"),
+                "datetime64[ns]",
+                np.dtype("datetime64[ns]"),
+                {},
+            ),
+            (
+                pd.to_datetime(["2020-01-14 10:00", "2020-01-15 11:11"]).as_unit("ns"),
+                object,
+                np.dtype("datetime64[ns]"),
+                {("infer_objects", False): np.dtype("object")},
+            ),
+            (
+                pd.period_range("1/1/2011", freq="M", periods=3),
+                None,
+                pd.PeriodDtype("M"),
+                {},
+            ),
+            (
+                pd.arrays.IntervalArray([pd.Interval(0, 1), pd.Interval(1, 5)]),
+                None,
+                pd.IntervalDtype("int64", "right"),
+                {},
+            ),
+        ],
+    )
+    @pytest.mark.parametrize("params", product(*[(True, False)] * 5))
+    def test_convert_dtypes(
+        self,
+        data,
+        maindtype,
+        expected_default,
+        expected_other,
+        params,
+        using_infer_string,
+    ):
+        if (
+            hasattr(data, "dtype")
+            and lib.is_np_dtype(data.dtype, "M")
+            and isinstance(maindtype, pd.DatetimeTZDtype)
+        ):
+            # this astype is deprecated in favor of tz_localize
+            msg = "Cannot use .astype to convert from timezone-naive dtype"
+            with pytest.raises(TypeError, match=msg):
+                pd.Series(data, dtype=maindtype)
+            return
+
+        if maindtype is not None:
+            series = pd.Series(data, dtype=maindtype)
+        else:
+            series = pd.Series(data)
+
+        result = series.convert_dtypes(*params)
+
+        param_names = [
+            "infer_objects",
+            "convert_string",
+            "convert_integer",
+            "convert_boolean",
+            "convert_floating",
+        ]
+        params_dict = dict(zip(param_names, params))
+
+        expected_dtype = expected_default
+        for spec, dtype in expected_other.items():
+            if all(params_dict[key] is val for key, val in zip(spec[::2], spec[1::2])):
+                expected_dtype = dtype
+        if (
+            using_infer_string
+            and expected_default == "string"
+            and expected_dtype == object
+            and params[0]
+            and not params[1]
+        ):
+            # If convert_string=False and infer_objects=True, we end up with the
+            # default string dtype instead of preserving object for string data
+            expected_dtype = pd.StringDtype(na_value=np.nan)
+
+        expected = pd.Series(data, dtype=expected_dtype)
+        tm.assert_series_equal(result, expected)
+
+        # Test that it is a copy
+        copy = series.copy(deep=True)
+
+        if result.notna().sum() > 0 and result.dtype in ["interval[int64, right]"]:
+            with pytest.raises(TypeError, match="Invalid value"):
+                result[result.notna()] = np.nan
+        else:
+            result[result.notna()] = np.nan
+
+        # Make sure original not changed
+        tm.assert_series_equal(series, copy)
+
+    def test_convert_string_dtype(self, nullable_string_dtype):
+        # https://github.com/pandas-dev/pandas/issues/31731 -> converting columns
+        # that are already string dtype
+        df = pd.DataFrame(
+            {"A": ["a", "b", pd.NA], "B": ["ä", "ö", "ü"]}, dtype=nullable_string_dtype
+        )
+        result = df.convert_dtypes()
+        tm.assert_frame_equal(df, result)
+
+    def test_convert_bool_dtype(self):
+        # GH32287
+        df = pd.DataFrame({"A": pd.array([True])})
+        tm.assert_frame_equal(df, df.convert_dtypes())
+
+    def test_convert_byte_string_dtype(self):
+        # GH-43183
+        byte_str = b"binary-string"
+
+        df = pd.DataFrame(data={"A": byte_str}, index=[0])
+        result = df.convert_dtypes()
+        expected = df
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "infer_objects, dtype", [(True, "Int64"), (False, "object")]
+    )
+    def test_convert_dtype_object_with_na(self, infer_objects, dtype):
+        # GH#48791
+        ser = pd.Series([1, pd.NA])
+        result = ser.convert_dtypes(infer_objects=infer_objects)
+        expected = pd.Series([1, pd.NA], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "infer_objects, dtype", [(True, "Float64"), (False, "object")]
+    )
+    def test_convert_dtype_object_with_na_float(self, infer_objects, dtype):
+        # GH#48791
+        ser = pd.Series([1.5, pd.NA])
+        result = ser.convert_dtypes(infer_objects=infer_objects)
+        expected = pd.Series([1.5, pd.NA], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_convert_dtypes_pyarrow_to_np_nullable(self):
+        # GH 53648
+        pytest.importorskip("pyarrow")
+        ser = pd.Series(range(2), dtype="int32[pyarrow]")
+        result = ser.convert_dtypes(dtype_backend="numpy_nullable")
+        expected = pd.Series(range(2), dtype="Int32")
+        tm.assert_series_equal(result, expected)
+
+    def test_convert_dtypes_pyarrow_null(self):
+        # GH#55346
+        pa = pytest.importorskip("pyarrow")
+        ser = pd.Series([None, None])
+        result = ser.convert_dtypes(dtype_backend="pyarrow")
+        expected = pd.Series([None, None], dtype=pd.ArrowDtype(pa.null()))
+        tm.assert_series_equal(result, expected)
+
+    def test_convert_dtype_pyarrow_timezone_preserve(self):
+        # GH 60237
+        pytest.importorskip("pyarrow")
+        ser = pd.Series(
+            pd.to_datetime(range(5), utc=True, unit="h"),
+            dtype="timestamp[ns, tz=UTC][pyarrow]",
+        )
+        result = ser.convert_dtypes(dtype_backend="pyarrow")
+        expected = ser.copy()
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_copy.py b/pandas/pandas/tests/series/methods/test_copy.py
new file mode 100644
index 0000000000000..ad5417d330d51
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_copy.py
@@ -0,0 +1,77 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestCopy:
+    @pytest.mark.parametrize("deep", ["default", None, False, True])
+    def test_copy(self, deep):
+        ser = Series(np.arange(10), dtype="float64")
+
+        # default deep is True
+        if deep == "default":
+            ser2 = ser.copy()
+        else:
+            ser2 = ser.copy(deep=deep)
+
+        # INFO(CoW) a shallow copy doesn't yet copy the data
+        # but parent will not be modified (CoW)
+        if deep is None or deep is False:
+            assert np.may_share_memory(ser.values, ser2.values)
+        else:
+            assert not np.may_share_memory(ser.values, ser2.values)
+
+        ser2[::2] = np.nan
+
+        # Did not modify original Series
+        assert np.isnan(ser2[0])
+        assert not np.isnan(ser[0])
+
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+    @pytest.mark.parametrize("deep", ["default", None, False, True])
+    def test_copy_tzaware(self, deep):
+        # GH#11794
+        # copy of tz-aware
+        expected = Series([Timestamp("2012/01/01", tz="UTC")])
+        expected2 = Series([Timestamp("1999/01/01", tz="UTC")])
+
+        ser = Series([Timestamp("2012/01/01", tz="UTC")])
+
+        if deep == "default":
+            ser2 = ser.copy()
+        else:
+            ser2 = ser.copy(deep=deep)
+
+        # INFO(CoW) a shallow copy doesn't yet copy the data
+        # but parent will not be modified (CoW)
+        if deep is None or deep is False:
+            assert np.may_share_memory(ser.values, ser2.values)
+        else:
+            assert not np.may_share_memory(ser.values, ser2.values)
+
+        ser2[0] = Timestamp("1999/01/01", tz="UTC")
+
+        # Did not modify original Series
+        tm.assert_series_equal(ser2, expected2)
+        tm.assert_series_equal(ser, expected)
+
+    def test_copy_name(self, datetime_series):
+        result = datetime_series.copy()
+        assert result.name == datetime_series.name
+
+    def test_copy_index_name_checking(self, datetime_series):
+        # don't want to be able to modify the index stored elsewhere after
+        # making a copy
+
+        datetime_series.index.name = None
+        assert datetime_series.index.name is None
+        assert datetime_series is datetime_series
+
+        cp = datetime_series.copy()
+        cp.index.name = "foo"
+        assert datetime_series.index.name is None
diff --git a/pandas/pandas/tests/series/methods/test_count.py b/pandas/pandas/tests/series/methods/test_count.py
new file mode 100644
index 0000000000000..2172a3205ef55
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_count.py
@@ -0,0 +1,24 @@
+import numpy as np
+
+from pandas import (
+    Categorical,
+    Series,
+)
+
+
+class TestSeriesCount:
+    def test_count(self, datetime_series):
+        assert datetime_series.count() == len(datetime_series)
+
+        datetime_series[::2] = np.nan
+
+        assert datetime_series.count() == np.isfinite(datetime_series).sum()
+
+    def test_count_categorical(self):
+        ser = Series(
+            Categorical(
+                [np.nan, 1, 2, np.nan], categories=[5, 4, 3, 2, 1], ordered=True
+            )
+        )
+        result = ser.count()
+        assert result == 2
diff --git a/pandas/pandas/tests/series/methods/test_cov_corr.py b/pandas/pandas/tests/series/methods/test_cov_corr.py
new file mode 100644
index 0000000000000..7a4d48fb76940
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_cov_corr.py
@@ -0,0 +1,186 @@
+import math
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestSeriesCov:
+    def test_cov(self, datetime_series):
+        # full overlap
+        tm.assert_almost_equal(
+            datetime_series.cov(datetime_series), datetime_series.std() ** 2
+        )
+
+        # partial overlap
+        tm.assert_almost_equal(
+            datetime_series[:15].cov(datetime_series[5:]),
+            datetime_series[5:15].std() ** 2,
+        )
+
+        # No overlap
+        assert np.isnan(datetime_series[::2].cov(datetime_series[1::2]))
+
+        # all NA
+        cp = datetime_series[:10].copy()
+        cp[:] = np.nan
+        assert isna(cp.cov(cp))
+
+        # min_periods
+        assert isna(datetime_series[:15].cov(datetime_series[5:], min_periods=12))
+
+        ts1 = datetime_series[:15].reindex(datetime_series.index)
+        ts2 = datetime_series[5:].reindex(datetime_series.index)
+        assert isna(ts1.cov(ts2, min_periods=12))
+
+    @pytest.mark.parametrize("test_ddof", [None, 0, 1, 2, 3])
+    @pytest.mark.parametrize("dtype", ["float64", "Float64"])
+    def test_cov_ddof(self, test_ddof, dtype):
+        # GH#34611
+        np_array1 = np.random.default_rng(2).random(10)
+        np_array2 = np.random.default_rng(2).random(10)
+
+        s1 = Series(np_array1, dtype=dtype)
+        s2 = Series(np_array2, dtype=dtype)
+
+        result = s1.cov(s2, ddof=test_ddof)
+        expected = np.cov(np_array1, np_array2, ddof=test_ddof)[0][1]
+        assert math.isclose(expected, result)
+
+
+class TestSeriesCorr:
+    def test_corr(self, datetime_series, any_float_dtype):
+        stats = pytest.importorskip("scipy.stats")
+
+        datetime_series = datetime_series.astype(any_float_dtype)
+
+        # full overlap
+        tm.assert_almost_equal(datetime_series.corr(datetime_series), 1)
+
+        # partial overlap
+        tm.assert_almost_equal(datetime_series[:15].corr(datetime_series[5:]), 1)
+
+        assert isna(datetime_series[:15].corr(datetime_series[5:], min_periods=12))
+
+        ts1 = datetime_series[:15].reindex(datetime_series.index)
+        ts2 = datetime_series[5:].reindex(datetime_series.index)
+        assert isna(ts1.corr(ts2, min_periods=12))
+
+        # No overlap
+        assert np.isnan(datetime_series[::2].corr(datetime_series[1::2]))
+
+        # all NA
+        cp = datetime_series[:10].copy()
+        cp[:] = np.nan
+        assert isna(cp.corr(cp))
+
+        A = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        result = A.corr(A)
+        expected, _ = stats.pearsonr(A, A)
+        tm.assert_almost_equal(result, expected)
+
+    def test_corr_rank(self):
+        stats = pytest.importorskip("scipy.stats")
+
+        # kendall and spearman
+        B = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        A = Series(
+            np.concatenate([np.arange(5, dtype=np.float64)] * 2),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        result = A.corr(B, method="kendall")
+        expected = stats.kendalltau(A, B)[0]
+        tm.assert_almost_equal(result, expected)
+
+        result = A.corr(B, method="spearman")
+        expected = stats.spearmanr(A, B)[0]
+        tm.assert_almost_equal(result, expected)
+
+        # results from R
+        A = Series(
+            [
+                -0.89926396,
+                0.94209606,
+                -1.03289164,
+                -0.95445587,
+                0.76910310,
+                -0.06430576,
+                -2.09704447,
+                0.40660407,
+                -0.89926396,
+                0.94209606,
+            ]
+        )
+        B = Series(
+            [
+                -1.01270225,
+                -0.62210117,
+                -1.56895827,
+                0.59592943,
+                -0.01680292,
+                1.17258718,
+                -1.06009347,
+                -0.10222060,
+                -0.89076239,
+                0.89372375,
+            ]
+        )
+        kexp = 0.4319297
+        sexp = 0.5853767
+        tm.assert_almost_equal(A.corr(B, method="kendall"), kexp)
+        tm.assert_almost_equal(A.corr(B, method="spearman"), sexp)
+
+    def test_corr_invalid_method(self):
+        # GH PR #22298
+        s1 = Series(np.random.default_rng(2).standard_normal(10))
+        s2 = Series(np.random.default_rng(2).standard_normal(10))
+        msg = "method must be either 'pearson', 'spearman', 'kendall', or a callable, "
+        with pytest.raises(ValueError, match=msg):
+            s1.corr(s2, method="____")
+
+    def test_corr_callable_method(self, datetime_series):
+        # simple correlation example
+        # returns 1 if exact equality, 0 otherwise
+        my_corr = lambda a, b: 1.0 if (a == b).all() else 0.0
+
+        # simple example
+        s1 = Series([1, 2, 3, 4, 5])
+        s2 = Series([5, 4, 3, 2, 1])
+        expected = 0
+        tm.assert_almost_equal(s1.corr(s2, method=my_corr), expected)
+
+        # full overlap
+        tm.assert_almost_equal(
+            datetime_series.corr(datetime_series, method=my_corr), 1.0
+        )
+
+        # partial overlap
+        tm.assert_almost_equal(
+            datetime_series[:15].corr(datetime_series[5:], method=my_corr), 1.0
+        )
+
+        # No overlap
+        assert np.isnan(
+            datetime_series[::2].corr(datetime_series[1::2], method=my_corr)
+        )
+
+        # dataframe example
+        df = pd.DataFrame([s1, s2])
+        expected = pd.DataFrame([{0: 1.0, 1: 0}, {0: 0, 1: 1.0}])
+        tm.assert_almost_equal(df.transpose().corr(method=my_corr), expected)
diff --git a/pandas/pandas/tests/series/methods/test_describe.py b/pandas/pandas/tests/series/methods/test_describe.py
new file mode 100644
index 0000000000000..79ec11feb5308
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_describe.py
@@ -0,0 +1,203 @@
+import numpy as np
+import pytest
+
+from pandas.compat.numpy import np_version_gte1p25
+
+from pandas.core.dtypes.common import (
+    is_complex_dtype,
+    is_extension_array_dtype,
+)
+
+from pandas import (
+    NA,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesDescribe:
+    def test_describe_ints(self):
+        ser = Series([0, 1, 2, 3, 4], name="int_data")
+        result = ser.describe()
+        expected = Series(
+            [5, 2, ser.std(), 0, 1, 2, 3, 4],
+            name="int_data",
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_bools(self):
+        ser = Series([True, True, False, False, False], name="bool_data")
+        result = ser.describe()
+        expected = Series(
+            [5, 2, False, 3], name="bool_data", index=["count", "unique", "top", "freq"]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_strs(self):
+        ser = Series(["a", "a", "b", "c", "d"], name="str_data")
+        result = ser.describe()
+        expected = Series(
+            [5, 4, "a", 2], name="str_data", index=["count", "unique", "top", "freq"]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_timedelta64(self):
+        ser = Series(
+            [
+                Timedelta("1 days"),
+                Timedelta("2 days"),
+                Timedelta("3 days"),
+                Timedelta("4 days"),
+                Timedelta("5 days"),
+            ],
+            name="timedelta_data",
+        )
+        result = ser.describe()
+        expected = Series(
+            [5, ser[2], ser.std(), ser[0], ser[1], ser[2], ser[3], ser[4]],
+            name="timedelta_data",
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_period(self):
+        ser = Series(
+            [Period("2020-01", "M"), Period("2020-01", "M"), Period("2019-12", "M")],
+            name="period_data",
+        )
+        result = ser.describe()
+        expected = Series(
+            [3, 2, ser[0], 2],
+            name="period_data",
+            index=["count", "unique", "top", "freq"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_empty_object(self):
+        # https://github.com/pandas-dev/pandas/issues/27183
+        s = Series([None, None], dtype=object)
+        result = s.describe()
+        expected = Series(
+            [0, 0, np.nan, np.nan],
+            dtype=object,
+            index=["count", "unique", "top", "freq"],
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = s[:0].describe()
+        tm.assert_series_equal(result, expected)
+        # ensure NaN, not None
+        assert np.isnan(result.iloc[2])
+        assert np.isnan(result.iloc[3])
+
+    def test_describe_with_tz(self, tz_naive_fixture):
+        # GH 21332
+        tz = tz_naive_fixture
+        name = str(tz_naive_fixture)
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s = Series(date_range(start, end, tz=tz), name=name)
+        result = s.describe()
+        expected = Series(
+            [
+                5,
+                Timestamp(2018, 1, 3).tz_localize(tz),
+                start.tz_localize(tz),
+                s[1],
+                s[2],
+                s[3],
+                end.tz_localize(tz),
+            ],
+            name=name,
+            index=["count", "mean", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_with_tz_numeric(self):
+        name = tz = "CET"
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s = Series(date_range(start, end, tz=tz), name=name)
+
+        result = s.describe()
+
+        expected = Series(
+            [
+                5,
+                Timestamp("2018-01-03 00:00:00", tz=tz),
+                Timestamp("2018-01-01 00:00:00", tz=tz),
+                Timestamp("2018-01-02 00:00:00", tz=tz),
+                Timestamp("2018-01-03 00:00:00", tz=tz),
+                Timestamp("2018-01-04 00:00:00", tz=tz),
+                Timestamp("2018-01-05 00:00:00", tz=tz),
+            ],
+            name=name,
+            index=["count", "mean", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_datetime_is_numeric_includes_datetime(self):
+        s = Series(date_range("2012", periods=3))
+        result = s.describe()
+        expected = Series(
+            [
+                3,
+                Timestamp("2012-01-02"),
+                Timestamp("2012-01-01"),
+                Timestamp("2012-01-01T12:00:00"),
+                Timestamp("2012-01-02"),
+                Timestamp("2012-01-02T12:00:00"),
+                Timestamp("2012-01-03"),
+            ],
+            index=["count", "mean", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings("ignore:Casting complex values to real discards")
+    def test_numeric_result_dtype(self, any_numeric_dtype):
+        # GH#48340 - describe should always return float on non-complex numeric input
+        if is_extension_array_dtype(any_numeric_dtype):
+            dtype = "Float64"
+        else:
+            dtype = "complex128" if is_complex_dtype(any_numeric_dtype) else None
+
+        ser = Series([0, 1], dtype=any_numeric_dtype)
+        if dtype == "complex128" and np_version_gte1p25:
+            with pytest.raises(
+                TypeError, match=r"^a must be an array of real numbers$"
+            ):
+                ser.describe()
+            return
+        result = ser.describe()
+        expected = Series(
+            [
+                2.0,
+                0.5,
+                ser.std(),
+                0,
+                0.25,
+                0.5,
+                0.75,
+                1.0,
+            ],
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+            dtype=dtype,
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_describe_one_element_ea(self):
+        # GH#52515
+        ser = Series([0.0], dtype="Float64")
+        with tm.assert_produces_warning(None):
+            result = ser.describe()
+        expected = Series(
+            [1, 0, NA, 0, 0, 0, 0, 0],
+            dtype="Float64",
+            index=["count", "mean", "std", "min", "25%", "50%", "75%", "max"],
+        )
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_diff.py b/pandas/pandas/tests/series/methods/test_diff.py
new file mode 100644
index 0000000000000..89c34aeb64206
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_diff.py
@@ -0,0 +1,91 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    TimedeltaIndex,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesDiff:
+    def test_diff_series_requires_integer(self):
+        series = Series(np.random.default_rng(2).standard_normal(2))
+        with pytest.raises(ValueError, match="periods must be an integer"):
+            series.diff(1.5)
+
+    def test_diff_np(self):
+        # TODO(__array_function__): could make np.diff return a Series
+        #  matching ser.diff()
+
+        ser = Series(np.arange(5))
+
+        res = np.diff(ser)
+        expected = np.array([1, 1, 1, 1])
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_diff_int(self):
+        # int dtype
+        a = 10000000000000000
+        b = a + 1
+        ser = Series([a, b])
+
+        result = ser.diff()
+        assert result[1] == 1
+
+    def test_diff_tz(self):
+        # Combined datetime diff, normal diff and boolean diff test
+        ts = Series(
+            np.arange(10, dtype=np.float64),
+            index=date_range("2020-01-01", periods=10),
+            name="ts",
+        )
+        ts.diff()
+
+        # neg n
+        result = ts.diff(-1)
+        expected = ts - ts.shift(-1)
+        tm.assert_series_equal(result, expected)
+
+        # 0
+        result = ts.diff(0)
+        expected = ts - ts
+        tm.assert_series_equal(result, expected)
+
+    def test_diff_dt64(self):
+        # datetime diff (GH#3100)
+        ser = Series(date_range("20130102", periods=5))
+        result = ser.diff()
+        expected = ser - ser.shift(1)
+        tm.assert_series_equal(result, expected)
+
+        # timedelta diff
+        result = result - result.shift(1)  # previous result
+        expected = expected.diff()  # previously expected
+        tm.assert_series_equal(result, expected)
+
+    def test_diff_dt64tz(self):
+        # with tz
+        ser = Series(
+            date_range("2000-01-01 09:00:00", periods=5, tz="US/Eastern"), name="foo"
+        )
+        result = ser.diff()
+        expected = Series(TimedeltaIndex(["NaT"] + ["1 days"] * 4), name="foo")
+        tm.assert_series_equal(result, expected)
+
+    def test_diff_bool(self):
+        # boolean series (test for fixing #17294)
+        data = [False, True, True, False, False]
+        output = [np.nan, True, False, True, False]
+        ser = Series(data)
+        result = ser.diff()
+        expected = Series(output)
+        tm.assert_series_equal(result, expected)
+
+    def test_diff_object_dtype(self):
+        # object series
+        ser = Series([False, True, 5.0, np.nan, True, False])
+        result = ser.diff()
+        expected = ser - ser.shift(1)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_drop.py b/pandas/pandas/tests/series/methods/test_drop.py
new file mode 100644
index 0000000000000..d2a5a3324e886
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_drop.py
@@ -0,0 +1,99 @@
+import pytest
+
+from pandas import (
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.api.types import is_bool_dtype
+
+
+@pytest.mark.parametrize(
+    "data, index, drop_labels, axis, expected_data, expected_index",
+    [
+        # Unique Index
+        ([1, 2], ["one", "two"], ["two"], 0, [1], ["one"]),
+        ([1, 2], ["one", "two"], ["two"], "rows", [1], ["one"]),
+        ([1, 1, 2], ["one", "two", "one"], ["two"], 0, [1, 2], ["one", "one"]),
+        # GH 5248 Non-Unique Index
+        ([1, 1, 2], ["one", "two", "one"], "two", 0, [1, 2], ["one", "one"]),
+        ([1, 1, 2], ["one", "two", "one"], ["one"], 0, [1], ["two"]),
+        ([1, 1, 2], ["one", "two", "one"], "one", 0, [1], ["two"]),
+    ],
+)
+def test_drop_unique_and_non_unique_index(
+    data, index, axis, drop_labels, expected_data, expected_index
+):
+    ser = Series(data=data, index=index)
+    result = ser.drop(drop_labels, axis=axis)
+    expected = Series(data=expected_data, index=expected_index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "drop_labels, axis, error_type, error_desc",
+    [
+        # single string/tuple-like
+        ("bc", 0, KeyError, "not found in axis"),
+        # bad axis
+        (("a",), 0, KeyError, "not found in axis"),
+        ("one", "columns", ValueError, "No axis named columns"),
+    ],
+)
+def test_drop_exception_raised(drop_labels, axis, error_type, error_desc):
+    ser = Series(range(3), index=list("abc"))
+    with pytest.raises(error_type, match=error_desc):
+        ser.drop(drop_labels, axis=axis)
+
+
+def test_drop_with_ignore_errors():
+    # errors='ignore'
+    ser = Series(range(3), index=list("abc"))
+    result = ser.drop("bc", errors="ignore")
+    tm.assert_series_equal(result, ser)
+    result = ser.drop(["a", "d"], errors="ignore")
+    expected = ser.iloc[1:]
+    tm.assert_series_equal(result, expected)
+
+    # GH 8522
+    ser = Series([2, 3], index=[True, False])
+    assert is_bool_dtype(ser.index)
+    assert ser.index.dtype == bool
+    result = ser.drop(True)
+    expected = Series([3], index=[False])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("index", [[1, 2, 3], [1, 1, 3]])
+@pytest.mark.parametrize("drop_labels", [[], [1], [3]])
+def test_drop_empty_list(index, drop_labels):
+    # GH 21494
+    expected_index = [i for i in index if i not in drop_labels]
+    series = Series(index=index, dtype=object).drop(drop_labels)
+    expected = Series(index=expected_index, dtype=object)
+    tm.assert_series_equal(series, expected)
+
+
+@pytest.mark.parametrize(
+    "data, index, drop_labels",
+    [
+        (None, [1, 2, 3], [1, 4]),
+        (None, [1, 2, 2], [1, 4]),
+        ([2, 3], [0, 1], [False, True]),
+    ],
+)
+def test_drop_non_empty_list(data, index, drop_labels):
+    # GH 21494 and GH 16877
+    dtype = object if data is None else None
+    ser = Series(data=data, index=index, dtype=dtype)
+    with pytest.raises(KeyError, match="not found in axis"):
+        ser.drop(drop_labels)
+
+
+def test_drop_index_ea_dtype(any_numeric_ea_dtype):
+    # GH#45860
+    df = Series(100, index=Index([1, 2, 2], dtype=any_numeric_ea_dtype))
+    idx = Index([df.index[1]])
+    result = df.drop(idx)
+    expected = Series(100, index=Index([1], dtype=any_numeric_ea_dtype))
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_drop_duplicates.py b/pandas/pandas/tests/series/methods/test_drop_duplicates.py
new file mode 100644
index 0000000000000..2dbd61530dc41
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_drop_duplicates.py
@@ -0,0 +1,275 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", Series([False, False, False, False, True, True, False])),
+        ("last", Series([False, True, True, False, False, False, False])),
+        (False, Series([False, True, True, False, True, True, False])),
+    ],
+)
+def test_drop_duplicates(any_numpy_dtype, keep, expected):
+    tc = Series([1, 0, 3, 5, 3, 0, 4], dtype=np.dtype(any_numpy_dtype))
+
+    if tc.dtype == "bool":
+        pytest.skip("tested separately in test_drop_duplicates_bool")
+
+    tm.assert_series_equal(tc.duplicated(keep=keep), expected)
+    tm.assert_series_equal(tc.drop_duplicates(keep=keep), tc[~expected])
+    sc = tc.copy()
+    return_value = sc.drop_duplicates(keep=keep, inplace=True)
+    assert return_value is None
+    tm.assert_series_equal(sc, tc[~expected])
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", [False, False, True, True]),
+        ("last", [True, True, False, False]),
+        (False, [True, True, True, True]),
+    ],
+)
+def test_drop_duplicates_bool(keep, expected):
+    tc = Series([True, False, True, False])
+    expected = Series(expected)
+    tm.assert_series_equal(tc.duplicated(keep=keep), expected)
+    tm.assert_series_equal(tc.drop_duplicates(keep=keep), tc[~expected])
+    sc = tc.copy()
+    return_value = sc.drop_duplicates(keep=keep, inplace=True)
+    tm.assert_series_equal(sc, tc[~expected])
+    assert return_value is None
+
+
+@pytest.mark.parametrize("values", [[], list(range(5))])
+def test_drop_duplicates_no_duplicates(any_numpy_dtype, keep, values):
+    tc = Series(values, dtype=np.dtype(any_numpy_dtype))
+    expected = Series([False] * len(tc), dtype="bool")
+
+    if tc.dtype == "bool":
+        # 0 -> False and 1-> True
+        # any other value would be duplicated
+        tc = tc[:2]
+        expected = expected[:2]
+
+    tm.assert_series_equal(tc.duplicated(keep=keep), expected)
+
+    result_dropped = tc.drop_duplicates(keep=keep)
+    tm.assert_series_equal(result_dropped, tc)
+
+    # validate shallow copy
+    assert result_dropped is not tc
+
+
+class TestSeriesDropDuplicates:
+    @pytest.fixture(
+        params=["int_", "uint", "float64", "str_", "timedelta64[h]", "datetime64[D]"]
+    )
+    def dtype(self, request):
+        """
+        Fixture that provides different data types for testing.
+        """
+        return request.param
+
+    @pytest.fixture
+    def cat_series_unused_category(self, dtype, ordered):
+        """
+        Fixture that creates a Categorical Series with some unused categories.
+        """
+        # Test case 1
+        cat_array = np.array([1, 2, 3, 4, 5], dtype=np.dtype(dtype))
+
+        input1 = np.array([1, 2, 3, 3], dtype=np.dtype(dtype))
+        cat = Categorical(input1, categories=cat_array, ordered=ordered)
+        tc1 = Series(cat)
+        return tc1
+
+    def test_drop_duplicates_categorical_non_bool(self, cat_series_unused_category):
+        tc1 = cat_series_unused_category
+
+        expected = Series([False, False, False, True])
+
+        result = tc1.duplicated()
+        tm.assert_series_equal(result, expected)
+
+        result = tc1.drop_duplicates()
+        tm.assert_series_equal(result, tc1[~expected])
+
+        sc = tc1.copy()
+        return_value = sc.drop_duplicates(inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc1[~expected])
+
+    def test_drop_duplicates_categorical_non_bool_keeplast(
+        self, cat_series_unused_category
+    ):
+        tc1 = cat_series_unused_category
+
+        expected = Series([False, False, True, False])
+
+        result = tc1.duplicated(keep="last")
+        tm.assert_series_equal(result, expected)
+
+        result = tc1.drop_duplicates(keep="last")
+        tm.assert_series_equal(result, tc1[~expected])
+
+        sc = tc1.copy()
+        return_value = sc.drop_duplicates(keep="last", inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc1[~expected])
+
+    def test_drop_duplicates_categorical_non_bool_keepfalse(
+        self, cat_series_unused_category
+    ):
+        tc1 = cat_series_unused_category
+
+        expected = Series([False, False, True, True])
+
+        result = tc1.duplicated(keep=False)
+        tm.assert_series_equal(result, expected)
+
+        result = tc1.drop_duplicates(keep=False)
+        tm.assert_series_equal(result, tc1[~expected])
+
+        sc = tc1.copy()
+        return_value = sc.drop_duplicates(keep=False, inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc1[~expected])
+
+    @pytest.fixture
+    def cat_series(self, dtype, ordered):
+        """
+        Fixture that creates a Categorical Series with no unused categories.
+        """
+        cat_array = np.array([1, 2, 3, 4, 5], dtype=np.dtype(dtype))
+
+        input2 = np.array([1, 2, 3, 5, 3, 2, 4], dtype=np.dtype(dtype))
+        cat = Categorical(input2, categories=cat_array, ordered=ordered)
+        tc2 = Series(cat)
+        return tc2
+
+    def test_drop_duplicates_categorical_non_bool2(self, cat_series):
+        tc2 = cat_series
+
+        expected = Series([False, False, False, False, True, True, False])
+
+        result = tc2.duplicated()
+        tm.assert_series_equal(result, expected)
+
+        result = tc2.drop_duplicates()
+        tm.assert_series_equal(result, tc2[~expected])
+
+        sc = tc2.copy()
+        return_value = sc.drop_duplicates(inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc2[~expected])
+
+    def test_drop_duplicates_categorical_non_bool2_keeplast(self, cat_series):
+        tc2 = cat_series
+
+        expected = Series([False, True, True, False, False, False, False])
+
+        result = tc2.duplicated(keep="last")
+        tm.assert_series_equal(result, expected)
+
+        result = tc2.drop_duplicates(keep="last")
+        tm.assert_series_equal(result, tc2[~expected])
+
+        sc = tc2.copy()
+        return_value = sc.drop_duplicates(keep="last", inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc2[~expected])
+
+    def test_drop_duplicates_categorical_non_bool2_keepfalse(self, cat_series):
+        tc2 = cat_series
+
+        expected = Series([False, True, True, False, True, True, False])
+
+        result = tc2.duplicated(keep=False)
+        tm.assert_series_equal(result, expected)
+
+        result = tc2.drop_duplicates(keep=False)
+        tm.assert_series_equal(result, tc2[~expected])
+
+        sc = tc2.copy()
+        return_value = sc.drop_duplicates(keep=False, inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc2[~expected])
+
+    def test_drop_duplicates_categorical_bool(self, ordered):
+        tc = Series(
+            Categorical(
+                [True, False, True, False], categories=[True, False], ordered=ordered
+            )
+        )
+
+        expected = Series([False, False, True, True])
+        tm.assert_series_equal(tc.duplicated(), expected)
+        tm.assert_series_equal(tc.drop_duplicates(), tc[~expected])
+        sc = tc.copy()
+        return_value = sc.drop_duplicates(inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc[~expected])
+
+        expected = Series([True, True, False, False])
+        tm.assert_series_equal(tc.duplicated(keep="last"), expected)
+        tm.assert_series_equal(tc.drop_duplicates(keep="last"), tc[~expected])
+        sc = tc.copy()
+        return_value = sc.drop_duplicates(keep="last", inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc[~expected])
+
+        expected = Series([True, True, True, True])
+        tm.assert_series_equal(tc.duplicated(keep=False), expected)
+        tm.assert_series_equal(tc.drop_duplicates(keep=False), tc[~expected])
+        sc = tc.copy()
+        return_value = sc.drop_duplicates(keep=False, inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(sc, tc[~expected])
+
+    def test_drop_duplicates_categorical_bool_na(self, nulls_fixture):
+        # GH#44351
+        ser = Series(
+            Categorical(
+                [True, False, True, False, nulls_fixture],
+                categories=[True, False],
+                ordered=True,
+            )
+        )
+        result = ser.drop_duplicates()
+        expected = Series(
+            Categorical([True, False, np.nan], categories=[True, False], ordered=True),
+            index=[0, 1, 4],
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_drop_duplicates_ignore_index(self):
+        # GH#48304
+        ser = Series([1, 2, 2, 3])
+        result = ser.drop_duplicates(ignore_index=True)
+        expected = Series([1, 2, 3])
+        tm.assert_series_equal(result, expected)
+
+    def test_duplicated_arrow_dtype(self):
+        pytest.importorskip("pyarrow")
+        ser = Series([True, False, None, False], dtype="bool[pyarrow]")
+        result = ser.drop_duplicates()
+        expected = Series([True, False, None], dtype="bool[pyarrow]")
+        tm.assert_series_equal(result, expected)
+
+    def test_drop_duplicates_arrow_strings(self):
+        # GH#54904
+        pa = pytest.importorskip("pyarrow")
+        ser = Series(["a", "a"], dtype=pd.ArrowDtype(pa.string()))
+        result = ser.drop_duplicates()
+        expecetd = Series(["a"], dtype=pd.ArrowDtype(pa.string()))
+        tm.assert_series_equal(result, expecetd)
diff --git a/pandas/pandas/tests/series/methods/test_dropna.py b/pandas/pandas/tests/series/methods/test_dropna.py
new file mode 100644
index 0000000000000..d03fcac24003e
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_dropna.py
@@ -0,0 +1,117 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    IntervalIndex,
+    NaT,
+    Period,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestDropna:
+    def test_dropna_empty(self):
+        ser = Series([], dtype=object)
+
+        assert len(ser.dropna()) == 0
+        return_value = ser.dropna(inplace=True)
+        assert return_value is None
+        assert len(ser) == 0
+
+        # invalid axis
+        msg = "No axis named 1 for object type Series"
+        with pytest.raises(ValueError, match=msg):
+            ser.dropna(axis=1)
+
+    def test_dropna_preserve_name(self, datetime_series):
+        datetime_series[:5] = np.nan
+        result = datetime_series.dropna()
+        assert result.name == datetime_series.name
+        name = datetime_series.name
+        ts = datetime_series.copy()
+        return_value = ts.dropna(inplace=True)
+        assert return_value is None
+        assert ts.name == name
+
+    def test_dropna_no_nan(self):
+        for ser in [
+            Series([1, 2, 3], name="x"),
+            Series([False, True, False], name="x"),
+        ]:
+            result = ser.dropna()
+            tm.assert_series_equal(result, ser)
+            assert result is not ser
+
+            s2 = ser.copy()
+            return_value = s2.dropna(inplace=True)
+            assert return_value is None
+            tm.assert_series_equal(s2, ser)
+
+    def test_dropna_intervals(self):
+        ser = Series(
+            [np.nan, 1, 2, 3],
+            IntervalIndex.from_arrays([np.nan, 0, 1, 2], [np.nan, 1, 2, 3]),
+        )
+
+        result = ser.dropna()
+        expected = ser.iloc[1:]
+        tm.assert_series_equal(result, expected)
+
+    def test_dropna_period_dtype(self):
+        # GH#13737
+        ser = Series([Period("2011-01", freq="M"), Period("NaT", freq="M")])
+        result = ser.dropna()
+        expected = Series([Period("2011-01", freq="M")])
+
+        tm.assert_series_equal(result, expected)
+
+    def test_datetime64_tz_dropna(self, unit):
+        # DatetimeLikeBlock
+        ser = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                NaT,
+                Timestamp("2011-01-03 10:00"),
+                NaT,
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        result = ser.dropna()
+        expected = Series(
+            [Timestamp("2011-01-01 10:00"), Timestamp("2011-01-03 10:00")],
+            index=[0, 2],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        # DatetimeTZBlock
+        idx = DatetimeIndex(
+            ["2011-01-01 10:00", NaT, "2011-01-03 10:00", NaT], tz="Asia/Tokyo"
+        ).as_unit(unit)
+        ser = Series(idx)
+        assert ser.dtype == f"datetime64[{unit}, Asia/Tokyo]"
+        result = ser.dropna()
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz="Asia/Tokyo"),
+                Timestamp("2011-01-03 10:00", tz="Asia/Tokyo"),
+            ],
+            index=[0, 2],
+            dtype=f"datetime64[{unit}, Asia/Tokyo]",
+        )
+        assert result.dtype == f"datetime64[{unit}, Asia/Tokyo]"
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("val", [1, 1.5])
+    def test_dropna_ignore_index(self, val):
+        # GH#31725
+        ser = Series([1, 2, val], index=[3, 2, 1])
+        result = ser.dropna(ignore_index=True)
+        expected = Series([1, 2, val])
+        tm.assert_series_equal(result, expected)
+
+        ser.dropna(ignore_index=True, inplace=True)
+        tm.assert_series_equal(ser, expected)
diff --git a/pandas/pandas/tests/series/methods/test_dtypes.py b/pandas/pandas/tests/series/methods/test_dtypes.py
new file mode 100644
index 0000000000000..82260bc2a65b9
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_dtypes.py
@@ -0,0 +1,7 @@
+import numpy as np
+
+
+class TestSeriesDtypes:
+    def test_dtype(self, datetime_series):
+        assert datetime_series.dtype == np.dtype("float64")
+        assert datetime_series.dtypes == np.dtype("float64")
diff --git a/pandas/pandas/tests/series/methods/test_duplicated.py b/pandas/pandas/tests/series/methods/test_duplicated.py
new file mode 100644
index 0000000000000..f5a387a7f302e
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_duplicated.py
@@ -0,0 +1,79 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    Categorical,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", [False, False, True, False, True]),
+        ("last", [True, True, False, False, False]),
+        (False, [True, True, True, False, True]),
+    ],
+)
+def test_duplicated_keep(keep, expected):
+    ser = Series(["a", "b", "b", "c", "a"], name="name")
+
+    result = ser.duplicated(keep=keep)
+    expected = Series(expected, name="name")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keep, expected",
+    [
+        ("first", [False, False, True, False, True]),
+        ("last", [True, True, False, False, False]),
+        (False, [True, True, True, False, True]),
+    ],
+)
+def test_duplicated_nan_none(keep, expected):
+    ser = Series([np.nan, 3, 3, None, np.nan], dtype=object)
+
+    result = ser.duplicated(keep=keep)
+    expected = Series(expected)
+    tm.assert_series_equal(result, expected)
+
+
+def test_duplicated_categorical_bool_na(nulls_fixture):
+    # GH#44351
+    ser = Series(
+        Categorical(
+            [True, False, True, False, nulls_fixture],
+            categories=[True, False],
+            ordered=True,
+        )
+    )
+    result = ser.duplicated()
+    expected = Series([False, False, True, True, False])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "keep, vals",
+    [
+        ("last", [True, True, False]),
+        ("first", [False, True, True]),
+        (False, [True, True, True]),
+    ],
+)
+def test_duplicated_mask(keep, vals):
+    # GH#48150
+    ser = Series([1, 2, NA, NA, NA], dtype="Int64")
+    result = ser.duplicated(keep=keep)
+    expected = Series([False, False] + vals)
+    tm.assert_series_equal(result, expected)
+
+
+def test_duplicated_mask_no_duplicated_na(keep):
+    # GH#48150
+    ser = Series([1, 2, NA], dtype="Int64")
+    result = ser.duplicated(keep=keep)
+    expected = Series([False, False, False])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_equals.py b/pandas/pandas/tests/series/methods/test_equals.py
new file mode 100644
index 0000000000000..0c52eacd7e516
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_equals.py
@@ -0,0 +1,133 @@
+import copy
+
+import numpy as np
+import pytest
+
+from pandas._libs.missing import is_matching_na
+
+from pandas.core.dtypes.common import is_float
+
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+)
+
+
+@pytest.mark.parametrize(
+    "arr, idx",
+    [
+        ([1, 2, 3, 4], [0, 2, 1, 3]),
+        ([1, np.nan, 3, np.nan], [0, 2, 1, 3]),
+        (
+            [1, np.nan, 3, np.nan],
+            MultiIndex.from_tuples([(0, "a"), (1, "b"), (2, "c"), (3, "c")]),
+        ),
+    ],
+)
+def test_equals(arr, idx):
+    s1 = Series(arr, index=idx)
+    s2 = s1.copy()
+    assert s1.equals(s2)
+
+    s1[1] = 9
+    assert not s1.equals(s2)
+
+
+@pytest.mark.parametrize(
+    "val", [1, 1.1, 1 + 1j, True, "abc", [1, 2], (1, 2), {1, 2}, {"a": 1}, None]
+)
+def test_equals_list_array(val):
+    # GH20676 Verify equals operator for list of Numpy arrays
+    arr = np.array([1, 2])
+    s1 = Series([arr, arr])
+    s2 = s1.copy()
+    assert s1.equals(s2)
+
+    s1[1] = val
+    assert not s1.equals(s2)
+
+
+def test_equals_false_negative():
+    # GH8437 Verify false negative behavior of equals function for dtype object
+    arr = [False, np.nan]
+    s1 = Series(arr)
+    s2 = s1.copy()
+    s3 = Series(index=range(2), dtype=object)
+    s4 = s3.copy()
+    s5 = s3.copy()
+    s6 = s3.copy()
+
+    s3[:-1] = s4[:-1] = s5[0] = s6[0] = False
+    assert s1.equals(s1)
+    assert s1.equals(s2)
+    assert s1.equals(s3)
+    assert s1.equals(s4)
+    assert s1.equals(s5)
+    assert s5.equals(s6)
+
+
+def test_equals_matching_nas():
+    # matching but not identical NAs
+    left = Series([np.datetime64("NaT")], dtype=object)
+    right = Series([np.datetime64("NaT")], dtype=object)
+    assert left.equals(right)
+    assert Index(left).equals(Index(right))
+    assert left.array.equals(right.array)
+
+    left = Series([np.timedelta64("NaT")], dtype=object)
+    right = Series([np.timedelta64("NaT")], dtype=object)
+    assert left.equals(right)
+    assert Index(left).equals(Index(right))
+    assert left.array.equals(right.array)
+
+    left = Series([np.float64("NaN")], dtype=object)
+    right = Series([np.float64("NaN")], dtype=object)
+    assert left.equals(right)
+    assert Index(left, dtype=left.dtype).equals(Index(right, dtype=right.dtype))
+    assert left.array.equals(right.array)
+
+
+def test_equals_mismatched_nas(nulls_fixture, nulls_fixture2):
+    # GH#39650
+    left = nulls_fixture
+    right = nulls_fixture2
+    if hasattr(right, "copy"):
+        right = right.copy()
+    else:
+        right = copy.copy(right)
+
+    ser = Series([left], dtype=object)
+    ser2 = Series([right], dtype=object)
+
+    if is_matching_na(left, right):
+        assert ser.equals(ser2)
+    elif (left is None and is_float(right)) or (right is None and is_float(left)):
+        assert ser.equals(ser2)
+    else:
+        assert not ser.equals(ser2)
+
+
+def test_equals_none_vs_nan():
+    # GH#39650
+    ser = Series([1, None], dtype=object)
+    ser2 = Series([1, np.nan], dtype=object)
+
+    assert ser.equals(ser2)
+    assert Index(ser, dtype=ser.dtype).equals(Index(ser2, dtype=ser2.dtype))
+    assert ser.array.equals(ser2.array)
+
+
+def test_equals_None_vs_float():
+    # GH#44190
+    left = Series([-np.inf, np.nan, -1.0, 0.0, 1.0, 10 / 3, np.inf], dtype=object)
+    right = Series([None] * len(left))
+
+    # these series were found to be equal due to a bug, check that they are correctly
+    # found to not equal
+    assert not left.equals(right)
+    assert not right.equals(left)
+    assert not left.to_frame().equals(right.to_frame())
+    assert not right.to_frame().equals(left.to_frame())
+    assert not Index(left, dtype="object").equals(Index(right, dtype="object"))
+    assert not Index(right, dtype="object").equals(Index(left, dtype="object"))
diff --git a/pandas/pandas/tests/series/methods/test_explode.py b/pandas/pandas/tests/series/methods/test_explode.py
new file mode 100644
index 0000000000000..15d615fc35081
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_explode.py
@@ -0,0 +1,176 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_basic():
+    s = pd.Series([[0, 1, 2], np.nan, [], (3, 4)], index=list("abcd"), name="foo")
+    result = s.explode()
+    expected = pd.Series(
+        [0, 1, 2, np.nan, np.nan, 3, 4], index=list("aaabcdd"), dtype=object, name="foo"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_mixed_type():
+    s = pd.Series(
+        [[0, 1, 2], np.nan, None, np.array([]), pd.Series(["a", "b"])], name="foo"
+    )
+    result = s.explode()
+    expected = pd.Series(
+        [0, 1, 2, np.nan, None, np.nan, "a", "b"],
+        index=[0, 0, 0, 1, 2, 3, 4, 4],
+        dtype=object,
+        name="foo",
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_empty():
+    s = pd.Series(dtype=object)
+    result = s.explode()
+    expected = s.copy()
+    tm.assert_series_equal(result, expected)
+
+
+def test_nested_lists():
+    s = pd.Series([[[1, 2, 3]], [1, 2], 1])
+    result = s.explode()
+    expected = pd.Series([[1, 2, 3], 1, 2, 1], index=[0, 1, 1, 2])
+    tm.assert_series_equal(result, expected)
+
+
+def test_multi_index():
+    s = pd.Series(
+        [[0, 1, 2], np.nan, [], (3, 4)],
+        name="foo",
+        index=pd.MultiIndex.from_product([list("ab"), range(2)], names=["foo", "bar"]),
+    )
+    result = s.explode()
+    index = pd.MultiIndex.from_tuples(
+        [("a", 0), ("a", 0), ("a", 0), ("a", 1), ("b", 0), ("b", 1), ("b", 1)],
+        names=["foo", "bar"],
+    )
+    expected = pd.Series(
+        [0, 1, 2, np.nan, np.nan, 3, 4], index=index, dtype=object, name="foo"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_large():
+    s = pd.Series([range(256)]).explode()
+    result = s.explode()
+    tm.assert_series_equal(result, s)
+
+
+def test_invert_array():
+    df = pd.DataFrame({"a": pd.date_range("20190101", periods=3, tz="UTC")})
+
+    listify = df.apply(lambda x: x.array, axis=1)
+    result = listify.explode()
+    tm.assert_series_equal(result, df["a"].rename())
+
+
+@pytest.mark.parametrize(
+    "data", [[1, 2, 3], pd.date_range("2019", periods=3, tz="UTC")]
+)
+def test_non_object_dtype(data):
+    ser = pd.Series(data)
+    result = ser.explode()
+    tm.assert_series_equal(result, ser)
+
+
+def test_typical_usecase():
+    df = pd.DataFrame(
+        [{"var1": "a,b,c", "var2": 1}, {"var1": "d,e,f", "var2": 2}],
+        columns=["var1", "var2"],
+    )
+    exploded = df.var1.str.split(",").explode()
+    result = df[["var2"]].join(exploded)
+    expected = pd.DataFrame(
+        {"var2": [1, 1, 1, 2, 2, 2], "var1": list("abcdef")},
+        columns=["var2", "var1"],
+        index=[0, 0, 0, 1, 1, 1],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_nested_EA():
+    # a nested EA array
+    s = pd.Series(
+        [
+            pd.date_range("20170101", periods=3, tz="UTC"),
+            pd.date_range("20170104", periods=3, tz="UTC"),
+        ]
+    )
+    result = s.explode()
+    expected = pd.Series(
+        pd.date_range("20170101", periods=6, tz="UTC"), index=[0, 0, 0, 1, 1, 1]
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_duplicate_index():
+    # GH 28005
+    s = pd.Series([[1, 2], [3, 4]], index=[0, 0])
+    result = s.explode()
+    expected = pd.Series([1, 2, 3, 4], index=[0, 0, 0, 0], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_ignore_index():
+    # GH 34932
+    s = pd.Series([[1, 2], [3, 4]])
+    result = s.explode(ignore_index=True)
+    expected = pd.Series([1, 2, 3, 4], index=[0, 1, 2, 3], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_explode_sets():
+    # https://github.com/pandas-dev/pandas/issues/35614
+    s = pd.Series([{"a", "b", "c"}], index=[1])
+    result = s.explode().sort_values()
+    expected = pd.Series(["a", "b", "c"], index=[1, 1, 1])
+    tm.assert_series_equal(result, expected)
+
+
+def test_explode_scalars_can_ignore_index():
+    # https://github.com/pandas-dev/pandas/issues/40487
+    s = pd.Series([1, 2, 3], index=["a", "b", "c"])
+    result = s.explode(ignore_index=True)
+    expected = pd.Series([1, 2, 3])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ignore_index", [True, False])
+def test_explode_pyarrow_list_type(ignore_index):
+    # GH 53602
+    pa = pytest.importorskip("pyarrow")
+
+    data = [
+        [None, None],
+        [1],
+        [],
+        [2, 3],
+        None,
+    ]
+    ser = pd.Series(data, dtype=pd.ArrowDtype(pa.list_(pa.int64())))
+    result = ser.explode(ignore_index=ignore_index)
+    expected = pd.Series(
+        data=[None, None, 1, None, 2, 3, None],
+        index=None if ignore_index else [0, 0, 1, 2, 3, 3, 4],
+        dtype=pd.ArrowDtype(pa.int64()),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ignore_index", [True, False])
+def test_explode_pyarrow_non_list_type(ignore_index):
+    pa = pytest.importorskip("pyarrow")
+    data = [1, 2, 3]
+    ser = pd.Series(data, dtype=pd.ArrowDtype(pa.int64()))
+    result = ser.explode(ignore_index=ignore_index)
+    expected = pd.Series([1, 2, 3], dtype="int64[pyarrow]", index=[0, 1, 2])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_fillna.py b/pandas/pandas/tests/series/methods/test_fillna.py
new file mode 100644
index 0000000000000..f53d75df83124
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_fillna.py
@@ -0,0 +1,1033 @@
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    NaT,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    isna,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import period_array
+
+
+class TestSeriesFillNA:
+    def test_fillna_nat(self):
+        series = Series([0, 1, 2, NaT._value], dtype="M8[ns]")
+
+        filled = series.ffill()
+        filled2 = series.fillna(value=series.values[2])
+
+        expected = series.copy()
+        expected.iloc[3] = expected.iloc[2]
+
+        tm.assert_series_equal(filled, expected)
+        tm.assert_series_equal(filled2, expected)
+
+        df = DataFrame({"A": series})
+        filled = df.ffill()
+        filled2 = df.fillna(value=series.values[2])
+        expected = DataFrame({"A": expected})
+        tm.assert_frame_equal(filled, expected)
+        tm.assert_frame_equal(filled2, expected)
+
+        series = Series([NaT._value, 0, 1, 2], dtype="M8[ns]")
+
+        filled = series.bfill()
+        filled2 = series.fillna(value=series[1])
+
+        expected = series.copy()
+        expected[0] = expected[1]
+
+        tm.assert_series_equal(filled, expected)
+        tm.assert_series_equal(filled2, expected)
+
+        df = DataFrame({"A": series})
+        filled = df.bfill()
+        filled2 = df.fillna(value=series[1])
+        expected = DataFrame({"A": expected})
+        tm.assert_frame_equal(filled, expected)
+        tm.assert_frame_equal(filled2, expected)
+
+    def test_fillna(self):
+        ts = Series(
+            [0.0, 1.0, 2.0, 3.0, 4.0], index=date_range("2020-01-01", periods=5)
+        )
+
+        tm.assert_series_equal(ts, ts.ffill())
+
+        ts.iloc[2] = np.nan
+
+        exp = Series([0.0, 1.0, 1.0, 3.0, 4.0], index=ts.index)
+        tm.assert_series_equal(ts.ffill(), exp)
+
+        exp = Series([0.0, 1.0, 3.0, 3.0, 4.0], index=ts.index)
+        tm.assert_series_equal(ts.bfill(), exp)
+
+        exp = Series([0.0, 1.0, 5.0, 3.0, 4.0], index=ts.index)
+        tm.assert_series_equal(ts.fillna(value=5), exp)
+
+    def test_fillna_nonscalar(self):
+        # GH#5703
+        s1 = Series([np.nan])
+        s2 = Series([1])
+        result = s1.fillna(s2)
+        expected = Series([1.0])
+        tm.assert_series_equal(result, expected)
+        result = s1.fillna({})
+        tm.assert_series_equal(result, s1)
+        result = s1.fillna(Series((), dtype=object))
+        tm.assert_series_equal(result, s1)
+        result = s2.fillna(s1)
+        tm.assert_series_equal(result, s2)
+        result = s1.fillna({0: 1})
+        tm.assert_series_equal(result, expected)
+        result = s1.fillna({1: 1})
+        tm.assert_series_equal(result, Series([np.nan]))
+        result = s1.fillna({0: 1, 1: 1})
+        tm.assert_series_equal(result, expected)
+        result = s1.fillna(Series({0: 1, 1: 1}))
+        tm.assert_series_equal(result, expected)
+        result = s1.fillna(Series({0: 1, 1: 1}, index=[4, 5]))
+        tm.assert_series_equal(result, s1)
+
+    def test_fillna_aligns(self):
+        s1 = Series([0, 1, 2], list("abc"))
+        s2 = Series([0, np.nan, 2], list("bac"))
+        result = s2.fillna(s1)
+        expected = Series([0, 0, 2.0], list("bac"))
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_limit(self):
+        ser = Series(np.nan, index=[0, 1, 2])
+        result = ser.fillna(999, limit=1)
+        expected = Series([999, np.nan, np.nan], index=[0, 1, 2])
+        tm.assert_series_equal(result, expected)
+
+        result = ser.fillna(999, limit=2)
+        expected = Series([999, 999, np.nan], index=[0, 1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_dont_cast_strings(self):
+        # GH#9043
+        # make sure a string representation of int/float values can be filled
+        # correctly without raising errors or being converted
+        vals = ["0", "1.5", "-0.3"]
+        for val in vals:
+            ser = Series([0, 1, np.nan, np.nan, 4], dtype="float64")
+            result = ser.fillna(val)
+            expected = Series([0, 1, val, val, 4], dtype="object")
+            tm.assert_series_equal(result, expected)
+
+    def test_fillna_consistency(self):
+        # GH#16402
+        # fillna with a tz aware to a tz-naive, should result in object
+
+        ser = Series([Timestamp("20130101"), NaT])
+
+        result = ser.fillna(Timestamp("20130101", tz="US/Eastern"))
+        expected = Series(
+            [Timestamp("20130101"), Timestamp("2013-01-01", tz="US/Eastern")],
+            dtype="object",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = ser.where([True, False], Timestamp("20130101", tz="US/Eastern"))
+        tm.assert_series_equal(result, expected)
+
+        result = ser.where([True, False], Timestamp("20130101", tz="US/Eastern"))
+        tm.assert_series_equal(result, expected)
+
+        # with a non-datetime
+        result = ser.fillna("foo")
+        expected = Series([Timestamp("20130101"), "foo"])
+        tm.assert_series_equal(result, expected)
+
+        # assignment
+        ser2 = ser.copy()
+        with pytest.raises(TypeError, match="Invalid value"):
+            ser2[1] = "foo"
+
+    def test_timedelta_fillna(self, frame_or_series, unit):
+        # GH#3371
+        ser = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130102"),
+                Timestamp("20130103 9:01:01"),
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        td = ser.diff()
+        obj = frame_or_series(td).copy()
+
+        # reg fillna
+        result = obj.fillna(Timedelta(seconds=0))
+        expected = Series(
+            [
+                timedelta(0),
+                timedelta(0),
+                timedelta(1),
+                timedelta(days=1, seconds=9 * 3600 + 60 + 1),
+            ],
+            dtype=f"m8[{unit}]",
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        # GH#45746 pre-1.? ints were interpreted as seconds.  then that was
+        #  deprecated and changed to raise. In 2.0 it casts to common dtype,
+        #  consistent with every other dtype's behavior
+        res = obj.fillna(1)
+        expected = obj.astype(object).fillna(1)
+        tm.assert_equal(res, expected)
+
+        result = obj.fillna(Timedelta(seconds=1))
+        expected = Series(
+            [
+                timedelta(seconds=1),
+                timedelta(0),
+                timedelta(1),
+                timedelta(days=1, seconds=9 * 3600 + 60 + 1),
+            ],
+            dtype=f"m8[{unit}]",
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        result = obj.fillna(timedelta(days=1, seconds=1))
+        expected = Series(
+            [
+                timedelta(days=1, seconds=1),
+                timedelta(0),
+                timedelta(1),
+                timedelta(days=1, seconds=9 * 3600 + 60 + 1),
+            ],
+            dtype=f"m8[{unit}]",
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        result = obj.fillna(np.timedelta64(10**9))
+        expected = Series(
+            [
+                timedelta(seconds=1),
+                timedelta(0),
+                timedelta(1),
+                timedelta(days=1, seconds=9 * 3600 + 60 + 1),
+            ],
+            dtype=f"m8[{unit}]",
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        result = obj.fillna(NaT)
+        expected = Series(
+            [
+                NaT,
+                timedelta(0),
+                timedelta(1),
+                timedelta(days=1, seconds=9 * 3600 + 60 + 1),
+            ],
+            dtype=f"m8[{unit}]",
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        # ffill
+        td[2] = np.nan
+        obj = frame_or_series(td).copy()
+        result = obj.ffill()
+        expected = td.fillna(Timedelta(seconds=0))
+        expected[0] = np.nan
+        expected = frame_or_series(expected)
+
+        tm.assert_equal(result, expected)
+
+        # bfill
+        td[2] = np.nan
+        obj = frame_or_series(td)
+        result = obj.bfill()
+        expected = td.fillna(Timedelta(seconds=0))
+        expected[2] = timedelta(days=1, seconds=9 * 3600 + 60 + 1)
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+    def test_datetime64_fillna(self):
+        ser = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130102"),
+                Timestamp("20130103 9:01:01"),
+            ]
+        )
+        ser[2] = np.nan
+
+        # ffill
+        result = ser.ffill()
+        expected = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130103 9:01:01"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+        # bfill
+        result = ser.bfill()
+        expected = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130103 9:01:01"),
+                Timestamp("20130103 9:01:01"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            False,
+            True,
+        ],
+    )
+    @pytest.mark.parametrize("tz", [None, "UTC"])
+    def test_datetime64_fillna_mismatched_reso_no_rounding(self, tz, scalar):
+        # GH#56410
+        dti = date_range("2016-01-01", periods=3, unit="s", tz=tz)
+        item = Timestamp("2016-02-03 04:05:06.789", tz=tz)
+        vec = date_range(item, periods=3, unit="ms")
+
+        exp_dtype = "M8[ms]" if tz is None else "M8[ms, UTC]"
+        expected = Series([item, dti[1], dti[2]], dtype=exp_dtype)
+
+        ser = Series(dti)
+        ser[0] = NaT
+        ser2 = ser.copy()
+
+        res = ser.fillna(item)
+        res2 = ser2.fillna(Series(vec))
+
+        if scalar:
+            tm.assert_series_equal(res, expected)
+        else:
+            tm.assert_series_equal(res2, expected)
+
+    @pytest.mark.parametrize(
+        "scalar",
+        [
+            False,
+            True,
+        ],
+    )
+    def test_timedelta64_fillna_mismatched_reso_no_rounding(self, scalar):
+        # GH#56410
+        tdi = date_range("2016-01-01", periods=3, unit="s") - Timestamp("1970-01-01")
+        item = Timestamp("2016-02-03 04:05:06.789") - Timestamp("1970-01-01")
+        vec = timedelta_range(item, periods=3, unit="ms")
+
+        expected = Series([item, tdi[1], tdi[2]], dtype="m8[ms]")
+
+        ser = Series(tdi)
+        ser[0] = NaT
+        ser2 = ser.copy()
+
+        res = ser.fillna(item)
+        res2 = ser2.fillna(Series(vec))
+
+        if scalar:
+            tm.assert_series_equal(res, expected)
+        else:
+            tm.assert_series_equal(res2, expected)
+
+    def test_datetime64_fillna_backfill(self):
+        # GH#6587
+        # make sure that we are treating as integer when filling
+        ser = Series([NaT, NaT, "2013-08-05 15:30:00.000001"], dtype="M8[ns]")
+
+        expected = Series(
+            [
+                "2013-08-05 15:30:00.000001",
+                "2013-08-05 15:30:00.000001",
+                "2013-08-05 15:30:00.000001",
+            ],
+            dtype="M8[ns]",
+        )
+        result = ser.bfill()
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("tz", ["US/Eastern", "Asia/Tokyo"])
+    def test_datetime64_tz_fillna(self, tz, unit):
+        # DatetimeLikeBlock
+        ser = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                NaT,
+                Timestamp("2011-01-03 10:00"),
+                NaT,
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        null_loc = Series([False, True, False, True])
+
+        result = ser.fillna(Timestamp("2011-01-02 10:00"))
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00"),
+                Timestamp("2011-01-03 10:00"),
+                Timestamp("2011-01-02 10:00"),
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(expected, result)
+        # check s is not changed
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(Timestamp("2011-01-02 10:00", tz=tz))
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00", tz=tz),
+                Timestamp("2011-01-03 10:00"),
+                Timestamp("2011-01-02 10:00", tz=tz),
+            ],
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna("AAA")
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                "AAA",
+                Timestamp("2011-01-03 10:00"),
+                "AAA",
+            ],
+            dtype=object,
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(
+            {
+                1: Timestamp("2011-01-02 10:00", tz=tz),
+                3: Timestamp("2011-01-04 10:00"),
+            }
+        )
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00", tz=tz),
+                Timestamp("2011-01-03 10:00"),
+                Timestamp("2011-01-04 10:00"),
+            ]
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(
+            {1: Timestamp("2011-01-02 10:00"), 3: Timestamp("2011-01-04 10:00")}
+        )
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00"),
+                Timestamp("2011-01-02 10:00"),
+                Timestamp("2011-01-03 10:00"),
+                Timestamp("2011-01-04 10:00"),
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        # DatetimeTZBlock
+        idx = DatetimeIndex(
+            ["2011-01-01 10:00", NaT, "2011-01-03 10:00", NaT], tz=tz
+        ).as_unit(unit)
+        ser = Series(idx)
+        assert ser.dtype == f"datetime64[{unit}, {tz}]"
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(Timestamp("2011-01-02 10:00"))
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                Timestamp("2011-01-02 10:00"),
+                Timestamp("2011-01-03 10:00", tz=tz),
+                Timestamp("2011-01-02 10:00"),
+            ]
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(Timestamp("2011-01-02 10:00", tz=tz))
+        idx = DatetimeIndex(
+            [
+                "2011-01-01 10:00",
+                "2011-01-02 10:00",
+                "2011-01-03 10:00",
+                "2011-01-02 10:00",
+            ],
+            tz=tz,
+        ).as_unit(unit)
+        expected = Series(idx)
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(Timestamp("2011-01-02 10:00", tz=tz).to_pydatetime())
+        idx = DatetimeIndex(
+            [
+                "2011-01-01 10:00",
+                "2011-01-02 10:00",
+                "2011-01-03 10:00",
+                "2011-01-02 10:00",
+            ],
+            tz=tz,
+        ).as_unit(unit)
+        expected = Series(idx)
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna("AAA")
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                "AAA",
+                Timestamp("2011-01-03 10:00", tz=tz),
+                "AAA",
+            ],
+            dtype=object,
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(
+            {
+                1: Timestamp("2011-01-02 10:00", tz=tz),
+                3: Timestamp("2011-01-04 10:00"),
+            }
+        )
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                Timestamp("2011-01-02 10:00", tz=tz),
+                Timestamp("2011-01-03 10:00", tz=tz),
+                Timestamp("2011-01-04 10:00"),
+            ]
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        result = ser.fillna(
+            {
+                1: Timestamp("2011-01-02 10:00", tz=tz),
+                3: Timestamp("2011-01-04 10:00", tz=tz),
+            }
+        )
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                Timestamp("2011-01-02 10:00", tz=tz),
+                Timestamp("2011-01-03 10:00", tz=tz),
+                Timestamp("2011-01-04 10:00", tz=tz),
+            ]
+        ).dt.as_unit(unit)
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        # filling with a naive/other zone, coerce to object
+        result = ser.fillna(Timestamp("20130101"))
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                Timestamp("2013-01-01"),
+                Timestamp("2011-01-03 10:00", tz=tz),
+                Timestamp("2013-01-01"),
+            ]
+        )
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+        # pre-2.0 fillna with mixed tzs would cast to object, in 2.0
+        #  it retains dtype.
+        result = ser.fillna(Timestamp("20130101", tz="US/Pacific"))
+        expected = Series(
+            [
+                Timestamp("2011-01-01 10:00", tz=tz),
+                Timestamp("2013-01-01", tz="US/Pacific").tz_convert(tz),
+                Timestamp("2011-01-03 10:00", tz=tz),
+                Timestamp("2013-01-01", tz="US/Pacific").tz_convert(tz),
+            ]
+        ).dt.as_unit(unit)
+        tm.assert_series_equal(expected, result)
+        tm.assert_series_equal(isna(ser), null_loc)
+
+    def test_fillna_dt64tz_with_method(self):
+        # with timezone
+        # GH#15855
+        ser = Series([Timestamp("2012-11-11 00:00:00+01:00"), NaT])
+        exp = Series(
+            [
+                Timestamp("2012-11-11 00:00:00+01:00"),
+                Timestamp("2012-11-11 00:00:00+01:00"),
+            ]
+        )
+        tm.assert_series_equal(ser.ffill(), exp)
+
+        ser = Series([NaT, Timestamp("2012-11-11 00:00:00+01:00")])
+        exp = Series(
+            [
+                Timestamp("2012-11-11 00:00:00+01:00"),
+                Timestamp("2012-11-11 00:00:00+01:00"),
+            ]
+        )
+        tm.assert_series_equal(ser.bfill(), exp)
+
+    def test_fillna_pytimedelta(self):
+        # GH#8209
+        ser = Series([np.nan, Timedelta("1 days")], index=["A", "B"])
+
+        result = ser.fillna(timedelta(1))
+        expected = Series(Timedelta("1 days"), index=["A", "B"])
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_period(self):
+        # GH#13737
+        ser = Series([Period("2011-01", freq="M"), Period("NaT", freq="M")])
+
+        res = ser.fillna(Period("2012-01", freq="M"))
+        exp = Series([Period("2011-01", freq="M"), Period("2012-01", freq="M")])
+        tm.assert_series_equal(res, exp)
+        assert res.dtype == "Period[M]"
+
+    def test_fillna_dt64_timestamp(self, frame_or_series):
+        ser = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130102"),
+                Timestamp("20130103 9:01:01"),
+            ]
+        )
+        ser[2] = np.nan
+        obj = frame_or_series(ser)
+
+        # reg fillna
+        result = obj.fillna(Timestamp("20130104"))
+        expected = Series(
+            [
+                Timestamp("20130101"),
+                Timestamp("20130101"),
+                Timestamp("20130104"),
+                Timestamp("20130103 9:01:01"),
+            ]
+        )
+        expected = frame_or_series(expected)
+        tm.assert_equal(result, expected)
+
+        result = obj.fillna(NaT)
+        expected = obj
+        tm.assert_equal(result, expected)
+
+    def test_fillna_dt64_non_nao(self):
+        # GH#27419
+        ser = Series([Timestamp("2010-01-01"), NaT, Timestamp("2000-01-01")])
+        val = np.datetime64("1975-04-05", "ms")
+
+        result = ser.fillna(val)
+        expected = Series(
+            [Timestamp("2010-01-01"), Timestamp("1975-04-05"), Timestamp("2000-01-01")]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_numeric_inplace(self):
+        x = Series([np.nan, 1.0, np.nan, 3.0, np.nan], ["z", "a", "b", "c", "d"])
+        y = x.copy()
+
+        return_value = y.fillna(value=0, inplace=True)
+        assert return_value is None
+
+        expected = x.fillna(value=0)
+        tm.assert_series_equal(y, expected)
+
+    # ---------------------------------------------------------------
+    # CategoricalDtype
+
+    @pytest.mark.parametrize(
+        "fill_value, expected_output",
+        [
+            ("a", ["a", "a", "b", "a", "a"]),
+            ({1: "a", 3: "b", 4: "b"}, ["a", "a", "b", "b", "b"]),
+            ({1: "a"}, ["a", "a", "b", np.nan, np.nan]),
+            ({1: "a", 3: "b"}, ["a", "a", "b", "b", np.nan]),
+            (Series("a"), ["a", np.nan, "b", np.nan, np.nan]),
+            (Series("a", index=[1]), ["a", "a", "b", np.nan, np.nan]),
+            (Series({1: "a", 3: "b"}), ["a", "a", "b", "b", np.nan]),
+            (Series(["a", "b"], index=[3, 4]), ["a", np.nan, "b", "a", "b"]),
+        ],
+    )
+    def test_fillna_categorical(self, fill_value, expected_output):
+        # GH#17033
+        # Test fillna for a Categorical series
+        data = ["a", np.nan, "b", np.nan, np.nan]
+        ser = Series(Categorical(data, categories=["a", "b"]))
+        exp = Series(Categorical(expected_output, categories=["a", "b"]))
+        result = ser.fillna(fill_value)
+        tm.assert_series_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "fill_value, expected_output",
+        [
+            (["a", "b", "c", "d", "e"], ["a", "b", "b", "d", "e"]),
+            (["b", "d", "a", "d", "a"], ["a", "d", "b", "d", "a"]),
+            (
+                Categorical(
+                    ["b", "d", "a", "d", "a"], categories=["b", "c", "d", "e", "a"]
+                ),
+                ["a", "d", "b", "d", "a"],
+            ),
+        ],
+    )
+    def test_fillna_categorical_with_new_categories(self, fill_value, expected_output):
+        # GH#26215
+        data = ["a", np.nan, "b", np.nan, np.nan]
+        ser = Series(Categorical(data, categories=["a", "b", "c", "d", "e"]))
+        exp = Series(Categorical(expected_output, categories=["a", "b", "c", "d", "e"]))
+        fill_value = Series(fill_value)
+        result = ser.fillna(fill_value)
+        tm.assert_series_equal(result, exp)
+
+    def test_fillna_categorical_raises(self):
+        data = ["a", np.nan, "b", np.nan, np.nan]
+        ser = Series(Categorical(data, categories=["a", "b"]))
+        cat = ser._values
+
+        msg = "Cannot setitem on a Categorical with a new category"
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna("d")
+
+        msg2 = "Length of 'value' does not match."
+        with pytest.raises(ValueError, match=msg2):
+            cat.fillna(Series("d"))
+
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna({1: "d", 3: "a"})
+
+        msg = '"value" parameter must be a scalar or dict, but you passed a "list"'
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna(["a", "b"])
+
+        msg = '"value" parameter must be a scalar or dict, but you passed a "tuple"'
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna(("a", "b"))
+
+        msg = (
+            '"value" parameter must be a scalar, dict '
+            'or Series, but you passed a "DataFrame"'
+        )
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna(DataFrame({1: ["a"], 3: ["b"]}))
+
+    @pytest.mark.parametrize("dtype", [float, "float32", "float64"])
+    @pytest.mark.parametrize("scalar", [True, False])
+    def test_fillna_float_casting(self, dtype, any_real_numpy_dtype, scalar):
+        # GH-43424
+        ser = Series([np.nan, 1.2], dtype=dtype)
+        fill_values = Series([2, 2], dtype=any_real_numpy_dtype)
+        if scalar:
+            fill_values = fill_values.dtype.type(2)
+
+        result = ser.fillna(fill_values)
+        expected = Series([2.0, 1.2], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        ser = Series([np.nan, 1.2], dtype=dtype)
+        mask = ser.isna().to_numpy()
+        ser[mask] = fill_values
+        tm.assert_series_equal(ser, expected)
+
+        ser = Series([np.nan, 1.2], dtype=dtype)
+        ser.mask(mask, fill_values, inplace=True)
+        tm.assert_series_equal(ser, expected)
+
+        ser = Series([np.nan, 1.2], dtype=dtype)
+        res = ser.where(~mask, fill_values)
+        tm.assert_series_equal(res, expected)
+
+    def test_fillna_f32_upcast_with_dict(self):
+        # GH-43424
+        ser = Series([np.nan, 1.2], dtype=np.float32)
+        result = ser.fillna({0: 1})
+        expected = Series([1.0, 1.2], dtype=np.float32)
+        tm.assert_series_equal(result, expected)
+
+    # ---------------------------------------------------------------
+    # Invalid Usages
+
+    def test_fillna_listlike_invalid(self):
+        ser = Series(np.random.default_rng(2).integers(-100, 100, 50))
+        msg = '"value" parameter must be a scalar or dict, but you passed a "list"'
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna([1, 2])
+
+        msg = '"value" parameter must be a scalar or dict, but you passed a "tuple"'
+        with pytest.raises(TypeError, match=msg):
+            ser.fillna((1, 2))
+
+    def test_fillna_method_and_limit_invalid(self):
+        # related GH#9217, make sure limit is an int and greater than 0
+        ser = Series([1, 2, 3, None])
+        msg = "|".join(
+            [
+                r"Cannot specify both 'value' and 'method'\.",
+                "Limit must be greater than 0",
+                "Limit must be an integer",
+            ]
+        )
+        for limit in [-1, 0, 1.0, 2.0]:
+            with pytest.raises(ValueError, match=msg):
+                ser.fillna(1, limit=limit)
+
+    def test_fillna_datetime64_with_timezone_tzinfo(self):
+        # https://github.com/pandas-dev/pandas/issues/38851
+        # different tzinfos representing UTC treated as equal
+        ser = Series(date_range("2020", periods=3, tz="UTC"))
+        expected = ser.copy()
+        ser[1] = NaT
+        result = ser.fillna(datetime(2020, 1, 2, tzinfo=timezone.utc))
+        tm.assert_series_equal(result, expected)
+
+        # pre-2.0 we cast to object with mixed tzs, in 2.0 we retain dtype
+        ts = Timestamp("2000-01-01", tz="US/Pacific")
+        ser2 = Series(ser._values.tz_convert("dateutil/US/Pacific"))
+        assert ser2.dtype.kind == "M"
+        result = ser2.fillna(ts)
+        expected = Series(
+            [ser2[0], ts.tz_convert(ser2.dtype.tz), ser2[2]],
+            dtype=ser2.dtype,
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "input, input_fillna, expected_data, expected_categories",
+        [
+            (["A", "B", None, "A"], "B", ["A", "B", "B", "A"], ["A", "B"]),
+            (["A", "B", np.nan, "A"], "B", ["A", "B", "B", "A"], ["A", "B"]),
+        ],
+    )
+    def test_fillna_categorical_accept_same_type(
+        self, input, input_fillna, expected_data, expected_categories
+    ):
+        # GH32414
+        cat = Categorical(input)
+        ser = Series(cat).fillna(input_fillna)
+        filled = cat.fillna(ser)
+        result = cat.fillna(filled)
+        expected = Categorical(expected_data, categories=expected_categories)
+        tm.assert_categorical_equal(result, expected)
+
+
+class TestFillnaPad:
+    def test_fillna_bug(self):
+        ser = Series([np.nan, 1.0, np.nan, 3.0, np.nan], ["z", "a", "b", "c", "d"])
+        filled = ser.ffill()
+        expected = Series([np.nan, 1.0, 1.0, 3.0, 3.0], ser.index)
+        tm.assert_series_equal(filled, expected)
+
+        filled = ser.bfill()
+        expected = Series([1.0, 1.0, 3.0, 3.0, np.nan], ser.index)
+        tm.assert_series_equal(filled, expected)
+
+    def test_ffill_mixed_dtypes_without_missing_data(self):
+        # GH#14956
+        series = Series([datetime(2015, 1, 1, tzinfo=timezone.utc), 1])
+        result = series.ffill()
+        tm.assert_series_equal(series, result)
+
+    def test_pad_nan(self):
+        x = Series(
+            [np.nan, 1.0, np.nan, 3.0, np.nan], ["z", "a", "b", "c", "d"], dtype=float
+        )
+
+        return_value = x.ffill(inplace=True)
+        assert return_value is None
+
+        expected = Series(
+            [np.nan, 1.0, 1.0, 3.0, 3.0], ["z", "a", "b", "c", "d"], dtype=float
+        )
+        tm.assert_series_equal(x[1:], expected[1:])
+        assert np.isnan(x.iloc[0]), np.isnan(expected.iloc[0])
+
+    def test_series_fillna_limit(self):
+        index = np.arange(10)
+        s = Series(np.random.default_rng(2).standard_normal(10), index=index)
+
+        result = s[:2].reindex(index)
+        result = result.ffill(limit=5)
+
+        expected = s[:2].reindex(index).ffill()
+        expected[-3:] = np.nan
+        tm.assert_series_equal(result, expected)
+
+        result = s[-2:].reindex(index)
+        result = result.bfill(limit=5)
+
+        expected = s[-2:].reindex(index).bfill()
+        expected[:3] = np.nan
+        tm.assert_series_equal(result, expected)
+
+    def test_series_pad_backfill_limit(self):
+        index = np.arange(10)
+        s = Series(np.random.default_rng(2).standard_normal(10), index=index)
+
+        result = s[:2].reindex(index, method="pad", limit=5)
+
+        expected = s[:2].reindex(index).ffill()
+        expected[-3:] = np.nan
+        tm.assert_series_equal(result, expected)
+
+        result = s[-2:].reindex(index, method="backfill", limit=5)
+
+        expected = s[-2:].reindex(index).bfill()
+        expected[:3] = np.nan
+        tm.assert_series_equal(result, expected)
+
+    def test_fillna_int(self):
+        ser = Series(np.random.default_rng(2).integers(-100, 100, 50))
+        return_value = ser.ffill(inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(ser.ffill(inplace=False), ser)
+
+    def test_datetime64tz_fillna_round_issue(self):
+        # GH#14872
+
+        data = Series(
+            [NaT, NaT, datetime(2016, 12, 12, 22, 24, 6, 100001, tzinfo=timezone.utc)]
+        )
+
+        filled = data.bfill()
+
+        expected = Series(
+            [
+                datetime(2016, 12, 12, 22, 24, 6, 100001, tzinfo=timezone.utc),
+                datetime(2016, 12, 12, 22, 24, 6, 100001, tzinfo=timezone.utc),
+                datetime(2016, 12, 12, 22, 24, 6, 100001, tzinfo=timezone.utc),
+            ]
+        )
+
+        tm.assert_series_equal(filled, expected)
+
+    def test_fillna_parr(self):
+        # GH-24537
+        dti = date_range(
+            Timestamp.max - Timedelta(nanoseconds=10), periods=5, freq="ns"
+        )
+        ser = Series(dti.to_period("ns"))
+        ser[2] = NaT
+        arr = period_array(
+            [
+                Timestamp("2262-04-11 23:47:16.854775797"),
+                Timestamp("2262-04-11 23:47:16.854775798"),
+                Timestamp("2262-04-11 23:47:16.854775798"),
+                Timestamp("2262-04-11 23:47:16.854775800"),
+                Timestamp("2262-04-11 23:47:16.854775801"),
+            ],
+            freq="ns",
+        )
+        expected = Series(arr)
+
+        filled = ser.ffill()
+
+        tm.assert_series_equal(filled, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected_data, method, kwargs",
+    (
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 3.0, 3.0, 3.0, 7.0, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "inside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 3.0, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "inside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, 7.0],
+            "ffill",
+            {"limit_area": "outside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, np.nan],
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+            [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            range(5),
+            range(5),
+            "ffill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, 7.0, 7.0, 7.0, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "inside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, np.nan, 3.0, np.nan, np.nan, 7.0, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "inside", "limit": 1},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [3.0, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "outside"},
+        ),
+        (
+            [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+            [np.nan, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, np.nan, np.nan],
+            "bfill",
+            {"limit_area": "outside", "limit": 1},
+        ),
+    ),
+)
+def test_ffill_bfill_limit_area(data, expected_data, method, kwargs):
+    # GH#56492
+    s = Series(data)
+    expected = Series(expected_data)
+    result = getattr(s, method)(**kwargs)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_get_numeric_data.py b/pandas/pandas/tests/series/methods/test_get_numeric_data.py
new file mode 100644
index 0000000000000..4a11d7905f506
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_get_numeric_data.py
@@ -0,0 +1,31 @@
+from pandas import (
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestGetNumericData:
+    def test_get_numeric_data_preserve_dtype(self):
+        # get the numeric data
+        obj = Series([1, 2, 3])
+        result = obj._get_numeric_data()
+        tm.assert_series_equal(result, obj)
+
+        # returned object is a shallow copy
+        result.iloc[0] = 0
+        assert obj.iloc[0] == 1
+
+        obj = Series([1, "2", 3.0])
+        result = obj._get_numeric_data()
+        expected = Series([], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        obj = Series([True, False, True])
+        result = obj._get_numeric_data()
+        tm.assert_series_equal(result, obj)
+
+        obj = Series(date_range("20130101", periods=3))
+        result = obj._get_numeric_data()
+        expected = Series([], dtype="M8[ns]")
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_head_tail.py b/pandas/pandas/tests/series/methods/test_head_tail.py
new file mode 100644
index 0000000000000..d9f8d85eda350
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_head_tail.py
@@ -0,0 +1,8 @@
+import pandas._testing as tm
+
+
+def test_head_tail(string_series):
+    tm.assert_series_equal(string_series.head(), string_series[:5])
+    tm.assert_series_equal(string_series.head(0), string_series[0:0])
+    tm.assert_series_equal(string_series.tail(), string_series[-5:])
+    tm.assert_series_equal(string_series.tail(0), string_series[0:0])
diff --git a/pandas/pandas/tests/series/methods/test_infer_objects.py b/pandas/pandas/tests/series/methods/test_infer_objects.py
new file mode 100644
index 0000000000000..0d411e08a0107
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_infer_objects.py
@@ -0,0 +1,56 @@
+import numpy as np
+
+from pandas import (
+    Series,
+    interval_range,
+)
+import pandas._testing as tm
+
+
+class TestInferObjects:
+    def test_copy(self, index_or_series):
+        # GH#50096
+        # case where we don't need to do inference because it is already non-object
+        obj = index_or_series(np.array([1, 2, 3], dtype="int64"))
+
+        result = obj.infer_objects()
+        assert tm.shares_memory(result, obj)
+
+        # case where we try to do inference but can't do better than object
+        obj2 = index_or_series(np.array(["foo", 2], dtype=object))
+        result2 = obj2.infer_objects()
+        assert tm.shares_memory(result2, obj2)
+
+    def test_infer_objects_series(self, index_or_series):
+        # GH#11221
+        actual = index_or_series(np.array([1, 2, 3], dtype="O")).infer_objects()
+        expected = index_or_series([1, 2, 3])
+        tm.assert_equal(actual, expected)
+
+        actual = index_or_series(np.array([1, 2, 3, None], dtype="O")).infer_objects()
+        expected = index_or_series([1.0, 2.0, 3.0, np.nan])
+        tm.assert_equal(actual, expected)
+
+        # only soft conversions, unconvertible pass thru unchanged
+
+        obj = index_or_series(np.array([1, 2, 3, None, "a"], dtype="O"))
+        actual = obj.infer_objects()
+        expected = index_or_series([1, 2, 3, None, "a"], dtype=object)
+
+        assert actual.dtype == "object"
+        tm.assert_equal(actual, expected)
+
+    def test_infer_objects_interval(self, index_or_series):
+        # GH#50090
+        ii = interval_range(1, 10)
+        obj = index_or_series(ii)
+
+        result = obj.astype(object).infer_objects()
+        tm.assert_equal(result, obj)
+
+    def test_infer_objects_bytes(self):
+        # GH#49650
+        ser = Series([b"a"], dtype="bytes")
+        expected = ser.copy()
+        result = ser.infer_objects()
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_info.py b/pandas/pandas/tests/series/methods/test_info.py
new file mode 100644
index 0000000000000..88f2cf384fc79
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_info.py
@@ -0,0 +1,192 @@
+from io import StringIO
+from string import ascii_uppercase
+import textwrap
+
+import numpy as np
+import pytest
+
+from pandas._config import using_string_dtype
+
+from pandas.compat import (
+    HAS_PYARROW,
+    PYPY,
+)
+
+from pandas import (
+    CategoricalIndex,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+)
+
+
+def test_info_categorical_column_just_works():
+    n = 2500
+    data = np.array(list("abcdefghij")).take(
+        np.random.default_rng(2).integers(0, 10, size=n, dtype=int)
+    )
+    s = Series(data).astype("category")
+    s.isna()
+    buf = StringIO()
+    s.info(buf=buf)
+
+    s2 = s[s == "d"]
+    buf = StringIO()
+    s2.info(buf=buf)
+
+
+def test_info_categorical():
+    # GH14298
+    idx = CategoricalIndex(["a", "b"])
+    s = Series(np.zeros(2), index=idx)
+    buf = StringIO()
+    s.info(buf=buf)
+
+
+@pytest.mark.parametrize("verbose", [True, False])
+def test_info_series(
+    lexsorted_two_level_string_multiindex, verbose, using_infer_string
+):
+    index = lexsorted_two_level_string_multiindex
+    ser = Series(range(len(index)), index=index, name="sth")
+    buf = StringIO()
+    ser.info(verbose=verbose, buf=buf)
+    result = buf.getvalue()
+
+    expected = textwrap.dedent(
+        """\
+        <class 'pandas.Series'>
+        MultiIndex: 10 entries, ('foo', 'one') to ('qux', 'three')
+        """
+    )
+    if verbose:
+        expected += textwrap.dedent(
+            """\
+            Series name: sth
+            Non-Null Count  Dtype
+            --------------  -----
+            10 non-null     int64
+            """
+        )
+    qualifier = "" if using_infer_string and HAS_PYARROW else "+"
+    expected += textwrap.dedent(
+        f"""\
+        dtypes: int64(1)
+        memory usage: {ser.memory_usage()}.0{qualifier} bytes
+        """
+    )
+    assert result == expected
+
+
+def test_info_memory():
+    s = Series([1, 2], dtype="i8")
+    buf = StringIO()
+    s.info(buf=buf)
+    result = buf.getvalue()
+    memory_bytes = float(s.memory_usage())
+    expected = textwrap.dedent(
+        f"""\
+    <class 'pandas.Series'>
+    RangeIndex: 2 entries, 0 to 1
+    Series name: None
+    Non-Null Count  Dtype
+    --------------  -----
+    2 non-null      int64
+    dtypes: int64(1)
+    memory usage: {memory_bytes} bytes
+    """
+    )
+    assert result == expected
+
+
+def test_info_wide():
+    s = Series(np.random.default_rng(2).standard_normal(101))
+    msg = "Argument `max_cols` can only be passed in DataFrame.info, not Series.info"
+    with pytest.raises(ValueError, match=msg):
+        s.info(max_cols=1)
+
+
+def test_info_shows_dtypes():
+    dtypes = [
+        "int64",
+        "float64",
+        "datetime64[ns]",
+        "timedelta64[ns]",
+        "complex128",
+        "object",
+        "bool",
+    ]
+    n = 10
+    for dtype in dtypes:
+        s = Series(np.random.default_rng(2).integers(2, size=n).astype(dtype))
+        buf = StringIO()
+        s.info(buf=buf)
+        res = buf.getvalue()
+        name = f"{n:d} non-null     {dtype}"
+        assert name in res
+
+
+@pytest.mark.xfail(PYPY, reason="on PyPy deep=True doesn't change result")
+def test_info_memory_usage_deep_not_pypy():
+    s_with_object_index = Series({"a": [1]}, index=["foo"])
+    assert s_with_object_index.memory_usage(
+        index=True, deep=True
+    ) > s_with_object_index.memory_usage(index=True)
+
+    s_object = Series({"a": ["a"]})
+    assert s_object.memory_usage(deep=True) > s_object.memory_usage()
+
+
+@pytest.mark.xfail(not PYPY, reason="on PyPy deep=True does not change result")
+def test_info_memory_usage_deep_pypy():
+    s_with_object_index = Series({"a": [1]}, index=["foo"])
+    assert s_with_object_index.memory_usage(
+        index=True, deep=True
+    ) == s_with_object_index.memory_usage(index=True)
+
+    s_object = Series({"a": ["a"]})
+    assert s_object.memory_usage(deep=True) == s_object.memory_usage()
+
+
+@pytest.mark.parametrize(
+    "index, plus",
+    [
+        ([1, 2, 3], False),
+        (Index(list("ABC"), dtype="str"), not (using_string_dtype() and HAS_PYARROW)),
+        (Index(list("ABC"), dtype=object), True),
+        (MultiIndex.from_product([range(3), range(3)]), False),
+        (
+            MultiIndex.from_product([range(3), ["foo", "bar"]]),
+            not (using_string_dtype() and HAS_PYARROW),
+        ),
+    ],
+)
+def test_info_memory_usage_qualified(index, plus):
+    buf = StringIO()
+    series = Series(1, index=index)
+    series.info(buf=buf)
+    if plus:
+        assert "+" in buf.getvalue()
+    else:
+        assert "+" not in buf.getvalue()
+
+
+def test_info_memory_usage_bug_on_multiindex():
+    # GH 14308
+    # memory usage introspection should not materialize .values
+    N = 100
+    M = len(ascii_uppercase)
+    index = MultiIndex.from_product(
+        [list(ascii_uppercase), date_range("20160101", periods=N)],
+        names=["id", "date"],
+    )
+    s = Series(np.random.default_rng(2).standard_normal(N * M), index=index)
+
+    unstacked = s.unstack("id")
+    assert s.values.nbytes == unstacked.values.nbytes
+    assert s.memory_usage(deep=True) > unstacked.memory_usage(deep=True).sum()
+
+    # high upper bound
+    diff = unstacked.memory_usage(deep=True).sum() - s.memory_usage(deep=True)
+    assert diff < 2000
diff --git a/pandas/pandas/tests/series/methods/test_interpolate.py b/pandas/pandas/tests/series/methods/test_interpolate.py
new file mode 100644
index 0000000000000..ff7f8d0b7fa72
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_interpolate.py
@@ -0,0 +1,808 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(
+    params=[
+        "linear",
+        "index",
+        "values",
+        "nearest",
+        "slinear",
+        "zero",
+        "quadratic",
+        "cubic",
+        "barycentric",
+        "krogh",
+        "polynomial",
+        "spline",
+        "piecewise_polynomial",
+        "from_derivatives",
+        "pchip",
+        "akima",
+        "cubicspline",
+    ]
+)
+def nontemporal_method(request):
+    """Fixture that returns an (method name, required kwargs) pair.
+
+    This fixture does not include method 'time' as a parameterization; that
+    method requires a Series with a DatetimeIndex, and is generally tested
+    separately from these non-temporal methods.
+    """
+    method = request.param
+    kwargs = {"order": 1} if method in ("spline", "polynomial") else {}
+    return method, kwargs
+
+
+@pytest.fixture(
+    params=[
+        "linear",
+        "slinear",
+        "zero",
+        "quadratic",
+        "cubic",
+        "barycentric",
+        "krogh",
+        "polynomial",
+        "spline",
+        "piecewise_polynomial",
+        "from_derivatives",
+        "pchip",
+        "akima",
+        "cubicspline",
+    ]
+)
+def interp_methods_ind(request):
+    """Fixture that returns a (method name, required kwargs) pair to
+    be tested for various Index types.
+
+    This fixture does not include methods - 'time', 'index', 'nearest',
+    'values' as a parameterization
+    """
+    method = request.param
+    kwargs = {"order": 1} if method in ("spline", "polynomial") else {}
+    return method, kwargs
+
+
+class TestSeriesInterpolateData:
+    @pytest.mark.xfail(reason="EA.fillna does not handle 'linear' method")
+    def test_interpolate_period_values(self):
+        orig = Series(date_range("2012-01-01", periods=5))
+        ser = orig.copy()
+        ser[2] = pd.NaT
+
+        # period cast
+        ser_per = ser.dt.to_period("D")
+        res_per = ser_per.interpolate()
+        expected_per = orig.dt.to_period("D")
+        tm.assert_series_equal(res_per, expected_per)
+
+    def test_interpolate(self, datetime_series):
+        ts = Series(np.arange(len(datetime_series), dtype=float), datetime_series.index)
+
+        ts_copy = ts.copy()
+
+        # Set data between Tuesday and Thursday to NaN for 2 consecutive weeks.
+        # Linear interpolation should fill in the missing values correctly,
+        # as the index is equally-spaced within each week.
+        ts_copy[1:4] = np.nan
+        ts_copy[6:9] = np.nan
+
+        linear_interp = ts_copy.interpolate(method="linear")
+        tm.assert_series_equal(linear_interp, ts)
+
+        ord_ts = Series(
+            [d.toordinal() for d in datetime_series.index], index=datetime_series.index
+        ).astype(float)
+
+        ord_ts_copy = ord_ts.copy()
+        ord_ts_copy[5:10] = np.nan
+
+        time_interp = ord_ts_copy.interpolate(method="time")
+        tm.assert_series_equal(time_interp, ord_ts)
+
+    def test_interpolate_time_raises_for_non_timeseries(self):
+        # When method='time' is used on a non-TimeSeries that contains a null
+        # value, a ValueError should be raised.
+        non_ts = Series([0, 1, 2, np.nan])
+        msg = "time-weighted interpolation only works on Series.* with a DatetimeIndex"
+        with pytest.raises(ValueError, match=msg):
+            non_ts.interpolate(method="time")
+
+    def test_interpolate_cubicspline(self):
+        pytest.importorskip("scipy")
+        ser = Series([10, 11, 12, 13])
+
+        expected = Series(
+            [11.00, 11.25, 11.50, 11.75, 12.00, 12.25, 12.50, 12.75, 13.00],
+            index=Index([1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5, 2.75, 3.0]),
+        )
+        # interpolate at new_index
+        new_index = ser.index.union(Index([1.25, 1.5, 1.75, 2.25, 2.5, 2.75])).astype(
+            float
+        )
+        result = ser.reindex(new_index).interpolate(method="cubicspline").loc[1:3]
+        tm.assert_series_equal(result, expected)
+
+    def test_interpolate_pchip(self):
+        pytest.importorskip("scipy")
+        ser = Series(np.sort(np.random.default_rng(2).uniform(size=100)))
+
+        # interpolate at new_index
+        new_index = ser.index.union(
+            Index([49.25, 49.5, 49.75, 50.25, 50.5, 50.75])
+        ).astype(float)
+        interp_s = ser.reindex(new_index).interpolate(method="pchip")
+        # does not blow up, GH5977
+        interp_s.loc[49:51]
+
+    def test_interpolate_akima(self):
+        pytest.importorskip("scipy")
+        ser = Series([10, 11, 12, 13])
+
+        # interpolate at new_index where `der` is zero
+        expected = Series(
+            [11.00, 11.25, 11.50, 11.75, 12.00, 12.25, 12.50, 12.75, 13.00],
+            index=Index([1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5, 2.75, 3.0]),
+        )
+        new_index = ser.index.union(Index([1.25, 1.5, 1.75, 2.25, 2.5, 2.75])).astype(
+            float
+        )
+        interp_s = ser.reindex(new_index).interpolate(method="akima")
+        tm.assert_series_equal(interp_s.loc[1:3], expected)
+
+        # interpolate at new_index where `der` is a non-zero int
+        expected = Series(
+            [11.0, 1.0, 1.0, 1.0, 12.0, 1.0, 1.0, 1.0, 13.0],
+            index=Index([1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5, 2.75, 3.0]),
+        )
+        new_index = ser.index.union(Index([1.25, 1.5, 1.75, 2.25, 2.5, 2.75])).astype(
+            float
+        )
+        interp_s = ser.reindex(new_index).interpolate(method="akima", der=1)
+        tm.assert_series_equal(interp_s.loc[1:3], expected)
+
+    def test_interpolate_piecewise_polynomial(self):
+        pytest.importorskip("scipy")
+        ser = Series([10, 11, 12, 13])
+
+        expected = Series(
+            [11.00, 11.25, 11.50, 11.75, 12.00, 12.25, 12.50, 12.75, 13.00],
+            index=Index([1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5, 2.75, 3.0]),
+        )
+        # interpolate at new_index
+        new_index = ser.index.union(Index([1.25, 1.5, 1.75, 2.25, 2.5, 2.75])).astype(
+            float
+        )
+        interp_s = ser.reindex(new_index).interpolate(method="piecewise_polynomial")
+        tm.assert_series_equal(interp_s.loc[1:3], expected)
+
+    def test_interpolate_from_derivatives(self):
+        pytest.importorskip("scipy")
+        ser = Series([10, 11, 12, 13])
+
+        expected = Series(
+            [11.00, 11.25, 11.50, 11.75, 12.00, 12.25, 12.50, 12.75, 13.00],
+            index=Index([1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5, 2.75, 3.0]),
+        )
+        # interpolate at new_index
+        new_index = ser.index.union(Index([1.25, 1.5, 1.75, 2.25, 2.5, 2.75])).astype(
+            float
+        )
+        interp_s = ser.reindex(new_index).interpolate(method="from_derivatives")
+        tm.assert_series_equal(interp_s.loc[1:3], expected)
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {},
+            pytest.param(
+                {"method": "polynomial", "order": 1}, marks=td.skip_if_no("scipy")
+            ),
+        ],
+    )
+    def test_interpolate_corners(self, kwargs):
+        s = Series([np.nan, np.nan])
+        tm.assert_series_equal(s.interpolate(**kwargs), s)
+
+        s = Series([], dtype=object).interpolate()
+        tm.assert_series_equal(s.interpolate(**kwargs), s)
+
+    def test_interpolate_index_values(self):
+        s = Series(np.nan, index=np.sort(np.random.default_rng(2).random(30)))
+        s.loc[::3] = np.random.default_rng(2).standard_normal(10)
+
+        vals = s.index.values.astype(float)
+
+        result = s.interpolate(method="index")
+
+        bad = isna(s)
+        good = ~bad
+        expected = Series(
+            np.interp(vals[bad], vals[good], s.values[good]), index=s.index[bad]
+        )
+
+        tm.assert_series_equal(result[bad], expected)
+
+        # 'values' is synonymous with 'index' for the method kwarg
+        other_result = s.interpolate(method="values")
+
+        tm.assert_series_equal(other_result, result)
+        tm.assert_series_equal(other_result[bad], expected)
+
+    def test_interpolate_non_ts(self):
+        s = Series([1, 3, np.nan, np.nan, np.nan, 11])
+        msg = (
+            "time-weighted interpolation only works on Series or DataFrames "
+            "with a DatetimeIndex"
+        )
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="time")
+
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {},
+            pytest.param(
+                {"method": "polynomial", "order": 1}, marks=td.skip_if_no("scipy")
+            ),
+        ],
+    )
+    def test_nan_interpolate(self, kwargs):
+        s = Series([0, 1, np.nan, 3])
+        result = s.interpolate(**kwargs)
+        expected = Series([0.0, 1.0, 2.0, 3.0])
+        tm.assert_series_equal(result, expected)
+
+    def test_nan_irregular_index(self):
+        s = Series([1, 2, np.nan, 4], index=[1, 3, 5, 9])
+        result = s.interpolate()
+        expected = Series([1.0, 2.0, 2.6666666666666665, 4.0], index=[1, 3, 5, 9])
+        tm.assert_series_equal(result, expected)
+
+    def test_nan_str_index(self):
+        s = Series([0, 1, 2, np.nan], index=list("abcd"))
+        result = s.interpolate()
+        expected = Series([0.0, 1.0, 2.0, 2.0], index=list("abcd"))
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_quad(self):
+        pytest.importorskip("scipy")
+        sq = Series([1, 4, np.nan, 16], index=[1, 2, 3, 4])
+        result = sq.interpolate(method="quadratic")
+        expected = Series([1.0, 4.0, 9.0, 16.0], index=[1, 2, 3, 4])
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_scipy_basic(self):
+        pytest.importorskip("scipy")
+        s = Series([1, 3, np.nan, 12, np.nan, 25])
+        # slinear
+        expected = Series([1.0, 3.0, 7.5, 12.0, 18.5, 25.0])
+        result = s.interpolate(method="slinear")
+        tm.assert_series_equal(result, expected)
+        result = s.interpolate(method="slinear")
+        tm.assert_series_equal(result, expected)
+        # nearest
+        expected = Series([1, 3, 3, 12, 12, 25.0])
+        result = s.interpolate(method="nearest")
+        tm.assert_series_equal(result, expected.astype("float"))
+
+        result = s.interpolate(method="nearest")
+        tm.assert_series_equal(result, expected)
+        # zero
+        expected = Series([1, 3, 3, 12, 12, 25.0])
+        result = s.interpolate(method="zero")
+        tm.assert_series_equal(result, expected.astype("float"))
+
+        result = s.interpolate(method="zero")
+        tm.assert_series_equal(result, expected)
+        # quadratic
+        # GH #15662.
+        expected = Series([1, 3.0, 6.823529, 12.0, 18.058824, 25.0])
+        result = s.interpolate(method="quadratic")
+        tm.assert_series_equal(result, expected)
+
+        result = s.interpolate(method="quadratic")
+        tm.assert_series_equal(result, expected)
+        # cubic
+        expected = Series([1.0, 3.0, 6.8, 12.0, 18.2, 25.0])
+        result = s.interpolate(method="cubic")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_limit(self):
+        s = Series([1, 3, np.nan, np.nan, np.nan, 11])
+
+        expected = Series([1.0, 3.0, 5.0, 7.0, np.nan, 11.0])
+        result = s.interpolate(method="linear", limit=2)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("limit", [-1, 0])
+    def test_interpolate_invalid_nonpositive_limit(self, nontemporal_method, limit):
+        # GH 9217: make sure limit is greater than zero.
+        s = Series([1, 2, np.nan, 4])
+        method, kwargs = nontemporal_method
+        with pytest.raises(ValueError, match="Limit must be greater than 0"):
+            s.interpolate(limit=limit, method=method, **kwargs)
+
+    def test_interpolate_invalid_float_limit(self, nontemporal_method):
+        # GH 9217: make sure limit is an integer.
+        s = Series([1, 2, np.nan, 4])
+        method, kwargs = nontemporal_method
+        limit = 2.0
+        with pytest.raises(ValueError, match="Limit must be an integer"):
+            s.interpolate(limit=limit, method=method, **kwargs)
+
+    @pytest.mark.parametrize("invalid_method", [None, "nonexistent_method"])
+    def test_interp_invalid_method(self, invalid_method):
+        s = Series([1, 3, np.nan, 12, np.nan, 25])
+
+        msg = "Can not interpolate with method=nonexistent_method"
+        if invalid_method is None:
+            msg = "'method' should be a string, not None"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method=invalid_method)
+
+        # When an invalid method and invalid limit (such as -1) are
+        # provided, the error message reflects the invalid method.
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method=invalid_method, limit=-1)
+
+    def test_interp_limit_forward(self):
+        s = Series([1, 3, np.nan, np.nan, np.nan, 11])
+
+        # Provide 'forward' (the default) explicitly here.
+        expected = Series([1.0, 3.0, 5.0, 7.0, np.nan, 11.0])
+
+        result = s.interpolate(method="linear", limit=2, limit_direction="forward")
+        tm.assert_series_equal(result, expected)
+
+        result = s.interpolate(method="linear", limit=2, limit_direction="FORWARD")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_unlimited(self):
+        # these test are for issue #16282 default Limit=None is unlimited
+        s = Series([np.nan, 1.0, 3.0, np.nan, np.nan, np.nan, 11.0, np.nan])
+        expected = Series([1.0, 1.0, 3.0, 5.0, 7.0, 9.0, 11.0, 11.0])
+        result = s.interpolate(method="linear", limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, 1.0, 3.0, 5.0, 7.0, 9.0, 11.0, 11.0])
+        result = s.interpolate(method="linear", limit_direction="forward")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([1.0, 1.0, 3.0, 5.0, 7.0, 9.0, 11.0, np.nan])
+        result = s.interpolate(method="linear", limit_direction="backward")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_limit_bad_direction(self):
+        s = Series([1, 3, np.nan, np.nan, np.nan, 11])
+
+        msg = (
+            r"Invalid limit_direction: expecting one of \['forward', "
+            r"'backward', 'both'\], got 'abc'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="linear", limit=2, limit_direction="abc")
+
+        # raises an error even if no limit is specified.
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="linear", limit_direction="abc")
+
+    # limit_area introduced GH #16284
+    def test_interp_limit_area(self):
+        # These tests are for issue #9218 -- fill NaNs in both directions.
+        s = Series([np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan])
+
+        expected = Series([np.nan, np.nan, 3.0, 4.0, 5.0, 6.0, 7.0, np.nan, np.nan])
+        result = s.interpolate(method="linear", limit_area="inside")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(
+            [np.nan, np.nan, 3.0, 4.0, np.nan, np.nan, 7.0, np.nan, np.nan]
+        )
+        result = s.interpolate(method="linear", limit_area="inside", limit=1)
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, np.nan, 3.0, 4.0, np.nan, 6.0, 7.0, np.nan, np.nan])
+        result = s.interpolate(
+            method="linear", limit_area="inside", limit_direction="both", limit=1
+        )
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, 7.0])
+        result = s.interpolate(method="linear", limit_area="outside")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(
+            [np.nan, np.nan, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, np.nan]
+        )
+        result = s.interpolate(method="linear", limit_area="outside", limit=1)
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, 7.0, np.nan])
+        result = s.interpolate(
+            method="linear", limit_area="outside", limit_direction="both", limit=1
+        )
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([3.0, 3.0, 3.0, np.nan, np.nan, np.nan, 7.0, np.nan, np.nan])
+        result = s.interpolate(
+            method="linear", limit_area="outside", limit_direction="backward"
+        )
+        tm.assert_series_equal(result, expected)
+
+        # raises an error even if limit type is wrong.
+        msg = r"Invalid limit_area: expecting one of \['inside', 'outside'\], got abc"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="linear", limit_area="abc")
+
+    @pytest.mark.parametrize(
+        "data, kwargs",
+        (
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "pad", "limit_area": "inside"},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "pad", "limit_area": "inside", "limit": 1},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "pad", "limit_area": "outside"},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "pad", "limit_area": "outside", "limit": 1},
+            ),
+            (
+                [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan],
+                {"method": "pad", "limit_area": "outside", "limit": 1},
+            ),
+            (
+                range(5),
+                {"method": "pad", "limit_area": "outside", "limit": 1},
+            ),
+        ),
+    )
+    def test_interp_limit_area_with_pad(self, data, kwargs):
+        # GH26796
+
+        s = Series(data)
+        msg = "Can not interpolate with method=pad"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(**kwargs)
+
+    @pytest.mark.parametrize(
+        "data, kwargs",
+        (
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "bfill", "limit_area": "inside"},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "bfill", "limit_area": "inside", "limit": 1},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "bfill", "limit_area": "outside"},
+            ),
+            (
+                [np.nan, np.nan, 3, np.nan, np.nan, np.nan, 7, np.nan, np.nan],
+                {"method": "bfill", "limit_area": "outside", "limit": 1},
+            ),
+        ),
+    )
+    def test_interp_limit_area_with_backfill(self, data, kwargs):
+        # GH26796
+        s = Series(data)
+
+        msg = "Can not interpolate with method=bfill"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(**kwargs)
+
+    def test_interp_limit_direction(self):
+        # These tests are for issue #9218 -- fill NaNs in both directions.
+        s = Series([1, 3, np.nan, np.nan, np.nan, 11])
+
+        expected = Series([1.0, 3.0, np.nan, 7.0, 9.0, 11.0])
+        result = s.interpolate(method="linear", limit=2, limit_direction="backward")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([1.0, 3.0, 5.0, np.nan, 9.0, 11.0])
+        result = s.interpolate(method="linear", limit=1, limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+        # Check that this works on a longer series of nans.
+        s = Series([1, 3, np.nan, np.nan, np.nan, 7, 9, np.nan, np.nan, 12, np.nan])
+
+        expected = Series([1.0, 3.0, 4.0, 5.0, 6.0, 7.0, 9.0, 10.0, 11.0, 12.0, 12.0])
+        result = s.interpolate(method="linear", limit=2, limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(
+            [1.0, 3.0, 4.0, np.nan, 6.0, 7.0, 9.0, 10.0, 11.0, 12.0, 12.0]
+        )
+        result = s.interpolate(method="linear", limit=1, limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_limit_to_ends(self):
+        # These test are for issue #10420 -- flow back to beginning.
+        s = Series([np.nan, np.nan, 5, 7, 9, np.nan])
+
+        expected = Series([5.0, 5.0, 5.0, 7.0, 9.0, np.nan])
+        result = s.interpolate(method="linear", limit=2, limit_direction="backward")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([5.0, 5.0, 5.0, 7.0, 9.0, 9.0])
+        result = s.interpolate(method="linear", limit=2, limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_limit_before_ends(self):
+        # These test are for issue #11115 -- limit ends properly.
+        s = Series([np.nan, np.nan, 5, 7, np.nan, np.nan])
+
+        expected = Series([np.nan, np.nan, 5.0, 7.0, 7.0, np.nan])
+        result = s.interpolate(method="linear", limit=1, limit_direction="forward")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, 5.0, 5.0, 7.0, np.nan, np.nan])
+        result = s.interpolate(method="linear", limit=1, limit_direction="backward")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([np.nan, 5.0, 5.0, 7.0, 7.0, np.nan])
+        result = s.interpolate(method="linear", limit=1, limit_direction="both")
+        tm.assert_series_equal(result, expected)
+
+    def test_interp_all_good(self):
+        pytest.importorskip("scipy")
+        s = Series([1, 2, 3])
+        result = s.interpolate(method="polynomial", order=1)
+        tm.assert_series_equal(result, s)
+
+        # non-scipy
+        result = s.interpolate()
+        tm.assert_series_equal(result, s)
+
+    @pytest.mark.parametrize(
+        "check_scipy", [False, pytest.param(True, marks=td.skip_if_no("scipy"))]
+    )
+    def test_interp_multiIndex(self, check_scipy):
+        idx = MultiIndex.from_tuples([(0, "a"), (1, "b"), (2, "c")])
+        s = Series([1, 2, np.nan], index=idx)
+
+        expected = s.copy()
+        expected.loc[2] = 2
+        result = s.interpolate()
+        tm.assert_series_equal(result, expected)
+
+        msg = "Only `method=linear` interpolation is supported on MultiIndexes"
+        if check_scipy:
+            with pytest.raises(ValueError, match=msg):
+                s.interpolate(method="polynomial", order=1)
+
+    def test_interp_nonmono_raise(self):
+        pytest.importorskip("scipy")
+        s = Series([1, np.nan, 3], index=[0, 2, 1])
+        msg = "krogh interpolation requires that the index be monotonic"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="krogh")
+
+    @pytest.mark.parametrize("method", ["nearest", "pad"])
+    def test_interp_datetime64(self, method, tz_naive_fixture):
+        pytest.importorskip("scipy")
+        df = Series(
+            [1, np.nan, 3], index=date_range("1/1/2000", periods=3, tz=tz_naive_fixture)
+        )
+
+        if method == "nearest":
+            result = df.interpolate(method=method)
+            expected = Series(
+                [1.0, 1.0, 3.0],
+                index=date_range("1/1/2000", periods=3, tz=tz_naive_fixture),
+            )
+            tm.assert_series_equal(result, expected)
+        else:
+            msg = "Can not interpolate with method=pad"
+            with pytest.raises(ValueError, match=msg):
+                df.interpolate(method=method)
+
+    def test_interp_pad_datetime64tz_values(self):
+        # GH#27628 missing.interpolate_2d should handle datetimetz values
+        dti = date_range("2015-04-05", periods=3, tz="US/Central")
+        ser = Series(dti)
+        ser[1] = pd.NaT
+
+        msg = "Can not interpolate with method=pad"
+        with pytest.raises(ValueError, match=msg):
+            ser.interpolate(method="pad")
+
+    def test_interp_limit_no_nans(self):
+        # GH 7173
+        s = Series([1.0, 2.0, 3.0])
+        result = s.interpolate(limit=1)
+        expected = s
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("method", ["polynomial", "spline"])
+    def test_no_order(self, method):
+        # see GH-10633, GH-24014
+        pytest.importorskip("scipy")
+        s = Series([0, 1, np.nan, 3])
+        msg = "You must specify the order of the spline or polynomial"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method=method)
+
+    @pytest.mark.parametrize("order", [-1, -1.0, 0, 0.0, np.nan])
+    def test_interpolate_spline_invalid_order(self, order):
+        pytest.importorskip("scipy")
+        s = Series([0, 1, np.nan, 3])
+        msg = "order needs to be specified and greater than 0"
+        with pytest.raises(ValueError, match=msg):
+            s.interpolate(method="spline", order=order)
+
+    def test_spline(self):
+        pytest.importorskip("scipy")
+        s = Series([1, 2, np.nan, 4, 5, np.nan, 7])
+        result = s.interpolate(method="spline", order=1)
+        expected = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0])
+        tm.assert_series_equal(result, expected)
+
+    def test_spline_extrapolate(self):
+        pytest.importorskip("scipy")
+        s = Series([1, 2, 3, 4, np.nan, 6, np.nan])
+        result3 = s.interpolate(method="spline", order=1, ext=3)
+        expected3 = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 6.0])
+        tm.assert_series_equal(result3, expected3)
+
+        result1 = s.interpolate(method="spline", order=1, ext=0)
+        expected1 = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0])
+        tm.assert_series_equal(result1, expected1)
+
+    def test_spline_smooth(self):
+        pytest.importorskip("scipy")
+        s = Series([1, 2, np.nan, 4, 5.1, np.nan, 7])
+        assert (
+            s.interpolate(method="spline", order=3, s=0)[5]
+            != s.interpolate(method="spline", order=3)[5]
+        )
+
+    def test_spline_interpolation(self):
+        # Explicit cast to float to avoid implicit cast when setting np.nan
+        pytest.importorskip("scipy")
+        s = Series(np.arange(10) ** 2, dtype="float")
+        s[np.random.default_rng(2).integers(0, 9, 3)] = np.nan
+        result1 = s.interpolate(method="spline", order=1)
+        expected1 = s.interpolate(method="spline", order=1)
+        tm.assert_series_equal(result1, expected1)
+
+    def test_interp_timedelta64(self):
+        # GH 6424
+        df = Series([1, np.nan, 3], index=pd.to_timedelta([1, 2, 3]))
+        result = df.interpolate(method="time")
+        expected = Series([1.0, 2.0, 3.0], index=pd.to_timedelta([1, 2, 3]))
+        tm.assert_series_equal(result, expected)
+
+        # test for non uniform spacing
+        df = Series([1, np.nan, 3], index=pd.to_timedelta([1, 2, 4]))
+        result = df.interpolate(method="time")
+        expected = Series([1.0, 1.666667, 3.0], index=pd.to_timedelta([1, 2, 4]))
+        tm.assert_series_equal(result, expected)
+
+    def test_series_interpolate_method_values(self):
+        # GH#1646
+        rng = date_range("1/1/2000", "1/20/2000", freq="D")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        ts[::2] = np.nan
+
+        result = ts.interpolate(method="values")
+        exp = ts.interpolate()
+        tm.assert_series_equal(result, exp)
+
+    def test_series_interpolate_intraday(self):
+        # #1698
+        index = date_range("1/1/2012", periods=4, freq="12D")
+        ts = Series([0, 12, 24, 36], index)
+        new_index = index.append(index + pd.DateOffset(days=1)).sort_values()
+
+        exp = ts.reindex(new_index).interpolate(method="time")
+
+        index = date_range("1/1/2012", periods=4, freq="12h")
+        ts = Series([0, 12, 24, 36], index)
+        new_index = index.append(index + pd.DateOffset(hours=1)).sort_values()
+        result = ts.reindex(new_index).interpolate(method="time")
+
+        tm.assert_numpy_array_equal(result.values, exp.values)
+
+    @pytest.mark.parametrize(
+        "ind",
+        [
+            ["a", "b", "c", "d"],
+            pd.period_range(start="2019-01-01", periods=4),
+            pd.interval_range(start=0, end=4),
+        ],
+    )
+    def test_interp_non_timedelta_index(self, interp_methods_ind, ind):
+        # gh 21662
+        df = pd.DataFrame([0, 1, np.nan, 3], index=ind)
+
+        method, kwargs = interp_methods_ind
+        if method == "pchip":
+            pytest.importorskip("scipy")
+
+        if method == "linear":
+            result = df[0].interpolate(**kwargs)
+            expected = Series([0.0, 1.0, 2.0, 3.0], name=0, index=ind)
+            tm.assert_series_equal(result, expected)
+        else:
+            expected_error = (
+                "Index column must be numeric or datetime type when "
+                f"using {method} method other than linear. "
+                "Try setting a numeric or datetime index column before "
+                "interpolating."
+            )
+            with pytest.raises(ValueError, match=expected_error):
+                df[0].interpolate(method=method, **kwargs)
+
+    def test_interpolate_timedelta_index(self, request, interp_methods_ind):
+        """
+        Tests for non numerical index types  - object, period, timedelta
+        Note that all methods except time, index, nearest and values
+        are tested here.
+        """
+        # gh 21662
+        pytest.importorskip("scipy")
+        ind = pd.timedelta_range(start=1, periods=4)
+        df = pd.DataFrame([0, 1, np.nan, 3], index=ind)
+
+        method, kwargs = interp_methods_ind
+
+        if method in {"cubic", "zero"}:
+            request.applymarker(
+                pytest.mark.xfail(
+                    reason=f"{method} interpolation is not supported for TimedeltaIndex"
+                )
+            )
+        result = df[0].interpolate(method=method, **kwargs)
+        expected = Series([0.0, 1.0, 2.0, 3.0], name=0, index=ind)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ascending, expected_values",
+        [(True, [1, 2, 3, 9, 10]), (False, [10, 9, 3, 2, 1])],
+    )
+    def test_interpolate_unsorted_index(self, ascending, expected_values):
+        # GH 21037
+        ts = Series(data=[10, 9, np.nan, 2, 1], index=[10, 9, 3, 2, 1])
+        result = ts.sort_index(ascending=ascending).interpolate(method="index")
+        expected = Series(data=expected_values, index=expected_values, dtype=float)
+        tm.assert_series_equal(result, expected)
+
+    def test_interpolate_asfreq_raises(self):
+        ser = Series(["a", None, "b"], dtype=object)
+        msg = "Can not interpolate with method=asfreq"
+        with pytest.raises(ValueError, match=msg):
+            ser.interpolate(method="asfreq")
+
+    def test_interpolate_fill_value(self):
+        # GH#54920
+        pytest.importorskip("scipy")
+        ser = Series([np.nan, 0, 1, np.nan, 3, np.nan])
+        result = ser.interpolate(method="nearest", fill_value=0)
+        expected = Series([np.nan, 0, 1, 1, 3, 0])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_is_monotonic.py b/pandas/pandas/tests/series/methods/test_is_monotonic.py
new file mode 100644
index 0000000000000..073ec4172aff6
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_is_monotonic.py
@@ -0,0 +1,26 @@
+import numpy as np
+
+from pandas import (
+    Series,
+    date_range,
+)
+
+
+class TestIsMonotonic:
+    def test_is_monotonic_numeric(self):
+        ser = Series(np.random.default_rng(2).integers(0, 10, size=1000))
+        assert not ser.is_monotonic_increasing
+        ser = Series(np.arange(1000))
+        assert ser.is_monotonic_increasing is True
+        assert ser.is_monotonic_increasing is True
+        ser = Series(np.arange(1000, 0, -1))
+        assert ser.is_monotonic_decreasing is True
+
+    def test_is_monotonic_dt64(self):
+        ser = Series(date_range("20130101", periods=10))
+        assert ser.is_monotonic_increasing is True
+        assert ser.is_monotonic_increasing is True
+
+        ser = Series(list(reversed(ser)))
+        assert ser.is_monotonic_increasing is False
+        assert ser.is_monotonic_decreasing is True
diff --git a/pandas/pandas/tests/series/methods/test_is_unique.py b/pandas/pandas/tests/series/methods/test_is_unique.py
new file mode 100644
index 0000000000000..edf3839c2cebb
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_is_unique.py
@@ -0,0 +1,40 @@
+import numpy as np
+import pytest
+
+from pandas import Series
+
+
+@pytest.mark.parametrize(
+    "data, expected",
+    [
+        (np.random.default_rng(2).integers(0, 10, size=1000), False),
+        (np.arange(1000), True),
+        ([], True),
+        ([np.nan], True),
+        (["foo", "bar", np.nan], True),
+        (["foo", "foo", np.nan], False),
+        (["foo", "bar", np.nan, np.nan], False),
+    ],
+)
+def test_is_unique(data, expected):
+    # GH#11946 / GH#25180
+    ser = Series(data)
+    assert ser.is_unique is expected
+
+
+def test_is_unique_class_ne(capsys):
+    # GH#20661
+    class Foo:
+        def __init__(self, val) -> None:
+            self._value = val
+
+        def __ne__(self, other):
+            raise Exception("NEQ not supported")
+
+    with capsys.disabled():
+        li = [Foo(i) for i in range(5)]
+        ser = Series(li, index=list(range(5)))
+
+    ser.is_unique
+    captured = capsys.readouterr()
+    assert len(captured.err) == 0
diff --git a/pandas/pandas/tests/series/methods/test_isin.py b/pandas/pandas/tests/series/methods/test_isin.py
new file mode 100644
index 0000000000000..4f8484252ba8f
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_isin.py
@@ -0,0 +1,269 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core import algorithms
+from pandas.core.arrays import PeriodArray
+
+
+class TestSeriesIsIn:
+    def test_isin(self):
+        s = Series(["A", "B", "C", "a", "B", "B", "A", "C"])
+
+        result = s.isin(["A", "C"])
+        expected = Series([True, False, True, False, False, False, True, True])
+        tm.assert_series_equal(result, expected)
+
+        # GH#16012
+        # This specific issue has to have a series over 1e6 in len, but the
+        # comparison array (in_list) must be large enough so that numpy doesn't
+        # do a manual masking trick that will avoid this issue altogether
+        s = Series(list("abcdefghijk" * 10**5))
+        # If numpy doesn't do the manual comparison/mask, these
+        # unorderable mixed types are what cause the exception in numpy
+        in_list = [-1, "a", "b", "G", "Y", "Z", "E", "K", "E", "S", "I", "R", "R"] * 6
+
+        assert s.isin(in_list).sum() == 200000
+
+    def test_isin_with_string_scalar(self):
+        # GH#4763
+        s = Series(["A", "B", "C", "a", "B", "B", "A", "C"])
+        msg = (
+            r"only list-like objects are allowed to be passed to isin\(\), "
+            r"you passed a `str`"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s.isin("a")
+
+        s = Series(["aaa", "b", "c"])
+        with pytest.raises(TypeError, match=msg):
+            s.isin("aaa")
+
+    def test_isin_datetimelike_mismatched_reso(self):
+        expected = Series([True, True, False, False, False])
+
+        ser = Series(date_range("jan-01-2013", "jan-05-2013"))
+
+        # fails on dtype conversion in the first place
+        day_values = np.asarray(ser[0:2].values).astype("datetime64[D]")
+        result = ser.isin(day_values)
+        tm.assert_series_equal(result, expected)
+
+        dta = ser[:2]._values.astype("M8[s]")
+        result = ser.isin(dta)
+        tm.assert_series_equal(result, expected)
+
+    def test_isin_datetimelike_mismatched_reso_list(self):
+        expected = Series([True, True, False, False, False])
+
+        ser = Series(date_range("jan-01-2013", "jan-05-2013"))
+
+        dta = ser[:2]._values.astype("M8[s]")
+        result = ser.isin(list(dta))
+        tm.assert_series_equal(result, expected)
+
+    def test_isin_with_i8(self):
+        # GH#5021
+
+        expected = Series([True, True, False, False, False])
+        expected2 = Series([False, True, False, False, False])
+
+        # datetime64[ns]
+        s = Series(date_range("jan-01-2013", "jan-05-2013"))
+
+        result = s.isin(s[0:2])
+        tm.assert_series_equal(result, expected)
+
+        result = s.isin(s[0:2].values)
+        tm.assert_series_equal(result, expected)
+
+        result = s.isin([s[1]])
+        tm.assert_series_equal(result, expected2)
+
+        result = s.isin([np.datetime64(s[1])])
+        tm.assert_series_equal(result, expected2)
+
+        result = s.isin(set(s[0:2]))
+        tm.assert_series_equal(result, expected)
+
+        # timedelta64[ns]
+        s = Series(pd.to_timedelta(range(5), unit="D"))
+        result = s.isin(s[0:2])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("empty", [[], Series(dtype=object), np.array([])])
+    def test_isin_empty(self, empty):
+        # see GH#16991
+        s = Series(["a", "b"])
+        expected = Series([False, False])
+
+        result = s.isin(empty)
+        tm.assert_series_equal(expected, result)
+
+    def test_isin_read_only(self):
+        # https://github.com/pandas-dev/pandas/issues/37174
+        arr = np.array([1, 2, 3])
+        arr.setflags(write=False)
+        s = Series([1, 2, 3])
+        result = s.isin(arr)
+        expected = Series([True, True, True])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", [object, None])
+    def test_isin_dt64_values_vs_ints(self, dtype):
+        # GH#36621 dont cast integers to datetimes for isin
+        dti = date_range("2013-01-01", "2013-01-05")
+        ser = Series(dti)
+
+        comps = np.asarray([1356998400000000000], dtype=dtype)
+
+        res = dti.isin(comps)
+        expected = np.array([False] * len(dti), dtype=bool)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = ser.isin(comps)
+        tm.assert_series_equal(res, Series(expected))
+
+        res = pd.core.algorithms.isin(ser, comps)
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_isin_tzawareness_mismatch(self):
+        dti = date_range("2013-01-01", "2013-01-05")
+        ser = Series(dti)
+
+        other = dti.tz_localize("UTC")
+
+        res = dti.isin(other)
+        expected = np.array([False] * len(dti), dtype=bool)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = ser.isin(other)
+        tm.assert_series_equal(res, Series(expected))
+
+        res = pd.core.algorithms.isin(ser, other)
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_isin_period_freq_mismatch(self):
+        dti = date_range("2013-01-01", "2013-01-05")
+        pi = dti.to_period("M")
+        ser = Series(pi)
+
+        # We construct another PeriodIndex with the same i8 values
+        #  but different dtype
+        dtype = dti.to_period("Y").dtype
+        other = PeriodArray._simple_new(pi.asi8, dtype=dtype)
+
+        res = pi.isin(other)
+        expected = np.array([False] * len(pi), dtype=bool)
+        tm.assert_numpy_array_equal(res, expected)
+
+        res = ser.isin(other)
+        tm.assert_series_equal(res, Series(expected))
+
+        res = pd.core.algorithms.isin(ser, other)
+        tm.assert_numpy_array_equal(res, expected)
+
+    @pytest.mark.parametrize("values", [[-9.0, 0.0], [-9, 0]])
+    def test_isin_float_in_int_series(self, values):
+        # GH#19356 GH#21804
+        ser = Series(values)
+        result = ser.isin([-9, -0.5])
+        expected = Series([True, False])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["boolean", "Int64", "Float64"])
+    @pytest.mark.parametrize(
+        "data,values,expected",
+        [
+            ([0, 1, 0], [1], [False, True, False]),
+            ([0, 1, 0], [1, pd.NA], [False, True, False]),
+            ([0, pd.NA, 0], [1, 0], [True, False, True]),
+            ([0, 1, pd.NA], [1, pd.NA], [False, True, True]),
+            ([0, 1, pd.NA], [1, np.nan], [False, True, False]),
+            ([0, pd.NA, pd.NA], [np.nan, pd.NaT, None], [False, False, False]),
+        ],
+    )
+    def test_isin_masked_types(self, dtype, data, values, expected):
+        # GH#42405
+        ser = Series(data, dtype=dtype)
+
+        result = ser.isin(values)
+        expected = Series(expected, dtype="boolean")
+
+        tm.assert_series_equal(result, expected)
+
+
+def test_isin_large_series_mixed_dtypes_and_nan(monkeypatch):
+    # https://github.com/pandas-dev/pandas/issues/37094
+    # combination of object dtype for the values
+    # and > _MINIMUM_COMP_ARR_LEN elements
+    min_isin_comp = 5
+    ser = Series([1, 2, np.nan] * min_isin_comp)
+    with monkeypatch.context() as m:
+        m.setattr(algorithms, "_MINIMUM_COMP_ARR_LEN", min_isin_comp)
+        result = ser.isin({"foo", "bar"})
+    expected = Series([False] * 3 * min_isin_comp)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype, data, values, expected",
+    [
+        ("boolean", [pd.NA, False, True], [False, pd.NA], [True, True, False]),
+        ("Int64", [pd.NA, 2, 1], [1, pd.NA], [True, False, True]),
+        ("boolean", [pd.NA, False, True], [pd.NA, True, "a", 20], [True, False, True]),
+        ("boolean", [pd.NA, False, True], [], [False, False, False]),
+        ("Float64", [20.0, 30.0, pd.NA], [pd.NA], [False, False, True]),
+    ],
+)
+def test_isin_large_series_and_pdNA(dtype, data, values, expected, monkeypatch):
+    # https://github.com/pandas-dev/pandas/issues/60678
+    # combination of  large series (> _MINIMUM_COMP_ARR_LEN elements) and
+    # values contains pdNA
+    min_isin_comp = 2
+    ser = Series(data, dtype=dtype)
+    expected = Series(expected, dtype="boolean")
+
+    with monkeypatch.context() as m:
+        m.setattr(algorithms, "_MINIMUM_COMP_ARR_LEN", min_isin_comp)
+        result = ser.isin(values)
+    tm.assert_series_equal(result, expected)
+
+
+def test_isin_complex_numbers():
+    # GH 17927
+    array = [0, 1j, 1j, 1, 1 + 1j, 1 + 2j, 1 + 1j]
+    result = Series(array).isin([1j, 1 + 1j, 1 + 2j])
+    expected = Series([False, True, True, False, True, True, True], dtype=bool)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,is_in",
+    [([1, [2]], [1]), (["simple str", [{"values": 3}]], ["simple str"])],
+)
+def test_isin_filtering_with_mixed_object_types(data, is_in):
+    # GH 20883
+
+    ser = Series(data)
+    result = ser.isin(is_in)
+    expected = Series([True, False])
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("data", [[1, 2, 3], [1.0, 2.0, 3.0]])
+@pytest.mark.parametrize("isin", [[1, 2], [1.0, 2.0]])
+def test_isin_filtering_on_iterable(data, isin):
+    # GH 50234
+
+    ser = Series(data)
+    result = ser.isin(i for i in isin)
+    expected_result = Series([True, True, False])
+
+    tm.assert_series_equal(result, expected_result)
diff --git a/pandas/pandas/tests/series/methods/test_isna.py b/pandas/pandas/tests/series/methods/test_isna.py
new file mode 100644
index 0000000000000..92bf2945cc0d1
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_isna.py
@@ -0,0 +1,36 @@
+"""
+We also test Series.notna in this file.
+"""
+
+import numpy as np
+
+from pandas import (
+    Period,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestIsna:
+    def test_isna_period_dtype(self):
+        # GH#13737
+        ser = Series([Period("2011-01", freq="M"), Period("NaT", freq="M")])
+
+        expected = Series([False, True])
+
+        result = ser.isna()
+        tm.assert_series_equal(result, expected)
+
+        result = ser.notna()
+        tm.assert_series_equal(result, ~expected)
+
+    def test_isna(self):
+        ser = Series([0, 5.4, 3, np.nan, -0.001])
+        expected = Series([False, False, False, True, False])
+        tm.assert_series_equal(ser.isna(), expected)
+        tm.assert_series_equal(ser.notna(), ~expected)
+
+        ser = Series(["hi", "", np.nan])
+        expected = Series([False, False, True])
+        tm.assert_series_equal(ser.isna(), expected)
+        tm.assert_series_equal(ser.notna(), ~expected)
diff --git a/pandas/pandas/tests/series/methods/test_item.py b/pandas/pandas/tests/series/methods/test_item.py
new file mode 100644
index 0000000000000..e927fa69db358
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_item.py
@@ -0,0 +1,60 @@
+"""
+Series.item method, mainly testing that we get python scalars as opposed to
+numpy scalars.
+"""
+
+import pytest
+
+from pandas import (
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+)
+
+
+class TestItem:
+    def test_item(self):
+        # We are testing that we get python scalars as opposed to numpy scalars
+        ser = Series([1])
+        result = ser.item()
+        assert result == 1
+        assert result == ser.iloc[0]
+        assert isinstance(result, int)  # i.e. not np.int64
+
+        ser = Series([0.5], index=[3])
+        result = ser.item()
+        assert isinstance(result, float)
+        assert result == 0.5
+
+        ser = Series([1, 2])
+        msg = "can only convert an array of size 1"
+        with pytest.raises(ValueError, match=msg):
+            ser.item()
+
+        dti = date_range("2016-01-01", periods=2)
+        with pytest.raises(ValueError, match=msg):
+            dti.item()
+        with pytest.raises(ValueError, match=msg):
+            Series(dti).item()
+
+        val = dti[:1].item()
+        assert isinstance(val, Timestamp)
+        val = Series(dti)[:1].item()
+        assert isinstance(val, Timestamp)
+
+        tdi = dti - dti
+        with pytest.raises(ValueError, match=msg):
+            tdi.item()
+        with pytest.raises(ValueError, match=msg):
+            Series(tdi).item()
+
+        val = tdi[:1].item()
+        assert isinstance(val, Timedelta)
+        val = Series(tdi)[:1].item()
+        assert isinstance(val, Timedelta)
+
+        # Case where ser[0] would not work
+        ser = Series(dti, index=[5, 6])
+        val = ser.iloc[:1].item()
+        assert val == dti[0]
diff --git a/pandas/pandas/tests/series/methods/test_map.py b/pandas/pandas/tests/series/methods/test_map.py
new file mode 100644
index 0000000000000..84b60a2afe6eb
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_map.py
@@ -0,0 +1,606 @@
+from collections import (
+    Counter,
+    defaultdict,
+)
+from decimal import Decimal
+import math
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    bdate_range,
+    date_range,
+    isna,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+def test_series_map_box_timedelta():
+    # GH#11349
+    ser = Series(timedelta_range("1 day 1 s", periods=5, freq="h"))
+
+    def f(x):
+        return x.total_seconds()
+
+    ser.map(f)
+
+
+def test_map_callable(datetime_series):
+    with np.errstate(all="ignore"):
+        tm.assert_series_equal(datetime_series.map(np.sqrt), np.sqrt(datetime_series))
+
+    # map function element-wise
+    tm.assert_series_equal(datetime_series.map(math.exp), np.exp(datetime_series))
+
+    # empty series
+    s = Series(dtype=object, name="foo", index=Index([], name="bar"))
+    rs = s.map(lambda x: x)
+    tm.assert_series_equal(s, rs)
+
+    # check all metadata (GH 9322)
+    assert s is not rs
+    assert s.index is rs.index
+    assert s.dtype == rs.dtype
+    assert s.name == rs.name
+
+    # index but no data
+    s = Series(index=[1, 2, 3], dtype=np.float64)
+    rs = s.map(lambda x: x)
+    tm.assert_series_equal(s, rs)
+
+
+def test_map_same_length_inference_bug():
+    s = Series([1, 2])
+
+    def f(x):
+        return (x, x + 1)
+
+    s = Series([1, 2, 3])
+    result = s.map(f)
+    expected = Series([(1, 2), (2, 3), (3, 4)])
+    tm.assert_series_equal(result, expected)
+
+    s = Series(["foo,bar"])
+    result = s.map(lambda x: x.split(","))
+    expected = Series([("foo", "bar")])
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_map_box_timestamps():
+    # GH#2689, GH#2627
+    ser = Series(date_range("1/1/2000", periods=3))
+
+    def func(x):
+        return (x.hour, x.day, x.month)
+
+    result = ser.map(func)
+    expected = Series([(0, 1, 1), (0, 2, 1), (0, 3, 1)])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_series_stringdtype(any_string_dtype, using_infer_string):
+    # map test on StringDType, GH#40823
+    ser1 = Series(
+        data=["cat", "dog", "rabbit"],
+        index=["id1", "id2", "id3"],
+        dtype=any_string_dtype,
+    )
+    ser2 = Series(["id3", "id2", "id1", "id7000"], dtype=any_string_dtype)
+    result = ser2.map(ser1)
+
+    item = pd.NA
+    if ser2.dtype == object:
+        item = np.nan
+
+    expected = Series(data=["rabbit", "dog", "cat", item], dtype=any_string_dtype)
+    if using_infer_string and any_string_dtype == "object":
+        expected = expected.astype("str")
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected_dtype",
+    [(["1-1", "1-1", np.nan], "category"), (["1-1", "1-2", np.nan], "str")],
+)
+def test_map_categorical_with_nan_values(data, expected_dtype):
+    # GH 20714 bug fixed in: GH 24275
+    def func(val):
+        return val.split("-")[0]
+
+    s = Series(data, dtype="category")
+
+    result = s.map(func, na_action="ignore")
+    expected = Series(["1", "1", np.nan], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_empty_integer_series():
+    # GH52384
+    s = Series([], dtype=int)
+    result = s.map(lambda x: x)
+    tm.assert_series_equal(result, s)
+
+
+def test_map_empty_integer_series_with_datetime_index():
+    # GH 21245
+    s = Series([], index=date_range(start="2018-01-01", periods=0), dtype=int)
+    result = s.map(lambda x: x)
+    tm.assert_series_equal(result, s)
+
+
+@pytest.mark.parametrize("func", [str, lambda x: str(x)])
+def test_map_simple_str_callables_same_as_astype(
+    string_series, func, using_infer_string
+):
+    # test that we are evaluating row-by-row first
+    # before vectorized evaluation
+    result = string_series.map(func)
+    expected = string_series.astype(str if not using_infer_string else "str")
+    tm.assert_series_equal(result, expected)
+
+
+def test_list_raises(string_series):
+    with pytest.raises(TypeError, match="'list' object is not callable"):
+        string_series.map([lambda x: x])
+
+
+def test_map():
+    data = {
+        "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+        "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+        "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+        "D": bdate_range("1/1/2009", periods=5),
+    }
+
+    source = Series(data["B"], index=data["C"])
+    target = Series(data["C"][:4], index=data["D"][:4])
+
+    merged = target.map(source)
+
+    for k, v in merged.items():
+        assert v == source[target[k]]
+
+    # input could be a dict
+    merged = target.map(source.to_dict())
+
+    for k, v in merged.items():
+        assert v == source[target[k]]
+
+
+def test_map_datetime(datetime_series):
+    # function
+    result = datetime_series.map(lambda x: x * 2)
+    tm.assert_series_equal(result, datetime_series * 2)
+
+
+def test_map_category():
+    # GH 10324
+    a = Series([1, 2, 3, 4])
+    b = Series(["even", "odd", "even", "odd"], dtype="category")
+    c = Series(["even", "odd", "even", "odd"])
+
+    exp = Series(["odd", "even", "odd", np.nan], dtype="category")
+    tm.assert_series_equal(a.map(b), exp)
+    exp = Series(["odd", "even", "odd", np.nan])
+    tm.assert_series_equal(a.map(c), exp)
+
+
+def test_map_category_numeric():
+    a = Series(["a", "b", "c", "d"])
+    b = Series([1, 2, 3, 4], index=pd.CategoricalIndex(["b", "c", "d", "e"]))
+    c = Series([1, 2, 3, 4], index=Index(["b", "c", "d", "e"]))
+
+    exp = Series([np.nan, 1, 2, 3])
+    tm.assert_series_equal(a.map(b), exp)
+    exp = Series([np.nan, 1, 2, 3])
+    tm.assert_series_equal(a.map(c), exp)
+
+
+def test_map_category_string():
+    a = Series(["a", "b", "c", "d"])
+    b = Series(
+        ["B", "C", "D", "E"],
+        dtype="category",
+        index=pd.CategoricalIndex(["b", "c", "d", "e"]),
+    )
+    c = Series(["B", "C", "D", "E"], index=Index(["b", "c", "d", "e"]))
+
+    exp = Series(
+        pd.Categorical([np.nan, "B", "C", "D"], categories=["B", "C", "D", "E"])
+    )
+    tm.assert_series_equal(a.map(b), exp)
+    exp = Series([np.nan, "B", "C", "D"])
+    tm.assert_series_equal(a.map(c), exp)
+
+
+def test_map_empty(request, index):
+    if isinstance(index, MultiIndex):
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Initializing a Series from a MultiIndex is not supported"
+            )
+        )
+
+    s = Series(index)
+    result = s.map({})
+
+    expected = Series(np.nan, index=s.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_compat():
+    # related GH 8024
+    s = Series([True, True, False], index=[1, 2, 3])
+    result = s.map({True: "foo", False: "bar"})
+    expected = Series(["foo", "foo", "bar"], index=[1, 2, 3])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_int():
+    left = Series({"a": 1.0, "b": 2.0, "c": 3.0, "d": 4})
+    right = Series({1: 11, 2: 22, 3: 33})
+
+    assert left.dtype == np.float64
+    assert issubclass(right.dtype.type, np.integer)
+
+    merged = left.map(right)
+    assert merged.dtype == np.float64
+    assert isna(merged["d"])
+    assert not isna(merged["c"])
+
+
+def test_map_type_inference():
+    s = Series(range(3))
+    s2 = s.map(lambda x: np.where(x == 0, 0, 1))
+    assert issubclass(s2.dtype.type, np.integer)
+
+
+def test_map_decimal(string_series):
+    result = string_series.map(lambda x: Decimal(str(x)))
+    assert result.dtype == np.object_
+    assert isinstance(result.iloc[0], Decimal)
+
+
+def test_map_na_exclusion():
+    s = Series([1.5, np.nan, 3, np.nan, 5])
+
+    result = s.map(lambda x: x * 2, na_action="ignore")
+    exp = s * 2
+    tm.assert_series_equal(result, exp)
+
+
+def test_map_dict_with_tuple_keys():
+    """
+    Due to new MultiIndex-ing behaviour in v0.14.0,
+    dicts with tuple keys passed to map were being
+    converted to a multi-index, preventing tuple values
+    from being mapped properly.
+    """
+    # GH 18496
+    df = DataFrame({"a": [(1,), (2,), (3, 4), (5, 6)]})
+    label_mappings = {(1,): "A", (2,): "B", (3, 4): "A", (5, 6): "B"}
+
+    df["labels"] = df["a"].map(label_mappings)
+    df["expected_labels"] = Series(["A", "B", "A", "B"], index=df.index)
+    # All labels should be filled now
+    tm.assert_series_equal(df["labels"], df["expected_labels"], check_names=False)
+
+
+def test_map_counter():
+    s = Series(["a", "b", "c"], index=[1, 2, 3])
+    counter = Counter()
+    counter["b"] = 5
+    counter["c"] += 1
+    result = s.map(counter)
+    expected = Series([0, 5, 1], index=[1, 2, 3])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_defaultdict():
+    s = Series([1, 2, 3], index=["a", "b", "c"])
+    default_dict = defaultdict(lambda: "blank")
+    default_dict[1] = "stuff"
+    result = s.map(default_dict)
+    expected = Series(["stuff", "blank", "blank"], index=["a", "b", "c"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_dict_na_key():
+    # https://github.com/pandas-dev/pandas/issues/17648
+    # Checks that np.nan key is appropriately mapped
+    s = Series([1, 2, np.nan])
+    expected = Series(["a", "b", "c"])
+    result = s.map({1: "a", 2: "b", np.nan: "c"})
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_defaultdict_na_key(na_action):
+    # GH 48813
+    s = Series([1, 2, np.nan])
+    default_map = defaultdict(lambda: "missing", {1: "a", 2: "b", np.nan: "c"})
+    result = s.map(default_map, na_action=na_action)
+    expected = Series({0: "a", 1: "b", 2: "c" if na_action is None else np.nan})
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_defaultdict_missing_key(na_action):
+    # GH 48813
+    s = Series([1, 2, np.nan])
+    default_map = defaultdict(lambda: "missing", {1: "a", 2: "b", 3: "c"})
+    result = s.map(default_map, na_action=na_action)
+    expected = Series({0: "a", 1: "b", 2: "missing" if na_action is None else np.nan})
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_defaultdict_unmutated(na_action):
+    # GH 48813
+    s = Series([1, 2, np.nan])
+    default_map = defaultdict(lambda: "missing", {1: "a", 2: "b", np.nan: "c"})
+    expected_default_map = default_map.copy()
+    s.map(default_map, na_action=na_action)
+    assert default_map == expected_default_map
+
+
+@pytest.mark.parametrize("arg_func", [dict, Series])
+def test_map_dict_ignore_na(arg_func):
+    # GH#47527
+    mapping = arg_func({1: 10, np.nan: 42})
+    ser = Series([1, np.nan, 2])
+    result = ser.map(mapping, na_action="ignore")
+    expected = Series([10, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_defaultdict_ignore_na():
+    # GH#47527
+    mapping = defaultdict(int, {1: 10, np.nan: 42})
+    ser = Series([1, np.nan, 2])
+    result = ser.map(mapping)
+    expected = Series([10, 42, 0])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na_action, expected",
+    [(None, Series([10.0, 42.0, np.nan])), ("ignore", Series([10, np.nan, np.nan]))],
+)
+def test_map_categorical_na_ignore(na_action, expected):
+    # GH#47527
+    values = pd.Categorical([1, np.nan, 2], categories=[10, 1, 2])
+    ser = Series(values)
+    result = ser.map({1: 10, np.nan: 42}, na_action=na_action)
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_dict_subclass_with_missing():
+    """
+    Test Series.map with a dictionary subclass that defines __missing__,
+    i.e. sets a default value (GH #15999).
+    """
+
+    class DictWithMissing(dict):
+        def __missing__(self, key):
+            return "missing"
+
+    s = Series([1, 2, 3])
+    dictionary = DictWithMissing({3: "three"})
+    result = s.map(dictionary)
+    expected = Series(["missing", "missing", "three"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_dict_subclass_without_missing():
+    class DictWithoutMissing(dict):
+        pass
+
+    s = Series([1, 2, 3])
+    dictionary = DictWithoutMissing({3: "three"})
+    result = s.map(dictionary)
+    expected = Series([np.nan, np.nan, "three"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_abc_mapping(non_dict_mapping_subclass):
+    # https://github.com/pandas-dev/pandas/issues/29733
+    # Check collections.abc.Mapping support as mapper for Series.map
+    s = Series([1, 2, 3])
+    not_a_dictionary = non_dict_mapping_subclass({3: "three"})
+    result = s.map(not_a_dictionary)
+    expected = Series([np.nan, np.nan, "three"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_abc_mapping_with_missing(non_dict_mapping_subclass):
+    # https://github.com/pandas-dev/pandas/issues/29733
+    # Check collections.abc.Mapping support as mapper for Series.map
+    class NonDictMappingWithMissing(non_dict_mapping_subclass):
+        def __missing__(self, key):
+            return "missing"
+
+    s = Series([1, 2, 3])
+    not_a_dictionary = NonDictMappingWithMissing({3: "three"})
+    result = s.map(not_a_dictionary)
+    # __missing__ is a dict concept, not a Mapping concept,
+    # so it should not change the result!
+    expected = Series([np.nan, np.nan, "three"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_box_dt64(unit):
+    vals = [pd.Timestamp("2011-01-01"), pd.Timestamp("2011-01-02")]
+    ser = Series(vals).dt.as_unit(unit)
+    assert ser.dtype == f"datetime64[{unit}]"
+    # boxed value must be Timestamp instance
+    res = ser.map(lambda x: f"{type(x).__name__}_{x.day}_{x.tz}")
+    exp = Series(["Timestamp_1_None", "Timestamp_2_None"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_map_box_dt64tz(unit):
+    vals = [
+        pd.Timestamp("2011-01-01", tz="US/Eastern"),
+        pd.Timestamp("2011-01-02", tz="US/Eastern"),
+    ]
+    ser = Series(vals).dt.as_unit(unit)
+    assert ser.dtype == f"datetime64[{unit}, US/Eastern]"
+    res = ser.map(lambda x: f"{type(x).__name__}_{x.day}_{x.tz}")
+    exp = Series(["Timestamp_1_US/Eastern", "Timestamp_2_US/Eastern"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_map_box_td64(unit):
+    # timedelta
+    vals = [pd.Timedelta("1 days"), pd.Timedelta("2 days")]
+    ser = Series(vals).dt.as_unit(unit)
+    assert ser.dtype == f"timedelta64[{unit}]"
+    res = ser.map(lambda x: f"{type(x).__name__}_{x.days}")
+    exp = Series(["Timedelta_1", "Timedelta_2"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_map_box_period():
+    # period
+    vals = [pd.Period("2011-01-01", freq="M"), pd.Period("2011-01-02", freq="M")]
+    ser = Series(vals)
+    assert ser.dtype == "Period[M]"
+    res = ser.map(lambda x: f"{type(x).__name__}_{x.freqstr}")
+    exp = Series(["Period_M", "Period_M"])
+    tm.assert_series_equal(res, exp)
+
+
+def test_map_categorical(na_action, using_infer_string):
+    values = pd.Categorical(list("ABBABCD"), categories=list("DCBA"), ordered=True)
+    s = Series(values, name="XX", index=list("abcdefg"))
+
+    result = s.map(lambda x: x.lower(), na_action=na_action)
+    exp_values = pd.Categorical(list("abbabcd"), categories=list("dcba"), ordered=True)
+    exp = Series(exp_values, name="XX", index=list("abcdefg"))
+    tm.assert_series_equal(result, exp)
+    tm.assert_categorical_equal(result.values, exp_values)
+
+    result = s.map(lambda x: "A", na_action=na_action)
+    exp = Series(["A"] * 7, name="XX", index=list("abcdefg"))
+    tm.assert_series_equal(result, exp)
+    assert result.dtype == object if not using_infer_string else "str"
+
+
+@pytest.mark.parametrize(
+    "na_action, expected",
+    (
+        [None, Series(["A", "B", "nan"], name="XX")],
+        [
+            "ignore",
+            Series(
+                ["A", "B", np.nan],
+                name="XX",
+                dtype=pd.CategoricalDtype(list("DCBA"), True),
+            ),
+        ],
+    ),
+)
+def test_map_categorical_na_action(na_action, expected):
+    dtype = pd.CategoricalDtype(list("DCBA"), ordered=True)
+    values = pd.Categorical(list("AB") + [np.nan], dtype=dtype)
+    s = Series(values, name="XX")
+    result = s.map(str, na_action=na_action)
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_datetimetz():
+    values = date_range("2011-01-01", "2011-01-02", freq="h").tz_localize("Asia/Tokyo")
+    s = Series(values, name="XX")
+
+    # keep tz
+    result = s.map(lambda x: x + pd.offsets.Day())
+    exp_values = date_range("2011-01-02", "2011-01-03", freq="h").tz_localize(
+        "Asia/Tokyo"
+    )
+    exp = Series(exp_values, name="XX")
+    tm.assert_series_equal(result, exp)
+
+    result = s.map(lambda x: x.hour)
+    exp = Series(list(range(24)) + [0], name="XX", dtype=np.int64)
+    tm.assert_series_equal(result, exp)
+
+    # not vectorized
+    def f(x):
+        if not isinstance(x, pd.Timestamp):
+            raise ValueError
+        return str(x.tz)
+
+    result = s.map(f)
+    exp = Series(["Asia/Tokyo"] * 25, name="XX")
+    tm.assert_series_equal(result, exp)
+
+
+@pytest.mark.parametrize(
+    "vals,mapping,exp",
+    [
+        (list("abc"), {np.nan: "not NaN"}, [np.nan] * 3 + ["not NaN"]),
+        (list("abc"), {"a": "a letter"}, ["a letter"] + [np.nan] * 3),
+        (list(range(3)), {0: 42}, [42] + [np.nan] * 3),
+    ],
+)
+def test_map_missing_mixed(vals, mapping, exp):
+    # GH20495
+    s = Series(vals + [np.nan])
+    result = s.map(mapping)
+    exp = Series(exp)
+    tm.assert_series_equal(result, exp)
+
+
+def test_map_scalar_on_date_time_index_aware_series():
+    # GH 25959
+    # Calling map on a localized time series should not cause an error
+    series = Series(
+        np.arange(10, dtype=np.float64),
+        index=date_range("2020-01-01", periods=10, tz="UTC"),
+        name="ts",
+    )
+    result = Series(series.index).map(lambda x: 1)
+    tm.assert_series_equal(result, Series(np.ones(len(series)), dtype="int64"))
+
+
+def test_map_float_to_string_precision():
+    # GH 13228
+    ser = Series(1 / 3)
+    result = ser.map(lambda val: str(val)).to_dict()
+    expected = {0: "0.3333333333333333"}
+    assert result == expected
+
+
+def test_map_to_timedelta():
+    list_of_valid_strings = ["00:00:01", "00:00:02"]
+    a = pd.to_timedelta(list_of_valid_strings)
+    b = Series(list_of_valid_strings).map(pd.to_timedelta)
+    tm.assert_series_equal(Series(a), b)
+
+    list_of_strings = ["00:00:01", np.nan, pd.NaT, pd.NaT]
+
+    a = pd.to_timedelta(list_of_strings)
+    ser = Series(list_of_strings)
+    b = ser.map(pd.to_timedelta)
+    tm.assert_series_equal(Series(a), b)
+
+
+def test_map_type():
+    # GH 46719
+    s = Series([3, "string", float], index=["a", "b", "c"])
+    result = s.map(type)
+    expected = Series([int, str, type], index=["a", "b", "c"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_map_kwargs():
+    # GH 59814
+    result = Series([2, 4, 5]).map(lambda x, y: x + y, y=2)
+    expected = Series([4, 6, 7])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_matmul.py b/pandas/pandas/tests/series/methods/test_matmul.py
new file mode 100644
index 0000000000000..4ca3ad3f7031e
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_matmul.py
@@ -0,0 +1,82 @@
+import operator
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestMatmul:
+    def test_matmul(self):
+        # matmul test is for GH#10259
+        a = Series(
+            np.random.default_rng(2).standard_normal(4), index=["p", "q", "r", "s"]
+        )
+        b = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            index=["1", "2", "3"],
+            columns=["p", "q", "r", "s"],
+        ).T
+
+        # Series @ DataFrame -> Series
+        result = operator.matmul(a, b)
+        expected = Series(np.dot(a.values, b.values), index=["1", "2", "3"])
+        tm.assert_series_equal(result, expected)
+
+        # DataFrame @ Series -> Series
+        result = operator.matmul(b.T, a)
+        expected = Series(np.dot(b.T.values, a.T.values), index=["1", "2", "3"])
+        tm.assert_series_equal(result, expected)
+
+        # Series @ Series -> scalar
+        result = operator.matmul(a, a)
+        expected = np.dot(a.values, a.values)
+        tm.assert_almost_equal(result, expected)
+
+        # GH#21530
+        # vector (1D np.array) @ Series (__rmatmul__)
+        result = operator.matmul(a.values, a)
+        expected = np.dot(a.values, a.values)
+        tm.assert_almost_equal(result, expected)
+
+        # GH#21530
+        # vector (1D list) @ Series (__rmatmul__)
+        result = operator.matmul(a.values.tolist(), a)
+        expected = np.dot(a.values, a.values)
+        tm.assert_almost_equal(result, expected)
+
+        # GH#21530
+        # matrix (2D np.array) @ Series (__rmatmul__)
+        result = operator.matmul(b.T.values, a)
+        expected = np.dot(b.T.values, a.values)
+        tm.assert_almost_equal(result, expected)
+
+        # GH#21530
+        # matrix (2D nested lists) @ Series (__rmatmul__)
+        result = operator.matmul(b.T.values.tolist(), a)
+        expected = np.dot(b.T.values, a.values)
+        tm.assert_almost_equal(result, expected)
+
+        # mixed dtype DataFrame @ Series
+        a["p"] = int(a.p)
+        result = operator.matmul(b.T, a)
+        expected = Series(np.dot(b.T.values, a.T.values), index=["1", "2", "3"])
+        tm.assert_series_equal(result, expected)
+
+        # different dtypes DataFrame @ Series
+        a = a.astype(int)
+        result = operator.matmul(b.T, a)
+        expected = Series(np.dot(b.T.values, a.T.values), index=["1", "2", "3"])
+        tm.assert_series_equal(result, expected)
+
+        msg = r"Dot product shape mismatch, \(4,\) vs \(3,\)"
+        # exception raised is of type Exception
+        with pytest.raises(Exception, match=msg):
+            a.dot(a.values[:3])
+        msg = "matrices are not aligned"
+        with pytest.raises(ValueError, match=msg):
+            a.dot(b.T)
diff --git a/pandas/pandas/tests/series/methods/test_nlargest.py b/pandas/pandas/tests/series/methods/test_nlargest.py
new file mode 100644
index 0000000000000..67ba1d7ca51b7
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_nlargest.py
@@ -0,0 +1,202 @@
+"""
+Note: for naming purposes, most tests are title with as e.g. "test_nlargest_foo"
+but are implicitly also testing nsmallest_foo.
+"""
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+
+def assert_check_nselect_boundary(vals, dtype, method):
+    # helper function for 'test_boundary_{dtype}' tests
+    ser = Series(vals, dtype=dtype)
+    result = getattr(ser, method)(3)
+    expected_idxr = range(3) if method == "nsmallest" else range(3, 0, -1)
+    expected = ser.loc[expected_idxr]
+    tm.assert_series_equal(result, expected)
+
+
+class TestSeriesNLargestNSmallest:
+    @pytest.mark.parametrize(
+        "r",
+        [
+            Series([3.0, 2, 1, 2, "5"], dtype="object"),
+            Series([3.0, 2, 1, 2, 5], dtype="object"),
+            # not supported on some archs
+            # Series([3., 2, 1, 2, 5], dtype='complex256'),
+            Series([3.0, 2, 1, 2, 5], dtype="complex128"),
+            Series(list("abcde")),
+            Series(list("abcde"), dtype="category"),
+        ],
+    )
+    @pytest.mark.parametrize("method", ["nlargest", "nsmallest"])
+    @pytest.mark.parametrize("arg", [2, 5, 0, -1])
+    def test_nlargest_error(self, r, method, arg):
+        dt = r.dtype
+        msg = f"Cannot use method 'n(largest|smallest)' with dtype {dt}"
+        with pytest.raises(TypeError, match=msg):
+            getattr(r, method)(arg)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            pd.to_datetime(["2003", "2002", "2001", "2002", "2005"]),
+            pd.to_datetime(["2003", "2002", "2001", "2002", "2005"], utc=True),
+            pd.to_timedelta(["3D", "2D", "1D", "2D", "5D"]),
+            np.array([3, 2, 1, 2, 5], dtype="int8"),
+            np.array([3, 2, 1, 2, 5], dtype="int16"),
+            np.array([3, 2, 1, 2, 5], dtype="int32"),
+            np.array([3, 2, 1, 2, 5], dtype="int64"),
+            np.array([3, 2, 1, 2, 5], dtype="uint8"),
+            np.array([3, 2, 1, 2, 5], dtype="uint16"),
+            np.array([3, 2, 1, 2, 5], dtype="uint32"),
+            np.array([3, 2, 1, 2, 5], dtype="uint64"),
+            np.array([3, 2, 1, 2, 5], dtype="float32"),
+            np.array([3, 2, 1, 2, 5], dtype="float64"),
+        ],
+    )
+    def test_nsmallest_nlargest(self, data):
+        # float, int, datetime64 (use i8), timedelts64 (same),
+        # object that are numbers, object that are strings
+        ser = Series(data)
+
+        tm.assert_series_equal(ser.nsmallest(2), ser.iloc[[2, 1]])
+        tm.assert_series_equal(ser.nsmallest(2, keep="last"), ser.iloc[[2, 3]])
+
+        empty = ser.iloc[0:0]
+        tm.assert_series_equal(ser.nsmallest(0), empty)
+        tm.assert_series_equal(ser.nsmallest(-1), empty)
+        tm.assert_series_equal(ser.nlargest(0), empty)
+        tm.assert_series_equal(ser.nlargest(-1), empty)
+
+        tm.assert_series_equal(ser.nsmallest(len(ser)), ser.sort_values())
+        tm.assert_series_equal(ser.nsmallest(len(ser) + 1), ser.sort_values())
+        tm.assert_series_equal(ser.nlargest(len(ser)), ser.iloc[[4, 0, 1, 3, 2]])
+        tm.assert_series_equal(ser.nlargest(len(ser) + 1), ser.iloc[[4, 0, 1, 3, 2]])
+
+    def test_nlargest_misc(self):
+        ser = Series([3.0, np.nan, 1, 2, 5])
+        result = ser.nlargest()
+        expected = ser.iloc[[4, 0, 3, 2, 1]]
+        tm.assert_series_equal(result, expected)
+        result = ser.nsmallest()
+        expected = ser.iloc[[2, 3, 0, 4, 1]]
+        tm.assert_series_equal(result, expected)
+
+        msg = 'keep must be either "first", "last"'
+        with pytest.raises(ValueError, match=msg):
+            ser.nsmallest(keep="invalid")
+        with pytest.raises(ValueError, match=msg):
+            ser.nlargest(keep="invalid")
+
+        # GH#15297
+        ser = Series([1] * 5, index=[1, 2, 3, 4, 5])
+        expected_first = Series([1] * 3, index=[1, 2, 3])
+        expected_last = Series([1] * 3, index=[5, 4, 3])
+
+        result = ser.nsmallest(3)
+        tm.assert_series_equal(result, expected_first)
+
+        result = ser.nsmallest(3, keep="last")
+        tm.assert_series_equal(result, expected_last)
+
+        result = ser.nlargest(3)
+        tm.assert_series_equal(result, expected_first)
+
+        result = ser.nlargest(3, keep="last")
+        tm.assert_series_equal(result, expected_last)
+
+    @pytest.mark.parametrize("n", range(1, 5))
+    def test_nlargest_n(self, n):
+        # GH 13412
+        ser = Series([1, 4, 3, 2], index=[0, 0, 1, 1])
+        result = ser.nlargest(n)
+        expected = ser.sort_values(ascending=False).head(n)
+        tm.assert_series_equal(result, expected)
+
+        result = ser.nsmallest(n)
+        expected = ser.sort_values().head(n)
+        tm.assert_series_equal(result, expected)
+
+    def test_nlargest_boundary_integer(self, nselect_method, any_int_numpy_dtype):
+        # GH#21426
+        dtype_info = np.iinfo(any_int_numpy_dtype)
+        min_val, max_val = dtype_info.min, dtype_info.max
+        vals = [min_val, min_val + 1, max_val - 1, max_val]
+        assert_check_nselect_boundary(vals, any_int_numpy_dtype, nselect_method)
+
+    def test_nlargest_boundary_float(self, nselect_method, float_numpy_dtype):
+        # GH#21426
+        dtype_info = np.finfo(float_numpy_dtype)
+        min_val, max_val = dtype_info.min, dtype_info.max
+        min_2nd, max_2nd = np.nextafter([min_val, max_val], 0, dtype=float_numpy_dtype)
+        vals = [min_val, min_2nd, max_2nd, max_val]
+        assert_check_nselect_boundary(vals, float_numpy_dtype, nselect_method)
+
+    @pytest.mark.parametrize("dtype", ["datetime64[ns]", "timedelta64[ns]"])
+    def test_nlargest_boundary_datetimelike(self, nselect_method, dtype):
+        # GH#21426
+        # use int64 bounds and +1 to min_val since true minimum is NaT
+        # (include min_val/NaT at end to maintain same expected_idxr)
+        dtype_info = np.iinfo("int64")
+        min_val, max_val = dtype_info.min, dtype_info.max
+        vals = [min_val + 1, min_val + 2, max_val - 1, max_val, min_val]
+        assert_check_nselect_boundary(vals, dtype, nselect_method)
+
+    def test_nlargest_duplicate_keep_all_ties(self):
+        # see GH#16818
+        ser = Series([10, 9, 8, 7, 7, 7, 7, 6])
+        result = ser.nlargest(4, keep="all")
+        expected = Series([10, 9, 8, 7, 7, 7, 7])
+        tm.assert_series_equal(result, expected)
+
+        result = ser.nsmallest(2, keep="all")
+        expected = Series([6, 7, 7, 7, 7], index=[7, 3, 4, 5, 6])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data,expected", [([True, False], [True]), ([True, False, True, True], [True])]
+    )
+    def test_nlargest_boolean(self, data, expected):
+        # GH#26154 : ensure True > False
+        ser = Series(data)
+        result = ser.nlargest(1)
+        expected = Series(expected)
+        tm.assert_series_equal(result, expected)
+
+    def test_nlargest_nullable(self, any_numeric_ea_dtype):
+        # GH#42816
+        dtype = any_numeric_ea_dtype
+        if dtype.startswith("UInt"):
+            # Can't cast from negative float to uint on some platforms
+            arr = np.random.default_rng(2).integers(1, 10, 10)
+        else:
+            arr = np.random.default_rng(2).standard_normal(10)
+        arr = arr.astype(dtype.lower(), copy=False)
+
+        ser = Series(arr, dtype=dtype, copy=True)
+        ser[1] = pd.NA
+        result = ser.nlargest(5)
+
+        expected = (
+            Series(np.delete(arr, 1), index=ser.index.delete(1))
+            .nlargest(5)
+            .astype(dtype)
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_nsmallest_nan_when_keep_is_all(self):
+        # GH#46589
+        s = Series([1, 2, 3, 3, 3, None])
+        result = s.nsmallest(3, keep="all")
+        expected = Series([1.0, 2.0, 3.0, 3.0, 3.0])
+        tm.assert_series_equal(result, expected)
+
+        s = Series([1, 2, None, None, None])
+        result = s.nsmallest(3, keep="all")
+        expected = Series([1, 2, None, None, None])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_nunique.py b/pandas/pandas/tests/series/methods/test_nunique.py
new file mode 100644
index 0000000000000..826132eb28162
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_nunique.py
@@ -0,0 +1,24 @@
+import numpy as np
+
+from pandas import (
+    Categorical,
+    Series,
+)
+
+
+def test_nunique():
+    # basics.rst doc example
+    series = Series(np.random.default_rng(2).standard_normal(500))
+    series[20:500] = np.nan
+    series[10:20] = 5000
+    result = series.nunique()
+    assert result == 11
+
+
+def test_nunique_categorical():
+    # GH#18051
+    ser = Series(Categorical([]))
+    assert ser.nunique() == 0
+
+    ser = Series(Categorical([np.nan]))
+    assert ser.nunique() == 0
diff --git a/pandas/pandas/tests/series/methods/test_pct_change.py b/pandas/pandas/tests/series/methods/test_pct_change.py
new file mode 100644
index 0000000000000..6279cf64818b8
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_pct_change.py
@@ -0,0 +1,76 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesPctChange:
+    def test_pct_change(self, datetime_series):
+        rs = datetime_series.pct_change()
+        tm.assert_series_equal(rs, datetime_series / datetime_series.shift(1) - 1)
+
+        rs = datetime_series.pct_change(2)
+        filled = datetime_series.ffill()
+        tm.assert_series_equal(rs, filled / filled.shift(2) - 1)
+
+        rs = datetime_series.pct_change(freq="5D")
+        filled = datetime_series.ffill()
+        tm.assert_series_equal(
+            rs, (filled / filled.shift(freq="5D") - 1).reindex_like(filled)
+        )
+
+    def test_pct_change_with_duplicate_axis(self):
+        # GH#28664
+        common_idx = date_range("2019-11-14", periods=5, freq="D")
+        result = Series(range(5), common_idx).pct_change(freq="B")
+
+        # the reason that the expected should be like this is documented at PR 28681
+        expected = Series([np.nan, np.inf, np.nan, np.nan, 3.0], common_idx)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_pct_change_shift_over_nas(self):
+        s = Series([1.0, 1.5, np.nan, 2.5, 3.0])
+        chg = s.pct_change()
+        expected = Series([np.nan, 0.5, np.nan, np.nan, 0.2])
+        tm.assert_series_equal(chg, expected)
+
+    @pytest.mark.parametrize("freq, periods", [("5B", 5), ("3B", 3), ("14B", 14)])
+    def test_pct_change_periods_freq(self, freq, periods, datetime_series):
+        # GH#7292
+        rs_freq = datetime_series.pct_change(freq=freq)
+        rs_periods = datetime_series.pct_change(periods)
+        tm.assert_series_equal(rs_freq, rs_periods)
+
+        empty_ts = Series(index=datetime_series.index, dtype=object)
+        rs_freq = empty_ts.pct_change(freq=freq)
+        rs_periods = empty_ts.pct_change(periods)
+        tm.assert_series_equal(rs_freq, rs_periods)
+
+
+def test_pct_change_with_duplicated_indices():
+    # GH30463
+    s = Series([np.nan, 1, 2, 3, 9, 18], index=["a", "b"] * 3)
+    result = s.pct_change()
+    expected = Series([np.nan, np.nan, 1.0, 0.5, 2.0, 1.0], index=["a", "b"] * 3)
+    tm.assert_series_equal(result, expected)
+
+
+def test_pct_change_no_warning_na_beginning():
+    # GH#54981
+    ser = Series([None, None, 1, 2, 3])
+    result = ser.pct_change()
+    expected = Series([np.nan, np.nan, np.nan, 1, 0.5])
+    tm.assert_series_equal(result, expected)
+
+
+def test_pct_change_empty():
+    # GH 57056
+    ser = Series([], dtype="float64")
+    expected = ser.copy()
+    result = ser.pct_change(periods=0)
+    tm.assert_series_equal(expected, result)
diff --git a/pandas/pandas/tests/series/methods/test_pop.py b/pandas/pandas/tests/series/methods/test_pop.py
new file mode 100644
index 0000000000000..7453f98ab3735
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_pop.py
@@ -0,0 +1,13 @@
+from pandas import Series
+import pandas._testing as tm
+
+
+def test_pop():
+    # GH#6600
+    ser = Series([0, 4, 0], index=["A", "B", "C"], name=4)
+
+    result = ser.pop("B")
+    assert result == 4
+
+    expected = Series([0, 0], index=["A", "C"], name=4)
+    tm.assert_series_equal(ser, expected)
diff --git a/pandas/pandas/tests/series/methods/test_quantile.py b/pandas/pandas/tests/series/methods/test_quantile.py
new file mode 100644
index 0000000000000..fa0563271d7df
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_quantile.py
@@ -0,0 +1,247 @@
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.common import is_integer
+
+import pandas as pd
+from pandas import (
+    Index,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.indexes.datetimes import Timestamp
+
+
+class TestSeriesQuantile:
+    def test_quantile(self, datetime_series):
+        q = datetime_series.quantile(0.1)
+        assert q == np.percentile(datetime_series.dropna(), 10)
+
+        q = datetime_series.quantile(0.9)
+        assert q == np.percentile(datetime_series.dropna(), 90)
+
+        # object dtype
+        q = Series(datetime_series, dtype=object).quantile(0.9)
+        assert q == np.percentile(datetime_series.dropna(), 90)
+
+        # datetime64[ns] dtype
+        dts = datetime_series.index.to_series()
+        q = dts.quantile(0.2)
+        assert q == Timestamp("2000-01-10 19:12:00")
+
+        # timedelta64[ns] dtype
+        tds = dts.diff()
+        q = tds.quantile(0.25)
+        assert q == pd.to_timedelta("24:00:00")
+
+        # GH7661
+        result = Series([np.timedelta64("NaT")]).sum()
+        assert result == pd.Timedelta(0)
+
+        msg = "percentiles should all be in the interval \\[0, 1\\]"
+        for invalid in [-1, 2, [0.5, -1], [0.5, 2]]:
+            with pytest.raises(ValueError, match=msg):
+                datetime_series.quantile(invalid)
+
+        s = Series(np.random.default_rng(2).standard_normal(100))
+        percentile_array = [-0.5, 0.25, 1.5]
+        with pytest.raises(ValueError, match=msg):
+            s.quantile(percentile_array)
+
+    def test_quantile_multi(self, datetime_series, unit):
+        datetime_series.index = datetime_series.index.as_unit(unit)
+        qs = [0.1, 0.9]
+        result = datetime_series.quantile(qs)
+        expected = Series(
+            [
+                np.percentile(datetime_series.dropna(), 10),
+                np.percentile(datetime_series.dropna(), 90),
+            ],
+            index=qs,
+            name=datetime_series.name,
+        )
+        tm.assert_series_equal(result, expected)
+
+        dts = datetime_series.index.to_series()
+        dts.name = "xxx"
+        result = dts.quantile((0.2, 0.2))
+        expected = Series(
+            [Timestamp("2000-01-10 19:12:00"), Timestamp("2000-01-10 19:12:00")],
+            index=[0.2, 0.2],
+            name="xxx",
+            dtype=f"M8[{unit}]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = datetime_series.quantile([])
+        expected = Series(
+            [], name=datetime_series.name, index=Index([], dtype=float), dtype="float64"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_interpolation(self, datetime_series):
+        # see gh-10174
+
+        # interpolation = linear (default case)
+        q = datetime_series.quantile(0.1, interpolation="linear")
+        assert q == np.percentile(datetime_series.dropna(), 10)
+        q1 = datetime_series.quantile(0.1)
+        assert q1 == np.percentile(datetime_series.dropna(), 10)
+
+        # test with and without interpolation keyword
+        assert q == q1
+
+    def test_quantile_interpolation_dtype(self):
+        # GH #10174
+
+        # interpolation = linear (default case)
+        q = Series([1, 3, 4]).quantile(0.5, interpolation="lower")
+        assert q == np.percentile(np.array([1, 3, 4]), 50)
+        assert is_integer(q)
+
+        q = Series([1, 3, 4]).quantile(0.5, interpolation="higher")
+        assert q == np.percentile(np.array([1, 3, 4]), 50)
+        assert is_integer(q)
+
+    def test_quantile_nan(self):
+        # GH 13098
+        ser = Series([1, 2, 3, 4, np.nan])
+        result = ser.quantile(0.5)
+        expected = 2.5
+        assert result == expected
+
+        # all nan/empty
+        s1 = Series([], dtype=object)
+        cases = [s1, Series([np.nan, np.nan])]
+
+        for ser in cases:
+            res = ser.quantile(0.5)
+            assert np.isnan(res)
+
+            res = ser.quantile([0.5])
+            tm.assert_series_equal(res, Series([np.nan], index=[0.5]))
+
+            res = ser.quantile([0.2, 0.3])
+            tm.assert_series_equal(res, Series([np.nan, np.nan], index=[0.2, 0.3]))
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            [
+                Timestamp("2011-01-01"),
+                Timestamp("2011-01-02"),
+                Timestamp("2011-01-03"),
+            ],
+            [
+                Timestamp("2011-01-01", tz="US/Eastern"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+                Timestamp("2011-01-03", tz="US/Eastern"),
+            ],
+            [pd.Timedelta("1 days"), pd.Timedelta("2 days"), pd.Timedelta("3 days")],
+            # NaT
+            [
+                Timestamp("2011-01-01"),
+                Timestamp("2011-01-02"),
+                Timestamp("2011-01-03"),
+                pd.NaT,
+            ],
+            [
+                Timestamp("2011-01-01", tz="US/Eastern"),
+                Timestamp("2011-01-02", tz="US/Eastern"),
+                Timestamp("2011-01-03", tz="US/Eastern"),
+                pd.NaT,
+            ],
+            [
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+                pd.Timedelta("3 days"),
+                pd.NaT,
+            ],
+        ],
+    )
+    def test_quantile_box(self, case):
+        ser = Series(case, name="XXX")
+        res = ser.quantile(0.5)
+        assert res == case[1]
+
+        res = ser.quantile([0.5])
+        exp = Series([case[1]], index=[0.5], name="XXX")
+        tm.assert_series_equal(res, exp)
+
+    def test_datetime_timedelta_quantiles(self):
+        # covers #9694
+        assert pd.isna(Series([], dtype="M8[ns]").quantile(0.5))
+        assert pd.isna(Series([], dtype="m8[ns]").quantile(0.5))
+
+    def test_quantile_nat(self):
+        res = Series([pd.NaT, pd.NaT]).quantile(0.5)
+        assert res is pd.NaT
+
+        res = Series([pd.NaT, pd.NaT]).quantile([0.5])
+        tm.assert_series_equal(res, Series([pd.NaT], index=[0.5]))
+
+    @pytest.mark.parametrize(
+        "values, dtype",
+        [([0, 0, 0, 1, 2, 3], "Sparse[int]"), ([0.0, None, 1.0, 2.0], "Sparse[float]")],
+    )
+    def test_quantile_sparse(self, values, dtype):
+        ser = Series(values, dtype=dtype)
+        result = ser.quantile([0.5])
+        expected = Series(np.asarray(ser)).quantile([0.5]).astype("Sparse[float]")
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_empty_float64(self):
+        # floats
+        ser = Series([], dtype="float64")
+
+        res = ser.quantile(0.5)
+        assert np.isnan(res)
+
+        res = ser.quantile([0.5])
+        exp = Series([np.nan], index=[0.5])
+        tm.assert_series_equal(res, exp)
+
+    def test_quantile_empty_int64(self):
+        # int
+        ser = Series([], dtype="int64")
+
+        res = ser.quantile(0.5)
+        assert np.isnan(res)
+
+        res = ser.quantile([0.5])
+        exp = Series([np.nan], index=[0.5])
+        tm.assert_series_equal(res, exp)
+
+    def test_quantile_empty_dt64(self):
+        # datetime
+        ser = Series([], dtype="datetime64[ns]")
+
+        res = ser.quantile(0.5)
+        assert res is pd.NaT
+
+        res = ser.quantile([0.5])
+        exp = Series([pd.NaT], index=[0.5], dtype=ser.dtype)
+        tm.assert_series_equal(res, exp)
+
+    @pytest.mark.parametrize("dtype", [int, float, "Int64"])
+    def test_quantile_dtypes(self, dtype):
+        result = Series([1, 2, 3], dtype=dtype).quantile(np.arange(0, 1, 0.25))
+        expected = Series(np.arange(1, 3, 0.5), index=np.arange(0, 1, 0.25))
+        if dtype == "Int64":
+            expected = expected.astype("Float64")
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_all_na(self, any_int_ea_dtype):
+        # GH#50681
+        ser = Series([pd.NA, pd.NA], dtype=any_int_ea_dtype)
+        with tm.assert_produces_warning(None):
+            result = ser.quantile([0.1, 0.5])
+        expected = Series([pd.NA, pd.NA], dtype=any_int_ea_dtype, index=[0.1, 0.5])
+        tm.assert_series_equal(result, expected)
+
+    def test_quantile_dtype_size(self, any_int_ea_dtype):
+        # GH#50681
+        ser = Series([pd.NA, pd.NA, 1], dtype=any_int_ea_dtype)
+        result = ser.quantile([0.1, 0.5])
+        expected = Series([1, 1], dtype=any_int_ea_dtype, index=[0.1, 0.5])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_rank.py b/pandas/pandas/tests/series/methods/test_rank.py
new file mode 100644
index 0000000000000..7c6a7893ba3a0
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_rank.py
@@ -0,0 +1,570 @@
+from itertools import chain
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs.algos import (
+    Infinity,
+    NegInfinity,
+)
+import pandas.util._test_decorators as td
+
+from pandas import (
+    NA,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.types import CategoricalDtype
+
+
+@pytest.fixture
+def ser():
+    return Series([1, 3, 4, 2, np.nan, 2, 1, 5, np.nan, 3])
+
+
+@pytest.fixture(
+    params=[
+        ["average", np.array([1.5, 5.5, 7.0, 3.5, np.nan, 3.5, 1.5, 8.0, np.nan, 5.5])],
+        ["min", np.array([1, 5, 7, 3, np.nan, 3, 1, 8, np.nan, 5])],
+        ["max", np.array([2, 6, 7, 4, np.nan, 4, 2, 8, np.nan, 6])],
+        ["first", np.array([1, 5, 7, 3, np.nan, 4, 2, 8, np.nan, 6])],
+        ["dense", np.array([1, 3, 4, 2, np.nan, 2, 1, 5, np.nan, 3])],
+    ],
+    ids=lambda x: x[0],
+)
+def results(request):
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        "object",
+        "float64",
+        "int64",
+        "Float64",
+        "Int64",
+        pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        pytest.param("int64[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        pytest.param("string[pyarrow]", marks=td.skip_if_no("pyarrow")),
+        "string[python]",
+        "str",
+    ]
+)
+def dtype(request):
+    return request.param
+
+
+def expected_dtype(dtype, method, pct=False):
+    exp_dtype = "float64"
+    # elif dtype in ["Int64", "Float64", "string[pyarrow]", "string[python]"]:
+    if dtype in ["string[pyarrow]"]:
+        exp_dtype = "Float64"
+    elif dtype in ["float64[pyarrow]", "int64[pyarrow]"]:
+        if method == "average" or pct:
+            exp_dtype = "double[pyarrow]"
+        else:
+            exp_dtype = "uint64[pyarrow]"
+
+    return exp_dtype
+
+
+class TestSeriesRank:
+    def test_rank(self, datetime_series):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        datetime_series[::2] = np.nan
+        datetime_series[:10:3] = 4.0
+
+        ranks = datetime_series.rank()
+        oranks = datetime_series.astype("O").rank()
+
+        tm.assert_series_equal(ranks, oranks)
+
+        mask = np.isnan(datetime_series)
+        filled = datetime_series.fillna(np.inf)
+
+        # rankdata returns a ndarray
+        exp = Series(sp_stats.rankdata(filled), index=filled.index, name="ts")
+        exp[mask] = np.nan
+
+        tm.assert_series_equal(ranks, exp)
+
+        iseries = Series(np.arange(5).repeat(2))
+
+        iranks = iseries.rank()
+        exp = iseries.astype(float).rank()
+        tm.assert_series_equal(iranks, exp)
+        iseries = Series(np.arange(5)) + 1.0
+        exp = iseries / 5.0
+        iranks = iseries.rank(pct=True)
+
+        tm.assert_series_equal(iranks, exp)
+
+        iseries = Series(np.repeat(1, 100))
+        exp = Series(np.repeat(0.505, 100))
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        # Explicit cast to float to avoid implicit cast when setting nan
+        iseries = iseries.astype("float")
+        iseries[1] = np.nan
+        exp = Series(np.repeat(50.0 / 99.0, 100))
+        exp[1] = np.nan
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        iseries = Series(np.arange(5)) + 1.0
+        iseries[4] = np.nan
+        exp = iseries / 4.0
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        iseries = Series(np.repeat(np.nan, 100))
+        exp = iseries
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        # Explicit cast to float to avoid implicit cast when setting nan
+        iseries = Series(np.arange(5), dtype="float") + 1
+        iseries[4] = np.nan
+        exp = iseries / 4.0
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        rng = date_range("1/1/1990", periods=5)
+        # Explicit cast to float to avoid implicit cast when setting nan
+        iseries = Series(np.arange(5), rng, dtype="float") + 1
+        iseries.iloc[4] = np.nan
+        exp = iseries / 4.0
+        iranks = iseries.rank(pct=True)
+        tm.assert_series_equal(iranks, exp)
+
+        iseries = Series([1e-50, 1e-100, 1e-20, 1e-2, 1e-20 + 1e-30, 1e-1])
+        exp = Series([2, 1, 3, 5, 4, 6.0])
+        iranks = iseries.rank()
+        tm.assert_series_equal(iranks, exp)
+
+        # GH 5968
+        iseries = Series(["3 day", "1 day 10m", "-2 day", NaT], dtype="m8[ns]")
+        exp = Series([3, 2, 1, np.nan])
+        iranks = iseries.rank()
+        tm.assert_series_equal(iranks, exp)
+
+        values = np.array(
+            [-50, -1, -1e-20, -1e-25, -1e-50, 0, 1e-40, 1e-20, 1e-10, 2, 40],
+            dtype="float64",
+        )
+        random_order = np.random.default_rng(2).permutation(len(values))
+        iseries = Series(values[random_order])
+        exp = Series(random_order + 1.0, dtype="float64")
+        iranks = iseries.rank()
+        tm.assert_series_equal(iranks, exp)
+
+    def test_rank_categorical(self):
+        # GH issue #15420 rank incorrectly orders ordered categories
+
+        # Test ascending/descending ranking for ordered categoricals
+        exp = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0])
+        exp_desc = Series([6.0, 5.0, 4.0, 3.0, 2.0, 1.0])
+        ordered = Series(
+            ["first", "second", "third", "fourth", "fifth", "sixth"]
+        ).astype(
+            CategoricalDtype(
+                categories=["first", "second", "third", "fourth", "fifth", "sixth"],
+                ordered=True,
+            )
+        )
+        tm.assert_series_equal(ordered.rank(), exp)
+        tm.assert_series_equal(ordered.rank(ascending=False), exp_desc)
+
+        # Unordered categoricals should be ranked as objects
+        unordered = Series(
+            ["first", "second", "third", "fourth", "fifth", "sixth"]
+        ).astype(
+            CategoricalDtype(
+                categories=["first", "second", "third", "fourth", "fifth", "sixth"],
+                ordered=False,
+            )
+        )
+        exp_unordered = Series([2.0, 4.0, 6.0, 3.0, 1.0, 5.0])
+        res = unordered.rank()
+        tm.assert_series_equal(res, exp_unordered)
+
+        unordered1 = Series([1, 2, 3, 4, 5, 6]).astype(
+            CategoricalDtype([1, 2, 3, 4, 5, 6], False)
+        )
+        exp_unordered1 = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0])
+        res1 = unordered1.rank()
+        tm.assert_series_equal(res1, exp_unordered1)
+
+        # Test na_option for rank data
+        na_ser = Series(
+            ["first", "second", "third", "fourth", "fifth", "sixth", np.nan]
+        ).astype(
+            CategoricalDtype(
+                ["first", "second", "third", "fourth", "fifth", "sixth", "seventh"],
+                True,
+            )
+        )
+
+        exp_top = Series([2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 1.0])
+        exp_bot = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0])
+        exp_keep = Series([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, np.nan])
+
+        tm.assert_series_equal(na_ser.rank(na_option="top"), exp_top)
+        tm.assert_series_equal(na_ser.rank(na_option="bottom"), exp_bot)
+        tm.assert_series_equal(na_ser.rank(na_option="keep"), exp_keep)
+
+        # Test na_option for rank data with ascending False
+        exp_top = Series([7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0])
+        exp_bot = Series([6.0, 5.0, 4.0, 3.0, 2.0, 1.0, 7.0])
+        exp_keep = Series([6.0, 5.0, 4.0, 3.0, 2.0, 1.0, np.nan])
+
+        tm.assert_series_equal(na_ser.rank(na_option="top", ascending=False), exp_top)
+        tm.assert_series_equal(
+            na_ser.rank(na_option="bottom", ascending=False), exp_bot
+        )
+        tm.assert_series_equal(na_ser.rank(na_option="keep", ascending=False), exp_keep)
+
+        # Test invalid values for na_option
+        msg = "na_option must be one of 'keep', 'top', or 'bottom'"
+
+        with pytest.raises(ValueError, match=msg):
+            na_ser.rank(na_option="bad", ascending=False)
+
+        # invalid type
+        with pytest.raises(ValueError, match=msg):
+            na_ser.rank(na_option=True, ascending=False)
+
+        # Test with pct=True
+        na_ser = Series(["first", "second", "third", "fourth", np.nan]).astype(
+            CategoricalDtype(["first", "second", "third", "fourth"], True)
+        )
+        exp_top = Series([0.4, 0.6, 0.8, 1.0, 0.2])
+        exp_bot = Series([0.2, 0.4, 0.6, 0.8, 1.0])
+        exp_keep = Series([0.25, 0.5, 0.75, 1.0, np.nan])
+
+        tm.assert_series_equal(na_ser.rank(na_option="top", pct=True), exp_top)
+        tm.assert_series_equal(na_ser.rank(na_option="bottom", pct=True), exp_bot)
+        tm.assert_series_equal(na_ser.rank(na_option="keep", pct=True), exp_keep)
+
+    def test_rank_nullable_integer(self):
+        # GH 56976
+        exp = Series([np.nan, 2, np.nan, 3, 3, 2, 3, 1])
+        exp = exp.astype("Int64")
+        result = exp.rank(na_option="keep")
+
+        expected = Series([np.nan, 2.5, np.nan, 5.0, 5.0, 2.5, 5.0, 1.0])
+
+        tm.assert_series_equal(result, expected)
+
+    def test_rank_signature(self):
+        s = Series([0, 1])
+        s.rank(method="average")
+        msg = "No axis named average for object type Series"
+        with pytest.raises(ValueError, match=msg):
+            s.rank("average")
+
+    def test_rank_tie_methods(self, ser, results, dtype, using_infer_string):
+        method, exp = results
+        if dtype == "int64" or (not using_infer_string and dtype == "str"):
+            pytest.skip("int64/str does not support NaN")
+
+        ser = ser if dtype is None else ser.astype(dtype)
+        result = ser.rank(method=method)
+        tm.assert_series_equal(result, Series(exp, dtype=expected_dtype(dtype, method)))
+
+    @pytest.mark.parametrize("na_option", ["top", "bottom", "keep"])
+    @pytest.mark.parametrize(
+        "dtype, na_value, pos_inf, neg_inf",
+        [
+            ("object", None, Infinity(), NegInfinity()),
+            ("float64", np.nan, np.inf, -np.inf),
+            ("Float64", NA, np.inf, -np.inf),
+            pytest.param(
+                "float64[pyarrow]",
+                NA,
+                np.inf,
+                -np.inf,
+                marks=td.skip_if_no("pyarrow"),
+            ),
+        ],
+    )
+    def test_rank_tie_methods_on_infs_nans(
+        self, rank_method, na_option, ascending, dtype, na_value, pos_inf, neg_inf
+    ):
+        pytest.importorskip("scipy")
+        if dtype == "float64[pyarrow]":
+            if rank_method == "average":
+                exp_dtype = "float64[pyarrow]"
+            else:
+                exp_dtype = "uint64[pyarrow]"
+        else:
+            exp_dtype = "float64"
+
+        chunk = 3
+        in_arr = [neg_inf] * chunk + [na_value] * chunk + [pos_inf] * chunk
+        iseries = Series(in_arr, dtype=dtype)
+        exp_ranks = {
+            "average": ([2, 2, 2], [5, 5, 5], [8, 8, 8]),
+            "min": ([1, 1, 1], [4, 4, 4], [7, 7, 7]),
+            "max": ([3, 3, 3], [6, 6, 6], [9, 9, 9]),
+            "first": ([1, 2, 3], [4, 5, 6], [7, 8, 9]),
+            "dense": ([1, 1, 1], [2, 2, 2], [3, 3, 3]),
+        }
+        ranks = exp_ranks[rank_method]
+        if na_option == "top":
+            order = [ranks[1], ranks[0], ranks[2]]
+        elif na_option == "bottom":
+            order = [ranks[0], ranks[2], ranks[1]]
+        else:
+            order = [ranks[0], [np.nan] * chunk, ranks[1]]
+        expected = order if ascending else order[::-1]
+        expected = list(chain.from_iterable(expected))
+        result = iseries.rank(
+            method=rank_method, na_option=na_option, ascending=ascending
+        )
+        tm.assert_series_equal(result, Series(expected, dtype=exp_dtype))
+
+    def test_rank_desc_mix_nans_infs(self):
+        # GH 19538
+        # check descending ranking when mix nans and infs
+        iseries = Series([1, np.nan, np.inf, -np.inf, 25])
+        result = iseries.rank(ascending=False)
+        exp = Series([3, np.nan, 1, 4, 2], dtype="float64")
+        tm.assert_series_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "op, value",
+        [
+            [operator.add, 0],
+            [operator.add, 1e6],
+            [operator.mul, 1e-6],
+        ],
+    )
+    def test_rank_methods_series(self, rank_method, op, value):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        xs = np.random.default_rng(2).standard_normal(9)
+        xs = np.concatenate([xs[i:] for i in range(0, 9, 2)])  # add duplicates
+        np.random.default_rng(2).shuffle(xs)
+
+        index = [chr(ord("a") + i) for i in range(len(xs))]
+        vals = op(xs, value)
+        ts = Series(vals, index=index)
+        result = ts.rank(method=rank_method)
+        sprank = sp_stats.rankdata(
+            vals, rank_method if rank_method != "first" else "ordinal"
+        )
+        expected = Series(sprank, index=index).astype("float64")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ser, exp",
+        [
+            ([1], [1]),
+            ([2], [1]),
+            ([0], [1]),
+            ([2, 2], [1, 1]),
+            ([1, 2, 3], [1, 2, 3]),
+            ([4, 2, 1], [3, 2, 1]),
+            ([1, 1, 5, 5, 3], [1, 1, 3, 3, 2]),
+            ([-5, -4, -3, -2, -1], [1, 2, 3, 4, 5]),
+        ],
+    )
+    def test_rank_dense_method(self, dtype, ser, exp):
+        if ser[0] < 0 and dtype.startswith("str"):
+            exp = exp[::-1]
+        s = Series(ser).astype(dtype)
+        result = s.rank(method="dense")
+        expected = Series(exp).astype(expected_dtype(dtype, "dense"))
+        tm.assert_series_equal(result, expected)
+
+    def test_rank_descending(self, ser, results, dtype, using_infer_string):
+        method, _ = results
+        if dtype == "int64" or (not using_infer_string and dtype == "str"):
+            s = ser.dropna()
+        else:
+            s = ser.astype(dtype)
+
+        res = s.rank(ascending=False)
+        if dtype.startswith("str"):
+            expected = (s.astype("float64").max() - s.astype("float64")).rank()
+        else:
+            expected = (s.max() - s).rank()
+        tm.assert_series_equal(res, expected.astype(expected_dtype(dtype, "average")))
+
+        if dtype.startswith("str"):
+            expected = (s.astype("float64").max() - s.astype("float64")).rank(
+                method=method
+            )
+        else:
+            expected = (s.max() - s).rank(method=method)
+        res2 = s.rank(method=method, ascending=False)
+        tm.assert_series_equal(res2, expected.astype(expected_dtype(dtype, method)))
+
+    def test_rank_int(self, ser, results):
+        method, exp = results
+        s = ser.dropna().astype("i8")
+
+        result = s.rank(method=method)
+        expected = Series(exp).dropna()
+        expected.index = result.index
+        tm.assert_series_equal(result, expected)
+
+    def test_rank_object_bug(self):
+        # GH 13445
+
+        # smoke tests
+        Series([np.nan] * 32).astype(object).rank(ascending=True)
+        Series([np.nan] * 32).astype(object).rank(ascending=False)
+
+    def test_rank_modify_inplace(self):
+        # GH 18521
+        # Check rank does not mutate series
+        s = Series([Timestamp("2017-01-05 10:20:27.569000"), NaT])
+        expected = s.copy()
+
+        s.rank()
+        result = s
+        tm.assert_series_equal(result, expected)
+
+    def test_rank_ea_small_values(self):
+        # GH#52471
+        ser = Series(
+            [5.4954145e29, -9.791984e-21, 9.3715776e-26, NA, 1.8790257e-28],
+            dtype="Float64",
+        )
+        result = ser.rank(method="min")
+        expected = Series([4, 1, 3, np.nan, 2])
+        tm.assert_series_equal(result, expected)
+
+
+# GH15630, pct should be on 100% basis when method='dense'
+
+
+@pytest.mark.parametrize(
+    "ser, exp",
+    [
+        ([1], [1.0]),
+        ([1, 2], [1.0 / 2, 2.0 / 2]),
+        ([2, 2], [1.0, 1.0]),
+        ([1, 2, 3], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([1, 2, 2], [1.0 / 2, 2.0 / 2, 2.0 / 2]),
+        ([4, 2, 1], [3.0 / 3, 2.0 / 3, 1.0 / 3]),
+        ([1, 1, 5, 5, 3], [1.0 / 3, 1.0 / 3, 3.0 / 3, 3.0 / 3, 2.0 / 3]),
+        ([1, 1, 3, 3, 5, 5], [1.0 / 3, 1.0 / 3, 2.0 / 3, 2.0 / 3, 3.0 / 3, 3.0 / 3]),
+        ([-5, -4, -3, -2, -1], [1.0 / 5, 2.0 / 5, 3.0 / 5, 4.0 / 5, 5.0 / 5]),
+    ],
+)
+def test_rank_dense_pct(dtype, ser, exp):
+    if ser[0] < 0 and dtype.startswith("str"):
+        exp = exp[::-1]
+    s = Series(ser).astype(dtype)
+    result = s.rank(method="dense", pct=True)
+    expected = Series(exp).astype(expected_dtype(dtype, "dense", pct=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser, exp",
+    [
+        ([1], [1.0]),
+        ([1, 2], [1.0 / 2, 2.0 / 2]),
+        ([2, 2], [1.0 / 2, 1.0 / 2]),
+        ([1, 2, 3], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([1, 2, 2], [1.0 / 3, 2.0 / 3, 2.0 / 3]),
+        ([4, 2, 1], [3.0 / 3, 2.0 / 3, 1.0 / 3]),
+        ([1, 1, 5, 5, 3], [1.0 / 5, 1.0 / 5, 4.0 / 5, 4.0 / 5, 3.0 / 5]),
+        ([1, 1, 3, 3, 5, 5], [1.0 / 6, 1.0 / 6, 3.0 / 6, 3.0 / 6, 5.0 / 6, 5.0 / 6]),
+        ([-5, -4, -3, -2, -1], [1.0 / 5, 2.0 / 5, 3.0 / 5, 4.0 / 5, 5.0 / 5]),
+    ],
+)
+def test_rank_min_pct(dtype, ser, exp):
+    if ser[0] < 0 and dtype.startswith("str"):
+        exp = exp[::-1]
+    s = Series(ser).astype(dtype)
+    result = s.rank(method="min", pct=True)
+    expected = Series(exp).astype(expected_dtype(dtype, "min", pct=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser, exp",
+    [
+        ([1], [1.0]),
+        ([1, 2], [1.0 / 2, 2.0 / 2]),
+        ([2, 2], [1.0, 1.0]),
+        ([1, 2, 3], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([1, 2, 2], [1.0 / 3, 3.0 / 3, 3.0 / 3]),
+        ([4, 2, 1], [3.0 / 3, 2.0 / 3, 1.0 / 3]),
+        ([1, 1, 5, 5, 3], [2.0 / 5, 2.0 / 5, 5.0 / 5, 5.0 / 5, 3.0 / 5]),
+        ([1, 1, 3, 3, 5, 5], [2.0 / 6, 2.0 / 6, 4.0 / 6, 4.0 / 6, 6.0 / 6, 6.0 / 6]),
+        ([-5, -4, -3, -2, -1], [1.0 / 5, 2.0 / 5, 3.0 / 5, 4.0 / 5, 5.0 / 5]),
+    ],
+)
+def test_rank_max_pct(dtype, ser, exp):
+    if ser[0] < 0 and dtype.startswith("str"):
+        exp = exp[::-1]
+    s = Series(ser).astype(dtype)
+    result = s.rank(method="max", pct=True)
+    expected = Series(exp).astype(expected_dtype(dtype, "max", pct=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser, exp",
+    [
+        ([1], [1.0]),
+        ([1, 2], [1.0 / 2, 2.0 / 2]),
+        ([2, 2], [1.5 / 2, 1.5 / 2]),
+        ([1, 2, 3], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([1, 2, 2], [1.0 / 3, 2.5 / 3, 2.5 / 3]),
+        ([4, 2, 1], [3.0 / 3, 2.0 / 3, 1.0 / 3]),
+        ([1, 1, 5, 5, 3], [1.5 / 5, 1.5 / 5, 4.5 / 5, 4.5 / 5, 3.0 / 5]),
+        ([1, 1, 3, 3, 5, 5], [1.5 / 6, 1.5 / 6, 3.5 / 6, 3.5 / 6, 5.5 / 6, 5.5 / 6]),
+        ([-5, -4, -3, -2, -1], [1.0 / 5, 2.0 / 5, 3.0 / 5, 4.0 / 5, 5.0 / 5]),
+    ],
+)
+def test_rank_average_pct(dtype, ser, exp):
+    if ser[0] < 0 and dtype.startswith("str"):
+        exp = exp[::-1]
+    s = Series(ser).astype(dtype)
+    result = s.rank(method="average", pct=True)
+    expected = Series(exp).astype(expected_dtype(dtype, "average", pct=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser, exp",
+    [
+        ([1], [1.0]),
+        ([1, 2], [1.0 / 2, 2.0 / 2]),
+        ([2, 2], [1.0 / 2, 2.0 / 2.0]),
+        ([1, 2, 3], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([1, 2, 2], [1.0 / 3, 2.0 / 3, 3.0 / 3]),
+        ([4, 2, 1], [3.0 / 3, 2.0 / 3, 1.0 / 3]),
+        ([1, 1, 5, 5, 3], [1.0 / 5, 2.0 / 5, 4.0 / 5, 5.0 / 5, 3.0 / 5]),
+        ([1, 1, 3, 3, 5, 5], [1.0 / 6, 2.0 / 6, 3.0 / 6, 4.0 / 6, 5.0 / 6, 6.0 / 6]),
+        ([-5, -4, -3, -2, -1], [1.0 / 5, 2.0 / 5, 3.0 / 5, 4.0 / 5, 5.0 / 5]),
+    ],
+)
+def test_rank_first_pct(dtype, ser, exp):
+    if ser[0] < 0 and dtype.startswith("str"):
+        exp = exp[::-1]
+    s = Series(ser).astype(dtype)
+    result = s.rank(method="first", pct=True)
+    expected = Series(exp).astype(expected_dtype(dtype, "first", pct=True))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.single_cpu
+def test_pct_max_many_rows():
+    # GH 18271
+    s = Series(np.arange(2**24 + 1))
+    result = s.rank(pct=True).max()
+    assert result == 1
diff --git a/pandas/pandas/tests/series/methods/test_reindex.py b/pandas/pandas/tests/series/methods/test_reindex.py
new file mode 100644
index 0000000000000..442d73cadfe47
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_reindex.py
@@ -0,0 +1,436 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    Categorical,
+    Float64Dtype,
+    Index,
+    MultiIndex,
+    NaT,
+    Period,
+    PeriodIndex,
+    RangeIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+
+
+def test_reindex(datetime_series, string_series):
+    identity = string_series.reindex(string_series.index)
+
+    assert tm.shares_memory(string_series.index, identity.index)
+
+    assert identity.index.is_(string_series.index)
+    assert identity.index.identical(string_series.index)
+
+    subIndex = string_series.index[10:20]
+    subSeries = string_series.reindex(subIndex)
+
+    for idx, val in subSeries.items():
+        assert val == string_series[idx]
+
+    subIndex2 = datetime_series.index[10:20]
+    subTS = datetime_series.reindex(subIndex2)
+
+    for idx, val in subTS.items():
+        assert val == datetime_series[idx]
+    stuffSeries = datetime_series.reindex(subIndex)
+
+    assert np.isnan(stuffSeries).all()
+
+    # This is extremely important for the Cython code to not screw up
+    nonContigIndex = datetime_series.index[::2]
+    subNonContig = datetime_series.reindex(nonContigIndex)
+    for idx, val in subNonContig.items():
+        assert val == datetime_series[idx]
+
+    # return a copy the same index here
+    result = datetime_series.reindex()
+    assert result is not datetime_series
+
+
+def test_reindex_nan():
+    ts = Series([2, 3, 5, 7], index=[1, 4, np.nan, 8])
+
+    i, j = [np.nan, 1, np.nan, 8, 4, np.nan], [2, 0, 2, 3, 1, 2]
+    tm.assert_series_equal(ts.reindex(i), ts.iloc[j])
+
+    ts.index = ts.index.astype("object")
+
+    # reindex coerces index.dtype to float, loc/iloc doesn't
+    tm.assert_series_equal(ts.reindex(i), ts.iloc[j], check_index_type=False)
+
+
+def test_reindex_series_add_nat():
+    rng = date_range("1/1/2000 00:00:00", periods=10, freq="10s")
+    series = Series(rng)
+
+    result = series.reindex(range(15))
+    assert np.issubdtype(result.dtype, np.dtype("M8[ns]"))
+
+    mask = result.isna()
+    assert mask[-5:].all()
+    assert not mask[:-5].any()
+
+
+def test_reindex_with_datetimes():
+    rng = date_range("1/1/2000", periods=20)
+    ts = Series(np.random.default_rng(2).standard_normal(20), index=rng)
+
+    result = ts.reindex(list(ts.index[5:10]))
+    expected = ts[5:10]
+    expected.index = expected.index._with_freq(None)
+    tm.assert_series_equal(result, expected)
+
+    result = ts[list(ts.index[5:10])]
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_corner(datetime_series):
+    # (don't forget to fix this) I think it's fixed
+    empty = Series(index=[])
+    empty.reindex(datetime_series.index, method="pad")  # it works
+
+    # corner case: pad empty series
+    reindexed = empty.reindex(datetime_series.index, method="pad")
+
+    # pass non-Index
+    reindexed = datetime_series.reindex(list(datetime_series.index))
+    datetime_series.index = datetime_series.index._with_freq(None)
+    tm.assert_series_equal(datetime_series, reindexed)
+
+    # bad fill method
+    ts = datetime_series[::2]
+    msg = (
+        r"Invalid fill method\. Expecting pad \(ffill\), backfill "
+        r"\(bfill\) or nearest\. Got foo"
+    )
+    with pytest.raises(ValueError, match=msg):
+        ts.reindex(datetime_series.index, method="foo")
+
+
+def test_reindex_pad():
+    s = Series(np.arange(10), dtype="int64")
+    s2 = s[::2]
+
+    reindexed = s2.reindex(s.index, method="pad")
+    reindexed2 = s2.reindex(s.index, method="ffill")
+    tm.assert_series_equal(reindexed, reindexed2)
+
+    expected = Series([0, 0, 2, 2, 4, 4, 6, 6, 8, 8])
+    tm.assert_series_equal(reindexed, expected)
+
+
+def test_reindex_pad2():
+    # GH4604
+    s = Series([1, 2, 3, 4, 5], index=["a", "b", "c", "d", "e"])
+    new_index = ["a", "g", "c", "f"]
+    expected = Series([1, 1, 3, 3.0], index=new_index)
+
+    # this changes dtype because the ffill happens after
+    result = s.reindex(new_index).ffill()
+    tm.assert_series_equal(result, expected)
+
+    result = s.reindex(new_index).ffill()
+    tm.assert_series_equal(result, expected)
+
+    expected = Series([1, 5, 3, 5], index=new_index)
+    result = s.reindex(new_index, method="ffill")
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_inference():
+    # inference of new dtype
+    s = Series([True, False, False, True], index=list("abcd"))
+    new_index = "agc"
+    result = s.reindex(list(new_index)).ffill()
+    expected = Series([True, True, False], index=list(new_index), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_downcasting():
+    # GH4618 shifted series downcasting
+    s = Series(False, index=range(5))
+    result = s.shift(1).bfill()
+    expected = Series(False, index=range(5), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_nearest():
+    s = Series(np.arange(10, dtype="int64"))
+    target = [0.1, 0.9, 1.5, 2.0]
+    result = s.reindex(target, method="nearest")
+    expected = Series(np.around(target).astype("int64"), target)
+    tm.assert_series_equal(expected, result)
+
+    result = s.reindex(target, method="nearest", tolerance=0.2)
+    expected = Series([0, 1, np.nan, 2], target)
+    tm.assert_series_equal(expected, result)
+
+    result = s.reindex(target, method="nearest", tolerance=[0.3, 0.01, 0.4, 3])
+    expected = Series([0, np.nan, np.nan, 2], target)
+    tm.assert_series_equal(expected, result)
+
+
+def test_reindex_int(datetime_series):
+    ts = datetime_series[::2]
+    int_ts = Series(np.zeros(len(ts), dtype=int), index=ts.index)
+
+    # this should work fine
+    reindexed_int = int_ts.reindex(datetime_series.index)
+
+    # if NaNs introduced
+    assert reindexed_int.dtype == np.float64
+
+    # NO NaNs introduced
+    reindexed_int = int_ts.reindex(int_ts.index[::2])
+    assert reindexed_int.dtype == np.dtype(int)
+
+
+def test_reindex_bool(datetime_series):
+    # A series other than float, int, string, or object
+    ts = datetime_series[::2]
+    bool_ts = Series(np.zeros(len(ts), dtype=bool), index=ts.index)
+
+    # this should work fine
+    reindexed_bool = bool_ts.reindex(datetime_series.index)
+
+    # if NaNs introduced
+    assert reindexed_bool.dtype == np.object_
+
+    # NO NaNs introduced
+    reindexed_bool = bool_ts.reindex(bool_ts.index[::2])
+    assert reindexed_bool.dtype == np.bool_
+
+
+def test_reindex_bool_pad(datetime_series):
+    # fail
+    ts = datetime_series[5:]
+    bool_ts = Series(np.zeros(len(ts), dtype=bool), index=ts.index)
+    filled_bool = bool_ts.reindex(datetime_series.index, method="pad")
+    assert isna(filled_bool[:5]).all()
+
+
+def test_reindex_categorical():
+    index = date_range("20000101", periods=3)
+
+    # reindexing to an invalid Categorical
+    s = Series(["a", "b", "c"], dtype="category")
+    result = s.reindex(index)
+    expected = Series(
+        Categorical(values=[np.nan, np.nan, np.nan], categories=["a", "b", "c"])
+    )
+    expected.index = index
+    tm.assert_series_equal(result, expected)
+
+    # partial reindexing
+    expected = Series(
+        Categorical(values=["b", "c"], categories=["a", "b", "c"]), index=range(1, 3)
+    )
+    result = s.reindex([1, 2])
+    tm.assert_series_equal(result, expected)
+
+    expected = Series(
+        Categorical(values=["c", np.nan], categories=["a", "b", "c"]), index=range(2, 4)
+    )
+    result = s.reindex([2, 3])
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_astype_order_consistency():
+    # GH#17444
+    ser = Series([1, 2, 3], index=[2, 0, 1])
+    new_index = [0, 1, 2]
+    temp_dtype = "category"
+    new_dtype = str
+    result = ser.reindex(new_index).astype(temp_dtype).astype(new_dtype)
+    expected = ser.astype(temp_dtype).reindex(new_index).astype(new_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_fill_value():
+    # -----------------------------------------------------------
+    # floats
+    floats = Series([1.0, 2.0, 3.0])
+    result = floats.reindex([1, 2, 3])
+    expected = Series([2.0, 3.0, np.nan], index=range(1, 4))
+    tm.assert_series_equal(result, expected)
+
+    result = floats.reindex([1, 2, 3], fill_value=0)
+    expected = Series([2.0, 3.0, 0], index=range(1, 4))
+    tm.assert_series_equal(result, expected)
+
+    # -----------------------------------------------------------
+    # ints
+    ints = Series([1, 2, 3])
+
+    result = ints.reindex([1, 2, 3])
+    expected = Series([2.0, 3.0, np.nan], index=range(1, 4))
+    tm.assert_series_equal(result, expected)
+
+    # don't upcast
+    result = ints.reindex([1, 2, 3], fill_value=0)
+    expected = Series([2, 3, 0], index=range(1, 4))
+    assert issubclass(result.dtype.type, np.integer)
+    tm.assert_series_equal(result, expected)
+
+    # -----------------------------------------------------------
+    # objects
+    objects = Series([1, 2, 3], dtype=object)
+
+    result = objects.reindex([1, 2, 3])
+    expected = Series([2, 3, np.nan], index=range(1, 4), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    result = objects.reindex([1, 2, 3], fill_value="foo")
+    expected = Series([2, 3, "foo"], index=range(1, 4), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    # ------------------------------------------------------------
+    # bools
+    bools = Series([True, False, True])
+
+    result = bools.reindex([1, 2, 3])
+    expected = Series([False, True, np.nan], index=range(1, 4), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    result = bools.reindex([1, 2, 3], fill_value=False)
+    expected = Series([False, True, False], index=range(1, 4))
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", ["datetime64[ns]", "timedelta64[ns]"])
+@pytest.mark.parametrize("fill_value", ["string", 0, Timedelta(0)])
+def test_reindex_fill_value_datetimelike_upcast(dtype, fill_value):
+    # https://github.com/pandas-dev/pandas/issues/42921
+    if dtype == "timedelta64[ns]" and fill_value == Timedelta(0):
+        # use the scalar that is not compatible with the dtype for this test
+        fill_value = Timestamp(0)
+
+    ser = Series([NaT], dtype=dtype)
+
+    result = ser.reindex([0, 1], fill_value=fill_value)
+    expected = Series([NaT, fill_value], index=range(2), dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_datetimeindexes_tz_naive_and_aware():
+    # GH 8306
+    idx = date_range("20131101", tz="America/Chicago", periods=7)
+    newidx = date_range("20131103", periods=10, freq="h")
+    s = Series(range(7), index=idx)
+    msg = (
+        r"Cannot compare dtypes datetime64\[ns, America/Chicago\] "
+        r"and datetime64\[ns\]"
+    )
+    with pytest.raises(TypeError, match=msg):
+        s.reindex(newidx, method="ffill")
+
+
+def test_reindex_empty_series_tz_dtype():
+    # GH 20869
+    result = Series(dtype="datetime64[ns, UTC]").reindex([0, 1])
+    expected = Series([NaT] * 2, dtype="datetime64[ns, UTC]")
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "p_values, o_values, values, expected_values",
+    [
+        (
+            [Period("2019Q1", "Q-DEC"), Period("2019Q2", "Q-DEC")],
+            [Period("2019Q1", "Q-DEC"), Period("2019Q2", "Q-DEC"), "All"],
+            [1.0, 1.0],
+            [1.0, 1.0, np.nan],
+        ),
+        (
+            [Period("2019Q1", "Q-DEC"), Period("2019Q2", "Q-DEC")],
+            [Period("2019Q1", "Q-DEC"), Period("2019Q2", "Q-DEC")],
+            [1.0, 1.0],
+            [1.0, 1.0],
+        ),
+    ],
+)
+def test_reindex_periodindex_with_object(p_values, o_values, values, expected_values):
+    # GH#28337
+    period_index = PeriodIndex(p_values)
+    object_index = Index(o_values)
+
+    ser = Series(values, index=period_index)
+    result = ser.reindex(object_index)
+    expected = Series(expected_values, index=object_index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_too_many_args():
+    # GH 40980
+    ser = Series([1, 2])
+    msg = r"reindex\(\) takes from 1 to 2 positional arguments but 3 were given"
+    with pytest.raises(TypeError, match=msg):
+        ser.reindex([2, 3], False)
+
+
+def test_reindex_double_index():
+    # GH 40980
+    ser = Series([1, 2])
+    msg = r"reindex\(\) got multiple values for argument 'index'"
+    with pytest.raises(TypeError, match=msg):
+        ser.reindex([2, 3], index=[3, 4])
+
+
+def test_reindex_no_posargs():
+    # GH 40980
+    ser = Series([1, 2])
+    result = ser.reindex(index=[1, 0])
+    expected = Series([2, 1], index=[1, 0])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("values", [[["a"], ["x"]], [[], []]])
+def test_reindex_empty_with_level(values):
+    # GH41170
+    ser = Series(
+        range(len(values[0])), index=MultiIndex.from_arrays(values), dtype="object"
+    )
+    result = ser.reindex(np.array(["b"]), level=0)
+    expected = Series(
+        index=MultiIndex(levels=[["b"], values[1]], codes=[[], []]), dtype="object"
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_missing_category():
+    # GH#18185
+    ser = Series([1, 2, 3, 1], dtype="category")
+    msg = r"Cannot setitem on a Categorical with a new category \(-1\)"
+    with pytest.raises(TypeError, match=msg):
+        ser.reindex([1, 2, 3, 4, 5], fill_value=-1)
+
+
+def test_reindexing_with_float64_NA_log():
+    # GH 47055
+    s = Series([1.0, NA], dtype=Float64Dtype())
+    s_reindex = s.reindex(range(3))
+    result = s_reindex.values._data
+    expected = np.array([1, np.nan, np.nan])
+    tm.assert_numpy_array_equal(result, expected)
+    with tm.assert_produces_warning(None):
+        result_log = np.log(s_reindex)
+        expected_log = Series([0, np.nan, np.nan], dtype=Float64Dtype())
+        tm.assert_series_equal(result_log, expected_log)
+
+
+@pytest.mark.parametrize("dtype", ["timedelta64", "datetime64"])
+def test_reindex_expand_nonnano_nat(dtype):
+    # GH 53497
+    ser = Series(np.array([1], dtype=f"{dtype}[s]"))
+    result = ser.reindex(RangeIndex(2))
+    expected = Series(
+        np.array([1, getattr(np, dtype)("nat", "s")], dtype=f"{dtype}[s]")
+    )
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_reindex_like.py b/pandas/pandas/tests/series/methods/test_reindex_like.py
new file mode 100644
index 0000000000000..10b8ac5817636
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_reindex_like.py
@@ -0,0 +1,45 @@
+from datetime import datetime
+
+import numpy as np
+
+from pandas import Series
+import pandas._testing as tm
+
+
+def test_reindex_like(datetime_series):
+    other = datetime_series[::2]
+    tm.assert_series_equal(
+        datetime_series.reindex(other.index), datetime_series.reindex_like(other)
+    )
+
+    # GH#7179
+    day1 = datetime(2013, 3, 5)
+    day2 = datetime(2013, 5, 5)
+    day3 = datetime(2014, 3, 5)
+
+    series1 = Series([5, None, None], [day1, day2, day3])
+    series2 = Series([None, None], [day1, day3])
+
+    with tm.assert_produces_warning(FutureWarning):
+        result = series1.reindex_like(series2, method="pad")
+    expected = Series([5, np.nan], index=[day1, day3])
+    tm.assert_series_equal(result, expected)
+
+
+def test_reindex_like_nearest():
+    ser = Series(np.arange(10, dtype="int64"))
+
+    target = [0.1, 0.9, 1.5, 2.0]
+    other = ser.reindex(target, method="nearest")
+    expected = Series(np.around(target).astype("int64"), target)
+
+    with tm.assert_produces_warning(FutureWarning):
+        result = ser.reindex_like(other, method="nearest")
+    tm.assert_series_equal(expected, result)
+
+    with tm.assert_produces_warning(FutureWarning):
+        result = ser.reindex_like(other, method="nearest", tolerance=1)
+    tm.assert_series_equal(expected, result)
+    with tm.assert_produces_warning(FutureWarning):
+        result = ser.reindex_like(other, method="nearest", tolerance=[1, 2, 3, 4])
+    tm.assert_series_equal(expected, result)
diff --git a/pandas/pandas/tests/series/methods/test_rename.py b/pandas/pandas/tests/series/methods/test_rename.py
new file mode 100644
index 0000000000000..f5a97d61990a4
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_rename.py
@@ -0,0 +1,179 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+    array,
+)
+import pandas._testing as tm
+
+
+class TestRename:
+    def test_rename(self, datetime_series):
+        ts = datetime_series
+        renamer = lambda x: x.strftime("%Y%m%d")
+        renamed = ts.rename(renamer)
+        assert renamed.index[0] == renamer(ts.index[0])
+
+        # dict
+        rename_dict = dict(zip(ts.index, renamed.index))
+        renamed2 = ts.rename(rename_dict)
+        tm.assert_series_equal(renamed, renamed2)
+
+    def test_rename_partial_dict(self):
+        # partial dict
+        ser = Series(np.arange(4), index=["a", "b", "c", "d"], dtype="int64")
+        renamed = ser.rename({"b": "foo", "d": "bar"})
+        tm.assert_index_equal(renamed.index, Index(["a", "foo", "c", "bar"]))
+
+    def test_rename_retain_index_name(self):
+        # index with name
+        renamer = Series(
+            np.arange(4), index=Index(["a", "b", "c", "d"], name="name"), dtype="int64"
+        )
+        renamed = renamer.rename({})
+        assert renamed.index.name == renamer.index.name
+
+    def test_rename_by_series(self):
+        ser = Series(range(5), name="foo")
+        renamer = Series({1: 10, 2: 20})
+        result = ser.rename(renamer)
+        expected = Series(range(5), index=[0, 10, 20, 3, 4], name="foo")
+        tm.assert_series_equal(result, expected)
+
+    def test_rename_set_name(self, using_infer_string):
+        ser = Series(range(4), index=list("abcd"))
+        for name in ["foo", 123, 123.0, datetime(2001, 11, 11), ("foo",)]:
+            result = ser.rename(name)
+            assert result.name == name
+            if using_infer_string:
+                tm.assert_extension_array_equal(result.index.values, ser.index.values)
+            else:
+                tm.assert_numpy_array_equal(result.index.values, ser.index.values)
+            assert ser.name is None
+
+    def test_rename_set_name_inplace(self, using_infer_string):
+        ser = Series(range(3), index=list("abc"))
+        for name in ["foo", 123, 123.0, datetime(2001, 11, 11), ("foo",)]:
+            ser.rename(name, inplace=True)
+            assert ser.name == name
+            exp = np.array(["a", "b", "c"], dtype=np.object_)
+            if using_infer_string:
+                exp = array(exp, dtype="str")
+                tm.assert_extension_array_equal(ser.index.values, exp)
+            else:
+                tm.assert_numpy_array_equal(ser.index.values, exp)
+
+    def test_rename_axis_supported(self):
+        # Supporting axis for compatibility, detailed in GH-18589
+        ser = Series(range(5))
+        ser.rename({}, axis=0)
+        ser.rename({}, axis="index")
+
+        with pytest.raises(ValueError, match="No axis named 5"):
+            ser.rename({}, axis=5)
+
+    def test_rename_inplace(self, datetime_series):
+        renamer = lambda x: x.strftime("%Y%m%d")
+        expected = renamer(datetime_series.index[0])
+
+        datetime_series.rename(renamer, inplace=True)
+        assert datetime_series.index[0] == expected
+
+    def test_rename_with_custom_indexer(self):
+        # GH 27814
+        class MyIndexer:
+            pass
+
+        ix = MyIndexer()
+        ser = Series([1, 2, 3]).rename(ix)
+        assert ser.name is ix
+
+    def test_rename_with_custom_indexer_inplace(self):
+        # GH 27814
+        class MyIndexer:
+            pass
+
+        ix = MyIndexer()
+        ser = Series([1, 2, 3])
+        ser.rename(ix, inplace=True)
+        assert ser.name is ix
+
+    def test_rename_callable(self):
+        # GH 17407
+        ser = Series(range(1, 6), index=Index(range(2, 7), name="IntIndex"))
+        result = ser.rename(str)
+        expected = ser.rename(lambda i: str(i))
+        tm.assert_series_equal(result, expected)
+
+        assert result.name == expected.name
+
+    def test_rename_none(self):
+        # GH 40977
+        ser = Series([1, 2], name="foo")
+        result = ser.rename(None)
+        expected = Series([1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_rename_series_with_multiindex(self):
+        # issue #43659
+        arrays = [
+            ["bar", "baz", "baz", "foo", "qux"],
+            ["one", "one", "two", "two", "one"],
+        ]
+
+        index = MultiIndex.from_arrays(arrays, names=["first", "second"])
+        ser = Series(np.ones(5), index=index)
+        result = ser.rename(index={"one": "yes"}, level="second", errors="raise")
+
+        arrays_expected = [
+            ["bar", "baz", "baz", "foo", "qux"],
+            ["yes", "yes", "two", "two", "yes"],
+        ]
+
+        index_expected = MultiIndex.from_arrays(
+            arrays_expected, names=["first", "second"]
+        )
+        series_expected = Series(np.ones(5), index=index_expected)
+
+        tm.assert_series_equal(result, series_expected)
+
+    def test_rename_series_with_multiindex_keeps_ea_dtypes(self):
+        # GH21055
+        arrays = [
+            Index([1, 2, 3], dtype="Int64").astype("category"),
+            Index([1, 2, 3], dtype="Int64"),
+        ]
+        mi = MultiIndex.from_arrays(arrays, names=["A", "B"])
+        ser = Series(1, index=mi)
+        result = ser.rename({1: 4}, level=1)
+
+        arrays_expected = [
+            Index([1, 2, 3], dtype="Int64").astype("category"),
+            Index([4, 2, 3], dtype="Int64"),
+        ]
+        mi_expected = MultiIndex.from_arrays(arrays_expected, names=["A", "B"])
+        expected = Series(1, index=mi_expected)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_rename_error_arg(self):
+        # GH 46889
+        ser = Series(["foo", "bar"])
+        match = re.escape("[2] not found in axis")
+        with pytest.raises(KeyError, match=match):
+            ser.rename({2: 9}, errors="raise")
+
+    def test_rename_copy_false(self):
+        # GH 46889
+        ser = Series(["foo", "bar"])
+        ser_orig = ser.copy()
+        shallow_copy = ser.rename({1: 9})
+        ser[0] = "foobar"
+        assert ser_orig[0] == shallow_copy[0]
+        assert ser_orig[1] == shallow_copy[9]
diff --git a/pandas/pandas/tests/series/methods/test_rename_axis.py b/pandas/pandas/tests/series/methods/test_rename_axis.py
new file mode 100644
index 0000000000000..58c095d697ede
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_rename_axis.py
@@ -0,0 +1,47 @@
+import pytest
+
+from pandas import (
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestSeriesRenameAxis:
+    def test_rename_axis_mapper(self):
+        # GH 19978
+        mi = MultiIndex.from_product([["a", "b", "c"], [1, 2]], names=["ll", "nn"])
+        ser = Series(list(range(len(mi))), index=mi)
+
+        result = ser.rename_axis(index={"ll": "foo"})
+        assert result.index.names == ["foo", "nn"]
+
+        result = ser.rename_axis(index=str.upper, axis=0)
+        assert result.index.names == ["LL", "NN"]
+
+        result = ser.rename_axis(index=["foo", "goo"])
+        assert result.index.names == ["foo", "goo"]
+
+        with pytest.raises(TypeError, match="unexpected"):
+            ser.rename_axis(columns="wrong")
+
+    def test_rename_axis_inplace(self, datetime_series):
+        # GH 15704
+        expected = datetime_series.rename_axis("foo")
+        result = datetime_series
+        no_return = result.rename_axis("foo", inplace=True)
+
+        assert no_return is None
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("kwargs", [{"mapper": None}, {"index": None}, {}])
+    def test_rename_axis_none(self, kwargs):
+        # GH 25034
+        index = Index(list("abc"), name="foo")
+        ser = Series([1, 2, 3], index=index)
+
+        result = ser.rename_axis(**kwargs)
+        expected_index = index.rename(None) if kwargs else index
+        expected = Series([1, 2, 3], index=expected_index)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_repeat.py b/pandas/pandas/tests/series/methods/test_repeat.py
new file mode 100644
index 0000000000000..8ecc8052ff49c
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_repeat.py
@@ -0,0 +1,40 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestRepeat:
+    def test_repeat(self):
+        ser = Series(np.random.default_rng(2).standard_normal(3), index=["a", "b", "c"])
+
+        reps = ser.repeat(5)
+        exp = Series(ser.values.repeat(5), index=ser.index.values.repeat(5))
+        tm.assert_series_equal(reps, exp)
+
+        to_rep = [2, 3, 4]
+        reps = ser.repeat(to_rep)
+        exp = Series(ser.values.repeat(to_rep), index=ser.index.values.repeat(to_rep))
+        tm.assert_series_equal(reps, exp)
+
+    def test_numpy_repeat(self):
+        ser = Series(np.arange(3), name="x")
+        expected = Series(
+            ser.values.repeat(2), name="x", index=ser.index.values.repeat(2)
+        )
+        tm.assert_series_equal(np.repeat(ser, 2), expected)
+
+        msg = "the 'axis' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.repeat(ser, 2, axis=0)
+
+    def test_repeat_with_multiindex(self):
+        # GH#9361, fixed by  GH#7891
+        m_idx = MultiIndex.from_tuples([(1, 2), (3, 4), (5, 6), (7, 8)])
+        data = ["a", "b", "c", "d"]
+        m_df = Series(data, index=m_idx)
+        assert m_df.repeat(3).shape == (3 * len(data),)
diff --git a/pandas/pandas/tests/series/methods/test_replace.py b/pandas/pandas/tests/series/methods/test_replace.py
new file mode 100644
index 0000000000000..abd5d075ea3d5
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_replace.py
@@ -0,0 +1,717 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core.arrays import IntervalArray
+
+
+class TestSeriesReplace:
+    def test_replace_explicit_none(self):
+        # GH#36984 if the user explicitly passes value=None, give it to them
+        ser = pd.Series([0, 0, ""], dtype=object)
+        result = ser.replace("", None)
+        expected = pd.Series([0, 0, None], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        # Cast column 2 to object to avoid implicit cast when setting entry to ""
+        df = pd.DataFrame(np.zeros((3, 3))).astype({2: object})
+        df.iloc[2, 2] = ""
+        result = df.replace("", None)
+        expected = pd.DataFrame(
+            {
+                0: np.zeros(3),
+                1: np.zeros(3),
+                2: np.array([0.0, 0.0, None], dtype=object),
+            }
+        )
+        assert expected.iloc[2, 2] is None
+        tm.assert_frame_equal(result, expected)
+
+        # GH#19998 same thing with object dtype
+        ser = pd.Series([10, 20, 30, "a", "a", "b", "a"])
+        result = ser.replace("a", None)
+        expected = pd.Series([10, 20, 30, None, None, "b", None])
+        assert expected.iloc[-1] is None
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_noop_doesnt_downcast(self):
+        # GH#44498
+        ser = pd.Series([None, None, pd.Timestamp("2021-12-16 17:31")], dtype=object)
+        res = ser.replace({np.nan: None})  # should be a no-op
+        tm.assert_series_equal(res, ser)
+        assert res.dtype == object
+
+        # same thing but different calling convention
+        res = ser.replace(np.nan, None)
+        tm.assert_series_equal(res, ser)
+        assert res.dtype == object
+
+    def test_replace(self):
+        N = 50
+        ser = pd.Series(np.random.default_rng(2).standard_normal(N))
+        ser[0:4] = np.nan
+        ser[6:10] = 0
+
+        # replace list with a single value
+        return_value = ser.replace([np.nan], -1, inplace=True)
+        assert return_value is None
+
+        exp = ser.fillna(-1)
+        tm.assert_series_equal(ser, exp)
+
+        rs = ser.replace(0.0, np.nan)
+        ser[ser == 0.0] = np.nan
+        tm.assert_series_equal(rs, ser)
+
+        ser = pd.Series(
+            np.fabs(np.random.default_rng(2).standard_normal(N)),
+            pd.date_range("2020-01-01", periods=N),
+            dtype=object,
+        )
+        ser[:5] = np.nan
+        ser[6:10] = "foo"
+        ser[20:30] = "bar"
+
+        # replace list with a single value
+
+        rs = ser.replace([np.nan, "foo", "bar"], -1)
+
+        assert (rs[:5] == -1).all()
+        assert (rs[6:10] == -1).all()
+        assert (rs[20:30] == -1).all()
+        assert (pd.isna(ser[:5])).all()
+
+        # replace with different values
+        rs = ser.replace({np.nan: -1, "foo": -2, "bar": -3})
+
+        assert (rs[:5] == -1).all()
+        assert (rs[6:10] == -2).all()
+        assert (rs[20:30] == -3).all()
+        assert (pd.isna(ser[:5])).all()
+
+        # replace with different values with 2 lists
+        rs2 = ser.replace([np.nan, "foo", "bar"], [-1, -2, -3])
+        tm.assert_series_equal(rs, rs2)
+
+        # replace inplace
+        return_value = ser.replace([np.nan, "foo", "bar"], -1, inplace=True)
+        assert return_value is None
+
+        assert (ser[:5] == -1).all()
+        assert (ser[6:10] == -1).all()
+        assert (ser[20:30] == -1).all()
+
+    def test_replace_nan_with_inf(self):
+        ser = pd.Series([np.nan, 0, np.inf])
+        tm.assert_series_equal(ser.replace(np.nan, 0), ser.fillna(0))
+
+        ser = pd.Series([np.nan, 0, "foo", "bar", np.inf, None, pd.NaT])
+        tm.assert_series_equal(ser.replace(np.nan, 0), ser.fillna(0))
+        filled = ser.copy()
+        filled[4] = 0
+        tm.assert_series_equal(ser.replace(np.inf, 0), filled)
+
+    def test_replace_listlike_value_listlike_target(self, datetime_series):
+        ser = pd.Series(datetime_series.index)
+        tm.assert_series_equal(ser.replace(np.nan, 0), ser.fillna(0))
+
+        # malformed
+        msg = r"Replacement lists must match in length\. Expecting 3 got 2"
+        with pytest.raises(ValueError, match=msg):
+            ser.replace([1, 2, 3], [np.nan, 0])
+
+        # ser is dt64 so can't hold 1 or 2, so this replace is a no-op
+        result = ser.replace([1, 2], [np.nan, 0])
+        tm.assert_series_equal(result, ser)
+
+        ser = pd.Series([0, 1, 2, 3, 4])
+        result = ser.replace([0, 1, 2, 3, 4], [4, 3, 2, 1, 0])
+        tm.assert_series_equal(result, pd.Series([4, 3, 2, 1, 0]))
+
+    def test_replace_gh5319(self):
+        # API change from 0.12?
+        # GH 5319
+        ser = pd.Series([0, np.nan, 2, 3, 4])
+        msg = (
+            "Series.replace must specify either 'value', "
+            "a dict-like 'to_replace', or dict-like 'regex'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.replace([np.nan])
+
+        with pytest.raises(ValueError, match=msg):
+            ser.replace(np.nan)
+
+    def test_replace_datetime64(self):
+        # GH 5797
+        ser = pd.Series(pd.date_range("20130101", periods=5))
+        expected = ser.copy()
+        expected.loc[2] = pd.Timestamp("20120101")
+        result = ser.replace({pd.Timestamp("20130103"): pd.Timestamp("20120101")})
+        tm.assert_series_equal(result, expected)
+        result = ser.replace(pd.Timestamp("20130103"), pd.Timestamp("20120101"))
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_nat_with_tz(self):
+        # GH 11792: Test with replacing NaT in a list with tz data
+        ts = pd.Timestamp("2015/01/01", tz="UTC")
+        s = pd.Series([pd.NaT, pd.Timestamp("2015/01/01", tz="UTC")])
+        result = s.replace([np.nan, pd.NaT], pd.Timestamp.min)
+        expected = pd.Series([pd.Timestamp.min, ts], dtype=object)
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_timedelta_td64(self):
+        tdi = pd.timedelta_range(0, periods=5)
+        ser = pd.Series(tdi)
+
+        # Using a single dict argument means we go through replace_list
+        result = ser.replace({ser[1]: ser[3]})
+
+        expected = pd.Series([ser[0], ser[3], ser[2], ser[3], ser[4]])
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_with_single_list(self):
+        ser = pd.Series([0, 1, 2, 3, 4])
+        msg = (
+            "Series.replace must specify either 'value', "
+            "a dict-like 'to_replace', or dict-like 'regex'"
+        )
+        with pytest.raises(ValueError, match=msg):
+            ser.replace([1, 2, 3])
+
+        s = ser.copy()
+        with pytest.raises(ValueError, match=msg):
+            s.replace([1, 2, 3], inplace=True)
+
+    def test_replace_mixed_types(self):
+        ser = pd.Series(np.arange(5), dtype="int64")
+
+        def check_replace(to_rep, val, expected):
+            sc = ser.copy()
+            result = ser.replace(to_rep, val)
+            return_value = sc.replace(to_rep, val, inplace=True)
+            assert return_value is None
+            tm.assert_series_equal(expected, result)
+            tm.assert_series_equal(expected, sc)
+
+        # 3.0 can still be held in our int64 series, so we do not upcast GH#44940
+        tr, v = [3], [3.0]
+        check_replace(tr, v, ser)
+        # Note this matches what we get with the scalars 3 and 3.0
+        check_replace(tr[0], v[0], ser)
+
+        # MUST upcast to float
+        e = pd.Series([0, 1, 2, 3.5, 4])
+        tr, v = [3], [3.5]
+        check_replace(tr, v, e)
+
+        # casts to object
+        e = pd.Series([0, 1, 2, 3.5, "a"])
+        tr, v = [3, 4], [3.5, "a"]
+        check_replace(tr, v, e)
+
+        # again casts to object
+        e = pd.Series([0, 1, 2, 3.5, pd.Timestamp("20130101")])
+        tr, v = [3, 4], [3.5, pd.Timestamp("20130101")]
+        check_replace(tr, v, e)
+
+        # casts to object
+        e = pd.Series([0, 1, 2, 3.5, True], dtype="object")
+        tr, v = [3, 4], [3.5, True]
+        check_replace(tr, v, e)
+
+        # test an object with dates + floats + integers + strings
+        dr = pd.Series(pd.date_range("1/1/2001", "1/10/2001", freq="D"))
+        result = dr.astype(object).replace([dr[0], dr[1], dr[2]], [1.0, 2, "a"])
+        expected = pd.Series([1.0, 2, "a"] + dr[3:].tolist(), dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_bool_with_string_no_op(self):
+        s = pd.Series([True, False, True])
+        result = s.replace("fun", "in-the-sun")
+        tm.assert_series_equal(s, result)
+
+    def test_replace_bool_with_string(self):
+        # nonexistent elements
+        s = pd.Series([True, False, True])
+        result = s.replace(True, "2u")
+        expected = pd.Series(["2u", False, "2u"])
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_bool_with_bool(self):
+        s = pd.Series([True, False, True])
+        result = s.replace(True, False)
+        expected = pd.Series([False] * len(s))
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_with_dict_with_bool_keys(self):
+        s = pd.Series([True, False, True])
+        result = s.replace({"asdf": "asdb", True: "yes"})
+        expected = pd.Series(["yes", False, "yes"])
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_Int_with_na(self, any_int_ea_dtype):
+        # GH 38267
+        result = pd.Series([0, None], dtype=any_int_ea_dtype).replace(0, pd.NA)
+        expected = pd.Series([pd.NA, pd.NA], dtype=any_int_ea_dtype)
+        tm.assert_series_equal(result, expected)
+        result = pd.Series([0, 1], dtype=any_int_ea_dtype).replace(0, pd.NA)
+        result.replace(1, pd.NA, inplace=True)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace2(self):
+        N = 50
+        ser = pd.Series(
+            np.fabs(np.random.default_rng(2).standard_normal(N)),
+            pd.date_range("2020-01-01", periods=N),
+            dtype=object,
+        )
+        ser[:5] = np.nan
+        ser[6:10] = "foo"
+        ser[20:30] = "bar"
+
+        # replace list with a single value
+        rs = ser.replace([np.nan, "foo", "bar"], -1)
+
+        assert (rs[:5] == -1).all()
+        assert (rs[6:10] == -1).all()
+        assert (rs[20:30] == -1).all()
+        assert (pd.isna(ser[:5])).all()
+
+        # replace with different values
+        rs = ser.replace({np.nan: -1, "foo": -2, "bar": -3})
+
+        assert (rs[:5] == -1).all()
+        assert (rs[6:10] == -2).all()
+        assert (rs[20:30] == -3).all()
+        assert (pd.isna(ser[:5])).all()
+
+        # replace with different values with 2 lists
+        rs2 = ser.replace([np.nan, "foo", "bar"], [-1, -2, -3])
+        tm.assert_series_equal(rs, rs2)
+
+        # replace inplace
+        return_value = ser.replace([np.nan, "foo", "bar"], -1, inplace=True)
+        assert return_value is None
+        assert (ser[:5] == -1).all()
+        assert (ser[6:10] == -1).all()
+        assert (ser[20:30] == -1).all()
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    def test_replace_cascade(self, inplace):
+        # Test that replaced values are not replaced again
+        # GH #50778
+        ser = pd.Series([1, 2, 3])
+        expected = pd.Series([2, 3, 4])
+
+        res = ser.replace([1, 2, 3], [2, 3, 4], inplace=inplace)
+        if inplace:
+            tm.assert_series_equal(ser, expected)
+        else:
+            tm.assert_series_equal(res, expected)
+
+    def test_replace_with_dictlike_and_string_dtype(self, nullable_string_dtype):
+        # GH 32621, GH#44940
+        ser = pd.Series(["one", "two", np.nan], dtype=nullable_string_dtype)
+        expected = pd.Series(["1", "2", np.nan], dtype=nullable_string_dtype)
+        result = ser.replace({"one": "1", "two": "2"})
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_with_empty_dictlike(self):
+        # GH 15289
+        s = pd.Series(list("abcd"))
+        tm.assert_series_equal(s, s.replace({}))
+
+        empty_series = pd.Series([])
+        tm.assert_series_equal(s, s.replace(empty_series))
+
+    def test_replace_string_with_number(self):
+        # GH 15743
+        s = pd.Series([1, 2, 3])
+        result = s.replace("2", np.nan)
+        expected = pd.Series([1, 2, 3])
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_replacer_equals_replacement(self):
+        # GH 20656
+        # make sure all replacers are matching against original values
+        s = pd.Series(["a", "b"])
+        expected = pd.Series(["b", "a"])
+        result = s.replace({"a": "b", "b": "a"})
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_unicode_with_number(self):
+        # GH 15743
+        s = pd.Series([1, 2, 3])
+        result = s.replace("2", np.nan)
+        expected = pd.Series([1, 2, 3])
+        tm.assert_series_equal(expected, result)
+
+    def test_replace_mixed_types_with_string(self):
+        # Testing mixed
+        s = pd.Series([1, 2, 3, "4", 4, 5])
+        result = s.replace([2, "4"], np.nan)
+        expected = pd.Series([1, np.nan, 3, np.nan, 4, 5], dtype=object)
+        tm.assert_series_equal(expected, result)
+
+    @pytest.mark.parametrize(
+        "categorical, numeric",
+        [
+            (["A"], [1]),
+            (["A", "B"], [1, 2]),
+        ],
+    )
+    def test_replace_categorical(self, categorical, numeric):
+        # GH 24971, GH#23305
+        ser = pd.Series(pd.Categorical(categorical, categories=["A", "B"]))
+        result = ser.cat.rename_categories({"A": 1, "B": 2})
+        expected = pd.Series(numeric).astype("category")
+        if 2 not in expected.cat.categories:
+            # i.e. categories should be [1, 2] even if there are no "B"s present
+            # GH#44940
+            expected = expected.cat.add_categories(2)
+        tm.assert_series_equal(expected, result, check_categorical=False)
+
+    def test_replace_categorical_inplace(self):
+        # GH 53358
+        data = ["a", "b", "c"]
+        data_exp = ["b", "b", "c"]
+        result = pd.Series(data, dtype="category")
+        result.replace(to_replace="a", value="b", inplace=True)
+        expected = pd.Series(pd.Categorical(data_exp, categories=data))
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_categorical_single(self):
+        # GH 26988
+        dti = pd.date_range("2016-01-01", periods=3, tz="US/Pacific")
+        s = pd.Series(dti)
+        c = s.astype("category")
+
+        expected = c.copy()
+        expected = expected.cat.add_categories("foo")
+        expected[2] = "foo"
+        expected = expected.cat.remove_unused_categories()
+        assert c[2] != "foo"
+
+        result = c.cat.rename_categories({c.values[2]: "foo"})
+        tm.assert_series_equal(expected, result)
+        assert c[2] != "foo"  # ensure non-inplace call does not alter original
+
+    def test_replace_with_no_overflowerror(self):
+        # GH 25616
+        # casts to object without Exception from OverflowError
+        s = pd.Series([0, 1, 2, 3, 4])
+        result = s.replace([3], ["100000000000000000000"])
+        expected = pd.Series([0, 1, 2, "100000000000000000000", 4])
+        tm.assert_series_equal(result, expected)
+
+        s = pd.Series([0, "100000000000000000000", "100000000000000000001"])
+        result = s.replace(["100000000000000000000"], [1])
+        expected = pd.Series([0, 1, "100000000000000000001"])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ser, to_replace, exp",
+        [
+            ([1, 2, 3], {1: 2, 2: 3, 3: 4}, [2, 3, 4]),
+            (["1", "2", "3"], {"1": "2", "2": "3", "3": "4"}, ["2", "3", "4"]),
+        ],
+    )
+    def test_replace_commutative(self, ser, to_replace, exp):
+        # GH 16051
+        # DataFrame.replace() overwrites when values are non-numeric
+
+        series = pd.Series(ser)
+
+        expected = pd.Series(exp)
+        result = series.replace(to_replace)
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ser, exp", [([1, 2, 3], [1, True, 3]), (["x", 2, 3], ["x", True, 3])]
+    )
+    def test_replace_no_cast(self, ser, exp):
+        # GH 9113
+        # BUG: replace int64 dtype with bool coerces to int64
+
+        series = pd.Series(ser)
+        result = series.replace(2, True)
+        expected = pd.Series(exp)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_invalid_to_replace(self):
+        # GH 18634
+        # API: replace() should raise an exception if invalid argument is given
+        series = pd.Series(["a", "b", "c "])
+        msg = (
+            r"Expecting 'to_replace' to be either a scalar, array-like, "
+            r"dict or None, got invalid type.*"
+        )
+        with pytest.raises(TypeError, match=msg):
+            series.replace(lambda x: x.strip())
+
+    @pytest.mark.parametrize("frame", [False, True])
+    def test_replace_nonbool_regex(self, frame):
+        obj = pd.Series(["a", "b", "c "])
+        if frame:
+            obj = obj.to_frame()
+
+        msg = "'to_replace' must be 'None' if 'regex' is not a bool"
+        with pytest.raises(ValueError, match=msg):
+            obj.replace(to_replace=["a"], regex="foo")
+
+    @pytest.mark.parametrize("frame", [False, True])
+    def test_replace_empty_copy(self, frame):
+        obj = pd.Series([], dtype=np.float64)
+        if frame:
+            obj = obj.to_frame()
+
+        res = obj.replace(4, 5, inplace=True)
+        assert res is None
+
+        res = obj.replace(4, 5, inplace=False)
+        tm.assert_equal(res, obj)
+        assert res is not obj
+
+    def test_replace_only_one_dictlike_arg(self, fixed_now_ts):
+        # GH#33340
+
+        ser = pd.Series([1, 2, "A", fixed_now_ts, True])
+        to_replace = {0: 1, 2: "A"}
+        value = "foo"
+        msg = "Series.replace cannot use dict-like to_replace and non-None value"
+        with pytest.raises(ValueError, match=msg):
+            ser.replace(to_replace, value)
+
+        to_replace = 1
+        value = {0: "foo", 2: "bar"}
+        msg = "Series.replace cannot use dict-value and non-None to_replace"
+        with pytest.raises(ValueError, match=msg):
+            ser.replace(to_replace, value)
+
+    def test_replace_dict_like_with_dict_like(self):
+        # GH 59452
+        s = pd.Series([1, 2, 3, 4, 5])
+        to_replace = pd.Series([1])
+        value = pd.Series([75])
+        msg = "to_replace and value cannot be dict-like for Series.replace"
+        with pytest.raises(ValueError, match=msg):
+            s.replace(to_replace, value)
+
+    def test_replace_extension_other(self, frame_or_series):
+        # https://github.com/pandas-dev/pandas/issues/34530
+        obj = frame_or_series(pd.array([1, 2, 3], dtype="Int64"))
+        result = obj.replace("", "")  # no exception
+        # should not have changed dtype
+        tm.assert_equal(obj, result)
+
+    def test_replace_with_compiled_regex(self):
+        # https://github.com/pandas-dev/pandas/issues/35680
+        s = pd.Series(["a", "b", "c"])
+        regex = re.compile("^a$")
+        result = s.replace({regex: "z"}, regex=True)
+        expected = pd.Series(["z", "b", "c"])
+        tm.assert_series_equal(result, expected)
+
+    def test_pandas_replace_na(self):
+        # GH#43344
+        # GH#56599
+        ser = pd.Series(["AA", "BB", "CC", "DD", "EE", "", pd.NA, "AA"], dtype="string")
+        regex_mapping = {
+            "AA": "CC",
+            "BB": "CC",
+            "EE": "CC",
+            "CC": "CC-REPL",
+        }
+        result = ser.replace(regex_mapping, regex=True)
+        exp = pd.Series(
+            ["CC", "CC", "CC-REPL", "DD", "CC", "", pd.NA, "CC"], dtype="string"
+        )
+        tm.assert_series_equal(result, exp)
+
+    @pytest.mark.parametrize(
+        "dtype, input_data, to_replace, expected_data",
+        [
+            ("bool", [True, False], {True: False}, [False, False]),
+            ("int64", [1, 2], {1: 10, 2: 20}, [10, 20]),
+            ("Int64", [1, 2], {1: 10, 2: 20}, [10, 20]),
+            ("float64", [1.1, 2.2], {1.1: 10.1, 2.2: 20.5}, [10.1, 20.5]),
+            ("Float64", [1.1, 2.2], {1.1: 10.1, 2.2: 20.5}, [10.1, 20.5]),
+            ("string", ["one", "two"], {"one": "1", "two": "2"}, ["1", "2"]),
+            (
+                pd.IntervalDtype("int64"),
+                IntervalArray([pd.Interval(1, 2), pd.Interval(2, 3)]),
+                {pd.Interval(1, 2): pd.Interval(10, 20)},
+                IntervalArray([pd.Interval(10, 20), pd.Interval(2, 3)]),
+            ),
+            (
+                pd.IntervalDtype("float64"),
+                IntervalArray([pd.Interval(1.0, 2.7), pd.Interval(2.8, 3.1)]),
+                {pd.Interval(1.0, 2.7): pd.Interval(10.6, 20.8)},
+                IntervalArray([pd.Interval(10.6, 20.8), pd.Interval(2.8, 3.1)]),
+            ),
+            (
+                pd.PeriodDtype("M"),
+                [pd.Period("2020-05", freq="M")],
+                {pd.Period("2020-05", freq="M"): pd.Period("2020-06", freq="M")},
+                [pd.Period("2020-06", freq="M")],
+            ),
+        ],
+    )
+    def test_replace_dtype(self, dtype, input_data, to_replace, expected_data):
+        # GH#33484
+        ser = pd.Series(input_data, dtype=dtype)
+        result = ser.replace(to_replace)
+        expected = pd.Series(expected_data, dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_string_dtype(self):
+        # GH#40732, GH#44940
+        ser = pd.Series(["one", "two", np.nan], dtype="string")
+        res = ser.replace({"one": "1", "two": "2"})
+        expected = pd.Series(["1", "2", np.nan], dtype="string")
+        tm.assert_series_equal(res, expected)
+
+        # GH#31644
+        ser2 = pd.Series(["A", np.nan], dtype="string")
+        res2 = ser2.replace("A", "B")
+        expected2 = pd.Series(["B", np.nan], dtype="string")
+        tm.assert_series_equal(res2, expected2)
+
+        ser3 = pd.Series(["A", "B"], dtype="string")
+        res3 = ser3.replace("A", pd.NA)
+        expected3 = pd.Series([pd.NA, "B"], dtype="string")
+        tm.assert_series_equal(res3, expected3)
+
+    def test_replace_string_dtype_list_to_replace(self):
+        # GH#41215, GH#44940
+        ser = pd.Series(["abc", "def"], dtype="string")
+        res = ser.replace(["abc", "any other string"], "xyz")
+        expected = pd.Series(["xyz", "def"], dtype="string")
+        tm.assert_series_equal(res, expected)
+
+    def test_replace_string_dtype_regex(self):
+        # GH#31644
+        ser = pd.Series(["A", "B"], dtype="string")
+        res = ser.replace(r".", "C", regex=True)
+        expected = pd.Series(["C", "C"], dtype="string")
+        tm.assert_series_equal(res, expected)
+
+    def test_replace_nullable_numeric(self):
+        # GH#40732, GH#44940
+
+        floats = pd.Series([1.0, 2.0, 3.999, 4.4], dtype=pd.Float64Dtype())
+        assert floats.replace({1.0: 9}).dtype == floats.dtype
+        assert floats.replace(1.0, 9).dtype == floats.dtype
+        assert floats.replace({1.0: 9.0}).dtype == floats.dtype
+        assert floats.replace(1.0, 9.0).dtype == floats.dtype
+
+        res = floats.replace(to_replace=[1.0, 2.0], value=[9.0, 10.0])
+        assert res.dtype == floats.dtype
+
+        ints = pd.Series([1, 2, 3, 4], dtype=pd.Int64Dtype())
+        assert ints.replace({1: 9}).dtype == ints.dtype
+        assert ints.replace(1, 9).dtype == ints.dtype
+        assert ints.replace({1: 9.0}).dtype == ints.dtype
+        assert ints.replace(1, 9.0).dtype == ints.dtype
+
+        # nullable (for now) raises instead of casting
+        with pytest.raises(TypeError, match="Invalid value"):
+            ints.replace({1: 9.5})
+        with pytest.raises(TypeError, match="Invalid value"):
+            ints.replace(1, 9.5)
+
+    @pytest.mark.parametrize("regex", [False, True])
+    def test_replace_regex_dtype_series(self, regex):
+        # GH-48644
+        series = pd.Series(["0"], dtype=object)
+        expected = pd.Series([1], dtype=object)
+        result = series.replace(to_replace="0", value=1, regex=regex)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("regex", [False, True])
+    def test_replace_regex_dtype_series_string(self, regex):
+        series = pd.Series(["0"], dtype="str")
+        expected = pd.Series([1], dtype=object)
+        result = series.replace(to_replace="0", value=1, regex=regex)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_different_int_types(self, any_int_numpy_dtype):
+        # GH#45311
+        labs = pd.Series([1, 1, 1, 0, 0, 2, 2, 2], dtype=any_int_numpy_dtype)
+
+        maps = pd.Series([0, 2, 1], dtype=any_int_numpy_dtype)
+        map_dict = dict(zip(maps.values, maps.index))
+
+        result = labs.replace(map_dict)
+        expected = labs.replace({0: 0, 2: 1, 1: 2})
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("val", [2, np.nan, 2.0])
+    def test_replace_value_none_dtype_numeric(self, val):
+        # GH#48231
+        ser = pd.Series([1, val])
+        result = ser.replace(val, None)
+        expected = pd.Series([1, None], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_change_dtype_series(self):
+        # GH#25797
+        df = pd.DataFrame({"Test": ["0.5", True, "0.6"]}, dtype=object)
+        df["Test"] = df["Test"].replace([True], [np.nan])
+        expected = pd.DataFrame({"Test": ["0.5", np.nan, "0.6"]}, dtype=object)
+        tm.assert_frame_equal(df, expected)
+
+        df = pd.DataFrame({"Test": ["0.5", None, "0.6"]}, dtype=object)
+        df["Test"] = df["Test"].replace([None], [np.nan])
+        tm.assert_frame_equal(df, expected)
+
+        df = pd.DataFrame({"Test": ["0.5", None, "0.6"]}, dtype=object)
+        df["Test"] = df["Test"].fillna(np.nan)
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("dtype", ["object", "Int64"])
+    def test_replace_na_in_obj_column(self, dtype):
+        # GH#47480
+        ser = pd.Series([0, 1, pd.NA], dtype=dtype)
+        expected = pd.Series([0, 2, pd.NA], dtype=dtype)
+        result = ser.replace(to_replace=1, value=2)
+        tm.assert_series_equal(result, expected)
+
+        ser.replace(to_replace=1, value=2, inplace=True)
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize("val", [0, 0.5])
+    def test_replace_numeric_column_with_na(self, val):
+        # GH#50758
+        ser = pd.Series([val, 1])
+        expected = pd.Series([val, pd.NA])
+        result = ser.replace(to_replace=1, value=pd.NA)
+        tm.assert_series_equal(result, expected)
+
+        ser.replace(to_replace=1, value=pd.NA, inplace=True)
+        tm.assert_series_equal(ser, expected)
+
+    def test_replace_ea_float_with_bool(self):
+        # GH#55398
+        ser = pd.Series([0.0], dtype="Float64")
+        expected = ser.copy()
+        result = ser.replace(False, 1.0)
+        tm.assert_series_equal(result, expected)
+
+        ser = pd.Series([False], dtype="boolean")
+        expected = ser.copy()
+        result = ser.replace(0.0, True)
+        tm.assert_series_equal(result, expected)
+
+    def test_replace_all_NA(self):
+        # GH#60688
+        df = pd.Series([pd.NA, pd.NA])
+        result = df.replace({r"^#": "$"}, regex=True)
+        expected = pd.Series([pd.NA, pd.NA])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_reset_index.py b/pandas/pandas/tests/series/methods/test_reset_index.py
new file mode 100644
index 0000000000000..d42aafc001680
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_reset_index.py
@@ -0,0 +1,225 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    RangeIndex,
+    Series,
+    date_range,
+    option_context,
+)
+import pandas._testing as tm
+
+
+class TestResetIndex:
+    def test_reset_index_dti_round_trip(self):
+        dti = date_range(start="1/1/2001", end="6/1/2001", freq="D")._with_freq(None)
+        d1 = DataFrame({"v": np.random.default_rng(2).random(len(dti))}, index=dti)
+        d2 = d1.reset_index()
+        assert d2.dtypes.iloc[0] == np.dtype("M8[ns]")
+        d3 = d2.set_index("index")
+        tm.assert_frame_equal(d1, d3, check_names=False)
+
+        # GH#2329
+        stamp = datetime(2012, 11, 22)
+        df = DataFrame([[stamp, 12.1]], columns=["Date", "Value"])
+        df = df.set_index("Date")
+
+        assert df.index[0] == stamp
+        assert df.reset_index()["Date"].iloc[0] == stamp
+
+    def test_reset_index(self):
+        df = DataFrame(
+            1.1 * np.arange(120).reshape((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=Index([f"i-{i}" for i in range(30)], dtype=object),
+        )[:5]
+        ser = df.stack()
+        ser.index.names = ["hash", "category"]
+
+        ser.name = "value"
+        df = ser.reset_index()
+        assert "value" in df
+
+        df = ser.reset_index(name="value2")
+        assert "value2" in df
+
+        # check inplace
+        s = ser.reset_index(drop=True)
+        s2 = ser
+        return_value = s2.reset_index(drop=True, inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(s, s2)
+
+        # level
+        index = MultiIndex(
+            levels=[["bar"], ["one", "two", "three"], [0, 1]],
+            codes=[[0, 0, 0, 0, 0, 0], [0, 1, 2, 0, 1, 2], [0, 1, 0, 1, 0, 1]],
+        )
+        s = Series(np.random.default_rng(2).standard_normal(6), index=index)
+        rs = s.reset_index(level=1)
+        assert len(rs.columns) == 2
+
+        rs = s.reset_index(level=[0, 2], drop=True)
+        tm.assert_index_equal(rs.index, Index(index.get_level_values(1)))
+        assert isinstance(rs, Series)
+
+    def test_reset_index_name(self):
+        s = Series([1, 2, 3], index=Index(range(3), name="x"))
+        assert s.reset_index().index.name is None
+        assert s.reset_index(drop=True).index.name is None
+
+    def test_reset_index_level(self):
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "B", "C"])
+
+        for levels in ["A", "B"], [0, 1]:
+            # With MultiIndex
+            s = df.set_index(["A", "B"])["C"]
+
+            result = s.reset_index(level=levels[0])
+            tm.assert_frame_equal(result, df.set_index("B"))
+
+            result = s.reset_index(level=levels[:1])
+            tm.assert_frame_equal(result, df.set_index("B"))
+
+            result = s.reset_index(level=levels)
+            tm.assert_frame_equal(result, df)
+
+            result = df.set_index(["A", "B"]).reset_index(level=levels, drop=True)
+            tm.assert_frame_equal(result, df[["C"]])
+
+            with pytest.raises(KeyError, match="Level E "):
+                s.reset_index(level=["A", "E"])
+
+            # With single-level Index
+            s = df.set_index("A")["B"]
+
+            result = s.reset_index(level=levels[0])
+            tm.assert_frame_equal(result, df[["A", "B"]])
+
+            result = s.reset_index(level=levels[:1])
+            tm.assert_frame_equal(result, df[["A", "B"]])
+
+            result = s.reset_index(level=levels[0], drop=True)
+            tm.assert_series_equal(result, df["B"])
+
+            with pytest.raises(IndexError, match="Too many levels"):
+                s.reset_index(level=[0, 1, 2])
+
+        # Check that .reset_index([],drop=True) doesn't fail
+        result = Series(range(4)).reset_index([], drop=True)
+        expected = Series(range(4))
+        tm.assert_series_equal(result, expected)
+
+    def test_reset_index_range(self):
+        # GH 12071
+        s = Series(range(2), name="A", dtype="int64")
+        series_result = s.reset_index()
+        assert isinstance(series_result.index, RangeIndex)
+        series_expected = DataFrame(
+            [[0, 0], [1, 1]], columns=["index", "A"], index=RangeIndex(stop=2)
+        )
+        tm.assert_frame_equal(series_result, series_expected)
+
+    def test_reset_index_drop_errors(self):
+        #  GH 20925
+
+        # KeyError raised for series index when passed level name is missing
+        s = Series(range(4))
+        with pytest.raises(KeyError, match="does not match index name"):
+            s.reset_index("wrong", drop=True)
+        with pytest.raises(KeyError, match="does not match index name"):
+            s.reset_index("wrong")
+
+        # KeyError raised for series when level to be dropped is missing
+        s = Series(range(4), index=MultiIndex.from_product([[1, 2]] * 2))
+        with pytest.raises(KeyError, match="not found"):
+            s.reset_index("wrong", drop=True)
+
+    def test_reset_index_with_drop(self):
+        arrays = [
+            ["bar", "bar", "baz", "baz", "qux", "qux", "foo", "foo"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+        ]
+        tuples = zip(*arrays)
+        index = MultiIndex.from_tuples(tuples)
+        data = np.random.default_rng(2).standard_normal(8)
+        ser = Series(data, index=index)
+        ser.iloc[3] = np.nan
+
+        deleveled = ser.reset_index()
+        assert isinstance(deleveled, DataFrame)
+        assert len(deleveled.columns) == len(ser.index.levels) + 1
+        assert deleveled.index.name == ser.index.name
+
+        deleveled = ser.reset_index(drop=True)
+        assert isinstance(deleveled, Series)
+        assert deleveled.index.name == ser.index.name
+
+    def test_reset_index_inplace_and_drop_ignore_name(self):
+        # GH#44575
+        ser = Series(range(2), name="old")
+        ser.reset_index(name="new", drop=True, inplace=True)
+        expected = Series(range(2), name="old")
+        tm.assert_series_equal(ser, expected)
+
+    def test_reset_index_drop_infer_string(self):
+        # GH#56160
+        pytest.importorskip("pyarrow")
+        ser = Series(["a", "b", "c"], dtype=object)
+        with option_context("future.infer_string", True):
+            result = ser.reset_index(drop=True)
+        tm.assert_series_equal(result, ser)
+
+
+@pytest.mark.parametrize(
+    "array, dtype",
+    [
+        (["a", "b"], object),
+        (
+            pd.period_range("12-1-2000", periods=2, freq="Q-DEC"),
+            pd.PeriodDtype(freq="Q-DEC"),
+        ),
+    ],
+)
+def test_reset_index_dtypes_on_empty_series_with_multiindex(
+    array, dtype, using_infer_string
+):
+    # GH 19602 - Preserve dtype on empty Series with MultiIndex
+    idx = MultiIndex.from_product([[0, 1], [0.5, 1.0], array])
+    result = Series(dtype=object, index=idx)[:0].reset_index().dtypes
+    exp = "str" if using_infer_string else object
+    expected = Series(
+        {
+            "level_0": np.int64,
+            "level_1": np.float64,
+            "level_2": exp if dtype == object else dtype,
+            0: object,
+        }
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "names, expected_names",
+    [
+        (["A", "A"], ["A", "A"]),
+        (["level_1", None], ["level_1", "level_1"]),
+    ],
+)
+@pytest.mark.parametrize("allow_duplicates", [False, True])
+def test_column_name_duplicates(names, expected_names, allow_duplicates):
+    # GH#44755 reset_index with duplicate column labels
+    s = Series([1], index=MultiIndex.from_arrays([[1], [1]], names=names))
+    if allow_duplicates:
+        result = s.reset_index(allow_duplicates=True)
+        expected = DataFrame([[1, 1, 1]], columns=expected_names + [0])
+        tm.assert_frame_equal(result, expected)
+    else:
+        with pytest.raises(ValueError, match="cannot insert"):
+            s.reset_index()
diff --git a/pandas/pandas/tests/series/methods/test_round.py b/pandas/pandas/tests/series/methods/test_round.py
new file mode 100644
index 0000000000000..c330b7a7dfbbb
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_round.py
@@ -0,0 +1,74 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestSeriesRound:
+    def test_round(self, datetime_series):
+        datetime_series.index.name = "index_name"
+        result = datetime_series.round(2)
+        expected = Series(
+            np.round(datetime_series.values, 2), index=datetime_series.index, name="ts"
+        )
+        tm.assert_series_equal(result, expected)
+        assert result.name == datetime_series.name
+
+    def test_round_numpy(self, any_float_dtype):
+        # See GH#12600
+        ser = Series([1.53, 1.36, 0.06], dtype=any_float_dtype)
+        out = np.round(ser, decimals=0)
+        expected = Series([2.0, 1.0, 0.0], dtype=any_float_dtype)
+        tm.assert_series_equal(out, expected)
+
+        msg = "the 'out' parameter is not supported"
+        with pytest.raises(ValueError, match=msg):
+            np.round(ser, decimals=0, out=ser)
+
+    def test_round_numpy_with_nan(self, any_float_dtype):
+        # See GH#14197
+        ser = Series([1.53, np.nan, 0.06], dtype=any_float_dtype)
+        with tm.assert_produces_warning(None):
+            result = ser.round()
+        expected = Series([2.0, np.nan, 0.0], dtype=any_float_dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_round_builtin(self, any_float_dtype):
+        ser = Series(
+            [1.123, 2.123, 3.123],
+            index=range(3),
+            dtype=any_float_dtype,
+        )
+        result = round(ser)
+        expected_rounded0 = Series(
+            [1.0, 2.0, 3.0], index=range(3), dtype=any_float_dtype
+        )
+        tm.assert_series_equal(result, expected_rounded0)
+
+        decimals = 2
+        expected_rounded = Series(
+            [1.12, 2.12, 3.12], index=range(3), dtype=any_float_dtype
+        )
+        result = round(ser, decimals)
+        tm.assert_series_equal(result, expected_rounded)
+
+    @pytest.mark.parametrize("method", ["round", "floor", "ceil"])
+    @pytest.mark.parametrize("freq", ["s", "5s", "min", "5min", "h", "5h"])
+    def test_round_nat(self, method, freq, unit):
+        # GH14940, GH#56158
+        ser = Series([pd.NaT], dtype=f"M8[{unit}]")
+        expected = Series(pd.NaT, dtype=f"M8[{unit}]")
+        round_method = getattr(ser.dt, method)
+        result = round_method(freq)
+        tm.assert_series_equal(result, expected)
+
+    def test_round_ea_boolean(self):
+        # GH#55936
+        ser = Series([True, False], dtype="boolean")
+        expected = ser.copy()
+        result = ser.round(2)
+        tm.assert_series_equal(result, expected)
+        result.iloc[0] = False
+        tm.assert_series_equal(ser, expected)
diff --git a/pandas/pandas/tests/series/methods/test_searchsorted.py b/pandas/pandas/tests/series/methods/test_searchsorted.py
new file mode 100644
index 0000000000000..239496052b99b
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_searchsorted.py
@@ -0,0 +1,77 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.types import is_scalar
+
+
+class TestSeriesSearchSorted:
+    def test_searchsorted(self):
+        ser = Series([1, 2, 3])
+
+        result = ser.searchsorted(1, side="left")
+        assert is_scalar(result)
+        assert result == 0
+
+        result = ser.searchsorted(1, side="right")
+        assert is_scalar(result)
+        assert result == 1
+
+    def test_searchsorted_numeric_dtypes_scalar(self):
+        ser = Series([1, 2, 90, 1000, 3e9])
+        res = ser.searchsorted(30)
+        assert is_scalar(res)
+        assert res == 2
+
+        res = ser.searchsorted([30])
+        exp = np.array([2], dtype=np.intp)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_searchsorted_numeric_dtypes_vector(self):
+        ser = Series([1, 2, 90, 1000, 3e9])
+        res = ser.searchsorted([91, 2e6])
+        exp = np.array([3, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_searchsorted_datetime64_scalar(self):
+        ser = Series(date_range("20120101", periods=10, freq="2D"))
+        val = Timestamp("20120102")
+        res = ser.searchsorted(val)
+        assert is_scalar(res)
+        assert res == 1
+
+    def test_searchsorted_datetime64_scalar_mixed_timezones(self):
+        # GH 30086
+        ser = Series(date_range("20120101", periods=10, freq="2D", tz="UTC"))
+        val = Timestamp("20120102", tz="America/New_York")
+        res = ser.searchsorted(val)
+        assert is_scalar(res)
+        assert res == 1
+
+    def test_searchsorted_datetime64_list(self):
+        ser = Series(date_range("20120101", periods=10, freq="2D"))
+        vals = [Timestamp("20120102"), Timestamp("20120104")]
+        res = ser.searchsorted(vals)
+        exp = np.array([1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_searchsorted_sorter(self):
+        # GH8490
+        ser = Series([3, 1, 2])
+        res = ser.searchsorted([0, 3], sorter=np.argsort(ser))
+        exp = np.array([0, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_searchsorted_dataframe_fail(self):
+        # GH#49620
+        ser = Series([1, 2, 3, 4, 5])
+        vals = pd.DataFrame([[1, 2], [3, 4]])
+        msg = "Value must be 1-D array-like or scalar, DataFrame is not supported"
+        with pytest.raises(ValueError, match=msg):
+            ser.searchsorted(vals)
diff --git a/pandas/pandas/tests/series/methods/test_set_name.py b/pandas/pandas/tests/series/methods/test_set_name.py
new file mode 100644
index 0000000000000..137207053c225
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_set_name.py
@@ -0,0 +1,21 @@
+from datetime import datetime
+
+from pandas import Series
+
+
+class TestSetName:
+    def test_set_name(self):
+        ser = Series([1, 2, 3])
+        ser2 = ser._set_name("foo")
+        assert ser2.name == "foo"
+        assert ser.name is None
+        assert ser is not ser2
+
+    def test_set_name_attribute(self):
+        ser = Series([1, 2, 3])
+        ser2 = Series([1, 2, 3], name="bar")
+        for name in [7, 7.0, "name", datetime(2001, 1, 1), (1,), "\u05d0"]:
+            ser.name = name
+            assert ser.name == name
+            ser2.name = name
+            assert ser2.name == name
diff --git a/pandas/pandas/tests/series/methods/test_size.py b/pandas/pandas/tests/series/methods/test_size.py
new file mode 100644
index 0000000000000..043e4b66dbf16
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_size.py
@@ -0,0 +1,20 @@
+import pytest
+
+from pandas import Series
+
+
+@pytest.mark.parametrize(
+    "data, index, expected",
+    [
+        ([1, 2, 3], None, 3),
+        ({"a": 1, "b": 2, "c": 3}, None, 3),
+        ([1, 2, 3], ["x", "y", "z"], 3),
+        ([1, 2, 3, 4, 5], ["x", "y", "z", "w", "n"], 5),
+        ([1, 2, 3, 4], ["x", "y", "z", "w"], 4),
+    ],
+)
+def test_series(data, index, expected):
+    # GH#52897
+    ser = Series(data, index=index)
+    assert ser.size == expected
+    assert isinstance(ser.size, int)
diff --git a/pandas/pandas/tests/series/methods/test_sort_index.py b/pandas/pandas/tests/series/methods/test_sort_index.py
new file mode 100644
index 0000000000000..d6817aa179b7b
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_sort_index.py
@@ -0,0 +1,337 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    IntervalIndex,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(params=["quicksort", "mergesort", "heapsort", "stable"])
+def sort_kind(request):
+    return request.param
+
+
+class TestSeriesSortIndex:
+    def test_sort_index_name(self, datetime_series):
+        result = datetime_series.sort_index(ascending=False)
+        assert result.name == datetime_series.name
+
+    def test_sort_index(self, datetime_series):
+        datetime_series.index = datetime_series.index._with_freq(None)
+
+        rindex = list(datetime_series.index)
+        np.random.default_rng(2).shuffle(rindex)
+
+        random_order = datetime_series.reindex(rindex)
+        sorted_series = random_order.sort_index()
+        tm.assert_series_equal(sorted_series, datetime_series)
+
+        # descending
+        sorted_series = random_order.sort_index(ascending=False)
+        tm.assert_series_equal(
+            sorted_series, datetime_series.reindex(datetime_series.index[::-1])
+        )
+
+        # compat on level
+        sorted_series = random_order.sort_index(level=0)
+        tm.assert_series_equal(sorted_series, datetime_series)
+
+        # compat on axis
+        sorted_series = random_order.sort_index(axis=0)
+        tm.assert_series_equal(sorted_series, datetime_series)
+
+        msg = "No axis named 1 for object type Series"
+        with pytest.raises(ValueError, match=msg):
+            random_order.sort_values(axis=1)
+
+        sorted_series = random_order.sort_index(level=0, axis=0)
+        tm.assert_series_equal(sorted_series, datetime_series)
+
+        with pytest.raises(ValueError, match=msg):
+            random_order.sort_index(level=0, axis=1)
+
+    def test_sort_index_inplace(self, datetime_series):
+        datetime_series.index = datetime_series.index._with_freq(None)
+
+        # For GH#11402
+        rindex = list(datetime_series.index)
+        np.random.default_rng(2).shuffle(rindex)
+
+        # descending
+        random_order = datetime_series.reindex(rindex)
+        result = random_order.sort_index(ascending=False, inplace=True)
+
+        assert result is None
+        expected = datetime_series.reindex(datetime_series.index[::-1])
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(random_order, expected)
+
+        # ascending
+        random_order = datetime_series.reindex(rindex)
+        result = random_order.sort_index(ascending=True, inplace=True)
+
+        assert result is None
+        expected = datetime_series.copy()
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(random_order, expected)
+
+    def test_sort_index_level(self):
+        mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+        s = Series([1, 2], mi)
+        backwards = s.iloc[[1, 0]]
+
+        res = s.sort_index(level="A")
+        tm.assert_series_equal(backwards, res)
+
+        res = s.sort_index(level=["A", "B"])
+        tm.assert_series_equal(backwards, res)
+
+        res = s.sort_index(level="A", sort_remaining=False)
+        tm.assert_series_equal(s, res)
+
+        res = s.sort_index(level=["A", "B"], sort_remaining=False)
+        tm.assert_series_equal(s, res)
+
+    @pytest.mark.parametrize("level", ["A", 0])  # GH#21052
+    def test_sort_index_multiindex(self, level):
+        mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+        s = Series([1, 2], mi)
+        backwards = s.iloc[[1, 0]]
+
+        # implicit sort_remaining=True
+        res = s.sort_index(level=level)
+        tm.assert_series_equal(backwards, res)
+
+        # GH#13496
+        # sort has no effect without remaining lvls
+        res = s.sort_index(level=level, sort_remaining=False)
+        tm.assert_series_equal(s, res)
+
+    def test_sort_index_kind(self, sort_kind):
+        # GH#14444 & GH#13589:  Add support for sort algo choosing
+        series = Series(index=[3, 2, 1, 4, 3], dtype=object)
+        expected_series = Series(index=[1, 2, 3, 3, 4], dtype=object)
+
+        index_sorted_series = series.sort_index(kind=sort_kind)
+        tm.assert_series_equal(expected_series, index_sorted_series)
+
+    def test_sort_index_na_position(self):
+        series = Series(index=[3, 2, 1, 4, 3, np.nan], dtype=object)
+        expected_series_first = Series(index=[np.nan, 1, 2, 3, 3, 4], dtype=object)
+
+        index_sorted_series = series.sort_index(na_position="first")
+        tm.assert_series_equal(expected_series_first, index_sorted_series)
+
+        expected_series_last = Series(index=[1, 2, 3, 3, 4, np.nan], dtype=object)
+
+        index_sorted_series = series.sort_index(na_position="last")
+        tm.assert_series_equal(expected_series_last, index_sorted_series)
+
+    def test_sort_index_intervals(self):
+        s = Series(
+            [np.nan, 1, 2, 3], IntervalIndex.from_arrays([0, 1, 2, 3], [1, 2, 3, 4])
+        )
+
+        result = s.sort_index()
+        expected = s
+        tm.assert_series_equal(result, expected)
+
+        result = s.sort_index(ascending=False)
+        expected = Series(
+            [3, 2, 1, np.nan], IntervalIndex.from_arrays([3, 2, 1, 0], [4, 3, 2, 1])
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize(
+        "original_list, sorted_list, ascending, ignore_index, output_index",
+        [
+            ([2, 3, 6, 1], [2, 3, 6, 1], True, True, [0, 1, 2, 3]),
+            ([2, 3, 6, 1], [2, 3, 6, 1], True, False, [0, 1, 2, 3]),
+            ([2, 3, 6, 1], [1, 6, 3, 2], False, True, [0, 1, 2, 3]),
+            ([2, 3, 6, 1], [1, 6, 3, 2], False, False, [3, 2, 1, 0]),
+        ],
+    )
+    def test_sort_index_ignore_index(
+        self, inplace, original_list, sorted_list, ascending, ignore_index, output_index
+    ):
+        # GH 30114
+        ser = Series(original_list)
+        expected = Series(sorted_list, index=output_index)
+        kwargs = {
+            "ascending": ascending,
+            "ignore_index": ignore_index,
+            "inplace": inplace,
+        }
+
+        if inplace:
+            result_ser = ser.copy()
+            result_ser.sort_index(**kwargs)
+        else:
+            result_ser = ser.sort_index(**kwargs)
+
+        tm.assert_series_equal(result_ser, expected)
+        tm.assert_series_equal(ser, Series(original_list))
+
+    def test_sort_index_ascending_list(self):
+        # GH#16934
+
+        # Set up a Series with a three level MultiIndex
+        arrays = [
+            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
+            ["one", "two", "one", "two", "one", "two", "one", "two"],
+            [4, 3, 2, 1, 4, 3, 2, 1],
+        ]
+        tuples = zip(*arrays)
+        mi = MultiIndex.from_tuples(tuples, names=["first", "second", "third"])
+        ser = Series(range(8), index=mi)
+
+        # Sort with boolean ascending
+        result = ser.sort_index(level=["third", "first"], ascending=False)
+        expected = ser.iloc[[4, 0, 5, 1, 6, 2, 7, 3]]
+        tm.assert_series_equal(result, expected)
+
+        # Sort with list of boolean ascending
+        result = ser.sort_index(level=["third", "first"], ascending=[False, True])
+        expected = ser.iloc[[0, 4, 1, 5, 2, 6, 3, 7]]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ascending",
+        [
+            None,
+            (True, None),
+            (False, "True"),
+        ],
+    )
+    def test_sort_index_ascending_bad_value_raises(self, ascending):
+        ser = Series(range(10), index=[0, 3, 2, 1, 4, 5, 7, 6, 8, 9])
+        match = 'For argument "ascending" expected type bool'
+        with pytest.raises(ValueError, match=match):
+            ser.sort_index(ascending=ascending)
+
+
+class TestSeriesSortIndexKey:
+    def test_sort_index_multiindex_key(self):
+        mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+        s = Series([1, 2], mi)
+        backwards = s.iloc[[1, 0]]
+
+        result = s.sort_index(level="C", key=lambda x: -x)
+        tm.assert_series_equal(s, result)
+
+        result = s.sort_index(level="C", key=lambda x: x)  # nothing happens
+        tm.assert_series_equal(backwards, result)
+
+    def test_sort_index_multiindex_key_multi_level(self):
+        mi = MultiIndex.from_tuples([[1, 1, 3], [1, 1, 1]], names=list("ABC"))
+        s = Series([1, 2], mi)
+        backwards = s.iloc[[1, 0]]
+
+        result = s.sort_index(level=["A", "C"], key=lambda x: -x)
+        tm.assert_series_equal(s, result)
+
+        result = s.sort_index(level=["A", "C"], key=lambda x: x)  # nothing happens
+        tm.assert_series_equal(backwards, result)
+
+    def test_sort_index_key(self):
+        series = Series(np.arange(6, dtype="int64"), index=list("aaBBca"))
+
+        result = series.sort_index()
+        expected = series.iloc[[2, 3, 0, 1, 5, 4]]
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_index(key=lambda x: x.str.lower())
+        expected = series.iloc[[0, 1, 5, 2, 3, 4]]
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_index(key=lambda x: x.str.lower(), ascending=False)
+        expected = series.iloc[[4, 2, 3, 0, 1, 5]]
+        tm.assert_series_equal(result, expected)
+
+    def test_sort_index_key_int(self):
+        series = Series(np.arange(6, dtype="int64"), index=np.arange(6, dtype="int64"))
+
+        result = series.sort_index()
+        tm.assert_series_equal(result, series)
+
+        result = series.sort_index(key=lambda x: -x)
+        expected = series.sort_index(ascending=False)
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_index(key=lambda x: 2 * x)
+        tm.assert_series_equal(result, series)
+
+    def test_sort_index_kind_key(self, sort_kind, sort_by_key):
+        # GH #14444 & #13589:  Add support for sort algo choosing
+        series = Series(index=[3, 2, 1, 4, 3], dtype=object)
+        expected_series = Series(index=[1, 2, 3, 3, 4], dtype=object)
+
+        index_sorted_series = series.sort_index(kind=sort_kind, key=sort_by_key)
+        tm.assert_series_equal(expected_series, index_sorted_series)
+
+    def test_sort_index_kind_neg_key(self, sort_kind):
+        # GH #14444 & #13589:  Add support for sort algo choosing
+        series = Series(index=[3, 2, 1, 4, 3], dtype=object)
+        expected_series = Series(index=[4, 3, 3, 2, 1], dtype=object)
+
+        index_sorted_series = series.sort_index(kind=sort_kind, key=lambda x: -x)
+        tm.assert_series_equal(expected_series, index_sorted_series)
+
+    def test_sort_index_na_position_key(self, sort_by_key):
+        series = Series(index=[3, 2, 1, 4, 3, np.nan], dtype=object)
+        expected_series_first = Series(index=[np.nan, 1, 2, 3, 3, 4], dtype=object)
+
+        index_sorted_series = series.sort_index(na_position="first", key=sort_by_key)
+        tm.assert_series_equal(expected_series_first, index_sorted_series)
+
+        expected_series_last = Series(index=[1, 2, 3, 3, 4, np.nan], dtype=object)
+
+        index_sorted_series = series.sort_index(na_position="last", key=sort_by_key)
+        tm.assert_series_equal(expected_series_last, index_sorted_series)
+
+    def test_changes_length_raises(self):
+        s = Series([1, 2, 3])
+        with pytest.raises(ValueError, match="change the shape"):
+            s.sort_index(key=lambda x: x[:1])
+
+    def test_sort_values_key_type(self):
+        s = Series([1, 2, 3], DatetimeIndex(["2008-10-24", "2008-11-23", "2007-12-22"]))
+
+        result = s.sort_index(key=lambda x: x.month)
+        expected = s.iloc[[0, 1, 2]]
+        tm.assert_series_equal(result, expected)
+
+        result = s.sort_index(key=lambda x: x.day)
+        expected = s.iloc[[2, 1, 0]]
+        tm.assert_series_equal(result, expected)
+
+        result = s.sort_index(key=lambda x: x.year)
+        expected = s.iloc[[2, 0, 1]]
+        tm.assert_series_equal(result, expected)
+
+        result = s.sort_index(key=lambda x: x.month_name())
+        expected = s.iloc[[2, 1, 0]]
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ascending",
+        [
+            [True, False],
+            [False, True],
+        ],
+    )
+    def test_sort_index_multi_already_monotonic(self, ascending):
+        # GH 56049
+        mi = MultiIndex.from_product([[1, 2], [3, 4]])
+        ser = Series(range(len(mi)), index=mi)
+        result = ser.sort_index(ascending=ascending)
+        if ascending == [True, False]:
+            expected = ser.take([1, 0, 3, 2])
+        elif ascending == [False, True]:
+            expected = ser.take([2, 3, 0, 1])
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_sort_values.py b/pandas/pandas/tests/series/methods/test_sort_values.py
new file mode 100644
index 0000000000000..995d9d176fc2b
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_sort_values.py
@@ -0,0 +1,235 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestSeriesSortValues:
+    def test_sort_values(self, datetime_series):
+        # check indexes are reordered corresponding with the values
+        ser = Series([3, 2, 4, 1], ["A", "B", "C", "D"])
+        expected = Series([1, 2, 3, 4], ["D", "B", "A", "C"])
+        result = ser.sort_values()
+        tm.assert_series_equal(expected, result)
+
+        ts = datetime_series.copy()
+        ts[:5] = np.nan
+        vals = ts.values
+
+        result = ts.sort_values()
+        assert np.isnan(result[-5:]).all()
+        tm.assert_numpy_array_equal(result[:-5].values, np.sort(vals[5:]))
+
+        # na_position
+        result = ts.sort_values(na_position="first")
+        assert np.isnan(result[:5]).all()
+        tm.assert_numpy_array_equal(result[5:].values, np.sort(vals[5:]))
+
+        # something object-type
+        ser = Series(["A", "B"], [1, 2])
+        # no failure
+        ser.sort_values()
+
+        # ascending=False
+        ordered = ts.sort_values(ascending=False)
+        expected = np.sort(ts.dropna().values)[::-1]
+        tm.assert_almost_equal(expected, ordered.dropna().values)
+        ordered = ts.sort_values(ascending=False, na_position="first")
+        tm.assert_almost_equal(expected, ordered.dropna().values)
+
+        # ascending=[False] should behave the same as ascending=False
+        ordered = ts.sort_values(ascending=[False])
+        expected = ts.sort_values(ascending=False)
+        tm.assert_series_equal(expected, ordered)
+        ordered = ts.sort_values(ascending=[False], na_position="first")
+        expected = ts.sort_values(ascending=False, na_position="first")
+        tm.assert_series_equal(expected, ordered)
+
+        msg = 'For argument "ascending" expected type bool, received type NoneType.'
+        with pytest.raises(ValueError, match=msg):
+            ts.sort_values(ascending=None)
+        msg = r"Length of ascending \(0\) must be 1 for Series"
+        with pytest.raises(ValueError, match=msg):
+            ts.sort_values(ascending=[])
+        msg = r"Length of ascending \(3\) must be 1 for Series"
+        with pytest.raises(ValueError, match=msg):
+            ts.sort_values(ascending=[1, 2, 3])
+        msg = r"Length of ascending \(2\) must be 1 for Series"
+        with pytest.raises(ValueError, match=msg):
+            ts.sort_values(ascending=[False, False])
+        msg = 'For argument "ascending" expected type bool, received type str.'
+        with pytest.raises(ValueError, match=msg):
+            ts.sort_values(ascending="foobar")
+
+        # inplace=True
+        ts = datetime_series.copy()
+        return_value = ts.sort_values(ascending=False, inplace=True)
+        assert return_value is None
+        tm.assert_series_equal(ts, datetime_series.sort_values(ascending=False))
+        tm.assert_index_equal(
+            ts.index, datetime_series.sort_values(ascending=False).index
+        )
+
+        # GH#5856/5853
+        # Series.sort_values operating on a view
+        df = DataFrame(np.random.default_rng(2).standard_normal((10, 4)))
+        s = df.iloc[:, 0]
+        s.sort_values(inplace=True)
+        tm.assert_series_equal(s, df.iloc[:, 0].sort_values())
+
+    def test_sort_values_categorical(self):
+        cat = Series(Categorical(["a", "b", "b", "a"], ordered=False))
+
+        # sort in the categories order
+        expected = Series(
+            Categorical(["a", "a", "b", "b"], ordered=False), index=[0, 3, 1, 2]
+        )
+        result = cat.sort_values()
+        tm.assert_series_equal(result, expected)
+
+        cat = Series(Categorical(["a", "c", "b", "d"], ordered=True))
+        res = cat.sort_values()
+        exp = np.array(["a", "b", "c", "d"], dtype=np.object_)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+
+        cat = Series(
+            Categorical(
+                ["a", "c", "b", "d"], categories=["a", "b", "c", "d"], ordered=True
+            )
+        )
+        res = cat.sort_values()
+        exp = np.array(["a", "b", "c", "d"], dtype=np.object_)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+
+        res = cat.sort_values(ascending=False)
+        exp = np.array(["d", "c", "b", "a"], dtype=np.object_)
+        tm.assert_numpy_array_equal(res.__array__(), exp)
+
+        raw_cat1 = Categorical(
+            ["a", "b", "c", "d"], categories=["a", "b", "c", "d"], ordered=False
+        )
+        raw_cat2 = Categorical(
+            ["a", "b", "c", "d"], categories=["d", "c", "b", "a"], ordered=True
+        )
+        s = ["a", "b", "c", "d"]
+        df = DataFrame(
+            {"unsort": raw_cat1, "sort": raw_cat2, "string": s, "values": [1, 2, 3, 4]}
+        )
+
+        # Cats must be sorted in a dataframe
+        res = df.sort_values(by=["string"], ascending=False)
+        exp = np.array(["d", "c", "b", "a"], dtype=np.object_)
+        tm.assert_numpy_array_equal(res["sort"].values.__array__(), exp)
+        assert res["sort"].dtype == "category"
+
+        res = df.sort_values(by=["sort"], ascending=False)
+        exp = df.sort_values(by=["string"], ascending=True)
+        tm.assert_series_equal(res["values"], exp["values"])
+        assert res["sort"].dtype == "category"
+        assert res["unsort"].dtype == "category"
+
+        # unordered cat, but we allow this
+        df.sort_values(by=["unsort"], ascending=False)
+
+        # multi-columns sort
+        # GH#7848
+        df = DataFrame(
+            {"id": [6, 5, 4, 3, 2, 1], "raw_grade": ["a", "b", "b", "a", "a", "e"]}
+        )
+        df["grade"] = Categorical(df["raw_grade"], ordered=True)
+        df["grade"] = df["grade"].cat.set_categories(["b", "e", "a"])
+
+        # sorts 'grade' according to the order of the categories
+        result = df.sort_values(by=["grade"])
+        expected = df.iloc[[1, 2, 5, 0, 3, 4]]
+        tm.assert_frame_equal(result, expected)
+
+        # multi
+        result = df.sort_values(by=["grade", "id"])
+        expected = df.iloc[[2, 1, 5, 4, 3, 0]]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("inplace", [True, False])
+    @pytest.mark.parametrize(
+        "original_list, sorted_list, ignore_index, output_index",
+        [
+            ([2, 3, 6, 1], [6, 3, 2, 1], True, [0, 1, 2, 3]),
+            ([2, 3, 6, 1], [6, 3, 2, 1], False, [2, 1, 0, 3]),
+        ],
+    )
+    def test_sort_values_ignore_index(
+        self, inplace, original_list, sorted_list, ignore_index, output_index
+    ):
+        # GH 30114
+        ser = Series(original_list)
+        expected = Series(sorted_list, index=output_index)
+        kwargs = {"ignore_index": ignore_index, "inplace": inplace}
+
+        if inplace:
+            result_ser = ser.copy()
+            result_ser.sort_values(ascending=False, **kwargs)
+        else:
+            result_ser = ser.sort_values(ascending=False, **kwargs)
+
+        tm.assert_series_equal(result_ser, expected)
+        tm.assert_series_equal(ser, Series(original_list))
+
+    def test_mergesort_descending_stability(self):
+        # GH 28697
+        s = Series([1, 2, 1, 3], ["first", "b", "second", "c"])
+        result = s.sort_values(ascending=False, kind="mergesort")
+        expected = Series([3, 2, 1, 1], ["c", "b", "first", "second"])
+        tm.assert_series_equal(result, expected)
+
+    def test_sort_values_validate_ascending_for_value_error(self):
+        # GH41634
+        ser = Series([23, 7, 21])
+
+        msg = 'For argument "ascending" expected type bool, received type str.'
+        with pytest.raises(ValueError, match=msg):
+            ser.sort_values(ascending="False")
+
+    def test_sort_values_validate_ascending_functional(self, ascending):
+        # GH41634
+        ser = Series([23, 7, 21])
+        expected = np.sort(ser.values)
+
+        sorted_ser = ser.sort_values(ascending=ascending)
+        if not ascending:
+            expected = expected[::-1]
+
+        result = sorted_ser.values
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestSeriesSortingKey:
+    def test_sort_values_key(self):
+        series = Series(np.array(["Hello", "goodbye"]))
+
+        result = series.sort_values(axis=0)
+        expected = series
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_values(axis=0, key=lambda x: x.str.lower())
+        expected = series[::-1]
+        tm.assert_series_equal(result, expected)
+
+    def test_sort_values_key_nan(self):
+        series = Series(np.array([0, 5, np.nan, 3, 2, np.nan]))
+
+        result = series.sort_values(axis=0)
+        expected = series.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_values(axis=0, key=lambda x: x + 5)
+        expected = series.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_series_equal(result, expected)
+
+        result = series.sort_values(axis=0, key=lambda x: -x, ascending=False)
+        expected = series.iloc[[0, 4, 3, 1, 2, 5]]
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_to_csv.py b/pandas/pandas/tests/series/methods/test_to_csv.py
new file mode 100644
index 0000000000000..3e3eb36112680
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_to_csv.py
@@ -0,0 +1,177 @@
+from datetime import datetime
+from io import StringIO
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+from pandas.io.common import get_handle
+
+
+class TestSeriesToCSV:
+    def read_csv(self, path, **kwargs):
+        params = {"index_col": 0, "header": None}
+        params.update(**kwargs)
+
+        header = params.get("header")
+        out = pd.read_csv(path, **params).squeeze("columns")
+
+        if header is None:
+            out.name = out.index.name = None
+
+        return out
+
+    def test_from_csv(self, datetime_series, string_series, temp_file):
+        # freq doesn't round-trip
+        datetime_series.index = datetime_series.index._with_freq(None)
+
+        path = temp_file
+        datetime_series.to_csv(path, header=False)
+        ts = self.read_csv(path, parse_dates=True)
+        expected = datetime_series.copy()
+        expected.index = expected.index.as_unit("s")
+        tm.assert_series_equal(expected, ts, check_names=False)
+
+        assert ts.name is None
+        assert ts.index.name is None
+
+        # see gh-10483
+        datetime_series.to_csv(path, header=True)
+        ts_h = self.read_csv(path, header=0)
+        assert ts_h.name == "ts"
+
+        string_series.to_csv(path, header=False)
+        series = self.read_csv(path)
+        tm.assert_series_equal(string_series, series, check_names=False)
+
+        assert series.name is None
+        assert series.index.name is None
+
+        string_series.to_csv(path, header=True)
+        series_h = self.read_csv(path, header=0)
+        assert series_h.name == "series"
+
+        with open(path, "w", encoding="utf-8") as outfile:
+            outfile.write("1998-01-01|1.0\n1999-01-01|2.0")
+
+        series = self.read_csv(path, sep="|", parse_dates=True)
+        check_series = Series({datetime(1998, 1, 1): 1.0, datetime(1999, 1, 1): 2.0})
+        check_series.index = check_series.index.as_unit("s")
+        tm.assert_series_equal(check_series, series)
+
+        series = self.read_csv(path, sep="|", parse_dates=False)
+        check_series = Series({"1998-01-01": 1.0, "1999-01-01": 2.0})
+        tm.assert_series_equal(check_series, series)
+
+    def test_to_csv(self, datetime_series, temp_file):
+        datetime_series.to_csv(temp_file, header=False)
+
+        with open(temp_file, newline=None, encoding="utf-8") as f:
+            lines = f.readlines()
+        assert lines[1] != "\n"
+
+        datetime_series.to_csv(temp_file, index=False, header=False)
+        arr = np.loadtxt(temp_file)
+        tm.assert_almost_equal(arr, datetime_series.values)
+
+    def test_to_csv_unicode_index(self):
+        buf = StringIO()
+        s = Series(["\u05d0", "d2"], index=["\u05d0", "\u05d1"])
+
+        s.to_csv(buf, encoding="UTF-8", header=False)
+        buf.seek(0)
+
+        s2 = self.read_csv(buf, index_col=0, encoding="UTF-8")
+        tm.assert_series_equal(s, s2)
+
+    def test_to_csv_float_format(self, temp_file):
+        ser = Series([0.123456, 0.234567, 0.567567])
+        ser.to_csv(temp_file, float_format="%.2f", header=False)
+
+        rs = self.read_csv(temp_file)
+        xp = Series([0.12, 0.23, 0.57])
+        tm.assert_series_equal(rs, xp)
+
+    def test_to_csv_list_entries(self):
+        s = Series(["jack and jill", "jesse and frank"])
+
+        split = s.str.split(r"\s+and\s+")
+
+        buf = StringIO()
+        split.to_csv(buf, header=False)
+
+    def test_to_csv_path_is_none(self):
+        # GH 8215
+        # Series.to_csv() was returning None, inconsistent with
+        # DataFrame.to_csv() which returned string
+        s = Series([1, 2, 3])
+        csv_str = s.to_csv(path_or_buf=None, header=False)
+        assert isinstance(csv_str, str)
+
+    @pytest.mark.parametrize(
+        "s,encoding",
+        [
+            (
+                Series([0.123456, 0.234567, 0.567567], index=["A", "B", "C"], name="X"),
+                None,
+            ),
+            # GH 21241, 21118
+            (Series(["abc", "def", "ghi"], name="X"), "ascii"),
+            (Series(["123", "你好", "世界"], name="中文"), "gb2312"),
+            (
+                Series(["123", "Γειά σου", "Κόσμε"], name="Ελληνικά"),  # noqa: RUF001
+                "cp737",
+            ),
+        ],
+    )
+    def test_to_csv_compression(self, s, encoding, compression, temp_file):
+        filename = temp_file
+        s.to_csv(filename, compression=compression, encoding=encoding, header=True)
+        # test the round trip - to_csv -> read_csv
+        result = pd.read_csv(
+            filename,
+            compression=compression,
+            encoding=encoding,
+            index_col=0,
+        ).squeeze("columns")
+        tm.assert_series_equal(s, result)
+
+        # test the round trip using file handle - to_csv -> read_csv
+        with get_handle(
+            filename, "w", compression=compression, encoding=encoding
+        ) as handles:
+            s.to_csv(handles.handle, encoding=encoding, header=True)
+
+        result = pd.read_csv(
+            filename,
+            compression=compression,
+            encoding=encoding,
+            index_col=0,
+        ).squeeze("columns")
+        tm.assert_series_equal(s, result)
+
+        # explicitly ensure file was compressed
+        with tm.decompress_file(filename, compression) as fh:
+            text = fh.read().decode(encoding or "utf8")
+            assert s.name in text
+
+        with tm.decompress_file(filename, compression) as fh:
+            tm.assert_series_equal(
+                s,
+                pd.read_csv(fh, index_col=0, encoding=encoding).squeeze("columns"),
+            )
+
+    def test_to_csv_interval_index(self, using_infer_string, temp_file):
+        # GH 28210
+        s = Series(["foo", "bar", "baz"], index=pd.interval_range(0, 3))
+
+        s.to_csv(temp_file, header=False)
+        result = self.read_csv(temp_file, index_col=0)
+
+        # can't roundtrip intervalindex via read_csv so check string repr (GH 23595)
+        expected = s
+        expected.index = expected.index.astype("str")
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_to_dict.py b/pandas/pandas/tests/series/methods/test_to_dict.py
new file mode 100644
index 0000000000000..41c01f4537f23
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_to_dict.py
@@ -0,0 +1,38 @@
+import collections
+
+import numpy as np
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestSeriesToDict:
+    @pytest.mark.parametrize(
+        "mapping", (dict, collections.defaultdict(list), collections.OrderedDict)
+    )
+    def test_to_dict(self, mapping, datetime_series):
+        # GH#16122
+        result = Series(datetime_series.to_dict(into=mapping), name="ts")
+        expected = datetime_series.copy()
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+        from_method = Series(datetime_series.to_dict(into=collections.Counter))
+        from_constructor = Series(collections.Counter(datetime_series.items()))
+        tm.assert_series_equal(from_method, from_constructor)
+
+    @pytest.mark.parametrize(
+        "input",
+        (
+            {"a": np.int64(64), "b": 10},
+            {"a": np.int64(64), "b": 10, "c": "ABC"},
+            {"a": np.uint64(64), "b": 10, "c": "ABC"},
+        ),
+    )
+    def test_to_dict_return_types(self, input):
+        # GH25969
+
+        d = Series(input).to_dict()
+        assert isinstance(d["a"], int)
+        assert isinstance(d["b"], int)
diff --git a/pandas/pandas/tests/series/methods/test_to_frame.py b/pandas/pandas/tests/series/methods/test_to_frame.py
new file mode 100644
index 0000000000000..0eadf696b34cc
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_to_frame.py
@@ -0,0 +1,63 @@
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestToFrame:
+    def test_to_frame_respects_name_none(self):
+        # GH#44212 if we explicitly pass name=None, then that should be respected,
+        #  not changed to 0
+        # GH-45448 this is first deprecated & enforced in 2.0
+        ser = Series(range(3))
+        result = ser.to_frame(None)
+
+        exp_index = Index([None], dtype=object)
+        tm.assert_index_equal(result.columns, exp_index)
+
+        result = ser.rename("foo").to_frame(None)
+        exp_index = Index([None], dtype=object)
+        tm.assert_index_equal(result.columns, exp_index)
+
+    def test_to_frame(self, datetime_series):
+        datetime_series.name = None
+        rs = datetime_series.to_frame()
+        xp = DataFrame(datetime_series.values, index=datetime_series.index)
+        tm.assert_frame_equal(rs, xp)
+
+        datetime_series.name = "testname"
+        rs = datetime_series.to_frame()
+        xp = DataFrame(
+            {"testname": datetime_series.values}, index=datetime_series.index
+        )
+        tm.assert_frame_equal(rs, xp)
+
+        rs = datetime_series.to_frame(name="testdifferent")
+        xp = DataFrame(
+            {"testdifferent": datetime_series.values}, index=datetime_series.index
+        )
+        tm.assert_frame_equal(rs, xp)
+
+    @pytest.mark.filterwarnings(
+        "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+    )
+    def test_to_frame_expanddim(self):
+        # GH#9762
+
+        class SubclassedSeries(Series):
+            @property
+            def _constructor_expanddim(self):
+                return SubclassedFrame
+
+        class SubclassedFrame(DataFrame):
+            pass
+
+        ser = SubclassedSeries([1, 2, 3], name="X")
+        result = ser.to_frame()
+        assert isinstance(result, SubclassedFrame)
+        expected = SubclassedFrame({"X": [1, 2, 3]})
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_to_numpy.py b/pandas/pandas/tests/series/methods/test_to_numpy.py
new file mode 100644
index 0000000000000..4bc7631090761
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_to_numpy.py
@@ -0,0 +1,49 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    NA,
+    Series,
+    Timedelta,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("dtype", ["int64", "float64"])
+def test_to_numpy_na_value(dtype):
+    # GH#48951
+    ser = Series([1, 2, NA, 4])
+    result = ser.to_numpy(dtype=dtype, na_value=0)
+    expected = np.array([1, 2, 0, 4], dtype=dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_to_numpy_cast_before_setting_na():
+    # GH#50600
+    ser = Series([1])
+    result = ser.to_numpy(dtype=np.float64, na_value=np.nan)
+    expected = np.array([1.0])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_to_numpy_arrow_dtype_given():
+    # GH#57121
+    ser = Series([1, NA], dtype="int64[pyarrow]")
+    result = ser.to_numpy(dtype="float64")
+    expected = np.array([1.0, np.nan])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_astype_ea_int_to_td_ts():
+    # GH#57093
+    ser = Series([1, None], dtype="Int64")
+    result = ser.astype("m8[ns]")
+    expected = Series([1, Timedelta("nat")], dtype="m8[ns]")
+    tm.assert_series_equal(result, expected)
+
+    result = ser.astype("M8[ns]")
+    expected = Series([1, Timedelta("nat")], dtype="M8[ns]")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_tolist.py b/pandas/pandas/tests/series/methods/test_tolist.py
new file mode 100644
index 0000000000000..4af473528e238
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_tolist.py
@@ -0,0 +1,36 @@
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    Interval,
+    Period,
+    Series,
+    Timedelta,
+    Timestamp,
+)
+
+
+@pytest.mark.parametrize(
+    "values, dtype, expected_dtype",
+    (
+        ([1], "int64", int),
+        ([1], "Int64", int),
+        ([1.0], "float64", float),
+        ([1.0], "Float64", float),
+        (["abc"], "object", str),
+        (["abc"], "string", str),
+        ([Interval(1, 3)], "interval", Interval),
+        ([Period("2000-01-01", "D")], "period[D]", Period),
+        ([Timedelta(days=1)], "timedelta64[ns]", Timedelta),
+        ([Timestamp("2000-01-01")], "datetime64[ns]", Timestamp),
+        pytest.param([1], "int64[pyarrow]", int, marks=td.skip_if_no("pyarrow")),
+        pytest.param([1.0], "float64[pyarrow]", float, marks=td.skip_if_no("pyarrow")),
+        pytest.param(["abc"], "string[pyarrow]", str, marks=td.skip_if_no("pyarrow")),
+    ),
+)
+def test_tolist_scalar_dtype(values, dtype, expected_dtype):
+    # GH49890
+    ser = Series(values, dtype=dtype)
+    result_dtype = type(ser.tolist()[0])
+    assert result_dtype == expected_dtype
diff --git a/pandas/pandas/tests/series/methods/test_truncate.py b/pandas/pandas/tests/series/methods/test_truncate.py
new file mode 100644
index 0000000000000..33eb5c10ae163
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_truncate.py
@@ -0,0 +1,67 @@
+from datetime import datetime
+
+import pytest
+
+import pandas as pd
+from pandas import (
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestTruncate:
+    def test_truncate_datetimeindex_tz(self):
+        # GH 9243
+        idx = date_range("4/1/2005", "4/30/2005", freq="D", tz="US/Pacific")
+        s = Series(range(len(idx)), index=idx)
+        with pytest.raises(TypeError, match="Cannot compare tz-naive"):
+            # GH#36148 as of 2.0 we require tzawareness compat
+            s.truncate(datetime(2005, 4, 2), datetime(2005, 4, 4))
+
+        lb = idx[1]
+        ub = idx[3]
+        result = s.truncate(lb.to_pydatetime(), ub.to_pydatetime())
+        expected = Series([1, 2, 3], index=idx[1:4])
+        tm.assert_series_equal(result, expected)
+
+    def test_truncate_periodindex(self):
+        # GH 17717
+        idx1 = pd.PeriodIndex(
+            [pd.Period("2017-09-02"), pd.Period("2017-09-02"), pd.Period("2017-09-03")]
+        )
+        series1 = Series([1, 2, 3], index=idx1)
+        result1 = series1.truncate(after="2017-09-02")
+
+        expected_idx1 = pd.PeriodIndex(
+            [pd.Period("2017-09-02"), pd.Period("2017-09-02")]
+        )
+        tm.assert_series_equal(result1, Series([1, 2], index=expected_idx1))
+
+        idx2 = pd.PeriodIndex(
+            [pd.Period("2017-09-03"), pd.Period("2017-09-02"), pd.Period("2017-09-03")]
+        )
+        series2 = Series([1, 2, 3], index=idx2)
+        result2 = series2.sort_index().truncate(after="2017-09-02")
+
+        expected_idx2 = pd.PeriodIndex([pd.Period("2017-09-02")])
+        tm.assert_series_equal(result2, Series([2], index=expected_idx2))
+
+    def test_truncate_one_element_series(self):
+        # GH 35544
+        series = Series([0.1], index=pd.DatetimeIndex(["2020-08-04"]))
+        before = pd.Timestamp("2020-08-02")
+        after = pd.Timestamp("2020-08-04")
+
+        result = series.truncate(before=before, after=after)
+
+        # the input Series and the expected Series are the same
+        tm.assert_series_equal(result, series)
+
+    def test_truncate_index_only_one_unique_value(self):
+        # GH 42365
+        obj = Series(0, index=date_range("2021-06-30", "2021-06-30")).repeat(5)
+
+        truncated = obj.truncate("2021-06-28", "2021-07-01")
+
+        tm.assert_series_equal(truncated, obj)
diff --git a/pandas/pandas/tests/series/methods/test_tz_localize.py b/pandas/pandas/tests/series/methods/test_tz_localize.py
new file mode 100644
index 0000000000000..53288e8a1f8e7
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_tz_localize.py
@@ -0,0 +1,122 @@
+from datetime import timezone
+
+import pytest
+
+from pandas._libs.tslibs import timezones
+
+from pandas import (
+    DatetimeIndex,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestTZLocalize:
+    def test_series_tz_localize_ambiguous_bool(self):
+        # make sure that we are correctly accepting bool values as ambiguous
+
+        # GH#14402
+        ts = Timestamp("2015-11-01 01:00:03")
+        expected0 = Timestamp("2015-11-01 01:00:03-0500", tz="US/Central")
+        expected1 = Timestamp("2015-11-01 01:00:03-0600", tz="US/Central")
+
+        ser = Series([ts])
+        expected0 = Series([expected0])
+        expected1 = Series([expected1])
+
+        with tm.external_error_raised(ValueError):
+            ser.dt.tz_localize("US/Central")
+
+        result = ser.dt.tz_localize("US/Central", ambiguous=True)
+        tm.assert_series_equal(result, expected0)
+
+        result = ser.dt.tz_localize("US/Central", ambiguous=[True])
+        tm.assert_series_equal(result, expected0)
+
+        result = ser.dt.tz_localize("US/Central", ambiguous=False)
+        tm.assert_series_equal(result, expected1)
+
+        result = ser.dt.tz_localize("US/Central", ambiguous=[False])
+        tm.assert_series_equal(result, expected1)
+
+    def test_series_tz_localize_matching_index(self):
+        # Matching the index of the result with that of the original series
+        # GH 43080
+        dt_series = Series(
+            date_range(start="2021-01-01T02:00:00", periods=5, freq="1D"),
+            index=[2, 6, 7, 8, 11],
+            dtype="category",
+        )
+        result = dt_series.dt.tz_localize("Europe/Berlin")
+        expected = Series(
+            date_range(
+                start="2021-01-01T02:00:00", periods=5, freq="1D", tz="Europe/Berlin"
+            ),
+            index=[2, 6, 7, 8, 11],
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "method, exp",
+        [
+            ["shift_forward", "2015-03-29 03:00:00"],
+            ["shift_backward", "2015-03-29 01:59:59.999999999"],
+            ["NaT", NaT],
+            ["raise", None],
+            ["foo", "invalid"],
+        ],
+    )
+    def test_tz_localize_nonexistent(self, warsaw, method, exp, unit):
+        # GH 8917
+        tz = warsaw
+        n = 60
+        dti = date_range(start="2015-03-29 02:00:00", periods=n, freq="min", unit=unit)
+        ser = Series(1, index=dti)
+        df = ser.to_frame()
+
+        if method == "raise":
+            with tm.external_error_raised(ValueError):
+                dti.tz_localize(tz, nonexistent=method)
+            with tm.external_error_raised(ValueError):
+                ser.tz_localize(tz, nonexistent=method)
+            with tm.external_error_raised(ValueError):
+                df.tz_localize(tz, nonexistent=method)
+
+        elif exp == "invalid":
+            msg = (
+                "The nonexistent argument must be one of "
+                "'raise', 'NaT', 'shift_forward', 'shift_backward' "
+                "or a timedelta object"
+            )
+            with pytest.raises(ValueError, match=msg):
+                dti.tz_localize(tz, nonexistent=method)
+            with pytest.raises(ValueError, match=msg):
+                ser.tz_localize(tz, nonexistent=method)
+            with pytest.raises(ValueError, match=msg):
+                df.tz_localize(tz, nonexistent=method)
+
+        else:
+            result = ser.tz_localize(tz, nonexistent=method)
+            expected = Series(1, index=DatetimeIndex([exp] * n, tz=tz).as_unit(unit))
+            tm.assert_series_equal(result, expected)
+
+            result = df.tz_localize(tz, nonexistent=method)
+            expected = expected.to_frame()
+            tm.assert_frame_equal(result, expected)
+
+            res_index = dti.tz_localize(tz, nonexistent=method)
+            tm.assert_index_equal(res_index, expected.index)
+
+    @pytest.mark.parametrize("tzstr", ["US/Eastern", "dateutil/US/Eastern"])
+    def test_series_tz_localize_empty(self, tzstr):
+        # GH#2248
+        ser = Series(dtype=object)
+
+        ser2 = ser.tz_localize("utc")
+        assert ser2.index.tz == timezone.utc
+
+        ser2 = ser.tz_localize(tzstr)
+        timezones.tz_compare(ser2.index.tz, timezones.maybe_get_tz(tzstr))
diff --git a/pandas/pandas/tests/series/methods/test_unique.py b/pandas/pandas/tests/series/methods/test_unique.py
new file mode 100644
index 0000000000000..14e2473325032
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_unique.py
@@ -0,0 +1,76 @@
+import numpy as np
+
+from pandas import (
+    Categorical,
+    IntervalIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+class TestUnique:
+    def test_unique_uint64(self):
+        ser = Series([1, 2, 2**63, 2**63], dtype=np.uint64)
+        res = ser.unique()
+        exp = np.array([1, 2, 2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_unique_data_ownership(self):
+        # it works! GH#1807
+        Series(Series(["a", "c", "b"]).unique()).sort_values()
+
+    def test_unique(self):
+        # GH#714 also, dtype=float
+        ser = Series([1.2345] * 100)
+        ser[::2] = np.nan
+        result = ser.unique()
+        assert len(result) == 2
+
+        # explicit f4 dtype
+        ser = Series([1.2345] * 100, dtype="f4")
+        ser[::2] = np.nan
+        result = ser.unique()
+        assert len(result) == 2
+
+    def test_unique_nan_object_dtype(self):
+        # NAs in object arrays GH#714
+        ser = Series(["foo"] * 100, dtype="O")
+        ser[::2] = np.nan
+        result = ser.unique()
+        assert len(result) == 2
+
+    def test_unique_none(self):
+        # decision about None
+        ser = Series([1, 2, 3, None, None, None], dtype=object)
+        result = ser.unique()
+        expected = np.array([1, 2, 3, None], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_unique_categorical(self):
+        # GH#18051
+        cat = Categorical([])
+        ser = Series(cat)
+        result = ser.unique()
+        tm.assert_categorical_equal(result, cat)
+
+        cat = Categorical([np.nan])
+        ser = Series(cat)
+        result = ser.unique()
+        tm.assert_categorical_equal(result, cat)
+
+    def test_tz_unique(self):
+        # GH 46128
+        dti1 = date_range("2016-01-01", periods=3)
+        ii1 = IntervalIndex.from_breaks(dti1)
+        ser1 = Series(ii1)
+        uni1 = ser1.unique()
+        tm.assert_interval_array_equal(ser1.array, uni1)
+
+        dti2 = date_range("2016-01-01", periods=3, tz="US/Eastern")
+        ii2 = IntervalIndex.from_breaks(dti2)
+        ser2 = Series(ii2)
+        uni2 = ser2.unique()
+        tm.assert_interval_array_equal(ser2.array, uni2)
+
+        assert uni1.dtype != uni2.dtype
diff --git a/pandas/pandas/tests/series/methods/test_unstack.py b/pandas/pandas/tests/series/methods/test_unstack.py
new file mode 100644
index 0000000000000..f61e20c43657d
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_unstack.py
@@ -0,0 +1,169 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_unstack_preserves_object():
+    mi = MultiIndex.from_product([["bar", "foo"], ["one", "two"]])
+
+    ser = Series(np.arange(4.0), index=mi, dtype=object)
+
+    res1 = ser.unstack()
+    assert (res1.dtypes == object).all()
+
+    res2 = ser.unstack(level=0)
+    assert (res2.dtypes == object).all()
+
+
+def test_unstack():
+    index = MultiIndex(
+        levels=[["bar", "foo"], ["one", "three", "two"]],
+        codes=[[1, 1, 0, 0], [0, 1, 0, 2]],
+    )
+
+    s = Series(np.arange(4.0), index=index)
+    unstacked = s.unstack()
+
+    expected = DataFrame(
+        [[2.0, np.nan, 3.0], [0.0, 1.0, np.nan]],
+        index=["bar", "foo"],
+        columns=["one", "three", "two"],
+    )
+
+    tm.assert_frame_equal(unstacked, expected)
+
+    unstacked = s.unstack(level=0)
+    tm.assert_frame_equal(unstacked, expected.T)
+
+    index = MultiIndex(
+        levels=[["bar"], ["one", "two", "three"], [0, 1]],
+        codes=[[0, 0, 0, 0, 0, 0], [0, 1, 2, 0, 1, 2], [0, 1, 0, 1, 0, 1]],
+    )
+    s = Series(np.random.default_rng(2).standard_normal(6), index=index)
+    exp_index = MultiIndex(
+        levels=[["one", "two", "three"], [0, 1]],
+        codes=[[0, 1, 2, 0, 1, 2], [0, 1, 0, 1, 0, 1]],
+    )
+    expected = DataFrame({"bar": s.values}, index=exp_index).sort_index(level=0)
+    unstacked = s.unstack(0).sort_index()
+    tm.assert_frame_equal(unstacked, expected)
+
+    # GH5873
+    idx = MultiIndex.from_arrays([[101, 102], [3.5, np.nan]])
+    ts = Series([1, 2], index=idx)
+    left = ts.unstack()
+    right = DataFrame(
+        [[np.nan, 1], [2, np.nan]], index=[101, 102], columns=[np.nan, 3.5]
+    )
+    tm.assert_frame_equal(left, right)
+
+    idx = MultiIndex.from_arrays(
+        [
+            ["cat", "cat", "cat", "dog", "dog"],
+            ["a", "a", "b", "a", "b"],
+            [1, 2, 1, 1, np.nan],
+        ]
+    )
+    ts = Series([1.0, 1.1, 1.2, 1.3, 1.4], index=idx)
+    right = DataFrame(
+        [[1.0, 1.3], [1.1, np.nan], [np.nan, 1.4], [1.2, np.nan]],
+        columns=["cat", "dog"],
+    )
+    tpls = [("a", 1), ("a", 2), ("b", np.nan), ("b", 1)]
+    right.index = MultiIndex.from_tuples(tpls)
+    tm.assert_frame_equal(ts.unstack(level=0), right)
+
+
+def test_unstack_tuplename_in_multiindex():
+    # GH 19966
+    idx = MultiIndex.from_product(
+        [["a", "b", "c"], [1, 2, 3]], names=[("A", "a"), ("B", "b")]
+    )
+    ser = Series(1, index=idx)
+    result = ser.unstack(("A", "a"))
+
+    expected = DataFrame(
+        [[1, 1, 1], [1, 1, 1], [1, 1, 1]],
+        columns=MultiIndex.from_tuples([("a",), ("b",), ("c",)], names=[("A", "a")]),
+        index=Index([1, 2, 3], name=("B", "b")),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "unstack_idx, expected_values, expected_index, expected_columns",
+    [
+        (
+            ("A", "a"),
+            [[1, 1], [1, 1], [1, 1], [1, 1]],
+            MultiIndex.from_tuples([(1, 3), (1, 4), (2, 3), (2, 4)], names=["B", "C"]),
+            MultiIndex.from_tuples([("a",), ("b",)], names=[("A", "a")]),
+        ),
+        (
+            (("A", "a"), "B"),
+            [[1, 1, 1, 1], [1, 1, 1, 1]],
+            Index([3, 4], name="C"),
+            MultiIndex.from_tuples(
+                [("a", 1), ("a", 2), ("b", 1), ("b", 2)], names=[("A", "a"), "B"]
+            ),
+        ),
+    ],
+)
+def test_unstack_mixed_type_name_in_multiindex(
+    unstack_idx, expected_values, expected_index, expected_columns
+):
+    # GH 19966
+    idx = MultiIndex.from_product(
+        [["a", "b"], [1, 2], [3, 4]], names=[("A", "a"), "B", "C"]
+    )
+    ser = Series(1, index=idx)
+    result = ser.unstack(unstack_idx)
+
+    expected = DataFrame(
+        expected_values, columns=expected_columns, index=expected_index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_multi_index_categorical_values():
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((10, 4)),
+        columns=Index(list("ABCD")),
+        index=date_range("2000-01-01", periods=10, freq="B"),
+    )
+    mi = df.stack().index.rename(["major", "minor"])
+    ser = Series(["foo"] * len(mi), index=mi, name="category", dtype="category")
+
+    result = ser.unstack()
+
+    dti = ser.index.levels[0]
+    c = pd.Categorical(["foo"] * len(dti))
+    expected = DataFrame(
+        {"A": c, "B": c, "C": c, "D": c},
+        columns=Index(list("ABCD"), name="minor"),
+        index=dti.rename("major"),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_unstack_mixed_level_names():
+    # GH#48763
+    arrays = [["a", "a"], [1, 2], ["red", "blue"]]
+    idx = MultiIndex.from_arrays(arrays, names=("x", 0, "y"))
+    ser = Series([1, 2], index=idx)
+    result = ser.unstack("x")
+    expected = DataFrame(
+        [[1], [2]],
+        columns=Index(["a"], name="x"),
+        index=MultiIndex.from_tuples([(1, "red"), (2, "blue")], names=[0, "y"]),
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_update.py b/pandas/pandas/tests/series/methods/test_update.py
new file mode 100644
index 0000000000000..9b5fb098bf3ee
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_update.py
@@ -0,0 +1,135 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    CategoricalDtype,
+    DataFrame,
+    NaT,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+class TestUpdate:
+    def test_update(self):
+        s = Series([1.5, np.nan, 3.0, 4.0, np.nan])
+        s2 = Series([np.nan, 3.5, np.nan, 5.0])
+        s.update(s2)
+
+        expected = Series([1.5, 3.5, 3.0, 5.0, np.nan])
+        tm.assert_series_equal(s, expected)
+
+        # GH 3217
+        df = DataFrame([{"a": 1}, {"a": 3, "b": 2}])
+        df["c"] = np.nan
+        # Cast to object to avoid upcast when setting "foo"
+        df["c"] = df["c"].astype(object)
+        df_orig = df.copy()
+
+        with tm.raises_chained_assignment_error():
+            df["c"].update(Series(["foo"], index=[0]))
+        expected = df_orig
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "other, dtype, expected, raises",
+        [
+            # other is int
+            ([61, 63], "int32", Series([10, 61, 12], dtype="int32"), False),
+            ([61, 63], "int64", Series([10, 61, 12]), False),
+            ([61, 63], float, Series([10.0, 61.0, 12.0]), False),
+            ([61, 63], object, Series([10, 61, 12], dtype=object), False),
+            # other is float, but can be cast to int
+            ([61.0, 63.0], "int32", Series([10, 61, 12], dtype="int32"), False),
+            ([61.0, 63.0], "int64", Series([10, 61, 12]), False),
+            ([61.0, 63.0], float, Series([10.0, 61.0, 12.0]), False),
+            ([61.0, 63.0], object, Series([10, 61.0, 12], dtype=object), False),
+            # others is float, cannot be cast to int
+            ([61.1, 63.1], "int32", Series([10.0, 61.1, 12.0]), True),
+            ([61.1, 63.1], "int64", Series([10.0, 61.1, 12.0]), True),
+            ([61.1, 63.1], float, Series([10.0, 61.1, 12.0]), False),
+            ([61.1, 63.1], object, Series([10, 61.1, 12], dtype=object), False),
+            # other is object, cannot be cast
+            ([(61,), (63,)], "int32", Series([10, (61,), 12]), True),
+            ([(61,), (63,)], "int64", Series([10, (61,), 12]), True),
+            ([(61,), (63,)], float, Series([10.0, (61,), 12.0]), True),
+            ([(61,), (63,)], object, Series([10, (61,), 12]), False),
+        ],
+    )
+    def test_update_dtypes(self, other, dtype, expected, raises):
+        ser = Series([10, 11, 12], dtype=dtype)
+        other = Series(other, index=[1, 3])
+        if raises:
+            with pytest.raises(TypeError, match="Invalid value"):
+                ser.update(other)
+        else:
+            ser.update(other)
+            tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize(
+        "values, other, expected",
+        [
+            # update by key
+            (
+                {"a": 1, "b": 2, "c": 3, "d": 4},
+                {"b": 5, "c": np.nan},
+                {"a": 1, "b": 5, "c": 3, "d": 4},
+            ),
+            # update by position
+            ([1, 2, 3, 4], [np.nan, 5, 1], [1, 5, 1, 4]),
+        ],
+    )
+    def test_update_from_non_series(self, values, other, expected):
+        # GH 33215
+        series = Series(values)
+        series.update(other)
+        expected = Series(expected)
+        tm.assert_series_equal(series, expected)
+
+    @pytest.mark.parametrize(
+        "data, other, expected, dtype",
+        [
+            (["a", None], [None, "b"], ["a", "b"], "string[python]"),
+            pytest.param(
+                ["a", None],
+                [None, "b"],
+                ["a", "b"],
+                "string[pyarrow]",
+                marks=td.skip_if_no("pyarrow"),
+            ),
+            ([1, None], [None, 2], [1, 2], "Int64"),
+            ([True, None], [None, False], [True, False], "boolean"),
+            (
+                ["a", None],
+                [None, "b"],
+                ["a", "b"],
+                CategoricalDtype(categories=["a", "b"]),
+            ),
+            (
+                [Timestamp(year=2020, month=1, day=1, tz="Europe/London"), NaT],
+                [NaT, Timestamp(year=2020, month=1, day=1, tz="Europe/London")],
+                [Timestamp(year=2020, month=1, day=1, tz="Europe/London")] * 2,
+                "datetime64[ns, Europe/London]",
+            ),
+        ],
+    )
+    def test_update_extension_array_series(self, data, other, expected, dtype):
+        result = Series(data, dtype=dtype)
+        other = Series(other, dtype=dtype)
+        expected = Series(expected, dtype=dtype)
+
+        result.update(other)
+        tm.assert_series_equal(result, expected)
+
+    def test_update_with_categorical_type(self):
+        # GH 25744
+        dtype = CategoricalDtype(["a", "b", "c", "d"])
+        s1 = Series(["a", "b", "c"], index=[1, 2, 3], dtype=dtype)
+        s2 = Series(["b", "a"], index=[1, 2], dtype=dtype)
+        s1.update(s2)
+        result = s1
+        expected = Series(["b", "a", "c"], index=[1, 2, 3], dtype=dtype)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_value_counts.py b/pandas/pandas/tests/series/methods/test_value_counts.py
new file mode 100644
index 0000000000000..7f882fa348b7e
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_value_counts.py
@@ -0,0 +1,255 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    Index,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestSeriesValueCounts:
+    def test_value_counts_datetime(self, unit):
+        # most dtypes are tested in tests/base
+        values = [
+            pd.Timestamp("2011-01-01 09:00"),
+            pd.Timestamp("2011-01-01 10:00"),
+            pd.Timestamp("2011-01-01 11:00"),
+            pd.Timestamp("2011-01-01 09:00"),
+            pd.Timestamp("2011-01-01 09:00"),
+            pd.Timestamp("2011-01-01 11:00"),
+        ]
+
+        exp_idx = pd.DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 11:00", "2011-01-01 10:00"],
+            name="xxx",
+        ).as_unit(unit)
+        exp = Series([3, 2, 1], index=exp_idx, name="count")
+
+        ser = Series(values, name="xxx").dt.as_unit(unit)
+        tm.assert_series_equal(ser.value_counts(), exp)
+        # check DatetimeIndex outputs the same result
+        idx = pd.DatetimeIndex(values, name="xxx").as_unit(unit)
+        tm.assert_series_equal(idx.value_counts(), exp)
+
+        # normalize
+        exp = Series(np.array([3.0, 2.0, 1]) / 6.0, index=exp_idx, name="proportion")
+        tm.assert_series_equal(ser.value_counts(normalize=True), exp)
+        tm.assert_series_equal(idx.value_counts(normalize=True), exp)
+
+    def test_value_counts_datetime_tz(self, unit):
+        values = [
+            pd.Timestamp("2011-01-01 09:00", tz="US/Eastern"),
+            pd.Timestamp("2011-01-01 10:00", tz="US/Eastern"),
+            pd.Timestamp("2011-01-01 11:00", tz="US/Eastern"),
+            pd.Timestamp("2011-01-01 09:00", tz="US/Eastern"),
+            pd.Timestamp("2011-01-01 09:00", tz="US/Eastern"),
+            pd.Timestamp("2011-01-01 11:00", tz="US/Eastern"),
+        ]
+
+        exp_idx = pd.DatetimeIndex(
+            ["2011-01-01 09:00", "2011-01-01 11:00", "2011-01-01 10:00"],
+            tz="US/Eastern",
+            name="xxx",
+        ).as_unit(unit)
+        exp = Series([3, 2, 1], index=exp_idx, name="count")
+
+        ser = Series(values, name="xxx").dt.as_unit(unit)
+        tm.assert_series_equal(ser.value_counts(), exp)
+        idx = pd.DatetimeIndex(values, name="xxx").as_unit(unit)
+        tm.assert_series_equal(idx.value_counts(), exp)
+
+        exp = Series(np.array([3.0, 2.0, 1]) / 6.0, index=exp_idx, name="proportion")
+        tm.assert_series_equal(ser.value_counts(normalize=True), exp)
+        tm.assert_series_equal(idx.value_counts(normalize=True), exp)
+
+    def test_value_counts_period(self):
+        values = [
+            pd.Period("2011-01", freq="M"),
+            pd.Period("2011-02", freq="M"),
+            pd.Period("2011-03", freq="M"),
+            pd.Period("2011-01", freq="M"),
+            pd.Period("2011-01", freq="M"),
+            pd.Period("2011-03", freq="M"),
+        ]
+
+        exp_idx = pd.PeriodIndex(
+            ["2011-01", "2011-03", "2011-02"], freq="M", name="xxx"
+        )
+        exp = Series([3, 2, 1], index=exp_idx, name="count")
+
+        ser = Series(values, name="xxx")
+        tm.assert_series_equal(ser.value_counts(), exp)
+        # check DatetimeIndex outputs the same result
+        idx = pd.PeriodIndex(values, name="xxx")
+        tm.assert_series_equal(idx.value_counts(), exp)
+
+        # normalize
+        exp = Series(np.array([3.0, 2.0, 1]) / 6.0, index=exp_idx, name="proportion")
+        tm.assert_series_equal(ser.value_counts(normalize=True), exp)
+        tm.assert_series_equal(idx.value_counts(normalize=True), exp)
+
+    def test_value_counts_categorical_ordered(self):
+        # most dtypes are tested in tests/base
+        values = Categorical([1, 2, 3, 1, 1, 3], ordered=True)
+
+        exp_idx = CategoricalIndex(
+            [1, 3, 2], categories=[1, 2, 3], ordered=True, name="xxx"
+        )
+        exp = Series([3, 2, 1], index=exp_idx, name="count")
+
+        ser = Series(values, name="xxx")
+        tm.assert_series_equal(ser.value_counts(), exp)
+        # check CategoricalIndex outputs the same result
+        idx = CategoricalIndex(values, name="xxx")
+        tm.assert_series_equal(idx.value_counts(), exp)
+
+        # normalize
+        exp = Series(np.array([3.0, 2.0, 1]) / 6.0, index=exp_idx, name="proportion")
+        tm.assert_series_equal(ser.value_counts(normalize=True), exp)
+        tm.assert_series_equal(idx.value_counts(normalize=True), exp)
+
+    def test_value_counts_categorical_not_ordered(self):
+        values = Categorical([1, 2, 3, 1, 1, 3], ordered=False)
+
+        exp_idx = CategoricalIndex(
+            [1, 3, 2], categories=[1, 2, 3], ordered=False, name="xxx"
+        )
+        exp = Series([3, 2, 1], index=exp_idx, name="count")
+
+        ser = Series(values, name="xxx")
+        tm.assert_series_equal(ser.value_counts(), exp)
+        # check CategoricalIndex outputs the same result
+        idx = CategoricalIndex(values, name="xxx")
+        tm.assert_series_equal(idx.value_counts(), exp)
+
+        # normalize
+        exp = Series(np.array([3.0, 2.0, 1]) / 6.0, index=exp_idx, name="proportion")
+        tm.assert_series_equal(ser.value_counts(normalize=True), exp)
+        tm.assert_series_equal(idx.value_counts(normalize=True), exp)
+
+    def test_value_counts_categorical(self):
+        # GH#12835
+        cats = Categorical(list("abcccb"), categories=list("cabd"))
+        ser = Series(cats, name="xxx")
+        res = ser.value_counts(sort=False)
+
+        exp_index = CategoricalIndex(
+            list("cabd"), categories=cats.categories, name="xxx"
+        )
+        exp = Series([3, 1, 2, 0], name="count", index=exp_index)
+        tm.assert_series_equal(res, exp)
+
+        res = ser.value_counts(sort=True)
+
+        exp_index = CategoricalIndex(
+            list("cbad"), categories=cats.categories, name="xxx"
+        )
+        exp = Series([3, 2, 1, 0], name="count", index=exp_index)
+        tm.assert_series_equal(res, exp)
+
+        # check object dtype handles the Series.name as the same
+        # (tested in tests/base)
+        ser = Series(["a", "b", "c", "c", "c", "b"], name="xxx")
+        res = ser.value_counts()
+        exp = Series([3, 2, 1], name="count", index=Index(["c", "b", "a"], name="xxx"))
+        tm.assert_series_equal(res, exp)
+
+    def test_value_counts_categorical_with_nan(self):
+        # see GH#9443
+
+        # sanity check
+        ser = Series(["a", "b", "a"], dtype="category")
+        exp = Series([2, 1], index=CategoricalIndex(["a", "b"]), name="count")
+
+        res = ser.value_counts(dropna=True)
+        tm.assert_series_equal(res, exp)
+
+        res = ser.value_counts(dropna=True)
+        tm.assert_series_equal(res, exp)
+
+        # same Series via two different constructions --> same behaviour
+        series = [
+            Series(["a", "b", None, "a", None, None], dtype="category"),
+            Series(
+                Categorical(["a", "b", None, "a", None, None], categories=["a", "b"])
+            ),
+        ]
+
+        for ser in series:
+            # None is a NaN value, so we exclude its count here
+            exp = Series([2, 1], index=CategoricalIndex(["a", "b"]), name="count")
+            res = ser.value_counts(dropna=True)
+            tm.assert_series_equal(res, exp)
+
+            # we don't exclude the count of None and sort by counts
+            exp = Series(
+                [3, 2, 1], index=CategoricalIndex([np.nan, "a", "b"]), name="count"
+            )
+            res = ser.value_counts(dropna=False)
+            tm.assert_series_equal(res, exp)
+
+            # When we aren't sorting by counts, and np.nan isn't a
+            # category, it should be last.
+            exp = Series(
+                [2, 1, 3], index=CategoricalIndex(["a", "b", np.nan]), name="count"
+            )
+            res = ser.value_counts(dropna=False, sort=False)
+            tm.assert_series_equal(res, exp)
+
+    @pytest.mark.parametrize(
+        "ser, dropna, exp",
+        [
+            (
+                Series([False, True, True, pd.NA]),
+                False,
+                Series([2, 1, 1], index=[True, False, pd.NA], name="count"),
+            ),
+            (
+                Series([False, True, True, pd.NA]),
+                True,
+                Series([2, 1], index=Index([True, False], dtype=object), name="count"),
+            ),
+            (
+                Series(range(3), index=[True, False, np.nan]).index,
+                False,
+                Series([1, 1, 1], index=[True, False, np.nan], name="count"),
+            ),
+        ],
+    )
+    def test_value_counts_bool_with_nan(self, ser, dropna, exp):
+        # GH32146
+        out = ser.value_counts(dropna=dropna)
+        tm.assert_series_equal(out, exp)
+
+    @pytest.mark.parametrize("dtype", [np.complex128, np.complex64])
+    def test_value_counts_complex_numbers(self, dtype):
+        # GH 17927
+        input_array = np.array([1 + 1j, 1 + 1j, 1, 3j, 3j, 3j], dtype=dtype)
+        result = Series(input_array).value_counts()
+        expected = Series(
+            [3, 2, 1], index=Index([3j, 1 + 1j, 1], dtype=dtype), name="count"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_value_counts_masked(self):
+        # GH#54984
+        dtype = "Int64"
+        ser = Series([1, 2, None, 2, None, 3], dtype=dtype)
+        result = ser.value_counts(dropna=False)
+        expected = Series(
+            [2, 2, 1, 1],
+            index=Index([2, None, 1, 3], dtype=dtype),
+            dtype=dtype,
+            name="count",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = ser.value_counts(dropna=True)
+        expected = Series(
+            [2, 1, 1], index=Index([2, 1, 3], dtype=dtype), dtype=dtype, name="count"
+        )
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/methods/test_values.py b/pandas/pandas/tests/series/methods/test_values.py
new file mode 100644
index 0000000000000..cb1595e68264f
--- /dev/null
+++ b/pandas/pandas/tests/series/methods/test_values.py
@@ -0,0 +1,29 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    IntervalIndex,
+    Series,
+    period_range,
+)
+import pandas._testing as tm
+
+
+class TestValues:
+    @pytest.mark.parametrize(
+        "data",
+        [
+            period_range("2000", periods=4),
+            IntervalIndex.from_breaks([1, 2, 3, 4]),
+        ],
+    )
+    def test_values_object_extension_dtypes(self, data):
+        # https://github.com/pandas-dev/pandas/issues/23995
+        result = Series(data).values
+        expected = np.array(data.astype(object))
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_values(self, datetime_series):
+        tm.assert_almost_equal(
+            datetime_series.values, list(datetime_series), check_dtype=False
+        )
diff --git a/pandas/pandas/tests/series/test_api.py b/pandas/pandas/tests/series/test_api.py
new file mode 100644
index 0000000000000..4b369bb0bc869
--- /dev/null
+++ b/pandas/pandas/tests/series/test_api.py
@@ -0,0 +1,278 @@
+import inspect
+import pydoc
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+class TestSeriesMisc:
+    def test_tab_completion(self):
+        # GH 9910
+        s = Series(list("abcd"))
+        # Series of str values should have .str but not .dt/.cat in __dir__
+        assert "str" in dir(s)
+        assert "dt" not in dir(s)
+        assert "cat" not in dir(s)
+
+    def test_tab_completion_dt(self):
+        # similarly for .dt
+        s = Series(date_range("1/1/2015", periods=5))
+        assert "dt" in dir(s)
+        assert "str" not in dir(s)
+        assert "cat" not in dir(s)
+
+    def test_tab_completion_cat(self):
+        # Similarly for .cat, but with the twist that str and dt should be
+        # there if the categories are of that type first cat and str.
+        s = Series(list("abbcd"), dtype="category")
+        assert "cat" in dir(s)
+        assert "str" in dir(s)  # as it is a string categorical
+        assert "dt" not in dir(s)
+
+    def test_tab_completion_cat_str(self):
+        # similar to cat and str
+        s = Series(date_range("1/1/2015", periods=5)).astype("category")
+        assert "cat" in dir(s)
+        assert "str" not in dir(s)
+        assert "dt" in dir(s)  # as it is a datetime categorical
+
+    def test_tab_completion_with_categorical(self):
+        # test the tab completion display
+        ok_for_cat = [
+            "categories",
+            "codes",
+            "ordered",
+            "set_categories",
+            "add_categories",
+            "remove_categories",
+            "rename_categories",
+            "reorder_categories",
+            "remove_unused_categories",
+            "as_ordered",
+            "as_unordered",
+        ]
+
+        s = Series(list("aabbcde")).astype("category")
+        results = sorted({r for r in s.cat.__dir__() if not r.startswith("_")})
+        tm.assert_almost_equal(results, sorted(set(ok_for_cat)))
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            Index(list("ab") * 5, dtype="category"),
+            Index([str(i) for i in range(10)]),
+            Index(["foo", "bar", "baz"] * 2),
+            date_range("2020-01-01", periods=10),
+            period_range("2020-01-01", periods=10, freq="D"),
+            timedelta_range("1 day", periods=10),
+            Index(np.arange(10), dtype=np.uint64),
+            Index(np.arange(10), dtype=np.int64),
+            Index(np.arange(10), dtype=np.float64),
+            Index([True, False]),
+            Index([f"a{i}" for i in range(101)]),
+            pd.MultiIndex.from_tuples(zip("ABCD", "EFGH")),
+            pd.MultiIndex.from_tuples(zip([0, 1, 2, 3], "EFGH")),
+        ],
+    )
+    def test_index_tab_completion(self, index):
+        # dir contains string-like values of the Index.
+        s = Series(index=index, dtype=object)
+        dir_s = dir(s)
+        for i, x in enumerate(s.index.unique(level=0)):
+            if i < 100:
+                assert not isinstance(x, str) or not x.isidentifier() or x in dir_s
+            else:
+                assert x not in dir_s
+
+    @pytest.mark.parametrize("ser", [Series(dtype=object), Series([1])])
+    def test_not_hashable(self, ser):
+        msg = "unhashable type: 'Series'"
+        with pytest.raises(TypeError, match=msg):
+            hash(ser)
+
+    def test_contains(self, datetime_series):
+        tm.assert_contains_all(datetime_series.index, datetime_series)
+
+    def test_axis_alias(self):
+        s = Series([1, 2, np.nan])
+        tm.assert_series_equal(s.dropna(axis="rows"), s.dropna(axis="index"))
+        assert s.dropna().sum(axis="rows") == 3
+        assert s._get_axis_number("rows") == 0
+        assert s._get_axis_name("rows") == "index"
+
+    def test_class_axis(self):
+        # https://github.com/pandas-dev/pandas/issues/18147
+        # no exception and no empty docstring
+        assert pydoc.getdoc(Series.index)
+
+    def test_ndarray_compat(self):
+        # test numpy compat with Series as sub-class of NDFrame
+        tsdf = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 3)),
+            columns=["A", "B", "C"],
+            index=date_range("1/1/2000", periods=1000),
+        )
+
+        def f(x):
+            return x[x.idxmax()]
+
+        result = tsdf.apply(f)
+        expected = tsdf.max()
+        tm.assert_series_equal(result, expected)
+
+    def test_ndarray_compat_like_func(self):
+        # using an ndarray like function
+        s = Series(np.random.default_rng(2).standard_normal(10))
+        result = Series(np.ones_like(s))
+        expected = Series(1, index=range(10), dtype="float64")
+        tm.assert_series_equal(result, expected)
+
+    def test_empty_method(self):
+        s_empty = Series(dtype=object)
+        assert s_empty.empty
+
+    @pytest.mark.parametrize("dtype", ["int64", object])
+    def test_empty_method_full_series(self, dtype):
+        full_series = Series(index=[1], dtype=dtype)
+        assert not full_series.empty
+
+    @pytest.mark.parametrize("dtype", [None, "Int64"])
+    def test_integer_series_size(self, dtype):
+        # GH 25580
+        s = Series(range(9), dtype=dtype)
+        assert s.size == 9
+
+    def test_attrs(self):
+        s = Series([0, 1], name="abc")
+        assert s.attrs == {}
+        s.attrs["version"] = 1
+        result = s + 1
+        assert result.attrs == {"version": 1}
+
+    def test_inspect_getmembers(self):
+        # GH38782
+        ser = Series(dtype=object)
+        inspect.getmembers(ser)
+
+    def test_unknown_attribute(self):
+        # GH#9680
+        tdi = timedelta_range(start=0, periods=10, freq="1s")
+        ser = Series(np.random.default_rng(2).normal(size=10), index=tdi)
+        assert "foo" not in ser.__dict__
+        msg = "'Series' object has no attribute 'foo'"
+        with pytest.raises(AttributeError, match=msg):
+            ser.foo
+
+    @pytest.mark.parametrize("op", ["year", "day", "second", "weekday"])
+    def test_datetime_series_no_datelike_attrs(self, op, datetime_series):
+        # GH#7206
+        msg = f"'Series' object has no attribute '{op}'"
+        with pytest.raises(AttributeError, match=msg):
+            getattr(datetime_series, op)
+
+    def test_series_datetimelike_attribute_access(self):
+        # attribute access should still work!
+        ser = Series({"year": 2000, "month": 1, "day": 10})
+        assert ser.year == 2000
+        assert ser.month == 1
+        assert ser.day == 10
+
+    def test_series_datetimelike_attribute_access_invalid(self):
+        ser = Series({"year": 2000, "month": 1, "day": 10})
+        msg = "'Series' object has no attribute 'weekday'"
+        with pytest.raises(AttributeError, match=msg):
+            ser.weekday
+
+    @pytest.mark.parametrize(
+        "kernel, has_numeric_only",
+        [
+            ("skew", True),
+            ("var", True),
+            ("all", False),
+            ("prod", True),
+            ("any", False),
+            ("idxmin", False),
+            ("quantile", False),
+            ("idxmax", False),
+            ("min", True),
+            ("sem", True),
+            ("mean", True),
+            ("nunique", False),
+            ("max", True),
+            ("sum", True),
+            ("count", False),
+            ("median", True),
+            ("std", True),
+            ("rank", True),
+            ("pct_change", False),
+            ("cummax", False),
+            ("shift", False),
+            ("diff", False),
+            ("cumsum", False),
+            ("cummin", False),
+            ("cumprod", False),
+            ("fillna", False),
+            ("ffill", False),
+            ("bfill", False),
+            ("sample", False),
+            ("tail", False),
+            ("take", False),
+            ("head", False),
+            ("cov", False),
+            ("corr", False),
+        ],
+    )
+    @pytest.mark.parametrize("dtype", [bool, int, float, object])
+    def test_numeric_only(self, kernel, has_numeric_only, dtype):
+        # GH#47500
+        ser = Series([0, 1, 1], dtype=dtype)
+        if kernel == "corrwith":
+            args = (ser,)
+        elif kernel == "corr":
+            args = (ser,)
+        elif kernel == "cov":
+            args = (ser,)
+        elif kernel == "nth":
+            args = (0,)
+        elif kernel == "fillna":
+            args = (True,)
+        elif kernel == "fillna":
+            args = ("ffill",)
+        elif kernel == "take":
+            args = ([0],)
+        elif kernel == "quantile":
+            args = (0.5,)
+        else:
+            args = ()
+        method = getattr(ser, kernel)
+        if not has_numeric_only:
+            msg = (
+                "(got an unexpected keyword argument 'numeric_only'"
+                "|too many arguments passed in)"
+            )
+            with pytest.raises(TypeError, match=msg):
+                method(*args, numeric_only=True)
+        elif dtype is object:
+            msg = f"Series.{kernel} does not allow numeric_only=True with non-numeric"
+            with pytest.raises(TypeError, match=msg):
+                method(*args, numeric_only=True)
+        else:
+            result = method(*args, numeric_only=True)
+            expected = method(*args, numeric_only=False)
+            if isinstance(expected, Series):
+                # transformer
+                tm.assert_series_equal(result, expected)
+            else:
+                # reducer
+                assert result == expected
diff --git a/pandas/pandas/tests/series/test_arithmetic.py b/pandas/pandas/tests/series/test_arithmetic.py
new file mode 100644
index 0000000000000..e7d284bd47e21
--- /dev/null
+++ b/pandas/pandas/tests/series/test_arithmetic.py
@@ -0,0 +1,960 @@
+from datetime import (
+    date,
+    timedelta,
+    timezone,
+)
+from decimal import Decimal
+import operator
+
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+from pandas._libs.tslibs import IncompatibleFrequency
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DatetimeTZDtype,
+    Index,
+    Series,
+    Timedelta,
+    bdate_range,
+    date_range,
+    isna,
+)
+import pandas._testing as tm
+from pandas.core import ops
+from pandas.core.computation import expressions as expr
+from pandas.core.computation.check import NUMEXPR_INSTALLED
+
+
+@pytest.fixture(autouse=True, params=[0, 1000000], ids=["numexpr", "python"])
+def switch_numexpr_min_elements(request, monkeypatch):
+    with monkeypatch.context() as m:
+        m.setattr(expr, "_MIN_ELEMENTS", request.param)
+        yield
+
+
+def _permute(obj):
+    return obj.take(np.random.default_rng(2).permutation(len(obj)))
+
+
+class TestSeriesFlexArithmetic:
+    @pytest.mark.parametrize(
+        "ts",
+        [
+            (lambda x: x, lambda x: x * 2, False),
+            (lambda x: x, lambda x: x[::2], False),
+            (lambda x: x, lambda x: 5, True),
+            (
+                lambda x: Series(range(10), dtype=np.float64),
+                lambda x: Series(range(10), dtype=np.float64),
+                True,
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "opname", ["add", "sub", "mul", "floordiv", "truediv", "pow"]
+    )
+    def test_flex_method_equivalence(self, opname, ts):
+        # check that Series.{opname} behaves like Series.__{opname}__,
+        tser = Series(
+            np.arange(20, dtype=np.float64),
+            index=date_range("2020-01-01", periods=20),
+            name="ts",
+        )
+
+        series = ts[0](tser)
+        other = ts[1](tser)
+        check_reverse = ts[2]
+
+        op = getattr(Series, opname)
+        alt = getattr(operator, opname)
+
+        result = op(series, other)
+        expected = alt(series, other)
+        tm.assert_almost_equal(result, expected)
+        if check_reverse:
+            rop = getattr(Series, "r" + opname)
+            result = rop(series, other)
+            expected = alt(other, series)
+            tm.assert_almost_equal(result, expected)
+
+    def test_flex_method_subclass_metadata_preservation(self, all_arithmetic_operators):
+        # GH 13208
+        class MySeries(Series):
+            _metadata = ["x"]
+
+            @property
+            def _constructor(self):
+                return MySeries
+
+        opname = all_arithmetic_operators
+        op = getattr(Series, opname)
+        m = MySeries([1, 2, 3], name="test")
+        m.x = 42
+        result = op(m, 1)
+        assert result.x == 42
+
+    def test_flex_add_scalar_fill_value(self):
+        # GH12723
+        ser = Series([0, 1, np.nan, 3, 4, 5])
+
+        exp = ser.fillna(0).add(2)
+        res = ser.add(2, fill_value=0)
+        tm.assert_series_equal(res, exp)
+
+    pairings = [(Series.div, operator.truediv, 1), (Series.rdiv, ops.rtruediv, 1)]
+    for op in ["add", "sub", "mul", "pow", "truediv", "floordiv"]:
+        fv = 0
+        lop = getattr(Series, op)
+        lequiv = getattr(operator, op)
+        rop = getattr(Series, "r" + op)
+        # bind op at definition time...
+        requiv = lambda x, y, op=op: getattr(operator, op)(y, x)
+        pairings.append((lop, lequiv, fv))
+        pairings.append((rop, requiv, fv))
+
+    @pytest.mark.parametrize("op, equiv_op, fv", pairings)
+    def test_operators_combine(self, op, equiv_op, fv):
+        def _check_fill(meth, op, a, b, fill_value=0):
+            exp_index = a.index.union(b.index)
+            a = a.reindex(exp_index)
+            b = b.reindex(exp_index)
+
+            amask = isna(a)
+            bmask = isna(b)
+
+            exp_values = []
+            for i in range(len(exp_index)):
+                with np.errstate(all="ignore"):
+                    if amask[i]:
+                        if bmask[i]:
+                            exp_values.append(np.nan)
+                            continue
+                        exp_values.append(op(fill_value, b[i]))
+                    elif bmask[i]:
+                        if amask[i]:
+                            exp_values.append(np.nan)
+                            continue
+                        exp_values.append(op(a[i], fill_value))
+                    else:
+                        exp_values.append(op(a[i], b[i]))
+
+            result = meth(a, b, fill_value=fill_value)
+            expected = Series(exp_values, exp_index)
+            tm.assert_series_equal(result, expected)
+
+        a = Series([np.nan, 1.0, 2.0, 3.0, np.nan], index=np.arange(5))
+        b = Series([np.nan, 1, np.nan, 3, np.nan, 4.0], index=np.arange(6))
+
+        result = op(a, b)
+        exp = equiv_op(a, b)
+        tm.assert_series_equal(result, exp)
+        _check_fill(op, equiv_op, a, b, fill_value=fv)
+        # should accept axis=0 or axis='rows'
+        op(a, b, axis=0)
+
+
+class TestSeriesArithmetic:
+    # Some of these may end up in tests/arithmetic, but are not yet sorted
+
+    def test_add_series_with_period_index(self):
+        rng = pd.period_range("1/1/2000", "1/1/2010", freq="Y")
+        ts = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        result = ts + ts[::2]
+        expected = ts + ts
+        expected.iloc[1::2] = np.nan
+        tm.assert_series_equal(result, expected)
+
+        result = ts + _permute(ts[::2])
+        tm.assert_series_equal(result, expected)
+
+        msg = "Input has different freq=D from Period\\(freq=Y-DEC\\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            ts + ts.asfreq("D", how="end")
+
+    @pytest.mark.parametrize(
+        "target_add,input_value,expected_value",
+        [
+            ("!", ["hello", "world"], ["hello!", "world!"]),
+            ("m", ["hello", "world"], ["hellom", "worldm"]),
+        ],
+    )
+    def test_string_addition(self, target_add, input_value, expected_value):
+        # GH28658 - ensure adding 'm' does not raise an error
+        a = Series(input_value)
+
+        result = a + target_add
+        expected = Series(expected_value)
+        tm.assert_series_equal(result, expected)
+
+    def test_divmod(self):
+        # GH#25557
+        a = Series([1, 1, 1, np.nan], index=["a", "b", "c", "d"])
+        b = Series([2, np.nan, 1, np.nan], index=["a", "b", "d", "e"])
+
+        result = a.divmod(b)
+        expected = divmod(a, b)
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+        result = a.rdivmod(b)
+        expected = divmod(b, a)
+        tm.assert_series_equal(result[0], expected[0])
+        tm.assert_series_equal(result[1], expected[1])
+
+    @pytest.mark.parametrize("index", [None, range(9)])
+    def test_series_integer_mod(self, index):
+        # GH#24396
+        s1 = Series(range(1, 10))
+        s2 = Series("foo", index=index)
+
+        msg = "not all arguments converted during string formatting|'mod' not supported"
+
+        with pytest.raises(TypeError, match=msg):
+            s2 % s1
+
+    def test_add_with_duplicate_index(self):
+        # GH14227
+        s1 = Series([1, 2], index=[1, 1])
+        s2 = Series([10, 10], index=[1, 2])
+        result = s1 + s2
+        expected = Series([11, 12, np.nan], index=[1, 1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_add_na_handling(self):
+        ser = Series(
+            [Decimal("1.3"), Decimal("2.3")], index=[date(2012, 1, 1), date(2012, 1, 2)]
+        )
+
+        result = ser + ser.shift(1)
+        result2 = ser.shift(1) + ser
+        assert isna(result.iloc[0])
+        assert isna(result2.iloc[0])
+
+    def test_add_corner_cases(self, datetime_series):
+        empty = Series([], index=Index([]), dtype=np.float64)
+
+        result = datetime_series + empty
+        assert np.isnan(result).all()
+
+        result = empty + empty
+        assert len(result) == 0
+
+    def test_add_float_plus_int(self, datetime_series):
+        # float + int
+        int_ts = datetime_series.astype(int)[:-5]
+        added = datetime_series + int_ts
+        expected = Series(
+            datetime_series.values[:-5] + int_ts.values,
+            index=datetime_series.index[:-5],
+            name="ts",
+        )
+        tm.assert_series_equal(added[:-5], expected)
+
+    def test_mul_empty_int_corner_case(self):
+        s1 = Series([], [], dtype=np.int32)
+        s2 = Series({"x": 0.0})
+        tm.assert_series_equal(s1 * s2, Series([np.nan], index=["x"]))
+
+    def test_sub_datetimelike_align(self):
+        # GH#7500
+        # datetimelike ops need to align
+        dt = Series(date_range("2012-1-1", periods=3, freq="D"))
+        dt.iloc[2] = np.nan
+        dt2 = dt[::-1]
+
+        expected = Series([timedelta(0), timedelta(0), pd.NaT])
+        # name is reset
+        result = dt2 - dt
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(expected, name=0)
+        result = (dt2.to_frame() - dt.to_frame())[0]
+        tm.assert_series_equal(result, expected)
+
+    def test_alignment_doesnt_change_tz(self):
+        # GH#33671
+        dti = date_range("2016-01-01", periods=10, tz="CET")
+        dti_utc = dti.tz_convert("UTC")
+        ser = Series(10, index=dti)
+        ser_utc = Series(10, index=dti_utc)
+
+        # we don't care about the result, just that original indexes are unchanged
+        ser * ser_utc
+
+        assert ser.index is dti
+        assert ser_utc.index is dti_utc
+
+    def test_alignment_categorical(self):
+        # GH13365
+        cat = Categorical(["3z53", "3z53", "LoJG", "LoJG", "LoJG", "N503"])
+        ser1 = Series(2, index=cat)
+        ser2 = Series(2, index=cat[:-1])
+        result = ser1 * ser2
+
+        exp_index = ["3z53"] * 4 + ["LoJG"] * 9 + ["N503"]
+        exp_index = pd.CategoricalIndex(exp_index, categories=cat.categories)
+        exp_values = [4.0] * 13 + [np.nan]
+        expected = Series(exp_values, exp_index)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_arithmetic_with_duplicate_index(self):
+        # GH#8363
+        # integer ops with a non-unique index
+        index = [2, 2, 3, 3, 4]
+        ser = Series(np.arange(1, 6, dtype="int64"), index=index)
+        other = Series(np.arange(5, dtype="int64"), index=index)
+        result = ser - other
+        expected = Series(1, index=[2, 2, 3, 3, 4])
+        tm.assert_series_equal(result, expected)
+
+        # GH#8363
+        # datetime ops with a non-unique index
+        ser = Series(date_range("20130101 09:00:00", periods=5), index=index)
+        other = Series(date_range("20130101", periods=5), index=index)
+        result = ser - other
+        expected = Series(Timedelta("9 hours"), index=[2, 2, 3, 3, 4])
+        tm.assert_series_equal(result, expected)
+
+    def test_masked_and_non_masked_propagate_na(self):
+        # GH#45810
+        ser1 = Series([0, np.nan], dtype="float")
+        ser2 = Series([0, 1], dtype="Int64")
+        result = ser1 * ser2
+        expected = Series([0, pd.NA], dtype="Float64")
+        tm.assert_series_equal(result, expected)
+
+    def test_mask_div_propagate_na_for_non_na_dtype(self):
+        # GH#42630
+        ser1 = Series([15, pd.NA, 5, 4], dtype="Int64")
+        ser2 = Series([15, 5, np.nan, 4])
+        result = ser1 / ser2
+        expected = Series([1.0, pd.NA, pd.NA, 1.0], dtype="Float64")
+        tm.assert_series_equal(result, expected)
+
+        result = ser2 / ser1
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("val, dtype", [(3, "Int64"), (3.5, "Float64")])
+    def test_add_list_to_masked_array(self, val, dtype):
+        # GH#22962
+        ser = Series([1, None, 3], dtype="Int64")
+        result = ser + [1, None, val]
+        expected = Series([2, None, 3 + val], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+        result = [1, None, val] + ser
+        tm.assert_series_equal(result, expected)
+
+    def test_add_list_to_masked_array_boolean(self, request):
+        # GH#22962
+        warning = (
+            UserWarning
+            if request.node.callspec.id == "numexpr" and NUMEXPR_INSTALLED
+            else None
+        )
+        ser = Series([True, None, False], dtype="boolean")
+        msg = "operator is not supported by numexpr for the bool dtype"
+        with tm.assert_produces_warning(warning, match=msg):
+            result = ser + [True, None, True]
+        expected = Series([True, None, True], dtype="boolean")
+        tm.assert_series_equal(result, expected)
+
+        with tm.assert_produces_warning(warning, match=msg):
+            result = [True, None, True] + ser
+        tm.assert_series_equal(result, expected)
+
+
+# ------------------------------------------------------------------
+# Comparisons
+
+
+class TestSeriesFlexComparison:
+    @pytest.mark.parametrize("axis", [0, None, "index"])
+    def test_comparison_flex_basic(self, axis, comparison_op):
+        left = Series(np.random.default_rng(2).standard_normal(10))
+        right = Series(np.random.default_rng(2).standard_normal(10))
+        result = getattr(left, comparison_op.__name__)(right, axis=axis)
+        expected = comparison_op(left, right)
+        tm.assert_series_equal(result, expected)
+
+    def test_comparison_bad_axis(self, comparison_op):
+        left = Series(np.random.default_rng(2).standard_normal(10))
+        right = Series(np.random.default_rng(2).standard_normal(10))
+
+        msg = "No axis named 1 for object type"
+        with pytest.raises(ValueError, match=msg):
+            getattr(left, comparison_op.__name__)(right, axis=1)
+
+    @pytest.mark.parametrize(
+        "values, op",
+        [
+            ([False, False, True, False], "eq"),
+            ([True, True, False, True], "ne"),
+            ([False, False, True, False], "le"),
+            ([False, False, False, False], "lt"),
+            ([False, True, True, False], "ge"),
+            ([False, True, False, False], "gt"),
+        ],
+    )
+    def test_comparison_flex_alignment(self, values, op):
+        left = Series([1, 3, 2], index=list("abc"))
+        right = Series([2, 2, 2], index=list("bcd"))
+        result = getattr(left, op)(right)
+        expected = Series(values, index=list("abcd"))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "values, op, fill_value",
+        [
+            ([False, False, True, True], "eq", 2),
+            ([True, True, False, False], "ne", 2),
+            ([False, False, True, True], "le", 0),
+            ([False, False, False, True], "lt", 0),
+            ([True, True, True, False], "ge", 0),
+            ([True, True, False, False], "gt", 0),
+        ],
+    )
+    def test_comparison_flex_alignment_fill(self, values, op, fill_value):
+        left = Series([1, 3, 2], index=list("abc"))
+        right = Series([2, 2, 2], index=list("bcd"))
+        result = getattr(left, op)(right, fill_value=fill_value)
+        expected = Series(values, index=list("abcd"))
+        tm.assert_series_equal(result, expected)
+
+
+class TestSeriesComparison:
+    def test_comparison_different_length(self):
+        a = Series(["a", "b", "c"])
+        b = Series(["b", "a"])
+        msg = "only compare identically-labeled Series"
+        with pytest.raises(ValueError, match=msg):
+            a < b
+
+        a = Series([1, 2])
+        b = Series([2, 3, 4])
+        with pytest.raises(ValueError, match=msg):
+            a == b
+
+    @pytest.mark.parametrize("opname", ["eq", "ne", "gt", "lt", "ge", "le"])
+    def test_ser_flex_cmp_return_dtypes(self, opname):
+        # GH#15115
+        ser = Series([1, 3, 2], index=range(3))
+        const = 2
+        result = getattr(ser, opname)(const).dtypes
+        expected = np.dtype("bool")
+        assert result == expected
+
+    @pytest.mark.parametrize("opname", ["eq", "ne", "gt", "lt", "ge", "le"])
+    def test_ser_flex_cmp_return_dtypes_empty(self, opname):
+        # GH#15115 empty Series case
+        ser = Series([1, 3, 2], index=range(3))
+        empty = ser.iloc[:0]
+        const = 2
+        result = getattr(empty, opname)(const).dtypes
+        expected = np.dtype("bool")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "names", [(None, None, None), ("foo", "bar", None), ("baz", "baz", "baz")]
+    )
+    def test_ser_cmp_result_names(self, names, comparison_op):
+        # datetime64 dtype
+        op = comparison_op
+        dti = date_range("1949-06-07 03:00:00", freq="h", periods=5, name=names[0])
+        ser = Series(dti).rename(names[1])
+        result = op(ser, dti)
+        assert result.name == names[2]
+
+        # datetime64tz dtype
+        dti = dti.tz_localize("US/Central")
+        dti = pd.DatetimeIndex(dti, freq="infer")  # freq not preserved by tz_localize
+        ser = Series(dti).rename(names[1])
+        result = op(ser, dti)
+        assert result.name == names[2]
+
+        # timedelta64 dtype
+        tdi = dti - dti.shift(1)
+        ser = Series(tdi).rename(names[1])
+        result = op(ser, tdi)
+        assert result.name == names[2]
+
+        # interval dtype
+        if op in [operator.eq, operator.ne]:
+            # interval dtype comparisons not yet implemented
+            ii = pd.interval_range(start=0, periods=5, name=names[0])
+            ser = Series(ii).rename(names[1])
+            result = op(ser, ii)
+            assert result.name == names[2]
+
+        # categorical
+        if op in [operator.eq, operator.ne]:
+            # categorical dtype comparisons raise for inequalities
+            cidx = tdi.astype("category")
+            ser = Series(cidx).rename(names[1])
+            result = op(ser, cidx)
+            assert result.name == names[2]
+
+    def test_comparisons(self):
+        s = Series(["a", "b", "c"])
+        s2 = Series([False, True, False])
+
+        # it works!
+        exp = Series([False, False, False])
+        tm.assert_series_equal(s == s2, exp)
+        tm.assert_series_equal(s2 == s, exp)
+
+    # -----------------------------------------------------------------
+    # Categorical Dtype Comparisons
+
+    def test_categorical_comparisons(self):
+        # GH#8938
+        # allow equality comparisons
+        a = Series(list("abc"), dtype="category")
+        b = Series(list("abc"), dtype="object")
+        c = Series(["a", "b", "cc"], dtype="object")
+        d = Series(list("acb"), dtype="object")
+        e = Categorical(list("abc"))
+        f = Categorical(list("acb"))
+
+        # vs scalar
+        assert not (a == "a").all()
+        assert ((a != "a") == ~(a == "a")).all()
+
+        assert not ("a" == a).all()
+        assert (a == "a")[0]
+        assert ("a" == a)[0]
+        assert not ("a" != a)[0]
+
+        # vs list-like
+        assert (a == a).all()
+        assert not (a != a).all()
+
+        assert (a == list(a)).all()
+        assert (a == b).all()
+        assert (b == a).all()
+        assert ((~(a == b)) == (a != b)).all()
+        assert ((~(b == a)) == (b != a)).all()
+
+        assert not (a == c).all()
+        assert not (c == a).all()
+        assert not (a == d).all()
+        assert not (d == a).all()
+
+        # vs a cat-like
+        assert (a == e).all()
+        assert (e == a).all()
+        assert not (a == f).all()
+        assert not (f == a).all()
+
+        assert (~(a == e) == (a != e)).all()
+        assert (~(e == a) == (e != a)).all()
+        assert (~(a == f) == (a != f)).all()
+        assert (~(f == a) == (f != a)).all()
+
+        # non-equality is not comparable
+        msg = "can only compare equality or not"
+        with pytest.raises(TypeError, match=msg):
+            a < b
+        with pytest.raises(TypeError, match=msg):
+            b < a
+        with pytest.raises(TypeError, match=msg):
+            a > b
+        with pytest.raises(TypeError, match=msg):
+            b > a
+
+    def test_unequal_categorical_comparison_raises_type_error(self):
+        # unequal comparison should raise for unordered cats
+        cat = Series(Categorical(list("abc")))
+        msg = "can only compare equality or not"
+        with pytest.raises(TypeError, match=msg):
+            cat > "b"
+
+        cat = Series(Categorical(list("abc"), ordered=False))
+        with pytest.raises(TypeError, match=msg):
+            cat > "b"
+
+        # https://github.com/pandas-dev/pandas/issues/9836#issuecomment-92123057
+        # and following comparisons with scalars not in categories should raise
+        # for unequal comps, but not for equal/not equal
+        cat = Series(Categorical(list("abc"), ordered=True))
+
+        msg = "Invalid comparison between dtype=category and str"
+        with pytest.raises(TypeError, match=msg):
+            cat < "d"
+        with pytest.raises(TypeError, match=msg):
+            cat > "d"
+        with pytest.raises(TypeError, match=msg):
+            "d" < cat
+        with pytest.raises(TypeError, match=msg):
+            "d" > cat
+
+        tm.assert_series_equal(cat == "d", Series([False, False, False]))
+        tm.assert_series_equal(cat != "d", Series([True, True, True]))
+
+    # -----------------------------------------------------------------
+
+    def test_comparison_tuples(self):
+        # GH#11339
+        # comparisons vs tuple
+        s = Series([(1, 1), (1, 2)])
+
+        result = s == (1, 2)
+        expected = Series([False, True])
+        tm.assert_series_equal(result, expected)
+
+        result = s != (1, 2)
+        expected = Series([True, False])
+        tm.assert_series_equal(result, expected)
+
+        result = s == (0, 0)
+        expected = Series([False, False])
+        tm.assert_series_equal(result, expected)
+
+        result = s != (0, 0)
+        expected = Series([True, True])
+        tm.assert_series_equal(result, expected)
+
+        s = Series([(1, 1), (1, 1)])
+
+        result = s == (1, 1)
+        expected = Series([True, True])
+        tm.assert_series_equal(result, expected)
+
+        result = s != (1, 1)
+        expected = Series([False, False])
+        tm.assert_series_equal(result, expected)
+
+    def test_comparison_frozenset(self):
+        ser = Series([frozenset([1]), frozenset([1, 2])])
+
+        result = ser == frozenset([1])
+        expected = Series([True, False])
+        tm.assert_series_equal(result, expected)
+
+    def test_comparison_operators_with_nas(self, comparison_op):
+        ser = Series(bdate_range("1/1/2000", periods=10), dtype=object)
+        ser[::2] = np.nan
+
+        # test that comparisons work
+        val = ser[5]
+
+        result = comparison_op(ser, val)
+        expected = comparison_op(ser.dropna(), val).reindex(ser.index)
+
+        if comparison_op is operator.ne:
+            expected = expected.fillna(True).astype(bool)
+        else:
+            expected = expected.fillna(False).astype(bool)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_ne(self):
+        ts = Series([3, 4, 5, 6, 7], [3, 4, 5, 6, 7], dtype=float)
+        expected = np.array([True, True, False, True, True])
+        tm.assert_numpy_array_equal(ts.index != 5, expected)
+        tm.assert_numpy_array_equal(~(ts.index == 5), expected)
+
+    @pytest.mark.parametrize("right_data", [[2, 2, 2], [2, 2, 2, 2]])
+    def test_comp_ops_df_compat(self, right_data, frame_or_series):
+        # GH 1134
+        # GH 50083 to clarify that index and columns must be identically labeled
+        left = Series([1, 2, 3], index=list("ABC"), name="x")
+        right = Series(right_data, index=list("ABDC")[: len(right_data)], name="x")
+        if frame_or_series is not Series:
+            msg = (
+                rf"Can only compare identically-labeled \(both index and columns\) "
+                f"{frame_or_series.__name__} objects"
+            )
+            left = left.to_frame()
+            right = right.to_frame()
+        else:
+            msg = (
+                f"Can only compare identically-labeled {frame_or_series.__name__} "
+                f"objects"
+            )
+
+        with pytest.raises(ValueError, match=msg):
+            left == right
+        with pytest.raises(ValueError, match=msg):
+            right == left
+
+        with pytest.raises(ValueError, match=msg):
+            left != right
+        with pytest.raises(ValueError, match=msg):
+            right != left
+
+        with pytest.raises(ValueError, match=msg):
+            left < right
+        with pytest.raises(ValueError, match=msg):
+            right < left
+
+    def test_compare_series_interval_keyword(self):
+        # GH#25338
+        ser = Series(["IntervalA", "IntervalB", "IntervalC"])
+        result = ser == "IntervalA"
+        expected = Series([True, False, False])
+        tm.assert_series_equal(result, expected)
+
+
+# ------------------------------------------------------------------
+# Unsorted
+#  These arithmetic tests were previously in other files, eventually
+#  should be parametrized and put into tests.arithmetic
+
+
+class TestTimeSeriesArithmetic:
+    def test_series_add_tz_mismatch_converts_to_utc(self):
+        rng = date_range("1/1/2011", periods=100, freq="h", tz="utc")
+
+        perm = np.random.default_rng(2).permutation(100)[:90]
+        ser1 = Series(
+            np.random.default_rng(2).standard_normal(90),
+            index=rng.take(perm).tz_convert("US/Eastern"),
+        )
+
+        perm = np.random.default_rng(2).permutation(100)[:90]
+        ser2 = Series(
+            np.random.default_rng(2).standard_normal(90),
+            index=rng.take(perm).tz_convert("Europe/Berlin"),
+        )
+
+        result = ser1 + ser2
+
+        uts1 = ser1.tz_convert("utc")
+        uts2 = ser2.tz_convert("utc")
+        expected = uts1 + uts2
+
+        # sort since input indexes are not equal
+        expected = expected.sort_index()
+
+        assert result.index.tz is timezone.utc
+        tm.assert_series_equal(result, expected)
+
+    def test_series_add_aware_naive_raises(self):
+        rng = date_range("1/1/2011", periods=10, freq="h")
+        ser = Series(np.random.default_rng(2).standard_normal(len(rng)), index=rng)
+
+        ser_utc = ser.tz_localize("utc")
+
+        msg = "Cannot join tz-naive with tz-aware DatetimeIndex"
+        with pytest.raises(Exception, match=msg):
+            ser + ser_utc
+
+        with pytest.raises(Exception, match=msg):
+            ser_utc + ser
+
+    # TODO: belongs in tests/arithmetic?
+    def test_datetime_understood(self, unit):
+        # Ensures it doesn't fail to create the right series
+        # reported in issue#16726
+        series = Series(date_range("2012-01-01", periods=3, unit=unit))
+        offset = pd.offsets.DateOffset(days=6)
+        result = series - offset
+        exp_dti = pd.to_datetime(["2011-12-26", "2011-12-27", "2011-12-28"]).as_unit(
+            unit
+        )
+        expected = Series(exp_dti)
+        tm.assert_series_equal(result, expected)
+
+    def test_align_date_objects_with_datetimeindex(self):
+        rng = date_range("1/1/2000", periods=20)
+        ts = Series(np.random.default_rng(2).standard_normal(20), index=rng)
+
+        ts_slice = ts[5:]
+        ts2 = ts_slice.copy()
+        ts2.index = [x.date() for x in ts2.index]
+
+        result = ts + ts2
+        result2 = ts2 + ts
+        expected = ts + ts[5:]
+        expected.index = expected.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
+
+
+class TestNamePreservation:
+    @pytest.mark.parametrize("box", [list, tuple, np.array, Index, Series, pd.array])
+    @pytest.mark.parametrize("flex", [True, False])
+    def test_series_ops_name_retention(self, flex, box, names, all_binary_operators):
+        # GH#33930 consistent name-retention
+        op = all_binary_operators
+
+        left = Series(range(10), name=names[0])
+        right = Series(range(10), name=names[1])
+
+        name = op.__name__.strip("_")
+        is_logical = name in ["and", "rand", "xor", "rxor", "or", "ror"]
+
+        msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+
+        right = box(right)
+        if flex:
+            if is_logical:
+                # Series doesn't have these as flex methods
+                return
+            result = getattr(left, name)(right)
+        else:
+            if is_logical and box in [list, tuple]:
+                with pytest.raises(TypeError, match=msg):
+                    # GH#52264 logical ops with dtype-less sequences deprecated
+                    op(left, right)
+                return
+            result = op(left, right)
+
+        assert isinstance(result, Series)
+        if box in [Index, Series]:
+            assert result.name is names[2] or result.name == names[2]
+        else:
+            assert result.name is names[0] or result.name == names[0]
+
+    def test_binop_maybe_preserve_name(self, datetime_series):
+        # names match, preserve
+        result = datetime_series * datetime_series
+        assert result.name == datetime_series.name
+        result = datetime_series.mul(datetime_series)
+        assert result.name == datetime_series.name
+
+        result = datetime_series * datetime_series[:-2]
+        assert result.name == datetime_series.name
+
+        # names don't match, don't preserve
+        cp = datetime_series.copy()
+        cp.name = "something else"
+        result = datetime_series + cp
+        assert result.name is None
+        result = datetime_series.add(cp)
+        assert result.name is None
+
+        ops = ["add", "sub", "mul", "div", "truediv", "floordiv", "mod", "pow"]
+        ops = ops + ["r" + op for op in ops]
+        for op in ops:
+            # names match, preserve
+            ser = datetime_series.copy()
+            result = getattr(ser, op)(ser)
+            assert result.name == datetime_series.name
+
+            # names don't match, don't preserve
+            cp = datetime_series.copy()
+            cp.name = "changed"
+            result = getattr(ser, op)(cp)
+            assert result.name is None
+
+    def test_scalarop_preserve_name(self, datetime_series):
+        result = datetime_series * 2
+        assert result.name == datetime_series.name
+
+
+class TestInplaceOperations:
+    @pytest.mark.parametrize(
+        "dtype1, dtype2, dtype_expected, dtype_mul",
+        (
+            ("Int64", "Int64", "Int64", "Int64"),
+            ("float", "float", "float", "float"),
+            ("Int64", "float", "Float64", "Float64"),
+            ("Int64", "Float64", "Float64", "Float64"),
+        ),
+    )
+    def test_series_inplace_ops(self, dtype1, dtype2, dtype_expected, dtype_mul):
+        # GH 37910
+
+        ser1 = Series([1], dtype=dtype1)
+        ser2 = Series([2], dtype=dtype2)
+        ser1 += ser2
+        expected = Series([3], dtype=dtype_expected)
+        tm.assert_series_equal(ser1, expected)
+
+        ser1 -= ser2
+        expected = Series([1], dtype=dtype_expected)
+        tm.assert_series_equal(ser1, expected)
+
+        ser1 *= ser2
+        expected = Series([2], dtype=dtype_mul)
+        tm.assert_series_equal(ser1, expected)
+
+
+def test_none_comparison(request, series_with_simple_index):
+    series = series_with_simple_index
+
+    if len(series) < 1:
+        request.applymarker(
+            pytest.mark.xfail(reason="Test doesn't make sense on empty data")
+        )
+
+    # bug brought up by #1079
+    # changed from TypeError in 0.17.0
+    series.iloc[0] = np.nan
+
+    # noinspection PyComparisonWithNone
+    result = series == None  # noqa: E711
+    assert not result.iat[0]
+    assert not result.iat[1]
+
+    # noinspection PyComparisonWithNone
+    result = series != None  # noqa: E711
+    assert result.iat[0]
+    assert result.iat[1]
+
+    result = None == series  # noqa: E711
+    assert not result.iat[0]
+    assert not result.iat[1]
+
+    result = None != series  # noqa: E711
+    assert result.iat[0]
+    assert result.iat[1]
+
+    if lib.is_np_dtype(series.dtype, "M") or isinstance(series.dtype, DatetimeTZDtype):
+        # Following DatetimeIndex (and Timestamp) convention,
+        # inequality comparisons with Series[datetime64] raise
+        msg = "Invalid comparison"
+        with pytest.raises(TypeError, match=msg):
+            None > series
+        with pytest.raises(TypeError, match=msg):
+            series > None
+    else:
+        result = None > series
+        assert not result.iat[0]
+        assert not result.iat[1]
+
+        result = series < None
+        assert not result.iat[0]
+        assert not result.iat[1]
+
+
+def test_series_varied_multiindex_alignment():
+    # GH 20414
+    s1 = Series(
+        range(8),
+        index=pd.MultiIndex.from_product(
+            [list("ab"), list("xy"), [1, 2]], names=["ab", "xy", "num"]
+        ),
+    )
+    s2 = Series(
+        [1000 * i for i in range(1, 5)],
+        index=pd.MultiIndex.from_product([list("xy"), [1, 2]], names=["xy", "num"]),
+    )
+    result = s1.loc[pd.IndexSlice[["a"], :, :]] + s2
+    expected = Series(
+        [1000, 2001, 3002, 4003],
+        index=pd.MultiIndex.from_tuples(
+            [("a", "x", 1), ("a", "x", 2), ("a", "y", 1), ("a", "y", 2)],
+            names=["ab", "xy", "num"],
+        ),
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_rmod_consistent_large_series():
+    # GH 29602
+    result = Series([2] * 10001).rmod(-1)
+    expected = Series([1] * 10001)
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/test_arrow_interface.py b/pandas/pandas/tests/series/test_arrow_interface.py
new file mode 100644
index 0000000000000..e73cf9bee6aeb
--- /dev/null
+++ b/pandas/pandas/tests/series/test_arrow_interface.py
@@ -0,0 +1,61 @@
+import ctypes
+
+import pytest
+
+import pandas as pd
+
+pa = pytest.importorskip("pyarrow", minversion="16.0")
+
+
+def test_series_arrow_interface():
+    s = pd.Series([1, 4, 2])
+
+    capsule = s.__arrow_c_stream__()
+    assert (
+        ctypes.pythonapi.PyCapsule_IsValid(
+            ctypes.py_object(capsule), b"arrow_array_stream"
+        )
+        == 1
+    )
+
+    ca = pa.chunked_array(s)
+    expected = pa.chunked_array([[1, 4, 2]])
+    assert ca.equals(expected)
+    ca = pa.chunked_array(s, type=pa.int32())
+    expected = pa.chunked_array([[1, 4, 2]], type=pa.int32())
+    assert ca.equals(expected)
+
+
+def test_series_arrow_interface_arrow_dtypes():
+    s = pd.Series([1, 4, 2], dtype="Int64[pyarrow]")
+
+    capsule = s.__arrow_c_stream__()
+    assert (
+        ctypes.pythonapi.PyCapsule_IsValid(
+            ctypes.py_object(capsule), b"arrow_array_stream"
+        )
+        == 1
+    )
+
+    ca = pa.chunked_array(s)
+    expected = pa.chunked_array([[1, 4, 2]])
+    assert ca.equals(expected)
+    ca = pa.chunked_array(s, type=pa.int32())
+    expected = pa.chunked_array([[1, 4, 2]], type=pa.int32())
+    assert ca.equals(expected)
+
+
+def test_series_arrow_interface_stringdtype():
+    s = pd.Series(["foo", "bar"], dtype="string[pyarrow]")
+
+    capsule = s.__arrow_c_stream__()
+    assert (
+        ctypes.pythonapi.PyCapsule_IsValid(
+            ctypes.py_object(capsule), b"arrow_array_stream"
+        )
+        == 1
+    )
+
+    ca = pa.chunked_array(s)
+    expected = pa.chunked_array([["foo", "bar"]], type=pa.large_string())
+    assert ca.equals(expected)
diff --git a/pandas/pandas/tests/series/test_constructors.py b/pandas/pandas/tests/series/test_constructors.py
new file mode 100644
index 0000000000000..a2be698c0ec28
--- /dev/null
+++ b/pandas/pandas/tests/series/test_constructors.py
@@ -0,0 +1,2280 @@
+from collections import OrderedDict
+from collections.abc import Iterator
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+from dateutil.tz import tzoffset
+import numpy as np
+from numpy import ma
+import pytest
+
+from pandas._libs import (
+    iNaT,
+    lib,
+)
+from pandas.compat import HAS_PYARROW
+from pandas.compat.numpy import np_version_gt2
+from pandas.errors import IntCastingNaNError
+
+from pandas.core.dtypes.dtypes import CategoricalDtype
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    DatetimeIndex,
+    DatetimeTZDtype,
+    Index,
+    Interval,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    Period,
+    RangeIndex,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.arrays import (
+    IntegerArray,
+    IntervalArray,
+    period_array,
+)
+from pandas.core.internals.blocks import NumpyBlock
+
+
+class TestSeriesConstructors:
+    def test_from_ints_with_non_nano_dt64_dtype(self, index_or_series):
+        values = np.arange(10)
+
+        res = index_or_series(values, dtype="M8[s]")
+        expected = index_or_series(values.astype("M8[s]"))
+        tm.assert_equal(res, expected)
+
+        res = index_or_series(list(values), dtype="M8[s]")
+        tm.assert_equal(res, expected)
+
+    def test_from_na_value_and_interval_of_datetime_dtype(self):
+        # GH#41805
+        ser = Series([None], dtype="interval[datetime64[ns]]")
+        assert ser.isna().all()
+        assert ser.dtype == "interval[datetime64[ns], right]"
+
+    def test_infer_with_date_and_datetime(self):
+        # GH#49341 pre-2.0 we inferred datetime-and-date to datetime64, which
+        #  was inconsistent with Index behavior
+        ts = Timestamp(2016, 1, 1)
+        vals = [ts.to_pydatetime(), ts.date()]
+
+        ser = Series(vals)
+        expected = Series(vals, dtype=object)
+        tm.assert_series_equal(ser, expected)
+
+        idx = Index(vals)
+        expected = Index(vals, dtype=object)
+        tm.assert_index_equal(idx, expected)
+
+    def test_unparsable_strings_with_dt64_dtype(self):
+        # pre-2.0 these would be silently ignored and come back with object dtype
+        vals = ["aa"]
+        msg = "^Unknown datetime string format, unable to parse: aa$"
+        with pytest.raises(ValueError, match=msg):
+            Series(vals, dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            Series(np.array(vals, dtype=object), dtype="datetime64[ns]")
+
+    def test_invalid_dtype_conversion_datetime_to_timedelta(self):
+        # GH#60728
+        vals = Series([NaT, Timestamp(2025, 1, 1)], dtype="datetime64[ns]")
+        msg = r"^Cannot cast DatetimeArray to dtype timedelta64\[ns\]$"
+        with pytest.raises(TypeError, match=msg):
+            Series(vals, dtype="timedelta64[ns]")
+
+    @pytest.mark.parametrize(
+        "constructor",
+        [
+            # NOTE: some overlap with test_constructor_empty but that test does not
+            # test for None or an empty generator.
+            # test_constructor_pass_none tests None but only with the index also
+            # passed.
+            (lambda idx: Series(index=idx)),
+            (lambda idx: Series(None, index=idx)),
+            (lambda idx: Series({}, index=idx)),
+            (lambda idx: Series((), index=idx)),
+            (lambda idx: Series([], index=idx)),
+            (lambda idx: Series((_ for _ in []), index=idx)),
+            (lambda idx: Series(data=None, index=idx)),
+            (lambda idx: Series(data={}, index=idx)),
+            (lambda idx: Series(data=(), index=idx)),
+            (lambda idx: Series(data=[], index=idx)),
+            (lambda idx: Series(data=(_ for _ in []), index=idx)),
+        ],
+    )
+    @pytest.mark.parametrize("empty_index", [None, []])
+    def test_empty_constructor(self, constructor, empty_index):
+        # GH 49573 (addition of empty_index parameter)
+        expected = Series(index=empty_index)
+        result = constructor(empty_index)
+
+        assert result.dtype == object
+        assert len(result.index) == 0
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_invalid_dtype(self):
+        # GH15520
+        msg = "not understood"
+        invalid_list = [Timestamp, "Timestamp", list]
+        for dtype in invalid_list:
+            with pytest.raises(TypeError, match=msg):
+                Series([], name="time", dtype=dtype)
+
+    def test_invalid_compound_dtype(self):
+        # GH#13296
+        c_dtype = np.dtype([("a", "i8"), ("b", "f4")])
+        cdt_arr = np.array([(1, 0.4), (256, -13)], dtype=c_dtype)
+
+        with pytest.raises(ValueError, match="Use DataFrame instead"):
+            Series(cdt_arr, index=["A", "B"])
+
+    def test_scalar_conversion(self):
+        # Pass in scalar is disabled
+        scalar = Series(0.5)
+        assert not isinstance(scalar, float)
+
+    def test_scalar_extension_dtype(self, ea_scalar_and_dtype):
+        # GH 28401
+
+        ea_scalar, ea_dtype = ea_scalar_and_dtype
+
+        ser = Series(ea_scalar, index=range(3))
+        expected = Series([ea_scalar] * 3, dtype=ea_dtype)
+
+        assert ser.dtype == ea_dtype
+        tm.assert_series_equal(ser, expected)
+
+    def test_constructor(self, datetime_series, using_infer_string):
+        empty_series = Series()
+        assert datetime_series.index._is_all_dates
+
+        # Pass in Series
+        derived = Series(datetime_series)
+        assert derived.index._is_all_dates
+
+        tm.assert_index_equal(derived.index, datetime_series.index)
+        # Ensure new index is not created
+        assert id(datetime_series.index) == id(derived.index)
+
+        # Mixed type Series
+        mixed = Series(["hello", np.nan], index=[0, 1])
+        assert mixed.dtype == np.object_ if not using_infer_string else "str"
+        assert np.isnan(mixed[1])
+
+        assert not empty_series.index._is_all_dates
+        assert not Series().index._is_all_dates
+
+        # exception raised is of type ValueError GH35744
+        with pytest.raises(
+            ValueError,
+            match=r"Data must be 1-dimensional, got ndarray of shape \(3, 3\) instead",
+        ):
+            Series(np.random.default_rng(2).standard_normal((3, 3)), index=np.arange(3))
+
+        mixed.name = "Series"
+        rs = Series(mixed).name
+        xp = "Series"
+        assert rs == xp
+
+        # raise on MultiIndex GH4187
+        m = MultiIndex.from_arrays([[1, 2], [3, 4]])
+        msg = "initializing a Series from a MultiIndex is not supported"
+        with pytest.raises(NotImplementedError, match=msg):
+            Series(m)
+
+    def test_constructor_index_ndim_gt_1_raises(self):
+        # GH#18579
+        df = DataFrame([[1, 2], [3, 4], [5, 6]], index=[3, 6, 9])
+        with pytest.raises(ValueError, match="Index data must be 1-dimensional"):
+            Series([1, 3, 2], index=df)
+
+    @pytest.mark.parametrize("input_class", [list, dict, OrderedDict])
+    def test_constructor_empty(self, input_class, using_infer_string):
+        empty = Series()
+        empty2 = Series(input_class())
+
+        # these are Index() and RangeIndex() which don't compare type equal
+        # but are just .equals
+        tm.assert_series_equal(empty, empty2, check_index_type=False)
+
+        # With explicit dtype:
+        empty = Series(dtype="float64")
+        empty2 = Series(input_class(), dtype="float64")
+        tm.assert_series_equal(empty, empty2, check_index_type=False)
+
+        # GH 18515 : with dtype=category:
+        empty = Series(dtype="category")
+        empty2 = Series(input_class(), dtype="category")
+        tm.assert_series_equal(empty, empty2, check_index_type=False)
+
+        if input_class is not list:
+            # With index:
+            empty = Series(index=range(10))
+            empty2 = Series(input_class(), index=range(10))
+            tm.assert_series_equal(empty, empty2)
+
+            # With index and dtype float64:
+            empty = Series(np.nan, index=range(10))
+            empty2 = Series(input_class(), index=range(10), dtype="float64")
+            tm.assert_series_equal(empty, empty2)
+
+            # GH 19853 : with empty string, index and dtype str
+            empty = Series("", dtype=str, index=range(3))
+            if using_infer_string:
+                empty2 = Series("", index=range(3), dtype="str")
+            else:
+                empty2 = Series("", index=range(3))
+            tm.assert_series_equal(empty, empty2)
+
+    @pytest.mark.parametrize("input_arg", [np.nan, float("nan")])
+    def test_constructor_nan(self, input_arg):
+        empty = Series(dtype="float64", index=range(10))
+        empty2 = Series(input_arg, index=range(10))
+
+        tm.assert_series_equal(empty, empty2, check_index_type=False)
+
+    @pytest.mark.parametrize(
+        "dtype",
+        ["f8", "i8", "M8[ns]", "m8[ns]", "category", "object", "datetime64[ns, UTC]"],
+    )
+    @pytest.mark.parametrize("index", [None, Index([])])
+    def test_constructor_dtype_only(self, dtype, index):
+        # GH-20865
+        result = Series(dtype=dtype, index=index)
+        assert result.dtype == dtype
+        assert len(result) == 0
+
+    def test_constructor_no_data_index_order(self):
+        result = Series(index=["b", "a", "c"])
+        assert result.index.tolist() == ["b", "a", "c"]
+
+    def test_constructor_no_data_string_type(self):
+        # GH 22477
+        result = Series(index=[1], dtype=str)
+        assert np.isnan(result.iloc[0])
+
+    @pytest.mark.parametrize("item", ["entry", "ѐ", 13])
+    def test_constructor_string_element_string_type(self, item):
+        # GH 22477
+        result = Series(item, index=[1], dtype=str)
+        assert result.iloc[0] == str(item)
+
+    def test_constructor_dtype_str_na_values(self, string_dtype):
+        # https://github.com/pandas-dev/pandas/issues/21083
+        ser = Series(["x", None], dtype=string_dtype)
+        result = ser.isna()
+        expected = Series([False, True])
+        tm.assert_series_equal(result, expected)
+        assert ser.iloc[1] is None
+
+        ser = Series(["x", np.nan], dtype=string_dtype)
+        assert np.isnan(ser.iloc[1])
+
+    def test_constructor_series(self):
+        index1 = ["d", "b", "a", "c"]
+        index2 = sorted(index1)
+        s1 = Series([4, 7, -5, 3], index=index1)
+        s2 = Series(s1, index=index2)
+
+        tm.assert_series_equal(s2, s1.sort_index())
+
+    def test_constructor_iterable(self):
+        # GH 21987
+        class Iter:
+            def __iter__(self) -> Iterator:
+                yield from range(10)
+
+        expected = Series(list(range(10)), dtype="int64")
+        result = Series(Iter(), dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_sequence(self):
+        # GH 21987
+        expected = Series(list(range(10)), dtype="int64")
+        result = Series(range(10), dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_single_str(self):
+        # GH 21987
+        expected = Series(["abc"])
+        result = Series("abc")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_list_like(self):
+        # make sure that we are coercing different
+        # list-likes to standard dtypes and not
+        # platform specific
+        expected = Series([1, 2, 3], dtype="int64")
+        for obj in [[1, 2, 3], (1, 2, 3), np.array([1, 2, 3], dtype="int64")]:
+            result = Series(obj, index=[0, 1, 2])
+            tm.assert_series_equal(result, expected)
+
+    def test_constructor_boolean_index(self):
+        # GH#18579
+        s1 = Series([1, 2, 3], index=[4, 5, 6])
+
+        index = s1 == 2
+        result = Series([1, 3, 2], index=index)
+        expected = Series([1, 3, 2], index=[False, True, False])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["bool", "int32", "int64", "float64"])
+    def test_constructor_index_dtype(self, dtype):
+        # GH 17088
+
+        s = Series(Index([0, 2, 4]), dtype=dtype)
+        assert s.dtype == dtype
+
+    @pytest.mark.parametrize(
+        "input_vals",
+        [
+            [1, 2],
+            ["1", "2"],
+            list(date_range("1/1/2011", periods=2, freq="h")),
+            list(date_range("1/1/2011", periods=2, freq="h", tz="US/Eastern")),
+            [Interval(left=0, right=5)],
+        ],
+    )
+    def test_constructor_list_str(self, input_vals, string_dtype):
+        # GH 16605
+        # Ensure that data elements from a list are converted to strings
+        # when dtype is str, 'str', or 'U'
+        result = Series(input_vals, dtype=string_dtype)
+        expected = Series(input_vals).astype(string_dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_list_str_na(self, string_dtype):
+        result = Series([1.0, 2.0, np.nan], dtype=string_dtype)
+        expected = Series(["1.0", "2.0", np.nan], dtype=object)
+        tm.assert_series_equal(result, expected)
+        assert np.isnan(result[2])
+
+    def test_constructor_generator(self):
+        gen = (i for i in range(10))
+
+        result = Series(gen)
+        exp = Series(range(10))
+        tm.assert_series_equal(result, exp)
+
+        # same but with non-default index
+        gen = (i for i in range(10))
+        result = Series(gen, index=range(10, 20))
+        exp.index = range(10, 20)
+        tm.assert_series_equal(result, exp)
+
+    def test_constructor_map(self):
+        # GH8909
+        m = (x for x in range(10))
+
+        result = Series(m)
+        exp = Series(range(10))
+        tm.assert_series_equal(result, exp)
+
+        # same but with non-default index
+        m = (x for x in range(10))
+        result = Series(m, index=range(10, 20))
+        exp.index = range(10, 20)
+        tm.assert_series_equal(result, exp)
+
+    def test_constructor_categorical(self):
+        cat = Categorical([0, 1, 2, 0, 1, 2], ["a", "b", "c"])
+        res = Series(cat)
+        tm.assert_categorical_equal(res.values, cat)
+
+        # can cast to a new dtype
+        result = Series(Categorical([1, 2, 3]), dtype="int64")
+        expected = Series([1, 2, 3], dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_construct_from_categorical_with_dtype(self):
+        # GH12574
+        ser = Series(Categorical([1, 2, 3]), dtype="category")
+        assert isinstance(ser.dtype, CategoricalDtype)
+
+    def test_construct_intlist_values_category_dtype(self):
+        ser = Series([1, 2, 3], dtype="category")
+        assert isinstance(ser.dtype, CategoricalDtype)
+
+    def test_constructor_categorical_with_coercion(self):
+        factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"])
+        # test basic creation / coercion of categoricals
+        s = Series(factor, name="A")
+        assert s.dtype == "category"
+        assert len(s) == len(factor)
+
+        # in a frame
+        df = DataFrame({"A": factor})
+        result = df["A"]
+        tm.assert_series_equal(result, s)
+        result = df.iloc[:, 0]
+        tm.assert_series_equal(result, s)
+        assert len(df) == len(factor)
+
+        df = DataFrame({"A": s})
+        result = df["A"]
+        tm.assert_series_equal(result, s)
+        assert len(df) == len(factor)
+
+        # multiples
+        df = DataFrame({"A": s, "B": s, "C": 1})
+        result1 = df["A"]
+        result2 = df["B"]
+        tm.assert_series_equal(result1, s)
+        tm.assert_series_equal(result2, s, check_names=False)
+        assert result2.name == "B"
+        assert len(df) == len(factor)
+
+    def test_constructor_categorical_with_coercion2(self):
+        # GH8623
+        x = DataFrame(
+            [[1, "John P. Doe"], [2, "Jane Dove"], [1, "John P. Doe"]],
+            columns=["person_id", "person_name"],
+        )
+        x["person_name"] = Categorical(x.person_name)  # doing this breaks transform
+
+        expected = x.iloc[0].person_name
+        result = x.person_name.iloc[0]
+        assert result == expected
+
+        result = x.person_name[0]
+        assert result == expected
+
+        result = x.person_name.loc[0]
+        assert result == expected
+
+    def test_constructor_series_to_categorical(self):
+        # see GH#16524: test conversion of Series to Categorical
+        series = Series(["a", "b", "c"])
+
+        result = Series(series, dtype="category")
+        expected = Series(["a", "b", "c"], dtype="category")
+
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_categorical_dtype(self):
+        result = Series(
+            ["a", "b"], dtype=CategoricalDtype(["a", "b", "c"], ordered=True)
+        )
+        assert isinstance(result.dtype, CategoricalDtype)
+        tm.assert_index_equal(result.cat.categories, Index(["a", "b", "c"]))
+        assert result.cat.ordered
+
+        result = Series(["a", "b"], dtype=CategoricalDtype(["b", "a"]))
+        assert isinstance(result.dtype, CategoricalDtype)
+        tm.assert_index_equal(result.cat.categories, Index(["b", "a"]))
+        assert result.cat.ordered is False
+
+        # GH 19565 - Check broadcasting of scalar with Categorical dtype
+        result = Series(
+            "a", index=[0, 1], dtype=CategoricalDtype(["a", "b"], ordered=True)
+        )
+        expected = Series(
+            ["a", "a"], index=[0, 1], dtype=CategoricalDtype(["a", "b"], ordered=True)
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_categorical_string(self):
+        # GH 26336: the string 'category' maintains existing CategoricalDtype
+        cdt = CategoricalDtype(categories=list("dabc"), ordered=True)
+        expected = Series(list("abcabc"), dtype=cdt)
+
+        # Series(Categorical, dtype='category') keeps existing dtype
+        cat = Categorical(list("abcabc"), dtype=cdt)
+        result = Series(cat, dtype="category")
+        tm.assert_series_equal(result, expected)
+
+        # Series(Series[Categorical], dtype='category') keeps existing dtype
+        result = Series(result, dtype="category")
+        tm.assert_series_equal(result, expected)
+
+    def test_categorical_sideeffects_free(self):
+        # Passing a categorical to a Series and then changing values in either
+        # the series or the categorical should not change the values in the
+        # other one, IF you specify copy!
+        cat = Categorical(["a", "b", "c", "a"])
+        s = Series(cat, copy=True)
+        assert s.cat is not cat
+        s = s.cat.rename_categories([1, 2, 3])
+        exp_s = np.array([1, 2, 3, 1], dtype=np.int64)
+        exp_cat = np.array(["a", "b", "c", "a"], dtype=np.object_)
+        tm.assert_numpy_array_equal(s.__array__(), exp_s)
+        tm.assert_numpy_array_equal(cat.__array__(), exp_cat)
+
+        # setting
+        s[0] = 2
+        exp_s2 = np.array([2, 2, 3, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(s.__array__(), exp_s2)
+        tm.assert_numpy_array_equal(cat.__array__(), exp_cat)
+
+        # however, copy is False by default
+        # so this WILL change values
+        cat = Categorical(["a", "b", "c", "a"])
+        s = Series(cat, copy=False)
+        assert s.values is cat
+        s = s.cat.rename_categories([1, 2, 3])
+        assert s.values is not cat
+        exp_s = np.array([1, 2, 3, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(s.__array__(), exp_s)
+
+        s[0] = 2
+        exp_s2 = np.array([2, 2, 3, 1], dtype=np.int64)
+        tm.assert_numpy_array_equal(s.__array__(), exp_s2)
+
+    def test_unordered_compare_equal(self):
+        left = Series(["a", "b", "c"], dtype=CategoricalDtype(["a", "b"]))
+        right = Series(Categorical(["a", "b", np.nan], categories=["a", "b"]))
+        tm.assert_series_equal(left, right)
+
+    def test_constructor_maskedarray(self):
+        data = ma.masked_all((3,), dtype=float)
+        result = Series(data)
+        expected = Series([np.nan, np.nan, np.nan])
+        tm.assert_series_equal(result, expected)
+
+        data[0] = 0.0
+        data[2] = 2.0
+        index = ["a", "b", "c"]
+        result = Series(data, index=index)
+        expected = Series([0.0, np.nan, 2.0], index=index)
+        tm.assert_series_equal(result, expected)
+
+        data[1] = 1.0
+        result = Series(data, index=index)
+        expected = Series([0.0, 1.0, 2.0], index=index)
+        tm.assert_series_equal(result, expected)
+
+        data = ma.masked_all((3,), dtype=int)
+        result = Series(data)
+        expected = Series([np.nan, np.nan, np.nan], dtype=float)
+        tm.assert_series_equal(result, expected)
+
+        data[0] = 0
+        data[2] = 2
+        index = ["a", "b", "c"]
+        result = Series(data, index=index)
+        expected = Series([0, np.nan, 2], index=index, dtype=float)
+        tm.assert_series_equal(result, expected)
+
+        data[1] = 1
+        result = Series(data, index=index)
+        expected = Series([0, 1, 2], index=index, dtype=int)
+        with pytest.raises(AssertionError, match="Series classes are different"):
+            # TODO should this be raising at all?
+            # https://github.com/pandas-dev/pandas/issues/56131
+            tm.assert_series_equal(result, expected)
+
+        data = ma.masked_all((3,), dtype=bool)
+        result = Series(data)
+        expected = Series([np.nan, np.nan, np.nan], dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        data[0] = True
+        data[2] = False
+        index = ["a", "b", "c"]
+        result = Series(data, index=index)
+        expected = Series([True, np.nan, False], index=index, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+        data[1] = True
+        result = Series(data, index=index)
+        expected = Series([True, True, False], index=index, dtype=bool)
+        with pytest.raises(AssertionError, match="Series classes are different"):
+            # TODO should this be raising at all?
+            # https://github.com/pandas-dev/pandas/issues/56131
+            tm.assert_series_equal(result, expected)
+
+        data = ma.masked_all((3,), dtype="M8[ns]")
+        result = Series(data)
+        expected = Series([iNaT, iNaT, iNaT], dtype="M8[ns]")
+        tm.assert_series_equal(result, expected)
+
+        data[0] = datetime(2001, 1, 1)
+        data[2] = datetime(2001, 1, 3)
+        index = ["a", "b", "c"]
+        result = Series(data, index=index)
+        expected = Series(
+            [datetime(2001, 1, 1), iNaT, datetime(2001, 1, 3)],
+            index=index,
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+        data[1] = datetime(2001, 1, 2)
+        result = Series(data, index=index)
+        expected = Series(
+            [datetime(2001, 1, 1), datetime(2001, 1, 2), datetime(2001, 1, 3)],
+            index=index,
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_maskedarray_hardened(self):
+        # Check numpy masked arrays with hard masks -- from GH24574
+        data = ma.masked_all((3,), dtype=float).harden_mask()
+        result = Series(data)
+        expected = Series([np.nan, np.nan, np.nan])
+        tm.assert_series_equal(result, expected)
+
+    def test_series_ctor_plus_datetimeindex(self):
+        rng = date_range("20090415", "20090519", freq="B")
+        data = {k: 1 for k in rng}
+
+        result = Series(data, index=rng)
+        assert result.index.is_(rng)
+
+    def test_constructor_default_index(self):
+        s = Series([0, 1, 2])
+        tm.assert_index_equal(s.index, Index(range(3)), exact=True)
+
+    @pytest.mark.parametrize(
+        "input",
+        [
+            [1, 2, 3],
+            (1, 2, 3),
+            list(range(3)),
+            Categorical(["a", "b", "a"]),
+            (i for i in range(3)),
+            (x for x in range(3)),
+        ],
+    )
+    def test_constructor_index_mismatch(self, input):
+        # GH 19342
+        # test that construction of a Series with an index of different length
+        # raises an error
+        msg = r"Length of values \(3\) does not match length of index \(4\)"
+        with pytest.raises(ValueError, match=msg):
+            Series(input, index=np.arange(4))
+
+    def test_constructor_numpy_scalar(self):
+        # GH 19342
+        # construction with a numpy scalar
+        # should not raise
+        result = Series(np.array(100), index=np.arange(4), dtype="int64")
+        expected = Series(100, index=np.arange(4), dtype="int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_broadcast_list(self):
+        # GH 19342
+        # construction with single-element container and index
+        # should raise
+        msg = r"Length of values \(1\) does not match length of index \(3\)"
+        with pytest.raises(ValueError, match=msg):
+            Series(["foo"], index=["a", "b", "c"])
+
+    def test_constructor_corner(self):
+        df = DataFrame(range(5), index=date_range("2020-01-01", periods=5))
+        objs = [df, df]
+        s = Series(objs, index=[0, 1])
+        assert isinstance(s, Series)
+
+    def test_constructor_sanitize(self):
+        s = Series(np.array([1.0, 1.0, 8.0]), dtype="i8")
+        assert s.dtype == np.dtype("i8")
+
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            Series(np.array([1.0, 1.0, np.nan]), copy=True, dtype="i8")
+
+    def test_constructor_copy(self):
+        # GH15125
+        # test dtype parameter has no side effects on copy=True
+        for data in [[1.0], np.array([1.0])]:
+            x = Series(data)
+            y = Series(x, copy=True, dtype=float)
+
+            # copy=True maintains original data in Series
+            tm.assert_series_equal(x, y)
+
+            # changes to origin of copy does not affect the copy
+            x[0] = 2.0
+            assert not x.equals(y)
+            assert x[0] == 2.0
+            assert y[0] == 1.0
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("20170101", periods=3, tz="US/Eastern"),
+            date_range("20170101", periods=3),
+            timedelta_range("1 day", periods=3),
+            period_range("2012Q1", periods=3, freq="Q"),
+            Index(list("abc")),
+            Index([1, 2, 3]),
+            RangeIndex(0, 3),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_constructor_limit_copies(self, index):
+        # GH 17449
+        # limit copies of input
+        s = Series(index)
+
+        # we make 1 copy; this is just a smoke test here
+        assert s._mgr.blocks[0].values is not index
+
+    def test_constructor_shallow_copy(self):
+        # constructing a Series from Series with copy=False should still
+        # give a "shallow" copy (share data, not attributes)
+        # https://github.com/pandas-dev/pandas/issues/49523
+        s = Series([1, 2, 3])
+        s_orig = s.copy()
+        s2 = Series(s)
+        assert s2._mgr is not s._mgr
+        # Overwriting index of s2 doesn't change s
+        s2.index = ["a", "b", "c"]
+        tm.assert_series_equal(s, s_orig)
+
+    def test_constructor_pass_none(self):
+        s = Series(None, index=range(5))
+        assert s.dtype == np.float64
+
+        s = Series(None, index=range(5), dtype=object)
+        assert s.dtype == np.object_
+
+        # GH 7431
+        # inference on the index
+        s = Series(index=np.array([None]))
+        expected = Series(index=Index([None]))
+        tm.assert_series_equal(s, expected)
+
+    def test_constructor_pass_nan_nat(self):
+        # GH 13467
+        exp = Series([np.nan, np.nan], dtype=np.float64)
+        assert exp.dtype == np.float64
+        tm.assert_series_equal(Series([np.nan, np.nan]), exp)
+        tm.assert_series_equal(Series(np.array([np.nan, np.nan])), exp)
+
+        exp = Series([NaT, NaT])
+        assert exp.dtype == "datetime64[s]"
+        tm.assert_series_equal(Series([NaT, NaT]), exp)
+        tm.assert_series_equal(Series(np.array([NaT, NaT])), exp)
+
+        tm.assert_series_equal(Series([NaT, np.nan]), exp)
+        tm.assert_series_equal(Series(np.array([NaT, np.nan])), exp)
+
+        tm.assert_series_equal(Series([np.nan, NaT]), exp)
+        tm.assert_series_equal(Series(np.array([np.nan, NaT])), exp)
+
+    def test_constructor_cast(self):
+        msg = "could not convert string to float"
+        with pytest.raises(ValueError, match=msg):
+            Series(["a", "b", "c"], dtype=float)
+
+    def test_constructor_signed_int_overflow_raises(self):
+        # GH#41734 disallow silent overflow, enforced in 2.0
+        if np_version_gt2:
+            msg = "The elements provided in the data cannot all be casted to the dtype"
+            err = OverflowError
+        else:
+            msg = "Values are too large to be losslessly converted"
+            err = ValueError
+        with pytest.raises(err, match=msg):
+            Series([1, 200, 923442], dtype="int8")
+
+        with pytest.raises(err, match=msg):
+            Series([1, 200, 923442], dtype="uint8")
+
+    @pytest.mark.parametrize(
+        "values",
+        [
+            np.array([1], dtype=np.uint16),
+            np.array([1], dtype=np.uint32),
+            np.array([1], dtype=np.uint64),
+            [np.uint16(1)],
+            [np.uint32(1)],
+            [np.uint64(1)],
+        ],
+    )
+    def test_constructor_numpy_uints(self, values):
+        # GH#47294
+        value = values[0]
+        result = Series(values)
+
+        assert result[0].dtype == value.dtype
+        assert result[0] == value
+
+    def test_constructor_unsigned_dtype_overflow(self, any_unsigned_int_numpy_dtype):
+        # see gh-15832
+        if np_version_gt2:
+            msg = (
+                f"The elements provided in the data cannot "
+                f"all be casted to the dtype {any_unsigned_int_numpy_dtype}"
+            )
+        else:
+            msg = "Trying to coerce negative values to unsigned integers"
+        with pytest.raises(OverflowError, match=msg):
+            Series([-1], dtype=any_unsigned_int_numpy_dtype)
+
+    def test_constructor_floating_data_int_dtype(self, frame_or_series):
+        # GH#40110
+        arr = np.random.default_rng(2).standard_normal(2)
+
+        # Long-standing behavior (for Series, new in 2.0 for DataFrame)
+        #  has been to ignore the dtype on these;
+        #  not clear if this is what we want long-term
+        # expected = frame_or_series(arr)
+
+        # GH#49599 as of 2.0 we raise instead of silently retaining float dtype
+        msg = "Trying to coerce float values to integer"
+        with pytest.raises(ValueError, match=msg):
+            frame_or_series(arr, dtype="i8")
+
+        with pytest.raises(ValueError, match=msg):
+            frame_or_series(list(arr), dtype="i8")
+
+        # pre-2.0, when we had NaNs, we silently ignored the integer dtype
+        arr[0] = np.nan
+        # expected = frame_or_series(arr)
+
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            frame_or_series(arr, dtype="i8")
+
+        exc = IntCastingNaNError
+        if frame_or_series is Series:
+            # TODO: try to align these
+            exc = ValueError
+            msg = "cannot convert float NaN to integer"
+        with pytest.raises(exc, match=msg):
+            # same behavior if we pass list instead of the ndarray
+            frame_or_series(list(arr), dtype="i8")
+
+        # float array that can be losslessly cast to integers
+        arr = np.array([1.0, 2.0], dtype="float64")
+        expected = frame_or_series(arr.astype("i8"))
+
+        obj = frame_or_series(arr, dtype="i8")
+        tm.assert_equal(obj, expected)
+
+        obj = frame_or_series(list(arr), dtype="i8")
+        tm.assert_equal(obj, expected)
+
+    def test_constructor_coerce_float_fail(self, any_int_numpy_dtype):
+        # see gh-15832
+        # Updated: make sure we treat this list the same as we would treat
+        #  the equivalent ndarray
+        # GH#49599 pre-2.0 we silently retained float dtype, in 2.0 we raise
+        vals = [1, 2, 3.5]
+
+        msg = "Trying to coerce float values to integer"
+        with pytest.raises(ValueError, match=msg):
+            Series(vals, dtype=any_int_numpy_dtype)
+        with pytest.raises(ValueError, match=msg):
+            Series(np.array(vals), dtype=any_int_numpy_dtype)
+
+    def test_constructor_coerce_float_valid(self, float_numpy_dtype):
+        s = Series([1, 2, 3.5], dtype=float_numpy_dtype)
+        expected = Series([1, 2, 3.5]).astype(float_numpy_dtype)
+        tm.assert_series_equal(s, expected)
+
+    def test_constructor_invalid_coerce_ints_with_float_nan(self, any_int_numpy_dtype):
+        # GH 22585
+        # Updated: make sure we treat this list the same as we would treat the
+        # equivalent ndarray
+        vals = [1, 2, np.nan]
+        # pre-2.0 this would return with a float dtype, in 2.0 we raise
+
+        msg = "cannot convert float NaN to integer"
+        with pytest.raises(ValueError, match=msg):
+            Series(vals, dtype=any_int_numpy_dtype)
+        msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+        with pytest.raises(IntCastingNaNError, match=msg):
+            Series(np.array(vals), dtype=any_int_numpy_dtype)
+
+    def test_constructor_dtype_no_cast(self):
+        # see gh-1572
+        s = Series([1, 2, 3])
+        s2 = Series(s, dtype=np.int64)
+
+        s2[1] = 5
+        assert s[1] == 2
+
+    def test_constructor_datelike_coercion(self):
+        # GH 9477
+        # incorrectly inferring on dateimelike looking when object dtype is
+        # specified
+        s = Series([Timestamp("20130101"), "NOV"], dtype=object)
+        assert s.iloc[0] == Timestamp("20130101")
+        assert s.iloc[1] == "NOV"
+        assert s.dtype == object
+
+    def test_constructor_datelike_coercion2(self):
+        # the dtype was being reset on the slicing and re-inferred to datetime
+        # even thought the blocks are mixed
+        belly = "216 3T19".split()
+        wing1 = "2T15 4H19".split()
+        wing2 = "416 4T20".split()
+        mat = pd.to_datetime("2016-01-22 2019-09-07".split())
+        df = DataFrame({"wing1": wing1, "wing2": wing2, "mat": mat}, index=belly)
+
+        result = df.loc["3T19"]
+        assert result.dtype == object
+        result = df.loc["216"]
+        assert result.dtype == object
+
+    def test_constructor_mixed_int_and_timestamp(self, frame_or_series):
+        # specifically Timestamp with nanos, not datetimes
+        objs = [Timestamp(9), 10, NaT._value]
+        result = frame_or_series(objs, dtype="M8[ns]")
+
+        expected = frame_or_series([Timestamp(9), Timestamp(10), NaT])
+        tm.assert_equal(result, expected)
+
+    def test_constructor_datetimes_with_nulls(self):
+        # gh-15869
+        for arr in [
+            np.array([None, None, None, None, datetime.now(), None]),
+            np.array([None, None, datetime.now(), None]),
+        ]:
+            result = Series(arr)
+            assert result.dtype == "M8[us]"
+
+    def test_constructor_dtype_datetime64(self):
+        s = Series(iNaT, dtype="M8[ns]", index=range(5))
+        assert isna(s).all()
+
+        # in theory this should be all nulls, but since
+        # we are not specifying a dtype is ambiguous
+        s = Series(iNaT, index=range(5))
+        assert not isna(s).all()
+
+        s = Series(np.nan, dtype="M8[ns]", index=range(5))
+        assert isna(s).all()
+
+        s = Series([datetime(2001, 1, 2, 0, 0), iNaT], dtype="M8[ns]")
+        assert isna(s[1])
+        assert s.dtype == "M8[ns]"
+
+        s = Series([datetime(2001, 1, 2, 0, 0), np.nan], dtype="M8[ns]")
+        assert isna(s[1])
+        assert s.dtype == "M8[ns]"
+
+    def test_constructor_dtype_datetime64_10(self):
+        # GH3416
+        pydates = [datetime(2013, 1, 1), datetime(2013, 1, 2), datetime(2013, 1, 3)]
+        dates = [np.datetime64(x) for x in pydates]
+
+        ser = Series(dates)
+        assert ser.dtype == "M8[us]"
+
+        ser.iloc[0] = np.nan
+        assert ser.dtype == "M8[us]"
+
+        # GH3414 related
+        expected = Series(pydates, dtype="datetime64[ms]")
+
+        result = Series(Series(dates).astype(np.int64) / 1000, dtype="M8[ms]")
+        tm.assert_series_equal(result, expected)
+
+        result = Series(dates, dtype="datetime64[ms]")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(
+            [NaT, datetime(2013, 1, 2), datetime(2013, 1, 3)], dtype="datetime64[ns]"
+        )
+        result = Series([np.nan] + dates[1:], dtype="datetime64[ns]")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dtype_datetime64_11(self):
+        pydates = [datetime(2013, 1, 1), datetime(2013, 1, 2), datetime(2013, 1, 3)]
+        dates = [np.datetime64(x) for x in pydates]
+
+        dts = Series(dates, dtype="datetime64[ns]")
+
+        # valid astype
+        dts.astype("int64")
+
+        # invalid casting
+        msg = r"Converting from datetime64\[ns\] to int32 is not supported"
+        with pytest.raises(TypeError, match=msg):
+            dts.astype("int32")
+
+        # ints are ok
+        # we test with np.int64 to get similar results on
+        # windows / 32-bit platforms
+        result = Series(dts, dtype=np.int64)
+        expected = Series(dts.astype(np.int64))
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dtype_datetime64_9(self):
+        # invalid dates can be help as object
+        result = Series([datetime(2, 1, 1)])
+        assert result[0] == datetime(2, 1, 1, 0, 0)
+
+        result = Series([datetime(3000, 1, 1)])
+        assert result[0] == datetime(3000, 1, 1, 0, 0)
+
+    def test_constructor_dtype_datetime64_8(self):
+        # don't mix types
+        result = Series([Timestamp("20130101"), 1], index=["a", "b"])
+        assert result["a"] == Timestamp("20130101")
+        assert result["b"] == 1
+
+    def test_constructor_dtype_datetime64_7(self):
+        # GH6529
+        # coerce datetime64 non-ns properly
+        dates = date_range("01-Jan-2015", "01-Dec-2015", freq="ME")
+        values2 = dates.view(np.ndarray).astype("datetime64[ns]")
+        expected = Series(values2, index=dates)
+
+        for unit in ["s", "D", "ms", "us", "ns"]:
+            dtype = np.dtype(f"M8[{unit}]")
+            values1 = dates.view(np.ndarray).astype(dtype)
+            result = Series(values1, dates)
+            if unit == "D":
+                # for unit="D" we cast to nearest-supported reso, i.e. "s"
+                dtype = np.dtype("M8[s]")
+            assert result.dtype == dtype
+            tm.assert_series_equal(result, expected.astype(dtype))
+
+        # GH 13876
+        # coerce to non-ns to object properly
+        expected = Series(values2, index=dates, dtype=object)
+        for dtype in ["s", "D", "ms", "us", "ns"]:
+            values1 = dates.view(np.ndarray).astype(f"M8[{dtype}]")
+            result = Series(values1, index=dates, dtype=object)
+            tm.assert_series_equal(result, expected)
+
+        # leave datetime.date alone
+        dates2 = np.array([d.date() for d in dates.to_pydatetime()], dtype=object)
+        series1 = Series(dates2, dates)
+        tm.assert_numpy_array_equal(series1.values, dates2)
+        assert series1.dtype == object
+
+    def test_constructor_dtype_datetime64_6(self):
+        # as of 2.0, these no longer infer datetime64 based on the strings,
+        #  matching the Index behavior
+
+        ser = Series([None, NaT, "2013-08-05 15:30:00.000001"])
+        assert ser.dtype == object
+
+        ser = Series([np.nan, NaT, "2013-08-05 15:30:00.000001"])
+        assert ser.dtype == object
+
+        ser = Series([NaT, None, "2013-08-05 15:30:00.000001"])
+        assert ser.dtype == object
+
+        ser = Series([NaT, np.nan, "2013-08-05 15:30:00.000001"])
+        assert ser.dtype == object
+
+    def test_constructor_dtype_datetime64_5(self):
+        # tz-aware (UTC and other tz's)
+        # GH 8411
+        dr = date_range("20130101", periods=3)
+        assert Series(dr).iloc[0].tz is None
+        dr = date_range("20130101", periods=3, tz="UTC")
+        assert str(Series(dr).iloc[0].tz) == "UTC"
+        dr = date_range("20130101", periods=3, tz="US/Eastern")
+        assert str(Series(dr).iloc[0].tz) == "US/Eastern"
+
+    def test_constructor_dtype_datetime64_4(self):
+        # non-convertible
+        ser = Series([1479596223000, -1479590, NaT])
+        assert ser.dtype == "object"
+        assert ser[2] is NaT
+        assert "NaT" in str(ser)
+
+    def test_constructor_dtype_datetime64_3(self):
+        # if we passed a NaT it remains
+        ser = Series([datetime(2010, 1, 1), datetime(2, 1, 1), NaT])
+        assert ser.dtype == "M8[us]"
+        assert ser[2] is NaT
+        assert "NaT" in str(ser)
+
+    def test_constructor_dtype_datetime64_2(self):
+        # if we passed a nan it remains
+        ser = Series([datetime(2010, 1, 1), datetime(2, 1, 1), np.nan])
+        assert ser.dtype == "M8[us]"
+        assert ser[2] is NaT
+        assert "NaT" in str(ser)
+
+    def test_constructor_with_datetime_tz(self):
+        # 8260
+        # support datetime64 with tz
+
+        dr = date_range("20130101", periods=3, tz="US/Eastern")
+        s = Series(dr)
+        assert s.dtype.name == "datetime64[ns, US/Eastern]"
+        assert s.dtype == "datetime64[ns, US/Eastern]"
+        assert isinstance(s.dtype, DatetimeTZDtype)
+        assert "datetime64[ns, US/Eastern]" in str(s)
+
+        # export
+        result = s.values
+        assert isinstance(result, np.ndarray)
+        assert result.dtype == "datetime64[ns]"
+
+        exp = DatetimeIndex(result)
+        exp = exp.tz_localize("UTC").tz_convert(tz=s.dt.tz)
+        tm.assert_index_equal(dr, exp)
+
+        # indexing
+        result = s.iloc[0]
+        assert result == Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern")
+        result = s[0]
+        assert result == Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern")
+
+        result = s[Series([True, True, False], index=s.index)]
+        tm.assert_series_equal(result, s[0:2])
+
+        result = s.iloc[0:1]
+        tm.assert_series_equal(result, Series(dr[0:1]))
+
+        # concat
+        result = pd.concat([s.iloc[0:1], s.iloc[1:]])
+        tm.assert_series_equal(result, s)
+
+        # short str
+        assert "datetime64[ns, US/Eastern]" in str(s)
+
+        # formatting with NaT
+        result = s.shift()
+        assert "datetime64[ns, US/Eastern]" in str(result)
+        assert "NaT" in str(result)
+
+        result = DatetimeIndex(s, freq="infer")
+        tm.assert_index_equal(result, dr)
+
+    def test_constructor_with_datetime_tz5(self):
+        # long str
+        ser = Series(date_range("20130101", periods=1000, tz="US/Eastern"))
+        assert "datetime64[ns, US/Eastern]" in str(ser)
+
+    def test_constructor_with_datetime_tz4(self):
+        # inference
+        ser = Series(
+            [
+                Timestamp("2013-01-01 13:00:00-0800", tz="US/Pacific"),
+                Timestamp("2013-01-02 14:00:00-0800", tz="US/Pacific"),
+            ]
+        )
+        assert ser.dtype == "datetime64[s, US/Pacific]"
+        assert lib.infer_dtype(ser, skipna=True) == "datetime64"
+
+    def test_constructor_with_datetime_tz3(self):
+        ser = Series(
+            [
+                Timestamp("2013-01-01 13:00:00-0800", tz="US/Pacific"),
+                Timestamp("2013-01-02 14:00:00-0800", tz="US/Eastern"),
+            ]
+        )
+        assert ser.dtype == "object"
+        assert lib.infer_dtype(ser, skipna=True) == "datetime"
+
+    def test_constructor_with_datetime_tz2(self):
+        # with all NaT
+        ser = Series(NaT, index=[0, 1], dtype="datetime64[ns, US/Eastern]")
+        dti = DatetimeIndex(["NaT", "NaT"], tz="US/Eastern").as_unit("ns")
+        expected = Series(dti)
+        tm.assert_series_equal(ser, expected)
+
+    def test_constructor_no_partial_datetime_casting(self):
+        # GH#40111
+        vals = [
+            "nan",
+            Timestamp("1990-01-01"),
+            "2015-03-14T16:15:14.123-08:00",
+            "2019-03-04T21:56:32.620-07:00",
+            None,
+        ]
+        ser = Series(vals)
+        assert all(ser[i] is vals[i] for i in range(len(vals)))
+
+    @pytest.mark.parametrize("arr_dtype", [np.int64, np.float64])
+    @pytest.mark.parametrize("kind", ["M", "m"])
+    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
+    def test_construction_to_datetimelike_unit(self, arr_dtype, kind, unit):
+        # tests all units
+        # gh-19223
+        # TODO: GH#19223 was about .astype, doesn't belong here
+        dtype = f"{kind}8[{unit}]"
+        arr = np.array([1, 2, 3], dtype=arr_dtype)
+        ser = Series(arr)
+        result = ser.astype(dtype)
+
+        expected = Series(arr.astype(dtype))
+
+        if unit in ["ns", "us", "ms", "s"]:
+            assert result.dtype == dtype
+            assert expected.dtype == dtype
+        else:
+            # Otherwise we cast to nearest-supported unit, i.e. seconds
+            assert result.dtype == f"{kind}8[s]"
+            assert expected.dtype == f"{kind}8[s]"
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("arg", ["2013-01-01 00:00:00", NaT, np.nan, None])
+    def test_constructor_with_naive_string_and_datetimetz_dtype(self, arg):
+        # GH 17415: With naive string
+        result = Series([arg], dtype="datetime64[ns, CET]")
+        expected = Series([Timestamp(arg)], dtype="M8[ns]").dt.tz_localize("CET")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_datetime64_bigendian(self):
+        # GH#30976
+        ms = np.datetime64(1, "ms")
+        arr = np.array([np.datetime64(1, "ms")], dtype=">M8[ms]")
+
+        result = Series(arr)
+        expected = Series([Timestamp(ms)]).astype("M8[ms]")
+        assert expected.dtype == "M8[ms]"
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("interval_constructor", [IntervalIndex, IntervalArray])
+    def test_construction_interval(self, interval_constructor):
+        # construction from interval & array of intervals
+        intervals = interval_constructor.from_breaks(np.arange(3), closed="right")
+        result = Series(intervals)
+        assert result.dtype == "interval[int64, right]"
+        tm.assert_index_equal(Index(result.values), Index(intervals))
+
+    @pytest.mark.parametrize(
+        "data_constructor", [list, np.array], ids=["list", "ndarray[object]"]
+    )
+    def test_constructor_infer_interval(self, data_constructor):
+        # GH 23563: consistent closed results in interval dtype
+        data = [Interval(0, 1), Interval(0, 2), None]
+        result = Series(data_constructor(data))
+        expected = Series(IntervalArray(data))
+        assert result.dtype == "interval[float64, right]"
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data_constructor", [list, np.array], ids=["list", "ndarray[object]"]
+    )
+    def test_constructor_interval_mixed_closed(self, data_constructor):
+        # GH 23563: mixed closed results in object dtype (not interval dtype)
+        data = [Interval(0, 1, closed="both"), Interval(0, 2, closed="neither")]
+        result = Series(data_constructor(data))
+        assert result.dtype == object
+        assert result.tolist() == data
+
+    def test_construction_consistency(self):
+        # make sure that we are not re-localizing upon construction
+        # GH 14928
+        ser = Series(date_range("20130101", periods=3, tz="US/Eastern"))
+
+        result = Series(ser, dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
+
+        result = Series(ser.dt.tz_convert("UTC"), dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
+
+        # Pre-2.0 dt64 values were treated as utc, which was inconsistent
+        #  with DatetimeIndex, which treats them as wall times, see GH#33401
+        result = Series(ser.values, dtype=ser.dtype)
+        expected = Series(ser.values).dt.tz_localize(ser.dtype.tz)
+        tm.assert_series_equal(result, expected)
+
+        with tm.assert_produces_warning(None):
+            # one suggested alternative to the deprecated (changed in 2.0) usage
+            middle = Series(ser.values).dt.tz_localize("UTC")
+            result = middle.dt.tz_convert(ser.dtype.tz)
+        tm.assert_series_equal(result, ser)
+
+        with tm.assert_produces_warning(None):
+            # the other suggested alternative to the deprecated usage
+            result = Series(ser.values.view("int64"), dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
+
+    @pytest.mark.parametrize(
+        "data_constructor", [list, np.array], ids=["list", "ndarray[object]"]
+    )
+    def test_constructor_infer_period(self, data_constructor):
+        data = [Period("2000", "D"), Period("2001", "D"), None]
+        result = Series(data_constructor(data))
+        expected = Series(period_array(data))
+        tm.assert_series_equal(result, expected)
+        assert result.dtype == "Period[D]"
+
+    @pytest.mark.xfail(reason="PeriodDtype Series not supported yet")
+    def test_construct_from_ints_including_iNaT_scalar_period_dtype(self):
+        series = Series([0, 1000, 2000, pd._libs.iNaT], dtype="period[D]")
+
+        val = series[3]
+        assert isna(val)
+
+        series[2] = val
+        assert isna(series[2])
+
+    def test_constructor_period_incompatible_frequency(self):
+        data = [Period("2000", "D"), Period("2001", "Y")]
+        result = Series(data)
+        assert result.dtype == object
+        assert result.tolist() == data
+
+    def test_constructor_periodindex(self):
+        # GH7932
+        # converting a PeriodIndex when put in a Series
+
+        pi = period_range("20130101", periods=5, freq="D")
+        s = Series(pi)
+        assert s.dtype == "Period[D]"
+        expected = Series(pi.astype(object))
+        assert expected.dtype == object
+
+    def test_constructor_dict(self):
+        d = {"a": 0.0, "b": 1.0, "c": 2.0}
+
+        result = Series(d)
+        expected = Series(d, index=sorted(d.keys()))
+        tm.assert_series_equal(result, expected)
+
+        result = Series(d, index=["b", "c", "d", "a"])
+        expected = Series([1, 2, np.nan, 0], index=["b", "c", "d", "a"])
+        tm.assert_series_equal(result, expected)
+
+        pidx = period_range("2020-01-01", periods=10, freq="D")
+        d = {pidx[0]: 0, pidx[1]: 1}
+        result = Series(d, index=pidx)
+        expected = Series(np.nan, pidx, dtype=np.float64)
+        expected.iloc[0] = 0
+        expected.iloc[1] = 1
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_list_value_explicit_dtype(self):
+        # GH 18625
+        d = {"a": [[2], [3], [4]]}
+        result = Series(d, index=["a"], dtype="object")
+        expected = Series(d, index=["a"])
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_order(self):
+        # GH19018
+        # initialization ordering: by insertion order
+        d = {"b": 1, "a": 0, "c": 2}
+        result = Series(d)
+        expected = Series([1, 0, 2], index=list("bac"))
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_extension(self, ea_scalar_and_dtype):
+        ea_scalar, ea_dtype = ea_scalar_and_dtype
+        d = {"a": ea_scalar}
+        result = Series(d, index=["a"])
+        expected = Series(ea_scalar, index=["a"], dtype=ea_dtype)
+
+        assert result.dtype == ea_dtype
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("value", [2, np.nan, None, float("nan")])
+    def test_constructor_dict_nan_key(self, value):
+        # GH 18480
+        d = {1: "a", value: "b", float("nan"): "c", 4: "d"}
+        result = Series(d).sort_values()
+        expected = Series(["a", "b", "c", "d"], index=[1, value, np.nan, 4])
+        tm.assert_series_equal(result, expected)
+
+        # MultiIndex:
+        d = {(1, 1): "a", (2, np.nan): "b", (3, value): "c"}
+        result = Series(d).sort_values()
+        expected = Series(
+            ["a", "b", "c"], index=Index([(1, 1), (2, np.nan), (3, value)])
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_datetime64_index(self):
+        # GH 9456
+
+        dates_as_str = ["1984-02-19", "1988-11-06", "1989-12-03", "1990-03-15"]
+        values = [42544017.198965244, 1234565, 40512335.181958228, -1]
+
+        def create_data(constructor):
+            return dict(zip((constructor(x) for x in dates_as_str), values))
+
+        data_datetime64 = create_data(np.datetime64)
+        data_datetime = create_data(lambda x: datetime.strptime(x, "%Y-%m-%d"))
+        data_Timestamp = create_data(Timestamp)
+
+        expected = Series(values, (Timestamp(x) for x in dates_as_str))
+
+        result_datetime64 = Series(data_datetime64)
+        result_datetime = Series(data_datetime)
+        result_Timestamp = Series(data_Timestamp)
+
+        tm.assert_series_equal(result_datetime64, expected)
+        tm.assert_series_equal(
+            result_datetime, expected.set_axis(expected.index.as_unit("us"))
+        )
+        tm.assert_series_equal(result_Timestamp, expected)
+
+    def test_constructor_dict_tuple_indexer(self):
+        # GH 12948
+        data = {(1, 1, None): -1.0}
+        result = Series(data)
+        expected = Series(
+            -1.0, index=MultiIndex(levels=[[1], [1], [np.nan]], codes=[[0], [0], [-1]])
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_mapping(self, non_dict_mapping_subclass):
+        # GH 29788
+        ndm = non_dict_mapping_subclass({3: "three"})
+        result = Series(ndm)
+        expected = Series(["three"], index=[3])
+
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_list_of_tuples(self):
+        data = [(1, 1), (2, 2), (2, 3)]
+        s = Series(data)
+        assert list(s) == data
+
+    def test_constructor_tuple_of_tuples(self):
+        data = ((1, 1), (2, 2), (2, 3))
+        s = Series(data)
+        assert tuple(s) == data
+
+    def test_constructor_dict_of_tuples(self):
+        data = {(1, 2): 3, (None, 5): 6}
+        result = Series(data).sort_values()
+        expected = Series([3, 6], index=MultiIndex.from_tuples([(1, 2), (None, 5)]))
+        tm.assert_series_equal(result, expected)
+
+    # https://github.com/pandas-dev/pandas/issues/22698
+    @pytest.mark.filterwarnings("ignore:elementwise comparison:FutureWarning")
+    def test_fromDict(self, using_infer_string):
+        data = {"a": 0, "b": 1, "c": 2, "d": 3}
+
+        series = Series(data)
+        tm.assert_is_sorted(series.index)
+
+        data = {"a": 0, "b": "1", "c": "2", "d": datetime.now()}
+        series = Series(data)
+        assert series.dtype == np.object_
+
+        data = {"a": 0, "b": "1", "c": "2", "d": "3"}
+        series = Series(data)
+        assert series.dtype == np.object_ if not using_infer_string else "str"
+
+        data = {"a": "0", "b": "1"}
+        series = Series(data, dtype=float)
+        assert series.dtype == np.float64
+
+    def test_fromValue(self, datetime_series, using_infer_string):
+        nans = Series(np.nan, index=datetime_series.index, dtype=np.float64)
+        assert nans.dtype == np.float64
+        assert len(nans) == len(datetime_series)
+
+        strings = Series("foo", index=datetime_series.index)
+        assert strings.dtype == np.object_ if not using_infer_string else "str"
+        assert len(strings) == len(datetime_series)
+
+        d = datetime.now()
+        dates = Series(d, index=datetime_series.index)
+        assert dates.dtype == "M8[us]"
+        assert len(dates) == len(datetime_series)
+
+        # GH12336
+        # Test construction of categorical series from value
+        categorical = Series(0, index=datetime_series.index, dtype="category")
+        expected = Series(0, index=datetime_series.index).astype("category")
+        assert categorical.dtype == "category"
+        assert len(categorical) == len(datetime_series)
+        tm.assert_series_equal(categorical, expected)
+
+    def test_constructor_dtype_timedelta64(self):
+        # basic
+        td = Series([timedelta(days=i) for i in range(3)])
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([timedelta(days=1)])
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([timedelta(days=1), timedelta(days=2), np.timedelta64(1, "s")])
+
+        assert td.dtype == "timedelta64[ns]"
+
+        # mixed with NaT
+        td = Series([timedelta(days=1), NaT], dtype="m8[ns]")
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([timedelta(days=1), np.nan], dtype="m8[ns]")
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([np.timedelta64(300000000), NaT], dtype="m8[ns]")
+        assert td.dtype == "timedelta64[ns]"
+
+        # improved inference
+        # GH5689
+        td = Series([np.timedelta64(300000000), NaT])
+        assert td.dtype == "timedelta64[ns]"
+
+        # because iNaT is int, not coerced to timedelta
+        td = Series([np.timedelta64(300000000), iNaT])
+        assert td.dtype == "object"
+
+        td = Series([np.timedelta64(300000000), np.nan])
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([NaT, np.timedelta64(300000000)])
+        assert td.dtype == "timedelta64[ns]"
+
+        td = Series([np.timedelta64(1, "s")])
+        assert td.dtype == "timedelta64[ns]"
+
+        # valid astype
+        td.astype("int64")
+
+        # invalid casting
+        msg = r"Converting from timedelta64\[ns\] to int32 is not supported"
+        with pytest.raises(TypeError, match=msg):
+            td.astype("int32")
+
+        # this is an invalid casting
+        msg = "|".join(
+            [
+                "Could not convert object to NumPy timedelta",
+                "Could not convert 'foo' to NumPy timedelta",
+            ]
+        )
+        with pytest.raises(ValueError, match=msg):
+            Series([timedelta(days=1), "foo"], dtype="m8[ns]")
+
+        # leave as object here
+        td = Series([timedelta(days=i) for i in range(3)] + ["foo"])
+        assert td.dtype == "object"
+
+        # as of 2.0, these no longer infer timedelta64 based on the strings,
+        #  matching Index behavior
+        ser = Series([None, NaT, "1 Day"])
+        assert ser.dtype == object
+
+        ser = Series([np.nan, NaT, "1 Day"])
+        assert ser.dtype == object
+
+        ser = Series([NaT, None, "1 Day"])
+        assert ser.dtype == object
+
+        ser = Series([NaT, np.nan, "1 Day"])
+        assert ser.dtype == object
+
+    # GH 16406
+    def test_constructor_mixed_tz(self):
+        s = Series([Timestamp("20130101"), Timestamp("20130101", tz="US/Eastern")])
+        expected = Series(
+            [Timestamp("20130101"), Timestamp("20130101", tz="US/Eastern")],
+            dtype="object",
+        )
+        tm.assert_series_equal(s, expected)
+
+    def test_NaT_scalar(self):
+        series = Series([0, 1000, 2000, iNaT], dtype="M8[ns]")
+
+        val = series[3]
+        assert isna(val)
+
+        series[2] = val
+        assert isna(series[2])
+
+    def test_NaT_cast(self):
+        # GH10747
+        result = Series([np.nan]).astype("M8[ns]")
+        expected = Series([NaT], dtype="M8[ns]")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_name_hashable(self):
+        for n in [777, 777.0, "name", datetime(2001, 11, 11), (1,), "\u05d0"]:
+            for data in [[1, 2, 3], np.ones(3), {"a": 0, "b": 1}]:
+                s = Series(data, name=n)
+                assert s.name == n
+
+    def test_constructor_name_unhashable(self):
+        msg = r"Series\.name must be a hashable type"
+        for n in [["name_list"], np.ones(2), {1: 2}]:
+            for data in [["name_list"], np.ones(2), {1: 2}]:
+                with pytest.raises(TypeError, match=msg):
+                    Series(data, name=n)
+
+    def test_auto_conversion(self):
+        series = Series(list(date_range("1/1/2000", periods=10)))
+        assert series.dtype == "M8[ns]"
+
+    def test_convert_non_ns(self):
+        # convert from a numpy array of non-ns timedelta64
+        arr = np.array([1, 2, 3], dtype="timedelta64[s]")
+        ser = Series(arr)
+        assert ser.dtype == arr.dtype
+
+        tdi = timedelta_range("00:00:01", periods=3, freq="s").as_unit("s")
+        expected = Series(tdi)
+        assert expected.dtype == arr.dtype
+        tm.assert_series_equal(ser, expected)
+
+        # convert from a numpy array of non-ns datetime64
+        arr = np.array(
+            ["2013-01-01", "2013-01-02", "2013-01-03"], dtype="datetime64[D]"
+        )
+        ser = Series(arr)
+        expected = Series(date_range("20130101", periods=3, freq="D"), dtype="M8[s]")
+        assert expected.dtype == "M8[s]"
+        tm.assert_series_equal(ser, expected)
+
+        arr = np.array(
+            ["2013-01-01 00:00:01", "2013-01-01 00:00:02", "2013-01-01 00:00:03"],
+            dtype="datetime64[s]",
+        )
+        ser = Series(arr)
+        expected = Series(
+            date_range("20130101 00:00:01", periods=3, freq="s"), dtype="M8[s]"
+        )
+        assert expected.dtype == "M8[s]"
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("1/1/2000", periods=10),
+            timedelta_range("1 day", periods=10),
+            period_range("2000-Q1", periods=10, freq="Q"),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_constructor_cant_cast_datetimelike(self, index):
+        # floats are not ok
+        # strip Index to convert PeriodIndex -> Period
+        # We don't care whether the error message says
+        # PeriodIndex or PeriodArray
+        msg = f"Cannot cast {type(index).__name__.rstrip('Index')}.*? to "
+
+        with pytest.raises(TypeError, match=msg):
+            Series(index, dtype=float)
+
+        # ints are ok
+        # we test with np.int64 to get similar results on
+        # windows / 32-bit platforms
+        result = Series(index, dtype=np.int64)
+        expected = Series(index.astype(np.int64))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "index",
+        [
+            date_range("1/1/2000", periods=10),
+            timedelta_range("1 day", periods=10),
+            period_range("2000-Q1", periods=10, freq="Q"),
+        ],
+        ids=lambda x: type(x).__name__,
+    )
+    def test_constructor_cast_object(self, index):
+        s = Series(index, dtype=object)
+        exp = Series(index).astype(object)
+        tm.assert_series_equal(s, exp)
+
+        s = Series(Index(index, dtype=object), dtype=object)
+        exp = Series(index).astype(object)
+        tm.assert_series_equal(s, exp)
+
+        s = Series(index.astype(object), dtype=object)
+        exp = Series(index).astype(object)
+        tm.assert_series_equal(s, exp)
+
+    @pytest.mark.parametrize("dtype", [np.datetime64, np.timedelta64])
+    def test_constructor_generic_timestamp_no_frequency(self, dtype, request):
+        # see gh-15524, gh-15987
+        msg = "dtype has no unit. Please pass in"
+
+        if np.dtype(dtype).name not in ["timedelta64", "datetime64"]:
+            mark = pytest.mark.xfail(reason="GH#33890 Is assigned ns unit")
+            request.applymarker(mark)
+
+        with pytest.raises(ValueError, match=msg):
+            Series([], dtype=dtype)
+
+    @pytest.mark.parametrize("unit", ["ps", "as", "fs", "Y", "M", "W", "D", "h", "m"])
+    @pytest.mark.parametrize("kind", ["m", "M"])
+    def test_constructor_generic_timestamp_bad_frequency(self, kind, unit):
+        # see gh-15524, gh-15987
+        # as of 2.0 we raise on any non-supported unit rather than silently
+        #  cast to nanos; previously we only raised for frequencies higher
+        #  than ns
+        dtype = f"{kind}8[{unit}]"
+
+        msg = "dtype=.* is not supported. Supported resolutions are"
+        with pytest.raises(TypeError, match=msg):
+            Series([], dtype=dtype)
+
+        with pytest.raises(TypeError, match=msg):
+            # pre-2.0 the DataFrame cast raised but the Series case did not
+            DataFrame([[0]], dtype=dtype)
+
+    @pytest.mark.parametrize("dtype", [None, "uint8", "category"])
+    def test_constructor_range_dtype(self, dtype):
+        # GH 16804
+        expected = Series([0, 1, 2, 3, 4], dtype=dtype or "int64")
+        result = Series(range(5), dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_range_overflows(self):
+        # GH#30173 range objects that overflow int64
+        rng = range(2**63, 2**63 + 4)
+        ser = Series(rng)
+        expected = Series(list(rng))
+        tm.assert_series_equal(ser, expected)
+        assert list(ser) == list(rng)
+        assert ser.dtype == np.uint64
+
+        rng2 = range(2**63 + 4, 2**63, -1)
+        ser2 = Series(rng2)
+        expected2 = Series(list(rng2))
+        tm.assert_series_equal(ser2, expected2)
+        assert list(ser2) == list(rng2)
+        assert ser2.dtype == np.uint64
+
+        rng3 = range(-(2**63), -(2**63) - 4, -1)
+        ser3 = Series(rng3)
+        expected3 = Series(list(rng3))
+        tm.assert_series_equal(ser3, expected3)
+        assert list(ser3) == list(rng3)
+        assert ser3.dtype == object
+
+        rng4 = range(2**73, 2**73 + 4)
+        ser4 = Series(rng4)
+        expected4 = Series(list(rng4))
+        tm.assert_series_equal(ser4, expected4)
+        assert list(ser4) == list(rng4)
+        assert ser4.dtype == object
+
+    def test_constructor_tz_mixed_data(self):
+        # GH 13051
+        dt_list = [
+            Timestamp("2016-05-01 02:03:37"),
+            Timestamp("2016-04-30 19:03:37-0700", tz="US/Pacific"),
+        ]
+        result = Series(dt_list)
+        expected = Series(dt_list, dtype=object)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("pydt", [True, False])
+    def test_constructor_data_aware_dtype_naive(self, tz_aware_fixture, pydt):
+        # GH#25843, GH#41555, GH#33401
+        tz = tz_aware_fixture
+        ts = Timestamp("2019", tz=tz)
+        if pydt:
+            ts = ts.to_pydatetime()
+
+        msg = (
+            "Cannot convert timezone-aware data to timezone-naive dtype. "
+            r"Use pd.Series\(values\).dt.tz_localize\(None\) instead."
+        )
+        with pytest.raises(ValueError, match=msg):
+            Series([ts], dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            Series(np.array([ts], dtype=object), dtype="datetime64[ns]")
+
+        with pytest.raises(ValueError, match=msg):
+            Series({0: ts}, dtype="datetime64[ns]")
+
+        msg = "Cannot unbox tzaware Timestamp to tznaive dtype"
+        with pytest.raises(TypeError, match=msg):
+            Series(ts, index=[0], dtype="datetime64[ns]")
+
+    def test_constructor_datetime64(self):
+        rng = date_range("1/1/2000 00:00:00", "1/1/2000 1:59:50", freq="10s")
+        dates = np.asarray(rng)
+
+        series = Series(dates)
+        assert np.issubdtype(series.dtype, np.dtype("M8[ns]"))
+
+    def test_constructor_datetimelike_scalar_to_string_dtype(
+        self, nullable_string_dtype
+    ):
+        # https://github.com/pandas-dev/pandas/pull/33846
+        result = Series("M", index=[1, 2, 3], dtype=nullable_string_dtype)
+        expected = Series(["M", "M", "M"], index=[1, 2, 3], dtype=nullable_string_dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("box", [lambda x: x, np.datetime64])
+    def test_constructor_sparse_datetime64(self, box):
+        # https://github.com/pandas-dev/pandas/issues/35762
+        values = [box("2012-01-01"), box("2013-01-01")]
+        dtype = pd.SparseDtype("datetime64[ns]")
+        result = Series(values, dtype=dtype)
+        arr = pd.arrays.SparseArray(values, dtype=dtype)
+        expected = Series(arr)
+        tm.assert_series_equal(result, expected)
+
+    def test_construction_from_ordered_collection(self):
+        # https://github.com/pandas-dev/pandas/issues/36044
+        result = Series({"a": 1, "b": 2}.keys())
+        expected = Series(["a", "b"])
+        tm.assert_series_equal(result, expected)
+
+        result = Series({"a": 1, "b": 2}.values())
+        expected = Series([1, 2])
+        tm.assert_series_equal(result, expected)
+
+    def test_construction_from_large_int_scalar_no_overflow(self):
+        # https://github.com/pandas-dev/pandas/issues/36291
+        n = 1_000_000_000_000_000_000_000
+        result = Series(n, index=[0])
+        expected = Series(n)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_list_of_periods_infers_period_dtype(self):
+        series = Series(list(period_range("2000-01-01", periods=10, freq="D")))
+        assert series.dtype == "Period[D]"
+
+        series = Series(
+            [Period("2011-01-01", freq="D"), Period("2011-02-01", freq="D")]
+        )
+        assert series.dtype == "Period[D]"
+
+    def test_constructor_subclass_dict(self, dict_subclass):
+        data = dict_subclass((x, 10.0 * x) for x in range(10))
+        series = Series(data)
+        expected = Series(dict(data.items()))
+        tm.assert_series_equal(series, expected)
+
+    def test_constructor_ordereddict(self):
+        # GH3283
+        data = OrderedDict(
+            (f"col{i}", np.random.default_rng(2).random()) for i in range(12)
+        )
+
+        series = Series(data)
+        expected = Series(list(data.values()), list(data.keys()))
+        tm.assert_series_equal(series, expected)
+
+        # Test with subclass
+        class A(OrderedDict):
+            pass
+
+        series = Series(A(data))
+        tm.assert_series_equal(series, expected)
+
+    def test_constructor_dict_multiindex(self):
+        d = {("a", "a"): 0.0, ("b", "a"): 1.0, ("b", "c"): 2.0}
+        _d = sorted(d.items())
+        result = Series(d)
+        expected = Series(
+            [x[1] for x in _d], index=MultiIndex.from_tuples([x[0] for x in _d])
+        )
+        tm.assert_series_equal(result, expected)
+
+        d["z"] = 111.0
+        _d.insert(0, ("z", d["z"]))
+        result = Series(d)
+        expected = Series(
+            [x[1] for x in _d], index=Index([x[0] for x in _d], tupleize_cols=False)
+        )
+        result = result.reindex(index=expected.index)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_multiindex_reindex_flat(self):
+        # construction involves reindexing with a MultiIndex corner case
+        data = {("i", "i"): 0, ("i", "j"): 1, ("j", "i"): 2, "j": np.nan}
+        expected = Series(data)
+
+        result = Series(expected[:-1].to_dict(), index=expected.index)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_dict_timedelta_index(self):
+        # GH #12169 : Resample category data with timedelta index
+        # construct Series from dict as data and TimedeltaIndex as index
+        # will result NaN in result Series data
+        expected = Series(
+            data=["A", "B", "C"], index=pd.to_timedelta([0, 10, 20], unit="s")
+        )
+
+        result = Series(
+            data={
+                pd.to_timedelta(0, unit="s"): "A",
+                pd.to_timedelta(10, unit="s"): "B",
+                pd.to_timedelta(20, unit="s"): "C",
+            },
+            index=pd.to_timedelta([0, 10, 20], unit="s"),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_infer_index_tz(self):
+        values = [188.5, 328.25]
+        tzinfo = tzoffset(None, 7200)
+        index = [
+            datetime(2012, 5, 11, 11, tzinfo=tzinfo),
+            datetime(2012, 5, 11, 12, tzinfo=tzinfo),
+        ]
+        series = Series(data=values, index=index)
+
+        assert series.index.tz == tzinfo
+
+        # it works! GH#2443
+        repr(series.index[0])
+
+    def test_constructor_with_pandas_dtype(self):
+        # going through 2D->1D path
+        vals = [(1,), (2,), (3,)]
+        ser = Series(vals)
+        dtype = ser.array.dtype  # NumpyEADtype
+        ser2 = Series(vals, dtype=dtype)
+        tm.assert_series_equal(ser, ser2)
+
+    def test_constructor_int_dtype_missing_values(self):
+        # GH#43017
+        result = Series(index=[0], dtype="int64")
+        expected = Series(np.nan, index=[0], dtype="float64")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_bool_dtype_missing_values(self):
+        # GH#43018
+        result = Series(index=[0], dtype="bool")
+        expected = Series(True, index=[0], dtype="bool")
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_int64_dtype(self, any_int_dtype):
+        # GH#44923
+        result = Series(["0", "1", "2"], dtype=any_int_dtype)
+        expected = Series([0, 1, 2], dtype=any_int_dtype)
+        tm.assert_series_equal(result, expected)
+
+    def test_constructor_raise_on_lossy_conversion_of_strings(self):
+        # GH#44923
+        if not np_version_gt2:
+            raises = pytest.raises(
+                ValueError, match="string values cannot be losslessly cast to int8"
+            )
+        else:
+            raises = pytest.raises(
+                OverflowError, match="The elements provided in the data"
+            )
+        with raises:
+            Series(["128"], dtype="int8")
+
+    def test_constructor_dtype_timedelta_alternative_construct(self):
+        # GH#35465
+        result = Series([1000000, 200000, 3000000], dtype="timedelta64[ns]")
+        expected = Series(pd.to_timedelta([1000000, 200000, 3000000], unit="ns"))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.xfail(
+        reason="Not clear what the correct expected behavior should be with "
+        "integers now that we support non-nano. ATM (2022-10-08) we treat ints "
+        "as nanoseconds, then cast to the requested dtype. xref #48312"
+    )
+    def test_constructor_dtype_timedelta_ns_s(self):
+        # GH#35465
+        result = Series([1000000, 200000, 3000000], dtype="timedelta64[ns]")
+        expected = Series([1000000, 200000, 3000000], dtype="timedelta64[s]")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.xfail(
+        reason="Not clear what the correct expected behavior should be with "
+        "integers now that we support non-nano. ATM (2022-10-08) we treat ints "
+        "as nanoseconds, then cast to the requested dtype. xref #48312"
+    )
+    def test_constructor_dtype_timedelta_ns_s_astype_int64(self):
+        # GH#35465
+        result = Series([1000000, 200000, 3000000], dtype="timedelta64[ns]").astype(
+            "int64"
+        )
+        expected = Series([1000000, 200000, 3000000], dtype="timedelta64[s]").astype(
+            "int64"
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.filterwarnings(
+        "ignore:elementwise comparison failed:DeprecationWarning"
+    )
+    @pytest.mark.parametrize("func", [Series, DataFrame, Index, pd.array])
+    def test_constructor_mismatched_null_nullable_dtype(
+        self, func, any_numeric_ea_dtype
+    ):
+        # GH#44514
+        msg = "|".join(
+            [
+                "cannot safely cast non-equivalent object",
+                r"int\(\) argument must be a string, a bytes-like object "
+                "or a (real )?number",
+                r"Cannot cast array data from dtype\('O'\) to dtype\('float64'\) "
+                "according to the rule 'safe'",
+                "object cannot be converted to a FloatingDtype",
+                "'values' contains non-numeric NA",
+            ]
+        )
+
+        for null in tm.NP_NAT_OBJECTS + [NaT]:
+            with pytest.raises(TypeError, match=msg):
+                func([null, 1.0, 3.0], dtype=any_numeric_ea_dtype)
+
+    def test_series_constructor_ea_int_from_bool(self):
+        # GH#42137
+        result = Series([True, False, True, pd.NA], dtype="Int64")
+        expected = Series([1, 0, 1, pd.NA], dtype="Int64")
+        tm.assert_series_equal(result, expected)
+
+        result = Series([True, False, True], dtype="Int64")
+        expected = Series([1, 0, 1], dtype="Int64")
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_ea_int_from_string_bool(self):
+        # GH#42137
+        with pytest.raises(ValueError, match="invalid literal"):
+            Series(["True", "False", "True", pd.NA], dtype="Int64")
+
+    @pytest.mark.parametrize("val", [1, 1.0])
+    def test_series_constructor_overflow_uint_ea(self, val):
+        # GH#38798
+        max_val = np.iinfo(np.uint64).max - 1
+        result = Series([max_val, val], dtype="UInt64")
+        expected = Series(np.array([max_val, 1], dtype="uint64"), dtype="UInt64")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("val", [1, 1.0])
+    def test_series_constructor_overflow_uint_ea_with_na(self, val):
+        # GH#38798
+        max_val = np.iinfo(np.uint64).max - 1
+        result = Series([max_val, val, pd.NA], dtype="UInt64")
+        expected = Series(
+            IntegerArray(
+                np.array([max_val, 1, 0], dtype="uint64"),
+                np.array([0, 0, 1], dtype=np.bool_),
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_overflow_uint_with_nan(self):
+        # GH#38798
+        max_val = np.iinfo(np.uint64).max - 1
+        result = Series([max_val, np.nan], dtype="UInt64")
+        expected = Series(
+            IntegerArray(
+                np.array([max_val, 1], dtype="uint64"),
+                np.array([0, 1], dtype=np.bool_),
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_ea_all_na(self):
+        # GH#38798
+        result = Series([np.nan, np.nan], dtype="UInt64")
+        expected = Series(
+            IntegerArray(
+                np.array([1, 1], dtype="uint64"),
+                np.array([1, 1], dtype=np.bool_),
+            )
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_series_from_index_dtype_equal_does_not_copy(self):
+        # GH#52008
+        idx = Index([1, 2, 3])
+        expected = idx.copy(deep=True)
+        ser = Series(idx, dtype="int64")
+        ser.iloc[0] = 100
+        tm.assert_index_equal(idx, expected)
+
+    def test_series_string_inference(self):
+        # GH#54430
+        with pd.option_context("future.infer_string", True):
+            ser = Series(["a", "b"])
+        dtype = pd.StringDtype("pyarrow" if HAS_PYARROW else "python", na_value=np.nan)
+        expected = Series(["a", "b"], dtype=dtype)
+        tm.assert_series_equal(ser, expected)
+
+        expected = Series(["a", 1], dtype="object")
+        with pd.option_context("future.infer_string", True):
+            ser = Series(["a", 1])
+        tm.assert_series_equal(ser, expected)
+
+    @pytest.mark.parametrize("na_value", [None, np.nan, pd.NA])
+    def test_series_string_with_na_inference(self, na_value):
+        # GH#54430
+        with pd.option_context("future.infer_string", True):
+            ser = Series(["a", na_value])
+        dtype = pd.StringDtype("pyarrow" if HAS_PYARROW else "python", na_value=np.nan)
+        expected = Series(["a", None], dtype=dtype)
+        tm.assert_series_equal(ser, expected)
+
+    def test_series_string_inference_scalar(self):
+        # GH#54430
+        with pd.option_context("future.infer_string", True):
+            ser = Series("a", index=[1])
+        dtype = pd.StringDtype("pyarrow" if HAS_PYARROW else "python", na_value=np.nan)
+        expected = Series("a", index=[1], dtype=dtype)
+        tm.assert_series_equal(ser, expected)
+
+    def test_series_string_inference_array_string_dtype(self):
+        # GH#54496
+        with pd.option_context("future.infer_string", True):
+            ser = Series(np.array(["a", "b"]))
+        dtype = pd.StringDtype("pyarrow" if HAS_PYARROW else "python", na_value=np.nan)
+        expected = Series(["a", "b"], dtype=dtype)
+        tm.assert_series_equal(ser, expected)
+
+    def test_series_string_inference_storage_definition(self):
+        # https://github.com/pandas-dev/pandas/issues/54793
+        # but after PDEP-14 (string dtype), it was decided to keep dtype="string"
+        # returning the NA string dtype, so expected is changed from
+        # "string[pyarrow_numpy]" to "string[python]"
+        expected = Series(["a", "b"], dtype="string[python]")
+        with pd.option_context("future.infer_string", True):
+            result = Series(["a", "b"], dtype="string")
+        tm.assert_series_equal(result, expected)
+
+        expected = Series(["a", "b"], dtype=pd.StringDtype(na_value=np.nan))
+        with pd.option_context("future.infer_string", True):
+            result = Series(["a", "b"], dtype="str")
+        tm.assert_series_equal(result, expected)
+
+    def test_series_constructor_infer_string_scalar(self):
+        # GH#55537
+        with pd.option_context("future.infer_string", True):
+            ser = Series("a", index=[1, 2], dtype="string[python]")
+        expected = Series(["a", "a"], index=[1, 2], dtype="string[python]")
+        tm.assert_series_equal(ser, expected)
+        assert ser.dtype.storage == "python"
+
+    def test_series_string_inference_na_first(self):
+        # GH#55655
+        with pd.option_context("future.infer_string", True):
+            result = Series([pd.NA, "b"])
+        dtype = pd.StringDtype("pyarrow" if HAS_PYARROW else "python", na_value=np.nan)
+        expected = Series([None, "b"], dtype=dtype)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("klass", [Series, Index])
+    def test_inference_on_pandas_objects(self, klass):
+        # GH#56012
+        obj = klass([Timestamp("2019-12-31")], dtype=object)
+        # This doesn't do inference
+        result = Series(obj)
+        assert result.dtype == np.object_
+
+
+class TestSeriesConstructorIndexCoercion:
+    def test_series_constructor_datetimelike_index_coercion(self):
+        idx = date_range("2020-01-01", periods=5)
+        ser = Series(
+            np.random.default_rng(2).standard_normal(len(idx)), idx.astype(object)
+        )
+        # as of 2.0, we no longer silently cast the object-dtype index
+        #  to DatetimeIndex GH#39307, GH#23598
+        assert not isinstance(ser.index, DatetimeIndex)
+
+    @pytest.mark.parametrize("container", [None, np.array, Series, Index])
+    @pytest.mark.parametrize("data", [1.0, range(4)])
+    def test_series_constructor_infer_multiindex(self, container, data):
+        indexes = [["a", "a", "b", "b"], ["x", "y", "x", "y"]]
+        if container is not None:
+            indexes = [container(ind) for ind in indexes]
+
+        multi = Series(data, index=indexes)
+        assert isinstance(multi.index, MultiIndex)
+
+    # TODO: make this not cast to object in pandas 3.0
+    @pytest.mark.skipif(
+        not np_version_gt2, reason="StringDType only available in numpy 2 and above"
+    )
+    @pytest.mark.parametrize(
+        "data",
+        [
+            ["a", "b", "c"],
+            ["a", "b", np.nan],
+        ],
+    )
+    def test_np_string_array_object_cast(self, data):
+        from numpy.dtypes import StringDType
+
+        arr = np.array(data, dtype=StringDType())
+        res = Series(arr)
+        assert res.dtype == np.object_
+        assert (res == data).all()
+
+
+class TestSeriesConstructorInternals:
+    def test_constructor_no_pandas_array(self):
+        ser = Series([1, 2, 3])
+        result = Series(ser.array)
+        tm.assert_series_equal(ser, result)
+        assert isinstance(result._mgr.blocks[0], NumpyBlock)
+        assert result._mgr.blocks[0].is_numeric
+
+    def test_from_array(self):
+        result = Series(pd.array(["1h", "2h"], dtype="timedelta64[ns]"))
+        assert result._mgr.blocks[0].is_extension is False
+
+        result = Series(pd.array(["2015"], dtype="datetime64[ns]"))
+        assert result._mgr.blocks[0].is_extension is False
+
+    def test_from_list_dtype(self):
+        result = Series(["1h", "2h"], dtype="timedelta64[ns]")
+        assert result._mgr.blocks[0].is_extension is False
+
+        result = Series(["2015"], dtype="datetime64[ns]")
+        assert result._mgr.blocks[0].is_extension is False
+
+
+def test_constructor(rand_series_with_duplicate_datetimeindex):
+    dups = rand_series_with_duplicate_datetimeindex
+    assert isinstance(dups, Series)
+    assert isinstance(dups.index, DatetimeIndex)
+
+
+@pytest.mark.parametrize(
+    "input_dict,expected",
+    [
+        ({0: 0}, np.array([[0]], dtype=np.int64)),
+        ({"a": "a"}, np.array([["a"]], dtype=object)),
+        ({1: 1}, np.array([[1]], dtype=np.int64)),
+    ],
+)
+def test_numpy_array(input_dict, expected):
+    result = np.array([Series(input_dict)])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_index_ordered_dict_keys():
+    # GH 22077
+
+    param_index = OrderedDict(
+        [
+            ((("a", "b"), ("c", "d")), 1),
+            ((("a", None), ("c", "d")), 2),
+        ]
+    )
+    series = Series([1, 2], index=param_index.keys())
+    expected = Series(
+        [1, 2],
+        index=MultiIndex.from_tuples(
+            [(("a", "b"), ("c", "d")), (("a", None), ("c", "d"))]
+        ),
+    )
+    tm.assert_series_equal(series, expected)
+
+
+@pytest.mark.parametrize(
+    "input_list",
+    [
+        [1, complex("nan"), 2],
+        [1 + 1j, complex("nan"), 2 + 2j],
+    ],
+)
+def test_series_with_complex_nan(input_list):
+    # GH#53627
+    ser = Series(input_list)
+    result = Series(ser.array)
+    assert ser.dtype == "complex128"
+    tm.assert_series_equal(ser, result)
+
+
+def test_dict_keys_rangeindex():
+    result = Series({0: 1, 1: 2})
+    expected = Series([1, 2], index=RangeIndex(2))
+    tm.assert_series_equal(result, expected, check_index_type=True)
diff --git a/pandas/pandas/tests/series/test_cumulative.py b/pandas/pandas/tests/series/test_cumulative.py
new file mode 100644
index 0000000000000..db83cf1112e74
--- /dev/null
+++ b/pandas/pandas/tests/series/test_cumulative.py
@@ -0,0 +1,284 @@
+"""
+Tests for Series cumulative operations.
+
+See also
+--------
+tests.frame.test_cumulative
+"""
+
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+methods = {
+    "cumsum": np.cumsum,
+    "cumprod": np.cumprod,
+    "cummin": np.minimum.accumulate,
+    "cummax": np.maximum.accumulate,
+}
+
+
+class TestSeriesCumulativeOps:
+    @pytest.mark.parametrize("func", [np.cumsum, np.cumprod])
+    def test_datetime_series(self, datetime_series, func):
+        tm.assert_numpy_array_equal(
+            func(datetime_series).values,
+            func(np.array(datetime_series)),
+            check_dtype=True,
+        )
+
+        # with missing values
+        ts = datetime_series.copy()
+        ts[::2] = np.nan
+
+        result = func(ts)[1::2]
+        expected = func(np.array(ts.dropna()))
+
+        tm.assert_numpy_array_equal(result.values, expected, check_dtype=False)
+
+    @pytest.mark.parametrize("method", ["cummin", "cummax"])
+    def test_cummin_cummax(self, datetime_series, method):
+        ufunc = methods[method]
+
+        result = getattr(datetime_series, method)().values
+        expected = ufunc(np.array(datetime_series))
+
+        tm.assert_numpy_array_equal(result, expected)
+        ts = datetime_series.copy()
+        ts[::2] = np.nan
+        result = getattr(ts, method)()[1::2]
+        expected = ufunc(ts.dropna())
+
+        result.index = result.index._with_freq(None)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ts",
+        [
+            pd.Timedelta(0),
+            pd.Timestamp("1999-12-31"),
+            pd.Timestamp("1999-12-31").tz_localize("US/Pacific"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "method, skipna, exp_tdi",
+        [
+            ["cummax", True, ["NaT", "2 days", "NaT", "2 days", "NaT", "3 days"]],
+            ["cummin", True, ["NaT", "2 days", "NaT", "1 days", "NaT", "1 days"]],
+            [
+                "cummax",
+                False,
+                ["NaT", "NaT", "NaT", "NaT", "NaT", "NaT"],
+            ],
+            [
+                "cummin",
+                False,
+                ["NaT", "NaT", "NaT", "NaT", "NaT", "NaT"],
+            ],
+        ],
+    )
+    def test_cummin_cummax_datetimelike(self, ts, method, skipna, exp_tdi):
+        # with ts==pd.Timedelta(0), we are testing td64; with naive Timestamp
+        #  we are testing datetime64[ns]; with Timestamp[US/Pacific]
+        #  we are testing dt64tz
+        tdi = pd.to_timedelta(["NaT", "2 days", "NaT", "1 days", "NaT", "3 days"])
+        ser = pd.Series(tdi + ts)
+
+        exp_tdi = pd.to_timedelta(exp_tdi)
+        expected = pd.Series(exp_tdi + ts)
+        result = getattr(ser, method)(skipna=skipna)
+        tm.assert_series_equal(expected, result)
+
+    def test_cumsum_datetimelike(self):
+        # GH#57956
+        df = pd.DataFrame(
+            [
+                [pd.Timedelta(0), pd.Timedelta(days=1)],
+                [pd.Timedelta(days=2), pd.NaT],
+                [pd.Timedelta(hours=-6), pd.Timedelta(hours=12)],
+            ]
+        )
+        result = df.cumsum()
+        expected = pd.DataFrame(
+            [
+                [pd.Timedelta(0), pd.Timedelta(days=1)],
+                [pd.Timedelta(days=2), pd.NaT],
+                [pd.Timedelta(days=1, hours=18), pd.Timedelta(days=1, hours=12)],
+            ]
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "func, exp",
+        [
+            ("cummin", "2012-1-1"),
+            ("cummax", "2012-1-2"),
+        ],
+    )
+    def test_cummin_cummax_period(self, func, exp):
+        # GH#28385
+        ser = pd.Series(
+            [pd.Period("2012-1-1", freq="D"), pd.NaT, pd.Period("2012-1-2", freq="D")]
+        )
+        result = getattr(ser, func)(skipna=False)
+        expected = pd.Series([pd.Period("2012-1-1", freq="D"), pd.NaT, pd.NaT])
+        tm.assert_series_equal(result, expected)
+
+        result = getattr(ser, func)(skipna=True)
+        exp = pd.Period(exp, freq="D")
+        expected = pd.Series([pd.Period("2012-1-1", freq="D"), pd.NaT, exp])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arg",
+        [
+            [False, False, False, True, True, False, False],
+            [False, False, False, False, False, False, False],
+        ],
+    )
+    @pytest.mark.parametrize(
+        "func", [lambda x: x, lambda x: ~x], ids=["identity", "inverse"]
+    )
+    @pytest.mark.parametrize("method", methods.keys())
+    def test_cummethods_bool(self, arg, func, method):
+        # GH#6270
+        # checking Series method vs the ufunc applied to the values
+
+        ser = func(pd.Series(arg))
+        ufunc = methods[method]
+
+        exp_vals = ufunc(ser.values)
+        expected = pd.Series(exp_vals)
+
+        result = getattr(ser, method)()
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "method, expected",
+        [
+            ["cumsum", pd.Series([0, 1, np.nan, 1], dtype=object)],
+            ["cumprod", pd.Series([False, 0, np.nan, 0])],
+            ["cummin", pd.Series([False, False, np.nan, False])],
+            ["cummax", pd.Series([False, True, np.nan, True])],
+        ],
+    )
+    def test_cummethods_bool_in_object_dtype(self, method, expected):
+        ser = pd.Series([False, True, np.nan, False])
+        result = getattr(ser, method)()
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "method, order",
+        [
+            ["cummax", "abc"],
+            ["cummin", "cba"],
+        ],
+    )
+    def test_cummax_cummin_on_ordered_categorical(self, method, order):
+        # GH#52335
+        cat = pd.CategoricalDtype(list(order), ordered=True)
+        ser = pd.Series(
+            list("ababcab"),
+            dtype=cat,
+        )
+        result = getattr(ser, method)()
+        expected = pd.Series(
+            list("abbbccc"),
+            dtype=cat,
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "skip, exp",
+        [
+            [True, ["a", np.nan, "b", "b", "c"]],
+            [False, ["a", np.nan, np.nan, np.nan, np.nan]],
+        ],
+    )
+    @pytest.mark.parametrize(
+        "method, order",
+        [
+            ["cummax", "abc"],
+            ["cummin", "cba"],
+        ],
+    )
+    def test_cummax_cummin_ordered_categorical_nan(self, skip, exp, method, order):
+        # GH#52335
+        cat = pd.CategoricalDtype(list(order), ordered=True)
+        ser = pd.Series(
+            ["a", np.nan, "b", "a", "c"],
+            dtype=cat,
+        )
+        result = getattr(ser, method)(skipna=skip)
+        expected = pd.Series(
+            exp,
+            dtype=cat,
+        )
+        tm.assert_series_equal(
+            result,
+            expected,
+        )
+
+    def test_cumprod_timedelta(self):
+        # GH#48111
+        ser = pd.Series([pd.Timedelta(days=1), pd.Timedelta(days=3)])
+        with pytest.raises(TypeError, match="cumprod not supported for Timedelta"):
+            ser.cumprod()
+
+    @pytest.mark.parametrize(
+        "data, op, skipna, expected_data",
+        [
+            ([], "cumsum", True, []),
+            ([], "cumsum", False, []),
+            (["x", "z", "y"], "cumsum", True, ["x", "xz", "xzy"]),
+            (["x", "z", "y"], "cumsum", False, ["x", "xz", "xzy"]),
+            (["x", pd.NA, "y"], "cumsum", True, ["x", pd.NA, "xy"]),
+            (["x", pd.NA, "y"], "cumsum", False, ["x", pd.NA, pd.NA]),
+            ([pd.NA, "x", "y"], "cumsum", True, [pd.NA, "x", "xy"]),
+            ([pd.NA, "x", "y"], "cumsum", False, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cumsum", True, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cumsum", False, [pd.NA, pd.NA, pd.NA]),
+            ([], "cummin", True, []),
+            ([], "cummin", False, []),
+            (["y", "z", "x"], "cummin", True, ["y", "y", "x"]),
+            (["y", "z", "x"], "cummin", False, ["y", "y", "x"]),
+            (["y", pd.NA, "x"], "cummin", True, ["y", pd.NA, "x"]),
+            (["y", pd.NA, "x"], "cummin", False, ["y", pd.NA, pd.NA]),
+            ([pd.NA, "y", "x"], "cummin", True, [pd.NA, "y", "x"]),
+            ([pd.NA, "y", "x"], "cummin", False, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cummin", True, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cummin", False, [pd.NA, pd.NA, pd.NA]),
+            ([], "cummax", True, []),
+            ([], "cummax", False, []),
+            (["x", "z", "y"], "cummax", True, ["x", "z", "z"]),
+            (["x", "z", "y"], "cummax", False, ["x", "z", "z"]),
+            (["x", pd.NA, "y"], "cummax", True, ["x", pd.NA, "y"]),
+            (["x", pd.NA, "y"], "cummax", False, ["x", pd.NA, pd.NA]),
+            ([pd.NA, "x", "y"], "cummax", True, [pd.NA, "x", "y"]),
+            ([pd.NA, "x", "y"], "cummax", False, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cummax", True, [pd.NA, pd.NA, pd.NA]),
+            ([pd.NA, pd.NA, pd.NA], "cummax", False, [pd.NA, pd.NA, pd.NA]),
+        ],
+    )
+    def test_cum_methods_ea_strings(
+        self, string_dtype_no_object, data, op, skipna, expected_data
+    ):
+        # https://github.com/pandas-dev/pandas/pull/60633 - pyarrow
+        # https://github.com/pandas-dev/pandas/pull/60938 - Python
+        ser = pd.Series(data, dtype=string_dtype_no_object)
+        method = getattr(ser, op)
+        expected = pd.Series(expected_data, dtype=string_dtype_no_object)
+        result = method(skipna=skipna)
+        tm.assert_series_equal(result, expected)
+
+    def test_cumprod_pyarrow_strings(self, pyarrow_string_dtype, skipna):
+        # https://github.com/pandas-dev/pandas/pull/60633
+        ser = pd.Series(list("xyz"), dtype=pyarrow_string_dtype)
+        msg = re.escape(f"operation 'cumprod' not supported for dtype '{ser.dtype}'")
+        with pytest.raises(TypeError, match=msg):
+            ser.cumprod(skipna=skipna)
diff --git a/pandas/pandas/tests/series/test_formats.py b/pandas/pandas/tests/series/test_formats.py
new file mode 100644
index 0000000000000..eb81840f6f8f9
--- /dev/null
+++ b/pandas/pandas/tests/series/test_formats.py
@@ -0,0 +1,568 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Index,
+    Series,
+    date_range,
+    option_context,
+    period_range,
+    timedelta_range,
+)
+
+
+class TestSeriesRepr:
+    def test_multilevel_name_print_0(self):
+        # GH#55415 None does not get printed, but 0 does
+        # (matching DataFrame and flat index behavior)
+        mi = pd.MultiIndex.from_product([range(2, 3), range(3, 4)], names=[0, None])
+        ser = Series(1.5, index=mi)
+
+        res = repr(ser)
+        expected = "0   \n2  3    1.5\ndtype: float64"
+        assert res == expected
+
+    def test_multilevel_name_print(self, lexsorted_two_level_string_multiindex):
+        index = lexsorted_two_level_string_multiindex
+        ser = Series(range(len(index)), index=index, name="sth")
+        expected = [
+            "first  second",
+            "foo    one       0",
+            "       two       1",
+            "       three     2",
+            "bar    one       3",
+            "       two       4",
+            "baz    two       5",
+            "       three     6",
+            "qux    one       7",
+            "       two       8",
+            "       three     9",
+            "Name: sth, dtype: int64",
+        ]
+        expected = "\n".join(expected)
+        assert repr(ser) == expected
+
+    def test_small_name_printing(self):
+        # Test small Series.
+        s = Series([0, 1, 2])
+
+        s.name = "test"
+        assert "Name: test" in repr(s)
+
+        s.name = None
+        assert "Name:" not in repr(s)
+
+    def test_big_name_printing(self):
+        # Test big Series (diff code path).
+        s = Series(range(1000))
+
+        s.name = "test"
+        assert "Name: test" in repr(s)
+
+        s.name = None
+        assert "Name:" not in repr(s)
+
+    def test_empty_name_printing(self):
+        s = Series(index=date_range("20010101", "20020101"), name="test", dtype=object)
+        assert "Name: test" in repr(s)
+
+    @pytest.mark.parametrize("args", [(), (0, -1)])
+    def test_float_range(self, args):
+        str(
+            Series(
+                np.random.default_rng(2).standard_normal(1000),
+                index=np.arange(1000, *args),
+            )
+        )
+
+    def test_empty_object(self):
+        # empty
+        str(Series(dtype=object))
+
+    def test_string(self, string_series):
+        str(string_series)
+        str(string_series.astype(int))
+
+        # with NaNs
+        string_series[5:7] = np.nan
+        str(string_series)
+
+    def test_object(self, object_series):
+        str(object_series)
+
+    def test_datetime(self, datetime_series):
+        str(datetime_series)
+        # with Nones
+        ots = datetime_series.astype("O")
+        ots[::2] = None
+        repr(ots)
+
+    @pytest.mark.parametrize(
+        "name",
+        [
+            "",
+            1,
+            1.2,
+            "foo",
+            "\u03b1\u03b2\u03b3",
+            "loooooooooooooooooooooooooooooooooooooooooooooooooooong",
+            ("foo", "bar", "baz"),
+            (1, 2),
+            ("foo", 1, 2.3),
+            ("\u03b1", "\u03b2", "\u03b3"),
+            ("\u03b1", "bar"),
+        ],
+    )
+    def test_various_names(self, name, string_series):
+        # various names
+        string_series.name = name
+        repr(string_series)
+
+    def test_tuple_name(self):
+        biggie = Series(
+            np.random.default_rng(2).standard_normal(1000),
+            index=np.arange(1000),
+            name=("foo", "bar", "baz"),
+        )
+        repr(biggie)
+
+    @pytest.mark.parametrize("arg", [100, 1001])
+    def test_tidy_repr_name_0(self, arg):
+        # tidy repr
+        ser = Series(np.random.default_rng(2).standard_normal(arg), name=0)
+        rep_str = repr(ser)
+        assert "Name: 0" in rep_str
+
+    def test_newline(self, any_string_dtype):
+        ser = Series(
+            ["a\n\r\tb"],
+            name="a\n\r\td",
+            index=Index(["a\n\r\tf"], dtype=any_string_dtype),
+            dtype=any_string_dtype,
+        )
+        assert "\t" not in repr(ser)
+        assert "\r" not in repr(ser)
+        assert "a\n" not in repr(ser)
+
+    @pytest.mark.parametrize(
+        "name, expected",
+        [
+            ["foo", "Series([], Name: foo, dtype: int64)"],
+            [None, "Series([], dtype: int64)"],
+        ],
+    )
+    def test_empty_int64(self, name, expected):
+        # with empty series (#4651)
+        s = Series([], dtype=np.int64, name=name)
+        assert repr(s) == expected
+
+    def test_repr_bool_fails(self, capsys):
+        s = Series(
+            [
+                DataFrame(np.random.default_rng(2).standard_normal((2, 2)))
+                for i in range(5)
+            ]
+        )
+
+        # It works (with no Cython exception barf)!
+        repr(s)
+
+        captured = capsys.readouterr()
+        assert captured.err == ""
+
+    def test_repr_name_iterable_indexable(self):
+        s = Series([1, 2, 3], name=np.int64(3))
+
+        # it works!
+        repr(s)
+
+        s.name = ("\u05d0",) * 2
+        repr(s)
+
+    def test_repr_max_rows(self):
+        # GH 6863
+        with option_context("display.max_rows", None):
+            str(Series(range(1001)))  # should not raise exception
+
+    def test_unicode_string_with_unicode(self):
+        df = Series(["\u05d0"], name="\u05d1")
+        str(df)
+
+        ser = Series(["\u03c3"] * 10)
+        repr(ser)
+
+        ser2 = Series(["\u05d0"] * 1000)
+        ser2.name = "title1"
+        repr(ser2)
+
+    def test_str_to_bytes_raises(self):
+        # GH 26447
+        df = Series(["abc"], name="abc")
+        msg = "^'str' object cannot be interpreted as an integer$"
+        with pytest.raises(TypeError, match=msg):
+            bytes(df)
+
+    def test_timeseries_repr_object_dtype(self):
+        index = Index(
+            [datetime(2000, 1, 1) + timedelta(i) for i in range(1000)], dtype=object
+        )
+        ts = Series(np.random.default_rng(2).standard_normal(len(index)), index)
+        repr(ts)
+
+        ts = Series(
+            np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+        )
+        assert repr(ts).splitlines()[-1].startswith("Freq:")
+
+        ts2 = ts.iloc[np.random.default_rng(2).integers(0, len(ts) - 1, 400)]
+        repr(ts2).splitlines()[-1]
+
+    def test_latex_repr(self):
+        pytest.importorskip("jinja2")  # uses Styler implementation
+        result = r"""\begin{tabular}{ll}
+\toprule
+ & 0 \\
+\midrule
+0 & $\alpha$ \\
+1 & b \\
+2 & c \\
+\bottomrule
+\end{tabular}
+"""
+        with option_context(
+            "styler.format.escape", None, "styler.render.repr", "latex"
+        ):
+            s = Series([r"$\alpha$", "b", "c"])
+            assert result == s._repr_latex_()
+
+        assert s._repr_latex_() is None
+
+    def test_index_repr_in_frame_with_nan(self):
+        # see gh-25061
+        i = Index([1, np.nan])
+        s = Series([1, 2], index=i)
+        exp = """1.0    1\nNaN    2\ndtype: int64"""
+
+        assert repr(s) == exp
+
+    def test_series_repr_nat(self):
+        series = Series([0, 1000, 2000, pd.NaT._value], dtype="M8[ns]")
+
+        result = repr(series)
+        expected = (
+            "0   1970-01-01 00:00:00.000000\n"
+            "1   1970-01-01 00:00:00.000001\n"
+            "2   1970-01-01 00:00:00.000002\n"
+            "3                          NaT\n"
+            "dtype: datetime64[ns]"
+        )
+        assert result == expected
+
+    def test_float_repr(self):
+        # GH#35603
+        # check float format when cast to object
+        ser = Series([1.0]).astype(object)
+        expected = "0    1.0\ndtype: object"
+        assert repr(ser) == expected
+
+    def test_different_null_objects(self):
+        # GH#45263
+        ser = Series([1, 2, 3, 4], [True, None, np.nan, pd.NaT])
+        result = repr(ser)
+        expected = "True    1\nNone    2\nNaN     3\nNaT     4\ndtype: int64"
+        assert result == expected
+
+
+class TestCategoricalRepr:
+    def test_categorical_repr_unicode(self):
+        # see gh-21002
+
+        class County:
+            name = "San Sebastián"
+            state = "PR"
+
+            def __repr__(self) -> str:
+                return self.name + ", " + self.state
+
+        cat = Categorical([County() for _ in range(61)])
+        idx = Index(cat)
+        ser = idx.to_series()
+
+        repr(ser)
+        str(ser)
+
+    def test_categorical_repr(self, using_infer_string):
+        a = Series(Categorical([1, 2, 3, 4]))
+        exp = (
+            "0    1\n1    2\n2    3\n3    4\n"
+            "dtype: category\nCategories (4, int64): [1, 2, 3, 4]"
+        )
+
+        assert exp == a.__str__()
+
+        a = Series(Categorical(["a", "b"] * 25))
+        if using_infer_string:
+            exp = (
+                "0     a\n1     b\n"
+                "     ..\n"
+                "48    a\n49    b\n"
+                "Length: 50, dtype: category\nCategories (2, str): [a, b]"
+            )
+        else:
+            exp = (
+                "0     a\n1     b\n"
+                "     ..\n"
+                "48    a\n49    b\n"
+                "Length: 50, dtype: category\nCategories (2, object): ['a', 'b']"
+            )
+        with option_context("display.max_rows", 5):
+            assert exp == repr(a)
+
+        levs = list("abcdefghijklmnopqrstuvwxyz")
+        a = Series(Categorical(["a", "b"], categories=levs, ordered=True))
+        if using_infer_string:
+            exp = (
+                "0    a\n1    b\n"
+                "dtype: category\n"
+                "Categories (26, str): [a < b < c < d ... w < x < y < z]"
+            )
+        else:
+            exp = (
+                "0    a\n1    b\n"
+                "dtype: category\n"
+                "Categories (26, object): ['a' < 'b' < 'c' < 'd' ... "
+                "'w' < 'x' < 'y' < 'z']"
+            )
+        assert exp == a.__str__()
+
+    def test_categorical_series_repr(self):
+        s = Series(Categorical([1, 2, 3]))
+        exp = """0    1
+1    2
+2    3
+dtype: category
+Categories (3, int64): [1, 2, 3]"""
+
+        assert repr(s) == exp
+
+        s = Series(Categorical(np.arange(10)))
+        exp = f"""0    0
+1    1
+2    2
+3    3
+4    4
+5    5
+6    6
+7    7
+8    8
+9    9
+dtype: category
+Categories (10, {np.dtype(int)}): [0, 1, 2, 3, ..., 6, 7, 8, 9]"""
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_ordered(self):
+        s = Series(Categorical([1, 2, 3], ordered=True))
+        exp = """0    1
+1    2
+2    3
+dtype: category
+Categories (3, int64): [1 < 2 < 3]"""
+
+        assert repr(s) == exp
+
+        s = Series(Categorical(np.arange(10), ordered=True))
+        exp = f"""0    0
+1    1
+2    2
+3    3
+4    4
+5    5
+6    6
+7    7
+8    8
+9    9
+dtype: category
+Categories (10, {np.dtype(int)}): [0 < 1 < 2 < 3 ... 6 < 7 < 8 < 9]"""
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_datetime(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        s = Series(Categorical(idx))
+        exp = """0   2011-01-01 09:00:00
+1   2011-01-01 10:00:00
+2   2011-01-01 11:00:00
+3   2011-01-01 12:00:00
+4   2011-01-01 13:00:00
+dtype: category
+Categories (5, datetime64[ns]): [2011-01-01 09:00:00, 2011-01-01 10:00:00, 2011-01-01 11:00:00,
+                                 2011-01-01 12:00:00, 2011-01-01 13:00:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        s = Series(Categorical(idx))
+        exp = """0   2011-01-01 09:00:00-05:00
+1   2011-01-01 10:00:00-05:00
+2   2011-01-01 11:00:00-05:00
+3   2011-01-01 12:00:00-05:00
+4   2011-01-01 13:00:00-05:00
+dtype: category
+Categories (5, datetime64[ns, US/Eastern]): [2011-01-01 09:00:00-05:00, 2011-01-01 10:00:00-05:00,
+                                             2011-01-01 11:00:00-05:00, 2011-01-01 12:00:00-05:00,
+                                             2011-01-01 13:00:00-05:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_datetime_ordered(self):
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5)
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0   2011-01-01 09:00:00
+1   2011-01-01 10:00:00
+2   2011-01-01 11:00:00
+3   2011-01-01 12:00:00
+4   2011-01-01 13:00:00
+dtype: category
+Categories (5, datetime64[ns]): [2011-01-01 09:00:00 < 2011-01-01 10:00:00 < 2011-01-01 11:00:00 <
+                                 2011-01-01 12:00:00 < 2011-01-01 13:00:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+        idx = date_range("2011-01-01 09:00", freq="h", periods=5, tz="US/Eastern")
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0   2011-01-01 09:00:00-05:00
+1   2011-01-01 10:00:00-05:00
+2   2011-01-01 11:00:00-05:00
+3   2011-01-01 12:00:00-05:00
+4   2011-01-01 13:00:00-05:00
+dtype: category
+Categories (5, datetime64[ns, US/Eastern]): [2011-01-01 09:00:00-05:00 < 2011-01-01 10:00:00-05:00 <
+                                             2011-01-01 11:00:00-05:00 < 2011-01-01 12:00:00-05:00 <
+                                             2011-01-01 13:00:00-05:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_period(self):
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        s = Series(Categorical(idx))
+        exp = """0    2011-01-01 09:00
+1    2011-01-01 10:00
+2    2011-01-01 11:00
+3    2011-01-01 12:00
+4    2011-01-01 13:00
+dtype: category
+Categories (5, period[h]): [2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00,
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        s = Series(Categorical(idx))
+        exp = """0    2011-01
+1    2011-02
+2    2011-03
+3    2011-04
+4    2011-05
+dtype: category
+Categories (5, period[M]): [2011-01, 2011-02, 2011-03, 2011-04, 2011-05]"""
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_period_ordered(self):
+        idx = period_range("2011-01-01 09:00", freq="h", periods=5)
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0    2011-01-01 09:00
+1    2011-01-01 10:00
+2    2011-01-01 11:00
+3    2011-01-01 12:00
+4    2011-01-01 13:00
+dtype: category
+Categories (5, period[h]): [2011-01-01 09:00 < 2011-01-01 10:00 < 2011-01-01 11:00 < 2011-01-01 12:00 <
+                            2011-01-01 13:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+        idx = period_range("2011-01", freq="M", periods=5)
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0    2011-01
+1    2011-02
+2    2011-03
+3    2011-04
+4    2011-05
+dtype: category
+Categories (5, period[M]): [2011-01 < 2011-02 < 2011-03 < 2011-04 < 2011-05]"""
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_timedelta(self):
+        idx = timedelta_range("1 days", periods=5)
+        s = Series(Categorical(idx))
+        exp = """0   1 days
+1   2 days
+2   3 days
+3   4 days
+4   5 days
+dtype: category
+Categories (5, timedelta64[ns]): [1 days, 2 days, 3 days, 4 days, 5 days]"""
+
+        assert repr(s) == exp
+
+        idx = timedelta_range("1 hours", periods=10)
+        s = Series(Categorical(idx))
+        exp = """0   0 days 01:00:00
+1   1 days 01:00:00
+2   2 days 01:00:00
+3   3 days 01:00:00
+4   4 days 01:00:00
+5   5 days 01:00:00
+6   6 days 01:00:00
+7   7 days 01:00:00
+8   8 days 01:00:00
+9   9 days 01:00:00
+dtype: category
+Categories (10, timedelta64[ns]): [0 days 01:00:00, 1 days 01:00:00, 2 days 01:00:00,
+                                   3 days 01:00:00, ..., 6 days 01:00:00, 7 days 01:00:00,
+                                   8 days 01:00:00, 9 days 01:00:00]"""  # noqa: E501
+
+        assert repr(s) == exp
+
+    def test_categorical_series_repr_timedelta_ordered(self):
+        idx = timedelta_range("1 days", periods=5)
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0   1 days
+1   2 days
+2   3 days
+3   4 days
+4   5 days
+dtype: category
+Categories (5, timedelta64[ns]): [1 days < 2 days < 3 days < 4 days < 5 days]"""
+
+        assert repr(s) == exp
+
+        idx = timedelta_range("1 hours", periods=10)
+        s = Series(Categorical(idx, ordered=True))
+        exp = """0   0 days 01:00:00
+1   1 days 01:00:00
+2   2 days 01:00:00
+3   3 days 01:00:00
+4   4 days 01:00:00
+5   5 days 01:00:00
+6   6 days 01:00:00
+7   7 days 01:00:00
+8   8 days 01:00:00
+9   9 days 01:00:00
+dtype: category
+Categories (10, timedelta64[ns]): [0 days 01:00:00 < 1 days 01:00:00 < 2 days 01:00:00 <
+                                   3 days 01:00:00 ... 6 days 01:00:00 < 7 days 01:00:00 <
+                                   8 days 01:00:00 < 9 days 01:00:00]"""  # noqa: E501
+
+        assert repr(s) == exp
diff --git a/pandas/pandas/tests/series/test_iteration.py b/pandas/pandas/tests/series/test_iteration.py
new file mode 100644
index 0000000000000..1e0fa7fae107e
--- /dev/null
+++ b/pandas/pandas/tests/series/test_iteration.py
@@ -0,0 +1,33 @@
+class TestIteration:
+    def test_keys(self, datetime_series):
+        assert datetime_series.keys() is datetime_series.index
+
+    def test_iter_datetimes(self, datetime_series):
+        for i, val in enumerate(datetime_series):
+            assert val == datetime_series.iloc[i]
+
+    def test_iter_strings(self, string_series):
+        for i, val in enumerate(string_series):
+            assert val == string_series.iloc[i]
+
+    def test_iteritems_datetimes(self, datetime_series):
+        for idx, val in datetime_series.items():
+            assert val == datetime_series[idx]
+
+    def test_iteritems_strings(self, string_series):
+        for idx, val in string_series.items():
+            assert val == string_series[idx]
+
+        # assert is lazy (generators don't define reverse, lists do)
+        assert not hasattr(string_series.items(), "reverse")
+
+    def test_items_datetimes(self, datetime_series):
+        for idx, val in datetime_series.items():
+            assert val == datetime_series[idx]
+
+    def test_items_strings(self, string_series):
+        for idx, val in string_series.items():
+            assert val == string_series[idx]
+
+        # assert is lazy (generators don't define reverse, lists do)
+        assert not hasattr(string_series.items(), "reverse")
diff --git a/pandas/pandas/tests/series/test_logical_ops.py b/pandas/pandas/tests/series/test_logical_ops.py
new file mode 100644
index 0000000000000..8f63819b09238
--- /dev/null
+++ b/pandas/pandas/tests/series/test_logical_ops.py
@@ -0,0 +1,548 @@
+from datetime import datetime
+import operator
+
+import numpy as np
+import pytest
+
+from pandas import (
+    ArrowDtype,
+    DataFrame,
+    Index,
+    Series,
+    StringDtype,
+    bdate_range,
+)
+import pandas._testing as tm
+from pandas.core import ops
+
+
+class TestSeriesLogicalOps:
+    @pytest.mark.parametrize("bool_op", [operator.and_, operator.or_, operator.xor])
+    def test_bool_operators_with_nas(self, bool_op):
+        # boolean &, |, ^ should work with object arrays and propagate NAs
+        ser = Series(bdate_range("1/1/2000", periods=10), dtype=object)
+        ser[::2] = np.nan
+
+        mask = ser.isna()
+        filled = ser.fillna(ser[0])
+
+        result = bool_op(ser < ser[9], ser > ser[3])
+
+        expected = bool_op(filled < filled[9], filled > filled[3])
+        expected[mask] = False
+        tm.assert_series_equal(result, expected)
+
+    def test_logical_operators_bool_dtype_with_empty(self):
+        # GH#9016: support bitwise op for integer types
+        index = list("bca")
+
+        s_tft = Series([True, False, True], index=index)
+        s_fff = Series([False, False, False], index=index)
+        s_empty = Series([], dtype=object)
+
+        res = s_tft & s_empty
+        expected = s_fff.sort_index()
+        tm.assert_series_equal(res, expected)
+
+        res = s_tft | s_empty
+        expected = s_tft.sort_index()
+        tm.assert_series_equal(res, expected)
+
+    def test_logical_operators_int_dtype_with_int_dtype(self):
+        # GH#9016: support bitwise op for integer types
+
+        s_0123 = Series(range(4), dtype="int64")
+        s_3333 = Series([3] * 4)
+        s_4444 = Series([4] * 4)
+
+        res = s_0123 & s_3333
+        expected = Series(range(4), dtype="int64")
+        tm.assert_series_equal(res, expected)
+
+        res = s_0123 | s_4444
+        expected = Series(range(4, 8), dtype="int64")
+        tm.assert_series_equal(res, expected)
+
+        s_1111 = Series([1] * 4, dtype="int8")
+        res = s_0123 & s_1111
+        expected = Series([0, 1, 0, 1], dtype="int64")
+        tm.assert_series_equal(res, expected)
+
+        res = s_0123.astype(np.int16) | s_1111.astype(np.int32)
+        expected = Series([1, 1, 3, 3], dtype="int32")
+        tm.assert_series_equal(res, expected)
+
+    def test_logical_operators_int_dtype_with_int_scalar(self):
+        # GH#9016: support bitwise op for integer types
+        s_0123 = Series(range(4), dtype="int64")
+
+        res = s_0123 & 0
+        expected = Series([0] * 4)
+        tm.assert_series_equal(res, expected)
+
+        res = s_0123 & 1
+        expected = Series([0, 1, 0, 1])
+        tm.assert_series_equal(res, expected)
+
+    def test_logical_operators_int_dtype_with_float(self):
+        # GH#9016: support bitwise op for integer types
+        s_0123 = Series(range(4), dtype="int64")
+
+        err_msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+
+        msg = "Cannot perform.+with a dtyped.+array and scalar of type"
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & np.nan
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & 3.14
+        msg = "unsupported operand type.+for &:"
+        with pytest.raises(TypeError, match=err_msg):
+            s_0123 & [0.1, 4, 3.14, 2]
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & np.array([0.1, 4, 3.14, 2])
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & Series([0.1, 4, -3.14, 2])
+
+    def test_logical_operators_int_dtype_with_str(self):
+        s_1111 = Series([1] * 4, dtype="int8")
+
+        err_msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+
+        msg = "Cannot perform 'and_' with a dtyped.+array and scalar of type"
+        with pytest.raises(TypeError, match=msg):
+            s_1111 & "a"
+        with pytest.raises(TypeError, match=err_msg):
+            s_1111 & ["a", "b", "c", "d"]
+
+    def test_logical_operators_int_dtype_with_bool(self):
+        # GH#9016: support bitwise op for integer types
+        s_0123 = Series(range(4), dtype="int64")
+
+        expected = Series([False] * 4)
+
+        result = s_0123 & False
+        tm.assert_series_equal(result, expected)
+
+        msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & [False]
+
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & (False,)
+
+        result = s_0123 ^ False
+        expected = Series([False, True, True, True])
+        tm.assert_series_equal(result, expected)
+
+    def test_logical_operators_int_dtype_with_object(self):
+        # GH#9016: support bitwise op for integer types
+        s_0123 = Series(range(4), dtype="int64")
+
+        result = s_0123 & Series([False, np.nan, False, False])
+        expected = Series([False] * 4)
+        tm.assert_series_equal(result, expected)
+
+        s_abNd = Series(["a", "b", np.nan, "d"])
+        with pytest.raises(
+            TypeError, match="unsupported.* 'int' and 'str'|'rand_' not supported"
+        ):
+            s_0123 & s_abNd
+
+    def test_logical_operators_bool_dtype_with_int(self):
+        index = list("bca")
+
+        s_tft = Series([True, False, True], index=index)
+        s_fff = Series([False, False, False], index=index)
+
+        res = s_tft & 0
+        expected = s_fff
+        tm.assert_series_equal(res, expected)
+
+        res = s_tft & 1
+        expected = s_tft
+        tm.assert_series_equal(res, expected)
+
+    def test_logical_ops_bool_dtype_with_ndarray(self):
+        # make sure we operate on ndarray the same as Series
+        left = Series([True, True, True, False, True])
+        right = [True, False, None, True, np.nan]
+
+        msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+
+        expected = Series([True, False, False, False, False])
+        with pytest.raises(TypeError, match=msg):
+            left & right
+        result = left & np.array(right)
+        tm.assert_series_equal(result, expected)
+        result = left & Index(right)
+        tm.assert_series_equal(result, expected)
+        result = left & Series(right)
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([True, True, True, True, True])
+        with pytest.raises(TypeError, match=msg):
+            left | right
+        result = left | np.array(right)
+        tm.assert_series_equal(result, expected)
+        result = left | Index(right)
+        tm.assert_series_equal(result, expected)
+        result = left | Series(right)
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([False, True, True, True, True])
+        with pytest.raises(TypeError, match=msg):
+            left ^ right
+        result = left ^ np.array(right)
+        tm.assert_series_equal(result, expected)
+        result = left ^ Index(right)
+        tm.assert_series_equal(result, expected)
+        result = left ^ Series(right)
+        tm.assert_series_equal(result, expected)
+
+    def test_logical_operators_int_dtype_with_bool_dtype_and_reindex(self):
+        # GH#9016: support bitwise op for integer types
+
+        index = list("bca")
+
+        s_tft = Series([True, False, True], index=index)
+        s_tft = Series([True, False, True], index=index)
+        s_tff = Series([True, False, False], index=index)
+
+        s_0123 = Series(range(4), dtype="int64")
+
+        # s_0123 will be all false now because of reindexing like s_tft
+        expected = Series([False] * 7, index=[0, 1, 2, 3, "a", "b", "c"])
+        result = s_tft & s_0123
+        tm.assert_series_equal(result, expected)
+
+        # GH#52538: no longer to object type when reindex is needed;
+        # matches DataFrame behavior
+        msg = r"unsupported operand type\(s\) for &: 'float' and 'bool'"
+        with pytest.raises(TypeError, match=msg):
+            s_0123 & s_tft
+
+        s_a0b1c0 = Series([1], list("b"))
+
+        res = s_tft & s_a0b1c0
+        expected = s_tff.reindex(list("abc"))
+        tm.assert_series_equal(res, expected)
+
+        res = s_tft | s_a0b1c0
+        expected = s_tft.reindex(list("abc"))
+        tm.assert_series_equal(res, expected)
+
+    def test_scalar_na_logical_ops_corners(self):
+        s = Series([2, 3, 4, 5, 6, 7, 8, 9, 10])
+
+        msg = "Cannot perform.+with a dtyped.+array and scalar of type"
+        with pytest.raises(TypeError, match=msg):
+            s & datetime(2005, 1, 1)
+
+        s = Series([2, 3, 4, 5, 6, 7, 8, 9, datetime(2005, 1, 1)])
+        s[::2] = np.nan
+
+        expected = Series(True, index=s.index)
+        expected[::2] = False
+
+        msg = (
+            r"Logical ops \(and, or, xor\) between Pandas objects and "
+            "dtype-less sequences"
+        )
+        with pytest.raises(TypeError, match=msg):
+            s & list(s)
+
+    def test_scalar_na_logical_ops_corners_aligns(self):
+        s = Series([2, 3, 4, 5, 6, 7, 8, 9, datetime(2005, 1, 1)])
+        s[::2] = np.nan
+        d = DataFrame({"A": s})
+
+        expected = DataFrame(False, index=range(9), columns=["A"] + list(range(9)))
+
+        result = s & d
+        tm.assert_frame_equal(result, expected)
+
+        result = d & s
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("op", [operator.and_, operator.or_, operator.xor])
+    def test_logical_ops_with_index(self, op):
+        # GH#22092, GH#19792
+        ser = Series([True, True, False, False])
+        idx1 = Index([True, False, True, False])
+        idx2 = Index([1, 0, 1, 0])
+
+        expected = Series([op(ser[n], idx1[n]) for n in range(len(ser))])
+
+        result = op(ser, idx1)
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([op(ser[n], idx2[n]) for n in range(len(ser))], dtype=bool)
+
+        result = op(ser, idx2)
+        tm.assert_series_equal(result, expected)
+
+    def test_reversed_xor_with_index_returns_series(self):
+        # GH#22092, GH#19792 pre-2.0 these were aliased to setops
+        ser = Series([True, True, False, False])
+        idx1 = Index([True, False, True, False], dtype=bool)
+        idx2 = Index([1, 0, 1, 0])
+
+        expected = Series([False, True, True, False])
+        result = idx1 ^ ser
+        tm.assert_series_equal(result, expected)
+
+        result = idx2 ^ ser
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "op",
+        [
+            ops.rand_,
+            ops.ror_,
+        ],
+    )
+    def test_reversed_logical_op_with_index_returns_series(self, op):
+        # GH#22092, GH#19792
+        ser = Series([True, True, False, False])
+        idx1 = Index([True, False, True, False])
+        idx2 = Index([1, 0, 1, 0])
+
+        expected = Series(op(idx1.values, ser.values))
+        result = op(ser, idx1)
+        tm.assert_series_equal(result, expected)
+
+        expected = op(ser, Series(idx2))
+        result = op(ser, idx2)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "op, expected",
+        [
+            (ops.rand_, [False, False]),
+            (ops.ror_, [True, True]),
+            (ops.rxor, [True, True]),
+        ],
+    )
+    def test_reverse_ops_with_index(self, op, expected):
+        # https://github.com/pandas-dev/pandas/pull/23628
+        # multi-set Index ops are buggy, so let's avoid duplicates...
+        # GH#49503
+        ser = Series([True, False])
+        idx = Index([False, True])
+
+        result = op(ser, idx)
+        expected = Series(expected)
+        tm.assert_series_equal(result, expected)
+
+    def test_logical_ops_label_based(self, using_infer_string):
+        # GH#4947
+        # logical ops should be label based
+
+        a = Series([True, False, True], list("bca"))
+        b = Series([False, True, False], list("abc"))
+
+        expected = Series([False, True, False], list("abc"))
+        result = a & b
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([True, True, False], list("abc"))
+        result = a | b
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([True, False, False], list("abc"))
+        result = a ^ b
+        tm.assert_series_equal(result, expected)
+
+        # rhs is bigger
+        a = Series([True, False, True], list("bca"))
+        b = Series([False, True, False, True], list("abcd"))
+
+        expected = Series([False, True, False, False], list("abcd"))
+        result = a & b
+        tm.assert_series_equal(result, expected)
+
+        expected = Series([True, True, False, False], list("abcd"))
+        result = a | b
+        tm.assert_series_equal(result, expected)
+
+        # filling
+
+        # vs empty
+        empty = Series([], dtype=object)
+
+        result = a & empty
+        expected = Series([False, False, False], list("abc"))
+        tm.assert_series_equal(result, expected)
+
+        result = a | empty
+        expected = Series([True, True, False], list("abc"))
+        tm.assert_series_equal(result, expected)
+
+        # vs non-matching
+        result = a & Series([1], ["z"])
+        expected = Series([False, False, False, False], list("abcz"))
+        tm.assert_series_equal(result, expected)
+
+        result = a | Series([1], ["z"])
+        expected = Series([True, True, False, False], list("abcz"))
+        tm.assert_series_equal(result, expected)
+
+        # identity
+        # we would like s[s|e] == s to hold for any e, whether empty or not
+        for e in [
+            empty.copy(),
+            Series([1], ["z"]),
+            Series(np.nan, b.index),
+            Series(np.nan, a.index),
+        ]:
+            result = a[a | e]
+            tm.assert_series_equal(result, a[a])
+
+        for e in [Series(["z"])]:
+            if using_infer_string:
+                # TODO(infer_string) should this behave differently?
+                # -> https://github.com/pandas-dev/pandas/issues/60234
+                with pytest.raises(
+                    TypeError, match="not supported for dtype|unsupported operand type"
+                ):
+                    result = a[a | e]
+            else:
+                result = a[a | e]
+            tm.assert_series_equal(result, a[a])
+
+        # vs scalars
+        index = list("bca")
+        t = Series([True, False, True])
+
+        for v in [True, 1, 2]:
+            result = Series([True, False, True], index=index) | v
+            expected = Series([True, True, True], index=index)
+            tm.assert_series_equal(result, expected)
+
+        msg = "Cannot perform.+with a dtyped.+array and scalar of type"
+        for v in [np.nan, "foo"]:
+            with pytest.raises(TypeError, match=msg):
+                t | v
+
+        for v in [False, 0]:
+            result = Series([True, False, True], index=index) | v
+            expected = Series([True, False, True], index=index)
+            tm.assert_series_equal(result, expected)
+
+        for v in [True, 1]:
+            result = Series([True, False, True], index=index) & v
+            expected = Series([True, False, True], index=index)
+            tm.assert_series_equal(result, expected)
+
+        for v in [False, 0]:
+            result = Series([True, False, True], index=index) & v
+            expected = Series([False, False, False], index=index)
+            tm.assert_series_equal(result, expected)
+        msg = "Cannot perform.+with a dtyped.+array and scalar of type"
+        for v in [np.nan]:
+            with pytest.raises(TypeError, match=msg):
+                t & v
+
+    def test_logical_ops_df_compat(self):
+        # GH#1134
+        s1 = Series([True, False, True], index=list("ABC"), name="x")
+        s2 = Series([True, True, False], index=list("ABD"), name="x")
+
+        exp = Series([True, False, False, False], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s1 & s2, exp)
+        tm.assert_series_equal(s2 & s1, exp)
+
+        # True | np.nan => True
+        exp_or1 = Series([True, True, True, False], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s1 | s2, exp_or1)
+        # np.nan | True => np.nan, filled with False
+        exp_or = Series([True, True, False, False], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s2 | s1, exp_or)
+
+        # DataFrame doesn't fill nan with False
+        tm.assert_frame_equal(s1.to_frame() & s2.to_frame(), exp.to_frame())
+        tm.assert_frame_equal(s2.to_frame() & s1.to_frame(), exp.to_frame())
+
+        exp = DataFrame({"x": [True, True, np.nan, np.nan]}, index=list("ABCD"))
+        tm.assert_frame_equal(s1.to_frame() | s2.to_frame(), exp_or1.to_frame())
+        tm.assert_frame_equal(s2.to_frame() | s1.to_frame(), exp_or.to_frame())
+
+        # different length
+        s3 = Series([True, False, True], index=list("ABC"), name="x")
+        s4 = Series([True, True, True, True], index=list("ABCD"), name="x")
+
+        exp = Series([True, False, True, False], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s3 & s4, exp)
+        tm.assert_series_equal(s4 & s3, exp)
+
+        # np.nan | True => np.nan, filled with False
+        exp_or1 = Series([True, True, True, False], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s3 | s4, exp_or1)
+        # True | np.nan => True
+        exp_or = Series([True, True, True, True], index=list("ABCD"), name="x")
+        tm.assert_series_equal(s4 | s3, exp_or)
+
+        tm.assert_frame_equal(s3.to_frame() & s4.to_frame(), exp.to_frame())
+        tm.assert_frame_equal(s4.to_frame() & s3.to_frame(), exp.to_frame())
+
+        tm.assert_frame_equal(s3.to_frame() | s4.to_frame(), exp_or1.to_frame())
+        tm.assert_frame_equal(s4.to_frame() | s3.to_frame(), exp_or.to_frame())
+
+    def test_int_dtype_different_index_not_bool(self):
+        # GH 52500
+        ser1 = Series([1, 2, 3], index=[10, 11, 23], name="a")
+        ser2 = Series([10, 20, 30], index=[11, 10, 23], name="a")
+        result = np.bitwise_xor(ser1, ser2)
+        expected = Series([21, 8, 29], index=[10, 11, 23], name="a")
+        tm.assert_series_equal(result, expected)
+
+        result = ser1 ^ ser2
+        tm.assert_series_equal(result, expected)
+
+    # TODO: this belongs in comparison tests
+    def test_pyarrow_numpy_string_invalid(self):
+        # GH#56008
+        pa = pytest.importorskip("pyarrow")
+        ser = Series([False, True])
+        ser2 = Series(["a", "b"], dtype=StringDtype(na_value=np.nan))
+        result = ser == ser2
+        expected_eq = Series(False, index=ser.index)
+        tm.assert_series_equal(result, expected_eq)
+
+        result = ser != ser2
+        expected_ne = Series(True, index=ser.index)
+        tm.assert_series_equal(result, expected_ne)
+
+        with pytest.raises(TypeError, match="Invalid comparison"):
+            ser > ser2
+
+        # GH#59505
+        ser3 = ser2.astype("string[pyarrow]")
+        result3_eq = ser3 == ser
+        tm.assert_series_equal(result3_eq, expected_eq.astype("bool[pyarrow]"))
+        result3_ne = ser3 != ser
+        tm.assert_series_equal(result3_ne, expected_ne.astype("bool[pyarrow]"))
+
+        with pytest.raises(TypeError, match="Invalid comparison"):
+            ser > ser3
+
+        ser4 = ser2.astype(ArrowDtype(pa.string()))
+        result4_eq = ser4 == ser
+        tm.assert_series_equal(result4_eq, expected_eq.astype("bool[pyarrow]"))
+        result4_ne = ser4 != ser
+        tm.assert_series_equal(result4_ne, expected_ne.astype("bool[pyarrow]"))
+
+        with pytest.raises(TypeError, match="Invalid comparison"):
+            ser > ser4
diff --git a/pandas/pandas/tests/series/test_missing.py b/pandas/pandas/tests/series/test_missing.py
new file mode 100644
index 0000000000000..1c88329a83b0e
--- /dev/null
+++ b/pandas/pandas/tests/series/test_missing.py
@@ -0,0 +1,88 @@
+from datetime import timedelta
+
+import numpy as np
+import pytest
+
+from pandas._libs import iNaT
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    Index,
+    NaT,
+    Series,
+    isna,
+)
+import pandas._testing as tm
+
+
+class TestSeriesMissingData:
+    def test_categorical_nan_handling(self):
+        # NaNs are represented as -1 in labels
+        s = Series(Categorical(["a", "b", np.nan, "a"]))
+        tm.assert_index_equal(s.cat.categories, Index(["a", "b"]))
+        tm.assert_numpy_array_equal(
+            s.values.codes, np.array([0, 1, -1, 0], dtype=np.int8)
+        )
+
+    def test_timedelta64_nan(self):
+        td = Series([timedelta(days=i) for i in range(10)])
+
+        # nan ops on timedeltas
+        td1 = td.copy()
+        td1[0] = np.nan
+        assert isna(td1[0])
+        assert td1[0]._value == iNaT
+        td1[0] = td[0]
+        assert not isna(td1[0])
+
+        # GH#16674 iNaT is treated as an integer when given by the user
+        with pytest.raises(TypeError, match="Invalid value"):
+            td1[1] = iNaT
+
+        td1[2] = NaT
+        assert isna(td1[2])
+        assert td1[2]._value == iNaT
+        td1[2] = td[2]
+        assert not isna(td1[2])
+
+        # boolean setting
+        # GH#2899 boolean setting
+        td3 = np.timedelta64(timedelta(days=3))
+        td7 = np.timedelta64(timedelta(days=7))
+        td[(td > td3) & (td < td7)] = np.nan
+        assert isna(td).sum() == 3
+
+    @pytest.mark.xfail(
+        reason="Chained inequality raises when trying to define 'selector'"
+    )
+    def test_logical_range_select(self, datetime_series):
+        # NumPy limitation =(
+        # https://github.com/pandas-dev/pandas/commit/9030dc021f07c76809848925cb34828f6c8484f3
+
+        selector = -0.5 <= datetime_series <= 0.5
+        expected = (datetime_series >= -0.5) & (datetime_series <= 0.5)
+        tm.assert_series_equal(selector, expected)
+
+    def test_valid(self, datetime_series):
+        ts = datetime_series.copy()
+        ts.index = ts.index._with_freq(None)
+        ts[::2] = np.nan
+
+        result = ts.dropna()
+        assert len(result) == ts.count()
+        tm.assert_series_equal(result, ts[1::2])
+        tm.assert_series_equal(result, ts[pd.notna(ts)])
+
+
+def test_hasnans_uncached_for_series():
+    # GH#19700
+    # set float64 dtype to avoid upcast when setting nan
+    idx = Index([0, 1], dtype="float64")
+    assert idx.hasnans is False
+    assert "hasnans" in idx._cache
+    ser = idx.to_series()
+    assert ser.hasnans is False
+    assert not hasattr(ser, "_cache")
+    ser.iloc[-1] = np.nan
+    assert ser.hasnans is True
diff --git a/pandas/pandas/tests/series/test_npfuncs.py b/pandas/pandas/tests/series/test_npfuncs.py
new file mode 100644
index 0000000000000..11a51c4700d5c
--- /dev/null
+++ b/pandas/pandas/tests/series/test_npfuncs.py
@@ -0,0 +1,46 @@
+"""
+Tests for np.foo applied to Series, not necessarily ufuncs.
+"""
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestPtp:
+    def test_ptp(self):
+        # GH#21614
+        N = 1000
+        arr = np.random.default_rng(2).standard_normal(N)
+        ser = Series(arr)
+        assert np.ptp(ser) == np.ptp(arr)
+
+
+def test_numpy_unique(datetime_series):
+    # it works!
+    np.unique(datetime_series)
+
+
+@pytest.mark.parametrize("index", [["a", "b", "c", "d", "e"], None])
+def test_numpy_argwhere(index):
+    # GH#35331
+
+    s = Series(range(5), index=index, dtype=np.int64)
+
+    result = np.argwhere(s > 2).astype(np.int64)
+    expected = np.array([[3], [4]], dtype=np.int64)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@td.skip_if_no("pyarrow")
+def test_log_arrow_backed_missing_value():
+    # GH#56285
+    ser = Series([1, 2, None], dtype="float64[pyarrow]")
+    result = np.log(ser)
+    expected = np.log(Series([1, 2, None], dtype="float64"))
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/test_reductions.py b/pandas/pandas/tests/series/test_reductions.py
new file mode 100644
index 0000000000000..86ce60b1fc12b
--- /dev/null
+++ b/pandas/pandas/tests/series/test_reductions.py
@@ -0,0 +1,210 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("operation, expected", [("min", "a"), ("max", "b")])
+def test_reductions_series_strings(operation, expected):
+    # GH#31746
+    ser = Series(["a", "b"], dtype="string")
+    res_operation_serie = getattr(ser, operation)()
+    assert res_operation_serie == expected
+
+
+@pytest.mark.parametrize("as_period", [True, False])
+def test_mode_extension_dtype(as_period):
+    # GH#41927 preserve dt64tz dtype
+    ser = Series([pd.Timestamp(1979, 4, n) for n in range(1, 5)])
+
+    if as_period:
+        ser = ser.dt.to_period("D")
+    else:
+        ser = ser.dt.tz_localize("US/Central")
+
+    res = ser.mode()
+    assert res.dtype == ser.dtype
+    tm.assert_series_equal(res, ser)
+
+
+def test_mode_nullable_dtype(any_numeric_ea_dtype):
+    # GH#55340
+    ser = Series([1, 3, 2, pd.NA, 3, 2, pd.NA], dtype=any_numeric_ea_dtype)
+    result = ser.mode(dropna=False)
+    expected = Series([2, 3, pd.NA], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.mode(dropna=True)
+    expected = Series([2, 3], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    ser[-1] = pd.NA
+
+    result = ser.mode(dropna=True)
+    expected = Series([2, 3], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.mode(dropna=False)
+    expected = Series([pd.NA], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_mode_infer_string():
+    # GH#56183
+    pytest.importorskip("pyarrow")
+    ser = Series(["a", "b"], dtype=object)
+    with pd.option_context("future.infer_string", True):
+        result = ser.mode()
+    expected = Series(["a", "b"], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_reductions_td64_with_nat():
+    # GH#8617
+    ser = Series([0, pd.NaT], dtype="m8[ns]")
+    exp = ser[0]
+    assert ser.median() == exp
+    assert ser.min() == exp
+    assert ser.max() == exp
+
+
+def test_td64_sum_empty(skipna):
+    # GH#37151
+    ser = Series([], dtype="timedelta64[ns]")
+
+    result = ser.sum(skipna=skipna)
+    assert isinstance(result, pd.Timedelta)
+    assert result == pd.Timedelta(0)
+
+
+def test_td64_summation_overflow():
+    # GH#9442
+    ser = Series(pd.date_range("20130101", periods=100000, freq="h"))
+    ser[0] += pd.Timedelta("1s 1ms")
+
+    # mean
+    result = (ser - ser.min()).mean()
+    expected = pd.Timedelta((pd.TimedeltaIndex(ser - ser.min()).asi8 / len(ser)).sum())
+
+    # the computation is converted to float so
+    # might be some loss of precision
+    assert np.allclose(result._value / 1000, expected._value / 1000)
+
+    # sum
+    msg = "overflow in timedelta operation"
+    with pytest.raises(ValueError, match=msg):
+        (ser - ser.min()).sum()
+
+    s1 = ser[0:10000]
+    with pytest.raises(ValueError, match=msg):
+        (s1 - s1.min()).sum()
+    s2 = ser[0:1000]
+    (s2 - s2.min()).sum()
+
+
+def test_prod_numpy16_bug():
+    ser = Series([1.0, 1.0, 1.0], index=range(3))
+    result = ser.prod()
+
+    assert not isinstance(result, Series)
+
+
+@pytest.mark.parametrize("func", [np.any, np.all])
+@pytest.mark.parametrize("kwargs", [{"keepdims": True}, {"out": object()}])
+def test_validate_any_all_out_keepdims_raises(kwargs, func):
+    ser = Series([1, 2])
+    param = next(iter(kwargs))
+    name = func.__name__
+
+    msg = (
+        f"the '{param}' parameter is not "
+        "supported in the pandas "
+        rf"implementation of {name}\(\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        func(ser, **kwargs)
+
+
+def test_validate_sum_initial():
+    ser = Series([1, 2])
+    msg = (
+        r"the 'initial' parameter is not "
+        r"supported in the pandas "
+        r"implementation of sum\(\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        np.sum(ser, initial=10)
+
+
+def test_validate_median_initial():
+    ser = Series([1, 2])
+    msg = (
+        r"the 'overwrite_input' parameter is not "
+        r"supported in the pandas "
+        r"implementation of median\(\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        # It seems like np.median doesn't dispatch, so we use the
+        # method instead of the ufunc.
+        ser.median(overwrite_input=True)
+
+
+def test_validate_stat_keepdims():
+    ser = Series([1, 2])
+    msg = (
+        r"the 'keepdims' parameter is not "
+        r"supported in the pandas "
+        r"implementation of sum\(\)"
+    )
+    with pytest.raises(ValueError, match=msg):
+        np.sum(ser, keepdims=True)
+
+
+def test_mean_with_convertible_string_raises():
+    # GH#44008
+    ser = Series(["1", "2"])
+    assert ser.sum() == "12"
+
+    msg = "Could not convert string '12' to numeric|does not support|Cannot perform"
+    with pytest.raises(TypeError, match=msg):
+        ser.mean()
+
+    df = ser.to_frame()
+    msg = r"Could not convert \['12'\] to numeric|does not support|Cannot perform"
+    with pytest.raises(TypeError, match=msg):
+        df.mean()
+
+
+def test_mean_dont_convert_j_to_complex():
+    # GH#36703
+    df = pd.DataFrame([{"db": "J", "numeric": 123}])
+    msg = r"Could not convert \['J'\] to numeric|does not support|Cannot perform"
+    with pytest.raises(TypeError, match=msg):
+        df.mean()
+
+    with pytest.raises(TypeError, match=msg):
+        df.agg("mean")
+
+    msg = "Could not convert string 'J' to numeric|does not support|Cannot perform"
+    with pytest.raises(TypeError, match=msg):
+        df["db"].mean()
+    msg = "Could not convert string 'J' to numeric|ufunc 'divide'|Cannot perform"
+    with pytest.raises(TypeError, match=msg):
+        np.mean(df["db"].astype("string").array)
+
+
+def test_median_with_convertible_string_raises():
+    # GH#34671 this _could_ return a string "2", but definitely not float 2.0
+    msg = r"Cannot convert \['1' '2' '3'\] to numeric|does not support|Cannot perform"
+    ser = Series(["1", "2", "3"])
+    with pytest.raises(TypeError, match=msg):
+        ser.median()
+
+    msg = (
+        r"Cannot convert \[\['1' '2' '3'\]\] to numeric|does not support|Cannot perform"
+    )
+    df = ser.to_frame()
+    with pytest.raises(TypeError, match=msg):
+        df.median()
diff --git a/pandas/pandas/tests/series/test_subclass.py b/pandas/pandas/tests/series/test_subclass.py
new file mode 100644
index 0000000000000..c2d5afcf884b1
--- /dev/null
+++ b/pandas/pandas/tests/series/test_subclass.py
@@ -0,0 +1,82 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:Passing a BlockManager|Passing a SingleBlockManager:DeprecationWarning"
+)
+
+
+class TestSeriesSubclassing:
+    @pytest.mark.parametrize(
+        "idx_method, indexer, exp_data, exp_idx",
+        [
+            ["loc", ["a", "b"], [1, 2], "ab"],
+            ["iloc", [2, 3], [3, 4], "cd"],
+        ],
+    )
+    def test_indexing_sliced(self, idx_method, indexer, exp_data, exp_idx):
+        s = tm.SubclassedSeries([1, 2, 3, 4], index=list("abcd"))
+        res = getattr(s, idx_method)[indexer]
+        exp = tm.SubclassedSeries(exp_data, index=list(exp_idx))
+        tm.assert_series_equal(res, exp)
+
+    def test_to_frame(self):
+        s = tm.SubclassedSeries([1, 2, 3, 4], index=list("abcd"), name="xxx")
+        res = s.to_frame()
+        exp = tm.SubclassedDataFrame({"xxx": [1, 2, 3, 4]}, index=list("abcd"))
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_unstack(self):
+        # GH 15564
+        s = tm.SubclassedSeries([1, 2, 3, 4], index=[list("aabb"), list("xyxy")])
+
+        res = s.unstack()
+        exp = tm.SubclassedDataFrame({"x": [1, 3], "y": [2, 4]}, index=["a", "b"])
+
+        tm.assert_frame_equal(res, exp)
+
+    def test_subclass_empty_repr(self):
+        sub_series = tm.SubclassedSeries()
+        assert "SubclassedSeries" in repr(sub_series)
+
+    def test_asof(self):
+        N = 3
+        rng = pd.date_range("1/1/1990", periods=N, freq="53s")
+        s = tm.SubclassedSeries({"A": [np.nan, np.nan, np.nan]}, index=rng)
+
+        result = s.asof(rng[-2:])
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_explode(self):
+        s = tm.SubclassedSeries([[1, 2, 3], "foo", [], [3, 4]])
+        result = s.explode()
+        assert isinstance(result, tm.SubclassedSeries)
+
+    def test_equals(self):
+        # https://github.com/pandas-dev/pandas/pull/34402
+        # allow subclass in both directions
+        s1 = pd.Series([1, 2, 3])
+        s2 = tm.SubclassedSeries([1, 2, 3])
+        assert s1.equals(s2)
+        assert s2.equals(s1)
+
+
+class SubclassedSeries(pd.Series):
+    @property
+    def _constructor(self):
+        def _new(*args, **kwargs):
+            # some constructor logic that accesses the Series' name
+            if self.name == "test":
+                return pd.Series(*args, **kwargs)
+            return SubclassedSeries(*args, **kwargs)
+
+        return _new
+
+
+def test_constructor_from_dict():
+    # https://github.com/pandas-dev/pandas/issues/52445
+    result = SubclassedSeries({"a": 1, "b": 2, "c": 3})
+    assert isinstance(result, SubclassedSeries)
diff --git a/pandas/pandas/tests/series/test_ufunc.py b/pandas/pandas/tests/series/test_ufunc.py
new file mode 100644
index 0000000000000..a5976bb2518c9
--- /dev/null
+++ b/pandas/pandas/tests/series/test_ufunc.py
@@ -0,0 +1,467 @@
+from collections import deque
+import re
+import string
+
+import numpy as np
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.arrays import SparseArray
+
+
+@pytest.fixture(params=[np.add, np.logaddexp])
+def ufunc(request):
+    # dunder op
+    return request.param
+
+
+@pytest.fixture(
+    params=[pytest.param(True, marks=pytest.mark.fails_arm_wheels), False],
+    ids=["sparse", "dense"],
+)
+def sparse(request):
+    return request.param
+
+
+@pytest.fixture
+def arrays_for_binary_ufunc():
+    """
+    A pair of random, length-100 integer-dtype arrays, that are mostly 0.
+    """
+    a1 = np.random.default_rng(2).integers(0, 10, 100, dtype="int64")
+    a2 = np.random.default_rng(2).integers(0, 10, 100, dtype="int64")
+    a1[::3] = 0
+    a2[::4] = 0
+    return a1, a2
+
+
+@pytest.mark.parametrize("ufunc", [np.positive, np.floor, np.exp])
+def test_unary_ufunc(ufunc, sparse):
+    # Test that ufunc(pd.Series) == pd.Series(ufunc)
+    arr = np.random.default_rng(2).integers(0, 10, 10, dtype="int64")
+    arr[::2] = 0
+    if sparse:
+        arr = SparseArray(arr, dtype=pd.SparseDtype("int64", 0))
+
+    index = list(string.ascii_letters[:10])
+    name = "name"
+    series = pd.Series(arr, index=index, name=name)
+
+    result = ufunc(series)
+    expected = pd.Series(ufunc(arr), index=index, name=name)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("flip", [True, False], ids=["flipped", "straight"])
+def test_binary_ufunc_with_array(flip, sparse, ufunc, arrays_for_binary_ufunc):
+    # Test that ufunc(pd.Series(a), array) == pd.Series(ufunc(a, b))
+    a1, a2 = arrays_for_binary_ufunc
+    if sparse:
+        a1 = SparseArray(a1, dtype=pd.SparseDtype("int64", 0))
+        a2 = SparseArray(a2, dtype=pd.SparseDtype("int64", 0))
+
+    name = "name"  # op(pd.Series, array) preserves the name.
+    series = pd.Series(a1, name=name)
+    other = a2
+
+    array_args = (a1, a2)
+    series_args = (series, other)  # ufunc(series, array)
+
+    if flip:
+        array_args = reversed(array_args)
+        series_args = reversed(series_args)  # ufunc(array, series)
+
+    expected = pd.Series(ufunc(*array_args), name=name)
+    result = ufunc(*series_args)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("flip", [True, False], ids=["flipped", "straight"])
+def test_binary_ufunc_with_index(flip, sparse, ufunc, arrays_for_binary_ufunc):
+    # Test that
+    #   * func(pd.Series(a), pd.Series(b)) == pd.Series(ufunc(a, b))
+    #   * ufunc(Index, pd.Series) dispatches to pd.Series (returns a pd.Series)
+    a1, a2 = arrays_for_binary_ufunc
+    if sparse:
+        a1 = SparseArray(a1, dtype=pd.SparseDtype("int64", 0))
+        a2 = SparseArray(a2, dtype=pd.SparseDtype("int64", 0))
+
+    name = "name"  # op(pd.Series, array) preserves the name.
+    series = pd.Series(a1, name=name)
+
+    other = pd.Index(a2, name=name).astype("int64")
+
+    array_args = (a1, a2)
+    series_args = (series, other)  # ufunc(series, array)
+
+    if flip:
+        array_args = reversed(array_args)
+        series_args = reversed(series_args)  # ufunc(array, series)
+
+    expected = pd.Series(ufunc(*array_args), name=name)
+    result = ufunc(*series_args)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("shuffle", [True, False], ids=["unaligned", "aligned"])
+@pytest.mark.parametrize("flip", [True, False], ids=["flipped", "straight"])
+def test_binary_ufunc_with_series(
+    flip, shuffle, sparse, ufunc, arrays_for_binary_ufunc
+):
+    # Test that
+    #   * func(pd.Series(a), pd.Series(b)) == pd.Series(ufunc(a, b))
+    #   with alignment between the indices
+    a1, a2 = arrays_for_binary_ufunc
+    if sparse:
+        a1 = SparseArray(a1, dtype=pd.SparseDtype("int64", 0))
+        a2 = SparseArray(a2, dtype=pd.SparseDtype("int64", 0))
+
+    name = "name"  # op(pd.Series, array) preserves the name.
+    series = pd.Series(a1, name=name)
+    other = pd.Series(a2, name=name)
+
+    idx = np.random.default_rng(2).permutation(len(a1))
+
+    if shuffle:
+        other = other.take(idx)
+        if flip:
+            index = other.align(series)[0].index
+        else:
+            index = series.align(other)[0].index
+    else:
+        index = series.index
+
+    array_args = (a1, a2)
+    series_args = (series, other)  # ufunc(series, array)
+
+    if flip:
+        array_args = tuple(reversed(array_args))
+        series_args = tuple(reversed(series_args))  # ufunc(array, series)
+
+    expected = pd.Series(ufunc(*array_args), index=index, name=name)
+    result = ufunc(*series_args)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("flip", [True, False])
+def test_binary_ufunc_scalar(ufunc, sparse, flip, arrays_for_binary_ufunc):
+    # Test that
+    #   * ufunc(pd.Series, scalar) == pd.Series(ufunc(array, scalar))
+    #   * ufunc(pd.Series, scalar) == ufunc(scalar, pd.Series)
+    arr, _ = arrays_for_binary_ufunc
+    if sparse:
+        arr = SparseArray(arr)
+    other = 2
+    series = pd.Series(arr, name="name")
+
+    series_args = (series, other)
+    array_args = (arr, other)
+
+    if flip:
+        series_args = tuple(reversed(series_args))
+        array_args = tuple(reversed(array_args))
+
+    expected = pd.Series(ufunc(*array_args), name="name")
+    result = ufunc(*series_args)
+
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("ufunc", [np.divmod])  # TODO: np.modf, np.frexp
+@pytest.mark.parametrize("shuffle", [True, False])
+@pytest.mark.filterwarnings("ignore:divide by zero:RuntimeWarning")
+def test_multiple_output_binary_ufuncs(ufunc, sparse, shuffle, arrays_for_binary_ufunc):
+    # Test that
+    #  the same conditions from binary_ufunc_scalar apply to
+    #  ufuncs with multiple outputs.
+
+    a1, a2 = arrays_for_binary_ufunc
+    # work around https://github.com/pandas-dev/pandas/issues/26987
+    a1[a1 == 0] = 1
+    a2[a2 == 0] = 1
+
+    if sparse:
+        a1 = SparseArray(a1, dtype=pd.SparseDtype("int64", 0))
+        a2 = SparseArray(a2, dtype=pd.SparseDtype("int64", 0))
+
+    s1 = pd.Series(a1)
+    s2 = pd.Series(a2)
+
+    if shuffle:
+        # ensure we align before applying the ufunc
+        s2 = s2.sample(frac=1)
+
+    expected = ufunc(a1, a2)
+    assert isinstance(expected, tuple)
+
+    result = ufunc(s1, s2)
+    assert isinstance(result, tuple)
+    tm.assert_series_equal(result[0], pd.Series(expected[0]))
+    tm.assert_series_equal(result[1], pd.Series(expected[1]))
+
+
+def test_multiple_output_ufunc(sparse, arrays_for_binary_ufunc):
+    # Test that the same conditions from unary input apply to multi-output
+    # ufuncs
+    arr, _ = arrays_for_binary_ufunc
+
+    if sparse:
+        arr = SparseArray(arr)
+
+    series = pd.Series(arr, name="name")
+    result = np.modf(series)
+    expected = np.modf(arr)
+
+    assert isinstance(result, tuple)
+    assert isinstance(expected, tuple)
+
+    tm.assert_series_equal(result[0], pd.Series(expected[0], name="name"))
+    tm.assert_series_equal(result[1], pd.Series(expected[1], name="name"))
+
+
+def test_binary_ufunc_drops_series_name(ufunc, sparse, arrays_for_binary_ufunc):
+    # Drop the names when they differ.
+    a1, a2 = arrays_for_binary_ufunc
+    s1 = pd.Series(a1, name="a")
+    s2 = pd.Series(a2, name="b")
+
+    result = ufunc(s1, s2)
+    assert result.name is None
+
+
+def test_object_series_ok():
+    class Dummy:
+        def __init__(self, value) -> None:
+            self.value = value
+
+        def __add__(self, other):
+            return self.value + other.value
+
+    arr = np.array([Dummy(0), Dummy(1)])
+    ser = pd.Series(arr)
+    tm.assert_series_equal(np.add(ser, ser), pd.Series(np.add(ser, arr)))
+    tm.assert_series_equal(np.add(ser, Dummy(1)), pd.Series(np.add(ser, Dummy(1))))
+
+
+@pytest.fixture(
+    params=[
+        pd.array([1, 3, 2], dtype=np.int64),
+        pd.array([1, 3, 2], dtype="Int64"),
+        pd.array([1, 3, 2], dtype="Float32"),
+        pd.array([1, 10, 2], dtype="Sparse[int]"),
+        pd.to_datetime(["2000", "2010", "2001"]),
+        pd.to_datetime(["2000", "2010", "2001"]).tz_localize("CET"),
+        pd.to_datetime(["2000", "2010", "2001"]).to_period(freq="D"),
+        pd.to_timedelta(["1 Day", "3 Days", "2 Days"]),
+        pd.IntervalIndex([pd.Interval(0, 1), pd.Interval(2, 3), pd.Interval(1, 2)]),
+    ],
+    ids=lambda x: str(x.dtype),
+)
+def values_for_np_reduce(request):
+    # min/max tests assume that these are monotonic increasing
+    return request.param
+
+
+class TestNumpyReductions:
+    # TODO: cases with NAs, axis kwarg for DataFrame
+
+    def test_multiply(self, values_for_np_reduce, box_with_array, request):
+        box = box_with_array
+        values = values_for_np_reduce
+
+        with tm.assert_produces_warning(None):
+            obj = box(values)
+
+        if isinstance(values, pd.core.arrays.SparseArray):
+            mark = pytest.mark.xfail(reason="SparseArray has no 'prod'")
+            request.applymarker(mark)
+
+        if values.dtype.kind in "iuf":
+            result = np.multiply.reduce(obj)
+            if box is pd.DataFrame:
+                expected = obj.prod(numeric_only=False)
+                tm.assert_series_equal(result, expected)
+            elif box is pd.Index:
+                # Index has no 'prod'
+                expected = obj._values.prod()
+                assert result == expected
+            else:
+                expected = obj.prod()
+                assert result == expected
+        else:
+            msg = "|".join(
+                [
+                    "does not support operation",
+                    "unsupported operand type",
+                    "ufunc 'multiply' cannot use operands",
+                ]
+            )
+            with pytest.raises(TypeError, match=msg):
+                np.multiply.reduce(obj)
+
+    def test_add(self, values_for_np_reduce, box_with_array):
+        box = box_with_array
+        values = values_for_np_reduce
+
+        with tm.assert_produces_warning(None):
+            obj = box(values)
+
+        if values.dtype.kind in "miuf":
+            result = np.add.reduce(obj)
+            if box is pd.DataFrame:
+                expected = obj.sum(numeric_only=False)
+                tm.assert_series_equal(result, expected)
+            elif box is pd.Index:
+                # Index has no 'sum'
+                expected = obj._values.sum()
+                assert result == expected
+            else:
+                expected = obj.sum()
+                assert result == expected
+        else:
+            msg = "|".join(
+                [
+                    "does not support operation",
+                    "unsupported operand type",
+                    "ufunc 'add' cannot use operands",
+                ]
+            )
+            with pytest.raises(TypeError, match=msg):
+                np.add.reduce(obj)
+
+    def test_max(self, values_for_np_reduce, box_with_array):
+        box = box_with_array
+        values = values_for_np_reduce
+
+        same_type = True
+        if box is pd.Index and values.dtype.kind in ["i", "f"]:
+            # ATM Index casts to object, so we get python ints/floats
+            same_type = False
+
+        with tm.assert_produces_warning(None):
+            obj = box(values)
+
+        result = np.maximum.reduce(obj)
+        if box is pd.DataFrame:
+            # TODO: cases with axis kwarg
+            expected = obj.max(numeric_only=False)
+            tm.assert_series_equal(result, expected)
+        else:
+            expected = values[1]
+            assert result == expected
+            if same_type:
+                # check we have e.g. Timestamp instead of dt64
+                assert type(result) == type(expected)
+
+    def test_min(self, values_for_np_reduce, box_with_array):
+        box = box_with_array
+        values = values_for_np_reduce
+
+        same_type = True
+        if box is pd.Index and values.dtype.kind in ["i", "f"]:
+            # ATM Index casts to object, so we get python ints/floats
+            same_type = False
+
+        with tm.assert_produces_warning(None):
+            obj = box(values)
+
+        result = np.minimum.reduce(obj)
+        if box is pd.DataFrame:
+            expected = obj.min(numeric_only=False)
+            tm.assert_series_equal(result, expected)
+        else:
+            expected = values[0]
+            assert result == expected
+            if same_type:
+                # check we have e.g. Timestamp instead of dt64
+                assert type(result) == type(expected)
+
+
+@pytest.mark.parametrize("type_", [list, deque, tuple])
+def test_binary_ufunc_other_types(type_):
+    a = pd.Series([1, 2, 3], name="name")
+    b = type_([3, 4, 5])
+
+    result = np.add(a, b)
+    expected = pd.Series(np.add(a.to_numpy(), b), name="name")
+    tm.assert_series_equal(result, expected)
+
+
+def test_object_dtype_ok():
+    class Thing:
+        def __init__(self, value) -> None:
+            self.value = value
+
+        def __add__(self, other):
+            other = getattr(other, "value", other)
+            return type(self)(self.value + other)
+
+        def __eq__(self, other) -> bool:
+            return type(other) is Thing and self.value == other.value
+
+        def __repr__(self) -> str:
+            return f"Thing({self.value})"
+
+    s = pd.Series([Thing(1), Thing(2)])
+    result = np.add(s, Thing(1))
+    expected = pd.Series([Thing(2), Thing(3)])
+    tm.assert_series_equal(result, expected)
+
+
+def test_outer():
+    # https://github.com/pandas-dev/pandas/issues/27186
+    ser = pd.Series([1, 2, 3])
+    obj = np.array([1, 2, 3])
+
+    with pytest.raises(NotImplementedError, match=""):
+        np.subtract.outer(ser, obj)
+
+
+def test_np_matmul():
+    # GH26650
+    df1 = pd.DataFrame(data=[[-1, 1, 10]])
+    df2 = pd.DataFrame(data=[-1, 1, 10])
+    expected = pd.DataFrame(data=[102])
+
+    result = np.matmul(df1, df2)
+    tm.assert_frame_equal(expected, result)
+
+
+@pytest.mark.parametrize("box", [pd.Index, pd.Series])
+def test_np_matmul_1D(box):
+    result = np.matmul(box([1, 2]), box([2, 3]))
+    assert result == 8
+    assert isinstance(result, np.int64)
+
+
+def test_array_ufuncs_for_many_arguments():
+    # GH39853
+    def add3(x, y, z):
+        return x + y + z
+
+    ufunc = np.frompyfunc(add3, 3, 1)
+    ser = pd.Series([1, 2])
+
+    result = ufunc(ser, ser, 1)
+    expected = pd.Series([3, 5], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    df = pd.DataFrame([[1, 2]])
+
+    msg = (
+        "Cannot apply ufunc <ufunc 'add3 (vectorized)'> "
+        "to mixed DataFrame and Series inputs."
+    )
+    with pytest.raises(NotImplementedError, match=re.escape(msg)):
+        ufunc(ser, ser, df)
+
+
+@pytest.mark.xfail(reason="see https://github.com/pandas-dev/pandas/pull/51082")
+def test_np_fix():
+    # np.fix is not a ufunc but is composed of several ufunc calls under the hood
+    # with `out` and `where` keywords
+    ser = pd.Series([-1.5, -0.5, 0.5, 1.5])
+    result = np.fix(ser)
+    expected = pd.Series([-1.0, -0.0, 0.0, 1.0])
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/series/test_unary.py b/pandas/pandas/tests/series/test_unary.py
new file mode 100644
index 0000000000000..8f153788e413c
--- /dev/null
+++ b/pandas/pandas/tests/series/test_unary.py
@@ -0,0 +1,50 @@
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+class TestSeriesUnaryOps:
+    # __neg__, __pos__, __invert__
+
+    def test_neg(self):
+        ser = Series(range(5), dtype="float64", name="series")
+        tm.assert_series_equal(-ser, -1 * ser)
+
+    def test_invert(self):
+        ser = Series(range(5), dtype="float64", name="series")
+        tm.assert_series_equal(-(ser < 0), ~(ser < 0))
+
+    @pytest.mark.parametrize(
+        "source, neg_target, abs_target",
+        [
+            ([1, 2, 3], [-1, -2, -3], [1, 2, 3]),
+            ([1, 2, None], [-1, -2, None], [1, 2, None]),
+        ],
+    )
+    def test_all_numeric_unary_operators(
+        self, any_numeric_ea_dtype, source, neg_target, abs_target
+    ):
+        # GH38794
+        dtype = any_numeric_ea_dtype
+        ser = Series(source, dtype=dtype)
+        neg_result, pos_result, abs_result = -ser, +ser, abs(ser)
+        if dtype.startswith("U"):
+            neg_target = -Series(source, dtype=dtype)
+        else:
+            neg_target = Series(neg_target, dtype=dtype)
+
+        abs_target = Series(abs_target, dtype=dtype)
+
+        tm.assert_series_equal(neg_result, neg_target)
+        tm.assert_series_equal(pos_result, ser)
+        tm.assert_series_equal(abs_result, abs_target)
+
+    @pytest.mark.parametrize("op", ["__neg__", "__abs__"])
+    def test_unary_float_op_mask(self, float_ea_dtype, op):
+        dtype = float_ea_dtype
+        ser = Series([1.1, 2.2, 3.3], dtype=dtype)
+        result = getattr(ser, op)()
+        target = result.copy(deep=True)
+        ser[0] = None
+        tm.assert_series_equal(result, target)
diff --git a/pandas/pandas/tests/series/test_validate.py b/pandas/pandas/tests/series/test_validate.py
new file mode 100644
index 0000000000000..3c867f7582b7d
--- /dev/null
+++ b/pandas/pandas/tests/series/test_validate.py
@@ -0,0 +1,26 @@
+import pytest
+
+
+@pytest.mark.parametrize(
+    "func",
+    [
+        "reset_index",
+        "_set_name",
+        "sort_values",
+        "sort_index",
+        "rename",
+        "dropna",
+        "drop_duplicates",
+    ],
+)
+@pytest.mark.parametrize("inplace", [1, "True", [1, 2, 3], 5.0])
+def test_validate_bool_args(string_series, func, inplace):
+    """Tests for error handling related to data types of method arguments."""
+    msg = 'For argument "inplace" expected type bool'
+    kwargs = {"inplace": inplace}
+
+    if func == "_set_name":
+        kwargs["name"] = "hello"
+
+    with pytest.raises(ValueError, match=msg):
+        getattr(string_series, func)(**kwargs)
diff --git a/pandas/pandas/tests/strings/__init__.py b/pandas/pandas/tests/strings/__init__.py
new file mode 100644
index 0000000000000..6c4bec6a23789
--- /dev/null
+++ b/pandas/pandas/tests/strings/__init__.py
@@ -0,0 +1,23 @@
+import numpy as np
+
+import pandas as pd
+
+
+def is_object_or_nan_string_dtype(dtype):
+    """
+    Check if string-like dtype is following NaN semantics, i.e. is object
+    dtype or a NaN-variant of the StringDtype.
+    """
+    return (isinstance(dtype, np.dtype) and dtype == "object") or (
+        dtype.na_value is np.nan
+    )
+
+
+def _convert_na_value(ser, expected):
+    if ser.dtype != object:
+        if ser.dtype.na_value is np.nan:
+            expected = expected.fillna(np.nan)
+        else:
+            # GH#18463
+            expected = expected.fillna(pd.NA)
+    return expected
diff --git a/pandas/pandas/tests/strings/conftest.py b/pandas/pandas/tests/strings/conftest.py
new file mode 100644
index 0000000000000..5bcbb16da3be9
--- /dev/null
+++ b/pandas/pandas/tests/strings/conftest.py
@@ -0,0 +1,133 @@
+import pytest
+
+from pandas import Series
+from pandas.core.strings.accessor import StringMethods
+
+_any_string_method = [
+    ("cat", (), {"sep": ","}),
+    ("cat", (Series(list("zyx")),), {"sep": ",", "join": "left"}),
+    ("center", (10,), {}),
+    ("contains", ("a",), {}),
+    ("count", ("a",), {}),
+    ("decode", ("UTF-8",), {}),
+    ("encode", ("UTF-8",), {}),
+    ("endswith", ("a",), {}),
+    ("endswith", ((),), {}),
+    ("endswith", (("a",),), {}),
+    ("endswith", (("a", "b"),), {}),
+    ("endswith", (("a", "MISSING"),), {}),
+    ("endswith", ("a",), {"na": True}),
+    ("endswith", ("a",), {"na": False}),
+    ("extract", ("([a-z]*)",), {"expand": False}),
+    ("extract", ("([a-z]*)",), {"expand": True}),
+    ("extractall", ("([a-z]*)",), {}),
+    ("find", ("a",), {}),
+    ("findall", ("a",), {}),
+    ("get", (0,), {}),
+    # because "index" (and "rindex") fail intentionally
+    # if the string is not found, search only for empty string
+    ("index", ("",), {}),
+    ("join", (",",), {}),
+    ("ljust", (10,), {}),
+    ("match", ("a",), {}),
+    ("fullmatch", ("a",), {}),
+    ("normalize", ("NFC",), {}),
+    ("pad", (10,), {}),
+    ("partition", (" ",), {"expand": False}),
+    ("partition", (" ",), {"expand": True}),
+    ("repeat", (3,), {}),
+    ("replace", ("a", "z"), {}),
+    ("rfind", ("a",), {}),
+    ("rindex", ("",), {}),
+    ("rjust", (10,), {}),
+    ("rpartition", (" ",), {"expand": False}),
+    ("rpartition", (" ",), {"expand": True}),
+    ("slice", (0, 1), {}),
+    ("slice_replace", (0, 1, "z"), {}),
+    ("split", (" ",), {"expand": False}),
+    ("split", (" ",), {"expand": True}),
+    ("startswith", ("a",), {}),
+    ("startswith", (("a",),), {}),
+    ("startswith", (("a", "b"),), {}),
+    ("startswith", (("a", "MISSING"),), {}),
+    ("startswith", ((),), {}),
+    ("startswith", ("a",), {"na": True}),
+    ("startswith", ("a",), {"na": False}),
+    ("removeprefix", ("a",), {}),
+    ("removesuffix", ("a",), {}),
+    # translating unicode points of "a" to "d"
+    ("translate", ({97: 100},), {}),
+    ("wrap", (2,), {}),
+    ("zfill", (10,), {}),
+] + list(
+    zip(
+        [
+            # methods without positional arguments: zip with empty tuple and empty dict
+            "capitalize",
+            "cat",
+            "get_dummies",
+            "isalnum",
+            "isalpha",
+            "isascii",
+            "isdecimal",
+            "isdigit",
+            "islower",
+            "isnumeric",
+            "isspace",
+            "istitle",
+            "isupper",
+            "len",
+            "lower",
+            "lstrip",
+            "partition",
+            "rpartition",
+            "rsplit",
+            "rstrip",
+            "slice",
+            "slice_replace",
+            "split",
+            "strip",
+            "swapcase",
+            "title",
+            "upper",
+            "casefold",
+        ],
+        [()] * 100,
+        [{}] * 100,
+    )
+)
+ids, _, _ = zip(*_any_string_method)  # use method name as fixture-id
+missing_methods = {f for f in dir(StringMethods) if not f.startswith("_")} - set(ids)
+
+# test that the above list captures all methods of StringMethods
+assert not missing_methods
+
+
+@pytest.fixture(params=_any_string_method, ids=ids)
+def any_string_method(request):
+    """
+    Fixture for all public methods of `StringMethods`
+
+    This fixture returns a tuple of the method name and sample arguments
+    necessary to call the method.
+
+    Returns
+    -------
+    method_name : str
+        The name of the method in `StringMethods`
+    args : tuple
+        Sample values for the positional arguments
+    kwargs : dict
+        Sample values for the keyword arguments
+
+    Examples
+    --------
+    >>> def test_something(any_string_method):
+    ...     s = Series(["a", "b", np.nan, "d"])
+    ...
+    ...     method_name, args, kwargs = any_string_method
+    ...     method = getattr(s.str, method_name)
+    ...     # will not raise
+    ...     method(*args, **kwargs)
+    """
+    return request.param
diff --git a/pandas/pandas/tests/strings/test_api.py b/pandas/pandas/tests/strings/test_api.py
new file mode 100644
index 0000000000000..4a1b97606db2b
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_api.py
@@ -0,0 +1,216 @@
+import weakref
+
+import numpy as np
+import pytest
+
+from pandas import (
+    CategoricalDtype,
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+    option_context,
+)
+from pandas.core.strings.accessor import StringMethods
+
+# subset of the full set from pandas/conftest.py
+_any_allowed_skipna_inferred_dtype = [
+    ("string", ["a", np.nan, "c"]),
+    ("bytes", [b"a", np.nan, b"c"]),
+    ("empty", [np.nan, np.nan, np.nan]),
+    ("empty", []),
+    ("mixed-integer", ["a", np.nan, 2]),
+]
+ids, _ = zip(*_any_allowed_skipna_inferred_dtype)  # use inferred type as id
+
+
+@pytest.fixture(params=_any_allowed_skipna_inferred_dtype, ids=ids)
+def any_allowed_skipna_inferred_dtype(request):
+    """
+    Fixture for all (inferred) dtypes allowed in StringMethods.__init__
+
+    The covered (inferred) types are:
+    * 'string'
+    * 'empty'
+    * 'bytes'
+    * 'mixed'
+    * 'mixed-integer'
+
+    Returns
+    -------
+    inferred_dtype : str
+        The string for the inferred dtype from _libs.lib.infer_dtype
+    values : np.ndarray
+        An array of object dtype that will be inferred to have
+        `inferred_dtype`
+
+    Examples
+    --------
+    >>> from pandas._libs import lib
+    >>>
+    >>> def test_something(any_allowed_skipna_inferred_dtype):
+    ...     inferred_dtype, values = any_allowed_skipna_inferred_dtype
+    ...     # will pass
+    ...     assert lib.infer_dtype(values, skipna=True) == inferred_dtype
+    ...
+    ...     # constructor for .str-accessor will also pass
+    ...     Series(values).str
+    """
+    inferred_dtype, values = request.param
+    values = np.array(values, dtype=object)  # object dtype to avoid casting
+
+    # correctness of inference tested in tests/dtypes/test_inference.py
+    return inferred_dtype, values
+
+
+def test_api(any_string_dtype):
+    # GH 6106, GH 9322
+    assert Series.str is StringMethods
+    assert isinstance(Series([""], dtype=any_string_dtype).str, StringMethods)
+
+
+def test_no_circular_reference(any_string_dtype):
+    # GH 47667
+    ser = Series([""], dtype=any_string_dtype)
+    ref = weakref.ref(ser)
+    ser.str  # Used to cache and cause circular reference
+    del ser
+    assert ref() is None
+
+
+def test_api_mi_raises():
+    # GH 23679
+    mi = MultiIndex.from_arrays([["a", "b", "c"]])
+    msg = "Can only use .str accessor with Index, not MultiIndex"
+    with pytest.raises(AttributeError, match=msg):
+        mi.str
+    assert not hasattr(mi, "str")
+
+
+@pytest.mark.parametrize("dtype", [object, "category"])
+def test_api_per_dtype(index_or_series, dtype, any_skipna_inferred_dtype):
+    # one instance of parametrized fixture
+    box = index_or_series
+    inferred_dtype, values = any_skipna_inferred_dtype
+
+    t = box(values, dtype=dtype)  # explicit dtype to avoid casting
+
+    types_passing_constructor = [
+        "string",
+        "unicode",
+        "empty",
+        "bytes",
+        "mixed",
+        "mixed-integer",
+    ]
+    if inferred_dtype in types_passing_constructor:
+        # GH 6106
+        assert isinstance(t.str, StringMethods)
+    else:
+        # GH 9184, GH 23011, GH 23163
+        msg = "Can only use .str accessor with string values.*"
+        with pytest.raises(AttributeError, match=msg):
+            t.str
+        assert not hasattr(t, "str")
+
+
+@pytest.mark.parametrize("dtype", [object, "category"])
+def test_api_per_method(
+    index_or_series,
+    dtype,
+    any_allowed_skipna_inferred_dtype,
+    any_string_method,
+    request,
+    using_infer_string,
+):
+    # this test does not check correctness of the different methods,
+    # just that the methods work on the specified (inferred) dtypes,
+    # and raise on all others
+    box = index_or_series
+
+    # one instance of each parametrized fixture
+    inferred_dtype, values = any_allowed_skipna_inferred_dtype
+    method_name, args, kwargs = any_string_method
+
+    reason = None
+    if box is Index and values.size == 0:
+        if method_name in ["partition", "rpartition"] and kwargs.get("expand", True):
+            raises = TypeError
+            reason = "Method cannot deal with empty Index"
+        elif method_name == "split" and kwargs.get("expand", None):
+            raises = TypeError
+            reason = "Split fails on empty Series when expand=True"
+        elif method_name == "get_dummies":
+            raises = ValueError
+            reason = "Need to fortify get_dummies corner cases"
+
+    elif (
+        box is Index
+        and inferred_dtype == "empty"
+        and dtype == object
+        and method_name == "get_dummies"
+    ):
+        raises = ValueError
+        reason = "Need to fortify get_dummies corner cases"
+
+    if reason is not None:
+        mark = pytest.mark.xfail(raises=raises, reason=reason)
+        request.applymarker(mark)
+
+    t = box(values, dtype=dtype)  # explicit dtype to avoid casting
+    method = getattr(t.str, method_name)
+
+    if using_infer_string and dtype == "category":
+        string_allowed = method_name not in ["decode"]
+    else:
+        string_allowed = True
+    bytes_allowed = method_name in ["decode", "get", "len", "slice"]
+    # as of v0.23.4, all methods except 'cat' are very lenient with the
+    # allowed data types, just returning NaN for entries that error.
+    # This could be changed with an 'errors'-kwarg to the `str`-accessor,
+    # see discussion in GH 13877
+    mixed_allowed = method_name not in ["cat"]
+
+    allowed_types = (
+        ["empty"]
+        + ["string", "unicode"] * string_allowed
+        + ["bytes"] * bytes_allowed
+        + ["mixed", "mixed-integer"] * mixed_allowed
+    )
+
+    if inferred_dtype in allowed_types:
+        # xref GH 23555, GH 23556
+        with option_context("future.no_silent_downcasting", True):
+            method(*args, **kwargs)  # works!
+    else:
+        # GH 23011, GH 23163
+        msg = (
+            f"Cannot use .str.{method_name} with values of "
+            f"inferred dtype {inferred_dtype!r}."
+            "|a bytes-like object is required, not 'str'"
+        )
+        with pytest.raises(TypeError, match=msg):
+            method(*args, **kwargs)
+
+
+def test_api_for_categorical(any_string_method, any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/10661
+    s = Series(list("aabb"), dtype=any_string_dtype)
+    s = s + " " + s
+    c = s.astype("category")
+    c = c.astype(CategoricalDtype(c.dtype.categories.astype("object")))
+    assert isinstance(c.str, StringMethods)
+
+    method_name, args, kwargs = any_string_method
+
+    result = getattr(c.str, method_name)(*args, **kwargs)
+    expected = getattr(s.astype("object").str, method_name)(*args, **kwargs)
+
+    if isinstance(result, DataFrame):
+        tm.assert_frame_equal(result, expected)
+    elif isinstance(result, Series):
+        tm.assert_series_equal(result, expected)
+    else:
+        # str.cat(others=None) returns string, for example
+        assert result == expected
diff --git a/pandas/pandas/tests/strings/test_case_justify.py b/pandas/pandas/tests/strings/test_case_justify.py
new file mode 100644
index 0000000000000..819556f961fa3
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_case_justify.py
@@ -0,0 +1,423 @@
+from datetime import datetime
+import operator
+
+import numpy as np
+import pytest
+
+from pandas import (
+    Series,
+    _testing as tm,
+)
+
+
+def test_title(any_string_dtype):
+    s = Series(["FOO", "BAR", np.nan, "Blah", "blurg"], dtype=any_string_dtype)
+    result = s.str.title()
+    expected = Series(["Foo", "Bar", np.nan, "Blah", "Blurg"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_title_mixed_object():
+    s = Series(["FOO", np.nan, "bar", True, datetime.today(), "blah", None, 1, 2.0])
+    result = s.str.title()
+    expected = Series(
+        ["Foo", np.nan, "Bar", np.nan, np.nan, "Blah", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_almost_equal(result, expected)
+
+
+def test_lower_upper(any_string_dtype):
+    s = Series(["om", np.nan, "nom", "nom"], dtype=any_string_dtype)
+
+    result = s.str.upper()
+    expected = Series(["OM", np.nan, "NOM", "NOM"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = result.str.lower()
+    tm.assert_series_equal(result, s)
+
+
+def test_lower_upper_mixed_object():
+    s = Series(["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0])
+
+    result = s.str.upper()
+    expected = Series(
+        ["A", np.nan, "B", np.nan, np.nan, "FOO", None, np.nan, np.nan], dtype=object
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.lower()
+    expected = Series(
+        ["a", np.nan, "b", np.nan, np.nan, "foo", None, np.nan, np.nan], dtype=object
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, expected",
+    [
+        (
+            ["FOO", "BAR", np.nan, "Blah", "blurg"],
+            ["Foo", "Bar", np.nan, "Blah", "Blurg"],
+        ),
+        (["a", "b", "c"], ["A", "B", "C"]),
+        (["a b", "a bc. de"], ["A b", "A bc. de"]),
+    ],
+)
+def test_capitalize(data, expected, any_string_dtype):
+    s = Series(data, dtype=any_string_dtype)
+    result = s.str.capitalize()
+    expected = Series(expected, dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_capitalize_mixed_object():
+    s = Series(["FOO", np.nan, "bar", True, datetime.today(), "blah", None, 1, 2.0])
+    result = s.str.capitalize()
+    expected = Series(
+        ["Foo", np.nan, "Bar", np.nan, np.nan, "Blah", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_swapcase(any_string_dtype):
+    s = Series(["FOO", "BAR", np.nan, "Blah", "blurg"], dtype=any_string_dtype)
+    result = s.str.swapcase()
+    expected = Series(["foo", "bar", np.nan, "bLAH", "BLURG"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_swapcase_mixed_object():
+    s = Series(["FOO", np.nan, "bar", True, datetime.today(), "Blah", None, 1, 2.0])
+    result = s.str.swapcase()
+    expected = Series(
+        ["foo", np.nan, "BAR", np.nan, np.nan, "bLAH", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_casefold():
+    # GH25405
+    expected = Series(["ss", np.nan, "case", "ssd"])
+    s = Series(["ß", np.nan, "case", "ßd"])
+    result = s.str.casefold()
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_casemethods(any_string_dtype):
+    values = ["aaa", "bbb", "CCC", "Dddd", "eEEE"]
+    s = Series(values, dtype=any_string_dtype)
+    assert s.str.lower().tolist() == [v.lower() for v in values]
+    assert s.str.upper().tolist() == [v.upper() for v in values]
+    assert s.str.title().tolist() == [v.title() for v in values]
+    assert s.str.capitalize().tolist() == [v.capitalize() for v in values]
+    assert s.str.swapcase().tolist() == [v.swapcase() for v in values]
+
+
+def test_pad(any_string_dtype):
+    s = Series(["a", "b", np.nan, "c", np.nan, "eeeeee"], dtype=any_string_dtype)
+
+    result = s.str.pad(5, side="left")
+    expected = Series(
+        ["    a", "    b", np.nan, "    c", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="right")
+    expected = Series(
+        ["a    ", "b    ", np.nan, "c    ", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="both")
+    expected = Series(
+        ["  a  ", "  b  ", np.nan, "  c  ", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_pad_mixed_object():
+    s = Series(["a", np.nan, "b", True, datetime.today(), "ee", None, 1, 2.0])
+
+    result = s.str.pad(5, side="left")
+    expected = Series(
+        ["    a", np.nan, "    b", np.nan, np.nan, "   ee", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="right")
+    expected = Series(
+        ["a    ", np.nan, "b    ", np.nan, np.nan, "ee   ", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="both")
+    expected = Series(
+        ["  a  ", np.nan, "  b  ", np.nan, np.nan, "  ee ", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_pad_fillchar(any_string_dtype):
+    s = Series(["a", "b", np.nan, "c", np.nan, "eeeeee"], dtype=any_string_dtype)
+
+    result = s.str.pad(5, side="left", fillchar="X")
+    expected = Series(
+        ["XXXXa", "XXXXb", np.nan, "XXXXc", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="right", fillchar="X")
+    expected = Series(
+        ["aXXXX", "bXXXX", np.nan, "cXXXX", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.pad(5, side="both", fillchar="X")
+    expected = Series(
+        ["XXaXX", "XXbXX", np.nan, "XXcXX", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_pad_fillchar_bad_arg_raises(any_string_dtype):
+    s = Series(["a", "b", np.nan, "c", np.nan, "eeeeee"], dtype=any_string_dtype)
+
+    msg = "fillchar must be a character, not str"
+    with pytest.raises(TypeError, match=msg):
+        s.str.pad(5, fillchar="XY")
+
+    msg = "fillchar must be a character, not int"
+    with pytest.raises(TypeError, match=msg):
+        s.str.pad(5, fillchar=5)
+
+
+@pytest.mark.parametrize("method_name", ["center", "ljust", "rjust", "zfill", "pad"])
+def test_pad_width_bad_arg_raises(method_name, any_string_dtype):
+    # see gh-13598
+    s = Series(["1", "22", "a", "bb"], dtype=any_string_dtype)
+    op = operator.methodcaller(method_name, "f")
+
+    msg = "width must be of integer type, not str"
+    with pytest.raises(TypeError, match=msg):
+        op(s.str)
+
+
+def test_center_ljust_rjust(any_string_dtype):
+    s = Series(["a", "b", np.nan, "c", np.nan, "eeeeee"], dtype=any_string_dtype)
+
+    result = s.str.center(5)
+    expected = Series(
+        ["  a  ", "  b  ", np.nan, "  c  ", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.ljust(5)
+    expected = Series(
+        ["a    ", "b    ", np.nan, "c    ", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.rjust(5)
+    expected = Series(
+        ["    a", "    b", np.nan, "    c", np.nan, "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_center_ljust_rjust_mixed_object():
+    s = Series(["a", np.nan, "b", True, datetime.today(), "c", "eee", None, 1, 2.0])
+
+    result = s.str.center(5)
+    expected = Series(
+        [
+            "  a  ",
+            np.nan,
+            "  b  ",
+            np.nan,
+            np.nan,
+            "  c  ",
+            " eee ",
+            None,
+            np.nan,
+            np.nan,
+        ],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.ljust(5)
+    expected = Series(
+        [
+            "a    ",
+            np.nan,
+            "b    ",
+            np.nan,
+            np.nan,
+            "c    ",
+            "eee  ",
+            None,
+            np.nan,
+            np.nan,
+        ],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.rjust(5)
+    expected = Series(
+        [
+            "    a",
+            np.nan,
+            "    b",
+            np.nan,
+            np.nan,
+            "    c",
+            "  eee",
+            None,
+            np.nan,
+            np.nan,
+        ],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_center_ljust_rjust_fillchar(any_string_dtype):
+    # GH#54533, GH#54792
+    s = Series(["a", "bb", "cccc", "ddddd", "eeeeee"], dtype=any_string_dtype)
+
+    result = s.str.center(5, fillchar="X")
+    expected = Series(
+        ["XXaXX", "XXbbX", "Xcccc", "ddddd", "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.center(5, "X") for v in np.array(s)], dtype=np.object_)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.object_), expected)
+
+    result = s.str.ljust(5, fillchar="X")
+    expected = Series(
+        ["aXXXX", "bbXXX", "ccccX", "ddddd", "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.ljust(5, "X") for v in np.array(s)], dtype=np.object_)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.object_), expected)
+
+    result = s.str.rjust(5, fillchar="X")
+    expected = Series(
+        ["XXXXa", "XXXbb", "Xcccc", "ddddd", "eeeeee"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.rjust(5, "X") for v in np.array(s)], dtype=np.object_)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.object_), expected)
+
+
+def test_center_ljust_rjust_fillchar_bad_arg_raises(any_string_dtype):
+    s = Series(["a", "bb", "cccc", "ddddd", "eeeeee"], dtype=any_string_dtype)
+
+    # If fillchar is not a character, normal str raises TypeError
+    # 'aaa'.ljust(5, 'XY')
+    # TypeError: must be char, not str
+    template = "fillchar must be a character, not {dtype}"
+
+    with pytest.raises(TypeError, match=template.format(dtype="str")):
+        s.str.center(5, fillchar="XY")
+
+    with pytest.raises(TypeError, match=template.format(dtype="str")):
+        s.str.ljust(5, fillchar="XY")
+
+    with pytest.raises(TypeError, match=template.format(dtype="str")):
+        s.str.rjust(5, fillchar="XY")
+
+    with pytest.raises(TypeError, match=template.format(dtype="int")):
+        s.str.center(5, fillchar=1)
+
+    with pytest.raises(TypeError, match=template.format(dtype="int")):
+        s.str.ljust(5, fillchar=1)
+
+    with pytest.raises(TypeError, match=template.format(dtype="int")):
+        s.str.rjust(5, fillchar=1)
+
+
+def test_zfill(any_string_dtype):
+    s = Series(["1", "22", "aaa", "333", "45678"], dtype=any_string_dtype)
+
+    result = s.str.zfill(5)
+    expected = Series(
+        ["00001", "00022", "00aaa", "00333", "45678"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.zfill(5) for v in np.array(s)], dtype=np.object_)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.object_), expected)
+
+    result = s.str.zfill(3)
+    expected = Series(["001", "022", "aaa", "333", "45678"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.zfill(3) for v in np.array(s)], dtype=np.object_)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.object_), expected)
+
+    s = Series(["1", np.nan, "aaa", np.nan, "45678"], dtype=any_string_dtype)
+    result = s.str.zfill(5)
+    expected = Series(
+        ["00001", np.nan, "00aaa", np.nan, "45678"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_wrap(any_string_dtype):
+    # test values are: two words less than width, two words equal to width,
+    # two words greater than width, one word less than width, one word
+    # equal to width, one word greater than width, multiple tokens with
+    # trailing whitespace equal to width
+    s = Series(
+        [
+            "hello world",
+            "hello world!",
+            "hello world!!",
+            "abcdefabcde",
+            "abcdefabcdef",
+            "abcdefabcdefa",
+            "ab ab ab ab ",
+            "ab ab ab ab a",
+            "\t",
+        ],
+        dtype=any_string_dtype,
+    )
+
+    # expected values
+    expected = Series(
+        [
+            "hello world",
+            "hello world!",
+            "hello\nworld!!",
+            "abcdefabcde",
+            "abcdefabcdef",
+            "abcdefabcdef\na",
+            "ab ab ab ab",
+            "ab ab ab ab\na",
+            "",
+        ],
+        dtype=any_string_dtype,
+    )
+
+    result = s.str.wrap(12, break_long_words=True)
+    tm.assert_series_equal(result, expected)
+
+
+def test_wrap_unicode(any_string_dtype):
+    # test with pre and post whitespace (non-unicode), NaN, and non-ascii Unicode
+    s = Series(
+        ["  pre  ", np.nan, "\xac\u20ac\U00008000 abadcafe"], dtype=any_string_dtype
+    )
+    expected = Series(
+        ["  pre", np.nan, "\xac\u20ac\U00008000 ab\nadcafe"], dtype=any_string_dtype
+    )
+    result = s.str.wrap(6)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/strings/test_cat.py b/pandas/pandas/tests/strings/test_cat.py
new file mode 100644
index 0000000000000..68ca807bde145
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_cat.py
@@ -0,0 +1,430 @@
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+    concat,
+    option_context,
+)
+
+
+@pytest.fixture
+def index_or_series2(index_or_series):
+    return index_or_series
+
+
+@pytest.mark.parametrize("other", [None, Series, Index])
+def test_str_cat_name(index_or_series, other):
+    # GH 21053
+    box = index_or_series
+    values = ["a", "b"]
+    if other:
+        other = other(values)
+    else:
+        other = values
+    result = box(values, name="name").str.cat(other, sep=",")
+    assert result.name == "name"
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+def test_str_cat(index_or_series, infer_string):
+    with option_context("future.infer_string", infer_string):
+        box = index_or_series
+        # test_cat above tests "str_cat" from ndarray;
+        # here testing "str.cat" from Series/Index to ndarray/list
+        s = box(["a", "a", "b", "b", "c", np.nan])
+
+        # single array
+        result = s.str.cat()
+        expected = "aabbc"
+        assert result == expected
+
+        result = s.str.cat(na_rep="-")
+        expected = "aabbc-"
+        assert result == expected
+
+        result = s.str.cat(sep="_", na_rep="NA")
+        expected = "a_a_b_b_c_NA"
+        assert result == expected
+
+        t = np.array(["a", np.nan, "b", "d", "foo", np.nan], dtype=object)
+        expected = box(["aa", "a-", "bb", "bd", "cfoo", "--"])
+
+        # Series/Index with array
+        result = s.str.cat(t, na_rep="-")
+        tm.assert_equal(result, expected)
+
+        # Series/Index with list
+        result = s.str.cat(list(t), na_rep="-")
+        tm.assert_equal(result, expected)
+
+        # errors for incorrect lengths
+        rgx = r"If `others` contains arrays or lists \(or other list-likes.*"
+        z = Series(["1", "2", "3"])
+
+        with pytest.raises(ValueError, match=rgx):
+            s.str.cat(z.values)
+
+        with pytest.raises(ValueError, match=rgx):
+            s.str.cat(list(z))
+
+
+def test_str_cat_raises_intuitive_error(index_or_series):
+    # GH 11334
+    box = index_or_series
+    s = box(["a", "b", "c", "d"])
+    message = "Did you mean to supply a `sep` keyword?"
+    with pytest.raises(ValueError, match=message):
+        s.str.cat("|")
+    with pytest.raises(ValueError, match=message):
+        s.str.cat("    ")
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+@pytest.mark.parametrize("sep", ["", None])
+@pytest.mark.parametrize("dtype_target", ["object", "category"])
+@pytest.mark.parametrize("dtype_caller", ["object", "category"])
+def test_str_cat_categorical(
+    index_or_series, dtype_caller, dtype_target, sep, infer_string
+):
+    box = index_or_series
+
+    with option_context("future.infer_string", infer_string):
+        s = Index(["a", "a", "b", "a"], dtype=dtype_caller)
+        s = s if box == Index else Series(s, index=s, dtype=s.dtype)
+        t = Index(["b", "a", "b", "c"], dtype=dtype_target)
+
+        expected = Index(
+            ["ab", "aa", "bb", "ac"], dtype=object if dtype_caller == "object" else None
+        )
+        expected = (
+            expected
+            if box == Index
+            else Series(
+                expected, index=Index(s, dtype=dtype_caller), dtype=expected.dtype
+            )
+        )
+
+        # Series/Index with unaligned Index -> t.values
+        result = s.str.cat(t.values, sep=sep)
+        tm.assert_equal(result, expected)
+
+        # Series/Index with Series having matching Index
+        t = Series(t.values, index=Index(s, dtype=dtype_caller))
+        result = s.str.cat(t, sep=sep)
+        tm.assert_equal(result, expected)
+
+        # Series/Index with Series.values
+        result = s.str.cat(t.values, sep=sep)
+        tm.assert_equal(result, expected)
+
+        # Series/Index with Series having different Index
+        t = Series(t.values, index=t.values)
+        expected = Index(
+            ["aa", "aa", "bb", "bb", "aa"],
+            dtype=object if dtype_caller == "object" else None,
+        )
+        dtype = object if dtype_caller == "object" else s.dtype.categories.dtype
+        expected = (
+            expected
+            if box == Index
+            else Series(
+                expected,
+                index=Index(expected.str[:1], dtype=dtype),
+                dtype=expected.dtype,
+            )
+        )
+
+        result = s.str.cat(t, sep=sep)
+        tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [[1, 2, 3], [0.1, 0.2, 0.3], [1, 2, "b"]],
+    ids=["integers", "floats", "mixed"],
+)
+# without dtype=object, np.array would cast [1, 2, 'b'] to ['1', '2', 'b']
+@pytest.mark.parametrize(
+    "box",
+    [Series, Index, list, lambda x: np.array(x, dtype=object)],
+    ids=["Series", "Index", "list", "np.array"],
+)
+def test_str_cat_wrong_dtype_raises(box, data):
+    # GH 22722
+    s = Series(["a", "b", "c"])
+    t = box(data)
+
+    msg = "Concatenation requires list-likes containing only strings.*"
+    with pytest.raises(TypeError, match=msg):
+        # need to use outer and na_rep, as otherwise Index would not raise
+        s.str.cat(t, join="outer", na_rep="-")
+
+
+def test_str_cat_mixed_inputs(index_or_series):
+    box = index_or_series
+    s = Index(["a", "b", "c", "d"])
+    s = s if box == Index else Series(s, index=s)
+
+    t = Series(["A", "B", "C", "D"], index=s.values)
+    d = concat([t, Series(s, index=s)], axis=1)
+
+    expected = Index(["aAa", "bBb", "cCc", "dDd"])
+    expected = expected if box == Index else Series(expected.values, index=s.values)
+
+    # Series/Index with DataFrame
+    result = s.str.cat(d)
+    tm.assert_equal(result, expected)
+
+    # Series/Index with two-dimensional ndarray
+    result = s.str.cat(d.values)
+    tm.assert_equal(result, expected)
+
+    # Series/Index with list of Series
+    result = s.str.cat([t, s])
+    tm.assert_equal(result, expected)
+
+    # Series/Index with mixed list of Series/array
+    result = s.str.cat([t, s.values])
+    tm.assert_equal(result, expected)
+
+    # Series/Index with list of Series; different indexes
+    t.index = ["b", "c", "d", "a"]
+    expected = box(["aDa", "bAb", "cBc", "dCd"])
+    expected = expected if box == Index else Series(expected.values, index=s.values)
+    result = s.str.cat([t, s])
+    tm.assert_equal(result, expected)
+
+    # Series/Index with mixed list; different index
+    result = s.str.cat([t, s.values])
+    tm.assert_equal(result, expected)
+
+    # Series/Index with DataFrame; different indexes
+    d.index = ["b", "c", "d", "a"]
+    expected = box(["aDd", "bAa", "cBb", "dCc"])
+    expected = expected if box == Index else Series(expected.values, index=s.values)
+    result = s.str.cat(d)
+    tm.assert_equal(result, expected)
+
+    # errors for incorrect lengths
+    rgx = r"If `others` contains arrays or lists \(or other list-likes.*"
+    z = Series(["1", "2", "3"])
+    e = concat([z, z], axis=1)
+
+    # two-dimensional ndarray
+    with pytest.raises(ValueError, match=rgx):
+        s.str.cat(e.values)
+
+    # list of list-likes
+    with pytest.raises(ValueError, match=rgx):
+        s.str.cat([z.values, s.values])
+
+    # mixed list of Series/list-like
+    with pytest.raises(ValueError, match=rgx):
+        s.str.cat([z.values, s])
+
+    # errors for incorrect arguments in list-like
+    rgx = "others must be Series, Index, DataFrame,.*"
+    # make sure None/NaN do not crash checks in _get_series_list
+    u = Series(["a", np.nan, "c", None])
+
+    # mix of string and Series
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat([u, "u"])
+
+    # DataFrame in list
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat([u, d])
+
+    # 2-dim ndarray in list
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat([u, d.values])
+
+    # nested lists
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat([u, [u, d]])
+
+    # forbidden input type: set
+    # GH 23009
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat(set(u))
+
+    # forbidden input type: set in list
+    # GH 23009
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat([u, set(u)])
+
+    # other forbidden input type, e.g. int
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat(1)
+
+    # nested list-likes
+    with pytest.raises(TypeError, match=rgx):
+        s.str.cat(iter([t.values, list(s)]))
+
+
+def test_str_cat_align_indexed(index_or_series, join_type):
+    # https://github.com/pandas-dev/pandas/issues/18657
+    box = index_or_series
+
+    s = Series(["a", "b", "c", "d"], index=["a", "b", "c", "d"])
+    t = Series(["D", "A", "E", "B"], index=["d", "a", "e", "b"])
+    sa, ta = s.align(t, join=join_type)
+    # result after manual alignment of inputs
+    expected = sa.str.cat(ta, na_rep="-")
+
+    if box == Index:
+        s = Index(s)
+        sa = Index(sa)
+        expected = Index(expected)
+
+    result = s.str.cat(t, join=join_type, na_rep="-")
+    tm.assert_equal(result, expected)
+
+
+def test_str_cat_align_mixed_inputs(join_type):
+    s = Series(["a", "b", "c", "d"])
+    t = Series(["d", "a", "e", "b"], index=[3, 0, 4, 1])
+    d = concat([t, t], axis=1)
+
+    expected_outer = Series(["aaa", "bbb", "c--", "ddd", "-ee"])
+    expected = expected_outer.loc[s.index.join(t.index, how=join_type)]
+
+    # list of Series
+    result = s.str.cat([t, t], join=join_type, na_rep="-")
+    tm.assert_series_equal(result, expected)
+
+    # DataFrame
+    result = s.str.cat(d, join=join_type, na_rep="-")
+    tm.assert_series_equal(result, expected)
+
+    # mixed list of indexed/unindexed
+    u = np.array(["A", "B", "C", "D"])
+    expected_outer = Series(["aaA", "bbB", "c-C", "ddD", "-e-"])
+    # joint index of rhs [t, u]; u will be forced have index of s
+    rhs_idx = (
+        t.index.intersection(s.index)
+        if join_type == "inner"
+        else t.index.union(s.index)
+        if join_type == "outer"
+        else t.index.append(s.index.difference(t.index))
+    )
+
+    expected = expected_outer.loc[s.index.join(rhs_idx, how=join_type)]
+    result = s.str.cat([t, u], join=join_type, na_rep="-")
+    tm.assert_series_equal(result, expected)
+
+    with pytest.raises(TypeError, match="others must be Series,.*"):
+        # nested lists are forbidden
+        s.str.cat([t, list(u)], join=join_type)
+
+    # errors for incorrect lengths
+    rgx = r"If `others` contains arrays or lists \(or other list-likes.*"
+    z = Series(["1", "2", "3"]).values
+
+    # unindexed object of wrong length
+    with pytest.raises(ValueError, match=rgx):
+        s.str.cat(z, join=join_type)
+
+    # unindexed object of wrong length in list
+    with pytest.raises(ValueError, match=rgx):
+        s.str.cat([t, z], join=join_type)
+
+
+def test_str_cat_all_na(index_or_series, index_or_series2):
+    # GH 24044
+    box = index_or_series
+    other = index_or_series2
+
+    # check that all NaNs in caller / target work
+    s = Index(["a", "b", "c", "d"])
+    s = s if box == Index else Series(s, index=s)
+    t = other([np.nan] * 4, dtype=object)
+    # add index of s for alignment
+    t = t if other == Index else Series(t, index=s)
+
+    # all-NA target
+    if box == Series:
+        expected = Series([np.nan] * 4, index=s.index, dtype=s.dtype)
+    else:  # box == Index
+        # TODO: Strimg option, this should return string dtype
+        expected = Index([np.nan] * 4, dtype=object)
+    result = s.str.cat(t, join="left")
+    tm.assert_equal(result, expected)
+
+    # all-NA caller (only for Series)
+    if other == Series:
+        expected = Series([np.nan] * 4, dtype=object, index=t.index)
+        result = t.str.cat(s, join="left")
+        tm.assert_series_equal(result, expected)
+
+
+def test_str_cat_special_cases():
+    s = Series(["a", "b", "c", "d"])
+    t = Series(["d", "a", "e", "b"], index=[3, 0, 4, 1])
+
+    # iterator of elements with different types
+    expected = Series(["aaa", "bbb", "c-c", "ddd", "-e-"])
+    result = s.str.cat(iter([t, s.values]), join="outer", na_rep="-")
+    tm.assert_series_equal(result, expected)
+
+    # right-align with different indexes in others
+    expected = Series(["aa-", "d-d"], index=[0, 3])
+    result = s.str.cat([t.loc[[0]], t.loc[[3]]], join="right", na_rep="-")
+    tm.assert_series_equal(result, expected)
+
+
+def test_cat_on_filtered_index():
+    df = DataFrame(
+        index=MultiIndex.from_product(
+            [[2011, 2012], [1, 2, 3]], names=["year", "month"]
+        )
+    )
+
+    df = df.reset_index()
+    df = df[df.month > 1]
+
+    str_year = df.year.astype("str")
+    str_month = df.month.astype("str")
+    str_both = str_year.str.cat(str_month, sep=" ")
+
+    assert str_both.loc[1] == "2011 2"
+
+    str_multiple = str_year.str.cat([str_month, str_month], sep=" ")
+
+    assert str_multiple.loc[1] == "2011 2 2"
+
+
+@pytest.mark.parametrize("klass", [tuple, list, np.array, Series, Index])
+def test_cat_different_classes(klass):
+    # https://github.com/pandas-dev/pandas/issues/33425
+    s = Series(["a", "b", "c"])
+    result = s.str.cat(klass(["x", "y", "z"]))
+    expected = Series(["ax", "by", "cz"])
+    tm.assert_series_equal(result, expected)
+
+
+def test_cat_on_series_dot_str():
+    # GH 28277
+    ps = Series(["AbC", "de", "FGHI", "j", "kLLLm"])
+
+    message = re.escape(
+        "others must be Series, Index, DataFrame, np.ndarray "
+        "or list-like (either containing only strings or "
+        "containing only objects of type Series/Index/"
+        "np.ndarray[1-dim])"
+    )
+    with pytest.raises(TypeError, match=message):
+        ps.str.cat(others=ps.str)
diff --git a/pandas/pandas/tests/strings/test_extract.py b/pandas/pandas/tests/strings/test_extract.py
new file mode 100644
index 0000000000000..1dc7dbabe85b0
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_extract.py
@@ -0,0 +1,723 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+from pandas.core.dtypes.dtypes import ArrowDtype
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+)
+
+
+def test_extract_expand_kwarg_wrong_type_raises(any_string_dtype):
+    # TODO: should this raise TypeError
+    values = Series(["fooBAD__barBAD", np.nan, "foo"], dtype=any_string_dtype)
+    with pytest.raises(ValueError, match="expand must be True or False"):
+        values.str.extract(".*(BAD[_]+).*(BAD)", expand=None)
+
+
+def test_extract_expand_kwarg(any_string_dtype):
+    s = Series(["fooBAD__barBAD", np.nan, "foo"], dtype=any_string_dtype)
+    expected = DataFrame(["BAD__", np.nan, np.nan], dtype=any_string_dtype)
+
+    result = s.str.extract(".*(BAD[_]+).*")
+    tm.assert_frame_equal(result, expected)
+
+    result = s.str.extract(".*(BAD[_]+).*", expand=True)
+    tm.assert_frame_equal(result, expected)
+
+    expected = DataFrame(
+        [["BAD__", "BAD"], [np.nan, np.nan], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    result = s.str.extract(".*(BAD[_]+).*(BAD)", expand=False)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_expand_False_mixed_object():
+    ser = Series(
+        ["aBAD_BAD", np.nan, "BAD_b_BAD", True, datetime.today(), "foo", None, 1, 2.0]
+    )
+
+    # two groups
+    result = ser.str.extract(".*(BAD[_]+).*(BAD)", expand=False)
+    er = [np.nan, np.nan]  # empty row
+    expected = DataFrame(
+        [["BAD_", "BAD"], er, ["BAD_", "BAD"], er, er, er, er, er, er], dtype=object
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # single group
+    result = ser.str.extract(".*(BAD[_]+).*BAD", expand=False)
+    expected = Series(
+        ["BAD_", np.nan, "BAD_", np.nan, np.nan, np.nan, None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_extract_expand_index_raises():
+    # GH9980
+    # Index only works with one regex group since
+    # multi-group would expand to a frame
+    idx = Index(["A1", "A2", "A3", "A4", "B5"])
+    msg = "only one regex group is supported with Index"
+    with pytest.raises(ValueError, match=msg):
+        idx.str.extract("([AB])([123])", expand=False)
+
+
+def test_extract_expand_no_capture_groups_raises(index_or_series, any_string_dtype):
+    s_or_idx = index_or_series(["A1", "B2", "C3"], dtype=any_string_dtype)
+    msg = "pattern contains no capture groups"
+
+    # no groups
+    with pytest.raises(ValueError, match=msg):
+        s_or_idx.str.extract("[ABC][123]", expand=False)
+
+    # only non-capturing groups
+    with pytest.raises(ValueError, match=msg):
+        s_or_idx.str.extract("(?:[AB]).*", expand=False)
+
+
+def test_extract_expand_single_capture_group(index_or_series, any_string_dtype):
+    # single group renames series/index properly
+    s_or_idx = index_or_series(["A1", "A2"], dtype=any_string_dtype)
+    result = s_or_idx.str.extract(r"(?P<uno>A)\d", expand=False)
+
+    expected = index_or_series(["A", "A"], name="uno", dtype=any_string_dtype)
+    if index_or_series == Series:
+        tm.assert_series_equal(result, expected)
+    else:
+        tm.assert_index_equal(result, expected)
+
+
+def test_extract_expand_capture_groups(any_string_dtype):
+    s = Series(["A1", "B2", "C3"], dtype=any_string_dtype)
+    # one group, no matches
+    result = s.str.extract("(_)", expand=False)
+    expected = Series([np.nan, np.nan, np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # two groups, no matches
+    result = s.str.extract("(_)(_)", expand=False)
+    expected = DataFrame(
+        [[np.nan, np.nan], [np.nan, np.nan], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one group, some matches
+    result = s.str.extract("([AB])[123]", expand=False)
+    expected = Series(["A", "B", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # two groups, some matches
+    result = s.str.extract("([AB])([123])", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one named group
+    result = s.str.extract("(?P<letter>[AB])", expand=False)
+    expected = Series(["A", "B", np.nan], name="letter", dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # two named groups
+    result = s.str.extract("(?P<letter>[AB])(?P<number>[123])", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # mix named and unnamed groups
+    result = s.str.extract("([AB])(?P<number>[123])", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]],
+        columns=[0, "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one normal group, one non-capturing group
+    result = s.str.extract("([AB])(?:[123])", expand=False)
+    expected = Series(["A", "B", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # two normal groups, one non-capturing group
+    s = Series(["A11", "B22", "C33"], dtype=any_string_dtype)
+    result = s.str.extract("([AB])([123])(?:[123])", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one optional group followed by one normal group
+    s = Series(["A1", "B2", "3"], dtype=any_string_dtype)
+    result = s.str.extract("(?P<letter>[AB])?(?P<number>[123])", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, "3"]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one normal group followed by one optional group
+    s = Series(["A1", "B2", "C"], dtype=any_string_dtype)
+    result = s.str.extract("(?P<letter>[ABC])(?P<number>[123])?", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], ["C", np.nan]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_expand_capture_groups_index(index, any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/6348
+    # not passing index to the extractor
+    data = ["A1", "B2", "C"]
+
+    if len(index) == 0:
+        pytest.skip("Test requires len(index) > 0")
+    while len(index) < len(data):
+        index = index.repeat(2)
+
+    index = index[: len(data)]
+    ser = Series(data, index=index, dtype=any_string_dtype)
+
+    result = ser.str.extract(r"(\d)", expand=False)
+    expected = Series(["1", "2", np.nan], index=index, dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.extract(r"(?P<letter>\D)(?P<number>\d)?", expand=False)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], ["C", np.nan]],
+        columns=["letter", "number"],
+        index=index,
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_single_series_name_is_preserved(any_string_dtype):
+    s = Series(["a3", "b3", "c2"], name="bob", dtype=any_string_dtype)
+    result = s.str.extract(r"(?P<sue>[a-z])", expand=False)
+    expected = Series(["a", "b", "c"], name="sue", dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_extract_expand_True(any_string_dtype):
+    # Contains tests like those in test_match and some others.
+    s = Series(["fooBAD__barBAD", np.nan, "foo"], dtype=any_string_dtype)
+
+    result = s.str.extract(".*(BAD[_]+).*(BAD)", expand=True)
+    expected = DataFrame(
+        [["BAD__", "BAD"], [np.nan, np.nan], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_expand_True_mixed_object():
+    er = [np.nan, np.nan]  # empty row
+    mixed = Series(
+        [
+            "aBAD_BAD",
+            np.nan,
+            "BAD_b_BAD",
+            True,
+            datetime.today(),
+            "foo",
+            None,
+            1,
+            2.0,
+        ]
+    )
+
+    result = mixed.str.extract(".*(BAD[_]+).*(BAD)", expand=True)
+    expected = DataFrame(
+        [["BAD_", "BAD"], er, ["BAD_", "BAD"], er, er, er, er, er, er], dtype=object
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_expand_True_single_capture_group_raises(
+    index_or_series, any_string_dtype
+):
+    # these should work for both Series and Index
+    # no groups
+    s_or_idx = index_or_series(["A1", "B2", "C3"], dtype=any_string_dtype)
+    msg = "pattern contains no capture groups"
+    with pytest.raises(ValueError, match=msg):
+        s_or_idx.str.extract("[ABC][123]", expand=True)
+
+    # only non-capturing groups
+    with pytest.raises(ValueError, match=msg):
+        s_or_idx.str.extract("(?:[AB]).*", expand=True)
+
+
+def test_extract_expand_True_single_capture_group(index_or_series, any_string_dtype):
+    # single group renames series/index properly
+    s_or_idx = index_or_series(["A1", "A2"], dtype=any_string_dtype)
+    result = s_or_idx.str.extract(r"(?P<uno>A)\d", expand=True)
+    expected = DataFrame({"uno": ["A", "A"]}, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("name", [None, "series_name"])
+def test_extract_series(name, any_string_dtype):
+    # extract should give the same result whether or not the series has a name.
+    s = Series(["A1", "B2", "C3"], name=name, dtype=any_string_dtype)
+
+    # one group, no matches
+    result = s.str.extract("(_)", expand=True)
+    expected = DataFrame([np.nan, np.nan, np.nan], dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+    # two groups, no matches
+    result = s.str.extract("(_)(_)", expand=True)
+    expected = DataFrame(
+        [[np.nan, np.nan], [np.nan, np.nan], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one group, some matches
+    result = s.str.extract("([AB])[123]", expand=True)
+    expected = DataFrame(["A", "B", np.nan], dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+    # two groups, some matches
+    result = s.str.extract("([AB])([123])", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one named group
+    result = s.str.extract("(?P<letter>[AB])", expand=True)
+    expected = DataFrame({"letter": ["A", "B", np.nan]}, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+    # two named groups
+    result = s.str.extract("(?P<letter>[AB])(?P<number>[123])", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # mix named and unnamed groups
+    result = s.str.extract("([AB])(?P<number>[123])", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]],
+        columns=[0, "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one normal group, one non-capturing group
+    result = s.str.extract("([AB])(?:[123])", expand=True)
+    expected = DataFrame(["A", "B", np.nan], dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_optional_groups(any_string_dtype):
+    # two normal groups, one non-capturing group
+    s = Series(["A11", "B22", "C33"], dtype=any_string_dtype)
+    result = s.str.extract("([AB])([123])(?:[123])", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one optional group followed by one normal group
+    s = Series(["A1", "B2", "3"], dtype=any_string_dtype)
+    result = s.str.extract("(?P<letter>[AB])?(?P<number>[123])", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], [np.nan, "3"]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one normal group followed by one optional group
+    s = Series(["A1", "B2", "C"], dtype=any_string_dtype)
+    result = s.str.extract("(?P<letter>[ABC])(?P<number>[123])?", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], ["C", np.nan]],
+        columns=["letter", "number"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_dataframe_capture_groups_index(index, any_string_dtype):
+    # GH6348
+    # not passing index to the extractor
+
+    data = ["A1", "B2", "C"]
+
+    if len(index) < len(data):
+        pytest.skip(f"Index needs more than {len(data)} values")
+
+    index = index[: len(data)]
+    s = Series(data, index=index, dtype=any_string_dtype)
+
+    result = s.str.extract(r"(\d)", expand=True)
+    expected = DataFrame(["1", "2", np.nan], index=index, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+    result = s.str.extract(r"(?P<letter>\D)(?P<number>\d)?", expand=True)
+    expected = DataFrame(
+        [["A", "1"], ["B", "2"], ["C", np.nan]],
+        columns=["letter", "number"],
+        index=index,
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extract_single_group_returns_frame(any_string_dtype):
+    # GH11386 extract should always return DataFrame, even when
+    # there is only one group. Prior to v0.18.0, extract returned
+    # Series when there was only one group in the regex.
+    s = Series(["a3", "b3", "c2"], name="series_name", dtype=any_string_dtype)
+    result = s.str.extract(r"(?P<letter>[a-z])", expand=True)
+    expected = DataFrame({"letter": ["a", "b", "c"]}, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extractall(any_string_dtype):
+    data = [
+        "dave@google.com",
+        "tdhock5@gmail.com",
+        "maudelaperriere@gmail.com",
+        "rob@gmail.com some text steve@gmail.com",
+        "a@b.com some text c@d.com and e@f.com",
+        np.nan,
+        "",
+    ]
+    expected_tuples = [
+        ("dave", "google", "com"),
+        ("tdhock5", "gmail", "com"),
+        ("maudelaperriere", "gmail", "com"),
+        ("rob", "gmail", "com"),
+        ("steve", "gmail", "com"),
+        ("a", "b", "com"),
+        ("c", "d", "com"),
+        ("e", "f", "com"),
+    ]
+    pat = r"""
+    (?P<user>[a-z0-9]+)
+    @
+    (?P<domain>[a-z]+)
+    \.
+    (?P<tld>[a-z]{2,4})
+    """
+    expected_columns = ["user", "domain", "tld"]
+    s = Series(data, dtype=any_string_dtype)
+    # extractall should return a DataFrame with one row for each match, indexed by the
+    # subject from which the match came.
+    expected_index = MultiIndex.from_tuples(
+        [(0, 0), (1, 0), (2, 0), (3, 0), (3, 1), (4, 0), (4, 1), (4, 2)],
+        names=(None, "match"),
+    )
+    expected = DataFrame(
+        expected_tuples, expected_index, expected_columns, dtype=any_string_dtype
+    )
+    result = s.str.extractall(pat, flags=re.VERBOSE)
+    tm.assert_frame_equal(result, expected)
+
+    # The index of the input Series should be used to construct the index of the output
+    # DataFrame:
+    mi = MultiIndex.from_tuples(
+        [
+            ("single", "Dave"),
+            ("single", "Toby"),
+            ("single", "Maude"),
+            ("multiple", "robAndSteve"),
+            ("multiple", "abcdef"),
+            ("none", "missing"),
+            ("none", "empty"),
+        ]
+    )
+    s = Series(data, index=mi, dtype=any_string_dtype)
+    expected_index = MultiIndex.from_tuples(
+        [
+            ("single", "Dave", 0),
+            ("single", "Toby", 0),
+            ("single", "Maude", 0),
+            ("multiple", "robAndSteve", 0),
+            ("multiple", "robAndSteve", 1),
+            ("multiple", "abcdef", 0),
+            ("multiple", "abcdef", 1),
+            ("multiple", "abcdef", 2),
+        ],
+        names=(None, None, "match"),
+    )
+    expected = DataFrame(
+        expected_tuples, expected_index, expected_columns, dtype=any_string_dtype
+    )
+    result = s.str.extractall(pat, flags=re.VERBOSE)
+    tm.assert_frame_equal(result, expected)
+
+    # MultiIndexed subject with names.
+    s = Series(data, index=mi, dtype=any_string_dtype)
+    s.index.names = ("matches", "description")
+    expected_index.names = ("matches", "description", "match")
+    expected = DataFrame(
+        expected_tuples, expected_index, expected_columns, dtype=any_string_dtype
+    )
+    result = s.str.extractall(pat, flags=re.VERBOSE)
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "pat,expected_names",
+    [
+        # optional groups.
+        ("(?P<letter>[AB])?(?P<number>[123])", ["letter", "number"]),
+        # only one of two groups has a name.
+        ("([AB])?(?P<number>[123])", [0, "number"]),
+    ],
+)
+def test_extractall_column_names(pat, expected_names, any_string_dtype):
+    s = Series(["", "A1", "32"], dtype=any_string_dtype)
+
+    result = s.str.extractall(pat)
+    expected = DataFrame(
+        [("A", "1"), (np.nan, "3"), (np.nan, "2")],
+        index=MultiIndex.from_tuples([(1, 0), (2, 0), (2, 1)], names=(None, "match")),
+        columns=expected_names,
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extractall_single_group(any_string_dtype):
+    s = Series(["a3", "b3", "d4c2"], name="series_name", dtype=any_string_dtype)
+    expected_index = MultiIndex.from_tuples(
+        [(0, 0), (1, 0), (2, 0), (2, 1)], names=(None, "match")
+    )
+
+    # extractall(one named group) returns DataFrame with one named column.
+    result = s.str.extractall(r"(?P<letter>[a-z])")
+    expected = DataFrame(
+        {"letter": ["a", "b", "d", "c"]}, index=expected_index, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # extractall(one un-named group) returns DataFrame with one un-named column.
+    result = s.str.extractall(r"([a-z])")
+    expected = DataFrame(
+        ["a", "b", "d", "c"], index=expected_index, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extractall_single_group_with_quantifier(any_string_dtype):
+    # GH#13382
+    # extractall(one un-named group with quantifier) returns DataFrame with one un-named
+    # column.
+    s = Series(["ab3", "abc3", "d4cd2"], name="series_name", dtype=any_string_dtype)
+    result = s.str.extractall(r"([a-z]+)")
+    expected = DataFrame(
+        ["ab", "abc", "d", "cd"],
+        index=MultiIndex.from_tuples(
+            [(0, 0), (1, 0), (2, 0), (2, 1)], names=(None, "match")
+        ),
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, names",
+    [
+        ([], (None,)),
+        ([], ("i1",)),
+        ([], (None, "i2")),
+        ([], ("i1", "i2")),
+        (["a3", "b3", "d4c2"], (None,)),
+        (["a3", "b3", "d4c2"], ("i1", "i2")),
+        (["a3", "b3", "d4c2"], (None, "i2")),
+    ],
+)
+def test_extractall_no_matches(data, names, any_string_dtype):
+    # GH19075 extractall with no matches should return a valid MultiIndex
+    n = len(data)
+    if len(names) == 1:
+        index = Index(range(n), name=names[0])
+    else:
+        tuples = (tuple([i] * (n - 1)) for i in range(n))
+        index = MultiIndex.from_tuples(tuples, names=names)
+    s = Series(data, name="series_name", index=index, dtype=any_string_dtype)
+    expected_index = MultiIndex.from_tuples([], names=(names + ("match",)))
+
+    # one un-named group.
+    result = s.str.extractall("(z)")
+    expected = DataFrame(columns=range(1), index=expected_index, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+    # two un-named groups.
+    result = s.str.extractall("(z)(z)")
+    expected = DataFrame(columns=range(2), index=expected_index, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, expected, check_column_type=True)
+
+    # one named group.
+    result = s.str.extractall("(?P<first>z)")
+    expected = DataFrame(
+        columns=["first"], index=expected_index, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # two named groups.
+    result = s.str.extractall("(?P<first>z)(?P<second>z)")
+    expected = DataFrame(
+        columns=["first", "second"], index=expected_index, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # one named, one un-named.
+    result = s.str.extractall("(z)(?P<second>z)")
+    expected = DataFrame(
+        columns=[0, "second"], index=expected_index, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extractall_stringindex(any_string_dtype):
+    s = Series(["a1a2", "b1", "c1"], name="xxx", dtype=any_string_dtype)
+    result = s.str.extractall(r"[ab](?P<digit>\d)")
+    expected = DataFrame(
+        {"digit": ["1", "2", "1"]},
+        index=MultiIndex.from_tuples([(0, 0), (0, 1), (1, 0)], names=[None, "match"]),
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+    # index should return the same result as the default index without name thus
+    # index.name doesn't affect to the result
+    if any_string_dtype == "object":
+        for idx in [
+            Index(["a1a2", "b1", "c1"], dtype=object),
+            Index(["a1a2", "b1", "c1"], name="xxx", dtype=object),
+        ]:
+            result = idx.str.extractall(r"[ab](?P<digit>\d)")
+            tm.assert_frame_equal(result, expected)
+
+    s = Series(
+        ["a1a2", "b1", "c1"],
+        name="s_name",
+        index=Index(["XX", "yy", "zz"], name="idx_name"),
+        dtype=any_string_dtype,
+    )
+    result = s.str.extractall(r"[ab](?P<digit>\d)")
+    expected = DataFrame(
+        {"digit": ["1", "2", "1"]},
+        index=MultiIndex.from_tuples(
+            [("XX", 0), ("XX", 1), ("yy", 0)], names=["idx_name", "match"]
+        ),
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_extractall_no_capture_groups_raises(any_string_dtype):
+    # Does not make sense to use extractall with a regex that has no capture groups.
+    # (it returns DataFrame with one column for each capture group)
+    s = Series(["a3", "b3", "d4c2"], name="series_name", dtype=any_string_dtype)
+    with pytest.raises(ValueError, match="no capture groups"):
+        s.str.extractall(r"[a-z]")
+
+
+def test_extract_index_one_two_groups():
+    s = Series(["a3", "b3", "d4c2"], index=["A3", "B3", "D4"], name="series_name")
+    r = s.index.str.extract(r"([A-Z])", expand=True)
+    e = DataFrame(["A", "B", "D"])
+    tm.assert_frame_equal(r, e)
+
+    # Prior to v0.18.0, index.str.extract(regex with one group)
+    # returned Index. With more than one group, extract raised an
+    # error (GH9980). Now extract always returns DataFrame.
+    r = s.index.str.extract(r"(?P<letter>[A-Z])(?P<digit>[0-9])", expand=True)
+    e_list = [("A", "3"), ("B", "3"), ("D", "4")]
+    e = DataFrame(e_list, columns=["letter", "digit"])
+    tm.assert_frame_equal(r, e)
+
+
+def test_extractall_same_as_extract(any_string_dtype):
+    s = Series(["a3", "b3", "c2"], name="series_name", dtype=any_string_dtype)
+
+    pattern_two_noname = r"([a-z])([0-9])"
+    extract_two_noname = s.str.extract(pattern_two_noname, expand=True)
+    has_multi_index = s.str.extractall(pattern_two_noname)
+    no_multi_index = has_multi_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_two_noname, no_multi_index)
+
+    pattern_two_named = r"(?P<letter>[a-z])(?P<digit>[0-9])"
+    extract_two_named = s.str.extract(pattern_two_named, expand=True)
+    has_multi_index = s.str.extractall(pattern_two_named)
+    no_multi_index = has_multi_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_two_named, no_multi_index)
+
+    pattern_one_named = r"(?P<group_name>[a-z])"
+    extract_one_named = s.str.extract(pattern_one_named, expand=True)
+    has_multi_index = s.str.extractall(pattern_one_named)
+    no_multi_index = has_multi_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_one_named, no_multi_index)
+
+    pattern_one_noname = r"([a-z])"
+    extract_one_noname = s.str.extract(pattern_one_noname, expand=True)
+    has_multi_index = s.str.extractall(pattern_one_noname)
+    no_multi_index = has_multi_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_one_noname, no_multi_index)
+
+
+def test_extractall_same_as_extract_subject_index(any_string_dtype):
+    # same as above tests, but s has an MultiIndex.
+    mi = MultiIndex.from_tuples(
+        [("A", "first"), ("B", "second"), ("C", "third")],
+        names=("capital", "ordinal"),
+    )
+    s = Series(["a3", "b3", "c2"], index=mi, name="series_name", dtype=any_string_dtype)
+
+    pattern_two_noname = r"([a-z])([0-9])"
+    extract_two_noname = s.str.extract(pattern_two_noname, expand=True)
+    has_match_index = s.str.extractall(pattern_two_noname)
+    no_match_index = has_match_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_two_noname, no_match_index)
+
+    pattern_two_named = r"(?P<letter>[a-z])(?P<digit>[0-9])"
+    extract_two_named = s.str.extract(pattern_two_named, expand=True)
+    has_match_index = s.str.extractall(pattern_two_named)
+    no_match_index = has_match_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_two_named, no_match_index)
+
+    pattern_one_named = r"(?P<group_name>[a-z])"
+    extract_one_named = s.str.extract(pattern_one_named, expand=True)
+    has_match_index = s.str.extractall(pattern_one_named)
+    no_match_index = has_match_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_one_named, no_match_index)
+
+    pattern_one_noname = r"([a-z])"
+    extract_one_noname = s.str.extract(pattern_one_noname, expand=True)
+    has_match_index = s.str.extractall(pattern_one_noname)
+    no_match_index = has_match_index.xs(0, level="match")
+    tm.assert_frame_equal(extract_one_noname, no_match_index)
+
+
+def test_extractall_preserves_dtype():
+    # Ensure that when extractall is called on a series with specific dtypes set, that
+    # the dtype is preserved in the resulting DataFrame's column.
+    pa = pytest.importorskip("pyarrow")
+
+    result = Series(["abc", "ab"], dtype=ArrowDtype(pa.string())).str.extractall("(ab)")
+    assert result.dtypes[0] == "string[pyarrow]"
diff --git a/pandas/pandas/tests/strings/test_find_replace.py b/pandas/pandas/tests/strings/test_find_replace.py
new file mode 100644
index 0000000000000..30e6ebf0eed13
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_find_replace.py
@@ -0,0 +1,1089 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    Series,
+    _testing as tm,
+)
+from pandas.tests.strings import (
+    _convert_na_value,
+    is_object_or_nan_string_dtype,
+)
+
+# --------------------------------------------------------------------------------------
+# str.contains
+# --------------------------------------------------------------------------------------
+
+
+def test_contains(any_string_dtype):
+    values = np.array(
+        ["foo", np.nan, "fooommm__foo", "mmm_", "foommm[_]+bar"], dtype=np.object_
+    )
+    values = Series(values, dtype=any_string_dtype)
+    pat = "mmm[_]+"
+
+    result = values.str.contains(pat)
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected = Series([False, False, True, True, False], dtype=bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series(
+            np.array([False, np.nan, True, True, False], dtype=np.object_),
+            dtype=expected_dtype,
+        )
+
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.contains(pat, regex=False)
+    if any_string_dtype == "str":
+        expected = Series([False, False, False, False, True], dtype=bool)
+    else:
+        expected = Series(
+            np.array([False, np.nan, False, False, True], dtype=np.object_),
+            dtype=expected_dtype,
+        )
+    tm.assert_series_equal(result, expected)
+
+    values = Series(
+        np.array(["foo", "xyz", "fooommm__foo", "mmm_"], dtype=object),
+        dtype=any_string_dtype,
+    )
+    result = values.str.contains(pat)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series(np.array([False, False, True, True]), dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # case insensitive using regex
+    values = Series(
+        np.array(["Foo", "xYz", "fOOomMm__fOo", "MMM_"], dtype=object),
+        dtype=any_string_dtype,
+    )
+
+    result = values.str.contains("FOO|mmm", case=False)
+    expected = Series(np.array([True, False, True, True]), dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # case insensitive without regex
+    result = values.str.contains("foo", regex=False, case=False)
+    expected = Series(np.array([True, False, True, False]), dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    # unicode
+    values = Series(
+        np.array(["foo", np.nan, "fooommm__foo", "mmm_"], dtype=np.object_),
+        dtype=any_string_dtype,
+    )
+    pat = "mmm[_]+"
+
+    result = values.str.contains(pat)
+    if any_string_dtype == "str":
+        expected = Series([False, False, True, True], dtype=bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series(
+            np.array([False, np.nan, True, True], dtype=np.object_),
+            dtype=expected_dtype,
+        )
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.contains(pat, na=False)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series(np.array([False, False, True, True]), dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    values = Series(
+        np.array(["foo", "xyz", "fooommm__foo", "mmm_"], dtype=np.object_),
+        dtype=any_string_dtype,
+    )
+    result = values.str.contains(pat)
+    expected = Series(np.array([False, False, True, True]), dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_contains_object_mixed():
+    mixed = Series(
+        np.array(
+            ["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0],
+            dtype=object,
+        )
+    )
+    result = mixed.str.contains("o")
+    expected = Series(
+        np.array(
+            [False, np.nan, False, np.nan, np.nan, True, None, np.nan, np.nan],
+            dtype=np.object_,
+        )
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_contains_na_kwarg_for_object_category():
+    # gh 22158
+
+    # na for category
+    values = Series(["a", "b", "c", "a", np.nan], dtype="category")
+    result = values.str.contains("a", na=True)
+    expected = Series([True, False, False, True, True])
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.contains("a", na=False)
+    expected = Series([True, False, False, True, False])
+    tm.assert_series_equal(result, expected)
+
+    # na for objects
+    values = Series(["a", "b", "c", "a", np.nan])
+    result = values.str.contains("a", na=True)
+    expected = Series([True, False, False, True, True])
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.contains("a", na=False)
+    expected = Series([True, False, False, True, False])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "na, expected",
+    [
+        (None, pd.NA),
+        (True, True),
+        (False, False),
+        (0, False),
+        (3, True),
+        (np.nan, pd.NA),
+    ],
+)
+@pytest.mark.parametrize("regex", [True, False])
+def test_contains_na_kwarg_for_nullable_string_dtype(
+    nullable_string_dtype, na, expected, regex
+):
+    # https://github.com/pandas-dev/pandas/pull/41025#issuecomment-824062416
+
+    values = Series(["a", "b", "c", "a", np.nan], dtype=nullable_string_dtype)
+
+    msg = (
+        "Allowing a non-bool 'na' in obj.str.contains is deprecated and "
+        "will raise in a future version"
+    )
+    warn = None
+    if not pd.isna(na) and not isinstance(na, bool):
+        warn = FutureWarning
+    with tm.assert_produces_warning(warn, match=msg):
+        result = values.str.contains("a", na=na, regex=regex)
+    expected = Series([True, False, False, True, expected], dtype="boolean")
+    tm.assert_series_equal(result, expected)
+
+
+def test_contains_moar(any_string_dtype):
+    # PR #1179
+    s = Series(
+        ["A", "B", "C", "Aaba", "Baca", "", np.nan, "CABA", "dog", "cat"],
+        dtype=any_string_dtype,
+    )
+
+    result = s.str.contains("a")
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected_dtype = bool
+        na_value = False
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        na_value = np.nan
+    expected = Series(
+        [False, False, False, True, True, False, na_value, False, False, True],
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("a", case=False)
+    expected = Series(
+        [True, False, False, True, True, False, na_value, True, False, True],
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("Aa")
+    expected = Series(
+        [False, False, False, True, False, False, na_value, False, False, False],
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("ba")
+    expected = Series(
+        [False, False, False, True, False, False, na_value, False, False, False],
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("ba", case=False)
+    expected = Series(
+        [False, False, False, True, True, False, na_value, True, False, False],
+        dtype=expected_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_contains_nan(any_string_dtype):
+    # PR #14171
+    s = Series([np.nan, np.nan, np.nan], dtype=any_string_dtype)
+
+    result = s.str.contains("foo", na=False)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series([False, False, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("foo", na=True)
+    expected = Series([True, True, True], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    msg = (
+        "Allowing a non-bool 'na' in obj.str.contains is deprecated and "
+        "will raise in a future version"
+    )
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = s.str.contains("foo", na="foo")
+    if any_string_dtype == "object":
+        expected = Series(["foo", "foo", "foo"], dtype=np.object_)
+    elif any_string_dtype.na_value is np.nan:
+        expected = Series([True, True, True], dtype=np.bool_)
+    else:
+        expected = Series([True, True, True], dtype="boolean")
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.contains("foo")
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected = Series([False, False, False], dtype=bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series([np.nan, np.nan, np.nan], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.startswith
+# --------------------------------------------------------------------------------------
+
+
+def test_startswith_endswith_validate_na(any_string_dtype):
+    # GH#59615
+    ser = Series(
+        ["om", np.nan, "foo_nom", "nom", "bar_foo", np.nan, "foo"],
+        dtype=any_string_dtype,
+    )
+
+    msg = "Allowing a non-bool 'na' in obj.str.startswith is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        ser.str.startswith("kapow", na="baz")
+    msg = "Allowing a non-bool 'na' in obj.str.endswith is deprecated"
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        ser.str.endswith("bar", na="baz")
+
+
+@pytest.mark.parametrize("pat", ["foo", ("foo", "baz")])
+@pytest.mark.parametrize("dtype", ["object", "category"])
+@pytest.mark.parametrize("null_value", [None, np.nan, pd.NA])
+@pytest.mark.parametrize("na", [True, False])
+def test_startswith(pat, dtype, null_value, na, using_infer_string):
+    # add category dtype parametrizations for GH-36241
+    values = Series(
+        ["om", null_value, "foo_nom", "nom", "bar_foo", null_value, "foo"],
+        dtype=dtype,
+    )
+
+    result = values.str.startswith(pat)
+    exp = Series([False, np.nan, True, False, False, np.nan, True])
+    if dtype == "object" and null_value is pd.NA:
+        # GH#18463
+        exp = exp.fillna(null_value)
+    elif dtype == "object" and null_value is None:
+        exp[exp.isna()] = None
+    elif using_infer_string and dtype == "category":
+        exp = exp.fillna(False).astype(bool)
+    tm.assert_series_equal(result, exp)
+
+    result = values.str.startswith(pat, na=na)
+    exp = Series([False, na, True, False, False, na, True])
+    tm.assert_series_equal(result, exp)
+
+    # mixed
+    mixed = np.array(
+        ["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0],
+        dtype=np.object_,
+    )
+    rs = Series(mixed).str.startswith("f")
+    xp = Series([False, np.nan, False, np.nan, np.nan, True, None, np.nan, np.nan])
+    tm.assert_series_equal(rs, xp)
+
+
+@pytest.mark.parametrize("na", [None, True, False])
+def test_startswith_string_dtype(any_string_dtype, na):
+    values = Series(
+        ["om", None, "foo_nom", "nom", "bar_foo", None, "foo", "regex", "rege."],
+        dtype=any_string_dtype,
+    )
+    result = values.str.startswith("foo", na=na)
+
+    expected_dtype = (
+        (object if na is None else bool)
+        if is_object_or_nan_string_dtype(any_string_dtype)
+        else "boolean"
+    )
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected_dtype = bool
+        if na is None:
+            na = False
+    exp = Series(
+        [False, na, True, False, False, na, True, False, False], dtype=expected_dtype
+    )
+    tm.assert_series_equal(result, exp)
+
+    result = values.str.startswith("rege.", na=na)
+    exp = Series(
+        [False, na, False, False, False, na, False, False, True], dtype=expected_dtype
+    )
+    tm.assert_series_equal(result, exp)
+
+
+# --------------------------------------------------------------------------------------
+# str.endswith
+# --------------------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize("pat", ["foo", ("foo", "baz")])
+@pytest.mark.parametrize("dtype", ["object", "category"])
+@pytest.mark.parametrize("null_value", [None, np.nan, pd.NA])
+@pytest.mark.parametrize("na", [True, False])
+def test_endswith(pat, dtype, null_value, na, using_infer_string):
+    # add category dtype parametrizations for GH-36241
+    values = Series(
+        ["om", null_value, "foo_nom", "nom", "bar_foo", null_value, "foo"],
+        dtype=dtype,
+    )
+
+    result = values.str.endswith(pat)
+    exp = Series([False, np.nan, False, False, True, np.nan, True])
+    if dtype == "object" and null_value is pd.NA:
+        # GH#18463
+        exp = exp.fillna(null_value)
+    elif dtype == "object" and null_value is None:
+        exp[exp.isna()] = None
+    elif using_infer_string and dtype == "category":
+        exp = exp.fillna(False).astype(bool)
+    tm.assert_series_equal(result, exp)
+
+    result = values.str.endswith(pat, na=na)
+    exp = Series([False, na, False, False, True, na, True])
+    tm.assert_series_equal(result, exp)
+
+    # mixed
+    mixed = np.array(
+        ["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0],
+        dtype=object,
+    )
+    rs = Series(mixed).str.endswith("f")
+    xp = Series([False, np.nan, False, np.nan, np.nan, False, None, np.nan, np.nan])
+    tm.assert_series_equal(rs, xp)
+
+
+@pytest.mark.parametrize("na", [None, True, False])
+def test_endswith_string_dtype(any_string_dtype, na):
+    values = Series(
+        ["om", None, "foo_nom", "nom", "bar_foo", None, "foo", "regex", "rege."],
+        dtype=any_string_dtype,
+    )
+    result = values.str.endswith("foo", na=na)
+    expected_dtype = (
+        (object if na is None else bool)
+        if is_object_or_nan_string_dtype(any_string_dtype)
+        else "boolean"
+    )
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected_dtype = bool
+        if na is None:
+            na = False
+    exp = Series(
+        [False, na, False, False, True, na, True, False, False], dtype=expected_dtype
+    )
+    tm.assert_series_equal(result, exp)
+
+    result = values.str.endswith("rege.", na=na)
+    exp = Series(
+        [False, na, False, False, False, na, False, False, True], dtype=expected_dtype
+    )
+    tm.assert_series_equal(result, exp)
+
+
+# --------------------------------------------------------------------------------------
+# str.replace
+# --------------------------------------------------------------------------------------
+def test_replace_dict_invalid(any_string_dtype):
+    # GH 51914
+    series = Series(data=["A", "B_junk", "C_gunk"], name="my_messy_col")
+    msg = "repl cannot be used when pat is a dictionary"
+
+    with pytest.raises(ValueError, match=msg):
+        series.str.replace(pat={"A": "a", "B": "b"}, repl="A")
+
+
+def test_replace_dict(any_string_dtype):
+    # GH 51914
+    series = Series(data=["A", "B", "C"], name="my_messy_col")
+    new_series = series.str.replace(pat={"A": "a", "B": "b"})
+    expected = Series(data=["a", "b", "C"], name="my_messy_col")
+    tm.assert_series_equal(new_series, expected)
+
+
+def test_replace(any_string_dtype):
+    ser = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+
+    result = ser.str.replace("BAD[_]*", "", regex=True)
+    expected = Series(["foobar", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_max_replacements(any_string_dtype):
+    ser = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+
+    expected = Series(["foobarBAD", np.nan], dtype=any_string_dtype)
+    result = ser.str.replace("BAD[_]*", "", n=1, regex=True)
+    tm.assert_series_equal(result, expected)
+
+    expected = Series(["foo__barBAD", np.nan], dtype=any_string_dtype)
+    result = ser.str.replace("BAD", "", n=1, regex=False)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_mixed_object():
+    ser = Series(
+        ["aBAD", np.nan, "bBAD", True, datetime.today(), "fooBAD", None, 1, 2.0]
+    )
+    result = Series(ser).str.replace("BAD[_]*", "", regex=True)
+    expected = Series(
+        ["a", np.nan, "b", np.nan, np.nan, "foo", None, np.nan, np.nan], dtype=object
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_unicode(any_string_dtype):
+    ser = Series([b"abcd,\xc3\xa0".decode("utf-8")], dtype=any_string_dtype)
+    expected = Series([b"abcd, \xc3\xa0".decode("utf-8")], dtype=any_string_dtype)
+    result = ser.str.replace(r"(?<=\w),(?=\w)", ", ", flags=re.UNICODE, regex=True)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("repl", [None, 3, {"a": "b"}])
+@pytest.mark.parametrize("data", [["a", "b", None], ["a", "b", "c", "ad"]])
+def test_replace_wrong_repl_type_raises(any_string_dtype, index_or_series, repl, data):
+    # https://github.com/pandas-dev/pandas/issues/13438
+    msg = "repl must be a string or callable"
+    obj = index_or_series(data, dtype=any_string_dtype)
+    with pytest.raises(TypeError, match=msg):
+        obj.str.replace("a", repl)
+
+
+def test_replace_callable(any_string_dtype):
+    # GH 15055
+    ser = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+
+    # test with callable
+    repl = lambda m: m.group(0).swapcase()
+    result = ser.str.replace("[a-z][A-Z]{2}", repl, n=2, regex=True)
+    expected = Series(["foObaD__baRbaD", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "repl", [lambda: None, lambda m, x: None, lambda m, x, y=None: None]
+)
+def test_replace_callable_raises(any_string_dtype, repl):
+    # GH 15055
+    values = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+
+    # test with wrong number of arguments, raising an error
+    msg = (
+        r"((takes)|(missing)) (?(2)from \d+ to )?\d+ "
+        r"(?(3)required )positional arguments?"
+    )
+    with pytest.raises(TypeError, match=msg):
+        values.str.replace("a", repl, regex=True)
+
+
+def test_replace_callable_named_groups(any_string_dtype):
+    # test regex named groups
+    ser = Series(["Foo Bar Baz", np.nan], dtype=any_string_dtype)
+    pat = r"(?P<first>\w+) (?P<middle>\w+) (?P<last>\w+)"
+    repl = lambda m: m.group("middle").swapcase()
+    result = ser.str.replace(pat, repl, regex=True)
+    expected = Series(["bAR", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_compiled_regex(any_string_dtype):
+    # GH 15446
+    ser = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+
+    # test with compiled regex
+    pat = re.compile(r"BAD_*")
+    result = ser.str.replace(pat, "", regex=True)
+    expected = Series(["foobar", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.replace(pat, "", n=1, regex=True)
+    expected = Series(["foobarBAD", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_compiled_regex_mixed_object():
+    pat = re.compile(r"BAD_*")
+    ser = Series(
+        ["aBAD", np.nan, "bBAD", True, datetime.today(), "fooBAD", None, 1, 2.0]
+    )
+    result = Series(ser).str.replace(pat, "", regex=True)
+    expected = Series(
+        ["a", np.nan, "b", np.nan, np.nan, "foo", None, np.nan, np.nan], dtype=object
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_compiled_regex_unicode(any_string_dtype):
+    ser = Series([b"abcd,\xc3\xa0".decode("utf-8")], dtype=any_string_dtype)
+    expected = Series([b"abcd, \xc3\xa0".decode("utf-8")], dtype=any_string_dtype)
+    pat = re.compile(r"(?<=\w),(?=\w)", flags=re.UNICODE)
+    result = ser.str.replace(pat, ", ", regex=True)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_compiled_regex_raises(any_string_dtype):
+    # case and flags provided to str.replace will have no effect
+    # and will produce warnings
+    ser = Series(["fooBAD__barBAD__bad", np.nan], dtype=any_string_dtype)
+    pat = re.compile(r"BAD_*")
+
+    msg = "case and flags cannot be set when pat is a compiled regex"
+
+    with pytest.raises(ValueError, match=msg):
+        ser.str.replace(pat, "", flags=re.IGNORECASE, regex=True)
+
+    with pytest.raises(ValueError, match=msg):
+        ser.str.replace(pat, "", case=False, regex=True)
+
+    with pytest.raises(ValueError, match=msg):
+        ser.str.replace(pat, "", case=True, regex=True)
+
+
+def test_replace_compiled_regex_callable(any_string_dtype):
+    # test with callable
+    ser = Series(["fooBAD__barBAD", np.nan], dtype=any_string_dtype)
+    repl = lambda m: m.group(0).swapcase()
+    pat = re.compile("[a-z][A-Z]{2}")
+    result = ser.str.replace(pat, repl, n=2, regex=True)
+    expected = Series(["foObaD__baRbaD", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("regex,expected_val", [(True, "bao"), (False, "foo")])
+def test_replace_literal(regex, expected_val, any_string_dtype):
+    # GH16808 literal replace (regex=False vs regex=True)
+    ser = Series(["f.o", "foo", np.nan], dtype=any_string_dtype)
+    expected = Series(["bao", expected_val, np.nan], dtype=any_string_dtype)
+    result = ser.str.replace("f.", "ba", regex=regex)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_literal_callable_raises(any_string_dtype):
+    ser = Series([], dtype=any_string_dtype)
+    repl = lambda m: m.group(0).swapcase()
+
+    msg = "Cannot use a callable replacement when regex=False"
+    with pytest.raises(ValueError, match=msg):
+        ser.str.replace("abc", repl, regex=False)
+
+
+def test_replace_literal_compiled_raises(any_string_dtype):
+    ser = Series([], dtype=any_string_dtype)
+    pat = re.compile("[a-z][A-Z]{2}")
+
+    msg = "Cannot use a compiled regex as replacement pattern with regex=False"
+    with pytest.raises(ValueError, match=msg):
+        ser.str.replace(pat, "", regex=False)
+
+
+def test_replace_moar(any_string_dtype):
+    # PR #1179
+    ser = Series(
+        ["A", "B", "C", "Aaba", "Baca", "", np.nan, "CABA", "dog", "cat"],
+        dtype=any_string_dtype,
+    )
+
+    result = ser.str.replace("A", "YYY")
+    expected = Series(
+        ["YYY", "B", "C", "YYYaba", "Baca", "", np.nan, "CYYYBYYY", "dog", "cat"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.replace("A", "YYY", case=False)
+    expected = Series(
+        [
+            "YYY",
+            "B",
+            "C",
+            "YYYYYYbYYY",
+            "BYYYcYYY",
+            "",
+            np.nan,
+            "CYYYBYYY",
+            "dog",
+            "cYYYt",
+        ],
+        dtype=any_string_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.replace("^.a|dog", "XX-XX ", case=False, regex=True)
+    expected = Series(
+        [
+            "A",
+            "B",
+            "C",
+            "XX-XX ba",
+            "XX-XX ca",
+            "",
+            np.nan,
+            "XX-XX BA",
+            "XX-XX ",
+            "XX-XX t",
+        ],
+        dtype=any_string_dtype,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_not_case_sensitive_not_regex(any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/41602
+    ser = Series(["A.", "a.", "Ab", "ab", np.nan], dtype=any_string_dtype)
+
+    result = ser.str.replace("a", "c", case=False, regex=False)
+    expected = Series(["c.", "c.", "cb", "cb", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.replace("a.", "c.", case=False, regex=False)
+    expected = Series(["c.", "c.", "Ab", "ab", np.nan], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_replace_regex(any_string_dtype):
+    # https://github.com/pandas-dev/pandas/pull/24809
+    s = Series(["a", "b", "ac", np.nan, ""], dtype=any_string_dtype)
+    result = s.str.replace("^.$", "a", regex=True)
+    expected = Series(["a", "a", "ac", np.nan, ""], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("regex", [True, False])
+def test_replace_regex_single_character(regex, any_string_dtype):
+    # https://github.com/pandas-dev/pandas/pull/24809, enforced in 2.0
+    # GH 24804
+    s = Series(["a.b", ".", "b", np.nan, ""], dtype=any_string_dtype)
+
+    result = s.str.replace(".", "a", regex=regex)
+    if regex:
+        expected = Series(["aaa", "a", "a", np.nan, ""], dtype=any_string_dtype)
+    else:
+        expected = Series(["aab", "a", "b", np.nan, ""], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.match
+# --------------------------------------------------------------------------------------
+
+
+def test_match(any_string_dtype):
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected_dtype = bool
+        na_value = False
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        na_value = np.nan
+
+    values = Series(["fooBAD__barBAD", np.nan, "foo"], dtype=any_string_dtype)
+    result = values.str.match(".*(BAD[_]+).*(BAD)")
+    expected = Series([True, na_value, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    values = Series(
+        ["fooBAD__barBAD", "BAD_BADleroybrown", np.nan, "foo"], dtype=any_string_dtype
+    )
+    result = values.str.match(".*BAD[_]+.*BAD")
+    expected = Series([True, True, na_value, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.match("BAD[_]+.*BAD")
+    expected = Series([False, True, na_value, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    values = Series(
+        ["fooBAD__barBAD", "^BAD_BADleroybrown", np.nan, "foo"], dtype=any_string_dtype
+    )
+    result = values.str.match("^BAD[_]+.*BAD")
+    expected = Series([False, False, na_value, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = values.str.match("\\^BAD[_]+.*BAD")
+    expected = Series([False, True, na_value, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_match_mixed_object():
+    mixed = Series(
+        [
+            "aBAD_BAD",
+            np.nan,
+            "BAD_b_BAD",
+            True,
+            datetime.today(),
+            "foo",
+            None,
+            1,
+            2.0,
+        ]
+    )
+    result = Series(mixed).str.match(".*(BAD[_]+).*(BAD)")
+    expected = Series([True, np.nan, True, np.nan, np.nan, False, None, np.nan, np.nan])
+    assert isinstance(result, Series)
+    tm.assert_series_equal(result, expected)
+
+
+def test_match_na_kwarg(any_string_dtype):
+    # GH #6609
+    s = Series(["a", "b", np.nan], dtype=any_string_dtype)
+
+    result = s.str.match("a", na=False)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series([True, False, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = s.str.match("a")
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected_dtype = bool
+        na_value = False
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        na_value = np.nan
+
+    expected = Series([True, False, na_value], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_match_case_kwarg(any_string_dtype):
+    values = Series(["ab", "AB", "abc", "ABC"], dtype=any_string_dtype)
+    result = values.str.match("ab", case=False)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series([True, True, True, True], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.fullmatch
+# --------------------------------------------------------------------------------------
+
+
+def test_fullmatch(any_string_dtype):
+    # GH 32806
+    ser = Series(
+        ["fooBAD__barBAD", "BAD_BADleroybrown", np.nan, "foo"], dtype=any_string_dtype
+    )
+    result = ser.str.fullmatch(".*BAD[_]+.*BAD")
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected = Series([True, False, False, False], dtype=bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series([True, False, np.nan, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_fullmatch_dollar_literal(any_string_dtype):
+    # GH 56652
+    ser = Series(["foo", "foo$foo", np.nan, "foo$"], dtype=any_string_dtype)
+    result = ser.str.fullmatch("foo\\$")
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected = Series([False, False, False, True], dtype=bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series([False, False, np.nan, True], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_fullmatch_na_kwarg(any_string_dtype):
+    ser = Series(
+        ["fooBAD__barBAD", "BAD_BADleroybrown", np.nan, "foo"], dtype=any_string_dtype
+    )
+    result = ser.str.fullmatch(".*BAD[_]+.*BAD", na=False)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series([True, False, False, False], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_fullmatch_case_kwarg(any_string_dtype):
+    ser = Series(["ab", "AB", "abc", "ABC"], dtype=any_string_dtype)
+    expected_dtype = (
+        np.bool_ if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+
+    expected = Series([True, False, False, False], dtype=expected_dtype)
+
+    result = ser.str.fullmatch("ab", case=True)
+    tm.assert_series_equal(result, expected)
+
+    expected = Series([True, True, False, False], dtype=expected_dtype)
+
+    result = ser.str.fullmatch("ab", case=False)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.fullmatch("ab", flags=re.IGNORECASE)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.findall
+# --------------------------------------------------------------------------------------
+
+
+def test_findall(any_string_dtype):
+    ser = Series(["fooBAD__barBAD", np.nan, "foo", "BAD"], dtype=any_string_dtype)
+    result = ser.str.findall("BAD[_]*")
+    expected = Series([["BAD__", "BAD"], np.nan, [], ["BAD"]])
+    expected = _convert_na_value(ser, expected)
+    tm.assert_series_equal(result, expected)
+
+
+def test_findall_mixed_object():
+    ser = Series(
+        [
+            "fooBAD__barBAD",
+            np.nan,
+            "foo",
+            True,
+            datetime.today(),
+            "BAD",
+            None,
+            1,
+            2.0,
+        ]
+    )
+
+    result = ser.str.findall("BAD[_]*")
+    expected = Series(
+        [
+            ["BAD__", "BAD"],
+            np.nan,
+            [],
+            np.nan,
+            np.nan,
+            ["BAD"],
+            None,
+            np.nan,
+            np.nan,
+        ]
+    )
+
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.find
+# --------------------------------------------------------------------------------------
+
+
+def test_find(any_string_dtype):
+    ser = Series(
+        ["ABCDEFG", "BCDEFEF", "DEFGHIJEF", "EFGHEF", "XXXX"], dtype=any_string_dtype
+    )
+    expected_dtype = (
+        np.int64 if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+
+    result = ser.str.find("EF")
+    expected = Series([4, 3, 1, 0, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.find("EF") for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+    result = ser.str.rfind("EF")
+    expected = Series([4, 5, 7, 4, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.rfind("EF") for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+    result = ser.str.find("EF", 3)
+    expected = Series([4, 3, 7, 4, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.find("EF", 3) for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+    result = ser.str.rfind("EF", 3)
+    expected = Series([4, 5, 7, 4, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.rfind("EF", 3) for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+    result = ser.str.find("EF", 3, 6)
+    expected = Series([4, 3, -1, 4, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.find("EF", 3, 6) for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+    result = ser.str.rfind("EF", 3, 6)
+    expected = Series([4, 3, -1, 4, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+    expected = np.array([v.rfind("EF", 3, 6) for v in np.array(ser)], dtype=np.int64)
+    tm.assert_numpy_array_equal(np.array(result, dtype=np.int64), expected)
+
+
+def test_find_bad_arg_raises(any_string_dtype):
+    ser = Series([], dtype=any_string_dtype)
+    with pytest.raises(TypeError, match="expected a string object, not int"):
+        ser.str.find(0)
+
+    with pytest.raises(TypeError, match="expected a string object, not int"):
+        ser.str.rfind(0)
+
+
+def test_find_nan(any_string_dtype):
+    ser = Series(
+        ["ABCDEFG", np.nan, "DEFGHIJEF", np.nan, "XXXX"], dtype=any_string_dtype
+    )
+    expected_dtype = (
+        np.float64 if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+
+    result = ser.str.find("EF")
+    expected = Series([4, np.nan, 1, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rfind("EF")
+    expected = Series([4, np.nan, 7, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.find("EF", 3)
+    expected = Series([4, np.nan, 7, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rfind("EF", 3)
+    expected = Series([4, np.nan, 7, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.find("EF", 3, 6)
+    expected = Series([4, np.nan, -1, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.rfind("EF", 3, 6)
+    expected = Series([4, np.nan, -1, np.nan, -1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+# str.translate
+# --------------------------------------------------------------------------------------
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+def test_translate(index_or_series, any_string_dtype, infer_string):
+    obj = index_or_series(
+        ["abcdefg", "abcc", "cdddfg", "cdefggg"], dtype=any_string_dtype
+    )
+    table = str.maketrans("abc", "cde")
+    result = obj.str.translate(table)
+    expected = index_or_series(
+        ["cdedefg", "cdee", "edddfg", "edefggg"], dtype=any_string_dtype
+    )
+    tm.assert_equal(result, expected)
+
+
+def test_translate_mixed_object():
+    # Series with non-string values
+    s = Series(["a", "b", "c", 1.2])
+    table = str.maketrans("abc", "cde")
+    expected = Series(["c", "d", "e", np.nan], dtype=object)
+    result = s.str.translate(table)
+    tm.assert_series_equal(result, expected)
+
+
+# --------------------------------------------------------------------------------------
+
+
+def test_flags_kwarg(any_string_dtype):
+    data = {
+        "Dave": "dave@google.com",
+        "Steve": "steve@gmail.com",
+        "Rob": "rob@gmail.com",
+        "Wes": np.nan,
+    }
+    data = Series(data, dtype=any_string_dtype)
+
+    pat = r"([A-Z0-9._%+-]+)@([A-Z0-9.-]+)\.([A-Z]{2,4})"
+
+    result = data.str.extract(pat, flags=re.IGNORECASE, expand=True)
+    assert result.iloc[0].tolist() == ["dave", "google", "com"]
+
+    result = data.str.match(pat, flags=re.IGNORECASE)
+    assert result.iloc[0]
+
+    result = data.str.fullmatch(pat, flags=re.IGNORECASE)
+    assert result.iloc[0]
+
+    result = data.str.findall(pat, flags=re.IGNORECASE)
+    assert result.iloc[0][0] == ("dave", "google", "com")
+
+    result = data.str.count(pat, flags=re.IGNORECASE)
+    assert result.iloc[0] == 1
+
+    msg = "has match groups"
+    with tm.assert_produces_warning(UserWarning, match=msg):
+        result = data.str.contains(pat, flags=re.IGNORECASE)
+    assert result.iloc[0]
diff --git a/pandas/pandas/tests/strings/test_get_dummies.py b/pandas/pandas/tests/strings/test_get_dummies.py
new file mode 100644
index 0000000000000..16e10c6fcdccd
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_get_dummies.py
@@ -0,0 +1,102 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+)
+
+
+def test_get_dummies(any_string_dtype):
+    s = Series(["a|b", "a|c", np.nan], dtype=any_string_dtype)
+    result = s.str.get_dummies("|")
+    expected = DataFrame([[1, 1, 0], [1, 0, 1], [0, 0, 0]], columns=list("abc"))
+    tm.assert_frame_equal(result, expected)
+
+    s = Series(["a;b", "a", 7], dtype=any_string_dtype)
+    result = s.str.get_dummies(";")
+    expected = DataFrame([[0, 1, 1], [0, 1, 0], [1, 0, 0]], columns=list("7ab"))
+    tm.assert_frame_equal(result, expected)
+
+
+def test_get_dummies_index():
+    # GH9980, GH8028
+    idx = Index(["a|b", "a|c", "b|c"])
+    result = idx.str.get_dummies("|")
+
+    expected = MultiIndex.from_tuples(
+        [(1, 1, 0), (1, 0, 1), (0, 1, 1)], names=("a", "b", "c")
+    )
+    tm.assert_index_equal(result, expected)
+
+
+# GH#47872
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        np.uint8,
+        np.int16,
+        np.uint16,
+        np.int32,
+        np.uint32,
+        np.int64,
+        np.uint64,
+        bool,
+        "Int8",
+        "Int16",
+        "Int32",
+        "Int64",
+        "boolean",
+    ],
+)
+def test_get_dummies_with_dtype(any_string_dtype, dtype):
+    s = Series(["a|b", "a|c", np.nan], dtype=any_string_dtype)
+    result = s.str.get_dummies("|", dtype=dtype)
+    expected = DataFrame(
+        [[1, 1, 0], [1, 0, 1], [0, 0, 0]], columns=list("abc"), dtype=dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# GH#47872
+@td.skip_if_no("pyarrow")
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "int8[pyarrow]",
+        "uint8[pyarrow]",
+        "int16[pyarrow]",
+        "uint16[pyarrow]",
+        "int32[pyarrow]",
+        "uint32[pyarrow]",
+        "int64[pyarrow]",
+        "uint64[pyarrow]",
+        "bool[pyarrow]",
+    ],
+)
+def test_get_dummies_with_pyarrow_dtype(any_string_dtype, dtype):
+    s = Series(["a|b", "a|c", np.nan], dtype=any_string_dtype)
+    result = s.str.get_dummies("|", dtype=dtype)
+    expected = DataFrame(
+        [[1, 1, 0], [1, 0, 1], [0, 0, 0]],
+        columns=list("abc"),
+        dtype=dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+# GH#47872
+def test_get_dummies_with_str_dtype(any_string_dtype):
+    s = Series(["a|b", "a|c", np.nan], dtype=any_string_dtype)
+
+    msg = "Only numeric or boolean dtypes are supported for 'dtype'"
+    with pytest.raises(ValueError, match=msg):
+        s.str.get_dummies("|", dtype=str)
+
+    with pytest.raises(ValueError, match=msg):
+        s.str.get_dummies("|", dtype="datetime64[ns]")
diff --git a/pandas/pandas/tests/strings/test_split_partition.py b/pandas/pandas/tests/strings/test_split_partition.py
new file mode 100644
index 0000000000000..4fab6e7778002
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_split_partition.py
@@ -0,0 +1,725 @@
+from datetime import datetime
+import re
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    _testing as tm,
+)
+from pandas.tests.strings import (
+    _convert_na_value,
+    is_object_or_nan_string_dtype,
+)
+
+
+@pytest.mark.parametrize("method", ["split", "rsplit"])
+def test_split(any_string_dtype, method):
+    values = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype=any_string_dtype)
+
+    result = getattr(values.str, method)("_")
+    exp = Series([["a", "b", "c"], ["c", "d", "e"], np.nan, ["f", "g", "h"]])
+    exp = _convert_na_value(values, exp)
+    tm.assert_series_equal(result, exp)
+
+
+@pytest.mark.parametrize("method", ["split", "rsplit"])
+def test_split_more_than_one_char(any_string_dtype, method):
+    # more than one char
+    values = Series(["a__b__c", "c__d__e", np.nan, "f__g__h"], dtype=any_string_dtype)
+    result = getattr(values.str, method)("__")
+    exp = Series([["a", "b", "c"], ["c", "d", "e"], np.nan, ["f", "g", "h"]])
+    exp = _convert_na_value(values, exp)
+    tm.assert_series_equal(result, exp)
+
+    result = getattr(values.str, method)("__", expand=False)
+    tm.assert_series_equal(result, exp)
+
+
+def test_split_more_regex_split(any_string_dtype):
+    # regex split
+    values = Series(["a,b_c", "c_d,e", np.nan, "f,g,h"], dtype=any_string_dtype)
+    result = values.str.split("[,_]")
+    exp = Series([["a", "b", "c"], ["c", "d", "e"], np.nan, ["f", "g", "h"]])
+    exp = _convert_na_value(values, exp)
+    tm.assert_series_equal(result, exp)
+
+
+def test_split_regex(any_string_dtype):
+    # GH 43563
+    # explicit regex = True split
+    values = Series("xxxjpgzzz.jpg", dtype=any_string_dtype)
+    result = values.str.split(r"\.jpg", regex=True)
+    exp = Series([["xxxjpgzzz", ""]])
+    tm.assert_series_equal(result, exp)
+
+
+def test_split_regex_explicit(any_string_dtype):
+    # explicit regex = True split with compiled regex
+    regex_pat = re.compile(r".jpg")
+    values = Series("xxxjpgzzz.jpg", dtype=any_string_dtype)
+    result = values.str.split(regex_pat)
+    exp = Series([["xx", "zzz", ""]])
+    tm.assert_series_equal(result, exp)
+
+    # explicit regex = False split
+    result = values.str.split(r"\.jpg", regex=False)
+    exp = Series([["xxxjpgzzz.jpg"]])
+    tm.assert_series_equal(result, exp)
+
+    # non explicit regex split, pattern length == 1
+    result = values.str.split(r".")
+    exp = Series([["xxxjpgzzz", "jpg"]])
+    tm.assert_series_equal(result, exp)
+
+    # non explicit regex split, pattern length != 1
+    result = values.str.split(r".jpg")
+    exp = Series([["xx", "zzz", ""]])
+    tm.assert_series_equal(result, exp)
+
+    # regex=False with pattern compiled regex raises error
+    with pytest.raises(
+        ValueError,
+        match="Cannot use a compiled regex as replacement pattern with regex=False",
+    ):
+        values.str.split(regex_pat, regex=False)
+
+
+@pytest.mark.parametrize("expand", [None, False])
+@pytest.mark.parametrize("method", ["split", "rsplit"])
+def test_split_object_mixed(expand, method):
+    mixed = Series(["a_b_c", np.nan, "d_e_f", True, datetime.today(), None, 1, 2.0])
+    result = getattr(mixed.str, method)("_", expand=expand)
+    exp = Series(
+        [
+            ["a", "b", "c"],
+            np.nan,
+            ["d", "e", "f"],
+            np.nan,
+            np.nan,
+            None,
+            np.nan,
+            np.nan,
+        ]
+    )
+    assert isinstance(result, Series)
+    tm.assert_almost_equal(result, exp)
+
+
+@pytest.mark.parametrize("method", ["split", "rsplit"])
+@pytest.mark.parametrize("n", [None, 0])
+def test_split_n(any_string_dtype, method, n):
+    s = Series(["a b", pd.NA, "b c"], dtype=any_string_dtype)
+    expected = Series([["a", "b"], pd.NA, ["b", "c"]])
+    result = getattr(s.str, method)(" ", n=n)
+    expected = _convert_na_value(s, expected)
+    tm.assert_series_equal(result, expected)
+
+
+def test_rsplit(any_string_dtype):
+    # regex split is not supported by rsplit
+    values = Series(["a,b_c", "c_d,e", np.nan, "f,g,h"], dtype=any_string_dtype)
+    result = values.str.rsplit("[,_]")
+    exp = Series([["a,b_c"], ["c_d,e"], np.nan, ["f,g,h"]])
+    exp = _convert_na_value(values, exp)
+    tm.assert_series_equal(result, exp)
+
+
+def test_rsplit_max_number(any_string_dtype):
+    # setting max number of splits, make sure it's from reverse
+    values = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype=any_string_dtype)
+    result = values.str.rsplit("_", n=1)
+    exp = Series([["a_b", "c"], ["c_d", "e"], np.nan, ["f_g", "h"]])
+    exp = _convert_na_value(values, exp)
+    tm.assert_series_equal(result, exp)
+
+
+def test_split_blank_string(any_string_dtype):
+    # expand blank split GH 20067
+    values = Series([""], name="test", dtype=any_string_dtype)
+    result = values.str.split(expand=True)
+    exp = DataFrame([[]], dtype=any_string_dtype)  # NOTE: this is NOT an empty df
+    tm.assert_frame_equal(result, exp)
+
+
+def test_split_blank_string_with_non_empty(any_string_dtype):
+    values = Series(["a b c", "a b", "", " "], name="test", dtype=any_string_dtype)
+    result = values.str.split(expand=True)
+    exp = DataFrame(
+        [
+            ["a", "b", "c"],
+            ["a", "b", None],
+            [None, None, None],
+            [None, None, None],
+        ],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+
+@pytest.mark.parametrize("method", ["split", "rsplit"])
+def test_split_noargs(any_string_dtype, method):
+    # #1859
+    s = Series(["Wes McKinney", "Travis  Oliphant"], dtype=any_string_dtype)
+    result = getattr(s.str, method)()
+    expected = ["Travis", "Oliphant"]
+    assert result[1] == expected
+
+
+@pytest.mark.parametrize(
+    "data, pat",
+    [
+        (["bd asdf jfg", "kjasdflqw asdfnfk"], None),
+        (["bd asdf jfg", "kjasdflqw asdfnfk"], "asdf"),
+        (["bd_asdf_jfg", "kjasdflqw_asdfnfk"], "_"),
+    ],
+)
+@pytest.mark.parametrize("n", [-1, 0])
+def test_split_maxsplit(data, pat, any_string_dtype, n):
+    # re.split 0, str.split -1
+    s = Series(data, dtype=any_string_dtype)
+
+    result = s.str.split(pat=pat, n=n)
+    xp = s.str.split(pat=pat)
+    tm.assert_series_equal(result, xp)
+
+
+@pytest.mark.parametrize(
+    "data, pat, expected_val",
+    [
+        (
+            ["split once", "split once too!"],
+            None,
+            "once too!",
+        ),
+        (
+            ["split_once", "split_once_too!"],
+            "_",
+            "once_too!",
+        ),
+    ],
+)
+def test_split_no_pat_with_nonzero_n(data, pat, expected_val, any_string_dtype):
+    s = Series(data, dtype=any_string_dtype)
+    result = s.str.split(pat=pat, n=1)
+    expected = Series({0: ["split", "once"], 1: ["split", expected_val]})
+    tm.assert_series_equal(expected, result, check_index_type=False)
+
+
+def test_split_to_dataframe_no_splits(any_string_dtype):
+    s = Series(["nosplit", "alsonosplit"], dtype=any_string_dtype)
+    result = s.str.split("_", expand=True)
+    exp = DataFrame({0: Series(["nosplit", "alsonosplit"], dtype=any_string_dtype)})
+    tm.assert_frame_equal(result, exp)
+
+
+def test_split_to_dataframe(any_string_dtype):
+    s = Series(["some_equal_splits", "with_no_nans"], dtype=any_string_dtype)
+    result = s.str.split("_", expand=True)
+    exp = DataFrame(
+        {0: ["some", "with"], 1: ["equal", "no"], 2: ["splits", "nans"]},
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+
+def test_split_to_dataframe_unequal_splits(any_string_dtype):
+    s = Series(
+        ["some_unequal_splits", "one_of_these_things_is_not"], dtype=any_string_dtype
+    )
+    result = s.str.split("_", expand=True)
+    exp = DataFrame(
+        {
+            0: ["some", "one"],
+            1: ["unequal", "of"],
+            2: ["splits", "these"],
+            3: [None, "things"],
+            4: [None, "is"],
+            5: [None, "not"],
+        },
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+
+def test_split_to_dataframe_with_index(any_string_dtype):
+    s = Series(
+        ["some_splits", "with_index"], index=["preserve", "me"], dtype=any_string_dtype
+    )
+    result = s.str.split("_", expand=True)
+    exp = DataFrame(
+        {0: ["some", "with"], 1: ["splits", "index"]},
+        index=["preserve", "me"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+    with pytest.raises(ValueError, match="expand must be"):
+        s.str.split("_", expand="not_a_boolean")
+
+
+def test_split_to_multiindex_expand_no_splits():
+    # https://github.com/pandas-dev/pandas/issues/23677
+
+    idx = Index(["nosplit", "alsonosplit", np.nan])
+    result = idx.str.split("_", expand=True)
+    exp = idx
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 1
+
+
+def test_split_to_multiindex_expand():
+    idx = Index(["some_equal_splits", "with_no_nans", np.nan, None])
+    result = idx.str.split("_", expand=True)
+    exp = MultiIndex.from_tuples(
+        [
+            ("some", "equal", "splits"),
+            ("with", "no", "nans"),
+            [np.nan, np.nan, np.nan],
+            [None, None, None],
+        ]
+    )
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 3
+
+
+def test_split_to_multiindex_expand_unequal_splits():
+    idx = Index(["some_unequal_splits", "one_of_these_things_is_not", np.nan, None])
+    result = idx.str.split("_", expand=True)
+    exp = MultiIndex.from_tuples(
+        [
+            ("some", "unequal", "splits", np.nan, np.nan, np.nan),
+            ("one", "of", "these", "things", "is", "not"),
+            (np.nan, np.nan, np.nan, np.nan, np.nan, np.nan),
+            (None, None, None, None, None, None),
+        ]
+    )
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 6
+
+    with pytest.raises(ValueError, match="expand must be"):
+        idx.str.split("_", expand="not_a_boolean")
+
+
+def test_rsplit_to_dataframe_expand_no_splits(any_string_dtype):
+    s = Series(["nosplit", "alsonosplit"], dtype=any_string_dtype)
+    result = s.str.rsplit("_", expand=True)
+    exp = DataFrame({0: Series(["nosplit", "alsonosplit"])}, dtype=any_string_dtype)
+    tm.assert_frame_equal(result, exp)
+
+
+def test_rsplit_to_dataframe_expand(any_string_dtype):
+    s = Series(["some_equal_splits", "with_no_nans"], dtype=any_string_dtype)
+    result = s.str.rsplit("_", expand=True)
+    exp = DataFrame(
+        {0: ["some", "with"], 1: ["equal", "no"], 2: ["splits", "nans"]},
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+    result = s.str.rsplit("_", expand=True, n=2)
+    exp = DataFrame(
+        {0: ["some", "with"], 1: ["equal", "no"], 2: ["splits", "nans"]},
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+    result = s.str.rsplit("_", expand=True, n=1)
+    exp = DataFrame(
+        {0: ["some_equal", "with_no"], 1: ["splits", "nans"]}, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, exp)
+
+
+def test_rsplit_to_dataframe_expand_with_index(any_string_dtype):
+    s = Series(
+        ["some_splits", "with_index"], index=["preserve", "me"], dtype=any_string_dtype
+    )
+    result = s.str.rsplit("_", expand=True)
+    exp = DataFrame(
+        {0: ["some", "with"], 1: ["splits", "index"]},
+        index=["preserve", "me"],
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, exp)
+
+
+def test_rsplit_to_multiindex_expand_no_split():
+    idx = Index(["nosplit", "alsonosplit"])
+    result = idx.str.rsplit("_", expand=True)
+    exp = idx
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 1
+
+
+def test_rsplit_to_multiindex_expand():
+    idx = Index(["some_equal_splits", "with_no_nans"])
+    result = idx.str.rsplit("_", expand=True)
+    exp = MultiIndex.from_tuples([("some", "equal", "splits"), ("with", "no", "nans")])
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 3
+
+
+def test_rsplit_to_multiindex_expand_n():
+    idx = Index(["some_equal_splits", "with_no_nans"])
+    result = idx.str.rsplit("_", expand=True, n=1)
+    exp = MultiIndex.from_tuples([("some_equal", "splits"), ("with_no", "nans")])
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 2
+
+
+def test_split_nan_expand(any_string_dtype):
+    # gh-18450
+    s = Series(["foo,bar,baz", np.nan], dtype=any_string_dtype)
+    result = s.str.split(",", expand=True)
+    exp = DataFrame(
+        [["foo", "bar", "baz"], [np.nan, np.nan, np.nan]], dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, exp)
+
+    # check that these are actually np.nan/pd.NA and not None
+    # TODO see GH 18463
+    # tm.assert_frame_equal does not differentiate
+    if is_object_or_nan_string_dtype(any_string_dtype):
+        assert all(np.isnan(x) for x in result.iloc[1])
+    else:
+        assert all(x is pd.NA for x in result.iloc[1])
+
+
+def test_split_with_name_series(any_string_dtype):
+    # GH 12617
+
+    # should preserve name
+    s = Series(["a,b", "c,d"], name="xxx", dtype=any_string_dtype)
+    res = s.str.split(",")
+    exp = Series([["a", "b"], ["c", "d"]], name="xxx")
+    tm.assert_series_equal(res, exp)
+
+    res = s.str.split(",", expand=True)
+    exp = DataFrame([["a", "b"], ["c", "d"]], dtype=any_string_dtype)
+    tm.assert_frame_equal(res, exp)
+
+
+def test_split_with_name_index():
+    # GH 12617
+    idx = Index(["a,b", "c,d"], name="xxx")
+    res = idx.str.split(",")
+    exp = Index([["a", "b"], ["c", "d"]], name="xxx")
+    assert res.nlevels == 1
+    tm.assert_index_equal(res, exp)
+
+    res = idx.str.split(",", expand=True)
+    exp = MultiIndex.from_tuples([("a", "b"), ("c", "d")])
+    assert res.nlevels == 2
+    tm.assert_index_equal(res, exp)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            [
+                ("a", "__", "b__c"),
+                ("c", "__", "d__e"),
+                np.nan,
+                ("f", "__", "g__h"),
+                None,
+            ],
+        ],
+        [
+            "rpartition",
+            [
+                ("a__b", "__", "c"),
+                ("c__d", "__", "e"),
+                np.nan,
+                ("f__g", "__", "h"),
+                None,
+            ],
+        ],
+    ],
+)
+def test_partition_series_more_than_one_char(method, exp, any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    # more than one char
+    s = Series(["a__b__c", "c__d__e", np.nan, "f__g__h", None], dtype=any_string_dtype)
+    result = getattr(s.str, method)("__", expand=False)
+    expected = Series(exp)
+    expected = _convert_na_value(s, expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            [("a", " ", "b c"), ("c", " ", "d e"), np.nan, ("f", " ", "g h"), None],
+        ],
+        [
+            "rpartition",
+            [("a b", " ", "c"), ("c d", " ", "e"), np.nan, ("f g", " ", "h"), None],
+        ],
+    ],
+)
+def test_partition_series_none(any_string_dtype, method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    # None
+    s = Series(["a b c", "c d e", np.nan, "f g h", None], dtype=any_string_dtype)
+    result = getattr(s.str, method)(expand=False)
+    expected = Series(exp)
+    expected = _convert_na_value(s, expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            [("abc", "", ""), ("cde", "", ""), np.nan, ("fgh", "", ""), None],
+        ],
+        [
+            "rpartition",
+            [("", "", "abc"), ("", "", "cde"), np.nan, ("", "", "fgh"), None],
+        ],
+    ],
+)
+def test_partition_series_not_split(any_string_dtype, method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    # Not split
+    s = Series(["abc", "cde", np.nan, "fgh", None], dtype=any_string_dtype)
+    result = getattr(s.str, method)("_", expand=False)
+    expected = Series(exp)
+    expected = _convert_na_value(s, expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            [("a", "_", "b_c"), ("c", "_", "d_e"), np.nan, ("f", "_", "g_h")],
+        ],
+        [
+            "rpartition",
+            [("a_b", "_", "c"), ("c_d", "_", "e"), np.nan, ("f_g", "_", "h")],
+        ],
+    ],
+)
+def test_partition_series_unicode(any_string_dtype, method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    # unicode
+    s = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype=any_string_dtype)
+
+    result = getattr(s.str, method)("_", expand=False)
+    expected = Series(exp)
+    expected = _convert_na_value(s, expected)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["partition", "rpartition"])
+def test_partition_series_stdlib(any_string_dtype, method):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    # compare to standard lib
+    s = Series(["A_B_C", "B_C_D", "E_F_G", "EFGHEF"], dtype=any_string_dtype)
+    result = getattr(s.str, method)("_", expand=False).tolist()
+    assert result == [getattr(v, method)("_") for v in s]
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            [("a", "_", "b_c"), ("c", "_", "d_e"), ("f", "_", "g_h"), np.nan, None],
+        ],
+        [
+            "rpartition",
+            [("a_b", "_", "c"), ("c_d", "_", "e"), ("f_g", "_", "h"), np.nan, None],
+        ],
+    ],
+)
+def test_partition_index(method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+
+    values = Index(["a_b_c", "c_d_e", "f_g_h", np.nan, None])
+
+    result = getattr(values.str, method)("_", expand=False)
+    exp = Index(np.array(exp, dtype=object), dtype=object)
+    tm.assert_index_equal(result, exp)
+    assert result.nlevels == 1
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            {
+                0: ["a", "c", np.nan, "f", None],
+                1: ["_", "_", np.nan, "_", None],
+                2: ["b_c", "d_e", np.nan, "g_h", None],
+            },
+        ],
+        [
+            "rpartition",
+            {
+                0: ["a_b", "c_d", np.nan, "f_g", None],
+                1: ["_", "_", np.nan, "_", None],
+                2: ["c", "e", np.nan, "h", None],
+            },
+        ],
+    ],
+)
+def test_partition_to_dataframe(any_string_dtype, method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+
+    s = Series(["a_b_c", "c_d_e", np.nan, "f_g_h", None], dtype=any_string_dtype)
+    result = getattr(s.str, method)("_")
+    expected = DataFrame(
+        exp,
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        [
+            "partition",
+            {
+                0: ["a", "c", np.nan, "f", None],
+                1: ["_", "_", np.nan, "_", None],
+                2: ["b_c", "d_e", np.nan, "g_h", None],
+            },
+        ],
+        [
+            "rpartition",
+            {
+                0: ["a_b", "c_d", np.nan, "f_g", None],
+                1: ["_", "_", np.nan, "_", None],
+                2: ["c", "e", np.nan, "h", None],
+            },
+        ],
+    ],
+)
+def test_partition_to_dataframe_from_series(any_string_dtype, method, exp):
+    # https://github.com/pandas-dev/pandas/issues/23558
+    s = Series(["a_b_c", "c_d_e", np.nan, "f_g_h", None], dtype=any_string_dtype)
+    result = getattr(s.str, method)("_", expand=True)
+    expected = DataFrame(
+        exp,
+        dtype=any_string_dtype,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_partition_with_name(any_string_dtype):
+    # GH 12617
+
+    s = Series(["a,b", "c,d"], name="xxx", dtype=any_string_dtype)
+    result = s.str.partition(",")
+    expected = DataFrame(
+        {0: ["a", "c"], 1: [",", ","], 2: ["b", "d"]}, dtype=any_string_dtype
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_partition_with_name_expand(any_string_dtype):
+    # GH 12617
+    # should preserve name
+    s = Series(["a,b", "c,d"], name="xxx", dtype=any_string_dtype)
+    result = s.str.partition(",", expand=False)
+    expected = Series([("a", ",", "b"), ("c", ",", "d")], name="xxx")
+    tm.assert_series_equal(result, expected)
+
+
+def test_partition_index_with_name():
+    idx = Index(["a,b", "c,d"], name="xxx")
+    result = idx.str.partition(",")
+    expected = MultiIndex.from_tuples([("a", ",", "b"), ("c", ",", "d")])
+    assert result.nlevels == 3
+    tm.assert_index_equal(result, expected)
+
+
+def test_partition_index_with_name_expand_false():
+    idx = Index(["a,b", "c,d"], name="xxx")
+    # should preserve name
+    result = idx.str.partition(",", expand=False)
+    expected = Index(np.array([("a", ",", "b"), ("c", ",", "d")]), name="xxx")
+    assert result.nlevels == 1
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["partition", "rpartition"])
+def test_partition_sep_kwarg(any_string_dtype, method):
+    # GH 22676; depr kwarg "pat" in favor of "sep"
+    s = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype=any_string_dtype)
+
+    expected = getattr(s.str, method)(sep="_")
+    result = getattr(s.str, method)("_")
+    tm.assert_frame_equal(result, expected)
+
+
+def test_get():
+    ser = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"])
+    result = ser.str.split("_").str.get(1)
+    expected = Series(["b", "d", np.nan, "g"], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_get_mixed_object():
+    ser = Series(["a_b_c", np.nan, "c_d_e", True, datetime.today(), None, 1, 2.0])
+    result = ser.str.split("_").str.get(1)
+    expected = Series(
+        ["b", np.nan, "d", np.nan, np.nan, None, np.nan, np.nan], dtype=object
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("idx", [2, -3])
+def test_get_bounds(idx):
+    ser = Series(["1_2_3_4_5", "6_7_8_9_10", "11_12"])
+    result = ser.str.split("_").str.get(idx)
+    expected = Series(["3", "8", np.nan], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "idx, exp", [[2, [3, 3, np.nan, "b"]], [-1, [3, 3, np.nan, np.nan]]]
+)
+def test_get_complex(idx, exp):
+    # GH 20671, getting value not in dict raising `KeyError`
+    ser = Series([(1, 2, 3), [1, 2, 3], {1, 2, 3}, {1: "a", 2: "b", 3: "c"}])
+
+    result = ser.str.get(idx)
+    expected = Series(exp)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("to_type", [tuple, list, np.array])
+def test_get_complex_nested(to_type):
+    ser = Series([to_type([to_type([1, 2])])])
+
+    result = ser.str.get(0)
+    expected = Series([to_type([1, 2])])
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.get(1)
+    expected = Series([np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_get_strings(any_string_dtype):
+    ser = Series(["a", "ab", np.nan, "abc"], dtype=any_string_dtype)
+    result = ser.str.get(2)
+    expected = Series([np.nan, np.nan, np.nan, "c"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/strings/test_string_array.py b/pandas/pandas/tests/strings/test_string_array.py
new file mode 100644
index 0000000000000..c5414022e664b
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_string_array.py
@@ -0,0 +1,112 @@
+import numpy as np
+import pytest
+
+from pandas._libs import lib
+
+from pandas import (
+    NA,
+    DataFrame,
+    Series,
+    _testing as tm,
+    option_context,
+)
+
+
+def test_string_array(nullable_string_dtype, any_string_method):
+    method_name, args, kwargs = any_string_method
+
+    data = ["a", "bb", np.nan, "ccc"]
+    a = Series(data, dtype=object)
+    b = Series(data, dtype=nullable_string_dtype)
+
+    if method_name == "decode":
+        with pytest.raises(TypeError, match="a bytes-like object is required"):
+            getattr(b.str, method_name)(*args, **kwargs)
+        return
+
+    expected = getattr(a.str, method_name)(*args, **kwargs)
+    result = getattr(b.str, method_name)(*args, **kwargs)
+
+    if isinstance(expected, Series):
+        if expected.dtype == "object" and lib.is_string_array(
+            expected.dropna().values,
+        ):
+            assert result.dtype == nullable_string_dtype
+            result = result.astype(object)
+
+        elif expected.dtype == "object" and lib.is_bool_array(
+            expected.values, skipna=True
+        ):
+            assert result.dtype == "boolean"
+            expected = expected.astype("boolean")
+
+        elif expected.dtype == "bool":
+            assert result.dtype == "boolean"
+            result = result.astype("bool")
+
+        elif expected.dtype == "float" and expected.isna().any():
+            assert result.dtype == "Int64"
+            result = result.astype("float")
+
+        if expected.dtype == object:
+            # GH#18463
+            expected[expected.isna()] = NA
+
+    elif isinstance(expected, DataFrame):
+        columns = expected.select_dtypes(include="object").columns
+        assert all(result[columns].dtypes == nullable_string_dtype)
+        result[columns] = result[columns].astype(object)
+        with option_context("future.no_silent_downcasting", True):
+            expected[columns] = expected[columns].fillna(NA)  # GH#18463
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method,expected",
+    [
+        ("count", [2, None]),
+        ("find", [0, None]),
+        ("index", [0, None]),
+        ("rindex", [2, None]),
+    ],
+)
+def test_string_array_numeric_integer_array(nullable_string_dtype, method, expected):
+    s = Series(["aba", None], dtype=nullable_string_dtype)
+    result = getattr(s.str, method)("a")
+    expected = Series(expected, dtype="Int64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method,expected",
+    [
+        ("isdigit", [False, None, True]),
+        ("isalpha", [True, None, False]),
+        ("isascii", [True, None, True]),
+        ("isalnum", [True, None, True]),
+        ("isnumeric", [False, None, True]),
+    ],
+)
+def test_string_array_boolean_array(nullable_string_dtype, method, expected):
+    s = Series(["a", None, "1"], dtype=nullable_string_dtype)
+    result = getattr(s.str, method)()
+    expected = Series(expected, dtype="boolean")
+    tm.assert_series_equal(result, expected)
+
+
+def test_string_array_extract(nullable_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/30969
+    # Only expand=False & multiple groups was failing
+
+    a = Series(["a1", "b2", "cc"], dtype=nullable_string_dtype)
+    b = Series(["a1", "b2", "cc"], dtype="object")
+    pat = r"(\w)(\d)"
+
+    result = a.str.extract(pat, expand=False)
+    expected = b.str.extract(pat, expand=False)
+    expected = expected.fillna(NA)  # GH#18463
+    assert all(result.dtypes == nullable_string_dtype)
+
+    result = result.astype(object)
+    tm.assert_equal(result, expected)
diff --git a/pandas/pandas/tests/strings/test_strings.py b/pandas/pandas/tests/strings/test_strings.py
new file mode 100644
index 0000000000000..025f837982595
--- /dev/null
+++ b/pandas/pandas/tests/strings/test_strings.py
@@ -0,0 +1,780 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+from pandas.core.strings.accessor import StringMethods
+from pandas.tests.strings import is_object_or_nan_string_dtype
+
+
+@pytest.mark.parametrize("pattern", [0, True, Series(["foo", "bar"])])
+def test_startswith_endswith_non_str_patterns(pattern):
+    # GH3485
+    ser = Series(["foo", "bar"])
+    msg = f"expected a string or tuple, not {type(pattern).__name__}"
+    with pytest.raises(TypeError, match=msg):
+        ser.str.startswith(pattern)
+    with pytest.raises(TypeError, match=msg):
+        ser.str.endswith(pattern)
+
+
+def test_iter_raises():
+    # GH 54173
+    ser = Series(["foo", "bar"])
+    with pytest.raises(TypeError, match="'StringMethods' object is not iterable"):
+        iter(ser.str)
+
+
+# test integer/float dtypes (inferred by constructor) and mixed
+
+
+def test_count(any_string_dtype):
+    ser = Series(["foo", "foofoo", np.nan, "foooofooofommmfoo"], dtype=any_string_dtype)
+    result = ser.str.count("f[o]+")
+    expected_dtype = (
+        np.float64 if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+    expected = Series([1, 2, np.nan, 4], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_count_mixed_object():
+    ser = Series(
+        ["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0],
+        dtype=object,
+    )
+    result = ser.str.count("a")
+    expected = Series([1, np.nan, 0, np.nan, np.nan, 0, np.nan, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_repeat(any_string_dtype):
+    ser = Series(["a", "b", np.nan, "c", np.nan, "d"], dtype=any_string_dtype)
+
+    result = ser.str.repeat(3)
+    expected = Series(
+        ["aaa", "bbb", np.nan, "ccc", np.nan, "ddd"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.repeat([1, 2, 3, 4, 5, 6])
+    expected = Series(
+        ["a", "bb", np.nan, "cccc", np.nan, "dddddd"], dtype=any_string_dtype
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_repeat_mixed_object():
+    ser = Series(["a", np.nan, "b", True, datetime.today(), "foo", None, 1, 2.0])
+    result = ser.str.repeat(3)
+    expected = Series(
+        ["aaa", np.nan, "bbb", np.nan, np.nan, "foofoofoo", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("arg, repeat", [[None, 4], ["b", None]])
+def test_repeat_with_null(any_string_dtype, arg, repeat):
+    # GH: 31632
+    ser = Series(["a", arg], dtype=any_string_dtype)
+    result = ser.str.repeat([3, repeat])
+    expected = Series(["aaa", None], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_empty_str_methods(any_string_dtype):
+    empty_str = empty = Series(dtype=any_string_dtype)
+    empty_inferred_str = Series(dtype="str")
+    if is_object_or_nan_string_dtype(any_string_dtype):
+        empty_int = Series(dtype="int64")
+        empty_bool = Series(dtype=bool)
+    else:
+        empty_int = Series(dtype="Int64")
+        empty_bool = Series(dtype="boolean")
+    empty_object = Series(dtype=object)
+    empty_bytes = Series(dtype=object)
+    empty_df = DataFrame()
+
+    # GH7241
+    # (extract) on empty series
+
+    tm.assert_series_equal(empty_str, empty.str.cat(empty))
+    assert "" == empty.str.cat()
+    tm.assert_series_equal(empty_str, empty.str.title())
+    tm.assert_series_equal(empty_int, empty.str.count("a"))
+    tm.assert_series_equal(empty_bool, empty.str.contains("a"))
+    tm.assert_series_equal(empty_bool, empty.str.startswith("a"))
+    tm.assert_series_equal(empty_bool, empty.str.endswith("a"))
+    tm.assert_series_equal(empty_str, empty.str.lower())
+    tm.assert_series_equal(empty_str, empty.str.upper())
+    tm.assert_series_equal(empty_str, empty.str.replace("a", "b"))
+    tm.assert_series_equal(empty_str, empty.str.repeat(3))
+    tm.assert_series_equal(empty_bool, empty.str.match("^a"))
+    tm.assert_frame_equal(
+        DataFrame(columns=range(1), dtype=any_string_dtype),
+        empty.str.extract("()", expand=True),
+    )
+    tm.assert_frame_equal(
+        DataFrame(columns=range(2), dtype=any_string_dtype),
+        empty.str.extract("()()", expand=True),
+    )
+    tm.assert_series_equal(empty_str, empty.str.extract("()", expand=False))
+    tm.assert_frame_equal(
+        DataFrame(columns=range(2), dtype=any_string_dtype),
+        empty.str.extract("()()", expand=False),
+    )
+    tm.assert_frame_equal(empty_df.set_axis([], axis=1), empty.str.get_dummies())
+    tm.assert_series_equal(empty_str, empty_str.str.join(""))
+    tm.assert_series_equal(empty_int, empty.str.len())
+    tm.assert_series_equal(empty_object, empty_str.str.findall("a"))
+    tm.assert_series_equal(empty_int, empty.str.find("a"))
+    tm.assert_series_equal(empty_int, empty.str.rfind("a"))
+    tm.assert_series_equal(empty_str, empty.str.pad(42))
+    tm.assert_series_equal(empty_str, empty.str.center(42))
+    tm.assert_series_equal(empty_object, empty.str.split("a"))
+    tm.assert_series_equal(empty_object, empty.str.rsplit("a"))
+    tm.assert_series_equal(empty_object, empty.str.partition("a", expand=False))
+    tm.assert_frame_equal(empty_df, empty.str.partition("a"))
+    tm.assert_series_equal(empty_object, empty.str.rpartition("a", expand=False))
+    tm.assert_frame_equal(empty_df, empty.str.rpartition("a"))
+    tm.assert_series_equal(empty_str, empty.str.slice(stop=1))
+    tm.assert_series_equal(empty_str, empty.str.slice(step=1))
+    tm.assert_series_equal(empty_str, empty.str.strip())
+    tm.assert_series_equal(empty_str, empty.str.lstrip())
+    tm.assert_series_equal(empty_str, empty.str.rstrip())
+    tm.assert_series_equal(empty_str, empty.str.wrap(42))
+    tm.assert_series_equal(empty_str, empty.str.get(0))
+    tm.assert_series_equal(empty_inferred_str, empty_bytes.str.decode("ascii"))
+    tm.assert_series_equal(empty_bytes, empty.str.encode("ascii"))
+    # ismethods should always return boolean (GH 29624)
+    tm.assert_series_equal(empty_bool, empty.str.isalnum())
+    tm.assert_series_equal(empty_bool, empty.str.isalpha())
+    tm.assert_series_equal(empty_bool, empty.str.isascii())
+    tm.assert_series_equal(empty_bool, empty.str.isdigit())
+    tm.assert_series_equal(empty_bool, empty.str.isspace())
+    tm.assert_series_equal(empty_bool, empty.str.islower())
+    tm.assert_series_equal(empty_bool, empty.str.isupper())
+    tm.assert_series_equal(empty_bool, empty.str.istitle())
+    tm.assert_series_equal(empty_bool, empty.str.isnumeric())
+    tm.assert_series_equal(empty_bool, empty.str.isdecimal())
+    tm.assert_series_equal(empty_str, empty.str.capitalize())
+    tm.assert_series_equal(empty_str, empty.str.swapcase())
+    tm.assert_series_equal(empty_str, empty.str.normalize("NFC"))
+
+    table = str.maketrans("a", "b")
+    tm.assert_series_equal(empty_str, empty.str.translate(table))
+
+
+@pytest.mark.parametrize(
+    "method, expected",
+    [
+        ("isascii", [True, True, True, True, True, True, True, True, True, True]),
+        ("isalnum", [True, True, True, True, True, False, True, True, False, False]),
+        ("isalpha", [True, True, True, False, False, False, True, False, False, False]),
+        (
+            "isdigit",
+            [False, False, False, True, False, False, False, True, False, False],
+        ),
+        (
+            "isnumeric",
+            [False, False, False, True, False, False, False, True, False, False],
+        ),
+        (
+            "isspace",
+            [False, False, False, False, False, False, False, False, False, True],
+        ),
+        (
+            "islower",
+            [False, True, False, False, False, False, False, False, False, False],
+        ),
+        (
+            "isupper",
+            [True, False, False, False, True, False, True, False, False, False],
+        ),
+        (
+            "istitle",
+            [True, False, True, False, True, False, False, False, False, False],
+        ),
+    ],
+)
+def test_ismethods(method, expected, any_string_dtype):
+    ser = Series(
+        ["A", "b", "Xy", "4", "3A", "", "TT", "55", "-", "  "], dtype=any_string_dtype
+    )
+    expected_dtype = (
+        "bool" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series(expected, dtype=expected_dtype)
+    result = getattr(ser.str, method)()
+    tm.assert_series_equal(result, expected)
+
+    # compare with standard library
+    expected_stdlib = [getattr(item, method)() for item in ser]
+    assert list(result) == expected_stdlib
+
+    # with missing value
+    ser.iloc[[1, 2, 3, 4]] = np.nan
+    result = getattr(ser.str, method)()
+    if ser.dtype == "object":
+        expected = expected.astype(object)
+        expected.iloc[[1, 2, 3, 4]] = np.nan
+    elif ser.dtype == "str":
+        # NaN propagates as False
+        expected.iloc[[1, 2, 3, 4]] = False
+    else:
+        # nullable dtypes propagate NaN
+        expected.iloc[[1, 2, 3, 4]] = np.nan
+
+
+@pytest.mark.parametrize(
+    "method, expected",
+    [
+        ("isnumeric", [False, True, True, False, True, True, False]),
+        ("isdecimal", [False, True, False, False, False, True, False]),
+    ],
+)
+def test_isnumeric_unicode(method, expected, any_string_dtype):
+    # 0x00bc: ¼ VULGAR FRACTION ONE QUARTER
+    # 0x2605: ★ not number
+    # 0x1378: ፸ ETHIOPIC NUMBER SEVENTY
+    # 0xFF13: ３ Em 3  # noqa: RUF003
+    ser = Series(
+        ["A", "3", "¼", "★", "፸", "３", "four"],  # noqa: RUF001
+        dtype=any_string_dtype,
+    )
+    expected_dtype = (
+        "bool" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+    )
+    expected = Series(expected, dtype=expected_dtype)
+    result = getattr(ser.str, method)()
+    tm.assert_series_equal(result, expected)
+
+    # compare with standard library
+    expected = [getattr(item, method)() for item in ser]
+    assert list(result) == expected
+
+
+@pytest.mark.parametrize(
+    "method, expected",
+    [
+        ("isnumeric", [False, np.nan, True, False, np.nan, True, False]),
+        ("isdecimal", [False, np.nan, False, False, np.nan, True, False]),
+    ],
+)
+def test_isnumeric_unicode_missing(method, expected, any_string_dtype):
+    values = ["A", np.nan, "¼", "★", np.nan, "３", "four"]  # noqa: RUF001
+    ser = Series(values, dtype=any_string_dtype)
+    if any_string_dtype == "str":
+        # NaN propagates as False
+        expected = Series(expected, dtype=object).fillna(False).astype(bool)
+    else:
+        expected_dtype = (
+            "object" if is_object_or_nan_string_dtype(any_string_dtype) else "boolean"
+        )
+        expected = Series(expected, dtype=expected_dtype)
+    result = getattr(ser.str, method)()
+    tm.assert_series_equal(result, expected)
+
+
+def test_spilt_join_roundtrip(any_string_dtype):
+    ser = Series(["a_b_c", "c_d_e", np.nan, "f_g_h"], dtype=any_string_dtype)
+    result = ser.str.split("_").str.join("_")
+    expected = ser.astype(object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_spilt_join_roundtrip_mixed_object():
+    ser = Series(
+        ["a_b", np.nan, "asdf_cas_asdf", True, datetime.today(), "foo", None, 1, 2.0]
+    )
+    result = ser.str.split("_").str.join("_")
+    expected = Series(
+        ["a_b", np.nan, "asdf_cas_asdf", np.nan, np.nan, "foo", None, np.nan, np.nan],
+        dtype=object,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_len(any_string_dtype):
+    ser = Series(
+        ["foo", "fooo", "fooooo", np.nan, "fooooooo", "foo\n", "あ"],
+        dtype=any_string_dtype,
+    )
+    result = ser.str.len()
+    expected_dtype = (
+        "float64" if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+    expected = Series([3, 4, 6, np.nan, 8, 4, 1], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_len_mixed():
+    ser = Series(
+        ["a_b", np.nan, "asdf_cas_asdf", True, datetime.today(), "foo", None, 1, 2.0]
+    )
+    result = ser.str.len()
+    expected = Series([3, np.nan, 13, np.nan, np.nan, 3, np.nan, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method,sub,start,end,expected",
+    [
+        ("index", "EF", None, None, [4, 3, 1, 0]),
+        ("rindex", "EF", None, None, [4, 5, 7, 4]),
+        ("index", "EF", 3, None, [4, 3, 7, 4]),
+        ("rindex", "EF", 3, None, [4, 5, 7, 4]),
+        ("index", "E", 4, 8, [4, 5, 7, 4]),
+        ("rindex", "E", 0, 5, [4, 3, 1, 4]),
+    ],
+)
+def test_index(method, sub, start, end, index_or_series, any_string_dtype, expected):
+    obj = index_or_series(
+        ["ABCDEFG", "BCDEFEF", "DEFGHIJEF", "EFGHEF"], dtype=any_string_dtype
+    )
+    expected_dtype = (
+        np.int64 if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+    expected = index_or_series(expected, dtype=expected_dtype)
+
+    result = getattr(obj.str, method)(sub, start, end)
+
+    if index_or_series is Series:
+        tm.assert_series_equal(result, expected)
+    else:
+        tm.assert_index_equal(result, expected)
+
+    # compare with standard library
+    expected = [getattr(item, method)(sub, start, end) for item in obj]
+    assert list(result) == expected
+
+
+def test_index_not_found_raises(index_or_series, any_string_dtype):
+    obj = index_or_series(
+        ["ABCDEFG", "BCDEFEF", "DEFGHIJEF", "EFGHEF"], dtype=any_string_dtype
+    )
+    with pytest.raises(ValueError, match="substring not found"):
+        obj.str.index("DE")
+
+
+@pytest.mark.parametrize("method", ["index", "rindex"])
+def test_index_wrong_type_raises(index_or_series, any_string_dtype, method):
+    obj = index_or_series([], dtype=any_string_dtype)
+    msg = "expected a string object, not int"
+
+    with pytest.raises(TypeError, match=msg):
+        getattr(obj.str, method)(0)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        ["index", [1, 1, 0]],
+        ["rindex", [3, 1, 2]],
+    ],
+)
+def test_index_missing(any_string_dtype, method, exp):
+    ser = Series(["abcb", "ab", "bcbe", np.nan], dtype=any_string_dtype)
+    expected_dtype = (
+        np.float64 if is_object_or_nan_string_dtype(any_string_dtype) else "Int64"
+    )
+
+    result = getattr(ser.str, method)("b")
+    expected = Series(exp + [np.nan], dtype=expected_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_pipe_failures(any_string_dtype):
+    # #2119
+    ser = Series(["A|B|C"], dtype=any_string_dtype)
+
+    result = ser.str.split("|")
+    expected = Series([["A", "B", "C"]], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str.replace("|", " ", regex=False)
+    expected = Series(["A B C"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start, stop, step, expected",
+    [
+        (2, 5, None, ["foo", "bar", np.nan, "baz"]),
+        (0, 3, -1, ["", "", np.nan, ""]),
+        (None, None, -1, ["owtoofaa", "owtrabaa", np.nan, "xuqzabaa"]),
+        (None, 2, -1, ["owtoo", "owtra", np.nan, "xuqza"]),
+        (3, 10, 2, ["oto", "ato", np.nan, "aqx"]),
+        (3, 0, -1, ["ofa", "aba", np.nan, "aba"]),
+    ],
+)
+def test_slice(start, stop, step, expected, any_string_dtype):
+    ser = Series(["aafootwo", "aabartwo", np.nan, "aabazqux"], dtype=any_string_dtype)
+    result = ser.str.slice(start, stop, step)
+    expected = Series(expected, dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start, stop, step, expected",
+    [
+        (2, 5, None, ["foo", np.nan, "bar", np.nan, np.nan, None, np.nan, np.nan]),
+        (4, 1, -1, ["oof", np.nan, "rab", np.nan, np.nan, None, np.nan, np.nan]),
+    ],
+)
+def test_slice_mixed_object(start, stop, step, expected):
+    ser = Series(["aafootwo", np.nan, "aabartwo", True, datetime.today(), None, 1, 2.0])
+    result = ser.str.slice(start, stop, step)
+    expected = Series(expected, dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "start,stop,repl,expected",
+    [
+        (2, 3, None, ["shrt", "a it longer", "evnlongerthanthat", "", np.nan]),
+        (2, 3, "z", ["shzrt", "a zit longer", "evznlongerthanthat", "z", np.nan]),
+        (2, 2, "z", ["shzort", "a zbit longer", "evzenlongerthanthat", "z", np.nan]),
+        (2, 1, "z", ["shzort", "a zbit longer", "evzenlongerthanthat", "z", np.nan]),
+        (-1, None, "z", ["shorz", "a bit longez", "evenlongerthanthaz", "z", np.nan]),
+        (None, -2, "z", ["zrt", "zer", "zat", "z", np.nan]),
+        (6, 8, "z", ["shortz", "a bit znger", "evenlozerthanthat", "z", np.nan]),
+        (-10, 3, "z", ["zrt", "a zit longer", "evenlongzerthanthat", "z", np.nan]),
+    ],
+)
+def test_slice_replace(start, stop, repl, expected, any_string_dtype):
+    ser = Series(
+        ["short", "a bit longer", "evenlongerthanthat", "", np.nan],
+        dtype=any_string_dtype,
+    )
+    expected = Series(expected, dtype=any_string_dtype)
+    result = ser.str.slice_replace(start, stop, repl)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        ["strip", ["aa", "bb", np.nan, "cc"]],
+        ["lstrip", ["aa   ", "bb \n", np.nan, "cc  "]],
+        ["rstrip", ["  aa", " bb", np.nan, "cc"]],
+    ],
+)
+def test_strip_lstrip_rstrip(any_string_dtype, method, exp):
+    ser = Series(["  aa   ", " bb \n", np.nan, "cc  "], dtype=any_string_dtype)
+
+    result = getattr(ser.str, method)()
+    expected = Series(exp, dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        ["strip", ["aa", np.nan, "bb"]],
+        ["lstrip", ["aa  ", np.nan, "bb \t\n"]],
+        ["rstrip", ["  aa", np.nan, " bb"]],
+    ],
+)
+def test_strip_lstrip_rstrip_mixed_object(method, exp):
+    ser = Series(["  aa  ", np.nan, " bb \t\n", True, datetime.today(), None, 1, 2.0])
+
+    result = getattr(ser.str, method)()
+    expected = Series(exp + [np.nan, np.nan, None, np.nan, np.nan], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, exp",
+    [
+        ["strip", ["ABC", " BNSD", "LDFJH "]],
+        ["lstrip", ["ABCxx", " BNSD", "LDFJH xx"]],
+        ["rstrip", ["xxABC", "xx BNSD", "LDFJH "]],
+    ],
+)
+def test_strip_lstrip_rstrip_args(any_string_dtype, method, exp):
+    ser = Series(["xxABCxx", "xx BNSD", "LDFJH xx"], dtype=any_string_dtype)
+
+    result = getattr(ser.str, method)("x")
+    expected = Series(exp, dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "prefix, expected", [("a", ["b", " b c", "bc"]), ("ab", ["", "a b c", "bc"])]
+)
+def test_removeprefix(any_string_dtype, prefix, expected):
+    ser = Series(["ab", "a b c", "bc"], dtype=any_string_dtype)
+    result = ser.str.removeprefix(prefix)
+    ser_expected = Series(expected, dtype=any_string_dtype)
+    tm.assert_series_equal(result, ser_expected)
+
+
+@pytest.mark.parametrize(
+    "suffix, expected", [("c", ["ab", "a b ", "b"]), ("bc", ["ab", "a b c", ""])]
+)
+def test_removesuffix(any_string_dtype, suffix, expected):
+    ser = Series(["ab", "a b c", "bc"], dtype=any_string_dtype)
+    result = ser.str.removesuffix(suffix)
+    ser_expected = Series(expected, dtype=any_string_dtype)
+    tm.assert_series_equal(result, ser_expected)
+
+
+def test_string_slice_get_syntax(any_string_dtype):
+    ser = Series(
+        ["YYY", "B", "C", "YYYYYYbYYY", "BYYYcYYY", np.nan, "CYYYBYYY", "dog", "cYYYt"],
+        dtype=any_string_dtype,
+    )
+
+    result = ser.str[0]
+    expected = ser.str.get(0)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str[:3]
+    expected = ser.str.slice(stop=3)
+    tm.assert_series_equal(result, expected)
+
+    result = ser.str[2::-1]
+    expected = ser.str.slice(start=2, step=-1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_string_slice_out_of_bounds_nested():
+    ser = Series([(1, 2), (1,), (3, 4, 5)])
+    result = ser.str[1]
+    expected = Series([2, np.nan, 4])
+    tm.assert_series_equal(result, expected)
+
+
+def test_string_slice_out_of_bounds(any_string_dtype):
+    ser = Series(["foo", "b", "ba"], dtype=any_string_dtype)
+    result = ser.str[1]
+    expected = Series(["o", np.nan, "a"], dtype=any_string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_encode_decode(any_string_dtype):
+    ser = Series(["a", "b", "a\xe4"], dtype=any_string_dtype).str.encode("utf-8")
+    result = ser.str.decode("utf-8")
+    expected = Series(["a", "b", "a\xe4"], dtype="str")
+    tm.assert_series_equal(result, expected)
+
+
+def test_encode_errors_kwarg(any_string_dtype):
+    ser = Series(["a", "b", "a\x9d"], dtype=any_string_dtype)
+
+    msg = (
+        r"'charmap' codec can't encode character '\\x9d' in position 1: "
+        "character maps to <undefined>"
+    )
+    with pytest.raises(UnicodeEncodeError, match=msg):
+        ser.str.encode("cp1252")
+
+    result = ser.str.encode("cp1252", "ignore")
+    expected = ser.map(lambda x: x.encode("cp1252", "ignore"))
+    tm.assert_series_equal(result, expected)
+
+
+def test_decode_errors_kwarg():
+    ser = Series([b"a", b"b", b"a\x9d"])
+
+    msg = (
+        "'charmap' codec can't decode byte 0x9d in position 1: "
+        "character maps to <undefined>"
+    )
+    with pytest.raises(UnicodeDecodeError, match=msg):
+        ser.str.decode("cp1252")
+
+    result = ser.str.decode("cp1252", "ignore")
+    expected = ser.map(lambda x: x.decode("cp1252", "ignore")).astype("str")
+    tm.assert_series_equal(result, expected)
+
+
+def test_decode_string_dtype(string_dtype):
+    # https://github.com/pandas-dev/pandas/pull/60940
+    ser = Series([b"a", b"b"])
+    result = ser.str.decode("utf-8", dtype=string_dtype)
+    expected = Series(["a", "b"], dtype=string_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_decode_object_dtype(object_dtype):
+    # https://github.com/pandas-dev/pandas/pull/60940
+    ser = Series([b"a", rb"\ud800"])
+    result = ser.str.decode("utf-8", dtype=object_dtype)
+    expected = Series(["a", r"\ud800"], dtype=object_dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_decode_bad_dtype():
+    # https://github.com/pandas-dev/pandas/pull/60940
+    ser = Series([b"a", b"b"])
+    msg = "dtype must be string or object, got dtype='int64'"
+    with pytest.raises(ValueError, match=msg):
+        ser.str.decode("utf-8", dtype="int64")
+
+
+@pytest.mark.parametrize(
+    "form, expected",
+    [
+        ("NFKC", ["ABC", "ABC", "123", np.nan, "アイエ"]),
+        ("NFC", ["ABC", "ＡＢＣ", "１２３", np.nan, "ｱｲｴ"]),  # noqa: RUF001
+    ],
+)
+def test_normalize(form, expected, any_string_dtype):
+    ser = Series(
+        ["ABC", "ＡＢＣ", "１２３", np.nan, "ｱｲｴ"],  # noqa: RUF001
+        index=["a", "b", "c", "d", "e"],
+        dtype=any_string_dtype,
+    )
+    expected = Series(expected, index=["a", "b", "c", "d", "e"], dtype=any_string_dtype)
+    result = ser.str.normalize(form)
+    tm.assert_series_equal(result, expected)
+
+
+def test_normalize_bad_arg_raises(any_string_dtype):
+    ser = Series(
+        ["ABC", "ＡＢＣ", "１２３", np.nan, "ｱｲｴ"],  # noqa: RUF001
+        index=["a", "b", "c", "d", "e"],
+        dtype=any_string_dtype,
+    )
+    with pytest.raises(ValueError, match="invalid normalization form"):
+        ser.str.normalize("xxx")
+
+
+def test_normalize_index():
+    idx = Index(["ＡＢＣ", "１２３", "ｱｲｴ"])  # noqa: RUF001
+    expected = Index(["ABC", "123", "アイエ"])
+    result = idx.str.normalize("NFKC")
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values,inferred_type",
+    [
+        (["a", "b"], "string"),
+        (["a", "b", 1], "mixed-integer"),
+        (["a", "b", 1.3], "mixed"),
+        (["a", "b", 1.3, 1], "mixed-integer"),
+        (["aa", datetime(2011, 1, 1)], "mixed"),
+    ],
+)
+def test_index_str_accessor_visibility(values, inferred_type, index_or_series):
+    obj = index_or_series(values)
+    if index_or_series is Index:
+        assert obj.inferred_type == inferred_type
+
+    assert isinstance(obj.str, StringMethods)
+
+
+@pytest.mark.parametrize(
+    "values,inferred_type",
+    [
+        ([1, np.nan], "floating"),
+        ([datetime(2011, 1, 1)], "datetime64"),
+        ([timedelta(1)], "timedelta64"),
+    ],
+)
+def test_index_str_accessor_non_string_values_raises(
+    values, inferred_type, index_or_series
+):
+    obj = index_or_series(values)
+    if index_or_series is Index:
+        assert obj.inferred_type == inferred_type
+
+    msg = "Can only use .str accessor with string values"
+    with pytest.raises(AttributeError, match=msg):
+        obj.str
+
+
+def test_index_str_accessor_multiindex_raises():
+    # MultiIndex has mixed dtype, but not allow to use accessor
+    idx = MultiIndex.from_tuples([("a", "b"), ("a", "b")])
+    assert idx.inferred_type == "mixed"
+
+    msg = "Can only use .str accessor with Index, not MultiIndex"
+    with pytest.raises(AttributeError, match=msg):
+        idx.str
+
+
+def test_str_accessor_no_new_attributes(any_string_dtype):
+    # https://github.com/pandas-dev/pandas/issues/10673
+    ser = Series(list("aabbcde"), dtype=any_string_dtype)
+    with pytest.raises(AttributeError, match="You cannot add any new attribute"):
+        ser.str.xlabel = "a"
+
+
+def test_cat_on_bytes_raises():
+    lhs = Series(np.array(list("abc"), "S1").astype(object))
+    rhs = Series(np.array(list("def"), "S1").astype(object))
+    msg = "Cannot use .str.cat with values of inferred dtype 'bytes'"
+    with pytest.raises(TypeError, match=msg):
+        lhs.str.cat(rhs)
+
+
+def test_str_accessor_in_apply_func():
+    # https://github.com/pandas-dev/pandas/issues/38979
+    df = DataFrame(zip("abc", "def"))
+    expected = Series(["A/D", "B/E", "C/F"])
+    result = df.apply(lambda f: "/".join(f.str.upper()), axis=1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_zfill():
+    # https://github.com/pandas-dev/pandas/issues/20868
+    value = Series(["-1", "1", "1000", 10, np.nan])
+    expected = Series(["-01", "001", "1000", np.nan, np.nan], dtype=object)
+    tm.assert_series_equal(value.str.zfill(3), expected)
+
+    value = Series(["-2", "+5"])
+    expected = Series(["-0002", "+0005"])
+    tm.assert_series_equal(value.str.zfill(5), expected)
+
+
+def test_zfill_with_non_integer_argument():
+    value = Series(["-2", "+5"])
+    wid = "a"
+    msg = f"width must be of integer type, not {type(wid).__name__}"
+    with pytest.raises(TypeError, match=msg):
+        value.str.zfill(wid)
+
+
+def test_zfill_with_leading_sign():
+    value = Series(["-cat", "-1", "+dog"])
+    expected = Series(["-0cat", "-0001", "+0dog"])
+    tm.assert_series_equal(value.str.zfill(5), expected)
+
+
+def test_get_with_dict_label():
+    # GH47911
+    s = Series(
+        [
+            {"name": "Hello", "value": "World"},
+            {"name": "Goodbye", "value": "Planet"},
+            {"value": "Sea"},
+        ]
+    )
+    result = s.str.get("name")
+    expected = Series(["Hello", "Goodbye", None], dtype=object)
+    tm.assert_series_equal(result, expected)
+    result = s.str.get("value")
+    expected = Series(["World", "Planet", "Sea"], dtype=object)
+    tm.assert_series_equal(result, expected)
+
+
+def test_series_str_decode():
+    # GH 22613
+    result = Series([b"x", b"y"]).str.decode(encoding="UTF-8", errors="strict")
+    expected = Series(["x", "y"], dtype="str")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/test_aggregation.py b/pandas/pandas/tests/test_aggregation.py
new file mode 100644
index 0000000000000..3a01805cc2365
--- /dev/null
+++ b/pandas/pandas/tests/test_aggregation.py
@@ -0,0 +1,93 @@
+import numpy as np
+import pytest
+
+from pandas.core.apply import (
+    _make_unique_kwarg_list,
+    maybe_mangle_lambdas,
+)
+
+
+def test_maybe_mangle_lambdas_passthrough():
+    assert maybe_mangle_lambdas("mean") == "mean"
+    assert maybe_mangle_lambdas(lambda x: x).__name__ == "<lambda>"
+    # don't mangle single lambda.
+    assert maybe_mangle_lambdas([lambda x: x])[0].__name__ == "<lambda>"
+
+
+def test_maybe_mangle_lambdas_listlike():
+    aggfuncs = [lambda x: 1, lambda x: 2]
+    result = maybe_mangle_lambdas(aggfuncs)
+    assert result[0].__name__ == "<lambda_0>"
+    assert result[1].__name__ == "<lambda_1>"
+    assert aggfuncs[0](None) == result[0](None)
+    assert aggfuncs[1](None) == result[1](None)
+
+
+def test_maybe_mangle_lambdas():
+    func = {"A": [lambda x: 0, lambda x: 1]}
+    result = maybe_mangle_lambdas(func)
+    assert result["A"][0].__name__ == "<lambda_0>"
+    assert result["A"][1].__name__ == "<lambda_1>"
+
+
+def test_maybe_mangle_lambdas_args():
+    func = {"A": [lambda x, a, b=1: (0, a, b), lambda x: 1]}
+    result = maybe_mangle_lambdas(func)
+    assert result["A"][0].__name__ == "<lambda_0>"
+    assert result["A"][1].__name__ == "<lambda_1>"
+
+    assert func["A"][0](0, 1) == (0, 1, 1)
+    assert func["A"][0](0, 1, 2) == (0, 1, 2)
+    assert func["A"][0](0, 2, b=3) == (0, 2, 3)
+
+
+def test_maybe_mangle_lambdas_named():
+    func = {"C": np.mean, "D": {"foo": np.mean, "bar": np.mean}}
+    result = maybe_mangle_lambdas(func)
+    assert result == func
+
+
+@pytest.mark.parametrize(
+    "order, expected_reorder",
+    [
+        (
+            [
+                ("height", "<lambda>"),
+                ("height", "max"),
+                ("weight", "max"),
+                ("height", "<lambda>"),
+                ("weight", "<lambda>"),
+            ],
+            [
+                ("height", "<lambda>_0"),
+                ("height", "max"),
+                ("weight", "max"),
+                ("height", "<lambda>_1"),
+                ("weight", "<lambda>"),
+            ],
+        ),
+        (
+            [
+                ("col2", "min"),
+                ("col1", "<lambda>"),
+                ("col1", "<lambda>"),
+                ("col1", "<lambda>"),
+            ],
+            [
+                ("col2", "min"),
+                ("col1", "<lambda>_0"),
+                ("col1", "<lambda>_1"),
+                ("col1", "<lambda>_2"),
+            ],
+        ),
+        (
+            [("col", "<lambda>"), ("col", "<lambda>"), ("col", "<lambda>")],
+            [("col", "<lambda>_0"), ("col", "<lambda>_1"), ("col", "<lambda>_2")],
+        ),
+    ],
+)
+def test_make_unique(order, expected_reorder):
+    # GH 27519, test if make_unique function reorders correctly
+    result = _make_unique_kwarg_list(order)
+
+    assert result == expected_reorder
diff --git a/pandas/pandas/tests/test_algos.py b/pandas/pandas/tests/test_algos.py
new file mode 100644
index 0000000000000..611b92eb022d6
--- /dev/null
+++ b/pandas/pandas/tests/test_algos.py
@@ -0,0 +1,2033 @@
+from datetime import datetime
+import struct
+
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    algos as libalgos,
+    hashtable as ht,
+)
+
+from pandas.core.dtypes.common import (
+    is_bool_dtype,
+    is_complex_dtype,
+    is_float_dtype,
+    is_integer_dtype,
+    is_object_dtype,
+)
+from pandas.core.dtypes.dtypes import (
+    CategoricalDtype,
+    DatetimeTZDtype,
+)
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    CategoricalIndex,
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    IntervalIndex,
+    MultiIndex,
+    NaT,
+    Period,
+    PeriodIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    cut,
+    date_range,
+    timedelta_range,
+    to_datetime,
+    to_timedelta,
+)
+import pandas._testing as tm
+import pandas.core.algorithms as algos
+from pandas.core.arrays import (
+    DatetimeArray,
+    TimedeltaArray,
+)
+import pandas.core.common as com
+
+
+class TestFactorize:
+    def test_factorize_complex(self):
+        # GH#17927
+        array = np.array([1, 2, 2 + 1j], dtype=complex)
+        labels, uniques = algos.factorize(array)
+
+        expected_labels = np.array([0, 1, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(labels, expected_labels)
+
+        expected_uniques = np.array([(1 + 0j), (2 + 0j), (2 + 1j)], dtype=complex)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_factorize(self, index_or_series_obj, sort):
+        obj = index_or_series_obj
+        result_codes, result_uniques = obj.factorize(sort=sort)
+
+        constructor = Index
+        if isinstance(obj, MultiIndex):
+            constructor = MultiIndex.from_tuples
+        expected_arr = obj.unique()
+        if expected_arr.dtype == np.float16:
+            expected_arr = expected_arr.astype(np.float32)
+        expected_uniques = constructor(expected_arr)
+        if (
+            isinstance(obj, Index)
+            and expected_uniques.dtype == bool
+            and obj.dtype == object
+        ):
+            expected_uniques = expected_uniques.astype(object)
+
+        if sort:
+            expected_uniques = expected_uniques.sort_values()
+
+        # construct an integer ndarray so that
+        # `expected_uniques.take(expected_codes)` is equal to `obj`
+        expected_uniques_list = list(expected_uniques)
+        expected_codes = [expected_uniques_list.index(val) for val in obj]
+        expected_codes = np.asarray(expected_codes, dtype=np.intp)
+
+        tm.assert_numpy_array_equal(result_codes, expected_codes)
+        tm.assert_index_equal(result_uniques, expected_uniques, exact=True)
+
+    def test_series_factorize_use_na_sentinel_false(self):
+        # GH#35667
+        values = np.array([1, 2, 1, np.nan])
+        ser = Series(values)
+        codes, uniques = ser.factorize(use_na_sentinel=False)
+
+        expected_codes = np.array([0, 1, 0, 2], dtype=np.intp)
+        expected_uniques = Index([1.0, 2.0, np.nan])
+
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_index_equal(uniques, expected_uniques)
+
+    def test_basic(self):
+        items = np.array(["a", "b", "b", "a", "a", "c", "c", "c"], dtype=object)
+        codes, uniques = algos.factorize(items)
+        tm.assert_numpy_array_equal(uniques, np.array(["a", "b", "c"], dtype=object))
+
+        codes, uniques = algos.factorize(items, sort=True)
+        exp = np.array([0, 1, 1, 0, 0, 2, 2, 2], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = np.array(["a", "b", "c"], dtype=object)
+        tm.assert_numpy_array_equal(uniques, exp)
+
+        arr = np.arange(5, dtype=np.intp)[::-1]
+
+        codes, uniques = algos.factorize(arr)
+        exp = np.array([0, 1, 2, 3, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = np.array([4, 3, 2, 1, 0], dtype=arr.dtype)
+        tm.assert_numpy_array_equal(uniques, exp)
+
+        codes, uniques = algos.factorize(arr, sort=True)
+        exp = np.array([4, 3, 2, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = np.array([0, 1, 2, 3, 4], dtype=arr.dtype)
+        tm.assert_numpy_array_equal(uniques, exp)
+
+        arr = np.arange(5.0)[::-1]
+
+        codes, uniques = algos.factorize(arr)
+        exp = np.array([0, 1, 2, 3, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = np.array([4.0, 3.0, 2.0, 1.0, 0.0], dtype=arr.dtype)
+        tm.assert_numpy_array_equal(uniques, exp)
+
+        codes, uniques = algos.factorize(arr, sort=True)
+        exp = np.array([4, 3, 2, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = np.array([0.0, 1.0, 2.0, 3.0, 4.0], dtype=arr.dtype)
+        tm.assert_numpy_array_equal(uniques, exp)
+
+    def test_mixed(self):
+        # doc example reshaping.rst
+        x = Series(["A", "A", np.nan, "B", 3.14, np.inf])
+        codes, uniques = algos.factorize(x)
+
+        exp = np.array([0, 0, -1, 1, 2, 3], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = Index(["A", "B", 3.14, np.inf])
+        tm.assert_index_equal(uniques, exp)
+
+        codes, uniques = algos.factorize(x, sort=True)
+        exp = np.array([2, 2, -1, 3, 0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = Index([3.14, np.inf, "A", "B"])
+        tm.assert_index_equal(uniques, exp)
+
+    def test_factorize_datetime64(self):
+        # M8
+        v1 = Timestamp("20130101 09:00:00.00004")
+        v2 = Timestamp("20130101")
+        x = Series([v1, v1, v1, v2, v2, v1])
+        codes, uniques = algos.factorize(x)
+
+        exp = np.array([0, 0, 0, 1, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = DatetimeIndex([v1, v2])
+        tm.assert_index_equal(uniques, exp)
+
+        codes, uniques = algos.factorize(x, sort=True)
+        exp = np.array([1, 1, 1, 0, 0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        exp = DatetimeIndex([v2, v1])
+        tm.assert_index_equal(uniques, exp)
+
+    def test_factorize_period(self):
+        # period
+        v1 = Period("201302", freq="M")
+        v2 = Period("201303", freq="M")
+        x = Series([v1, v1, v1, v2, v2, v1])
+
+        # periods are not 'sorted' as they are converted back into an index
+        codes, uniques = algos.factorize(x)
+        exp = np.array([0, 0, 0, 1, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        tm.assert_index_equal(uniques, PeriodIndex([v1, v2]))
+
+        codes, uniques = algos.factorize(x, sort=True)
+        exp = np.array([0, 0, 0, 1, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        tm.assert_index_equal(uniques, PeriodIndex([v1, v2]))
+
+    def test_factorize_timedelta(self):
+        # GH 5986
+        v1 = to_timedelta("1 day 1 min")
+        v2 = to_timedelta("1 day")
+        x = Series([v1, v2, v1, v1, v2, v2, v1])
+        codes, uniques = algos.factorize(x)
+        exp = np.array([0, 1, 0, 0, 1, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        tm.assert_index_equal(uniques, to_timedelta([v1, v2]))
+
+        codes, uniques = algos.factorize(x, sort=True)
+        exp = np.array([1, 0, 1, 1, 0, 0, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, exp)
+        tm.assert_index_equal(uniques, to_timedelta([v2, v1]))
+
+    def test_factorize_nan(self):
+        # nan should map to na_sentinel, not reverse_indexer[na_sentinel]
+        # rizer.factorize should not raise an exception if na_sentinel indexes
+        # outside of reverse_indexer
+        key = np.array([1, 2, 1, np.nan], dtype="O")
+        rizer = ht.ObjectFactorizer(len(key))
+        for na_sentinel in (-1, 20):
+            ids = rizer.factorize(key, na_sentinel=na_sentinel)
+            expected = np.array([0, 1, 0, na_sentinel], dtype=np.intp)
+            assert len(set(key)) == len(set(expected))
+            tm.assert_numpy_array_equal(pd.isna(key), expected == na_sentinel)
+            tm.assert_numpy_array_equal(ids, expected)
+
+    def test_factorizer_with_mask(self):
+        # GH#49549
+        data = np.array([1, 2, 3, 1, 1, 0], dtype="int64")
+        mask = np.array([False, False, False, False, False, True])
+        rizer = ht.Int64Factorizer(len(data))
+        result = rizer.factorize(data, mask=mask)
+        expected = np.array([0, 1, 2, 0, 0, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+        expected_uniques = np.array([1, 2, 3], dtype="int64")
+        tm.assert_numpy_array_equal(rizer.uniques.to_array(), expected_uniques)
+
+    def test_factorizer_object_with_nan(self):
+        # GH#49549
+        data = np.array([1, 2, 3, 1, np.nan])
+        rizer = ht.ObjectFactorizer(len(data))
+        result = rizer.factorize(data.astype(object))
+        expected = np.array([0, 1, 2, 0, -1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+        expected_uniques = np.array([1, 2, 3], dtype=object)
+        tm.assert_numpy_array_equal(rizer.uniques.to_array(), expected_uniques)
+
+    @pytest.mark.parametrize(
+        "data, expected_codes, expected_uniques",
+        [
+            (
+                [(1, 1), (1, 2), (0, 0), (1, 2), "nonsense"],
+                [0, 1, 2, 1, 3],
+                [(1, 1), (1, 2), (0, 0), "nonsense"],
+            ),
+            (
+                [(1, 1), (1, 2), (0, 0), (1, 2), (1, 2, 3)],
+                [0, 1, 2, 1, 3],
+                [(1, 1), (1, 2), (0, 0), (1, 2, 3)],
+            ),
+            ([(1, 1), (1, 2), (0, 0), (1, 2)], [0, 1, 2, 1], [(1, 1), (1, 2), (0, 0)]),
+        ],
+    )
+    def test_factorize_tuple_list(self, data, expected_codes, expected_uniques):
+        # GH9454
+        data = com.asarray_tuplesafe(data, dtype=object)
+        codes, uniques = pd.factorize(data)
+
+        tm.assert_numpy_array_equal(codes, np.array(expected_codes, dtype=np.intp))
+
+        expected_uniques_array = com.asarray_tuplesafe(expected_uniques, dtype=object)
+        tm.assert_numpy_array_equal(uniques, expected_uniques_array)
+
+    def test_complex_sorting(self):
+        # gh 12666 - check no segfault
+        x17 = np.array([complex(i) for i in range(17)], dtype=object)
+
+        msg = "'[<>]' not supported between instances of .*"
+        with pytest.raises(TypeError, match=msg):
+            algos.factorize(x17[::-1], sort=True)
+
+    def test_numeric_dtype_factorize(self, any_real_numpy_dtype):
+        # GH41132
+        dtype = any_real_numpy_dtype
+        data = np.array([1, 2, 2, 1], dtype=dtype)
+        expected_codes = np.array([0, 1, 1, 0], dtype=np.intp)
+        expected_uniques = np.array([1, 2], dtype=dtype)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_float64_factorize(self, writable):
+        data = np.array([1.0, 1e8, 1.0, 1e-8, 1e8, 1.0], dtype=np.float64)
+        data.setflags(write=writable)
+        expected_codes = np.array([0, 1, 0, 2, 1, 0], dtype=np.intp)
+        expected_uniques = np.array([1.0, 1e8, 1e-8], dtype=np.float64)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_uint64_factorize(self, writable):
+        data = np.array([2**64 - 1, 1, 2**64 - 1], dtype=np.uint64)
+        data.setflags(write=writable)
+        expected_codes = np.array([0, 1, 0], dtype=np.intp)
+        expected_uniques = np.array([2**64 - 1, 1], dtype=np.uint64)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_int64_factorize(self, writable):
+        data = np.array([2**63 - 1, -(2**63), 2**63 - 1], dtype=np.int64)
+        data.setflags(write=writable)
+        expected_codes = np.array([0, 1, 0], dtype=np.intp)
+        expected_uniques = np.array([2**63 - 1, -(2**63)], dtype=np.int64)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_string_factorize(self, writable):
+        data = np.array(["a", "c", "a", "b", "c"], dtype=object)
+        data.setflags(write=writable)
+        expected_codes = np.array([0, 1, 0, 2, 1], dtype=np.intp)
+        expected_uniques = np.array(["a", "c", "b"], dtype=object)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_object_factorize(self, writable):
+        data = np.array(["a", "c", None, np.nan, "a", "b", NaT, "c"], dtype=object)
+        data.setflags(write=writable)
+        expected_codes = np.array([0, 1, -1, -1, 0, 2, -1, 1], dtype=np.intp)
+        expected_uniques = np.array(["a", "c", "b"], dtype=object)
+
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_datetime64_factorize(self, writable):
+        # GH35650 Verify whether read-only datetime64 array can be factorized
+        data = np.array([np.datetime64("2020-01-01T00:00:00.000")], dtype="M8[ns]")
+        data.setflags(write=writable)
+        expected_codes = np.array([0], dtype=np.intp)
+        expected_uniques = np.array(
+            ["2020-01-01T00:00:00.000000000"], dtype="datetime64[ns]"
+        )
+
+        codes, uniques = pd.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    def test_factorize_rangeindex(self, sort):
+        # increasing -> sort doesn't matter
+        ri = pd.RangeIndex.from_range(range(10))
+        expected = np.arange(10, dtype=np.intp), ri
+
+        result = algos.factorize(ri, sort=sort)
+        tm.assert_numpy_array_equal(result[0], expected[0])
+        tm.assert_index_equal(result[1], expected[1], exact=True)
+
+        result = ri.factorize(sort=sort)
+        tm.assert_numpy_array_equal(result[0], expected[0])
+        tm.assert_index_equal(result[1], expected[1], exact=True)
+
+    def test_factorize_rangeindex_decreasing(self, sort):
+        # decreasing -> sort matters
+        ri = pd.RangeIndex.from_range(range(10))
+        expected = np.arange(10, dtype=np.intp), ri
+
+        ri2 = ri[::-1]
+        expected = expected[0], ri2
+        if sort:
+            expected = expected[0][::-1], expected[1][::-1]
+
+        result = algos.factorize(ri2, sort=sort)
+        tm.assert_numpy_array_equal(result[0], expected[0])
+        tm.assert_index_equal(result[1], expected[1], exact=True)
+
+        result = ri2.factorize(sort=sort)
+        tm.assert_numpy_array_equal(result[0], expected[0])
+        tm.assert_index_equal(result[1], expected[1], exact=True)
+
+    def test_deprecate_order(self):
+        # gh 19727 - check warning is raised for deprecated keyword, order.
+        # Test not valid once order keyword is removed.
+        data = np.array([2**63, 1, 2**63], dtype=np.uint64)
+        with pytest.raises(TypeError, match="got an unexpected keyword"):
+            algos.factorize(data, order=True)
+        with tm.assert_produces_warning(False):
+            algos.factorize(data)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            np.array([0, 1, 0], dtype="u8"),
+            np.array([-(2**63), 1, -(2**63)], dtype="i8"),
+            np.array(["__nan__", "foo", "__nan__"], dtype="object"),
+        ],
+    )
+    def test_parametrized_factorize_na_value_default(self, data):
+        # arrays that include the NA default for that type, but isn't used.
+        codes, uniques = algos.factorize(data)
+        expected_uniques = data[[0, 1]]
+        expected_codes = np.array([0, 1, 0], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    @pytest.mark.parametrize(
+        "data, na_value",
+        [
+            (np.array([0, 1, 0, 2], dtype="u8"), 0),
+            (np.array([1, 0, 1, 2], dtype="u8"), 1),
+            (np.array([-(2**63), 1, -(2**63), 0], dtype="i8"), -(2**63)),
+            (np.array([1, -(2**63), 1, 0], dtype="i8"), 1),
+            (np.array(["a", "", "a", "b"], dtype=object), "a"),
+            (np.array([(), ("a", 1), (), ("a", 2)], dtype=object), ()),
+            (np.array([("a", 1), (), ("a", 1), ("a", 2)], dtype=object), ("a", 1)),
+        ],
+    )
+    def test_parametrized_factorize_na_value(self, data, na_value):
+        codes, uniques = algos.factorize_array(data, na_value=na_value)
+        expected_uniques = data[[1, 3]]
+        expected_codes = np.array([-1, 0, -1, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_numpy_array_equal(uniques, expected_uniques)
+
+    @pytest.mark.parametrize(
+        "data, uniques",
+        [
+            (
+                np.array(["b", "a", None, "b"], dtype=object),
+                np.array(["b", "a"], dtype=object),
+            ),
+            (
+                pd.array([2, 1, np.nan, 2], dtype="Int64"),
+                pd.array([2, 1], dtype="Int64"),
+            ),
+        ],
+        ids=["numpy_array", "extension_array"],
+    )
+    def test_factorize_use_na_sentinel(self, sort, data, uniques):
+        codes, uniques = algos.factorize(data, sort=sort, use_na_sentinel=True)
+        if sort:
+            expected_codes = np.array([1, 0, -1, 1], dtype=np.intp)
+            expected_uniques = algos.safe_sort(uniques)
+        else:
+            expected_codes = np.array([0, 1, -1, 0], dtype=np.intp)
+            expected_uniques = uniques
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        if isinstance(data, np.ndarray):
+            tm.assert_numpy_array_equal(uniques, expected_uniques)
+        else:
+            tm.assert_extension_array_equal(uniques, expected_uniques)
+
+    @pytest.mark.parametrize(
+        "data, expected_codes, expected_uniques",
+        [
+            (
+                ["a", None, "b", "a"],
+                np.array([0, 1, 2, 0], dtype=np.dtype("intp")),
+                np.array(["a", np.nan, "b"], dtype=object),
+            ),
+            (
+                ["a", np.nan, "b", "a"],
+                np.array([0, 1, 2, 0], dtype=np.dtype("intp")),
+                np.array(["a", np.nan, "b"], dtype=object),
+            ),
+        ],
+    )
+    def test_object_factorize_use_na_sentinel_false(
+        self, data, expected_codes, expected_uniques
+    ):
+        codes, uniques = algos.factorize(
+            np.array(data, dtype=object), use_na_sentinel=False
+        )
+
+        tm.assert_numpy_array_equal(uniques, expected_uniques, strict_nan=True)
+        tm.assert_numpy_array_equal(codes, expected_codes, strict_nan=True)
+
+    @pytest.mark.parametrize(
+        "data, expected_codes, expected_uniques",
+        [
+            (
+                np.array([1, None, 1, 2], dtype=object),
+                np.array([0, 1, 0, 2], dtype=np.dtype("intp")),
+                np.array([1, np.nan, 2], dtype="O"),
+            ),
+            (
+                np.array([1, np.nan, 1, 2], dtype=np.float64),
+                np.array([0, 1, 0, 2], dtype=np.dtype("intp")),
+                np.array([1, np.nan, 2], dtype=np.float64),
+            ),
+        ],
+    )
+    def test_int_factorize_use_na_sentinel_false(
+        self, data, expected_codes, expected_uniques
+    ):
+        codes, uniques = algos.factorize(data, use_na_sentinel=False)
+
+        tm.assert_numpy_array_equal(uniques, expected_uniques, strict_nan=True)
+        tm.assert_numpy_array_equal(codes, expected_codes, strict_nan=True)
+
+    @pytest.mark.parametrize(
+        "data, expected_codes, expected_uniques",
+        [
+            (
+                Index(Categorical(["a", "a", "b"])),
+                np.array([0, 0, 1], dtype=np.intp),
+                CategoricalIndex(["a", "b"], categories=["a", "b"], dtype="category"),
+            ),
+            (
+                Series(Categorical(["a", "a", "b"])),
+                np.array([0, 0, 1], dtype=np.intp),
+                CategoricalIndex(["a", "b"], categories=["a", "b"], dtype="category"),
+            ),
+            (
+                Series(DatetimeIndex(["2017", "2017"], tz="US/Eastern")),
+                np.array([0, 0], dtype=np.intp),
+                DatetimeIndex(["2017"], tz="US/Eastern"),
+            ),
+        ],
+    )
+    def test_factorize_mixed_values(self, data, expected_codes, expected_uniques):
+        # GH 19721
+        codes, uniques = algos.factorize(data)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+        tm.assert_index_equal(uniques, expected_uniques)
+
+    def test_factorize_interval_non_nano(self, unit):
+        # GH#56099
+        left = DatetimeIndex(["2016-01-01", np.nan, "2015-10-11"]).as_unit(unit)
+        right = DatetimeIndex(["2016-01-02", np.nan, "2015-10-15"]).as_unit(unit)
+        idx = IntervalIndex.from_arrays(left, right)
+        codes, cats = idx.factorize()
+        assert cats.dtype == f"interval[datetime64[{unit}], right]"
+
+        ts = Timestamp(0).as_unit(unit)
+        idx2 = IntervalIndex.from_arrays(left - ts, right - ts)
+        codes2, cats2 = idx2.factorize()
+        assert cats2.dtype == f"interval[timedelta64[{unit}], right]"
+
+        idx3 = IntervalIndex.from_arrays(
+            left.tz_localize("US/Pacific"), right.tz_localize("US/Pacific")
+        )
+        codes3, cats3 = idx3.factorize()
+        assert cats3.dtype == f"interval[datetime64[{unit}, US/Pacific], right]"
+
+
+class TestUnique:
+    def test_ints(self):
+        arr = np.random.default_rng(2).integers(0, 100, size=50)
+
+        result = algos.unique(arr)
+        assert isinstance(result, np.ndarray)
+
+    def test_objects(self):
+        arr = np.random.default_rng(2).integers(0, 100, size=50).astype("O")
+
+        result = algos.unique(arr)
+        assert isinstance(result, np.ndarray)
+
+    def test_object_refcount_bug(self):
+        lst = np.array(["A", "B", "C", "D", "E"], dtype=object)
+        for i in range(1000):
+            len(algos.unique(lst))
+
+    def test_index_returned(self, index):
+        # GH#57043
+        index = index.repeat(2)
+        result = algos.unique(index)
+
+        # dict.fromkeys preserves the order
+        unique_values = list(dict.fromkeys(index.values))
+        if isinstance(index, MultiIndex):
+            expected = MultiIndex.from_tuples(unique_values, names=index.names)
+        else:
+            expected = Index(unique_values, dtype=index.dtype)
+            if isinstance(index.dtype, DatetimeTZDtype):
+                expected = expected.normalize()
+        tm.assert_index_equal(result, expected, exact=True)
+
+    def test_dtype_preservation(self, any_numpy_dtype):
+        # GH 15442
+        if any_numpy_dtype in (tm.BYTES_DTYPES + tm.STRING_DTYPES):
+            data = [1, 2, 2]
+            uniques = [1, 2]
+        elif is_integer_dtype(any_numpy_dtype):
+            data = [1, 2, 2]
+            uniques = [1, 2]
+        elif is_float_dtype(any_numpy_dtype):
+            data = [1, 2, 2]
+            uniques = [1.0, 2.0]
+        elif is_complex_dtype(any_numpy_dtype):
+            data = [complex(1, 0), complex(2, 0), complex(2, 0)]
+            uniques = [complex(1, 0), complex(2, 0)]
+        elif is_bool_dtype(any_numpy_dtype):
+            data = [True, True, False]
+            uniques = [True, False]
+        elif is_object_dtype(any_numpy_dtype):
+            data = ["A", "B", "B"]
+            uniques = ["A", "B"]
+        else:
+            # datetime64[ns]/M8[ns]/timedelta64[ns]/m8[ns] tested elsewhere
+            data = [1, 2, 2]
+            uniques = [1, 2]
+
+        result = Series(data, dtype=any_numpy_dtype).unique()
+        expected = np.array(uniques, dtype=any_numpy_dtype)
+
+        if any_numpy_dtype in tm.STRING_DTYPES:
+            expected = expected.astype(object)
+
+        if expected.dtype.kind in ["m", "M"]:
+            # We get TimedeltaArray/DatetimeArray
+            assert isinstance(result, (DatetimeArray, TimedeltaArray))
+            result = np.array(result)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_datetime64_dtype_array_returned(self):
+        # GH 9431
+        dt_arr = np.array(
+            [
+                "2015-01-03T00:00:00.000000000",
+                "2015-01-01T00:00:00.000000000",
+            ],
+            dtype="M8[ns]",
+        )
+
+        dt_index = to_datetime(
+            [
+                "2015-01-03T00:00:00.000000000",
+                "2015-01-01T00:00:00.000000000",
+                "2015-01-01T00:00:00.000000000",
+            ]
+        )
+        result = algos.unique(dt_index)
+        expected = to_datetime(dt_arr)
+        tm.assert_index_equal(result, expected, exact=True)
+
+        s = Series(dt_index)
+        result = algos.unique(s)
+        tm.assert_numpy_array_equal(result, dt_arr)
+        assert result.dtype == dt_arr.dtype
+
+        arr = s.values
+        result = algos.unique(arr)
+        tm.assert_numpy_array_equal(result, dt_arr)
+        assert result.dtype == dt_arr.dtype
+
+    def test_datetime_non_ns(self):
+        a = np.array(["2000", "2000", "2001"], dtype="datetime64[s]")
+        result = pd.unique(a)
+        expected = np.array(["2000", "2001"], dtype="datetime64[s]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_timedelta_non_ns(self):
+        a = np.array(["2000", "2000", "2001"], dtype="timedelta64[s]")
+        result = pd.unique(a)
+        expected = np.array([2000, 2001], dtype="timedelta64[s]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_timedelta64_dtype_array_returned(self):
+        # GH 9431
+        td_arr = np.array([31200, 45678, 10000], dtype="m8[ns]")
+
+        td_index = to_timedelta([31200, 45678, 31200, 10000, 45678])
+        result = algos.unique(td_index)
+        expected = to_timedelta(td_arr)
+        tm.assert_index_equal(result, expected)
+        assert result.dtype == expected.dtype
+
+        s = Series(td_index)
+        result = algos.unique(s)
+        tm.assert_numpy_array_equal(result, td_arr)
+        assert result.dtype == td_arr.dtype
+
+        arr = s.values
+        result = algos.unique(arr)
+        tm.assert_numpy_array_equal(result, td_arr)
+        assert result.dtype == td_arr.dtype
+
+    def test_uint64_overflow(self):
+        s = Series([1, 2, 2**63, 2**63], dtype=np.uint64)
+        exp = np.array([1, 2, 2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(algos.unique(s), exp)
+
+    def test_nan_in_object_array(self):
+        duplicated_items = ["a", np.nan, "c", "c"]
+        result = pd.unique(np.array(duplicated_items, dtype=object))
+        expected = np.array(["a", np.nan, "c"], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_categorical(self):
+        # we are expecting to return in the order
+        # of appearance
+        expected = Categorical(list("bac"))
+
+        # we are expecting to return in the order
+        # of the categories
+        expected_o = Categorical(list("bac"), categories=list("abc"), ordered=True)
+
+        # GH 15939
+        c = Categorical(list("baabc"))
+        result = c.unique()
+        tm.assert_categorical_equal(result, expected)
+
+        result = algos.unique(c)
+        tm.assert_categorical_equal(result, expected)
+
+        c = Categorical(list("baabc"), ordered=True)
+        result = c.unique()
+        tm.assert_categorical_equal(result, expected_o)
+
+        result = algos.unique(c)
+        tm.assert_categorical_equal(result, expected_o)
+
+        # Series of categorical dtype
+        s = Series(Categorical(list("baabc")), name="foo")
+        result = s.unique()
+        tm.assert_categorical_equal(result, expected)
+
+        result = pd.unique(s)
+        tm.assert_categorical_equal(result, expected)
+
+        # CI -> return CI
+        ci = CategoricalIndex(Categorical(list("baabc"), categories=list("abc")))
+        expected = CategoricalIndex(expected)
+        result = ci.unique()
+        tm.assert_index_equal(result, expected)
+
+        result = pd.unique(ci)
+        tm.assert_index_equal(result, expected)
+
+    def test_datetime64tz_aware(self, unit):
+        # GH 15939
+
+        dti = Index(
+            [
+                Timestamp("20160101", tz="US/Eastern"),
+                Timestamp("20160101", tz="US/Eastern"),
+            ]
+        ).as_unit(unit)
+        ser = Series(dti)
+
+        result = ser.unique()
+        expected = dti[:1]._data
+        tm.assert_extension_array_equal(result, expected)
+
+        result = dti.unique()
+        expected = dti[:1]
+        tm.assert_index_equal(result, expected)
+
+        result = pd.unique(ser)
+        expected = dti[:1]._data
+        tm.assert_extension_array_equal(result, expected)
+
+        result = pd.unique(dti)
+        expected = dti[:1]
+        tm.assert_index_equal(result, expected)
+
+    def test_order_of_appearance(self):
+        # 9346
+        # light testing of guarantee of order of appearance
+        # these also are the doc-examples
+        result = pd.unique(Series([2, 1, 3, 3]))
+        tm.assert_numpy_array_equal(result, np.array([2, 1, 3], dtype="int64"))
+
+        result = pd.unique(Series([2] + [1] * 5))
+        tm.assert_numpy_array_equal(result, np.array([2, 1], dtype="int64"))
+
+        data = np.array(["a", "a", "b", "c"], dtype=object)
+        result = pd.unique(data)
+        expected = np.array(["a", "b", "c"], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = pd.unique(Series(Categorical(list("aabc"))))
+        expected = Categorical(list("abc"))
+        tm.assert_categorical_equal(result, expected)
+
+    def test_order_of_appearance_dt64(self, unit):
+        ser = Series([Timestamp("20160101"), Timestamp("20160101")]).dt.as_unit(unit)
+        result = pd.unique(ser)
+        expected = np.array(["2016-01-01T00:00:00.000000000"], dtype=f"M8[{unit}]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_order_of_appearance_dt64tz(self, unit):
+        dti = DatetimeIndex(
+            [
+                Timestamp("20160101", tz="US/Eastern"),
+                Timestamp("20160101", tz="US/Eastern"),
+            ]
+        ).as_unit(unit)
+        result = pd.unique(dti)
+        expected = DatetimeIndex(
+            ["2016-01-01 00:00:00"], dtype=f"datetime64[{unit}, US/Eastern]", freq=None
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arg ,expected",
+        [
+            (("1", "1", "2"), np.array(["1", "2"], dtype=object)),
+            (("foo",), np.array(["foo"], dtype=object)),
+        ],
+    )
+    def test_tuple_with_strings(self, arg, expected):
+        # see GH 17108
+        arg = com.asarray_tuplesafe(arg, dtype=object)
+        result = pd.unique(arg)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_obj_none_preservation(self):
+        # GH 20866
+        arr = np.array(["foo", None], dtype=object)
+        result = pd.unique(arr)
+        expected = np.array(["foo", None], dtype=object)
+
+        tm.assert_numpy_array_equal(result, expected, strict_nan=True)
+
+    def test_signed_zero(self):
+        # GH 21866
+        a = np.array([-0.0, 0.0])
+        result = pd.unique(a)
+        expected = np.array([-0.0])  # 0.0 and -0.0 are equivalent
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_different_nans(self):
+        # GH 21866
+        # create different nans from bit-patterns:
+        NAN1 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000000))[0]
+        NAN2 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000001))[0]
+        assert NAN1 != NAN1
+        assert NAN2 != NAN2
+        a = np.array([NAN1, NAN2])  # NAN1 and NAN2 are equivalent
+        result = pd.unique(a)
+        expected = np.array([np.nan])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("el_type", [np.float64, object])
+    def test_first_nan_kept(self, el_type):
+        # GH 22295
+        # create different nans from bit-patterns:
+        bits_for_nan1 = 0xFFF8000000000001
+        bits_for_nan2 = 0x7FF8000000000001
+        NAN1 = struct.unpack("d", struct.pack("=Q", bits_for_nan1))[0]
+        NAN2 = struct.unpack("d", struct.pack("=Q", bits_for_nan2))[0]
+        assert NAN1 != NAN1
+        assert NAN2 != NAN2
+        a = np.array([NAN1, NAN2], dtype=el_type)
+        result = pd.unique(a)
+        assert result.size == 1
+        # use bit patterns to identify which nan was kept:
+        result_nan_bits = struct.unpack("=Q", struct.pack("d", result[0]))[0]
+        assert result_nan_bits == bits_for_nan1
+
+    def test_do_not_mangle_na_values(self, unique_nulls_fixture, unique_nulls_fixture2):
+        # GH 22295
+        if unique_nulls_fixture is unique_nulls_fixture2:
+            return  # skip it, values not unique
+        a = np.array([unique_nulls_fixture, unique_nulls_fixture2], dtype=object)
+        result = pd.unique(a)
+        assert result.size == 2
+        assert a[0] is unique_nulls_fixture
+        assert a[1] is unique_nulls_fixture2
+
+    def test_unique_masked(self, any_numeric_ea_dtype):
+        # GH#48019
+        ser = Series([1, pd.NA, 2] * 3, dtype=any_numeric_ea_dtype)
+        result = pd.unique(ser)
+        expected = pd.array([1, pd.NA, 2], dtype=any_numeric_ea_dtype)
+        tm.assert_extension_array_equal(result, expected)
+
+    def test_unique_NumpyExtensionArray(self):
+        arr_complex = pd.array(
+            [1 + 1j, 2, 3]
+        )  # NumpyEADtype('complex128') => NumpyExtensionArray
+        result = pd.unique(arr_complex)
+        expected = pd.array([1 + 1j, 2 + 0j, 3 + 0j])
+        tm.assert_extension_array_equal(result, expected)
+
+
+def test_nunique_ints(index_or_series_or_array):
+    # GH#36327
+    values = index_or_series_or_array(np.random.default_rng(2).integers(0, 20, 30))
+    result = algos.nunique_ints(values)
+    expected = len(algos.unique(values))
+    assert result == expected
+
+
+class TestIsin:
+    def test_invalid(self):
+        msg = (
+            r"only list-like objects are allowed to be passed to isin\(\), "
+            r"you passed a `int`"
+        )
+        with pytest.raises(TypeError, match=msg):
+            algos.isin(1, 1)
+        with pytest.raises(TypeError, match=msg):
+            algos.isin(1, [1])
+        with pytest.raises(TypeError, match=msg):
+            algos.isin([1], 1)
+
+    def test_basic(self):
+        result = algos.isin(np.array([1, 2]), [1])
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(Series([1, 2]), [1])
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(Series([1, 2]), Series([1]))
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(Series([1, 2]), {1})
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        arg = np.array(["a", "b"], dtype=object)
+        result = algos.isin(arg, ["a"])
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(Series(arg), Series(["a"]))
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(Series(arg), {"a"})
+        expected = np.array([True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(arg, [1])
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_i8(self):
+        arr = date_range("20130101", periods=3).values
+        result = algos.isin(arr, [arr[0]])
+        expected = np.array([True, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(arr, arr[0:2])
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(arr, set(arr[0:2]))
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        arr = timedelta_range("1 day", periods=3).values
+        result = algos.isin(arr, [arr[0]])
+        expected = np.array([True, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(arr, arr[0:2])
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.isin(arr, set(arr[0:2]))
+        expected = np.array([True, True, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype1", ["m8[ns]", "M8[ns]", "M8[ns, UTC]", "period[D]"])
+    @pytest.mark.parametrize("dtype", ["i8", "f8", "u8"])
+    def test_isin_datetimelike_values_numeric_comps(self, dtype, dtype1):
+        # Anything but object and we get all-False shortcut
+
+        dta = date_range("2013-01-01", periods=3)._values
+        arr = Series(dta.view("i8")).array.view(dtype1)
+
+        comps = arr.view("i8").astype(dtype)
+
+        result = algos.isin(comps, arr)
+        expected = np.zeros(comps.shape, dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_large(self):
+        s = date_range("20000101", periods=2000000, freq="s").values
+        result = algos.isin(s, s[0:2])
+        expected = np.zeros(len(s), dtype=bool)
+        expected[0] = True
+        expected[1] = True
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["m8[ns]", "M8[ns]", "M8[ns, UTC]", "period[D]"])
+    def test_isin_datetimelike_all_nat(self, dtype):
+        # GH#56427
+        dta = date_range("2013-01-01", periods=3)._values
+        arr = Series(dta.view("i8")).array.view(dtype)
+
+        arr[0] = NaT
+        result = algos.isin(arr, [NaT])
+        expected = np.array([True, False, False], dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", ["m8[ns]", "M8[ns]", "M8[ns, UTC]"])
+    def test_isin_datetimelike_strings_returns_false(self, dtype):
+        # GH#53111
+        dta = date_range("2013-01-01", periods=3)._values
+        arr = Series(dta.view("i8")).array.view(dtype)
+
+        vals = [str(x) for x in arr]
+        res = algos.isin(arr, vals)
+        assert not res.any()
+
+        vals2 = np.array(vals, dtype=str)
+        res2 = algos.isin(arr, vals2)
+        assert not res2.any()
+
+    def test_isin_dt64tz_with_nat(self):
+        # the all-NaT values used to get inferred to tznaive, which was evaluated
+        #  as non-matching GH#56427
+        dti = date_range("2016-01-01", periods=3, tz="UTC")
+        ser = Series(dti)
+        ser[0] = NaT
+
+        res = algos.isin(ser._values, [NaT])
+        exp = np.array([True, False, False], dtype=bool)
+        tm.assert_numpy_array_equal(res, exp)
+
+    def test_categorical_from_codes(self):
+        # GH 16639
+        vals = np.array([0, 1, 2, 0])
+        cats = ["a", "b", "c"]
+        Sd = Series(Categorical([1]).from_codes(vals, cats))
+        St = Series(Categorical([1]).from_codes(np.array([0, 1]), cats))
+        expected = np.array([True, True, False, True])
+        result = algos.isin(Sd, St)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_categorical_isin(self):
+        vals = np.array([0, 1, 2, 0])
+        cats = ["a", "b", "c"]
+        cat = Categorical([1]).from_codes(vals, cats)
+        other = Categorical([1]).from_codes(np.array([0, 1]), cats)
+
+        expected = np.array([True, True, False, True])
+        result = algos.isin(cat, other)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_same_nan_is_in(self):
+        # GH 22160
+        # nan is special, because from " a is b" doesn't follow "a == b"
+        # at least, isin() should follow python's "np.nan in [nan] == True"
+        # casting to -> np.float64 -> another float-object somewhere on
+        # the way could lead jeopardize this behavior
+        comps = np.array([np.nan], dtype=object)  # could be casted to float64
+        values = [np.nan]
+        expected = np.array([True])
+        result = algos.isin(comps, values)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_same_nan_is_in_large(self):
+        # https://github.com/pandas-dev/pandas/issues/22205
+        s = np.tile(1.0, 1_000_001)
+        s[0] = np.nan
+        result = algos.isin(s, np.array([np.nan, 1]))
+        expected = np.ones(len(s), dtype=bool)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_same_nan_is_in_large_series(self):
+        # https://github.com/pandas-dev/pandas/issues/22205
+        s = np.tile(1.0, 1_000_001)
+        series = Series(s)
+        s[0] = np.nan
+        result = series.isin(np.array([np.nan, 1]))
+        expected = Series(np.ones(len(s), dtype=bool))
+        tm.assert_series_equal(result, expected)
+
+    def test_same_object_is_in(self):
+        # GH 22160
+        # there could be special treatment for nans
+        # the user however could define a custom class
+        # with similar behavior, then we at least should
+        # fall back to usual python's behavior: "a in [a] == True"
+        class LikeNan:
+            def __eq__(self, other) -> bool:
+                return False
+
+            def __hash__(self):
+                return 0
+
+        a, b = LikeNan(), LikeNan()
+
+        arg = np.array([a], dtype=object)
+
+        # same object -> True
+        tm.assert_numpy_array_equal(algos.isin(arg, [a]), np.array([True]))
+        # different objects -> False
+        tm.assert_numpy_array_equal(algos.isin(arg, [b]), np.array([False]))
+
+    def test_different_nans(self):
+        # GH 22160
+        # all nans are handled as equivalent
+
+        comps = [float("nan")]
+        values = [float("nan")]
+        assert comps[0] is not values[0]  # different nan-objects
+
+        # as list of python-objects:
+        result = algos.isin(np.array(comps), values)
+        tm.assert_numpy_array_equal(np.array([True]), result)
+
+        # as object-array:
+        result = algos.isin(
+            np.asarray(comps, dtype=object), np.asarray(values, dtype=object)
+        )
+        tm.assert_numpy_array_equal(np.array([True]), result)
+
+        # as float64-array:
+        result = algos.isin(
+            np.asarray(comps, dtype=np.float64), np.asarray(values, dtype=np.float64)
+        )
+        tm.assert_numpy_array_equal(np.array([True]), result)
+
+    def test_no_cast(self):
+        # GH 22160
+        # ensure 42 is not casted to a string
+        comps = np.array(["ss", 42], dtype=object)
+        values = ["42"]
+        expected = np.array([False, False])
+
+        result = algos.isin(comps, values)
+        tm.assert_numpy_array_equal(expected, result)
+
+    @pytest.mark.parametrize("empty", [[], Series(dtype=object), np.array([])])
+    def test_empty(self, empty):
+        # see gh-16991
+        vals = Index(["a", "b"])
+        expected = np.array([False, False])
+
+        result = algos.isin(vals, empty)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_different_nan_objects(self):
+        # GH 22119
+        comps = np.array(["nan", np.nan * 1j, float("nan")], dtype=object)
+        vals = np.array([float("nan")], dtype=object)
+        expected = np.array([False, False, True])
+        result = algos.isin(comps, vals)
+        tm.assert_numpy_array_equal(expected, result)
+
+    def test_different_nans_as_float64(self):
+        # GH 21866
+        # create different nans from bit-patterns,
+        # these nans will land in different buckets in the hash-table
+        # if no special care is taken
+        NAN1 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000000))[0]
+        NAN2 = struct.unpack("d", struct.pack("=Q", 0x7FF8000000000001))[0]
+        assert NAN1 != NAN1
+        assert NAN2 != NAN2
+
+        # check that NAN1 and NAN2 are equivalent:
+        arr = np.array([NAN1, NAN2], dtype=np.float64)
+        lookup1 = np.array([NAN1], dtype=np.float64)
+        result = algos.isin(arr, lookup1)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        lookup2 = np.array([NAN2], dtype=np.float64)
+        result = algos.isin(arr, lookup2)
+        expected = np.array([True, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_isin_int_df_string_search(self):
+        """Comparing df with int`s (1,2) with a string at isin() ("1")
+        -> should not match values because int 1 is not equal str 1"""
+        df = DataFrame({"values": [1, 2]})
+        result = df.isin(["1"])
+        expected_false = DataFrame({"values": [False, False]})
+        tm.assert_frame_equal(result, expected_false)
+
+    def test_isin_nan_df_string_search(self):
+        """Comparing df with nan value (np.nan,2) with a string at isin() ("NaN")
+        -> should not match values because np.nan is not equal str NaN"""
+        df = DataFrame({"values": [np.nan, 2]})
+        result = df.isin(np.array(["NaN"], dtype=object))
+        expected_false = DataFrame({"values": [False, False]})
+        tm.assert_frame_equal(result, expected_false)
+
+    def test_isin_float_df_string_search(self):
+        """Comparing df with floats (1.4245,2.32441) with a string at isin() ("1.4245")
+        -> should not match values because float 1.4245 is not equal str 1.4245"""
+        df = DataFrame({"values": [1.4245, 2.32441]})
+        result = df.isin(np.array(["1.4245"], dtype=object))
+        expected_false = DataFrame({"values": [False, False]})
+        tm.assert_frame_equal(result, expected_false)
+
+    def test_isin_unsigned_dtype(self):
+        # GH#46485
+        ser = Series([1378774140726870442], dtype=np.uint64)
+        result = ser.isin([1378774140726870528])
+        expected = Series(False)
+        tm.assert_series_equal(result, expected)
+
+
+class TestValueCounts:
+    def test_value_counts(self):
+        arr = np.random.default_rng(1234).standard_normal(4)
+        factor = cut(arr, 4)
+
+        # assert isinstance(factor, n)
+        result = algos.value_counts_internal(factor)
+        breaks = [-1.606, -1.018, -0.431, 0.155, 0.741]
+        index = IntervalIndex.from_breaks(breaks).astype(CategoricalDtype(ordered=True))
+        expected = Series([1, 0, 2, 1], index=index, name="count")
+        tm.assert_series_equal(result.sort_index(), expected.sort_index())
+
+    def test_value_counts_bins(self):
+        s = [1, 2, 3, 4]
+        result = algos.value_counts_internal(s, bins=1)
+        expected = Series(
+            [4], index=IntervalIndex.from_tuples([(0.996, 4.0)]), name="count"
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = algos.value_counts_internal(s, bins=2, sort=False)
+        expected = Series(
+            [2, 2],
+            index=IntervalIndex.from_tuples([(0.996, 2.5), (2.5, 4.0)]),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_value_counts_dtypes(self):
+        result = algos.value_counts_internal(np.array([1, 1.0]))
+        assert len(result) == 1
+
+        result = algos.value_counts_internal(np.array([1, 1.0]), bins=1)
+        assert len(result) == 1
+
+        result = algos.value_counts_internal(Series([1, 1.0, "1"]))  # object
+        assert len(result) == 2
+
+        msg = "bins argument only works with numeric data"
+        with pytest.raises(TypeError, match=msg):
+            algos.value_counts_internal(np.array(["1", 1], dtype=object), bins=1)
+
+    def test_value_counts_nat(self):
+        td = Series([np.timedelta64(10000), NaT], dtype="timedelta64[ns]")
+        dt = to_datetime(["NaT", "2014-01-01"])
+
+        for ser in [td, dt]:
+            vc = algos.value_counts_internal(ser)
+            vc_with_na = algos.value_counts_internal(ser, dropna=False)
+            assert len(vc) == 1
+            assert len(vc_with_na) == 2
+
+        exp_dt = Series({Timestamp("2014-01-01 00:00:00"): 1}, name="count")
+        result_dt = algos.value_counts_internal(dt)
+        tm.assert_series_equal(result_dt, exp_dt)
+
+        exp_td = Series([1], index=[np.timedelta64(10000)], name="count")
+        result_td = algos.value_counts_internal(td)
+        tm.assert_series_equal(result_td, exp_td)
+
+    @pytest.mark.parametrize("dtype", [object, "M8[us]"])
+    def test_value_counts_datetime_outofbounds(self, dtype):
+        # GH 13663
+        ser = Series(
+            [
+                datetime(3000, 1, 1),
+                datetime(5000, 1, 1),
+                datetime(5000, 1, 1),
+                datetime(6000, 1, 1),
+                datetime(3000, 1, 1),
+                datetime(3000, 1, 1),
+            ],
+            dtype=dtype,
+        )
+
+        res = ser.value_counts()
+
+        exp_index = Index(
+            [datetime(3000, 1, 1), datetime(5000, 1, 1), datetime(6000, 1, 1)],
+            dtype=dtype,
+        )
+        exp = Series([3, 2, 1], index=exp_index, name="count")
+        tm.assert_series_equal(res, exp)
+
+    def test_categorical(self):
+        s = Series(Categorical(list("aaabbc")))
+        result = s.value_counts()
+        expected = Series(
+            [3, 2, 1], index=CategoricalIndex(["a", "b", "c"]), name="count"
+        )
+
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        # preserve order?
+        s = s.cat.as_ordered()
+        result = s.value_counts()
+        expected.index = expected.index.as_ordered()
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_categorical_nans(self):
+        s = Series(Categorical(list("aaaaabbbcc")))  # 4,3,2,1 (nan)
+        s.iloc[1] = np.nan
+        result = s.value_counts()
+        expected = Series(
+            [4, 3, 2],
+            index=CategoricalIndex(["a", "b", "c"], categories=["a", "b", "c"]),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected, check_index_type=True)
+        result = s.value_counts(dropna=False)
+        expected = Series(
+            [4, 3, 2, 1], index=CategoricalIndex(["a", "b", "c", np.nan]), name="count"
+        )
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        # out of order
+        s = Series(
+            Categorical(list("aaaaabbbcc"), ordered=True, categories=["b", "a", "c"])
+        )
+        s.iloc[1] = np.nan
+        result = s.value_counts()
+        expected = Series(
+            [4, 3, 2],
+            index=CategoricalIndex(
+                ["a", "b", "c"],
+                categories=["b", "a", "c"],
+                ordered=True,
+            ),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+        result = s.value_counts(dropna=False)
+        expected = Series(
+            [4, 3, 2, 1],
+            index=CategoricalIndex(
+                ["a", "b", "c", np.nan], categories=["b", "a", "c"], ordered=True
+            ),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_categorical_zeroes(self):
+        # keep the `d` category with 0
+        s = Series(Categorical(list("bbbaac"), categories=list("abcd"), ordered=True))
+        result = s.value_counts()
+        expected = Series(
+            [3, 2, 1, 0],
+            index=Categorical(
+                ["b", "a", "c", "d"], categories=list("abcd"), ordered=True
+            ),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected, check_index_type=True)
+
+    def test_value_counts_dropna(self):
+        # https://github.com/pandas-dev/pandas/issues/9443#issuecomment-73719328
+
+        tm.assert_series_equal(
+            Series([True, True, False]).value_counts(dropna=True),
+            Series([2, 1], index=[True, False], name="count"),
+        )
+        tm.assert_series_equal(
+            Series([True, True, False]).value_counts(dropna=False),
+            Series([2, 1], index=[True, False], name="count"),
+        )
+
+        tm.assert_series_equal(
+            Series([True] * 3 + [False] * 2 + [None] * 5).value_counts(dropna=True),
+            Series([3, 2], index=Index([True, False], dtype=object), name="count"),
+        )
+        tm.assert_series_equal(
+            Series([True] * 5 + [False] * 3 + [None] * 2).value_counts(dropna=False),
+            Series([5, 3, 2], index=[True, False, None], name="count"),
+        )
+        tm.assert_series_equal(
+            Series([10.3, 5.0, 5.0]).value_counts(dropna=True),
+            Series([2, 1], index=[5.0, 10.3], name="count"),
+        )
+        tm.assert_series_equal(
+            Series([10.3, 5.0, 5.0]).value_counts(dropna=False),
+            Series([2, 1], index=[5.0, 10.3], name="count"),
+        )
+
+        tm.assert_series_equal(
+            Series([10.3, 5.0, 5.0, None]).value_counts(dropna=True),
+            Series([2, 1], index=[5.0, 10.3], name="count"),
+        )
+
+        result = Series([10.3, 10.3, 5.0, 5.0, 5.0, None]).value_counts(dropna=False)
+        expected = Series([3, 2, 1], index=[5.0, 10.3, None], name="count")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("dtype", (np.float64, object, "M8[ns]"))
+    def test_value_counts_normalized(self, dtype):
+        # GH12558
+        s = Series([1] * 2 + [2] * 3 + [np.nan] * 5)
+        s_typed = s.astype(dtype)
+        result = s_typed.value_counts(normalize=True, dropna=False)
+        expected = Series(
+            [0.5, 0.3, 0.2],
+            index=Series([np.nan, 2.0, 1.0], dtype=dtype),
+            name="proportion",
+        )
+        tm.assert_series_equal(result, expected)
+
+        result = s_typed.value_counts(normalize=True, dropna=True)
+        expected = Series(
+            [0.6, 0.4], index=Series([2.0, 1.0], dtype=dtype), name="proportion"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_value_counts_uint64(self):
+        arr = np.array([2**63], dtype=np.uint64)
+        expected = Series([1], index=[2**63], name="count")
+        result = algos.value_counts_internal(arr)
+
+        tm.assert_series_equal(result, expected)
+
+        arr = np.array([-1, 2**63], dtype=object)
+        expected = Series([1, 1], index=[-1, 2**63], name="count")
+        result = algos.value_counts_internal(arr)
+
+        tm.assert_series_equal(result, expected)
+
+    def test_value_counts_series(self):
+        # GH#54857
+        values = np.array([3, 1, 2, 3, 4, np.nan])
+        result = Series(values).value_counts(bins=3)
+        expected = Series(
+            [2, 2, 1],
+            index=IntervalIndex.from_tuples(
+                [(0.996, 2.0), (2.0, 3.0), (3.0, 4.0)], dtype="interval[float64, right]"
+            ),
+            name="count",
+        )
+        tm.assert_series_equal(result, expected)
+
+
+class TestDuplicated:
+    def test_duplicated_with_nas(self):
+        keys = np.array([0, 1, np.nan, 0, 2, np.nan], dtype=object)
+
+        result = algos.duplicated(keys)
+        expected = np.array([False, False, False, True, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.duplicated(keys, keep="first")
+        expected = np.array([False, False, False, True, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.duplicated(keys, keep="last")
+        expected = np.array([True, False, True, False, False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.duplicated(keys, keep=False)
+        expected = np.array([True, False, True, True, False, True])
+        tm.assert_numpy_array_equal(result, expected)
+
+        keys = np.empty(8, dtype=object)
+        for i, t in enumerate(
+            zip([0, 0, np.nan, np.nan] * 2, [0, np.nan, 0, np.nan] * 2)
+        ):
+            keys[i] = t
+
+        result = algos.duplicated(keys)
+        falses = [False] * 4
+        trues = [True] * 4
+        expected = np.array(falses + trues)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.duplicated(keys, keep="last")
+        expected = np.array(trues + falses)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.duplicated(keys, keep=False)
+        expected = np.array(trues + trues)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            np.array([1, 2, 1, 5, 3, 2, 4, 1, 5, 6]),
+            np.array([1.1, 2.2, 1.1, np.nan, 3.3, 2.2, 4.4, 1.1, np.nan, 6.6]),
+            np.array(
+                [
+                    1 + 1j,
+                    2 + 2j,
+                    1 + 1j,
+                    5 + 5j,
+                    3 + 3j,
+                    2 + 2j,
+                    4 + 4j,
+                    1 + 1j,
+                    5 + 5j,
+                    6 + 6j,
+                ]
+            ),
+            np.array(["a", "b", "a", "e", "c", "b", "d", "a", "e", "f"], dtype=object),
+            np.array([1, 2**63, 1, 3**5, 10, 2**63, 39, 1, 3**5, 7], dtype=np.uint64),
+        ],
+    )
+    def test_numeric_object_likes(self, case):
+        exp_first = np.array(
+            [False, False, True, False, False, True, False, True, True, False]
+        )
+        exp_last = np.array(
+            [True, True, True, True, False, False, False, False, False, False]
+        )
+        exp_false = exp_first | exp_last
+
+        res_first = algos.duplicated(case, keep="first")
+        tm.assert_numpy_array_equal(res_first, exp_first)
+
+        res_last = algos.duplicated(case, keep="last")
+        tm.assert_numpy_array_equal(res_last, exp_last)
+
+        res_false = algos.duplicated(case, keep=False)
+        tm.assert_numpy_array_equal(res_false, exp_false)
+
+        # index
+        for idx in [Index(case), Index(case, dtype="category")]:
+            res_first = idx.duplicated(keep="first")
+            tm.assert_numpy_array_equal(res_first, exp_first)
+
+            res_last = idx.duplicated(keep="last")
+            tm.assert_numpy_array_equal(res_last, exp_last)
+
+            res_false = idx.duplicated(keep=False)
+            tm.assert_numpy_array_equal(res_false, exp_false)
+
+        # series
+        for s in [Series(case), Series(case, dtype="category")]:
+            res_first = s.duplicated(keep="first")
+            tm.assert_series_equal(res_first, Series(exp_first))
+
+            res_last = s.duplicated(keep="last")
+            tm.assert_series_equal(res_last, Series(exp_last))
+
+            res_false = s.duplicated(keep=False)
+            tm.assert_series_equal(res_false, Series(exp_false))
+
+    def test_datetime_likes(self):
+        dt = [
+            "2011-01-01",
+            "2011-01-02",
+            "2011-01-01",
+            "NaT",
+            "2011-01-03",
+            "2011-01-02",
+            "2011-01-04",
+            "2011-01-01",
+            "NaT",
+            "2011-01-06",
+        ]
+        td = [
+            "1 days",
+            "2 days",
+            "1 days",
+            "NaT",
+            "3 days",
+            "2 days",
+            "4 days",
+            "1 days",
+            "NaT",
+            "6 days",
+        ]
+
+        cases = [
+            np.array([Timestamp(d) for d in dt]),
+            np.array([Timestamp(d, tz="US/Eastern") for d in dt]),
+            np.array([Period(d, freq="D") for d in dt]),
+            np.array([np.datetime64(d) for d in dt]),
+            np.array([Timedelta(d) for d in td]),
+        ]
+
+        exp_first = np.array(
+            [False, False, True, False, False, True, False, True, True, False]
+        )
+        exp_last = np.array(
+            [True, True, True, True, False, False, False, False, False, False]
+        )
+        exp_false = exp_first | exp_last
+
+        for case in cases:
+            res_first = algos.duplicated(case, keep="first")
+            tm.assert_numpy_array_equal(res_first, exp_first)
+
+            res_last = algos.duplicated(case, keep="last")
+            tm.assert_numpy_array_equal(res_last, exp_last)
+
+            res_false = algos.duplicated(case, keep=False)
+            tm.assert_numpy_array_equal(res_false, exp_false)
+
+            # index
+            for idx in [
+                Index(case),
+                Index(case, dtype="category"),
+                Index(case, dtype=object),
+            ]:
+                res_first = idx.duplicated(keep="first")
+                tm.assert_numpy_array_equal(res_first, exp_first)
+
+                res_last = idx.duplicated(keep="last")
+                tm.assert_numpy_array_equal(res_last, exp_last)
+
+                res_false = idx.duplicated(keep=False)
+                tm.assert_numpy_array_equal(res_false, exp_false)
+
+            # series
+            for s in [
+                Series(case),
+                Series(case, dtype="category"),
+                Series(case, dtype=object),
+            ]:
+                res_first = s.duplicated(keep="first")
+                tm.assert_series_equal(res_first, Series(exp_first))
+
+                res_last = s.duplicated(keep="last")
+                tm.assert_series_equal(res_last, Series(exp_last))
+
+                res_false = s.duplicated(keep=False)
+                tm.assert_series_equal(res_false, Series(exp_false))
+
+    @pytest.mark.parametrize("case", [Index([1, 2, 3]), pd.RangeIndex(0, 3)])
+    def test_unique_index(self, case):
+        assert case.is_unique is True
+        tm.assert_numpy_array_equal(case.duplicated(), np.array([False, False, False]))
+
+    @pytest.mark.parametrize(
+        "arr, uniques",
+        [
+            (
+                [(0, 0), (0, 1), (1, 0), (1, 1), (0, 0), (0, 1), (1, 0), (1, 1)],
+                [(0, 0), (0, 1), (1, 0), (1, 1)],
+            ),
+            (
+                [("b", "c"), ("a", "b"), ("a", "b"), ("b", "c")],
+                [("b", "c"), ("a", "b")],
+            ),
+            ([("a", 1), ("b", 2), ("a", 3), ("a", 1)], [("a", 1), ("b", 2), ("a", 3)]),
+        ],
+    )
+    def test_unique_tuples(self, arr, uniques):
+        # https://github.com/pandas-dev/pandas/issues/16519
+        expected = np.empty(len(uniques), dtype=object)
+        expected[:] = uniques
+
+        msg = (
+            r"unique requires a Series, Index, ExtensionArray, np.ndarray "
+            r"or NumpyExtensionArray got list"
+        )
+        with pytest.raises(TypeError, match=msg):
+            # GH#52986
+            pd.unique(arr)
+
+        res = pd.unique(com.asarray_tuplesafe(arr, dtype=object))
+        tm.assert_numpy_array_equal(res, expected)
+
+    @pytest.mark.parametrize(
+        "array,expected",
+        [
+            (
+                [1 + 1j, 0, 1, 1j, 1 + 2j, 1 + 2j],
+                np.array([(1 + 1j), 0j, (1 + 0j), 1j, (1 + 2j)], dtype=complex),
+            )
+        ],
+    )
+    def test_unique_complex_numbers(self, array, expected):
+        # GH 17927
+        msg = (
+            r"unique requires a Series, Index, ExtensionArray, np.ndarray "
+            r"or NumpyExtensionArray got list"
+        )
+
+        with pytest.raises(TypeError, match=msg):
+            # GH#52986
+            pd.unique(array)
+
+        res = pd.unique(np.array(array))
+        tm.assert_numpy_array_equal(res, expected)
+
+
+class TestHashTable:
+    @pytest.mark.parametrize(
+        "htable, data",
+        [
+            (
+                ht.PyObjectHashTable,
+                np.array([f"foo_{i}" for i in range(1000)], dtype=object),
+            ),
+            (
+                ht.StringHashTable,
+                np.array([f"foo_{i}" for i in range(1000)], dtype=object),
+            ),
+            (ht.Float64HashTable, np.arange(1000, dtype=np.float64)),
+            (ht.Int64HashTable, np.arange(1000, dtype=np.int64)),
+            (ht.UInt64HashTable, np.arange(1000, dtype=np.uint64)),
+        ],
+    )
+    def test_hashtable_unique(self, htable, data, writable):
+        # output of maker has guaranteed unique elements
+        s = Series(data, dtype=data.dtype)
+        if htable == ht.Float64HashTable:
+            # add NaN for float column
+            s.loc[500] = np.nan
+        elif htable == ht.PyObjectHashTable:
+            # use different NaN types for object column
+            s.loc[500:502] = [np.nan, None, NaT]
+
+        # create duplicated selection
+        s_duplicated = s.sample(frac=3, replace=True).reset_index(drop=True)
+        s_duplicated.values.setflags(write=writable)
+
+        # drop_duplicates has own cython code (hash_table_func_helper.pxi)
+        # and is tested separately; keeps first occurrence like ht.unique()
+        expected_unique = s_duplicated.drop_duplicates(keep="first").values
+        result_unique = htable().unique(s_duplicated.values)
+        tm.assert_numpy_array_equal(result_unique, expected_unique)
+
+        # test return_inverse=True
+        # reconstruction can only succeed if the inverse is correct
+        result_unique, result_inverse = htable().unique(
+            s_duplicated.values, return_inverse=True
+        )
+        tm.assert_numpy_array_equal(result_unique, expected_unique)
+        reconstr = result_unique[result_inverse]
+        tm.assert_numpy_array_equal(reconstr, s_duplicated.values)
+
+    @pytest.mark.parametrize(
+        "htable, data",
+        [
+            (
+                ht.PyObjectHashTable,
+                np.array([f"foo_{i}" for i in range(1000)], dtype=object),
+            ),
+            (
+                ht.StringHashTable,
+                np.array([f"foo_{i}" for i in range(1000)], dtype=object),
+            ),
+            (ht.Float64HashTable, np.arange(1000, dtype=np.float64)),
+            (ht.Int64HashTable, np.arange(1000, dtype=np.int64)),
+            (ht.UInt64HashTable, np.arange(1000, dtype=np.uint64)),
+        ],
+    )
+    def test_hashtable_factorize(self, htable, writable, data):
+        # output of maker has guaranteed unique elements
+        s = Series(data, dtype=data.dtype)
+        if htable == ht.Float64HashTable:
+            # add NaN for float column
+            s.loc[500] = np.nan
+        elif htable == ht.PyObjectHashTable:
+            # use different NaN types for object column
+            s.loc[500:502] = [np.nan, None, NaT]
+
+        # create duplicated selection
+        s_duplicated = s.sample(frac=3, replace=True).reset_index(drop=True)
+        s_duplicated.values.setflags(write=writable)
+        na_mask = s_duplicated.isna().values
+
+        result_unique, result_inverse = htable().factorize(s_duplicated.values)
+
+        # drop_duplicates has own cython code (hash_table_func_helper.pxi)
+        # and is tested separately; keeps first occurrence like ht.factorize()
+        # since factorize removes all NaNs, we do the same here
+        expected_unique = s_duplicated.dropna().drop_duplicates().values
+        tm.assert_numpy_array_equal(result_unique, expected_unique)
+
+        # reconstruction can only succeed if the inverse is correct. Since
+        # factorize removes the NaNs, those have to be excluded here as well
+        result_reconstruct = result_unique[result_inverse[~na_mask]]
+        expected_reconstruct = s_duplicated.dropna().values
+        tm.assert_numpy_array_equal(result_reconstruct, expected_reconstruct)
+
+
+class TestRank:
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            [np.nan, np.nan, 5.0, 5.0, 5.0, np.nan, 1, 2, 3, np.nan],
+            [4.0, np.nan, 5.0, 5.0, 5.0, np.nan, 1, 2, 4.0, np.nan],
+        ],
+    )
+    def test_scipy_compat(self, arr):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        arr = np.array(arr)
+
+        mask = ~np.isfinite(arr)
+        result = libalgos.rank_1d(arr)
+        arr[mask] = np.inf
+        exp = sp_stats.rankdata(arr)
+        exp[mask] = np.nan
+        tm.assert_almost_equal(result, exp)
+
+    def test_basic(self, writable, any_int_numpy_dtype):
+        exp = np.array([1, 2], dtype=np.float64)
+
+        data = np.array([1, 100], dtype=any_int_numpy_dtype)
+        data.setflags(write=writable)
+        ser = Series(data)
+        result = algos.rank(ser)
+        tm.assert_numpy_array_equal(result, exp)
+
+    @pytest.mark.parametrize("dtype", [np.float64, np.uint64])
+    def test_uint64_overflow(self, dtype):
+        exp = np.array([1, 2], dtype=np.float64)
+
+        s = Series([1, 2**63], dtype=dtype)
+        tm.assert_numpy_array_equal(algos.rank(s), exp)
+
+    def test_too_many_ndims(self):
+        arr = np.array([[[1, 2, 3], [4, 5, 6], [7, 8, 9]]])
+        msg = "Array with ndim > 2 are not supported"
+
+        with pytest.raises(TypeError, match=msg):
+            algos.rank(arr)
+
+    @pytest.mark.single_cpu
+    def test_pct_max_many_rows(self):
+        # GH 18271
+        values = np.arange(2**24 + 1)
+        result = algos.rank(values, pct=True).max()
+        assert result == 1
+
+        values = np.arange(2**25 + 2).reshape(2**24 + 1, 2)
+        result = algos.rank(values, pct=True).max()
+        assert result == 1
+
+
+class TestMode:
+    def test_no_mode(self):
+        exp = Series([], dtype=np.float64, index=Index([], dtype=int))
+        tm.assert_numpy_array_equal(algos.mode(np.array([])), exp.values)
+
+    def test_mode_single(self, any_real_numpy_dtype):
+        # GH 15714
+        exp_single = [1]
+        data_single = [1]
+
+        exp_multi = [1]
+        data_multi = [1, 1]
+
+        ser = Series(data_single, dtype=any_real_numpy_dtype)
+        exp = Series(exp_single, dtype=any_real_numpy_dtype)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+        ser = Series(data_multi, dtype=any_real_numpy_dtype)
+        exp = Series(exp_multi, dtype=any_real_numpy_dtype)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_mode_obj_int(self):
+        exp = Series([1], dtype=int)
+        tm.assert_numpy_array_equal(algos.mode(exp.values), exp.values)
+
+        exp = Series(["a", "b", "c"], dtype=object)
+        tm.assert_numpy_array_equal(algos.mode(exp.values), exp.values)
+
+    def test_number_mode(self, any_real_numpy_dtype):
+        exp_single = [1]
+        data_single = [1] * 5 + [2] * 3
+
+        exp_multi = [1, 3]
+        data_multi = [1] * 5 + [2] * 3 + [3] * 5
+
+        ser = Series(data_single, dtype=any_real_numpy_dtype)
+        exp = Series(exp_single, dtype=any_real_numpy_dtype)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+        ser = Series(data_multi, dtype=any_real_numpy_dtype)
+        exp = Series(exp_multi, dtype=any_real_numpy_dtype)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_strobj_mode(self):
+        exp = ["b"]
+        data = ["a"] * 2 + ["b"] * 3
+
+        ser = Series(data, dtype="c")
+        exp = Series(exp, dtype="c")
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    @pytest.mark.parametrize("dt", [str, object])
+    def test_strobj_multi_char(self, dt, using_infer_string):
+        exp = ["bar"]
+        data = ["foo"] * 2 + ["bar"] * 3
+
+        ser = Series(data, dtype=dt)
+        exp = Series(exp, dtype=dt)
+        if using_infer_string and dt is str:
+            tm.assert_extension_array_equal(algos.mode(ser.values), exp.values)
+        else:
+            tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_datelike_mode(self):
+        exp = Series(["1900-05-03", "2011-01-03", "2013-01-02"], dtype="M8[ns]")
+        ser = Series(["2011-01-03", "2013-01-02", "1900-05-03"], dtype="M8[ns]")
+        tm.assert_extension_array_equal(algos.mode(ser.values), exp._values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+        exp = Series(["2011-01-03", "2013-01-02"], dtype="M8[ns]")
+        ser = Series(
+            ["2011-01-03", "2013-01-02", "1900-05-03", "2011-01-03", "2013-01-02"],
+            dtype="M8[ns]",
+        )
+        tm.assert_extension_array_equal(algos.mode(ser.values), exp._values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_timedelta_mode(self):
+        exp = Series(["-1 days", "0 days", "1 days"], dtype="timedelta64[ns]")
+        ser = Series(["1 days", "-1 days", "0 days"], dtype="timedelta64[ns]")
+        tm.assert_extension_array_equal(algos.mode(ser.values), exp._values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+        exp = Series(["2 min", "1 day"], dtype="timedelta64[ns]")
+        ser = Series(
+            ["1 day", "1 day", "-1 day", "-1 day 2 min", "2 min", "2 min"],
+            dtype="timedelta64[ns]",
+        )
+        tm.assert_extension_array_equal(algos.mode(ser.values), exp._values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_mixed_dtype(self):
+        exp = Series(["foo"], dtype=object)
+        ser = Series([1, "foo", "foo"])
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_uint64_overflow(self):
+        exp = Series([2**63], dtype=np.uint64)
+        ser = Series([1, 2**63, 2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+        exp = Series([1, 2**63], dtype=np.uint64)
+        ser = Series([1, 2**63], dtype=np.uint64)
+        tm.assert_numpy_array_equal(algos.mode(ser.values), exp.values)
+        tm.assert_series_equal(ser.mode(), exp)
+
+    def test_categorical(self):
+        c = Categorical([1, 2])
+        exp = c
+        res = Series(c).mode()._values
+        tm.assert_categorical_equal(res, exp)
+
+        c = Categorical([1, "a", "a"])
+        exp = Categorical(["a"], categories=[1, "a"])
+        res = Series(c).mode()._values
+        tm.assert_categorical_equal(res, exp)
+
+        c = Categorical([1, 1, 2, 3, 3])
+        exp = Categorical([1, 3], categories=[1, 2, 3])
+        res = Series(c).mode()._values
+        tm.assert_categorical_equal(res, exp)
+
+    def test_index(self):
+        idx = Index([1, 2, 3])
+        exp = Series([1, 2, 3], dtype=np.int64)
+        tm.assert_numpy_array_equal(algos.mode(idx), exp.values)
+
+        idx = Index([1, "a", "a"])
+        exp = Series(["a"], dtype=object)
+        tm.assert_numpy_array_equal(algos.mode(idx), exp.values)
+
+        idx = Index([1, 1, 2, 3, 3])
+        exp = Series([1, 3], dtype=np.int64)
+        tm.assert_numpy_array_equal(algos.mode(idx), exp.values)
+
+        idx = Index(
+            ["1 day", "1 day", "-1 day", "-1 day 2 min", "2 min", "2 min"],
+            dtype="timedelta64[ns]",
+        )
+        with pytest.raises(AttributeError, match="TimedeltaIndex"):
+            # algos.mode expects Arraylike, does *not* unwrap TimedeltaIndex
+            algos.mode(idx)
+
+    def test_ser_mode_with_name(self):
+        # GH 46737
+        ser = Series([1, 1, 3], name="foo")
+        result = ser.mode()
+        expected = Series([1], name="foo")
+        tm.assert_series_equal(result, expected)
+
+
+class TestDiff:
+    @pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+    def test_diff_datetimelike_nat(self, dtype):
+        # NaT - NaT is NaT, not 0
+        arr = np.arange(12).astype(np.int64).view(dtype).reshape(3, 4)
+        arr[:, 2] = arr.dtype.type("NaT", "ns")
+        result = algos.diff(arr, 1, axis=0)
+
+        expected = np.ones(arr.shape, dtype="timedelta64[ns]") * 4
+        expected[:, 2] = np.timedelta64("NaT", "ns")
+        expected[0, :] = np.timedelta64("NaT", "ns")
+
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.diff(arr.T, 1, axis=1)
+        tm.assert_numpy_array_equal(result, expected.T)
+
+    def test_diff_ea_axis(self):
+        dta = date_range("2016-01-01", periods=3, tz="US/Pacific")._data
+
+        msg = "cannot diff DatetimeArray on axis=1"
+        with pytest.raises(ValueError, match=msg):
+            algos.diff(dta, 1, axis=1)
+
+    @pytest.mark.parametrize("dtype", ["int8", "int16"])
+    def test_diff_low_precision_int(self, dtype):
+        arr = np.array([0, 1, 1, 0, 0], dtype=dtype)
+        result = algos.diff(arr, 1)
+        expected = np.array([np.nan, 1, 0, -1, 0], dtype="float32")
+        tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("op", [np.array, pd.array])
+def test_union_with_duplicates(op):
+    # GH#36289
+    lvals = op([3, 1, 3, 4])
+    rvals = op([2, 3, 1, 1])
+    expected = op([3, 3, 1, 1, 4, 2])
+    if isinstance(expected, np.ndarray):
+        result = algos.union_with_duplicates(lvals, rvals)
+        tm.assert_numpy_array_equal(result, expected)
+    else:
+        result = algos.union_with_duplicates(lvals, rvals)
+        tm.assert_extension_array_equal(result, expected)
diff --git a/pandas/pandas/tests/test_common.py b/pandas/pandas/tests/test_common.py
new file mode 100644
index 0000000000000..9f68ef5ffdc47
--- /dev/null
+++ b/pandas/pandas/tests/test_common.py
@@ -0,0 +1,269 @@
+import collections
+from functools import partial
+import string
+import subprocess
+import sys
+
+import numpy as np
+import pytest
+
+from pandas.compat import WASM
+
+import pandas as pd
+from pandas import Series
+import pandas._testing as tm
+from pandas.core import ops
+import pandas.core.common as com
+from pandas.util.version import Version
+
+
+class TestGetCallableName:
+    def fn(self, x):
+        return x
+
+    partial1 = partial(fn)
+    partial2 = partial(partial1)
+    lambda_ = lambda x: x
+
+    class SomeCall:
+        def __call__(self):
+            # This shouldn't actually get called below; SomeCall.__init__
+            #  should.
+            raise NotImplementedError
+
+    @pytest.mark.parametrize(
+        "func, expected",
+        [
+            (fn, "fn"),
+            (partial1, "fn"),
+            (partial2, "fn"),
+            (lambda_, "<lambda>"),
+            (SomeCall(), "SomeCall"),
+            (1, None),
+        ],
+    )
+    def test_get_callable_name(self, func, expected):
+        assert com.get_callable_name(func) == expected
+
+
+class TestRandomState:
+    def test_seed(self):
+        seed = 5
+        assert com.random_state(seed).uniform() == np.random.RandomState(seed).uniform()
+
+    def test_object(self):
+        seed = 10
+        state_obj = np.random.RandomState(seed)
+        assert (
+            com.random_state(state_obj).uniform()
+            == np.random.RandomState(seed).uniform()
+        )
+
+    def test_default(self):
+        assert com.random_state() is np.random
+
+    def test_array_like(self):
+        state = np.random.default_rng(None).integers(0, 2**31, size=624, dtype="uint32")
+        assert (
+            com.random_state(state).uniform() == np.random.RandomState(state).uniform()
+        )
+
+    def test_bit_generators(self):
+        seed = 3
+        assert (
+            com.random_state(np.random.MT19937(seed)).uniform()
+            == np.random.RandomState(np.random.MT19937(seed)).uniform()
+        )
+
+        seed = 11
+        assert (
+            com.random_state(np.random.PCG64(seed)).uniform()
+            == np.random.RandomState(np.random.PCG64(seed)).uniform()
+        )
+
+    @pytest.mark.parametrize("state", ["test", 5.5])
+    def test_error(self, state):
+        msg = (
+            "random_state must be an integer, array-like, a BitGenerator, Generator, "
+            "a numpy RandomState, or None"
+        )
+        with pytest.raises(ValueError, match=msg):
+            com.random_state(state)
+
+
+@pytest.mark.parametrize("args, expected", [((1, 2, None), True), ((1, 2, 3), False)])
+def test_any_none(args, expected):
+    assert com.any_none(*args) is expected
+
+
+@pytest.mark.parametrize(
+    "args, expected",
+    [((1, 2, 3), True), ((1, 2, None), False), ((None, None, None), False)],
+)
+def test_all_not_none(args, expected):
+    assert com.all_not_none(*args) is expected
+
+
+@pytest.mark.parametrize(
+    "left, right, expected",
+    [
+        (Series([1], name="x"), Series([2], name="x"), "x"),
+        (Series([1], name="x"), Series([2], name="y"), None),
+        (Series([1]), Series([2], name="x"), None),
+        (Series([1], name="x"), Series([2]), None),
+        (Series([1], name="x"), [2], "x"),
+        ([1], Series([2], name="y"), "y"),
+        # matching NAs
+        (Series([1], name=np.nan), pd.Index([], name=np.nan), np.nan),
+        (Series([1], name=np.nan), pd.Index([], name=pd.NaT), None),
+        (Series([1], name=pd.NA), pd.Index([], name=pd.NA), pd.NA),
+        # tuple name GH#39757
+        (
+            Series([1], name=np.int64(1)),
+            pd.Index([], name=(np.int64(1), np.int64(2))),
+            None,
+        ),
+        (
+            Series([1], name=(np.int64(1), np.int64(2))),
+            pd.Index([], name=(np.int64(1), np.int64(2))),
+            (np.int64(1), np.int64(2)),
+        ),
+        pytest.param(
+            Series([1], name=(np.float64("nan"), np.int64(2))),
+            pd.Index([], name=(np.float64("nan"), np.int64(2))),
+            (np.float64("nan"), np.int64(2)),
+            marks=pytest.mark.xfail(
+                reason="Not checking for matching NAs inside tuples."
+            ),
+        ),
+    ],
+)
+def test_maybe_match_name(left, right, expected):
+    res = ops.common._maybe_match_name(left, right)
+    assert res is expected or res == expected
+
+
+@pytest.mark.parametrize(
+    "into, msg",
+    [
+        (
+            # uninitialized defaultdict
+            collections.defaultdict,
+            r"to_dict\(\) only accepts initialized defaultdicts",
+        ),
+        (
+            # non-mapping subtypes,, instance
+            [],
+            "unsupported type: <class 'list'>",
+        ),
+        (
+            # non-mapping subtypes, class
+            list,
+            "unsupported type: <class 'list'>",
+        ),
+    ],
+)
+def test_standardize_mapping_type_error(into, msg):
+    with pytest.raises(TypeError, match=msg):
+        com.standardize_mapping(into)
+
+
+def test_standardize_mapping():
+    fill = {"bad": "data"}
+    assert com.standardize_mapping(fill) == dict
+
+    # Convert instance to type
+    assert com.standardize_mapping({}) == dict
+
+    dd = collections.defaultdict(list)
+    assert isinstance(com.standardize_mapping(dd), partial)
+
+
+def test_git_version():
+    # GH 21295
+    git_version = pd.__git_version__
+    assert len(git_version) == 40
+    assert all(c in string.hexdigits for c in git_version)
+
+
+def test_version_tag():
+    version = Version(pd.__version__)
+    try:
+        version > Version("0.0.1")
+    except TypeError as err:
+        raise ValueError(
+            "No git tags exist, please sync tags between upstream and your repo"
+        ) from err
+
+
+@pytest.mark.parametrize(
+    "obj", [(obj,) for obj in pd.__dict__.values() if callable(obj)]
+)
+def test_serializable(obj):
+    # GH 35611
+    unpickled = tm.round_trip_pickle(obj)
+    assert type(obj) == type(unpickled)
+
+
+class TestIsBoolIndexer:
+    def test_non_bool_array_with_na(self):
+        # in particular, this should not raise
+        arr = np.array(["A", "B", np.nan], dtype=object)
+        assert not com.is_bool_indexer(arr)
+
+    def test_list_subclass(self):
+        # GH#42433
+
+        class MyList(list):
+            pass
+
+        val = MyList(["a"])
+
+        assert not com.is_bool_indexer(val)
+
+        val = MyList([True])
+        assert com.is_bool_indexer(val)
+
+    def test_frozenlist(self):
+        # GH#42461
+        data = {"col1": [1, 2], "col2": [3, 4]}
+        df = pd.DataFrame(data=data)
+
+        frozen = df.index.names[1:]
+        assert not com.is_bool_indexer(frozen)
+
+        result = df[frozen]
+        expected = df[[]]
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("with_exception", [True, False])
+def test_temp_setattr(with_exception):
+    # GH#45954
+    ser = Series(dtype=object)
+    ser.name = "first"
+    # Raise a ValueError in either case to satisfy pytest.raises
+    match = "Inside exception raised" if with_exception else "Outside exception raised"
+    with pytest.raises(ValueError, match=match):
+        with com.temp_setattr(ser, "name", "second"):
+            assert ser.name == "second"
+            if with_exception:
+                raise ValueError("Inside exception raised")
+        raise ValueError("Outside exception raised")
+    assert ser.name == "first"
+
+
+@pytest.mark.skipif(WASM, reason="Can't start subprocesses in WASM")
+@pytest.mark.single_cpu
+def test_str_size():
+    # GH#21758
+    a = "a"
+    expected = sys.getsizeof(a)
+    pyexe = sys.executable.replace("\\", "/")
+    call = [
+        pyexe,
+        "-c",
+        "a='a';import sys;sys.getsizeof(a);import pandas;print(sys.getsizeof(a));",
+    ]
+    result = subprocess.check_output(call).decode()[-4:-1].strip("\n")
+    assert int(result) == int(expected)
diff --git a/pandas/pandas/tests/test_downstream.py b/pandas/pandas/tests/test_downstream.py
new file mode 100644
index 0000000000000..76fad35304fe6
--- /dev/null
+++ b/pandas/pandas/tests/test_downstream.py
@@ -0,0 +1,321 @@
+"""
+Testing that we work in the downstream packages
+"""
+
+import array
+from functools import partial
+import subprocess
+import sys
+
+import numpy as np
+import pytest
+
+from pandas.errors import IntCastingNaNError
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    TimedeltaIndex,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+
+@pytest.fixture
+def df():
+    return DataFrame({"A": [1, 2, 3]})
+
+
+def test_dask(df):
+    # dask sets "compute.use_numexpr" to False, so catch the current value
+    # and ensure to reset it afterwards to avoid impacting other tests
+    olduse = pd.get_option("compute.use_numexpr")
+
+    try:
+        pytest.importorskip("toolz")
+        dd = pytest.importorskip("dask.dataframe")
+
+        ddf = dd.from_pandas(df, npartitions=3)
+        assert ddf.A is not None
+        assert ddf.compute() is not None
+    finally:
+        pd.set_option("compute.use_numexpr", olduse)
+
+
+# TODO(CoW) see https://github.com/pandas-dev/pandas/pull/51082
+@pytest.mark.skip(reason="not implemented with CoW")
+def test_dask_ufunc():
+    # dask sets "compute.use_numexpr" to False, so catch the current value
+    # and ensure to reset it afterwards to avoid impacting other tests
+    olduse = pd.get_option("compute.use_numexpr")
+
+    try:
+        da = pytest.importorskip("dask.array")
+        dd = pytest.importorskip("dask.dataframe")
+
+        s = Series([1.5, 2.3, 3.7, 4.0])
+        ds = dd.from_pandas(s, npartitions=2)
+
+        result = da.log(ds).compute()
+        expected = np.log(s)
+        tm.assert_series_equal(result, expected)
+    finally:
+        pd.set_option("compute.use_numexpr", olduse)
+
+
+def test_construct_dask_float_array_int_dtype_match_ndarray():
+    # GH#40110 make sure we treat a float-dtype dask array with the same
+    #  rules we would for an ndarray
+    dd = pytest.importorskip("dask.dataframe")
+
+    arr = np.array([1, 2.5, 3])
+    darr = dd.from_array(arr)
+
+    res = Series(darr)
+    expected = Series(arr)
+    tm.assert_series_equal(res, expected)
+
+    # GH#49599 in 2.0 we raise instead of silently ignoring the dtype
+    msg = "Trying to coerce float values to integers"
+    with pytest.raises(ValueError, match=msg):
+        Series(darr, dtype="i8")
+
+    msg = r"Cannot convert non-finite values \(NA or inf\) to integer"
+    arr[2] = np.nan
+    with pytest.raises(IntCastingNaNError, match=msg):
+        Series(darr, dtype="i8")
+    # which is the same as we get with a numpy input
+    with pytest.raises(IntCastingNaNError, match=msg):
+        Series(arr, dtype="i8")
+
+
+def test_xarray(df):
+    pytest.importorskip("xarray")
+
+    assert df.to_xarray() is not None
+
+
+def test_xarray_cftimeindex_nearest():
+    # https://github.com/pydata/xarray/issues/3751
+    cftime = pytest.importorskip("cftime")
+    xarray = pytest.importorskip("xarray")
+
+    times = xarray.cftime_range("0001", periods=2)
+    key = cftime.DatetimeGregorian(2000, 1, 1)
+    result = times.get_indexer([key], method="nearest")
+    expected = 1
+    assert result == expected
+
+
+@pytest.mark.single_cpu
+def test_oo_optimizable():
+    # GH 21071
+    subprocess.check_call([sys.executable, "-OO", "-c", "import pandas"])
+
+
+@pytest.mark.single_cpu
+def test_oo_optimized_datetime_index_unpickle():
+    # GH 42866
+    subprocess.check_call(
+        [
+            sys.executable,
+            "-OO",
+            "-c",
+            (
+                "import pandas as pd, pickle; "
+                "pickle.loads(pickle.dumps(pd.date_range('2021-01-01', periods=1)))"
+            ),
+        ]
+    )
+
+
+def test_statsmodels():
+    smf = pytest.importorskip("statsmodels.formula.api")
+
+    df = DataFrame(
+        {"Lottery": range(5), "Literacy": range(5), "Pop1831": range(100, 105)}
+    )
+    smf.ols("Lottery ~ Literacy + np.log(Pop1831)", data=df).fit()
+
+
+def test_scikit_learn():
+    pytest.importorskip("sklearn")
+    from sklearn import (
+        datasets,
+        svm,
+    )
+
+    digits = datasets.load_digits()
+    clf = svm.SVC(gamma=0.001, C=100.0)
+    clf.fit(digits.data[:-1], digits.target[:-1])
+    clf.predict(digits.data[-1:])
+
+
+def test_seaborn(mpl_cleanup):
+    seaborn = pytest.importorskip("seaborn")
+    tips = DataFrame(
+        {"day": pd.date_range("2023", freq="D", periods=5), "total_bill": range(5)}
+    )
+    seaborn.stripplot(x="day", y="total_bill", data=tips)
+
+
+def test_pandas_datareader():
+    pytest.importorskip("pandas_datareader")
+
+
+@pytest.mark.filterwarnings("ignore:Passing a BlockManager:DeprecationWarning")
+def test_pyarrow(df):
+    pyarrow = pytest.importorskip("pyarrow")
+    table = pyarrow.Table.from_pandas(df)
+    result = table.to_pandas()
+    tm.assert_frame_equal(result, df)
+
+
+def test_yaml_dump(df):
+    # GH#42748
+    yaml = pytest.importorskip("yaml")
+
+    dumped = yaml.dump(df)
+
+    loaded = yaml.load(dumped, Loader=yaml.Loader)
+    tm.assert_frame_equal(df, loaded)
+
+    loaded2 = yaml.load(dumped, Loader=yaml.UnsafeLoader)
+    tm.assert_frame_equal(df, loaded2)
+
+
+@pytest.mark.single_cpu
+def test_missing_required_dependency():
+    # GH 23868
+    # To ensure proper isolation, we pass these flags
+    # -S : disable site-packages
+    # -s : disable user site-packages
+    # -E : disable PYTHON* env vars, especially PYTHONPATH
+    # https://github.com/MacPython/pandas-wheels/pull/50
+
+    pyexe = sys.executable.replace("\\", "/")
+
+    # We skip this test if pandas is installed as a site package. We first
+    # import the package normally and check the path to the module before
+    # executing the test which imports pandas with site packages disabled.
+    call = [pyexe, "-c", "import pandas;print(pandas.__file__)"]
+    output = subprocess.check_output(call).decode()
+    if "site-packages" in output:
+        pytest.skip("pandas installed as site package")
+
+    # This test will fail if pandas is installed as a site package. The flags
+    # prevent pandas being imported and the test will report Failed: DID NOT
+    # RAISE <class 'subprocess.CalledProcessError'>
+    call = [pyexe, "-sSE", "-c", "import pandas"]
+
+    msg = (
+        rf"Command '\['{pyexe}', '-sSE', '-c', 'import pandas'\]' "
+        "returned non-zero exit status 1."
+    )
+
+    with pytest.raises(subprocess.CalledProcessError, match=msg) as exc:
+        subprocess.check_output(call, stderr=subprocess.STDOUT)
+
+    output = exc.value.stdout.decode()
+    for name in ["numpy", "dateutil"]:
+        assert name in output
+
+
+def test_frame_setitem_dask_array_into_new_col(request):
+    # GH#47128
+
+    # dask sets "compute.use_numexpr" to False, so catch the current value
+    # and ensure to reset it afterwards to avoid impacting other tests
+    olduse = pd.get_option("compute.use_numexpr")
+
+    try:
+        dask = pytest.importorskip("dask")
+        da = pytest.importorskip("dask.array")
+        if Version(dask.__version__) <= Version("2025.1.0") and Version(
+            np.__version__
+        ) >= Version("2.1"):
+            request.applymarker(
+                pytest.mark.xfail(reason="loc.__setitem__ incorrectly mutated column c")
+            )
+
+        dda = da.array([1, 2])
+        df = DataFrame({"a": ["a", "b"]})
+        df["b"] = dda
+        df["c"] = dda
+        df.loc[[False, True], "b"] = 100
+        result = df.loc[[1], :]
+        expected = DataFrame({"a": ["b"], "b": [100], "c": [2]}, index=[1])
+        tm.assert_frame_equal(result, expected)
+    finally:
+        pd.set_option("compute.use_numexpr", olduse)
+
+
+def test_pandas_priority():
+    # GH#48347
+
+    class MyClass:
+        __pandas_priority__ = 5000
+
+        def __radd__(self, other):
+            return self
+
+    left = MyClass()
+    right = Series(range(3))
+
+    assert right.__add__(left) is NotImplemented
+    assert right + left is left
+
+
+@pytest.mark.parametrize("dtype", ["M8[ns]", "m8[ns]"])
+@pytest.mark.parametrize(
+    "box", [memoryview, partial(array.array, "i"), "dask", "xarray"]
+)
+def test_from_obscure_array(dtype, box):
+    # GH#24539 recognize e.g xarray, dask, ...
+    # Note: we dont do this for PeriodArray bc _from_sequence won't accept
+    #  an array of integers
+    # TODO: could check with arraylike of Period objects
+    # GH#24539 recognize e.g xarray, dask, ...
+    arr = np.array([1, 2, 3], dtype=np.int64)
+    if box == "dask":
+        da = pytest.importorskip("dask.array")
+        data = da.array(arr)
+    elif box == "xarray":
+        xr = pytest.importorskip("xarray")
+        data = xr.DataArray(arr)
+    else:
+        data = box(arr)
+
+    if not isinstance(data, memoryview):
+        # FIXME(GH#44431) these raise on memoryview and attempted fix
+        #  fails on py3.10
+        func = {"M8[ns]": pd.to_datetime, "m8[ns]": pd.to_timedelta}[dtype]
+        result = func(arr).array
+        expected = func(data).array
+        tm.assert_equal(result, expected)
+
+    # Let's check the Indexes while we're here
+    idx_cls = {"M8[ns]": DatetimeIndex, "m8[ns]": TimedeltaIndex}[dtype]
+    result = idx_cls(arr)
+    expected = idx_cls(data)
+    tm.assert_index_equal(result, expected)
+
+
+def test_xarray_coerce_unit():
+    # GH44053
+    xr = pytest.importorskip("xarray")
+
+    arr = xr.DataArray([1, 2, 3])
+    result = pd.to_datetime(arr, unit="ns")
+    expected = DatetimeIndex(
+        [
+            "1970-01-01 00:00:00.000000001",
+            "1970-01-01 00:00:00.000000002",
+            "1970-01-01 00:00:00.000000003",
+        ],
+        dtype="datetime64[ns]",
+        freq=None,
+    )
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/test_errors.py b/pandas/pandas/tests/test_errors.py
new file mode 100644
index 0000000000000..c5c4b234eb129
--- /dev/null
+++ b/pandas/pandas/tests/test_errors.py
@@ -0,0 +1,104 @@
+import pytest
+
+from pandas.errors import (
+    AbstractMethodError,
+    UndefinedVariableError,
+)
+
+import pandas as pd
+
+
+@pytest.mark.parametrize(
+    "exc",
+    [
+        "AttributeConflictWarning",
+        "CSSWarning",
+        "CategoricalConversionWarning",
+        "ClosedFileError",
+        "DataError",
+        "DatabaseError",
+        "DtypeWarning",
+        "EmptyDataError",
+        "IncompatibilityWarning",
+        "IndexingError",
+        "InvalidColumnName",
+        "InvalidComparison",
+        "InvalidVersion",
+        "LossySetitemError",
+        "MergeError",
+        "NoBufferPresent",
+        "NumExprClobberingError",
+        "NumbaUtilError",
+        "OptionError",
+        "OutOfBoundsDatetime",
+        "ParserError",
+        "ParserWarning",
+        "PerformanceWarning",
+        "PossibleDataLossError",
+        "PossiblePrecisionLoss",
+        "PyperclipException",
+        "SpecificationError",
+        "UnsortedIndexError",
+        "UnsupportedFunctionCall",
+        "ValueLabelTypeMismatch",
+    ],
+)
+def test_exception_importable(exc):
+    from pandas import errors
+
+    err = getattr(errors, exc)
+    assert err is not None
+
+    # check that we can raise on them
+
+    msg = "^$"
+
+    with pytest.raises(err, match=msg):
+        raise err()
+
+
+def test_catch_oob():
+    from pandas import errors
+
+    msg = "Cannot cast 1500-01-01 00:00:00 to unit='ns' without overflow"
+    with pytest.raises(errors.OutOfBoundsDatetime, match=msg):
+        pd.Timestamp("15000101").as_unit("ns")
+
+
+@pytest.mark.parametrize("is_local", [True, False])
+def test_catch_undefined_variable_error(is_local):
+    variable_name = "x"
+    if is_local:
+        msg = f"local variable '{variable_name}' is not defined"
+    else:
+        msg = f"name '{variable_name}' is not defined"
+
+    with pytest.raises(UndefinedVariableError, match=msg):
+        raise UndefinedVariableError(variable_name, is_local)
+
+
+class Foo:
+    @classmethod
+    def classmethod(cls):
+        raise AbstractMethodError(cls, methodtype="classmethod")
+
+    @property
+    def property(self):
+        raise AbstractMethodError(self, methodtype="property")
+
+    def method(self):
+        raise AbstractMethodError(self)
+
+
+def test_AbstractMethodError_classmethod():
+    xpr = "This classmethod must be defined in the concrete class Foo"
+    with pytest.raises(AbstractMethodError, match=xpr):
+        Foo.classmethod()
+
+    xpr = "This property must be defined in the concrete class Foo"
+    with pytest.raises(AbstractMethodError, match=xpr):
+        Foo().property
+
+    xpr = "This method must be defined in the concrete class Foo"
+    with pytest.raises(AbstractMethodError, match=xpr):
+        Foo().method()
diff --git a/pandas/pandas/tests/test_expressions.py b/pandas/pandas/tests/test_expressions.py
new file mode 100644
index 0000000000000..8f275345a7819
--- /dev/null
+++ b/pandas/pandas/tests/test_expressions.py
@@ -0,0 +1,464 @@
+import operator
+import re
+
+import numpy as np
+import pytest
+
+from pandas import option_context
+import pandas._testing as tm
+from pandas.core.api import DataFrame
+from pandas.core.computation import expressions as expr
+
+
+@pytest.fixture
+def _frame():
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((10001, 4)),
+        columns=list("ABCD"),
+        dtype="float64",
+    )
+
+
+@pytest.fixture
+def _frame2():
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((100, 4)),
+        columns=list("ABCD"),
+        dtype="float64",
+    )
+
+
+@pytest.fixture
+def _mixed(_frame):
+    return DataFrame(
+        {
+            "A": _frame["A"],
+            "B": _frame["B"].astype("float32"),
+            "C": _frame["C"].astype("int64"),
+            "D": _frame["D"].astype("int32"),
+        }
+    )
+
+
+@pytest.fixture
+def _mixed2(_frame2):
+    return DataFrame(
+        {
+            "A": _frame2["A"],
+            "B": _frame2["B"].astype("float32"),
+            "C": _frame2["C"].astype("int64"),
+            "D": _frame2["D"].astype("int32"),
+        }
+    )
+
+
+@pytest.fixture
+def _integer():
+    return DataFrame(
+        np.random.default_rng(2).integers(1, 100, size=(10001, 4)),
+        columns=list("ABCD"),
+        dtype="int64",
+    )
+
+
+@pytest.fixture
+def _integer_integers(_integer):
+    # integers to get a case with zeros
+    return _integer * np.random.default_rng(2).integers(0, 2, size=np.shape(_integer))
+
+
+@pytest.fixture
+def _integer2():
+    return DataFrame(
+        np.random.default_rng(2).integers(1, 100, size=(101, 4)),
+        columns=list("ABCD"),
+        dtype="int64",
+    )
+
+
+@pytest.fixture
+def _array(_frame):
+    return _frame["A"].to_numpy()
+
+
+@pytest.fixture
+def _array2(_frame2):
+    return _frame2["A"].to_numpy()
+
+
+@pytest.fixture
+def _array_mixed(_mixed):
+    return _mixed["D"].to_numpy()
+
+
+@pytest.fixture
+def _array_mixed2(_mixed2):
+    return _mixed2["D"].to_numpy()
+
+
+@pytest.mark.skipif(not expr.USE_NUMEXPR, reason="not using numexpr")
+class TestExpressions:
+    @staticmethod
+    def call_op(df, other, flex: bool, opname: str):
+        if flex:
+            op = lambda x, y: getattr(x, opname)(y)
+            op.__name__ = opname
+        else:
+            op = getattr(operator, opname)
+
+        with option_context("compute.use_numexpr", False):
+            expected = op(df, other)
+
+        expr.get_test_result()
+
+        result = op(df, other)
+        return result, expected
+
+    @pytest.mark.parametrize(
+        "fixture",
+        [
+            "_integer",
+            "_integer2",
+            "_integer_integers",
+            "_frame",
+            "_frame2",
+            "_mixed",
+            "_mixed2",
+        ],
+    )
+    @pytest.mark.parametrize("flex", [True, False])
+    @pytest.mark.parametrize(
+        "arith", ["add", "sub", "mul", "mod", "truediv", "floordiv"]
+    )
+    def test_run_arithmetic(self, request, fixture, flex, arith, monkeypatch):
+        df = request.getfixturevalue(fixture)
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 0)
+            result, expected = self.call_op(df, df, flex, arith)
+
+            if arith == "truediv":
+                assert all(x.kind == "f" for x in expected.dtypes.values)
+            tm.assert_equal(expected, result)
+
+            for i in range(len(df.columns)):
+                result, expected = self.call_op(
+                    df.iloc[:, i], df.iloc[:, i], flex, arith
+                )
+                if arith == "truediv":
+                    assert expected.dtype.kind == "f"
+                tm.assert_equal(expected, result)
+
+    @pytest.mark.parametrize(
+        "fixture",
+        [
+            "_integer",
+            "_integer2",
+            "_integer_integers",
+            "_frame",
+            "_frame2",
+            "_mixed",
+            "_mixed2",
+        ],
+    )
+    @pytest.mark.parametrize("flex", [True, False])
+    def test_run_binary(self, request, fixture, flex, comparison_op, monkeypatch):
+        """
+        tests solely that the result is the same whether or not numexpr is
+        enabled.  Need to test whether the function does the correct thing
+        elsewhere.
+        """
+        df = request.getfixturevalue(fixture)
+        arith = comparison_op.__name__
+        with option_context("compute.use_numexpr", False):
+            other = df + 1
+
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 0)
+            expr.set_test_mode(True)
+
+            result, expected = self.call_op(df, other, flex, arith)
+
+            used_numexpr = expr.get_test_result()
+            assert used_numexpr, "Did not use numexpr as expected."
+            tm.assert_equal(expected, result)
+
+            for i in range(len(df.columns)):
+                binary_comp = other.iloc[:, i] + 1
+                self.call_op(df.iloc[:, i], binary_comp, flex, "add")
+
+    def test_invalid(self):
+        array = np.random.default_rng(2).standard_normal(1_000_001)
+        array2 = np.random.default_rng(2).standard_normal(100)
+
+        # no op
+        result = expr._can_use_numexpr(operator.add, None, array, array, "evaluate")
+        assert not result
+
+        # min elements
+        result = expr._can_use_numexpr(operator.add, "+", array2, array2, "evaluate")
+        assert not result
+
+        # ok, we only check on first part of expression
+        result = expr._can_use_numexpr(operator.add, "+", array, array2, "evaluate")
+        assert result
+
+    @pytest.mark.filterwarnings("ignore:invalid value encountered in:RuntimeWarning")
+    @pytest.mark.parametrize(
+        "opname,op_str",
+        [("add", "+"), ("sub", "-"), ("mul", "*"), ("truediv", "/"), ("pow", "**")],
+    )
+    @pytest.mark.parametrize(
+        "left_fix,right_fix", [("_array", "_array2"), ("_array_mixed", "_array_mixed2")]
+    )
+    def test_binary_ops(self, request, opname, op_str, left_fix, right_fix):
+        left = request.getfixturevalue(left_fix)
+        right = request.getfixturevalue(right_fix)
+
+        def testit(left, right, opname, op_str):
+            if opname == "pow":
+                left = np.abs(left)
+
+            op = getattr(operator, opname)
+
+            # array has 0s
+            result = expr.evaluate(op, left, left, use_numexpr=True)
+            expected = expr.evaluate(op, left, left, use_numexpr=False)
+            tm.assert_numpy_array_equal(result, expected)
+
+            result = expr._can_use_numexpr(op, op_str, right, right, "evaluate")
+            assert not result
+
+        with option_context("compute.use_numexpr", False):
+            testit(left, right, opname, op_str)
+
+        expr.set_numexpr_threads(1)
+        testit(left, right, opname, op_str)
+        expr.set_numexpr_threads()
+        testit(left, right, opname, op_str)
+
+    @pytest.mark.parametrize(
+        "left_fix,right_fix", [("_array", "_array2"), ("_array_mixed", "_array_mixed2")]
+    )
+    def test_comparison_ops(self, request, comparison_op, left_fix, right_fix):
+        left = request.getfixturevalue(left_fix)
+        right = request.getfixturevalue(right_fix)
+
+        def testit():
+            f12 = left + 1
+            f22 = right + 1
+
+            op = comparison_op
+
+            result = expr.evaluate(op, left, f12, use_numexpr=True)
+            expected = expr.evaluate(op, left, f12, use_numexpr=False)
+            tm.assert_numpy_array_equal(result, expected)
+
+            result = expr._can_use_numexpr(op, op, right, f22, "evaluate")
+            assert not result
+
+        with option_context("compute.use_numexpr", False):
+            testit()
+
+        expr.set_numexpr_threads(1)
+        testit()
+        expr.set_numexpr_threads()
+        testit()
+
+    @pytest.mark.parametrize("cond", [True, False])
+    @pytest.mark.parametrize("fixture", ["_frame", "_frame2", "_mixed", "_mixed2"])
+    def test_where(self, request, cond, fixture):
+        df = request.getfixturevalue(fixture)
+
+        def testit():
+            c = np.empty(df.shape, dtype=np.bool_)
+            c.fill(cond)
+            result = expr.where(c, df.values, df.values + 1)
+            expected = np.where(c, df.values, df.values + 1)
+            tm.assert_numpy_array_equal(result, expected)
+
+        with option_context("compute.use_numexpr", False):
+            testit()
+
+        expr.set_numexpr_threads(1)
+        testit()
+        expr.set_numexpr_threads()
+        testit()
+
+    @pytest.mark.parametrize(
+        "op_str,opname", [("/", "truediv"), ("//", "floordiv"), ("**", "pow")]
+    )
+    def test_bool_ops_raise_on_arithmetic(self, op_str, opname):
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).random(10) > 0.5,
+                "b": np.random.default_rng(2).random(10) > 0.5,
+            }
+        )
+
+        msg = f"operator '{opname}' not implemented for bool dtypes"
+        f = getattr(operator, opname)
+        err_msg = re.escape(msg)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(df, df)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(df.a, df.b)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(df.a, True)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(False, df.a)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(False, df)
+
+        with pytest.raises(NotImplementedError, match=err_msg):
+            f(df, True)
+
+    @pytest.mark.parametrize(
+        "op_str,opname", [("+", "add"), ("*", "mul"), ("-", "sub")]
+    )
+    def test_bool_ops_warn_on_arithmetic(self, op_str, opname, monkeypatch):
+        n = 10
+        df = DataFrame(
+            {
+                "a": np.random.default_rng(2).random(n) > 0.5,
+                "b": np.random.default_rng(2).random(n) > 0.5,
+            }
+        )
+
+        subs = {"+": "|", "*": "&", "-": "^"}
+        sub_funcs = {"|": "or_", "&": "and_", "^": "xor"}
+
+        f = getattr(operator, opname)
+        fe = getattr(operator, sub_funcs[subs[op_str]])
+
+        if op_str == "-":
+            # raises TypeError
+            return
+
+        msg = "operator is not supported by numexpr"
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 5)
+            with option_context("compute.use_numexpr", True):
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(df, df)
+                    e = fe(df, df)
+                    tm.assert_frame_equal(r, e)
+
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(df.a, df.b)
+                    e = fe(df.a, df.b)
+                    tm.assert_series_equal(r, e)
+
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(df.a, True)
+                    e = fe(df.a, True)
+                    tm.assert_series_equal(r, e)
+
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(False, df.a)
+                    e = fe(False, df.a)
+                    tm.assert_series_equal(r, e)
+
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(False, df)
+                    e = fe(False, df)
+                    tm.assert_frame_equal(r, e)
+
+                with tm.assert_produces_warning(UserWarning, match=msg):
+                    r = f(df, True)
+                    e = fe(df, True)
+                    tm.assert_frame_equal(r, e)
+
+    @pytest.mark.parametrize(
+        "test_input,expected",
+        [
+            (
+                DataFrame(
+                    [[0, 1, 2, "aa"], [0, 1, 2, "aa"]], columns=["a", "b", "c", "dtype"]
+                ),
+                DataFrame([[False, False], [False, False]], columns=["a", "dtype"]),
+            ),
+            (
+                DataFrame(
+                    [[0, 3, 2, "aa"], [0, 4, 2, "aa"], [0, 1, 1, "bb"]],
+                    columns=["a", "b", "c", "dtype"],
+                ),
+                DataFrame(
+                    [[False, False], [False, False], [False, False]],
+                    columns=["a", "dtype"],
+                ),
+            ),
+        ],
+    )
+    def test_bool_ops_column_name_dtype(self, test_input, expected):
+        # GH 22383 - .ne fails if columns containing column name 'dtype'
+        result = test_input.loc[:, ["a", "dtype"]].ne(test_input.loc[:, ["a", "dtype"]])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arith", ("add", "sub", "mul", "mod", "truediv", "floordiv")
+    )
+    @pytest.mark.parametrize("axis", (0, 1))
+    def test_frame_series_axis(self, axis, arith, _frame, monkeypatch):
+        # GH#26736 Dataframe.floordiv(Series, axis=1) fails
+
+        df = _frame
+        if axis == 1:
+            other = df.iloc[0, :]
+        else:
+            other = df.iloc[:, 0]
+
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 0)
+
+            op_func = getattr(df, arith)
+
+            with option_context("compute.use_numexpr", False):
+                expected = op_func(other, axis=axis)
+
+            result = op_func(other, axis=axis)
+            tm.assert_frame_equal(expected, result)
+
+    @pytest.mark.parametrize(
+        "op",
+        [
+            "__mod__",
+            "__rmod__",
+            "__floordiv__",
+            "__rfloordiv__",
+        ],
+    )
+    @pytest.mark.parametrize("scalar", [-5, 5])
+    def test_python_semantics_with_numexpr_installed(
+        self, op, box_with_array, scalar, monkeypatch
+    ):
+        # https://github.com/pandas-dev/pandas/issues/36047
+        with monkeypatch.context() as m:
+            m.setattr(expr, "_MIN_ELEMENTS", 0)
+            data = np.arange(-50, 50)
+            obj = box_with_array(data)
+            method = getattr(obj, op)
+            result = method(scalar)
+
+            # compare result with numpy
+            with option_context("compute.use_numexpr", False):
+                expected = method(scalar)
+
+            tm.assert_equal(result, expected)
+
+            # compare result element-wise with Python
+            for i, elem in enumerate(data):
+                if box_with_array == DataFrame:
+                    scalar_result = result.iloc[i, 0]
+                else:
+                    scalar_result = result[i]
+                try:
+                    expected = getattr(int(elem), op)(scalar)
+                except ZeroDivisionError:
+                    pass
+                else:
+                    assert scalar_result == expected
diff --git a/pandas/pandas/tests/test_flags.py b/pandas/pandas/tests/test_flags.py
new file mode 100644
index 0000000000000..9294b3fc3319b
--- /dev/null
+++ b/pandas/pandas/tests/test_flags.py
@@ -0,0 +1,48 @@
+import pytest
+
+import pandas as pd
+
+
+class TestFlags:
+    def test_equality(self):
+        a = pd.DataFrame().set_flags(allows_duplicate_labels=True).flags
+        b = pd.DataFrame().set_flags(allows_duplicate_labels=False).flags
+
+        assert a == a
+        assert b == b
+        assert a != b
+        assert a != 2
+
+    def test_set(self):
+        df = pd.DataFrame().set_flags(allows_duplicate_labels=True)
+        a = df.flags
+        a.allows_duplicate_labels = False
+        assert a.allows_duplicate_labels is False
+        a["allows_duplicate_labels"] = True
+        assert a.allows_duplicate_labels is True
+
+    def test_repr(self):
+        a = repr(pd.DataFrame({"A"}).set_flags(allows_duplicate_labels=True).flags)
+        assert a == "<Flags(allows_duplicate_labels=True)>"
+        a = repr(pd.DataFrame({"A"}).set_flags(allows_duplicate_labels=False).flags)
+        assert a == "<Flags(allows_duplicate_labels=False)>"
+
+    def test_obj_ref(self):
+        df = pd.DataFrame()
+        flags = df.flags
+        del df
+        with pytest.raises(ValueError, match="object has been deleted"):
+            flags.allows_duplicate_labels = True
+
+    def test_getitem(self):
+        df = pd.DataFrame()
+        flags = df.flags
+        assert flags["allows_duplicate_labels"] is True
+        flags["allows_duplicate_labels"] = False
+        assert flags["allows_duplicate_labels"] is False
+
+        with pytest.raises(KeyError, match="a"):
+            flags["a"]
+
+        with pytest.raises(ValueError, match="a"):
+            flags["a"] = 10
diff --git a/pandas/pandas/tests/test_multilevel.py b/pandas/pandas/tests/test_multilevel.py
new file mode 100644
index 0000000000000..a23e6d9b3973a
--- /dev/null
+++ b/pandas/pandas/tests/test_multilevel.py
@@ -0,0 +1,347 @@
+import datetime
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+)
+import pandas._testing as tm
+
+
+class TestMultiLevel:
+    def test_reindex_level(self, multiindex_year_month_day_dataframe_random_data):
+        # axis=0
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        month_sums = ymd.groupby("month").sum()
+        result = month_sums.reindex(ymd.index, level=1)
+        expected = ymd.groupby(level="month").transform("sum")
+
+        tm.assert_frame_equal(result, expected)
+
+        # Series
+        result = month_sums["A"].reindex(ymd.index, level=1)
+        expected = ymd["A"].groupby(level="month").transform("sum")
+        tm.assert_series_equal(result, expected, check_names=False)
+
+    def test_reindex(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        expected = frame.iloc[[0, 3]]
+        reindexed = frame.loc[[("foo", "one"), ("bar", "one")]]
+        tm.assert_frame_equal(reindexed, expected)
+
+    def test_reindex_preserve_levels(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        new_index = ymd.index[::10]
+        chunk = ymd.reindex(new_index)
+        assert chunk.index.is_(new_index)
+
+        chunk = ymd.loc[new_index]
+        assert chunk.index.equals(new_index)
+
+        ymdT = ymd.T
+        chunk = ymdT.reindex(columns=new_index)
+        assert chunk.columns.is_(new_index)
+
+        chunk = ymdT.loc[:, new_index]
+        assert chunk.columns.equals(new_index)
+
+    def test_groupby_transform(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        s = frame["A"]
+        grouper = s.index.get_level_values(0)
+
+        grouped = s.groupby(grouper, group_keys=False)
+
+        applied = grouped.apply(lambda x: x * 2)
+        expected = grouped.transform(lambda x: x * 2)
+        result = applied.reindex(expected.index)
+        tm.assert_series_equal(result, expected, check_names=False)
+
+    def test_groupby_corner(self):
+        midx = MultiIndex(
+            levels=[["foo"], ["bar"], ["baz"]],
+            codes=[[0], [0], [0]],
+            names=["one", "two", "three"],
+        )
+        df = DataFrame(
+            [np.random.default_rng(2).random(4)],
+            columns=["a", "b", "c", "d"],
+            index=midx,
+        )
+        # should work
+        df.groupby(level="three")
+
+    def test_setitem_with_expansion_multiindex_columns(
+        self, multiindex_year_month_day_dataframe_random_data
+    ):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        df = ymd[:5].T
+        df[2000, 1, 10] = df[2000, 1, 7]
+        assert isinstance(df.columns, MultiIndex)
+        assert (df[2000, 1, 10] == df[2000, 1, 7]).all()
+
+    def test_alignment(self):
+        x = Series(
+            data=[1, 2, 3], index=MultiIndex.from_tuples([("A", 1), ("A", 2), ("B", 3)])
+        )
+
+        y = Series(
+            data=[4, 5, 6], index=MultiIndex.from_tuples([("Z", 1), ("Z", 2), ("B", 3)])
+        )
+
+        res = x - y
+        exp_index = x.index.union(y.index)
+        exp = x.reindex(exp_index) - y.reindex(exp_index)
+        tm.assert_series_equal(res, exp)
+
+        # hit non-monotonic code path
+        res = x[::-1] - y[::-1]
+        exp_index = x.index.union(y.index)
+        exp = x.reindex(exp_index) - y.reindex(exp_index)
+        tm.assert_series_equal(res, exp)
+
+    def test_groupby_multilevel(self, multiindex_year_month_day_dataframe_random_data):
+        ymd = multiindex_year_month_day_dataframe_random_data
+
+        result = ymd.groupby(level=[0, 1]).mean()
+
+        k1 = ymd.index.get_level_values(0)
+        k2 = ymd.index.get_level_values(1)
+
+        expected = ymd.groupby([k1, k2]).mean()
+
+        tm.assert_frame_equal(result, expected)
+        assert result.index.names == ymd.index.names[:2]
+
+        result2 = ymd.groupby(level=ymd.index.names[:2]).mean()
+        tm.assert_frame_equal(result, result2)
+
+    def test_multilevel_consolidate(self):
+        index = MultiIndex.from_tuples(
+            [("foo", "one"), ("foo", "two"), ("bar", "one"), ("bar", "two")]
+        )
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((4, 4)), index=index, columns=index
+        )
+        df["Totals", ""] = df.sum(axis=1)
+        df = df._consolidate()
+
+    def test_level_with_tuples(self):
+        index = MultiIndex(
+            levels=[[("foo", "bar", 0), ("foo", "baz", 0), ("foo", "qux", 0)], [0, 1]],
+            codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+        )
+
+        series = Series(np.random.default_rng(2).standard_normal(6), index=index)
+        frame = DataFrame(np.random.default_rng(2).standard_normal((6, 4)), index=index)
+
+        result = series[("foo", "bar", 0)]
+        result2 = series.loc[("foo", "bar", 0)]
+        expected = series[:2]
+        expected.index = expected.index.droplevel(0)
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
+
+        with pytest.raises(KeyError, match=r"^\(\('foo', 'bar', 0\), 2\)$"):
+            series[("foo", "bar", 0), 2]
+
+        result = frame.loc[("foo", "bar", 0)]
+        result2 = frame.xs(("foo", "bar", 0))
+        expected = frame[:2]
+        expected.index = expected.index.droplevel(0)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+        index = MultiIndex(
+            levels=[[("foo", "bar"), ("foo", "baz"), ("foo", "qux")], [0, 1]],
+            codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]],
+        )
+
+        series = Series(np.random.default_rng(2).standard_normal(6), index=index)
+        frame = DataFrame(np.random.default_rng(2).standard_normal((6, 4)), index=index)
+
+        result = series[("foo", "bar")]
+        result2 = series.loc[("foo", "bar")]
+        expected = series[:2]
+        expected.index = expected.index.droplevel(0)
+        tm.assert_series_equal(result, expected)
+        tm.assert_series_equal(result2, expected)
+
+        result = frame.loc[("foo", "bar")]
+        result2 = frame.xs(("foo", "bar"))
+        expected = frame[:2]
+        expected.index = expected.index.droplevel(0)
+        tm.assert_frame_equal(result, expected)
+        tm.assert_frame_equal(result2, expected)
+
+    def test_reindex_level_partial_selection(self, multiindex_dataframe_random_data):
+        frame = multiindex_dataframe_random_data
+
+        result = frame.reindex(["foo", "qux"], level=0)
+        expected = frame.iloc[[0, 1, 2, 7, 8, 9]]
+        tm.assert_frame_equal(result, expected)
+
+        result = frame.T.reindex(["foo", "qux"], axis=1, level=0)
+        tm.assert_frame_equal(result, expected.T)
+
+        result = frame.loc[["foo", "qux"]]
+        tm.assert_frame_equal(result, expected)
+
+        result = frame["A"].loc[["foo", "qux"]]
+        tm.assert_series_equal(result, expected["A"])
+
+        result = frame.T.loc[:, ["foo", "qux"]]
+        tm.assert_frame_equal(result, expected.T)
+
+    @pytest.mark.parametrize("d", [4, "d"])
+    def test_empty_frame_groupby_dtypes_consistency(self, d):
+        # GH 20888
+        group_keys = ["a", "b", "c"]
+        df = DataFrame({"a": [1], "b": [2], "c": [3], "d": [d]})
+
+        g = df[df.a == 2].groupby(group_keys)
+        result = g.first().index
+        expected = MultiIndex(
+            levels=[[1], [2], [3]], codes=[[], [], []], names=["a", "b", "c"]
+        )
+
+        tm.assert_index_equal(result, expected)
+
+    def test_duplicate_groupby_issues(self):
+        idx_tp = [
+            ("600809", "20061231"),
+            ("600809", "20070331"),
+            ("600809", "20070630"),
+            ("600809", "20070331"),
+        ]
+        dt = ["demo", "demo", "demo", "demo"]
+
+        idx = MultiIndex.from_tuples(idx_tp, names=["STK_ID", "RPT_Date"])
+        s = Series(dt, index=idx)
+
+        result = s.groupby(s.index).first()
+        assert len(result) == 3
+
+    def test_subsets_multiindex_dtype(self):
+        # GH 20757
+        data = [["x", 1]]
+        columns = [("a", "b", np.nan), ("a", "c", 0.0)]
+        df = DataFrame(data, columns=MultiIndex.from_tuples(columns))
+        expected = df.dtypes.a.b
+        result = df.a.b.dtypes
+        tm.assert_series_equal(result, expected)
+
+    def test_datetime_object_multiindex(self):
+        data_dic = {
+            (0, datetime.date(2018, 3, 3)): {"A": 1, "B": 10},
+            (0, datetime.date(2018, 3, 4)): {"A": 2, "B": 11},
+            (1, datetime.date(2018, 3, 3)): {"A": 3, "B": 12},
+            (1, datetime.date(2018, 3, 4)): {"A": 4, "B": 13},
+        }
+        result = DataFrame.from_dict(data_dic, orient="index")
+        data = {"A": [1, 2, 3, 4], "B": [10, 11, 12, 13]}
+        index = [
+            [0, 0, 1, 1],
+            [
+                datetime.date(2018, 3, 3),
+                datetime.date(2018, 3, 4),
+                datetime.date(2018, 3, 3),
+                datetime.date(2018, 3, 4),
+            ],
+        ]
+        expected = DataFrame(data=data, index=index)
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_multiindex_with_na(self):
+        df = DataFrame(
+            [
+                ["A", np.nan, 1.23, 4.56],
+                ["A", "G", 1.23, 4.56],
+                ["A", "D", 9.87, 10.54],
+            ],
+            columns=["pivot_0", "pivot_1", "col_1", "col_2"],
+        ).set_index(["pivot_0", "pivot_1"])
+
+        df.at[("A", "F"), "col_2"] = 0.0
+
+        expected = DataFrame(
+            [
+                ["A", np.nan, 1.23, 4.56],
+                ["A", "G", 1.23, 4.56],
+                ["A", "D", 9.87, 10.54],
+                ["A", "F", np.nan, 0.0],
+            ],
+            columns=["pivot_0", "pivot_1", "col_1", "col_2"],
+        ).set_index(["pivot_0", "pivot_1"])
+
+        tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize("na", [None, np.nan])
+    def test_multiindex_insert_level_with_na(self, na):
+        # GH 59003
+        df = DataFrame([0], columns=[["A"], ["B"]])
+        df[na, "B"] = 1
+        tm.assert_frame_equal(df[na], DataFrame([1], columns=["B"]))
+
+    def test_multiindex_dt_with_nan(self):
+        # GH#60388
+        df = DataFrame(
+            [
+                [1, np.nan, 5, np.nan],
+                [2, np.nan, 6, np.nan],
+                [np.nan, 3, np.nan, 7],
+                [np.nan, 4, np.nan, 8],
+            ],
+            index=Series(["a", "b", "c", "d"], dtype=object, name="sub"),
+            columns=MultiIndex.from_product(
+                [
+                    ["value1", "value2"],
+                    [datetime.datetime(2024, 11, 1), datetime.datetime(2024, 11, 2)],
+                ],
+                names=[None, "Date"],
+            ),
+        )
+        df = df.reset_index()
+        result = df[df.columns[0]]
+        expected = Series(["a", "b", "c", "d"], name=("sub", np.nan))
+        tm.assert_series_equal(result, expected)
+
+
+class TestSorted:
+    """everything you wanted to test about sorting"""
+
+    def test_sort_non_lexsorted(self):
+        # degenerate case where we sort but don't
+        # have a satisfying result :<
+        # GH 15797
+        idx = MultiIndex(
+            [["A", "B", "C"], ["c", "b", "a"]], [[0, 1, 2, 0, 1, 2], [0, 2, 1, 1, 0, 2]]
+        )
+
+        df = DataFrame({"col": range(len(idx))}, index=idx, dtype="int64")
+        assert df.index.is_monotonic_increasing is False
+
+        sorted = df.sort_index()
+        assert sorted.index.is_monotonic_increasing is True
+
+        expected = DataFrame(
+            {"col": [1, 4, 5, 2]},
+            index=MultiIndex.from_tuples(
+                [("B", "a"), ("B", "c"), ("C", "a"), ("C", "b")]
+            ),
+            dtype="int64",
+        )
+        result = sorted.loc[pd.IndexSlice["B":"C", "a":"c"], :]
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/test_nanops.py b/pandas/pandas/tests/test_nanops.py
new file mode 100644
index 0000000000000..e7ed8e855a762
--- /dev/null
+++ b/pandas/pandas/tests/test_nanops.py
@@ -0,0 +1,1276 @@
+from functools import partial
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import is_integer_dtype
+
+import pandas as pd
+from pandas import (
+    Series,
+    isna,
+)
+import pandas._testing as tm
+from pandas.core import nanops
+
+use_bn = nanops._USE_BOTTLENECK
+
+
+@pytest.fixture
+def disable_bottleneck(monkeypatch):
+    with monkeypatch.context() as m:
+        m.setattr(nanops, "_USE_BOTTLENECK", False)
+        yield
+
+
+@pytest.fixture
+def arr_shape():
+    return 11, 7
+
+
+@pytest.fixture
+def arr_float(arr_shape):
+    return np.random.default_rng(2).standard_normal(arr_shape)
+
+
+@pytest.fixture
+def arr_complex(arr_float):
+    return arr_float + arr_float * 1j
+
+
+@pytest.fixture
+def arr_int(arr_shape):
+    return np.random.default_rng(2).integers(-10, 10, arr_shape)
+
+
+@pytest.fixture
+def arr_bool(arr_shape):
+    return np.random.default_rng(2).integers(0, 2, arr_shape) == 0
+
+
+@pytest.fixture
+def arr_str(arr_float):
+    return np.abs(arr_float).astype("S")
+
+
+@pytest.fixture
+def arr_utf(arr_float):
+    return np.abs(arr_float).astype("U")
+
+
+@pytest.fixture
+def arr_date(arr_shape):
+    return np.random.default_rng(2).integers(0, 20000, arr_shape).astype("M8[ns]")
+
+
+@pytest.fixture
+def arr_tdelta(arr_shape):
+    return np.random.default_rng(2).integers(0, 20000, arr_shape).astype("m8[ns]")
+
+
+@pytest.fixture
+def arr_nan(arr_shape):
+    return np.tile(np.nan, arr_shape)
+
+
+@pytest.fixture
+def arr_float_nan(arr_float, arr_nan):
+    return np.vstack([arr_float, arr_nan])
+
+
+@pytest.fixture
+def arr_nan_float1(arr_nan, arr_float):
+    return np.vstack([arr_nan, arr_float])
+
+
+@pytest.fixture
+def arr_nan_nan(arr_nan):
+    return np.vstack([arr_nan, arr_nan])
+
+
+@pytest.fixture
+def arr_inf(arr_float):
+    return arr_float * np.inf
+
+
+@pytest.fixture
+def arr_float_inf(arr_float, arr_inf):
+    return np.vstack([arr_float, arr_inf])
+
+
+@pytest.fixture
+def arr_nan_inf(arr_nan, arr_inf):
+    return np.vstack([arr_nan, arr_inf])
+
+
+@pytest.fixture
+def arr_float_nan_inf(arr_float, arr_nan, arr_inf):
+    return np.vstack([arr_float, arr_nan, arr_inf])
+
+
+@pytest.fixture
+def arr_nan_nan_inf(arr_nan, arr_inf):
+    return np.vstack([arr_nan, arr_nan, arr_inf])
+
+
+@pytest.fixture
+def arr_obj(
+    arr_float, arr_int, arr_bool, arr_complex, arr_str, arr_utf, arr_date, arr_tdelta
+):
+    return np.vstack(
+        [
+            arr_float.astype("O"),
+            arr_int.astype("O"),
+            arr_bool.astype("O"),
+            arr_complex.astype("O"),
+            arr_str.astype("O"),
+            arr_utf.astype("O"),
+            arr_date.astype("O"),
+            arr_tdelta.astype("O"),
+        ]
+    )
+
+
+@pytest.fixture
+def arr_nan_nanj(arr_nan):
+    with np.errstate(invalid="ignore"):
+        return arr_nan + arr_nan * 1j
+
+
+@pytest.fixture
+def arr_complex_nan(arr_complex, arr_nan_nanj):
+    with np.errstate(invalid="ignore"):
+        return np.vstack([arr_complex, arr_nan_nanj])
+
+
+@pytest.fixture
+def arr_nan_infj(arr_inf):
+    with np.errstate(invalid="ignore"):
+        return arr_inf * 1j
+
+
+@pytest.fixture
+def arr_complex_nan_infj(arr_complex, arr_nan_infj):
+    with np.errstate(invalid="ignore"):
+        return np.vstack([arr_complex, arr_nan_infj])
+
+
+@pytest.fixture
+def arr_float_1d(arr_float):
+    return arr_float[:, 0]
+
+
+@pytest.fixture
+def arr_nan_1d(arr_nan):
+    return arr_nan[:, 0]
+
+
+@pytest.fixture
+def arr_float_nan_1d(arr_float_nan):
+    return arr_float_nan[:, 0]
+
+
+@pytest.fixture
+def arr_float1_nan_1d(arr_float1_nan):
+    return arr_float1_nan[:, 0]
+
+
+@pytest.fixture
+def arr_nan_float1_1d(arr_nan_float1):
+    return arr_nan_float1[:, 0]
+
+
+class TestnanopsDataFrame:
+    def setup_method(self):
+        nanops._USE_BOTTLENECK = False
+
+        arr_shape = (11, 7)
+
+        self.arr_float = np.random.default_rng(2).standard_normal(arr_shape)
+        self.arr_float1 = np.random.default_rng(2).standard_normal(arr_shape)
+        self.arr_complex = self.arr_float + self.arr_float1 * 1j
+        self.arr_int = np.random.default_rng(2).integers(-10, 10, arr_shape)
+        self.arr_bool = np.random.default_rng(2).integers(0, 2, arr_shape) == 0
+        self.arr_str = np.abs(self.arr_float).astype("S")
+        self.arr_utf = np.abs(self.arr_float).astype("U")
+        self.arr_date = (
+            np.random.default_rng(2).integers(0, 20000, arr_shape).astype("M8[ns]")
+        )
+        self.arr_tdelta = (
+            np.random.default_rng(2).integers(0, 20000, arr_shape).astype("m8[ns]")
+        )
+
+        self.arr_nan = np.tile(np.nan, arr_shape)
+        self.arr_float_nan = np.vstack([self.arr_float, self.arr_nan])
+        self.arr_float1_nan = np.vstack([self.arr_float1, self.arr_nan])
+        self.arr_nan_float1 = np.vstack([self.arr_nan, self.arr_float1])
+        self.arr_nan_nan = np.vstack([self.arr_nan, self.arr_nan])
+
+        self.arr_inf = self.arr_float * np.inf
+        self.arr_float_inf = np.vstack([self.arr_float, self.arr_inf])
+
+        self.arr_nan_inf = np.vstack([self.arr_nan, self.arr_inf])
+        self.arr_float_nan_inf = np.vstack([self.arr_float, self.arr_nan, self.arr_inf])
+        self.arr_nan_nan_inf = np.vstack([self.arr_nan, self.arr_nan, self.arr_inf])
+        self.arr_obj = np.vstack(
+            [
+                self.arr_float.astype("O"),
+                self.arr_int.astype("O"),
+                self.arr_bool.astype("O"),
+                self.arr_complex.astype("O"),
+                self.arr_str.astype("O"),
+                self.arr_utf.astype("O"),
+                self.arr_date.astype("O"),
+                self.arr_tdelta.astype("O"),
+            ]
+        )
+
+        with np.errstate(invalid="ignore"):
+            self.arr_nan_nanj = self.arr_nan + self.arr_nan * 1j
+            self.arr_complex_nan = np.vstack([self.arr_complex, self.arr_nan_nanj])
+
+            self.arr_nan_infj = self.arr_inf * 1j
+            self.arr_complex_nan_infj = np.vstack([self.arr_complex, self.arr_nan_infj])
+
+        self.arr_float_2d = self.arr_float
+        self.arr_float1_2d = self.arr_float1
+
+        self.arr_nan_2d = self.arr_nan
+        self.arr_float_nan_2d = self.arr_float_nan
+        self.arr_float1_nan_2d = self.arr_float1_nan
+        self.arr_nan_float1_2d = self.arr_nan_float1
+
+        self.arr_float_1d = self.arr_float[:, 0]
+        self.arr_float1_1d = self.arr_float1[:, 0]
+
+        self.arr_nan_1d = self.arr_nan[:, 0]
+        self.arr_float_nan_1d = self.arr_float_nan[:, 0]
+        self.arr_float1_nan_1d = self.arr_float1_nan[:, 0]
+        self.arr_nan_float1_1d = self.arr_nan_float1[:, 0]
+
+    def teardown_method(self):
+        nanops._USE_BOTTLENECK = use_bn
+
+    def check_results(self, targ, res, axis, check_dtype=True):
+        res = getattr(res, "asm8", res)
+
+        if (
+            axis != 0
+            and hasattr(targ, "shape")
+            and targ.ndim
+            and targ.shape != res.shape
+        ):
+            res = np.split(res, [targ.shape[0]], axis=0)[0]
+
+        try:
+            tm.assert_almost_equal(targ, res, check_dtype=check_dtype)
+        except AssertionError:
+            # handle timedelta dtypes
+            if hasattr(targ, "dtype") and targ.dtype == "m8[ns]":
+                raise
+
+            # There are sometimes rounding errors with
+            # complex and object dtypes.
+            # If it isn't one of those, re-raise the error.
+            if not hasattr(res, "dtype") or res.dtype.kind not in ["c", "O"]:
+                raise
+            # convert object dtypes to something that can be split into
+            # real and imaginary parts
+            if res.dtype.kind == "O":
+                if targ.dtype.kind != "O":
+                    res = res.astype(targ.dtype)
+                else:
+                    cast_dtype = "c16" if hasattr(np, "complex128") else "f8"
+                    res = res.astype(cast_dtype)
+                    targ = targ.astype(cast_dtype)
+            # there should never be a case where numpy returns an object
+            # but nanops doesn't, so make that an exception
+            elif targ.dtype.kind == "O":
+                raise
+            tm.assert_almost_equal(np.real(targ), np.real(res), check_dtype=check_dtype)
+            tm.assert_almost_equal(np.imag(targ), np.imag(res), check_dtype=check_dtype)
+
+    def check_fun_data(
+        self,
+        testfunc,
+        targfunc,
+        testar,
+        testarval,
+        targarval,
+        skipna,
+        check_dtype=True,
+        empty_targfunc=None,
+        **kwargs,
+    ):
+        for axis in list(range(targarval.ndim)) + [None]:
+            targartempval = targarval if skipna else testarval
+            if skipna and empty_targfunc and isna(targartempval).all():
+                targ = empty_targfunc(targartempval, axis=axis, **kwargs)
+            else:
+                targ = targfunc(targartempval, axis=axis, **kwargs)
+
+            if targartempval.dtype == object and (
+                targfunc is np.any or targfunc is np.all
+            ):
+                # GH#12863 the numpy functions will retain e.g. floatiness
+                if isinstance(targ, np.ndarray):
+                    targ = targ.astype(bool)
+                else:
+                    targ = bool(targ)
+
+            if testfunc.__name__ in ["nanargmax", "nanargmin"] and (
+                testar.startswith("arr_nan")
+                or (testar.endswith("nan") and (not skipna or axis == 1))
+            ):
+                with pytest.raises(ValueError, match="Encountered .* NA value"):
+                    testfunc(testarval, axis=axis, skipna=skipna, **kwargs)
+                return
+            res = testfunc(testarval, axis=axis, skipna=skipna, **kwargs)
+
+            if (
+                isinstance(targ, np.complex128)
+                and isinstance(res, float)
+                and np.isnan(targ)
+                and np.isnan(res)
+            ):
+                # GH#18463
+                targ = res
+
+            self.check_results(targ, res, axis, check_dtype=check_dtype)
+            if skipna:
+                res = testfunc(testarval, axis=axis, **kwargs)
+                self.check_results(targ, res, axis, check_dtype=check_dtype)
+            if axis is None:
+                res = testfunc(testarval, skipna=skipna, **kwargs)
+                self.check_results(targ, res, axis, check_dtype=check_dtype)
+            if skipna and axis is None:
+                res = testfunc(testarval, **kwargs)
+                self.check_results(targ, res, axis, check_dtype=check_dtype)
+
+        if testarval.ndim <= 1:
+            return
+
+        # Recurse on lower-dimension
+        testarval2 = np.take(testarval, 0, axis=-1)
+        targarval2 = np.take(targarval, 0, axis=-1)
+        self.check_fun_data(
+            testfunc,
+            targfunc,
+            testar,
+            testarval2,
+            targarval2,
+            skipna=skipna,
+            check_dtype=check_dtype,
+            empty_targfunc=empty_targfunc,
+            **kwargs,
+        )
+
+    def check_fun(
+        self, testfunc, targfunc, testar, skipna, empty_targfunc=None, **kwargs
+    ):
+        targar = testar
+        if testar.endswith("_nan") and hasattr(self, testar[:-4]):
+            targar = testar[:-4]
+
+        testarval = getattr(self, testar)
+        targarval = getattr(self, targar)
+        self.check_fun_data(
+            testfunc,
+            targfunc,
+            testar,
+            testarval,
+            targarval,
+            skipna=skipna,
+            empty_targfunc=empty_targfunc,
+            **kwargs,
+        )
+
+    def check_funs(
+        self,
+        testfunc,
+        targfunc,
+        skipna,
+        allow_complex=True,
+        allow_all_nan=True,
+        allow_date=True,
+        allow_tdelta=True,
+        allow_obj=True,
+        **kwargs,
+    ):
+        self.check_fun(testfunc, targfunc, "arr_float", skipna, **kwargs)
+        self.check_fun(testfunc, targfunc, "arr_float_nan", skipna, **kwargs)
+        self.check_fun(testfunc, targfunc, "arr_int", skipna, **kwargs)
+        self.check_fun(testfunc, targfunc, "arr_bool", skipna, **kwargs)
+        objs = [
+            self.arr_float.astype("O"),
+            self.arr_int.astype("O"),
+            self.arr_bool.astype("O"),
+        ]
+
+        if allow_all_nan:
+            self.check_fun(testfunc, targfunc, "arr_nan", skipna, **kwargs)
+
+        if allow_complex:
+            self.check_fun(testfunc, targfunc, "arr_complex", skipna, **kwargs)
+            self.check_fun(testfunc, targfunc, "arr_complex_nan", skipna, **kwargs)
+            if allow_all_nan:
+                self.check_fun(testfunc, targfunc, "arr_nan_nanj", skipna, **kwargs)
+            objs += [self.arr_complex.astype("O")]
+
+        if allow_date:
+            targfunc(self.arr_date)
+            self.check_fun(testfunc, targfunc, "arr_date", skipna, **kwargs)
+            objs += [self.arr_date.astype("O")]
+
+        if allow_tdelta:
+            try:
+                targfunc(self.arr_tdelta)
+            except TypeError:
+                pass
+            else:
+                self.check_fun(testfunc, targfunc, "arr_tdelta", skipna, **kwargs)
+                objs += [self.arr_tdelta.astype("O")]
+
+        if allow_obj:
+            self.arr_obj = np.vstack(objs)
+            # some nanops handle object dtypes better than their numpy
+            # counterparts, so the numpy functions need to be given something
+            # else
+            if allow_obj == "convert":
+                targfunc = partial(
+                    self._badobj_wrap, func=targfunc, allow_complex=allow_complex
+                )
+            self.check_fun(testfunc, targfunc, "arr_obj", skipna, **kwargs)
+
+    def _badobj_wrap(self, value, func, allow_complex=True, **kwargs):
+        if value.dtype.kind == "O":
+            if allow_complex:
+                value = value.astype("c16")
+            else:
+                value = value.astype("f8")
+        return func(value, **kwargs)
+
+    @pytest.mark.parametrize(
+        "nan_op,np_op", [(nanops.nanany, np.any), (nanops.nanall, np.all)]
+    )
+    def test_nan_funcs(self, nan_op, np_op, skipna):
+        self.check_funs(nan_op, np_op, skipna, allow_all_nan=False, allow_date=False)
+
+    def test_nansum(self, skipna):
+        self.check_funs(
+            nanops.nansum,
+            np.sum,
+            skipna,
+            allow_date=False,
+            check_dtype=False,
+            empty_targfunc=np.nansum,
+        )
+
+    def test_nanmean(self, skipna):
+        self.check_funs(
+            nanops.nanmean, np.mean, skipna, allow_obj=False, allow_date=False
+        )
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_nanmedian(self, skipna):
+        self.check_funs(
+            nanops.nanmedian,
+            np.median,
+            skipna,
+            allow_complex=False,
+            allow_date=False,
+            allow_obj="convert",
+        )
+
+    @pytest.mark.parametrize("ddof", range(3))
+    def test_nanvar(self, ddof, skipna):
+        self.check_funs(
+            nanops.nanvar,
+            np.var,
+            skipna,
+            allow_complex=False,
+            allow_date=False,
+            allow_obj="convert",
+            ddof=ddof,
+        )
+
+    @pytest.mark.parametrize("ddof", range(3))
+    def test_nanstd(self, ddof, skipna):
+        self.check_funs(
+            nanops.nanstd,
+            np.std,
+            skipna,
+            allow_complex=False,
+            allow_date=False,
+            allow_obj="convert",
+            ddof=ddof,
+        )
+
+    @pytest.mark.parametrize("ddof", range(3))
+    def test_nansem(self, ddof, skipna):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        with np.errstate(invalid="ignore"):
+            self.check_funs(
+                nanops.nansem,
+                sp_stats.sem,
+                skipna,
+                allow_complex=False,
+                allow_date=False,
+                allow_tdelta=False,
+                allow_obj="convert",
+                ddof=ddof,
+            )
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    @pytest.mark.parametrize(
+        "nan_op,np_op", [(nanops.nanmin, np.min), (nanops.nanmax, np.max)]
+    )
+    def test_nanops_with_warnings(self, nan_op, np_op, skipna):
+        self.check_funs(nan_op, np_op, skipna, allow_obj=False)
+
+    def _argminmax_wrap(self, value, axis=None, func=None):
+        res = func(value, axis)
+        nans = np.min(value, axis)
+        nullnan = isna(nans)
+        if res.ndim:
+            res[nullnan] = -1
+        elif (hasattr(nullnan, "all") and nullnan.all()) or (
+            not hasattr(nullnan, "all") and nullnan
+        ):
+            res = -1
+        return res
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_nanargmax(self, skipna):
+        func = partial(self._argminmax_wrap, func=np.argmax)
+        self.check_funs(nanops.nanargmax, func, skipna, allow_obj=False)
+
+    @pytest.mark.filterwarnings("ignore::RuntimeWarning")
+    def test_nanargmin(self, skipna):
+        func = partial(self._argminmax_wrap, func=np.argmin)
+        self.check_funs(nanops.nanargmin, func, skipna, allow_obj=False)
+
+    def _skew_kurt_wrap(self, values, axis=None, func=None):
+        if not isinstance(values.dtype.type, np.floating):
+            values = values.astype("f8")
+        result = func(values, axis=axis, bias=False)
+        # fix for handling cases where all elements in an axis are the same
+        if isinstance(result, np.ndarray):
+            result[np.max(values, axis=axis) == np.min(values, axis=axis)] = 0
+            return result
+        elif np.max(values) == np.min(values):
+            return 0.0
+        return result
+
+    def test_nanskew(self, skipna):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        func = partial(self._skew_kurt_wrap, func=sp_stats.skew)
+        with np.errstate(invalid="ignore"):
+            self.check_funs(
+                nanops.nanskew,
+                func,
+                skipna,
+                allow_complex=False,
+                allow_date=False,
+                allow_tdelta=False,
+            )
+
+    def test_nankurt(self, skipna):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        func1 = partial(sp_stats.kurtosis, fisher=True)
+        func = partial(self._skew_kurt_wrap, func=func1)
+        with np.errstate(invalid="ignore"):
+            self.check_funs(
+                nanops.nankurt,
+                func,
+                skipna,
+                allow_complex=False,
+                allow_date=False,
+                allow_tdelta=False,
+            )
+
+    def test_nanprod(self, skipna):
+        self.check_funs(
+            nanops.nanprod,
+            np.prod,
+            skipna,
+            allow_date=False,
+            allow_tdelta=False,
+            empty_targfunc=np.nanprod,
+        )
+
+    def check_nancorr_nancov_2d(self, checkfun, targ0, targ1, **kwargs):
+        res00 = checkfun(self.arr_float_2d, self.arr_float1_2d, **kwargs)
+        res01 = checkfun(
+            self.arr_float_2d,
+            self.arr_float1_2d,
+            min_periods=len(self.arr_float_2d) - 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ0, res00)
+        tm.assert_almost_equal(targ0, res01)
+
+        res10 = checkfun(self.arr_float_nan_2d, self.arr_float1_nan_2d, **kwargs)
+        res11 = checkfun(
+            self.arr_float_nan_2d,
+            self.arr_float1_nan_2d,
+            min_periods=len(self.arr_float_2d) - 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ1, res10)
+        tm.assert_almost_equal(targ1, res11)
+
+        targ2 = np.nan
+        res20 = checkfun(self.arr_nan_2d, self.arr_float1_2d, **kwargs)
+        res21 = checkfun(self.arr_float_2d, self.arr_nan_2d, **kwargs)
+        res22 = checkfun(self.arr_nan_2d, self.arr_nan_2d, **kwargs)
+        res23 = checkfun(self.arr_float_nan_2d, self.arr_nan_float1_2d, **kwargs)
+        res24 = checkfun(
+            self.arr_float_nan_2d,
+            self.arr_nan_float1_2d,
+            min_periods=len(self.arr_float_2d) - 1,
+            **kwargs,
+        )
+        res25 = checkfun(
+            self.arr_float_2d,
+            self.arr_float1_2d,
+            min_periods=len(self.arr_float_2d) + 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ2, res20)
+        tm.assert_almost_equal(targ2, res21)
+        tm.assert_almost_equal(targ2, res22)
+        tm.assert_almost_equal(targ2, res23)
+        tm.assert_almost_equal(targ2, res24)
+        tm.assert_almost_equal(targ2, res25)
+
+    def check_nancorr_nancov_1d(self, checkfun, targ0, targ1, **kwargs):
+        res00 = checkfun(self.arr_float_1d, self.arr_float1_1d, **kwargs)
+        res01 = checkfun(
+            self.arr_float_1d,
+            self.arr_float1_1d,
+            min_periods=len(self.arr_float_1d) - 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ0, res00)
+        tm.assert_almost_equal(targ0, res01)
+
+        res10 = checkfun(self.arr_float_nan_1d, self.arr_float1_nan_1d, **kwargs)
+        res11 = checkfun(
+            self.arr_float_nan_1d,
+            self.arr_float1_nan_1d,
+            min_periods=len(self.arr_float_1d) - 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ1, res10)
+        tm.assert_almost_equal(targ1, res11)
+
+        targ2 = np.nan
+        res20 = checkfun(self.arr_nan_1d, self.arr_float1_1d, **kwargs)
+        res21 = checkfun(self.arr_float_1d, self.arr_nan_1d, **kwargs)
+        res22 = checkfun(self.arr_nan_1d, self.arr_nan_1d, **kwargs)
+        res23 = checkfun(self.arr_float_nan_1d, self.arr_nan_float1_1d, **kwargs)
+        res24 = checkfun(
+            self.arr_float_nan_1d,
+            self.arr_nan_float1_1d,
+            min_periods=len(self.arr_float_1d) - 1,
+            **kwargs,
+        )
+        res25 = checkfun(
+            self.arr_float_1d,
+            self.arr_float1_1d,
+            min_periods=len(self.arr_float_1d) + 1,
+            **kwargs,
+        )
+        tm.assert_almost_equal(targ2, res20)
+        tm.assert_almost_equal(targ2, res21)
+        tm.assert_almost_equal(targ2, res22)
+        tm.assert_almost_equal(targ2, res23)
+        tm.assert_almost_equal(targ2, res24)
+        tm.assert_almost_equal(targ2, res25)
+
+    def test_nancorr(self):
+        targ0 = np.corrcoef(self.arr_float_2d, self.arr_float1_2d)[0, 1]
+        targ1 = np.corrcoef(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0, 1]
+        self.check_nancorr_nancov_2d(nanops.nancorr, targ0, targ1)
+        targ0 = np.corrcoef(self.arr_float_1d, self.arr_float1_1d)[0, 1]
+        targ1 = np.corrcoef(self.arr_float_1d.flat, self.arr_float1_1d.flat)[0, 1]
+        self.check_nancorr_nancov_1d(nanops.nancorr, targ0, targ1, method="pearson")
+
+    def test_nancorr_pearson(self):
+        targ0 = np.corrcoef(self.arr_float_2d, self.arr_float1_2d)[0, 1]
+        targ1 = np.corrcoef(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0, 1]
+        self.check_nancorr_nancov_2d(nanops.nancorr, targ0, targ1, method="pearson")
+        targ0 = np.corrcoef(self.arr_float_1d, self.arr_float1_1d)[0, 1]
+        targ1 = np.corrcoef(self.arr_float_1d.flat, self.arr_float1_1d.flat)[0, 1]
+        self.check_nancorr_nancov_1d(nanops.nancorr, targ0, targ1, method="pearson")
+
+    def test_nancorr_kendall(self):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        targ0 = sp_stats.kendalltau(self.arr_float_2d, self.arr_float1_2d)[0]
+        targ1 = sp_stats.kendalltau(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0]
+        self.check_nancorr_nancov_2d(nanops.nancorr, targ0, targ1, method="kendall")
+        targ0 = sp_stats.kendalltau(self.arr_float_1d, self.arr_float1_1d)[0]
+        targ1 = sp_stats.kendalltau(self.arr_float_1d.flat, self.arr_float1_1d.flat)[0]
+        self.check_nancorr_nancov_1d(nanops.nancorr, targ0, targ1, method="kendall")
+
+    def test_nancorr_spearman(self):
+        sp_stats = pytest.importorskip("scipy.stats")
+
+        targ0 = sp_stats.spearmanr(self.arr_float_2d, self.arr_float1_2d)[0]
+        targ1 = sp_stats.spearmanr(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0]
+        self.check_nancorr_nancov_2d(nanops.nancorr, targ0, targ1, method="spearman")
+        targ0 = sp_stats.spearmanr(self.arr_float_1d, self.arr_float1_1d)[0]
+        targ1 = sp_stats.spearmanr(self.arr_float_1d.flat, self.arr_float1_1d.flat)[0]
+        self.check_nancorr_nancov_1d(nanops.nancorr, targ0, targ1, method="spearman")
+
+    def test_invalid_method(self):
+        pytest.importorskip("scipy")
+        targ0 = np.corrcoef(self.arr_float_2d, self.arr_float1_2d)[0, 1]
+        targ1 = np.corrcoef(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0, 1]
+        msg = "Unknown method 'foo', expected one of 'kendall', 'spearman'"
+        with pytest.raises(ValueError, match=msg):
+            self.check_nancorr_nancov_1d(nanops.nancorr, targ0, targ1, method="foo")
+
+    def test_nancov(self):
+        targ0 = np.cov(self.arr_float_2d, self.arr_float1_2d)[0, 1]
+        targ1 = np.cov(self.arr_float_2d.flat, self.arr_float1_2d.flat)[0, 1]
+        self.check_nancorr_nancov_2d(nanops.nancov, targ0, targ1)
+        targ0 = np.cov(self.arr_float_1d, self.arr_float1_1d)[0, 1]
+        targ1 = np.cov(self.arr_float_1d.flat, self.arr_float1_1d.flat)[0, 1]
+        self.check_nancorr_nancov_1d(nanops.nancov, targ0, targ1)
+
+
+@pytest.mark.parametrize(
+    "arr, correct",
+    [
+        ("arr_complex", False),
+        ("arr_int", False),
+        ("arr_bool", False),
+        ("arr_str", False),
+        ("arr_utf", False),
+        ("arr_complex_nan", False),
+        ("arr_nan_nanj", False),
+        ("arr_nan_infj", True),
+        ("arr_complex_nan_infj", True),
+    ],
+)
+def test_has_infs_non_float(request, arr, correct, disable_bottleneck):
+    val = request.getfixturevalue(arr)
+    while getattr(val, "ndim", True):
+        res0 = nanops._has_infs(val)
+        if correct:
+            assert res0
+        else:
+            assert not res0
+
+        if not hasattr(val, "ndim"):
+            break
+
+        # Reduce dimension for next step in the loop
+        val = np.take(val, 0, axis=-1)
+
+
+@pytest.mark.parametrize(
+    "arr, correct",
+    [
+        ("arr_float", False),
+        ("arr_nan", False),
+        ("arr_float_nan", False),
+        ("arr_nan_nan", False),
+        ("arr_float_inf", True),
+        ("arr_inf", True),
+        ("arr_nan_inf", True),
+        ("arr_float_nan_inf", True),
+        ("arr_nan_nan_inf", True),
+    ],
+)
+@pytest.mark.parametrize("astype", [None, "f4", "f2"])
+def test_has_infs_floats(request, arr, correct, astype, disable_bottleneck):
+    val = request.getfixturevalue(arr)
+    if astype is not None:
+        val = val.astype(astype)
+    while getattr(val, "ndim", True):
+        res0 = nanops._has_infs(val)
+        if correct:
+            assert res0
+        else:
+            assert not res0
+
+        if not hasattr(val, "ndim"):
+            break
+
+        # Reduce dimension for next step in the loop
+        val = np.take(val, 0, axis=-1)
+
+
+@pytest.mark.parametrize(
+    "fixture", ["arr_float", "arr_complex", "arr_int", "arr_bool", "arr_str", "arr_utf"]
+)
+def test_bn_ok_dtype(fixture, request, disable_bottleneck):
+    obj = request.getfixturevalue(fixture)
+    assert nanops._bn_ok_dtype(obj.dtype, "test")
+
+
+@pytest.mark.parametrize(
+    "fixture",
+    [
+        "arr_date",
+        "arr_tdelta",
+        "arr_obj",
+    ],
+)
+def test_bn_not_ok_dtype(fixture, request, disable_bottleneck):
+    obj = request.getfixturevalue(fixture)
+    assert not nanops._bn_ok_dtype(obj.dtype, "test")
+
+
+class TestEnsureNumeric:
+    def test_numeric_values(self):
+        # Test integer
+        assert nanops._ensure_numeric(1) == 1
+
+        # Test float
+        assert nanops._ensure_numeric(1.1) == 1.1
+
+        # Test complex
+        assert nanops._ensure_numeric(1 + 2j) == 1 + 2j
+
+    def test_ndarray(self):
+        # Test numeric ndarray
+        values = np.array([1, 2, 3])
+        assert np.allclose(nanops._ensure_numeric(values), values)
+
+        # Test object ndarray
+        o_values = values.astype(object)
+        assert np.allclose(nanops._ensure_numeric(o_values), values)
+
+        # Test convertible string ndarray
+        s_values = np.array(["1", "2", "3"], dtype=object)
+        msg = r"Could not convert \['1' '2' '3'\] to numeric"
+        with pytest.raises(TypeError, match=msg):
+            nanops._ensure_numeric(s_values)
+
+        # Test non-convertible string ndarray
+        s_values = np.array(["foo", "bar", "baz"], dtype=object)
+        msg = r"Could not convert .* to numeric"
+        with pytest.raises(TypeError, match=msg):
+            nanops._ensure_numeric(s_values)
+
+    def test_convertable_values(self):
+        with pytest.raises(TypeError, match="Could not convert string '1' to numeric"):
+            nanops._ensure_numeric("1")
+        with pytest.raises(
+            TypeError, match="Could not convert string '1.1' to numeric"
+        ):
+            nanops._ensure_numeric("1.1")
+        with pytest.raises(
+            TypeError, match=r"Could not convert string '1\+1j' to numeric"
+        ):
+            nanops._ensure_numeric("1+1j")
+
+    def test_non_convertable_values(self):
+        msg = "Could not convert string 'foo' to numeric"
+        with pytest.raises(TypeError, match=msg):
+            nanops._ensure_numeric("foo")
+
+        # with the wrong type, python raises TypeError for us
+        msg = "argument must be a string or a number"
+        with pytest.raises(TypeError, match=msg):
+            nanops._ensure_numeric({})
+        with pytest.raises(TypeError, match=msg):
+            nanops._ensure_numeric([])
+
+
+class TestNanvarFixedValues:
+    # xref GH10242
+    # Samples from a normal distribution.
+    @pytest.fixture
+    def variance(self):
+        return 3.0
+
+    @pytest.fixture
+    def samples(self, variance):
+        return self.prng.normal(scale=variance**0.5, size=100000)
+
+    def test_nanvar_all_finite(self, samples, variance):
+        actual_variance = nanops.nanvar(samples)
+        tm.assert_almost_equal(actual_variance, variance, rtol=1e-2)
+
+    def test_nanvar_nans(self, samples, variance):
+        samples_test = np.nan * np.ones(2 * samples.shape[0])
+        samples_test[::2] = samples
+
+        actual_variance = nanops.nanvar(samples_test, skipna=True)
+        tm.assert_almost_equal(actual_variance, variance, rtol=1e-2)
+
+        actual_variance = nanops.nanvar(samples_test, skipna=False)
+        tm.assert_almost_equal(actual_variance, np.nan, rtol=1e-2)
+
+    def test_nanstd_nans(self, samples, variance):
+        samples_test = np.nan * np.ones(2 * samples.shape[0])
+        samples_test[::2] = samples
+
+        actual_std = nanops.nanstd(samples_test, skipna=True)
+        tm.assert_almost_equal(actual_std, variance**0.5, rtol=1e-2)
+
+        actual_std = nanops.nanvar(samples_test, skipna=False)
+        tm.assert_almost_equal(actual_std, np.nan, rtol=1e-2)
+
+    def test_nanvar_axis(self, samples, variance):
+        # Generate some sample data.
+        samples_unif = self.prng.uniform(size=samples.shape[0])
+        samples = np.vstack([samples, samples_unif])
+
+        actual_variance = nanops.nanvar(samples, axis=1)
+        tm.assert_almost_equal(
+            actual_variance, np.array([variance, 1.0 / 12]), rtol=1e-2
+        )
+
+    def test_nanvar_ddof(self):
+        n = 5
+        samples = self.prng.uniform(size=(10000, n + 1))
+        samples[:, -1] = np.nan  # Force use of our own algorithm.
+
+        variance_0 = nanops.nanvar(samples, axis=1, skipna=True, ddof=0).mean()
+        variance_1 = nanops.nanvar(samples, axis=1, skipna=True, ddof=1).mean()
+        variance_2 = nanops.nanvar(samples, axis=1, skipna=True, ddof=2).mean()
+
+        # The unbiased estimate.
+        var = 1.0 / 12
+        tm.assert_almost_equal(variance_1, var, rtol=1e-2)
+
+        # The underestimated variance.
+        tm.assert_almost_equal(variance_0, (n - 1.0) / n * var, rtol=1e-2)
+
+        # The overestimated variance.
+        tm.assert_almost_equal(variance_2, (n - 1.0) / (n - 2.0) * var, rtol=1e-2)
+
+    @pytest.mark.parametrize("axis", range(2))
+    @pytest.mark.parametrize("ddof", range(3))
+    def test_ground_truth(self, axis, ddof):
+        # Test against values that were precomputed with Numpy.
+        samples = np.empty((4, 4))
+        samples[:3, :3] = np.array(
+            [
+                [0.97303362, 0.21869576, 0.55560287],
+                [0.72980153, 0.03109364, 0.99155171],
+                [0.09317602, 0.60078248, 0.15871292],
+            ]
+        )
+        samples[3] = samples[:, 3] = np.nan
+
+        # Actual variances along axis=0, 1 for ddof=0, 1, 2
+        variance = np.array(
+            [
+                [
+                    [0.13762259, 0.05619224, 0.11568816],
+                    [0.20643388, 0.08428837, 0.17353224],
+                    [0.41286776, 0.16857673, 0.34706449],
+                ],
+                [
+                    [0.09519783, 0.16435395, 0.05082054],
+                    [0.14279674, 0.24653093, 0.07623082],
+                    [0.28559348, 0.49306186, 0.15246163],
+                ],
+            ]
+        )
+
+        # Test nanvar.
+        var = nanops.nanvar(samples, skipna=True, axis=axis, ddof=ddof)
+        tm.assert_almost_equal(var[:3], variance[axis, ddof])
+        assert np.isnan(var[3])
+
+        # Test nanstd.
+        std = nanops.nanstd(samples, skipna=True, axis=axis, ddof=ddof)
+        tm.assert_almost_equal(std[:3], variance[axis, ddof] ** 0.5)
+        assert np.isnan(std[3])
+
+    @pytest.mark.parametrize("ddof", range(3))
+    def test_nanstd_roundoff(self, ddof):
+        # Regression test for GH 10242 (test data taken from GH 10489). Ensure
+        # that variance is stable.
+        data = Series(766897346 * np.ones(10))
+        result = data.std(ddof=ddof)
+        assert result == 0.0
+
+    @property
+    def prng(self):
+        return np.random.default_rng(2)
+
+
+class TestNanskewFixedValues:
+    # xref GH 11974
+    # Test data + skewness value (computed with scipy.stats.skew)
+    @pytest.fixture
+    def samples(self):
+        return np.sin(np.linspace(0, 1, 200))
+
+    @pytest.fixture
+    def actual_skew(self):
+        return -0.1875895205961754
+
+    @pytest.mark.parametrize("val", [3075.2, 3075.3, 3075.5])
+    def test_constant_series(self, val):
+        # xref GH 11974
+        data = val * np.ones(300)
+        skew = nanops.nanskew(data)
+        assert skew == 0.0
+
+    def test_all_finite(self):
+        alpha, beta = 0.3, 0.1
+        left_tailed = self.prng.beta(alpha, beta, size=100)
+        assert nanops.nanskew(left_tailed) < 0
+
+        alpha, beta = 0.1, 0.3
+        right_tailed = self.prng.beta(alpha, beta, size=100)
+        assert nanops.nanskew(right_tailed) > 0
+
+    def test_ground_truth(self, samples, actual_skew):
+        skew = nanops.nanskew(samples)
+        tm.assert_almost_equal(skew, actual_skew)
+
+    def test_axis(self, samples, actual_skew):
+        samples = np.vstack([samples, np.nan * np.ones(len(samples))])
+        skew = nanops.nanskew(samples, axis=1)
+        tm.assert_almost_equal(skew, np.array([actual_skew, np.nan]))
+
+    def test_nans(self, samples):
+        samples = np.hstack([samples, np.nan])
+        skew = nanops.nanskew(samples, skipna=False)
+        assert np.isnan(skew)
+
+    def test_nans_skipna(self, samples, actual_skew):
+        samples = np.hstack([samples, np.nan])
+        skew = nanops.nanskew(samples, skipna=True)
+        tm.assert_almost_equal(skew, actual_skew)
+
+    @property
+    def prng(self):
+        return np.random.default_rng(2)
+
+
+class TestNankurtFixedValues:
+    # xref GH 11974
+    # Test data + kurtosis value (computed with scipy.stats.kurtosis)
+    @pytest.fixture
+    def samples(self):
+        return np.sin(np.linspace(0, 1, 200))
+
+    @pytest.fixture
+    def actual_kurt(self):
+        return -1.2058303433799713
+
+    @pytest.mark.parametrize("val", [3075.2, 3075.3, 3075.5])
+    def test_constant_series(self, val):
+        # xref GH 11974
+        data = val * np.ones(300)
+        kurt = nanops.nankurt(data)
+        assert kurt == 0.0
+
+    def test_all_finite(self):
+        alpha, beta = 0.3, 0.1
+        left_tailed = self.prng.beta(alpha, beta, size=100)
+        assert nanops.nankurt(left_tailed) < 2
+
+        alpha, beta = 0.1, 0.3
+        right_tailed = self.prng.beta(alpha, beta, size=100)
+        assert nanops.nankurt(right_tailed) < 0
+
+    def test_ground_truth(self, samples, actual_kurt):
+        kurt = nanops.nankurt(samples)
+        tm.assert_almost_equal(kurt, actual_kurt)
+
+    def test_axis(self, samples, actual_kurt):
+        samples = np.vstack([samples, np.nan * np.ones(len(samples))])
+        kurt = nanops.nankurt(samples, axis=1)
+        tm.assert_almost_equal(kurt, np.array([actual_kurt, np.nan]))
+
+    def test_nans(self, samples):
+        samples = np.hstack([samples, np.nan])
+        kurt = nanops.nankurt(samples, skipna=False)
+        assert np.isnan(kurt)
+
+    def test_nans_skipna(self, samples, actual_kurt):
+        samples = np.hstack([samples, np.nan])
+        kurt = nanops.nankurt(samples, skipna=True)
+        tm.assert_almost_equal(kurt, actual_kurt)
+
+    @property
+    def prng(self):
+        return np.random.default_rng(2)
+
+
+class TestDatetime64NaNOps:
+    # Enabling mean changes the behavior of DataFrame.mean
+    # See https://github.com/pandas-dev/pandas/issues/24752
+    def test_nanmean(self, unit):
+        dti = pd.date_range("2016-01-01", periods=3).as_unit(unit)
+        expected = dti[1]
+
+        for obj in [dti, dti._data]:
+            result = nanops.nanmean(obj)
+            assert result == expected
+
+        dti2 = dti.insert(1, pd.NaT)
+
+        for obj in [dti2, dti2._data]:
+            result = nanops.nanmean(obj)
+            assert result == expected
+
+    @pytest.mark.parametrize("constructor", ["M8", "m8"])
+    def test_nanmean_skipna_false(self, constructor, unit):
+        dtype = f"{constructor}[{unit}]"
+        arr = np.arange(12).astype(np.int64).view(dtype).reshape(4, 3)
+
+        arr[-1, -1] = "NaT"
+
+        result = nanops.nanmean(arr, skipna=False)
+        assert np.isnat(result)
+        assert result.dtype == dtype
+
+        result = nanops.nanmean(arr, axis=0, skipna=False)
+        expected = np.array([4, 5, "NaT"], dtype=arr.dtype)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = nanops.nanmean(arr, axis=1, skipna=False)
+        expected = np.array([arr[0, 1], arr[1, 1], arr[2, 1], arr[-1, -1]])
+        tm.assert_numpy_array_equal(result, expected)
+
+
+def test_use_bottleneck():
+    if nanops._BOTTLENECK_INSTALLED:
+        with pd.option_context("use_bottleneck", True):
+            assert pd.get_option("use_bottleneck")
+
+        with pd.option_context("use_bottleneck", False):
+            assert not pd.get_option("use_bottleneck")
+
+
+@pytest.mark.parametrize(
+    "numpy_op, expected",
+    [
+        (np.sum, 10),
+        (np.nansum, 10),
+        (np.mean, 2.5),
+        (np.nanmean, 2.5),
+        (np.median, 2.5),
+        (np.nanmedian, 2.5),
+        (np.min, 1),
+        (np.max, 4),
+        (np.nanmin, 1),
+        (np.nanmax, 4),
+    ],
+)
+def test_numpy_ops(numpy_op, expected):
+    # GH8383
+    result = numpy_op(Series([1, 2, 3, 4]))
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "operation",
+    [
+        nanops.nanany,
+        nanops.nanall,
+        nanops.nansum,
+        nanops.nanmean,
+        nanops.nanmedian,
+        nanops.nanstd,
+        nanops.nanvar,
+        nanops.nansem,
+        nanops.nanargmax,
+        nanops.nanargmin,
+        nanops.nanmax,
+        nanops.nanmin,
+        nanops.nanskew,
+        nanops.nankurt,
+        nanops.nanprod,
+    ],
+)
+def test_nanops_independent_of_mask_param(operation):
+    # GH22764
+    ser = Series([1, 2, np.nan, 3, np.nan, 4])
+    mask = ser.isna()
+    median_expected = operation(ser._values)
+    median_result = operation(ser._values, mask=mask)
+    assert median_expected == median_result
+
+
+@pytest.mark.parametrize("min_count", [-1, 0])
+def test_check_below_min_count_negative_or_zero_min_count(min_count):
+    # GH35227
+    result = nanops.check_below_min_count((21, 37), None, min_count)
+    expected_result = False
+    assert result == expected_result
+
+
+@pytest.mark.parametrize(
+    "mask", [None, np.array([False, False, True]), np.array([True] + 9 * [False])]
+)
+@pytest.mark.parametrize("min_count, expected_result", [(1, False), (101, True)])
+def test_check_below_min_count_positive_min_count(mask, min_count, expected_result):
+    # GH35227
+    shape = (10, 10)
+    result = nanops.check_below_min_count(shape, mask, min_count)
+    assert result == expected_result
+
+
+@td.skip_if_windows
+@td.skip_if_32bit
+@pytest.mark.parametrize("min_count, expected_result", [(1, False), (2812191852, True)])
+def test_check_below_min_count_large_shape(min_count, expected_result):
+    # GH35227 large shape used to show that the issue is fixed
+    shape = (2244367, 1253)
+    result = nanops.check_below_min_count(shape, mask=None, min_count=min_count)
+    assert result == expected_result
+
+
+@pytest.mark.parametrize("func", ["nanmean", "nansum"])
+def test_check_bottleneck_disallow(any_real_numpy_dtype, func):
+    # GH 42878 bottleneck sometimes produces unreliable results for mean and sum
+    assert not nanops._bn_ok_dtype(np.dtype(any_real_numpy_dtype).type, func)
+
+
+@pytest.mark.parametrize("val", [2**55, -(2**55), 20150515061816532])
+def test_nanmean_overflow(disable_bottleneck, val):
+    # GH 10155
+    # In the previous implementation mean can overflow for int dtypes, it
+    # is now consistent with numpy
+
+    ser = Series(val, index=range(500), dtype=np.int64)
+    result = ser.mean()
+    np_result = ser.values.mean()
+    assert result == val
+    assert result == np_result
+    assert result.dtype == np.float64
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        np.int16,
+        np.int32,
+        np.int64,
+        np.float32,
+        np.float64,
+        getattr(np, "float128", None),
+    ],
+)
+@pytest.mark.parametrize("method", ["mean", "std", "var", "skew", "kurt", "min", "max"])
+def test_returned_dtype(disable_bottleneck, dtype, method):
+    if dtype is None:
+        pytest.skip("np.float128 not available")
+
+    ser = Series(range(10), dtype=dtype)
+    result = getattr(ser, method)()
+    if is_integer_dtype(dtype) and method not in ["min", "max"]:
+        assert result.dtype == np.float64
+    else:
+        assert result.dtype == dtype
diff --git a/pandas/pandas/tests/test_optional_dependency.py b/pandas/pandas/tests/test_optional_dependency.py
new file mode 100644
index 0000000000000..9127981d1845d
--- /dev/null
+++ b/pandas/pandas/tests/test_optional_dependency.py
@@ -0,0 +1,100 @@
+import sys
+import types
+
+import pytest
+
+from pandas.compat._optional import (
+    VERSIONS,
+    import_optional_dependency,
+)
+
+import pandas._testing as tm
+
+
+def test_import_optional():
+    match = "Missing .*notapackage.* pip .* conda .* notapackage"
+    with pytest.raises(ImportError, match=match) as exc_info:
+        import_optional_dependency("notapackage")
+    # The original exception should be there as context:
+    assert isinstance(exc_info.value.__context__, ImportError)
+
+    result = import_optional_dependency("notapackage", errors="ignore")
+    assert result is None
+
+
+def test_xlrd_version_fallback():
+    pytest.importorskip("xlrd")
+    import_optional_dependency("xlrd")
+
+
+def test_bad_version(monkeypatch):
+    name = "fakemodule"
+    module = types.ModuleType(name)
+    module.__version__ = "0.9.0"
+    sys.modules[name] = module
+    monkeypatch.setitem(VERSIONS, name, "1.0.0")
+
+    match = "Pandas requires .*1.0.0.* of .fakemodule.*'0.9.0'"
+    with pytest.raises(ImportError, match=match):
+        import_optional_dependency("fakemodule")
+
+    # Test min_version parameter
+    result = import_optional_dependency("fakemodule", min_version="0.8")
+    assert result is module
+
+    with tm.assert_produces_warning(UserWarning, match=match):
+        result = import_optional_dependency("fakemodule", errors="warn")
+    assert result is None
+
+    module.__version__ = "1.0.0"  # exact match is OK
+    result = import_optional_dependency("fakemodule")
+    assert result is module
+
+    with pytest.raises(ImportError, match="Pandas requires version '1.1.0'"):
+        import_optional_dependency("fakemodule", min_version="1.1.0")
+
+    with tm.assert_produces_warning(UserWarning, match="Pandas requires version"):
+        result = import_optional_dependency(
+            "fakemodule", errors="warn", min_version="1.1.0"
+        )
+    assert result is None
+
+    result = import_optional_dependency(
+        "fakemodule", errors="ignore", min_version="1.1.0"
+    )
+    assert result is None
+
+
+def test_submodule(monkeypatch):
+    # Create a fake module with a submodule
+    name = "fakemodule"
+    module = types.ModuleType(name)
+    module.__version__ = "0.9.0"
+    sys.modules[name] = module
+    sub_name = "submodule"
+    submodule = types.ModuleType(sub_name)
+    setattr(module, sub_name, submodule)
+    sys.modules[f"{name}.{sub_name}"] = submodule
+    monkeypatch.setitem(VERSIONS, name, "1.0.0")
+
+    match = "Pandas requires .*1.0.0.* of .fakemodule.*'0.9.0'"
+    with pytest.raises(ImportError, match=match):
+        import_optional_dependency("fakemodule.submodule")
+
+    with tm.assert_produces_warning(UserWarning, match=match):
+        result = import_optional_dependency("fakemodule.submodule", errors="warn")
+    assert result is None
+
+    module.__version__ = "1.0.0"  # exact match is OK
+    result = import_optional_dependency("fakemodule.submodule")
+    assert result is submodule
+
+
+def test_no_version_raises(monkeypatch):
+    name = "fakemodule"
+    module = types.ModuleType(name)
+    sys.modules[name] = module
+    monkeypatch.setitem(VERSIONS, name, "1.0.0")
+
+    with pytest.raises(ImportError, match="Can't determine .* fakemodule"):
+        import_optional_dependency(name)
diff --git a/pandas/pandas/tests/test_register_accessor.py b/pandas/pandas/tests/test_register_accessor.py
new file mode 100644
index 0000000000000..9deff56139394
--- /dev/null
+++ b/pandas/pandas/tests/test_register_accessor.py
@@ -0,0 +1,123 @@
+from collections.abc import Generator
+import contextlib
+import weakref
+
+import pytest
+
+import pandas as pd
+import pandas._testing as tm
+from pandas.core import accessor
+
+
+def test_dirname_mixin() -> None:
+    # GH37173
+
+    class X(accessor.DirNamesMixin):
+        x = 1
+        y: int
+
+        def __init__(self) -> None:
+            self.z = 3
+
+    result = [attr_name for attr_name in dir(X()) if not attr_name.startswith("_")]
+
+    assert result == ["x", "z"]
+
+
+@contextlib.contextmanager
+def ensure_removed(obj, attr) -> Generator[None, None, None]:
+    """Ensure that an attribute added to 'obj' during the test is
+    removed when we're done
+    """
+    try:
+        yield
+    finally:
+        try:
+            delattr(obj, attr)
+        except AttributeError:
+            pass
+        obj._accessors.discard(attr)
+
+
+class MyAccessor:
+    def __init__(self, obj) -> None:
+        self.obj = obj
+        self.item = "item"
+
+    @property
+    def prop(self):
+        return self.item
+
+    def method(self):
+        return self.item
+
+
+@pytest.mark.parametrize(
+    "obj, registrar",
+    [
+        (pd.Series, pd.api.extensions.register_series_accessor),
+        (pd.DataFrame, pd.api.extensions.register_dataframe_accessor),
+        (pd.Index, pd.api.extensions.register_index_accessor),
+    ],
+)
+def test_register(obj, registrar):
+    with ensure_removed(obj, "mine"):
+        before = set(dir(obj))
+        registrar("mine")(MyAccessor)
+        o = obj([]) if obj is not pd.Series else obj([], dtype=object)
+        assert o.mine.prop == "item"
+        after = set(dir(obj))
+        assert (before ^ after) == {"mine"}
+        assert "mine" in obj._accessors
+
+
+def test_accessor_works():
+    with ensure_removed(pd.Series, "mine"):
+        pd.api.extensions.register_series_accessor("mine")(MyAccessor)
+
+        s = pd.Series([1, 2])
+        assert s.mine.obj is s
+
+        assert s.mine.prop == "item"
+        assert s.mine.method() == "item"
+
+
+def test_overwrite_warns():
+    match = r".*MyAccessor.*fake.*Series.*"
+    with tm.assert_produces_warning(UserWarning, match=match):
+        with ensure_removed(pd.Series, "fake"):
+            setattr(pd.Series, "fake", 123)
+            pd.api.extensions.register_series_accessor("fake")(MyAccessor)
+            s = pd.Series([1, 2])
+            assert s.fake.prop == "item"
+
+
+def test_raises_attribute_error():
+    with ensure_removed(pd.Series, "bad"):
+
+        @pd.api.extensions.register_series_accessor("bad")
+        class Bad:
+            def __init__(self, data) -> None:
+                raise AttributeError("whoops")
+
+        with pytest.raises(AttributeError, match="whoops"):
+            pd.Series([], dtype=object).bad
+
+
+@pytest.mark.parametrize(
+    "klass, registrar",
+    [
+        (pd.Series, pd.api.extensions.register_series_accessor),
+        (pd.DataFrame, pd.api.extensions.register_dataframe_accessor),
+        (pd.Index, pd.api.extensions.register_index_accessor),
+    ],
+)
+def test_no_circular_reference(klass, registrar):
+    # GH 41357
+    with ensure_removed(klass, "access"):
+        registrar("access")(MyAccessor)
+        obj = klass([0])
+        ref = weakref.ref(obj)
+        assert obj.access.obj is obj
+        del obj
+        assert ref() is None
diff --git a/pandas/pandas/tests/test_sorting.py b/pandas/pandas/tests/test_sorting.py
new file mode 100644
index 0000000000000..869d41efa6c28
--- /dev/null
+++ b/pandas/pandas/tests/test_sorting.py
@@ -0,0 +1,487 @@
+from collections import defaultdict
+from datetime import datetime
+from itertools import product
+
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    MultiIndex,
+    Series,
+    array,
+    concat,
+    merge,
+)
+import pandas._testing as tm
+from pandas.core.algorithms import safe_sort
+import pandas.core.common as com
+from pandas.core.sorting import (
+    _decons_group_index,
+    get_group_index,
+    is_int64_overflow_possible,
+    lexsort_indexer,
+    nargsort,
+)
+
+
+@pytest.fixture
+def left_right():
+    low, high, n = -1 << 10, 1 << 10, 1 << 20
+    left = DataFrame(
+        np.random.default_rng(2).integers(low, high, (n, 7)), columns=list("ABCDEFG")
+    )
+    left["left"] = left.sum(axis=1)
+    right = left.sample(
+        frac=1, random_state=np.random.default_rng(2), ignore_index=True
+    )
+    right.columns = right.columns[:-1].tolist() + ["right"]
+    right["right"] *= -1
+    return left, right
+
+
+class TestSorting:
+    @pytest.mark.slow
+    def test_int64_overflow(self):
+        B = np.concatenate((np.arange(1000), np.arange(1000), np.arange(500)))
+        A = np.arange(2500)
+        df = DataFrame(
+            {
+                "A": A,
+                "B": B,
+                "C": A,
+                "D": B,
+                "E": A,
+                "F": B,
+                "G": A,
+                "H": B,
+                "values": np.random.default_rng(2).standard_normal(2500),
+            }
+        )
+
+        lg = df.groupby(["A", "B", "C", "D", "E", "F", "G", "H"])
+        rg = df.groupby(["H", "G", "F", "E", "D", "C", "B", "A"])
+
+        left = lg.sum()["values"]
+        right = rg.sum()["values"]
+
+        exp_index, _ = left.index.sortlevel()
+        tm.assert_index_equal(left.index, exp_index)
+
+        exp_index, _ = right.index.sortlevel(0)
+        tm.assert_index_equal(right.index, exp_index)
+
+        tups = list(map(tuple, df[["A", "B", "C", "D", "E", "F", "G", "H"]].values))
+        tups = com.asarray_tuplesafe(tups)
+
+        expected = df.groupby(tups).sum()["values"]
+
+        for k, v in expected.items():
+            assert left[k] == right[k[::-1]]
+            assert left[k] == v
+        assert len(left) == len(right)
+
+    def test_int64_overflow_groupby_large_range(self):
+        # GH9096
+        values = range(55109)
+        data = DataFrame.from_dict({"a": values, "b": values, "c": values, "d": values})
+        grouped = data.groupby(["a", "b", "c", "d"])
+        assert len(grouped) == len(values)
+
+    @pytest.mark.parametrize("agg", ["mean", "median"])
+    def test_int64_overflow_groupby_large_df_shuffled(self, agg):
+        rs = np.random.default_rng(2)
+        arr = rs.integers(-1 << 12, 1 << 12, (1 << 15, 5))
+        i = rs.choice(len(arr), len(arr) * 4)
+        arr = np.vstack((arr, arr[i]))  # add some duplicate rows
+
+        i = rs.permutation(len(arr))
+        arr = arr[i]  # shuffle rows
+
+        df = DataFrame(arr, columns=list("abcde"))
+        df["jim"], df["joe"] = np.zeros((2, len(df)))
+        gr = df.groupby(list("abcde"))
+
+        # verify this is testing what it is supposed to test!
+        assert is_int64_overflow_possible(
+            tuple(ping.ngroups for ping in gr._grouper.groupings)
+        )
+
+        mi = MultiIndex.from_arrays(
+            [ar.ravel() for ar in np.array_split(np.unique(arr, axis=0), 5, axis=1)],
+            names=list("abcde"),
+        )
+
+        res = DataFrame(
+            np.zeros((len(mi), 2)), columns=["jim", "joe"], index=mi
+        ).sort_index()
+
+        tm.assert_frame_equal(getattr(gr, agg)(), res)
+
+    @pytest.mark.parametrize(
+        "order, na_position, exp",
+        [
+            [
+                True,
+                "last",
+                list(range(5, 105)) + list(range(5)) + list(range(105, 110)),
+            ],
+            [
+                True,
+                "first",
+                list(range(5)) + list(range(105, 110)) + list(range(5, 105)),
+            ],
+            [
+                False,
+                "last",
+                list(range(104, 4, -1)) + list(range(5)) + list(range(105, 110)),
+            ],
+            [
+                False,
+                "first",
+                list(range(5)) + list(range(105, 110)) + list(range(104, 4, -1)),
+            ],
+        ],
+    )
+    def test_lexsort_indexer(self, order, na_position, exp):
+        keys = [[np.nan] * 5 + list(range(100)) + [np.nan] * 5]
+        result = lexsort_indexer(keys, orders=order, na_position=na_position)
+        tm.assert_numpy_array_equal(result, np.array(exp, dtype=np.intp))
+
+    @pytest.mark.parametrize(
+        "ascending, na_position, exp",
+        [
+            [
+                True,
+                "last",
+                list(range(5, 105)) + list(range(5)) + list(range(105, 110)),
+            ],
+            [
+                True,
+                "first",
+                list(range(5)) + list(range(105, 110)) + list(range(5, 105)),
+            ],
+            [
+                False,
+                "last",
+                list(range(104, 4, -1)) + list(range(5)) + list(range(105, 110)),
+            ],
+            [
+                False,
+                "first",
+                list(range(5)) + list(range(105, 110)) + list(range(104, 4, -1)),
+            ],
+        ],
+    )
+    def test_nargsort(self, ascending, na_position, exp):
+        # list places NaNs last, np.array(..., dtype="O") may not place NaNs first
+        items = np.array([np.nan] * 5 + list(range(100)) + [np.nan] * 5, dtype="O")
+
+        # mergesort is the most difficult to get right because we want it to be
+        # stable.
+
+        # According to numpy/core/tests/test_multiarray, """The number of
+        # sorted items must be greater than ~50 to check the actual algorithm
+        # because quick and merge sort fall over to insertion sort for small
+        # arrays."""
+
+        result = nargsort(
+            items, kind="mergesort", ascending=ascending, na_position=na_position
+        )
+        tm.assert_numpy_array_equal(result, np.array(exp), check_dtype=False)
+
+
+class TestMerge:
+    def test_int64_overflow_outer_merge(self):
+        # #2690, combinatorial explosion
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((1000, 7)),
+            columns=list("ABCDEF") + ["G1"],
+        )
+        df2 = DataFrame(
+            np.random.default_rng(3).standard_normal((1000, 7)),
+            columns=list("ABCDEF") + ["G2"],
+        )
+        result = merge(df1, df2, how="outer")
+        assert len(result) == 2000
+
+    @pytest.mark.slow
+    def test_int64_overflow_check_sum_col(self, left_right):
+        left, right = left_right
+
+        out = merge(left, right, how="outer")
+        assert len(out) == len(left)
+        tm.assert_series_equal(out["left"], -out["right"], check_names=False)
+        result = out.iloc[:, :-2].sum(axis=1)
+        tm.assert_series_equal(out["left"], result, check_names=False)
+        assert result.name is None
+
+    @pytest.mark.slow
+    def test_int64_overflow_how_merge(self, left_right, join_type):
+        left, right = left_right
+
+        out = merge(left, right, how="outer")
+        out.sort_values(out.columns.tolist(), inplace=True)
+        tm.assert_frame_equal(out, merge(left, right, how=join_type, sort=True))
+
+    @pytest.mark.slow
+    def test_int64_overflow_sort_false_order(self, left_right):
+        left, right = left_right
+
+        # check that left merge w/ sort=False maintains left frame order
+        out = merge(left, right, how="left", sort=False)
+        tm.assert_frame_equal(left, out[left.columns.tolist()])
+
+        out = merge(right, left, how="left", sort=False)
+        tm.assert_frame_equal(right, out[right.columns.tolist()])
+
+    @pytest.mark.slow
+    def test_int64_overflow_one_to_many_none_match(self, join_type, sort):
+        # one-2-many/none match
+        how = join_type
+        low, high, n = -1 << 10, 1 << 10, 1 << 11
+        left = DataFrame(
+            np.random.default_rng(2).integers(low, high, (n, 7)).astype("int64"),
+            columns=list("ABCDEFG"),
+        )
+
+        # confirm that this is checking what it is supposed to check
+        shape = left.apply(Series.nunique).values
+        assert is_int64_overflow_possible(shape)
+
+        # add duplicates to left frame
+        left = concat([left, left], ignore_index=True)
+
+        right = DataFrame(
+            np.random.default_rng(3).integers(low, high, (n // 2, 7)).astype("int64"),
+            columns=list("ABCDEFG"),
+        )
+
+        # add duplicates & overlap with left to the right frame
+        i = np.random.default_rng(4).choice(len(left), n)
+        right = concat([right, right, left.iloc[i]], ignore_index=True)
+
+        left["left"] = np.random.default_rng(2).standard_normal(len(left))
+        right["right"] = np.random.default_rng(2).standard_normal(len(right))
+
+        # shuffle left & right frames
+        left = left.sample(
+            frac=1, ignore_index=True, random_state=np.random.default_rng(5)
+        )
+        right = right.sample(
+            frac=1, ignore_index=True, random_state=np.random.default_rng(6)
+        )
+
+        # manually compute outer merge
+        ldict, rdict = defaultdict(list), defaultdict(list)
+
+        for idx, row in left.set_index(list("ABCDEFG")).iterrows():
+            ldict[idx].append(row["left"])
+
+        for idx, row in right.set_index(list("ABCDEFG")).iterrows():
+            rdict[idx].append(row["right"])
+
+        vals = []
+        for k, lval in ldict.items():
+            rval = rdict.get(k, [np.nan])
+            for lv, rv in product(lval, rval):
+                vals.append(
+                    k
+                    + (
+                        lv,
+                        rv,
+                    )
+                )
+
+        for k, rval in rdict.items():
+            if k not in ldict:
+                vals.extend(
+                    k
+                    + (
+                        np.nan,
+                        rv,
+                    )
+                    for rv in rval
+                )
+
+        out = DataFrame(vals, columns=list("ABCDEFG") + ["left", "right"])
+        out = out.sort_values(out.columns.to_list(), ignore_index=True)
+
+        jmask = {
+            "left": out["left"].notna(),
+            "right": out["right"].notna(),
+            "inner": out["left"].notna() & out["right"].notna(),
+            "outer": np.ones(len(out), dtype="bool"),
+        }
+
+        mask = jmask[how]
+        frame = out[mask].sort_values(out.columns.to_list(), ignore_index=True)
+        assert mask.all() ^ mask.any() or how == "outer"
+
+        res = merge(left, right, how=how, sort=sort)
+        if sort:
+            kcols = list("ABCDEFG")
+            tm.assert_frame_equal(
+                res[kcols], res[kcols].sort_values(kcols, kind="mergesort")
+            )
+
+        # as in GH9092 dtypes break with outer/right join
+        # 2021-12-18: dtype does not break anymore
+        tm.assert_frame_equal(
+            frame, res.sort_values(res.columns.to_list(), ignore_index=True)
+        )
+
+
+@pytest.mark.parametrize(
+    "codes_list, shape",
+    [
+        [
+            [
+                np.tile([0, 1, 2, 3, 0, 1, 2, 3], 100).astype(np.int64),
+                np.tile([0, 2, 4, 3, 0, 1, 2, 3], 100).astype(np.int64),
+                np.tile([5, 1, 0, 2, 3, 0, 5, 4], 100).astype(np.int64),
+            ],
+            (4, 5, 6),
+        ],
+        [
+            [
+                np.tile(np.arange(10000, dtype=np.int64), 5),
+                np.tile(np.arange(10000, dtype=np.int64), 5),
+            ],
+            (10000, 10000),
+        ],
+    ],
+)
+def test_decons(codes_list, shape):
+    group_index = get_group_index(codes_list, shape, sort=True, xnull=True)
+    codes_list2 = _decons_group_index(group_index, shape)
+
+    for a, b in zip(codes_list, codes_list2):
+        tm.assert_numpy_array_equal(a, b)
+
+
+class TestSafeSort:
+    @pytest.mark.parametrize(
+        "arg, exp",
+        [
+            [[3, 1, 2, 0, 4], [0, 1, 2, 3, 4]],
+            [
+                np.array(list("baaacb"), dtype=object),
+                np.array(list("aaabbc"), dtype=object),
+            ],
+            [[], []],
+        ],
+    )
+    def test_basic_sort(self, arg, exp):
+        result = safe_sort(np.array(arg))
+        expected = np.array(exp)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("verify", [True, False])
+    @pytest.mark.parametrize(
+        "codes, exp_codes",
+        [
+            [[0, 1, 1, 2, 3, 0, -1, 4], [3, 1, 1, 2, 0, 3, -1, 4]],
+            [[], []],
+        ],
+    )
+    def test_codes(self, verify, codes, exp_codes):
+        values = np.array([3, 1, 2, 0, 4])
+        expected = np.array([0, 1, 2, 3, 4])
+
+        result, result_codes = safe_sort(
+            values, codes, use_na_sentinel=True, verify=verify
+        )
+        expected_codes = np.array(exp_codes, dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+        tm.assert_numpy_array_equal(result_codes, expected_codes)
+
+    def test_codes_out_of_bound(self):
+        values = np.array([3, 1, 2, 0, 4])
+        expected = np.array([0, 1, 2, 3, 4])
+
+        # out of bound indices
+        codes = [0, 101, 102, 2, 3, 0, 99, 4]
+        result, result_codes = safe_sort(values, codes, use_na_sentinel=True)
+        expected_codes = np.array([3, -1, -1, 2, 0, 3, -1, 4], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+        tm.assert_numpy_array_equal(result_codes, expected_codes)
+
+    @pytest.mark.parametrize("codes", [[-1, -1], [2, -1], [2, 2]])
+    def test_codes_empty_array_out_of_bound(self, codes):
+        empty_values = np.array([])
+        expected_codes = -np.ones_like(codes, dtype=np.intp)
+        _, result_codes = safe_sort(empty_values, codes)
+        tm.assert_numpy_array_equal(result_codes, expected_codes)
+
+    def test_mixed_integer(self):
+        values = np.array(["b", 1, 0, "a", 0, "b"], dtype=object)
+        result = safe_sort(values)
+        expected = np.array([0, 0, 1, "a", "b", "b"], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_mixed_integer_with_codes(self):
+        values = np.array(["b", 1, 0, "a"], dtype=object)
+        codes = [0, 1, 2, 3, 0, -1, 1]
+        result, result_codes = safe_sort(values, codes)
+        expected = np.array([0, 1, "a", "b"], dtype=object)
+        expected_codes = np.array([3, 1, 0, 2, 3, -1, 1], dtype=np.intp)
+        tm.assert_numpy_array_equal(result, expected)
+        tm.assert_numpy_array_equal(result_codes, expected_codes)
+
+    def test_unsortable(self):
+        # GH 13714
+        arr = np.array([1, 2, datetime.now(), 0, 3], dtype=object)
+        msg = "'[<>]' not supported between instances of .*"
+        with pytest.raises(TypeError, match=msg):
+            safe_sort(arr)
+
+    @pytest.mark.parametrize(
+        "arg, codes, err, msg",
+        [
+            [1, None, TypeError, "Only np.ndarray, ExtensionArray, and Index"],
+            [np.array([0, 1, 2]), 1, TypeError, "Only list-like objects or None"],
+            [np.array([0, 1, 2, 1]), [0, 1], ValueError, "values should be unique"],
+        ],
+    )
+    def test_exceptions(self, arg, codes, err, msg):
+        with pytest.raises(err, match=msg):
+            safe_sort(values=arg, codes=codes)
+
+    @pytest.mark.parametrize(
+        "arg, exp", [[[1, 3, 2], [1, 2, 3]], [[1, 3, np.nan, 2], [1, 2, 3, np.nan]]]
+    )
+    def test_extension_array(self, arg, exp):
+        a = array(arg, dtype="Int64")
+        result = safe_sort(a)
+        expected = array(exp, dtype="Int64")
+        tm.assert_extension_array_equal(result, expected)
+
+    @pytest.mark.parametrize("verify", [True, False])
+    def test_extension_array_codes(self, verify):
+        a = array([1, 3, 2], dtype="Int64")
+        result, codes = safe_sort(a, [0, 1, -1, 2], use_na_sentinel=True, verify=verify)
+        expected_values = array([1, 2, 3], dtype="Int64")
+        expected_codes = np.array([0, 2, -1, 1], dtype=np.intp)
+        tm.assert_extension_array_equal(result, expected_values)
+        tm.assert_numpy_array_equal(codes, expected_codes)
+
+
+def test_mixed_str_null(nulls_fixture):
+    values = np.array(["b", nulls_fixture, "a", "b"], dtype=object)
+    result = safe_sort(values)
+    expected = np.array(["a", "b", "b", nulls_fixture], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_safe_sort_multiindex():
+    # GH#48412
+    arr1 = Series([2, 1, NA, NA], dtype="Int64")
+    arr2 = [2, 1, 3, 3]
+    midx = MultiIndex.from_arrays([arr1, arr2])
+    result = safe_sort(midx)
+    expected = MultiIndex.from_arrays(
+        [Series([1, 2, NA, NA], dtype="Int64"), [1, 2, 3, 3]]
+    )
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/test_take.py b/pandas/pandas/tests/test_take.py
new file mode 100644
index 0000000000000..451ef42fff3d1
--- /dev/null
+++ b/pandas/pandas/tests/test_take.py
@@ -0,0 +1,317 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas._libs import iNaT
+
+from pandas import array
+import pandas._testing as tm
+import pandas.core.algorithms as algos
+
+
+@pytest.fixture(
+    params=[
+        (np.int8, np.int16(127), np.int8),
+        (np.int8, np.int16(128), np.int16),
+        (np.int32, 1, np.int32),
+        (np.int32, 2.0, np.float64),
+        (np.int32, 3.0 + 4.0j, np.complex128),
+        (np.int32, True, np.object_),
+        (np.int32, "", np.object_),
+        (np.float64, 1, np.float64),
+        (np.float64, 2.0, np.float64),
+        (np.float64, 3.0 + 4.0j, np.complex128),
+        (np.float64, True, np.object_),
+        (np.float64, "", np.object_),
+        (np.complex128, 1, np.complex128),
+        (np.complex128, 2.0, np.complex128),
+        (np.complex128, 3.0 + 4.0j, np.complex128),
+        (np.complex128, True, np.object_),
+        (np.complex128, "", np.object_),
+        (np.bool_, 1, np.object_),
+        (np.bool_, 2.0, np.object_),
+        (np.bool_, 3.0 + 4.0j, np.object_),
+        (np.bool_, True, np.bool_),
+        (np.bool_, "", np.object_),
+    ]
+)
+def dtype_fill_out_dtype(request):
+    return request.param
+
+
+class TestTake:
+    def test_1d_fill_nonna(self, dtype_fill_out_dtype):
+        dtype, fill_value, out_dtype = dtype_fill_out_dtype
+        data = np.random.default_rng(2).integers(0, 2, 4).astype(dtype)
+        indexer = [2, 1, 0, -1]
+
+        result = algos.take_nd(data, indexer, fill_value=fill_value)
+        assert (result[[0, 1, 2]] == data[[2, 1, 0]]).all()
+        assert result[3] == fill_value
+        assert result.dtype == out_dtype
+
+        indexer = [2, 1, 0, 1]
+
+        result = algos.take_nd(data, indexer, fill_value=fill_value)
+        assert (result[[0, 1, 2, 3]] == data[indexer]).all()
+        assert result.dtype == dtype
+
+    def test_2d_fill_nonna(self, dtype_fill_out_dtype):
+        dtype, fill_value, out_dtype = dtype_fill_out_dtype
+        data = np.random.default_rng(2).integers(0, 2, (5, 3)).astype(dtype)
+        indexer = [2, 1, 0, -1]
+
+        result = algos.take_nd(data, indexer, axis=0, fill_value=fill_value)
+        assert (result[[0, 1, 2], :] == data[[2, 1, 0], :]).all()
+        assert (result[3, :] == fill_value).all()
+        assert result.dtype == out_dtype
+
+        result = algos.take_nd(data, indexer, axis=1, fill_value=fill_value)
+        assert (result[:, [0, 1, 2]] == data[:, [2, 1, 0]]).all()
+        assert (result[:, 3] == fill_value).all()
+        assert result.dtype == out_dtype
+
+        indexer = [2, 1, 0, 1]
+        result = algos.take_nd(data, indexer, axis=0, fill_value=fill_value)
+        assert (result[[0, 1, 2, 3], :] == data[indexer, :]).all()
+        assert result.dtype == dtype
+
+        result = algos.take_nd(data, indexer, axis=1, fill_value=fill_value)
+        assert (result[:, [0, 1, 2, 3]] == data[:, indexer]).all()
+        assert result.dtype == dtype
+
+    def test_3d_fill_nonna(self, dtype_fill_out_dtype):
+        dtype, fill_value, out_dtype = dtype_fill_out_dtype
+
+        data = np.random.default_rng(2).integers(0, 2, (5, 4, 3)).astype(dtype)
+        indexer = [2, 1, 0, -1]
+
+        result = algos.take_nd(data, indexer, axis=0, fill_value=fill_value)
+        assert (result[[0, 1, 2], :, :] == data[[2, 1, 0], :, :]).all()
+        assert (result[3, :, :] == fill_value).all()
+        assert result.dtype == out_dtype
+
+        result = algos.take_nd(data, indexer, axis=1, fill_value=fill_value)
+        assert (result[:, [0, 1, 2], :] == data[:, [2, 1, 0], :]).all()
+        assert (result[:, 3, :] == fill_value).all()
+        assert result.dtype == out_dtype
+
+        result = algos.take_nd(data, indexer, axis=2, fill_value=fill_value)
+        assert (result[:, :, [0, 1, 2]] == data[:, :, [2, 1, 0]]).all()
+        assert (result[:, :, 3] == fill_value).all()
+        assert result.dtype == out_dtype
+
+        indexer = [2, 1, 0, 1]
+        result = algos.take_nd(data, indexer, axis=0, fill_value=fill_value)
+        assert (result[[0, 1, 2, 3], :, :] == data[indexer, :, :]).all()
+        assert result.dtype == dtype
+
+        result = algos.take_nd(data, indexer, axis=1, fill_value=fill_value)
+        assert (result[:, [0, 1, 2, 3], :] == data[:, indexer, :]).all()
+        assert result.dtype == dtype
+
+        result = algos.take_nd(data, indexer, axis=2, fill_value=fill_value)
+        assert (result[:, :, [0, 1, 2, 3]] == data[:, :, indexer]).all()
+        assert result.dtype == dtype
+
+    def test_1d_other_dtypes(self):
+        arr = np.random.default_rng(2).standard_normal(10).astype(np.float32)
+
+        indexer = [1, 2, 3, -1]
+        result = algos.take_nd(arr, indexer)
+        expected = arr.take(indexer)
+        expected[-1] = np.nan
+        tm.assert_almost_equal(result, expected)
+
+    def test_2d_other_dtypes(self):
+        arr = np.random.default_rng(2).standard_normal((10, 5)).astype(np.float32)
+
+        indexer = [1, 2, 3, -1]
+
+        # axis=0
+        result = algos.take_nd(arr, indexer, axis=0)
+        expected = arr.take(indexer, axis=0)
+        expected[-1] = np.nan
+        tm.assert_almost_equal(result, expected)
+
+        # axis=1
+        result = algos.take_nd(arr, indexer, axis=1)
+        expected = arr.take(indexer, axis=1)
+        expected[:, -1] = np.nan
+        tm.assert_almost_equal(result, expected)
+
+    def test_1d_bool(self):
+        arr = np.array([0, 1, 0], dtype=bool)
+
+        result = algos.take_nd(arr, [0, 2, 2, 1])
+        expected = arr.take([0, 2, 2, 1])
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.take_nd(arr, [0, 2, -1])
+        assert result.dtype == np.object_
+
+    def test_2d_bool(self):
+        arr = np.array([[0, 1, 0], [1, 0, 1], [0, 1, 1]], dtype=bool)
+
+        result = algos.take_nd(arr, [0, 2, 2, 1])
+        expected = arr.take([0, 2, 2, 1], axis=0)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.take_nd(arr, [0, 2, 2, 1], axis=1)
+        expected = arr.take([0, 2, 2, 1], axis=1)
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = algos.take_nd(arr, [0, 2, -1])
+        assert result.dtype == np.object_
+
+    def test_2d_float32(self):
+        arr = np.random.default_rng(2).standard_normal((4, 3)).astype(np.float32)
+        indexer = [0, 2, -1, 1, -1]
+
+        # axis=0
+        result = algos.take_nd(arr, indexer, axis=0)
+
+        expected = arr.take(indexer, axis=0)
+        expected[[2, 4], :] = np.nan
+        tm.assert_almost_equal(result, expected)
+
+        # axis=1
+        result = algos.take_nd(arr, indexer, axis=1)
+        expected = arr.take(indexer, axis=1)
+        expected[:, [2, 4]] = np.nan
+        tm.assert_almost_equal(result, expected)
+
+    def test_2d_datetime64(self):
+        # 2005/01/01 - 2006/01/01
+        arr = (
+            np.random.default_rng(2).integers(11_045_376, 11_360_736, (5, 3))
+            * 100_000_000_000
+        )
+        arr = arr.view(dtype="datetime64[ns]")
+        indexer = [0, 2, -1, 1, -1]
+
+        # axis=0
+        result = algos.take_nd(arr, indexer, axis=0)
+        expected = arr.take(indexer, axis=0)
+        expected.view(np.int64)[[2, 4], :] = iNaT
+        tm.assert_almost_equal(result, expected)
+
+        result = algos.take_nd(arr, indexer, axis=0, fill_value=datetime(2007, 1, 1))
+        expected = arr.take(indexer, axis=0)
+        expected[[2, 4], :] = datetime(2007, 1, 1)
+        tm.assert_almost_equal(result, expected)
+
+        # axis=1
+        result = algos.take_nd(arr, indexer, axis=1)
+        expected = arr.take(indexer, axis=1)
+        expected.view(np.int64)[:, [2, 4]] = iNaT
+        tm.assert_almost_equal(result, expected)
+
+        result = algos.take_nd(arr, indexer, axis=1, fill_value=datetime(2007, 1, 1))
+        expected = arr.take(indexer, axis=1)
+        expected[:, [2, 4]] = datetime(2007, 1, 1)
+        tm.assert_almost_equal(result, expected)
+
+    def test_take_axis_0(self):
+        arr = np.arange(12).reshape(4, 3)
+        result = algos.take(arr, [0, -1])
+        expected = np.array([[0, 1, 2], [9, 10, 11]])
+        tm.assert_numpy_array_equal(result, expected)
+
+        # allow_fill=True
+        result = algos.take(arr, [0, -1], allow_fill=True, fill_value=0)
+        expected = np.array([[0, 1, 2], [0, 0, 0]])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_take_axis_1(self):
+        arr = np.arange(12).reshape(4, 3)
+        result = algos.take(arr, [0, -1], axis=1)
+        expected = np.array([[0, 2], [3, 5], [6, 8], [9, 11]])
+        tm.assert_numpy_array_equal(result, expected)
+
+        # allow_fill=True
+        result = algos.take(arr, [0, -1], axis=1, allow_fill=True, fill_value=0)
+        expected = np.array([[0, 0], [3, 0], [6, 0], [9, 0]])
+        tm.assert_numpy_array_equal(result, expected)
+
+        # GH#26976 make sure we validate along the correct axis
+        with pytest.raises(IndexError, match="indices are out-of-bounds"):
+            algos.take(arr, [0, 3], axis=1, allow_fill=True, fill_value=0)
+
+    def test_take_non_hashable_fill_value(self):
+        arr = np.array([1, 2, 3])
+        indexer = np.array([1, -1])
+        with pytest.raises(ValueError, match="fill_value must be a scalar"):
+            algos.take(arr, indexer, allow_fill=True, fill_value=[1])
+
+        # with object dtype it is allowed
+        arr = np.array([1, 2, 3], dtype=object)
+        result = algos.take(arr, indexer, allow_fill=True, fill_value=[1])
+        expected = np.array([2, [1]], dtype=object)
+        tm.assert_numpy_array_equal(result, expected)
+
+
+class TestExtensionTake:
+    # The take method found in pd.api.extensions
+
+    def test_bounds_check_large(self):
+        arr = np.array([1, 2])
+
+        msg = "indices are out-of-bounds"
+        with pytest.raises(IndexError, match=msg):
+            algos.take(arr, [2, 3], allow_fill=True)
+
+        msg = "index 2 is out of bounds for( axis 0 with)? size 2"
+        with pytest.raises(IndexError, match=msg):
+            algos.take(arr, [2, 3], allow_fill=False)
+
+    def test_bounds_check_small(self):
+        arr = np.array([1, 2, 3], dtype=np.int64)
+        indexer = [0, -1, -2]
+
+        msg = r"'indices' contains values less than allowed \(-2 < -1\)"
+        with pytest.raises(ValueError, match=msg):
+            algos.take(arr, indexer, allow_fill=True)
+
+        result = algos.take(arr, indexer)
+        expected = np.array([1, 3, 2], dtype=np.int64)
+        tm.assert_numpy_array_equal(result, expected)
+
+    @pytest.mark.parametrize("allow_fill", [True, False])
+    def test_take_empty(self, allow_fill):
+        arr = np.array([], dtype=np.int64)
+        # empty take is ok
+        result = algos.take(arr, [], allow_fill=allow_fill)
+        tm.assert_numpy_array_equal(arr, result)
+
+        msg = "|".join(
+            [
+                "cannot do a non-empty take from an empty axes.",
+                "indices are out-of-bounds",
+            ]
+        )
+        with pytest.raises(IndexError, match=msg):
+            algos.take(arr, [0], allow_fill=allow_fill)
+
+    def test_take_na_empty(self):
+        result = algos.take(np.array([]), [-1, -1], allow_fill=True, fill_value=0.0)
+        expected = np.array([0.0, 0.0])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_take_coerces_list(self):
+        # GH#52981 coercing is deprecated, disabled in 3.0
+        arr = [1, 2, 3]
+        msg = (
+            "pd.api.extensions.take requires a numpy.ndarray, ExtensionArray, "
+            "Index, Series, or NumpyExtensionArray got list"
+        )
+        with pytest.raises(TypeError, match=msg):
+            algos.take(arr, [0, 0])
+
+    def test_take_NumpyExtensionArray(self):
+        # GH#59177
+        arr = array([1 + 1j, 2, 3])  # NumpyEADtype('complex128') (NumpyExtensionArray)
+        assert algos.take(arr, [2]) == 2
+        arr = array([1, 2, 3])  # Int64Dtype() (ExtensionArray)
+        assert algos.take(arr, [2]) == 2
diff --git a/pandas/pandas/tests/tools/__init__.py b/pandas/pandas/tests/tools/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tools/test_to_datetime.py b/pandas/pandas/tests/tools/test_to_datetime.py
new file mode 100644
index 0000000000000..e039f54960389
--- /dev/null
+++ b/pandas/pandas/tests/tools/test_to_datetime.py
@@ -0,0 +1,3715 @@
+"""test to_datetime"""
+
+import calendar
+from collections import deque
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+    timezone,
+)
+from decimal import Decimal
+import locale
+import zoneinfo
+
+from dateutil.parser import parse
+import numpy as np
+import pytest
+
+from pandas._libs import tslib
+from pandas._libs.tslibs import (
+    iNaT,
+    parsing,
+)
+from pandas.compat import WASM
+from pandas.errors import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+)
+import pandas.util._test_decorators as td
+
+from pandas.core.dtypes.common import is_datetime64_ns_dtype
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+    isna,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.core.arrays import DatetimeArray
+from pandas.core.tools import datetimes as tools
+from pandas.core.tools.datetimes import start_caching_at
+
+PARSING_ERR_MSG = (
+    r"You might want to try:\n"
+    r"    - passing `format` if your strings have a consistent format;\n"
+    r"    - passing `format=\'ISO8601\'` if your strings are all ISO8601 "
+    r"but not necessarily in exactly the same format;\n"
+    r"    - passing `format=\'mixed\'`, and the format will be inferred "
+    r"for each element individually. You might want to use `dayfirst` "
+    r"alongside this."
+)
+
+
+class TestTimeConversionFormats:
+    def test_to_datetime_readonly(self, writable):
+        # GH#34857
+        arr = np.array([], dtype=object)
+        arr.setflags(write=writable)
+        result = to_datetime(arr)
+        expected = to_datetime([])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "format, expected",
+        [
+            [
+                "%d/%m/%Y",
+                [Timestamp("20000101"), Timestamp("20000201"), Timestamp("20000301")],
+            ],
+            [
+                "%m/%d/%Y",
+                [Timestamp("20000101"), Timestamp("20000102"), Timestamp("20000103")],
+            ],
+        ],
+    )
+    def test_to_datetime_format(self, cache, index_or_series, format, expected):
+        values = index_or_series(["1/1/2000", "1/2/2000", "1/3/2000"])
+        result = to_datetime(values, format=format, cache=cache)
+        expected = index_or_series(expected)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arg, expected, format",
+        [
+            ["1/1/2000", "20000101", "%d/%m/%Y"],
+            ["1/1/2000", "20000101", "%m/%d/%Y"],
+            ["1/2/2000", "20000201", "%d/%m/%Y"],
+            ["1/2/2000", "20000102", "%m/%d/%Y"],
+            ["1/3/2000", "20000301", "%d/%m/%Y"],
+            ["1/3/2000", "20000103", "%m/%d/%Y"],
+        ],
+    )
+    def test_to_datetime_format_scalar(self, cache, arg, expected, format):
+        result = to_datetime(arg, format=format, cache=cache)
+        expected = Timestamp(expected)
+        assert result == expected
+
+    def test_to_datetime_format_YYYYMMDD(self, cache):
+        ser = Series([19801222, 19801222] + [19810105] * 5)
+        expected = Series([Timestamp(x) for x in ser.apply(str)])
+
+        result = to_datetime(ser, format="%Y%m%d", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+        result = to_datetime(ser.apply(str), format="%Y%m%d", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_format_YYYYMMDD_with_nat(self, cache):
+        # Explicit cast to float to explicit cast when setting np.nan
+        ser = Series([19801222, 19801222] + [19810105] * 5, dtype="float")
+        # with NaT
+        expected = Series(
+            [Timestamp("19801222"), Timestamp("19801222")]
+            + [Timestamp("19810105")] * 5,
+            dtype="M8[s]",
+        )
+        expected[2] = np.nan
+        ser[2] = np.nan
+
+        result = to_datetime(ser, format="%Y%m%d", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+        # string with NaT
+        ser2 = ser.apply(str)
+        ser2[2] = "nat"
+        with pytest.raises(
+            ValueError,
+            match=(
+                'unconverted data remains when parsing with format "%Y%m%d": ".0". '
+            ),
+        ):
+            # https://github.com/pandas-dev/pandas/issues/50051
+            to_datetime(ser2, format="%Y%m%d", cache=cache)
+
+    def test_to_datetime_format_YYYYMM_with_nat(self, cache):
+        # https://github.com/pandas-dev/pandas/issues/50237
+        # Explicit cast to float to explicit cast when setting np.nan
+        ser = Series([198012, 198012] + [198101] * 5, dtype="float")
+        expected = Series(
+            [Timestamp("19801201"), Timestamp("19801201")]
+            + [Timestamp("19810101")] * 5,
+            dtype="M8[s]",
+        )
+        expected[2] = np.nan
+        ser[2] = np.nan
+        result = to_datetime(ser, format="%Y%m", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_format_YYYYMMDD_oob_for_ns(self, cache):
+        # coercion
+        # GH 7930, GH 14487
+        ser = Series([20121231, 20141231, 99991231])
+        result = to_datetime(ser, format="%Y%m%d", errors="raise", cache=cache)
+        expected = Series(
+            np.array(["2012-12-31", "2014-12-31", "9999-12-31"], dtype="M8[s]"),
+            dtype="M8[s]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_format_YYYYMMDD_coercion(self, cache):
+        # coercion
+        # GH 7930
+        ser = Series([20121231, 20141231, 999999999999999999999999999991231])
+        result = to_datetime(ser, format="%Y%m%d", errors="coerce", cache=cache)
+        expected = Series(["20121231", "20141231", "NaT"], dtype="M8[s]")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "input_s",
+        [
+            # Null values with Strings
+            ["19801222", "20010112", None],
+            ["19801222", "20010112", np.nan],
+            ["19801222", "20010112", NaT],
+            ["19801222", "20010112", "NaT"],
+            # Null values with Integers
+            [19801222, 20010112, None],
+            [19801222, 20010112, np.nan],
+            [19801222, 20010112, NaT],
+            [19801222, 20010112, "NaT"],
+        ],
+    )
+    def test_to_datetime_format_YYYYMMDD_with_none(self, input_s):
+        # GH 30011
+        # format='%Y%m%d'
+        # with None
+        expected = Series([Timestamp("19801222"), Timestamp("20010112"), NaT])
+        result = Series(to_datetime(input_s, format="%Y%m%d"))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "input_s, expected",
+        [
+            # NaN before strings with invalid date values
+            [
+                ["19801222", np.nan, "20010012", "10019999"],
+                [Timestamp("19801222"), np.nan, np.nan, np.nan],
+            ],
+            # NaN after strings with invalid date values
+            [
+                ["19801222", "20010012", "10019999", np.nan],
+                [Timestamp("19801222"), np.nan, np.nan, np.nan],
+            ],
+            # NaN before integers with invalid date values
+            [
+                [20190813, np.nan, 20010012, 20019999],
+                [Timestamp("20190813"), np.nan, np.nan, np.nan],
+            ],
+            # NaN after integers with invalid date values
+            [
+                [20190813, 20010012, np.nan, 20019999],
+                [Timestamp("20190813"), np.nan, np.nan, np.nan],
+            ],
+        ],
+    )
+    def test_to_datetime_format_YYYYMMDD_overflow(self, input_s, expected):
+        # GH 25512
+        # format='%Y%m%d', errors='coerce'
+        input_s = Series(input_s)
+        result = to_datetime(input_s, format="%Y%m%d", errors="coerce")
+        expected = Series(expected)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data, format, expected",
+        [
+            ([pd.NA], "%Y%m%d%H%M%S", ["NaT"]),
+            ([pd.NA], None, ["NaT"]),
+            (
+                [pd.NA, "20210202202020"],
+                "%Y%m%d%H%M%S",
+                ["NaT", "2021-02-02 20:20:20"],
+            ),
+            (["201010", pd.NA], "%y%m%d", ["2020-10-10", "NaT"]),
+            (["201010", pd.NA], "%d%m%y", ["2010-10-20", "NaT"]),
+            ([None, np.nan, pd.NA], None, ["NaT", "NaT", "NaT"]),
+            ([None, np.nan, pd.NA], "%Y%m%d", ["NaT", "NaT", "NaT"]),
+        ],
+    )
+    def test_to_datetime_with_NA(self, data, format, expected):
+        # GH#42957
+        result = to_datetime(data, format=format)
+        expected = DatetimeIndex(expected)
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_with_NA_with_warning(self):
+        # GH#42957
+        result = to_datetime(["201010", pd.NA])
+        expected = DatetimeIndex(["2010-10-20", "NaT"])
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_format_integer(self, cache):
+        # GH 10178
+        ser = Series([2000, 2001, 2002])
+        expected = Series([Timestamp(x) for x in ser.apply(str)])
+
+        result = to_datetime(ser, format="%Y", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+        ser = Series([200001, 200105, 200206])
+        expected = Series([Timestamp(x[:4] + "-" + x[4:]) for x in ser.apply(str)])
+
+        result = to_datetime(ser, format="%Y%m", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_format_microsecond(self, cache):
+        month_abbr = calendar.month_abbr[4]
+        val = f"01-{month_abbr}-2011 00:00:01.978"
+
+        format = "%d-%b-%Y %H:%M:%S.%f"
+        result = to_datetime(val, format=format, cache=cache)
+        exp = datetime.strptime(val, format)
+        assert result == exp
+
+    @pytest.mark.parametrize(
+        "value, format, dt",
+        [
+            ["01/10/2010 15:20", "%m/%d/%Y %H:%M", Timestamp("2010-01-10 15:20")],
+            ["01/10/2010 05:43", "%m/%d/%Y %I:%M", Timestamp("2010-01-10 05:43")],
+            [
+                "01/10/2010 13:56:01",
+                "%m/%d/%Y %H:%M:%S",
+                Timestamp("2010-01-10 13:56:01"),
+            ],
+            # The 3 tests below are locale-dependent.
+            # They pass, except when the machine locale is zh_CN or it_IT .
+            pytest.param(
+                "01/10/2010 08:14 PM",
+                "%m/%d/%Y %I:%M %p",
+                Timestamp("2010-01-10 20:14"),
+                marks=pytest.mark.xfail(
+                    locale.getlocale()[0] in ("zh_CN", "it_IT"),
+                    reason="fail on a CI build with LC_ALL=zh_CN.utf8/it_IT.utf8",
+                    strict=False,
+                ),
+            ),
+            pytest.param(
+                "01/10/2010 07:40 AM",
+                "%m/%d/%Y %I:%M %p",
+                Timestamp("2010-01-10 07:40"),
+                marks=pytest.mark.xfail(
+                    locale.getlocale()[0] in ("zh_CN", "it_IT"),
+                    reason="fail on a CI build with LC_ALL=zh_CN.utf8/it_IT.utf8",
+                    strict=False,
+                ),
+            ),
+            pytest.param(
+                "01/10/2010 09:12:56 AM",
+                "%m/%d/%Y %I:%M:%S %p",
+                Timestamp("2010-01-10 09:12:56"),
+                marks=pytest.mark.xfail(
+                    locale.getlocale()[0] in ("zh_CN", "it_IT"),
+                    reason="fail on a CI build with LC_ALL=zh_CN.utf8/it_IT.utf8",
+                    strict=False,
+                ),
+            ),
+        ],
+    )
+    def test_to_datetime_format_time(self, cache, value, format, dt):
+        assert to_datetime(value, format=format, cache=cache) == dt
+
+    @td.skip_if_not_us_locale
+    def test_to_datetime_with_non_exact(self, cache):
+        # GH 10834
+        # 8904
+        # exact kw
+        ser = Series(
+            ["19MAY11", "foobar19MAY11", "19MAY11:00:00:00", "19MAY11 00:00:00Z"]
+        )
+        result = to_datetime(ser, format="%d%b%y", exact=False, cache=cache)
+        expected = to_datetime(
+            ser.str.extract(r"(\d+\w+\d+)", expand=False), format="%d%b%y", cache=cache
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "format, expected",
+        [
+            ("%Y-%m-%d", Timestamp(2000, 1, 3)),
+            ("%Y-%d-%m", Timestamp(2000, 3, 1)),
+            ("%Y-%m-%d %H", Timestamp(2000, 1, 3, 12)),
+            ("%Y-%d-%m %H", Timestamp(2000, 3, 1, 12)),
+            ("%Y-%m-%d %H:%M", Timestamp(2000, 1, 3, 12, 34)),
+            ("%Y-%d-%m %H:%M", Timestamp(2000, 3, 1, 12, 34)),
+            ("%Y-%m-%d %H:%M:%S", Timestamp(2000, 1, 3, 12, 34, 56)),
+            ("%Y-%d-%m %H:%M:%S", Timestamp(2000, 3, 1, 12, 34, 56)),
+            ("%Y-%m-%d %H:%M:%S.%f", Timestamp(2000, 1, 3, 12, 34, 56, 123456)),
+            ("%Y-%d-%m %H:%M:%S.%f", Timestamp(2000, 3, 1, 12, 34, 56, 123456)),
+            (
+                "%Y-%m-%d %H:%M:%S.%f%z",
+                Timestamp(2000, 1, 3, 12, 34, 56, 123456, tz="UTC+01:00"),
+            ),
+            (
+                "%Y-%d-%m %H:%M:%S.%f%z",
+                Timestamp(2000, 3, 1, 12, 34, 56, 123456, tz="UTC+01:00"),
+            ),
+        ],
+    )
+    def test_non_exact_doesnt_parse_whole_string(self, cache, format, expected):
+        # https://github.com/pandas-dev/pandas/issues/50412
+        # the formats alternate between ISO8601 and non-ISO8601 to check both paths
+        result = to_datetime(
+            "2000-01-03 12:34:56.123456+01:00", format=format, exact=False
+        )
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "arg",
+        [
+            "2012-01-01 09:00:00.000000001",
+            "2012-01-01 09:00:00.000001",
+            "2012-01-01 09:00:00.001",
+            "2012-01-01 09:00:00.001000",
+            "2012-01-01 09:00:00.001000000",
+        ],
+    )
+    def test_parse_nanoseconds_with_formula(self, cache, arg):
+        # GH8989
+        # truncating the nanoseconds when a format was provided
+        expected = to_datetime(arg, cache=cache)
+        result = to_datetime(arg, format="%Y-%m-%d %H:%M:%S.%f", cache=cache)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "value,fmt,expected",
+        [
+            ["2009324", "%Y%W%w", "2009-08-13"],
+            ["2013020", "%Y%U%w", "2013-01-13"],
+        ],
+    )
+    def test_to_datetime_format_weeks(self, value, fmt, expected, cache):
+        assert to_datetime(value, format=fmt, cache=cache) == Timestamp(expected)
+
+    @pytest.mark.parametrize(
+        "fmt,dates,expected_dates",
+        [
+            [
+                "%Y-%m-%d %H:%M:%S %Z",
+                ["2010-01-01 12:00:00 UTC"] * 2,
+                [Timestamp("2010-01-01 12:00:00", tz="UTC")] * 2,
+            ],
+            [
+                "%Y-%m-%d %H:%M:%S%z",
+                ["2010-01-01 12:00:00+0100"] * 2,
+                [
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=60))
+                    )
+                ]
+                * 2,
+            ],
+            [
+                "%Y-%m-%d %H:%M:%S %z",
+                ["2010-01-01 12:00:00 +0100"] * 2,
+                [
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=60))
+                    )
+                ]
+                * 2,
+            ],
+            [
+                "%Y-%m-%d %H:%M:%S %z",
+                ["2010-01-01 12:00:00 Z", "2010-01-01 12:00:00 Z"],
+                [
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=0))
+                    ),
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=0))
+                    ),
+                ],
+            ],
+        ],
+    )
+    def test_to_datetime_parse_tzname_or_tzoffset(self, fmt, dates, expected_dates):
+        # GH 13486
+        result = to_datetime(dates, format=fmt)
+        expected = Index(expected_dates)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "fmt,dates,expected_dates",
+        [
+            [
+                "%Y-%m-%d %H:%M:%S %Z",
+                [
+                    "2010-01-01 12:00:00 UTC",
+                    "2010-01-01 12:00:00 GMT",
+                    "2010-01-01 12:00:00 US/Pacific",
+                ],
+                [
+                    Timestamp("2010-01-01 12:00:00", tz="UTC"),
+                    Timestamp("2010-01-01 12:00:00", tz="GMT"),
+                    Timestamp("2010-01-01 12:00:00", tz="US/Pacific"),
+                ],
+            ],
+            [
+                "%Y-%m-%d %H:%M:%S %z",
+                ["2010-01-01 12:00:00 +0100", "2010-01-01 12:00:00 -0100"],
+                [
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=60))
+                    ),
+                    Timestamp(
+                        "2010-01-01 12:00:00", tzinfo=timezone(timedelta(minutes=-60))
+                    ),
+                ],
+            ],
+        ],
+    )
+    def test_to_datetime_parse_tzname_or_tzoffset_utc_false_removed(
+        self, fmt, dates, expected_dates
+    ):
+        # GH#13486, GH#50887, GH#57275
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(dates, format=fmt)
+
+    def test_to_datetime_parse_tzname_or_tzoffset_different_tz_to_utc(self):
+        # GH 32792
+        dates = [
+            "2010-01-01 12:00:00 +0100",
+            "2010-01-01 12:00:00 -0100",
+            "2010-01-01 12:00:00 +0300",
+            "2010-01-01 12:00:00 +0400",
+        ]
+        expected_dates = [
+            "2010-01-01 11:00:00+00:00",
+            "2010-01-01 13:00:00+00:00",
+            "2010-01-01 09:00:00+00:00",
+            "2010-01-01 08:00:00+00:00",
+        ]
+        fmt = "%Y-%m-%d %H:%M:%S %z"
+
+        result = to_datetime(dates, format=fmt, utc=True)
+        expected = DatetimeIndex(expected_dates)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "offset", ["+0", "-1foo", "UTCbar", ":10", "+01:000:01", ""]
+    )
+    def test_to_datetime_parse_timezone_malformed(self, offset):
+        fmt = "%Y-%m-%d %H:%M:%S %z"
+        date = "2010-01-01 12:00:00 " + offset
+
+        msg = "|".join(
+            [
+                r'^time data ".*" doesn\'t match format ".*". ' f"{PARSING_ERR_MSG}$",
+                r'^unconverted data remains when parsing with format ".*": ".*". '
+                f"{PARSING_ERR_MSG}$",
+            ]
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([date], format=fmt)
+
+    def test_to_datetime_parse_timezone_keeps_name(self):
+        # GH 21697
+        fmt = "%Y-%m-%d %H:%M:%S %z"
+        arg = Index(["2010-01-01 12:00:00 Z"], name="foo")
+        result = to_datetime(arg, format=fmt)
+        expected = DatetimeIndex(["2010-01-01 12:00:00"], tz="UTC", name="foo")
+        tm.assert_index_equal(result, expected)
+
+
+class TestToDatetime:
+    @pytest.mark.filterwarnings("ignore:Could not infer format")
+    def test_to_datetime_overflow(self):
+        # we should get an OutOfBoundsDatetime, NOT OverflowError
+        # TODO: Timestamp raises ValueError("could not convert string to Timestamp")
+        #  can we make these more consistent?
+        arg = "08335394550"
+        msg = 'Parsing "08335394550" to datetime overflows'
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(arg)
+
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime([arg])
+
+        res = to_datetime(arg, errors="coerce")
+        assert res is NaT
+        res = to_datetime([arg], errors="coerce")
+        exp = Index([NaT], dtype="M8[s]")
+        tm.assert_index_equal(res, exp)
+
+    def test_to_datetime_mixed_datetime_and_string(self):
+        # GH#47018 adapted old doctest with new behavior
+        d1 = datetime(2020, 1, 1, 17, tzinfo=timezone(-timedelta(hours=1)))
+        d2 = datetime(2020, 1, 1, 18, tzinfo=timezone(-timedelta(hours=1)))
+        res = to_datetime(["2020-01-01 17:00 -0100", d2])
+        expected = to_datetime([d1, d2]).tz_convert(timezone(timedelta(minutes=-60)))
+        tm.assert_index_equal(res, expected)
+
+    def test_to_datetime_mixed_string_and_numeric(self):
+        # GH#55780 np.array(vals) would incorrectly cast the number to str
+        vals = ["2016-01-01", 0]
+        expected = DatetimeIndex([Timestamp(x) for x in vals])
+        result = to_datetime(vals, format="mixed")
+        result2 = to_datetime(vals[::-1], format="mixed")[::-1]
+        result3 = DatetimeIndex(vals)
+        result4 = DatetimeIndex(vals[::-1])[::-1]
+
+        tm.assert_index_equal(result, expected)
+        tm.assert_index_equal(result2, expected)
+        tm.assert_index_equal(result3, expected)
+        tm.assert_index_equal(result4, expected)
+
+    @pytest.mark.parametrize(
+        "format", ["%Y-%m-%d", "%Y-%d-%m"], ids=["ISO8601", "non-ISO8601"]
+    )
+    def test_to_datetime_mixed_date_and_string(self, format):
+        # https://github.com/pandas-dev/pandas/issues/50108
+        d1 = date(2020, 1, 2)
+        res = to_datetime(["2020-01-01", d1], format=format)
+        expected = DatetimeIndex(["2020-01-01", "2020-01-02"], dtype="M8[s]")
+        tm.assert_index_equal(res, expected)
+
+    @pytest.mark.parametrize(
+        "fmt",
+        ["%Y-%d-%m %H:%M:%S%z", "%Y-%m-%d %H:%M:%S%z"],
+        ids=["non-ISO8601 format", "ISO8601 format"],
+    )
+    @pytest.mark.parametrize(
+        "utc, args, expected",
+        [
+            pytest.param(
+                True,
+                ["2000-01-01 01:00:00-08:00", "2000-01-01 02:00:00-08:00"],
+                DatetimeIndex(
+                    ["2000-01-01 09:00:00+00:00", "2000-01-01 10:00:00+00:00"],
+                    dtype="datetime64[us, UTC]",
+                ),
+                id="all tz-aware, with utc",
+            ),
+            pytest.param(
+                False,
+                ["2000-01-01 01:00:00+00:00", "2000-01-01 02:00:00+00:00"],
+                DatetimeIndex(
+                    ["2000-01-01 01:00:00+00:00", "2000-01-01 02:00:00+00:00"],
+                ).as_unit("us"),
+                id="all tz-aware, without utc",
+            ),
+            pytest.param(
+                True,
+                ["2000-01-01 01:00:00-08:00", "2000-01-01 02:00:00+00:00"],
+                DatetimeIndex(
+                    ["2000-01-01 09:00:00+00:00", "2000-01-01 02:00:00+00:00"],
+                    dtype="datetime64[us, UTC]",
+                ),
+                id="all tz-aware, mixed offsets, with utc",
+            ),
+            pytest.param(
+                True,
+                ["2000-01-01 01:00:00", "2000-01-01 02:00:00+00:00"],
+                DatetimeIndex(
+                    ["2000-01-01 01:00:00+00:00", "2000-01-01 02:00:00+00:00"],
+                    dtype="datetime64[us, UTC]",
+                ),
+                id="tz-aware string, naive pydatetime, with utc",
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "constructor",
+        [Timestamp, lambda x: Timestamp(x).to_pydatetime()],
+    )
+    def test_to_datetime_mixed_datetime_and_string_with_format(
+        self, fmt, utc, args, expected, constructor
+    ):
+        # https://github.com/pandas-dev/pandas/issues/49298
+        # https://github.com/pandas-dev/pandas/issues/50254
+        # note: ISO8601 formats go down a fastpath, so we need to check both
+        # a ISO8601 format and a non-ISO8601 one
+        ts1 = constructor(args[0])
+        ts2 = args[1]
+        result = to_datetime([ts1, ts2], format=fmt, utc=utc)
+        if constructor is Timestamp:
+            expected = expected.as_unit("s")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "fmt",
+        ["%Y-%d-%m %H:%M:%S%z", "%Y-%m-%d %H:%M:%S%z"],
+        ids=["non-ISO8601 format", "ISO8601 format"],
+    )
+    @pytest.mark.parametrize(
+        "constructor",
+        [Timestamp, lambda x: Timestamp(x).to_pydatetime()],
+    )
+    def test_to_datetime_mixed_dt_and_str_with_format_mixed_offsets_utc_false_removed(
+        self, fmt, constructor
+    ):
+        # https://github.com/pandas-dev/pandas/issues/49298
+        # https://github.com/pandas-dev/pandas/issues/50254
+        # GH#57275
+        # note: ISO8601 formats go down a fastpath, so we need to check both
+        # a ISO8601 format and a non-ISO8601 one
+        args = ["2000-01-01 01:00:00", "2000-01-01 02:00:00+00:00"]
+        ts1 = constructor(args[0])
+        ts2 = args[1]
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([ts1, ts2], format=fmt, utc=False)
+
+    @pytest.mark.parametrize(
+        "fmt, expected",
+        [
+            pytest.param(
+                "%Y-%m-%d %H:%M:%S%z",
+                [
+                    Timestamp("2000-01-01 09:00:00+0100", tz="UTC+01:00"),
+                    Timestamp("2000-01-02 02:00:00+0200", tz="UTC+02:00"),
+                    NaT,
+                ],
+                id="ISO8601, non-UTC",
+            ),
+            pytest.param(
+                "%Y-%d-%m %H:%M:%S%z",
+                [
+                    Timestamp("2000-01-01 09:00:00+0100", tz="UTC+01:00"),
+                    Timestamp("2000-02-01 02:00:00+0200", tz="UTC+02:00"),
+                    NaT,
+                ],
+                id="non-ISO8601, non-UTC",
+            ),
+        ],
+    )
+    def test_to_datetime_mixed_offsets_with_none_tz_utc_false_removed(
+        self, fmt, expected
+    ):
+        # https://github.com/pandas-dev/pandas/issues/50071
+        # GH#57275
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(
+                ["2000-01-01 09:00:00+01:00", "2000-01-02 02:00:00+02:00", None],
+                format=fmt,
+                utc=False,
+            )
+
+    @pytest.mark.parametrize(
+        "fmt, expected",
+        [
+            pytest.param(
+                "%Y-%m-%d %H:%M:%S%z",
+                DatetimeIndex(
+                    ["2000-01-01 08:00:00+00:00", "2000-01-02 00:00:00+00:00", "NaT"],
+                    dtype="datetime64[s, UTC]",
+                ),
+                id="ISO8601, UTC",
+            ),
+            pytest.param(
+                "%Y-%d-%m %H:%M:%S%z",
+                DatetimeIndex(
+                    ["2000-01-01 08:00:00+00:00", "2000-02-01 00:00:00+00:00", "NaT"],
+                    dtype="datetime64[s, UTC]",
+                ),
+                id="non-ISO8601, UTC",
+            ),
+        ],
+    )
+    def test_to_datetime_mixed_offsets_with_none(self, fmt, expected):
+        # https://github.com/pandas-dev/pandas/issues/50071
+        result = to_datetime(
+            ["2000-01-01 09:00:00+01:00", "2000-01-02 02:00:00+02:00", None],
+            format=fmt,
+            utc=True,
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "fmt",
+        ["%Y-%d-%m %H:%M:%S%z", "%Y-%m-%d %H:%M:%S%z"],
+        ids=["non-ISO8601 format", "ISO8601 format"],
+    )
+    @pytest.mark.parametrize(
+        "args",
+        [
+            pytest.param(
+                ["2000-01-01 01:00:00-08:00", "2000-01-01 02:00:00-07:00"],
+                id="all tz-aware, mixed timezones, without utc",
+            ),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "constructor",
+        [Timestamp, lambda x: Timestamp(x).to_pydatetime()],
+    )
+    def test_to_datetime_mixed_datetime_and_string_with_format_raises(
+        self, fmt, args, constructor
+    ):
+        # https://github.com/pandas-dev/pandas/issues/49298
+        # note: ISO8601 formats go down a fastpath, so we need to check both
+        # a ISO8601 format and a non-ISO8601 one
+        ts1 = constructor(args[0])
+        ts2 = constructor(args[1])
+        with pytest.raises(
+            ValueError, match="cannot be converted to datetime64 unless utc=True"
+        ):
+            to_datetime([ts1, ts2], format=fmt, utc=False)
+
+    def test_to_datetime_np_str(self):
+        # GH#32264
+        # GH#48969
+        value = np.str_("2019-02-04 10:18:46.297000+0000")
+
+        ser = Series([value])
+
+        exp = Timestamp("2019-02-04 10:18:46.297000", tz="UTC")
+
+        assert to_datetime(value) == exp
+        assert to_datetime(ser.iloc[0]) == exp
+
+        res = to_datetime([value])
+        expected = Index([exp])
+        tm.assert_index_equal(res, expected)
+
+        res = to_datetime(ser)
+        expected = Series(expected)
+        tm.assert_series_equal(res, expected)
+
+    @pytest.mark.parametrize(
+        "s, _format, dt",
+        [
+            ["2015-1-1", "%G-%V-%u", datetime(2014, 12, 29, 0, 0)],
+            ["2015-1-4", "%G-%V-%u", datetime(2015, 1, 1, 0, 0)],
+            ["2015-1-7", "%G-%V-%u", datetime(2015, 1, 4, 0, 0)],
+            ["2024-52-1", "%G-%V-%u", datetime(2024, 12, 23, 0, 0)],
+            ["2024-52-7", "%G-%V-%u", datetime(2024, 12, 29, 0, 0)],
+            ["2025-1-1", "%G-%V-%u", datetime(2024, 12, 30, 0, 0)],
+            ["2020-53-1", "%G-%V-%u", datetime(2020, 12, 28, 0, 0)],
+        ],
+    )
+    def test_to_datetime_iso_week_year_format(self, s, _format, dt):
+        # See GH#16607
+        assert to_datetime(s, format=_format) == dt
+
+    @pytest.mark.parametrize(
+        "msg, s, _format",
+        [
+            [
+                "Week 53 does not exist in ISO year 2024",
+                "2024 53 1",
+                "%G %V %u",
+            ],
+            [
+                "Week 53 does not exist in ISO year 2023",
+                "2023 53 1",
+                "%G %V %u",
+            ],
+        ],
+    )
+    def test_invalid_iso_week_53(self, msg, s, _format):
+        # See GH#60885
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(s, format=_format)
+
+    @pytest.mark.parametrize(
+        "msg, s, _format",
+        [
+            [
+                "ISO week directive '%V' is incompatible with the year directive "
+                "'%Y'. Use the ISO year '%G' instead.",
+                "1999 50",
+                "%Y %V",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 51",
+                "%G %V",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 Monday",
+                "%G %A",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 Mon",
+                "%G %a",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 6",
+                "%G %w",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 6",
+                "%G %u",
+            ],
+            [
+                "ISO year directive '%G' must be used with the ISO week directive "
+                "'%V' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "2051",
+                "%G",
+            ],
+            [
+                "Day of the year directive '%j' is not compatible with ISO year "
+                "directive '%G'. Use '%Y' instead.",
+                "1999 51 6 256",
+                "%G %V %u %j",
+            ],
+            [
+                "ISO week directive '%V' is incompatible with the year directive "
+                "'%Y'. Use the ISO year '%G' instead.",
+                "1999 51 Sunday",
+                "%Y %V %A",
+            ],
+            [
+                "ISO week directive '%V' is incompatible with the year directive "
+                "'%Y'. Use the ISO year '%G' instead.",
+                "1999 51 Sun",
+                "%Y %V %a",
+            ],
+            [
+                "ISO week directive '%V' is incompatible with the year directive "
+                "'%Y'. Use the ISO year '%G' instead.",
+                "1999 51 1",
+                "%Y %V %w",
+            ],
+            [
+                "ISO week directive '%V' is incompatible with the year directive "
+                "'%Y'. Use the ISO year '%G' instead.",
+                "1999 51 1",
+                "%Y %V %u",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "20",
+                "%V",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 51 Sunday",
+                "%V %A",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 51 Sun",
+                "%V %a",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 51 1",
+                "%V %w",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "1999 51 1",
+                "%V %u",
+            ],
+            [
+                "Day of the year directive '%j' is not compatible with ISO year "
+                "directive '%G'. Use '%Y' instead.",
+                "1999 50",
+                "%G %j",
+            ],
+            [
+                "ISO week directive '%V' must be used with the ISO year directive "
+                "'%G' and a weekday directive '%A', '%a', '%w', or '%u'.",
+                "20 Monday",
+                "%V %A",
+            ],
+        ],
+    )
+    @pytest.mark.parametrize("errors", ["raise", "coerce"])
+    def test_error_iso_week_year(self, msg, s, _format, errors):
+        # See GH#16607, GH#50308
+        # This test checks for errors thrown when giving the wrong format
+        # However, as discussed on PR#25541, overriding the locale
+        # causes a different error to be thrown due to the format being
+        # locale specific, but the test data is in english.
+        # Therefore, the tests only run when locale is not overwritten,
+        # as a sort of solution to this problem.
+        if locale.getlocale() != ("zh_CN", "UTF-8") and locale.getlocale() != (
+            "it_IT",
+            "UTF-8",
+        ):
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(s, format=_format, errors=errors)
+
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_to_datetime_dtarr(self, tz):
+        # DatetimeArray
+        dti = date_range("1965-04-03", periods=19, freq="2W", tz=tz)
+        arr = dti._data
+
+        result = to_datetime(arr)
+        assert result is arr
+
+    # Doesn't work on Windows since tzpath not set correctly
+    @td.skip_if_windows
+    @pytest.mark.parametrize("utc", [True, False])
+    @pytest.mark.parametrize("tz", [None, "US/Central"])
+    def test_to_datetime_arrow(self, tz, utc, index_or_series):
+        pa = pytest.importorskip("pyarrow")
+
+        dti = date_range("1965-04-03", periods=19, freq="2W", tz=tz)
+        dti = index_or_series(dti)
+
+        dti_arrow = dti.astype(pd.ArrowDtype(pa.timestamp(unit="ns", tz=tz)))
+
+        result = to_datetime(dti_arrow, utc=utc)
+        expected = to_datetime(dti, utc=utc).astype(
+            pd.ArrowDtype(pa.timestamp(unit="ns", tz=tz if not utc else "UTC"))
+        )
+        if not utc and index_or_series is not Series:
+            # Doesn't hold for utc=True, since that will astype
+            # to_datetime also returns a new object for series
+            assert result is dti_arrow
+        if index_or_series is Series:
+            tm.assert_series_equal(result, expected)
+        else:
+            tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_pydatetime(self):
+        actual = to_datetime(datetime(2008, 1, 15))
+        assert actual == datetime(2008, 1, 15)
+
+    def test_to_datetime_YYYYMMDD(self):
+        actual = to_datetime("20080115")
+        assert actual == datetime(2008, 1, 15)
+
+    @td.skip_if_windows  # `tm.set_timezone` does not work in windows
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    def test_to_datetime_now(self):
+        # See GH#18666
+        with tm.set_timezone("US/Eastern"):
+            # GH#18705
+            now = Timestamp("now")
+            pdnow = to_datetime("now")
+            pdnow2 = to_datetime(["now"])[0]
+
+            # These should all be equal with infinite perf; this gives
+            # a generous margin of 10 seconds
+            assert abs(pdnow._value - now._value) < 1e10
+            assert abs(pdnow2._value - now._value) < 1e10
+
+            assert pdnow.tzinfo is None
+            assert pdnow2.tzinfo is None
+
+    @td.skip_if_windows  # `tm.set_timezone` does not work on Windows
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    @pytest.mark.parametrize("tz", ["Pacific/Auckland", "US/Samoa"])
+    def test_to_datetime_today(self, tz):
+        # See GH#18666
+        # Test with one timezone far ahead of UTC and another far behind, so
+        # one of these will _almost_ always be in a different day from UTC.
+        # Unfortunately this test between 12 and 1 AM Samoa time
+        # this both of these timezones _and_ UTC will all be in the same day,
+        # so this test will not detect the regression introduced in #18666.
+        with tm.set_timezone(tz):
+            nptoday = np.datetime64("today").astype("datetime64[us]").astype(np.int64)
+            pdtoday = to_datetime("today")
+            pdtoday2 = to_datetime(["today"])[0]
+
+            tstoday = Timestamp("today")
+            tstoday2 = Timestamp.today()
+
+            # These should all be equal with infinite perf; this gives
+            # a generous margin of 10 seconds
+            assert abs(pdtoday.normalize()._value - nptoday) < 1e10
+            assert abs(pdtoday2.normalize()._value - nptoday) < 1e10
+            assert abs(pdtoday._value - tstoday._value) < 1e10
+            assert abs(pdtoday._value - tstoday2._value) < 1e10
+
+            assert pdtoday.tzinfo is None
+            assert pdtoday2.tzinfo is None
+
+    @pytest.mark.parametrize("arg", ["now", "today"])
+    def test_to_datetime_today_now_unicode_bytes(self, arg):
+        to_datetime([arg])
+
+    @pytest.mark.filterwarnings("ignore:Timestamp.utcnow is deprecated:FutureWarning")
+    @pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+    @pytest.mark.parametrize(
+        "format, expected_ds",
+        [
+            ("%Y-%m-%d %H:%M:%S%z", "2020-01-03"),
+            ("%Y-%d-%m %H:%M:%S%z", "2020-03-01"),
+            (None, "2020-01-03"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "string, attribute",
+        [
+            ("now", "utcnow"),
+            ("today", "today"),
+        ],
+    )
+    def test_to_datetime_now_with_format(self, format, expected_ds, string, attribute):
+        # https://github.com/pandas-dev/pandas/issues/50359
+        result = to_datetime(["2020-01-03 00:00:00Z", string], format=format, utc=True)
+        expected = DatetimeIndex(
+            [expected_ds, getattr(Timestamp, attribute)()], dtype="datetime64[s, UTC]"
+        )
+        assert (expected - result).max().total_seconds() < 1
+
+    @pytest.mark.parametrize(
+        "dt", [np.datetime64("2000-01-01"), np.datetime64("2000-01-02")]
+    )
+    def test_to_datetime_dt64s(self, cache, dt):
+        assert to_datetime(dt, cache=cache) == Timestamp(dt)
+
+    @pytest.mark.parametrize(
+        "arg, format",
+        [
+            ("2001-01-01", "%Y-%m-%d"),
+            ("01-01-2001", "%d-%m-%Y"),
+        ],
+    )
+    def test_to_datetime_dt64s_and_str(self, arg, format):
+        # https://github.com/pandas-dev/pandas/issues/50036
+        result = to_datetime([arg, np.datetime64("2020-01-01")], format=format)
+        expected = DatetimeIndex(["2001-01-01", "2020-01-01"])
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dt", [np.datetime64("1000-01-01"), np.datetime64("5000-01-02")]
+    )
+    @pytest.mark.parametrize("errors", ["raise", "coerce"])
+    def test_to_datetime_dt64s_out_of_ns_bounds(self, cache, dt, errors):
+        # GH#50369 We cast to the nearest supported reso, i.e. "s"
+        ts = to_datetime(dt, errors=errors, cache=cache)
+        assert isinstance(ts, Timestamp)
+        assert ts.unit == "s"
+        assert ts.asm8 == dt
+
+        ts = Timestamp(dt)
+        assert ts.unit == "s"
+        assert ts.asm8 == dt
+
+    @pytest.mark.skip_ubsan
+    def test_to_datetime_dt64d_out_of_bounds(self, cache):
+        dt64 = np.datetime64(np.iinfo(np.int64).max, "D")
+
+        msg = "Out of bounds second timestamp: 25252734927768524-07-27"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            Timestamp(dt64)
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(dt64, errors="raise", cache=cache)
+
+        assert to_datetime(dt64, errors="coerce", cache=cache) is NaT
+
+    @pytest.mark.parametrize("unit", ["s", "D"])
+    def test_to_datetime_array_of_dt64s(self, cache, unit):
+        # https://github.com/pandas-dev/pandas/issues/31491
+        # Need at least 50 to ensure cache is used.
+        dts = [
+            np.datetime64("2000-01-01", unit),
+            np.datetime64("2000-01-02", unit),
+        ] * 30
+        # Assuming all datetimes are in bounds, to_datetime() returns
+        # an array that is equal to Timestamp() parsing
+        result = to_datetime(dts, cache=cache)
+        expected = DatetimeIndex([Timestamp(x).asm8 for x in dts], dtype="M8[s]")
+
+        tm.assert_index_equal(result, expected)
+
+        # A list of datetimes where the last one is out of bounds
+        dts_with_oob = dts + [np.datetime64("9999-01-01")]
+
+        # As of GH#51978 we do not raise in this case
+        to_datetime(dts_with_oob, errors="raise")
+
+        result = to_datetime(dts_with_oob, errors="coerce", cache=cache)
+        expected = DatetimeIndex(np.array(dts_with_oob, dtype="M8[s]"))
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_tz(self, cache):
+        # xref 8260
+        # uniform returns a DatetimeIndex
+        arr = [
+            Timestamp("2013-01-01 13:00:00-0800", tz="US/Pacific"),
+            Timestamp("2013-01-02 14:00:00-0800", tz="US/Pacific"),
+        ]
+        result = to_datetime(arr, cache=cache)
+        expected = DatetimeIndex(
+            ["2013-01-01 13:00:00", "2013-01-02 14:00:00"], tz="US/Pacific"
+        ).as_unit("s")
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_tz_mixed(self, cache):
+        # mixed tzs will raise if errors='raise'
+        # https://github.com/pandas-dev/pandas/issues/50585
+        arr = [
+            Timestamp("2013-01-01 13:00:00", tz="US/Pacific"),
+            Timestamp("2013-01-02 14:00:00", tz="US/Eastern"),
+        ]
+        msg = (
+            "Tz-aware datetime.datetime cannot be "
+            "converted to datetime64 unless utc=True"
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(arr, cache=cache)
+
+        result = to_datetime(arr, cache=cache, errors="coerce")
+        expected = DatetimeIndex(
+            ["2013-01-01 13:00:00-08:00", "NaT"], dtype="datetime64[s, US/Pacific]"
+        )
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_different_offsets_removed(self, cache):
+        # inspired by asv timeseries.ToDatetimeNONISO8601 benchmark
+        # see GH-26097 for more
+        # GH#57275
+        ts_string_1 = "March 1, 2018 12:00:00+0400"
+        ts_string_2 = "March 1, 2018 12:00:00+0500"
+        arr = [ts_string_1] * 5 + [ts_string_2] * 5
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(arr, cache=cache)
+
+    def test_to_datetime_tz_pytz(self, cache):
+        # see gh-8260
+        pytz = pytest.importorskip("pytz")
+        us_eastern = pytz.timezone("US/Eastern")
+        arr = np.array(
+            [
+                us_eastern.localize(
+                    datetime(year=2000, month=1, day=1, hour=3, minute=0)
+                ),
+                us_eastern.localize(
+                    datetime(year=2000, month=6, day=1, hour=3, minute=0)
+                ),
+            ],
+            dtype=object,
+        )
+        result = to_datetime(arr, utc=True, cache=cache)
+        expected = DatetimeIndex(
+            ["2000-01-01 08:00:00+00:00", "2000-06-01 07:00:00+00:00"],
+            dtype="datetime64[us, UTC]",
+            freq=None,
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "init_constructor, end_constructor",
+        [
+            (Index, DatetimeIndex),
+            (list, DatetimeIndex),
+            (np.array, DatetimeIndex),
+            (Series, Series),
+        ],
+    )
+    def test_to_datetime_utc_true(self, cache, init_constructor, end_constructor):
+        # See gh-11934 & gh-6415
+        data = ["20100102 121314", "20100102 121315"]
+        expected_data = [
+            Timestamp("2010-01-02 12:13:14", tz="utc"),
+            Timestamp("2010-01-02 12:13:15", tz="utc"),
+        ]
+
+        result = to_datetime(
+            init_constructor(data), format="%Y%m%d %H%M%S", utc=True, cache=cache
+        )
+        expected = end_constructor(expected_data)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "scalar, expected",
+        [
+            ["20100102 121314", Timestamp("2010-01-02 12:13:14", tz="utc")],
+            ["20100102 121315", Timestamp("2010-01-02 12:13:15", tz="utc")],
+        ],
+    )
+    def test_to_datetime_utc_true_scalar(self, cache, scalar, expected):
+        # Test scalar case as well
+        result = to_datetime(scalar, format="%Y%m%d %H%M%S", utc=True, cache=cache)
+        assert result == expected
+
+    def test_to_datetime_utc_true_with_series_single_value(self, cache):
+        # GH 15760 UTC=True with Series
+        ts = 1.5e18
+        result = to_datetime(Series([ts]), utc=True, cache=cache)
+        expected = Series([Timestamp(ts, tz="utc")])
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_utc_true_with_series_tzaware_string(self, cache):
+        ts = "2013-01-01 00:00:00-01:00"
+        expected_ts = "2013-01-01 01:00:00"
+        data = Series([ts] * 3)
+        result = to_datetime(data, utc=True, cache=cache)
+        expected = Series([Timestamp(expected_ts, tz="utc")] * 3)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "date, dtype",
+        [
+            ("2013-01-01 01:00:00", "datetime64[ns]"),
+            ("2013-01-01 01:00:00", "datetime64[ns, UTC]"),
+        ],
+    )
+    def test_to_datetime_utc_true_with_series_datetime_ns(self, cache, date, dtype):
+        expected = Series(
+            [Timestamp("2013-01-01 01:00:00", tz="UTC")], dtype="M8[ns, UTC]"
+        )
+        result = to_datetime(Series([date], dtype=dtype), utc=True, cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_tz_psycopg2(self, request, cache):
+        # xref 8260
+        psycopg2_tz = pytest.importorskip("psycopg2.tz")
+
+        # misc cases
+        tz1 = psycopg2_tz.FixedOffsetTimezone(offset=-300, name=None)
+        tz2 = psycopg2_tz.FixedOffsetTimezone(offset=-240, name=None)
+        arr = np.array(
+            [
+                datetime(2000, 1, 1, 3, 0, tzinfo=tz1),
+                datetime(2000, 6, 1, 3, 0, tzinfo=tz2),
+            ],
+            dtype=object,
+        )
+
+        result = to_datetime(arr, errors="coerce", utc=True, cache=cache)
+        expected = DatetimeIndex(
+            ["2000-01-01 08:00:00+00:00", "2000-06-01 07:00:00+00:00"],
+            dtype="datetime64[us, UTC]",
+            freq=None,
+        )
+        tm.assert_index_equal(result, expected)
+
+        # dtype coercion
+        i = DatetimeIndex(
+            ["2000-01-01 08:00:00"],
+            tz=psycopg2_tz.FixedOffsetTimezone(offset=-300, name=None),
+        ).as_unit("us")
+        assert not is_datetime64_ns_dtype(i)
+
+        # tz coercion
+        result = to_datetime(i, errors="coerce", cache=cache)
+        tm.assert_index_equal(result, i)
+
+        result = to_datetime(i, errors="coerce", utc=True, cache=cache)
+        expected = DatetimeIndex(["2000-01-01 13:00:00"], dtype="datetime64[us, UTC]")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("arg", [True, False])
+    def test_datetime_bool(self, cache, arg):
+        # GH13176
+        msg = r"dtype bool cannot be converted to datetime64\[ns\]"
+        with pytest.raises(TypeError, match=msg):
+            to_datetime(arg)
+        assert to_datetime(arg, errors="coerce", cache=cache) is NaT
+
+    def test_datetime_bool_arrays_mixed(self, cache):
+        msg = f"{type(cache)} is not convertible to datetime"
+        with pytest.raises(TypeError, match=msg):
+            to_datetime([False, datetime.today()], cache=cache)
+        with pytest.raises(
+            ValueError,
+            match=(
+                r'^time data "True" doesn\'t match format "%Y%m%d". '
+                f"{PARSING_ERR_MSG}$"
+            ),
+        ):
+            to_datetime(["20130101", True], cache=cache)
+        tm.assert_index_equal(
+            to_datetime([0, False, NaT, 0.0], errors="coerce", cache=cache),
+            DatetimeIndex(
+                [to_datetime(0, cache=cache), NaT, NaT, to_datetime(0, cache=cache)]
+            ),
+        )
+
+    @pytest.mark.parametrize("arg", [bool, to_datetime])
+    def test_datetime_invalid_datatype(self, arg):
+        # GH13176
+        msg = "is not convertible to datetime"
+        with pytest.raises(TypeError, match=msg):
+            to_datetime(arg)
+
+    @pytest.mark.parametrize("errors", ["coerce", "raise"])
+    def test_invalid_format_raises(self, errors):
+        # https://github.com/pandas-dev/pandas/issues/50255
+        with pytest.raises(
+            ValueError, match="':' is a bad directive in format 'H%:M%:S%"
+        ):
+            to_datetime(["00:00:00"], format="H%:M%:S%", errors=errors)
+
+    @pytest.mark.parametrize("value", ["a", "00:01:99"])
+    @pytest.mark.parametrize("format", [None, "%H:%M:%S"])
+    def test_datetime_invalid_scalar(self, value, format):
+        # GH24763
+        res = to_datetime(value, errors="coerce", format=format)
+        assert res is NaT
+
+        msg = "|".join(
+            [
+                r'^time data "a" doesn\'t match format "%H:%M:%S". '
+                f"{PARSING_ERR_MSG}$",
+                r'^Given date string "a" not likely a datetime$',
+                r'^unconverted data remains when parsing with format "%H:%M:%S": "9". '
+                f"{PARSING_ERR_MSG}$",
+                r"^second must be in 0..59: 00:01:99$",
+            ]
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(value, errors="raise", format=format)
+
+    @pytest.mark.parametrize("value", ["3000/12/11 00:00:00"])
+    @pytest.mark.parametrize("format", [None, "%H:%M:%S"])
+    def test_datetime_outofbounds_scalar(self, value, format):
+        # GH24763
+        res = to_datetime(value, errors="coerce", format=format)
+        if format is None:
+            assert isinstance(res, Timestamp)
+            assert res == Timestamp(value)
+        else:
+            assert res is NaT
+
+        if format is not None:
+            msg = r'^time data ".*" doesn\'t match format ".*"'
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(value, errors="raise", format=format)
+        else:
+            res = to_datetime(value, errors="raise", format=format)
+            assert isinstance(res, Timestamp)
+            assert res == Timestamp(value)
+
+    @pytest.mark.parametrize(
+        ("values"), [(["a"]), (["00:01:99"]), (["a", "b", "99:00:00"])]
+    )
+    @pytest.mark.parametrize("format", [(None), ("%H:%M:%S")])
+    def test_datetime_invalid_index(self, values, format):
+        # GH24763
+        # Not great to have logic in tests, but this one's hard to
+        # parametrise over
+        if format is None and len(values) > 1:
+            warn = UserWarning
+        else:
+            warn = None
+
+        with tm.assert_produces_warning(
+            warn, match="Could not infer format", raise_on_extra_warnings=False
+        ):
+            res = to_datetime(values, errors="coerce", format=format)
+        tm.assert_index_equal(res, DatetimeIndex([NaT] * len(values)))
+
+        msg = "|".join(
+            [
+                r'^Given date string "a" not likely a datetime$',
+                r'^time data "a" doesn\'t match format "%H:%M:%S". '
+                f"{PARSING_ERR_MSG}$",
+                r'^unconverted data remains when parsing with format "%H:%M:%S": "9". '
+                f"{PARSING_ERR_MSG}$",
+                r"^second must be in 0..59: 00:01:99$",
+            ]
+        )
+        with pytest.raises(ValueError, match=msg):
+            with tm.assert_produces_warning(
+                warn, match="Could not infer format", raise_on_extra_warnings=False
+            ):
+                to_datetime(values, errors="raise", format=format)
+
+    @pytest.mark.parametrize("utc", [True, None])
+    @pytest.mark.parametrize("format", ["%Y%m%d %H:%M:%S", None])
+    @pytest.mark.parametrize("constructor", [list, tuple, np.array, Index, deque])
+    def test_to_datetime_cache(self, utc, format, constructor):
+        date = "20130101 00:00:00"
+        test_dates = [date] * 10**5
+        data = constructor(test_dates)
+
+        result = to_datetime(data, utc=utc, format=format, cache=True)
+        expected = to_datetime(data, utc=utc, format=format, cache=False)
+
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_from_deque(self):
+        # GH 29403
+        result = to_datetime(deque([Timestamp("2010-06-02 09:30:00")] * 51))
+        expected = to_datetime([Timestamp("2010-06-02 09:30:00")] * 51)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("utc", [True, None])
+    @pytest.mark.parametrize("format", ["%Y%m%d %H:%M:%S", None])
+    def test_to_datetime_cache_series(self, utc, format):
+        date = "20130101 00:00:00"
+        test_dates = [date] * 10**5
+        data = Series(test_dates)
+        result = to_datetime(data, utc=utc, format=format, cache=True)
+        expected = to_datetime(data, utc=utc, format=format, cache=False)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_cache_scalar(self):
+        date = "20130101 00:00:00"
+        result = to_datetime(date, cache=True)
+        expected = Timestamp("20130101 00:00:00")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "datetimelikes,expected_values,exp_unit",
+        (
+            (
+                (None, np.nan) + (NaT,) * start_caching_at,
+                (NaT,) * (start_caching_at + 2),
+                "s",
+            ),
+            (
+                (None, Timestamp("2012-07-26")) + (NaT,) * start_caching_at,
+                (NaT, Timestamp("2012-07-26")) + (NaT,) * start_caching_at,
+                "s",
+            ),
+            (
+                (None,)
+                + (NaT,) * start_caching_at
+                + ("2012 July 26", Timestamp("2012-07-26")),
+                (NaT,) * (start_caching_at + 1)
+                + (Timestamp("2012-07-26"), Timestamp("2012-07-26")),
+                "s",
+            ),
+        ),
+    )
+    def test_convert_object_to_datetime_with_cache(
+        self, datetimelikes, expected_values, exp_unit
+    ):
+        # GH#39882
+        ser = Series(
+            datetimelikes,
+            dtype="object",
+        )
+        result_series = to_datetime(ser, errors="coerce")
+        expected_series = Series(
+            expected_values,
+            dtype=f"datetime64[{exp_unit}]",
+        )
+        tm.assert_series_equal(result_series, expected_series)
+
+    @pytest.mark.parametrize(
+        "input",
+        [
+            Series([NaT] * 20 + [None] * 20, dtype="object"),
+            Series([NaT] * 60 + [None] * 60, dtype="object"),
+            Series([None] * 20),
+            Series([None] * 60),
+            Series([""] * 20),
+            Series([""] * 60),
+            Series([pd.NA] * 20),
+            Series([pd.NA] * 60),
+            Series([np.nan] * 20),
+            Series([np.nan] * 60),
+        ],
+    )
+    def test_to_datetime_converts_null_like_to_nat(self, cache, input):
+        # GH35888
+        expected = Series([NaT] * len(input), dtype="M8[s]")
+        result = to_datetime(input, cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "date, format",
+        [
+            ("2017-20", "%Y-%W"),
+            ("20 Sunday", "%W %A"),
+            ("20 Sun", "%W %a"),
+            ("2017-21", "%Y-%U"),
+            ("20 Sunday", "%U %A"),
+            ("20 Sun", "%U %a"),
+        ],
+    )
+    def test_week_without_day_and_calendar_year(self, date, format):
+        # GH16774
+
+        msg = "Cannot use '%W' or '%U' without day and year"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(date, format=format)
+
+    def test_to_datetime_coerce(self):
+        # GH#26122, GH#57275
+        ts_strings = [
+            "March 1, 2018 12:00:00+0400",
+            "March 1, 2018 12:00:00+0500",
+            "20100240",
+        ]
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(ts_strings, errors="coerce")
+
+    @pytest.mark.parametrize(
+        "string_arg, format",
+        [("March 1, 2018", "%B %d, %Y"), ("2018-03-01", "%Y-%m-%d")],
+    )
+    @pytest.mark.parametrize(
+        "outofbounds",
+        [
+            datetime(9999, 1, 1),
+            date(9999, 1, 1),
+            np.datetime64("9999-01-01"),
+            "January 1, 9999",
+            "9999-01-01",
+        ],
+    )
+    def test_to_datetime_coerce_oob(self, string_arg, format, outofbounds):
+        # https://github.com/pandas-dev/pandas/issues/50255
+        ts_strings = [string_arg, outofbounds]
+        result = to_datetime(ts_strings, errors="coerce", format=format)
+        if isinstance(outofbounds, str) and (
+            format.startswith("%B") ^ outofbounds.startswith("J")
+        ):
+            # the strings don't match the given format, so they raise and we coerce
+            expected = DatetimeIndex([datetime(2018, 3, 1), NaT], dtype="M8[s]")
+        elif isinstance(outofbounds, datetime):
+            expected = DatetimeIndex(
+                [datetime(2018, 3, 1), outofbounds], dtype="M8[us]"
+            )
+        else:
+            expected = DatetimeIndex([datetime(2018, 3, 1), outofbounds], dtype="M8[s]")
+        tm.assert_index_equal(result, expected)
+
+    def test_to_datetime_malformed_no_raise(self):
+        # GH 28299
+        # GH 48633
+        ts_strings = ["200622-12-31", "111111-24-11"]
+        with tm.assert_produces_warning(
+            UserWarning, match="Could not infer format", raise_on_extra_warnings=False
+        ):
+            result = to_datetime(ts_strings, errors="coerce")
+        # TODO: should Index get "s" by default here?
+        exp = Index([NaT, NaT], dtype="M8[s]")
+        tm.assert_index_equal(result, exp)
+
+    def test_to_datetime_malformed_raise(self):
+        # GH 48633
+        ts_strings = ["200622-12-31", "111111-24-11"]
+        msg = (
+            'Parsed string "200622-12-31" gives an invalid tzoffset, which must '
+            r"be between -timedelta\(hours=24\) and timedelta\(hours=24\)"
+        )
+        with pytest.raises(
+            ValueError,
+            match=msg,
+        ):
+            with tm.assert_produces_warning(
+                UserWarning, match="Could not infer format"
+            ):
+                to_datetime(
+                    ts_strings,
+                    errors="raise",
+                )
+
+    def test_iso_8601_strings_with_same_offset(self):
+        # GH 17697, 11736
+        ts_str = "2015-11-18 15:30:00+05:30"
+        result = to_datetime(ts_str)
+        expected = Timestamp(ts_str)
+        assert result == expected
+
+        expected = DatetimeIndex([Timestamp(ts_str)] * 2)
+        result = to_datetime([ts_str] * 2)
+        tm.assert_index_equal(result, expected)
+
+        result = DatetimeIndex([ts_str] * 2)
+        tm.assert_index_equal(result, expected)
+
+    def test_iso_8601_strings_with_different_offsets_removed(self):
+        # GH#17697, GH#11736, GH#50887, GH#57275
+        ts_strings = ["2015-11-18 15:30:00+05:30", "2015-11-18 16:30:00+06:30", NaT]
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(ts_strings)
+
+    def test_iso_8601_strings_with_different_offsets_utc(self):
+        ts_strings = ["2015-11-18 15:30:00+05:30", "2015-11-18 16:30:00+06:30", NaT]
+        result = to_datetime(ts_strings, utc=True)
+        expected = DatetimeIndex(
+            [Timestamp(2015, 11, 18, 10), Timestamp(2015, 11, 18, 10), NaT], tz="UTC"
+        ).as_unit("s")
+        tm.assert_index_equal(result, expected)
+
+    def test_mixed_offsets_with_native_datetime_utc_false_raises(self):
+        # GH#25978, GH#57275
+
+        vals = [
+            "nan",
+            Timestamp("1990-01-01"),
+            "2015-03-14T16:15:14.123-08:00",
+            "2019-03-04T21:56:32.620-07:00",
+            None,
+            "today",
+            "now",
+        ]
+        ser = Series(vals)
+        assert all(ser[i] is vals[i] for i in range(len(vals)))  # GH#40111
+
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(ser)
+
+    def test_non_iso_strings_with_tz_offset(self):
+        result = to_datetime(["March 1, 2018 12:00:00+0400"] * 2)
+        expected = DatetimeIndex(
+            [datetime(2018, 3, 1, 12, tzinfo=timezone(timedelta(minutes=240)))] * 2
+        ).as_unit("s")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ts, expected",
+        [
+            (Timestamp("2018-01-01"), Timestamp("2018-01-01", tz="UTC")),
+            (
+                Timestamp("2018-01-01", tz="US/Pacific"),
+                Timestamp("2018-01-01 08:00", tz="UTC"),
+            ),
+        ],
+    )
+    def test_timestamp_utc_true(self, ts, expected):
+        # GH 24415
+        result = to_datetime(ts, utc=True)
+        assert result == expected
+
+    @pytest.mark.parametrize("dt_str", ["00010101", "13000101", "30000101", "99990101"])
+    def test_to_datetime_with_format_out_of_bounds(self, dt_str):
+        # GH 9107
+        res = to_datetime(dt_str, format="%Y%m%d")
+        dtobj = datetime.strptime(dt_str, "%Y%m%d")
+        expected = Timestamp(dtobj).as_unit("s")
+        assert res == expected
+        assert res.unit == expected.unit
+
+    def test_to_datetime_utc(self):
+        arr = np.array([parse("2012-06-13T01:39:00Z")], dtype=object)
+
+        result = to_datetime(arr, utc=True)
+        assert result.tz is timezone.utc
+
+    def test_to_datetime_fixed_offset(self):
+        from pandas.tests.indexes.datetimes.test_timezones import FixedOffset
+
+        fixed_off = FixedOffset(-420, "-07:00")
+
+        dates = [
+            datetime(2000, 1, 1, tzinfo=fixed_off),
+            datetime(2000, 1, 2, tzinfo=fixed_off),
+            datetime(2000, 1, 3, tzinfo=fixed_off),
+        ]
+        result = to_datetime(dates)
+        assert result.tz == fixed_off
+
+    @pytest.mark.parametrize(
+        "date",
+        [
+            ["2020-10-26 00:00:00+06:00", "2020-10-26 00:00:00+01:00"],
+            ["2020-10-26 00:00:00+06:00", Timestamp("2018-01-01", tz="US/Pacific")],
+            [
+                "2020-10-26 00:00:00+06:00",
+                datetime(2020, 1, 1, 18).astimezone(
+                    zoneinfo.ZoneInfo("Australia/Melbourne")
+                ),
+            ],
+        ],
+    )
+    def test_to_datetime_mixed_offsets_with_utc_false_removed(self, date):
+        # GH#50887, GH#57275
+        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(date, utc=False)
+
+
+class TestToDatetimeUnit:
+    @pytest.mark.parametrize("unit", ["Y", "M"])
+    @pytest.mark.parametrize("item", [150, float(150)])
+    def test_to_datetime_month_or_year_unit_int(self, cache, unit, item, request):
+        # GH#50870 Note we have separate tests that pd.Timestamp gets these right
+        ts = Timestamp(item, unit=unit)
+        expected = DatetimeIndex([ts], dtype="M8[ns]")
+
+        result = to_datetime([item], unit=unit, cache=cache)
+        tm.assert_index_equal(result, expected)
+
+        result = to_datetime(np.array([item], dtype=object), unit=unit, cache=cache)
+        tm.assert_index_equal(result, expected)
+
+        result = to_datetime(np.array([item]), unit=unit, cache=cache)
+        tm.assert_index_equal(result, expected)
+
+        # with a nan!
+        result = to_datetime(np.array([item, np.nan]), unit=unit, cache=cache)
+        assert result.isna()[1]
+        tm.assert_index_equal(result[:1], expected)
+
+    @pytest.mark.parametrize("unit", ["Y", "M"])
+    def test_to_datetime_month_or_year_unit_non_round_float(self, cache, unit):
+        # GH#50301
+        # Match Timestamp behavior in disallowing non-round floats with
+        #  Y or M unit
+        msg = f"Conversion of non-round float with unit={unit} is ambiguous"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([1.5], unit=unit, errors="raise")
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(np.array([1.5]), unit=unit, errors="raise")
+
+        msg = r"Given date string \"1.5\" not likely a datetime"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(["1.5"], unit=unit, errors="raise")
+
+        res = to_datetime([1.5], unit=unit, errors="coerce")
+        expected = Index([NaT], dtype="M8[ns]")
+        tm.assert_index_equal(res, expected)
+
+        # In 3.0, the string "1.5" is parsed as as it would be without unit,
+        #  which fails. With errors="coerce" this becomes NaT.
+        res = to_datetime(["1.5"], unit=unit, errors="coerce")
+        expected = to_datetime([NaT]).as_unit("ns")
+        tm.assert_index_equal(res, expected)
+
+        # round floats are OK
+        res = to_datetime([1.0], unit=unit)
+        expected = to_datetime([1], unit=unit)
+        tm.assert_index_equal(res, expected)
+
+    def test_unit(self, cache):
+        # GH 11758
+        # test proper behavior with errors
+        msg = "cannot specify both format and unit"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([1], unit="D", format="%Y%m%d", cache=cache)
+
+    def test_unit_array_mixed_nans(self, cache):
+        values = [11111111111111111, 1, 1.0, iNaT, NaT, np.nan, "NaT", ""]
+
+        result = to_datetime(values, unit="D", errors="coerce", cache=cache)
+        expected = DatetimeIndex(
+            ["NaT", "1970-01-02", "1970-01-02", "NaT", "NaT", "NaT", "NaT", "NaT"],
+            dtype="M8[ns]",
+        )
+        tm.assert_index_equal(result, expected)
+
+        msg = "cannot convert input 11111111111111111 with the unit 'D'"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(values, unit="D", errors="raise", cache=cache)
+
+    def test_unit_array_mixed_nans_large_int(self, cache):
+        values = [1420043460000000000000000, iNaT, NaT, np.nan, "NaT"]
+
+        result = to_datetime(values, errors="coerce", unit="s", cache=cache)
+        expected = DatetimeIndex(["NaT", "NaT", "NaT", "NaT", "NaT"], dtype="M8[ns]")
+        tm.assert_index_equal(result, expected)
+
+        msg = "cannot convert input 1420043460000000000000000 with the unit 's'"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(values, errors="raise", unit="s", cache=cache)
+
+    def test_to_datetime_invalid_str_not_out_of_bounds_valuerror(self, cache):
+        # if we have a string, then we raise a ValueError
+        # and NOT an OutOfBoundsDatetime
+        msg = "Unknown datetime string format, unable to parse: foo"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime("foo", errors="raise", unit="s", cache=cache)
+
+    @pytest.mark.parametrize("error", ["raise", "coerce"])
+    def test_unit_consistency(self, cache, error):
+        # consistency of conversions
+        expected = Timestamp("1970-05-09 14:25:11")
+        result = to_datetime(11111111, unit="s", errors=error, cache=cache)
+        assert result == expected
+        assert isinstance(result, Timestamp)
+
+    @pytest.mark.parametrize("errors", ["raise", "coerce"])
+    @pytest.mark.parametrize("dtype", ["float64", "int64"])
+    def test_unit_with_numeric(self, cache, errors, dtype):
+        # GH 13180
+        # coercions from floats/ints are ok
+        expected = DatetimeIndex(
+            ["2015-06-19 05:33:20", "2015-05-27 22:33:20"], dtype="M8[ns]"
+        )
+        arr = np.array([1.434692e18, 1.432766e18]).astype(dtype)
+        result = to_datetime(arr, errors=errors, cache=cache)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "exp, arr, warning",
+        [
+            [
+                ["NaT", "2015-06-19 05:33:20", "2015-05-27 22:33:20"],
+                ["foo", 1.434692e18, 1.432766e18],
+                UserWarning,
+            ],
+            [
+                ["2015-06-19 05:33:20", "2015-05-27 22:33:20", "NaT", "NaT"],
+                [1.434692e18, 1.432766e18, "foo", "NaT"],
+                None,
+            ],
+        ],
+    )
+    def test_unit_with_numeric_coerce(self, cache, exp, arr, warning):
+        # but we want to make sure that we are coercing
+        # if we have ints/strings
+        expected = DatetimeIndex(exp, dtype="M8[ns]")
+        with tm.assert_produces_warning(warning, match="Could not infer format"):
+            result = to_datetime(arr, errors="coerce", cache=cache)
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "arr",
+        [
+            [Timestamp("20130101"), 1.434692e18, 1.432766e18],
+            [1.434692e18, 1.432766e18, Timestamp("20130101")],
+        ],
+    )
+    def test_unit_mixed(self, cache, arr):
+        # GH#50453 pre-2.0 with mixed numeric/datetimes and errors="coerce"
+        #  the numeric entries would be coerced to NaT, was never clear exactly
+        #  why.
+        # mixed integers/datetimes
+        expected = Index([Timestamp(x) for x in arr], dtype="M8[ns]")
+        result = to_datetime(arr, errors="coerce", cache=cache)
+        tm.assert_index_equal(result, expected)
+
+        # GH#49037 pre-2.0 this raised, but it always worked with Series,
+        #  was never clear why it was disallowed
+        result = to_datetime(arr, errors="raise", cache=cache)
+        tm.assert_index_equal(result, expected)
+
+        result = DatetimeIndex(arr)
+        tm.assert_index_equal(result, expected)
+
+    def test_unit_rounding(self, cache):
+        # GH 14156 & GH 20445: argument will incur floating point errors
+        # but no premature rounding
+        value = 1434743731.8770001
+        result = to_datetime(value, unit="s", cache=cache)
+        expected = Timestamp("2015-06-19 19:55:31.877000093")
+        assert result == expected
+
+        alt = Timestamp(value, unit="s")
+        assert alt == result
+
+    @pytest.mark.parametrize("dtype", [int, float])
+    def test_to_datetime_unit(self, dtype):
+        epoch = 1370745748
+        ser = Series([epoch + t for t in range(20)]).astype(dtype)
+        result = to_datetime(ser, unit="s")
+        expected = Series(
+            [
+                Timestamp("2013-06-09 02:42:28") + timedelta(seconds=t)
+                for t in range(20)
+            ],
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("null", [iNaT, np.nan])
+    def test_to_datetime_unit_with_nulls(self, null):
+        epoch = 1370745748
+        ser = Series([epoch + t for t in range(20)] + [null])
+        result = to_datetime(ser, unit="s")
+        expected = Series(
+            [Timestamp("2013-06-09 02:42:28") + timedelta(seconds=t) for t in range(20)]
+            + [NaT],
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_unit_fractional_seconds(self):
+        # GH13834
+        epoch = 1370745748
+        ser = Series([epoch + t for t in np.arange(0, 2, 0.25)] + [iNaT]).astype(float)
+        result = to_datetime(ser, unit="s")
+        expected = Series(
+            [
+                Timestamp("2013-06-09 02:42:28") + timedelta(seconds=t)
+                for t in np.arange(0, 2, 0.25)
+            ]
+            + [NaT],
+            dtype="M8[ns]",
+        )
+        # GH20455 argument will incur floating point errors but no premature rounding
+        result = result.round("ms")
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_unit_na_values(self):
+        result = to_datetime([1, 2, "NaT", NaT, np.nan], unit="D")
+        expected = DatetimeIndex(
+            [Timestamp("1970-01-02"), Timestamp("1970-01-03")] + ["NaT"] * 3,
+            dtype="M8[ns]",
+        )
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize("bad_val", ["foo", 111111111])
+    def test_to_datetime_unit_invalid(self, bad_val):
+        if bad_val == "foo":
+            msg = f"Unknown datetime string format, unable to parse: {bad_val}"
+        else:
+            msg = "cannot convert input 111111111 with the unit 'D'"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([1, 2, bad_val], unit="D")
+
+    @pytest.mark.parametrize("bad_val", ["foo", 111111111])
+    def test_to_timestamp_unit_coerce(self, bad_val):
+        # coerce we can process
+        expected = DatetimeIndex(
+            [Timestamp("1970-01-02"), Timestamp("1970-01-03")] + ["NaT"] * 1,
+            dtype="M8[ns]",
+        )
+        result = to_datetime([1, 2, bad_val], unit="D", errors="coerce")
+        tm.assert_index_equal(result, expected)
+
+    def test_float_to_datetime_raise_near_bounds(self):
+        # GH50183
+        msg = "cannot convert input with unit 'D'"
+        oneday_in_ns = 1e9 * 60 * 60 * 24
+        tsmax_in_days = 2**63 / oneday_in_ns  # 2**63 ns, in days
+        # just in bounds
+        should_succeed = Series(
+            [0, tsmax_in_days - 0.005, -tsmax_in_days + 0.005], dtype=float
+        )
+        expected = (should_succeed * oneday_in_ns).astype(np.int64)
+        for error_mode in ["raise", "coerce"]:
+            result1 = to_datetime(should_succeed, unit="D", errors=error_mode)
+            # Cast to `np.float64` so that `rtol` and inexact checking kick in
+            # (`check_exact` doesn't take place for integer dtypes)
+            tm.assert_almost_equal(
+                result1.astype(np.int64).astype(np.float64),
+                expected.astype(np.float64),
+                rtol=1e-10,
+            )
+        # just out of bounds
+        should_fail1 = Series([0, tsmax_in_days + 0.005], dtype=float)
+        should_fail2 = Series([0, -tsmax_in_days - 0.005], dtype=float)
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(should_fail1, unit="D", errors="raise")
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(should_fail2, unit="D", errors="raise")
+
+
+class TestToDatetimeDataFrame:
+    @pytest.fixture
+    def df(self):
+        return DataFrame(
+            {
+                "year": [2015, 2016],
+                "month": [2, 3],
+                "day": [4, 5],
+                "hour": [6, 7],
+                "minute": [58, 59],
+                "second": [10, 11],
+                "ms": [1, 1],
+                "us": [2, 2],
+                "ns": [3, 3],
+            }
+        )
+
+    def test_dataframe(self, df, cache):
+        result = to_datetime(
+            {"year": df["year"], "month": df["month"], "day": df["day"]}, cache=cache
+        )
+        expected = Series(
+            [Timestamp("20150204 00:00:00"), Timestamp("20160305 00:0:00")]
+        )
+        tm.assert_series_equal(result, expected)
+
+        # dict-like
+        result = to_datetime(df[["year", "month", "day"]].to_dict(), cache=cache)
+        expected.index = Index([0, 1])
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_dict_with_constructable(self, df, cache):
+        # dict but with constructable
+        df2 = df[["year", "month", "day"]].to_dict()
+        df2["month"] = 2
+        result = to_datetime(df2, cache=cache)
+        expected2 = Series(
+            [Timestamp("20150204 00:00:00"), Timestamp("20160205 00:0:00")],
+            index=Index([0, 1]),
+        )
+        tm.assert_series_equal(result, expected2)
+
+    @pytest.mark.parametrize(
+        "unit",
+        [
+            {
+                "year": "years",
+                "month": "months",
+                "day": "days",
+                "hour": "hours",
+                "minute": "minutes",
+                "second": "seconds",
+            },
+            {
+                "year": "year",
+                "month": "month",
+                "day": "day",
+                "hour": "hour",
+                "minute": "minute",
+                "second": "second",
+            },
+        ],
+    )
+    def test_dataframe_field_aliases_column_subset(self, df, cache, unit):
+        # unit mappings
+        result = to_datetime(df[list(unit.keys())].rename(columns=unit), cache=cache)
+        expected = Series(
+            [Timestamp("20150204 06:58:10"), Timestamp("20160305 07:59:11")],
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_field_aliases(self, df, cache):
+        d = {
+            "year": "year",
+            "month": "month",
+            "day": "day",
+            "hour": "hour",
+            "minute": "minute",
+            "second": "second",
+            "ms": "ms",
+            "us": "us",
+            "ns": "ns",
+        }
+
+        result = to_datetime(df.rename(columns=d), cache=cache)
+        expected = Series(
+            [
+                Timestamp("20150204 06:58:10.001002003"),
+                Timestamp("20160305 07:59:11.001002003"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_str_dtype(self, df, cache):
+        # coerce back to int
+        result = to_datetime(df.astype(str), cache=cache)
+        expected = Series(
+            [
+                Timestamp("20150204 06:58:10.001002003"),
+                Timestamp("20160305 07:59:11.001002003"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_float32_dtype(self, df, cache):
+        # GH#60506
+        # coerce to float64
+        result = to_datetime(df.astype(np.float32), cache=cache)
+        expected = Series(
+            [
+                Timestamp("20150204 06:58:10.001002003"),
+                Timestamp("20160305 07:59:11.001002003"),
+            ]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_coerce(self, cache):
+        # passing coerce
+        df2 = DataFrame({"year": [2015, 2016], "month": [2, 20], "day": [4, 5]})
+
+        msg = (
+            r'^cannot assemble the datetimes: time data ".+" doesn\'t '
+            r'match format "%Y%m%d"\.'
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df2, cache=cache)
+
+        result = to_datetime(df2, errors="coerce", cache=cache)
+        expected = Series([Timestamp("20150204 00:00:00"), NaT])
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_extra_keys_raises(self, df, cache):
+        # extra columns
+        msg = r"extra keys have been passed to the datetime assemblage: \[foo\]"
+        df2 = df.copy()
+        df2["foo"] = 1
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df2, cache=cache)
+
+    @pytest.mark.parametrize(
+        "cols",
+        [
+            ["year"],
+            ["year", "month"],
+            ["year", "month", "second"],
+            ["month", "day"],
+            ["year", "day", "second"],
+        ],
+    )
+    def test_dataframe_missing_keys_raises(self, df, cache, cols):
+        # not enough
+        msg = (
+            r"to assemble mappings requires at least that \[year, month, "
+            r"day\] be specified: \[.+\] is missing"
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df[cols], cache=cache)
+
+    def test_dataframe_duplicate_columns_raises(self, cache):
+        # duplicates
+        msg = "cannot assemble with duplicate keys"
+        df2 = DataFrame({"year": [2015, 2016], "month": [2, 20], "day": [4, 5]})
+        df2.columns = ["year", "year", "day"]
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df2, cache=cache)
+
+        df2 = DataFrame(
+            {"year": [2015, 2016], "month": [2, 20], "day": [4, 5], "hour": [4, 5]}
+        )
+        df2.columns = ["year", "month", "day", "day"]
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df2, cache=cache)
+
+    def test_dataframe_int16(self, cache):
+        # GH#13451
+        df = DataFrame({"year": [2015, 2016], "month": [2, 3], "day": [4, 5]})
+
+        # int16
+        result = to_datetime(df.astype("int16"), cache=cache)
+        expected = Series(
+            [Timestamp("20150204 00:00:00"), Timestamp("20160305 00:00:00")]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_mixed(self, cache):
+        # mixed dtypes
+        df = DataFrame({"year": [2015, 2016], "month": [2, 3], "day": [4, 5]})
+        df["month"] = df["month"].astype("int8")
+        df["day"] = df["day"].astype("int8")
+        result = to_datetime(df, cache=cache)
+        expected = Series(
+            [Timestamp("20150204 00:00:00"), Timestamp("20160305 00:00:00")]
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_dataframe_float(self, cache):
+        # float
+        df = DataFrame({"year": [2000, 2001], "month": [1.5, 1], "day": [1, 1]})
+        msg = (
+            r"^cannot assemble the datetimes: unconverted data remains when parsing "
+            r'with format ".*": "1".'
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(df, cache=cache)
+
+    def test_dataframe_utc_true(self):
+        # GH#23760
+        df = DataFrame({"year": [2015, 2016], "month": [2, 3], "day": [4, 5]})
+        result = to_datetime(df, utc=True)
+        expected = Series(
+            np.array(["2015-02-04", "2016-03-05"], dtype="datetime64[s]")
+        ).dt.tz_localize("UTC")
+        tm.assert_series_equal(result, expected)
+
+
+class TestToDatetimeMisc:
+    def test_to_datetime_barely_out_of_bounds(self):
+        # GH#19529
+        # GH#19382 close enough to bounds that dropping nanos would result
+        # in an in-bounds datetime
+        arr = np.array(["2262-04-11 23:47:16.854775808"], dtype=object)
+
+        msg = "^Out of bounds nanosecond timestamp: .*"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            to_datetime(arr)
+
+    @pytest.mark.parametrize(
+        "arg, exp_str",
+        [
+            ["2012-01-01 00:00:00", "2012-01-01 00:00:00"],
+            ["20121001", "2012-10-01"],  # bad iso 8601
+        ],
+    )
+    def test_to_datetime_iso8601(self, cache, arg, exp_str):
+        result = to_datetime([arg], cache=cache)
+        exp = Timestamp(exp_str)
+        assert result[0] == exp
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2012", "%Y-%m"),
+            ("2012-01", "%Y-%m-%d"),
+            ("2012-01-01", "%Y-%m-%d %H"),
+            ("2012-01-01 10", "%Y-%m-%d %H:%M"),
+            ("2012-01-01 10:00", "%Y-%m-%d %H:%M:%S"),
+            ("2012-01-01 10:00:00", "%Y-%m-%d %H:%M:%S.%f"),
+            ("2012-01-01 10:00:00.123", "%Y-%m-%d %H:%M:%S.%f%z"),
+            (0, "%Y-%m-%d"),
+        ],
+    )
+    @pytest.mark.parametrize("exact", [True, False])
+    def test_to_datetime_iso8601_fails(self, input, format, exact):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        # `format` is longer than the string, so this fails regardless of `exact`
+        with pytest.raises(
+            ValueError,
+            match=(rf"time data \"{input}\" doesn't match format " rf"\"{format}\""),
+        ):
+            to_datetime(input, format=format, exact=exact)
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2012-01-01", "%Y-%m"),
+            ("2012-01-01 10", "%Y-%m-%d"),
+            ("2012-01-01 10:00", "%Y-%m-%d %H"),
+            ("2012-01-01 10:00:00", "%Y-%m-%d %H:%M"),
+            (0, "%Y-%m-%d"),
+        ],
+    )
+    def test_to_datetime_iso8601_exact_fails(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        # `format` is shorter than the date string, so only fails with `exact=True`
+        msg = "|".join(
+            [
+                '^unconverted data remains when parsing with format ".*": ".*". '
+                f"{PARSING_ERR_MSG}$",
+                f'^time data ".*" doesn\'t match format ".*". {PARSING_ERR_MSG}$',
+            ]
+        )
+        with pytest.raises(
+            ValueError,
+            match=(msg),
+        ):
+            to_datetime(input, format=format)
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2012-01-01", "%Y-%m"),
+            ("2012-01-01 00", "%Y-%m-%d"),
+            ("2012-01-01 00:00", "%Y-%m-%d %H"),
+            ("2012-01-01 00:00:00", "%Y-%m-%d %H:%M"),
+        ],
+    )
+    def test_to_datetime_iso8601_non_exact(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        expected = Timestamp(2012, 1, 1)
+        result = to_datetime(input, format=format, exact=False)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2020-01", "%Y/%m"),
+            ("2020-01-01", "%Y/%m/%d"),
+            ("2020-01-01 00", "%Y/%m/%dT%H"),
+            ("2020-01-01T00", "%Y/%m/%d %H"),
+            ("2020-01-01 00:00", "%Y/%m/%dT%H:%M"),
+            ("2020-01-01T00:00", "%Y/%m/%d %H:%M"),
+            ("2020-01-01 00:00:00", "%Y/%m/%dT%H:%M:%S"),
+            ("2020-01-01T00:00:00", "%Y/%m/%d %H:%M:%S"),
+        ],
+    )
+    def test_to_datetime_iso8601_separator(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        with pytest.raises(
+            ValueError,
+            match=(rf"time data \"{input}\" doesn\'t match format " rf"\"{format}\""),
+        ):
+            to_datetime(input, format=format)
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2020-01", "%Y-%m"),
+            ("2020-01-01", "%Y-%m-%d"),
+            ("2020-01-01 00", "%Y-%m-%d %H"),
+            ("2020-01-01T00", "%Y-%m-%dT%H"),
+            ("2020-01-01 00:00", "%Y-%m-%d %H:%M"),
+            ("2020-01-01T00:00", "%Y-%m-%dT%H:%M"),
+            ("2020-01-01 00:00:00", "%Y-%m-%d %H:%M:%S"),
+            ("2020-01-01T00:00:00", "%Y-%m-%dT%H:%M:%S"),
+            ("2020-01-01T00:00:00.000", "%Y-%m-%dT%H:%M:%S.%f"),
+            ("2020-01-01T00:00:00.000000", "%Y-%m-%dT%H:%M:%S.%f"),
+            ("2020-01-01T00:00:00.000000000", "%Y-%m-%dT%H:%M:%S.%f"),
+        ],
+    )
+    def test_to_datetime_iso8601_valid(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        expected = Timestamp(2020, 1, 1)
+        result = to_datetime(input, format=format)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2020-1", "%Y-%m"),
+            ("2020-1-1", "%Y-%m-%d"),
+            ("2020-1-1 0", "%Y-%m-%d %H"),
+            ("2020-1-1T0", "%Y-%m-%dT%H"),
+            ("2020-1-1 0:0", "%Y-%m-%d %H:%M"),
+            ("2020-1-1T0:0", "%Y-%m-%dT%H:%M"),
+            ("2020-1-1 0:0:0", "%Y-%m-%d %H:%M:%S"),
+            ("2020-1-1T0:0:0", "%Y-%m-%dT%H:%M:%S"),
+            ("2020-1-1T0:0:0.000", "%Y-%m-%dT%H:%M:%S.%f"),
+            ("2020-1-1T0:0:0.000000", "%Y-%m-%dT%H:%M:%S.%f"),
+            ("2020-1-1T0:0:0.000000000", "%Y-%m-%dT%H:%M:%S.%f"),
+        ],
+    )
+    def test_to_datetime_iso8601_non_padded(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/21422
+        expected = Timestamp(2020, 1, 1)
+        result = to_datetime(input, format=format)
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "input, format",
+        [
+            ("2020-01-01T00:00:00.000000000+00:00", "%Y-%m-%dT%H:%M:%S.%f%z"),
+            ("2020-01-01T00:00:00+00:00", "%Y-%m-%dT%H:%M:%S%z"),
+            ("2020-01-01T00:00:00Z", "%Y-%m-%dT%H:%M:%S%z"),
+        ],
+    )
+    def test_to_datetime_iso8601_with_timezone_valid(self, input, format):
+        # https://github.com/pandas-dev/pandas/issues/12649
+        expected = Timestamp(2020, 1, 1, tzinfo=timezone.utc)
+        result = to_datetime(input, format=format)
+        assert result == expected
+
+    def test_to_datetime_default(self, cache):
+        rs = to_datetime("2001", cache=cache)
+        xp = datetime(2001, 1, 1)
+        assert rs == xp
+
+    @pytest.mark.xfail(reason="fails to enforce dayfirst=True, which would raise")
+    def test_to_datetime_respects_dayfirst(self, cache):
+        # dayfirst is essentially broken
+
+        # The msg here is not important since it isn't actually raised yet.
+        msg = "Invalid date specified"
+        with pytest.raises(ValueError, match=msg):
+            # if dayfirst is respected, then this would parse as month=13, which
+            #  would raise
+            with tm.assert_produces_warning(UserWarning, match="Provide format"):
+                to_datetime("01-13-2012", dayfirst=True, cache=cache)
+
+    def test_to_datetime_on_datetime64_series(self, cache):
+        # #2699
+        ser = Series(date_range("1/1/2000", periods=10))
+
+        result = to_datetime(ser, cache=cache)
+        assert result[0] == ser[0]
+
+    def test_to_datetime_with_space_in_series(self, cache):
+        # GH 6428
+        ser = Series(["10/18/2006", "10/18/2008", " "])
+        msg = (
+            r'^time data " " doesn\'t match format "%m/%d/%Y". ' rf"{PARSING_ERR_MSG}$"
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(ser, errors="raise", cache=cache)
+        result_coerce = to_datetime(ser, errors="coerce", cache=cache)
+        expected_coerce = Series(
+            [datetime(2006, 10, 18), datetime(2008, 10, 18), NaT]
+        ).dt.as_unit("s")
+        tm.assert_series_equal(result_coerce, expected_coerce)
+
+    @td.skip_if_not_us_locale
+    def test_to_datetime_with_apply(self, cache):
+        # this is only locale tested with US/None locales
+        # GH 5195
+        # with a format and coerce a single item to_datetime fails
+        td = Series(["May 04", "Jun 02", "Dec 11"], index=[1, 2, 3])
+        expected = to_datetime(td, format="%b %y", cache=cache)
+        result = td.apply(to_datetime, format="%b %y", cache=cache)
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_timezone_name(self):
+        # https://github.com/pandas-dev/pandas/issues/49748
+        result = to_datetime("2020-01-01 00:00:00UTC", format="%Y-%m-%d %H:%M:%S%Z")
+        expected = Timestamp(2020, 1, 1).tz_localize("UTC")
+        assert result == expected
+
+    @td.skip_if_not_us_locale
+    @pytest.mark.parametrize("errors", ["raise", "coerce"])
+    def test_to_datetime_with_apply_with_empty_str(self, cache, errors):
+        # this is only locale tested with US/None locales
+        # GH 5195, GH50251
+        # with a format and coerce a single item to_datetime fails
+        td = Series(["May 04", "Jun 02", ""], index=[1, 2, 3])
+        expected = to_datetime(td, format="%b %y", errors=errors, cache=cache)
+
+        result = td.apply(
+            lambda x: to_datetime(x, format="%b %y", errors="coerce", cache=cache)
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_empty_stt(self, cache):
+        # empty string
+        result = to_datetime("", cache=cache)
+        assert result is NaT
+
+    def test_to_datetime_empty_str_list(self, cache):
+        result = to_datetime(["", ""], cache=cache)
+        assert isna(result).all()
+
+    def test_to_datetime_zero(self, cache):
+        # ints
+        result = Timestamp(0)
+        expected = to_datetime(0, cache=cache)
+        assert result == expected
+
+    def test_to_datetime_strings(self, cache):
+        # GH 3888 (strings)
+        expected = to_datetime(["2012"], cache=cache)[0]
+        result = to_datetime("2012", cache=cache)
+        assert result == expected
+
+    def test_to_datetime_strings_variation(self, cache):
+        array = ["2012", "20120101", "20120101 12:01:01"]
+        expected = [to_datetime(dt_str, cache=cache) for dt_str in array]
+        result = [Timestamp(date_str) for date_str in array]
+        tm.assert_almost_equal(result, expected)
+
+    @pytest.mark.parametrize("result", [Timestamp("2012"), to_datetime("2012")])
+    def test_to_datetime_strings_vs_constructor(self, result):
+        expected = Timestamp(2012, 1, 1)
+        assert result == expected
+
+    def test_to_datetime_unprocessable_input(self, cache):
+        # GH 4928
+        # GH 21864
+        msg = '^Given date string "1" not likely a datetime$'
+        with pytest.raises(ValueError, match=msg):
+            to_datetime([1, "1"], errors="raise", cache=cache)
+
+    def test_to_datetime_other_datetime64_units(self):
+        # 5/25/2012
+        scalar = np.int64(1337904000000000).view("M8[us]")
+        as_obj = scalar.astype("O")
+
+        index = DatetimeIndex([scalar])
+        assert index[0] == scalar.astype("O")
+
+        value = Timestamp(scalar)
+        assert value == as_obj
+
+    def test_to_datetime_list_of_integers(self):
+        rng = date_range("1/1/2000", periods=20)
+        rng = DatetimeIndex(rng.values)
+
+        ints = list(rng.asi8)
+
+        result = DatetimeIndex(ints)
+
+        tm.assert_index_equal(rng, result)
+
+    def test_to_datetime_overflow(self):
+        # gh-17637
+        # we are overflowing Timedelta range here
+        msg = "Cannot cast 139999 days 00:00:00 to unit='ns' without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            date_range(start="1/1/1700", freq="B", periods=100000)
+
+    def test_string_invalid_operation(self, cache):
+        invalid = np.array(["87156549591102612381000001219H5"], dtype=object)
+        # GH #51084
+
+        with pytest.raises(ValueError, match="Unknown datetime string format"):
+            to_datetime(invalid, errors="raise", cache=cache)
+
+    def test_string_na_nat_conversion(self, cache):
+        # GH #999, #858
+
+        strings = np.array(["1/1/2000", "1/2/2000", np.nan, "1/4/2000"], dtype=object)
+
+        expected = np.empty(4, dtype="M8[s]")
+        for i, val in enumerate(strings):
+            if isna(val):
+                expected[i] = iNaT
+            else:
+                expected[i] = parse(val)
+
+        result = tslib.array_to_datetime(strings)[0]
+        tm.assert_almost_equal(result, expected)
+
+        result2 = to_datetime(strings, cache=cache)
+        assert isinstance(result2, DatetimeIndex)
+        tm.assert_numpy_array_equal(result, result2.values)
+
+    def test_string_na_nat_conversion_malformed(self, cache):
+        malformed = np.array(["1/100/2000", np.nan], dtype=object)
+
+        # GH 10636, default is now 'raise'
+        msg = r"Unknown datetime string format"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(malformed, errors="raise", cache=cache)
+
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(malformed, errors="raise", cache=cache)
+
+    def test_string_na_nat_conversion_with_name(self, cache):
+        idx = ["a", "b", "c", "d", "e"]
+        series = Series(
+            ["1/1/2000", np.nan, "1/3/2000", np.nan, "1/5/2000"], index=idx, name="foo"
+        )
+        dseries = Series(
+            [
+                to_datetime("1/1/2000", cache=cache),
+                np.nan,
+                to_datetime("1/3/2000", cache=cache),
+                np.nan,
+                to_datetime("1/5/2000", cache=cache),
+            ],
+            index=idx,
+            name="foo",
+        )
+
+        result = to_datetime(series, cache=cache)
+        dresult = to_datetime(dseries, cache=cache)
+
+        expected = Series(np.empty(5, dtype="M8[s]"), index=idx)
+        for i in range(5):
+            x = series.iloc[i]
+            if isna(x):
+                expected.iloc[i] = NaT
+            else:
+                expected.iloc[i] = to_datetime(x, cache=cache)
+
+        tm.assert_series_equal(result, expected, check_names=False)
+        assert result.name == "foo"
+
+        tm.assert_series_equal(dresult, expected, check_names=False)
+        assert dresult.name == "foo"
+
+    @pytest.mark.parametrize(
+        "unit",
+        ["h", "m", "s", "ms", "us", "ns"],
+    )
+    def test_dti_constructor_numpy_timeunits(self, cache, unit):
+        # GH 9114
+        dtype = np.dtype(f"M8[{unit}]")
+        base = to_datetime(["2000-01-01T00:00", "2000-01-02T00:00", "NaT"], cache=cache)
+
+        values = base.values.astype(dtype)
+
+        if unit in ["h", "m"]:
+            # we cast to closest supported unit
+            unit = "s"
+        exp_dtype = np.dtype(f"M8[{unit}]")
+        expected = DatetimeIndex(base.astype(exp_dtype))
+        assert expected.dtype == exp_dtype
+
+        tm.assert_index_equal(DatetimeIndex(values), expected)
+        tm.assert_index_equal(to_datetime(values, cache=cache), expected)
+
+    def test_dayfirst(self, cache):
+        # GH 5917
+        arr = ["10/02/2014", "11/02/2014", "12/02/2014"]
+        expected = DatetimeIndex(
+            [datetime(2014, 2, 10), datetime(2014, 2, 11), datetime(2014, 2, 12)]
+        ).as_unit("s")
+        idx1 = DatetimeIndex(arr, dayfirst=True)
+        idx2 = DatetimeIndex(np.array(arr), dayfirst=True)
+        idx3 = to_datetime(arr, dayfirst=True, cache=cache)
+        idx4 = to_datetime(np.array(arr), dayfirst=True, cache=cache)
+        idx5 = DatetimeIndex(Index(arr), dayfirst=True)
+        idx6 = DatetimeIndex(Series(arr), dayfirst=True)
+        tm.assert_index_equal(expected, idx1)
+        tm.assert_index_equal(expected, idx2)
+        tm.assert_index_equal(expected, idx3)
+        tm.assert_index_equal(expected, idx4)
+        tm.assert_index_equal(expected, idx5)
+        tm.assert_index_equal(expected, idx6)
+
+    def test_dayfirst_warnings_valid_input(self):
+        # GH 12585
+        warning_msg = (
+            "Parsing dates in .* format when dayfirst=.* was specified. "
+            "Pass `dayfirst=.*` or specify a format to silence this warning."
+        )
+
+        # CASE 1: valid input
+        arr = ["31/12/2014", "10/03/2011"]
+        expected = DatetimeIndex(
+            ["2014-12-31", "2011-03-10"], dtype="datetime64[s]", freq=None
+        )
+
+        # A. dayfirst arg correct, no warning
+        res1 = to_datetime(arr, dayfirst=True)
+        tm.assert_index_equal(expected, res1)
+
+        # B. dayfirst arg incorrect, warning
+        with tm.assert_produces_warning(UserWarning, match=warning_msg):
+            res2 = to_datetime(arr, dayfirst=False)
+        tm.assert_index_equal(expected, res2)
+
+    def test_dayfirst_warnings_invalid_input(self):
+        # CASE 2: invalid input
+        # cannot consistently process with single format
+        # ValueError *always* raised
+
+        # first in DD/MM/YYYY, second in MM/DD/YYYY
+        arr = ["31/12/2014", "03/30/2011"]
+
+        with pytest.raises(
+            ValueError,
+            match=(
+                r'^time data "03/30/2011" doesn\'t match format '
+                rf'"%d/%m/%Y". {PARSING_ERR_MSG}$'
+            ),
+        ):
+            to_datetime(arr, dayfirst=True)
+
+    @pytest.mark.parametrize("klass", [DatetimeIndex, DatetimeArray._from_sequence])
+    def test_to_datetime_dta_tz(self, klass):
+        # GH#27733
+        dti = date_range("2015-04-05", periods=3).rename("foo")
+        expected = dti.tz_localize("UTC")
+
+        obj = klass(dti)
+        expected = klass(expected)
+
+        result = to_datetime(obj, utc=True)
+        tm.assert_equal(result, expected)
+
+
+class TestGuessDatetimeFormat:
+    @pytest.mark.parametrize(
+        "test_list",
+        [
+            [
+                "2011-12-30 00:00:00.000000",
+                "2011-12-30 00:00:00.000000",
+                "2011-12-30 00:00:00.000000",
+            ],
+            [np.nan, np.nan, "2011-12-30 00:00:00.000000"],
+            ["", "2011-12-30 00:00:00.000000"],
+            ["NaT", "2011-12-30 00:00:00.000000"],
+            ["2011-12-30 00:00:00.000000", "random_string"],
+            ["now", "2011-12-30 00:00:00.000000"],
+            ["today", "2011-12-30 00:00:00.000000"],
+        ],
+    )
+    def test_guess_datetime_format_for_array(self, test_list):
+        expected_format = "%Y-%m-%d %H:%M:%S.%f"
+        test_array = np.array(test_list, dtype=object)
+        assert tools._guess_datetime_format_for_array(test_array) == expected_format
+
+    @td.skip_if_not_us_locale
+    def test_guess_datetime_format_for_array_all_nans(self):
+        format_for_string_of_nans = tools._guess_datetime_format_for_array(
+            np.array([np.nan, np.nan, np.nan], dtype="O")
+        )
+        assert format_for_string_of_nans is None
+
+
+class TestToDatetimeInferFormat:
+    @pytest.mark.parametrize(
+        "test_format", ["%m-%d-%Y", "%m/%d/%Y %H:%M:%S.%f", "%Y-%m-%dT%H:%M:%S.%f"]
+    )
+    def test_to_datetime_infer_datetime_format_consistent_format(
+        self, cache, test_format
+    ):
+        ser = Series(date_range("20000101", periods=50, freq="h"))
+
+        s_as_dt_strings = ser.apply(lambda x: x.strftime(test_format))
+
+        with_format = to_datetime(s_as_dt_strings, format=test_format, cache=cache)
+        without_format = to_datetime(s_as_dt_strings, cache=cache)
+
+        # Whether the format is explicitly passed, or
+        # it is inferred, the results should all be the same
+        tm.assert_series_equal(with_format, without_format)
+
+    def test_to_datetime_inconsistent_format(self, cache):
+        data = ["01/01/2011 00:00:00", "01-02-2011 00:00:00", "2011-01-03T00:00:00"]
+        ser = Series(np.array(data))
+        msg = (
+            r'^time data "01-02-2011 00:00:00" doesn\'t match format '
+            rf'"%m/%d/%Y %H:%M:%S". {PARSING_ERR_MSG}$'
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(ser, cache=cache)
+
+    def test_to_datetime_consistent_format(self, cache):
+        data = ["Jan/01/2011", "Feb/01/2011", "Mar/01/2011"]
+        ser = Series(np.array(data))
+        result = to_datetime(ser, cache=cache)
+        expected = Series(
+            ["2011-01-01", "2011-02-01", "2011-03-01"], dtype="datetime64[s]"
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_series_with_nans(self, cache):
+        ser = Series(
+            np.array(
+                ["01/01/2011 00:00:00", np.nan, "01/03/2011 00:00:00", np.nan],
+                dtype=object,
+            )
+        )
+        result = to_datetime(ser, cache=cache)
+        expected = Series(["2011-01-01", NaT, "2011-01-03", NaT], dtype="datetime64[s]")
+        tm.assert_series_equal(result, expected)
+
+    def test_to_datetime_series_start_with_nans(self, cache):
+        ser = Series(
+            np.array(
+                [
+                    np.nan,
+                    np.nan,
+                    "01/01/2011 00:00:00",
+                    "01/02/2011 00:00:00",
+                    "01/03/2011 00:00:00",
+                ],
+                dtype=object,
+            )
+        )
+
+        result = to_datetime(ser, cache=cache)
+        expected = Series(
+            [NaT, NaT, "2011-01-01", "2011-01-02", "2011-01-03"], dtype="datetime64[s]"
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "tz_name, offset",
+        [("UTC", 0), ("UTC-3", 180), ("UTC+3", -180)],
+    )
+    def test_infer_datetime_format_tz_name(self, tz_name, offset):
+        # GH 33133
+        ser = Series([f"2019-02-02 08:07:13 {tz_name}"])
+        result = to_datetime(ser)
+        tz = timezone(timedelta(minutes=offset))
+        expected = Series([Timestamp("2019-02-02 08:07:13").tz_localize(tz)])
+        expected = expected.dt.as_unit("s")
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "ts,zero_tz",
+        [
+            ("2019-02-02 08:07:13", "Z"),
+            ("2019-02-02 08:07:13", ""),
+            ("2019-02-02 08:07:13.012345", "Z"),
+            ("2019-02-02 08:07:13.012345", ""),
+        ],
+    )
+    def test_infer_datetime_format_zero_tz(self, ts, zero_tz):
+        # GH 41047
+        ser = Series([ts + zero_tz])
+        result = to_datetime(ser)
+        tz = timezone.utc if zero_tz == "Z" else None
+        expected = Series([Timestamp(ts, tz=tz)])
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("format", [None, "%Y-%m-%d"])
+    def test_to_datetime_iso8601_noleading_0s(self, cache, format):
+        # GH 11871
+        ser = Series(["2014-1-1", "2014-2-2", "2015-3-3"])
+        expected = Series(
+            [
+                Timestamp("2014-01-01"),
+                Timestamp("2014-02-02"),
+                Timestamp("2015-03-03"),
+            ]
+        )
+        result = to_datetime(ser, format=format, cache=cache)
+        tm.assert_series_equal(result, expected)
+
+
+class TestDaysInMonth:
+    # tests for issue #10154
+
+    @pytest.mark.parametrize(
+        "arg, format",
+        [
+            ["2015-02-29", None],
+            ["2015-02-29", "%Y-%m-%d"],
+            ["2015-02-32", "%Y-%m-%d"],
+            ["2015-04-31", "%Y-%m-%d"],
+        ],
+    )
+    def test_day_not_in_month_coerce(self, cache, arg, format):
+        assert isna(to_datetime(arg, errors="coerce", format=format, cache=cache))
+
+    def test_day_not_in_month_raise(self, cache):
+        msg = "day is out of range for month: 2015-02-29"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime("2015-02-29", errors="raise", cache=cache)
+
+    @pytest.mark.parametrize(
+        "arg, format, msg",
+        [
+            (
+                "2015-02-29",
+                "%Y-%m-%d",
+                f"^day is out of range for month. {PARSING_ERR_MSG}$",
+            ),
+            (
+                "2015-29-02",
+                "%Y-%d-%m",
+                f"^day is out of range for month. {PARSING_ERR_MSG}$",
+            ),
+            (
+                "2015-02-32",
+                "%Y-%m-%d",
+                '^unconverted data remains when parsing with format "%Y-%m-%d": "2". '
+                f"{PARSING_ERR_MSG}$",
+            ),
+            (
+                "2015-32-02",
+                "%Y-%d-%m",
+                '^time data "2015-32-02" doesn\'t match format "%Y-%d-%m". '
+                f"{PARSING_ERR_MSG}$",
+            ),
+            (
+                "2015-04-31",
+                "%Y-%m-%d",
+                f"^day is out of range for month. {PARSING_ERR_MSG}$",
+            ),
+            (
+                "2015-31-04",
+                "%Y-%d-%m",
+                f"^day is out of range for month. {PARSING_ERR_MSG}$",
+            ),
+        ],
+    )
+    def test_day_not_in_month_raise_value(self, cache, arg, format, msg):
+        # https://github.com/pandas-dev/pandas/issues/50462
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(arg, errors="raise", format=format, cache=cache)
+
+
+class TestDatetimeParsingWrappers:
+    @pytest.mark.parametrize(
+        "date_str, expected",
+        [
+            ("2011-01-01", datetime(2011, 1, 1)),
+            ("2Q2005", datetime(2005, 4, 1)),
+            ("2Q05", datetime(2005, 4, 1)),
+            ("2005Q1", datetime(2005, 1, 1)),
+            ("05Q1", datetime(2005, 1, 1)),
+            ("2011Q3", datetime(2011, 7, 1)),
+            ("11Q3", datetime(2011, 7, 1)),
+            ("3Q2011", datetime(2011, 7, 1)),
+            ("3Q11", datetime(2011, 7, 1)),
+            # quarterly without space
+            ("2000Q4", datetime(2000, 10, 1)),
+            ("00Q4", datetime(2000, 10, 1)),
+            ("4Q2000", datetime(2000, 10, 1)),
+            ("4Q00", datetime(2000, 10, 1)),
+            ("2000q4", datetime(2000, 10, 1)),
+            ("2000-Q4", datetime(2000, 10, 1)),
+            ("00-Q4", datetime(2000, 10, 1)),
+            ("4Q-2000", datetime(2000, 10, 1)),
+            ("4Q-00", datetime(2000, 10, 1)),
+            ("00q4", datetime(2000, 10, 1)),
+            ("2005", datetime(2005, 1, 1)),
+            ("2005-11", datetime(2005, 11, 1)),
+            ("2005 11", datetime(2005, 11, 1)),
+            ("11-2005", datetime(2005, 11, 1)),
+            ("11 2005", datetime(2005, 11, 1)),
+            ("200511", datetime(2020, 5, 11)),
+            ("20051109", datetime(2005, 11, 9)),
+            ("20051109 10:15", datetime(2005, 11, 9, 10, 15)),
+            ("20051109 08H", datetime(2005, 11, 9, 8, 0)),
+            ("2005-11-09 10:15", datetime(2005, 11, 9, 10, 15)),
+            ("2005-11-09 08H", datetime(2005, 11, 9, 8, 0)),
+            ("2005/11/09 10:15", datetime(2005, 11, 9, 10, 15)),
+            ("2005/11/09 10:15:32", datetime(2005, 11, 9, 10, 15, 32)),
+            ("2005/11/09 10:15:32 AM", datetime(2005, 11, 9, 10, 15, 32)),
+            ("2005/11/09 10:15:32 PM", datetime(2005, 11, 9, 22, 15, 32)),
+            ("2005/11/09 08H", datetime(2005, 11, 9, 8, 0)),
+            ("Thu Sep 25 10:36:28 2003", datetime(2003, 9, 25, 10, 36, 28)),
+            ("Thu Sep 25 2003", datetime(2003, 9, 25)),
+            ("Sep 25 2003", datetime(2003, 9, 25)),
+            ("January 1 2014", datetime(2014, 1, 1)),
+            # GH#10537
+            ("2014-06", datetime(2014, 6, 1)),
+            ("06-2014", datetime(2014, 6, 1)),
+            ("2014-6", datetime(2014, 6, 1)),
+            ("6-2014", datetime(2014, 6, 1)),
+            ("20010101 12", datetime(2001, 1, 1, 12)),
+            ("20010101 1234", datetime(2001, 1, 1, 12, 34)),
+            ("20010101 123456", datetime(2001, 1, 1, 12, 34, 56)),
+        ],
+    )
+    def test_parsers(self, date_str, expected, cache):
+        # dateutil >= 2.5.0 defaults to yearfirst=True
+        # https://github.com/dateutil/dateutil/issues/217
+        yearfirst = True
+
+        result1, reso_attrname = parsing.parse_datetime_string_with_reso(
+            date_str, yearfirst=yearfirst
+        )
+
+        reso = {
+            "nanosecond": "ns",
+            "microsecond": "us",
+            "millisecond": "ms",
+            "second": "s",
+        }.get(reso_attrname, "s")
+        result2 = to_datetime(date_str, yearfirst=yearfirst)
+        result3 = to_datetime([date_str], yearfirst=yearfirst)
+        # result5 is used below
+        result4 = to_datetime(
+            np.array([date_str], dtype=object), yearfirst=yearfirst, cache=cache
+        )
+        result6 = DatetimeIndex([date_str], yearfirst=yearfirst)
+        # result7 is used below
+        result8 = DatetimeIndex(Index([date_str]), yearfirst=yearfirst)
+        result9 = DatetimeIndex(Series([date_str]), yearfirst=yearfirst)
+
+        for res in [result1, result2]:
+            assert res == expected
+        for res in [result3, result4, result6, result8, result9]:
+            exp = DatetimeIndex([Timestamp(expected)]).as_unit(reso)
+            tm.assert_index_equal(res, exp)
+
+        # these really need to have yearfirst, but we don't support
+        if not yearfirst:
+            result5 = Timestamp(date_str)
+            assert result5 == expected
+            result7 = date_range(date_str, freq="S", periods=1, yearfirst=yearfirst)
+            assert result7 == expected
+
+    def test_na_values_with_cache(
+        self, cache, unique_nulls_fixture, unique_nulls_fixture2
+    ):
+        # GH22305
+        expected = Index([NaT, NaT], dtype="datetime64[s]")
+        result = to_datetime([unique_nulls_fixture, unique_nulls_fixture2], cache=cache)
+        tm.assert_index_equal(result, expected)
+
+    def test_parsers_nat(self):
+        # Test that each of several string-accepting methods return pd.NaT
+        result1, _ = parsing.parse_datetime_string_with_reso("NaT")
+        result2 = to_datetime("NaT")
+        result3 = Timestamp("NaT")
+        result4 = DatetimeIndex(["NaT"])[0]
+        assert result1 is NaT
+        assert result2 is NaT
+        assert result3 is NaT
+        assert result4 is NaT
+
+    @pytest.mark.parametrize(
+        "date_str, dayfirst, yearfirst, expected",
+        [
+            ("10-11-12", False, False, datetime(2012, 10, 11)),
+            ("10-11-12", True, False, datetime(2012, 11, 10)),
+            ("10-11-12", False, True, datetime(2010, 11, 12)),
+            ("10-11-12", True, True, datetime(2010, 12, 11)),
+            ("20/12/21", False, False, datetime(2021, 12, 20)),
+            ("20/12/21", True, False, datetime(2021, 12, 20)),
+            ("20/12/21", False, True, datetime(2020, 12, 21)),
+            ("20/12/21", True, True, datetime(2020, 12, 21)),
+            # GH 58859
+            ("20201012", True, False, datetime(2020, 12, 10)),
+        ],
+    )
+    def test_parsers_dayfirst_yearfirst(
+        self, cache, date_str, dayfirst, yearfirst, expected
+    ):
+        # OK
+        # 2.5.1 10-11-12   [dayfirst=0, yearfirst=0] -> 2012-10-11 00:00:00
+        # 2.5.2 10-11-12   [dayfirst=0, yearfirst=1] -> 2012-10-11 00:00:00
+        # 2.5.3 10-11-12   [dayfirst=0, yearfirst=0] -> 2012-10-11 00:00:00
+
+        # OK
+        # 2.5.1 10-11-12   [dayfirst=0, yearfirst=1] -> 2010-11-12 00:00:00
+        # 2.5.2 10-11-12   [dayfirst=0, yearfirst=1] -> 2010-11-12 00:00:00
+        # 2.5.3 10-11-12   [dayfirst=0, yearfirst=1] -> 2010-11-12 00:00:00
+
+        # bug fix in 2.5.2
+        # 2.5.1 10-11-12   [dayfirst=1, yearfirst=1] -> 2010-11-12 00:00:00
+        # 2.5.2 10-11-12   [dayfirst=1, yearfirst=1] -> 2010-12-11 00:00:00
+        # 2.5.3 10-11-12   [dayfirst=1, yearfirst=1] -> 2010-12-11 00:00:00
+
+        # OK
+        # 2.5.1 10-11-12   [dayfirst=1, yearfirst=0] -> 2012-11-10 00:00:00
+        # 2.5.2 10-11-12   [dayfirst=1, yearfirst=0] -> 2012-11-10 00:00:00
+        # 2.5.3 10-11-12   [dayfirst=1, yearfirst=0] -> 2012-11-10 00:00:00
+
+        # OK
+        # 2.5.1 20/12/21   [dayfirst=0, yearfirst=0] -> 2021-12-20 00:00:00
+        # 2.5.2 20/12/21   [dayfirst=0, yearfirst=0] -> 2021-12-20 00:00:00
+        # 2.5.3 20/12/21   [dayfirst=0, yearfirst=0] -> 2021-12-20 00:00:00
+
+        # OK
+        # 2.5.1 20/12/21   [dayfirst=0, yearfirst=1] -> 2020-12-21 00:00:00
+        # 2.5.2 20/12/21   [dayfirst=0, yearfirst=1] -> 2020-12-21 00:00:00
+        # 2.5.3 20/12/21   [dayfirst=0, yearfirst=1] -> 2020-12-21 00:00:00
+
+        # revert of bug in 2.5.2
+        # 2.5.1 20/12/21   [dayfirst=1, yearfirst=1] -> 2020-12-21 00:00:00
+        # 2.5.2 20/12/21   [dayfirst=1, yearfirst=1] -> month must be in 1..12
+        # 2.5.3 20/12/21   [dayfirst=1, yearfirst=1] -> 2020-12-21 00:00:00
+
+        # OK
+        # 2.5.1 20/12/21   [dayfirst=1, yearfirst=0] -> 2021-12-20 00:00:00
+        # 2.5.2 20/12/21   [dayfirst=1, yearfirst=0] -> 2021-12-20 00:00:00
+        # 2.5.3 20/12/21   [dayfirst=1, yearfirst=0] -> 2021-12-20 00:00:00
+
+        # str : dayfirst, yearfirst, expected
+
+        # compare with dateutil result
+        dateutil_result = parse(date_str, dayfirst=dayfirst, yearfirst=yearfirst)
+        assert dateutil_result == expected
+
+        result1, _ = parsing.parse_datetime_string_with_reso(
+            date_str, dayfirst=dayfirst, yearfirst=yearfirst
+        )
+
+        # we don't support dayfirst/yearfirst here:
+        if not dayfirst and not yearfirst:
+            result2 = Timestamp(date_str)
+            assert result2 == expected
+
+        result3 = to_datetime(
+            date_str, dayfirst=dayfirst, yearfirst=yearfirst, cache=cache
+        )
+
+        result4 = DatetimeIndex([date_str], dayfirst=dayfirst, yearfirst=yearfirst)[0]
+
+        assert result1 == expected
+        assert result3 == expected
+        assert result4 == expected
+
+    @pytest.mark.parametrize(
+        "date_str, exp_def",
+        [["10:15", datetime(1, 1, 1, 10, 15)], ["9:05", datetime(1, 1, 1, 9, 5)]],
+    )
+    def test_parsers_timestring(self, date_str, exp_def):
+        # must be the same as dateutil result
+        exp_now = parse(date_str)
+
+        result1, _ = parsing.parse_datetime_string_with_reso(date_str)
+        result2 = to_datetime(date_str)
+        result3 = to_datetime([date_str])
+        result4 = Timestamp(date_str)
+        result5 = DatetimeIndex([date_str])[0]
+        # parse time string return time string based on default date
+        # others are not, and can't be changed because it is used in
+        # time series plot
+        assert result1 == exp_def
+        assert result2 == exp_now
+        assert result3 == exp_now
+        assert result4 == exp_now
+        assert result5 == exp_now
+
+    @pytest.mark.parametrize(
+        "dt_string, tz, dt_string_repr",
+        [
+            (
+                "2013-01-01 05:45+0545",
+                timezone(timedelta(minutes=345)),
+                "Timestamp('2013-01-01 05:45:00+0545', tz='UTC+05:45')",
+            ),
+            (
+                "2013-01-01 05:30+0530",
+                timezone(timedelta(minutes=330)),
+                "Timestamp('2013-01-01 05:30:00+0530', tz='UTC+05:30')",
+            ),
+        ],
+    )
+    def test_parsers_timezone_minute_offsets_roundtrip(
+        self, cache, dt_string, tz, dt_string_repr
+    ):
+        # GH11708
+        base = to_datetime("2013-01-01 00:00:00", cache=cache)
+        base = base.tz_localize("UTC").tz_convert(tz)
+        dt_time = to_datetime(dt_string, cache=cache)
+        assert base == dt_time
+        assert dt_string_repr == repr(dt_time)
+
+
+@pytest.fixture(params=["D", "s", "ms", "us", "ns"])
+def units(request):
+    """Day and some time units.
+
+    * D
+    * s
+    * ms
+    * us
+    * ns
+    """
+    return request.param
+
+
+@pytest.fixture
+def julian_dates():
+    return date_range("2014-1-1", periods=10).to_julian_date().values
+
+
+class TestOrigin:
+    def test_origin_and_unit(self):
+        # GH#42624
+        ts = to_datetime(1, unit="s", origin=1)
+        expected = Timestamp("1970-01-01 00:00:02")
+        assert ts == expected
+
+        ts = to_datetime(1, unit="s", origin=1_000_000_000)
+        expected = Timestamp("2001-09-09 01:46:41")
+        assert ts == expected
+
+    def test_julian(self, julian_dates):
+        # gh-11276, gh-11745
+        # for origin as julian
+
+        result = Series(to_datetime(julian_dates, unit="D", origin="julian"))
+        expected = Series(
+            to_datetime(julian_dates - Timestamp(0).to_julian_date(), unit="D")
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_unix(self):
+        result = Series(to_datetime([0, 1, 2], unit="D", origin="unix"))
+        expected = Series(
+            [Timestamp("1970-01-01"), Timestamp("1970-01-02"), Timestamp("1970-01-03")],
+            dtype="M8[ns]",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_julian_round_trip(self):
+        result = to_datetime(2456658, origin="julian", unit="D")
+        assert result.to_julian_date() == 2456658
+
+        # out-of-bounds
+        msg = "1 is Out of Bounds for origin='julian'"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(1, origin="julian", unit="D")
+
+    def test_invalid_unit(self, units, julian_dates):
+        # checking for invalid combination of origin='julian' and unit != D
+        if units != "D":
+            msg = "unit must be 'D' for origin='julian'"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(julian_dates, unit=units, origin="julian")
+
+    @pytest.mark.parametrize("unit", ["ns", "D"])
+    def test_invalid_origin(self, unit):
+        # need to have a numeric specified
+        msg = "it must be numeric with a unit specified"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime("2005-01-01", origin="1960-01-01", unit=unit)
+
+    @pytest.mark.parametrize(
+        "epochs",
+        [
+            Timestamp(1960, 1, 1),
+            datetime(1960, 1, 1),
+            "1960-01-01",
+            np.datetime64("1960-01-01"),
+        ],
+    )
+    def test_epoch(self, units, epochs):
+        epoch_1960 = Timestamp(1960, 1, 1)
+        units_from_epochs = np.arange(5, dtype=np.int64)
+        expected = Series(
+            [pd.Timedelta(x, unit=units) + epoch_1960 for x in units_from_epochs]
+        )
+
+        result = Series(to_datetime(units_from_epochs, unit=units, origin=epochs))
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "origin, exc",
+        [
+            ("random_string", ValueError),
+            ("epoch", ValueError),
+            ("13-24-1990", ValueError),
+            (datetime(1, 1, 1), OutOfBoundsDatetime),
+        ],
+    )
+    def test_invalid_origins(self, origin, exc, units):
+        msg = "|".join(
+            [
+                f"origin {origin} is Out of Bounds",
+                f"origin {origin} cannot be converted to a Timestamp",
+                "Cannot cast .* to unit='ns' without overflow",
+            ]
+        )
+        with pytest.raises(exc, match=msg):
+            to_datetime(list(range(5)), unit=units, origin=origin)
+
+    def test_invalid_origins_tzinfo(self):
+        # GH16842
+        with pytest.raises(ValueError, match="must be tz-naive"):
+            to_datetime(1, unit="D", origin=datetime(2000, 1, 1, tzinfo=timezone.utc))
+
+    def test_incorrect_value_exception(self):
+        # GH47495
+        msg = "Unknown datetime string format, unable to parse: yesterday"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(["today", "yesterday"])
+
+    @pytest.mark.parametrize(
+        "format, warning",
+        [
+            (None, UserWarning),
+            ("%Y-%m-%d %H:%M:%S", None),
+            ("%Y-%d-%m %H:%M:%S", None),
+        ],
+    )
+    def test_to_datetime_out_of_bounds_with_format_arg(self, format, warning):
+        # see gh-23830
+        if format is None:
+            res = to_datetime("2417-10-10 00:00:00.00", format=format)
+            assert isinstance(res, Timestamp)
+            assert res.year == 2417
+            assert res.month == 10
+            assert res.day == 10
+        else:
+            msg = "unconverted data remains when parsing with format.*"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime("2417-10-10 00:00:00.00", format=format)
+
+    @pytest.mark.parametrize(
+        "arg, origin, expected_str",
+        [
+            [200 * 365, "unix", "2169-11-13 00:00:00"],
+            [200 * 365, "1870-01-01", "2069-11-13 00:00:00"],
+            [300 * 365, "1870-01-01", "2169-10-20 00:00:00"],
+        ],
+    )
+    def test_processing_order(self, arg, origin, expected_str):
+        # make sure we handle out-of-bounds *before*
+        # constructing the dates
+
+        result = to_datetime(arg, unit="D", origin=origin)
+        expected = Timestamp(expected_str)
+        assert result == expected
+
+        result = to_datetime(200 * 365, unit="D", origin="1870-01-01")
+        expected = Timestamp("2069-11-13 00:00:00")
+        assert result == expected
+
+        result = to_datetime(300 * 365, unit="D", origin="1870-01-01")
+        expected = Timestamp("2169-10-20 00:00:00")
+        assert result == expected
+
+    @pytest.mark.parametrize(
+        "offset,utc,exp",
+        [
+            ["Z", True, "2019-01-01T00:00:00.000Z"],
+            ["Z", None, "2019-01-01T00:00:00.000Z"],
+            ["-01:00", True, "2019-01-01T01:00:00.000Z"],
+            ["-01:00", None, "2019-01-01T00:00:00.000-01:00"],
+        ],
+    )
+    def test_arg_tz_ns_unit(self, offset, utc, exp):
+        # GH 25546
+        arg = "2019-01-01T00:00:00.000" + offset
+        result = to_datetime([arg], unit="ns", utc=utc)
+        expected = to_datetime([exp]).as_unit("ns")
+        tm.assert_index_equal(result, expected)
+
+
+class TestShouldCache:
+    @pytest.mark.parametrize(
+        "listlike,do_caching",
+        [
+            ([1, 2, 3, 4, 5, 6, 7, 8, 9, 0], False),
+            ([1, 1, 1, 1, 4, 5, 6, 7, 8, 9], True),
+        ],
+    )
+    def test_should_cache(self, listlike, do_caching):
+        assert (
+            tools.should_cache(listlike, check_count=len(listlike), unique_share=0.7)
+            == do_caching
+        )
+
+    @pytest.mark.parametrize(
+        "unique_share,check_count, err_message",
+        [
+            (0.5, 11, r"check_count must be in next bounds: \[0; len\(arg\)\]"),
+            (10, 2, r"unique_share must be in next bounds: \(0; 1\)"),
+        ],
+    )
+    def test_should_cache_errors(self, unique_share, check_count, err_message):
+        arg = [5] * 10
+
+        with pytest.raises(AssertionError, match=err_message):
+            tools.should_cache(arg, unique_share, check_count)
+
+    @pytest.mark.parametrize(
+        "listlike",
+        [
+            (deque([Timestamp("2010-06-02 09:30:00")] * 51)),
+            ([Timestamp("2010-06-02 09:30:00")] * 51),
+            (tuple([Timestamp("2010-06-02 09:30:00")] * 51)),
+        ],
+    )
+    def test_no_slicing_errors_in_should_cache(self, listlike):
+        # GH#29403
+        assert tools.should_cache(listlike) is True
+
+
+def test_nullable_integer_to_datetime():
+    # Test for #30050
+    ser = Series([1, 2, None, 2**61, None])
+    ser = ser.astype("Int64")
+    ser_copy = ser.copy()
+
+    res = to_datetime(ser, unit="ns")
+
+    expected = Series(
+        [
+            np.datetime64("1970-01-01 00:00:00.000000001"),
+            np.datetime64("1970-01-01 00:00:00.000000002"),
+            np.datetime64("NaT"),
+            np.datetime64("2043-01-25 23:56:49.213693952"),
+            np.datetime64("NaT"),
+        ]
+    )
+    tm.assert_series_equal(res, expected)
+    # Check that ser isn't mutated
+    tm.assert_series_equal(ser, ser_copy)
+
+
+@pytest.mark.parametrize("klass", [np.array, list])
+def test_na_to_datetime(nulls_fixture, klass):
+    if isinstance(nulls_fixture, Decimal):
+        with pytest.raises(TypeError, match="not convertible to datetime"):
+            to_datetime(klass([nulls_fixture]))
+
+    else:
+        result = to_datetime(klass([nulls_fixture]))
+
+        assert result[0] is NaT
+
+
+@pytest.mark.parametrize("errors", ["raise", "coerce"])
+@pytest.mark.parametrize(
+    "args, format",
+    [
+        (["03/24/2016", "03/25/2016", ""], "%m/%d/%Y"),
+        (["2016-03-24", "2016-03-25", ""], "%Y-%m-%d"),
+    ],
+    ids=["non-ISO8601", "ISO8601"],
+)
+def test_empty_string_datetime(errors, args, format):
+    # GH13044, GH50251
+    td = Series(args)
+
+    # coerce empty string to pd.NaT
+    result = to_datetime(td, format=format, errors=errors)
+    expected = Series(["2016-03-24", "2016-03-25", NaT], dtype="datetime64[s]")
+    tm.assert_series_equal(expected, result)
+
+
+def test_empty_string_datetime_coerce__unit():
+    # GH13044
+    # coerce empty string to pd.NaT
+    result = to_datetime([1, ""], unit="s", errors="coerce")
+    expected = DatetimeIndex(["1970-01-01 00:00:01", "NaT"], dtype="datetime64[ns]")
+    tm.assert_index_equal(expected, result)
+
+    # verify that no exception is raised even when errors='raise' is set
+    result = to_datetime([1, ""], unit="s", errors="raise")
+    tm.assert_index_equal(expected, result)
+
+
+def test_to_datetime_monotonic_increasing_index(cache):
+    # GH28238
+    cstart = start_caching_at
+    times = date_range(Timestamp("1980"), periods=cstart, freq="YS")
+    times = times.to_frame(index=False, name="DT").sample(n=cstart, random_state=1)
+    times.index = times.index.to_series().astype(float) / 1000
+    result = to_datetime(times.iloc[:, 0], cache=cache)
+    expected = times.iloc[:, 0]
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "series_length",
+    [40, start_caching_at, (start_caching_at + 1), (start_caching_at + 5)],
+)
+def test_to_datetime_cache_coerce_50_lines_outofbounds(series_length):
+    # GH#45319
+    ser = Series(
+        [datetime.fromisoformat("1446-04-12 00:00:00+00:00")]
+        + ([datetime.fromisoformat("1991-10-20 00:00:00+00:00")] * series_length),
+        dtype=object,
+    )
+    result1 = to_datetime(ser, errors="coerce", utc=True)
+
+    expected1 = Series([Timestamp(x) for x in ser])
+    assert expected1.dtype == "M8[us, UTC]"
+    tm.assert_series_equal(result1, expected1)
+
+    result3 = to_datetime(ser, errors="raise", utc=True)
+    tm.assert_series_equal(result3, expected1)
+
+
+def test_to_datetime_format_f_parse_nanos():
+    # GH 48767
+    timestamp = "15/02/2020 02:03:04.123456789"
+    timestamp_format = "%d/%m/%Y %H:%M:%S.%f"
+    result = to_datetime(timestamp, format=timestamp_format)
+    expected = Timestamp(
+        year=2020,
+        month=2,
+        day=15,
+        hour=2,
+        minute=3,
+        second=4,
+        microsecond=123456,
+        nanosecond=789,
+    )
+    assert result == expected
+
+
+def test_to_datetime_mixed_iso8601():
+    # https://github.com/pandas-dev/pandas/issues/50411
+    result = to_datetime(["2020-01-01", "2020-01-01 05:00:00"], format="ISO8601")
+    expected = DatetimeIndex(["2020-01-01 00:00:00", "2020-01-01 05:00:00"])
+    tm.assert_index_equal(result, expected)
+
+
+def test_to_datetime_mixed_other():
+    # https://github.com/pandas-dev/pandas/issues/50411
+    result = to_datetime(["01/11/2000", "12 January 2000"], format="mixed")
+    expected = DatetimeIndex(["2000-01-11", "2000-01-12"])
+    tm.assert_index_equal(result, expected)
+
+
+@pytest.mark.parametrize("exact", [True, False])
+@pytest.mark.parametrize("format", ["ISO8601", "mixed"])
+def test_to_datetime_mixed_or_iso_exact(exact, format):
+    msg = "Cannot use 'exact' when 'format' is 'mixed' or 'ISO8601'"
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(["2020-01-01"], exact=exact, format=format)
+
+
+def test_to_datetime_mixed_not_necessarily_iso8601_raise():
+    # https://github.com/pandas-dev/pandas/issues/50411
+    with pytest.raises(ValueError, match="Time data 01-01-2000 is not ISO8601 format"):
+        to_datetime(["2020-01-01", "01-01-2000"], format="ISO8601")
+
+
+def test_to_datetime_mixed_not_necessarily_iso8601_coerce():
+    # https://github.com/pandas-dev/pandas/issues/50411
+    result = to_datetime(
+        ["2020-01-01", "01-01-2000"], format="ISO8601", errors="coerce"
+    )
+    tm.assert_index_equal(result, DatetimeIndex(["2020-01-01 00:00:00", NaT]))
+
+
+def test_unknown_tz_raises():
+    # GH#18702, GH#51476
+    dtstr = "2014 Jan 9 05:15 FAKE"
+    msg = '.*un-recognized timezone "FAKE".'
+    with pytest.raises(ValueError, match=msg):
+        Timestamp(dtstr)
+
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(dtstr)
+    with pytest.raises(ValueError, match=msg):
+        to_datetime([dtstr])
+
+
+def test_unformatted_input_raises():
+    valid, invalid = "2024-01-01", "N"
+    ser = Series([valid] * start_caching_at + [invalid])
+    msg = 'time data "N" doesn\'t match format "%Y-%m-%d"'
+
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(ser, format="%Y-%m-%d", exact=True, cache=True)
+
+
+def test_from_numeric_arrow_dtype(any_numeric_ea_dtype):
+    # GH 52425
+    pytest.importorskip("pyarrow")
+    ser = Series([1, 2], dtype=f"{any_numeric_ea_dtype.lower()}[pyarrow]")
+    result = to_datetime(ser)
+    expected = Series([1, 2], dtype="datetime64[ns]")
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_datetime_with_empty_str_utc_false_format_mixed():
+    # GH 50887
+    vals = ["2020-01-01 00:00+00:00", ""]
+    result = to_datetime(vals, format="mixed")
+    expected = Index([Timestamp("2020-01-01 00:00+00:00"), "NaT"], dtype="M8[s, UTC]")
+    tm.assert_index_equal(result, expected)
+
+    # Check that a couple of other similar paths work the same way
+    alt = to_datetime(vals)
+    tm.assert_index_equal(alt, expected)
+    alt2 = DatetimeIndex(vals)
+    tm.assert_index_equal(alt2, expected)
+
+
+def test_to_datetime_with_empty_str_utc_false_offsets_and_format_mixed():
+    # GH#50887, GH#57275
+    msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(
+            ["2020-01-01 00:00+00:00", "2020-01-01 00:00+02:00", ""], format="mixed"
+        )
+
+
+def test_to_datetime_mixed_tzs_mixed_types():
+    # GH#55793, GH#55693 mismatched tzs but one is str and other is
+    #  datetime object
+    ts = Timestamp("2016-01-02 03:04:05", tz="US/Pacific")
+    dtstr = "2023-10-30 15:06+01"
+    arr = [ts, dtstr]
+
+    msg = (
+        "Mixed timezones detected. Pass utc=True in to_datetime or tz='UTC' "
+        "in DatetimeIndex to convert to a common timezone"
+    )
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(arr)
+    with pytest.raises(ValueError, match=msg):
+        to_datetime(arr, format="mixed")
+    with pytest.raises(ValueError, match=msg):
+        DatetimeIndex(arr)
+
+
+def test_to_datetime_mixed_types_matching_tzs():
+    # GH#55793
+    dtstr = "2023-11-01 09:22:03-07:00"
+    ts = Timestamp(dtstr)
+    arr = [ts, dtstr]
+    res1 = to_datetime(arr)
+    res2 = to_datetime(arr[::-1])[::-1]
+    res3 = to_datetime(arr, format="mixed")
+    res4 = DatetimeIndex(arr)
+
+    expected = DatetimeIndex([ts, ts])
+    tm.assert_index_equal(res1, expected)
+    tm.assert_index_equal(res2, expected)
+    tm.assert_index_equal(res3, expected)
+    tm.assert_index_equal(res4, expected)
+
+
+dtstr = "2020-01-01 00:00+00:00"
+ts = Timestamp(dtstr)
+
+
+@pytest.mark.filterwarnings("ignore:Could not infer format:UserWarning")
+@pytest.mark.parametrize(
+    "aware_val",
+    [dtstr, Timestamp(dtstr)],
+    ids=lambda x: type(x).__name__,
+)
+@pytest.mark.parametrize(
+    "naive_val",
+    [dtstr[:-6], ts.tz_localize(None), ts.date(), ts.asm8, ts.value, float(ts.value)],
+    ids=lambda x: type(x).__name__,
+)
+@pytest.mark.parametrize("naive_first", [True, False])
+def test_to_datetime_mixed_awareness_mixed_types(aware_val, naive_val, naive_first):
+    # GH#55793, GH#55693, GH#57275
+    # Empty string parses to NaT
+    vals = [aware_val, naive_val, ""]
+
+    vec = vals
+    if naive_first:
+        # alas, the behavior is order-dependent, so we test both ways
+        vec = [naive_val, aware_val, ""]
+
+    # both_strs-> paths that were previously already deprecated with warning
+    #  issued in _array_to_datetime_object
+    both_strs = isinstance(aware_val, str) and isinstance(naive_val, str)
+    has_numeric = isinstance(naive_val, (int, float))
+    both_datetime = isinstance(naive_val, datetime) and isinstance(aware_val, datetime)
+
+    mixed_msg = (
+        "Mixed timezones detected. Pass utc=True in to_datetime or tz='UTC' "
+        "in DatetimeIndex to convert to a common timezone"
+    )
+
+    first_non_null = next(x for x in vec if x != "")
+    # if first_non_null is a not a string, _guess_datetime_format_for_array
+    #  doesn't guess a format so we don't go through array_strptime
+    if not isinstance(first_non_null, str):
+        # that case goes through array_strptime which has different behavior
+        msg = mixed_msg
+        if naive_first and isinstance(aware_val, Timestamp):
+            if isinstance(naive_val, Timestamp):
+                msg = "Tz-aware datetime.datetime cannot be converted to datetime64"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(vec)
+        else:
+            if not naive_first and both_datetime:
+                msg = "Cannot mix tz-aware with tz-naive values"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(vec)
+
+        # No warning/error with utc=True
+        to_datetime(vec, utc=True)
+
+    elif has_numeric and vec.index(aware_val) < vec.index(naive_val):
+        msg = "time data .* doesn't match format"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec)
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec, utc=True)
+
+    elif both_strs and vec.index(aware_val) < vec.index(naive_val):
+        msg = r"time data \"2020-01-01 00:00\" doesn't match format"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec)
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec, utc=True)
+
+    elif both_strs and vec.index(naive_val) < vec.index(aware_val):
+        msg = "unconverted data remains when parsing with format"
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec)
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec, utc=True)
+
+    else:
+        msg = mixed_msg
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec)
+
+        # No warning/error with utc=True
+        to_datetime(vec, utc=True)
+
+    if both_strs:
+        msg = mixed_msg
+        with pytest.raises(ValueError, match=msg):
+            to_datetime(vec, format="mixed")
+        with pytest.raises(ValueError, match=msg):
+            DatetimeIndex(vec)
+    else:
+        msg = mixed_msg
+        if naive_first and isinstance(aware_val, Timestamp):
+            if isinstance(naive_val, Timestamp):
+                msg = "Tz-aware datetime.datetime cannot be converted to datetime64"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(vec, format="mixed")
+            with pytest.raises(ValueError, match=msg):
+                DatetimeIndex(vec)
+        else:
+            if not naive_first and both_datetime:
+                msg = "Cannot mix tz-aware with tz-naive values"
+            with pytest.raises(ValueError, match=msg):
+                to_datetime(vec, format="mixed")
+            with pytest.raises(ValueError, match=msg):
+                DatetimeIndex(vec)
+
+
+def test_to_datetime_wrapped_datetime64_ps():
+    # GH#60341
+    result = to_datetime([np.datetime64(1901901901901, "ps")])
+    expected = DatetimeIndex(
+        ["1970-01-01 00:00:01.901901901"], dtype="datetime64[ns]", freq=None
+    )
+    tm.assert_index_equal(result, expected)
diff --git a/pandas/pandas/tests/tools/test_to_numeric.py b/pandas/pandas/tests/tools/test_to_numeric.py
new file mode 100644
index 0000000000000..893f526fb3eb0
--- /dev/null
+++ b/pandas/pandas/tests/tools/test_to_numeric.py
@@ -0,0 +1,921 @@
+import decimal
+
+import numpy as np
+from numpy import iinfo
+import pytest
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+from pandas import (
+    ArrowDtype,
+    DataFrame,
+    Index,
+    Series,
+    option_context,
+    to_numeric,
+)
+import pandas._testing as tm
+
+
+@pytest.fixture(params=[None, "raise", "coerce"])
+def errors(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def signed(request):
+    return request.param
+
+
+@pytest.fixture(params=[lambda x: x, str], ids=["identity", "str"])
+def transform(request):
+    return request.param
+
+
+@pytest.fixture(params=[47393996303418497800, 100000000000000000000])
+def large_val(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def multiple_elts(request):
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        (lambda x: Index(x, name="idx"), tm.assert_index_equal),
+        (lambda x: Series(x, name="ser"), tm.assert_series_equal),
+        (lambda x: np.array(Index(x).values), tm.assert_numpy_array_equal),
+    ]
+)
+def transform_assert_equal(request):
+    return request.param
+
+
+@pytest.mark.parametrize(
+    "input_kwargs,result_kwargs",
+    [
+        ({}, {"dtype": np.int64}),
+        ({"errors": "coerce", "downcast": "integer"}, {"dtype": np.int8}),
+    ],
+)
+def test_empty(input_kwargs, result_kwargs):
+    # see gh-16302
+    ser = Series([], dtype=object)
+    result = to_numeric(ser, **input_kwargs)
+
+    expected = Series([], **result_kwargs)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "infer_string", [False, pytest.param(True, marks=td.skip_if_no("pyarrow"))]
+)
+@pytest.mark.parametrize("last_val", ["7", 7])
+def test_series(last_val, infer_string):
+    with option_context("future.infer_string", infer_string):
+        ser = Series(["1", "-3.14", last_val])
+        result = to_numeric(ser)
+
+    expected = Series([1, -3.14, 7])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        [1, 3, 4, 5],
+        [1.0, 3.0, 4.0, 5.0],
+        # Bool is regarded as numeric.
+        [True, False, True, True],
+    ],
+)
+def test_series_numeric(data):
+    ser = Series(data, index=list("ABCD"), name="EFG")
+
+    result = to_numeric(ser)
+    tm.assert_series_equal(result, ser)
+
+
+@pytest.mark.parametrize(
+    "data,msg",
+    [
+        ([1, -3.14, "apple"], 'Unable to parse string "apple" at position 2'),
+        (
+            ["orange", 1, -3.14, "apple"],
+            'Unable to parse string "orange" at position 0',
+        ),
+    ],
+)
+def test_error(data, msg):
+    ser = Series(data)
+
+    with pytest.raises(ValueError, match=msg):
+        to_numeric(ser, errors="raise")
+
+
+def test_ignore_error():
+    ser = Series([1, -3.14, "apple"])
+    result = to_numeric(ser, errors="coerce")
+
+    expected = Series([1, -3.14, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "errors,exp",
+    [
+        ("raise", 'Unable to parse string "apple" at position 2'),
+        # Coerces to float.
+        ("coerce", [1.0, 0.0, np.nan]),
+    ],
+)
+def test_bool_handling(errors, exp):
+    ser = Series([True, False, "apple"])
+
+    if isinstance(exp, str):
+        with pytest.raises(ValueError, match=exp):
+            to_numeric(ser, errors=errors)
+    else:
+        result = to_numeric(ser, errors=errors)
+        expected = Series(exp)
+
+        tm.assert_series_equal(result, expected)
+
+
+def test_list():
+    ser = ["1", "-3.14", "7"]
+    res = to_numeric(ser)
+
+    expected = np.array([1, -3.14, 7])
+    tm.assert_numpy_array_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "data,arr_kwargs",
+    [
+        ([1, 3, 4, 5], {"dtype": np.int64}),
+        ([1.0, 3.0, 4.0, 5.0], {}),
+        # Boolean is regarded as numeric.
+        ([True, False, True, True], {}),
+    ],
+)
+def test_list_numeric(data, arr_kwargs):
+    result = to_numeric(data)
+    expected = np.array(data, **arr_kwargs)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("kwargs", [{"dtype": "O"}, {}])
+def test_numeric(kwargs):
+    data = [1, -3.14, 7]
+
+    ser = Series(data, **kwargs)
+    result = to_numeric(ser)
+
+    expected = Series(data)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "columns",
+    [
+        # One column.
+        "a",
+        # Multiple columns.
+        ["a", "b"],
+    ],
+)
+def test_numeric_df_columns(columns):
+    # see gh-14827
+    df = DataFrame(
+        {
+            "a": [1.2, decimal.Decimal("3.14"), decimal.Decimal("infinity"), "0.1"],
+            "b": [1.0, 2.0, 3.0, 4.0],
+        }
+    )
+
+    expected = DataFrame({"a": [1.2, 3.14, np.inf, 0.1], "b": [1.0, 2.0, 3.0, 4.0]})
+    df[columns] = df[columns].apply(to_numeric)
+
+    tm.assert_frame_equal(df, expected)
+
+
+@pytest.mark.parametrize(
+    "data,exp_data",
+    [
+        (
+            [[decimal.Decimal("3.14"), 1.0], decimal.Decimal("1.6"), 0.1],
+            [[3.14, 1.0], 1.6, 0.1],
+        ),
+        ([np.array([decimal.Decimal("3.14"), 1.0]), 0.1], [[3.14, 1.0], 0.1]),
+    ],
+)
+def test_numeric_embedded_arr_likes(data, exp_data):
+    # Test to_numeric with embedded lists and arrays
+    df = DataFrame({"a": data})
+    df["a"] = df["a"].apply(to_numeric)
+
+    expected = DataFrame({"a": exp_data})
+    tm.assert_frame_equal(df, expected)
+
+
+def test_all_nan():
+    ser = Series(["a", "b", "c"])
+    result = to_numeric(ser, errors="coerce")
+
+    expected = Series([np.nan, np.nan, np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_type_check(errors):
+    # see gh-11776
+    df = DataFrame({"a": [1, -3.14, 7], "b": ["4", "5", "6"]})
+    kwargs = {"errors": errors} if errors is not None else {}
+    with pytest.raises(TypeError, match="1-d array"):
+        to_numeric(df, **kwargs)
+
+
+@pytest.mark.parametrize("val", [1, 1.1, 20001])
+def test_scalar(val, signed, transform):
+    val = -val if signed else val
+    assert to_numeric(transform(val)) == float(val)
+
+
+def test_really_large_scalar(large_val, signed, transform, errors):
+    # see gh-24910
+    kwargs = {"errors": errors} if errors is not None else {}
+    val = -large_val if signed else large_val
+
+    val = transform(val)
+    val_is_string = isinstance(val, str)
+
+    if val_is_string and errors in (None, "raise"):
+        msg = "Integer out of range. at position 0"
+        with pytest.raises(ValueError, match=msg):
+            to_numeric(val, **kwargs)
+    else:
+        expected = float(val) if (errors == "coerce" and val_is_string) else val
+        tm.assert_almost_equal(to_numeric(val, **kwargs), expected)
+
+
+def test_really_large_in_arr(large_val, signed, transform, multiple_elts, errors):
+    # see gh-24910
+    kwargs = {"errors": errors} if errors is not None else {}
+    val = -large_val if signed else large_val
+    val = transform(val)
+
+    extra_elt = "string"
+    arr = [val] + multiple_elts * [extra_elt]
+
+    val_is_string = isinstance(val, str)
+    coercing = errors == "coerce"
+
+    if errors in (None, "raise") and (val_is_string or multiple_elts):
+        if val_is_string:
+            msg = "Integer out of range. at position 0"
+        else:
+            msg = 'Unable to parse string "string" at position 1'
+
+        with pytest.raises(ValueError, match=msg):
+            to_numeric(arr, **kwargs)
+    else:
+        result = to_numeric(arr, **kwargs)
+
+        exp_val = float(val) if (coercing and val_is_string) else val
+        expected = [exp_val]
+
+        if multiple_elts:
+            if coercing:
+                expected.append(np.nan)
+                exp_dtype = float
+            else:
+                expected.append(extra_elt)
+                exp_dtype = object
+        else:
+            exp_dtype = float if isinstance(exp_val, (int, float)) else object
+
+        tm.assert_almost_equal(result, np.array(expected, dtype=exp_dtype))
+
+
+def test_really_large_in_arr_consistent(large_val, signed, multiple_elts, errors):
+    # see gh-24910
+    #
+    # Even if we discover that we have to hold float, does not mean
+    # we should be lenient on subsequent elements that fail to be integer.
+    kwargs = {"errors": errors} if errors is not None else {}
+    arr = [str(-large_val if signed else large_val)]
+
+    if multiple_elts:
+        arr.insert(0, large_val)
+
+    if errors in (None, "raise"):
+        index = int(multiple_elts)
+        msg = f"Integer out of range. at position {index}"
+
+        with pytest.raises(ValueError, match=msg):
+            to_numeric(arr, **kwargs)
+    else:
+        result = to_numeric(arr, **kwargs)
+        expected = [float(i) for i in arr]
+        exp_dtype = float
+
+        tm.assert_almost_equal(result, np.array(expected, dtype=exp_dtype))
+
+
+@pytest.mark.parametrize(
+    "errors,checker",
+    [
+        ("raise", 'Unable to parse string "fail" at position 0'),
+        ("coerce", lambda x: np.isnan(x)),
+    ],
+)
+def test_scalar_fail(errors, checker):
+    scalar = "fail"
+
+    if isinstance(checker, str):
+        with pytest.raises(ValueError, match=checker):
+            to_numeric(scalar, errors=errors)
+    else:
+        assert checker(to_numeric(scalar, errors=errors))
+
+
+@pytest.mark.parametrize("data", [[1, 2, 3], [1.0, np.nan, 3, np.nan]])
+def test_numeric_dtypes(data, transform_assert_equal):
+    transform, assert_equal = transform_assert_equal
+    data = transform(data)
+
+    result = to_numeric(data)
+    assert_equal(result, data)
+
+
+@pytest.mark.parametrize(
+    "data,exp",
+    [
+        (["1", "2", "3"], np.array([1, 2, 3], dtype="int64")),
+        (["1.5", "2.7", "3.4"], np.array([1.5, 2.7, 3.4])),
+    ],
+)
+def test_str(data, exp, transform_assert_equal):
+    transform, assert_equal = transform_assert_equal
+    result = to_numeric(transform(data))
+
+    expected = transform(exp)
+    assert_equal(result, expected)
+
+
+def test_datetime_like(tz_naive_fixture, transform_assert_equal):
+    transform, assert_equal = transform_assert_equal
+    idx = pd.date_range("20130101", periods=3, tz=tz_naive_fixture)
+
+    result = to_numeric(transform(idx))
+    expected = transform(idx.asi8)
+    assert_equal(result, expected)
+
+
+def test_timedelta(transform_assert_equal):
+    transform, assert_equal = transform_assert_equal
+    idx = pd.timedelta_range("1 days", periods=3, freq="D")
+
+    result = to_numeric(transform(idx))
+    expected = transform(idx.asi8)
+    assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "scalar",
+    [
+        pd.Timedelta(1, "D"),
+        pd.Timestamp("2017-01-01T12"),
+        pd.Timestamp("2017-01-01T12", tz="US/Pacific"),
+    ],
+)
+def test_timedelta_timestamp_scalar(scalar):
+    # GH#59944
+    result = to_numeric(scalar)
+    expected = to_numeric(Series(scalar))[0]
+    assert result == expected
+
+
+def test_period(request, transform_assert_equal):
+    transform, assert_equal = transform_assert_equal
+
+    idx = pd.period_range("2011-01", periods=3, freq="M", name="")
+    inp = transform(idx)
+
+    if not isinstance(inp, Index):
+        request.applymarker(
+            pytest.mark.xfail(reason="Missing PeriodDtype support in to_numeric")
+        )
+    result = to_numeric(inp)
+    expected = transform(idx.asi8)
+    assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "errors,expected",
+    [
+        ("raise", "Invalid object type at position 0"),
+        ("coerce", Series([np.nan, 1.0, np.nan])),
+    ],
+)
+def test_non_hashable(errors, expected):
+    # see gh-13324
+    ser = Series([[10.0, 2], 1.0, "apple"])
+
+    if isinstance(expected, str):
+        with pytest.raises(TypeError, match=expected):
+            to_numeric(ser, errors=errors)
+    else:
+        result = to_numeric(ser, errors=errors)
+        tm.assert_series_equal(result, expected)
+
+
+def test_downcast_invalid_cast():
+    # see gh-13352
+    data = ["1", 2, 3]
+    invalid_downcast = "unsigned-integer"
+    msg = "invalid downcasting method provided"
+
+    with pytest.raises(ValueError, match=msg):
+        to_numeric(data, downcast=invalid_downcast)
+
+
+def test_errors_invalid_value():
+    # see gh-26466
+    data = ["1", 2, 3]
+    invalid_error_value = "invalid"
+    msg = "invalid error value specified"
+
+    with pytest.raises(ValueError, match=msg):
+        to_numeric(data, errors=invalid_error_value)
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        ["1", 2, 3],
+        [1, 2, 3],
+        np.array(["1970-01-02", "1970-01-03", "1970-01-04"], dtype="datetime64[D]"),
+    ],
+)
+@pytest.mark.parametrize(
+    "kwargs,exp_dtype",
+    [
+        # Basic function tests.
+        ({}, np.int64),
+        ({"downcast": None}, np.int64),
+        # Support below np.float32 is rare and far between.
+        ({"downcast": "float"}, np.dtype(np.float32).char),
+        # Basic dtype support.
+        ({"downcast": "unsigned"}, np.dtype(np.typecodes["UnsignedInteger"][0])),
+    ],
+)
+def test_downcast_basic(data, kwargs, exp_dtype):
+    # see gh-13352
+    result = to_numeric(data, **kwargs)
+    expected = np.array([1, 2, 3], dtype=exp_dtype)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("signed_downcast", ["integer", "signed"])
+@pytest.mark.parametrize(
+    "data",
+    [
+        ["1", 2, 3],
+        [1, 2, 3],
+        np.array(["1970-01-02", "1970-01-03", "1970-01-04"], dtype="datetime64[D]"),
+    ],
+)
+def test_signed_downcast(data, signed_downcast):
+    # see gh-13352
+    smallest_int_dtype = np.dtype(np.typecodes["Integer"][0])
+    expected = np.array([1, 2, 3], dtype=smallest_int_dtype)
+
+    res = to_numeric(data, downcast=signed_downcast)
+    tm.assert_numpy_array_equal(res, expected)
+
+
+def test_ignore_downcast_neg_to_unsigned():
+    # Cannot cast to an unsigned integer
+    # because we have a negative number.
+    data = ["-1", 2, 3]
+    expected = np.array([-1, 2, 3], dtype=np.int64)
+
+    res = to_numeric(data, downcast="unsigned")
+    tm.assert_numpy_array_equal(res, expected)
+
+
+# Warning in 32 bit platforms
+@pytest.mark.parametrize("downcast", ["integer", "signed", "unsigned"])
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        (["1.1", 2, 3], np.array([1.1, 2, 3], dtype=np.float64)),
+        (
+            [10000.0, 20000, 3000, 40000.36, 50000, 50000.00],
+            np.array(
+                [10000.0, 20000, 3000, 40000.36, 50000, 50000.00], dtype=np.float64
+            ),
+        ),
+    ],
+)
+def test_ignore_downcast_cannot_convert_float(data, expected, downcast):
+    # Cannot cast to an integer (signed or unsigned)
+    # because we have a float number.
+    res = to_numeric(data, downcast=downcast)
+    tm.assert_numpy_array_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "downcast,expected_dtype",
+    [("integer", np.int16), ("signed", np.int16), ("unsigned", np.uint16)],
+)
+def test_downcast_not8bit(downcast, expected_dtype):
+    # the smallest integer dtype need not be np.(u)int8
+    data = ["256", 257, 258]
+
+    expected = np.array([256, 257, 258], dtype=expected_dtype)
+    res = to_numeric(data, downcast=downcast)
+    tm.assert_numpy_array_equal(res, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype,downcast,min_max",
+    [
+        ("int8", "integer", [iinfo(np.int8).min, iinfo(np.int8).max]),
+        ("int16", "integer", [iinfo(np.int16).min, iinfo(np.int16).max]),
+        ("int32", "integer", [iinfo(np.int32).min, iinfo(np.int32).max]),
+        ("int64", "integer", [iinfo(np.int64).min, iinfo(np.int64).max]),
+        ("uint8", "unsigned", [iinfo(np.uint8).min, iinfo(np.uint8).max]),
+        ("uint16", "unsigned", [iinfo(np.uint16).min, iinfo(np.uint16).max]),
+        ("uint32", "unsigned", [iinfo(np.uint32).min, iinfo(np.uint32).max]),
+        ("uint64", "unsigned", [iinfo(np.uint64).min, iinfo(np.uint64).max]),
+        ("int16", "integer", [iinfo(np.int8).min, iinfo(np.int8).max + 1]),
+        ("int32", "integer", [iinfo(np.int16).min, iinfo(np.int16).max + 1]),
+        ("int64", "integer", [iinfo(np.int32).min, iinfo(np.int32).max + 1]),
+        ("int16", "integer", [iinfo(np.int8).min - 1, iinfo(np.int16).max]),
+        ("int32", "integer", [iinfo(np.int16).min - 1, iinfo(np.int32).max]),
+        ("int64", "integer", [iinfo(np.int32).min - 1, iinfo(np.int64).max]),
+        ("uint16", "unsigned", [iinfo(np.uint8).min, iinfo(np.uint8).max + 1]),
+        ("uint32", "unsigned", [iinfo(np.uint16).min, iinfo(np.uint16).max + 1]),
+        ("uint64", "unsigned", [iinfo(np.uint32).min, iinfo(np.uint32).max + 1]),
+    ],
+)
+def test_downcast_limits(dtype, downcast, min_max):
+    # see gh-14404: test the limits of each downcast.
+    series = to_numeric(Series(min_max), downcast=downcast)
+    assert series.dtype == dtype
+
+
+def test_downcast_float64_to_float32():
+    # GH-43693: Check float64 preservation when >= 16,777,217
+    series = Series([16777217.0, np.finfo(np.float64).max, np.nan], dtype=np.float64)
+    result = to_numeric(series, downcast="float")
+
+    assert series.dtype == result.dtype
+
+
+def test_downcast_uint64():
+    # see gh-14422:
+    # BUG: to_numeric doesn't work uint64 numbers
+    ser = Series([0, 9223372036854775808])
+    result = to_numeric(ser, downcast="unsigned")
+    expected = Series([0, 9223372036854775808], dtype=np.uint64)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data,exp_data",
+    [
+        (
+            [200, 300, "", "NaN", 30000000000000000000],
+            [200, 300, np.nan, np.nan, 30000000000000000000],
+        ),
+        (
+            ["12345678901234567890", "1234567890", "ITEM"],
+            [12345678901234567890, 1234567890, np.nan],
+        ),
+    ],
+)
+def test_coerce_uint64_conflict(data, exp_data):
+    # see gh-17007 and gh-17125
+    #
+    # Still returns float despite the uint64-nan conflict,
+    # which would normally force the casting to object.
+    result = to_numeric(Series(data), errors="coerce")
+    expected = Series(exp_data, dtype=float)
+    tm.assert_series_equal(result, expected)
+
+
+def test_non_coerce_uint64_conflict():
+    # see gh-17007 and gh-17125
+    #
+    # For completeness.
+    ser = Series(["12345678901234567890", "1234567890", "ITEM"])
+
+    with pytest.raises(ValueError, match="Unable to parse string"):
+        to_numeric(ser, errors="raise")
+
+
+@pytest.mark.parametrize("dc1", ["integer", "float", "unsigned"])
+@pytest.mark.parametrize("dc2", ["integer", "float", "unsigned"])
+def test_downcast_empty(dc1, dc2):
+    # GH32493
+
+    tm.assert_numpy_array_equal(
+        to_numeric([], downcast=dc1),
+        to_numeric([], downcast=dc2),
+        check_dtype=False,
+    )
+
+
+def test_failure_to_convert_uint64_string_to_NaN():
+    # GH 32394
+    result = to_numeric("uint64", errors="coerce")
+    assert np.isnan(result)
+
+    ser = Series([32, 64, np.nan])
+    result = to_numeric(Series(["32", "64", "uint64"]), errors="coerce")
+    tm.assert_series_equal(result, ser)
+
+
+@pytest.mark.parametrize(
+    "strrep",
+    [
+        "243.164",
+        "245.968",
+        "249.585",
+        "259.745",
+        "265.742",
+        "272.567",
+        "279.196",
+        "280.366",
+        "275.034",
+        "271.351",
+        "272.889",
+        "270.627",
+        "280.828",
+        "290.383",
+        "308.153",
+        "319.945",
+        "336.0",
+        "344.09",
+        "351.385",
+        "356.178",
+        "359.82",
+        "361.03",
+        "367.701",
+        "380.812",
+        "387.98",
+        "391.749",
+        "391.171",
+        "385.97",
+        "385.345",
+        "386.121",
+        "390.996",
+        "399.734",
+        "413.073",
+        "421.532",
+        "430.221",
+        "437.092",
+        "439.746",
+        "446.01",
+        "451.191",
+        "460.463",
+        "469.779",
+        "472.025",
+        "479.49",
+        "474.864",
+        "467.54",
+        "471.978",
+    ],
+)
+def test_precision_float_conversion(strrep):
+    # GH 31364
+    result = to_numeric(strrep)
+
+    assert result == float(strrep)
+
+
+@pytest.mark.parametrize(
+    "values, expected",
+    [
+        (["1", "2", None], Series([1, 2, np.nan], dtype="Int64")),
+        (["1", "2", "3"], Series([1, 2, 3], dtype="Int64")),
+        (["1", "2", 3], Series([1, 2, 3], dtype="Int64")),
+        (["1", "2", 3.5], Series([1, 2, 3.5], dtype="Float64")),
+        (["1", None, 3.5], Series([1, np.nan, 3.5], dtype="Float64")),
+        (["1", "2", "3.5"], Series([1, 2, 3.5], dtype="Float64")),
+    ],
+)
+def test_to_numeric_from_nullable_string(values, nullable_string_dtype, expected):
+    # https://github.com/pandas-dev/pandas/issues/37262
+    s = Series(values, dtype=nullable_string_dtype)
+    result = to_numeric(s)
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_numeric_from_nullable_string_coerce(nullable_string_dtype):
+    # GH#52146
+    values = ["a", "1"]
+    ser = Series(values, dtype=nullable_string_dtype)
+    result = to_numeric(ser, errors="coerce")
+    expected = Series([pd.NA, 1], dtype="Int64")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "data, input_dtype, downcast, expected_dtype",
+    (
+        ([1, 1], "Int64", "integer", "Int8"),
+        ([1.0, pd.NA], "Float64", "integer", "Int8"),
+        ([1.0, 1.1], "Float64", "integer", "Float64"),
+        ([1, pd.NA], "Int64", "integer", "Int8"),
+        ([450, 300], "Int64", "integer", "Int16"),
+        ([1, 1], "Float64", "integer", "Int8"),
+        ([np.iinfo(np.int64).max - 1, 1], "Int64", "integer", "Int64"),
+        ([1, 1], "Int64", "signed", "Int8"),
+        ([1.0, 1.0], "Float32", "signed", "Int8"),
+        ([1.0, 1.1], "Float64", "signed", "Float64"),
+        ([1, pd.NA], "Int64", "signed", "Int8"),
+        ([450, -300], "Int64", "signed", "Int16"),
+        ([np.iinfo(np.uint64).max - 1, 1], "UInt64", "signed", "UInt64"),
+        ([1, 1], "Int64", "unsigned", "UInt8"),
+        ([1.0, 1.0], "Float32", "unsigned", "UInt8"),
+        ([1.0, 1.1], "Float64", "unsigned", "Float64"),
+        ([1, pd.NA], "Int64", "unsigned", "UInt8"),
+        ([450, -300], "Int64", "unsigned", "Int64"),
+        ([-1, -1], "Int32", "unsigned", "Int32"),
+        ([1, 1], "Float64", "float", "Float32"),
+        ([1, 1.1], "Float64", "float", "Float32"),
+        ([1, 1], "Float32", "float", "Float32"),
+        ([1, 1.1], "Float32", "float", "Float32"),
+    ),
+)
+def test_downcast_nullable_numeric(data, input_dtype, downcast, expected_dtype):
+    arr = pd.array(data, dtype=input_dtype)
+    result = to_numeric(arr, downcast=downcast)
+    expected = pd.array(data, dtype=expected_dtype)
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_downcast_nullable_mask_is_copied():
+    # GH38974
+
+    arr = pd.array([1, 2, pd.NA], dtype="Int64")
+
+    result = to_numeric(arr, downcast="integer")
+    expected = pd.array([1, 2, pd.NA], dtype="Int8")
+    tm.assert_extension_array_equal(result, expected)
+
+    arr[1] = pd.NA  # should not modify result
+    tm.assert_extension_array_equal(result, expected)
+
+
+def test_to_numeric_scientific_notation():
+    # GH 15898
+    result = to_numeric("1.7e+308")
+    expected = np.float64(1.7e308)
+    assert result == expected
+
+
+@pytest.mark.parametrize("val", [9876543210.0, 2.0**128])
+def test_to_numeric_large_float_not_downcast_to_float_32(val):
+    # GH 19729
+    expected = Series([val])
+    result = to_numeric(expected, downcast="float")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "val, dtype", [(1, "Int64"), (1.5, "Float64"), (True, "boolean")]
+)
+def test_to_numeric_dtype_backend(val, dtype):
+    # GH#50505
+    ser = Series([val], dtype=object)
+    result = to_numeric(ser, dtype_backend="numpy_nullable")
+    expected = Series([val], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "val, dtype",
+    [
+        (1, "Int64"),
+        (1.5, "Float64"),
+        (True, "boolean"),
+        (1, "int64[pyarrow]"),
+        (1.5, "float64[pyarrow]"),
+        (True, "bool[pyarrow]"),
+    ],
+)
+def test_to_numeric_dtype_backend_na(val, dtype):
+    # GH#50505
+    if "pyarrow" in dtype:
+        pytest.importorskip("pyarrow")
+        dtype_backend = "pyarrow"
+    else:
+        dtype_backend = "numpy_nullable"
+    ser = Series([val, None], dtype=object)
+    result = to_numeric(ser, dtype_backend=dtype_backend)
+    expected = Series([val, pd.NA], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "val, dtype, downcast",
+    [
+        (1, "Int8", "integer"),
+        (1.5, "Float32", "float"),
+        (1, "Int8", "signed"),
+        (1, "int8[pyarrow]", "integer"),
+        (1.5, "float[pyarrow]", "float"),
+        (1, "int8[pyarrow]", "signed"),
+    ],
+)
+def test_to_numeric_dtype_backend_downcasting(val, dtype, downcast):
+    # GH#50505
+    if "pyarrow" in dtype:
+        pytest.importorskip("pyarrow")
+        dtype_backend = "pyarrow"
+    else:
+        dtype_backend = "numpy_nullable"
+    ser = Series([val, None], dtype=object)
+    result = to_numeric(ser, dtype_backend=dtype_backend, downcast=downcast)
+    expected = Series([val, pd.NA], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "smaller, dtype_backend",
+    [["UInt8", "numpy_nullable"], ["uint8[pyarrow]", "pyarrow"]],
+)
+def test_to_numeric_dtype_backend_downcasting_uint(smaller, dtype_backend):
+    # GH#50505
+    if dtype_backend == "pyarrow":
+        pytest.importorskip("pyarrow")
+    ser = Series([1, pd.NA], dtype="UInt64")
+    result = to_numeric(ser, dtype_backend=dtype_backend, downcast="unsigned")
+    expected = Series([1, pd.NA], dtype=smaller)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "Int64",
+        "UInt64",
+        "Float64",
+        "boolean",
+        "int64[pyarrow]",
+        "uint64[pyarrow]",
+        "float64[pyarrow]",
+        "bool[pyarrow]",
+    ],
+)
+def test_to_numeric_dtype_backend_already_nullable(dtype):
+    # GH#50505
+    if "pyarrow" in dtype:
+        pytest.importorskip("pyarrow")
+    ser = Series([1, pd.NA], dtype=dtype)
+    result = to_numeric(ser, dtype_backend="numpy_nullable")
+    expected = Series([1, pd.NA], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_to_numeric_dtype_backend_error(dtype_backend):
+    # GH#50505
+    ser = Series(["a", "b", ""])
+    expected = ser.copy()
+    with pytest.raises(ValueError, match="Unable to parse string"):
+        to_numeric(ser, dtype_backend=dtype_backend)
+
+    result = to_numeric(ser, dtype_backend=dtype_backend, errors="coerce")
+    if dtype_backend == "pyarrow":
+        dtype = "double[pyarrow]"
+    else:
+        dtype = "Float64"
+    expected = Series([np.nan, np.nan, np.nan], dtype=dtype)
+    tm.assert_series_equal(result, expected)
+
+
+def test_invalid_dtype_backend():
+    ser = Series([1, 2, 3])
+    msg = (
+        "dtype_backend numpy is invalid, only 'numpy_nullable' and "
+        "'pyarrow' are allowed."
+    )
+    with pytest.raises(ValueError, match=msg):
+        to_numeric(ser, dtype_backend="numpy")
+
+
+def test_coerce_pyarrow_backend():
+    # GH 52588
+    pa = pytest.importorskip("pyarrow")
+    ser = Series(list("12x"), dtype=ArrowDtype(pa.string()))
+    result = to_numeric(ser, errors="coerce", dtype_backend="pyarrow")
+    expected = Series([1, 2, None], dtype=ArrowDtype(pa.int64()))
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/tools/test_to_time.py b/pandas/pandas/tests/tools/test_to_time.py
new file mode 100644
index 0000000000000..5e61f5e1a3029
--- /dev/null
+++ b/pandas/pandas/tests/tools/test_to_time.py
@@ -0,0 +1,70 @@
+from datetime import time
+import locale
+
+import numpy as np
+import pytest
+
+from pandas.compat import PY311
+
+from pandas import Series
+import pandas._testing as tm
+from pandas.core.tools.times import to_time
+
+# The tests marked with this are locale-dependent.
+# They pass, except when the machine locale is zh_CN or it_IT.
+fails_on_non_english = pytest.mark.xfail(
+    locale.getlocale()[0] in ("zh_CN", "it_IT"),
+    reason="fail on a CI build with LC_ALL=zh_CN.utf8/it_IT.utf8",
+    strict=False,
+)
+
+
+class TestToTime:
+    @pytest.mark.parametrize(
+        "time_string",
+        [
+            "14:15",
+            "1415",
+            pytest.param("2:15pm", marks=fails_on_non_english),
+            pytest.param("0215pm", marks=fails_on_non_english),
+            "14:15:00",
+            "141500",
+            pytest.param("2:15:00pm", marks=fails_on_non_english),
+            pytest.param("021500pm", marks=fails_on_non_english),
+            time(14, 15),
+        ],
+    )
+    def test_parsers_time(self, time_string):
+        # GH#11818
+        assert to_time(time_string) == time(14, 15)
+
+    def test_odd_format(self):
+        new_string = "14.15"
+        msg = r"Cannot convert arg \['14\.15'\] to a time"
+        if not PY311:
+            with pytest.raises(ValueError, match=msg):
+                to_time(new_string)
+        assert to_time(new_string, format="%H.%M") == time(14, 15)
+
+    def test_arraylike(self):
+        arg = ["14:15", "20:20"]
+        expected_arr = [time(14, 15), time(20, 20)]
+        assert to_time(arg) == expected_arr
+        assert to_time(arg, format="%H:%M") == expected_arr
+        assert to_time(arg, infer_time_format=True) == expected_arr
+        assert to_time(arg, format="%I:%M%p", errors="coerce") == [None, None]
+
+        with pytest.raises(ValueError, match="errors must be"):
+            to_time(arg, format="%I:%M%p", errors="ignore")
+
+        msg = "Cannot convert.+to a time with given format"
+        with pytest.raises(ValueError, match=msg):
+            to_time(arg, format="%I:%M%p", errors="raise")
+
+        tm.assert_series_equal(
+            to_time(Series(arg, name="test")), Series(expected_arr, name="test")
+        )
+
+        res = to_time(np.array(arg))
+        assert isinstance(res, list)
+        assert res == expected_arr
diff --git a/pandas/pandas/tests/tools/test_to_timedelta.py b/pandas/pandas/tests/tools/test_to_timedelta.py
new file mode 100644
index 0000000000000..9ec2689069da9
--- /dev/null
+++ b/pandas/pandas/tests/tools/test_to_timedelta.py
@@ -0,0 +1,310 @@
+from datetime import (
+    time,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    WASM,
+)
+from pandas.errors import OutOfBoundsTimedelta
+
+import pandas as pd
+from pandas import (
+    Series,
+    TimedeltaIndex,
+    isna,
+    to_timedelta,
+)
+import pandas._testing as tm
+from pandas.core.arrays import TimedeltaArray
+
+
+class TestTimedeltas:
+    def test_to_timedelta_dt64_raises(self):
+        # Passing datetime64-dtype data to TimedeltaIndex is no longer
+        #  supported GH#29794
+        msg = r"dtype datetime64\[ns\] cannot be converted to timedelta64\[ns\]"
+
+        ser = Series([pd.NaT], dtype="M8[ns]")
+        with pytest.raises(TypeError, match=msg):
+            to_timedelta(ser)
+        with pytest.raises(TypeError, match=msg):
+            ser.to_frame().apply(to_timedelta)
+
+    def test_to_timedelta_readonly(self, writable):
+        # GH#34857
+        arr = np.array([], dtype=object)
+        arr.setflags(write=writable)
+        result = to_timedelta(arr)
+        expected = to_timedelta([])
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timedelta_null(self):
+        result = to_timedelta(["", ""])
+        assert isna(result).all()
+
+    def test_to_timedelta_same_np_timedelta64(self):
+        # pass thru
+        result = to_timedelta(np.array([np.timedelta64(1, "s")]))
+        expected = pd.Index(np.array([np.timedelta64(1, "s")]))
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timedelta_series(self):
+        # Series
+        expected = Series([timedelta(days=1), timedelta(days=1, seconds=1)])
+
+        msg = "'d' is deprecated and will be removed in a future version."
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = to_timedelta(Series(["1d", "1days 00:00:01"]))
+        tm.assert_series_equal(result, expected)
+
+    def test_to_timedelta_units(self):
+        # with units
+        result = TimedeltaIndex(
+            [np.timedelta64(0, "ns"), np.timedelta64(10, "s").astype("m8[ns]")]
+        )
+        expected = to_timedelta([0, 10], unit="s")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "dtype, unit",
+        [
+            ["int64", "s"],
+            ["int64", "m"],
+            ["int64", "h"],
+            ["timedelta64[s]", "s"],
+            ["timedelta64[D]", "D"],
+        ],
+    )
+    def test_to_timedelta_units_dtypes(self, dtype, unit):
+        # arrays of various dtypes
+        arr = np.array([1] * 5, dtype=dtype)
+        result = to_timedelta(arr, unit=unit)
+        exp_dtype = "m8[ns]" if dtype == "int64" else "m8[s]"
+        expected = TimedeltaIndex([np.timedelta64(1, unit)] * 5, dtype=exp_dtype)
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timedelta_oob_non_nano(self):
+        arr = np.array([pd.NaT._value + 1], dtype="timedelta64[m]")
+
+        msg = (
+            "Cannot convert -9223372036854775807 minutes to "
+            r"timedelta64\[s\] without overflow"
+        )
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            to_timedelta(arr)
+
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            TimedeltaIndex(arr)
+
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            TimedeltaArray._from_sequence(arr, dtype="m8[s]")
+
+    @pytest.mark.parametrize("box", [lambda x: x, pd.DataFrame])
+    @pytest.mark.parametrize("errors", ["raise", "coerce"])
+    def test_to_timedelta_dataframe(self, box, errors):
+        # GH 11776
+        arg = box(np.arange(10).reshape(2, 5))
+        with pytest.raises(TypeError, match="1-d array"):
+            to_timedelta(arg, errors=errors)
+
+    def test_to_timedelta_invalid_errors(self):
+        # bad value for errors parameter
+        msg = "errors must be one of"
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(["foo"], errors="never")
+
+    @pytest.mark.parametrize("arg", [[1, 2], 1])
+    def test_to_timedelta_invalid_unit(self, arg):
+        # these will error
+        msg = "invalid unit abbreviation: foo"
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(arg, unit="foo")
+
+    def test_to_timedelta_time(self):
+        # time not supported ATM
+        msg = (
+            "Value must be Timedelta, string, integer, float, timedelta or convertible"
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(time(second=1))
+        assert to_timedelta(time(second=1), errors="coerce") is pd.NaT
+
+    def test_to_timedelta_bad_value(self):
+        msg = "Could not convert 'foo' to NumPy timedelta"
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(["foo", "bar"])
+
+    def test_to_timedelta_bad_value_coerce(self):
+        tm.assert_index_equal(
+            TimedeltaIndex([pd.NaT, pd.NaT]),
+            to_timedelta(["foo", "bar"], errors="coerce"),
+        )
+
+        tm.assert_index_equal(
+            TimedeltaIndex(["1 day", pd.NaT, "1 min"]),
+            to_timedelta(["1 day", "bar", "1 min"], errors="coerce"),
+        )
+
+    @pytest.mark.parametrize(
+        "val, errors",
+        [
+            ("1M", True),
+            ("1 M", True),
+            ("1Y", True),
+            ("1 Y", True),
+            ("1y", True),
+            ("1 y", True),
+            ("1m", False),
+            ("1 m", False),
+            ("1 day", False),
+            ("2day", False),
+        ],
+    )
+    def test_unambiguous_timedelta_values(self, val, errors):
+        # GH36666 Deprecate use of strings denoting units with 'M', 'Y', 'm' or 'y'
+        # in pd.to_timedelta
+        msg = "Units 'M', 'Y' and 'y' do not represent unambiguous timedelta"
+        if errors:
+            with pytest.raises(ValueError, match=msg):
+                to_timedelta(val)
+        else:
+            # check it doesn't raise
+            to_timedelta(val)
+
+    def test_to_timedelta_via_apply(self):
+        # GH 5458
+        expected = Series([np.timedelta64(1, "s")])
+        result = Series(["00:00:01"]).apply(to_timedelta)
+        tm.assert_series_equal(result, expected)
+
+        result = Series([to_timedelta("00:00:01")])
+        tm.assert_series_equal(result, expected)
+
+    def test_to_timedelta_inference_without_warning(self):
+        # GH#41731 inference produces a warning in the Series constructor,
+        #  but _not_ in to_timedelta
+        vals = ["00:00:01", pd.NaT]
+        with tm.assert_produces_warning(None):
+            result = to_timedelta(vals)
+
+        expected = TimedeltaIndex([pd.Timedelta(seconds=1), pd.NaT])
+        tm.assert_index_equal(result, expected)
+
+    def test_to_timedelta_on_missing_values(self):
+        # GH5438
+        timedelta_NaT = np.timedelta64("NaT")
+
+        actual = to_timedelta(Series(["00:00:01", np.nan]))
+        expected = Series(
+            [np.timedelta64(1000000000, "ns"), timedelta_NaT],
+            dtype=f"{tm.ENDIAN}m8[ns]",
+        )
+        tm.assert_series_equal(actual, expected)
+
+        ser = Series(["00:00:01", pd.NaT], dtype="m8[ns]")
+        actual = to_timedelta(ser)
+        tm.assert_series_equal(actual, expected)
+
+    @pytest.mark.parametrize("val", [np.nan, pd.NaT, pd.NA])
+    def test_to_timedelta_on_missing_values_scalar(self, val):
+        actual = to_timedelta(val)
+        assert actual._value == np.timedelta64("NaT").astype("int64")
+
+    @pytest.mark.parametrize("val", [np.nan, pd.NaT, pd.NA])
+    def test_to_timedelta_on_missing_values_list(self, val):
+        actual = to_timedelta([val])
+        assert actual[0]._value == np.timedelta64("NaT").astype("int64")
+
+    @pytest.mark.skipif(WASM, reason="No fp exception support in WASM")
+    @pytest.mark.xfail(not IS64, reason="Floating point error")
+    def test_to_timedelta_float(self):
+        # https://github.com/pandas-dev/pandas/issues/25077
+        arr = np.arange(0, 1, 1e-6)[-10:]
+        result = to_timedelta(arr, unit="s")
+        expected_asi8 = np.arange(999990000, 10**9, 1000, dtype="int64")
+        tm.assert_numpy_array_equal(result.asi8, expected_asi8)
+
+    def test_to_timedelta_coerce_strings_unit(self):
+        arr = np.array([1, 2, "error"], dtype=object)
+        result = to_timedelta(arr, unit="ns", errors="coerce")
+        expected = to_timedelta([1, 2, pd.NaT], unit="ns")
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "expected_val, result_val", [[timedelta(days=2), 2], [None, None]]
+    )
+    def test_to_timedelta_nullable_int64_dtype(self, expected_val, result_val):
+        # GH 35574
+        expected = Series([timedelta(days=1), expected_val])
+        result = to_timedelta(Series([1, result_val], dtype="Int64"), unit="days")
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        ("input", "expected"),
+        [
+            ("8:53:08.71800000001", "8:53:08.718"),
+            ("8:53:08.718001", "8:53:08.718001"),
+            ("8:53:08.7180000001", "8:53:08.7180000001"),
+            ("-8:53:08.71800000001", "-8:53:08.718"),
+            ("8:53:08.7180000089", "8:53:08.718000008"),
+        ],
+    )
+    @pytest.mark.parametrize("func", [pd.Timedelta, to_timedelta])
+    def test_to_timedelta_precision_over_nanos(self, input, expected, func):
+        # GH: 36738
+        expected = pd.Timedelta(expected)
+        result = func(input)
+        assert result == expected
+
+    def test_to_timedelta_zerodim(self, fixed_now_ts):
+        # ndarray.item() incorrectly returns int for dt64[ns] and td64[ns]
+        dt64 = fixed_now_ts.to_datetime64()
+        arg = np.array(dt64)
+
+        msg = (
+            "Value must be Timedelta, string, integer, float, timedelta "
+            "or convertible, not datetime64"
+        )
+        with pytest.raises(ValueError, match=msg):
+            to_timedelta(arg)
+
+        arg2 = arg.view("m8[ns]")
+        result = to_timedelta(arg2)
+        assert isinstance(result, pd.Timedelta)
+        assert result._value == dt64.view("i8")
+
+    def test_to_timedelta_numeric_ea(self, any_numeric_ea_dtype):
+        # GH#48796
+        ser = Series([1, pd.NA], dtype=any_numeric_ea_dtype)
+        result = to_timedelta(ser)
+        expected = Series([pd.Timedelta(1, unit="ns"), pd.NaT])
+        tm.assert_series_equal(result, expected)
+
+    def test_to_timedelta_fraction(self):
+        result = to_timedelta(1.0 / 3, unit="h")
+        expected = pd.Timedelta("0 days 00:19:59.999999998")
+        assert result == expected
+
+
+def test_from_numeric_arrow_dtype(any_numeric_ea_dtype):
+    # GH 52425
+    pytest.importorskip("pyarrow")
+    ser = Series([1, 2], dtype=f"{any_numeric_ea_dtype.lower()}[pyarrow]")
+    result = to_timedelta(ser)
+    expected = Series([1, 2], dtype="timedelta64[ns]")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("unit", ["ns", "ms"])
+def test_from_timedelta_arrow_dtype(unit):
+    # GH 54298
+    pytest.importorskip("pyarrow")
+    expected = Series([timedelta(1)], dtype=f"duration[{unit}][pyarrow]")
+    result = to_timedelta(expected)
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/tseries/__init__.py b/pandas/pandas/tests/tseries/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tseries/frequencies/__init__.py b/pandas/pandas/tests/tseries/frequencies/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tseries/frequencies/test_freq_code.py b/pandas/pandas/tests/tseries/frequencies/test_freq_code.py
new file mode 100644
index 0000000000000..16b7190753ee2
--- /dev/null
+++ b/pandas/pandas/tests/tseries/frequencies/test_freq_code.py
@@ -0,0 +1,69 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    Period,
+    to_offset,
+)
+
+
+@pytest.mark.parametrize(
+    "freqstr,exp_freqstr",
+    [("D", "D"), ("W", "D"), ("ME", "D"), ("s", "s"), ("min", "s"), ("h", "s")],
+)
+def test_get_to_timestamp_base(freqstr, exp_freqstr):
+    off = to_offset(freqstr)
+    per = Period._from_ordinal(1, off)
+    exp_code = to_offset(exp_freqstr)._period_dtype_code
+
+    result_code = per._dtype._get_to_timestamp_base()
+    assert result_code == exp_code
+
+
+@pytest.mark.parametrize(
+    "args,expected",
+    [
+        ((1.5, "min"), (90, "s")),
+        ((62.4, "min"), (3744, "s")),
+        ((1.04, "h"), (3744, "s")),
+        ((1, "D"), (1, "D")),
+        ((0.342931, "h"), (1234551600, "us")),
+        ((1.2345, "D"), (106660800, "ms")),
+    ],
+)
+def test_resolution_bumping(args, expected):
+    # see gh-14378
+    off = to_offset(str(args[0]) + args[1])
+    assert off.n == expected[0]
+    assert off._prefix == expected[1]
+
+
+@pytest.mark.parametrize(
+    "args",
+    [
+        (0.5, "ns"),
+        # Too much precision in the input can prevent.
+        (0.3429324798798269273987982, "h"),
+    ],
+)
+def test_cat(args):
+    msg = "Invalid frequency"
+
+    with pytest.raises(ValueError, match=msg):
+        to_offset(str(args[0]) + args[1])
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected",
+    [
+        ("1h", "2021-01-01T09:00:00"),
+        ("1D", "2021-01-02T08:00:00"),
+        ("1W", "2021-01-03T08:00:00"),
+        ("1ME", "2021-01-31T08:00:00"),
+        ("1YE", "2021-12-31T08:00:00"),
+    ],
+)
+def test_compatibility(freqstr, expected):
+    ts_np = np.datetime64("2021-01-01T08:00:00.00")
+    do = to_offset(freqstr)
+    assert ts_np + do == np.datetime64(expected)
diff --git a/pandas/pandas/tests/tseries/frequencies/test_frequencies.py b/pandas/pandas/tests/tseries/frequencies/test_frequencies.py
new file mode 100644
index 0000000000000..f0af290b2fb69
--- /dev/null
+++ b/pandas/pandas/tests/tseries/frequencies/test_frequencies.py
@@ -0,0 +1,29 @@
+import pytest
+
+from pandas._libs.tslibs import offsets
+
+from pandas.tseries.frequencies import (
+    is_subperiod,
+    is_superperiod,
+)
+
+
+@pytest.mark.parametrize(
+    "p1,p2,expected",
+    [
+        # Input validation.
+        (offsets.MonthEnd(), None, False),
+        (offsets.YearEnd(), None, False),
+        (None, offsets.YearEnd(), False),
+        (None, offsets.MonthEnd(), False),
+        (None, None, False),
+        (offsets.YearEnd(), offsets.MonthEnd(), True),
+        (offsets.Hour(), offsets.Minute(), True),
+        (offsets.Second(), offsets.Milli(), True),
+        (offsets.Milli(), offsets.Micro(), True),
+        (offsets.Micro(), offsets.Nano(), True),
+    ],
+)
+def test_super_sub_symmetry(p1, p2, expected):
+    assert is_superperiod(p1, p2) is expected
+    assert is_subperiod(p2, p1) is expected
diff --git a/pandas/pandas/tests/tseries/frequencies/test_inference.py b/pandas/pandas/tests/tseries/frequencies/test_inference.py
new file mode 100644
index 0000000000000..dad5c73b89626
--- /dev/null
+++ b/pandas/pandas/tests/tseries/frequencies/test_inference.py
@@ -0,0 +1,544 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.ccalendar import (
+    DAYS,
+    MONTHS,
+)
+from pandas._libs.tslibs.offsets import _get_offset
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+from pandas.compat import is_platform_windows
+
+from pandas import (
+    DatetimeIndex,
+    Index,
+    RangeIndex,
+    Series,
+    Timestamp,
+    date_range,
+    period_range,
+)
+from pandas.core.arrays import (
+    DatetimeArray,
+    TimedeltaArray,
+)
+from pandas.core.tools.datetimes import to_datetime
+
+from pandas.tseries import (
+    frequencies,
+    offsets,
+)
+
+
+@pytest.fixture(
+    params=[
+        (timedelta(1), "D"),
+        (timedelta(hours=1), "h"),
+        (timedelta(minutes=1), "min"),
+        (timedelta(seconds=1), "s"),
+        (np.timedelta64(1, "ns"), "ns"),
+        (timedelta(microseconds=1), "us"),
+        (timedelta(microseconds=1000), "ms"),
+    ]
+)
+def base_delta_code_pair(request):
+    return request.param
+
+
+freqs = (
+    [f"QE-{month}" for month in MONTHS]
+    + [f"{annual}-{month}" for annual in ["YE", "BYE"] for month in MONTHS]
+    + ["ME", "BME", "BMS"]
+    + [f"WOM-{count}{day}" for count in range(1, 5) for day in DAYS]
+    + [f"W-{day}" for day in DAYS]
+)
+
+
+@pytest.mark.parametrize("freq", freqs)
+@pytest.mark.parametrize("periods", [5, 7])
+def test_infer_freq_range(periods, freq):
+    freq = freq.upper()
+
+    gen = date_range("1/1/2000", periods=periods, freq=freq)
+    index = DatetimeIndex(gen.values)
+
+    if not freq.startswith("QE-"):
+        assert frequencies.infer_freq(index) == gen.freqstr
+    else:
+        inf_freq = frequencies.infer_freq(index)
+        is_dec_range = inf_freq == "QE-DEC" and gen.freqstr in (
+            "QE",
+            "QE-DEC",
+            "QE-SEP",
+            "QE-JUN",
+            "QE-MAR",
+        )
+        is_nov_range = inf_freq == "QE-NOV" and gen.freqstr in (
+            "QE-NOV",
+            "QE-AUG",
+            "QE-MAY",
+            "QE-FEB",
+        )
+        is_oct_range = inf_freq == "QE-OCT" and gen.freqstr in (
+            "QE-OCT",
+            "QE-JUL",
+            "QE-APR",
+            "QE-JAN",
+        )
+        assert is_dec_range or is_nov_range or is_oct_range
+
+
+def test_raise_if_period_index():
+    index = period_range(start="1/1/1990", periods=20, freq="M")
+    msg = "Check the `freq` attribute instead of using infer_freq"
+
+    with pytest.raises(TypeError, match=msg):
+        frequencies.infer_freq(index)
+
+
+def test_raise_if_too_few():
+    index = DatetimeIndex(["12/31/1998", "1/3/1999"])
+    msg = "Need at least 3 dates to infer frequency"
+
+    with pytest.raises(ValueError, match=msg):
+        frequencies.infer_freq(index)
+
+
+def test_business_daily():
+    index = DatetimeIndex(["01/01/1999", "1/4/1999", "1/5/1999"])
+    assert frequencies.infer_freq(index) == "B"
+
+
+def test_business_daily_look_alike():
+    # see gh-16624
+    #
+    # Do not infer "B when "weekend" (2-day gap) in wrong place.
+    index = DatetimeIndex(["12/31/1998", "1/3/1999", "1/4/1999"])
+    assert frequencies.infer_freq(index) is None
+
+
+def test_day_corner():
+    index = DatetimeIndex(["1/1/2000", "1/2/2000", "1/3/2000"])
+    assert frequencies.infer_freq(index) == "D"
+
+
+def test_non_datetime_index():
+    dates = to_datetime(["1/1/2000", "1/2/2000", "1/3/2000"])
+    assert frequencies.infer_freq(dates) == "D"
+
+
+def test_fifth_week_of_month_infer():
+    # see gh-9425
+    #
+    # Only attempt to infer up to WOM-4.
+    index = DatetimeIndex(["2014-03-31", "2014-06-30", "2015-03-30"])
+    assert frequencies.infer_freq(index) is None
+
+
+def test_week_of_month_fake():
+    # All of these dates are on same day
+    # of week and are 4 or 5 weeks apart.
+    index = DatetimeIndex(["2013-08-27", "2013-10-01", "2013-10-29", "2013-11-26"])
+    assert frequencies.infer_freq(index) != "WOM-4TUE"
+
+
+def test_fifth_week_of_month():
+    # see gh-9425
+    #
+    # Only supports freq up to WOM-4.
+    msg = (
+        "Of the four parameters: start, end, periods, "
+        "and freq, exactly three must be specified"
+    )
+
+    with pytest.raises(ValueError, match=msg):
+        date_range("2014-01-01", freq="WOM-5MON")
+
+
+def test_monthly_ambiguous():
+    rng = DatetimeIndex(["1/31/2000", "2/29/2000", "3/31/2000"])
+    assert rng.inferred_freq == "ME"
+
+
+def test_annual_ambiguous():
+    rng = DatetimeIndex(["1/31/2000", "1/31/2001", "1/31/2002"])
+    assert rng.inferred_freq == "YE-JAN"
+
+
+@pytest.mark.parametrize("count", range(1, 5))
+def test_infer_freq_delta(base_delta_code_pair, count):
+    b = Timestamp(datetime.now())
+    base_delta, code = base_delta_code_pair
+
+    inc = base_delta * count
+    index = DatetimeIndex([b + inc * j for j in range(3)])
+
+    exp_freq = f"{count:d}{code}" if count > 1 else code
+    assert frequencies.infer_freq(index) == exp_freq
+
+
+@pytest.mark.parametrize(
+    "constructor",
+    [
+        lambda now, delta: DatetimeIndex(
+            [now + delta * 7] + [now + delta * j for j in range(3)]
+        ),
+        lambda now, delta: DatetimeIndex(
+            [now + delta * j for j in range(3)] + [now + delta * 7]
+        ),
+    ],
+)
+def test_infer_freq_custom(base_delta_code_pair, constructor):
+    b = Timestamp(datetime.now())
+    base_delta, _ = base_delta_code_pair
+
+    index = constructor(b, base_delta)
+    assert frequencies.infer_freq(index) is None
+
+
+@pytest.mark.parametrize(
+    "expected,dates",
+    list(
+        {
+            "YS-JAN": ["2009-01-01", "2010-01-01", "2011-01-01", "2012-01-01"],
+            "QE-OCT": ["2009-01-31", "2009-04-30", "2009-07-31", "2009-10-31"],
+            "ME": ["2010-11-30", "2010-12-31", "2011-01-31", "2011-02-28"],
+            "W-SAT": ["2010-12-25", "2011-01-01", "2011-01-08", "2011-01-15"],
+            "D": ["2011-01-01", "2011-01-02", "2011-01-03", "2011-01-04"],
+            "h": [
+                "2011-12-31 22:00",
+                "2011-12-31 23:00",
+                "2012-01-01 00:00",
+                "2012-01-01 01:00",
+            ],
+        }.items()
+    ),
+)
+def test_infer_freq_tz(tz_naive_fixture, expected, dates, unit):
+    # see gh-7310, GH#55609
+    tz = tz_naive_fixture
+    idx = DatetimeIndex(dates, tz=tz).as_unit(unit)
+    assert idx.inferred_freq == expected
+
+
+def test_infer_freq_tz_series(tz_naive_fixture):
+    # infer_freq should work with both tz-naive and tz-aware series. See gh-52456
+    tz = tz_naive_fixture
+    idx = date_range("2021-01-01", "2021-01-04", tz=tz)
+    series = idx.to_series().reset_index(drop=True)
+    inferred_freq = frequencies.infer_freq(series)
+    assert inferred_freq == "D"
+
+
+@pytest.mark.parametrize(
+    "date_pair",
+    [
+        ["2013-11-02", "2013-11-5"],  # Fall DST
+        ["2014-03-08", "2014-03-11"],  # Spring DST
+        ["2014-01-01", "2014-01-03"],  # Regular Time
+    ],
+)
+@pytest.mark.parametrize(
+    "freq",
+    ["h", "3h", "10min", "3601s", "3600001ms", "3600000001us", "3600000000001ns"],
+)
+def test_infer_freq_tz_transition(tz_naive_fixture, date_pair, freq):
+    # see gh-8772
+    tz = tz_naive_fixture
+    idx = date_range(date_pair[0], date_pair[1], freq=freq, tz=tz)
+    assert idx.inferred_freq == freq
+
+
+def test_infer_freq_tz_transition_custom():
+    index = date_range("2013-11-03", periods=5, freq="3h").tz_localize(
+        "America/Chicago"
+    )
+    assert index.inferred_freq is None
+
+
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        # Hourly freq in a day must result in "h"
+        (
+            [
+                "2014-07-01 09:00",
+                "2014-07-01 10:00",
+                "2014-07-01 11:00",
+                "2014-07-01 12:00",
+                "2014-07-01 13:00",
+                "2014-07-01 14:00",
+            ],
+            "h",
+        ),
+        (
+            [
+                "2014-07-01 09:00",
+                "2014-07-01 10:00",
+                "2014-07-01 11:00",
+                "2014-07-01 12:00",
+                "2014-07-01 13:00",
+                "2014-07-01 14:00",
+                "2014-07-01 15:00",
+                "2014-07-01 16:00",
+                "2014-07-02 09:00",
+                "2014-07-02 10:00",
+                "2014-07-02 11:00",
+            ],
+            "bh",
+        ),
+        (
+            [
+                "2014-07-04 09:00",
+                "2014-07-04 10:00",
+                "2014-07-04 11:00",
+                "2014-07-04 12:00",
+                "2014-07-04 13:00",
+                "2014-07-04 14:00",
+                "2014-07-04 15:00",
+                "2014-07-04 16:00",
+                "2014-07-07 09:00",
+                "2014-07-07 10:00",
+                "2014-07-07 11:00",
+            ],
+            "bh",
+        ),
+        (
+            [
+                "2014-07-04 09:00",
+                "2014-07-04 10:00",
+                "2014-07-04 11:00",
+                "2014-07-04 12:00",
+                "2014-07-04 13:00",
+                "2014-07-04 14:00",
+                "2014-07-04 15:00",
+                "2014-07-04 16:00",
+                "2014-07-07 09:00",
+                "2014-07-07 10:00",
+                "2014-07-07 11:00",
+                "2014-07-07 12:00",
+                "2014-07-07 13:00",
+                "2014-07-07 14:00",
+                "2014-07-07 15:00",
+                "2014-07-07 16:00",
+                "2014-07-08 09:00",
+                "2014-07-08 10:00",
+                "2014-07-08 11:00",
+                "2014-07-08 12:00",
+                "2014-07-08 13:00",
+                "2014-07-08 14:00",
+                "2014-07-08 15:00",
+                "2014-07-08 16:00",
+            ],
+            "bh",
+        ),
+    ],
+)
+def test_infer_freq_business_hour(data, expected):
+    # see gh-7905
+    idx = DatetimeIndex(data)
+    assert idx.inferred_freq == expected
+
+
+def test_not_monotonic():
+    rng = DatetimeIndex(["1/31/2000", "1/31/2001", "1/31/2002"])
+    rng = rng[::-1]
+
+    assert rng.inferred_freq == "-1YE-JAN"
+
+
+def test_non_datetime_index2():
+    rng = DatetimeIndex(["1/31/2000", "1/31/2001", "1/31/2002"])
+    vals = rng.to_pydatetime()
+
+    result = frequencies.infer_freq(vals)
+    assert result == rng.inferred_freq
+
+
+@pytest.mark.parametrize(
+    "idx",
+    [
+        Index(np.arange(5), dtype=np.int64),
+        Index(np.arange(5), dtype=np.float64),
+        period_range("2020-01-01", periods=5),
+        RangeIndex(5),
+    ],
+)
+def test_invalid_index_types(idx):
+    # see gh-48439
+    msg = "|".join(
+        [
+            "cannot infer freq from a non-convertible",
+            "Check the `freq` attribute instead of using infer_freq",
+        ]
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        frequencies.infer_freq(idx)
+
+
+@pytest.mark.skipif(is_platform_windows(), reason="see gh-10822: Windows issue")
+def test_invalid_index_types_unicode():
+    # see gh-10822
+    #
+    # Odd error message on conversions to datetime for unicode.
+    msg = "Unknown datetime string format"
+
+    with pytest.raises(ValueError, match=msg):
+        frequencies.infer_freq(Index(["ZqgszYBfuL"]))
+
+
+def test_string_datetime_like_compat():
+    # see gh-6463
+    data = ["2004-01", "2004-02", "2004-03", "2004-04"]
+
+    expected = frequencies.infer_freq(data)
+    result = frequencies.infer_freq(Index(data))
+
+    assert result == expected
+
+
+def test_series():
+    # see gh-6407
+    s = Series(date_range("20130101", "20130110"))
+    inferred = frequencies.infer_freq(s)
+    assert inferred == "D"
+
+
+@pytest.mark.parametrize("end", [10, 10.0])
+def test_series_invalid_type(end):
+    # see gh-6407
+    msg = "cannot infer freq from a non-convertible dtype on a Series"
+    s = Series(np.arange(end))
+
+    with pytest.raises(TypeError, match=msg):
+        frequencies.infer_freq(s)
+
+
+def test_series_inconvertible_string(using_infer_string):
+    # see gh-6407
+    if using_infer_string:
+        msg = "cannot infer freq from"
+
+        with pytest.raises(TypeError, match=msg):
+            frequencies.infer_freq(Series(["foo", "bar"]))
+    else:
+        msg = "Unknown datetime string format"
+
+        with pytest.raises(ValueError, match=msg):
+            frequencies.infer_freq(Series(["foo", "bar"]))
+
+
+@pytest.mark.parametrize("freq", [None, "ms"])
+def test_series_period_index(freq):
+    # see gh-6407
+    #
+    # Cannot infer on PeriodIndex
+    msg = "cannot infer freq from a non-convertible dtype on a Series"
+    s = Series(period_range("2013", periods=10, freq=freq))
+
+    with pytest.raises(TypeError, match=msg):
+        frequencies.infer_freq(s)
+
+
+@pytest.mark.parametrize("freq", ["ME", "ms", "s"])
+def test_series_datetime_index(freq):
+    s = Series(date_range("20130101", periods=10, freq=freq))
+    inferred = frequencies.infer_freq(s)
+    assert inferred == freq
+
+
+@pytest.mark.parametrize(
+    "offset_func",
+    [
+        _get_offset,
+        lambda freq: date_range("2011-01-01", periods=5, freq=freq),
+    ],
+)
+@pytest.mark.parametrize(
+    "freq",
+    [
+        "WEEKDAY",
+        "EOM",
+        "W@MON",
+        "W@TUE",
+        "W@WED",
+        "W@THU",
+        "W@FRI",
+        "W@SAT",
+        "W@SUN",
+        "QE@JAN",
+        "QE@FEB",
+        "QE@MAR",
+        "YE@JAN",
+        "YE@FEB",
+        "YE@MAR",
+        "YE@APR",
+        "YE@MAY",
+        "YE@JUN",
+        "YE@JUL",
+        "YE@AUG",
+        "YE@SEP",
+        "YE@OCT",
+        "YE@NOV",
+        "YE@DEC",
+        "WOM@1MON",
+        "WOM@2MON",
+        "WOM@3MON",
+        "WOM@4MON",
+        "WOM@1TUE",
+        "WOM@2TUE",
+        "WOM@3TUE",
+        "WOM@4TUE",
+        "WOM@1WED",
+        "WOM@2WED",
+        "WOM@3WED",
+        "WOM@4WED",
+        "WOM@1THU",
+        "WOM@2THU",
+        "WOM@3THU",
+        "WOM@4THU",
+        "WOM@1FRI",
+        "WOM@2FRI",
+        "WOM@3FRI",
+        "WOM@4FRI",
+    ],
+)
+def test_legacy_offset_warnings(offset_func, freq):
+    with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+        offset_func(freq)
+
+
+def test_ms_vs_capital_ms():
+    left = _get_offset("ms")
+    right = _get_offset("MS")
+
+    assert left == offsets.Milli()
+    assert right == offsets.MonthBegin()
+
+
+def test_infer_freq_non_nano():
+    arr = np.arange(10).astype(np.int64).view("M8[s]")
+    dta = DatetimeArray._simple_new(arr, dtype=arr.dtype)
+    res = frequencies.infer_freq(dta)
+    assert res == "s"
+
+    arr2 = arr.view("m8[ms]")
+    tda = TimedeltaArray._simple_new(arr2, dtype=arr2.dtype)
+    res2 = frequencies.infer_freq(tda)
+    assert res2 == "ms"
+
+
+def test_infer_freq_non_nano_tzaware(tz_aware_fixture):
+    tz = tz_aware_fixture
+
+    dti = date_range("2016-01-01", periods=365, freq="B", tz=tz)
+    dta = dti._data.as_unit("s")
+
+    res = frequencies.infer_freq(dta)
+    assert res == "B"
diff --git a/pandas/pandas/tests/tseries/holiday/__init__.py b/pandas/pandas/tests/tseries/holiday/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tseries/holiday/test_calendar.py b/pandas/pandas/tests/tseries/holiday/test_calendar.py
new file mode 100644
index 0000000000000..90e2e117852a2
--- /dev/null
+++ b/pandas/pandas/tests/tseries/holiday/test_calendar.py
@@ -0,0 +1,119 @@
+from datetime import datetime
+
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    offsets,
+    to_datetime,
+)
+import pandas._testing as tm
+
+from pandas.tseries.holiday import (
+    AbstractHolidayCalendar,
+    Holiday,
+    Timestamp,
+    USFederalHolidayCalendar,
+    USLaborDay,
+    USThanksgivingDay,
+    get_calendar,
+)
+
+
+@pytest.mark.parametrize(
+    "transform", [lambda x: x, lambda x: x.strftime("%Y-%m-%d"), lambda x: Timestamp(x)]
+)
+def test_calendar(transform):
+    start_date = datetime(2012, 1, 1)
+    end_date = datetime(2012, 12, 31)
+
+    calendar = USFederalHolidayCalendar()
+    holidays = calendar.holidays(transform(start_date), transform(end_date))
+
+    expected = [
+        datetime(2012, 1, 2),
+        datetime(2012, 1, 16),
+        datetime(2012, 2, 20),
+        datetime(2012, 5, 28),
+        datetime(2012, 7, 4),
+        datetime(2012, 9, 3),
+        datetime(2012, 10, 8),
+        datetime(2012, 11, 12),
+        datetime(2012, 11, 22),
+        datetime(2012, 12, 25),
+    ]
+
+    assert list(holidays.to_pydatetime()) == expected
+
+
+def test_calendar_caching():
+    # see gh-9552.
+
+    class TestCalendar(AbstractHolidayCalendar):
+        def __init__(self, name=None, rules=None) -> None:
+            super().__init__(name=name, rules=rules)
+
+    jan1 = TestCalendar(rules=[Holiday("jan1", year=2015, month=1, day=1)])
+    jan2 = TestCalendar(rules=[Holiday("jan2", year=2015, month=1, day=2)])
+
+    # Getting holidays for Jan 1 should not alter results for Jan 2.
+    expected = DatetimeIndex(["01-Jan-2015"]).as_unit("us")
+    tm.assert_index_equal(jan1.holidays(), expected)
+
+    expected2 = DatetimeIndex(["02-Jan-2015"]).as_unit("us")
+    tm.assert_index_equal(jan2.holidays(), expected2)
+
+
+def test_calendar_observance_dates():
+    # see gh-11477
+    us_fed_cal = get_calendar("USFederalHolidayCalendar")
+    holidays0 = us_fed_cal.holidays(
+        datetime(2015, 7, 3), datetime(2015, 7, 3)
+    )  # <-- same start and end dates
+    holidays1 = us_fed_cal.holidays(
+        datetime(2015, 7, 3), datetime(2015, 7, 6)
+    )  # <-- different start and end dates
+    holidays2 = us_fed_cal.holidays(
+        datetime(2015, 7, 3), datetime(2015, 7, 3)
+    )  # <-- same start and end dates
+
+    # These should all produce the same result.
+    #
+    # In addition, calling with different start and end
+    # dates should not alter the output if we call the
+    # function again with the same start and end date.
+    tm.assert_index_equal(holidays0, holidays1)
+    tm.assert_index_equal(holidays0, holidays2)
+
+
+def test_rule_from_name():
+    us_fed_cal = get_calendar("USFederalHolidayCalendar")
+    assert us_fed_cal.rule_from_name("Thanksgiving Day") == USThanksgivingDay
+
+
+def test_calendar_2031():
+    # See gh-27790
+    #
+    # Labor Day 2031 is on September 1. Saturday before is August 30.
+    # Next working day after August 30 ought to be Tuesday, September 2.
+
+    class testCalendar(AbstractHolidayCalendar):
+        rules = [USLaborDay]
+
+    cal = testCalendar()
+    workDay = offsets.CustomBusinessDay(calendar=cal)
+    Sat_before_Labor_Day_2031 = to_datetime("2031-08-30")
+    next_working_day = Sat_before_Labor_Day_2031 + 0 * workDay
+    assert next_working_day == to_datetime("2031-09-02")
+
+
+def test_no_holidays_calendar():
+    # Test for issue #31415
+
+    class NoHolidaysCalendar(AbstractHolidayCalendar):
+        pass
+
+    cal = NoHolidaysCalendar()
+    holidays = cal.holidays(Timestamp("01-Jan-2020"), Timestamp("01-Jan-2021"))
+    empty_index = DatetimeIndex([])  # Type is DatetimeIndex since return_name=False
+    tm.assert_index_equal(holidays, empty_index)
diff --git a/pandas/pandas/tests/tseries/holiday/test_federal.py b/pandas/pandas/tests/tseries/holiday/test_federal.py
new file mode 100644
index 0000000000000..2565877f8a2a4
--- /dev/null
+++ b/pandas/pandas/tests/tseries/holiday/test_federal.py
@@ -0,0 +1,58 @@
+from datetime import datetime
+
+from pandas import DatetimeIndex
+import pandas._testing as tm
+
+from pandas.tseries.holiday import (
+    AbstractHolidayCalendar,
+    USFederalHolidayCalendar,
+    USMartinLutherKingJr,
+    USMemorialDay,
+)
+
+
+def test_no_mlk_before_1986():
+    # see gh-10278
+    class MLKCalendar(AbstractHolidayCalendar):
+        rules = [USMartinLutherKingJr]
+
+    holidays = MLKCalendar().holidays(start="1984", end="1988").to_pydatetime().tolist()
+
+    # Testing to make sure holiday is not incorrectly observed before 1986.
+    assert holidays == [datetime(1986, 1, 20, 0, 0), datetime(1987, 1, 19, 0, 0)]
+
+
+def test_memorial_day():
+    class MemorialDay(AbstractHolidayCalendar):
+        rules = [USMemorialDay]
+
+    holidays = MemorialDay().holidays(start="1971", end="1980").to_pydatetime().tolist()
+
+    # Fixes 5/31 error and checked manually against Wikipedia.
+    assert holidays == [
+        datetime(1971, 5, 31, 0, 0),
+        datetime(1972, 5, 29, 0, 0),
+        datetime(1973, 5, 28, 0, 0),
+        datetime(1974, 5, 27, 0, 0),
+        datetime(1975, 5, 26, 0, 0),
+        datetime(1976, 5, 31, 0, 0),
+        datetime(1977, 5, 30, 0, 0),
+        datetime(1978, 5, 29, 0, 0),
+        datetime(1979, 5, 28, 0, 0),
+    ]
+
+
+def test_federal_holiday_inconsistent_returntype():
+    # GH 49075 test case
+    # Instantiate two calendars to rule out _cache
+    cal1 = USFederalHolidayCalendar()
+    cal2 = USFederalHolidayCalendar()
+
+    results_2018 = cal1.holidays(start=datetime(2018, 8, 1), end=datetime(2018, 8, 31))
+    results_2019 = cal2.holidays(start=datetime(2019, 8, 1), end=datetime(2019, 8, 31))
+    expected_results = DatetimeIndex([], dtype="datetime64[ns]", freq=None)
+
+    # Check against expected results to ensure both date
+    # ranges generate expected results as per GH49075 submission
+    tm.assert_index_equal(results_2018, expected_results)
+    tm.assert_index_equal(results_2019, expected_results)
diff --git a/pandas/pandas/tests/tseries/holiday/test_holiday.py b/pandas/pandas/tests/tseries/holiday/test_holiday.py
new file mode 100644
index 0000000000000..ffe6ff0b51bcf
--- /dev/null
+++ b/pandas/pandas/tests/tseries/holiday/test_holiday.py
@@ -0,0 +1,355 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import pytest
+
+from pandas import (
+    DatetimeIndex,
+    Series,
+)
+import pandas._testing as tm
+
+from pandas.tseries.holiday import (
+    MO,
+    SA,
+    AbstractHolidayCalendar,
+    DateOffset,
+    EasterMonday,
+    GoodFriday,
+    Holiday,
+    HolidayCalendarFactory,
+    Timestamp,
+    USColumbusDay,
+    USFederalHolidayCalendar,
+    USLaborDay,
+    USMartinLutherKingJr,
+    USMemorialDay,
+    USPresidentsDay,
+    USThanksgivingDay,
+    get_calendar,
+    next_monday,
+)
+
+
+@pytest.mark.parametrize(
+    "holiday,start_date,end_date,expected",
+    [
+        (
+            USMemorialDay,
+            datetime(2011, 1, 1),
+            datetime(2020, 12, 31),
+            [
+                datetime(2011, 5, 30),
+                datetime(2012, 5, 28),
+                datetime(2013, 5, 27),
+                datetime(2014, 5, 26),
+                datetime(2015, 5, 25),
+                datetime(2016, 5, 30),
+                datetime(2017, 5, 29),
+                datetime(2018, 5, 28),
+                datetime(2019, 5, 27),
+                datetime(2020, 5, 25),
+            ],
+        ),
+        (
+            Holiday("July 4th Eve", month=7, day=3),
+            "2001-01-01",
+            "2003-03-03",
+            [Timestamp("2001-07-03 00:00:00"), Timestamp("2002-07-03 00:00:00")],
+        ),
+        (
+            Holiday("July 4th Eve", month=7, day=3, days_of_week=(0, 1, 2, 3)),
+            "2001-01-01",
+            "2008-03-03",
+            [
+                Timestamp("2001-07-03 00:00:00"),
+                Timestamp("2002-07-03 00:00:00"),
+                Timestamp("2003-07-03 00:00:00"),
+                Timestamp("2006-07-03 00:00:00"),
+                Timestamp("2007-07-03 00:00:00"),
+            ],
+        ),
+        (
+            EasterMonday,
+            datetime(2011, 1, 1),
+            datetime(2020, 12, 31),
+            [
+                Timestamp("2011-04-25 00:00:00"),
+                Timestamp("2012-04-09 00:00:00"),
+                Timestamp("2013-04-01 00:00:00"),
+                Timestamp("2014-04-21 00:00:00"),
+                Timestamp("2015-04-06 00:00:00"),
+                Timestamp("2016-03-28 00:00:00"),
+                Timestamp("2017-04-17 00:00:00"),
+                Timestamp("2018-04-02 00:00:00"),
+                Timestamp("2019-04-22 00:00:00"),
+                Timestamp("2020-04-13 00:00:00"),
+            ],
+        ),
+        (
+            GoodFriday,
+            datetime(2011, 1, 1),
+            datetime(2020, 12, 31),
+            [
+                Timestamp("2011-04-22 00:00:00"),
+                Timestamp("2012-04-06 00:00:00"),
+                Timestamp("2013-03-29 00:00:00"),
+                Timestamp("2014-04-18 00:00:00"),
+                Timestamp("2015-04-03 00:00:00"),
+                Timestamp("2016-03-25 00:00:00"),
+                Timestamp("2017-04-14 00:00:00"),
+                Timestamp("2018-03-30 00:00:00"),
+                Timestamp("2019-04-19 00:00:00"),
+                Timestamp("2020-04-10 00:00:00"),
+            ],
+        ),
+        (
+            USThanksgivingDay,
+            datetime(2011, 1, 1),
+            datetime(2020, 12, 31),
+            [
+                datetime(2011, 11, 24),
+                datetime(2012, 11, 22),
+                datetime(2013, 11, 28),
+                datetime(2014, 11, 27),
+                datetime(2015, 11, 26),
+                datetime(2016, 11, 24),
+                datetime(2017, 11, 23),
+                datetime(2018, 11, 22),
+                datetime(2019, 11, 28),
+                datetime(2020, 11, 26),
+            ],
+        ),
+    ],
+)
+def test_holiday_dates(holiday, start_date, end_date, expected):
+    assert list(holiday.dates(start_date, end_date)) == expected
+
+    # Verify that timezone info is preserved.
+    assert list(
+        holiday.dates(
+            Timestamp(start_date, tz=timezone.utc), Timestamp(end_date, tz=timezone.utc)
+        )
+    ) == [dt.replace(tzinfo=timezone.utc) for dt in expected]
+
+
+@pytest.mark.parametrize(
+    "holiday,start,expected",
+    [
+        (USMemorialDay, datetime(2015, 7, 1), []),
+        (USMemorialDay, "2015-05-25", [Timestamp("2015-05-25")]),
+        (USLaborDay, datetime(2015, 7, 1), []),
+        (USLaborDay, "2015-09-07", [Timestamp("2015-09-07")]),
+        (USColumbusDay, datetime(2015, 7, 1), []),
+        (USColumbusDay, "2015-10-12", [Timestamp("2015-10-12")]),
+        (USThanksgivingDay, datetime(2015, 7, 1), []),
+        (USThanksgivingDay, "2015-11-26", [Timestamp("2015-11-26")]),
+        (USMartinLutherKingJr, datetime(2015, 7, 1), []),
+        (USMartinLutherKingJr, "2015-01-19", [Timestamp("2015-01-19")]),
+        (USPresidentsDay, datetime(2015, 7, 1), []),
+        (USPresidentsDay, "2015-02-16", [Timestamp("2015-02-16")]),
+        (GoodFriday, datetime(2015, 7, 1), []),
+        (GoodFriday, "2015-04-03", [Timestamp("2015-04-03")]),
+        (EasterMonday, "2015-04-06", [Timestamp("2015-04-06")]),
+        (EasterMonday, datetime(2015, 7, 1), []),
+        (EasterMonday, "2015-04-05", []),
+        ("New Year's Day", "2015-01-01", [Timestamp("2015-01-01")]),
+        ("New Year's Day", "2010-12-31", [Timestamp("2010-12-31")]),
+        ("New Year's Day", datetime(2015, 7, 1), []),
+        ("New Year's Day", "2011-01-01", []),
+        ("Independence Day", "2015-07-03", [Timestamp("2015-07-03")]),
+        ("Independence Day", datetime(2015, 7, 1), []),
+        ("Independence Day", "2015-07-04", []),
+        ("Veterans Day", "2012-11-12", [Timestamp("2012-11-12")]),
+        ("Veterans Day", datetime(2015, 7, 1), []),
+        ("Veterans Day", "2012-11-11", []),
+        ("Christmas Day", "2011-12-26", [Timestamp("2011-12-26")]),
+        ("Christmas Day", datetime(2015, 7, 1), []),
+        ("Christmas Day", "2011-12-25", []),
+        ("Juneteenth National Independence Day", "2020-06-19", []),
+        (
+            "Juneteenth National Independence Day",
+            "2021-06-18",
+            [Timestamp("2021-06-18")],
+        ),
+        ("Juneteenth National Independence Day", "2022-06-19", []),
+        (
+            "Juneteenth National Independence Day",
+            "2022-06-20",
+            [Timestamp("2022-06-20")],
+        ),
+    ],
+)
+def test_holidays_within_dates(holiday, start, expected):
+    # see gh-11477
+    #
+    # Fix holiday behavior where holiday.dates returned dates outside
+    # start/end date, or observed rules could not be applied because the
+    # holiday was not in the original date range (e.g., 7/4/2015 -> 7/3/2015).
+    if isinstance(holiday, str):
+        calendar = get_calendar("USFederalHolidayCalendar")
+        holiday = calendar.rule_from_name(holiday)
+
+    assert list(holiday.dates(start, start)) == expected
+
+    # Verify that timezone info is preserved.
+    assert list(
+        holiday.dates(
+            Timestamp(start, tz=timezone.utc), Timestamp(start, tz=timezone.utc)
+        )
+    ) == [dt.replace(tzinfo=timezone.utc) for dt in expected]
+
+
+@pytest.mark.parametrize(
+    "transform", [lambda x: x.strftime("%Y-%m-%d"), lambda x: Timestamp(x)]
+)
+def test_argument_types(transform):
+    start_date = datetime(2011, 1, 1)
+    end_date = datetime(2020, 12, 31)
+
+    holidays = USThanksgivingDay.dates(start_date, end_date)
+    holidays2 = USThanksgivingDay.dates(transform(start_date), transform(end_date))
+    tm.assert_index_equal(holidays, holidays2)
+
+
+@pytest.mark.parametrize(
+    "name,kwargs",
+    [
+        ("One-Time", {"year": 2012, "month": 5, "day": 28}),
+        (
+            "Range",
+            {
+                "month": 5,
+                "day": 28,
+                "start_date": datetime(2012, 1, 1),
+                "end_date": datetime(2012, 12, 31),
+                "offset": DateOffset(weekday=MO(1)),
+            },
+        ),
+    ],
+)
+def test_special_holidays(name, kwargs):
+    base_date = [datetime(2012, 5, 28)]
+    holiday = Holiday(name, **kwargs)
+
+    start_date = datetime(2011, 1, 1)
+    end_date = datetime(2020, 12, 31)
+
+    assert base_date == holiday.dates(start_date, end_date)
+
+
+def test_get_calendar():
+    class TestCalendar(AbstractHolidayCalendar):
+        rules = []
+
+    calendar = get_calendar("TestCalendar")
+    assert TestCalendar == type(calendar)
+
+
+def test_factory():
+    class_1 = HolidayCalendarFactory(
+        "MemorialDay", AbstractHolidayCalendar, USMemorialDay
+    )
+    class_2 = HolidayCalendarFactory(
+        "Thanksgiving", AbstractHolidayCalendar, USThanksgivingDay
+    )
+    class_3 = HolidayCalendarFactory("Combined", class_1, class_2)
+
+    assert len(class_1.rules) == 1
+    assert len(class_2.rules) == 1
+    assert len(class_3.rules) == 2
+
+
+def test_both_offset_observance_raises():
+    # see gh-10217
+    msg = "Cannot use both offset and observance"
+    with pytest.raises(NotImplementedError, match=msg):
+        Holiday(
+            "Cyber Monday",
+            month=11,
+            day=1,
+            offset=[DateOffset(weekday=SA(4))],
+            observance=next_monday,
+        )
+
+
+def test_list_of_list_of_offsets_raises():
+    # see gh-29049
+    # Test that the offsets of offsets are forbidden
+    holiday1 = Holiday(
+        "Holiday1",
+        month=USThanksgivingDay.month,
+        day=USThanksgivingDay.day,
+        offset=[USThanksgivingDay.offset, DateOffset(1)],
+    )
+    msg = "Only BaseOffsets and flat lists of them are supported for offset."
+    with pytest.raises(ValueError, match=msg):
+        Holiday(
+            "Holiday2",
+            month=holiday1.month,
+            day=holiday1.day,
+            offset=[holiday1.offset, DateOffset(3)],
+        )
+
+
+def test_half_open_interval_with_observance():
+    # Prompted by GH 49075
+    # Check for holidays that have a half-open date interval where
+    # they have either a start_date or end_date defined along
+    # with a defined observance pattern to make sure that the return type
+    # for Holiday.dates() remains consistent before & after the year that
+    # marks the 'edge' of the half-open date interval.
+
+    holiday_1 = Holiday(
+        "Arbitrary Holiday - start 2022-03-14",
+        start_date=datetime(2022, 3, 14),
+        month=3,
+        day=14,
+        observance=next_monday,
+    )
+    holiday_2 = Holiday(
+        "Arbitrary Holiday 2 - end 2022-03-20",
+        end_date=datetime(2022, 3, 20),
+        month=3,
+        day=20,
+        observance=next_monday,
+    )
+
+    class TestHolidayCalendar(AbstractHolidayCalendar):
+        rules = [
+            USMartinLutherKingJr,
+            holiday_1,
+            holiday_2,
+            USLaborDay,
+        ]
+
+    start = Timestamp("2022-08-01")
+    end = Timestamp("2022-08-31")
+    year_offset = DateOffset(years=5)
+    expected_results = DatetimeIndex([], dtype="datetime64[ns]", freq=None)
+    test_cal = TestHolidayCalendar()
+
+    date_interval_low = test_cal.holidays(start - year_offset, end - year_offset)
+    date_window_edge = test_cal.holidays(start, end)
+    date_interval_high = test_cal.holidays(start + year_offset, end + year_offset)
+
+    tm.assert_index_equal(date_interval_low, expected_results)
+    tm.assert_index_equal(date_window_edge, expected_results)
+    tm.assert_index_equal(date_interval_high, expected_results)
+
+
+def test_holidays_with_timezone_specified_but_no_occurences():
+    # GH 54580
+    # _apply_rule() in holiday.py was silently dropping timezones if you passed it
+    # an empty list of holiday dates that had timezone information
+    start_date = Timestamp("2018-01-01", tz="America/Chicago")
+    end_date = Timestamp("2018-01-11", tz="America/Chicago")
+    test_case = USFederalHolidayCalendar().holidays(
+        start_date, end_date, return_name=True
+    )
+    expected_results = Series("New Year's Day", index=[start_date])
+    expected_results.index = expected_results.index.as_unit("ns")
+
+    tm.assert_equal(test_case, expected_results)
diff --git a/pandas/pandas/tests/tseries/holiday/test_observance.py b/pandas/pandas/tests/tseries/holiday/test_observance.py
new file mode 100644
index 0000000000000..83038ad254b77
--- /dev/null
+++ b/pandas/pandas/tests/tseries/holiday/test_observance.py
@@ -0,0 +1,105 @@
+from datetime import datetime
+
+import pytest
+
+from pandas.tseries.holiday import (
+    after_nearest_workday,
+    before_nearest_workday,
+    nearest_workday,
+    next_monday,
+    next_monday_or_tuesday,
+    next_workday,
+    previous_friday,
+    previous_workday,
+    sunday_to_monday,
+    weekend_to_monday,
+)
+
+_WEDNESDAY = datetime(2014, 4, 9)
+_THURSDAY = datetime(2014, 4, 10)
+_FRIDAY = datetime(2014, 4, 11)
+_SATURDAY = datetime(2014, 4, 12)
+_SUNDAY = datetime(2014, 4, 13)
+_MONDAY = datetime(2014, 4, 14)
+_TUESDAY = datetime(2014, 4, 15)
+_NEXT_WEDNESDAY = datetime(2014, 4, 16)
+
+
+@pytest.mark.parametrize("day", [_SATURDAY, _SUNDAY])
+def test_next_monday(day):
+    assert next_monday(day) == _MONDAY
+
+
+@pytest.mark.parametrize(
+    "day,expected", [(_SATURDAY, _MONDAY), (_SUNDAY, _TUESDAY), (_MONDAY, _TUESDAY)]
+)
+def test_next_monday_or_tuesday(day, expected):
+    assert next_monday_or_tuesday(day) == expected
+
+
+@pytest.mark.parametrize("day", [_SATURDAY, _SUNDAY])
+def test_previous_friday(day):
+    assert previous_friday(day) == _FRIDAY
+
+
+def test_sunday_to_monday():
+    assert sunday_to_monday(_SUNDAY) == _MONDAY
+
+
+@pytest.mark.parametrize(
+    "day,expected", [(_SATURDAY, _FRIDAY), (_SUNDAY, _MONDAY), (_MONDAY, _MONDAY)]
+)
+def test_nearest_workday(day, expected):
+    assert nearest_workday(day) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected", [(_SATURDAY, _MONDAY), (_SUNDAY, _MONDAY), (_MONDAY, _MONDAY)]
+)
+def test_weekend_to_monday(day, expected):
+    assert weekend_to_monday(day) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected",
+    [
+        (_WEDNESDAY, _THURSDAY),
+        (_THURSDAY, _FRIDAY),
+        (_SATURDAY, _MONDAY),
+        (_SUNDAY, _MONDAY),
+        (_MONDAY, _TUESDAY),
+        (_TUESDAY, _NEXT_WEDNESDAY),  # WED is same week as TUE
+    ],
+)
+def test_next_workday(day, expected):
+    assert next_workday(day) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected", [(_SATURDAY, _FRIDAY), (_SUNDAY, _FRIDAY), (_TUESDAY, _MONDAY)]
+)
+def test_previous_workday(day, expected):
+    assert previous_workday(day) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected",
+    [
+        (_THURSDAY, _WEDNESDAY),
+        (_FRIDAY, _THURSDAY),
+        (_SATURDAY, _THURSDAY),
+        (_SUNDAY, _FRIDAY),
+        (_MONDAY, _FRIDAY),  # last week Friday
+        (_TUESDAY, _MONDAY),
+        (_NEXT_WEDNESDAY, _TUESDAY),  # WED is same week as TUE
+    ],
+)
+def test_before_nearest_workday(day, expected):
+    assert before_nearest_workday(day) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected", [(_SATURDAY, _MONDAY), (_SUNDAY, _TUESDAY), (_FRIDAY, _MONDAY)]
+)
+def test_after_nearest_workday(day, expected):
+    assert after_nearest_workday(day) == expected
diff --git a/pandas/pandas/tests/tseries/offsets/__init__.py b/pandas/pandas/tests/tseries/offsets/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tseries/offsets/common.py b/pandas/pandas/tests/tseries/offsets/common.py
new file mode 100644
index 0000000000000..e0838cceb4c7b
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/common.py
@@ -0,0 +1,38 @@
+"""
+Assertion helpers and base class for offsets tests
+"""
+
+from __future__ import annotations
+
+
+def assert_offset_equal(offset, base, expected):
+    actual = offset + base
+    actual_swapped = base + offset
+    actual_apply = offset._apply(base)
+    try:
+        assert actual == expected
+        assert actual_swapped == expected
+        assert actual_apply == expected
+    except AssertionError as err:
+        raise AssertionError(
+            f"\nExpected: {expected}\nActual: {actual}\nFor Offset: {offset})"
+            f"\nAt Date: {base}"
+        ) from err
+
+
+def assert_is_on_offset(offset, date, expected):
+    actual = offset.is_on_offset(date)
+    assert actual == expected, (
+        f"\nExpected: {expected}\nActual: {actual}\nFor Offset: {offset})"
+        f"\nAt Date: {date}"
+    )
+
+
+class WeekDay:
+    MON = 0
+    TUE = 1
+    WED = 2
+    THU = 3
+    FRI = 4
+    SAT = 5
+    SUN = 6
diff --git a/pandas/pandas/tests/tseries/offsets/data/cday-0.14.1.pickle b/pandas/pandas/tests/tseries/offsets/data/cday-0.14.1.pickle
new file mode 100644
index 0000000000000..48488099482e4
Binary files /dev/null and b/pandas/pandas/tests/tseries/offsets/data/cday-0.14.1.pickle differ
diff --git a/pandas/pandas/tests/tseries/offsets/test_business_day.py b/pandas/pandas/tests/tseries/offsets/test_business_day.py
new file mode 100644
index 0000000000000..b1ab5fc64804b
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_business_day.py
@@ -0,0 +1,237 @@
+"""
+Tests for offsets.BDay
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import (
+    ApplyTypeError,
+    BDay,
+    BMonthEnd,
+)
+
+from pandas import (
+    DatetimeIndex,
+    Timedelta,
+    _testing as tm,
+)
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries import offsets
+
+
+@pytest.fixture
+def dt():
+    return datetime(2008, 1, 1)
+
+
+@pytest.fixture
+def _offset():
+    return BDay
+
+
+@pytest.fixture
+def offset(_offset):
+    return _offset()
+
+
+@pytest.fixture
+def offset2(_offset):
+    return _offset(2)
+
+
+class TestBusinessDay:
+    def test_different_normalize_equals(self, _offset, offset2):
+        # GH#21404 changed __eq__ to return False when `normalize` does not match
+        offset = _offset()
+        offset2 = _offset(normalize=True)
+        assert offset != offset2
+
+    def test_repr(self, offset, offset2):
+        assert repr(offset) == "<BusinessDay>"
+        assert repr(offset2) == "<2 * BusinessDays>"
+
+        expected = "<BusinessDay: offset=datetime.timedelta(days=1)>"
+        assert repr(offset + timedelta(1)) == expected
+
+    def test_with_offset(self, dt, offset):
+        offset = offset + timedelta(hours=2)
+
+        assert (dt + offset) == datetime(2008, 1, 2, 2)
+
+    @pytest.mark.parametrize(
+        "td",
+        [
+            Timedelta(hours=2),
+            Timedelta(hours=2).to_pytimedelta(),
+            Timedelta(hours=2).to_timedelta64(),
+        ],
+        ids=lambda x: type(x),
+    )
+    def test_with_offset_index(self, td, dt, offset):
+        dti = DatetimeIndex([dt])
+        expected = DatetimeIndex([datetime(2008, 1, 2, 2)])
+
+        result = dti + (td + offset)
+        tm.assert_index_equal(result, expected)
+
+        result = dti + (offset + td)
+        tm.assert_index_equal(result, expected)
+
+    def test_eq(self, offset2):
+        assert offset2 == offset2
+
+    def test_hash(self, offset2):
+        assert hash(offset2) == hash(offset2)
+
+    def test_add_datetime(self, dt, offset2):
+        assert offset2 + dt == datetime(2008, 1, 3)
+        assert offset2 + np.datetime64("2008-01-01 00:00:00") == datetime(2008, 1, 3)
+
+    def testRollback1(self, dt, _offset):
+        assert _offset(10).rollback(dt) == dt
+
+    def testRollback2(self, _offset):
+        assert _offset(10).rollback(datetime(2008, 1, 5)) == datetime(2008, 1, 4)
+
+    def testRollforward1(self, dt, _offset):
+        assert _offset(10).rollforward(dt) == dt
+
+    def testRollforward2(self, _offset):
+        assert _offset(10).rollforward(datetime(2008, 1, 5)) == datetime(2008, 1, 7)
+
+    def test_roll_date_object(self, offset):
+        dt = date(2012, 9, 15)
+
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 9, 14)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 9, 17)
+
+        offset = offsets.Day()
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 9, 15)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 9, 15)
+
+    @pytest.mark.parametrize(
+        "dt, expected",
+        [
+            (datetime(2008, 1, 1), True),
+            (datetime(2008, 1, 5), False),
+        ],
+    )
+    def test_is_on_offset(self, offset, dt, expected):
+        assert_is_on_offset(offset, dt, expected)
+
+    apply_cases: list[tuple[int, dict[datetime, datetime]]] = [
+        (
+            1,
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 2),
+                datetime(2008, 1, 4): datetime(2008, 1, 7),
+                datetime(2008, 1, 5): datetime(2008, 1, 7),
+                datetime(2008, 1, 6): datetime(2008, 1, 7),
+                datetime(2008, 1, 7): datetime(2008, 1, 8),
+            },
+        ),
+        (
+            2,
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 3),
+                datetime(2008, 1, 4): datetime(2008, 1, 8),
+                datetime(2008, 1, 5): datetime(2008, 1, 8),
+                datetime(2008, 1, 6): datetime(2008, 1, 8),
+                datetime(2008, 1, 7): datetime(2008, 1, 9),
+            },
+        ),
+        (
+            -1,
+            {
+                datetime(2008, 1, 1): datetime(2007, 12, 31),
+                datetime(2008, 1, 4): datetime(2008, 1, 3),
+                datetime(2008, 1, 5): datetime(2008, 1, 4),
+                datetime(2008, 1, 6): datetime(2008, 1, 4),
+                datetime(2008, 1, 7): datetime(2008, 1, 4),
+                datetime(2008, 1, 8): datetime(2008, 1, 7),
+            },
+        ),
+        (
+            -2,
+            {
+                datetime(2008, 1, 1): datetime(2007, 12, 28),
+                datetime(2008, 1, 4): datetime(2008, 1, 2),
+                datetime(2008, 1, 5): datetime(2008, 1, 3),
+                datetime(2008, 1, 6): datetime(2008, 1, 3),
+                datetime(2008, 1, 7): datetime(2008, 1, 3),
+                datetime(2008, 1, 8): datetime(2008, 1, 4),
+                datetime(2008, 1, 9): datetime(2008, 1, 7),
+            },
+        ),
+        (
+            0,
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 1, 4): datetime(2008, 1, 4),
+                datetime(2008, 1, 5): datetime(2008, 1, 7),
+                datetime(2008, 1, 6): datetime(2008, 1, 7),
+                datetime(2008, 1, 7): datetime(2008, 1, 7),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", apply_cases)
+    def test_apply(self, case, _offset):
+        n, cases = case
+        offset = _offset(n)
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_apply_large_n(self, _offset):
+        dt = datetime(2012, 10, 23)
+
+        result = dt + _offset(10)
+        assert result == datetime(2012, 11, 6)
+
+        result = dt + _offset(100) - _offset(100)
+        assert result == dt
+
+        off = _offset() * 6
+        rs = datetime(2012, 1, 1) - off
+        xp = datetime(2011, 12, 23)
+        assert rs == xp
+
+        st = datetime(2011, 12, 18)
+        rs = st + off
+        xp = datetime(2011, 12, 26)
+        assert rs == xp
+
+        off = _offset() * 10
+        rs = datetime(2014, 1, 5) + off  # see #5890
+        xp = datetime(2014, 1, 17)
+        assert rs == xp
+
+    def test_apply_corner(self, _offset):
+        if _offset is BDay:
+            msg = "Only know how to combine business day with datetime or timedelta"
+        else:
+            msg = (
+                "Only know how to combine trading day "
+                "with datetime, datetime64 or timedelta"
+            )
+        with pytest.raises(ApplyTypeError, match=msg):
+            _offset()._apply(BMonthEnd())
diff --git a/pandas/pandas/tests/tseries/offsets/test_business_hour.py b/pandas/pandas/tests/tseries/offsets/test_business_hour.py
new file mode 100644
index 0000000000000..1b488dc9a47d4
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_business_hour.py
@@ -0,0 +1,1451 @@
+"""
+Tests for offsets.BusinessHour
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    time as dt_time,
+)
+
+import pytest
+
+from pandas._libs.tslibs import (
+    Timedelta,
+    Timestamp,
+)
+from pandas._libs.tslibs.offsets import (
+    BDay,
+    BusinessHour,
+    Nano,
+)
+
+from pandas import (
+    DatetimeIndex,
+    _testing as tm,
+    date_range,
+)
+from pandas.tests.tseries.offsets.common import assert_offset_equal
+
+
+@pytest.fixture
+def dt():
+    return datetime(2014, 7, 1, 10, 00)
+
+
+@pytest.fixture
+def _offset():
+    return BusinessHour
+
+
+@pytest.fixture
+def offset1():
+    return BusinessHour()
+
+
+@pytest.fixture
+def offset2():
+    return BusinessHour(n=3)
+
+
+@pytest.fixture
+def offset3():
+    return BusinessHour(n=-1)
+
+
+@pytest.fixture
+def offset4():
+    return BusinessHour(n=-4)
+
+
+@pytest.fixture
+def offset5():
+    return BusinessHour(start=dt_time(11, 0), end=dt_time(14, 30))
+
+
+@pytest.fixture
+def offset6():
+    return BusinessHour(start="20:00", end="05:00")
+
+
+@pytest.fixture
+def offset7():
+    return BusinessHour(n=-2, start=dt_time(21, 30), end=dt_time(6, 30))
+
+
+@pytest.fixture
+def offset8():
+    return BusinessHour(start=["09:00", "13:00"], end=["12:00", "17:00"])
+
+
+@pytest.fixture
+def offset9():
+    return BusinessHour(n=3, start=["09:00", "22:00"], end=["13:00", "03:00"])
+
+
+@pytest.fixture
+def offset10():
+    return BusinessHour(n=-1, start=["23:00", "13:00"], end=["02:00", "17:00"])
+
+
+class TestBusinessHour:
+    @pytest.mark.parametrize(
+        "start,end,match",
+        [
+            (
+                dt_time(11, 0, 5),
+                "17:00",
+                "time data must be specified only with hour and minute",
+            ),
+            ("AAA", "17:00", "time data must match '%H:%M' format"),
+            ("14:00:05", "17:00", "time data must match '%H:%M' format"),
+            ([], "17:00", "Must include at least 1 start time"),
+            ("09:00", [], "Must include at least 1 end time"),
+            (
+                ["09:00", "11:00"],
+                "17:00",
+                "number of starting time and ending time must be the same",
+            ),
+            (
+                ["09:00", "11:00"],
+                ["10:00"],
+                "number of starting time and ending time must be the same",
+            ),
+            (
+                ["09:00", "11:00"],
+                ["12:00", "20:00"],
+                r"invalid starting and ending time\(s\): opening hours should not "
+                "touch or overlap with one another",
+            ),
+            (
+                ["12:00", "20:00"],
+                ["09:00", "11:00"],
+                r"invalid starting and ending time\(s\): opening hours should not "
+                "touch or overlap with one another",
+            ),
+        ],
+    )
+    def test_constructor_errors(self, start, end, match):
+        with pytest.raises(ValueError, match=match):
+            BusinessHour(start=start, end=end)
+
+    def test_different_normalize_equals(self, _offset):
+        # GH#21404 changed __eq__ to return False when `normalize` does not match
+        offset = _offset()
+        offset2 = _offset(normalize=True)
+        assert offset != offset2
+
+    def test_repr(
+        self,
+        offset1,
+        offset2,
+        offset3,
+        offset4,
+        offset5,
+        offset6,
+        offset7,
+        offset8,
+        offset9,
+        offset10,
+    ):
+        assert repr(offset1) == "<BusinessHour: bh=09:00-17:00>"
+        assert repr(offset2) == "<3 * BusinessHours: bh=09:00-17:00>"
+        assert repr(offset3) == "<-1 * BusinessHour: bh=09:00-17:00>"
+        assert repr(offset4) == "<-4 * BusinessHours: bh=09:00-17:00>"
+
+        assert repr(offset5) == "<BusinessHour: bh=11:00-14:30>"
+        assert repr(offset6) == "<BusinessHour: bh=20:00-05:00>"
+        assert repr(offset7) == "<-2 * BusinessHours: bh=21:30-06:30>"
+        assert repr(offset8) == "<BusinessHour: bh=09:00-12:00,13:00-17:00>"
+        assert repr(offset9) == "<3 * BusinessHours: bh=09:00-13:00,22:00-03:00>"
+        assert repr(offset10) == "<-1 * BusinessHour: bh=13:00-17:00,23:00-02:00>"
+
+    def test_with_offset(self, dt):
+        expected = Timestamp("2014-07-01 13:00")
+
+        assert dt + BusinessHour() * 3 == expected
+        assert dt + BusinessHour(n=3) == expected
+
+    @pytest.mark.parametrize(
+        "offset_name",
+        ["offset1", "offset2", "offset3", "offset4", "offset8", "offset9", "offset10"],
+    )
+    def test_eq_attribute(self, offset_name, request):
+        offset = request.getfixturevalue(offset_name)
+        assert offset == offset
+
+    @pytest.mark.parametrize(
+        "offset1,offset2",
+        [
+            (BusinessHour(start="09:00"), BusinessHour()),
+            (
+                BusinessHour(start=["23:00", "13:00"], end=["12:00", "17:00"]),
+                BusinessHour(start=["13:00", "23:00"], end=["17:00", "12:00"]),
+            ),
+        ],
+    )
+    def test_eq(self, offset1, offset2):
+        assert offset1 == offset2
+
+    @pytest.mark.parametrize(
+        "offset1,offset2",
+        [
+            (BusinessHour(), BusinessHour(-1)),
+            (BusinessHour(start="09:00"), BusinessHour(start="09:01")),
+            (
+                BusinessHour(start="09:00", end="17:00"),
+                BusinessHour(start="17:00", end="09:01"),
+            ),
+            (
+                BusinessHour(start=["13:00", "23:00"], end=["18:00", "07:00"]),
+                BusinessHour(start=["13:00", "23:00"], end=["17:00", "12:00"]),
+            ),
+        ],
+    )
+    def test_neq(self, offset1, offset2):
+        assert offset1 != offset2
+
+    @pytest.mark.parametrize(
+        "offset_name",
+        ["offset1", "offset2", "offset3", "offset4", "offset8", "offset9", "offset10"],
+    )
+    def test_hash(self, offset_name, request):
+        offset = request.getfixturevalue(offset_name)
+        assert offset == offset
+
+    def test_add_datetime(
+        self,
+        dt,
+        offset1,
+        offset2,
+        offset3,
+        offset4,
+        offset8,
+        offset9,
+        offset10,
+    ):
+        assert offset1 + dt == datetime(2014, 7, 1, 11)
+        assert offset2 + dt == datetime(2014, 7, 1, 13)
+        assert offset3 + dt == datetime(2014, 6, 30, 17)
+        assert offset4 + dt == datetime(2014, 6, 30, 14)
+        assert offset8 + dt == datetime(2014, 7, 1, 11)
+        assert offset9 + dt == datetime(2014, 7, 1, 22)
+        assert offset10 + dt == datetime(2014, 7, 1, 1)
+
+    def test_sub(self, dt, offset2, _offset):
+        off = offset2
+        msg = "Cannot subtract datetime from offset"
+        with pytest.raises(TypeError, match=msg):
+            off - dt
+        assert 2 * off - off == off
+
+        assert dt - offset2 == dt + _offset(-3)
+
+    def test_multiply_by_zero(self, dt, offset1, offset2):
+        assert dt - 0 * offset1 == dt
+        assert dt + 0 * offset1 == dt
+        assert dt - 0 * offset2 == dt
+        assert dt + 0 * offset2 == dt
+
+    def testRollback1(
+        self,
+        dt,
+        _offset,
+        offset1,
+        offset2,
+        offset3,
+        offset4,
+        offset5,
+        offset6,
+        offset7,
+        offset8,
+        offset9,
+        offset10,
+    ):
+        assert offset1.rollback(dt) == dt
+        assert offset2.rollback(dt) == dt
+        assert offset3.rollback(dt) == dt
+        assert offset4.rollback(dt) == dt
+        assert offset5.rollback(dt) == datetime(2014, 6, 30, 14, 30)
+        assert offset6.rollback(dt) == datetime(2014, 7, 1, 5, 0)
+        assert offset7.rollback(dt) == datetime(2014, 7, 1, 6, 30)
+        assert offset8.rollback(dt) == dt
+        assert offset9.rollback(dt) == dt
+        assert offset10.rollback(dt) == datetime(2014, 7, 1, 2)
+
+        datet = datetime(2014, 7, 1, 0)
+        assert offset1.rollback(datet) == datetime(2014, 6, 30, 17)
+        assert offset2.rollback(datet) == datetime(2014, 6, 30, 17)
+        assert offset3.rollback(datet) == datetime(2014, 6, 30, 17)
+        assert offset4.rollback(datet) == datetime(2014, 6, 30, 17)
+        assert offset5.rollback(datet) == datetime(2014, 6, 30, 14, 30)
+        assert offset6.rollback(datet) == datet
+        assert offset7.rollback(datet) == datet
+        assert offset8.rollback(datet) == datetime(2014, 6, 30, 17)
+        assert offset9.rollback(datet) == datet
+        assert offset10.rollback(datet) == datet
+
+        assert _offset(5).rollback(dt) == dt
+
+    def testRollback2(self, _offset):
+        assert _offset(-3).rollback(datetime(2014, 7, 5, 15, 0)) == datetime(
+            2014, 7, 4, 17, 0
+        )
+
+    def testRollforward1(
+        self,
+        dt,
+        _offset,
+        offset1,
+        offset2,
+        offset3,
+        offset4,
+        offset5,
+        offset6,
+        offset7,
+        offset8,
+        offset9,
+        offset10,
+    ):
+        assert offset1.rollforward(dt) == dt
+        assert offset2.rollforward(dt) == dt
+        assert offset3.rollforward(dt) == dt
+        assert offset4.rollforward(dt) == dt
+        assert offset5.rollforward(dt) == datetime(2014, 7, 1, 11, 0)
+        assert offset6.rollforward(dt) == datetime(2014, 7, 1, 20, 0)
+        assert offset7.rollforward(dt) == datetime(2014, 7, 1, 21, 30)
+        assert offset8.rollforward(dt) == dt
+        assert offset9.rollforward(dt) == dt
+        assert offset10.rollforward(dt) == datetime(2014, 7, 1, 13)
+
+        datet = datetime(2014, 7, 1, 0)
+        assert offset1.rollforward(datet) == datetime(2014, 7, 1, 9)
+        assert offset2.rollforward(datet) == datetime(2014, 7, 1, 9)
+        assert offset3.rollforward(datet) == datetime(2014, 7, 1, 9)
+        assert offset4.rollforward(datet) == datetime(2014, 7, 1, 9)
+        assert offset5.rollforward(datet) == datetime(2014, 7, 1, 11)
+        assert offset6.rollforward(datet) == datet
+        assert offset7.rollforward(datet) == datet
+        assert offset8.rollforward(datet) == datetime(2014, 7, 1, 9)
+        assert offset9.rollforward(datet) == datet
+        assert offset10.rollforward(datet) == datet
+
+        assert _offset(5).rollforward(dt) == dt
+
+    def testRollforward2(self, _offset):
+        assert _offset(-3).rollforward(datetime(2014, 7, 5, 16, 0)) == datetime(
+            2014, 7, 7, 9
+        )
+
+    def test_roll_date_object(self):
+        offset = BusinessHour()
+
+        dt = datetime(2014, 7, 6, 15, 0)
+
+        result = offset.rollback(dt)
+        assert result == datetime(2014, 7, 4, 17)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2014, 7, 7, 9)
+
+    normalize_cases = []
+    normalize_cases.append(
+        (
+            BusinessHour(normalize=True),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 2),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 2),
+                datetime(2014, 7, 1, 0): datetime(2014, 7, 1),
+                datetime(2014, 7, 4, 15): datetime(2014, 7, 4),
+                datetime(2014, 7, 4, 15, 59): datetime(2014, 7, 4),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7),
+                datetime(2014, 7, 5, 23): datetime(2014, 7, 7),
+                datetime(2014, 7, 6, 10): datetime(2014, 7, 7),
+            },
+        )
+    )
+
+    normalize_cases.append(
+        (
+            BusinessHour(-1, normalize=True),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 6, 30),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 10): datetime(2014, 6, 30),
+                datetime(2014, 7, 1, 0): datetime(2014, 6, 30),
+                datetime(2014, 7, 7, 10): datetime(2014, 7, 4),
+                datetime(2014, 7, 7, 10, 1): datetime(2014, 7, 7),
+                datetime(2014, 7, 5, 23): datetime(2014, 7, 4),
+                datetime(2014, 7, 6, 10): datetime(2014, 7, 4),
+            },
+        )
+    )
+
+    normalize_cases.append(
+        (
+            BusinessHour(1, normalize=True, start="17:00", end="04:00"),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 2),
+                datetime(2014, 7, 2, 2): datetime(2014, 7, 2),
+                datetime(2014, 7, 2, 3): datetime(2014, 7, 2),
+                datetime(2014, 7, 4, 23): datetime(2014, 7, 5),
+                datetime(2014, 7, 5, 2): datetime(2014, 7, 5),
+                datetime(2014, 7, 7, 2): datetime(2014, 7, 7),
+                datetime(2014, 7, 7, 17): datetime(2014, 7, 7),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", normalize_cases)
+    def test_normalize(self, case):
+        offset, cases = case
+        for dt, expected in cases.items():
+            assert offset._apply(dt) == expected
+
+    on_offset_cases = []
+    on_offset_cases.append(
+        (
+            BusinessHour(),
+            {
+                datetime(2014, 7, 1, 9): True,
+                datetime(2014, 7, 1, 8, 59): False,
+                datetime(2014, 7, 1, 8): False,
+                datetime(2014, 7, 1, 17): True,
+                datetime(2014, 7, 1, 17, 1): False,
+                datetime(2014, 7, 1, 18): False,
+                datetime(2014, 7, 5, 9): False,
+                datetime(2014, 7, 6, 12): False,
+            },
+        )
+    )
+
+    on_offset_cases.append(
+        (
+            BusinessHour(start="10:00", end="15:00"),
+            {
+                datetime(2014, 7, 1, 9): False,
+                datetime(2014, 7, 1, 10): True,
+                datetime(2014, 7, 1, 15): True,
+                datetime(2014, 7, 1, 15, 1): False,
+                datetime(2014, 7, 5, 12): False,
+                datetime(2014, 7, 6, 12): False,
+            },
+        )
+    )
+
+    on_offset_cases.append(
+        (
+            BusinessHour(start="19:00", end="05:00"),
+            {
+                datetime(2014, 7, 1, 9, 0): False,
+                datetime(2014, 7, 1, 10, 0): False,
+                datetime(2014, 7, 1, 15): False,
+                datetime(2014, 7, 1, 15, 1): False,
+                datetime(2014, 7, 5, 12, 0): False,
+                datetime(2014, 7, 6, 12, 0): False,
+                datetime(2014, 7, 1, 19, 0): True,
+                datetime(2014, 7, 2, 0, 0): True,
+                datetime(2014, 7, 4, 23): True,
+                datetime(2014, 7, 5, 1): True,
+                datetime(2014, 7, 5, 5, 0): True,
+                datetime(2014, 7, 6, 23, 0): False,
+                datetime(2014, 7, 7, 3, 0): False,
+            },
+        )
+    )
+
+    on_offset_cases.append(
+        (
+            BusinessHour(start=["09:00", "13:00"], end=["12:00", "17:00"]),
+            {
+                datetime(2014, 7, 1, 9): True,
+                datetime(2014, 7, 1, 8, 59): False,
+                datetime(2014, 7, 1, 8): False,
+                datetime(2014, 7, 1, 17): True,
+                datetime(2014, 7, 1, 17, 1): False,
+                datetime(2014, 7, 1, 18): False,
+                datetime(2014, 7, 5, 9): False,
+                datetime(2014, 7, 6, 12): False,
+                datetime(2014, 7, 1, 12, 30): False,
+            },
+        )
+    )
+
+    on_offset_cases.append(
+        (
+            BusinessHour(start=["19:00", "23:00"], end=["21:00", "05:00"]),
+            {
+                datetime(2014, 7, 1, 9, 0): False,
+                datetime(2014, 7, 1, 10, 0): False,
+                datetime(2014, 7, 1, 15): False,
+                datetime(2014, 7, 1, 15, 1): False,
+                datetime(2014, 7, 5, 12, 0): False,
+                datetime(2014, 7, 6, 12, 0): False,
+                datetime(2014, 7, 1, 19, 0): True,
+                datetime(2014, 7, 2, 0, 0): True,
+                datetime(2014, 7, 4, 23): True,
+                datetime(2014, 7, 5, 1): True,
+                datetime(2014, 7, 5, 5, 0): True,
+                datetime(2014, 7, 6, 23, 0): False,
+                datetime(2014, 7, 7, 3, 0): False,
+                datetime(2014, 7, 4, 22): False,
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, cases = case
+        for dt, expected in cases.items():
+            assert offset.is_on_offset(dt) == expected
+
+    apply_cases = [
+        (
+            BusinessHour(),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 12),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 1, 19): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 2, 9),
+                datetime(2014, 7, 1, 16, 30, 15): datetime(2014, 7, 2, 9, 30, 15),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 2, 12),
+                # out of business hours
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 10),
+                # saturday
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 10),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 10),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7, 9, 30),
+                datetime(2014, 7, 4, 16, 30, 30): datetime(2014, 7, 7, 9, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(4),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 2, 9),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 2, 11),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 2, 12),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 13),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 13),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7, 12, 30),
+                datetime(2014, 7, 4, 16, 30, 30): datetime(2014, 7, 7, 12, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(-1),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 10),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 1, 12),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 1, 10): datetime(2014, 6, 30, 17),
+                datetime(2014, 7, 1, 16, 30, 15): datetime(2014, 7, 1, 15, 30, 15),
+                datetime(2014, 7, 1, 9, 30, 15): datetime(2014, 6, 30, 16, 30, 15),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 1, 5): datetime(2014, 6, 30, 16),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 2, 10),
+                # out of business hours
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 16),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 2, 16),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 16),
+                # saturday
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 4, 16),
+                datetime(2014, 7, 7, 9): datetime(2014, 7, 4, 16),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 4, 16, 30),
+                datetime(2014, 7, 7, 9, 30, 30): datetime(2014, 7, 4, 16, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(-4),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 6, 30, 15),
+                datetime(2014, 7, 1, 13): datetime(2014, 6, 30, 17),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 1, 11),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1, 12),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 13),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 1, 13),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 4, 13),
+                datetime(2014, 7, 4, 18): datetime(2014, 7, 4, 13),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 4, 13, 30),
+                datetime(2014, 7, 7, 9, 30, 30): datetime(2014, 7, 4, 13, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(start="13:00", end="16:00"),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 1, 19): datetime(2014, 7, 2, 14),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 2, 14),
+                datetime(2014, 7, 1, 15, 30, 15): datetime(2014, 7, 2, 13, 30, 15),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 14),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 14),
+            },
+        ),
+        (
+            BusinessHour(n=2, start="13:00", end="16:00"),
+            {
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 15),
+                datetime(2014, 7, 2, 14, 30): datetime(2014, 7, 3, 13, 30),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 15),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 15),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 15),
+                datetime(2014, 7, 4, 14, 30): datetime(2014, 7, 7, 13, 30),
+                datetime(2014, 7, 4, 14, 30, 30): datetime(2014, 7, 7, 13, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=-1, start="13:00", end="16:00"),
+            {
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 2, 13): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 2, 15): datetime(2014, 7, 2, 14),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 16): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 13, 30, 15): datetime(2014, 7, 1, 15, 30, 15),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 4, 15),
+                datetime(2014, 7, 7, 11): datetime(2014, 7, 4, 15),
+            },
+        ),
+        (
+            BusinessHour(n=-3, start="10:00", end="16:00"),
+            {
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 13),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 2, 11),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 1, 13),
+                datetime(2014, 7, 2, 13): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 2, 11, 30): datetime(2014, 7, 1, 14, 30),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 13),
+                datetime(2014, 7, 4, 10): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 4, 13),
+                datetime(2014, 7, 4, 16): datetime(2014, 7, 4, 13),
+                datetime(2014, 7, 4, 12, 30): datetime(2014, 7, 3, 15, 30),
+                datetime(2014, 7, 4, 12, 30, 30): datetime(2014, 7, 3, 15, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(start="19:00", end="05:00"),
+            {
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 20),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 2, 20),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 20),
+                datetime(2014, 7, 2, 13): datetime(2014, 7, 2, 20),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 20),
+                datetime(2014, 7, 2, 4, 30): datetime(2014, 7, 2, 19, 30),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 1),
+                datetime(2014, 7, 4, 10): datetime(2014, 7, 4, 20),
+                datetime(2014, 7, 4, 23): datetime(2014, 7, 5, 0),
+                datetime(2014, 7, 5, 0): datetime(2014, 7, 5, 1),
+                datetime(2014, 7, 5, 4): datetime(2014, 7, 7, 19),
+                datetime(2014, 7, 5, 4, 30): datetime(2014, 7, 7, 19, 30),
+                datetime(2014, 7, 5, 4, 30, 30): datetime(2014, 7, 7, 19, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=-1, start="19:00", end="05:00"),
+            {
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 4),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 13): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 20): datetime(2014, 7, 2, 5),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 19, 30): datetime(2014, 7, 2, 4, 30),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 23),
+                datetime(2014, 7, 3, 6): datetime(2014, 7, 3, 4),
+                datetime(2014, 7, 4, 23): datetime(2014, 7, 4, 22),
+                datetime(2014, 7, 5, 0): datetime(2014, 7, 4, 23),
+                datetime(2014, 7, 5, 4): datetime(2014, 7, 5, 3),
+                datetime(2014, 7, 7, 19, 30): datetime(2014, 7, 5, 4, 30),
+                datetime(2014, 7, 7, 19, 30, 30): datetime(2014, 7, 5, 4, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=4, start="00:00", end="23:00"),
+            {
+                datetime(2014, 7, 3, 22): datetime(2014, 7, 4, 3),
+                datetime(2014, 7, 4, 22): datetime(2014, 7, 7, 3),
+                datetime(2014, 7, 3, 22, 30): datetime(2014, 7, 4, 3, 30),
+                datetime(2014, 7, 3, 22, 20): datetime(2014, 7, 4, 3, 20),
+                datetime(2014, 7, 4, 22, 30, 30): datetime(2014, 7, 7, 3, 30, 30),
+                datetime(2014, 7, 4, 22, 30, 20): datetime(2014, 7, 7, 3, 30, 20),
+            },
+        ),
+        (
+            BusinessHour(n=-4, start="00:00", end="23:00"),
+            {
+                datetime(2014, 7, 4, 3): datetime(2014, 7, 3, 22),
+                datetime(2014, 7, 7, 3): datetime(2014, 7, 4, 22),
+                datetime(2014, 7, 4, 3, 30): datetime(2014, 7, 3, 22, 30),
+                datetime(2014, 7, 4, 3, 20): datetime(2014, 7, 3, 22, 20),
+                datetime(2014, 7, 7, 3, 30, 30): datetime(2014, 7, 4, 22, 30, 30),
+                datetime(2014, 7, 7, 3, 30, 20): datetime(2014, 7, 4, 22, 30, 20),
+            },
+        ),
+        (
+            BusinessHour(start=["09:00", "14:00"], end=["12:00", "18:00"]),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 1, 19): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1, 17),
+                datetime(2014, 7, 1, 16, 30, 15): datetime(2014, 7, 1, 17, 30, 15),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2, 9),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 2, 14),
+                # out of business hours
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 10),
+                # saturday
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 10),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 9),
+                datetime(2014, 7, 4, 17, 30): datetime(2014, 7, 7, 9, 30),
+                datetime(2014, 7, 4, 17, 30, 30): datetime(2014, 7, 7, 9, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=4, start=["09:00", "14:00"], end=["12:00", "18:00"]),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 17),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 2, 9),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 2, 10),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 2, 11),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 2, 14),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 2, 17),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 15),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 15),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 15),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 15),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 15),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 14),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7, 11, 30),
+                datetime(2014, 7, 4, 16, 30, 30): datetime(2014, 7, 7, 11, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=-4, start=["09:00", "14:00"], end=["12:00", "18:00"]),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 6, 30, 16),
+                datetime(2014, 7, 1, 13): datetime(2014, 6, 30, 17),
+                datetime(2014, 7, 1, 15): datetime(2014, 6, 30, 18),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1, 10),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 11),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 1, 12),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 12),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 2, 12),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 12),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 4, 12),
+                datetime(2014, 7, 4, 18): datetime(2014, 7, 4, 12),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 4, 14, 30),
+                datetime(2014, 7, 7, 9, 30, 30): datetime(2014, 7, 4, 14, 30, 30),
+            },
+        ),
+        (
+            BusinessHour(n=-1, start=["19:00", "03:00"], end=["01:00", "05:00"]),
+            {
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1, 4),
+                datetime(2014, 7, 2, 14): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 13): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 20): datetime(2014, 7, 2, 5),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 2, 4),
+                datetime(2014, 7, 2, 4): datetime(2014, 7, 2, 1),
+                datetime(2014, 7, 2, 19, 30): datetime(2014, 7, 2, 4, 30),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 2, 23),
+                datetime(2014, 7, 3, 6): datetime(2014, 7, 3, 4),
+                datetime(2014, 7, 4, 23): datetime(2014, 7, 4, 22),
+                datetime(2014, 7, 5, 0): datetime(2014, 7, 4, 23),
+                datetime(2014, 7, 5, 4): datetime(2014, 7, 5, 0),
+                datetime(2014, 7, 7, 3, 30): datetime(2014, 7, 5, 0, 30),
+                datetime(2014, 7, 7, 19, 30): datetime(2014, 7, 7, 4, 30),
+                datetime(2014, 7, 7, 19, 30, 30): datetime(2014, 7, 7, 4, 30, 30),
+            },
+        ),
+    ]
+
+    # long business hours (see gh-26381)
+
+    # multiple business hours
+
+    @pytest.mark.parametrize("case", apply_cases)
+    def test_apply(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    apply_large_n_cases = [
+        (
+            # A week later
+            BusinessHour(40),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 8, 11),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 8, 13),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 8, 15),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 8, 16),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 9, 9),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 9, 11),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 9, 9),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 10, 9),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 10, 9),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 10, 9),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 14, 9),
+                datetime(2014, 7, 4, 18): datetime(2014, 7, 14, 9),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 14, 9, 30),
+                datetime(2014, 7, 7, 9, 30, 30): datetime(2014, 7, 14, 9, 30, 30),
+            },
+        ),
+        (
+            # 3 days and 1 hour before
+            BusinessHour(-25),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 6, 26, 10),
+                datetime(2014, 7, 1, 13): datetime(2014, 6, 26, 12),
+                datetime(2014, 7, 1, 9): datetime(2014, 6, 25, 16),
+                datetime(2014, 7, 1, 10): datetime(2014, 6, 25, 17),
+                datetime(2014, 7, 3, 11): datetime(2014, 6, 30, 10),
+                datetime(2014, 7, 3, 8): datetime(2014, 6, 27, 16),
+                datetime(2014, 7, 3, 19): datetime(2014, 6, 30, 16),
+                datetime(2014, 7, 3, 23): datetime(2014, 6, 30, 16),
+                datetime(2014, 7, 4, 9): datetime(2014, 6, 30, 16),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 6, 18): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 1, 16, 30),
+                datetime(2014, 7, 7, 10, 30, 30): datetime(2014, 7, 2, 9, 30, 30),
+            },
+        ),
+        (
+            # 5 days and 3 hours later
+            BusinessHour(28, start="21:00", end="02:00"),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 9, 0),
+                datetime(2014, 7, 1, 22): datetime(2014, 7, 9, 1),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 9, 21),
+                datetime(2014, 7, 2, 2): datetime(2014, 7, 10, 0),
+                datetime(2014, 7, 3, 21): datetime(2014, 7, 11, 0),
+                datetime(2014, 7, 4, 1): datetime(2014, 7, 11, 23),
+                datetime(2014, 7, 4, 2): datetime(2014, 7, 12, 0),
+                datetime(2014, 7, 4, 3): datetime(2014, 7, 12, 0),
+                datetime(2014, 7, 5, 1): datetime(2014, 7, 14, 23),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 15, 0),
+                datetime(2014, 7, 6, 18): datetime(2014, 7, 15, 0),
+                datetime(2014, 7, 7, 1): datetime(2014, 7, 15, 0),
+                datetime(2014, 7, 7, 23, 30): datetime(2014, 7, 15, 21, 30),
+            },
+        ),
+        (
+            # large n for multiple opening hours (3 days and 1 hour before)
+            BusinessHour(n=-25, start=["09:00", "14:00"], end=["12:00", "19:00"]),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 6, 26, 10),
+                datetime(2014, 7, 1, 13): datetime(2014, 6, 26, 11),
+                datetime(2014, 7, 1, 9): datetime(2014, 6, 25, 18),
+                datetime(2014, 7, 1, 10): datetime(2014, 6, 25, 19),
+                datetime(2014, 7, 3, 11): datetime(2014, 6, 30, 10),
+                datetime(2014, 7, 3, 8): datetime(2014, 6, 27, 18),
+                datetime(2014, 7, 3, 19): datetime(2014, 6, 30, 18),
+                datetime(2014, 7, 3, 23): datetime(2014, 6, 30, 18),
+                datetime(2014, 7, 4, 9): datetime(2014, 6, 30, 18),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 1, 18),
+                datetime(2014, 7, 6, 18): datetime(2014, 7, 1, 18),
+                datetime(2014, 7, 7, 9, 30): datetime(2014, 7, 1, 18, 30),
+                datetime(2014, 7, 7, 10, 30, 30): datetime(2014, 7, 2, 9, 30, 30),
+            },
+        ),
+        (
+            # 5 days and 3 hours later
+            BusinessHour(28, start=["21:00", "03:00"], end=["01:00", "04:00"]),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 9, 0),
+                datetime(2014, 7, 1, 22): datetime(2014, 7, 9, 3),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 9, 21),
+                datetime(2014, 7, 2, 2): datetime(2014, 7, 9, 23),
+                datetime(2014, 7, 3, 21): datetime(2014, 7, 11, 0),
+                datetime(2014, 7, 4, 1): datetime(2014, 7, 11, 23),
+                datetime(2014, 7, 4, 2): datetime(2014, 7, 11, 23),
+                datetime(2014, 7, 4, 3): datetime(2014, 7, 11, 23),
+                datetime(2014, 7, 4, 21): datetime(2014, 7, 12, 0),
+                datetime(2014, 7, 5, 0): datetime(2014, 7, 14, 22),
+                datetime(2014, 7, 5, 1): datetime(2014, 7, 14, 23),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 14, 23),
+                datetime(2014, 7, 6, 18): datetime(2014, 7, 14, 23),
+                datetime(2014, 7, 7, 1): datetime(2014, 7, 14, 23),
+                datetime(2014, 7, 7, 23, 30): datetime(2014, 7, 15, 21, 30),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", apply_large_n_cases)
+    def test_apply_large_n(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_apply_nanoseconds(self):
+        tests = [
+            (
+                BusinessHour(),
+                {
+                    Timestamp("2014-07-04 15:00") + Nano(5): Timestamp(
+                        "2014-07-04 16:00"
+                    )
+                    + Nano(5),
+                    Timestamp("2014-07-04 16:00") + Nano(5): Timestamp(
+                        "2014-07-07 09:00"
+                    )
+                    + Nano(5),
+                    Timestamp("2014-07-04 16:00") - Nano(5): Timestamp(
+                        "2014-07-04 17:00"
+                    )
+                    - Nano(5),
+                },
+            ),
+            (
+                BusinessHour(-1),
+                {
+                    Timestamp("2014-07-04 15:00") + Nano(5): Timestamp(
+                        "2014-07-04 14:00"
+                    )
+                    + Nano(5),
+                    Timestamp("2014-07-04 10:00") + Nano(5): Timestamp(
+                        "2014-07-04 09:00"
+                    )
+                    + Nano(5),
+                    Timestamp("2014-07-04 10:00") - Nano(5): Timestamp(
+                        "2014-07-03 17:00"
+                    )
+                    - Nano(5),
+                },
+            ),
+        ]
+
+        for offset, cases in tests:
+            for base, expected in cases.items():
+                assert_offset_equal(offset, base, expected)
+
+    @pytest.mark.parametrize("td_unit", ["s", "ms", "us", "ns"])
+    def test_bday_ignores_timedeltas(self, unit, td_unit):
+        # GH#55608
+        idx = date_range("2010/02/01", "2010/02/10", freq="12h", unit=unit)
+        td = Timedelta(3, unit="h").as_unit(td_unit)
+        off = BDay(offset=td)
+        t1 = idx + off
+
+        exp_unit = tm.get_finest_unit(td.unit, idx.unit)
+
+        expected = DatetimeIndex(
+            [
+                "2010-02-02 03:00:00",
+                "2010-02-02 15:00:00",
+                "2010-02-03 03:00:00",
+                "2010-02-03 15:00:00",
+                "2010-02-04 03:00:00",
+                "2010-02-04 15:00:00",
+                "2010-02-05 03:00:00",
+                "2010-02-05 15:00:00",
+                "2010-02-08 03:00:00",
+                "2010-02-08 15:00:00",
+                "2010-02-08 03:00:00",
+                "2010-02-08 15:00:00",
+                "2010-02-08 03:00:00",
+                "2010-02-08 15:00:00",
+                "2010-02-09 03:00:00",
+                "2010-02-09 15:00:00",
+                "2010-02-10 03:00:00",
+                "2010-02-10 15:00:00",
+                "2010-02-11 03:00:00",
+            ],
+            freq=None,
+        ).as_unit(exp_unit)
+        tm.assert_index_equal(t1, expected)
+
+        # TODO(GH#55564): as_unit will be unnecessary
+        pointwise = DatetimeIndex([x + off for x in idx]).as_unit(exp_unit)
+        tm.assert_index_equal(pointwise, expected)
+
+    def test_add_bday_offset_nanos(self):
+        # GH#55608
+        idx = date_range("2010/02/01", "2010/02/10", freq="12h", unit="ns")
+        off = BDay(offset=Timedelta(3, unit="ns"))
+
+        result = idx + off
+        expected = DatetimeIndex([x + off for x in idx])
+        tm.assert_index_equal(result, expected)
+
+
+class TestOpeningTimes:
+    # opening time should be affected by sign of n, not by n's value and end
+    opening_time_cases = [
+        (
+            [
+                BusinessHour(),
+                BusinessHour(n=2),
+                BusinessHour(n=4),
+                BusinessHour(end="10:00"),
+                BusinessHour(n=2, end="4:00"),
+                BusinessHour(n=4, end="15:00"),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 1, 9),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 1, 9),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 1, 9),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 1, 9),
+                ),
+                # if timestamp is on opening time, next opening time is
+                # as it is
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 2, 10): (
+                    datetime(2014, 7, 3, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                # 2014-07-05 is saturday
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 4, 9),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 4, 9),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 4, 9),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 4, 9),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 4, 9),
+                ),
+                datetime(2014, 7, 7, 9, 1): (
+                    datetime(2014, 7, 8, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(start="11:15"),
+                BusinessHour(n=2, start="11:15"),
+                BusinessHour(n=3, start="11:15"),
+                BusinessHour(start="11:15", end="10:00"),
+                BusinessHour(n=2, start="11:15", end="4:00"),
+                BusinessHour(n=3, start="11:15", end="15:00"),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 1, 11, 15),
+                    datetime(2014, 6, 30, 11, 15),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 11, 15),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 11, 15),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 11, 15),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 11, 15),
+                ),
+                datetime(2014, 7, 2, 10): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 11, 15),
+                ),
+                datetime(2014, 7, 2, 11, 15): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 2, 11, 15),
+                ),
+                datetime(2014, 7, 2, 11, 15, 1): (
+                    datetime(2014, 7, 3, 11, 15),
+                    datetime(2014, 7, 2, 11, 15),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 11, 15),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 4, 11, 15),
+                    datetime(2014, 7, 3, 11, 15),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 11, 15),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 11, 15),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 11, 15),
+                ),
+                datetime(2014, 7, 7, 9, 1): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 11, 15),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(-1),
+                BusinessHour(n=-2),
+                BusinessHour(n=-4),
+                BusinessHour(n=-1, end="10:00"),
+                BusinessHour(n=-2, end="4:00"),
+                BusinessHour(n=-4, end="15:00"),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 1, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 1, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 1, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 1, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 2, 9),
+                ),
+                datetime(2014, 7, 2, 10): (
+                    datetime(2014, 7, 2, 9),
+                    datetime(2014, 7, 3, 9),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 4, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 4, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 4, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 4, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 4, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 7, 9): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 7, 9),
+                ),
+                datetime(2014, 7, 7, 9, 1): (
+                    datetime(2014, 7, 7, 9),
+                    datetime(2014, 7, 8, 9),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(start="17:00", end="05:00"),
+                BusinessHour(n=3, start="17:00", end="03:00"),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 6, 30, 17),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 2, 17),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 2, 17),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 2, 17),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 17),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 4, 17): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 3, 17),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 7, 17, 1): (
+                    datetime(2014, 7, 8, 17),
+                    datetime(2014, 7, 7, 17),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(-1, start="17:00", end="05:00"),
+                BusinessHour(n=-2, start="17:00", end="03:00"),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 6, 30, 17),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 2, 16, 59): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 17),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 3, 17),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 17),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 17),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 17),
+                ),
+                datetime(2014, 7, 7, 18): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 8, 17),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(start=["11:15", "15:00"], end=["13:00", "20:00"]),
+                BusinessHour(n=3, start=["11:15", "15:00"], end=["12:00", "20:00"]),
+                BusinessHour(start=["11:15", "15:00"], end=["13:00", "17:00"]),
+                BusinessHour(n=2, start=["11:15", "15:00"], end=["12:00", "03:00"]),
+                BusinessHour(n=3, start=["11:15", "15:00"], end=["13:00", "16:00"]),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 1, 11, 15),
+                    datetime(2014, 6, 30, 15),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 15),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 15),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 15),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 15),
+                ),
+                datetime(2014, 7, 2, 10): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 1, 15),
+                ),
+                datetime(2014, 7, 2, 11, 15): (
+                    datetime(2014, 7, 2, 11, 15),
+                    datetime(2014, 7, 2, 11, 15),
+                ),
+                datetime(2014, 7, 2, 11, 15, 1): (
+                    datetime(2014, 7, 2, 15),
+                    datetime(2014, 7, 2, 11, 15),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 15),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 4, 11, 15),
+                    datetime(2014, 7, 3, 15),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 15),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 15),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 15),
+                ),
+                datetime(2014, 7, 7, 9, 1): (
+                    datetime(2014, 7, 7, 11, 15),
+                    datetime(2014, 7, 4, 15),
+                ),
+                datetime(2014, 7, 7, 12): (
+                    datetime(2014, 7, 7, 15),
+                    datetime(2014, 7, 7, 11, 15),
+                ),
+            },
+        ),
+        (
+            [
+                BusinessHour(n=-1, start=["17:00", "08:00"], end=["05:00", "10:00"]),
+                BusinessHour(n=-2, start=["08:00", "17:00"], end=["10:00", "03:00"]),
+            ],
+            {
+                datetime(2014, 7, 1, 11): (
+                    datetime(2014, 7, 1, 8),
+                    datetime(2014, 7, 1, 17),
+                ),
+                datetime(2014, 7, 1, 18): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 8),
+                ),
+                datetime(2014, 7, 1, 23): (
+                    datetime(2014, 7, 1, 17),
+                    datetime(2014, 7, 2, 8),
+                ),
+                datetime(2014, 7, 2, 8): (
+                    datetime(2014, 7, 2, 8),
+                    datetime(2014, 7, 2, 8),
+                ),
+                datetime(2014, 7, 2, 9): (
+                    datetime(2014, 7, 2, 8),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 2, 16, 59): (
+                    datetime(2014, 7, 2, 8),
+                    datetime(2014, 7, 2, 17),
+                ),
+                datetime(2014, 7, 5, 10): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 8),
+                ),
+                datetime(2014, 7, 4, 10): (
+                    datetime(2014, 7, 4, 8),
+                    datetime(2014, 7, 4, 17),
+                ),
+                datetime(2014, 7, 4, 23): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 8),
+                ),
+                datetime(2014, 7, 6, 10): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 8),
+                ),
+                datetime(2014, 7, 7, 5): (
+                    datetime(2014, 7, 4, 17),
+                    datetime(2014, 7, 7, 8),
+                ),
+                datetime(2014, 7, 7, 18): (
+                    datetime(2014, 7, 7, 17),
+                    datetime(2014, 7, 8, 8),
+                ),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", opening_time_cases)
+    def test_opening_time(self, case):
+        _offsets, cases = case
+        for offset in _offsets:
+            for dt, (exp_next, exp_prev) in cases.items():
+                assert offset._next_opening_time(dt) == exp_next
+                assert offset._prev_opening_time(dt) == exp_prev
diff --git a/pandas/pandas/tests/tseries/offsets/test_business_month.py b/pandas/pandas/tests/tseries/offsets/test_business_month.py
new file mode 100644
index 0000000000000..3ae2a115d46f7
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_business_month.py
@@ -0,0 +1,218 @@
+"""
+Tests for the following offsets:
+- BMonthBegin
+- BMonthEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+import pandas as pd
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    BMonthBegin,
+    BMonthEnd,
+)
+
+
+@pytest.mark.parametrize("n", [-2, 1])
+@pytest.mark.parametrize(
+    "cls",
+    [
+        BMonthBegin,
+        BMonthEnd,
+    ],
+)
+def test_apply_index(cls, n):
+    offset = cls(n=n)
+    rng = pd.date_range(start="1/1/2000", periods=100000, freq="min")
+    ser = pd.Series(rng)
+
+    res = rng + offset
+    assert res.freq is None  # not retained
+    assert res[0] == rng[0] + offset
+    assert res[-1] == rng[-1] + offset
+    res2 = ser + offset
+    # apply_index is only for indexes, not series, so no res2_v2
+    assert res2.iloc[0] == ser.iloc[0] + offset
+    assert res2.iloc[-1] == ser.iloc[-1] + offset
+
+
+class TestBMonthBegin:
+    def test_offsets_compare_equal(self):
+        # root cause of #456
+        offset1 = BMonthBegin()
+        offset2 = BMonthBegin()
+        assert not offset1 != offset2
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2006, 9, 1): datetime(2006, 10, 2),
+                datetime(2007, 1, 1): datetime(2007, 2, 1),
+                datetime(2006, 12, 1): datetime(2007, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthBegin(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2006, 10, 2): datetime(2006, 10, 2),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2006, 9, 15): datetime(2006, 10, 2),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthBegin(2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 3, 3),
+                datetime(2008, 1, 15): datetime(2008, 3, 3),
+                datetime(2006, 12, 29): datetime(2007, 2, 1),
+                datetime(2006, 12, 31): datetime(2007, 2, 1),
+                datetime(2007, 1, 1): datetime(2007, 3, 1),
+                datetime(2006, 11, 1): datetime(2007, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthBegin(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 1),
+                datetime(2008, 6, 30): datetime(2008, 6, 2),
+                datetime(2008, 6, 1): datetime(2008, 5, 1),
+                datetime(2008, 3, 10): datetime(2008, 3, 3),
+                datetime(2008, 12, 31): datetime(2008, 12, 1),
+                datetime(2006, 12, 29): datetime(2006, 12, 1),
+                datetime(2006, 12, 30): datetime(2006, 12, 1),
+                datetime(2007, 1, 1): datetime(2006, 12, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (BMonthBegin(), datetime(2007, 12, 31), False),
+        (BMonthBegin(), datetime(2008, 1, 1), True),
+        (BMonthBegin(), datetime(2001, 4, 2), True),
+        (BMonthBegin(), datetime(2008, 3, 3), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+
+class TestBMonthEnd:
+    def test_normalize(self):
+        dt = datetime(2007, 1, 1, 3)
+
+        result = dt + BMonthEnd(normalize=True)
+        expected = dt.replace(hour=0) + BMonthEnd()
+        assert result == expected
+
+    def test_offsets_compare_equal(self):
+        # root cause of #456
+        offset1 = BMonthEnd()
+        offset2 = BMonthEnd()
+        assert not offset1 != offset2
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 2, 29),
+                datetime(2006, 12, 29): datetime(2007, 1, 31),
+                datetime(2006, 12, 31): datetime(2007, 1, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 31),
+                datetime(2006, 12, 1): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2006, 12, 29): datetime(2006, 12, 29),
+                datetime(2006, 12, 31): datetime(2007, 1, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthEnd(2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 29),
+                datetime(2008, 1, 31): datetime(2008, 3, 31),
+                datetime(2006, 12, 29): datetime(2007, 2, 28),
+                datetime(2006, 12, 31): datetime(2007, 2, 28),
+                datetime(2007, 1, 1): datetime(2007, 2, 28),
+                datetime(2006, 11, 1): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BMonthEnd(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 29),
+                datetime(2008, 6, 30): datetime(2008, 5, 30),
+                datetime(2008, 12, 31): datetime(2008, 11, 28),
+                datetime(2006, 12, 29): datetime(2006, 11, 30),
+                datetime(2006, 12, 30): datetime(2006, 12, 29),
+                datetime(2007, 1, 1): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (BMonthEnd(), datetime(2007, 12, 31), True),
+        (BMonthEnd(), datetime(2008, 1, 1), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_business_quarter.py b/pandas/pandas/tests/tseries/offsets/test_business_quarter.py
new file mode 100644
index 0000000000000..ab3e55c4989fb
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_business_quarter.py
@@ -0,0 +1,299 @@
+"""
+Tests for the following offsets:
+- BQuarterBegin
+- BQuarterEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    BQuarterBegin,
+    BQuarterEnd,
+)
+
+
+def test_quarterly_dont_normalize():
+    date = datetime(2012, 3, 31, 5, 30)
+
+    offsets = (BQuarterEnd, BQuarterBegin)
+
+    for klass in offsets:
+        result = date + klass()
+        assert result.time() == date.time()
+
+
+@pytest.mark.parametrize("offset", [BQuarterBegin(), BQuarterEnd()])
+def test_on_offset(offset):
+    dates = [
+        datetime(2016, m, d)
+        for m in [10, 11, 12]
+        for d in [1, 2, 3, 28, 29, 30, 31]
+        if not (m == 11 and d == 31)
+    ]
+    for date in dates:
+        res = offset.is_on_offset(date)
+        slow_version = date == (date + offset) - offset
+        assert res == slow_version
+
+
+class TestBQuarterBegin:
+    def test_repr(self):
+        expected = "<BusinessQuarterBegin: startingMonth=3>"
+        assert repr(BQuarterBegin()) == expected
+        expected = "<BusinessQuarterBegin: startingMonth=3>"
+        assert repr(BQuarterBegin(startingMonth=3)) == expected
+        expected = "<BusinessQuarterBegin: startingMonth=1>"
+        assert repr(BQuarterBegin(startingMonth=1)) == expected
+
+    def test_offset_corner_case(self):
+        # corner
+        offset = BQuarterBegin(n=-1, startingMonth=1)
+        assert datetime(2007, 4, 3) + offset == datetime(2007, 4, 2)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BQuarterBegin(startingMonth=1),
+            {
+                datetime(2008, 1, 1): datetime(2008, 4, 1),
+                datetime(2008, 1, 31): datetime(2008, 4, 1),
+                datetime(2008, 2, 15): datetime(2008, 4, 1),
+                datetime(2008, 2, 29): datetime(2008, 4, 1),
+                datetime(2008, 3, 15): datetime(2008, 4, 1),
+                datetime(2008, 3, 31): datetime(2008, 4, 1),
+                datetime(2008, 4, 15): datetime(2008, 7, 1),
+                datetime(2007, 3, 15): datetime(2007, 4, 2),
+                datetime(2007, 2, 28): datetime(2007, 4, 2),
+                datetime(2007, 1, 1): datetime(2007, 4, 2),
+                datetime(2007, 4, 15): datetime(2007, 7, 2),
+                datetime(2007, 7, 1): datetime(2007, 7, 2),
+                datetime(2007, 4, 1): datetime(2007, 4, 2),
+                datetime(2007, 4, 2): datetime(2007, 7, 2),
+                datetime(2008, 4, 30): datetime(2008, 7, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterBegin(startingMonth=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2008, 1, 15): datetime(2008, 2, 1),
+                datetime(2008, 2, 29): datetime(2008, 5, 1),
+                datetime(2008, 3, 15): datetime(2008, 5, 1),
+                datetime(2008, 3, 31): datetime(2008, 5, 1),
+                datetime(2008, 4, 15): datetime(2008, 5, 1),
+                datetime(2008, 8, 15): datetime(2008, 11, 3),
+                datetime(2008, 9, 15): datetime(2008, 11, 3),
+                datetime(2008, 11, 1): datetime(2008, 11, 3),
+                datetime(2008, 4, 30): datetime(2008, 5, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterBegin(startingMonth=1, n=0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2007, 12, 31): datetime(2008, 1, 1),
+                datetime(2008, 2, 15): datetime(2008, 4, 1),
+                datetime(2008, 2, 29): datetime(2008, 4, 1),
+                datetime(2008, 1, 15): datetime(2008, 4, 1),
+                datetime(2008, 2, 27): datetime(2008, 4, 1),
+                datetime(2008, 3, 15): datetime(2008, 4, 1),
+                datetime(2007, 4, 1): datetime(2007, 4, 2),
+                datetime(2007, 4, 2): datetime(2007, 4, 2),
+                datetime(2007, 7, 1): datetime(2007, 7, 2),
+                datetime(2007, 4, 15): datetime(2007, 7, 2),
+                datetime(2007, 7, 2): datetime(2007, 7, 2),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterBegin(startingMonth=1, n=-1),
+            {
+                datetime(2008, 1, 1): datetime(2007, 10, 1),
+                datetime(2008, 1, 31): datetime(2008, 1, 1),
+                datetime(2008, 2, 15): datetime(2008, 1, 1),
+                datetime(2008, 2, 29): datetime(2008, 1, 1),
+                datetime(2008, 3, 15): datetime(2008, 1, 1),
+                datetime(2008, 3, 31): datetime(2008, 1, 1),
+                datetime(2008, 4, 15): datetime(2008, 4, 1),
+                datetime(2007, 7, 3): datetime(2007, 7, 2),
+                datetime(2007, 4, 3): datetime(2007, 4, 2),
+                datetime(2007, 7, 2): datetime(2007, 4, 2),
+                datetime(2008, 4, 1): datetime(2008, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterBegin(startingMonth=1, n=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 7, 1),
+                datetime(2008, 1, 15): datetime(2008, 7, 1),
+                datetime(2008, 2, 29): datetime(2008, 7, 1),
+                datetime(2008, 3, 15): datetime(2008, 7, 1),
+                datetime(2007, 3, 31): datetime(2007, 7, 2),
+                datetime(2007, 4, 15): datetime(2007, 10, 1),
+                datetime(2008, 4, 30): datetime(2008, 10, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+
+class TestBQuarterEnd:
+    def test_repr(self):
+        expected = "<BusinessQuarterEnd: startingMonth=3>"
+        assert repr(BQuarterEnd()) == expected
+        expected = "<BusinessQuarterEnd: startingMonth=3>"
+        assert repr(BQuarterEnd(startingMonth=3)) == expected
+        expected = "<BusinessQuarterEnd: startingMonth=1>"
+        assert repr(BQuarterEnd(startingMonth=1)) == expected
+
+    def test_offset_corner_case(self):
+        # corner
+        offset = BQuarterEnd(n=-1, startingMonth=1)
+        assert datetime(2010, 1, 31) + offset == datetime(2010, 1, 29)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BQuarterEnd(startingMonth=1),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 4, 30),
+                datetime(2008, 2, 15): datetime(2008, 4, 30),
+                datetime(2008, 2, 29): datetime(2008, 4, 30),
+                datetime(2008, 3, 15): datetime(2008, 4, 30),
+                datetime(2008, 3, 31): datetime(2008, 4, 30),
+                datetime(2008, 4, 15): datetime(2008, 4, 30),
+                datetime(2008, 4, 30): datetime(2008, 7, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterEnd(startingMonth=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 29),
+                datetime(2008, 1, 31): datetime(2008, 2, 29),
+                datetime(2008, 2, 15): datetime(2008, 2, 29),
+                datetime(2008, 2, 29): datetime(2008, 5, 30),
+                datetime(2008, 3, 15): datetime(2008, 5, 30),
+                datetime(2008, 3, 31): datetime(2008, 5, 30),
+                datetime(2008, 4, 15): datetime(2008, 5, 30),
+                datetime(2008, 4, 30): datetime(2008, 5, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterEnd(startingMonth=1, n=0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2008, 2, 15): datetime(2008, 4, 30),
+                datetime(2008, 2, 29): datetime(2008, 4, 30),
+                datetime(2008, 3, 15): datetime(2008, 4, 30),
+                datetime(2008, 3, 31): datetime(2008, 4, 30),
+                datetime(2008, 4, 15): datetime(2008, 4, 30),
+                datetime(2008, 4, 30): datetime(2008, 4, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterEnd(startingMonth=1, n=-1),
+            {
+                datetime(2008, 1, 1): datetime(2007, 10, 31),
+                datetime(2008, 1, 31): datetime(2007, 10, 31),
+                datetime(2008, 2, 15): datetime(2008, 1, 31),
+                datetime(2008, 2, 29): datetime(2008, 1, 31),
+                datetime(2008, 3, 15): datetime(2008, 1, 31),
+                datetime(2008, 3, 31): datetime(2008, 1, 31),
+                datetime(2008, 4, 15): datetime(2008, 1, 31),
+                datetime(2008, 4, 30): datetime(2008, 1, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BQuarterEnd(startingMonth=1, n=2),
+            {
+                datetime(2008, 1, 31): datetime(2008, 7, 31),
+                datetime(2008, 2, 15): datetime(2008, 7, 31),
+                datetime(2008, 2, 29): datetime(2008, 7, 31),
+                datetime(2008, 3, 15): datetime(2008, 7, 31),
+                datetime(2008, 3, 31): datetime(2008, 7, 31),
+                datetime(2008, 4, 15): datetime(2008, 7, 31),
+                datetime(2008, 4, 30): datetime(2008, 10, 31),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (BQuarterEnd(1, startingMonth=1), datetime(2008, 1, 31), True),
+        (BQuarterEnd(1, startingMonth=1), datetime(2007, 12, 31), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2008, 2, 29), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2007, 3, 30), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2007, 3, 31), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2008, 4, 30), True),
+        (BQuarterEnd(1, startingMonth=1), datetime(2008, 5, 30), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2007, 6, 29), False),
+        (BQuarterEnd(1, startingMonth=1), datetime(2007, 6, 30), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2008, 1, 31), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2007, 12, 31), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2008, 2, 29), True),
+        (BQuarterEnd(1, startingMonth=2), datetime(2007, 3, 30), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2007, 3, 31), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2008, 4, 30), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2008, 5, 30), True),
+        (BQuarterEnd(1, startingMonth=2), datetime(2007, 6, 29), False),
+        (BQuarterEnd(1, startingMonth=2), datetime(2007, 6, 30), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2008, 1, 31), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2007, 12, 31), True),
+        (BQuarterEnd(1, startingMonth=3), datetime(2008, 2, 29), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2007, 3, 30), True),
+        (BQuarterEnd(1, startingMonth=3), datetime(2007, 3, 31), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2008, 4, 30), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2008, 5, 30), False),
+        (BQuarterEnd(1, startingMonth=3), datetime(2007, 6, 29), True),
+        (BQuarterEnd(1, startingMonth=3), datetime(2007, 6, 30), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_business_year.py b/pandas/pandas/tests/tseries/offsets/test_business_year.py
new file mode 100644
index 0000000000000..cf12b166b30e4
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_business_year.py
@@ -0,0 +1,216 @@
+"""
+Tests for the following offsets:
+- BYearBegin
+- BYearEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    BYearBegin,
+    BYearEnd,
+)
+
+
+class TestBYearBegin:
+    def test_misspecified(self):
+        msg = "Month must go from 1 to 12"
+        with pytest.raises(ValueError, match=msg):
+            BYearBegin(month=13)
+        with pytest.raises(ValueError, match=msg):
+            BYearEnd(month=13)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BYearBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2009, 1, 1),
+                datetime(2008, 6, 30): datetime(2009, 1, 1),
+                datetime(2008, 12, 31): datetime(2009, 1, 1),
+                datetime(2011, 1, 1): datetime(2011, 1, 3),
+                datetime(2011, 1, 3): datetime(2012, 1, 2),
+                datetime(2005, 12, 30): datetime(2006, 1, 2),
+                datetime(2005, 12, 31): datetime(2006, 1, 2),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearBegin(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 6, 30): datetime(2009, 1, 1),
+                datetime(2008, 12, 31): datetime(2009, 1, 1),
+                datetime(2005, 12, 30): datetime(2006, 1, 2),
+                datetime(2005, 12, 31): datetime(2006, 1, 2),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearBegin(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 1, 2),
+                datetime(2009, 1, 4): datetime(2009, 1, 1),
+                datetime(2009, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 6, 30): datetime(2008, 1, 1),
+                datetime(2008, 12, 31): datetime(2008, 1, 1),
+                datetime(2006, 12, 29): datetime(2006, 1, 2),
+                datetime(2006, 12, 30): datetime(2006, 1, 2),
+                datetime(2006, 1, 1): datetime(2005, 1, 3),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearBegin(-2),
+            {
+                datetime(2007, 1, 1): datetime(2005, 1, 3),
+                datetime(2007, 6, 30): datetime(2006, 1, 2),
+                datetime(2008, 12, 31): datetime(2007, 1, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+
+class TestBYearEnd:
+    offset_cases = []
+    offset_cases.append(
+        (
+            BYearEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 12, 31),
+                datetime(2008, 12, 31): datetime(2009, 12, 31),
+                datetime(2005, 12, 30): datetime(2006, 12, 29),
+                datetime(2005, 12, 31): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 12, 31),
+                datetime(2008, 12, 31): datetime(2008, 12, 31),
+                datetime(2005, 12, 31): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearEnd(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 29),
+                datetime(2008, 6, 30): datetime(2007, 12, 31),
+                datetime(2008, 12, 31): datetime(2007, 12, 31),
+                datetime(2006, 12, 29): datetime(2005, 12, 30),
+                datetime(2006, 12, 30): datetime(2006, 12, 29),
+                datetime(2007, 1, 1): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearEnd(-2),
+            {
+                datetime(2007, 1, 1): datetime(2005, 12, 30),
+                datetime(2008, 6, 30): datetime(2006, 12, 29),
+                datetime(2008, 12, 31): datetime(2006, 12, 29),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (BYearEnd(), datetime(2007, 12, 31), True),
+        (BYearEnd(), datetime(2008, 1, 1), False),
+        (BYearEnd(), datetime(2006, 12, 31), False),
+        (BYearEnd(), datetime(2006, 12, 29), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+
+class TestBYearEndLagged:
+    def test_bad_month_fail(self):
+        msg = "Month must go from 1 to 12"
+        with pytest.raises(ValueError, match=msg):
+            BYearEnd(month=13)
+        with pytest.raises(ValueError, match=msg):
+            BYearEnd(month=0)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            BYearEnd(month=6),
+            {
+                datetime(2008, 1, 1): datetime(2008, 6, 30),
+                datetime(2007, 6, 30): datetime(2008, 6, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            BYearEnd(n=-1, month=6),
+            {
+                datetime(2008, 1, 1): datetime(2007, 6, 29),
+                datetime(2007, 6, 30): datetime(2007, 6, 29),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_roll(self):
+        offset = BYearEnd(month=6)
+        date = datetime(2009, 11, 30)
+
+        assert offset.rollforward(date) == datetime(2010, 6, 30)
+        assert offset.rollback(date) == datetime(2009, 6, 30)
+
+    on_offset_cases = [
+        (BYearEnd(month=2), datetime(2007, 2, 28), True),
+        (BYearEnd(month=6), datetime(2007, 6, 30), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_common.py b/pandas/pandas/tests/tseries/offsets/test_common.py
new file mode 100644
index 0000000000000..34181f28bb1a0
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_common.py
@@ -0,0 +1,265 @@
+from datetime import datetime
+
+from dateutil.tz.tz import tzlocal
+import pytest
+
+from pandas._libs.tslibs import (
+    OutOfBoundsDatetime,
+    Timestamp,
+)
+from pandas.compat import (
+    IS64,
+    WASM,
+    is_platform_windows,
+)
+
+from pandas.tseries.offsets import (
+    FY5253,
+    BDay,
+    BMonthBegin,
+    BMonthEnd,
+    BQuarterBegin,
+    BQuarterEnd,
+    BusinessHour,
+    BYearBegin,
+    BYearEnd,
+    CBMonthBegin,
+    CBMonthEnd,
+    CDay,
+    CustomBusinessHour,
+    DateOffset,
+    FY5253Quarter,
+    LastWeekOfMonth,
+    MonthBegin,
+    MonthEnd,
+    QuarterEnd,
+    SemiMonthBegin,
+    SemiMonthEnd,
+    Week,
+    WeekOfMonth,
+    YearBegin,
+    YearEnd,
+)
+
+
+def _get_offset(klass, value=1, normalize=False):
+    # create instance from offset class
+    if klass is FY5253:
+        klass = klass(
+            n=value,
+            startingMonth=1,
+            weekday=1,
+            variation="last",
+            normalize=normalize,
+        )
+    elif klass is FY5253Quarter:
+        klass = klass(
+            n=value,
+            startingMonth=1,
+            weekday=1,
+            qtr_with_extra_week=1,
+            variation="last",
+            normalize=normalize,
+        )
+    elif klass is LastWeekOfMonth:
+        klass = klass(n=value, weekday=5, normalize=normalize)
+    elif klass is WeekOfMonth:
+        klass = klass(n=value, week=1, weekday=5, normalize=normalize)
+    elif klass is Week:
+        klass = klass(n=value, weekday=5, normalize=normalize)
+    elif klass is DateOffset:
+        klass = klass(days=value, normalize=normalize)
+    else:
+        klass = klass(value, normalize=normalize)
+    return klass
+
+
+@pytest.fixture(
+    params=[
+        BDay,
+        BusinessHour,
+        BMonthEnd,
+        BMonthBegin,
+        BQuarterEnd,
+        BQuarterBegin,
+        BYearEnd,
+        BYearBegin,
+        CDay,
+        CustomBusinessHour,
+        CBMonthEnd,
+        CBMonthBegin,
+        MonthEnd,
+        MonthBegin,
+        SemiMonthBegin,
+        SemiMonthEnd,
+        QuarterEnd,
+        LastWeekOfMonth,
+        WeekOfMonth,
+        Week,
+        YearBegin,
+        YearEnd,
+        FY5253,
+        FY5253Quarter,
+        DateOffset,
+    ]
+)
+def _offset(request):
+    return request.param
+
+
+@pytest.mark.skipif(WASM, reason="OverflowError received on WASM")
+def test_apply_out_of_range(request, tz_naive_fixture, _offset):
+    tz = tz_naive_fixture
+
+    # try to create an out-of-bounds result timestamp; if we can't create
+    # the offset skip
+    try:
+        if _offset in (BusinessHour, CustomBusinessHour):
+            # Using 10000 in BusinessHour fails in tz check because of DST
+            # difference
+            offset = _get_offset(_offset, value=100000)
+        else:
+            offset = _get_offset(_offset, value=10000)
+
+        result = Timestamp("20080101") + offset
+        assert isinstance(result, datetime)
+        assert result.tzinfo is None
+
+        # Check tz is preserved
+        t = Timestamp("20080101", tz=tz)
+        result = t + offset
+        assert isinstance(result, datetime)
+        if tz is not None:
+            assert t.tzinfo is not None
+
+        if (
+            isinstance(tz, tzlocal)
+            and ((not IS64) or WASM)
+            and _offset is not DateOffset
+        ):
+            # If we hit OutOfBoundsDatetime on non-64 bit machines
+            # we'll drop out of the try clause before the next test
+            request.applymarker(
+                pytest.mark.xfail(reason="OverflowError inside tzlocal past 2038")
+            )
+        elif (
+            isinstance(tz, tzlocal)
+            and is_platform_windows()
+            and _offset in (QuarterEnd, BQuarterBegin, BQuarterEnd)
+        ):
+            request.applymarker(
+                pytest.mark.xfail(reason="After GH#49737 t.tzinfo is None on CI")
+            )
+        assert str(t.tzinfo) == str(result.tzinfo)
+
+    except OutOfBoundsDatetime:
+        pass
+    except (ValueError, KeyError):
+        # we are creating an invalid offset
+        # so ignore
+        pass
+
+
+def test_offsets_compare_equal(_offset):
+    # root cause of GH#456: __ne__ was not implemented
+    offset1 = _offset()
+    offset2 = _offset()
+    assert not offset1 != offset2
+    assert offset1 == offset2
+
+
+@pytest.mark.parametrize(
+    "date, offset2",
+    [
+        [Timestamp(2008, 1, 1), BDay(2)],
+        [Timestamp(2014, 7, 1, 10, 00), BusinessHour(n=3)],
+        [
+            Timestamp(2014, 7, 1, 10),
+            CustomBusinessHour(
+                holidays=["2014-06-27", Timestamp(2014, 6, 30), Timestamp("2014-07-02")]
+            ),
+        ],
+        [Timestamp(2008, 1, 2), SemiMonthEnd(2)],
+        [Timestamp(2008, 1, 2), SemiMonthBegin(2)],
+        [Timestamp(2008, 1, 2), Week(2)],
+        [Timestamp(2008, 1, 2), WeekOfMonth(2)],
+        [Timestamp(2008, 1, 2), LastWeekOfMonth(2)],
+    ],
+)
+def test_rsub(date, offset2):
+    assert date - offset2 == (-offset2)._apply(date)
+
+
+@pytest.mark.parametrize(
+    "date, offset2",
+    [
+        [Timestamp(2008, 1, 1), BDay(2)],
+        [Timestamp(2014, 7, 1, 10, 00), BusinessHour(n=3)],
+        [
+            Timestamp(2014, 7, 1, 10),
+            CustomBusinessHour(
+                holidays=["2014-06-27", Timestamp(2014, 6, 30), Timestamp("2014-07-02")]
+            ),
+        ],
+        [Timestamp(2008, 1, 2), SemiMonthEnd(2)],
+        [Timestamp(2008, 1, 2), SemiMonthBegin(2)],
+        [Timestamp(2008, 1, 2), Week(2)],
+        [Timestamp(2008, 1, 2), WeekOfMonth(2)],
+        [Timestamp(2008, 1, 2), LastWeekOfMonth(2)],
+    ],
+)
+def test_radd(date, offset2):
+    assert date + offset2 == offset2 + date
+
+
+@pytest.mark.parametrize(
+    "date, offset_box, offset2",
+    [
+        [Timestamp(2008, 1, 1), BDay, BDay(2)],
+        [Timestamp(2008, 1, 2), SemiMonthEnd, SemiMonthEnd(2)],
+        [Timestamp(2008, 1, 2), SemiMonthBegin, SemiMonthBegin(2)],
+        [Timestamp(2008, 1, 2), Week, Week(2)],
+        [Timestamp(2008, 1, 2), WeekOfMonth, WeekOfMonth(2)],
+        [Timestamp(2008, 1, 2), LastWeekOfMonth, LastWeekOfMonth(2)],
+    ],
+)
+def test_sub(date, offset_box, offset2):
+    off = offset2
+    msg = "Cannot subtract datetime from offset"
+    with pytest.raises(TypeError, match=msg):
+        off - date
+
+    assert 2 * off - off == off
+    assert date - offset2 == date + offset_box(-2)
+    assert date - offset2 == date - (2 * off - off)
+
+
+@pytest.mark.parametrize(
+    "offset_box, offset1",
+    [
+        [BDay, BDay()],
+        [LastWeekOfMonth, LastWeekOfMonth()],
+        [WeekOfMonth, WeekOfMonth()],
+        [Week, Week()],
+        [SemiMonthBegin, SemiMonthBegin()],
+        [SemiMonthEnd, SemiMonthEnd()],
+        [CustomBusinessHour, CustomBusinessHour(weekmask="Tue Wed Thu Fri")],
+        [BusinessHour, BusinessHour()],
+    ],
+)
+def test_Mult1(offset_box, offset1):
+    dt = Timestamp(2008, 1, 2)
+    assert dt + 10 * offset1 == dt + offset_box(10)
+    assert dt + 5 * offset1 == dt + offset_box(5)
+
+
+def test_compare_str(_offset):
+    # GH#23524
+    # comparing to strings that cannot be cast to DateOffsets should
+    #  not raise for __eq__ or __ne__
+    off = _get_offset(_offset)
+
+    assert not off == "infer"
+    assert off != "foo"
+    # Note: inequalities are only implemented for Tick subclasses;
+    #  tests for this are in test_ticks
diff --git a/pandas/pandas/tests/tseries/offsets/test_custom_business_day.py b/pandas/pandas/tests/tseries/offsets/test_custom_business_day.py
new file mode 100644
index 0000000000000..d2f309dd3f33c
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_custom_business_day.py
@@ -0,0 +1,99 @@
+"""
+Tests for offsets.CustomBusinessDay / CDay
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import CDay
+
+from pandas import (
+    _testing as tm,
+    read_pickle,
+)
+from pandas.tests.tseries.offsets.common import assert_offset_equal
+
+from pandas.tseries.holiday import USFederalHolidayCalendar
+
+
+@pytest.fixture
+def offset():
+    return CDay()
+
+
+@pytest.fixture
+def offset2():
+    return CDay(2)
+
+
+class TestCustomBusinessDay:
+    def test_repr(self, offset, offset2):
+        assert repr(offset) == "<CustomBusinessDay>"
+        assert repr(offset2) == "<2 * CustomBusinessDays>"
+
+        expected = "<BusinessDay: offset=datetime.timedelta(days=1)>"
+        assert repr(offset + timedelta(1)) == expected
+
+    def test_holidays(self):
+        # Define a TradingDay offset
+        holidays = ["2012-05-01", datetime(2013, 5, 1), np.datetime64("2014-05-01")]
+        tday = CDay(holidays=holidays)
+        for year in range(2012, 2015):
+            dt = datetime(year, 4, 30)
+            xp = datetime(year, 5, 2)
+            rs = dt + tday
+            assert rs == xp
+
+    def test_weekmask(self):
+        weekmask_saudi = "Sat Sun Mon Tue Wed"  # Thu-Fri Weekend
+        weekmask_uae = "1111001"  # Fri-Sat Weekend
+        weekmask_egypt = [1, 1, 1, 1, 0, 0, 1]  # Fri-Sat Weekend
+        bday_saudi = CDay(weekmask=weekmask_saudi)
+        bday_uae = CDay(weekmask=weekmask_uae)
+        bday_egypt = CDay(weekmask=weekmask_egypt)
+        dt = datetime(2013, 5, 1)
+        xp_saudi = datetime(2013, 5, 4)
+        xp_uae = datetime(2013, 5, 2)
+        xp_egypt = datetime(2013, 5, 2)
+        assert xp_saudi == dt + bday_saudi
+        assert xp_uae == dt + bday_uae
+        assert xp_egypt == dt + bday_egypt
+        xp2 = datetime(2013, 5, 5)
+        assert xp2 == dt + 2 * bday_saudi
+        assert xp2 == dt + 2 * bday_uae
+        assert xp2 == dt + 2 * bday_egypt
+
+    def test_weekmask_and_holidays(self):
+        weekmask_egypt = "Sun Mon Tue Wed Thu"  # Fri-Sat Weekend
+        holidays = ["2012-05-01", datetime(2013, 5, 1), np.datetime64("2014-05-01")]
+        bday_egypt = CDay(holidays=holidays, weekmask=weekmask_egypt)
+        dt = datetime(2013, 4, 30)
+        xp_egypt = datetime(2013, 5, 5)
+        assert xp_egypt == dt + 2 * bday_egypt
+
+    @pytest.mark.filterwarnings("ignore:Non:pandas.errors.PerformanceWarning")
+    def test_calendar(self):
+        calendar = USFederalHolidayCalendar()
+        dt = datetime(2014, 1, 17)
+        assert_offset_equal(CDay(calendar=calendar), dt, datetime(2014, 1, 21))
+
+    def test_roundtrip_pickle(self, offset, offset2):
+        def _check_roundtrip(obj):
+            unpickled = tm.round_trip_pickle(obj)
+            assert unpickled == obj
+
+        _check_roundtrip(offset)
+        _check_roundtrip(offset2)
+        _check_roundtrip(offset * 2)
+
+    def test_pickle_compat_0_14_1(self, datapath):
+        hdays = [datetime(2013, 1, 1) for ele in range(4)]
+        pth = datapath("tseries", "offsets", "data", "cday-0.14.1.pickle")
+        cday0_14_1 = read_pickle(pth)
+        cday = CDay(holidays=hdays)
+        assert cday == cday0_14_1
diff --git a/pandas/pandas/tests/tseries/offsets/test_custom_business_hour.py b/pandas/pandas/tests/tseries/offsets/test_custom_business_hour.py
new file mode 100644
index 0000000000000..360ed70fa5b9e
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_custom_business_hour.py
@@ -0,0 +1,324 @@
+"""
+Tests for offsets.CustomBusinessHour
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    time as dt_time,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas._libs.tslibs.offsets import (
+    BusinessHour,
+    CustomBusinessHour,
+    Nano,
+)
+
+from pandas.tests.tseries.offsets.common import assert_offset_equal
+
+from pandas.tseries.holiday import USFederalHolidayCalendar
+
+holidays = ["2014-06-27", datetime(2014, 6, 30), np.datetime64("2014-07-02")]
+
+
+@pytest.fixture
+def dt():
+    return datetime(2014, 7, 1, 10, 00)
+
+
+@pytest.fixture
+def _offset():
+    return CustomBusinessHour
+
+
+# 2014 Calendar to check custom holidays
+#   Sun Mon Tue Wed Thu Fri Sat
+#  6/22  23  24  25  26  27  28
+#    29  30 7/1   2   3   4   5
+#     6   7   8   9  10  11  12
+@pytest.fixture
+def offset1():
+    return CustomBusinessHour(weekmask="Tue Wed Thu Fri")
+
+
+@pytest.fixture
+def offset2():
+    return CustomBusinessHour(holidays=holidays)
+
+
+class TestCustomBusinessHour:
+    def test_constructor_errors(self):
+        msg = "time data must be specified only with hour and minute"
+        with pytest.raises(ValueError, match=msg):
+            CustomBusinessHour(start=dt_time(11, 0, 5))
+        msg = "time data must match '%H:%M' format"
+        with pytest.raises(ValueError, match=msg):
+            CustomBusinessHour(start="AAA")
+        msg = "time data must match '%H:%M' format"
+        with pytest.raises(ValueError, match=msg):
+            CustomBusinessHour(start="14:00:05")
+
+    def test_different_normalize_equals(self, _offset):
+        # GH#21404 changed __eq__ to return False when `normalize` does not match
+        offset = _offset()
+        offset2 = _offset(normalize=True)
+        assert offset != offset2
+
+    def test_repr(self, offset1, offset2):
+        assert repr(offset1) == "<CustomBusinessHour: cbh=09:00-17:00>"
+        assert repr(offset2) == "<CustomBusinessHour: cbh=09:00-17:00>"
+
+    def test_with_offset(self, dt):
+        expected = Timestamp("2014-07-01 13:00")
+
+        assert dt + CustomBusinessHour() * 3 == expected
+        assert dt + CustomBusinessHour(n=3) == expected
+
+    def test_eq(self, offset1, offset2):
+        for offset in [offset1, offset2]:
+            assert offset == offset
+
+        assert CustomBusinessHour() != CustomBusinessHour(-1)
+        assert CustomBusinessHour(start="09:00") == CustomBusinessHour()
+        assert CustomBusinessHour(start="09:00") != CustomBusinessHour(start="09:01")
+        assert CustomBusinessHour(start="09:00", end="17:00") != CustomBusinessHour(
+            start="17:00", end="09:01"
+        )
+
+        assert CustomBusinessHour(weekmask="Tue Wed Thu Fri") != CustomBusinessHour(
+            weekmask="Mon Tue Wed Thu Fri"
+        )
+        assert CustomBusinessHour(holidays=["2014-06-27"]) != CustomBusinessHour(
+            holidays=["2014-06-28"]
+        )
+
+    def test_hash(self, offset1, offset2):
+        assert hash(offset1) == hash(offset1)
+        assert hash(offset2) == hash(offset2)
+
+    def test_add_dateime(self, dt, offset1, offset2):
+        assert offset1 + dt == datetime(2014, 7, 1, 11)
+        assert offset2 + dt == datetime(2014, 7, 1, 11)
+
+    def testRollback1(self, dt, offset1, offset2):
+        assert offset1.rollback(dt) == dt
+        assert offset2.rollback(dt) == dt
+
+        d = datetime(2014, 7, 1, 0)
+
+        # 2014/07/01 is Tuesday, 06/30 is Monday(holiday)
+        assert offset1.rollback(d) == datetime(2014, 6, 27, 17)
+
+        # 2014/6/30 and 2014/6/27 are holidays
+        assert offset2.rollback(d) == datetime(2014, 6, 26, 17)
+
+    def testRollback2(self, _offset):
+        assert _offset(-3).rollback(datetime(2014, 7, 5, 15, 0)) == datetime(
+            2014, 7, 4, 17, 0
+        )
+
+    def testRollforward1(self, dt, offset1, offset2):
+        assert offset1.rollforward(dt) == dt
+        assert offset2.rollforward(dt) == dt
+
+        d = datetime(2014, 7, 1, 0)
+        assert offset1.rollforward(d) == datetime(2014, 7, 1, 9)
+        assert offset2.rollforward(d) == datetime(2014, 7, 1, 9)
+
+    def testRollforward2(self, _offset):
+        assert _offset(-3).rollforward(datetime(2014, 7, 5, 16, 0)) == datetime(
+            2014, 7, 7, 9
+        )
+
+    def test_roll_date_object(self):
+        offset = BusinessHour()
+
+        dt = datetime(2014, 7, 6, 15, 0)
+
+        result = offset.rollback(dt)
+        assert result == datetime(2014, 7, 4, 17)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2014, 7, 7, 9)
+
+    normalize_cases = [
+        (
+            CustomBusinessHour(normalize=True, holidays=holidays),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 3),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 3),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 3),
+                datetime(2014, 7, 1, 0): datetime(2014, 7, 1),
+                datetime(2014, 7, 4, 15): datetime(2014, 7, 4),
+                datetime(2014, 7, 4, 15, 59): datetime(2014, 7, 4),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7),
+                datetime(2014, 7, 5, 23): datetime(2014, 7, 7),
+                datetime(2014, 7, 6, 10): datetime(2014, 7, 7),
+            },
+        ),
+        (
+            CustomBusinessHour(-1, normalize=True, holidays=holidays),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 6, 26),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 10): datetime(2014, 6, 26),
+                datetime(2014, 7, 1, 0): datetime(2014, 6, 26),
+                datetime(2014, 7, 7, 10): datetime(2014, 7, 4),
+                datetime(2014, 7, 7, 10, 1): datetime(2014, 7, 7),
+                datetime(2014, 7, 5, 23): datetime(2014, 7, 4),
+                datetime(2014, 7, 6, 10): datetime(2014, 7, 4),
+            },
+        ),
+        (
+            CustomBusinessHour(
+                1, normalize=True, start="17:00", end="04:00", holidays=holidays
+            ),
+            {
+                datetime(2014, 7, 1, 8): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 1),
+                datetime(2014, 7, 1, 23): datetime(2014, 7, 2),
+                datetime(2014, 7, 2, 2): datetime(2014, 7, 2),
+                datetime(2014, 7, 2, 3): datetime(2014, 7, 3),
+                datetime(2014, 7, 4, 23): datetime(2014, 7, 5),
+                datetime(2014, 7, 5, 2): datetime(2014, 7, 5),
+                datetime(2014, 7, 7, 2): datetime(2014, 7, 7),
+                datetime(2014, 7, 7, 17): datetime(2014, 7, 7),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("norm_cases", normalize_cases)
+    def test_normalize(self, norm_cases):
+        offset, cases = norm_cases
+        for dt, expected in cases.items():
+            assert offset._apply(dt) == expected
+
+    @pytest.mark.parametrize(
+        "dt, expected",
+        [
+            [datetime(2014, 7, 1, 9), False],
+            [datetime(2014, 7, 1, 10), True],
+            [datetime(2014, 7, 1, 15), True],
+            [datetime(2014, 7, 1, 15, 1), False],
+            [datetime(2014, 7, 5, 12), False],
+            [datetime(2014, 7, 6, 12), False],
+        ],
+    )
+    def test_is_on_offset(self, dt, expected):
+        offset = CustomBusinessHour(start="10:00", end="15:00", holidays=holidays)
+        assert offset.is_on_offset(dt) == expected
+
+    apply_cases = [
+        (
+            CustomBusinessHour(holidays=holidays),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 12),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 1, 14),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 1, 16),
+                datetime(2014, 7, 1, 19): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 3, 9),
+                datetime(2014, 7, 1, 16, 30, 15): datetime(2014, 7, 3, 9, 30, 15),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 3, 10),
+                # out of business hours
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 10),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 10),
+                # saturday
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 10),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 10),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7, 9, 30),
+                datetime(2014, 7, 4, 16, 30, 30): datetime(2014, 7, 7, 9, 30, 30),
+            },
+        ),
+        (
+            CustomBusinessHour(4, holidays=holidays),
+            {
+                datetime(2014, 7, 1, 11): datetime(2014, 7, 1, 15),
+                datetime(2014, 7, 1, 13): datetime(2014, 7, 3, 9),
+                datetime(2014, 7, 1, 15): datetime(2014, 7, 3, 11),
+                datetime(2014, 7, 1, 16): datetime(2014, 7, 3, 12),
+                datetime(2014, 7, 1, 17): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 11): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 8): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 19): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 2, 23): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 3, 0): datetime(2014, 7, 3, 13),
+                datetime(2014, 7, 5, 15): datetime(2014, 7, 7, 13),
+                datetime(2014, 7, 4, 17): datetime(2014, 7, 7, 13),
+                datetime(2014, 7, 4, 16, 30): datetime(2014, 7, 7, 12, 30),
+                datetime(2014, 7, 4, 16, 30, 30): datetime(2014, 7, 7, 12, 30, 30),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("apply_case", apply_cases)
+    def test_apply(self, apply_case):
+        offset, cases = apply_case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    nano_cases = [
+        (
+            CustomBusinessHour(holidays=holidays),
+            {
+                Timestamp("2014-07-01 15:00") + Nano(5): Timestamp("2014-07-01 16:00")
+                + Nano(5),
+                Timestamp("2014-07-01 16:00") + Nano(5): Timestamp("2014-07-03 09:00")
+                + Nano(5),
+                Timestamp("2014-07-01 16:00") - Nano(5): Timestamp("2014-07-01 17:00")
+                - Nano(5),
+            },
+        ),
+        (
+            CustomBusinessHour(-1, holidays=holidays),
+            {
+                Timestamp("2014-07-01 15:00") + Nano(5): Timestamp("2014-07-01 14:00")
+                + Nano(5),
+                Timestamp("2014-07-01 10:00") + Nano(5): Timestamp("2014-07-01 09:00")
+                + Nano(5),
+                Timestamp("2014-07-01 10:00") - Nano(5): Timestamp("2014-06-26 17:00")
+                - Nano(5),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("nano_case", nano_cases)
+    def test_apply_nanoseconds(self, nano_case):
+        offset, cases = nano_case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_us_federal_holiday_with_datetime(self):
+        # GH 16867
+        bhour_us = CustomBusinessHour(calendar=USFederalHolidayCalendar())
+        t0 = datetime(2014, 1, 17, 15)
+        result = t0 + bhour_us * 8
+        expected = Timestamp("2014-01-21 15:00:00")
+        assert result == expected
+
+
+@pytest.mark.parametrize(
+    "weekmask, expected_time, mult",
+    [
+        ["Mon Tue Wed Thu Fri Sat", "2018-11-10 09:00:00", 10],
+        ["Tue Wed Thu Fri Sat", "2018-11-13 08:00:00", 18],
+    ],
+)
+def test_custom_businesshour_weekmask_and_holidays(weekmask, expected_time, mult):
+    # GH 23542
+    holidays = ["2018-11-09"]
+    bh = CustomBusinessHour(
+        start="08:00", end="17:00", weekmask=weekmask, holidays=holidays
+    )
+    result = Timestamp("2018-11-08 08:00") + mult * bh
+    expected = Timestamp(expected_time)
+    assert result == expected
diff --git a/pandas/pandas/tests/tseries/offsets/test_custom_business_month.py b/pandas/pandas/tests/tseries/offsets/test_custom_business_month.py
new file mode 100644
index 0000000000000..fd6565e3908f3
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_custom_business_month.py
@@ -0,0 +1,414 @@
+"""
+Tests for the following offsets:
+- CustomBusinessMonthBase
+- CustomBusinessMonthBegin
+- CustomBusinessMonthEnd
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import (
+    CBMonthBegin,
+    CBMonthEnd,
+    CDay,
+)
+
+import pandas._testing as tm
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries import offsets
+
+
+@pytest.fixture
+def dt():
+    return datetime(2008, 1, 1)
+
+
+class TestCommonCBM:
+    @pytest.mark.parametrize("offset2", [CBMonthBegin(2), CBMonthEnd(2)])
+    def test_eq(self, offset2):
+        assert offset2 == offset2
+
+    @pytest.mark.parametrize("offset2", [CBMonthBegin(2), CBMonthEnd(2)])
+    def test_hash(self, offset2):
+        assert hash(offset2) == hash(offset2)
+
+    @pytest.mark.parametrize("_offset", [CBMonthBegin, CBMonthEnd])
+    def test_roundtrip_pickle(self, _offset):
+        def _check_roundtrip(obj):
+            unpickled = tm.round_trip_pickle(obj)
+            assert unpickled == obj
+
+        _check_roundtrip(_offset())
+        _check_roundtrip(_offset(2))
+        _check_roundtrip(_offset() * 2)
+
+    @pytest.mark.parametrize("_offset", [CBMonthBegin, CBMonthEnd])
+    def test_copy(self, _offset):
+        # GH 17452
+        off = _offset(weekmask="Mon Wed Fri")
+        assert off == off.copy()
+
+
+class TestCustomBusinessMonthBegin:
+    def test_different_normalize_equals(self):
+        # GH#21404 changed __eq__ to return False when `normalize` does not match
+        offset = CBMonthBegin()
+        offset2 = CBMonthBegin(normalize=True)
+        assert offset != offset2
+
+    def test_repr(self):
+        assert repr(CBMonthBegin()) == "<CustomBusinessMonthBegin>"
+        assert repr(CBMonthBegin(2)) == "<2 * CustomBusinessMonthBegins>"
+
+    def test_add_datetime(self, dt):
+        assert CBMonthBegin(2) + dt == datetime(2008, 3, 3)
+
+    def testRollback1(self):
+        assert CDay(10).rollback(datetime(2007, 12, 31)) == datetime(2007, 12, 31)
+
+    def testRollback2(self, dt):
+        assert CBMonthBegin(10).rollback(dt) == datetime(2008, 1, 1)
+
+    def testRollforward1(self, dt):
+        assert CBMonthBegin(10).rollforward(dt) == datetime(2008, 1, 1)
+
+    def test_roll_date_object(self):
+        offset = CBMonthBegin()
+
+        dt = date(2012, 9, 15)
+
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 9, 3)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 10, 1)
+
+        offset = offsets.Day()
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 9, 15)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 9, 15)
+
+    on_offset_cases = [
+        (CBMonthBegin(), datetime(2008, 1, 1), True),
+        (CBMonthBegin(), datetime(2008, 1, 31), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    apply_cases = [
+        (
+            CBMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 1),
+                datetime(2008, 2, 7): datetime(2008, 3, 3),
+            },
+        ),
+        (
+            2 * CBMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 3, 3),
+                datetime(2008, 2, 7): datetime(2008, 4, 1),
+            },
+        ),
+        (
+            -CBMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2007, 12, 3),
+                datetime(2008, 2, 8): datetime(2008, 2, 1),
+            },
+        ),
+        (
+            -2 * CBMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2007, 11, 1),
+                datetime(2008, 2, 9): datetime(2008, 1, 1),
+            },
+        ),
+        (
+            CBMonthBegin(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 1, 7): datetime(2008, 2, 1),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", apply_cases)
+    def test_apply(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_apply_large_n(self):
+        dt = datetime(2012, 10, 23)
+
+        result = dt + CBMonthBegin(10)
+        assert result == datetime(2013, 8, 1)
+
+        result = dt + CDay(100) - CDay(100)
+        assert result == dt
+
+        off = CBMonthBegin() * 6
+        rs = datetime(2012, 1, 1) - off
+        xp = datetime(2011, 7, 1)
+        assert rs == xp
+
+        st = datetime(2011, 12, 18)
+        rs = st + off
+
+        xp = datetime(2012, 6, 1)
+        assert rs == xp
+
+    def test_holidays(self):
+        # Define a TradingDay offset
+        holidays = ["2012-02-01", datetime(2012, 2, 2), np.datetime64("2012-03-01")]
+        bm_offset = CBMonthBegin(holidays=holidays)
+        dt = datetime(2012, 1, 1)
+
+        assert dt + bm_offset == datetime(2012, 1, 2)
+        assert dt + 2 * bm_offset == datetime(2012, 2, 3)
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            (
+                CBMonthBegin(n=1, offset=timedelta(days=5)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 4, 1) + timedelta(days=5),
+                    datetime(2021, 4, 17): datetime(2021, 5, 3) + timedelta(days=5),
+                },
+            ),
+            (
+                CBMonthBegin(n=2, offset=timedelta(days=40)),
+                {
+                    datetime(2021, 3, 10): datetime(2021, 5, 3) + timedelta(days=40),
+                    datetime(2021, 4, 30): datetime(2021, 6, 1) + timedelta(days=40),
+                },
+            ),
+            (
+                CBMonthBegin(n=1, offset=timedelta(days=-5)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 4, 1) - timedelta(days=5),
+                    datetime(2021, 4, 11): datetime(2021, 5, 3) - timedelta(days=5),
+                },
+            ),
+            (
+                -2 * CBMonthBegin(n=1, offset=timedelta(days=10)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 1, 1) + timedelta(days=10),
+                    datetime(2021, 4, 3): datetime(2021, 3, 1) + timedelta(days=10),
+                },
+            ),
+            (
+                CBMonthBegin(n=0, offset=timedelta(days=1)),
+                {
+                    datetime(2021, 3, 2): datetime(2021, 4, 1) + timedelta(days=1),
+                    datetime(2021, 4, 1): datetime(2021, 4, 1) + timedelta(days=1),
+                },
+            ),
+            (
+                CBMonthBegin(
+                    n=1, holidays=["2021-04-01", "2021-04-02"], offset=timedelta(days=1)
+                ),
+                {
+                    datetime(2021, 3, 2): datetime(2021, 4, 5) + timedelta(days=1),
+                },
+            ),
+        ],
+    )
+    def test_apply_with_extra_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+
+class TestCustomBusinessMonthEnd:
+    def test_different_normalize_equals(self):
+        # GH#21404 changed __eq__ to return False when `normalize` does not match
+        offset = CBMonthEnd()
+        offset2 = CBMonthEnd(normalize=True)
+        assert offset != offset2
+
+    def test_repr(self):
+        assert repr(CBMonthEnd()) == "<CustomBusinessMonthEnd>"
+        assert repr(CBMonthEnd(2)) == "<2 * CustomBusinessMonthEnds>"
+
+    def test_add_datetime(self, dt):
+        assert CBMonthEnd(2) + dt == datetime(2008, 2, 29)
+
+    def testRollback1(self):
+        assert CDay(10).rollback(datetime(2007, 12, 31)) == datetime(2007, 12, 31)
+
+    def testRollback2(self, dt):
+        assert CBMonthEnd(10).rollback(dt) == datetime(2007, 12, 31)
+
+    def testRollforward1(self, dt):
+        assert CBMonthEnd(10).rollforward(dt) == datetime(2008, 1, 31)
+
+    def test_roll_date_object(self):
+        offset = CBMonthEnd()
+
+        dt = date(2012, 9, 15)
+
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 8, 31)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 9, 28)
+
+        offset = offsets.Day()
+        result = offset.rollback(dt)
+        assert result == datetime(2012, 9, 15)
+
+        result = offset.rollforward(dt)
+        assert result == datetime(2012, 9, 15)
+
+    on_offset_cases = [
+        (CBMonthEnd(), datetime(2008, 1, 31), True),
+        (CBMonthEnd(), datetime(2008, 1, 1), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    apply_cases = [
+        (
+            CBMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 2, 7): datetime(2008, 2, 29),
+            },
+        ),
+        (
+            2 * CBMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 29),
+                datetime(2008, 2, 7): datetime(2008, 3, 31),
+            },
+        ),
+        (
+            -CBMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2007, 12, 31),
+                datetime(2008, 2, 8): datetime(2008, 1, 31),
+            },
+        ),
+        (
+            -2 * CBMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2007, 11, 30),
+                datetime(2008, 2, 9): datetime(2007, 12, 31),
+            },
+        ),
+        (
+            CBMonthEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 2, 7): datetime(2008, 2, 29),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", apply_cases)
+    def test_apply(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    def test_apply_large_n(self):
+        dt = datetime(2012, 10, 23)
+
+        result = dt + CBMonthEnd(10)
+        assert result == datetime(2013, 7, 31)
+
+        result = dt + CDay(100) - CDay(100)
+        assert result == dt
+
+        off = CBMonthEnd() * 6
+        rs = datetime(2012, 1, 1) - off
+        xp = datetime(2011, 7, 29)
+        assert rs == xp
+
+        st = datetime(2011, 12, 18)
+        rs = st + off
+        xp = datetime(2012, 5, 31)
+        assert rs == xp
+
+    def test_holidays(self):
+        # Define a TradingDay offset
+        holidays = ["2012-01-31", datetime(2012, 2, 28), np.datetime64("2012-02-29")]
+        bm_offset = CBMonthEnd(holidays=holidays)
+        dt = datetime(2012, 1, 1)
+        assert dt + bm_offset == datetime(2012, 1, 30)
+        assert dt + 2 * bm_offset == datetime(2012, 2, 27)
+
+    @pytest.mark.parametrize(
+        "case",
+        [
+            (
+                CBMonthEnd(n=1, offset=timedelta(days=5)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 3, 31) + timedelta(days=5),
+                    datetime(2021, 4, 17): datetime(2021, 4, 30) + timedelta(days=5),
+                },
+            ),
+            (
+                CBMonthEnd(n=2, offset=timedelta(days=40)),
+                {
+                    datetime(2021, 3, 10): datetime(2021, 4, 30) + timedelta(days=40),
+                    datetime(2021, 4, 30): datetime(2021, 6, 30) + timedelta(days=40),
+                },
+            ),
+            (
+                CBMonthEnd(n=1, offset=timedelta(days=-5)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 3, 31) - timedelta(days=5),
+                    datetime(2021, 4, 11): datetime(2021, 4, 30) - timedelta(days=5),
+                },
+            ),
+            (
+                -2 * CBMonthEnd(n=1, offset=timedelta(days=10)),
+                {
+                    datetime(2021, 3, 1): datetime(2021, 1, 29) + timedelta(days=10),
+                    datetime(2021, 4, 3): datetime(2021, 2, 26) + timedelta(days=10),
+                },
+            ),
+            (
+                CBMonthEnd(n=0, offset=timedelta(days=1)),
+                {
+                    datetime(2021, 3, 2): datetime(2021, 3, 31) + timedelta(days=1),
+                    datetime(2021, 4, 1): datetime(2021, 4, 30) + timedelta(days=1),
+                },
+            ),
+            (
+                CBMonthEnd(n=1, holidays=["2021-03-31"], offset=timedelta(days=1)),
+                {
+                    datetime(2021, 3, 2): datetime(2021, 3, 30) + timedelta(days=1),
+                },
+            ),
+        ],
+    )
+    def test_apply_with_extra_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_dst.py b/pandas/pandas/tests/tseries/offsets/test_dst.py
new file mode 100644
index 0000000000000..e75958843040d
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_dst.py
@@ -0,0 +1,265 @@
+"""
+Tests for DateOffset additions over Daylight Savings Time
+"""
+
+from datetime import timedelta
+
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas._libs.tslibs.offsets import (
+    BMonthBegin,
+    BMonthEnd,
+    BQuarterBegin,
+    BQuarterEnd,
+    BYearBegin,
+    BYearEnd,
+    CBMonthBegin,
+    CBMonthEnd,
+    CustomBusinessDay,
+    DateOffset,
+    Day,
+    MonthBegin,
+    MonthEnd,
+    QuarterBegin,
+    QuarterEnd,
+    SemiMonthBegin,
+    SemiMonthEnd,
+    Week,
+    YearBegin,
+    YearEnd,
+)
+
+from pandas import DatetimeIndex
+import pandas._testing as tm
+
+pytz = pytest.importorskip("pytz")
+
+
+def get_utc_offset_hours(ts):
+    # take a Timestamp and compute total hours of utc offset
+    o = ts.utcoffset()
+    return (o.days * 24 * 3600 + o.seconds) / 3600.0
+
+
+class TestDST:
+    # one microsecond before the DST transition
+    ts_pre_fallback = "2013-11-03 01:59:59.999999"
+    ts_pre_springfwd = "2013-03-10 01:59:59.999999"
+
+    # test both basic names and dateutil timezones
+    timezone_utc_offsets = {
+        pytz.timezone("US/Eastern"): {
+            "utc_offset_daylight": -4,
+            "utc_offset_standard": -5,
+        },
+        "dateutil/US/Pacific": {"utc_offset_daylight": -7, "utc_offset_standard": -8},
+    }
+    valid_date_offsets_singular = [
+        "weekday",
+        "day",
+        "hour",
+        "minute",
+        "second",
+        "microsecond",
+    ]
+    valid_date_offsets_plural = [
+        "weeks",
+        "days",
+        "hours",
+        "minutes",
+        "seconds",
+        "milliseconds",
+        "microseconds",
+    ]
+
+    def _test_all_offsets(self, n, performance_warning, **kwds):
+        valid_offsets = (
+            self.valid_date_offsets_plural
+            if n > 1
+            else self.valid_date_offsets_singular
+        )
+
+        for name in valid_offsets:
+            self._test_offset(
+                offset_name=name,
+                offset_n=n,
+                performance_warning=performance_warning,
+                **kwds,
+            )
+
+    def _test_offset(
+        self, offset_name, offset_n, tstart, expected_utc_offset, performance_warning
+    ):
+        offset = DateOffset(**{offset_name: offset_n})
+
+        if (
+            offset_name in ["hour", "minute", "second", "microsecond"]
+            and offset_n == 1
+            and tstart
+            == Timestamp(
+                "2013-11-03 01:59:59.999999-0500", tz=pytz.timezone("US/Eastern")
+            )
+        ):
+            # This addition results in an ambiguous wall time
+            err_msg = {
+                "hour": "2013-11-03 01:59:59.999999",
+                "minute": "2013-11-03 01:01:59.999999",
+                "second": "2013-11-03 01:59:01.999999",
+                "microsecond": "2013-11-03 01:59:59.000001",
+            }[offset_name]
+            with pytest.raises(ValueError, match=err_msg):
+                tstart + offset
+            # While we're here, let's check that we get the same behavior in a
+            #  vectorized path
+            dti = DatetimeIndex([tstart])
+            warn_msg = "Non-vectorized DateOffset"
+            with pytest.raises(ValueError, match=err_msg):
+                with tm.assert_produces_warning(performance_warning, match=warn_msg):
+                    dti + offset
+            return
+
+        t = tstart + offset
+        if expected_utc_offset is not None:
+            assert get_utc_offset_hours(t) == expected_utc_offset
+
+        if offset_name == "weeks":
+            # dates should match
+            assert t.date() == timedelta(days=7 * offset.kwds["weeks"]) + tstart.date()
+            # expect the same day of week, hour of day, minute, second, ...
+            assert (
+                t.dayofweek == tstart.dayofweek
+                and t.hour == tstart.hour
+                and t.minute == tstart.minute
+                and t.second == tstart.second
+            )
+        elif offset_name == "days":
+            # dates should match
+            assert timedelta(offset.kwds["days"]) + tstart.date() == t.date()
+            # expect the same hour of day, minute, second, ...
+            assert (
+                t.hour == tstart.hour
+                and t.minute == tstart.minute
+                and t.second == tstart.second
+            )
+        elif offset_name in self.valid_date_offsets_singular:
+            # expect the singular offset value to match between tstart and t
+            datepart_offset = getattr(
+                t, offset_name if offset_name != "weekday" else "dayofweek"
+            )
+            assert datepart_offset == offset.kwds[offset_name]
+        else:
+            # the offset should be the same as if it was done in UTC
+            assert t == (tstart.tz_convert("UTC") + offset).tz_convert(
+                pytz.timezone("US/Pacific")
+            )
+
+    def _make_timestamp(self, string, hrs_offset, tz):
+        if hrs_offset >= 0:
+            offset_string = f"{hrs_offset:02d}00"
+        else:
+            offset_string = f"-{(hrs_offset * -1):02}00"
+        return Timestamp(string + offset_string).tz_convert(tz)
+
+    def test_springforward_plural(self, performance_warning):
+        # test moving from standard to daylight savings
+        for tz, utc_offsets in self.timezone_utc_offsets.items():
+            hrs_pre = utc_offsets["utc_offset_standard"]
+            hrs_post = utc_offsets["utc_offset_daylight"]
+            self._test_all_offsets(
+                n=3,
+                performance_warning=performance_warning,
+                tstart=self._make_timestamp(self.ts_pre_springfwd, hrs_pre, tz),
+                expected_utc_offset=hrs_post,
+            )
+
+    def test_fallback_singular(self, performance_warning):
+        # in the case of singular offsets, we don't necessarily know which utc
+        # offset the new Timestamp will wind up in (the tz for 1 month may be
+        # different from 1 second) so we don't specify an expected_utc_offset
+        for tz, utc_offsets in self.timezone_utc_offsets.items():
+            hrs_pre = utc_offsets["utc_offset_standard"]
+            self._test_all_offsets(
+                n=1,
+                performance_warning=performance_warning,
+                tstart=self._make_timestamp(self.ts_pre_fallback, hrs_pre, tz),
+                expected_utc_offset=None,
+            )
+
+    def test_springforward_singular(self, performance_warning):
+        for tz, utc_offsets in self.timezone_utc_offsets.items():
+            hrs_pre = utc_offsets["utc_offset_standard"]
+            self._test_all_offsets(
+                n=1,
+                performance_warning=performance_warning,
+                tstart=self._make_timestamp(self.ts_pre_springfwd, hrs_pre, tz),
+                expected_utc_offset=None,
+            )
+
+    offset_classes = {
+        MonthBegin: ["11/2/2012", "12/1/2012"],
+        MonthEnd: ["11/2/2012", "11/30/2012"],
+        BMonthBegin: ["11/2/2012", "12/3/2012"],
+        BMonthEnd: ["11/2/2012", "11/30/2012"],
+        CBMonthBegin: ["11/2/2012", "12/3/2012"],
+        CBMonthEnd: ["11/2/2012", "11/30/2012"],
+        SemiMonthBegin: ["11/2/2012", "11/15/2012"],
+        SemiMonthEnd: ["11/2/2012", "11/15/2012"],
+        Week: ["11/2/2012", "11/9/2012"],
+        YearBegin: ["11/2/2012", "1/1/2013"],
+        YearEnd: ["11/2/2012", "12/31/2012"],
+        BYearBegin: ["11/2/2012", "1/1/2013"],
+        BYearEnd: ["11/2/2012", "12/31/2012"],
+        QuarterBegin: ["11/2/2012", "12/1/2012"],
+        QuarterEnd: ["11/2/2012", "12/31/2012"],
+        BQuarterBegin: ["11/2/2012", "12/3/2012"],
+        BQuarterEnd: ["11/2/2012", "12/31/2012"],
+        Day: ["11/4/2012", "11/4/2012 23:00"],
+    }.items()
+
+    @pytest.mark.parametrize("tup", offset_classes)
+    def test_all_offset_classes(self, tup):
+        offset, test_values = tup
+
+        first = Timestamp(test_values[0], tz="US/Eastern") + offset()
+        second = Timestamp(test_values[1], tz="US/Eastern")
+        assert first == second
+
+
+@pytest.mark.parametrize(
+    "original_dt, target_dt, offset, tz",
+    [
+        (
+            Timestamp("2021-10-01 01:15"),
+            Timestamp("2021-10-31 01:15"),
+            MonthEnd(1),
+            "Europe/London",
+        ),
+        (
+            Timestamp("2010-12-05 02:59"),
+            Timestamp("2010-10-31 02:59"),
+            SemiMonthEnd(-3),
+            "Europe/Paris",
+        ),
+        (
+            Timestamp("2021-10-31 01:20"),
+            Timestamp("2021-11-07 01:20"),
+            CustomBusinessDay(2, weekmask="Sun Mon"),
+            "US/Eastern",
+        ),
+        (
+            Timestamp("2020-04-03 01:30"),
+            Timestamp("2020-11-01 01:30"),
+            YearBegin(1, month=11),
+            "America/Chicago",
+        ),
+    ],
+)
+def test_nontick_offset_with_ambiguous_time_error(original_dt, target_dt, offset, tz):
+    # .apply for non-Tick offsets throws ValueError when the target dt
+    # is dst-ambiguous
+    localized_dt = original_dt.tz_localize(tz)
+
+    msg = f"Cannot infer dst time from {target_dt}, try using the 'ambiguous' argument"
+    with pytest.raises(ValueError, match=msg):
+        localized_dt + offset
diff --git a/pandas/pandas/tests/tseries/offsets/test_easter.py b/pandas/pandas/tests/tseries/offsets/test_easter.py
new file mode 100644
index 0000000000000..ada72d94434a3
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_easter.py
@@ -0,0 +1,34 @@
+"""
+Tests for the following offsets:
+- Easter
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+from pandas.tests.tseries.offsets.common import assert_offset_equal
+
+from pandas.tseries.offsets import Easter
+
+
+class TestEaster:
+    @pytest.mark.parametrize(
+        "offset,date,expected",
+        [
+            (Easter(), datetime(2010, 1, 1), datetime(2010, 4, 4)),
+            (Easter(), datetime(2010, 4, 5), datetime(2011, 4, 24)),
+            (Easter(2), datetime(2010, 1, 1), datetime(2011, 4, 24)),
+            (Easter(), datetime(2010, 4, 4), datetime(2011, 4, 24)),
+            (Easter(2), datetime(2010, 4, 4), datetime(2012, 4, 8)),
+            (-Easter(), datetime(2011, 1, 1), datetime(2010, 4, 4)),
+            (-Easter(), datetime(2010, 4, 5), datetime(2010, 4, 4)),
+            (-Easter(2), datetime(2011, 1, 1), datetime(2009, 4, 12)),
+            (-Easter(), datetime(2010, 4, 4), datetime(2009, 4, 12)),
+            (-Easter(2), datetime(2010, 4, 4), datetime(2008, 3, 23)),
+        ],
+    )
+    def test_offset(self, offset, date, expected):
+        assert_offset_equal(offset, date, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_fiscal.py b/pandas/pandas/tests/tseries/offsets/test_fiscal.py
new file mode 100644
index 0000000000000..f442b363f737d
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_fiscal.py
@@ -0,0 +1,642 @@
+"""
+Tests for Fiscal Year and Fiscal Quarter offset classes
+"""
+
+from datetime import datetime
+
+from dateutil.relativedelta import relativedelta
+import pytest
+
+from pandas import Timestamp
+from pandas.tests.tseries.offsets.common import (
+    WeekDay,
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    FY5253,
+    FY5253Quarter,
+)
+
+
+def makeFY5253LastOfMonthQuarter(*args, **kwds):
+    return FY5253Quarter(*args, variation="last", **kwds)
+
+
+def makeFY5253NearestEndMonthQuarter(*args, **kwds):
+    return FY5253Quarter(*args, variation="nearest", **kwds)
+
+
+def makeFY5253NearestEndMonth(*args, **kwds):
+    return FY5253(*args, variation="nearest", **kwds)
+
+
+def makeFY5253LastOfMonth(*args, **kwds):
+    return FY5253(*args, variation="last", **kwds)
+
+
+def test_get_offset_name():
+    assert (
+        makeFY5253LastOfMonthQuarter(
+            weekday=1, startingMonth=3, qtr_with_extra_week=4
+        ).freqstr
+        == "REQ-L-MAR-TUE-4"
+    )
+    assert (
+        makeFY5253NearestEndMonthQuarter(
+            weekday=1, startingMonth=3, qtr_with_extra_week=3
+        ).freqstr
+        == "REQ-N-MAR-TUE-3"
+    )
+
+
+class TestFY5253LastOfMonth:
+    offset_lom_sat_aug = makeFY5253LastOfMonth(1, startingMonth=8, weekday=WeekDay.SAT)
+    offset_lom_sat_sep = makeFY5253LastOfMonth(1, startingMonth=9, weekday=WeekDay.SAT)
+
+    on_offset_cases = [
+        # From Wikipedia (see:
+        # https://en.wikipedia.org/wiki/4%E2%80%934%E2%80%935_calendar#Last_Saturday_of_the_month_at_fiscal_year_end)
+        (offset_lom_sat_aug, datetime(2006, 8, 26), True),
+        (offset_lom_sat_aug, datetime(2007, 8, 25), True),
+        (offset_lom_sat_aug, datetime(2008, 8, 30), True),
+        (offset_lom_sat_aug, datetime(2009, 8, 29), True),
+        (offset_lom_sat_aug, datetime(2010, 8, 28), True),
+        (offset_lom_sat_aug, datetime(2011, 8, 27), True),
+        (offset_lom_sat_aug, datetime(2012, 8, 25), True),
+        (offset_lom_sat_aug, datetime(2013, 8, 31), True),
+        (offset_lom_sat_aug, datetime(2014, 8, 30), True),
+        (offset_lom_sat_aug, datetime(2015, 8, 29), True),
+        (offset_lom_sat_aug, datetime(2016, 8, 27), True),
+        (offset_lom_sat_aug, datetime(2017, 8, 26), True),
+        (offset_lom_sat_aug, datetime(2018, 8, 25), True),
+        (offset_lom_sat_aug, datetime(2019, 8, 31), True),
+        (offset_lom_sat_aug, datetime(2006, 8, 27), False),
+        (offset_lom_sat_aug, datetime(2007, 8, 28), False),
+        (offset_lom_sat_aug, datetime(2008, 8, 31), False),
+        (offset_lom_sat_aug, datetime(2009, 8, 30), False),
+        (offset_lom_sat_aug, datetime(2010, 8, 29), False),
+        (offset_lom_sat_aug, datetime(2011, 8, 28), False),
+        (offset_lom_sat_aug, datetime(2006, 8, 25), False),
+        (offset_lom_sat_aug, datetime(2007, 8, 24), False),
+        (offset_lom_sat_aug, datetime(2008, 8, 29), False),
+        (offset_lom_sat_aug, datetime(2009, 8, 28), False),
+        (offset_lom_sat_aug, datetime(2010, 8, 27), False),
+        (offset_lom_sat_aug, datetime(2011, 8, 26), False),
+        (offset_lom_sat_aug, datetime(2019, 8, 30), False),
+        # From GMCR (see for example:
+        # http://yahoo.brand.edgar-online.com/Default.aspx?
+        # companyid=3184&formtypeID=7)
+        (offset_lom_sat_sep, datetime(2010, 9, 25), True),
+        (offset_lom_sat_sep, datetime(2011, 9, 24), True),
+        (offset_lom_sat_sep, datetime(2012, 9, 29), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    def test_apply(self):
+        offset_lom_aug_sat = makeFY5253LastOfMonth(startingMonth=8, weekday=WeekDay.SAT)
+        offset_lom_aug_sat_1 = makeFY5253LastOfMonth(
+            n=1, startingMonth=8, weekday=WeekDay.SAT
+        )
+
+        date_seq_lom_aug_sat = [
+            datetime(2006, 8, 26),
+            datetime(2007, 8, 25),
+            datetime(2008, 8, 30),
+            datetime(2009, 8, 29),
+            datetime(2010, 8, 28),
+            datetime(2011, 8, 27),
+            datetime(2012, 8, 25),
+            datetime(2013, 8, 31),
+            datetime(2014, 8, 30),
+            datetime(2015, 8, 29),
+            datetime(2016, 8, 27),
+        ]
+
+        tests = [
+            (offset_lom_aug_sat, date_seq_lom_aug_sat),
+            (offset_lom_aug_sat_1, date_seq_lom_aug_sat),
+            (offset_lom_aug_sat, [datetime(2006, 8, 25)] + date_seq_lom_aug_sat),
+            (offset_lom_aug_sat_1, [datetime(2006, 8, 27)] + date_seq_lom_aug_sat[1:]),
+            (
+                makeFY5253LastOfMonth(n=-1, startingMonth=8, weekday=WeekDay.SAT),
+                list(reversed(date_seq_lom_aug_sat)),
+            ),
+        ]
+        for test in tests:
+            offset, data = test
+            current = data[0]
+            for datum in data[1:]:
+                current = current + offset
+                assert current == datum
+
+
+class TestFY5253NearestEndMonth:
+    def test_get_year_end(self):
+        assert makeFY5253NearestEndMonth(
+            startingMonth=8, weekday=WeekDay.SAT
+        ).get_year_end(datetime(2013, 1, 1)) == datetime(2013, 8, 31)
+        assert makeFY5253NearestEndMonth(
+            startingMonth=8, weekday=WeekDay.SUN
+        ).get_year_end(datetime(2013, 1, 1)) == datetime(2013, 9, 1)
+        assert makeFY5253NearestEndMonth(
+            startingMonth=8, weekday=WeekDay.FRI
+        ).get_year_end(datetime(2013, 1, 1)) == datetime(2013, 8, 30)
+
+        offset_n = FY5253(weekday=WeekDay.TUE, startingMonth=12, variation="nearest")
+        assert offset_n.get_year_end(datetime(2012, 1, 1)) == datetime(2013, 1, 1)
+        assert offset_n.get_year_end(datetime(2012, 1, 10)) == datetime(2013, 1, 1)
+
+        assert offset_n.get_year_end(datetime(2013, 1, 1)) == datetime(2013, 12, 31)
+        assert offset_n.get_year_end(datetime(2013, 1, 2)) == datetime(2013, 12, 31)
+        assert offset_n.get_year_end(datetime(2013, 1, 3)) == datetime(2013, 12, 31)
+        assert offset_n.get_year_end(datetime(2013, 1, 10)) == datetime(2013, 12, 31)
+
+        JNJ = FY5253(n=1, startingMonth=12, weekday=6, variation="nearest")
+        assert JNJ.get_year_end(datetime(2006, 1, 1)) == datetime(2006, 12, 31)
+
+    offset_lom_aug_sat = makeFY5253NearestEndMonth(
+        1, startingMonth=8, weekday=WeekDay.SAT
+    )
+    offset_lom_aug_thu = makeFY5253NearestEndMonth(
+        1, startingMonth=8, weekday=WeekDay.THU
+    )
+    offset_n = FY5253(weekday=WeekDay.TUE, startingMonth=12, variation="nearest")
+
+    on_offset_cases = [
+        #    From Wikipedia (see:
+        #    https://en.wikipedia.org/wiki/4%E2%80%934%E2%80%935_calendar
+        #    #Saturday_nearest_the_end_of_month)
+        #    2006-09-02   2006 September 2
+        #    2007-09-01   2007 September 1
+        #    2008-08-30   2008 August 30    (leap year)
+        #    2009-08-29   2009 August 29
+        #    2010-08-28   2010 August 28
+        #    2011-09-03   2011 September 3
+        #    2012-09-01   2012 September 1  (leap year)
+        #    2013-08-31   2013 August 31
+        #    2014-08-30   2014 August 30
+        #    2015-08-29   2015 August 29
+        #    2016-09-03   2016 September 3  (leap year)
+        #    2017-09-02   2017 September 2
+        #    2018-09-01   2018 September 1
+        #    2019-08-31   2019 August 31
+        (offset_lom_aug_sat, datetime(2006, 9, 2), True),
+        (offset_lom_aug_sat, datetime(2007, 9, 1), True),
+        (offset_lom_aug_sat, datetime(2008, 8, 30), True),
+        (offset_lom_aug_sat, datetime(2009, 8, 29), True),
+        (offset_lom_aug_sat, datetime(2010, 8, 28), True),
+        (offset_lom_aug_sat, datetime(2011, 9, 3), True),
+        (offset_lom_aug_sat, datetime(2016, 9, 3), True),
+        (offset_lom_aug_sat, datetime(2017, 9, 2), True),
+        (offset_lom_aug_sat, datetime(2018, 9, 1), True),
+        (offset_lom_aug_sat, datetime(2019, 8, 31), True),
+        (offset_lom_aug_sat, datetime(2006, 8, 27), False),
+        (offset_lom_aug_sat, datetime(2007, 8, 28), False),
+        (offset_lom_aug_sat, datetime(2008, 8, 31), False),
+        (offset_lom_aug_sat, datetime(2009, 8, 30), False),
+        (offset_lom_aug_sat, datetime(2010, 8, 29), False),
+        (offset_lom_aug_sat, datetime(2011, 8, 28), False),
+        (offset_lom_aug_sat, datetime(2006, 8, 25), False),
+        (offset_lom_aug_sat, datetime(2007, 8, 24), False),
+        (offset_lom_aug_sat, datetime(2008, 8, 29), False),
+        (offset_lom_aug_sat, datetime(2009, 8, 28), False),
+        (offset_lom_aug_sat, datetime(2010, 8, 27), False),
+        (offset_lom_aug_sat, datetime(2011, 8, 26), False),
+        (offset_lom_aug_sat, datetime(2019, 8, 30), False),
+        # From Micron, see:
+        # http://google.brand.edgar-online.com/?sym=MU&formtypeID=7
+        (offset_lom_aug_thu, datetime(2012, 8, 30), True),
+        (offset_lom_aug_thu, datetime(2011, 9, 1), True),
+        (offset_n, datetime(2012, 12, 31), False),
+        (offset_n, datetime(2013, 1, 1), True),
+        (offset_n, datetime(2013, 1, 2), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    def test_apply(self):
+        date_seq_nem_8_sat = [
+            datetime(2006, 9, 2),
+            datetime(2007, 9, 1),
+            datetime(2008, 8, 30),
+            datetime(2009, 8, 29),
+            datetime(2010, 8, 28),
+            datetime(2011, 9, 3),
+        ]
+
+        JNJ = [
+            datetime(2005, 1, 2),
+            datetime(2006, 1, 1),
+            datetime(2006, 12, 31),
+            datetime(2007, 12, 30),
+            datetime(2008, 12, 28),
+            datetime(2010, 1, 3),
+            datetime(2011, 1, 2),
+            datetime(2012, 1, 1),
+            datetime(2012, 12, 30),
+        ]
+
+        DEC_SAT = FY5253(n=-1, startingMonth=12, weekday=5, variation="nearest")
+
+        tests = [
+            (
+                makeFY5253NearestEndMonth(startingMonth=8, weekday=WeekDay.SAT),
+                date_seq_nem_8_sat,
+            ),
+            (
+                makeFY5253NearestEndMonth(n=1, startingMonth=8, weekday=WeekDay.SAT),
+                date_seq_nem_8_sat,
+            ),
+            (
+                makeFY5253NearestEndMonth(startingMonth=8, weekday=WeekDay.SAT),
+                [datetime(2006, 9, 1)] + date_seq_nem_8_sat,
+            ),
+            (
+                makeFY5253NearestEndMonth(n=1, startingMonth=8, weekday=WeekDay.SAT),
+                [datetime(2006, 9, 3)] + date_seq_nem_8_sat[1:],
+            ),
+            (
+                makeFY5253NearestEndMonth(n=-1, startingMonth=8, weekday=WeekDay.SAT),
+                list(reversed(date_seq_nem_8_sat)),
+            ),
+            (
+                makeFY5253NearestEndMonth(n=1, startingMonth=12, weekday=WeekDay.SUN),
+                JNJ,
+            ),
+            (
+                makeFY5253NearestEndMonth(n=-1, startingMonth=12, weekday=WeekDay.SUN),
+                list(reversed(JNJ)),
+            ),
+            (
+                makeFY5253NearestEndMonth(n=1, startingMonth=12, weekday=WeekDay.SUN),
+                [datetime(2005, 1, 2), datetime(2006, 1, 1)],
+            ),
+            (
+                makeFY5253NearestEndMonth(n=1, startingMonth=12, weekday=WeekDay.SUN),
+                [datetime(2006, 1, 2), datetime(2006, 12, 31)],
+            ),
+            (DEC_SAT, [datetime(2013, 1, 15), datetime(2012, 12, 29)]),
+        ]
+        for test in tests:
+            offset, data = test
+            current = data[0]
+            for datum in data[1:]:
+                current = current + offset
+                assert current == datum
+
+
+class TestFY5253LastOfMonthQuarter:
+    def test_equality(self):
+        assert makeFY5253LastOfMonthQuarter(
+            startingMonth=1, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        ) == makeFY5253LastOfMonthQuarter(
+            startingMonth=1, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+        assert makeFY5253LastOfMonthQuarter(
+            startingMonth=1, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        ) != makeFY5253LastOfMonthQuarter(
+            startingMonth=1, weekday=WeekDay.SUN, qtr_with_extra_week=4
+        )
+        assert makeFY5253LastOfMonthQuarter(
+            startingMonth=1, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        ) != makeFY5253LastOfMonthQuarter(
+            startingMonth=2, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+
+    def test_offset(self):
+        offset = makeFY5253LastOfMonthQuarter(
+            1, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+        offset2 = makeFY5253LastOfMonthQuarter(
+            2, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+        offset4 = makeFY5253LastOfMonthQuarter(
+            4, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+
+        offset_neg1 = makeFY5253LastOfMonthQuarter(
+            -1, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+        offset_neg2 = makeFY5253LastOfMonthQuarter(
+            -2, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+
+        GMCR = [
+            datetime(2010, 3, 27),
+            datetime(2010, 6, 26),
+            datetime(2010, 9, 25),
+            datetime(2010, 12, 25),
+            datetime(2011, 3, 26),
+            datetime(2011, 6, 25),
+            datetime(2011, 9, 24),
+            datetime(2011, 12, 24),
+            datetime(2012, 3, 24),
+            datetime(2012, 6, 23),
+            datetime(2012, 9, 29),
+            datetime(2012, 12, 29),
+            datetime(2013, 3, 30),
+            datetime(2013, 6, 29),
+        ]
+
+        assert_offset_equal(offset, base=GMCR[0], expected=GMCR[1])
+        assert_offset_equal(
+            offset, base=GMCR[0] + relativedelta(days=-1), expected=GMCR[0]
+        )
+        assert_offset_equal(offset, base=GMCR[1], expected=GMCR[2])
+
+        assert_offset_equal(offset2, base=GMCR[0], expected=GMCR[2])
+        assert_offset_equal(offset4, base=GMCR[0], expected=GMCR[4])
+
+        assert_offset_equal(offset_neg1, base=GMCR[-1], expected=GMCR[-2])
+        assert_offset_equal(
+            offset_neg1, base=GMCR[-1] + relativedelta(days=+1), expected=GMCR[-1]
+        )
+        assert_offset_equal(offset_neg2, base=GMCR[-1], expected=GMCR[-3])
+
+        date = GMCR[0] + relativedelta(days=-1)
+        for expected in GMCR:
+            assert_offset_equal(offset, date, expected)
+            date = date + offset
+
+        date = GMCR[-1] + relativedelta(days=+1)
+        for expected in reversed(GMCR):
+            assert_offset_equal(offset_neg1, date, expected)
+            date = date + offset_neg1
+
+    lomq_aug_sat_4 = makeFY5253LastOfMonthQuarter(
+        1, startingMonth=8, weekday=WeekDay.SAT, qtr_with_extra_week=4
+    )
+    lomq_sep_sat_4 = makeFY5253LastOfMonthQuarter(
+        1, startingMonth=9, weekday=WeekDay.SAT, qtr_with_extra_week=4
+    )
+
+    on_offset_cases = [
+        # From Wikipedia
+        (lomq_aug_sat_4, datetime(2006, 8, 26), True),
+        (lomq_aug_sat_4, datetime(2007, 8, 25), True),
+        (lomq_aug_sat_4, datetime(2008, 8, 30), True),
+        (lomq_aug_sat_4, datetime(2009, 8, 29), True),
+        (lomq_aug_sat_4, datetime(2010, 8, 28), True),
+        (lomq_aug_sat_4, datetime(2011, 8, 27), True),
+        (lomq_aug_sat_4, datetime(2019, 8, 31), True),
+        (lomq_aug_sat_4, datetime(2006, 8, 27), False),
+        (lomq_aug_sat_4, datetime(2007, 8, 28), False),
+        (lomq_aug_sat_4, datetime(2008, 8, 31), False),
+        (lomq_aug_sat_4, datetime(2009, 8, 30), False),
+        (lomq_aug_sat_4, datetime(2010, 8, 29), False),
+        (lomq_aug_sat_4, datetime(2011, 8, 28), False),
+        (lomq_aug_sat_4, datetime(2006, 8, 25), False),
+        (lomq_aug_sat_4, datetime(2007, 8, 24), False),
+        (lomq_aug_sat_4, datetime(2008, 8, 29), False),
+        (lomq_aug_sat_4, datetime(2009, 8, 28), False),
+        (lomq_aug_sat_4, datetime(2010, 8, 27), False),
+        (lomq_aug_sat_4, datetime(2011, 8, 26), False),
+        (lomq_aug_sat_4, datetime(2019, 8, 30), False),
+        # From GMCR
+        (lomq_sep_sat_4, datetime(2010, 9, 25), True),
+        (lomq_sep_sat_4, datetime(2011, 9, 24), True),
+        (lomq_sep_sat_4, datetime(2012, 9, 29), True),
+        (lomq_sep_sat_4, datetime(2013, 6, 29), True),
+        (lomq_sep_sat_4, datetime(2012, 6, 23), True),
+        (lomq_sep_sat_4, datetime(2012, 6, 30), False),
+        (lomq_sep_sat_4, datetime(2013, 3, 30), True),
+        (lomq_sep_sat_4, datetime(2012, 3, 24), True),
+        (lomq_sep_sat_4, datetime(2012, 12, 29), True),
+        (lomq_sep_sat_4, datetime(2011, 12, 24), True),
+        # INTC (extra week in Q1)
+        # See: http://www.intc.com/releasedetail.cfm?ReleaseID=542844
+        (
+            makeFY5253LastOfMonthQuarter(
+                1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+            ),
+            datetime(2011, 4, 2),
+            True,
+        ),
+        # see: http://google.brand.edgar-online.com/?sym=INTC&formtypeID=7
+        (
+            makeFY5253LastOfMonthQuarter(
+                1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+            ),
+            datetime(2012, 12, 29),
+            True,
+        ),
+        (
+            makeFY5253LastOfMonthQuarter(
+                1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+            ),
+            datetime(2011, 12, 31),
+            True,
+        ),
+        (
+            makeFY5253LastOfMonthQuarter(
+                1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+            ),
+            datetime(2010, 12, 25),
+            True,
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    def test_year_has_extra_week(self):
+        # End of long Q1
+        assert makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(2011, 4, 2))
+
+        # Start of long Q1
+        assert makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(2010, 12, 26))
+
+        # End of year before year with long Q1
+        assert not makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(2010, 12, 25))
+
+        for year in [
+            x for x in range(1994, 2011 + 1) if x not in [2011, 2005, 2000, 1994]
+        ]:
+            assert not makeFY5253LastOfMonthQuarter(
+                1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+            ).year_has_extra_week(datetime(year, 4, 2))
+
+        # Other long years
+        assert makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(2005, 4, 2))
+
+        assert makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(2000, 4, 2))
+
+        assert makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        ).year_has_extra_week(datetime(1994, 4, 2))
+
+    def test_get_weeks(self):
+        sat_dec_1 = makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=1
+        )
+        sat_dec_4 = makeFY5253LastOfMonthQuarter(
+            1, startingMonth=12, weekday=WeekDay.SAT, qtr_with_extra_week=4
+        )
+
+        assert sat_dec_1.get_weeks(datetime(2011, 4, 2)) == [14, 13, 13, 13]
+        assert sat_dec_4.get_weeks(datetime(2011, 4, 2)) == [13, 13, 13, 14]
+        assert sat_dec_1.get_weeks(datetime(2010, 12, 25)) == [13, 13, 13, 13]
+
+
+class TestFY5253NearestEndMonthQuarter:
+    offset_nem_sat_aug_4 = makeFY5253NearestEndMonthQuarter(
+        1, startingMonth=8, weekday=WeekDay.SAT, qtr_with_extra_week=4
+    )
+    offset_nem_thu_aug_4 = makeFY5253NearestEndMonthQuarter(
+        1, startingMonth=8, weekday=WeekDay.THU, qtr_with_extra_week=4
+    )
+    offset_n = FY5253(weekday=WeekDay.TUE, startingMonth=12, variation="nearest")
+
+    on_offset_cases = [
+        # From Wikipedia
+        (offset_nem_sat_aug_4, datetime(2006, 9, 2), True),
+        (offset_nem_sat_aug_4, datetime(2007, 9, 1), True),
+        (offset_nem_sat_aug_4, datetime(2008, 8, 30), True),
+        (offset_nem_sat_aug_4, datetime(2009, 8, 29), True),
+        (offset_nem_sat_aug_4, datetime(2010, 8, 28), True),
+        (offset_nem_sat_aug_4, datetime(2011, 9, 3), True),
+        (offset_nem_sat_aug_4, datetime(2016, 9, 3), True),
+        (offset_nem_sat_aug_4, datetime(2017, 9, 2), True),
+        (offset_nem_sat_aug_4, datetime(2018, 9, 1), True),
+        (offset_nem_sat_aug_4, datetime(2019, 8, 31), True),
+        (offset_nem_sat_aug_4, datetime(2006, 8, 27), False),
+        (offset_nem_sat_aug_4, datetime(2007, 8, 28), False),
+        (offset_nem_sat_aug_4, datetime(2008, 8, 31), False),
+        (offset_nem_sat_aug_4, datetime(2009, 8, 30), False),
+        (offset_nem_sat_aug_4, datetime(2010, 8, 29), False),
+        (offset_nem_sat_aug_4, datetime(2011, 8, 28), False),
+        (offset_nem_sat_aug_4, datetime(2006, 8, 25), False),
+        (offset_nem_sat_aug_4, datetime(2007, 8, 24), False),
+        (offset_nem_sat_aug_4, datetime(2008, 8, 29), False),
+        (offset_nem_sat_aug_4, datetime(2009, 8, 28), False),
+        (offset_nem_sat_aug_4, datetime(2010, 8, 27), False),
+        (offset_nem_sat_aug_4, datetime(2011, 8, 26), False),
+        (offset_nem_sat_aug_4, datetime(2019, 8, 30), False),
+        # From Micron, see:
+        # http://google.brand.edgar-online.com/?sym=MU&formtypeID=7
+        (offset_nem_thu_aug_4, datetime(2012, 8, 30), True),
+        (offset_nem_thu_aug_4, datetime(2011, 9, 1), True),
+        # See: http://google.brand.edgar-online.com/?sym=MU&formtypeID=13
+        (offset_nem_thu_aug_4, datetime(2013, 5, 30), True),
+        (offset_nem_thu_aug_4, datetime(2013, 2, 28), True),
+        (offset_nem_thu_aug_4, datetime(2012, 11, 29), True),
+        (offset_nem_thu_aug_4, datetime(2012, 5, 31), True),
+        (offset_nem_thu_aug_4, datetime(2007, 3, 1), True),
+        (offset_nem_thu_aug_4, datetime(1994, 3, 3), True),
+        (offset_n, datetime(2012, 12, 31), False),
+        (offset_n, datetime(2013, 1, 1), True),
+        (offset_n, datetime(2013, 1, 2), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+    def test_offset(self):
+        offset = makeFY5253NearestEndMonthQuarter(
+            1, startingMonth=8, weekday=WeekDay.THU, qtr_with_extra_week=4
+        )
+
+        MU = [
+            datetime(2012, 5, 31),
+            datetime(2012, 8, 30),
+            datetime(2012, 11, 29),
+            datetime(2013, 2, 28),
+            datetime(2013, 5, 30),
+        ]
+
+        date = MU[0] + relativedelta(days=-1)
+        for expected in MU:
+            assert_offset_equal(offset, date, expected)
+            date = date + offset
+
+        assert_offset_equal(offset, datetime(2012, 5, 31), datetime(2012, 8, 30))
+        assert_offset_equal(offset, datetime(2012, 5, 30), datetime(2012, 5, 31))
+
+        offset2 = FY5253Quarter(
+            weekday=5, startingMonth=12, variation="last", qtr_with_extra_week=4
+        )
+
+        assert_offset_equal(offset2, datetime(2013, 1, 15), datetime(2013, 3, 30))
+
+
+def test_bunched_yearends():
+    # GH#14774 cases with two fiscal year-ends in the same calendar-year
+    fy = FY5253(n=1, weekday=5, startingMonth=12, variation="nearest")
+    dt = Timestamp("2004-01-01")
+    assert fy.rollback(dt) == Timestamp("2002-12-28")
+    assert (-fy)._apply(dt) == Timestamp("2002-12-28")
+    assert dt - fy == Timestamp("2002-12-28")
+
+    assert fy.rollforward(dt) == Timestamp("2004-01-03")
+    assert fy._apply(dt) == Timestamp("2004-01-03")
+    assert fy + dt == Timestamp("2004-01-03")
+    assert dt + fy == Timestamp("2004-01-03")
+
+    # Same thing, but starting from a Timestamp in the previous year.
+    dt = Timestamp("2003-12-31")
+    assert fy.rollback(dt) == Timestamp("2002-12-28")
+    assert (-fy)._apply(dt) == Timestamp("2002-12-28")
+    assert dt - fy == Timestamp("2002-12-28")
+
+
+def test_fy5253_last_onoffset():
+    # GH#18877 dates on the year-end but not normalized to midnight
+    offset = FY5253(n=-5, startingMonth=5, variation="last", weekday=0)
+    ts = Timestamp("1984-05-28 06:29:43.955911354+0200", tz="Europe/San_Marino")
+    fast = offset.is_on_offset(ts)
+    slow = (ts + offset) - offset == ts
+    assert fast == slow
+
+
+def test_fy5253_nearest_onoffset():
+    # GH#18877 dates on the year-end but not normalized to midnight
+    offset = FY5253(n=3, startingMonth=7, variation="nearest", weekday=2)
+    ts = Timestamp("2032-07-28 00:12:59.035729419+0000", tz="Africa/Dakar")
+    fast = offset.is_on_offset(ts)
+    slow = (ts + offset) - offset == ts
+    assert fast == slow
+
+
+def test_fy5253qtr_onoffset_nearest():
+    # GH#19036
+    ts = Timestamp("1985-09-02 23:57:46.232550356-0300", tz="Atlantic/Bermuda")
+    offset = FY5253Quarter(
+        n=3, qtr_with_extra_week=1, startingMonth=2, variation="nearest", weekday=0
+    )
+    fast = offset.is_on_offset(ts)
+    slow = (ts + offset) - offset == ts
+    assert fast == slow
+
+
+def test_fy5253qtr_onoffset_last():
+    # GH#19036
+    offset = FY5253Quarter(
+        n=-2, qtr_with_extra_week=1, startingMonth=7, variation="last", weekday=2
+    )
+    ts = Timestamp("2011-01-26 19:03:40.331096129+0200", tz="Africa/Windhoek")
+    slow = (ts + offset) - offset == ts
+    fast = offset.is_on_offset(ts)
+    assert fast == slow
diff --git a/pandas/pandas/tests/tseries/offsets/test_index.py b/pandas/pandas/tests/tseries/offsets/test_index.py
new file mode 100644
index 0000000000000..4fb9815ba92bb
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_index.py
@@ -0,0 +1,58 @@
+"""
+Tests for offset behavior with indices.
+"""
+
+import pytest
+
+from pandas import (
+    Series,
+    date_range,
+)
+
+from pandas.tseries.offsets import (
+    BMonthBegin,
+    BMonthEnd,
+    BQuarterBegin,
+    BQuarterEnd,
+    BYearBegin,
+    BYearEnd,
+    MonthBegin,
+    MonthEnd,
+    QuarterBegin,
+    QuarterEnd,
+    YearBegin,
+    YearEnd,
+)
+
+
+@pytest.mark.parametrize("n", [-2, 1])
+@pytest.mark.parametrize(
+    "cls",
+    [
+        MonthBegin,
+        MonthEnd,
+        BMonthBegin,
+        BMonthEnd,
+        QuarterBegin,
+        QuarterEnd,
+        BQuarterBegin,
+        BQuarterEnd,
+        YearBegin,
+        YearEnd,
+        BYearBegin,
+        BYearEnd,
+    ],
+)
+def test_apply_index(cls, n):
+    offset = cls(n=n)
+    rng = date_range(start="1/1/2000", periods=100000, freq="min")
+    ser = Series(rng)
+
+    res = rng + offset
+    assert res.freq is None  # not retained
+    assert res[0] == rng[0] + offset
+    assert res[-1] == rng[-1] + offset
+    res2 = ser + offset
+    # apply_index is only for indexes, not series, so no res2_v2
+    assert res2.iloc[0] == ser.iloc[0] + offset
+    assert res2.iloc[-1] == ser.iloc[-1] + offset
diff --git a/pandas/pandas/tests/tseries/offsets/test_month.py b/pandas/pandas/tests/tseries/offsets/test_month.py
new file mode 100644
index 0000000000000..4dd494d0872a1
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_month.py
@@ -0,0 +1,667 @@
+"""
+Tests for the following offsets:
+- SemiMonthBegin
+- SemiMonthEnd
+- MonthBegin
+- MonthEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas._libs.tslibs.offsets import (
+    MonthBegin,
+    MonthEnd,
+    SemiMonthBegin,
+    SemiMonthEnd,
+)
+
+from pandas import (
+    DatetimeIndex,
+    Series,
+    _testing as tm,
+)
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+
+class TestSemiMonthEnd:
+    def test_offset_whole_year(self):
+        dates = (
+            datetime(2007, 12, 31),
+            datetime(2008, 1, 15),
+            datetime(2008, 1, 31),
+            datetime(2008, 2, 15),
+            datetime(2008, 2, 29),
+            datetime(2008, 3, 15),
+            datetime(2008, 3, 31),
+            datetime(2008, 4, 15),
+            datetime(2008, 4, 30),
+            datetime(2008, 5, 15),
+            datetime(2008, 5, 31),
+            datetime(2008, 6, 15),
+            datetime(2008, 6, 30),
+            datetime(2008, 7, 15),
+            datetime(2008, 7, 31),
+            datetime(2008, 8, 15),
+            datetime(2008, 8, 31),
+            datetime(2008, 9, 15),
+            datetime(2008, 9, 30),
+            datetime(2008, 10, 15),
+            datetime(2008, 10, 31),
+            datetime(2008, 11, 15),
+            datetime(2008, 11, 30),
+            datetime(2008, 12, 15),
+            datetime(2008, 12, 31),
+        )
+
+        for base, exp_date in zip(dates[:-1], dates[1:]):
+            assert_offset_equal(SemiMonthEnd(), base, exp_date)
+
+        # ensure .apply_index works as expected
+        shift = DatetimeIndex(dates[:-1])
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = SemiMonthEnd() + shift
+
+        exp = DatetimeIndex(dates[1:])
+        tm.assert_index_equal(result, exp)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            SemiMonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 15),
+                datetime(2008, 1, 15): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 2, 15),
+                datetime(2006, 12, 14): datetime(2006, 12, 15),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2007, 1, 15),
+                datetime(2007, 1, 1): datetime(2007, 1, 15),
+                datetime(2006, 12, 1): datetime(2006, 12, 15),
+                datetime(2006, 12, 15): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(day_of_month=20),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 20),
+                datetime(2008, 1, 15): datetime(2008, 1, 20),
+                datetime(2008, 1, 21): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 2, 20),
+                datetime(2006, 12, 14): datetime(2006, 12, 20),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2007, 1, 20),
+                datetime(2007, 1, 1): datetime(2007, 1, 20),
+                datetime(2006, 12, 1): datetime(2006, 12, 20),
+                datetime(2006, 12, 15): datetime(2006, 12, 20),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 15),
+                datetime(2008, 1, 16): datetime(2008, 1, 31),
+                datetime(2008, 1, 15): datetime(2008, 1, 15),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2006, 12, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 15),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(0, day_of_month=16),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 16),
+                datetime(2008, 1, 16): datetime(2008, 1, 16),
+                datetime(2008, 1, 15): datetime(2008, 1, 16),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2006, 12, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 16),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 2, 29),
+                datetime(2006, 12, 29): datetime(2007, 1, 15),
+                datetime(2006, 12, 31): datetime(2007, 1, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 31),
+                datetime(2007, 1, 16): datetime(2007, 2, 15),
+                datetime(2006, 11, 1): datetime(2006, 11, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 6, 15),
+                datetime(2008, 12, 31): datetime(2008, 12, 15),
+                datetime(2006, 12, 29): datetime(2006, 12, 15),
+                datetime(2006, 12, 30): datetime(2006, 12, 15),
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(-1, day_of_month=4),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+                datetime(2007, 1, 4): datetime(2006, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 6, 4),
+                datetime(2008, 12, 31): datetime(2008, 12, 4),
+                datetime(2006, 12, 5): datetime(2006, 12, 4),
+                datetime(2006, 12, 30): datetime(2006, 12, 4),
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            SemiMonthEnd(-2),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 15),
+                datetime(2008, 6, 30): datetime(2008, 5, 31),
+                datetime(2008, 3, 15): datetime(2008, 2, 15),
+                datetime(2008, 12, 31): datetime(2008, 11, 30),
+                datetime(2006, 12, 29): datetime(2006, 11, 30),
+                datetime(2006, 12, 14): datetime(2006, 11, 15),
+                datetime(2007, 1, 1): datetime(2006, 12, 15),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_apply_index(self, case):
+        # https://github.com/pandas-dev/pandas/issues/34580
+        offset, cases = case
+        shift = DatetimeIndex(cases.keys())
+        exp = DatetimeIndex(cases.values())
+
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = offset + shift
+        tm.assert_index_equal(result, exp)
+
+    on_offset_cases = [
+        (datetime(2007, 12, 31), True),
+        (datetime(2007, 12, 15), True),
+        (datetime(2007, 12, 14), False),
+        (datetime(2007, 12, 1), False),
+        (datetime(2008, 2, 29), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        dt, expected = case
+        assert_is_on_offset(SemiMonthEnd(), dt, expected)
+
+    @pytest.mark.parametrize("klass", [Series, DatetimeIndex])
+    def test_vectorized_offset_addition(self, klass):
+        shift = klass(
+            [
+                Timestamp("2000-01-15 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-15", tz="US/Central"),
+            ],
+            name="a",
+        )
+
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = shift + SemiMonthEnd()
+            result2 = SemiMonthEnd() + shift
+
+        exp = klass(
+            [
+                Timestamp("2000-01-31 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-29", tz="US/Central"),
+            ],
+            name="a",
+        )
+        tm.assert_equal(result, exp)
+        tm.assert_equal(result2, exp)
+
+        shift = klass(
+            [
+                Timestamp("2000-01-01 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-01", tz="US/Central"),
+            ],
+            name="a",
+        )
+
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = shift + SemiMonthEnd()
+            result2 = SemiMonthEnd() + shift
+
+        exp = klass(
+            [
+                Timestamp("2000-01-15 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-15", tz="US/Central"),
+            ],
+            name="a",
+        )
+        tm.assert_equal(result, exp)
+        tm.assert_equal(result2, exp)
+
+
+class TestSemiMonthBegin:
+    def test_offset_whole_year(self):
+        dates = (
+            datetime(2007, 12, 15),
+            datetime(2008, 1, 1),
+            datetime(2008, 1, 15),
+            datetime(2008, 2, 1),
+            datetime(2008, 2, 15),
+            datetime(2008, 3, 1),
+            datetime(2008, 3, 15),
+            datetime(2008, 4, 1),
+            datetime(2008, 4, 15),
+            datetime(2008, 5, 1),
+            datetime(2008, 5, 15),
+            datetime(2008, 6, 1),
+            datetime(2008, 6, 15),
+            datetime(2008, 7, 1),
+            datetime(2008, 7, 15),
+            datetime(2008, 8, 1),
+            datetime(2008, 8, 15),
+            datetime(2008, 9, 1),
+            datetime(2008, 9, 15),
+            datetime(2008, 10, 1),
+            datetime(2008, 10, 15),
+            datetime(2008, 11, 1),
+            datetime(2008, 11, 15),
+            datetime(2008, 12, 1),
+            datetime(2008, 12, 15),
+        )
+
+        for base, exp_date in zip(dates[:-1], dates[1:]):
+            assert_offset_equal(SemiMonthBegin(), base, exp_date)
+
+        # ensure .apply_index works as expected
+        shift = DatetimeIndex(dates[:-1])
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = SemiMonthBegin() + shift
+
+        exp = DatetimeIndex(dates[1:])
+        tm.assert_index_equal(result, exp)
+
+    offset_cases = [
+        (
+            SemiMonthBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 15),
+                datetime(2008, 1, 15): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 14): datetime(2006, 12, 15),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2007, 1, 1): datetime(2007, 1, 15),
+                datetime(2006, 12, 1): datetime(2006, 12, 15),
+                datetime(2006, 12, 15): datetime(2007, 1, 1),
+            },
+        ),
+        (
+            SemiMonthBegin(day_of_month=20),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 20),
+                datetime(2008, 1, 15): datetime(2008, 1, 20),
+                datetime(2008, 1, 21): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 14): datetime(2006, 12, 20),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2007, 1, 1): datetime(2007, 1, 20),
+                datetime(2006, 12, 1): datetime(2006, 12, 20),
+                datetime(2006, 12, 15): datetime(2006, 12, 20),
+            },
+        ),
+        (
+            SemiMonthBegin(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 1, 16): datetime(2008, 2, 1),
+                datetime(2008, 1, 15): datetime(2008, 1, 15),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 2): datetime(2006, 12, 15),
+                datetime(2007, 1, 1): datetime(2007, 1, 1),
+            },
+        ),
+        (
+            SemiMonthBegin(0, day_of_month=16),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 1, 16): datetime(2008, 1, 16),
+                datetime(2008, 1, 15): datetime(2008, 1, 16),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2006, 12, 29): datetime(2007, 1, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2007, 1, 5): datetime(2007, 1, 16),
+                datetime(2007, 1, 1): datetime(2007, 1, 1),
+            },
+        ),
+        (
+            SemiMonthBegin(2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 15),
+                datetime(2006, 12, 1): datetime(2007, 1, 1),
+                datetime(2006, 12, 29): datetime(2007, 1, 15),
+                datetime(2006, 12, 15): datetime(2007, 1, 15),
+                datetime(2007, 1, 1): datetime(2007, 2, 1),
+                datetime(2007, 1, 16): datetime(2007, 2, 15),
+                datetime(2006, 11, 1): datetime(2006, 12, 1),
+            },
+        ),
+        (
+            SemiMonthBegin(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 15),
+                datetime(2008, 6, 30): datetime(2008, 6, 15),
+                datetime(2008, 6, 14): datetime(2008, 6, 1),
+                datetime(2008, 12, 31): datetime(2008, 12, 15),
+                datetime(2006, 12, 29): datetime(2006, 12, 15),
+                datetime(2006, 12, 15): datetime(2006, 12, 1),
+                datetime(2007, 1, 1): datetime(2006, 12, 15),
+            },
+        ),
+        (
+            SemiMonthBegin(-1, day_of_month=4),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 4),
+                datetime(2007, 1, 4): datetime(2007, 1, 1),
+                datetime(2008, 6, 30): datetime(2008, 6, 4),
+                datetime(2008, 12, 31): datetime(2008, 12, 4),
+                datetime(2006, 12, 5): datetime(2006, 12, 4),
+                datetime(2006, 12, 30): datetime(2006, 12, 4),
+                datetime(2006, 12, 2): datetime(2006, 12, 1),
+                datetime(2007, 1, 1): datetime(2006, 12, 4),
+            },
+        ),
+        (
+            SemiMonthBegin(-2),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 1),
+                datetime(2008, 6, 30): datetime(2008, 6, 1),
+                datetime(2008, 6, 14): datetime(2008, 5, 15),
+                datetime(2008, 12, 31): datetime(2008, 12, 1),
+                datetime(2006, 12, 29): datetime(2006, 12, 1),
+                datetime(2006, 12, 15): datetime(2006, 11, 15),
+                datetime(2007, 1, 1): datetime(2006, 12, 1),
+            },
+        ),
+    ]
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_apply_index(self, case):
+        offset, cases = case
+        shift = DatetimeIndex(cases.keys())
+
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = offset + shift
+
+        exp = DatetimeIndex(cases.values())
+        tm.assert_index_equal(result, exp)
+
+    on_offset_cases = [
+        (datetime(2007, 12, 1), True),
+        (datetime(2007, 12, 15), True),
+        (datetime(2007, 12, 14), False),
+        (datetime(2007, 12, 31), False),
+        (datetime(2008, 2, 15), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        dt, expected = case
+        assert_is_on_offset(SemiMonthBegin(), dt, expected)
+
+    @pytest.mark.parametrize("klass", [Series, DatetimeIndex])
+    def test_vectorized_offset_addition(self, klass):
+        shift = klass(
+            [
+                Timestamp("2000-01-15 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-15", tz="US/Central"),
+            ],
+            name="a",
+        )
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = shift + SemiMonthBegin()
+            result2 = SemiMonthBegin() + shift
+
+        exp = klass(
+            [
+                Timestamp("2000-02-01 00:15:00", tz="US/Central"),
+                Timestamp("2000-03-01", tz="US/Central"),
+            ],
+            name="a",
+        )
+        tm.assert_equal(result, exp)
+        tm.assert_equal(result2, exp)
+
+        shift = klass(
+            [
+                Timestamp("2000-01-01 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-01", tz="US/Central"),
+            ],
+            name="a",
+        )
+        with tm.assert_produces_warning(None):
+            # GH#22535 check that we don't get a FutureWarning from adding
+            # an integer array to PeriodIndex
+            result = shift + SemiMonthBegin()
+            result2 = SemiMonthBegin() + shift
+
+        exp = klass(
+            [
+                Timestamp("2000-01-15 00:15:00", tz="US/Central"),
+                Timestamp("2000-02-15", tz="US/Central"),
+            ],
+            name="a",
+        )
+        tm.assert_equal(result, exp)
+        tm.assert_equal(result2, exp)
+
+
+class TestMonthBegin:
+    offset_cases = []
+    # NOTE: I'm not entirely happy with the logic here for Begin -ss
+    # see thread 'offset conventions' on the ML
+    offset_cases.append(
+        (
+            MonthBegin(),
+            {
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2008, 2, 1): datetime(2008, 3, 1),
+                datetime(2006, 12, 31): datetime(2007, 1, 1),
+                datetime(2006, 12, 1): datetime(2007, 1, 1),
+                datetime(2007, 1, 31): datetime(2007, 2, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthBegin(0),
+            {
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2006, 12, 3): datetime(2007, 1, 1),
+                datetime(2007, 1, 31): datetime(2007, 2, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthBegin(2),
+            {
+                datetime(2008, 2, 29): datetime(2008, 4, 1),
+                datetime(2008, 1, 31): datetime(2008, 3, 1),
+                datetime(2006, 12, 31): datetime(2007, 2, 1),
+                datetime(2007, 12, 28): datetime(2008, 2, 1),
+                datetime(2007, 1, 1): datetime(2007, 3, 1),
+                datetime(2006, 11, 1): datetime(2007, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthBegin(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 1),
+                datetime(2008, 5, 31): datetime(2008, 5, 1),
+                datetime(2008, 12, 31): datetime(2008, 12, 1),
+                datetime(2006, 12, 29): datetime(2006, 12, 1),
+                datetime(2006, 1, 2): datetime(2006, 1, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+
+class TestMonthEnd:
+    def test_day_of_month(self):
+        dt = datetime(2007, 1, 1)
+        offset = MonthEnd()
+
+        result = dt + offset
+        assert result == Timestamp(2007, 1, 31)
+
+        result = result + offset
+        assert result == Timestamp(2007, 2, 28)
+
+    def test_normalize(self):
+        dt = datetime(2007, 1, 1, 3)
+
+        result = dt + MonthEnd(normalize=True)
+        expected = dt.replace(hour=0) + MonthEnd()
+        assert result == expected
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            MonthEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 2, 29),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2007, 1, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 31),
+                datetime(2006, 12, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2006, 12, 29): datetime(2006, 12, 31),
+                datetime(2006, 12, 31): datetime(2006, 12, 31),
+                datetime(2007, 1, 1): datetime(2007, 1, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthEnd(2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 29),
+                datetime(2008, 1, 31): datetime(2008, 3, 31),
+                datetime(2006, 12, 29): datetime(2007, 1, 31),
+                datetime(2006, 12, 31): datetime(2007, 2, 28),
+                datetime(2007, 1, 1): datetime(2007, 2, 28),
+                datetime(2006, 11, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            MonthEnd(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 5, 31),
+                datetime(2008, 12, 31): datetime(2008, 11, 30),
+                datetime(2006, 12, 29): datetime(2006, 11, 30),
+                datetime(2006, 12, 30): datetime(2006, 11, 30),
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (MonthEnd(), datetime(2007, 12, 31), True),
+        (MonthEnd(), datetime(2008, 1, 1), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_offsets.py b/pandas/pandas/tests/tseries/offsets/test_offsets.py
new file mode 100644
index 0000000000000..7480b99595066
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_offsets.py
@@ -0,0 +1,1191 @@
+"""
+Tests of pandas.tseries.offsets
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    NaT,
+    Timedelta,
+    Timestamp,
+    conversion,
+    timezones,
+)
+import pandas._libs.tslibs.offsets as liboffsets
+from pandas._libs.tslibs.offsets import (
+    _get_offset,
+    _offset_map,
+    to_offset,
+)
+from pandas._libs.tslibs.period import INVALID_FREQ_ERR_MSG
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.tests.tseries.offsets.common import WeekDay
+
+from pandas.tseries import offsets
+from pandas.tseries.offsets import (
+    FY5253,
+    BDay,
+    BMonthEnd,
+    BusinessHour,
+    CustomBusinessDay,
+    CustomBusinessHour,
+    CustomBusinessMonthBegin,
+    CustomBusinessMonthEnd,
+    DateOffset,
+    Easter,
+    FY5253Quarter,
+    LastWeekOfMonth,
+    MonthBegin,
+    Nano,
+    Tick,
+    Week,
+    WeekOfMonth,
+)
+
+_ARITHMETIC_DATE_OFFSET = [
+    "years",
+    "months",
+    "weeks",
+    "days",
+    "hours",
+    "minutes",
+    "seconds",
+    "milliseconds",
+    "microseconds",
+]
+
+
+def _create_offset(klass, value=1, normalize=False):
+    # create instance from offset class
+    if klass is FY5253:
+        klass = klass(
+            n=value,
+            startingMonth=1,
+            weekday=1,
+            variation="last",
+            normalize=normalize,
+        )
+    elif klass is FY5253Quarter:
+        klass = klass(
+            n=value,
+            startingMonth=1,
+            weekday=1,
+            qtr_with_extra_week=1,
+            variation="last",
+            normalize=normalize,
+        )
+    elif klass is LastWeekOfMonth:
+        klass = klass(n=value, weekday=5, normalize=normalize)
+    elif klass is WeekOfMonth:
+        klass = klass(n=value, week=1, weekday=5, normalize=normalize)
+    elif klass is Week:
+        klass = klass(n=value, weekday=5, normalize=normalize)
+    elif klass is DateOffset:
+        klass = klass(days=value, normalize=normalize)
+    else:
+        klass = klass(value, normalize=normalize)
+    return klass
+
+
+@pytest.fixture(
+    params=[
+        getattr(offsets, o)
+        for o in offsets.__all__
+        if issubclass(getattr(offsets, o), liboffsets.MonthOffset)
+        and o != "MonthOffset"
+    ]
+)
+def month_classes(request):
+    """
+    Fixture for month based datetime offsets available for a time series.
+    """
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        getattr(offsets, o) for o in offsets.__all__ if o not in ("Tick", "BaseOffset")
+    ]
+)
+def offset_types(request):
+    """
+    Fixture for all the datetime offsets available for a time series.
+    """
+    return request.param
+
+
+@pytest.fixture
+def dt():
+    return Timestamp(datetime(2008, 1, 2))
+
+
+@pytest.fixture
+def expecteds():
+    # executed value created by _create_offset
+    # are applied to 2011/01/01 09:00 (Saturday)
+    # used for .apply and .rollforward
+    return {
+        "Day": Timestamp("2011-01-02 09:00:00"),
+        "DateOffset": Timestamp("2011-01-02 09:00:00"),
+        "BusinessDay": Timestamp("2011-01-03 09:00:00"),
+        "CustomBusinessDay": Timestamp("2011-01-03 09:00:00"),
+        "CustomBusinessMonthEnd": Timestamp("2011-01-31 09:00:00"),
+        "CustomBusinessMonthBegin": Timestamp("2011-01-03 09:00:00"),
+        "MonthBegin": Timestamp("2011-02-01 09:00:00"),
+        "BusinessMonthBegin": Timestamp("2011-01-03 09:00:00"),
+        "MonthEnd": Timestamp("2011-01-31 09:00:00"),
+        "SemiMonthEnd": Timestamp("2011-01-15 09:00:00"),
+        "SemiMonthBegin": Timestamp("2011-01-15 09:00:00"),
+        "BusinessMonthEnd": Timestamp("2011-01-31 09:00:00"),
+        "YearBegin": Timestamp("2012-01-01 09:00:00"),
+        "BYearBegin": Timestamp("2011-01-03 09:00:00"),
+        "YearEnd": Timestamp("2011-12-31 09:00:00"),
+        "BYearEnd": Timestamp("2011-12-30 09:00:00"),
+        "QuarterBegin": Timestamp("2011-03-01 09:00:00"),
+        "BQuarterBegin": Timestamp("2011-03-01 09:00:00"),
+        "QuarterEnd": Timestamp("2011-03-31 09:00:00"),
+        "BQuarterEnd": Timestamp("2011-03-31 09:00:00"),
+        "BusinessHour": Timestamp("2011-01-03 10:00:00"),
+        "CustomBusinessHour": Timestamp("2011-01-03 10:00:00"),
+        "WeekOfMonth": Timestamp("2011-01-08 09:00:00"),
+        "LastWeekOfMonth": Timestamp("2011-01-29 09:00:00"),
+        "FY5253Quarter": Timestamp("2011-01-25 09:00:00"),
+        "FY5253": Timestamp("2011-01-25 09:00:00"),
+        "Week": Timestamp("2011-01-08 09:00:00"),
+        "Easter": Timestamp("2011-04-24 09:00:00"),
+        "Hour": Timestamp("2011-01-01 10:00:00"),
+        "Minute": Timestamp("2011-01-01 09:01:00"),
+        "Second": Timestamp("2011-01-01 09:00:01"),
+        "Milli": Timestamp("2011-01-01 09:00:00.001000"),
+        "Micro": Timestamp("2011-01-01 09:00:00.000001"),
+        "Nano": Timestamp("2011-01-01T09:00:00.000000001"),
+    }
+
+
+class TestCommon:
+    def test_immutable(self, offset_types):
+        # GH#21341 check that __setattr__ raises
+        offset = _create_offset(offset_types)
+        msg = "objects is not writable|DateOffset objects are immutable"
+        with pytest.raises(AttributeError, match=msg):
+            offset.normalize = True
+        with pytest.raises(AttributeError, match=msg):
+            offset.n = 91
+
+    def test_return_type(self, offset_types):
+        offset = _create_offset(offset_types)
+
+        # make sure that we are returning a Timestamp
+        result = Timestamp("20080101") + offset
+        assert isinstance(result, Timestamp)
+
+        # make sure that we are returning NaT
+        assert NaT + offset is NaT
+        assert offset + NaT is NaT
+
+        assert NaT - offset is NaT
+        assert (-offset)._apply(NaT) is NaT
+
+    def test_offset_n(self, offset_types):
+        offset = _create_offset(offset_types)
+        assert offset.n == 1
+
+        neg_offset = offset * -1
+        assert neg_offset.n == -1
+
+        mul_offset = offset * 3
+        assert mul_offset.n == 3
+
+    def test_offset_timedelta64_arg(self, offset_types):
+        # check that offset._validate_n raises TypeError on a timedelt64
+        #  object
+        off = _create_offset(offset_types)
+
+        td64 = np.timedelta64(4567, "s")
+        with pytest.raises(TypeError, match="argument must be an integer"):
+            type(off)(n=td64, **off.kwds)
+
+    def test_offset_mul_ndarray(self, offset_types):
+        off = _create_offset(offset_types)
+
+        expected = np.array([[off, off * 2], [off * 3, off * 4]])
+
+        result = np.array([[1, 2], [3, 4]]) * off
+        tm.assert_numpy_array_equal(result, expected)
+
+        result = off * np.array([[1, 2], [3, 4]])
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_offset_freqstr(self, offset_types):
+        offset = _create_offset(offset_types)
+
+        freqstr = offset.freqstr
+        if freqstr not in ("<Easter>", "<DateOffset: days=1>", "LWOM-SAT"):
+            code = _get_offset(freqstr)
+            assert offset.rule_code == code
+
+    def _check_offsetfunc_works(self, offset, funcname, dt, expected, normalize=False):
+        if normalize and issubclass(offset, Tick):
+            # normalize=True disallowed for Tick subclasses GH#21427
+            return
+
+        offset_s = _create_offset(offset, normalize=normalize)
+        func = getattr(offset_s, funcname)
+
+        result = func(dt)
+        assert isinstance(result, Timestamp)
+        assert result == expected
+
+        result = func(Timestamp(dt))
+        assert isinstance(result, Timestamp)
+        assert result == expected
+
+        # see gh-14101
+        ts = Timestamp(dt) + Nano(5)
+        # test nanosecond is preserved
+        with tm.assert_produces_warning(None):
+            result = func(ts)
+
+        assert isinstance(result, Timestamp)
+        if normalize is False:
+            assert result == expected + Nano(5)
+        else:
+            assert result == expected
+
+        if isinstance(dt, np.datetime64):
+            # test tz when input is datetime or Timestamp
+            return
+
+        for tz in [
+            None,
+            "UTC",
+            "Asia/Tokyo",
+            "US/Eastern",
+            "dateutil/Asia/Tokyo",
+            "dateutil/US/Pacific",
+        ]:
+            expected_localize = expected.tz_localize(tz)
+            tz_obj = timezones.maybe_get_tz(tz)
+            dt_tz = conversion.localize_pydatetime(dt, tz_obj)
+
+            result = func(dt_tz)
+            assert isinstance(result, Timestamp)
+            assert result == expected_localize
+
+            result = func(Timestamp(dt, tz=tz))
+            assert isinstance(result, Timestamp)
+            assert result == expected_localize
+
+            # see gh-14101
+            ts = Timestamp(dt, tz=tz) + Nano(5)
+            # test nanosecond is preserved
+            with tm.assert_produces_warning(None):
+                result = func(ts)
+            assert isinstance(result, Timestamp)
+            if normalize is False:
+                assert result == expected_localize + Nano(5)
+            else:
+                assert result == expected_localize
+
+    def test_apply(self, offset_types, expecteds):
+        sdt = datetime(2011, 1, 1, 9, 0)
+        ndt = np.datetime64("2011-01-01 09:00")
+
+        expected = expecteds[offset_types.__name__]
+        expected_norm = Timestamp(expected.date())
+
+        for dt in [sdt, ndt]:
+            self._check_offsetfunc_works(offset_types, "_apply", dt, expected)
+
+            self._check_offsetfunc_works(
+                offset_types, "_apply", dt, expected_norm, normalize=True
+            )
+
+    def test_rollforward(self, offset_types, expecteds):
+        expecteds = expecteds.copy()
+
+        # result will not be changed if the target is on the offset
+        no_changes = [
+            "Day",
+            "MonthBegin",
+            "SemiMonthBegin",
+            "YearBegin",
+            "Week",
+            "Hour",
+            "Minute",
+            "Second",
+            "Milli",
+            "Micro",
+            "Nano",
+            "DateOffset",
+        ]
+        for n in no_changes:
+            expecteds[n] = Timestamp("2011/01/01 09:00")
+
+        expecteds["BusinessHour"] = Timestamp("2011-01-03 09:00:00")
+        expecteds["CustomBusinessHour"] = Timestamp("2011-01-03 09:00:00")
+
+        # but be changed when normalize=True
+        norm_expected = expecteds.copy()
+        for k in norm_expected:
+            norm_expected[k] = Timestamp(norm_expected[k].date())
+
+        normalized = {
+            "Day": Timestamp("2011-01-02 00:00:00"),
+            "DateOffset": Timestamp("2011-01-02 00:00:00"),
+            "MonthBegin": Timestamp("2011-02-01 00:00:00"),
+            "SemiMonthBegin": Timestamp("2011-01-15 00:00:00"),
+            "YearBegin": Timestamp("2012-01-01 00:00:00"),
+            "Week": Timestamp("2011-01-08 00:00:00"),
+            "Hour": Timestamp("2011-01-01 00:00:00"),
+            "Minute": Timestamp("2011-01-01 00:00:00"),
+            "Second": Timestamp("2011-01-01 00:00:00"),
+            "Milli": Timestamp("2011-01-01 00:00:00"),
+            "Micro": Timestamp("2011-01-01 00:00:00"),
+        }
+        norm_expected.update(normalized)
+
+        sdt = datetime(2011, 1, 1, 9, 0)
+        ndt = np.datetime64("2011-01-01 09:00")
+
+        for dt in [sdt, ndt]:
+            expected = expecteds[offset_types.__name__]
+            self._check_offsetfunc_works(offset_types, "rollforward", dt, expected)
+            expected = norm_expected[offset_types.__name__]
+            self._check_offsetfunc_works(
+                offset_types, "rollforward", dt, expected, normalize=True
+            )
+
+    def test_rollback(self, offset_types):
+        expecteds = {
+            "BusinessDay": Timestamp("2010-12-31 09:00:00"),
+            "CustomBusinessDay": Timestamp("2010-12-31 09:00:00"),
+            "CustomBusinessMonthEnd": Timestamp("2010-12-31 09:00:00"),
+            "CustomBusinessMonthBegin": Timestamp("2010-12-01 09:00:00"),
+            "BusinessMonthBegin": Timestamp("2010-12-01 09:00:00"),
+            "MonthEnd": Timestamp("2010-12-31 09:00:00"),
+            "SemiMonthEnd": Timestamp("2010-12-31 09:00:00"),
+            "BusinessMonthEnd": Timestamp("2010-12-31 09:00:00"),
+            "BYearBegin": Timestamp("2010-01-01 09:00:00"),
+            "YearEnd": Timestamp("2010-12-31 09:00:00"),
+            "BYearEnd": Timestamp("2010-12-31 09:00:00"),
+            "QuarterBegin": Timestamp("2010-12-01 09:00:00"),
+            "BQuarterBegin": Timestamp("2010-12-01 09:00:00"),
+            "QuarterEnd": Timestamp("2010-12-31 09:00:00"),
+            "BQuarterEnd": Timestamp("2010-12-31 09:00:00"),
+            "BusinessHour": Timestamp("2010-12-31 17:00:00"),
+            "CustomBusinessHour": Timestamp("2010-12-31 17:00:00"),
+            "WeekOfMonth": Timestamp("2010-12-11 09:00:00"),
+            "LastWeekOfMonth": Timestamp("2010-12-25 09:00:00"),
+            "FY5253Quarter": Timestamp("2010-10-26 09:00:00"),
+            "FY5253": Timestamp("2010-01-26 09:00:00"),
+            "Easter": Timestamp("2010-04-04 09:00:00"),
+        }
+
+        # result will not be changed if the target is on the offset
+        for n in [
+            "Day",
+            "MonthBegin",
+            "SemiMonthBegin",
+            "YearBegin",
+            "Week",
+            "Hour",
+            "Minute",
+            "Second",
+            "Milli",
+            "Micro",
+            "Nano",
+            "DateOffset",
+        ]:
+            expecteds[n] = Timestamp("2011/01/01 09:00")
+
+        # but be changed when normalize=True
+        norm_expected = expecteds.copy()
+        for k in norm_expected:
+            norm_expected[k] = Timestamp(norm_expected[k].date())
+
+        normalized = {
+            "Day": Timestamp("2010-12-31 00:00:00"),
+            "DateOffset": Timestamp("2010-12-31 00:00:00"),
+            "MonthBegin": Timestamp("2010-12-01 00:00:00"),
+            "SemiMonthBegin": Timestamp("2010-12-15 00:00:00"),
+            "YearBegin": Timestamp("2010-01-01 00:00:00"),
+            "Week": Timestamp("2010-12-25 00:00:00"),
+            "Hour": Timestamp("2011-01-01 00:00:00"),
+            "Minute": Timestamp("2011-01-01 00:00:00"),
+            "Second": Timestamp("2011-01-01 00:00:00"),
+            "Milli": Timestamp("2011-01-01 00:00:00"),
+            "Micro": Timestamp("2011-01-01 00:00:00"),
+        }
+        norm_expected.update(normalized)
+
+        sdt = datetime(2011, 1, 1, 9, 0)
+        ndt = np.datetime64("2011-01-01 09:00")
+
+        for dt in [sdt, ndt]:
+            expected = expecteds[offset_types.__name__]
+            self._check_offsetfunc_works(offset_types, "rollback", dt, expected)
+
+            expected = norm_expected[offset_types.__name__]
+            self._check_offsetfunc_works(
+                offset_types, "rollback", dt, expected, normalize=True
+            )
+
+    def test_is_on_offset(self, offset_types, expecteds):
+        dt = expecteds[offset_types.__name__]
+        offset_s = _create_offset(offset_types)
+        assert offset_s.is_on_offset(dt)
+
+        # when normalize=True, is_on_offset checks time is 00:00:00
+        if issubclass(offset_types, Tick):
+            # normalize=True disallowed for Tick subclasses GH#21427
+            return
+        offset_n = _create_offset(offset_types, normalize=True)
+        assert not offset_n.is_on_offset(dt)
+
+        if offset_types in (BusinessHour, CustomBusinessHour):
+            # In default BusinessHour (9:00-17:00), normalized time
+            # cannot be in business hour range
+            return
+        date = datetime(dt.year, dt.month, dt.day)
+        assert offset_n.is_on_offset(date)
+
+    def test_add(self, offset_types, tz_naive_fixture, expecteds):
+        tz = tz_naive_fixture
+        dt = datetime(2011, 1, 1, 9, 0)
+
+        offset_s = _create_offset(offset_types)
+        expected = expecteds[offset_types.__name__]
+
+        result_dt = dt + offset_s
+        result_ts = Timestamp(dt) + offset_s
+        for result in [result_dt, result_ts]:
+            assert isinstance(result, Timestamp)
+            assert result == expected
+
+        expected_localize = expected.tz_localize(tz)
+        result = Timestamp(dt, tz=tz) + offset_s
+        assert isinstance(result, Timestamp)
+        assert result == expected_localize
+
+        # normalize=True, disallowed for Tick subclasses GH#21427
+        if issubclass(offset_types, Tick):
+            return
+        offset_s = _create_offset(offset_types, normalize=True)
+        expected = Timestamp(expected.date())
+
+        result_dt = dt + offset_s
+        result_ts = Timestamp(dt) + offset_s
+        for result in [result_dt, result_ts]:
+            assert isinstance(result, Timestamp)
+            assert result == expected
+
+        expected_localize = expected.tz_localize(tz)
+        result = Timestamp(dt, tz=tz) + offset_s
+        assert isinstance(result, Timestamp)
+        assert result == expected_localize
+
+    def test_add_empty_datetimeindex(
+        self, performance_warning, offset_types, tz_naive_fixture
+    ):
+        # GH#12724, GH#30336
+        offset_s = _create_offset(offset_types)
+
+        dti = DatetimeIndex([], tz=tz_naive_fixture).as_unit("ns")
+
+        if not isinstance(
+            offset_s,
+            (
+                Easter,
+                WeekOfMonth,
+                LastWeekOfMonth,
+                CustomBusinessDay,
+                BusinessHour,
+                CustomBusinessHour,
+                CustomBusinessMonthBegin,
+                CustomBusinessMonthEnd,
+                FY5253,
+                FY5253Quarter,
+            ),
+        ):
+            # We don't have an optimized apply_index
+            performance_warning = False
+
+        # stacklevel checking is slow, and we have ~800 of variants of this
+        #  test, so let's only check the stacklevel in a subset of them
+        check_stacklevel = tz_naive_fixture is None
+        with tm.assert_produces_warning(
+            performance_warning, check_stacklevel=check_stacklevel
+        ):
+            result = dti + offset_s
+        tm.assert_index_equal(result, dti)
+        with tm.assert_produces_warning(
+            performance_warning, check_stacklevel=check_stacklevel
+        ):
+            result = offset_s + dti
+        tm.assert_index_equal(result, dti)
+
+        dta = dti._data
+        with tm.assert_produces_warning(
+            performance_warning, check_stacklevel=check_stacklevel
+        ):
+            result = dta + offset_s
+        tm.assert_equal(result, dta)
+        with tm.assert_produces_warning(
+            performance_warning, check_stacklevel=check_stacklevel
+        ):
+            result = offset_s + dta
+        tm.assert_equal(result, dta)
+
+    def test_pickle_roundtrip(self, offset_types):
+        off = _create_offset(offset_types)
+        res = tm.round_trip_pickle(off)
+        assert off == res
+        if type(off) is not DateOffset:
+            for attr in off._attributes:
+                if attr == "calendar":
+                    # np.busdaycalendar __eq__ will return False;
+                    #  we check holidays and weekmask attrs so are OK
+                    continue
+                # Make sure nothings got lost from _params (which __eq__) is based on
+                assert getattr(off, attr) == getattr(res, attr)
+
+    def test_pickle_dateoffset_odd_inputs(self):
+        # GH#34511
+        off = DateOffset(months=12)
+        res = tm.round_trip_pickle(off)
+        assert off == res
+
+        base_dt = datetime(2020, 1, 1)
+        assert base_dt + off == base_dt + res
+
+    def test_offsets_hashable(self, offset_types):
+        # GH: 37267
+        off = _create_offset(offset_types)
+        assert hash(off) is not None
+
+    # TODO: belongs in arithmetic tests?
+    @pytest.mark.filterwarnings(
+        "ignore:Non-vectorized DateOffset being applied to Series or DatetimeIndex"
+    )
+    @pytest.mark.parametrize("unit", ["s", "ms", "us"])
+    def test_add_dt64_ndarray_non_nano(self, offset_types, unit):
+        # check that the result with non-nano matches nano
+        off = _create_offset(offset_types)
+
+        dti = date_range("2016-01-01", periods=35, freq="D", unit=unit)
+
+        result = (dti + off)._with_freq(None)
+
+        exp_unit = unit
+        if isinstance(off, Tick) and off._creso > dti._data._creso:
+            # cast to higher reso like we would with Timedelta scalar
+            exp_unit = Timedelta(off).unit
+        # TODO(GH#55564): as_unit will be unnecessary
+        expected = DatetimeIndex([x + off for x in dti]).as_unit(exp_unit)
+
+        tm.assert_index_equal(result, expected)
+
+
+class TestDateOffset:
+    def setup_method(self):
+        _offset_map.clear()
+
+    def test_repr(self):
+        repr(DateOffset())
+        repr(DateOffset(2))
+        repr(2 * DateOffset())
+        repr(2 * DateOffset(months=2))
+
+    def test_mul(self):
+        assert DateOffset(2) == 2 * DateOffset(1)
+        assert DateOffset(2) == DateOffset(1) * 2
+
+    @pytest.mark.parametrize("kwd", sorted(liboffsets._relativedelta_kwds))
+    def test_constructor(self, kwd, request):
+        if kwd == "millisecond":
+            request.applymarker(
+                pytest.mark.xfail(
+                    raises=NotImplementedError,
+                    reason="Constructing DateOffset object with `millisecond` is not "
+                    "yet supported.",
+                )
+            )
+        offset = DateOffset(**{kwd: 2})
+        assert offset.kwds == {kwd: 2}
+        assert getattr(offset, kwd) == 2
+
+    def test_default_constructor(self, dt):
+        assert (dt + DateOffset(2)) == datetime(2008, 1, 4)
+
+    def test_copy(self):
+        assert DateOffset(months=2).copy() == DateOffset(months=2)
+        assert DateOffset(milliseconds=1).copy() == DateOffset(milliseconds=1)
+
+    @pytest.mark.parametrize(
+        "arithmatic_offset_type, expected",
+        zip(
+            _ARITHMETIC_DATE_OFFSET,
+            [
+                "2009-01-02",
+                "2008-02-02",
+                "2008-01-09",
+                "2008-01-03",
+                "2008-01-02 01:00:00",
+                "2008-01-02 00:01:00",
+                "2008-01-02 00:00:01",
+                "2008-01-02 00:00:00.001000000",
+                "2008-01-02 00:00:00.000001000",
+            ],
+        ),
+    )
+    def test_add(self, arithmatic_offset_type, expected, dt):
+        assert DateOffset(**{arithmatic_offset_type: 1}) + dt == Timestamp(expected)
+        assert dt + DateOffset(**{arithmatic_offset_type: 1}) == Timestamp(expected)
+
+    @pytest.mark.parametrize(
+        "arithmatic_offset_type, expected",
+        zip(
+            _ARITHMETIC_DATE_OFFSET,
+            [
+                "2007-01-02",
+                "2007-12-02",
+                "2007-12-26",
+                "2008-01-01",
+                "2008-01-01 23:00:00",
+                "2008-01-01 23:59:00",
+                "2008-01-01 23:59:59",
+                "2008-01-01 23:59:59.999000000",
+                "2008-01-01 23:59:59.999999000",
+            ],
+        ),
+    )
+    def test_sub(self, arithmatic_offset_type, expected, dt):
+        assert dt - DateOffset(**{arithmatic_offset_type: 1}) == Timestamp(expected)
+        with pytest.raises(TypeError, match="Cannot subtract datetime from offset"):
+            DateOffset(**{arithmatic_offset_type: 1}) - dt
+
+    @pytest.mark.parametrize(
+        "arithmatic_offset_type, n, expected",
+        zip(
+            _ARITHMETIC_DATE_OFFSET,
+            range(1, 10),
+            [
+                "2009-01-02",
+                "2008-03-02",
+                "2008-01-23",
+                "2008-01-06",
+                "2008-01-02 05:00:00",
+                "2008-01-02 00:06:00",
+                "2008-01-02 00:00:07",
+                "2008-01-02 00:00:00.008000000",
+                "2008-01-02 00:00:00.000009000",
+            ],
+        ),
+    )
+    def test_mul_add(self, arithmatic_offset_type, n, expected, dt):
+        assert DateOffset(**{arithmatic_offset_type: 1}) * n + dt == Timestamp(expected)
+        assert n * DateOffset(**{arithmatic_offset_type: 1}) + dt == Timestamp(expected)
+        assert dt + DateOffset(**{arithmatic_offset_type: 1}) * n == Timestamp(expected)
+        assert dt + n * DateOffset(**{arithmatic_offset_type: 1}) == Timestamp(expected)
+
+    @pytest.mark.parametrize(
+        "arithmatic_offset_type, n, expected",
+        zip(
+            _ARITHMETIC_DATE_OFFSET,
+            range(1, 10),
+            [
+                "2007-01-02",
+                "2007-11-02",
+                "2007-12-12",
+                "2007-12-29",
+                "2008-01-01 19:00:00",
+                "2008-01-01 23:54:00",
+                "2008-01-01 23:59:53",
+                "2008-01-01 23:59:59.992000000",
+                "2008-01-01 23:59:59.999991000",
+            ],
+        ),
+    )
+    def test_mul_sub(self, arithmatic_offset_type, n, expected, dt):
+        assert dt - DateOffset(**{arithmatic_offset_type: 1}) * n == Timestamp(expected)
+        assert dt - n * DateOffset(**{arithmatic_offset_type: 1}) == Timestamp(expected)
+
+    def test_leap_year(self):
+        d = datetime(2008, 1, 31)
+        assert (d + DateOffset(months=1)) == datetime(2008, 2, 29)
+
+    def test_eq(self):
+        offset1 = DateOffset(days=1)
+        offset2 = DateOffset(days=365)
+
+        assert offset1 != offset2
+
+        assert DateOffset(milliseconds=3) != DateOffset(milliseconds=7)
+
+    @pytest.mark.parametrize(
+        "offset_kwargs, expected_arg",
+        [
+            ({"microseconds": 1, "milliseconds": 1}, "2022-01-01 00:00:00.001001"),
+            ({"seconds": 1, "milliseconds": 1}, "2022-01-01 00:00:01.001"),
+            ({"minutes": 1, "milliseconds": 1}, "2022-01-01 00:01:00.001"),
+            ({"hours": 1, "milliseconds": 1}, "2022-01-01 01:00:00.001"),
+            ({"days": 1, "milliseconds": 1}, "2022-01-02 00:00:00.001"),
+            ({"weeks": 1, "milliseconds": 1}, "2022-01-08 00:00:00.001"),
+            ({"months": 1, "milliseconds": 1}, "2022-02-01 00:00:00.001"),
+            ({"years": 1, "milliseconds": 1}, "2023-01-01 00:00:00.001"),
+        ],
+    )
+    def test_milliseconds_combination(self, offset_kwargs, expected_arg):
+        # GH 49897
+        offset = DateOffset(**offset_kwargs)
+        ts = Timestamp("2022-01-01")
+        result = ts + offset
+        expected = Timestamp(expected_arg)
+
+        assert result == expected
+
+    def test_offset_invalid_arguments(self):
+        msg = "^Invalid argument/s or bad combination of arguments"
+        with pytest.raises(ValueError, match=msg):
+            DateOffset(picoseconds=1)
+
+
+class TestOffsetNames:
+    def test_get_offset_name(self):
+        assert BDay().freqstr == "B"
+        assert BDay(2).freqstr == "2B"
+        assert BMonthEnd().freqstr == "BME"
+        assert Week(weekday=0).freqstr == "W-MON"
+        assert Week(weekday=1).freqstr == "W-TUE"
+        assert Week(weekday=2).freqstr == "W-WED"
+        assert Week(weekday=3).freqstr == "W-THU"
+        assert Week(weekday=4).freqstr == "W-FRI"
+
+        assert LastWeekOfMonth(weekday=WeekDay.SUN).freqstr == "LWOM-SUN"
+
+
+def test_get_offset():
+    with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+        _get_offset("gibberish")
+    with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+        _get_offset("QS-JAN-B")
+
+    pairs = [
+        ("B", BDay()),
+        ("BME", BMonthEnd()),
+        ("W-MON", Week(weekday=0)),
+        ("W-TUE", Week(weekday=1)),
+        ("W-WED", Week(weekday=2)),
+        ("W-THU", Week(weekday=3)),
+        ("W-FRI", Week(weekday=4)),
+    ]
+
+    for name, expected in pairs:
+        offset = _get_offset(name)
+        assert offset == expected, (
+            f"Expected {name!r} to yield {expected!r} (actual: {offset!r})"
+        )
+
+
+def test_get_offset_legacy():
+    pairs = [("w@Sat", Week(weekday=5))]
+    for name, expected in pairs:
+        with pytest.raises(ValueError, match=INVALID_FREQ_ERR_MSG):
+            _get_offset(name)
+
+
+class TestOffsetAliases:
+    def setup_method(self):
+        _offset_map.clear()
+
+    def test_alias_equality(self):
+        for k, v in _offset_map.items():
+            if v is None:
+                continue
+            assert k == v.copy()
+
+    def test_rule_code(self):
+        lst = ["ME", "MS", "BME", "BMS", "D", "B", "h", "min", "s", "ms", "us"]
+        for k in lst:
+            assert k == _get_offset(k).rule_code
+            # should be cached - this is kind of an internals test...
+            assert k in _offset_map
+            assert k == (_get_offset(k) * 3).rule_code
+
+        suffix_lst = ["MON", "TUE", "WED", "THU", "FRI", "SAT", "SUN"]
+        base = "W"
+        for v in suffix_lst:
+            alias = "-".join([base, v])
+            assert alias == _get_offset(alias).rule_code
+            assert alias == (_get_offset(alias) * 5).rule_code
+
+        suffix_lst = [
+            "JAN",
+            "FEB",
+            "MAR",
+            "APR",
+            "MAY",
+            "JUN",
+            "JUL",
+            "AUG",
+            "SEP",
+            "OCT",
+            "NOV",
+            "DEC",
+        ]
+        base_lst = ["YE", "YS", "BYE", "BYS", "QE", "QS", "BQE", "BQS"]
+        for base in base_lst:
+            for v in suffix_lst:
+                alias = "-".join([base, v])
+                assert alias == _get_offset(alias).rule_code
+                assert alias == (_get_offset(alias) * 5).rule_code
+
+
+def test_freq_offsets():
+    off = BDay(1, offset=timedelta(0, 1800))
+    assert off.freqstr == "B+30Min"
+
+    off = BDay(1, offset=timedelta(0, -1800))
+    assert off.freqstr == "B-30Min"
+
+
+class TestReprNames:
+    def test_str_for_named_is_name(self):
+        # look at all the amazing combinations!
+        month_prefixes = ["YE", "YS", "BYE", "BYS", "QE", "BQE", "BQS", "QS"]
+        names = [
+            prefix + "-" + month
+            for prefix in month_prefixes
+            for month in [
+                "JAN",
+                "FEB",
+                "MAR",
+                "APR",
+                "MAY",
+                "JUN",
+                "JUL",
+                "AUG",
+                "SEP",
+                "OCT",
+                "NOV",
+                "DEC",
+            ]
+        ]
+        days = ["MON", "TUE", "WED", "THU", "FRI", "SAT", "SUN"]
+        names += ["W-" + day for day in days]
+        names += ["WOM-" + week + day for week in ("1", "2", "3", "4") for day in days]
+        _offset_map.clear()
+        for name in names:
+            offset = _get_offset(name)
+            assert offset.freqstr == name
+
+
+# ---------------------------------------------------------------------
+
+
+def test_valid_default_arguments(offset_types):
+    # GH#19142 check that the calling the constructors without passing
+    # any keyword arguments produce valid offsets
+    cls = offset_types
+    cls()
+
+
+@pytest.mark.parametrize("kwd", sorted(liboffsets._relativedelta_kwds))
+def test_valid_month_attributes(kwd, month_classes):
+    # GH#18226
+    cls = month_classes
+    # check that we cannot create e.g. MonthEnd(weeks=3)
+    msg = rf"__init__\(\) got an unexpected keyword argument '{kwd}'"
+    with pytest.raises(TypeError, match=msg):
+        cls(**{kwd: 3})
+
+
+def test_month_offset_name(month_classes):
+    # GH#33757 off.name with n != 1 should not raise AttributeError
+    obj = month_classes(1)
+    obj2 = month_classes(2)
+    assert obj2.name == obj.name
+
+
+@pytest.mark.parametrize("kwd", sorted(liboffsets._relativedelta_kwds))
+def test_valid_relativedelta_kwargs(kwd, request):
+    if kwd == "millisecond":
+        request.applymarker(
+            pytest.mark.xfail(
+                raises=NotImplementedError,
+                reason="Constructing DateOffset object with `millisecond` is not "
+                "yet supported.",
+            )
+        )
+    # Check that all the arguments specified in liboffsets._relativedelta_kwds
+    # are in fact valid relativedelta keyword args
+    DateOffset(**{kwd: 1})
+
+
+@pytest.mark.parametrize("kwd", sorted(liboffsets._relativedelta_kwds))
+def test_valid_tick_attributes(kwd, tick_classes):
+    # GH#18226
+    cls = tick_classes
+    # check that we cannot create e.g. Hour(weeks=3)
+    msg = rf"__init__\(\) got an unexpected keyword argument '{kwd}'"
+    with pytest.raises(TypeError, match=msg):
+        cls(**{kwd: 3})
+
+
+def test_validate_n_error():
+    with pytest.raises(TypeError, match="argument must be an integer"):
+        DateOffset(n="Doh!")
+
+    with pytest.raises(TypeError, match="argument must be an integer"):
+        MonthBegin(n=timedelta(1))
+
+    with pytest.raises(TypeError, match="argument must be an integer"):
+        BDay(n=np.array([1, 2], dtype=np.int64))
+
+
+def test_require_integers(offset_types):
+    cls = offset_types
+    with pytest.raises(ValueError, match="argument must be an integer"):
+        cls(n=1.5)
+
+
+def test_tick_normalize_raises(tick_classes):
+    # check that trying to create a Tick object with normalize=True raises
+    # GH#21427
+    cls = tick_classes
+    msg = "Tick offset with `normalize=True` are not allowed."
+    with pytest.raises(ValueError, match=msg):
+        cls(n=3, normalize=True)
+
+
+@pytest.mark.parametrize(
+    "offset_kwargs, expected_arg",
+    [
+        ({"nanoseconds": 1}, "1970-01-01 00:00:00.000000001"),
+        ({"nanoseconds": 5}, "1970-01-01 00:00:00.000000005"),
+        ({"nanoseconds": -1}, "1969-12-31 23:59:59.999999999"),
+        ({"microseconds": 1}, "1970-01-01 00:00:00.000001"),
+        ({"microseconds": -1}, "1969-12-31 23:59:59.999999"),
+        ({"seconds": 1}, "1970-01-01 00:00:01"),
+        ({"seconds": -1}, "1969-12-31 23:59:59"),
+        ({"minutes": 1}, "1970-01-01 00:01:00"),
+        ({"minutes": -1}, "1969-12-31 23:59:00"),
+        ({"hours": 1}, "1970-01-01 01:00:00"),
+        ({"hours": -1}, "1969-12-31 23:00:00"),
+        ({"days": 1}, "1970-01-02 00:00:00"),
+        ({"days": -1}, "1969-12-31 00:00:00"),
+        ({"weeks": 1}, "1970-01-08 00:00:00"),
+        ({"weeks": -1}, "1969-12-25 00:00:00"),
+        ({"months": 1}, "1970-02-01 00:00:00"),
+        ({"months": -1}, "1969-12-01 00:00:00"),
+        ({"years": 1}, "1971-01-01 00:00:00"),
+        ({"years": -1}, "1969-01-01 00:00:00"),
+    ],
+)
+def test_dateoffset_add_sub(offset_kwargs, expected_arg):
+    offset = DateOffset(**offset_kwargs)
+    ts = Timestamp(0)
+    result = ts + offset
+    expected = Timestamp(expected_arg)
+    assert result == expected
+    result -= offset
+    assert result == ts
+    result = offset + ts
+    assert result == expected
+
+
+def test_dateoffset_add_sub_timestamp_with_nano():
+    offset = DateOffset(minutes=2, nanoseconds=9)
+    ts = Timestamp(4)
+    result = ts + offset
+    expected = Timestamp("1970-01-01 00:02:00.000000013")
+    assert result == expected
+    result -= offset
+    assert result == ts
+    result = offset + ts
+    assert result == expected
+
+    offset2 = DateOffset(minutes=2, nanoseconds=9, hour=1)
+    assert offset2._use_relativedelta
+    with tm.assert_produces_warning(None):
+        # no warning about Discarding nonzero nanoseconds
+        result2 = ts + offset2
+    expected2 = Timestamp("1970-01-01 01:02:00.000000013")
+    assert result2 == expected2
+
+
+@pytest.mark.parametrize(
+    "attribute",
+    [
+        "hours",
+        "days",
+        "weeks",
+        "months",
+        "years",
+    ],
+)
+def test_dateoffset_immutable(attribute):
+    offset = DateOffset(**{attribute: 0})
+    msg = "DateOffset objects are immutable"
+    with pytest.raises(AttributeError, match=msg):
+        setattr(offset, attribute, 5)
+
+
+def test_dateoffset_misc():
+    oset = offsets.DateOffset(months=2, days=4)
+    # it works
+    oset.freqstr
+
+    assert not offsets.DateOffset(months=2) == 2
+
+
+@pytest.mark.parametrize("n", [-1, 1, 3])
+def test_construct_int_arg_no_kwargs_assumed_days(n):
+    # GH 45890, 45643
+    offset = DateOffset(n)
+    assert offset._offset == timedelta(1)
+    result = Timestamp(2022, 1, 2) + offset
+    expected = Timestamp(2022, 1, 2 + n)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "offset, expected",
+    [
+        (
+            DateOffset(minutes=7, nanoseconds=18),
+            Timestamp("2022-01-01 00:07:00.000000018"),
+        ),
+        (DateOffset(nanoseconds=3), Timestamp("2022-01-01 00:00:00.000000003")),
+    ],
+)
+def test_dateoffset_add_sub_timestamp_series_with_nano(offset, expected):
+    # GH 47856
+    start_time = Timestamp("2022-01-01")
+    teststamp = start_time
+    testseries = Series([start_time])
+    testseries = testseries + offset
+    assert testseries[0] == expected
+    testseries -= offset
+    assert testseries[0] == teststamp
+    testseries = offset + testseries
+    assert testseries[0] == expected
+
+
+@pytest.mark.parametrize(
+    "n_months, scaling_factor, start_timestamp, expected_timestamp",
+    [
+        (1, 2, "2020-01-30", "2020-03-30"),
+        (2, 1, "2020-01-30", "2020-03-30"),
+        (1, 0, "2020-01-30", "2020-01-30"),
+        (2, 0, "2020-01-30", "2020-01-30"),
+        (1, -1, "2020-01-30", "2019-12-30"),
+        (2, -1, "2020-01-30", "2019-11-30"),
+    ],
+)
+def test_offset_multiplication(
+    n_months, scaling_factor, start_timestamp, expected_timestamp
+):
+    # GH 47953
+    mo1 = DateOffset(months=n_months)
+
+    startscalar = Timestamp(start_timestamp)
+    startarray = Series([startscalar])
+
+    resultscalar = startscalar + (mo1 * scaling_factor)
+    resultarray = startarray + (mo1 * scaling_factor)
+
+    expectedscalar = Timestamp(expected_timestamp)
+    expectedarray = Series([expectedscalar])
+    assert resultscalar == expectedscalar
+
+    tm.assert_series_equal(resultarray, expectedarray)
+
+
+def test_dateoffset_operations_on_dataframes(performance_warning):
+    # GH 47953
+    df = DataFrame({"T": [Timestamp("2019-04-30")], "D": [DateOffset(months=1)]})
+    frameresult1 = df["T"] + 26 * df["D"]
+    df2 = DataFrame(
+        {
+            "T": [Timestamp("2019-04-30"), Timestamp("2019-04-30")],
+            "D": [DateOffset(months=1), DateOffset(months=1)],
+        }
+    )
+    expecteddate = Timestamp("2021-06-30")
+    with tm.assert_produces_warning(performance_warning):
+        frameresult2 = df2["T"] + 26 * df2["D"]
+
+    assert frameresult1[0] == expecteddate
+    assert frameresult2[0] == expecteddate
+
+
+def test_is_yqm_start_end():
+    freq_m = to_offset("ME")
+    bm = to_offset("BME")
+    qfeb = to_offset("QE-FEB")
+    qsfeb = to_offset("QS-FEB")
+    bq = to_offset("BQE")
+    bqs_apr = to_offset("BQS-APR")
+    as_nov = to_offset("YS-NOV")
+
+    tests = [
+        (freq_m.is_month_start(Timestamp("2013-06-01")), 1),
+        (bm.is_month_start(Timestamp("2013-06-01")), 0),
+        (freq_m.is_month_start(Timestamp("2013-06-03")), 0),
+        (bm.is_month_start(Timestamp("2013-06-03")), 1),
+        (qfeb.is_month_end(Timestamp("2013-02-28")), 1),
+        (qfeb.is_quarter_end(Timestamp("2013-02-28")), 1),
+        (qfeb.is_year_end(Timestamp("2013-02-28")), 1),
+        (qfeb.is_month_start(Timestamp("2013-03-01")), 1),
+        (qfeb.is_quarter_start(Timestamp("2013-03-01")), 1),
+        (qfeb.is_year_start(Timestamp("2013-03-01")), 1),
+        (qsfeb.is_month_end(Timestamp("2013-03-31")), 1),
+        (qsfeb.is_quarter_end(Timestamp("2013-03-31")), 0),
+        (qsfeb.is_year_end(Timestamp("2013-03-31")), 0),
+        (qsfeb.is_month_start(Timestamp("2013-02-01")), 1),
+        (qsfeb.is_quarter_start(Timestamp("2013-02-01")), 1),
+        (qsfeb.is_year_start(Timestamp("2013-02-01")), 1),
+        (bq.is_month_end(Timestamp("2013-06-30")), 0),
+        (bq.is_quarter_end(Timestamp("2013-06-30")), 0),
+        (bq.is_year_end(Timestamp("2013-06-30")), 0),
+        (bq.is_month_end(Timestamp("2013-06-28")), 1),
+        (bq.is_quarter_end(Timestamp("2013-06-28")), 1),
+        (bq.is_year_end(Timestamp("2013-06-28")), 0),
+        (bqs_apr.is_month_end(Timestamp("2013-06-30")), 0),
+        (bqs_apr.is_quarter_end(Timestamp("2013-06-30")), 0),
+        (bqs_apr.is_year_end(Timestamp("2013-06-30")), 0),
+        (bqs_apr.is_month_end(Timestamp("2013-06-28")), 1),
+        (bqs_apr.is_quarter_end(Timestamp("2013-06-28")), 1),
+        (bqs_apr.is_year_end(Timestamp("2013-03-29")), 1),
+        (as_nov.is_year_start(Timestamp("2013-11-01")), 1),
+        (as_nov.is_year_end(Timestamp("2013-10-31")), 1),
+        (Timestamp("2012-02-01").days_in_month, 29),
+        (Timestamp("2013-02-01").days_in_month, 28),
+    ]
+
+    for ts, value in tests:
+        assert ts == value
+
+
+@pytest.mark.parametrize("left", [DateOffset(1), Nano(1)])
+@pytest.mark.parametrize("right", [DateOffset(1), Nano(1)])
+def test_multiply_dateoffset_typeerror(left, right):
+    with pytest.raises(TypeError, match="Cannot multiply"):
+        left * right
diff --git a/pandas/pandas/tests/tseries/offsets/test_offsets_properties.py b/pandas/pandas/tests/tseries/offsets/test_offsets_properties.py
new file mode 100644
index 0000000000000..809d8f87b2c02
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_offsets_properties.py
@@ -0,0 +1,73 @@
+"""
+Behavioral based tests for offsets and date_range.
+
+This file is adapted from https://github.com/pandas-dev/pandas/pull/18761 -
+which was more ambitious but less idiomatic in its use of Hypothesis.
+
+You may wish to consult the previous version for inspiration on further
+tests, or when trying to pin down the bugs exposed by the tests below.
+"""
+
+import zoneinfo
+
+from hypothesis import (
+    assume,
+    given,
+)
+import pytest
+
+from pandas.compat import WASM
+
+import pandas as pd
+from pandas._testing._hypothesis import (
+    DATETIME_JAN_1_1900_OPTIONAL_TZ,
+    YQM_OFFSET,
+)
+
+# ----------------------------------------------------------------
+# Offset-specific behaviour tests
+
+
+@pytest.mark.arm_slow
+@given(DATETIME_JAN_1_1900_OPTIONAL_TZ, YQM_OFFSET)
+def test_on_offset_implementations(dt, offset):
+    assume(not offset.normalize)
+    # This case is flaky in CI 2024-11-04
+    assume(
+        not (
+            WASM
+            and isinstance(dt.tzinfo, zoneinfo.ZoneInfo)
+            and dt.tzinfo.key == "Indian/Cocos"
+            and isinstance(offset, pd.offsets.MonthBegin)
+        )
+    )
+    # check that the class-specific implementations of is_on_offset match
+    # the general case definition:
+    #   (dt + offset) - offset == dt
+    try:
+        compare = (dt + offset) - offset
+    except ValueError:
+        # When dt + offset does not exist or is DST-ambiguous, assume(False) to
+        # indicate to hypothesis that this is not a valid test case
+        # DST-ambiguous example (GH41906):
+        # dt = datetime.datetime(1900, 1, 1, tzinfo=ZoneInfo('Africa/Kinshasa'))
+        # offset = MonthBegin(66)
+        assume(False)
+
+    assert offset.is_on_offset(dt) == (compare == dt)
+
+
+@given(YQM_OFFSET)
+def test_shift_across_dst(offset):
+    # GH#18319 check that 1) timezone is correctly normalized and
+    # 2) that hour is not incorrectly changed by this normalization
+    assume(not offset.normalize)
+
+    # Note that dti includes a transition across DST boundary
+    dti = pd.date_range(
+        start="2017-10-30 12:00:00", end="2017-11-06", freq="D", tz="US/Eastern"
+    )
+    assert (dti.hour == 12).all()  # we haven't screwed up yet
+
+    res = dti + offset
+    assert (res.hour == 12).all()
diff --git a/pandas/pandas/tests/tseries/offsets/test_quarter.py b/pandas/pandas/tests/tseries/offsets/test_quarter.py
new file mode 100644
index 0000000000000..b92ff9d39a3ca
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_quarter.py
@@ -0,0 +1,287 @@
+"""
+Tests for the following offsets:
+- QuarterBegin
+- QuarterEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import pytest
+
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    QuarterBegin,
+    QuarterEnd,
+)
+
+
+@pytest.mark.parametrize("klass", (QuarterBegin, QuarterEnd))
+def test_quarterly_dont_normalize(klass):
+    date = datetime(2012, 3, 31, 5, 30)
+    result = date + klass()
+    assert result.time() == date.time()
+
+
+@pytest.mark.parametrize("offset", [QuarterBegin(), QuarterEnd()])
+@pytest.mark.parametrize(
+    "date",
+    [
+        datetime(2016, m, d)
+        for m in [10, 11, 12]
+        for d in [1, 2, 3, 28, 29, 30, 31]
+        if not (m == 11 and d == 31)
+    ],
+)
+def test_on_offset(offset, date):
+    res = offset.is_on_offset(date)
+    slow_version = date == (date + offset) - offset
+    assert res == slow_version
+
+
+class TestQuarterBegin:
+    def test_repr(self):
+        expected = "<QuarterBegin: startingMonth=3>"
+        assert repr(QuarterBegin()) == expected
+        expected = "<QuarterBegin: startingMonth=3>"
+        assert repr(QuarterBegin(startingMonth=3)) == expected
+        expected = "<QuarterBegin: startingMonth=1>"
+        assert repr(QuarterBegin(startingMonth=1)) == expected
+
+    def test_offset_corner_case(self):
+        # corner
+        offset = QuarterBegin(n=-1, startingMonth=1)
+        assert datetime(2010, 2, 1) + offset == datetime(2010, 1, 1)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            QuarterBegin(startingMonth=1),
+            {
+                datetime(2007, 12, 1): datetime(2008, 1, 1),
+                datetime(2008, 1, 1): datetime(2008, 4, 1),
+                datetime(2008, 2, 15): datetime(2008, 4, 1),
+                datetime(2008, 2, 29): datetime(2008, 4, 1),
+                datetime(2008, 3, 15): datetime(2008, 4, 1),
+                datetime(2008, 3, 31): datetime(2008, 4, 1),
+                datetime(2008, 4, 15): datetime(2008, 7, 1),
+                datetime(2008, 4, 1): datetime(2008, 7, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterBegin(startingMonth=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 1),
+                datetime(2008, 1, 31): datetime(2008, 2, 1),
+                datetime(2008, 1, 15): datetime(2008, 2, 1),
+                datetime(2008, 2, 29): datetime(2008, 5, 1),
+                datetime(2008, 3, 15): datetime(2008, 5, 1),
+                datetime(2008, 3, 31): datetime(2008, 5, 1),
+                datetime(2008, 4, 15): datetime(2008, 5, 1),
+                datetime(2008, 4, 30): datetime(2008, 5, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterBegin(startingMonth=1, n=0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 12, 1): datetime(2009, 1, 1),
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 2, 15): datetime(2008, 4, 1),
+                datetime(2008, 2, 29): datetime(2008, 4, 1),
+                datetime(2008, 3, 15): datetime(2008, 4, 1),
+                datetime(2008, 3, 31): datetime(2008, 4, 1),
+                datetime(2008, 4, 15): datetime(2008, 7, 1),
+                datetime(2008, 4, 30): datetime(2008, 7, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterBegin(startingMonth=1, n=-1),
+            {
+                datetime(2008, 1, 1): datetime(2007, 10, 1),
+                datetime(2008, 1, 31): datetime(2008, 1, 1),
+                datetime(2008, 2, 15): datetime(2008, 1, 1),
+                datetime(2008, 2, 29): datetime(2008, 1, 1),
+                datetime(2008, 3, 15): datetime(2008, 1, 1),
+                datetime(2008, 3, 31): datetime(2008, 1, 1),
+                datetime(2008, 4, 15): datetime(2008, 4, 1),
+                datetime(2008, 4, 30): datetime(2008, 4, 1),
+                datetime(2008, 7, 1): datetime(2008, 4, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterBegin(startingMonth=1, n=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 7, 1),
+                datetime(2008, 2, 15): datetime(2008, 7, 1),
+                datetime(2008, 2, 29): datetime(2008, 7, 1),
+                datetime(2008, 3, 15): datetime(2008, 7, 1),
+                datetime(2008, 3, 31): datetime(2008, 7, 1),
+                datetime(2008, 4, 15): datetime(2008, 10, 1),
+                datetime(2008, 4, 1): datetime(2008, 10, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+
+class TestQuarterEnd:
+    def test_repr(self):
+        expected = "<QuarterEnd: startingMonth=3>"
+        assert repr(QuarterEnd()) == expected
+        expected = "<QuarterEnd: startingMonth=3>"
+        assert repr(QuarterEnd(startingMonth=3)) == expected
+        expected = "<QuarterEnd: startingMonth=1>"
+        assert repr(QuarterEnd(startingMonth=1)) == expected
+
+    def test_offset_corner_case(self):
+        # corner
+        offset = QuarterEnd(n=-1, startingMonth=1)
+        assert datetime(2010, 2, 1) + offset == datetime(2010, 1, 31)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            QuarterEnd(startingMonth=1),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 4, 30),
+                datetime(2008, 2, 15): datetime(2008, 4, 30),
+                datetime(2008, 2, 29): datetime(2008, 4, 30),
+                datetime(2008, 3, 15): datetime(2008, 4, 30),
+                datetime(2008, 3, 31): datetime(2008, 4, 30),
+                datetime(2008, 4, 15): datetime(2008, 4, 30),
+                datetime(2008, 4, 30): datetime(2008, 7, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterEnd(startingMonth=2),
+            {
+                datetime(2008, 1, 1): datetime(2008, 2, 29),
+                datetime(2008, 1, 31): datetime(2008, 2, 29),
+                datetime(2008, 2, 15): datetime(2008, 2, 29),
+                datetime(2008, 2, 29): datetime(2008, 5, 31),
+                datetime(2008, 3, 15): datetime(2008, 5, 31),
+                datetime(2008, 3, 31): datetime(2008, 5, 31),
+                datetime(2008, 4, 15): datetime(2008, 5, 31),
+                datetime(2008, 4, 30): datetime(2008, 5, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterEnd(startingMonth=1, n=0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 31),
+                datetime(2008, 1, 31): datetime(2008, 1, 31),
+                datetime(2008, 2, 15): datetime(2008, 4, 30),
+                datetime(2008, 2, 29): datetime(2008, 4, 30),
+                datetime(2008, 3, 15): datetime(2008, 4, 30),
+                datetime(2008, 3, 31): datetime(2008, 4, 30),
+                datetime(2008, 4, 15): datetime(2008, 4, 30),
+                datetime(2008, 4, 30): datetime(2008, 4, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterEnd(startingMonth=1, n=-1),
+            {
+                datetime(2008, 1, 1): datetime(2007, 10, 31),
+                datetime(2008, 1, 31): datetime(2007, 10, 31),
+                datetime(2008, 2, 15): datetime(2008, 1, 31),
+                datetime(2008, 2, 29): datetime(2008, 1, 31),
+                datetime(2008, 3, 15): datetime(2008, 1, 31),
+                datetime(2008, 3, 31): datetime(2008, 1, 31),
+                datetime(2008, 4, 15): datetime(2008, 1, 31),
+                datetime(2008, 4, 30): datetime(2008, 1, 31),
+                datetime(2008, 7, 1): datetime(2008, 4, 30),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            QuarterEnd(startingMonth=1, n=2),
+            {
+                datetime(2008, 1, 31): datetime(2008, 7, 31),
+                datetime(2008, 2, 15): datetime(2008, 7, 31),
+                datetime(2008, 2, 29): datetime(2008, 7, 31),
+                datetime(2008, 3, 15): datetime(2008, 7, 31),
+                datetime(2008, 3, 31): datetime(2008, 7, 31),
+                datetime(2008, 4, 15): datetime(2008, 7, 31),
+                datetime(2008, 4, 30): datetime(2008, 10, 31),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (QuarterEnd(1, startingMonth=1), datetime(2008, 1, 31), True),
+        (QuarterEnd(1, startingMonth=1), datetime(2007, 12, 31), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2008, 2, 29), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2007, 3, 30), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2007, 3, 31), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2008, 4, 30), True),
+        (QuarterEnd(1, startingMonth=1), datetime(2008, 5, 30), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2008, 5, 31), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2007, 6, 29), False),
+        (QuarterEnd(1, startingMonth=1), datetime(2007, 6, 30), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2008, 1, 31), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2007, 12, 31), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2008, 2, 29), True),
+        (QuarterEnd(1, startingMonth=2), datetime(2007, 3, 30), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2007, 3, 31), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2008, 4, 30), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2008, 5, 30), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2008, 5, 31), True),
+        (QuarterEnd(1, startingMonth=2), datetime(2007, 6, 29), False),
+        (QuarterEnd(1, startingMonth=2), datetime(2007, 6, 30), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2008, 1, 31), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2007, 12, 31), True),
+        (QuarterEnd(1, startingMonth=3), datetime(2008, 2, 29), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2007, 3, 30), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2007, 3, 31), True),
+        (QuarterEnd(1, startingMonth=3), datetime(2008, 4, 30), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2008, 5, 30), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2008, 5, 31), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2007, 6, 29), False),
+        (QuarterEnd(1, startingMonth=3), datetime(2007, 6, 30), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
diff --git a/pandas/pandas/tests/tseries/offsets/test_ticks.py b/pandas/pandas/tests/tseries/offsets/test_ticks.py
new file mode 100644
index 0000000000000..46b6846ad1ec2
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_ticks.py
@@ -0,0 +1,386 @@
+"""
+Tests for offsets.Tick and subclasses
+"""
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+from hypothesis import (
+    assume,
+    example,
+    given,
+)
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.offsets import delta_to_tick
+
+from pandas import (
+    Timedelta,
+    Timestamp,
+)
+import pandas._testing as tm
+from pandas._testing._hypothesis import INT_NEG_999_TO_POS_999
+from pandas.tests.tseries.offsets.common import assert_offset_equal
+
+from pandas.tseries import offsets
+from pandas.tseries.offsets import (
+    Hour,
+    Micro,
+    Milli,
+    Minute,
+    Nano,
+    Second,
+)
+
+# ---------------------------------------------------------------------
+# Test Helpers
+
+tick_classes = [Hour, Minute, Second, Milli, Micro, Nano]
+
+
+# ---------------------------------------------------------------------
+
+
+def test_apply_ticks():
+    result = offsets.Hour(3) + offsets.Hour(4)
+    exp = offsets.Hour(7)
+    assert result == exp
+
+
+def test_delta_to_tick():
+    delta = timedelta(3)
+
+    tick = delta_to_tick(delta)
+    assert tick == offsets.Day(3)
+
+    td = Timedelta(nanoseconds=5)
+    tick = delta_to_tick(td)
+    assert tick == Nano(5)
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+@example(n=2, m=3)
+@example(n=800, m=300)
+@example(n=1000, m=5)
+@given(n=INT_NEG_999_TO_POS_999, m=INT_NEG_999_TO_POS_999)
+def test_tick_add_sub(cls, n, m):
+    # For all Tick subclasses and all integers n, m, we should have
+    # tick(n) + tick(m) == tick(n+m)
+    # tick(n) - tick(m) == tick(n-m)
+    left = cls(n)
+    right = cls(m)
+    expected = cls(n + m)
+
+    assert left + right == expected
+
+    expected = cls(n - m)
+    assert left - right == expected
+
+
+@pytest.mark.arm_slow
+@pytest.mark.parametrize("cls", tick_classes)
+@example(n=2, m=3)
+@given(n=INT_NEG_999_TO_POS_999, m=INT_NEG_999_TO_POS_999)
+def test_tick_equality(cls, n, m):
+    assume(m != n)
+    # tick == tock iff tick.n == tock.n
+    left = cls(n)
+    right = cls(m)
+    assert left != right
+
+    right = cls(n)
+    assert left == right
+    assert not left != right
+
+    if n != 0:
+        assert cls(n) != cls(-n)
+
+
+# ---------------------------------------------------------------------
+
+
+def test_Hour():
+    assert_offset_equal(Hour(), datetime(2010, 1, 1), datetime(2010, 1, 1, 1))
+    assert_offset_equal(Hour(-1), datetime(2010, 1, 1, 1), datetime(2010, 1, 1))
+    assert_offset_equal(2 * Hour(), datetime(2010, 1, 1), datetime(2010, 1, 1, 2))
+    assert_offset_equal(-1 * Hour(), datetime(2010, 1, 1, 1), datetime(2010, 1, 1))
+
+    assert Hour(3) + Hour(2) == Hour(5)
+    assert Hour(3) - Hour(2) == Hour()
+
+    assert Hour(4) != Hour(1)
+
+
+def test_Minute():
+    assert_offset_equal(Minute(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 1))
+    assert_offset_equal(Minute(-1), datetime(2010, 1, 1, 0, 1), datetime(2010, 1, 1))
+    assert_offset_equal(2 * Minute(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 2))
+    assert_offset_equal(-1 * Minute(), datetime(2010, 1, 1, 0, 1), datetime(2010, 1, 1))
+
+    assert Minute(3) + Minute(2) == Minute(5)
+    assert Minute(3) - Minute(2) == Minute()
+    assert Minute(5) != Minute()
+
+
+def test_Second():
+    assert_offset_equal(Second(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 1))
+    assert_offset_equal(Second(-1), datetime(2010, 1, 1, 0, 0, 1), datetime(2010, 1, 1))
+    assert_offset_equal(
+        2 * Second(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 2)
+    )
+    assert_offset_equal(
+        -1 * Second(), datetime(2010, 1, 1, 0, 0, 1), datetime(2010, 1, 1)
+    )
+
+    assert Second(3) + Second(2) == Second(5)
+    assert Second(3) - Second(2) == Second()
+
+
+def test_Millisecond():
+    assert_offset_equal(
+        Milli(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 0, 1000)
+    )
+    assert_offset_equal(
+        Milli(-1), datetime(2010, 1, 1, 0, 0, 0, 1000), datetime(2010, 1, 1)
+    )
+    assert_offset_equal(
+        Milli(2), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 0, 2000)
+    )
+    assert_offset_equal(
+        2 * Milli(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 0, 2000)
+    )
+    assert_offset_equal(
+        -1 * Milli(), datetime(2010, 1, 1, 0, 0, 0, 1000), datetime(2010, 1, 1)
+    )
+
+    assert Milli(3) + Milli(2) == Milli(5)
+    assert Milli(3) - Milli(2) == Milli()
+
+
+def test_MillisecondTimestampArithmetic():
+    assert_offset_equal(
+        Milli(), Timestamp("2010-01-01"), Timestamp("2010-01-01 00:00:00.001")
+    )
+    assert_offset_equal(
+        Milli(-1), Timestamp("2010-01-01 00:00:00.001"), Timestamp("2010-01-01")
+    )
+
+
+def test_Microsecond():
+    assert_offset_equal(Micro(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 0, 1))
+    assert_offset_equal(
+        Micro(-1), datetime(2010, 1, 1, 0, 0, 0, 1), datetime(2010, 1, 1)
+    )
+
+    assert_offset_equal(
+        2 * Micro(), datetime(2010, 1, 1), datetime(2010, 1, 1, 0, 0, 0, 2)
+    )
+    assert_offset_equal(
+        -1 * Micro(), datetime(2010, 1, 1, 0, 0, 0, 1), datetime(2010, 1, 1)
+    )
+
+    assert Micro(3) + Micro(2) == Micro(5)
+    assert Micro(3) - Micro(2) == Micro()
+
+
+def test_NanosecondGeneric():
+    timestamp = Timestamp(datetime(2010, 1, 1))
+    assert timestamp.nanosecond == 0
+
+    result = timestamp + Nano(10)
+    assert result.nanosecond == 10
+
+    reverse_result = Nano(10) + timestamp
+    assert reverse_result.nanosecond == 10
+
+
+def test_Nanosecond():
+    timestamp = Timestamp(datetime(2010, 1, 1))
+    assert_offset_equal(Nano(), timestamp, timestamp + np.timedelta64(1, "ns"))
+    assert_offset_equal(Nano(-1), timestamp + np.timedelta64(1, "ns"), timestamp)
+    assert_offset_equal(2 * Nano(), timestamp, timestamp + np.timedelta64(2, "ns"))
+    assert_offset_equal(-1 * Nano(), timestamp + np.timedelta64(1, "ns"), timestamp)
+
+    assert Nano(3) + Nano(2) == Nano(5)
+    assert Nano(3) - Nano(2) == Nano()
+
+    # GH9284
+    assert Nano(1) + Nano(10) == Nano(11)
+    assert Nano(5) + Micro(1) == Nano(1005)
+    assert Micro(5) + Nano(1) == Nano(5001)
+
+
+@pytest.mark.parametrize(
+    "kls, expected",
+    [
+        (Hour, Timedelta(hours=5)),
+        (Minute, Timedelta(hours=2, minutes=3)),
+        (Second, Timedelta(hours=2, seconds=3)),
+        (Milli, Timedelta(hours=2, milliseconds=3)),
+        (Micro, Timedelta(hours=2, microseconds=3)),
+        (Nano, Timedelta(hours=2, nanoseconds=3)),
+    ],
+)
+def test_tick_addition(kls, expected):
+    offset = kls(3)
+    td = Timedelta(hours=2)
+
+    for other in [td, td.to_pytimedelta(), td.to_timedelta64()]:
+        result = offset + other
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+        result = other + offset
+        assert isinstance(result, Timedelta)
+        assert result == expected
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_tick_division(cls):
+    off = cls(10)
+
+    assert off / cls(5) == 2
+    assert off / 2 == cls(5)
+    assert off / 2.0 == cls(5)
+
+    assert off / off._as_pd_timedelta == 1
+    assert off / off._as_pd_timedelta.to_timedelta64() == 1
+
+    assert off / Nano(1) == off._as_pd_timedelta / Nano(1)._as_pd_timedelta
+
+    if cls is not Nano:
+        # A case where we end up with a smaller class
+        result = off / 1000
+        assert isinstance(result, offsets.Tick)
+        assert not isinstance(result, cls)
+        assert result._as_pd_timedelta == off._as_pd_timedelta / 1000
+
+    if cls._nanos_inc < Timedelta(seconds=1)._value:
+        # Case where we end up with a bigger class
+        result = off / 0.001
+        assert isinstance(result, offsets.Tick)
+        assert not isinstance(result, cls)
+        assert result._as_pd_timedelta == off._as_pd_timedelta / 0.001
+
+
+def test_tick_mul_float():
+    off = Micro(2)
+
+    # Case where we retain type
+    result = off * 1.5
+    expected = Micro(3)
+    assert result == expected
+    assert isinstance(result, Micro)
+
+    # Case where we bump up to the next type
+    result = off * 1.25
+    expected = Nano(2500)
+    assert result == expected
+    assert isinstance(result, Nano)
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_tick_rdiv(cls):
+    off = cls(10)
+    delta = off._as_pd_timedelta
+    td64 = delta.to_timedelta64()
+    instance__type = ".".join([cls.__module__, cls.__name__])
+    msg = (
+        f"unsupported operand type\\(s\\) for \\/: 'int'|'float' and '{instance__type}'"
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        2 / off
+    with pytest.raises(TypeError, match=msg):
+        2.0 / off
+
+    assert (td64 * 2.5) / off == 2.5
+
+    if cls is not Nano:
+        # skip pytimedelta for Nano since it gets dropped
+        assert (delta.to_pytimedelta() * 2) / off == 2
+
+    result = np.array([2 * td64, td64]) / off
+    expected = np.array([2.0, 1.0])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize("cls1", tick_classes)
+@pytest.mark.parametrize("cls2", tick_classes)
+def test_tick_zero(cls1, cls2):
+    assert cls1(0) == cls2(0)
+    assert cls1(0) + cls2(0) == cls1(0)
+
+    if cls1 is not Nano:
+        assert cls1(2) + cls2(0) == cls1(2)
+
+    if cls1 is Nano:
+        assert cls1(2) + Nano(0) == cls1(2)
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_tick_equalities(cls):
+    assert cls() == cls(1)
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_compare_ticks(cls):
+    three = cls(3)
+    four = cls(4)
+
+    assert three < cls(4)
+    assert cls(3) < four
+    assert four > cls(3)
+    assert cls(4) > three
+    assert cls(3) == cls(3)
+    assert cls(3) != cls(4)
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_compare_ticks_to_strs(cls):
+    # GH#23524
+    off = cls(19)
+
+    # These tests should work with any strings, but we particularly are
+    #  interested in "infer" as that comparison is convenient to make in
+    #  Datetime/Timedelta Array/Index constructors
+    assert not off == "infer"
+    assert not "foo" == off
+
+    instance_type = ".".join([cls.__module__, cls.__name__])
+    msg = (
+        "'<'|'<='|'>'|'>=' not supported between instances of "
+        f"'str' and '{instance_type}'|'{instance_type}' and 'str'"
+    )
+
+    for left, right in [("infer", off), (off, "infer")]:
+        with pytest.raises(TypeError, match=msg):
+            left < right
+        with pytest.raises(TypeError, match=msg):
+            left <= right
+        with pytest.raises(TypeError, match=msg):
+            left > right
+        with pytest.raises(TypeError, match=msg):
+            left >= right
+
+
+@pytest.mark.parametrize("cls", tick_classes)
+def test_compare_ticks_to_timedeltalike(cls):
+    off = cls(19)
+
+    td = off._as_pd_timedelta
+
+    others = [td, td.to_timedelta64()]
+    if cls is not Nano:
+        others.append(td.to_pytimedelta())
+
+    for other in others:
+        assert off == other
+        assert not off != other
+        assert not off < other
+        assert not off > other
+        assert off <= other
+        assert off >= other
diff --git a/pandas/pandas/tests/tseries/offsets/test_week.py b/pandas/pandas/tests/tseries/offsets/test_week.py
new file mode 100644
index 0000000000000..d34a7953121c1
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_week.py
@@ -0,0 +1,342 @@
+"""
+Tests for the following offsets:
+- Week
+- WeekOfMonth
+- LastWeekOfMonth
+"""
+
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import pytest
+
+from pandas._libs.tslibs import Timestamp
+from pandas._libs.tslibs.offsets import (
+    Day,
+    LastWeekOfMonth,
+    Week,
+    WeekOfMonth,
+)
+
+from pandas.tests.tseries.offsets.common import (
+    WeekDay,
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+
+class TestWeek:
+    def test_repr(self):
+        assert repr(Week(weekday=0)) == "<Week: weekday=0>"
+        assert repr(Week(n=-1, weekday=0)) == "<-1 * Week: weekday=0>"
+        assert repr(Week(n=-2, weekday=0)) == "<-2 * Weeks: weekday=0>"
+
+    def test_corner(self):
+        with pytest.raises(ValueError, match="Day must be"):
+            Week(weekday=7)
+
+        with pytest.raises(ValueError, match="Day must be"):
+            Week(weekday=-1)
+
+    offset_cases = []
+    # not business week
+    offset_cases.append(
+        (
+            Week(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 8),
+                datetime(2008, 1, 4): datetime(2008, 1, 11),
+                datetime(2008, 1, 5): datetime(2008, 1, 12),
+                datetime(2008, 1, 6): datetime(2008, 1, 13),
+                datetime(2008, 1, 7): datetime(2008, 1, 14),
+            },
+        )
+    )
+
+    # Mon
+    offset_cases.append(
+        (
+            Week(weekday=0),
+            {
+                datetime(2007, 12, 31): datetime(2008, 1, 7),
+                datetime(2008, 1, 4): datetime(2008, 1, 7),
+                datetime(2008, 1, 5): datetime(2008, 1, 7),
+                datetime(2008, 1, 6): datetime(2008, 1, 7),
+                datetime(2008, 1, 7): datetime(2008, 1, 14),
+            },
+        )
+    )
+
+    # n=0 -> roll forward. Mon
+    offset_cases.append(
+        (
+            Week(0, weekday=0),
+            {
+                datetime(2007, 12, 31): datetime(2007, 12, 31),
+                datetime(2008, 1, 4): datetime(2008, 1, 7),
+                datetime(2008, 1, 5): datetime(2008, 1, 7),
+                datetime(2008, 1, 6): datetime(2008, 1, 7),
+                datetime(2008, 1, 7): datetime(2008, 1, 7),
+            },
+        )
+    )
+
+    # n=0 -> roll forward. Mon
+    offset_cases.append(
+        (
+            Week(-2, weekday=1),
+            {
+                datetime(2010, 4, 6): datetime(2010, 3, 23),
+                datetime(2010, 4, 8): datetime(2010, 3, 30),
+                datetime(2010, 4, 5): datetime(2010, 3, 23),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    @pytest.mark.parametrize("weekday", range(7))
+    def test_is_on_offset(self, weekday):
+        offset = Week(weekday=weekday)
+
+        for day in range(1, 8):
+            date = datetime(2008, 1, day)
+            expected = day % 7 == weekday
+        assert_is_on_offset(offset, date, expected)
+
+    @pytest.mark.parametrize(
+        "n,date",
+        [
+            (2, "1862-01-13 09:03:34.873477378+0210"),
+            (-2, "1856-10-24 16:18:36.556360110-0717"),
+        ],
+    )
+    def test_is_on_offset_weekday_none(self, n, date):
+        # GH 18510 Week with weekday = None, normalize = False
+        # should always be is_on_offset
+        offset = Week(n=n, weekday=None)
+        ts = Timestamp(date, tz="Africa/Lusaka")
+        fast = offset.is_on_offset(ts)
+        slow = (ts + offset) - offset == ts
+        assert fast == slow
+
+    def test_week_add_invalid(self):
+        # Week with weekday should raise TypeError and _not_ AttributeError
+        #  when adding invalid offset
+        offset = Week(weekday=1)
+        other = Day()
+        with pytest.raises(TypeError, match="Cannot add"):
+            offset + other
+
+
+class TestWeekOfMonth:
+    def test_constructor(self):
+        with pytest.raises(ValueError, match="^Week"):
+            WeekOfMonth(n=1, week=4, weekday=0)
+
+        with pytest.raises(ValueError, match="^Week"):
+            WeekOfMonth(n=1, week=-1, weekday=0)
+
+        with pytest.raises(ValueError, match="^Day"):
+            WeekOfMonth(n=1, week=0, weekday=-1)
+
+        with pytest.raises(ValueError, match="^Day"):
+            WeekOfMonth(n=1, week=0, weekday=-7)
+
+    def test_repr(self):
+        assert (
+            repr(WeekOfMonth(weekday=1, week=2)) == "<WeekOfMonth: week=2, weekday=1>"
+        )
+
+    def test_offset(self):
+        date1 = datetime(2011, 1, 4)  # 1st Tuesday of Month
+        date2 = datetime(2011, 1, 11)  # 2nd Tuesday of Month
+        date3 = datetime(2011, 1, 18)  # 3rd Tuesday of Month
+        date4 = datetime(2011, 1, 25)  # 4th Tuesday of Month
+
+        # see for loop for structure
+        test_cases = [
+            (-2, 2, 1, date1, datetime(2010, 11, 16)),
+            (-2, 2, 1, date2, datetime(2010, 11, 16)),
+            (-2, 2, 1, date3, datetime(2010, 11, 16)),
+            (-2, 2, 1, date4, datetime(2010, 12, 21)),
+            (-1, 2, 1, date1, datetime(2010, 12, 21)),
+            (-1, 2, 1, date2, datetime(2010, 12, 21)),
+            (-1, 2, 1, date3, datetime(2010, 12, 21)),
+            (-1, 2, 1, date4, datetime(2011, 1, 18)),
+            (0, 0, 1, date1, datetime(2011, 1, 4)),
+            (0, 0, 1, date2, datetime(2011, 2, 1)),
+            (0, 0, 1, date3, datetime(2011, 2, 1)),
+            (0, 0, 1, date4, datetime(2011, 2, 1)),
+            (0, 1, 1, date1, datetime(2011, 1, 11)),
+            (0, 1, 1, date2, datetime(2011, 1, 11)),
+            (0, 1, 1, date3, datetime(2011, 2, 8)),
+            (0, 1, 1, date4, datetime(2011, 2, 8)),
+            (0, 0, 1, date1, datetime(2011, 1, 4)),
+            (0, 1, 1, date2, datetime(2011, 1, 11)),
+            (0, 2, 1, date3, datetime(2011, 1, 18)),
+            (0, 3, 1, date4, datetime(2011, 1, 25)),
+            (1, 0, 0, date1, datetime(2011, 2, 7)),
+            (1, 0, 0, date2, datetime(2011, 2, 7)),
+            (1, 0, 0, date3, datetime(2011, 2, 7)),
+            (1, 0, 0, date4, datetime(2011, 2, 7)),
+            (1, 0, 1, date1, datetime(2011, 2, 1)),
+            (1, 0, 1, date2, datetime(2011, 2, 1)),
+            (1, 0, 1, date3, datetime(2011, 2, 1)),
+            (1, 0, 1, date4, datetime(2011, 2, 1)),
+            (1, 0, 2, date1, datetime(2011, 1, 5)),
+            (1, 0, 2, date2, datetime(2011, 2, 2)),
+            (1, 0, 2, date3, datetime(2011, 2, 2)),
+            (1, 0, 2, date4, datetime(2011, 2, 2)),
+            (1, 2, 1, date1, datetime(2011, 1, 18)),
+            (1, 2, 1, date2, datetime(2011, 1, 18)),
+            (1, 2, 1, date3, datetime(2011, 2, 15)),
+            (1, 2, 1, date4, datetime(2011, 2, 15)),
+            (2, 2, 1, date1, datetime(2011, 2, 15)),
+            (2, 2, 1, date2, datetime(2011, 2, 15)),
+            (2, 2, 1, date3, datetime(2011, 3, 15)),
+            (2, 2, 1, date4, datetime(2011, 3, 15)),
+        ]
+
+        for n, week, weekday, dt, expected in test_cases:
+            offset = WeekOfMonth(n, week=week, weekday=weekday)
+            assert_offset_equal(offset, dt, expected)
+
+        # try subtracting
+        result = datetime(2011, 2, 1) - WeekOfMonth(week=1, weekday=2)
+        assert result == datetime(2011, 1, 12)
+
+        result = datetime(2011, 2, 3) - WeekOfMonth(week=0, weekday=2)
+        assert result == datetime(2011, 2, 2)
+
+    on_offset_cases = [
+        (0, 0, datetime(2011, 2, 7), True),
+        (0, 0, datetime(2011, 2, 6), False),
+        (0, 0, datetime(2011, 2, 14), False),
+        (1, 0, datetime(2011, 2, 14), True),
+        (0, 1, datetime(2011, 2, 1), True),
+        (0, 1, datetime(2011, 2, 8), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        week, weekday, dt, expected = case
+        offset = WeekOfMonth(week=week, weekday=weekday)
+        assert offset.is_on_offset(dt) == expected
+
+    @pytest.mark.parametrize(
+        "n,week,date,tz",
+        [
+            (2, 2, "1916-05-15 01:14:49.583410462+0422", "Asia/Qyzylorda"),
+            (-3, 1, "1980-12-08 03:38:52.878321185+0500", "Asia/Oral"),
+        ],
+    )
+    def test_is_on_offset_nanoseconds(self, n, week, date, tz):
+        # GH 18864
+        # Make sure that nanoseconds don't trip up is_on_offset (and with it apply)
+        offset = WeekOfMonth(n=n, week=week, weekday=0)
+        ts = Timestamp(date, tz=tz)
+        fast = offset.is_on_offset(ts)
+        slow = (ts + offset) - offset == ts
+        assert fast == slow
+
+
+class TestLastWeekOfMonth:
+    def test_constructor(self):
+        with pytest.raises(ValueError, match="^N cannot be 0"):
+            LastWeekOfMonth(n=0, weekday=1)
+
+        with pytest.raises(ValueError, match="^Day"):
+            LastWeekOfMonth(n=1, weekday=-1)
+
+        with pytest.raises(ValueError, match="^Day"):
+            LastWeekOfMonth(n=1, weekday=7)
+
+    def test_offset(self):
+        # Saturday
+        last_sat = datetime(2013, 8, 31)
+        next_sat = datetime(2013, 9, 28)
+        offset_sat = LastWeekOfMonth(n=1, weekday=5)
+
+        one_day_before = last_sat + timedelta(days=-1)
+        assert one_day_before + offset_sat == last_sat
+
+        one_day_after = last_sat + timedelta(days=+1)
+        assert one_day_after + offset_sat == next_sat
+
+        # Test On that day
+        assert last_sat + offset_sat == next_sat
+
+        # Thursday
+
+        offset_thur = LastWeekOfMonth(n=1, weekday=3)
+        last_thurs = datetime(2013, 1, 31)
+        next_thurs = datetime(2013, 2, 28)
+
+        one_day_before = last_thurs + timedelta(days=-1)
+        assert one_day_before + offset_thur == last_thurs
+
+        one_day_after = last_thurs + timedelta(days=+1)
+        assert one_day_after + offset_thur == next_thurs
+
+        # Test on that day
+        assert last_thurs + offset_thur == next_thurs
+
+        three_before = last_thurs + timedelta(days=-3)
+        assert three_before + offset_thur == last_thurs
+
+        two_after = last_thurs + timedelta(days=+2)
+        assert two_after + offset_thur == next_thurs
+
+        offset_sunday = LastWeekOfMonth(n=1, weekday=WeekDay.SUN)
+        assert datetime(2013, 7, 31) + offset_sunday == datetime(2013, 8, 25)
+
+    on_offset_cases = [
+        (WeekDay.SUN, datetime(2013, 1, 27), True),
+        (WeekDay.SAT, datetime(2013, 3, 30), True),
+        (WeekDay.MON, datetime(2013, 2, 18), False),  # Not the last Mon
+        (WeekDay.SUN, datetime(2013, 2, 25), False),  # Not a SUN
+        (WeekDay.MON, datetime(2013, 2, 25), True),
+        (WeekDay.SAT, datetime(2013, 11, 30), True),
+        (WeekDay.SAT, datetime(2006, 8, 26), True),
+        (WeekDay.SAT, datetime(2007, 8, 25), True),
+        (WeekDay.SAT, datetime(2008, 8, 30), True),
+        (WeekDay.SAT, datetime(2009, 8, 29), True),
+        (WeekDay.SAT, datetime(2010, 8, 28), True),
+        (WeekDay.SAT, datetime(2011, 8, 27), True),
+        (WeekDay.SAT, datetime(2019, 8, 31), True),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        weekday, dt, expected = case
+        offset = LastWeekOfMonth(weekday=weekday)
+        assert offset.is_on_offset(dt) == expected
+
+    @pytest.mark.parametrize(
+        "n,weekday,date,tz",
+        [
+            (4, 6, "1917-05-27 20:55:27.084284178+0200", "Europe/Warsaw"),
+            (-4, 5, "2005-08-27 05:01:42.799392561-0500", "America/Rainy_River"),
+        ],
+    )
+    def test_last_week_of_month_on_offset(self, n, weekday, date, tz):
+        # GH 19036, GH 18977 _adjust_dst was incorrect for LastWeekOfMonth
+        offset = LastWeekOfMonth(n=n, weekday=weekday)
+        ts = Timestamp(date, tz=tz)
+        slow = (ts + offset) - offset == ts
+        fast = offset.is_on_offset(ts)
+        assert fast == slow
+
+    def test_repr(self):
+        assert (
+            repr(LastWeekOfMonth(n=2, weekday=1)) == "<2 * LastWeekOfMonths: weekday=1>"
+        )
diff --git a/pandas/pandas/tests/tseries/offsets/test_year.py b/pandas/pandas/tests/tseries/offsets/test_year.py
new file mode 100644
index 0000000000000..9d2a0b20e1e7c
--- /dev/null
+++ b/pandas/pandas/tests/tseries/offsets/test_year.py
@@ -0,0 +1,340 @@
+"""
+Tests for the following offsets:
+- YearBegin
+- YearEnd
+"""
+
+from __future__ import annotations
+
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+from pandas import Timestamp
+from pandas.tests.tseries.offsets.common import (
+    assert_is_on_offset,
+    assert_offset_equal,
+)
+
+from pandas.tseries.offsets import (
+    YearBegin,
+    YearEnd,
+)
+
+
+class TestYearBegin:
+    def test_misspecified(self):
+        with pytest.raises(ValueError, match="Month must go from 1 to 12"):
+            YearBegin(month=13)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            YearBegin(),
+            {
+                datetime(2008, 1, 1): datetime(2009, 1, 1),
+                datetime(2008, 6, 30): datetime(2009, 1, 1),
+                datetime(2008, 12, 31): datetime(2009, 1, 1),
+                datetime(2005, 12, 30): datetime(2006, 1, 1),
+                datetime(2005, 12, 31): datetime(2006, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 1, 1),
+                datetime(2008, 6, 30): datetime(2009, 1, 1),
+                datetime(2008, 12, 31): datetime(2009, 1, 1),
+                datetime(2005, 12, 30): datetime(2006, 1, 1),
+                datetime(2005, 12, 31): datetime(2006, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(3),
+            {
+                datetime(2008, 1, 1): datetime(2011, 1, 1),
+                datetime(2008, 6, 30): datetime(2011, 1, 1),
+                datetime(2008, 12, 31): datetime(2011, 1, 1),
+                datetime(2005, 12, 30): datetime(2008, 1, 1),
+                datetime(2005, 12, 31): datetime(2008, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 1, 1),
+                datetime(2007, 1, 15): datetime(2007, 1, 1),
+                datetime(2008, 6, 30): datetime(2008, 1, 1),
+                datetime(2008, 12, 31): datetime(2008, 1, 1),
+                datetime(2006, 12, 29): datetime(2006, 1, 1),
+                datetime(2006, 12, 30): datetime(2006, 1, 1),
+                datetime(2007, 1, 1): datetime(2006, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(-2),
+            {
+                datetime(2007, 1, 1): datetime(2005, 1, 1),
+                datetime(2008, 6, 30): datetime(2007, 1, 1),
+                datetime(2008, 12, 31): datetime(2007, 1, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(month=4),
+            {
+                datetime(2007, 4, 1): datetime(2008, 4, 1),
+                datetime(2007, 4, 15): datetime(2008, 4, 1),
+                datetime(2007, 3, 1): datetime(2007, 4, 1),
+                datetime(2007, 12, 15): datetime(2008, 4, 1),
+                datetime(2012, 1, 31): datetime(2012, 4, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(0, month=4),
+            {
+                datetime(2007, 4, 1): datetime(2007, 4, 1),
+                datetime(2007, 3, 1): datetime(2007, 4, 1),
+                datetime(2007, 12, 15): datetime(2008, 4, 1),
+                datetime(2012, 1, 31): datetime(2012, 4, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(4, month=4),
+            {
+                datetime(2007, 4, 1): datetime(2011, 4, 1),
+                datetime(2007, 4, 15): datetime(2011, 4, 1),
+                datetime(2007, 3, 1): datetime(2010, 4, 1),
+                datetime(2007, 12, 15): datetime(2011, 4, 1),
+                datetime(2012, 1, 31): datetime(2015, 4, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(-1, month=4),
+            {
+                datetime(2007, 4, 1): datetime(2006, 4, 1),
+                datetime(2007, 3, 1): datetime(2006, 4, 1),
+                datetime(2007, 12, 15): datetime(2007, 4, 1),
+                datetime(2012, 1, 31): datetime(2011, 4, 1),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearBegin(-3, month=4),
+            {
+                datetime(2007, 4, 1): datetime(2004, 4, 1),
+                datetime(2007, 3, 1): datetime(2004, 4, 1),
+                datetime(2007, 12, 15): datetime(2005, 4, 1),
+                datetime(2012, 1, 31): datetime(2009, 4, 1),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (YearBegin(), datetime(2007, 1, 3), False),
+        (YearBegin(), datetime(2008, 1, 1), True),
+        (YearBegin(), datetime(2006, 12, 31), False),
+        (YearBegin(), datetime(2006, 1, 2), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+
+class TestYearEnd:
+    def test_misspecified(self):
+        with pytest.raises(ValueError, match="Month must go from 1 to 12"):
+            YearEnd(month=13)
+
+    offset_cases = []
+    offset_cases.append(
+        (
+            YearEnd(),
+            {
+                datetime(2008, 1, 1): datetime(2008, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 12, 31),
+                datetime(2008, 12, 31): datetime(2009, 12, 31),
+                datetime(2005, 12, 30): datetime(2005, 12, 31),
+                datetime(2005, 12, 31): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(0),
+            {
+                datetime(2008, 1, 1): datetime(2008, 12, 31),
+                datetime(2008, 6, 30): datetime(2008, 12, 31),
+                datetime(2008, 12, 31): datetime(2008, 12, 31),
+                datetime(2005, 12, 30): datetime(2005, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(-1),
+            {
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+                datetime(2008, 6, 30): datetime(2007, 12, 31),
+                datetime(2008, 12, 31): datetime(2007, 12, 31),
+                datetime(2006, 12, 29): datetime(2005, 12, 31),
+                datetime(2006, 12, 30): datetime(2005, 12, 31),
+                datetime(2007, 1, 1): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(-2),
+            {
+                datetime(2007, 1, 1): datetime(2005, 12, 31),
+                datetime(2008, 6, 30): datetime(2006, 12, 31),
+                datetime(2008, 12, 31): datetime(2006, 12, 31),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (YearEnd(), datetime(2007, 12, 31), True),
+        (YearEnd(), datetime(2008, 1, 1), False),
+        (YearEnd(), datetime(2006, 12, 31), True),
+        (YearEnd(), datetime(2006, 12, 29), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+
+class TestYearEndDiffMonth:
+    offset_cases = []
+    offset_cases.append(
+        (
+            YearEnd(month=3),
+            {
+                datetime(2008, 1, 1): datetime(2008, 3, 31),
+                datetime(2008, 2, 15): datetime(2008, 3, 31),
+                datetime(2008, 3, 31): datetime(2009, 3, 31),
+                datetime(2008, 3, 30): datetime(2008, 3, 31),
+                datetime(2005, 3, 31): datetime(2006, 3, 31),
+                datetime(2006, 7, 30): datetime(2007, 3, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(0, month=3),
+            {
+                datetime(2008, 1, 1): datetime(2008, 3, 31),
+                datetime(2008, 2, 28): datetime(2008, 3, 31),
+                datetime(2008, 3, 31): datetime(2008, 3, 31),
+                datetime(2005, 3, 30): datetime(2005, 3, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(-1, month=3),
+            {
+                datetime(2007, 1, 1): datetime(2006, 3, 31),
+                datetime(2008, 2, 28): datetime(2007, 3, 31),
+                datetime(2008, 3, 31): datetime(2007, 3, 31),
+                datetime(2006, 3, 29): datetime(2005, 3, 31),
+                datetime(2006, 3, 30): datetime(2005, 3, 31),
+                datetime(2007, 3, 1): datetime(2006, 3, 31),
+            },
+        )
+    )
+
+    offset_cases.append(
+        (
+            YearEnd(-2, month=3),
+            {
+                datetime(2007, 1, 1): datetime(2005, 3, 31),
+                datetime(2008, 6, 30): datetime(2007, 3, 31),
+                datetime(2008, 3, 31): datetime(2006, 3, 31),
+            },
+        )
+    )
+
+    @pytest.mark.parametrize("case", offset_cases)
+    def test_offset(self, case):
+        offset, cases = case
+        for base, expected in cases.items():
+            assert_offset_equal(offset, base, expected)
+
+    on_offset_cases = [
+        (YearEnd(month=3), datetime(2007, 3, 31), True),
+        (YearEnd(month=3), datetime(2008, 1, 1), False),
+        (YearEnd(month=3), datetime(2006, 3, 31), True),
+        (YearEnd(month=3), datetime(2006, 3, 29), False),
+    ]
+
+    @pytest.mark.parametrize("case", on_offset_cases)
+    def test_is_on_offset(self, case):
+        offset, dt, expected = case
+        assert_is_on_offset(offset, dt, expected)
+
+
+def test_add_out_of_pydatetime_range():
+    # GH#50348 don't raise in Timestamp.replace
+    ts = Timestamp(np.datetime64("-20000-12-31"))
+    off = YearEnd()
+
+    result = ts + off
+    # TODO(cython3): "arg: datetime" annotation will impose
+    # datetime limitations on Timestamp. The fused type below works in cy3
+    # ctypedef fused datetimelike:
+    #     _Timestamp
+    #     datetime
+    # expected = Timestamp(np.datetime64("-19999-12-31"))
+    # assert result == expected
+    assert result.year in (-19999, 1973)
+    assert result.month == 12
+    assert result.day == 31
diff --git a/pandas/pandas/tests/tslibs/__init__.py b/pandas/pandas/tests/tslibs/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/tslibs/test_api.py b/pandas/pandas/tests/tslibs/test_api.py
new file mode 100644
index 0000000000000..42d055326c2a5
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_api.py
@@ -0,0 +1,65 @@
+"""Tests that the tslibs API is locked down"""
+
+from pandas._libs import tslibs
+
+
+def test_namespace():
+    submodules = [
+        "base",
+        "ccalendar",
+        "conversion",
+        "dtypes",
+        "fields",
+        "nattype",
+        "np_datetime",
+        "offsets",
+        "parsing",
+        "period",
+        "strptime",
+        "vectorized",
+        "timedeltas",
+        "timestamps",
+        "timezones",
+        "tzconversion",
+    ]
+
+    api = [
+        "BaseOffset",
+        "NaT",
+        "NaTType",
+        "iNaT",
+        "nat_strings",
+        "OutOfBoundsDatetime",
+        "OutOfBoundsTimedelta",
+        "Period",
+        "IncompatibleFrequency",
+        "Resolution",
+        "Tick",
+        "Timedelta",
+        "dt64arr_to_periodarr",
+        "Timestamp",
+        "is_date_array_normalized",
+        "ints_to_pydatetime",
+        "normalize_i8_timestamps",
+        "get_resolution",
+        "delta_to_nanoseconds",
+        "ints_to_pytimedelta",
+        "localize_pydatetime",
+        "tz_convert_from_utc",
+        "tz_convert_from_utc_single",
+        "to_offset",
+        "tz_compare",
+        "is_unitless",
+        "astype_overflowsafe",
+        "get_unit_from_dtype",
+        "periods_per_day",
+        "periods_per_second",
+        "guess_datetime_format",
+        "add_overflowsafe",
+        "get_supported_dtype",
+        "is_supported_dtype",
+    ]
+
+    expected = set(submodules + api)
+    names = [x for x in dir(tslibs) if not x.startswith("__")]
+    assert set(names) == expected
diff --git a/pandas/pandas/tests/tslibs/test_array_to_datetime.py b/pandas/pandas/tests/tslibs/test_array_to_datetime.py
new file mode 100644
index 0000000000000..fc0000553049e
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_array_to_datetime.py
@@ -0,0 +1,300 @@
+from datetime import (
+    date,
+    datetime,
+    timedelta,
+    timezone,
+)
+
+from dateutil.tz.tz import tzoffset
+import numpy as np
+import pytest
+
+from pandas._libs import (
+    NaT,
+    iNaT,
+    tslib,
+)
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+
+from pandas import Timestamp
+import pandas._testing as tm
+
+creso_infer = NpyDatetimeUnit.NPY_FR_GENERIC.value
+
+
+class TestArrayToDatetimeResolutionInference:
+    # TODO: tests that include tzs, ints
+
+    def test_infer_all_nat(self):
+        arr = np.array([NaT, np.nan], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        assert result.dtype == "M8[s]"
+
+    def test_infer_homogeoneous_datetimes(self):
+        dt = datetime(2023, 10, 27, 18, 3, 5, 678000)
+        arr = np.array([dt, dt, dt], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([dt, dt, dt], dtype="M8[us]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_infer_homogeoneous_date_objects(self):
+        dt = datetime(2023, 10, 27, 18, 3, 5, 678000)
+        dt2 = dt.date()
+        arr = np.array([None, dt2, dt2, dt2], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([np.datetime64("NaT"), dt2, dt2, dt2], dtype="M8[s]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_infer_homogeoneous_dt64(self):
+        dt = datetime(2023, 10, 27, 18, 3, 5, 678000)
+        dt64 = np.datetime64(dt, "ms")
+        arr = np.array([None, dt64, dt64, dt64], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([np.datetime64("NaT"), dt64, dt64, dt64], dtype="M8[ms]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_infer_homogeoneous_timestamps(self):
+        dt = datetime(2023, 10, 27, 18, 3, 5, 678000)
+        ts = Timestamp(dt).as_unit("ns")
+        arr = np.array([None, ts, ts, ts], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([np.datetime64("NaT")] + [ts.asm8] * 3, dtype="M8[ns]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_infer_homogeoneous_datetimes_strings(self):
+        item = "2023-10-27 18:03:05.678000"
+        arr = np.array([None, item, item, item], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([np.datetime64("NaT"), item, item, item], dtype="M8[us]")
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_infer_heterogeneous(self):
+        dtstr = "2023-10-27 18:03:05.678000"
+
+        arr = np.array([dtstr, dtstr[:-3], dtstr[:-7], None], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array(arr, dtype="M8[us]")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result, tz = tslib.array_to_datetime(arr[::-1], creso=creso_infer)
+        assert tz is None
+        tm.assert_numpy_array_equal(result, expected[::-1])
+
+    @pytest.mark.parametrize(
+        "item", [float("nan"), NaT.value, float(NaT.value), "NaT", ""]
+    )
+    def test_infer_with_nat_int_float_str(self, item):
+        # floats/ints get inferred to nanos *unless* they are NaN/iNaT,
+        # similar NaT string gets treated like NaT scalar (ignored for resolution)
+        dt = datetime(2023, 11, 15, 15, 5, 6)
+
+        arr = np.array([dt, item], dtype=object)
+        result, tz = tslib.array_to_datetime(arr, creso=creso_infer)
+        assert tz is None
+        expected = np.array([dt, np.datetime64("NaT")], dtype="M8[us]")
+        tm.assert_numpy_array_equal(result, expected)
+
+        result2, tz2 = tslib.array_to_datetime(arr[::-1], creso=creso_infer)
+        assert tz2 is None
+        tm.assert_numpy_array_equal(result2, expected[::-1])
+
+
+class TestArrayToDatetimeWithTZResolutionInference:
+    def test_array_to_datetime_with_tz_resolution(self):
+        tz = tzoffset("custom", 3600)
+        vals = np.array(["2016-01-01 02:03:04.567", NaT], dtype=object)
+        res = tslib.array_to_datetime_with_tz(vals, tz, False, False, creso_infer)
+        assert res.dtype == "M8[ms]"
+
+        vals2 = np.array([datetime(2016, 1, 1, 2, 3, 4), NaT], dtype=object)
+        res2 = tslib.array_to_datetime_with_tz(vals2, tz, False, False, creso_infer)
+        assert res2.dtype == "M8[us]"
+
+        vals3 = np.array([NaT, np.datetime64(12345, "s")], dtype=object)
+        res3 = tslib.array_to_datetime_with_tz(vals3, tz, False, False, creso_infer)
+        assert res3.dtype == "M8[s]"
+
+    def test_array_to_datetime_with_tz_resolution_all_nat(self):
+        tz = tzoffset("custom", 3600)
+        vals = np.array(["NaT"], dtype=object)
+        res = tslib.array_to_datetime_with_tz(vals, tz, False, False, creso_infer)
+        assert res.dtype == "M8[s]"
+
+        vals2 = np.array([NaT, NaT], dtype=object)
+        res2 = tslib.array_to_datetime_with_tz(vals2, tz, False, False, creso_infer)
+        assert res2.dtype == "M8[s]"
+
+
+@pytest.mark.parametrize(
+    "data,expected",
+    [
+        (
+            ["01-01-2013", "01-02-2013"],
+            [
+                "2013-01-01T00:00:00.000000000",
+                "2013-01-02T00:00:00.000000000",
+            ],
+        ),
+        (
+            ["Mon Sep 16 2013", "Tue Sep 17 2013"],
+            [
+                "2013-09-16T00:00:00.000000000",
+                "2013-09-17T00:00:00.000000000",
+            ],
+        ),
+    ],
+)
+def test_parsing_valid_dates(data, expected):
+    arr = np.array(data, dtype=object)
+    result, _ = tslib.array_to_datetime(arr)
+
+    expected = np.array(expected, dtype="M8[s]")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "dt_string, expected_tz",
+    [
+        ["01-01-2013 08:00:00+08:00", 480],
+        ["2013-01-01T08:00:00.000000000+0800", 480],
+        ["2012-12-31T16:00:00.000000000-0800", -480],
+        ["12-31-2012 23:00:00-01:00", -60],
+    ],
+)
+def test_parsing_timezone_offsets(dt_string, expected_tz):
+    # All of these datetime strings with offsets are equivalent
+    # to the same datetime after the timezone offset is added.
+    arr = np.array(["01-01-2013 00:00:00"], dtype=object)
+    expected, _ = tslib.array_to_datetime(arr)
+    if "000000000" in dt_string:
+        expected = expected.astype("M8[ns]")
+
+    arr = np.array([dt_string], dtype=object)
+    result, result_tz = tslib.array_to_datetime(arr)
+
+    tm.assert_numpy_array_equal(result, expected)
+    assert result_tz == timezone(timedelta(minutes=expected_tz))
+
+
+def test_parsing_non_iso_timezone_offset():
+    dt_string = "01-01-2013T00:00:00.000000000+0000"
+    arr = np.array([dt_string], dtype=object)
+
+    with tm.assert_produces_warning(None):
+        # GH#50949 should not get tzlocal-deprecation warning here
+        result, result_tz = tslib.array_to_datetime(arr)
+    expected = np.array([np.datetime64("2013-01-01 00:00:00.000000000")])
+
+    tm.assert_numpy_array_equal(result, expected)
+    assert result_tz is timezone.utc
+
+
+def test_parsing_different_timezone_offsets():
+    # see gh-17697
+    data = ["2015-11-18 15:30:00+05:30", "2015-11-18 15:30:00+06:30"]
+    data = np.array(data, dtype=object)
+
+    msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+    with pytest.raises(ValueError, match=msg):
+        tslib.array_to_datetime(data)
+
+
+@pytest.mark.parametrize(
+    "invalid_date,exp_unit",
+    [
+        (date(1000, 1, 1), "s"),
+        (datetime(1000, 1, 1), "us"),
+        ("1000-01-01", "s"),
+        ("Jan 1, 1000", "s"),
+        (np.datetime64("1000-01-01"), "s"),
+    ],
+)
+@pytest.mark.parametrize("errors", ["coerce", "raise"])
+def test_coerce_outside_ns_bounds(invalid_date, exp_unit, errors):
+    arr = np.array([invalid_date], dtype="object")
+
+    result, _ = tslib.array_to_datetime(arr, errors=errors)
+    out_reso = np.datetime_data(result.dtype)[0]
+    assert out_reso == exp_unit
+    ts = Timestamp(invalid_date)
+    assert ts.unit == exp_unit
+
+    expected = np.array([ts._value], dtype=f"M8[{exp_unit}]")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_coerce_outside_ns_bounds_one_valid():
+    arr = np.array(["1/1/1000", "1/1/2000"], dtype=object)
+    result, _ = tslib.array_to_datetime(arr, errors="coerce")
+
+    expected = ["1000-01-01T00:00:00.000000000", "2000-01-01T00:00:00.000000000"]
+    expected = np.array(expected, dtype="M8[s]")
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_coerce_of_invalid_datetimes():
+    arr = np.array(["01-01-2013", "not_a_date", "1"], dtype=object)
+    # With coercing, the invalid dates becomes iNaT
+    result, _ = tslib.array_to_datetime(arr, errors="coerce")
+    expected = ["2013-01-01T00:00:00.000000000", iNaT, iNaT]
+    tm.assert_numpy_array_equal(result, np.array(expected, dtype="M8[s]"))
+
+    # With coercing, the invalid dates becomes iNaT
+    result, _ = tslib.array_to_datetime(arr, errors="coerce")
+    expected = ["2013-01-01T00:00:00.000000000", iNaT, iNaT]
+
+    tm.assert_numpy_array_equal(result, np.array(expected, dtype="M8[s]"))
+
+
+def test_to_datetime_barely_out_of_bounds():
+    # see gh-19382, gh-19529
+    #
+    # Close enough to bounds that dropping nanos
+    # would result in an in-bounds datetime.
+    arr = np.array(["2262-04-11 23:47:16.854775808"], dtype=object)
+    msg = "^Out of bounds nanosecond timestamp: 2262-04-11 23:47:16$"
+
+    with pytest.raises(tslib.OutOfBoundsDatetime, match=msg):
+        tslib.array_to_datetime(arr)
+
+
+@pytest.mark.parametrize(
+    "timestamp",
+    [
+        # Close enough to bounds that scaling micros to nanos overflows
+        # but adding nanos would result in an in-bounds datetime.
+        "1677-09-21T00:12:43.145224193",
+        "1677-09-21T00:12:43.145224999",
+        # this always worked
+        "1677-09-21T00:12:43.145225000",
+    ],
+)
+def test_to_datetime_barely_inside_bounds(timestamp):
+    # see gh-57150
+    result, _ = tslib.array_to_datetime(np.array([timestamp], dtype=object))
+    tm.assert_numpy_array_equal(result, np.array([timestamp], dtype="M8[ns]"))
+
+
+class SubDatetime(datetime):
+    pass
+
+
+@pytest.mark.parametrize("klass", [SubDatetime, datetime, Timestamp])
+def test_datetime_subclass(klass):
+    # GH 25851
+    # ensure that subclassed datetime works with
+    # array_to_datetime
+
+    arr = np.array([klass(2000, 1, 1)], dtype=object)
+    result, _ = tslib.array_to_datetime(arr)
+
+    expected = np.array(["2000-01-01T00:00:00.000000"], dtype="M8[us]")
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/tslibs/test_ccalendar.py b/pandas/pandas/tests/tslibs/test_ccalendar.py
new file mode 100644
index 0000000000000..8dd1bd47e4728
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_ccalendar.py
@@ -0,0 +1,63 @@
+from datetime import (
+    date,
+    datetime,
+)
+
+from hypothesis import given
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import ccalendar
+
+from pandas._testing._hypothesis import DATETIME_IN_PD_TIMESTAMP_RANGE_NO_TZ
+
+
+@pytest.mark.parametrize(
+    "date_tuple,expected",
+    [
+        ((2001, 3, 1), 60),
+        ((2004, 3, 1), 61),
+        ((1907, 12, 31), 365),  # End-of-year, non-leap year.
+        ((2004, 12, 31), 366),  # End-of-year, leap year.
+    ],
+)
+def test_get_day_of_year_numeric(date_tuple, expected):
+    assert ccalendar.get_day_of_year(*date_tuple) == expected
+
+
+def test_get_day_of_year_dt():
+    dt = datetime.fromordinal(1 + np.random.default_rng(2).integers(365 * 4000))
+    result = ccalendar.get_day_of_year(dt.year, dt.month, dt.day)
+
+    expected = (dt - dt.replace(month=1, day=1)).days + 1
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "input_date_tuple, expected_iso_tuple",
+    [
+        [(2020, 1, 1), (2020, 1, 3)],
+        [(2019, 12, 31), (2020, 1, 2)],
+        [(2019, 12, 30), (2020, 1, 1)],
+        [(2009, 12, 31), (2009, 53, 4)],
+        [(2010, 1, 1), (2009, 53, 5)],
+        [(2010, 1, 3), (2009, 53, 7)],
+        [(2010, 1, 4), (2010, 1, 1)],
+        [(2006, 1, 1), (2005, 52, 7)],
+        [(2005, 12, 31), (2005, 52, 6)],
+        [(2008, 12, 28), (2008, 52, 7)],
+        [(2008, 12, 29), (2009, 1, 1)],
+    ],
+)
+def test_dt_correct_iso_8601_year_week_and_day(input_date_tuple, expected_iso_tuple):
+    result = ccalendar.get_iso_calendar(*input_date_tuple)
+    expected_from_date_isocalendar = date(*input_date_tuple).isocalendar()
+    assert result == expected_from_date_isocalendar
+    assert result == expected_iso_tuple
+
+
+@given(DATETIME_IN_PD_TIMESTAMP_RANGE_NO_TZ)
+def test_isocalendar(dt):
+    expected = dt.isocalendar()
+    result = ccalendar.get_iso_calendar(dt.year, dt.month, dt.day)
+    assert result == expected
diff --git a/pandas/pandas/tests/tslibs/test_conversion.py b/pandas/pandas/tests/tslibs/test_conversion.py
new file mode 100644
index 0000000000000..f62910b5e1f1c
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_conversion.py
@@ -0,0 +1,167 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    OutOfBoundsTimedelta,
+    astype_overflowsafe,
+    conversion,
+    iNaT,
+    timezones,
+    tz_convert_from_utc,
+    tzconversion,
+)
+
+from pandas import (
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def _compare_utc_to_local(tz_didx):
+    def f(x):
+        return tzconversion.tz_convert_from_utc_single(x, tz_didx.tz)
+
+    result = tz_convert_from_utc(tz_didx.asi8, tz_didx.tz)
+    expected = np.vectorize(f)(tz_didx.asi8)
+
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def _compare_local_to_utc(tz_didx, naive_didx):
+    # Check that tz_localize behaves the same vectorized and pointwise.
+    err1 = err2 = None
+    try:
+        result = tzconversion.tz_localize_to_utc(naive_didx.asi8, tz_didx.tz)
+        err1 = None
+    except Exception as err:
+        err1 = err
+
+    try:
+        expected = naive_didx.map(lambda x: x.tz_localize(tz_didx.tz)).asi8
+    except Exception as err:
+        err2 = err
+
+    if err1 is not None:
+        assert type(err1) == type(err2)
+    else:
+        assert err2 is None
+        tm.assert_numpy_array_equal(result, expected)
+
+
+def test_tz_localize_to_utc_copies():
+    # GH#46460
+    arr = np.arange(5, dtype="i8")
+    result = tz_convert_from_utc(arr, tz=timezone.utc)
+    tm.assert_numpy_array_equal(result, arr)
+    assert not np.shares_memory(arr, result)
+
+    result = tz_convert_from_utc(arr, tz=None)
+    tm.assert_numpy_array_equal(result, arr)
+    assert not np.shares_memory(arr, result)
+
+
+def test_tz_convert_single_matches_tz_convert_hourly(tz_aware_fixture):
+    tz = tz_aware_fixture
+    tz_didx = date_range("2014-03-01", "2015-01-10", freq="h", tz=tz)
+    naive_didx = date_range("2014-03-01", "2015-01-10", freq="h")
+
+    _compare_utc_to_local(tz_didx)
+    _compare_local_to_utc(tz_didx, naive_didx)
+
+
+@pytest.mark.parametrize("freq", ["D", "YE"])
+def test_tz_convert_single_matches_tz_convert(tz_aware_fixture, freq):
+    tz = tz_aware_fixture
+    tz_didx = date_range("2018-01-01", "2020-01-01", freq=freq, tz=tz)
+    naive_didx = date_range("2018-01-01", "2020-01-01", freq=freq)
+
+    _compare_utc_to_local(tz_didx)
+    _compare_local_to_utc(tz_didx, naive_didx)
+
+
+@pytest.mark.parametrize(
+    "arr",
+    [
+        pytest.param([], id="empty"),
+        pytest.param([iNaT], id="all_nat"),
+    ],
+)
+def test_tz_convert_corner(arr):
+    arr = np.array([iNaT], dtype=np.int64)
+    result = tz_convert_from_utc(arr, timezones.maybe_get_tz("Asia/Tokyo"))
+    tm.assert_numpy_array_equal(result, arr)
+
+
+def test_tz_convert_readonly():
+    # GH#35530
+    arr = np.array([0], dtype=np.int64)
+    arr.setflags(write=False)
+    result = tz_convert_from_utc(arr, timezone.utc)
+    tm.assert_numpy_array_equal(result, arr)
+
+
+@pytest.mark.parametrize("copy", [True, False])
+@pytest.mark.parametrize("dtype", ["M8[ns]", "M8[s]"])
+def test_length_zero_copy(dtype, copy):
+    arr = np.array([], dtype=dtype)
+    result = astype_overflowsafe(arr, copy=copy, dtype=np.dtype("M8[ns]"))
+    if copy:
+        assert not np.shares_memory(result, arr)
+    elif arr.dtype == result.dtype:
+        assert result is arr
+    else:
+        assert not np.shares_memory(result, arr)
+
+
+def test_ensure_datetime64ns_bigendian():
+    # GH#29684
+    arr = np.array([np.datetime64(1, "ms")], dtype=">M8[ms]")
+    result = astype_overflowsafe(arr, dtype=np.dtype("M8[ns]"))
+
+    expected = np.array([np.datetime64(1, "ms")], dtype="M8[ns]")
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_ensure_timedelta64ns_overflows():
+    arr = np.arange(10).astype("m8[Y]") * 100
+    msg = r"Cannot convert 300 years to timedelta64\[ns\] without overflow"
+    with pytest.raises(OutOfBoundsTimedelta, match=msg):
+        astype_overflowsafe(arr, dtype=np.dtype("m8[ns]"))
+
+
+class SubDatetime(datetime):
+    pass
+
+
+@pytest.mark.parametrize(
+    "dt, expected",
+    [
+        pytest.param(
+            Timestamp("2000-01-01"),
+            Timestamp("2000-01-01", tz=timezone.utc),
+            id="timestamp",
+        ),
+        pytest.param(
+            datetime(2000, 1, 1),
+            datetime(2000, 1, 1, tzinfo=timezone.utc),
+            id="datetime",
+        ),
+        pytest.param(
+            SubDatetime(2000, 1, 1),
+            SubDatetime(2000, 1, 1, tzinfo=timezone.utc),
+            id="subclassed_datetime",
+        ),
+    ],
+)
+def test_localize_pydatetime_dt_types(dt, expected):
+    # GH 25851
+    # ensure that subclassed datetime works with
+    # localize_pydatetime
+    result = conversion.localize_pydatetime(dt, timezone.utc)
+    assert result == expected
diff --git a/pandas/pandas/tests/tslibs/test_fields.py b/pandas/pandas/tests/tslibs/test_fields.py
new file mode 100644
index 0000000000000..da67c093b8f4d
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_fields.py
@@ -0,0 +1,40 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import fields
+
+import pandas._testing as tm
+
+
+@pytest.fixture
+def dtindex():
+    dtindex = np.arange(5, dtype=np.int64) * 10**9 * 3600 * 24 * 32
+    dtindex.flags.writeable = False
+    return dtindex
+
+
+def test_get_date_name_field_readonly(dtindex):
+    # https://github.com/vaexio/vaex/issues/357
+    #  fields functions shouldn't raise when we pass read-only data
+    result = fields.get_date_name_field(dtindex, "month_name")
+    expected = np.array(["January", "February", "March", "April", "May"], dtype=object)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_get_date_field_readonly(dtindex):
+    result = fields.get_date_field(dtindex, "Y")
+    expected = np.array([1970, 1970, 1970, 1970, 1970], dtype=np.int32)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_get_start_end_field_readonly(dtindex):
+    result = fields.get_start_end_field(dtindex, "is_month_start", None)
+    expected = np.array([True, False, False, False, False], dtype=np.bool_)
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_get_timedelta_field_readonly(dtindex):
+    # treat dtindex as timedeltas for this next one
+    result = fields.get_timedelta_field(dtindex, "seconds")
+    expected = np.array([0] * 5, dtype=np.int32)
+    tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/tslibs/test_libfrequencies.py b/pandas/pandas/tests/tslibs/test_libfrequencies.py
new file mode 100644
index 0000000000000..effd3b4b8b4e5
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_libfrequencies.py
@@ -0,0 +1,27 @@
+import pytest
+
+from pandas._libs.tslibs.parsing import get_rule_month
+
+from pandas.tseries import offsets
+
+
+@pytest.mark.parametrize(
+    "obj,expected",
+    [
+        ("W", "DEC"),
+        (offsets.Week().freqstr, "DEC"),
+        ("D", "DEC"),
+        (offsets.Day().freqstr, "DEC"),
+        ("Q", "DEC"),
+        (offsets.QuarterEnd(startingMonth=12).freqstr, "DEC"),
+        ("Q-JAN", "JAN"),
+        (offsets.QuarterEnd(startingMonth=1).freqstr, "JAN"),
+        ("Y-DEC", "DEC"),
+        (offsets.YearEnd().freqstr, "DEC"),
+        ("Y-MAY", "MAY"),
+        (offsets.YearEnd(month=5).freqstr, "MAY"),
+    ],
+)
+def test_get_rule_month(obj, expected):
+    result = get_rule_month(obj)
+    assert result == expected
diff --git a/pandas/pandas/tests/tslibs/test_liboffsets.py b/pandas/pandas/tests/tslibs/test_liboffsets.py
new file mode 100644
index 0000000000000..f311284b9dc63
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_liboffsets.py
@@ -0,0 +1,172 @@
+"""
+Tests for helper functions in the cython tslibs.offsets
+"""
+
+from datetime import datetime
+
+import pytest
+
+from pandas._libs.tslibs.ccalendar import (
+    get_firstbday,
+    get_lastbday,
+)
+import pandas._libs.tslibs.offsets as liboffsets
+from pandas._libs.tslibs.offsets import roll_qtrday
+
+from pandas import Timestamp
+
+
+@pytest.fixture(params=["start", "end", "business_start", "business_end"])
+def day_opt(request):
+    return request.param
+
+
+@pytest.mark.parametrize(
+    "dt,exp_week_day,exp_last_day",
+    [
+        (datetime(2017, 11, 30), 3, 30),  # Business day.
+        (datetime(1993, 10, 31), 6, 29),  # Non-business day.
+    ],
+)
+def test_get_last_bday(dt, exp_week_day, exp_last_day):
+    assert dt.weekday() == exp_week_day
+    assert get_lastbday(dt.year, dt.month) == exp_last_day
+
+
+@pytest.mark.parametrize(
+    "dt,exp_week_day,exp_first_day",
+    [
+        (datetime(2017, 4, 1), 5, 3),  # Non-weekday.
+        (datetime(1993, 10, 1), 4, 1),  # Business day.
+    ],
+)
+def test_get_first_bday(dt, exp_week_day, exp_first_day):
+    assert dt.weekday() == exp_week_day
+    assert get_firstbday(dt.year, dt.month) == exp_first_day
+
+
+@pytest.mark.parametrize(
+    "months,day_opt,expected",
+    [
+        (0, 15, datetime(2017, 11, 15)),
+        (0, None, datetime(2017, 11, 30)),
+        (1, "start", datetime(2017, 12, 1)),
+        (-145, "end", datetime(2005, 10, 31)),
+        (0, "business_end", datetime(2017, 11, 30)),
+        (0, "business_start", datetime(2017, 11, 1)),
+    ],
+)
+def test_shift_month_dt(months, day_opt, expected):
+    dt = datetime(2017, 11, 30)
+    assert liboffsets.shift_month(dt, months, day_opt=day_opt) == expected
+
+
+@pytest.mark.parametrize(
+    "months,day_opt,expected",
+    [
+        (1, "start", Timestamp("1929-06-01")),
+        (-3, "end", Timestamp("1929-02-28")),
+        (25, None, Timestamp("1931-06-5")),
+        (-1, 31, Timestamp("1929-04-30")),
+    ],
+)
+def test_shift_month_ts(months, day_opt, expected):
+    ts = Timestamp("1929-05-05")
+    assert liboffsets.shift_month(ts, months, day_opt=day_opt) == expected
+
+
+def test_shift_month_error():
+    dt = datetime(2017, 11, 15)
+    day_opt = "this should raise"
+
+    with pytest.raises(ValueError, match=day_opt):
+        liboffsets.shift_month(dt, 3, day_opt=day_opt)
+
+
+@pytest.mark.parametrize(
+    "other,expected",
+    [
+        # Before March 1.
+        (datetime(2017, 2, 10), {2: 1, -7: -7, 0: 0}),
+        # After March 1.
+        (Timestamp("2014-03-15", tz="US/Eastern"), {2: 2, -7: -6, 0: 1}),
+    ],
+)
+@pytest.mark.parametrize("n", [2, -7, 0])
+def test_roll_qtrday_year(other, expected, n):
+    month = 3
+    day_opt = "start"  # `other` will be compared to March 1.
+
+    assert roll_qtrday(other, n, month, day_opt, modby=12) == expected[n]
+
+
+@pytest.mark.parametrize(
+    "other,expected",
+    [
+        # Before June 30.
+        (datetime(1999, 6, 29), {5: 4, -7: -7, 0: 0}),
+        # After June 30.
+        (Timestamp(2072, 8, 24, 6, 17, 18), {5: 5, -7: -6, 0: 1}),
+    ],
+)
+@pytest.mark.parametrize("n", [5, -7, 0])
+def test_roll_qtrday_year2(other, expected, n):
+    month = 6
+    day_opt = "end"  # `other` will be compared to June 30.
+
+    assert roll_qtrday(other, n, month, day_opt, modby=12) == expected[n]
+
+
+def test_get_day_of_month_error():
+    # get_day_of_month is not directly exposed.
+    # We test it via roll_qtrday.
+    dt = datetime(2017, 11, 15)
+    day_opt = "foo"
+
+    with pytest.raises(ValueError, match=day_opt):
+        # To hit the raising case we need month == dt.month and n > 0.
+        roll_qtrday(dt, n=3, month=11, day_opt=day_opt, modby=12)
+
+
+@pytest.mark.parametrize("month", [3, 5])
+# (other.month % 3) < (month % 3)  # (other.month % 3) > (month % 3)
+@pytest.mark.parametrize("n", [4, -3])
+def test_roll_qtr_day_not_mod_unequal(day_opt, month, n):
+    expected = {3: {-3: -2, 4: 4}, 5: {-3: -3, 4: 3}}
+
+    other = Timestamp(2072, 10, 1, 6, 17, 18)  # Saturday.
+    assert roll_qtrday(other, n, month, day_opt, modby=3) == expected[month][n]
+
+
+@pytest.mark.parametrize(
+    "other,month,exp_dict",
+    [
+        # Monday.
+        (datetime(1999, 5, 31), 2, {-1: {"start": 0, "business_start": 0}}),
+        # Saturday.
+        (
+            Timestamp(2072, 10, 1, 6, 17, 18),
+            4,
+            {2: {"end": 1, "business_end": 1, "business_start": 1}},
+        ),
+        # First business day.
+        (
+            Timestamp(2072, 10, 3, 6, 17, 18),
+            4,
+            {2: {"end": 1, "business_end": 1}, -1: {"start": 0}},
+        ),
+    ],
+)
+@pytest.mark.parametrize("n", [2, -1])
+def test_roll_qtr_day_mod_equal(other, month, exp_dict, n, day_opt):
+    # All cases have (other.month % 3) == (month % 3).
+    expected = exp_dict.get(n, {}).get(day_opt, n)
+    assert roll_qtrday(other, n, month, day_opt, modby=3) == expected
+
+
+@pytest.mark.parametrize(
+    "n,expected", [(42, {29: 42, 1: 42, 31: 41}), (-4, {29: -4, 1: -3, 31: -4})]
+)
+@pytest.mark.parametrize("compare", [29, 1, 31])
+def test_roll_convention(n, expected, compare):
+    assert liboffsets.roll_convention(29, n, compare) == expected[compare]
diff --git a/pandas/pandas/tests/tslibs/test_np_datetime.py b/pandas/pandas/tests/tslibs/test_np_datetime.py
new file mode 100644
index 0000000000000..02edf1a093877
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_np_datetime.py
@@ -0,0 +1,222 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas._libs.tslibs.np_datetime import (
+    OutOfBoundsDatetime,
+    OutOfBoundsTimedelta,
+    astype_overflowsafe,
+    is_unitless,
+    py_get_unit_from_dtype,
+    py_td64_to_tdstruct,
+)
+
+import pandas._testing as tm
+
+
+def test_is_unitless():
+    dtype = np.dtype("M8[ns]")
+    assert not is_unitless(dtype)
+
+    dtype = np.dtype("datetime64")
+    assert is_unitless(dtype)
+
+    dtype = np.dtype("m8[ns]")
+    assert not is_unitless(dtype)
+
+    dtype = np.dtype("timedelta64")
+    assert is_unitless(dtype)
+
+    msg = "dtype must be datetime64 or timedelta64"
+    with pytest.raises(ValueError, match=msg):
+        is_unitless(np.dtype(np.int64))
+
+    msg = "Argument 'dtype' has incorrect type"
+    with pytest.raises(TypeError, match=msg):
+        is_unitless("foo")
+
+
+def test_get_unit_from_dtype():
+    # datetime64
+    assert py_get_unit_from_dtype(np.dtype("M8[Y]")) == NpyDatetimeUnit.NPY_FR_Y.value
+    assert py_get_unit_from_dtype(np.dtype("M8[M]")) == NpyDatetimeUnit.NPY_FR_M.value
+    assert py_get_unit_from_dtype(np.dtype("M8[W]")) == NpyDatetimeUnit.NPY_FR_W.value
+    # B has been deprecated and removed -> no 3
+    assert py_get_unit_from_dtype(np.dtype("M8[D]")) == NpyDatetimeUnit.NPY_FR_D.value
+    assert py_get_unit_from_dtype(np.dtype("M8[h]")) == NpyDatetimeUnit.NPY_FR_h.value
+    assert py_get_unit_from_dtype(np.dtype("M8[m]")) == NpyDatetimeUnit.NPY_FR_m.value
+    assert py_get_unit_from_dtype(np.dtype("M8[s]")) == NpyDatetimeUnit.NPY_FR_s.value
+    assert py_get_unit_from_dtype(np.dtype("M8[ms]")) == NpyDatetimeUnit.NPY_FR_ms.value
+    assert py_get_unit_from_dtype(np.dtype("M8[us]")) == NpyDatetimeUnit.NPY_FR_us.value
+    assert py_get_unit_from_dtype(np.dtype("M8[ns]")) == NpyDatetimeUnit.NPY_FR_ns.value
+    assert py_get_unit_from_dtype(np.dtype("M8[ps]")) == NpyDatetimeUnit.NPY_FR_ps.value
+    assert py_get_unit_from_dtype(np.dtype("M8[fs]")) == NpyDatetimeUnit.NPY_FR_fs.value
+    assert py_get_unit_from_dtype(np.dtype("M8[as]")) == NpyDatetimeUnit.NPY_FR_as.value
+
+    # timedelta64
+    assert py_get_unit_from_dtype(np.dtype("m8[Y]")) == NpyDatetimeUnit.NPY_FR_Y.value
+    assert py_get_unit_from_dtype(np.dtype("m8[M]")) == NpyDatetimeUnit.NPY_FR_M.value
+    assert py_get_unit_from_dtype(np.dtype("m8[W]")) == NpyDatetimeUnit.NPY_FR_W.value
+    # B has been deprecated and removed -> no 3
+    assert py_get_unit_from_dtype(np.dtype("m8[D]")) == NpyDatetimeUnit.NPY_FR_D.value
+    assert py_get_unit_from_dtype(np.dtype("m8[h]")) == NpyDatetimeUnit.NPY_FR_h.value
+    assert py_get_unit_from_dtype(np.dtype("m8[m]")) == NpyDatetimeUnit.NPY_FR_m.value
+    assert py_get_unit_from_dtype(np.dtype("m8[s]")) == NpyDatetimeUnit.NPY_FR_s.value
+    assert py_get_unit_from_dtype(np.dtype("m8[ms]")) == NpyDatetimeUnit.NPY_FR_ms.value
+    assert py_get_unit_from_dtype(np.dtype("m8[us]")) == NpyDatetimeUnit.NPY_FR_us.value
+    assert py_get_unit_from_dtype(np.dtype("m8[ns]")) == NpyDatetimeUnit.NPY_FR_ns.value
+    assert py_get_unit_from_dtype(np.dtype("m8[ps]")) == NpyDatetimeUnit.NPY_FR_ps.value
+    assert py_get_unit_from_dtype(np.dtype("m8[fs]")) == NpyDatetimeUnit.NPY_FR_fs.value
+    assert py_get_unit_from_dtype(np.dtype("m8[as]")) == NpyDatetimeUnit.NPY_FR_as.value
+
+
+def test_td64_to_tdstruct():
+    val = 12454636234  # arbitrary value
+
+    res1 = py_td64_to_tdstruct(val, NpyDatetimeUnit.NPY_FR_ns.value)
+    exp1 = {
+        "days": 0,
+        "hrs": 0,
+        "min": 0,
+        "sec": 12,
+        "ms": 454,
+        "us": 636,
+        "ns": 234,
+        "seconds": 12,
+        "microseconds": 454636,
+        "nanoseconds": 234,
+    }
+    assert res1 == exp1
+
+    res2 = py_td64_to_tdstruct(val, NpyDatetimeUnit.NPY_FR_us.value)
+    exp2 = {
+        "days": 0,
+        "hrs": 3,
+        "min": 27,
+        "sec": 34,
+        "ms": 636,
+        "us": 234,
+        "ns": 0,
+        "seconds": 12454,
+        "microseconds": 636234,
+        "nanoseconds": 0,
+    }
+    assert res2 == exp2
+
+    res3 = py_td64_to_tdstruct(val, NpyDatetimeUnit.NPY_FR_ms.value)
+    exp3 = {
+        "days": 144,
+        "hrs": 3,
+        "min": 37,
+        "sec": 16,
+        "ms": 234,
+        "us": 0,
+        "ns": 0,
+        "seconds": 13036,
+        "microseconds": 234000,
+        "nanoseconds": 0,
+    }
+    assert res3 == exp3
+
+    # Note this out of bounds for nanosecond Timedelta
+    res4 = py_td64_to_tdstruct(val, NpyDatetimeUnit.NPY_FR_s.value)
+    exp4 = {
+        "days": 144150,
+        "hrs": 21,
+        "min": 10,
+        "sec": 34,
+        "ms": 0,
+        "us": 0,
+        "ns": 0,
+        "seconds": 76234,
+        "microseconds": 0,
+        "nanoseconds": 0,
+    }
+    assert res4 == exp4
+
+
+class TestAstypeOverflowSafe:
+    def test_pass_non_dt64_array(self):
+        # check that we raise, not segfault
+        arr = np.arange(5)
+        dtype = np.dtype("M8[ns]")
+
+        msg = (
+            "astype_overflowsafe values.dtype and dtype must be either "
+            "both-datetime64 or both-timedelta64"
+        )
+        with pytest.raises(TypeError, match=msg):
+            astype_overflowsafe(arr, dtype, copy=True)
+
+        with pytest.raises(TypeError, match=msg):
+            astype_overflowsafe(arr, dtype, copy=False)
+
+    def test_pass_non_dt64_dtype(self):
+        # check that we raise, not segfault
+        arr = np.arange(5, dtype="i8").view("M8[D]")
+        dtype = np.dtype("m8[ns]")
+
+        msg = (
+            "astype_overflowsafe values.dtype and dtype must be either "
+            "both-datetime64 or both-timedelta64"
+        )
+        with pytest.raises(TypeError, match=msg):
+            astype_overflowsafe(arr, dtype, copy=True)
+
+        with pytest.raises(TypeError, match=msg):
+            astype_overflowsafe(arr, dtype, copy=False)
+
+    def test_astype_overflowsafe_dt64(self):
+        dtype = np.dtype("M8[ns]")
+
+        dt = np.datetime64("2262-04-05", "D")
+        arr = dt + np.arange(10, dtype="m8[D]")
+
+        # arr.astype silently overflows, so this
+        wrong = arr.astype(dtype)
+        roundtrip = wrong.astype(arr.dtype)
+        assert not (wrong == roundtrip).all()
+
+        msg = "Out of bounds nanosecond timestamp"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            astype_overflowsafe(arr, dtype)
+
+        # But converting to microseconds is fine, and we match numpy's results.
+        dtype2 = np.dtype("M8[us]")
+        result = astype_overflowsafe(arr, dtype2)
+        expected = arr.astype(dtype2)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_overflowsafe_td64(self):
+        dtype = np.dtype("m8[ns]")
+
+        dt = np.datetime64("2262-04-05", "D")
+        arr = dt + np.arange(10, dtype="m8[D]")
+        arr = arr.view("m8[D]")
+
+        # arr.astype silently overflows, so this
+        wrong = arr.astype(dtype)
+        roundtrip = wrong.astype(arr.dtype)
+        assert not (wrong == roundtrip).all()
+
+        msg = r"Cannot convert 106752 days to timedelta64\[ns\] without overflow"
+        with pytest.raises(OutOfBoundsTimedelta, match=msg):
+            astype_overflowsafe(arr, dtype)
+
+        # But converting to microseconds is fine, and we match numpy's results.
+        dtype2 = np.dtype("m8[us]")
+        result = astype_overflowsafe(arr, dtype2)
+        expected = arr.astype(dtype2)
+        tm.assert_numpy_array_equal(result, expected)
+
+    def test_astype_overflowsafe_disallow_rounding(self):
+        arr = np.array([-1500, 1500], dtype="M8[ns]")
+        dtype = np.dtype("M8[us]")
+
+        msg = "Cannot losslessly cast '-1500 ns' to us"
+        with pytest.raises(ValueError, match=msg):
+            astype_overflowsafe(arr, dtype, round_ok=False)
+
+        result = astype_overflowsafe(arr, dtype, round_ok=True)
+        expected = arr.astype(dtype)
+        tm.assert_numpy_array_equal(result, expected)
diff --git a/pandas/pandas/tests/tslibs/test_npy_units.py b/pandas/pandas/tests/tslibs/test_npy_units.py
new file mode 100644
index 0000000000000..6d05dc79fbb2c
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_npy_units.py
@@ -0,0 +1,27 @@
+import numpy as np
+
+from pandas._libs.tslibs.dtypes import abbrev_to_npy_unit
+from pandas._libs.tslibs.vectorized import is_date_array_normalized
+
+# a datetime64 ndarray which *is* normalized
+day_arr = np.arange(10, dtype="i8").view("M8[D]")
+
+
+class TestIsDateArrayNormalized:
+    def test_is_date_array_normalized_day(self):
+        arr = day_arr
+        abbrev = "D"
+        unit = abbrev_to_npy_unit(abbrev)
+        result = is_date_array_normalized(arr.view("i8"), None, unit)
+        assert result is True
+
+    def test_is_date_array_normalized_seconds(self):
+        abbrev = "s"
+        arr = day_arr.astype(f"M8[{abbrev}]")
+        unit = abbrev_to_npy_unit(abbrev)
+        result = is_date_array_normalized(arr.view("i8"), None, unit)
+        assert result is True
+
+        arr[0] += np.timedelta64(1, abbrev)
+        result2 = is_date_array_normalized(arr.view("i8"), None, unit)
+        assert result2 is False
diff --git a/pandas/pandas/tests/tslibs/test_parse_iso8601.py b/pandas/pandas/tests/tslibs/test_parse_iso8601.py
new file mode 100644
index 0000000000000..1992faae2ea6a
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_parse_iso8601.py
@@ -0,0 +1,119 @@
+from datetime import datetime
+
+import pytest
+
+from pandas._libs import tslib
+
+from pandas import Timestamp
+
+
+@pytest.mark.parametrize(
+    "date_str, exp",
+    [
+        ("2011-01-02", datetime(2011, 1, 2)),
+        ("2011-1-2", datetime(2011, 1, 2)),
+        ("2011-01", datetime(2011, 1, 1)),
+        ("2011-1", datetime(2011, 1, 1)),
+        ("2011 01 02", datetime(2011, 1, 2)),
+        ("2011.01.02", datetime(2011, 1, 2)),
+        ("2011/01/02", datetime(2011, 1, 2)),
+        ("2011\\01\\02", datetime(2011, 1, 2)),
+        ("2013-01-01 05:30:00", datetime(2013, 1, 1, 5, 30)),
+        ("2013-1-1 5:30:00", datetime(2013, 1, 1, 5, 30)),
+        ("2013-1-1 5:30:00+01:00", Timestamp(2013, 1, 1, 5, 30, tz="UTC+01:00")),
+    ],
+)
+def test_parsers_iso8601(date_str, exp):
+    # see gh-12060
+    #
+    # Test only the ISO parser - flexibility to
+    # different separators and leading zero's.
+    actual = tslib._test_parse_iso8601(date_str)
+    assert actual == exp
+
+
+@pytest.mark.parametrize(
+    "date_str",
+    [
+        "2011-01/02",
+        "2011=11=11",
+        "201401",
+        "201111",
+        "200101",
+        # Mixed separated and unseparated.
+        "2005-0101",
+        "200501-01",
+        "20010101 12:3456",
+        "20010101 1234:56",
+        # HHMMSS must have two digits in
+        # each component if unseparated.
+        "20010101 1",
+        "20010101 123",
+        "20010101 12345",
+        "20010101 12345Z",
+    ],
+)
+def test_parsers_iso8601_invalid(date_str):
+    msg = f'Error parsing datetime string "{date_str}"'
+
+    with pytest.raises(ValueError, match=msg):
+        tslib._test_parse_iso8601(date_str)
+
+
+def test_parsers_iso8601_invalid_offset_invalid():
+    date_str = "2001-01-01 12-34-56"
+    msg = f'Timezone hours offset out of range in datetime string "{date_str}"'
+
+    with pytest.raises(ValueError, match=msg):
+        tslib._test_parse_iso8601(date_str)
+
+
+def test_parsers_iso8601_leading_space():
+    # GH#25895 make sure isoparser doesn't overflow with long input
+    date_str, expected = ("2013-1-1 5:30:00", datetime(2013, 1, 1, 5, 30))
+    actual = tslib._test_parse_iso8601(" " * 200 + date_str)
+    assert actual == expected
+
+
+@pytest.mark.parametrize(
+    "date_str, timespec, exp",
+    [
+        ("2023-01-01 00:00:00", "auto", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00", "seconds", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00", "milliseconds", "2023-01-01T00:00:00.000"),
+        ("2023-01-01 00:00:00", "microseconds", "2023-01-01T00:00:00.000000"),
+        ("2023-01-01 00:00:00", "nanoseconds", "2023-01-01T00:00:00.000000000"),
+        ("2023-01-01 00:00:00.001", "auto", "2023-01-01T00:00:00.001000"),
+        ("2023-01-01 00:00:00.001", "seconds", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00.001", "milliseconds", "2023-01-01T00:00:00.001"),
+        ("2023-01-01 00:00:00.001", "microseconds", "2023-01-01T00:00:00.001000"),
+        ("2023-01-01 00:00:00.001", "nanoseconds", "2023-01-01T00:00:00.001000000"),
+        ("2023-01-01 00:00:00.000001", "auto", "2023-01-01T00:00:00.000001"),
+        ("2023-01-01 00:00:00.000001", "seconds", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00.000001", "milliseconds", "2023-01-01T00:00:00.000"),
+        ("2023-01-01 00:00:00.000001", "microseconds", "2023-01-01T00:00:00.000001"),
+        ("2023-01-01 00:00:00.000001", "nanoseconds", "2023-01-01T00:00:00.000001000"),
+        ("2023-01-01 00:00:00.000000001", "auto", "2023-01-01T00:00:00.000000001"),
+        ("2023-01-01 00:00:00.000000001", "seconds", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00.000000001", "milliseconds", "2023-01-01T00:00:00.000"),
+        ("2023-01-01 00:00:00.000000001", "microseconds", "2023-01-01T00:00:00.000000"),
+        (
+            "2023-01-01 00:00:00.000000001",
+            "nanoseconds",
+            "2023-01-01T00:00:00.000000001",
+        ),
+        ("2023-01-01 00:00:00.000001001", "auto", "2023-01-01T00:00:00.000001001"),
+        ("2023-01-01 00:00:00.000001001", "seconds", "2023-01-01T00:00:00"),
+        ("2023-01-01 00:00:00.000001001", "milliseconds", "2023-01-01T00:00:00.000"),
+        ("2023-01-01 00:00:00.000001001", "microseconds", "2023-01-01T00:00:00.000001"),
+        (
+            "2023-01-01 00:00:00.000001001",
+            "nanoseconds",
+            "2023-01-01T00:00:00.000001001",
+        ),
+    ],
+)
+def test_iso8601_formatter(date_str: str, timespec: str, exp: str):
+    # GH#53020
+    ts = Timestamp(date_str)
+    assert ts.isoformat(timespec=timespec) == exp
diff --git a/pandas/pandas/tests/tslibs/test_parsing.py b/pandas/pandas/tests/tslibs/test_parsing.py
new file mode 100644
index 0000000000000..bc5cd5fcccbf8
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_parsing.py
@@ -0,0 +1,422 @@
+"""
+Tests for Timestamp parsing, aimed at pandas/_libs/tslibs/parsing.pyx
+"""
+
+from datetime import datetime
+import re
+
+from dateutil.parser import parse as du_parse
+from hypothesis import given
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    parsing,
+    strptime,
+)
+from pandas._libs.tslibs.parsing import parse_datetime_string_with_reso
+from pandas.compat import (
+    ISMUSL,
+    WASM,
+    is_platform_windows,
+)
+import pandas.util._test_decorators as td
+
+# Usually we wouldn't want this import in this test file (which is targeted at
+#  tslibs.parsing), but it is convenient to test the Timestamp constructor at
+#  the same time as the other parsing functions.
+from pandas import Timestamp
+import pandas._testing as tm
+from pandas._testing._hypothesis import DATETIME_NO_TZ
+
+
+@pytest.mark.skipif(WASM, reason="tzset is not available on WASM")
+@pytest.mark.skipif(
+    is_platform_windows() or ISMUSL,
+    reason="TZ setting incorrect on Windows and MUSL Linux",
+)
+def test_parsing_tzlocal_deprecated():
+    # GH#50791
+    msg = "|".join(
+        [
+            r"Parsing 'EST' as tzlocal \(dependent on system timezone\) "
+            r"is no longer supported\. "
+            "Pass the 'tz' keyword or call tz_localize after construction instead",
+            ".*included an un-recognized timezone",
+        ]
+    )
+    dtstr = "Jan 15 2004 03:00 EST"
+
+    with tm.set_timezone("US/Eastern"):
+        with pytest.raises(ValueError, match=msg):
+            parse_datetime_string_with_reso(dtstr)
+
+        with pytest.raises(ValueError, match=msg):
+            parsing.py_parse_datetime_string(dtstr)
+
+        with pytest.raises(ValueError, match=msg):
+            Timestamp(dtstr)
+
+
+def test_parse_datetime_string_with_reso():
+    (parsed, reso) = parse_datetime_string_with_reso("4Q1984")
+    (parsed_lower, reso_lower) = parse_datetime_string_with_reso("4q1984")
+
+    assert reso == reso_lower
+    assert parsed == parsed_lower
+
+
+def test_parse_datetime_string_with_reso_nanosecond_reso():
+    # GH#46811
+    parsed, reso = parse_datetime_string_with_reso("2022-04-20 09:19:19.123456789")
+    assert reso == "nanosecond"
+
+
+def test_parse_datetime_string_with_reso_invalid_type():
+    # Raise on invalid input, don't just return it
+    msg = "Argument 'date_string' has incorrect type (expected str, got tuple)"
+    with pytest.raises(TypeError, match=re.escape(msg)):
+        parse_datetime_string_with_reso((4, 5))
+
+
+@pytest.mark.parametrize(
+    "dashed,normal", [("1988-Q2", "1988Q2"), ("2Q-1988", "2Q1988")]
+)
+def test_parse_time_quarter_with_dash(dashed, normal):
+    # see gh-9688
+    (parsed_dash, reso_dash) = parse_datetime_string_with_reso(dashed)
+    (parsed, reso) = parse_datetime_string_with_reso(normal)
+
+    assert parsed_dash == parsed
+    assert reso_dash == reso
+
+
+@pytest.mark.parametrize("dashed", ["-2Q1992", "2-Q1992", "4-4Q1992"])
+def test_parse_time_quarter_with_dash_error(dashed):
+    msg = f"Unknown datetime string format, unable to parse: {dashed}"
+
+    with pytest.raises(parsing.DateParseError, match=msg):
+        parse_datetime_string_with_reso(dashed)
+
+
+@pytest.mark.parametrize(
+    "date_string,expected",
+    [
+        ("123.1234", False),
+        ("-50000", False),
+        ("999", False),
+        ("m", False),
+        ("T", False),
+        ("Mon Sep 16, 2013", True),
+        ("2012-01-01", True),
+        ("01/01/2012", True),
+        ("01012012", True),
+        ("0101", True),
+        ("1-1", True),
+    ],
+)
+def test_does_not_convert_mixed_integer(date_string, expected):
+    assert parsing._does_string_look_like_datetime(date_string) is expected
+
+
+@pytest.mark.parametrize(
+    "date_str,kwargs,msg",
+    [
+        (
+            "2013Q5",
+            {},
+            (
+                "Incorrect quarterly string is given, "
+                "quarter must be between 1 and 4: 2013Q5"
+            ),
+        ),
+        # see gh-5418
+        (
+            "2013Q1",
+            {"freq": "INVLD-L-DEC-SAT"},
+            ("Unable to retrieve month information from given freq: INVLD-L-DEC-SAT"),
+        ),
+    ],
+)
+def test_parsers_quarterly_with_freq_error(date_str, kwargs, msg):
+    with pytest.raises(parsing.DateParseError, match=msg):
+        parsing.parse_datetime_string_with_reso(date_str, **kwargs)
+
+
+@pytest.mark.parametrize(
+    "date_str,freq,expected",
+    [
+        ("2013Q2", None, datetime(2013, 4, 1)),
+        ("2013Q2", "Y-APR", datetime(2012, 8, 1)),
+        ("2013-Q2", "Y-DEC", datetime(2013, 4, 1)),
+    ],
+)
+def test_parsers_quarterly_with_freq(date_str, freq, expected):
+    result, _ = parsing.parse_datetime_string_with_reso(date_str, freq=freq)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "date_str", ["2Q 2005", "2Q-200Y", "2Q-200", "22Q2005", "2Q200.", "6Q-20"]
+)
+def test_parsers_quarter_invalid(date_str):
+    if date_str == "6Q-20":
+        msg = (
+            "Incorrect quarterly string is given, quarter "
+            f"must be between 1 and 4: {date_str}"
+        )
+    else:
+        msg = f"Unknown datetime string format, unable to parse: {date_str}"
+
+    with pytest.raises(ValueError, match=msg):
+        parsing.parse_datetime_string_with_reso(date_str)
+
+
+@pytest.mark.parametrize(
+    "date_str,expected",
+    [("201101", datetime(2011, 1, 1, 0, 0)), ("200005", datetime(2000, 5, 1, 0, 0))],
+)
+def test_parsers_month_freq(date_str, expected):
+    result, _ = parsing.parse_datetime_string_with_reso(date_str, freq="ME")
+    assert result == expected
+
+
+@td.skip_if_not_us_locale
+@pytest.mark.parametrize(
+    "string,fmt",
+    [
+        ("20111230", "%Y%m%d"),
+        ("201112300000", "%Y%m%d%H%M"),
+        ("20111230000000", "%Y%m%d%H%M%S"),
+        ("20111230T00", "%Y%m%dT%H"),
+        ("20111230T0000", "%Y%m%dT%H%M"),
+        ("20111230T000000", "%Y%m%dT%H%M%S"),
+        ("2011-12-30", "%Y-%m-%d"),
+        ("2011", "%Y"),
+        ("2011-01", "%Y-%m"),
+        ("30-12-2011", "%d-%m-%Y"),
+        ("2011-12-30 00:00:00", "%Y-%m-%d %H:%M:%S"),
+        ("2011-12-30T00:00:00", "%Y-%m-%dT%H:%M:%S"),
+        ("2011-12-30T00:00:00UTC", "%Y-%m-%dT%H:%M:%S%Z"),
+        ("2011-12-30T00:00:00Z", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+9", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+09", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+090", None),
+        ("2011-12-30T00:00:00+0900", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00-0900", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+09:00", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+09:000", None),
+        ("2011-12-30T00:00:00+9:0", "%Y-%m-%dT%H:%M:%S%z"),
+        ("2011-12-30T00:00:00+09:", None),
+        ("2011-12-30T00:00:00.000000UTC", "%Y-%m-%dT%H:%M:%S.%f%Z"),
+        ("2011-12-30T00:00:00.000000Z", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+9", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+09", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+090", None),
+        ("2011-12-30T00:00:00.000000+0900", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000-0900", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+09:00", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+09:000", None),
+        ("2011-12-30T00:00:00.000000+9:0", "%Y-%m-%dT%H:%M:%S.%f%z"),
+        ("2011-12-30T00:00:00.000000+09:", None),
+        ("2011-12-30 00:00:00.000000", "%Y-%m-%d %H:%M:%S.%f"),
+        ("Tue 24 Aug 2021 01:30:48", "%a %d %b %Y %H:%M:%S"),
+        ("Tuesday 24 Aug 2021 01:30:48", "%A %d %b %Y %H:%M:%S"),
+        ("Tue 24 Aug 2021 01:30:48 AM", "%a %d %b %Y %I:%M:%S %p"),
+        ("Tuesday 24 Aug 2021 01:30:48 AM", "%A %d %b %Y %I:%M:%S %p"),
+        ("27.03.2003 14:55:00.000", "%d.%m.%Y %H:%M:%S.%f"),  # GH50317
+        ("2023-11-09T20:23:46Z", "%Y-%m-%dT%H:%M:%S%z"),  # GH57452
+    ],
+)
+def test_guess_datetime_format_with_parseable_formats(string, fmt):
+    with tm.maybe_produces_warning(
+        UserWarning, fmt is not None and re.search(r"%d.*%m", fmt)
+    ):
+        result = parsing.guess_datetime_format(string)
+    assert result == fmt
+
+
+@pytest.mark.parametrize("dayfirst,expected", [(True, "%d/%m/%Y"), (False, "%m/%d/%Y")])
+def test_guess_datetime_format_with_dayfirst(dayfirst, expected):
+    ambiguous_string = "01/01/2011"
+    result = parsing.guess_datetime_format(ambiguous_string, dayfirst=dayfirst)
+    assert result == expected
+
+
+@td.skip_if_not_us_locale
+@pytest.mark.parametrize(
+    "string,fmt",
+    [
+        ("30/Dec/2011", "%d/%b/%Y"),
+        ("30/December/2011", "%d/%B/%Y"),
+        ("30/Dec/2011 00:00:00", "%d/%b/%Y %H:%M:%S"),
+    ],
+)
+def test_guess_datetime_format_with_locale_specific_formats(string, fmt):
+    result = parsing.guess_datetime_format(string)
+    assert result == fmt
+
+
+@pytest.mark.parametrize(
+    "invalid_dt",
+    [
+        "01/2013",
+        "12:00:00",
+        "1/1/1/1",
+        "this_is_not_a_datetime",
+        "51a",
+        "13/2019",
+        "202001",  # YYYYMM isn't ISO8601
+        "2020/01",  # YYYY/MM isn't ISO8601 either
+        "87156549591102612381000001219H5",
+    ],
+)
+def test_guess_datetime_format_invalid_inputs(invalid_dt):
+    # A datetime string must include a year, month and a day for it to be
+    # guessable, in addition to being a string that looks like a datetime.
+    assert parsing.guess_datetime_format(invalid_dt) is None
+
+
+@pytest.mark.parametrize("invalid_type_dt", [9, datetime(2011, 1, 1)])
+def test_guess_datetime_format_wrong_type_inputs(invalid_type_dt):
+    # A datetime string must include a year, month and a day for it to be
+    # guessable, in addition to being a string that looks like a datetime.
+    with pytest.raises(
+        TypeError,
+        match=r"^Argument 'dt_str' has incorrect type \(expected str, got .*\)$",
+    ):
+        parsing.guess_datetime_format(invalid_type_dt)
+
+
+@pytest.mark.parametrize(
+    "string,fmt,dayfirst,warning",
+    [
+        ("2011-1-1", "%Y-%m-%d", False, None),
+        ("2011-1-1", "%Y-%d-%m", True, None),
+        ("1/1/2011", "%m/%d/%Y", False, None),
+        ("1/1/2011", "%d/%m/%Y", True, None),
+        ("30-1-2011", "%d-%m-%Y", False, UserWarning),
+        ("30-1-2011", "%d-%m-%Y", True, None),
+        ("2011-1-1 0:0:0", "%Y-%m-%d %H:%M:%S", False, None),
+        ("2011-1-1 0:0:0", "%Y-%d-%m %H:%M:%S", True, None),
+        ("2011-1-3T00:00:0", "%Y-%m-%dT%H:%M:%S", False, None),
+        ("2011-1-3T00:00:0", "%Y-%d-%mT%H:%M:%S", True, None),
+        ("2011-1-1 00:00:00", "%Y-%m-%d %H:%M:%S", False, None),
+        ("2011-1-1 00:00:00", "%Y-%d-%m %H:%M:%S", True, None),
+    ],
+)
+def test_guess_datetime_format_no_padding(string, fmt, dayfirst, warning):
+    # see gh-11142
+    msg = (
+        rf"Parsing dates in {fmt} format when dayfirst=False \(the default\) "
+        "was specified. "
+        "Pass `dayfirst=True` or specify a format to silence this warning."
+    )
+    with tm.assert_produces_warning(warning, match=msg):
+        result = parsing.guess_datetime_format(string, dayfirst=dayfirst)
+    assert result == fmt
+
+
+def test_try_parse_dates():
+    arr = np.array(["5/1/2000", "6/1/2000", "7/1/2000"], dtype=object)
+    result = parsing.try_parse_dates(arr, parser=lambda x: du_parse(x, dayfirst=True))
+
+    expected = np.array([du_parse(d, dayfirst=True) for d in arr])
+    tm.assert_numpy_array_equal(result, expected)
+
+
+def test_parse_datetime_string_with_reso_check_instance_type_raise_exception():
+    # issue 20684
+    msg = "Argument 'date_string' has incorrect type (expected str, got tuple)"
+    with pytest.raises(TypeError, match=re.escape(msg)):
+        parse_datetime_string_with_reso((1, 2, 3))
+
+    result = parse_datetime_string_with_reso("2019")
+    expected = (datetime(2019, 1, 1), "year")
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "fmt,expected",
+    [
+        ("%Y %m %d %H:%M:%S", True),
+        ("%Y/%m/%d %H:%M:%S", True),
+        (r"%Y\%m\%d %H:%M:%S", True),
+        ("%Y-%m-%d %H:%M:%S", True),
+        ("%Y.%m.%d %H:%M:%S", True),
+        ("%Y%m%d %H:%M:%S", True),
+        ("%Y-%m-%dT%H:%M:%S", True),
+        ("%Y-%m-%dT%H:%M:%S%z", True),
+        ("%Y-%m-%dT%H:%M:%S%Z", False),
+        ("%Y-%m-%dT%H:%M:%S.%f", True),
+        ("%Y-%m-%dT%H:%M:%S.%f%z", True),
+        ("%Y-%m-%dT%H:%M:%S.%f%Z", False),
+        ("%Y%m%d", True),
+        ("%Y%m", False),
+        ("%Y", True),
+        ("%Y-%m-%d", True),
+        ("%Y-%m", True),
+    ],
+)
+def test_is_iso_format(fmt, expected):
+    # see gh-41047
+    result = strptime._test_format_is_iso(fmt)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "input",
+    [
+        "2018-01-01T00:00:00.123456789",
+        "2018-01-01T00:00:00.123456",
+        "2018-01-01T00:00:00.123",
+    ],
+)
+def test_guess_datetime_format_f(input):
+    # https://github.com/pandas-dev/pandas/issues/49043
+    result = parsing.guess_datetime_format(input)
+    expected = "%Y-%m-%dT%H:%M:%S.%f"
+    assert result == expected
+
+
+def _helper_hypothesis_delimited_date(call, date_string, **kwargs):
+    msg, result = None, None
+    try:
+        result = call(date_string, **kwargs)
+    except ValueError as err:
+        msg = str(err)
+    return msg, result
+
+
+@given(DATETIME_NO_TZ)
+@pytest.mark.parametrize("delimiter", list(" -./"))
+@pytest.mark.parametrize("dayfirst", [True, False])
+@pytest.mark.parametrize(
+    "date_format",
+    ["%d %m %Y", "%m %d %Y", "%m %Y", "%Y %m %d", "%y %m %d", "%Y%m%d", "%y%m%d"],
+)
+def test_hypothesis_delimited_date(
+    request, date_format, dayfirst, delimiter, test_datetime
+):
+    if date_format == "%m %Y" and delimiter == ".":
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="parse_datetime_string cannot reliably tell whether "
+                "e.g. %m.%Y is a float or a date"
+            )
+        )
+    date_string = test_datetime.strftime(date_format.replace(" ", delimiter))
+
+    except_out_dateutil, result = _helper_hypothesis_delimited_date(
+        parsing.py_parse_datetime_string, date_string, dayfirst=dayfirst
+    )
+    except_in_dateutil, expected = _helper_hypothesis_delimited_date(
+        du_parse,
+        date_string,
+        default=datetime(1, 1, 1),
+        dayfirst=dayfirst,
+        yearfirst=False,
+    )
+
+    assert except_out_dateutil == except_in_dateutil
+    assert result == expected
diff --git a/pandas/pandas/tests/tslibs/test_period.py b/pandas/pandas/tests/tslibs/test_period.py
new file mode 100644
index 0000000000000..715e2d3da88db
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_period.py
@@ -0,0 +1,123 @@
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    iNaT,
+    to_offset,
+)
+from pandas._libs.tslibs.period import (
+    extract_ordinals,
+    get_period_field_arr,
+    period_asfreq,
+    period_ordinal,
+)
+
+import pandas._testing as tm
+
+
+def get_freq_code(freqstr: str) -> int:
+    off = to_offset(freqstr, is_period=True)
+    # error: "BaseOffset" has no attribute "_period_dtype_code"
+    code = off._period_dtype_code  # type: ignore[attr-defined]
+    return code
+
+
+@pytest.mark.parametrize(
+    "freq1,freq2,expected",
+    [
+        ("D", "h", 24),
+        ("D", "min", 1440),
+        ("D", "s", 86400),
+        ("D", "ms", 86400000),
+        ("D", "us", 86400000000),
+        ("D", "ns", 86400000000000),
+        ("h", "min", 60),
+        ("h", "s", 3600),
+        ("h", "ms", 3600000),
+        ("h", "us", 3600000000),
+        ("h", "ns", 3600000000000),
+        ("min", "s", 60),
+        ("min", "ms", 60000),
+        ("min", "us", 60000000),
+        ("min", "ns", 60000000000),
+        ("s", "ms", 1000),
+        ("s", "us", 1000000),
+        ("s", "ns", 1000000000),
+        ("ms", "us", 1000),
+        ("ms", "ns", 1000000),
+        ("us", "ns", 1000),
+    ],
+)
+def test_intra_day_conversion_factors(freq1, freq2, expected):
+    assert (
+        period_asfreq(1, get_freq_code(freq1), get_freq_code(freq2), False) == expected
+    )
+
+
+@pytest.mark.parametrize(
+    "freq,expected", [("Y", 0), ("M", 0), ("W", 1), ("D", 0), ("B", 0)]
+)
+def test_period_ordinal_start_values(freq, expected):
+    # information for Jan. 1, 1970.
+    assert period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq_code(freq)) == expected
+
+
+@pytest.mark.parametrize(
+    "dt,expected",
+    [
+        ((1970, 1, 4, 0, 0, 0, 0, 0), 1),
+        ((1970, 1, 5, 0, 0, 0, 0, 0), 2),
+        ((2013, 10, 6, 0, 0, 0, 0, 0), 2284),
+        ((2013, 10, 7, 0, 0, 0, 0, 0), 2285),
+    ],
+)
+def test_period_ordinal_week(dt, expected):
+    args = dt + (get_freq_code("W"),)
+    assert period_ordinal(*args) == expected
+
+
+@pytest.mark.parametrize(
+    "day,expected",
+    [
+        # Thursday (Oct. 3, 2013).
+        (3, 11415),
+        # Friday (Oct. 4, 2013).
+        (4, 11416),
+        # Saturday (Oct. 5, 2013).
+        (5, 11417),
+        # Sunday (Oct. 6, 2013).
+        (6, 11417),
+        # Monday (Oct. 7, 2013).
+        (7, 11417),
+        # Tuesday (Oct. 8, 2013).
+        (8, 11418),
+    ],
+)
+def test_period_ordinal_business_day(day, expected):
+    # 5000 is PeriodDtypeCode for BusinessDay
+    args = (2013, 10, day, 0, 0, 0, 0, 0, 5000)
+    assert period_ordinal(*args) == expected
+
+
+class TestExtractOrdinals:
+    def test_extract_ordinals_raises(self):
+        # with non-object, make sure we raise TypeError, not segfault
+        arr = np.arange(5)
+        freq = to_offset("D")
+        with pytest.raises(TypeError, match="values must be object-dtype"):
+            extract_ordinals(arr, freq)
+
+    def test_extract_ordinals_2d(self):
+        freq = to_offset("D")
+        arr = np.empty(10, dtype=object)
+        arr[:] = iNaT
+
+        res = extract_ordinals(arr, freq)
+        res2 = extract_ordinals(arr.reshape(5, 2), freq)
+        tm.assert_numpy_array_equal(res, res2.reshape(-1))
+
+
+def test_get_period_field_array_raises_on_out_of_range():
+    msg = "Buffer dtype mismatch, expected 'const int64_t' but got 'double'"
+    with pytest.raises(ValueError, match=msg):
+        get_period_field_arr(-1, np.empty(1), 0)
diff --git a/pandas/pandas/tests/tslibs/test_resolution.py b/pandas/pandas/tests/tslibs/test_resolution.py
new file mode 100644
index 0000000000000..0e7705ad7ed94
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_resolution.py
@@ -0,0 +1,56 @@
+import datetime
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs import (
+    Resolution,
+    get_resolution,
+)
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+
+
+def test_get_resolution_nano():
+    # don't return the fallback RESO_DAY
+    arr = np.array([1], dtype=np.int64)
+    res = get_resolution(arr)
+    assert res == Resolution.RESO_NS
+
+
+def test_get_resolution_non_nano_data():
+    arr = np.array([1], dtype=np.int64)
+    res = get_resolution(arr, None, NpyDatetimeUnit.NPY_FR_us.value)
+    assert res == Resolution.RESO_US
+
+    res = get_resolution(arr, datetime.timezone.utc, NpyDatetimeUnit.NPY_FR_us.value)
+    assert res == Resolution.RESO_US
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected",
+    [
+        ("Y", "year"),
+        ("Q", "quarter"),
+        ("M", "month"),
+        ("D", "day"),
+        ("h", "hour"),
+        ("min", "minute"),
+        ("s", "second"),
+        ("ms", "millisecond"),
+        ("us", "microsecond"),
+        ("ns", "nanosecond"),
+    ],
+)
+def test_get_attrname_from_abbrev(freqstr, expected):
+    reso = Resolution.get_reso_from_freqstr(freqstr)
+    assert reso.attr_abbrev == freqstr
+    assert reso.attrname == expected
+
+
+@pytest.mark.parametrize("freq", ["H", "S"])
+def test_unit_H_S_raises(freq):
+    # GH#59143
+    msg = f"Invalid frequency: {freq}"
+
+    with pytest.raises(ValueError, match=msg):
+        Resolution.get_reso_from_freqstr(freq)
diff --git a/pandas/pandas/tests/tslibs/test_strptime.py b/pandas/pandas/tests/tslibs/test_strptime.py
new file mode 100644
index 0000000000000..d726006b03f6d
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_strptime.py
@@ -0,0 +1,110 @@
+from datetime import (
+    datetime,
+    timezone,
+)
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.dtypes import NpyDatetimeUnit
+from pandas._libs.tslibs.strptime import array_strptime
+
+from pandas import (
+    NaT,
+    Timestamp,
+)
+import pandas._testing as tm
+
+creso_infer = NpyDatetimeUnit.NPY_FR_GENERIC.value
+
+
+class TestArrayStrptimeResolutionInference:
+    def test_array_strptime_resolution_all_nat(self):
+        arr = np.array([NaT, np.nan], dtype=object)
+
+        fmt = "%Y-%m-%d %H:%M:%S"
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        assert res.dtype == "M8[s]"
+
+        res, _ = array_strptime(arr, fmt=fmt, utc=True, creso=creso_infer)
+        assert res.dtype == "M8[s]"
+
+    @pytest.mark.parametrize("tz", [None, timezone.utc])
+    def test_array_strptime_resolution_inference_homogeneous_strings(self, tz):
+        dt = datetime(2016, 1, 2, 3, 4, 5, 678900, tzinfo=tz)
+
+        fmt = "%Y-%m-%d %H:%M:%S"
+        dtstr = dt.strftime(fmt)
+        arr = np.array([dtstr] * 3, dtype=object)
+        expected = np.array([dt.replace(tzinfo=None)] * 3, dtype="M8[s]")
+
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+        fmt = "%Y-%m-%d %H:%M:%S.%f"
+        dtstr = dt.strftime(fmt)
+        arr = np.array([dtstr] * 3, dtype=object)
+        expected = np.array([dt.replace(tzinfo=None)] * 3, dtype="M8[us]")
+
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+        fmt = "ISO8601"
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+    @pytest.mark.parametrize("tz", [None, timezone.utc])
+    def test_array_strptime_resolution_mixed(self, tz):
+        dt = datetime(2016, 1, 2, 3, 4, 5, 678900, tzinfo=tz)
+
+        ts = Timestamp(dt).as_unit("ns")
+
+        arr = np.array([dt, ts], dtype=object)
+        expected = np.array(
+            [Timestamp(dt).as_unit("ns").asm8, ts.asm8],
+            dtype="M8[ns]",
+        )
+
+        fmt = "%Y-%m-%d %H:%M:%S"
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+        fmt = "ISO8601"
+        res, _ = array_strptime(arr, fmt=fmt, utc=False, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+    def test_array_strptime_resolution_todaynow(self):
+        # specifically case where today/now is the *first* item
+        vals = np.array(["today", np.datetime64("2017-01-01", "us")], dtype=object)
+
+        now = Timestamp("now").asm8
+        res, _ = array_strptime(vals, fmt="%Y-%m-%d", utc=False, creso=creso_infer)
+        res2, _ = array_strptime(
+            vals[::-1], fmt="%Y-%m-%d", utc=False, creso=creso_infer
+        )
+
+        # 1s is an arbitrary cutoff for call overhead; in local testing the
+        #  actual difference is about 250us
+        tolerance = np.timedelta64(1, "s")
+
+        assert res.dtype == "M8[us]"
+        assert abs(res[0] - now) < tolerance
+        assert res[1] == vals[1]
+
+        assert res2.dtype == "M8[us]"
+        assert abs(res2[1] - now) < tolerance * 2
+        assert res2[0] == vals[1]
+
+    def test_array_strptime_str_outside_nano_range(self):
+        vals = np.array(["2401-09-15"], dtype=object)
+        expected = np.array(["2401-09-15"], dtype="M8[s]")
+        fmt = "ISO8601"
+        res, _ = array_strptime(vals, fmt=fmt, creso=creso_infer)
+        tm.assert_numpy_array_equal(res, expected)
+
+        # non-iso -> different path
+        vals2 = np.array(["Sep 15, 2401"], dtype=object)
+        expected2 = np.array(["2401-09-15"], dtype="M8[s]")
+        fmt2 = "%b %d, %Y"
+        res2, _ = array_strptime(vals2, fmt=fmt2, creso=creso_infer)
+        tm.assert_numpy_array_equal(res2, expected2)
diff --git a/pandas/pandas/tests/tslibs/test_timedeltas.py b/pandas/pandas/tests/tslibs/test_timedeltas.py
new file mode 100644
index 0000000000000..4784a6d0d600d
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_timedeltas.py
@@ -0,0 +1,149 @@
+import re
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.timedeltas import (
+    array_to_timedelta64,
+    delta_to_nanoseconds,
+    ints_to_pytimedelta,
+)
+
+from pandas import (
+    Timedelta,
+    offsets,
+)
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "obj,expected",
+    [
+        (np.timedelta64(14, "D"), 14 * 24 * 3600 * 1e9),
+        (Timedelta(minutes=-7), -7 * 60 * 1e9),
+        (Timedelta(minutes=-7).to_pytimedelta(), -7 * 60 * 1e9),
+        (Timedelta(seconds=1234e-9), 1234),  # GH43764, GH40946
+        (
+            Timedelta(seconds=1e-9, milliseconds=1e-5, microseconds=1e-1),
+            111,
+        ),  # GH43764
+        (
+            Timedelta(days=1, seconds=1e-9, milliseconds=1e-5, microseconds=1e-1),
+            24 * 3600e9 + 111,
+        ),  # GH43764
+        (offsets.Nano(125), 125),
+    ],
+)
+def test_delta_to_nanoseconds(obj, expected):
+    result = delta_to_nanoseconds(obj)
+    assert result == expected
+
+
+def test_delta_to_nanoseconds_error():
+    obj = np.array([123456789], dtype="m8[ns]")
+
+    with pytest.raises(TypeError, match="<class 'numpy.ndarray'>"):
+        delta_to_nanoseconds(obj)
+
+    with pytest.raises(TypeError, match="float"):
+        delta_to_nanoseconds(1.5)
+    with pytest.raises(TypeError, match="int"):
+        delta_to_nanoseconds(1)
+    with pytest.raises(TypeError, match="int"):
+        delta_to_nanoseconds(np.int64(2))
+    with pytest.raises(TypeError, match="int"):
+        delta_to_nanoseconds(np.int32(3))
+
+
+def test_delta_to_nanoseconds_td64_MY_raises():
+    msg = (
+        "delta_to_nanoseconds does not support Y or M units, "
+        "as their duration in nanoseconds is ambiguous"
+    )
+
+    td = np.timedelta64(1234, "Y")
+
+    with pytest.raises(ValueError, match=msg):
+        delta_to_nanoseconds(td)
+
+    td = np.timedelta64(1234, "M")
+
+    with pytest.raises(ValueError, match=msg):
+        delta_to_nanoseconds(td)
+
+
+@pytest.mark.parametrize("unit", ["Y", "M"])
+def test_unsupported_td64_unit_raises(unit):
+    # GH 52806
+    with pytest.raises(
+        ValueError,
+        match=f"Unit {unit} is not supported. "
+        "Only unambiguous timedelta values durations are supported. "
+        "Allowed units are 'W', 'D', 'h', 'm', 's', 'ms', 'us', 'ns'",
+    ):
+        Timedelta(np.timedelta64(1, unit))
+
+
+def test_huge_nanoseconds_overflow():
+    # GH 32402
+    assert delta_to_nanoseconds(Timedelta(1e10)) == 1e10
+    assert delta_to_nanoseconds(Timedelta(nanoseconds=1e10)) == 1e10
+
+
+@pytest.mark.parametrize(
+    "kwargs", [{"Seconds": 1}, {"seconds": 1, "Nanoseconds": 1}, {"Foo": 2}]
+)
+def test_kwarg_assertion(kwargs):
+    err_message = (
+        "cannot construct a Timedelta from the passed arguments, "
+        "allowed keywords are "
+        "[weeks, days, hours, minutes, seconds, "
+        "milliseconds, microseconds, nanoseconds]"
+    )
+
+    with pytest.raises(ValueError, match=re.escape(err_message)):
+        Timedelta(**kwargs)
+
+
+class TestArrayToTimedelta64:
+    def test_array_to_timedelta64_string_with_unit_2d_raises(self):
+        # check the 'unit is not None and errors != "coerce"' path
+        #  in array_to_timedelta64 raises correctly with 2D values
+        values = np.array([["1", 2], [3, "4"]], dtype=object)
+        with pytest.raises(ValueError, match="unit must not be specified"):
+            array_to_timedelta64(values, unit="s")
+
+    def test_array_to_timedelta64_non_object_raises(self):
+        # check we raise, not segfault
+        values = np.arange(5)
+
+        msg = "'values' must have object dtype"
+        with pytest.raises(TypeError, match=msg):
+            array_to_timedelta64(values)
+
+
+@pytest.mark.parametrize("unit", ["s", "ms", "us"])
+def test_ints_to_pytimedelta(unit):
+    # tests for non-nanosecond cases
+    arr = np.arange(6, dtype=np.int64).view(f"m8[{unit}]")
+
+    res = ints_to_pytimedelta(arr, box=False)
+    # For non-nanosecond, .astype(object) gives pytimedelta objects
+    #  instead of integers
+    expected = arr.astype(object)
+    tm.assert_numpy_array_equal(res, expected)
+
+    res = ints_to_pytimedelta(arr, box=True)
+    expected = np.array([Timedelta(x) for x in arr], dtype=object)
+    tm.assert_numpy_array_equal(res, expected)
+
+
+@pytest.mark.parametrize("unit", ["Y", "M", "ps", "fs", "as"])
+def test_ints_to_pytimedelta_unsupported(unit):
+    arr = np.arange(6, dtype=np.int64).view(f"m8[{unit}]")
+
+    with pytest.raises(NotImplementedError, match=r"\d{1,2}"):
+        ints_to_pytimedelta(arr, box=False)
+    msg = "Only resolutions 's', 'ms', 'us', 'ns' are supported"
+    with pytest.raises(NotImplementedError, match=msg):
+        ints_to_pytimedelta(arr, box=True)
diff --git a/pandas/pandas/tests/tslibs/test_timezones.py b/pandas/pandas/tests/tslibs/test_timezones.py
new file mode 100644
index 0000000000000..60bbcf08ce8e7
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_timezones.py
@@ -0,0 +1,171 @@
+from datetime import (
+    datetime,
+    timedelta,
+    timezone,
+)
+
+import dateutil.tz
+import pytest
+
+from pandas._libs.tslibs import (
+    conversion,
+    timezones,
+)
+from pandas.compat import is_platform_windows
+
+from pandas import Timestamp
+
+
+def test_is_utc(utc_fixture):
+    tz = timezones.maybe_get_tz(utc_fixture)
+    assert timezones.is_utc(tz)
+
+
+def test_cache_keys_are_distinct_for_pytz_vs_dateutil():
+    pytz = pytest.importorskip("pytz")
+    for tz_name in pytz.common_timezones:
+        tz_p = timezones.maybe_get_tz(tz_name)
+        tz_d = timezones.maybe_get_tz("dateutil/" + tz_name)
+
+    if tz_d is None:
+        pytest.skip(tz_name + ": dateutil does not know about this one")
+
+    if not (tz_name == "UTC" and is_platform_windows()):
+        # they both end up as tzwin("UTC") on windows
+        assert timezones._p_tz_cache_key(tz_p) != timezones._p_tz_cache_key(tz_d)
+
+
+def test_tzlocal_repr():
+    # see gh-13583
+    ts = Timestamp("2011-01-01", tz=dateutil.tz.tzlocal())
+    assert ts.tz == dateutil.tz.tzlocal()
+    assert "tz='tzlocal()')" in repr(ts)
+
+
+def test_tzlocal_maybe_get_tz():
+    # see gh-13583
+    tz = timezones.maybe_get_tz("tzlocal()")
+    assert tz == dateutil.tz.tzlocal()
+
+
+def test_tzlocal_offset():
+    # see gh-13583
+    #
+    # Get offset using normal datetime for test.
+    ts = Timestamp("2011-01-01", tz=dateutil.tz.tzlocal())
+
+    offset = dateutil.tz.tzlocal().utcoffset(datetime(2011, 1, 1))
+    offset = offset.total_seconds()
+
+    assert ts._value + offset == Timestamp("2011-01-01")._value
+
+
+def test_tzlocal_is_not_utc():
+    # even if the machine running the test is localized to UTC
+    tz = dateutil.tz.tzlocal()
+    assert not timezones.is_utc(tz)
+
+    assert not timezones.tz_compare(tz, dateutil.tz.tzutc())
+
+
+def test_tz_compare_utc(utc_fixture, utc_fixture2):
+    tz = timezones.maybe_get_tz(utc_fixture)
+    tz2 = timezones.maybe_get_tz(utc_fixture2)
+    assert timezones.tz_compare(tz, tz2)
+
+
+@pytest.fixture(
+    params=[
+        ("pytz/US/Eastern", lambda tz, x: tz.localize(x)),
+        (dateutil.tz.gettz("US/Eastern"), lambda tz, x: x.replace(tzinfo=tz)),
+    ]
+)
+def infer_setup(request):
+    eastern, localize = request.param
+    if isinstance(eastern, str) and eastern.startswith("pytz/"):
+        pytz = pytest.importorskip("pytz")
+        eastern = pytz.timezone(eastern.removeprefix("pytz/"))
+
+    start_naive = datetime(2001, 1, 1)
+    end_naive = datetime(2009, 1, 1)
+
+    start = localize(eastern, start_naive)
+    end = localize(eastern, end_naive)
+
+    return eastern, localize, start, end, start_naive, end_naive
+
+
+def test_infer_tz_compat(infer_setup):
+    eastern, _, start, end, start_naive, end_naive = infer_setup
+
+    assert (
+        timezones.infer_tzinfo(start, end)
+        is conversion.localize_pydatetime(start_naive, eastern).tzinfo
+    )
+    assert (
+        timezones.infer_tzinfo(start, None)
+        is conversion.localize_pydatetime(start_naive, eastern).tzinfo
+    )
+    assert (
+        timezones.infer_tzinfo(None, end)
+        is conversion.localize_pydatetime(end_naive, eastern).tzinfo
+    )
+
+
+def test_infer_tz_utc_localize(infer_setup):
+    _, _, start, end, start_naive, end_naive = infer_setup
+    utc = timezone.utc
+
+    start = start_naive.astimezone(utc)
+    end = end_naive.astimezone(utc)
+
+    assert timezones.infer_tzinfo(start, end) is utc
+
+
+@pytest.mark.parametrize("ordered", [True, False])
+def test_infer_tz_mismatch(infer_setup, ordered):
+    eastern, _, _, _, start_naive, end_naive = infer_setup
+    msg = "Inputs must both have the same timezone"
+
+    utc = timezone.utc
+    start = start_naive.astimezone(utc)
+    end = conversion.localize_pydatetime(end_naive, eastern)
+
+    args = (start, end) if ordered else (end, start)
+
+    with pytest.raises(AssertionError, match=msg):
+        timezones.infer_tzinfo(*args)
+
+
+def test_maybe_get_tz_invalid_types():
+    with pytest.raises(TypeError, match="<class 'float'>"):
+        timezones.maybe_get_tz(44.0)
+
+    with pytest.raises(TypeError, match="<class 'module'>"):
+        timezones.maybe_get_tz(pytest)
+
+    msg = "<class 'pandas.Timestamp'>"
+    with pytest.raises(TypeError, match=msg):
+        timezones.maybe_get_tz(Timestamp("2021-01-01", tz="UTC"))
+
+
+def test_maybe_get_tz_offset_only():
+    # see gh-36004
+
+    # timezone.utc
+    tz = timezones.maybe_get_tz(timezone.utc)
+    assert tz == timezone(timedelta(hours=0, minutes=0))
+
+    # without UTC+- prefix
+    tz = timezones.maybe_get_tz("+01:15")
+    assert tz == timezone(timedelta(hours=1, minutes=15))
+
+    tz = timezones.maybe_get_tz("-01:15")
+    assert tz == timezone(-timedelta(hours=1, minutes=15))
+
+    # with UTC+- prefix
+    tz = timezones.maybe_get_tz("UTC+02:45")
+    assert tz == timezone(timedelta(hours=2, minutes=45))
+
+    tz = timezones.maybe_get_tz("UTC-02:45")
+    assert tz == timezone(-timedelta(hours=2, minutes=45))
diff --git a/pandas/pandas/tests/tslibs/test_to_offset.py b/pandas/pandas/tests/tslibs/test_to_offset.py
new file mode 100644
index 0000000000000..67521c7e2a3ac
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_to_offset.py
@@ -0,0 +1,239 @@
+import re
+
+import pytest
+
+from pandas._libs.tslibs import (
+    Timedelta,
+    offsets,
+    to_offset,
+)
+
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "freq_input,expected",
+    [
+        (to_offset("10us"), offsets.Micro(10)),
+        (offsets.Hour(), offsets.Hour()),
+        ("2h30min", offsets.Minute(150)),
+        ("2h 30min", offsets.Minute(150)),
+        ("2h30min15s", offsets.Second(150 * 60 + 15)),
+        ("2h 60min", offsets.Hour(3)),
+        ("2h 20.5min", offsets.Second(8430)),
+        ("1.5min", offsets.Second(90)),
+        ("0.5s", offsets.Milli(500)),
+        ("15ms500us", offsets.Micro(15500)),
+        ("10s75ms", offsets.Milli(10075)),
+        ("1s0.25ms", offsets.Micro(1000250)),
+        ("2800ns", offsets.Nano(2800)),
+        ("2SME", offsets.SemiMonthEnd(2)),
+        ("2SME-16", offsets.SemiMonthEnd(2, day_of_month=16)),
+        ("2SMS-14", offsets.SemiMonthBegin(2, day_of_month=14)),
+        ("2SMS-15", offsets.SemiMonthBegin(2)),
+        ("LWOM-MON", offsets.LastWeekOfMonth()),
+    ],
+)
+def test_to_offset(freq_input, expected):
+    result = to_offset(freq_input)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected", [("-1s", -1), ("-2SME", -2), ("-1SMS", -1), ("-5min10s", -310)]
+)
+def test_to_offset_negative(freqstr, expected):
+    result = to_offset(freqstr)
+    assert result.n == expected
+
+
+@pytest.mark.filterwarnings("ignore:.*'m' is deprecated.*:FutureWarning")
+@pytest.mark.parametrize(
+    "freqstr",
+    [
+        "2h20m",
+        "us1",
+        "-us",
+        "3us1",
+        "-2-3us",
+        "-2D:3h",
+        "1.5.0s",
+        "2SMS-15-15",
+        "2SMS-15D",
+        "100foo",
+        # Invalid leading +/- signs.
+        "+-1D",
+        "-+1h",
+        "+1",
+        "-7",
+        "+D",
+        "-m",
+        # Invalid shortcut anchors.
+        "SME-0",
+        "SME-28",
+        "SME-29",
+        "SME-FOO",
+        "BSM",
+        "SME--1",
+        "SMS-1",
+        "SMS-28",
+        "SMS-30",
+        "SMS-BAR",
+        "SMS-BYR",
+        "BSMS",
+        "SMS--2",
+    ],
+)
+def test_to_offset_invalid(freqstr):
+    # see gh-13930
+
+    # We escape string because some of our
+    # inputs contain regex special characters.
+    msg = re.escape(f"Invalid frequency: {freqstr}")
+    with pytest.raises(ValueError, match=msg):
+        to_offset(freqstr)
+
+
+def test_to_offset_no_evaluate():
+    msg = str(("", ""))
+    with pytest.raises(TypeError, match=msg):
+        to_offset(("", ""))
+
+
+def test_to_offset_tuple_unsupported():
+    with pytest.raises(TypeError, match="pass as a string instead"):
+        to_offset((5, "T"))
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected",
+    [
+        ("2D 3h", offsets.Hour(51)),
+        ("2 D3 h", offsets.Hour(51)),
+        ("2 D 3 h", offsets.Hour(51)),
+        ("  2 D 3 h  ", offsets.Hour(51)),
+        ("   h    ", offsets.Hour()),
+        (" 3  h    ", offsets.Hour(3)),
+    ],
+)
+def test_to_offset_whitespace(freqstr, expected):
+    result = to_offset(freqstr)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected", [("00h 00min 01s", 1), ("-00h 03min 14s", -194)]
+)
+def test_to_offset_leading_zero(freqstr, expected):
+    result = to_offset(freqstr)
+    assert result.n == expected
+
+
+@pytest.mark.parametrize(
+    "freqstr,expected,wrn", [("+1d", 1, FutureWarning), ("+2h30min", 150, None)]
+)
+def test_to_offset_leading_plus(freqstr, expected, wrn):
+    msg = "'d' is deprecated and will be removed in a future version."
+
+    with tm.assert_produces_warning(wrn, match=msg):
+        result = to_offset(freqstr)
+    assert result.n == expected
+
+
+@pytest.mark.parametrize(
+    "kwargs,expected",
+    [
+        ({"days": 1, "seconds": 1}, offsets.Second(86401)),
+        ({"days": -1, "seconds": 1}, offsets.Second(-86399)),
+        ({"hours": 1, "minutes": 10}, offsets.Minute(70)),
+        ({"hours": 1, "minutes": -10}, offsets.Minute(50)),
+        ({"weeks": 1}, offsets.Day(7)),
+        ({"hours": 1}, offsets.Hour(1)),
+        ({"hours": 1}, to_offset("60min")),
+        ({"microseconds": 1}, offsets.Micro(1)),
+        ({"microseconds": 0}, offsets.Nano(0)),
+    ],
+)
+def test_to_offset_pd_timedelta(kwargs, expected):
+    # see gh-9064
+    td = Timedelta(**kwargs)
+    result = to_offset(td)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "shortcut,expected",
+    [
+        ("W", offsets.Week(weekday=6)),
+        ("W-SUN", offsets.Week(weekday=6)),
+        ("QE", offsets.QuarterEnd(startingMonth=12)),
+        ("QE-DEC", offsets.QuarterEnd(startingMonth=12)),
+        ("QE-MAY", offsets.QuarterEnd(startingMonth=5)),
+        ("SME", offsets.SemiMonthEnd(day_of_month=15)),
+        ("SME-15", offsets.SemiMonthEnd(day_of_month=15)),
+        ("SME-1", offsets.SemiMonthEnd(day_of_month=1)),
+        ("SME-27", offsets.SemiMonthEnd(day_of_month=27)),
+        ("SMS-2", offsets.SemiMonthBegin(day_of_month=2)),
+        ("SMS-27", offsets.SemiMonthBegin(day_of_month=27)),
+    ],
+)
+def test_anchored_shortcuts(shortcut, expected):
+    result = to_offset(shortcut)
+    assert result == expected
+
+
+@pytest.mark.parametrize(
+    "freq_depr",
+    [
+        "2ye-mar",
+        "2ys",
+        "2qe",
+        "2qs-feb",
+        "2bqs",
+        "2sms",
+        "1sme",
+        "2bms",
+        "2cbme",
+        "2me",
+    ],
+)
+def test_to_offset_lowercase_frequency_raises(freq_depr):
+    msg = f"Invalid frequency: {freq_depr}"
+
+    with pytest.raises(ValueError, match=msg):
+        to_offset(freq_depr)
+
+
+@pytest.mark.parametrize("freq_depr", ["2MIN", "2Us", "2NS"])
+def test_to_offset_uppercase_frequency_deprecated(freq_depr):
+    # GH#54939
+    depr_msg = f"'{freq_depr[1:]}' is deprecated and will be removed in a "
+    f"future version, please use '{freq_depr.lower()[1:]}' instead."
+
+    with tm.assert_produces_warning(FutureWarning, match=depr_msg):
+        to_offset(freq_depr)
+
+
+@pytest.mark.parametrize(
+    "freq_depr,expected",
+    [
+        ("2w", offsets.Week(2, weekday=6)),
+        ("2b", offsets.BusinessDay(2)),
+        ("2d", offsets.Day(2)),
+    ],
+)
+def test_to_offset_lowercase_frequency_deprecated(freq_depr, expected):
+    # GH#54939, GH#58998
+    msg = f"'{freq_depr[1:]}' is deprecated and will be removed in a future version."
+
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        result = to_offset(freq_depr)
+    assert result == expected
+
+
+@pytest.mark.parametrize("freq", ["2H", "2BH", "2S"])
+def test_to_offset_uppercase_frequency_raises(freq):
+    msg = f"Invalid frequency: {freq}"
+
+    with pytest.raises(ValueError, match=msg):
+        to_offset(freq)
diff --git a/pandas/pandas/tests/tslibs/test_tzconversion.py b/pandas/pandas/tests/tslibs/test_tzconversion.py
new file mode 100644
index 0000000000000..f32829b4e0b21
--- /dev/null
+++ b/pandas/pandas/tests/tslibs/test_tzconversion.py
@@ -0,0 +1,26 @@
+import zoneinfo
+
+import numpy as np
+import pytest
+
+from pandas._libs.tslibs.tzconversion import tz_localize_to_utc
+
+
+class TestTZLocalizeToUTC:
+    def test_tz_localize_to_utc_ambiguous_infer(self):
+        # val is a timestamp that is ambiguous when localized to US/Eastern
+        val = 1_320_541_200_000_000_000
+        vals = np.array([val, val - 1, val], dtype=np.int64)
+
+        with pytest.raises(ValueError, match="2011-11-06 01:00:00"):
+            tz_localize_to_utc(vals, zoneinfo.ZoneInfo("US/Eastern"), ambiguous="infer")
+
+        with pytest.raises(ValueError, match="are no repeated times"):
+            tz_localize_to_utc(
+                vals[:1], zoneinfo.ZoneInfo("US/Eastern"), ambiguous="infer"
+            )
+
+        vals[1] += 1
+        msg = "There are 2 dst switches when there should only be 1"
+        with pytest.raises(ValueError, match=msg):
+            tz_localize_to_utc(vals, zoneinfo.ZoneInfo("US/Eastern"), ambiguous="infer")
diff --git a/pandas/pandas/tests/util/__init__.py b/pandas/pandas/tests/util/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/util/conftest.py b/pandas/pandas/tests/util/conftest.py
new file mode 100644
index 0000000000000..2e931ff42fe15
--- /dev/null
+++ b/pandas/pandas/tests/util/conftest.py
@@ -0,0 +1,46 @@
+import pytest
+
+
+@pytest.fixture(params=[True, False])
+def check_dtype(request):
+    """
+    Fixture returning `True` or `False`, determining whether to check
+    if the `dtype` is identical or not, when comparing two data structures,
+    e.g. `Series`, `SparseArray` or `DataFrame`.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def check_exact(request):
+    """
+    Fixture returning `True` or `False`, determining whether to
+    compare floating point numbers exactly or not.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def check_index_type(request):
+    """
+    Fixture returning `True` or `False`, determining whether to check
+    if the `Index` types are identical or not.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[0.5e-3, 0.5e-5])
+def rtol(request):
+    """
+    Fixture returning 0.5e-3 or 0.5e-5. Those values are used as relative tolerance.
+    """
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def check_categorical(request):
+    """
+    Fixture returning `True` or `False`, determining whether to
+    compare internal `Categorical` exactly or not.
+    """
+    return request.param
diff --git a/pandas/pandas/tests/util/test_assert_almost_equal.py b/pandas/pandas/tests/util/test_assert_almost_equal.py
new file mode 100644
index 0000000000000..091670ed69f11
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_almost_equal.py
@@ -0,0 +1,586 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    DataFrame,
+    Index,
+    NaT,
+    Series,
+    Timestamp,
+)
+import pandas._testing as tm
+
+
+def _assert_almost_equal_both(a, b, **kwargs):
+    """
+    Check that two objects are approximately equal.
+
+    This check is performed commutatively.
+
+    Parameters
+    ----------
+    a : object
+        The first object to compare.
+    b : object
+        The second object to compare.
+    **kwargs
+        The arguments passed to `tm.assert_almost_equal`.
+    """
+    tm.assert_almost_equal(a, b, **kwargs)
+    tm.assert_almost_equal(b, a, **kwargs)
+
+
+def _assert_not_almost_equal(a, b, **kwargs):
+    """
+    Check that two objects are not approximately equal.
+
+    Parameters
+    ----------
+    a : object
+        The first object to compare.
+    b : object
+        The second object to compare.
+    **kwargs
+        The arguments passed to `tm.assert_almost_equal`.
+    """
+    try:
+        tm.assert_almost_equal(a, b, **kwargs)
+        msg = f"{a} and {b} were approximately equal when they shouldn't have been"
+        pytest.fail(reason=msg)
+    except AssertionError:
+        pass
+
+
+def _assert_not_almost_equal_both(a, b, **kwargs):
+    """
+    Check that two objects are not approximately equal.
+
+    This check is performed commutatively.
+
+    Parameters
+    ----------
+    a : object
+        The first object to compare.
+    b : object
+        The second object to compare.
+    **kwargs
+        The arguments passed to `tm.assert_almost_equal`.
+    """
+    _assert_not_almost_equal(a, b, **kwargs)
+    _assert_not_almost_equal(b, a, **kwargs)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (1.1, 1.1),
+        (1.1, 1.100001),
+        (np.int16(1), 1.000001),
+        (np.float64(1.1), 1.1),
+        (np.uint32(5), 5),
+    ],
+)
+def test_assert_almost_equal_numbers(a, b):
+    _assert_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (1.1, 1),
+        (1.1, True),
+        (1, 2),
+        (1.0001, np.int16(1)),
+        # The following two examples are not "almost equal" due to tol.
+        (0.1, 0.1001),
+        (0.0011, 0.0012),
+    ],
+)
+def test_assert_not_almost_equal_numbers(a, b):
+    _assert_not_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (1.1, 1.1),
+        (1.1, 1.100001),
+        (1.1, 1.1001),
+        (0.000001, 0.000005),
+        (1000.0, 1000.0005),
+        # Testing this example, as per #13357
+        (0.000011, 0.000012),
+    ],
+)
+def test_assert_almost_equal_numbers_atol(a, b):
+    # Equivalent to the deprecated check_less_precise=True, enforced in 2.0
+    _assert_almost_equal_both(a, b, rtol=0.5e-3, atol=0.5e-3)
+
+
+@pytest.mark.parametrize("a,b", [(1.1, 1.11), (0.1, 0.101), (0.000011, 0.001012)])
+def test_assert_not_almost_equal_numbers_atol(a, b):
+    _assert_not_almost_equal_both(a, b, atol=1e-3)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (1.1, 1.1),
+        (1.1, 1.100001),
+        (1.1, 1.1001),
+        (1000.0, 1000.0005),
+        (1.1, 1.11),
+        (0.1, 0.101),
+    ],
+)
+def test_assert_almost_equal_numbers_rtol(a, b):
+    _assert_almost_equal_both(a, b, rtol=0.05)
+
+
+@pytest.mark.parametrize("a,b", [(0.000011, 0.000012), (0.000001, 0.000005)])
+def test_assert_not_almost_equal_numbers_rtol(a, b):
+    _assert_not_almost_equal_both(a, b, rtol=0.05)
+
+
+@pytest.mark.parametrize(
+    "a,b,rtol",
+    [
+        (1.00001, 1.00005, 0.001),
+        (-0.908356 + 0.2j, -0.908358 + 0.2j, 1e-3),
+        (0.1 + 1.009j, 0.1 + 1.006j, 0.1),
+        (0.1001 + 2.0j, 0.1 + 2.001j, 0.01),
+    ],
+)
+def test_assert_almost_equal_complex_numbers(a, b, rtol):
+    _assert_almost_equal_both(a, b, rtol=rtol)
+    _assert_almost_equal_both(np.complex64(a), np.complex64(b), rtol=rtol)
+    _assert_almost_equal_both(np.complex128(a), np.complex128(b), rtol=rtol)
+
+
+@pytest.mark.parametrize(
+    "a,b,rtol",
+    [
+        (0.58310768, 0.58330768, 1e-7),
+        (-0.908 + 0.2j, -0.978 + 0.2j, 0.001),
+        (0.1 + 1j, 0.1 + 2j, 0.01),
+        (-0.132 + 1.001j, -0.132 + 1.005j, 1e-5),
+        (0.58310768j, 0.58330768j, 1e-9),
+    ],
+)
+def test_assert_not_almost_equal_complex_numbers(a, b, rtol):
+    _assert_not_almost_equal_both(a, b, rtol=rtol)
+    _assert_not_almost_equal_both(np.complex64(a), np.complex64(b), rtol=rtol)
+    _assert_not_almost_equal_both(np.complex128(a), np.complex128(b), rtol=rtol)
+
+
+@pytest.mark.parametrize("a,b", [(0, 0), (0, 0.0), (0, np.float64(0)), (0.00000001, 0)])
+def test_assert_almost_equal_numbers_with_zeros(a, b):
+    _assert_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize("a,b", [(0.001, 0), (1, 0)])
+def test_assert_not_almost_equal_numbers_with_zeros(a, b):
+    _assert_not_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize("a,b", [(1, "abc"), (1, [1]), (1, object())])
+def test_assert_not_almost_equal_numbers_with_mixed(a, b):
+    _assert_not_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize(
+    "left_dtype", ["M8[ns]", "m8[ns]", "float64", "int64", "object"]
+)
+@pytest.mark.parametrize(
+    "right_dtype", ["M8[ns]", "m8[ns]", "float64", "int64", "object"]
+)
+def test_assert_almost_equal_edge_case_ndarrays(left_dtype, right_dtype):
+    # Empty compare.
+    _assert_almost_equal_both(
+        np.array([], dtype=left_dtype),
+        np.array([], dtype=right_dtype),
+        check_dtype=False,
+    )
+
+
+def test_assert_almost_equal_sets():
+    # GH#51727
+    _assert_almost_equal_both({1, 2, 3}, {1, 2, 3})
+
+
+def test_assert_almost_not_equal_sets():
+    # GH#51727
+    msg = r"{1, 2, 3} != {1, 2, 4}"
+    with pytest.raises(AssertionError, match=msg):
+        _assert_almost_equal_both({1, 2, 3}, {1, 2, 4})
+
+
+def test_assert_almost_equal_dicts():
+    _assert_almost_equal_both({"a": 1, "b": 2}, {"a": 1, "b": 2})
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        ({"a": 1, "b": 2}, {"a": 1, "b": 3}),
+        ({"a": 1, "b": 2}, {"a": 1, "b": 2, "c": 3}),
+        ({"a": 1}, 1),
+        ({"a": 1}, "abc"),
+        ({"a": 1}, [1]),
+    ],
+)
+def test_assert_not_almost_equal_dicts(a, b):
+    _assert_not_almost_equal_both(a, b)
+
+
+@pytest.mark.parametrize("val", [1, 2])
+def test_assert_almost_equal_dict_like_object(val):
+    dict_val = 1
+    real_dict = {"a": val}
+
+    class DictLikeObj:
+        def keys(self):
+            return ("a",)
+
+        def __getitem__(self, item):
+            if item == "a":
+                return dict_val
+
+    func = (
+        _assert_almost_equal_both if val == dict_val else _assert_not_almost_equal_both
+    )
+    func(real_dict, DictLikeObj(), check_dtype=False)
+
+
+def test_assert_almost_equal_strings():
+    _assert_almost_equal_both("abc", "abc")
+
+
+@pytest.mark.parametrize("b", ["abcd", "abd", 1, [1]])
+def test_assert_not_almost_equal_strings(b):
+    _assert_not_almost_equal_both("abc", b)
+
+
+@pytest.mark.parametrize("box", [list, np.array])
+def test_assert_almost_equal_iterables(box):
+    _assert_almost_equal_both(box([1, 2, 3]), box([1, 2, 3]))
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        # Class is different.
+        (np.array([1, 2, 3]), [1, 2, 3]),
+        # Dtype is different.
+        (np.array([1, 2, 3]), np.array([1.0, 2.0, 3.0])),
+        # Can't compare generators.
+        (iter([1, 2, 3]), [1, 2, 3]),
+        ([1, 2, 3], [1, 2, 4]),
+        ([1, 2, 3], [1, 2, 3, 4]),
+        ([1, 2, 3], 1),
+    ],
+)
+def test_assert_not_almost_equal_iterables(a, b):
+    _assert_not_almost_equal(a, b)
+
+
+def test_assert_almost_equal_null():
+    _assert_almost_equal_both(None, None)
+
+
+@pytest.mark.parametrize("a,b", [(None, np.nan), (None, 0), (np.nan, 0)])
+def test_assert_not_almost_equal_null(a, b):
+    _assert_not_almost_equal(a, b)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (np.inf, np.inf),
+        (np.inf, float("inf")),
+        (np.array([np.inf, np.nan, -np.inf]), np.array([np.inf, np.nan, -np.inf])),
+    ],
+)
+def test_assert_almost_equal_inf(a, b):
+    _assert_almost_equal_both(a, b)
+
+
+objs = [NA, np.nan, NaT, None, np.datetime64("NaT"), np.timedelta64("NaT")]
+
+
+@pytest.mark.parametrize("left", objs)
+@pytest.mark.parametrize("right", objs)
+def test_mismatched_na_assert_almost_equal(left, right):
+    left_arr = np.array([left], dtype=object)
+    right_arr = np.array([right], dtype=object)
+
+    msg = "Mismatched null-like values"
+
+    if left is right:
+        _assert_almost_equal_both(left, right, check_dtype=False)
+        tm.assert_numpy_array_equal(left_arr, right_arr)
+        tm.assert_index_equal(
+            Index(left_arr, dtype=object), Index(right_arr, dtype=object)
+        )
+        tm.assert_series_equal(
+            Series(left_arr, dtype=object), Series(right_arr, dtype=object)
+        )
+        tm.assert_frame_equal(
+            DataFrame(left_arr, dtype=object), DataFrame(right_arr, dtype=object)
+        )
+
+    else:
+        with pytest.raises(AssertionError, match=msg):
+            _assert_almost_equal_both(left, right, check_dtype=False)
+
+        # TODO: to get the same deprecation in assert_numpy_array_equal we need
+        #  to change/deprecate the default for strict_nan to become True
+        # TODO: to get the same deprecation in assert_index_equal we need to
+        #  change/deprecate array_equivalent_object to be stricter, as
+        #  assert_index_equal uses Index.equal which uses array_equivalent.
+        with pytest.raises(AssertionError, match="Series are different"):
+            tm.assert_series_equal(
+                Series(left_arr, dtype=object), Series(right_arr, dtype=object)
+            )
+        with pytest.raises(AssertionError, match="DataFrame.iloc.* are different"):
+            tm.assert_frame_equal(
+                DataFrame(left_arr, dtype=object), DataFrame(right_arr, dtype=object)
+            )
+
+
+def test_assert_not_almost_equal_inf():
+    _assert_not_almost_equal_both(np.inf, 0)
+
+
+@pytest.mark.parametrize(
+    "a,b",
+    [
+        (Index([1.0, 1.1]), Index([1.0, 1.100001])),
+        (Series([1.0, 1.1]), Series([1.0, 1.100001])),
+        (np.array([1.1, 2.000001]), np.array([1.1, 2.0])),
+        (DataFrame({"a": [1.0, 1.1]}), DataFrame({"a": [1.0, 1.100001]})),
+    ],
+)
+def test_assert_almost_equal_pandas(a, b):
+    _assert_almost_equal_both(a, b)
+
+
+def test_assert_almost_equal_object():
+    a = [Timestamp("2011-01-01"), Timestamp("2011-01-01")]
+    b = [Timestamp("2011-01-01"), Timestamp("2011-01-01")]
+    _assert_almost_equal_both(a, b)
+
+
+def test_assert_almost_equal_value_mismatch():
+    msg = "expected 2\\.00000 but got 1\\.00000, with rtol=1e-05, atol=1e-08"
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(1, 2)
+
+
+@pytest.mark.parametrize(
+    "a,b,klass1,klass2",
+    [(np.array([1]), 1, "ndarray", "int"), (1, np.array([1]), "int", "ndarray")],
+)
+def test_assert_almost_equal_class_mismatch(a, b, klass1, klass2):
+    msg = f"""numpy array are different
+
+numpy array classes are different
+\\[left\\]:  {klass1}
+\\[right\\]: {klass2}"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(a, b)
+
+
+def test_assert_almost_equal_value_mismatch1():
+    msg = """numpy array are different
+
+numpy array values are different \\(66\\.66667 %\\)
+\\[left\\]:  \\[nan, 2\\.0, 3\\.0\\]
+\\[right\\]: \\[1\\.0, nan, 3\\.0\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(np.array([np.nan, 2, 3]), np.array([1, np.nan, 3]))
+
+
+def test_assert_almost_equal_value_mismatch2():
+    msg = """numpy array are different
+
+numpy array values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[1, 2\\]
+\\[right\\]: \\[1, 3\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(np.array([1, 2]), np.array([1, 3]))
+
+
+def test_assert_almost_equal_value_mismatch3():
+    msg = """numpy array are different
+
+numpy array values are different \\(16\\.66667 %\\)
+\\[left\\]:  \\[\\[1, 2\\], \\[3, 4\\], \\[5, 6\\]\\]
+\\[right\\]: \\[\\[1, 3\\], \\[3, 4\\], \\[5, 6\\]\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(
+            np.array([[1, 2], [3, 4], [5, 6]]), np.array([[1, 3], [3, 4], [5, 6]])
+        )
+
+
+def test_assert_almost_equal_value_mismatch4():
+    msg = """numpy array are different
+
+numpy array values are different \\(25\\.0 %\\)
+\\[left\\]:  \\[\\[1, 2\\], \\[3, 4\\]\\]
+\\[right\\]: \\[\\[1, 3\\], \\[3, 4\\]\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(np.array([[1, 2], [3, 4]]), np.array([[1, 3], [3, 4]]))
+
+
+def test_assert_almost_equal_shape_mismatch_override():
+    msg = """Index are different
+
+Index shapes are different
+\\[left\\]:  \\(2L*,\\)
+\\[right\\]: \\(3L*,\\)"""
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(np.array([1, 2]), np.array([3, 4, 5]), obj="Index")
+
+
+def test_assert_almost_equal_unicode():
+    # see gh-20503
+    msg = """numpy array are different
+
+numpy array values are different \\(33\\.33333 %\\)
+\\[left\\]:  \\[á, à, ä\\]
+\\[right\\]: \\[á, à, å\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(np.array(["á", "à", "ä"]), np.array(["á", "à", "å"]))
+
+
+def test_assert_almost_equal_timestamp():
+    a = np.array([Timestamp("2011-01-01"), Timestamp("2011-01-01")])
+    b = np.array([Timestamp("2011-01-01"), Timestamp("2011-01-02")])
+
+    msg = """numpy array are different
+
+numpy array values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[2011-01-01 00:00:00, 2011-01-01 00:00:00\\]
+\\[right\\]: \\[2011-01-01 00:00:00, 2011-01-02 00:00:00\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal(a, b)
+
+
+def test_assert_almost_equal_iterable_length_mismatch():
+    msg = """Iterable are different
+
+Iterable length are different
+\\[left\\]:  2
+\\[right\\]: 3"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal([1, 2], [3, 4, 5])
+
+
+def test_assert_almost_equal_iterable_values_mismatch():
+    msg = """Iterable are different
+
+Iterable values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[1, 2\\]
+\\[right\\]: \\[1, 3\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_almost_equal([1, 2], [1, 3])
+
+
+subarr = np.empty(2, dtype=object)
+subarr[:] = [np.array([None, "b"], dtype=object), np.array(["c", "d"], dtype=object)]
+
+NESTED_CASES = [
+    # nested array
+    (
+        np.array([np.array([50, 70, 90]), np.array([20, 30])], dtype=object),
+        np.array([np.array([50, 70, 90]), np.array([20, 30])], dtype=object),
+    ),
+    # >1 level of nesting
+    (
+        np.array(
+            [
+                np.array([np.array([50, 70]), np.array([90])], dtype=object),
+                np.array([np.array([20, 30])], dtype=object),
+            ],
+            dtype=object,
+        ),
+        np.array(
+            [
+                np.array([np.array([50, 70]), np.array([90])], dtype=object),
+                np.array([np.array([20, 30])], dtype=object),
+            ],
+            dtype=object,
+        ),
+    ),
+    # lists
+    (
+        np.array([[50, 70, 90], [20, 30]], dtype=object),
+        np.array([[50, 70, 90], [20, 30]], dtype=object),
+    ),
+    # mixed array/list
+    (
+        np.array([np.array([1, 2, 3]), np.array([4, 5])], dtype=object),
+        np.array([[1, 2, 3], [4, 5]], dtype=object),
+    ),
+    (
+        np.array([np.array([], dtype=object), None], dtype=object),
+        np.array([[], None], dtype=object),
+    ),
+    (
+        np.array(
+            [
+                np.array([np.array([1, 2, 3]), np.array([4, 5])], dtype=object),
+                np.array(
+                    [np.array([6]), np.array([7, 8]), np.array([9])], dtype=object
+                ),
+            ],
+            dtype=object,
+        ),
+        np.array([[[1, 2, 3], [4, 5]], [[6], [7, 8], [9]]], dtype=object),
+    ),
+    # same-length lists
+    (
+        np.array([subarr, None], dtype=object),
+        np.array([[[None, "b"], ["c", "d"]], None], dtype=object),
+    ),
+    # dicts
+    (
+        np.array([{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object),
+        np.array([{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object),
+    ),
+    (
+        np.array([{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object),
+        np.array([{"f1": 1, "f2": ["a", "b"]}], dtype=object),
+    ),
+    # array/list of dicts
+    (
+        np.array(
+            [
+                np.array(
+                    [{"f1": 1, "f2": np.array(["a", "b"], dtype=object)}], dtype=object
+                ),
+                np.array([], dtype=object),
+            ],
+            dtype=object,
+        ),
+        np.array([[{"f1": 1, "f2": ["a", "b"]}], []], dtype=object),
+    ),
+]
+
+
+@pytest.mark.filterwarnings("ignore:elementwise comparison failed:DeprecationWarning")
+@pytest.mark.parametrize("a,b", NESTED_CASES)
+def test_assert_almost_equal_array_nested(a, b):
+    _assert_almost_equal_both(a, b)
diff --git a/pandas/pandas/tests/util/test_assert_attr_equal.py b/pandas/pandas/tests/util/test_assert_attr_equal.py
new file mode 100644
index 0000000000000..bbbb0bf2172b1
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_attr_equal.py
@@ -0,0 +1,33 @@
+from types import SimpleNamespace
+
+import pytest
+
+from pandas.core.dtypes.common import is_float
+
+import pandas._testing as tm
+
+
+def test_assert_attr_equal(nulls_fixture):
+    obj = SimpleNamespace()
+    obj.na_value = nulls_fixture
+    tm.assert_attr_equal("na_value", obj, obj)
+
+
+def test_assert_attr_equal_different_nulls(nulls_fixture, nulls_fixture2):
+    obj = SimpleNamespace()
+    obj.na_value = nulls_fixture
+
+    obj2 = SimpleNamespace()
+    obj2.na_value = nulls_fixture2
+
+    if nulls_fixture is nulls_fixture2:
+        tm.assert_attr_equal("na_value", obj, obj2)
+    elif is_float(nulls_fixture) and is_float(nulls_fixture2):
+        # we consider float("nan") and np.float64("nan") to be equivalent
+        tm.assert_attr_equal("na_value", obj, obj2)
+    elif type(nulls_fixture) is type(nulls_fixture2):
+        # e.g. Decimal("NaN")
+        tm.assert_attr_equal("na_value", obj, obj2)
+    else:
+        with pytest.raises(AssertionError, match='"na_value" are different'):
+            tm.assert_attr_equal("na_value", obj, obj2)
diff --git a/pandas/pandas/tests/util/test_assert_categorical_equal.py b/pandas/pandas/tests/util/test_assert_categorical_equal.py
new file mode 100644
index 0000000000000..c171564574708
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_categorical_equal.py
@@ -0,0 +1,88 @@
+import pytest
+
+from pandas import Categorical
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize("c", [None, [1, 2, 3, 4, 5]])
+def test_categorical_equal(c):
+    c = Categorical([1, 2, 3, 4], categories=c)
+    tm.assert_categorical_equal(c, c)
+
+
+@pytest.mark.parametrize("check_category_order", [True, False])
+def test_categorical_equal_order_mismatch(check_category_order):
+    c1 = Categorical([1, 2, 3, 4], categories=[1, 2, 3, 4])
+    c2 = Categorical([1, 2, 3, 4], categories=[4, 3, 2, 1])
+    kwargs = {"check_category_order": check_category_order}
+
+    if check_category_order:
+        msg = """Categorical\\.categories are different
+
+Categorical\\.categories values are different \\(100\\.0 %\\)
+\\[left\\]:  Index\\(\\[1, 2, 3, 4\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[4, 3, 2, 1\\], dtype='int64'\\)"""
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_categorical_equal(c1, c2, **kwargs)
+    else:
+        tm.assert_categorical_equal(c1, c2, **kwargs)
+
+
+def test_categorical_equal_categories_mismatch():
+    msg = """Categorical\\.categories are different
+
+Categorical\\.categories values are different \\(25\\.0 %\\)
+\\[left\\]:  Index\\(\\[1, 2, 3, 4\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[1, 2, 3, 5\\], dtype='int64'\\)"""
+
+    c1 = Categorical([1, 2, 3, 4])
+    c2 = Categorical([1, 2, 3, 5])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_categorical_equal(c1, c2)
+
+
+def test_categorical_equal_codes_mismatch():
+    categories = [1, 2, 3, 4]
+    msg = """Categorical\\.codes are different
+
+Categorical\\.codes values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[0, 1, 3, 2\\]
+\\[right\\]: \\[0, 1, 2, 3\\]"""
+
+    c1 = Categorical([1, 2, 4, 3], categories=categories)
+    c2 = Categorical([1, 2, 3, 4], categories=categories)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_categorical_equal(c1, c2)
+
+
+def test_categorical_equal_ordered_mismatch():
+    data = [1, 2, 3, 4]
+    msg = """Categorical are different
+
+Attribute "ordered" are different
+\\[left\\]:  False
+\\[right\\]: True"""
+
+    c1 = Categorical(data, ordered=False)
+    c2 = Categorical(data, ordered=True)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_categorical_equal(c1, c2)
+
+
+@pytest.mark.parametrize("obj", ["index", "foo", "pandas"])
+def test_categorical_equal_object_override(obj):
+    data = [1, 2, 3, 4]
+    msg = f"""{obj} are different
+
+Attribute "ordered" are different
+\\[left\\]:  False
+\\[right\\]: True"""
+
+    c1 = Categorical(data, ordered=False)
+    c2 = Categorical(data, ordered=True)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_categorical_equal(c1, c2, obj=obj)
diff --git a/pandas/pandas/tests/util/test_assert_extension_array_equal.py b/pandas/pandas/tests/util/test_assert_extension_array_equal.py
new file mode 100644
index 0000000000000..5d82ae9af0e95
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_extension_array_equal.py
@@ -0,0 +1,125 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    Timestamp,
+    array,
+)
+import pandas._testing as tm
+from pandas.core.arrays.sparse import SparseArray
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {},  # Default is check_exact=False
+        {"check_exact": False},
+        {"check_exact": True},
+    ],
+)
+def test_assert_extension_array_equal_not_exact(kwargs):
+    # see gh-23709
+    arr1 = SparseArray([-0.17387645482451206, 0.3414148016424936])
+    arr2 = SparseArray([-0.17387645482451206, 0.3414148016424937])
+
+    if kwargs.get("check_exact", False):
+        msg = """\
+ExtensionArray are different
+
+ExtensionArray values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[-0\\.17387645482.*, 0\\.341414801642.*\\]
+\\[right\\]: \\[-0\\.17387645482.*, 0\\.341414801642.*\\]"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_extension_array_equal(arr1, arr2, **kwargs)
+    else:
+        tm.assert_extension_array_equal(arr1, arr2, **kwargs)
+
+
+@pytest.mark.parametrize("decimals", range(10))
+def test_assert_extension_array_equal_less_precise(decimals):
+    rtol = 0.5 * 10**-decimals
+    arr1 = SparseArray([0.5, 0.123456])
+    arr2 = SparseArray([0.5, 0.123457])
+
+    if decimals >= 5:
+        msg = """\
+ExtensionArray are different
+
+ExtensionArray values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[0\\.5, 0\\.123456\\]
+\\[right\\]: \\[0\\.5, 0\\.123457\\]"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_extension_array_equal(arr1, arr2, rtol=rtol)
+    else:
+        tm.assert_extension_array_equal(arr1, arr2, rtol=rtol)
+
+
+def test_assert_extension_array_equal_dtype_mismatch(check_dtype):
+    end = 5
+    kwargs = {"check_dtype": check_dtype}
+
+    arr1 = SparseArray(np.arange(end, dtype="int64"))
+    arr2 = SparseArray(np.arange(end, dtype="int32"))
+
+    if check_dtype:
+        msg = """\
+ExtensionArray are different
+
+Attribute "dtype" are different
+\\[left\\]:  Sparse\\[int64, 0\\]
+\\[right\\]: Sparse\\[int32, 0\\]"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_extension_array_equal(arr1, arr2, **kwargs)
+    else:
+        tm.assert_extension_array_equal(arr1, arr2, **kwargs)
+
+
+def test_assert_extension_array_equal_missing_values():
+    arr1 = SparseArray([np.nan, 1, 2, np.nan])
+    arr2 = SparseArray([np.nan, 1, 2, 3])
+
+    msg = """\
+ExtensionArray NA mask are different
+
+ExtensionArray NA mask values are different \\(25\\.0 %\\)
+\\[left\\]:  \\[True, False, False, True\\]
+\\[right\\]: \\[True, False, False, False\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_extension_array_equal(arr1, arr2)
+
+
+@pytest.mark.parametrize("side", ["left", "right"])
+def test_assert_extension_array_equal_non_extension_array(side):
+    numpy_array = np.arange(5)
+    extension_array = SparseArray(numpy_array)
+
+    msg = f"{side} is not an ExtensionArray"
+    args = (
+        (numpy_array, extension_array)
+        if side == "left"
+        else (extension_array, numpy_array)
+    )
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_extension_array_equal(*args)
+
+
+def test_assert_extension_array_equal_ignore_dtype_mismatch(any_int_dtype):
+    # https://github.com/pandas-dev/pandas/issues/35715
+    left = array([1, 2, 3], dtype="Int64")
+    right = array([1, 2, 3], dtype=any_int_dtype)
+    tm.assert_extension_array_equal(left, right, check_dtype=False)
+
+
+def test_assert_extension_array_equal_time_units():
+    # https://github.com/pandas-dev/pandas/issues/55730
+    timestamp = Timestamp("2023-11-04T12")
+    naive = array([timestamp], dtype="datetime64[ns]")
+    utc = array([timestamp], dtype="datetime64[ns, UTC]")
+
+    tm.assert_extension_array_equal(naive, utc, check_dtype=False)
+    tm.assert_extension_array_equal(utc, naive, check_dtype=False)
diff --git a/pandas/pandas/tests/util/test_assert_frame_equal.py b/pandas/pandas/tests/util/test_assert_frame_equal.py
new file mode 100644
index 0000000000000..ea954756d63c8
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_frame_equal.py
@@ -0,0 +1,397 @@
+import pytest
+
+import pandas as pd
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+@pytest.fixture(params=[True, False])
+def by_blocks_fixture(request):
+    return request.param
+
+
+def _assert_frame_equal_both(a, b, **kwargs):
+    """
+    Check that two DataFrame equal.
+
+    This check is performed commutatively.
+
+    Parameters
+    ----------
+    a : DataFrame
+        The first DataFrame to compare.
+    b : DataFrame
+        The second DataFrame to compare.
+    kwargs : dict
+        The arguments passed to `tm.assert_frame_equal`.
+    """
+    tm.assert_frame_equal(a, b, **kwargs)
+    tm.assert_frame_equal(b, a, **kwargs)
+
+
+@pytest.mark.parametrize("check_like", [True, False])
+def test_frame_equal_row_order_mismatch(check_like, frame_or_series):
+    df1 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["a", "b", "c"])
+    df2 = DataFrame({"A": [3, 2, 1], "B": [6, 5, 4]}, index=["c", "b", "a"])
+
+    if not check_like:  # Do not ignore row-column orderings.
+        msg = f"{frame_or_series.__name__}.index are different"
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_frame_equal(
+                df1, df2, check_like=check_like, obj=frame_or_series.__name__
+            )
+    else:
+        _assert_frame_equal_both(
+            df1, df2, check_like=check_like, obj=frame_or_series.__name__
+        )
+
+
+@pytest.mark.parametrize(
+    "df1,df2",
+    [
+        ({"A": [1, 2, 3]}, {"A": [1, 2, 3, 4]}),
+        ({"A": [1, 2, 3], "B": [4, 5, 6]}, {"A": [1, 2, 3]}),
+    ],
+)
+def test_frame_equal_shape_mismatch(df1, df2, frame_or_series):
+    df1 = DataFrame(df1)
+    df2 = DataFrame(df2)
+    msg = f"{frame_or_series.__name__} are different"
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2, obj=frame_or_series.__name__)
+
+
+@pytest.mark.parametrize(
+    "df1,df2,msg",
+    [
+        # Index
+        (
+            DataFrame.from_records({"a": [1, 2], "c": ["l1", "l2"]}, index=["a"]),
+            DataFrame.from_records({"a": [1.0, 2.0], "c": ["l1", "l2"]}, index=["a"]),
+            "DataFrame\\.index are different",
+        ),
+        # MultiIndex
+        (
+            DataFrame.from_records(
+                {"a": [1, 2], "b": [2.1, 1.5], "c": ["l1", "l2"]}, index=["a", "b"]
+            ),
+            DataFrame.from_records(
+                {"a": [1.0, 2.0], "b": [2.1, 1.5], "c": ["l1", "l2"]}, index=["a", "b"]
+            ),
+            "DataFrame\\.index level \\[0\\] are different",
+        ),
+    ],
+)
+def test_frame_equal_index_dtype_mismatch(df1, df2, msg, check_index_type):
+    kwargs = {"check_index_type": check_index_type}
+
+    if check_index_type:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_frame_equal(df1, df2, **kwargs)
+    else:
+        tm.assert_frame_equal(df1, df2, **kwargs)
+
+
+def test_empty_dtypes(check_dtype):
+    columns = ["col1", "col2"]
+    df1 = DataFrame(columns=columns)
+    df2 = DataFrame(columns=columns)
+
+    kwargs = {"check_dtype": check_dtype}
+    df1["col1"] = df1["col1"].astype("int64")
+
+    if check_dtype:
+        msg = r"Attributes of DataFrame\..* are different"
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_frame_equal(df1, df2, **kwargs)
+    else:
+        tm.assert_frame_equal(df1, df2, **kwargs)
+
+
+@pytest.mark.parametrize("check_like", [True, False])
+def test_frame_equal_index_mismatch(check_like, frame_or_series, using_infer_string):
+    if using_infer_string:
+        dtype = "str"
+    else:
+        dtype = "object"
+    msg = f"""{frame_or_series.__name__}\\.index are different
+
+{frame_or_series.__name__}\\.index values are different \\(33\\.33333 %\\)
+\\[left\\]:  Index\\(\\['a', 'b', 'c'\\], dtype='{dtype}'\\)
+\\[right\\]: Index\\(\\['a', 'b', 'd'\\], dtype='{dtype}'\\)
+At positional index 2, first diff: c != d"""
+
+    df1 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["a", "b", "c"])
+    df2 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["a", "b", "d"])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(
+            df1, df2, check_like=check_like, obj=frame_or_series.__name__
+        )
+
+
+@pytest.mark.parametrize("check_like", [True, False])
+def test_frame_equal_columns_mismatch(check_like, frame_or_series, using_infer_string):
+    if using_infer_string:
+        dtype = "str"
+    else:
+        dtype = "object"
+    msg = f"""{frame_or_series.__name__}\\.columns are different
+
+{frame_or_series.__name__}\\.columns values are different \\(50\\.0 %\\)
+\\[left\\]:  Index\\(\\['A', 'B'\\], dtype='{dtype}'\\)
+\\[right\\]: Index\\(\\['A', 'b'\\], dtype='{dtype}'\\)"""
+
+    df1 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]}, index=["a", "b", "c"])
+    df2 = DataFrame({"A": [1, 2, 3], "b": [4, 5, 6]}, index=["a", "b", "c"])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(
+            df1, df2, check_like=check_like, obj=frame_or_series.__name__
+        )
+
+
+def test_frame_equal_block_mismatch(by_blocks_fixture, frame_or_series):
+    obj = frame_or_series.__name__
+    msg = f"""{obj}\\.iloc\\[:, 1\\] \\(column name="B"\\) are different
+
+{obj}\\.iloc\\[:, 1\\] \\(column name="B"\\) values are different \\(33\\.33333 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\[4, 5, 6\\]
+\\[right\\]: \\[4, 5, 7\\]"""
+
+    df1 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 6]})
+    df2 = DataFrame({"A": [1, 2, 3], "B": [4, 5, 7]})
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2, by_blocks=by_blocks_fixture, obj=obj)
+
+
+@pytest.mark.parametrize(
+    "df1,df2,msg",
+    [
+        (
+            {"A": ["á", "à", "ä"], "E": ["é", "è", "ë"]},
+            {"A": ["á", "à", "ä"], "E": ["é", "è", "e̊"]},
+            """{obj}\\.iloc\\[:, 1\\] \\(column name="E"\\) are different
+
+{obj}\\.iloc\\[:, 1\\] \\(column name="E"\\) values are different \\(33\\.33333 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\[é, è, ë\\]
+\\[right\\]: \\[é, è, e̊\\]""",
+        ),
+        (
+            {"A": ["á", "à", "ä"], "E": ["é", "è", "ë"]},
+            {"A": ["a", "a", "a"], "E": ["e", "e", "e"]},
+            """{obj}\\.iloc\\[:, 0\\] \\(column name="A"\\) are different
+
+{obj}\\.iloc\\[:, 0\\] \\(column name="A"\\) values are different \\(100\\.0 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\[á, à, ä\\]
+\\[right\\]: \\[a, a, a\\]""",
+        ),
+    ],
+)
+def test_frame_equal_unicode(df1, df2, msg, by_blocks_fixture, frame_or_series):
+    # see gh-20503
+    #
+    # Test ensures that `tm.assert_frame_equals` raises the right exception
+    # when comparing DataFrames containing differing unicode objects.
+    df1 = DataFrame(df1)
+    df2 = DataFrame(df2)
+    msg = msg.format(obj=frame_or_series.__name__)
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(
+            df1, df2, by_blocks=by_blocks_fixture, obj=frame_or_series.__name__
+        )
+
+
+def test_assert_frame_equal_extension_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/32747
+    left = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    right = left.astype(int)
+
+    msg = (
+        "Attributes of DataFrame\\.iloc\\[:, 0\\] "
+        '\\(column name="a"\\) are different\n\n'
+        'Attribute "dtype" are different\n'
+        "\\[left\\]:  Int64\n"
+        "\\[right\\]: int[32|64]"
+    )
+
+    tm.assert_frame_equal(left, right, check_dtype=False)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(left, right, check_dtype=True)
+
+
+def test_assert_frame_equal_interval_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/32747
+    left = DataFrame({"a": [pd.Interval(0, 1)]}, dtype="interval")
+    right = left.astype(object)
+
+    msg = (
+        "Attributes of DataFrame\\.iloc\\[:, 0\\] "
+        '\\(column name="a"\\) are different\n\n'
+        'Attribute "dtype" are different\n'
+        "\\[left\\]:  interval\\[int64, right\\]\n"
+        "\\[right\\]: object"
+    )
+
+    tm.assert_frame_equal(left, right, check_dtype=False)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(left, right, check_dtype=True)
+
+
+def test_assert_frame_equal_ignore_extension_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/35715
+    left = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    right = DataFrame({"a": [1, 2, 3]}, dtype="Int32")
+    tm.assert_frame_equal(left, right, check_dtype=False)
+
+
+def test_assert_frame_equal_ignore_extension_dtype_mismatch_cross_class():
+    # https://github.com/pandas-dev/pandas/issues/35715
+    left = DataFrame({"a": [1, 2, 3]}, dtype="Int64")
+    right = DataFrame({"a": [1, 2, 3]}, dtype="int64")
+    tm.assert_frame_equal(left, right, check_dtype=False)
+
+
+@pytest.mark.parametrize(
+    "dtype", ["timedelta64[ns]", "datetime64[ns, UTC]", "Period[D]"]
+)
+def test_assert_frame_equal_datetime_like_dtype_mismatch(dtype):
+    df1 = DataFrame({"a": []}, dtype=dtype)
+    df2 = DataFrame({"a": []})
+    tm.assert_frame_equal(df1, df2, check_dtype=False)
+
+
+def test_allows_duplicate_labels():
+    left = DataFrame()
+    right = DataFrame().set_flags(allows_duplicate_labels=False)
+    tm.assert_frame_equal(left, left)
+    tm.assert_frame_equal(right, right)
+    tm.assert_frame_equal(left, right, check_flags=False)
+    tm.assert_frame_equal(right, left, check_flags=False)
+
+    with pytest.raises(AssertionError, match="<Flags"):
+        tm.assert_frame_equal(left, right)
+
+    with pytest.raises(AssertionError, match="<Flags"):
+        tm.assert_frame_equal(left, right)
+
+
+def test_assert_frame_equal_columns_mixed_dtype():
+    # GH#39168
+    df = DataFrame([[0, 1, 2]], columns=["foo", "bar", 42], index=[1, "test", 2])
+    tm.assert_frame_equal(df, df, check_like=True)
+
+
+def test_frame_equal_extension_dtype(frame_or_series, any_numeric_ea_dtype):
+    # GH#39410
+    obj = frame_or_series([1, 2], dtype=any_numeric_ea_dtype)
+    tm.assert_equal(obj, obj, check_exact=True)
+
+
+@pytest.mark.parametrize("indexer", [(0, 1), (1, 0)])
+def test_frame_equal_mixed_dtypes(frame_or_series, any_numeric_ea_dtype, indexer):
+    dtypes = (any_numeric_ea_dtype, "int64")
+    obj1 = frame_or_series([1, 2], dtype=dtypes[indexer[0]])
+    obj2 = frame_or_series([1, 2], dtype=dtypes[indexer[1]])
+    tm.assert_equal(obj1, obj2, check_exact=True, check_dtype=False)
+
+
+def test_assert_frame_equal_check_like_different_indexes():
+    # GH#39739
+    df1 = DataFrame(index=pd.Index([], dtype="object"))
+    df2 = DataFrame(index=pd.RangeIndex(start=0, stop=0, step=1))
+    with pytest.raises(AssertionError, match="DataFrame.index are different"):
+        tm.assert_frame_equal(df1, df2, check_like=True)
+
+
+def test_assert_frame_equal_checking_allow_dups_flag():
+    # GH#45554
+    left = DataFrame([[1, 2], [3, 4]])
+    left.flags.allows_duplicate_labels = False
+
+    right = DataFrame([[1, 2], [3, 4]])
+    right.flags.allows_duplicate_labels = True
+    tm.assert_frame_equal(left, right, check_flags=False)
+
+    with pytest.raises(AssertionError, match="allows_duplicate_labels"):
+        tm.assert_frame_equal(left, right, check_flags=True)
+
+
+def test_assert_frame_equal_check_like_categorical_midx():
+    # GH#48975
+    left = DataFrame(
+        [[1], [2], [3]],
+        index=pd.MultiIndex.from_arrays(
+            [
+                pd.Categorical(["a", "b", "c"]),
+                pd.Categorical(["a", "b", "c"]),
+            ]
+        ),
+    )
+    right = DataFrame(
+        [[3], [2], [1]],
+        index=pd.MultiIndex.from_arrays(
+            [
+                pd.Categorical(["c", "b", "a"]),
+                pd.Categorical(["c", "b", "a"]),
+            ]
+        ),
+    )
+    tm.assert_frame_equal(left, right, check_like=True)
+
+
+def test_assert_frame_equal_ea_column_definition_in_exception_mask():
+    # GH#50323
+    df1 = DataFrame({"a": pd.Series([pd.NA, 1], dtype="Int64")})
+    df2 = DataFrame({"a": pd.Series([1, 1], dtype="Int64")})
+
+    msg = r'DataFrame.iloc\[:, 0\] \(column name="a"\) NA mask values are different'
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2)
+
+
+def test_assert_frame_equal_ea_column_definition_in_exception():
+    # GH#50323
+    df1 = DataFrame({"a": pd.Series([pd.NA, 1], dtype="Int64")})
+    df2 = DataFrame({"a": pd.Series([pd.NA, 2], dtype="Int64")})
+
+    msg = r'DataFrame.iloc\[:, 0\] \(column name="a"\) values are different'
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2, check_exact=True)
+
+
+def test_assert_frame_equal_ts_column():
+    # GH#50323
+    df1 = DataFrame({"a": [pd.Timestamp("2019-12-31"), pd.Timestamp("2020-12-31")]})
+    df2 = DataFrame({"a": [pd.Timestamp("2020-12-31"), pd.Timestamp("2020-12-31")]})
+
+    msg = r'DataFrame.iloc\[:, 0\] \(column name="a"\) values are different'
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2)
+
+
+def test_assert_frame_equal_set():
+    # GH#51727
+    df1 = DataFrame({"set_column": [{1, 2, 3}, {4, 5, 6}]})
+    df2 = DataFrame({"set_column": [{1, 2, 3}, {4, 5, 6}]})
+    tm.assert_frame_equal(df1, df2)
+
+
+def test_assert_frame_equal_set_mismatch():
+    # GH#51727
+    df1 = DataFrame({"set_column": [{1, 2, 3}, {4, 5, 6}]})
+    df2 = DataFrame({"set_column": [{1, 2, 3}, {4, 5, 7}]})
+
+    msg = r'DataFrame.iloc\[:, 0\] \(column name="set_column"\) values are different'
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_frame_equal(df1, df2)
diff --git a/pandas/pandas/tests/util/test_assert_index_equal.py b/pandas/pandas/tests/util/test_assert_index_equal.py
new file mode 100644
index 0000000000000..ab52d6c8e9f39
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_index_equal.py
@@ -0,0 +1,319 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    NA,
+    Categorical,
+    CategoricalIndex,
+    Index,
+    MultiIndex,
+    NaT,
+    RangeIndex,
+)
+import pandas._testing as tm
+
+
+def test_index_equal_levels_mismatch():
+    msg = """Index are different
+
+Index levels are different
+\\[left\\]:  1, Index\\(\\[1, 2, 3\\], dtype='int64'\\)
+\\[right\\]: 2, MultiIndex\\(\\[\\('A', 1\\),
+            \\('A', 2\\),
+            \\('B', 3\\),
+            \\('B', 4\\)\\],
+           \\)"""
+
+    idx1 = Index([1, 2, 3])
+    idx2 = MultiIndex.from_tuples([("A", 1), ("A", 2), ("B", 3), ("B", 4)])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, exact=False)
+
+
+def test_index_equal_values_mismatch(check_exact):
+    msg = """MultiIndex level \\[1\\] are different
+
+MultiIndex level \\[1\\] values are different \\(25\\.0 %\\)
+\\[left\\]:  Index\\(\\[2, 2, 3, 4\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[1, 2, 3, 4\\], dtype='int64'\\)"""
+
+    idx1 = MultiIndex.from_tuples([("A", 2), ("A", 2), ("B", 3), ("B", 4)])
+    idx2 = MultiIndex.from_tuples([("A", 1), ("A", 2), ("B", 3), ("B", 4)])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, check_exact=check_exact)
+
+
+def test_index_equal_length_mismatch(check_exact):
+    msg = """Index are different
+
+Index length are different
+\\[left\\]:  3, Index\\(\\[1, 2, 3\\], dtype='int64'\\)
+\\[right\\]: 4, Index\\(\\[1, 2, 3, 4\\], dtype='int64'\\)"""
+
+    idx1 = Index([1, 2, 3])
+    idx2 = Index([1, 2, 3, 4])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, check_exact=check_exact)
+
+
+@pytest.mark.parametrize("exact", [False, "equiv"])
+def test_index_equal_class(exact):
+    idx1 = Index([0, 1, 2])
+    idx2 = RangeIndex(3)
+
+    tm.assert_index_equal(idx1, idx2, exact=exact)
+
+
+def test_int_float_index_equal_class_mismatch(check_exact):
+    msg = """Index are different
+
+Attribute "inferred_type" are different
+\\[left\\]:  integer
+\\[right\\]: floating"""
+
+    idx1 = Index([1, 2, 3])
+    idx2 = Index([1, 2, 3], dtype=np.float64)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, exact=True, check_exact=check_exact)
+
+
+def test_range_index_equal_class_mismatch(check_exact):
+    msg = """Index are different
+
+Index classes are different
+\\[left\\]:  Index\\(\\[1, 2, 3\\], dtype='int64'\\)
+\\[right\\]: """
+
+    idx1 = Index([1, 2, 3])
+    idx2 = RangeIndex(range(3))
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, exact=True, check_exact=check_exact)
+
+
+def test_index_equal_values_close(check_exact):
+    idx1 = Index([1, 2, 3.0])
+    idx2 = Index([1, 2, 3.0000000001])
+
+    if check_exact:
+        msg = """Index are different
+
+Index values are different \\(33\\.33333 %\\)
+\\[left\\]:  Index\\(\\[1.0, 2.0, 3.0], dtype='float64'\\)
+\\[right\\]: Index\\(\\[1.0, 2.0, 3.0000000001\\], dtype='float64'\\)"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(idx1, idx2, check_exact=check_exact)
+    else:
+        tm.assert_index_equal(idx1, idx2, check_exact=check_exact)
+
+
+def test_index_equal_values_less_close(check_exact, rtol):
+    idx1 = Index([1, 2, 3.0])
+    idx2 = Index([1, 2, 3.0001])
+    kwargs = {"check_exact": check_exact, "rtol": rtol}
+
+    if check_exact or rtol < 0.5e-3:
+        msg = """Index are different
+
+Index values are different \\(33\\.33333 %\\)
+\\[left\\]:  Index\\(\\[1.0, 2.0, 3.0], dtype='float64'\\)
+\\[right\\]: Index\\(\\[1.0, 2.0, 3.0001\\], dtype='float64'\\)"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(idx1, idx2, **kwargs)
+    else:
+        tm.assert_index_equal(idx1, idx2, **kwargs)
+
+
+def test_index_equal_values_too_far(check_exact, rtol):
+    idx1 = Index([1, 2, 3])
+    idx2 = Index([1, 2, 4])
+    kwargs = {"check_exact": check_exact, "rtol": rtol}
+
+    msg = """Index are different
+
+Index values are different \\(33\\.33333 %\\)
+\\[left\\]:  Index\\(\\[1, 2, 3\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[1, 2, 4\\], dtype='int64'\\)"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, **kwargs)
+
+
+@pytest.mark.parametrize("check_order", [True, False])
+def test_index_equal_value_order_mismatch(check_exact, rtol, check_order):
+    idx1 = Index([1, 2, 3])
+    idx2 = Index([3, 2, 1])
+
+    msg = """Index are different
+
+Index values are different \\(66\\.66667 %\\)
+\\[left\\]:  Index\\(\\[1, 2, 3\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[3, 2, 1\\], dtype='int64'\\)"""
+
+    if check_order:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(
+                idx1, idx2, check_exact=check_exact, rtol=rtol, check_order=True
+            )
+    else:
+        tm.assert_index_equal(
+            idx1, idx2, check_exact=check_exact, rtol=rtol, check_order=False
+        )
+
+
+def test_index_equal_level_values_mismatch(check_exact, rtol):
+    idx1 = MultiIndex.from_tuples([("A", 2), ("A", 2), ("B", 3), ("B", 4)])
+    idx2 = MultiIndex.from_tuples([("A", 1), ("A", 2), ("B", 3), ("B", 4)])
+    kwargs = {"check_exact": check_exact, "rtol": rtol}
+
+    msg = """MultiIndex level \\[1\\] are different
+
+MultiIndex level \\[1\\] values are different \\(25\\.0 %\\)
+\\[left\\]:  Index\\(\\[2, 2, 3, 4\\], dtype='int64'\\)
+\\[right\\]: Index\\(\\[1, 2, 3, 4\\], dtype='int64'\\)"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2, **kwargs)
+
+
+@pytest.mark.parametrize(
+    "name1,name2",
+    [(None, "x"), ("x", "x"), (np.nan, np.nan), (NaT, NaT), (np.nan, NaT)],
+)
+def test_index_equal_names(name1, name2):
+    idx1 = Index([1, 2, 3], name=name1)
+    idx2 = Index([1, 2, 3], name=name2)
+
+    if name1 == name2 or name1 is name2:
+        tm.assert_index_equal(idx1, idx2)
+    else:
+        name1 = "'x'" if name1 == "x" else name1
+        name2 = "'x'" if name2 == "x" else name2
+        msg = f"""Index are different
+
+Attribute "names" are different
+\\[left\\]:  \\[{name1}\\]
+\\[right\\]: \\[{name2}\\]"""
+
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(idx1, idx2)
+
+
+def test_index_equal_category_mismatch(check_categorical, using_infer_string):
+    if using_infer_string:
+        dtype = "str"
+    else:
+        dtype = "object"
+    msg = f"""Index are different
+
+Attribute "dtype" are different
+\\[left\\]:  CategoricalDtype\\(categories=\\['a', 'b'\\], ordered=False, \
+categories_dtype={dtype}\\)
+\\[right\\]: CategoricalDtype\\(categories=\\['a', 'b', 'c'\\], \
+ordered=False, categories_dtype={dtype}\\)"""
+
+    idx1 = Index(Categorical(["a", "b"]))
+    idx2 = Index(Categorical(["a", "b"], categories=["a", "b", "c"]))
+
+    if check_categorical:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(idx1, idx2, check_categorical=check_categorical)
+    else:
+        tm.assert_index_equal(idx1, idx2, check_categorical=check_categorical)
+
+
+@pytest.mark.parametrize("exact", [False, True])
+def test_index_equal_range_categories(check_categorical, exact):
+    # GH41263
+    msg = """\
+Index are different
+
+Index classes are different
+\\[left\\]:  RangeIndex\\(start=0, stop=10, step=1\\)
+\\[right\\]: Index\\(\\[0, 1, 2, 3, 4, 5, 6, 7, 8, 9\\], dtype='int64'\\)"""
+
+    rcat = CategoricalIndex(RangeIndex(10))
+    icat = CategoricalIndex(list(range(10)))
+
+    if check_categorical and exact:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_index_equal(rcat, icat, check_categorical=True, exact=True)
+    else:
+        tm.assert_index_equal(
+            rcat, icat, check_categorical=check_categorical, exact=exact
+        )
+
+
+def test_assert_index_equal_different_inferred_types():
+    # GH#31884
+    msg = """\
+Index are different
+
+Attribute "inferred_type" are different
+\\[left\\]:  mixed
+\\[right\\]: datetime"""
+
+    idx1 = Index([NA, np.datetime64("nat")])
+    idx2 = Index([NA, NaT])
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_index_equal(idx1, idx2)
+
+
+def test_assert_index_equal_different_names_check_order_false():
+    # GH#47328
+    idx1 = Index([1, 3], name="a")
+    idx2 = Index([3, 1], name="b")
+    with pytest.raises(AssertionError, match='"names" are different'):
+        tm.assert_index_equal(idx1, idx2, check_order=False, check_names=True)
+
+
+def test_assert_index_equal_mixed_dtype():
+    # GH#39168
+    idx = Index(["foo", "bar", 42])
+    tm.assert_index_equal(idx, idx, check_order=False)
+
+
+def test_assert_index_equal_ea_dtype_order_false(any_numeric_ea_dtype):
+    # GH#47207
+    idx1 = Index([1, 3], dtype=any_numeric_ea_dtype)
+    idx2 = Index([3, 1], dtype=any_numeric_ea_dtype)
+    tm.assert_index_equal(idx1, idx2, check_order=False)
+
+
+def test_assert_index_equal_object_ints_order_false():
+    # GH#47207
+    idx1 = Index([1, 3], dtype="object")
+    idx2 = Index([3, 1], dtype="object")
+    tm.assert_index_equal(idx1, idx2, check_order=False)
+
+
+@pytest.mark.parametrize("check_categorical", [True, False])
+@pytest.mark.parametrize("check_names", [True, False])
+def test_assert_ea_index_equal_non_matching_na(check_names, check_categorical):
+    # GH#48608
+    idx1 = Index([1, 2], dtype="Int64")
+    idx2 = Index([1, NA], dtype="Int64")
+    with pytest.raises(AssertionError, match="50.0 %"):
+        tm.assert_index_equal(
+            idx1, idx2, check_names=check_names, check_categorical=check_categorical
+        )
+
+
+@pytest.mark.parametrize("check_categorical", [True, False])
+def test_assert_multi_index_dtype_check_categorical(check_categorical):
+    # GH#52126
+    idx1 = MultiIndex.from_arrays([Categorical(np.array([1, 2], dtype=np.uint64))])
+    idx2 = MultiIndex.from_arrays([Categorical(np.array([1, 2], dtype=np.int64))])
+    if check_categorical:
+        with pytest.raises(
+            AssertionError, match=r"^MultiIndex level \[0\] are different"
+        ):
+            tm.assert_index_equal(idx1, idx2, check_categorical=check_categorical)
+    else:
+        tm.assert_index_equal(idx1, idx2, check_categorical=check_categorical)
diff --git a/pandas/pandas/tests/util/test_assert_interval_array_equal.py b/pandas/pandas/tests/util/test_assert_interval_array_equal.py
new file mode 100644
index 0000000000000..aad27672c0f6f
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_interval_array_equal.py
@@ -0,0 +1,100 @@
+import pytest
+
+from pandas import (
+    Interval,
+    interval_range,
+)
+import pandas._testing as tm
+from pandas.arrays import IntervalArray
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"start": 0, "periods": 4},
+        {"start": 1, "periods": 5},
+        {"start": 5, "end": 10, "closed": "left"},
+    ],
+)
+def test_interval_array_equal(kwargs):
+    arr = interval_range(**kwargs).values
+    tm.assert_interval_array_equal(arr, arr)
+
+
+def test_interval_array_equal_closed_mismatch():
+    kwargs = {"start": 0, "periods": 5}
+    arr1 = interval_range(closed="left", **kwargs).values
+    arr2 = interval_range(closed="right", **kwargs).values
+
+    msg = """\
+IntervalArray are different
+
+Attribute "closed" are different
+\\[left\\]:  left
+\\[right\\]: right"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_interval_array_equal(arr1, arr2)
+
+
+def test_interval_array_equal_periods_mismatch():
+    kwargs = {"start": 0}
+    arr1 = interval_range(periods=5, **kwargs).values
+    arr2 = interval_range(periods=6, **kwargs).values
+
+    msg = """\
+IntervalArray.left are different
+
+IntervalArray.left shapes are different
+\\[left\\]:  \\(5,\\)
+\\[right\\]: \\(6,\\)"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_interval_array_equal(arr1, arr2)
+
+
+def test_interval_array_equal_end_mismatch():
+    kwargs = {"start": 0, "periods": 5}
+    arr1 = interval_range(end=10, **kwargs).values
+    arr2 = interval_range(end=20, **kwargs).values
+
+    msg = """\
+IntervalArray.left are different
+
+IntervalArray.left values are different \\(80.0 %\\)
+\\[left\\]:  \\[0, 2, 4, 6, 8\\]
+\\[right\\]: \\[0, 4, 8, 12, 16\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_interval_array_equal(arr1, arr2)
+
+
+def test_interval_array_equal_start_mismatch():
+    kwargs = {"periods": 4}
+    arr1 = interval_range(start=0, **kwargs).values
+    arr2 = interval_range(start=1, **kwargs).values
+
+    msg = """\
+IntervalArray.left are different
+
+IntervalArray.left values are different \\(100.0 %\\)
+\\[left\\]:  \\[0, 1, 2, 3\\]
+\\[right\\]: \\[1, 2, 3, 4\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_interval_array_equal(arr1, arr2)
+
+
+def test_interval_array_equal_end_mismatch_only():
+    arr1 = IntervalArray([Interval(0, 1), Interval(0, 5)])
+    arr2 = IntervalArray([Interval(0, 1), Interval(0, 6)])
+
+    msg = """\
+IntervalArray.right are different
+
+IntervalArray.right values are different \\(50.0 %\\)
+\\[left\\]:  \\[1, 5\\]
+\\[right\\]: \\[1, 6\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_interval_array_equal(arr1, arr2)
diff --git a/pandas/pandas/tests/util/test_assert_numpy_array_equal.py b/pandas/pandas/tests/util/test_assert_numpy_array_equal.py
new file mode 100644
index 0000000000000..660402ee857e3
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_numpy_array_equal.py
@@ -0,0 +1,223 @@
+import copy
+
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import Timestamp
+import pandas._testing as tm
+
+
+def test_assert_numpy_array_equal_shape_mismatch():
+    msg = """numpy array are different
+
+numpy array shapes are different
+\\[left\\]:  \\(2L*,\\)
+\\[right\\]: \\(3L*,\\)"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(np.array([1, 2]), np.array([3, 4, 5]))
+
+
+def test_assert_numpy_array_equal_bad_type():
+    expected = "Expected type"
+
+    with pytest.raises(AssertionError, match=expected):
+        tm.assert_numpy_array_equal(1, 2)
+
+
+@pytest.mark.parametrize(
+    "a,b,klass1,klass2",
+    [(np.array([1]), 1, "ndarray", "int"), (1, np.array([1]), "int", "ndarray")],
+)
+def test_assert_numpy_array_equal_class_mismatch(a, b, klass1, klass2):
+    msg = f"""numpy array are different
+
+numpy array classes are different
+\\[left\\]:  {klass1}
+\\[right\\]: {klass2}"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(a, b)
+
+
+def test_assert_numpy_array_equal_value_mismatch1():
+    msg = """numpy array are different
+
+numpy array values are different \\(66\\.66667 %\\)
+\\[left\\]:  \\[nan, 2\\.0, 3\\.0\\]
+\\[right\\]: \\[1\\.0, nan, 3\\.0\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(np.array([np.nan, 2, 3]), np.array([1, np.nan, 3]))
+
+
+def test_assert_numpy_array_equal_value_mismatch2():
+    msg = """numpy array are different
+
+numpy array values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[1, 2\\]
+\\[right\\]: \\[1, 3\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(np.array([1, 2]), np.array([1, 3]))
+
+
+def test_assert_numpy_array_equal_value_mismatch3():
+    msg = """numpy array are different
+
+numpy array values are different \\(16\\.66667 %\\)
+\\[left\\]:  \\[\\[1, 2\\], \\[3, 4\\], \\[5, 6\\]\\]
+\\[right\\]: \\[\\[1, 3\\], \\[3, 4\\], \\[5, 6\\]\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(
+            np.array([[1, 2], [3, 4], [5, 6]]), np.array([[1, 3], [3, 4], [5, 6]])
+        )
+
+
+def test_assert_numpy_array_equal_value_mismatch4():
+    msg = """numpy array are different
+
+numpy array values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[1\\.1, 2\\.000001\\]
+\\[right\\]: \\[1\\.1, 2.0\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(np.array([1.1, 2.000001]), np.array([1.1, 2.0]))
+
+
+def test_assert_numpy_array_equal_value_mismatch5():
+    msg = """numpy array are different
+
+numpy array values are different \\(16\\.66667 %\\)
+\\[left\\]:  \\[\\[1, 2\\], \\[3, 4\\], \\[5, 6\\]\\]
+\\[right\\]: \\[\\[1, 3\\], \\[3, 4\\], \\[5, 6\\]\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(
+            np.array([[1, 2], [3, 4], [5, 6]]), np.array([[1, 3], [3, 4], [5, 6]])
+        )
+
+
+def test_assert_numpy_array_equal_value_mismatch6():
+    msg = """numpy array are different
+
+numpy array values are different \\(25\\.0 %\\)
+\\[left\\]:  \\[\\[1, 2\\], \\[3, 4\\]\\]
+\\[right\\]: \\[\\[1, 3\\], \\[3, 4\\]\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(
+            np.array([[1, 2], [3, 4]]), np.array([[1, 3], [3, 4]])
+        )
+
+
+def test_assert_numpy_array_equal_shape_mismatch_override():
+    msg = """Index are different
+
+Index shapes are different
+\\[left\\]:  \\(2L*,\\)
+\\[right\\]: \\(3L*,\\)"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(np.array([1, 2]), np.array([3, 4, 5]), obj="Index")
+
+
+def test_numpy_array_equal_unicode():
+    # see gh-20503
+    #
+    # Test ensures that `tm.assert_numpy_array_equals` raises the right
+    # exception when comparing np.arrays containing differing unicode objects.
+    msg = """numpy array are different
+
+numpy array values are different \\(33\\.33333 %\\)
+\\[left\\]:  \\[á, à, ä\\]
+\\[right\\]: \\[á, à, å\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(
+            np.array(["á", "à", "ä"]), np.array(["á", "à", "å"])
+        )
+
+
+def test_numpy_array_equal_object():
+    a = np.array([Timestamp("2011-01-01"), Timestamp("2011-01-01")])
+    b = np.array([Timestamp("2011-01-01"), Timestamp("2011-01-02")])
+
+    msg = """numpy array are different
+
+numpy array values are different \\(50\\.0 %\\)
+\\[left\\]:  \\[2011-01-01 00:00:00, 2011-01-01 00:00:00\\]
+\\[right\\]: \\[2011-01-01 00:00:00, 2011-01-02 00:00:00\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(a, b)
+
+
+@pytest.mark.parametrize("other_type", ["same", "copy"])
+@pytest.mark.parametrize("check_same", ["same", "copy"])
+def test_numpy_array_equal_copy_flag(other_type, check_same):
+    a = np.array([1, 2, 3])
+    msg = None
+
+    if other_type == "same":
+        other = a.view()
+    else:
+        other = a.copy()
+
+    if check_same != other_type:
+        msg = (
+            r"array\(\[1, 2, 3\]\) is not array\(\[1, 2, 3\]\)"
+            if check_same == "same"
+            else r"array\(\[1, 2, 3\]\) is array\(\[1, 2, 3\]\)"
+        )
+
+    if msg is not None:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_numpy_array_equal(a, other, check_same=check_same)
+    else:
+        tm.assert_numpy_array_equal(a, other, check_same=check_same)
+
+
+def test_numpy_array_equal_contains_na():
+    # https://github.com/pandas-dev/pandas/issues/31881
+    a = np.array([True, False])
+    b = np.array([True, pd.NA], dtype=object)
+
+    msg = """numpy array are different
+
+numpy array values are different \\(50.0 %\\)
+\\[left\\]:  \\[True, False\\]
+\\[right\\]: \\[True, <NA>\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(a, b)
+
+
+def test_numpy_array_equal_identical_na(nulls_fixture):
+    a = np.array([nulls_fixture], dtype=object)
+
+    tm.assert_numpy_array_equal(a, a)
+
+    # matching but not the identical object
+    if hasattr(nulls_fixture, "copy"):
+        other = nulls_fixture.copy()
+    else:
+        other = copy.copy(nulls_fixture)
+    b = np.array([other], dtype=object)
+    tm.assert_numpy_array_equal(a, b)
+
+
+def test_numpy_array_equal_different_na():
+    a = np.array([np.nan], dtype=object)
+    b = np.array([pd.NA], dtype=object)
+
+    msg = """numpy array are different
+
+numpy array values are different \\(100.0 %\\)
+\\[left\\]:  \\[nan\\]
+\\[right\\]: \\[<NA>\\]"""
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_numpy_array_equal(a, b)
diff --git a/pandas/pandas/tests/util/test_assert_produces_warning.py b/pandas/pandas/tests/util/test_assert_produces_warning.py
new file mode 100644
index 0000000000000..5b917dbbe7ba7
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_produces_warning.py
@@ -0,0 +1,271 @@
+""" "
+Test module for testing ``pandas._testing.assert_produces_warning``.
+"""
+
+import warnings
+
+import pytest
+
+from pandas.errors import (
+    DtypeWarning,
+    PerformanceWarning,
+)
+
+import pandas._testing as tm
+
+
+@pytest.fixture(
+    params=[
+        (RuntimeWarning, UserWarning),
+        (UserWarning, FutureWarning),
+        (FutureWarning, RuntimeWarning),
+        (DeprecationWarning, PerformanceWarning),
+        (PerformanceWarning, FutureWarning),
+        (DtypeWarning, DeprecationWarning),
+        (ResourceWarning, DeprecationWarning),
+        (FutureWarning, DeprecationWarning),
+    ],
+    ids=lambda x: type(x).__name__,
+)
+def pair_different_warnings(request):
+    """
+    Return pair or different warnings.
+
+    Useful for testing how several different warnings are handled
+    in tm.assert_produces_warning.
+    """
+    return request.param
+
+
+def f():
+    warnings.warn("f1", FutureWarning)
+    warnings.warn("f2", RuntimeWarning)
+
+
+def test_assert_produces_warning_honors_filter():
+    # Raise by default.
+    msg = r"Caused unexpected warning\(s\)"
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(RuntimeWarning):
+            f()
+
+    with tm.assert_produces_warning(RuntimeWarning, raise_on_extra_warnings=False):
+        f()
+
+
+@pytest.mark.parametrize(
+    "category",
+    [
+        RuntimeWarning,
+        ResourceWarning,
+        UserWarning,
+        FutureWarning,
+        DeprecationWarning,
+        PerformanceWarning,
+        DtypeWarning,
+    ],
+)
+@pytest.mark.parametrize(
+    "message, match",
+    [
+        ("", None),
+        ("", ""),
+        ("Warning message", r".*"),
+        ("Warning message", "War"),
+        ("Warning message", r"[Ww]arning"),
+        ("Warning message", "age"),
+        ("Warning message", r"age$"),
+        ("Message 12-234 with numbers", r"\d{2}-\d{3}"),
+        ("Message 12-234 with numbers", r"^Mes.*\d{2}-\d{3}"),
+        ("Message 12-234 with numbers", r"\d{2}-\d{3}\s\S+"),
+        ("Message, which we do not match", None),
+    ],
+)
+def test_catch_warning_category_and_match(category, message, match):
+    with tm.assert_produces_warning(category, match=match):
+        warnings.warn(message, category)
+
+
+def test_fail_to_match_runtime_warning():
+    category = RuntimeWarning
+    match = "Did not see this warning"
+    unmatched = (
+        r"Did not see warning 'RuntimeWarning' matching 'Did not see this warning'. "
+        r"The emitted warning messages are "
+        r"\[RuntimeWarning\('This is not a match.'\), "
+        r"RuntimeWarning\('Another unmatched warning.'\)\]"
+    )
+    with pytest.raises(AssertionError, match=unmatched):
+        with tm.assert_produces_warning(category, match=match):
+            warnings.warn("This is not a match.", category)
+            warnings.warn("Another unmatched warning.", category)
+
+
+def test_fail_to_match_future_warning():
+    category = FutureWarning
+    match = "Warning"
+    unmatched = (
+        r"Did not see warning 'FutureWarning' matching 'Warning'. "
+        r"The emitted warning messages are "
+        r"\[FutureWarning\('This is not a match.'\), "
+        r"FutureWarning\('Another unmatched warning.'\)\]"
+    )
+    with pytest.raises(AssertionError, match=unmatched):
+        with tm.assert_produces_warning(category, match=match):
+            warnings.warn("This is not a match.", category)
+            warnings.warn("Another unmatched warning.", category)
+
+
+def test_fail_to_match_resource_warning():
+    category = ResourceWarning
+    match = r"\d+"
+    unmatched = (
+        r"Did not see warning 'ResourceWarning' matching '\\d\+'. "
+        r"The emitted warning messages are "
+        r"\[ResourceWarning\('This is not a match.'\), "
+        r"ResourceWarning\('Another unmatched warning.'\)\]"
+    )
+    with pytest.raises(AssertionError, match=unmatched):
+        with tm.assert_produces_warning(category, match=match):
+            warnings.warn("This is not a match.", category)
+            warnings.warn("Another unmatched warning.", category)
+
+
+def test_fail_to_catch_actual_warning(pair_different_warnings):
+    expected_category, actual_category = pair_different_warnings
+    match = "Did not see expected warning of class"
+    with pytest.raises(AssertionError, match=match):
+        with tm.assert_produces_warning(expected_category):
+            warnings.warn("warning message", actual_category)
+
+
+def test_ignore_extra_warning(pair_different_warnings):
+    expected_category, extra_category = pair_different_warnings
+    with tm.assert_produces_warning(expected_category, raise_on_extra_warnings=False):
+        warnings.warn("Expected warning", expected_category)
+        warnings.warn("Unexpected warning OK", extra_category)
+
+
+def test_raise_on_extra_warning(pair_different_warnings):
+    expected_category, extra_category = pair_different_warnings
+    match = r"Caused unexpected warning\(s\)"
+    with pytest.raises(AssertionError, match=match):
+        with tm.assert_produces_warning(expected_category):
+            warnings.warn("Expected warning", expected_category)
+            warnings.warn("Unexpected warning NOT OK", extra_category)
+
+
+def test_same_category_different_messages_first_match():
+    category = UserWarning
+    with tm.assert_produces_warning(category, match=r"^Match this"):
+        warnings.warn("Match this", category)
+        warnings.warn("Do not match that", category)
+        warnings.warn("Do not match that either", category)
+
+
+def test_same_category_different_messages_last_match():
+    category = DeprecationWarning
+    with tm.assert_produces_warning(category, match=r"^Match this"):
+        warnings.warn("Do not match that", category)
+        warnings.warn("Do not match that either", category)
+        warnings.warn("Match this", category)
+
+
+def test_match_multiple_warnings():
+    # https://github.com/pandas-dev/pandas/issues/47829
+    category = (FutureWarning, UserWarning)
+    with tm.assert_produces_warning(category, match=r"^Match this"):
+        warnings.warn("Match this", FutureWarning)
+        warnings.warn("Match this too", UserWarning)
+
+
+def test_must_match_multiple_warnings():
+    # https://github.com/pandas-dev/pandas/issues/56555
+    category = (FutureWarning, UserWarning)
+    msg = "Did not see expected warning of class 'UserWarning'"
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(category, match=r"^Match this"):
+            warnings.warn("Match this", FutureWarning)
+
+
+def test_must_match_multiple_warnings_messages():
+    # https://github.com/pandas-dev/pandas/issues/56555
+    category = (FutureWarning, UserWarning)
+    msg = r"The emitted warning messages are \[UserWarning\('Not this'\)\]"
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(category, match=r"^Match this"):
+            warnings.warn("Match this", FutureWarning)
+            warnings.warn("Not this", UserWarning)
+
+
+def test_allow_partial_match_for_multiple_warnings():
+    # https://github.com/pandas-dev/pandas/issues/56555
+    category = (FutureWarning, UserWarning)
+    with tm.assert_produces_warning(
+        category, match=r"^Match this", must_find_all_warnings=False
+    ):
+        warnings.warn("Match this", FutureWarning)
+
+
+def test_allow_partial_match_for_multiple_warnings_messages():
+    # https://github.com/pandas-dev/pandas/issues/56555
+    category = (FutureWarning, UserWarning)
+    with tm.assert_produces_warning(
+        category, match=r"^Match this", must_find_all_warnings=False
+    ):
+        warnings.warn("Match this", FutureWarning)
+        warnings.warn("Not this", UserWarning)
+
+
+def test_right_category_wrong_match_raises(pair_different_warnings):
+    target_category, other_category = pair_different_warnings
+    with pytest.raises(AssertionError, match="Did not see warning.*matching"):
+        with tm.assert_produces_warning(target_category, match=r"^Match this"):
+            warnings.warn("Do not match it", target_category)
+            warnings.warn("Match this", other_category)
+
+
+@pytest.mark.parametrize("false_or_none", [False, None])
+class TestFalseOrNoneExpectedWarning:
+    def test_raise_on_warning(self, false_or_none):
+        msg = r"Caused unexpected warning\(s\)"
+        with pytest.raises(AssertionError, match=msg):
+            with tm.assert_produces_warning(false_or_none):
+                f()
+
+    def test_no_raise_without_warning(self, false_or_none):
+        with tm.assert_produces_warning(false_or_none):
+            pass
+
+    def test_no_raise_with_false_raise_on_extra(self, false_or_none):
+        with tm.assert_produces_warning(false_or_none, raise_on_extra_warnings=False):
+            f()
+
+
+def test_raises_during_exception():
+    msg = "Did not see expected warning of class 'UserWarning'"
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(UserWarning):
+            raise ValueError
+
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(UserWarning):
+            warnings.warn("FutureWarning", FutureWarning)
+            raise IndexError
+
+    msg = "Caused unexpected warning"
+    with pytest.raises(AssertionError, match=msg):
+        with tm.assert_produces_warning(None):
+            warnings.warn("FutureWarning", FutureWarning)
+            raise SystemError
+
+
+def test_passes_during_exception():
+    with pytest.raises(SyntaxError, match="Error"):
+        with tm.assert_produces_warning(None):
+            raise SyntaxError("Error")
+
+    with pytest.raises(ValueError, match="Error"):
+        with tm.assert_produces_warning(FutureWarning, match="FutureWarning"):
+            warnings.warn("FutureWarning", FutureWarning)
+            raise ValueError("Error")
diff --git a/pandas/pandas/tests/util/test_assert_series_equal.py b/pandas/pandas/tests/util/test_assert_series_equal.py
new file mode 100644
index 0000000000000..c3cd90f2edfb3
--- /dev/null
+++ b/pandas/pandas/tests/util/test_assert_series_equal.py
@@ -0,0 +1,518 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    Categorical,
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+
+def _assert_series_equal_both(a, b, **kwargs):
+    """
+    Check that two Series equal.
+
+    This check is performed commutatively.
+
+    Parameters
+    ----------
+    a : Series
+        The first Series to compare.
+    b : Series
+        The second Series to compare.
+    kwargs : dict
+        The arguments passed to `tm.assert_series_equal`.
+    """
+    tm.assert_series_equal(a, b, **kwargs)
+    tm.assert_series_equal(b, a, **kwargs)
+
+
+def _assert_not_series_equal(a, b, **kwargs):
+    """
+    Check that two Series are not equal.
+
+    Parameters
+    ----------
+    a : Series
+        The first Series to compare.
+    b : Series
+        The second Series to compare.
+    kwargs : dict
+        The arguments passed to `tm.assert_series_equal`.
+    """
+    try:
+        tm.assert_series_equal(a, b, **kwargs)
+        msg = "The two Series were equal when they shouldn't have been"
+
+        pytest.fail(msg=msg)
+    except AssertionError:
+        pass
+
+
+def _assert_not_series_equal_both(a, b, **kwargs):
+    """
+    Check that two Series are not equal.
+
+    This check is performed commutatively.
+
+    Parameters
+    ----------
+    a : Series
+        The first Series to compare.
+    b : Series
+        The second Series to compare.
+    kwargs : dict
+        The arguments passed to `tm.assert_series_equal`.
+    """
+    _assert_not_series_equal(a, b, **kwargs)
+    _assert_not_series_equal(b, a, **kwargs)
+
+
+@pytest.mark.parametrize("data", [range(3), list("abc"), list("áàä")])
+def test_series_equal(data):
+    _assert_series_equal_both(Series(data), Series(data))
+
+
+@pytest.mark.parametrize(
+    "data1,data2",
+    [
+        (range(3), range(1, 4)),
+        (list("abc"), list("xyz")),
+        (list("áàä"), list("éèë")),
+        (list("áàä"), list(b"aaa")),
+        (range(3), range(4)),
+    ],
+)
+def test_series_not_equal_value_mismatch(data1, data2):
+    _assert_not_series_equal_both(Series(data1), Series(data2))
+
+
+@pytest.mark.parametrize(
+    "kwargs",
+    [
+        {"dtype": "float64"},  # dtype mismatch
+        {"index": [1, 2, 4]},  # index mismatch
+        {"name": "foo"},  # name mismatch
+    ],
+)
+def test_series_not_equal_metadata_mismatch(kwargs):
+    data = range(3)
+    s1 = Series(data)
+
+    s2 = Series(data, **kwargs)
+    _assert_not_series_equal_both(s1, s2)
+
+
+@pytest.mark.parametrize("data1,data2", [(0.12345, 0.12346), (0.1235, 0.1236)])
+@pytest.mark.parametrize("decimals", [0, 1, 2, 3, 5, 10])
+def test_less_precise(data1, data2, any_float_dtype, decimals):
+    rtol = 10**-decimals
+    s1 = Series([data1], dtype=any_float_dtype)
+    s2 = Series([data2], dtype=any_float_dtype)
+
+    if decimals in (5, 10) or (decimals >= 3 and abs(data1 - data2) >= 0.0005):
+        msg = "Series values are different"
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_series_equal(s1, s2, rtol=rtol)
+    else:
+        _assert_series_equal_both(s1, s2, rtol=rtol)
+
+
+@pytest.mark.parametrize(
+    "s1,s2,msg",
+    [
+        # Index
+        (
+            Series(["l1", "l2"], index=[1, 2]),
+            Series(["l1", "l2"], index=[1.0, 2.0]),
+            "Series\\.index are different",
+        ),
+        # MultiIndex
+        (
+            DataFrame.from_records(
+                {"a": [1, 2], "b": [2.1, 1.5], "c": ["l1", "l2"]}, index=["a", "b"]
+            ).c,
+            DataFrame.from_records(
+                {"a": [1.0, 2.0], "b": [2.1, 1.5], "c": ["l1", "l2"]}, index=["a", "b"]
+            ).c,
+            "Series\\.index level \\[0\\] are different",
+        ),
+    ],
+)
+def test_series_equal_index_dtype(s1, s2, msg, check_index_type):
+    kwargs = {"check_index_type": check_index_type}
+
+    if check_index_type:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_series_equal(s1, s2, **kwargs)
+    else:
+        tm.assert_series_equal(s1, s2, **kwargs)
+
+
+@pytest.mark.parametrize("check_like", [True, False])
+def test_series_equal_order_mismatch(check_like):
+    s1 = Series([1, 2, 3], index=["a", "b", "c"])
+    s2 = Series([3, 2, 1], index=["c", "b", "a"])
+
+    if not check_like:  # Do not ignore index ordering.
+        with pytest.raises(AssertionError, match="Series.index are different"):
+            tm.assert_series_equal(s1, s2, check_like=check_like)
+    else:
+        _assert_series_equal_both(s1, s2, check_like=check_like)
+
+
+@pytest.mark.parametrize("check_index", [True, False])
+def test_series_equal_index_mismatch(check_index):
+    s1 = Series([1, 2, 3], index=["a", "b", "c"])
+    s2 = Series([1, 2, 3], index=["c", "b", "a"])
+
+    if check_index:  # Do not ignore index.
+        with pytest.raises(AssertionError, match="Series.index are different"):
+            tm.assert_series_equal(s1, s2, check_index=check_index)
+    else:
+        _assert_series_equal_both(s1, s2, check_index=check_index)
+
+
+def test_series_invalid_param_combination():
+    left = Series(dtype=object)
+    right = Series(dtype=object)
+    with pytest.raises(
+        ValueError, match="check_like must be False if check_index is False"
+    ):
+        tm.assert_series_equal(left, right, check_index=False, check_like=True)
+
+
+def test_series_equal_length_mismatch(rtol):
+    msg = """Series are different
+
+Series length are different
+\\[left\\]:  3, RangeIndex\\(start=0, stop=3, step=1\\)
+\\[right\\]: 4, RangeIndex\\(start=0, stop=4, step=1\\)"""
+
+    s1 = Series([1, 2, 3])
+    s2 = Series([1, 2, 3, 4])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s1, s2, rtol=rtol)
+
+
+def test_series_equal_numeric_values_mismatch(rtol):
+    msg = """Series are different
+
+Series values are different \\(33\\.33333 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\[1, 2, 3\\]
+\\[right\\]: \\[1, 2, 4\\]"""
+
+    s1 = Series([1, 2, 3])
+    s2 = Series([1, 2, 4])
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s1, s2, rtol=rtol)
+
+
+def test_series_equal_categorical_values_mismatch(rtol, using_infer_string):
+    if using_infer_string:
+        msg = """Series are different
+
+Series values are different \\(66\\.66667 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\['a', 'b', 'c'\\]
+Categories \\(3, str\\): \\[a, b, c\\]
+\\[right\\]: \\['a', 'c', 'b'\\]
+Categories \\(3, str\\): \\[a, b, c\\]"""
+    else:
+        msg = """Series are different
+
+Series values are different \\(66\\.66667 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\['a', 'b', 'c'\\]
+Categories \\(3, object\\): \\['a', 'b', 'c'\\]
+\\[right\\]: \\['a', 'c', 'b'\\]
+Categories \\(3, object\\): \\['a', 'b', 'c'\\]"""
+
+    s1 = Series(Categorical(["a", "b", "c"]))
+    s2 = Series(Categorical(["a", "c", "b"]))
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s1, s2, rtol=rtol)
+
+
+def test_series_equal_datetime_values_mismatch(rtol):
+    msg = """Series are different
+
+Series values are different \\(100.0 %\\)
+\\[index\\]: \\[0, 1, 2\\]
+\\[left\\]:  \\[1514764800000000000, 1514851200000000000, 1514937600000000000\\]
+\\[right\\]: \\[1549065600000000000, 1549152000000000000, 1549238400000000000\\]"""
+
+    s1 = Series(pd.date_range("2018-01-01", periods=3, freq="D"))
+    s2 = Series(pd.date_range("2019-02-02", periods=3, freq="D"))
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s1, s2, rtol=rtol)
+
+
+def test_series_equal_categorical_mismatch(check_categorical, using_infer_string):
+    if using_infer_string:
+        dtype = "str"
+    else:
+        dtype = "object"
+    msg = f"""Attributes of Series are different
+
+Attribute "dtype" are different
+\\[left\\]:  CategoricalDtype\\(categories=\\['a', 'b'\\], ordered=False, \
+categories_dtype={dtype}\\)
+\\[right\\]: CategoricalDtype\\(categories=\\['a', 'b', 'c'\\], \
+ordered=False, categories_dtype={dtype}\\)"""
+
+    s1 = Series(Categorical(["a", "b"]))
+    s2 = Series(Categorical(["a", "b"], categories=list("abc")))
+
+    if check_categorical:
+        with pytest.raises(AssertionError, match=msg):
+            tm.assert_series_equal(s1, s2, check_categorical=check_categorical)
+    else:
+        _assert_series_equal_both(s1, s2, check_categorical=check_categorical)
+
+
+def test_assert_series_equal_extension_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/32747
+    left = Series(pd.array([1, 2, 3], dtype="Int64"))
+    right = left.astype(int)
+
+    msg = """Attributes of Series are different
+
+Attribute "dtype" are different
+\\[left\\]:  Int64
+\\[right\\]: int[32|64]"""
+
+    tm.assert_series_equal(left, right, check_dtype=False)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(left, right, check_dtype=True)
+
+
+def test_assert_series_equal_interval_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/32747
+    left = Series([pd.Interval(0, 1)], dtype="interval")
+    right = left.astype(object)
+
+    msg = """Attributes of Series are different
+
+Attribute "dtype" are different
+\\[left\\]:  interval\\[int64, right\\]
+\\[right\\]: object"""
+
+    tm.assert_series_equal(left, right, check_dtype=False)
+
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(left, right, check_dtype=True)
+
+
+def test_series_equal_series_type():
+    class MySeries(Series):
+        pass
+
+    s1 = Series([1, 2])
+    s2 = Series([1, 2])
+    s3 = MySeries([1, 2])
+
+    tm.assert_series_equal(s1, s2, check_series_type=False)
+    tm.assert_series_equal(s1, s2, check_series_type=True)
+
+    tm.assert_series_equal(s1, s3, check_series_type=False)
+    tm.assert_series_equal(s3, s1, check_series_type=False)
+
+    with pytest.raises(AssertionError, match="Series classes are different"):
+        tm.assert_series_equal(s1, s3, check_series_type=True)
+
+    with pytest.raises(AssertionError, match="Series classes are different"):
+        tm.assert_series_equal(s3, s1, check_series_type=True)
+
+
+def test_series_equal_exact_for_nonnumeric():
+    # https://github.com/pandas-dev/pandas/issues/35446
+    s1 = Series(["a", "b"])
+    s2 = Series(["a", "b"])
+    s3 = Series(["b", "a"])
+
+    tm.assert_series_equal(s1, s2, check_exact=True)
+    tm.assert_series_equal(s2, s1, check_exact=True)
+
+    msg = """Series are different
+
+Series values are different \\(100\\.0 %\\)
+\\[index\\]: \\[0, 1\\]
+\\[left\\]:  \\[a, b\\]
+\\[right\\]: \\[b, a\\]"""
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s1, s3, check_exact=True)
+
+    msg = """Series are different
+
+Series values are different \\(100\\.0 %\\)
+\\[index\\]: \\[0, 1\\]
+\\[left\\]:  \\[b, a\\]
+\\[right\\]: \\[a, b\\]"""
+    with pytest.raises(AssertionError, match=msg):
+        tm.assert_series_equal(s3, s1, check_exact=True)
+
+
+def test_assert_series_equal_ignore_extension_dtype_mismatch():
+    # https://github.com/pandas-dev/pandas/issues/35715
+    left = Series([1, 2, 3], dtype="Int64")
+    right = Series([1, 2, 3], dtype="Int32")
+    tm.assert_series_equal(left, right, check_dtype=False)
+
+
+def test_assert_series_equal_ignore_extension_dtype_mismatch_cross_class():
+    # https://github.com/pandas-dev/pandas/issues/35715
+    left = Series([1, 2, 3], dtype="Int64")
+    right = Series([1, 2, 3], dtype="int64")
+    tm.assert_series_equal(left, right, check_dtype=False)
+
+
+def test_allows_duplicate_labels():
+    left = Series([1])
+    right = Series([1]).set_flags(allows_duplicate_labels=False)
+    tm.assert_series_equal(left, left)
+    tm.assert_series_equal(right, right)
+    tm.assert_series_equal(left, right, check_flags=False)
+    tm.assert_series_equal(right, left, check_flags=False)
+
+    with pytest.raises(AssertionError, match="<Flags"):
+        tm.assert_series_equal(left, right)
+
+    with pytest.raises(AssertionError, match="<Flags"):
+        tm.assert_series_equal(left, right)
+
+
+def test_assert_series_equal_identical_na(nulls_fixture):
+    ser = Series([nulls_fixture])
+
+    tm.assert_series_equal(ser, ser.copy())
+
+    # while we're here do Index too
+    idx = pd.Index(ser)
+    tm.assert_index_equal(idx, idx.copy(deep=True))
+
+
+def test_identical_nested_series_is_equal():
+    # GH#22400
+    x = Series(
+        [
+            0,
+            0.0131142231938,
+            1.77774652865e-05,
+            np.array([0.4722720840328748, 0.4216929783681722]),
+        ]
+    )
+    y = Series(
+        [
+            0,
+            0.0131142231938,
+            1.77774652865e-05,
+            np.array([0.4722720840328748, 0.4216929783681722]),
+        ]
+    )
+    # These two arrays should be equal, nesting could cause issue
+
+    tm.assert_series_equal(x, x)
+    tm.assert_series_equal(x, x, check_exact=True)
+    tm.assert_series_equal(x, y)
+    tm.assert_series_equal(x, y, check_exact=True)
+
+
+@pytest.mark.parametrize("dtype", ["datetime64", "timedelta64"])
+def test_check_dtype_false_different_reso(dtype):
+    # GH 52449
+    ser_s = Series([1000213, 2131232, 21312331]).astype(f"{dtype}[s]")
+    ser_ms = ser_s.astype(f"{dtype}[ms]")
+    with pytest.raises(AssertionError, match="Attributes of Series are different"):
+        tm.assert_series_equal(ser_s, ser_ms)
+    tm.assert_series_equal(ser_ms, ser_s, check_dtype=False)
+
+    ser_ms -= Series([1, 1, 1]).astype(f"{dtype}[ms]")
+
+    with pytest.raises(AssertionError, match="Series are different"):
+        tm.assert_series_equal(ser_s, ser_ms)
+
+    with pytest.raises(AssertionError, match="Series are different"):
+        tm.assert_series_equal(ser_s, ser_ms, check_dtype=False)
+
+
+@pytest.mark.parametrize("dtype", ["Int64", "int64"])
+def test_large_unequal_ints(dtype):
+    # https://github.com/pandas-dev/pandas/issues/55882
+    left = Series([1577840521123000], dtype=dtype)
+    right = Series([1577840521123543], dtype=dtype)
+    with pytest.raises(AssertionError, match="Series are different"):
+        tm.assert_series_equal(left, right)
+
+
+@pytest.mark.parametrize("dtype", [None, object])
+@pytest.mark.parametrize("check_exact", [True, False])
+@pytest.mark.parametrize("val", [3, 3.5])
+def test_ea_and_numpy_no_dtype_check(val, check_exact, dtype):
+    # GH#56651
+    left = Series([1, 2, val], dtype=dtype)
+    right = Series(pd.array([1, 2, val]))
+    tm.assert_series_equal(left, right, check_dtype=False, check_exact=check_exact)
+
+
+def test_assert_series_equal_int_tol():
+    # GH#56646
+    left = Series([81, 18, 121, 38, 74, 72, 81, 81, 146, 81, 81, 170, 74, 74])
+    right = Series([72, 9, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72])
+    tm.assert_series_equal(left, right, rtol=1.5)
+
+    tm.assert_frame_equal(left.to_frame(), right.to_frame(), rtol=1.5)
+    tm.assert_extension_array_equal(
+        left.astype("Int64").values, right.astype("Int64").values, rtol=1.5
+    )
+
+
+@pytest.mark.parametrize(
+    "left_idx, right_idx",
+    [
+        (
+            pd.Index([0, 0.2, 0.4, 0.6, 0.8, 1]),
+            pd.Index(np.linspace(0, 1, 6)),
+        ),
+        (
+            pd.MultiIndex.from_arrays([[0, 0, 0, 0, 1, 1], [0, 0.2, 0.4, 0.6, 0.8, 1]]),
+            pd.MultiIndex.from_arrays([[0, 0, 0, 0, 1, 1], np.linspace(0, 1, 6)]),
+        ),
+        (
+            pd.MultiIndex.from_arrays(
+                [["a", "a", "a", "b", "b", "b"], [1, 2, 3, 4, 5, 10000000000001]]
+            ),
+            pd.MultiIndex.from_arrays(
+                [["a", "a", "a", "b", "b", "b"], [1, 2, 3, 4, 5, 10000000000002]]
+            ),
+        ),
+        pytest.param(
+            pd.Index([1, 2, 3, 4, 5, 10000000000001]),
+            pd.Index([1, 2, 3, 4, 5, 10000000000002]),
+            marks=pytest.mark.xfail(reason="check_exact_index defaults to True"),
+        ),
+        pytest.param(
+            pd.MultiIndex.from_arrays(
+                [[0, 0, 0, 0, 1, 1], [1, 2, 3, 4, 5, 10000000000001]]
+            ),
+            pd.MultiIndex.from_arrays(
+                [[0, 0, 0, 0, 1, 1], [1, 2, 3, 4, 5, 10000000000002]]
+            ),
+            marks=pytest.mark.xfail(reason="check_exact_index defaults to True"),
+        ),
+    ],
+)
+def test_assert_series_equal_check_exact_index_default(left_idx, right_idx):
+    # GH#57067
+    ser1 = Series(np.zeros(6, dtype=int), left_idx)
+    ser2 = Series(np.zeros(6, dtype=int), right_idx)
+    tm.assert_series_equal(ser1, ser2)
+    tm.assert_frame_equal(ser1.to_frame(), ser2.to_frame())
diff --git a/pandas/pandas/tests/util/test_deprecate.py b/pandas/pandas/tests/util/test_deprecate.py
new file mode 100644
index 0000000000000..92f422b8269f5
--- /dev/null
+++ b/pandas/pandas/tests/util/test_deprecate.py
@@ -0,0 +1,63 @@
+from textwrap import dedent
+
+import pytest
+
+from pandas.util._decorators import deprecate
+
+import pandas._testing as tm
+
+
+def new_func():
+    """
+    This is the summary. The deprecate directive goes next.
+
+    This is the extended summary. The deprecate directive goes before this.
+    """
+    return "new_func called"
+
+
+def new_func_no_docstring():
+    return "new_func_no_docstring called"
+
+
+def new_func_wrong_docstring():
+    """Summary should be in the next line."""
+    return "new_func_wrong_docstring called"
+
+
+def new_func_with_deprecation():
+    """
+    This is the summary. The deprecate directive goes next.
+
+    .. deprecated:: 1.0
+        Use new_func instead.
+
+    This is the extended summary. The deprecate directive goes before this.
+    """
+
+
+def test_deprecate_ok():
+    depr_func = deprecate("depr_func", new_func, "1.0", msg="Use new_func instead.")
+
+    with tm.assert_produces_warning(FutureWarning):
+        result = depr_func()
+
+    assert result == "new_func called"
+    assert depr_func.__doc__ == dedent(new_func_with_deprecation.__doc__)
+
+
+def test_deprecate_no_docstring():
+    depr_func = deprecate(
+        "depr_func", new_func_no_docstring, "1.0", msg="Use new_func instead."
+    )
+    with tm.assert_produces_warning(FutureWarning):
+        result = depr_func()
+    assert result == "new_func_no_docstring called"
+
+
+def test_deprecate_wrong_docstring():
+    msg = "deprecate needs a correctly formatted docstring"
+    with pytest.raises(AssertionError, match=msg):
+        deprecate(
+            "depr_func", new_func_wrong_docstring, "1.0", msg="Use new_func instead."
+        )
diff --git a/pandas/pandas/tests/util/test_deprecate_kwarg.py b/pandas/pandas/tests/util/test_deprecate_kwarg.py
new file mode 100644
index 0000000000000..b165e9fba0e4f
--- /dev/null
+++ b/pandas/pandas/tests/util/test_deprecate_kwarg.py
@@ -0,0 +1,90 @@
+import pytest
+
+from pandas.util._decorators import deprecate_kwarg
+
+import pandas._testing as tm
+
+
+@deprecate_kwarg("old", "new")
+def _f1(new=False):
+    return new
+
+
+_f2_mappings = {"yes": True, "no": False}
+
+
+@deprecate_kwarg("old", "new", _f2_mappings)
+def _f2(new=False):
+    return new
+
+
+def _f3_mapping(x):
+    return x + 1
+
+
+@deprecate_kwarg("old", "new", _f3_mapping)
+def _f3(new=0):
+    return new
+
+
+@pytest.mark.parametrize("key,klass", [("old", FutureWarning), ("new", None)])
+def test_deprecate_kwarg(key, klass):
+    x = 78
+
+    with tm.assert_produces_warning(klass):
+        assert _f1(**{key: x}) == x
+
+
+@pytest.mark.parametrize("key", list(_f2_mappings.keys()))
+def test_dict_deprecate_kwarg(key):
+    with tm.assert_produces_warning(FutureWarning):
+        assert _f2(old=key) == _f2_mappings[key]
+
+
+@pytest.mark.parametrize("key", ["bogus", 12345, -1.23])
+def test_missing_deprecate_kwarg(key):
+    with tm.assert_produces_warning(FutureWarning):
+        assert _f2(old=key) == key
+
+
+@pytest.mark.parametrize("x", [1, -1.4, 0])
+def test_callable_deprecate_kwarg(x):
+    with tm.assert_produces_warning(FutureWarning):
+        assert _f3(old=x) == _f3_mapping(x)
+
+
+def test_callable_deprecate_kwarg_fail():
+    msg = "((can only|cannot) concatenate)|(must be str)|(Can't convert)"
+
+    with pytest.raises(TypeError, match=msg):
+        _f3(old="hello")
+
+
+def test_bad_deprecate_kwarg():
+    msg = "mapping from old to new argument values must be dict or callable!"
+
+    with pytest.raises(TypeError, match=msg):
+
+        @deprecate_kwarg("old", "new", 0)
+        def f4(new=None):
+            return new
+
+
+@deprecate_kwarg("old", None)
+def _f4(old=True, unchanged=True):
+    return old, unchanged
+
+
+@pytest.mark.parametrize("key", ["old", "unchanged"])
+def test_deprecate_keyword(key):
+    x = 9
+
+    if key == "old":
+        klass = FutureWarning
+        expected = (x, True)
+    else:
+        klass = None
+        expected = (True, x)
+
+    with tm.assert_produces_warning(klass):
+        assert _f4(**{key: x}) == expected
diff --git a/pandas/pandas/tests/util/test_deprecate_nonkeyword_arguments.py b/pandas/pandas/tests/util/test_deprecate_nonkeyword_arguments.py
new file mode 100644
index 0000000000000..f81d32b574682
--- /dev/null
+++ b/pandas/pandas/tests/util/test_deprecate_nonkeyword_arguments.py
@@ -0,0 +1,140 @@
+"""
+Tests for the `deprecate_nonkeyword_arguments` decorator
+"""
+
+import inspect
+
+from pandas.util._decorators import deprecate_nonkeyword_arguments
+
+import pandas._testing as tm
+
+
+@deprecate_nonkeyword_arguments(
+    version="1.1", allowed_args=["a", "b"], name="f_add_inputs"
+)
+def f(a, b=0, c=0, d=0):
+    return a + b + c + d
+
+
+def test_f_signature():
+    assert str(inspect.signature(f)) == "(a, b=0, *, c=0, d=0)"
+
+
+def test_one_argument():
+    with tm.assert_produces_warning(None):
+        assert f(19) == 19
+
+
+def test_one_and_one_arguments():
+    with tm.assert_produces_warning(None):
+        assert f(19, d=6) == 25
+
+
+def test_two_arguments():
+    with tm.assert_produces_warning(None):
+        assert f(1, 5) == 6
+
+
+def test_two_and_two_arguments():
+    with tm.assert_produces_warning(None):
+        assert f(1, 3, c=3, d=5) == 12
+
+
+def test_three_arguments():
+    with tm.assert_produces_warning(FutureWarning):
+        assert f(6, 3, 3) == 12
+
+
+def test_four_arguments():
+    with tm.assert_produces_warning(FutureWarning):
+        assert f(1, 2, 3, 4) == 10
+
+
+def test_three_arguments_with_name_in_warning():
+    msg = (
+        "Starting with pandas version 1.1 all arguments of f_add_inputs "
+        "except for the arguments 'a' and 'b' will be keyword-only."
+    )
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        assert f(6, 3, 3) == 12
+
+
+@deprecate_nonkeyword_arguments(version="1.1")
+def g(a, b=0, c=0, d=0):
+    with tm.assert_produces_warning(None):
+        return a + b + c + d
+
+
+def test_g_signature():
+    assert str(inspect.signature(g)) == "(a, *, b=0, c=0, d=0)"
+
+
+def test_one_and_three_arguments_default_allowed_args():
+    with tm.assert_produces_warning(None):
+        assert g(1, b=3, c=3, d=5) == 12
+
+
+def test_three_arguments_default_allowed_args():
+    with tm.assert_produces_warning(FutureWarning):
+        assert g(6, 3, 3) == 12
+
+
+def test_three_positional_argument_with_warning_message_analysis():
+    msg = (
+        "Starting with pandas version 1.1 all arguments of g "
+        "except for the argument 'a' will be keyword-only."
+    )
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        assert g(6, 3, 3) == 12
+
+
+@deprecate_nonkeyword_arguments(version="1.1")
+def h(a=0, b=0, c=0, d=0):
+    return a + b + c + d
+
+
+def test_h_signature():
+    assert str(inspect.signature(h)) == "(*, a=0, b=0, c=0, d=0)"
+
+
+def test_all_keyword_arguments():
+    with tm.assert_produces_warning(None):
+        assert h(a=1, b=2) == 3
+
+
+def test_one_positional_argument():
+    with tm.assert_produces_warning(FutureWarning):
+        assert h(23) == 23
+
+
+def test_one_positional_argument_with_warning_message_analysis():
+    msg = "Starting with pandas version 1.1 all arguments of h will be keyword-only."
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        assert h(19) == 19
+
+
+@deprecate_nonkeyword_arguments(version="1.1")
+def i(a=0, /, b=0, *, c=0, d=0):
+    return a + b + c + d
+
+
+def test_i_signature():
+    assert str(inspect.signature(i)) == "(*, a=0, b=0, c=0, d=0)"
+
+
+class Foo:
+    @deprecate_nonkeyword_arguments(version=None, allowed_args=["self", "bar"])
+    def baz(self, bar=None, foobar=None): ...
+
+
+def test_foo_signature():
+    assert str(inspect.signature(Foo.baz)) == "(self, bar=None, *, foobar=None)"
+
+
+def test_class():
+    msg = (
+        r"In a future version of pandas all arguments of Foo\.baz "
+        r"except for the argument \'bar\' will be keyword-only"
+    )
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        Foo().baz("qux", "quox")
diff --git a/pandas/pandas/tests/util/test_doc.py b/pandas/pandas/tests/util/test_doc.py
new file mode 100644
index 0000000000000..50859564e654f
--- /dev/null
+++ b/pandas/pandas/tests/util/test_doc.py
@@ -0,0 +1,90 @@
+from textwrap import dedent
+
+from pandas.util._decorators import doc
+
+
+@doc(method="cumsum", operation="sum")
+def cumsum(whatever):
+    """
+    This is the {method} method.
+
+    It computes the cumulative {operation}.
+    """
+
+
+@doc(
+    cumsum,
+    dedent(
+        """
+        Examples
+        --------
+
+        >>> cumavg([1, 2, 3])
+        2
+        """
+    ),
+    method="cumavg",
+    operation="average",
+)
+def cumavg(whatever):
+    pass
+
+
+@doc(cumsum, method="cummax", operation="maximum")
+def cummax(whatever):
+    pass
+
+
+@doc(cummax, method="cummin", operation="minimum")
+def cummin(whatever):
+    pass
+
+
+def test_docstring_formatting():
+    docstr = dedent(
+        """
+        This is the cumsum method.
+
+        It computes the cumulative sum.
+        """
+    )
+    assert cumsum.__doc__ == docstr
+
+
+def test_docstring_appending():
+    docstr = dedent(
+        """
+        This is the cumavg method.
+
+        It computes the cumulative average.
+
+        Examples
+        --------
+
+        >>> cumavg([1, 2, 3])
+        2
+        """
+    )
+    assert cumavg.__doc__ == docstr
+
+
+def test_doc_template_from_func():
+    docstr = dedent(
+        """
+        This is the cummax method.
+
+        It computes the cumulative maximum.
+        """
+    )
+    assert cummax.__doc__ == docstr
+
+
+def test_inherit_doc_template():
+    docstr = dedent(
+        """
+        This is the cummin method.
+
+        It computes the cumulative minimum.
+        """
+    )
+    assert cummin.__doc__ == docstr
diff --git a/pandas/pandas/tests/util/test_hashing.py b/pandas/pandas/tests/util/test_hashing.py
new file mode 100644
index 0000000000000..e654534ccd453
--- /dev/null
+++ b/pandas/pandas/tests/util/test_hashing.py
@@ -0,0 +1,418 @@
+import numpy as np
+import pytest
+
+import pandas as pd
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    period_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+from pandas.core.util.hashing import hash_tuples
+from pandas.util import (
+    hash_array,
+    hash_pandas_object,
+)
+
+
+@pytest.fixture(
+    params=[
+        Series([1, 2, 3] * 3, dtype="int32"),
+        Series([None, 2.5, 3.5] * 3, dtype="float32"),
+        Series(["a", "b", "c"] * 3, dtype="category"),
+        Series(["d", "e", "f"] * 3),
+        Series([True, False, True] * 3),
+        Series(pd.date_range("20130101", periods=9)),
+        Series(pd.date_range("20130101", periods=9, tz="US/Eastern")),
+        Series(timedelta_range("2000", periods=9)),
+    ]
+)
+def series(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def index(request):
+    return request.param
+
+
+def test_consistency():
+    # Check that our hash doesn't change because of a mistake
+    # in the actual code; this is the ground truth.
+    result = hash_pandas_object(Index(["foo", "bar", "baz"]))
+    expected = Series(
+        np.array(
+            [3600424527151052760, 1374399572096150070, 477881037637427054],
+            dtype="uint64",
+        ),
+        index=["foo", "bar", "baz"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_hash_array(series):
+    arr = series.values
+    tm.assert_numpy_array_equal(hash_array(arr), hash_array(arr))
+
+
+@pytest.mark.parametrize("dtype", ["U", object])
+def test_hash_array_mixed(dtype):
+    result1 = hash_array(np.array(["3", "4", "All"]))
+    result2 = hash_array(np.array([3, 4, "All"], dtype=dtype))
+
+    tm.assert_numpy_array_equal(result1, result2)
+
+
+@pytest.mark.parametrize("val", [5, "foo", pd.Timestamp("20130101")])
+def test_hash_array_errors(val):
+    msg = "must pass a ndarray-like"
+    with pytest.raises(TypeError, match=msg):
+        hash_array(val)
+
+
+def test_hash_array_index_exception():
+    # GH42003 TypeError instead of AttributeError
+    obj = pd.DatetimeIndex(["2018-10-28 01:20:00"], tz="Europe/Berlin")
+
+    msg = "Use hash_pandas_object instead"
+    with pytest.raises(TypeError, match=msg):
+        hash_array(obj)
+
+
+def test_hash_tuples():
+    tuples = [(1, "one"), (1, "two"), (2, "one")]
+    result = hash_tuples(tuples)
+
+    expected = hash_pandas_object(MultiIndex.from_tuples(tuples)).values
+    tm.assert_numpy_array_equal(result, expected)
+
+    # We only need to support MultiIndex and list-of-tuples
+    msg = "|".join(["object is not iterable", "zip argument #1 must support iteration"])
+    with pytest.raises(TypeError, match=msg):
+        hash_tuples(tuples[0])
+
+
+@pytest.mark.parametrize("val", [5, "foo", pd.Timestamp("20130101")])
+def test_hash_tuples_err(val):
+    msg = "must be convertible to a list-of-tuples"
+    with pytest.raises(TypeError, match=msg):
+        hash_tuples(val)
+
+
+def test_multiindex_unique():
+    mi = MultiIndex.from_tuples([(118, 472), (236, 118), (51, 204), (102, 51)])
+    assert mi.is_unique is True
+
+    result = hash_pandas_object(mi)
+    assert result.is_unique is True
+
+
+def test_multiindex_objects():
+    mi = MultiIndex(
+        levels=[["b", "d", "a"], [1, 2, 3]],
+        codes=[[0, 1, 0, 2], [2, 0, 0, 1]],
+        names=["col1", "col2"],
+    )
+    recons = mi._sort_levels_monotonic()
+
+    # These are equal.
+    assert mi.equals(recons)
+    assert Index(mi.values).equals(Index(recons.values))
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        Series([1, 2, 3]),
+        Series([1.0, 1.5, 3.2]),
+        Series([1.0, 1.5, np.nan]),
+        Series([1.0, 1.5, 3.2], index=[1.5, 1.1, 3.3]),
+        Series(["a", "b", "c"]),
+        Series(["a", np.nan, "c"]),
+        Series(["a", None, "c"]),
+        Series([True, False, True]),
+        Series(dtype=object),
+        DataFrame({"x": ["a", "b", "c"], "y": [1, 2, 3]}),
+        DataFrame(),
+        DataFrame(np.full((10, 4), np.nan)),
+        DataFrame(
+            {
+                "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+                "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+                "C": Index(["foo1", "foo2", "foo3", "foo4", "foo5"], dtype=object),
+                "D": pd.date_range("20130101", periods=5),
+            }
+        ),
+        DataFrame(range(5), index=pd.date_range("2020-01-01", periods=5)),
+        Series(range(5), index=pd.date_range("2020-01-01", periods=5)),
+        Series(period_range("2020-01-01", periods=10, freq="D")),
+        Series(pd.date_range("20130101", periods=3, tz="US/Eastern")),
+    ],
+)
+def test_hash_pandas_object(obj, index):
+    a = hash_pandas_object(obj, index=index)
+    b = hash_pandas_object(obj, index=index)
+    tm.assert_series_equal(a, b)
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        Series([1, 2, 3]),
+        Series([1.0, 1.5, 3.2]),
+        Series([1.0, 1.5, np.nan]),
+        Series([1.0, 1.5, 3.2], index=[1.5, 1.1, 3.3]),
+        Series(["a", "b", "c"]),
+        Series(["a", np.nan, "c"]),
+        Series(["a", None, "c"]),
+        Series([True, False, True]),
+        DataFrame({"x": ["a", "b", "c"], "y": [1, 2, 3]}),
+        DataFrame(np.full((10, 4), np.nan)),
+        DataFrame(
+            {
+                "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+                "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+                "C": Index(["foo1", "foo2", "foo3", "foo4", "foo5"], dtype=object),
+                "D": pd.date_range("20130101", periods=5),
+            }
+        ),
+        DataFrame(range(5), index=pd.date_range("2020-01-01", periods=5)),
+        Series(range(5), index=pd.date_range("2020-01-01", periods=5)),
+        Series(period_range("2020-01-01", periods=10, freq="D")),
+        Series(pd.date_range("20130101", periods=3, tz="US/Eastern")),
+    ],
+)
+def test_hash_pandas_object_diff_index_non_empty(obj):
+    a = hash_pandas_object(obj, index=True)
+    b = hash_pandas_object(obj, index=False)
+    assert not (a == b).all()
+
+
+@pytest.mark.parametrize(
+    "obj",
+    [
+        Index([1, 2, 3]),
+        Index([True, False, True]),
+        timedelta_range("1 day", periods=2),
+        period_range("2020-01-01", freq="D", periods=2),
+        MultiIndex.from_product(
+            [range(5), ["foo", "bar", "baz"], pd.date_range("20130101", periods=2)]
+        ),
+        MultiIndex.from_product([pd.CategoricalIndex(list("aabc")), range(3)]),
+    ],
+)
+def test_hash_pandas_index(obj, index):
+    a = hash_pandas_object(obj, index=index)
+    b = hash_pandas_object(obj, index=index)
+    tm.assert_series_equal(a, b)
+
+
+def test_hash_pandas_series(series, index):
+    a = hash_pandas_object(series, index=index)
+    b = hash_pandas_object(series, index=index)
+    tm.assert_series_equal(a, b)
+
+
+def test_hash_pandas_series_diff_index(series):
+    a = hash_pandas_object(series, index=True)
+    b = hash_pandas_object(series, index=False)
+    assert not (a == b).all()
+
+
+@pytest.mark.parametrize("klass", [Index, Series])
+@pytest.mark.parametrize("dtype", ["float64", "object"])
+def test_hash_pandas_empty_object(klass, dtype, index):
+    # These are by-definition the same with
+    # or without the index as the data is empty.
+    obj = klass([], dtype=dtype)
+    a = hash_pandas_object(obj, index=index)
+    b = hash_pandas_object(obj, index=index)
+    tm.assert_series_equal(a, b)
+
+
+@pytest.mark.parametrize(
+    "s1",
+    [
+        ["a", "b", "c", "d"],
+        [1000, 2000, 3000, 4000],
+        pd.date_range(0, periods=4),
+    ],
+)
+@pytest.mark.parametrize("categorize", [True, False])
+def test_categorical_consistency(s1, categorize):
+    # see gh-15143
+    #
+    # Check that categoricals hash consistent with their values,
+    # not codes. This should work for categoricals of any dtype.
+    s1 = Series(s1)
+    s2 = s1.astype("category").cat.set_categories(s1)
+    s3 = s2.cat.set_categories(list(reversed(s1)))
+
+    # These should all hash identically.
+    h1 = hash_pandas_object(s1, categorize=categorize)
+    h2 = hash_pandas_object(s2, categorize=categorize)
+    h3 = hash_pandas_object(s3, categorize=categorize)
+
+    tm.assert_series_equal(h1, h2)
+    tm.assert_series_equal(h1, h3)
+
+
+def test_categorical_with_nan_consistency(unit):
+    dti = pd.date_range("2012-01-01", periods=5, name="B", unit=unit)
+    cat = pd.Categorical.from_codes([-1, 0, 1, 2, 3, 4], categories=dti)
+    expected = hash_array(cat, categorize=False)
+
+    ts = pd.Timestamp("2012-01-01").as_unit(unit)
+    cat2 = pd.Categorical.from_codes([-1, 0], categories=[ts])
+    result = hash_array(cat2, categorize=False)
+
+    assert result[0] in expected
+    assert result[1] in expected
+
+
+def test_pandas_errors():
+    msg = "Unexpected type for hashing"
+    with pytest.raises(TypeError, match=msg):
+        hash_pandas_object(pd.Timestamp("20130101"))
+
+
+def test_hash_keys():
+    # Using different hash keys, should have
+    # different hashes for the same data.
+    #
+    # This only matters for object dtypes.
+    obj = Series(list("abc"))
+
+    a = hash_pandas_object(obj, hash_key="9876543210123456")
+    b = hash_pandas_object(obj, hash_key="9876543210123465")
+
+    assert (a != b).all()
+
+
+def test_df_hash_keys():
+    # DataFrame version of the test_hash_keys.
+    # https://github.com/pandas-dev/pandas/issues/41404
+    obj = DataFrame({"x": np.arange(3), "y": list("abc")})
+
+    a = hash_pandas_object(obj, hash_key="9876543210123456")
+    b = hash_pandas_object(obj, hash_key="9876543210123465")
+
+    assert (a != b).all()
+
+
+def test_df_encoding():
+    # Check that DataFrame recognizes optional encoding.
+    # https://github.com/pandas-dev/pandas/issues/41404
+    # https://github.com/pandas-dev/pandas/pull/42049
+    obj = DataFrame({"x": np.arange(3), "y": list("a+c")})
+
+    a = hash_pandas_object(obj, encoding="utf8")
+    b = hash_pandas_object(obj, encoding="utf7")
+
+    # Note that the "+" is encoded as "+-" in utf-7.
+    assert a[0] == b[0]
+    assert a[1] != b[1]
+    assert a[2] == b[2]
+
+
+def test_invalid_key():
+    # This only matters for object dtypes.
+    msg = "key should be a 16-byte string encoded"
+
+    with pytest.raises(ValueError, match=msg):
+        hash_pandas_object(Series(list("abc")), hash_key="foo")
+
+
+def test_already_encoded(index):
+    # If already encoded, then ok.
+    obj = Series(list("abc")).str.encode("utf8")
+    a = hash_pandas_object(obj, index=index)
+    b = hash_pandas_object(obj, index=index)
+    tm.assert_series_equal(a, b)
+
+
+def test_alternate_encoding(index):
+    obj = Series(list("abc"))
+    a = hash_pandas_object(obj, index=index)
+    b = hash_pandas_object(obj, index=index)
+    tm.assert_series_equal(a, b)
+
+
+@pytest.mark.parametrize("l_exp", range(8))
+@pytest.mark.parametrize("l_add", [0, 1])
+def test_same_len_hash_collisions(l_exp, l_add):
+    length = 2 ** (l_exp + 8) + l_add
+    idx = np.array([str(i) for i in range(length)], dtype=object)
+
+    result = hash_array(idx, "utf8")
+    assert not result[0] == result[1]
+
+
+def test_hash_collisions():
+    # Hash collisions are bad.
+    #
+    # https://github.com/pandas-dev/pandas/issues/14711#issuecomment-264885726
+    hashes = [
+        "Ingrid-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",
+        "Tim-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",
+    ]
+
+    # These should be different.
+    result1 = hash_array(np.asarray(hashes[0:1], dtype=object), "utf8")
+    expected1 = np.array([14963968704024874985], dtype=np.uint64)
+    tm.assert_numpy_array_equal(result1, expected1)
+
+    result2 = hash_array(np.asarray(hashes[1:2], dtype=object), "utf8")
+    expected2 = np.array([16428432627716348016], dtype=np.uint64)
+    tm.assert_numpy_array_equal(result2, expected2)
+
+    result = hash_array(np.asarray(hashes, dtype=object), "utf8")
+    tm.assert_numpy_array_equal(result, np.concatenate([expected1, expected2], axis=0))
+
+
+@pytest.mark.parametrize(
+    "data, result_data",
+    [
+        [[tuple("1"), tuple("2")], [10345501319357378243, 8331063931016360761]],
+        [[(1,), (2,)], [9408946347443669104, 3278256261030523334]],
+    ],
+)
+def test_hash_with_tuple(data, result_data):
+    # GH#28969 array containing a tuple raises on call to arr.astype(str)
+    #  apparently a numpy bug github.com/numpy/numpy/issues/9441
+
+    df = DataFrame({"data": data})
+    result = hash_pandas_object(df)
+    expected = Series(result_data, dtype=np.uint64)
+    tm.assert_series_equal(result, expected)
+
+
+def test_hashable_tuple_args():
+    # require that the elements of such tuples are themselves hashable
+
+    df3 = DataFrame(
+        {
+            "data": [
+                (
+                    1,
+                    [],
+                ),
+                (
+                    2,
+                    {},
+                ),
+            ]
+        }
+    )
+    with pytest.raises(TypeError, match="unhashable type: 'list'"):
+        hash_pandas_object(df3)
+
+
+def test_hash_object_none_key():
+    # https://github.com/pandas-dev/pandas/issues/30887
+    result = pd.util.hash_pandas_object(Series(["a", "b"]), hash_key=None)
+    expected = Series([4578374827886788867, 17338122309987883691], dtype="uint64")
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/util/test_numba.py b/pandas/pandas/tests/util/test_numba.py
new file mode 100644
index 0000000000000..27b68ff0f6044
--- /dev/null
+++ b/pandas/pandas/tests/util/test_numba.py
@@ -0,0 +1,12 @@
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import option_context
+
+
+@td.skip_if_installed("numba")
+def test_numba_not_installed_option_context():
+    with pytest.raises(ImportError, match="Missing optional"):
+        with option_context("compute.use_numba", True):
+            pass
diff --git a/pandas/pandas/tests/util/test_rewrite_warning.py b/pandas/pandas/tests/util/test_rewrite_warning.py
new file mode 100644
index 0000000000000..f847a06d8ea8d
--- /dev/null
+++ b/pandas/pandas/tests/util/test_rewrite_warning.py
@@ -0,0 +1,39 @@
+import warnings
+
+import pytest
+
+from pandas.util._exceptions import rewrite_warning
+
+import pandas._testing as tm
+
+
+@pytest.mark.parametrize(
+    "target_category, target_message, hit",
+    [
+        (FutureWarning, "Target message", True),
+        (FutureWarning, "Target", True),
+        (FutureWarning, "get mess", True),
+        (FutureWarning, "Missed message", False),
+        (DeprecationWarning, "Target message", False),
+    ],
+)
+@pytest.mark.parametrize(
+    "new_category",
+    [
+        None,
+        DeprecationWarning,
+    ],
+)
+def test_rewrite_warning(target_category, target_message, hit, new_category):
+    new_message = "Rewritten message"
+    if hit:
+        expected_category = new_category if new_category else target_category
+        expected_message = new_message
+    else:
+        expected_category = FutureWarning
+        expected_message = "Target message"
+    with tm.assert_produces_warning(expected_category, match=expected_message):
+        with rewrite_warning(
+            target_message, target_category, new_message, new_category
+        ):
+            warnings.warn(message="Target message", category=FutureWarning)
diff --git a/pandas/pandas/tests/util/test_shares_memory.py b/pandas/pandas/tests/util/test_shares_memory.py
new file mode 100644
index 0000000000000..8f1ac93b40247
--- /dev/null
+++ b/pandas/pandas/tests/util/test_shares_memory.py
@@ -0,0 +1,32 @@
+import numpy as np
+
+import pandas.util._test_decorators as td
+
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_shares_memory_interval():
+    obj = pd.interval_range(1, 5)
+
+    assert tm.shares_memory(obj, obj)
+    assert tm.shares_memory(obj, obj._data)
+    assert tm.shares_memory(obj, obj[::-1])
+    assert tm.shares_memory(obj, obj[:2])
+
+    assert not tm.shares_memory(obj, obj._data.copy())
+
+
+@td.skip_if_no("pyarrow")
+def test_shares_memory_string():
+    # GH#55823
+    import pyarrow as pa
+
+    obj = pd.array(["a", "b"], dtype=pd.StringDtype("pyarrow", na_value=pd.NA))
+    assert tm.shares_memory(obj, obj)
+
+    obj = pd.array(["a", "b"], dtype=pd.StringDtype("pyarrow", na_value=np.nan))
+    assert tm.shares_memory(obj, obj)
+
+    obj = pd.array(["a", "b"], dtype=pd.ArrowDtype(pa.string()))
+    assert tm.shares_memory(obj, obj)
diff --git a/pandas/pandas/tests/util/test_show_versions.py b/pandas/pandas/tests/util/test_show_versions.py
new file mode 100644
index 0000000000000..72c9db23b2108
--- /dev/null
+++ b/pandas/pandas/tests/util/test_show_versions.py
@@ -0,0 +1,81 @@
+import json
+import os
+import re
+
+from pandas.util._print_versions import (
+    _get_dependency_info,
+    _get_sys_info,
+)
+
+import pandas as pd
+
+
+def test_show_versions(tmpdir):
+    # GH39701
+    as_json = os.path.join(tmpdir, "test_output.json")
+
+    pd.show_versions(as_json=as_json)
+
+    with open(as_json, encoding="utf-8") as fd:
+        # check if file output is valid JSON, will raise an exception if not
+        result = json.load(fd)
+
+    # Basic check that each version element is found in output
+    expected = {
+        "system": _get_sys_info(),
+        "dependencies": _get_dependency_info(),
+    }
+
+    assert result == expected
+
+
+def test_show_versions_console_json(capsys):
+    # GH39701
+    pd.show_versions(as_json=True)
+    stdout = capsys.readouterr().out
+
+    # check valid json is printed to the console if as_json is True
+    result = json.loads(stdout)
+
+    # Basic check that each version element is found in output
+    expected = {
+        "system": _get_sys_info(),
+        "dependencies": _get_dependency_info(),
+    }
+
+    assert result == expected
+
+
+def test_show_versions_console(capsys):
+    # gh-32041
+    # gh-32041
+    pd.show_versions(as_json=False)
+    result = capsys.readouterr().out
+
+    # check header
+    assert "INSTALLED VERSIONS" in result
+
+    # check full commit hash
+    assert re.search(r"commit\s*:\s[0-9a-f]{40}\n", result)
+
+    # check required dependency
+    # 2020-12-09 npdev has "dirty" in the tag
+    # 2022-05-25 npdev released with RC wo/ "dirty".
+    # Just ensure we match [0-9]+\..* since npdev version is variable
+    assert re.search(r"numpy\s*:\s[0-9]+\..*\n", result)
+
+    # check optional dependency
+    assert re.search(r"pyarrow\s*:\s([0-9]+.*|None)\n", result)
+
+
+def test_json_output_match(capsys, tmpdir):
+    # GH39701
+    pd.show_versions(as_json=True)
+    result_console = capsys.readouterr().out
+
+    out_path = os.path.join(tmpdir, "test_json.json")
+    pd.show_versions(as_json=out_path)
+    with open(out_path, encoding="utf-8") as out_fd:
+        result_file = out_fd.read()
+
+    assert result_console == result_file
diff --git a/pandas/pandas/tests/util/test_util.py b/pandas/pandas/tests/util/test_util.py
new file mode 100644
index 0000000000000..dfb8587d3924e
--- /dev/null
+++ b/pandas/pandas/tests/util/test_util.py
@@ -0,0 +1,58 @@
+import os
+
+import pytest
+
+from pandas import (
+    array,
+    compat,
+)
+import pandas._testing as tm
+
+
+def test_numpy_err_state_is_default():
+    expected = {"over": "warn", "divide": "warn", "invalid": "warn", "under": "ignore"}
+    import numpy as np
+
+    # The error state should be unchanged after that import.
+    assert np.geterr() == expected
+
+
+def test_convert_rows_list_to_csv_str():
+    rows_list = ["aaa", "bbb", "ccc"]
+    ret = tm.convert_rows_list_to_csv_str(rows_list)
+
+    if compat.is_platform_windows():
+        expected = "aaa\r\nbbb\r\nccc\r\n"
+    else:
+        expected = "aaa\nbbb\nccc\n"
+
+    assert ret == expected
+
+
+@pytest.mark.parametrize("strict_data_files", [True, False])
+def test_datapath_missing(datapath):
+    with pytest.raises(ValueError, match="Could not find file"):
+        datapath("not_a_file")
+
+
+def test_datapath(datapath):
+    args = ("io", "data", "csv", "iris.csv")
+
+    result = datapath(*args)
+    expected = os.path.join(os.path.dirname(os.path.dirname(__file__)), *args)
+
+    assert result == expected
+
+
+def test_external_error_raised():
+    with tm.external_error_raised(TypeError):
+        raise TypeError("Should not check this error message, so it will pass")
+
+
+def test_is_sorted():
+    arr = array([1, 2, 3], dtype="Int64")
+    tm.assert_is_sorted(arr)
+
+    arr = array([4, 2, 3], dtype="Int64")
+    with pytest.raises(AssertionError, match="ExtensionArray are different"):
+        tm.assert_is_sorted(arr)
diff --git a/pandas/pandas/tests/util/test_validate_args.py b/pandas/pandas/tests/util/test_validate_args.py
new file mode 100644
index 0000000000000..eef0931ec28ef
--- /dev/null
+++ b/pandas/pandas/tests/util/test_validate_args.py
@@ -0,0 +1,70 @@
+import pytest
+
+from pandas.util._validators import validate_args
+
+
+@pytest.fixture
+def _fname():
+    return "func"
+
+
+def test_bad_min_fname_arg_count(_fname):
+    msg = "'max_fname_arg_count' must be non-negative"
+
+    with pytest.raises(ValueError, match=msg):
+        validate_args(_fname, (None,), -1, "foo")
+
+
+def test_bad_arg_length_max_value_single(_fname):
+    args = (None, None)
+    compat_args = ("foo",)
+
+    min_fname_arg_count = 0
+    max_length = len(compat_args) + min_fname_arg_count
+    actual_length = len(args) + min_fname_arg_count
+    msg = (
+        rf"{_fname}\(\) takes at most {max_length} "
+        rf"argument \({actual_length} given\)"
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        validate_args(_fname, args, min_fname_arg_count, compat_args)
+
+
+def test_bad_arg_length_max_value_multiple(_fname):
+    args = (None, None)
+    compat_args = {"foo": None}
+
+    min_fname_arg_count = 2
+    max_length = len(compat_args) + min_fname_arg_count
+    actual_length = len(args) + min_fname_arg_count
+    msg = (
+        rf"{_fname}\(\) takes at most {max_length} "
+        rf"arguments \({actual_length} given\)"
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        validate_args(_fname, args, min_fname_arg_count, compat_args)
+
+
+@pytest.mark.parametrize("i", range(1, 3))
+def test_not_all_defaults(i, _fname):
+    bad_arg = "foo"
+    msg = (
+        f"the '{bad_arg}' parameter is not supported "
+        rf"in the pandas implementation of {_fname}\(\)"
+    )
+
+    compat_args = {"foo": 2, "bar": -1, "baz": 3}
+    arg_vals = (1, -1, 3)
+
+    with pytest.raises(ValueError, match=msg):
+        validate_args(_fname, arg_vals[:i], 2, compat_args)
+
+
+def test_validation(_fname):
+    # No exceptions should be raised.
+    validate_args(_fname, (None,), 2, {"out": None})
+
+    compat_args = {"axis": 1, "out": None}
+    validate_args(_fname, (1, None), 2, compat_args)
diff --git a/pandas/pandas/tests/util/test_validate_args_and_kwargs.py b/pandas/pandas/tests/util/test_validate_args_and_kwargs.py
new file mode 100644
index 0000000000000..215026d648471
--- /dev/null
+++ b/pandas/pandas/tests/util/test_validate_args_and_kwargs.py
@@ -0,0 +1,84 @@
+import pytest
+
+from pandas.util._validators import validate_args_and_kwargs
+
+
+@pytest.fixture
+def _fname():
+    return "func"
+
+
+def test_invalid_total_length_max_length_one(_fname):
+    compat_args = ("foo",)
+    kwargs = {"foo": "FOO"}
+    args = ("FoO", "BaZ")
+
+    min_fname_arg_count = 0
+    max_length = len(compat_args) + min_fname_arg_count
+    actual_length = len(kwargs) + len(args) + min_fname_arg_count
+
+    msg = (
+        rf"{_fname}\(\) takes at most {max_length} "
+        rf"argument \({actual_length} given\)"
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        validate_args_and_kwargs(_fname, args, kwargs, min_fname_arg_count, compat_args)
+
+
+def test_invalid_total_length_max_length_multiple(_fname):
+    compat_args = ("foo", "bar", "baz")
+    kwargs = {"foo": "FOO", "bar": "BAR"}
+    args = ("FoO", "BaZ")
+
+    min_fname_arg_count = 2
+    max_length = len(compat_args) + min_fname_arg_count
+    actual_length = len(kwargs) + len(args) + min_fname_arg_count
+
+    msg = (
+        rf"{_fname}\(\) takes at most {max_length} "
+        rf"arguments \({actual_length} given\)"
+    )
+
+    with pytest.raises(TypeError, match=msg):
+        validate_args_and_kwargs(_fname, args, kwargs, min_fname_arg_count, compat_args)
+
+
+@pytest.mark.parametrize("args,kwargs", [((), {"foo": -5, "bar": 2}), ((-5, 2), {})])
+def test_missing_args_or_kwargs(args, kwargs, _fname):
+    bad_arg = "bar"
+    min_fname_arg_count = 2
+
+    compat_args = {"foo": -5, bad_arg: 1}
+
+    msg = (
+        rf"the '{bad_arg}' parameter is not supported "
+        rf"in the pandas implementation of {_fname}\(\)"
+    )
+
+    with pytest.raises(ValueError, match=msg):
+        validate_args_and_kwargs(_fname, args, kwargs, min_fname_arg_count, compat_args)
+
+
+def test_duplicate_argument(_fname):
+    min_fname_arg_count = 2
+
+    compat_args = {"foo": None, "bar": None, "baz": None}
+    kwargs = {"foo": None, "bar": None}
+    args = (None,)  # duplicate value for "foo"
+
+    msg = rf"{_fname}\(\) got multiple values for keyword argument 'foo'"
+
+    with pytest.raises(TypeError, match=msg):
+        validate_args_and_kwargs(_fname, args, kwargs, min_fname_arg_count, compat_args)
+
+
+def test_validation(_fname):
+    # No exceptions should be raised.
+    compat_args = {"foo": 1, "bar": None, "baz": -2}
+    kwargs = {"baz": -2}
+
+    args = (1, None)
+    min_fname_arg_count = 2
+
+    validate_args_and_kwargs(_fname, args, kwargs, min_fname_arg_count, compat_args)
diff --git a/pandas/pandas/tests/util/test_validate_inclusive.py b/pandas/pandas/tests/util/test_validate_inclusive.py
new file mode 100644
index 0000000000000..c1254c614ab30
--- /dev/null
+++ b/pandas/pandas/tests/util/test_validate_inclusive.py
@@ -0,0 +1,40 @@
+import numpy as np
+import pytest
+
+from pandas.util._validators import validate_inclusive
+
+import pandas as pd
+
+
+@pytest.mark.parametrize(
+    "invalid_inclusive",
+    (
+        "ccc",
+        2,
+        object(),
+        None,
+        np.nan,
+        pd.NA,
+        pd.DataFrame(),
+    ),
+)
+def test_invalid_inclusive(invalid_inclusive):
+    with pytest.raises(
+        ValueError,
+        match="Inclusive has to be either 'both', 'neither', 'left' or 'right'",
+    ):
+        validate_inclusive(invalid_inclusive)
+
+
+@pytest.mark.parametrize(
+    "valid_inclusive, expected_tuple",
+    (
+        ("left", (True, False)),
+        ("right", (False, True)),
+        ("both", (True, True)),
+        ("neither", (False, False)),
+    ),
+)
+def test_valid_inclusive(valid_inclusive, expected_tuple):
+    resultant_tuple = validate_inclusive(valid_inclusive)
+    assert expected_tuple == resultant_tuple
diff --git a/pandas/pandas/tests/util/test_validate_kwargs.py b/pandas/pandas/tests/util/test_validate_kwargs.py
new file mode 100644
index 0000000000000..dba447e30cf57
--- /dev/null
+++ b/pandas/pandas/tests/util/test_validate_kwargs.py
@@ -0,0 +1,69 @@
+import pytest
+
+from pandas.util._validators import (
+    validate_bool_kwarg,
+    validate_kwargs,
+)
+
+
+@pytest.fixture
+def _fname():
+    return "func"
+
+
+def test_bad_kwarg(_fname):
+    good_arg = "f"
+    bad_arg = good_arg + "o"
+
+    compat_args = {good_arg: "foo", bad_arg + "o": "bar"}
+    kwargs = {good_arg: "foo", bad_arg: "bar"}
+
+    msg = rf"{_fname}\(\) got an unexpected keyword argument '{bad_arg}'"
+
+    with pytest.raises(TypeError, match=msg):
+        validate_kwargs(_fname, kwargs, compat_args)
+
+
+@pytest.mark.parametrize("i", range(1, 3))
+def test_not_all_none(i, _fname):
+    bad_arg = "foo"
+    msg = (
+        rf"the '{bad_arg}' parameter is not supported "
+        rf"in the pandas implementation of {_fname}\(\)"
+    )
+
+    compat_args = {"foo": 1, "bar": "s", "baz": None}
+
+    kwarg_keys = ("foo", "bar", "baz")
+    kwarg_vals = (2, "s", None)
+
+    kwargs = dict(zip(kwarg_keys[:i], kwarg_vals[:i]))
+
+    with pytest.raises(ValueError, match=msg):
+        validate_kwargs(_fname, kwargs, compat_args)
+
+
+def test_validation(_fname):
+    # No exceptions should be raised.
+    compat_args = {"f": None, "b": 1, "ba": "s"}
+
+    kwargs = {"f": None, "b": 1}
+    validate_kwargs(_fname, kwargs, compat_args)
+
+
+@pytest.mark.parametrize("name", ["inplace", "copy"])
+@pytest.mark.parametrize("value", [1, "True", [1, 2, 3], 5.0])
+def test_validate_bool_kwarg_fail(name, value):
+    msg = (
+        f'For argument "{name}" expected type bool, '
+        f"received type {type(value).__name__}"
+    )
+
+    with pytest.raises(ValueError, match=msg):
+        validate_bool_kwarg(value, name)
+
+
+@pytest.mark.parametrize("name", ["inplace", "copy"])
+@pytest.mark.parametrize("value", [True, False, None])
+def test_validate_bool_kwarg(name, value):
+    assert validate_bool_kwarg(value, name) == value
diff --git a/pandas/pandas/tests/window/__init__.py b/pandas/pandas/tests/window/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/window/conftest.py b/pandas/pandas/tests/window/conftest.py
new file mode 100644
index 0000000000000..fe873b3b74254
--- /dev/null
+++ b/pandas/pandas/tests/window/conftest.py
@@ -0,0 +1,124 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Series,
+    bdate_range,
+)
+
+
+@pytest.fixture(params=[True, False])
+def raw(request):
+    """raw keyword argument for rolling.apply"""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        "sum",
+        "mean",
+        "median",
+        "max",
+        "min",
+        "var",
+        "std",
+        "kurt",
+        "skew",
+        "count",
+        "sem",
+    ]
+)
+def arithmetic_win_operators(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def center(request):
+    return request.param
+
+
+@pytest.fixture(params=[None, 1])
+def min_periods(request):
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def adjust(request):
+    """adjust keyword argument for ewm"""
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def ignore_na(request):
+    """ignore_na keyword argument for ewm"""
+    return request.param
+
+
+@pytest.fixture(params=[True, False])
+def numeric_only(request):
+    """numeric_only keyword argument"""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        pytest.param("numba", marks=[td.skip_if_no("numba"), pytest.mark.single_cpu]),
+        "cython",
+    ]
+)
+def engine(request):
+    """engine keyword argument for rolling.apply"""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        pytest.param(
+            ("numba", True), marks=[td.skip_if_no("numba"), pytest.mark.single_cpu]
+        ),
+        ("cython", True),
+        ("cython", False),
+    ]
+)
+def engine_and_raw(request):
+    """engine and raw keyword arguments for rolling.apply"""
+    return request.param
+
+
+@pytest.fixture(params=["1 day", timedelta(days=1), np.timedelta64(1, "D")])
+def halflife_with_times(request):
+    """Halflife argument for EWM when times is specified."""
+    return request.param
+
+
+@pytest.fixture
+def series():
+    """Make mocked series as fixture."""
+    arr = np.random.default_rng(2).standard_normal(100)
+    locs = np.arange(20, 40)
+    arr[locs] = np.nan
+    series = Series(arr, index=bdate_range(datetime(2009, 1, 1), periods=100))
+    return series
+
+
+@pytest.fixture
+def frame():
+    """Make mocked frame as fixture."""
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((100, 10)),
+        index=bdate_range(datetime(2009, 1, 1), periods=100),
+    )
+
+
+@pytest.fixture(params=[None, 1, 2, 5, 10])
+def step(request):
+    """step keyword argument for rolling window operations."""
+    return request.param
diff --git a/pandas/pandas/tests/window/moments/__init__.py b/pandas/pandas/tests/window/moments/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/pandas/tests/window/moments/conftest.py b/pandas/pandas/tests/window/moments/conftest.py
new file mode 100644
index 0000000000000..fccf80c3c7a58
--- /dev/null
+++ b/pandas/pandas/tests/window/moments/conftest.py
@@ -0,0 +1,72 @@
+import itertools
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    notna,
+)
+
+
+def create_series():
+    return [
+        Series(dtype=np.float64, name="a"),
+        Series([np.nan] * 5),
+        Series([1.0] * 5),
+        Series(range(5, 0, -1)),
+        Series(range(5)),
+        Series([np.nan, 1.0, np.nan, 1.0, 1.0]),
+        Series([np.nan, 1.0, np.nan, 2.0, 3.0]),
+        Series([np.nan, 1.0, np.nan, 3.0, 2.0]),
+    ]
+
+
+def create_dataframes():
+    return [
+        DataFrame(columns=["a", "a"]),
+        DataFrame(np.arange(15).reshape((5, 3)), columns=["a", "a", 99]),
+    ] + [DataFrame(s) for s in create_series()]
+
+
+def is_constant(x):
+    values = x.values.ravel("K")
+    return len(set(values[notna(values)])) == 1
+
+
+@pytest.fixture(
+    params=(
+        obj
+        for obj in itertools.chain(create_series(), create_dataframes())
+        if is_constant(obj)
+    ),
+)
+def consistent_data(request):
+    return request.param
+
+
+@pytest.fixture(params=create_series())
+def series_data(request):
+    return request.param
+
+
+@pytest.fixture(params=itertools.chain(create_series(), create_dataframes()))
+def all_data(request):
+    """
+    Test:
+        - Empty Series / DataFrame
+        - All NaN
+        - All consistent value
+        - Monotonically decreasing
+        - Monotonically increasing
+        - Monotonically consistent with NaNs
+        - Monotonically increasing with NaNs
+        - Monotonically decreasing with NaNs
+    """
+    return request.param
+
+
+@pytest.fixture(params=[0, 2])
+def min_periods(request):
+    return request.param
diff --git a/pandas/pandas/tests/window/moments/test_moments_consistency_ewm.py b/pandas/pandas/tests/window/moments/test_moments_consistency_ewm.py
new file mode 100644
index 0000000000000..49dee50954f4f
--- /dev/null
+++ b/pandas/pandas/tests/window/moments/test_moments_consistency_ewm.py
@@ -0,0 +1,243 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    concat,
+)
+import pandas._testing as tm
+
+
+def create_mock_weights(obj, com, adjust, ignore_na):
+    if isinstance(obj, DataFrame):
+        if not len(obj.columns):
+            return DataFrame(index=obj.index, columns=obj.columns)
+        w = concat(
+            [
+                create_mock_series_weights(
+                    obj.iloc[:, i], com=com, adjust=adjust, ignore_na=ignore_na
+                )
+                for i in range(len(obj.columns))
+            ],
+            axis=1,
+        )
+        w.index = obj.index
+        w.columns = obj.columns
+        return w
+    else:
+        return create_mock_series_weights(obj, com, adjust, ignore_na)
+
+
+def create_mock_series_weights(s, com, adjust, ignore_na):
+    w = Series(np.nan, index=s.index, name=s.name)
+    alpha = 1.0 / (1.0 + com)
+    if adjust:
+        count = 0
+        for i in range(len(s)):
+            if s.iat[i] == s.iat[i]:
+                w.iat[i] = pow(1.0 / (1.0 - alpha), count)
+                count += 1
+            elif not ignore_na:
+                count += 1
+    else:
+        sum_wts = 0.0
+        prev_i = -1
+        count = 0
+        for i in range(len(s)):
+            if s.iat[i] == s.iat[i]:
+                if prev_i == -1:
+                    w.iat[i] = 1.0
+                else:
+                    w.iat[i] = alpha * sum_wts / pow(1.0 - alpha, count - prev_i)
+                sum_wts += w.iat[i]
+                prev_i = count
+                count += 1
+            elif not ignore_na:
+                count += 1
+    return w
+
+
+def test_ewm_consistency_mean(all_data, adjust, ignore_na, min_periods):
+    com = 3.0
+
+    result = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).mean()
+    weights = create_mock_weights(all_data, com=com, adjust=adjust, ignore_na=ignore_na)
+    expected = all_data.multiply(weights).cumsum().divide(weights.cumsum()).ffill()
+    expected[
+        all_data.expanding().count() < (max(min_periods, 1) if min_periods else 1)
+    ] = np.nan
+    tm.assert_equal(result, expected.astype("float64"))
+
+
+def test_ewm_consistency_consistent(consistent_data, adjust, ignore_na, min_periods):
+    com = 3.0
+
+    count_x = consistent_data.expanding().count()
+    mean_x = consistent_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).mean()
+    # check that correlation of a series with itself is either 1 or NaN
+    corr_x_x = consistent_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).corr(consistent_data)
+    exp = (
+        consistent_data.max()
+        if isinstance(consistent_data, Series)
+        else consistent_data.max().max()
+    )
+
+    # check mean of constant series
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = exp
+    tm.assert_equal(mean_x, expected)
+
+    # check correlation of constant series with itself is NaN
+    expected[:] = np.nan
+    tm.assert_equal(corr_x_x, expected)
+
+
+def test_ewm_consistency_var_debiasing_factors(
+    all_data, adjust, ignore_na, min_periods
+):
+    com = 3.0
+
+    # check variance debiasing factors
+    var_unbiased_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=False)
+    var_biased_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=True)
+
+    weights = create_mock_weights(all_data, com=com, adjust=adjust, ignore_na=ignore_na)
+    cum_sum = weights.cumsum().ffill()
+    cum_sum_sq = (weights * weights).cumsum().ffill()
+    numerator = cum_sum * cum_sum
+    denominator = numerator - cum_sum_sq
+    denominator[denominator <= 0.0] = np.nan
+    var_debiasing_factors_x = numerator / denominator
+
+    tm.assert_equal(var_unbiased_x, var_biased_x * var_debiasing_factors_x)
+
+
+@pytest.mark.parametrize("bias", [True, False])
+def test_moments_consistency_var(all_data, adjust, ignore_na, min_periods, bias):
+    com = 3.0
+
+    mean_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).mean()
+    var_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=bias)
+    assert not (var_x < 0).any().any()
+
+    if bias:
+        # check that biased var(x) == mean(x^2) - mean(x)^2
+        mean_x2 = (
+            (all_data * all_data)
+            .ewm(com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na)
+            .mean()
+        )
+        tm.assert_equal(var_x, mean_x2 - (mean_x * mean_x))
+
+
+@pytest.mark.parametrize("bias", [True, False])
+def test_moments_consistency_var_constant(
+    consistent_data, adjust, ignore_na, min_periods, bias
+):
+    com = 3.0
+    count_x = consistent_data.expanding(min_periods=min_periods).count()
+    var_x = consistent_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=bias)
+
+    # check that variance of constant series is identically 0
+    assert not (var_x > 0).any().any()
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = 0.0
+    if not bias:
+        expected[count_x < 2] = np.nan
+    tm.assert_equal(var_x, expected)
+
+
+@pytest.mark.parametrize("bias", [True, False])
+def test_ewm_consistency_std(all_data, adjust, ignore_na, min_periods, bias):
+    com = 3.0
+    var_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=bias)
+    assert not (var_x < 0).any().any()
+
+    std_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).std(bias=bias)
+    assert not (std_x < 0).any().any()
+
+    # check that var(x) == std(x)^2
+    tm.assert_equal(var_x, std_x * std_x)
+
+    cov_x_x = all_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).cov(all_data, bias=bias)
+    assert not (cov_x_x < 0).any().any()
+
+    # check that var(x) == cov(x, x)
+    tm.assert_equal(var_x, cov_x_x)
+
+
+@pytest.mark.parametrize("bias", [True, False])
+def test_ewm_consistency_series_cov_corr(
+    series_data, adjust, ignore_na, min_periods, bias
+):
+    com = 3.0
+
+    var_x_plus_y = (
+        (series_data + series_data)
+        .ewm(com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na)
+        .var(bias=bias)
+    )
+    var_x = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=bias)
+    var_y = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).var(bias=bias)
+    cov_x_y = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).cov(series_data, bias=bias)
+    # check that cov(x, y) == (var(x+y) - var(x) -
+    # var(y)) / 2
+    tm.assert_equal(cov_x_y, 0.5 * (var_x_plus_y - var_x - var_y))
+
+    # check that corr(x, y) == cov(x, y) / (std(x) *
+    # std(y))
+    corr_x_y = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).corr(series_data)
+    std_x = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).std(bias=bias)
+    std_y = series_data.ewm(
+        com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+    ).std(bias=bias)
+    tm.assert_equal(corr_x_y, cov_x_y / (std_x * std_y))
+
+    if bias:
+        # check that biased cov(x, y) == mean(x*y) -
+        # mean(x)*mean(y)
+        mean_x = series_data.ewm(
+            com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+        ).mean()
+        mean_y = series_data.ewm(
+            com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na
+        ).mean()
+        mean_x_times_y = (
+            (series_data * series_data)
+            .ewm(com=com, min_periods=min_periods, adjust=adjust, ignore_na=ignore_na)
+            .mean()
+        )
+        tm.assert_equal(cov_x_y, mean_x_times_y - (mean_x * mean_y))
diff --git a/pandas/pandas/tests/window/moments/test_moments_consistency_expanding.py b/pandas/pandas/tests/window/moments/test_moments_consistency_expanding.py
new file mode 100644
index 0000000000000..7d2fa1ad5d211
--- /dev/null
+++ b/pandas/pandas/tests/window/moments/test_moments_consistency_expanding.py
@@ -0,0 +1,144 @@
+import numpy as np
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+def no_nans(x):
+    return x.notna().all().all()
+
+
+def all_na(x):
+    return x.isnull().all().all()
+
+
+@pytest.mark.parametrize("f", [lambda v: Series(v).sum(), np.nansum, np.sum])
+def test_expanding_apply_consistency_sum_nans(request, all_data, min_periods, f):
+    if f is np.sum:
+        if not no_nans(all_data) and not (
+            all_na(all_data) and not all_data.empty and min_periods > 0
+        ):
+            request.applymarker(
+                pytest.mark.xfail(reason="np.sum has different behavior with NaNs")
+            )
+    expanding_f_result = all_data.expanding(min_periods=min_periods).sum()
+    expanding_apply_f_result = all_data.expanding(min_periods=min_periods).apply(
+        func=f, raw=True
+    )
+    tm.assert_equal(expanding_f_result, expanding_apply_f_result)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_moments_consistency_var(all_data, min_periods, ddof):
+    var_x = all_data.expanding(min_periods=min_periods).var(ddof=ddof)
+    assert not (var_x < 0).any().any()
+
+    if ddof == 0:
+        # check that biased var(x) == mean(x^2) - mean(x)^2
+        mean_x2 = (all_data * all_data).expanding(min_periods=min_periods).mean()
+        mean_x = all_data.expanding(min_periods=min_periods).mean()
+        tm.assert_equal(var_x, mean_x2 - (mean_x * mean_x))
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_moments_consistency_var_constant(consistent_data, min_periods, ddof):
+    count_x = consistent_data.expanding(min_periods=min_periods).count()
+    var_x = consistent_data.expanding(min_periods=min_periods).var(ddof=ddof)
+
+    # check that variance of constant series is identically 0
+    assert not (var_x > 0).any().any()
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = 0.0
+    if ddof == 1:
+        expected[count_x < 2] = np.nan
+    tm.assert_equal(var_x, expected)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_expanding_consistency_var_std_cov(all_data, min_periods, ddof):
+    var_x = all_data.expanding(min_periods=min_periods).var(ddof=ddof)
+    assert not (var_x < 0).any().any()
+
+    std_x = all_data.expanding(min_periods=min_periods).std(ddof=ddof)
+    assert not (std_x < 0).any().any()
+
+    # check that var(x) == std(x)^2
+    tm.assert_equal(var_x, std_x * std_x)
+
+    cov_x_x = all_data.expanding(min_periods=min_periods).cov(all_data, ddof=ddof)
+    assert not (cov_x_x < 0).any().any()
+
+    # check that var(x) == cov(x, x)
+    tm.assert_equal(var_x, cov_x_x)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_expanding_consistency_series_cov_corr(series_data, min_periods, ddof):
+    var_x_plus_y = (
+        (series_data + series_data).expanding(min_periods=min_periods).var(ddof=ddof)
+    )
+    var_x = series_data.expanding(min_periods=min_periods).var(ddof=ddof)
+    var_y = series_data.expanding(min_periods=min_periods).var(ddof=ddof)
+    cov_x_y = series_data.expanding(min_periods=min_periods).cov(series_data, ddof=ddof)
+    # check that cov(x, y) == (var(x+y) - var(x) -
+    # var(y)) / 2
+    tm.assert_equal(cov_x_y, 0.5 * (var_x_plus_y - var_x - var_y))
+
+    # check that corr(x, y) == cov(x, y) / (std(x) *
+    # std(y))
+    corr_x_y = series_data.expanding(min_periods=min_periods).corr(series_data)
+    std_x = series_data.expanding(min_periods=min_periods).std(ddof=ddof)
+    std_y = series_data.expanding(min_periods=min_periods).std(ddof=ddof)
+    tm.assert_equal(corr_x_y, cov_x_y / (std_x * std_y))
+
+    if ddof == 0:
+        # check that biased cov(x, y) == mean(x*y) -
+        # mean(x)*mean(y)
+        mean_x = series_data.expanding(min_periods=min_periods).mean()
+        mean_y = series_data.expanding(min_periods=min_periods).mean()
+        mean_x_times_y = (
+            (series_data * series_data).expanding(min_periods=min_periods).mean()
+        )
+        tm.assert_equal(cov_x_y, mean_x_times_y - (mean_x * mean_y))
+
+
+def test_expanding_consistency_mean(all_data, min_periods):
+    result = all_data.expanding(min_periods=min_periods).mean()
+    expected = (
+        all_data.expanding(min_periods=min_periods).sum()
+        / all_data.expanding(min_periods=min_periods).count()
+    )
+    tm.assert_equal(result, expected.astype("float64"))
+
+
+def test_expanding_consistency_constant(consistent_data, min_periods):
+    count_x = consistent_data.expanding().count()
+    mean_x = consistent_data.expanding(min_periods=min_periods).mean()
+    # check that correlation of a series with itself is either 1 or NaN
+    corr_x_x = consistent_data.expanding(min_periods=min_periods).corr(consistent_data)
+
+    exp = (
+        consistent_data.max()
+        if isinstance(consistent_data, Series)
+        else consistent_data.max().max()
+    )
+
+    # check mean of constant series
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = exp
+    tm.assert_equal(mean_x, expected)
+
+    # check correlation of constant series with itself is NaN
+    expected[:] = np.nan
+    tm.assert_equal(corr_x_x, expected)
+
+
+def test_expanding_consistency_var_debiasing_factors(all_data, min_periods):
+    # check variance debiasing factors
+    var_unbiased_x = all_data.expanding(min_periods=min_periods).var()
+    var_biased_x = all_data.expanding(min_periods=min_periods).var(ddof=0)
+    var_debiasing_factors_x = all_data.expanding().count() / (
+        all_data.expanding().count() - 1.0
+    ).replace(0.0, np.nan)
+    tm.assert_equal(var_unbiased_x, var_biased_x * var_debiasing_factors_x)
diff --git a/pandas/pandas/tests/window/moments/test_moments_consistency_rolling.py b/pandas/pandas/tests/window/moments/test_moments_consistency_rolling.py
new file mode 100644
index 0000000000000..be22338c00cb2
--- /dev/null
+++ b/pandas/pandas/tests/window/moments/test_moments_consistency_rolling.py
@@ -0,0 +1,244 @@
+import numpy as np
+import pytest
+
+from pandas import Series
+import pandas._testing as tm
+
+
+def no_nans(x):
+    return x.notna().all().all()
+
+
+def all_na(x):
+    return x.isnull().all().all()
+
+
+@pytest.fixture(params=[(1, 0), (5, 1)])
+def rolling_consistency_cases(request):
+    """window, min_periods"""
+    return request.param
+
+
+@pytest.mark.parametrize("f", [lambda v: Series(v).sum(), np.nansum, np.sum])
+def test_rolling_apply_consistency_sum(
+    request, all_data, rolling_consistency_cases, center, f
+):
+    window, min_periods = rolling_consistency_cases
+
+    if f is np.sum:
+        if not no_nans(all_data) and not (
+            all_na(all_data) and not all_data.empty and min_periods > 0
+        ):
+            request.applymarker(
+                pytest.mark.xfail(reason="np.sum has different behavior with NaNs")
+            )
+    rolling_f_result = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).sum()
+    rolling_apply_f_result = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).apply(func=f, raw=True)
+    tm.assert_equal(rolling_f_result, rolling_apply_f_result)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_moments_consistency_var(all_data, rolling_consistency_cases, center, ddof):
+    window, min_periods = rolling_consistency_cases
+
+    var_x = all_data.rolling(window=window, min_periods=min_periods, center=center).var(
+        ddof=ddof
+    )
+    assert not (var_x < 0).any().any()
+
+    if ddof == 0:
+        # check that biased var(x) == mean(x^2) - mean(x)^2
+        mean_x = all_data.rolling(
+            window=window, min_periods=min_periods, center=center
+        ).mean()
+        mean_x2 = (
+            (all_data * all_data)
+            .rolling(window=window, min_periods=min_periods, center=center)
+            .mean()
+        )
+        tm.assert_equal(var_x, mean_x2 - (mean_x * mean_x))
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_moments_consistency_var_constant(
+    consistent_data, rolling_consistency_cases, center, ddof
+):
+    window, min_periods = rolling_consistency_cases
+
+    count_x = consistent_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).count()
+    var_x = consistent_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).var(ddof=ddof)
+
+    # check that variance of constant series is identically 0
+    assert not (var_x > 0).any().any()
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = 0.0
+    if ddof == 1:
+        expected[count_x < 2] = np.nan
+    tm.assert_equal(var_x, expected)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_rolling_consistency_var_std_cov(
+    all_data, rolling_consistency_cases, center, ddof
+):
+    window, min_periods = rolling_consistency_cases
+
+    var_x = all_data.rolling(window=window, min_periods=min_periods, center=center).var(
+        ddof=ddof
+    )
+    assert not (var_x < 0).any().any()
+
+    std_x = all_data.rolling(window=window, min_periods=min_periods, center=center).std(
+        ddof=ddof
+    )
+    assert not (std_x < 0).any().any()
+
+    # check that var(x) == std(x)^2
+    tm.assert_equal(var_x, std_x * std_x)
+
+    cov_x_x = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).cov(all_data, ddof=ddof)
+    assert not (cov_x_x < 0).any().any()
+
+    # check that var(x) == cov(x, x)
+    tm.assert_equal(var_x, cov_x_x)
+
+
+@pytest.mark.parametrize("ddof", [0, 1])
+def test_rolling_consistency_series_cov_corr(
+    series_data, rolling_consistency_cases, center, ddof
+):
+    window, min_periods = rolling_consistency_cases
+
+    var_x_plus_y = (
+        (series_data + series_data)
+        .rolling(window=window, min_periods=min_periods, center=center)
+        .var(ddof=ddof)
+    )
+    var_x = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).var(ddof=ddof)
+    var_y = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).var(ddof=ddof)
+    cov_x_y = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).cov(series_data, ddof=ddof)
+    # check that cov(x, y) == (var(x+y) - var(x) -
+    # var(y)) / 2
+    tm.assert_equal(cov_x_y, 0.5 * (var_x_plus_y - var_x - var_y))
+
+    # check that corr(x, y) == cov(x, y) / (std(x) *
+    # std(y))
+    corr_x_y = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).corr(series_data)
+    std_x = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).std(ddof=ddof)
+    std_y = series_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).std(ddof=ddof)
+    tm.assert_equal(corr_x_y, cov_x_y / (std_x * std_y))
+
+    if ddof == 0:
+        # check that biased cov(x, y) == mean(x*y) -
+        # mean(x)*mean(y)
+        mean_x = series_data.rolling(
+            window=window, min_periods=min_periods, center=center
+        ).mean()
+        mean_y = series_data.rolling(
+            window=window, min_periods=min_periods, center=center
+        ).mean()
+        mean_x_times_y = (
+            (series_data * series_data)
+            .rolling(window=window, min_periods=min_periods, center=center)
+            .mean()
+        )
+        tm.assert_equal(cov_x_y, mean_x_times_y - (mean_x * mean_y))
+
+
+def test_rolling_consistency_mean(all_data, rolling_consistency_cases, center):
+    window, min_periods = rolling_consistency_cases
+
+    result = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).mean()
+    expected = (
+        all_data.rolling(window=window, min_periods=min_periods, center=center)
+        .sum()
+        .divide(
+            all_data.rolling(
+                window=window, min_periods=min_periods, center=center
+            ).count()
+        )
+    )
+    tm.assert_equal(result, expected.astype("float64"))
+
+
+def test_rolling_consistency_constant(
+    consistent_data, rolling_consistency_cases, center
+):
+    window, min_periods = rolling_consistency_cases
+
+    count_x = consistent_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).count()
+    mean_x = consistent_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).mean()
+    # check that correlation of a series with itself is either 1 or NaN
+    corr_x_x = consistent_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).corr(consistent_data)
+
+    exp = (
+        consistent_data.max()
+        if isinstance(consistent_data, Series)
+        else consistent_data.max().max()
+    )
+
+    # check mean of constant series
+    expected = consistent_data * np.nan
+    expected[count_x >= max(min_periods, 1)] = exp
+    tm.assert_equal(mean_x, expected)
+
+    # check correlation of constant series with itself is NaN
+    expected[:] = np.nan
+    tm.assert_equal(corr_x_x, expected)
+
+
+def test_rolling_consistency_var_debiasing_factors(
+    all_data, rolling_consistency_cases, center
+):
+    window, min_periods = rolling_consistency_cases
+
+    # check variance debiasing factors
+    var_unbiased_x = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).var()
+    var_biased_x = all_data.rolling(
+        window=window, min_periods=min_periods, center=center
+    ).var(ddof=0)
+    var_debiasing_factors_x = (
+        all_data.rolling(window=window, min_periods=min_periods, center=center)
+        .count()
+        .divide(
+            (
+                all_data.rolling(
+                    window=window, min_periods=min_periods, center=center
+                ).count()
+                - 1.0
+            ).replace(0.0, np.nan)
+        )
+    )
+    tm.assert_equal(var_unbiased_x, var_biased_x * var_debiasing_factors_x)
diff --git a/pandas/pandas/tests/window/test_api.py b/pandas/pandas/tests/window/test_api.py
new file mode 100644
index 0000000000000..877b50e37670c
--- /dev/null
+++ b/pandas/pandas/tests/window/test_api.py
@@ -0,0 +1,385 @@
+import numpy as np
+import pytest
+
+from pandas.errors import (
+    DataError,
+    SpecificationError,
+)
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Period,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    timedelta_range,
+)
+import pandas._testing as tm
+
+
+def test_getitem(step):
+    frame = DataFrame(np.random.default_rng(2).standard_normal((5, 5)))
+    r = frame.rolling(window=5, step=step)
+    tm.assert_index_equal(r._selected_obj.columns, frame[::step].columns)
+
+    r = frame.rolling(window=5, step=step)[1]
+    assert r._selected_obj.name == frame[::step].columns[1]
+
+    # technically this is allowed
+    r = frame.rolling(window=5, step=step)[1, 3]
+    tm.assert_index_equal(r._selected_obj.columns, frame[::step].columns[[1, 3]])
+
+    r = frame.rolling(window=5, step=step)[[1, 3]]
+    tm.assert_index_equal(r._selected_obj.columns, frame[::step].columns[[1, 3]])
+
+
+def test_select_bad_cols():
+    df = DataFrame([[1, 2]], columns=["A", "B"])
+    g = df.rolling(window=5)
+    with pytest.raises(KeyError, match="Columns not found: 'C'"):
+        g[["C"]]
+    with pytest.raises(KeyError, match="^[^A]+$"):
+        # A should not be referenced as a bad column...
+        # will have to rethink regex if you change message!
+        g[["A", "C"]]
+
+
+def test_attribute_access():
+    df = DataFrame([[1, 2]], columns=["A", "B"])
+    r = df.rolling(window=5)
+    tm.assert_series_equal(r.A.sum(), r["A"].sum())
+    msg = "'Rolling' object has no attribute 'F'"
+    with pytest.raises(AttributeError, match=msg):
+        r.F
+
+
+def tests_skip_nuisance(step):
+    df = DataFrame({"A": range(5), "B": range(5, 10), "C": "foo"})
+    r = df.rolling(window=3, step=step)
+    result = r[["A", "B"]].sum()
+    expected = DataFrame(
+        {"A": [np.nan, np.nan, 3, 6, 9], "B": [np.nan, np.nan, 18, 21, 24]},
+        columns=list("AB"),
+    )[::step]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_sum_object_str_raises(step):
+    df = DataFrame({"A": range(5), "B": range(5, 10), "C": "foo"})
+    r = df.rolling(window=3, step=step)
+    with pytest.raises(
+        DataError, match="Cannot aggregate non-numeric type: object|str"
+    ):
+        # GH#42738, enforced in 2.0
+        r.sum()
+
+
+def test_agg(step):
+    df = DataFrame({"A": range(5), "B": range(0, 10, 2)})
+
+    r = df.rolling(window=3, step=step)
+    a_mean = r["A"].mean()
+    a_std = r["A"].std()
+    a_sum = r["A"].sum()
+    b_mean = r["B"].mean()
+    b_std = r["B"].std()
+
+    result = r.aggregate([np.mean, lambda x: np.std(x, ddof=1)])
+    expected = concat([a_mean, a_std, b_mean, b_std], axis=1)
+    expected.columns = MultiIndex.from_product([["A", "B"], ["mean", "<lambda>"]])
+    tm.assert_frame_equal(result, expected)
+
+    result = r.aggregate({"A": np.mean, "B": lambda x: np.std(x, ddof=1)})
+
+    expected = concat([a_mean, b_std], axis=1)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+    result = r.aggregate({"A": ["mean", "std"]})
+    expected = concat([a_mean, a_std], axis=1)
+    expected.columns = MultiIndex.from_tuples([("A", "mean"), ("A", "std")])
+    tm.assert_frame_equal(result, expected)
+
+    result = r["A"].aggregate(["mean", "sum"])
+    expected = concat([a_mean, a_sum], axis=1)
+    expected.columns = ["mean", "sum"]
+    tm.assert_frame_equal(result, expected)
+
+    msg = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        # using a dict with renaming
+        r.aggregate({"A": {"mean": "mean", "sum": "sum"}})
+
+    with pytest.raises(SpecificationError, match=msg):
+        r.aggregate(
+            {"A": {"mean": "mean", "sum": "sum"}, "B": {"mean2": "mean", "sum2": "sum"}}
+        )
+
+    result = r.aggregate({"A": ["mean", "std"], "B": ["mean", "std"]})
+    expected = concat([a_mean, a_std, b_mean, b_std], axis=1)
+
+    exp_cols = [("A", "mean"), ("A", "std"), ("B", "mean"), ("B", "std")]
+    expected.columns = MultiIndex.from_tuples(exp_cols)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+def test_agg_apply(raw):
+    # passed lambda
+    df = DataFrame({"A": range(5), "B": range(0, 10, 2)})
+
+    r = df.rolling(window=3)
+    a_sum = r["A"].sum()
+
+    result = r.agg({"A": np.sum, "B": lambda x: np.std(x, ddof=1)})
+    rcustom = r["B"].apply(lambda x: np.std(x, ddof=1), raw=raw)
+    expected = concat([a_sum, rcustom], axis=1)
+    tm.assert_frame_equal(result, expected, check_like=True)
+
+
+def test_agg_consistency(step):
+    df = DataFrame({"A": range(5), "B": range(0, 10, 2)})
+    r = df.rolling(window=3, step=step)
+
+    result = r.agg([np.sum, np.mean]).columns
+    expected = MultiIndex.from_product([list("AB"), ["sum", "mean"]])
+    tm.assert_index_equal(result, expected)
+
+    result = r["A"].agg([np.sum, np.mean]).columns
+    expected = Index(["sum", "mean"])
+    tm.assert_index_equal(result, expected)
+
+    result = r.agg({"A": [np.sum, np.mean]}).columns
+    expected = MultiIndex.from_tuples([("A", "sum"), ("A", "mean")])
+    tm.assert_index_equal(result, expected)
+
+
+def test_agg_nested_dicts():
+    # API change for disallowing these types of nested dicts
+    df = DataFrame({"A": range(5), "B": range(0, 10, 2)})
+    r = df.rolling(window=3)
+
+    msg = "nested renamer is not supported"
+    with pytest.raises(SpecificationError, match=msg):
+        r.aggregate({"r1": {"A": ["mean", "sum"]}, "r2": {"B": ["mean", "sum"]}})
+
+    expected = concat(
+        [r["A"].mean(), r["A"].std(), r["B"].mean(), r["B"].std()], axis=1
+    )
+    expected.columns = MultiIndex.from_tuples(
+        [("ra", "mean"), ("ra", "std"), ("rb", "mean"), ("rb", "std")]
+    )
+    with pytest.raises(SpecificationError, match=msg):
+        r[["A", "B"]].agg({"A": {"ra": ["mean", "std"]}, "B": {"rb": ["mean", "std"]}})
+
+    with pytest.raises(SpecificationError, match=msg):
+        r.agg({"A": {"ra": ["mean", "std"]}, "B": {"rb": ["mean", "std"]}})
+
+
+@pytest.mark.parametrize(
+    "func,window_size",
+    [
+        (
+            "rolling",
+            2,
+        ),
+        (
+            "expanding",
+            None,
+        ),
+    ],
+)
+def test_pipe(func, window_size):
+    # Issue #57076
+    df = DataFrame(
+        {
+            "B": np.random.default_rng(2).standard_normal(10),
+            "C": np.random.default_rng(2).standard_normal(10),
+        }
+    )
+    r = getattr(df, func)(window_size)
+
+    expected = r.max() - r.mean()
+    result = r.pipe(lambda x: x.max() - x.mean())
+    tm.assert_frame_equal(result, expected)
+
+    expected = r.max() - 2 * r.min()
+    result = r.pipe(lambda x, k: x.max() - k * x.min(), k=2)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_count_nonnumeric_types(step):
+    # GH12541
+    cols = [
+        "int",
+        "float",
+        "string",
+        "datetime",
+        "timedelta",
+        "periods",
+        "fl_inf",
+        "fl_nan",
+        "str_nan",
+        "dt_nat",
+        "periods_nat",
+    ]
+    dt_nat_col = [Timestamp("20170101"), Timestamp("20170203"), Timestamp(None)]
+
+    df = DataFrame(
+        {
+            "int": [1, 2, 3],
+            "float": [4.0, 5.0, 6.0],
+            "string": list("abc"),
+            "datetime": date_range("20170101", periods=3),
+            "timedelta": timedelta_range("1 s", periods=3, freq="s"),
+            "periods": [
+                Period("2012-01"),
+                Period("2012-02"),
+                Period("2012-03"),
+            ],
+            "fl_inf": [1.0, 2.0, np.inf],
+            "fl_nan": [1.0, 2.0, np.nan],
+            "str_nan": ["aa", "bb", np.nan],
+            "dt_nat": dt_nat_col,
+            "periods_nat": [
+                Period("2012-01"),
+                Period("2012-02"),
+                Period(None),
+            ],
+        },
+        columns=cols,
+    )
+
+    expected = DataFrame(
+        {
+            "int": [1.0, 2.0, 2.0],
+            "float": [1.0, 2.0, 2.0],
+            "string": [1.0, 2.0, 2.0],
+            "datetime": [1.0, 2.0, 2.0],
+            "timedelta": [1.0, 2.0, 2.0],
+            "periods": [1.0, 2.0, 2.0],
+            "fl_inf": [1.0, 2.0, 2.0],
+            "fl_nan": [1.0, 2.0, 1.0],
+            "str_nan": [1.0, 2.0, 1.0],
+            "dt_nat": [1.0, 2.0, 1.0],
+            "periods_nat": [1.0, 2.0, 1.0],
+        },
+        columns=cols,
+    )[::step]
+
+    result = df.rolling(window=2, min_periods=0, step=step).count()
+    tm.assert_frame_equal(result, expected)
+
+    result = df.rolling(1, min_periods=0, step=step).count()
+    expected = df.notna().astype(float)[::step]
+    tm.assert_frame_equal(result, expected)
+
+
+def test_preserve_metadata():
+    # GH 10565
+    s = Series(np.arange(100), name="foo")
+
+    s2 = s.rolling(30).sum()
+    s3 = s.rolling(20).sum()
+    assert s2.name == "foo"
+    assert s3.name == "foo"
+
+
+@pytest.mark.parametrize(
+    "func,window_size,expected_vals",
+    [
+        (
+            "rolling",
+            2,
+            [
+                [np.nan, np.nan, np.nan, np.nan],
+                [15.0, 20.0, 25.0, 20.0],
+                [25.0, 30.0, 35.0, 30.0],
+                [np.nan, np.nan, np.nan, np.nan],
+                [20.0, 30.0, 35.0, 30.0],
+                [35.0, 40.0, 60.0, 40.0],
+                [60.0, 80.0, 85.0, 80],
+            ],
+        ),
+        (
+            "expanding",
+            None,
+            [
+                [10.0, 10.0, 20.0, 20.0],
+                [15.0, 20.0, 25.0, 20.0],
+                [20.0, 30.0, 30.0, 20.0],
+                [10.0, 10.0, 30.0, 30.0],
+                [20.0, 30.0, 35.0, 30.0],
+                [26.666667, 40.0, 50.0, 30.0],
+                [40.0, 80.0, 60.0, 30.0],
+            ],
+        ),
+    ],
+)
+def test_multiple_agg_funcs(func, window_size, expected_vals):
+    # GH 15072
+    df = DataFrame(
+        [
+            ["A", 10, 20],
+            ["A", 20, 30],
+            ["A", 30, 40],
+            ["B", 10, 30],
+            ["B", 30, 40],
+            ["B", 40, 80],
+            ["B", 80, 90],
+        ],
+        columns=["stock", "low", "high"],
+    )
+
+    f = getattr(df.groupby("stock"), func)
+    if window_size:
+        window = f(window_size)
+    else:
+        window = f()
+
+    index = MultiIndex.from_tuples(
+        [("A", 0), ("A", 1), ("A", 2), ("B", 3), ("B", 4), ("B", 5), ("B", 6)],
+        names=["stock", None],
+    )
+    columns = MultiIndex.from_tuples(
+        [("low", "mean"), ("low", "max"), ("high", "mean"), ("high", "min")]
+    )
+    expected = DataFrame(expected_vals, index=index, columns=columns)
+
+    result = window.agg({"low": ["mean", "max"], "high": ["mean", "min"]})
+
+    tm.assert_frame_equal(result, expected)
+
+
+def test_dont_modify_attributes_after_methods(
+    arithmetic_win_operators, closed, center, min_periods, step
+):
+    # GH 39554
+    roll_obj = Series(range(1)).rolling(
+        1, center=center, closed=closed, min_periods=min_periods, step=step
+    )
+    expected = {attr: getattr(roll_obj, attr) for attr in roll_obj._attributes}
+    getattr(roll_obj, arithmetic_win_operators)()
+    result = {attr: getattr(roll_obj, attr) for attr in roll_obj._attributes}
+    assert result == expected
+
+
+def test_rolling_min_min_periods(step):
+    a = Series([1, 2, 3, 4, 5])
+    result = a.rolling(window=100, min_periods=1, step=step).min()
+    expected = Series(np.ones(len(a)))[::step]
+    tm.assert_series_equal(result, expected)
+    msg = "min_periods 5 must be <= window 3"
+    with pytest.raises(ValueError, match=msg):
+        Series([1, 2, 3]).rolling(window=3, min_periods=5, step=step).min()
+
+
+def test_rolling_max_min_periods(step):
+    a = Series([1, 2, 3, 4, 5], dtype=np.float64)
+    result = a.rolling(window=100, min_periods=1, step=step).max()
+    expected = a[::step]
+    tm.assert_almost_equal(result, expected)
+    msg = "min_periods 5 must be <= window 3"
+    with pytest.raises(ValueError, match=msg):
+        Series([1, 2, 3]).rolling(window=3, min_periods=5, step=step).max()
diff --git a/pandas/pandas/tests/window/test_apply.py b/pandas/pandas/tests/window/test_apply.py
new file mode 100644
index 0000000000000..2398713585cfb
--- /dev/null
+++ b/pandas/pandas/tests/window/test_apply.py
@@ -0,0 +1,318 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+# suppress warnings about empty slices, as we are deliberately testing
+# with a 0-length Series
+pytestmark = pytest.mark.filterwarnings(
+    "ignore:.*(empty slice|0 for slice).*:RuntimeWarning"
+)
+
+
+def f(x):
+    return x[np.isfinite(x)].mean()
+
+
+@pytest.mark.parametrize("bad_raw", [None, 1, 0])
+def test_rolling_apply_invalid_raw(bad_raw):
+    with pytest.raises(ValueError, match="raw parameter must be `True` or `False`"):
+        Series(range(3)).rolling(1).apply(len, raw=bad_raw)
+
+
+def test_rolling_apply_out_of_bounds(engine_and_raw):
+    # gh-1850
+    engine, raw = engine_and_raw
+
+    vals = Series([1, 2, 3, 4])
+
+    result = vals.rolling(10).apply(np.sum, engine=engine, raw=raw)
+    assert result.isna().all()
+
+    result = vals.rolling(10, min_periods=1).apply(np.sum, engine=engine, raw=raw)
+    expected = Series([1, 3, 6, 10], dtype=float)
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize("window", [2, "2s"])
+def test_rolling_apply_with_pandas_objects(window):
+    # 5071
+    df = DataFrame(
+        {
+            "A": np.random.default_rng(2).standard_normal(5),
+            "B": np.random.default_rng(2).integers(0, 10, size=5),
+        },
+        index=date_range("20130101", periods=5, freq="s"),
+    )
+
+    # we have an equal spaced timeseries index
+    # so simulate removing the first period
+    def f(x):
+        if x.index[0] == df.index[0]:
+            return np.nan
+        return x.iloc[-1]
+
+    result = df.rolling(window).apply(f, raw=False)
+    expected = df.iloc[2:].reindex_like(df)
+    tm.assert_frame_equal(result, expected)
+
+    with tm.external_error_raised(AttributeError):
+        df.rolling(window).apply(f, raw=True)
+
+
+def test_rolling_apply(engine_and_raw, step):
+    engine, raw = engine_and_raw
+
+    expected = Series([], dtype="float64")
+    result = expected.rolling(10, step=step).apply(
+        lambda x: x.mean(), engine=engine, raw=raw
+    )
+    tm.assert_series_equal(result, expected)
+
+    # gh-8080
+    s = Series([None, None, None])
+    result = s.rolling(2, min_periods=0, step=step).apply(
+        lambda x: len(x), engine=engine, raw=raw
+    )
+    expected = Series([1.0, 2.0, 2.0])[::step]
+    tm.assert_series_equal(result, expected)
+
+    result = s.rolling(2, min_periods=0, step=step).apply(len, engine=engine, raw=raw)
+    tm.assert_series_equal(result, expected)
+
+
+def test_all_apply(engine_and_raw):
+    engine, raw = engine_and_raw
+
+    df = (
+        DataFrame(
+            {"A": date_range("20130101", periods=5, freq="s"), "B": range(5)}
+        ).set_index("A")
+        * 2
+    )
+    er = df.rolling(window=1)
+    r = df.rolling(window="1s")
+
+    result = r.apply(lambda x: 1, engine=engine, raw=raw)
+    expected = er.apply(lambda x: 1, engine=engine, raw=raw)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ragged_apply(engine_and_raw):
+    engine, raw = engine_and_raw
+
+    df = DataFrame({"B": range(5)})
+    df.index = [
+        Timestamp("20130101 09:00:00"),
+        Timestamp("20130101 09:00:02"),
+        Timestamp("20130101 09:00:03"),
+        Timestamp("20130101 09:00:05"),
+        Timestamp("20130101 09:00:06"),
+    ]
+
+    f = lambda x: 1
+    result = df.rolling(window="1s", min_periods=1).apply(f, engine=engine, raw=raw)
+    expected = df.copy()
+    expected["B"] = 1.0
+    tm.assert_frame_equal(result, expected)
+
+    result = df.rolling(window="2s", min_periods=1).apply(f, engine=engine, raw=raw)
+    expected = df.copy()
+    expected["B"] = 1.0
+    tm.assert_frame_equal(result, expected)
+
+    result = df.rolling(window="5s", min_periods=1).apply(f, engine=engine, raw=raw)
+    expected = df.copy()
+    expected["B"] = 1.0
+    tm.assert_frame_equal(result, expected)
+
+
+def test_invalid_engine():
+    with pytest.raises(ValueError, match="engine must be either 'numba' or 'cython'"):
+        Series(range(1)).rolling(1).apply(lambda x: x, engine="foo")
+
+
+def test_invalid_engine_kwargs_cython():
+    with pytest.raises(ValueError, match="cython engine does not accept engine_kwargs"):
+        Series(range(1)).rolling(1).apply(
+            lambda x: x, engine="cython", engine_kwargs={"nopython": False}
+        )
+
+
+def test_invalid_raw_numba():
+    with pytest.raises(
+        ValueError, match="raw must be `True` when using the numba engine"
+    ):
+        Series(range(1)).rolling(1).apply(lambda x: x, raw=False, engine="numba")
+
+
+@pytest.mark.parametrize("args_kwargs", [[None, {"par": 10}], [(10,), None]])
+def test_rolling_apply_args_kwargs(args_kwargs):
+    # GH 33433
+    def numpysum(x, par):
+        return np.sum(x + par)
+
+    df = DataFrame({"gr": [1, 1], "a": [1, 2]})
+
+    idx = Index(["gr", "a"])
+    expected = DataFrame([[11.0, 11.0], [11.0, 12.0]], columns=idx)
+
+    result = df.rolling(1).apply(numpysum, args=args_kwargs[0], kwargs=args_kwargs[1])
+    tm.assert_frame_equal(result, expected)
+
+    midx = MultiIndex.from_tuples([(1, 0), (1, 1)], names=["gr", None])
+    expected = Series([11.0, 12.0], index=midx, name="a")
+
+    gb_rolling = df.groupby("gr")["a"].rolling(1)
+
+    result = gb_rolling.apply(numpysum, args=args_kwargs[0], kwargs=args_kwargs[1])
+    tm.assert_series_equal(result, expected)
+
+
+def test_nans(raw):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = obj.rolling(50, min_periods=30).apply(f, raw=raw)
+    tm.assert_almost_equal(result.iloc[-1], np.mean(obj[10:-10]))
+
+    # min_periods is working correctly
+    result = obj.rolling(20, min_periods=15).apply(f, raw=raw)
+    assert isna(result.iloc[23])
+    assert not isna(result.iloc[24])
+
+    assert not isna(result.iloc[-6])
+    assert isna(result.iloc[-5])
+
+    obj2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = obj2.rolling(10, min_periods=5).apply(f, raw=raw)
+    assert isna(result.iloc[3])
+    assert notna(result.iloc[4])
+
+    result0 = obj.rolling(20, min_periods=0).apply(f, raw=raw)
+    result1 = obj.rolling(20, min_periods=1).apply(f, raw=raw)
+    tm.assert_almost_equal(result0, result1)
+
+
+def test_center(raw):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = obj.rolling(20, min_periods=15, center=True).apply(f, raw=raw)
+    expected = (
+        concat([obj, Series([np.nan] * 9)])
+        .rolling(20, min_periods=15)
+        .apply(f, raw=raw)
+        .iloc[9:]
+        .reset_index(drop=True)
+    )
+    tm.assert_series_equal(result, expected)
+
+
+def test_series(raw, series):
+    result = series.rolling(50).apply(f, raw=raw)
+    assert isinstance(result, Series)
+    tm.assert_almost_equal(result.iloc[-1], np.mean(series[-50:]))
+
+
+def test_frame(raw, frame):
+    result = frame.rolling(50).apply(f, raw=raw)
+    assert isinstance(result, DataFrame)
+    tm.assert_series_equal(
+        result.iloc[-1, :],
+        frame.iloc[-50:, :].apply(np.mean, axis=0, raw=raw),
+        check_names=False,
+    )
+
+
+def test_time_rule_series(raw, series):
+    win = 25
+    minp = 10
+    ser = series[::2].resample("B").mean()
+    series_result = ser.rolling(window=win, min_periods=minp).apply(f, raw=raw)
+    last_date = series_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_series = series[::2].truncate(prev_date, last_date)
+    tm.assert_almost_equal(series_result.iloc[-1], np.mean(trunc_series))
+
+
+def test_time_rule_frame(raw, frame):
+    win = 25
+    minp = 10
+    frm = frame[::2].resample("B").mean()
+    frame_result = frm.rolling(window=win, min_periods=minp).apply(f, raw=raw)
+    last_date = frame_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_frame = frame[::2].truncate(prev_date, last_date)
+    tm.assert_series_equal(
+        frame_result.xs(last_date),
+        trunc_frame.apply(np.mean, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize("minp", [0, 99, 100])
+def test_min_periods(raw, series, minp, step):
+    result = series.rolling(len(series) + 1, min_periods=minp, step=step).apply(
+        f, raw=raw
+    )
+    expected = series.rolling(len(series), min_periods=minp, step=step).apply(
+        f, raw=raw
+    )
+    nan_mask = isna(result)
+    tm.assert_series_equal(nan_mask, isna(expected))
+
+    nan_mask = ~nan_mask
+    tm.assert_almost_equal(result[nan_mask], expected[nan_mask])
+
+
+def test_center_reindex_series(raw, series):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+    minp = 10
+
+    series_xp = (
+        series.reindex(list(series.index) + s)
+        .rolling(window=25, min_periods=minp)
+        .apply(f, raw=raw)
+        .shift(-12)
+        .reindex(series.index)
+    )
+    series_rs = series.rolling(window=25, min_periods=minp, center=True).apply(
+        f, raw=raw
+    )
+    tm.assert_series_equal(series_xp, series_rs)
+
+
+def test_center_reindex_frame(raw):
+    # shifter index
+    frame = DataFrame(range(100), index=date_range("2020-01-01", freq="D", periods=100))
+    s = [f"x{x:d}" for x in range(12)]
+    minp = 10
+
+    frame_xp = (
+        frame.reindex(list(frame.index) + s)
+        .rolling(window=25, min_periods=minp)
+        .apply(f, raw=raw)
+        .shift(-12)
+        .reindex(frame.index)
+    )
+    frame_rs = frame.rolling(window=25, min_periods=minp, center=True).apply(f, raw=raw)
+    tm.assert_frame_equal(frame_xp, frame_rs)
diff --git a/pandas/pandas/tests/window/test_base_indexer.py b/pandas/pandas/tests/window/test_base_indexer.py
new file mode 100644
index 0000000000000..4f91e56a7d82b
--- /dev/null
+++ b/pandas/pandas/tests/window/test_base_indexer.py
@@ -0,0 +1,519 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    MultiIndex,
+    Series,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.indexers import (
+    BaseIndexer,
+    FixedForwardWindowIndexer,
+)
+from pandas.core.indexers.objects import (
+    ExpandingIndexer,
+    FixedWindowIndexer,
+    VariableOffsetWindowIndexer,
+)
+
+from pandas.tseries.offsets import BusinessDay
+
+
+def test_bad_get_window_bounds_signature():
+    class BadIndexer(BaseIndexer):
+        def get_window_bounds(self):
+            return None
+
+    indexer = BadIndexer()
+    with pytest.raises(ValueError, match="BadIndexer does not implement"):
+        Series(range(5)).rolling(indexer)
+
+
+def test_expanding_indexer():
+    s = Series(range(10))
+    indexer = ExpandingIndexer()
+    result = s.rolling(indexer).mean()
+    expected = s.expanding().mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_indexer_constructor_arg():
+    # Example found in computation.rst
+    use_expanding = [True, False, True, False, True]
+    df = DataFrame({"values": range(5)})
+
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            start = np.empty(num_values, dtype=np.int64)
+            end = np.empty(num_values, dtype=np.int64)
+            for i in range(num_values):
+                if self.use_expanding[i]:
+                    start[i] = 0
+                    end[i] = i + 1
+                else:
+                    start[i] = i
+                    end[i] = i + self.window_size
+            return start, end
+
+    indexer = CustomIndexer(window_size=1, use_expanding=use_expanding)
+    result = df.rolling(indexer).sum()
+    expected = DataFrame({"values": [0.0, 1.0, 3.0, 3.0, 10.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_indexer_accepts_rolling_args():
+    df = DataFrame({"values": range(5)})
+
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            start = np.empty(num_values, dtype=np.int64)
+            end = np.empty(num_values, dtype=np.int64)
+            for i in range(num_values):
+                if (
+                    center
+                    and min_periods == 1
+                    and closed == "both"
+                    and step == 1
+                    and i == 2
+                ):
+                    start[i] = 0
+                    end[i] = num_values
+                else:
+                    start[i] = i
+                    end[i] = i + self.window_size
+            return start, end
+
+    indexer = CustomIndexer(window_size=1)
+    result = df.rolling(
+        indexer, center=True, min_periods=1, closed="both", step=1
+    ).sum()
+    expected = DataFrame({"values": [0.0, 1.0, 10.0, 3.0, 4.0]})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func,np_func,expected,np_kwargs",
+    [
+        ("count", len, [3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 2.0, np.nan], {}),
+        ("min", np.min, [0.0, 1.0, 2.0, 3.0, 4.0, 6.0, 6.0, 7.0, 8.0, np.nan], {}),
+        (
+            "max",
+            np.max,
+            [2.0, 3.0, 4.0, 100.0, 100.0, 100.0, 8.0, 9.0, 9.0, np.nan],
+            {},
+        ),
+        (
+            "std",
+            np.std,
+            [
+                1.0,
+                1.0,
+                1.0,
+                55.71654452,
+                54.85739087,
+                53.9845657,
+                1.0,
+                1.0,
+                0.70710678,
+                np.nan,
+            ],
+            {"ddof": 1},
+        ),
+        (
+            "var",
+            np.var,
+            [
+                1.0,
+                1.0,
+                1.0,
+                3104.333333,
+                3009.333333,
+                2914.333333,
+                1.0,
+                1.0,
+                0.500000,
+                np.nan,
+            ],
+            {"ddof": 1},
+        ),
+        (
+            "median",
+            np.median,
+            [1.0, 2.0, 3.0, 4.0, 6.0, 7.0, 7.0, 8.0, 8.5, np.nan],
+            {},
+        ),
+    ],
+)
+def test_rolling_forward_window(
+    frame_or_series, func, np_func, expected, np_kwargs, step
+):
+    # GH 32865
+    values = np.arange(10.0)
+    values[5] = 100.0
+
+    indexer = FixedForwardWindowIndexer(window_size=3)
+
+    match = "Forward-looking windows can't have center=True"
+    rolling = frame_or_series(values).rolling(window=indexer, center=True)
+    with pytest.raises(ValueError, match=match):
+        getattr(rolling, func)()
+
+    match = "Forward-looking windows don't support setting the closed argument"
+    rolling = frame_or_series(values).rolling(window=indexer, closed="right")
+    with pytest.raises(ValueError, match=match):
+        getattr(rolling, func)()
+
+    rolling = frame_or_series(values).rolling(window=indexer, min_periods=2, step=step)
+    result = getattr(rolling, func)()
+
+    # Check that the function output matches the explicitly provided array
+    expected = frame_or_series(expected)[::step]
+    tm.assert_equal(result, expected)
+
+    # Check that the rolling function output matches applying an alternative
+    # function to the rolling window object
+    expected2 = frame_or_series(rolling.apply(lambda x: np_func(x, **np_kwargs)))
+    tm.assert_equal(result, expected2)
+
+    # Check that the function output matches applying an alternative function
+    # if min_periods isn't specified
+    # GH 39604: After count-min_periods deprecation, apply(lambda x: len(x))
+    # is equivalent to count after setting min_periods=0
+    min_periods = 0 if func == "count" else None
+    rolling3 = frame_or_series(values).rolling(window=indexer, min_periods=min_periods)
+    result3 = getattr(rolling3, func)()
+    expected3 = frame_or_series(rolling3.apply(lambda x: np_func(x, **np_kwargs)))
+    tm.assert_equal(result3, expected3)
+
+
+def test_rolling_forward_skewness(frame_or_series, step):
+    values = np.arange(10.0)
+    values[5] = 100.0
+
+    indexer = FixedForwardWindowIndexer(window_size=5)
+    rolling = frame_or_series(values).rolling(window=indexer, min_periods=3, step=step)
+    result = rolling.skew()
+
+    expected = frame_or_series(
+        [
+            0.0,
+            2.232396,
+            2.229508,
+            2.228340,
+            2.229091,
+            2.231989,
+            0.0,
+            0.0,
+            np.nan,
+            np.nan,
+        ]
+    )[::step]
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func,expected",
+    [
+        ("cov", [2.0, 2.0, 2.0, 97.0, 2.0, -93.0, 2.0, 2.0, np.nan, np.nan]),
+        (
+            "corr",
+            [
+                1.0,
+                1.0,
+                1.0,
+                0.8704775290207161,
+                0.018229084250926637,
+                -0.861357304646493,
+                1.0,
+                1.0,
+                np.nan,
+                np.nan,
+            ],
+        ),
+    ],
+)
+def test_rolling_forward_cov_corr(func, expected):
+    values1 = np.arange(10).reshape(-1, 1)
+    values2 = values1 * 2
+    values1[5, 0] = 100
+    values = np.concatenate([values1, values2], axis=1)
+
+    indexer = FixedForwardWindowIndexer(window_size=3)
+    rolling = DataFrame(values).rolling(window=indexer, min_periods=3)
+    # We are interested in checking only pairwise covariance / correlation
+    result = getattr(rolling, func)().loc[(slice(None), 1), 0]
+    result = result.reset_index(drop=True)
+    expected = Series(expected).reset_index(drop=True)
+    expected.name = result.name
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "closed,expected_data",
+    [
+        ["right", [0.0, 1.0, 2.0, 3.0, 7.0, 12.0, 6.0, 7.0, 8.0, 9.0]],
+        ["left", [0.0, 0.0, 1.0, 2.0, 5.0, 9.0, 5.0, 6.0, 7.0, 8.0]],
+    ],
+)
+def test_non_fixed_variable_window_indexer(closed, expected_data):
+    index = date_range("2020", periods=10)
+    df = DataFrame(range(10), index=index)
+    offset = BusinessDay(1)
+    indexer = VariableOffsetWindowIndexer(index=index, offset=offset)
+    result = df.rolling(indexer, closed=closed).sum()
+    expected = DataFrame(expected_data, index=index)
+    tm.assert_frame_equal(result, expected)
+
+
+def test_variableoffsetwindowindexer_not_dti():
+    # GH 54379
+    with pytest.raises(ValueError, match="index must be a DatetimeIndex."):
+        VariableOffsetWindowIndexer(index="foo", offset=BusinessDay(1))
+
+
+def test_variableoffsetwindowindexer_not_offset():
+    # GH 54379
+    idx = date_range("2020", periods=10)
+    with pytest.raises(ValueError, match="offset must be a DateOffset-like object."):
+        VariableOffsetWindowIndexer(index=idx, offset="foo")
+
+
+def test_fixed_forward_indexer_count(step):
+    # GH: 35579
+    df = DataFrame({"b": [None, None, None, 7]})
+    indexer = FixedForwardWindowIndexer(window_size=2)
+    result = df.rolling(window=indexer, min_periods=0, step=step).count()
+    expected = DataFrame({"b": [0.0, 0.0, 1.0, 1.0]})[::step]
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("end_value", "values"), [(1, [0.0, 1, 1, 3, 2]), (-1, [0.0, 1, 0, 3, 1])]
+)
+@pytest.mark.parametrize(("func", "args"), [("median", []), ("quantile", [0.5])])
+def test_indexer_quantile_sum(end_value, values, func, args):
+    # GH 37153
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            start = np.empty(num_values, dtype=np.int64)
+            end = np.empty(num_values, dtype=np.int64)
+            for i in range(num_values):
+                if self.use_expanding[i]:
+                    start[i] = 0
+                    end[i] = max(i + end_value, 1)
+                else:
+                    start[i] = i
+                    end[i] = i + self.window_size
+            return start, end
+
+    use_expanding = [True, False, True, False, True]
+    df = DataFrame({"values": range(5)})
+
+    indexer = CustomIndexer(window_size=1, use_expanding=use_expanding)
+    result = getattr(df.rolling(indexer), func)(*args)
+    expected = DataFrame({"values": values})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "indexer_class", [FixedWindowIndexer, FixedForwardWindowIndexer, ExpandingIndexer]
+)
+@pytest.mark.parametrize("window_size", [1, 2, 12])
+@pytest.mark.parametrize(
+    "df_data",
+    [
+        {"a": [1, 1], "b": [0, 1]},
+        {"a": [1, 2], "b": [0, 1]},
+        {"a": [1] * 16, "b": [np.nan, 1, 2, np.nan] + list(range(4, 16))},
+    ],
+)
+def test_indexers_are_reusable_after_groupby_rolling(
+    indexer_class, window_size, df_data
+):
+    # GH 43267
+    df = DataFrame(df_data)
+    num_trials = 3
+    indexer = indexer_class(window_size=window_size)
+    original_window_size = indexer.window_size
+    for i in range(num_trials):
+        df.groupby("a")["b"].rolling(window=indexer, min_periods=1).mean()
+        assert indexer.window_size == original_window_size
+
+
+@pytest.mark.parametrize(
+    "window_size, num_values, expected_start, expected_end",
+    [
+        (1, 1, [0], [1]),
+        (1, 2, [0, 1], [1, 2]),
+        (2, 1, [0], [1]),
+        (2, 2, [0, 1], [2, 2]),
+        (5, 12, range(12), list(range(5, 12)) + [12] * 5),
+        (12, 5, range(5), [5] * 5),
+        (0, 0, np.array([]), np.array([])),
+        (1, 0, np.array([]), np.array([])),
+        (0, 1, [0], [0]),
+    ],
+)
+def test_fixed_forward_indexer_bounds(
+    window_size, num_values, expected_start, expected_end, step
+):
+    # GH 43267
+    indexer = FixedForwardWindowIndexer(window_size=window_size)
+    start, end = indexer.get_window_bounds(num_values=num_values, step=step)
+
+    tm.assert_numpy_array_equal(
+        start, np.array(expected_start[::step]), check_dtype=False
+    )
+    tm.assert_numpy_array_equal(end, np.array(expected_end[::step]), check_dtype=False)
+    assert len(start) == len(end)
+
+
+@pytest.mark.parametrize(
+    "df, window_size, expected",
+    [
+        (
+            DataFrame({"b": [0, 1, 2], "a": [1, 2, 2]}),
+            2,
+            Series(
+                [0, 1.5, 2.0],
+                index=MultiIndex.from_arrays([[1, 2, 2], range(3)], names=["a", None]),
+                name="b",
+                dtype=np.float64,
+            ),
+        ),
+        (
+            DataFrame(
+                {
+                    "b": [np.nan, 1, 2, np.nan] + list(range(4, 18)),
+                    "a": [1] * 7 + [2] * 11,
+                    "c": range(18),
+                }
+            ),
+            12,
+            Series(
+                [
+                    3.6,
+                    3.6,
+                    4.25,
+                    5.0,
+                    5.0,
+                    5.5,
+                    6.0,
+                    12.0,
+                    12.5,
+                    13.0,
+                    13.5,
+                    14.0,
+                    14.5,
+                    15.0,
+                    15.5,
+                    16.0,
+                    16.5,
+                    17.0,
+                ],
+                index=MultiIndex.from_arrays(
+                    [[1] * 7 + [2] * 11, range(18)], names=["a", None]
+                ),
+                name="b",
+                dtype=np.float64,
+            ),
+        ),
+    ],
+)
+def test_rolling_groupby_with_fixed_forward_specific(df, window_size, expected):
+    # GH 43267
+    indexer = FixedForwardWindowIndexer(window_size=window_size)
+    result = df.groupby("a")["b"].rolling(window=indexer, min_periods=1).mean()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "group_keys",
+    [
+        (1,),
+        (1, 2),
+        (2, 1),
+        (1, 1, 2),
+        (1, 2, 1),
+        (1, 1, 2, 2),
+        (1, 2, 3, 2, 3),
+        (1, 1, 2) * 4,
+        (1, 2, 3) * 5,
+    ],
+)
+@pytest.mark.parametrize("window_size", [1, 2, 3, 4, 5, 8, 20])
+def test_rolling_groupby_with_fixed_forward_many(group_keys, window_size):
+    # GH 43267
+    df = DataFrame(
+        {
+            "a": np.array(list(group_keys)),
+            "b": np.arange(len(group_keys), dtype=np.float64) + 17,
+            "c": np.arange(len(group_keys), dtype=np.int64),
+        }
+    )
+
+    indexer = FixedForwardWindowIndexer(window_size=window_size)
+    result = df.groupby("a")["b"].rolling(window=indexer, min_periods=1).sum()
+    result.index.names = ["a", "c"]
+
+    groups = df.groupby("a")[["a", "b", "c"]]
+    manual = concat(
+        [
+            g.assign(
+                b=[
+                    g["b"].iloc[i : i + window_size].sum(min_count=1)
+                    for i in range(len(g))
+                ]
+            )
+            for _, g in groups
+        ]
+    )
+    manual = manual.set_index(["a", "c"])["b"]
+
+    tm.assert_series_equal(result, manual)
+
+
+def test_unequal_start_end_bounds():
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            return np.array([1]), np.array([1, 2])
+
+    indexer = CustomIndexer()
+    roll = Series(1).rolling(indexer)
+    match = "start"
+    with pytest.raises(ValueError, match=match):
+        roll.mean()
+
+    with pytest.raises(ValueError, match=match):
+        next(iter(roll))
+
+    with pytest.raises(ValueError, match=match):
+        roll.corr(pairwise=True)
+
+    with pytest.raises(ValueError, match=match):
+        roll.cov(pairwise=True)
+
+
+def test_unequal_bounds_to_object():
+    # GH 44470
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            return np.array([1]), np.array([2])
+
+    indexer = CustomIndexer()
+    roll = Series([1, 1]).rolling(indexer)
+    match = "start and end"
+    with pytest.raises(ValueError, match=match):
+        roll.mean()
+
+    with pytest.raises(ValueError, match=match):
+        next(iter(roll))
+
+    with pytest.raises(ValueError, match=match):
+        roll.corr(pairwise=True)
+
+    with pytest.raises(ValueError, match=match):
+        roll.cov(pairwise=True)
diff --git a/pandas/pandas/tests/window/test_cython_aggregations.py b/pandas/pandas/tests/window/test_cython_aggregations.py
new file mode 100644
index 0000000000000..feb25a294c540
--- /dev/null
+++ b/pandas/pandas/tests/window/test_cython_aggregations.py
@@ -0,0 +1,113 @@
+from functools import partial
+import sys
+
+import numpy as np
+import pytest
+
+import pandas._libs.window.aggregations as window_aggregations
+
+from pandas import Series
+import pandas._testing as tm
+
+
+def _get_rolling_aggregations():
+    # list pairs of name and function
+    # each function has this signature:
+    # (const float64_t[:] values, ndarray[int64_t] start,
+    #  ndarray[int64_t] end, int64_t minp) -> np.ndarray
+    named_roll_aggs = (
+        [
+            ("roll_sum", window_aggregations.roll_sum),
+            ("roll_mean", window_aggregations.roll_mean),
+        ]
+        + [
+            (f"roll_var({ddof})", partial(window_aggregations.roll_var, ddof=ddof))
+            for ddof in [0, 1]
+        ]
+        + [
+            ("roll_skew", window_aggregations.roll_skew),
+            ("roll_kurt", window_aggregations.roll_kurt),
+            ("roll_median_c", window_aggregations.roll_median_c),
+            ("roll_max", window_aggregations.roll_max),
+            ("roll_min", window_aggregations.roll_min),
+            ("roll_first", window_aggregations.roll_first),
+            ("roll_last", window_aggregations.roll_last),
+        ]
+        + [
+            (
+                f"roll_quantile({quantile},{interpolation})",
+                partial(
+                    window_aggregations.roll_quantile,
+                    quantile=quantile,
+                    interpolation=interpolation,
+                ),
+            )
+            for quantile in [0.0001, 0.5, 0.9999]
+            for interpolation in window_aggregations.interpolation_types
+        ]
+        + [
+            (
+                f"roll_rank({percentile},{method},{ascending})",
+                partial(
+                    window_aggregations.roll_rank,
+                    percentile=percentile,
+                    method=method,
+                    ascending=ascending,
+                ),
+            )
+            for percentile in [True, False]
+            for method in window_aggregations.rolling_rank_tiebreakers.keys()
+            for ascending in [True, False]
+        ]
+    )
+    # unzip to a list of 2 tuples, names and functions
+    unzipped = list(zip(*named_roll_aggs))
+    return {"ids": unzipped[0], "params": unzipped[1]}
+
+
+_rolling_aggregations = _get_rolling_aggregations()
+
+
+@pytest.fixture(
+    params=_rolling_aggregations["params"], ids=_rolling_aggregations["ids"]
+)
+def rolling_aggregation(request):
+    """Make a rolling aggregation function as fixture."""
+    return request.param
+
+
+def test_rolling_aggregation_boundary_consistency(rolling_aggregation):
+    # GH-45647
+    minp, step, width, size, selection = 0, 1, 3, 11, [2, 7]
+    values = np.arange(1, 1 + size, dtype=np.float64)
+    end = np.arange(width, size, step, dtype=np.int64)
+    start = end - width
+    selarr = np.array(selection, dtype=np.int32)
+    result = Series(rolling_aggregation(values, start[selarr], end[selarr], minp))
+    expected = Series(rolling_aggregation(values, start, end, minp)[selarr])
+    tm.assert_equal(expected, result)
+
+
+def test_rolling_aggregation_with_unused_elements(rolling_aggregation):
+    # GH-45647
+    minp, width = 0, 5  # width at least 4 for kurt
+    size = 2 * width + 5
+    values = np.arange(1, size + 1, dtype=np.float64)
+    values[width : width + 2] = sys.float_info.min
+    values[width + 2] = np.nan
+    values[width + 3 : width + 5] = sys.float_info.max
+    start = np.array([0, size - width], dtype=np.int64)
+    end = np.array([width, size], dtype=np.int64)
+    loc = np.array(
+        [j for i in range(len(start)) for j in range(start[i], end[i])],
+        dtype=np.int32,
+    )
+    result = Series(rolling_aggregation(values, start, end, minp))
+    compact_values = np.array(values[loc], dtype=np.float64)
+    compact_start = np.arange(0, len(start) * width, width, dtype=np.int64)
+    compact_end = compact_start + width
+    expected = Series(
+        rolling_aggregation(compact_values, compact_start, compact_end, minp)
+    )
+    assert np.isfinite(expected.values).all(), "Not all expected values are finite"
+    tm.assert_equal(expected, result)
diff --git a/pandas/pandas/tests/window/test_dtypes.py b/pandas/pandas/tests/window/test_dtypes.py
new file mode 100644
index 0000000000000..4007320b5de33
--- /dev/null
+++ b/pandas/pandas/tests/window/test_dtypes.py
@@ -0,0 +1,173 @@
+import numpy as np
+import pytest
+
+from pandas.errors import DataError
+
+from pandas.core.dtypes.common import pandas_dtype
+
+from pandas import (
+    NA,
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+
+# gh-12373 : rolling functions error on float32 data
+# make sure rolling functions works for different dtypes
+#
+# further note that we are only checking rolling for fully dtype
+# compliance (though both expanding and ewm inherit)
+
+
+def get_dtype(dtype, coerce_int=None):
+    if coerce_int is False and "int" in dtype:
+        return None
+    return pandas_dtype(dtype)
+
+
+@pytest.fixture(
+    params=[
+        "object",
+        "category",
+        "int8",
+        "int16",
+        "int32",
+        "int64",
+        "uint8",
+        "uint16",
+        "uint32",
+        "uint64",
+        "float16",
+        "float32",
+        "float64",
+        "m8[ns]",
+        "M8[ns]",
+        "datetime64[ns, UTC]",
+    ]
+)
+def dtypes(request):
+    """Dtypes for window tests"""
+    return request.param
+
+
+@pytest.mark.parametrize(
+    "method, data, expected_data, coerce_int, min_periods",
+    [
+        ("count", np.arange(5), [1, 2, 2, 2, 2], True, 0),
+        ("count", np.arange(10, 0, -2), [1, 2, 2, 2, 2], True, 0),
+        ("count", [0, 1, 2, np.nan, 4], [1, 2, 2, 1, 1], False, 0),
+        ("max", np.arange(5), [np.nan, 1, 2, 3, 4], True, None),
+        ("max", np.arange(10, 0, -2), [np.nan, 10, 8, 6, 4], True, None),
+        ("max", [0, 1, 2, np.nan, 4], [np.nan, 1, 2, np.nan, np.nan], False, None),
+        ("min", np.arange(5), [np.nan, 0, 1, 2, 3], True, None),
+        ("min", np.arange(10, 0, -2), [np.nan, 8, 6, 4, 2], True, None),
+        ("min", [0, 1, 2, np.nan, 4], [np.nan, 0, 1, np.nan, np.nan], False, None),
+        ("sum", np.arange(5), [np.nan, 1, 3, 5, 7], True, None),
+        ("sum", np.arange(10, 0, -2), [np.nan, 18, 14, 10, 6], True, None),
+        ("sum", [0, 1, 2, np.nan, 4], [np.nan, 1, 3, np.nan, np.nan], False, None),
+        ("mean", np.arange(5), [np.nan, 0.5, 1.5, 2.5, 3.5], True, None),
+        ("mean", np.arange(10, 0, -2), [np.nan, 9, 7, 5, 3], True, None),
+        ("mean", [0, 1, 2, np.nan, 4], [np.nan, 0.5, 1.5, np.nan, np.nan], False, None),
+        ("std", np.arange(5), [np.nan] + [np.sqrt(0.5)] * 4, True, None),
+        ("std", np.arange(10, 0, -2), [np.nan] + [np.sqrt(2)] * 4, True, None),
+        (
+            "std",
+            [0, 1, 2, np.nan, 4],
+            [np.nan] + [np.sqrt(0.5)] * 2 + [np.nan] * 2,
+            False,
+            None,
+        ),
+        ("var", np.arange(5), [np.nan, 0.5, 0.5, 0.5, 0.5], True, None),
+        ("var", np.arange(10, 0, -2), [np.nan, 2, 2, 2, 2], True, None),
+        ("var", [0, 1, 2, np.nan, 4], [np.nan, 0.5, 0.5, np.nan, np.nan], False, None),
+        ("median", np.arange(5), [np.nan, 0.5, 1.5, 2.5, 3.5], True, None),
+        ("median", np.arange(10, 0, -2), [np.nan, 9, 7, 5, 3], True, None),
+        (
+            "median",
+            [0, 1, 2, np.nan, 4],
+            [np.nan, 0.5, 1.5, np.nan, np.nan],
+            False,
+            None,
+        ),
+    ],
+)
+def test_series_dtypes(
+    method, data, expected_data, coerce_int, dtypes, min_periods, step
+):
+    ser = Series(data, dtype=get_dtype(dtypes, coerce_int=coerce_int))
+    rolled = ser.rolling(2, min_periods=min_periods, step=step)
+
+    if dtypes in ("m8[ns]", "M8[ns]", "datetime64[ns, UTC]") and method != "count":
+        msg = "No numeric types to aggregate"
+        with pytest.raises(DataError, match=msg):
+            getattr(rolled, method)()
+    else:
+        result = getattr(rolled, method)()
+        expected = Series(expected_data, dtype="float64")[::step]
+        tm.assert_almost_equal(result, expected)
+
+
+def test_series_nullable_int(any_signed_int_ea_dtype, step):
+    # GH 43016
+    ser = Series([0, 1, NA], dtype=any_signed_int_ea_dtype)
+    result = ser.rolling(2, step=step).mean()
+    expected = Series([np.nan, 0.5, np.nan])[::step]
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method, expected_data, min_periods",
+    [
+        ("count", {0: Series([1, 2, 2, 2, 2]), 1: Series([1, 2, 2, 2, 2])}, 0),
+        (
+            "max",
+            {0: Series([np.nan, 2, 4, 6, 8]), 1: Series([np.nan, 3, 5, 7, 9])},
+            None,
+        ),
+        (
+            "min",
+            {0: Series([np.nan, 0, 2, 4, 6]), 1: Series([np.nan, 1, 3, 5, 7])},
+            None,
+        ),
+        (
+            "sum",
+            {0: Series([np.nan, 2, 6, 10, 14]), 1: Series([np.nan, 4, 8, 12, 16])},
+            None,
+        ),
+        (
+            "mean",
+            {0: Series([np.nan, 1, 3, 5, 7]), 1: Series([np.nan, 2, 4, 6, 8])},
+            None,
+        ),
+        (
+            "std",
+            {
+                0: Series([np.nan] + [np.sqrt(2)] * 4),
+                1: Series([np.nan] + [np.sqrt(2)] * 4),
+            },
+            None,
+        ),
+        (
+            "var",
+            {0: Series([np.nan, 2, 2, 2, 2]), 1: Series([np.nan, 2, 2, 2, 2])},
+            None,
+        ),
+        (
+            "median",
+            {0: Series([np.nan, 1, 3, 5, 7]), 1: Series([np.nan, 2, 4, 6, 8])},
+            None,
+        ),
+    ],
+)
+def test_dataframe_dtypes(method, expected_data, dtypes, min_periods, step):
+    df = DataFrame(np.arange(10).reshape((5, 2)), dtype=get_dtype(dtypes))
+    rolled = df.rolling(2, min_periods=min_periods, step=step)
+
+    if dtypes in ("m8[ns]", "M8[ns]", "datetime64[ns, UTC]") and method != "count":
+        msg = "Cannot aggregate non-numeric type"
+        with pytest.raises(DataError, match=msg):
+            getattr(rolled, method)()
+    else:
+        result = getattr(rolled, method)()
+        expected = DataFrame(expected_data, dtype="float64")[::step]
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_ewm.py b/pandas/pandas/tests/window/test_ewm.py
new file mode 100644
index 0000000000000..4ea6c805a2ee4
--- /dev/null
+++ b/pandas/pandas/tests/window/test_ewm.py
@@ -0,0 +1,737 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+
+
+def test_doc_string():
+    df = DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    df
+    df.ewm(com=0.5).mean()
+
+
+def test_constructor(frame_or_series):
+    c = frame_or_series(range(5)).ewm
+
+    # valid
+    c(com=0.5)
+    c(span=1.5)
+    c(alpha=0.5)
+    c(halflife=0.75)
+    c(com=0.5, span=None)
+    c(alpha=0.5, com=None)
+    c(halflife=0.75, alpha=None)
+
+    # not valid: mutually exclusive
+    msg = "comass, span, halflife, and alpha are mutually exclusive"
+    with pytest.raises(ValueError, match=msg):
+        c(com=0.5, alpha=0.5)
+    with pytest.raises(ValueError, match=msg):
+        c(span=1.5, halflife=0.75)
+    with pytest.raises(ValueError, match=msg):
+        c(alpha=0.5, span=1.5)
+
+    # not valid: com < 0
+    msg = "comass must satisfy: comass >= 0"
+    with pytest.raises(ValueError, match=msg):
+        c(com=-0.5)
+
+    # not valid: span < 1
+    msg = "span must satisfy: span >= 1"
+    with pytest.raises(ValueError, match=msg):
+        c(span=0.5)
+
+    # not valid: halflife <= 0
+    msg = "halflife must satisfy: halflife > 0"
+    with pytest.raises(ValueError, match=msg):
+        c(halflife=0)
+
+    # not valid: alpha <= 0 or alpha > 1
+    msg = "alpha must satisfy: 0 < alpha <= 1"
+    for alpha in (-0.5, 1.5):
+        with pytest.raises(ValueError, match=msg):
+            c(alpha=alpha)
+
+
+def test_ewma_times_not_datetime_type():
+    msg = r"times must be datetime64 dtype."
+    with pytest.raises(ValueError, match=msg):
+        Series(range(5)).ewm(times=np.arange(5))
+
+
+def test_ewma_times_not_same_length():
+    msg = "times must be the same length as the object."
+    with pytest.raises(ValueError, match=msg):
+        Series(range(5)).ewm(times=np.arange(4).astype("datetime64[ns]"))
+
+
+def test_ewma_halflife_not_correct_type():
+    msg = "halflife must be a timedelta convertible object"
+    with pytest.raises(ValueError, match=msg):
+        Series(range(5)).ewm(halflife=1, times=np.arange(5).astype("datetime64[ns]"))
+
+
+def test_ewma_halflife_without_times(halflife_with_times):
+    msg = "halflife can only be a timedelta convertible argument if times is not None."
+    with pytest.raises(ValueError, match=msg):
+        Series(range(5)).ewm(halflife=halflife_with_times)
+
+
+@pytest.mark.parametrize(
+    "times",
+    [
+        np.arange(10).astype("datetime64[D]").astype("datetime64[ns]"),
+        date_range("2000", freq="D", periods=10),
+        date_range("2000", freq="D", periods=10).tz_localize("UTC"),
+    ],
+)
+@pytest.mark.parametrize("min_periods", [0, 2])
+def test_ewma_with_times_equal_spacing(halflife_with_times, times, min_periods):
+    halflife = halflife_with_times
+    data = np.arange(10.0)
+    data[::2] = np.nan
+    df = DataFrame({"A": data})
+    result = df.ewm(halflife=halflife, min_periods=min_periods, times=times).mean()
+    expected = df.ewm(halflife=1.0, min_periods=min_periods).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ewma_with_times_variable_spacing(tz_aware_fixture, unit, adjust):
+    # GH 54328
+    tz = tz_aware_fixture
+    halflife = "23 days"
+    times = (
+        DatetimeIndex(["2020-01-01", "2020-01-10T00:04:05", "2020-02-23T05:00:23"])
+        .tz_localize(tz)
+        .as_unit(unit)
+    )
+    data = np.arange(3)
+    df = DataFrame(data)
+    result = df.ewm(halflife=halflife, times=times, adjust=adjust).mean()
+    if adjust:
+        expected = DataFrame([0.0, 0.5674161888241773, 1.545239952073459])
+    else:
+        expected = DataFrame([0.0, 0.23762518642226227, 1.534926369128742])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_ewm_with_nat_raises(halflife_with_times):
+    # GH#38535
+    ser = Series(range(1))
+    times = DatetimeIndex(["NaT"])
+    with pytest.raises(ValueError, match="Cannot convert NaT values to integer"):
+        ser.ewm(com=0.1, halflife=halflife_with_times, times=times)
+
+
+def test_ewm_with_times_getitem(halflife_with_times):
+    # GH 40164
+    halflife = halflife_with_times
+    data = np.arange(10.0)
+    data[::2] = np.nan
+    times = date_range("2000", freq="D", periods=10)
+    df = DataFrame({"A": data, "B": data})
+    result = df.ewm(halflife=halflife, times=times)["A"].mean()
+    expected = df.ewm(halflife=1.0)["A"].mean()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("arg", ["com", "halflife", "span", "alpha"])
+def test_ewm_getitem_attributes_retained(arg, adjust, ignore_na):
+    # GH 40164
+    kwargs = {arg: 1, "adjust": adjust, "ignore_na": ignore_na}
+    ewm = DataFrame({"A": range(1), "B": range(1)}).ewm(**kwargs)
+    expected = {attr: getattr(ewm, attr) for attr in ewm._attributes}
+    ewm_slice = ewm["A"]
+    result = {attr: getattr(ewm, attr) for attr in ewm_slice._attributes}
+    assert result == expected
+
+
+def test_ewma_times_adjust_false_with_disallowed_com():
+    # GH 54328
+    with pytest.raises(
+        NotImplementedError,
+        match=(
+            "None of com, span, or alpha can be specified "
+            "if times is provided and adjust=False"
+        ),
+    ):
+        Series(range(1)).ewm(
+            0.1,
+            adjust=False,
+            times=date_range("2000", freq="D", periods=1),
+            halflife="1D",
+        )
+
+
+def test_ewma_times_adjust_false_with_disallowed_alpha():
+    # GH 54328
+    with pytest.raises(
+        NotImplementedError,
+        match=(
+            "None of com, span, or alpha can be specified "
+            "if times is provided and adjust=False"
+        ),
+    ):
+        Series(range(1)).ewm(
+            0.1,
+            adjust=False,
+            times=date_range("2000", freq="D", periods=1),
+            alpha=0.5,
+            halflife="1D",
+        )
+
+
+def test_ewma_times_adjust_false_with_disallowed_span():
+    # GH 54328
+    with pytest.raises(
+        NotImplementedError,
+        match=(
+            "None of com, span, or alpha can be specified "
+            "if times is provided and adjust=False"
+        ),
+    ):
+        Series(range(1)).ewm(
+            0.1,
+            adjust=False,
+            times=date_range("2000", freq="D", periods=1),
+            span=10,
+            halflife="1D",
+        )
+
+
+def test_times_string_col_raises():
+    # GH 43265
+    df = DataFrame(
+        {"A": np.arange(10.0), "time_col": date_range("2000", freq="D", periods=10)}
+    )
+    with pytest.raises(ValueError, match="times must be datetime64"):
+        df.ewm(halflife="1 day", min_periods=0, times="time_col")
+
+
+def test_ewm_sum_adjust_false_notimplemented():
+    data = Series(range(1)).ewm(com=1, adjust=False)
+    with pytest.raises(NotImplementedError, match="sum is not"):
+        data.sum()
+
+
+@pytest.mark.parametrize("method", ["sum", "std", "var", "cov", "corr"])
+def test_times_only_mean_implemented(frame_or_series, method):
+    # GH 51695
+    halflife = "1 day"
+    times = date_range("2000", freq="D", periods=10)
+    ewm = frame_or_series(range(10)).ewm(halflife=halflife, times=times)
+    with pytest.raises(
+        NotImplementedError, match=f"{method} is not implemented with times"
+    ):
+        getattr(ewm, method)()
+
+
+@pytest.mark.parametrize(
+    "expected_data, ignore",
+    [[[10.0, 5.0, 2.5, 11.25], False], [[10.0, 5.0, 5.0, 12.5], True]],
+)
+def test_ewm_sum(expected_data, ignore):
+    # xref from Numbagg tests
+    # https://github.com/numbagg/numbagg/blob/v0.2.1/numbagg/test/test_moving.py#L50
+    data = Series([10, 0, np.nan, 10])
+    result = data.ewm(alpha=0.5, ignore_na=ignore).sum()
+    expected = Series(expected_data)
+    tm.assert_series_equal(result, expected)
+
+
+def test_ewma_adjust():
+    vals = Series(np.zeros(1000))
+    vals[5] = 1
+    result = vals.ewm(span=100, adjust=False).mean().sum()
+    assert np.abs(result - 1) < 1e-2
+
+
+def test_ewma_cases(adjust, ignore_na):
+    # try adjust/ignore_na args matrix
+
+    s = Series([1.0, 2.0, 4.0, 8.0])
+
+    if adjust:
+        expected = Series([1.0, 1.6, 2.736842, 4.923077])
+    else:
+        expected = Series([1.0, 1.333333, 2.222222, 4.148148])
+
+    result = s.ewm(com=2.0, adjust=adjust, ignore_na=ignore_na).mean()
+    tm.assert_series_equal(result, expected)
+
+
+def test_ewma_nan_handling():
+    s = Series([1.0] + [np.nan] * 5 + [1.0])
+    result = s.ewm(com=5).mean()
+    tm.assert_series_equal(result, Series([1.0] * len(s)))
+
+    s = Series([np.nan] * 2 + [1.0] + [np.nan] * 2 + [1.0])
+    result = s.ewm(com=5).mean()
+    tm.assert_series_equal(result, Series([np.nan] * 2 + [1.0] * 4))
+
+
+@pytest.mark.parametrize(
+    "s, adjust, ignore_na, w",
+    [
+        (
+            [np.nan, 1.0, 101.0],
+            True,
+            False,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))), 1.0],
+        ),
+        (
+            [np.nan, 1.0, 101.0],
+            True,
+            True,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))), 1.0],
+        ),
+        (
+            [np.nan, 1.0, 101.0],
+            False,
+            False,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))), (1.0 / (1.0 + 2.0))],
+        ),
+        (
+            [np.nan, 1.0, 101.0],
+            False,
+            True,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))), (1.0 / (1.0 + 2.0))],
+        ),
+        (
+            [1.0, np.nan, 101.0],
+            True,
+            False,
+            [(1.0 - (1.0 / (1.0 + 2.0))) ** 2, np.nan, 1.0],
+        ),
+        (
+            [1.0, np.nan, 101.0],
+            True,
+            True,
+            [(1.0 - (1.0 / (1.0 + 2.0))), np.nan, 1.0],
+        ),
+        (
+            [1.0, np.nan, 101.0],
+            False,
+            False,
+            [(1.0 - (1.0 / (1.0 + 2.0))) ** 2, np.nan, (1.0 / (1.0 + 2.0))],
+        ),
+        (
+            [1.0, np.nan, 101.0],
+            False,
+            True,
+            [(1.0 - (1.0 / (1.0 + 2.0))), np.nan, (1.0 / (1.0 + 2.0))],
+        ),
+        (
+            [np.nan, 1.0, np.nan, np.nan, 101.0, np.nan],
+            True,
+            False,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))) ** 3, np.nan, np.nan, 1.0, np.nan],
+        ),
+        (
+            [np.nan, 1.0, np.nan, np.nan, 101.0, np.nan],
+            True,
+            True,
+            [np.nan, (1.0 - (1.0 / (1.0 + 2.0))), np.nan, np.nan, 1.0, np.nan],
+        ),
+        (
+            [np.nan, 1.0, np.nan, np.nan, 101.0, np.nan],
+            False,
+            False,
+            [
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))) ** 3,
+                np.nan,
+                np.nan,
+                (1.0 / (1.0 + 2.0)),
+                np.nan,
+            ],
+        ),
+        (
+            [np.nan, 1.0, np.nan, np.nan, 101.0, np.nan],
+            False,
+            True,
+            [
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))),
+                np.nan,
+                np.nan,
+                (1.0 / (1.0 + 2.0)),
+                np.nan,
+            ],
+        ),
+        (
+            [1.0, np.nan, 101.0, 50.0],
+            True,
+            False,
+            [
+                (1.0 - (1.0 / (1.0 + 2.0))) ** 3,
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))),
+                1.0,
+            ],
+        ),
+        (
+            [1.0, np.nan, 101.0, 50.0],
+            True,
+            True,
+            [
+                (1.0 - (1.0 / (1.0 + 2.0))) ** 2,
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))),
+                1.0,
+            ],
+        ),
+        (
+            [1.0, np.nan, 101.0, 50.0],
+            False,
+            False,
+            [
+                (1.0 - (1.0 / (1.0 + 2.0))) ** 3,
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))) * (1.0 / (1.0 + 2.0)),
+                (1.0 / (1.0 + 2.0))
+                * ((1.0 - (1.0 / (1.0 + 2.0))) ** 2 + (1.0 / (1.0 + 2.0))),
+            ],
+        ),
+        (
+            [1.0, np.nan, 101.0, 50.0],
+            False,
+            True,
+            [
+                (1.0 - (1.0 / (1.0 + 2.0))) ** 2,
+                np.nan,
+                (1.0 - (1.0 / (1.0 + 2.0))) * (1.0 / (1.0 + 2.0)),
+                (1.0 / (1.0 + 2.0)),
+            ],
+        ),
+    ],
+)
+def test_ewma_nan_handling_cases(s, adjust, ignore_na, w):
+    # GH 7603
+    s = Series(s)
+    expected = (s.multiply(w).cumsum() / Series(w).cumsum()).ffill()
+    result = s.ewm(com=2.0, adjust=adjust, ignore_na=ignore_na).mean()
+
+    tm.assert_series_equal(result, expected)
+    if ignore_na is False:
+        # check that ignore_na defaults to False
+        result = s.ewm(com=2.0, adjust=adjust).mean()
+        tm.assert_series_equal(result, expected)
+
+
+def test_ewm_alpha():
+    # GH 10789
+    arr = np.random.default_rng(2).standard_normal(100)
+    locs = np.arange(20, 40)
+    arr[locs] = np.nan
+
+    s = Series(arr)
+    a = s.ewm(alpha=0.61722699889169674).mean()
+    b = s.ewm(com=0.62014947789973052).mean()
+    c = s.ewm(span=2.240298955799461).mean()
+    d = s.ewm(halflife=0.721792864318).mean()
+    tm.assert_series_equal(a, b)
+    tm.assert_series_equal(a, c)
+    tm.assert_series_equal(a, d)
+
+
+def test_ewm_domain_checks():
+    # GH 12492
+    arr = np.random.default_rng(2).standard_normal(100)
+    locs = np.arange(20, 40)
+    arr[locs] = np.nan
+
+    s = Series(arr)
+    msg = "comass must satisfy: comass >= 0"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(com=-0.1)
+    s.ewm(com=0.0)
+    s.ewm(com=0.1)
+
+    msg = "span must satisfy: span >= 1"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(span=-0.1)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(span=0.0)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(span=0.9)
+    s.ewm(span=1.0)
+    s.ewm(span=1.1)
+
+    msg = "halflife must satisfy: halflife > 0"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(halflife=-0.1)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(halflife=0.0)
+    s.ewm(halflife=0.1)
+
+    msg = "alpha must satisfy: 0 < alpha <= 1"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(alpha=-0.1)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(alpha=0.0)
+    s.ewm(alpha=0.1)
+    s.ewm(alpha=1.0)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(alpha=1.1)
+
+
+@pytest.mark.parametrize("method", ["mean", "std", "var"])
+def test_ew_empty_series(method):
+    vals = Series([], dtype=np.float64)
+
+    ewm = vals.ewm(3)
+    result = getattr(ewm, method)()
+    tm.assert_almost_equal(result, vals)
+
+
+@pytest.mark.parametrize("min_periods", [0, 1])
+@pytest.mark.parametrize("name", ["mean", "var", "std"])
+def test_ew_min_periods(min_periods, name):
+    # excluding NaNs correctly
+    arr = np.random.default_rng(2).standard_normal(50)
+    arr[:10] = np.nan
+    arr[-10:] = np.nan
+    s = Series(arr)
+
+    # check min_periods
+    # GH 7898
+    result = getattr(s.ewm(com=50, min_periods=2), name)()
+    assert result[:11].isna().all()
+    assert not result[11:].isna().any()
+
+    result = getattr(s.ewm(com=50, min_periods=min_periods), name)()
+    if name == "mean":
+        assert result[:10].isna().all()
+        assert not result[10:].isna().any()
+    else:
+        # ewm.std, ewm.var (with bias=False) require at least
+        # two values
+        assert result[:11].isna().all()
+        assert not result[11:].isna().any()
+
+    # check series of length 0
+    result = getattr(Series(dtype=object).ewm(com=50, min_periods=min_periods), name)()
+    tm.assert_series_equal(result, Series(dtype="float64"))
+
+    # check series of length 1
+    result = getattr(Series([1.0]).ewm(50, min_periods=min_periods), name)()
+    if name == "mean":
+        tm.assert_series_equal(result, Series([1.0]))
+    else:
+        # ewm.std, ewm.var with bias=False require at least
+        # two values
+        tm.assert_series_equal(result, Series([np.nan]))
+
+    # pass in ints
+    result2 = getattr(Series(np.arange(50)).ewm(span=10), name)()
+    assert result2.dtype == np.float64
+
+
+@pytest.mark.parametrize("name", ["cov", "corr"])
+def test_ewm_corr_cov(name):
+    A = Series(np.random.default_rng(2).standard_normal(50), index=range(50))
+    B = A[2:] + np.random.default_rng(2).standard_normal(48)
+
+    A[:10] = np.nan
+    B.iloc[-10:] = np.nan
+
+    result = getattr(A.ewm(com=20, min_periods=5), name)(B)
+    assert np.isnan(result.values[:14]).all()
+    assert not np.isnan(result.values[14:]).any()
+
+
+@pytest.mark.parametrize("min_periods", [0, 1, 2])
+@pytest.mark.parametrize("name", ["cov", "corr"])
+def test_ewm_corr_cov_min_periods(name, min_periods):
+    # GH 7898
+    A = Series(np.random.default_rng(2).standard_normal(50), index=range(50))
+    B = A[2:] + np.random.default_rng(2).standard_normal(48)
+
+    A[:10] = np.nan
+    B.iloc[-10:] = np.nan
+
+    result = getattr(A.ewm(com=20, min_periods=min_periods), name)(B)
+    # binary functions (ewmcov, ewmcorr) with bias=False require at
+    # least two values
+    assert np.isnan(result.values[:11]).all()
+    assert not np.isnan(result.values[11:]).any()
+
+    # check series of length 0
+    empty = Series([], dtype=np.float64)
+    result = getattr(empty.ewm(com=50, min_periods=min_periods), name)(empty)
+    tm.assert_series_equal(result, empty)
+
+    # check series of length 1
+    result = getattr(Series([1.0]).ewm(com=50, min_periods=min_periods), name)(
+        Series([1.0])
+    )
+    tm.assert_series_equal(result, Series([np.nan]))
+
+
+@pytest.mark.parametrize("name", ["cov", "corr"])
+def test_different_input_array_raise_exception(name):
+    A = Series(np.random.default_rng(2).standard_normal(50), index=range(50))
+    A[:10] = np.nan
+
+    msg = "other must be a DataFrame or Series"
+    # exception raised is Exception
+    with pytest.raises(ValueError, match=msg):
+        getattr(A.ewm(com=20, min_periods=5), name)(
+            np.random.default_rng(2).standard_normal(50)
+        )
+
+
+@pytest.mark.parametrize("name", ["var", "std", "mean"])
+def test_ewma_series(series, name):
+    series_result = getattr(series.ewm(com=10), name)()
+    assert isinstance(series_result, Series)
+
+
+@pytest.mark.parametrize("name", ["var", "std", "mean"])
+def test_ewma_frame(frame, name):
+    frame_result = getattr(frame.ewm(com=10), name)()
+    assert isinstance(frame_result, DataFrame)
+
+
+def test_ewma_span_com_args(series):
+    A = series.ewm(com=9.5).mean()
+    B = series.ewm(span=20).mean()
+    tm.assert_almost_equal(A, B)
+    msg = "comass, span, halflife, and alpha are mutually exclusive"
+    with pytest.raises(ValueError, match=msg):
+        series.ewm(com=9.5, span=20)
+
+    msg = "Must pass one of comass, span, halflife, or alpha"
+    with pytest.raises(ValueError, match=msg):
+        series.ewm().mean()
+
+
+def test_ewma_halflife_arg(series):
+    A = series.ewm(com=13.932726172912965).mean()
+    B = series.ewm(halflife=10.0).mean()
+    tm.assert_almost_equal(A, B)
+    msg = "comass, span, halflife, and alpha are mutually exclusive"
+    with pytest.raises(ValueError, match=msg):
+        series.ewm(span=20, halflife=50)
+    with pytest.raises(ValueError, match=msg):
+        series.ewm(com=9.5, halflife=50)
+    with pytest.raises(ValueError, match=msg):
+        series.ewm(com=9.5, span=20, halflife=50)
+    msg = "Must pass one of comass, span, halflife, or alpha"
+    with pytest.raises(ValueError, match=msg):
+        series.ewm()
+
+
+def test_ewm_alpha_arg(series):
+    # GH 10789
+    s = series
+    msg = "Must pass one of comass, span, halflife, or alpha"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm()
+
+    msg = "comass, span, halflife, and alpha are mutually exclusive"
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(com=10.0, alpha=0.5)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(span=10.0, alpha=0.5)
+    with pytest.raises(ValueError, match=msg):
+        s.ewm(halflife=10.0, alpha=0.5)
+
+
+@pytest.mark.parametrize("func", ["cov", "corr"])
+def test_ewm_pairwise_cov_corr(func, frame):
+    result = getattr(frame.ewm(span=10, min_periods=5), func)()
+    result = result.loc[(slice(None), 1), 5]
+    result.index = result.index.droplevel(1)
+    expected = getattr(frame[1].ewm(span=10, min_periods=5), func)(frame[5])
+    tm.assert_series_equal(result, expected, check_names=False)
+
+
+def test_numeric_only_frame(arithmetic_win_operators, numeric_only):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    df = DataFrame({"a": [1], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    ewm = df.ewm(span=2, min_periods=1)
+    op = getattr(ewm, kernel, None)
+    if op is not None:
+        result = op(numeric_only=numeric_only)
+
+        columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+        expected = df[columns].agg([kernel]).reset_index(drop=True).astype(float)
+        assert list(expected.columns) == columns
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+def test_numeric_only_corr_cov_frame(kernel, numeric_only, use_arg):
+    # GH#46560
+    df = DataFrame({"a": [1, 2, 3], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    arg = (df,) if use_arg else ()
+    ewm = df.ewm(span=2, min_periods=1)
+    op = getattr(ewm, kernel)
+    result = op(*arg, numeric_only=numeric_only)
+
+    # Compare result to op using float dtypes, dropping c when numeric_only is True
+    columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+    df2 = df[columns].astype(float)
+    arg2 = (df2,) if use_arg else ()
+    ewm2 = df2.ewm(span=2, min_periods=1)
+    op2 = getattr(ewm2, kernel)
+    expected = op2(*arg2, numeric_only=numeric_only)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_series(arithmetic_win_operators, numeric_only, dtype):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    ser = Series([1], dtype=dtype)
+    ewm = ser.ewm(span=2, min_periods=1)
+    op = getattr(ewm, kernel, None)
+    if op is None:
+        # Nothing to test
+        pytest.skip("No op to test")
+    if numeric_only and dtype is object:
+        msg = f"ExponentialMovingWindow.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(numeric_only=numeric_only)
+    else:
+        result = op(numeric_only=numeric_only)
+        expected = ser.agg([kernel]).reset_index(drop=True).astype(float)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_corr_cov_series(kernel, use_arg, numeric_only, dtype):
+    # GH#46560
+    ser = Series([1, 2, 3], dtype=dtype)
+    arg = (ser,) if use_arg else ()
+    ewm = ser.ewm(span=2, min_periods=1)
+    op = getattr(ewm, kernel)
+    if numeric_only and dtype is object:
+        msg = f"ExponentialMovingWindow.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(*arg, numeric_only=numeric_only)
+    else:
+        result = op(*arg, numeric_only=numeric_only)
+
+        ser2 = ser.astype(float)
+        arg2 = (ser2,) if use_arg else ()
+        ewm2 = ser2.ewm(span=2, min_periods=1)
+        op2 = getattr(ewm2, kernel)
+        expected = op2(*arg2, numeric_only=numeric_only)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_expanding.py b/pandas/pandas/tests/window/test_expanding.py
new file mode 100644
index 0000000000000..39cedc3b692da
--- /dev/null
+++ b/pandas/pandas/tests/window/test_expanding.py
@@ -0,0 +1,793 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    Series,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+
+def test_doc_string():
+    df = DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    df
+    df.expanding(2).sum()
+
+
+def test_constructor(frame_or_series):
+    # GH 12669
+
+    c = frame_or_series(range(5)).expanding
+
+    # valid
+    c(min_periods=1)
+
+
+@pytest.mark.parametrize("w", [2.0, "foo", np.array([2])])
+def test_constructor_invalid(frame_or_series, w):
+    # not valid
+
+    c = frame_or_series(range(5)).expanding
+    msg = "min_periods must be an integer"
+    with pytest.raises(ValueError, match=msg):
+        c(min_periods=w)
+
+
+@pytest.mark.parametrize(
+    "expander",
+    [
+        1,
+        pytest.param(
+            "ls",
+            marks=pytest.mark.xfail(
+                reason="GH#16425 expanding with offset not supported"
+            ),
+        ),
+    ],
+)
+def test_empty_df_expanding(expander):
+    # GH 15819 Verifies that datetime and integer expanding windows can be
+    # applied to empty DataFrames
+
+    expected = DataFrame()
+    result = DataFrame().expanding(expander).sum()
+    tm.assert_frame_equal(result, expected)
+
+    # Verifies that datetime and integer expanding windows can be applied
+    # to empty DataFrames with datetime index
+    expected = DataFrame(index=DatetimeIndex([]))
+    result = DataFrame(index=DatetimeIndex([])).expanding(expander).sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_missing_minp_zero():
+    # https://github.com/pandas-dev/pandas/pull/18921
+    # minp=0
+    x = Series([np.nan])
+    result = x.expanding(min_periods=0).sum()
+    expected = Series([0.0])
+    tm.assert_series_equal(result, expected)
+
+    # minp=1
+    result = x.expanding(min_periods=1).sum()
+    expected = Series([np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_expanding():
+    # see gh-23372.
+    df = DataFrame(np.ones((10, 20)))
+
+    expected = DataFrame(
+        {i: [np.nan] * 2 + [float(j) for j in range(3, 11)] for i in range(20)}
+    )
+    result = df.expanding(3).sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_expanding_count_with_min_periods(frame_or_series):
+    # GH 26996
+    result = frame_or_series(range(5)).expanding(min_periods=3).count()
+    expected = frame_or_series([np.nan, np.nan, 3.0, 4.0, 5.0])
+    tm.assert_equal(result, expected)
+
+
+def test_expanding_count_default_min_periods_with_null_values(frame_or_series):
+    # GH 26996
+    values = [1, 2, 3, np.nan, 4, 5, 6]
+    expected_counts = [1.0, 2.0, 3.0, 3.0, 4.0, 5.0, 6.0]
+
+    result = frame_or_series(values).expanding().count()
+    expected = frame_or_series(expected_counts)
+    tm.assert_equal(result, expected)
+
+
+def test_expanding_count_with_min_periods_exceeding_series_length(frame_or_series):
+    # GH 25857
+    result = frame_or_series(range(5)).expanding(min_periods=6).count()
+    expected = frame_or_series([np.nan, np.nan, np.nan, np.nan, np.nan])
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df,expected,min_periods",
+    [
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [1, 2, 3], "B": [4, 5, 6]}, [0, 1, 2]),
+            ],
+            3,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [1, 2, 3], "B": [4, 5, 6]}, [0, 1, 2]),
+            ],
+            2,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [1, 2, 3], "B": [4, 5, 6]}, [0, 1, 2]),
+            ],
+            1,
+        ),
+        ({"A": [1], "B": [4]}, [], 2),
+        (None, [({}, [])], 1),
+        (
+            {"A": [1, np.nan, 3], "B": [np.nan, 5, 6]},
+            [
+                ({"A": [1.0], "B": [np.nan]}, [0]),
+                ({"A": [1, np.nan], "B": [np.nan, 5]}, [0, 1]),
+                ({"A": [1, np.nan, 3], "B": [np.nan, 5, 6]}, [0, 1, 2]),
+            ],
+            3,
+        ),
+        (
+            {"A": [1, np.nan, 3], "B": [np.nan, 5, 6]},
+            [
+                ({"A": [1.0], "B": [np.nan]}, [0]),
+                ({"A": [1, np.nan], "B": [np.nan, 5]}, [0, 1]),
+                ({"A": [1, np.nan, 3], "B": [np.nan, 5, 6]}, [0, 1, 2]),
+            ],
+            2,
+        ),
+        (
+            {"A": [1, np.nan, 3], "B": [np.nan, 5, 6]},
+            [
+                ({"A": [1.0], "B": [np.nan]}, [0]),
+                ({"A": [1, np.nan], "B": [np.nan, 5]}, [0, 1]),
+                ({"A": [1, np.nan, 3], "B": [np.nan, 5, 6]}, [0, 1, 2]),
+            ],
+            1,
+        ),
+    ],
+)
+def test_iter_expanding_dataframe(df, expected, min_periods):
+    # GH 11704
+    df = DataFrame(df)
+    expecteds = [DataFrame(values, index=index) for (values, index) in expected]
+
+    for expected, actual in zip(expecteds, df.expanding(min_periods)):
+        tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "ser,expected,min_periods",
+    [
+        (Series([1, 2, 3]), [([1], [0]), ([1, 2], [0, 1]), ([1, 2, 3], [0, 1, 2])], 3),
+        (Series([1, 2, 3]), [([1], [0]), ([1, 2], [0, 1]), ([1, 2, 3], [0, 1, 2])], 2),
+        (Series([1, 2, 3]), [([1], [0]), ([1, 2], [0, 1]), ([1, 2, 3], [0, 1, 2])], 1),
+        (Series([1, 2]), [([1], [0]), ([1, 2], [0, 1])], 2),
+        (Series([np.nan, 2]), [([np.nan], [0]), ([np.nan, 2], [0, 1])], 2),
+        (Series([], dtype="int64"), [], 2),
+    ],
+)
+def test_iter_expanding_series(ser, expected, min_periods):
+    # GH 11704
+    expecteds = [Series(values, index=index) for (values, index) in expected]
+
+    for expected, actual in zip(expecteds, ser.expanding(min_periods)):
+        tm.assert_series_equal(actual, expected)
+
+
+def test_center_invalid():
+    # GH 20647
+    df = DataFrame()
+    with pytest.raises(TypeError, match=".* got an unexpected keyword"):
+        df.expanding(center=True)
+
+
+def test_expanding_sem(frame_or_series):
+    # GH: 26476
+    obj = frame_or_series([0, 1, 2])
+    result = obj.expanding().sem()
+    if isinstance(result, DataFrame):
+        result = Series(result[0].values)
+    expected = Series([np.nan] + [0.707107] * 2)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["skew", "kurt"])
+def test_expanding_skew_kurt_numerical_stability(method):
+    # GH: 6929
+    s = Series(np.random.default_rng(2).random(10))
+    expected = getattr(s.expanding(3), method)()
+    s = s + 5000
+    result = getattr(s.expanding(3), method)()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("window", [1, 3, 10, 20])
+@pytest.mark.parametrize("method", ["min", "max", "average"])
+@pytest.mark.parametrize("pct", [True, False])
+@pytest.mark.parametrize("test_data", ["default", "duplicates", "nans"])
+def test_rank(window, method, pct, ascending, test_data):
+    length = 20
+    if test_data == "default":
+        ser = Series(data=np.random.default_rng(2).random(length))
+    elif test_data == "duplicates":
+        ser = Series(data=np.random.default_rng(2).choice(3, length))
+    elif test_data == "nans":
+        ser = Series(
+            data=np.random.default_rng(2).choice(
+                [1.0, 0.25, 0.75, np.nan, np.inf, -np.inf], length
+            )
+        )
+
+    expected = ser.expanding(window).apply(
+        lambda x: x.rank(method=method, pct=pct, ascending=ascending).iloc[-1]
+    )
+    result = ser.expanding(window).rank(method=method, pct=pct, ascending=ascending)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_expanding_corr(series):
+    A = series.dropna()
+    B = (A + np.random.default_rng(2).standard_normal(len(A)))[:-5]
+
+    result = A.expanding().corr(B)
+
+    rolling_result = A.rolling(window=len(A), min_periods=1).corr(B)
+
+    tm.assert_almost_equal(rolling_result, result)
+
+
+def test_expanding_count(series):
+    result = series.expanding(min_periods=0).count()
+    tm.assert_almost_equal(
+        result, series.rolling(window=len(series), min_periods=0).count()
+    )
+
+
+def test_expanding_quantile(series):
+    result = series.expanding().quantile(0.5)
+
+    rolling_result = series.rolling(window=len(series), min_periods=1).quantile(0.5)
+
+    tm.assert_almost_equal(result, rolling_result)
+
+
+def test_expanding_cov(series):
+    A = series
+    B = (A + np.random.default_rng(2).standard_normal(len(A)))[:-5]
+
+    result = A.expanding().cov(B)
+
+    rolling_result = A.rolling(window=len(A), min_periods=1).cov(B)
+
+    tm.assert_almost_equal(rolling_result, result)
+
+
+def test_expanding_cov_pairwise(frame):
+    result = frame.expanding().cov()
+
+    rolling_result = frame.rolling(window=len(frame), min_periods=1).cov()
+
+    tm.assert_frame_equal(result, rolling_result)
+
+
+def test_expanding_corr_pairwise(frame):
+    result = frame.expanding().corr()
+
+    rolling_result = frame.rolling(window=len(frame), min_periods=1).corr()
+    tm.assert_frame_equal(result, rolling_result)
+
+
+@pytest.mark.parametrize(
+    "func,static_comp",
+    [
+        ("sum", lambda x: np.sum(x, axis=0)),
+        ("mean", lambda x: np.mean(x, axis=0)),
+        ("max", lambda x: np.max(x, axis=0)),
+        ("min", lambda x: np.min(x, axis=0)),
+    ],
+    ids=["sum", "mean", "max", "min"],
+)
+def test_expanding_func(func, static_comp, frame_or_series):
+    data = frame_or_series(np.array(list(range(10)) + [np.nan] * 10))
+
+    obj = data.expanding(min_periods=1)
+    result = getattr(obj, func)()
+    assert isinstance(result, frame_or_series)
+
+    expected = static_comp(data[:11])
+    if frame_or_series is Series:
+        tm.assert_almost_equal(result[10], expected)
+    else:
+        tm.assert_series_equal(result.iloc[10], expected, check_names=False)
+
+
+@pytest.mark.parametrize(
+    "func,static_comp",
+    [("sum", np.sum), ("mean", np.mean), ("max", np.max), ("min", np.min)],
+    ids=["sum", "mean", "max", "min"],
+)
+def test_expanding_min_periods(func, static_comp):
+    ser = Series(np.random.default_rng(2).standard_normal(50))
+
+    result = getattr(ser.expanding(min_periods=30), func)()
+    assert result[:29].isna().all()
+    tm.assert_almost_equal(result.iloc[-1], static_comp(ser[:50]))
+
+    # min_periods is working correctly
+    result = getattr(ser.expanding(min_periods=15), func)()
+    assert isna(result.iloc[13])
+    assert notna(result.iloc[14])
+
+    ser2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = getattr(ser2.expanding(min_periods=5), func)()
+    assert isna(result[3])
+    assert notna(result[4])
+
+    # min_periods=0
+    result0 = getattr(ser.expanding(min_periods=0), func)()
+    result1 = getattr(ser.expanding(min_periods=1), func)()
+    tm.assert_almost_equal(result0, result1)
+
+    result = getattr(ser.expanding(min_periods=1), func)()
+    tm.assert_almost_equal(result.iloc[-1], static_comp(ser[:50]))
+
+
+def test_expanding_apply(engine_and_raw, frame_or_series):
+    engine, raw = engine_and_raw
+    data = frame_or_series(np.array(list(range(10)) + [np.nan] * 10))
+    result = data.expanding(min_periods=1).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    assert isinstance(result, frame_or_series)
+
+    if frame_or_series is Series:
+        tm.assert_almost_equal(result[9], np.mean(data[:11], axis=0))
+    else:
+        tm.assert_series_equal(
+            result.iloc[9], np.mean(data[:11], axis=0), check_names=False
+        )
+
+
+def test_expanding_min_periods_apply(engine_and_raw):
+    engine, raw = engine_and_raw
+    ser = Series(np.random.default_rng(2).standard_normal(50))
+
+    result = ser.expanding(min_periods=30).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    assert result[:29].isna().all()
+    tm.assert_almost_equal(result.iloc[-1], np.mean(ser[:50]))
+
+    # min_periods is working correctly
+    result = ser.expanding(min_periods=15).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    assert isna(result.iloc[13])
+    assert notna(result.iloc[14])
+
+    ser2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = ser2.expanding(min_periods=5).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    assert isna(result[3])
+    assert notna(result[4])
+
+    # min_periods=0
+    result0 = ser.expanding(min_periods=0).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    result1 = ser.expanding(min_periods=1).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    tm.assert_almost_equal(result0, result1)
+
+    result = ser.expanding(min_periods=1).apply(
+        lambda x: x.mean(), raw=raw, engine=engine
+    )
+    tm.assert_almost_equal(result.iloc[-1], np.mean(ser[:50]))
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: (x.expanding(min_periods=5).cov(x, pairwise=True)),
+        lambda x: (x.expanding(min_periods=5).corr(x, pairwise=True)),
+    ],
+)
+def test_moment_functions_zero_length_pairwise(f):
+    df1 = DataFrame()
+    df2 = DataFrame(columns=Index(["a"], name="foo"), index=Index([], name="bar"))
+    df2["a"] = df2["a"].astype("float64")
+
+    df1_expected = DataFrame(index=MultiIndex.from_product([df1.index, df1.columns]))
+    df2_expected = DataFrame(
+        index=MultiIndex.from_product([df2.index, df2.columns], names=["bar", "foo"]),
+        columns=Index(["a"], name="foo"),
+        dtype="float64",
+    )
+
+    df1_result = f(df1)
+    tm.assert_frame_equal(df1_result, df1_expected)
+
+    df2_result = f(df2)
+    tm.assert_frame_equal(df2_result, df2_expected)
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: x.expanding().count(),
+        lambda x: x.expanding(min_periods=5).cov(x, pairwise=False),
+        lambda x: x.expanding(min_periods=5).corr(x, pairwise=False),
+        lambda x: x.expanding(min_periods=5).max(),
+        lambda x: x.expanding(min_periods=5).min(),
+        lambda x: x.expanding(min_periods=5).first(),
+        lambda x: x.expanding(min_periods=5).last(),
+        lambda x: x.expanding(min_periods=5).sum(),
+        lambda x: x.expanding(min_periods=5).mean(),
+        lambda x: x.expanding(min_periods=5).std(),
+        lambda x: x.expanding(min_periods=5).var(),
+        lambda x: x.expanding(min_periods=5).skew(),
+        lambda x: x.expanding(min_periods=5).kurt(),
+        lambda x: x.expanding(min_periods=5).quantile(0.5),
+        lambda x: x.expanding(min_periods=5).median(),
+        lambda x: x.expanding(min_periods=5).apply(sum, raw=False),
+        lambda x: x.expanding(min_periods=5).apply(sum, raw=True),
+    ],
+)
+def test_moment_functions_zero_length(f):
+    # GH 8056
+    s = Series(dtype=np.float64)
+    s_expected = s
+    df1 = DataFrame()
+    df1_expected = df1
+    df2 = DataFrame(columns=["a"])
+    df2["a"] = df2["a"].astype("float64")
+    df2_expected = df2
+
+    s_result = f(s)
+    tm.assert_series_equal(s_result, s_expected)
+
+    df1_result = f(df1)
+    tm.assert_frame_equal(df1_result, df1_expected)
+
+    df2_result = f(df2)
+    tm.assert_frame_equal(df2_result, df2_expected)
+
+
+def test_expanding_apply_empty_series(engine_and_raw):
+    engine, raw = engine_and_raw
+    ser = Series([], dtype=np.float64)
+    tm.assert_series_equal(
+        ser, ser.expanding().apply(lambda x: x.mean(), raw=raw, engine=engine)
+    )
+
+
+def test_expanding_apply_min_periods_0(engine_and_raw):
+    # GH 8080
+    engine, raw = engine_and_raw
+    s = Series([None, None, None])
+    result = s.expanding(min_periods=0).apply(lambda x: len(x), raw=raw, engine=engine)
+    expected = Series([1.0, 2.0, 3.0])
+    tm.assert_series_equal(result, expected)
+
+
+def test_expanding_cov_diff_index():
+    # GH 7512
+    s1 = Series([1, 2, 3], index=range(3))
+    s2 = Series([1, 3], index=range(0, 4, 2))
+    result = s1.expanding().cov(s2)
+    expected = Series([None, None, 2.0])
+    tm.assert_series_equal(result, expected)
+
+    s2a = Series([1, None, 3], index=[0, 1, 2])
+    result = s1.expanding().cov(s2a)
+    tm.assert_series_equal(result, expected)
+
+    s1 = Series([7, 8, 10], index=[0, 1, 3])
+    s2 = Series([7, 9, 10], index=[0, 2, 3])
+    result = s1.expanding().cov(s2)
+    expected = Series([None, None, None, 4.5], index=list(range(4)))
+    tm.assert_series_equal(result, expected)
+
+
+def test_expanding_corr_diff_index():
+    # GH 7512
+    s1 = Series([1, 2, 3], index=range(3))
+    s2 = Series([1, 3], index=range(0, 4, 2))
+    result = s1.expanding().corr(s2)
+    expected = Series([None, None, 1.0])
+    tm.assert_series_equal(result, expected)
+
+    s2a = Series([1, None, 3], index=[0, 1, 2])
+    result = s1.expanding().corr(s2a)
+    tm.assert_series_equal(result, expected)
+
+    s1 = Series([7, 8, 10], index=[0, 1, 3])
+    s2 = Series([7, 9, 10], index=[0, 2, 3])
+    result = s1.expanding().corr(s2)
+    expected = Series([None, None, None, 1.0], index=list(range(4)))
+    tm.assert_series_equal(result, expected)
+
+
+def test_expanding_cov_pairwise_diff_length():
+    # GH 7512
+    df1 = DataFrame([[1, 5], [3, 2], [3, 9]], columns=Index(["A", "B"], name="foo"))
+    df1a = DataFrame(
+        [[1, 5], [3, 9]], index=[0, 2], columns=Index(["A", "B"], name="foo")
+    )
+    df2 = DataFrame(
+        [[5, 6], [None, None], [2, 1]], columns=Index(["X", "Y"], name="foo")
+    )
+    df2a = DataFrame(
+        [[5, 6], [2, 1]], index=[0, 2], columns=Index(["X", "Y"], name="foo")
+    )
+    # xref gh-15826
+    # .loc is not preserving the names
+    result1 = df1.expanding().cov(df2, pairwise=True).loc[2]
+    result2 = df1.expanding().cov(df2a, pairwise=True).loc[2]
+    result3 = df1a.expanding().cov(df2, pairwise=True).loc[2]
+    result4 = df1a.expanding().cov(df2a, pairwise=True).loc[2]
+    expected = DataFrame(
+        [[-3.0, -6.0], [-5.0, -10.0]],
+        columns=Index(["A", "B"], name="foo"),
+        index=Index(["X", "Y"], name="foo"),
+    )
+    tm.assert_frame_equal(result1, expected)
+    tm.assert_frame_equal(result2, expected)
+    tm.assert_frame_equal(result3, expected)
+    tm.assert_frame_equal(result4, expected)
+
+
+def test_expanding_corr_pairwise_diff_length():
+    # GH 7512
+    df1 = DataFrame(
+        [[1, 2], [3, 2], [3, 4]], columns=["A", "B"], index=Index(range(3), name="bar")
+    )
+    df1a = DataFrame(
+        [[1, 2], [3, 4]], index=Index([0, 2], name="bar"), columns=["A", "B"]
+    )
+    df2 = DataFrame(
+        [[5, 6], [None, None], [2, 1]],
+        columns=["X", "Y"],
+        index=Index(range(3), name="bar"),
+    )
+    df2a = DataFrame(
+        [[5, 6], [2, 1]], index=Index([0, 2], name="bar"), columns=["X", "Y"]
+    )
+    result1 = df1.expanding().corr(df2, pairwise=True).loc[2]
+    result2 = df1.expanding().corr(df2a, pairwise=True).loc[2]
+    result3 = df1a.expanding().corr(df2, pairwise=True).loc[2]
+    result4 = df1a.expanding().corr(df2a, pairwise=True).loc[2]
+    expected = DataFrame(
+        [[-1.0, -1.0], [-1.0, -1.0]], columns=["A", "B"], index=Index(["X", "Y"])
+    )
+    tm.assert_frame_equal(result1, expected)
+    tm.assert_frame_equal(result2, expected)
+    tm.assert_frame_equal(result3, expected)
+    tm.assert_frame_equal(result4, expected)
+
+
+@pytest.mark.parametrize(
+    "values,method,expected",
+    [
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "first",
+            [float("nan"), float("nan"), 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "first",
+            [
+                float("nan"),
+                float("nan"),
+                float("nan"),
+                float("nan"),
+                1.0,
+                1.0,
+                1.0,
+                1.0,
+                1.0,
+                1.0,
+            ],
+        ),
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "last",
+            [float("nan"), float("nan"), 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "last",
+            [
+                float("nan"),
+                float("nan"),
+                float("nan"),
+                float("nan"),
+                5.0,
+                5.0,
+                7.0,
+                7.0,
+                9.0,
+                9.0,
+            ],
+        ),
+    ],
+)
+def test_expanding_first_last(values, method, expected):
+    # GH#33155
+    x = Series(values)
+    result = getattr(x.expanding(3), method)()
+    expected = Series(expected)
+    tm.assert_almost_equal(result, expected)
+
+    x = DataFrame({"A": values})
+    result = getattr(x.expanding(3), method)()
+    expected = DataFrame({"A": expected})
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values,method,expected",
+    [
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "first",
+            [1.0] * 10,
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "first",
+            [1.0] * 10,
+        ),
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "last",
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "last",
+            [1.0, 1.0, 3.0, 3.0, 5.0, 5.0, 7.0, 7.0, 9.0, 9.0],
+        ),
+    ],
+)
+def test_expanding_first_last_no_minp(values, method, expected):
+    # GH#33155
+    x = Series(values)
+    result = getattr(x.expanding(min_periods=0), method)()
+    expected = Series(expected)
+    tm.assert_almost_equal(result, expected)
+
+    x = DataFrame({"A": values})
+    result = getattr(x.expanding(min_periods=0), method)()
+    expected = DataFrame({"A": expected})
+    tm.assert_almost_equal(result, expected)
+
+
+def test_expanding_apply_args_kwargs(engine_and_raw):
+    def mean_w_arg(x, const):
+        return np.mean(x) + const
+
+    engine, raw = engine_and_raw
+
+    df = DataFrame(np.random.default_rng(2).random((20, 3)))
+
+    expected = df.expanding().apply(np.mean, engine=engine, raw=raw) + 20.0
+
+    result = df.expanding().apply(mean_w_arg, engine=engine, raw=raw, args=(20,))
+    tm.assert_frame_equal(result, expected)
+
+    result = df.expanding().apply(mean_w_arg, raw=raw, kwargs={"const": 20})
+    tm.assert_frame_equal(result, expected)
+
+
+def test_numeric_only_frame(arithmetic_win_operators, numeric_only):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    df = DataFrame({"a": [1], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    expanding = df.expanding()
+    op = getattr(expanding, kernel, None)
+    if op is not None:
+        result = op(numeric_only=numeric_only)
+
+        columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+        expected = df[columns].agg([kernel]).reset_index(drop=True).astype(float)
+        assert list(expected.columns) == columns
+
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+def test_numeric_only_corr_cov_frame(kernel, numeric_only, use_arg):
+    # GH#46560
+    df = DataFrame({"a": [1, 2, 3], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    arg = (df,) if use_arg else ()
+    expanding = df.expanding()
+    op = getattr(expanding, kernel)
+    result = op(*arg, numeric_only=numeric_only)
+
+    # Compare result to op using float dtypes, dropping c when numeric_only is True
+    columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+    df2 = df[columns].astype(float)
+    arg2 = (df2,) if use_arg else ()
+    expanding2 = df2.expanding()
+    op2 = getattr(expanding2, kernel)
+    expected = op2(*arg2, numeric_only=numeric_only)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_series(arithmetic_win_operators, numeric_only, dtype):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    ser = Series([1], dtype=dtype)
+    expanding = ser.expanding()
+    op = getattr(expanding, kernel)
+    if numeric_only and dtype is object:
+        msg = f"Expanding.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(numeric_only=numeric_only)
+    else:
+        result = op(numeric_only=numeric_only)
+        expected = ser.agg([kernel]).reset_index(drop=True).astype(float)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_corr_cov_series(kernel, use_arg, numeric_only, dtype):
+    # GH#46560
+    ser = Series([1, 2, 3], dtype=dtype)
+    arg = (ser,) if use_arg else ()
+    expanding = ser.expanding()
+    op = getattr(expanding, kernel)
+    if numeric_only and dtype is object:
+        msg = f"Expanding.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(*arg, numeric_only=numeric_only)
+    else:
+        result = op(*arg, numeric_only=numeric_only)
+
+        ser2 = ser.astype(float)
+        arg2 = (ser2,) if use_arg else ()
+        expanding2 = ser2.expanding()
+        op2 = getattr(expanding2, kernel)
+        expected = op2(*arg2, numeric_only=numeric_only)
+        tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_groupby.py b/pandas/pandas/tests/window/test_groupby.py
new file mode 100644
index 0000000000000..392239b8adadd
--- /dev/null
+++ b/pandas/pandas/tests/window/test_groupby.py
@@ -0,0 +1,1367 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    NamedAgg,
+    Series,
+    Timestamp,
+    date_range,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.api.indexers import BaseIndexer
+from pandas.core.groupby.groupby import get_groupby
+
+
+@pytest.fixture
+def times_frame():
+    """Frame for testing times argument in EWM groupby."""
+    return DataFrame(
+        {
+            "A": ["a", "b", "c", "a", "b", "c", "a", "b", "c", "a"],
+            "B": [0, 0, 0, 1, 1, 1, 2, 2, 2, 3],
+            "C": to_datetime(
+                [
+                    "2020-01-01",
+                    "2020-01-01",
+                    "2020-01-01",
+                    "2020-01-02",
+                    "2020-01-10",
+                    "2020-01-22",
+                    "2020-01-03",
+                    "2020-01-23",
+                    "2020-01-23",
+                    "2020-01-04",
+                ]
+            ),
+        }
+    )
+
+
+@pytest.fixture
+def roll_frame():
+    return DataFrame({"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)})
+
+
+class TestRolling:
+    def test_groupby_unsupported_argument(self, roll_frame):
+        msg = r"groupby\(\) got an unexpected keyword argument 'foo'"
+        with pytest.raises(TypeError, match=msg):
+            roll_frame.groupby("A", foo=1)
+
+    def test_getitem(self, roll_frame):
+        g = roll_frame.groupby("A")
+        g_mutated = get_groupby(roll_frame, by="A")
+
+        expected = g_mutated.B.apply(lambda x: x.rolling(2).mean())
+
+        result = g.rolling(2).mean().B
+        tm.assert_series_equal(result, expected)
+
+        result = g.rolling(2).B.mean()
+        tm.assert_series_equal(result, expected)
+
+        result = g.B.rolling(2).mean()
+        tm.assert_series_equal(result, expected)
+
+        result = roll_frame.B.groupby(roll_frame.A).rolling(2).mean()
+        tm.assert_series_equal(result, expected)
+
+    def test_getitem_multiple(self, roll_frame):
+        # GH 13174
+        g = roll_frame.groupby("A")
+        r = g.rolling(2, min_periods=0)
+        g_mutated = get_groupby(roll_frame, by="A")
+        expected = g_mutated.B.apply(lambda x: x.rolling(2, min_periods=0).count())
+
+        result = r.B.count()
+        tm.assert_series_equal(result, expected)
+
+        result = r.B.count()
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            "sum",
+            "mean",
+            "min",
+            "max",
+            "first",
+            "last",
+            "count",
+            "kurt",
+            "skew",
+        ],
+    )
+    def test_rolling(self, f, roll_frame):
+        g = roll_frame.groupby("A", group_keys=False)
+        r = g.rolling(window=4)
+
+        result = getattr(r, f)()
+        expected = g.apply(lambda x: getattr(x.rolling(4), f)())
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([roll_frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f", ["std", "var"])
+    def test_rolling_ddof(self, f, roll_frame):
+        g = roll_frame.groupby("A", group_keys=False)
+        r = g.rolling(window=4)
+
+        result = getattr(r, f)(ddof=1)
+        expected = g.apply(lambda x: getattr(x.rolling(4), f)(ddof=1))
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([roll_frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "interpolation", ["linear", "lower", "higher", "midpoint", "nearest"]
+    )
+    def test_rolling_quantile(self, interpolation, roll_frame):
+        g = roll_frame.groupby("A", group_keys=False)
+        r = g.rolling(window=4)
+
+        result = r.quantile(0.4, interpolation=interpolation)
+        expected = g.apply(
+            lambda x: x.rolling(4).quantile(0.4, interpolation=interpolation)
+        )
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([roll_frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f, expected_val", [["corr", 1], ["cov", 0.5]])
+    def test_rolling_corr_cov_other_same_size_as_groups(self, f, expected_val):
+        # GH 42915
+        df = DataFrame(
+            {"value": range(10), "idx1": [1] * 5 + [2] * 5, "idx2": [1, 2, 3, 4, 5] * 2}
+        ).set_index(["idx1", "idx2"])
+        other = DataFrame({"value": range(5), "idx2": [1, 2, 3, 4, 5]}).set_index(
+            "idx2"
+        )
+        result = getattr(df.groupby(level=0).rolling(2), f)(other)
+        expected_data = ([np.nan] + [expected_val] * 4) * 2
+        expected = DataFrame(
+            expected_data,
+            columns=["value"],
+            index=MultiIndex.from_arrays(
+                [
+                    [1] * 5 + [2] * 5,
+                    [1] * 5 + [2] * 5,
+                    list(range(1, 6)) * 2,
+                ],
+                names=["idx1", "idx1", "idx2"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f", ["corr", "cov"])
+    def test_rolling_corr_cov_other_diff_size_as_groups(self, f, roll_frame):
+        g = roll_frame.groupby("A")
+        r = g.rolling(window=4)
+
+        result = getattr(r, f)(roll_frame)
+
+        def func(x):
+            return getattr(x.rolling(4), f)(roll_frame)
+
+        expected = g.apply(func)
+        # GH 39591: The grouped column should be all np.nan
+        # (groupby.apply inserts 0s for cov)
+        expected["A"] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f", ["corr", "cov"])
+    def test_rolling_corr_cov_pairwise(self, f, roll_frame):
+        g = roll_frame.groupby("A")
+        r = g.rolling(window=4)
+
+        result = getattr(r.B, f)(pairwise=True)
+
+        def func(x):
+            return getattr(x.B.rolling(4), f)(pairwise=True)
+
+        expected = g.apply(func)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "func, expected_values",
+        [("cov", [[1.0, 1.0], [1.0, 4.0]]), ("corr", [[1.0, 0.5], [0.5, 1.0]])],
+    )
+    def test_rolling_corr_cov_unordered(self, func, expected_values):
+        # GH 43386
+        df = DataFrame(
+            {
+                "a": ["g1", "g2", "g1", "g1"],
+                "b": [0, 0, 1, 2],
+                "c": [2, 0, 6, 4],
+            }
+        )
+        rol = df.groupby("a").rolling(3)
+        result = getattr(rol, func)()
+        expected = DataFrame(
+            {
+                "b": 4 * [np.nan] + expected_values[0] + 2 * [np.nan],
+                "c": 4 * [np.nan] + expected_values[1] + 2 * [np.nan],
+            },
+            index=MultiIndex.from_tuples(
+                [
+                    ("g1", 0, "b"),
+                    ("g1", 0, "c"),
+                    ("g1", 2, "b"),
+                    ("g1", 2, "c"),
+                    ("g1", 3, "b"),
+                    ("g1", 3, "c"),
+                    ("g2", 1, "b"),
+                    ("g2", 1, "c"),
+                ],
+                names=["a", None, None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_rolling_apply(self, raw, roll_frame):
+        g = roll_frame.groupby("A", group_keys=False)
+        r = g.rolling(window=4)
+
+        # reduction
+        result = r.apply(lambda x: x.sum(), raw=raw)
+        expected = g.apply(lambda x: x.rolling(4).apply(lambda y: y.sum(), raw=raw))
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([roll_frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    def test_rolling_apply_mutability(self):
+        # GH 14013
+        df = DataFrame({"A": ["foo"] * 3 + ["bar"] * 3, "B": [1] * 6})
+        g = df.groupby("A")
+
+        mi = MultiIndex.from_tuples(
+            [("bar", 3), ("bar", 4), ("bar", 5), ("foo", 0), ("foo", 1), ("foo", 2)]
+        )
+
+        mi.names = ["A", None]
+        # Grouped column should not be a part of the output
+        expected = DataFrame([np.nan, 2.0, 2.0] * 2, columns=["B"], index=mi)
+
+        result = g.rolling(window=2).sum()
+        tm.assert_frame_equal(result, expected)
+
+        # Call an arbitrary function on the groupby
+        g.sum()
+
+        # Make sure nothing has been mutated
+        result = g.rolling(window=2).sum()
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("expected_value,raw_value", [[1.0, True], [0.0, False]])
+    def test_groupby_rolling(self, expected_value, raw_value):
+        # GH 31754
+
+        def isnumpyarray(x):
+            return int(isinstance(x, np.ndarray))
+
+        df = DataFrame({"id": [1, 1, 1], "value": [1, 2, 3]})
+        result = df.groupby("id").value.rolling(1).apply(isnumpyarray, raw=raw_value)
+        expected = Series(
+            [expected_value] * 3,
+            index=MultiIndex.from_tuples(((1, 0), (1, 1), (1, 2)), names=["id", None]),
+            name="value",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_center_center(self):
+        # GH 35552
+        series = Series(range(1, 6))
+        result = series.groupby(series).rolling(center=True, window=3).mean()
+        expected = Series(
+            [np.nan] * 5,
+            index=MultiIndex.from_tuples(((1, 0), (2, 1), (3, 2), (4, 3), (5, 4))),
+        )
+        tm.assert_series_equal(result, expected)
+
+        series = Series(range(1, 5))
+        result = series.groupby(series).rolling(center=True, window=3).mean()
+        expected = Series(
+            [np.nan] * 4,
+            index=MultiIndex.from_tuples(((1, 0), (2, 1), (3, 2), (4, 3))),
+        )
+        tm.assert_series_equal(result, expected)
+
+        df = DataFrame({"a": ["a"] * 5 + ["b"] * 6, "b": range(11)})
+        result = df.groupby("a").rolling(center=True, window=3).mean()
+        expected = DataFrame(
+            [np.nan, 1, 2, 3, np.nan, np.nan, 6, 7, 8, 9, np.nan],
+            index=MultiIndex.from_tuples(
+                (
+                    ("a", 0),
+                    ("a", 1),
+                    ("a", 2),
+                    ("a", 3),
+                    ("a", 4),
+                    ("b", 5),
+                    ("b", 6),
+                    ("b", 7),
+                    ("b", 8),
+                    ("b", 9),
+                    ("b", 10),
+                ),
+                names=["a", None],
+            ),
+            columns=["b"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({"a": ["a"] * 5 + ["b"] * 5, "b": range(10)})
+        result = df.groupby("a").rolling(center=True, window=3).mean()
+        expected = DataFrame(
+            [np.nan, 1, 2, 3, np.nan, np.nan, 6, 7, 8, np.nan],
+            index=MultiIndex.from_tuples(
+                (
+                    ("a", 0),
+                    ("a", 1),
+                    ("a", 2),
+                    ("a", 3),
+                    ("a", 4),
+                    ("b", 5),
+                    ("b", 6),
+                    ("b", 7),
+                    ("b", 8),
+                    ("b", 9),
+                ),
+                names=["a", None],
+            ),
+            columns=["b"],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_center_on(self):
+        # GH 37141
+        df = DataFrame(
+            data={
+                "Date": date_range("2020-01-01", "2020-01-10"),
+                "gb": ["group_1"] * 6 + ["group_2"] * 4,
+                "value": range(10),
+            }
+        )
+        result = (
+            df.groupby("gb")
+            .rolling(6, on="Date", center=True, min_periods=1)
+            .value.mean()
+        )
+        mi = MultiIndex.from_arrays([df["gb"], df["Date"]], names=["gb", "Date"])
+        expected = Series(
+            [1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 7.0, 7.5, 7.5, 7.5],
+            name="value",
+            index=mi,
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("min_periods", [5, 4, 3])
+    def test_groupby_rolling_center_min_periods(self, min_periods):
+        # GH 36040
+        df = DataFrame({"group": ["A"] * 10 + ["B"] * 10, "data": range(20)})
+
+        window_size = 5
+        result = (
+            df.groupby("group")
+            .rolling(window_size, center=True, min_periods=min_periods)
+            .mean()
+        )
+        result = result.reset_index()[["group", "data"]]
+
+        grp_A_mean = [1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 7.5, 8.0]
+        grp_B_mean = [x + 10.0 for x in grp_A_mean]
+
+        num_nans = max(0, min_periods - 3)  # For window_size of 5
+        nans = [np.nan] * num_nans
+        grp_A_expected = nans + grp_A_mean[num_nans : 10 - num_nans] + nans
+        grp_B_expected = nans + grp_B_mean[num_nans : 10 - num_nans] + nans
+
+        expected = DataFrame(
+            {"group": ["A"] * 10 + ["B"] * 10, "data": grp_A_expected + grp_B_expected}
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_subselect_rolling(self):
+        # GH 35486
+        df = DataFrame(
+            {"a": [1, 2, 3, 2], "b": [4.0, 2.0, 3.0, 1.0], "c": [10, 20, 30, 20]}
+        )
+        result = df.groupby("a")[["b"]].rolling(2).max()
+        expected = DataFrame(
+            [np.nan, np.nan, 2.0, np.nan],
+            columns=["b"],
+            index=MultiIndex.from_tuples(
+                ((1, 0), (2, 1), (2, 3), (3, 2)), names=["a", None]
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        result = df.groupby("a")["b"].rolling(2).max()
+        expected = Series(
+            [np.nan, np.nan, 2.0, np.nan],
+            index=MultiIndex.from_tuples(
+                ((1, 0), (2, 1), (2, 3), (3, 2)), names=["a", None]
+            ),
+            name="b",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_custom_indexer(self):
+        # GH 35557
+        class SimpleIndexer(BaseIndexer):
+            def get_window_bounds(
+                self,
+                num_values=0,
+                min_periods=None,
+                center=None,
+                closed=None,
+                step=None,
+            ):
+                min_periods = self.window_size if min_periods is None else 0
+                end = np.arange(num_values, dtype=np.int64) + 1
+                start = end - self.window_size
+                start[start < 0] = min_periods
+                return start, end
+
+        df = DataFrame(
+            {"a": [1.0, 2.0, 3.0, 4.0, 5.0] * 3}, index=[0] * 5 + [1] * 5 + [2] * 5
+        )
+        result = (
+            df.groupby(df.index)
+            .rolling(SimpleIndexer(window_size=3), min_periods=1)
+            .sum()
+        )
+        expected = df.groupby(df.index).rolling(window=3, min_periods=1).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_subset_with_closed(self):
+        # GH 35549
+        df = DataFrame(
+            {
+                "column1": range(8),
+                "column2": range(8),
+                "group": ["A"] * 4 + ["B"] * 4,
+                "date": [
+                    Timestamp(date)
+                    for date in ["2019-01-01", "2019-01-01", "2019-01-02", "2019-01-02"]
+                ]
+                * 2,
+            }
+        )
+        result = (
+            df.groupby("group").rolling("1D", on="date", closed="left")["column1"].sum()
+        )
+        expected = Series(
+            [np.nan, np.nan, 1.0, 1.0, np.nan, np.nan, 9.0, 9.0],
+            index=MultiIndex.from_frame(
+                df[["group", "date"]],
+                names=["group", "date"],
+            ),
+            name="column1",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_agg_namedagg(self):
+        # GH#28333
+        df = DataFrame(
+            {
+                "kind": ["cat", "dog", "cat", "dog", "cat", "dog"],
+                "height": [9.1, 6.0, 9.5, 34.0, 12.0, 8.0],
+                "weight": [7.9, 7.5, 9.9, 198.0, 10.0, 42.0],
+            }
+        )
+        result = (
+            df.groupby("kind")
+            .rolling(2)
+            .agg(
+                total_weight=NamedAgg(column="weight", aggfunc=sum),
+                min_height=NamedAgg(column="height", aggfunc=min),
+            )
+        )
+        expected = DataFrame(
+            {
+                "total_weight": [np.nan, 17.8, 19.9, np.nan, 205.5, 240.0],
+                "min_height": [np.nan, 9.1, 9.5, np.nan, 6.0, 8.0],
+            },
+            index=MultiIndex(
+                [["cat", "dog"], [0, 1, 2, 3, 4, 5]],
+                [[0, 0, 0, 1, 1, 1], [0, 2, 4, 1, 3, 5]],
+                names=["kind", None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_subset_rolling_subset_with_closed(self):
+        # GH 35549
+        df = DataFrame(
+            {
+                "column1": range(8),
+                "column2": range(8),
+                "group": ["A"] * 4 + ["B"] * 4,
+                "date": [
+                    Timestamp(date)
+                    for date in ["2019-01-01", "2019-01-01", "2019-01-02", "2019-01-02"]
+                ]
+                * 2,
+            }
+        )
+
+        result = (
+            df.groupby("group")[["column1", "date"]]
+            .rolling("1D", on="date", closed="left")["column1"]
+            .sum()
+        )
+        expected = Series(
+            [np.nan, np.nan, 1.0, 1.0, np.nan, np.nan, 9.0, 9.0],
+            index=MultiIndex.from_frame(
+                df[["group", "date"]],
+                names=["group", "date"],
+            ),
+            name="column1",
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize("func", ["max", "min"])
+    def test_groupby_rolling_index_changed(self, func):
+        # GH: #36018 nlevels of MultiIndex changed
+        ds = Series(
+            [1, 2, 2],
+            index=MultiIndex.from_tuples(
+                [("a", "x"), ("a", "y"), ("c", "z")], names=["1", "2"]
+            ),
+            name="a",
+        )
+
+        result = getattr(ds.groupby(ds).rolling(2), func)()
+        expected = Series(
+            [np.nan, np.nan, 2.0],
+            index=MultiIndex.from_tuples(
+                [(1, "a", "x"), (2, "a", "y"), (2, "c", "z")], names=["a", "1", "2"]
+            ),
+            name="a",
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_empty_frame(self):
+        # GH 36197
+        expected = DataFrame({"s1": []})
+        result = expected.groupby("s1").rolling(window=1).sum()
+        # GH 32262
+        expected = expected.drop(columns="s1")
+        # GH-38057 from_tuples gives empty object dtype, we now get float/int levels
+        # expected.index = MultiIndex.from_tuples([], names=["s1", None])
+        expected.index = MultiIndex.from_product(
+            [Index([], dtype="float64"), Index([], dtype="int64")], names=["s1", None]
+        )
+        tm.assert_frame_equal(result, expected)
+
+        expected = DataFrame({"s1": [], "s2": []})
+        result = expected.groupby(["s1", "s2"]).rolling(window=1).sum()
+        # GH 32262
+        expected = expected.drop(columns=["s1", "s2"])
+        expected.index = MultiIndex.from_product(
+            [
+                Index([], dtype="float64"),
+                Index([], dtype="float64"),
+                Index([], dtype="int64"),
+            ],
+            names=["s1", "s2", None],
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_string_index(self):
+        # GH: 36727
+        df = DataFrame(
+            [
+                ["A", "group_1", Timestamp(2019, 1, 1, 9)],
+                ["B", "group_1", Timestamp(2019, 1, 2, 9)],
+                ["Z", "group_2", Timestamp(2019, 1, 3, 9)],
+                ["H", "group_1", Timestamp(2019, 1, 6, 9)],
+                ["E", "group_2", Timestamp(2019, 1, 20, 9)],
+            ],
+            columns=["index", "group", "eventTime"],
+        ).set_index("index")
+
+        groups = df.groupby("group")
+        df["count_to_date"] = groups.cumcount()
+        rolling_groups = groups.rolling("10D", on="eventTime")
+        result = rolling_groups.apply(lambda df: df.shape[0])
+        expected = DataFrame(
+            [
+                ["A", "group_1", Timestamp(2019, 1, 1, 9), 1.0],
+                ["B", "group_1", Timestamp(2019, 1, 2, 9), 2.0],
+                ["H", "group_1", Timestamp(2019, 1, 6, 9), 3.0],
+                ["Z", "group_2", Timestamp(2019, 1, 3, 9), 1.0],
+                ["E", "group_2", Timestamp(2019, 1, 20, 9), 1.0],
+            ],
+            columns=["index", "group", "eventTime", "count_to_date"],
+        ).set_index(["group", "index"])
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_no_sort(self):
+        # GH 36889
+        result = (
+            DataFrame({"foo": [2, 1], "bar": [2, 1]})
+            .groupby("foo", sort=False)
+            .rolling(1)
+            .min()
+        )
+        expected = DataFrame(
+            np.array([[2.0, 2.0], [1.0, 1.0]]),
+            columns=["foo", "bar"],
+            index=MultiIndex.from_tuples([(2, 0), (1, 1)], names=["foo", None]),
+        )
+        # GH 32262
+        expected = expected.drop(columns="foo")
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_count_closed_on(self, unit):
+        # GH 35869
+        df = DataFrame(
+            {
+                "column1": range(6),
+                "column2": range(6),
+                "group": 3 * ["A", "B"],
+                "date": date_range(end="20190101", periods=6, unit=unit),
+            }
+        )
+        msg = "'d' is deprecated and will be removed in a future version."
+
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            result = (
+                df.groupby("group")
+                .rolling("3d", on="date", closed="left")["column1"]
+                .count()
+            )
+        dti = DatetimeIndex(
+            [
+                "2018-12-27",
+                "2018-12-29",
+                "2018-12-31",
+                "2018-12-28",
+                "2018-12-30",
+                "2019-01-01",
+            ],
+            dtype=f"M8[{unit}]",
+        )
+        mi = MultiIndex.from_arrays(
+            [
+                ["A", "A", "A", "B", "B", "B"],
+                dti,
+            ],
+            names=["group", "date"],
+        )
+        expected = Series(
+            [np.nan, 1.0, 1.0, np.nan, 1.0, 1.0],
+            name="column1",
+            index=mi,
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        ("func", "kwargs"),
+        [("rolling", {"window": 2, "min_periods": 1}), ("expanding", {})],
+    )
+    def test_groupby_rolling_sem(self, func, kwargs):
+        # GH: 26476
+        df = DataFrame(
+            [["a", 1], ["a", 2], ["b", 1], ["b", 2], ["b", 3]], columns=["a", "b"]
+        )
+        result = getattr(df.groupby("a"), func)(**kwargs).sem()
+        expected = DataFrame(
+            {"a": [np.nan] * 5, "b": [np.nan, 0.70711, np.nan, 0.70711, 0.70711]},
+            index=MultiIndex.from_tuples(
+                [("a", 0), ("a", 1), ("b", 2), ("b", 3), ("b", 4)], names=["a", None]
+            ),
+        )
+        # GH 32262
+        expected = expected.drop(columns="a")
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        ("rollings", "key"), [({"on": "a"}, "a"), ({"on": None}, "index")]
+    )
+    def test_groupby_rolling_nans_in_index(self, rollings, key):
+        # GH: 34617
+        df = DataFrame(
+            {
+                "a": to_datetime(["2020-06-01 12:00", "2020-06-01 14:00", np.nan]),
+                "b": [1, 2, 3],
+                "c": [1, 1, 1],
+            }
+        )
+        if key == "index":
+            df = df.set_index("a")
+        with pytest.raises(ValueError, match=f"{key} values must not have NaT"):
+            df.groupby("c").rolling("60min", **rollings)
+
+    @pytest.mark.parametrize("group_keys", [True, False])
+    def test_groupby_rolling_group_keys(self, group_keys):
+        # GH 37641
+        # GH 38523: GH 37641 actually was not a bug.
+        # group_keys only applies to groupby.apply directly
+        arrays = [["val1", "val1", "val2"], ["val1", "val1", "val2"]]
+        index = MultiIndex.from_arrays(arrays, names=("idx1", "idx2"))
+
+        s = Series([1, 2, 3], index=index)
+        result = s.groupby(["idx1", "idx2"], group_keys=group_keys).rolling(1).mean()
+        expected = Series(
+            [1.0, 2.0, 3.0],
+            index=MultiIndex.from_tuples(
+                [
+                    ("val1", "val1", "val1", "val1"),
+                    ("val1", "val1", "val1", "val1"),
+                    ("val2", "val2", "val2", "val2"),
+                ],
+                names=["idx1", "idx2", "idx1", "idx2"],
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_index_level_and_column_label(self):
+        # The groupby keys should not appear as a resulting column
+        arrays = [["val1", "val1", "val2"], ["val1", "val1", "val2"]]
+        index = MultiIndex.from_arrays(arrays, names=("idx1", "idx2"))
+
+        df = DataFrame({"A": [1, 1, 2], "B": range(3)}, index=index)
+        result = df.groupby(["idx1", "A"]).rolling(1).mean()
+        expected = DataFrame(
+            {"B": [0.0, 1.0, 2.0]},
+            index=MultiIndex.from_tuples(
+                [
+                    ("val1", 1, "val1", "val1"),
+                    ("val1", 1, "val1", "val1"),
+                    ("val2", 2, "val2", "val2"),
+                ],
+                names=["idx1", "A", "idx1", "idx2"],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_rolling_resulting_multiindex(self):
+        # a few different cases checking the created MultiIndex of the result
+        # https://github.com/pandas-dev/pandas/pull/38057
+
+        # grouping by 1 columns -> 2-level MI as result
+        df = DataFrame({"a": np.arange(8.0), "b": [1, 2] * 4})
+        result = df.groupby("b").rolling(3).mean()
+        expected_index = MultiIndex.from_tuples(
+            [(1, 0), (1, 2), (1, 4), (1, 6), (2, 1), (2, 3), (2, 5), (2, 7)],
+            names=["b", None],
+        )
+        tm.assert_index_equal(result.index, expected_index)
+
+    def test_groupby_rolling_resulting_multiindex2(self):
+        # grouping by 2 columns -> 3-level MI as result
+        df = DataFrame({"a": np.arange(12.0), "b": [1, 2] * 6, "c": [1, 2, 3, 4] * 3})
+        result = df.groupby(["b", "c"]).rolling(2).sum()
+        expected_index = MultiIndex.from_tuples(
+            [
+                (1, 1, 0),
+                (1, 1, 4),
+                (1, 1, 8),
+                (1, 3, 2),
+                (1, 3, 6),
+                (1, 3, 10),
+                (2, 2, 1),
+                (2, 2, 5),
+                (2, 2, 9),
+                (2, 4, 3),
+                (2, 4, 7),
+                (2, 4, 11),
+            ],
+            names=["b", "c", None],
+        )
+        tm.assert_index_equal(result.index, expected_index)
+
+    def test_groupby_rolling_resulting_multiindex3(self):
+        # grouping with 1 level on dataframe with 2-level MI -> 3-level MI as result
+        df = DataFrame({"a": np.arange(8.0), "b": [1, 2] * 4, "c": [1, 2, 3, 4] * 2})
+        df = df.set_index("c", append=True)
+        result = df.groupby("b").rolling(3).mean()
+        expected_index = MultiIndex.from_tuples(
+            [
+                (1, 0, 1),
+                (1, 2, 3),
+                (1, 4, 1),
+                (1, 6, 3),
+                (2, 1, 2),
+                (2, 3, 4),
+                (2, 5, 2),
+                (2, 7, 4),
+            ],
+            names=["b", None, "c"],
+        )
+        tm.assert_index_equal(result.index, expected_index, exact="equiv")
+
+    def test_groupby_rolling_object_doesnt_affect_groupby_apply(self, roll_frame):
+        # GH 39732
+        g = roll_frame.groupby("A", group_keys=False)
+        expected = g.apply(lambda x: x.rolling(4).sum()).index
+        _ = g.rolling(window=4)
+        result = g.apply(lambda x: x.rolling(4).sum()).index
+        tm.assert_index_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        ("window", "min_periods", "closed", "expected"),
+        [
+            (2, 0, "left", [None, 0.0, 1.0, 1.0, None, 0.0, 1.0, 1.0]),
+            (2, 2, "left", [None, None, 1.0, 1.0, None, None, 1.0, 1.0]),
+            (4, 4, "left", [None, None, None, None, None, None, None, None]),
+            (4, 4, "right", [None, None, None, 5.0, None, None, None, 5.0]),
+        ],
+    )
+    def test_groupby_rolling_var(self, window, min_periods, closed, expected):
+        df = DataFrame([1, 2, 3, 4, 5, 6, 7, 8])
+        result = (
+            df.groupby([1, 2, 1, 2, 1, 2, 1, 2])
+            .rolling(window=window, min_periods=min_periods, closed=closed)
+            .var(0)
+        )
+        expected_result = DataFrame(
+            np.array(expected, dtype="float64"),
+            index=MultiIndex(
+                levels=[np.array([1, 2]), [0, 1, 2, 3, 4, 5, 6, 7]],
+                codes=[[0, 0, 0, 0, 1, 1, 1, 1], [0, 2, 4, 6, 1, 3, 5, 7]],
+            ),
+        )
+        tm.assert_frame_equal(result, expected_result)
+
+    @pytest.mark.parametrize(
+        "columns", [MultiIndex.from_tuples([("A", ""), ("B", "C")]), ["A", "B"]]
+    )
+    def test_by_column_not_in_values(self, columns):
+        # GH 32262
+        df = DataFrame([[1, 0]] * 20 + [[2, 0]] * 12 + [[3, 0]] * 8, columns=columns)
+        g = df.groupby("A")
+        original_obj = g.obj.copy(deep=True)
+        r = g.rolling(4)
+        result = r.sum()
+        assert "A" not in result.columns
+        tm.assert_frame_equal(g.obj, original_obj)
+
+    def test_groupby_level(self):
+        # GH 38523, 38787
+        arrays = [
+            ["Falcon", "Falcon", "Parrot", "Parrot"],
+            ["Captive", "Wild", "Captive", "Wild"],
+        ]
+        index = MultiIndex.from_arrays(arrays, names=("Animal", "Type"))
+        df = DataFrame({"Max Speed": [390.0, 350.0, 30.0, 20.0]}, index=index)
+        result = df.groupby(level=0)["Max Speed"].rolling(2).sum()
+        expected = Series(
+            [np.nan, 740.0, np.nan, 50.0],
+            index=MultiIndex.from_tuples(
+                [
+                    ("Falcon", "Falcon", "Captive"),
+                    ("Falcon", "Falcon", "Wild"),
+                    ("Parrot", "Parrot", "Captive"),
+                    ("Parrot", "Parrot", "Wild"),
+                ],
+                names=["Animal", "Animal", "Type"],
+            ),
+            name="Max Speed",
+        )
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "by, expected_data",
+        [
+            [["id"], {"num": [100.0, 150.0, 150.0, 200.0]}],
+            [
+                ["id", "index"],
+                {
+                    "date": [
+                        Timestamp("2018-01-01"),
+                        Timestamp("2018-01-02"),
+                        Timestamp("2018-01-01"),
+                        Timestamp("2018-01-02"),
+                    ],
+                    "num": [100.0, 200.0, 150.0, 250.0],
+                },
+            ],
+        ],
+    )
+    def test_as_index_false(self, by, expected_data, unit):
+        # GH 39433
+        data = [
+            ["A", "2018-01-01", 100.0],
+            ["A", "2018-01-02", 200.0],
+            ["B", "2018-01-01", 150.0],
+            ["B", "2018-01-02", 250.0],
+        ]
+        df = DataFrame(data, columns=["id", "date", "num"])
+        df["date"] = df["date"].astype(f"M8[{unit}]")
+        df = df.set_index(["date"])
+
+        gp_by = [getattr(df, attr) for attr in by]
+        result = (
+            df.groupby(gp_by, as_index=False).rolling(window=2, min_periods=1).mean()
+        )
+
+        expected = {"id": ["A", "A", "B", "B"]}
+        expected.update(expected_data)
+        expected = DataFrame(
+            expected,
+            index=df.index,
+        )
+        if "date" in expected_data:
+            expected["date"] = expected["date"].astype(f"M8[{unit}]")
+        tm.assert_frame_equal(result, expected)
+
+    def test_nan_and_zero_endpoints(self, any_int_numpy_dtype):
+        # https://github.com/twosigma/pandas/issues/53
+        typ = np.dtype(any_int_numpy_dtype).type
+        size = 1000
+        idx = np.repeat(typ(0), size)
+        idx[-1] = 1
+
+        val = 5e25
+        arr = np.repeat(val, size)
+        arr[0] = np.nan
+        arr[-1] = 0
+
+        df = DataFrame(
+            {
+                "index": idx,
+                "adl2": arr,
+            }
+        ).set_index("index")
+        result = df.groupby("index")["adl2"].rolling(window=10, min_periods=1).mean()
+        expected = Series(
+            arr,
+            name="adl2",
+            index=MultiIndex.from_arrays(
+                [
+                    Index([0] * 999 + [1], dtype=typ, name="index"),
+                    Index([0] * 999 + [1], dtype=typ, name="index"),
+                ],
+            ),
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_groupby_rolling_non_monotonic(self):
+        # GH 43909
+
+        shuffled = [3, 0, 1, 2]
+        sec = 1_000
+        df = DataFrame(
+            [{"t": Timestamp(2 * x * sec), "x": x + 1, "c": 42} for x in shuffled]
+        )
+        with pytest.raises(ValueError, match=r".* must be monotonic"):
+            df.groupby("c").rolling(on="t", window="3s")
+
+    def test_groupby_monotonic(self):
+        # GH 15130
+        # we don't need to validate monotonicity when grouping
+
+        # GH 43909 we should raise an error here to match
+        # behaviour of non-groupby rolling.
+
+        data = [
+            ["David", "1/1/2015", 100],
+            ["David", "1/5/2015", 500],
+            ["David", "5/30/2015", 50],
+            ["David", "7/25/2015", 50],
+            ["Ryan", "1/4/2014", 100],
+            ["Ryan", "1/19/2015", 500],
+            ["Ryan", "3/31/2016", 50],
+            ["Joe", "7/1/2015", 100],
+            ["Joe", "9/9/2015", 500],
+            ["Joe", "10/15/2015", 50],
+        ]
+
+        df = DataFrame(data=data, columns=["name", "date", "amount"])
+        df["date"] = to_datetime(df["date"])
+        df = df.sort_values("date")
+
+        expected = (
+            df.set_index("date")
+            .groupby("name")
+            .apply(lambda x: x.rolling("180D")["amount"].sum())
+        )
+        result = df.groupby("name").rolling("180D", on="date")["amount"].sum()
+        tm.assert_series_equal(result, expected)
+
+    def test_datelike_on_monotonic_within_each_group(self):
+        # GH 13966 (similar to #15130, closed by #15175)
+
+        # superseded by 43909
+        # GH 46061: OK if the on is monotonic relative to each each group
+
+        dates = date_range(start="2016-01-01 09:30:00", periods=20, freq="s")
+        df = DataFrame(
+            {
+                "A": [1] * 20 + [2] * 12 + [3] * 8,
+                "B": np.concatenate((dates, dates)),
+                "C": np.arange(40),
+            }
+        )
+
+        expected = (
+            df.set_index("B").groupby("A").apply(lambda x: x.rolling("4s")["C"].mean())
+        )
+        result = df.groupby("A").rolling("4s", on="B").C.mean()
+        tm.assert_series_equal(result, expected)
+
+    def test_datelike_on_not_monotonic_within_each_group(self):
+        # GH 46061
+        df = DataFrame(
+            {
+                "A": [1] * 3 + [2] * 3,
+                "B": [Timestamp(year, 1, 1) for year in [2020, 2021, 2019]] * 2,
+                "C": range(6),
+            }
+        )
+        with pytest.raises(ValueError, match="Each group within B must be monotonic."):
+            df.groupby("A").rolling("365D", on="B")
+
+
+class TestExpanding:
+    @pytest.fixture
+    def frame(self):
+        return DataFrame({"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)})
+
+    @pytest.mark.parametrize(
+        "f", ["sum", "mean", "min", "max", "first", "last", "count", "kurt", "skew"]
+    )
+    def test_expanding(self, f, frame):
+        g = frame.groupby("A", group_keys=False)
+        r = g.expanding()
+
+        result = getattr(r, f)()
+        expected = g.apply(lambda x: getattr(x.expanding(), f)())
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f", ["std", "var"])
+    def test_expanding_ddof(self, f, frame):
+        g = frame.groupby("A", group_keys=False)
+        r = g.expanding()
+
+        result = getattr(r, f)(ddof=0)
+        expected = g.apply(lambda x: getattr(x.expanding(), f)(ddof=0))
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "interpolation", ["linear", "lower", "higher", "midpoint", "nearest"]
+    )
+    def test_expanding_quantile(self, interpolation, frame):
+        g = frame.groupby("A", group_keys=False)
+        r = g.expanding()
+
+        result = r.quantile(0.4, interpolation=interpolation)
+        expected = g.apply(
+            lambda x: x.expanding().quantile(0.4, interpolation=interpolation)
+        )
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("f", ["corr", "cov"])
+    def test_expanding_corr_cov(self, f, frame):
+        g = frame.groupby("A")
+        r = g.expanding()
+
+        result = getattr(r, f)(frame)
+
+        def func_0(x):
+            return getattr(x.expanding(), f)(frame)
+
+        expected = g.apply(func_0)
+        # GH 39591: groupby.apply returns 1 instead of nan for windows
+        # with all nan values
+        null_idx = list(range(20, 61)) + list(range(72, 113))
+        expected.iloc[null_idx, 1] = np.nan
+        # GH 39591: The grouped column should be all np.nan
+        # (groupby.apply inserts 0s for cov)
+        expected["A"] = np.nan
+        tm.assert_frame_equal(result, expected)
+
+        result = getattr(r.B, f)(pairwise=True)
+
+        def func_1(x):
+            return getattr(x.B.expanding(), f)(pairwise=True)
+
+        expected = g.apply(func_1)
+        tm.assert_series_equal(result, expected)
+
+    def test_expanding_apply(self, raw, frame):
+        g = frame.groupby("A", group_keys=False)
+        r = g.expanding()
+
+        # reduction
+        result = r.apply(lambda x: x.sum(), raw=raw)
+        expected = g.apply(lambda x: x.expanding().apply(lambda y: y.sum(), raw=raw))
+        # GH 39732
+        expected_index = MultiIndex.from_arrays([frame["A"], range(40)])
+        expected.index = expected_index
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_expanding_agg_namedagg(self):
+        # GH#28333
+        df = DataFrame(
+            {
+                "kind": ["cat", "dog", "cat", "dog", "cat", "dog"],
+                "height": [9.1, 6.0, 9.5, 34.0, 12.0, 8.0],
+                "weight": [7.9, 7.5, 9.9, 198.0, 10.0, 42.0],
+            }
+        )
+        result = (
+            df.groupby("kind")
+            .expanding(1)
+            .agg(
+                total_weight=NamedAgg(column="weight", aggfunc=sum),
+                min_height=NamedAgg(column="height", aggfunc=min),
+            )
+        )
+        expected = DataFrame(
+            {
+                "total_weight": [7.9, 17.8, 27.8, 7.5, 205.5, 247.5],
+                "min_height": [9.1, 9.1, 9.1, 6.0, 6.0, 6.0],
+            },
+            index=MultiIndex(
+                [["cat", "dog"], [0, 1, 2, 3, 4, 5]],
+                [[0, 0, 0, 1, 1, 1], [0, 2, 4, 1, 3, 5]],
+                names=["kind", None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+class TestEWM:
+    @pytest.mark.parametrize(
+        "method, expected_data",
+        [
+            ["mean", [0.0, 0.6666666666666666, 1.4285714285714286, 2.2666666666666666]],
+            ["std", [np.nan, 0.707107, 0.963624, 1.177164]],
+            ["var", [np.nan, 0.5, 0.9285714285714286, 1.3857142857142857]],
+        ],
+    )
+    def test_methods(self, method, expected_data):
+        # GH 16037
+        df = DataFrame({"A": ["a"] * 4, "B": range(4)})
+        result = getattr(df.groupby("A").ewm(com=1.0), method)()
+        expected = DataFrame(
+            {"B": expected_data},
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", 0),
+                    ("a", 1),
+                    ("a", 2),
+                    ("a", 3),
+                ],
+                names=["A", None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_groupby_ewm_agg_namedagg(self):
+        # GH#28333
+        df = DataFrame({"A": ["a"] * 4, "B": range(4)})
+        result = (
+            df.groupby("A")
+            .ewm(com=1.0)
+            .agg(
+                B_mean=NamedAgg(column="B", aggfunc="mean"),
+                B_std=NamedAgg(column="B", aggfunc="std"),
+                B_var=NamedAgg(column="B", aggfunc="var"),
+            )
+        )
+        expected = DataFrame(
+            {
+                "B_mean": [
+                    0.0,
+                    0.6666666666666666,
+                    1.4285714285714286,
+                    2.2666666666666666,
+                ],
+                "B_std": [np.nan, 0.707107, 0.963624, 1.177164],
+                "B_var": [np.nan, 0.5, 0.9285714285714286, 1.3857142857142857],
+            },
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", 0),
+                    ("a", 1),
+                    ("a", 2),
+                    ("a", 3),
+                ],
+                names=["A", None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "method, expected_data",
+        [["corr", [np.nan, 1.0, 1.0, 1]], ["cov", [np.nan, 0.5, 0.928571, 1.385714]]],
+    )
+    def test_pairwise_methods(self, method, expected_data):
+        # GH 16037
+        df = DataFrame({"A": ["a"] * 4, "B": range(4)})
+        result = getattr(df.groupby("A").ewm(com=1.0), method)()
+        expected = DataFrame(
+            {"B": expected_data},
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", 0, "B"),
+                    ("a", 1, "B"),
+                    ("a", 2, "B"),
+                    ("a", 3, "B"),
+                ],
+                names=["A", None, None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.groupby("A")[["B"]].apply(
+            lambda x: getattr(x.ewm(com=1.0), method)()
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_times(self, times_frame):
+        # GH 40951
+        halflife = "23 days"
+        # GH#42738
+        times = times_frame.pop("C")
+        result = times_frame.groupby("A").ewm(halflife=halflife, times=times).mean()
+        expected = DataFrame(
+            {
+                "B": [
+                    0.0,
+                    0.507534,
+                    1.020088,
+                    1.537661,
+                    0.0,
+                    0.567395,
+                    1.221209,
+                    0.0,
+                    0.653141,
+                    1.195003,
+                ]
+            },
+            index=MultiIndex.from_tuples(
+                [
+                    ("a", 0),
+                    ("a", 3),
+                    ("a", 6),
+                    ("a", 9),
+                    ("b", 1),
+                    ("b", 4),
+                    ("b", 7),
+                    ("c", 2),
+                    ("c", 5),
+                    ("c", 8),
+                ],
+                names=["A", None],
+            ),
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_times_array(self, times_frame):
+        # GH 40951
+        halflife = "23 days"
+        times = times_frame.pop("C")
+        gb = times_frame.groupby("A")
+        result = gb.ewm(halflife=halflife, times=times).mean()
+        expected = gb.ewm(halflife=halflife, times=times.values).mean()
+        tm.assert_frame_equal(result, expected)
+
+    def test_dont_mutate_obj_after_slicing(self):
+        # GH 43355
+        df = DataFrame(
+            {
+                "id": ["a", "a", "b", "b", "b"],
+                "timestamp": date_range("2021-9-1", periods=5, freq="h"),
+                "y": range(5),
+            }
+        )
+        grp = df.groupby("id").rolling("1h", on="timestamp")
+        result = grp.count()
+        expected_df = DataFrame(
+            {
+                "timestamp": date_range("2021-9-1", periods=5, freq="h"),
+                "y": [1.0] * 5,
+            },
+            index=MultiIndex.from_arrays(
+                [["a", "a", "b", "b", "b"], list(range(5))], names=["id", None]
+            ),
+        )
+        tm.assert_frame_equal(result, expected_df)
+
+        result = grp["y"].count()
+        expected_series = Series(
+            [1.0] * 5,
+            index=MultiIndex.from_arrays(
+                [
+                    ["a", "a", "b", "b", "b"],
+                    date_range("2021-9-1", periods=5, freq="h"),
+                ],
+                names=["id", "timestamp"],
+            ),
+            name="y",
+        )
+        tm.assert_series_equal(result, expected_series)
+        # This is the key test
+        result = grp.count()
+        tm.assert_frame_equal(result, expected_df)
+
+
+def test_rolling_corr_with_single_integer_in_index():
+    # GH 44078
+    df = DataFrame({"a": [(1,), (1,), (1,)], "b": [4, 5, 6]})
+    gb = df.groupby(["a"])
+    result = gb.rolling(2).corr(other=df)
+    index = MultiIndex.from_tuples([((1,), 0), ((1,), 1), ((1,), 2)], names=["a", None])
+    expected = DataFrame(
+        {"a": [np.nan, np.nan, np.nan], "b": [np.nan, 1.0, 1.0]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_corr_with_tuples_in_index():
+    # GH 44078
+    df = DataFrame(
+        {
+            "a": [
+                (
+                    1,
+                    2,
+                ),
+                (
+                    1,
+                    2,
+                ),
+                (
+                    1,
+                    2,
+                ),
+            ],
+            "b": [4, 5, 6],
+        }
+    )
+    gb = df.groupby(["a"])
+    result = gb.rolling(2).corr(other=df)
+    index = MultiIndex.from_tuples(
+        [((1, 2), 0), ((1, 2), 1), ((1, 2), 2)], names=["a", None]
+    )
+    expected = DataFrame(
+        {"a": [np.nan, np.nan, np.nan], "b": [np.nan, 1.0, 1.0]}, index=index
+    )
+    tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_numba.py b/pandas/pandas/tests/window/test_numba.py
new file mode 100644
index 0000000000000..887aeca6590dc
--- /dev/null
+++ b/pandas/pandas/tests/window/test_numba.py
@@ -0,0 +1,583 @@
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+from pandas.errors import NumbaUtilError
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    Series,
+    option_context,
+    to_datetime,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [pytest.mark.single_cpu]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+@pytest.fixture(params=["single", "table"])
+def method(request):
+    """method keyword in rolling/expanding/ewm constructor"""
+    return request.param
+
+
+@pytest.fixture(
+    params=[
+        ["sum", {}],
+        ["mean", {}],
+        ["median", {}],
+        ["max", {}],
+        ["min", {}],
+        ["var", {}],
+        ["var", {"ddof": 0}],
+        ["std", {}],
+        ["std", {"ddof": 0}],
+    ]
+)
+def arithmetic_numba_supported_operators(request):
+    return request.param
+
+
+@pytest.fixture
+def roll_frame():
+    return DataFrame({"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)})
+
+
+@td.skip_if_no("numba")
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+class TestEngine:
+    @pytest.mark.parametrize("jit", [True, False])
+    def test_numba_vs_cython_apply(self, jit, nogil, parallel, nopython, center, step):
+        def f(x, *args):
+            arg_sum = 0
+            for arg in args:
+                arg_sum += arg
+            return np.mean(x) + arg_sum
+
+        if jit:
+            import numba
+
+            f = numba.jit(f)
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        args = (2,)
+
+        s = Series(range(10))
+        result = s.rolling(2, center=center, step=step).apply(
+            f, args=args, engine="numba", engine_kwargs=engine_kwargs, raw=True
+        )
+        expected = s.rolling(2, center=center, step=step).apply(
+            f, engine="cython", args=args, raw=True
+        )
+        tm.assert_series_equal(result, expected)
+
+    def test_apply_numba_with_kwargs(self, roll_frame):
+        # GH 58995
+        # rolling apply
+        def func(sr, a=0):
+            return sr.sum() + a
+
+        data = DataFrame(range(10))
+
+        result = data.rolling(5).apply(func, engine="numba", raw=True, kwargs={"a": 1})
+        expected = data.rolling(5).sum() + 1
+        tm.assert_frame_equal(result, expected)
+
+        result = data.rolling(5).apply(func, engine="numba", raw=True, args=(1,))
+        tm.assert_frame_equal(result, expected)
+
+        # expanding apply
+
+        result = data.expanding().apply(func, engine="numba", raw=True, kwargs={"a": 1})
+        expected = data.expanding().sum() + 1
+        tm.assert_frame_equal(result, expected)
+
+        result = data.expanding().apply(func, engine="numba", raw=True, args=(1,))
+        tm.assert_frame_equal(result, expected)
+
+        # groupby rolling
+        result = (
+            roll_frame.groupby("A")
+            .rolling(5)
+            .apply(func, engine="numba", raw=True, kwargs={"a": 1})
+        )
+        expected = roll_frame.groupby("A").rolling(5).sum() + 1
+        tm.assert_frame_equal(result, expected)
+
+        result = (
+            roll_frame.groupby("A")
+            .rolling(5)
+            .apply(func, engine="numba", raw=True, args=(1,))
+        )
+        tm.assert_frame_equal(result, expected)
+        # groupby expanding
+
+        result = (
+            roll_frame.groupby("A")
+            .expanding()
+            .apply(func, engine="numba", raw=True, kwargs={"a": 1})
+        )
+        expected = roll_frame.groupby("A").expanding().sum() + 1
+        tm.assert_frame_equal(result, expected)
+
+        result = (
+            roll_frame.groupby("A")
+            .expanding()
+            .apply(func, engine="numba", raw=True, args=(1,))
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_numba_min_periods(self):
+        # GH 58868
+        def last_row(x):
+            assert len(x) == 3
+            return x[-1]
+
+        df = DataFrame([[1, 2], [3, 4], [5, 6], [7, 8]])
+
+        result = df.rolling(3, method="table", min_periods=3).apply(
+            last_row, raw=True, engine="numba"
+        )
+
+        expected = DataFrame([[np.nan, np.nan], [np.nan, np.nan], [5, 6], [7, 8]])
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data",
+        [
+            DataFrame(np.eye(5)),
+            DataFrame(
+                [
+                    [5, 7, 7, 7, np.nan, np.inf, 4, 3, 3, 3],
+                    [5, 7, 7, 7, np.nan, np.inf, 7, 3, 3, 3],
+                    [np.nan, np.nan, 5, 6, 7, 5, 5, 5, 5, 5],
+                ]
+            ).T,
+            Series(range(5), name="foo"),
+            Series([20, 10, 10, np.inf, 1, 1, 2, 3]),
+            Series([20, 10, 10, np.nan, 10, 1, 2, 3]),
+        ],
+    )
+    def test_numba_vs_cython_rolling_methods(
+        self,
+        data,
+        nogil,
+        parallel,
+        nopython,
+        arithmetic_numba_supported_operators,
+        step,
+    ):
+        method, kwargs = arithmetic_numba_supported_operators
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        roll = data.rolling(3, step=step)
+        result = getattr(roll, method)(
+            engine="numba", engine_kwargs=engine_kwargs, **kwargs
+        )
+        expected = getattr(roll, method)(engine="cython", **kwargs)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "data", [DataFrame(np.eye(5)), Series(range(5), name="foo")]
+    )
+    def test_numba_vs_cython_expanding_methods(
+        self, data, nogil, parallel, nopython, arithmetic_numba_supported_operators
+    ):
+        method, kwargs = arithmetic_numba_supported_operators
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        data = DataFrame(np.eye(5))
+        expand = data.expanding()
+        result = getattr(expand, method)(
+            engine="numba", engine_kwargs=engine_kwargs, **kwargs
+        )
+        expected = getattr(expand, method)(engine="cython", **kwargs)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize("jit", [True, False])
+    def test_cache_apply(self, jit, nogil, parallel, nopython, step):
+        # Test that the functions are cached correctly if we switch functions
+        def func_1(x):
+            return np.mean(x) + 4
+
+        def func_2(x):
+            return np.std(x) * 5
+
+        if jit:
+            import numba
+
+            func_1 = numba.jit(func_1)
+            func_2 = numba.jit(func_2)
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        roll = Series(range(10)).rolling(2, step=step)
+        result = roll.apply(
+            func_1, engine="numba", engine_kwargs=engine_kwargs, raw=True
+        )
+        expected = roll.apply(func_1, engine="cython", raw=True)
+        tm.assert_series_equal(result, expected)
+
+        result = roll.apply(
+            func_2, engine="numba", engine_kwargs=engine_kwargs, raw=True
+        )
+        expected = roll.apply(func_2, engine="cython", raw=True)
+        tm.assert_series_equal(result, expected)
+        # This run should use the cached func_1
+        result = roll.apply(
+            func_1, engine="numba", engine_kwargs=engine_kwargs, raw=True
+        )
+        expected = roll.apply(func_1, engine="cython", raw=True)
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "window,window_kwargs",
+        [
+            ["rolling", {"window": 3, "min_periods": 0}],
+            ["expanding", {}],
+        ],
+    )
+    def test_dont_cache_args(
+        self, window, window_kwargs, nogil, parallel, nopython, method
+    ):
+        # GH 42287
+
+        def add(values, x):
+            return np.sum(values) + x
+
+        engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+        df = DataFrame({"value": [0, 0, 0]})
+        result = getattr(df, window)(method=method, **window_kwargs).apply(
+            add, raw=True, engine="numba", engine_kwargs=engine_kwargs, args=(1,)
+        )
+        expected = DataFrame({"value": [1.0, 1.0, 1.0]})
+        tm.assert_frame_equal(result, expected)
+
+        result = getattr(df, window)(method=method, **window_kwargs).apply(
+            add, raw=True, engine="numba", engine_kwargs=engine_kwargs, args=(2,)
+        )
+        expected = DataFrame({"value": [2.0, 2.0, 2.0]})
+        tm.assert_frame_equal(result, expected)
+
+    def test_dont_cache_engine_kwargs(self):
+        # If the user passes a different set of engine_kwargs don't return the same
+        # jitted function
+        nogil = False
+        parallel = True
+        nopython = True
+
+        def func(x):
+            return nogil + parallel + nopython
+
+        engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+        df = DataFrame({"value": [0, 0, 0]})
+        result = df.rolling(1).apply(
+            func, raw=True, engine="numba", engine_kwargs=engine_kwargs
+        )
+        expected = DataFrame({"value": [2.0, 2.0, 2.0]})
+        tm.assert_frame_equal(result, expected)
+
+        parallel = False
+        engine_kwargs = {"nopython": nopython, "nogil": nogil, "parallel": parallel}
+        result = df.rolling(1).apply(
+            func, raw=True, engine="numba", engine_kwargs=engine_kwargs
+        )
+        expected = DataFrame({"value": [1.0, 1.0, 1.0]})
+        tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("numba")
+class TestEWM:
+    @pytest.mark.parametrize(
+        "grouper", [lambda x: x, lambda x: x.groupby("A")], ids=["None", "groupby"]
+    )
+    @pytest.mark.parametrize("method", ["mean", "sum"])
+    def test_invalid_engine(self, grouper, method):
+        df = DataFrame({"A": ["a", "b", "a", "b"], "B": range(4)})
+        with pytest.raises(ValueError, match="engine must be either"):
+            getattr(grouper(df).ewm(com=1.0), method)(engine="foo")
+
+    @pytest.mark.parametrize(
+        "grouper", [lambda x: x, lambda x: x.groupby("A")], ids=["None", "groupby"]
+    )
+    @pytest.mark.parametrize("method", ["mean", "sum"])
+    def test_invalid_engine_kwargs(self, grouper, method):
+        df = DataFrame({"A": ["a", "b", "a", "b"], "B": range(4)})
+        with pytest.raises(ValueError, match="cython engine does not"):
+            getattr(grouper(df).ewm(com=1.0), method)(
+                engine="cython", engine_kwargs={"nopython": True}
+            )
+
+    @pytest.mark.parametrize("grouper", ["None", "groupby"])
+    @pytest.mark.parametrize("method", ["mean", "sum"])
+    def test_cython_vs_numba(
+        self, grouper, method, nogil, parallel, nopython, ignore_na, adjust
+    ):
+        df = DataFrame({"B": range(4)})
+        if grouper == "None":
+            grouper = lambda x: x
+        else:
+            df["A"] = ["a", "b", "a", "b"]
+            grouper = lambda x: x.groupby("A")
+        if method == "sum":
+            adjust = True
+        ewm = grouper(df).ewm(com=1.0, adjust=adjust, ignore_na=ignore_na)
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        result = getattr(ewm, method)(engine="numba", engine_kwargs=engine_kwargs)
+        expected = getattr(ewm, method)(engine="cython")
+
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("grouper", ["None", "groupby"])
+    def test_cython_vs_numba_times(self, grouper, nogil, parallel, nopython, ignore_na):
+        # GH 40951
+
+        df = DataFrame({"B": [0, 0, 1, 1, 2, 2]})
+        if grouper == "None":
+            grouper = lambda x: x
+        else:
+            grouper = lambda x: x.groupby("A")
+            df["A"] = ["a", "b", "a", "b", "b", "a"]
+
+        halflife = "23 days"
+        times = to_datetime(
+            [
+                "2020-01-01",
+                "2020-01-01",
+                "2020-01-02",
+                "2020-01-10",
+                "2020-02-23",
+                "2020-01-03",
+            ]
+        )
+        ewm = grouper(df).ewm(
+            halflife=halflife, adjust=True, ignore_na=ignore_na, times=times
+        )
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        result = ewm.mean(engine="numba", engine_kwargs=engine_kwargs)
+        expected = ewm.mean(engine="cython")
+
+        tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("numba")
+def test_use_global_config():
+    def f(x):
+        return np.mean(x) + 2
+
+    s = Series(range(10))
+    with option_context("compute.use_numba", True):
+        result = s.rolling(2).apply(f, engine=None, raw=True)
+    expected = s.rolling(2).apply(f, engine="numba", raw=True)
+    tm.assert_series_equal(expected, result)
+
+
+@td.skip_if_no("numba")
+def test_invalid_kwargs_nopython():
+    with pytest.raises(TypeError, match="got an unexpected keyword argument 'a'"):
+        Series(range(1)).rolling(1).apply(
+            lambda x: x, kwargs={"a": 1}, engine="numba", raw=True
+        )
+    with pytest.raises(
+        NumbaUtilError, match="numba does not support keyword-only arguments"
+    ):
+        Series(range(1)).rolling(1).apply(
+            lambda x, *, a: x, kwargs={"a": 1}, engine="numba", raw=True
+        )
+
+    tm.assert_series_equal(
+        Series(range(1), dtype=float) + 1,
+        Series(range(1))
+        .rolling(1)
+        .apply(lambda x, a: (x + a).sum(), kwargs={"a": 1}, engine="numba", raw=True),
+    )
+
+
+@td.skip_if_no("numba")
+@pytest.mark.slow
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+class TestTableMethod:
+    def test_table_series_valueerror(self):
+        def f(x):
+            return np.sum(x, axis=0) + 1
+
+        with pytest.raises(
+            ValueError, match="method='table' not applicable for Series objects."
+        ):
+            Series(range(1)).rolling(1, method="table").apply(
+                f, engine="numba", raw=True
+            )
+
+    def test_table_method_rolling_methods(
+        self,
+        nogil,
+        parallel,
+        nopython,
+        arithmetic_numba_supported_operators,
+        step,
+    ):
+        method, kwargs = arithmetic_numba_supported_operators
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        df = DataFrame(np.eye(3))
+        roll_table = df.rolling(2, method="table", min_periods=0, step=step)
+        if method in ("var", "std"):
+            with pytest.raises(NotImplementedError, match=f"{method} not supported"):
+                getattr(roll_table, method)(
+                    engine_kwargs=engine_kwargs, engine="numba", **kwargs
+                )
+        else:
+            roll_single = df.rolling(2, method="single", min_periods=0, step=step)
+            result = getattr(roll_table, method)(
+                engine_kwargs=engine_kwargs, engine="numba", **kwargs
+            )
+            expected = getattr(roll_single, method)(
+                engine_kwargs=engine_kwargs, engine="numba", **kwargs
+            )
+            tm.assert_frame_equal(result, expected)
+
+    def test_table_method_rolling_apply(self, nogil, parallel, nopython, step):
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        def f(x):
+            return np.sum(x, axis=0) + 1
+
+        df = DataFrame(np.eye(3))
+        result = df.rolling(2, method="table", min_periods=0, step=step).apply(
+            f, raw=True, engine_kwargs=engine_kwargs, engine="numba"
+        )
+        expected = df.rolling(2, method="single", min_periods=0, step=step).apply(
+            f, raw=True, engine_kwargs=engine_kwargs, engine="numba"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_table_method_rolling_apply_col_order(self):
+        # GH#59666
+        def f(x):
+            return np.nanmean(x[:, 0] - x[:, 1])
+
+        df = DataFrame(
+            {
+                "a": [1, 2, 3, 4, 5, 6],
+                "b": [6, 7, 8, 5, 6, 7],
+            }
+        )
+        result = df.rolling(3, method="table", min_periods=0)[["a", "b"]].apply(
+            f, raw=True, engine="numba"
+        )
+        expected = DataFrame(
+            {
+                "a": [-5, -5, -5, -3.66667, -2.33333, -1],
+                "b": [-5, -5, -5, -3.66667, -2.33333, -1],
+            }
+        )
+        tm.assert_almost_equal(result, expected)
+        result = df.rolling(3, method="table", min_periods=0)[["b", "a"]].apply(
+            f, raw=True, engine="numba"
+        )
+        expected = DataFrame(
+            {
+                "b": [5, 5, 5, 3.66667, 2.33333, 1],
+                "a": [5, 5, 5, 3.66667, 2.33333, 1],
+            }
+        )
+        tm.assert_almost_equal(result, expected)
+
+    def test_table_method_rolling_weighted_mean(self, step):
+        def weighted_mean(x):
+            arr = np.ones((1, x.shape[1]))
+            arr[:, :2] = (x[:, :2] * x[:, 2]).sum(axis=0) / x[:, 2].sum()
+            return arr
+
+        df = DataFrame([[1, 2, 0.6], [2, 3, 0.4], [3, 4, 0.2], [4, 5, 0.7]])
+        result = df.rolling(2, method="table", min_periods=0, step=step).apply(
+            weighted_mean, raw=True, engine="numba"
+        )
+        expected = DataFrame(
+            [
+                [1.0, 2.0, 1.0],
+                [1.8, 2.0, 1.0],
+                [3.333333, 2.333333, 1.0],
+                [1.555556, 7, 1.0],
+            ]
+        )[::step]
+        tm.assert_frame_equal(result, expected)
+
+    def test_table_method_expanding_apply(self, nogil, parallel, nopython):
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        def f(x):
+            return np.sum(x, axis=0) + 1
+
+        df = DataFrame(np.eye(3))
+        result = df.expanding(method="table").apply(
+            f, raw=True, engine_kwargs=engine_kwargs, engine="numba"
+        )
+        expected = df.expanding(method="single").apply(
+            f, raw=True, engine_kwargs=engine_kwargs, engine="numba"
+        )
+        tm.assert_frame_equal(result, expected)
+
+    def test_table_method_expanding_methods(
+        self, nogil, parallel, nopython, arithmetic_numba_supported_operators
+    ):
+        method, kwargs = arithmetic_numba_supported_operators
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        df = DataFrame(np.eye(3))
+        expand_table = df.expanding(method="table")
+        if method in ("var", "std"):
+            with pytest.raises(NotImplementedError, match=f"{method} not supported"):
+                getattr(expand_table, method)(
+                    engine_kwargs=engine_kwargs, engine="numba", **kwargs
+                )
+        else:
+            expand_single = df.expanding(method="single")
+            result = getattr(expand_table, method)(
+                engine_kwargs=engine_kwargs, engine="numba", **kwargs
+            )
+            expected = getattr(expand_single, method)(
+                engine_kwargs=engine_kwargs, engine="numba", **kwargs
+            )
+            tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("data", [np.eye(3), np.ones((2, 3)), np.ones((3, 2))])
+    @pytest.mark.parametrize("method", ["mean", "sum"])
+    def test_table_method_ewm(self, data, method, nogil, parallel, nopython):
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        df = DataFrame(data)
+
+        result = getattr(df.ewm(com=1, method="table"), method)(
+            engine_kwargs=engine_kwargs, engine="numba"
+        )
+        expected = getattr(df.ewm(com=1, method="single"), method)(
+            engine_kwargs=engine_kwargs, engine="numba"
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+@td.skip_if_no("numba")
+def test_npfunc_no_warnings():
+    df = DataFrame({"col1": [1, 2, 3, 4, 5]})
+    with tm.assert_produces_warning(False):
+        df.col1.rolling(2).apply(np.prod, raw=True, engine="numba")
diff --git a/pandas/pandas/tests/window/test_online.py b/pandas/pandas/tests/window/test_online.py
new file mode 100644
index 0000000000000..43d55a7992b3c
--- /dev/null
+++ b/pandas/pandas/tests/window/test_online.py
@@ -0,0 +1,112 @@
+import numpy as np
+import pytest
+
+from pandas.compat import is_platform_arm
+
+from pandas import (
+    DataFrame,
+    Series,
+)
+import pandas._testing as tm
+from pandas.util.version import Version
+
+pytestmark = [pytest.mark.single_cpu]
+
+numba = pytest.importorskip("numba")
+pytestmark.append(
+    pytest.mark.skipif(
+        Version(numba.__version__) == Version("0.61") and is_platform_arm(),
+        reason=f"Segfaults on ARM platforms with numba {numba.__version__}",
+    )
+)
+
+
+@pytest.mark.filterwarnings("ignore")
+# Filter warnings when parallel=True and the function can't be parallelized by Numba
+class TestEWM:
+    def test_invalid_update(self):
+        df = DataFrame({"a": range(5), "b": range(5)})
+        online_ewm = df.head(2).ewm(0.5).online()
+        with pytest.raises(
+            ValueError,
+            match="Must call mean with update=None first before passing update",
+        ):
+            online_ewm.mean(update=df.head(1))
+
+    @pytest.mark.slow
+    @pytest.mark.parametrize(
+        "obj", [DataFrame({"a": range(5), "b": range(5)}), Series(range(5), name="foo")]
+    )
+    def test_online_vs_non_online_mean(
+        self, obj, nogil, parallel, nopython, adjust, ignore_na
+    ):
+        expected = obj.ewm(0.5, adjust=adjust, ignore_na=ignore_na).mean()
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+
+        online_ewm = (
+            obj.head(2)
+            .ewm(0.5, adjust=adjust, ignore_na=ignore_na)
+            .online(engine_kwargs=engine_kwargs)
+        )
+        # Test resetting once
+        for _ in range(2):
+            result = online_ewm.mean()
+            tm.assert_equal(result, expected.head(2))
+
+            result = online_ewm.mean(update=obj.tail(3))
+            tm.assert_equal(result, expected.tail(3))
+
+            online_ewm.reset()
+
+    @pytest.mark.xfail(raises=NotImplementedError)
+    @pytest.mark.parametrize(
+        "obj", [DataFrame({"a": range(5), "b": range(5)}), Series(range(5), name="foo")]
+    )
+    def test_update_times_mean(
+        self, obj, nogil, parallel, nopython, adjust, ignore_na, halflife_with_times
+    ):
+        times = Series(
+            np.array(
+                ["2020-01-01", "2020-01-05", "2020-01-07", "2020-01-17", "2020-01-21"],
+                dtype="datetime64[ns]",
+            )
+        )
+        expected = obj.ewm(
+            0.5,
+            adjust=adjust,
+            ignore_na=ignore_na,
+            times=times,
+            halflife=halflife_with_times,
+        ).mean()
+
+        engine_kwargs = {"nogil": nogil, "parallel": parallel, "nopython": nopython}
+        online_ewm = (
+            obj.head(2)
+            .ewm(
+                0.5,
+                adjust=adjust,
+                ignore_na=ignore_na,
+                times=times.head(2),
+                halflife=halflife_with_times,
+            )
+            .online(engine_kwargs=engine_kwargs)
+        )
+        # Test resetting once
+        for _ in range(2):
+            result = online_ewm.mean()
+            tm.assert_equal(result, expected.head(2))
+
+            result = online_ewm.mean(update=obj.tail(3), update_times=times.tail(3))
+            tm.assert_equal(result, expected.tail(3))
+
+            online_ewm.reset()
+
+    @pytest.mark.parametrize("method", ["aggregate", "std", "corr", "cov", "var"])
+    def test_ewm_notimplementederror_raises(self, method):
+        ser = Series(range(10))
+        kwargs = {}
+        if method == "aggregate":
+            kwargs["func"] = lambda x: x
+
+        with pytest.raises(NotImplementedError, match=".* is not implemented."):
+            getattr(ser.ewm(1).online(), method)(**kwargs)
diff --git a/pandas/pandas/tests/window/test_pairwise.py b/pandas/pandas/tests/window/test_pairwise.py
new file mode 100644
index 0000000000000..d23c6501ed1d1
--- /dev/null
+++ b/pandas/pandas/tests/window/test_pairwise.py
@@ -0,0 +1,445 @@
+import numpy as np
+import pytest
+
+from pandas.compat import IS64
+
+from pandas import (
+    DataFrame,
+    Index,
+    MultiIndex,
+    Series,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.core.algorithms import safe_sort
+
+
+@pytest.fixture(
+    params=[
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=[1, 0]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=[1, 1]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=["C", "C"]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=[1.0, 0]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=[0.0, 1]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=["C", 1]),
+        DataFrame([[2.0, 4.0], [1.0, 2.0], [5.0, 2.0], [8.0, 1.0]], columns=[1, 0.0]),
+        DataFrame([[2, 4.0], [1, 2.0], [5, 2.0], [8, 1.0]], columns=[0, 1.0]),
+        DataFrame([[2, 4], [1, 2], [5, 2], [8, 1.0]], columns=[1.0, "X"]),
+    ]
+)
+def pairwise_frames(request):
+    """Pairwise frames test_pairwise"""
+    return request.param
+
+
+@pytest.fixture
+def pairwise_target_frame():
+    """Pairwise target frame for test_pairwise"""
+    return DataFrame([[2, 4], [1, 2], [5, 2], [8, 1]], columns=[0, 1])
+
+
+@pytest.fixture
+def pairwise_other_frame():
+    """Pairwise other frame for test_pairwise"""
+    return DataFrame(
+        [[None, 1, 1], [None, 1, 2], [None, 3, 2], [None, 8, 1]],
+        columns=["Y", "Z", "X"],
+    )
+
+
+def test_rolling_cov(series):
+    A = series
+    B = A + np.random.default_rng(2).standard_normal(len(A))
+
+    result = A.rolling(window=50, min_periods=25).cov(B)
+    tm.assert_almost_equal(result.iloc[-1], np.cov(A[-50:], B[-50:])[0, 1])
+
+
+def test_rolling_corr(series):
+    A = series
+    B = A + np.random.default_rng(2).standard_normal(len(A))
+
+    result = A.rolling(window=50, min_periods=25).corr(B)
+    tm.assert_almost_equal(result.iloc[-1], np.corrcoef(A[-50:], B[-50:])[0, 1])
+
+
+def test_rolling_corr_bias_correction():
+    # test for correct bias correction
+    a = Series(
+        np.arange(20, dtype=np.float64), index=date_range("2020-01-01", periods=20)
+    )
+    b = a.copy()
+    a[:5] = np.nan
+    b[:10] = np.nan
+
+    result = a.rolling(window=len(a), min_periods=1).corr(b)
+    tm.assert_almost_equal(result.iloc[-1], a.corr(b))
+
+
+@pytest.mark.parametrize("func", ["cov", "corr"])
+def test_rolling_pairwise_cov_corr(func, frame):
+    result = getattr(frame.rolling(window=10, min_periods=5), func)()
+    result = result.loc[(slice(None), 1), 5]
+    result.index = result.index.droplevel(1)
+    expected = getattr(frame[1].rolling(window=10, min_periods=5), func)(frame[5])
+    tm.assert_series_equal(result, expected, check_names=False)
+
+
+@pytest.mark.parametrize("method", ["corr", "cov"])
+def test_flex_binary_frame(method, frame):
+    series = frame[1]
+
+    res = getattr(series.rolling(window=10), method)(frame)
+    res2 = getattr(frame.rolling(window=10), method)(series)
+    exp = frame.apply(lambda x: getattr(series.rolling(window=10), method)(x))
+
+    tm.assert_frame_equal(res, exp)
+    tm.assert_frame_equal(res2, exp)
+
+    frame2 = DataFrame(
+        np.random.default_rng(2).standard_normal(frame.shape),
+        index=frame.index,
+        columns=frame.columns,
+    )
+
+    res3 = getattr(frame.rolling(window=10), method)(frame2)
+    res3.columns = Index(list(res3.columns))
+    exp = DataFrame(
+        {k: getattr(frame[k].rolling(window=10), method)(frame2[k]) for k in frame}
+    )
+    tm.assert_frame_equal(res3, exp)
+
+
+@pytest.mark.parametrize("window", range(7))
+def test_rolling_corr_with_zero_variance(window):
+    # GH 18430
+    s = Series(np.zeros(20))
+    other = Series(np.arange(20))
+
+    assert s.rolling(window=window).corr(other=other).isna().all()
+
+
+def test_corr_sanity():
+    # GH 3155
+    df = DataFrame(
+        np.array(
+            [
+                [0.87024726, 0.18505595],
+                [0.64355431, 0.3091617],
+                [0.92372966, 0.50552513],
+                [0.00203756, 0.04520709],
+                [0.84780328, 0.33394331],
+                [0.78369152, 0.63919667],
+            ]
+        )
+    )
+
+    res = df[0].rolling(5, center=True).corr(df[1])
+    assert all(np.abs(np.nan_to_num(x)) <= 1 for x in res)
+
+    df = DataFrame(np.random.default_rng(2).random((30, 2)))
+    res = df[0].rolling(5, center=True).corr(df[1])
+    assert all(np.abs(np.nan_to_num(x)) <= 1 for x in res)
+
+
+def test_rolling_cov_diff_length():
+    # GH 7512
+    s1 = Series([1, 2, 3], index=range(3))
+    s2 = Series([1, 3], index=range(0, 4, 2))
+    result = s1.rolling(window=3, min_periods=2).cov(s2)
+    expected = Series([None, None, 2.0])
+    tm.assert_series_equal(result, expected)
+
+    s2a = Series([1, None, 3], index=range(3))
+    result = s1.rolling(window=3, min_periods=2).cov(s2a)
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_corr_diff_length():
+    # GH 7512
+    s1 = Series([1, 2, 3], index=range(3))
+    s2 = Series([1, 3], index=range(0, 4, 2))
+    result = s1.rolling(window=3, min_periods=2).corr(s2)
+    expected = Series([None, None, 1.0])
+    tm.assert_series_equal(result, expected)
+
+    s2a = Series([1, None, 3], index=range(3))
+    result = s1.rolling(window=3, min_periods=2).corr(s2a)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: (x.rolling(window=10, min_periods=5).cov(x, pairwise=True)),
+        lambda x: (x.rolling(window=10, min_periods=5).corr(x, pairwise=True)),
+    ],
+)
+def test_rolling_functions_window_non_shrinkage_binary(f):
+    # corr/cov return a MI DataFrame
+    df = DataFrame(
+        [[1, 5], [3, 2], [3, 9], [-1, 0]],
+        columns=Index(["A", "B"], name="foo"),
+        index=Index(range(4), name="bar"),
+    )
+    df_expected = DataFrame(
+        columns=Index(["A", "B"], name="foo"),
+        index=MultiIndex.from_product([df.index, df.columns], names=["bar", "foo"]),
+        dtype="float64",
+    )
+    df_result = f(df)
+    tm.assert_frame_equal(df_result, df_expected)
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: (x.rolling(window=10, min_periods=5).cov(x, pairwise=True)),
+        lambda x: (x.rolling(window=10, min_periods=5).corr(x, pairwise=True)),
+    ],
+)
+def test_moment_functions_zero_length_pairwise(f):
+    df1 = DataFrame()
+    df2 = DataFrame(columns=Index(["a"], name="foo"), index=Index([], name="bar"))
+    df2["a"] = df2["a"].astype("float64")
+
+    df1_expected = DataFrame(index=MultiIndex.from_product([df1.index, df1.columns]))
+    df2_expected = DataFrame(
+        index=MultiIndex.from_product([df2.index, df2.columns], names=["bar", "foo"]),
+        columns=Index(["a"], name="foo"),
+        dtype="float64",
+    )
+
+    df1_result = f(df1)
+    tm.assert_frame_equal(df1_result, df1_expected)
+
+    df2_result = f(df2)
+    tm.assert_frame_equal(df2_result, df2_expected)
+
+
+class TestPairwise:
+    # GH 7738
+    @pytest.mark.parametrize("f", [lambda x: x.cov(), lambda x: x.corr()])
+    def test_no_flex(self, pairwise_frames, pairwise_target_frame, f):
+        # DataFrame methods (which do not call flex_binary_moment())
+
+        result = f(pairwise_frames)
+        tm.assert_index_equal(result.index, pairwise_frames.columns)
+        tm.assert_index_equal(result.columns, pairwise_frames.columns)
+        expected = f(pairwise_target_frame)
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            lambda x: x.expanding().cov(pairwise=True),
+            lambda x: x.expanding().corr(pairwise=True),
+            lambda x: x.rolling(window=3).cov(pairwise=True),
+            lambda x: x.rolling(window=3).corr(pairwise=True),
+            lambda x: x.ewm(com=3).cov(pairwise=True),
+            lambda x: x.ewm(com=3).corr(pairwise=True),
+        ],
+    )
+    def test_pairwise_with_self(self, pairwise_frames, pairwise_target_frame, f):
+        # DataFrame with itself, pairwise=True
+        # note that we may construct the 1st level of the MI
+        # in a non-monotonic way, so compare accordingly
+        result = f(pairwise_frames)
+        tm.assert_index_equal(
+            result.index.levels[0], pairwise_frames.index, check_names=False
+        )
+        tm.assert_index_equal(
+            safe_sort(result.index.levels[1]),
+            safe_sort(pairwise_frames.columns.unique()),
+        )
+        tm.assert_index_equal(result.columns, pairwise_frames.columns)
+        expected = f(pairwise_target_frame)
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            lambda x: x.expanding().cov(pairwise=False),
+            lambda x: x.expanding().corr(pairwise=False),
+            lambda x: x.rolling(window=3).cov(pairwise=False),
+            lambda x: x.rolling(window=3).corr(pairwise=False),
+            lambda x: x.ewm(com=3).cov(pairwise=False),
+            lambda x: x.ewm(com=3).corr(pairwise=False),
+        ],
+    )
+    def test_no_pairwise_with_self(self, pairwise_frames, pairwise_target_frame, f):
+        # DataFrame with itself, pairwise=False
+        result = f(pairwise_frames)
+        tm.assert_index_equal(result.index, pairwise_frames.index)
+        tm.assert_index_equal(result.columns, pairwise_frames.columns)
+        expected = f(pairwise_target_frame)
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            lambda x, y: x.expanding().cov(y, pairwise=True),
+            lambda x, y: x.expanding().corr(y, pairwise=True),
+            lambda x, y: x.rolling(window=3).cov(y, pairwise=True),
+            # TODO: We're missing a flag somewhere in meson
+            pytest.param(
+                lambda x, y: x.rolling(window=3).corr(y, pairwise=True),
+                marks=pytest.mark.xfail(
+                    not IS64, reason="Precision issues on 32 bit", strict=False
+                ),
+            ),
+            lambda x, y: x.ewm(com=3).cov(y, pairwise=True),
+            lambda x, y: x.ewm(com=3).corr(y, pairwise=True),
+        ],
+    )
+    def test_pairwise_with_other(
+        self, pairwise_frames, pairwise_target_frame, pairwise_other_frame, f
+    ):
+        # DataFrame with another DataFrame, pairwise=True
+        result = f(pairwise_frames, pairwise_other_frame)
+        tm.assert_index_equal(
+            result.index.levels[0], pairwise_frames.index, check_names=False
+        )
+        tm.assert_index_equal(
+            safe_sort(result.index.levels[1]),
+            safe_sort(pairwise_other_frame.columns.unique()),
+        )
+        expected = f(pairwise_target_frame, pairwise_other_frame)
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    @pytest.mark.filterwarnings("ignore:RuntimeWarning")
+    @pytest.mark.parametrize(
+        "f",
+        [
+            lambda x, y: x.expanding().cov(y, pairwise=False),
+            lambda x, y: x.expanding().corr(y, pairwise=False),
+            lambda x, y: x.rolling(window=3).cov(y, pairwise=False),
+            lambda x, y: x.rolling(window=3).corr(y, pairwise=False),
+            lambda x, y: x.ewm(com=3).cov(y, pairwise=False),
+            lambda x, y: x.ewm(com=3).corr(y, pairwise=False),
+        ],
+    )
+    def test_no_pairwise_with_other(self, pairwise_frames, pairwise_other_frame, f):
+        # DataFrame with another DataFrame, pairwise=False
+        result = (
+            f(pairwise_frames, pairwise_other_frame)
+            if pairwise_frames.columns.is_unique
+            else None
+        )
+        if result is not None:
+            # we can have int and str columns
+            expected_index = pairwise_frames.index.union(pairwise_other_frame.index)
+            expected_columns = pairwise_frames.columns.union(
+                pairwise_other_frame.columns
+            )
+            tm.assert_index_equal(result.index, expected_index)
+            tm.assert_index_equal(result.columns, expected_columns)
+        else:
+            with pytest.raises(ValueError, match="'arg1' columns are not unique"):
+                f(pairwise_frames, pairwise_other_frame)
+            with pytest.raises(ValueError, match="'arg2' columns are not unique"):
+                f(pairwise_other_frame, pairwise_frames)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            lambda x, y: x.expanding().cov(y),
+            lambda x, y: x.expanding().corr(y),
+            lambda x, y: x.rolling(window=3).cov(y),
+            lambda x, y: x.rolling(window=3).corr(y),
+            lambda x, y: x.ewm(com=3).cov(y),
+            lambda x, y: x.ewm(com=3).corr(y),
+        ],
+    )
+    def test_pairwise_with_series(self, pairwise_frames, pairwise_target_frame, f):
+        # DataFrame with a Series
+        result = f(pairwise_frames, Series([1, 1, 3, 8]))
+        tm.assert_index_equal(result.index, pairwise_frames.index)
+        tm.assert_index_equal(result.columns, pairwise_frames.columns)
+        expected = f(pairwise_target_frame, Series([1, 1, 3, 8]))
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+        result = f(Series([1, 1, 3, 8]), pairwise_frames)
+        tm.assert_index_equal(result.index, pairwise_frames.index)
+        tm.assert_index_equal(result.columns, pairwise_frames.columns)
+        expected = f(Series([1, 1, 3, 8]), pairwise_target_frame)
+        # since we have sorted the results
+        # we can only compare non-nans
+        result = result.dropna().values
+        expected = expected.dropna().values
+        tm.assert_numpy_array_equal(result, expected, check_dtype=False)
+
+    def test_corr_freq_memory_error(self):
+        # GH 31789
+        s = Series(range(5), index=date_range("2020", periods=5))
+        result = s.rolling("12h").corr(s)
+        expected = Series([np.nan] * 5, index=date_range("2020", periods=5))
+        tm.assert_series_equal(result, expected)
+
+    def test_cov_mulittindex(self):
+        # GH 34440
+
+        columns = MultiIndex.from_product([list("ab"), list("xy"), list("AB")])
+        index = range(3)
+        df = DataFrame(np.arange(24).reshape(3, 8), index=index, columns=columns)
+
+        result = df.ewm(alpha=0.1).cov()
+
+        index = MultiIndex.from_product([range(3), list("ab"), list("xy"), list("AB")])
+        columns = MultiIndex.from_product([list("ab"), list("xy"), list("AB")])
+        expected = DataFrame(
+            np.vstack(
+                (
+                    np.full((8, 8), np.nan),
+                    np.full((8, 8), 32.000000),
+                    np.full((8, 8), 63.881919),
+                )
+            ),
+            index=index,
+            columns=columns,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_multindex_columns_pairwise_func(self):
+        # GH 21157
+        columns = MultiIndex.from_arrays([["M", "N"], ["P", "Q"]], names=["a", "b"])
+        df = DataFrame(np.ones((5, 2)), columns=columns)
+        result = df.rolling(3).corr()
+        expected = DataFrame(
+            np.nan,
+            index=MultiIndex.from_arrays(
+                [
+                    np.repeat(np.arange(5, dtype=np.int64), 2),
+                    ["M", "N"] * 5,
+                    ["P", "Q"] * 5,
+                ],
+                names=[None, "a", "b"],
+            ),
+            columns=columns,
+        )
+        tm.assert_frame_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_rolling.py b/pandas/pandas/tests/window/test_rolling.py
new file mode 100644
index 0000000000000..2aaa35ec5ec2c
--- /dev/null
+++ b/pandas/pandas/tests/window/test_rolling.py
@@ -0,0 +1,1911 @@
+from datetime import (
+    datetime,
+    timedelta,
+)
+
+import numpy as np
+import pytest
+
+from pandas.compat import (
+    IS64,
+    is_platform_arm,
+    is_platform_power,
+    is_platform_riscv64,
+)
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    MultiIndex,
+    Series,
+    Timedelta,
+    Timestamp,
+    date_range,
+    period_range,
+)
+import pandas._testing as tm
+from pandas.api.indexers import BaseIndexer
+from pandas.core.indexers.objects import VariableOffsetWindowIndexer
+
+from pandas.tseries.offsets import BusinessDay
+
+
+def test_doc_string():
+    df = DataFrame({"B": [0, 1, 2, np.nan, 4]})
+    df
+    df.rolling(2).sum()
+    df.rolling(2, min_periods=1).sum()
+
+
+def test_constructor(frame_or_series):
+    # GH 12669
+
+    c = frame_or_series(range(5)).rolling
+
+    # valid
+    c(0)
+    c(window=2)
+    c(window=2, min_periods=1)
+    c(window=2, min_periods=1, center=True)
+    c(window=2, min_periods=1, center=False)
+
+    # GH 13383
+
+    msg = "window must be an integer 0 or greater"
+
+    with pytest.raises(ValueError, match=msg):
+        c(-1)
+
+
+@pytest.mark.parametrize("w", [2.0, "foo", np.array([2])])
+def test_invalid_constructor(frame_or_series, w):
+    # not valid
+
+    c = frame_or_series(range(5)).rolling
+
+    msg = "|".join(
+        [
+            "window must be an integer",
+            "passed window foo is not compatible with a datetimelike index",
+        ]
+    )
+    with pytest.raises(ValueError, match=msg):
+        c(window=w)
+
+    msg = "min_periods must be an integer"
+    with pytest.raises(ValueError, match=msg):
+        c(window=2, min_periods=w)
+
+    msg = "center must be a boolean"
+    with pytest.raises(ValueError, match=msg):
+        c(window=2, min_periods=1, center=w)
+
+
+@pytest.mark.parametrize(
+    "window",
+    [
+        timedelta(days=3),
+        Timedelta(days=3),
+        "3D",
+        VariableOffsetWindowIndexer(
+            index=date_range("2015-12-25", periods=5), offset=BusinessDay(1)
+        ),
+    ],
+)
+def test_freq_window_not_implemented(window):
+    # GH 15354
+    df = DataFrame(
+        np.arange(10),
+        index=date_range("2015-12-24", periods=10, freq="D"),
+    )
+    with pytest.raises(
+        NotImplementedError, match="^step (not implemented|is not supported)"
+    ):
+        df.rolling(window, step=3).sum()
+
+
+@pytest.mark.parametrize("agg", ["cov", "corr"])
+def test_step_not_implemented_for_cov_corr(agg):
+    # GH 15354
+    roll = DataFrame(range(2)).rolling(1, step=2)
+    with pytest.raises(NotImplementedError, match="step not implemented"):
+        getattr(roll, agg)()
+
+
+@pytest.mark.parametrize("window", [timedelta(days=3), Timedelta(days=3)])
+def test_constructor_with_timedelta_window(window):
+    # GH 15440
+    n = 10
+    df = DataFrame(
+        {"value": np.arange(n)},
+        index=date_range("2015-12-24", periods=n, freq="D"),
+    )
+    expected_data = np.append([0.0, 1.0], np.arange(3.0, 27.0, 3))
+
+    result = df.rolling(window=window).sum()
+    expected = DataFrame(
+        {"value": expected_data},
+        index=date_range("2015-12-24", periods=n, freq="D"),
+    )
+    tm.assert_frame_equal(result, expected)
+    expected = df.rolling("3D").sum()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("window", [timedelta(days=3), Timedelta(days=3), "3D"])
+def test_constructor_timedelta_window_and_minperiods(window, raw):
+    # GH 15305
+    n = 10
+    df = DataFrame(
+        {"value": np.arange(n)},
+        index=date_range("2017-08-08", periods=n, freq="D"),
+    )
+    expected = DataFrame(
+        {"value": np.append([np.nan, 1.0], np.arange(3.0, 27.0, 3))},
+        index=date_range("2017-08-08", periods=n, freq="D"),
+    )
+    result_roll_sum = df.rolling(window=window, min_periods=2).sum()
+    result_roll_generic = df.rolling(window=window, min_periods=2).apply(sum, raw=raw)
+    tm.assert_frame_equal(result_roll_sum, expected)
+    tm.assert_frame_equal(result_roll_generic, expected)
+
+
+def test_closed_fixed(closed, arithmetic_win_operators):
+    # GH 34315
+    func_name = arithmetic_win_operators
+    df_fixed = DataFrame({"A": [0, 1, 2, 3, 4]})
+    df_time = DataFrame({"A": [0, 1, 2, 3, 4]}, index=date_range("2020", periods=5))
+
+    result = getattr(
+        df_fixed.rolling(2, closed=closed, min_periods=1),
+        func_name,
+    )()
+    expected = getattr(
+        df_time.rolling("2D", closed=closed, min_periods=1),
+        func_name,
+    )().reset_index(drop=True)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "closed, window_selections",
+    [
+        (
+            "both",
+            [
+                [True, True, False, False, False],
+                [True, True, True, False, False],
+                [False, True, True, True, False],
+                [False, False, True, True, True],
+                [False, False, False, True, True],
+            ],
+        ),
+        (
+            "left",
+            [
+                [True, False, False, False, False],
+                [True, True, False, False, False],
+                [False, True, True, False, False],
+                [False, False, True, True, False],
+                [False, False, False, True, True],
+            ],
+        ),
+        (
+            "right",
+            [
+                [True, True, False, False, False],
+                [False, True, True, False, False],
+                [False, False, True, True, False],
+                [False, False, False, True, True],
+                [False, False, False, False, True],
+            ],
+        ),
+        (
+            "neither",
+            [
+                [True, False, False, False, False],
+                [False, True, False, False, False],
+                [False, False, True, False, False],
+                [False, False, False, True, False],
+                [False, False, False, False, True],
+            ],
+        ),
+    ],
+)
+def test_datetimelike_centered_selections(
+    closed, window_selections, arithmetic_win_operators
+):
+    # GH 34315
+    func_name = arithmetic_win_operators
+    df_time = DataFrame(
+        {"A": [0.0, 1.0, 2.0, 3.0, 4.0]}, index=date_range("2020", periods=5)
+    )
+
+    expected = DataFrame(
+        {"A": [getattr(df_time["A"].iloc[s], func_name)() for s in window_selections]},
+        index=date_range("2020", periods=5),
+    )
+
+    if func_name == "sem":
+        kwargs = {"ddof": 0}
+    else:
+        kwargs = {}
+
+    result = getattr(
+        df_time.rolling("2D", closed=closed, min_periods=1, center=True),
+        func_name,
+    )(**kwargs)
+
+    tm.assert_frame_equal(result, expected, check_dtype=False)
+
+
+@pytest.mark.parametrize(
+    "window,closed,expected",
+    [
+        ("3s", "right", [3.0, 3.0, 3.0]),
+        ("3s", "both", [3.0, 3.0, 3.0]),
+        ("3s", "left", [3.0, 3.0, 3.0]),
+        ("3s", "neither", [3.0, 3.0, 3.0]),
+        ("2s", "right", [3.0, 2.0, 2.0]),
+        ("2s", "both", [3.0, 3.0, 3.0]),
+        ("2s", "left", [1.0, 3.0, 3.0]),
+        ("2s", "neither", [1.0, 2.0, 2.0]),
+    ],
+)
+def test_datetimelike_centered_offset_covers_all(
+    window, closed, expected, frame_or_series
+):
+    # GH 42753
+
+    index = [
+        Timestamp("20130101 09:00:01"),
+        Timestamp("20130101 09:00:02"),
+        Timestamp("20130101 09:00:02"),
+    ]
+    df = frame_or_series([1, 1, 1], index=index)
+
+    result = df.rolling(window, closed=closed, center=True).sum()
+    expected = frame_or_series(expected, index=index)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "window,closed,expected",
+    [
+        ("2D", "right", [4, 4, 4, 4, 4, 4, 2, 2]),
+        ("2D", "left", [2, 2, 4, 4, 4, 4, 4, 4]),
+        ("2D", "both", [4, 4, 6, 6, 6, 6, 4, 4]),
+        ("2D", "neither", [2, 2, 2, 2, 2, 2, 2, 2]),
+    ],
+)
+def test_datetimelike_nonunique_index_centering(
+    window, closed, expected, frame_or_series
+):
+    index = DatetimeIndex(
+        [
+            "2020-01-01",
+            "2020-01-01",
+            "2020-01-02",
+            "2020-01-02",
+            "2020-01-03",
+            "2020-01-03",
+            "2020-01-04",
+            "2020-01-04",
+        ]
+    )
+
+    df = frame_or_series([1] * 8, index=index, dtype=float)
+    expected = frame_or_series(expected, index=index, dtype=float)
+
+    result = df.rolling(window, center=True, closed=closed).sum()
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "closed,expected",
+    [
+        ("left", [np.nan, np.nan, 1, 1, 1, 10, 14, 14, 18, 21]),
+        ("neither", [np.nan, np.nan, 1, 1, 1, 9, 5, 5, 13, 8]),
+        ("right", [0, 1, 3, 6, 10, 14, 11, 18, 21, 17]),
+        ("both", [0, 1, 3, 6, 10, 15, 20, 27, 26, 30]),
+    ],
+)
+def test_variable_window_nonunique(closed, expected, frame_or_series):
+    # GH 20712
+    index = DatetimeIndex(
+        [
+            "2011-01-01",
+            "2011-01-01",
+            "2011-01-02",
+            "2011-01-02",
+            "2011-01-02",
+            "2011-01-03",
+            "2011-01-04",
+            "2011-01-04",
+            "2011-01-05",
+            "2011-01-06",
+        ]
+    )
+
+    df = frame_or_series(range(10), index=index, dtype=float)
+    expected = frame_or_series(expected, index=index, dtype=float)
+
+    result = df.rolling("2D", closed=closed).sum()
+
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "closed,expected",
+    [
+        ("left", [np.nan, np.nan, 1, 1, 1, 10, 15, 15, 18, 21]),
+        ("neither", [np.nan, np.nan, 1, 1, 1, 10, 15, 15, 13, 8]),
+        ("right", [0, 1, 3, 6, 10, 15, 21, 28, 21, 17]),
+        ("both", [0, 1, 3, 6, 10, 15, 21, 28, 26, 30]),
+    ],
+)
+def test_variable_offset_window_nonunique(closed, expected, frame_or_series):
+    # GH 20712
+    index = DatetimeIndex(
+        [
+            "2011-01-01",
+            "2011-01-01",
+            "2011-01-02",
+            "2011-01-02",
+            "2011-01-02",
+            "2011-01-03",
+            "2011-01-04",
+            "2011-01-04",
+            "2011-01-05",
+            "2011-01-06",
+        ]
+    )
+
+    df = frame_or_series(range(10), index=index, dtype=float)
+    expected = frame_or_series(expected, index=index, dtype=float)
+
+    offset = BusinessDay(2)
+    indexer = VariableOffsetWindowIndexer(index=index, offset=offset)
+    result = df.rolling(indexer, closed=closed, min_periods=1).sum()
+
+    tm.assert_equal(result, expected)
+
+
+def test_even_number_window_alignment():
+    # see discussion in GH 38780
+    s = Series(range(3), index=date_range(start="2020-01-01", freq="D", periods=3))
+
+    # behavior of index- and datetime-based windows differs here!
+    # s.rolling(window=2, min_periods=1, center=True).mean()
+
+    result = s.rolling(window="2D", min_periods=1, center=True).mean()
+
+    expected = Series([0.5, 1.5, 2], index=s.index)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_closed_fixed_binary_col(center, step):
+    # GH 34315
+    data = [0, 1, 1, 0, 0, 1, 0, 1]
+    df = DataFrame(
+        {"binary_col": data},
+        index=date_range(start="2020-01-01", freq="min", periods=len(data)),
+    )
+
+    if center:
+        expected_data = [2 / 3, 0.5, 0.4, 0.5, 0.428571, 0.5, 0.571429, 0.5]
+    else:
+        expected_data = [np.nan, 0, 0.5, 2 / 3, 0.5, 0.4, 0.5, 0.428571]
+
+    expected = DataFrame(
+        expected_data,
+        columns=["binary_col"],
+        index=date_range(start="2020-01-01", freq="min", periods=len(expected_data)),
+    )[::step]
+
+    rolling = df.rolling(
+        window=len(df), closed="left", min_periods=1, center=center, step=step
+    )
+    result = rolling.mean()
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("closed", ["neither", "left"])
+def test_closed_empty(closed, arithmetic_win_operators):
+    # GH 26005
+    func_name = arithmetic_win_operators
+    ser = Series(data=np.arange(5), index=date_range("2000", periods=5, freq="2D"))
+    roll = ser.rolling("1D", closed=closed)
+
+    result = getattr(roll, func_name)()
+    expected = Series([np.nan] * 5, index=ser.index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_closed_one_entry(func):
+    # GH24718
+    ser = Series(data=[2], index=date_range("2000", periods=1))
+    result = getattr(ser.rolling("10D", closed="left"), func)()
+    tm.assert_series_equal(result, Series([np.nan], index=ser.index))
+
+
+@pytest.mark.parametrize("func", ["min", "max"])
+def test_closed_one_entry_groupby(func):
+    # GH24718
+    ser = DataFrame(
+        data={"A": [1, 1, 2], "B": [3, 2, 1]},
+        index=date_range("2000", periods=3),
+    )
+    result = getattr(
+        ser.groupby("A", sort=False)["B"].rolling("10D", closed="left"), func
+    )()
+    exp_idx = MultiIndex.from_arrays(arrays=[[1, 1, 2], ser.index], names=("A", None))
+    expected = Series(data=[np.nan, 3, np.nan], index=exp_idx, name="B")
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("input_dtype", ["int", "float"])
+@pytest.mark.parametrize(
+    "func,closed,expected",
+    [
+        ("min", "right", [0.0, 0, 0, 1, 2, 3, 4, 5, 6, 7]),
+        ("min", "both", [0.0, 0, 0, 0, 1, 2, 3, 4, 5, 6]),
+        ("min", "neither", [np.nan, 0, 0, 1, 2, 3, 4, 5, 6, 7]),
+        ("min", "left", [np.nan, 0, 0, 0, 1, 2, 3, 4, 5, 6]),
+        ("max", "right", [0.0, 1, 2, 3, 4, 5, 6, 7, 8, 9]),
+        ("max", "both", [0.0, 1, 2, 3, 4, 5, 6, 7, 8, 9]),
+        ("max", "neither", [np.nan, 0, 1, 2, 3, 4, 5, 6, 7, 8]),
+        ("max", "left", [np.nan, 0, 1, 2, 3, 4, 5, 6, 7, 8]),
+    ],
+)
+def test_closed_min_max_datetime(input_dtype, func, closed, expected):
+    # see gh-21704
+    ser = Series(
+        data=np.arange(10).astype(input_dtype),
+        index=date_range("2000", periods=10),
+    )
+
+    result = getattr(ser.rolling("3D", closed=closed), func)()
+    expected = Series(expected, index=ser.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_closed_uneven():
+    # see gh-21704
+    ser = Series(data=np.arange(10), index=date_range("2000", periods=10))
+
+    # uneven
+    ser = ser.drop(index=ser.index[[1, 5]])
+    result = ser.rolling("3D", closed="left").min()
+    expected = Series([np.nan, 0, 0, 2, 3, 4, 6, 6], index=ser.index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "func,closed,expected",
+    [
+        ("min", "right", [np.nan, 0, 0, 1, 2, 3, 4, 5, np.nan, np.nan]),
+        ("min", "both", [np.nan, 0, 0, 0, 1, 2, 3, 4, 5, np.nan]),
+        ("min", "neither", [np.nan, np.nan, 0, 1, 2, 3, 4, 5, np.nan, np.nan]),
+        ("min", "left", [np.nan, np.nan, 0, 0, 1, 2, 3, 4, 5, np.nan]),
+        ("max", "right", [np.nan, 1, 2, 3, 4, 5, 6, 6, np.nan, np.nan]),
+        ("max", "both", [np.nan, 1, 2, 3, 4, 5, 6, 6, 6, np.nan]),
+        ("max", "neither", [np.nan, np.nan, 1, 2, 3, 4, 5, 6, np.nan, np.nan]),
+        ("max", "left", [np.nan, np.nan, 1, 2, 3, 4, 5, 6, 6, np.nan]),
+    ],
+)
+def test_closed_min_max_minp(func, closed, expected):
+    # see gh-21704
+    ser = Series(data=np.arange(10), index=date_range("2000", periods=10))
+    # Explicit cast to float to avoid implicit cast when setting nan
+    ser = ser.astype("float")
+    ser[ser.index[-3:]] = np.nan
+    result = getattr(ser.rolling("3D", min_periods=2, closed=closed), func)()
+    expected = Series(expected, index=ser.index)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "closed,expected",
+    [
+        ("right", [0, 0.5, 1, 2, 3, 4, 5, 6, 7, 8]),
+        ("both", [0, 0.5, 1, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5]),
+        ("neither", [np.nan, 0, 0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5]),
+        ("left", [np.nan, 0, 0.5, 1, 2, 3, 4, 5, 6, 7]),
+    ],
+)
+def test_closed_median_quantile(closed, expected):
+    # GH 26005
+    ser = Series(data=np.arange(10), index=date_range("2000", periods=10))
+    roll = ser.rolling("3D", closed=closed)
+    expected = Series(expected, index=ser.index)
+
+    result = roll.median()
+    tm.assert_series_equal(result, expected)
+
+    result = roll.quantile(0.5)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("roller", ["1s", 1])
+def tests_empty_df_rolling(roller):
+    # GH 15819 Verifies that datetime and integer rolling windows can be
+    # applied to empty DataFrames
+    expected = DataFrame()
+    result = DataFrame().rolling(roller).sum()
+    tm.assert_frame_equal(result, expected)
+
+    # Verifies that datetime and integer rolling windows can be applied to
+    # empty DataFrames with datetime index
+    expected = DataFrame(index=DatetimeIndex([]))
+    result = DataFrame(index=DatetimeIndex([])).rolling(roller).sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_empty_window_median_quantile():
+    # GH 26005
+    expected = Series([np.nan, np.nan, np.nan])
+    roll = Series(np.arange(3)).rolling(0)
+
+    result = roll.median()
+    tm.assert_series_equal(result, expected)
+
+    result = roll.quantile(0.1)
+    tm.assert_series_equal(result, expected)
+
+
+def test_missing_minp_zero():
+    # https://github.com/pandas-dev/pandas/pull/18921
+    # minp=0
+    x = Series([np.nan])
+    result = x.rolling(1, min_periods=0).sum()
+    expected = Series([0.0])
+    tm.assert_series_equal(result, expected)
+
+    # minp=1
+    result = x.rolling(1, min_periods=1).sum()
+    expected = Series([np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+def test_missing_minp_zero_variable():
+    # https://github.com/pandas-dev/pandas/pull/18921
+    x = Series(
+        [np.nan] * 4,
+        index=DatetimeIndex(["2017-01-01", "2017-01-04", "2017-01-06", "2017-01-07"]),
+    )
+    result = x.rolling(Timedelta("2D"), min_periods=0).sum()
+    expected = Series(0.0, index=x.index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_multi_index_names():
+    # GH 16789, 16825
+    cols = MultiIndex.from_product([["A", "B"], ["C", "D", "E"]], names=["1", "2"])
+    df = DataFrame(np.ones((10, 6)), columns=cols)
+    result = df.rolling(3).cov()
+
+    tm.assert_index_equal(result.columns, df.columns)
+    assert result.index.names == [None, "1", "2"]
+
+
+def test_rolling_axis_sum():
+    # see gh-23372.
+    df = DataFrame(np.ones((10, 20)))
+    expected = DataFrame({i: [np.nan] * 2 + [3.0] * 8 for i in range(20)})
+    result = df.rolling(3).sum()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_axis_count():
+    # see gh-26055
+    df = DataFrame({"x": range(3), "y": range(3)})
+
+    expected = DataFrame({"x": [1.0, 2.0, 2.0], "y": [1.0, 2.0, 2.0]})
+    result = df.rolling(2, min_periods=0).count()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_readonly_array():
+    # GH-27766
+    arr = np.array([1, 3, np.nan, 3, 5])
+    arr.setflags(write=False)
+    result = Series(arr).rolling(2).mean()
+    expected = Series([np.nan, 2, np.nan, np.nan, 4])
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_datetime(tz_naive_fixture):
+    # GH-28192
+    tz = tz_naive_fixture
+    df = DataFrame(
+        {i: [1] * 2 for i in date_range("2019-8-01", "2019-08-03", freq="D", tz=tz)}
+    )
+
+    result = df.T.rolling("2D").sum().T
+    expected = DataFrame(
+        {
+            **{
+                i: [1.0] * 2
+                for i in date_range("2019-8-01", periods=1, freq="D", tz=tz)
+            },
+            **{
+                i: [2.0] * 2
+                for i in date_range("2019-8-02", "2019-8-03", freq="D", tz=tz)
+            },
+        }
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_window_as_string(center):
+    # see gh-22590
+    date_today = datetime.now()
+    days = date_range(date_today, date_today + timedelta(365), freq="D")
+
+    data = np.ones(len(days))
+    df = DataFrame({"DateCol": days, "metric": data})
+
+    df.set_index("DateCol", inplace=True)
+    result = df.rolling(window="21D", min_periods=2, closed="left", center=center)[
+        "metric"
+    ].agg("max")
+
+    index = days.rename("DateCol")
+    index = index._with_freq(None)
+    expected_data = np.ones(len(days), dtype=np.float64)
+    if not center:
+        expected_data[:2] = np.nan
+    expected = Series(expected_data, index=index, name="metric")
+    tm.assert_series_equal(result, expected)
+
+
+def test_min_periods1():
+    # GH#6795
+    df = DataFrame([0, 1, 2, 1, 0], columns=["a"])
+    result = df["a"].rolling(3, center=True, min_periods=1).max()
+    expected = Series([1.0, 2.0, 2.0, 2.0, 1.0], name="a")
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_count_with_min_periods(frame_or_series):
+    # GH 26996
+    result = frame_or_series(range(5)).rolling(3, min_periods=3).count()
+    expected = frame_or_series([np.nan, np.nan, 3.0, 3.0, 3.0])
+    tm.assert_equal(result, expected)
+
+
+def test_rolling_count_default_min_periods_with_null_values(frame_or_series):
+    # GH 26996
+    values = [1, 2, 3, np.nan, 4, 5, 6]
+    expected_counts = [1.0, 2.0, 3.0, 2.0, 2.0, 2.0, 3.0]
+
+    # GH 31302
+    result = frame_or_series(values).rolling(3, min_periods=0).count()
+    expected = frame_or_series(expected_counts)
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "df,expected,window,min_periods",
+    [
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [1, 2, 3], "B": [4, 5, 6]}, [0, 1, 2]),
+            ],
+            3,
+            None,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [2, 3], "B": [5, 6]}, [1, 2]),
+            ],
+            2,
+            1,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [2, 3], "B": [5, 6]}, [1, 2]),
+            ],
+            2,
+            2,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [2], "B": [5]}, [1]),
+                ({"A": [3], "B": [6]}, [2]),
+            ],
+            1,
+            1,
+        ),
+        (
+            {"A": [1, 2, 3], "B": [4, 5, 6]},
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [2], "B": [5]}, [1]),
+                ({"A": [3], "B": [6]}, [2]),
+            ],
+            1,
+            0,
+        ),
+        ({"A": [1], "B": [4]}, [], 2, None),
+        ({"A": [1], "B": [4]}, [], 2, 1),
+        (None, [({}, [])], 2, None),
+        (
+            {"A": [1, np.nan, 3], "B": [np.nan, 5, 6]},
+            [
+                ({"A": [1.0], "B": [np.nan]}, [0]),
+                ({"A": [1, np.nan], "B": [np.nan, 5]}, [0, 1]),
+                ({"A": [1, np.nan, 3], "B": [np.nan, 5, 6]}, [0, 1, 2]),
+            ],
+            3,
+            2,
+        ),
+    ],
+)
+def test_iter_rolling_dataframe(df, expected, window, min_periods):
+    # GH 11704
+    df = DataFrame(df)
+    expecteds = [DataFrame(values, index=index) for (values, index) in expected]
+
+    for expected, actual in zip(expecteds, df.rolling(window, min_periods=min_periods)):
+        tm.assert_frame_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "expected,window",
+    [
+        (
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [2, 3], "B": [5, 6]}, [1, 2]),
+            ],
+            "2D",
+        ),
+        (
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [1, 2], "B": [4, 5]}, [0, 1]),
+                ({"A": [1, 2, 3], "B": [4, 5, 6]}, [0, 1, 2]),
+            ],
+            "3D",
+        ),
+        (
+            [
+                ({"A": [1], "B": [4]}, [0]),
+                ({"A": [2], "B": [5]}, [1]),
+                ({"A": [3], "B": [6]}, [2]),
+            ],
+            "1D",
+        ),
+    ],
+)
+def test_iter_rolling_on_dataframe(expected, window):
+    # GH 11704, 40373
+    df = DataFrame(
+        {
+            "A": [1, 2, 3, 4, 5],
+            "B": [4, 5, 6, 7, 8],
+            "C": date_range(start="2016-01-01", periods=5, freq="D"),
+        }
+    )
+
+    expecteds = [
+        DataFrame(values, index=df.loc[index, "C"]) for (values, index) in expected
+    ]
+    for expected, actual in zip(expecteds, df.rolling(window, on="C")):
+        tm.assert_frame_equal(actual, expected)
+
+
+def test_iter_rolling_on_dataframe_unordered():
+    # GH 43386
+    df = DataFrame({"a": ["x", "y", "x"], "b": [0, 1, 2]})
+    results = list(df.groupby("a").rolling(2))
+    expecteds = [df.iloc[idx, [1]] for idx in [[0], [0, 2], [1]]]
+    for result, expected in zip(results, expecteds):
+        tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "ser,expected,window, min_periods",
+    [
+        (
+            Series([1, 2, 3]),
+            [([1], [0]), ([1, 2], [0, 1]), ([1, 2, 3], [0, 1, 2])],
+            3,
+            None,
+        ),
+        (
+            Series([1, 2, 3]),
+            [([1], [0]), ([1, 2], [0, 1]), ([1, 2, 3], [0, 1, 2])],
+            3,
+            1,
+        ),
+        (
+            Series([1, 2, 3]),
+            [([1], [0]), ([1, 2], [0, 1]), ([2, 3], [1, 2])],
+            2,
+            1,
+        ),
+        (
+            Series([1, 2, 3]),
+            [([1], [0]), ([1, 2], [0, 1]), ([2, 3], [1, 2])],
+            2,
+            2,
+        ),
+        (Series([1, 2, 3]), [([1], [0]), ([2], [1]), ([3], [2])], 1, 0),
+        (Series([1, 2, 3]), [([1], [0]), ([2], [1]), ([3], [2])], 1, 1),
+        (Series([1, 2]), [([1], [0]), ([1, 2], [0, 1])], 2, 0),
+        (Series([], dtype="int64"), [], 2, 1),
+    ],
+)
+def test_iter_rolling_series(ser, expected, window, min_periods):
+    # GH 11704
+    expecteds = [Series(values, index=index) for (values, index) in expected]
+
+    for expected, actual in zip(
+        expecteds, ser.rolling(window, min_periods=min_periods)
+    ):
+        tm.assert_series_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "expected,expected_index,window",
+    [
+        (
+            [[0], [1], [2], [3], [4]],
+            [
+                date_range("2020-01-01", periods=1, freq="D"),
+                date_range("2020-01-02", periods=1, freq="D"),
+                date_range("2020-01-03", periods=1, freq="D"),
+                date_range("2020-01-04", periods=1, freq="D"),
+                date_range("2020-01-05", periods=1, freq="D"),
+            ],
+            "1D",
+        ),
+        (
+            [[0], [0, 1], [1, 2], [2, 3], [3, 4]],
+            [
+                date_range("2020-01-01", periods=1, freq="D"),
+                date_range("2020-01-01", periods=2, freq="D"),
+                date_range("2020-01-02", periods=2, freq="D"),
+                date_range("2020-01-03", periods=2, freq="D"),
+                date_range("2020-01-04", periods=2, freq="D"),
+            ],
+            "2D",
+        ),
+        (
+            [[0], [0, 1], [0, 1, 2], [1, 2, 3], [2, 3, 4]],
+            [
+                date_range("2020-01-01", periods=1, freq="D"),
+                date_range("2020-01-01", periods=2, freq="D"),
+                date_range("2020-01-01", periods=3, freq="D"),
+                date_range("2020-01-02", periods=3, freq="D"),
+                date_range("2020-01-03", periods=3, freq="D"),
+            ],
+            "3D",
+        ),
+    ],
+)
+def test_iter_rolling_datetime(expected, expected_index, window):
+    # GH 11704
+    ser = Series(range(5), index=date_range(start="2020-01-01", periods=5, freq="D"))
+
+    expecteds = [
+        Series(values, index=idx) for (values, idx) in zip(expected, expected_index)
+    ]
+
+    for expected, actual in zip(expecteds, ser.rolling(window)):
+        tm.assert_series_equal(actual, expected)
+
+
+@pytest.mark.parametrize(
+    "grouping,_index",
+    [
+        (
+            {"level": 0},
+            MultiIndex.from_tuples(
+                [(0, 0), (0, 0), (1, 1), (1, 1), (1, 1)], names=[None, None]
+            ),
+        ),
+        (
+            {"by": "X"},
+            MultiIndex.from_tuples(
+                [(0, 0), (1, 0), (2, 1), (3, 1), (4, 1)], names=["X", None]
+            ),
+        ),
+    ],
+)
+def test_rolling_positional_argument(grouping, _index, raw):
+    # GH 34605
+
+    def scaled_sum(*args):
+        if len(args) < 2:
+            raise ValueError("The function needs two arguments")
+        array, scale = args
+        return array.sum() / scale
+
+    df = DataFrame(data={"X": range(5)}, index=[0, 0, 1, 1, 1])
+
+    expected = DataFrame(data={"X": [0.0, 0.5, 1.0, 1.5, 2.0]}, index=_index)
+    # GH 40341
+    if "by" in grouping:
+        expected = expected.drop(columns="X", errors="ignore")
+    result = df.groupby(**grouping).rolling(1).apply(scaled_sum, raw=raw, args=(2,))
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("add", [0.0, 2.0])
+def test_rolling_numerical_accuracy_kahan_mean(add, unit):
+    # GH: 36031 implementing kahan summation
+    dti = DatetimeIndex(
+        [
+            Timestamp("19700101 09:00:00"),
+            Timestamp("19700101 09:00:03"),
+            Timestamp("19700101 09:00:06"),
+        ]
+    ).as_unit(unit)
+    df = DataFrame(
+        {"A": [3002399751580331.0 + add, -0.0, -0.0]},
+        index=dti,
+    )
+    result = (
+        df.resample("1s").ffill().rolling("3s", closed="left", min_periods=3).mean()
+    )
+    dates = date_range("19700101 09:00:00", periods=7, freq="s", unit=unit)
+    expected = DataFrame(
+        {
+            "A": [
+                np.nan,
+                np.nan,
+                np.nan,
+                3002399751580330.5,
+                2001599834386887.25,
+                1000799917193443.625,
+                0.0,
+            ]
+        },
+        index=dates,
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_numerical_accuracy_kahan_sum():
+    # GH: 13254
+    df = DataFrame([2.186, -1.647, 0.0, 0.0, 0.0, 0.0], columns=["x"])
+    result = df["x"].rolling(3).sum()
+    expected = Series([np.nan, np.nan, 0.539, -1.647, 0.0, 0.0], name="x")
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_numerical_accuracy_jump():
+    # GH: 32761
+    index = date_range(start="2020-01-01", end="2020-01-02", freq="60s").append(
+        DatetimeIndex(["2020-01-03"])
+    )
+    data = np.random.default_rng(2).random(len(index))
+
+    df = DataFrame({"data": data}, index=index)
+    result = df.rolling("60s").mean()
+    tm.assert_frame_equal(result, df[["data"]])
+
+
+def test_rolling_numerical_accuracy_small_values():
+    # GH: 10319
+    s = Series(
+        data=[0.00012456, 0.0003, -0.0, -0.0],
+        index=date_range("1999-02-03", "1999-02-06"),
+    )
+    result = s.rolling(1).mean()
+    tm.assert_series_equal(result, s)
+
+
+def test_rolling_numerical_too_large_numbers():
+    # GH: 11645
+    dates = date_range("2015-01-01", periods=10, freq="D")
+    ds = Series(data=range(10), index=dates, dtype=np.float64)
+    ds.iloc[2] = -9e33
+    result = ds.rolling(5).mean()
+    expected = Series(
+        [
+            np.nan,
+            np.nan,
+            np.nan,
+            np.nan,
+            -1.8e33,
+            -1.8e33,
+            -1.8e33,
+            5.0,
+            6.0,
+            7.0,
+        ],
+        index=dates,
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("index", "window"),
+    [
+        (
+            period_range(start="2020-01-01 08:00", end="2020-01-01 08:08", freq="min"),
+            "2min",
+        ),
+        (
+            period_range(
+                start="2020-01-01 08:00", end="2020-01-01 12:00", freq="30min"
+            ),
+            "1h",
+        ),
+    ],
+)
+@pytest.mark.parametrize(
+    ("func", "values"),
+    [
+        ("min", [np.nan, 0, 0, 1, 2, 3, 4, 5, 6]),
+        ("max", [np.nan, 0, 1, 2, 3, 4, 5, 6, 7]),
+        ("sum", [np.nan, 0, 1, 3, 5, 7, 9, 11, 13]),
+    ],
+)
+def test_rolling_period_index(index, window, func, values):
+    # GH: 34225
+    ds = Series([0, 1, 2, 3, 4, 5, 6, 7, 8], index=index)
+    result = getattr(ds.rolling(window, closed="left"), func)()
+    expected = Series(values, index=index)
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_sem(frame_or_series):
+    # GH: 26476
+    obj = frame_or_series([0, 1, 2])
+    result = obj.rolling(2, min_periods=1).sem()
+    if isinstance(result, DataFrame):
+        result = Series(result[0].values)
+    expected = Series([np.nan] + [0.7071067811865476] * 2)
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.xfail(
+    is_platform_arm() or is_platform_power() or is_platform_riscv64(),
+    reason="GH 38921",
+)
+@pytest.mark.parametrize(
+    ("func", "third_value", "values"),
+    [
+        ("var", 1, [5e33, 0, 0.5, 0.5, 2, 0]),
+        ("std", 1, [7.071068e16, 0, 0.7071068, 0.7071068, 1.414214, 0]),
+        ("var", 2, [5e33, 0.5, 0, 0.5, 2, 0]),
+        ("std", 2, [7.071068e16, 0.7071068, 0, 0.7071068, 1.414214, 0]),
+    ],
+)
+def test_rolling_var_numerical_issues(func, third_value, values):
+    # GH: 37051
+    ds = Series([99999999999999999, 1, third_value, 2, 3, 1, 1])
+    result = getattr(ds.rolling(2), func)()
+    expected = Series([np.nan] + values)
+    tm.assert_series_equal(result, expected)
+    # GH 42064
+    # new `roll_var` will output 0.0 correctly
+    tm.assert_series_equal(result == 0, expected == 0)
+
+
+def test_timeoffset_as_window_parameter_for_corr(unit):
+    # GH: 28266
+    dti = DatetimeIndex(
+        [
+            Timestamp("20130101 09:00:00"),
+            Timestamp("20130102 09:00:02"),
+            Timestamp("20130103 09:00:03"),
+            Timestamp("20130105 09:00:05"),
+            Timestamp("20130106 09:00:06"),
+        ]
+    ).as_unit(unit)
+    mi = MultiIndex.from_product([dti, ["B", "A"]])
+
+    exp = DataFrame(
+        {
+            "B": [
+                np.nan,
+                np.nan,
+                0.9999999999999998,
+                -1.0,
+                1.0,
+                -0.3273268353539892,
+                0.9999999999999998,
+                1.0,
+                0.9999999999999998,
+                1.0,
+            ],
+            "A": [
+                np.nan,
+                np.nan,
+                -1.0,
+                1.0000000000000002,
+                -0.3273268353539892,
+                0.9999999999999966,
+                1.0,
+                1.0000000000000002,
+                1.0,
+                1.0000000000000002,
+            ],
+        },
+        index=mi,
+    )
+
+    df = DataFrame(
+        {"B": [0, 1, 2, 4, 3], "A": [7, 4, 6, 9, 3]},
+        index=dti,
+    )
+
+    res = df.rolling(window="3D").corr()
+
+    tm.assert_frame_equal(exp, res)
+
+
+@pytest.mark.parametrize("method", ["var", "sum", "mean", "skew", "kurt", "min", "max"])
+def test_rolling_decreasing_indices(method):
+    """
+    Make sure that decreasing indices give the same results as increasing indices.
+
+    GH 36933
+    """
+    df = DataFrame({"values": np.arange(-15, 10) ** 2})
+    df_reverse = DataFrame({"values": df["values"][::-1]}, index=df.index[::-1])
+
+    increasing = getattr(df.rolling(window=5), method)()
+    decreasing = getattr(df_reverse.rolling(window=5), method)()
+
+    assert np.abs(decreasing.values[::-1][:-4] - increasing.values[4:]).max() < 1e-12
+
+
+@pytest.mark.parametrize(
+    "window,closed,expected",
+    [
+        ("2s", "right", [1.0, 3.0, 5.0, 3.0]),
+        ("2s", "left", [0.0, 1.0, 3.0, 5.0]),
+        ("2s", "both", [1.0, 3.0, 6.0, 5.0]),
+        ("2s", "neither", [0.0, 1.0, 2.0, 3.0]),
+        ("3s", "right", [1.0, 3.0, 6.0, 5.0]),
+        ("3s", "left", [1.0, 3.0, 6.0, 5.0]),
+        ("3s", "both", [1.0, 3.0, 6.0, 5.0]),
+        ("3s", "neither", [1.0, 3.0, 6.0, 5.0]),
+    ],
+)
+def test_rolling_decreasing_indices_centered(window, closed, expected, frame_or_series):
+    """
+    Ensure that a symmetrical inverted index return same result as non-inverted.
+    """
+    #  GH 43927
+
+    index = date_range("2020", periods=4, freq="1s")
+    df_inc = frame_or_series(range(4), index=index)
+    df_dec = frame_or_series(range(4), index=index[::-1])
+
+    expected_inc = frame_or_series(expected, index=index)
+    expected_dec = frame_or_series(expected, index=index[::-1])
+
+    result_inc = df_inc.rolling(window, closed=closed, center=True).sum()
+    result_dec = df_dec.rolling(window, closed=closed, center=True).sum()
+
+    tm.assert_equal(result_inc, expected_inc)
+    tm.assert_equal(result_dec, expected_dec)
+
+
+@pytest.mark.parametrize(
+    "window,expected",
+    [
+        ("1ns", [1.0, 1.0, 1.0, 1.0]),
+        ("3ns", [2.0, 3.0, 3.0, 2.0]),
+    ],
+)
+def test_rolling_center_nanosecond_resolution(
+    window, closed, expected, frame_or_series
+):
+    index = date_range("2020", periods=4, freq="1ns")
+    df = frame_or_series([1, 1, 1, 1], index=index, dtype=float)
+    expected = frame_or_series(expected, index=index, dtype=float)
+    result = df.rolling(window, closed=closed, center=True).sum()
+    tm.assert_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "method,expected",
+    [
+        (
+            "var",
+            [
+                float("nan"),
+                43.0,
+                float("nan"),
+                136.333333,
+                43.5,
+                94.966667,
+                182.0,
+                318.0,
+            ],
+        ),
+        (
+            "mean",
+            [float("nan"), 7.5, float("nan"), 21.5, 6.0, 9.166667, 13.0, 17.5],
+        ),
+        (
+            "sum",
+            [float("nan"), 30.0, float("nan"), 86.0, 30.0, 55.0, 91.0, 140.0],
+        ),
+        (
+            "skew",
+            [
+                float("nan"),
+                0.709296,
+                float("nan"),
+                0.407073,
+                0.984656,
+                0.919184,
+                0.874674,
+                0.842418,
+            ],
+        ),
+        (
+            "kurt",
+            [
+                float("nan"),
+                -0.5916711736073559,
+                float("nan"),
+                -1.0028993131317954,
+                -0.06103844629409494,
+                -0.254143227116194,
+                -0.37362637362637585,
+                -0.45439658241367054,
+            ],
+        ),
+    ],
+)
+def test_rolling_non_monotonic(method, expected):
+    """
+    Make sure the (rare) branch of non-monotonic indices is covered by a test.
+
+    output from 1.1.3 is assumed to be the expected output. Output of sum/mean has
+    manually been verified.
+
+    GH 36933.
+    """
+    # Based on an example found in computation.rst
+    use_expanding = [True, False, True, False, True, True, True, True]
+    df = DataFrame({"values": np.arange(len(use_expanding)) ** 2})
+
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            start = np.empty(num_values, dtype=np.int64)
+            end = np.empty(num_values, dtype=np.int64)
+            for i in range(num_values):
+                if self.use_expanding[i]:
+                    start[i] = 0
+                    end[i] = i + 1
+                else:
+                    start[i] = i
+                    end[i] = i + self.window_size
+            return start, end
+
+    indexer = CustomIndexer(window_size=4, use_expanding=use_expanding)
+
+    result = getattr(df.rolling(indexer), method)()
+    expected = DataFrame({"values": expected})
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("index", "window"),
+    [
+        ([0, 1, 2, 3, 4], 2),
+        (date_range("2001-01-01", freq="D", periods=5), "2D"),
+    ],
+)
+def test_rolling_corr_timedelta_index(index, window):
+    # GH: 31286
+    x = Series([1, 2, 3, 4, 5], index=index)
+    y = x.copy()
+    x.iloc[0:2] = 0.0
+    result = x.rolling(window).corr(y)
+    expected = Series([np.nan, np.nan, 1, 1, 1], index=index)
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values,method,expected",
+    [
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "first",
+            [float("nan"), float("nan"), 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "first",
+            [float("nan")] * 10,
+        ),
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "last",
+            [float("nan"), float("nan"), 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "last",
+            [float("nan")] * 10,
+        ),
+    ],
+)
+def test_rolling_first_last(values, method, expected):
+    # GH#33155
+    x = Series(values)
+    result = getattr(x.rolling(3), method)()
+    expected = Series(expected)
+    tm.assert_almost_equal(result, expected)
+
+    x = DataFrame({"A": values})
+    result = getattr(x.rolling(3), method)()
+    expected = DataFrame({"A": expected})
+    tm.assert_almost_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "values,method,expected",
+    [
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "first",
+            [1.0, 1.0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "first",
+            [1.0, 1.0, 1.0, 3.0, 3.0, 5.0, 5.0, 7.0, 7.0, 9.0],
+        ),
+        (
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+            "last",
+            [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0],
+        ),
+        (
+            [1.0, np.nan, 3.0, np.nan, 5.0, np.nan, 7.0, np.nan, 9.0, np.nan],
+            "last",
+            [1.0, 1.0, 3.0, 3.0, 5.0, 5.0, 7.0, 7.0, 9.0, 9.0],
+        ),
+    ],
+)
+def test_rolling_first_last_no_minp(values, method, expected):
+    # GH#33155
+    x = Series(values)
+    result = getattr(x.rolling(3, min_periods=0), method)()
+    expected = Series(expected)
+    tm.assert_almost_equal(result, expected)
+
+    x = DataFrame({"A": values})
+    result = getattr(x.rolling(3, min_periods=0), method)()
+    expected = DataFrame({"A": expected})
+    tm.assert_almost_equal(result, expected)
+
+
+def test_groupby_rolling_nan_included():
+    # GH 35542
+    data = {"group": ["g1", np.nan, "g1", "g2", np.nan], "B": [0, 1, 2, 3, 4]}
+    df = DataFrame(data)
+    result = df.groupby("group", dropna=False).rolling(1, min_periods=1).mean()
+    expected = DataFrame(
+        {"B": [0.0, 2.0, 3.0, 1.0, 4.0]},
+        # GH-38057 from_tuples puts the NaNs in the codes, result expects them
+        # to be in the levels, at the moment
+        # index=MultiIndex.from_tuples(
+        #     [("g1", 0), ("g1", 2), ("g2", 3), (np.nan, 1), (np.nan, 4)],
+        #     names=["group", None],
+        # ),
+        index=MultiIndex(
+            [["g1", "g2", np.nan], [0, 1, 2, 3, 4]],
+            [[0, 0, 1, 2, 2], [0, 2, 3, 1, 4]],
+            names=["group", None],
+        ),
+    )
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("method", ["skew", "kurt"])
+def test_rolling_skew_kurt_numerical_stability(method):
+    # GH#6929
+    ser = Series(np.random.default_rng(2).random(10))
+    ser_copy = ser.copy()
+    expected = getattr(ser.rolling(3), method)()
+    tm.assert_series_equal(ser, ser_copy)
+    ser = ser + 50000
+    result = getattr(ser.rolling(3), method)()
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    ("method", "values"),
+    [
+        ("skew", [2.0, 0.854563, 0.0, 1.999984]),
+        ("kurt", [4.0, -1.289256, -1.2, 3.999946]),
+    ],
+)
+def test_rolling_skew_kurt_large_value_range(method, values):
+    # GH: 37557
+    s = Series([3000000, 1, 1, 2, 3, 4, 999])
+    result = getattr(s.rolling(4), method)()
+    expected = Series([np.nan] * 3 + values)
+    tm.assert_series_equal(result, expected)
+
+
+def test_invalid_method():
+    with pytest.raises(ValueError, match="method must be 'table' or 'single"):
+        Series(range(1)).rolling(1, method="foo")
+
+
+def test_rolling_descending_date_order_with_offset(frame_or_series):
+    # GH#40002
+    msg = "'d' is deprecated and will be removed in a future version."
+
+    with tm.assert_produces_warning(FutureWarning, match=msg):
+        idx = date_range(start="2020-01-01", end="2020-01-03", freq="1d")
+        obj = frame_or_series(range(1, 4), index=idx)
+        result = obj.rolling("1d", closed="left").sum()
+
+    expected = frame_or_series([np.nan, 1, 2], index=idx)
+    tm.assert_equal(result, expected)
+
+    result = obj.iloc[::-1].rolling("1D", closed="left").sum()
+    idx = date_range(start="2020-01-03", end="2020-01-01", freq="-1D")
+    expected = frame_or_series([np.nan, 3, 2], index=idx)
+    tm.assert_equal(result, expected)
+
+
+def test_rolling_var_floating_artifact_precision():
+    # GH 37051
+    s = Series([7, 5, 5, 5])
+    result = s.rolling(3).var()
+    expected = Series([np.nan, np.nan, 4 / 3, 0])
+    tm.assert_series_equal(result, expected, atol=1.0e-15, rtol=1.0e-15)
+    # GH 42064
+    # new `roll_var` will output 0.0 correctly
+    tm.assert_series_equal(result == 0, expected == 0)
+
+
+def test_rolling_std_small_values():
+    # GH 37051
+    s = Series(
+        [
+            0.00000054,
+            0.00000053,
+            0.00000054,
+        ]
+    )
+    result = s.rolling(2).std()
+    expected = Series([np.nan, 7.071068e-9, 7.071068e-9])
+    tm.assert_series_equal(result, expected, atol=1.0e-15, rtol=1.0e-15)
+
+
+@pytest.mark.parametrize(
+    "start, exp_values",
+    [
+        (1, [0.03, 0.0155, 0.0155, 0.011, 0.01025]),
+        (2, [0.001, 0.001, 0.0015, 0.00366666]),
+    ],
+)
+def test_rolling_mean_all_nan_window_floating_artifacts(start, exp_values):
+    # GH#41053
+    df = DataFrame(
+        [
+            0.03,
+            0.03,
+            0.001,
+            np.nan,
+            0.002,
+            0.008,
+            np.nan,
+            np.nan,
+            np.nan,
+            np.nan,
+            np.nan,
+            np.nan,
+            0.005,
+            0.2,
+        ]
+    )
+
+    values = exp_values + [
+        0.00366666,
+        0.005,
+        0.005,
+        0.008,
+        np.nan,
+        np.nan,
+        0.005,
+        0.102500,
+    ]
+    expected = DataFrame(
+        values,
+        index=list(range(start, len(values) + start)),
+    )
+    result = df.iloc[start:].rolling(5, min_periods=0).mean()
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_sum_all_nan_window_floating_artifacts():
+    # GH#41053
+    df = DataFrame([0.002, 0.008, 0.005, np.nan, np.nan, np.nan])
+    result = df.rolling(3, min_periods=0).sum()
+    expected = DataFrame([0.002, 0.010, 0.015, 0.013, 0.005, 0.0])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_rolling_zero_window():
+    # GH 22719
+    s = Series(range(1))
+    result = s.rolling(0).min()
+    expected = Series([np.nan])
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("window", [1, 3, 10, 20])
+@pytest.mark.parametrize("method", ["min", "max", "average"])
+@pytest.mark.parametrize("pct", [True, False])
+@pytest.mark.parametrize("test_data", ["default", "duplicates", "nans"])
+def test_rank(window, method, pct, ascending, test_data):
+    length = 20
+    if test_data == "default":
+        ser = Series(data=np.random.default_rng(2).random(length))
+    elif test_data == "duplicates":
+        ser = Series(data=np.random.default_rng(2).choice(3, length))
+    elif test_data == "nans":
+        ser = Series(
+            data=np.random.default_rng(2).choice(
+                [1.0, 0.25, 0.75, np.nan, np.inf, -np.inf], length
+            )
+        )
+
+    expected = ser.rolling(window).apply(
+        lambda x: x.rank(method=method, pct=pct, ascending=ascending).iloc[-1]
+    )
+    result = ser.rolling(window).rank(method=method, pct=pct, ascending=ascending)
+
+    tm.assert_series_equal(result, expected)
+
+
+def test_rolling_quantile_np_percentile():
+    # #9413: Tests that rolling window's quantile default behavior
+    # is analogous to Numpy's percentile
+    row = 10
+    col = 5
+    idx = date_range("20100101", periods=row, freq="B")
+    df = DataFrame(
+        np.random.default_rng(2).random(row * col).reshape((row, -1)), index=idx
+    )
+
+    df_quantile = df.quantile([0.25, 0.5, 0.75], axis=0)
+    np_percentile = np.percentile(df, [25, 50, 75], axis=0)
+
+    tm.assert_almost_equal(df_quantile.values, np.array(np_percentile))
+
+
+@pytest.mark.parametrize("quantile", [0.0, 0.1, 0.45, 0.5, 1])
+@pytest.mark.parametrize(
+    "interpolation", ["linear", "lower", "higher", "nearest", "midpoint"]
+)
+@pytest.mark.parametrize(
+    "data",
+    [
+        [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0],
+        [8.0, 1.0, 3.0, 4.0, 5.0, 2.0, 6.0, 7.0],
+        [0.0, np.nan, 0.2, np.nan, 0.4],
+        [np.nan, np.nan, np.nan, np.nan],
+        [np.nan, 0.1, np.nan, 0.3, 0.4, 0.5],
+        [0.5],
+        [np.nan, 0.7, 0.6],
+    ],
+)
+def test_rolling_quantile_interpolation_options(quantile, interpolation, data):
+    # Tests that rolling window's quantile behavior is analogous to
+    # Series' quantile for each interpolation option
+    s = Series(data)
+
+    q1 = s.quantile(quantile, interpolation)
+    q2 = s.expanding(min_periods=1).quantile(quantile, interpolation).iloc[-1]
+
+    if np.isnan(q1):
+        assert np.isnan(q2)
+    elif not IS64:
+        # Less precision on 32-bit
+        assert np.allclose([q1], [q2], rtol=1e-07, atol=0)
+    else:
+        assert q1 == q2
+
+
+def test_invalid_quantile_value():
+    data = np.arange(5)
+    s = Series(data)
+
+    msg = "Interpolation 'invalid' is not supported"
+    with pytest.raises(ValueError, match=msg):
+        s.rolling(len(data), min_periods=1).quantile(0.5, interpolation="invalid")
+
+
+def test_rolling_quantile_param():
+    ser = Series([0.0, 0.1, 0.5, 0.9, 1.0])
+    msg = "quantile value -0.1 not in \\[0, 1\\]"
+    with pytest.raises(ValueError, match=msg):
+        ser.rolling(3).quantile(-0.1)
+
+    msg = "quantile value 10.0 not in \\[0, 1\\]"
+    with pytest.raises(ValueError, match=msg):
+        ser.rolling(3).quantile(10.0)
+
+    msg = "must be real number, not str"
+    with pytest.raises(TypeError, match=msg):
+        ser.rolling(3).quantile("foo")
+
+
+def test_rolling_std_1obs():
+    vals = Series([1.0, 2.0, 3.0, 4.0, 5.0])
+
+    result = vals.rolling(1, min_periods=1).std()
+    expected = Series([np.nan] * 5)
+    tm.assert_series_equal(result, expected)
+
+    result = vals.rolling(1, min_periods=1).std(ddof=0)
+    expected = Series([0.0] * 5)
+    tm.assert_series_equal(result, expected)
+
+    result = Series([np.nan, np.nan, 3, 4, 5]).rolling(3, min_periods=2).std()
+    assert np.isnan(result[2])
+
+
+def test_rolling_std_neg_sqrt():
+    # unit test from Bottleneck
+
+    # Test move_nanstd for neg sqrt.
+
+    a = Series(
+        [
+            0.0011448196318903589,
+            0.00028718669878572767,
+            0.00028718669878572767,
+            0.00028718669878572767,
+            0.00028718669878572767,
+        ]
+    )
+    b = a.rolling(window=3).std()
+    assert np.isfinite(b[2:]).all()
+
+    b = a.ewm(span=3).std()
+    assert np.isfinite(b[2:]).all()
+
+
+def test_step_not_integer_raises():
+    with pytest.raises(ValueError, match="step must be an integer"):
+        DataFrame(range(2)).rolling(1, step="foo")
+
+
+def test_step_not_positive_raises():
+    with pytest.raises(ValueError, match="step must be >= 0"):
+        DataFrame(range(2)).rolling(1, step=-1)
+
+
+@pytest.mark.parametrize(
+    ["values", "window", "min_periods", "expected"],
+    [
+        [
+            [20, 10, 10, np.inf, 1, 1, 2, 3],
+            3,
+            1,
+            [np.nan, 50, 100 / 3, 0, 40.5, 0, 1 / 3, 1],
+        ],
+        [
+            [20, 10, 10, np.nan, 10, 1, 2, 3],
+            3,
+            1,
+            [np.nan, 50, 100 / 3, 0, 0, 40.5, 73 / 3, 1],
+        ],
+        [
+            [np.nan, 5, 6, 7, 5, 5, 5],
+            3,
+            3,
+            [np.nan] * 3 + [1, 1, 4 / 3, 0],
+        ],
+        [
+            [5, 7, 7, 7, np.nan, np.inf, 4, 3, 3, 3],
+            3,
+            3,
+            [np.nan] * 2 + [4 / 3, 0] + [np.nan] * 4 + [1 / 3, 0],
+        ],
+        [
+            [5, 7, 7, 7, np.nan, np.inf, 7, 3, 3, 3],
+            3,
+            3,
+            [np.nan] * 2 + [4 / 3, 0] + [np.nan] * 4 + [16 / 3, 0],
+        ],
+        [
+            [5, 7] * 4,
+            3,
+            3,
+            [np.nan] * 2 + [4 / 3] * 6,
+        ],
+        [
+            [5, 7, 5, np.nan, 7, 5, 7],
+            3,
+            2,
+            [np.nan, 2, 4 / 3] + [2] * 3 + [4 / 3],
+        ],
+    ],
+)
+def test_rolling_var_same_value_count_logic(values, window, min_periods, expected):
+    # GH 42064.
+
+    expected = Series(expected)
+    sr = Series(values)
+
+    # With new algo implemented, result will be set to .0 in rolling var
+    # if sufficient amount of consecutively same values are found.
+    result_var = sr.rolling(window, min_periods=min_periods).var()
+
+    # use `assert_series_equal` twice to check for equality,
+    # because `check_exact=True` will fail in 32-bit tests due to
+    # precision loss.
+
+    # 1. result should be close to correct value
+    # non-zero values can still differ slightly from "truth"
+    # as the result of online algorithm
+    tm.assert_series_equal(result_var, expected)
+    # 2. zeros should be exactly the same since the new algo takes effect here
+    tm.assert_series_equal(expected == 0, result_var == 0)
+
+    # std should also pass as it's just a sqrt of var
+    result_std = sr.rolling(window, min_periods=min_periods).std()
+    tm.assert_series_equal(result_std, np.sqrt(expected))
+    tm.assert_series_equal(expected == 0, result_std == 0)
+
+
+def test_rolling_mean_sum_floating_artifacts():
+    # GH 42064.
+
+    sr = Series([1 / 3, 4, 0, 0, 0, 0, 0])
+    r = sr.rolling(3)
+    result = r.mean()
+    assert (result[-3:] == 0).all()
+    result = r.sum()
+    assert (result[-3:] == 0).all()
+
+
+def test_rolling_skew_kurt_floating_artifacts():
+    # GH 42064 46431
+
+    sr = Series([1 / 3, 4, 0, 0, 0, 0, 0])
+    r = sr.rolling(4)
+    result = r.skew()
+    assert (result[-2:] == 0).all()
+    result = r.kurt()
+    assert (result[-2:] == -3).all()
+
+
+def test_numeric_only_frame(arithmetic_win_operators, numeric_only):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    df = DataFrame({"a": [1], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    rolling = df.rolling(2, min_periods=1)
+    op = getattr(rolling, kernel)
+    result = op(numeric_only=numeric_only)
+
+    columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+    expected = df[columns].agg([kernel]).reset_index(drop=True).astype(float)
+    assert list(expected.columns) == columns
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+def test_numeric_only_corr_cov_frame(kernel, numeric_only, use_arg):
+    # GH#46560
+    df = DataFrame({"a": [1, 2, 3], "b": 2, "c": 3})
+    df["c"] = df["c"].astype(object)
+    arg = (df,) if use_arg else ()
+    rolling = df.rolling(2, min_periods=1)
+    op = getattr(rolling, kernel)
+    result = op(*arg, numeric_only=numeric_only)
+
+    # Compare result to op using float dtypes, dropping c when numeric_only is True
+    columns = ["a", "b"] if numeric_only else ["a", "b", "c"]
+    df2 = df[columns].astype(float)
+    arg2 = (df2,) if use_arg else ()
+    rolling2 = df2.rolling(2, min_periods=1)
+    op2 = getattr(rolling2, kernel)
+    expected = op2(*arg2, numeric_only=numeric_only)
+
+    tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_series(arithmetic_win_operators, numeric_only, dtype):
+    # GH#46560
+    kernel = arithmetic_win_operators
+    ser = Series([1], dtype=dtype)
+    rolling = ser.rolling(2, min_periods=1)
+    op = getattr(rolling, kernel)
+    if numeric_only and dtype is object:
+        msg = f"Rolling.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(numeric_only=numeric_only)
+    else:
+        result = op(numeric_only=numeric_only)
+        expected = ser.agg([kernel]).reset_index(drop=True).astype(float)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("kernel", ["corr", "cov"])
+@pytest.mark.parametrize("use_arg", [True, False])
+@pytest.mark.parametrize("dtype", [int, object])
+def test_numeric_only_corr_cov_series(kernel, use_arg, numeric_only, dtype):
+    # GH#46560
+    ser = Series([1, 2, 3], dtype=dtype)
+    arg = (ser,) if use_arg else ()
+    rolling = ser.rolling(2, min_periods=1)
+    op = getattr(rolling, kernel)
+    if numeric_only and dtype is object:
+        msg = f"Rolling.{kernel} does not implement numeric_only"
+        with pytest.raises(NotImplementedError, match=msg):
+            op(*arg, numeric_only=numeric_only)
+    else:
+        result = op(*arg, numeric_only=numeric_only)
+
+        ser2 = ser.astype(float)
+        arg2 = (ser2,) if use_arg else ()
+        rolling2 = ser2.rolling(2, min_periods=1)
+        op2 = getattr(rolling2, kernel)
+        expected = op2(*arg2, numeric_only=numeric_only)
+        tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("tz", [None, "UTC", "Europe/Prague"])
+def test_rolling_timedelta_window_non_nanoseconds(unit, tz):
+    # Test Sum, GH#55106
+    df_time = DataFrame(
+        {"A": range(5)}, index=date_range("2013-01-01", freq="1s", periods=5, tz=tz)
+    )
+    sum_in_nanosecs = df_time.rolling("1s").sum()
+    # microseconds / milliseconds should not break the correct rolling
+    df_time.index = df_time.index.as_unit(unit)
+    sum_in_microsecs = df_time.rolling("1s").sum()
+    sum_in_microsecs.index = sum_in_microsecs.index.as_unit("ns")
+    tm.assert_frame_equal(sum_in_nanosecs, sum_in_microsecs)
+
+    # Test max, GH#55026
+    ref_dates = date_range("2023-01-01", "2023-01-10", unit="ns", tz=tz)
+    ref_series = Series(0, index=ref_dates)
+    ref_series.iloc[0] = 1
+    ref_max_series = ref_series.rolling(Timedelta(days=4)).max()
+
+    dates = date_range("2023-01-01", "2023-01-10", unit=unit, tz=tz)
+    series = Series(0, index=dates)
+    series.iloc[0] = 1
+    max_series = series.rolling(Timedelta(days=4)).max()
+
+    ref_df = DataFrame(ref_max_series)
+    df = DataFrame(max_series)
+    df.index = df.index.as_unit("ns")
+
+    tm.assert_frame_equal(ref_df, df)
diff --git a/pandas/pandas/tests/window/test_rolling_functions.py b/pandas/pandas/tests/window/test_rolling_functions.py
new file mode 100644
index 0000000000000..6820ab7332975
--- /dev/null
+++ b/pandas/pandas/tests/window/test_rolling_functions.py
@@ -0,0 +1,535 @@
+from datetime import datetime
+
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Series,
+    concat,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+@pytest.mark.parametrize(
+    "compare_func, roll_func, kwargs",
+    [
+        [np.mean, "mean", {}],
+        [np.nansum, "sum", {}],
+        [
+            lambda x: np.isfinite(x).astype(float).sum(),
+            "count",
+            {},
+        ],
+        [np.median, "median", {}],
+        [np.min, "min", {}],
+        [np.max, "max", {}],
+        [lambda x: np.std(x, ddof=1), "std", {}],
+        [lambda x: np.std(x, ddof=0), "std", {"ddof": 0}],
+        [lambda x: np.var(x, ddof=1), "var", {}],
+        [lambda x: np.var(x, ddof=0), "var", {"ddof": 0}],
+    ],
+)
+def test_series(series, compare_func, roll_func, kwargs, step):
+    result = getattr(series.rolling(50, step=step), roll_func)(**kwargs)
+    assert isinstance(result, Series)
+    end = range(0, len(series), step or 1)[-1] + 1
+    tm.assert_almost_equal(result.iloc[-1], compare_func(series[end - 50 : end]))
+
+
+@pytest.mark.parametrize(
+    "compare_func, roll_func, kwargs",
+    [
+        [np.mean, "mean", {}],
+        [np.nansum, "sum", {}],
+        [
+            lambda x: np.isfinite(x).astype(float).sum(),
+            "count",
+            {},
+        ],
+        [np.median, "median", {}],
+        [np.min, "min", {}],
+        [np.max, "max", {}],
+        [lambda x: np.std(x, ddof=1), "std", {}],
+        [lambda x: np.std(x, ddof=0), "std", {"ddof": 0}],
+        [lambda x: np.var(x, ddof=1), "var", {}],
+        [lambda x: np.var(x, ddof=0), "var", {"ddof": 0}],
+    ],
+)
+def test_frame(raw, frame, compare_func, roll_func, kwargs, step):
+    result = getattr(frame.rolling(50, step=step), roll_func)(**kwargs)
+    assert isinstance(result, DataFrame)
+    end = range(0, len(frame), step or 1)[-1] + 1
+    tm.assert_series_equal(
+        result.iloc[-1, :],
+        frame.iloc[end - 50 : end, :].apply(compare_func, axis=0, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize(
+    "compare_func, roll_func, kwargs, minp",
+    [
+        [np.mean, "mean", {}, 10],
+        [np.nansum, "sum", {}, 10],
+        [lambda x: np.isfinite(x).astype(float).sum(), "count", {}, 0],
+        [np.median, "median", {}, 10],
+        [np.min, "min", {}, 10],
+        [np.max, "max", {}, 10],
+        [lambda x: np.std(x, ddof=1), "std", {}, 10],
+        [lambda x: np.std(x, ddof=0), "std", {"ddof": 0}, 10],
+        [lambda x: np.var(x, ddof=1), "var", {}, 10],
+        [lambda x: np.var(x, ddof=0), "var", {"ddof": 0}, 10],
+    ],
+)
+def test_time_rule_series(series, compare_func, roll_func, kwargs, minp):
+    win = 25
+    ser = series[::2].resample("B").mean()
+    series_result = getattr(ser.rolling(window=win, min_periods=minp), roll_func)(
+        **kwargs
+    )
+    last_date = series_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_series = series[::2].truncate(prev_date, last_date)
+    tm.assert_almost_equal(series_result.iloc[-1], compare_func(trunc_series))
+
+
+@pytest.mark.parametrize(
+    "compare_func, roll_func, kwargs, minp",
+    [
+        [np.mean, "mean", {}, 10],
+        [np.nansum, "sum", {}, 10],
+        [lambda x: np.isfinite(x).astype(float).sum(), "count", {}, 0],
+        [np.median, "median", {}, 10],
+        [np.min, "min", {}, 10],
+        [np.max, "max", {}, 10],
+        [lambda x: np.std(x, ddof=1), "std", {}, 10],
+        [lambda x: np.std(x, ddof=0), "std", {"ddof": 0}, 10],
+        [lambda x: np.var(x, ddof=1), "var", {}, 10],
+        [lambda x: np.var(x, ddof=0), "var", {"ddof": 0}, 10],
+    ],
+)
+def test_time_rule_frame(raw, frame, compare_func, roll_func, kwargs, minp):
+    win = 25
+    frm = frame[::2].resample("B").mean()
+    frame_result = getattr(frm.rolling(window=win, min_periods=minp), roll_func)(
+        **kwargs
+    )
+    last_date = frame_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_frame = frame[::2].truncate(prev_date, last_date)
+    tm.assert_series_equal(
+        frame_result.xs(last_date),
+        trunc_frame.apply(compare_func, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize(
+    "compare_func, roll_func, kwargs",
+    [
+        [np.mean, "mean", {}],
+        [np.nansum, "sum", {}],
+        [np.median, "median", {}],
+        [np.min, "min", {}],
+        [np.max, "max", {}],
+        [lambda x: np.std(x, ddof=1), "std", {}],
+        [lambda x: np.std(x, ddof=0), "std", {"ddof": 0}],
+        [lambda x: np.var(x, ddof=1), "var", {}],
+        [lambda x: np.var(x, ddof=0), "var", {"ddof": 0}],
+    ],
+)
+def test_nans(compare_func, roll_func, kwargs):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = getattr(obj.rolling(50, min_periods=30), roll_func)(**kwargs)
+    tm.assert_almost_equal(result.iloc[-1], compare_func(obj[10:-10]))
+
+    # min_periods is working correctly
+    result = getattr(obj.rolling(20, min_periods=15), roll_func)(**kwargs)
+    assert isna(result.iloc[23])
+    assert not isna(result.iloc[24])
+
+    assert not isna(result.iloc[-6])
+    assert isna(result.iloc[-5])
+
+    obj2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = getattr(obj2.rolling(10, min_periods=5), roll_func)(**kwargs)
+    assert isna(result.iloc[3])
+    assert notna(result.iloc[4])
+
+    if roll_func != "sum":
+        result0 = getattr(obj.rolling(20, min_periods=0), roll_func)(**kwargs)
+        result1 = getattr(obj.rolling(20, min_periods=1), roll_func)(**kwargs)
+        tm.assert_almost_equal(result0, result1)
+
+
+def test_nans_count():
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+    result = obj.rolling(50, min_periods=30).count()
+    tm.assert_almost_equal(
+        result.iloc[-1], np.isfinite(obj[10:-10]).astype(float).sum()
+    )
+
+
+@pytest.mark.parametrize(
+    "roll_func, kwargs",
+    [
+        ["mean", {}],
+        ["sum", {}],
+        ["median", {}],
+        ["min", {}],
+        ["max", {}],
+        ["std", {}],
+        ["std", {"ddof": 0}],
+        ["var", {}],
+        ["var", {"ddof": 0}],
+    ],
+)
+@pytest.mark.parametrize("minp", [0, 99, 100])
+def test_min_periods(series, minp, roll_func, kwargs, step):
+    result = getattr(
+        series.rolling(len(series) + 1, min_periods=minp, step=step), roll_func
+    )(**kwargs)
+    expected = getattr(
+        series.rolling(len(series), min_periods=minp, step=step), roll_func
+    )(**kwargs)
+    nan_mask = isna(result)
+    tm.assert_series_equal(nan_mask, isna(expected))
+
+    nan_mask = ~nan_mask
+    tm.assert_almost_equal(result[nan_mask], expected[nan_mask])
+
+
+def test_min_periods_count(series, step):
+    result = series.rolling(len(series) + 1, min_periods=0, step=step).count()
+    expected = series.rolling(len(series), min_periods=0, step=step).count()
+    nan_mask = isna(result)
+    tm.assert_series_equal(nan_mask, isna(expected))
+
+    nan_mask = ~nan_mask
+    tm.assert_almost_equal(result[nan_mask], expected[nan_mask])
+
+
+@pytest.mark.parametrize(
+    "roll_func, kwargs, minp",
+    [
+        ["mean", {}, 15],
+        ["sum", {}, 15],
+        ["count", {}, 0],
+        ["median", {}, 15],
+        ["min", {}, 15],
+        ["max", {}, 15],
+        ["std", {}, 15],
+        ["std", {"ddof": 0}, 15],
+        ["var", {}, 15],
+        ["var", {"ddof": 0}, 15],
+    ],
+)
+def test_center(roll_func, kwargs, minp):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = getattr(obj.rolling(20, min_periods=minp, center=True), roll_func)(
+        **kwargs
+    )
+    expected = (
+        getattr(
+            concat([obj, Series([np.nan] * 9)]).rolling(20, min_periods=minp), roll_func
+        )(**kwargs)
+        .iloc[9:]
+        .reset_index(drop=True)
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize(
+    "roll_func, kwargs, minp, fill_value",
+    [
+        ["mean", {}, 10, None],
+        ["sum", {}, 10, None],
+        ["count", {}, 0, 0],
+        ["median", {}, 10, None],
+        ["min", {}, 10, None],
+        ["max", {}, 10, None],
+        ["std", {}, 10, None],
+        ["std", {"ddof": 0}, 10, None],
+        ["var", {}, 10, None],
+        ["var", {"ddof": 0}, 10, None],
+    ],
+)
+def test_center_reindex_series(series, roll_func, kwargs, minp, fill_value):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    series_xp = (
+        getattr(
+            series.reindex(list(series.index) + s).rolling(window=25, min_periods=minp),
+            roll_func,
+        )(**kwargs)
+        .shift(-12)
+        .reindex(series.index)
+    )
+    series_rs = getattr(
+        series.rolling(window=25, min_periods=minp, center=True), roll_func
+    )(**kwargs)
+    if fill_value is not None:
+        series_xp = series_xp.fillna(fill_value)
+    tm.assert_series_equal(series_xp, series_rs)
+
+
+@pytest.mark.parametrize(
+    "roll_func, kwargs, minp, fill_value",
+    [
+        ["mean", {}, 10, None],
+        ["sum", {}, 10, None],
+        ["count", {}, 0, 0],
+        ["median", {}, 10, None],
+        ["min", {}, 10, None],
+        ["max", {}, 10, None],
+        ["std", {}, 10, None],
+        ["std", {"ddof": 0}, 10, None],
+        ["var", {}, 10, None],
+        ["var", {"ddof": 0}, 10, None],
+    ],
+)
+def test_center_reindex_frame(frame, roll_func, kwargs, minp, fill_value):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    frame_xp = (
+        getattr(
+            frame.reindex(list(frame.index) + s).rolling(window=25, min_periods=minp),
+            roll_func,
+        )(**kwargs)
+        .shift(-12)
+        .reindex(frame.index)
+    )
+    frame_rs = getattr(
+        frame.rolling(window=25, min_periods=minp, center=True), roll_func
+    )(**kwargs)
+    if fill_value is not None:
+        frame_xp = frame_xp.fillna(fill_value)
+    tm.assert_frame_equal(frame_xp, frame_rs)
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: x.rolling(window=10, min_periods=5).cov(x, pairwise=False),
+        lambda x: x.rolling(window=10, min_periods=5).corr(x, pairwise=False),
+        lambda x: x.rolling(window=10, min_periods=5).max(),
+        lambda x: x.rolling(window=10, min_periods=5).min(),
+        lambda x: x.rolling(window=10, min_periods=5).sum(),
+        lambda x: x.rolling(window=10, min_periods=5).mean(),
+        lambda x: x.rolling(window=10, min_periods=5).std(),
+        lambda x: x.rolling(window=10, min_periods=5).var(),
+        lambda x: x.rolling(window=10, min_periods=5).skew(),
+        lambda x: x.rolling(window=10, min_periods=5).kurt(),
+        lambda x: x.rolling(window=10, min_periods=5).first(),
+        lambda x: x.rolling(window=10, min_periods=5).last(),
+        lambda x: x.rolling(window=10, min_periods=5).quantile(q=0.5),
+        lambda x: x.rolling(window=10, min_periods=5).median(),
+        lambda x: x.rolling(window=10, min_periods=5).apply(sum, raw=False),
+        lambda x: x.rolling(window=10, min_periods=5).apply(sum, raw=True),
+        pytest.param(
+            lambda x: x.rolling(win_type="boxcar", window=10, min_periods=5).mean(),
+            marks=td.skip_if_no("scipy"),
+        ),
+    ],
+)
+def test_rolling_functions_window_non_shrinkage(f):
+    # GH 7764
+    s = Series(range(4))
+    s_expected = Series(np.nan, index=s.index)
+    df = DataFrame([[1, 5], [3, 2], [3, 9], [-1, 0]], columns=["A", "B"])
+    df_expected = DataFrame(np.nan, index=df.index, columns=df.columns)
+
+    s_result = f(s)
+    tm.assert_series_equal(s_result, s_expected)
+
+    df_result = f(df)
+    tm.assert_frame_equal(df_result, df_expected)
+
+
+def test_rolling_max_gh6297(step):
+    """Replicate result expected in GH #6297"""
+    indices = [datetime(1975, 1, i) for i in range(1, 6)]
+    # So that we can have 2 datapoints on one of the days
+    indices.append(datetime(1975, 1, 3, 6, 0))
+    series = Series(range(1, 7), index=indices)
+    # Use floats instead of ints as values
+    series = series.map(lambda x: float(x))
+    # Sort chronologically
+    series = series.sort_index()
+
+    expected = Series(
+        [1.0, 2.0, 6.0, 4.0, 5.0],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )[::step]
+    x = series.resample("D").max().rolling(window=1, step=step).max()
+    tm.assert_series_equal(expected, x)
+
+
+def test_rolling_max_resample(step):
+    indices = [datetime(1975, 1, i) for i in range(1, 6)]
+    # So that we can have 3 datapoints on last day (4, 10, and 20)
+    indices.append(datetime(1975, 1, 5, 1))
+    indices.append(datetime(1975, 1, 5, 2))
+    series = Series(list(range(5)) + [10, 20], index=indices)
+    # Use floats instead of ints as values
+    series = series.map(lambda x: float(x))
+    # Sort chronologically
+    series = series.sort_index()
+
+    # Default how should be max
+    expected = Series(
+        [0.0, 1.0, 2.0, 3.0, 20.0],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )[::step]
+    x = series.resample("D").max().rolling(window=1, step=step).max()
+    tm.assert_series_equal(expected, x)
+
+    # Now specify median (10.0)
+    expected = Series(
+        [0.0, 1.0, 2.0, 3.0, 10.0],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )[::step]
+    x = series.resample("D").median().rolling(window=1, step=step).max()
+    tm.assert_series_equal(expected, x)
+
+    # Now specify mean (4+10+20)/3
+    v = (4.0 + 10.0 + 20.0) / 3.0
+    expected = Series(
+        [0.0, 1.0, 2.0, 3.0, v],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )[::step]
+    x = series.resample("D").mean().rolling(window=1, step=step).max()
+    tm.assert_series_equal(expected, x)
+
+
+def test_rolling_min_resample(step):
+    indices = [datetime(1975, 1, i) for i in range(1, 6)]
+    # So that we can have 3 datapoints on last day (4, 10, and 20)
+    indices.append(datetime(1975, 1, 5, 1))
+    indices.append(datetime(1975, 1, 5, 2))
+    series = Series(list(range(5)) + [10, 20], index=indices)
+    # Use floats instead of ints as values
+    series = series.map(lambda x: float(x))
+    # Sort chronologically
+    series = series.sort_index()
+
+    # Default how should be min
+    expected = Series(
+        [0.0, 1.0, 2.0, 3.0, 4.0],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )[::step]
+    r = series.resample("D").min().rolling(window=1, step=step)
+    tm.assert_series_equal(expected, r.min())
+
+
+def test_rolling_median_resample():
+    indices = [datetime(1975, 1, i) for i in range(1, 6)]
+    # So that we can have 3 datapoints on last day (4, 10, and 20)
+    indices.append(datetime(1975, 1, 5, 1))
+    indices.append(datetime(1975, 1, 5, 2))
+    series = Series(list(range(5)) + [10, 20], index=indices)
+    # Use floats instead of ints as values
+    series = series.map(lambda x: float(x))
+    # Sort chronologically
+    series = series.sort_index()
+
+    # Default how should be median
+    expected = Series(
+        [0.0, 1.0, 2.0, 3.0, 10],
+        index=DatetimeIndex([datetime(1975, 1, i, 0) for i in range(1, 6)], freq="D"),
+    )
+    x = series.resample("D").median().rolling(window=1).median()
+    tm.assert_series_equal(expected, x)
+
+
+def test_rolling_median_memory_error():
+    # GH11722
+    n = 20000
+    Series(np.random.default_rng(2).standard_normal(n)).rolling(
+        window=2, center=False
+    ).median()
+    Series(np.random.default_rng(2).standard_normal(n)).rolling(
+        window=2, center=False
+    ).median()
+
+
+def test_rolling_min_max_numeric_types(any_real_numpy_dtype):
+    # GH12373
+
+    # Just testing that these don't throw exceptions and that
+    # the return type is float64. Other tests will cover quantitative
+    # correctness
+    result = (
+        DataFrame(np.arange(20, dtype=any_real_numpy_dtype)).rolling(window=5).max()
+    )
+    assert result.dtypes[0] == np.dtype("f8")
+    result = (
+        DataFrame(np.arange(20, dtype=any_real_numpy_dtype)).rolling(window=5).min()
+    )
+    assert result.dtypes[0] == np.dtype("f8")
+
+
+@pytest.mark.parametrize(
+    "f",
+    [
+        lambda x: x.rolling(window=10, min_periods=0).count(),
+        lambda x: x.rolling(window=10, min_periods=5).cov(x, pairwise=False),
+        lambda x: x.rolling(window=10, min_periods=5).corr(x, pairwise=False),
+        lambda x: x.rolling(window=10, min_periods=5).max(),
+        lambda x: x.rolling(window=10, min_periods=5).min(),
+        lambda x: x.rolling(window=10, min_periods=5).sum(),
+        lambda x: x.rolling(window=10, min_periods=5).mean(),
+        lambda x: x.rolling(window=10, min_periods=5).std(),
+        lambda x: x.rolling(window=10, min_periods=5).var(),
+        lambda x: x.rolling(window=10, min_periods=5).skew(),
+        lambda x: x.rolling(window=10, min_periods=5).kurt(),
+        lambda x: x.rolling(window=10, min_periods=5).first(),
+        lambda x: x.rolling(window=10, min_periods=5).last(),
+        lambda x: x.rolling(window=10, min_periods=5).quantile(0.5),
+        lambda x: x.rolling(window=10, min_periods=5).median(),
+        lambda x: x.rolling(window=10, min_periods=5).apply(sum, raw=False),
+        lambda x: x.rolling(window=10, min_periods=5).apply(sum, raw=True),
+        pytest.param(
+            lambda x: x.rolling(win_type="boxcar", window=10, min_periods=5).mean(),
+            marks=td.skip_if_no("scipy"),
+        ),
+    ],
+)
+def test_moment_functions_zero_length(f):
+    # GH 8056
+    s = Series(dtype=np.float64)
+    s_expected = s
+    df1 = DataFrame()
+    df1_expected = df1
+    df2 = DataFrame(columns=["a"])
+    df2["a"] = df2["a"].astype("float64")
+    df2_expected = df2
+
+    s_result = f(s)
+    tm.assert_series_equal(s_result, s_expected)
+
+    df1_result = f(df1)
+    tm.assert_frame_equal(df1_result, df1_expected)
+
+    df2_result = f(df2)
+    tm.assert_frame_equal(df2_result, df2_expected)
diff --git a/pandas/pandas/tests/window/test_rolling_quantile.py b/pandas/pandas/tests/window/test_rolling_quantile.py
new file mode 100644
index 0000000000000..66713f1cfaa8d
--- /dev/null
+++ b/pandas/pandas/tests/window/test_rolling_quantile.py
@@ -0,0 +1,175 @@
+from functools import partial
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    concat,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+def scoreatpercentile(a, per):
+    values = np.sort(a, axis=0)
+
+    idx = int(per / 1.0 * (values.shape[0] - 1))
+
+    if idx == values.shape[0] - 1:
+        retval = values[-1]
+
+    else:
+        qlow = idx / (values.shape[0] - 1)
+        qhig = (idx + 1) / (values.shape[0] - 1)
+        vlow = values[idx]
+        vhig = values[idx + 1]
+        retval = vlow + (vhig - vlow) * (per - qlow) / (qhig - qlow)
+
+    return retval
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_series(series, q, step):
+    compare_func = partial(scoreatpercentile, per=q)
+    result = series.rolling(50, step=step).quantile(q)
+    assert isinstance(result, Series)
+    end = range(0, len(series), step or 1)[-1] + 1
+    tm.assert_almost_equal(result.iloc[-1], compare_func(series[end - 50 : end]))
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_frame(raw, frame, q, step):
+    compare_func = partial(scoreatpercentile, per=q)
+    result = frame.rolling(50, step=step).quantile(q)
+    assert isinstance(result, DataFrame)
+    end = range(0, len(frame), step or 1)[-1] + 1
+    tm.assert_series_equal(
+        result.iloc[-1, :],
+        frame.iloc[end - 50 : end, :].apply(compare_func, axis=0, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_time_rule_series(series, q):
+    compare_func = partial(scoreatpercentile, per=q)
+    win = 25
+    ser = series[::2].resample("B").mean()
+    series_result = ser.rolling(window=win, min_periods=10).quantile(q)
+    last_date = series_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_series = series[::2].truncate(prev_date, last_date)
+    tm.assert_almost_equal(series_result.iloc[-1], compare_func(trunc_series))
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_time_rule_frame(raw, frame, q):
+    compare_func = partial(scoreatpercentile, per=q)
+    win = 25
+    frm = frame[::2].resample("B").mean()
+    frame_result = frm.rolling(window=win, min_periods=10).quantile(q)
+    last_date = frame_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_frame = frame[::2].truncate(prev_date, last_date)
+    tm.assert_series_equal(
+        frame_result.xs(last_date),
+        trunc_frame.apply(compare_func, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_nans(q):
+    compare_func = partial(scoreatpercentile, per=q)
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = obj.rolling(50, min_periods=30).quantile(q)
+    tm.assert_almost_equal(result.iloc[-1], compare_func(obj[10:-10]))
+
+    # min_periods is working correctly
+    result = obj.rolling(20, min_periods=15).quantile(q)
+    assert isna(result.iloc[23])
+    assert not isna(result.iloc[24])
+
+    assert not isna(result.iloc[-6])
+    assert isna(result.iloc[-5])
+
+    obj2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = obj2.rolling(10, min_periods=5).quantile(q)
+    assert isna(result.iloc[3])
+    assert notna(result.iloc[4])
+
+    result0 = obj.rolling(20, min_periods=0).quantile(q)
+    result1 = obj.rolling(20, min_periods=1).quantile(q)
+    tm.assert_almost_equal(result0, result1)
+
+
+@pytest.mark.parametrize("minp", [0, 99, 100])
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_min_periods(series, minp, q, step):
+    result = series.rolling(len(series) + 1, min_periods=minp, step=step).quantile(q)
+    expected = series.rolling(len(series), min_periods=minp, step=step).quantile(q)
+    nan_mask = isna(result)
+    tm.assert_series_equal(nan_mask, isna(expected))
+
+    nan_mask = ~nan_mask
+    tm.assert_almost_equal(result[nan_mask], expected[nan_mask])
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_center(q):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = obj.rolling(20, center=True).quantile(q)
+    expected = (
+        concat([obj, Series([np.nan] * 9)])
+        .rolling(20)
+        .quantile(q)
+        .iloc[9:]
+        .reset_index(drop=True)
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_center_reindex_series(series, q):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    series_xp = (
+        series.reindex(list(series.index) + s)
+        .rolling(window=25)
+        .quantile(q)
+        .shift(-12)
+        .reindex(series.index)
+    )
+
+    series_rs = series.rolling(window=25, center=True).quantile(q)
+    tm.assert_series_equal(series_xp, series_rs)
+
+
+@pytest.mark.parametrize("q", [0.0, 0.1, 0.5, 0.9, 1.0])
+def test_center_reindex_frame(frame, q):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    frame_xp = (
+        frame.reindex(list(frame.index) + s)
+        .rolling(window=25)
+        .quantile(q)
+        .shift(-12)
+        .reindex(frame.index)
+    )
+    frame_rs = frame.rolling(window=25, center=True).quantile(q)
+    tm.assert_frame_equal(frame_xp, frame_rs)
diff --git a/pandas/pandas/tests/window/test_rolling_skew_kurt.py b/pandas/pandas/tests/window/test_rolling_skew_kurt.py
new file mode 100644
index 0000000000000..79c14f243e7cc
--- /dev/null
+++ b/pandas/pandas/tests/window/test_rolling_skew_kurt.py
@@ -0,0 +1,227 @@
+from functools import partial
+
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    concat,
+    isna,
+    notna,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+@pytest.mark.parametrize("sp_func, roll_func", [["kurtosis", "kurt"], ["skew", "skew"]])
+def test_series(series, sp_func, roll_func):
+    sp_stats = pytest.importorskip("scipy.stats")
+
+    compare_func = partial(getattr(sp_stats, sp_func), bias=False)
+    result = getattr(series.rolling(50), roll_func)()
+    assert isinstance(result, Series)
+    tm.assert_almost_equal(result.iloc[-1], compare_func(series[-50:]))
+
+
+@pytest.mark.parametrize("sp_func, roll_func", [["kurtosis", "kurt"], ["skew", "skew"]])
+def test_frame(raw, frame, sp_func, roll_func):
+    sp_stats = pytest.importorskip("scipy.stats")
+
+    compare_func = partial(getattr(sp_stats, sp_func), bias=False)
+    result = getattr(frame.rolling(50), roll_func)()
+    assert isinstance(result, DataFrame)
+    tm.assert_series_equal(
+        result.iloc[-1, :],
+        frame.iloc[-50:, :].apply(compare_func, axis=0, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize("sp_func, roll_func", [["kurtosis", "kurt"], ["skew", "skew"]])
+def test_time_rule_series(series, sp_func, roll_func):
+    sp_stats = pytest.importorskip("scipy.stats")
+
+    compare_func = partial(getattr(sp_stats, sp_func), bias=False)
+    win = 25
+    ser = series[::2].resample("B").mean()
+    series_result = getattr(ser.rolling(window=win, min_periods=10), roll_func)()
+    last_date = series_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_series = series[::2].truncate(prev_date, last_date)
+    tm.assert_almost_equal(series_result.iloc[-1], compare_func(trunc_series))
+
+
+@pytest.mark.parametrize("sp_func, roll_func", [["kurtosis", "kurt"], ["skew", "skew"]])
+def test_time_rule_frame(raw, frame, sp_func, roll_func):
+    sp_stats = pytest.importorskip("scipy.stats")
+
+    compare_func = partial(getattr(sp_stats, sp_func), bias=False)
+    win = 25
+    frm = frame[::2].resample("B").mean()
+    frame_result = getattr(frm.rolling(window=win, min_periods=10), roll_func)()
+    last_date = frame_result.index[-1]
+    prev_date = last_date - 24 * offsets.BDay()
+
+    trunc_frame = frame[::2].truncate(prev_date, last_date)
+    tm.assert_series_equal(
+        frame_result.xs(last_date),
+        trunc_frame.apply(compare_func, raw=raw),
+        check_names=False,
+    )
+
+
+@pytest.mark.parametrize("sp_func, roll_func", [["kurtosis", "kurt"], ["skew", "skew"]])
+def test_nans(sp_func, roll_func):
+    sp_stats = pytest.importorskip("scipy.stats")
+
+    compare_func = partial(getattr(sp_stats, sp_func), bias=False)
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = getattr(obj.rolling(50, min_periods=30), roll_func)()
+    tm.assert_almost_equal(result.iloc[-1], compare_func(obj[10:-10]))
+
+    # min_periods is working correctly
+    result = getattr(obj.rolling(20, min_periods=15), roll_func)()
+    assert isna(result.iloc[23])
+    assert not isna(result.iloc[24])
+
+    assert not isna(result.iloc[-6])
+    assert isna(result.iloc[-5])
+
+    obj2 = Series(np.random.default_rng(2).standard_normal(20))
+    result = getattr(obj2.rolling(10, min_periods=5), roll_func)()
+    assert isna(result.iloc[3])
+    assert notna(result.iloc[4])
+
+    result0 = getattr(obj.rolling(20, min_periods=0), roll_func)()
+    result1 = getattr(obj.rolling(20, min_periods=1), roll_func)()
+    tm.assert_almost_equal(result0, result1)
+
+
+@pytest.mark.parametrize("minp", [0, 99, 100])
+@pytest.mark.parametrize("roll_func", ["kurt", "skew"])
+def test_min_periods(series, minp, roll_func, step):
+    result = getattr(
+        series.rolling(len(series) + 1, min_periods=minp, step=step), roll_func
+    )()
+    expected = getattr(
+        series.rolling(len(series), min_periods=minp, step=step), roll_func
+    )()
+    nan_mask = isna(result)
+    tm.assert_series_equal(nan_mask, isna(expected))
+
+    nan_mask = ~nan_mask
+    tm.assert_almost_equal(result[nan_mask], expected[nan_mask])
+
+
+@pytest.mark.parametrize("roll_func", ["kurt", "skew"])
+def test_center(roll_func):
+    obj = Series(np.random.default_rng(2).standard_normal(50))
+    obj[:10] = np.nan
+    obj[-10:] = np.nan
+
+    result = getattr(obj.rolling(20, center=True), roll_func)()
+    expected = (
+        getattr(concat([obj, Series([np.nan] * 9)]).rolling(20), roll_func)()
+        .iloc[9:]
+        .reset_index(drop=True)
+    )
+    tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("roll_func", ["kurt", "skew"])
+def test_center_reindex_series(series, roll_func):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    series_xp = (
+        getattr(
+            series.reindex(list(series.index) + s).rolling(window=25),
+            roll_func,
+        )()
+        .shift(-12)
+        .reindex(series.index)
+    )
+    series_rs = getattr(series.rolling(window=25, center=True), roll_func)()
+    tm.assert_series_equal(series_xp, series_rs)
+
+
+@pytest.mark.slow
+@pytest.mark.parametrize("roll_func", ["kurt", "skew"])
+def test_center_reindex_frame(frame, roll_func):
+    # shifter index
+    s = [f"x{x:d}" for x in range(12)]
+
+    frame_xp = (
+        getattr(
+            frame.reindex(list(frame.index) + s).rolling(window=25),
+            roll_func,
+        )()
+        .shift(-12)
+        .reindex(frame.index)
+    )
+    frame_rs = getattr(frame.rolling(window=25, center=True), roll_func)()
+    tm.assert_frame_equal(frame_xp, frame_rs)
+
+
+def test_rolling_skew_edge_cases(step):
+    expected = Series([np.nan] * 4 + [0.0])[::step]
+    # yields all NaN (0 variance)
+    d = Series([1] * 5)
+    x = d.rolling(window=5, step=step).skew()
+    # index 4 should be 0 as it contains 5 same obs
+    tm.assert_series_equal(expected, x)
+
+    expected = Series([np.nan] * 5)[::step]
+    # yields all NaN (window too small)
+    d = Series(np.random.default_rng(2).standard_normal(5))
+    x = d.rolling(window=2, step=step).skew()
+    tm.assert_series_equal(expected, x)
+
+    # yields [NaN, NaN, NaN, 0.177994, 1.548824]
+    d = Series([-1.50837035, -0.1297039, 0.19501095, 1.73508164, 0.41941401])
+    expected = Series([np.nan, np.nan, np.nan, 0.177994, 1.548824])[::step]
+    x = d.rolling(window=4, step=step).skew()
+    tm.assert_series_equal(expected, x)
+
+
+def test_rolling_kurt_edge_cases(step):
+    expected = Series([np.nan] * 4 + [-3.0])[::step]
+
+    # yields all NaN (0 variance)
+    d = Series([1] * 5)
+    x = d.rolling(window=5, step=step).kurt()
+    tm.assert_series_equal(expected, x)
+
+    # yields all NaN (window too small)
+    expected = Series([np.nan] * 5)[::step]
+    d = Series(np.random.default_rng(2).standard_normal(5))
+    x = d.rolling(window=3, step=step).kurt()
+    tm.assert_series_equal(expected, x)
+
+    # yields [NaN, NaN, NaN, 1.224307, 2.671499]
+    d = Series([-1.50837035, -0.1297039, 0.19501095, 1.73508164, 0.41941401])
+    expected = Series([np.nan, np.nan, np.nan, 1.224307, 2.671499])[::step]
+    x = d.rolling(window=4, step=step).kurt()
+    tm.assert_series_equal(expected, x)
+
+
+def test_rolling_skew_eq_value_fperr(step):
+    # #18804 all rolling skew for all equal values should return Nan
+    # #46717 update: all equal values should return 0 instead of NaN
+    a = Series([1.1] * 15).rolling(window=10, step=step).skew()
+    assert (a[a.index >= 9] == 0).all()
+    assert a[a.index < 9].isna().all()
+
+
+def test_rolling_kurt_eq_value_fperr(step):
+    # #18804 all rolling kurt for all equal values should return Nan
+    # #46717 update: all equal values should return -3 instead of NaN
+    a = Series([1.1] * 15).rolling(window=10, step=step).kurt()
+    assert (a[a.index >= 9] == -3).all()
+    assert a[a.index < 9].isna().all()
diff --git a/pandas/pandas/tests/window/test_timeseries_window.py b/pandas/pandas/tests/window/test_timeseries_window.py
new file mode 100644
index 0000000000000..043f369566a5d
--- /dev/null
+++ b/pandas/pandas/tests/window/test_timeseries_window.py
@@ -0,0 +1,747 @@
+import numpy as np
+import pytest
+
+import pandas.util._test_decorators as td
+
+from pandas import (
+    DataFrame,
+    DatetimeIndex,
+    Index,
+    MultiIndex,
+    NaT,
+    Series,
+    Timestamp,
+    date_range,
+)
+import pandas._testing as tm
+
+from pandas.tseries import offsets
+
+
+@pytest.fixture
+def regular():
+    return DataFrame(
+        {"A": date_range("20130101", periods=5, freq="s"), "B": range(5)}
+    ).set_index("A")
+
+
+@pytest.fixture
+def ragged():
+    df = DataFrame({"B": range(5)})
+    df.index = [
+        Timestamp("20130101 09:00:00"),
+        Timestamp("20130101 09:00:02"),
+        Timestamp("20130101 09:00:03"),
+        Timestamp("20130101 09:00:05"),
+        Timestamp("20130101 09:00:06"),
+    ]
+    return df
+
+
+class TestRollingTS:
+    # rolling time-series friendly
+    # xref GH13327
+
+    def test_doc_string(self):
+        df = DataFrame(
+            {"B": [0, 1, 2, np.nan, 4]},
+            index=[
+                Timestamp("20130101 09:00:00"),
+                Timestamp("20130101 09:00:02"),
+                Timestamp("20130101 09:00:03"),
+                Timestamp("20130101 09:00:05"),
+                Timestamp("20130101 09:00:06"),
+            ],
+        )
+        df
+        df.rolling("2s").sum()
+
+    def test_invalid_window_non_int(self, regular):
+        # not a valid freq
+        msg = "passed window foobar is not compatible with a datetimelike index"
+        with pytest.raises(ValueError, match=msg):
+            regular.rolling(window="foobar")
+        # not a datetimelike index
+        msg = "window must be an integer"
+        with pytest.raises(ValueError, match=msg):
+            regular.reset_index().rolling(window="foobar")
+
+    @pytest.mark.parametrize("freq", ["2MS", offsets.MonthBegin(2)])
+    def test_invalid_window_nonfixed(self, freq, regular):
+        # non-fixed freqs
+        msg = "\\<2 \\* MonthBegins\\> is a non-fixed frequency"
+        with pytest.raises(ValueError, match=msg):
+            regular.rolling(window=freq)
+
+    @pytest.mark.parametrize("freq", ["1D", offsets.Day(2), "2ms"])
+    def test_valid_window(self, freq, regular):
+        regular.rolling(window=freq)
+
+    @pytest.mark.parametrize("minp", [1.0, "foo", np.array([1, 2, 3])])
+    def test_invalid_minp(self, minp, regular):
+        # non-integer min_periods
+        msg = (
+            r"local variable 'minp' referenced before assignment|"
+            "min_periods must be an integer"
+        )
+        with pytest.raises(ValueError, match=msg):
+            regular.rolling(window="1D", min_periods=minp)
+
+    def test_on(self, regular):
+        df = regular
+
+        # not a valid column
+        msg = (
+            r"invalid on specified as foobar, must be a column "
+            "\\(of DataFrame\\), an Index or None"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.rolling(window="2s", on="foobar")
+
+        # column is valid
+        df = df.copy()
+        df["C"] = date_range("20130101", periods=len(df))
+        df.rolling(window="2D", on="C").sum()
+
+        # invalid columns
+        msg = "window must be an integer"
+        with pytest.raises(ValueError, match=msg):
+            df.rolling(window="2d", on="B")
+
+        # ok even though on non-selected
+        df.rolling(window="2D", on="C").B.sum()
+
+    def test_monotonic_on(self):
+        # on/index must be monotonic
+        df = DataFrame(
+            {"A": date_range("20130101", periods=5, freq="s"), "B": range(5)}
+        )
+
+        assert df.A.is_monotonic_increasing
+        df.rolling("2s", on="A").sum()
+
+        df = df.set_index("A")
+        assert df.index.is_monotonic_increasing
+        df.rolling("2s").sum()
+
+    def test_non_monotonic_on(self):
+        # GH 19248
+        df = DataFrame(
+            {"A": date_range("20130101", periods=5, freq="s"), "B": range(5)}
+        )
+        df = df.set_index("A")
+        non_monotonic_index = df.index.to_list()
+        non_monotonic_index[0] = non_monotonic_index[3]
+        df.index = non_monotonic_index
+
+        assert not df.index.is_monotonic_increasing
+
+        msg = "index values must be monotonic"
+        with pytest.raises(ValueError, match=msg):
+            df.rolling("2s").sum()
+
+        df = df.reset_index()
+
+        msg = (
+            r"invalid on specified as A, must be a column "
+            "\\(of DataFrame\\), an Index or None"
+        )
+        with pytest.raises(ValueError, match=msg):
+            df.rolling("2s", on="A").sum()
+
+    def test_frame_on(self):
+        df = DataFrame(
+            {"B": range(5), "C": date_range("20130101 09:00:00", periods=5, freq="3s")}
+        )
+
+        df["A"] = [
+            Timestamp("20130101 09:00:00"),
+            Timestamp("20130101 09:00:02"),
+            Timestamp("20130101 09:00:03"),
+            Timestamp("20130101 09:00:05"),
+            Timestamp("20130101 09:00:06"),
+        ]
+
+        # we are doing simulating using 'on'
+        expected = df.set_index("A").rolling("2s").B.sum().reset_index(drop=True)
+
+        result = df.rolling("2s", on="A").B.sum()
+        tm.assert_series_equal(result, expected)
+
+        # test as a frame
+        # we should be ignoring the 'on' as an aggregation column
+        # note that the expected is setting, computing, and resetting
+        # so the columns need to be switched compared
+        # to the actual result where they are ordered as in the
+        # original
+        expected = (
+            df.set_index("A").rolling("2s")[["B"]].sum().reset_index()[["B", "A"]]
+        )
+
+        result = df.rolling("2s", on="A")[["B"]].sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_frame_on2(self, unit):
+        # using multiple aggregation columns
+        dti = DatetimeIndex(
+            [
+                Timestamp("20130101 09:00:00"),
+                Timestamp("20130101 09:00:02"),
+                Timestamp("20130101 09:00:03"),
+                Timestamp("20130101 09:00:05"),
+                Timestamp("20130101 09:00:06"),
+            ]
+        ).as_unit(unit)
+        df = DataFrame(
+            {
+                "A": [0, 1, 2, 3, 4],
+                "B": [0, 1, 2, np.nan, 4],
+                "C": dti,
+            },
+            columns=["A", "C", "B"],
+        )
+
+        expected1 = DataFrame(
+            {"A": [0.0, 1, 3, 3, 7], "B": [0, 1, 3, np.nan, 4], "C": df["C"]},
+            columns=["A", "C", "B"],
+        )
+
+        result = df.rolling("2s", on="C").sum()
+        expected = expected1
+        tm.assert_frame_equal(result, expected)
+
+        expected = Series([0, 1, 3, np.nan, 4], name="B")
+        result = df.rolling("2s", on="C").B.sum()
+        tm.assert_series_equal(result, expected)
+
+        expected = expected1[["A", "B", "C"]]
+        result = df.rolling("2s", on="C")[["A", "B", "C"]].sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_basic_regular(self, regular):
+        df = regular.copy()
+
+        df.index = date_range("20130101", periods=5, freq="D")
+        expected = df.rolling(window=1, min_periods=1).sum()
+        result = df.rolling(window="1D").sum()
+        tm.assert_frame_equal(result, expected)
+
+        df.index = date_range("20130101", periods=5, freq="2D")
+        expected = df.rolling(window=1, min_periods=1).sum()
+        result = df.rolling(window="2D", min_periods=1).sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.rolling(window=1, min_periods=1).sum()
+        result = df.rolling(window="2D", min_periods=1).sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.rolling(window=1).sum()
+        result = df.rolling(window="2D").sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_min_periods(self, regular):
+        # compare for min_periods
+        df = regular
+
+        # these slightly different
+        expected = df.rolling(2, min_periods=1).sum()
+        result = df.rolling("2s").sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.rolling(2, min_periods=1).sum()
+        result = df.rolling("2s", min_periods=1).sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_closed(self, regular, unit):
+        # xref GH13965
+
+        dti = DatetimeIndex(
+            [
+                Timestamp("20130101 09:00:01"),
+                Timestamp("20130101 09:00:02"),
+                Timestamp("20130101 09:00:03"),
+                Timestamp("20130101 09:00:04"),
+                Timestamp("20130101 09:00:06"),
+            ]
+        ).as_unit(unit)
+
+        df = DataFrame(
+            {"A": [1] * 5},
+            index=dti,
+        )
+
+        # closed must be 'right', 'left', 'both', 'neither'
+        msg = "closed must be 'right', 'left', 'both' or 'neither'"
+        with pytest.raises(ValueError, match=msg):
+            regular.rolling(window="2s", closed="blabla")
+
+        expected = df.copy()
+        expected["A"] = [1.0, 2, 2, 2, 1]
+        result = df.rolling("2s", closed="right").sum()
+        tm.assert_frame_equal(result, expected)
+
+        # default should be 'right'
+        result = df.rolling("2s").sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.copy()
+        expected["A"] = [1.0, 2, 3, 3, 2]
+        result = df.rolling("2s", closed="both").sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.copy()
+        expected["A"] = [np.nan, 1.0, 2, 2, 1]
+        result = df.rolling("2s", closed="left").sum()
+        tm.assert_frame_equal(result, expected)
+
+        expected = df.copy()
+        expected["A"] = [np.nan, 1.0, 1, 1, np.nan]
+        result = df.rolling("2s", closed="neither").sum()
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_sum(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 3, 3, 7]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=2).sum()
+        expected = df.copy()
+        expected["B"] = [np.nan, np.nan, 3, np.nan, 7]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="3s", min_periods=1).sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 3, 5, 7]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="3s").sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 3, 5, 7]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="4s", min_periods=1).sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 3, 6, 9]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="4s", min_periods=3).sum()
+        expected = df.copy()
+        expected["B"] = [np.nan, np.nan, 3, 6, 9]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).sum()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 3, 6, 10]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_mean(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).mean()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).mean()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 1.5, 3.0, 3.5]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_median(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).median()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).median()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 1.5, 3.0, 3.5]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_quantile(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).quantile(0.5)
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).quantile(0.5)
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 1.5, 3.0, 3.5]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_std(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).std(ddof=0)
+        expected = df.copy()
+        expected["B"] = [0.0] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="1s", min_periods=1).std(ddof=1)
+        expected = df.copy()
+        expected["B"] = [np.nan] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="3s", min_periods=1).std(ddof=0)
+        expected = df.copy()
+        expected["B"] = [0.0] + [0.5] * 4
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).std(ddof=1)
+        expected = df.copy()
+        expected["B"] = [np.nan, 0.707107, 1.0, 1.0, 1.290994]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_var(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).var(ddof=0)
+        expected = df.copy()
+        expected["B"] = [0.0] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="1s", min_periods=1).var(ddof=1)
+        expected = df.copy()
+        expected["B"] = [np.nan] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="3s", min_periods=1).var(ddof=0)
+        expected = df.copy()
+        expected["B"] = [0.0] + [0.25] * 4
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).var(ddof=1)
+        expected = df.copy()
+        expected["B"] = [np.nan, 0.5, 1.0, 1.0, 1 + 2 / 3.0]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_skew(self, ragged):
+        df = ragged
+        result = df.rolling(window="3s", min_periods=1).skew()
+        expected = df.copy()
+        expected["B"] = [np.nan] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).skew()
+        expected = df.copy()
+        expected["B"] = [np.nan] * 2 + [0.0, 0.0, 0.0]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_kurt(self, ragged):
+        df = ragged
+        result = df.rolling(window="3s", min_periods=1).kurt()
+        expected = df.copy()
+        expected["B"] = [np.nan] * 5
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).kurt()
+        expected = df.copy()
+        expected["B"] = [np.nan] * 4 + [-1.2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_count(self, ragged):
+        df = ragged
+        result = df.rolling(window="1s", min_periods=1).count()
+        expected = df.copy()
+        expected["B"] = [1.0, 1, 1, 1, 1]
+        tm.assert_frame_equal(result, expected)
+
+        df = ragged
+        result = df.rolling(window="1s").count()
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).count()
+        expected = df.copy()
+        expected["B"] = [1.0, 1, 2, 1, 2]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=2).count()
+        expected = df.copy()
+        expected["B"] = [np.nan, np.nan, 2, np.nan, 2]
+        tm.assert_frame_equal(result, expected)
+
+    def test_regular_min(self):
+        df = DataFrame(
+            {"A": date_range("20130101", periods=5, freq="s"), "B": [0.0, 1, 2, 3, 4]}
+        ).set_index("A")
+        result = df.rolling("1s").min()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame(
+            {"A": date_range("20130101", periods=5, freq="s"), "B": [5, 4, 3, 4, 5]}
+        ).set_index("A")
+
+        tm.assert_frame_equal(result, expected)
+        result = df.rolling("2s").min()
+        expected = df.copy()
+        expected["B"] = [5.0, 4, 3, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling("5s").min()
+        expected = df.copy()
+        expected["B"] = [5.0, 4, 3, 3, 3]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_min(self, ragged):
+        df = ragged
+
+        result = df.rolling(window="1s", min_periods=1).min()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).min()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 1, 3, 3]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).min()
+        expected = df.copy()
+        expected["B"] = [0.0, 0, 0, 1, 1]
+        tm.assert_frame_equal(result, expected)
+
+    def test_perf_min(self):
+        N = 10000
+
+        dfp = DataFrame(
+            {"B": np.random.default_rng(2).standard_normal(N)},
+            index=date_range("20130101", periods=N, freq="s"),
+        )
+        expected = dfp.rolling(2, min_periods=1).min()
+        result = dfp.rolling("2s").min()
+        assert ((result - expected) < 0.01).all().all()
+
+        expected = dfp.rolling(200, min_periods=1).min()
+        result = dfp.rolling("200s").min()
+        assert ((result - expected) < 0.01).all().all()
+
+    def test_ragged_max(self, ragged):
+        df = ragged
+
+        result = df.rolling(window="1s", min_periods=1).max()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).max()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).max()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_first(self, ragged):
+        df = ragged
+
+        result = df.rolling(window="1s", min_periods=1).first()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).first()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 1, 3, 3]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).first()
+        expected = df.copy()
+        expected["B"] = [0.0, 0, 0, 1, 1]
+        tm.assert_frame_equal(result, expected)
+
+    def test_ragged_last(self, ragged):
+        df = ragged
+
+        result = df.rolling(window="1s", min_periods=1).last()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="2s", min_periods=1).last()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.rolling(window="5s", min_periods=1).last()
+        expected = df.copy()
+        expected["B"] = [0.0, 1, 2, 3, 4]
+        tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "freq, op, result_data",
+        [
+            ("ms", "min", [0.0] * 10),
+            ("ms", "mean", [0.0] * 9 + [2.0 / 9]),
+            ("ms", "max", [0.0] * 9 + [2.0]),
+            ("s", "min", [0.0] * 10),
+            ("s", "mean", [0.0] * 9 + [2.0 / 9]),
+            ("s", "max", [0.0] * 9 + [2.0]),
+            ("min", "min", [0.0] * 10),
+            ("min", "mean", [0.0] * 9 + [2.0 / 9]),
+            ("min", "max", [0.0] * 9 + [2.0]),
+            ("h", "min", [0.0] * 10),
+            ("h", "mean", [0.0] * 9 + [2.0 / 9]),
+            ("h", "max", [0.0] * 9 + [2.0]),
+            ("D", "min", [0.0] * 10),
+            ("D", "mean", [0.0] * 9 + [2.0 / 9]),
+            ("D", "max", [0.0] * 9 + [2.0]),
+        ],
+    )
+    def test_freqs_ops(self, freq, op, result_data):
+        # GH 21096
+        index = date_range(start="2018-1-1 01:00:00", freq=f"1{freq}", periods=10)
+        # Explicit cast to float to avoid implicit cast when setting nan
+        s = Series(data=0, index=index, dtype="float")
+        s.iloc[1] = np.nan
+        s.iloc[-1] = 2
+        result = getattr(s.rolling(window=f"10{freq}"), op)()
+        expected = Series(data=result_data, index=index)
+
+        tm.assert_series_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "f",
+        [
+            "sum",
+            "mean",
+            "count",
+            "median",
+            "std",
+            "var",
+            "kurt",
+            "skew",
+            "min",
+            "max",
+            "first",
+            "last",
+        ],
+    )
+    def test_all(self, f, regular):
+        # simple comparison of integer vs time-based windowing
+        df = regular * 2
+        er = df.rolling(window=1)
+        r = df.rolling(window="1s")
+
+        result = getattr(r, f)()
+        expected = getattr(er, f)()
+        tm.assert_frame_equal(result, expected)
+
+        result = r.quantile(0.5)
+        expected = er.quantile(0.5)
+        tm.assert_frame_equal(result, expected)
+
+    def test_all2(self, arithmetic_win_operators):
+        f = arithmetic_win_operators
+        # more sophisticated comparison of integer vs.
+        # time-based windowing
+        df = DataFrame(
+            {"B": np.arange(50)}, index=date_range("20130101", periods=50, freq="h")
+        )
+        # in-range data
+        dft = df.between_time("09:00", "16:00")
+
+        r = dft.rolling(window="5h")
+
+        result = getattr(r, f)()
+
+        # we need to roll the days separately
+        # to compare with a time-based roll
+        # finally groupby-apply will return a multi-index
+        # so we need to drop the day
+        def agg_by_day(x):
+            x = x.between_time("09:00", "16:00")
+            return getattr(x.rolling(5, min_periods=1), f)()
+
+        expected = (
+            df.groupby(df.index.day).apply(agg_by_day).reset_index(level=0, drop=True)
+        )
+
+        tm.assert_frame_equal(result, expected)
+
+    def test_rolling_cov_offset(self):
+        # GH16058
+
+        idx = date_range("2017-01-01", periods=24, freq="1h")
+        ss = Series(np.arange(len(idx)), index=idx)
+
+        result = ss.rolling("2h").cov()
+        expected = Series([np.nan] + [0.5] * (len(idx) - 1), index=idx)
+        tm.assert_series_equal(result, expected)
+
+        expected2 = ss.rolling(2, min_periods=1).cov()
+        tm.assert_series_equal(result, expected2)
+
+        result = ss.rolling("3h").cov()
+        expected = Series([np.nan, 0.5] + [1.0] * (len(idx) - 2), index=idx)
+        tm.assert_series_equal(result, expected)
+
+        expected2 = ss.rolling(3, min_periods=1).cov()
+        tm.assert_series_equal(result, expected2)
+
+    def test_rolling_on_decreasing_index(self, unit):
+        # GH-19248, GH-32385
+        index = DatetimeIndex(
+            [
+                Timestamp("20190101 09:00:30"),
+                Timestamp("20190101 09:00:27"),
+                Timestamp("20190101 09:00:20"),
+                Timestamp("20190101 09:00:18"),
+                Timestamp("20190101 09:00:10"),
+            ]
+        ).as_unit(unit)
+
+        df = DataFrame({"column": [3, 4, 4, 5, 6]}, index=index)
+        result = df.rolling("5s").min()
+        expected = DataFrame({"column": [3.0, 3.0, 4.0, 4.0, 6.0]}, index=index)
+        tm.assert_frame_equal(result, expected)
+
+    def test_rolling_on_empty(self):
+        # GH-32385
+        df = DataFrame({"column": []}, index=[])
+        result = df.rolling("5s").min()
+        expected = DataFrame({"column": []}, index=[])
+        tm.assert_frame_equal(result, expected)
+
+    def test_rolling_on_multi_index_level(self):
+        # GH-15584
+        df = DataFrame(
+            {"column": range(6)},
+            index=MultiIndex.from_product(
+                [date_range("20190101", periods=3), range(2)], names=["date", "seq"]
+            ),
+        )
+        result = df.rolling("10D", on=df.index.get_level_values("date")).sum()
+        expected = DataFrame(
+            {"column": [0.0, 1.0, 3.0, 6.0, 10.0, 15.0]}, index=df.index
+        )
+        tm.assert_frame_equal(result, expected)
+
+
+def test_nat_axis_error():
+    idx = [Timestamp("2020"), NaT]
+    df = DataFrame(np.eye(2), index=idx)
+    with pytest.raises(ValueError, match="index values must not have NaT"):
+        df.rolling("D").mean()
+
+
+@td.skip_if_no("pyarrow")
+def test_arrow_datetime_axis():
+    # GH 55849
+    expected = Series(
+        np.arange(5, dtype=np.float64),
+        index=Index(
+            date_range("2020-01-01", periods=5), dtype="timestamp[ns][pyarrow]"
+        ),
+    )
+    result = expected.rolling("1D").sum()
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tests/window/test_win_type.py b/pandas/pandas/tests/window/test_win_type.py
new file mode 100644
index 0000000000000..574dfc34b6d26
--- /dev/null
+++ b/pandas/pandas/tests/window/test_win_type.py
@@ -0,0 +1,670 @@
+import numpy as np
+import pytest
+
+from pandas import (
+    DataFrame,
+    Series,
+    Timedelta,
+    concat,
+    date_range,
+)
+import pandas._testing as tm
+from pandas.api.indexers import BaseIndexer
+
+
+@pytest.fixture(
+    params=[
+        "triang",
+        "blackman",
+        "hamming",
+        "bartlett",
+        "bohman",
+        "blackmanharris",
+        "nuttall",
+        "barthann",
+    ]
+)
+def win_types(request):
+    return request.param
+
+
+@pytest.fixture(params=["kaiser", "gaussian", "general_gaussian", "exponential"])
+def win_types_special(request):
+    return request.param
+
+
+def test_constructor(frame_or_series):
+    # GH 12669
+    pytest.importorskip("scipy")
+    c = frame_or_series(range(5)).rolling
+
+    # valid
+    c(win_type="boxcar", window=2, min_periods=1)
+    c(win_type="boxcar", window=2, min_periods=1, center=True)
+    c(win_type="boxcar", window=2, min_periods=1, center=False)
+
+
+@pytest.mark.parametrize("w", [2.0, "foo", np.array([2])])
+def test_invalid_constructor(frame_or_series, w):
+    # not valid
+    pytest.importorskip("scipy")
+    c = frame_or_series(range(5)).rolling
+    with pytest.raises(ValueError, match="min_periods must be an integer"):
+        c(win_type="boxcar", window=2, min_periods=w)
+    with pytest.raises(ValueError, match="center must be a boolean"):
+        c(win_type="boxcar", window=2, min_periods=1, center=w)
+
+
+@pytest.mark.parametrize("wt", ["foobar", 1])
+def test_invalid_constructor_wintype(frame_or_series, wt):
+    pytest.importorskip("scipy")
+    c = frame_or_series(range(5)).rolling
+    with pytest.raises(ValueError, match="Invalid win_type"):
+        c(win_type=wt, window=2)
+
+
+def test_constructor_with_win_type(frame_or_series, win_types):
+    # GH 12669
+    pytest.importorskip("scipy")
+    c = frame_or_series(range(5)).rolling
+    c(win_type=win_types, window=2)
+
+
+@pytest.mark.parametrize("arg", ["median", "kurt", "skew"])
+def test_agg_function_support(arg):
+    pytest.importorskip("scipy")
+    df = DataFrame({"A": np.arange(5)})
+    roll = df.rolling(2, win_type="triang")
+
+    msg = f"'{arg}' is not a valid function for 'Window' object"
+    with pytest.raises(AttributeError, match=msg):
+        roll.agg(arg)
+
+    with pytest.raises(AttributeError, match=msg):
+        roll.agg([arg])
+
+    with pytest.raises(AttributeError, match=msg):
+        roll.agg({"A": arg})
+
+
+def test_invalid_scipy_arg():
+    # This error is raised by scipy
+    pytest.importorskip("scipy")
+    msg = r"boxcar\(\) got an unexpected"
+    with pytest.raises(TypeError, match=msg):
+        Series(range(3)).rolling(1, win_type="boxcar").mean(foo="bar")
+
+
+def test_constructor_with_win_type_invalid(frame_or_series):
+    # GH 13383
+    pytest.importorskip("scipy")
+    c = frame_or_series(range(5)).rolling
+
+    msg = "window must be an integer 0 or greater"
+
+    with pytest.raises(ValueError, match=msg):
+        c(-1, win_type="boxcar")
+
+
+def test_window_with_args(step):
+    # make sure that we are aggregating window functions correctly with arg
+    pytest.importorskip("scipy")
+    r = Series(np.random.default_rng(2).standard_normal(100)).rolling(
+        window=10, min_periods=1, win_type="gaussian", step=step
+    )
+    expected = concat([r.mean(std=10), r.mean(std=0.01)], axis=1)
+    expected.columns = ["<lambda>", "<lambda>"]
+    result = r.aggregate([lambda x: x.mean(std=10), lambda x: x.mean(std=0.01)])
+    tm.assert_frame_equal(result, expected)
+
+    def a(x):
+        return x.mean(std=10)
+
+    def b(x):
+        return x.mean(std=0.01)
+
+    expected = concat([r.mean(std=10), r.mean(std=0.01)], axis=1)
+    expected.columns = ["a", "b"]
+    result = r.aggregate([a, b])
+    tm.assert_frame_equal(result, expected)
+
+
+def test_win_type_with_method_invalid():
+    pytest.importorskip("scipy")
+    with pytest.raises(
+        NotImplementedError, match="'single' is the only supported method type."
+    ):
+        Series(range(1)).rolling(1, win_type="triang", method="table")
+
+
+@pytest.mark.parametrize("arg", [2000000000, "2s", Timedelta("2s")])
+def test_consistent_win_type_freq(arg):
+    # GH 15969
+    pytest.importorskip("scipy")
+    s = Series(range(1))
+    with pytest.raises(ValueError, match="Invalid win_type freq"):
+        s.rolling(arg, win_type="freq")
+
+
+def test_win_type_freq_return_none():
+    # GH 48838
+    freq_roll = Series(range(2), index=date_range("2020", periods=2)).rolling("2s")
+    assert freq_roll.win_type is None
+
+
+def test_win_type_not_implemented():
+    pytest.importorskip("scipy")
+
+    class CustomIndexer(BaseIndexer):
+        def get_window_bounds(self, num_values, min_periods, center, closed, step):
+            return np.array([0, 1]), np.array([1, 2])
+
+    df = DataFrame({"values": range(2)})
+    indexer = CustomIndexer()
+    with pytest.raises(NotImplementedError, match="BaseIndexer subclasses not"):
+        df.rolling(indexer, win_type="boxcar")
+
+
+def test_cmov_mean(step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    vals = np.array([6.95, 15.21, 4.72, 9.12, 13.81, 13.49, 16.68, 9.48, 10.63, 14.48])
+    result = Series(vals).rolling(5, center=True, step=step).mean()
+    expected_values = [
+        np.nan,
+        np.nan,
+        9.962,
+        11.27,
+        11.564,
+        12.516,
+        12.818,
+        12.952,
+        np.nan,
+        np.nan,
+    ]
+    expected = Series(expected_values)[::step]
+    tm.assert_series_equal(expected, result)
+
+
+def test_cmov_window(step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    vals = np.array([6.95, 15.21, 4.72, 9.12, 13.81, 13.49, 16.68, 9.48, 10.63, 14.48])
+    result = Series(vals).rolling(5, win_type="boxcar", center=True, step=step).mean()
+    expected_values = [
+        np.nan,
+        np.nan,
+        9.962,
+        11.27,
+        11.564,
+        12.516,
+        12.818,
+        12.952,
+        np.nan,
+        np.nan,
+    ]
+    expected = Series(expected_values)[::step]
+    tm.assert_series_equal(expected, result)
+
+
+def test_cmov_window_corner(step):
+    # GH 8238
+    # all nan
+    pytest.importorskip("scipy")
+    vals = Series([np.nan] * 10)
+    result = vals.rolling(5, center=True, win_type="boxcar", step=step).mean()
+    assert np.isnan(result).all()
+
+    # empty
+    vals = Series([], dtype=object)
+    result = vals.rolling(5, center=True, win_type="boxcar", step=step).mean()
+    assert len(result) == 0
+
+    # shorter than window
+    vals = Series(np.random.default_rng(2).standard_normal(5))
+    result = vals.rolling(10, win_type="boxcar", step=step).mean()
+    assert np.isnan(result).all()
+    assert len(result) == len(range(0, 5, step or 1))
+
+
+@pytest.mark.parametrize(
+    "f,xp",
+    [
+        (
+            "mean",
+            [
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+                [9.252, 9.392],
+                [8.644, 9.906],
+                [8.87, 10.208],
+                [6.81, 8.588],
+                [7.792, 8.644],
+                [9.05, 7.824],
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+            ],
+        ),
+        (
+            "std",
+            [
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+                [3.789706, 4.068313],
+                [3.429232, 3.237411],
+                [3.589269, 3.220810],
+                [3.405195, 2.380655],
+                [3.281839, 2.369869],
+                [3.676846, 1.801799],
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+            ],
+        ),
+        (
+            "var",
+            [
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+                [14.36187, 16.55117],
+                [11.75963, 10.48083],
+                [12.88285, 10.37362],
+                [11.59535, 5.66752],
+                [10.77047, 5.61628],
+                [13.51920, 3.24648],
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+            ],
+        ),
+        (
+            "sum",
+            [
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+                [46.26, 46.96],
+                [43.22, 49.53],
+                [44.35, 51.04],
+                [34.05, 42.94],
+                [38.96, 43.22],
+                [45.25, 39.12],
+                [np.nan, np.nan],
+                [np.nan, np.nan],
+            ],
+        ),
+    ],
+)
+def test_cmov_window_frame(f, xp, step):
+    # Gh 8238
+    pytest.importorskip("scipy")
+    df = DataFrame(
+        np.array(
+            [
+                [12.18, 3.64],
+                [10.18, 9.16],
+                [13.24, 14.61],
+                [4.51, 8.11],
+                [6.15, 11.44],
+                [9.14, 6.21],
+                [11.31, 10.67],
+                [2.94, 6.51],
+                [9.42, 8.39],
+                [12.44, 7.34],
+            ]
+        )
+    )
+    xp = DataFrame(np.array(xp))[::step]
+
+    roll = df.rolling(5, win_type="boxcar", center=True, step=step)
+    rs = getattr(roll, f)()
+
+    tm.assert_frame_equal(xp, rs)
+
+
+@pytest.mark.parametrize("min_periods", [0, 1, 2, 3, 4, 5])
+def test_cmov_window_na_min_periods(step, min_periods):
+    pytest.importorskip("scipy")
+    vals = Series(np.random.default_rng(2).standard_normal(10))
+    vals[4] = np.nan
+    vals[8] = np.nan
+
+    xp = vals.rolling(5, min_periods=min_periods, center=True, step=step).mean()
+    rs = vals.rolling(
+        5, win_type="boxcar", min_periods=min_periods, center=True, step=step
+    ).mean()
+    tm.assert_series_equal(xp, rs)
+
+
+def test_cmov_window_regular(win_types, step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    vals = np.array([6.95, 15.21, 4.72, 9.12, 13.81, 13.49, 16.68, 9.48, 10.63, 14.48])
+    xps = {
+        "hamming": [
+            np.nan,
+            np.nan,
+            8.71384,
+            9.56348,
+            12.38009,
+            14.03687,
+            13.8567,
+            11.81473,
+            np.nan,
+            np.nan,
+        ],
+        "triang": [
+            np.nan,
+            np.nan,
+            9.28667,
+            10.34667,
+            12.00556,
+            13.33889,
+            13.38,
+            12.33667,
+            np.nan,
+            np.nan,
+        ],
+        "barthann": [
+            np.nan,
+            np.nan,
+            8.4425,
+            9.1925,
+            12.5575,
+            14.3675,
+            14.0825,
+            11.5675,
+            np.nan,
+            np.nan,
+        ],
+        "bohman": [
+            np.nan,
+            np.nan,
+            7.61599,
+            9.1764,
+            12.83559,
+            14.17267,
+            14.65923,
+            11.10401,
+            np.nan,
+            np.nan,
+        ],
+        "blackmanharris": [
+            np.nan,
+            np.nan,
+            6.97691,
+            9.16438,
+            13.05052,
+            14.02156,
+            15.10512,
+            10.74574,
+            np.nan,
+            np.nan,
+        ],
+        "nuttall": [
+            np.nan,
+            np.nan,
+            7.04618,
+            9.16786,
+            13.02671,
+            14.03559,
+            15.05657,
+            10.78514,
+            np.nan,
+            np.nan,
+        ],
+        "blackman": [
+            np.nan,
+            np.nan,
+            7.73345,
+            9.17869,
+            12.79607,
+            14.20036,
+            14.57726,
+            11.16988,
+            np.nan,
+            np.nan,
+        ],
+        "bartlett": [
+            np.nan,
+            np.nan,
+            8.4425,
+            9.1925,
+            12.5575,
+            14.3675,
+            14.0825,
+            11.5675,
+            np.nan,
+            np.nan,
+        ],
+    }
+
+    xp = Series(xps[win_types])[::step]
+    rs = Series(vals).rolling(5, win_type=win_types, center=True, step=step).mean()
+    tm.assert_series_equal(xp, rs)
+
+
+def test_cmov_window_regular_linear_range(win_types, step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    vals = np.array(range(10), dtype=float)
+    rs = Series(vals).rolling(5, win_type=win_types, center=True, step=step).mean()
+    xp = vals
+    xp[:2] = np.nan
+    xp[-2:] = np.nan
+    xp = Series(xp)[::step]
+
+    tm.assert_series_equal(xp, rs)
+
+
+def test_cmov_window_regular_missing_data(win_types, step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    vals = np.array(
+        [6.95, 15.21, 4.72, 9.12, 13.81, 13.49, 16.68, np.nan, 10.63, 14.48]
+    )
+    xps = {
+        "bartlett": [
+            np.nan,
+            np.nan,
+            9.70333,
+            10.5225,
+            8.4425,
+            9.1925,
+            12.5575,
+            14.3675,
+            15.61667,
+            13.655,
+        ],
+        "blackman": [
+            np.nan,
+            np.nan,
+            9.04582,
+            11.41536,
+            7.73345,
+            9.17869,
+            12.79607,
+            14.20036,
+            15.8706,
+            13.655,
+        ],
+        "barthann": [
+            np.nan,
+            np.nan,
+            9.70333,
+            10.5225,
+            8.4425,
+            9.1925,
+            12.5575,
+            14.3675,
+            15.61667,
+            13.655,
+        ],
+        "bohman": [
+            np.nan,
+            np.nan,
+            8.9444,
+            11.56327,
+            7.61599,
+            9.1764,
+            12.83559,
+            14.17267,
+            15.90976,
+            13.655,
+        ],
+        "hamming": [
+            np.nan,
+            np.nan,
+            9.59321,
+            10.29694,
+            8.71384,
+            9.56348,
+            12.38009,
+            14.20565,
+            15.24694,
+            13.69758,
+        ],
+        "nuttall": [
+            np.nan,
+            np.nan,
+            8.47693,
+            12.2821,
+            7.04618,
+            9.16786,
+            13.02671,
+            14.03673,
+            16.08759,
+            13.65553,
+        ],
+        "triang": [
+            np.nan,
+            np.nan,
+            9.33167,
+            9.76125,
+            9.28667,
+            10.34667,
+            12.00556,
+            13.82125,
+            14.49429,
+            13.765,
+        ],
+        "blackmanharris": [
+            np.nan,
+            np.nan,
+            8.42526,
+            12.36824,
+            6.97691,
+            9.16438,
+            13.05052,
+            14.02175,
+            16.1098,
+            13.65509,
+        ],
+    }
+
+    xp = Series(xps[win_types])[::step]
+    rs = Series(vals).rolling(5, win_type=win_types, min_periods=3, step=step).mean()
+    tm.assert_series_equal(xp, rs)
+
+
+def test_cmov_window_special(win_types_special, step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    kwds = {
+        "kaiser": {"beta": 1.0},
+        "gaussian": {"std": 1.0},
+        "general_gaussian": {"p": 2.0, "sig": 2.0},
+        "exponential": {"tau": 10},
+    }
+
+    vals = np.array([6.95, 15.21, 4.72, 9.12, 13.81, 13.49, 16.68, 9.48, 10.63, 14.48])
+
+    xps = {
+        "gaussian": [
+            np.nan,
+            np.nan,
+            8.97297,
+            9.76077,
+            12.24763,
+            13.89053,
+            13.65671,
+            12.01002,
+            np.nan,
+            np.nan,
+        ],
+        "general_gaussian": [
+            np.nan,
+            np.nan,
+            9.85011,
+            10.71589,
+            11.73161,
+            13.08516,
+            12.95111,
+            12.74577,
+            np.nan,
+            np.nan,
+        ],
+        "kaiser": [
+            np.nan,
+            np.nan,
+            9.86851,
+            11.02969,
+            11.65161,
+            12.75129,
+            12.90702,
+            12.83757,
+            np.nan,
+            np.nan,
+        ],
+        "exponential": [
+            np.nan,
+            np.nan,
+            9.83364,
+            11.10472,
+            11.64551,
+            12.66138,
+            12.92379,
+            12.83770,
+            np.nan,
+            np.nan,
+        ],
+    }
+
+    xp = Series(xps[win_types_special])[::step]
+    rs = (
+        Series(vals)
+        .rolling(5, win_type=win_types_special, center=True, step=step)
+        .mean(**kwds[win_types_special])
+    )
+    tm.assert_series_equal(xp, rs)
+
+
+def test_cmov_window_special_linear_range(win_types_special, step):
+    # GH 8238
+    pytest.importorskip("scipy")
+    kwds = {
+        "kaiser": {"beta": 1.0},
+        "gaussian": {"std": 1.0},
+        "general_gaussian": {"p": 2.0, "sig": 2.0},
+        "slepian": {"width": 0.5},
+        "exponential": {"tau": 10},
+    }
+
+    vals = np.array(range(10), dtype=float)
+    rs = (
+        Series(vals)
+        .rolling(5, win_type=win_types_special, center=True, step=step)
+        .mean(**kwds[win_types_special])
+    )
+    xp = vals
+    xp[:2] = np.nan
+    xp[-2:] = np.nan
+    xp = Series(xp)[::step]
+    tm.assert_series_equal(xp, rs)
+
+
+def test_weighted_var_big_window_no_segfault(win_types, center):
+    # GitHub Issue #46772
+    pytest.importorskip("scipy")
+    x = Series(0)
+    result = x.rolling(window=16, center=center, win_type=win_types).var()
+    expected = Series(np.nan)
+
+    tm.assert_series_equal(result, expected)
diff --git a/pandas/pandas/tseries/__init__.py b/pandas/pandas/tseries/__init__.py
new file mode 100644
index 0000000000000..c00843ecac418
--- /dev/null
+++ b/pandas/pandas/tseries/__init__.py
@@ -0,0 +1,12 @@
+# ruff: noqa: TC004
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    # import modules that have public classes/functions:
+    from pandas.tseries import (
+        frequencies,
+        offsets,
+    )
+
+    # and mark only those modules as public
+    __all__ = ["frequencies", "offsets"]
diff --git a/pandas/pandas/tseries/api.py b/pandas/pandas/tseries/api.py
new file mode 100644
index 0000000000000..5ea899f1610a7
--- /dev/null
+++ b/pandas/pandas/tseries/api.py
@@ -0,0 +1,10 @@
+"""
+Timeseries API
+"""
+
+from pandas._libs.tslibs.parsing import guess_datetime_format
+
+from pandas.tseries import offsets
+from pandas.tseries.frequencies import infer_freq
+
+__all__ = ["guess_datetime_format", "infer_freq", "offsets"]
diff --git a/pandas/pandas/tseries/frequencies.py b/pandas/pandas/tseries/frequencies.py
new file mode 100644
index 0000000000000..88ea1bfa3c6ed
--- /dev/null
+++ b/pandas/pandas/tseries/frequencies.py
@@ -0,0 +1,610 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+from pandas._libs import lib
+from pandas._libs.algos import unique_deltas
+from pandas._libs.tslibs import (
+    Timestamp,
+    get_unit_from_dtype,
+    periods_per_day,
+    tz_convert_from_utc,
+)
+from pandas._libs.tslibs.ccalendar import (
+    DAYS,
+    MONTH_ALIASES,
+    MONTH_NUMBERS,
+    MONTHS,
+    int_to_weekday,
+)
+from pandas._libs.tslibs.dtypes import OFFSET_TO_PERIOD_FREQSTR
+from pandas._libs.tslibs.fields import (
+    build_field_sarray,
+    month_position_check,
+)
+from pandas._libs.tslibs.offsets import (
+    DateOffset,
+    Day,
+    to_offset,
+)
+from pandas._libs.tslibs.parsing import get_rule_month
+from pandas.util._decorators import cache_readonly
+
+from pandas.core.dtypes.common import is_numeric_dtype
+from pandas.core.dtypes.dtypes import (
+    DatetimeTZDtype,
+    PeriodDtype,
+)
+from pandas.core.dtypes.generic import (
+    ABCIndex,
+    ABCSeries,
+)
+
+from pandas.core.algorithms import unique
+
+if TYPE_CHECKING:
+    from pandas._typing import npt
+
+    from pandas import (
+        DatetimeIndex,
+        Series,
+        TimedeltaIndex,
+    )
+    from pandas.core.arrays.datetimelike import DatetimeLikeArrayMixin
+# --------------------------------------------------------------------
+# Offset related functions
+
+_need_suffix = ["QS", "BQE", "BQS", "YS", "BYE", "BYS"]
+
+for _prefix in _need_suffix:
+    for _m in MONTHS:
+        key = f"{_prefix}-{_m}"
+        OFFSET_TO_PERIOD_FREQSTR[key] = OFFSET_TO_PERIOD_FREQSTR[_prefix]
+
+for _prefix in ["Y", "Q"]:
+    for _m in MONTHS:
+        _alias = f"{_prefix}-{_m}"
+        OFFSET_TO_PERIOD_FREQSTR[_alias] = _alias
+
+for _d in DAYS:
+    OFFSET_TO_PERIOD_FREQSTR[f"W-{_d}"] = f"W-{_d}"
+
+
+def get_period_alias(offset_str: str) -> str | None:
+    """
+    Alias to closest period strings BQ->Q etc.
+    """
+    return OFFSET_TO_PERIOD_FREQSTR.get(offset_str, None)
+
+
+# ---------------------------------------------------------------------
+# Period codes
+
+
+def infer_freq(
+    index: DatetimeIndex | TimedeltaIndex | Series | DatetimeLikeArrayMixin,
+) -> str | None:
+    """
+    Infer the most likely frequency given the input index.
+
+    This method attempts to deduce the most probable frequency (e.g., 'D' for daily,
+    'H' for hourly) from a sequence of datetime-like objects. It is particularly useful
+    when the frequency of a time series is not explicitly set or known but can be
+    inferred from its values.
+
+    Parameters
+    ----------
+    index : DatetimeIndex, TimedeltaIndex, Series or array-like
+      If passed a Series will use the values of the series (NOT THE INDEX).
+
+    Returns
+    -------
+    str or None
+        None if no discernible frequency.
+
+    Raises
+    ------
+    TypeError
+        If the index is not datetime-like.
+    ValueError
+        If there are fewer than three values.
+
+    See Also
+    --------
+    date_range : Return a fixed frequency DatetimeIndex.
+    timedelta_range : Return a fixed frequency TimedeltaIndex with day as the default.
+    period_range : Return a fixed frequency PeriodIndex.
+    DatetimeIndex.freq : Return the frequency object if it is set, otherwise None.
+
+    Examples
+    --------
+    >>> idx = pd.date_range(start="2020/12/01", end="2020/12/30", periods=30)
+    >>> pd.infer_freq(idx)
+    'D'
+    """
+    from pandas.core.api import DatetimeIndex
+
+    if isinstance(index, ABCSeries):
+        values = index._values
+        if not (
+            lib.is_np_dtype(values.dtype, "mM")
+            or isinstance(values.dtype, DatetimeTZDtype)
+            or values.dtype == object
+        ):
+            raise TypeError(
+                "cannot infer freq from a non-convertible dtype "
+                f"on a Series of {index.dtype}"
+            )
+        index = values
+
+    inferer: _FrequencyInferer
+
+    if not hasattr(index, "dtype"):
+        pass
+    elif isinstance(index.dtype, PeriodDtype):
+        raise TypeError(
+            "PeriodIndex given. Check the `freq` attribute instead of using infer_freq."
+        )
+    elif lib.is_np_dtype(index.dtype, "m"):
+        # Allow TimedeltaIndex and TimedeltaArray
+        inferer = _TimedeltaFrequencyInferer(index)
+        return inferer.get_freq()
+
+    elif is_numeric_dtype(index.dtype):
+        raise TypeError(
+            f"cannot infer freq from a non-convertible index of dtype {index.dtype}"
+        )
+
+    if not isinstance(index, DatetimeIndex):
+        index = DatetimeIndex(index)
+
+    inferer = _FrequencyInferer(index)
+    return inferer.get_freq()
+
+
+class _FrequencyInferer:
+    """
+    Not sure if I can avoid the state machine here
+    """
+
+    def __init__(self, index) -> None:
+        self.index = index
+        self.i8values = index.asi8
+
+        # For get_unit_from_dtype we need the dtype to the underlying ndarray,
+        #  which for tz-aware is not the same as index.dtype
+        if isinstance(index, ABCIndex):
+            # error: Item "ndarray[Any, Any]" of "Union[ExtensionArray,
+            # ndarray[Any, Any]]" has no attribute "_ndarray"
+            self._creso = get_unit_from_dtype(
+                index._data._ndarray.dtype  # type: ignore[union-attr]
+            )
+        else:
+            # otherwise we have DTA/TDA
+            self._creso = get_unit_from_dtype(index._ndarray.dtype)
+
+        # This moves the values, which are implicitly in UTC, to the
+        # the timezone so they are in local time
+        if hasattr(index, "tz"):
+            if index.tz is not None:
+                self.i8values = tz_convert_from_utc(
+                    self.i8values, index.tz, reso=self._creso
+                )
+
+        if len(index) < 3:
+            raise ValueError("Need at least 3 dates to infer frequency")
+
+        self.is_monotonic = (
+            self.index._is_monotonic_increasing or self.index._is_monotonic_decreasing
+        )
+
+    @cache_readonly
+    def deltas(self) -> npt.NDArray[np.int64]:
+        return unique_deltas(self.i8values)
+
+    @cache_readonly
+    def deltas_asi8(self) -> npt.NDArray[np.int64]:
+        # NB: we cannot use self.i8values here because we may have converted
+        #  the tz in __init__
+        return unique_deltas(self.index.asi8)
+
+    @cache_readonly
+    def is_unique(self) -> bool:
+        return len(self.deltas) == 1
+
+    @cache_readonly
+    def is_unique_asi8(self) -> bool:
+        return len(self.deltas_asi8) == 1
+
+    def get_freq(self) -> str | None:
+        """
+        Find the appropriate frequency string to describe the inferred
+        frequency of self.i8values
+
+        Returns
+        -------
+        str or None
+        """
+        if not self.is_monotonic or not self.index._is_unique:
+            return None
+
+        delta = self.deltas[0]
+        ppd = periods_per_day(self._creso)
+        if delta and _is_multiple(delta, ppd):
+            return self._infer_daily_rule()
+
+        # Business hourly, maybe. 17: one day / 65: one weekend
+        if self.hour_deltas in ([1, 17], [1, 65], [1, 17, 65]):
+            return "bh"
+
+        # Possibly intraday frequency.  Here we use the
+        # original .asi8 values as the modified values
+        # will not work around DST transitions.  See #8772
+        if not self.is_unique_asi8:
+            return None
+
+        delta = self.deltas_asi8[0]
+        pph = ppd // 24
+        ppm = pph // 60
+        pps = ppm // 60
+        if _is_multiple(delta, pph):
+            # Hours
+            return _maybe_add_count("h", delta / pph)
+        elif _is_multiple(delta, ppm):
+            # Minutes
+            return _maybe_add_count("min", delta / ppm)
+        elif _is_multiple(delta, pps):
+            # Seconds
+            return _maybe_add_count("s", delta / pps)
+        elif _is_multiple(delta, (pps // 1000)):
+            # Milliseconds
+            return _maybe_add_count("ms", delta / (pps // 1000))
+        elif _is_multiple(delta, (pps // 1_000_000)):
+            # Microseconds
+            return _maybe_add_count("us", delta / (pps // 1_000_000))
+        else:
+            # Nanoseconds
+            return _maybe_add_count("ns", delta)
+
+    @cache_readonly
+    def day_deltas(self) -> list[int]:
+        ppd = periods_per_day(self._creso)
+        return [x / ppd for x in self.deltas]
+
+    @cache_readonly
+    def hour_deltas(self) -> list[int]:
+        pph = periods_per_day(self._creso) // 24
+        return [x / pph for x in self.deltas]
+
+    @cache_readonly
+    def fields(self) -> np.ndarray:  # structured array of fields
+        return build_field_sarray(self.i8values, reso=self._creso)
+
+    @cache_readonly
+    def rep_stamp(self) -> Timestamp:
+        return Timestamp(self.i8values[0], unit=self.index.unit)
+
+    def month_position_check(self) -> str | None:
+        return month_position_check(self.fields, self.index.dayofweek)
+
+    @cache_readonly
+    def mdiffs(self) -> npt.NDArray[np.int64]:
+        nmonths = self.fields["Y"] * 12 + self.fields["M"]
+        return unique_deltas(nmonths.astype("i8"))
+
+    @cache_readonly
+    def ydiffs(self) -> npt.NDArray[np.int64]:
+        return unique_deltas(self.fields["Y"].astype("i8"))
+
+    def _infer_daily_rule(self) -> str | None:
+        annual_rule = self._get_annual_rule()
+        if annual_rule:
+            nyears = self.ydiffs[0]
+            month = MONTH_ALIASES[self.rep_stamp.month]
+            alias = f"{annual_rule}-{month}"
+            return _maybe_add_count(alias, nyears)
+
+        quarterly_rule = self._get_quarterly_rule()
+        if quarterly_rule:
+            nquarters = self.mdiffs[0] / 3
+            mod_dict = {0: 12, 2: 11, 1: 10}
+            month = MONTH_ALIASES[mod_dict[self.rep_stamp.month % 3]]
+            alias = f"{quarterly_rule}-{month}"
+            return _maybe_add_count(alias, nquarters)
+
+        monthly_rule = self._get_monthly_rule()
+        if monthly_rule:
+            return _maybe_add_count(monthly_rule, self.mdiffs[0])
+
+        if self.is_unique:
+            return self._get_daily_rule()
+
+        if self._is_business_daily():
+            return "B"
+
+        wom_rule = self._get_wom_rule()
+        if wom_rule:
+            return wom_rule
+
+        return None
+
+    def _get_daily_rule(self) -> str | None:
+        ppd = periods_per_day(self._creso)
+        days = self.deltas[0] / ppd
+        if days % 7 == 0:
+            # Weekly
+            wd = int_to_weekday[self.rep_stamp.weekday()]
+            alias = f"W-{wd}"
+            return _maybe_add_count(alias, days / 7)
+        else:
+            return _maybe_add_count("D", days)
+
+    def _get_annual_rule(self) -> str | None:
+        if len(self.ydiffs) > 1:
+            return None
+
+        if len(unique(self.fields["M"])) > 1:
+            return None
+
+        pos_check = self.month_position_check()
+
+        if pos_check is None:
+            return None
+        else:
+            return {"cs": "YS", "bs": "BYS", "ce": "YE", "be": "BYE"}.get(pos_check)
+
+    def _get_quarterly_rule(self) -> str | None:
+        if len(self.mdiffs) > 1:
+            return None
+
+        if not self.mdiffs[0] % 3 == 0:
+            return None
+
+        pos_check = self.month_position_check()
+
+        if pos_check is None:
+            return None
+        else:
+            return {"cs": "QS", "bs": "BQS", "ce": "QE", "be": "BQE"}.get(pos_check)
+
+    def _get_monthly_rule(self) -> str | None:
+        if len(self.mdiffs) > 1:
+            return None
+        pos_check = self.month_position_check()
+
+        if pos_check is None:
+            return None
+        else:
+            return {"cs": "MS", "bs": "BMS", "ce": "ME", "be": "BME"}.get(pos_check)
+
+    def _is_business_daily(self) -> bool:
+        # quick check: cannot be business daily
+        if self.day_deltas != [1, 3]:
+            return False
+
+        # probably business daily, but need to confirm
+        first_weekday = self.index[0].weekday()
+        shifts = np.diff(self.i8values)
+        ppd = periods_per_day(self._creso)
+        shifts = np.floor_divide(shifts, ppd)
+        weekdays = np.mod(first_weekday + np.cumsum(shifts), 7)
+
+        return bool(
+            np.all(
+                ((weekdays == 0) & (shifts == 3))
+                | ((weekdays > 0) & (weekdays <= 4) & (shifts == 1))
+            )
+        )
+
+    def _get_wom_rule(self) -> str | None:
+        weekdays = unique(self.index.weekday)
+        if len(weekdays) > 1:
+            return None
+
+        week_of_months = unique((self.index.day - 1) // 7)
+        # Only attempt to infer up to WOM-4. See #9425
+        week_of_months = week_of_months[week_of_months < 4]
+        if len(week_of_months) == 0 or len(week_of_months) > 1:
+            return None
+
+        # get which week
+        week = week_of_months[0] + 1
+        wd = int_to_weekday[weekdays[0]]
+
+        return f"WOM-{week}{wd}"
+
+
+class _TimedeltaFrequencyInferer(_FrequencyInferer):
+    def _infer_daily_rule(self):
+        if self.is_unique:
+            return self._get_daily_rule()
+
+
+def _is_multiple(us, mult: int) -> bool:
+    return us % mult == 0
+
+
+def _maybe_add_count(base: str, count: float) -> str:
+    if count != 1:
+        assert count == int(count)
+        count = int(count)
+        return f"{count}{base}"
+    else:
+        return base
+
+
+# ----------------------------------------------------------------------
+# Frequency comparison
+
+
+def is_subperiod(source, target) -> bool:
+    """
+    Returns True if downsampling is possible between source and target
+    frequencies
+
+    Parameters
+    ----------
+    source : str or DateOffset
+        Frequency converting from
+    target : str or DateOffset
+        Frequency converting to
+
+    Returns
+    -------
+    bool
+    """
+    if target is None or source is None:
+        return False
+    source = _maybe_coerce_freq(source)
+    target = _maybe_coerce_freq(target)
+
+    if _is_annual(target):
+        if _is_quarterly(source):
+            return _quarter_months_conform(
+                get_rule_month(source), get_rule_month(target)
+            )
+        return source in {"D", "C", "B", "M", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_quarterly(target):
+        return source in {"D", "C", "B", "M", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_monthly(target):
+        return source in {"D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_weekly(target):
+        return source in {target, "D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif target == "B":
+        return source in {"B", "h", "min", "s", "ms", "us", "ns"}
+    elif target == "C":
+        return source in {"C", "h", "min", "s", "ms", "us", "ns"}
+    elif target == "D":
+        return source in {"D", "h", "min", "s", "ms", "us", "ns"}
+    elif target == "h":
+        return source in {"h", "min", "s", "ms", "us", "ns"}
+    elif target == "min":
+        return source in {"min", "s", "ms", "us", "ns"}
+    elif target == "s":
+        return source in {"s", "ms", "us", "ns"}
+    elif target == "ms":
+        return source in {"ms", "us", "ns"}
+    elif target == "us":
+        return source in {"us", "ns"}
+    elif target == "ns":
+        return source in {"ns"}
+    else:
+        return False
+
+
+def is_superperiod(source, target) -> bool:
+    """
+    Returns True if upsampling is possible between source and target
+    frequencies
+
+    Parameters
+    ----------
+    source : str or DateOffset
+        Frequency converting from
+    target : str or DateOffset
+        Frequency converting to
+
+    Returns
+    -------
+    bool
+    """
+    if target is None or source is None:
+        return False
+    source = _maybe_coerce_freq(source)
+    target = _maybe_coerce_freq(target)
+
+    if _is_annual(source):
+        if _is_annual(target):
+            return get_rule_month(source) == get_rule_month(target)
+
+        if _is_quarterly(target):
+            smonth = get_rule_month(source)
+            tmonth = get_rule_month(target)
+            return _quarter_months_conform(smonth, tmonth)
+        return target in {"D", "C", "B", "M", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_quarterly(source):
+        return target in {"D", "C", "B", "M", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_monthly(source):
+        return target in {"D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif _is_weekly(source):
+        return target in {source, "D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif source == "B":
+        return target in {"D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif source == "C":
+        return target in {"D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif source == "D":
+        return target in {"D", "C", "B", "h", "min", "s", "ms", "us", "ns"}
+    elif source == "h":
+        return target in {"h", "min", "s", "ms", "us", "ns"}
+    elif source == "min":
+        return target in {"min", "s", "ms", "us", "ns"}
+    elif source == "s":
+        return target in {"s", "ms", "us", "ns"}
+    elif source == "ms":
+        return target in {"ms", "us", "ns"}
+    elif source == "us":
+        return target in {"us", "ns"}
+    elif source == "ns":
+        return target in {"ns"}
+    else:
+        return False
+
+
+def _maybe_coerce_freq(code) -> str:
+    """we might need to coerce a code to a rule_code
+    and uppercase it
+
+    Parameters
+    ----------
+    source : str or DateOffset
+        Frequency converting from
+
+    Returns
+    -------
+    str
+    """
+    assert code is not None
+    if isinstance(code, DateOffset):
+        code = PeriodDtype(to_offset(code.name))._freqstr
+    if code in {"h", "min", "s", "ms", "us", "ns"}:
+        return code
+    else:
+        return code.upper()
+
+
+def _quarter_months_conform(source: str, target: str) -> bool:
+    snum = MONTH_NUMBERS[source]
+    tnum = MONTH_NUMBERS[target]
+    return snum % 3 == tnum % 3
+
+
+def _is_annual(rule: str) -> bool:
+    rule = rule.upper()
+    return rule == "Y" or rule.startswith("Y-")
+
+
+def _is_quarterly(rule: str) -> bool:
+    rule = rule.upper()
+    return rule == "Q" or rule.startswith(("Q-", "BQ"))
+
+
+def _is_monthly(rule: str) -> bool:
+    rule = rule.upper()
+    return rule in ("M", "BM")
+
+
+def _is_weekly(rule: str) -> bool:
+    rule = rule.upper()
+    return rule == "W" or rule.startswith("W-")
+
+
+__all__ = [
+    "Day",
+    "get_period_alias",
+    "infer_freq",
+    "is_subperiod",
+    "is_superperiod",
+    "to_offset",
+]
diff --git a/pandas/pandas/tseries/holiday.py b/pandas/pandas/tseries/holiday.py
new file mode 100644
index 0000000000000..2d195fbbc4e84
--- /dev/null
+++ b/pandas/pandas/tseries/holiday.py
@@ -0,0 +1,659 @@
+from __future__ import annotations
+
+from datetime import (
+    datetime,
+    timedelta,
+)
+from typing import TYPE_CHECKING
+import warnings
+
+from dateutil.relativedelta import (
+    FR,
+    MO,
+    SA,
+    SU,
+    TH,
+    TU,
+    WE,
+)
+import numpy as np
+
+from pandas._libs.tslibs.offsets import BaseOffset
+from pandas.errors import PerformanceWarning
+
+from pandas import (
+    DateOffset,
+    DatetimeIndex,
+    Series,
+    Timestamp,
+    concat,
+    date_range,
+)
+
+from pandas.tseries.offsets import (
+    Day,
+    Easter,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+
+def next_monday(dt: datetime) -> datetime:
+    """
+    If holiday falls on Saturday, use following Monday instead;
+    if holiday falls on Sunday, use Monday instead
+    """
+    if dt.weekday() == 5:
+        return dt + timedelta(2)
+    elif dt.weekday() == 6:
+        return dt + timedelta(1)
+    return dt
+
+
+def next_monday_or_tuesday(dt: datetime) -> datetime:
+    """
+    For second holiday of two adjacent ones!
+    If holiday falls on Saturday, use following Monday instead;
+    if holiday falls on Sunday or Monday, use following Tuesday instead
+    (because Monday is already taken by adjacent holiday on the day before)
+    """
+    dow = dt.weekday()
+    if dow in (5, 6):
+        return dt + timedelta(2)
+    if dow == 0:
+        return dt + timedelta(1)
+    return dt
+
+
+def previous_friday(dt: datetime) -> datetime:
+    """
+    If holiday falls on Saturday or Sunday, use previous Friday instead.
+    """
+    if dt.weekday() == 5:
+        return dt - timedelta(1)
+    elif dt.weekday() == 6:
+        return dt - timedelta(2)
+    return dt
+
+
+def sunday_to_monday(dt: datetime) -> datetime:
+    """
+    If holiday falls on Sunday, use day thereafter (Monday) instead.
+    """
+    if dt.weekday() == 6:
+        return dt + timedelta(1)
+    return dt
+
+
+def weekend_to_monday(dt: datetime) -> datetime:
+    """
+    If holiday falls on Sunday or Saturday,
+    use day thereafter (Monday) instead.
+    Needed for holidays such as Christmas observation in Europe
+    """
+    if dt.weekday() == 6:
+        return dt + timedelta(1)
+    elif dt.weekday() == 5:
+        return dt + timedelta(2)
+    return dt
+
+
+def nearest_workday(dt: datetime) -> datetime:
+    """
+    If holiday falls on Saturday, use day before (Friday) instead;
+    if holiday falls on Sunday, use day thereafter (Monday) instead.
+    """
+    if dt.weekday() == 5:
+        return dt - timedelta(1)
+    elif dt.weekday() == 6:
+        return dt + timedelta(1)
+    return dt
+
+
+def next_workday(dt: datetime) -> datetime:
+    """
+    returns next workday used for observances
+    """
+    dt += timedelta(days=1)
+    while dt.weekday() > 4:
+        # Mon-Fri are 0-4
+        dt += timedelta(days=1)
+    return dt
+
+
+def previous_workday(dt: datetime) -> datetime:
+    """
+    returns previous workday used for observances
+    """
+    dt -= timedelta(days=1)
+    while dt.weekday() > 4:
+        # Mon-Fri are 0-4
+        dt -= timedelta(days=1)
+    return dt
+
+
+def before_nearest_workday(dt: datetime) -> datetime:
+    """
+    returns previous workday before nearest workday
+    """
+    return previous_workday(nearest_workday(dt))
+
+
+def after_nearest_workday(dt: datetime) -> datetime:
+    """
+    returns next workday after nearest workday
+    needed for Boxing day or multiple holidays in a series
+    """
+    return next_workday(nearest_workday(dt))
+
+
+class Holiday:
+    """
+    Class that defines a holiday with start/end dates and rules
+    for observance.
+    """
+
+    start_date: Timestamp | None
+    end_date: Timestamp | None
+    days_of_week: tuple[int, ...] | None
+
+    def __init__(
+        self,
+        name: str,
+        year=None,
+        month=None,
+        day=None,
+        offset: BaseOffset | list[BaseOffset] | None = None,
+        observance: Callable | None = None,
+        start_date=None,
+        end_date=None,
+        days_of_week: tuple | None = None,
+    ) -> None:
+        """
+        Parameters
+        ----------
+        name : str
+            Name of the holiday , defaults to class name
+        year : int, default None
+            Year of the holiday
+        month : int, default None
+            Month of the holiday
+        day : int, default None
+            Day of the holiday
+        offset : list of pandas.tseries.offsets or
+                class from pandas.tseries.offsets, default None
+            Computes offset from date
+        observance : function, default None
+            Computes when holiday is given a pandas Timestamp
+        start_date : datetime-like, default None
+            First date the holiday is observed
+        end_date : datetime-like, default None
+            Last date the holiday is observed
+        days_of_week : tuple of int or dateutil.relativedelta weekday strs, default None
+            Provide a tuple of days e.g  (0,1,2,3,) for Monday Through Thursday
+            Monday=0,..,Sunday=6
+
+        Examples
+        --------
+        >>> from dateutil.relativedelta import MO
+
+        >>> USMemorialDay = pd.tseries.holiday.Holiday(
+        ...     "Memorial Day", month=5, day=31, offset=pd.DateOffset(weekday=MO(-1))
+        ... )
+        >>> USMemorialDay
+        Holiday: Memorial Day (month=5, day=31, offset=<DateOffset: weekday=MO(-1)>)
+
+        >>> USLaborDay = pd.tseries.holiday.Holiday(
+        ...     "Labor Day", month=9, day=1, offset=pd.DateOffset(weekday=MO(1))
+        ... )
+        >>> USLaborDay
+        Holiday: Labor Day (month=9, day=1, offset=<DateOffset: weekday=MO(+1)>)
+
+        >>> July3rd = pd.tseries.holiday.Holiday("July 3rd", month=7, day=3)
+        >>> July3rd
+        Holiday: July 3rd (month=7, day=3, )
+
+        >>> NewYears = pd.tseries.holiday.Holiday(
+        ...     "New Years Day",
+        ...     month=1,
+        ...     day=1,
+        ...     observance=pd.tseries.holiday.nearest_workday,
+        ... )
+        >>> NewYears  # doctest: +SKIP
+        Holiday: New Years Day (
+            month=1, day=1, observance=<function nearest_workday at 0x66545e9bc440>
+        )
+
+        >>> July3rd = pd.tseries.holiday.Holiday(
+        ...     "July 3rd", month=7, day=3, days_of_week=(0, 1, 2, 3)
+        ... )
+        >>> July3rd
+        Holiday: July 3rd (month=7, day=3, )
+        """
+        if offset is not None:
+            if observance is not None:
+                raise NotImplementedError("Cannot use both offset and observance.")
+            if not (
+                isinstance(offset, BaseOffset)
+                or (
+                    isinstance(offset, list)
+                    and all(isinstance(off, BaseOffset) for off in offset)
+                )
+            ):
+                raise ValueError(
+                    "Only BaseOffsets and flat lists of them are supported for offset."
+                )
+
+        self.name = name
+        self.year = year
+        self.month = month
+        self.day = day
+        self.offset = offset
+        self.start_date = (
+            Timestamp(start_date) if start_date is not None else start_date
+        )
+        self.end_date = Timestamp(end_date) if end_date is not None else end_date
+        self.observance = observance
+        assert days_of_week is None or type(days_of_week) == tuple
+        self.days_of_week = days_of_week
+
+    def __repr__(self) -> str:
+        info = ""
+        if self.year is not None:
+            info += f"year={self.year}, "
+        info += f"month={self.month}, day={self.day}, "
+
+        if self.offset is not None:
+            info += f"offset={self.offset}"
+
+        if self.observance is not None:
+            info += f"observance={self.observance}"
+
+        repr = f"Holiday: {self.name} ({info})"
+        return repr
+
+    def dates(
+        self, start_date, end_date, return_name: bool = False
+    ) -> Series | DatetimeIndex:
+        """
+        Calculate holidays observed between start date and end date
+
+        Parameters
+        ----------
+        start_date : starting date, datetime-like, optional
+        end_date : ending date, datetime-like, optional
+        return_name : bool, optional, default=False
+            If True, return a series that has dates and holiday names.
+            False will only return dates.
+
+        Returns
+        -------
+        Series or DatetimeIndex
+            Series if return_name is True
+        """
+        start_date = Timestamp(start_date)
+        end_date = Timestamp(end_date)
+
+        filter_start_date = start_date
+        filter_end_date = end_date
+
+        if self.year is not None:
+            dt = Timestamp(datetime(self.year, self.month, self.day))
+            dti = DatetimeIndex([dt])
+            if return_name:
+                return Series(self.name, index=dti)
+            else:
+                return dti
+
+        dates = self._reference_dates(start_date, end_date)
+        holiday_dates = self._apply_rule(dates)
+        if self.days_of_week is not None:
+            holiday_dates = holiday_dates[
+                np.isin(
+                    # error: "DatetimeIndex" has no attribute "dayofweek"
+                    holiday_dates.dayofweek,  # type: ignore[attr-defined]
+                    self.days_of_week,
+                ).ravel()
+            ]
+
+        if self.start_date is not None:
+            filter_start_date = max(
+                self.start_date.tz_localize(filter_start_date.tz), filter_start_date
+            )
+        if self.end_date is not None:
+            filter_end_date = min(
+                self.end_date.tz_localize(filter_end_date.tz), filter_end_date
+            )
+        holiday_dates = holiday_dates[
+            (holiday_dates >= filter_start_date) & (holiday_dates <= filter_end_date)
+        ]
+        if return_name:
+            return Series(self.name, index=holiday_dates)
+        return holiday_dates
+
+    def _reference_dates(
+        self, start_date: Timestamp, end_date: Timestamp
+    ) -> DatetimeIndex:
+        """
+        Get reference dates for the holiday.
+
+        Return reference dates for the holiday also returning the year
+        prior to the start_date and year following the end_date.  This ensures
+        that any offsets to be applied will yield the holidays within
+        the passed in dates.
+        """
+        if self.start_date is not None:
+            start_date = self.start_date.tz_localize(start_date.tz)
+
+        if self.end_date is not None:
+            end_date = self.end_date.tz_localize(start_date.tz)
+
+        year_offset = DateOffset(years=1)
+        reference_start_date = Timestamp(
+            datetime(start_date.year - 1, self.month, self.day)
+        )
+
+        reference_end_date = Timestamp(
+            datetime(end_date.year + 1, self.month, self.day)
+        )
+        # Don't process unnecessary holidays
+        dates = date_range(
+            start=reference_start_date,
+            end=reference_end_date,
+            freq=year_offset,
+            tz=start_date.tz,
+        )
+
+        return dates
+
+    def _apply_rule(self, dates: DatetimeIndex) -> DatetimeIndex:
+        """
+        Apply the given offset/observance to a DatetimeIndex of dates.
+
+        Parameters
+        ----------
+        dates : DatetimeIndex
+            Dates to apply the given offset/observance rule
+
+        Returns
+        -------
+        Dates with rules applied
+        """
+        if dates.empty:
+            return dates.copy()
+
+        if self.observance is not None:
+            return dates.map(lambda d: self.observance(d))
+
+        if self.offset is not None:
+            if not isinstance(self.offset, list):
+                offsets = [self.offset]
+            else:
+                offsets = self.offset
+            for offset in offsets:
+                # if we are adding a non-vectorized value
+                # ignore the PerformanceWarnings:
+                with warnings.catch_warnings():
+                    warnings.simplefilter("ignore", PerformanceWarning)
+                    dates += offset
+        return dates
+
+
+holiday_calendars = {}
+
+
+def register(cls) -> None:
+    try:
+        name = cls.name
+    except AttributeError:
+        name = cls.__name__
+    holiday_calendars[name] = cls
+
+
+def get_calendar(name: str) -> AbstractHolidayCalendar:
+    """
+    Return an instance of a calendar based on its name.
+
+    Parameters
+    ----------
+    name : str
+        Calendar name to return an instance of
+    """
+    return holiday_calendars[name]()
+
+
+class HolidayCalendarMetaClass(type):
+    def __new__(cls, clsname: str, bases, attrs):
+        calendar_class = super().__new__(cls, clsname, bases, attrs)
+        register(calendar_class)
+        return calendar_class
+
+
+class AbstractHolidayCalendar(metaclass=HolidayCalendarMetaClass):
+    """
+    Abstract interface to create holidays following certain rules.
+    """
+
+    rules: list[Holiday] = []
+    start_date = Timestamp(datetime(1970, 1, 1))
+    end_date = Timestamp(datetime(2200, 12, 31))
+    _cache = None
+
+    def __init__(self, name: str = "", rules=None) -> None:
+        """
+        Initializes holiday object with a given set a rules.  Normally
+        classes just have the rules defined within them.
+
+        Parameters
+        ----------
+        name : str
+            Name of the holiday calendar, defaults to class name
+        rules : array of Holiday objects
+            A set of rules used to create the holidays.
+        """
+        super().__init__()
+        if not name:
+            name = type(self).__name__
+        self.name = name
+
+        if rules is not None:
+            self.rules = rules
+
+    def rule_from_name(self, name: str) -> Holiday | None:
+        for rule in self.rules:
+            if rule.name == name:
+                return rule
+
+        return None
+
+    def holidays(self, start=None, end=None, return_name: bool = False):
+        """
+        Returns a curve with holidays between start_date and end_date
+
+        Parameters
+        ----------
+        start : starting date, datetime-like, optional
+        end : ending date, datetime-like, optional
+        return_name : bool, optional
+            If True, return a series that has dates and holiday names.
+            False will only return a DatetimeIndex of dates.
+
+        Returns
+        -------
+            DatetimeIndex of holidays
+        """
+        if self.rules is None:
+            raise Exception(
+                f"Holiday Calendar {self.name} does not have any rules specified"
+            )
+
+        if start is None:
+            start = AbstractHolidayCalendar.start_date
+
+        if end is None:
+            end = AbstractHolidayCalendar.end_date
+
+        start = Timestamp(start)
+        end = Timestamp(end)
+
+        # If we don't have a cache or the dates are outside the prior cache, we
+        # get them again
+        if self._cache is None or start < self._cache[0] or end > self._cache[1]:
+            pre_holidays = [
+                rule.dates(start, end, return_name=True) for rule in self.rules
+            ]
+            if pre_holidays:
+                # error: Argument 1 to "concat" has incompatible type
+                # "List[Union[Series, DatetimeIndex]]"; expected
+                # "Union[Iterable[DataFrame], Mapping[<nothing>, DataFrame]]"
+                holidays = concat(pre_holidays)  # type: ignore[arg-type]
+            else:
+                # error: Incompatible types in assignment (expression has type
+                # "Series", variable has type "DataFrame")
+                holidays = Series(index=DatetimeIndex([]), dtype=object)  # type: ignore[assignment]
+
+            self._cache = (start, end, holidays.sort_index())
+
+        holidays = self._cache[2]
+        holidays = holidays[start:end]
+
+        if return_name:
+            return holidays
+        else:
+            return holidays.index
+
+    @staticmethod
+    def merge_class(base, other):
+        """
+        Merge holiday calendars together. The base calendar
+        will take precedence to other. The merge will be done
+        based on each holiday's name.
+
+        Parameters
+        ----------
+        base : AbstractHolidayCalendar
+          instance/subclass or array of Holiday objects
+        other : AbstractHolidayCalendar
+          instance/subclass or array of Holiday objects
+        """
+        try:
+            other = other.rules
+        except AttributeError:
+            pass
+
+        if not isinstance(other, list):
+            other = [other]
+        other_holidays = {holiday.name: holiday for holiday in other}
+
+        try:
+            base = base.rules
+        except AttributeError:
+            pass
+
+        if not isinstance(base, list):
+            base = [base]
+        base_holidays = {holiday.name: holiday for holiday in base}
+
+        other_holidays.update(base_holidays)
+        return list(other_holidays.values())
+
+    def merge(self, other, inplace: bool = False):
+        """
+        Merge holiday calendars together.  The caller's class
+        rules take precedence.  The merge will be done
+        based on each holiday's name.
+
+        Parameters
+        ----------
+        other : holiday calendar
+        inplace : bool (default=False)
+            If True set rule_table to holidays, else return array of Holidays
+        """
+        holidays = self.merge_class(self, other)
+        if inplace:
+            self.rules = holidays
+        else:
+            return holidays
+
+
+USMemorialDay = Holiday(
+    "Memorial Day", month=5, day=31, offset=DateOffset(weekday=MO(-1))
+)
+USLaborDay = Holiday("Labor Day", month=9, day=1, offset=DateOffset(weekday=MO(1)))
+USColumbusDay = Holiday(
+    "Columbus Day", month=10, day=1, offset=DateOffset(weekday=MO(2))
+)
+USThanksgivingDay = Holiday(
+    "Thanksgiving Day", month=11, day=1, offset=DateOffset(weekday=TH(4))
+)
+USMartinLutherKingJr = Holiday(
+    "Birthday of Martin Luther King, Jr.",
+    start_date=datetime(1986, 1, 1),
+    month=1,
+    day=1,
+    offset=DateOffset(weekday=MO(3)),
+)
+USPresidentsDay = Holiday(
+    "Washington's Birthday", month=2, day=1, offset=DateOffset(weekday=MO(3))
+)
+GoodFriday = Holiday("Good Friday", month=1, day=1, offset=[Easter(), Day(-2)])
+
+EasterMonday = Holiday("Easter Monday", month=1, day=1, offset=[Easter(), Day(1)])
+
+
+class USFederalHolidayCalendar(AbstractHolidayCalendar):
+    """
+    US Federal Government Holiday Calendar based on rules specified by:
+    https://www.opm.gov/policy-data-oversight/pay-leave/federal-holidays/
+    """
+
+    rules = [
+        Holiday("New Year's Day", month=1, day=1, observance=nearest_workday),
+        USMartinLutherKingJr,
+        USPresidentsDay,
+        USMemorialDay,
+        Holiday(
+            "Juneteenth National Independence Day",
+            month=6,
+            day=19,
+            start_date="2021-06-18",
+            observance=nearest_workday,
+        ),
+        Holiday("Independence Day", month=7, day=4, observance=nearest_workday),
+        USLaborDay,
+        USColumbusDay,
+        Holiday("Veterans Day", month=11, day=11, observance=nearest_workday),
+        USThanksgivingDay,
+        Holiday("Christmas Day", month=12, day=25, observance=nearest_workday),
+    ]
+
+
+def HolidayCalendarFactory(name: str, base, other, base_class=AbstractHolidayCalendar):
+    rules = AbstractHolidayCalendar.merge_class(base, other)
+    calendar_class = type(name, (base_class,), {"rules": rules, "name": name})
+    return calendar_class
+
+
+__all__ = [
+    "FR",
+    "MO",
+    "SA",
+    "SU",
+    "TH",
+    "TU",
+    "WE",
+    "HolidayCalendarFactory",
+    "after_nearest_workday",
+    "before_nearest_workday",
+    "get_calendar",
+    "nearest_workday",
+    "next_monday",
+    "next_monday_or_tuesday",
+    "next_workday",
+    "previous_friday",
+    "previous_workday",
+    "register",
+    "sunday_to_monday",
+    "weekend_to_monday",
+]
diff --git a/pandas/pandas/tseries/offsets.py b/pandas/pandas/tseries/offsets.py
new file mode 100644
index 0000000000000..a065137e6971c
--- /dev/null
+++ b/pandas/pandas/tseries/offsets.py
@@ -0,0 +1,91 @@
+from __future__ import annotations
+
+from pandas._libs.tslibs.offsets import (
+    FY5253,
+    BaseOffset,
+    BDay,
+    BMonthBegin,
+    BMonthEnd,
+    BQuarterBegin,
+    BQuarterEnd,
+    BusinessDay,
+    BusinessHour,
+    BusinessMonthBegin,
+    BusinessMonthEnd,
+    BYearBegin,
+    BYearEnd,
+    CBMonthBegin,
+    CBMonthEnd,
+    CDay,
+    CustomBusinessDay,
+    CustomBusinessHour,
+    CustomBusinessMonthBegin,
+    CustomBusinessMonthEnd,
+    DateOffset,
+    Day,
+    Easter,
+    FY5253Quarter,
+    Hour,
+    LastWeekOfMonth,
+    Micro,
+    Milli,
+    Minute,
+    MonthBegin,
+    MonthEnd,
+    Nano,
+    QuarterBegin,
+    QuarterEnd,
+    Second,
+    SemiMonthBegin,
+    SemiMonthEnd,
+    Tick,
+    Week,
+    WeekOfMonth,
+    YearBegin,
+    YearEnd,
+)
+
+__all__ = [
+    "FY5253",
+    "BDay",
+    "BMonthBegin",
+    "BMonthEnd",
+    "BQuarterBegin",
+    "BQuarterEnd",
+    "BYearBegin",
+    "BYearEnd",
+    "BaseOffset",
+    "BusinessDay",
+    "BusinessHour",
+    "BusinessMonthBegin",
+    "BusinessMonthEnd",
+    "CBMonthBegin",
+    "CBMonthEnd",
+    "CDay",
+    "CustomBusinessDay",
+    "CustomBusinessHour",
+    "CustomBusinessMonthBegin",
+    "CustomBusinessMonthEnd",
+    "DateOffset",
+    "Day",
+    "Easter",
+    "FY5253Quarter",
+    "Hour",
+    "LastWeekOfMonth",
+    "Micro",
+    "Milli",
+    "Minute",
+    "MonthBegin",
+    "MonthEnd",
+    "Nano",
+    "QuarterBegin",
+    "QuarterEnd",
+    "Second",
+    "SemiMonthBegin",
+    "SemiMonthEnd",
+    "Tick",
+    "Week",
+    "WeekOfMonth",
+    "YearBegin",
+    "YearEnd",
+]
diff --git a/pandas/pandas/util/__init__.py b/pandas/pandas/util/__init__.py
new file mode 100644
index 0000000000000..dfab207635fec
--- /dev/null
+++ b/pandas/pandas/util/__init__.py
@@ -0,0 +1,29 @@
+def __getattr__(key: str):
+    # These imports need to be lazy to avoid circular import errors
+    if key == "hash_array":
+        from pandas.core.util.hashing import hash_array
+
+        return hash_array
+    if key == "hash_pandas_object":
+        from pandas.core.util.hashing import hash_pandas_object
+
+        return hash_pandas_object
+    if key == "Appender":
+        from pandas.util._decorators import Appender
+
+        return Appender
+    if key == "Substitution":
+        from pandas.util._decorators import Substitution
+
+        return Substitution
+
+    if key == "cache_readonly":
+        from pandas.util._decorators import cache_readonly
+
+        return cache_readonly
+
+    raise AttributeError(f"module 'pandas.util' has no attribute '{key}'")
+
+
+def __dir__() -> list[str]:
+    return list(globals().keys()) + ["hash_array", "hash_pandas_object"]
diff --git a/pandas/pandas/util/_decorators.py b/pandas/pandas/util/_decorators.py
new file mode 100644
index 0000000000000..a1a0d51a7c72b
--- /dev/null
+++ b/pandas/pandas/util/_decorators.py
@@ -0,0 +1,528 @@
+from __future__ import annotations
+
+from functools import wraps
+import inspect
+from textwrap import dedent
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    cast,
+)
+import warnings
+
+from pandas._libs.properties import cache_readonly
+from pandas._typing import (
+    F,
+    T,
+)
+from pandas.util._exceptions import find_stack_level
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Callable,
+        Mapping,
+    )
+
+
+def deprecate(
+    name: str,
+    alternative: Callable[..., Any],
+    version: str,
+    alt_name: str | None = None,
+    klass: type[Warning] | None = None,
+    stacklevel: int = 2,
+    msg: str | None = None,
+) -> Callable[[F], F]:
+    """
+    Return a new function that emits a deprecation warning on use.
+
+    To use this method for a deprecated function, another function
+    `alternative` with the same signature must exist. The deprecated
+    function will emit a deprecation warning, and in the docstring
+    it will contain the deprecation directive with the provided version
+    so it can be detected for future removal.
+
+    Parameters
+    ----------
+    name : str
+        Name of function to deprecate.
+    alternative : func
+        Function to use instead.
+    version : str
+        Version of pandas in which the method has been deprecated.
+    alt_name : str, optional
+        Name to use in preference of alternative.__name__.
+    klass : Warning, default FutureWarning
+    stacklevel : int, default 2
+    msg : str
+        The message to display in the warning.
+        Default is '{name} is deprecated. Use {alt_name} instead.'
+    """
+    alt_name = alt_name or alternative.__name__
+    klass = klass or FutureWarning
+    warning_msg = msg or f"{name} is deprecated, use {alt_name} instead."
+
+    @wraps(alternative)
+    def wrapper(*args, **kwargs) -> Callable[..., Any]:
+        warnings.warn(warning_msg, klass, stacklevel=stacklevel)
+        return alternative(*args, **kwargs)
+
+    # adding deprecated directive to the docstring
+    msg = msg or f"Use `{alt_name}` instead."
+    doc_error_msg = (
+        "deprecate needs a correctly formatted docstring in "
+        "the target function (should have a one liner short "
+        "summary, and opening quotes should be in their own "
+        f"line). Found:\n{alternative.__doc__}"
+    )
+
+    # when python is running in optimized mode (i.e. `-OO`), docstrings are
+    # removed, so we check that a docstring with correct formatting is used
+    # but we allow empty docstrings
+    if alternative.__doc__:
+        if alternative.__doc__.count("\n") < 3:
+            raise AssertionError(doc_error_msg)
+        empty1, summary, empty2, doc_string = alternative.__doc__.split("\n", 3)
+        if empty1 or (empty2 and not summary):
+            raise AssertionError(doc_error_msg)
+        wrapper.__doc__ = dedent(
+            f"""
+        {summary.strip()}
+
+        .. deprecated:: {version}
+            {msg}
+
+        {dedent(doc_string)}"""
+        )
+    # error: Incompatible return value type (got "Callable[[VarArg(Any), KwArg(Any)],
+    # Callable[...,Any]]", expected "Callable[[F], F]")
+    return wrapper  # type: ignore[return-value]
+
+
+def deprecate_kwarg(
+    old_arg_name: str,
+    new_arg_name: str | None,
+    mapping: Mapping[Any, Any] | Callable[[Any], Any] | None = None,
+    stacklevel: int = 2,
+) -> Callable[[F], F]:
+    """
+    Decorator to deprecate a keyword argument of a function.
+
+    Parameters
+    ----------
+    old_arg_name : str
+        Name of argument in function to deprecate
+    new_arg_name : str or None
+        Name of preferred argument in function. Use None to raise warning that
+        ``old_arg_name`` keyword is deprecated.
+    mapping : dict or callable
+        If mapping is present, use it to translate old arguments to
+        new arguments. A callable must do its own value checking;
+        values not found in a dict will be forwarded unchanged.
+
+    Examples
+    --------
+    The following deprecates 'cols', using 'columns' instead
+
+    >>> @deprecate_kwarg(old_arg_name="cols", new_arg_name="columns")
+    ... def f(columns=""):
+    ...     print(columns)
+    >>> f(columns="should work ok")
+    should work ok
+
+    >>> f(cols="should raise warning")  # doctest: +SKIP
+    FutureWarning: cols is deprecated, use columns instead
+      warnings.warn(msg, FutureWarning)
+    should raise warning
+
+    >>> f(cols="should error", columns="can't pass do both")  # doctest: +SKIP
+    TypeError: Can only specify 'cols' or 'columns', not both
+
+    >>> @deprecate_kwarg("old", "new", {"yes": True, "no": False})
+    ... def f(new=False):
+    ...     print("yes!" if new else "no!")
+    >>> f(old="yes")  # doctest: +SKIP
+    FutureWarning: old='yes' is deprecated, use new=True instead
+      warnings.warn(msg, FutureWarning)
+    yes!
+
+    To raise a warning that a keyword will be removed entirely in the future
+
+    >>> @deprecate_kwarg(old_arg_name="cols", new_arg_name=None)
+    ... def f(cols="", another_param=""):
+    ...     print(cols)
+    >>> f(cols="should raise warning")  # doctest: +SKIP
+    FutureWarning: the 'cols' keyword is deprecated and will be removed in a
+    future version please takes steps to stop use of 'cols'
+    should raise warning
+    >>> f(another_param="should not raise warning")  # doctest: +SKIP
+    should not raise warning
+
+    >>> f(cols="should raise warning", another_param="")  # doctest: +SKIP
+    FutureWarning: the 'cols' keyword is deprecated and will be removed in a
+    future version please takes steps to stop use of 'cols'
+    should raise warning
+    """
+    if mapping is not None and not hasattr(mapping, "get") and not callable(mapping):
+        raise TypeError(
+            "mapping from old to new argument values must be dict or callable!"
+        )
+
+    def _deprecate_kwarg(func: F) -> F:
+        @wraps(func)
+        def wrapper(*args, **kwargs) -> Callable[..., Any]:
+            old_arg_value = kwargs.pop(old_arg_name, None)
+
+            if old_arg_value is not None:
+                if new_arg_name is None:
+                    msg = (
+                        f"the {old_arg_name!r} keyword is deprecated and "
+                        "will be removed in a future version. Please take "
+                        f"steps to stop the use of {old_arg_name!r}"
+                    )
+                    warnings.warn(msg, FutureWarning, stacklevel=stacklevel)
+                    kwargs[old_arg_name] = old_arg_value
+                    return func(*args, **kwargs)
+
+                elif mapping is not None:
+                    if callable(mapping):
+                        new_arg_value = mapping(old_arg_value)
+                    else:
+                        new_arg_value = mapping.get(old_arg_value, old_arg_value)
+                    msg = (
+                        f"the {old_arg_name}={old_arg_value!r} keyword is "
+                        "deprecated, use "
+                        f"{new_arg_name}={new_arg_value!r} instead."
+                    )
+                else:
+                    new_arg_value = old_arg_value
+                    msg = (
+                        f"the {old_arg_name!r} keyword is deprecated, "
+                        f"use {new_arg_name!r} instead."
+                    )
+
+                warnings.warn(msg, FutureWarning, stacklevel=stacklevel)
+                if kwargs.get(new_arg_name) is not None:
+                    msg = (
+                        f"Can only specify {old_arg_name!r} "
+                        f"or {new_arg_name!r}, not both."
+                    )
+                    raise TypeError(msg)
+                kwargs[new_arg_name] = new_arg_value
+            return func(*args, **kwargs)
+
+        return cast(F, wrapper)
+
+    return _deprecate_kwarg
+
+
+def _format_argument_list(allow_args: list[str]) -> str:
+    """
+    Convert the allow_args argument (either string or integer) of
+    `deprecate_nonkeyword_arguments` function to a string describing
+    it to be inserted into warning message.
+
+    Parameters
+    ----------
+    allowed_args : list, tuple or int
+        The `allowed_args` argument for `deprecate_nonkeyword_arguments`,
+        but None value is not allowed.
+
+    Returns
+    -------
+    str
+        The substring describing the argument list in best way to be
+        inserted to the warning message.
+
+    Examples
+    --------
+    `format_argument_list([])` -> ''
+    `format_argument_list(['a'])` -> "except for the arguments 'a'"
+    `format_argument_list(['a', 'b'])` -> "except for the arguments 'a' and 'b'"
+    `format_argument_list(['a', 'b', 'c'])` ->
+        "except for the arguments 'a', 'b' and 'c'"
+    """
+    if "self" in allow_args:
+        allow_args.remove("self")
+    if not allow_args:
+        return ""
+    elif len(allow_args) == 1:
+        return f" except for the argument '{allow_args[0]}'"
+    else:
+        last = allow_args[-1]
+        args = ", ".join(["'" + x + "'" for x in allow_args[:-1]])
+        return f" except for the arguments {args} and '{last}'"
+
+
+def future_version_msg(version: str | None) -> str:
+    """Specify which version of pandas the deprecation will take place in."""
+    if version is None:
+        return "In a future version of pandas"
+    else:
+        return f"Starting with pandas version {version}"
+
+
+def deprecate_nonkeyword_arguments(
+    version: str | None,
+    allowed_args: list[str] | None = None,
+    name: str | None = None,
+) -> Callable[[F], F]:
+    """
+    Decorator to deprecate a use of non-keyword arguments of a function.
+
+    Parameters
+    ----------
+    version : str, optional
+        The version in which positional arguments will become
+        keyword-only. If None, then the warning message won't
+        specify any particular version.
+
+    allowed_args : list, optional
+        In case of list, it must be the list of names of some
+        first arguments of the decorated functions that are
+        OK to be given as positional arguments. In case of None value,
+        defaults to list of all arguments not having the
+        default value.
+
+    name : str, optional
+        The specific name of the function to show in the warning
+        message. If None, then the Qualified name of the function
+        is used.
+    """
+
+    def decorate(func):
+        old_sig = inspect.signature(func)
+
+        if allowed_args is not None:
+            allow_args = allowed_args
+        else:
+            allow_args = [
+                p.name
+                for p in old_sig.parameters.values()
+                if p.kind in (p.POSITIONAL_ONLY, p.POSITIONAL_OR_KEYWORD)
+                and p.default is p.empty
+            ]
+
+        new_params = [
+            p.replace(kind=p.KEYWORD_ONLY)
+            if (
+                p.kind in (p.POSITIONAL_ONLY, p.POSITIONAL_OR_KEYWORD)
+                and p.name not in allow_args
+            )
+            else p
+            for p in old_sig.parameters.values()
+        ]
+        new_params.sort(key=lambda p: p.kind)
+        new_sig = old_sig.replace(parameters=new_params)
+
+        num_allow_args = len(allow_args)
+        msg = (
+            f"{future_version_msg(version)} all arguments of "
+            f"{name or func.__qualname__}{{arguments}} will be keyword-only."
+        )
+
+        @wraps(func)
+        def wrapper(*args, **kwargs):
+            if len(args) > num_allow_args:
+                warnings.warn(
+                    msg.format(arguments=_format_argument_list(allow_args)),
+                    FutureWarning,
+                    stacklevel=find_stack_level(),
+                )
+            return func(*args, **kwargs)
+
+        # error: "Callable[[VarArg(Any), KwArg(Any)], Any]" has no
+        # attribute "__signature__"
+        wrapper.__signature__ = new_sig  # type: ignore[attr-defined]
+        return wrapper
+
+    return decorate
+
+
+def doc(*docstrings: None | str | Callable, **params: object) -> Callable[[F], F]:
+    """
+    A decorator to take docstring templates, concatenate them and perform string
+    substitution on them.
+
+    This decorator will add a variable "_docstring_components" to the wrapped
+    callable to keep track the original docstring template for potential usage.
+    If it should be consider as a template, it will be saved as a string.
+    Otherwise, it will be saved as callable, and later user __doc__ and dedent
+    to get docstring.
+
+    Parameters
+    ----------
+    *docstrings : None, str, or callable
+        The string / docstring / docstring template to be appended in order
+        after default docstring under callable.
+    **params
+        The string which would be used to format docstring template.
+    """
+
+    def decorator(decorated: F) -> F:
+        # collecting docstring and docstring templates
+        docstring_components: list[str | Callable] = []
+        if decorated.__doc__:
+            docstring_components.append(dedent(decorated.__doc__))
+
+        for docstring in docstrings:
+            if docstring is None:
+                continue
+            if hasattr(docstring, "_docstring_components"):
+                docstring_components.extend(
+                    docstring._docstring_components  # pyright: ignore[reportAttributeAccessIssue]
+                )
+            elif isinstance(docstring, str) or docstring.__doc__:
+                docstring_components.append(docstring)
+
+        params_applied = [
+            component.format(**params)
+            if isinstance(component, str) and len(params) > 0
+            else component
+            for component in docstring_components
+        ]
+
+        decorated.__doc__ = "".join(
+            [
+                component
+                if isinstance(component, str)
+                else dedent(component.__doc__ or "")
+                for component in params_applied
+            ]
+        )
+
+        # error: "F" has no attribute "_docstring_components"
+        decorated._docstring_components = (  # type: ignore[attr-defined]
+            docstring_components
+        )
+        return decorated
+
+    return decorator
+
+
+# Substitution and Appender are derived from matplotlib.docstring (1.1.0)
+# module https://matplotlib.org/users/license.html
+
+
+class Substitution:
+    """
+    A decorator to take a function's docstring and perform string
+    substitution on it.
+
+    This decorator should be robust even if func.__doc__ is None
+    (for example, if -OO was passed to the interpreter)
+
+    Usage: construct a docstring.Substitution with a sequence or
+    dictionary suitable for performing substitution; then
+    decorate a suitable function with the constructed object. e.g.
+
+    sub_author_name = Substitution(author='Jason')
+
+    @sub_author_name
+    def some_function(x):
+        "%(author)s wrote this function"
+
+    # note that some_function.__doc__ is now "Jason wrote this function"
+
+    One can also use positional arguments.
+
+    sub_first_last_names = Substitution('Edgar Allen', 'Poe')
+
+    @sub_first_last_names
+    def some_function(x):
+        "%s %s wrote the Raven"
+    """
+
+    def __init__(self, *args, **kwargs) -> None:
+        if args and kwargs:
+            raise AssertionError("Only positional or keyword args are allowed")
+
+        self.params = args or kwargs
+
+    def __call__(self, func: F) -> F:
+        func.__doc__ = func.__doc__ and func.__doc__ % self.params
+        return func
+
+    def update(self, *args, **kwargs) -> None:
+        """
+        Update self.params with supplied args.
+        """
+        if isinstance(self.params, dict):
+            self.params.update(*args, **kwargs)
+
+
+class Appender:
+    """
+    A function decorator that will append an addendum to the docstring
+    of the target function.
+
+    This decorator should be robust even if func.__doc__ is None
+    (for example, if -OO was passed to the interpreter).
+
+    Usage: construct a docstring.Appender with a string to be joined to
+    the original docstring. An optional 'join' parameter may be supplied
+    which will be used to join the docstring and addendum. e.g.
+
+    add_copyright = Appender("Copyright (c) 2009", join='\n')
+
+    @add_copyright
+    def my_dog(has='fleas'):
+        "This docstring will have a copyright below"
+        pass
+    """
+
+    addendum: str | None
+
+    def __init__(self, addendum: str | None, join: str = "", indents: int = 0) -> None:
+        if indents > 0:
+            self.addendum = indent(addendum, indents=indents)
+        else:
+            self.addendum = addendum
+        self.join = join
+
+    def __call__(self, func: T) -> T:
+        func.__doc__ = func.__doc__ if func.__doc__ else ""
+        self.addendum = self.addendum if self.addendum else ""
+        docitems = [func.__doc__, self.addendum]
+        func.__doc__ = dedent(self.join.join(docitems))
+        return func
+
+
+def indent(text: str | None, indents: int = 1) -> str:
+    if not text or not isinstance(text, str):
+        return ""
+    jointext = "".join(["\n"] + ["    "] * indents)
+    return jointext.join(text.split("\n"))
+
+
+__all__ = [
+    "Appender",
+    "Substitution",
+    "cache_readonly",
+    "deprecate",
+    "deprecate_kwarg",
+    "deprecate_nonkeyword_arguments",
+    "doc",
+    "future_version_msg",
+]
+
+
+def set_module(module) -> Callable[[F], F]:
+    """Private decorator for overriding __module__ on a function or class.
+
+    Example usage::
+
+        @set_module("pandas")
+        def example():
+            pass
+
+
+        assert example.__module__ == "pandas"
+    """
+
+    def decorator(func: F) -> F:
+        if module is not None:
+            func.__module__ = module
+        return func
+
+    return decorator
diff --git a/pandas/pandas/util/_doctools.py b/pandas/pandas/util/_doctools.py
new file mode 100644
index 0000000000000..828b7a4591bf2
--- /dev/null
+++ b/pandas/pandas/util/_doctools.py
@@ -0,0 +1,206 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import numpy as np
+
+import pandas as pd
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+    from matplotlib.figure import Figure
+
+
+class TablePlotter:
+    """
+    Layout some DataFrames in vertical/horizontal layout for explanation.
+    Used in merging.rst
+    """
+
+    def __init__(
+        self,
+        cell_width: float = 0.37,
+        cell_height: float = 0.25,
+        font_size: float = 7.5,
+    ) -> None:
+        self.cell_width = cell_width
+        self.cell_height = cell_height
+        self.font_size = font_size
+
+    def _shape(self, df: pd.DataFrame) -> tuple[int, int]:
+        """
+        Calculate table shape considering index levels.
+        """
+        row, col = df.shape
+        return row + df.columns.nlevels, col + df.index.nlevels
+
+    def _get_cells(self, left, right, vertical) -> tuple[int, int]:
+        """
+        Calculate appropriate figure size based on left and right data.
+        """
+        if vertical:
+            # calculate required number of cells
+            vcells = max(sum(self._shape(df)[0] for df in left), self._shape(right)[0])
+            hcells = max(self._shape(df)[1] for df in left) + self._shape(right)[1]
+        else:
+            vcells = max([self._shape(df)[0] for df in left] + [self._shape(right)[0]])
+            hcells = sum([self._shape(df)[1] for df in left] + [self._shape(right)[1]])
+        return hcells, vcells
+
+    def plot(
+        self, left, right, labels: Iterable[str] = (), vertical: bool = True
+    ) -> Figure:
+        """
+        Plot left / right DataFrames in specified layout.
+
+        Parameters
+        ----------
+        left : list of DataFrames before operation is applied
+        right : DataFrame of operation result
+        labels : list of str to be drawn as titles of left DataFrames
+        vertical : bool, default True
+            If True, use vertical layout. If False, use horizontal layout.
+        """
+        from matplotlib import gridspec
+        import matplotlib.pyplot as plt
+
+        if not isinstance(left, list):
+            left = [left]
+        left = [self._conv(df) for df in left]
+        right = self._conv(right)
+
+        hcells, vcells = self._get_cells(left, right, vertical)
+
+        if vertical:
+            figsize = self.cell_width * hcells, self.cell_height * vcells
+        else:
+            # include margin for titles
+            figsize = self.cell_width * hcells, self.cell_height * vcells
+        fig = plt.figure(figsize=figsize)
+
+        if vertical:
+            gs = gridspec.GridSpec(len(left), hcells)
+            # left
+            max_left_cols = max(self._shape(df)[1] for df in left)
+            max_left_rows = max(self._shape(df)[0] for df in left)
+            for i, (_left, _label) in enumerate(zip(left, labels)):
+                ax = fig.add_subplot(gs[i, 0:max_left_cols])
+                self._make_table(ax, _left, title=_label, height=1.0 / max_left_rows)
+            # right
+            ax = plt.subplot(gs[:, max_left_cols:])
+            self._make_table(ax, right, title="Result", height=1.05 / vcells)
+            fig.subplots_adjust(top=0.9, bottom=0.05, left=0.05, right=0.95)
+        else:
+            max_rows = max(self._shape(df)[0] for df in left + [right])
+            height = 1.0 / np.max(max_rows)
+            gs = gridspec.GridSpec(1, hcells)
+            # left
+            i = 0
+            for df, _label in zip(left, labels):
+                sp = self._shape(df)
+                ax = fig.add_subplot(gs[0, i : i + sp[1]])
+                self._make_table(ax, df, title=_label, height=height)
+                i += sp[1]
+            # right
+            ax = plt.subplot(gs[0, i:])
+            self._make_table(ax, right, title="Result", height=height)
+            fig.subplots_adjust(top=0.85, bottom=0.05, left=0.05, right=0.95)
+
+        return fig
+
+    def _conv(self, data):
+        """
+        Convert each input to appropriate for table outplot.
+        """
+        if isinstance(data, pd.Series):
+            if data.name is None:
+                data = data.to_frame(name="")
+            else:
+                data = data.to_frame()
+        data = data.fillna("NaN")
+        return data
+
+    def _insert_index(self, data):
+        # insert is destructive
+        data = data.copy()
+        idx_nlevels = data.index.nlevels
+        if idx_nlevels == 1:
+            data.insert(0, "Index", data.index)
+        else:
+            for i in range(idx_nlevels):
+                data.insert(i, f"Index{i}", data.index._get_level_values(i))
+
+        col_nlevels = data.columns.nlevels
+        if col_nlevels > 1:
+            col = data.columns._get_level_values(0)
+            values = [
+                data.columns._get_level_values(i)._values for i in range(1, col_nlevels)
+            ]
+            col_df = pd.DataFrame(values)
+            data.columns = col_df.columns
+            data = pd.concat([col_df, data])
+            data.columns = col
+        return data
+
+    def _make_table(self, ax, df, title: str, height: float | None = None) -> None:
+        if df is None:
+            ax.set_visible(False)
+            return
+
+        from pandas import plotting
+
+        idx_nlevels = df.index.nlevels
+        col_nlevels = df.columns.nlevels
+        # must be convert here to get index levels for colorization
+        df = self._insert_index(df)
+        tb = plotting.table(ax, df, loc=9)
+        tb.set_fontsize(self.font_size)
+
+        if height is None:
+            height = 1.0 / (len(df) + 1)
+
+        props = tb.properties()
+        for (r, c), cell in props["celld"].items():
+            if c == -1:
+                cell.set_visible(False)
+            elif r < col_nlevels and c < idx_nlevels:
+                cell.set_visible(False)
+            elif r < col_nlevels or c < idx_nlevels:
+                cell.set_facecolor("#AAAAAA")
+            cell.set_height(height)
+
+        ax.set_title(title, size=self.font_size)
+        ax.axis("off")
+
+
+def main() -> None:
+    import matplotlib.pyplot as plt
+
+    p = TablePlotter()
+
+    df1 = pd.DataFrame({"A": [10, 11, 12], "B": [20, 21, 22], "C": [30, 31, 32]})
+    df2 = pd.DataFrame({"A": [10, 12], "C": [30, 32]})
+
+    p.plot([df1, df2], pd.concat([df1, df2]), labels=["df1", "df2"], vertical=True)
+    plt.show()
+
+    df3 = pd.DataFrame({"X": [10, 12], "Z": [30, 32]})
+
+    p.plot(
+        [df1, df3], pd.concat([df1, df3], axis=1), labels=["df1", "df2"], vertical=False
+    )
+    plt.show()
+
+    idx = pd.MultiIndex.from_tuples(
+        [(1, "A"), (1, "B"), (1, "C"), (2, "A"), (2, "B"), (2, "C")]
+    )
+    column = pd.MultiIndex.from_tuples([(1, "A"), (1, "B")])
+    df3 = pd.DataFrame({"v1": [1, 2, 3, 4, 5, 6], "v2": [5, 6, 7, 8, 9, 10]}, index=idx)
+    df3.columns = column
+    p.plot(df3, df3, labels=["df3"])
+    plt.show()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pandas/pandas/util/_exceptions.py b/pandas/pandas/util/_exceptions.py
new file mode 100644
index 0000000000000..b3c8e54d3ca7f
--- /dev/null
+++ b/pandas/pandas/util/_exceptions.py
@@ -0,0 +1,106 @@
+from __future__ import annotations
+
+import contextlib
+import inspect
+import os
+import re
+from typing import (
+    TYPE_CHECKING,
+    Any,
+)
+import warnings
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+    from types import FrameType
+
+
+@contextlib.contextmanager
+def rewrite_exception(old_name: str, new_name: str) -> Generator[None]:
+    """
+    Rewrite the message of an exception.
+    """
+    try:
+        yield
+    except Exception as err:
+        if not err.args:
+            raise
+        msg = str(err.args[0])
+        msg = msg.replace(old_name, new_name)
+        args: tuple[Any, ...] = (msg,)
+        if len(err.args) > 1:
+            args = args + err.args[1:]
+        err.args = args
+        raise
+
+
+def find_stack_level() -> int:
+    """
+    Find the first place in the stack that is not inside pandas
+    (tests notwithstanding).
+    """
+
+    import pandas as pd
+
+    pkg_dir = os.path.dirname(pd.__file__)
+    test_dir = os.path.join(pkg_dir, "tests")
+
+    # https://stackoverflow.com/questions/17407119/python-inspect-stack-is-slow
+    frame: FrameType | None = inspect.currentframe()
+    try:
+        n = 0
+        while frame:
+            filename = inspect.getfile(frame)
+            if filename.startswith(pkg_dir) and not filename.startswith(test_dir):
+                frame = frame.f_back
+                n += 1
+            else:
+                break
+    finally:
+        # See note in
+        # https://docs.python.org/3/library/inspect.html#inspect.Traceback
+        del frame
+    return n
+
+
+@contextlib.contextmanager
+def rewrite_warning(
+    target_message: str,
+    target_category: type[Warning],
+    new_message: str,
+    new_category: type[Warning] | None = None,
+) -> Generator[None]:
+    """
+    Rewrite the message of a warning.
+
+    Parameters
+    ----------
+    target_message : str
+        Warning message to match.
+    target_category : Warning
+        Warning type to match.
+    new_message : str
+        New warning message to emit.
+    new_category : Warning or None, default None
+        New warning type to emit. When None, will be the same as target_category.
+    """
+    if new_category is None:
+        new_category = target_category
+    with warnings.catch_warnings(record=True) as record:
+        yield
+    if len(record) > 0:
+        match = re.compile(target_message)
+        for warning in record:
+            if warning.category is target_category and re.search(
+                match, str(warning.message)
+            ):
+                category = new_category
+                message: Warning | str = new_message
+            else:
+                category, message = warning.category, warning.message
+            warnings.warn_explicit(
+                message=message,
+                category=category,
+                filename=warning.filename,
+                lineno=warning.lineno,
+            )
diff --git a/pandas/pandas/util/_print_versions.py b/pandas/pandas/util/_print_versions.py
new file mode 100644
index 0000000000000..bd20660bdbba6
--- /dev/null
+++ b/pandas/pandas/util/_print_versions.py
@@ -0,0 +1,162 @@
+from __future__ import annotations
+
+import codecs
+import json
+import locale
+import os
+import platform
+import struct
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from pandas._typing import JSONSerializable
+
+from pandas.compat._optional import (
+    VERSIONS,
+    get_version,
+    import_optional_dependency,
+)
+
+
+def _get_commit_hash() -> str | None:
+    """
+    Use vendored versioneer code to get git hash, which handles
+    git worktree correctly.
+    """
+    try:
+        from pandas._version_meson import (  # pyright: ignore [reportMissingImports]
+            __git_version__,
+        )
+
+        return __git_version__
+    except ImportError:
+        from pandas._version import get_versions
+
+        versions = get_versions()  # type: ignore[no-untyped-call]
+        return versions["full-revisionid"]
+
+
+def _get_sys_info() -> dict[str, JSONSerializable]:
+    """
+    Returns system information as a JSON serializable dictionary.
+    """
+    uname_result = platform.uname()
+    language_code, encoding = locale.getlocale()
+    return {
+        "commit": _get_commit_hash(),
+        "python": platform.python_version(),
+        "python-bits": struct.calcsize("P") * 8,
+        "OS": uname_result.system,
+        "OS-release": uname_result.release,
+        "Version": uname_result.version,
+        "machine": uname_result.machine,
+        "processor": uname_result.processor,
+        "byteorder": sys.byteorder,
+        "LC_ALL": os.environ.get("LC_ALL"),
+        "LANG": os.environ.get("LANG"),
+        "LOCALE": {"language-code": language_code, "encoding": encoding},
+    }
+
+
+def _get_dependency_info() -> dict[str, JSONSerializable]:
+    """
+    Returns dependency information as a JSON serializable dictionary.
+    """
+    deps = [
+        "pandas",
+        # required
+        "numpy",
+        "dateutil",
+        # install / build,
+        "pip",
+        "Cython",
+        # docs
+        "sphinx",
+        # Other, not imported.
+        "IPython",
+    ]
+    # Optional dependencies
+    deps.extend(list(VERSIONS))
+
+    result: dict[str, JSONSerializable] = {}
+    for modname in deps:
+        try:
+            mod = import_optional_dependency(modname, errors="ignore")
+        except Exception:
+            # Dependency conflicts may cause a non ImportError
+            result[modname] = "N/A"
+        else:
+            result[modname] = get_version(mod) if mod else None
+    return result
+
+
+def show_versions(as_json: str | bool = False) -> None:
+    """
+    Provide useful information, important for bug reports.
+
+    It comprises info about hosting operation system, pandas version,
+    and versions of other installed relative packages.
+
+    Parameters
+    ----------
+    as_json : str or bool, default False
+        * If False, outputs info in a human readable form to the console.
+        * If str, it will be considered as a path to a file.
+          Info will be written to that file in JSON format.
+        * If True, outputs info in JSON format to the console.
+
+    See Also
+    --------
+    get_option : Retrieve the value of the specified option.
+    set_option : Set the value of the specified option or options.
+
+    Examples
+    --------
+    >>> pd.show_versions()  # doctest: +SKIP
+    Your output may look something like this:
+    INSTALLED VERSIONS
+    ------------------
+    commit           : 37ea63d540fd27274cad6585082c91b1283f963d
+    python           : 3.10.6.final.0
+    python-bits      : 64
+    OS               : Linux
+    OS-release       : 5.10.102.1-microsoft-standard-WSL2
+    Version          : #1 SMP Wed Mar 2 00:30:59 UTC 2022
+    machine          : x86_64
+    processor        : x86_64
+    byteorder        : little
+    LC_ALL           : None
+    LANG             : en_GB.UTF-8
+    LOCALE           : en_GB.UTF-8
+    pandas           : 2.0.1
+    numpy            : 1.24.3
+    ...
+    """
+    sys_info = _get_sys_info()
+    deps = _get_dependency_info()
+
+    if as_json:
+        j = {"system": sys_info, "dependencies": deps}
+
+        if as_json is True:
+            sys.stdout.writelines(json.dumps(j, indent=2))
+        else:
+            assert isinstance(as_json, str)  # needed for mypy
+            with codecs.open(as_json, "wb", encoding="utf8") as f:
+                json.dump(j, f, indent=2)
+
+    else:
+        assert isinstance(sys_info["LOCALE"], dict)  # needed for mypy
+        language_code = sys_info["LOCALE"]["language-code"]
+        encoding = sys_info["LOCALE"]["encoding"]
+        sys_info["LOCALE"] = f"{language_code}.{encoding}"
+
+        maxlen = max(len(x) for x in deps)
+        print("\nINSTALLED VERSIONS")
+        print("------------------")
+        for k, v in sys_info.items():
+            print(f"{k:<{maxlen}}: {v}")
+        print("")
+        for k, v in deps.items():
+            print(f"{k:<{maxlen}}: {v}")
diff --git a/pandas/pandas/util/_test_decorators.py b/pandas/pandas/util/_test_decorators.py
new file mode 100644
index 0000000000000..1c17587db72d4
--- /dev/null
+++ b/pandas/pandas/util/_test_decorators.py
@@ -0,0 +1,147 @@
+"""
+This module provides decorator functions which can be applied to test objects
+in order to skip those objects when certain conditions occur. A sample use case
+is to detect if the platform is missing ``matplotlib``. If so, any test objects
+which require ``matplotlib`` and decorated with ``@td.skip_if_no("matplotlib")``
+will be skipped by ``pytest`` during the execution of the test suite.
+
+To illustrate, after importing this module:
+
+import pandas.util._test_decorators as td
+
+The decorators can be applied to classes:
+
+@td.skip_if_no("package")
+class Foo:
+    ...
+
+Or individual functions:
+
+@td.skip_if_no("package")
+def test_foo():
+    ...
+
+For more information, refer to the ``pytest`` documentation on ``skipif``.
+"""
+
+from __future__ import annotations
+
+import locale
+from typing import TYPE_CHECKING
+
+import pytest
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+    from pandas._typing import F
+
+from pandas.compat import (
+    IS64,
+    WASM,
+    is_platform_windows,
+)
+from pandas.compat._optional import import_optional_dependency
+
+
+def skip_if_installed(package: str) -> pytest.MarkDecorator:
+    """
+    Skip a test if a package is installed.
+
+    Parameters
+    ----------
+    package : str
+        The name of the package.
+
+    Returns
+    -------
+    pytest.MarkDecorator
+        a pytest.mark.skipif to use as either a test decorator or a
+        parametrization mark.
+    """
+    return pytest.mark.skipif(
+        bool(import_optional_dependency(package, errors="ignore")),
+        reason=f"Skipping because {package} is installed.",
+    )
+
+
+def skip_if_no(package: str, min_version: str | None = None) -> pytest.MarkDecorator:
+    """
+    Generic function to help skip tests when required packages are not
+    present on the testing system.
+
+    This function returns a pytest mark with a skip condition that will be
+    evaluated during test collection. An attempt will be made to import the
+    specified ``package`` and optionally ensure it meets the ``min_version``
+
+    The mark can be used as either a decorator for a test class or to be
+    applied to parameters in pytest.mark.parametrize calls or parametrized
+    fixtures. Use pytest.importorskip if an imported moduled is later needed
+    or for test functions.
+
+    If the import and version check are unsuccessful, then the test function
+    (or test case when used in conjunction with parametrization) will be
+    skipped.
+
+    Parameters
+    ----------
+    package: str
+        The name of the required package.
+    min_version: str or None, default None
+        Optional minimum version of the package.
+
+    Returns
+    -------
+    pytest.MarkDecorator
+        a pytest.mark.skipif to use as either a test decorator or a
+        parametrization mark.
+    """
+    msg = f"Could not import '{package}'"
+    if min_version:
+        msg += f" satisfying a min_version of {min_version}"
+    return pytest.mark.skipif(
+        not bool(
+            import_optional_dependency(
+                package, errors="ignore", min_version=min_version
+            )
+        ),
+        reason=msg,
+    )
+
+
+skip_if_32bit = pytest.mark.skipif(not IS64, reason="skipping for 32 bit")
+skip_if_windows = pytest.mark.skipif(is_platform_windows(), reason="Running on Windows")
+skip_if_not_us_locale = pytest.mark.skipif(
+    locale.getlocale()[0] != "en_US",
+    reason=f"Set local {locale.getlocale()[0]} is not en_US",
+)
+skip_if_wasm = pytest.mark.skipif(
+    WASM,
+    reason="does not support wasm",
+)
+
+
+def parametrize_fixture_doc(*args) -> Callable[[F], F]:
+    """
+    Intended for use as a decorator for parametrized fixture,
+    this function will wrap the decorated function with a pytest
+    ``parametrize_fixture_doc`` mark. That mark will format
+    initial fixture docstring by replacing placeholders {0}, {1} etc
+    with parameters passed as arguments.
+
+    Parameters
+    ----------
+    args: iterable
+        Positional arguments for docstring.
+
+    Returns
+    -------
+    function
+        The decorated function wrapped within a pytest
+        ``parametrize_fixture_doc`` mark
+    """
+
+    def documented_fixture(fixture):
+        fixture.__doc__ = fixture.__doc__.format(*args)
+        return fixture
+
+    return documented_fixture
diff --git a/pandas/pandas/util/_tester.py b/pandas/pandas/util/_tester.py
new file mode 100644
index 0000000000000..c0e9756372f47
--- /dev/null
+++ b/pandas/pandas/util/_tester.py
@@ -0,0 +1,58 @@
+"""
+Entrypoint for testing from the top-level namespace.
+"""
+
+from __future__ import annotations
+
+import os
+import sys
+
+from pandas.compat._optional import import_optional_dependency
+
+PKG = os.path.dirname(os.path.dirname(__file__))
+
+
+def test(extra_args: list[str] | None = None, run_doctests: bool = False) -> None:
+    """
+    Run the pandas test suite using pytest.
+
+    By default, runs with the marks -m "not slow and not network and not db"
+
+    Parameters
+    ----------
+    extra_args : list[str], default None
+        Extra marks to run the tests.
+    run_doctests : bool, default False
+        Whether to only run the Python and Cython doctests. If you would like to run
+        both doctests/regular tests, just append "--doctest-modules"/"--doctest-cython"
+        to extra_args.
+
+    See Also
+    --------
+    pytest.main : The main entry point for pytest testing framework.
+
+    Examples
+    --------
+    >>> pd.test()  # doctest: +SKIP
+    running: pytest...
+    """
+    pytest = import_optional_dependency("pytest")
+    import_optional_dependency("hypothesis")
+    cmd = ["-m not slow and not network and not db"]
+    if extra_args:
+        if not isinstance(extra_args, list):
+            extra_args = [extra_args]
+        cmd = extra_args
+    if run_doctests:
+        cmd = [
+            "--doctest-modules",
+            "--doctest-cython",
+            f"--ignore={os.path.join(PKG, 'tests')}",
+        ]
+    cmd += [PKG]
+    joined = " ".join(cmd)
+    print(f"running: pytest {joined}")
+    sys.exit(pytest.main(cmd))
+
+
+__all__ = ["test"]
diff --git a/pandas/pandas/util/_validators.py b/pandas/pandas/util/_validators.py
new file mode 100644
index 0000000000000..9aab19fe340ec
--- /dev/null
+++ b/pandas/pandas/util/_validators.py
@@ -0,0 +1,454 @@
+"""
+Module that contains many useful utilities
+for validating data or function arguments
+"""
+
+from __future__ import annotations
+
+from collections.abc import (
+    Iterable,
+    Sequence,
+)
+from typing import (
+    TypeVar,
+    overload,
+)
+
+import numpy as np
+
+from pandas._libs import lib
+
+from pandas.core.dtypes.common import (
+    is_bool,
+    is_integer,
+)
+
+BoolishT = TypeVar("BoolishT", bool, int)
+BoolishNoneT = TypeVar("BoolishNoneT", bool, int, None)
+
+
+def _check_arg_length(fname, args, max_fname_arg_count, compat_args) -> None:
+    """
+    Checks whether 'args' has length of at most 'compat_args'. Raises
+    a TypeError if that is not the case, similar to in Python when a
+    function is called with too many arguments.
+    """
+    if max_fname_arg_count < 0:
+        raise ValueError("'max_fname_arg_count' must be non-negative")
+
+    if len(args) > len(compat_args):
+        max_arg_count = len(compat_args) + max_fname_arg_count
+        actual_arg_count = len(args) + max_fname_arg_count
+        argument = "argument" if max_arg_count == 1 else "arguments"
+
+        raise TypeError(
+            f"{fname}() takes at most {max_arg_count} {argument} "
+            f"({actual_arg_count} given)"
+        )
+
+
+def _check_for_default_values(fname, arg_val_dict, compat_args) -> None:
+    """
+    Check that the keys in `arg_val_dict` are mapped to their
+    default values as specified in `compat_args`.
+
+    Note that this function is to be called only when it has been
+    checked that arg_val_dict.keys() is a subset of compat_args
+    """
+    for key in arg_val_dict:
+        # try checking equality directly with '=' operator,
+        # as comparison may have been overridden for the left
+        # hand object
+        try:
+            v1 = arg_val_dict[key]
+            v2 = compat_args[key]
+
+            # check for None-ness otherwise we could end up
+            # comparing a numpy array vs None
+            if (v1 is not None and v2 is None) or (v1 is None and v2 is not None):
+                match = False
+            else:
+                match = v1 == v2
+
+            if not is_bool(match):
+                raise ValueError("'match' is not a boolean")
+
+        # could not compare them directly, so try comparison
+        # using the 'is' operator
+        except ValueError:
+            match = arg_val_dict[key] is compat_args[key]
+
+        if not match:
+            raise ValueError(
+                f"the '{key}' parameter is not supported in "
+                f"the pandas implementation of {fname}()"
+            )
+
+
+def validate_args(fname, args, max_fname_arg_count, compat_args) -> None:
+    """
+    Checks whether the length of the `*args` argument passed into a function
+    has at most `len(compat_args)` arguments and whether or not all of these
+    elements in `args` are set to their default values.
+
+    Parameters
+    ----------
+    fname : str
+        The name of the function being passed the `*args` parameter
+    args : tuple
+        The `*args` parameter passed into a function
+    max_fname_arg_count : int
+        The maximum number of arguments that the function `fname`
+        can accept, excluding those in `args`. Used for displaying
+        appropriate error messages. Must be non-negative.
+    compat_args : dict
+        A dictionary of keys and their associated default values.
+        In order to accommodate buggy behaviour in some versions of `numpy`,
+        where a signature displayed keyword arguments but then passed those
+        arguments **positionally** internally when calling downstream
+        implementations, a dict ensures that the original
+        order of the keyword arguments is enforced.
+
+    Raises
+    ------
+    TypeError
+        If `args` contains more values than there are `compat_args`
+    ValueError
+        If `args` contains values that do not correspond to those
+        of the default values specified in `compat_args`
+    """
+    _check_arg_length(fname, args, max_fname_arg_count, compat_args)
+
+    # We do this so that we can provide a more informative
+    # error message about the parameters that we are not
+    # supporting in the pandas implementation of 'fname'
+    kwargs = dict(zip(compat_args, args))
+    _check_for_default_values(fname, kwargs, compat_args)
+
+
+def _check_for_invalid_keys(fname, kwargs, compat_args) -> None:
+    """
+    Checks whether 'kwargs' contains any keys that are not
+    in 'compat_args' and raises a TypeError if there is one.
+    """
+    # set(dict) --> set of the dictionary's keys
+    diff = set(kwargs) - set(compat_args)
+
+    if diff:
+        bad_arg = next(iter(diff))
+        raise TypeError(f"{fname}() got an unexpected keyword argument '{bad_arg}'")
+
+
+def validate_kwargs(fname, kwargs, compat_args) -> None:
+    """
+    Checks whether parameters passed to the **kwargs argument in a
+    function `fname` are valid parameters as specified in `*compat_args`
+    and whether or not they are set to their default values.
+
+    Parameters
+    ----------
+    fname : str
+        The name of the function being passed the `**kwargs` parameter
+    kwargs : dict
+        The `**kwargs` parameter passed into `fname`
+    compat_args: dict
+        A dictionary of keys that `kwargs` is allowed to have and their
+        associated default values
+
+    Raises
+    ------
+    TypeError if `kwargs` contains keys not in `compat_args`
+    ValueError if `kwargs` contains keys in `compat_args` that do not
+    map to the default values specified in `compat_args`
+    """
+    kwds = kwargs.copy()
+    _check_for_invalid_keys(fname, kwargs, compat_args)
+    _check_for_default_values(fname, kwds, compat_args)
+
+
+def validate_args_and_kwargs(
+    fname, args, kwargs, max_fname_arg_count, compat_args
+) -> None:
+    """
+    Checks whether parameters passed to the *args and **kwargs argument in a
+    function `fname` are valid parameters as specified in `*compat_args`
+    and whether or not they are set to their default values.
+
+    Parameters
+    ----------
+    fname: str
+        The name of the function being passed the `**kwargs` parameter
+    args: tuple
+        The `*args` parameter passed into a function
+    kwargs: dict
+        The `**kwargs` parameter passed into `fname`
+    max_fname_arg_count: int
+        The minimum number of arguments that the function `fname`
+        requires, excluding those in `args`. Used for displaying
+        appropriate error messages. Must be non-negative.
+    compat_args: dict
+        A dictionary of keys that `kwargs` is allowed to
+        have and their associated default values.
+
+    Raises
+    ------
+    TypeError if `args` contains more values than there are
+    `compat_args` OR `kwargs` contains keys not in `compat_args`
+    ValueError if `args` contains values not at the default value (`None`)
+    `kwargs` contains keys in `compat_args` that do not map to the default
+    value as specified in `compat_args`
+
+    See Also
+    --------
+    validate_args : Purely args validation.
+    validate_kwargs : Purely kwargs validation.
+
+    """
+    # Check that the total number of arguments passed in (i.e.
+    # args and kwargs) does not exceed the length of compat_args
+    _check_arg_length(
+        fname, args + tuple(kwargs.values()), max_fname_arg_count, compat_args
+    )
+
+    # Check there is no overlap with the positional and keyword
+    # arguments, similar to what is done in actual Python functions
+    args_dict = dict(zip(compat_args, args))
+
+    for key in args_dict:
+        if key in kwargs:
+            raise TypeError(
+                f"{fname}() got multiple values for keyword argument '{key}'"
+            )
+
+    kwargs.update(args_dict)
+    validate_kwargs(fname, kwargs, compat_args)
+
+
+def validate_bool_kwarg(
+    value: BoolishNoneT,
+    arg_name: str,
+    none_allowed: bool = True,
+    int_allowed: bool = False,
+) -> BoolishNoneT:
+    """
+    Ensure that argument passed in arg_name can be interpreted as boolean.
+
+    Parameters
+    ----------
+    value : bool
+        Value to be validated.
+    arg_name : str
+        Name of the argument. To be reflected in the error message.
+    none_allowed : bool, default True
+        Whether to consider None to be a valid boolean.
+    int_allowed : bool, default False
+        Whether to consider integer value to be a valid boolean.
+
+    Returns
+    -------
+    value
+        The same value as input.
+
+    Raises
+    ------
+    ValueError
+        If the value is not a valid boolean.
+    """
+    good_value = is_bool(value)
+    if none_allowed:
+        good_value = good_value or (value is None)
+
+    if int_allowed:
+        good_value = good_value or isinstance(value, int)
+
+    if not good_value:
+        raise ValueError(
+            f'For argument "{arg_name}" expected type bool, received '
+            f"type {type(value).__name__}."
+        )
+    return value
+
+
+def validate_fillna_kwargs(value, method, validate_scalar_dict_value: bool = True):
+    """
+    Validate the keyword arguments to 'fillna'.
+
+    This checks that exactly one of 'value' and 'method' is specified.
+    If 'method' is specified, this validates that it's a valid method.
+
+    Parameters
+    ----------
+    value, method : object
+        The 'value' and 'method' keyword arguments for 'fillna'.
+    validate_scalar_dict_value : bool, default True
+        Whether to validate that 'value' is a scalar or dict. Specifically,
+        validate that it is not a list or tuple.
+
+    Returns
+    -------
+    value, method : object
+    """
+    from pandas.core.missing import clean_fill_method
+
+    if value is None and method is None:
+        raise ValueError("Must specify a fill 'value' or 'method'.")
+    if value is None and method is not None:
+        method = clean_fill_method(method)
+
+    elif value is not None and method is None:
+        if validate_scalar_dict_value and isinstance(value, (list, tuple)):
+            raise TypeError(
+                '"value" parameter must be a scalar or dict, but '
+                f'you passed a "{type(value).__name__}"'
+            )
+
+    elif value is not None and method is not None:
+        raise ValueError("Cannot specify both 'value' and 'method'.")
+
+    return value, method
+
+
+def validate_percentile(q: float | Iterable[float]) -> np.ndarray:
+    """
+    Validate percentiles (used by describe and quantile).
+
+    This function checks if the given float or iterable of floats is a valid percentile
+    otherwise raises a ValueError.
+
+    Parameters
+    ----------
+    q: float or iterable of floats
+        A single percentile or an iterable of percentiles.
+
+    Returns
+    -------
+    ndarray
+        An ndarray of the percentiles if valid.
+
+    Raises
+    ------
+    ValueError if percentiles are not in given interval([0, 1]).
+    """
+    q_arr = np.asarray(q)
+    # Don't change this to an f-string. The string formatting
+    # is too expensive for cases where we don't need it.
+    msg = "percentiles should all be in the interval [0, 1]"
+    if q_arr.ndim == 0:
+        if not 0 <= q_arr <= 1:
+            raise ValueError(msg)
+    elif not all(0 <= qs <= 1 for qs in q_arr):
+        raise ValueError(msg)
+    return q_arr
+
+
+@overload
+def validate_ascending(ascending: BoolishT) -> BoolishT: ...
+
+
+@overload
+def validate_ascending(ascending: Sequence[BoolishT]) -> list[BoolishT]: ...
+
+
+def validate_ascending(
+    ascending: bool | int | Sequence[BoolishT],
+) -> bool | int | list[BoolishT]:
+    """Validate ``ascending`` kwargs for ``sort_index`` method."""
+    kwargs = {"none_allowed": False, "int_allowed": True}
+    if not isinstance(ascending, Sequence):
+        return validate_bool_kwarg(ascending, "ascending", **kwargs)
+
+    return [validate_bool_kwarg(item, "ascending", **kwargs) for item in ascending]
+
+
+def validate_endpoints(closed: str | None) -> tuple[bool, bool]:
+    """
+    Check that the `closed` argument is among [None, "left", "right"]
+
+    Parameters
+    ----------
+    closed : {None, "left", "right"}
+
+    Returns
+    -------
+    left_closed : bool
+    right_closed : bool
+
+    Raises
+    ------
+    ValueError : if argument is not among valid values
+    """
+    left_closed = False
+    right_closed = False
+
+    if closed is None:
+        left_closed = True
+        right_closed = True
+    elif closed == "left":
+        left_closed = True
+    elif closed == "right":
+        right_closed = True
+    else:
+        raise ValueError("Closed has to be either 'left', 'right' or None")
+
+    return left_closed, right_closed
+
+
+def validate_inclusive(inclusive: str | None) -> tuple[bool, bool]:
+    """
+    Check that the `inclusive` argument is among {"both", "neither", "left", "right"}.
+
+    Parameters
+    ----------
+    inclusive : {"both", "neither", "left", "right"}
+
+    Returns
+    -------
+    left_right_inclusive : tuple[bool, bool]
+
+    Raises
+    ------
+    ValueError : if argument is not among valid values
+    """
+    left_right_inclusive: tuple[bool, bool] | None = None
+
+    if isinstance(inclusive, str):
+        left_right_inclusive = {
+            "both": (True, True),
+            "left": (True, False),
+            "right": (False, True),
+            "neither": (False, False),
+        }.get(inclusive)
+
+    if left_right_inclusive is None:
+        raise ValueError(
+            "Inclusive has to be either 'both', 'neither', 'left' or 'right'"
+        )
+
+    return left_right_inclusive
+
+
+def validate_insert_loc(loc: int, length: int) -> int:
+    """
+    Check that we have an integer between -length and length, inclusive.
+
+    Standardize negative loc to within [0, length].
+
+    The exceptions we raise on failure match np.insert.
+    """
+    if not is_integer(loc):
+        raise TypeError(f"loc must be an integer between -{length} and {length}")
+
+    if loc < 0:
+        loc += length
+    if not 0 <= loc <= length:
+        raise IndexError(f"loc must be an integer between -{length} and {length}")
+    return loc  # pyright: ignore[reportReturnType]
+
+
+def check_dtype_backend(dtype_backend) -> None:
+    if dtype_backend is not lib.no_default:
+        if dtype_backend not in ["numpy_nullable", "pyarrow"]:
+            raise ValueError(
+                f"dtype_backend {dtype_backend} is invalid, only 'numpy_nullable' and "
+                f"'pyarrow' are allowed.",
+            )
diff --git a/pandas/pandas/util/version/__init__.py b/pandas/pandas/util/version/__init__.py
new file mode 100644
index 0000000000000..bd741140f6542
--- /dev/null
+++ b/pandas/pandas/util/version/__init__.py
@@ -0,0 +1,453 @@
+# Vendored from https://github.com/pypa/packaging/blob/main/src/packaging/_structures.py
+# and https://github.com/pypa/packaging/blob/main/src/packaging/version.py
+# changeset 24e5350b2ff3c5c7a36676c2af5f2cb39fd1baf8
+
+# This file is dual licensed under the terms of the Apache License, Version
+# 2.0, and the BSD License. Licence at LICENSES/PACKAGING_LICENSE
+from __future__ import annotations
+
+from collections.abc import Callable
+import itertools
+import re
+from typing import (
+    Any,
+    NamedTuple,
+    SupportsInt,
+    Union,
+)
+
+__all__ = ["VERSION_PATTERN", "InvalidVersion", "Version", "parse"]
+
+
+class InfinityType:
+    def __repr__(self) -> str:
+        return "Infinity"
+
+    def __hash__(self) -> int:
+        return hash(repr(self))
+
+    def __lt__(self, other: object) -> bool:
+        return False
+
+    def __le__(self, other: object) -> bool:
+        return False
+
+    def __eq__(self, other: object) -> bool:
+        return isinstance(other, type(self))
+
+    def __gt__(self, other: object) -> bool:
+        return True
+
+    def __ge__(self, other: object) -> bool:
+        return True
+
+    def __neg__(self: object) -> NegativeInfinityType:
+        return NegativeInfinity
+
+
+Infinity = InfinityType()
+
+
+class NegativeInfinityType:
+    def __repr__(self) -> str:
+        return "-Infinity"
+
+    def __hash__(self) -> int:
+        return hash(repr(self))
+
+    def __lt__(self, other: object) -> bool:
+        return True
+
+    def __le__(self, other: object) -> bool:
+        return True
+
+    def __eq__(self, other: object) -> bool:
+        return isinstance(other, type(self))
+
+    def __gt__(self, other: object) -> bool:
+        return False
+
+    def __ge__(self, other: object) -> bool:
+        return False
+
+    def __neg__(self: object) -> InfinityType:
+        return Infinity
+
+
+NegativeInfinity = NegativeInfinityType()
+
+
+LocalType = tuple[Union[int, str], ...]
+
+CmpPrePostDevType = Union[InfinityType, NegativeInfinityType, tuple[str, int]]
+CmpLocalType = Union[
+    NegativeInfinityType,
+    tuple[Union[tuple[int, str], tuple[NegativeInfinityType, Union[int, str]]], ...],
+]
+CmpKey = tuple[
+    int,
+    tuple[int, ...],
+    CmpPrePostDevType,
+    CmpPrePostDevType,
+    CmpPrePostDevType,
+    CmpLocalType,
+]
+VersionComparisonMethod = Callable[[CmpKey, CmpKey], bool]
+
+
+class _Version(NamedTuple):
+    epoch: int
+    release: tuple[int, ...]
+    dev: tuple[str, int] | None
+    pre: tuple[str, int] | None
+    post: tuple[str, int] | None
+    local: LocalType | None
+
+
+def parse(version: str) -> Version:
+    return Version(version)
+
+
+# The docstring is from an older version of the packaging library to avoid
+# errors in the docstring validation.
+class InvalidVersion(ValueError):
+    """
+    An invalid version was found, users should refer to PEP 440.
+
+    The ``InvalidVersion`` exception is raised when a version string is
+    improperly formatted. Pandas uses this exception to ensure that all
+    version strings are PEP 440 compliant.
+
+    See Also
+    --------
+    util.version.Version : Class for handling and parsing version strings.
+
+    Examples
+    --------
+    >>> pd.util.version.Version("1.")
+    Traceback (most recent call last):
+    InvalidVersion: Invalid version: '1.'
+    """
+
+
+class _BaseVersion:
+    _key: tuple[Any, ...]
+
+    def __hash__(self) -> int:
+        return hash(self._key)
+
+    # Please keep the duplicated `isinstance` check
+    # in the six comparisons hereunder
+    # unless you find a way to avoid adding overhead function calls.
+    def __lt__(self, other: _BaseVersion) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key < other._key
+
+    def __le__(self, other: _BaseVersion) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key <= other._key
+
+    def __eq__(self, other: object) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key == other._key
+
+    def __ge__(self, other: _BaseVersion) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key >= other._key
+
+    def __gt__(self, other: _BaseVersion) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key > other._key
+
+    def __ne__(self, other: object) -> bool:
+        if not isinstance(other, _BaseVersion):
+            return NotImplemented
+
+        return self._key != other._key
+
+
+# Deliberately not anchored to the start and end of the string, to make it
+# easier for 3rd party code to reuse
+_VERSION_PATTERN = r"""
+    v?
+    (?:
+        (?:(?P<epoch>[0-9]+)!)?                           # epoch
+        (?P<release>[0-9]+(?:\.[0-9]+)*)                  # release segment
+        (?P<pre>                                          # pre-release
+            [-_\.]?
+            (?P<pre_l>alpha|a|beta|b|preview|pre|c|rc)
+            [-_\.]?
+            (?P<pre_n>[0-9]+)?
+        )?
+        (?P<post>                                         # post release
+            (?:-(?P<post_n1>[0-9]+))
+            |
+            (?:
+                [-_\.]?
+                (?P<post_l>post|rev|r)
+                [-_\.]?
+                (?P<post_n2>[0-9]+)?
+            )
+        )?
+        (?P<dev>                                          # dev release
+            [-_\.]?
+            (?P<dev_l>dev)
+            [-_\.]?
+            (?P<dev_n>[0-9]+)?
+        )?
+    )
+    (?:\+(?P<local>[a-z0-9]+(?:[-_\.][a-z0-9]+)*))?       # local version
+"""
+
+VERSION_PATTERN = _VERSION_PATTERN
+
+
+class Version(_BaseVersion):
+    _regex = re.compile(r"^\s*" + VERSION_PATTERN + r"\s*$", re.VERBOSE | re.IGNORECASE)
+    _key: CmpKey
+
+    def __init__(self, version: str) -> None:
+        # Validate the version and parse it into pieces
+        match = self._regex.search(version)
+        if not match:
+            raise InvalidVersion(f"Invalid version: '{version}'")
+
+        # Store the parsed out pieces of the version
+        self._version = _Version(
+            epoch=int(match.group("epoch")) if match.group("epoch") else 0,
+            release=tuple(int(i) for i in match.group("release").split(".")),
+            pre=_parse_letter_version(match.group("pre_l"), match.group("pre_n")),
+            post=_parse_letter_version(
+                match.group("post_l"), match.group("post_n1") or match.group("post_n2")
+            ),
+            dev=_parse_letter_version(match.group("dev_l"), match.group("dev_n")),
+            local=_parse_local_version(match.group("local")),
+        )
+
+        # Generate a key which will be used for sorting
+        self._key = _cmpkey(
+            self._version.epoch,
+            self._version.release,
+            self._version.pre,
+            self._version.post,
+            self._version.dev,
+            self._version.local,
+        )
+
+    def __repr__(self) -> str:
+        return f"<Version('{self}')>"
+
+    def __str__(self) -> str:
+        parts = []
+
+        # Epoch
+        if self.epoch != 0:
+            parts.append(f"{self.epoch}!")
+
+        # Release segment
+        parts.append(".".join(str(x) for x in self.release))
+
+        # Pre-release
+        if self.pre is not None:
+            parts.append("".join(str(x) for x in self.pre))
+
+        # Post-release
+        if self.post is not None:
+            parts.append(f".post{self.post}")
+
+        # Development release
+        if self.dev is not None:
+            parts.append(f".dev{self.dev}")
+
+        # Local version segment
+        if self.local is not None:
+            parts.append(f"+{self.local}")
+
+        return "".join(parts)
+
+    @property
+    def epoch(self) -> int:
+        return self._version.epoch
+
+    @property
+    def release(self) -> tuple[int, ...]:
+        return self._version.release
+
+    @property
+    def pre(self) -> tuple[str, int] | None:
+        return self._version.pre
+
+    @property
+    def post(self) -> int | None:
+        return self._version.post[1] if self._version.post else None
+
+    @property
+    def dev(self) -> int | None:
+        return self._version.dev[1] if self._version.dev else None
+
+    @property
+    def local(self) -> str | None:
+        if self._version.local:
+            return ".".join(str(x) for x in self._version.local)
+        else:
+            return None
+
+    @property
+    def public(self) -> str:
+        return str(self).split("+", 1)[0]
+
+    @property
+    def base_version(self) -> str:
+        parts = []
+
+        # Epoch
+        if self.epoch != 0:
+            parts.append(f"{self.epoch}!")
+
+        # Release segment
+        parts.append(".".join(str(x) for x in self.release))
+
+        return "".join(parts)
+
+    @property
+    def is_prerelease(self) -> bool:
+        return self.dev is not None or self.pre is not None
+
+    @property
+    def is_postrelease(self) -> bool:
+        return self.post is not None
+
+    @property
+    def is_devrelease(self) -> bool:
+        return self.dev is not None
+
+    @property
+    def major(self) -> int:
+        return self.release[0] if len(self.release) >= 1 else 0
+
+    @property
+    def minor(self) -> int:
+        return self.release[1] if len(self.release) >= 2 else 0
+
+    @property
+    def micro(self) -> int:
+        return self.release[2] if len(self.release) >= 3 else 0
+
+
+def _parse_letter_version(
+    letter: str | None, number: str | bytes | SupportsInt | None
+) -> tuple[str, int] | None:
+    if letter:
+        # We consider there to be an implicit 0 in a pre-release if there is
+        # not a numeral associated with it.
+        if number is None:
+            number = 0
+
+        # We normalize any letters to their lower case form
+        letter = letter.lower()
+
+        # We consider some words to be alternate spellings of other words and
+        # in those cases we want to normalize the spellings to our preferred
+        # spelling.
+        if letter == "alpha":
+            letter = "a"
+        elif letter == "beta":
+            letter = "b"
+        elif letter in ["c", "pre", "preview"]:
+            letter = "rc"
+        elif letter in ["rev", "r"]:
+            letter = "post"
+
+        return letter, int(number)
+    if not letter and number:
+        # We assume if we are given a number, but we are not given a letter
+        # then this is using the implicit post release syntax (e.g. 1.0-1)
+        letter = "post"
+
+        return letter, int(number)
+
+    return None
+
+
+_local_version_separators = re.compile(r"[\._-]")
+
+
+def _parse_local_version(local: str | None) -> LocalType | None:
+    if local is not None:
+        return tuple(
+            part.lower() if not part.isdigit() else int(part)
+            for part in _local_version_separators.split(local)
+        )
+    return None
+
+
+def _cmpkey(
+    epoch: int,
+    release: tuple[int, ...],
+    pre: tuple[str, int] | None,
+    post: tuple[str, int] | None,
+    dev: tuple[str, int] | None,
+    local: LocalType | None,
+) -> CmpKey:
+    # When we compare a release version, we want to compare it with all of the
+    # trailing zeros removed. So we'll use a reverse the list, drop all the now
+    # leading zeros until we come to something non zero, then take the rest
+    # re-reverse it back into the correct order and make it a tuple and use
+    # that for our sorting key.
+    _release = tuple(
+        reversed(list(itertools.dropwhile(lambda x: x == 0, reversed(release))))
+    )
+
+    # We need to "trick" the sorting algorithm to put 1.0.dev0 before 1.0a0.
+    # We'll do this by abusing the pre segment, but we _only_ want to do this
+    # if there is not a pre or a post segment. If we have one of those then
+    # the normal sorting rules will handle this case correctly.
+    if pre is None and post is None and dev is not None:
+        _pre: CmpPrePostDevType = NegativeInfinity
+    # Versions without a pre-release (except as noted above) should sort after
+    # those with one.
+    elif pre is None:
+        _pre = Infinity
+    else:
+        _pre = pre
+
+    # Versions without a post segment should sort before those with one.
+    if post is None:
+        _post: CmpPrePostDevType = NegativeInfinity
+
+    else:
+        _post = post
+
+    # Versions without a development segment should sort after those with one.
+    if dev is None:
+        _dev: CmpPrePostDevType = Infinity
+
+    else:
+        _dev = dev
+
+    if local is None:
+        # Versions without a local segment should sort before those with one.
+        _local: CmpLocalType = NegativeInfinity
+    else:
+        # Versions with a local segment need that segment parsed to implement
+        # the sorting rules in PEP440.
+        # - Alpha numeric segments sort before numeric segments
+        # - Alpha numeric segments sort lexicographically
+        # - Numeric segments sort numerically
+        # - Shorter versions sort before longer versions when the prefixes
+        #   match exactly
+        _local = tuple(
+            (i, "") if isinstance(i, int) else (NegativeInfinity, i) for i in local
+        )
+
+    return epoch, _release, _pre, _post, _dev, _local
diff --git a/pandas/pyproject.toml b/pandas/pyproject.toml
new file mode 100644
index 0000000000000..b7d53b0d8934a
--- /dev/null
+++ b/pandas/pyproject.toml
@@ -0,0 +1,746 @@
+[build-system]
+# Minimum requirements for the build system to execute.
+# See https://github.com/scipy/scipy/pull/12940 for the AIX issue.
+requires = [
+    "meson-python>=0.13.1",
+    "meson>=1.2.1,<2",
+    "wheel",
+    "Cython~=3.0.5",  # Note: sync with setup.py, environment.yml and asv.conf.json
+    # Force numpy higher than 2.0rc1, so that built wheels are compatible
+    # with both numpy 1 and 2
+    "numpy>=2.0.0rc1",
+    "versioneer[toml]"
+]
+
+build-backend = "mesonpy"
+
+[project]
+name = 'pandas'
+dynamic = [
+  'version'
+]
+description = 'Powerful data structures for data analysis, time series, and statistics'
+readme = 'README.md'
+authors = [
+  { name = 'The Pandas Development Team', email='pandas-dev@python.org' },
+]
+license = {file = 'LICENSE'}
+requires-python = '>=3.10'
+dependencies = [
+  "numpy>=1.23.5; python_version<'3.12'",
+  "numpy>=1.26.0; python_version>='3.12'",
+  "python-dateutil>=2.8.2",
+  "tzdata>=2022.7"
+]
+classifiers = [
+    'Development Status :: 5 - Production/Stable',
+    'Environment :: Console',
+    'Intended Audience :: Science/Research',
+    'License :: OSI Approved :: BSD License',
+    'Operating System :: OS Independent',
+    'Programming Language :: Cython',
+    'Programming Language :: Python',
+    'Programming Language :: Python :: 3',
+    'Programming Language :: Python :: 3 :: Only',
+    'Programming Language :: Python :: 3.10',
+    'Programming Language :: Python :: 3.11',
+    'Programming Language :: Python :: 3.12',
+    'Programming Language :: Python :: 3.13',
+    'Topic :: Scientific/Engineering'
+]
+
+[project.urls]
+homepage = 'https://pandas.pydata.org'
+documentation = 'https://pandas.pydata.org/docs/'
+repository = 'https://github.com/pandas-dev/pandas'
+
+[project.entry-points."pandas_plotting_backends"]
+matplotlib = "pandas:plotting._matplotlib"
+
+[project.optional-dependencies]
+test = ['hypothesis>=6.84.0', 'pytest>=7.3.2', 'pytest-xdist>=3.4.0']
+pyarrow = ['pyarrow>=10.0.1']
+performance = ['bottleneck>=1.3.6', 'numba>=0.56.4', 'numexpr>=2.8.4']
+computation = ['scipy>=1.10.0', 'xarray>=2022.12.0']
+fss = ['fsspec>=2022.11.0']
+aws = ['s3fs>=2022.11.0']
+gcp = ['gcsfs>=2022.11.0']
+excel = ['odfpy>=1.4.1', 'openpyxl>=3.1.0', 'python-calamine>=0.1.7', 'pyxlsb>=1.0.10', 'xlrd>=2.0.1', 'xlsxwriter>=3.0.5']
+parquet = ['pyarrow>=10.0.1']
+feather = ['pyarrow>=10.0.1']
+hdf5 = [# blosc only available on conda (https://github.com/Blosc/python-blosc/issues/297)
+        #'blosc>=1.20.1',
+        'tables>=3.8.0']
+spss = ['pyreadstat>=1.2.0']
+postgresql = ['SQLAlchemy>=2.0.0', 'psycopg2>=2.9.6', 'adbc-driver-postgresql>=0.10.0']
+mysql = ['SQLAlchemy>=2.0.0', 'pymysql>=1.0.2']
+sql-other = ['SQLAlchemy>=2.0.0', 'adbc-driver-postgresql>=0.10.0', 'adbc-driver-sqlite>=0.8.0']
+html = ['beautifulsoup4>=4.11.2', 'html5lib>=1.1', 'lxml>=4.9.2']
+xml = ['lxml>=4.9.2']
+plot = ['matplotlib>=3.6.3']
+output-formatting = ['jinja2>=3.1.2', 'tabulate>=0.9.0']
+clipboard = ['PyQt5>=5.15.9', 'qtpy>=2.3.0']
+compression = ['zstandard>=0.19.0']
+timezone = ['pytz>=2023.4']
+all = ['adbc-driver-postgresql>=0.10.0',
+       'adbc-driver-sqlite>=0.8.0',
+       'beautifulsoup4>=4.11.2',
+       # blosc only available on conda (https://github.com/Blosc/python-blosc/issues/297)
+       #'blosc>=1.21.3',
+       'bottleneck>=1.3.6',
+       'fastparquet>=2023.10.0',
+       'fsspec>=2022.11.0',
+       'gcsfs>=2022.11.0',
+       'html5lib>=1.1',
+       'hypothesis>=6.84.0',
+       'jinja2>=3.1.2',
+       'lxml>=4.9.2',
+       'matplotlib>=3.6.3',
+       'numba>=0.56.4',
+       'numexpr>=2.8.4',
+       'odfpy>=1.4.1',
+       'openpyxl>=3.1.0',
+       'psycopg2>=2.9.6',
+       'pyarrow>=10.0.1',
+       'pymysql>=1.0.2',
+       'PyQt5>=5.15.9',
+       'pyreadstat>=1.2.0',
+       'pytest>=7.3.2',
+       'pytest-xdist>=3.4.0',
+       'python-calamine>=0.1.7',
+       'pytz>=2023.4',
+       'pyxlsb>=1.0.10',
+       'qtpy>=2.3.0',
+       'scipy>=1.10.0',
+       's3fs>=2022.11.0',
+       'SQLAlchemy>=2.0.0',
+       'tables>=3.8.0',
+       'tabulate>=0.9.0',
+       'xarray>=2022.12.0',
+       'xlrd>=2.0.1',
+       'xlsxwriter>=3.0.5',
+       'zstandard>=0.19.0']
+
+# TODO: Remove after setuptools support is dropped.
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+include = ["pandas", "pandas.*"]
+namespaces = false
+
+[tool.setuptools.exclude-package-data]
+"*" = ["*.c", "*.h"]
+
+# See the docstring in versioneer.py for instructions. Note that you must
+# re-run 'versioneer.py setup' after changing this section, and commit the
+# resulting files.
+[tool.versioneer]
+VCS = "git"
+style = "pep440"
+versionfile_source = "pandas/_version.py"
+versionfile_build = "pandas/_version.py"
+tag_prefix = "v"
+parentdir_prefix = "pandas-"
+
+[tool.meson-python.args]
+setup = ['--vsenv'] # For Windows
+
+[tool.cibuildwheel]
+skip = "cp36-* cp37-* cp38-* cp39-* pp* *_i686 *_ppc64le *_s390x"
+build-verbosity = "3"
+environment = {LDFLAGS="-Wl,--strip-all"}
+test-requires = "hypothesis>=6.84.0 pytest>=7.3.2 pytest-xdist>=3.4.0"
+test-command = """
+  PANDAS_CI='1' python -c 'import pandas as pd; \
+  pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db", "-n 2", "--no-strict-data-files"]); \
+  pd.test(extra_args=["-m not clipboard and single_cpu and not slow and not network and not db", "--no-strict-data-files"]);' \
+  """
+free-threaded-support = true
+before-build = "PACKAGE_DIR={package} bash {package}/scripts/cibw_before_build.sh"
+
+[tool.cibuildwheel.windows]
+before-build = "pip install delvewheel && bash {package}/scripts/cibw_before_build_windows.sh"
+before-test = "bash {package}/scripts/cibw_before_test_windows.sh"
+test-command = """
+  set PANDAS_CI='1' && \
+  python -c "import pandas as pd; \
+  pd.test(extra_args=['--no-strict-data-files', '-m not clipboard and not single_cpu and not slow and not network and not db']);" \
+  """
+repair-wheel-command = "delvewheel repair -w {dest_dir} {wheel}"
+
+[[tool.cibuildwheel.overrides]]
+select = "*-manylinux_aarch64*"
+test-command = """
+  PANDAS_CI='1' python -c 'import pandas as pd; \
+  pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db and not fails_arm_wheels", "-n 2", "--no-strict-data-files"]); \
+  pd.test(extra_args=["-m not clipboard and single_cpu and not slow and not network and not db", "--no-strict-data-files"]);' \
+  """
+
+[[tool.cibuildwheel.overrides]]
+select = "*-musllinux*"
+before-test = "apk update && apk add musl-locales"
+
+[[tool.cibuildwheel.overrides]]
+# Don't strip wheels on macOS.
+# macOS doesn't support stripping wheels with linker
+# https://github.com/MacPython/numpy-wheels/pull/87#issuecomment-624878264
+select = "*-macosx*"
+environment = {CFLAGS="-g0"}
+
+[[tool.cibuildwheel.overrides]]
+select = "*pyodide*"
+test-requires = "pytest>=7.3.2 hypothesis>=6.84.0"
+# Pyodide repairs wheels on its own, using auditwheel-emscripten
+repair-wheel-command = ""
+test-command = """
+  PANDAS_CI='1' python -c 'import pandas as pd; \
+  pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db", "--no-strict-data-files"]);' \
+  """
+
+[tool.ruff]
+line-length = 88
+target-version = "py310"
+fix = true
+
+[tool.ruff.lint]
+unfixable = []
+typing-modules = ["pandas._typing"]
+
+select = [
+  # pyflakes
+  "F",
+  # pycodestyle
+  "E", "W",
+  # flake8-2020
+  "YTT",
+  # flake8-bugbear
+  "B",
+  # flake8-quotes
+  "Q",
+  # flake8-debugger
+  "T10",
+  # flake8-gettext
+  "INT",
+  # pylint
+  "PL",
+  # flake8-pytest-style
+  "PT",
+  # misc lints
+  "PIE",
+  # flake8-pyi
+  "PYI",
+  # tidy imports
+  "TID",
+  # implicit string concatenation
+  "ISC",
+  # type-checking imports
+  "TCH",
+  # comprehensions
+  "C4",
+  # pygrep-hooks
+  "PGH",
+  # Ruff-specific rules
+  "RUF",
+  # flake8-bandit: exec-builtin
+  "S102",
+  # numpy-legacy-random
+  "NPY002",
+  # Perflint
+  "PERF",
+  # flynt
+  "FLY",
+  # flake8-logging-format
+  "G",
+  # flake8-future-annotations
+  "FA",
+  # unconventional-import-alias
+  "ICN001",
+  # flake8-slots
+  "SLOT",
+  # flake8-raise
+  "RSE"
+]
+
+ignore = [
+  ### Intentionally disabled
+  # module level import not at top of file
+  "E402",
+  # do not assign a lambda expression, use a def
+  "E731",
+  # controversial
+  "B007",
+  # controversial
+  "B008",
+  # getattr is used to side-step mypy
+  "B010",
+  # tests use comparisons but not their returned value
+  "B015",
+  # Function definition does not bind loop variable
+  "B023",
+  # Only works with python >=3.10
+  "B905",
+  # Too many arguments to function call
+  "PLR0913",
+  # Too many returns
+  "PLR0911",
+  # Too many branches
+  "PLR0912",
+  # Too many statements
+  "PLR0915",
+  # Redefined loop name
+  "PLW2901",
+  # Global statements are discouraged
+  "PLW0603",
+  # Use `typing.NamedTuple` instead of `collections.namedtuple`
+  "PYI024",
+  # Use of possibly insecure function; consider using ast.literal_eval
+  "S307",
+  # while int | float can be shortened to float, the former is more explicit
+  "PYI041",
+  # incorrect-dict-iterator, flags valid Series.items usage
+  "PERF102",
+  # try-except-in-loop, becomes useless in Python 3.11
+  "PERF203",
+  # pytest-parametrize-names-wrong-type
+  "PT006",
+  # pytest-parametrize-values-wrong-type
+  "PT007",
+  # pytest-patch-with-lambda
+  "PT008",
+  # pytest-raises-with-multiple-statements
+  "PT012",
+  # pytest-assert-in-except
+  "PT017",
+  # pytest-composite-assertion
+  "PT018",
+  # pytest-fixture-param-without-value
+  "PT019",
+  # The following rules may cause conflicts when used with the formatter:
+  "ISC001",
+  # if-stmt-min-max
+  "PLR1730",
+
+  ### TODO: Enable gradually
+  # Useless statement
+  "B018",
+  # Magic number
+  "PLR2004",
+  # comparison-with-itself
+  "PLR0124",
+  # collection-literal-concatenation
+  "RUF005",
+  # pairwise-over-zipped (>=PY310 only)
+  "RUF007",
+  # mutable-class-default
+  "RUF012",
+  # type-comparison
+  "E721",
+  # repeated-equality-comparison
+  "PLR1714",
+  # self-or-cls-assignment
+  "PLW0642",
+  # literal-membership
+  "PLR6201", # 847 errors
+  # Method could be a function, class method, or static method
+  "PLR6301", # 11411 errors
+  # Private name import
+  "PLC2701", # 27 errors
+  # Too many positional arguments (6/5)
+  "PLR0917", # 470 errors
+  # compare-to-empty-string
+  "PLC1901",
+  # `tempfile.NamedTemporaryFile` in text mode without explicit `encoding` argument
+  "PLW1514", # 1 error
+  # Object does not implement `__hash__` method
+  "PLW1641", # 16 errors
+  # Bad or misspelled dunder method name
+  "PLW3201", # 69 errors, seems to be all false positive
+  # Unnecessary lookup of dictionary value by key
+  "PLR1733", # 5 errors, it seems like we wannt to ignore these
+  # Unnecessary lookup of list item by index
+  "PLR1736", # 4 errors, we're currently having inline pylint ignore
+  # Unpacking a dictionary in iteration without calling `.items()`
+  "PLE1141", # autofixable
+  # import-outside-toplevel
+  "PLC0415",
+  # unnecessary-dunder-call
+  "PLC2801",
+  # comparison-with-itself
+  "PLR0124",
+  # too-many-public-methods
+  "PLR0904",
+  # too-many-return-statements
+  "PLR0911",
+  # too-many-branches
+  "PLR0912",
+  # too-many-arguments
+  "PLR0913",
+  # too-many-locals
+  "PLR0914",
+  # too-many-statements
+  "PLR0915",
+  # too-many-boolean-expressions
+  "PLR0916",
+  # too-many-nested-blocks
+  "PLR1702",
+  # redefined-argument-from-local
+  "PLR1704",
+  # unnecessary-lambda
+  "PLW0108",
+  # global-statement
+  "PLW0603",
+]
+
+exclude = [
+  "doc/sphinxext/*.py",
+  "doc/build/*.py",
+  "doc/temp/*.py",
+  ".eggs/*.py",
+  # vendored files
+  "pandas/util/version/*",
+  "pandas/io/clipboard/__init__.py",
+  # exclude asv benchmark environments from linting
+  "env",
+]
+
+[tool.ruff.lint.flake8-tidy-imports.banned-api]
+"urllib.request.urlopen".msg = "Use pandas.io.common.urlopen instead of urllib.request.urlopen"
+# numpy.random is banned but np.random is not. Is this intentional?
+# "numpy.random".msg = "Do not use numpy.random"
+"pytest.warns".msg = "Use tm.assert_produces_warning instead of pytest.warns"
+"pytest.xfail".msg = "Use pytest.mark.xfail instead of pytest.xfail"
+"conftest".msg = "No direct imports from conftest"
+"numpy.testing".msg = "Do not use numpy.testing"
+# "numpy.array_equal".msg = "Do not use numpy.array_equal" # Used in pandas/core
+"unittest.mock".msg = "use pytest builtin monkeypatch fixture instead"
+"os.remove".msg = "Do not use os.remove"
+
+
+
+[tool.ruff.lint.flake8-import-conventions.aliases]
+"pandas.core.construction.array" = "pd_array"
+
+[tool.ruff.lint.per-file-ignores]
+# relative imports allowed for asv_bench
+"asv_bench/*" = ["TID", "NPY002"]
+# to be enabled gradually
+"pandas/core/*" = ["PLR5501"]
+"pandas/tests/*" = ["B028", "FLY"]
+"scripts/*" = ["B028"]
+# Keep this one enabled
+"pandas/_typing.py" = ["TCH"]
+
+[tool.ruff.lint.flake8-pytest-style]
+fixture-parentheses = false
+mark-parentheses = false
+
+[tool.ruff.format]
+docstring-code-format = true
+
+[tool.pytest.ini_options]
+# sync minversion with pyproject.toml & install.rst
+minversion = "7.3.2"
+addopts = "--strict-markers --strict-config --capture=no --durations=30 --junitxml=test-data.xml"
+empty_parameter_set_mark = "fail_at_collect"
+xfail_strict = true
+testpaths = "pandas"
+doctest_optionflags = [
+  "NORMALIZE_WHITESPACE",
+  "IGNORE_EXCEPTION_DETAIL",
+  "ELLIPSIS",
+]
+filterwarnings = [
+  "error:::pandas",
+  "error::ResourceWarning",
+  "error::pytest.PytestUnraisableExceptionWarning",
+  # TODO(PY311-minimum): Specify EncodingWarning
+  # Ignore 3rd party EncodingWarning but raise on pandas'
+  "ignore:.*encoding.* argument not specified",
+  "error:.*encoding.* argument not specified::pandas",
+  "ignore:.*ssl.SSLSocket:pytest.PytestUnraisableExceptionWarning",
+  "ignore:.*ssl.SSLSocket:ResourceWarning",
+  # GH 44844: Can remove once minimum matplotlib version >= 3.7
+  "ignore:.*FileIO:pytest.PytestUnraisableExceptionWarning",
+  "ignore:.*BufferedRandom:ResourceWarning",
+  "ignore::ResourceWarning:asyncio",
+  # From plotting doctests
+  "ignore:More than 20 figures have been opened:RuntimeWarning",
+  "ignore:.*urllib3:DeprecationWarning:botocore",
+  "ignore:Setuptools is replacing distutils.:UserWarning:_distutils_hack",
+  # https://github.com/PyTables/PyTables/issues/822
+  "ignore:a closed node found in the registry:UserWarning:tables",
+]
+junit_family = "xunit2"
+markers = [
+  "single_cpu: tests that should run on a single cpu only",
+  "slow: mark a test as slow",
+  "network: mark a test as network",
+  "db: tests requiring a database (mysql or postgres)",
+  "clipboard: mark a pd.read_clipboard test",
+  "arm_slow: mark a test as slow for arm64 architecture",
+  "skip_ubsan: Tests known to fail UBSAN check",
+  # TODO: someone should investigate this ...
+  # these tests only fail in the wheel builder and don't fail in regular
+  # ARM CI
+  "fails_arm_wheels: Tests that fail in the ARM wheel build only",
+]
+
+[tool.mypy]
+# Import discovery
+mypy_path = "typings"
+files = ["pandas", "typings"]
+namespace_packages = false
+explicit_package_bases = false
+ignore_missing_imports = true
+follow_imports = "normal"
+follow_imports_for_stubs = false
+no_site_packages = false
+no_silence_site_packages = false
+# Platform configuration
+python_version = "3.11"
+platform = "linux-64"
+# Disallow dynamic typing
+disallow_any_unimported = false # TODO
+disallow_any_expr = false # TODO
+disallow_any_decorated = false # TODO
+disallow_any_explicit = false # TODO
+disallow_any_generics = false # TODO
+disallow_subclassing_any = false # TODO
+# Untyped definitions and calls
+disallow_untyped_calls = true
+disallow_untyped_defs = true
+disallow_incomplete_defs = true
+check_untyped_defs = true
+disallow_untyped_decorators = true
+# None and Optional handling
+no_implicit_optional = true
+strict_optional = true
+# Configuring warnings
+warn_redundant_casts = true
+warn_unused_ignores = true
+warn_no_return = true
+warn_return_any = false # TODO
+warn_unreachable = false # GH#27396
+# Suppressing errors
+ignore_errors = false
+enable_error_code = "ignore-without-code"
+# Miscellaneous strictness flags
+allow_untyped_globals = false
+allow_redefinition = false
+local_partial_types = false
+implicit_reexport = true
+strict_equality = true
+# Configuring error messages
+show_error_context = false
+show_column_numbers = false
+show_error_codes = true
+
+[[tool.mypy.overrides]]
+module = [
+  "pandas._config.config", # TODO
+  "pandas._libs.*",
+  "pandas._testing.*", # TODO
+  "pandas.compat.numpy.function", # TODO
+  "pandas.core._numba.executor", # TODO
+  "pandas.core.array_algos.masked_reductions", # TODO
+  "pandas.core.array_algos.putmask", # TODO
+  "pandas.core.array_algos.quantile", # TODO
+  "pandas.core.array_algos.replace", # TODO
+  "pandas.core.array_algos.take", # TODO
+  "pandas.core.arrays.*", # TODO
+  "pandas.core.computation.*", # TODO
+  "pandas.core.dtypes.astype", # TODO
+  "pandas.core.dtypes.cast", # TODO
+  "pandas.core.dtypes.common", # TODO
+  "pandas.core.dtypes.concat", # TODO
+  "pandas.core.dtypes.dtypes", # TODO
+  "pandas.core.dtypes.generic", # TODO
+  "pandas.core.dtypes.missing", # TODO
+  "pandas.core.groupby.generic", # TODO
+  "pandas.core.groupby.grouper", # TODO
+  "pandas.core.groupby.groupby", # TODO
+  "pandas.core.groupby.ops", # TODO
+  "pandas.core.indexers.*", # TODO
+  "pandas.core.indexes.*", # TODO
+  "pandas.core.interchange.column", # TODO
+  "pandas.core.interchange.dataframe_protocol", # TODO
+  "pandas.core.interchange.from_dataframe", # TODO
+  "pandas.core.internals.*", # TODO
+  "pandas.core.ops.array_ops", # TODO
+  "pandas.core.ops.common", # TODO
+  "pandas.core.ops.missing", # TODO
+  "pandas.core.reshape.*", # TODO
+  "pandas.core.strings.*", # TODO
+  "pandas.core.tools.*", # TODO
+  "pandas.core.window.common", # TODO
+  "pandas.core.window.ewm", # TODO
+  "pandas.core.window.expanding", # TODO
+  "pandas.core.window.numba_", # TODO
+  "pandas.core.window.online", # TODO
+  "pandas.core.window.rolling", # TODO
+  "pandas.core.accessor", # TODO
+  "pandas.core.algorithms", # TODO
+  "pandas.core.apply", # TODO
+  "pandas.core.arraylike", # TODO
+  "pandas.core.base", # TODO
+  "pandas.core.common", # TODO
+  "pandas.core.construction", # TODO
+  "pandas.core.flags", # TODO
+  "pandas.core.frame", # TODO
+  "pandas.core.generic", # TODO
+  "pandas.core.indexing", # TODO
+  "pandas.core.missing", # TODO
+  "pandas.core.nanops", # TODO
+  "pandas.core.resample", # TODO
+  "pandas.core.roperator", # TODO
+  "pandas.core.sample", # TODO
+  "pandas.core.series", # TODO
+  "pandas.core.sorting", # TODO
+  "pandas.errors", # TODO
+  "pandas.io.clipboard", # TODO
+  "pandas.io.excel._base", # TODO
+  "pandas.io.excel._odfreader", # TODO
+  "pandas.io.excel._openpyxl", # TODO
+  "pandas.io.excel._pyxlsb", # TODO
+  "pandas.io.excel._xlrd", # TODO
+  "pandas.io.excel._xlsxwriter", # TODO
+  "pandas.io.formats.excel", # TODO
+  "pandas.io.formats.format", # TODO
+  "pandas.io.formats.style", # TODO
+  "pandas.io.formats.style_render", # TODO
+  "pandas.io.formats.xml", # TODO
+  "pandas.io.json.*", # TODO
+  "pandas.io.parsers.*", # TODO
+  "pandas.io.sas.sas_xport", # TODO
+  "pandas.io.sas.sas7bdat", # TODO
+  "pandas.io.clipboards", # TODO
+  "pandas.io.html", # TODO
+  "pandas.io.parquet", # TODO
+  "pandas.io.pytables", # TODO
+  "pandas.io.sql", # TODO
+  "pandas.io.xml", # TODO
+  "pandas.plotting.*", # TODO
+  "pandas.tests.*",
+  "pandas.tseries.frequencies", # TODO
+  "pandas.tseries.holiday", # TODO
+  "pandas.util._decorators", # TODO
+  "pandas.util._doctools", # TODO
+  "pandas.util._test_decorators", # TODO
+  "pandas.util._validators", # TODO
+  "pandas.util", # TODO
+  "pandas._version",
+  "pandas.conftest",
+  "pandas"
+]
+disallow_untyped_calls = false
+disallow_untyped_defs = false
+disallow_incomplete_defs = false
+
+[[tool.mypy.overrides]]
+module = [
+  "pandas.tests.*",
+  "pandas._version",
+  "pandas.io.clipboard",
+]
+check_untyped_defs = false
+
+[[tool.mypy.overrides]]
+module = [
+  "pandas.tests.apply.test_series_apply",
+  "pandas.tests.arithmetic.conftest",
+  "pandas.tests.arrays.sparse.test_combine_concat",
+  "pandas.tests.dtypes.test_common",
+  "pandas.tests.frame.methods.test_to_records",
+  "pandas.tests.groupby.test_rank",
+  "pandas.tests.groupby.transform.test_transform",
+  "pandas.tests.indexes.interval.test_interval",
+  "pandas.tests.indexing.test_categorical",
+  "pandas.tests.io.excel.test_writers",
+  "pandas.tests.reductions.test_reductions",
+  "pandas.tests.test_expressions",
+]
+ignore_errors = true
+
+# To be kept consistent with "Import Formatting" section in contributing.rst
+[tool.isort]
+known_pre_libs = "pandas._config"
+known_pre_core = ["pandas._libs", "pandas._typing", "pandas.util._*", "pandas.compat", "pandas.errors"]
+known_dtypes = "pandas.core.dtypes"
+known_post_core = ["pandas.tseries", "pandas.io", "pandas.plotting"]
+sections = ["FUTURE", "STDLIB", "THIRDPARTY" ,"PRE_LIBS" , "PRE_CORE", "DTYPES", "FIRSTPARTY", "POST_CORE", "LOCALFOLDER"]
+profile = "black"
+combine_as_imports = true
+force_grid_wrap = 2
+force_sort_within_sections = true
+skip_glob = "env"
+skip = "pandas/__init__.py"
+
+[tool.pyright]
+pythonVersion = "3.11"
+typeCheckingMode = "basic"
+useLibraryCodeForTypes = false
+include = ["pandas", "typings"]
+exclude = ["pandas/tests", "pandas/io/clipboard", "pandas/util/version", "pandas/core/_numba/extensions.py"]
+# enable subset of "strict"
+reportDuplicateImport = true
+reportInconsistentConstructor = true
+reportInvalidStubStatement = true
+reportOverlappingOverload = true
+reportPropertyTypeMismatch = true
+reportUntypedClassDecorator = true
+reportUntypedFunctionDecorator = true
+reportUntypedNamedTuple = true
+reportUnusedImport = true
+disableBytesTypePromotions = true
+# disable subset of "basic"
+reportArgumentType = false
+reportAssignmentType = false
+reportAttributeAccessIssue = false
+reportCallIssue = false
+reportGeneralTypeIssues = false
+reportIndexIssue = false
+reportMissingModuleSource = false
+reportOperatorIssue = false
+reportOptionalCall = false
+reportOptionalIterable = false
+reportOptionalMemberAccess = false
+reportOptionalOperand = false
+reportOptionalSubscript = false
+reportPrivateImportUsage = false
+reportRedeclaration = false
+reportReturnType = false
+reportUnboundVariable = false
+
+[tool.coverage.run]
+branch = true
+omit = ["pandas/_typing.py", "pandas/_version.py"]
+plugins = ["Cython.Coverage"]
+source = ["pandas"]
+
+[tool.coverage.report]
+ignore_errors = false
+show_missing = true
+omit = ["pandas/_version.py"]
+exclude_lines = [
+  # Have to re-enable the standard pragma
+  "pragma: no cover",
+  # Don't complain about missing debug-only code:s
+  "def __repr__",
+  "if self.debug",
+  # Don't complain if tests don't hit defensive assertion code:
+  "raise AssertionError",
+  "raise NotImplementedError",
+  "AbstractMethodError",
+  # Don't complain if non-runnable code isn't run:
+  "if 0:",
+  "if __name__ == .__main__.:",
+  "if TYPE_CHECKING:",
+]
+
+[tool.coverage.html]
+directory = "coverage_html_report"
+
+[tool.codespell]
+ignore-words-list = "blocs, coo, hist, nd, sav, ser, recuse, nin, timere, expec, expecs, indext, SME, NotIn, tructures, tru, indx, abd, ABD"
+ignore-regex = 'https://([\w/\.])+'
diff --git a/pandas/pyright_reportGeneralTypeIssues.json b/pandas/pyright_reportGeneralTypeIssues.json
new file mode 100644
index 0000000000000..9103d4d533cf3
--- /dev/null
+++ b/pandas/pyright_reportGeneralTypeIssues.json
@@ -0,0 +1,118 @@
+{
+    "typeCheckingMode": "off",
+    "reportArgumentType": true,
+    "reportAssignmentType": true,
+    "reportAttributeAccessIssue": true,
+    "reportCallIssue": true,
+    "reportGeneralTypeIssues": true,
+    "reportIndexIssue": true,
+    "reportOperatorIssue": true,
+    "reportRedeclaration": true,
+    "reportReturnType": true,
+    "useLibraryCodeForTypes": false,
+    "analyzeUnannotatedFunctions": false,
+    "disableBytesTypePromotions": true,
+    "include":
+      [
+        "pandas",
+        "typings"
+      ],
+    "exclude":
+      [
+        "pandas/tests",
+
+        "pandas/io/clipboard",
+        "pandas/util/version",
+
+        "pandas/_testing/__init__.py",
+        "pandas/_testing/_io.py",
+        "pandas/compat/pickle_compat.py",
+        "pandas/core/_numba/extensions.py",
+        "pandas/core/_numba/kernels/sum_.py",
+        "pandas/core/_numba/kernels/var_.py",
+        "pandas/core/algorithms.py",
+        "pandas/core/apply.py",
+        "pandas/core/array_algos/take.py",
+        "pandas/core/arrays/_mixins.py",
+        "pandas/core/arrays/arrow/array.py",
+        "pandas/core/arrays/base.py",
+        "pandas/core/arrays/boolean.py",
+        "pandas/core/arrays/categorical.py",
+        "pandas/core/arrays/datetimelike.py",
+        "pandas/core/arrays/datetimes.py",
+        "pandas/core/arrays/interval.py",
+        "pandas/core/arrays/masked.py",
+        "pandas/core/arrays/period.py",
+        "pandas/core/arrays/sparse/accessor.py",
+        "pandas/core/arrays/sparse/array.py",
+        "pandas/core/arrays/string_.py",
+        "pandas/core/arrays/string_arrow.py",
+        "pandas/core/arrays/timedeltas.py",
+        "pandas/core/computation/align.py",
+        "pandas/core/construction.py",
+        "pandas/core/dtypes/cast.py",
+        "pandas/core/dtypes/common.py",
+        "pandas/core/dtypes/concat.py",
+        "pandas/core/dtypes/dtypes.py",
+        "pandas/core/frame.py",
+        "pandas/core/generic.py",
+        "pandas/core/groupby/generic.py",
+        "pandas/core/groupby/groupby.py",
+        "pandas/core/groupby/grouper.py",
+        "pandas/core/groupby/ops.py",
+        "pandas/core/indexers/utils.py",
+        "pandas/core/indexes/base.py",
+        "pandas/core/indexes/category.py",
+        "pandas/core/indexes/datetimelike.py",
+        "pandas/core/indexes/datetimes.py",
+        "pandas/core/indexes/extension.py",
+        "pandas/core/indexes/interval.py",
+        "pandas/core/indexes/multi.py",
+        "pandas/core/indexes/period.py",
+        "pandas/core/indexing.py",
+        "pandas/core/internals/api.py",
+        "pandas/core/internals/blocks.py",
+        "pandas/core/internals/construction.py",
+        "pandas/core/internals/managers.py",
+        "pandas/core/missing.py",
+        "pandas/core/nanops.py",
+        "pandas/core/ops/array_ops.py",
+        "pandas/core/resample.py",
+        "pandas/core/reshape/concat.py",
+        "pandas/core/reshape/merge.py",
+        "pandas/core/reshape/pivot.py",
+        "pandas/core/reshape/tile.py",
+        "pandas/core/series.py",
+        "pandas/core/sorting.py",
+        "pandas/core/strings/accessor.py",
+        "pandas/core/tools/datetimes.py",
+        "pandas/core/tools/numeric.py",
+        "pandas/core/util/hashing.py",
+        "pandas/core/window/ewm.py",
+        "pandas/core/window/rolling.py",
+        "pandas/io/common.py",
+        "pandas/io/excel/_base.py",
+        "pandas/io/excel/_odfreader.py",
+        "pandas/io/formats/excel.py",
+        "pandas/io/formats/format.py",
+        "pandas/io/formats/info.py",
+        "pandas/io/formats/printing.py",
+        "pandas/io/formats/style.py",
+        "pandas/io/formats/style_render.py",
+        "pandas/io/json/_json.py",
+        "pandas/io/json/_normalize.py",
+        "pandas/io/parsers/arrow_parser_wrapper.py",
+        "pandas/io/parsers/base_parser.py",
+        "pandas/io/parsers/c_parser_wrapper.py",
+        "pandas/io/pytables.py",
+        "pandas/io/sql.py",
+        "pandas/io/stata.py",
+        "pandas/plotting/_matplotlib/boxplot.py",
+        "pandas/plotting/_matplotlib/core.py",
+        "pandas/plotting/_matplotlib/misc.py",
+        "pandas/plotting/_matplotlib/timeseries.py",
+        "pandas/plotting/_matplotlib/tools.py",
+        "pandas/tseries/frequencies.py",
+        "pandas/tseries/holiday.py",
+      ],
+}
diff --git a/pandas/requirements-dev.txt b/pandas/requirements-dev.txt
new file mode 100644
index 0000000000000..fb4d9cdb589ca
--- /dev/null
+++ b/pandas/requirements-dev.txt
@@ -0,0 +1,91 @@
+# This file is auto-generated from environment.yml, do not modify.
+# See that file for comments about the need/usage of each dependency.
+
+pip
+versioneer[toml]
+cython~=3.0.5
+meson[ninja]==1.2.1
+meson-python==0.13.1
+pytest>=7.3.2
+pytest-cov
+pytest-xdist>=3.4.0
+pytest-qt>=4.4.0
+pytest-localserver
+PyQt5>=5.15.9
+coverage
+python-dateutil
+numpy<2
+beautifulsoup4>=4.11.2
+blosc
+bottleneck>=1.3.6
+fastparquet>=2023.10.0
+fsspec>=2022.11.0
+html5lib>=1.1
+hypothesis>=6.84.0
+gcsfs>=2022.11.0
+ipython
+pickleshare
+jinja2>=3.1.2
+lxml>=4.9.2
+matplotlib>=3.6.3
+numba>=0.56.4
+numexpr>=2.8.4
+openpyxl>=3.1.0
+odfpy>=1.4.1
+py
+psycopg2-binary>=2.9.6
+pyarrow>=10.0.1
+pymysql>=1.0.2
+pyreadstat>=1.2.0
+tables>=3.8.0
+python-calamine>=0.1.7
+pytz>=2023.4
+pyxlsb>=1.0.10
+s3fs>=2022.11.0
+scipy>=1.10.0
+SQLAlchemy>=2.0.0
+tabulate>=0.9.0
+xarray>=2022.12.0, <=2024.9.0
+xlrd>=2.0.1
+xlsxwriter>=3.0.5
+zstandard>=0.19.0
+dask
+seaborn
+moto
+flask
+asv>=0.6.1
+flake8==7.1.0
+mypy==1.13.0
+tokenize-rt
+pre-commit>=4.0.1
+gitpython
+gitdb
+google-auth
+natsort
+numpydoc
+pydata-sphinx-theme==0.16
+pytest-cython
+sphinx
+sphinx-design
+sphinx-copybutton
+types-python-dateutil
+types-PyMySQL
+types-pytz
+types-PyYAML
+types-setuptools
+nbconvert>=7.11.0
+nbsphinx
+pandoc
+ipywidgets
+nbformat
+notebook>=7.0.6
+ipykernel
+markdown
+feedparser
+pyyaml
+requests
+pygments
+adbc-driver-postgresql>=0.10.0
+adbc-driver-sqlite>=0.8.0
+typing_extensions; python_version<"3.11"
+tzdata>=2022.7
diff --git a/pandas/scripts/__init__.py b/pandas/scripts/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/scripts/check_for_inconsistent_pandas_namespace.py b/pandas/scripts/check_for_inconsistent_pandas_namespace.py
new file mode 100644
index 0000000000000..ec0a4a408c800
--- /dev/null
+++ b/pandas/scripts/check_for_inconsistent_pandas_namespace.py
@@ -0,0 +1,140 @@
+"""
+Check that test suite file doesn't use the pandas namespace inconsistently.
+
+We check for cases of ``Series`` and ``pd.Series`` appearing in the same file
+(likewise for other pandas objects).
+
+This is meant to be run as a pre-commit hook - to run it manually, you can do:
+
+    pre-commit run inconsistent-namespace-usage --all-files
+
+To automatically fixup a given file, you can pass `--replace`, e.g.
+
+    python scripts/check_for_inconsistent_pandas_namespace.py test_me.py --replace
+
+though note that you may need to manually fixup some imports and that you will also
+need the additional dependency `tokenize-rt` (which is left out from the pre-commit
+hook so that it uses the same virtualenv as the other local ones).
+
+The general structure is similar to that of some plugins from
+https://github.com/asottile/pyupgrade .
+"""
+
+import argparse
+import ast
+from collections.abc import (
+    MutableMapping,
+    Sequence,
+)
+import sys
+from typing import NamedTuple
+
+ERROR_MESSAGE = (
+    "{path}:{lineno}:{col_offset}: "
+    "Found both '{prefix}.{name}' and '{name}' in {path}"
+)
+
+
+class OffsetWithNamespace(NamedTuple):
+    lineno: int
+    col_offset: int
+    namespace: str
+
+
+class Visitor(ast.NodeVisitor):
+    def __init__(self) -> None:
+        self.pandas_namespace: MutableMapping[OffsetWithNamespace, str] = {}
+        self.imported_from_pandas: set[str] = set()
+
+    def visit_Attribute(self, node: ast.Attribute) -> None:
+        if isinstance(node.value, ast.Name) and node.value.id in {"pandas", "pd"}:
+            offset_with_namespace = OffsetWithNamespace(
+                node.lineno, node.col_offset, node.value.id
+            )
+            self.pandas_namespace[offset_with_namespace] = node.attr
+        self.generic_visit(node)
+
+    def visit_ImportFrom(self, node: ast.ImportFrom) -> None:
+        if node.module is not None and "pandas" in node.module:
+            self.imported_from_pandas.update(name.name for name in node.names)
+        self.generic_visit(node)
+
+
+def replace_inconsistent_pandas_namespace(visitor: Visitor, content: str) -> str:
+    from tokenize_rt import (
+        reversed_enumerate,
+        src_to_tokens,
+        tokens_to_src,
+    )
+
+    tokens = src_to_tokens(content)
+    for n, i in reversed_enumerate(tokens):
+        offset_with_namespace = OffsetWithNamespace(i.offset[0], i.offset[1], i.src)
+        if (
+            offset_with_namespace in visitor.pandas_namespace
+            and visitor.pandas_namespace[offset_with_namespace]
+            in visitor.imported_from_pandas
+        ):
+            # Replace `pd`
+            tokens[n] = i._replace(src="")
+            # Replace `.`
+            tokens[n + 1] = tokens[n + 1]._replace(src="")
+
+    new_src: str = tokens_to_src(tokens)
+    return new_src
+
+
+def check_for_inconsistent_pandas_namespace(
+    content: str, path: str, *, replace: bool
+) -> str | None:
+    tree = ast.parse(content)
+
+    visitor = Visitor()
+    visitor.visit(tree)
+
+    inconsistencies = visitor.imported_from_pandas.intersection(
+        visitor.pandas_namespace.values()
+    )
+
+    if not inconsistencies:
+        # No inconsistent namespace usage, nothing to replace.
+        return None
+
+    if not replace:
+        inconsistency = inconsistencies.pop()
+        lineno, col_offset, prefix = next(
+            key for key, val in visitor.pandas_namespace.items() if val == inconsistency
+        )
+        msg = ERROR_MESSAGE.format(
+            lineno=lineno,
+            col_offset=col_offset,
+            prefix=prefix,
+            name=inconsistency,
+            path=path,
+        )
+        sys.stdout.write(msg)
+        sys.exit(1)
+
+    return replace_inconsistent_pandas_namespace(visitor, content)
+
+
+def main(argv: Sequence[str] | None = None) -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("paths", nargs="*")
+    parser.add_argument("--replace", action="store_true")
+    args = parser.parse_args(argv)
+
+    for path in args.paths:
+        with open(path, encoding="utf-8") as fd:
+            content = fd.read()
+        new_content = check_for_inconsistent_pandas_namespace(
+            content, path, replace=args.replace
+        )
+        if not args.replace or new_content is None:
+            continue
+        with open(path, "w", encoding="utf-8") as fd:
+            fd.write(new_content)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pandas/scripts/check_test_naming.py b/pandas/scripts/check_test_naming.py
new file mode 100644
index 0000000000000..f9190643b3246
--- /dev/null
+++ b/pandas/scripts/check_test_naming.py
@@ -0,0 +1,155 @@
+"""
+Check that test names start with `test`, and that test classes start with `Test`.
+
+This is meant to be run as a pre-commit hook - to run it manually, you can do:
+
+    pre-commit run check-test-naming --all-files
+
+NOTE: if this finds a false positive, you can add the comment `# not a test` to the
+class or function definition. Though hopefully that shouldn't be necessary.
+"""
+from __future__ import annotations
+
+import argparse
+import ast
+import os
+from pathlib import Path
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import (
+        Iterator,
+        Sequence,
+    )
+
+PRAGMA = "# not a test"
+
+
+def _find_names(node: ast.Module) -> Iterator[str]:
+    for _node in ast.walk(node):
+        if isinstance(_node, ast.Name):
+            yield _node.id
+        elif isinstance(_node, ast.Attribute):
+            yield _node.attr
+
+
+def _is_fixture(node: ast.expr) -> bool:
+    if isinstance(node, ast.Call):
+        node = node.func
+    return (
+        isinstance(node, ast.Attribute)
+        and node.attr == "fixture"
+        and isinstance(node.value, ast.Name)
+        and node.value.id == "pytest"
+    )
+
+
+def _is_register_dtype(node):
+    return isinstance(node, ast.Name) and node.id == "register_extension_dtype"
+
+
+def is_misnamed_test_func(
+    node: ast.expr | ast.stmt, names: Sequence[str], line: str
+) -> bool:
+    return (
+        isinstance(node, ast.FunctionDef)
+        and not node.name.startswith("test")
+        and names.count(node.name) == 0
+        and not any(_is_fixture(decorator) for decorator in node.decorator_list)
+        and PRAGMA not in line
+        and node.name
+        not in ("teardown_method", "setup_method", "teardown_class", "setup_class")
+    )
+
+
+def is_misnamed_test_class(
+    node: ast.expr | ast.stmt, names: Sequence[str], line: str
+) -> bool:
+    return (
+        isinstance(node, ast.ClassDef)
+        and not node.name.startswith("Test")
+        and names.count(node.name) == 0
+        and not any(_is_register_dtype(decorator) for decorator in node.decorator_list)
+        and PRAGMA not in line
+    )
+
+
+def main(content: str, file: str) -> int:
+    lines = content.splitlines()
+    tree = ast.parse(content)
+    names = list(_find_names(tree))
+    ret = 0
+    for node in tree.body:
+        if is_misnamed_test_func(node, names, lines[node.lineno - 1]):
+            print(
+                f"{file}:{node.lineno}:{node.col_offset} "
+                "found test function which does not start with 'test'"
+            )
+            ret = 1
+        elif is_misnamed_test_class(node, names, lines[node.lineno - 1]):
+            print(
+                f"{file}:{node.lineno}:{node.col_offset} "
+                "found test class which does not start with 'Test'"
+            )
+            ret = 1
+        if (
+            isinstance(node, ast.ClassDef)
+            and names.count(node.name) == 0
+            and not any(
+                _is_register_dtype(decorator) for decorator in node.decorator_list
+            )
+            and PRAGMA not in lines[node.lineno - 1]
+        ):
+            for _node in node.body:
+                if is_misnamed_test_func(_node, names, lines[_node.lineno - 1]):
+                    # It could be that this function is used somewhere by the
+                    # parent class. For example, there might be a base class
+                    # with
+                    #
+                    # class Foo:
+                    #     def foo(self):
+                    #          assert 1+1==2
+                    #     def test_foo(self):
+                    #         self.foo()
+                    #
+                    # and then some subclass overwrites `foo`. So, we check that
+                    # `self.foo` doesn't appear in any of the test classes.
+                    # Note some false negatives might get through, but that's OK.
+                    # This is good enough that has helped identify several examples
+                    # of tests not being run.
+                    assert isinstance(_node, ast.FunctionDef)  # help mypy
+                    should_continue = False
+                    for _file in (Path("pandas") / "tests").rglob("*.py"):
+                        with open(os.path.join(_file), encoding="utf-8") as fd:
+                            _content = fd.read()
+                        if f"self.{_node.name}" in _content:
+                            should_continue = True
+                            break
+                    if should_continue:
+                        continue
+
+                    print(
+                        f"{file}:{_node.lineno}:{_node.col_offset} "
+                        "found test function which does not start with 'test'"
+                    )
+                    ret = 1
+    return ret
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("paths", nargs="*")
+    args = parser.parse_args()
+
+    ret = 0
+
+    for file in args.paths:
+        filename = os.path.basename(file)
+        if not (filename.startswith("test") and filename.endswith(".py")):
+            continue
+        with open(file, encoding="utf-8") as fd:
+            content = fd.read()
+        ret |= main(content, file)
+
+    sys.exit(ret)
diff --git a/pandas/scripts/cibw_before_build.sh b/pandas/scripts/cibw_before_build.sh
new file mode 100644
index 0000000000000..d326dd3637314
--- /dev/null
+++ b/pandas/scripts/cibw_before_build.sh
@@ -0,0 +1,13 @@
+#!/bin/bash
+# Add 3rd party licenses, like numpy does
+for file in $PACKAGE_DIR/LICENSES/*; do
+  cat $file >> $PACKAGE_DIR/LICENSE
+done
+
+# TODO: Delete when there's a PyPI Cython release that supports free-threaded Python 3.13.
+FREE_THREADED_BUILD="$(python -c"import sysconfig; print(bool(sysconfig.get_config_var('Py_GIL_DISABLED')))")"
+if [[ $FREE_THREADED_BUILD == "True" ]]; then
+    python -m pip install -U pip
+    python -m pip install -i https://pypi.anaconda.org/scientific-python-nightly-wheels/simple cython
+    python -m pip install numpy ninja meson-python versioneer[toml]
+fi
diff --git a/pandas/scripts/cibw_before_build_windows.sh b/pandas/scripts/cibw_before_build_windows.sh
new file mode 100644
index 0000000000000..f9e1e68d8efba
--- /dev/null
+++ b/pandas/scripts/cibw_before_build_windows.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+# Add 3rd party licenses, like numpy does
+for file in $PACKAGE_DIR/LICENSES/*; do
+  cat $file >> $PACKAGE_DIR/LICENSE
+done
+
+# TODO: Delete when there's a PyPI Cython release that supports free-threaded Python 3.13
+# and a NumPy Windows wheel for the free-threaded build on PyPI.
+FREE_THREADED_BUILD="$(python -c"import sysconfig; print(bool(sysconfig.get_config_var('Py_GIL_DISABLED')))")"
+if [[ $FREE_THREADED_BUILD == "True" ]]; then
+    python -m pip install -U pip
+    python -m pip install -i https://pypi.anaconda.org/scientific-python-nightly-wheels/simple numpy cython
+    python -m pip install ninja meson-python versioneer[toml]
+fi
diff --git a/pandas/scripts/cibw_before_test_windows.sh b/pandas/scripts/cibw_before_test_windows.sh
new file mode 100644
index 0000000000000..8878e3950452f
--- /dev/null
+++ b/pandas/scripts/cibw_before_test_windows.sh
@@ -0,0 +1,6 @@
+#!/bin/bash
+# TODO: Delete when there's a NumPy Windows wheel for the free-threaded build on PyPI.
+FREE_THREADED_BUILD="$(python -c"import sysconfig; print(bool(sysconfig.get_config_var('Py_GIL_DISABLED')))")"
+if [[ $FREE_THREADED_BUILD == "True" ]]; then
+    python -m pip install -i https://pypi.anaconda.org/scientific-python-nightly-wheels/simple numpy
+fi
diff --git a/pandas/scripts/download_wheels.sh b/pandas/scripts/download_wheels.sh
new file mode 100755
index 0000000000000..3dcae0cadcdcf
--- /dev/null
+++ b/pandas/scripts/download_wheels.sh
@@ -0,0 +1,30 @@
+#!/bin/bash
+#
+# Download all wheels for a pandas version.
+#
+# This script is mostly useful during the release process, when wheels
+# generated by the MacPython repo need to be downloaded locally to then
+# be uploaded to the PyPI.
+#
+# There is no API to access the wheel files, so the script downloads the
+# website, extracts the file urls from the html, and then downloads it
+# one by one to the dist/ directory where they would be generated.
+
+VERSION=$1
+BASE_DIR=$(dirname -- $0)
+mkdir -p $BASE_DIR/../dist
+DIST_DIR="$(realpath $BASE_DIR/../dist)"
+
+if [ -z "$VERSION" ]; then
+    printf "Usage:\n\t%s <version>\n\nWhere <version> is for example 1.5.3"  "$0"
+    exit 1
+fi
+
+curl "https://anaconda.org/multibuild-wheels-staging/pandas/files?version=${VERSION}" | \
+    grep "href=\"/multibuild-wheels-staging/pandas/${VERSION}" | \
+    sed -r 's/.*<a href="([^"]+\.(whl|tar.gz))">.*/\1/g' | \
+    awk '{print "https://anaconda.org" $0 }' | \
+    xargs wget -P "$DIST_DIR"
+
+printf '\nWheels downloaded to %s\nYou can upload them to PyPI using:\n\n' "$DIST_DIR"
+printf "\ttwine upload %s/pandas-%s*.{whl,tar.gz} --skip-existing" "$DIST_DIR" "$VERSION"
diff --git a/pandas/scripts/generate_pip_deps_from_conda.py b/pandas/scripts/generate_pip_deps_from_conda.py
new file mode 100755
index 0000000000000..a57876902ad36
--- /dev/null
+++ b/pandas/scripts/generate_pip_deps_from_conda.py
@@ -0,0 +1,156 @@
+#!/usr/bin/env python3
+"""
+Convert the conda environment.yml to the pip requirements-dev.txt,
+or check that they have the same packages (for the CI)
+
+Usage:
+
+    Generate `requirements-dev.txt`
+    $ python scripts/generate_pip_deps_from_conda.py
+
+    Compare and fail (exit status != 0) if `requirements-dev.txt` has not been
+    generated with this script:
+    $ python scripts/generate_pip_deps_from_conda.py --compare
+"""
+import argparse
+import pathlib
+import re
+import sys
+
+if sys.version_info >= (3, 11):
+    import tomllib
+else:
+    import tomli as tomllib
+import yaml
+
+EXCLUDE = {"python", "c-compiler", "cxx-compiler"}
+REMAP_VERSION = {"tzdata": "2022.7"}
+CONDA_TO_PIP = {
+    "versioneer": "versioneer[toml]",
+    "meson": "meson[ninja]",
+    "pytables": "tables",
+    "psycopg2": "psycopg2-binary",
+    "dask-core": "dask",
+    "seaborn-base": "seaborn",
+    "sqlalchemy": "SQLAlchemy",
+    "pyqt": "PyQt5",
+}
+
+
+def conda_package_to_pip(package: str):
+    """
+    Convert a conda package to its pip equivalent.
+
+    In most cases they are the same, those are the exceptions:
+    - Packages that should be excluded (in `EXCLUDE`)
+    - Packages that should be renamed (in `CONDA_TO_PIP`)
+    - A package requiring a specific version, in conda is defined with a single
+      equal (e.g. ``pandas=1.0``) and in pip with two (e.g. ``pandas==1.0``)
+    """
+    package = re.sub("(?<=[^<>~])=", "==", package).strip()
+
+    for compare in ("<=", ">=", "=="):
+        if compare in package:
+            pkg, version = package.split(compare)
+            if pkg in EXCLUDE:
+                return
+            if pkg in REMAP_VERSION:
+                return "".join((pkg, compare, REMAP_VERSION[pkg]))
+            if pkg in CONDA_TO_PIP:
+                return "".join((CONDA_TO_PIP[pkg], compare, version))
+
+    if package in EXCLUDE:
+        return
+
+    if package in CONDA_TO_PIP:
+        return CONDA_TO_PIP[package]
+
+    return package
+
+
+def generate_pip_from_conda(
+    conda_path: pathlib.Path, pip_path: pathlib.Path, compare: bool = False
+) -> bool:
+    """
+    Generate the pip dependencies file from the conda file, or compare that
+    they are synchronized (``compare=True``).
+
+    Parameters
+    ----------
+    conda_path : pathlib.Path
+        Path to the conda file with dependencies (e.g. `environment.yml`).
+    pip_path : pathlib.Path
+        Path to the pip file with dependencies (e.g. `requirements-dev.txt`).
+    compare : bool, default False
+        Whether to generate the pip file (``False``) or to compare if the
+        pip file has been generated with this script and the last version
+        of the conda file (``True``).
+
+    Returns
+    -------
+    bool
+        True if the comparison fails, False otherwise
+    """
+    with conda_path.open() as file:
+        deps = yaml.safe_load(file)["dependencies"]
+
+    pip_deps = []
+    for dep in deps:
+        if isinstance(dep, str):
+            conda_dep = conda_package_to_pip(dep)
+            if conda_dep:
+                pip_deps.append(conda_dep)
+        elif isinstance(dep, dict) and len(dep) == 1 and "pip" in dep:
+            pip_deps.extend(dep["pip"])
+        else:
+            raise ValueError(f"Unexpected dependency {dep}")
+
+    header = (
+        f"# This file is auto-generated from {conda_path.name}, do not modify.\n"
+        "# See that file for comments about the need/usage of each dependency.\n\n"
+    )
+    pip_content = header + "\n".join(pip_deps) + "\n"
+
+    # add setuptools to requirements-dev.txt
+    with open(pathlib.Path(conda_path.parent, "pyproject.toml"), "rb") as fd:
+        meta = tomllib.load(fd)
+    for requirement in meta["build-system"]["requires"]:
+        if "setuptools" in requirement:
+            pip_content += requirement
+            pip_content += "\n"
+
+    if compare:
+        with pip_path.open() as file:
+            return pip_content != file.read()
+
+    with pip_path.open("w") as file:
+        file.write(pip_content)
+    return False
+
+
+if __name__ == "__main__":
+    argparser = argparse.ArgumentParser(
+        description="convert (or compare) conda file to pip"
+    )
+    argparser.add_argument(
+        "--compare",
+        action="store_true",
+        help="compare whether the two files are equivalent",
+    )
+    args = argparser.parse_args()
+
+    conda_fname = "environment.yml"
+    pip_fname = "requirements-dev.txt"
+    repo_path = pathlib.Path(__file__).parent.parent.absolute()
+    res = generate_pip_from_conda(
+        pathlib.Path(repo_path, conda_fname),
+        pathlib.Path(repo_path, pip_fname),
+        compare=args.compare,
+    )
+    if res:
+        msg = (
+            f"`{pip_fname}` has to be generated with `{__file__}` after "
+            f"`{conda_fname}` is modified.\n"
+        )
+        sys.stderr.write(msg)
+    sys.exit(res)
diff --git a/pandas/scripts/pandas_errors_documented.py b/pandas/scripts/pandas_errors_documented.py
new file mode 100644
index 0000000000000..b68da137717de
--- /dev/null
+++ b/pandas/scripts/pandas_errors_documented.py
@@ -0,0 +1,55 @@
+"""
+Check that doc/source/reference/testing.rst documents
+all exceptions and warnings in pandas/errors/__init__.py.
+
+This is meant to be run as a pre-commit hook - to run it manually, you can do:
+
+    pre-commit run pandas-errors-documented --all-files
+"""
+from __future__ import annotations
+
+import argparse
+import ast
+import pathlib
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+API_PATH = pathlib.Path("doc/source/reference/testing.rst").resolve()
+
+
+def get_defined_errors(content: str) -> set[str]:
+    errors = set()
+    for node in ast.walk(ast.parse(content)):
+        if isinstance(node, ast.ClassDef):
+            errors.add(node.name)
+        elif isinstance(node, ast.ImportFrom) and node.module != "__future__":
+            for alias in node.names:
+                errors.add(alias.name)
+    return errors
+
+
+def main(argv: Sequence[str] | None = None) -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("path")
+    args = parser.parse_args(argv)
+    with open(args.path, encoding="utf-8") as f:
+        file_errors = get_defined_errors(f.read())
+    with open(API_PATH, encoding="utf-8") as f:
+        doc_errors = {
+            line.split(".")[1].strip() for line in f.readlines() if "errors" in line
+        }
+    missing = file_errors.difference(doc_errors)
+    if missing:
+        sys.stdout.write(
+            f"The following exceptions and/or warnings are not documented "
+            f"in {API_PATH}: {missing}"
+        )
+        sys.exit(1)
+    sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pandas/scripts/run_stubtest.py b/pandas/scripts/run_stubtest.py
new file mode 100644
index 0000000000000..df88c61061f12
--- /dev/null
+++ b/pandas/scripts/run_stubtest.py
@@ -0,0 +1,100 @@
+import os
+from pathlib import Path
+import sys
+import tempfile
+import warnings
+
+from mypy import stubtest
+
+import pandas as pd
+
+pd_version = getattr(pd, "__version__", "")
+
+# fail early if pandas is not installed
+if not pd_version:
+    # fail on the CI, soft fail during local development
+    warnings.warn("You need to install the development version of pandas")
+    if pd.compat.is_ci_environment():
+        sys.exit(1)
+    else:
+        sys.exit(0)
+
+# GH 48260
+if "dev" not in pd_version:
+    warnings.warn(
+        f"stubtest may fail as {pd_version} is not a dev version. "
+        f"Please install a pandas dev version or see https://pandas.pydata.org/"
+        f"pandas-docs/stable/development/contributing_codebase.html"
+        f"#validating-type-hints on how to skip the stubtest"
+    )
+
+
+_ALLOWLIST = [  # should be empty
+    # TODO (child classes implement these methods)
+    "pandas._libs.hashtable.HashTable.__contains__",
+    "pandas._libs.hashtable.HashTable.__len__",
+    "pandas._libs.hashtable.HashTable.factorize",
+    "pandas._libs.hashtable.HashTable.get_item",
+    "pandas._libs.hashtable.HashTable.get_labels",
+    "pandas._libs.hashtable.HashTable.get_na",
+    "pandas._libs.hashtable.HashTable.get_state",
+    "pandas._libs.hashtable.HashTable.lookup",
+    "pandas._libs.hashtable.HashTable.map_locations",
+    "pandas._libs.hashtable.HashTable.set_item",
+    "pandas._libs.hashtable.HashTable.set_na",
+    "pandas._libs.hashtable.HashTable.sizeof",
+    "pandas._libs.hashtable.HashTable.unique",
+    "pandas._libs.hashtable.HashTable.hash_inner_join",
+    # stubtest might be too sensitive
+    "pandas._libs.lib.NoDefault",
+    "pandas._libs.lib._NoDefault.no_default",
+    # stubtest/Cython is not recognizing the default value for the dtype parameter
+    "pandas._libs.lib.map_infer_mask",
+    # internal type alias (should probably be private)
+    "pandas._libs.lib.ndarray_obj_2d",
+    # runtime argument "owner" has a default value but stub argument does not
+    "pandas._libs.properties.AxisProperty.__get__",
+    "pandas._libs.properties.cache_readonly.deleter",
+    "pandas._libs.properties.cache_readonly.getter",
+    "pandas._libs.properties.cache_readonly.setter",
+    # TODO (child classes implement these methods)
+    "pandas._libs.sparse.SparseIndex.__init__",
+    "pandas._libs.sparse.SparseIndex.equals",
+    "pandas._libs.sparse.SparseIndex.indices",
+    "pandas._libs.sparse.SparseIndex.intersect",
+    "pandas._libs.sparse.SparseIndex.lookup",
+    "pandas._libs.sparse.SparseIndex.lookup_array",
+    "pandas._libs.sparse.SparseIndex.make_union",
+    "pandas._libs.sparse.SparseIndex.nbytes",
+    "pandas._libs.sparse.SparseIndex.ngaps",
+    "pandas._libs.sparse.SparseIndex.to_block_index",
+    "pandas._libs.sparse.SparseIndex.to_int_index",
+    # TODO (decorator changes argument names)
+    "pandas._libs.tslibs.offsets.BusinessHour.rollback",
+    "pandas._libs.tslibs.offsets.BusinessHour.rollforward ",
+    # type alias
+    "pandas._libs.tslibs.timedeltas.UnitChoices",
+]
+
+if __name__ == "__main__":
+    # find pyi files
+    root = Path.cwd()
+    pyi_modules = [
+        str(pyi.relative_to(root).with_suffix("")).replace(os.sep, ".")
+        for pyi in root.glob("pandas/**/*.pyi")
+    ]
+
+    # create allowlist
+    with tempfile.NamedTemporaryFile(mode="w+t") as allow:
+        allow.write("\n".join(_ALLOWLIST))
+        allow.flush()
+
+        args = pyi_modules + [
+            "--ignore-missing-stub",
+            "--concise",
+            "--mypy-config-file",
+            "pyproject.toml",
+            "--allowlist",
+            allow.name,
+        ]
+        sys.exit(stubtest.test_stubs(stubtest.parse_options(args)))
diff --git a/pandas/scripts/run_vulture.py b/pandas/scripts/run_vulture.py
new file mode 100644
index 0000000000000..edf0a1588305c
--- /dev/null
+++ b/pandas/scripts/run_vulture.py
@@ -0,0 +1,21 @@
+"""Look for unreachable code."""
+
+import argparse
+import sys
+
+from vulture import Vulture
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("files", nargs="*")
+    args = parser.parse_args()
+
+    v = Vulture()
+    v.scavenge(args.files)
+    ret = 0
+    for item in v.get_unused_code(min_confidence=100):
+        if item.typ == "unreachable_code":
+            print(item.get_report())
+            ret = 1
+
+    sys.exit(ret)
diff --git a/pandas/scripts/sort_whatsnew_note.py b/pandas/scripts/sort_whatsnew_note.py
new file mode 100644
index 0000000000000..428ffca83ea26
--- /dev/null
+++ b/pandas/scripts/sort_whatsnew_note.py
@@ -0,0 +1,81 @@
+"""
+Sort whatsnew note blocks by issue number.
+
+NOTE: this assumes that each entry is on its own line, and ends with an issue number.
+If that's not the case, then an entry might not get sorted. However, virtually all
+recent-enough whatsnew entries follow this pattern. So, although not perfect, this
+script should be good enough to significantly reduce merge conflicts.
+
+For example:
+
+- Fixed bug in resample (:issue:`321`)
+- Fixed bug in groupby (:issue:`123`)
+
+would become
+
+- Fixed bug in groupby (:issue:`123`)
+- Fixed bug in resample (:issue:`321`)
+
+The motivation is to reduce merge conflicts by reducing the chances that multiple
+contributors will edit the same line of code.
+
+You can run this manually with
+
+    pre-commit run sort-whatsnew-items --all-files
+"""
+from __future__ import annotations
+
+import argparse
+import re
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+
+# Check line starts with `-` and ends with e.g. `(:issue:`12345`)`,
+# possibly with a trailing full stop.
+pattern = re.compile(r"-.*\(:issue:`(\d+)`\)\.?$")
+
+
+def sort_whatsnew_note(content: str) -> int:
+    new_lines = []
+    block: list[str] = []
+    lines = content.splitlines(keepends=True)
+    for line in lines:
+        if line.startswith("- ") and pattern.search(line) is not None:
+            block.append(line)
+        else:
+            block = sorted(block)
+            new_lines.extend(block)
+            new_lines.append(line)
+            block = []
+    if sorted(new_lines) != sorted(lines):  # pragma: no cover
+        # Defensive check - this script should only reorder lines, not modify any
+        # content.
+        raise AssertionError(
+            "Script modified content of file. Something is wrong, please don't "
+            "trust it."
+        )
+    return "".join(new_lines)
+
+
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("paths", nargs="*")
+    args = parser.parse_args(argv)
+    ret = 0
+    for path in args.paths:
+        with open(path, encoding="utf-8") as fd:
+            content = fd.read()
+        new_content = sort_whatsnew_note(content)
+        if content != new_content:
+            ret |= 1
+            with open(path, "w", encoding="utf-8") as fd:
+                fd.write(new_content)
+    return ret
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/pandas/scripts/tests/__init__.py b/pandas/scripts/tests/__init__.py
new file mode 100644
index 0000000000000..e69de29bb2d1d
diff --git a/pandas/scripts/tests/conftest.py b/pandas/scripts/tests/conftest.py
new file mode 100644
index 0000000000000..d43741f925eb5
--- /dev/null
+++ b/pandas/scripts/tests/conftest.py
@@ -0,0 +1,8 @@
+# pyproject.toml defines addopts: --strict-data-files
+# strict-data-files is defined & used in pandas/conftest.py
+def pytest_addoption(parser) -> None:
+    parser.addoption(
+        "--strict-data-files",
+        action="store_true",
+        help="Unused",
+    )
diff --git a/pandas/scripts/tests/data/deps_expected_duplicate_package.yaml b/pandas/scripts/tests/data/deps_expected_duplicate_package.yaml
new file mode 100644
index 0000000000000..72721c2842707
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_expected_duplicate_package.yaml
@@ -0,0 +1,4 @@
+# Test: duplicate package
+dependencies:
+  - jinja2>=3.0.0
+  - jinja2>=3.0.0
diff --git a/pandas/scripts/tests/data/deps_expected_no_version.yaml b/pandas/scripts/tests/data/deps_expected_no_version.yaml
new file mode 100644
index 0000000000000..843e48330a928
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_expected_no_version.yaml
@@ -0,0 +1,5 @@
+# Test: empty version
+dependencies:
+  - jinja2>=3.0.0
+  - scipy>=1.7.1
+  - SQLAlchemy>=1.4.16
diff --git a/pandas/scripts/tests/data/deps_expected_random.yaml b/pandas/scripts/tests/data/deps_expected_random.yaml
new file mode 100644
index 0000000000000..d1db7989a95a4
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_expected_random.yaml
@@ -0,0 +1,54 @@
+# Test: random
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.8
+
+  # build dependencies
+  - versioneer[toml]
+  - cython>=0.29.32
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - psutil
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+  - pytz
+
+  # optional dependencies
+  - beautifulsoup4>=5.9.3
+  - blosc
+  - bottleneck>=1.3.2
+  - fastparquet>=0.6.3
+  - fsspec>=2021.07.0
+  - html5lib>=1.1
+  - hypothesis>=6.34.2
+  - gcsfs>=2021.07.0
+  - jinja2>=3.0.0
+  - lxml>=4.6.3
+  - matplotlib>=3.6.1
+  - numba>=0.53.1
+  - numexpr>=2.7.3
+  - openpyxl>=3.0.7
+  - odfpy>=1.4.1
+  - psycopg2>=2.8.6
+  - pyarrow<11, >=7.0.0
+  - pymysql>=1.0.2
+  - pyreadstat>=1.1.2
+  - pytables>=3.6.1
+  - python-calamine>=0.1.7
+  - pyxlsb>=1.0.8
+  - s3fs>=2021.08.0
+  - scipy>=1.7.1
+  - sqlalchemy>=1.4.16
+  - tabulate>=0.8.9
+  - xarray>=0.21.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=1.4.3
+  - zstandard>=0.15.2
diff --git a/pandas/scripts/tests/data/deps_expected_range.yaml b/pandas/scripts/tests/data/deps_expected_range.yaml
new file mode 100644
index 0000000000000..c8e25076ef3b0
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_expected_range.yaml
@@ -0,0 +1,5 @@
+# Test: range
+dependencies:
+  - jinja2<8, >=3.0.0
+  - scipy<9, >=1.7.1
+  - SQLAlchemy<2.0, >=1.4.16
diff --git a/pandas/scripts/tests/data/deps_expected_same_version.yaml b/pandas/scripts/tests/data/deps_expected_same_version.yaml
new file mode 100644
index 0000000000000..e07b221ecd44f
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_expected_same_version.yaml
@@ -0,0 +1,3 @@
+# Test: same version
+dependencies:
+  - jinja2>=3.0.0
diff --git a/pandas/scripts/tests/data/deps_minimum.toml b/pandas/scripts/tests/data/deps_minimum.toml
new file mode 100644
index 0000000000000..0a53225a5d995
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_minimum.toml
@@ -0,0 +1,410 @@
+[build-system]
+# Minimum requirements for the build system to execute.
+# See https://github.com/scipy/scipy/pull/12940 for the AIX issue.
+requires = [
+    "setuptools>=61.0.0",
+    "wheel",
+    "Cython>=0.29.32,<3",  # Note: sync with setup.py, environment.yml and asv.conf.json
+    "oldest-supported-numpy>=2022.8.16",
+    "versioneer[toml]"
+]
+# build-backend = "setuptools.build_meta"
+
+[project]
+name = 'pandas'
+dynamic = [
+  'version'
+]
+description = 'Powerful data structures for data analysis, time series, and statistics'
+readme = 'README.md'
+authors = [
+  { name = 'The Pandas Development Team', email='pandas-dev@python.org' },
+]
+license = {file = 'LICENSE'}
+requires-python = '>=3.8'
+dependencies = [
+  "numpy>=1.20.3; python_version<'3.10'",
+  "numpy>=1.21.0; python_version>='3.10'",
+  "numpy>=1.23.2; python_version>='3.11'",
+  "python-dateutil>=2.8.2",
+  "pytz>=2020.1"
+]
+classifiers = [
+    'Development Status :: 5 - Production/Stable',
+    'Environment :: Console',
+    'Intended Audience :: Science/Research',
+    'License :: OSI Approved :: BSD License',
+    'Operating System :: OS Independent',
+    'Programming Language :: Cython',
+    'Programming Language :: Python',
+    'Programming Language :: Python :: 3',
+    'Programming Language :: Python :: 3 :: Only',
+    'Programming Language :: Python :: 3.10',
+    'Programming Language :: Python :: 3.11',
+    'Topic :: Scientific/Engineering'
+]
+
+[project.urls]
+homepage = 'https://pandas.pydata.org'
+documentation = 'https://pandas.pydata.org/docs/'
+repository = 'https://github.com/pandas-dev/pandas'
+
+[project.entry-points."pandas_plotting_backends"]
+matplotlib = "pandas:plotting._matplotlib"
+
+[project.optional-dependencies]
+test = ['hypothesis>=6.34.2', 'pytest>=7.3.2', 'pytest-xdist>=3.4.0']
+performance = ['bottleneck>=1.3.2', 'numba>=0.53.1', 'numexpr>=2.7.1']
+timezone = ['tzdata>=2022.1']
+computation = ['scipy>=1.7.1', 'xarray>=0.21.0']
+fss = ['fsspec>=2021.07.0']
+aws = ['s3fs>=2021.08.0']
+gcp = ['gcsfs>=2021.07.0']
+excel = ['odfpy>=1.4.1', 'openpyxl>=3.0.7', 'python-calamine>=0.1.7', 'pyxlsb>=1.0.8', 'xlrd>=2.0.1', 'xlsxwriter>=1.4.3']
+parquet = ['pyarrow>=7.0.0']
+feather = ['pyarrow>=7.0.0']
+hdf5 = [# blosc only available on conda (https://github.com/Blosc/python-blosc/issues/297)
+        #'blosc>=1.20.1',
+        'tables>=3.6.1']
+spss = ['pyreadstat>=1.1.2']
+postgresql = ['SQLAlchemy>=1.4.16', 'psycopg2>=2.8.6']
+mysql = ['SQLAlchemy>=1.4.16', 'pymysql>=1.0.2']
+sql-other = ['SQLAlchemy>=1.4.16']
+html = ['beautifulsoup4>=4.9.3', 'html5lib>=1.1', 'lxml>=4.6.3']
+xml = ['lxml>=4.6.3']
+plot = ['matplotlib>=3.6.1']
+output_formatting = ['jinja2>=3.0.0', 'tabulate>=0.8.9']
+clipboard = ['PyQt5>=5.15.1', 'qtpy>=2.3.0']
+compression = ['zstandard>=0.15.2']
+all = ['beautifulsoup4>=5.9.3',
+       # blosc only available on conda (https://github.com/Blosc/python-blosc/issues/297)
+       #'blosc>=1.21.0',
+       'bottleneck>=1.3.2',
+       'fastparquet>=0.6.3',
+       'fsspec>=2021.07.0',
+       'gcsfs>=2021.07.0',
+       'html5lib>=1.1',
+       'hypothesis>=6.34.2',
+       'jinja2>=3.0.0',
+       'lxml>=4.6.3',
+       'matplotlib>=3.6.1',
+       'numba>=0.53.1',
+       'numexpr>=2.7.3',
+       'odfpy>=1.4.1',
+       'openpyxl>=3.0.7',
+       'psycopg2>=2.8.6',
+       'pyarrow>=7.0.0',
+       'pymysql>=1.0.2',
+       'PyQt5>=5.15.1',
+       'pyreadstat>=1.1.2',
+       'pytest>=7.3.2',
+       'pytest-xdist>=3.4.0',
+       'python-calamine>=0.1.7',
+       'pyxlsb>=1.0.8',
+       'qtpy>=2.3.0',
+       'scipy>=1.7.1',
+       's3fs>=2021.08.0',
+       'SQLAlchemy>=1.4.16',
+       'tables>=3.6.1',
+       'tabulate>=0.8.9',
+       'tzdata>=2022.1',
+       'xarray>=0.21.0',
+       'xlrd>=2.0.1',
+       'xlsxwriter>=1.4.3',
+       'zstandard>=0.15.2']
+
+# TODO: Remove after setuptools support is dropped.
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+include = ["pandas", "pandas.*"]
+namespaces = false
+
+[tool.setuptools.exclude-package-data]
+"*" = ["*.c", "*.h"]
+
+# See the docstring in versioneer.py for instructions. Note that you must
+# re-run 'versioneer.py setup' after changing this section, and commit the
+# resulting files.
+[tool.versioneer]
+VCS = "git"
+style = "pep440"
+versionfile_source = "pandas/_version.py"
+versionfile_build = "pandas/_version.py"
+tag_prefix = "v"
+parentdir_prefix = "pandas-"
+
+[tool.cibuildwheel]
+skip = "cp36-* cp37-* pp37-* *-manylinux_i686 *_ppc64le *_s390x *-musllinux*"
+build-verbosity = "3"
+test-requires = "hypothesis>=6.34.2 pytest>=7.3.2 pytest-xdist>=3.4.0"
+test-command = "python {project}/ci/test_wheels.py"
+
+[tool.cibuildwheel.macos]
+archs = "x86_64 arm64"
+test-skip = "*_arm64"
+
+[tool.cibuildwheel.windows]
+repair-wheel-command = "python ci/fix_wheels.py {wheel} {dest_dir}"
+
+[[tool.cibuildwheel.overrides]]
+select = "*-win*"
+# We test separately for Windows, since we use
+# the base windows docker image to check if any dlls are
+# missing from the wheel
+test-command = ""
+
+[[tool.cibuildwheel.overrides]]
+select = "*-win32"
+environment = { IS_32_BIT="true" }
+
+[tool.ruff]
+line-length = 88
+update-check = false
+target-version = "py38"
+
+select = [
+  # pyflakes
+  "F",
+  # pycodestyle
+  "E",
+  "W",
+  # flake8-2020
+  "YTT",
+  # flake8-bugbear
+  "B",
+  # flake8-quotes
+  "Q",
+  # pylint
+  "PLE", "PLR", "PLW",
+]
+
+ignore = [
+  # space before : (needed for how black formats slicing)
+  # "E203",  # not yet implemented
+  # module level import not at top of file
+  "E402",
+  # do not assign a lambda expression, use a def
+  "E731",
+  # line break before binary operator
+  # "W503",  # not yet implemented
+  # line break after binary operator
+  # "W504",  # not yet implemented
+  # controversial
+  "B006",
+  # controversial
+  "B007",
+  # controversial
+  "B008",
+  # setattr is used to side-step mypy
+  "B009",
+  # getattr is used to side-step mypy
+  "B010",
+  # tests use assert False
+  "B011",
+  # tests use comparisons but not their returned value
+  "B015",
+  # false positives
+  "B019",
+  # Loop control variable overrides iterable it iterates
+  "B020",
+  # Function definition does not bind loop variable
+  "B023",
+  # Functions defined inside a loop must not use variables redefined in the loop
+  # "B301",  # not yet implemented
+
+  # Additional checks that don't pass yet
+  # Within an except clause, raise exceptions with ...
+  "B904",
+]
+
+exclude = [
+  "doc/sphinxext/*.py",
+  "doc/build/*.py",
+  "doc/temp/*.py",
+  ".eggs/*.py",
+  "versioneer.py",
+  # exclude asv benchmark environments from linting
+  "env",
+]
+
+[tool.pytest.ini_options]
+# sync minversion with pyproject.toml & install.rst
+minversion = "7.0"
+addopts = "--strict-data-files --strict-markers --strict-config --capture=no --durations=30 --junitxml=test-data.xml"
+empty_parameter_set_mark = "fail_at_collect"
+xfail_strict = true
+testpaths = "pandas"
+doctest_optionflags = [
+  "NORMALIZE_WHITESPACE",
+  "IGNORE_EXCEPTION_DETAIL",
+  "ELLIPSIS",
+]
+filterwarnings = [
+  # Will be fixed in numba 0.56: https://github.com/numba/numba/issues/7758
+  "ignore:`np.MachAr` is deprecated:DeprecationWarning:numba",
+  "ignore:.*urllib3:DeprecationWarning:botocore",
+  "ignore:Setuptools is replacing distutils.:UserWarning:_distutils_hack",
+  # https://github.com/PyTables/PyTables/issues/822
+  "ignore:a closed node found in the registry:UserWarning:tables",
+  "ignore:`np.object` is a deprecated:DeprecationWarning:tables",
+  "ignore:tostring:DeprecationWarning:tables",
+  "ignore:distutils Version classes are deprecated:DeprecationWarning:numexpr",
+  "ignore:distutils Version classes are deprecated:DeprecationWarning:fastparquet",
+  "ignore:distutils Version classes are deprecated:DeprecationWarning:fsspec",
+]
+junit_family = "xunit2"
+markers = [
+  "single_cpu: tests that should run on a single cpu only",
+  "slow: mark a test as slow",
+  "network: mark a test as network",
+  "db: tests requiring a database (mysql or postgres)",
+  "clipboard: mark a pd.read_clipboard test",
+  "arm_slow: mark a test as slow for arm64 architecture",
+  "skip_ubsan: tests known to invoke undefined behavior",
+]
+
+[tool.mypy]
+# Import discovery
+mypy_path = "typings"
+files = ["pandas", "typings"]
+namespace_packages = false
+explicit_package_bases = false
+ignore_missing_imports = true
+follow_imports = "normal"
+follow_imports_for_stubs = false
+no_site_packages = false
+no_silence_site_packages = false
+# Platform configuration
+python_version = "3.8"
+platform = "linux-64"
+# Disallow dynamic typing
+disallow_any_unimported = false # TODO
+disallow_any_expr = false # TODO
+disallow_any_decorated = false # TODO
+disallow_any_explicit = false # TODO
+disallow_any_generics = false # TODO
+disallow_subclassing_any = false # TODO
+# Untyped definitions and calls
+disallow_untyped_calls = false # TODO
+disallow_untyped_defs = false # TODO
+disallow_incomplete_defs = false # TODO
+check_untyped_defs = true
+disallow_untyped_decorators = true
+# None and Optional handling
+no_implicit_optional = true
+strict_optional = true
+# Configuring warnings
+warn_redundant_casts = true
+warn_unused_ignores = true
+warn_no_return = true
+warn_return_any = false # TODO
+warn_unreachable = false # GH#27396
+# Suppressing errors
+ignore_errors = false
+enable_error_code = "ignore-without-code"
+# Miscellaneous strictness flags
+allow_untyped_globals = false
+allow_redefinition = false
+local_partial_types = false
+implicit_reexport = true
+strict_equality = true
+# Configuring error messages
+show_error_context = false
+show_column_numbers = false
+show_error_codes = true
+
+[[tool.mypy.overrides]]
+module = [
+  "pandas.tests.*",
+  "pandas._version",
+  "pandas.io.clipboard",
+]
+check_untyped_defs = false
+
+[[tool.mypy.overrides]]
+module = [
+  "pandas.tests.apply.test_series_apply",
+  "pandas.tests.arithmetic.conftest",
+  "pandas.tests.arrays.sparse.test_combine_concat",
+  "pandas.tests.dtypes.test_common",
+  "pandas.tests.frame.methods.test_to_records",
+  "pandas.tests.groupby.test_rank",
+  "pandas.tests.groupby.transform.test_transform",
+  "pandas.tests.indexes.interval.test_interval",
+  "pandas.tests.indexing.test_categorical",
+  "pandas.tests.io.excel.test_writers",
+  "pandas.tests.reductions.test_reductions",
+  "pandas.tests.test_expressions",
+]
+ignore_errors = true
+
+# To be kept consistent with "Import Formatting" section in contributing.rst
+[tool.isort]
+known_pre_libs = "pandas._config"
+known_pre_core = ["pandas._libs", "pandas._typing", "pandas.util._*", "pandas.compat", "pandas.errors"]
+known_dtypes = "pandas.core.dtypes"
+known_post_core = ["pandas.tseries", "pandas.io", "pandas.plotting"]
+sections = ["FUTURE", "STDLIB", "THIRDPARTY" ,"PRE_LIBS" , "PRE_CORE", "DTYPES", "FIRSTPARTY", "POST_CORE", "LOCALFOLDER"]
+profile = "black"
+combine_as_imports = true
+force_grid_wrap = 2
+force_sort_within_sections = true
+skip_glob = "env"
+skip = "pandas/__init__.py"
+
+[tool.pyright]
+pythonVersion = "3.8"
+typeCheckingMode = "basic"
+include = ["pandas", "typings"]
+exclude = ["pandas/tests", "pandas/io/clipboard", "pandas/util/version"]
+# enable subset of "strict"
+reportDuplicateImport = true
+reportInvalidStubStatement = true
+reportOverlappingOverload = true
+reportPropertyTypeMismatch = true
+reportUntypedClassDecorator = true
+reportUntypedFunctionDecorator = true
+reportUntypedNamedTuple = true
+reportUnusedImport = true
+# disable subset of "basic"
+reportGeneralTypeIssues = false
+reportMissingModuleSource = false
+reportOptionalCall = false
+reportOptionalIterable = false
+reportOptionalMemberAccess = false
+reportOptionalOperand = false
+reportOptionalSubscript = false
+reportPrivateImportUsage = false
+reportUnboundVariable = false
+
+[tool.coverage.run]
+branch = true
+omit = ["pandas/_typing.py", "pandas/_version.py"]
+plugins = ["Cython.Coverage"]
+source = ["pandas"]
+
+[tool.coverage.report]
+ignore_errors = false
+show_missing = true
+omit = ["pandas/_version.py"]
+exclude_lines = [
+  # Have to re-enable the standard pragma
+  "pragma: no cover",
+  # Don't complain about missing debug-only code:s
+  "def __repr__",
+  # Don't complain if tests don't hit defensive assertion code:
+  "raise AssertionError",
+  "raise NotImplementedError",
+  "AbstractMethodError",
+  # Don't complain if non-runnable code isn't run:
+  "if TYPE_CHECKING:",
+]
+
+[tool.coverage.html]
+directory = "coverage_html_report"
+
+[tool.codespell]
+ignore-words-list = "blocs, coo, hist, nd, sav, ser, recuse"
+ignore-regex = 'https://([\w/\.])+'
diff --git a/pandas/scripts/tests/data/deps_unmodified_duplicate_package.yaml b/pandas/scripts/tests/data/deps_unmodified_duplicate_package.yaml
new file mode 100644
index 0000000000000..72721c2842707
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_unmodified_duplicate_package.yaml
@@ -0,0 +1,4 @@
+# Test: duplicate package
+dependencies:
+  - jinja2>=3.0.0
+  - jinja2>=3.0.0
diff --git a/pandas/scripts/tests/data/deps_unmodified_no_version.yaml b/pandas/scripts/tests/data/deps_unmodified_no_version.yaml
new file mode 100644
index 0000000000000..c57b49a003efd
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_unmodified_no_version.yaml
@@ -0,0 +1,5 @@
+# Test: empty version
+dependencies:
+  - jinja2
+  - scipy
+  - SQLAlchemy
diff --git a/pandas/scripts/tests/data/deps_unmodified_random.yaml b/pandas/scripts/tests/data/deps_unmodified_random.yaml
new file mode 100644
index 0000000000000..afb28dd2c08bb
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_unmodified_random.yaml
@@ -0,0 +1,54 @@
+# Test: random
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.8
+
+  # build dependencies
+  - versioneer[toml]
+  - cython>=0.29.32
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=3.4.0
+  - psutil
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy
+  - pytz
+
+  # optional dependencies
+  - beautifulsoup4
+  - blosc
+  - bottleneck>=1.3.2
+  - fastparquet>=0.6.3
+  - fsspec>=2021.07.0
+  - html5lib>=1.1
+  - hypothesis
+  - gcsfs>=2021.07.0
+  - jinja2
+  - lxml>=4.6.3
+  - matplotlib>=3.6.1
+  - numba
+  - numexpr>=2.7.3
+  - openpyxl>=3.0.7
+  - odfpy>=1.4.1
+  - psycopg2
+  - pyarrow<11, >=7.0.0
+  - pymysql>=1.0.2
+  - pyreadstat>=1.1.2
+  - pytables>=3.6.1
+  - python-calamine>=0.1.7
+  - pyxlsb>=1.0.8
+  - s3fs>=2021.08.0
+  - scipy>=1.7.1
+  - sqlalchemy>=1.4.16
+  - tabulate>=0.8.9
+  - xarray>=0.21.0
+  - xlrd>=2.0.1
+  - xlsxwriter>=1.4.3
+  - zstandard>=0.15.2
diff --git a/pandas/scripts/tests/data/deps_unmodified_range.yaml b/pandas/scripts/tests/data/deps_unmodified_range.yaml
new file mode 100644
index 0000000000000..22882af2cbc4b
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_unmodified_range.yaml
@@ -0,0 +1,5 @@
+# Test: range
+dependencies:
+  - jinja2<8
+  - scipy<9
+  - SQLAlchemy<2.0
diff --git a/pandas/scripts/tests/data/deps_unmodified_same_version.yaml b/pandas/scripts/tests/data/deps_unmodified_same_version.yaml
new file mode 100644
index 0000000000000..e07b221ecd44f
--- /dev/null
+++ b/pandas/scripts/tests/data/deps_unmodified_same_version.yaml
@@ -0,0 +1,3 @@
+# Test: same version
+dependencies:
+  - jinja2>=3.0.0
diff --git a/pandas/scripts/tests/test_check_test_naming.py b/pandas/scripts/tests/test_check_test_naming.py
new file mode 100644
index 0000000000000..dbd803ce4dd31
--- /dev/null
+++ b/pandas/scripts/tests/test_check_test_naming.py
@@ -0,0 +1,43 @@
+import pytest
+
+from scripts.check_test_naming import main
+
+
+@pytest.mark.parametrize(
+    "src, expected_out, expected_ret",
+    [
+        (
+            "def foo(): pass\n",
+            "t.py:1:0 found test function which does not start with 'test'\n",
+            1,
+        ),
+        (
+            "class Foo:\n    def test_foo(): pass\n",
+            "t.py:1:0 found test class which does not start with 'Test'\n",
+            1,
+        ),
+        ("def test_foo(): pass\n", "", 0),
+        ("class TestFoo:\n    def test_foo(): pass\n", "", 0),
+        (
+            "def foo():\n    pass\ndef test_foo():\n    foo()\n",
+            "",
+            0,
+        ),
+        (
+            "class Foo:  # not a test\n"
+            "    pass\n"
+            "def test_foo():\n"
+            "    Class.foo()\n",
+            "",
+            0,
+        ),
+        ("@pytest.fixture\ndef foo(): pass\n", "", 0),
+        ("@pytest.fixture()\ndef foo(): pass\n", "", 0),
+        ("@register_extension_dtype\nclass Foo: pass\n", "", 0),
+    ],
+)
+def test_main(capsys, src, expected_out, expected_ret):
+    ret = main(src, "t.py")
+    out, _ = capsys.readouterr()
+    assert out == expected_out
+    assert ret == expected_ret
diff --git a/pandas/scripts/tests/test_inconsistent_namespace_check.py b/pandas/scripts/tests/test_inconsistent_namespace_check.py
new file mode 100644
index 0000000000000..64f66e6168efe
--- /dev/null
+++ b/pandas/scripts/tests/test_inconsistent_namespace_check.py
@@ -0,0 +1,61 @@
+import pytest
+
+from scripts.check_for_inconsistent_pandas_namespace import (
+    check_for_inconsistent_pandas_namespace,
+)
+
+BAD_FILE_0 = (
+    "from pandas import Categorical\n"
+    "cat_0 = Categorical()\n"
+    "cat_1 = pd.Categorical()"
+)
+BAD_FILE_1 = (
+    "from pandas import Categorical\n"
+    "cat_0 = pd.Categorical()\n"
+    "cat_1 = Categorical()"
+)
+BAD_FILE_2 = (
+    "from pandas import Categorical\n"
+    "cat_0 = pandas.Categorical()\n"
+    "cat_1 = Categorical()"
+)
+GOOD_FILE_0 = (
+    "from pandas import Categorical\ncat_0 = Categorical()\ncat_1 = Categorical()"
+)
+GOOD_FILE_1 = "cat_0 = pd.Categorical()\ncat_1 = pd.Categorical()"
+GOOD_FILE_2 = "from array import array\nimport pandas as pd\narr = pd.array([])"
+PATH = "t.py"
+
+
+@pytest.mark.parametrize(
+    "content, expected",
+    [
+        (BAD_FILE_0, "t.py:3:8: Found both 'pd.Categorical' and 'Categorical' in t.py"),
+        (BAD_FILE_1, "t.py:2:8: Found both 'pd.Categorical' and 'Categorical' in t.py"),
+        (
+            BAD_FILE_2,
+            "t.py:2:8: Found both 'pandas.Categorical' and 'Categorical' in t.py",
+        ),
+    ],
+)
+def test_inconsistent_usage(content, expected, capsys):
+    with pytest.raises(SystemExit):
+        check_for_inconsistent_pandas_namespace(content, PATH, replace=False)
+    result, _ = capsys.readouterr()
+    assert result == expected
+
+
+@pytest.mark.parametrize("content", [GOOD_FILE_0, GOOD_FILE_1, GOOD_FILE_2])
+@pytest.mark.parametrize("replace", [True, False])
+def test_consistent_usage(content, replace):
+    # should not raise
+    check_for_inconsistent_pandas_namespace(content, PATH, replace=replace)
+
+
+@pytest.mark.parametrize("content", [BAD_FILE_0, BAD_FILE_1, BAD_FILE_2])
+def test_inconsistent_usage_with_replace(content):
+    result = check_for_inconsistent_pandas_namespace(content, PATH, replace=True)
+    expected = (
+        "from pandas import Categorical\ncat_0 = Categorical()\ncat_1 = Categorical()"
+    )
+    assert result == expected
diff --git a/pandas/scripts/tests/test_sort_whatsnew_note.py b/pandas/scripts/tests/test_sort_whatsnew_note.py
new file mode 100644
index 0000000000000..6b8ea3ef97237
--- /dev/null
+++ b/pandas/scripts/tests/test_sort_whatsnew_note.py
@@ -0,0 +1,30 @@
+from scripts.sort_whatsnew_note import sort_whatsnew_note
+
+
+def test_sort_whatsnew_note() -> None:
+    content = (
+        ".. _whatsnew_200:\n"
+        "\n"
+        "What's new in 2.0.0 (March XX, 2023)\n"
+        "------------------------------------\n"
+        "\n"
+        "Timedelta\n"
+        "^^^^^^^^^\n"
+        "- Bug in :meth:`Timedelta.round` (:issue:`51494`)\n"
+        "- Bug in :class:`TimedeltaIndex` (:issue:`51575`)\n"
+        "\n"
+    )
+    expected = (
+        ".. _whatsnew_200:\n"
+        "\n"
+        "What's new in 2.0.0 (March XX, 2023)\n"
+        "------------------------------------\n"
+        "\n"
+        "Timedelta\n"
+        "^^^^^^^^^\n"
+        "- Bug in :class:`TimedeltaIndex` (:issue:`51575`)\n"
+        "- Bug in :meth:`Timedelta.round` (:issue:`51494`)\n"
+        "\n"
+    )
+    result = sort_whatsnew_note(content)
+    assert result == expected
diff --git a/pandas/scripts/tests/test_validate_docstrings.py b/pandas/scripts/tests/test_validate_docstrings.py
new file mode 100644
index 0000000000000..3bffd1f1987aa
--- /dev/null
+++ b/pandas/scripts/tests/test_validate_docstrings.py
@@ -0,0 +1,529 @@
+import io
+import textwrap
+
+import pytest
+
+from scripts import validate_docstrings
+
+
+class BadDocstrings:
+    """Everything here has a bad docstring"""
+
+    def private_classes(self) -> None:
+        """
+        This mentions NDFrame, which is not correct.
+        """
+
+    def prefix_pandas(self) -> None:
+        """
+        Have `pandas` prefix in See Also section.
+
+        See Also
+        --------
+        pandas.Series.rename : Alter Series index labels or name.
+        DataFrame.head : The first `n` rows of the caller object.
+        """
+
+    def redundant_import(self, paramx=None, paramy=None) -> None:
+        """
+        A sample DataFrame method.
+
+        Should not import numpy and pandas.
+
+        Examples
+        --------
+        >>> import numpy as np
+        >>> import pandas as pd
+        >>> df = pd.DataFrame(np.ones((3, 3)),
+        ...                   columns=('a', 'b', 'c'))
+        >>> df.all(axis=1)
+        0    True
+        1    True
+        2    True
+        dtype: bool
+        >>> df.all(bool_only=True)
+        Series([], dtype: bool)
+        """
+
+    def unused_import(self) -> None:
+        """
+        Examples
+        --------
+        >>> import pandas as pdf
+        >>> df = pd.DataFrame(np.ones((3, 3)), columns=('a', 'b', 'c'))
+        """
+
+    def missing_whitespace_around_arithmetic_operator(self) -> None:
+        """
+        Examples
+        --------
+        >>> 2+5
+        7
+        """
+
+    def indentation_is_not_a_multiple_of_four(self) -> None:
+        """
+        Examples
+        --------
+        >>> if 2 + 5:
+        ...   pass
+        """
+
+    def missing_whitespace_after_comma(self) -> None:
+        """
+        Examples
+        --------
+        >>> df = pd.DataFrame(np.ones((3,3)),columns=('a','b', 'c'))
+        """
+
+    def write_array_like_with_hyphen_not_underscore(self) -> None:
+        """
+        In docstrings, use array-like over array_like
+        """
+
+    def leftover_files(self) -> None:
+        """
+        Examples
+        --------
+        >>> import pathlib
+        >>> pathlib.Path("foo.txt").touch()
+        """
+
+
+class TestValidator:
+    def _import_path(self, klass=None, func=None):
+        """
+        Build the required import path for tests in this module.
+
+        Parameters
+        ----------
+        klass : str
+            Class name of object in module.
+        func : str
+            Function name of object in module.
+
+        Returns
+        -------
+        str
+            Import path of specified object in this module
+        """
+        base_path = "scripts.tests.test_validate_docstrings"
+
+        if klass:
+            base_path = f"{base_path}.{klass}"
+
+        if func:
+            base_path = f"{base_path}.{func}"
+
+        return base_path
+
+    def test_bad_class(self, capsys) -> None:
+        errors = validate_docstrings.pandas_validate(
+            self._import_path(klass="BadDocstrings")
+        )["errors"]
+        assert isinstance(errors, list)
+        assert errors
+
+    @pytest.mark.parametrize(
+        "klass,func,msgs",
+        [
+            (
+                "BadDocstrings",
+                "private_classes",
+                (
+                    "Private classes (NDFrame) should not be mentioned in public "
+                    "docstrings",
+                ),
+            ),
+            (
+                "BadDocstrings",
+                "prefix_pandas",
+                (
+                    "pandas.Series.rename in `See Also` section "
+                    "does not need `pandas` prefix",
+                ),
+            ),
+            # Examples tests
+            (
+                "BadDocstrings",
+                "redundant_import",
+                ("Do not import numpy, as it is imported automatically",),
+            ),
+            (
+                "BadDocstrings",
+                "redundant_import",
+                ("Do not import pandas, as it is imported automatically",),
+            ),
+            (
+                "BadDocstrings",
+                "unused_import",
+                (
+                    "flake8 error: line 1, col 1: F401 'pandas as pdf' "
+                    "imported but unused",
+                ),
+            ),
+            (
+                "BadDocstrings",
+                "missing_whitespace_around_arithmetic_operator",
+                (
+                    "flake8 error: line 1, col 2: "
+                    "E226 missing whitespace around arithmetic operator",
+                ),
+            ),
+            (
+                "BadDocstrings",
+                "indentation_is_not_a_multiple_of_four",
+                # with flake8 3.9.0, the message ends with four spaces,
+                #  whereas in earlier versions, it ended with "four"
+                (
+                    "flake8 error: line 2, col 3: E111 indentation is not a "
+                    "multiple of 4",
+                ),
+            ),
+            (
+                "BadDocstrings",
+                "missing_whitespace_after_comma",
+                ("flake8 error: line 1, col 33: E231 missing whitespace after ','",),
+            ),
+            (
+                "BadDocstrings",
+                "write_array_like_with_hyphen_not_underscore",
+                ("Use 'array-like' rather than 'array_like' in docstrings",),
+            ),
+        ],
+    )
+    def test_bad_docstrings(self, capsys, klass, func, msgs) -> None:
+        result = validate_docstrings.pandas_validate(
+            self._import_path(klass=klass, func=func)
+        )
+        for msg in msgs:
+            assert msg in " ".join([err[1] for err in result["errors"]])
+
+    def test_validate_all_ignore_deprecated(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "pandas_validate",
+            lambda func_name: {
+                "docstring": "docstring1",
+                "errors": [
+                    ("ER01", "err desc"),
+                    ("ER02", "err desc"),
+                    ("ER03", "err desc"),
+                ],
+                "warnings": [],
+                "examples_errors": "",
+                "deprecated": True,
+            },
+        )
+        result = validate_docstrings.validate_all(prefix=None, ignore_deprecated=True)
+        assert len(result) == 0
+
+    def test_validate_all_ignore_errors(self, monkeypatch):
+        monkeypatch.setattr(
+            validate_docstrings,
+            "pandas_validate",
+            lambda func_name: {
+                "docstring": "docstring1",
+                "errors": [
+                    ("ER01", "err desc"),
+                    ("ER02", "err desc"),
+                    ("ER03", "err desc")
+                ],
+                "warnings": [],
+                "examples_errors": "",
+                "deprecated": True,
+                "file": "file1",
+                "file_line": "file_line1"
+            },
+        )
+        monkeypatch.setattr(
+            validate_docstrings,
+            "get_all_api_items",
+            lambda: [
+                (
+                    "pandas.DataFrame.align",
+                    "func",
+                    "current_section",
+                    "current_subsection",
+                ),
+                (
+                    "pandas.Index.all",
+                    "func",
+                    "current_section",
+                    "current_subsection",
+                ),
+            ],
+        )
+
+        exit_status = validate_docstrings.print_validate_all_results(
+            output_format="default",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={None: {"ER03"}},
+        )
+        # two functions * two not ignored errors
+        assert exit_status == 2 * 2
+
+        exit_status = validate_docstrings.print_validate_all_results(
+            output_format="default",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={
+                None: {"ER03"},
+                "pandas.DataFrame.align": {"ER01"},
+                # ignoring an error that is not requested should be of no effect
+                "pandas.Index.all": {"ER03"}
+            }
+        )
+        # two functions * two not global ignored errors - one function ignored error
+        assert exit_status == 2 * 2 - 1
+
+
+
+class TestApiItems:
+    @property
+    def api_doc(self):
+        return io.StringIO(
+            textwrap.dedent(
+                """
+            .. currentmodule:: itertools
+
+            Itertools
+            ---------
+
+            Infinite
+            ~~~~~~~~
+
+            .. autosummary::
+
+                cycle
+                count
+
+            Finite
+            ~~~~~~
+
+            .. autosummary::
+
+                chain
+
+            .. currentmodule:: random
+
+            Random
+            ------
+
+            All
+            ~~~
+
+            .. autosummary::
+
+                seed
+                randint
+            """
+            )
+        )
+
+    @pytest.mark.parametrize(
+        "idx,name",
+        [
+            (0, "itertools.cycle"),
+            (1, "itertools.count"),
+            (2, "itertools.chain"),
+            (3, "random.seed"),
+            (4, "random.randint"),
+        ],
+    )
+    def test_item_name(self, idx, name) -> None:
+        result = list(validate_docstrings.get_api_items(self.api_doc))
+        assert result[idx][0] == name
+
+    @pytest.mark.parametrize(
+        "idx,func",
+        [(0, "cycle"), (1, "count"), (2, "chain"), (3, "seed"), (4, "randint")],
+    )
+    def test_item_function(self, idx, func) -> None:
+        result = list(validate_docstrings.get_api_items(self.api_doc))
+        assert callable(result[idx][1])
+        assert result[idx][1].__name__ == func
+
+    @pytest.mark.parametrize(
+        "idx,section",
+        [
+            (0, "Itertools"),
+            (1, "Itertools"),
+            (2, "Itertools"),
+            (3, "Random"),
+            (4, "Random"),
+        ],
+    )
+    def test_item_section(self, idx, section) -> None:
+        result = list(validate_docstrings.get_api_items(self.api_doc))
+        assert result[idx][2] == section
+
+    @pytest.mark.parametrize(
+        "idx,subsection",
+        [(0, "Infinite"), (1, "Infinite"), (2, "Finite"), (3, "All"), (4, "All")],
+    )
+    def test_item_subsection(self, idx, subsection) -> None:
+        result = list(validate_docstrings.get_api_items(self.api_doc))
+        assert result[idx][3] == subsection
+
+
+class TestPandasDocstringClass:
+    @pytest.mark.parametrize(
+        "name", ["pandas.Series.str.isdecimal", "pandas.Series.str.islower"]
+    )
+    def test_encode_content_write_to_file(self, name) -> None:
+        # GH25466
+        docstr = validate_docstrings.PandasDocstring(name).validate_pep8()
+        # the list of pep8 errors should be empty
+        assert not list(docstr)
+
+
+class TestMainFunction:
+    def test_exit_status_for_main(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "pandas_validate",
+            lambda func_name: {
+                "docstring": "docstring1",
+                "errors": [
+                    ("ER01", "err desc"),
+                    ("ER02", "err desc"),
+                    ("ER03", "err desc"),
+                ],
+                "examples_errs": "",
+            },
+        )
+        exit_status = validate_docstrings.main(
+            func_name="docstring1",
+            prefix=None,
+            output_format="default",
+            ignore_deprecated=False,
+            ignore_errors={},
+        )
+        assert exit_status == 3
+
+    def test_exit_status_errors_for_validate_all(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "validate_all",
+            lambda prefix, ignore_deprecated=False, ignore_functions=None: {
+                "docstring1": {
+                    "errors": [
+                        ("ER01", "err desc"),
+                        ("ER02", "err desc"),
+                        ("ER03", "err desc"),
+                    ],
+                    "file": "module1.py",
+                    "file_line": 23,
+                },
+                "docstring2": {
+                    "errors": [("ER04", "err desc"), ("ER05", "err desc")],
+                    "file": "module2.py",
+                    "file_line": 925,
+                },
+            },
+        )
+        exit_status = validate_docstrings.main(
+            func_name=None,
+            prefix=None,
+            output_format="default",
+            ignore_deprecated=False,
+            ignore_errors={},
+        )
+        assert exit_status == 5
+
+    def test_no_exit_status_noerrors_for_validate_all(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "validate_all",
+            lambda prefix, ignore_deprecated=False, ignore_functions=None: {
+                "docstring1": {"errors": [], "warnings": [("WN01", "warn desc")]},
+                "docstring2": {"errors": []},
+            },
+        )
+        exit_status = validate_docstrings.main(
+            func_name=None,
+            output_format="default",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={},
+        )
+        assert exit_status == 0
+
+    def test_exit_status_for_validate_all_json(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "validate_all",
+            lambda prefix, ignore_deprecated=False, ignore_functions=None: {
+                "docstring1": {
+                    "errors": [
+                        ("ER01", "err desc"),
+                        ("ER02", "err desc"),
+                        ("ER03", "err desc"),
+                    ]
+                },
+                "docstring2": {"errors": [("ER04", "err desc"), ("ER05", "err desc")]},
+            },
+        )
+        exit_status = validate_docstrings.main(
+            func_name=None,
+            output_format="json",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={},
+        )
+        assert exit_status == 0
+
+    def test_errors_param_filters_errors(self, monkeypatch) -> None:
+        monkeypatch.setattr(
+            validate_docstrings,
+            "validate_all",
+            lambda prefix, ignore_deprecated=False, ignore_functions=None: {
+                "Series.foo": {
+                    "errors": [
+                        ("ER01", "err desc"),
+                        ("ER02", "err desc"),
+                        ("ER03", "err desc"),
+                    ],
+                    "file": "series.py",
+                    "file_line": 142,
+                },
+                "DataFrame.bar": {
+                    "errors": [("ER01", "err desc"), ("ER02", "err desc")],
+                    "file": "frame.py",
+                    "file_line": 598,
+                },
+                "Series.foobar": {
+                    "errors": [("ER01", "err desc")],
+                    "file": "series.py",
+                    "file_line": 279,
+                },
+            },
+        )
+        monkeypatch.setattr(
+            validate_docstrings,
+            "ERROR_MSGS",
+            {
+                "ER01": "err desc",
+                "ER02": "err desc",
+                "ER03": "err desc",
+            },
+        )
+        exit_status = validate_docstrings.main(
+            func_name=None,
+            output_format="default",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={None: {"ER02", "ER03"}},
+        )
+        assert exit_status == 3
+
+        exit_status = validate_docstrings.main(
+            func_name=None,
+            output_format="default",
+            prefix=None,
+            ignore_deprecated=False,
+            ignore_errors={None: {"ER01", "ER02"}},
+        )
+        assert exit_status == 1
diff --git a/pandas/scripts/tests/test_validate_exception_location.py b/pandas/scripts/tests/test_validate_exception_location.py
new file mode 100644
index 0000000000000..18bd118549cb5
--- /dev/null
+++ b/pandas/scripts/tests/test_validate_exception_location.py
@@ -0,0 +1,59 @@
+import pytest
+
+from scripts.validate_exception_location import (
+    ERROR_MESSAGE,
+    validate_exception_and_warning_placement,
+)
+
+PATH = "t.py"
+
+# ERRORS_IN_TESTING_RST is the set returned when parsing testing.rst for all the
+# exceptions and warnings.
+CUSTOM_EXCEPTION_NOT_IN_TESTING_RST = "MyException"
+CUSTOM_EXCEPTION__IN_TESTING_RST = "MyOldException"
+ERRORS_IN_TESTING_RST = {CUSTOM_EXCEPTION__IN_TESTING_RST}
+
+TEST_CODE = """
+import numpy as np
+import sys
+
+def my_func():
+  pass
+
+class {custom_name}({error_type}):
+  pass
+
+"""
+
+
+# Test with various python-defined exceptions to ensure they are all flagged.
+@pytest.fixture(params=["Exception", "ValueError", "Warning", "UserWarning"])
+def error_type(request):
+    return request.param
+
+
+def test_class_that_inherits_an_exception_and_is_not_in_the_testing_rst_is_flagged(
+    capsys, error_type
+) -> None:
+    content = TEST_CODE.format(
+        custom_name=CUSTOM_EXCEPTION_NOT_IN_TESTING_RST, error_type=error_type
+    )
+    expected_msg = ERROR_MESSAGE.format(errors=CUSTOM_EXCEPTION_NOT_IN_TESTING_RST)
+    with pytest.raises(SystemExit, match=None):
+        validate_exception_and_warning_placement(PATH, content, ERRORS_IN_TESTING_RST)
+    result_msg, _ = capsys.readouterr()
+    assert result_msg == expected_msg
+
+
+def test_class_that_inherits_an_exception_but_is_in_the_testing_rst_is_not_flagged(
+    capsys, error_type
+) -> None:
+    content = TEST_CODE.format(
+        custom_name=CUSTOM_EXCEPTION__IN_TESTING_RST, error_type=error_type
+    )
+    validate_exception_and_warning_placement(PATH, content, ERRORS_IN_TESTING_RST)
+
+
+def test_class_that_does_not_inherit_an_exception_is_not_flagged(capsys) -> None:
+    content = "class MyClass(NonExceptionClass): pass"
+    validate_exception_and_warning_placement(PATH, content, ERRORS_IN_TESTING_RST)
diff --git a/pandas/scripts/tests/test_validate_min_versions_in_sync.py b/pandas/scripts/tests/test_validate_min_versions_in_sync.py
new file mode 100644
index 0000000000000..b6e339c13889f
--- /dev/null
+++ b/pandas/scripts/tests/test_validate_min_versions_in_sync.py
@@ -0,0 +1,61 @@
+import pathlib
+import sys
+
+import pytest
+import yaml
+
+if sys.version_info >= (3, 11):
+    import tomllib
+else:
+    import tomli as tomllib
+
+from scripts.validate_min_versions_in_sync import (
+    get_toml_map_from,
+    get_yaml_map_from,
+    pin_min_versions_to_yaml_file,
+)
+
+
+@pytest.mark.parametrize(
+    "src_toml, src_yaml, expected_yaml",
+    [
+        (
+            pathlib.Path("scripts/tests/data/deps_minimum.toml"),
+            pathlib.Path("scripts/tests/data/deps_unmodified_random.yaml"),
+            pathlib.Path("scripts/tests/data/deps_expected_random.yaml"),
+        ),
+        (
+            pathlib.Path("scripts/tests/data/deps_minimum.toml"),
+            pathlib.Path("scripts/tests/data/deps_unmodified_same_version.yaml"),
+            pathlib.Path("scripts/tests/data/deps_expected_same_version.yaml"),
+        ),
+        (
+            pathlib.Path("scripts/tests/data/deps_minimum.toml"),
+            pathlib.Path("scripts/tests/data/deps_unmodified_duplicate_package.yaml"),
+            pathlib.Path("scripts/tests/data/deps_expected_duplicate_package.yaml"),
+        ),
+        (
+            pathlib.Path("scripts/tests/data/deps_minimum.toml"),
+            pathlib.Path("scripts/tests/data/deps_unmodified_no_version.yaml"),
+            pathlib.Path("scripts/tests/data/deps_expected_no_version.yaml"),
+        ),
+        (
+            pathlib.Path("scripts/tests/data/deps_minimum.toml"),
+            pathlib.Path("scripts/tests/data/deps_unmodified_range.yaml"),
+            pathlib.Path("scripts/tests/data/deps_expected_range.yaml"),
+        ),
+    ],
+)
+def test_pin_min_versions_to_yaml_file(src_toml, src_yaml, expected_yaml) -> None:
+    with open(src_toml, "rb") as toml_f:
+        toml_map = tomllib.load(toml_f)
+    with open(src_yaml, encoding="utf-8") as yaml_f:
+        yaml_file_data = yaml_f.read()
+    yaml_file = yaml.safe_load(yaml_file_data)
+    yaml_dependencies = yaml_file["dependencies"]
+    yaml_map = get_yaml_map_from(yaml_dependencies)
+    toml_map = get_toml_map_from(toml_map)
+    result_yaml_file = pin_min_versions_to_yaml_file(yaml_map, toml_map, yaml_file_data)
+    with open(expected_yaml, encoding="utf-8") as yaml_f:
+        dummy_yaml_expected_file_1 = yaml_f.read()
+    assert result_yaml_file == dummy_yaml_expected_file_1
diff --git a/pandas/scripts/tests/test_validate_unwanted_patterns.py b/pandas/scripts/tests/test_validate_unwanted_patterns.py
new file mode 100644
index 0000000000000..4c433d03aff4d
--- /dev/null
+++ b/pandas/scripts/tests/test_validate_unwanted_patterns.py
@@ -0,0 +1,298 @@
+import io
+
+import pytest
+
+from scripts import validate_unwanted_patterns
+
+
+class TestStringsWithWrongPlacedWhitespace:
+    @pytest.mark.parametrize(
+        "data",
+        [
+            (
+                """
+    msg = (
+        "foo\n"
+        " bar"
+    )
+    """
+            ),
+            (
+                """
+    msg = (
+        "foo"
+        "  bar"
+        "baz"
+    )
+    """
+            ),
+            (
+                """
+    msg = (
+        f"foo"
+        "  bar"
+    )
+    """
+            ),
+            (
+                """
+    msg = (
+        "foo"
+        f"  bar"
+    )
+    """
+            ),
+            (
+                """
+    msg = (
+        "foo"
+        rf"  bar"
+    )
+    """
+            ),
+        ],
+    )
+    def test_strings_with_wrong_placed_whitespace(self, data) -> None:
+        fd = io.StringIO(data.strip())
+        result = list(
+            validate_unwanted_patterns.strings_with_wrong_placed_whitespace(fd)
+        )
+        assert result == []
+
+    @pytest.mark.parametrize(
+        "data, expected",
+        [
+            (
+                (
+                    """
+    msg = (
+        "foo"
+        " bar"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    )
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        f"foo"
+        " bar"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    )
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        "foo"
+        f" bar"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    )
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        f"foo"
+        f" bar"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    )
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        "foo"
+        rf" bar"
+        " baz"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                    (
+                        4,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        "foo"
+        " bar"
+        rf" baz"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                    (
+                        4,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                ],
+            ),
+            (
+                (
+                    """
+    msg = (
+        "foo"
+        rf" bar"
+        rf" baz"
+    )
+    """
+                ),
+                [
+                    (
+                        3,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                    (
+                        4,
+                        (
+                            "String has a space at the beginning instead "
+                            "of the end of the previous string."
+                        ),
+                    ),
+                ],
+            ),
+        ],
+    )
+    def test_strings_with_wrong_placed_whitespace_raises(self, data, expected) -> None:
+        fd = io.StringIO(data.strip())
+        result = list(
+            validate_unwanted_patterns.strings_with_wrong_placed_whitespace(fd)
+        )
+        assert result == expected
+
+
+class TestNoDefaultUsedNotOnlyForTyping:
+    @pytest.mark.parametrize(
+        "data",
+        [
+            (
+                """
+def f(
+    a: int | NoDefault,
+    b: float | lib.NoDefault = 0.1,
+    c: pandas._libs.lib.NoDefault = lib.no_default,
+) -> lib.NoDefault | None:
+    pass
+"""
+            ),
+            (
+                """
+# var = lib.NoDefault
+# the above is incorrect
+a: NoDefault | int
+b: lib.NoDefault = lib.no_default
+"""
+            ),
+        ],
+    )
+    def test_nodefault_used_not_only_for_typing(self, data) -> None:
+        fd = io.StringIO(data.strip())
+        result = list(validate_unwanted_patterns.nodefault_used_not_only_for_typing(fd))
+        assert result == []
+
+    @pytest.mark.parametrize(
+        "data, expected",
+        [
+            (
+                (
+                    """
+def f(
+    a = lib.NoDefault,
+    b: Any
+        = pandas._libs.lib.NoDefault,
+):
+    pass
+"""
+                ),
+                [
+                    (2, "NoDefault is used not only for typing"),
+                    (4, "NoDefault is used not only for typing"),
+                ],
+            ),
+            (
+                (
+                    """
+a: Any = lib.NoDefault
+if a is NoDefault:
+    pass
+"""
+                ),
+                [
+                    (1, "NoDefault is used not only for typing"),
+                    (2, "NoDefault is used not only for typing"),
+                ],
+            ),
+        ],
+    )
+    def test_nodefault_used_not_only_for_typing_raises(self, data, expected) -> None:
+        fd = io.StringIO(data.strip())
+        result = list(validate_unwanted_patterns.nodefault_used_not_only_for_typing(fd))
+        assert result == expected
diff --git a/pandas/scripts/validate_docstrings.py b/pandas/scripts/validate_docstrings.py
new file mode 100755
index 0000000000000..944575dcc8659
--- /dev/null
+++ b/pandas/scripts/validate_docstrings.py
@@ -0,0 +1,533 @@
+#!/usr/bin/env python3
+"""
+Analyze docstrings to detect errors.
+
+If no argument is provided, it does a quick check of docstrings and returns
+a csv with all API functions and results of basic checks.
+
+If a function or method is provided in the form "pandas.function",
+"pandas.module.class.method", etc. a list of all errors in the docstring for
+the specified function or method.
+
+Usage::
+    $ ./validate_docstrings.py
+    $ ./validate_docstrings.py pandas.DataFrame.head
+"""
+from __future__ import annotations
+
+import argparse
+import collections
+import doctest
+import importlib
+import json
+import os
+import pathlib
+import subprocess
+import sys
+import tempfile
+
+import matplotlib
+import matplotlib.pyplot as plt
+from numpydoc.docscrape import get_doc_object
+from numpydoc.validate import (
+    ERROR_MSGS as NUMPYDOC_ERROR_MSGS,
+    Validator,
+    validate,
+)
+
+# With template backend, matplotlib plots nothing
+matplotlib.use("template")
+
+# Styler methods are Jinja2 objects who's docstrings we don't own.
+IGNORE_VALIDATION = {
+    "Styler.env",
+    "Styler.template_html",
+    "Styler.template_html_style",
+    "Styler.template_html_table",
+    "Styler.template_latex",
+    "Styler.template_typst",
+    "Styler.template_string",
+    "Styler.loader",
+    "errors.InvalidComparison",
+    "errors.LossySetitemError",
+    "errors.NoBufferPresent",
+    "errors.IncompatibilityWarning",
+    "errors.PyperclipException",
+    "errors.PyperclipWindowsException",
+}
+PRIVATE_CLASSES = ["NDFrame", "IndexOpsMixin"]
+ERROR_MSGS = {
+    "GL04": "Private classes ({mentioned_private_classes}) should not be "
+    "mentioned in public docstrings",
+    "PD01": "Use 'array-like' rather than 'array_like' in docstrings.",
+    "SA05": "{reference_name} in `See Also` section does not need `pandas` "
+    "prefix, use {right_reference} instead.",
+    "EX03": "flake8 error: line {line_number}, col {col_number}: {error_code} "
+    "{error_message}",
+    "EX04": "Do not import {imported_library}, as it is imported "
+    "automatically for the examples (numpy as np, pandas as pd)",
+}
+ALL_ERRORS = set(NUMPYDOC_ERROR_MSGS).union(set(ERROR_MSGS))
+duplicated_errors = set(NUMPYDOC_ERROR_MSGS).intersection(set(ERROR_MSGS))
+assert not duplicated_errors, (f"Errors {duplicated_errors} exist in both pandas "
+                               "and numpydoc, should they be removed from pandas?")
+
+
+def pandas_error(code, **kwargs):
+    """
+    Copy of the numpydoc error function, since ERROR_MSGS can't be updated
+    with our custom errors yet.
+    """
+    return code, ERROR_MSGS[code].format(**kwargs)
+
+
+def get_api_items(api_doc_fd):
+    """
+    Yield information about all public API items.
+
+    Parse api.rst file from the documentation, and extract all the functions,
+    methods, classes, attributes... This should include all pandas public API.
+
+    Parameters
+    ----------
+    api_doc_fd : file descriptor
+        A file descriptor of the API documentation page, containing the table
+        of contents with all the public API.
+
+    Yields
+    ------
+    name : str
+        The name of the object (e.g. 'pandas.Series.str.upper').
+    func : function
+        The object itself. In most cases this will be a function or method,
+        but it can also be classes, properties, cython objects...
+    section : str
+        The name of the section in the API page where the object item is
+        located.
+    subsection : str
+        The name of the subsection in the API page where the object item is
+        located.
+    """
+    current_module = "pandas"
+    previous_line = current_section = current_subsection = ""
+    position = None
+    for line in api_doc_fd:
+        line_stripped = line.strip()
+        if len(line_stripped) == len(previous_line):
+            if set(line_stripped) == set("-"):
+                current_section = previous_line
+                continue
+            if set(line_stripped) == set("~"):
+                current_subsection = previous_line
+                continue
+
+        if line_stripped.startswith(".. currentmodule::"):
+            current_module = line_stripped.replace(".. currentmodule::", "").strip()
+            continue
+
+        if line_stripped == ".. autosummary::":
+            position = "autosummary"
+            continue
+
+        if position == "autosummary":
+            if line_stripped == "":
+                position = "items"
+                continue
+
+        if position == "items":
+            if line_stripped == "":
+                position = None
+                continue
+            if line_stripped in IGNORE_VALIDATION:
+                continue
+            func = importlib.import_module(current_module)
+            for part in line_stripped.split("."):
+                func = getattr(func, part)
+
+            yield (
+                f"{current_module}.{line_stripped}",
+                func,
+                current_section,
+                current_subsection,
+            )
+
+        previous_line = line_stripped
+
+
+class PandasDocstring(Validator):
+    def __init__(self, func_name: str, doc_obj=None) -> None:
+        self.func_name = func_name
+        if doc_obj is None:
+            doc_obj = get_doc_object(Validator._load_obj(func_name))
+        super().__init__(doc_obj)
+
+    @property
+    def name(self):
+        return self.func_name
+
+    @property
+    def mentioned_private_classes(self):
+        return [klass for klass in PRIVATE_CLASSES if klass in self.raw_doc]
+
+    @property
+    def examples_source_code(self):
+        lines = doctest.DocTestParser().get_examples(self.raw_doc)
+        return [line.source for line in lines]
+
+    def validate_pep8(self):
+        if not self.examples:
+            return
+
+        # F401 is needed to not generate flake8 errors in examples
+        # that do not user numpy or pandas
+        content = "".join(
+            (
+                "import numpy as np  # noqa: F401\n",
+                "import pandas as pd  # noqa: F401\n",
+                *self.examples_source_code,
+            )
+        )
+
+        error_messages = []
+
+        file = tempfile.NamedTemporaryFile(mode="w", encoding="utf-8", delete=False)
+        try:
+            file.write(content)
+            file.flush()
+            cmd = [
+                sys.executable,
+                "-m",
+                "flake8",
+                "--format=%(row)d\t%(col)d\t%(code)s\t%(text)s",
+                "--max-line-length=88",
+                "--ignore=E203,E3,W503,W504,E402,E731,E128,E124,E704",
+                file.name,
+            ]
+            response = subprocess.run(cmd, capture_output=True, check=False, text=True)
+            for output in ("stdout", "stderr"):
+                out = getattr(response, output)
+                out = out.replace(file.name, "")
+                messages = out.strip("\n").splitlines()
+                if messages:
+                    error_messages.extend(messages)
+        finally:
+            file.close()
+            os.unlink(file.name)
+
+        for error_message in error_messages:
+            line_number, col_number, error_code, message = error_message.split(
+                "\t", maxsplit=3
+            )
+            # Note: we subtract 2 from the line number because
+            # 'import numpy as np\nimport pandas as pd\n'
+            # is prepended to the docstrings.
+            yield error_code, message, int(line_number) - 2, int(col_number)
+
+    def non_hyphenated_array_like(self):
+        return "array_like" in self.raw_doc
+
+
+def pandas_validate(func_name: str):
+    """
+    Call the numpydoc validation, and add the errors specific to pandas.
+
+    Parameters
+    ----------
+    func_name : str
+        Name of the object of the docstring to validate.
+
+    Returns
+    -------
+    dict
+        Information about the docstring and the errors found.
+    """
+    func_obj = Validator._load_obj(func_name)
+    # Some objects are instances, e.g. IndexSlice, which numpydoc can't validate
+    doc_obj = get_doc_object(func_obj, doc=func_obj.__doc__)
+    doc = PandasDocstring(func_name, doc_obj)
+    result = validate(doc_obj)
+    mentioned_errs = doc.mentioned_private_classes
+    if mentioned_errs:
+        result["errors"].append(
+            pandas_error("GL04", mentioned_private_classes=", ".join(mentioned_errs))
+        )
+
+    if doc.see_also:
+        result["errors"].extend(
+            pandas_error(
+                "SA05",
+                reference_name=rel_name,
+                right_reference=rel_name[len("pandas."):],
+            )
+            for rel_name in doc.see_also
+            if rel_name.startswith("pandas.")
+        )
+
+    result["examples_errs"] = ""
+    if doc.examples:
+        for error_code, error_message, line_number, col_number in doc.validate_pep8():
+            result["errors"].append(
+                pandas_error(
+                    "EX03",
+                    error_code=error_code,
+                    error_message=error_message,
+                    line_number=line_number,
+                    col_number=col_number,
+                )
+            )
+        examples_source_code = "".join(doc.examples_source_code)
+        result["errors"].extend(
+            pandas_error("EX04", imported_library=wrong_import)
+            for wrong_import in ("numpy", "pandas")
+            if f"import {wrong_import}" in examples_source_code
+        )
+
+    if doc.non_hyphenated_array_like():
+        result["errors"].append(pandas_error("PD01"))
+
+    plt.close("all")
+    return result
+
+
+def validate_all(prefix, ignore_deprecated=False):
+    """
+    Execute the validation of all docstrings, and return a dict with the
+    results.
+
+    Parameters
+    ----------
+    prefix : str or None
+        If provided, only the docstrings that start with this pattern will be
+        validated. If None, all docstrings will be validated.
+    ignore_deprecated: bool, default False
+        If True, deprecated objects are ignored when validating docstrings.
+
+    Returns
+    -------
+    dict
+        A dictionary with an item for every function/method... containing
+        all the validation information.
+    """
+    result = {}
+    seen = {}
+
+    for func_name, _, section, subsection in get_all_api_items():
+        if prefix and not func_name.startswith(prefix):
+            continue
+        doc_info = pandas_validate(func_name)
+        if ignore_deprecated and doc_info["deprecated"]:
+            continue
+        result[func_name] = doc_info
+
+        shared_code_key = doc_info["file"], doc_info["file_line"]
+        shared_code = seen.get(shared_code_key, "")
+        result[func_name].update(
+            {
+                "in_api": True,
+                "section": section,
+                "subsection": subsection,
+                "shared_code_with": shared_code,
+            }
+        )
+
+        seen[shared_code_key] = func_name
+
+    return result
+
+
+def get_all_api_items():
+    base_path = pathlib.Path(__file__).parent.parent
+    api_doc_fnames = pathlib.Path(base_path, "doc", "source", "reference")
+    for api_doc_fname in api_doc_fnames.glob("*.rst"):
+        with open(api_doc_fname, encoding="utf-8") as f:
+            yield from get_api_items(f)
+
+
+def print_validate_all_results(
+    output_format: str,
+    prefix: str | None,
+    ignore_deprecated: bool,
+    ignore_errors: dict[str, set[str]],
+):
+    if output_format not in ("default", "json", "actions"):
+        raise ValueError(f'Unknown output_format "{output_format}"')
+    if ignore_errors is None:
+        ignore_errors = {}
+
+    result = validate_all(prefix, ignore_deprecated)
+
+    if output_format == "json":
+        sys.stdout.write(json.dumps(result))
+        return 0
+
+    prefix = "##[error]" if output_format == "actions" else ""
+    exit_status = 0
+    for func_name, res in result.items():
+        error_messages = dict(res["errors"])
+        actual_failures = set(error_messages)
+        expected_failures = (ignore_errors.get(func_name, set())
+                             | ignore_errors.get(None, set()))
+        for err_code in actual_failures - expected_failures:
+            sys.stdout.write(
+                f'{prefix}{res["file"]}:{res["file_line"]}:'
+                f'{err_code}:{func_name}:{error_messages[err_code]}\n'
+            )
+            exit_status += 1
+        for err_code in ignore_errors.get(func_name, set()) - actual_failures:
+            sys.stdout.write(
+                f'{prefix}{res["file"]}:{res["file_line"]}:'
+                f"{err_code}:{func_name}:"
+                "EXPECTED TO FAIL, BUT NOT FAILING\n"
+            )
+            exit_status += 1
+
+    return exit_status
+
+
+def print_validate_one_results(func_name: str,
+                               ignore_errors: dict[str, set[str]]) -> int:
+    def header(title, width=80, char="#") -> str:
+        full_line = char * width
+        side_len = (width - len(title) - 2) // 2
+        adj = "" if len(title) % 2 == 0 else " "
+        title_line = f"{char * side_len} {title}{adj} {char * side_len}"
+
+        return f"\n{full_line}\n{title_line}\n{full_line}\n\n"
+
+    result = pandas_validate(func_name)
+
+    result["errors"] = [(code, message) for code, message in result["errors"]
+                        if code not in ignore_errors.get(None, set())]
+
+    sys.stderr.write(header(f"Docstring ({func_name})"))
+    sys.stderr.write(f"{result['docstring']}\n")
+
+    sys.stderr.write(header("Validation"))
+    if result["errors"]:
+        sys.stderr.write(f'{len(result["errors"])} Errors found for `{func_name}`:\n')
+        for err_code, err_desc in result["errors"]:
+            sys.stderr.write(f"\t{err_code}\t{err_desc}\n")
+    else:
+        sys.stderr.write(f'Docstring for "{func_name}" correct. :)\n')
+
+    if result["examples_errs"]:
+        sys.stderr.write(header("Doctests"))
+        sys.stderr.write(result["examples_errs"])
+
+    return len(result["errors"]) + len(result["examples_errs"])
+
+
+def _format_ignore_errors(raw_ignore_errors):
+    ignore_errors = collections.defaultdict(set)
+    if raw_ignore_errors:
+        for error_codes in raw_ignore_errors:
+            obj_name = None
+            if " " in error_codes:
+                obj_name, error_codes = error_codes.split(" ")
+
+            # function errors "pandas.Series PR01,SA01"
+            if obj_name:
+                if obj_name in ignore_errors:
+                    raise ValueError(
+                        f"Object `{obj_name}` is present in more than one "
+                        "--ignore_errors argument. Please use it once and specify "
+                        "the errors separated by commas.")
+                ignore_errors[obj_name] = set(error_codes.split(","))
+
+                unknown_errors = ignore_errors[obj_name] - ALL_ERRORS
+                if unknown_errors:
+                    raise ValueError(
+                        f"Object `{obj_name}` is ignoring errors {unknown_errors} "
+                        f"which are not known. Known errors are: {ALL_ERRORS}")
+
+            # global errors "PR02,ES01"
+            else:
+                ignore_errors[None].update(set(error_codes.split(",")))
+
+        unknown_errors = ignore_errors["*"] - ALL_ERRORS
+        if unknown_errors:
+            raise ValueError(
+                f"Unknown errors {unknown_errors} specified using --ignore_errors "
+                "Known errors are: {ALL_ERRORS}")
+
+    return ignore_errors
+
+
+def main(
+    func_name,
+    output_format,
+    prefix,
+    ignore_deprecated,
+    ignore_errors
+):
+    """
+    Main entry point. Call the validation for one or for all docstrings.
+    """
+    if func_name is None:
+        return print_validate_all_results(
+            output_format,
+            prefix,
+            ignore_deprecated,
+            ignore_errors
+        )
+    else:
+        return print_validate_one_results(func_name, ignore_errors)
+
+
+if __name__ == "__main__":
+    format_opts = "default", "json", "actions"
+    func_help = (
+        "function or method to validate (e.g. pandas.DataFrame.head) "
+        "if not provided, all docstrings are validated and returned "
+        "as JSON"
+    )
+    argparser = argparse.ArgumentParser(description="validate pandas docstrings")
+    argparser.add_argument("function", nargs="?", default=None, help=func_help)
+    argparser.add_argument(
+        "--format",
+        default="default",
+        choices=format_opts,
+        help="format of the output when validating "
+        "multiple docstrings (ignored when validating one). "
+        "It can be {str(format_opts)[1:-1]}",
+    )
+    argparser.add_argument(
+        "--prefix",
+        default=None,
+        help="pattern for the "
+        "docstring names, in order to decide which ones "
+        'will be validated. A prefix "pandas.Series.str."'
+        "will make the script validate all the docstrings "
+        "of methods starting by this pattern. It is "
+        "ignored if parameter function is provided",
+    )
+    argparser.add_argument(
+        "--ignore_deprecated",
+        default=False,
+        action="store_true",
+        help="if this flag is set, "
+        "deprecated objects are ignored when validating "
+        "all docstrings",
+    )
+    argparser.add_argument(
+        "--ignore_errors",
+        "-i",
+        default=None,
+        action="append",
+        help="comma-separated list of error codes "
+        "(e.g. 'PR02,SA01'), with optional object path "
+        "to ignore errors for a single object "
+        "(e.g. pandas.DataFrame.head PR02,SA01). "
+        "Partial validation for more than one function"
+        "can be achieved by repeating this parameter.",
+    )
+    args = argparser.parse_args(sys.argv[1:])
+
+    sys.exit(
+        main(args.function,
+             args.format,
+             args.prefix,
+             args.ignore_deprecated,
+             _format_ignore_errors(args.ignore_errors),
+             )
+    )
diff --git a/pandas/scripts/validate_exception_location.py b/pandas/scripts/validate_exception_location.py
new file mode 100644
index 0000000000000..ecba1eb424ad5
--- /dev/null
+++ b/pandas/scripts/validate_exception_location.py
@@ -0,0 +1,104 @@
+"""
+Validate that the exceptions and warnings are in appropriate places.
+
+Checks for classes that inherit a python exception and warning and
+flags them, unless they are exempted from checking. Exempt meaning
+the exception/warning is defined in testing.rst. Testing.rst contains
+a list of pandas defined exceptions and warnings. This list is kept
+current by other pre-commit hook, pandas_errors_documented.py.
+This hook maintains that errors.__init__.py and testing.rst are in-sync.
+Therefore, the exception or warning should be defined or imported in
+errors.__init__.py. Ideally, the exception or warning is defined unless
+there's special reason to import it.
+
+Prints the exception/warning that do not follow this convention.
+
+Usage::
+
+As a pre-commit hook:
+    pre-commit run validate-errors-locations --all-files
+"""
+from __future__ import annotations
+
+import argparse
+import ast
+import pathlib
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+API_PATH = pathlib.Path("doc/source/reference/testing.rst").resolve()
+ERROR_MESSAGE = (
+    "The following exception(s) and/or warning(s): {errors} exist(s) outside of "
+    "pandas/errors/__init__.py. Please either define them in "
+    "pandas/errors/__init__.py. Or, if not possible then import them in "
+    "pandas/errors/__init__.py.\n"
+)
+
+
+def get_warnings_and_exceptions_from_api_path() -> set[str]:
+    with open(API_PATH, encoding="utf-8") as f:
+        doc_errors = {
+            line.split(".")[1].strip() for line in f.readlines() if "errors" in line
+        }
+        return doc_errors
+
+
+class Visitor(ast.NodeVisitor):
+    def __init__(self, path: str, exception_set: set[str]) -> None:
+        self.path = path
+        self.exception_set = exception_set
+        self.found_exceptions = set()
+
+    def visit_ClassDef(self, node) -> None:
+        def is_an_exception_subclass(base_id: str) -> bool:
+            return base_id == "Exception" or base_id.endswith(("Warning", "Error"))
+
+        exception_classes = []
+
+        # Go through the class's bases and check if they are an Exception or Warning.
+        for base in node.bases:
+            base_id = getattr(base, "id", None)
+            if base_id and is_an_exception_subclass(base_id):
+                exception_classes.append(base_id)
+
+        # The class subclassed an Exception or Warning so add it to the list.
+        if exception_classes:
+            self.found_exceptions.add(node.name)
+
+
+def validate_exception_and_warning_placement(
+    file_path: str, file_content: str, errors: set[str]
+) -> None:
+    tree = ast.parse(file_content)
+    visitor = Visitor(file_path, errors)
+    visitor.visit(tree)
+
+    misplaced_exceptions = visitor.found_exceptions.difference(errors)
+
+    # If misplaced_exceptions isn't an empty list then there exists
+    # pandas-defined Exception or Warnings outside of pandas/errors/__init__.py, so
+    # we should flag them.
+    if misplaced_exceptions:
+        msg = ERROR_MESSAGE.format(errors=", ".join(misplaced_exceptions))
+        sys.stdout.write(msg)
+        sys.exit(1)
+
+
+def main(argv: Sequence[str] | None = None) -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("paths", nargs="*")
+    args = parser.parse_args(argv)
+
+    error_set = get_warnings_and_exceptions_from_api_path()
+
+    for path in args.paths:
+        with open(path, encoding="utf-8") as fd:
+            content = fd.read()
+        validate_exception_and_warning_placement(path, content, error_set)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pandas/scripts/validate_min_versions_in_sync.py b/pandas/scripts/validate_min_versions_in_sync.py
new file mode 100755
index 0000000000000..1001b00450354
--- /dev/null
+++ b/pandas/scripts/validate_min_versions_in_sync.py
@@ -0,0 +1,300 @@
+#!/usr/bin/env python3
+"""
+Check pandas required and optional dependencies are synced across:
+
+ci/deps/actions-.*-minimum_versions.yaml
+pandas/compat/_optional.py
+setup.cfg
+
+TODO: doc/source/getting_started/install.rst
+
+This is meant to be run as a pre-commit hook - to run it manually, you can do:
+
+    pre-commit run validate-min-versions-in-sync --all-files
+"""
+from __future__ import annotations
+
+import pathlib
+import sys
+
+import yaml
+
+if sys.version_info >= (3, 11):
+    import tomllib
+else:
+    import tomli as tomllib
+
+from typing import Any
+
+from scripts.generate_pip_deps_from_conda import CONDA_TO_PIP
+
+DOC_PATH = pathlib.Path("doc/source/getting_started/install.rst").resolve()
+CI_PATH = next(
+    pathlib.Path("ci/deps").absolute().glob("actions-*-minimum_versions.yaml")
+)
+CODE_PATH = pathlib.Path("pandas/compat/_optional.py").resolve()
+SETUP_PATH = pathlib.Path("pyproject.toml").resolve()
+YAML_PATH = pathlib.Path("ci/deps")
+ENV_PATH = pathlib.Path("environment.yml")
+EXCLUDE_DEPS = {"tzdata", "blosc", "pyqt", "pyqt5"}
+EXCLUSION_LIST = frozenset(["python=3.8[build=*_pypy]"])
+# pandas package is not available
+# in pre-commit environment
+sys.path.append("pandas/compat")
+sys.path.append("pandas/util")
+import _exceptions
+import version
+
+sys.modules["pandas.util.version"] = version
+sys.modules["pandas.util._exceptions"] = _exceptions
+import _optional
+
+
+def pin_min_versions_to_ci_deps() -> int:
+    """
+    Pin minimum versions to CI dependencies.
+
+    Pip dependencies are not pinned.
+    """
+    all_yaml_files = list(YAML_PATH.iterdir())
+    all_yaml_files.append(ENV_PATH)
+    toml_dependencies = {}
+    with open(SETUP_PATH, "rb") as toml_f:
+        toml_dependencies = tomllib.load(toml_f)
+    ret = 0
+    for curr_file in all_yaml_files:
+        with open(curr_file, encoding="utf-8") as yaml_f:
+            yaml_start_data = yaml_f.read()
+        yaml_file = yaml.safe_load(yaml_start_data)
+        yaml_dependencies = yaml_file["dependencies"]
+        yaml_map = get_yaml_map_from(yaml_dependencies)
+        toml_map = get_toml_map_from(toml_dependencies)
+        yaml_result_data = pin_min_versions_to_yaml_file(
+            yaml_map, toml_map, yaml_start_data
+        )
+        if yaml_result_data != yaml_start_data:
+            with open(curr_file, "w", encoding="utf-8") as f:
+                f.write(yaml_result_data)
+            ret |= 1
+    return ret
+
+
+def get_toml_map_from(toml_dic: dict[str, Any]) -> dict[str, str]:
+    toml_deps = {}
+    toml_dependencies = set(toml_dic["project"]["optional-dependencies"]["all"])
+    for dependency in toml_dependencies:
+        toml_package, toml_version = dependency.strip().split(">=")
+        toml_deps[toml_package] = toml_version
+    return toml_deps
+
+
+def get_operator_from(dependency: str) -> str | None:
+    if "<=" in dependency:
+        operator = "<="
+    elif ">=" in dependency:
+        operator = ">="
+    elif "=" in dependency:
+        operator = "="
+    elif ">" in dependency:
+        operator = ">"
+    elif "<" in dependency:
+        operator = "<"
+    else:
+        operator = None
+    return operator
+
+
+def get_yaml_map_from(
+    yaml_dic: list[str | dict[str, list[str]]]
+) -> dict[str, list[str] | None]:
+    yaml_map: dict[str, list[str] | None] = {}
+    for dependency in yaml_dic:
+        if (
+            isinstance(dependency, dict)
+            or dependency in EXCLUSION_LIST
+            or dependency in yaml_map
+        ):
+            continue
+        search_text = str(dependency)
+        operator = get_operator_from(search_text)
+        if "," in dependency:
+            yaml_dependency, yaml_version1 = search_text.split(",")
+            operator = get_operator_from(yaml_dependency)
+            assert operator is not None
+            yaml_package, yaml_version2 = yaml_dependency.split(operator)
+            yaml_version2 = operator + yaml_version2
+            yaml_map[yaml_package] = [yaml_version1, yaml_version2]
+        elif "[build=*_pypy]" in dependency:
+            search_text = search_text.replace("[build=*_pypy]", "")
+            yaml_package, yaml_version = search_text.split(operator)
+            yaml_version = operator + yaml_version
+            yaml_map[yaml_package] = [yaml_version]
+        elif operator is not None:
+            yaml_package, yaml_version = search_text.split(operator)
+            yaml_version = operator + yaml_version
+            yaml_map[yaml_package] = [yaml_version]
+        else:
+            yaml_package, yaml_version = search_text.strip(), None
+            yaml_map[yaml_package] = yaml_version
+    return yaml_map
+
+
+def clean_version_list(
+    yaml_versions: list[str], toml_version: version.Version
+) -> list[str]:
+    for i in range(len(yaml_versions)):
+        yaml_version = yaml_versions[i].strip()
+        operator = get_operator_from(yaml_version)
+        assert operator is not None
+        if "<=" in operator or ">=" in operator:
+            yaml_version = yaml_version[2:]
+        else:
+            yaml_version = yaml_version[1:]
+        yaml_version = version.parse(yaml_version)
+        if yaml_version < toml_version:
+            yaml_versions[i] = "-" + str(yaml_version)
+        elif yaml_version >= toml_version:
+            if ">" in operator:
+                yaml_versions[i] = "-" + str(yaml_version)
+    return yaml_versions
+
+
+def pin_min_versions_to_yaml_file(
+    yaml_map: dict[str, list[str] | None], toml_map: dict[str, str], yaml_file_data: str
+) -> str:
+    data = yaml_file_data
+    for yaml_package, yaml_versions in yaml_map.items():
+        if yaml_package in EXCLUSION_LIST:
+            continue
+        old_dep = yaml_package
+        if yaml_versions is not None:
+            old_dep = old_dep + ", ".join(yaml_versions)
+        if CONDA_TO_PIP.get(yaml_package, yaml_package) in toml_map:
+            min_dep = toml_map[CONDA_TO_PIP.get(yaml_package, yaml_package)]
+        elif yaml_package in toml_map:
+            min_dep = toml_map[yaml_package]
+        else:
+            continue
+        if yaml_versions is None:
+            new_dep = old_dep + ">=" + min_dep
+            data = data.replace(old_dep, new_dep, 1)
+            continue
+        toml_version = version.parse(min_dep)
+        yaml_versions_list = clean_version_list(yaml_versions, toml_version)
+        cleaned_yaml_versions = [x for x in yaml_versions_list if "-" not in x]
+        new_dep = yaml_package
+        for clean_yaml_version in cleaned_yaml_versions:
+            new_dep += clean_yaml_version + ", "
+        operator = get_operator_from(new_dep)
+        if operator != "=":
+            new_dep += ">=" + min_dep
+        else:
+            new_dep = new_dep[:-2]
+        data = data.replace(old_dep, new_dep)
+    return data
+
+
+def get_versions_from_code() -> dict[str, str]:
+    """Min versions for checking within pandas code."""
+    install_map = _optional.INSTALL_MAPPING
+    inverse_install_map = {v: k for k, v in install_map.items()}
+    versions = _optional.VERSIONS
+    for item in EXCLUDE_DEPS:
+        item = inverse_install_map.get(item, item)
+        versions.pop(item, None)
+    return {install_map.get(k, k).casefold(): v for k, v in versions.items()}
+
+
+def get_versions_from_ci(content: list[str]) -> tuple[dict[str, str], dict[str, str]]:
+    """Min versions in CI job for testing all optional dependencies."""
+    # Don't parse with pyyaml because it ignores comments we're looking for
+    seen_required = False
+    seen_optional = False
+    seen_test = False
+    required_deps = {}
+    optional_deps = {}
+    for line in content:
+        if "# test dependencies" in line:
+            seen_test = True
+        elif seen_test and "- pytest>=" in line:
+            # Only grab pytest
+            package, version = line.strip().split(">=")
+            package = package[2:]
+            optional_deps[package.casefold()] = version
+        elif "# required dependencies" in line:
+            seen_required = True
+        elif "# optional dependencies" in line:
+            seen_optional = True
+        elif "- pip:" in line:
+            continue
+        elif seen_required and line.strip():
+            if "==" in line:
+                package, version = line.strip().split("==", maxsplit=1)
+            else:
+                package, version = line.strip().split("=", maxsplit=1)
+            package = package.split()[-1]
+            if package in EXCLUDE_DEPS:
+                continue
+            if not seen_optional:
+                required_deps[package.casefold()] = version
+            else:
+                optional_deps[package.casefold()] = version
+    return required_deps, optional_deps
+
+
+def get_versions_from_toml() -> dict[str, str]:
+    """Min versions in pyproject.toml for pip install pandas[extra]."""
+    install_map = _optional.INSTALL_MAPPING
+    optional_dependencies = {}
+    with open(SETUP_PATH, "rb") as pyproject_f:
+        pyproject_toml = tomllib.load(pyproject_f)
+        opt_deps = pyproject_toml["project"]["optional-dependencies"]
+        dependencies = set(opt_deps["all"])
+
+        # remove pytest plugin dependencies
+        pytest_plugins = {dep for dep in opt_deps["test"] if dep.startswith("pytest-")}
+        dependencies = dependencies.difference(pytest_plugins)
+
+    for dependency in dependencies:
+        package, version = dependency.strip().split(">=")
+        optional_dependencies[install_map.get(package, package).casefold()] = version
+
+    for item in EXCLUDE_DEPS:
+        optional_dependencies.pop(item, None)
+    return optional_dependencies
+
+
+def main() -> int:
+    ret = 0
+    ret |= pin_min_versions_to_ci_deps()
+    with open(CI_PATH, encoding="utf-8") as f:
+        _, ci_optional = get_versions_from_ci(f.readlines())
+    code_optional = get_versions_from_code()
+    setup_optional = get_versions_from_toml()
+
+    diff = (ci_optional.items() | code_optional.items() | setup_optional.items()) - (
+        ci_optional.items() & code_optional.items() & setup_optional.items()
+    )
+
+    if diff:
+        packages = {package for package, _ in diff}
+        out = sys.stdout
+        out.write(
+            f"The follow minimum version differences were found between  "
+            f"{CI_PATH}, {CODE_PATH} AND {SETUP_PATH}. "
+            f"Please ensure these are aligned: \n\n"
+        )
+
+        for package in packages:
+            out.write(
+                f"{package}\n"
+                f"{CI_PATH}: {ci_optional.get(package, 'Not specified')}\n"
+                f"{CODE_PATH}: {code_optional.get(package, 'Not specified')}\n"
+                f"{SETUP_PATH}: {setup_optional.get(package, 'Not specified')}\n\n"
+            )
+        ret |= 1
+    return ret
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/pandas/scripts/validate_rst_title_capitalization.py b/pandas/scripts/validate_rst_title_capitalization.py
new file mode 100755
index 0000000000000..44318cd797163
--- /dev/null
+++ b/pandas/scripts/validate_rst_title_capitalization.py
@@ -0,0 +1,285 @@
+"""
+Validate that the titles in the rst files follow the proper capitalization convention.
+
+Print the titles that do not follow the convention.
+
+Usage::
+
+As pre-commit hook (recommended):
+    pre-commit run title-capitalization --all-files
+
+From the command-line:
+    python scripts/validate_rst_title_capitalization.py <rst file>
+"""
+from __future__ import annotations
+
+import argparse
+import re
+import sys
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+
+CAPITALIZATION_EXCEPTIONS = {
+    "pandas",
+    "pd",
+    "Python",
+    "IPython",
+    "PyTables",
+    "Excel",
+    "JSON",
+    "HTML",
+    "SAS",
+    "SQL",
+    "BigQuery",
+    "STATA",
+    "Interval",
+    "IntervalArray",
+    "PEP8",
+    "Period",
+    "Series",
+    "Index",
+    "DataFrame",
+    "DataFrames",
+    "C",
+    "Git",
+    "GitHub",
+    "NumPy",
+    "Apache",
+    "Arrow",
+    "Parquet",
+    "MultiIndex",
+    "NumFOCUS",
+    "sklearn",
+    "Docker",
+    "PeriodIndex",
+    "NA",
+    "NaN",
+    "NaT",
+    "ValueError",
+    "Boolean",
+    "BooleanArray",
+    "KeyError",
+    "API",
+    "FAQ",
+    "IO",
+    "Timedelta",
+    "TimedeltaIndex",
+    "DatetimeIndex",
+    "IntervalIndex",
+    "Categorical",
+    "CategoricalIndex",
+    "GroupBy",
+    "DataFrameGroupBy",
+    "SeriesGroupBy",
+    "SPSS",
+    "ORC",
+    "R",
+    "HDF5",
+    "HDFStore",
+    "CDay",
+    "CBMonthBegin",
+    "CBMonthEnd",
+    "BMonthBegin",
+    "BMonthEnd",
+    "BDay",
+    "FY5253Quarter",
+    "FY5253",
+    "YearBegin",
+    "YearEnd",
+    "BYearBegin",
+    "BYearEnd",
+    "YearOffset",
+    "QuarterBegin",
+    "QuarterEnd",
+    "BQuarterBegin",
+    "BQuarterEnd",
+    "QuarterOffset",
+    "LastWeekOfMonth",
+    "WeekOfMonth",
+    "SemiMonthBegin",
+    "SemiMonthEnd",
+    "SemiMonthOffset",
+    "CustomBusinessMonthBegin",
+    "CustomBusinessMonthEnd",
+    "BusinessMonthBegin",
+    "BusinessMonthEnd",
+    "MonthBegin",
+    "MonthEnd",
+    "MonthOffset",
+    "CustomBusinessHour",
+    "CustomBusinessDay",
+    "BusinessHour",
+    "BusinessDay",
+    "DateOffset",
+    "January",
+    "February",
+    "March",
+    "April",
+    "May",
+    "June",
+    "July",
+    "August",
+    "September",
+    "October",
+    "November",
+    "December",
+    "Float64Index",
+    "FloatIndex",
+    "TZ",
+    "GIL",
+    "strftime",
+    "XPORT",
+    "Unicode",
+    "East",
+    "Asian",
+    "None",
+    "URLs",
+    "UInt64",
+    "SciPy",
+    "Matplotlib",
+    "PyPy",
+    "SparseDataFrame",
+    "Google",
+    "CategoricalDtype",
+    "UTC",
+    "False",
+    "Styler",
+    "os",
+    "str",
+    "msgpack",
+    "ExtensionArray",
+    "LZMA",
+    "Numba",
+    "Timestamp",
+    "PyArrow",
+    "Gitpod",
+    "Liveserve",
+    "I",
+    "VSCode",
+}
+
+CAP_EXCEPTIONS_DICT = {word.lower(): word for word in CAPITALIZATION_EXCEPTIONS}
+
+err_msg = "Heading capitalization formatted incorrectly. Please correctly capitalize"
+
+symbols = ("*", "=", "-", "^", "~", "#", '"')
+
+
+def correct_title_capitalization(title: str) -> str:
+    """
+    Algorithm to create the correct capitalization for a given title.
+
+    Parameters
+    ----------
+    title : str
+        Heading string to correct.
+
+    Returns
+    -------
+    str
+        Correctly capitalized heading.
+    """
+
+    # Skip modification no matter what if title begins by ":" to exclude specific
+    # syntax that is needed to build links.
+    if title[0] == ":":
+        return title
+
+    # Strip all non-word characters from the beginning of the title to the
+    # first word character.
+    correct_title: str = re.sub(r"^\W*", "", title).capitalize()
+
+    # Remove a URL from the title. We do this because words in a URL must
+    # stay lowercase, even if they are a capitalization exception.
+    removed_https_title = re.sub(r"<https?:\/\/.*[\r\n]*>", "", correct_title)
+
+    # Split a title into a list using non-word character delimiters.
+    word_list = re.split(r"\W", removed_https_title)
+
+    for word in word_list:
+        if word.lower() in CAP_EXCEPTIONS_DICT:
+            correct_title = re.sub(
+                rf"\b{word}\b", CAP_EXCEPTIONS_DICT[word.lower()], correct_title
+            )
+
+    return correct_title
+
+
+def find_titles(rst_file: str) -> Iterable[tuple[str, int]]:
+    """
+    Algorithm to identify particular text that should be considered headings in an
+    RST file.
+
+    See <https://thomas-cokelaer.info/tutorials/sphinx/rest_syntax.html> for details
+    on what constitutes a string as a heading in RST.
+
+    Parameters
+    ----------
+    rst_file : str
+        RST file to scan through for headings.
+
+    Yields
+    -------
+    title : str
+        A heading found in the rst file.
+
+    line_number : int
+        The corresponding line number of the heading.
+    """
+
+    with open(rst_file, encoding="utf-8") as fd:
+        previous_line = ""
+        for i, line in enumerate(fd):
+            line_no_last_elem = line[:-1]
+            line_chars = set(line_no_last_elem)
+            if (
+                len(line_chars) == 1
+                and line_chars.pop() in symbols
+                and len(line_no_last_elem) == len(previous_line)
+            ):
+                yield re.sub(r"[`\*_]", "", previous_line), i
+            previous_line = line_no_last_elem
+
+
+def main(source_paths: list[str]) -> int:
+    """
+    The main method to print all headings with incorrect capitalization.
+
+    Parameters
+    ----------
+    source_paths : str
+        List of directories to validate, provided through command line arguments.
+
+    Returns
+    -------
+    int
+        Number of incorrect headings found overall.
+    """
+
+    number_of_errors: int = 0
+
+    for filename in source_paths:
+        for title, line_number in find_titles(filename):
+            if title != correct_title_capitalization(title):
+                print(
+                    f"""{filename}:{line_number}:{err_msg} "{title}" to "{
+                    correct_title_capitalization(title)}" """
+                )
+                number_of_errors += 1
+
+    return number_of_errors
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Validate heading capitalization")
+
+    parser.add_argument(
+        "paths", nargs="*", help="Source paths of file/directory to check."
+    )
+
+    args = parser.parse_args()
+
+    sys.exit(main(args.paths))
diff --git a/pandas/scripts/validate_unwanted_patterns.py b/pandas/scripts/validate_unwanted_patterns.py
new file mode 100755
index 0000000000000..d804e15f6d48f
--- /dev/null
+++ b/pandas/scripts/validate_unwanted_patterns.py
@@ -0,0 +1,426 @@
+#!/usr/bin/env python3
+"""
+Unwanted patterns test cases.
+
+The reason this file exist despite the fact we already have
+`ci/code_checks.sh`,
+(see https://github.com/pandas-dev/pandas/blob/master/ci/code_checks.sh)
+
+is that some of the test cases are more complex/impossible to validate via regex.
+So this file is somewhat an extensions to `ci/code_checks.sh`
+"""
+
+import argparse
+import ast
+from collections.abc import (
+    Callable,
+    Iterable,
+)
+import sys
+import token
+import tokenize
+from typing import IO
+
+PRIVATE_IMPORTS_TO_IGNORE: set[str] = {
+    "_extension_array_shared_docs",
+    "_index_shared_docs",
+    "_interval_shared_docs",
+    "_merge_doc",
+    "_shared_docs",
+    "_new_Index",
+    "_new_PeriodIndex",
+    "_pipe_template",
+    "_apply_groupings_depr",
+    "__main__",
+    "_transform_template",
+    "_get_plot_backend",
+    "_matplotlib",
+    "_arrow_utils",
+    "_registry",
+    "_test_parse_iso8601",
+    "_testing",
+    "_test_decorators",
+    "__version__",  # check np.__version__ in compat.numpy.function
+    "__git_version__",
+    "_arrow_dtype_mapping",
+    "_global_config",
+    "_chained_assignment_msg",
+    "_chained_assignment_method_msg",
+    "_version_meson",
+    # The numba extensions need this to mock the iloc object
+    "_iLocIndexer",
+    # TODO(4.0): GH#55043 - remove upon removal of CoW option
+    "_get_option",
+    "_fill_limit_area_1d",
+    "_make_block",
+}
+
+
+def _get_literal_string_prefix_len(token_string: str) -> int:
+    """
+    Getting the length of the literal string prefix.
+
+    Parameters
+    ----------
+    token_string : str
+        String to check.
+
+    Returns
+    -------
+    int
+        Length of the literal string prefix.
+
+    Examples
+    --------
+    >>> example_string = "'Hello world'"
+    >>> _get_literal_string_prefix_len(example_string)
+    0
+    >>> example_string = "r'Hello world'"
+    >>> _get_literal_string_prefix_len(example_string)
+    1
+    """
+    try:
+        return min(
+            token_string.find(quote)
+            for quote in (r"'", r'"')
+            if token_string.find(quote) >= 0
+        )
+    except ValueError:
+        return 0
+
+
+PRIVATE_FUNCTIONS_ALLOWED = {"sys._getframe"}  # no known alternative
+
+
+def private_function_across_module(file_obj: IO[str]) -> Iterable[tuple[int, str]]:
+    """
+    Checking that a private function is not used across modules.
+    Parameters
+    ----------
+    file_obj : IO
+        File-like object containing the Python code to validate.
+    Yields
+    ------
+    line_number : int
+        Line number of the private function that is used across modules.
+    msg : str
+        Explanation of the error.
+    """
+    contents = file_obj.read()
+    tree = ast.parse(contents)
+
+    imported_modules: set[str] = set()
+
+    for node in ast.walk(tree):
+        if isinstance(node, (ast.Import, ast.ImportFrom)):
+            for module in node.names:
+                module_fqdn = module.name if module.asname is None else module.asname
+                imported_modules.add(module_fqdn)
+
+        if not isinstance(node, ast.Call):
+            continue
+
+        try:
+            module_name = node.func.value.id
+            function_name = node.func.attr
+        except AttributeError:
+            continue
+
+        # Exception section #
+
+        # (Debatable) Class case
+        if module_name[0].isupper():
+            continue
+        # (Debatable) Dunder methods case
+        elif function_name.startswith("__") and function_name.endswith("__"):
+            continue
+        elif module_name + "." + function_name in PRIVATE_FUNCTIONS_ALLOWED:
+            continue
+
+        if module_name in imported_modules and function_name.startswith("_"):
+            yield (node.lineno, f"Private function '{module_name}.{function_name}'")
+
+
+def private_import_across_module(file_obj: IO[str]) -> Iterable[tuple[int, str]]:
+    """
+    Checking that a private function is not imported across modules.
+    Parameters
+    ----------
+    file_obj : IO
+        File-like object containing the Python code to validate.
+    Yields
+    ------
+    line_number : int
+        Line number of import statement, that imports the private function.
+    msg : str
+        Explanation of the error.
+    """
+    contents = file_obj.read()
+    tree = ast.parse(contents)
+
+    for node in ast.walk(tree):
+        if not isinstance(node, (ast.Import, ast.ImportFrom)):
+            continue
+
+        for module in node.names:
+            module_name = module.name.split(".")[-1]
+            if module_name in PRIVATE_IMPORTS_TO_IGNORE:
+                continue
+
+            if module_name.startswith("_"):
+                yield (node.lineno, f"Import of internal function {module_name!r}")
+
+
+def strings_with_wrong_placed_whitespace(
+    file_obj: IO[str],
+) -> Iterable[tuple[int, str]]:
+    """
+    Test case for leading spaces in concated strings.
+
+    For example:
+
+    >>> rule = (
+    ...    "We want the space at the end of the line, "
+    ...    "not at the beginning"
+    ... )
+
+    Instead of:
+
+    >>> rule = (
+    ...    "We want the space at the end of the line,"
+    ...    " not at the beginning"
+    ... )
+
+    Parameters
+    ----------
+    file_obj : IO
+        File-like object containing the Python code to validate.
+
+    Yields
+    ------
+    line_number : int
+        Line number of unconcatenated string.
+    msg : str
+        Explanation of the error.
+    """
+
+    def has_wrong_whitespace(first_line: str, second_line: str) -> bool:
+        """
+        Checking if the two lines are mattching the unwanted pattern.
+
+        Parameters
+        ----------
+        first_line : str
+            First line to check.
+        second_line : str
+            Second line to check.
+
+        Returns
+        -------
+        bool
+            True if the two received string match, an unwanted pattern.
+
+        Notes
+        -----
+        The unwanted pattern that we are trying to catch is if the spaces in
+        a string that is concatenated over multiple lines are placed at the
+        end of each string, unless this string is ending with a
+        newline character (\n).
+
+        For example, this is bad:
+
+        >>> rule = (
+        ...    "We want the space at the end of the line,"
+        ...    " not at the beginning"
+        ... )
+
+        And what we want is:
+
+        >>> rule = (
+        ...    "We want the space at the end of the line, "
+        ...    "not at the beginning"
+        ... )
+
+        And if the string is ending with a new line character (\n) we
+        do not want any trailing whitespaces after it.
+
+        For example, this is bad:
+
+        >>> rule = (
+        ...    "We want the space at the begging of "
+        ...    "the line if the previous line is ending with a \n "
+        ...    "not at the end, like always"
+        ... )
+
+        And what we do want is:
+
+        >>> rule = (
+        ...    "We want the space at the begging of "
+        ...    "the line if the previous line is ending with a \n"
+        ...    " not at the end, like always"
+        ... )
+        """
+        if first_line.endswith(r"\n"):
+            return False
+        elif first_line.startswith("  ") or second_line.startswith("  "):
+            return False
+        elif first_line.endswith("  ") or second_line.endswith("  "):
+            return False
+        elif (not first_line.endswith(" ")) and second_line.startswith(" "):
+            return True
+        return False
+
+    tokens: list = list(tokenize.generate_tokens(file_obj.readline))
+
+    for first_token, second_token, third_token in zip(tokens, tokens[1:], tokens[2:]):
+        # Checking if we are in a block of concated string
+        if (
+            first_token.type == third_token.type == token.STRING
+            and second_token.type == token.NL
+        ):
+            # Striping the quotes, with the string literal prefix
+            first_string: str = first_token.string[
+                _get_literal_string_prefix_len(first_token.string) + 1 : -1
+            ]
+            second_string: str = third_token.string[
+                _get_literal_string_prefix_len(third_token.string) + 1 : -1
+            ]
+
+            if has_wrong_whitespace(first_string, second_string):
+                yield (
+                    third_token.start[0],
+                    (
+                        "String has a space at the beginning instead "
+                        "of the end of the previous string."
+                    ),
+                )
+
+
+def nodefault_used_not_only_for_typing(file_obj: IO[str]) -> Iterable[tuple[int, str]]:
+    """Test case where pandas._libs.lib.NoDefault is not used for typing.
+
+    Parameters
+    ----------
+    file_obj : IO
+        File-like object containing the Python code to validate.
+
+    Yields
+    ------
+    line_number : int
+        Line number of misused lib.NoDefault.
+    msg : str
+        Explanation of the error.
+    """
+    contents = file_obj.read()
+    tree = ast.parse(contents)
+    in_annotation = False
+    nodes: list[tuple[bool, ast.AST]] = [(in_annotation, tree)]
+
+    while nodes:
+        in_annotation, node = nodes.pop()
+        if not in_annotation and (
+            (isinstance(node, ast.Name)  # Case `NoDefault`
+            and node.id == "NoDefault")
+            or (isinstance(node, ast.Attribute)  # Cases e.g. `lib.NoDefault`
+            and node.attr == "NoDefault")
+        ):
+            yield (node.lineno, "NoDefault is used not only for typing")
+
+        # This part is adapted from
+        # https://github.com/asottile/pyupgrade/blob/5495a248f2165941c5d3b82ac3226ba7ad1fa59d/pyupgrade/_data.py#L70-L113
+        for name in reversed(node._fields):
+            value = getattr(node, name)
+            if name in {"annotation", "returns"}:
+                next_in_annotation = True
+            else:
+                next_in_annotation = in_annotation
+            if isinstance(value, ast.AST):
+                nodes.append((next_in_annotation, value))
+            elif isinstance(value, list):
+                nodes.extend(
+                    (next_in_annotation, value)
+                    for value in reversed(value)
+                    if isinstance(value, ast.AST)
+                )
+
+
+def main(
+    function: Callable[[IO[str]], Iterable[tuple[int, str]]],
+    source_path: str,
+    output_format: str,
+) -> bool:
+    """
+    Main entry point of the script.
+
+    Parameters
+    ----------
+    function : Callable
+        Function to execute for the specified validation type.
+    source_path : str
+        Source path representing path to a file/directory.
+    output_format : str
+        Output format of the error message.
+    file_extensions_to_check : str
+        Comma separated values of what file extensions to check.
+    excluded_file_paths : str
+        Comma separated values of what file paths to exclude during the check.
+
+    Returns
+    -------
+    bool
+        True if found any patterns are found related to the given function.
+
+    Raises
+    ------
+    ValueError
+        If the `source_path` is not pointing to existing file/directory.
+    """
+    is_failed: bool = False
+
+    for file_path in source_path:
+        with open(file_path, encoding="utf-8") as file_obj:
+            for line_number, msg in function(file_obj):
+                is_failed = True
+                print(
+                    output_format.format(
+                        source_path=file_path, line_number=line_number, msg=msg
+                    )
+                )
+
+    return is_failed
+
+
+if __name__ == "__main__":
+    available_validation_types: list[str] = [
+        "private_function_across_module",
+        "private_import_across_module",
+        "strings_with_wrong_placed_whitespace",
+        "nodefault_used_not_only_for_typing",
+    ]
+
+    parser = argparse.ArgumentParser(description="Unwanted patterns checker.")
+
+    parser.add_argument("paths", nargs="*", help="Source paths of files to check.")
+    parser.add_argument(
+        "--format",
+        "-f",
+        default="{source_path}:{line_number}: {msg}",
+        help="Output format of the error message.",
+    )
+    parser.add_argument(
+        "--validation-type",
+        "-vt",
+        choices=available_validation_types,
+        required=True,
+        help="Validation test case to check.",
+    )
+
+    args = parser.parse_args()
+
+    sys.exit(
+        main(
+            function=globals().get(args.validation_type),
+            source_path=args.paths,
+            output_format=args.format,
+        )
+    )
diff --git a/pandas/setup.py b/pandas/setup.py
new file mode 100755
index 0000000000000..737ebd270d1e4
--- /dev/null
+++ b/pandas/setup.py
@@ -0,0 +1,686 @@
+#!/usr/bin/env python3
+
+"""
+Parts of this file were taken from the pyzmq project
+(https://github.com/zeromq/pyzmq) which have been permitted for use under the
+BSD license. Parts are from lxml (https://github.com/lxml/lxml)
+"""
+
+import argparse
+import multiprocessing
+import os
+from os.path import join as pjoin
+import platform
+import shutil
+import sys
+from sysconfig import get_config_vars
+
+import numpy
+from pkg_resources import parse_version
+from setuptools import (
+    Command,
+    Extension,
+    setup,
+)
+from setuptools.command.build_ext import build_ext as _build_ext
+import versioneer
+
+cmdclass = versioneer.get_cmdclass()
+
+
+def is_platform_windows():
+    return sys.platform in ("win32", "cygwin")
+
+
+def is_platform_mac():
+    return sys.platform == "darwin"
+
+
+# note: sync with pyproject.toml, environment.yml and asv.conf.json
+min_cython_ver = "3.0"
+
+try:
+    from Cython import (
+        Tempita,
+        __version__ as _CYTHON_VERSION,
+    )
+    from Cython.Build import cythonize
+
+    _CYTHON_INSTALLED = parse_version(_CYTHON_VERSION) >= parse_version(min_cython_ver)
+except ImportError:
+    _CYTHON_VERSION = None
+    _CYTHON_INSTALLED = False
+    cythonize = lambda x, *args, **kwargs: x  # dummy func
+
+
+_pxi_dep_template = {
+    "algos": ["_libs/algos_common_helper.pxi.in", "_libs/algos_take_helper.pxi.in"],
+    "hashtable": [
+        "_libs/hashtable_class_helper.pxi.in",
+        "_libs/hashtable_func_helper.pxi.in",
+        "_libs/khash_for_primitive_helper.pxi.in",
+    ],
+    "index": ["_libs/index_class_helper.pxi.in"],
+    "sparse": ["_libs/sparse_op_helper.pxi.in"],
+    "interval": ["_libs/intervaltree.pxi.in"],
+}
+
+_pxifiles = []
+_pxi_dep = {}
+for module, files in _pxi_dep_template.items():
+    pxi_files = [pjoin("pandas", x) for x in files]
+    _pxifiles.extend(pxi_files)
+    _pxi_dep[module] = pxi_files
+
+
+class build_ext(_build_ext):
+    @classmethod
+    def render_templates(cls, pxifiles) -> None:
+        for pxifile in pxifiles:
+            # build pxifiles first, template extension must be .pxi.in
+            assert pxifile.endswith(".pxi.in")
+            outfile = pxifile[:-3]
+
+            if (
+                os.path.exists(outfile)
+                and os.stat(pxifile).st_mtime < os.stat(outfile).st_mtime
+            ):
+                # if .pxi.in is not updated, no need to output .pxi
+                continue
+
+            with open(pxifile, encoding="utf-8") as f:
+                tmpl = f.read()
+            pyxcontent = Tempita.sub(tmpl)
+
+            with open(outfile, "w", encoding="utf-8") as f:
+                f.write(pyxcontent)
+
+    def build_extensions(self) -> None:
+        # if building from c files, don't need to
+        # generate template output
+        if _CYTHON_INSTALLED:
+            self.render_templates(_pxifiles)
+
+        super().build_extensions()
+
+
+class CleanCommand(Command):
+    """Custom command to clean the .so and .pyc files."""
+
+    user_options = [("all", "a", "")]
+
+    def initialize_options(self) -> None:
+        self.all = True
+        self._clean_me = []
+        self._clean_trees = []
+
+        base = pjoin("pandas", "_libs", "src")
+        parser = pjoin(base, "parser")
+        vendored = pjoin(base, "vendored")
+        dt = pjoin(base, "datetime")
+        ujson_python = pjoin(vendored, "ujson", "python")
+        ujson_lib = pjoin(vendored, "ujson", "lib")
+        self._clean_exclude = [
+            pjoin(vendored, "numpy", "datetime", "np_datetime.c"),
+            pjoin(vendored, "numpy", "datetime", "np_datetime_strings.c"),
+            pjoin(dt, "date_conversions.c"),
+            pjoin(parser, "tokenizer.c"),
+            pjoin(parser, "io.c"),
+            pjoin(ujson_python, "ujson.c"),
+            pjoin(ujson_python, "objToJSON.c"),
+            pjoin(ujson_python, "JSONtoObj.c"),
+            pjoin(ujson_lib, "ultrajsonenc.c"),
+            pjoin(ujson_lib, "ultrajsondec.c"),
+            pjoin(dt, "pd_datetime.c"),
+            pjoin(parser, "pd_parser.c"),
+        ]
+
+        for root, dirs, files in os.walk("pandas"):
+            for f in files:
+                filepath = pjoin(root, f)
+                if filepath in self._clean_exclude:
+                    continue
+
+                if os.path.splitext(f)[-1] in (
+                    ".pyc",
+                    ".so",
+                    ".o",
+                    ".pyo",
+                    ".pyd",
+                    ".c",
+                    ".cpp",
+                    ".orig",
+                ):
+                    self._clean_me.append(filepath)
+            self._clean_trees.append(pjoin(root, d) for d in dirs if d == "__pycache__")
+
+        # clean the generated pxi files
+        for pxifile in _pxifiles:
+            pxifile_replaced = pxifile.replace(".pxi.in", ".pxi")
+            self._clean_me.append(pxifile_replaced)
+
+        self._clean_trees.append(d for d in ("build", "dist") if os.path.exists(d))
+
+    def finalize_options(self) -> None:
+        pass
+
+    def run(self) -> None:
+        for clean_me in self._clean_me:
+            try:
+                os.unlink(clean_me)
+            except OSError:
+                pass
+        for clean_tree in self._clean_trees:
+            try:
+                shutil.rmtree(clean_tree)
+            except OSError:
+                pass
+
+
+# we need to inherit from the versioneer
+# class as it encodes the version info
+sdist_class = cmdclass["sdist"]
+
+
+class CheckSDist(sdist_class):
+    """Custom sdist that ensures Cython has compiled all pyx files to c."""
+
+    _pyxfiles = [
+        "pandas/_libs/arrays.pyx",
+        "pandas/_libs/lib.pyx",
+        "pandas/_libs/hashtable.pyx",
+        "pandas/_libs/tslib.pyx",
+        "pandas/_libs/index.pyx",
+        "pandas/_libs/internals.pyx",
+        "pandas/_libs/algos.pyx",
+        "pandas/_libs/join.pyx",
+        "pandas/_libs/indexing.pyx",
+        "pandas/_libs/interval.pyx",
+        "pandas/_libs/hashing.pyx",
+        "pandas/_libs/missing.pyx",
+        "pandas/_libs/testing.pyx",
+        "pandas/_libs/sparse.pyx",
+        "pandas/_libs/ops.pyx",
+        "pandas/_libs/parsers.pyx",
+        "pandas/_libs/tslibs/base.pyx",
+        "pandas/_libs/tslibs/ccalendar.pyx",
+        "pandas/_libs/tslibs/dtypes.pyx",
+        "pandas/_libs/tslibs/period.pyx",
+        "pandas/_libs/tslibs/strptime.pyx",
+        "pandas/_libs/tslibs/np_datetime.pyx",
+        "pandas/_libs/tslibs/timedeltas.pyx",
+        "pandas/_libs/tslibs/timestamps.pyx",
+        "pandas/_libs/tslibs/timezones.pyx",
+        "pandas/_libs/tslibs/conversion.pyx",
+        "pandas/_libs/tslibs/fields.pyx",
+        "pandas/_libs/tslibs/offsets.pyx",
+        "pandas/_libs/tslibs/parsing.pyx",
+        "pandas/_libs/tslibs/tzconversion.pyx",
+        "pandas/_libs/tslibs/vectorized.pyx",
+        "pandas/_libs/window/indexers.pyx",
+        "pandas/_libs/writers.pyx",
+        "pandas/_libs/sas.pyx",
+        "pandas/_libs/byteswap.pyx",
+    ]
+
+    _cpp_pyxfiles = [
+        "pandas/_libs/window/aggregations.pyx",
+    ]
+
+    def initialize_options(self) -> None:
+        sdist_class.initialize_options(self)
+
+    def run(self) -> None:
+        if "cython" in cmdclass:
+            self.run_command("cython")
+        else:
+            # If we are not running cython then
+            # compile the extensions correctly
+            pyx_files = [(self._pyxfiles, "c"), (self._cpp_pyxfiles, "cpp")]
+
+            for pyxfiles, extension in pyx_files:
+                for pyxfile in pyxfiles:
+                    sourcefile = pyxfile[:-3] + extension
+                    msg = (
+                        f"{extension}-source file '{sourcefile}' not found.\n"
+                        "Run 'setup.py cython' before sdist."
+                    )
+                    assert os.path.isfile(sourcefile), msg
+        sdist_class.run(self)
+
+
+class CheckingBuildExt(build_ext):
+    """
+    Subclass build_ext to get clearer report if Cython is necessary.
+    """
+
+    def check_cython_extensions(self, extensions) -> None:
+        for ext in extensions:
+            for src in ext.sources:
+                if not os.path.exists(src):
+                    print(f"{ext.name}: -> [{ext.sources}]")
+                    raise Exception(
+                        f"""Cython-generated file '{src}' not found.
+                Cython is required to compile pandas from a development branch.
+                Please install Cython or download a release package of pandas.
+                """
+                    )
+
+    def build_extensions(self) -> None:
+        self.check_cython_extensions(self.extensions)
+        build_ext.build_extensions(self)
+
+
+class CythonCommand(build_ext):
+    """
+    Custom command subclassed from Cython.Distutils.build_ext
+    to compile pyx->c, and stop there. All this does is override the
+    C-compile method build_extension() with a no-op.
+    """
+
+    def build_extension(self, ext) -> None:
+        pass
+
+
+class DummyBuildSrc(Command):
+    """numpy's build_src command interferes with Cython's build_ext."""
+
+    user_options = []
+
+    def initialize_options(self) -> None:
+        self.py_modules_dict = {}
+
+    def finalize_options(self) -> None:
+        pass
+
+    def run(self) -> None:
+        pass
+
+
+cmdclass["clean"] = CleanCommand
+cmdclass["build_ext"] = CheckingBuildExt
+
+if _CYTHON_INSTALLED:
+    suffix = ".pyx"
+    cmdclass["cython"] = CythonCommand
+else:
+    suffix = ".c"
+    cmdclass["build_src"] = DummyBuildSrc
+
+# ----------------------------------------------------------------------
+# Preparation of compiler arguments
+
+debugging_symbols_requested = "--with-debugging-symbols" in sys.argv
+if debugging_symbols_requested:
+    sys.argv.remove("--with-debugging-symbols")
+
+
+if sys.byteorder == "big":
+    endian_macro = [("__BIG_ENDIAN__", "1")]
+else:
+    endian_macro = [("__LITTLE_ENDIAN__", "1")]
+
+
+extra_compile_args = []
+extra_link_args = []
+if is_platform_windows():
+    if debugging_symbols_requested:
+        extra_compile_args.append("/Z7")
+        extra_link_args.append("/DEBUG")
+else:
+    # PANDAS_CI=1 is set in CI
+    if os.environ.get("PANDAS_CI", "0") == "1":
+        extra_compile_args.append("-Werror")
+    if debugging_symbols_requested:
+        extra_compile_args.append("-g3")
+        extra_compile_args.append("-UNDEBUG")
+        extra_compile_args.append("-O0")
+
+# Build for at least macOS 10.9 when compiling on a 10.9 system or above,
+# overriding CPython distuitls behaviour which is to target the version that
+# python was built for. This may be overridden by setting
+# MACOSX_DEPLOYMENT_TARGET before calling setup.py
+if is_platform_mac():
+    if "MACOSX_DEPLOYMENT_TARGET" not in os.environ:
+        current_system = platform.mac_ver()[0]
+        python_target = get_config_vars().get(
+            "MACOSX_DEPLOYMENT_TARGET", current_system
+        )
+        target_macos_version = "10.9"
+        parsed_macos_version = parse_version(target_macos_version)
+        if (
+            parse_version(str(python_target))
+            < parsed_macos_version
+            <= parse_version(current_system)
+        ):
+            os.environ["MACOSX_DEPLOYMENT_TARGET"] = target_macos_version
+
+    if sys.version_info[:2] == (3, 8):  # GH 33239
+        extra_compile_args.append("-Wno-error=deprecated-declarations")
+
+    # https://github.com/pandas-dev/pandas/issues/35559
+    extra_compile_args.append("-Wno-error=unreachable-code")
+
+# enable coverage by building cython files by setting the environment variable
+# "PANDAS_CYTHON_COVERAGE" (with a Truthy value) or by running build_ext
+# with `--with-cython-coverage`enabled
+linetrace = os.environ.get("PANDAS_CYTHON_COVERAGE", False)
+if "--with-cython-coverage" in sys.argv:
+    linetrace = True
+    sys.argv.remove("--with-cython-coverage")
+
+# Note: if not using `cythonize`, coverage can be enabled by
+# pinning `ext.cython_directives = directives` to each ext in extensions.
+# github.com/cython/cython/wiki/enhancements-compilerdirectives#in-setuppy
+directives = {"linetrace": False, "language_level": 3, "always_allow_keywords": True}
+macros = []
+if linetrace:
+    # https://pypkg.com/pypi/pytest-cython/f/tests/example-project/setup.py
+    directives["linetrace"] = True
+    macros = [("CYTHON_TRACE", "1"), ("CYTHON_TRACE_NOGIL", "1")]
+
+# silence build warnings about deprecated API usage
+# we can't do anything about these warnings because they stem from
+# cython+numpy version mismatches.
+macros.append(("NPY_NO_DEPRECATED_API", "0"))
+
+
+# ----------------------------------------------------------------------
+# Specification of Dependencies
+
+
+# TODO(cython#4518): Need to check to see if e.g. `linetrace` has changed and
+#  possibly re-compile.
+def maybe_cythonize(extensions, *args, **kwargs):
+    """
+    Render tempita templates before calling cythonize. This is skipped for
+
+    * clean
+    * sdist
+    """
+    if "clean" in sys.argv or "sdist" in sys.argv:
+        # See https://github.com/cython/cython/issues/1495
+        return extensions
+
+    elif not _CYTHON_INSTALLED:
+        # GH#28836 raise a helfpul error message
+        if _CYTHON_VERSION:
+            raise RuntimeError(
+                f"Cannot cythonize with old Cython version ({_CYTHON_VERSION} "
+                f"installed, needs {min_cython_ver})"
+            )
+        raise RuntimeError("Cannot cythonize without Cython installed.")
+
+    # reuse any parallel arguments provided for compilation to cythonize
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--parallel", "-j", type=int, default=1)
+    parsed, _ = parser.parse_known_args()
+
+    kwargs["nthreads"] = parsed.parallel
+    build_ext.render_templates(_pxifiles)
+    if debugging_symbols_requested:
+        kwargs["gdb_debug"] = True
+
+    return cythonize(extensions, *args, **kwargs)
+
+
+def srcpath(name=None, suffix=".pyx", subdir="src"):
+    return pjoin("pandas", subdir, name + suffix)
+
+
+lib_depends = ["pandas/_libs/include/pandas/parse_helper.h"]
+
+tseries_depends = [
+    "pandas/_libs/include/pandas/datetime/pd_datetime.h",
+]
+
+ext_data = {
+    "_libs.algos": {
+        "pyxfile": "_libs/algos",
+        "depends": _pxi_dep["algos"],
+    },
+    "_libs.arrays": {"pyxfile": "_libs/arrays"},
+    "_libs.groupby": {"pyxfile": "_libs/groupby"},
+    "_libs.hashing": {"pyxfile": "_libs/hashing", "depends": []},
+    "_libs.hashtable": {
+        "pyxfile": "_libs/hashtable",
+        "depends": (
+            [
+                "pandas/_libs/include/pandas/vendored/klib/khash_python.h",
+                "pandas/_libs/include/pandas/vendored/klib/khash.h",
+            ]
+            + _pxi_dep["hashtable"]
+        ),
+    },
+    "_libs.index": {
+        "pyxfile": "_libs/index",
+        "depends": _pxi_dep["index"],
+    },
+    "_libs.indexing": {"pyxfile": "_libs/indexing"},
+    "_libs.internals": {"pyxfile": "_libs/internals"},
+    "_libs.interval": {
+        "pyxfile": "_libs/interval",
+        "depends": _pxi_dep["interval"],
+    },
+    "_libs.join": {"pyxfile": "_libs/join"},
+    "_libs.lib": {
+        "pyxfile": "_libs/lib",
+        "depends": lib_depends + tseries_depends,
+    },
+    "_libs.missing": {"pyxfile": "_libs/missing", "depends": tseries_depends},
+    "_libs.parsers": {
+        "pyxfile": "_libs/parsers",
+        "depends": [
+            "pandas/_libs/src/parser/tokenizer.h",
+            "pandas/_libs/src/parser/io.h",
+            "pandas/_libs/src/pd_parser.h",
+        ],
+    },
+    "_libs.ops": {"pyxfile": "_libs/ops"},
+    "_libs.ops_dispatch": {"pyxfile": "_libs/ops_dispatch"},
+    "_libs.properties": {"pyxfile": "_libs/properties"},
+    "_libs.reshape": {"pyxfile": "_libs/reshape", "depends": []},
+    "_libs.sparse": {"pyxfile": "_libs/sparse", "depends": _pxi_dep["sparse"]},
+    "_libs.tslib": {
+        "pyxfile": "_libs/tslib",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.base": {"pyxfile": "_libs/tslibs/base"},
+    "_libs.tslibs.ccalendar": {"pyxfile": "_libs/tslibs/ccalendar"},
+    "_libs.tslibs.dtypes": {"pyxfile": "_libs/tslibs/dtypes"},
+    "_libs.tslibs.conversion": {
+        "pyxfile": "_libs/tslibs/conversion",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.fields": {
+        "pyxfile": "_libs/tslibs/fields",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.nattype": {"pyxfile": "_libs/tslibs/nattype"},
+    "_libs.tslibs.np_datetime": {
+        "pyxfile": "_libs/tslibs/np_datetime",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.offsets": {
+        "pyxfile": "_libs/tslibs/offsets",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.parsing": {
+        "pyxfile": "_libs/tslibs/parsing",
+        "sources": ["pandas/_libs/src/parser/tokenizer.c"],
+    },
+    "_libs.tslibs.period": {
+        "pyxfile": "_libs/tslibs/period",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.strptime": {
+        "pyxfile": "_libs/tslibs/strptime",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.timedeltas": {
+        "pyxfile": "_libs/tslibs/timedeltas",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.timestamps": {
+        "pyxfile": "_libs/tslibs/timestamps",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.timezones": {"pyxfile": "_libs/tslibs/timezones"},
+    "_libs.tslibs.tzconversion": {
+        "pyxfile": "_libs/tslibs/tzconversion",
+        "depends": tseries_depends,
+    },
+    "_libs.tslibs.vectorized": {
+        "pyxfile": "_libs/tslibs/vectorized",
+        "depends": tseries_depends,
+    },
+    "_libs.testing": {"pyxfile": "_libs/testing"},
+    "_libs.window.aggregations": {
+        "pyxfile": "_libs/window/aggregations",
+        "language": "c++",
+        "suffix": ".cpp",
+        "depends": ["pandas/_libs/include/pandas/skiplist.h"],
+    },
+    "_libs.window.indexers": {"pyxfile": "_libs/window/indexers"},
+    "_libs.writers": {"pyxfile": "_libs/writers"},
+    "_libs.sas": {"pyxfile": "_libs/sas"},
+    "_libs.byteswap": {"pyxfile": "_libs/byteswap"},
+}
+
+extensions = []
+
+for name, data in ext_data.items():
+    source_suffix = suffix if suffix == ".pyx" else data.get("suffix", ".c")
+
+    sources = [srcpath(data["pyxfile"], suffix=source_suffix, subdir="")]
+
+    sources.extend(data.get("sources", []))
+
+    include = ["pandas/_libs/include", numpy.get_include()]
+
+    undef_macros = []
+
+    if (
+        sys.platform == "zos"
+        and data.get("language") == "c++"
+        and os.path.basename(os.environ.get("CXX", "/bin/xlc++")) in ("xlc", "xlc++")
+    ):
+        data.get("macros", macros).append(("__s390__", "1"))
+        extra_compile_args.append("-qlanglvl=extended0x:nolibext")
+        undef_macros.append("_POSIX_THREADS")
+
+    obj = Extension(
+        f"pandas.{name}",
+        sources=sources,
+        depends=data.get("depends", []),
+        include_dirs=include,
+        language=data.get("language", "c"),
+        define_macros=data.get("macros", macros),
+        extra_compile_args=extra_compile_args,
+        extra_link_args=extra_link_args,
+        undef_macros=undef_macros,
+    )
+
+    extensions.append(obj)
+
+# ----------------------------------------------------------------------
+# ujson
+
+if suffix == ".pyx":
+    # undo dumb setuptools bug clobbering .pyx sources back to .c
+    for ext in extensions:
+        if ext.sources[0].endswith((".c", ".cpp")):
+            root, _ = os.path.splitext(ext.sources[0])
+            ext.sources[0] = root + suffix
+
+ujson_ext = Extension(
+    "pandas._libs.json",
+    depends=[
+        "pandas/_libs/include/pandas/vendored/ujson/lib/ultrajson.h",
+        "pandas/_libs/include/pandas/datetime/pd_datetime.h",
+    ],
+    sources=(
+        [
+            "pandas/_libs/src/vendored/ujson/python/ujson.c",
+            "pandas/_libs/src/vendored/ujson/python/objToJSON.c",
+            "pandas/_libs/src/vendored/ujson/python/JSONtoObj.c",
+            "pandas/_libs/src/vendored/ujson/lib/ultrajsonenc.c",
+            "pandas/_libs/src/vendored/ujson/lib/ultrajsondec.c",
+        ]
+    ),
+    include_dirs=[
+        "pandas/_libs/include",
+        numpy.get_include(),
+    ],
+    extra_compile_args=(extra_compile_args),
+    extra_link_args=extra_link_args,
+    define_macros=macros,
+)
+
+
+extensions.append(ujson_ext)
+
+# ----------------------------------------------------------------------
+
+# ----------------------------------------------------------------------
+# pd_datetime
+pd_dt_ext = Extension(
+    "pandas._libs.pandas_datetime",
+    depends=["pandas/_libs/tslibs/datetime/pd_datetime.h"],
+    sources=(
+        [
+            "pandas/_libs/src/vendored/numpy/datetime/np_datetime.c",
+            "pandas/_libs/src/vendored/numpy/datetime/np_datetime_strings.c",
+            "pandas/_libs/src/datetime/date_conversions.c",
+            "pandas/_libs/src/datetime/pd_datetime.c",
+        ]
+    ),
+    include_dirs=[
+        "pandas/_libs/include",
+        numpy.get_include(),
+    ],
+    extra_compile_args=(extra_compile_args),
+    extra_link_args=extra_link_args,
+    define_macros=macros,
+)
+
+
+extensions.append(pd_dt_ext)
+
+# ----------------------------------------------------------------------
+
+# ----------------------------------------------------------------------
+# pd_datetime
+pd_parser_ext = Extension(
+    "pandas._libs.pandas_parser",
+    depends=["pandas/_libs/include/pandas/parser/pd_parser.h"],
+    sources=(
+        [
+            "pandas/_libs/src/parser/tokenizer.c",
+            "pandas/_libs/src/parser/io.c",
+            "pandas/_libs/src/parser/pd_parser.c",
+        ]
+    ),
+    include_dirs=[
+        "pandas/_libs/include",
+    ],
+    extra_compile_args=(extra_compile_args),
+    extra_link_args=extra_link_args,
+    define_macros=macros,
+)
+
+
+extensions.append(pd_parser_ext)
+
+
+# ----------------------------------------------------------------------
+
+
+if __name__ == "__main__":
+    # Freeze to support parallel compilation when using spawn instead of fork
+    multiprocessing.freeze_support()
+    setup(
+        version=versioneer.get_version(),
+        ext_modules=maybe_cythonize(extensions, compiler_directives=directives),
+        cmdclass=cmdclass,
+    )
diff --git a/pandas/tests/frame/test_query_eval.py b/pandas/tests/frame/test_query_eval.py
index 375b9b00a4988..9b4be66f61914 100644
--- a/pandas/tests/frame/test_query_eval.py
+++ b/pandas/tests/frame/test_query_eval.py
@@ -160,21 +160,13 @@ def test_query_empty_string(self):
             df.query("")
 
     def test_query_duplicate_column_name(self, engine, parser):
-        df = DataFrame(
-            {
-                "A": range(3),
-                "B": range(3),
-                "C": range(3)
-            }
-        ).rename(columns={"B": "A"})
+        df = DataFrame({"A": range(3), "B": range(3), "C": range(3)}).rename(
+            columns={"B": "A"}
+        )
 
-        res = df.query('C == 1', engine=engine, parser=parser)
+        res = df.query("C == 1", engine=engine, parser=parser)
 
-        expect = DataFrame(
-            [[1, 1, 1]],
-            columns=["A", "A", "C"],
-            index=[1]
-        )
+        expect = DataFrame([[1, 1, 1]], columns=["A", "A", "C"], index=[1])
 
         tm.assert_frame_equal(res, expect)
 
@@ -1140,9 +1132,7 @@ def test_query_with_nested_special_character(self, parser, engine):
             [">=", operator.ge],
         ],
     )
-    def test_query_lex_compare_strings(
-        self, parser, engine, op, func
-    ):
+    def test_query_lex_compare_strings(self, parser, engine, op, func):
         a = Series(np.random.default_rng(2).choice(list("abcde"), 20))
         b = Series(np.arange(a.size))
         df = DataFrame({"X": a, "Y": b})
@@ -1406,7 +1396,7 @@ def test_expr_with_column_name_with_backtick_and_hash(self):
     def test_expr_with_column_name_with_backtick(self):
         # GH 59285
         df = DataFrame({"a`b": (1, 2, 3), "ab": (4, 5, 6)})
-        result = df.query("`a``b` < 2")  # noqa
+        result = df.query("`a``b` < 2")
         # Note: Formatting checks may wrongly consider the above ``inline code``.
         expected = df[df["a`b"] < 2]
         tm.assert_frame_equal(result, expected)
diff --git a/pandas/tests/io/parser/data/utf16_ex.txt b/pandas/tests/io/parser/data/utf16_ex.txt
index f0b452a2bd5ff..69c2b79713483 100644
Binary files a/pandas/tests/io/parser/data/utf16_ex.txt and b/pandas/tests/io/parser/data/utf16_ex.txt differ
diff --git a/pandas/typings/numba.pyi b/pandas/typings/numba.pyi
new file mode 100644
index 0000000000000..36cccb894049b
--- /dev/null
+++ b/pandas/typings/numba.pyi
@@ -0,0 +1,41 @@
+# pyright: reportIncompleteStub = false
+from typing import (
+    Any,
+    Callable,
+    Literal,
+    overload,
+)
+
+import numba
+
+from pandas._typing import F
+
+def __getattr__(name: str) -> Any: ...  # incomplete
+@overload
+def jit(signature_or_function: F) -> F: ...
+@overload
+def jit(
+    signature_or_function: str
+    | list[str]
+    | numba.core.types.abstract.Type
+    | list[numba.core.types.abstract.Type] = ...,
+    locals: dict = ...,  # TODO: Mapping of local variable names to Numba types
+    cache: bool = ...,
+    pipeline_class: numba.compiler.CompilerBase = ...,
+    boundscheck: bool | None = ...,
+    *,
+    nopython: bool = ...,
+    forceobj: bool = ...,
+    looplift: bool = ...,
+    error_model: Literal["python", "numpy"] = ...,
+    inline: Literal["never", "always"] | Callable = ...,
+    # TODO: If a callable is provided it will be called with the call expression
+    # node that is requesting inlining, the caller's IR and callee's IR as
+    # arguments, it is expected to return Truthy as to whether to inline.
+    target: Literal["cpu", "gpu", "npyufunc", "cuda"] = ...,  # deprecated
+    nogil: bool = ...,
+    parallel: bool = ...,
+) -> Callable[[F], F]: ...
+
+njit = jit
+generated_jit = jit
diff --git a/pandas/web/README.md b/pandas/web/README.md
new file mode 100644
index 0000000000000..8586a763c51d9
--- /dev/null
+++ b/pandas/web/README.md
@@ -0,0 +1,11 @@
+Directory containing the pandas website (hosted at https://pandas.pydata.org).
+
+The website sources are in `web/pandas/`, which also include a `config.yml` file
+containing the settings to build the website. The website is generated with the
+command `./pandas_web.py pandas`. See `./pandas_web.py --help` and the header of
+the script for more information and options.
+
+After building the website, a local http server is needed to access the built website (it is not possible to open the local files with the browser, since the links and
+the image sources are absolute to where they are served from). The easiest way
+to run an http server locally is to run `python -m http.server` from the
+`web/build/` directory.
diff --git a/pandas/web/pandas/_templates/layout.html b/pandas/web/pandas/_templates/layout.html
new file mode 100644
index 0000000000000..c26b093b0c4ba
--- /dev/null
+++ b/pandas/web/pandas/_templates/layout.html
@@ -0,0 +1,113 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <script defer data-domain="pandas.pydata.org" src="https://views.scientific-python.org/js/script.js"></script>
+        <title>pandas - Python Data Analysis Library</title>
+        <meta charset="utf-8">
+        <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
+        <link rel='shortcut icon' type='image/x-icon' id='favicon-tag' href='{{ base_url }}static/img/favicon.ico'/>
+        <link rel="stylesheet"
+              href="https://cdn.jsdelivr.net/npm/bootstrap@5.0.1/dist/css/bootstrap.min.css"
+              integrity="sha384-+0n0xVW2eSR5OomGNYDnhzAbDsOXxcvSN1TPprVMTNDbiYZCxYbOOl7+AMvyTG2x"
+              crossorigin="anonymous">
+        {% for stylesheet in static.css %}
+            <link rel="stylesheet"
+                  href="{{ base_url }}{{ stylesheet }}">
+        {% endfor %}
+        <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/bootstrap-icons@1.11.3/font/bootstrap-icons.min.css">
+    </head>
+    <body>
+        <header>
+            <nav class="navbar navbar-expand-md navbar-dark fixed-top bg-dark">
+                <div class="container">
+                    <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#nav-content" aria-controls="nav-content" aria-expanded="false" aria-label="Toggle navigation">
+                        <span class="navbar-toggler-icon"></span>
+                    </button>
+
+                    {% if static.logo %}<a class="navbar-brand" href="{{ base_url }}."><img alt="" src="{{ base_url }}{{ static.logo }}"/></a>{% endif %}
+
+                    <div class="collapse navbar-collapse" id="nav-content">   
+                        <ul class="navbar-nav ms-auto">
+                            {% for item in navbar %}
+                                {% if not item.has_subitems %}
+                                    <li class="nav-item">
+                                        <a class="nav-link" href="{% if not item.target.startswith("http") %}{{ base_url }}{% endif %}{{ item.target }}">{{ item.name }}</a>
+                                    </li>
+                                {% else %}
+                                    <li class="nav-item dropdown">
+                                        <a class="nav-link dropdown-toggle"
+                                           data-bs-toggle="dropdown"
+                                           id="{{ item.slug }}"
+                                           href="#"
+                                           role="button"
+                                           aria-haspopup="true"
+                                           aria-expanded="false">{{ item.name }}</a>
+                                        <div class="dropdown-menu" aria-labelledby="{{ item.slug }}">
+                                        {% for subitem in item.target %}
+                                            <a class="dropdown-item" href="{% if not subitem.target.startswith("http") %}{{ base_url }}{% endif %}{{ subitem.target }}">{{ subitem.name }}</a>
+                                        {% endfor %}
+                                        </div>
+                                    </li>
+                                {% endif %}
+                            {% endfor %}
+                        </ul>
+                    </div>
+                </div>
+            </nav>
+        </header>
+        <main role="main">
+            <div class="container">
+                {% block body %}{% endblock %}
+            </div>
+        </main>
+        <footer class="container pt-4 pt-md-5 border-top">
+            <ul class="list-inline social-buttons float-end">
+                <li class="list-inline-item">
+                    <a href="https://t.me/s/pandas_dev">
+                        <i class="fab bi bi-telegram"></i>
+                    </a>
+                </li>
+                <li class="list-inline-item">
+                    <a href="https://fosstodon.org/@pandas_dev" rel="me">
+                        <i class="fab bi bi-mastodon"></i>
+                    </a>
+                </li>
+                <li class="list-inline-item">
+                    <a href="https://x.com/pandas_dev">
+                        <i class="fab bi bi-twitter-x"></i>
+                    </a>
+                </li>
+                <li class="list-inline-item">
+                    <a href="https://github.com/pandas-dev/pandas">
+                        <i class="fab bi bi-github"></i>
+                    </a>
+                </li>
+                <li class="list-inline-item">
+                    <a href="https://stackoverflow.com/questions/tagged/pandas">
+                        <i class="fab bi bi-stack-overflow"></i>
+                    </a>
+                </li>
+            </ul>
+            <p>
+                &copy; {{ current_year }} pandas via <a href="https://numfocus.org">NumFOCUS, Inc.</a> Hosted by <a href="https://www.ovhcloud.com">OVHcloud</a>.
+            </p>
+        </footer>
+        
+        <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.0.1/dist/js/bootstrap.bundle.min.js" 
+                integrity="sha384-gtEjrD/SeCtmISkJkNUaaKMoLD0//ElJ19smozuHV6z3Iehds+3Ulb9Bn9Plx0x4" 
+                crossorigin="anonymous"></script>
+        <script type="text/javascript">
+            const faviconTag = document.getElementById('favicon-tag');
+            const matcher = window.matchMedia('(prefers-color-scheme:dark)');
+            
+            const onUpdate = () => {
+                if (matcher.matches) faviconTag.href = "{{ base_url }}/static/img/favicon_white.ico";
+                else faviconTag.href = "{{ base_url }}/static/img/favicon.ico";
+            };
+            
+            onUpdate();
+            
+            matcher.addEventListener('change', onUpdate);
+        </script>
+    </body>
+</html>
diff --git a/pandas/web/pandas/about/citing.md b/pandas/web/pandas/about/citing.md
new file mode 100644
index 0000000000000..a3c470d05e55f
--- /dev/null
+++ b/pandas/web/pandas/about/citing.md
@@ -0,0 +1,127 @@
+# Citing and logo
+
+## Citing pandas
+
+If you use _pandas_ for a scientific publication, we would appreciate citations to the published software and the
+following paper:
+
+- [pandas on Zenodo](https://zenodo.org/search?page=1&size=20&q=conceptrecid%3A%223509134%22&sort=-version&all_versions=True),
+   Please find us on Zenodo and replace with the citation for the version you are using. You can replace the full author
+   list from there with "The pandas development team" like in the example below.
+
+        @software{reback2020pandas,
+            author       = {The pandas development team},
+            title        = {pandas-dev/pandas: Pandas},
+            month        = feb,
+            year         = 2020,
+            publisher    = {Zenodo},
+            version      = {latest},
+            doi          = {10.5281/zenodo.3509134},
+            url          = {https://doi.org/10.5281/zenodo.3509134}
+        }
+
+- [Data structures for statistical computing in python](https://pub.curvenote.com/01908378-3686-7168-a380-d82bbf21c799/public/mckinney-57fc0d4e8a08cd7f26a4b8bf468a71f4.pdf),
+   McKinney, Proceedings of the 9th Python in Science Conference, Volume 445, 2010.
+
+        @InProceedings{ mckinney-proc-scipy-2010,
+          author    = { {W}es {M}c{K}inney },
+          title     = { {D}ata {S}tructures for {S}tatistical {C}omputing in {P}ython },
+          booktitle = { {P}roceedings of the 9th {P}ython in {S}cience {C}onference },
+          pages     = { 56 - 61 },
+          year      = { 2010 },
+          editor    = { {S}t\'efan van der {W}alt and {J}arrod {M}illman },
+          doi       = { 10.25080/Majora-92bf1922-00a }
+        }
+
+## Brand and logo
+
+When using the project name _pandas_, please use it in lower case, even at the beginning of a sentence.
+
+The official logos of _pandas_ are:
+
+### Primary logo
+
+<table class="table logo">
+    <tr>
+        <td>
+            <img alt="" src="{{ base_url }}static/img/pandas.svg"/>
+        </td>
+        <td style="background-color: #150458">
+            <img alt="" src="{{ base_url }}static/img/pandas_white.svg"/>
+        </td>
+    </tr>
+</table>
+
+### Secondary logo
+
+<table class="table logo">
+    <tr>
+        <td>
+            <img alt="" src="{{ base_url }}static/img/pandas_secondary.svg"/>
+        </td>
+        <td style="background-color: #150458">
+            <img alt="" src="{{ base_url }}static/img/pandas_secondary_white.svg"/>
+        </td>
+    </tr>
+</table>
+
+### Logo mark
+
+<table class="table logo">
+    <tr>
+        <td>
+            <img alt="" src="{{ base_url }}static/img/pandas_mark.svg"/>
+        </td>
+        <td style="background-color: #150458">
+            <img alt="" src="{{ base_url }}static/img/pandas_mark_white.svg"/>
+        </td>
+    </tr>
+</table>
+
+### Logo usage
+
+The pandas logo is available in full color and white accent.
+The full color logo should only appear against white backgrounds.
+The white accent logo should go against contrasting color background.
+
+When using the logo, please follow the next directives:
+
+- Primary logo should never be seen under 1 inch in size for printing and 72px for web
+- The secondary logo should never be seen under 0.75 inch in size for printing and 55px for web
+- Leave enough margin around the logo (leave the height of the logo in the top, bottom and both sides)
+- Do not distort the logo by changing its proportions
+- Do not place text or other elements on top of the logo
+
+### Colors
+
+<table class="table">
+    <tr>
+        <td style="text-align: center;">
+            <svg xmlns="http://www.w3.org/2000/svg" width="100" height="100">
+                <circle cx="50" cy="50" r="50" fill="#150458"/>
+            </svg>
+            <br/>
+            <b style="color: #150458;">Blue</b><br/>
+            RGB: R21 G4 B88<br/>
+            HEX: #150458
+        </td>
+        <td style="text-align: center;">
+            <svg xmlns="http://www.w3.org/2000/svg" width="100" height="100">
+                <circle cx="50" cy="50" r="50" fill="#ffca00"/>
+            </svg>
+            <br/>
+            <b style="color: #150458;">Yellow</b><br/>
+            RGB: R255 G202 B0<br/>
+            HEX: #FFCA00
+        </td>
+        <td style="text-align: center;">
+            <svg xmlns="http://www.w3.org/2000/svg" width="100" height="100">
+                <circle cx="50" cy="50" r="50" fill="#e70488"/>
+            </svg>
+            <br/>
+            <b style="color: #150458;">Pink</b><br/>
+            RGB: R231 G4 B136<br/>
+            HEX: #E70488
+        </td>
+    </tr>
+</table>
diff --git a/pandas/web/pandas/about/governance.md b/pandas/web/pandas/about/governance.md
new file mode 100644
index 0000000000000..d8777d1d0c15d
--- /dev/null
+++ b/pandas/web/pandas/about/governance.md
@@ -0,0 +1,317 @@
+# Project governance
+
+The official version of this document, along with a list of
+individuals and institutions in the roles defined in the governance
+section below, is contained in the
+[Project governance]({{ base_url }}about/governance.html)
+page of the pandas website.
+
+## The Project
+
+The pandas Project (The Project) is an open source software project affiliated
+with the 501(c)3 NumFOCUS Foundation. The goal of The Project is to develop open
+source software for data ingest, data preparation, data analysis, and data
+visualization for the Python programming language. The Software developed by
+The Project is released under the BSD (or similar) open source license,
+developed openly and hosted in public GitHub repositories under the [pandas
+GitHub organization](https://github.com/pandas-dev). Examples of Project Software
+include the main pandas code repository and the pandas-stubs library.
+
+Through its affiliation with NumFOCUS, The Project has the right to receive
+tax-deductible donations in the United States of America.
+
+The Project is developed by a team of distributed developers, called
+Contributors. Contributors are individuals who have contributed code,
+documentation, designs or other work to one or more Project repositories.
+Anyone can be a Contributor. Contributors can be affiliated with any legal
+entity or none. Contributors participate in the project by submitting,
+reviewing and discussing GitHub Pull Requests and Issues and participating in
+open and public Project discussions on GitHub, mailing lists, and
+elsewhere. The foundation of Project participation is openness and
+transparency.
+
+Here is a list of the current Contributors to the main pandas repository:
+
+[https://github.com/pandas-dev/pandas/graphs/contributors](https://github.com/pandas-dev/pandas/graphs/contributors)
+
+There are also many other Contributors listed in the logs of other repositories of
+the pandas project.
+
+The Project Community consists of all Contributors and Users of the Project.
+Contributors work on behalf of and are responsible to the larger Project
+Community and we strive to keep the barrier between Contributors and Users as
+low as possible.
+
+The Project is formally affiliated with the 501(c)3 NumFOCUS Foundation
+([https://numfocus.org](https://numfocus.org)), which serves as its fiscal
+sponsor, may hold project trademarks and other intellectual property, helps
+manage project donations and acts as a parent legal entity. NumFOCUS is the
+only legal entity that has a formal relationship with the project (see
+Institutional Partners section below).
+
+## Governance
+
+This section describes the governance and leadership model of The Project.
+
+The foundations of Project governance are:
+
+-   Openness & Transparency
+-   Active Contribution
+-   Institutional Neutrality
+
+Traditionally, Project leadership was provided by a BDFL (Wes McKinney) and
+subset of Contributors, called the Core Team, whose active and consistent
+contributions have been recognized by their receiving “commit rights” to the
+Project GitHub repositories. In general all Project decisions are made through
+consensus among the Core Team with input from the Community. The BDFL can, but
+rarely chooses to, override the Core Team and make a final decision on a
+matter.
+
+While this approach has served us well, as the Project grows and faces more
+legal and financial decisions and interacts with other institutions, we see a
+need for a more formal governance model. Moving forward The Project leadership
+will consist of a BDFL and Core Team. We view this governance model as the
+formalization of what we are already doing, rather than a change in direction.
+
+### BDFL
+
+The Project will have a BDFL (Benevolent Dictator for Life), who is currently
+Wes McKinney. As Dictator, the BDFL has the authority to make all final
+decisions for The Project. As Benevolent, the BDFL, in practice chooses to
+defer that authority to the consensus of the community discussion channels and
+the Core Team. It is expected, and in the past has been the case, that the BDFL
+will only rarely assert his/her final authority. Because it is rarely used, we
+refer to BDFL’s final authority as a “special” or “overriding” vote. When it
+does occur, the BDFL override typically happens in situations where there is a
+deadlock in the Core Team or if the Core Team ask the BDFL to make a decision
+on a specific matter. To ensure the benevolence of the BDFL, The Project
+encourages others to fork the project if they disagree with the overall
+direction the BDFL is taking. The BDFL is chair of the Core Team (see below)
+and may delegate his/her authority on a particular decision or set of decisions
+to any other Core Team Member at his/her discretion.
+
+The BDFL can appoint his/her successor, but it is expected that the Core Team
+would be consulted on this decision. If the BDFL is unable to appoint a
+successor (e.g. due to death or illness), the Core Team will choose a successor
+by voting with at least 2/3 of the Core Team members voting in favor of the
+chosen successor. At least 80% of the Core Team must participate in the
+vote. If no BDFL candidate receives 2/3 of the votes of the Core Team, the Core
+Team members shall propose the BDFL candidates to the Main NumFOCUS board, who
+will then make the final decision.
+
+### Core Team
+
+The Project's Core Team will consist of Project Contributors who have produced
+contributions that are substantial in quality and quantity, and sustained over
+at least one year. The overall role of the Core Team is to ensure, through
+working with the BDFL and taking input from the Community, the long-term
+well-being of the project, both technically and as a community.
+
+During the everyday project activities, Core Team participate in all
+discussions, code review and other project activities as peers with all other
+Contributors and the Community. In these everyday activities, Core Team do not
+have any special power or privilege through their membership on the Core
+Team. However, it is expected that because of the quality and quantity of their
+contributions and their expert knowledge of the Project Software that the Core
+Team will provide useful guidance, both technical and in terms of project
+direction, to potentially less experienced contributors.
+
+The Core Team and its Members play a special role in certain situations.
+In particular, the Core Team may:
+
+-   Make decisions about the overall scope, vision and direction of the
+    project.
+-   Make decisions about strategic collaborations with other organizations or
+    individuals.
+-   Make decisions about specific technical issues, features, bugs and pull
+    requests. They are the primary mechanism of guiding the code review process
+    and merging pull requests.
+-   Make decisions about the Services that are run by The Project and manage
+    those Services for the benefit of the Project and Community.
+-   Make decisions when regular community discussion doesn't produce consensus
+    on an issue in a reasonable time frame.
+
+### Core Team membership
+
+To become eligible for being a Core Team Member an individual must be a Project
+Contributor who has produced contributions that are substantial in quality and
+quantity, and sustained over at least one year. Potential Core Team Members are
+nominated by existing Core members and voted upon by the existing Core Team
+after asking if the potential Member is interested and willing to serve in that
+capacity. The Core Team will be initially formed from the set of existing
+Contributors who have been granted commit rights as of late 2015.
+
+When considering potential Members, the Core Team will look at candidates with
+a comprehensive view of their contributions. This will include but is not
+limited to code, code review, infrastructure work, mailing list and chat
+participation, community help/building, education and outreach, design work,
+etc. We are deliberately not setting arbitrary quantitative metrics (like “100
+commits in this repo”) to avoid encouraging behavior that plays to the metrics
+rather than the project’s overall well-being. We want to encourage a diverse
+array of backgrounds, viewpoints and talents in our team, which is why we
+explicitly do not define code as the sole metric on which Core Team membership
+will be evaluated.
+
+If a Core Team member becomes inactive in the project for a period of one year,
+they will be considered for removal from the Core Team. Before removal,
+inactive Member will be approached by the BDFL to see if they plan on returning
+to active participation. If not they will be removed immediately upon a Core
+Team vote. If they plan on returning to active participation soon, they will be
+given a grace period of one year. If they don't return to active participation
+within that time period they will be removed by vote of the Core Team without
+further grace period. All former Core Team members can be considered for
+membership again at any time in the future, like any other Project Contributor.
+Retired Core Team members will be listed on the project website, acknowledging
+the period during which they were active in the Core Team.
+
+The Core Team reserves the right to eject current Members, other than the BDFL,
+if they are deemed to be actively harmful to the project’s well-being, and
+attempts at communication and conflict resolution have failed.
+
+### Conflict of interest
+
+It is expected that the BDFL and Core Team Members will be employed at a wide
+range of companies, universities and non-profit organizations. Because of this,
+it is possible that Members will have conflict of interests. Such conflict of
+interests include, but are not limited to:
+
+-   Financial interests, such as investments, employment or contracting work,
+    outside of The Project that may influence their work on The Project.
+-   Access to proprietary information of their employer that could potentially
+    leak into their work with the Project.
+
+All members of the Core Team, BDFL included, shall disclose to the rest of the
+Core Team any conflict of interest they may have. Members with a conflict of
+interest in a particular issue may participate in Core Team discussions on that
+issue, but must recuse themselves from voting on the issue. If the BDFL has
+recused his/herself for a particular decision, they will appoint a substitute
+BDFL for that decision.
+
+### Private communications of the Core Team
+
+Unless specifically required, all Core Team discussions and activities will be
+public and done in collaboration and discussion with the Project Contributors
+and Community. The Core Team will have a private mailing list that will be used
+sparingly and only when a specific matter requires privacy. When private
+communications and decisions are needed, the Core Team will do its best to
+summarize those to the Community after eliding personal/private/sensitive
+information that should not be posted to the public internet.
+
+### Subcommittees
+
+The Core Team can create subcommittees that provide leadership and guidance for
+specific aspects of the project. Like the Core Team as a whole, subcommittees
+should conduct their business in an open and public manner unless privacy is
+specifically called for. Private subcommittee communications should happen on
+the main private mailing list of the Core Team unless specifically called for.
+
+Question: if the BDFL is not on a subcommittee, do they still have override
+authority?
+
+Suggestion: they do, but they should appoint a delegate who plays that role
+most of the time, and explicit BDFL intervention is sought only if the
+committee disagrees with that delegate’s decision and no resolution is possible
+within the team. This is different from a BDFL delegate for a specific decision
+(or a recusal situation), where the BDFL is literally giving up his/her
+authority to someone else in full. It’s more like what Linus Torvalds uses with his
+“lieutenants” model.
+
+### NumFOCUS Subcommittee
+
+The Core Team will maintain one narrowly focused subcommittee to manage its
+interactions with NumFOCUS.
+
+- The NumFOCUS Subcommittee is comprised of at least 5 persons who manage
+  project funding that comes through NumFOCUS. It is expected that these funds
+  will be spent in a manner that is consistent with the non-profit mission of
+  NumFOCUS and the direction of the Project as determined by the full Core
+  Team.
+- This Subcommittee shall NOT make decisions about the direction, scope or
+  technical direction of the Project.
+- This Subcommittee will have at least 5 members. No more than 2 Subcommittee
+  Members can report to one person (either directly or indirectly) through
+  employment or contracting work (including the reportee, i.e. the reportee + 1
+  is the max). This avoids effective majorities resting on one person.
+
+## Institutional Partners and Funding
+
+The BDFL and Core Team are the primary leadership for the project. No outside
+institution, individual or legal entity has the ability to own, control, usurp
+or influence the project other than by participating in the Project as
+Contributors and Core Team. However, because institutions are the primary
+funding mechanism for the project, it is important to formally acknowledge
+institutional participation in the project. These are Institutional Partners.
+
+An Institutional Contributor is any individual Project Contributor who
+contributes to the project as part of their official duties at an Institutional
+Partner. Likewise, an Institutional Core Team Member is any Core Team Member
+who contributes to the project as part of their official duties at an
+Institutional Partner.
+
+With these definitions, an Institutional Partner is any recognized legal entity
+in the United States or elsewhere that employs at least one Institutional
+Contributor or Institutional Core Team Member. Institutional Partners can be
+for-profit or non-profit entities.
+
+Institutions become eligible to become an Institutional Partner by employing
+individuals who actively contribute to The Project as part of their official
+duties. To state this another way, the only way for an Institutional Partner to
+influence the project is by actively contributing to the open development of
+the project, on equal terms with any other member of the community of
+Contributors and Core Team Members. Merely using pandas Software or Services in
+an institutional context does not allow an entity to become an Institutional
+Partner. Financial gifts do not enable an entity to become an Institutional
+Partner. Once an institution becomes eligible for Institutional Partnership,
+the Core Team must nominate and approve the Partnership.
+
+If an existing Institutional Partner no longer has a contributing employee,
+they will be given a one-year grace period for other employees to begin
+contributing.
+
+An Institutional Partner is free to pursue funding for their work on The
+Project through any legal means. This could involve a non-profit organization
+raising money from private foundations and donors or a for-profit company
+building proprietary products and services that leverage Project Software and
+Services. Funding acquired by Institutional Partners to work on The Project is
+called Institutional Funding. However, no funding obtained by an Institutional
+Partner can override The Project BDFL and Core Team. If a Partner has funding
+to do pandas work and the Core Team decides to not pursue that work as a
+project, the Partner is free to pursue it on their own. However in this
+situation, that part of the Partner’s work will not be under the pandas
+umbrella and cannot use the Project trademarks in a way that suggests a formal
+relationship.
+
+To acknowledge institutional contributions, there are two levels of
+Institutional Partners, with associated benefits:
+
+**Tier 1** = an institution with at least one Institutional Core Team Member
+
+- Acknowledged on the pandas website, in talks and T-shirts.
+- Ability to acknowledge their own funding sources on the pandas website, in
+  talks and T-shirts.
+- Ability to influence the project through the participation of their Core Team
+  Member.
+
+**Tier 2** = an institution with at least one Institutional Contributor
+
+## Breach
+
+Non-compliance with the terms of the governance documents shall be reported to
+the Core Team either through public or private channels as deemed appropriate.
+
+## Changing the Governance
+
+Changes to the governance are submitted via a GitHub pull request to The Project's
+[governance page](https://github.com/pandas-dev/pandas/blob/main/web/pandas/about/governance.md).
+The pull request is then refined in response to public comment and review, with
+the goal being consensus in the community.  After this open period, a Core Team
+Member proposes to the Core Team that the changes be ratified and the pull
+request merged (accepting the proposed changes) or proposes that the pull
+request be closed without merging (rejecting the proposed changes). The Member
+should state the final commit hash in the pull request being proposed for
+acceptance or rejection and briefly summarize the pull request. A minimum of
+80% of the Core Team must vote and at least 2/3 of the votes must be positive
+to carry out the proposed action (fractions of a vote rounded up to the nearest
+integer). Since the BDFL holds ultimate authority in The Project, the BDFL has
+authority to act alone in accepting or rejecting changes or overriding Core
+Team decisions.
diff --git a/pandas/web/pandas/about/index.md b/pandas/web/pandas/about/index.md
new file mode 100644
index 0000000000000..4c809a148b328
--- /dev/null
+++ b/pandas/web/pandas/about/index.md
@@ -0,0 +1,86 @@
+# About pandas
+
+## History of development
+
+In 2008, _pandas_ development began at [AQR Capital Management](https://www.aqr.com).
+By the end of 2009 it had been [open sourced](https://en.wikipedia.org/wiki/Open_source),
+and is actively supported today by a community of like-minded individuals around the world who
+contribute their valuable time and energy to help make open source _pandas_
+possible. Thank you to [all of our contributors](team.html).
+
+Since 2015, _pandas_ is a [NumFOCUS sponsored project](https://numfocus.org/sponsored-projects).
+This will help ensure the success of development of _pandas_ as a world-class open-source project.
+
+### Timeline
+
+- **2008**:  Development of _pandas_ started
+- **2009**: _pandas_ becomes open source
+- **2012**: First edition of _Python for Data Analysis_ is published
+- **2015**: _pandas_ becomes a [NumFOCUS sponsored project](https://numfocus.org/sponsored-projects)
+- **2018**: First in-person core developer sprint
+
+## Library Highlights
+
+- A fast and efficient **DataFrame** object for data manipulation with
+  integrated indexing;
+
+- Tools for **reading and writing data** between in-memory data structures and
+  different formats: CSV and text files, Microsoft Excel, SQL databases, and
+  the fast HDF5 format;
+
+- Intelligent **data alignment** and integrated handling of **missing data**:
+  gain automatic label-based alignment in computations and easily manipulate
+  messy data into an orderly form;
+
+- Flexible **reshaping** and pivoting of data sets;
+
+- Intelligent label-based **slicing**, **fancy indexing**, and **subsetting**
+  of large data sets;
+
+- Columns can be inserted and deleted from data structures for **size
+  mutability**;
+
+- Aggregating or transforming data with a powerful **group by** engine
+  allowing split-apply-combine operations on data sets;
+
+- High performance **merging and joining** of data sets;
+
+- **Hierarchical axis indexing** provides an intuitive way of working with
+  high-dimensional data in a lower-dimensional data structure;
+
+- **Time series**-functionality: date range generation and frequency
+  conversion, moving window statistics, date shifting and lagging.
+  Even create domain-specific time offsets and join time
+  series without losing data;
+
+- Highly **optimized for performance**, with critical code paths written in
+  [Cython](https://cython.org) or C.
+
+- Python with *pandas* is in use in a wide variety of **academic and
+  commercial** domains, including Finance, Neuroscience, Economics,
+  Statistics, Advertising, Web Analytics, and more.
+
+## Mission
+
+_pandas_ aims to be the fundamental high-level building block for doing practical,
+real world data analysis in Python.
+Additionally, it has the broader goal of becoming the most powerful and flexible
+open source data analysis / manipulation tool available in any language.
+
+## Vision
+
+A world where data analytics and manipulation software is:
+
+- Accessible to everyone
+- Free for users to use and modify
+- Flexible
+- Powerful
+- Easy to use
+- Fast
+
+## Values
+
+Is in the core of _pandas_ to be respectful and welcoming with everybody,
+users, contributors and the broader community. Regardless of level of experience,
+gender, gender identity and expression, sexual orientation, disability,
+personal appearance, body size, race, ethnicity, age, religion, or nationality.
diff --git a/pandas/web/pandas/about/roadmap.md b/pandas/web/pandas/about/roadmap.md
new file mode 100644
index 0000000000000..aba95ec2c03fc
--- /dev/null
+++ b/pandas/web/pandas/about/roadmap.md
@@ -0,0 +1,197 @@
+# Roadmap
+
+This page provides an overview of the major themes in pandas'
+development. Each of these items requires a relatively large amount of
+effort to implement. These may be achieved more quickly with dedicated
+funding or interest from contributors.
+
+An item being on the roadmap does not mean that it will *necessarily*
+happen, even with unlimited funding. During the implementation period we
+may discover issues preventing the adoption of the feature.
+
+Additionally, an item *not* being on the roadmap does not exclude it
+from inclusion in pandas. The roadmap is intended for larger,
+fundamental changes to the project that are likely to take months or
+years of developer time. Smaller-scoped items will continue to be
+tracked on our [issue tracker](https://github.com/pandas-dev/pandas/issues).
+
+The roadmap is defined as a set of major enhancement proposals named PDEPs.
+For more information about PDEPs, and how to submit one, please refer to
+[PEDP-1]({{ base_url }}pdeps/0001-purpose-and-guidelines.html).
+
+## PDEPs
+
+{% for pdep_type in ["Under discussion", "Accepted", "Implemented", "Rejected"] %}
+
+<h3 id="pdeps-{{pdep_type}}">{{ pdep_type.replace("_", " ").capitalize() }}</h3>
+
+<ul>
+{% for pdep in pdeps[pdep_type] %}
+    <li><a href="{% if not pdep.url.startswith("http") %}{{ base_url }}{% endif %}{{ pdep.url }}">{{ pdep.title }}</a></li>
+{% else %}
+    <li>There are currently no PDEPs with this status</li>
+{% endfor %}
+</ul>
+
+{% endfor %}
+
+## Roadmap points pending a PDEP
+
+<div class="alert alert-warning" role="alert">
+  pandas is in the process of moving roadmap points to PDEPs (implemented in
+  August 2022). During the transition, some roadmap points will exist as PDEPs,
+  while others will exist as sections below.
+</div>
+
+### Extensibility
+
+Pandas `extending.extension-types` allow
+for extending NumPy types with custom data types and array storage.
+Pandas uses extension types internally, and provides an interface for
+3rd-party libraries to define their own custom data types.
+
+Many parts of pandas still unintentionally convert data to a NumPy
+array. These problems are especially pronounced for nested data.
+
+We'd like to improve the handling of extension arrays throughout the
+library, making their behavior more consistent with the handling of
+NumPy arrays. We'll do this by cleaning up pandas' internals and
+adding new methods to the extension array interface.
+
+### String data type
+
+Currently, pandas stores text data in an `object` -dtype NumPy array.
+The current implementation has two primary drawbacks: First, `object`
+-dtype is not specific to strings: any Python object can be stored in an
+`object` -dtype array, not just strings. Second: this is not efficient.
+The NumPy memory model isn't especially well-suited to variable width
+text data.
+
+To solve the first issue, we propose a new extension type for string
+data. This will initially be opt-in, with users explicitly requesting
+`dtype="string"`. The array backing this string dtype may initially be
+the current implementation: an `object` -dtype NumPy array of Python
+strings.
+
+To solve the second issue (performance), we'll explore alternative
+in-memory array libraries (for example, Apache Arrow). As part of the
+work, we may need to implement certain operations expected by pandas
+users (for example the algorithm used in, `Series.str.upper`). That work
+may be done outside of pandas.
+
+### Apache Arrow interoperability
+
+[Apache Arrow](https://arrow.apache.org) is a cross-language development
+platform for in-memory data. The Arrow logical types are closely aligned
+with typical pandas use cases.
+
+We'd like to provide better-integrated support for Arrow memory and
+data types within pandas. This will let us take advantage of its I/O
+capabilities and provide for better interoperability with other
+languages and libraries using Arrow.
+
+### Decoupling of indexing and internals
+
+The code for getting and setting values in pandas' data structures
+needs refactoring. In particular, we must clearly separate code that
+converts keys (e.g., the argument to `DataFrame.loc`) to positions from
+code that uses these positions to get or set values. This is related to
+the proposed BlockManager rewrite. Currently, the BlockManager sometimes
+uses label-based, rather than position-based, indexing. We propose that
+it should only work with positional indexing, and the translation of
+keys to positions should be entirely done at a higher level.
+
+Indexing is a complicated API with many subtleties. This refactor will require care
+and attention. The following principles should inspire refactoring of indexing code and
+should result on cleaner, simpler, and more performant code.
+
+1. Label indexing must never involve looking in an axis twice for the same label(s).
+This implies that any validation step must either:
+
+  * limit validation to general features (e.g. dtype/structure of the key/index), or
+  * reuse the result for the actual indexing.
+
+2. Indexers must never rely on an explicit call to other indexers.
+For instance, it is OK to have some internal method of `.loc` call some
+internal method of `__getitem__` (or of their common base class),
+but never in the code flow of `.loc` should `the_obj[something]` appear.
+
+3. Execution of positional indexing must never involve labels (as currently, sadly, happens).
+That is, the code flow of a getter call (or a setter call in which the right hand side is non-indexed)
+to `.iloc` should never involve the axes of the object in any way.
+
+4. Indexing must never involve accessing/modifying values (i.e., act on `._data` or `.values`) more than once.
+The following steps must hence be clearly decoupled:
+
+  * find positions we need to access/modify on each axis
+  * (if we are accessing) derive the type of object we need to return (dimensionality)
+  * actually access/modify the values
+  * (if we are accessing) construct the return object
+
+5. As a corollary to the decoupling between 4.i and 4.iii, any code which deals on how data is stored
+(including any combination of handling multiple dtypes, and sparse storage, categoricals, third-party types)
+must be independent from code that deals with identifying affected rows/columns,
+and take place only once step 4.i is completed.
+
+  * In particular, such code should most probably not live in `pandas/core/indexing.py`
+  * ... and must not depend in any way on the type(s) of axes (e.g. no `MultiIndex` special cases)
+
+6. As a corollary to point 1.i, `Index` (sub)classes must provide separate methods for any desired validity check of label(s) which does not involve actual lookup,
+on the one side, and for any required conversion/adaptation/lookup of label(s), on the other.
+
+7. Use of trial and error should be limited, and anyway restricted to catch only exceptions
+which are actually expected (typically `KeyError`).
+
+  * In particular, code should never (intentionally) raise new exceptions in the `except` portion of a `try... exception`
+
+8. Any code portion which is not specific to setters and getters must be shared,
+and when small differences in behavior are expected (e.g. getting with `.loc` raises for
+missing labels, setting still doesn't), they can be managed with a specific parameter.
+
+### Numba-accelerated operations
+
+[Numba](https://numba.pydata.org) is a JIT compiler for Python code.
+We'd like to provide ways for users to apply their own Numba-jitted
+functions where pandas accepts user-defined functions (for example,
+`Series.apply`,
+`DataFrame.apply`,
+`DataFrame.applymap`, and in groupby and
+window contexts). This will improve the performance of
+user-defined-functions in these operations by staying within compiled
+code.
+
+### Documentation improvements
+
+We'd like to improve the content, structure, and presentation of the
+pandas documentation. Some specific goals include
+
+-   Overhaul the HTML theme with a modern, responsive design
+    (`15556`)
+-   Improve the "Getting Started" documentation, designing and writing
+    learning paths for users different backgrounds (e.g. brand new to
+    programming, familiar with other languages like R, already familiar
+    with Python).
+-   Improve the overall organization of the documentation and specific
+    subsections of the documentation to make navigation and finding
+    content easier.
+
+### Performance monitoring
+
+Pandas uses [airspeed velocity](https://asv.readthedocs.io/en/stable/)
+to monitor for performance regressions. ASV itself is a fabulous tool,
+but requires some additional work to be integrated into an open source
+project's workflow.
+
+The [asv-runner](https://github.com/asv-runner) organization, currently
+made up of pandas maintainers, provides tools built on top of ASV. We
+have a physical machine for running a number of project's benchmarks,
+and tools managing the benchmark runs and reporting on results.
+
+We'd like to fund improvements and maintenance of these tools to
+
+-   Be more stable. Currently, they're maintained on the nights and
+    weekends when a maintainer has free time.
+-   Tune the system for benchmarks to improve stability, following
+    <https://pyperf.readthedocs.io/en/latest/system.html>
+-   Build a GitHub bot to request ASV runs *before* a PR is merged.
+    Currently, the benchmarks are only run nightly.
diff --git a/pandas/web/pandas/about/sponsors.md b/pandas/web/pandas/about/sponsors.md
new file mode 100644
index 0000000000000..4473a16cfd590
--- /dev/null
+++ b/pandas/web/pandas/about/sponsors.md
@@ -0,0 +1,60 @@
+# Sponsors
+
+## NumFOCUS
+
+![](https://numfocus.org/wp-content/uploads/2018/01/optNumFocus_LRG.png)
+
+_pandas_ is a Sponsored Project of [NumFOCUS](https://numfocus.org/), a 501(c)(3) nonprofit charity in the United States.
+NumFOCUS provides _pandas_ with fiscal, legal, and administrative support to help ensure the
+health and sustainability of the project. Visit numfocus.org for more information.
+
+Donations to _pandas_ are managed by NumFOCUS. For donors in the United States, your gift is tax-deductible
+to the extent provided by law. As with any donation, you should consult with your tax adviser about your particular tax situation.
+
+## Become a sponsor
+
+As a free and open source project, _pandas_ relies on the support of the community of users for its development.
+If you work for an organization that uses and benefits from _pandas_, please consider supporting pandas. There
+are different ways, such as employing people to work on pandas, funding the project, or becoming a
+[NumFOCUS sponsor](https://numfocus.org/sponsors) to support the broader ecosystem. Please contact us at
+[admin@numfocus.org](mailto:admin@numfocus.org) to discuss.
+
+## Institutional partners
+
+Institutional partners are companies and universities that support the project by employing contributors.
+Current institutional partners include:
+
+<ul>
+    {% for company in sponsors.active if company.kind == "partner" %}
+        <li><a href="{{ company.url }}">{{ company.name }}</a>: {{ company.description }}</li>
+    {% endfor %}
+</ul>
+
+## Sponsors
+
+Sponsors are organizations that provide funding for pandas. Current sponsors include:
+
+<ul>
+    {% for company in sponsors.active if company.kind == "regular" %}
+        <li><a href="{{ company.url }}">{{ company.name }}</a>: {{ company.description }}</li>
+    {% endfor %}
+</ul>
+
+## In-kind sponsors
+
+In-kind sponsors are organizations that support pandas development with goods or services.
+Current in-kind sponsors include:
+
+<ul>
+    {% for company in sponsors.inkind %}
+        <li><a href="{{ company.url }}">{{ company.name }}</a>: {{ company.description }}</li>
+    {% endfor %}
+</ul>
+
+## Past institutional partners
+
+<ul>
+    {% for company in sponsors.past if company.kind == "partner" %}
+        <li><a href="{{ company.url }}">{{ company.name }}</a></li>
+    {% endfor %}
+</ul>
diff --git a/pandas/web/pandas/about/team.md b/pandas/web/pandas/about/team.md
new file mode 100644
index 0000000000000..7a19fd7af6595
--- /dev/null
+++ b/pandas/web/pandas/about/team.md
@@ -0,0 +1,85 @@
+# Team
+
+## Contributors
+
+_pandas_ is made with love by more than [2,000 volunteer contributors](https://github.com/pandas-dev/pandas/graphs/contributors).
+
+If you want to support pandas development, you can find information in the [donations page]({{ base_url }}donate.html).
+
+## Active maintainers
+
+<div class="card-group maintainers">
+    {% for username in maintainers.active %}
+        {% set person = maintainers.github_info.get(username) %}
+        <div class="card">
+            <img class="card-img-top" alt="" src="{{ person.avatar_url }}"/>
+            <div class="card-body">
+                <h6 class="card-title">
+                    {% if person.blog %}
+                        <a href="{{ person.blog }}">
+                            {{ person.name or person.login }}
+                        </a>
+                    {% else %}
+                        {{ person.name or person.login }}
+                    {% endif %}
+                </h6>
+                <p class="card-text small"><a href="{{ person.html_url }}">{{ person.login }}</a></p>
+            </div>
+        </div>
+    {% endfor %}
+</div>
+
+## Diversity and Inclusion
+
+> _pandas_ expressly welcomes and encourages contributions from anyone who faces under-representation, discrimination in the technology industry
+> or anyone willing to increase the diversity of our team.
+> We have identified visible gaps and obstacles in sustaining diversity and inclusion in the open-source communities and we are proactive in increasing
+> the diversity of our team.
+> We have a [code of conduct]({{ base_url }}community/coc.html) to ensure a friendly and welcoming environment.
+> Please send an email to [pandas-code-of-conduct-committee](mailto:pandas-coc@googlegroups.com), if you think we can do a
+> better job at achieving this goal.
+
+## Governance
+
+The project governance is available in the [project governance page]({{ base_url }}about/governance.html).
+
+## Workgroups
+
+{% for k, workgroup in workgroups.items() %}
+
+### {{ workgroup.name }}
+
+<ul>
+    <li><b>Contact:</b>
+        <a id="{{ workgroup.name|replace(' ', '-') }}" href="mailto:asp.{{ workgroup.contact }}">asp.{{ workgroup.contact }}</a>
+        <script TYPE="text/javascript">
+            var mail_tag_id = '{{ workgroup.name|replace(' ', '-') }}';
+            var mail_tag_element = document.getElementById( mail_tag_id );
+            mail_tag_element.innerHTML = mail_tag_element.innerHTML.replace(/^asp./, "");
+            mail_tag_element.setAttribute('href', "mailto:"+mail_tag_element.innerHTML);
+        </script>
+    </li>
+    <li><b>Responsibilities:</b> {{ workgroup.responsibilities }}</li>
+    <li><b>Members:</b>
+        <ul>
+            {% for person in workgroup.members %}
+                <li>{{ person }}{% if loop.first %} (lead){% endif %}</li>
+            {% endfor %}
+        </ul>
+    </li>
+</ul>
+
+{% endfor %}
+
+## Inactive maintainers
+
+<ul>
+    {% for username in maintainers.inactive %}
+        {% set person = maintainers.github_info.get(username) %}
+        <li>
+            <a href="{{ person.blog or person.html_url }}">
+                {{ person.name or person.login }}
+            </a>
+        </li>
+    {% endfor %}
+</ul>
diff --git a/pandas/web/pandas/community/benchmarks.md b/pandas/web/pandas/community/benchmarks.md
new file mode 100644
index 0000000000000..1e63832a5a2ba
--- /dev/null
+++ b/pandas/web/pandas/community/benchmarks.md
@@ -0,0 +1,79 @@
+# Benchmarks
+
+Benchmarks are tests to measure the performance of pandas. There are two different
+kinds of benchmarks relevant to pandas:
+
+* Internal pandas benchmarks to measure speed and memory usage over time
+* Community benchmarks comparing the speed or memory usage of different tools at
+  doing the same job
+
+## pandas benchmarks
+
+pandas benchmarks are implemented in the [asv_bench](https://github.com/pandas-dev/pandas/tree/main/asv_bench)
+directory of our repository. The benchmarks are implemented for the
+[airspeed velocity](https://asv.readthedocs.io/en/v0.6.1/) (asv for short) framework.
+
+The benchmarks can be run locally by any pandas developer. This can be done
+with the `asv run` command, and it can be useful to detect if local changes have
+an impact in performance, by running the benchmarks before and after the changes.
+More information on running the performance test suite is found
+[here](https://pandas.pydata.org/docs/dev/development/contributing_codebase.html#running-the-performance-test-suite).
+
+Note that benchmarks are not deterministic, and running in different hardware or
+running in the same hardware with different levels of stress have a big impact in
+the result. Even running the benchmarks with identical hardware and almost identical
+conditions produces significant differences when running the same exact code.
+
+## pandas benchmarks servers
+
+We currently have two physical servers running the benchmarks of pandas for every
+(or almost every) commit to the `main` branch. The servers run independently from
+each other. The original server has been running for a long time, and it is physically
+located with one of the pandas maintainers. The newer server is in a datacenter
+kindly sponsored by [OVHCloud](https://www.ovhcloud.com/). More information about
+pandas sponsors, and how your company can support the development of pandas is
+available at the [pandas sponsors]({{ base_url }}about/sponsors.html) page.
+
+Results of the benchmarks are available at:
+
+- Original server: [asv](https://asv-runner.github.io/asv-collection/pandas/)
+- OVH server: [asv](https://pandas.pydata.org/benchmarks/asv/) (benchmarks results can
+  also be visualized in this [Conbench PoC](http://57.128.112.95:5000/)
+
+### Original server configuration
+
+The machine can be configured with the Ansible playbook in
+[tomaugspurger/asv-runner](https://github.com/tomaugspurger/asv-runner).
+The results are published to another GitHub repository,
+[tomaugspurger/asv-collection](https://github.com/tomaugspurger/asv-collection).
+
+The benchmarks are scheduled by [Airflow](https://airflow.apache.org/).
+It has a dashboard for viewing and debugging the results.
+You’ll need to setup an SSH tunnel to view them:
+
+```
+ssh -L 8080:localhost:8080 pandas@panda.likescandy.com
+```
+
+### OVH server configuration
+
+The server used to run the benchmarks has been configured to reduce system
+noise and maximize the stability of the benchmarks times.
+
+The details on how the server is configured can be found in the
+[pandas-benchmarks repository](https://github.com/pandas-dev/pandas-benchmarks).
+There is a quick summary here:
+
+- CPU isolation: Avoid user space tasks to execute in the same CPU as benchmarks, possibly interrupting them during the execution (include all virtual CPUs using a physical core)
+- NoHZ: Stop the kernel tick that enables context switching in the isolated CPU
+- IRQ affinity: Ban benchmarks CPU to avoid many (but not all) kernel interruption in the isolated CPU
+- TurboBoost: Disable CPU scaling based on high CPU demand
+- P-States: Use "performance" governor to disable P-States and CPU frequency changes based on them
+- C-States: Set C-State to 0 and disable changes to avoid slower CPU after system inactivity
+
+## Community benchmarks
+
+The main benchmarks comparing dataframe tools that include pandas are:
+
+- [DuckDB (former H2O.ai) benchmarks](https://duckdblabs.github.io/db-benchmark/)
+- [TPCH benchmarks](https://pola.rs/posts/benchmarks/)
diff --git a/pandas/web/pandas/community/blog/2019-user-survey.md b/pandas/web/pandas/community/blog/2019-user-survey.md
new file mode 100644
index 0000000000000..821fdd01acf65
--- /dev/null
+++ b/pandas/web/pandas/community/blog/2019-user-survey.md
@@ -0,0 +1,172 @@
+Title: 2019 pandas user survey
+Date: 2019-08-22
+
+<style type="text/css">
+table td {
+    background: none;
+}
+
+table tr.even td {
+    background: none;
+}
+
+table {
+	text-shadow: none;
+}
+
+</style>
+
+# 2019 pandas user survey
+
+Pandas recently conducted a user survey to help guide future development.
+Thanks to everyone who participated! This post presents the high-level results.
+
+This analysis and the raw data can be found [on GitHub](https://github.com/pandas-dev/pandas-user-surveys) and run on Binder
+
+[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/pandas-dev/pandas-user-surveys/master?filepath=2019.ipynb)
+
+
+We had about 1250 responses over the 15 days we ran the survey in the summer of 2019.
+
+## About the Respondents
+
+There was a fair amount of representation across pandas experience and frequency of use, though the majority of respondents are on the more experienced side.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_4_0.png)
+
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_5_0.png)
+
+
+We included a few questions that were also asked in the [Python Developers Survey](https://www.jetbrains.com/research/python-developers-survey-2018/) so we could compare Pandas' population to Python's.
+
+90% of our respondents use Python as a primary language (compared with 84% from the PSF survey).
+
+
+
+
+
+    Yes    90.67%
+    No      9.33%
+    Name: Is Python your main language?, dtype: object
+
+
+
+Windows users are well represented (see [Steve Dower's talk](https://www.youtube.com/watch?v=uoI57uMdDD4) on this topic).
+
+
+
+
+
+    Linux      61.57%
+    Windows    60.21%
+    MacOS      42.75%
+    Name: What Operating Systems do you use?, dtype: object
+
+
+
+For environment isolation, [conda](https://conda.io/en/latest/) was the most popular.
+
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_13_0.png)
+
+
+Most respondents are Python 3 only.
+
+
+
+
+
+    3        92.39%
+    2 & 3     6.80%
+    2         0.81%
+    Name: Python 2 or 3?, dtype: object
+
+
+
+## Pandas APIs
+
+It can be hard for open source projects to know what features are actually used. We asked a few questions to get an idea.
+
+CSV and Excel are (for better or worse) the most popular formats.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_18_0.png)
+
+
+In preparation for a possible refactor of pandas internals, we wanted to get a sense for
+how common wide (100s of columns or more) DataFrames are.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_20_0.png)
+
+
+Pandas is slowly growing new extension types. Categoricals are the most popular,
+and the nullable integer type is already almost as popular as datetime with timezone.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_22_0.png)
+
+
+More and better examples seem to be a high-priority development item.
+Pandas recently received a NumFOCUS grant to improve our documentation,
+which we're using to write tutorial-style documentation, which should help
+meet this need.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_24_0.png)
+
+
+We also asked about specific, commonly-requested features.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_26_0.png)
+
+
+Of these, the clear standout is "scaling" to large datasets. A couple observations:
+
+1. Perhaps pandas' documentation should do a better job of promoting libraries that provide scalable dataframes (like [Dask](https://dask.org), [vaex](https://dask.org), and [modin](https://modin.readthedocs.io/en/latest/))
+2. Memory efficiency (perhaps from a native string data type, fewer internal copies, etc.) is a valuable goal.
+
+After that, the next-most critical improvement is integer missing values. Those were actually added in [Pandas 0.24](https://pandas.pydata.org/pandas-docs/stable/whatsnew/v0.24.0.html#optional-integer-na-support), but they're not the default, and there's still some incompatibilities with the rest of pandas API.
+
+Pandas is a less conservative library than, say, NumPy. We're approaching 1.0, but on the way we've made many deprecations and some outright API breaking changes. Fortunately, most people are OK with the tradeoff.
+
+
+
+
+
+    Yes    94.89%
+    No      5.11%
+    Name: Is Pandas stable enough for you?, dtype: object
+
+
+
+There's a perception (which is shared by many of the pandas maintainers) that the pandas API is too large. To measure that, we asked whether users thought that pandas' API was too large, too small, or just right.
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_31_0.png)
+
+
+Finally, we asked for an overall satisfaction with the library, from 1 (not very unsatisfied) to 5 (very satisfied).
+
+
+
+![png]({{ base_url }}/static/img/blog/2019-user-survey/2019_33_0.png)
+
+
+Most people are very satisfied. The average response is 4.39. I look forward to tracking this number over time.
+
+If you're analyzing the raw data, be sure to share the results with us [@pandas_dev](https://twitter.com/pandas_dev).
diff --git a/pandas/web/pandas/community/blog/asv-pandas-grant.md b/pandas/web/pandas/community/blog/asv-pandas-grant.md
new file mode 100644
index 0000000000000..205c2222fec94
--- /dev/null
+++ b/pandas/web/pandas/community/blog/asv-pandas-grant.md
@@ -0,0 +1,141 @@
+Title: Write up of the NumFOCUS grant to improve pandas benchmarks and diversity
+Date: 2022-04-01
+
+# Write up of the NumFOCUS grant to improve pandas benchmarks and diversity
+
+*By Lucy Jiménez and Dorothy Kabarozi B.*
+
+We want to share our experience working on **Improvements to the**
+**ASV benchmarking framework and diversity efforts** sponsored by
+[NumFOCUS](https://numfocus.org/) to the [pandas](https://pandas.pydata.org/)
+project.
+
+This grant focused on two aspects: the first one is to improve the
+[asv library](https://asv.readthedocs.io/en/stable/), a tool used by
+benchmarking Python packages and used by pandas; this project was
+unmaintained, and the codebase was quite old; additionally, it didn't
+adhere to modern standards, had Python 2 compatibility code that could
+be removed, and also the CI could be improved. The second aspect is
+encouraging more underrepresented groups to contribute to open source
+projects. This grant was held over 10 weeks, working around 20 hours a
+week. It was developed by Dorothy Kabarozi B. from Uganda and Lucy
+Jiménez from Colombia, under the mentoring of Marc Garcia.
+
+## Why were we part of the grant?
+
+Even when we come from different backgrounds, Dorothy from systems
+engineering and Lucy from computational chemistry, we have always been
+interested in participating and contributing to open source software
+projects. For that reason, we have been running the PyLadies meetups in
+our communities ([PyLadies Kampala](https://twitter.com/pyladieskla),
+[PyLadies Colombia](https://twitter.com/pyladies_co)) and have always
+been on the lookout for any opportunities that lead us to contribute.
+
+It all happened through Marc Garcia; he had put out a call ​through a post
+on social media to mentor ladies from diverse backgrounds. Dorothy got to
+be part of the pandas mentorship group. At the same time, Lucy was
+co-organizer of the SciPy Latam conference, and it is from here she met
+Marc, who was the speaker at that conference, and through this mutual
+connection, we were able to learn about this benchmarks grant.
+
+In brief, by attending conferences, meetups, and social media, you can
+make connections and links that will lead you to these opportunities.
+
+## Learning from the source code
+
+At the beginning of the grant, we started from the basics. We noticed that
+we could improve our skills in managing Git and GitHub. For example, we had
+some troubles with the git workflow, so we had to read and practice more
+about it. One of the valuable resources was the explanation from Marc about
+[how to make an open source contribution](https://tubedu.org/w/kjnHEg72j76StmSFmjzbnE),
+which we invite you to take a look at it.
+
+We learned a lot from the source code and gained immense knowledge about
+best practices and code quality through this grant. We have been working
+on: updating the code to improve the style to follow the PEP-8 guidelines,
+removing Python 2 compatibility code and six dependencies, and finding
+unused code and removing it. We also learned about GitHub actions, and we
+started building the CI on GitHub actions for the asv package; for that we
+have been working on add linting with Flake8, testing with pytest, building
+docs, and running CI on different python versions.
+
+Additionally, we were able to identify bugs in the source code, review
+pull request from other contributors, and create new issues, something we
+thought only maintainers could do but not contributors. Finally, not only
+is reviewing the code itself a learning experience, but also the structure
+and folder hierarchy in the project started to be more transparent.
+
+## Our experience
+
+For this grant, we had a fantastic Mentor, Marc Garcia. He was always
+willing to share his knowledge, explain unclear concepts and share helpful
+feedback. Whenever we would implement that feedback, it felt easier to work
+on more issues faster. We felt the growth from the time we started on this
+project, and we will carry it along as we contribute to more open source
+projects; this all goes back to Marc for his fantastic mentorship. It is
+also important to note that we received feedback from other contributors,
+stakeholders, and core devs during this process, which gave us a broader
+look at the work in open source projects.
+
+We also built a strong teamwork partnership. We helped each other a lot as
+we had numerous one-on-one calls to understand the tasks better. We always
+looked for ways to support each other from the technical side and encouraged
+each other when needed. For us, it was professional and human growth.
+
+## Running an open source software sprint
+
+The knowledge and experience acquired in this process allowed us to
+organize two virtual sprints. The events were carried out in the company
+of local PyLadies communities; the first one was on February 26th with
+[PyLadies Kampala](https://twitter.com/pyladieskla) and on March 21
+with [PyLadies Colombia](https://bit.ly/sprint-asv).
+
+While organizing these events, we learned how to organize and conduct a
+virtual sprint. Some participants in the sprint ultimately had no idea
+about open source, and it was great explaining open source concepts and
+taking them through the Git workflow. Finally, they were able to make their
+first contribution. We learned how to follow up on contributors, helping
+them along the way until their PRs were merged and by reviewing their
+contributions on GitHub.
+
+The most outstanding achievement was mentoring new contributors and
+sharing the knowledge acquired from this grant with others participants
+in our respective communities. Most new contributors after the experience
+have gone ahead to apply for outreach and the upcoming
+[Google Summer of Code](https://summerofcode.withgoogle.com/)
+to apply the skills they learned from these sprints.
+
+## Conclusion
+
+In conclusion, we learned a lot from this experience from the code part,
+the workflow on the open source projects, how to be resilient in difficult
+moments, and encouraging more women and people from our local communities
+to contribute to open source projects.
+
+Finally, if you want to be part of an open source project, we invite you
+to check out GitHub repos for different projects you are interested in and
+search for the easy issues to work on and get started. Also, you can contact
+the maintainers of the projects with specific questions, search for the
+open source communities in your country or contact us for more help.
+
+## Acknowledgments
+
+Many thanks to [NumFOCUS](https://numfocus.org/) for giving us this support
+through [Small Development Grants](https://numfocus.org/programs/small-development-grants)
+and Marc for the excellent mentoring he generously gave us throughout these
+weeks.
+
+We are looking forward to contributing more and impacting our communities
+and the open source community!
+
+___
+If you want to know more, please don't hesitate to connect with us through
+these channels:
+
+*Lucy Jiménez*
+* [Twitter](https://twitter.com/JimenezLucyJ)
+* [LinkedIn](https://www.linkedin.com/in/lucy-j/)
+
+*Dorothy Kabarozi*
+* [Twitter](https://twitter.com/kizdorothy)
+* [LinkedIn](https://www.linkedin.com/in/dorothy-kabarozi/)
diff --git a/pandas/web/pandas/community/blog/extension-arrays.md b/pandas/web/pandas/community/blog/extension-arrays.md
new file mode 100644
index 0000000000000..80a187bb3fc3c
--- /dev/null
+++ b/pandas/web/pandas/community/blog/extension-arrays.md
@@ -0,0 +1,218 @@
+Title: pandas extension arrays
+Date: 2019-01-04
+
+# pandas extension arrays
+
+Extensibility was a major theme in pandas development over the last couple of
+releases. This post introduces the pandas extension array interface: the
+motivation behind it and how it might affect you as a pandas user. Finally, we
+look at how extension arrays may shape the future of pandas.
+
+Extension Arrays are just one of the changes in pandas 0.24.0. See the
+[whatsnew][whatsnew] for a full changelog.
+
+## The Motivation
+
+Pandas is built on top of NumPy. You could roughly define a Series as a wrapper
+around a NumPy array, and a DataFrame as a collection of Series with a shared
+index. That's not entirely correct for several reasons, but I want to focus on
+the "wrapper around a NumPy array" part. It'd be more correct to say "wrapper
+around an array-like object".
+
+Pandas mostly uses NumPy's builtin data representation; we've restricted it in
+places and extended it in others. For example, pandas' early users cared greatly
+about timezone-aware datetimes, which NumPy doesn't support. So pandas
+internally defined a `DatetimeTZ` dtype (which mimics a NumPy dtype), and
+allowed you to use that dtype in `Index`, `Series`, and as a column in a
+`DataFrame`. That dtype carried around the tzinfo, but wasn't itself a valid
+NumPy dtype.
+
+As another example, consider `Categorical`. This actually composes *two* arrays:
+one for the `categories` and one for the `codes`. But it can be stored in a
+`DataFrame` like any other column.
+
+Each of these extension types pandas added is useful on its own, but carries a
+high maintenance cost. Large sections of the codebase need to be aware of how to
+handle a NumPy array or one of these other kinds of special arrays. This made
+adding new extension types to pandas very difficult.
+
+Anaconda, Inc. had a client who regularly dealt with datasets with IP addresses.
+They wondered if it made sense to add an [IPArray][IPArray] to pandas. In the
+end, we didn't think it passed the cost-benefit test for inclusion in pandas
+*itself*, but we were interested in defining an interface for third-party
+extensions to pandas. Any object implementing this interface would be allowed in
+pandas. I was able to write [cyberpandas][cyberpandas] outside of pandas, but it
+feels like using any other dtype built into pandas.
+
+## The Current State
+
+As of pandas 0.24.0, all of pandas' internal extension arrays (Categorical,
+Datetime with Timezone, Period, Interval, and Sparse) are now built on top of
+the ExtensionArray interface. Users shouldn't notice many changes. The main
+thing you'll notice is that things are cast to `object` dtype in fewer places,
+meaning your code will run faster and your types will be more stable. This
+includes storing `Period` and `Interval` data in `Series` (which were previously
+cast to object dtype).
+
+Additionally, we'll be able to add *new* extension arrays with relative ease.
+For example, 0.24.0 (optionally) solved one of pandas longest-standing pain
+points: missing values casting integer-dtype values to float.
+
+
+```python
+>>> int_ser = pd.Series([1, 2], index=[0, 2])
+>>> int_ser
+0    1
+2    2
+dtype: int64
+
+>>> int_ser.reindex([0, 1, 2])
+0    1.0
+1    NaN
+2    2.0
+dtype: float64
+```
+
+With the new [IntegerArray][IntegerArray] and nullable integer dtypes, we can
+natively represent integer data with missing values.
+
+```python
+>>> int_ser = pd.Series([1, 2], index=[0, 2], dtype=pd.Int64Dtype())
+>>> int_ser
+0    1
+2    2
+dtype: Int64
+
+>>> int_ser.reindex([0, 1, 2])
+0      1
+1    NaN
+2      2
+dtype: Int64
+```
+
+One thing it does slightly change how you should access the raw (unlabeled)
+arrays stored inside a Series or Index, which is occasionally useful. Perhaps
+the method you're calling only works with NumPy arrays, or perhaps you want to
+disable automatic alignment.
+
+In the past, you'd hear things like "Use `.values` to extract the NumPy array
+from a Series or DataFrame." If it were a good resource, they'd tell you that's
+not *entirely* true, since there are some exceptions. I'd like to delve into
+those exceptions.
+
+The fundamental problem with `.values` is that it serves two purposes:
+
+1. Extracting the array backing a Series, Index, or DataFrame
+2. Converting the Series, Index, or DataFrame to a NumPy array
+
+As we saw above, the "array" backing a Series or Index might not be a NumPy
+array, it may instead be an extension array (from pandas or a third-party
+library). For example, consider `Categorical`,
+
+```python
+>>> cat = pd.Categorical(['a', 'b', 'a'], categories=['a', 'b', 'c'])
+>>> ser = pd.Series(cat)
+>>> ser
+0    a
+1    b
+2    a
+dtype: category
+Categories (3, object): ['a', 'b', 'c']
+
+>>> ser.values
+[a, b, a]
+Categories (3, object): ['a', 'b', 'c']
+```
+
+In this case `.values` is a Categorical, not a NumPy array. For period-dtype
+data, `.values` returns a NumPy array of `Period` objects, which is expensive to
+create. For timezone-aware data, `.values` converts to UTC and *drops* the
+timezone info. These kind of surprises (different types, or expensive or lossy
+conversions) stem from trying to shoehorn these extension arrays into a NumPy
+array. But the entire point of an extension array is for representing data NumPy
+*can't* natively represent.
+
+To solve the `.values` problem, we've split its roles into two dedicated methods:
+
+1. Use `.array` to get a zero-copy reference to the underlying data
+2. Use `.to_numpy()` to get a (potentially expensive, lossy) NumPy array of the
+   data.
+
+So with our Categorical example,
+
+```python
+>>> ser.array
+[a, b, a]
+Categories (3, object): ['a', 'b', 'c']
+
+>>> ser.to_numpy()
+array(['a', 'b', 'a'], dtype=object)
+```
+
+To summarize:
+
+- `.array` will *always* be a an ExtensionArray, and is always a zero-copy
+   reference back to the data.
+- `.to_numpy()` is *always* a NumPy array, so you can reliably call
+   ndarray-specific methods on it.
+
+You shouldn't ever need `.values` anymore.
+
+## Possible Future Paths
+
+Extension Arrays open up quite a few exciting opportunities. Currently, pandas
+represents string data using Python objects in a NumPy array, which is slow.
+Libraries like [Apache Arrow][arrow] provide native support for variable-length
+strings, and the [Fletcher][fletcher] library provides pandas extension arrays
+for Arrow arrays. It will allow [GeoPandas][geopandas] to store geometry data
+more efficiently. Pandas (or third-party libraries) will be able to support
+nested data, data with units, geo data, GPU arrays. Keep an eye on the
+[pandas ecosystem][eco] page, which will keep track of third-party extension
+arrays. It's an exciting time for pandas development.
+
+## Other Thoughts
+
+I'd like to emphasize that this is an *interface*, and not a concrete array
+implementation. We are *not* reimplementing NumPy here in pandas. Rather, this
+is a way to take any array-like data structure (one or more NumPy arrays, an
+Apache Arrow array, a CuPy array) and place it inside a DataFrame. I think
+getting pandas out of the array business, and instead thinking about
+higher-level tabular data things, is a healthy development for the project.
+
+This works perfectly with NumPy's [`__array_ufunc__`][ufunc] protocol and
+[NEP-18][nep18]. You'll be able to use the familiar NumPy API on objects that
+aren't backed by NumPy memory.
+
+## Upgrade
+
+These new goodies are all available in the recently released pandas 0.24.
+
+conda:
+
+    conda install -c conda-forge pandas
+
+pip:
+
+    pip install --upgrade pandas
+
+As always, we're happy to hear feedback on the [mailing list][ml],
+[@pandas-dev][twitter], or [issue tracker][tracker].
+
+Thanks to the many contributors, maintainers, and [institutional
+partners][partners] involved in the pandas community.
+
+
+[IPArray]: https://github.com/pandas-dev/pandas/issues/18767
+[cyberpandas]: https://cyberpandas.readthedocs.io
+[IntegerArray]: http://pandas.pydata.org/pandas-docs/version/0.24/reference/api/pandas.arrays.IntegerArray.html
+[fletcher]: https://github.com/xhochy/fletcher
+[arrow]: https://arrow.apache.org
+[ufunc]: https://numpy.org/neps/nep-0013-ufunc-overrides.html
+[nep18]: https://www.numpy.org/neps/nep-0018-array-function-protocol.html
+[ml]: https://mail.python.org/mailman/listinfo/pandas-dev
+[twitter]: https://twitter.com/pandas_dev
+[tracker]: https://github.com/pandas-dev/pandas/issues
+[partners]: https://pandas.pydata.org/about/sponsors.html
+[eco]: http://pandas.pydata.org/pandas-docs/stable/ecosystem.html#extension-data-types
+[whatsnew]: http://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.0.html
+[geopandas]: https://github.com/geopandas/geopandas
diff --git a/pandas/web/pandas/community/blog/index.html b/pandas/web/pandas/community/blog/index.html
new file mode 100644
index 0000000000000..6f0bee2e11407
--- /dev/null
+++ b/pandas/web/pandas/community/blog/index.html
@@ -0,0 +1,14 @@
+{% extends "layout.html" %}
+
+{% block body %}
+    {% for post in blog.posts %}
+        <div class="card">
+            <div class="card-body">
+                <h5 class="card-title"><a href="{% if not post.link.startswith("http") %}{{ base_url }}{% endif %}{{ post.link }}" target="_blank">{{ post.title }}</a></h5>
+                <h6 class="card-subtitle text-muted small mb-4">Source: {{ post.feed }} | Author: {{ post.author }} | Published: {{ post.published.strftime("%b %d, %Y") }}</h6>
+                <div class="card-text mb-2">{{ post.summary }}</div>
+                <a class="card-link small" href="{% if not post.link.startswith("http") %}{{ base_url }}{% endif %}{{ post.link }}" target="_blank">Read more</a>
+            </div>
+        </div>
+    {% endfor %}
+{% endblock %}
diff --git a/pandas/web/pandas/community/blog/pandas-1.0.md b/pandas/web/pandas/community/blog/pandas-1.0.md
new file mode 100644
index 0000000000000..d190ed6e897b3
--- /dev/null
+++ b/pandas/web/pandas/community/blog/pandas-1.0.md
@@ -0,0 +1,31 @@
+Title: pandas 1.0
+Date: 2020-01-29
+
+# pandas 1.0
+
+Today pandas celebrates its 1.0.0 release. In many ways this is just a normal release with a host of new features, performance improvements, and bug fixes, which are documented in our [release notes](https://pandas.pydata.org/pandas-docs/version/1.0.0/whatsnew/v1.0.0.html). But it’s also something a bit more — a milestone for the project beyond just the commits. We wanted to take some time to reflect on where we've been and where we're going.
+
+## Reflections
+
+The world of scientific Python has changed a lot since pandas was started.  In 2011, [the ecosystem was fragmented](https://wesmckinney.com/blog/a-roadmap-for-rich-scientific-data-structures-in-python/): a standard *rich* data structure for statistics and data science had yet to emerge. This echos a similar story for NumPy, which consolidated array efforts that were [previously fragmented](https://numpy.org/old_array_packages.html).
+
+Over the subsequent years, pandas emerged as a *de facto* standard. It’s used by data scientists and analysts and as a data structure for other libraries to build on top of. StackOverflow [cited pandas](https://stackoverflow.blog/2017/09/14/python-growing-quickly/) as one of the reasons for Python being the fastest growing major programming language.
+
+![Growth of pandas](https://149351115.v2.pressablecdn.com/wp-content/uploads/2017/09/related_tags_over_time-1-1000x1000.png)
+
+Today, the ecosystem is in another phase of exploration.
+Several new DataFrame implementations are cropping up to fill needs not met by pandas.
+We're [working with those projects](https://datapythonista.me/blog/dataframe-summit-at-euroscipy.html) to establish shared standards and semantics for rich data structures.
+
+## Community and Project Health
+
+This release cycle is the first to involve any kind of grant funding for pandas. [Pandas received funding](https://chanzuckerberg.com/eoss/proposals/) as part of the CZI’s [*Essential Open Source Software for Science*](https://medium.com/@cziscience/the-invisible-foundations-of-biomedicine-4ab7f8d4f5dd) [program](https://medium.com/@cziscience/the-invisible-foundations-of-biomedicine-4ab7f8d4f5dd). The pandas project relies overwhelmingly on volunteer contributors. These volunteer contributions are shepherded and augmented by some maintainers who are given time from their employers — our [institutional partners](../about/sponsors.html). The largest work item in our grant award was library maintenance, which specifically includes working with community members to address our large backlog of open issues and pull requests.
+
+While a “1.0.0” version might seem arbitrary or anti-climactic (given that pandas as a codebase is nearly 12 years old), we see it as a symbolic milestone celebrating the growth of our core developer team and depth of our contributor base.  Few open source projects are ever truly “done” and pandas is no different. We recognize the essential role that pandas now occupies, and we intend to continue to evolve the project and adapt to the needs of the world’s data wranglers.
+
+## Going Forward
+
+Our [roadmap](https://pandas.pydata.org/pandas-docs/version/1.0.0/development/roadmap.html) contains an up-to-date listing of where we see the project heading over the next few years.
+Needless to say, there's still plenty to do.
+
+Check out the [release notes](https://pandas.pydata.org/pandas-docs/version/1.0.0/whatsnew/v1.0.0.html) and visit the [installation page](https://pandas.pydata.org/pandas-docs/version/1.0.0/getting_started/install.html) for instructions on updating to pandas 1.0.
diff --git a/pandas/web/pandas/community/coc.md b/pandas/web/pandas/community/coc.md
new file mode 100644
index 0000000000000..22cd77859c557
--- /dev/null
+++ b/pandas/web/pandas/community/coc.md
@@ -0,0 +1,65 @@
+# Code of conduct
+
+As contributors and maintainers of this project, and in the interest of
+fostering an open and welcoming community, we pledge to respect all people who
+contribute through reporting issues, posting feature requests, updating
+documentation, submitting pull requests or patches, and other activities.
+
+We are committed to making participation in this project a harassment-free
+experience for everyone, regardless of level of experience, gender, gender
+identity and expression, sexual orientation, disability, personal appearance,
+body size, race, ethnicity, age, religion, or nationality.
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery
+* Personal attacks
+* Trolling or insulting/derogatory comments
+* Public or private harassment
+* Publishing other's private information, such as physical or electronic
+  addresses, without explicit permission
+* Other unethical or unprofessional conduct
+
+Furthermore, we encourage inclusive behavior - for example,
+please don't say “hey guys!” but “hey everyone!”.
+
+Project maintainers have the right and responsibility to remove, edit, or
+reject comments, commits, code, wiki edits, issues, and other contributions
+that are not aligned to this Code of Conduct, or to ban temporarily or
+permanently any contributor for other behaviors that they deem inappropriate,
+threatening, offensive, or harmful.
+
+By adopting this Code of Conduct, project maintainers commit themselves to
+fairly and consistently applying these principles to every aspect of managing
+this project. Project maintainers who do not follow or enforce the Code of
+Conduct may be permanently removed from the project team.
+
+This Code of Conduct applies both within project spaces and in public spaces
+when an individual is representing the project or its community.
+
+A working group of community members is committed to promptly addressing any
+reported issues. The working group is made up of pandas contributors and users.
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported by contacting the working group by e-mail (pandas-coc@googlegroups.com).
+Messages sent to this e-mail address will not be publicly visible but only to
+the working group members. The working group currently includes
+
+<ul>
+    {% for person in maintainers.coc %}
+    <li>{{ person }}</li>
+    {% endfor %}
+</ul>
+
+All complaints will be reviewed and investigated and will result in a response
+that is deemed necessary and appropriate to the circumstances. Maintainers are
+obligated to maintain confidentiality with regard to the reporter of an
+incident.
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage],
+version 1.3.0, available at
+[https://www.contributor-covenant.org/version/1/3/0/][version],
+and the [Swift Code of Conduct][swift].
+
+[homepage]: https://www.contributor-covenant.org
+[version]: https://www.contributor-covenant.org/version/1/3/0/
+[swift]: https://swift.org/community/#code-of-conduct
diff --git a/pandas/web/pandas/community/ecosystem.md b/pandas/web/pandas/community/ecosystem.md
new file mode 100644
index 0000000000000..2ad8d6243db55
--- /dev/null
+++ b/pandas/web/pandas/community/ecosystem.md
@@ -0,0 +1,739 @@
+# Ecosystem
+
+Increasingly, packages are being built on top of pandas to address
+specific needs in data preparation, analysis and visualization. This is
+encouraging because it means pandas is not only helping users to handle
+their data tasks but also that it provides a better starting point for
+developers to build powerful and more focused data tools. The creation
+of libraries that complement pandas' functionality also allows pandas
+development to remain focused around its original requirements.
+
+This is a community-maintained list of projects that build on pandas in order
+to provide tools in the PyData space. The pandas core development team does not necessarily endorse any particular project on this list or have any knowledge of the maintenance status of any particular library.
+
+For a more complete list of projects that depend on pandas, see the [libraries.io usage page for
+pandas](https://libraries.io/pypi/pandas/usage) or [search pypi for
+pandas](https://pypi.org/search/?q=pandas).
+
+We'd like to make it easier for users to find these projects, if you
+know of other substantial projects that you feel should be on this list,
+please let us know.
+
+## Statistics and machine learning
+
+### [Statsmodels](https://www.statsmodels.org/)
+
+Statsmodels is the prominent Python "statistics and econometrics
+library" and it has a long-standing special relationship with pandas.
+Statsmodels provides powerful statistics, econometrics, analysis and
+modeling functionality that is out of pandas' scope. Statsmodels
+leverages pandas objects as the underlying data container for
+computation.
+
+### [Featuretools](https://github.com/alteryx/featuretools/)
+
+Featuretools is a Python library for automated feature engineering built
+on top of pandas. It excels at transforming temporal and relational
+datasets into feature matrices for machine learning using reusable
+feature engineering "primitives". Users can contribute their own
+primitives in Python and share them with the rest of the community.
+
+### [Compose](https://github.com/alteryx/compose)
+
+Compose is a machine learning tool for labeling data and prediction engineering.
+It allows you to structure the labeling process by parameterizing
+prediction problems and transforming time-driven relational data into
+target values with cutoff times that can be used for supervised learning.
+
+### [STUMPY](https://github.com/TDAmeritrade/stumpy)
+
+STUMPY is a powerful and scalable Python library for modern time series analysis.
+At its core, STUMPY efficiently computes something called a
+[matrix profile](https://stumpy.readthedocs.io/en/latest/Tutorial_The_Matrix_Profile.html),
+which can be used for a wide variety of time series data mining tasks.
+
+## Visualization
+
+### [Altair](https://altair-viz.github.io/)
+
+Altair is a declarative statistical visualization library for Python.
+With Altair, you can spend more time understanding your data and its
+meaning. Altair's API is simple, friendly and consistent and built on
+top of the powerful Vega-Lite JSON specification. This elegant
+simplicity produces beautiful and effective visualizations with a
+minimal amount of code. Altair works with Pandas DataFrames.
+
+### [Bokeh](https://docs.bokeh.org)
+
+Bokeh is a Python interactive visualization library for large datasets
+that natively uses the latest web technologies. Its goal is to provide
+elegant, concise construction of novel graphics in the style of
+Protovis/D3, while delivering high-performance interactivity over large
+data to thin clients.
+
+[Pandas-Bokeh](https://github.com/PatrikHlobil/Pandas-Bokeh) provides a
+high level API for Bokeh that can be loaded as a native Pandas plotting
+backend via
+
+```
+pd.set_option("plotting.backend", "pandas_bokeh")
+```
+
+It is very similar to the matplotlib plotting backend, but provides
+interactive web-based charts and maps.
+
+### [pygwalker](https://github.com/Kanaries/pygwalker)
+
+PyGWalker is an interactive data visualization and
+exploratory data analysis tool built upon Graphic Walker
+with support for visualization, cleaning, and annotation workflows.
+
+pygwalker can save interactively created charts
+to Graphic-Walker and Vega-Lite JSON.
+
+```
+import pygwalker as pyg
+pyg.walk(df)
+```
+
+### [seaborn](https://seaborn.pydata.org)
+
+Seaborn is a Python visualization library based on
+[matplotlib](https://matplotlib.org). It provides a high-level,
+dataset-oriented interface for creating attractive statistical graphics.
+The plotting functions in seaborn understand pandas objects and leverage
+pandas grouping operations internally to support concise specification
+of complex visualizations. Seaborn also goes beyond matplotlib and
+pandas with the option to perform statistical estimation while plotting,
+aggregating across observations and visualizing the fit of statistical
+models to emphasize patterns in a dataset.
+
+```
+import seaborn as sns
+sns.set_theme()
+```
+
+### [plotnine](https://github.com/has2k1/plotnine/)
+
+Hadley Wickham's [ggplot2](https://ggplot2.tidyverse.org/) is a
+foundational exploratory visualization package for the R language. Based
+on ["The Grammar of
+Graphics"](https://www.cs.uic.edu/~wilkinson/TheGrammarOfGraphics/GOG.html)
+it provides a powerful, declarative and extremely general way to
+generate bespoke plots of any kind of data.
+Various implementations to other languages are available.
+A good implementation for Python users is [has2k1/plotnine](https://github.com/has2k1/plotnine/).
+
+### [IPython Vega](https://github.com/vega/ipyvega)
+
+[IPython Vega](https://github.com/vega/ipyvega) leverages [Vega](https://github.com/vega/vega) to create plots within Jupyter Notebook.
+
+### [Plotly](https://plot.ly/python)
+
+[Plotly's](https://plot.ly/) [Python API](https://plot.ly/python/)
+enables interactive figures and web shareability. Maps, 2D, 3D, and
+live-streaming graphs are rendered with WebGL and
+[D3.js](https://d3js.org/). The library supports plotting directly from
+a pandas DataFrame and cloud-based collaboration. Users of [matplotlib,
+ggplot for Python, and
+Seaborn](https://plot.ly/python/matplotlib-to-plotly-tutorial/) can
+convert figures into interactive web-based plots. Plots can be drawn in
+[IPython Notebooks](https://plot.ly/ipython-notebooks/) , edited with R
+or MATLAB, modified in a GUI, or embedded in apps and dashboards. Plotly
+is free for unlimited sharing, and has
+[cloud](https://plot.ly/product/plans/),
+[offline](https://plot.ly/python/offline/), or
+[on-premise](https://plot.ly/product/enterprise/) accounts for private
+use.
+
+### [Lux](https://github.com/lux-org/lux)
+
+Lux is a Python library that facilitates fast and easy experimentation with data by automating the visual data exploration process. To use Lux, simply add an extra import alongside pandas:
+
+```python
+import lux
+import pandas as pd
+
+df = pd.read_csv("data.csv")
+df  # discover interesting insights!
+```
+
+By printing out a dataframe, Lux automatically [recommends a set of visualizations](https://raw.githubusercontent.com/lux-org/lux-resources/master/readme_img/demohighlight.gif) that highlights interesting trends and patterns in the dataframe. Users can leverage any existing pandas commands without modifying their code, while being able to visualize their pandas data structures (e.g., DataFrame, Series, Index) at the same time. Lux also offers a [powerful, intuitive language](https://lux-api.readthedocs.io/en/latest/source/guide/vis.html) that allow users to create  Altair, matplotlib, or Vega-Lite visualizations without having to think at the level of code.
+
+### [D-Tale](https://github.com/man-group/dtale)
+
+D-Tale is a lightweight web client for visualizing pandas data structures. It
+provides a rich spreadsheet-style grid which acts as a wrapper for a lot of
+pandas functionality (query, sort, describe, corr...) so users can quickly
+manipulate their data. There is also an interactive chart-builder using Plotly
+Dash allowing users to build nice portable visualizations. D-Tale can be
+invoked with the following command
+
+```python
+import dtale
+
+dtale.show(df)
+```
+
+D-Tale integrates seamlessly with Jupyter notebooks, Python terminals, Kaggle
+& Google Colab. Here are some demos of the [grid](http://alphatechadmin.pythonanywhere.com/dtale/main/1).
+
+### [hvplot](https://hvplot.holoviz.org/index.html)
+
+hvPlot is a high-level plotting API for the PyData ecosystem built on [HoloViews](https://holoviews.org/).
+It can be loaded as a native pandas plotting backend via
+
+```python
+pd.set_option("plotting.backend", "hvplot")
+```
+
+## IDE
+
+### [IPython](https://ipython.org/documentation.html)
+
+IPython is an interactive command shell and distributed computing
+environment. IPython tab completion works with Pandas methods and also
+attributes like DataFrame columns.
+
+### [Jupyter Notebook / Jupyter Lab](https://jupyter.org)
+
+Jupyter Notebook is a web application for creating Jupyter notebooks. A
+Jupyter notebook is a JSON document containing an ordered list of
+input/output cells which can contain code, text, mathematics, plots and
+rich media. Jupyter notebooks can be converted to a number of open
+standard output formats (HTML, HTML presentation slides, LaTeX, PDF,
+ReStructuredText, Markdown, Python) through 'Download As' in the web
+interface and `jupyter convert` in a shell.
+
+Pandas DataFrames implement `_repr_html_` and `_repr_latex` methods which
+are utilized by Jupyter Notebook for displaying (abbreviated) HTML or
+LaTeX tables. LaTeX output is properly escaped. (Note: HTML tables may
+or may not be compatible with non-HTML Jupyter output formats.)
+
+See [Options and Settings](https://pandas.pydata.org/docs/user_guide/options.html)
+for pandas `display.` settings.
+
+### [Spyder](https://www.spyder-ide.org/)
+
+Spyder is a cross-platform PyQt-based IDE combining the editing,
+analysis, debugging and profiling functionality of a software
+development tool with the data exploration, interactive execution, deep
+inspection and rich visualization capabilities of a scientific
+environment like MATLAB or Rstudio.
+
+Its [Variable
+Explorer](https://docs.spyder-ide.org/current/panes/variableexplorer.html) allows
+users to view, manipulate and edit pandas `Index`, `Series`, and
+`DataFrame` objects like a "spreadsheet", including copying and
+modifying values, sorting, displaying a "heatmap", converting data
+types and more. Pandas objects can also be renamed, duplicated, new
+columns added, copied/pasted to/from the clipboard (as TSV), and
+saved/loaded to/from a file. Spyder can also import data from a variety
+of plain text and binary files or the clipboard into a new pandas
+DataFrame via a sophisticated import wizard.
+
+Most pandas classes, methods and data attributes can be autocompleted in
+Spyder's [Editor](https://docs.spyder-ide.org/current/panes/editor.html) and [IPython
+Console](https://docs.spyder-ide.org/current/panes/ipythonconsole.html), and Spyder's
+[Help pane](https://docs.spyder-ide.org/current/panes/help.html) can retrieve and
+render Numpydoc documentation on pandas objects in rich text with Sphinx
+both automatically and on-demand.
+
+### [marimo](https://marimo.io)
+
+marimo is a reactive notebook for Python and SQL that enhances productivity when working with dataframes. It provides several features to make data manipulation and visualization more interactive and fun:
+
+1. Rich, interactive displays: marimo can display pandas dataframes in interactive tables or charts with filtering and sorting capabilities.
+2. Data selection: Users can select data in tables or pandas-backed plots, and the selections are automatically sent to Python as pandas dataframes.
+3. No-code transformations: Users can interactively transform pandas dataframes using a GUI, without writing code. The generated code can be copied and pasted into the notebook.
+4. Custom filters: marimo allows the creation of pandas-backed filters using UI elements like sliders and dropdowns.
+5. Dataset explorer: marimo automatically discovers and displays all dataframes in the notebook, allowing users to explore and visualize data interactively.
+6. SQL integration: marimo allows users to write SQL queries against any pandas dataframes existing in memory.
+
+## API
+
+### [pandas-datareader](https://github.com/pydata/pandas-datareader)
+
+`pandas-datareader` is a remote data access library for pandas
+(PyPI:`pandas-datareader`). It is based on functionality that was
+located in `pandas.io.data` and `pandas.io.wb` but was split off in
+v0.19. See more in the [pandas-datareader
+docs](https://pandas-datareader.readthedocs.io/en/latest/):
+
+The following data feeds are available:
+
+- Google Finance
+- Tiingo
+- Morningstar
+- IEX
+- Robinhood
+- Enigma
+- Quandl
+- FRED
+- Fama/French
+- World Bank
+- OECD
+- Eurostat
+- TSP Fund Data
+- Nasdaq Trader Symbol Definitions
+- Stooq Index Data
+- MOEX Data
+
+### [pandaSDMX](https://pandasdmx.readthedocs.io)
+
+pandaSDMX is a library to retrieve and acquire statistical data and
+metadata disseminated in [SDMX](https://sdmx.org) 2.1, an
+ISO-standard widely used by institutions such as statistics offices,
+central banks, and international organisations. pandaSDMX can expose
+datasets and related structural metadata including data flows,
+code-lists, and data structure definitions as pandas Series or
+MultiIndexed DataFrames.
+
+### [fredapi](https://github.com/mortada/fredapi)
+
+fredapi is a Python interface to the [Federal Reserve Economic Data
+(FRED)](https://fred.stlouisfed.org/) provided by the Federal Reserve
+Bank of St. Louis. It works with both the FRED database and ALFRED
+database that contains point-in-time data (i.e. historic data
+revisions). fredapi provides a wrapper in Python to the FRED HTTP API,
+and also provides several convenient methods for parsing and analyzing
+point-in-time data from ALFRED. fredapi makes use of pandas and returns
+data in a Series or DataFrame. This module requires a FRED API key that
+you can obtain for free on the FRED website.
+
+## Domain specific
+
+### [Geopandas](https://github.com/geopandas/geopandas)
+
+Geopandas extends pandas data objects to include geographic information
+which support geometric operations. If your work entails maps and
+geographical coordinates, and you love pandas, you should take a close
+look at Geopandas.
+
+### [gurobipy-pandas](https://github.com/Gurobi/gurobipy-pandas)
+
+gurobipy-pandas provides a convenient accessor API to connect pandas with
+gurobipy. It enables users to more easily and efficiently build mathematical
+optimization models from data stored in DataFrames and Series, and to read
+solutions back directly as pandas objects.
+
+### [staircase](https://github.com/staircase-dev/staircase)
+
+staircase is a data analysis package, built upon pandas and numpy, for modelling and
+manipulation of mathematical step functions. It provides a rich variety of arithmetic
+operations, relational operations, logical operations, statistical operations and
+aggregations for step functions defined over real numbers, datetime and timedelta domains.
+
+### [xarray](https://github.com/pydata/xarray)
+
+xarray brings the labeled data power of pandas to the physical sciences
+by providing N-dimensional variants of the core pandas data structures.
+It aims to provide a pandas-like and pandas-compatible toolkit for
+analytics on multi-dimensional arrays, rather than the tabular data for
+which pandas excels.
+
+## IO
+
+### [NTV-pandas](https://github.com/loco-philippe/ntv-pandas)
+
+NTV-pandas provides a JSON converter with more data types than the ones supported by pandas directly.
+
+It supports the following data types:
+
+- pandas data types
+- data types defined in the [NTV format](https://loco-philippe.github.io/ES/JSON%20semantic%20format%20(JSON-NTV).htm)
+- data types defined in [Table Schema specification](https://datapackage.org/standard/table-schema/)
+
+The interface is always reversible (conversion round trip) with two formats (JSON-NTV and JSON-TableSchema).
+
+Example:
+
+```python
+import ntv_pandas as npd
+
+jsn = df.npd.to_json(table=False)  # save df as a JSON-value (format Table Schema if table is True else format NTV )
+df  = npd.read_json(jsn)  # load a JSON-value as a `DataFrame`
+
+df.equals(npd.read_json(df.npd.to_json(df)))  # `True` in any case, whether `table=True` or not
+```
+
+### [BCPandas](https://github.com/yehoshuadimarsky/bcpandas)
+
+BCPandas provides high performance writes from pandas to Microsoft SQL Server,
+far exceeding the performance of the native ``df.to_sql`` method. Internally, it uses
+Microsoft's BCP utility, but the complexity is fully abstracted away from the end user.
+Rigorously tested, it is a complete replacement for ``df.to_sql``.
+
+### [Deltalake](https://pypi.org/project/deltalake)
+
+Deltalake python package lets you access tables stored in
+[Delta Lake](https://delta.io/) natively in Python without the need to use Spark or
+JVM. It provides the ``delta_table.to_pyarrow_table().to_pandas()`` method to convert
+any Delta table into Pandas dataframe.
+
+### [pandas-gbq](https://github.com/googleapis/python-bigquery-pandas)
+
+pandas-gbq provides high performance reads and writes to and from
+[Google BigQuery](https://cloud.google.com/bigquery/). Previously (before version 2.2.0),
+these methods were exposed as `pandas.read_gbq` and `DataFrame.to_gbq`.
+Use `pandas_gbq.read_gbq` and `pandas_gbq.to_gbq`, instead.
+
+
+### [ArcticDB](https://github.com/man-group/ArcticDB)
+
+ArcticDB is a serverless DataFrame database engine designed for the Python Data Science ecosystem. ArcticDB enables you to store, retrieve, and process pandas DataFrames at scale. It is a storage engine designed for object storage and also supports local-disk storage using LMDB. ArcticDB requires zero additional infrastructure beyond a running Python environment and access to object storage and can be installed in seconds. Please find full documentation [here](https://docs.arcticdb.io/latest/).
+
+#### ArcticDB Terminology
+
+ArcticDB is structured to provide a scalable and efficient way to manage and retrieve DataFrames, organized into several key components:
+
+- `Object Store` Collections of libraries. Used to separate logical environments from each other. Analogous to a database server.
+- `Library` Contains multiple symbols which are grouped in a certain way (different users, markets, etc). Analogous to a database.
+- `Symbol` Atomic unit of data storage. Identified by a string name. Data stored under a symbol strongly resembles a pandas DataFrame. Analogous to tables.
+- `Version` Every modifying action (write, append, update) performed on a symbol creates a new version of that object.
+
+#### Installation
+
+To install, simply run:
+
+```console
+pip install arcticdb
+```
+
+To get started, we can import ArcticDB and instantiate it:
+
+```python
+import arcticdb as adb
+import numpy as np
+import pandas as pd
+# this will set up the storage using the local file system
+arctic = adb.Arctic("lmdb://arcticdb_test")
+```
+
+> **Note:** ArcticDB supports any S3 API compatible storage, including AWS. ArcticDB also supports Azure Blob storage.  
+> ArcticDB also supports LMDB for local/file based storage - to use LMDB, pass an LMDB path as the URI: `adb.Arctic('lmdb://path/to/desired/database')`.
+
+#### Library Setup
+
+ArcticDB is geared towards storing many (potentially millions) of tables. Individual tables (DataFrames) are called symbols and are stored in collections called libraries. A single library can store many symbols. Libraries must first be initialized prior to use:
+
+```python
+lib = arctic.get_library('sample', create_if_missing=True)
+```
+
+#### Writing Data to ArcticDB
+
+Now we have a library set up, we can get to reading and writing data. ArcticDB has a set of simple functions for DataFrame storage. Let's write a DataFrame to storage.
+
+```python
+df = pd.DataFrame(
+    {
+        "a": list("abc"),
+        "b": list(range(1, 4)),
+        "c": np.arange(3, 6).astype("u1"),
+        "d": np.arange(4.0, 7.0, dtype="float64"),
+        "e": [True, False, True],
+        "f": pd.date_range("20130101", periods=3)
+    }
+)
+
+df
+df.dtypes
+```
+
+Write to ArcticDB.
+
+```python
+write_record = lib.write("test", df)
+```
+
+> **Note:** When writing pandas DataFrames, ArcticDB supports the following index types:
+>
+> - `pandas.Index` containing int64 (or the corresponding dedicated types Int64Index, UInt64Index)
+> - `RangeIndex`
+> - `DatetimeIndex`
+> - `MultiIndex` composed of above supported types
+>
+> The "row" concept in `head`/`tail` refers to the row number ('iloc'), not the value in the `pandas.Index` ('loc').
+
+#### Reading Data from ArcticDB
+
+Read the data back from storage:
+
+```python
+read_record = lib.read("test")
+read_record.data
+df.dtypes
+```
+
+ArcticDB also supports appending, updating, and querying data from storage to a pandas DataFrame. Please find more information [here](https://docs.arcticdb.io/latest/api/query_builder/).
+
+### [Hugging Face](https://huggingface.co/datasets)
+
+The Hugging Face Dataset Hub provides a large collection of ready-to-use datasets for machine learning shared by the community. The platform offers a user-friendly interface to explore, discover and visualize datasets, and provides tools to easily load and work with these datasets in Python thanks to the [huggingface_hub](https://github.com/huggingface/huggingface_hub) library.
+
+You can access datasets on Hugging Face using `hf://` paths in pandas, in the form `hf://datasets/username/dataset_name/...`.
+
+For example, here is how to load the [stanfordnlp/imdb dataset](https://huggingface.co/datasets/stanfordnlp/imdb):
+
+```python
+import pandas as pd
+
+# Load the IMDB dataset
+df = pd.read_parquet("hf://datasets/stanfordnlp/imdb/plain_text/train-00000-of-00001.parquet")
+```
+
+Tip: on a dataset page, click on "Use this dataset" to get the code to load it in pandas.
+
+To save a dataset on Hugging Face you need to [create a public or private dataset](https://huggingface.co/new-dataset) and [login](https://huggingface.co/docs/huggingface_hub/quick-start#login-command), and then you can use `df.to_csv/to_json/to_parquet`:
+
+```python
+# Save the dataset to my Hugging Face account
+df.to_parquet("hf://datasets/username/dataset_name/train.parquet")
+```
+
+You can find more information about the Hugging Face Dataset Hub in the [documentation](https://huggingface.co/docs/hub/en/datasets).
+
+## Out-of-core
+
+### [Bodo](https://github.com/bodo-ai/Bodo)
+
+
+Bodo is a high-performance compute engine for Python data processing.
+Using an auto-parallelizing just-in-time (JIT) compiler, Bodo simplifies scaling Pandas
+workloads from laptops to clusters without major code changes.
+Under the hood, Bodo relies on MPI-based high-performance computing (HPC) technology—making it
+both easier to use and often much faster than alternatives.
+Bodo also provides a SQL engine that can query distributed pandas dataframes efficiently.
+
+```python
+import pandas as pd
+import bodo
+
+@bodo.jit
+def process_data():
+    df = pd.read_parquet("my_data.pq")
+    df2 = pd.DataFrame({"A": df.apply(lambda r: 0 if r.A == 0 else (r.B // r.A), axis=1)})
+    df2.to_parquet("out.pq")
+
+process_data()
+```
+
+
+### [Cylon](https://cylondata.org/)
+
+Cylon is a fast, scalable, distributed memory parallel runtime with a pandas
+like Python DataFrame API. ”Core Cylon” is implemented with C++ using Apache
+Arrow format to represent the data in-memory. Cylon DataFrame API implements
+most of the core operators of pandas such as merge, filter, join, concat,
+group-by, drop_duplicates, etc. These operators are designed to work across
+thousands of cores to scale applications. It can interoperate with pandas
+DataFrame by reading data from pandas or converting data to pandas so users
+can selectively scale parts of their pandas DataFrame applications.
+
+```python
+from pycylon import read_csv, DataFrame, CylonEnv
+from pycylon.net import MPIConfig
+
+# Initialize Cylon distributed environment
+config: MPIConfig = MPIConfig()
+env: CylonEnv = CylonEnv(config=config, distributed=True)
+
+df1: DataFrame = read_csv('/tmp/csv1.csv')
+df2: DataFrame = read_csv('/tmp/csv2.csv')
+
+# Using 1000s of cores across the cluster to compute the join
+df3: Table = df1.join(other=df2, on=[0], algorithm="hash", env=env)
+
+print(df3)
+```
+
+### [Dask](https://docs.dask.org)
+
+Dask is a flexible parallel computing library for analytics. Dask
+provides a familiar `DataFrame` interface for out-of-core, parallel and
+distributed computing.
+
+### [Dask-ML](https://ml.dask.org)
+
+Dask-ML enables parallel and distributed machine learning using Dask
+alongside existing machine learning libraries like Scikit-Learn,
+XGBoost, and TensorFlow.
+
+### [Ibis](https://ibis-project.org/docs/)
+
+Ibis offers a standard way to write analytics code, that can be run in multiple engines. It helps in bridging the gap between local Python environments (like pandas) and remote storage and execution systems like Hadoop components (like HDFS, Impala, Hive, Spark) and SQL databases (Postgres, etc.).
+
+
+### [Koalas](https://koalas.readthedocs.io/en/latest/)
+
+Koalas provides a familiar pandas DataFrame interface on top of Apache
+Spark. It enables users to leverage multi-cores on one machine or a
+cluster of machines to speed up or scale their DataFrame code.
+
+### [Modin](https://github.com/modin-project/modin)
+
+The ``modin.pandas`` DataFrame is a parallel and distributed drop-in replacement
+for pandas. This means that you can use Modin with existing pandas code or write
+new code with the existing pandas API. Modin can leverage your entire machine or
+cluster to speed up and scale your pandas workloads, including traditionally
+time-consuming tasks like ingesting data (``read_csv``, ``read_excel``,
+``read_parquet``, etc.).
+
+```python
+# import pandas as pd
+import modin.pandas as pd
+
+df = pd.read_csv("big.csv")  # use all your cores!
+```
+
+### [Pandarallel](https://github.com/nalepae/pandarallel)
+
+Pandarallel provides a simple way to parallelize your pandas operations on all your CPUs by changing only one line of code.
+If also displays progress bars.
+
+```python
+from pandarallel import pandarallel
+
+pandarallel.initialize(progress_bar=True)
+
+# df.apply(func)
+df.parallel_apply(func)
+```
+
+### [Vaex](https://vaex.io/docs/)
+
+Increasingly, packages are being built on top of pandas to address
+specific needs in data preparation, analysis and visualization. Vaex is
+a python library for Out-of-Core DataFrames (similar to Pandas), to
+visualize and explore big tabular datasets. It can calculate statistics
+such as mean, sum, count, standard deviation etc, on an N-dimensional
+grid up to a billion (10^9) objects/rows per second. Visualization is
+done using histograms, density plots and 3d volume rendering, allowing
+interactive exploration of big data. Vaex uses memory mapping, zero
+memory copy policy and lazy computations for best performance (no memory
+wasted).
+
+- ``vaex.from_pandas``
+- ``vaex.to_pandas_df``
+
+### [Hail Query](https://hail.is/)
+
+An out-of-core, preemptible-safe, distributed, dataframe library serving
+the genetics community. Hail Query ships with on-disk data formats,
+in-memory data formats, an expression compiler, a query planner, and a
+distributed sort algorithm all designed to accelerate queries on large
+matrices of genome sequencing data.
+
+It is often easiest to use pandas to manipulate the summary statistics or
+other small aggregates produced by Hail. For this reason, Hail provides
+native import to and export from pandas DataFrames:
+
+- [`Table.from_pandas`](https://hail.is/docs/latest/hail.Table.html#hail.Table.from_pandas)
+- [`Table.to_pandas`](https://hail.is/docs/latest/hail.Table.html#hail.Table.to_pandas)
+
+## Data cleaning and validation
+
+### [pyjanitor](https://github.com/pyjanitor-devs/pyjanitor)
+
+Pyjanitor provides a clean API for cleaning data, using method chaining.
+
+### [Pandera](https://pandera.readthedocs.io/en/stable/)
+
+Pandera provides a flexible and expressive API for performing data validation on dataframes
+to make data processing pipelines more readable and robust.
+Dataframes contain information that pandera explicitly validates at runtime. This is useful in
+production-critical data pipelines or reproducible research settings.
+
+## Extension data types
+
+Pandas provides an interface for defining
+[extension types](https://pandas.pydata.org/docs/development/extending.html#extension-types) to extend NumPy's type system.
+The following libraries implement that interface to provide types not found in NumPy or pandas,
+which work well with pandas' data containers.
+
+### [awkward-pandas](https://awkward-pandas.readthedocs.io/)
+
+Awkward-pandas provides an extension type for storing [Awkward
+Arrays](https://awkward-array.org/) inside pandas' Series and
+DataFrame. It also provides an accessor for using awkward functions
+on Series that are of awkward type.
+
+### [db-dtypes](https://github.com/googleapis/python-db-dtypes-pandas)
+
+db-dtypes provides an extension types for working with types like
+DATE, TIME, and JSON from database systems. This package is used
+by pandas-gbq to provide natural dtypes for BigQuery data types without
+a natural numpy type.
+
+### [Pandas-Genomics](https://pandas-genomics.readthedocs.io/en/latest/)
+
+Pandas-Genomics provides an extension type and extension array for working
+ with genomics data.  It also includes `genomics` accessors for many useful properties
+ and methods related to QC and analysis of genomics data.
+
+### [Physipandas](https://github.com/mocquin/physipandas)
+
+Physipandas provides an extension for manipulating physical quantities
+ (like scalar and numpy.ndarray) in association with a physical unit
+ (like meter or joule) and additional features for integration of
+ `physipy` accessors with pandas Series and Dataframe.
+
+### [Pint-Pandas](https://github.com/hgrecco/pint-pandas)
+
+Pint-Pandas provides an extension type for storing numeric arrays with units.
+These arrays can be stored inside pandas' Series and DataFrame. Operations
+between Series and DataFrame columns which use pint's extension array are then
+units aware.
+
+### [Text Extensions](https://ibm.biz/text-extensions-for-pandas)
+
+Text Extensions for Pandas provides extension types to cover common data structures for representing natural language data, plus library integrations that convert the outputs of popular natural language processing libraries into pandas DataFrames.
+
+## Accessors
+
+A directory of projects providing
+[extension accessors](https://pandas.pydata.org/docs/development/extending.html#registering-custom-accessors).
+This is for users to discover new accessors and for library
+authors to coordinate on the namespace.
+
+  | Library                                                              | Accessor   | Classes               |
+  | -------------------------------------------------------------------- | ---------- | --------------------- |
+  | [awkward-pandas](https://awkward-pandas.readthedocs.io/en/latest/)   | `ak`       | `Series`              |
+  | [pdvega](https://altair-viz.github.io/pdvega/)                       | `vgplot`   | `Series`, `DataFrame` |
+  | [pandas-genomics](https://pandas-genomics.readthedocs.io/en/latest/) | `genomics` | `Series`, `DataFrame` |
+  | [pint-pandas](https://github.com/hgrecco/pint-pandas)                | `pint`     | `Series`, `DataFrame` |
+  | [physipandas](https://github.com/mocquin/physipandas)                | `physipy`  | `Series`, `DataFrame` |
+  | [composeml](https://github.com/alteryx/compose)                      | `slice`    | `DataFrame`           |
+  | [gurobipy-pandas](https://github.com/Gurobi/gurobipy-pandas)         | `gppd`     | `Series`, `DataFrame` |
+  | [staircase](https://www.staircase.dev/)                              | `sc`       | `Series`, `DataFrame` |
+  | [woodwork](https://github.com/alteryx/woodwork)                      | `slice`    | `Series`, `DataFrame` |
+
+## Development tools
+
+### [pandas-stubs](https://github.com/VirtusLab/pandas-stubs)
+
+While pandas repository is partially typed, the package itself doesn't expose this information for external use.
+Install pandas-stubs to enable basic type coverage of pandas API.
+
+Learn more by reading through these issues [14468](https://github.com/pandas-dev/pandas/issues/14468),
+[26766](https://github.com/pandas-dev/pandas/issues/26766), [28142](https://github.com/pandas-dev/pandas/issues/28142).
+
+See installation and usage instructions on the [GitHub page](https://github.com/VirtusLab/pandas-stubs).
+
+### [Hamilton](https://github.com/dagworks-inc/hamilton)
+
+Hamilton is a declarative dataflow framework that came out of Stitch Fix. It was designed to help one manage a
+Pandas code base, specifically with respect to feature engineering for machine learning models.
+
+It prescribes an opinionated paradigm, that ensures all code is:
+
+* unit testable
+* integration testing friendly
+* documentation friendly
+* transformation logic is reusable, as it is decoupled from the context of where it is used.
+* integratable with runtime data quality checks.
+
+This helps one to scale your pandas code base, at the same time, keeping maintenance costs low.
+
+For more information, see [documentation](https://hamilton.readthedocs.io/).
diff --git a/pandas/web/pandas/config.yml b/pandas/web/pandas/config.yml
new file mode 100644
index 0000000000000..41ba581852dd1
--- /dev/null
+++ b/pandas/web/pandas/config.yml
@@ -0,0 +1,221 @@
+main:
+  templates_path: _templates
+  base_template: "layout.html"
+  production_url: "https://pandas.pydata.org/"
+  ignore:
+  - _templates/layout.html
+  - config.yml
+  github_repo_url: pandas-dev/pandas
+  context_preprocessors:
+  - pandas_web.Preprocessors.current_year
+  - pandas_web.Preprocessors.navbar_add_info
+  - pandas_web.Preprocessors.blog_add_posts
+  - pandas_web.Preprocessors.maintainers_add_info
+  - pandas_web.Preprocessors.home_add_releases
+  - pandas_web.Preprocessors.roadmap_pdeps
+  markdown_extensions:
+  - toc
+  - tables
+  - fenced_code
+  - meta
+  - footnotes
+  - codehilite
+static:
+  logo: static/img/pandas_white.svg
+  css:
+  - static/css/pandas.css
+  - static/css/codehilite.css
+navbar:
+  - name: "About us"
+    target:
+    - name: "About pandas"
+      target: about/
+    - name: "Project roadmap"
+      target: about/roadmap.html
+    - name: "Governance"
+      target: about/governance.html
+    - name: "Team"
+      target: about/team.html
+    - name: "Sponsors"
+      target: about/sponsors.html
+    - name: "Citing and logo"
+      target: about/citing.html
+  - name: "Getting started"
+    target: getting_started.html
+  - name: "Documentation"
+    target: docs/
+  - name: "Community"
+    target:
+    - name: "Blog"
+      target: community/blog/
+    - name: "Ask a question (StackOverflow)"
+      target: https://stackoverflow.com/questions/tagged/pandas
+    - name: "Code of conduct"
+      target: community/coc.html
+    - name: "Ecosystem"
+      target: community/ecosystem.html
+    - name: "Benchmarks"
+      target: community/benchmarks.html
+  - name: "Contribute"
+    target: contribute.html
+blog:
+  num_posts: 50
+  posts_path: community/blog
+  author: "pandas team"
+  feed_name: "pandas blog"
+  feed:
+  - https://wesmckinney.com/feeds/pandas.atom.xml
+  - https://tomaugspurger.github.io/feed
+  - https://jorisvandenbossche.github.io/feeds/pandas.atom.xml
+  - https://datapythonista.me/blog/feeds/pandas.atom.xml
+  - https://numfocus.org/tag/pandas/feed/
+  - https://phofl.github.io/feeds/pandas.atom.xml
+maintainers:
+  active:
+  - jorisvandenbossche
+  - TomAugspurger
+  - jreback
+  - WillAyd
+  - mroeschke
+  - jbrockmendel
+  - datapythonista
+  - simonjayhawkins
+  - topper-123
+  - alimcmaster1
+  - bashtage
+  - Dr-Irv
+  - MarcoGorelli
+  - rhshadrach
+  - phofl
+  - attack68
+  - fangchenli
+  - lithomas1
+  - lukemanley
+  - noatamir
+  inactive:
+  - lodagro
+  - jseabold
+  - jtratner
+  - shoyer
+  - chris-b1
+  - sinhrks
+  - cpcloud
+  - toobaz
+  - jschendel
+  - charlesdong1991
+  - dsaxton
+  - wesm
+  - gfyoung
+  - mzeitlin11
+  - twoertwein
+workgroups:
+  coc:
+    name: Code of Conduct
+    contact: coc@pandas.pydata.org
+    responsibilities: "Make sure pandas is the welcoming and inclusive community we want it to be. Keeping the CoC updated, and addressing violation reports."
+    members:
+    - Tom Augspurger
+    - Bijay Regmi
+    - Wuraola Oyewusi
+    - Мария Чакчурина
+  finance:
+    name: Finance
+    contact: finance@pandas.pydata.org
+    responsibilities: "Manage the funding. Coordinate the request of grants. Approve the project expenses."
+    members:
+    - Matthew Roeschke
+    - Jeff Reback
+    - Joris Van den Bossche
+    - Patrick Hoefler
+  infrastructure:
+    name: Infrastructure
+    contact: infrastructure@pandas.pydata.org
+    responsibilities: "Keep the pandas infrastructure up and working. In particular the servers for the website, benchmarks, CI and others needed."
+    members:
+    - William Ayd
+    - Thomas Li
+  communications:
+    name: Communications
+    contact: communications@pandas.pydata.org
+    responsibilities: "Share relevant information with the broader community, mainly via our social networks, as well as being the main point of contact between NumFOCUS and the core team."
+    members:
+    - Marco Gorelli
+sponsors:
+  active:
+  - name: "NumFOCUS"
+    url: https://numfocus.org/
+    logo: static/img/partners/numfocus.svg
+    kind: numfocus
+  - name: "Two Sigma"
+    url: https://www.twosigma.com/
+    logo: static/img/partners/two_sigma.svg
+    kind: partner
+    description: "Jeff Reback"
+  - name: "Voltron Data"
+    url: https://voltrondata.com/
+    logo: static/img/partners/voltron_data.svg
+    kind: partner
+    description: "Joris Van den Bossche"
+  - name: "Coiled"
+    url: https://www.coiled.io
+    logo: static/img/partners/coiled.svg
+    kind: partner
+    description: "Patrick Hoefler"
+  - name: "Quansight"
+    url: https://quansight.com/
+    logo: static/img/partners/quansight_labs.svg
+    kind: partner
+    description: "Marco Gorelli"
+  - name: "Nvidia"
+    url: https://www.nvidia.com
+    logo: static/img/partners/nvidia.svg
+    kind: partner
+    description: "Matthew Roeschke"
+  - name: "Intel"
+    url: https://www.intel.com/
+    logo: /static/img/partners/intel.svg
+    kind: partner
+    description: "Brock Mendel"
+  - name: "Tidelift"
+    url: https://tidelift.com
+    logo: static/img/partners/tidelift.svg
+    kind: regular
+    description: "<i>pandas</i> is part of the <a href=\"https://tidelift.com/subscription/pkg/pypi-pandas?utm_source=pypi-pandas&utm_medium=referral&utm_campaign=readme\">Tidelift subscription</a>. You can support pandas by becoming a Tidelift subscriber."
+  - name: "Chan Zuckerberg Initiative"
+    url: https://chanzuckerberg.com/
+    logo: static/img/partners/czi.svg
+    kind: regular
+    description: "<i>pandas</i> is funded by the Essential Open Source Software for Science program of the Chan Zuckerberg Initiative. The funding is used for general maintenance, improve extension types, and a efficient string type."
+  - name: "Bodo"
+    url: https://www.bodo.ai/
+    logo: static/img/partners/bodo.svg
+    kind: regular
+    description: "Bodo's parallel computing platform uses pandas API, and Bodo financially supports pandas development to help improve pandas, in particular the pandas API"
+  inkind:  # not included in active so they don't appear in the home page
+  - name: "OVH"
+    url: https://us.ovhcloud.com/
+    description: "Website and documentation hosting."
+  - name: "Indeed"
+    url: https://opensource.indeedeng.io/
+    description: "<i>pandas</i> logo design"
+  past:
+  - name: "Paris-Saclay Center for Data Science"
+    url: https://www.datascience-paris-saclay.fr/
+    kind: partner
+  - name: "Anaconda"
+    url: https://www.anaconda.com/
+    kind: partner
+  - name: "RStudio"
+    url: https://www.rstudio.com/
+    kind: partner
+  - name: "Ursa Labs"
+    url: https://ursalabs.org/
+    kind: partner
+  - name: "Gousto"
+    url: https://www.gousto.co.uk/
+    kind: partner
+  - name: "d-fine GmbH"
+    url: https://www.d-fine.com/en/
+    kind: partner
+roadmap:
+  pdeps_path: pdeps
diff --git a/pandas/web/pandas/contribute.md b/pandas/web/pandas/contribute.md
new file mode 100644
index 0000000000000..3307ddcfb1a0a
--- /dev/null
+++ b/pandas/web/pandas/contribute.md
@@ -0,0 +1,55 @@
+# Contribute to pandas
+
+_pandas_ is and will always be **free**. To make the development sustainable, we need _pandas_ users, corporate
+and individual, to support the development by providing their time and money.
+
+You can find more information about current developers in the [team page]({{ base_url }}about/team.html),
+and about current sponsors in the [sponsors page]({{ base_url }}about/sponsors.html).
+
+<section>
+    <div class="container mt-5">
+      <div class="row text-center">
+        <div class="col-md-4">
+          <span class="fa-stack fa-4x">
+            <i class="fas fa-circle fa-stack-2x pink"></i>
+            <i class="fas fa-building fa-stack-1x fa-inverse"></i>
+          </span>
+          <h4 class="service-heading mt-3 fw-bold blue">Corporate support</h4>
+          <p class="text-muted">
+            pandas depends on companies and institutions using the software to support its development. Hiring
+            people to work on pandas, or letting existing employees to contribute to the
+            software. Or sponsoring pandas with funds, so the project can hire people to
+            progress on the <a href="{{ base_url }}about/roadmap.html">pandas roadmap</a>.
+          </p>
+          <p>More information in the <a href="{{ base_url }}about/sponsors.html">sponsors page</a></p>
+        </div>
+        <div class="col-md-4">
+          <span class="fa-stack fa-4x">
+            <i class="fas fa-circle fa-stack-2x pink"></i>
+            <i class="fas fa-users fa-stack-1x fa-inverse"></i>
+          </span>
+          <h4 class="service-heading mt-3 fw-bold blue">Individual contributors</h4>
+          <p class="text-muted">
+            pandas is mostly developed by volunteers. All kind of contributions are welcome,
+            such as contributions to the code, to the website (including graphical designers),
+            to the documentation (including translators) and others. There are tasks for all
+            levels, including beginners.
+          </p>
+          <p>More information in the <a href="{{ base_url }}docs/development/index.html">contributing page</a></p>
+        </div>
+        <div class="col-md-4">
+          <span class="fa-stack fa-4x">
+            <i class="fas fa-circle fa-stack-2x pink"></i>
+            <i class="fas fa-dollar-sign fa-stack-1x fa-inverse"></i>
+          </span>
+          <h4 class="service-heading mt-3 fw-bold blue">Donations</h4>
+          <p class="text-muted">
+            Individual donations are appreciated, and are used for things like the project
+            infrastructure, travel expenses for our volunteer contributors to attend
+            the in-person sprints, or to give small grants to develop features.
+          </p>
+          <p>Make your donation in the <a href="https://opencollective.com/pandas">donate page</a></p>
+        </div>
+      </div>
+    </div>
+</section>
diff --git a/pandas/web/pandas/getting_started.md b/pandas/web/pandas/getting_started.md
new file mode 100644
index 0000000000000..801081a9ef391
--- /dev/null
+++ b/pandas/web/pandas/getting_started.md
@@ -0,0 +1,32 @@
+# Getting started
+
+## Installation instructions
+
+To install pandas, please reference the [installation page]({{ base_url}}docs/getting_started/install.html)
+from the pandas documentation.
+
+## Tutorials
+
+You can learn more about pandas in the [tutorials]({{ base_url }}docs/getting_started/intro_tutorials/),
+and more about JupyterLab in the
+[JupyterLab documentation](https://jupyterlab.readthedocs.io/en/stable/user/interface.html).
+
+## Books
+
+The book we recommend to learn pandas is [Python for Data Analysis](https://amzn.to/3DyLaJc),
+by [Wes McKinney](https://wesmckinney.com/), creator of pandas.
+
+<a href="https://amzn.to/3DyLaJc">
+    <img alt="Python for Data Analysis" src="{{ base_url }}static/img/books/pydata_book.gif"/>
+</a>
+
+## Videos
+
+<iframe width="560" height="315" frameborder="0"
+src="https://www.youtube.com/embed/_T8LGqJtuGc"
+allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture"
+allowfullscreen></iframe>
+
+## Cheat sheet
+
+[pandas cheat sheet](https://pandas.pydata.org/Pandas_Cheat_Sheet.pdf)
diff --git a/pandas/web/pandas/index.html b/pandas/web/pandas/index.html
new file mode 100644
index 0000000000000..e8aab9e11144c
--- /dev/null
+++ b/pandas/web/pandas/index.html
@@ -0,0 +1,136 @@
+{% extends "layout.html" %}
+{% block body %}
+    <div class="container">
+        <div class="row">
+            <div class="col-md-9">
+                <section class="h-30 p-5 bg-light border rounded-3 text-center mb-4">
+                    <h1>pandas</h1>
+                    <p>
+                        <strong>pandas</strong> is a fast, powerful, flexible and easy to use open source data analysis and manipulation tool,<br/>
+                        built on top of the <a href="https://www.python.org">Python</a> programming language.
+                    </p>
+                    <p>
+                        <a class="btn btn-primary" href="{{ base_url }}getting_started.html">Install pandas now!</a>
+                    </p>
+                </section>
+
+                <div class="row">
+                    <div class="col-md-4">
+                        <h5>Getting started</h5>
+                        <ul>
+                            <!-- <li><a href="{{ base_url }}/try.html">Try pandas online</a></li> -->
+                            <li><a href="{{ base_url }}getting_started.html">Install pandas</a></li>
+                            <li><a href="{{ base_url }}docs/getting_started/index.html">Getting started</a></li>
+                        </ul>
+                    </div>
+                    <div class="col-md-4">
+                        <h5>Documentation</h5>
+                        <ul>
+                            <li><a href="{{ base_url }}docs/user_guide/index.html">User guide</a></li>
+                            <li><a href="{{ base_url }}docs/reference/index.html">API reference</a></li>
+                            <li><a href="{{ base_url }}docs/development/index.html">Contributing to pandas</a></li>
+                            <li><a href="{{ base_url }}docs/whatsnew/index.html">Release notes</a></li>
+                        </ul>
+                    </div>
+                    <div class="col-md-4">
+                        <h5>Community</h5>
+                        <ul>
+                            <li><a href="{{ base_url }}about/index.html">About pandas</a></li>
+                            <li><a href="https://stackoverflow.com/questions/tagged/pandas">Ask a question</a></li>
+                            <li><a href="{{ base_url }}community/ecosystem.html">Ecosystem</a></li>
+                        </ul>
+                    </div>
+                </div>
+                <section>
+                    <h5>With the support of:</h5>
+                    {% for row in sponsors.active | batch(6, "") %}
+                        <div class="row mx-auto h-100">
+                            {% for company in row %}
+                                <div class="col-6 col-md-2">
+                                    {% if company %}
+                                        <a href="{{ company.url }}" target="_blank">
+                                            <img class="img-fluid img-thumbnail py-5 mx-auto" alt="{{ company.name }}" src="{{ base_url }}{{ company.logo }}"/>
+                                        </a>
+                                    {% endif %}
+                                </div>
+                            {% endfor %}
+                        </div>
+                    {% endfor %}
+                    <p class="mt-4">The full list of companies supporting <i>pandas</i> is available in the <a href="{{ base_url }}about/sponsors.html">sponsors page</a>.
+                </section>
+            </div>
+            <div class="col-md-3">
+                {% if releases %}
+                    <h4>Latest version: {{ releases[0].name }}</h4>
+                    <ul>
+                        <li><a href="{{ base_url }}docs/whatsnew/v{{ releases[0].name }}.html">What's new in {{ releases[0].name }}</a></li>
+                        <li>Release date:<br/>{{ releases[0].published.strftime("%b %d, %Y") }}</li>
+                        <li><a href="{{ base_url}}docs/">Documentation (web)</a></li>
+                        <li><a href="{{ releases[0].url }}">Download source code</a></li>
+                    </ul>
+                {% endif %}
+                <h4>Follow us</h4>
+                <div class="text-center">
+                    <ul class="list-inline social-buttons">
+                        <li class="list-inline-item">
+                            <a href="https://t.me/s/pandas_dev">
+                                <i class="follow-us-button fab bi bi-telegram"></i>
+                            </a>
+                        </li>
+                        <li class="list-inline-item">
+                            <a href="https://fosstodon.org/@pandas_dev" rel="me">
+                                <i class="follow-us-button fab bi bi-mastodon"></i>
+                            </a>
+                        </li>
+                        <li class="list-inline-item">
+                            <a href="https://x.com/pandas_dev">
+                                <i class="follow-us-button fab bi bi-twitter-x"></i>
+                            </a>
+                        </li>
+                    </ul>
+
+                </div>
+				<h4>Recommended books</h4>
+				<p class="book">
+					<a href="https://amzn.to/3DyLaJc">
+						<img class="img-fluid" alt="Python for Data Analysis" src="{{ base_url }}static/img/books/pydata_book.gif"/>
+					</a>
+				</p>
+				<p class="book">
+					<a href="https://store.metasnake.com/effective-pandas-book/lhte7">
+						<img class="img-fluid" alt="Effective pandas 2" src="{{ base_url }}static/img/books/effective_pandas_2.gif"/>
+					</a>
+				</p>
+                {% if releases[1:5] %}
+                    <h4>Previous versions</h4>
+                    <ul>
+                        {% for release in releases[1:5] %}
+                            <li class="small">
+                                {{ release.name }} ({{ release.published.strftime("%b %d, %Y") }})<br/>
+                                <a href="https://pandas.pydata.org/pandas-docs/stable/whatsnew/{{ release.tag }}.html">changelog</a> |
+                                <a href="https://pandas.pydata.org/pandas-docs/version/{{ release.name }}/">docs</a> |
+                                <a href="{{ release.url }}">code</a>
+                            </li>
+                        {% endfor %}
+                    </ul>
+                {% endif %}
+                {% if releases[5:] %}
+                    <p class="text-center">
+                        <a data-bs-toggle="collapse" href="#show-more-releases" role="button" aria-expanded="false" aria-controls="show-more-releases">Show more</a>
+                    </p>
+                    <ul id="show-more-releases" class="collapse">
+                        {% for release in releases[5:] %}
+                            <li class="small">
+                                {{ release.name }} ({{ release.published.strftime("%b %d, %Y") }})<br/>
+                                <a href="https://pandas.pydata.org/pandas-docs/stable/whatsnew/{{ release.tag }}.html">changelog</a> |
+                                <a href="https://pandas.pydata.org/pandas-docs/version/{{ release.name }}/">docs</a> |
+                                <a href="{{ release.url }}">code</a>
+                            </li>
+                        {% endfor %}
+                    </ul>
+                {% endif %}
+            </div>
+        </div>
+    </div>
+
+{% endblock %}
diff --git a/pandas/web/pandas/pdeps/0001-purpose-and-guidelines.md b/pandas/web/pandas/pdeps/0001-purpose-and-guidelines.md
new file mode 100644
index 0000000000000..7f5f0326eba6c
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0001-purpose-and-guidelines.md
@@ -0,0 +1,288 @@
+# PDEP-1: Purpose and guidelines
+
+- Created: 3 August 2022
+- Status: Accepted
+- Discussion: [#47444](https://github.com/pandas-dev/pandas/pull/47444),
+  [#51417](https://github.com/pandas-dev/pandas/pull/51417)
+- Author: [Marc Garcia](https://github.com/datapythonista),
+  [Noa Tamir](https://github.com/noatamir)
+- Revision: 3
+
+## PDEP definition, purpose and scope
+
+A PDEP (pandas enhancement proposal) is a proposal for a **major** change in
+pandas, in a similar way as a Python [PEP](https://peps.python.org/pep-0001/)
+or a NumPy [NEP](https://numpy.org/neps/nep-0000.html).
+
+Bug fixes and conceptually minor changes (e.g. adding a parameter to a function) are out of the
+scope of PDEPs. A PDEP should be used for changes that are not immediate and not obvious, when
+everybody in the pandas community needs to be aware of the possibility of an upcoming change.
+Such changes require detailed documentation before being implemented and frequently lead to a
+significant discussion within the community.
+
+PDEPs are appropriate for user facing changes, internal changes and significant discussions.
+Examples of topics worth a PDEP could include substantial API changes, breaking behavior changes,
+moving a module from pandas to a separate repository, or a refactoring of the pandas block manager.
+It is not always trivial to know which issue has enough scope to require the full PDEP process.
+Some simple API changes have sufficient consensus among the core team, and minimal impact on the
+community. On the other hand, if an issue becomes controversial, i.e. it generated a significant
+discussion, one could suggest opening a PDEP to formalize and document the discussion, making it
+easier for the wider community to participate. For context, see
+[the list of issues that could have been a PDEP](#List-of-issues).
+
+## PDEP guidelines
+
+### Target audience
+
+A PDEP is a public document available to anyone, but the main stakeholders to
+consider when writing a PDEP are:
+
+- The core development team, who will have the final decision on whether a PDEP
+  is approved or not
+- Contributors to pandas and other related projects, and experienced users. Their
+  feedback is highly encouraged and appreciated, to make sure all points of views
+  are taken into consideration
+- The wider pandas community, in particular users, who may or may not have feedback
+  on the proposal, but should know and be able to understand the future direction of
+  the project
+
+### PDEP authors
+
+Anyone can propose a PDEP, but a core member should be engaged to advise on a proposal made by
+non-core contributors. To submit a PDEP as a community member, please propose the PDEP concept on
+[an issue](https://github.com/pandas-dev/pandas/issues/new/choose), and find a pandas team
+member to collaborate with. They can advise you on the PDEP process and should be listed as an
+advisor on the PDEP when it is submitted to the PDEP repository.
+
+### Workflow
+
+#### Rationale
+
+Our workflow was created to support and enable a consensus seeking process, and to provide clarity,
+for current and future authors, as well as voting members. It is not a strict policy, and we
+discourage any interpretation which seeks to take advantage of it in a way that could "force" or
+"sneak" decisions in one way or another. We expect and encourage transparency, active discussion,
+feedback, and compromise from all our community members.
+
+#### PDEP States
+
+The possible states of a PDEP are:
+
+- Draft
+- Under discussion
+- Accepted
+- Implemented
+- Rejected
+- Withdrawn
+
+Next is described the workflow that PDEPs can follow.
+
+#### Submitting a PDEP
+
+Proposing a PDEP is done by creating a PR adding a new file to `web/pandas/pdeps/`.
+The file is a markdown file, you can use `web/pandas/pdeps/0001-purpose-and-guidelines.md` as a reference
+for the expected format.
+
+The initial status of a PDEP will be `Status: Draft`. This will be changed to
+`Status: Under discussion` by the author(s), when they are ready to proceed with the decision
+making process.
+
+#### PDEP Discussion Timeline
+
+A PDEP discussion will remain open for up to 60 days. This period aims to enable participation
+from volunteers, who might not always be available to respond quickly, as well as provide ample
+time to make changes based on suggestions and considerations offered by the participants.
+Similarly, the following voting period will remain open for 15 days.
+
+To enable and encourage discussions on PDEPs, we follow a notification schedule. At each of the
+following steps, the pandas team, and the pandas-dev mailing list are notified via GitHub and
+E-mail:
+
+- Once a PDEP is ready for discussion.
+- After 30 days, with a note that there is at most 30 days remaining for discussion,
+  and that a vote will be called for if no discussion occurs in the next 15 days.
+- After 45 days, with a note that there is at most 15 days remaining for discussion,
+  and that a vote will be called for in 15 days.
+- Once the voting period starts, after 60 days or in case of an earlier vote, with 15 days
+  remaining for voting.
+- After 10 voting days, with 5 days remaining for voting.
+
+After 30 discussion days, in case 15 days passed without any new unaddressed comments,
+the authors may close the discussion period preemptively, by sending an early reminder
+of 15 days remaining until the voting period starts.
+
+#### Casting Votes
+
+As the voting period starts, a VOTE issue is created which links to the PDEP discussion pull request.
+Each voting member, including author(s) with voting rights, may cast a vote by adding one of the following comments:
+
+- +1: approve.
+- 0: abstain.
+  - Reason: A one sentence reason is required.
+- -1: disapprove
+  - Reason: A one sentence reason is required.
+
+A disapprove vote requires prior participation in the PDEP discussion issue.
+
+Comments made on the public VOTE issue by non-voting members will be deleted.
+
+Once the voting period ends, any voter may tally the votes in a comment, using the format: w-x-y-z,
+where w stands for the total of approving, x of abstaining, y of disapproving votes cast, and z
+of number of voting members who did not respond to the VOTE issue. The tally of the votes will state
+if a quorum has been reached or not.
+
+#### Quorum and Majority
+
+For a PDEP vote to result in accepting the proposal, a quorum is required. All votes (including
+abstentions) are counted towards the quorum. The quorum is computed as the lower of these two
+values:
+
+- 11 voting members.
+- 50% of voting members.
+
+Given a quorum, a majority of 70% of the non-abstaining votes is required as well, i.e. 70% of
+the approving and disapproving votes must be in favor.
+
+Thus, abstaining votes count towards a quorum, but not towards a majority. A voting member might
+choose to abstain when they have participated in the discussion, have some objections to the
+proposal, but do not wish to stop the proposal from moving forward, nor indicate their full
+support.
+
+If a quorum was not reached by the end of the voting period, the PDEP is not accepted. Its status
+will change to rejected.
+
+#### Accepted PDEP
+
+Once a PDEP is accepted, any contributions can be made toward the implementation of the PDEP,
+with an open-ended completion timeline. Development of pandas is difficult to understand and
+forecast, being that the contributors to pandas are a mix of volunteers and developers paid from different sources,
+with different priorities. For companies, institutions or individuals with interest in seeing a
+PDEP being implemented, or to in general see progress to the pandas roadmap, please check how
+you can help in the [contributing page](/contribute.html).
+
+#### Implemented PDEP
+
+Once a PDEP is implemented and available in the main branch of pandas, its
+status will be changed to `Status: Implemented`, so there is visibility that the PDEP
+is not part of the roadmap and future plans, but a change that has already
+happened. The first pandas version in which the PDEP implementation is
+available will also be included in the PDEP header with for example
+`Implemented: v2.0.0`.
+
+#### Rejected PDEP
+
+A PDEP can be rejected when the final decision is that its implementation is
+not in the best interests of the project. Rejected PDEPs are as useful as accepted
+PDEPs, since there are discussions that are worth having, and decisions about
+changes to pandas being made. They will be merged with `Status: Rejected`, so
+there is visibility on what was discussed and what was the outcome of the
+discussion. A PDEP can be rejected for different reasons, for example good ideas
+that are not backward-compatible, and the breaking changes are not considered worth
+implementing.
+
+The PDEP author(s) can also decide to withdraw the PDEP before a final decision
+is made (`Status: Withdrawn`), when the PDEP authors themselves have decided
+that the PDEP is actually a bad idea, or have accepted that it is not broadly
+supported or a competing proposal is a better alternative.
+
+The author(s) may choose to resubmit a rejected or withdrawn PDEP. We expect
+authors to use their judgement in that case, as to whether they believe more
+discussion, or an amended proposal has the potential to lead to a different
+result. A new PDEP is then created, which includes a link to the previously
+rejected PDEP.
+
+#### Invalid PDEP
+
+For submitted PDEPs that do not contain proper documentation, are out of scope, or
+are not useful to the community for any other reason, the PR will be closed after
+discussion with the author, instead of merging them as rejected. This is to avoid
+adding noise to the list of rejected PDEPs, which should contain documentation as
+good as an accepted PDEP, but where the final decision was to not implement the changes.
+
+## Evolution of PDEPs
+
+Most PDEPs are not expected to change after they are accepted. Once there is agreement on the changes,
+and they are implemented, the PDEP will be only useful to understand why the development happened,
+and the details of the discussion.
+
+But in some cases, a PDEP can be updated. For example, a PDEP defining a procedure or
+a policy, like this one (PDEP-1). Or cases when after attempting the implementation,
+new knowledge is obtained that makes the original PDEP obsolete, and changes are
+required. When there are specific changes to be made to the original PDEP, this will
+be edited, its `Revision: X` label will be increased by one, and a note will be added
+to the `PDEP-N history` section. This will let readers understand that the PDEP has
+changed and avoid confusion.
+
+## <a id="List-of-issues"></a> List of issues that could have been PDEPs for context
+### Clear examples for potential PDEPs:
+
+- Adding a new parameter to many existing methods, or deprecating one in many places. For example:
+  - The `numeric_only` deprecation ([GH-28900][28900]) affected many methods and could have been a PDEP.
+- Adding a new data type has impact on a variety of places that need to handle the data type.
+  Such wide-ranging impact would require a PDEP. For example:
+  - `Categorical` ([GH-7217][7217], [GH-8074][8074]), `StringDtype` ([GH-8640][8640]), `ArrowDtype`
+- A significant (breaking) change in existing behavior. For example:
+  - Copy/view changes ([GH-36195][36195])
+- Support of new Python features with a wide impact on the project. For example:
+  - Supporting typing within pandas vs. creation of `pandas-stubs` ([GH-43197][43197],
+    [GH-45253][45253])
+- New required dependency.
+- Removing module from the project or splitting it off to a separate repository:
+  - Moving rarely used I/O connectors to a separate repository [GH-28409](28409)
+- Significant changes to contributors' processes are not going to have an impact on users, but
+they do benefit from structured discussion among the contributors. For example:
+  - Changing the build system to meson ([GH-49115][49115])
+
+### Borderline examples:
+Small changes to core functionality, such as `DataFrame` and `Series`, should always be
+considered as a PDEP candidate as it will likely have a big impact on users. But the same types
+of changes in other functionalities would not be good PDEP candidates. That said, any discussion,
+no matter how small the change, which becomes controversial is a PDEP candidate. Consider if more
+attention and/or a formal decision-making process would help. Following are some examples we
+hope can help clarify our meaning here:
+
+- API breaking changes, or discussion thereof, could be a PDEP. For example:
+  - `value_counts` result rename ([GH-49497][49497]). The scope does not justify a PDEP at first, but later a
+    discussion about whether it should be executed as a breaking change or with deprecation
+    emerges, which could benefit from the PDEP process.
+- Adding new methods or parameters to an existing method typically will not require a PDEP for
+  non-core features. For example:
+  - Both `dropna(percentage)` ([GH-35299][35299]), and `Timestamp.normalize()` ([GH-8794][8794])
+    would not have required a PDEP.
+  - On the other hand, `DataFrame.assign()` might. While it is a single method without backwards
+    compatibility concerns, it is also a core feature and the discussion should be highly visible.
+- Deprecating or removing a single method would not require a PDEP in most cases.
+  - That said, `DataFrame.append` ([GH-35407][35407]) is an example of deprecations on core
+    features that would be a good candidate for a PDEP.
+- Changing the default value of parameters in a core pandas method is another edge case. For
+  example:
+  - Such changes for `dropna` in `DataFrame.groupby` and `Series.groupby` could be a PDEP.
+- New top level modules and/or exposing internal classes. For example:
+  - Add `pandas.api.typing` ([GH-48577][48577]) is relatively small and would not necessarily
+    require a PDEP.
+
+
+### PDEP-1 History
+
+- 3 August 2022: Initial version ([GH-47938][47938])
+- 15 February 2023: Version 2 ([GH-51417][51417]) clarifies the scope of PDEPs and adds examples
+- 09 June 2023: Version 3 ([GH-53576][53576]) defines a structured decision making process for PDEPs
+
+[7217]: https://github.com/pandas-dev/pandas/pull/7217
+[8074]: https://github.com/pandas-dev/pandas/issues/8074
+[8640]: https://github.com/pandas-dev/pandas/issues/8640
+[36195]: https://github.com/pandas-dev/pandas/issues/36195
+[43197]: https://github.com/pandas-dev/pandas/issues/43197
+[45253]: https://github.com/pandas-dev/pandas/issues/45253
+[49497]: https://github.com/pandas-dev/pandas/issues/49497
+[35299]: https://github.com/pandas-dev/pandas/issues/35299
+[8794]: https://github.com/pandas-dev/pandas/issues/8794
+[6249]: https://github.com/pandas-dev/pandas/issues/6249
+[48577]: https://github.com/pandas-dev/pandas/issues/48577
+[49115]: https://github.com/pandas-dev/pandas/pull/49115
+[28409]: https://github.com/pandas-dev/pandas/issues/28409
+[47938]: https://github.com/pandas-dev/pandas/pull/47938
+[51417]: https://github.com/pandas-dev/pandas/pull/51417
+[28900]: https://github.com/pandas-dev/pandas/issues/28900
+[35407]: https://github.com/pandas-dev/pandas/issues/35407
+[53576]: https://github.com/pandas-dev/pandas/pull/53576
diff --git a/pandas/web/pandas/pdeps/0004-consistent-to-datetime-parsing.md b/pandas/web/pandas/pdeps/0004-consistent-to-datetime-parsing.md
new file mode 100644
index 0000000000000..68c6dfa26d1f1
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0004-consistent-to-datetime-parsing.md
@@ -0,0 +1,104 @@
+# PDEP-4: Consistent datetime parsing
+
+- Created: 18 September 2022
+- Status: Implemented
+- Discussion: [#48621](https://github.com/pandas-dev/pandas/pull/48621)
+- Author: [Marco Gorelli](https://github.com/MarcoGorelli)
+- Revision: 2
+
+## Abstract
+
+The suggestion is that:
+
+- ``to_datetime`` becomes strict and uses the same datetime format to parse all elements in its input.
+  The format will either be inferred from the first non-NaN element (if `format` is not provided by the user), or from
+  `format`;
+- ``infer_datetime_format`` be deprecated (as a strict version of it will become the default);
+- an easy workaround for non-strict parsing be clearly documented.
+
+## Motivation and Scope
+
+Pandas date parsing is very flexible, but arguably too much so - see
+https://github.com/pandas-dev/pandas/issues/12585 and linked issues for how
+much confusion this causes. Pandas can swap format midway, and though this
+is documented, it regularly breaks users' expectations.
+
+Simple example:
+```ipython
+In [1]: pd.to_datetime(['12-01-2000 00:00:00', '13-01-2000 00:00:00'])
+Out[1]: DatetimeIndex(['2000-12-01', '2000-01-13'], dtype='datetime64[ns]', freq=None)
+```
+The user was almost certainly intending the data to be read as "12th of January, 13th of January".
+However, it's read as "1st of December, 13th of January". No warning or error is thrown.
+
+Currently, the only way to ensure consistent parsing is by explicitly passing
+``format=``. The argument ``infer_datetime_format``
+isn't strict, can be called together with ``format``, and can still break users' expectations:
+
+```ipython
+In [2]: pd.to_datetime(['12-01-2000 00:00:00', '13-01-2000 00:00:00'], infer_datetime_format=True)
+Out[2]: DatetimeIndex(['2000-12-01', '2000-01-13'], dtype='datetime64[ns]', freq=None)
+```
+
+## Detailed Description
+
+Concretely, the suggestion is:
+
+- if no ``format`` is specified, ``pandas`` will guess the format from the first non-NaN row
+  and parse the rest of the input according to that format. Errors will be handled
+  according to the ``errors`` argument - there will be no silent switching of format;
+- ``infer_datetime_format`` will be deprecated;
+- ``dayfirst`` and ``yearfirst`` will continue working as they currently do;
+- if the format cannot be guessed from the first non-NaN row, a ``UserWarning`` will be thrown,
+  encouraging users to explicitly pass in a format.
+  Note that this should only happen for invalid inputs such as `'a'`
+  (which would later throw a ``ParserError`` anyway), or inputs such as ``'00:12:13'``,
+  which would currently get converted to ``''2022-09-18 00:12:13''``.
+
+If a user has dates in a mixed format, they can still use flexible parsing and accept
+the risks that poses, e.g.:
+```ipython
+In [3]: pd.to_datetime(['12-01-2000 00:00:00', '13-01-2000 00:00:00'], format='mixed')
+Out[3]: DatetimeIndex(['2000-12-01', '2000-01-13'], dtype='datetime64[ns]', freq=None)
+```
+or, if their dates are all ISO8601,
+```ipython
+In [4]: pd.to_datetime(['2020-01-01', '2020-01-01 03:00'], format='ISO8601')
+Out[4]: DatetimeIndex(['2020-01-01 00:00:00', '2020-01-01 03:00:00'], dtype='datetime64[ns]', freq=None)
+```
+
+## Usage and Impact
+
+My expectation is that the impact would be a net-positive:
+
+- potentially severe bugs in people's code will be caught early;
+- users who actually want mixed formats can still parse them, but now they'd be forced to be
+  very explicit about it;
+- the codebase would be noticeably simplified.
+
+As far as I can tell, there is no chance of _introducing_ bugs.
+
+## Implementation
+
+The whatsnew notes read
+
+> In the next major version release, 2.0, several larger API changes are being considered without a formal deprecation.
+
+I'd suggest making this change as part of the above, because:
+
+- it would only help prevent bugs, not introduce any;
+- given the severity of bugs that can result from the current behaviour, waiting another 2 years until pandas 3.0.0
+  would potentially cause a lot of damage.
+
+Note that this wouldn't mean getting rid of ``dateutil.parser``, as that would still be used within ``guess_datetime_format``. With this proposal, however, subsequent rows would be parsed with the guessed format rather than repeatedly calling ``dateutil.parser`` and risk having it silently switch format
+
+Finally, the function ``from pandas._libs.tslibs.parsing import guess_datetime_format`` would be made public, under ``pandas.tools``.
+
+## Out of scope
+
+We could make ``guess_datetime_format`` smarter by using a random sample of elements to infer the format.
+
+### PDEP History
+
+- 18 September 2022: Initial draft
+- 25 January 2023: Amended to mention ``format='ISO8601'`` and ``format='mixed'`` options
diff --git a/pandas/web/pandas/pdeps/0005-no-default-index-mode.md b/pandas/web/pandas/pdeps/0005-no-default-index-mode.md
new file mode 100644
index 0000000000000..d543a4718e896
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0005-no-default-index-mode.md
@@ -0,0 +1,380 @@
+# PDEP-5: NoRowIndex
+
+- Created: 14 November 2022
+- Status: Withdrawn
+- Discussion: [#49693](https://github.com/pandas-dev/pandas/pull/49693)
+- Author: [Marco Gorelli](https://github.com/MarcoGorelli)
+- Revision: 2
+
+## Abstract
+
+The suggestion is to add a ``NoRowIndex`` class. Internally, it would act a bit like
+a ``RangeIndex``, but some methods would be stricter. This would be one
+step towards enabling users who do not want to think about indices to not need to.
+
+## Motivation
+
+The Index can be a source of confusion and frustration for pandas users. For example, let's consider the inputs
+
+```python
+In[37]: ser1 = pd.Series([10, 15, 20, 25], index=[1, 2, 3, 5])
+
+In[38]: ser2 = pd.Series([10, 15, 20, 25], index=[1, 2, 3, 4])
+```
+
+Then:
+
+- it can be unexpected that adding `Series` with the same length (but different indices) produces `NaN`s in the result (https://stackoverflow.com/q/66094702/4451315):
+
+  ```python
+  In [41]: ser1 + ser2
+  Out[41]:
+  1    20.0
+  2    30.0
+  3    40.0
+  4     NaN
+  5     NaN
+  dtype: float64
+  ```
+
+- concatenation, even with `ignore_index=True`, still aligns on the index (https://github.com/pandas-dev/pandas/issues/25349):
+
+    ```python
+    In [42]: pd.concat([ser1, ser2], axis=1, ignore_index=True)
+    Out[42]:
+          0     1
+    1  10.0  10.0
+    2  15.0  15.0
+    3  20.0  20.0
+    5  25.0   NaN
+    4   NaN  25.0
+    ```
+
+- it can be frustrating to have to repeatedly call `.reset_index()` (https://twitter.com/chowthedog/status/1559946277315641345):
+
+  ```python
+  In [3]: ser1.reset_index(drop=True) + ser2.reset_index(drop=True)
+  Out[3]:
+  0    20
+  1    30
+  2    40
+  3    50
+  dtype: int64
+  ```
+
+If a user did not want to think about row labels (which they may have ended up after slicing / concatenating operations),
+then ``NoRowIndex`` would enable the above to work in a more intuitive
+manner (details and examples to follow below).
+
+## Scope
+
+This proposal deals exclusively with the ``NoRowIndex`` class. To allow users to fully "opt-out" of having to think
+about row labels, the following could also be useful:
+- a ``pd.set_option('mode.no_row_index', True)`` mode which would default to creating new ``DataFrame``s and
+  ``Series`` with ``NoRowIndex`` instead of ``RangeIndex``;
+- giving ``as_index`` options to methods which currently create an index
+  (e.g. ``value_counts``, ``.sum()``, ``.pivot_table``) to just insert a new column instead of creating an
+  ``Index``.
+
+However, neither of the above will be discussed here.
+
+## Detailed Description
+
+The core pandas code would change as little as possible. The additional complexity should be handled
+within the ``NoRowIndex`` object. It would act just like ``RangeIndex``, but would be a bit stricter
+in some cases:
+- `name` could only be `None`;
+- `start` could only be `0`, `step` `1`;
+- when appending one ``NoRowIndex`` to another ``NoRowIndex``, the result would still be ``NoRowIndex``.
+  Appending a ``NoRowIndex`` to any other index (or vice-versa) would raise;
+- the ``NoRowIndex`` class would be preserved under slicing;
+- a ``NoRowIndex`` could only be aligned with another ``Index`` if it's also ``NoRowIndex`` and if it's of the same length;
+- ``DataFrame`` columns cannot be `NoRowIndex` (so ``transpose`` would need some adjustments when called on a ``NoRowIndex`` ``DataFrame``);
+- `insert` and `delete` should raise. As a consequence, if ``df`` is a ``DataFrame`` with a
+  ``NoRowIndex``, then `df.drop` with `axis=0` would always raise;
+- arithmetic operations (e.g. `NoRowIndex(3) + 2`) would always raise;
+- when printing a ``DataFrame``/``Series`` with a ``NoRowIndex``, then the row labels would not be printed;
+- a ``MultiIndex`` could not be created with a ``NoRowIndex`` as one of its levels.
+
+Let's go into more detail for some of these. In the examples that follow, the ``NoRowIndex`` will be passed explicitly,
+but this is not how users would be expected to use it (see "Usage and Impact" section for details).
+
+### NoRowIndex.append
+
+If one has two ``DataFrame``s with ``NoRowIndex``, then one would expect that concatenating them would
+result in a ``DataFrame`` which still has ``NoRowIndex``. To do this, the following rule could be introduced:
+
+> If appending a ``NoRowIndex`` of length ``y`` to a ``NoRowIndex`` of length ``x``, the result will be a
+  ``NoRowIndex`` of length ``x + y``.
+
+Example:
+
+```python
+In [6]: df1 = pd.DataFrame({'a': [1,  2], 'b': [4, 5]}, index=NoRowIndex(2))
+
+In [7]: df2 = pd.DataFrame({'a': [4], 'b': [0]}, index=NoRowIndex(1))
+
+In [8]: df1
+Out[8]:
+ a  b
+ 1  4
+ 2  5
+
+In [9]: df2
+Out[9]:
+ a  b
+ 4  0
+
+In [10]: pd.concat([df1, df2])
+Out[10]:
+ a  b
+ 1  4
+ 2  5
+ 4  0
+
+In [11]: pd.concat([df1, df2]).index
+Out[11]: NoRowIndex(len=3)
+```
+
+Appending anything other than another ``NoRowIndex`` would raise.
+
+### Slicing a ``NoRowIndex``
+
+If one has a ``DataFrame`` with ``NoRowIndex``, then one would expect that a slice of it would still have
+a ``NoRowIndex``. This could be accomplished with:
+
+> If a slice of length ``x`` is taken from a ``NoRowIndex`` of length ``y``, then one gets a
+ ``NoRowIndex`` of length ``x``. Label-based slicing would not be allowed.
+
+Example:
+
+```python
+In [12]: df = pd.DataFrame({'a': [1,  2, 3], 'b': [4, 5, 6]}, index=NoRowIndex(3))
+
+In [13]: df.loc[df['a']>1, 'b']
+Out[13]:
+5
+6
+Name: b, dtype: int64
+
+In [14]: df.loc[df['a']>1, 'b'].index
+Out[14]: NoRowIndex(len=2)
+```
+
+Slicing by label, however, would be disallowed:
+```python
+In [15]: df.loc[0, 'b']
+---------------------------------------------------------------------------
+IndexError: Cannot use label-based indexing on NoRowIndex!
+```
+
+Note too that:
+- other uses of ``.loc``, such as boolean masks, would still be allowed (see F.A.Q);
+- ``.iloc`` and ``.iat`` would keep working as before;
+- ``.at`` would raise.
+
+### Aligning ``NoRowIndex``s
+
+To minimise surprises, the rule would be:
+
+> A ``NoRowIndex`` can only be aligned with another ``NoRowIndex`` of the same length.
+> Attempting to align it with anything else would raise.
+
+Example:
+```python
+In [1]: ser1 = pd.Series([1, 2, 3], index=NoRowIndex(3))
+
+In [2]: ser2 = pd.Series([4, 5, 6], index=NoRowIndex(3))
+
+In [3]: ser1 + ser2  # works!
+Out[3]:
+5
+7
+9
+dtype: int64
+
+In [4]: ser1 + ser2.iloc[1:]  # errors!
+---------------------------------------------------------------------------
+TypeError: Cannot join NoRowIndex of different lengths
+```
+
+### Columns cannot be NoRowIndex
+
+This proposal deals exclusively with allowing users to not need to think about
+row labels. There's no suggestion to remove the column labels.
+
+In particular, calling ``transpose`` on a ``NoRowIndex`` ``DataFrame``
+would error. The error would come with a helpful error message, informing
+users that they should first set an index. E.g.:
+```python
+In [4]: df = pd.DataFrame({'a': [1,  2, 3], 'b': [4, 5, 6]}, index=NoRowIndex(3))
+
+In [5]: df.transpose()
+---------------------------------------------------------------------------
+ValueError: Columns cannot be NoRowIndex.
+If you got here via `transpose` or an `axis=1` operation, then you should first set an index, e.g.: `df.pipe(lambda _df: _df.set_axis(pd.RangeIndex(len(_df))))`
+```
+
+### DataFrameFormatter and SeriesFormatter changes
+
+When printing an object with a ``NoRowIndex``, then the row labels would not be shown:
+
+```python
+In [15]: df = pd.DataFrame({'a': [1,  2, 3], 'b': [4, 5, 6]}, index=NoRowIndex(3))
+
+In [16]: df
+Out[16]:
+ a  b
+ 1  4
+ 2  5
+ 3  6
+```
+
+Of the above changes, this may be the only one that would need implementing within
+``DataFrameFormatter`` / ``SerieFormatter``, as opposed to within ``NoRowIndex``.
+
+## Usage and Impact
+
+Users would not be expected to work with the ``NoRowIndex`` class itself directly.
+Usage would probably involve a mode which would change how the ``default_index``
+function to return a ``NoRowIndex`` rather than a ``RangeIndex``.
+Then, if a ``mode.no_row_index`` option was introduced and a user opted in to it with
+
+```python
+pd.set_option("mode.no_row_index", True)
+```
+
+then the following would all create a ``DataFrame`` with a ``NoRowIndex`` (as they
+all call ``default_index``):
+
+- ``df.reset_index()``;
+- ``pd.concat([df1, df2], ignore_index=True)``
+- ``df1.merge(df2, on=col)``;
+- ``df = pd.DataFrame({'col_1': [1, 2, 3]})``
+
+Further discussion of such a mode is out-of-scope for this proposal. A ``NoRowIndex`` would
+just be a first step towards getting there.
+
+## Implementation
+
+Draft pull request showing proof of concept: https://github.com/pandas-dev/pandas/pull/49693.
+
+Note that implementation details could well change even if this PDEP were
+accepted. For example, ``NoRowIndex`` would not necessarily need to subclass
+``RangeIndex``, and it would not necessarily need to be accessible to the user
+(``df.index`` could well return ``None``)
+
+## Likely FAQ
+
+**Q: Could not users just use ``RangeIndex``? Why do we need a new class?**
+
+**A**: ``RangeIndex`` is not preserved under slicing and appending, e.g.:
+  ```python
+  In[1]: ser = pd.Series([1, 2, 3])
+
+  In[2]: ser[ser != 2].index
+  Out[2]: Int64Index([0, 2], dtype="int64")
+  ```
+  If someone does not want to think about row labels and starts off
+  with a ``RangeIndex``, they'll very quickly lose it.
+
+**Q: Are indices not really powerful?**
+
+**A:** Yes! And they're also confusing to many users, even experienced developers.
+  Often users are using ``.reset_index`` to avoid issues with indices and alignment.
+  Such users would benefit from being able to not think about indices
+  and alignment. Indices would be here to stay, and ``NoRowIndex`` would not be the
+  default.
+
+**Q: How could one switch a ``NoRowIndex`` ``DataFrame`` back to one with an index?**
+
+**A:** The simplest way would probably be:
+  ```python
+  df.set_axis(pd.RangeIndex(len(df)))
+  ```
+  There's probably no need to introduce a new method for this.
+
+  Conversely, to get rid of the index, then if the ``mode.no_row_index`` option was introduced, then
+  one could simply do ``df.reset_index(drop=True)``.
+
+**Q: How would ``tz_localize`` and other methods which operate on the index work on a ``NoRowIndex`` ``DataFrame``?**
+
+**A:** Same way they work on other ``NumericIndex``s, which would typically be to raise:
+
+  ```python
+  In [2]: ser.tz_localize('UTC')
+  ---------------------------------------------------------------------------
+  TypeError: index is not a valid DatetimeIndex or PeriodIndex
+  ```
+
+**Q: Why not let transpose switch ``NoRowIndex`` to ``RangeIndex`` under the hood before swapping index and columns?**
+
+**A:** This is the kind of magic that can lead to surprising behaviour that's
+  difficult to debug. For example, ``df.transpose().transpose()`` would not
+  round-trip. It's easy enough to set an index after all, better to "force" users
+  to be intentional about what they want and end up with fewer surprises later
+  on.
+
+**Q: What would df.sum(), and other methods which introduce an index, return?**
+
+**A:** Such methods would still set an index and would work the same way they
+  do now. There may be some way to change that (e.g. introducing ``as_index``
+  arguments and introducing a mode to set its default) but that's out of scope
+  for this particular PDEP.
+
+**Q: How would a user opt-in to a ``NoRowIndex`` DataFrame?**
+
+**A:** This PDEP would only allow it via the constructor, passing
+  ``index=NoRowIndex(len(df))``. A mode could be introduced to toggle
+  making that the default, but would be out-of-scope for the current PDEP.
+
+**Q: Would ``.loc`` stop working?**
+
+**A:** No. It would only raise if used for label-based selection. Other uses
+  of ``.loc``, such as ``df.loc[:, col_1]`` or ``df.loc[boolean_mask, col_1]``, would
+  continue working.
+
+**Q: What's unintuitive about ``Series`` aligning indices when summing?**
+
+**A:** Not sure, but I once asked a group of experienced developers what the
+  output of
+  ```python
+  ser1 = pd.Series([1, 1, 1], index=[1, 2, 3])
+  ser2 = pd.Series([1, 1, 1], index=[3, 4, 5])
+  print(ser1 + ser2)
+  ```
+  would be, and _nobody_ got it right.
+
+## Reasons for withdrawal
+
+After some discussions, it has become clear there is not enough for support for the proposal in its current state.
+In short, it would add too much complexity to justify the potential benefits. It would unacceptably increase
+the maintenance burden, the testing requirements, and the benefits would be minimal.
+
+Concretely:
+- maintenance burden: it would not be possible to handle all the complexity within the ``NoRowIndex`` class itself, some
+  extra logic would need to go into the pandas core codebase, which is already very complex and hard to maintain;
+- the testing burden would be too high. Properly testing this would mean almost doubling the size of the test suite.
+  Coverage for options already is not great: for example [this issue](https://github.com/pandas-dev/pandas/issues/49732)
+  was caused by a PR which passed CI, but CI did not (and still does not) cover that option (plotting backends);
+- it will not benefit most users, as users do not tend to use nor discover options which are not the default;
+- it would be difficult to reconcile with some existing behaviours: for example, ``df.sum()`` returns a Series with the
+  column names in the index.
+
+In order to make no-index the pandas default and have a chance of benefiting users, a more comprehensive set of changes
+would need to made at the same time. This would require a proposal much larger in scope, and would be a much more radical change.
+It may be that this proposal will be revisited in the future, but in its current state (as an option) it cannot be accepted.
+
+This has still been a useful exercise, though, as it has resulted in two related proposals (see below).
+
+## Related proposals
+
+- Deprecate automatic alignment, at least in some cases: https://github.com/pandas-dev/pandas/issues/49939;
+- ``.value_counts`` behaviour change: https://github.com/pandas-dev/pandas/issues/49497
+
+## PDEP History
+
+- 14 November 2022: Initial draft
+- 18 November 2022: First revision (limited the proposal to a new class, leaving a ``mode`` to a separate proposal)
+- 14 December 2022: Withdrawal (difficulty reconciling with some existing methods, lack of strong support,
+  maintenance burden increasing unjustifiably)
diff --git a/pandas/web/pandas/pdeps/0006-ban-upcasting.md b/pandas/web/pandas/pdeps/0006-ban-upcasting.md
new file mode 100644
index 0000000000000..ae5872186bf23
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0006-ban-upcasting.md
@@ -0,0 +1,247 @@
+# PDEP-6: Ban upcasting in setitem-like operations
+
+- Created: 23 December 2022
+- Status: Implemented
+- Discussion: [#39584](https://github.com/pandas-dev/pandas/pull/50402)
+- Author: [Marco Gorelli](https://github.com/MarcoGorelli) ([original issue](https://github.com/pandas-dev/pandas/issues/39584) by [Joris Van den Bossche](https://github.com/jorisvandenbossche))
+- Revision: 1
+
+## Abstract
+
+The suggestion is that setitem-like operations would
+not change a ``Series``' dtype (nor that of a ``DataFrame``'s column).
+
+Current behaviour:
+```python
+In [1]: ser = pd.Series([1, 2, 3], dtype='int64')
+
+In [2]: ser[2] = 'potage'
+
+In [3]: ser  # dtype changed to 'object'!
+Out[3]:
+0         1
+1         2
+2    potage
+dtype: object
+```
+
+Suggested behaviour:
+
+```python
+In [1]: ser = pd.Series([1, 2, 3])
+
+In [2]: ser[2] = 'potage'  # raises!
+---------------------------------------------------------------------------
+ValueError: Invalid value 'potage' for dtype int64
+```
+
+## Motivation and Scope
+
+Currently, pandas is extremely flexible in handling different dtypes.
+However, this can potentially hide bugs, break user expectations, and copy data
+in what looks like it should be an inplace operation.
+
+An example of it hiding a bug is:
+```python
+In[9]: ser = pd.Series(pd.date_range("2000", periods=3))
+
+In[10]: ser[2] = "2000-01-04"  # works, is converted to datetime64
+
+In[11]: ser[2] = "2000-01-04x"  # typo - but pandas does not error, it upcasts to object
+```
+
+The scope of this PDEP is limited to setitem-like operations on Series (and DataFrame columns).
+For example, starting with:
+```python
+df = DataFrame({"a": [1, 2, np.nan], "b": [4, 5, 6]})
+ser = df["a"].copy()
+```
+then the following would all raise:
+
+* setitem-like operations:
+
+    - ``ser.fillna('foo', inplace=True)``
+    - ``ser.where(ser.isna(), 'foo', inplace=True)``
+    - ``ser.fillna('foo', inplace=False)``
+    - ``ser.where(ser.isna(), 'foo', inplace=False)``
+
+* setitem indexing operations (where ``indexer`` could be a slice, a mask,
+  a single value, a list or array of values, or any other allowed indexer):
+
+    - ``ser.iloc[indexer] = 'foo'``
+    - ``ser.loc[indexer] = 'foo'``
+    - ``df.iloc[indexer, 0] = 'foo'``
+    - ``df.loc[indexer, 'a'] = 'foo'``
+    - ``ser[indexer] = 'foo'``
+
+It may be desirable to expand the top list to ``Series.replace`` and ``Series.update``,
+but to keep the scope of the PDEP down, they are excluded for now.
+
+Examples of operations which would not raise are:
+
+- ``ser.diff()``
+- ``pd.concat([ser, ser.astype(object)])``
+- ``ser.mean()``
+- ``ser[0] = 3``  # same dtype
+- ``ser[0] = 3.``  # 3.0 is a 'round' float and so compatible with 'int64' dtype
+- ``df['a'] = pd.date_range(datetime(2020, 1, 1), periods=3)``
+- ``df.index.intersection(ser.index)``
+
+## Detailed description
+
+Concretely, the suggestion is:
+
+- If a ``Series`` is of a given dtype, then a ``setitem``-like operation should not change its dtype.
+- If a ``setitem``-like operation would previously have changed a ``Series``' dtype, it would now raise.
+
+For a start, this would involve:
+
+1. changing ``Block.setitem`` such that it does not have an ``except`` block in:
+
+    <!-- language: python -->
+
+        value = extract_array(value, extract_numpy=True)
+        try:
+            casted = np_can_hold_element(values.dtype, value)
+        except LossSetitiemError:
+            # current dtype cannot store value, coerce to common dtype
+            nb = self.coerce_to_target_dtype(value)
+            return nb.setitem(index, value)
+        else:
+
+2. making a similar change in:
+
+    - ``Block.where``
+    - ``Block.putmask``
+    - ``EABackedBlock.setitem``
+    - ``EABackedBlock.where``
+    - ``EABackedBlock.putmask``
+
+The above would already require several hundreds of tests to be adjusted. Note that once
+implementation starts, the list of locations to change may turn out to be slightly
+different.
+
+### Ban upcasting altogether, or just upcasting to ``object``?
+
+The trickiest part of this proposal concerns what to do when setting a float in an integer column:
+
+```python
+In[1]: ser = pd.Series([1, 2, 3])
+
+In [2]: ser
+Out[2]:
+0    1
+1    2
+2    3
+dtype: int64
+
+In[3]: ser[0] = 1.5  # what should this do?
+```
+
+The current behaviour is to upcast to 'float64':
+```python
+In [4]: ser
+Out[4]:
+0    1.5
+1    2.0
+2    3.0
+dtype: float64
+```
+
+This is not necessarily a sign of a bug, because the user might just be thinking of their ``Series`` as being
+numeric (without much regard for ``int`` vs ``float``) - ``'int64'`` is just what pandas happened to infer
+when constructing it.
+
+Possible options could be:
+
+1. Only accept round floats (e.g. ``1.0``) and raise on anything else (e.g. ``1.01``).
+2. Convert the float value to ``int`` before setting it (i.e. silently round all float values).
+3. Limit "banning upcasting" to when the upcasted dtype is ``object`` (i.e. preserve current behavior of upcasting the int64 Series to float64).
+
+Let us compare with what other libraries do:
+
+- ``numpy``: option 2
+- ``cudf``: option 2
+- ``polars``: option 2
+- ``R data.frame``: just upcasts (like pandas does now for non-nullable dtypes);
+- ``pandas`` (nullable dtypes): option 1
+- ``datatable``: option 1
+- ``DataFrames.jl``: option 1
+
+Option ``2`` would be a breaking behaviour change in pandas. Further,
+if the objective of this PDEP is to prevent bugs, then this is also not desirable:
+someone might set ``1.5`` and later be surprised to learn that they actually set ``1``.
+
+There are several downsides to option ``3``:
+
+- It would be inconsistent with the nullable dtypes' behaviour.
+- It would also add complexity to the codebase and to tests.
+- It would be hard to teach, as instead of being able to teach a simple rule,
+  There would be a rule with exceptions.
+- There would be a risk of loss of precision and or overflow.
+- It opens the door to other exceptions, such as not upcasting ``'int8'`` to ``'int16'``.
+
+Option ``1`` is the maximally safe one in terms of protecting users from bugs, being
+consistent with the current behaviour of nullable dtypes, and in being simple to teach.
+Therefore, the option chosen by this PDEP is option 1.
+
+## Usage and Impact
+
+This would make pandas stricter, so there should not be any risk of introducing bugs. If anything, this would help prevent bugs.
+
+Unfortunately, it would also risk annoying users who might have been intentionally upcasting.
+
+Given that users could still get the current behaviour by first explicitly casting the Series
+to float, it would be more beneficial to the community at large to err on the side
+of strictness.
+
+## Out of scope
+
+Enlargement. For example:
+```python
+ser = pd.Series([1, 2, 3])
+ser[len(ser)] = 4.5
+```
+There is arguably a larger conversation to be had about whether that should be allowed
+at all. To keep this proposal focused, it is intentionally excluded from the scope.
+
+## F.A.Q.
+
+**Q: What happens if setting ``1.0`` in an ``int8`` Series?**
+
+**A**: The current behavior would be to insert ``1.0`` as ``1`` and keep the dtype
+  as ``int8``. So, this would not change.
+
+**Q: What happens if setting ``1_000_000.0`` in an ``int8`` Series?**
+
+**A**: The current behavior would be to upcast to ``int32``. So under this PDEP,
+  it would instead raise.
+
+**Q: What happens in setting ``16.000000000000001`` in an ``int8`` Series?**
+
+**A**: As far as Python is concerned, ``16.000000000000001`` and ``16.0`` are the
+  same number. So, it would be inserted as ``16`` and the dtype would not change
+  (just like what happens now, there would be no change here).
+
+**Q: What if I want ``1.0000000001`` to be inserted as ``1.0`` in an ``int8`` Series?**
+
+**A**: You may want to define your own helper function, such as:
+
+```python
+def maybe_convert_to_int(x: int | float, tolerance: float):
+    if np.abs(x - round(x)) < tolerance:
+        return round(x)
+    return x
+```
+
+which you could adapt according to your needs.
+
+
+## Timeline
+
+Deprecate sometime in the 2.x releases (after 2.0.0 has already been released), and enforce in 3.0.0.
+
+### PDEP History
+
+- 23 December 2022: Initial draft
+- 4 July 2024: Change status to "implemented"
diff --git a/pandas/web/pandas/pdeps/0007-copy-on-write.md b/pandas/web/pandas/pdeps/0007-copy-on-write.md
new file mode 100644
index 0000000000000..f5adb6a571120
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0007-copy-on-write.md
@@ -0,0 +1,589 @@
+# PDEP-7: Consistent copy/view semantics in pandas with Copy-on-Write
+
+- Created: July 2021
+- Status: Implemented
+- Discussion: [#36195](https://github.com/pandas-dev/pandas/issues/36195)
+- Author: [Joris Van den Bossche](https://github.com/jorisvandenbossche)
+- Revision: 1
+
+## Abstract
+
+Short summary of the proposal:
+
+1. The result of _any_ indexing operation (subsetting a DataFrame or Series in any way,
+   i.e. including accessing a DataFrame column as a Series) or any method returning a
+   new DataFrame or Series, always _behaves as if_ it were a copy in terms of user
+   API.
+2. We implement Copy-on-Write (as implementation detail). This way, we can actually use
+   views as much as possible under the hood, while ensuring the user API behaves as a
+   copy.
+3. As a consequence, if you want to modify an object (DataFrame or Series), the only way
+   to do this is to directly modify that object itself .
+
+This addresses multiple aspects: 1) a clear and consistent user API (a clear rule: _any_
+subset or returned series/dataframe **always** behaves as a copy of the original, and
+thus never modifies the original) and 2) improving performance by avoiding excessive
+copies (e.g. a chained method workflow would no longer return an actual data copy at each
+step).
+
+Because every single indexing step behaves as a copy, this also means that with this
+proposal, "chained assignment" (with multiple setitem steps) will _never_ work and
+the `SettingWithCopyWarning` can be removed.
+
+## Background
+
+pandas' current behavior on whether indexing returns a view or copy is confusing. Even
+for experienced users, it's hard to tell whether a view or copy will be returned (see
+below for a summary). We'd like to provide an API that is consistent and sensible about
+returning views vs. copies.
+
+We also care about performance. Returning views from indexing operations is faster and
+reduces memory usage. The same is true for several methods that don't modify the data
+such as setting/resetting the index, renaming columns, etc. that can be used in a method
+chaining workflow and currently return a new copy at each step.
+
+Finally, there are API / usability issues around views. It can be challenging to know
+the user's intent in operations that modify a subset of a DataFrame (column and/or row
+selection), like:
+
+```python
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> df2 = df[["A", "B"]]
+>>> df2.loc[df2["A"] > 1, "A"] = 1
+```
+
+Did the user intend to modify `df` when they modified `df2` (setting aside issues with
+the current implementation)? In other words, if we had a perfectly consistent world
+where indexing the columns always returned views or always returned a copy, does the
+code above imply that the user wants to mutate `df`?
+
+There are two possible behaviours the user might intend:
+
+1. Case 1: I know my subset might be a view of the original and I want to modify the
+   original as well.
+2. Case 2: I just want to modify the subset without modifying the original.
+
+Today, pandas' inconsistency means _neither_ of these workflows is really possible. The
+first is difficult, because indexing operations often (though not always) return copies,
+and even when a view is returned you sometimes get a `SettingWithCopyWarning` when
+mutating. The second is somewhat possible, but requires many defensive copies (to avoid
+`SettingWithCopyWarning`, or to ensure that you have a copy when a view _was_ returned).
+
+## Proposal
+
+For these reasons (consistency, performance, code clarity), this PDEP proposes the
+following changes:
+
+1. The result of _any_ indexing operation (subsetting a DataFrame or Series in any way,
+   i.e. including accessing a DataFrame column as a Series) or any method returning a
+   new DataFrame or Series, always _behaves as if_ it were a copy in terms of user
+   API.
+2. We implement Copy-on-Write. This way, we can actually use views as much as possible
+   under the hood, while ensuring the user API behaves as a copy.
+
+The intent is to capture the performance benefits of views as much as possible, while
+providing consistent and clear behaviour to the user. This essentially makes returning
+views an internal optimization, without the user needing to know if the specific
+indexing operation would return a view or a copy. The new rule would be simple: any
+series/dataframe derived from another series/dataframe, through an indexing operation or
+a method, always behaves as a copy of the original series/dataframe.
+
+The mechanism to ensure this consistent behaviour, Copy-on-Write, would entail the
+following: the setitem operation (i.e. `df[..] = ..` or `df.loc[..] = ..` or
+`df.iloc[..] = ..`, or equivalent for Series) would check if the data that is being
+modified is a view on another dataframe (or is being viewed by another dataframe). If it
+is, then we would copy the data before mutating.
+
+Taking the example from above, if the user wishes to not mutate the parent, we no longer
+require a defensive copy just to avoid a `SettingWithCopyWarning`.
+
+```python
+# Case 2: The user does not want mutating df2 to mutate the parent df, via CoW
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> df2 = df[["A", "B"]]
+>>> df2.loc[df2["A"] > 1, "A"] = 1
+>>> df.iloc[1, 0]  # df was not mutated
+2
+```
+
+On the other hand, if the user actually wants to modify the original df, they can no
+longer rely on the fact that `df2` could be a view, as mutating a subset would now never
+mutate the parent. The only way to modify the original df is by combining all indexing
+steps in a single indexing operation on the original (no "chained" setitem):
+
+```python
+# Case 1: user wants mutations of df2 to be reflected in df -> no longer possible
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> df2 = df[["A", "B"]]
+>>> df2.loc[df2["A"] > 1, "A"] = 1  # mutating df2 will not mutate df
+>>> df.loc[df["A"] > 1, "A"] = 1  # need to directly mutate df instead
+```
+
+### This proposal also extends to methods
+
+In principle, there's nothing special about indexing when it comes to defensive copying.
+_Any_ method that returns a new series/dataframe without altering existing data (rename,
+set_index, assign, dropping columns, etc.) currently returns a copy by default and is a
+candidate for returning a view:
+
+```python
+>>> df2 = df.rename(columns=str.lower)
+>>> df3 = df2.set_index("a")
+```
+
+Now, generally, pandas users won't expect `df2` or `df3` to be a view such that mutating
+`df2` or `df3` would mutate `df`. Copy-on-Write allows us to also avoid
+unnecessary copies in methods such as the above (or in the variant using method chaining
+like `df.rename(columns=str.lower).set_index("a")`).
+
+### Propagating mutation forwards
+
+Thus far we have considered the (more common) case of taking a subset, mutating the
+subset, and how that should affect the parent. What about the other direction, where the
+parent is mutated?
+
+```python
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+>>> df2 = df[["A"]]
+>>> df.iloc[0, 0] = 10
+>>> df2.iloc[0, 0]  # what is this value?
+```
+
+Given that `df2` is _considered_ as a copy of df under this proposal (i.e. behaves as a
+copy), also mutating the parent `df` will not mutate the subset `df2`.
+
+### When do mutations propagate to other objects and when not?
+
+This proposal basically means that mutations _never_ propagate to _other_ objects (as
+would happen with views). The only way to modify a DataFrame or Series is to modify the
+object itself directly.
+
+But let's illustrate this in Python terms. Consider that we have a DataFrame `df1`, and we
+assign that to another name `df2`:
+
+```python
+>>> df1 = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+>>> df2 = df1
+```
+
+Although we have now two variables (`df1` and `df2`), this assignment follows the standard
+python semantics, and both names are pointing to the same object ("df1 and df2 are
+_identical_"):
+
+```python
+>>> id(df1) == id(df2)  # or: df1 is df2
+True
+```
+
+Thus, if you modify DataFrame `df2`, this is also reflected in the other variable `df1`, and
+the other way around (since it's the same object):
+
+```python
+>>> df1.iloc[0, 0]
+1
+>>> df2.iloc[0, 0] = 10
+>>> df1.iloc[0, 0]
+10
+```
+
+In summary, modifications are only "propagated" between _identical_ objects (not just
+equal (`==`), but identical (`is`) in python terms, see
+[docs](https://docs.python.org/3/reference/expressions.html#is)). Propagation is not
+really the proper term, since there is only one object that was modified.
+
+However, when in some way creating a new object (even though it might be a DataFrame
+with the same data, and thus be an "equal" DataFrame):
+
+```python
+>>> df1 = pd.DataFrame({"A": [1, 2], "B": [3, 4]})
+>>> df2 = df1[:]  # or df1.loc[...] with some indexer
+```
+
+Those objects are no longer identical:
+
+```python
+>>> id(df1) == id(df2)  # or df1 is df2
+False
+```
+
+And thus modifications to one will not propagate to the other:
+
+```python
+>>> df1.iloc[0, 0]
+1
+>>> df2.iloc[0, 0] = 10
+>>> df1.iloc[0, 0]  # not changed
+1
+```
+
+Currently, any getitem indexing operation returns _new_ objects, and also almost all
+DataFrame/Series methods return a _new_ object (except with `inplace=True` in some
+cases), and thus follow the above logic of never modifying its parent/child DataFrame or
+Series (using the lazy Copy-on-Write mechanism where possible).
+
+## Copy / view behaviour in NumPy versus pandas
+
+NumPy has the concept of "views" (an array that shares data with another array, viewing
+the same memory, see e.g.
+[this explanation](https://scipy-cookbook.readthedocs.io/items/ViewsVsCopies.html) for
+more details). Typically you create views as a slice of another array. But other
+indexing methods, often called "fancy indexing", do not return views but copies: using a
+list of indices or a boolean mask.
+
+Pandas, being built on NumPy, uses those concepts, and also exposes the behaviour
+consequences to its users. This basically means that pandas users, to understand the
+details of how indexing works, also need to understand those view / fancy indexing
+concepts of numpy.
+
+However, because DataFrames are not an array, the copy/view rules still differ from
+NumPy's rules with current pandas. Slicing rows generally gives a view (following
+NumPy), but slicing columns doesn't always give a view (this could be changed to match
+NumPy however, see "Alternatives" 1b below). Fancy indexing rows (e.g. with a list of
+(positional) labels) gives a copy, but fancy indexing columns _could_ give a view
+(currently this gives a copy as well, but one of the "Alternatives" (1b) is to have this
+always return a view).
+
+The proposal in this document is to decouple the pandas user-facing behaviour from those
+NumPy concepts. Creating a subset of a DataFrame with a slice or with a mask would
+behave in a similar way for the user (both return a new object and behave as a copy of
+the original). We still use the concept of views internally in pandas to optimize the
+implementation, but this becomes hidden from the user.
+
+## Alternatives
+
+The [original document](https://docs.google.com/document/d/1csGE4qigPR2vzmU2--jwURn3sK5k5eVewinxd8OUPk0/edit) and GitHub issue ([Proposal for future copy / view semantics in indexing operations - #36195](https://github.com/pandas-dev/pandas/issues/36195)) discussed several options for making the copy/view situation more consistent and clear:
+
+1. **Well-Defined copy/view rules:** ensure we have more consistent rules about which
+   operations result in a copy and which in a view, and then views result in mutating
+   the parent, copies not.
+   a. A minimal change would be to officialize the current behaviour. This comes down to
+      fixing some bugs and clearly documenting and testing which operations are views,
+      and which are copies.
+   b. An alternative would be to simplify the set of rules. For example: selecting
+      columns is always a view, subsetting rows is always a copy. Or: selecting columns
+      is always a view, subsetting rows as a slice is a view otherwise always a copy.
+
+2. **Copy-on-Write**: The setitem operation would check if it's a view on another
+   dataframe. If it is, then we would copy our data before mutating. (i.e. this
+   proposal)
+
+3. **Error-on-Write**: The setitem operation would check if it's a subset of another
+   dataframe (both view of copy). Only rather than copying in case of a view we would
+   raise an exception telling the user to either copy the data with
+   ``.copy_if_needed()`` (name TBD) or mark the frame as "a mutable view" with
+   ``.as_mutable_view()`` (name TBD).
+
+This document basically proposes an extended version of option 2 (Copy-on-Write). Some
+arguments in favor of Copy-on-Write compared to the other options:
+
+* Copy-on-Write will improve the copy/view efficiency of _methods_ (e.g. rename,
+  (re)set_index, drop columns, etc. See section above). This will result in
+  lower memory usage and better performance.
+
+* This proposal can also be seen as a clear "well-defined rule". Using Copy-on-Write
+  under the hood is an implementation detail to delay the actual copy until it is
+  needed. The rule of "always copy" is the simplest "well-defined rule" we can get.
+
+  Other "well-defined rule" ideas above would always include some specific cases (and
+  deviations from the NumPy rules). And even with clear rules a user still needs to know
+  the details of those rules to understand that `df['a'][df['b'] < 0] = 0` or
+  `df[df['b'] < 0]['a'] = 0` does something differently (switched order of column/row
+  indexing: the first mutates df (if selecting a column is a view) and the second
+  doesn't). While with the "always copy" rule with Copy-on-Write, neither of those
+  examples will work to update `df`.
+
+On the other hand, the proposal in this document does not give the user control over
+whether a subset should be a view (when possible) that mutates the parent when being
+mutated. The only way to modify the parent dataframe is with a direct indexing operation
+on this dataframe itself.
+
+See the GitHub comment with some more detailed argumentation:
+[https://github.com/pandas-dev/pandas/issues/36195#issuecomment-786654449](https://github.com/pandas-dev/pandas/issues/36195#issuecomment-786654449)
+
+## Disadvantages
+
+Other than the fact that this proposal would result in a backwards incompatible,
+breaking change in behaviour (see next section), there are some other potential
+disadvantages:
+
+* Deviation from NumPy: NumPy uses the copy and view concepts, while in this proposal
+  views would basically not exist anymore in pandas (for the user, at least; we would
+  still use it internally as an implementation detail)
+  * But as a counter argument: many pandas users are probably not familiar with those
+    concepts, and pandas already deviates from the exact rules in NumPy.
+* Performance cost of indexing and methods becomes harder to predict: because the copy
+  of the data doesn't happen at the moment when actually creating the new object, but
+  can happen at a later stage when modifying either the parent or child object, it
+  becomes less transparent about when pandas copies data (but in general we should copy
+  less often). This is somewhat mitigated because Copy-on-Write will only copy the columns
+  that are mutated. Unrelated columns won't get copied.
+* Increased memory usage for some use cases: while the majority of use cases will
+  see an improvement in memory usage with this proposal, there are a few use
+  cases where this might not be the case. Specifically in cases where pandas currently
+  does return a view (e.g. slicing rows) and in the case you are fine with (or don't care
+  about) the current behaviour of it being a view when mutating that subset (i.e.
+  mutating the sliced subset also mutates the parent dataframe), in such a case the
+  proposal would introduce a new copy compared to the current behaviour. There is a
+  workaround for this though: the copy is not needed if the previous object goes out
+  of scope, e.g. the variable is reassigned to something else.
+
+## Backward compatibility
+
+The proposal in this document is clearly a backwards incompatible change that breaks
+existing behaviour. Because of the current inconsistencies and subtleties around views
+vs. copies and mutation, it would be difficult to change anything without breaking
+changes. The current proposal is not the proposal with the minimal changes, though. A
+change like this will in any case need to be accompanied with a major version bump (for
+example pandas 3.0).
+
+Doing a traditional deprecation cycle that lives in several minor feature releases will
+be too noisy. Indexing is too common an operation to include a warning (even if we limit
+it to just those operations that previously returned views). However, this proposal is
+already implemented and thus available. Users can opt-in and test their code (this is
+possible starting with version 1.5 with `pd.options.mode.copy_on_write = True`).
+
+Further we will add a warning mode for pandas 2.2 that raises warnings for all cases that
+will change behaviour under the Copy-on-Write proposal. We can
+provide a clearly documented upgrade path to first enable the warnings, fix all
+warnings, and then enable the Copy-on-Write mode and ensure your code is still working,
+and then finally upgrade to the new major release.
+
+## Implementation
+
+The implementation is available since pandas 1.5 (and significantly improved starting
+with pandas 2.0). It uses weakrefs to keep track of whether the
+data of a Dataframe/Series are viewing the data of another (pandas) object or are being
+viewed by another object. This way, whenever the series/dataframe gets modified, we can
+check if its data first needs to be copied before mutating it
+(see [here](https://pandas.pydata.org/docs/development/copy_on_write.html)).
+
+To test the implementation and experiment with the new behaviour, you can
+enable it with the following option:
+
+```python
+>>> pd.options.mode.copy_on_write = True
+```
+
+after importing pandas (or setting the `PANDAS_COPY_ON_WRITE=1` environment variable
+before importing pandas).
+
+## Concrete examples
+
+### Chained assignment
+
+Consider a "classic" case of chained indexing, which was the original motivation for the SettingWithCopy warning:
+
+```python
+>>> df[df['B'] > 3]['B'] = 10
+```
+
+That is roughly equivalent to
+
+```python
+>>> df2 = df[df['B'] > 3]  # Copy under NumPy's rules
+>>> df2['B'] = 10  # Update (the copy) df2, df not changed
+>>> del df2  # All references to df2 are lost, goes out of scope
+```
+
+And so `df` is not modified. For this reason, the SettingWithCopyWarning was introduced.
+
+_With this proposal_, any result of an indexing operation behaves as a copy
+(Copy-on-Write), and thus chained assignment will _never_ work. Given that there is then
+no ambiguity, the idea is to drop the warning.
+
+The above example is a case where chained assignment doesn't work with current pandas.
+But there are of course also patterns with chained assignment that currently _do_ work
+and are used. _With this proposal_, any chained assignment will not work, and so those
+cases will stop working (e.g. the case above but switching the order):
+
+```python
+>>> df['B'][df['B'] > 3] = 10
+# or
+>>> df['B'][0:5] = 10
+```
+
+These cases will raise a warning ``ChainedAssignmentError``, because they can never
+accomplish what the user intended. There will be false-positive cases when these
+operations are triggered from Cython, because Cython uses a different reference counting
+mechanism. These cases should be rare, since calling pandas code from Cython does not
+have any performance benefits.
+
+### Filtered dataframe
+
+A typical example where the current SettingWithCopyWarning becomes annoying is when
+filtering a DataFrame (which always already returns a copy):
+
+```python
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> df_filtered = df[df["A"] > 1]
+>>> df_filtered["new_column"] = 1
+SettingWithCopyWarning:
+A value is trying to be set on a copy of a slice from a DataFrame.
+Try using .loc[row_indexer,col_indexer] = value instead
+```
+
+If you then modify your filtered dataframe (e.g. adding a column), you get the
+unnecessary SettingWithCopyWarning (with confusing message). The only way to get rid of
+the warning is by doing a defensive copy (`df_filtered = df[df["A"] > 1].copy()`, which
+results in copying the data twice in the current implementation, Copy-on-Write would
+not require ``.copy()`` anymore).
+
+_With this proposal_, the filtered dataframe is never a view and the above
+workflow would work as expected without warning (and thus without needing the extra
+copy).
+
+### Modifying a Series (from DataFrame column)
+
+_Currently_, accessing a column of a DataFrame as a Series is one of the few cases that
+is actually guaranteed to always be a view:
+
+```python
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> s = df["A"]
+>>> s.loc[0] = 0   # will also modify df (but no longer with this proposal)
+```
+
+_With this proposal_, any indexing operation results in a copy, so also accessing a
+column as a Series (in practice, it will still be a view of course, but behave as a copy
+through Copy-on-Write). In the above example, mutating `s` will no longer modify the
+parent `df`.
+
+This situation is similar as the "chained assignment" case above, except with
+an explicit intermediate variable. To actually change the original DataFrame,
+the solution is the same: mutate directly the DataFrame in a single step.
+For example:
+
+```python
+>>> df.loc[0, "A"] = 0
+```
+
+### "Shallow" copies
+
+_Currently_, it is possible to create a "shallow" copy of a DataFrame with
+`copy(deep=False)`. This creates a new DataFrame object but without copying the
+underlying index and data. Any changes to the data of the original will be reflected in
+the shallow copy (and vice versa). See the
+[docs](https://pandas.pydata.org/pandas-docs/version/1.5/reference/api/pandas.DataFrame.copy.html).
+
+```python
+>>> df = pd.DataFrame({"A": [1, 2], "B": [3, 4], "C": [5, 6]})
+>>> df2 = df.copy(deep=False)
+>>> df2.iloc[0, 0] = 0   # will also modify df (but no longer with this proposal)
+```
+
+_With this proposal_, this kind of shallow copy is no longer possible. Only "identical"
+objects (in Python terms: `df2 is df`) can share data without triggering Copy-on-Write.
+A shallow copy will rather become a "delayed" copy through Copy-on-Write.
+
+See
+[#36195 (comment)](https://github.com/pandas-dev/pandas/issues/36195#issuecomment-830579242)
+for a more detailed comment on this.
+
+### Methods returning a new DataFrame with the same data
+
+This example is already shown above as well, but so _currently_ almost all methods on a
+Series/DataFrame by default return a new object that is a copy of the original data:
+
+```python
+>>> df2 = df.rename(columns=str.lower)
+>>> df3 = df2.set_index("a")
+```
+
+In the above example, df2 holds a copy of the data of df, and df3 holds a copy of the
+data of df2. Mutating any of those DataFrames would not modify the parent dataframe.
+
+_With this proposal_, those methods would continue to return new objects, but would use
+the shallow copy mechanism with Copy-on-Write so that in practice, those methods don't
+need to copy the data at each step, while preserving the current behaviour.
+
+### Series and DataFrame constructors
+
+_Currently_, the Series and DataFrame constructors don't always copy the input
+(depending on the type of the input). For example:
+
+```python
+>>> s = pd.Series([1, 2, 3])
+>>> s2 = pd.Series(s)
+>>> s2.iloc[0] = 0   # will also modify the parent Series s
+>>> s
+0	0  # <-- modified
+1	2
+2	3
+dtype: int64
+```
+
+_With this proposal_, we can also use the shallow copy with Copy-on-Write approach _by
+default_ in the constructors. This would mean that by default, a new Series or DataFrame
+(like `s2` in the above example) would not modify the data from which it is being
+constructed (when being modified itself), honoring the proposed rules.
+
+## More background: Current behaviour of views vs copy
+
+To the best of our knowledge, indexing operations currently return views in the
+following cases:
+
+* Selecting a single column (as a Series) out of a DataFrame is always a view
+  (``df['a']``)
+* Slicing columns from a DataFrame creating a subset DataFrame (``df[['a':'b']]`` or
+  ``df.loc[:, 'a': 'b']``) is a view _if_ the the original DataFrame consists of a
+  single block (single dtype, consolidated) and _if_ you are slicing (so not a list
+  selection). In all other cases, getting a subset is always a copy.
+* Selecting rows _can_ return a view, when the row indexer is a `slice` object.
+
+Remaining operations (subsetting rows with a list indexer or boolean mask) in practice
+return a copy, and we will raise a SettingWithCopyWarning when the user tries to modify
+the subset.
+
+## More background: Previous attempts
+
+We've discussed this general issue before. [https://github.com/pandas-dev/pandas/issues/10954](https://github.com/pandas-dev/pandas/issues/10954) and a few pull requests ([https://github.com/pandas-dev/pandas/pull/12036](https://github.com/pandas-dev/pandas/pull/12036), [https://github.com/pandas-dev/pandas/pull/11207](https://github.com/pandas-dev/pandas/pull/11207), [https://github.com/pandas-dev/pandas/pull/11500](https://github.com/pandas-dev/pandas/pull/11500)).
+
+## Comparison with other languages / libraries
+
+### R
+
+For the user, R has somewhat similar behaviour. Most R objects can be considered
+immutable, through "copy-on-modify"
+([https://adv-r.hadley.nz/names-values.html#copy-on-modify](https://adv-r.hadley.nz/names-values.html#copy-on-modify)).
+But in contrast to Python, in R this is a language feature, and any assignment (binding
+a variable to a new name) or passing as function argument will essentially create a
+"copy" (when mutating such an object, at that point the actual data get copied and
+rebind to the name):
+
+```r
+x <- c(1, 2, 3)
+y <- x
+y[[1]] <- 10  # does not modify x
+```
+
+While if you would do the above example in Python with a list, x and y are "identical"
+and mutating one will also mutate the other.
+
+As a consequence of this language behaviour, modifying a `data.frame` will not modify
+other data.frames that might share memory (before being copied with "copy-on-modify").
+
+### Polars
+
+Polars ([https://github.com/pola-rs/polars](https://github.com/pola-rs/polars)) is a
+DataFrame library with a Python interface, mainly written in Rust on top of Arrow. It
+explicitly
+[mentions](https://pola-rs.github.io/polars-book/user-guide/introduction.html#current-status)
+"Copy-on-Write" semantics as one its features.
+
+Based on some experiments, the user-facing behaviour of Polars seems similar to the behaviour
+described in this proposal (mutating a DataFrame/Series never mutates a parent/child
+object, and so chained assignment also doesn't work)
+
+
+## PDEP-7 History
+
+- July 2021: Initial version
+- February 2023: Converted into a PDEP
+
+Note: this proposal has been discussed before it was turned into a PDEP. The main
+discussion happened in [GH-36195](https://github.com/pandas-dev/pandas/issues/36195).
+This document is modified from the original document discussing different options for
+clear copy/view semantics started by Tom Augspurger
+([google doc](https://docs.google.com/document/d/1csGE4qigPR2vzmU2--jwURn3sK5k5eVewinxd8OUPk0/edit)).
+
+Related mailing list discussion: [https://mail.python.org/pipermail/pandas-dev/2021-July/001358.html](https://t.co/vT5dOMhNjV?amp=1)
diff --git a/pandas/web/pandas/pdeps/0009-io-extensions.md b/pandas/web/pandas/pdeps/0009-io-extensions.md
new file mode 100644
index 0000000000000..aeda990cea7df
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0009-io-extensions.md
@@ -0,0 +1,406 @@
+# PDEP-9: Allow third-party projects to register pandas connectors with a standard API
+
+- Created: 5 March 2023
+- Status: Rejected
+- Discussion: [#51799](https://github.com/pandas-dev/pandas/pull/51799)
+              [#53005](https://github.com/pandas-dev/pandas/pull/53005)
+- Author: [Marc Garcia](https://github.com/datapythonista)
+- Revision: 1
+
+## PDEP Summary
+
+This document proposes that third-party projects implementing I/O or memory
+connectors to pandas can register them using Python's entrypoint system,
+and make them available to pandas users with the usual pandas I/O interface.
+For example, packages independent from pandas could implement readers from
+DuckDB and writers to Delta Lake, and when installed in the user environment
+the user would be able to use them as if they were implemented in pandas.
+For example:
+
+```python
+import pandas
+
+pandas.load_io_plugins()
+
+df = pandas.DataFrame.read_duckdb("SELECT * FROM 'my_dataset.parquet';")
+
+df.to_deltalake('/delta/my_dataset')
+```
+
+This would allow to easily extend the existing number of connectors, adding
+support to new formats and database engines, data lake technologies,
+out-of-core connectors, the new ADBC interface, and others, and at the
+same time reduce the maintenance cost of the pandas code base.
+
+## Current state
+
+pandas supports importing and exporting data from different formats using
+I/O connectors, currently implemented in `pandas/io`, as well as connectors
+to in-memory structures like Python structures or other library formats.
+In many cases, those connectors wrap an existing Python library, while in
+some others, pandas implements the logic to read and write to a particular
+format.
+
+In some cases, different engines exist for the same format. The API to use
+those connectors is `pandas.read_<format>(engine='<engine-name>', ...)` to
+import data, and `DataFrame.to_<format>(engine='<engine-name>', ...)` to
+export data.
+
+For objects exported to memory (like a Python dict) the API is the same as
+for I/O, `DataFrame.to_<format>(...)`. For formats imported from objects in
+memory, the API is different using the `from_` prefix instead of `read_`,
+`DataFrame.from_<format>(...)`.
+
+In some cases, the pandas API provides `DataFrame.to_*` methods that are not
+used to export the data to a disk or memory object, but instead to transform
+the index of a `DataFrame`: `DataFrame.to_period` and `DataFrame.to_timestamp`.
+
+Dependencies of the connectors are not loaded by default, and are
+imported when the connector is used. If the dependencies are not installed
+an `ImportError` is raised.
+
+```python
+>>> pandas.read_gbq(query)
+Traceback (most recent call last):
+  ...
+ImportError: Missing optional dependency 'pandas-gbq'.
+pandas-gbq is required to load data from Google BigQuery.
+See the docs: https://pandas-gbq.readthedocs.io.
+Use pip or conda to install pandas-gbq.
+```
+
+### Supported formats
+
+The list of formats can be found in the
+[IO guide](https://pandas.pydata.org/docs/dev/user_guide/io.html).
+A more detailed table, including in memory objects, and I/O connectors in the
+DataFrame styler is presented next:
+
+| Format       | Reader | Writer | Engines                                                                           |
+|--------------|--------|--------|-----------------------------------------------------------------------------------|
+| CSV          | X      | X      | `c`, `python`, `pyarrow`                                                          |
+| FWF          | X      |        | `c`, `python`, `pyarrow`                                                          |
+| JSON         | X      | X      | `ujson`, `pyarrow`                                                                |
+| HTML         | X      | X      | `lxml`, `bs4/html5lib` (parameter `flavor`)                                       |
+| LaTeX        |        | X      |                                                                                   |
+| XML          | X      | X      | `lxml`, `etree` (parameter `parser`)                                              |
+| Clipboard    | X      | X      |                                                                                   |
+| Excel        | X      | X      | `xlrd`, `openpyxl`, `odf`, `pyxlsb` (each engine supports different file formats) |
+| HDF5         | X      | X      |                                                                                   |
+| Feather      | X      | X      |                                                                                   |
+| Parquet      | X      | X      | `pyarrow`, `fastparquet`                                                          |
+| ORC          | X      | X      |                                                                                   |
+| Stata        | X      | X      |                                                                                   |
+| SAS          | X      |        |                                                                                   |
+| SPSS         | X      |        |                                                                                   |
+| Pickle       | X      | X      |                                                                                   |
+| SQL          | X      | X      | `sqlalchemy`, `dbapi2` (inferred from the type of the `con` parameter)            |
+| BigQuery     | X      | X      |                                                                                   |
+| dict         | X      | X      |                                                                                   |
+| records      | X      | X      |                                                                                   |
+| string       |        | X      |                                                                                   |
+| markdown     |        | X      |                                                                                   |
+| xarray       |        | X      |                                                                                   |
+
+At the time of writing this document, the `io/` module contains
+close to 100,000 lines of Python, C and Cython code.
+
+There is no objective criteria for when a format is included
+in pandas, and the list above is mostly the result of a developer
+being interested in implementing the connectors for a certain
+format in pandas.
+
+The number of existing formats available for data that can be processed with
+pandas is constantly increasing, and its difficult for pandas to keep up to
+date even with popular formats. It possibly makes sense to have connectors
+to PyArrow, PySpark, Iceberg, DuckDB, Hive, Polars, and many others.
+
+At the same time, some of the formats are not frequently used as shown in the
+[2019 user survey](https://pandas.pydata.org//community/blog/2019-user-survey.html).
+Those less popular formats include SPSS, SAS, Google BigQuery and
+Stata. Note that only I/O formats (and not memory formats like records or xarray)
+were included in the survey.
+
+The maintenance cost of supporting all formats is not only in maintaining the
+code and reviewing pull requests, but also it has a significant cost in time
+spent on CI systems installing dependencies, compiling code, running tests, etc.
+
+In some cases, the main maintainers of some of the connectors are not part of
+the pandas core development team, but people specialized in one of the formats.
+
+## Proposal
+
+While the current pandas approach has worked reasonably well, it is difficult
+to find a stable solution where the maintenance incurred in pandas is not
+too big, while at the same time users can interact with all different formats
+and representations they are interested in, in an easy and intuitive way.
+
+Third-party packages are already able to implement connectors to pandas, but
+there are some limitations to it:
+
+- Given the large number of formats supported by pandas itself, third-party
+  connectors are likely seen as second class citizens, not important enough
+  to be used, or not well supported.
+- There is no standard API for external I/O connectors, and users need
+  to learn each of them individually. Since the pandas I/O API is inconsistent
+  by using read/to instead of read/write or from/to, developers in many cases
+  ignore the convention. Also, even if developers follow the pandas convention
+  the namespaces would be different, since developers of connectors will rarely
+  monkeypatch their functions into the `pandas` or `DataFrame` namespace.
+- Method chaining is not possible with third-party I/O connectors to export
+  data, unless authors monkey patch the `DataFrame` class, which should not
+  be encouraged.
+
+This document proposes to open the development of pandas I/O connectors to
+third-party libraries in a standard way that overcomes those limitations.
+
+### Proposal implementation
+
+Implementing this proposal would not require major changes to pandas, and
+the API defined next would be used.
+
+#### User API
+
+Users will be able to install third-party packages implementing pandas
+connectors using the standard packaging tools (pip, conda, etc.). These
+connectors should implement entrypoints that pandas will use to
+automatically create the corresponding methods `pandas.read_*`,
+`pandas.DataFrame.to_*` and `pandas.Series.to_*`. Arbitrary function or
+method names will not be created by this interface, only the `read_*`
+and `to_*` pattern will be allowed.
+
+By simply installing the appropriate packages and calling the function
+`pandas.load_io_plugins()` users will be able to use code like this:
+
+```python
+import pandas
+
+pandas.load_io_plugins()
+
+df = pandas.read_duckdb("SELECT * FROM 'dataset.parquet';")
+
+df.to_hive(hive_conn, "hive_table")
+```
+
+This API allows for method chaining:
+
+```python
+(pandas.read_duckdb("SELECT * FROM 'dataset.parquet';")
+       .to_hive(hive_conn, "hive_table"))
+```
+
+The total number of I/O functions and methods is expected to be small, as users
+in general use only a small subset of formats. The number could actually be
+reduced from the current state if the less popular formats (such as SAS, SPSS,
+BigQuery, etc.) are removed from the pandas core into third-party packages.
+Moving these connectors is not part of this proposal, and could be discussed
+later in a separate proposal.
+
+#### Plugin registration
+
+Third-party packages would implement
+[entrypoints](https://setuptools.pypa.io/en/latest/userguide/entry_point.html#entry-points-for-plugins)
+to define the connectors that they implement, under a group `dataframe.io`.
+
+For example, a hypothetical project `pandas_duckdb` implementing a `read_duckdb`
+function, could use `pyproject.toml` to define the next entry point:
+
+```toml
+[project.entry-points."dataframe.io"]
+reader_duckdb = "pandas_duckdb:read_duckdb"
+```
+
+When the user calls `pandas.load_io_plugins()`, it would read the entrypoint registry for the
+`dataframe.io` group, and would dynamically create methods in the `pandas`,
+`pandas.DataFrame` and `pandas.Series` namespaces for them. Only entrypoints with
+name starting by `reader_` or `writer_` would be processed by pandas, and the functions
+registered in the entrypoint would be made available to pandas users in the corresponding
+pandas namespaces. The text after the keywords `reader_` and `writer_` would be used
+for the name of the function. In the example above, the entrypoint name `reader_duckdb`
+would create `pandas.read_duckdb`. An entrypoint with name `writer_hive` would create
+the methods `DataFrame.to_hive` and `Series.to_hive`.
+
+Entrypoints not starting with `reader_` or `writer_` would be ignored by this interface,
+but will not raise an exception since they can be used for future extensions of this
+API, or other related dataframe I/O interfaces.
+
+#### Internal API
+
+Connectors will use the dataframe interchange API to provide data to pandas. When
+data is read from a connector, and before returning it to the user as a response
+to `pandas.read_<format>`, data will be parsed from the data interchange interface
+and converted to a pandas DataFrame. In practice, connectors are likely to return
+a pandas DataFrame or a PyArrow Table, but the interface will support any object
+implementing the dataframe interchange API.
+
+#### Connector guidelines
+
+In order to provide a better and more consistent experience to users, guidelines
+will be created to unify terminology and behavior. Some of the topics to unify are
+defined next.
+
+**Guidelines to avoid name conflicts**. Since it is expected that more than one
+implementation exists for certain formats, as it already happens, guidelines on
+how to name connectors would be created. The easiest approach is probably to use
+as the format a string of the type `to_<format>_<implementation-id>` if it is
+expected that more than one connector can exist. For example, for LanceDB it is likely
+that only one connector exist, and the name `lance` can be used (which would create
+`pandas.read_lance` or `DataFrame.to_lance`. But if a new `csv` reader based in the
+Arrow2 Rust implementation, the guidelines can recommend to use `csv_arrow2` to
+create `pandas.read_csv_arrow2`, etc.
+
+**Existence and naming of parameters**, since many connectors are likely to provide
+similar features, like loading only a subset of columns in the data, or dealing
+with paths. Examples of recommendations to connector developers could be:
+
+- `columns`: Use this argument to let the user load a subset of columns. Allow a
+  list or tuple.
+- `path`: Use this argument if the dataset is a file in the file disk. Allow a string,
+  a `pathlib.Path` object, or a file descriptor. For a string object, allow URLs that
+  will be automatically download, compressed files that will be automatically
+  uncompressed, etc. Specific libraries can be recommended to deal with those in an
+  easier and more consistent way.
+- `schema`: For datasets that don't have a schema (e.g. `csv`), allow providing an
+  Apache Arrow schema instance, and automatically infer types if not provided.
+
+Note that the above are only examples of guidelines for illustration, and not
+a proposal of the guidelines, which would be developed independently after this
+PDEP is approved.
+
+**Connector registry and documentation**. To simplify the discovery of connectors
+and its documentation, connector developers can be encourage to register their
+projects in a central location, and to use a standard structure for documentation.
+This would allow the creation of a unified website to find the available
+connectors, and their documentation. It would also allow to customize the
+documentation for specific implementations, and include their final API.
+
+### Connector examples
+
+This section lists specific examples of connectors that could immediately
+benefit from this proposal.
+
+**PyArrow** currently provides `Table.from_pandas` and `Table.to_pandas`.
+With the new interface, it could also register `DataFrame.from_pyarrow`
+and `DataFrame.to_pyarrow`, so pandas users can use the converters with
+the interface they are used to, when PyArrow is installed in the environment.
+Better integration with PyArrow tables was discussed in
+[#51760](https://github.com/pandas-dev/pandas/issues/51760).
+
+_Current API_:
+
+```python
+pyarrow.Table.from_pandas(table.to_pandas()
+                               .query('my_col > 0'))
+```
+
+_Proposed API_:
+
+```python
+(pandas.read_pyarrow(table)
+       .query('my_col > 0')
+       .to_pyarrow())
+```
+
+**Polars**, **Vaex** and other dataframe frameworks could benefit from
+third-party projects that make the interoperability with pandas use a
+more explicitly API. Integration with Polars was requested in
+[#47368](https://github.com/pandas-dev/pandas/issues/47368).
+
+_Current API_:
+
+```python
+polars.DataFrame(df.to_pandas()
+                   .query('my_col > 0'))
+```
+
+_Proposed API_:
+
+```python
+(pandas.read_polars(df)
+       .query('my_col > 0')
+       .to_polars())
+```
+
+**DuckDB** provides an out-of-core engine able to push predicates before
+the data is loaded, making much better use of memory and significantly
+decreasing loading time. pandas, because of its eager nature is not able
+to easily implement this itself, but could benefit from a DuckDB loader.
+The loader can already be implemented inside pandas (it has already been
+proposed in [#45678](https://github.com/pandas-dev/pandas/issues/45678),
+or as a third-party extension with an arbitrary API. But this proposal would
+let the creation of a third-party extension with a standard and intuitive API:
+
+```python
+pandas.read_duckdb("SELECT *
+                    FROM 'dataset.parquet'
+                    WHERE my_col > 0")
+```
+
+**Out-of-core algorithms** push some operations like filtering or grouping
+to the loading of the data. While this is not currently possible, connectors
+implementing out-of-core algorithms could be developed using this interface.
+
+**Big data** systems such as Hive, Iceberg, Presto, etc. could benefit
+from a standard way to load data to pandas. Also regular **SQL databases**
+that can return their query results as Arrow, would benefit from better
+and faster connectors than the existing ones based on SQL Alchemy and
+Python structures.
+
+Any other format, including **domain-specific formats** could easily
+implement pandas connectors with a clear and intuitive API.
+
+### Limitations
+
+The implementation of this proposal has some limitations discussed here:
+
+- **Lack of support for multiple engines.** The current pandas I/O API
+  supports multiple engines for the same format (for the same function or
+  method name). For example `read_csv(engine='pyarrow', ...)`. Supporting
+  engines requires that all engines for a particular format use the same
+  signature (the same parameters), which is not ideal. Different connectors
+  are likely to have different parameters and using `*args` and `**kwargs`
+  provides users with a more complex and difficult experience. For this
+  reason this proposal prefers that function and method names are unique
+  instead of supporting an option for engines.
+- **Lack of support for type checking of connectors.** This PDEP proposes
+  creating functions and methods dynamically, and those are not supported
+  for type checking using stubs. This is already the case for other
+  dynamically created components of pandas, such as custom accessors.
+- **No improvements to the current I/O API**. In the discussions of this
+  proposal it has been considered to improve the current pandas I/O API to
+  fix the inconsistency of using `read` / `to` (instead of for example
+  `read` / `write`), avoid using `to_` prefixed methods for non-I/O
+  operations, or using a dedicated namespace (e.g. `DataFrame.io`) for
+  the connectors. All of these changes are out of scope for this PDEP.
+
+## Future plans
+
+This PDEP is exclusively to support a better API for existing of future
+connectors. It is out of scope for this PDEP to implement changes to any
+connectors existing in the pandas code base.
+
+Some ideas for future discussion related to this PDEP include:
+
+- Automatically loading of I/O plugins when pandas is imported.
+
+- Removing from the pandas code base some of the least frequently used connectors,
+such as SAS, SPSS or Google BigQuery, and move them to third-party connectors
+registered with this interface.
+
+- Discussing a better API for pandas connectors. For example, using `read_*`
+methods instead of `from_*` methods, renaming `to_*` methods not used as I/O
+connectors, using a consistent terminology like from/to, read/write, load/dump, etc.
+or using a dedicated namespace for connectors (e.g. `pandas.io` instead of the
+general `pandas` namespace).
+
+- Implement as I/O connectors some of the formats supported by the `DataFrame`
+constructor.
+
+## PDEP-9 History
+
+- 5 March 2023: Initial version
+- 30 May 2023: Major refactoring to use the pandas existing API,
+  the dataframe interchange API and to make the user be explicit to load
+  the plugins
+- 13 June 2023: The PDEP did not get any support after several iterations,
+  and its been closed as rejected by the author
diff --git a/pandas/web/pandas/pdeps/0010-required-pyarrow-dependency.md b/pandas/web/pandas/pdeps/0010-required-pyarrow-dependency.md
new file mode 100644
index 0000000000000..0c3bf3c776988
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0010-required-pyarrow-dependency.md
@@ -0,0 +1,214 @@
+# PDEP-10: PyArrow as a required dependency for default string inference implementation
+
+- Created: 17 April 2023
+- Status: Accepted
+- Discussion: [#52711](https://github.com/pandas-dev/pandas/pull/52711)
+              [#52509](https://github.com/pandas-dev/pandas/issues/52509)
+- Author: [Matthew Roeschke](https://github.com/mroeschke)
+          [Patrick Hoefler](https://github.com/phofl)
+- Revision: 1
+
+## Abstract
+
+This PDEP proposes that:
+
+- PyArrow becomes a required runtime dependency starting with pandas 3.0
+- The minimum version of PyArrow supported starting with pandas 3.0 is version 7 of PyArrow.
+- When the minimum version of PyArrow is bumped, PyArrow will be bumped to the highest version that has
+  been released for at least 2 years.
+- The pandas 2.1 release notes will have a big warning that PyArrow will become a required dependency starting
+  with pandas 3.0. We will pin a feedback issue on the pandas issue tracker. The note in the release notes will point
+  to that issue.
+- Starting in pandas 2.2, pandas raises a ``FutureWarning`` when PyArrow is not installed in the users
+  environment when pandas is imported. This will ensure that only one warning is raised and users can
+  easily silence it if necessary. This warning will point to the feedback issue.
+- Starting in pandas 3.0, the default type inferred for string data will be `ArrowDtype` with `pyarrow.string`
+  instead of `object`. Additionally, we will infer all dtypes that are listed below as well instead of storing as object.
+
+This will bring **immediate benefits to users**, as well as opening up the door for significant further
+benefits in the future.
+
+## Background
+
+PyArrow is an optional dependency of pandas that provides a wide range of supplemental features to pandas:
+
+- Since pandas 0.21.0, PyArrow provided I/O reading functionality for Parquet
+- Since pandas 1.2.0, pandas integrated PyArrow into the `ExtensionArray` interface to provide an
+  optional string data type backed by PyArrow
+- Since pandas 1.4.0, PyArrow provided I/0 reading functionality for CSV
+- Since pandas 1.5.0, pandas provided an `ArrowExtensionArray` and `ArrowDtype` to support all PyArrow
+  data types within the `ExtensionArray` interface
+- Since pandas 2.0.0, all I/O readers have the option to return PyArrow-backed data types, and many methods
+  now utilize PyArrow compute functions to
+accelerate PyArrow-backed data in pandas, notably string and datetime types.
+
+As of pandas 2.0, one can feasibly utilize PyArrow as an alternative data representation to NumPy with advantages such as:
+
+1. Consistent `NA` support for all data types;
+2. Broader support of data types such as `decimal`, `date` and nested types;
+3. Better interoperability with other dataframe libraries based on Arrow.
+
+## Motivation
+
+While all the functionality described in the previous paragraph is currently optional, PyArrow has significant
+integration into many areas of pandas. With our roadmap noting that pandas strives for better Apache Arrow
+interoperability [^1] and many projects [^2], within or beyond the Python ecosystem, adopting or interacting with
+the Arrow format, making PyArrow a required dependency provides an additional signal of confidence in the Arrow
+ecosystem (as well as improving interoperability with it).
+
+### Immediate User Benefit 1: pyarrow strings
+
+Currently, when users pass string data into pandas constructors without specifying a data type, the resulting data type
+is `object`, which has significantly much worse memory usage and performance as compared to pyarrow strings.
+With pyarrow string support available since 1.2.0, requiring pyarrow for 3.0 will allow pandas to default
+the inferred type to the more efficient pyarrow string type.
+
+```python
+In [1]: import pandas as pd
+
+In [2]: pd.Series(["a"]).dtype
+# Current behavior
+Out[2]: dtype('O')
+
+# Future behavior in 3.0
+Out[2]: string[pyarrow]
+```
+
+Dask developers investigated performance and memory of pyarrow strings [here](https://www.coiled.io/blog/pyarrow-strings-in-dask-dataframes),
+and found them to be a significant improvement over the current `object` dtype.
+
+Little demo:
+```python
+import string
+import random
+
+import pandas as pd
+
+
+def random_string() -> str:
+    return "".join(random.choices(string.printable, k=random.randint(10, 100)))
+
+
+ser_object = pd.Series([random_string() for _ in range(1_000_000)])
+ser_string = ser_object.astype("string[pyarrow]")\
+```
+
+PyArrow backed strings are significantly faster than NumPy object strings:
+
+*str.len*
+
+```python
+In[1]: %timeit ser_object.str.len()
+118 ms ± 260 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+In[2]: %timeit ser_string.str.len()
+24.2 ms ± 187 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+```
+
+*str.startswith*
+
+```python
+In[3]: %timeit ser_object.str.startswith("a")
+136 ms ± 300 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
+
+In[4]: %timeit ser_string.str.startswith("a")
+11 ms ± 19.8 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
+```
+
+### Immediate User Benefit 2: Nested Datatypes
+
+Currently, if you try storing `dict`s in a pandas `Series`, you will again get the horrendous `object` dtype:
+```python
+In [6]: pd.Series([{'a': 1, 'b': 2}, {'a': 2, 'b': 99}])
+Out[6]:
+0     {'a': 1, 'b': 2}
+1    {'a': 2, 'b': 99}
+dtype: object
+```
+
+If `pyarrow` were required, this could have been auto-inferred to be `pyarrow.struct`, which again
+would come with memory and performance improvements.
+
+### Immediate User Benefit 3: Interoperability
+
+Other Arrow-backed dataframe libraries are growing in popularity. Having the same memory representation
+would improve interoperability with them, as operations such as:
+```python
+import pandas as pd
+import polars as pl
+
+df = pd.DataFrame(
+  {
+    'a': ['one', 'two'],
+    'b': [{'name': 'Billy', 'age': 3}, {'name': 'Bob', 'age': 4}],
+  }
+)
+pl.from_pandas(df)
+```
+could be zero-copy. Users making use of multiple dataframe libraries would more easily be able to
+switch between them.
+
+### Future User Benefits:
+
+Requiring PyArrow would simplify the related development within pandas and potentially improve NumPy
+functionality that would be better suited by PyArrow including:
+
+- Avoiding runtime checking if PyArrow is available to perform PyArrow object inference during constructor or indexing operations
+
+- NumPy object dtype will be avoided as much as possible. This means that every dtype that has a PyArrow equivalent is inferred automatically as such. This includes:
+  - decimal
+  - binary
+  - nested types (list or dict data)
+  - strings
+  - time
+  - date
+
+#### Developer benefits
+
+First, this would simplify development of pyarrow-backed datatypes, as it would avoid
+optional dependency checks.
+
+Second, it could potentially remove redundant functionality:
+- fastparquet engine in `read_parquet`;
+- potentially simplifying the `read_csv` logic (needs more investigation);
+- factorization;
+- datetime/timezone ops.
+
+## Drawbacks
+
+Including PyArrow would naturally increase the installation size of pandas. For example, installing pandas and PyArrow
+using pip from wheels, numpy and pandas requires about `70MB`, and including PyArrow requires an additional `120MB`.
+An increase of installation size would have negative implication using pandas in space-constrained development or deployment environments
+such as AWS Lambda.
+
+Additionally, if a user is installing pandas in an environment where wheels are not available through a `pip install` or `conda install`,
+the user will need to also build Arrow C++ and related dependencies when installing from source. These environments include
+
+- Alpine linux (commonly used as a base for Docker containers)
+- Python development versions
+
+Lastly, pandas development and releases will need to be mindful of PyArrow's development and release cadance. For example when
+supporting a newly released Python version, pandas will also need to be mindful of PyArrow's wheel support for that Python version
+before releasing a new pandas version.
+
+## F.A.Q.
+
+**Q: Why can't pandas just use numpy string and numpy void datatypes instead of pyarrow string and pyarrow struct?**
+
+**A**: NumPy strings aren't yet available, whereas pyarrow strings are. NumPy void datatype would be different to pyarrow struct,
+  not bringing the same interoperabitlity benefit with other arrow-based dataframe libraries.
+
+**Q: Are all pyarrow dtypes ready? Isn't it too soon to make them the default?**
+
+**A**: They will likely be ready by 3.0 - however, we're not making them the default (yet).
+  For example, `pd.Series([1, 2, 3])` will continue to be auto-inferred to be
+  `np.int64`.  We will only change the default for dtypes which currently have no `numpy`-backed equivalent and which are
+  stored as `object` dtype, such as strings and nested datatypes.
+
+### PDEP-10 History
+
+- 17 April 2023: Initial version
+- 8 May 2023: Changed proposal to make pyarrow required in pandas 3.0 instead of 2.1
+
+[^1] <https://pandas.pydata.org/docs/development/roadmap.html#apache-arrow-interoperability>
+[^2] <https://arrow.apache.org/powered_by/>
diff --git a/pandas/web/pandas/pdeps/0012-compact-and-reversible-JSON-interface.md b/pandas/web/pandas/pdeps/0012-compact-and-reversible-JSON-interface.md
new file mode 100644
index 0000000000000..71f669825f979
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0012-compact-and-reversible-JSON-interface.md
@@ -0,0 +1,479 @@
+# PDEP-12: Compact and reversible JSON interface
+
+- Created: 16 June 2023
+- Status: Rejected
+- Discussion:
+    [#53252](https://github.com/pandas-dev/pandas/issues/53252)
+    [#55038](https://github.com/pandas-dev/pandas/issues/55038)
+- Author: [Philippe THOMY](https://github.com/loco-philippe)
+- Revision: 3
+
+##### Summary
+
+- [Abstract](./0012-compact-and-reversible-JSON-interface.md/#Abstract)
+  - [Problem description](./0012-compact-and-reversible-JSON-interface.md/#Problem-description)
+  - [Feature Description](./0012-compact-and-reversible-JSON-interface.md/#Feature-Description)
+- [Scope](./0012-compact-and-reversible-JSON-interface.md/#Scope)
+- [Motivation](./0012-compact-and-reversible-JSON-interface.md/#Motivation)
+  - [Why is it important to have a compact and reversible JSON interface ?](./0012-compact-and-reversible-JSON-interface.md/#Why-is-it-important-to-have-a-compact-and-reversible-JSON-interface-?)
+  - [Is it relevant to take an extended type into account ?](./0012-compact-and-reversible-JSON-interface.md/#Is-it-relevant-to-take-an-extended-type-into-account-?)
+  - [Is this only useful for pandas ?](./0012-compact-and-reversible-JSON-interface.md/#Is-this-only-useful-for-pandas-?)
+- [Description](./0012-compact-and-reversible-JSON-interface.md/#Description)
+  - [Data typing](./0012-compact-and-reversible-JSON-interface.md/#Data-typing)
+  - [Correspondence between TableSchema and pandas](./panda0012-compact-and-reversible-JSON-interfaces_PDEP.md/#Correspondence-between-TableSchema-and-pandas)
+  - [JSON format](./0012-compact-and-reversible-JSON-interface.md/#JSON-format)
+  - [Conversion](./0012-compact-and-reversible-JSON-interface.md/#Conversion)
+- [Usage and impact](./0012-compact-and-reversible-JSON-interface.md/#Usage-and-impact)
+  - [Usage](./0012-compact-and-reversible-JSON-interface.md/#Usage)
+  - [Compatibility](./0012-compact-and-reversible-JSON-interface.md/#Compatibility)
+  - [Impacts on the pandas framework](./0012-compact-and-reversible-JSON-interface.md/#Impacts-on-the-pandas-framework)
+  - [Risk to do / risk not to do](./0012-compact-and-reversible-JSON-interface.md/#Risk-to-do-/-risk-not-to-do)
+- [Implementation](./0012-compact-and-reversible-JSON-interface.md/#Implementation)
+  - [Modules](./0012-compact-and-reversible-JSON-interface.md/#Modules)
+  - [Implementation options](./0012-compact-and-reversible-JSON-interface.md/#Implementation-options)
+- [F.A.Q.](./0012-compact-and-reversible-JSON-interface.md/#F.A.Q.)
+- [Synthesis](./0012-compact-and-reversible-JSON-interface.md/Synthesis)
+- [Core team decision](./0012-compact-and-reversible-JSON-interface.md/#Core-team-decision)
+- [Timeline](./0012-compact-and-reversible-JSON-interface.md/#Timeline)
+- [PDEP history](./0012-compact-and-reversible-JSON-interface.md/#PDEP-history)
+
+-------------------------
+
+## Abstract
+
+### Problem description
+
+The `dtype` and "Python type" are not explicitly taken into account in the current JSON interface.
+
+So, the JSON interface is not always reversible and has inconsistencies related to the consideration of the `dtype`.
+
+Another consequence is the partial application of the Table Schema specification in the `orient="table"` option (6 Table Schema data types are taken into account out of the 24 defined).
+
+Some JSON-interface problems are detailed in the [linked NoteBook](https://nbviewer.org/github/loco-philippe/ntv-pandas/blob/main/example/example_json_pandas.ipynb#Current-Json-interface)
+
+### Feature Description
+
+To have a simple, compact and reversible solution, I propose to use the [JSON-NTV format (Named and Typed Value)](https://github.com/loco-philippe/NTV#readme) - which integrates the notion of type - and its JSON-TAB variation for tabular data (the JSON-NTV format is defined in an [IETF Internet-Draft](https://datatracker.ietf.org/doc/draft-thomy-json-ntv/) (not yet an RFC !!) ).
+
+This solution allows to include a large number of types (not necessarily pandas `dtype`) which allows to have:
+
+- a Table Schema JSON interface (`orient="table"`) which respects the Table Schema specification (going from 6 types to 20 types),
+- a global JSON interface for all pandas data formats.
+
+#### Global JSON interface example
+
+In the example below, a DataFrame with several data types is converted to JSON.
+
+The DataFrame resulting from this JSON is identical to the initial DataFrame (reversibility).
+
+With the existing JSON interface, this conversion is not possible.
+
+This example uses `ntv_pandas` module defined in the [ntv-pandas repository](https://github.com/loco-philippe/ntv-pandas#readme).
+
+Data example:
+
+```python
+In [1]: from shapely.geometry import Point
+        from datetime import date
+        import pandas as pd
+        import ntv_pandas as npd
+
+In [2]: data = {'index':           [100, 200, 300, 400, 500, 600],
+                'dates::date':     [date(1964,1,1), date(1985,2,5), date(2022,1,21), date(1964,1,1), date(1985,2,5), date(2022,1,21)],
+                'value':           [10, 10, 20, 20, 30, 30],
+                'value32':         pd.Series([12, 12, 22, 22, 32, 32], dtype='int32'),
+                'res':             [10, 20, 30, 10, 20, 30],
+                'coord::point':    [Point(1,2), Point(3,4), Point(5,6), Point(7,8), Point(3,4), Point(5,6)],
+                'names':           pd.Series(['john', 'eric', 'judith', 'mila', 'hector', 'maria'], dtype='string'),
+                'unique':          True }
+
+In [3]: df = pd.DataFrame(data).set_index('index')
+
+In [4]: df
+Out[4]:       dates::date  value  value32  res coord::point   names  unique
+        index
+        100    1964-01-01     10       12   10  POINT (1 2)    john    True
+        200    1985-02-05     10       12   20  POINT (3 4)    eric    True
+        300    2022-01-21     20       22   30  POINT (5 6)  judith    True
+        400    1964-01-01     20       22   10  POINT (7 8)    mila    True
+        500    1985-02-05     30       32   20  POINT (3 4)  hector    True
+        600    2022-01-21     30       32   30  POINT (5 6)   maria    True
+```
+
+JSON representation
+
+```python
+In [5]: df_to_json = npd.to_json(df)
+        pprint(df_to_json, width=120)
+Out[5]: {':tab': {'coord::point': [[1.0, 2.0], [3.0, 4.0], [5.0, 6.0], [7.0, 8.0], [3.0, 4.0], [5.0, 6.0]],
+                  'dates::date': ['1964-01-01', '1985-02-05', '2022-01-21', '1964-01-01', '1985-02-05', '2022-01-21'],
+                  'index': [100, 200, 300, 400, 500, 600],
+                  'names::string': ['john', 'eric', 'judith', 'mila', 'hector', 'maria'],
+                  'res': [10, 20, 30, 10, 20, 30],
+                  'unique': [True, True, True, True, True, True],
+                  'value': [10, 10, 20, 20, 30, 30],
+                  'value32::int32': [12, 12, 22, 22, 32, 32]}}
+```
+
+Reversibility
+
+```python
+In [5]: df_from_json = npd.read_json(df_to_json)
+        print('df created from JSON is equal to initial df ? ', df_from_json.equals(df))
+Out[5]: df created from JSON is equal to initial df ?  True
+```
+
+Several other examples are provided in the [linked NoteBook](https://nbviewer.org/github/loco-philippe/ntv-pandas/blob/main/example/example_ntv_pandas.ipynb)
+
+#### Table Schema JSON interface example
+
+In the example below, a DataFrame with several Table Schema data types is converted to JSON.
+
+The DataFrame resulting from this JSON is identical to the initial DataFrame (reversibility).
+
+With the existing Table Schema JSON interface, this conversion is not possible.
+
+```python
+In [1]: from shapely.geometry import Point
+        from datetime import date
+
+In [2]: df = pd.DataFrame({
+            'end february::date': ['date(2023,2,28)', 'date(2024,2,29)', 'date(2025,2,28)'],
+            'coordinates::point': ['Point([2.3, 48.9])', 'Point([5.4, 43.3])', 'Point([4.9, 45.8])'],
+            'contact::email':     ['john.doe@table.com', 'lisa.minelli@schema.com', 'walter.white@breaking.com']
+            })
+
+In [3]: df
+Out[3]: end february::date coordinates::point             contact::email
+        0         2023-02-28   POINT (2.3 48.9)         john.doe@table.com
+        1         2024-02-29   POINT (5.4 43.3)    lisa.minelli@schema.com
+        2         2025-02-28   POINT (4.9 45.8)  walter.white@breaking.com
+```
+
+JSON representation
+
+```python
+In [4]: df_to_table = npd.to_json(df, table=True)
+        pprint(df_to_table, width=140, sort_dicts=False)
+Out[4]: {'schema': {'fields': [{'name': 'index', 'type': 'integer'},
+                               {'name': 'end february', 'type': 'date'},
+                               {'name': 'coordinates', 'type': 'geopoint', 'format': 'array'},
+                               {'name': 'contact', 'type': 'string', 'format': 'email'}],
+                    'primaryKey': ['index'],
+                    'pandas_version': '1.4.0'},
+         'data': [{'index': 0, 'end february': '2023-02-28', 'coordinates': [2.3, 48.9], 'contact': 'john.doe@table.com'},
+                  {'index': 1, 'end february': '2024-02-29', 'coordinates': [5.4, 43.3], 'contact': 'lisa.minelli@schema.com'},
+                  {'index': 2, 'end february': '2025-02-28', 'coordinates': [4.9, 45.8], 'contact': 'walter.white@breaking.com'}]}
+```
+
+Reversibility
+
+```python
+In [5]: df_from_table = npd.read_json(df_to_table)
+        print('df created from JSON is equal to initial df ? ', df_from_table.equals(df))
+Out[5]: df created from JSON is equal to initial df ?  True
+```
+
+Several other examples are provided in the [linked NoteBook](https://nbviewer.org/github/loco-philippe/ntv-pandas/blob/main/example/example_table_pandas.ipynb)
+
+## Scope
+
+The objective is to make available the proposed JSON interface for any type of data and for `orient="table"` option or a new option `orient="ntv"`.
+
+The proposed interface is compatible with existing data.
+
+## Motivation
+
+### Why extend the `orient=table` option to other data types?
+
+- The Table Schema specification defines 24 data types, 6 are taken into account in the pandas interface
+
+### Why is it important to have a compact and reversible JSON interface ?
+
+- a reversible interface provides an exchange format.
+- a textual exchange format facilitates exchanges between platforms (e.g. OpenData)
+- a JSON exchange format can be used at API level
+
+### Is it relevant to take an extended type into account ?
+
+- it avoids the addition of an additional data schema
+- it increases the semantic scope of the data processed by pandas
+- it is an answer to several issues (e.g.  #12997, #14358, #16492, #35420, #35464, #36211, #39537, #49585, #50782, #51375, #52595, #53252)
+- the use of a complementary type avoids having to modify the pandas data model
+
+### Is this only useful for pandas ?
+
+- the JSON-TAB format is applicable to tabular data and multi-dimensional data.
+- this JSON interface can therefore be used for any application using tabular or multi-dimensional data. This would allow for example reversible data exchanges between pandas - DataFrame and Xarray - DataArray (Xarray issue under construction) [see example DataFrame / DataArray](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb#Multidimensional-data).
+
+## Description
+
+The proposed solution is based on several key points:
+
+- data typing
+- correspondence between TableSchema and pandas
+- JSON format for tabular data
+- conversion to and from JSON format
+
+### Data typing
+
+Data types are defined and managed in the NTV project (name, JSON encoder and decoder).
+
+Pandas `dtype` are compatible with NTV types :
+
+| **pandas  dtype**  | **NTV type**   |
+|--------------------|------------|
+| intxx              | intxx      |
+| uintxx             | uintxx     |
+| floatxx            | floatxx    |
+| datetime[ns]       | datetime   |
+| datetime[ns, <tz>] | datetimetz |
+| timedelta[ns]      | durationiso|
+| string             | string     |
+| boolean            | boolean    |
+
+Note:
+
+- datetime with timezone is a single NTV type (string ISO8601)
+- `CategoricalDtype` and `SparseDtype` are included in the tabular JSON format
+- `object` `dtype` is depending on the context (see below)
+- `PeriodDtype` and `IntervalDtype` are to be defined
+
+JSON types (implicit or explicit) are converted in `dtype` following pandas JSON interface:
+
+| **JSON type**  | **pandas  dtype** |
+|----------------|-------------------|
+| number         | int64 / float64   |
+| string         | string / object   |
+| array          | object            |
+| object         | object            |
+| true, false    | boolean           |
+| null           | NaT / NaN / None  |
+
+Note:
+
+- if an NTV type is defined, the `dtype` is adjusted accordingly
+- the consideration of null type data needs to be clarified
+
+The other NTV types are associated with `object` `dtype`.
+
+### Correspondence between TableSchema and pandas
+
+The TableSchema typing is carried by two attributes `format` and `type`.
+
+The table below shows the correspondence between TableSchema format / type and pandas NTVtype / dtype:
+
+| **format / type**  | **NTV type / dtype** |
+|--------------------|----------------------|
+| default / datetime |  / datetime64[ns]    |
+| default / number   |  / float64           |
+| default / integer  |  / int64             |
+| default / boolean  |  / bool              |
+| default / string   |  / object            |
+| default / duration |  / timedelta64[ns]   |
+| email   / string   | email   / string     |
+| uri     / string   | uri     / string     |
+| default / object   | object  / object     |
+| default / array    | array   / object     |
+| default / date     | date    / object     |
+| default / time     | time    / object     |
+| default / year     | year    / int64      |
+| default / yearmonth| month   / int64      |
+| array   / geopoint | point   / object     |
+| default / geojson  | geojson / object     |
+
+Note:
+
+- other TableSchema format are defined and are to be studied (uuid, binary, topojson, specific format for geopoint and datation)
+- the first six lines correspond to the existing
+
+### JSON format
+
+The JSON format for the TableSchema interface is the existing.
+
+The JSON format for the Global interface is defined in [JSON-TAB](https://github.com/loco-philippe/NTV/blob/v1.1.0/documentation/JSON-TAB-standard.pdf) specification.
+It includes the naming rules originally defined in the [JSON-ND project](https://github.com/glenkleidon/JSON-ND) and support for categorical data.
+The specification have to be updated to include sparse data.
+
+### Conversion
+
+When data is associated with a non-`object` `dtype`, pandas conversion methods are used.
+Otherwise, NTV conversion is used.
+
+#### pandas -> JSON
+
+- `NTV type` is not defined : use `to_json()`
+- `NTV type` is defined and `dtype` is not `object` : use `to_json()`
+- `NTV type` is defined and `dtype` is `object` : use NTV conversion (if pandas conversion does not exist)
+
+#### JSON -> pandas
+
+- `NTV type` is compatible with a `dtype` : use `read_json()`
+- `NTV type` is not compatible with a `dtype` : use NTV conversion (if pandas conversion does not exist)
+
+## Usage and Impact
+
+### Usage
+
+It seems to me that this proposal responds to important issues:
+
+- having an efficient text format for data exchange
+
+    The alternative CSV format is not reversible and obsolete (last revision in 2005). Current CSV tools do not comply with the standard.
+
+- taking into account "semantic" data in pandas objects
+
+- having a complete Table Schema interface
+
+### Compatibility
+
+Interface can be used without NTV type (compatibility with existing data - [see examples](https://nbviewer.org/github/loco-philippe/ntv-pandas/blob/main/example/example_ntv_pandas.ipynb#Appendix-:-Series-tests))
+
+If the interface is available, throw a new `orient` option in the JSON interface, the use of the feature is decoupled from the other features.
+
+### Impacts on the pandas framework
+
+Initially, the impacts are very limited:
+
+- modification of the `name` of `Series` or `DataFrame columns` (no functional impact),
+- added an option in the Json interface (e.g. `orient='ntv'`) and added associated methods (no functional interference with the other methods)
+
+In later stages, several developments could be considered:
+
+- validation of the `name` of `Series` or `DataFrame columns` ,
+- management of the NTV type as a "complementary-object-dtype"
+- functional extensions depending on the NTV type
+
+### Risk to do / risk not to do
+
+The JSON-NTV format and the JSON-TAB format are not (yet) recognized and used formats. The risk for pandas is that this function is not used (no functional impacts).
+
+On the other hand, the early use by pandas will allow a better consideration of the expectations and needs of pandas as well as a reflection on the evolution of the types supported by pandas.
+
+## Implementation
+
+### Modules
+
+Two modules are defined for NTV:
+
+- json-ntv
+
+    this module manages NTV data without dependency to another module
+
+- ntvconnector
+
+    those modules manage the conversion between objects and JSON data. They have dependency with objects modules (e.g. connectors with shapely location have dependency with shapely).
+
+The pandas integration of the JSON interface requires importing only the json-ntv module.
+
+### Implementation options
+
+The interface can be implemented as NTV connector (`SeriesConnector` and `DataFrameConnector`) and as a new pandas JSON interface `orient` option.
+
+Several pandas implementations are possible:
+
+1. External:
+
+    In this implementation, the interface is available only in the NTV side.
+    This option means that this evolution of the JSON interface is not useful or strategic for pandas.
+
+2. NTV side:
+
+    In this implementation, the interface is available in the both sides and the conversion is located inside NTV.
+    This option is the one that minimizes the impacts on the pandas side
+
+3. pandas side:
+
+    In this implementation, the interface is available in the both sides and the conversion is located inside pandas.
+    This option allows pandas to keep control of this evolution
+
+4. pandas restricted:
+
+    In this implementation, the pandas interface and the conversion are located inside pandas and only for non-object `dtype`.
+    This option makes it possible to offer a compact and reversible interface while prohibiting the introduction of types incompatible with the existing `dtype`
+
+## F.A.Q.
+
+**Q: Does `orient="table"` not do what you are proposing already?**
+
+**A**: In principle, yes, this option takes into account the notion of type.
+
+But this is very limited (see examples added in the [Notebook](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb)) :
+
+- **Types and Json interface**
+  - the only way to keep the types in the json interface is to use the `orient='table'` option
+  - few dtypes are not allowed in json-table interface : period, timedelta64, interval
+  - allowed types are not always kept in json-table interface
+  - data with 'object' dtype is kept only id data is string
+  - with categorical dtype, the underlying dtype is not included in json interface
+- **Data compactness**
+  - json-table interface is not compact (in the example in the [Notebook](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb#data-compactness))the size is triple or quadruple the size of the compact format
+- **Reversibility**
+  - Interface is reversible only with few dtypes : int64, float64, bool, string, datetime64 and partially categorical
+- **External types**
+  - the interface does not accept external types
+  - Table-schema defines 20 data types but the `orient="table"` interface takes into account 5 data types (see [table](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb#Converting-table-schema-type-to-pandas-dtype))
+  - to integrate external types, it is necessary to first create ExtensionArray and ExtensionDtype objects
+
+The current interface is not compatible with the data structure defined by table-schema. For this to be possible, it is necessary to integrate a "type extension" like the one proposed (this has moreover been partially achieved with the notion of `extDtype` found in the interface for several formats).
+
+**Q: In general, we should only have 1 `"table"` format for pandas in read_json/to_json. There is also the issue of backwards compatibility if we do change the format. The fact that the table interface is buggy is not a reason to add a new interface (I'd rather fix those bugs). Can the existing format be adapted in a way that fixes the type issues/issues with roundtripping?**
+
+**A**: I will add two additional remarks:
+
+- the types defined in Tableschema are partially taken into account (examples of types not taken into account in the interface: string-uri, array, date, time, year, geopoint, string-email):
+- the `read_json()` interface works too with the following data: `{'simple': [1,2,3] }` (contrary to what is indicated in the documentation) but it is impossible with `to_json()` to recreate this simple json.
+
+I think that the problem cannot be limited to bug fixes and that a clear strategy must be defined for the Json interface in particular with the gradual abandonment in open-data solutions of the obsolete CSV format in favor of a Json format.
+
+As stated, the proposed solution addresses several shortcomings of the current interface and could simply fit into the pandas environment (the other option would be to consider that the Json interface is a peripheral function of pandas and can remain external to pandas) regardless of the `orient='table'` option.
+
+It is nevertheless possible to merge the proposed format and the `orient='table'` format in order to have an explicit management of the notion of `extDtype`
+
+**Q: As far as I can tell, JSON NTV is not in any form a standardised JSON format. I believe that pandas (and geopandas, which is where I came from to this issue) should try to follow either de facto or de jure standards and do not opt in for a file format that does not have any community support at this moment. This can obviously change in the future and that is where this PR should be revised. Why would pandas use this standard?**
+
+**A**: As indicated in the issue (and detailed in [the attached Notebook](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb)), the json interface is not reversible (`to_json` then `read_json` does not always return the initial object) and several shortcomings and bugs are present. The main cause of this problem is that the data type is not taken into account in the JSON format (or very partially with the `orient='table'` option).
+
+The proposal made answers this problem ([the example at the beginning of Notebook](https://nbviewer.org/github/loco-philippe/NTV/blob/main/example/example_pandas.ipynb#0---Simple-example) simply and clearly illustrates the interest of the proposal).
+
+Regarding the underlying JSON-NTV format, its impact is quite low for tabular data (it is limited to adding the type in the field name).
+Nevertheless, the question is relevant: The JSON-NTV format ([IETF Internet-Draft](https://datatracker.ietf.org/doc/draft-thomy-json-ntv/)) is a shared, documented, supported and implemented format, but indeed the community support is for the moment reduced but it only asks to expand !!
+
+## Synthesis
+
+To conclude,
+
+- if it is important (or strategic) to have a reversible JSON interface for any type of data, the proposal can be allowed,
+- if not, a third-party package listed in the [ecosystem](https://pandas.pydata.org/community/ecosystem.html) that reads/writes this format to/from pandas DataFrames should be considered
+
+## Core team decision
+
+Vote was open from september-11 to september-26:
+
+- Final tally is 0 approvals, 5 abstentions, 7 disapprove. The quorum has been met. The PDEP fails.
+
+**Disapprove comments** :
+
+- 1 Given the newness of the proposed JSON NTV format, I would support (as described in the PDEP): "if not, a third-party package listed in the ecosystem that reads/writes this format to/from pandas DataFrames should be considered"
+- 2 Same reason as -1-, this should be a third party package for now
+- 3 Not mature enough, and not clear what the market size would be.
+- 4 for the same reason I left in the PDEP: "I think this (JSON-NTV format) does not meet the bar of being a commonly used format for implementation within pandas"
+- 5 agree with -4-
+- 6 agree with the other core-dev responders. I think work in the existing json interface is extremely valuable. A number of the original issues raised are just bug fixes / extensions of already existing functionality. Trying to start anew is likely not worth the migration effort. That said if a format is well supported in the community we can reconsider in the future (obviously json is well supported but the actual specification detailed here is too new / not accepted as a standard)
+- 7 while I do think having a more comprehensive JSON format would be worthwhile, making a new format part of pandas means an implicit endorsement of a standard that is still being reviewed by the broader community.
+
+**Decision**:
+
+- add the `ntv-pandas` package in the [ecosystem](https://pandas.pydata.org/community/ecosystem.html)
+- revisit again this PDEP at a later stage, for example in 1/2 to 1 year (based on the evolution of the Internet draft [JSON semantic format (JSON-NTV)](https://www.ietf.org/archive/id/draft-thomy-json-ntv-01.html) and the usage of the [ntv-pandas](https://github.com/loco-philippe/ntv-pandas#readme))
+
+## Timeline
+
+Not applicable
+
+## PDEP History
+
+- 16 June 2023: Initial draft
+- 22 July 2023: Add F.A.Q.
+- 06 September 2023: Add Table Schema extension
+- 01 Octobre: Add Core team decision
diff --git a/pandas/web/pandas/pdeps/0014-string-dtype.md b/pandas/web/pandas/pdeps/0014-string-dtype.md
new file mode 100644
index 0000000000000..5b74f71216454
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0014-string-dtype.md
@@ -0,0 +1,375 @@
+# PDEP-14: Dedicated string data type for pandas 3.0
+
+- Created: May 3, 2024
+- Status: Accepted
+- Discussion: https://github.com/pandas-dev/pandas/pull/58551
+- Author: [Joris Van den Bossche](https://github.com/jorisvandenbossche)
+- Revision: 1
+
+## Abstract
+
+This PDEP proposes to introduce a dedicated string dtype that will be used by
+default in pandas 3.0:
+
+* In pandas 3.0, enable a string dtype (`"str"`) by default, using PyArrow if available
+  or otherwise a string dtype using numpy object-dtype under the hood as fallback.
+* The default string dtype will use missing value semantics (using NaN) consistent
+  with the other default data types.
+
+This will give users a long-awaited proper string dtype for 3.0, while 1) not
+(yet) making PyArrow a _hard_ dependency, but only a dependency used by default,
+and 2) leaving room for future improvements (different missing value semantics,
+using NumPy 2.0 strings, etc).
+
+## Background
+
+Currently, pandas by default stores text data in an `object`-dtype NumPy array.
+The current implementation has two primary drawbacks. First, `object` dtype is
+not specific to strings: any Python object can be stored in an `object`-dtype
+array, not just strings, and seeing `object` as the dtype for a column with
+strings is confusing for users. Second: this is not efficient (all string
+methods on a Series are eventually calling Python methods on the individual
+string objects).
+
+To solve the first issue, a dedicated extension dtype for string data has
+already been
+[added in pandas 1.0](https://pandas.pydata.org/docs/whatsnew/v1.0.0.html#dedicated-string-data-type).
+This has always been opt-in for now, requiring users to explicitly request the
+dtype (with `dtype="string"` or `dtype=pd.StringDtype()`). The array backing
+this string dtype was initially almost the same as the default implementation,
+i.e. an `object`-dtype NumPy array of Python strings.
+
+To solve the second issue (performance), pandas contributed to the development
+of string kernels in the PyArrow package, and a variant of the string dtype
+backed by PyArrow was
+[added in pandas 1.3](https://pandas.pydata.org/docs/whatsnew/v1.3.0.html#pyarrow-backed-string-data-type).
+This could be specified with the `storage` keyword in the opt-in string dtype
+(`pd.StringDtype(storage="pyarrow")`).
+
+Since its introduction, the `StringDtype` has always been opt-in, and has used
+the experimental `pd.NA` sentinel for missing values (which was also [introduced
+in pandas 1.0](https://pandas.pydata.org/docs/whatsnew/v1.0.0.html#experimental-na-scalar-to-denote-missing-values)).
+However, up to this date, pandas has not yet taken the step to use `pd.NA` for
+for any default dtype, and thus the `StringDtype` deviates in missing value
+behaviour compared to the default data types.
+
+In 2023, [PDEP-10](https://pandas.pydata.org/pdeps/0010-required-pyarrow-dependency.html)
+proposed to start using a PyArrow-backed string dtype by default in pandas 3.0
+(i.e. infer this type for string data instead of object dtype). To ensure we
+could use the variant of `StringDtype` backed by PyArrow instead of Python
+objects (for better performance), it proposed to make `pyarrow` a new required
+runtime dependency of pandas.
+
+In the meantime, NumPy has also been working on a native variable-width string
+data type, which was made available [starting with NumPy
+2.0](https://numpy.org/devdocs/release/2.0.0-notes.html#stringdtype-has-been-added-to-numpy).
+This can provide a potential alternative to PyArrow for implementing a string
+data type in pandas that is not backed by Python objects.
+
+After acceptance of PDEP-10, two aspects of the proposal have been under
+reconsideration:
+
+- Based on feedback from users and maintainers from other packages (mostly
+  around installation complexity and size), it has been considered to relax the
+  new `pyarrow` requirement to not be a _hard_ runtime dependency. In addition,
+  NumPy 2.0 could in the future potentially reduce the need to make PyArrow a
+  required dependency specifically for a dedicated pandas string dtype.
+- PDEP-10 did not consider the usage of the experimental `pd.NA` as a
+  consequence of adopting one of the existing implementations of the
+  `StringDtype`.
+
+For the second aspect, another variant of the `StringDtype` was
+[introduced in pandas 2.1](https://pandas.pydata.org/docs/whatsnew/v2.1.0.html#whatsnew-210-enhancements-infer-strings)
+that is still backed by PyArrow but follows the default missing values semantics
+pandas uses for all other default data types (and using `NaN` as the missing
+value sentinel) ([GH-54792](https://github.com/pandas-dev/pandas/issues/54792)).
+At the time, the `storage` option for this new variant was called
+`"pyarrow_numpy"` to disambiguate from the existing `"pyarrow"` option using
+`pd.NA` (but this PDEP proposes a better naming scheme, see the "Naming"
+subsection below).
+
+This last dtype variant is what users currently (pandas 2.2) get for string data
+when enabling the ``future.infer_string`` option (to enable the behaviour which
+is intended to become the default in pandas 3.0).
+
+## Proposal
+
+To be able to move forward with a string data type in pandas 3.0, this PDEP proposes:
+
+1. For pandas 3.0, a `"str"` string dtype is enabled by default, i.e. this
+   string dtype will be used as the default dtype for text data when creating
+   pandas objects (e.g. inference in constructors, I/O functions).
+2. This default string dtype will follow the same behaviour for missing values
+   as other default data types, and use `NaN` as the missing value sentinel.
+3. The string dtype will use PyArrow if installed, and otherwise falls back to
+   an in-house functionally-equivalent (but slower) version. This fallback can
+   reuse (with minor code additions) the existing numpy object-dtype backed
+   StringArray for its implementation.
+4. Installation guidelines are updated to clearly encourage users to install
+   pyarrow for the default user experience.
+
+Those string dtypes enabled by default will then no longer be considered as
+experimental.
+
+### Default inference of a string dtype
+
+By default, pandas will infer this new string dtype instead of object dtype for
+string data (when creating pandas objects, such as in constructors or IO
+functions).
+
+In pandas 2.2, the existing `future.infer_string` option can be used to opt-in to the future
+default behaviour:
+
+```python
+>>> pd.options.future.infer_string = True
+>>> pd.Series(["a", "b", None])
+0      a
+1      b
+2    NaN
+dtype: string
+```
+
+Right now (pandas 2.2), the existing option only enables the PyArrow-based
+future dtype. For the remaining 2.x releases, this option will be expanded to
+also work when PyArrow is not installed to enable the object-dtype fallback in
+that case.
+
+### Missing value semantics
+
+As mentioned in the background section, the original `StringDtype` has always
+used the experimental `pd.NA` sentinel for missing values. In addition to using
+`pd.NA` as the scalar for a missing value, this essentially means that:
+
+- String columns follow ["NA-semantics"](https://pandas.pydata.org/docs/user_guide/missing_data.html#na-semantics)
+  for missing values, where `NA` propagates in boolean operations such as
+  comparisons or predicates.
+- Operations on the string column that give a numeric or boolean result use the
+  nullable Integer/Float/Boolean data types (e.g. `ser.str.len()` returns the
+  nullable `"Int64"` / `pd.Int64Dtype()` dtype instead of the numpy `int64`
+  dtype (or `float64` in case of missing values)).
+
+However, up to this date, all other default data types still use `NaN` semantics
+for missing values. Therefore, this proposal says that a new default string
+dtype should also still use the same default missing value semantics and return
+default data types when doing operations on the string column, to be consistent
+with the other default dtypes at this point.
+
+In practice, this means that the default string dtype will use `NaN` as
+the missing value sentinel, and:
+
+- String columns will follow NaN-semantics for missing values, where `NaN` gives
+  False in boolean operations such as comparisons or predicates.
+- Operations on the string column that give a numeric or boolean result will use
+  the default data types (i.e. numpy `int64`/`float64`/`bool`).
+
+Because the original `StringDtype` implementations already use `pd.NA` and
+return masked integer and boolean arrays in operations, a new variant of the
+existing dtypes that uses `NaN` and default data types was needed. The original
+variant of `StringDtype` using `pd.NA` will continue to be available for those
+who were already using it.
+
+### Object-dtype "fallback" implementation
+
+To avoid a hard dependency on PyArrow for pandas 3.0, this PDEP proposes to keep
+a "fallback" option in case PyArrow is not installed. The original `StringDtype`
+backed by a numpy object-dtype array of Python strings can be mostly reused for
+this (adding a new variant of the dtype) and a new `StringArray` subclass only
+needs minor changes to follow the above-mentioned missing value semantics
+([GH-58451](https://github.com/pandas-dev/pandas/pull/58451)).
+
+For pandas 3.0, this is the most realistic option given this implementation has
+already been available for a long time. Beyond 3.0, further improvements such as
+using NumPy 2.0 ([GH-58503](https://github.com/pandas-dev/pandas/issues/58503))
+or nanoarrow ([GH-58552](https://github.com/pandas-dev/pandas/issues/58552)) can
+still be explored, but at that point that is an implementation detail that
+should not have a direct impact on users (except for performance).
+
+For the original variant of `StringDtype` using `pd.NA`, currently the default
+storage is `"python"` (the object-dtype based implementation). Also for this
+variant, it is proposed to follow the same logic for determining the default
+storage, i.e. default to `"pyarrow"` if available, and otherwise
+fall back to `"python"`.
+
+### Naming
+
+Given the long history of this topic, the naming of the dtypes is a difficult
+topic.
+
+In the first place, it should be acknowledged that most users should not need to
+use storage-specific options. Users are expected to specify a generic name (such
+as `"str"` or `"string"`), and that will give them their default string dtype
+(which depends on whether PyArrow is installed or not).
+
+For the generic string alias to specify the dtype, `"string"` is already used
+for the `StringDtype` using `pd.NA`. This PDEP proposes to use `"str"` for the
+new default `StringDtype` using `NaN`. This ensures backwards compatibility for
+code using `dtype="string"`, and was also chosen because `dtype="str"` or
+`dtype=str` currently already works to ensure your data is converted to
+strings (only using object dtype for the result).
+
+But for testing purposes and advanced use cases that want control over the exact
+variant of the `StringDtype`, we need some way to specify this and distinguish
+them from the other string dtypes.
+
+Currently (pandas 2.2), `StringDtype(storage="pyarrow_numpy")` is used for the new variant using `NaN`,
+where the `"pyarrow_numpy"` storage was used to disambiguate from the existing
+`"pyarrow"` option using `pd.NA`. However, `"pyarrow_numpy"` is a rather confusing
+option and doesn't generalize well. Therefore, this PDEP proposes a new naming
+scheme as outlined below, and `"pyarrow_numpy"` will be deprecated as an alias
+in pandas 2.3 and removed in pandas 3.0.
+
+The `storage` keyword of `StringDtype` is kept to disambiguate the underlying
+storage of the string data (using pyarrow or python objects), but an additional
+`na_value` is introduced to disambiguate the the variants using NA semantics
+and NaN semantics.
+
+Overview of the different ways to specify a dtype and the resulting concrete
+dtype of the data:
+
+| User specification                          | Concrete dtype                                                | String alias                          | Note     |
+|---------------------------------------------|---------------------------------------------------------------|---------------------------------------|----------|
+| Unspecified (inference)                     | `StringDtype(storage="pyarrow"\|"python", na_value=np.nan)`   | "str"                                 | (1)      |
+| `"str"` or `StringDtype(na_value=np.nan)`   | `StringDtype(storage="pyarrow"\|"python", na_value=np.nan)`   | "str"                                 | (1)      |
+| `StringDtype("pyarrow", na_value=np.nan)`   | `StringDtype(storage="pyarrow", na_value=np.nan)`             | "str"                                 |          |
+| `StringDtype("python", na_value=np.nan)`    | `StringDtype(storage="python", na_value=np.nan)`              | "str"                                 |          |
+| `StringDtype("pyarrow")`                    | `StringDtype(storage="pyarrow", na_value=pd.NA)`              | "string[pyarrow]"                     |          |
+| `StringDtype("python")`                     | `StringDtype(storage="python", na_value=pd.NA)`               | "string[python]"                      |          |
+| `"string"` or `StringDtype()`               | `StringDtype(storage="pyarrow"\|"python", na_value=pd.NA)`    | "string[pyarrow]" or "string[python]" | (1)      |
+| `StringDtype("pyarrow_numpy")`              | `StringDtype(storage="pyarrow", na_value=np.nan)`             | "string[pyarrow_numpy]"               | (2)      |
+
+Notes:
+
+- (1) You get "pyarrow" or "python" depending on pyarrow being installed.
+- (2) "pyarrow_numpy" is kept temporarily because this is already in a released
+  version, but it will be deprecated in 2.x and removed for 3.0.
+
+For the new default string dtype, only the `"str"` alias can be used to
+specify the dtype as a string, i.e. pandas would not provide a way to make the
+underlying storage (pyarrow or python) explicit through the string alias. This
+string alias is only a convenience shortcut and for most users `"str"` is
+sufficient (they don't need to specify the storage), and the explicit
+`pd.StringDtype(storage=..., na_value=np.nan)` is still available for more
+fine-grained control.
+
+Also for the existing variant using `pd.NA`, specifying the storage through the
+string alias could be deprecated, but that is left for a separate decision.
+
+## Alternatives
+
+### Why not delay introducing a default string dtype?
+
+To avoid introducing a new string dtype while other discussions and changes are
+in flux (eventually making pyarrow a required dependency? adopting `pd.NA` as
+the default missing value sentinel? using the new NumPy 2.0 capabilities?
+overhauling all our dtypes to use a logical data type system?), introducing a
+default string dtype could also be delayed until there is more clarity in those
+other discussions. Specifically, it would avoid temporarily switching to use
+`NaN` for the string dtype, while in a future version we might switch back
+to `pd.NA` by default.
+
+However:
+
+1. Delaying has a cost: it further postpones introducing a dedicated string
+   dtype that has significant benefits for users, both in usability as (for the
+   part of the user base that has PyArrow installed) in performance.
+2. In case pandas eventually transitions to use `pd.NA` as the default missing value
+   sentinel, a migration path for _all_ pandas data types will be needed, and thus
+   the challenges around this will not be unique to the string dtype and
+   therefore not a reason to delay this.
+
+Making this change now for 3.0 will benefit the majority of users, and the PDEP
+author believes this is worth the cost of the added complexity around "yet
+another dtype" (also for other data types we already have multiple variants).
+
+### Why not use the existing StringDtype with `pd.NA`?
+
+Wouldn't adding even more variants of the string dtype make things only more
+confusing? Indeed, this proposal unfortunately introduces more variants of the
+string dtype. However, the reason for this is to ensure the actual default user
+experience is _less_ confusing, and the new string dtype fits better with the
+other default data types.
+
+If the new default string data type would use `pd.NA`, then after some
+operations, a user can easily end up with a DataFrame that mixes columns using
+`NaN` semantics and columns using `NA` semantics (and thus a DataFrame that
+could have columns with two different int64, two different float64, two different
+bool, etc dtypes). This would lead to a very confusing default experience.
+
+With the proposed new variant of the StringDtype, this will ensure that for the
+_default_ experience, a user will only see only 1 kind of integer dtype, only
+kind of 1 bool dtype, etc. For now, a user should only get columns using `pd.NA`
+when explicitly opting into this.
+
+### Naming alternatives
+
+An initial version of this PDEP proposed to use the `"string"` alias and the
+default `pd.StringDtype()` class constructor for the new default dtype.
+However, that caused a lot of discussion around backwards compatibility for
+existing users of `dtype=pd.StringDtype()` and `dtype="string"`, that uses
+`pd.NA` to represent missing values.
+
+During the discussion, several alternatives have been brought up. Both
+alternative keyword names as using a different constructor. In the end,
+this PDEP proposes to use a different string alias (`"str"`) but to keep
+using the existing `pd.StringDtype` (with the existing `storage` keyword but
+with an additional `na_value` keyword) for now to keep the changes as
+minimal as possible, leaving a larger overhaul of the dtype system (potentially
+including different constructor functions or namespace) for a future discussion.
+See [GH-58613](https://github.com/pandas-dev/pandas/issues/58613) for the full
+discussion.
+
+One consequence is that when using the class constructor for the default dtype,
+it has to be used with non-default arguments, i.e. a user needs to specify
+`pd.StringDtype(na_value=np.nan)` to get the default dtype using `NaN`.
+Therefore, the pandas documentation will focus on the usage of `dtype="str"`.
+
+## Backward compatibility
+
+The most visible backwards incompatible change will be that columns with string
+data will no longer have an `object` dtype. Therefore, code that assumes
+`object` dtype (such as `ser.dtype == object`) will need to be updated. This
+change is done as a hard break in a major release, as warning in advance for the
+changed inference is deemed too noisy.
+
+To allow testing code in advance, the
+`pd.options.future.infer_string = True` option is available for users.
+
+Otherwise, the actual string-specific functionality (such as the `.str` accessor
+methods) should generally all keep working as is.
+
+By preserving the current missing value semantics, this proposal is also mostly
+backwards compatible on this aspect. When storing strings in object dtype, pandas
+however did allow using `None` as the missing value indicator as well (and in
+certain cases such as the `shift` method, pandas even introduced this itself).
+For all the cases where currently `None` was used as the missing value sentinel,
+this will change to consistently use `NaN`.
+
+### For existing users of `StringDtype`
+
+Existing code that already opted in to use the `StringDtype` using `pd.NA`
+should generally keep working as is. The latest version of this PDEP preserves
+the behaviour of `dtype="string"` or `dtype=pd.StringDtype()` to mean the
+`pd.NA` variant of the dtype.
+
+It does propose the change the default storage to `"pyarrow"` (if available) for
+the opt-in `pd.NA` variant as well, but this should have limited, if any,
+user-visible impact.
+
+## Timeline
+
+The future PyArrow-backed string dtype was already made available behind a feature
+flag in pandas 2.1 (enabled by `pd.options.future.infer_string = True`).
+
+The variant using numpy object-dtype can also be backported to the 2.2.x branch
+to allow easier testing. It is proposed to release this as 2.3.0 (created from
+the 2.2.x branch, given that the main branch already includes many other changes
+targeted for 3.0), together with the changes to the naming scheme.
+
+The 2.3.0 release would then have all future string functionality available
+(both the pyarrow and object-dtype based variants of the default string dtype).
+
+For pandas 3.0, this `future.infer_string` flag becomes enabled by default.
+
+## PDEP-14 History
+
+- 3 May 2024: Initial version
diff --git a/pandas/web/pandas/pdeps/0017-backwards-compatibility-and-deprecation-policy.md b/pandas/web/pandas/pdeps/0017-backwards-compatibility-and-deprecation-policy.md
new file mode 100644
index 0000000000000..b8eba90f399c9
--- /dev/null
+++ b/pandas/web/pandas/pdeps/0017-backwards-compatibility-and-deprecation-policy.md
@@ -0,0 +1,74 @@
+# PDEP-17: Backwards compatibility and deprecation policy
+
+- Created: 27 June 2024
+- Status: Accepted
+- Discussion: [#59125](https://github.com/pandas-dev/pandas/issues/59125)
+- Author: [Abdulaziz Aloqeely](https://github.com/Aloqeely)
+- Revision: 1
+
+## Abstract
+
+This PDEP defines pandas' backwards compatibility and deprecation policy.
+
+The main additions to [pandas' current version policy](https://pandas.pydata.org/pandas-docs/version/2.2/development/policies.html) are:
+- Deprecated functionality should remain unchanged in at least 2 minor releases before being changed or removed.
+- Deprecations should initially use DeprecationWarning, and then be switched to FutureWarning in the last minor release before the major release they are planned to be removed in
+
+## Motivation
+
+Having a clear backwards compatibility and deprecation policy is crucial to having a healthy ecosystem. We want to ensure users can rely on pandas being stable while still allowing the library to evolve.
+
+This policy will ensure that users have enough time to deal with deprecations while also minimizing disruptions on downstream packages' users.
+
+## Scope
+
+This PDEP covers pandas' approach to backwards compatibility and the deprecation and removal process.
+
+## Background
+
+pandas uses a loose variant of semantic versioning.  
+A pandas release number is written in the format of ``MAJOR.MINOR.PATCH``.
+
+## General policy
+
+This policy applies to the [public API][1]. Anything not part of the [public API][1] or is marked as "Experimental" may be changed or removed at anytime.
+
+- Breaking backwards compatibility should benefit more than it harms users.
+- Breaking changes should go through a deprecation cycle before being implemented if possible.
+- Breaking changes should only occur in major releases.
+- No deprecations should be introduced in patch releases.
+- Deprecated functionality should remain unchanged in at least 2 minor releases before being changed or removed.
+
+Some bug fixes may require breaking backwards compatibility. In these cases, a deprecation cycle is not necessary. However, bug fixes which have a large impact on users might be treated as a breaking change. Whether or not a change is a bug fix or an API breaking change is a judgement call.
+
+## Deprecation process
+
+Deprecation provides a way to warn developers and give them time to adapt their code to the new functionality before the old behavior is eventually removed.
+
+A deprecation's warning message should:
+- Provide information on what is changing.
+- Mention how to achieve similar behavior if an alternative is available.
+- For large-scale deprecations, it is recommended to include a reason for the deprecation, alongside a discussion link to get user feedback.
+
+Additionally, when one introduces a deprecation, they should:
+- Use the appropriate warning class. More info on this can be found below.
+- Add the GitHub issue/PR number as a comment above the warning line.
+- Add an entry in the release notes.
+- Mention that the functionality is deprecated in the documentation using the ``.. deprecated::`` directive.
+
+### Which warning class to use
+
+Deprecations should initially use ``DeprecationWarning``, and then be switched to ``FutureWarning`` for broader visibility in the last minor release before the major release they are planned to be removed in.  
+This implementation detail can be ignored by using the appropriate ``PandasDeprecationWarning`` variable, which will be aliased to the proper warning class based on the pandas version.
+
+### Enforcement of deprecations
+
+When one enforces a deprecation, they should:
+- Add an entry in the release notes.
+- For API changes, replace the ``.. deprecated::`` directive in the documentation with a ``.. versionchanged::`` directive.
+
+### PDEP-17 History
+
+- 27 June 2024: Initial version.
+
+[1]: https://pandas.pydata.org/docs/reference/index.html
diff --git a/pandas/web/pandas/robots.txt b/pandas/web/pandas/robots.txt
new file mode 100644
index 0000000000000..91ac176b957a3
--- /dev/null
+++ b/pandas/web/pandas/robots.txt
@@ -0,0 +1,2 @@
+User-agent: *
+Disallow: /preview/
diff --git a/pandas/web/pandas/static/css/codehilite.css b/pandas/web/pandas/static/css/codehilite.css
new file mode 100644
index 0000000000000..4a0078e141d87
--- /dev/null
+++ b/pandas/web/pandas/static/css/codehilite.css
@@ -0,0 +1,75 @@
+pre { line-height: 125%; }
+td.linenos .normal { color: #666666; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+span.linenos { color: #666666; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+.codehilite .hll { background-color: #ffffcc }
+.codehilite { background: #f0f0f0; }
+.codehilite .c { color: #60a0b0; font-style: italic } /* Comment */
+.codehilite .err { border: 1px solid #FF0000 } /* Error */
+.codehilite .k { color: #007020; font-weight: bold } /* Keyword */
+.codehilite .o { color: #666666 } /* Operator */
+.codehilite .ch { color: #60a0b0; font-style: italic } /* Comment.Hashbang */
+.codehilite .cm { color: #60a0b0; font-style: italic } /* Comment.Multiline */
+.codehilite .cp { color: #007020 } /* Comment.Preproc */
+.codehilite .cpf { color: #60a0b0; font-style: italic } /* Comment.PreprocFile */
+.codehilite .c1 { color: #60a0b0; font-style: italic } /* Comment.Single */
+.codehilite .cs { color: #60a0b0; background-color: #fff0f0 } /* Comment.Special */
+.codehilite .gd { color: #A00000 } /* Generic.Deleted */
+.codehilite .ge { font-style: italic } /* Generic.Emph */
+.codehilite .gr { color: #FF0000 } /* Generic.Error */
+.codehilite .gh { color: #000080; font-weight: bold } /* Generic.Heading */
+.codehilite .gi { color: #00A000 } /* Generic.Inserted */
+.codehilite .go { color: #888888 } /* Generic.Output */
+.codehilite .gp { color: #c65d09; font-weight: bold } /* Generic.Prompt */
+.codehilite .gs { font-weight: bold } /* Generic.Strong */
+.codehilite .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
+.codehilite .gt { color: #0044DD } /* Generic.Traceback */
+.codehilite .kc { color: #007020; font-weight: bold } /* Keyword.Constant */
+.codehilite .kd { color: #007020; font-weight: bold } /* Keyword.Declaration */
+.codehilite .kn { color: #007020; font-weight: bold } /* Keyword.Namespace */
+.codehilite .kp { color: #007020 } /* Keyword.Pseudo */
+.codehilite .kr { color: #007020; font-weight: bold } /* Keyword.Reserved */
+.codehilite .kt { color: #902000 } /* Keyword.Type */
+.codehilite .m { color: #40a070 } /* Literal.Number */
+.codehilite .s { color: #4070a0 } /* Literal.String */
+.codehilite .na { color: #4070a0 } /* Name.Attribute */
+.codehilite .nb { color: #007020 } /* Name.Builtin */
+.codehilite .nc { color: #0e84b5; font-weight: bold } /* Name.Class */
+.codehilite .no { color: #60add5 } /* Name.Constant */
+.codehilite .nd { color: #555555; font-weight: bold } /* Name.Decorator */
+.codehilite .ni { color: #d55537; font-weight: bold } /* Name.Entity */
+.codehilite .ne { color: #007020 } /* Name.Exception */
+.codehilite .nf { color: #06287e } /* Name.Function */
+.codehilite .nl { color: #002070; font-weight: bold } /* Name.Label */
+.codehilite .nn { color: #0e84b5; font-weight: bold } /* Name.Namespace */
+.codehilite .nt { color: #062873; font-weight: bold } /* Name.Tag */
+.codehilite .nv { color: #bb60d5 } /* Name.Variable */
+.codehilite .ow { color: #007020; font-weight: bold } /* Operator.Word */
+.codehilite .w { color: #bbbbbb } /* Text.Whitespace */
+.codehilite .mb { color: #40a070 } /* Literal.Number.Bin */
+.codehilite .mf { color: #40a070 } /* Literal.Number.Float */
+.codehilite .mh { color: #40a070 } /* Literal.Number.Hex */
+.codehilite .mi { color: #40a070 } /* Literal.Number.Integer */
+.codehilite .mo { color: #40a070 } /* Literal.Number.Oct */
+.codehilite .sa { color: #4070a0 } /* Literal.String.Affix */
+.codehilite .sb { color: #4070a0 } /* Literal.String.Backtick */
+.codehilite .sc { color: #4070a0 } /* Literal.String.Char */
+.codehilite .dl { color: #4070a0 } /* Literal.String.Delimiter */
+.codehilite .sd { color: #4070a0; font-style: italic } /* Literal.String.Doc */
+.codehilite .s2 { color: #4070a0 } /* Literal.String.Double */
+.codehilite .se { color: #4070a0; font-weight: bold } /* Literal.String.Escape */
+.codehilite .sh { color: #4070a0 } /* Literal.String.Heredoc */
+.codehilite .si { color: #70a0d0; font-style: italic } /* Literal.String.Interpol */
+.codehilite .sx { color: #c65d09 } /* Literal.String.Other */
+.codehilite .sr { color: #235388 } /* Literal.String.Regex */
+.codehilite .s1 { color: #4070a0 } /* Literal.String.Single */
+.codehilite .ss { color: #517918 } /* Literal.String.Symbol */
+.codehilite .bp { color: #007020 } /* Name.Builtin.Pseudo */
+.codehilite .fm { color: #06287e } /* Name.Function.Magic */
+.codehilite .vc { color: #bb60d5 } /* Name.Variable.Class */
+.codehilite .vg { color: #bb60d5 } /* Name.Variable.Global */
+.codehilite .vi { color: #bb60d5 } /* Name.Variable.Instance */
+.codehilite .vm { color: #bb60d5 } /* Name.Variable.Magic */
+.codehilite .il { color: #40a070 } /* Literal.Number.Integer.Long */
+.codehilite > pre { padding: 1em; border: 0.1em solid darkgrey; border-radius: 0.4em; }
diff --git a/pandas/web/pandas/static/css/pandas.css b/pandas/web/pandas/static/css/pandas.css
new file mode 100644
index 0000000000000..ec9a4bd502dd1
--- /dev/null
+++ b/pandas/web/pandas/static/css/pandas.css
@@ -0,0 +1,105 @@
+body {
+    padding-top: 5em;
+    color: #444;
+}
+h1 {
+    font-size: 2.4rem;
+    font-weight: 700;
+    color: #130654;
+}
+h2 {
+    font-size: 1.8rem;
+    font-weight: 700;
+    color: #130654;
+}
+h3 {
+    font-size: 1.3rem;
+    font-weight: 600;
+    color: black;
+}
+h3 a {
+    color: black;
+    text-decoration: underline dotted !important;
+}
+a {
+    color: #130654;
+}
+code {
+    white-space: pre;
+}
+ol ol, ol ul, ul ol, ul ul {
+    margin-bottom: 1rem;
+}
+.blue {
+    color: #150458;
+}
+.pink {
+    color: #e70488;
+}
+.follow-us-button {
+    font-size: 2.4rem !important;
+    color: #0d6efd !important;
+}
+.follow-us-button:hover {
+    color: #0b5ed7 !important;
+}
+.fab {
+    font-size: 1.2rem;
+    color: #666;
+}
+.fab:hover {
+    color: #130654;
+}
+a.navbar-brand img {
+    height: 3rem;
+}
+a:link:not(.btn):not(.dropdown-item):not(.nav-link) {
+text-decoration: none;
+}
+a:visited:not(.btn):not(.dropdown-item):not(.nav-link) {
+text-decoration: none;
+}
+a:hover:not(.btn):not(.dropdown-item):not(.nav-link) {
+text-decoration: underline;
+}
+a:active:not(.btn):not(.dropdown-item):not(.nav-link) {
+text-decoration: underline;
+}
+div.card {
+    margin: 0 0 .2em .2em !important;
+}
+@media (min-width: 576px) {
+    .card-group.maintainers div.card {
+        min-width: 10rem;
+        max-width: 10rem;
+    }
+}
+div.card .card-title {
+    font-weight: 500;
+    color: #130654;
+}
+.book {
+    padding: 0 20%;
+}
+.bg-dark {
+    background-color: #130654 !important;
+}
+.navbar-dark .navbar-nav .nav-link {
+    color: rgba(255, 255, 255, .9);
+}
+.navbar-dark .navbar-nav .nav-link:hover {
+    color: white;
+}
+table.logo td {
+    text-align: center;
+}
+table.logo img {
+    height: 4rem;
+}
+blockquote {
+    background: #f9f9f9;
+    border-left: 5px solid #ccc;
+    padding-left:15px;
+    color: #787878;
+    font-size: 18px;
+ }
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_13_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_13_0.png
new file mode 100644
index 0000000000000..9ce2ff483f2c2
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_13_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_18_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_18_0.png
new file mode 100644
index 0000000000000..63b2c93b0573d
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_18_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_20_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_20_0.png
new file mode 100644
index 0000000000000..1c7abb0434dad
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_20_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_22_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_22_0.png
new file mode 100644
index 0000000000000..5ef3d69b48700
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_22_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_24_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_24_0.png
new file mode 100644
index 0000000000000..1a15be05af92d
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_24_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_26_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_26_0.png
new file mode 100644
index 0000000000000..4f8d9f2c439ae
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_26_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_31_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_31_0.png
new file mode 100644
index 0000000000000..6c8b5f1108f79
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_31_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_33_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_33_0.png
new file mode 100644
index 0000000000000..fd490d3e7255a
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_33_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_4_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_4_0.png
new file mode 100644
index 0000000000000..5276ed359badb
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_4_0.png differ
diff --git a/pandas/web/pandas/static/img/blog/2019-user-survey/2019_5_0.png b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_5_0.png
new file mode 100644
index 0000000000000..a252e1c9b3503
Binary files /dev/null and b/pandas/web/pandas/static/img/blog/2019-user-survey/2019_5_0.png differ
diff --git a/pandas/web/pandas/static/img/books/effective_pandas_2.gif b/pandas/web/pandas/static/img/books/effective_pandas_2.gif
new file mode 100644
index 0000000000000..ae3efbcbd18a8
Binary files /dev/null and b/pandas/web/pandas/static/img/books/effective_pandas_2.gif differ
diff --git a/pandas/web/pandas/static/img/books/pydata_book.gif b/pandas/web/pandas/static/img/books/pydata_book.gif
new file mode 100644
index 0000000000000..2c9ad9f1cefd7
Binary files /dev/null and b/pandas/web/pandas/static/img/books/pydata_book.gif differ
diff --git a/pandas/web/pandas/static/img/favicon.ico b/pandas/web/pandas/static/img/favicon.ico
new file mode 100644
index 0000000000000..0af2443dcaa3e
Binary files /dev/null and b/pandas/web/pandas/static/img/favicon.ico differ
diff --git a/pandas/web/pandas/static/img/favicon_white.ico b/pandas/web/pandas/static/img/favicon_white.ico
new file mode 100644
index 0000000000000..68223fc92f456
Binary files /dev/null and b/pandas/web/pandas/static/img/favicon_white.ico differ
diff --git a/pandas/web/pandas/static/img/pandas.svg b/pandas/web/pandas/static/img/pandas.svg
new file mode 100644
index 0000000000000..a7af4e4d2d401
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 818.63 331.21"><defs><style>.cls-1{fill:#130754;}.cls-2{fill:#ffca00;}.cls-3{fill:#e70488;}</style></defs><title>Artboard 63</title><path class="cls-1" d="M290.85,199.21c-10.27,0-20.73-4.25-27.28-12.58v45H243l0-111.09h18.6l.71,12.22c6.38-9.39,17.71-14.35,28.52-14.35,20.73,0,36,17.37,36,40.4S311.58,199.22,290.85,199.21Zm-6.37-65.55c-12.05,0-21.79,9.39-21.79,25.16S272.43,184,284.48,184s21.79-9.39,21.79-25.16S296.53,133.66,284.48,133.66Z"/><path class="cls-1" d="M404.36,197.1l-.71-12.22c-6.38,9.39-17.72,14.35-28.53,14.34-20.73,0-36-17.36-36-40.39s15.24-40.4,36-40.39c10.81,0,22.15,5,28.53,14.35l.71-12.22H423V197.1Zm-22.85-63.43c-12.05,0-21.79,9.39-21.8,25.16S369.45,184,381.5,184s21.8-9.39,21.8-25.16S393.56,133.67,381.51,133.67Z"/><path class="cls-1" d="M494.87,197.11V154.77c0-14.88-5.13-19.84-14.52-19.84-9.75,0-20.38,8.85-20.38,19.48v42.7H439.41V120.57H458.2l.89,14.18c5.14-9.75,16.65-16.3,28.35-16.3,20.37,0,28,14.18,28,33.13v45.54Z"/><path class="cls-1" d="M590.77,197.13l-.71-12.23c-6.38,9.39-17.72,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.39,36-40.39c10.27,0,20.72,4.26,27.28,12.58V90.83h20.56l0,106.3ZM567.92,133.7c-12,0-21.79,9.39-21.79,25.15S555.87,184,567.92,184s21.79-9.38,21.79-25.15S580,133.7,567.92,133.7Z"/><path class="cls-1" d="M686.6,197.14l-.71-12.22c-6.38,9.39-17.72,14.34-28.53,14.34-20.73,0-36-17.36-36-40.4s15.24-40.39,36-40.39c10.81,0,22.15,5,28.53,14.36l.71-12.23h18.6v76.53Zm-22.85-63.43c-12,0-21.79,9.39-21.8,25.16S651.7,184,663.74,184s21.8-9.39,21.8-25.16S675.8,133.71,663.75,133.71Z"/><path class="cls-1" d="M750.73,199.63a60.16,60.16,0,0,1-30.65-8.69l3.37-14.17c6.2,3.72,15.59,8.51,26.93,8.51,8.15,0,13.82-2.48,13.82-8.86,0-5.49-5.85-7.44-16.3-9.92-18.78-4.08-25.51-14-25.51-24.81,0-12.05,9.39-23.38,30.12-23.38,12.58,0,23.57,5.49,26,6.91l-3.37,13.47A44.59,44.59,0,0,0,753,132.31c-8.32,0-12.4,2.83-12.4,7.44,0,5.13,5.32,7.44,13.46,9.39,20.2,4.25,28.35,13.64,28.35,23.92C782.45,189.53,770.4,199.63,750.73,199.63Z"/><rect class="cls-1" x="74.88" y="68.42" width="24.09" height="50.02"/><rect class="cls-1" x="74.88" y="171.17" width="24.09" height="50.02"/><rect class="cls-2" x="74.88" y="133.04" width="24.09" height="23.6"/><rect class="cls-1" x="36.19" y="109.55" width="24.09" height="166.27"/><rect class="cls-1" x="112.78" y="212.44" width="24.09" height="50.02"/><rect class="cls-1" x="112.78" y="109.61" width="24.09" height="50.02"/><rect class="cls-3" x="112.78" y="174.23" width="24.09" height="23.6"/><rect class="cls-1" x="150.67" y="55.39" width="24.09" height="166.27"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/pandas_mark.svg b/pandas/web/pandas/static/img/pandas_mark.svg
new file mode 100644
index 0000000000000..1451f57de198e
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas_mark.svg
@@ -0,0 +1,111 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   id="Layer_1"
+   data-name="Layer 1"
+   viewBox="0 0 210.21 280.43"
+   version="1.1"
+   sodipodi:docname="pandas_mark.svg"
+   inkscape:version="0.92.4 (unknown)">
+  <metadata
+     id="metadata27">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <sodipodi:namedview
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1"
+     objecttolerance="10"
+     gridtolerance="10"
+     guidetolerance="10"
+     inkscape:pageopacity="0"
+     inkscape:pageshadow="2"
+     inkscape:window-width="1131"
+     inkscape:window-height="921"
+     id="namedview25"
+     showgrid="false"
+     inkscape:zoom="0.84156476"
+     inkscape:cx="107.48153"
+     inkscape:cy="140.215"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="Layer_1" />
+  <defs
+     id="defs4">
+    <style
+       id="style2">.cls-1{fill:#130754;}.cls-2{fill:#48e5ac;}.cls-3{fill:#e70488;}</style>
+  </defs>
+  <title
+     id="title6">Artboard 61</title>
+  <rect
+     class="cls-1"
+     x="74.51"
+     y="43.03"
+     width="24.09"
+     height="50.02"
+     id="rect8" />
+  <rect
+     class="cls-1"
+     x="74.51"
+     y="145.78"
+     width="24.09"
+     height="50.02"
+     id="rect10" />
+  <rect
+     class="cls-2"
+     x="74.51"
+     y="107.65"
+     width="24.09"
+     height="23.6"
+     id="rect12"
+     style="fill:#ffca00;fill-opacity:1" />
+  <rect
+     class="cls-1"
+     x="35.81"
+     y="84.15"
+     width="24.09"
+     height="166.27"
+     id="rect14" />
+  <rect
+     class="cls-1"
+     x="112.41"
+     y="187.05"
+     width="24.09"
+     height="50.02"
+     id="rect16" />
+  <rect
+     class="cls-1"
+     x="112.41"
+     y="84.21"
+     width="24.09"
+     height="50.02"
+     id="rect18" />
+  <rect
+     class="cls-3"
+     x="112.41"
+     y="148.84"
+     width="24.09"
+     height="23.6"
+     id="rect20" />
+  <rect
+     class="cls-1"
+     x="150.3"
+     y="30"
+     width="24.09"
+     height="166.27"
+     id="rect22" />
+</svg>
diff --git a/pandas/web/pandas/static/img/pandas_mark_white.svg b/pandas/web/pandas/static/img/pandas_mark_white.svg
new file mode 100644
index 0000000000000..ae50bf5430c3a
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas_mark_white.svg
@@ -0,0 +1,111 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   id="Layer_1"
+   data-name="Layer 1"
+   viewBox="0 0 210.21 280.43"
+   version="1.1"
+   sodipodi:docname="pandas_mark_white.svg"
+   inkscape:version="0.92.4 (unknown)">
+  <metadata
+     id="metadata27">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <sodipodi:namedview
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1"
+     objecttolerance="10"
+     gridtolerance="10"
+     guidetolerance="10"
+     inkscape:pageopacity="0"
+     inkscape:pageshadow="2"
+     inkscape:window-width="761"
+     inkscape:window-height="480"
+     id="namedview25"
+     showgrid="false"
+     inkscape:zoom="0.84156476"
+     inkscape:cx="105.105"
+     inkscape:cy="140.215"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="Layer_1" />
+  <defs
+     id="defs4">
+    <style
+       id="style2">.cls-1{fill:#fff;}.cls-2{fill:#48e5ac;}.cls-3{fill:#e70488;}</style>
+  </defs>
+  <title
+     id="title6">Artboard 61 copy</title>
+  <rect
+     class="cls-1"
+     x="74.51"
+     y="43.03"
+     width="24.09"
+     height="50.02"
+     id="rect8" />
+  <rect
+     class="cls-1"
+     x="74.51"
+     y="145.78"
+     width="24.09"
+     height="50.02"
+     id="rect10" />
+  <rect
+     class="cls-2"
+     x="74.51"
+     y="107.65"
+     width="24.09"
+     height="23.6"
+     id="rect12"
+     style="fill:#ffca00;fill-opacity:1" />
+  <rect
+     class="cls-1"
+     x="35.81"
+     y="84.15"
+     width="24.09"
+     height="166.27"
+     id="rect14" />
+  <rect
+     class="cls-1"
+     x="112.41"
+     y="187.05"
+     width="24.09"
+     height="50.02"
+     id="rect16" />
+  <rect
+     class="cls-1"
+     x="112.41"
+     y="84.21"
+     width="24.09"
+     height="50.02"
+     id="rect18" />
+  <rect
+     class="cls-3"
+     x="112.41"
+     y="148.84"
+     width="24.09"
+     height="23.6"
+     id="rect20" />
+  <rect
+     class="cls-1"
+     x="150.3"
+     y="30"
+     width="24.09"
+     height="166.27"
+     id="rect22" />
+</svg>
diff --git a/pandas/web/pandas/static/img/pandas_secondary.svg b/pandas/web/pandas/static/img/pandas_secondary.svg
new file mode 100644
index 0000000000000..e74404842e5b6
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas_secondary.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 664.97 470.93"><defs><style>.cls-1{fill:#130754;}.cls-2{fill:#ffca00;}.cls-3{fill:#e70488;}</style></defs><title>Artboard 57</title><path class="cls-1" d="M110.61,397.48c-10.28,0-20.73-4.25-27.29-12.58v45H62.76l0-111.08h18.6L82.09,331c6.38-9.39,17.72-14.35,28.53-14.34,20.72,0,36,17.36,36,40.4S131.33,397.49,110.61,397.48Zm-6.37-65.55c-12,0-21.8,9.39-21.8,25.16s9.74,25.16,21.79,25.16S126,372.86,126,357.09,116.28,331.93,104.24,331.93Z"/><path class="cls-1" d="M224.11,395.37l-.71-12.22c-6.38,9.39-17.72,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.4,36-40.39c10.8,0,22.14,5,28.52,14.35l.71-12.22h18.6v76.54Zm-22.85-63.43c-12,0-21.79,9.39-21.79,25.16s9.74,25.16,21.79,25.16,21.79-9.39,21.79-25.15S213.31,332,201.26,331.94Z"/><path class="cls-1" d="M314.62,395.39V353c0-14.88-5.14-19.84-14.53-19.84-9.74,0-20.37,8.85-20.38,19.48v42.7H259.17V318.84H278l.88,14.18c5.14-9.75,16.66-16.3,28.35-16.3,20.37,0,28,14.18,28,33.14v45.53Z"/><path class="cls-1" d="M410.52,395.4l-.71-12.23c-6.37,9.39-17.71,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.39,36-40.39c10.27,0,20.73,4.26,27.28,12.59V289.1h20.55l0,106.3ZM387.68,332c-12.05,0-21.8,9.39-21.8,25.16s9.74,25.15,21.79,25.16,21.79-9.39,21.79-25.16S399.72,332,387.68,332Z"/><path class="cls-1" d="M506.35,395.41l-.71-12.22c-6.38,9.39-17.72,14.35-28.52,14.34-20.73,0-36-17.36-36-40.39s15.24-40.4,36-40.39c10.8,0,22.14,5,28.52,14.35l.71-12.22H525v76.53ZM483.5,332c-12.05,0-21.79,9.39-21.79,25.16s9.74,25.16,21.79,25.16,21.79-9.39,21.79-25.16S495.55,332,483.5,332Z"/><path class="cls-1" d="M570.49,397.9a60.15,60.15,0,0,1-30.65-8.68L543.2,375c6.2,3.72,15.59,8.51,26.93,8.51,8.15,0,13.82-2.48,13.82-8.86,0-5.49-5.84-7.44-16.3-9.92-18.77-4.08-25.51-14-25.5-24.81,0-12,9.39-23.38,30.12-23.38,12.58,0,23.56,5.5,26,6.91L594.94,337a44.52,44.52,0,0,0-22.14-6.38c-8.33,0-12.41,2.83-12.41,7.44,0,5.13,5.32,7.44,13.47,9.39,20.19,4.25,28.34,13.64,28.34,23.92C602.2,387.81,590.15,397.9,570.49,397.9Z"/><rect class="cls-1" x="301.89" y="54.05" width="24.09" height="50.02"/><rect class="cls-1" x="301.89" y="156.8" width="24.09" height="50.02"/><rect class="cls-2" x="301.89" y="118.68" width="24.09" height="23.6"/><rect class="cls-1" x="263.19" y="95.18" width="24.09" height="166.27"/><rect class="cls-1" x="339.79" y="198.07" width="24.09" height="50.02"/><rect class="cls-1" x="339.79" y="95.24" width="24.09" height="50.02"/><rect class="cls-3" x="339.79" y="159.86" width="24.09" height="23.6"/><rect class="cls-1" x="377.68" y="41.03" width="24.09" height="166.27"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/pandas_secondary_white.svg b/pandas/web/pandas/static/img/pandas_secondary_white.svg
new file mode 100644
index 0000000000000..86bcca57a031e
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas_secondary_white.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 664.97 470.93"><defs><style>.cls-1{fill:#fff;}.cls-2{fill:#ffca00;}.cls-3{fill:#e70488;}</style></defs><title>Artboard 57 copy</title><path class="cls-1" d="M110.61,397.48c-10.28,0-20.73-4.25-27.29-12.58v45H62.76l0-111.08h18.6L82.09,331c6.38-9.39,17.72-14.35,28.53-14.34,20.72,0,36,17.36,36,40.4S131.33,397.49,110.61,397.48Zm-6.37-65.55c-12,0-21.8,9.39-21.8,25.16s9.74,25.16,21.79,25.16S126,372.86,126,357.09,116.28,331.93,104.24,331.93Z"/><path class="cls-1" d="M224.11,395.37l-.71-12.22c-6.38,9.39-17.72,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.4,36-40.39c10.8,0,22.14,5,28.52,14.35l.71-12.22h18.6v76.54Zm-22.85-63.43c-12,0-21.79,9.39-21.79,25.16s9.74,25.16,21.79,25.16,21.79-9.39,21.79-25.15S213.31,332,201.26,331.94Z"/><path class="cls-1" d="M314.62,395.39V353c0-14.88-5.14-19.84-14.53-19.84-9.74,0-20.37,8.85-20.38,19.48v42.7H259.17V318.84H278l.88,14.18c5.14-9.75,16.66-16.3,28.35-16.3,20.37,0,28,14.18,28,33.14v45.53Z"/><path class="cls-1" d="M410.52,395.4l-.71-12.23c-6.37,9.39-17.71,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.39,36-40.39c10.27,0,20.73,4.26,27.28,12.59V289.1h20.55l0,106.3ZM387.68,332c-12.05,0-21.8,9.39-21.8,25.16s9.74,25.15,21.79,25.16,21.79-9.39,21.79-25.16S399.72,332,387.68,332Z"/><path class="cls-1" d="M506.35,395.41l-.71-12.22c-6.38,9.39-17.72,14.35-28.52,14.34-20.73,0-36-17.36-36-40.39s15.24-40.4,36-40.39c10.8,0,22.14,5,28.52,14.35l.71-12.22H525v76.53ZM483.5,332c-12.05,0-21.79,9.39-21.79,25.16s9.74,25.16,21.79,25.16,21.79-9.39,21.79-25.16S495.55,332,483.5,332Z"/><path class="cls-1" d="M570.49,397.9a60.15,60.15,0,0,1-30.65-8.68L543.2,375c6.2,3.72,15.59,8.51,26.93,8.51,8.15,0,13.82-2.48,13.82-8.86,0-5.49-5.84-7.44-16.3-9.92-18.77-4.08-25.51-14-25.5-24.81,0-12,9.39-23.38,30.12-23.38,12.58,0,23.56,5.5,26,6.91L594.94,337a44.52,44.52,0,0,0-22.14-6.38c-8.33,0-12.41,2.83-12.41,7.44,0,5.13,5.32,7.44,13.47,9.39,20.19,4.25,28.34,13.64,28.34,23.92C602.2,387.81,590.15,397.9,570.49,397.9Z"/><rect class="cls-1" x="301.89" y="54.05" width="24.09" height="50.02"/><rect class="cls-1" x="301.89" y="156.8" width="24.09" height="50.02"/><rect class="cls-2" x="301.89" y="118.68" width="24.09" height="23.6"/><rect class="cls-1" x="263.19" y="95.18" width="24.09" height="166.27"/><rect class="cls-1" x="339.79" y="198.07" width="24.09" height="50.02"/><rect class="cls-1" x="339.79" y="95.24" width="24.09" height="50.02"/><rect class="cls-3" x="339.79" y="159.86" width="24.09" height="23.6"/><rect class="cls-1" x="377.68" y="41.03" width="24.09" height="166.27"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/pandas_white.svg b/pandas/web/pandas/static/img/pandas_white.svg
new file mode 100644
index 0000000000000..bc7c41651182d
--- /dev/null
+++ b/pandas/web/pandas/static/img/pandas_white.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 818.63 331.21"><defs><style>.cls-1{fill:#fff;}.cls-2{fill:#ffca00;}.cls-3{fill:#e70488;}</style></defs><title>Artboard 63 copy 2</title><path class="cls-1" d="M290.85,199.21c-10.27,0-20.73-4.25-27.28-12.58v45H243l0-111.09h18.6l.71,12.22c6.38-9.39,17.71-14.35,28.52-14.35,20.73,0,36,17.37,36,40.4S311.58,199.22,290.85,199.21Zm-6.37-65.55c-12.05,0-21.79,9.39-21.79,25.16S272.43,184,284.48,184s21.79-9.39,21.79-25.16S296.53,133.66,284.48,133.66Z"/><path class="cls-1" d="M404.36,197.1l-.71-12.22c-6.38,9.39-17.72,14.35-28.53,14.34-20.73,0-36-17.36-36-40.39s15.24-40.4,36-40.39c10.81,0,22.15,5,28.53,14.35l.71-12.22H423V197.1Zm-22.85-63.43c-12.05,0-21.79,9.39-21.8,25.16S369.45,184,381.5,184s21.8-9.39,21.8-25.16S393.56,133.67,381.51,133.67Z"/><path class="cls-1" d="M494.87,197.11V154.77c0-14.88-5.13-19.84-14.52-19.84-9.75,0-20.38,8.85-20.38,19.48v42.7H439.41V120.57H458.2l.89,14.18c5.14-9.75,16.65-16.3,28.35-16.3,20.37,0,28,14.18,28,33.13v45.54Z"/><path class="cls-1" d="M590.77,197.13l-.71-12.23c-6.38,9.39-17.72,14.35-28.52,14.35-20.73,0-36-17.37-36-40.4s15.24-40.39,36-40.39c10.27,0,20.72,4.26,27.28,12.58V90.83h20.56l0,106.3ZM567.92,133.7c-12,0-21.79,9.39-21.79,25.15S555.87,184,567.92,184s21.79-9.38,21.79-25.15S580,133.7,567.92,133.7Z"/><path class="cls-1" d="M686.6,197.14l-.71-12.22c-6.38,9.39-17.72,14.34-28.53,14.34-20.73,0-36-17.36-36-40.4s15.24-40.39,36-40.39c10.81,0,22.15,5,28.53,14.36l.71-12.23h18.6v76.53Zm-22.85-63.43c-12,0-21.79,9.39-21.8,25.16S651.7,184,663.74,184s21.8-9.39,21.8-25.16S675.8,133.71,663.75,133.71Z"/><path class="cls-1" d="M750.73,199.63a60.16,60.16,0,0,1-30.65-8.69l3.37-14.17c6.2,3.72,15.59,8.51,26.93,8.51,8.15,0,13.82-2.48,13.82-8.86,0-5.49-5.85-7.44-16.3-9.92-18.78-4.08-25.51-14-25.51-24.81,0-12.05,9.39-23.38,30.12-23.38,12.58,0,23.57,5.49,26,6.91l-3.37,13.47A44.59,44.59,0,0,0,753,132.31c-8.32,0-12.4,2.83-12.4,7.44,0,5.13,5.32,7.44,13.46,9.39,20.2,4.25,28.35,13.64,28.35,23.92C782.45,189.53,770.4,199.63,750.73,199.63Z"/><rect class="cls-1" x="74.88" y="68.42" width="24.09" height="50.02"/><rect class="cls-1" x="74.88" y="171.17" width="24.09" height="50.02"/><rect class="cls-2" x="74.88" y="133.04" width="24.09" height="23.6"/><rect class="cls-1" x="36.19" y="109.55" width="24.09" height="166.27"/><rect class="cls-1" x="112.78" y="212.44" width="24.09" height="50.02"/><rect class="cls-1" x="112.78" y="109.61" width="24.09" height="50.02"/><rect class="cls-3" x="112.78" y="174.23" width="24.09" height="23.6"/><rect class="cls-1" x="150.67" y="55.39" width="24.09" height="166.27"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/partners/bodo.svg b/pandas/web/pandas/static/img/partners/bodo.svg
new file mode 100644
index 0000000000000..9dc6cb47505a3
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/bodo.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" width="930.84" height="267.43" viewBox="0 0 930.84 267.43"><defs><style>.cls-1{fill:#5e5e5e;}.cls-2{fill:#1db100;}</style></defs><title>bodo-grey-green</title><path class="cls-1" d="M85.73,93a83.24,83.24,0,0,1,33.14,6.64,85.67,85.67,0,0,1,0,158.11,83.45,83.45,0,0,1-33.14,6.67,83.06,83.06,0,0,1-33.34-6.67A87.46,87.46,0,0,1,6.67,212,83.23,83.23,0,0,1,0,178.67V11.44A10.82,10.82,0,0,1,3.43,3.25a11.3,11.3,0,0,1,8-3.24,10.66,10.66,0,0,1,7.81,3.24,11,11,0,0,1,3.24,8.19V121.17a85.33,85.33,0,0,1,23.24-17.91,86.89,86.89,0,0,1,19.47-7.64A81.71,81.71,0,0,1,85.77,93m0,149a62.21,62.21,0,0,0,24.57-5,63.2,63.2,0,0,0,33.72-33.53,64,64,0,0,0,0-49.34,62.65,62.65,0,0,0-33.72-33.72,63.48,63.48,0,0,0-49.15,0,62.72,62.72,0,0,0-33.72,33.72,64,64,0,0,0,0,49.34A63.25,63.25,0,0,0,61.19,237,62.19,62.19,0,0,0,85.77,242"/><path class="cls-1" d="M264.79,264.43a83.15,83.15,0,0,1-33.33-6.67,88.56,88.56,0,0,1-27.27-18.29,84.93,84.93,0,0,1-18.48-27.05,85.76,85.76,0,1,1,158.12,0,84.61,84.61,0,0,1-18.48,27.05,87.75,87.75,0,0,1-27.43,18.29,83.53,83.53,0,0,1-33.15,6.67m0-148.59a61.81,61.81,0,0,0-24.58,5,62.65,62.65,0,0,0-33.72,33.72,63.48,63.48,0,0,0,0,49.15,62.72,62.72,0,0,0,33.72,33.72,63.48,63.48,0,0,0,49.15,0,62.74,62.74,0,0,0,33.75-33.76,63.48,63.48,0,0,0,0-49.15,62.65,62.65,0,0,0-33.72-33.72,62,62,0,0,0-24.58-5"/><path class="cls-1" d="M443.48,93a85.82,85.82,0,0,1,63.24,28.19V11.06A10.66,10.66,0,0,1,510,3.25,11.31,11.31,0,0,1,529.19,11V178.7A83.06,83.06,0,0,1,522.52,212,85.76,85.76,0,1,1,410.19,99.62,82.89,82.89,0,0,1,443.53,93m63.24,85.72a61.86,61.86,0,0,0-4.95-24.57,62.65,62.65,0,0,0-33.72-33.72,63.48,63.48,0,0,0-49.15,0,62.74,62.74,0,0,0-33.76,33.75,64,64,0,0,0,0,49.34,63.39,63.39,0,0,0,116.59,0,61.48,61.48,0,0,0,5-24.77"/><path class="cls-1" d="M622.94,264.43a83.06,83.06,0,0,1-33.34-6.67,88.6,88.6,0,0,1-27.24-18.29,84.77,84.77,0,0,1-18.48-27.05,85.76,85.76,0,1,1,158.11,0,84.42,84.42,0,0,1-18.47,27.05,87.91,87.91,0,0,1-27.44,18.29,83.45,83.45,0,0,1-33.14,6.67m0-148.59a62,62,0,0,0-24.58,5,62.72,62.72,0,0,0-33.72,33.72,63.48,63.48,0,0,0,0,49.15,62.78,62.78,0,0,0,33.72,33.72,63.48,63.48,0,0,0,49.15,0,62.72,62.72,0,0,0,33.68-33.76,63.48,63.48,0,0,0,0-49.15,62.65,62.65,0,0,0-33.72-33.72,61.86,61.86,0,0,0-24.57-5"/><path class="cls-2" d="M850.77,254.14a86.3,86.3,0,0,1-19.24,7.62,81.2,81.2,0,0,1-20.76,2.67,83.23,83.23,0,0,1-33.34-6.67A87.46,87.46,0,0,1,731.71,212a85.77,85.77,0,1,1,158.12-66.51,83.44,83.44,0,0,1,6.66,33.14V253a11.31,11.31,0,0,1-3.23,8,10.84,10.84,0,0,1-8.2,3.43,10.38,10.38,0,0,1-7.81-3.43,11.34,11.34,0,0,1-3.24-8V236.23a81.85,81.85,0,0,1-23.24,17.91M747.52,178.7A63.44,63.44,0,0,0,786.19,237a63.45,63.45,0,0,0,49.14,0,62.65,62.65,0,0,0,33.72-33.72,63.48,63.48,0,0,0,0-49.15,62.59,62.59,0,0,0-33.72-33.72,63.45,63.45,0,0,0-49.14,0,63.4,63.4,0,0,0-38.72,58.29"/><path class="cls-2" d="M919.35,78.12a11.07,11.07,0,0,1-11-11V62.93a11.33,11.33,0,0,1,3.24-8,10.31,10.31,0,0,1,7.81-3.43,11.71,11.71,0,0,1,11.43,11.43v4.19a10.31,10.31,0,0,1-3.43,7.81,11.31,11.31,0,0,1-8,3.24m0,186.31a10.68,10.68,0,0,1-7.81-3.24,11.14,11.14,0,0,1-3.24-8.19V104.41a11.12,11.12,0,0,1,3.24-8.19A10.6,10.6,0,0,1,919.4,93a11.22,11.22,0,0,1,8,3.24,10.75,10.75,0,0,1,3.43,8.19V253a10.76,10.76,0,0,1-3.43,8.19,11.31,11.31,0,0,1-8,3.24"/><path class="cls-2" d="M731.81,234.54a19.27,19.27,0,1,1-27.25,0h0a19.26,19.26,0,0,1,27.25,0"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/partners/coiled.svg b/pandas/web/pandas/static/img/partners/coiled.svg
new file mode 100644
index 0000000000000..2d76ce150084b
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/coiled.svg
@@ -0,0 +1,234 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   version="1.1"
+   id="svg18"
+   width="995.65601"
+   height="491.82666"
+   viewBox="0 0 995.65601 491.82666"
+   sodipodi:docname="coiled.svg"
+   inkscape:version="1.2.2 (b0a84865, 2022-12-01)"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <defs
+     id="defs22">
+    <clipPath
+       clipPathUnits="userSpaceOnUse"
+       id="clipPath36">
+      <path
+         d="M 0,368.87 H 746.742 V 0 H 0 Z"
+         id="path34" />
+    </clipPath>
+  </defs>
+  <sodipodi:namedview
+     id="namedview20"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     showgrid="false"
+     inkscape:zoom="0.47984385"
+     inkscape:cx="499.1207"
+     inkscape:cy="245.91333"
+     inkscape:window-width="1309"
+     inkscape:window-height="456"
+     inkscape:window-x="0"
+     inkscape:window-y="37"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="g24" />
+  <g
+     id="g24"
+     inkscape:groupmode="layer"
+     inkscape:label="Coiled Logo - 2022- Black and White"
+     transform="matrix(1.3333333,0,0,-1.3333333,0,491.82667)">
+    <g
+       id="g26"
+       transform="translate(241.374,83.57)">
+      <path
+         d="M 0,0 5.69,16.01 H 5.9 L 11.59,0 H -0.01 Z M 7.4,19.78 H 4.27 L -7.4,-12.32 h 3.06 l 3.42,9.68 h 13.45 l 3.42,-9.68 h 3.06 l -11.6,32.1 z"
+         style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+         id="path28" />
+    </g>
+    <g
+       id="g30">
+      <g
+         id="g32"
+         clip-path="url(#clipPath36)">
+        <g
+           id="g38"
+           transform="translate(297.394,80.14)">
+          <path
+             d="m 0,0 c -0.93,-1.99 -2.35,-3.56 -4.2,-4.63 -1.85,-1.07 -4.13,-1.64 -6.9,-1.64 h -6.33 v 26.83 h 6.83 c 2.63,0 4.77,-0.57 6.55,-1.57 1.78,-1 3.13,-2.63 3.99,-4.63 C 0.87,12.37 1.29,10.02 1.29,7.24 1.36,4.46 0.93,2.04 0.01,-0.02 m 2.47,15.83 c -1.14,2.42 -2.85,4.2 -4.98,5.48 -2.13,1.28 -4.77,1.92 -7.83,1.92 h -9.96 v -32.1 h 9.32 c 3.2,0 5.98,0.64 8.26,1.92 2.28,1.28 3.99,3.13 5.2,5.55 1.21,2.42 1.78,5.34 1.78,8.68 0,3.34 -0.64,6.19 -1.78,8.54"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path40" />
+        </g>
+        <g
+           id="g42"
+           transform="translate(321.8039,80.5201)">
+          <path
+             d="m 0,0 c 0,-1.35 -0.28,-2.49 -0.85,-3.63 -0.57,-1.07 -1.42,-1.99 -2.49,-2.63 -1.07,-0.64 -2.35,-1 -3.84,-1 -1.49,0 -2.92,0.36 -3.91,1.21 -1.07,0.78 -1.57,1.92 -1.57,3.42 0,1 0.28,1.78 0.78,2.42 0.5,0.64 1.28,1.14 2.21,1.49 0.93,0.36 2.13,0.64 3.49,0.78 0.5,0.07 1.07,0.14 1.71,0.21 0.64,0.07 1.28,0.21 1.92,0.28 0.64,0.14 1.21,0.28 1.71,0.43 0.5,0.14 0.85,0.36 1.07,0.57 V -0.01 H 0.02 Z m 1.98,11.09 c -0.57,1.07 -1.21,1.85 -2.06,2.49 -0.85,0.64 -1.78,1 -2.78,1.28 -1,0.21 -1.99,0.36 -2.85,0.36 -1.21,0 -2.42,-0.14 -3.63,-0.5 -1.21,-0.36 -2.28,-0.93 -3.2,-1.71 -0.93,-0.78 -1.71,-1.85 -2.28,-3.13 l 2.63,-0.93 c 0.43,1.07 1.21,1.92 2.28,2.7 1.07,0.78 2.49,1.14 4.27,1.14 1.78,0 3.2,-0.5 4.2,-1.42 1,-0.92 1.49,-2.28 1.49,-3.91 V 6.89 C 0.05,6.25 -0.23,5.82 -0.66,5.54 -1.16,5.26 -1.87,5.04 -2.79,4.97 -3.72,4.9 -4.85,4.69 -6.21,4.54 -7.56,4.4 -8.77,4.11 -9.91,3.83 c -1.14,-0.28 -2.13,-0.71 -2.92,-1.21 -0.85,-0.5 -1.49,-1.21 -1.92,-1.99 -0.43,-0.85 -0.71,-1.85 -0.71,-2.99 0,-1.57 0.36,-2.85 1.07,-3.91 0.71,-1.07 1.64,-1.92 2.85,-2.49 1.21,-0.57 2.49,-0.85 3.99,-0.85 1.35,0 2.49,0.21 3.49,0.64 1,0.43 1.85,1 2.49,1.64 0.64,0.64 1.14,1.35 1.49,2.06 H 0.06 V -9.11 H 2.84 V 7.26 C 2.77,8.68 2.56,10.04 1.99,11.1"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path44" />
+        </g>
+        <g
+           id="g46"
+           transform="translate(347.484,90.04)">
+          <path
+             d="M 0,0 -2.56,-0.71 C -2.77,0 -3.13,0.64 -3.63,1.28 -4.13,1.92 -4.7,2.35 -5.41,2.7 c -0.71,0.35 -1.64,0.5 -2.7,0.5 -1.64,0 -2.99,-0.36 -3.99,-1.14 -1.07,-0.78 -1.57,-1.78 -1.57,-2.99 0,-1.07 0.36,-1.85 1.07,-2.49 0.71,-0.64 1.85,-1.14 3.34,-1.49 l 3.63,-0.85 c 1.99,-0.5 3.56,-1.28 4.55,-2.35 0.99,-1.07 1.49,-2.42 1.49,-3.99 0,-1.35 -0.36,-2.56 -1.14,-3.63 -0.78,-1.07 -1.78,-1.85 -3.13,-2.49 -1.35,-0.64 -2.92,-0.93 -4.63,-0.93 -2.35,0 -4.27,0.5 -5.84,1.57 -1.49,1.07 -2.49,2.56 -2.92,4.63 l 2.7,0.64 c 0.36,-1.42 1,-2.49 2.06,-3.27 1,-0.71 2.35,-1.07 3.99,-1.07 1.85,0 3.34,0.43 4.41,1.28 1.07,0.85 1.64,1.85 1.64,3.13 0,1 -0.36,1.78 -1,2.49 -0.64,0.64 -1.64,1.14 -2.92,1.49 l -3.91,0.93 c -2.06,0.5 -3.63,1.28 -4.63,2.42 -1,1.14 -1.49,2.42 -1.49,4.06 0,1.35 0.36,2.49 1.07,3.49 0.71,1 1.71,1.78 2.99,2.35 1.28,0.57 2.7,0.85 4.34,0.85 2.21,0 3.99,-0.5 5.27,-1.49 1.28,-1 2.13,-2.56 2.7,-4.34"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path48" />
+        </g>
+        <g
+           id="g50"
+           transform="translate(371.824,71.26)">
+          <path
+             d="m 0,0 h -3.56 l -9.39,11.95 -2.78,-2.63 V 0 h -2.77 v 32.09 h 2.77 V 12.73 h 0.36 l 11.24,11.32 h 3.56 L -10.89,13.73 Z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path52" />
+        </g>
+        <g
+           id="g54"
+           transform="translate(413.244,93.32)">
+          <path
+             d="m 0,0 h -2.92 c -0.21,1.07 -0.64,2.13 -1.21,3.06 -0.57,0.93 -1.28,1.78 -2.13,2.49 -0.85,0.71 -1.85,1.28 -2.92,1.64 -1.07,0.36 -2.28,0.57 -3.56,0.57 -1.99,0 -3.84,-0.5 -5.55,-1.57 -1.71,-1.07 -2.99,-2.63 -3.99,-4.63 -1,-2.06 -1.49,-4.55 -1.49,-7.47 0,-2.92 0.5,-5.48 1.49,-7.54 1,-2.06 2.35,-3.56 3.99,-4.63 1.64,-1.07 3.49,-1.57 5.55,-1.57 1.28,0 2.42,0.21 3.56,0.57 1.07,0.36 2.06,0.93 2.92,1.64 0.86,0.71 1.57,1.49 2.13,2.49 0.57,0.93 0.93,1.99 1.21,3.06 H 0 c -0.28,-1.49 -0.71,-2.85 -1.49,-4.13 -0.78,-1.28 -1.64,-2.35 -2.7,-3.34 -1.07,-1 -2.35,-1.71 -3.77,-2.21 -1.42,-0.5 -2.99,-0.78 -4.7,-0.78 -2.7,0 -5.12,0.71 -7.19,2.06 -2.06,1.35 -3.7,3.27 -4.91,5.69 -1.21,2.42 -1.78,5.34 -1.78,8.75 0,3.41 0.57,6.26 1.78,8.75 1.21,2.49 2.85,4.34 4.91,5.69 2.06,1.35 4.48,2.06 7.19,2.06 1.71,0 3.27,-0.28 4.7,-0.78 1.42,-0.5 2.7,-1.28 3.77,-2.21 C -3.12,6.73 -2.2,5.6 -1.49,4.32 -0.78,3.04 -0.28,1.47 0,-0.02"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path56" />
+        </g>
+        <g
+           id="g58"
+           transform="translate(435.164,78.23)">
+          <path
+             d="m 0,0 c -0.57,-1.49 -1.49,-2.7 -2.63,-3.56 -1.14,-0.93 -2.56,-1.35 -4.27,-1.35 -1.71,0 -3.06,0.43 -4.27,1.35 -1.14,0.93 -2.06,2.06 -2.63,3.56 -0.57,1.49 -0.93,3.13 -0.93,4.91 0,1.78 0.28,3.49 0.93,4.98 0.57,1.49 1.49,2.7 2.63,3.63 1.14,0.93 2.56,1.35 4.27,1.35 1.71,0 3.06,-0.43 4.27,-1.35 C -1.49,12.59 -0.57,11.39 0,9.89 0.64,8.4 0.93,6.76 0.93,4.91 0.93,3.06 0.57,1.49 0,0 m 2.35,11.46 c -0.93,1.85 -2.13,3.34 -3.7,4.41 -1.57,1.07 -3.42,1.57 -5.48,1.57 -2.06,0 -3.91,-0.5 -5.48,-1.57 -1.57,-1.07 -2.85,-2.49 -3.7,-4.41 -0.85,-1.92 -1.35,-4.06 -1.35,-6.55 0,-2.49 0.43,-4.63 1.35,-6.48 0.93,-1.85 2.13,-3.34 3.7,-4.41 1.57,-1.07 3.42,-1.57 5.48,-1.57 2.06,0 3.91,0.5 5.48,1.57 1.57,1.07 2.85,2.49 3.7,4.41 0.93,1.85 1.35,3.99 1.35,6.48 0,2.49 -0.43,4.7 -1.35,6.55"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path60" />
+        </g>
+        <g
+           id="g62"
+           transform="translate(444.344,71.26)">
+          <path
+             d="m 0,0 v 24.05 h 2.7 v -3.7 h 0.21 c 0.5,1.28 1.28,2.21 2.42,2.92 1.14,0.71 2.49,1.07 4.06,1.07 1.57,0 2.99,-0.36 4.13,-1.14 1.07,-0.78 1.85,-1.85 2.42,-3.2 h 0.21 c 0.57,1.35 1.42,2.42 2.63,3.2 1.21,0.78 2.7,1.21 4.48,1.21 2.21,0 3.99,-0.71 5.27,-2.13 1.28,-1.42 1.92,-3.49 1.92,-6.12 V 0.01 h -2.78 v 16.15 c 0,1.92 -0.5,3.34 -1.42,4.27 -0.93,0.93 -2.28,1.42 -3.84,1.42 -1.85,0 -3.27,-0.57 -4.27,-1.71 -1,-1.14 -1.49,-2.56 -1.49,-4.27 V 0 H 13.8 v 16.37 c 0,1.64 -0.43,2.92 -1.35,3.91 -0.92,0.99 -2.21,1.49 -3.84,1.49 -1.14,0 -2.13,-0.28 -2.99,-0.78 -0.85,-0.5 -1.57,-1.28 -2.06,-2.28 -0.5,-1 -0.78,-2.06 -0.78,-3.34 l 0,-15.44 z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path64" />
+        </g>
+        <g
+           id="g66"
+           transform="translate(498.2141,78.24)">
+          <path
+             d="m 0,0 c -0.57,-1.49 -1.42,-2.7 -2.49,-3.56 -1.14,-0.85 -2.49,-1.35 -4.06,-1.35 -1.57,0 -2.85,0.43 -3.99,1.28 -1.07,0.85 -1.92,1.99 -2.49,3.56 -0.57,1.49 -0.85,3.2 -0.85,5.12 0,1.92 0.28,3.63 0.85,5.12 0.57,1.49 1.42,2.63 2.49,3.49 1.07,0.86 2.42,1.28 3.99,1.28 1.57,0 2.99,-0.43 4.06,-1.28 C -1.42,12.81 -0.57,11.6 0,10.1 0.57,8.61 0.85,6.97 0.85,5.12 0.85,3.27 0.57,1.49 0,0 m 2.42,11.52 c -0.85,1.85 -1.99,3.34 -3.49,4.34 -1.49,1.07 -3.2,1.57 -5.2,1.57 -1.49,0 -2.7,-0.28 -3.63,-0.78 -1,-0.5 -1.78,-1.14 -2.35,-1.92 -0.57,-0.78 -1.07,-1.49 -1.35,-2.21 h -0.36 v 4.55 h -2.7 v -33.09 h 2.78 V -2.5 h 0.28 c 0.36,-0.71 0.78,-1.49 1.35,-2.28 0.57,-0.78 1.42,-1.42 2.35,-1.92 0.93,-0.5 2.21,-0.78 3.63,-0.78 1.99,0 3.7,0.5 5.2,1.57 1.5,1.07 2.63,2.49 3.49,4.34 0.85,1.85 1.28,4.06 1.28,6.55 0,2.49 -0.5,4.7 -1.28,6.55"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path68" />
+        </g>
+        <g
+           id="g70"
+           transform="translate(521.7141,80.5201)">
+          <path
+             d="m 0,0 c 0,-1.35 -0.28,-2.49 -0.85,-3.63 -0.57,-1.07 -1.42,-1.99 -2.49,-2.63 -1.07,-0.64 -2.35,-1 -3.84,-1 -1.49,0 -2.92,0.36 -3.91,1.21 -1.07,0.78 -1.57,1.92 -1.57,3.42 0,1 0.28,1.78 0.78,2.42 0.5,0.64 1.28,1.14 2.21,1.49 0.93,0.36 2.13,0.64 3.49,0.78 0.5,0.07 1.07,0.14 1.71,0.21 0.64,0.07 1.28,0.21 1.85,0.28 0.64,0.14 1.21,0.28 1.71,0.43 0.5,0.15 0.85,0.36 1.07,0.57 V -0.01 H 0.02 Z m 1.98,11.09 c -0.57,1.07 -1.21,1.85 -2.06,2.49 -0.85,0.64 -1.78,1 -2.78,1.28 -1,0.21 -1.99,0.36 -2.85,0.36 -1.21,0 -2.42,-0.14 -3.63,-0.5 -1.21,-0.36 -2.28,-0.93 -3.2,-1.71 -0.93,-0.78 -1.71,-1.85 -2.28,-3.13 l 2.63,-0.93 c 0.43,1.07 1.21,1.92 2.28,2.7 1.07,0.78 2.49,1.14 4.27,1.14 1.78,0 3.2,-0.5 4.2,-1.42 1,-0.92 1.49,-2.28 1.49,-3.91 V 6.89 C 0.05,6.25 -0.23,5.82 -0.66,5.54 -1.16,5.26 -1.87,5.04 -2.79,4.97 -3.72,4.83 -4.85,4.69 -6.21,4.54 -7.56,4.4 -8.77,4.11 -9.91,3.83 c -1.14,-0.28 -2.13,-0.71 -2.92,-1.21 -0.85,-0.5 -1.49,-1.21 -1.92,-1.99 -0.43,-0.85 -0.71,-1.85 -0.71,-2.99 0,-1.57 0.36,-2.85 1.07,-3.91 0.71,-1.07 1.64,-1.92 2.85,-2.49 1.21,-0.57 2.49,-0.85 3.99,-0.85 1.35,0 2.49,0.21 3.49,0.64 1,0.43 1.85,1 2.49,1.64 0.64,0.64 1.14,1.35 1.49,2.06 H 0.06 V -9.11 H 2.84 V 7.26 C 2.77,8.68 2.48,10.04 1.99,11.1"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path72" />
+        </g>
+        <g
+           id="g74"
+           transform="translate(534.014,86.27)">
+          <path
+             d="M 0,0 V -15.02 H -2.78 V 9.03 h 2.7 V 5.26 h 0.21 c 0.57,1.21 1.42,2.21 2.63,2.99 1.21,0.71 2.7,1.14 4.41,1.14 1.71,0 3.06,-0.36 4.27,-1 1.21,-0.71 2.21,-1.64 2.85,-2.99 0.64,-1.35 1,-2.92 1,-4.84 V -15.03 H 12.51 V 0.41 c 0,1.99 -0.57,3.56 -1.64,4.7 C 9.8,6.25 8.31,6.82 6.46,6.82 5.18,6.82 4.11,6.54 3.12,6.04 2.12,5.54 1.34,4.69 0.77,3.69 0.27,2.62 -0.01,1.41 -0.01,-0.01"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path76" />
+        </g>
+        <g
+           id="g78"
+           transform="translate(557.214,62.22)">
+          <path
+             d="m 0,0 c -0.57,0 -1.07,0.07 -1.57,0.14 -0.5,0.07 -0.85,0.21 -1.14,0.36 l 0.78,2.49 C -1,2.71 -0.15,2.56 0.63,2.63 c 0.71,0.07 1.42,0.43 1.99,1 0.57,0.57 1.07,1.49 1.57,2.7 l 0.85,2.42 -8.97,24.34 h 2.99 L 6.18,12.74 h 0.21 l 7.12,20.35 H 16.5 L 6.32,5.19 C 5.89,4.05 5.39,3.06 4.75,2.27 4.11,1.49 3.4,0.92 2.62,0.56 1.91,0.2 0.98,-0.01 -0.01,-0.01"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path80" />
+        </g>
+        <g
+           id="g82"
+           transform="translate(85.574,158.35)">
+          <path
+             d="m 0,0 c -46.81,-0.45 -85.12,37.13 -85.57,83.94 -0.45,46.81 37.13,85.12 83.94,85.57 H 0 Z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path84" />
+        </g>
+        <g
+           id="g86"
+           transform="translate(155.384,134.8)">
+          <path
+             d="M 0,0 C 0,34.93 -21.42,66.28 -53.97,78.97 V -78.96 C -21.43,-66.27 0,-34.93 0,0"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path88" />
+        </g>
+        <g
+           id="g90"
+           transform="translate(101.534,258.03)">
+          <path
+             d="m 0,0 c 30.61,0 55.42,24.8 55.43,55.41 0,30.61 -24.8,55.42 -55.41,55.43 H 0 V 0.01 Z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path92" />
+        </g>
+        <g
+           id="g94"
+           transform="translate(46.304,64.28)">
+          <path
+             d="M 0,0 C -0.02,-27.08 15.11,-51.9 39.2,-64.28 V 64.26 C 15.13,51.88 0,27.08 0,0.01 Z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path96" />
+        </g>
+        <g
+           id="g98"
+           transform="translate(454.4311,181.9546)">
+          <path
+             d="m 0,0 c 0,22.36 -13.83,36.92 -35.1,36.92 -21.27,0 -35.1,-14.55 -35.1,-36.92 0,-22.37 13.83,-36.92 35.1,-36.92 C -13.83,-36.92 0,-22.38 0,0 m -84.02,0 c 0,30 19.46,49.64 48.92,49.64 C -5.64,49.64 14,30 14,0 c 0,-30 -19.46,-49.47 -49.1,-49.47 -29.46,0 -48.92,19.63 -48.92,49.47 z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path100" />
+        </g>
+        <path
+           d="m 483.231,227.965 h 13.83 v -91.84 h -13.83 z"
+           style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+           id="path102" />
+        <path
+           d="m 516.842,269.795 h 13.83 v -133.69 h -13.83 z"
+           style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+           id="path104" />
+        <g
+           id="g106"
+           transform="translate(733.2923,181.9546)">
+          <path
+             d="m 0,0 c 0,20.91 -14.55,36.92 -34.37,36.92 -19.82,0 -34.56,-16.01 -34.56,-36.92 0,-20.91 14.74,-36.92 34.56,-36.92 C -14.55,-36.92 0,-20.74 0,0 m -82.75,0 c 0,28.73 19.64,49.64 46.56,49.64 14.35,0.27 27.79,-6.97 35.46,-19.09 h 0.18 v 57.29 h 14 V -45.85 H -0.19 L 0,-30.03 h -0.19 c -7.72,-12.3 -21.31,-19.67 -35.83,-19.43 -26.91,0 -46.73,20.91 -46.73,49.47 z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path108" />
+        </g>
+        <g
+           id="g110"
+           transform="translate(355.9671,226.3063)">
+          <path
+             d="m 0,0 h -17.807 c -1.054,4.71 -2.884,8.848 -5.492,12.414 -2.611,3.566 -5.77,6.57 -9.478,9.013 -3.71,2.444 -7.815,4.281 -12.314,5.514 -4.5,1.231 -8.18,1.848 -13.063,1.848 -8.903,0 -16.957,-2.069 -24.16,-6.206 -7.205,-4.138 -12.925,-10.235 -17.161,-18.291 -4.236,-8.056 -6.354,-17.939 -6.354,-29.648 0,-11.71 2.118,-21.592 6.354,-29.648 4.236,-8.057 9.956,-14.154 17.161,-18.291 7.203,-4.138 15.257,-6.207 24.16,-6.207 4.883,0 9.574,0.616 14.074,1.849 4.499,1.231 8.604,3.07 12.314,5.513 3.708,2.444 6.867,5.458 9.478,9.046 2.608,3.588 4.438,7.715 5.492,12.381 H 1.011 c -1.341,-6.911 -3.782,-13.096 -7.324,-18.555 -3.543,-5.458 -7.947,-10.091 -13.211,-13.899 -5.266,-3.808 -11.166,-6.702 -17.7,-8.683 -6.534,-1.981 -13.51,-2.972 -20.93,-2.972 -12.541,0 -23.694,2.817 -33.458,8.453 -9.766,5.634 -17.448,13.645 -23.049,24.035 -5.6,10.388 -8.4,22.715 -8.4,36.978 0,14.263 2.8,26.588 8.4,36.978 5.601,10.388 13.283,18.399 23.049,24.035 9.764,5.635 20.917,8.452 33.458,8.452 7.42,0 13.385,-0.99 19.919,-2.972 6.534,-1.981 12.433,-4.875 17.7,-8.683 5.264,-3.807 9.668,-8.43 13.211,-13.866 C -3.782,13.15 -1.341,6.955 0,0"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path112" />
+        </g>
+        <g
+           id="g114"
+           transform="translate(562.7174,190.0971)">
+          <path
+             d="m 0,0 c 0.3,4.47 1.53,8.69 3.67,12.66 2.5,4.64 6.07,8.41 10.73,11.32 4.66,2.9 10.19,4.35 16.59,4.35 5.82,0 10.87,-1.23 15.17,-3.7 4.29,-2.46 7.61,-5.82 9.95,-10.09 2.34,-4.26 3.51,-9.1 3.51,-14.54 z m 75.97,-6.51 c 0,8.98 -1.29,16.54 -3.88,22.68 -2.59,6.14 -6.07,11.08 -10.43,14.82 -4.36,3.74 -9.2,6.45 -14.51,8.12 -5.32,1.67 -10.7,2.51 -16.16,2.51 -9.44,0 -17.7,-2.12 -24.77,-6.36 C -0.84,31.01 -6.34,25.12 -10.271,17.58 -14.2,10.04 -16.16,1.34 -16.16,-8.52 c 0,-9.86 1.96,-18.5 5.889,-25.91 3.931,-7.42 9.541,-13.18 16.831,-17.3 7.3,-4.11 15.98,-6.17 26.07,-6.17 7.09,0 13.47,1.01 19.15,3.04 5.68,2.02 10.449,4.86 14.31,8.52 3.86,3.66 6.61,7.91 8.24,12.75 l 0.99,4.06 -16.52,-0.05 c -1.37,-3.34 -3.3,-6.13 -5.79,-8.36 -2.5,-2.24 -5.47,-3.91 -8.9,-5.02 -3.43,-1.1 -7.25,-1.66 -11.48,-1.66 -6.5,0 -12.2,1.31 -17.1,3.92 -4.91,2.61 -8.74,6.43 -11.49,11.47 -2.49,4.56 -3.85,10.05 -4.08,16.45 h 76.01 z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path116" />
+        </g>
+        <g
+           id="g118"
+           transform="translate(481.7372,256.8414)">
+          <path
+             d="m 0,0 c 0,4.88 3.274,8.133 8.331,8.133 5.058,0 8.487,-3.253 8.487,-8.133 0,-4.88 -3.568,-8.28 -8.487,-8.28 C 3.413,-8.28 0,-5.026 0,0 Z"
+             style="fill:#000000;fill-opacity:1;fill-rule:nonzero;stroke:none"
+             id="path120" />
+        </g>
+      </g>
+    </g>
+  </g>
+</svg>
diff --git a/pandas/web/pandas/static/img/partners/czi.svg b/pandas/web/pandas/static/img/partners/czi.svg
new file mode 100644
index 0000000000000..b0ad9eb80580b
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/czi.svg
@@ -0,0 +1,38 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="112px" height="62px" viewBox="0 0 112 62" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 52.2 (67145) - http://www.bohemiancoding.com/sketch -->
+    <title>Group</title>
+    <desc>Created with Sketch.</desc>
+    <g id="Symbols" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="nav-elements-/-czi_logo_nav_dark" transform="translate(-48.000000, -17.000000)">
+            <g id="Group" transform="translate(48.000000, 17.000000)">
+                <path d="M13.118,16.05056 C11.6721111,16.69381 10.2262222,17.047985 8.20166667,17.047985 C3.64077778,17.047985 0.370222222,14.43236 0.370222222,9.79941 C0.370222222,5.565585 3.64077778,2.26176 8.15733333,2.26176 C9.38077778,2.26176 10.6928889,2.550835 11.627,2.949185 L12.3837778,2.195885 L13.0293333,2.195885 L13.2961111,7.62631 L12.4724444,7.67126 C11.0047778,4.545685 9.73622222,3.34831 7.93411111,3.34831 C5.70966667,3.34831 4.04133333,5.410585 4.04133333,9.466935 C4.04133333,13.678285 6.11022222,15.851385 8.869,15.851385 C9.44688889,15.851385 9.80311111,15.784735 10.3366667,15.60726 L12.9624444,12.260035 L13.5628889,12.348385 L13.118,16.05056 Z" id="Fill-1" fill="#231F20"/>
+                <path d="M21.1942889,8.26987 C20.3931778,8.26987 19.9039556,8.469045 19.1254,8.890645 L19.1254,15.585095 L20.3262889,16.07257 L20.3262889,16.715045 L14.7869556,16.715045 L14.7869556,16.07257 L16.2110667,15.585095 L16.2110667,2.83867 L14.6757333,1.907895 L14.6757333,1.331295 L18.7022889,0.289695 L19.1254,0.533045 L19.1254,7.959095 L19.3696222,8.00327 C20.3931778,7.09497 21.5948444,6.36337 22.9746222,6.36337 C24.8428444,6.36337 25.8220667,7.582445 25.8220667,9.39982 L25.8220667,15.585095 L27.2454,16.07257 L27.2454,16.715045 L21.7060667,16.715045 L21.7060667,16.07257 L22.9077333,15.585095 L22.9077333,10.375545 C22.9077333,9.08982 22.4402889,8.26987 21.1942889,8.26987" id="Fill-3" fill="#231F20"/>
+                <path d="M33.6751333,11.7944925 C31.2943556,11.8828425 30.4271333,12.7035675 30.4271333,13.8118175 C30.4271333,14.8309425 31.0058,15.3858425 32.0285778,15.3858425 C32.4742444,15.3858425 33.0746889,15.1858925 33.6751333,14.8092425 L33.6751333,11.7944925 Z M36.4339111,14.6092925 C36.4339111,15.3633675 36.7675778,15.8291425 38.0579111,15.8508425 L38.1022444,16.4274425 C37.4574667,16.6707925 36.4782444,16.9373925 35.6110222,16.9373925 C34.5874667,16.9373925 34.0088,16.5157925 33.8306889,15.4741925 L33.5639111,15.4300175 C32.5629111,16.3607925 31.6062444,16.9149175 30.4271333,16.9149175 C28.8256889,16.9149175 27.7126889,16.0507925 27.7126889,14.5209425 C27.7126889,12.6369175 29.6711333,11.3511925 33.6751333,11.0628925 L33.6751333,9.8438175 C33.6751333,8.1806675 33.0078,7.6265425 31.3394667,7.6265425 C30.8945778,7.6265425 30.4940222,7.7156675 30.2046889,7.8264925 C30.6052444,8.1589675 30.8276889,8.5356175 30.8276889,9.0230925 C30.8276889,9.7322175 30.3376889,10.2871175 29.4930222,10.2871175 C28.6919111,10.2871175 28.0914667,9.6663425 28.0914667,8.9347425 C28.0914667,7.4048925 30.5383556,6.3632925 32.8740222,6.3632925 C35.2991333,6.3632925 36.4339111,7.2723675 36.4339111,9.4446925 L36.4339111,14.6092925 Z" id="Fill-5" fill="#231F20"/>
+                <path d="M49.1152667,15.585095 L50.5168222,16.07257 L50.5168222,16.715045 L44.9992667,16.715045 L44.9992667,16.07257 L46.2009333,15.585095 L46.2009333,10.375545 C46.2009333,9.067345 45.7334889,8.26987 44.5100444,8.26987 C43.7532667,8.26987 43.2640444,8.49152 42.5080444,8.890645 L42.5080444,15.585095 L43.6871556,16.07257 L43.6871556,16.715045 L38.1696,16.715045 L38.1696,16.07257 L39.5937111,15.585095 L39.5937111,8.690695 L38.1921556,7.75992 L38.1921556,7.18332 L41.9962667,6.29672 L42.4186,6.562545 L42.2630444,8.15827 L42.5298222,8.20322 C43.8870444,6.96167 45.0443778,6.36337 46.2678222,6.36337 C48.0917111,6.36337 49.1152667,7.515795 49.1152667,9.400595 L49.1152667,15.585095 Z" id="Fill-7" fill="#231F20"/>
+                <path d="M12.5395667,38.507425 L0.148011111,38.507425 L0.0593444444,37.9533 L8.9579,25.583525 L5.55434444,25.6277 C4.33012222,25.6277 3.77401111,26.026825 2.81734444,27.268375 L1.39323333,29.041575 L0.8379,28.90905 L1.26023333,24.386925 L12.8957889,24.386925 L12.9844556,24.94105 L4.15201111,37.332525 L8.37923333,37.28835 C9.53656667,37.28835 10.0926778,36.822575 10.8261222,35.7585 L12.4726778,33.364525 L12.9844556,33.45365 L12.5395667,38.507425 Z" id="Fill-9" fill="#231F20"/>
+                <path d="M25.5768333,37.4432725 L25.5768333,37.9973975 L21.7501667,38.8839975 L21.3278333,38.6398725 L21.4833889,37.0441475 L21.2166111,36.9999725 C19.8593889,38.2415225 18.7689444,38.8398225 17.4786111,38.8398225 C15.6103889,38.8398225 14.6311667,37.5757975 14.6311667,35.7801225 L14.6311667,30.4163475 L13.2521667,29.5072725 L13.2521667,28.9756225 L17.1449444,28.0882475 L17.5455,28.3540725 L17.5455,34.7826975 C17.5455,36.1575475 18.0347222,36.9108475 19.2363889,36.9108475 C19.9923889,36.9108475 20.4823889,36.7116725 21.2383889,36.3125475 L21.2383889,30.4163475 L19.8368333,29.5072725 L19.8368333,28.9756225 L23.7303889,28.0882475 L24.1527222,28.3540725 L24.1527222,36.6233225 L25.5768333,37.4432725 Z" id="Fill-11" fill="#231F20"/>
+                <path d="M34.0978556,32.12313 C33.0968556,32.12313 32.5181889,31.39153 32.5181889,30.394105 C32.5181889,29.972505 32.6068556,29.57338 32.7631889,29.219205 C32.6294111,29.19673 32.5181889,29.17503 32.3626333,29.17503 C30.7160778,29.240905 29.2701889,30.65993 29.2701889,33.342205 C29.2701889,35.80283 30.7829667,37.088555 32.8075222,37.088555 C33.7198556,37.088555 34.3428556,36.888605 35.3881889,36.490255 L35.6106333,37.021905 C34.3646333,38.28593 32.9630778,38.861755 31.4277444,38.861755 C28.5585222,38.861755 26.2671889,36.800255 26.2671889,33.74133 C26.2671889,30.637455 28.6915222,28.15513 32.0071889,28.15513 C34.2090778,28.15513 35.5888556,29.219205 35.5888556,30.52663 C35.5888556,31.435705 34.9658556,32.12313 34.0978556,32.12313" id="Fill-13" fill="#231F20"/>
+                <polygon id="Fill-15" fill="#231F20" points="46.8457111 37.39902 47.9136 37.864795 47.9136 38.50727 42.9303778 38.50727 42.9303778 37.931445 43.7532667 37.509845 41.1726 34.00762 40.3940444 34.00762 40.3940444 37.376545 41.5062667 37.864795 41.5062667 38.50727 36.0556 38.50727 36.0556 37.864795 37.4797111 37.376545 37.4797111 24.63012 35.9443778 23.699345 35.9443778 23.122745 39.9717111 22.081145 40.3940444 22.324495 40.3940444 32.831945 41.2394889 32.831945 43.9539333 29.795495 42.7297111 29.04142 42.7297111 28.487295 47.4461556 28.487295 47.4461556 29.152245 46.0671556 29.68467 42.9972667 32.41112"/>
+                <path d="M54.8996,32.38911 C54.8778222,30.12766 54.2991556,29.063585 53.0531556,29.063585 C51.8297111,29.063585 50.9617111,30.371785 50.8730444,32.588285 L54.8996,32.38911 Z M50.8504889,33.386535 C50.8730444,35.802985 51.9183778,37.15536 54.1653778,37.15536 C55.1889333,37.15536 56.0787111,36.95541 57.3916,36.467935 L57.6366,36.977885 C56.2568222,38.107835 54.6102667,38.86191 53.1418222,38.86191 C49.9607111,38.86191 47.9361556,36.77871 47.9361556,33.696535 C47.9361556,30.50431 50.4499333,28.155285 53.2312667,28.155285 C56.5687111,28.155285 57.8582667,30.57096 57.7470444,33.386535 L50.8504889,33.386535 Z" id="Fill-17" fill="#231F20"/>
+                <path d="M66.0904222,31.657665 C65.2224222,31.657665 64.6888667,31.014415 64.5776444,30.261115 C63.8659778,30.416115 63.3090889,30.88189 62.8198667,31.524365 L62.8198667,37.37639 L64.3995333,37.86464 L64.3995333,38.507115 L58.4822,38.507115 L58.4822,37.86464 L59.9055333,37.37639 L59.9055333,30.482765 L58.5039778,29.55199 L58.5039778,28.97539 L62.3088667,28.08879 L62.7312,28.354615 L62.4862,30.482765 L62.7529778,30.549415 C63.6653111,29.04204 64.5776444,28.132965 65.8905333,28.132965 C66.9358667,28.132965 67.6926444,28.82039 67.6926444,29.88369 C67.6926444,30.90359 67.0027556,31.657665 66.0904222,31.657665" id="Fill-19" fill="#231F20"/>
+                <path d="M73.4766667,30.0172225 C72.9205556,30.0172225 72.4087778,30.1722225 71.9413333,30.3496975 L71.9413333,36.7783225 C72.4974444,37.4874475 73.1204444,37.8423975 73.9884444,37.8423975 C75.1675556,37.8423975 76.2128889,36.7116725 76.2128889,33.8960975 C76.2128889,31.2587725 75.2118889,30.0172225 73.4766667,30.0172225 M73.4097778,38.8622975 C72.1194444,38.8622975 70.3624444,38.5739975 69.027,38.1531725 L69.027,24.6301975 L67.4698889,23.6994225 L67.4698889,23.1228225 L71.519,22.0812225 L71.9413333,22.3245725 L71.9413333,29.4630975 L72.2088889,29.5072725 C73.1204444,28.6873225 74.0327778,28.1548975 75.3005556,28.1548975 C77.392,28.1548975 79.2827778,30.0830975 79.2827778,33.2094475 C79.2827778,36.1800225 77.0583333,38.8622975 73.4097778,38.8622975" id="Fill-21" fill="#231F20"/>
+                <path d="M87.4482,32.38911 C87.4256444,30.12766 86.8477556,29.063585 85.6009778,29.063585 C84.3783111,29.063585 83.5103111,30.371785 83.4216444,32.588285 L87.4482,32.38911 Z M83.3990889,33.386535 C83.4216444,35.802985 84.4669778,37.15536 86.7139778,37.15536 C87.7375333,37.15536 88.6273111,36.95541 89.9394222,36.467935 L90.1844222,36.977885 C88.8054222,38.107835 87.1588667,38.86191 85.6904222,38.86191 C82.5093111,38.86191 80.4847556,36.77871 80.4847556,33.696535 C80.4847556,30.50431 82.9985333,28.155285 85.7790889,28.155285 C89.1165333,28.155285 90.4068667,30.57096 90.2956444,33.386535 L83.3990889,33.386535 Z" id="Fill-23" fill="#231F20"/>
+                <path d="M98.6386333,31.657665 C97.7714111,31.657665 97.2370778,31.014415 97.1258556,30.261115 C96.4141889,30.416115 95.8580778,30.88189 95.3680778,31.524365 L95.3680778,37.37639 L96.9477444,37.86464 L96.9477444,38.507115 L91.0304111,38.507115 L91.0304111,37.86464 L92.4537444,37.37639 L92.4537444,30.482765 L91.0529667,29.55199 L91.0529667,28.97539 L94.8570778,28.08879 L95.2794111,28.354615 L95.0344111,30.482765 L95.3011889,30.549415 C96.2135222,29.04204 97.1258556,28.132965 98.4387444,28.132965 C99.4840778,28.132965 100.240856,28.82039 100.240856,29.88369 C100.240856,30.90359 99.5509667,31.657665 98.6386333,31.657665" id="Fill-25" fill="#231F20"/>
+                <path d="M105.268333,28.93075 C104.267333,28.93075 103.555667,29.574 103.555667,31.5022 C103.555667,33.519525 104.245556,34.07365 105.379556,34.07365 C106.380556,34.07365 107.026111,33.164575 107.026111,31.41385 C107.026111,29.773175 106.403111,28.93075 105.268333,28.93075 M106.580444,38.8399 L102.999556,38.795725 C102.710222,39.172375 102.509556,39.65985 102.509556,40.258925 C102.509556,41.41135 103.466222,42.2096 105.757556,42.2096 C107.604,42.2096 109.050667,41.300525 109.050667,40.23645 C109.050667,39.2615 108.293889,38.8616 106.580444,38.8399 M105.335222,34.849425 C104.578444,34.849425 104.111778,34.782775 103.711222,34.65025 C103.488778,34.937775 103.288889,35.337675 103.288889,35.6252 C103.288889,36.1801 103.688667,36.445925 104.756556,36.467625 L107.492778,36.5118 C109.806667,36.55675 111.008333,37.44335 111.008333,39.105725 C111.008333,41.27805 108.627556,43.1179 104.979,43.1179 C101.642333,43.1179 100.107,42.14295 100.107,40.857225 C100.107,40.037275 100.796889,39.2832 102.265333,38.729075 L102.242778,38.507425 C101.353,38.219125 100.952444,37.665 100.952444,37.044225 C100.952444,36.312625 101.575444,35.51515 102.887556,34.49525 L102.887556,34.36195 C101.709222,33.78535 100.885556,32.76545 100.885556,31.5022 C100.885556,29.4407 102.843222,28.154975 105.379556,28.154975 C106.870556,28.154975 107.804667,28.5541 108.583222,29.108225 L110.295889,28.154975 L111.564444,28.154975 L111.564444,30.083175 L111.119556,30.083175 L109.273111,29.729 L109.228,29.906475 C109.495556,30.371475 109.695444,30.859725 109.695444,31.56885 C109.695444,33.452875 108.027111,34.849425 105.335222,34.849425" id="Fill-27" fill="#231F20"/>
+                <polygon id="Fill-29" fill="#231F20" points="7.8232 59.5675075 7.8232 60.2991075 0.236755556 60.2991075 0.236755556 59.5675075 2.37253333 59.0800325 2.37253333 47.3976825 0.236755556 46.9094325 0.236755556 46.1786075 7.8232 46.1786075 7.8232 46.9094325 5.70997778 47.3976825 5.70997778 59.0800325"/>
+                <path d="M19.7256889,59.168615 L21.1272444,59.65609 L21.1272444,60.298565 L15.6096889,60.298565 L15.6096889,59.65609 L16.8113556,59.168615 L16.8113556,53.959065 C16.8113556,52.650865 16.3439111,51.85339 15.1204667,51.85339 C14.3636889,51.85339 13.8752444,52.07504 13.1184667,52.47339 L13.1184667,59.168615 L14.2975778,59.65609 L14.2975778,60.298565 L8.78002222,60.298565 L8.78002222,59.65609 L10.2041333,59.168615 L10.2041333,52.274215 L8.80257778,51.34344 L8.80257778,50.76684 L12.6066889,49.88024 L13.0298,50.146065 L12.8734667,51.74179 L13.1410222,51.78674 C14.4974667,50.54519 15.6548,49.94689 16.8782444,49.94689 C18.7021333,49.94689 19.7256889,51.099315 19.7256889,52.984115 L19.7256889,59.168615 Z" id="Fill-31" fill="#231F20"/>
+                <path d="M24.3088222,44.1832925 C25.3767111,44.1832925 26.1334889,44.9590675 26.1334889,45.9789675 C26.1334889,46.9763925 25.3323778,47.8846925 24.2201556,47.8846925 C23.1966,47.8846925 22.4180444,47.1313925 22.4180444,46.1114925 C22.4180444,45.1140675 23.2191556,44.1832925 24.3088222,44.1832925 Z M21.4388222,51.3435175 L21.4388222,50.7669175 L25.3767111,49.8803175 L25.7547111,50.1019675 L25.7547111,59.1686925 L27.1344889,59.6561675 L27.1344889,60.2986425 L21.3727111,60.2986425 L21.3727111,59.6561675 L22.8403778,59.1686925 L22.8403778,52.2742925 L21.4388222,51.3435175 Z" id="Fill-33" fill="#231F20"/>
+                <path d="M31.5837667,60.58733 C29.8259889,60.58733 28.9362111,59.744905 28.9362111,57.94923 L28.9362111,51.720555 L27.2235444,51.720555 L27.2235444,51.12148 L29.3811,49.63658 L31.0276556,47.13178 L31.8505444,47.13178 L31.8505444,50.23488 L35.2323222,50.23488 L35.2323222,51.720555 L31.8505444,51.720555 L31.8505444,56.906855 C31.8505444,57.838405 31.9843222,58.326655 32.4292111,58.636655 C32.8297667,58.90248 33.2521,58.96913 33.9863222,58.96913 C34.3425444,58.96913 34.8092111,58.946655 35.2766556,58.90248 L35.2992111,59.589905 C34.4755444,60.099855 32.8958778,60.58733 31.5837667,60.58733" id="Fill-35" fill="#231F20"/>
+                <path d="M38.4804778,44.1832925 C39.5483667,44.1832925 40.3051444,44.9590675 40.3051444,45.9789675 C40.3051444,46.9763925 39.5040333,47.8846925 38.3918111,47.8846925 C37.3682556,47.8846925 36.5897,47.1313925 36.5897,46.1114925 C36.5897,45.1140675 37.3908111,44.1832925 38.4804778,44.1832925 Z M35.6104778,51.3435175 L35.6104778,50.7669175 L39.5483667,49.8803175 L39.9263667,50.1019675 L39.9263667,59.1686925 L41.3061444,59.6561675 L41.3061444,60.2986425 L35.5443667,60.2986425 L35.5443667,59.6561675 L37.0120333,59.1686925 L37.0120333,52.2742925 L35.6104778,51.3435175 Z" id="Fill-37" fill="#231F20"/>
+                <path d="M47.9577,55.3780125 C45.5769222,55.4663625 44.7097,56.2863125 44.7097,57.3953375 C44.7097,58.4144625 45.2883667,58.9693625 46.3111444,58.9693625 C46.7568111,58.9693625 47.3572556,58.7694125 47.9577,58.3927625 L47.9577,55.3780125 Z M50.7164778,58.1928125 C50.7164778,58.9468875 51.0501444,59.4126625 52.3404778,59.4343625 L52.3848111,60.0109625 C51.7400333,60.2543125 50.7608111,60.5209125 49.8935889,60.5209125 C48.8700333,60.5209125 48.2913667,60.0993125 48.1132556,59.0577125 L47.8464778,59.0135375 C46.8454778,59.9443125 45.8888111,60.4984375 44.7097,60.4984375 C43.1082556,60.4984375 41.9952556,59.6343125 41.9952556,58.1044625 C41.9952556,56.2204375 43.9537,54.9347125 47.9577,54.6464125 L47.9577,53.4273375 C47.9577,51.7641875 47.2903667,51.2100625 45.6220333,51.2100625 C45.1771444,51.2100625 44.7765889,51.2991875 44.4872556,51.4100125 C44.8878111,51.7424875 45.1102556,52.1191375 45.1102556,52.6066125 C45.1102556,53.3157375 44.6202556,53.8706375 43.7755889,53.8706375 C42.9744778,53.8706375 42.3740333,53.2498625 42.3740333,52.5182625 C42.3740333,50.9884125 44.8209222,49.9468125 47.1565889,49.9468125 C49.5817,49.9468125 50.7164778,50.8558875 50.7164778,53.0282125 L50.7164778,58.1928125 Z" id="Fill-39" fill="#231F20"/>
+                <path d="M56.4567111,60.58733 C54.6989333,60.58733 53.8091556,59.744905 53.8091556,57.94923 L53.8091556,51.720555 L52.0964889,51.720555 L52.0964889,51.12148 L54.2540444,49.63658 L55.9006,47.13178 L56.7234889,47.13178 L56.7234889,50.23488 L60.1052667,50.23488 L60.1052667,51.720555 L56.7234889,51.720555 L56.7234889,56.906855 C56.7234889,57.838405 56.8572667,58.326655 57.3021556,58.636655 C57.7027111,58.90248 58.1250444,58.96913 58.8592667,58.96913 C59.2154889,58.96913 59.6821556,58.946655 60.1496,58.90248 L60.1721556,59.589905 C59.3484889,60.099855 57.7688222,60.58733 56.4567111,60.58733" id="Fill-41" fill="#231F20"/>
+                <path d="M63.3534222,44.1832925 C64.4213111,44.1832925 65.1780889,44.9590675 65.1780889,45.9789675 C65.1780889,46.9763925 64.3769778,47.8846925 63.2647556,47.8846925 C62.2412,47.8846925 61.4626444,47.1313925 61.4626444,46.1114925 C61.4626444,45.1140675 62.2637556,44.1832925 63.3534222,44.1832925 Z M60.4834222,51.3435175 L60.4834222,50.7669175 L64.4213111,49.8803175 L64.7993111,50.1019675 L64.7993111,59.1686925 L66.1790889,59.6561675 L66.1790889,60.2986425 L60.4173111,60.2986425 L60.4173111,59.6561675 L61.8849778,59.1686925 L61.8849778,52.2742925 L60.4834222,51.3435175 Z" id="Fill-43" fill="#231F20"/>
+                <polygon id="Fill-45" fill="#231F20" points="72.8530444 57.3727075 72.9860444 57.3727075 74.0096 54.7570825 75.0774889 51.5873325 73.7871556 50.8999075 73.7871556 50.2791325 78.1924889 50.2791325 78.1924889 50.9448575 76.7014889 51.6756825 72.8748222 60.5432325 71.2290444 60.5432325 67.3129333 51.4765075 65.9564889 50.9448575 65.9564889 50.2791325 71.9407111 50.2791325 71.9407111 50.8999075 70.5617111 51.4315575"/>
+                <path d="M85.2226667,54.180715 C85.2001111,51.92004 84.6222222,50.85519 83.3762222,50.85519 C82.1527778,50.85519 81.2847778,52.16339 81.1961111,54.37989 L85.2226667,54.180715 Z M81.1735556,55.17814 C81.1961111,57.59459 82.2414444,58.946965 84.4884444,58.946965 C85.512,58.946965 86.4017778,58.74779 87.7146667,58.25954 L87.9588889,58.76949 C86.5798889,59.900215 84.9333333,60.65429 83.4648889,60.65429 C80.2837778,60.65429 78.2592222,58.570315 78.2592222,55.48814 C78.2592222,52.29669 80.773,49.94689 83.5535556,49.94689 C86.891,49.94689 88.1813333,52.36334 88.0701111,55.17814 L81.1735556,55.17814 Z" id="Fill-47" fill="#231F20"/>
+                <path d="M99.2454556,53.522275 C98.7523444,53.86405 98.1752333,54.07795 97.3725667,54.095 C96.4065667,54.11515 95.6809,53.4781 95.6140111,52.53725 C95.5782333,52.037375 95.7462333,51.5592 96.0876778,51.1934 C96.4299,50.826825 96.8957889,50.62145 97.3990111,50.617575 C97.6541222,50.61525 97.9232333,50.667175 98.1542333,50.766375 C98.1542333,50.766375 98.0912333,51.291825 98.0702333,51.5282 L98.8651222,51.53285 L99.0074556,50.3502 L98.7842333,50.20915 C98.3665667,49.9534 97.8851222,49.81855 97.3920111,49.8232 C96.6741222,49.8294 95.9842333,50.13165 95.4973444,50.653225 C95.0034556,51.18255 94.7600111,51.87075 94.8121222,52.592275 C94.8572333,53.232425 95.1341222,53.816775 95.5899,54.2376 C96.0511222,54.664625 96.6725667,54.894025 97.3445667,54.887825 C97.3609,54.887825 97.3772333,54.887825 97.3943444,54.88705 C97.8011222,54.876975 98.2600111,54.799475 98.7647889,54.6003 L99.3380111,53.491275 C99.3380111,53.491275 99.3263444,53.4657 99.2454556,53.522275" id="Fill-49" fill="#FF414C"/>
+                <path d="M101.945278,56.363425 L101.848833,57.4523 L101.847278,57.4647 C101.808389,57.877 101.5875,58.226525 101.256167,58.4009 C100.993278,58.539625 100.7055,58.53885 100.4465,58.398575 C100.177389,58.252875 100.049056,58.03355 100.000833,57.756875 C99.9253889,57.318225 100.175833,56.860975 100.517278,56.630025 L105.846611,53.175075 C105.887056,53.468025 105.908056,53.7664 105.908056,54.069425 C105.908056,57.645275 102.991389,60.55385 99.4058333,60.55385 C95.8202778,60.55385 92.9028333,57.645275 92.9028333,54.069425 C92.9028333,50.49435 95.8202778,47.585 99.4058333,47.585 C102.086833,47.585 104.392944,49.2125 105.386167,51.5282 L106.036389,51.107375 C104.897722,48.5863 102.355167,46.8255 99.4058333,46.8255 C95.3971667,46.8255 92.1359444,50.075075 92.1359444,54.069425 C92.1359444,58.063775 95.3971667,61.31335 99.4058333,61.31335 C103.4145,61.31335 106.675722,58.063775 106.675722,54.069425 C106.675722,53.6083 106.631389,53.15725 106.548167,52.72015 C106.498389,52.434175 106.388722,52.054425 106.370056,51.9932 L104.8565,52.936375 L102.852167,54.201175 C102.781389,53.93535 102.619611,53.666425 102.277389,53.511425 C101.872167,53.32775 101.215722,53.469575 100.7335,53.742375 C100.7335,53.742375 102.053389,51.19805 102.271944,50.780325 C102.285944,50.752425 102.2665,50.72065 102.235389,50.72065 L102.1545,50.721425 L102.155278,50.719875 L99.7682778,50.719875 L99.6687222,51.54215 L100.042056,51.5406 L101.005722,51.5406 L99.1165,55.145125 C99.0799444,55.21565 99.1561667,55.290825 99.2269444,55.2544 L100.235722,54.720425 C100.721833,54.47165 101.532278,53.960925 101.932833,54.2562 C101.991167,54.298825 102.065833,54.41275 102.0705,54.531325 C102.072833,54.577825 102.0495,54.621225 102.010611,54.647575 L100.000056,56.0015 C99.3459444,56.4665 99.0908333,57.15935 99.1515,57.8119 C99.2020556,58.361375 99.5450556,58.834125 100.053722,59.110025 C100.321278,59.254175 100.609833,59.322375 100.897611,59.314625 C101.151167,59.30765 101.403944,59.241775 101.640389,59.117 C102.214389,58.815525 102.5955,58.228075 102.662389,57.546075 L102.824167,55.7132 L101.945278,56.363425 Z" id="Fill-51" fill="#FF414C"/>
+            </g>
+        </g>
+    </g>
+</svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/partners/intel.svg b/pandas/web/pandas/static/img/partners/intel.svg
new file mode 100644
index 0000000000000..61d4ac6ce192b
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/intel.svg
@@ -0,0 +1,64 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Generator: Adobe Illustrator 23.0.4, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+
+<svg
+   version="1.1"
+   id="Layer_1"
+   x="0px"
+   y="0px"
+   viewBox="0 0 450.00001 230.00001"
+   xml:space="preserve"
+   sodipodi:docname="intel.svg"
+   inkscape:version="1.2.2 (b0a8486541, 2022-12-01)"
+   width="450"
+   height="230"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"><defs
+   id="defs1066" /><sodipodi:namedview
+   id="namedview1064"
+   pagecolor="#ffffff"
+   bordercolor="#666666"
+   borderopacity="1.0"
+   inkscape:pageshadow="2"
+   inkscape:pageopacity="0.0"
+   inkscape:pagecheckerboard="0"
+   showgrid="false"
+   inkscape:zoom="0.86470488"
+   inkscape:cx="164.79611"
+   inkscape:cy="49.727949"
+   inkscape:window-width="1920"
+   inkscape:window-height="1051"
+   inkscape:window-x="0"
+   inkscape:window-y="0"
+   inkscape:window-maximized="1"
+   inkscape:current-layer="Layer_1"
+   inkscape:showpageshadow="2"
+   inkscape:deskcolor="#d1d1d1" />
+<style
+   type="text/css"
+   id="style1053">
+	.st0{fill:#0068B5;}
+</style>
+
+<rect
+   x="30.9"
+   y="41.799999"
+   class="st0"
+   width="28.1"
+   height="28.1"
+   id="rect1055"
+   style="fill:#00c7fd;fill-opacity:1" /><g
+   id="g1061"
+   transform="translate(26.2,36.6)">
+	<path
+   d="M 32.1,151.6 V 50.4 H 5.5 v 101.2 z m 176.8,1 v -24.8 c -3.9,0 -7.2,-0.2 -9.6,-0.6 -2.8,-0.4 -4.9,-1.4 -6.3,-2.8 -1.4,-1.4 -2.3,-3.4 -2.8,-6 -0.4,-2.5 -0.6,-5.8 -0.6,-9.8 V 73.2 h 19.3 V 50.4 H 189.6 V 10.9 h -26.7 v 97.9 c 0,8.3 0.7,15.3 2.1,20.9 1.4,5.5 3.8,10 7.1,13.4 3.3,3.4 7.7,5.8 13,7.3 5.4,1.5 12.2,2.2 20.3,2.2 z m 152.8,-1 V 3.1 H 335 V 151.6 Z M 137.2,60.3 c -7.4,-8 -17.8,-12 -31,-12 -6.4,0 -12.2,1.3 -17.5,3.9 C 83.5,54.8 79,58.4 75.5,63 L 74,64.9 V 63.2 50.4 H 47.7 v 101.2 h 26.5 v -53.9 3.7 c 0,-0.6 0,-1.2 0,-1.8 0.3,-9.5 2.6,-16.5 7,-21 4.7,-4.8 10.4,-7.2 16.9,-7.2 7.7,0 13.6,2.4 17.5,7 3.8,4.6 5.8,11.1 5.8,19.4 v 0 0.2 0 0 53.5 h 26.9 V 94.1 c 0.1,-14.4 -3.7,-25.8 -11.1,-33.8 z m 184,40.5 c 0,-7.3 -1.3,-14.1 -3.8,-20.5 -2.6,-6.3 -6.2,-11.9 -10.7,-16.7 -4.6,-4.8 -10.1,-8.5 -16.5,-11.2 -6.4,-2.7 -13.5,-4 -21.2,-4 -7.3,0 -14.2,1.4 -20.6,4.1 -6.4,2.8 -12,6.5 -16.7,11.2 -4.7,4.7 -8.5,10.3 -11.2,16.7 -2.8,6.4 -4.1,13.3 -4.1,20.6 0,7.3 1.3,14.2 3.9,20.6 2.6,6.4 6.3,12 10.9,16.7 4.6,4.7 10.3,8.5 16.9,11.2 6.6,2.8 13.9,4.2 21.7,4.2 22.6,0 36.6,-10.3 45,-19.9 l -19.2,-14.6 c -4,4.8 -13.6,11.3 -25.6,11.3 -7.5,0 -13.7,-1.7 -18.4,-5.2 -4.7,-3.4 -7.9,-8.2 -9.6,-14.1 l -0.3,-0.9 h 79.5 z m -79.3,-9.3 c 0,-7.4 8.5,-20.3 26.8,-20.4 18.3,0 26.9,12.9 26.9,20.3 z"
+   id="path1057"
+   style="fill:#0068b5;fill-opacity:1" />
+	<path
+   d="m 392.1,138.4 c -0.5,-1.2 -1.2,-2.2 -2.1,-3.1 -0.9,-0.9 -1.9,-1.6 -3.1,-2.1 -1.2,-0.5 -2.5,-0.8 -3.8,-0.8 -1.4,0 -2.6,0.3 -3.8,0.8 -1.2,0.5 -2.2,1.2 -3.1,2.1 -0.9,0.9 -1.6,1.9 -2.1,3.1 -0.5,1.2 -0.8,2.5 -0.8,3.8 0,1.4 0.3,2.6 0.8,3.8 0.5,1.2 1.2,2.2 2.1,3.1 0.9,0.9 1.9,1.6 3.1,2.1 1.2,0.5 2.5,0.8 3.8,0.8 1.4,0 2.6,-0.3 3.8,-0.8 1.2,-0.5 2.2,-1.2 3.1,-2.1 0.9,-0.9 1.6,-1.9 2.1,-3.1 0.5,-1.2 0.8,-2.5 0.8,-3.8 0,-1.3 -0.3,-2.6 -0.8,-3.8 z m -1.6,7 c -0.4,1 -1,1.9 -1.7,2.6 -0.7,0.7 -1.6,1.3 -2.6,1.7 -1,0.4 -2,0.6 -3.2,0.6 -1.1,0 -2.2,-0.2 -3.2,-0.6 -1,-0.4 -1.9,-1 -2.6,-1.7 -0.7,-0.7 -1.3,-1.6 -1.7,-2.6 -0.4,-1 -0.6,-2 -0.6,-3.2 0,-1.1 0.2,-2.2 0.6,-3.2 0.4,-1 1,-1.9 1.7,-2.6 0.7,-0.7 1.6,-1.3 2.6,-1.7 1,-0.4 2,-0.6 3.2,-0.6 1.1,0 2.2,0.2 3.2,0.6 1,0.4 1.9,1 2.6,1.7 0.7,0.7 1.3,1.6 1.7,2.6 0.4,1 0.6,2 0.6,3.2 0.1,1.2 -0.2,2.2 -0.6,3.2 z m -5.6,-2.4 c 0.8,-0.1 1.4,-0.4 1.9,-0.9 0.5,-0.5 0.8,-1.2 0.8,-2.2 0,-1.1 -0.3,-1.9 -1,-2.5 -0.6,-0.6 -1.7,-0.9 -3,-0.9 h -4.4 v 11.3 h 2.1 v -4.6 h 1.5 l 2.8,4.6 h 2.2 z m -1.1,-1.6 c -0.3,0 -0.6,0 -1,0 h -1.5 v -3.2 h 1.5 c 0.3,0 0.6,0 1,0 0.3,0 0.6,0.1 0.9,0.2 0.3,0.1 0.5,0.3 0.6,0.5 0.1,0.2 0.2,0.5 0.2,0.9 0,0.4 -0.1,0.7 -0.2,0.9 -0.2,0.2 -0.4,0.4 -0.6,0.5 -0.3,0.1 -0.6,0.2 -0.9,0.2 z"
+   id="path1059"
+   style="fill:#0068b5;fill-opacity:1" />
+</g>
+</svg>
diff --git a/pandas/web/pandas/static/img/partners/numfocus.svg b/pandas/web/pandas/static/img/partners/numfocus.svg
new file mode 100644
index 0000000000000..fcdd87b41e475
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/numfocus.svg
@@ -0,0 +1,60 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Generator: Adobe Illustrator 19.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px" viewBox="0 0 432 135.7" style="enable-background:new 0 0 432 135.7;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#F1563F;}
+	.st1{fill:#008896;}
+</style>
+<g>
+	<g>
+		<g>
+			<path class="st0" d="M97.9,12.2v51.9c0,12.7-6.8,19.7-19.1,19.7c-12.2,0-19-7-19-19.7V12.2h5v51.9c0,9.8,4.8,14.9,14,14.9     c9.2,0,14.1-5.2,14.1-14.9V12.2H97.9z"/>
+		</g>
+		<g>
+			<path class="st1" d="M329.8,29.8c0-0.3,0.1-0.7,0.1-1c0-8.3-6.9-16.7-20.1-16.7c-13.1,0-20.6,7.7-20.6,21.2v29.5     c0,13.5,7.4,21.2,20.3,21.2c13.4,0,20.4-8.4,20.4-16.7c0-0.3,0-0.8-0.1-1.4l-7.8,0c-0.7,4.6-1.7,10.4-12,10.4     c-9,0-13-4.1-13-13.4V33.3c0-9.2,4-13.4,12.7-13.4c7.7,0,11.8,3.4,12.2,10.1L329.8,29.8z"/>
+		</g>
+		<g>
+			<path class="st1" d="M376.2,12.4v50.3c0,13.6-7.3,21.2-20.5,21.2c-13.2,0-20.4-7.5-20.4-21.2V12.4h7.9v50.3     c0,9,4.1,13.4,12.5,13.4c8.4,0,12.6-4.5,12.6-13.4V12.4H376.2z"/>
+		</g>
+		<g>
+			<path class="st1" d="M414.9,22.6c-2-1-6-2.9-11.3-2.9c-8.4,0-12.6,3.4-12.6,10c0,7.1,4.8,9.1,12.5,11.8     c8.3,2.9,18.6,6.5,18.6,21.9c0,13-7.5,20.5-20.6,20.5c-8.2,0-14.2-2.7-17.3-6c-1.2-1.3-0.5-0.6-1.2-1.6l5.3-5.1     c1.9,2.2,5,5,12.8,5c8.7,0,13.2-4.1,13.2-12.3c0-9.9-6.6-12.3-14.3-15.1C392,46,383,42.8,383,30.1c0-11.3,7.7-18.1,20.6-18.1     c5.5,0,12.5,1.3,15.4,4.1L414.9,22.6z"/>
+		</g>
+		<g>
+			<path class="st1" d="M283.5,47.2c0-21.2-17.2-38.5-38.5-38.5c-21.2,0-38.5,17.2-38.5,38.5c0,21.2,17.2,38.5,38.5,38.5     C266.2,85.6,283.5,68.4,283.5,47.2z M213.1,47.2c0-17.6,14.3-31.9,31.9-31.9c17.6,0,31.9,14.3,31.9,31.9     c0,17.6-14.3,31.9-31.9,31.9C227.4,79.1,213.1,64.8,213.1,47.2z"/>
+		</g>
+		<g>
+			<path class="st0" d="M233.9,32.3c1.2,0,2.1-1,2.1-2.3c0-1.3-0.9-2.3-2.1-2.3h-7.3c-1.3,0-2.3,1-2.3,2.2v34.5c0,1.2,1,2.2,2.3,2.2     h7.3c1.2,0,2.1-1,2.1-2.3c0-1.3-1-2.3-2.1-2.3h-4.9V32.3H233.9z"/>
+		</g>
+		<g>
+			<path class="st0" d="M256.1,62c-1.2,0-2.2,1-2.2,2.3c0,1.3,1,2.3,2.2,2.3h7.3c1.3,0,2.3-1,2.3-2.2V29.9c0-1.2-1-2.2-2.3-2.2h-7.3     c-1.2,0-2.2,1-2.2,2.3c0,1.3,1,2.3,2.2,2.3h4.9V62H256.1z"/>
+		</g>
+		<polygon class="st1" points="208.7,19.8 208.7,12.1 171.8,12.1 171.8,83.7 179.7,83.7 179.7,51.5 196.2,51.5 196.2,43.9     179.7,43.9 179.7,19.8   "/>
+		<polygon class="st0" points="156.6,12.2 152.3,12.2 133.2,51.9 113.9,12.2 109.7,12.2 109.7,83.7 114.6,83.7 114.6,24.3     133.1,61.9 151.6,24.4 151.6,83.7 156.6,83.7   "/>
+		<polygon class="st0" points="44.6,83.7 48.1,83.7 48.1,12.1 43.1,12.1 43.2,70.5 14.2,12.2 10.1,12.2 10.1,83.7 15.1,83.7     14.9,23.1   "/>
+	</g>
+	<g id="XMLID_3_">
+		<path class="st1" d="M34.9,125.3c-1.2,0-2.3-0.2-3.2-0.5c-0.9-0.3-1.6-0.8-2.1-1.5c-0.5-0.7-0.9-1.4-1.1-2.3    c-0.2-0.9-0.4-1.9-0.4-3v-8.5c0-2.3,0.5-4.1,1.6-5.3c1.1-1.2,2.8-1.8,5.2-1.8c2.4,0,4.1,0.6,5.2,1.8c1.1,1.2,1.6,3,1.6,5.3v8.5    c0,2.3-0.5,4.1-1.6,5.4C39.1,124.7,37.3,125.3,34.9,125.3z M33.3,122.5c0.4,0.2,1,0.3,1.7,0.3c0.7,0,1.2-0.1,1.7-0.3    c0.4-0.2,0.8-0.5,1-0.9c0.2-0.4,0.4-0.8,0.5-1.3c0.1-0.5,0.1-1,0.1-1.7v-9.9c0-0.7,0-1.2-0.1-1.7c-0.1-0.5-0.2-0.9-0.5-1.2    c-0.2-0.4-0.6-0.7-1-0.8c-0.4-0.2-1-0.3-1.7-0.3c-0.7,0-1.2,0.1-1.7,0.3c-0.4,0.2-0.8,0.5-1,0.8c-0.2,0.4-0.4,0.8-0.5,1.2    c-0.1,0.5-0.1,1-0.1,1.7v9.9c0,0.7,0,1.3,0.1,1.7c0.1,0.5,0.2,0.9,0.5,1.3C32.5,122.1,32.8,122.4,33.3,122.5z"/>
+		<path class="st1" d="M46.7,125v-22.5h6.2c2.2,0,3.7,0.5,4.7,1.6c1,1.1,1.5,2.6,1.5,4.6c0,1.8-0.5,3.3-1.6,4.3    c-1,1-2.6,1.5-4.6,1.5h-2.7V125H46.7z M50.2,112.3h1.6c1.5,0,2.6-0.2,3.1-0.7c0.6-0.5,0.9-1.4,0.9-2.8c0-0.6,0-1,0-1.4    c0-0.4-0.1-0.7-0.2-1c-0.1-0.3-0.2-0.6-0.4-0.8c-0.2-0.2-0.4-0.3-0.7-0.5c-0.3-0.1-0.7-0.2-1.1-0.3c-0.4,0-0.9-0.1-1.5-0.1h-1.6    V112.3z"/>
+		<path class="st1" d="M63,125v-22.5h9.6v2.3h-6.2v7.4h5v2.2h-5v8.3h6.2v2.3H63z"/>
+		<path class="st1" d="M77.1,125v-22.5h2.4l7.1,15v-15h2.9V125h-2.2L80,109.7V125H77.1z"/>
+		<path class="st1" d="M109.6,125.3c-1,0-1.9-0.1-2.7-0.4c-0.8-0.3-1.4-0.6-1.9-1c-0.5-0.4-0.9-1-1.2-1.6c-0.3-0.6-0.5-1.3-0.7-2    c-0.1-0.7-0.2-1.5-0.2-2.4v-8c0-1,0.1-1.8,0.2-2.5c0.1-0.7,0.3-1.4,0.7-2.1c0.3-0.6,0.7-1.2,1.2-1.6c0.5-0.4,1.1-0.7,1.9-1    c0.8-0.2,1.7-0.4,2.7-0.4c2.2,0,3.8,0.5,4.8,1.6c1,1.1,1.5,2.7,1.5,4.8v1.8h-3.3V109c0-0.3,0-0.6,0-0.8c0-0.2,0-0.4,0-0.7    c0-0.3,0-0.5-0.1-0.7c0-0.2-0.1-0.4-0.2-0.6c-0.1-0.2-0.1-0.4-0.2-0.5c-0.1-0.1-0.2-0.3-0.4-0.4c-0.1-0.1-0.3-0.2-0.5-0.3    c-0.2-0.1-0.4-0.1-0.7-0.2c-0.3,0-0.6-0.1-0.9-0.1c-0.5,0-0.9,0-1.3,0.1c-0.4,0.1-0.7,0.2-0.9,0.4c-0.2,0.2-0.4,0.4-0.6,0.7    c-0.1,0.2-0.3,0.6-0.3,0.9c-0.1,0.4-0.1,0.8-0.1,1.1s0,0.8,0,1.3v8.9c0,1.7,0.2,2.9,0.7,3.5c0.5,0.7,1.3,1,2.5,1    c0.5,0,0.9,0,1.2-0.1c0.3-0.1,0.6-0.2,0.8-0.4c0.2-0.2,0.4-0.4,0.5-0.7c0.1-0.2,0.2-0.5,0.3-0.9c0.1-0.4,0.1-0.7,0.1-1.1    c0-0.3,0-0.8,0-1.3v-1.7h3.3v1.7c0,0.9-0.1,1.6-0.2,2.3c-0.1,0.7-0.3,1.3-0.6,1.9c-0.3,0.6-0.7,1.1-1.1,1.5    c-0.5,0.4-1.1,0.7-1.8,0.9C111.4,125.2,110.6,125.3,109.6,125.3z"/>
+		<path class="st1" d="M127.1,125.3c-1.2,0-2.3-0.2-3.2-0.5c-0.9-0.3-1.6-0.8-2.1-1.5c-0.5-0.7-0.9-1.4-1.1-2.3    c-0.2-0.9-0.4-1.9-0.4-3v-8.5c0-2.3,0.5-4.1,1.6-5.3c1.1-1.2,2.8-1.8,5.2-1.8c2.4,0,4.1,0.6,5.2,1.8c1.1,1.2,1.6,3,1.6,5.3v8.5    c0,2.3-0.5,4.1-1.6,5.4C131.2,124.7,129.5,125.3,127.1,125.3z M125.4,122.5c0.4,0.2,1,0.3,1.7,0.3c0.7,0,1.2-0.1,1.7-0.3    c0.4-0.2,0.8-0.5,1-0.9c0.2-0.4,0.4-0.8,0.5-1.3c0.1-0.5,0.1-1,0.1-1.7v-9.9c0-0.7,0-1.2-0.1-1.7c-0.1-0.5-0.2-0.9-0.5-1.2    c-0.2-0.4-0.6-0.7-1-0.8c-0.4-0.2-1-0.3-1.7-0.3c-0.7,0-1.2,0.1-1.7,0.3c-0.4,0.2-0.8,0.5-1,0.8c-0.2,0.4-0.4,0.8-0.5,1.2    c-0.1,0.5-0.1,1-0.1,1.7v9.9c0,0.7,0,1.3,0.1,1.7c0.1,0.5,0.2,0.9,0.5,1.3C124.6,122.1,125,122.4,125.4,122.5z"/>
+		<path class="st1" d="M138.9,125v-22.5h5.4c2.7,0,4.6,0.6,5.7,1.7c1.1,1.1,1.7,2.8,1.7,5.2v8.3c0,2.5-0.5,4.3-1.6,5.5    c-1.1,1.2-2.9,1.8-5.5,1.8H138.9z M142.3,122.8h2c0.5,0,1,0,1.4-0.1c0.4-0.1,0.7-0.2,1-0.3c0.3-0.1,0.5-0.3,0.7-0.6    c0.2-0.3,0.3-0.6,0.4-0.8c0.1-0.2,0.2-0.6,0.2-1.1c0-0.5,0.1-0.9,0.1-1.2c0-0.3,0-0.8,0-1.5v-7.3c0-0.5,0-1,0-1.4    c0-0.4-0.1-0.7-0.1-1.1c-0.1-0.4-0.2-0.7-0.3-0.9c-0.1-0.2-0.3-0.5-0.5-0.7c-0.2-0.2-0.4-0.4-0.7-0.5c-0.3-0.1-0.6-0.2-1-0.3    c-0.4-0.1-0.8-0.1-1.3-0.1h-1.9V122.8z"/>
+		<path class="st1" d="M156.6,125v-22.5h9.6v2.3h-6.2v7.4h5v2.2h-5v8.3h6.2v2.3H156.6z"/>
+		<path class="st1" d="M178.9,112.4v-2.3h9.4v2.3H178.9z M178.9,117.3v-2.3h9.4v2.3H178.9z"/>
+		<path class="st1" d="M202.1,125v-22.5h5.7c2.3,0,3.9,0.5,5,1.4c1.1,0.9,1.6,2.3,1.6,4.3c0,2.9-1.2,4.5-3.6,4.8    c1.5,0.3,2.5,1,3.2,1.9c0.7,0.9,1,2.2,1,3.8c0,2-0.5,3.6-1.6,4.7c-1,1.1-2.7,1.7-4.8,1.7H202.1z M205.6,111.9h2    c1.4,0,2.4-0.3,3-0.9c0.6-0.6,0.8-1.5,0.8-2.9c0-0.4,0-0.8-0.1-1.2s-0.2-0.6-0.3-0.8c-0.1-0.2-0.3-0.4-0.5-0.6    c-0.2-0.2-0.5-0.3-0.7-0.4c-0.2-0.1-0.5-0.2-0.9-0.2c-0.4,0-0.8-0.1-1.1-0.1c-0.4,0-0.8,0-1.4,0h-0.8V111.9z M205.6,122.8h2.3    c1.5,0,2.5-0.3,3.1-1c0.6-0.6,0.8-1.7,0.8-3.2c0-1.4-0.3-2.5-1-3.2c-0.7-0.7-1.7-1.1-3.2-1.1h-2.1V122.8z"/>
+		<path class="st1" d="M219.7,125v-22.5h9.6v2.3h-6.2v7.4h5v2.2h-5v8.3h6.2v2.3H219.7z"/>
+		<path class="st1" d="M236.4,125v-20.2h-4.6v-2.3h12.4v2.3h-4.4V125H236.4z"/>
+		<path class="st1" d="M250.2,125v-20.2h-4.6v-2.3h12.4v2.3h-4.4V125H250.2z"/>
+		<path class="st1" d="M261.5,125v-22.5h9.6v2.3H265v7.4h5v2.2h-5v8.3h6.2v2.3H261.5z"/>
+		<path class="st1" d="M275.6,125v-22.5h5c2.5,0,4.4,0.5,5.5,1.4c1.2,0.9,1.8,2.5,1.8,4.6c0,2.9-1,4.7-3.1,5.3l3.6,11.3H285    l-3.3-10.6H279V125H275.6z M279,112.2h1.3c1.5,0,2.6-0.3,3.2-0.8c0.6-0.5,1-1.5,1-2.9c0-1.4-0.3-2.3-0.8-2.9    c-0.6-0.6-1.6-0.8-3.1-0.8H279V112.2z"/>
+		<path class="st1" d="M307.5,125.3c-2.1,0-3.7-0.6-4.8-1.8c-1.1-1.2-1.7-2.8-1.8-4.8l3.1-0.8c0.2,3.2,1.4,4.8,3.5,4.8    c0.9,0,1.6-0.2,2-0.7c0.5-0.4,0.7-1.1,0.7-2c0-0.5-0.1-0.9-0.2-1.3c-0.1-0.4-0.3-0.8-0.6-1.1c-0.3-0.4-0.6-0.7-0.8-0.9    c-0.3-0.2-0.6-0.6-1.1-1l-4.2-3.4c-0.8-0.7-1.5-1.4-1.8-2.2c-0.4-0.8-0.6-1.7-0.6-2.8c0-1.6,0.5-2.9,1.6-3.8    c1-0.9,2.5-1.4,4.3-1.4c2,0,3.5,0.4,4.5,1.4c1,1,1.6,2.4,1.8,4.4l-2.9,0.7c0-0.5-0.1-0.9-0.2-1.3c-0.1-0.4-0.2-0.8-0.3-1.1    c-0.2-0.4-0.4-0.7-0.6-0.9c-0.2-0.2-0.5-0.4-0.9-0.6c-0.4-0.1-0.8-0.2-1.3-0.2c-1.8,0.1-2.7,0.9-2.7,2.5c0,0.7,0.1,1.2,0.4,1.7    c0.3,0.4,0.7,0.9,1.3,1.4l4.2,3.4c1.1,0.9,2,1.8,2.6,2.9c0.6,1,1,2.2,1,3.5c0,1.6-0.5,3-1.7,3.9    C310.7,124.8,309.3,125.3,307.5,125.3z"/>
+		<path class="st1" d="M323.9,125.3c-1,0-1.9-0.1-2.7-0.4c-0.8-0.3-1.4-0.6-1.9-1c-0.5-0.4-0.9-1-1.2-1.6c-0.3-0.6-0.5-1.3-0.7-2    c-0.1-0.7-0.2-1.5-0.2-2.4v-8c0-1,0.1-1.8,0.2-2.5c0.1-0.7,0.3-1.4,0.7-2.1c0.3-0.6,0.7-1.2,1.2-1.6c0.5-0.4,1.1-0.7,1.9-1    c0.8-0.2,1.7-0.4,2.7-0.4c2.2,0,3.9,0.5,4.8,1.6c1,1.1,1.5,2.7,1.5,4.8v1.8h-3.3V109c0-0.3,0-0.6,0-0.8c0-0.2,0-0.4,0-0.7    c0-0.3,0-0.5-0.1-0.7c0-0.2-0.1-0.4-0.2-0.6c-0.1-0.2-0.1-0.4-0.2-0.5c-0.1-0.1-0.2-0.3-0.4-0.4c-0.1-0.1-0.3-0.2-0.5-0.3    c-0.2-0.1-0.4-0.1-0.7-0.2c-0.3,0-0.5-0.1-0.9-0.1c-0.5,0-0.9,0-1.3,0.1c-0.4,0.1-0.7,0.2-0.9,0.4c-0.2,0.2-0.4,0.4-0.6,0.7    c-0.1,0.2-0.3,0.6-0.3,0.9c-0.1,0.4-0.1,0.8-0.1,1.1c0,0.4,0,0.8,0,1.3v8.9c0,1.7,0.2,2.9,0.7,3.5c0.5,0.7,1.3,1,2.5,1    c0.5,0,0.9,0,1.2-0.1c0.3-0.1,0.6-0.2,0.8-0.4c0.2-0.2,0.4-0.4,0.5-0.7c0.1-0.2,0.2-0.5,0.3-0.9c0.1-0.4,0.1-0.7,0.1-1.1    c0-0.3,0-0.8,0-1.3v-1.7h3.3v1.7c0,0.9-0.1,1.6-0.2,2.3c-0.1,0.7-0.3,1.3-0.6,1.9c-0.3,0.6-0.7,1.1-1.1,1.5    c-0.5,0.4-1.1,0.7-1.8,0.9C325.7,125.2,324.9,125.3,323.9,125.3z"/>
+		<path class="st1" d="M335.2,125v-22.5h3.4V125H335.2z"/>
+		<path class="st1" d="M344.2,125v-22.5h9.6v2.3h-6.2v7.4h5v2.2h-5v8.3h6.2v2.3H344.2z"/>
+		<path class="st1" d="M358.3,125v-22.5h2.4l7.1,15v-15h2.9V125h-2.2l-7.2-15.4V125H358.3z"/>
+		<path class="st1" d="M382.3,125.3c-1,0-1.9-0.1-2.7-0.4c-0.8-0.3-1.4-0.6-1.9-1c-0.5-0.4-0.9-1-1.2-1.6c-0.3-0.6-0.5-1.3-0.7-2    c-0.1-0.7-0.2-1.5-0.2-2.4v-8c0-1,0.1-1.8,0.2-2.5c0.1-0.7,0.3-1.4,0.7-2.1c0.3-0.6,0.7-1.2,1.2-1.6c0.5-0.4,1.1-0.7,1.9-1    c0.8-0.2,1.7-0.4,2.7-0.4c2.2,0,3.9,0.5,4.8,1.6c1,1.1,1.5,2.7,1.5,4.8v1.8h-3.3V109c0-0.3,0-0.6,0-0.8c0-0.2,0-0.4,0-0.7    c0-0.3,0-0.5-0.1-0.7c0-0.2-0.1-0.4-0.2-0.6c-0.1-0.2-0.1-0.4-0.2-0.5c-0.1-0.1-0.2-0.3-0.4-0.4c-0.1-0.1-0.3-0.2-0.5-0.3    c-0.2-0.1-0.4-0.1-0.7-0.2c-0.3,0-0.5-0.1-0.9-0.1c-0.5,0-0.9,0-1.3,0.1c-0.4,0.1-0.7,0.2-0.9,0.4c-0.2,0.2-0.4,0.4-0.6,0.7    c-0.1,0.2-0.3,0.6-0.3,0.9c-0.1,0.4-0.1,0.8-0.1,1.1c0,0.4,0,0.8,0,1.3v8.9c0,1.7,0.2,2.9,0.7,3.5c0.5,0.7,1.3,1,2.5,1    c0.5,0,0.9,0,1.2-0.1c0.3-0.1,0.6-0.2,0.8-0.4c0.2-0.2,0.4-0.4,0.5-0.7c0.1-0.2,0.2-0.5,0.3-0.9c0.1-0.4,0.1-0.7,0.1-1.1    c0-0.3,0-0.8,0-1.3v-1.7h3.3v1.7c0,0.9-0.1,1.6-0.2,2.3c-0.1,0.7-0.3,1.3-0.6,1.9c-0.3,0.6-0.7,1.1-1.1,1.5    c-0.5,0.4-1.1,0.7-1.8,0.9C384.1,125.2,383.2,125.3,382.3,125.3z"/>
+		<path class="st1" d="M393.4,125v-22.5h9.6v2.3h-6.2v7.4h5v2.2h-5v8.3h6.2v2.3H393.4z"/>
+	</g>
+</g>
+</svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/partners/nvidia.svg b/pandas/web/pandas/static/img/partners/nvidia.svg
new file mode 100644
index 0000000000000..59f9e19cf00ad
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/nvidia.svg
@@ -0,0 +1,56 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   version="1.1"
+   id="svg2"
+   x="0px"
+   y="0px"
+   width="450"
+   height="340"
+   viewBox="35.188 31.512 450.00001 340"
+   enable-background="new 35.188 31.512 351.46 258.785"
+   xml:space="preserve"
+   sodipodi:docname="nvidia.svg"
+   inkscape:version="1.0.1 (3bc2e813f5, 2020-09-07)"><metadata
+   id="metadata11"><rdf:RDF><cc:Work
+       rdf:about=""><dc:format>image/svg+xml</dc:format><dc:type
+         rdf:resource="http://purl.org/dc/dcmitype/StillImage" /><dc:title>generated by pstoedit version:3.44 from NVBadge_2D.eps</dc:title></cc:Work></rdf:RDF></metadata><defs
+   id="defs9" /><sodipodi:namedview
+   pagecolor="#ffffff"
+   bordercolor="#666666"
+   borderopacity="1"
+   objecttolerance="10"
+   gridtolerance="10"
+   guidetolerance="10"
+   inkscape:pageopacity="0"
+   inkscape:pageshadow="2"
+   inkscape:window-width="1920"
+   inkscape:window-height="1051"
+   id="namedview7"
+   showgrid="false"
+   inkscape:zoom="2.2870273"
+   inkscape:cx="163.94912"
+   inkscape:cy="154.68953"
+   inkscape:window-x="0"
+   inkscape:window-y="0"
+   inkscape:window-maximized="1"
+   inkscape:current-layer="svg2" />
+<title
+   id="title4">generated by pstoedit version:3.44 from NVBadge_2D.eps</title>
+
+
+<path
+   id="path17"
+   d="m 433.465,322.7165 c 0,3.771 -2.769,6.302 -6.047,6.302 v -0.023 c -3.371,0.023 -6.089,-2.508 -6.089,-6.278 0,-3.769 2.718,-6.293 6.089,-6.293 3.279,-0.001 6.047,2.523 6.047,6.292 z m 2.453,0 c 0,-5.175 -4.02,-8.179 -8.5,-8.179 -4.511,0 -8.531,3.004 -8.531,8.179 0,5.172 4.021,8.188 8.531,8.188 4.481,0 8.5,-3.016 8.5,-8.188 m -9.91,0.692 h 0.91 l 2.109,3.703 h 2.316 l -2.336,-3.859 c 1.207,-0.086 2.2,-0.661 2.2,-2.286 0,-2.019 -1.392,-2.668 -3.75,-2.668 h -3.411 v 8.813 h 1.961 v -3.703 m 10e-4,-1.492 v -2.122 h 1.364 c 0.742,0 1.753,0.06 1.753,0.965 0,0.985 -0.523,1.157 -1.398,1.157 h -1.719" /><path
+   id="path19"
+   d="m 378.676,277.6345 10.598,28.993 H 367.75 Z m -11.35,-11.289 -24.423,61.88 h 17.246 l 3.863,-10.934 h 28.903 l 3.656,10.934 h 18.722 l -24.605,-61.888 z m -49.033,61.903 h 17.497 v -61.922 l -17.5,-0.004 z m -121.467,-61.926 -14.598,49.078 -13.984,-49.074 -18.879,-0.004 19.972,61.926 h 25.207 l 20.133,-61.926 z m 70.725,13.484 h 7.52 c 10.91,0 17.966,4.898 17.966,17.609 0,12.714 -7.056,17.613 -17.966,17.613 h -7.52 z m -17.35,-13.484 v 61.926 h 28.366 c 15.113,0 20.048,-2.512 25.384,-8.148 3.769,-3.957 6.207,-12.641 6.207,-22.134 0,-8.707 -2.063,-16.468 -5.66,-21.304 -6.481,-8.649 -15.817,-10.34 -29.75,-10.34 z m -165.743,-0.086 v 62.012 h 17.645 v -47.086 l 13.672,0.004 c 4.527,0 7.754,1.128 9.934,3.457 2.765,2.945 3.894,7.699 3.894,16.395 v 27.23 h 17.098 v -34.262 c 0,-24.453 -15.586,-27.75 -30.836,-27.75 z m 137.583,0.086 0.007,61.926 h 17.489 v -61.926 z" /><path
+   id="path21"
+   fill="#77b900"
+   d="m 131.481,143.0215 c 0,0 22.504,-33.203 67.437,-36.638 v -12.046 c -49.769,3.997 -92.867,46.149 -92.867,46.149 0,0 24.41,70.565 92.867,77.026 v -12.804 c -50.237,-6.32 -67.437,-61.687 -67.437,-61.687 z m 67.437,36.223 v 11.726 c -37.968,-6.769 -48.507,-46.237 -48.507,-46.237 0,0 18.23,-20.195 48.507,-23.47 v 12.867 c -0.023,0 -0.039,-0.007 -0.058,-0.007 -15.891,-1.907 -28.305,12.938 -28.305,12.938 0,0 6.958,24.991 28.363,32.183 m 0,-107.125 v 22.218 c 1.461,-0.112 2.922,-0.207 4.391,-0.257 56.582,-1.907 93.449,46.406 93.449,46.406 0,0 -42.343,51.488 -86.457,51.488 -4.043,0 -7.828,-0.375 -11.383,-1.005 v 13.739 c 3.04,0.386 6.192,0.613 9.481,0.613 41.051,0 70.738,-20.965 99.484,-45.778 4.766,3.817 24.278,13.103 28.289,17.168 -27.332,22.883 -91.031,41.329 -127.144,41.329 -3.481,0 -6.824,-0.211 -10.11,-0.528 v 19.306 H 354.95 V 72.1195 Z m 0,49.144 v -14.879 c 1.446,-0.101 2.903,-0.179 4.391,-0.226 40.688,-1.278 67.382,34.965 67.382,34.965 0,0 -28.832,40.043 -59.746,40.043 -4.449,0 -8.438,-0.715 -12.028,-1.922 v -45.114 c 15.84,1.914 19.028,8.911 28.551,24.786 l 21.18,-17.859 c 0,0 -15.461,-20.277 -41.524,-20.277 -2.833,-0.001 -5.544,0.198 -8.206,0.483" />
+</svg>
diff --git a/pandas/web/pandas/static/img/partners/quansight_labs.svg b/pandas/web/pandas/static/img/partners/quansight_labs.svg
new file mode 100644
index 0000000000000..d49ab1b7d39ec
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/quansight_labs.svg
@@ -0,0 +1 @@
+<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 511.28 230.68"><defs><style>.cls-1{fill:#452391;}.cls-2{fill:#99c941;}.cls-3{fill:#973896;}</style></defs><path class="cls-1" d="M41.94,188.52c-2.37.51-7.48,2.71-12.51,5.54h3.46c5-2.71,10-5,14.08-6a15,15,0,0,0,1.62-.46Z"/><path class="cls-1" d="M21.51,200.85s9.82-6.64,19.13-6.73C52.76,194,64,212.43,78.34,212.43c13.42,0,19.3-9.22,22.1-16.83l-.58-.26c-5.22,7.72-10.57,10.32-19.14,10.32-15,0-20-15-40.06-14.73C30.05,191.06,21.41,199.35,21.51,200.85Z"/><path class="cls-1" d="M62.94,178.35c.13-.15.25-.31.37-.46a30.46,30.46,0,0,0,6.35-18.57,27.52,27.52,0,0,0-.42-4.5,30.27,30.27,0,0,0-3.75-10.91l.21.25C61,135.79,52.5,130.1,42.82,129.88h-1.3c-9.69.22-18.18,5.91-22.89,14.28l.21-.25A30.48,30.48,0,0,0,15,155.57a29.28,29.28,0,0,0-.31,3.85c.08,16.05,12.4,29.22,27.41,29.22A26.8,26.8,0,0,0,62.9,178.41ZM39.15,133.72l.38-.05.41,0,.69-.07.35,0,.44,0h1.5l.44,0,.34,0,.69.07.41,0,.38.05.43.08c11.63,2.15,16.16,15,16.16,25.45,0,6.22-1.59,13.22-5.3,18.35h0l-.33.45a16.86,16.86,0,0,1-13.87,6.76,18.31,18.31,0,0,1-2.76-.2c-.63-.05-1.26-.13-1.88-.22A16.13,16.13,0,0,1,32.19,182l.14,0c-6.93-4.77-9.77-14.46-9.77-22.76,0-10.48,4.53-23.3,16.16-25.45Z"/><path class="cls-1" d="M113.62,185.14c0,.43-.44.65-1.32.65H104a1.38,1.38,0,0,1-1.32-.9L101.38,182c-.28-.54-.55-.57-.82-.08q-3.54,4.68-10.27,4.68a10.93,10.93,0,0,1-8.42-3.4c-2.11-2.28-3.16-5.49-3.16-9.66v-16.5a3.43,3.43,0,0,0-2.71-3.7l-1.73-.49c-.87-.28-1.31-.58-1.31-.91s.44-.65,1.31-.65H85.61c.71,0,1.07.19,1.07.57a7.93,7.93,0,0,1-1,1.11c-.69.69-1,2.31-1,4.89v17q0,7.4,7.64,7.39A11.09,11.09,0,0,0,98,180.54c2.14-1.32,3.2-2.88,3.2-4.69v-18.8a3.43,3.43,0,0,0-2.71-3.7l-1.72-.49c-.88-.28-1.32-.58-1.32-.91s.44-.65,1.32-.65h9.61c.76,0,1.12.32,1.07,1l-.33,5.42v22.43a3.27,3.27,0,0,0,2.79,3.61l2.3.49C113.15,184.45,113.62,184.75,113.62,185.14Z"/><path class="cls-1" d="M149.59,185.14c0,.43-.44.65-1.31.65h-8.14a1.52,1.52,0,0,1-1.39-.9l-.66-1.15a1,1,0,0,0-.82-.58,1.9,1.9,0,0,0-1.07.41,17.43,17.43,0,0,1-10.26,3,11,11,0,0,1-7.15-2.21,8.23,8.23,0,0,1-3-6.66q0-5.91,5.67-8.37a31.87,31.87,0,0,1,5.42-1.56c3.07-.72,5.06-1.21,6-1.48,2.9-.88,4.35-2.27,4.35-4.19v-3.21a4.59,4.59,0,0,0-2.63-4.18,10,10,0,0,0-4.84-1.15c-3,0-5.25.65-6.7,2s-2.5,3.53-3.16,6.65a.59.59,0,0,1-.58.41.43.43,0,0,1-.49-.49l-.08-9.86c0-.49.27-.76.82-.82.88-.05,2.6-.25,5.18-.57a41.59,41.59,0,0,1,5.17-.42q6.57,0,9.69,2.55,3.54,3,3.53,9.69V180q0,3.21,2.63,3.7l2.47.49C149.15,184.4,149.59,184.7,149.59,185.14Zm-12.32-7.89v-8.13q0-1.06-.9-.66l-10.76,3.12q-4,1.15-4,4.77,0,7.14,6.41,7.14a11.85,11.85,0,0,0,6.57-1.76A5.21,5.21,0,0,0,137.27,177.25Z"/><path class="cls-1" d="M189.92,185.14c0,.43-.44.65-1.32.65H177.27c-.71,0-1.07-.19-1.07-.57a7.21,7.21,0,0,1,1-1.11c.68-.69,1-2.31,1-4.89v-17q0-7.4-7.63-7.39a11.26,11.26,0,0,0-5.67,1.72q-3.28,2.05-3.29,4.68V180a3.45,3.45,0,0,0,2.71,3.7l1.73.49c.87.28,1.31.58,1.31.91s-.44.65-1.31.65H151.4q-1.32,0-1.32-.57t1.32-1l1.72-.49a3.43,3.43,0,0,0,2.63-3.7V157c0-2.13-.93-3.34-2.79-3.61l-3.37-.49c-.88-.11-1.32-.41-1.32-.91s.44-.65,1.32-.65h9.28a1.36,1.36,0,0,1,1.31.9l1.32,2.87c.27.55.55.58.82.09q3.52-4.68,10.27-4.69a11,11,0,0,1,8.42,3.41q3.15,3.42,3.16,9.65V180a3.44,3.44,0,0,0,2.71,3.7l1.72.49C189.48,184.51,189.92,184.81,189.92,185.14Z"/><path class="cls-1" d="M218.58,176.18q0,10.44-12.73,10.43A43.59,43.59,0,0,1,194.27,185c-.71-.22-1.15-1-1.31-2.38,0-1,0-2-.09-3.08s-.19-2.51-.41-4.31c-.11-.66.08-1,.58-1s.79.3.9.91c1.15,3.28,2.77,5.53,4.85,6.73,1.8,1.1,4.49,1.64,8.05,1.64a8.34,8.34,0,0,0,4.6-1.31,4.68,4.68,0,0,0,2.38-4q0-3.53-5.34-5.83-4.84-2.06-9.78-4.19-5.34-3-5.33-8,0-9.69,11.41-9.7a86.77,86.77,0,0,1,11.17,1c.55.06.82.33.82.82l.17,8.13c0,.66-.17,1-.49,1s-.5-.25-.66-.74q-1.48-4.44-4.11-5.91c-1.53-.88-4-1.32-7.39-1.32q-6.4,0-6.41,4.93,0,2.79,5.34,5.09,9.45,4,10,4.36Q218.58,171,218.58,176.18Z"/><path class="cls-1" d="M242.07,185.22c0,.38-.44.57-1.31.57H223.51q-1.32,0-1.32-.57c0-.5.44-.82,1.32-1l2.54-.49q2.72-.5,2.71-3.7V157q0-3.12-2.71-3.61l-2.79-.49c-.87-.17-1.31-.47-1.31-.91s.44-.65,1.31-.65h10.6c.82,0,1.2.32,1.15,1l-.33,5.42V180c0,2.19.9,3.43,2.71,3.7l3.37.49C241.63,184.34,242.07,184.67,242.07,185.22ZM235,137.74a4.15,4.15,0,0,1-1.43,3.08,4.49,4.49,0,0,1-3.21,1.36,3.42,3.42,0,0,1-2.4-1,3.06,3.06,0,0,1-1-2.34,4.18,4.18,0,0,1,1.38-3.12,4.49,4.49,0,0,1,3.17-1.31Q235,134.46,235,137.74Z"/><path class="cls-1" d="M280.67,154c0,.54-.3.82-.9.82l-8.95.25a11.67,11.67,0,0,1-1.4,16.53A16,16,0,0,1,259,175h-2.38c-3.29,0-4.93,1.06-4.93,3.16q0,2.76,3.45,2.76c.33,0,1.67-.06,4-.17s4.25-.16,5.67-.16c4,0,6.95.63,8.87,1.89q3.38,2.3,3.37,7.8,0,6.74-6.74,10.18a27.43,27.43,0,0,1-12.73,2.72,17.13,17.13,0,0,1-9.11-2.22,8.74,8.74,0,0,1-4.36-7.89q0-5.74,5.42-8.7a4.85,4.85,0,0,1-2-4,5.83,5.83,0,0,1,3.78-5.91c.38-.16.58-.36.58-.57s-.17-.42-.51-.58c-4.86-2.14-7.3-5.69-7.3-10.68a10.82,10.82,0,0,1,4.25-8.83,16.31,16.31,0,0,1,10.52-3.41,18.24,18.24,0,0,1,7.18,1.4l13.7-.57c.6,0,.9.22.9.65Zm-8.21,37.61c0-2.13-.71-3.64-2.14-4.51s-3.86-1.32-7.31-1.32c-1.58,0-3.87,0-6.85.13l-5,.12q-2.55.83-2.55,5,0,3.85,4.35,5.91a18.64,18.64,0,0,0,7.89,1.56Q272.46,198.52,272.46,191.62Zm-4.76-28q0-9.94-9.33-9.94-8.09,0-8.09,8.13a11.06,11.06,0,0,0,2.4,7.47q2.38,2.81,7,2.8a7.56,7.56,0,0,0,5.94-2.3A8.88,8.88,0,0,0,267.7,163.62Z"/><path class="cls-1" d="M319.28,185.14c0,.43-.44.65-1.32.65H306.63c-.72,0-1.07-.19-1.07-.57a7.21,7.21,0,0,1,1-1.11q1-1,1-4.23V162.22q0-7.4-7.64-7.39a10.82,10.82,0,0,0-5.75,1.72q-3.2,2-3.2,4.68V180a3.44,3.44,0,0,0,2.71,3.7l1.73.49c.87.28,1.31.58,1.31.91s-.44.65-1.31.65H280.75c-.87,0-1.31-.19-1.31-.57s.44-.71,1.31-1l1.73-.49a3.43,3.43,0,0,0,2.63-3.7v-42c0-2.13-.91-3.34-2.71-3.61L279,134c-.88-.11-1.31-.41-1.31-.91s.43-.65,1.31-.65H290.2c.82,0,1.2.32,1.15,1l-.33,5.42v16a.39.39,0,0,0,.66.33q3.52-4.68,10.27-4.69a10.92,10.92,0,0,1,8.41,3.41q3.17,3.42,3.17,9.65V180a3.44,3.44,0,0,0,2.71,3.7l1.8.49C318.86,184.56,319.28,184.86,319.28,185.14Z"/><path class="cls-1" d="M343.34,183.71a.51.51,0,0,1-.16.65,11.25,11.25,0,0,1-6.74,1.84q-10.1,0-10.1-7.41V155.4h-6.49c-.55,0-.82-.25-.82-.77v-2.48c0-.57.27-.85.82-.85h4.76q1.39,0,2.22-1.89l3.86-9.2c.22-.49.49-.74.82-.74.5,0,.74.33.74,1V151.3h9.36c.55,0,.83.22.83.65v2.8c0,.43-.25.65-.74.65h-9.45v21.82q0,6.53,5.09,6.52a9.31,9.31,0,0,0,4.77-1.15c.33-.33.6-.3.82.08Z"/><path class="cls-1" d="M396,172.82l-2,12.07a1,1,0,0,1-1.15.9H360.92q-1.32,0-1.32-.57t1.32-1q2.54-.57,2.54-3.69V142.18q0-3.21-2.71-3.61l-3.37-.5c-.87-.11-1.31-.41-1.31-.9s.44-.66,1.31-.66h18.48q1.32,0,1.32.57c0,.5-.44.82-1.32,1l-2.71.49c-1.86.33-2.79,1.55-2.79,3.68v38.63c0,1,.58,1.47,1.73,1.47h10.26c3.95-.05,6.76-1,8.46-2.87q1.89-2,3.37-7.48c.22-.71.6-1.06,1.15-1.06S396.21,171.56,396,172.82Z"/><path class="cls-1" d="M431.47,185.14c0,.43-.44.65-1.32.65H422a1.52,1.52,0,0,1-1.39-.9l-.66-1.15a1,1,0,0,0-.82-.58,1.9,1.9,0,0,0-1.07.41,17.44,17.44,0,0,1-10.27,3,11,11,0,0,1-7.14-2.21,8.23,8.23,0,0,1-3-6.66q0-5.91,5.67-8.37a31.58,31.58,0,0,1,5.42-1.56q4.59-1.08,6-1.48c2.91-.88,4.36-2.27,4.36-4.19v-3.21a4.59,4.59,0,0,0-2.63-4.18,10,10,0,0,0-4.85-1.15c-3,0-5.24.65-6.69,2s-2.51,3.53-3.16,6.65a.59.59,0,0,1-.58.41.43.43,0,0,1-.49-.49l-.08-9.86c0-.49.27-.76.82-.82.87-.05,2.6-.25,5.17-.57a41.72,41.72,0,0,1,5.18-.42q6.57,0,9.69,2.55,3.53,3,3.53,9.69V180q0,3.21,2.63,3.7l2.46.49C431,184.4,431.47,184.7,431.47,185.14Zm-12.32-7.89v-8.13c0-.71-.3-.93-.91-.66l-10.75,3.12q-4,1.15-4,4.77,0,7.14,6.41,7.14a11.85,11.85,0,0,0,6.57-1.76A5.23,5.23,0,0,0,419.15,177.25Z"/><path class="cls-1" d="M465.88,168.79q0,8-4.35,12.9a13.71,13.71,0,0,1-10.68,4.92,26.07,26.07,0,0,1-6.82-.82,23.07,23.07,0,0,0-3.69-.82,5.45,5.45,0,0,0-2.47.44,14.15,14.15,0,0,1-1.56.79c-.65,0-1-.3-1-.9a2,2,0,0,1,.45-.86,2.8,2.8,0,0,0,.45-1.69V138.07c0-2.13-.9-3.34-2.71-3.61l-3.37-.49c-.87-.11-1.31-.41-1.31-.91s.44-.65,1.31-.65h11.17c.83,0,1.21.32,1.15,1l-.32,5.42v15.44c0,.28.1.44.32.5a.58.58,0,0,0,.58-.17,11.68,11.68,0,0,1,8.7-4.11,12.5,12.5,0,0,1,9.86,4.19Q465.89,159.52,465.88,168.79Zm-6.24,0q0-14.08-9.28-14.09-4.61,0-7.64,4.67a3.25,3.25,0,0,0-.57,1.72v14.17q0,4.42,1.8,6.31,1.89,2.13,6.57,2.13a7.52,7.52,0,0,0,6.66-4Q459.64,175.64,459.64,168.75Z"/><path class="cls-1" d="M496.6,176.18q0,10.44-12.73,10.43A43.48,43.48,0,0,1,472.29,185c-.71-.22-1.15-1-1.32-2.38,0-1,0-2-.08-3.08s-.19-2.51-.41-4.31c-.11-.66.08-1,.58-1s.79.3.9.91c1.15,3.28,2.76,5.53,4.85,6.73,1.8,1.1,4.48,1.64,8,1.64a8.31,8.31,0,0,0,4.6-1.31,4.67,4.67,0,0,0,2.39-4q0-3.53-5.34-5.83-4.85-2.06-9.78-4.19-5.34-3-5.34-8,0-9.69,11.42-9.7a86.77,86.77,0,0,1,11.17,1c.55.06.82.33.82.82l.17,8.13c0,.66-.17,1-.5,1s-.49-.25-.65-.74q-1.49-4.44-4.11-5.91c-1.53-.88-4-1.32-7.39-1.32q-6.4,0-6.41,4.93,0,2.79,5.34,5.09,9.45,4,10,4.36Q496.6,171,496.6,176.18Z"/><rect class="cls-1" x="179.9" y="34" width="154.24" height="4.36"/><path class="cls-1" d="M270.14,85.37,270.06,81c1.24,0,3.05,0,5.14,0,4.63,0,11,0,13.71-.28a14.89,14.89,0,0,0,8.51-4c3.41-3.08,5.1-7.06,5.16-12.18,0-1.64,0-3.28,0-4.92,0-2.58,0-5.25.09-7.9a17.67,17.67,0,0,1,17.56-17.56c3.14-.09,6.32-.08,9.4-.07h4v4.36c-1.33,0-2.66,0-4,0-3,0-6.19,0-9.27.07a13.26,13.26,0,0,0-13.32,13.33c-.08,2.58-.08,5.23-.09,7.78,0,1.66,0,3.31,0,5A20.12,20.12,0,0,1,300.34,80a19.34,19.34,0,0,1-10.94,5.06c-3,.34-9.21.33-14.21.31C273.12,85.36,271.33,85.35,270.14,85.37Z"/><path class="cls-1" d="M242.35,85.37c-1.22,0-2.51,0-3.89,0a82.68,82.68,0,0,1-9.55-.31A19.37,19.37,0,0,1,218,80c-4.3-3.88-6.52-9-6.6-15.36,0-1.65,0-3.31,0-5,0-2.56,0-5.2-.08-7.79a13.27,13.27,0,0,0-13.33-13.33c-3.13-.08-8.74-.08-13.24-.07-1.85,0-3.52,0-4.8,0l0-4.36h4.77c4.54,0,10.18,0,13.38.07a17.67,17.67,0,0,1,17.56,17.56c.07,2.65.08,5.32.09,7.91,0,1.64,0,3.28,0,4.91.07,5.12,1.76,9.1,5.17,12.18a14.86,14.86,0,0,0,8.51,4,84.4,84.4,0,0,0,9,.28c1.4,0,2.72,0,4,0Z"/><path class="cls-2" d="M256.86,97.52A14.47,14.47,0,0,1,242.39,82.9a14.26,14.26,0,0,1,14.67-14.32A14.46,14.46,0,0,1,271.33,83.1,14.69,14.69,0,0,1,256.86,97.52Z"/><path class="cls-1" d="M348.8,21.45A14.45,14.45,0,0,1,363.08,36a14.66,14.66,0,0,1-14.45,14.43,14.45,14.45,0,0,1-14.49-14.6A14.27,14.27,0,0,1,348.8,21.45Z"/><path class="cls-3" d="M164.39,20.16A15.75,15.75,0,0,1,179.9,36a16,16,0,0,1-15.78,15.67,15.75,15.75,0,0,1-15.74-16A15.55,15.55,0,0,1,164.39,20.16Z"/><path class="cls-1" d="M256.72,99.43a16.38,16.38,0,0,1,.08-32.76H257a16.37,16.37,0,0,1,16.16,16.45,16.54,16.54,0,0,1-16.38,16.31Zm.08-28.94a12.41,12.41,0,0,0-12.59,12.42,12.59,12.59,0,0,0,3.67,9,12.4,12.4,0,0,0,8.84,3.68v1.91l0-1.91a12.71,12.71,0,0,0,12.57-12.53A12.55,12.55,0,0,0,257,70.49Z"/><path class="cls-1" d="M348.46,52.3a16.13,16.13,0,0,1-11.53-4.8,16.38,16.38,0,0,1-4.79-11.72,16.19,16.19,0,0,1,16.39-16.24h.2A16.34,16.34,0,0,1,364.9,36,16.54,16.54,0,0,1,348.55,52.3Zm.07-28.94A12.41,12.41,0,0,0,336,35.8a12.57,12.57,0,0,0,3.68,9,12.34,12.34,0,0,0,8.82,3.67h.07a12.7,12.7,0,0,0,12.55-12.54,12.51,12.51,0,0,0-12.4-12.58Z"/><path class="cls-1" d="M164,53.59a17.44,17.44,0,0,1-12.46-5.18,17.71,17.71,0,0,1-5.16-12.69,17.49,17.49,0,0,1,17.7-17.47h.24A17.67,17.67,0,0,1,181.72,36,17.85,17.85,0,0,1,164,53.59Zm.08-31.52A13.7,13.7,0,0,0,150.2,35.75a13.89,13.89,0,0,0,4.05,10A13.63,13.63,0,0,0,164,49.77v1.91l0-1.91A14,14,0,0,0,177.9,36,13.59,13.59,0,0,0,174,26.27a13.74,13.74,0,0,0-9.72-4.2Z"/></svg>
\ No newline at end of file
diff --git a/pandas/web/pandas/static/img/partners/tidelift.svg b/pandas/web/pandas/static/img/partners/tidelift.svg
new file mode 100644
index 0000000000000..af12d68417235
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/tidelift.svg
@@ -0,0 +1,33 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 21.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Artwork" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 190.1 33" style="enable-background:new 0 0 190.1 33;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#4B5168;}
+	.st1{fill:#F6914D;}
+</style>
+<g>
+	<path class="st0" d="M33.4,27.7V5.3c0-2.3,0-2.3,2.4-2.3c2.4,0,2.4,0,2.4,2.3v22.4c0,2.3,0,2.3-2.4,2.3
+		C33.4,29.9,33.4,29.9,33.4,27.7z"/>
+	<path class="st0" d="M45,26.4V6.6c0-3.6,0-3.6,3.6-3.6h5.8c7.8,0,12.5,3.9,13,10.2c0.2,2.2,0.2,3.4,0,5.5
+		c-0.5,6.3-5.3,11.2-13,11.2h-5.8C45,29.9,45,29.9,45,26.4z M54.3,25.4c5.3,0,8-3,8.3-7.1c0.1-1.8,0.1-2.8,0-4.6
+		c-0.3-4.2-3-6.1-8.3-6.1h-4.5v17.8H54.3z"/>
+	<path class="st0" d="M73.8,26.4V6.6c0-3.6,0-3.6,3.6-3.6h13.5c2.3,0,2.3,0,2.3,2.2c0,2.2,0,2.2-2.3,2.2H78.6v6.9h11
+		c2.2,0,2.2,0,2.2,2.1c0,2.1,0,2.1-2.2,2.1h-11v6.9h12.3c2.3,0,2.3,0,2.3,2.2c0,2.3,0,2.3-2.3,2.3H77.4
+		C73.8,29.9,73.8,29.9,73.8,26.4z"/>
+	<path class="st0" d="M100,26.4v-21c0-2.3,0-2.3,2.4-2.3c2.4,0,2.4,0,2.4,2.3v20.2h11.9c2.4,0,2.4,0,2.4,2.2c0,2.2,0,2.2-2.4,2.2
+		h-13.1C100,29.9,100,29.9,100,26.4z"/>
+	<path class="st0" d="M125.8,27.7V5.3c0-2.3,0-2.3,2.4-2.3c2.4,0,2.4,0,2.4,2.3v22.4c0,2.3,0,2.3-2.4,2.3
+		C125.8,29.9,125.8,29.9,125.8,27.7z"/>
+	<path class="st0" d="M137.4,27.7V6.6c0-3.6,0-3.6,3.6-3.6h13.5c2.3,0,2.3,0,2.3,2.2c0,2.2,0,2.2-2.3,2.2h-12.2v7.2h11.3
+		c2.3,0,2.3,0,2.3,2.2c0,2.2,0,2.2-2.3,2.2h-11.3v8.6c0,2.3,0,2.3-2.4,2.3S137.4,29.9,137.4,27.7z"/>
+	<path class="st0" d="M24.2,3.1H5.5c-2.4,0-2.4,0-2.4,2.2c0,2.2,0,2.2,2.4,2.2h7v4.7v3.2l4.8-3.7v-1.1V7.5h7c2.4,0,2.4,0,2.4-2.2
+		C26.6,3.1,26.6,3.1,24.2,3.1z"/>
+	<path class="st1" d="M12.5,20v7.6c0,2.3,0,2.3,2.4,2.3c2.4,0,2.4,0,2.4-2.3V16.3L12.5,20z"/>
+	<g>
+		<path class="st0" d="M165.9,3.1h18.7c2.4,0,2.4,0,2.4,2.2c0,2.2,0,2.2-2.4,2.2h-7v4.7v3.2l-4.8-3.7v-1.1V7.5h-7
+			c-2.4,0-2.4,0-2.4-2.2C163.5,3.1,163.5,3.1,165.9,3.1z"/>
+		<path class="st1" d="M177.6,20v7.6c0,2.3,0,2.3-2.4,2.3c-2.4,0-2.4,0-2.4-2.3V16.3L177.6,20z"/>
+	</g>
+</g>
+</svg>
diff --git a/pandas/web/pandas/static/img/partners/two_sigma.svg b/pandas/web/pandas/static/img/partners/two_sigma.svg
new file mode 100644
index 0000000000000..d38df12766ed6
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/two_sigma.svg
@@ -0,0 +1 @@
+<svg width="230" height="42" viewBox="0 0 230 42" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><title>Logo</title><defs><path id="a" d="M19.436 21.668V1.025H0v20.643h19.435z"></path></defs><g fill="none" fill-rule="evenodd"><path fill="#2D2D2D" d="M59.06 13.464h-7.137v-3.155h17.811v3.155H62.6V30.95h-3.54zm14.01-3.155h3.745l4.747 15.66h.06l4.483-15.66h3.301l4.454 15.66h.059l4.777-15.66h3.716L95.895 30.95H92.09l-4.335-15.127h-.059L83.361 30.95h-3.804zm41.214-.355c5.986 0 10.527 4.158 10.527 10.556 0 6.55-4.541 10.794-10.527 10.794-5.985 0-10.558-4.245-10.558-10.794 0-6.398 4.573-10.556 10.558-10.556m0 18.285c3.892 0 6.93-2.89 6.93-7.729 0-4.658-3.007-7.518-6.93-7.518-3.922 0-6.93 2.86-6.93 7.518 0 4.839 3.038 7.73 6.93 7.73m40.846-17.931h3.539V30.95h-3.54V19.41zm18.744-.355c2.832 0 5.222.885 7.313 2.33 0 0-2.026 2.374-2.128 2.311-1.56-1-3.21-1.574-5.096-1.574-4.247 0-7.048 3.068-7.048 7.433 0 4.746 2.624 7.785 7.048 7.785 1.534 0 3.067-.385 4.13-1.003v-4.897h-5.19v-2.623h8.462v9.347c-2.007 1.416-4.63 2.24-7.49 2.24-6.46 0-10.587-4.363-10.587-10.85 0-6.075 4.187-10.499 10.586-10.499m12.506.355h3.57l6.812 9.701 6.811-9.701h3.541V30.95h-3.421V15.558l-6.962 9.73-6.958-9.73V30.95h-3.392z"></path><g transform="translate(210.418 9.283)"><mask id="b" fill="#fff"><use xlink:href="#a"></use></mask><path d="M7.639 1.025h4.158l7.64 20.643H15.63l-1.561-4.454H5.368l-1.533 4.454H0L7.639 1.025zM6.34 14.354h6.725L9.734 4.74h-.06L6.34 14.354z" fill="#2D2D2D" mask="url(#b)"></path></g><path d="M136.826 26.498c1.861 1.007 3.618 1.68 5.887 1.68 2.715 0 4.069-1.18 4.069-2.83 0-4.66-11.616-1.594-11.616-9.466 0-3.303 2.74-5.928 7.37-5.928 2.714 0 5.443.653 7.579 1.902l-2.314 2.361c-1.68-.72-3.11-1.137-5.146-1.137-2.389 0-3.806 1.21-3.806 2.744 0 4.63 11.62 1.473 11.62 9.494 0 3.393-2.567 5.985-7.756 5.985-3.035 0-6.33-1.076-8.273-2.419l2.386-2.386z" fill="#2D2D2D"></path><path fill="#009AA6" d="M20.625 0L0 20.63l20.625 20.628 20.63-20.628z"></path><path d="M9.748 26.478c-.16-6.605 7.789-5.746 7.789-9.13 0-1.1-.863-2.041-2.784-2.041-1.401 0-2.743.701-3.724 1.602l-1.46-1.463c1.259-1.18 3.223-2.14 5.284-2.14 3.304 0 4.986 1.842 4.986 4.003 0 4.986-7.728 4.104-7.728 8.27h7.607v1.98h-9.95l-.02-1.081zm15.937-.5c-1.521 0-2.423-.98-2.423-2.862 0-2.404 1.602-4.566 3.525-4.566 1.5 0 2.402.981 2.402 2.883 0 2.401-1.582 4.545-3.504 4.545zm9.713-9.25h-8.444v.003c-3.437.005-6.033 2.745-6.033 6.403 0 2.905 1.881 4.666 4.544 4.666 3.464 0 6.067-2.743 6.067-6.386 0-1.182-.313-2.173-.856-2.935h2.947l1.775-1.75z" fill="#FFF"></path></g></svg>
diff --git a/pandas/web/pandas/static/img/partners/voltron_data.svg b/pandas/web/pandas/static/img/partners/voltron_data.svg
new file mode 100644
index 0000000000000..0fb7dfd850166
--- /dev/null
+++ b/pandas/web/pandas/static/img/partners/voltron_data.svg
@@ -0,0 +1,52 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 26.3.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 1887.7 876.4" style="enable-background:new 0 0 1887.7 876.4;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#005050;}
+</style>
+<g>
+	<g>
+		<g>
+			<path class="st0" d="M943.8,0L943.8,0H791.1H638.5l76.3,132.2l76.3,132.2l76.3,132.2l76.4,132.2l76.3-132.2l76.3-132.2
+				l-76.3-132.2L943.8,0z M1000,385l-56.2,97.4L887.6,385L818,264.4l69.6-120.6l56.2-97.4l56.2,97.4l69.6,120.6L1000,385z"/>
+			<polygon class="st0" points="1082.5,0 1096,23.5 1208.4,23.5 1152.1,121.1 1165.6,144.6 1249.1,0 			"/>
+		</g>
+	</g>
+	<g>
+		<g>
+			<path class="st0" d="M3.2,758.3c-1-2.1-0.3-3.6,1.7-3.6h14.7c1.2-0.2,2.5,0.1,3.5,0.7c1,0.7,1.8,1.7,2.2,2.8l38.2,78.4h0.3
+				l38.3-78.4c0.4-1.2,1.2-2.2,2.2-2.8c1-0.7,2.3-0.9,3.5-0.7h14.7c2.1,0,2.9,1.5,1.7,3.6L66.8,873.9c-0.7,1.5-1.4,2-2.6,2h-1
+				c-1.2,0-1.9-0.5-2.5-2L3.2,758.3z"/>
+			<path class="st0" d="M245.4,752.4c16.2,0.3,31.6,7,43,18.6c11.3,11.6,17.7,27.1,17.7,43.4c0,16.2-6.3,31.8-17.7,43.4
+				c-11.3,11.6-26.8,18.3-43,18.6c-34.5,0-61.1-27.1-61.1-62C184.2,779.6,210.9,752.4,245.4,752.4z M245.4,858.4
+				c24.9,0,41.7-20.6,41.7-44s-16.7-44.1-41.7-44.1c-24.9,0-41.5,20.5-41.5,44.1S220.6,858.5,245.4,858.4L245.4,858.4z"/>
+			<path class="st0" d="M454.2,856.8c2.9,0,3.4,0.7,3.4,3.4v10.4c0,2.9-0.7,3.6-3.4,3.6h-70c-2.6,0-3.4-0.9-3.4-3.6V758.3
+				c0-2.7,0.8-3.6,3.4-3.6h12.6c2.5,0,3.4,0.9,3.4,3.6v98.5H454.2z"/>
+			<path class="st0" d="M593.9,754.7c2.9,0,3.6,0.9,3.6,3.6v10.9c0,2.9-0.9,3.6-3.6,3.6h-34.5v97.8c0,2.9-0.7,3.6-3.4,3.6h-12.6
+				c-2.6,0-3.4-0.9-3.4-3.6v-97.8h-34.7c-2.5,0-3.4-0.9-3.4-3.6v-10.9c0-2.7,0.9-3.6,3.4-3.6H593.9z"/>
+			<path class="st0" d="M686,870.7c0,2.9-0.9,3.6-3.4,3.6H670c-2.6,0-3.4-0.9-3.4-3.6V758.3c0-2.7,0.8-3.6,3.4-3.6h44.6
+				c20,0,42,9.9,42,37.5c0,21.9-13.8,32.6-29.6,36l28.5,41.5c1.9,2.9,0.7,4.5-2.6,4.5h-13.6c-1.2,0-2.4-0.4-3.5-1
+				c-1-0.6-1.9-1.5-2.5-2.6l-28.4-41h-19.1L686,870.7z M712.7,811.8c18.4,0,24.2-8.6,24.2-19.8s-5.7-19.4-24.2-19.4h-26.6v39.3
+				L712.7,811.8L712.7,811.8z"/>
+			<path class="st0" d="M885.7,752.4c16.2,0.3,31.6,7,43,18.6c11.3,11.6,17.7,27.1,17.7,43.4c0,16.2-6.3,31.8-17.7,43.4
+				c-11.3,11.6-26.8,18.3-43,18.6c-34.5,0-61.1-27.1-61.1-62C824.5,779.6,851.2,752.4,885.7,752.4z M885.7,858.4
+				c24.9,0,41.6-20.6,41.6-44s-16.7-44.1-41.6-44.1c-24.9,0-41.4,20.6-41.4,44.1C844.2,838.1,860.9,858.5,885.7,858.4L885.7,858.4z"
+				/>
+			<path class="st0" d="M1040.9,796.4h-0.3v74.2c0,2.9-0.9,3.6-3.4,3.6h-12.5c-2.9,0-3.6-0.9-3.6-3.6V754.7c0-1.2,0.5-1.7,1.9-1.7
+				h0.9c0.6,0,1.1,0.1,1.6,0.4c0.5,0.3,0.9,0.7,1.2,1.1l76.1,78h0.3v-74.2c0-2.7,0.7-3.6,3.4-3.6h12.5c2.9,0,3.6,0.9,3.6,3.6v115.8
+				c0,1.3-0.5,1.9-1.9,1.9h-0.9c-0.6,0-1.1-0.1-1.6-0.4c-0.5-0.3-0.9-0.6-1.2-1.1L1040.9,796.4z"/>
+			<path class="st0" d="M1335.7,754.7c34.5,0,60.3,24.9,60.3,59.8c0,34.8-25.7,59.8-60.3,59.8h-42.8c-2.6,0-3.4-0.9-3.4-3.6V758.3
+				c0-2.7,0.8-3.6,3.4-3.6H1335.7z M1309.1,772.5v84h26.6c24.8,0,40.6-18.6,40.6-42s-15.9-42.1-40.6-42.1L1309.1,772.5z"/>
+			<path class="st0" d="M1577,870.7c1,2.1,0.3,3.6-1.7,3.6h-14.7c-1.2,0.1-2.4-0.1-3.5-0.8c-1-0.7-1.8-1.6-2.2-2.8l-10.9-22.5h-55
+				l-10.9,22.5c-0.4,1.2-1.1,2.2-2.2,2.9c-1,0.7-2.3,0.9-3.5,0.7h-14.7c-2.1,0-2.9-1.5-1.7-3.6l57.4-115.6c0.7-1.5,1.4-2.1,2.6-2.1
+				h1c1.2,0,1.9,0.5,2.6,2.1L1577,870.7z M1516.6,790.1l-19.8,40.1h39.6l-19.6-40.1H1516.6z"/>
+			<path class="st0" d="M1714.4,754.7c2.9,0,3.6,0.9,3.6,3.6v10.9c0,2.9-0.9,3.6-3.6,3.6h-34.5v97.8c0,2.9-0.7,3.6-3.4,3.6h-12.7
+				c-2.5,0-3.4-0.9-3.4-3.6v-97.8h-34.7c-2.6,0-3.4-0.9-3.4-3.6v-10.9c0-2.7,0.9-3.6,3.4-3.6H1714.4z"/>
+			<path class="st0" d="M1884.4,870.7c1,2.1,0.3,3.6-1.7,3.6H1868c-1.2,0.1-2.4-0.1-3.5-0.8c-1-0.7-1.8-1.6-2.2-2.8l-10.9-22.5
+				h-54.9l-10.9,22.5c-0.3,1.2-1.1,2.2-2.2,2.9s-2.3,0.9-3.5,0.7h-14.7c-2.1,0-2.9-1.5-1.7-3.6l57.4-115.6c0.7-1.5,1.4-2.1,2.6-2.1
+				h1c1.2,0,1.9,0.5,2.6,2.1L1884.4,870.7z M1823.9,790.1l-19.8,40.1h39.6l-19.6-40.1H1823.9z"/>
+		</g>
+	</g>
+</g>
+</svg>
diff --git a/pandas/web/pandas/versions.json b/pandas/web/pandas/versions.json
new file mode 100644
index 0000000000000..2d2599ae8585b
--- /dev/null
+++ b/pandas/web/pandas/versions.json
@@ -0,0 +1,53 @@
+[
+    {
+        "name": "dev",
+        "version": "dev",
+        "url": "https://pandas.pydata.org/docs/dev/"
+    },
+    {
+        "name": "2.2 (stable)",
+        "version": "2.2",
+        "url": "https://pandas.pydata.org/docs/",
+        "preferred": true
+    },
+    {
+        "name": "2.1",
+        "version": "2.1",
+        "url": "https://pandas.pydata.org/pandas-docs/version/2.1/"
+    },
+    {
+        "name": "2.0",
+        "version": "2.0",
+        "url": "https://pandas.pydata.org/pandas-docs/version/2.0/"
+    },
+    {
+        "name": "1.5",
+        "version": "1.5",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.5/"
+    },
+    {
+        "name": "1.4",
+        "version": "1.4",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.4/"
+    },
+    {
+        "name": "1.3",
+        "version": "1.3",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.3/"
+    },
+    {
+        "name": "1.2",
+        "version": "1.2",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.2/"
+    },
+    {
+        "name": "1.1",
+        "version": "1.1",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.1/"
+    },
+    {
+        "name": "1.0",
+        "version": "1.0",
+        "url": "https://pandas.pydata.org/pandas-docs/version/1.0/"
+    }
+]
diff --git a/pandas/web/pandas_web.py b/pandas/web/pandas_web.py
new file mode 100755
index 0000000000000..b3872b829c73a
--- /dev/null
+++ b/pandas/web/pandas_web.py
@@ -0,0 +1,498 @@
+#!/usr/bin/env python3
+"""
+Simple static site generator for the pandas web.
+
+pandas_web.py takes a directory as parameter, and copies all the files into the
+target directory after converting markdown files into html and rendering both
+markdown and html files with a context. The context is obtained by parsing
+the file ``config.yml`` in the root of the source directory.
+
+The file should contain:
+```
+main:
+  template_path: <path_to_the_jinja2_templates_directory>
+  base_template: <template_file_all_other_files_will_extend>
+  ignore:
+  - <list_of_files_in_the_source_that_will_not_be_copied>
+  github_repo_url: <organization/repo-name>
+  context_preprocessors:
+  - <list_of_functions_that_will_enrich_the_context_parsed_in_this_file>
+  markdown_extensions:
+  - <list_of_markdown_extensions_that_will_be_loaded>
+```
+
+The rest of the items in the file will be added directly to the context.
+"""
+
+import argparse
+import collections
+import datetime
+import importlib
+import itertools
+import json
+import operator
+import os
+import pathlib
+import re
+import shutil
+import sys
+import time
+import typing
+
+import feedparser
+import jinja2
+import markdown
+from packaging import version
+import requests
+import yaml
+
+api_token = os.environ.get("GITHUB_TOKEN")
+if api_token is not None:
+    GITHUB_API_HEADERS = {"Authorization": f"Bearer {api_token}"}
+else:
+    GITHUB_API_HEADERS = {}
+
+
+class Preprocessors:
+    """
+    Built-in context preprocessors.
+
+    Context preprocessors are functions that receive the context used to
+    render the templates, and enriches it with additional information.
+
+    The original context is obtained by parsing ``config.yml``, and
+    anything else needed just be added with context preprocessors.
+    """
+
+    @staticmethod
+    def current_year(context):
+        """
+        Add the current year to the context, so it can be used for the copyright
+        note, or other places where it is needed.
+        """
+        context["current_year"] = datetime.datetime.now().year
+        return context
+
+    @staticmethod
+    def navbar_add_info(context):
+        """
+        Items in the main navigation bar can be direct links, or dropdowns with
+        subitems. This context preprocessor adds a boolean field
+        ``has_subitems`` that tells which one of them every element is. It
+        also adds a ``slug`` field to be used as a CSS id.
+        """
+        for i, item in enumerate(context["navbar"]):
+            context["navbar"][i] = dict(
+                item,
+                has_subitems=isinstance(item["target"], list),
+                slug=(item["name"].replace(" ", "-").lower()),
+            )
+        return context
+
+    @staticmethod
+    def blog_add_posts(context):
+        """
+        Given the blog feed defined in the configuration yaml, this context
+        preprocessor fetches the posts in the feeds, and returns the relevant
+        information for them (sorted from newest to oldest).
+        """
+        tag_expr = re.compile("<.*?>")
+        posts = []
+        # posts from the file system
+        if context["blog"]["posts_path"]:
+            posts_path = os.path.join(
+                context["source_path"], *context["blog"]["posts_path"].split("/")
+            )
+            for fname in os.listdir(posts_path):
+                if fname.startswith("index."):
+                    continue
+                link = (
+                    f"/{context['blog']['posts_path']}"
+                    f"/{os.path.splitext(fname)[0]}.html"
+                )
+                md = markdown.Markdown(
+                    extensions=context["main"]["markdown_extensions"]
+                )
+                with open(os.path.join(posts_path, fname), encoding="utf-8") as f:
+                    html = md.convert(f.read())
+                title = md.Meta["title"][0]
+                summary = re.sub(tag_expr, "", html)
+                try:
+                    body_position = summary.index(title) + len(title)
+                except ValueError as err:
+                    raise ValueError(
+                        f'Blog post "{fname}" should have a markdown header '
+                        f'corresponding to its "Title" element "{title}"'
+                    ) from err
+                summary = " ".join(summary[body_position:].split(" ")[:30])
+                posts.append(
+                    {
+                        "title": title,
+                        "author": context["blog"]["author"],
+                        "published": datetime.datetime.strptime(
+                            md.Meta["date"][0], "%Y-%m-%d"
+                        ),
+                        "feed": context["blog"]["feed_name"],
+                        "link": link,
+                        "description": summary,
+                        "summary": summary,
+                    }
+                )
+        # posts from rss feeds
+        for feed_url in context["blog"]["feed"]:
+            feed_data = feedparser.parse(feed_url)
+            for entry in feed_data.entries:
+                published = datetime.datetime.fromtimestamp(
+                    time.mktime(entry.published_parsed)
+                )
+                summary = re.sub(tag_expr, "", entry.summary)
+                posts.append(
+                    {
+                        "title": entry.title,
+                        "author": entry.author,
+                        "published": published,
+                        "feed": feed_data["feed"]["title"],
+                        "link": entry.link,
+                        "description": entry.description,
+                        "summary": summary,
+                    }
+                )
+        posts.sort(key=operator.itemgetter("published"), reverse=True)
+        context["blog"]["posts"] = posts[: context["blog"]["num_posts"]]
+        return context
+
+    @staticmethod
+    def maintainers_add_info(context):
+        """
+        Given the active maintainers defined in the yaml file, it fetches
+        the GitHub user information for them.
+        """
+        repeated = set(context["maintainers"]["active"]) & set(
+            context["maintainers"]["inactive"]
+        )
+        if repeated:
+            raise ValueError(f"Maintainers {repeated} are both active and inactive")
+
+        maintainers_info = {}
+        for user in (
+            context["maintainers"]["active"] + context["maintainers"]["inactive"]
+        ):
+            resp = requests.get(
+                f"https://api.github.com/users/{user}",
+                headers=GITHUB_API_HEADERS,
+                timeout=5,
+            )
+            if resp.status_code == 403:
+                sys.stderr.write(
+                    "WARN: GitHub API quota exceeded when fetching maintainers\n"
+                )
+                # if we exceed github api quota, we use the github info
+                # of maintainers saved with the website
+                resp_bkp = requests.get(
+                    context["main"]["production_url"] + "maintainers.json", timeout=5
+                )
+                resp_bkp.raise_for_status()
+                maintainers_info = resp_bkp.json()
+                break
+
+            resp.raise_for_status()
+            maintainers_info[user] = resp.json()
+
+        context["maintainers"]["github_info"] = maintainers_info
+
+        # save the data fetched from github to use it in case we exceed
+        # git github api quota in the future
+        with open(
+            pathlib.Path(context["target_path"]) / "maintainers.json",
+            "w",
+            encoding="utf-8",
+        ) as f:
+            json.dump(maintainers_info, f)
+
+        return context
+
+    @staticmethod
+    def home_add_releases(context):
+        context["releases"] = []
+
+        github_repo_url = context["main"]["github_repo_url"]
+        resp = requests.get(
+            f"https://api.github.com/repos/{github_repo_url}/releases",
+            headers=GITHUB_API_HEADERS,
+            timeout=5,
+        )
+        if resp.status_code == 403:
+            sys.stderr.write("WARN: GitHub API quota exceeded when fetching releases\n")
+            resp_bkp = requests.get(
+                context["main"]["production_url"] + "releases.json", timeout=5
+            )
+            resp_bkp.raise_for_status()
+            releases = resp_bkp.json()
+        else:
+            resp.raise_for_status()
+            releases = resp.json()
+
+        with open(
+            pathlib.Path(context["target_path"]) / "releases.json",
+            "w",
+            encoding="utf-8",
+        ) as f:
+            json.dump(releases, f, default=datetime.datetime.isoformat)
+
+        for release in releases:
+            if release["prerelease"]:
+                continue
+            published = datetime.datetime.strptime(
+                release["published_at"], "%Y-%m-%dT%H:%M:%SZ"
+            )
+            context["releases"].append(
+                {
+                    "name": release["tag_name"].lstrip("v"),
+                    "parsed_version": version.parse(release["tag_name"].lstrip("v")),
+                    "tag": release["tag_name"],
+                    "published": published,
+                    "url": (
+                        release["assets"][0]["browser_download_url"]
+                        if release["assets"]
+                        else ""
+                    ),
+                }
+            )
+        # sorting out obsolete versions
+        grouped_releases = itertools.groupby(
+            context["releases"],
+            key=lambda r: (r["parsed_version"].major, r["parsed_version"].minor),
+        )
+        context["releases"] = [
+            max(release_group, key=lambda r: r["parsed_version"].minor)
+            for _, release_group in grouped_releases
+        ]
+        # sorting releases by version number
+        context["releases"].sort(key=lambda r: r["parsed_version"], reverse=True)
+        return context
+
+    @staticmethod
+    def roadmap_pdeps(context):
+        """
+        PDEP's (pandas enhancement proposals) are not part of the bar
+        navigation. They are included as lists in the "Roadmap" page
+        and linked from there. This preprocessor obtains the list of
+        PDEP's in different status from the directory tree and GitHub.
+        """
+        KNOWN_STATUS = {
+            "Draft",
+            "Under discussion",
+            "Accepted",
+            "Implemented",
+            "Rejected",
+            "Withdrawn",
+        }
+        context["pdeps"] = collections.defaultdict(list)
+
+        # accepted, rejected and implemented
+        pdeps_path = (
+            pathlib.Path(context["source_path"]) / context["roadmap"]["pdeps_path"]
+        )
+        for pdep in sorted(pdeps_path.iterdir()):
+            if pdep.suffix != ".md":
+                continue
+            with pdep.open() as f:
+                title = f.readline()[2:]  # removing markdown title "# "
+                status = None
+                for line in f:
+                    if line.startswith("- Status: "):
+                        status = line.strip().split(": ", 1)[1]
+                        break
+                if status not in KNOWN_STATUS:
+                    raise RuntimeError(
+                        f'PDEP "{pdep}" status "{status}" is unknown. '
+                        f"Should be one of: {KNOWN_STATUS}"
+                    )
+            html_file = pdep.with_suffix(".html").name
+            context["pdeps"][status].append(
+                {
+                    "title": title,
+                    "url": f"pdeps/{html_file}",
+                }
+            )
+
+        # under discussion
+        github_repo_url = context["main"]["github_repo_url"]
+        resp = requests.get(
+            "https://api.github.com/search/issues?"
+            f"q=is:pr is:open label:PDEP draft:false repo:{github_repo_url}",
+            headers=GITHUB_API_HEADERS,
+            timeout=5,
+        )
+        if resp.status_code == 403:
+            sys.stderr.write("WARN: GitHub API quota exceeded when fetching pdeps\n")
+            resp_bkp = requests.get(
+                context["main"]["production_url"] + "pdeps.json", timeout=5
+            )
+            resp_bkp.raise_for_status()
+            pdeps = resp_bkp.json()
+        else:
+            resp.raise_for_status()
+            pdeps = resp.json()
+
+        with open(
+            pathlib.Path(context["target_path"]) / "pdeps.json", "w", encoding="utf-8"
+        ) as f:
+            json.dump(pdeps, f)
+
+        compiled_pattern = re.compile(r"^PDEP-(\d+)")
+
+        def sort_pdep(pdep: dict) -> int:
+            title = pdep["title"]
+            match = compiled_pattern.match(title)
+            if not match:
+                msg = f"""Could not find PDEP number in '{title}'. Please make sure to
+                write the title as: 'PDEP-num: {title}'."""
+                raise ValueError(msg)
+
+            return int(match[1])
+
+        context["pdeps"]["Under discussion"].extend(
+            {"title": pdep["title"], "url": pdep["html_url"]}
+            for pdep in sorted(pdeps["items"], key=sort_pdep)
+        )
+
+        return context
+
+
+def get_callable(obj_as_str: str) -> object:
+    """
+    Get a Python object from its string representation.
+
+    For example, for ``sys.stdout.write`` would import the module ``sys``
+    and return the ``write`` function.
+    """
+    components = obj_as_str.split(".")
+    attrs = []
+    while components:
+        try:
+            obj = importlib.import_module(".".join(components))
+        except ImportError:
+            attrs.insert(0, components.pop())
+        else:
+            break
+
+    if not obj:
+        raise ImportError(f'Could not import "{obj_as_str}"')
+
+    for attr in attrs:
+        obj = getattr(obj, attr)
+
+    return obj
+
+
+def get_context(config_fname: str, **kwargs):
+    """
+    Load the config yaml as the base context, and enrich it with the
+    information added by the context preprocessors defined in the file.
+    """
+    with open(config_fname, encoding="utf-8") as f:
+        context = yaml.safe_load(f)
+
+    context["source_path"] = os.path.dirname(config_fname)
+    context.update(kwargs)
+
+    preprocessors = (
+        get_callable(context_prep)
+        for context_prep in context["main"]["context_preprocessors"]
+    )
+    for preprocessor in preprocessors:
+        context = preprocessor(context)
+        msg = f"{preprocessor.__name__} is missing the return statement"
+        assert context is not None, msg
+
+    return context
+
+
+def get_source_files(source_path: str) -> typing.Generator[str, None, None]:
+    """
+    Generate the list of files present in the source directory.
+    """
+    for root, dirs, fnames in os.walk(source_path):
+        root_rel_path = os.path.relpath(root, source_path)
+        for fname in fnames:
+            yield os.path.join(root_rel_path, fname)
+
+
+def extend_base_template(content: str, base_template: str) -> str:
+    """
+    Wrap document to extend the base template, before it is rendered with
+    Jinja2.
+    """
+    result = '{% extends "' + base_template + '" %}'
+    result += "{% block body %}"
+    result += content
+    result += "{% endblock %}"
+    return result
+
+
+def main(
+    source_path: str,
+    target_path: str,
+) -> int:
+    """
+    Copy every file in the source directory to the target directory.
+
+    For ``.md`` and ``.html`` files, render them with the context
+    before copying them. ``.md`` files are transformed to HTML.
+    """
+    config_fname = os.path.join(source_path, "config.yml")
+
+    shutil.rmtree(target_path, ignore_errors=True)
+    os.makedirs(target_path, exist_ok=True)
+
+    sys.stderr.write("Generating context...\n")
+    context = get_context(config_fname, target_path=target_path)
+    sys.stderr.write("Context generated\n")
+
+    templates_path = os.path.join(source_path, context["main"]["templates_path"])
+    jinja_env = jinja2.Environment(loader=jinja2.FileSystemLoader(templates_path))
+
+    for fname in get_source_files(source_path):
+        if os.path.normpath(fname) in context["main"]["ignore"]:
+            continue
+
+        sys.stderr.write(f"Processing {fname}\n")
+        dirname = os.path.dirname(fname)
+        os.makedirs(os.path.join(target_path, dirname), exist_ok=True)
+
+        extension = os.path.splitext(fname)[-1]
+        if extension in (".html", ".md"):
+            with open(os.path.join(source_path, fname), encoding="utf-8") as f:
+                content = f.read()
+            if extension == ".md":
+                body = markdown.markdown(
+                    content, extensions=context["main"]["markdown_extensions"]
+                )
+                # Apply Bootstrap's table formatting manually
+                # Python-Markdown doesn't let us config table attributes by hand
+                body = body.replace("<table>", '<table class="table table-bordered">')
+                content = extend_base_template(body, context["main"]["base_template"])
+            context["base_url"] = "".join(["../"] * os.path.normpath(fname).count("/"))
+            content = jinja_env.from_string(content).render(**context)
+            fname_html = os.path.splitext(fname)[0] + ".html"
+            with open(
+                os.path.join(target_path, fname_html), "w", encoding="utf-8"
+            ) as f:
+                f.write(content)
+        else:
+            shutil.copy(
+                os.path.join(source_path, fname), os.path.join(target_path, dirname)
+            )
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Documentation builder.")
+    parser.add_argument(
+        "source_path", help="path to the source directory (must contain config.yml)"
+    )
+    parser.add_argument(
+        "--target-path", default="build", help="directory where to write the output"
+    )
+    args = parser.parse_args()
+    sys.exit(main(args.source_path, args.target_path))
diff --git a/pandas/web/tests/test_pandas_web.py b/pandas/web/tests/test_pandas_web.py
new file mode 100644
index 0000000000000..6b88263f98be6
--- /dev/null
+++ b/pandas/web/tests/test_pandas_web.py
@@ -0,0 +1,88 @@
+from unittest.mock import (  # noqa: TID251
+    mock_open,
+    patch,
+)
+
+import pytest
+import requests
+
+from web.pandas_web import Preprocessors
+
+
+class MockResponse:
+    def __init__(self, status_code: int, response: dict) -> None:
+        self.status_code = status_code
+        self._resp = response
+
+    def json(self):
+        return self._resp
+
+    @staticmethod
+    def raise_for_status() -> None:
+        return
+
+
+@pytest.fixture
+def context() -> dict:
+    return {
+        "main": {"github_repo_url": "pandas-dev/pandas"},
+        "target_path": "test_target_path",
+    }
+
+
+@pytest.fixture
+def mock_response(monkeypatch, request) -> None:
+    def mocked_resp(*args, **kwargs):
+        status_code, response = request.param
+        return MockResponse(status_code, response)
+
+    monkeypatch.setattr(requests, "get", mocked_resp)
+
+
+_releases_list = [
+    {
+        "prerelease": False,
+        "published_at": "2024-01-19T03:34:05Z",
+        "tag_name": "v1.5.6",
+        "assets": None,
+    },
+    {
+        "prerelease": False,
+        "published_at": "2023-11-10T19:07:37Z",
+        "tag_name": "v2.1.3",
+        "assets": None,
+    },
+    {
+        "prerelease": False,
+        "published_at": "2023-08-30T13:24:32Z",
+        "tag_name": "v2.1.0",
+        "assets": None,
+    },
+    {
+        "prerelease": False,
+        "published_at": "2023-04-30T13:24:32Z",
+        "tag_name": "v2.0.0",
+        "assets": None,
+    },
+    {
+        "prerelease": True,
+        "published_at": "2023-01-19T03:34:05Z",
+        "tag_name": "v1.5.3xd",
+        "assets": None,
+    },
+    {
+        "prerelease": False,
+        "published_at": "2027-01-19T03:34:05Z",
+        "tag_name": "v10.0.1",
+        "assets": None,
+    },
+]
+
+
+@pytest.mark.parametrize("mock_response", [(200, _releases_list)], indirect=True)
+def test_web_preprocessor_creates_releases(mock_response, context) -> None:
+    m = mock_open()
+    with patch("builtins.open", m):
+        context = Preprocessors.home_add_releases(context)
+        release_versions = [release["name"] for release in context["releases"]]
+        assert release_versions == ["10.0.1", "2.1.3", "2.0.0", "1.5.6"]